Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.

1 Four Ways to Represent a Function

1. (a) The point (  1, 2 ) is on the graph of f , so f ( 1)= 2 .

(b) When x=2 , y is about 2.8 , so f (2) 2.8 .
(c) f (x)=2 is equivalent to y=2 . When y=2 , we have x= 3 and x=1 .
(d) Reasonable estimates for x when y=0 are x= 2.5 and x=0.3 .
(e) The domain of f consists of all x  values on the graph of f . For this function, the domain is
 3 x 3 , or  3,3 . The range of f consists of all y  values on the graph of f . For this function,
the range is  2 y 3 , or  2,3 .
(f) As x increases from  1 to 3 , y increases from  2 to 3 . Thus, f is increasing on the interval  1,3
.

2. (a) The point (  4, 2 ) is on the graph of f , so f ( 4)= 2 . The point ( 3,4 ) is on the graph of g , so
g(3)=4 .
(b) We are looking for the values of x for which the y  values are equal. The y  values for f and g
are equal at the points (  2,1 ) and ( 2,2 ) , so the desired values of x are  2 and 2 .
(c) f (x)= 1 is equivalent to y= 1 . When y= 1 , we have x= 3 and x=4 .
(d) As x increases from 0 to 4 , y decreases from 3 to  1 . Thus, f is decreasing on the interval 0,4
.
(e) The domain of f consists of all x  values on the graph of f . For this function, the domain is
 4 x 4 , or  4,4 . The range of f consists of all y  values on the graph of f . For this function,
the range is  2 y 3 , or  2,3 .
(f) The domain of g is  4,3 and the range is 0.5,4 .

3. From Figure 1 in the text, the lowest point occurs at about ( t,a ) = ( 12, 85) . The highest point
occurs at about ( 17,115) . Thus, the range of the vertical ground acceleration is  85 a 115 . In
Figure 11, the range of the north south acceleration is approximately  325 a 485 . In Figure 12,
the range of the east west acceleration is approximately  210 a 200 .

4. Example 1: A car is driven at 60 mi / h for 2 hours. The distance d traveled by the car is a function
of the time t . The domain of the function is {t |0 t 2} , where t is measured in hours. The range of
the function is { d |0 d 120} , where d is measured in miles.

Example 2: At a certain university, the number of students N on campus at any time on a particular
day is a function of the time t after midnight. The domain of the function is {t |0 t 24} , where t is
measured in hours. The range of the function is { N |0 N k} , where N is an integer and k is the
largest number of students on campus at once.

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

Example 3: A certain employee is paid $8.00 per hour and works a maximum of 30 hours per week.
The number of hours worked is rounded down to the nearest quarter of an hour. This employee’s
gross weekly pay P is a function of the number of hours worked h . The domain of the function is
0,30 and the range of the function is { 0,2.00,4.00,... ,238.00,240.00} .

5. No, the curve is not the graph of a function because a vertical line intersects the curve more than
once. Hence, the curve fails the Vertical Line Test.

6. Yes, the curve is the graph of a function because it passes the Vertical Line Test. The domain is
[ 2,2] and the range is [ 1,2] .

7. Yes, the curve is the graph of a function because it passes the Vertical Line Test. The domain is
[ 3,2] and the range is  3, 2 )  [ 1,3] .

8. No, the curve is not the graph of a function since for x=0 ,  1 , and  2 , there are infinitely many
points on the curve.

9. The person’s weight increased to about 160 pounds at age 20 and stayed fairly steady for 10 years.
The person’s weight dropped to about 120 pounds for the next 5 years, then increased rapidly to
about 170 pounds. The next 30 years saw a gradual increase to 190 pounds. Possible reasons for the
drop in weight at 30 years of age: diet, exercise, health problems.

10. The salesman travels away from home from 8 to 9 A.M. and is then stationary until 10 : 00 . The
salesman travels farther away from 10 until noon. There is no change in his distance from home until
1 : 00 , at which time the distance from home decreases until 3 : 00 . Then the distance starts
increasing again, reaching the maximum distance away from home at 5 : 00 . There is no change from
5 until 6 , and then the distance decreases rapidly until 7 : 00 P.M., at which time the salesman
reaches home.

11. The water will cool down almost to freezing as the ice melts. Then, when the ice has melted, the
water will slowly warm up to room temperature.

2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

12. The summer solstice (the longest day of the year) is around June 21, and the winter solstice (the
shortest day) is around December 22.

13. Of course, this graph depends strongly on the geographical location!

14. The temperature of the pie would increase rapidly, level off to oven temperature, decrease rapidly,
and then level off to room temperature.

15.

16. (a)

(b)
(c)

3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

(d)

17. (a)
(b) From the graph, we estimate the number of cell phone subscribers in Malaysia to be about 540 in

1994 and 1450 in 1996.

18. (a)

(b) From the graph in part (a), we estimate the temperature at 11:00 A.M. to be about 84.5 C.

2
19. f (x)=3x  x+2.
2
f (2)=3(2)  2+2=12 2+2=12.
2
f ( 2)=3( 2)  ( 2)+2=12+2+2=16.

4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

2
f (a)=3a  a+2.
2 2
f ( a)=3( a)  ( a)+2=3a +a+2.
2 2 2 2
f (a+1)=3(a+1)  (a+1)+2=3(a +2a+1) a 1+2=3a +6a+3 a+1=3a +5a+4.
2 2
2 f (a)=2 f (a)=2(3a  a+2)=6a  2a+4.
2 2 2
f (2a)=3(2a)  (2a)+2=3(4a ) 2a+2=12a  2a+2.
2 22 2 4 2 4 2
f (a )=3(a )  (a )+2=3(a ) a +2=3a  a +2.
2
f (a)
2 2
= 3a  a+2 = 3a  a+2 ( 2
) ( 3a2 a+2)
4 3 2 3 2 2 4 3 2
=9a  3a +6a  3a +a  2a+6a  2a+4=9a  6a +13a  4a+4.
2 2 2 2 2
f (a+h)=3(a+h)  (a+h)+2=3(a +2ah+h ) a h+2=3a +6ah+3h  a h+2.

4 3 4
3 3
3 2
20. A spherical balloon with radius r+1 has volume V ( r+1 ) =  ( r+1 ) =  r +3r +3r+1 . We ( )
wish to find the amount of air needed to inflate the balloon from a radius of r to r+1 . Hence, we need
4 4 3 4
3
3 2
3 3
2
to find the difference V ( r+1 )  V ( r ) =  r +3r +3r+1   r =  3r +3r+1 . ( ) ( )
2
21. f (x)=x x , so f (2+h)=2+h (2+h) =2+h (4+4h+h )=2+h 4 4h h =
2 2 2
( h2+3h+2) ,
2 2 2 2 2
f (x+h)=x+h ( x+h ) =x+h (x +2xh+h )=x+h x  2xh h , and
2 2 2 2
f (x+h) f (x) x+h x  2xh h  x+x h 2xh h h(1 2x h)
= = = =1 2x h .
h h h h

x 2+h 2+h x+h

22. f (x)= , so f (2+h)= = , f (x+h)= , and
x+1 2+h+1 3+h x+h+1
x+h x

f (x+h) f (x) x+h+1 x+1 ( x+h ) ( x+1 )  x ( x+h+1 ) 1
= = = .
h h h ( x+h+1 ) ( x+1 ) ( x+h+1 ) ( x+1 )
1
23. f (x)=x/(3x 1) is defined for all x except when 0=3x 1
x= , so the domain is
3

{ x R| x
1
3 } =  ,
1
3

1
3
, .

/( 2
)
24. f (x)=(5x+4) x +3x+2 is defined for all x except when 0=x +3x+2

 1 , so the domain is { x R| x  2, 1} =( , 2) ( 2, 1) ( 1, ) .

2
0=(x+2)(x+1)
x= 2 or

25.
5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

3
f (t)= t + t is defined when t 0 . These values of t give real number results for t , whereas any
3
value of t gives a real number result for t . The domain is 0, ).

26. g(u)= u + 4 u is defined when u 0 and 4 u 0

u 4 . Thus, the domain is 0 u 4= 0,4 .

27. h(x)=1 / 4 2 2
x  5x is defined when x  5x>0
x(x 5)>0 . Note that x  5x 0 since that would
result in division by zero. The expression x(x 5) is positive if x<0 or x>5 . (See Appendix A for
2

methods for solving inequalities.) Thus, the domain is ( ,0) (5, ) .

2 2 2 2 2 2
28. h(x)= 4 x . Now y= 4 x  y =4 x
x +y =4 , so the graph is the top half of a circle of
radius 2 with center at the origin. The domain is { x|4 x2 0} = { x|4 x
2
} ={ x|2 x } =[ 2,2] .
From the graph, the range is 0 y 2 , or 0,2 .

29. f (x)=5 is defined for all real numbers, so the domain is R , or ( , ) . The graph of f is a
horizontal line with y  intercept 5 .

1
30. F(x)= (x+3) is defined for all real numbers, so the domain is R , or ( , ) . The graph of F is
2
3
a line with x  intercept  3 and y  intercept .
2

2
31. f (t)=t  6t is defined for all real numbers, so the domain is R , or ( , ) . The graph of f is a
2
parabola opening upward since the coefficient of t is positive. To find the t  intercepts, let y=0 and
2
solve for t . 0=t  6t=t(t 6)  t=0 and t=6 . The t  coordinate of the vertex is halfway between the t 
6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

2
intercepts, that is, at t=3 . Since f (3)=3  6 3= 9 , the vertex is (3, 9) .

2
4 t (2+t)(2 t)
32. H(t)= = , so for t 2 , H(t)=2+t . The domain is {t |t 2} . So the graph of H is
2 t 2 t
the same as the graph of the function f (t)=t+2 (a line) except for the hole at ( 2,4 ) .

33. g(x)= x 5 is defined when x 5 0 or x 5 , so the domain is 5, ) . Since y= x 5 

2 2
y =x 5  x=y +5 , we see that g is the top half of a parabola.

34.

F(x)= 2x+1 = { 2x+1

 (2x+1)
if 2x+1 0
if 2x+1<1
1
if x 
= { 2x+1
 2x 1
if x<
1
2

The domain is R , or ( , ).

35.
7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

G(x)=
3x+ x
x
. Since x = { x if x 0
 x if x<0
, we have

{ {
3x+x 4x
if x>0 if x>0
G(x)=
x
3x x
if x<0
=
x
2x
if x<0
= { 4 if x>0
2 if x<0
x x

Note that G is not defined for x=0 . The domain is ( ,0) (0, ).

36. g(x)=
x
x
2
. Since x = { x if x 0
 x if x<0
, we have

{
x

{
1
2 if x>0 if x>0
x x
g(x)=  x =
1
if x<0 if x<0
x
2  x

Note that g is not defined for x=0 . The domain is ( ,0) (0, ).

37. f (x)=
x
{ if x 0
x+1 if x>0
Domain is R , or ( , ) .

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

38. f (x)= {
2x+3 if x< 1
3 x if x  1
Domain is R , or ( , ) .

39. f (x)=
{ x+2 if x  1
x
2
if x> 1
2
Note that for x= 1 , both x+2 and x are equal to 1. Domain is R .

 1 if x  1
40. f (x)=

Domain is R .
{ 3x+2 if  1<x<1
7 2x if x 1

y y
2 1
41. Recall that the slope m of a line between the two points x ,y ( 1 1 ) and ( x ,y ) is m= x  x
2 2
2 1
and an

equation of the line connecting those two points is y y =m(x x ) . The slope of this line segment is
1 1
 6 1 7 7 7 4
= , so an equation is y 1= (x+2) . The function is f (x)= x ,  2 x 4 .
4 (  2 ) 6 6 6 3

3 (  2 ) 5 5
42. The slope of this line segment is = , so an equation is y+2= (x+3) . The function is
6 (  3) 9 9
5 1
f (x)= x ,  3 x 6 .
9 3

43. We need to solve the given equation for y .

9
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

2 2
x+ ( y 1 ) =0
( y 1 ) = x
y 1=  x
y=1  x . The expression with the positive radical
represents the top half of the parabola, and the one with the negative radical represents the bottom
half. Hence, we want f (x)=1  x . Note that the domain is x 0 .

2 2 2 2 2
44. ( x 1 ) +y =1
y= 1 (x 1) = 2x x . The top half is given by the function f (x)= 2x x ,
0 x 2 .

45. For  1 x 2 , the graph is the line with slope 1 and y intercept 1 , that is, the line y=x+1 . For
3 3 3
2<x 4 , the graph is the line with slope  and x intercept 4 , so y 0= (x 4)= x+6 . So the
2 2 2

function is f (x)=
{x+1

2
3
if  1 x 2
x+6 if 2<x 4

46. For x 0 , the graph is the line y=2 . For 0<x 1 , the graph is the line with slope  2 and y
intercept 2 , that is, the line y= 2x+2 . For x>1 , the graph is the line with slope 1 and x intercept 1 ,

that is, the line y=1( x 1 ) =x 1 . So the function is f (x)=

2 if x 0
{
 2x+2 if 0<x 1 .
x 1 if 1<x

47. Let the length and width of the rectangle be L and W . Then the perimeter is 2L+2W =20 and the
20 2L
area is A=LW . Solving the first equation for W in terms of L gives W = =10 L . Thus,
2
2
A(L)=L(10 L)=10L L . Since lengths are positive, the domain of A is 0<L<10 . If we further restrict
L to be larger than W , then 5<L<10 would be the domain.

48. Let the length and width of the rectangle be L and W . Then the area is LW =16 , so that W =16/L .
The perimeter is P=2L+2W , so P(L)=2L+2(16/L)=2L+32/L , and the domain of P is L>0 , since
lengths must be positive quantities. If we further restrict L to be larger than W , then L>4 would be
the domain.

49. Let the length of a side of the equilateral triangle be x . Then by the Pythagorean Theorem, the
2 1 2 2 2 2 1 2 3 2 3
height y of the triangle satisfies y + x =x , so that y =x  x = x and y= x . Using the
2 4 4 2
1 1 3 3 2
formula for the area A of a triangle, A= ( base ) ( height ) , we obtain A(x)= (x) x = x ,
2 2 2 4
with domain x>0 .

3 3
50. Let the volume of the cube be V and the length of an edge be L . Then V =L so L= V , and the
10
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

surface area is S(V )=6 ( 3 V ) 2=6V 2/3 , with domain V >0 .

51. Let each side of the base of the box have length x , and let the height of the box be h . Since the
2 2 2
volume is 2 , we know that 2=hx , so that h=2/x , and the surface area is S=x +4xh . Thus,
2 2 2
S(x)=x +4x(2/x )=x +(8/x) , with domain x>0 .

2
1  x 1 2
52. The area of the window is A=xh+  =xh+ , where h is the height of the
x
2 8 2
1 1
rectangular portion of the window. The perimeter is P=2h+x+  x=30
2h=30 x  x

2 2
1
h= ( 60 2x  x ) . Thus,
4

2
60 2x  x  x 1 2  2  2 4 2  2 2  +4
A(x)=x + =15x x x + x =15x x x =15x x
4 8 2 4 8 8 8 8

Since the lengths x and h must be positive quantities, we have x>0 and h>0 . For h>0 , we have 2h>0
1 60 60

30 x  x>0
60>2x+ x
x< . Hence, the domain of A is 0<x< .
2 2+ 2+

53. The height of the box is x and the length and width are L=20 2x , W =12 2x . Then V =LWx and
so
(
V (x)= ( 20 2x ) ( 12 2x ) (x)=4(10 x)(6 x)(x)=4x 60 16x+x =4x  64x +240x .
2
) 3 2

The sides L , W , and x must be positive. Thus, L>0

20 2x>0
x<10 ; W >0
12 2x>0
x<6 ; and
x>0 . Combining these restrictions gives us the domain 0<x<6 .
54.

{
$2.00 if 0.0<x 1.0
2.20 if 1.0<x 1.1
2.40 if 1.1<x 1.2
2.60 if 1.2<x 1.3
2.80 if 1.3<x 1.4
C(x)= 3.00 if 1.4<x 1.5
3.20 if 1.5<x 1.6
3.40 if 1.6<x 1.7
3.60 if 1.7<x 1.8
3.80 if 1.8<x 1.9
4.00 if 1.9<x<2.0

11
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

55. (a)
(b) On $14 , 000 , tax is assessed on $4000 , and 10%($4000)=$400 .
On $26 , 000 , tax is assessed on $16 , 000 , and 10%($10 , 000)+15%($6000)=$1000+$900=$1900 .
(c) As in part (b), there is $1000 tax assessed on $20,000 of income, so the graph of T is a line
segment from (10 , 000,0) to (20 , 000,1000) . The tax on $30,000 is $2500, so the graph of T for
x>20 , 000 is the ray with initial point (20 , 000,1000) that passes through (30 , 000,2500) .

56. One example is the amount paid for cable or telephone system repair in the home, usually
measured to the nearest quarter hour. Another example is the amount paid by a student in tuition fees,
if the fees vary according to the number of credits for which the student has registered.

57. f is an odd function because its graph is symmetric about the origin. g is an even function
because its graph is symmetric with respect to the y  axis.

58. f is not an even function since it is not symmetric with respect to the y  axis. f is not an odd
function since it is not symmetric about the origin. Hence, f is neither even nor odd. g is an even
function because its graph is symmetric with respect to the y  axis.

59. (a) Because an even function is symmetric with respect to the y  axis, and the point ( 5,3) is on
the graph of this even function, the point (  5,3) must also be on its graph.
(b) Because an odd function is symmetric with respect to the origin, and the point ( 5,3) is on the
graph of this odd function, the point (  5, 3) must also be on its graph.

60. (a) If f is even, we get the rest of the graph by reflecting about the y  axis.

12
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function


(b) If f is odd, we get the rest of the graph by rotating 180 about the origin.

 2
61. f (x)=x .
 2 1 1
f ( x) = (  x ) = 2
=
2
(  x) x
 2
=x = f (x)

So f is an even function.

 3
62. f (x)=x .
 3 1 1
f ( x) = (  x ) = 3
=
3
(  x)  x

=
1
3
= ( x 3) = f (x)
x

So f is odd.

2 2 2
63. f (x)=x +x , so f ( x)= (  x ) + (  x ) =x  x . Since this is neither f (x) nor  f (x) , the function f is
13
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.1 Four Ways to Represent a Function

neither even nor odd.

4 2
64. f (x)=x  4x .
f ( x) = (  x ) 4 4 (  x ) 2
4 2
=x  4x = f (x)

So f is even.

3
65. f (x)=x  x .
f ( x) = (  x ) 3 (  x ) = x3+x

= ( x3 x) = f (x)

So f is odd.

3 2 3 2 3 2
66. f (x)=3x +2x +1 , so f ( x)=3( x) +2( x) +1= 3x +2x +1 . Since this is neither f (x) nor  f (x) ,
the function f is neither even nor odd.

14
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

5
1. (a) f (x)= x is a root function with n=5 .
2
(b) g(x)= 1 x is an algebraic function because it is a root of a polynomial.
9 4
(c) h(x)=x +x is a polynomial of degree 9 .
2
x +1
(d) r(x)= 3
is a rational function because it is a ratio of polynomials.
x +x
(e) s(x)=tan 2x is a trigonometric function.
(f) t(x)=log x is a logarithmic function.
10

2. (a) y= ( x 6 ) / ( x+6 ) is a rational function because it is a ratio of polynomials.

2
(b) y=x+x /
x 1 is an algebraic function because it involves polynomials and roots of
polynomials.
x
(c) y=10 is an exponential function (notice that x is the exponent ).
10
(d) y=x is a power function (notice that x is the base ).
6 4
(e) y=2t +t   is a polynomial of degree 6 .
(f) y=cos  +sin  is a trigonometric function.

3. We notice from the figure that g and h are even functions (symmetric with respect to the y  axis)
5
and that f is an odd function (symmetric with respect to the origin). So (b) y=x must be f . Since
8 2
g is flatter than h near the origin, we must have (c) y=x matched with g and (a) y=x matched
with h .

4. (a) The graph of y=3x is a line (choice G ).

x
(b) y=3 is an exponential function (choice f ).
3
(c) y=x is an odd polynomial function or power function (choice F ).
3 1/3
(d) y= x =x is a root function (choice g ).

5. (a) An equation for the family of linear functions with slope 2 is y= f (x)=2x+b , where b is the y 
intercept.

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

(b) f (2)=1 means that the point ( 2,1 ) is on the graph of f . We can use the point slope form of a line
to obtain an equation for the family of linear functions through the point ( 2,1 ) . y 1=m ( x 2 ) , which
is equivalent to y=mx+ ( 1 2m ) in slope intercept form.

(c) To belong to both families, an equation must have slope m=2 , so the equation in part (b),
y=mx+ ( 1 2m ) , becomes y=2x 3 . It is the only function that belongs to both families.

6. All members of the family of linear functions f (x)=1+m(x+3) have graphs that are lines passing
through the point ( 3,1) .

7. All members of the family of linear functions f (x)=c x have graphs that are lines with slope  1 .
The y  intercept is c .

8. (a)

2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

(b) The slope of  4 means that for each increase of 1 dollar for a rental space, the number of spaces
rented decreases by 4 . The y  intercept of 200 is the number of spaces that would be occupied if
there were no charge for each space. The x  intercept of 50 is the smallest rental fee that results in no
spaces rented.

9. (a)
9 9 
(b) The slope of means that F increases degrees for each increase of 1 C. (Equivalently, F
5 5
increases by 9 when C increases by 5 and F decreases by 9 when C decreases by 5 .) The F 
intercept of 32 is the Fahrenheit temperature corresponding to a Celsius temperature of 0 .

10. (a) Let d= distance traveled (in miles) and t= time elapsed (in hours). At t=0 , d=0 and at t=50
1 5 5 40 0
minutes =50 = h, d=40 . Thus we have two points: ( 0,0 ) and ,40 , so m= =48 and
60 6 6 5
 0
6
so d=48t .

(b)

(c) The slope is 48 and represents the car’s speed in mi / h.

11. (a) Using N in place of x and T in place of y , we find the slope to be

T T 80 70 10 1 1 1 173
2 1
= = = . So a linear equation is T  80= ( N 173)  T  80= N 
N  N 173 113 60 6 6 6 6
2 1

3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

1 307 307
T= N+ =51.16 .
6 6 6
1
(b) The slope of means that the temperature in Fahrenheit degrees increases one sixth as rapidly
6
as the number of cricket chirps per minute. Said differently, each increase of 6 cricket chirps per

minute corresponds to an increase of 1 F.
1 307  
(c) When N=150 , the temperature is given approximately by T = ( 150 ) + =76.16 F  76 F.
6 6

12. (a) Let x denote the number of chairs produced in one day and y the associated cost. Using the
4800 2200 2600
points ( 100,2200 ) and ( 300,4800 ) we get the slope = =13 . So y 2200=13 ( x 100 )
300 100 200
 y=13x+900 .

(b) The slope of the line in part (a) is 13 and it represents the cost (in dollars) of producing each
additional chair.

(c) The y  intercept is 900 and it represents the fixed daily costs of operating the factory.

change in pressure 4.34

13. (a) We are given = =0.434 . Using P for pressure and d for
10 feet change in depth 10
depth with the point ( d,P ) = ( 0,15) , we have the slope intercept form of the line, P=0.434d+15 .
85
(b) When P=100 , then 100=0.434d+15 0.434d=85 d=  195.85 feet. Thus, the pressure is
0.434
2
100 lb/in at a depth of approximately 196 feet.

14. (a) Using d in place of x and C in place of y , we find the slope to be

C C 460 380 80 1
2 1
= = =
d  d 800 480 320 4
2 1
4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

1 1 1
So a linear equation is C 460= ( d 800 )  C 460= d 200 C= d+260 .
4 4 4

1
(b) Letting d=1500 we get C= ( 1500 ) +260=635 . The cost of driving 1500 miles is $635.
4

(c)
The slope of the line represents the cost per mile, $0.25 .
(d) The y  intercept represents the fixed cost, $260 .
(e) A linear function gives a suitable model in this situation because you have fixed monthly costs
such as insurance and car payments, as well as costs that increase as you drive, such as gasoline, oil,
and tires, and the cost of these for each additional mile driven is a constant.

15. (a) The data appear to be periodic and a sine or cosine function would make the best model. A
model of the form f (x)=a cos (bx)+c seems appropriate.
(b) The data appear to be decreasing in a linear fashion. A model of the form f (x)=mx+b seems
appropriate.

x
16. (a) The data appear to be increasing exponentially. A model of the form f (x)=a b or
x
f (x)=a b +c seems appropriate.
(b) The data appear to be decreasing similarly to the values of the reciprocal function. A model of the
form f (x)=a/x seems appropriate.

17. Some values are given to many decimal places. These are the results given by several computer
algebra systems  rounding is left to the reader.

(a)
A linear model does seem appropriate.
8.2 14.1
(b) Using the points ( 4000,14.1 ) and ( 60,000,8.2 ) , we obtain y 14.1= ( x 4000 ) or,
60,000 4000
equivalently, y  0.000105357x+14.521429 .

5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

(c) Using a computing device, we obtain the least squares regression line
y= 0.0000997855x+13.950764 .
The following commands and screens illustrate how to find the least squares regression line on a TI
83 Plus. Enter the data into list one (L1) and list two (L2). Press to enter the editor.

Find the regession line and store it in Y 1 . Press .

Note from the last figure that the regression line has been stored in Y 1 and that Plot1 has been turned
on (Plot1 is highlighted). You can turn on Plot1 from the Y= menu by placing the cursor on Plot1 and
pressing or by pressing .

Now press to produce a graph of the data and the regression line. Note that choice 9 of the
ZOOM menu automatically selects a window that displays all of the data.

(d) When x=25 , 000 , y 11.456 ; or about 11.5 per 100 population.
(e) When x=80 , 000 , y 5.968 ; or about a 6% chance.
(f) When x=200 , 000 , y is negative, so the model does not apply.

18. (a)
6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

(b)
Using a computing device, we obtain the least squares regression line y=4.856x 220.96 .

(c) When x=100 F, y=264.7 265 chirps / min.

19. (a)
A linear model does seem appropriate.

(b)
Using a computing device, we obtain the least squares regression line y=0.089119747x 158.2403249
, where x is the year and y is the height in feet.
(c) When x=2000 , the model gives y 20.00 ft. Note that the actual winning height for the 2000
Olympics is less than the winning height for 1996  so much for that prediction.
(d) When x=2100 , y 28.91 ft. This would be an increase of 9.49 ft from 1996 to 2100. Even though
there was an increase of 8.59 ft from 1900 to 1996, it is unlikely that a similar increase will occur
over the next 100 years.

20. By looking at the scatter plot of the data, we rule out the linear and logarithmic models.

7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.2 Mathematical Models: A Catalog of Essential Functions

We try various models:

Quadratic: 2
y=0.496x  62.2893x+1970.639
Cubic: 3 2
y=0.0201243201x  3.88037296x +247.6754468x 5163.935198
Quartic: 4 3 2
y=0.0002951049x  0.0654560995x +5.27525641x  180.2266511x+2203.210956
Exponential: y=2.41422994 ( 1.054516914 ) x
Power: 3
y=0.000022854971x .616078251

After examining the graphs of these models, we see that the cubic and quartic models are clearly the
best.

21.
3 2
Using a computing device, we obtain the cubic function y=ax +bx +cx+d with a=0.0012937 ,
b= 7.06142 , c=12 , 823 , and d= 7 , 743 , 770 . When x=1925 , y 1914 (million).

1.499661718
22. (a) T =1.000396048d
1.5 2 3
(b) The power model in part (a) is approximately T =d . Squaring both sides gives us T =d , so the
2 3
model matches Kepler’s Third Law, T =kd .

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

1. (a) If the graph of f is shifted 3 units upward, its equation becomes y= f (x)+3 .
(b) If the graph of f is shifted 3 units downward, its equation becomes y= f (x) 3 .
(c) If the graph of f is shifted 3 units to the right, its equation becomes y= f (x 3) .
(d) If the graph of f is shifted 3 units to the left, its equation becomes y= f (x+3) .
(e) If the graph of f is reflected about the x  axis, its equation becomes y= f (x) .
(f) If the graph of f is reflected about the y  axis, its equation becomes y= f ( x) .
(g) If the graph of f is stretched vertically by a factor of 3 , its equation becomes y=3 f (x) .
1
(h) If the graph of f is shrunk vertically by a factor of 3 , its equation becomes y= f (x) .
3

2. (a) To obtain the graph of y=5 f (x) from the graph of y= f (x) , stretch the graph vertically by a
factor of 5 .
(b) To obtain the graph of y= f (x 5) from the graph of y= f (x) , shift the graph 5 units to the right.
(c) To obtain the graph of y= f (x) from the graph of y= f (x) , reflect the graph about the x  axis.
(d) To obtain the graph of y= 5 f (x) from the graph of y= f (x) , stretch the graph vertically by a factor
of 5 and reflect it about the x  axis.
(e) To obtain the graph of y= f (5x) from the graph of y= f (x) , shrink the graph horizontally by a
factor of 5 .
(f) To obtain the graph of y=5 f (x) 3 from the graph of y= f (x) , stretch the graph vertically by a
factor of 5 and shift it 3 units downward.

3. (a) (graph 3) The graph of f is shifted 4 units to the right and has equation y= f (x 4) .
(b) (graph 1) The graph of f is shifted 3 units upward and has equation y= f (x)+3 .
1
(c) (graph 4) The graph of f is shrunk vertically by a factor of 3 and has equation y= f (x) .
3
(d) (graph 5) The graph of f is shifted 4 units to the left and reflected about the x  axis. Its equation
is y= f (x+4) .
(e) (graph 2) The graph of f is shifted 6 units to the left and stretched vertically by a factor of 2 . Its
equation is y=2 f (x+6) .

4. (a) To graph y= f (x+4) we shift the graph of f , 4 units to the left.

The point ( 2,1 ) on the graph of f corresponds to the point

( 2 4,1 ) = (  2,1 ) .
(b) To graph y= f (x)+4 we shift the graph of f , 4 units upward.

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

The point ( 2,1 ) on the graph of f corresponds to the point ( 2,1+4 ) = ( 2,5) .
(c) To graph y=2 f (x) we stretch the graph of f vertically by a factor of 2 .

The point ( 2,1 ) on the graph of f corresponds to the point ( 2,2 1 ) = ( 2,2 ) .
1
(d) To graph y= f (x)+3 , we shrink the graph of f vertically by a factor of 2 , then reflect the
2
resulting graph about the x  axis, then shift the resulting graph 3 units upward.

1
The point ( 2,1 ) on the graph of f corresponds to the point 2,  1+3 = ( 2,2.5) .
2

5. (a) To graph y= f (2x) we shrink the graph of f horizontally by a factor of 2 .

1
The point ( 4, 1 ) on the graph of f corresponds to the point  4, 1 = ( 2, 1 ) .
2
1
(b) To graph y= f x we stretch the graph of f horizontally by a factor of 2 .
2

The point ( 4, 1 ) on the graph of f corresponds to the point ( 2 4, 1 ) = ( 8, 1 ) .
(c) To graph y= f ( x) we reflect the graph of f about the y  axis.

2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

The point ( 4, 1 ) on the graph of f corresponds to the point (  1 4, 1 ) = (  4, 1 ) .
(d) To graph y= f ( x) we reflect the graph of f about the y  axis, then about the x  axis.

The point ( 4, 1 ) on the graph of f corresponds to the point (  1 4, 1  1 ) = (  4,1 ) .

2
6. The graph of y= f (x)= 3x x has been shifted 2 units to the right and stretched vertically by a
factor of 2 . Thus, a function describing the graph is

y = 2 f (x 2)=2 3(x 2) (x 2)2

= 2 3x 6 (x2 4x+4) =2  x2+7x 10

2
7. The graph of y= f (x)= 3x x has been shifted 4 units to the left, reflected about the x  axis, and
shifted downward 1 unit. Thus, a function describing the graph is

 1  f x+4  1

y= reflect shift shift

about 4units 1unit
x axis left down

This function can be written as

y = f (x+4) 1= 3(x+4) (x+4)  1=

2
3x+12 ( x2+8x+16)  1
2
=  x  5x 4  1

8. (a) The graph of y=2sin x can be obtained from the graph of y=sin x by stretching it vertically by
a factor of 2 .

3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

(b) The graph of y=1+ x can be obtained from the graph of y= x by shifting it upward 1 unit.

3 3
9. y= x : Start with the graph of y=x and reflect about the x  axis. Note: Reflecting about the y 
3 3
axis gives the same result since substituting  x for x gives us y=( x) = x .

2 2 2
10. y=1 x = x +1 : Start with the graph of y=x , reflect about the x  axis, and then shift 1 unit
upward.

2 2
11. y=(x+1) : Start with the graph of y=x and shift 1 unit to the left.

12.
4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

2 2 2 2
y=x  4x+3=(x  4x+4) 1=(x 2)  1 : Start with the graph of y=x , shift 2 units to the right, and then
shift 1 unit downward.

13. y=1+2cos x : Start with the graph of y=cos x , stretch vertically by a factor of 2 , and then shift 1
unit upward.

14. y=4sin 3x : Start with the graph of y=sin x , compress horizontally by a factor of 3 , and then
stretch vertically by a factor of 4 .

15. y=sin (x/2) : Start with the graph of y=sin x and stretch horizontally by a factor of 2 .

16. y=1/(x 4) : Start with the graph of y=1/x and shift 4 units to the right.

5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

17. y= x+3 : Start with the graph of y= x and shift 3 units to the left.

4 4
18. y=(x+2) +3 : Start with the graph of y=x , shift 2 units to the left, and then shift 3 units upward.

1 2 1 2 1 2 2
19. y= (x +8x)= (x +8x+16) 8= (x+4)  8 : Start with the graph of y=x , compress vertically by
2 2 2
a factor of 2 , shift 4 units to the left, and then shift 8 units downward.

3 3
20. y=1+ x 1 : Start with the graph of y= x , shift 1 unit to the right, and then shift 1 unit upward.

21. y=2/(x+1) : Start with the graph of y=1/x , shift 1 unit to the left, and then stretch vertically by a
factor of 2.

6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

1  
22. y= tan (x ) : Start with the graph of y=tan x , shift units to the right, and then compress
4 4 4
vertically by a factor of 4 .

23. y= sin x : Start with the graph of y=sin x and reflect all the parts of the graph below the x  axis
about the x  axis.

2 2 2 2
24. y= x  2x = x  2x+1 1 = (x 1)  1 : Start with the graph of y=x , shift 1 unit right, shift 1 unit
downward, and reflect the portion of the graph below the x  axis about the x  axis.


25. This is just like the solution to Example 4 except the amplitude of the curve (the 30 N curve in
2
Figure 9 on June 21) is 14 12=2 . So the function is L(t)=12+2sin (t 80) . March 31 is the
365
90 th day of the year, so the model gives L(90) 12.34 h. The daylight time (5:51 A.M. to 6:18 P.M.)
is 12 hours and 27 minutes, or 12.45 h. The model value differs from the actual value by
12.45 12.34
 0.009 , less than 1% .
12.45

26. Using a sine function to model the brightness of Delta Cephei as a function of time, we take its
period to be 5.4 days, its amplitude to be 0.35 (on the scale of magnitude), and its average magnitude
to be 4.0 . If we take t=0 at a time of average brightness, then the magnitude (brightness) as a
2
function of time t in days can be modeled by the formula M(t)=4.0+0.35sin t .
5.4

7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

27. (a) To obtain y= f ( x ) , the portion of the graph of y= f (x) to the right of the y  axis is reflected
about the y  axis.
(b) y=sin x

(c) y= x

28. The most important features of the given graph are the x  intercepts and the maximum and
minimum points. The graph of y=1 / f (x) has vertical asymptotes at the x  values where there are x 
intercepts on the graph of y= f (x) . The maximum of 1 on the graph of y= f (x) corresponds to a
minimum of 1/1=1 on y=1 / f (x) . Similarly, the minimum on the graph of y= f (x) corresponds to a
maximum on the graph of y=1 / f (x) . As the values of y get large (positively or negatively) on the
graph of y= f (x) , the values of y get close to zero on the graph of y=1 / f (x) .

29. Assuming that successive horizontal and vertical gridlines are a unit apart, we can make a table of
approximate values as follows.

x 0 1 2 3 4 5 6
f (x) 2 1.7 1.3 1.0 0.7 0.3 0
g(x) 2 2.7 3 2.8 2.4 1.7 0
g(x)+g(x) 4 4.4 4.3 3.8 3.1 2.0 0

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

Connecting the points ( x,f(x)+g(x) ) with a smooth curve gives an approximation to the graph of f +g .
Extra points can be plotted between those listed above if necessary.

30. First note that the domain of f +g is the intersection of the domains of f and g ; that is, f +g is
only defined where both f and g are defined. Taking the horizontal and vertical units of length to be
the distances between successive vertical and horizontal gridlines, we can make a table of
approximate values as follows:
x  2  1 0 1 2 2.5 3
f (x)  1 2.2 2.0 2.4 2.7 2.7 2.3
g(x) 1  1.3  1.2  0.6 0.3 0.5 0.7
f (x)+g(x) 0 0.9 0.8 1.8 3.0 3.2 3.0

Extra values of x (like the value 2.5 in the table above) can be added as needed.

3 2 2
31. f (x)=x +2x ; g(x)=3x  1 . D=R for both f and g .
3 2 2 3 2
( f +g)(x)=(x +2x )+(3x  1)=x +5x  1 , D=R .
3 2 2 3 2
( f  g)(x)=(x +2x ) (3x  1)=x  x +1 , D=R .
3 2 2 5 4 3 2
( fg)(x)=(x +2x )(3x  1)=3x +6x  x  2x , D=R .
3 2
f
g
(x)=
x +2x
2
3x  1
, D=
{ x| x 
1
3 } 2
since 3x  1 0 .

32. f (x)= 1+x , D=[ 1, ) ; g(x)= 1 x , D=( ,1] .

( f +g) (x)= 1+x + 1 x , D= (  ,1]
[ 1, ) =  1,1 .
( f  g) (x)= 1+x  1 x , D=  1,1 .
2
( fg) (x)= 1+x 1 x = 1 x , D=  1,1 .
9
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

f 1+x f
(x)= , D=[ 1,1) . We must exclude x=1 since it would make undefined.
g 1 x g

33. f (x)=x , g(x)=1/x

3 2
34. f (x)=x , g(x)= x

2
35. f (x)=2x  x ; g(x)=3x+2 . D=R for both f and g , and hence for their composites.
2 2 2
( f  g)(x)= f (g(x))= f (3x+2)=2(3x+2)  (3x+2)=2(9x +12x+4) 3x 2=18x +21x+6 .
2 2 2
(g f )(x)=g( f (x))=g(2x  x)=3(2x  x)+2=6x  3x+2 .
2 2 2 2 4 3 2 2 4 3
( f  f )(x)= f ( f (x))= f (2x  x)=2(2x  x)  (2x  x)=2(4x  4x +x ) 2x +x=8x  8x +x .
(g g)(x)=g(g(x))=g(3x+2)=3(3x+2)+2=9x+6+2=9x+8 .

3
36. f (x)=1 x , D=R ; g(x)=1/x , D={x| x 0} .

3 3
( f  g) (x)= f (g(x))= f ( 1/x ) =1 (1/x) =1 1/x , D={x| x 0} .

( g ( 3
) 3
f ) (x)=g( f (x))=g 1 x =1/(1 x ) , D={x|1 x  0}={x| x 1}.
3

3 33 9 6 3
( f f ) (x)= f ( f (x))= f (1 x )=1 (1 x ) [ =x  3x +3x ] , D=R .
( g g ) (x)=g(g(x))=g ( 1/x ) =1/ ( 1/x ) =x , D={x| x 0} because 0 is not in the domain of g .

37. f (x)=sin x , D=R ; g(x)=1 x , D=[0, ) .

( f  g) (x)= f (g(x))= f (1 x )=sin ( 1 x ) , D=[0, ] .
( g f ) (x)=g( f (x))=g ( sin x ) =1 sin x . For sin x to be defined, we must have sin x 0
x 0,  2 ,3   2 ,   4 ,5   4 , 3  ... , so
D= { x| x 2n , +2n ,where n is an integer} .

10
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

( f  f ) (x)= f ( f (x))= f ( sin x ) =sin ( sin x ) , D=R .

( g g) (x)=g(g(x))=g ( 1 x ) =1 1 x ,
D={ x 0|1 x 0} ={ x 0|1 x } ={ x 0| x 1} =[0,1] .

2
38. f (x)=1 3x , D=R ; g(x)=5x +3x+2 , D=R .
( f  g) (x) = f ( g(x) ) = f (5x2+3x+2)=1 3(5x2+3x+2)
2 2
=1 15x  9x 6= 15x  9x 5 , D=R .
( g f ) (x) =g ( f (x) ) =g ( 1 3x ) =5(1 3x)2+3(1 3x)+2=5(1 6x+9x2)+3 9x+2
2 2
=5 30x+45x  9x+5=45x  39x+10 , D=R .
( f  f ) (x)= f ( f (x) ) = f ( 1 3x ) =1 3(1 3x)=1 3+9x=9x 2, D=R .
( g g) (x) =g ( g(x) ) =g(5x2+3x+2)=5(5x2+3x+2)2+3(5x2+3x+2)+2
4 3 2 2
=5(25x +30x +29x +12x+4)+15x +9x+6+2
4 3 2 2
=125x +150x +145x +60x+20+15x +9x+8
4 3 2
=125x +150x +160x +69x+28, D=R .

1 x+1
39. f (x)=x+ , D= { x| x 0} ; g(x)= , D= { x| x  2} .
x x+2
x+1 x+1 1 x+1 x+2
= f (g(x))= f = + = +
( f  g)(x) x+2 x+2 x+1 x+2 x+1
x+2

=
(x+1)(x+1)+(x+2)(x+2)
=
x +2x+1 + x +4x+4(=
2
2x +6x+5 ) ( 2
) 2

(x+2)(x+1) (x+2)(x+1) (x+2)(x+1)

Since g(x) is not defined for x= 2 and f (g(x)) is not defined for x= 2 and x= 1 , the domain of (
f  g)(x) is D= { x| x  2, 1} .
2
1 x +1+x
x+ +1 2 2
1 x x x +x+1 x +x+1
(g f )(x)=g( f (x))=g x+ = = = = .
x 1 2 2 2
x +1+2x x +2x+1 (x+1)
x+ +2
x
x
Since f (x) is not defined for x=0 and g( f (x)) is not defined for x= 1 , the domain of (g f )(x) is
D= { x| x  1,0} .

11
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

1 1 1 1 1 1 x
= f ( f (x))= f x+ = x+ + =x+ + =x+ +
( f  f )(x) x x 1 x 2 x 2
x+ x +1 x +1
x
x

=
( 2
) (
x(x) x +1 +1 x +1 +x(x)
2
) =
4 2
x +x +x +1+x
2 2

2 2
x(x +1) x(x +1)
4 2
x +3x +1
= 2
, D= { x| x 0} .
x(x +1)

x+1 x+1+1(x+2)
+1
x+1 x+2 x+2 x+1+x+2 2x+3
(g g)(x)=g(g(x))=g = = = = . Since g(x) is not
x+2 x+1 x+1+2(x+2) x+1+2x+4 3x+5
+2
x+2 x+2
5
defined for x= 2 and g(g(x)) is not defined for x= , the domain of ( g g ) (x) is
3

{
D= x| x  2,
5
3
. }
40. f (x)= 2x+3 , D= { x| x 
3
2 } 2
; g(x)=x +1 , D=R .
2 2 2
( f  g)(x)= f (x +1)= 2(x +1)+3 = 2x +5 , D=R .
(g f )(x)=g ( 2x+3 ) = ( 2x+3 ) +1=(2x+3)+1=2x+4 , D=
2
{ x| x 
3
2 } .

( f  f )(x)= f (
2
2x+3 ) = 2 ( 2x+3 ) +3 = 2 2x+3 +3 , D=
2 2 4 2 4 2
{ x| x 
3
2 } .

(g g)(x)=g(x +1)=(x +1) +1=(x +2x +1)+1=x +2x +2 , D=R .

41.
( f  g h)(x) = f (g(h(x)))= f (g(x 1))= f (2(x 1))
=2(x 1)+1=2x 1

42.
( f  g h)(x) = f (g(h(x)))= f (g(1 x))= f ((1 x)2)
2 2
=2(1 x)  1=2x  4x+1

43.
12
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

( f  g h)(x) = f (g(h(x)))= f (g(x+3))= f ((x+3)2+2)

2
= f (x +6x+11)= ( x2+6x+11)  1 = 2
x +6x+10

2
44. ( f  g h)(x)= f (g(h(x)))= f ( g ( x+3 ) ) = f ( cos x+3 ) =
cos x+3 +1

2 10 2 10
45. Let g(x)=x +1 and f (x)=x . Then ( f  g)(x)= f (g(x))=(x +1) =F(x) .

46. Let g(x)= x and f (x)=sin x . Then ( f  g)(x)= f (g(x))=sin ( x ) =F(x) .

2
2 x x
47. Let g(x)=x and f (x)= . Then ( f  g)(x)= f (g(x))= 2 =G(x) .
x+4
x +4

48. Let g(x)=x+3 and f (x)=1/x . Then ( f  g)(x)= f (g(x))=1/(x+3)=G(x) .

49. Let g(t)=cos t and f (t)= t . Then ( f  g)(t)= f (g(t))= cos t =u(t) .

t tan t
50. Let g(t)=tan t and f (t)= . Then ( f  g)(t)= f (g(t))= =u(t) .
1+t 1+tan t
2
2 x x
51. Let h(x)=x , g(x)=3 , and f (x)=1 x . Then ( f  g h)(x)=1 3 =H(x) .

3 3
52. Let h(x)= x , g(x)=x 1 , and f (x)= x . Then ( f  g h)(x)= x  1 =H(x) .

4 4 4
53. Let h(x)= x , g(x)=sec x , and f (x)=x . Then ( f  g h)(x)= ( sec x ) =sec ( x ) =H(x) .

54. (a) f (g(1))= f (6)=5

(b) g( f (1))=g(3)=2
(c) f ( f (1))= f (3)=4
(d) g(g(1))=g(6)=3
(e) (g f )(3)=g( f (3))=g(4)=1
(f) ( f  g)(6)= f (g(6))= f (3)=4

55. (a) g(2)=5 , because the point ( 2,5) is on the graph of g . Thus, f (g(2))= f (5)=4 , because the
point ( 5,4 ) is on the graph of f .
(b) g( f (0))=g(0)=3

13
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

(c) ( f  g)(0)= f (g(0))= f (3)=0

(d) (g f )(6)=g( f (6))=g(6) . This value is not defined, because there is no point on the graph of g that
has x  coordinate 6 .
(e) (g g)( 2)=g(g( 2))=g(1)=4
(f) ( f  f )(4)= f ( f (4))= f (2)= 2

56. To find a particular value of f (g(x)) , say for x=0 , we note from the graph that g(0) 2.8 and
f (2.8)  0.5 . Thus, f (g ( 0 ) ) f (2.8)  0.5 . The other values listed in the table were obtained in a
similar fashion.
x g(x) f (g(x))
x g(x) f (g(x))
0 2.8  0.5
 5  0.2  4
1 2.2  1.7
 4 1.2  3.3
2 1.2  3.3
 3 2.2  1.7
3  0.2  4
 2 2.8  0.5
4  1.9  2.2
 1 3  0.2
5  4.1 1.9

57. (a) Using the relationship distance = rate  time with the radius r as the distance, we have
r(t)=60t .
2 2 2
(b) A= r  (A r)(t)=A(r(t))= (60t) =3600 t . This formula gives us the extent of the rippled area
2
(in cm ) at any time t .

58. (a) d=rt d(t)=350t

(b) There is a Pythagorean relationship involving the legs with lengths d and 1 and the hypotenuse
2 2 2 2
with length s : d +1 =s . Thus, s(d)= d +1 .
2
(c) (s d)(t)=s(d(t))=s(350t)= (350t) +1

59. (a)

14
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

H(t)=
(b)
{ 0 if t<0
1 if t 0

V (t)= { 0 if t<0
120 if t 0
so V (t)=120H(t) .

(c)
Starting with the formula in part (b), we replace 120 with 240 to reflect the different voltage. Also,
because we are starting 5 units to the right of t=0 , we replace t with t 5 Thus, the formula is
V (t)=240H(t 5) .

60. (a)
R(t) =tH(t)

= { 0 if t<0
t if t 0

(b) V (t)= {
0 if t<0
2t if 0 t 60
so V (t)=2tH(t) , t 60 .

(c) V (t)= { 0
4 ( t 7)
if t<7
if 7 t 32
so V (t)=4(t 7)H(t 7) , t 32 .

61. (a) By examining the variable terms in g and h , we deduce that we must square g to get the terms
2 2 2 2
4x and 4x in h . If we let f (x)=x +c , then ( f  g)(x)= f (g(x))= f (2x+1)=(2x+1) +c=4x +4x+ ( 1+c ) .
15
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.3 New Functions from Old Functions

2 2
Since h(x)=4x +4x+7 , we must have 1+c=7 . So c=6 and f (x)=x +6 .
(b) We need a function g so that f (g(x))=3(g(x))+5=h(x) . But
2 2 2 2
h(x)=3x +3x+2=3(x +x)+2=3(x +x 1)+5 , so we see that g(x)=x +x 1 .

62. We need a function g so that g( f (x))=g(x+4)=h(x)=4x 1=4(x+4) 17 . So we see that the function
g must be g(x)=4x 17 .

63. We need to examine h( x) .

h( x)=( f  g)( x)= f (g( x))= f (g(x)) [because g is even] =h(x)

Because h( x)=h(x) , h is an even function.

64. h( x)= f (g( x))= f ( g(x)) . At this point, we can’t simplify the expression, so we might try to find
2
a counterexample to show that h is not an odd function. Let g(x)=x , an odd function, and f (x)=x +x .
2
Then h(x)=x +x, which is neither even nor odd.
Now suppose f is an odd function. Then f ( g(x))= f (g(x))= h(x) . Hence, h( x)= h(x) , and so h is
odd if both f and g are odd.
Now suppose f is an even function. Then f ( g(x))= f (g(x))=h(x) . Hence, h( x)=h(x) , and so h is
even if g is odd and f is even.

16
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

4
1. f (x)=x +2
(a)  2,2 by  2,2

(b) 0,4 by 0,4

(c)  4,4 by  4,4

(d)  8,8 by  4,40

(e)  40,40 by  80,800

The most appropriate graph is produced in viewing rectangle (d).

2
2. f (x)=x +7x+6
(a)  5,5 by  5,5

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

(b) 0,10 by  20,100

(c)  15,8 by  20,100

(d)  10,3 by  100,20

The most appropriate graph is produced in viewing rectangle (c).

3
3. f (x)=10+25x x
(a)  4,4 by  4,4

(b)  10,10 by  10,10

2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

(c)  20,20 by  100,100

(d)  100,100 by  200,200

The most appropriate graph is produced in viewing rectangle (c) because the maximum and minimum
points are fairly easy to see and estimate.

2
4. f (x)= 8x x
(a)  4,4 by  4,4

(b)  5,5 by 0,100

3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

(c)  10,10 by  10,40

(d)  2,10 by  2,6

The most appropriate graph is produced in viewing rectangle (d).

2
5. Since the graph of f (x)=5+20x x is a parabola opening downward, an appropriate viewing
rectangle should include the maximum point.

3 2
6. An appropriate viewing rectangle for f (x)=x +30x +200x should include the high and low points.

3 2
7. f (x)=0.01x  x +5 . Graphing f in a standard viewing rectangle,  10,10 by  10,10 , shows us
what appears to be a parabola. But since this is a cubic polynomial, we know that a larger viewing
rectangle will reveal a minimum point as well as the maximum point. After some trial and error, we
choose the viewing rectangle  50,150 by  2000,2000 .

4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

8. f (x)=x(x+6)(x 9)

4 4
9. f (x)= 81 x is defined when
4 4 4 4 4
81 x  0 x  81 x  3 , so the domain of f is  3,3 . Also 0 81 x  81 =3 , so the range
is 0,3 .

10. f (x)= 0.1x+20 is defined when 0.1x+20 0 x  200 , so the domain of f is  200, ).

2
11. The graph of f (x)=x +(100/x) has a vertical asymptote of x=0 . As you zoom out, the graph of f
2
looks more and more like that of y=x .

5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

2
12. The graph of f (x)=x/(x +100) is symmetric with respect to the origin.

13. f (x)=cos ( 100x )

14. f (x)=3sin (120x)

15. f (x)=sin ( x/40 )

16. f (x)=tan ( 25x )

17.
6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

2
cos (x )
y=3

2
18. y=x +0.02sin (50x)

19. We must solve the given equation for y to obtain equations for the upper and lower halves of the
ellipse.
2 2
2 2 2 2 2 1 4x 1 4x
4x +2y =1 2y =1 4x  y=  y=
2 2

7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

2 2 2 2 2
20. y  9x =1  y =1+9x  y= 1+9x

3 2
21. From the graph of f (x)=x  9x  4 , we see that there is one solution of the equation f (x)=0 and it
is slightly larger than 9. By zooming in or using a root or zero feature, we obtain x 9.05.

3
22. We see that the graphs of f (x)=x and g(x)=4x 1 intersect three times. The x  coordinates of
these points (which are the solutions of the equation) are approximately  2.11,0.25, and 1.86 .
3
Alternatively, we could find these values by finding the zeros of h(x)=x  4x+1 .

2
23. We see that the graphs of f (x)=x and g(x)=sin x intersect twice. One solution is x=0. The other
solution of f =g is the x  coordinate of the point of intersection in the first quadrant. Using an
intersect feature or zooming in, we find this value to be approximately 0.88. Alternatively, we could
2
find that value by finding the positive zero of h(x)=x  sin x .

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

Note : After producing the graph on a TI 83 Plus, we can find the approximate value 0.88 by using
the following keystrokes:
. The ‘‘1’’ is just a guess for 0.88.

24. (a)
The x  coordinates of the three points of intersection are x  3.29 ,  2.36 and 1.20 .
(b) Using trial and error, we find that m 0.3365 . Note that m could also be negative.

3 2
25. g(x)=x /10 is larger than f (x)=10x whenever x>100 .

4 3 3
26. f (x)=x  100x is larger than g(x)=x whenever x>101 .

27.

We see from the graphs of y= sin x x and y=0.1 that there are two solutions to the equation
9
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

sin x x =0.1 : x  0.85 and x 0.85 . The condition sin x x <0.1 holds for any x lying between
these two values.

28.
5 3 5
P(x)=3x  5x +2x , Q(x)=3x . These graphs are significantly different only in the region close to the
origin. The larger a viewing rectangle one chooses, the more similar the two graphs look.

4 6
29. (a) The root functions y= x , y= x and y= x

3 5
(b) The root functions y=x , y= x and y= x

3 4 5
(c) The root functions y= x , y= x , y= x and y= x

(d)

For any n , the n th root of 0 is 0 and the n th root of 1 is 1 ; that is, all n th root functions pass
through the points ( 0,0 ) and ( 1,1 ) .
For odd n , the domain of the n th root function is R , while for even n , it is { x
R| x 0} .
Graphs of even root functions look similar to that of x , while those of odd root functions
3
resemble that of x .
10
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

n
As n increases, the graph of x becomes steeper near 0 and flatter for x>1 .

3
30. (a) The functions y=1/x and y=1/x

2 4
(b) The functions y=1/x and y=1/x

2 3 4
(c) The functions y=1/x , y=1/x , y=1/x and y=1/x

(d)

n
The graphs of all functions of the form y=1/x pass through the point ( 1,1 ) .
n
If n is even, the graph of the function is entirely above the x axis. The graphs of 1/x
for n even are similar to one another.
If n is odd, the function is positive for positive x and negative for negative x . The
n
graphs of 1/x for n odd are similar to one another.
n
As n increases, the graphs of 1/x approach 0 faster as x  .

4 2
31. f (x)=x +cx +x . If c<0 , there are three humps: two minimum points and a
maximum point. These humps get flatter as c increases, until at c=0 two of the humps
disappear and there is only one minimum point. This single hump then moves to the
right and approaches the origin as c increases.

11
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

2
32. f (x)= 1+cx . If c<0 , the function is only defined on  1  c ,1  c , and / /
its graph is the top half of an ellipse. If c=0 , the graph is the line y=1 . If c>0 , the
graph is the top half of a hyperbola. As c approaches 0 , these curves become flatter
and approach the line y=1 .

n  x
33. y=x 2 . As n increases, the maximum of the function moves further from the
origin, and gets larger. Note, however, that regardless of n , the function approaches 0
as x  .

x
34. y= . The ‘‘bullet’’ becomes broader as c increases.
2
c x

2 3 2
35. y =cx +x
If c<0 , the loop is to the right of the origin, and if c is positive, it is to the left. In both
cases, the closer c is to 0 , the larger the loop is.
12
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

(In the limiting case, c=0 , the loop is ‘‘infinite’’, that is, it doesn’t close.) Also, the
larger c is, the steeper the slope is on the loopless side of the origin.

36. (a) y=sin ( x )

This function is not periodic; it oscillates less frequently as x increases.

(b) y=sin x ( 2)
This function oscillates more frequently as x increases. Note also that this function
is even, whereas sin x is odd.

37. The graphing window is 95 pixels wide and we want to start with x=0 and end
with x=2 . Since there are 94 ‘‘gaps’’ between pixels, the distance between pixels is
2  0 2
. Thus, the x  values that the calculator actually plots are x=0+  n , where
94 94
n=0 , 1 , 2 , ... , 93 , 94 . For y=sin 2x , the actual points plotted by the calculator are
2 2
 sin 2  n for n=0 , 1 , ... , 94. For y=sin 96x , the points plotted
94 94
2 2
are  sin 96  n for n=0 , 1 , ... , 94. But
94 94
2 2 2 2
sin 96  n =sin 94  n+2  n =sin 2 n+2  n
94 94 94 94

13
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.4 Graphing Calculators and Computers

2
=sin 2  n [ by periodicity of sine ], n=0 , 1 , ... , 94
94

So the y  values, and hence the points, plotted for y=sin 96x are identical to those
plotted for y=sin 2x .
Note: Try graphing y=sin 94x . Can you see why all the y  values are zero?

38. As in Exercise 37, we know that the points being plotted for y=sin 45x are
2 2
 sin 45  n for n=0 , 1 , ... , 94 . But
94 94
2 2 2 2
sin 45  n =sin 47  n 2  n =sin n  2  n
94 94 94 94
2 2
=sin ( n ) cos 2  n  cos ( n ) sin 2  n [ Subtraction
94 94
formula for the sine]
2 2 2
=0 cos 2  n  ( 1)sin 2  n = sin 2  n , n=0 ,
94 94 94
1 , ... , 94

So the y  values, and hence the points, plotted for y=sin 45x lie on either y=sin 2x or
y= sin 2x .

14
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

x
1. (a) f (x)=a , a>0
(b) R
(c) ( 0, )
(d) See Figures (c), (b), and (a), respectively.

2. (a) The number e is the value of a such that the slope of the tangent line at x=0 on the graph of
x
y=a is exactly 1 .
(b) e 2.71828
x
(c) f (x)=e

3. All of these graphs approach 0 as x   , all of them pass through the point ( 0,1 ) , and all of
them are increasing and approach  as x  . The larger the base, the faster the function increases
for x>0 , and the faster it approaches 0 as x   .

 x x  x
4. The graph of e is the reflection of the graph of e about the y axis, and the graph of 8 is the
x x x
reflection of that of 8 about the y  axis. The graph of 8 increases more quickly than that of e for
x>0 , and approaches 0 faster as x   .

x x
5. The functions with bases greater than 1 ( 3 and 10 ) are increasing, while those with bases less
1 x 1 x 1 x x
than 1 and are decreasing. The graph of is the reflection of that of 3
3 10 3
1 x x
about the y axis, and the graph of is the reflection of that of 10 about the y axis. The
10
x x
graph of 10 increases more quickly than that of 3 for x>0 , and approaches 0 faster as x   .

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

6. Each of the graphs approaches  as x   , and each approaches 0 as x  . The smaller the
base, the faster the function grows as x   , and the faster it approaches 0 as x  .

x
7. We start with the graph of y=4 (Figure 3) and then shift 3 units downward. This shift doesn’t
x
affect the domain, but the range of y=4  3 is (  3, ) . There is a horizontal asymptote of y= 3.

x x
y=4 y=4  3

x
8. We start with the graph of y=4 (Figure 3) and then shift 3 units to the right. There is a horizontal
asymptote of y=0 .

x x 3
y=4 y=4

2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

x
9. We start with the graph of y=2 (Figure 2), reflect it about the y  axis, and then about the x  axis
  x
(or just rotate 180 to handle both reflections) to obtain the graph of y= 2 . In each graph, y=0 is the
horizontal asymptote.

x  x  x
y=2 y=2 y= 2

x
10. We start with the graph of y=e (Figure 13), vertically stretch by a factor of 2, and then shift 1 unit
upward. There is a horizontal asymptote of y=1.

x x
y=2e y=1+2e

x
11. We start with the graph of y=e (Figure 13), reflect it about the x  axis, and then shift 3 units
upward. Note the horizontal asymptote of y=3 .

x
y= e

x
y=3 e

x
12. We start with the graph of y=e (Figure 13), reflect it about the y  axis, and then about the x  axis
3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

  x
(or just rotate 180 to handle both reflections) to obtain the graph of y= e . Now shift this graph 1
unit upward, vertically stretch by a factor of 5 , and then shift 2 units upward.

 x
y= e

 x
y=2+5(1 e )

x
13. (a) To find the equation of the graph that results from shifting the graph of y=e 2 units downward,
x
we subtract 2 from the original function to get y=e  2 .
x
(b) To find the equation of the graph that results from shifting the graph of y=e 2 units to the right,
( x 2 )
we replace x with x 2 in the original function to get y=e .
x
(c) To find the equation of the graph that results from reflecting the graph of y=e about the x  axis,
x
we multiply the original function by  1 to get y= e .
x
(d) To find the equation of the graph that results from reflecting the graph of y=e about the y  axis,
 x
we replace x with  x in the original function to get y=e .
x
(e) To find the equation of the graph that results from reflecting the graph of y=e about the x  axis
x
and then about the y  axis, we first multiply the original function by  1 (to get y= e ) and then
 x
replace x with  x in this equation to get y= e .

14. (a) This reflection consists of first reflecting the graph about the x  axis (giving the graph with
x x
equation y= e ) and then shifting this graph 2 4=8 units upward. So the equation is y= e +8 .
(b) This reflection consists of first reflecting the graph about the y  axis (giving the graph with
 x  ( x 4 )
equation y=e ) and then shifting this graph 2 2=4 units to the right. So the equation is y=e .

x x x
15. (a) The denominator 1+e is never equal to zero because e >0 , so the domain of f (x)=1/(1+e ) is
R.
4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

x x x
(b) 1 e =0 e =1 x=0 , so the domain of f (x)=1/(1 e ) is (  ,0)
(0, ).

 t
16. (a) The sine and exponential functions have domain R , so g(t)=sin (e ) also has domain R .
t t t
(b) The function g(t)= 1 2 has domain {t |1 2  0}={t |2  1}={t |t 0}=(  ,0] .

x 1 6 3 6 3
17. Use y=Ca with the points ( 1,6 ) and ( 3,24 ) . 6=Ca C= and 24=Ca 24= a
a a
2 6 x
4=a a=2 [ since a>0 ] and C= =3 . The function is f (x)=3 2 .
2

x x 2 2 2
18. Given the y  intercept ( 0,2 ) , we have y=Ca =2a . Using the point 2, gives us =2a
9 9
1 2 1 1 x  x
=a a= [ since a>0 ]. The function is f (x)=2 or f (x)=2(3) .
9 3 3

x
19. If f (x)=5 , then
x+h
f (x+h) f (x) 5  5 5 5  5 5 5  1
= = = =5
x
x x h x x
( h
) 5 1
h
.
h h h h h

28 1 27
20. Suppose the month is February. Your payment on the 28th day would be 2 =2 =134 , 217 ,
728 cents, or $1 , 342 , 177 . 28 . Clearly, the second method of payment results in a larger amount
for any month.

2 24 24
21. 2 ft =24 in, f (24)=24 in =576 in =48 ft. g(24)=2 in =2 /(12 5280) mi  265 mi

x 5
22. We see from the graphs that for x less than about 1.8 , g(x)=5 > f (x)=x , and then near the point
( 1.8,17.1 ) the curves intersect. Then f (x)>g(x) from x 1.8 until x=5 . At ( 5,3125) there is another
point of intersection, and for x>5 we see that g(x)> f (x) . In fact, g increases much more rapidly than
f beyond that point.

5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

23. The graph of g finally surpasses that of f at x 35.8 .

x x 9
24. We graph y=e and y=1 , 000 , 000 , 000 and determine where e =1 10 . This seems to be true at
x 20.723 , so
6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

x 9
e >1 10 for x>20.723 .

25. (a) Fifteen hours represents 5 doubling periods (one doubling period is three hours).
5
100 2 =3200
(b) In t hours, there will be t/3 doubling periods. The initial population is 100 , so the population y at
t/3
time t is y=100 2 .
20/3
(c) t=20 y=100 2  10 , 159
x/3
(d) We graph y =100 2 and y =50 , 000 . The two curves intersect at x 26.9 , so the population
1 2
reaches 50 , 000 in about 26.9 hours.

1 4 1
26. (a) Sixty hours represents 4 half life periods. 2 = g
2 8
(b) In t hours, there will be t/15 half life periods. The initial mass is 2 g, so the mass y at time t is
1 t/15
y=2 .
2
1 96/15
(c) 4 days =4 24=96 hours. t=96 y=2  0.024 g
2
(d) y=0.01 t 114.7 hours

27. An exponential model is

7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.5 Exponential Functions

t  12
y=ab , where a=3.154832569 10 and b=1.017764706 . This model gives y(1993) 5498 million
and y(2010) 7417 million.

t  9
28. An exponential model is y=ab , where a=1.9976760197589 10 and b=1.0129334321697 . This
model gives y(1925) 111 million, y(2010) 330 million, and y(2020) 375 million.

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

1. (a) See Definition 1.

(b) It must pass the Horizontal Line Test.

 1  1  1
2. (a) f (y)=x f (x)=y for any y in B . The domain of f is B and the range of f is A .
(b) See the steps in (5).
(c) Reflect the graph of f about the line y=x .

3. f is not one to one because 2 6 , but f (2)=2.0= f (6) .

4. f is one to one since for any two different domain values, there are different range values.

5. No horizontal line intersects the graph of f more than once. Thus, by the Horizontal Line Test, f is
one to one.

6. The horizontal line y=0 (the x  axis) intersects the graph of f in more than one point. Thus, by the
Horizontal Line Test, f is not one to one.

7. The horizontal line y=0 (the x  axis) intersects the graph of f in more than one point. Thus, by the
Horizontal Line Test, f is not one to one.

8. No horizontal line intersects the graph of f more than once. Thus, by the Horizontal Line Test, f is
one to one.

1 1
9. The graph of f (x)= (x+5) is a line with slope . It passes the Horizontal Line Test, so f is one
2 2
to one.
1 1
Algebraic solution : If x  x , then x +5 x +5 x +5 
( ) (x +5 
) f x 
( ) f ( x ) , so f is
1 2 1 2 2 1 2 2 1 2
one to one.

2 b 4
10. The graph of f (x)=1+4x x is a parabola with axis of symmetry x= = =2 . Pick any x
2a 2( 1)
 values equidistant from 2 to find two equal function values. For example, f (1)=4 and f (3)=4 , so f
is not 1  1 .

11. g(x)= x  g( 1)=1=g(1) , so g is not one to one.

12. x  x  x  x  g x  g x
( ) ( ) , so g is 1  1.
1 2 1 2 1 2

13. A football will attain every height h up to its maximum height twice: once on the way up, and

1
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

again on the way down. Thus, even if t does not equal t , f t

1 2 ( ) may equal f ( t ) , so f is not 1 
1 2
1
.

14. f is not 1  1 because eventually we all stop growing and therefore, there are two times at which
we have the same height.

15. f does not pass the Horizontal Line Test, so f is not 1  1 .

16. f passes the Horizontal Line Test,

so f is 1  1 .

 1
17. Since f (2)=9 and f is 1  1 , we know that f (9)=2 . Remember, if the point ( 2,9 ) is on the
 1
graph of f , then the point ( 9,2 ) is on the graph of f .

18. (a) First, we must determine x such that f (x)=3 . By inspection, we see that if x=0 , then f (x)=3 .
 1
Since f is 1  1 ( f is an increasing function), it has an inverse, and f (3)=0 .
 1
(b) By the second cancellation equation in (4), we have f f (5) =5 . ( )
19. First, we must determine x such that g(x)=4 . By inspection, we see that if x=0 , then g(x)=4 .
 1
Since g is 1  1 ( g is an increasing function), it has an inverse, and g (4)=0 .

20. (a) f is 1  1 because it passes the Horizontal Line Test.

 1  1
(b) Domain of f =  3,3 = Range of f . Range of f =  2,2 = Domain of f .
 1
(c) Since f ( 2)=1 , f (1)= 2 .

21. We solve
2
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

5 9 9
C= (F 32) for F : C=F 32 F= C+32 . This gives us a formula for the inverse function, that
9 5 5
is, the Fahrenheit temperature F as a function of the Celsius temperature C . F  459.67
9 9
C+32  459.67 C  491.67 C  273.15 , the domain of the inverse function.
5 5

2 2 2 2
m 2 m 2 m m m
0 v 0 v 0 2 2 0 0
22. m=  1 =  =1  v =c 1  v=c 1 . This
2 2 2 2 2 2 2 2
1 v /c c m c m m m
 1
formula gives us the speed v of the particle in terms of its mass m , that is, v= f (m) .

2 2 1 2 10
23. f (x)= 10 3x  y= 10 3x (y 0) y =10 3x 3x=10 y  x= y+ . Interchange x and y
3 3
1 2 10  1 1 2 10  1
: y= x+ . So f (x)= x+ . Note that the domain of f is x 0 .
3 3 3 3

4x 1 4x 1
24. f (x)=  y=  y(2x+3)=4x 1  2xy+3y=4x 1  3y+1=4x 2xy  3y+1=(4 2y)x
2x+3 2x+3
3y+1 3x+1
x= . Interchange x and y : y= .
4 2y 4 2x
 1 3x+1
So f (x)= .
4 2x
3 3
x x 3 3 3
25. f (x)=e  y=e  ln y=x  x= ln y . Interchange x and y : y= ln x .
 1 3
So f (x)= ln x .

3 3 y 3 3 3 y 3
26. y= f (x)=2x +3 y 3=2x  =x  x= .
2 2
3 x 3  1 3 x 3
Interchange x and y : y= . So f (x)= .
2 2

y y x  1 x
27. y=ln ( x+3)  x+3=e  x=e  3 . Interchange x and y : y=e  3 . So f (x)=e  3 .

x
1+e x x x x x
28. y=  y ye =1+e  y 1=ye +e  y 1=e ( y+1 ) 
x
1 e

3
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

x y 1 y 1 x 1  1 x 1
e=  x=ln . Interchange x and y : y=ln . So f (x)=ln .
y+1 y+1 x+1 x+1
 1
Note that the domain of f is x >1 .

2 2 2 2 2 2
29. y= f (x)=1  1 y=  x=  x= , since x>0 . Interchange x and y : y=
2 2 1 y 1 y 1 x
x x
 1 2
. So f (x)= .
1 x

2 2 2 2 2
30. y= f (x)= x +2x , x>0 y>0 and y =x +2x x +2x y =0 . Now we use the quadratic formula:

x=
 2
2
( 2)
2  4 1  y
= 1
2
1+y . But x>0 , so the negative root is inadmissible. Interchange x
2 1
2  1 2
and y : y= 1+ 1+x . So f (x)= 1+ 1+x , x>0 .

31. The function f is one to one, so its inverse exists and the graph of its inverse can be obtained by
reflecting the graph of f about the line y=x .

4
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

32. The function f is one to one, so its inverse exists and the graph of its inverse can be obtained by
reflecting the graph of f about the line y=x . For the graph of 1/ f , the y  coordinates are simply the
1
reciprocals of f . For example, if f (5)=9, then 1/ f (5)= . If we draw the horizontal line y=1 , we see
9
that the only place where the graphs intersect is on that line.

y
33. (a) It is defined as the inverse of the exponential function with base a , that is, log x=y a =x .
a
(b) ( 0,
)
(c) 
(d) See Figure .

34. (a) The natural logarithm is the logarithm with base e , denoted ln x .
(b) The common logarithm is the logarithm with base 10 , denoted log x .
(c) See Figure .

6
35. (a) log 64=6 since 2 =64 .
2
1  2 1
(b) log = 2 since 6 = .
6 36 36

1 1/3
36. (a) log 2= since 8 =2 .
8 3
2
(b) ln e = 2

2
37. (a) log 1.25+log 80=log
10 10 10
( 1.25 80 ) =log 10100=log 1010 =2
3 200 2
(b) log 10+log 20 3log 2=log ( 10 20 )  log 2 =log =log 25=log 5 =2
5 5 5 5 5 5 8 5 5

5
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

38. (a) 2
( log 23+log 25) =2log 215=15 [ Or: 2 ( log 23+log 25) =2log 23 2log 25=3 5=15 ]
3
(b) e
3ln 2
=e
ln (2 )
=e
ln 8
=8 [ Or: e
3ln 2
= e ( ln 2) 3=23=8 ]
2 16
39. 2ln 4 ln 2=ln 4  ln 2=ln 16 ln 2=ln =ln 8
2

a b a b a b
40. ln x+aln y bln z=ln x+ln y  ln z =ln (x y ) ln z =ln (xy /z )
2
1 2 2 1/2 2 (1+x ) x
41. ln (1+x )+ ln x ln sin x=ln (1+x )+ln x  ln sin x=ln [(1+x ) x ] ln sin x=ln
2 sin x

ln 10
42. (a) log 10=  0.926628
12 ln 12
ln 8.4
(b) log 8.4=  3.070389
2 ln 2

ln x ln x
43. To graph these functions, we use log x= and log x= . These graphs all approach
1.5 ln 1.5 50 ln 50
+

as x 0 , and they all pass through the point ( 1,0 ) . Also, they are all increasing, and all
approach
as x
. The functions with larger bases increase extremely slowly, and the ones with
smaller bases do so somewhat more quickly. The functions with large bases approach the y axis
+
more closely as x 0 .

x
44. We see that the graph of ln x is the reflection of the graph of e about the line y=x , and that the
x x
graph of log x is the reflection of the graph of 10 about the same line. The graph of 10 increases
10
x
more quickly than that of e . Also note that log x
as x
more slowly than ln x .
10

6
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

36
45. 3 ft =36 in, so we need x such that log x=36 x=2 =68 , 719 , 476 , 736 . In miles, this is
2
1ft 1mi
68 , 719 , 476 , 736 in  1 , 084 , 587.7 mi.
12in 5280ft

46.

0.1
From the graphs, we see that f (x)=x >g(x)=ln x for approximately 0<x<3.06 , and then g(x)> f (x)
15
for 3.06<x<3.43 10 (approximately). At that point, the graph of f finally surpasses the graph of g
for good.

47. (a) Shift the graph of y=log x five units to the left to obtain the graph of y=log (x+5) . Note the
10 10
vertical asymptote of x= 5 .
y=log x
10

7
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

y=log (x+5)
10

(b) Reflect the graph of y=ln x about the x  axis to obtain the graph of y= ln x .
y=ln x

y= ln x

48. (a) Reflect the graph of y=ln x about the y  axis to obtain the graph of y=ln (  x ) .
y=ln x

y=ln ( x)

8
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

(b) Reflect the portion of the graph of y=ln x to

the right of the y  axis about the y  axis. The
graph of y=ln x is that reflection in addition
to the original portion.
y=ln x

y=ln x

1 1/2
49. (a) 2ln x=1  ln x=  x=e = e
2
 x
(b) e =5   x=ln 5  x= ln 5

2x+3 2x+3 1
50. (a) e  7=0  e =7  2x+3=ln 7  2x=ln 7 3  x= (ln 7 3)
2
 3  3 1  3
(b) ln (5 2x)= 3  5 2x=e  2x=5 e  x= (5 e )
2

x 5
51. (a) 2 =3 log 3=x 5 x=5+log 3 .
2 2
ln 3 ln 3
Or: 2
x 5
=3 ln 2( ) =ln 3
x 5
(x 5)ln 2=ln 3 x 5=
ln 2
 x=5+
ln 2
1 2
(b) ln x+ln (x 1)=ln (x(x 1))=1 x(x 1)=e  x  x e=0 . The quadratic formula (with a=1 , b= 1 ,
1
and c= e ) gives x= ( 1 1+4e ) , but we reject the negative root since the natural logarithm is not
2
1
defined for x<0 . So x= ( 1+ 1+4e ) .
2
9
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

ln (ln x) 1 1 ln x e e
52. (a) ln (ln x)=1 e =e  ln x=e =e e =e  x=e
ax bx ax bx bx
(b) e =Ce  ln e =ln [C(e )] ax=ln C+bx+ln e  ax=ln C+bx ax bx=ln C  (a b)x=ln C
ln C
x=
a b

x x
53. (a) e <10  ln e <ln 10  x<ln 10  x (
,ln 10)
ln x  1  1
(b) ln x> 1  e >e  x>e  x ( 1/e,
)

54. (a) 2<ln x<9 

2
e <e
ln x
<e 
9 2
e <x<e 
9
x ( e2,e9)
2 3x 2 3x 1 1
(b) e >4  ln e >ln 4  2 3x>ln 4   3x>ln 4 2  x< (ln 4 2)  x 
, (2 ln 4)
3 3

2x 2x 2x 1
55. (a) For f (x)= 3 e , we must have 3 e  0 e  3 2x ln 3  x ln 3 .
2
1
Thus, the domain of f is (
, ln 3] .
2
2x 2 2x 2x 2 2 1 2
(b) y= f (x)= 3 e [ note that y 0 ]  y =3 e  e =3 y  2x=ln (3 y ) x= ln (3 y ) .
2
1 2  1 1 2  1
Interchange x and y : y= ln (3 x ) . So f (x)= ln (3 x ) . For the domain of f , we must have
2 2
2 2  1
3 x >0 x <3 |x|< 3   3 <x< 3  0 x< 3 since x 0 . Note that the domain of f , [0, 3 ) ,
equals the range of f .

 2
56. (a) For f (x)=ln (2+ln x) , we must have 2+ln x>0 ln x> 2 x>e . Thus, the domain of f is
 2
(e ,
).
y x
y y e 2 e 2
(b) y= f (x)=ln (2+ln x)  e =2+ln x  ln x=e  2  x=e . Interchange x and y : y=e . So
x
 1 e 2  1
f (x)=e . The domain of f , as well as the range of f , is  .

3 2 3 2
57. We see that the graph of y= f (x)= x +x +x+1 is increasing, so f is 1  1 . Enter x= y +y +y+1
and use your CAS to solve the equation for y . Using Derive, we get two (irrelevant) solutions
involving imaginary expressions, as well as one which can be simplified to the following:
3
 1
y= f (x)=
6
4
( 3
D 27x +20 
2 3 2
D+27x  20 + 2
3
)

10
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

4 2
where D=3 3 27x  40x +16 . Maple and Mathematica each give two complex expressions and one
real expression, and the real expression is equivalent to that given by Derive. For example, Maple’s
2/3 1/3
1 M  8 2M 2 2 4
expression simplifies to , where M=108x +12 48 120x +81x  80 .
6 1/3
2M

6 4 3
58. (a) If we use Derive, then solving x=y +y for y gives us six solutions of the form y= B 1
3
A  A 
, where B {  2sin
A
3
,2sin
+
3 3
, 2cos +
3 6 } and A=sin
 1 27x 2
2
. The
6 4 3 A 
inverse for y=x +x ( x 0 ) is y= B 1 with B=2sin + , but because the domain of A
3 3 3
4 4
is 0, , this expression is only valid for x 0, .
27 27
6 4
Happily, Maple gives us the rest of the solution! We solve x=y +y for y to get the two real solutions


6 C
1/3
(C 2/3 2C 1/3+4) , where C=108x+12 3 x ( 27x 4 ) , and the inverse for y=x +x (
6 4
6 1/3
C
4
x 0 ) is the positive solution, whose domain is ,
.
27
Mathematica also gives two real solutions, equivalent to those of Maple. The positive one is
6 3  1/3
6
( 1/3 3
)
4 D +2 2 D  2 , where D= 2+27x+3 3 x 27x 4 . Although this expression also has
4
domain ,
, Mathematica is mysteriously able to plot the solution for all x 0 .
27

(b)

11
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

n
t/3 t/3 n t n
59. (a) n=100 2  =2  log =  t=3log . Using formula ( ), we can
100 2 100 3 2 100
ln (n/100)
write this as t=3 . This function tells us how long it will take to obtain n bacteria (given
ln 2
the number n ).
50,000 ln 500
(b) n=50 , 000 t=3log =3log 500=3  26.9 hours
2 100 2 ln 2

 t/a   t/a t
60. (a) Q=Q
0
( 1 e ) 
Q
=1 e
t/a
 e =1
Q
 
a
=ln 1
Q
 t= aln (1 Q/Q ) . This
0
Q Q Q
0 0 0
gives us the time t necessary to obtain a given charge Q .
(b) Q=0.9Q and a=2 t= 2ln 1 0.9 Q /Q
0 ( (
= 2ln 0.1 4.6 seconds.
0 0 ))
61. (a) To find the equation of the graph that results from shifting the graph of y=ln x3 units upward,
we add 3 to the original function to get y=ln x+3 .
(b) To find the equation of the graph that results from shifting the graph of y=ln x3 units to the left,
we replace x with x+3 in the original function to get y=ln ( x+3) .
(c) To find the equation of the graph that results from reflecting the graph of y=ln x about the x  axis,
we multiply the original equation by  1 to get y= ln x .
(d) To find the equation of the graph that results from reflecting the graph of y=ln x about the y  axis,
we replace x with  x in the original equation to get y=ln ( x) .
(e) To find the equation of the graph that results from reflecting the graph of y=ln x about the line y=x
x
, we interchange x and y in the original equation to get x=ln y y=e .
(f) To find the equation of the graph that results from reflecting the graph of y=ln x about the x  axis
and then about the line y=x , we first multiply the original equation by  1 and then interchange x and
 x
y in this equation to get x= ln y ln y= x y=e .
(g) To find the equation of the graph that results from reflecting the graph of y=ln x about the y  axis
and then about the line y=x , we first replace x with  x in the original equation and then interchange x
x x
and y to get x=ln ( y)  y=e  y= e .
(h) To find the equation of the graph that results from shifting the graph of y=ln x3 units to the left
and then reflecting it about the line y=x , we first replace x with x+3 in the original equation and then
12
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

x x
interchange x and y in this equation to get x=ln (y+3) y+3=e  y=e  3 .

62. (a) If the point ( x,y ) is on the graph of y= f (x) , then the point ( x c,y ) is that point shifted c units
 1
to the left. Since f is 1  1 , the point ( y,x ) is on the graph of y= f (x) and the point corresponding to
 1
( x c,y ) on the graph of f is ( y,x c ) on the graph of f. Thus, the curve’s reflection is shifted down
the same number of units as the curve itself is shifted to the left. So an expression for the inverse
 1  1
function is g (x)= f (x) c .
(b) If we compress (or stretch) a curve horizontally, the curve’s reflection in the line y=x is
compressed (or stretched) vertically by the same factor. Using this geometric principle, we see that
 1  1
the inverse of h(x)= f (cx) can be expressed as h (x)= ( 1/c ) f (x) .

 1 3   3   
63. (a) sin = since sin = and is in  , .
2 3 3 2 3 2 2
 1
(b) cos ( 1)= since cos  = 1 and  is in 0, .

    
64. (a) arctan( 1)= since tan  = 1 and  is in  , .
4 4 4 2 2
 1     3
(b) csc 2= since csc =2 and is in 0,   , .
6 6 6 2 2

 1     
65. (a) tan 3= since tan = 3 and is in .  ,
3 3 3 2 2
1   1   
(b) arcsin  = since sin  = and  is in  , .
2 4 4 2 4 2 2

 1     3
66. (a) sec 2= since sec = 2 and is in 0,   , .
4 4 4 2 2
    
(b) arcsin1= since sin =1 and is in  , .
2 2 2 2 2

 1
67. (a) sin (sin 0.7)=0.7 since 0.7 is in  1,1 .
 1 4  1    
(b) tan tan =tan 3= since is in  , .
3 3 3 2 2

2 2
68. (a) Let  =arctan2 , so tan  =2 sec  =1+tan  =1+4=5 sec  = 5  sec (arctan2)=sec  = 5
.
(b) Let

13
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

 1 5 5  1 5 2 5 2 119
 =sin . Then sin  = , so cos 2sin =cos 2 =1 2sin  =1 2 = .
13 13 13 13 169

 1    1 2 2
69. Let y=sin x . Then   y  cos y 0 , so cos (sin x)=cos y= 1 sin y = 1 x
2 2

 1  1 x
70. Let y=sin x . Then sin y=x , so from the triangle we see that tan (sin x)=tan y= .
2
1 x

 1  1 x
71. Let y=tan x . Then tan y=x , so from the triangle we see that sin (tan x)=sin y= .
2
1+x

72.

 1 2
Let y=cos x . Then cos y=x sin y= 1 x since 0 y  . So
 1 2
sin (2cos x)=sin 2y=2sin ycos y=2x 1 x .
73.

 1
The graph of sin x is the reflection of the graph of sin x about the line y=x .

74.

14
Stewart Calculus ET 5e 0534393217;1. Functions and Models; 1.6 Inverse Functions and Logarithms

 1
The graph of tan x is the reflection of the graph of tan x about the line y=x .

75.

 1
g(x)=sin (3x+1) .
Domain ( g ) = { x| 1 3x+1 1} = { x| 2 3x 0} =
   
{ x|
2
3 }
 x 0 = 
2
3
,0 .

Range ( g ) = { y|
2
 y
2 } =  ,
2 2
.

 1
76. (a) f (x)=sin sin x ( )

Since one function undoes what the other one does, we get the identity function, y=x , on the
restricted domain  1 x 1 .
 1
(b) g(x)=sin ( sin x )

This is similar to part (a), but with domain R . Equations for g on intervals of the form
  n n+1
 + n, + n , for any integer n , can be found using g ( x ) =( 1) x+( 1) n . The sine
2 2
function is monotonic on each of these intervals, and hence, so is g (but in a linear fashion).

15
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.1 The Tangent and Velocity Problems

1. (a) Using P ( 15,250 ) , we construct the following table:

t Q slope=m
PQ
694 250 444
5 ( 5,694 ) = = 44.4
5 15 10
444 250 194
10 ( 10,444 ) = = 38.8
10 15 5
111 250 139
20 ( 20,111 ) = = 27.8
20 15 5
28 250 222
25 ( 25,28 ) = = 22.2
25 15 10
0 250 250
30 ( 30,0 ) = = 16.6
30 15 15

(b) Using the values of t that correspond to the points closest to P ( t=10 and t=20 ), we have
 38.8+ (  27.8 )
= 33.3
2

(c)  300
From the graph, we can estimate the slope of the tangent line at P to be = 33.3 .
9

2.
2948 2530 418 2948 2661 287
(a) Slope = =  69.67 (b) Slope = = =71.75
42 36 6 42 38 4
2948 2806 142 3080 2948 132
(c) Slope = = =71 (d) Slope = = =66
42 40 2 44 42 2

From the data, we see that the patient’s heart rate is decreasing from 71 to 66 heartbeats / minute after
42 minutes. After being stable for a while, the patient’s heart rate is dropping.

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.1 The Tangent and Velocity Problems

1
3. (a) For the curve y=x/ ( 1+x ) and the point P 1,
2
x Q m
PQ
(i) 0.5 ( 0.5,0.333333) 0.333333
(ii) 0.9 ( 0.9,0.473684 ) 0.263158
(iii) 0.99 ( 0.99,0.497487) 0.251256
(iv) 0.999 ( 0.999,0.499750 ) 0.250125
(v) 1.1 ( 1.5,06 ) 0.2
(vi) 1.5 ( 1.1,0.523810 ) 0.238095
(vii) 1.01 ( 1.01,0.502488 ) 0.248756
(viii) 1.001 ( 1.001,0.500250 ) 0.249875

1
(b) The slope appears to be .
4
1 1 1 1
(c) y = (x 1) or y= x+ .
2 4 4 4

4. For the curve y=ln x and the point P(2,ln2) :

(a)
x Q m
PQ
(i) 1.5 ( 1.5,0.405465) 0.575364
(ii) 1.9 ( 1.9,0.641854 ) 0.512933
(iii) 1.99 ( 1.99,0.688135) 0.501254
(iv) 1.999 ( 1.999,0.692647) 0.500125
(v) 2.5 ( 2.5,0.916291 ) 0.446287
(vi) 2.1 ( 2.1,0.741937) 0.487902
(vii) 2.01 ( 2.01,0.698135) 0.498754
(viii) 2.001 ( 2.001,0.693647) 0.499875

1
(b) The slope appears to be .
2
1 1
(c) y ln 2= (x 2) or y= x 1+ln 2
2 2
(d)

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.1 The Tangent and Velocity Problems

2 2
5. (a) y=y(t)=40t 16t . At t=2 , y=40(2) 16(2) =16 . The average velocity between times 2 and 2+h
2 2
y(2+h) y(2) 40(2+h) 16(2+h)  16  24h 16h
is v = = = = 24 16h , if h 0 .
ave (2+h) 2 h h
(i) [2,2.5] : h=0.5 , v = 32 ft / save
(ii) [2,2.1] : h=0.1 , v = 25.6 ft / s ave

(iii) [2,2.05] : h=0.05 , vave= 24.8 ft / s (iv) [2,2.01] : h=0.01 , vave= 24.16 ft / s

(b) The instantaneous velocity when t=2 ( h approaches 0 ) is  24 ft / s.

6. The average velocity between t and t+h seconds is

2
58(t+h) 0.83(t+h)  58t 0.83t( =
2
)
58h 1.66th 0.83h
2
=58 1.66t 0.83h if h 0 .
h h
(a) Here t=1 , so the average velocity is 58 1.66 0.83h=56.34 0.83h .
(i) 1,2 : h=1,55.51 m / s (ii) 1,1.5 : h=0.5 , 55.925 m / s
(iii) 1,1.1 : h=0.1 , 56.257 m / s (iv) 1,1.01 : h=0.01 , 56.3317 m / s
(v) 1,1.001 : h=0.001 , 56.33917 m / s

(b) The instantaneous velocity after 1 second is 56.34 m / s.

7. (a)
13 7
(i) 1,3 : h=2 , v =
ft / s (ii) 1,2 : h=1 , v ft / s=
ave 6 ave 6
19 331
(iii) 1,1.5 : h=0.5 , v = ft / s (iv) 1,1.1 : h=0.1 , v = ft / s
ave 24 ave 600

(b) As h approaches 0 , the velocity approaches

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.1 The Tangent and Velocity Problems

3 1
= ft / s.
6 2

(c)

(d)

s(2+h) s(2)
8. Average velocity between times t=2 and t=2+h is given by .
h
(a)
(i) s(5) s(2) 178 32 146
h=3 v = = =  48.7 ft / s
av 5 2 3 3
(ii) s(4) s(2) 119 32 87
h=2 v = = = =43.5 ft / s
av 4 2 2 2
(iii) s(3) s(2) 70 32
h=1 v = = =38 ft / s
av 3 2 1

(b) Using the points ( 0.8,0 ) and ( 5,118 ) from the approximate tangent line, the instantaneous
118 0
velocity at t=2 is about  28 ft / s.
5 0.8

9. For the curve y=sin (10 /x) and the point P ( 1,0 ) :
(a)
x Q m
PQ
2 ( 2,0 ) 0

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.1 The Tangent and Velocity Problems

1.5 ( 1.5,0.8660 ) 1.7321

1.4 ( 1.4, 0.4339 )  1.0847
1.3 ( 1.3, 0.8230 )  2.7433
1.2 ( 1.2,0.8660 ) 4.3301
1.1 ( 1.1, 0.2817)  2.8173

x Q m
PQ
0.5 ( 0.5,0 ) 0
0.6 ( 0.6,0.8660 )  2.1651
0.7 ( 0.7,0.7818 )  2.6061
0.8 ( 0.8,1 )  5
0.9 ( 0.9, 0.3420 ) 3.4202

As x approaches 1 , the slopes do not appear to be approaching any particular value.

(b)
We see that problems with estimation are caused by the frequent oscillations of the graph. The
tangent is so steep at P that we need to take x  values much closer to 1 in order to get accurate
estimates of its slope.

(c) If we choose x=1.001 , then the point Q is ( 1.001, 0.0314 ) and m   31.3794 . If x=0.999 , then
PQ
Q is ( 0.999,0.0314 ) and m = 31.4422 . The average of these slopes is  31.4108 . So we estimate
PQ
that the slope of the tangent line at P is about  31.4 .

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

1. As x approaches 2 , f (x) approaches 5 . [Or, the values of f (x) can be made as close to 5 as we like
by taking x sufficiently close to 2 (but x 2 ).] Yes, the graph could have a hole at ( 2,5) and be
defined such that f (2)=3 .

2. As x approaches 1 from the left, f (x) approaches 3 ; and as x approaches 1 from the right, f (x)
approaches 7 . No, the limit does not exist because the left and right hand limits are different.

3. (a) lim f (x)= means that the values of f (x) can be made arbitrarily large (as large as we please)
x  3
by taking x sufficiently close to  3 (but not equal to  3 ).
(b) lim f (x)=  means that the values of f (x) can be made arbitrarily large negative by taking x
+
x 4
sufficiently close to 4 through values larger than 4 .

4. (a) lim f (x)=3

x 0
(b) lim f (x)=4

x 3
(c) lim f (x)=2
+
x 3
(d) lim f (x) does not exist because the limits in part (b) and part (c) are not equal.
x 3
(e) f (3)=3

5. (a) f (x) approaches 2 as x approaches 1 from the left, so lim f (x)=2 .


x 1
(b) f (x) approaches 3 as x approaches 1 from the right, so lim f (x)=3 .
+
x 1
(c) lim f (x) does not exist because the limits in part (a) and part (b) are not equal.
x 1
(d) f (x) approaches 4 as x approaches 5 from the left and from the right, so lim f (x)=4 .
x 5
(e) f (5) is not defined, so it doesn’t exist.

6. (a) lim g(x)= 1


x  2
(b) lim g(x)=1
+
x  2
(c) lim g(x) doesn’t exist
x  2
(d) g( 2)=1
(e) lim g(x)=1

x 2

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

(f) lim g(x)=2

+
x 2
(g) lim g(x) doesn’t exist
x 2
(h) g(2)=2
(i) lim g(x) doesn’t exist
+
x 4
(j) lim g(x)=2

x 4
(k) g(0) doesn’t exist
(l) lim g(x)=0
x 0

7. (a) lim g(t)= 1


t 0
(b) lim g(t)= 2
+
t 0
(c) lim g(t) does not exist because the limits in part (a) and part (b) are not equal.
t 0
(d) lim g(t)=2

t 2
(e) lim g(t)=0
+
t 2
(f) lim g(t) does not exist because the limits in part (d) and part (e) are not equal.
t 2
(g) g(2)=1
(h) lim g(t)=3
t 4

8. (a) lim R(x)= 

x 2
(b) lim R(x)=
x 5
(c) lim R(x)= 

x  3
(d) lim R(x)=
+
x  3
(e) The equations of the vertical asymptotes are x= 3 , x=2 , and x=5 .

9. (a) lim f (x)= 

x  7
(b) lim f (x)=
x  3
(c)
2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

lim f (x)=
x 0
(d) lim f (x)= 

x 6
(e) lim f (x)=
+
x 6
(f) The equations of the vertical asymptotes are x= 7 , x= 3 , x=0 , and x=6 .

10. lim f (t)=150 mg and lim f (t)=300 mg. These limits show that there is an abrupt change in the
 +
t 12
t 12
amount of drug in the patient’s bloodstream at t=12 h. The left hand limit represents the amount of
the drug just before the fourth injection. The right hand limit represents the amount of the drug just
after the fourth injection.

11.

(a) lim f (x)=1


x 0

(b) lim f (x)=0

+
x 0

(c) lim f (x) does not exist because the limits in part (a) and part (b) are not equal.
x 0

12. lim f (x) exists for all a except a= 1 .

x a

13. lim f (x)=4 , lim f (x)=2 ,

+ 
x 3 x 3
3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

lim f (x)=2 , f (3)=3 , f ( 2)=1

x  2

14. lim f (x)=1 , lim f (x)= 1 , lim f (x)=0 , lim f (x)=1 , f (2)=1 , f (0) is undefined
 +  +
x 0 x 0 x 2 x 2

2
x  2x
15. For f (x)= 2
:
x  x 2
x f (x)
2.5 0.714286
2.1 0.677419
2.05 0.672131
2.01 0.667774
2.005 0.667221
2.001 0.666778

x f (x)
1.9 0.655172
1.95 0.661017
1.99 0.665552
1.995 0.666110
1.999 0.666556
2
x  2x 2
It appears that lim =0.6= .
2 3
x 2 x  x 2

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

2
x  2x
16. For f (x)= 2
:
x  x 2
x f (x)
0 0
 0.5  1
 0.9  9
 0.95  19
 0.99  99
 0.999  999

x f (x)
 2 2
 1.5 3
 1.1 11
 1.01 101
 1.001 1001

2
x  2x  +
It appears that lim 2
does not exist since f (x)   as x  1 and f (x)  as x  1 .
x  1 x  x 2

x
e  1 x
17. For f (x)= :
2
x
x f (x)
1 0.718282
0.5 0.594885
0.1 0.517092
0.05 0.508439
0.01 0.501671

x f (x)
 1 0.367879

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

 0.5 0.426123
 0.1 0.483742
 0.05 0.491770
 0.01 0.498337

x
e  1 x 1
It appears that lim =0.5= .
x 0 2 2
x

18. For f (x)=xln x+x ( 2

):
x f (x)
1 0.693147
0.5  0.143841
0.1  0.220727
0.05  0.147347
0.01  0.045952
0.005  0.026467
0.001  0.006907

It appears that lim xln x+x =0 .

+
( 2
)
x 0

x+4  2
19. For f (x)= :
x
x f (x)
1 0.236068
0.5 0.242641
0.1 0.248457
0.05 0.249224
0.01 0.249844

x f (x)
 1 0.267949

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

 0.5 0.258343
 0.1 0.251582
 0.05 0.250786
 0.01 0.250156
x+4  2 1
It appears that lim =0.25= .
x 0 x 4

tan 3x
20. For f (x)= :
tan 5x
x f (x)
 0.2 0.439279
 0.1 0.566236
 0.05 0.591893
 0.01 0.599680
 0.001 0.599997
tan 3x 3
It appears that lim =0.6= .
x 0 tan 5x 5

6
x 1
21. For f (x)= 10
:
x  1
x f (x)
0.5 0.985337
0.9 0.719397
0.95 0.660186
0.99 0.612018
0.999 0.601200

x f (x)
1.5 0.183369
1.1 0.484119
1.05 0.540783
1.01 0.588022

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

1.001 0.598800
6
x 1 3
It appears that lim =0.6= .
10 5
x 1 x  1

x x
9 5
22. For f (x)= :
x
x f (x)
0.5 1.527864
0.1 0.711120
0.05 0.646496
0.01 0.599082
0.001 0.588906

x f (x)
 0.5 0.227761
 0.1 0.485984
 0.05 0.534447
 0.01 0.576706
 0.001 0.586669
x x
9 5
It appears that lim =0.59 . Later we will be able to show that the exact value is ln (9/5) .
x 0 x

6 + 6
23. lim = since ( x 5)  0 as x 5 and >0 for x>5 .
+ x 5 x 5
x 5

6  6
24. lim =  since ( x 5)  0 as x 5 and <0 for x<5 .
 x 5 x 5
x 5

2 x
25. lim = since the numerator is positive and the denominator approaches 0 through
2
x 1
(x 1)
positive values as x 1.

26.

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

x 1 2 x 1
lim 2
=  since x  0 as x 0 and 2
<0 for 0<x<1 and for  2<x<0 .
x 0 x (x+2) x (x+2)

x 1 + x 1
27. lim 2
=  since ( x+2 )  0 as x  2 and 2
<0 for  2<x<0 .
+
x  2 x (x+2) x (x+2)


28. lim csc x=lim ( 1/sin x ) = since sin x 0 as x  and sin x>0 for 0<x< .
 
x  x 


29. lim sec x= lim ( 1/cos x ) =  since cos x 0 as x (  /2) and cos x<0 for   <x<  /2 .
 
x (   /2 ) x (   /2 )

+ +
30. lim ln (x 5)=  since x 5 0 as x 5 .
+
x 5

3
31. (a) f (x)=1/(x  1)
x f (x)
0.5  1.14
0.9  3.69
0.99  33.7
0.999  333.7
0.9999  3333.7
0.99999  33 , 333.7

x f (x)
1.5 0.42
1.1 3.02
1.01 33.0
1.001 333.0
1.0001 3333.0
1.00001 33 , 333.3
From these calculations, it seems that lim f (x)=  and lim f (x)= .
 +
x 1 x 1
(b) If x is slightly smaller than 1 , then
9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

3 3
x  1 will be a negative number close to 0 , and the reciprocal of x  1 , that is, f (x) , will be a negative
number with large absolute value. So lim f (x)=  .

x 1
3
If x is slightly larger than 1 , then x  1 will be a small positive number, and its reciprocal, f (x) , will
be a large positive number. So lim f (x)= .
+
x 1
(c) It appears from the graph of f that lim f (x)=  and lim f (x)= .
 +
x 1 x 1

x x + +
32. (a) y= = . Therefore, as x  1 or x 2 , the denominator approaches 0 , and
2 (x 2)(x+1)
x  x 2
 
y>0 for x< 1 and for x>2 , so lim y=lim y= . Also, as x  1 or x 2 , the denominator
+ +
x  1 x 2
approaches 0 and y<0 for  1<x<2 , so lim y=lim y=  .
 
x  1 x 2

(b)

1/x
33. (a) Let h(x)= ( 1+x ) .
x h(x)
 0.001 2.71964
 0.0001 2.71842
 0.00001 2.71830
 0.000001 2.71828
0.000001 2.71828
0.00001 2.71827

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

0.0001 2.71815
0.001 2.71692
1/x
It appears that lim ( 1+x )  2.71828 , which is approximately e . In Section 7.4 we will see that the
x 0
value of the limit is exactly e .

(b)

x
34. For the curve y=2 and the points P ( 0,1 ) and Q x,2 ( x) :
x Q m
PQ

0.1 ( 0.1,1.0717735) 0.71773

0.01 ( 0.01,1.0069556 ) 0.69556
0.001 ( 0.001,1.0006934 ) 0.69339
0.0001 ( 0.0001,1.0000693) 0.69317
The slope appears to be about 0.693 .

35. (a)
x f (x)
1 0.998000
0.8 0.638259
0.6 0.358484
0.4 0.158680
0.2 0.038851
0.1 0.008928
0.05 0.001465
It appears that lim f (x)=0 .
x 0
(b)
x f (x)
0.04 0.000572

11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

0.02  0.000614
0.01  0.000907
0.005  0.000978
0.003  0.000993
0.001  0.001000
It appears that lim f (x)= 0.001 .
x 0

tan x x
36. h(x)=
3
x

(a)
x h(x)
1.0 0.55740773
0.5 0.37041992
0.1 0.33467209
0.05 0.33366700
0.01 0.33334667
0.005 0.33333667

1
(b) It seems that lim h(x)= .
x 0 3

(c) x h(x)
0.001 0.33333350
0.0005 0.33333344
0.0001 0.33333000
0.00005 0.33333600
0.00001 0.33300000
0.000001 0.00000000
Here the values will vary from one calculator to another. Every calculator will eventually give
false values .
(d) As in part (c), when we take a small enough viewing rectangle we get incorrect output.

12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

37. No matter how many times we zoom in toward the origin, the graphs of f (x)=sin ( /x) appear to
consist of almost vertical lines. This indicates more and more frequent oscillations as x 0 .

m  2 2 +
0
38. lim m=lim . As v c , 1 v /c  0 , and m  .
  2 2
v c v c 1 v /c

39.

There appear to be vertical asymptotes of the curve y=tan (2sin x) at x  0.90 and x  2.24 . To
find the exact equations of these asymptotes, we note that the graph of the tangent function has
13
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.2 The Limit of a Function

 
vertical asymptotes at x= + n . Thus, we must have 2sin x= + n , or equivalently,
2 2
    1
sin x= + n . Since  1 sin x 1 , we must have sin x= and so x= sin (corresponding
4 2 4 4
to x  0.90 ).

 1  1
Just as 150 is the reference angle for 30 ,   sin is the reference angle for sin . So
4 4
 1
x=   sin are also equations of the vertical asymptotes (corresponding to x  2.24 ).
4

(
40. (a) Let y= x  1 / ( x  1 ) .
3
)
x y
0.99 5.92531
0.999 5.99250
0.9999 5.99925
1.01 6.07531
1.001 6.00750
1.0001 6.00075

From the table and the graph, we guess that the limit of y as x approaches 1 is 6 .
3
x 1
(b) We need to have 5.5< <6.5 . From the graph we obtain the approximate points of
x 1
intersection P ( 0.9313853,5.5) and Q ( 1.0649004,6.5) . Now 1 0.9313853 0.0686 and
1.0649004 1 0.0649 , so by requiring that x be within 0.0649 of 1 , we ensure that y is within 0.5 of
6.

14
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

1. (a)
lim f (x)+h(x) =lim f (x)+lim h(x)
x a x a x a
= 3+8=5

2 2 2
(b) lim f (x) = lim f (x) =( 3) =9
x a x a
3 3
(c) lim h(x) = 3 lim h(x) = 8 =2
x a x a
1 1 1 1
(d) lim = = =
x a f (x) lim f (x)  3 3
x a
lim f (x)
f (x) x a  3 3
(e) lim = = =
x a h(x) lim h(x) 8 8
x a
lim g(x)
g(x) x a 0
(f) lim = = =0
x a f (x) lim f (x)  3
x a
(g) The limit does not exist, since lim g ( x ) =0 but lim f (x) 0 .
x a x a
2lim f (x)
2 f (x) x a 2 (  3) 6
(h) lim = = =
x a h(x) f (x) lim h(x) lim f (x) 8 (  3 ) 11
x a x a

2. (a) lim f (x)+g(x) =lim f (x)+lim g(x)=2+0=2

x 2 x 2 x 2
(b) lim g(x) does not exist since its left and right hand limits are not equal, so the given limit does
x 1
not exist.
(c) lim f (x)g(x) =lim f (x) lim g(x)=0 1.3=0
x 0 x 0 x 0
(d) Since lim g(x)=0 and g is in the denominator, but lim f (x)= 1 0 , the given limit does not
x  1 x  1
exist.
3 3 3
(e) lim x f (x)= lim x lim f (x) =2  2=16
x 2 x 2 x 2
(f) lim 3+ f (x) = 3+lim f (x) = 3+1 =2
x 1 x 1

3.

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

4 2 4 2
lim (3x +2x  x+1) =lim 3x +lim 2x  lim x+lim 1 [Limit Laws 1 and 2]
x  2 x  2 x  2 x  2 x  2
4 2
=3lim x +2lim x  lim x+lim 1 [3]
x  2 x  2 x  2 x  2
4 2 [9, 8, and 7]
=3( 2) +2( 2)  ( 2)+(1)
=48+8+2+1=59

4.

2 lim ( 2x2+1)
2x +1 x 2
lim = [Limit Law 5]
x 2
2
x +6x 4 lim ( x2+6x 4)
x 2
2
2lim x +lim 1
x 2 x 2
= [2, 1, and 3]
2
lim x +6lim x lim 4
x 2 x 2 x 2
2
2(2) +1 9 3
= = = [9, 7, and 8]
2 12 4
(2) +6(2) 4

5.
2 3 2 3
lim (x  4)(x +5x 1) =lim (x  4) lim (x +5x 1) [Limit Law 4]
x 3 x 3 x 3
2 3
= lim x  lim 4  lim x +5lim x lim 1 [2, 1, and 3]
x 3 x 3 x 3 x 3 x 3
2 3 [7, 8, and 9]
=(3  4) (3 +5 3 1)
=5 41=205

6.
2 3 5 2 3 5
lim (t +1) (t+3) =lim (t +1)  lim (t+3) [Limit Law 4]
t  1 t  1 t  1
2 3 5
= lim (t +1)  lim (t+3) [6]
t  1 t  1

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

2 3 5
= lim t +lim 1  lim t+lim 3 [1]
t  1 t  1 t  1 t  1
2 3 5
= ( 1) +1   1+3 =8 32=256 [9, 7, and 8]

7.
1+3x 3 1+3x 3
lim = lim [6]
2 4 2 4
x 1 1+4x +3x x 1 1+4x +3x
lim (1+3x) 3
x 1
= 2 4 [5]
lim (1+4x +3x )
x 1

lim 1+3lim x 3
x 1 x 1
= [2, 1, and 3]
2 4
lim 1+4lim x +3lim x
x 1 x 1 x 1
3 4 3 1 3 1
1+3(1)
= = = =
2 4 8 2 8 [7, 8, and 9]
1+4(1) +3(1)

8.

lim
4
u +3u+6 = lim ( u4+3u+6) [11]
u  2 u  2

4
= lim u +3lim u+lim 6 [1, 2, and 3]
u  2 u  2 u  2

4 [9, 8, and 7]
= (  2 ) +3 (  2 ) +6
= 16 6+6 = 16 =4

9.

lim 16 x =
2
lim

( 16 x2) [11]

x 4 x 4

2
= lim 16 lim x [2]
 
x 4 x 4

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

2 [7 and 9]
= 16 ( 4 ) =0

10. (a) The left hand side of the equation is not defined for x=2 , but the right hand side is.
(b) Since the equation holds for all x 2 , it follows that both sides of the equation approach the same
limit as x 2 , just as in Example 3. Remember that in finding lim f (x) , we never consider x=a .
x a

2
x +x 6 (x+3)(x 2)
11. lim =lim =lim (x+3)=2+3=5
x 2 x 2 x 2 x 2 x 2

2
x +5x+4 (x+4)(x+1) x+1  4+1  3 3
12. lim = lim = lim = = =
x  4
2
x  4 (x+4)(x 1) x  4 x 1  4 1  5 5
x +3x 4

2
x  x+6 2
13. lim does not exist since x 2 0 but x  x+6 8 as x 2.
x 2 x 2

2
x  4x x(x 4) x 4 4
14. lim =lim =lim = =
2 (x 4)(x+1) x 4 x+1 4+1 5
x 4 x  3x 4 x 4

2
t  9 (t+3)(t 3) t 3  3 3  6 6
15. lim =lim =lim = = =
3 (2t+1)(t+3) t  3 2t+1 2( 3)+1  5 5
2
t  3 2t +7t+3 t 

2
x  4x 2 2
16. lim 2
does not exist since x  3x 4 0 but x  4x 5 as x  1.
x  1 x  3x 4

2 2 2
(4+h)  16 (16+8h+h ) 16 8h+h h(8+h)
17. lim =lim =lim =lim =lim (8+h)=8+0=8
h 0 h h 0 h h 0 h h 0 h h 0

( x 1 ) ( x +x+1 )
3 2 2 2
x 1 x +x+1 1 +1+1 3
18. lim =lim =lim = =
2 (x 1)(x+1) x 1 x+1 1+1 2
x 1 x 1 x 1

19.

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

lim
4
( 1+h )  1
=lim
( 1+4h+6h2+4h3+h4)  1 =lim 4h+6h +4h +h
2 3 4

h 0 h h 0 h h 0 h
2 3
h(4+6h+4h +h )
=lim
h
2 3
=lim 4+6h+4h +h =4+0+0+0=4 ( )
h 0 h 0

20.

lim
(2+h)  8
=lim
3
8+12h+6h +h  8
=lim
(
12h+6h +h
2 3
) 2 3

h 0 h h 0 h h 0 h

=lim ( 12+6h+h2) =12+0+0=12

h 0

9 t ( 3+ t ) ( 3 t )
21. lim =lim =lim ( 3+ t ) =3+ 9 =6
t 9 3 t t 9 3 t t 9

22.
1+h  1 1+h  1 1+h +1 ( 1+h )  1 h
lim = lim  = lim = lim
h 0 h h 0 h 1+h +1 h 0 h ( 1+h +1 ) h 0 h( 1+h +1 )
1 1 1
= lim = =
h 0 1+h +1 1 +1 2

23.
x+2  3 x+2  3 x+2 +3 ( x+2 )  9
lim =lim  =lim
x 7 x 7 x 7 x 7 x+2 +3 x 7 ( x 7) ( x+2 +3)
x 7 1 1 1
=lim =lim = =
x 7 ( x 7) ( x+2 +3) x 7 x+2 +3 9 +3 6

24.

lim
x  16
4
=lim
( x+2 ) ( x 2 ) x +4 2 ( 2
) 2
=lim ( x+2 ) x +4 =lim ( x+2 ) lim x +4 ( ) ( )
x 2 x 2 x 2 x 2 x 2 x 2 x 2

(
= ( 2+2 ) 2 +4 =32
2
)
25.

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

1 1 x+4
+
4 x 4x x+4 1 1 1
lim =lim =lim =lim = =
x  4 4+x x  4 4+x x  4 4x(4+x) x  4 4x 4( 4) 16

26. lim
1

1
=lim
( t2+t )  t =lim t2 =lim 1 = 1 =1
t ( t +t ) t 0
t 0 t 2
t 0
2 t t(t+1) t 0 t+1 0+1
t +t

27.
2
x  81 ( x 9 ) ( x+9 ) ( x  3)( x +3) ( x+9 ) factor x 9 as a
lim =lim =lim
x 9 x  3 x 9 x 3 x 9 x 3 difference of squares

=lim ( x +3) ( x+9 ) = ( 9 +3) ( 9+9 ) =6 18=108

x 9

28.
 1  1 1 1
(3+h)  3 
lim 3+h 3 3 ( 3+h )  h
h =lim =lim =lim
h 0
h 0 h h 0 h(3+h)3 h 0 h(3+h)3
1 1 1 1
=lim  = = =
h 0 3(3+h) lim 3(3+h) 3(3+0) 9
h 0

29.

lim
1

1 1 1+t ( 1 1+t ) ( 1+ 1+t )  t
=lim =lim =lim
t 0 t 1+t t t 0 t 1+t t 0 t t+1 ( 1+ 1+t ) t 0 t 1+t ( 1+ 1+t )
 1  1 1
=lim = =
t 0 1+t ( 1+ 1+t ) 1+0 ( 1+ 1+0 ) 2

30.

lim
2
x x
=lim
(
x 1 x
3/2
) =lim x ( 1 x ) ( 1+ x +x ) [difference of cubes]
x 1 1 x x 1 1 x x 1 1 x
=lim x ( 1+ x +x ) =lim 1(1+1+1) =3
x 1 x 1

Another method: We ’’add and subtract’’ 1 in the numerator, and then split up the fraction:

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

lim
x x
2
=lim
( (
x  1 ) + 1 x
2
) =lim  1+
(1 x)(1+x)
x 1 1 x x 1 1 x x 1 1 x
( 1 x ) ( 1+ x ) ( 1+x )
=lim  1+ = 1+ ( 1+ 1 ) ( 1+1 ) =3
x 1 1 x

31. (a)
x 2
lim 
x 0 1+3x  1 3
(b)
x f (x)
 0.001 0.6661663
 0.0001 0.6666167
 0.00001 0.6666617
 0.000001 0.6666662
0.000001 0.6666672
0.00001 0.6666717
0.0001 0.6667167
0.001 0.6671663
2
The limit appears to be .
3
(c)
x 1+3x +1 x ( 1+3x +1 ) x ( 1+3x +1 )
lim  =lim =lim
x 0 1+3x  1 1+3x +1 x 0 ( 1+3x )  1 x 0 3x
1
= lim ( 1+3x +1 ) [Limit Law 3]
3 x 0
1
= lim (1+3x) +lim 1 [1 and 11]
3 x 0 x 0
1
= lim 1+3lim x +1 [1, 3, and 7]
3 x 0 x 0

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

1
=
3
( 1+3 0 +1 ) [7 and 8]
1 2
= (1+1)=
3 3

32. (a)
3+x  3
lim  0.29
x 0 x
(b)
x f (x)
 0.001 0.2886992
 0.0001 0.2886775
 0.00001 0.2886754
 0.000001 0.2886752
0.000001 0.2886751
0.00001 0.2886749
0.0001 0.2886727
0.001 0.2886511
The limit appears to be approximately 0.2887 .
(c)
3+x  3 3+x + 3 (3+x) 3 1
lim  =lim =lim
x 0 x 3+x + 3 x 0 x ( 3+x + 3 ) x 0 3+x + 3
lim 1
x 0
= [Limit Laws 5 and 1]
lim 3+x +lim 3
x 0 x 0
1
=
lim ( 3+x ) + 3 [7 and 11]
x 0
1
= [1, 7, and 8]
3+0 + 3

8
 Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

1
=
2 3

2 2 2 2 2 2
33. Let f (x)= x , g(x)=x cos 20 x and h(x)=x . Then  1 cos 20 x 1  x  x cos 20 x x
f (x) g(x) h(x) . So since lim f (x)=lim h(x)=0 , by the Squeeze Theorem we have lim g(x)=0 .
x 0 x 0 x 0

3 2 3 2 3 2
34. Let f (x)= x +x , g(x)= x +x sin ( /x) , and h(x)= x +x . Then  1 sin ( /x) 1
3 2 3 2 3 2
 x +x  x +x sin ( /x) x +x f ( x )  g(x) h(x) . So since lim f (x)=lim h(x)=0 , by the
x 0 x 0
Squeeze Theorem we have lim g(x)=0 .
x 0

2
35. 1 f (x) x +2x+2 for all x . Now lim 1=1 and
x  1

lim ( x +2x+2) =lim

2 2 2
x +2lim x+lim 2= (  1 ) +2( 1)+2=1 . Therefore, by the Squeeze Theorem,
x  1 x  1 x  1 x  1
lim f (x)=1 .
x  1

36. 3x
3
f (x) x +2 for 0 x 2 . Now lim 3x=3 and lim ( x3+2) =lim x3+lim 2=13+2=3 . Therefore,
x 1 x 1 x 1 x 1
by the Squeeze Theorem, lim f (x)=3 .
x 1

37.  1 cos (2/x) 1

4 4
 x  x cos (2/x) x . Since lim
4
(  x4) =0 and lim x4=0 , we have
x 0 x 0

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

4
lim x cos (2/x) =0 by the Squeeze Theorem.
x 0

 1 sin ( /x) 1 sin ( /x)

38.  1 sin ( /x) 1 e  e  e x /e xe  x e . Since lim ( x /e ) =0 and
+
x 0
sin ( /x )
lim ( x e ) =0 , we have lim xe =0 by the Squeeze Theorem.
+ +
x 0 x 0

39. If x> 4 , then x+4 =x+4 , so lim x+4 = lim ( x+4 ) = 4+4=0 .
+ +
x  4 x  4
If x< 4 , then x+4 = ( x+4 ) , so lim x+4 = lim  ( x+4 ) = (  4+4 ) =0 .
 
x  4 x  4
Since the right and left limits are equal, lim x+4 =0 .
x  4

x+4  ( x+4 )
40. If x< 4 , then x+4 = ( x+4 ) , so lim = lim = lim (  1 ) = 1 .
 x+4  x+4 
x  4 x  4 x  4

x 2 x 2
41. If x>2 , then x 2 =x 2 , so lim =lim =lim 1=1 . If x<2 , then x 2 = ( x 2 ) , so
+ x 2 + x 2 +
x 2 x 2 x 2
x 2  ( x 2 ) x 2
lim =lim =lim  1= 1 . The right and left limits are different, so lim does not
 x 2  x 2  x 2 x 2
x 2 x 2 x 2
exist.

2 2
3 2x  3x 2x  3x x ( 2x 3)
42. If x> , then 2x 3 =2x 3 , so lim = lim = lim = lim x=1.5 . If
2 + 2x 3 + 2x 3 + 2x 3 +
x 1.5 x 1.5 x 1.5 x 1.5
2 2
3 2x  3x 2x  3x x ( 2x 3)
x< , then 2x 3 =3 2x , so lim = lim = lim = lim  x= 1.5 . The
2  2x 3   ( 2x 3)   ( 2x 3) 
x 1.5 x 1.5 x 1.5 x 1.5
2
2x  3x
right and left limits are different, so lim does not exist.
x 1.5 2x 3

1 1 1 1 2
43. Since x = x for x<0 , we have lim  =lim  =lim , which does not
 x x  x  x  x
x 0 x 0 x 0
exist since the denominator approaches 0 and the numerator does not.

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

1 1 1 1
44. Since x =x for x>0 , we have lim  =lim  =lim 0=0 .
+ x x + x x +
x 0 x 0 x 0

45. (a)

(b)
(i) Since sgnx=1 for x>0 , lim sgnx=lim 1=1 .
+ +
x 0 x 0
(ii) Since sgnx= 1 for x<0 , lim sgn x=lim  1= 1 .
 
x 0 x 0
(iii) Since lim sgnx lim sgnx , lim sgnx does not exist.
 + x 0
x 0 x 0
(iv) Since sgnx =1 for x 0 , lim sgnx =lim 1=1 .
x 0 x 0

46. (a)
lim f (x) =lim
 
( 4 x2) =lim 4 lim x2  
x 2 x 2 x 2 x 2
=4 4=0

lim f (x) =lim ( x 1 ) =lim x lim 1

+ + + +
x 2 x 2 x 2 x 2
=2 1=1

(b) No, lim f (x) does not exist since lim f (x) lim f (x) .
x 2  +
x 2 x 2
(c)

47. (a)

11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

2 2
x 1 x 1
(i) lim =lim =lim ( x+1 ) =2
+ x 1 + x 1 +
x 1 x 1 x 1
2 2
x 1 x 1
(ii) lim =lim =lim  ( x+1 ) = 2
 x 1   ( x 1 ) 
x 1 x 1 x 1

(b) No, lim F(x) does not exist since lim F(x) lim F(x) .
x 1 + 
x 1 x 1
(c)

48. (a)
(i) 2 2
lim h(x)=lim x =0 =0
+ +
x 0 x 0
(ii) lim h(x)=lim x=0 , so lim h(x)=0 .
  x 0
x 0 x 0
(iii) 2 2
lim h(x)=lim x =1 =1
x 1 x 1
(iv) 2 2
lim h(x)=lim x =2 =4
 
x 2 x 2
(v) lim h(x)=lim ( 8 x ) =8 2=6
+ +
x 2 x 2
(vi) Since lim h(x) lim h(x) , lim h(x) does not exist.
 + x 2
x 2 x 2

(b)

49. (a)
12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

(i) [ x]= 2 for  2 x< 1 , so lim [ x]= lim (  2 ) = 2

+ +
x  2 x  2
(ii) [ x]= 3 for  3 x< 2 , so lim [ x]= lim (  3) = 3 .
 
x  2 x  2
The right and left limits are different, so lim [ x] does not exist.
x  2
(iii) [ x]= 3 for  3 x< 2 , so lim [ x]= lim (  3) = 3 .
x  2.4 x  2.4

(b)
(i) [ x]=n 1 for n 1 x<n , so lim [ x]=lim ( n 1 ) =n 1 .
 
x n x n
(ii) [ x]=n for n x<n+1 , so lim [ x]=lim n=n .
+ +
x n x n

(c) lim [ x] exists

a is not an integer.
x a

50. (a)

(b)
(i) lim f (x)=lim ( x [ x]) =lim x ( n 1 ) =n ( n 1 ) =1
  
x n x n x n
(ii) lim f (x)=lim ( x [ x]) =lim ( x n ) =n n=0
+ + +
x n x n x n

(c) lim f (x) exists

a is not an integer.
x a

51. The graph of f (x)=[ x]+[ x] is the same as the graph of g(x)= 1 with holes at each integer, since
f (a)=0 for any integer a . Thus, lim f (x)= 1 and lim f ( x ) = 1 , so lim f (x)= 1 . However,
 + x 2
x 2 x 2
f (2)=[2]+[ 2]=2+( 2)=0 , so lim f (x) f (2) .
x 2

2
v
52. lim L 1 =L 1 1 =0 . As the velocity approaches the speed of light, the length
 0 2 0
v c c
13
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

approaches 0 .
A left hand limit is necessary since L is not defined for v>c .

2 n
53. Since p(x) is a polynomial, p(x)=a +a x+a x +   +a x . Thus, by the Limit Laws,
0 1 2 n
2 n
lim p(x) = lim a +a x+a x +   +a x
x a 0 1 2 n
x a
2 n
= a0+a1lim x+a2lim x +   +anlim x
x a x a x a
2 n
= a +a a+a a +   +a a = p(a)
0 1 2 n
Thus, for any polynomial p , lim p(x)= p(a) .
x a

p(x)
54. Let r(x)= where p(x) and q(x) are any polynomials, and suppose that q(a) 0 . Thus,
q(x)
lim p(x)
p(x) x a p(a)
lim r(x)=lim = [Limit Law 5] = [ Exercise 53 ] =r(a) .
x a x a q(x) lim q ( x ) q(a)
x a

2 2
55. Observe that 0 f (x) x for all x , and lim 0=0=lim x . So, by the Squeeze Theorem,
x 0 x 0
lim f (x)=0 .
x 0

56. Let f (x)=[ x] and g(x)= [ x] . Then lim f (x) and lim g(x) do not exist (Example 10) but
x 3 x 3
lim f (x)+g(x) =lim ( [ x] [ x]) =lim 0=0 .
x 3 x 3 x 3

57. Let f (x)=H(x) and g(x)=1 H(x) , where H is the Heaviside function defined in Exercise 1.3.59.
Thus, either f or g is 0 for any value of x . Then lim f (x) and lim g(x) do not exist, but
x 0 x 0
lim f (x)g(x) =lim 0=0 .
x 0 x 0

58.
6 x  2 6 x  2 6 x +2 3 x +1
lim =lim  
x 2 3 x  1 x 2 3 x  1 6 x +2 3 x +1

14
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws

2
=lim
( )  22  3 x +1
6 x
=lim
6 x 4

3 x +1
2 2 6 x +2 3 x 1 6 x +2
x 2 ( 3 x )  1 x 2

( 2 x ) ( 3 x +1 ) 3 x +1 1
=lim =lim =
x 2 ( 2 x ) ( 6 x +2 ) x 2 6 x +2 2

59. Since the denominator approaches 0 as x  2 , the limit will exist only if the numerator also
approaches 0 as x  2 . In order for this to happen, we need lim ( 3x2+ax+a+3) =0

x  2
2
3( 2) +a( 2)+a+3=0
12 2a+a+3=0
a=15 . With a=15 , the limit becomes
2
3x +15x+18 3(x+2)(x+3) 3(x+3) 3( 2+3) 3
lim =lim =lim = = = 1 .
x  2
2
x  2 (x 1)(x+2) x  2 x 1  2 1  3
x +x 2

60. Solution 1: First, we find the coordinates of P and Q as functions of r . Then we can find the
equation of the line determined by these two points, and thus find the x intercept (the point R ), and
take the limit as r 0 .
2 2 2
The coordinates of P are ( 0,r ) . The point Q is the point of intersection of the two circles x +y =r
2 2 2 2 2 2 1 2
and (x 1) +y =1 . Eliminating y from these equations, we get r  x =1 (x 1)
r =1+2x 1
x= r .
2
Substituting back into the equation of the shrinking circle to find the y coordinate, we get
1 2 2 2 2 2 2 1 2 1 2
r +y =r
y =r 1 r
y=r 1 r (the positive y value). So the coordinates of
2 4 4
1 2 1 2
Q are r ,r 1 r . The equation of the line joining P and Q is thus
2 4
1 2
r 1 r  r
4
y r= ( x 0 ) . We set y=0 in order to find the x intercept, and get
1 2
r  0
2
1 2 1 2 1 2
r  r 1 r +1
2 2 4 1 2
x= r = =2 1 r +1 .
1 2 1 2 4
r 1 r  1 1 r  1
4 4
+ 1 2
Now we take the limit as r 0 : lim x=lim 2 1 r +1 =lim 2 ( 1 +1 ) =4 .
+ + 4 +
r 0 r 0 r 0
So the limiting position of R is the point ( 4,0 ) .
15
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.3 Calculating Limits Using the Limit Laws


Solution 2: We add a few lines to the diagram, as shown. Note that  PQS=90 (subtended by
diameter PS ).

So  SQR=90 = OQT (subtended by diameter OT ). It follows that  OQS= TQR . Also

 PSQ=90   SPQ= ORP . Since QOS is isosceles, so is QTR , implying that QT =TR . As the
circle C shrinks, the point Q plainly approaches the origin, so the point R must approach a point
2
twice as far from the origin as T , that is, the point ( 4,0 ) , as above.

16
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

1. (a) To have 5x+3 within a distance of 0.1 of 13 , we must have 12.9 5x+3 13.1
9.9 5x 10.1 1.98 x 2.02 . Thus, x must be within 0.02 units of 2 so that 5x+3 is within 0.1 of
13 .
(b) Use 0.01 in place of 0.1 in part (a) to obtain 0.002 .

2. (a) To have 6x 1 within a distance of 0.01 of 29 , we must have 28.99 6x 1 29.01
29.99 6x 30.01 4.9983 x 5.0016 . Thus, x must be within 0.0016 units of 5 so that 6x 1 is
within 0.01 of 29 .
(b) As in part (a) with 0.001 in place of 0.01 , we obtain 0.00016 .
(c) As in part (a) with 0.0001 in place of 0.01 , we obtain 0.000016 .

10 4
3. On the left side of x=2 , we need x 2 <  2 = . On the right side, we need
7 7
10 4
x 2 <  2 = . For both of these conditions to be satisfied at once, we need the more
3 3
4 4
restrictive of the two to hold, that is, x 2 < . So we can choose  = , or any smaller positive
7 7
number.

4. On the left side, we need x 5 < 4 5 =1 . On the right side, we need x 5 < 5.7 5 =0.7 . For both
conditions to be satisfied at once, we need the more restrictive condition to hold; that is, x 5 <0.7 .
So we can choose  =0.7 , or any smaller positive number.

5. The leftmost question mark is the solution of x =1.6 and the rightmost, x =2.4 . So the values are
2 2
1.6 =2.56 and 2.4 =5.76 . On the left side, we need x 4 < 2.56 4 =1.44 . On the right side, we need
x 4 < 5.76 4 =1.76 . To satisfy both conditions, we need the more restrictive condition to hold   
namely, x 4 <1.44 . Thus, we can choose  =1.44 , or any smaller positive number.

2 1 1
6. The left hand question mark is the positive solution of x = , that is, x= , and the right hand
2 2
2 3 3
question mark is the positive solution of x = , that is, x= . On the left side, we need
2 2
1
x 1 <  1  0.292 (rounding down to be safe). On the right side, we need
2
3
x 1 <  1  0.224 . The more restrictive of these two conditions must apply, so we choose
2
 =0.224 (or any smaller positive number).

7. 4x+1  3 <0.5  2.5< 4x+1 <3.5 . We plot the three parts of this inequality on the same screen
1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

and identify the x  coordinates of the points of intersection using the cursor. It appears that the
inequality holds for 1.3125 x 2.8125 . Since 2 1.3125 =0.6875 and 2 2.8125 =0.8125 , we
choose 0< <min { 0.6875,0.8125} =0.6875 .

1
8. sin x <0.1  0.4<sin x<0.6 . From the graph, we see that for this inequality to hold, we
2
need 0.42 x 0.64 . So since 0.5 0.42 =0.08 and 0.5 0.64 =0.14 , we choose
0<  min { 0.08,0.14} =0.08 .

9. For  =1 , the definition of a limit requires that we find  such that ( 4+x 3x3)  2 <1
3 3
1<4+x 3x <3 whenever 0< x 1 < . If we plot the graphs of y=1 , y=4+x 3x and y=3 on the same
screen, we see that we need 0.86 x 1.11 . So since 1 0.86 =0.14 and 1 1.11 =0.11 , we choose
 =0.11 (or any smaller positive number). For  =0.1 , we must find  such that ( 4+x 3x3)  2 <0.1
3
 1.9<4+x 3x <2.1 whenever 0< x 1 < . From the graph, we see that we need 0.988 x 1.012 .
So since 1 0.988 =0.012 and 1 1.012 =0.012 , we choose  =0.012 (or any smaller positive
number) for the inequality to hold.

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

x
e 1
10. For  =0.5 , the definition of a limit requires that we find  such that  1 <0.5
x
x x
e 1 e 1
0.5< <1.5 whenever 0< x 0 < . If we plot the graphs of y=0.5 , y= , and y=1.5 on the
x x
same screen, we see that we need  1.59 x 0.76 . So since 0 (  1.59 ) =1.59 and 0 0.76 =0.76 ,
we choose  =0.76 (or any smaller positive number). For  =0.1 , we must find  such that
x x
e 1 e 1
 1 <0.1  0.9< <1.1 whenever 0< x 0 < . From the graph, we see that we need
x x
 0.21 x 0.18 . So since 0 (  0.21 ) =0.21 and 0 0.18 =0.18 , we choose  =0.18 (or any smaller
positive number) for the inequality to hold.

x
11. From the graph, we see that >100 whenever 0.93 x 1.07 . So since
( x +1) ( x 1 )
2 2

1 0.93 =0.07 and 1 1.07 =0.07 , we can take  =0.07 (or any smaller positive number).

12. For M=100 , we need  0.0997<x<0 or 0<x<0.0997 . Thus, we choose  =0.0997 (or any smaller
2
positive number) so that if 0< x < , then cot x>100 .

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

For M=1000 , we need  0.0316<x<0 or 0<x<0.0316 . Thus, we choose  =0.0316 (or any smaller
2
positive number) so that if 0< x < , then cot x>1000 .

2 2 2 2 1000
13. (a) A= r and A=1000 cm   r =1000 r = 

1000
r= r>0  17.8412 cm.

2 2
(b) A 1000  5  5  r  1000 5 1000 5  r  1000+5
995 1005 1000 995
 r  17.7966 r 17.8858 .   0.04466 and
   
1005 1000
  0.04455 . So if the machinist gets the radius within 0.0445 cm of 17.8412 ,
 
2
the area will be within 5 cm of 1000 .
(c) x is the radius, f (x) is the area, a is the target radius given in part (a), L is the target area (1000) ,
 is the tolerance in the area ( 5 ), and  is the tolerance in the radius given in part (b).

2 2
14. (a) T =0.1w +2.155w+20 and T =200 0.1w +2.155w+20=200 [ by the quadratic formula or
from the graph] w 33.0 watts ( w>0 )

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

(b) From the graph, 199 T  201 32.89<w<33.11 .

(c) x is the input power, f (x) is the temperature, a is the target input power given in part (a), L is the
target temperature ( 200 ),  is the tolerance in the temperature ( 1 ), and  is the tolerance in the
power input in watts indicated in part (b) ( 0.11 watts).

15. Given  >0 , we need  >0 such that if 0< x 1 < , then (2x+3) 5 < . But (2x+3) 5 < 
2x 2 <  2 x 1 <  x 1 < /2 . So if we choose  = /2 , then 0< x 1 <  (2x+3) 5 < . Thus,
lim (2x+3)=5 by the definition of a limit.
x 1

1
16. Given  >0 , we need  >0 such that if 0< x ( 2) < , then ( x+3) 2 < . But
2
1 1 1
( x+3) 2 <  x+1 <  x+2 <  x ( 2) <2 . So if we choose  =2 , then
2 2 2
1 1
0< x ( 2) <  ( x+3) 2 < . Thus, lim ( x+3)=2 by the definition of a limit.
2 x  2 2

17. Given  >0 , we need  >0 such that if 0< x ( 3) < , then (1 4x) 13 < . But (1 4x) 13 < 
 4x 12 <   4 x+3 <  x ( 3) < /4 . So if we choose  = /4 , then 0< x ( 3) < 
(1 4x) 13 < . Thus, lim (1 4x)=13 by the definition of a limit.
x  3

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

18. Given  >0 , we need  >0 such that if 0< x 4 < , then (7 3x) ( 5) < . But (7 3x) ( 5) < 
 3x+12 <   3 x 4 <  x 4 < /3 . So if we choose  = /3 , then 0< x 4 < 
(7 3x) ( 5) < . Thus, lim (7 3x)= 5 by the definition of a limit.
x 4

x 3 1
19. Given  >0 , we need  >0 such that if 0< x 3 < , then  <  x 3 <  x 3 <5 .
5 5 5
x 3 x 3
So choose  =5 . Then 0< x 3 <  x 3 <5  <   < . By the definition of a
5 5 5
x 3
limit, lim = .
x 3 5 5

x 9 x 3
20. Given  >0 , we need  >0 such that if 0< x 6 < , then +3  <   < 
4 2 4 2
1 x 6
x 6 <  x 6 <4 . So choose  =4 . Then 0< x 6 <  x 6 <4  < 
4 4
x 6 x 9 x 9
 <  +3  < . By the definition of a limit, lim +3 = .
4 4 4 2 x 6 4 2

3
21. Given  >0 , we need  >0 such that if 0< x (  5) < , then 4 x  7 < 
5

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

3 3 5 5
 x 3 <  x+5 <  x (  5) <  . So choose  =  . Then x (  5) < 
5 5 3 3
3 3
4 x  7 < . Thus, lim 4 x =7 by the definition of a limit.
5 x  5 5

2
x +x 12
22. Given  >0 , we need  >0 such that if 0< x 3 < , then  7 < . Notice that if
x 3
2
x +x 12 (x+4)(x 3)
0< x 3 , then x 3 , so = =x+4 . Thus, when 0< x 3 , we have
x 3 x 3
2
x +x 12
 7 <  (x+4) 7 <  x 3 < . We take  = and see that 0< x 3 < 
x 3
2 2
x +x 12 x +x 12
 7 < . By the definition of a limit, lim =7 .
x 3 x 3 x 3

23. Given  >0 , we need  >0 such that if 0< x a < , then x a < . So  = will work.

24. Given  >0 , we need  >0 such that if 0< x a < , then c c < . But c c =0 , so this will be
true no matter what  we pick.

2 2
25. Given  >0 , we need  >0 such that if 0< x 0 < , then x  0 <  x <  x <  . Take  = 
2 2
. Then 0< x 0 <  x  0 < . Thus, lim x =0 by the definition of a limit.
x 0

3 3 3
26. Given  >0 , we need  >0 such that if 0< x 0 < , then x  0 <  x <  x <  . Take
3 3 3 3
 =  . Then 0< x 0 <  x  0 < = . Thus, lim x =0 by the definition of a limit.
x 0

27. Given  >0 , we need  >0 such that if 0< x 0 < , then x  0 < . But x = x . So this is
true if we pick  = . Thus, lim x =0 by the definition of a limit.
x 0

4 4 4
28. Given  >0 , we need  >0 such that if 9  <x<9 , then 9 x  0 <  9 x <  9 x< 
4 4 4 4
9  <x<9 . So take  = . Then 9  <x<9 9 x  0 < . Thus, lim 9 x =0 by the definition of a

x 9
limit.

29. Given  >0 , we need  >0 such that if 0< x 2 < , then ( x2 4x+5)  1 < 
2
x  4x+4 < 
7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

2
(x 2) < . So take  =  . Then 0< x 2 <  x 2 <  
2
(x 2) < . Thus, lim ( x2 4x+5) =1
x 2
by the definition of a limit.

30. Given  >0 , we need  >0 such that if 0< x 3 < , then x +x 4  8 <  x +x 12 <  ( 2
) 2

(x 3)(x+4) < . Notice that if x 3 <1 , then  1<x 3<1 6<x+4<8 x+4 <8 . So take
 =min { 1, /8} . Then 0< x 3 <  (x 3)(x+4)  8(x 3) =8
x 3 <8   . Thus, lim ( x2+x 4) =8
x 3
by the definition of a limit.

31. Given  >0 , we need  >0 such that if 0< x ( 2) < , then ( x2 1)  3 < or upon simplifying we
2
need x  4 < whenever 0< x+2 < . Notice that if x+2 <1 , then  1<x+2<1  5<x 2< 3
x 2 <5 . So take  =min { /5,1} . Then 0< x+2 <  x 2 <5 and x+2 < /5 , so
( x2 1)  3 = (x+2)(x 2) = x+2 x 2 <( /5)(5)= . Thus, by the definition of a limit, lim ( x2 1) =3
x  2
.

3
32. Given  >0 , we need  >0 such that if 0< x 2 < , then x  8 < . Now
3
x  8 = (x 2) x +2x+4 ( 2
)
. If x 2 <1 , that is, 1<x<3 , then x +2x+4<3 +2(3)+4=19 and so
2 2


3
x  8 = x 2 ( 2
)
x +2x+4 <19 x 2 . So if we take  =min 1,

19 { }
, then 0< x 2 < 
3
x  8 = x 2 ( 2
) 3
x +2x+4 <
19= . Thus, by the definition of a limit, lim x =8 .
19 x 2


33. Given  >0 , we let  =min

{ }
2,
8
. If 0< x 3 < , then x 3 <2

 2<x 3<2 4<x+3<8
2 2
x+3 <8 . Also x 3 < , so x  9 = x+3 x 3 <8
= . Thus, lim x =9 .
8 8 x 3

34. From the figure, our choices for  are  =3 9  and  = 9+  3 . The largest possible choice
1 2
for  is the minimum value of { , }; that is,  =min{ , }= = 9+  3 .
1 2 1 2 2

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

35. (a) The points of intersection in the graph are (x ,2.6) and (x ,3.4) with x  0.891 and x  1.093 .
1 2 1 2
Thus, we can take  to be the smaller of 1 x and x  1 . So  =x  1 0.093 .
1 2 2

3
(b) Solving x +x+1=3+ gives us two nonreal complex roots and one real root, which is
2/3

x( )=
( 216+108 +12 336+324 +81 )  12 . Thus,  =x( 2
) 1 .
1/3
6 ( 216+108 +12 336+324 +81 )
2

(c) If  =0.4 , then x( ) 1.093272342 and  =x( ) 1 0.093 , which agrees with our answer in part
(a).

36. 1. Guessing a value for  Let  >0 be given. We have to find a number  >0 such that
1 1 1 1 2 x x 2
 < whenever 0< x 2 < . But  = = < . We find a positive
x 2 x 2 2x 2x
1 x 2 
constant C such that <C  <C x 2 and we can make C x 2 < by taking x 2 < =
2x 2x C
1 1 1 1 1 1 1
. We restrict x to lie in the interval x 2 <1 1<x<3 so 1> >  < <  < . So
x 3 6 2x 2 2x 2
1
C= is suitable. Thus, we should choose  =min { 1,2 } .
2
2. Showing that  works Given  >0 we let  =min { 1,2 } . If 0< x 2 < , then x 2 <1 1<x<3
1 1 1 1 x 2 1
< (as in part 1). Also x 2 <2 , so  = <
2 = . This shows that
2x 2 x 2 2x 2

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

1
lim (1/x)= .
x 2 2

37. 1. Guessing a value for  Given  >0 , we must find  >0 such that x a < whenever
x a
0< x a < . But x  a = < (from the hint). Now if we can find a positive constant C
x+ a
x a x a
such that x + a >C then < < , and we take x a <C . We can find this number by
x+ a C
1 1 1 1 3
restricting x to lie in some interval centered at a . If x a < a , then  a<x a< a  a<x< a
2 2 2 2 2
1 1
 x+ a> a + a , and so C= a + a is a suitable choice for the constant. So
2 2

x a <
1
2
a+ a  . This suggests that we let  =min { 1
2
a,
1
2
a+ a  } .

2. Showing that  works Given  >0 , we let  =min { 1

2
a,
1
2
a+ a  } . If 0< x a < ,

1 1 1
then x a < a x+ a> a + a (as in part 1). Also x a < a + a  , so
2 2 2
x a ( a/2 + a ) 
x a = < = . Therefore, lim x = a by the definition of a limit.
x+ a ( a/2 + a ) x a
1 1
38. Suppose that lim H(t)=L . Given  = , there exists  >0 such that 0< t <  H(t) L < 
t 0 2 2
1 1 1 1 1
L <H(t)<L+ . For 0<t< , H(t)=1 , so 1<L+  L> . For   <t<0 , H(t)=0 , so L <0
2 2 2 2 2
1 1
L< . This contradicts L> . Therefore, lim H(t) does not exist.
2 2 t 0
1 1
39. Suppose that lim f (x)=L . Given  = , there exists  >0 such that 0< x <  f (x) L < .
x 0 2 2
1 1
Take any rational number r with 0< r < . Then f (r)=0 , so 0 L < , so L L < . Now take
2 2
1 1 1
any irrational number s with 0< s < . Then f (s)=1 , so 1 L < . Hence, 1 L< , so L> . This
2 2 2
1
contradicts L< , so lim f (x) does not exist.
2 x 0

40. First suppose that lim f (x)=L . Then, given  >0 there exists  >0 so that 0< x a < 
x a
f (x) L < . Then a  <x<a 0< x a < so f (x) L < . Thus,
10
 Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

lim f (x)=L . Also a<x<a+  0< x a < so f (x) L < . Hence, lim f (x)=L .
 +
x a x a
Now suppose lim f (x)=L=lim f (x) . Let  >0 be given. Since lim f (x)=L , there exists  >0 so that
 +  1
x a x a x a
a  <x<a f (x) L < . Since lim f (x)=L , there exists  >0 so that a<x<a+  f (x) L < . Let
1 + 2 2
x a
 be the smaller of  and  . Then 0< x a <  a  <x<a or a<x<a+ so f (x) L < . Hence,
1 2 1 2
lim f (x)=L . So we have proved that lim f (x)=L lim f (x)=L=lim f (x) .
x a x a  +
x a x a

1 4 1 1 1
41. >10 , 000 (x+3) <  x+3 <  x ( 3) <
4 10,000 4
10,000 10
(x+3)

4 1 4 1
42. Given M>0 , we need  >0 such that 0< x+3 <  1/(x+3) >M . Now >M (x+3) < 
4 M
(x+3)
1 1 1 1 1
x+3 < 4
. So take  = 4
. Then 0< x+3 < = 4
 >M , so lim = .
4 4
M M M (x+3) x  3 (x+3)

ln x M
43. Given M<0 we need  >0 so that ln x<M whenever 0<x< ; that is, x=e <e whenever 0<x< .
M M M
This suggests that we take  =e . If 0<x<e , then ln x<ln e =M . By the definition of a limit,
lim ln x=  .
+
x 0

44. (a) Let M be given. Since lim f (x)= , there exists  >0 such that 0< x a <  f (x)>M+1 c .
1 1
x a
Since lim g(x)=c , there exists  >0 such that 0< x a <  g(x) c <1 g(x)>c 1 . Let  be the
2 2
x a
smaller of  and  . Then 0< x a <  f (x)+g(x)>(M+1 c)+(c 1)=M . Thus, lim f (x)+g(x) = .
1 2
x a
(b) Let M>0 be given. Since lim g(x)=c>0 , there exists  >0 such that 0< x a <  g(x) c <c/2
1 1
x a
 g(x)>c/2 . Since lim f (x)= , there exists  >0 such that 0< x a <  f (x)>2M/c . Let
2 2
x a
2M c
 =min  , { } . Then 0< x a <  f (x) g(x)> =M , so lim f (x) g(x)= .
1 2 c 2 x a
(c) Let N<0 be given. Since lim g(x)=c<0 , there exists  >0 such that 0< x a <  g(x) c < c/2
1 1
x a
 g(x)<c/2 . Since lim f (x)= , there exists  >0 such that 0< x a <  f (x)>2N/c . (Note that
2 2
x a

11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.4 The Precise Definition of a Limit

c<0 and N<0 2N/c>0 .) Let  =min  , { } . Then 0< x a <  f (x)>2N/c
1 2
2N c
f (x) g(x)<
=N , so lim f (x) g(x)=  .
c 2 x a

12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

1. From Definition 1, lim f (x)= f (4) .

x 4

2. The graph of f has no hole, jump, or vertical asymptote.

3. (a) The following are the numbers at which f is discontinuous and the type of discontinuity at that
number:  4 (removable),  2 ( jump), 2 ( jump), 4 (infinite).
(b) f is continuous from the left at  2 since lim f (x)= f ( 2) . f is continuous from the right at 2 and

x  2
4 since lim f (x)= f (2) and lim f (x)= f (4) . It is continuous from neither side at  4 since f ( 4) is
+ +
x 2 x 4
undefined.

4. g is continuous on  4, 2 ) , (  2,2 ) , 2,4 ) , ( 4,6 ) , and ( 6,8 ) .

5. The graph of y= f (x) must have a discontinuity at x=3 and must show that lim f (x)= f (3) .

x 3

6.

7. (a)

(b) There are discontinuities at times t=1 , 2 , 3 , and 4 . A person parking in the lot would want to
keep in mind that the charge will jump at the beginning of each hour.

8. (a) Continuous; at the location in question, the temperature changes smoothly as time passes,
without any instantaneous jumps from one temperature to another.
1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

(b) Continuous; the temperature at a specific time changes smoothly as the distance due west from
New York City increases, without any instantaneous jumps.
(c) Discontinuous; as the distance due west from New York City increases, the altitude above sea
level may jump from one height to another without going through all of the intermediate values    at
a cliff, for example.
(d) Discontinuous; as the distance traveled increases, the cost of the ride jumps in small increments.
(e) Discontinuous; when the lights are switched on (or off ), the current suddenly changes between 0
and some nonzero value, without passing through all of the intermediate values. This is debatable,
though, depending on your definition of current.

9. Since f and g are continuous functions,

lim 2 f (x) g(x) = 2lim f (x) lim g(x) [by Limit Laws 2 and 3]
x 3 x 3 x 3
= 2 f (3) g(3) [by continuity of f and g at x=3 ]
= 2 5 g(3)=10 g(3)
Since it is given that lim 2 f (x) g(x) =4 , we have 10 g(3)=4 , so g(3)=6 .
x 3

10. lim f (x)=lim ( x2+ )

7 x =lim x +
2 2
lim 7 lim x =4 + 7 4 =16+ 3 = f (4) .
x 4 x 4 x 4 x 4 x 4
By the definition of continuity, f is continuous at a=4 .

11. lim f (x)=lim ( x+2x3) 4= lim x+2lim x

3 4
=  1+2( 1)
3 4 4
=( 3) =81= f ( 1) .
x  1 x  1 x  1 x  1
By the definition of continuity, f is continuous at a= 1 .

lim x+lim 1
x+1 x 4 x 4 4+1 5
12. lim g(x)=lim = = = =g(4) . So g is continuous at 4 .
2 2 2 31
x 4 x 4 2x  1 2lim x  lim 1 2(4)  1
x 4 x 4

lim (2x+3) 2lim x+lim 3

2x+3 x a x a x a
13. For a>2 , we have lim f (x)=lim = [ Limit Law 5] =
x a x a x 2 lim (x 2) lim x lim 2
x a x a x a
2a+3
[1, 2, and 3] = [ 7 and 8] = f (a) . Thus, f is continuous at x=a for every a in (2, ) ; that is, f is
a 2
continuous on (2, ) .

14. For a<3 , we have lim g(x)=lim 2 3 x =2lim 3 x [ Limit Law 3] =2 lim (3 x) [ 11]
x a x a x a x a

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

=2 lim 3 lim x [ 2] =2 3 a [ 7 and 8] =g(a) , so g is continuous at x=a for every a in (  ,3) .

x a x a
Also, lim g(x)=0=g(3) , so g is continuous from the left at 3 . Thus, g is continuous on (  ,3].

x 3

15. f (x)=ln x 2 is discontinuous at 2 since f (2)=ln 0 is not defined.

16. f (x)= { 1/(x 1)

2
if x 1
if x=1
is discontinuous at 1 because lim f (x) does not exist.
x 1

17. f (x)=
{ e
x
x

2
ifx<0
ifx 0
x
The left hand limit of f at a=0 is lim f (x)=lim e =1 . The right hand limit of f at a=0 is
 
x 0 x 0
2
lim f (x)=lim x =0 . Since these limits are not equal, lim f (x) does not exist and f is discontinuous
+ + x 0
x 0 x 0
at 0.

18.

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

{
2
x x
if x 1
f (x)= 2
x 1 if x=1
1
2
lim f (x) =lim x  x =lim x(x 1)
x 1
x 1 x  1 x 1 (x+1)(x 1)
2

x 1
=lim = ,
x 1 x+1 2

but f (1)=1 , so f is discontinous at 1.

{
2
x  x 12 if x  3 x 4 if x  3
19. f (x)=
 5
x+3 if x= 3
= {  5 if x= 3

So lim f (x)=lim (x 4)= 7 and f ( 3)= 5 .

x  3 x  3
Since lim f (x) f ( 3) , f is discontinuous at  3 .
x  3

20. f (x)=
{ 1+x
4 x
2

2
if x<1
if x 1
2
lim f (x)=lim (1+x )=1+1 =2 and
 
x 1 x 1

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

lim f (x)=lim (4 x)=4 1=3 .

+ +
x 1 x 1
Thus, f is discontinuous at 1 because lim f (x) does not exist.
x 1

x
21. F(x)= 2
is a rational function. So by Theorem 5 (or Theorem 7), F is continuous at every
x +5x+6
{ 2
}
number in its domain, x| x +5x+6 0 = { x|(x+3)(x+2) 0} = { x| x  3, 2} or
(   , 3) (  3, 2 ) (  2, ) .
3 3
22. By Theorem 7, the root function x and the polynomial function 1+x are continuous on R . By
part 4 of Theorem 4, the product G(x)= x 1+x
3
( 3
) is continuous on its domain, R .
2
23. By Theorem 5, the polynomials x and 2x 1 are continuous on (  , ) . By Theorem 7, the root
function x is continuous on [0, )] . By Theorem 9, the composite function 2x 1 is continuous on
1 2 1
its domain, [ , )] . By part 1 of Theorem 4, the sum R(x)=x + 2x 1 is continuous on [ , )] .
2 2

24. By Theorem 7, the trigonometric function sin x and the polynomial function x+1 are continuous
sin x
on R . By part 5 of Theorem 4, h ( x ) = is continuous on its domain, { x| x  1} .
x+1

25. By Theorem 5, the polynomial 5x is continuous on (   ) . By Theorems 9 and 7, sin 5x is ,

x
continuous on (   , ) . By Theorem 7, e is continuous on (   , ) . By part 4 of Theorem 4, the
x
product of e and sin 5x is continuous at all numbers which are in both of their domains, that is, on
(   , ) .
2  1
26. By Theorem 5, the polynomial x  1 is continuous on (   , ) . By Theorem 7, sin is
 1
continuous on its domain,  1,1 . By Theorem 9, sin ( x2 1) is continuous on its domain, which is
{ x| 1
2
} {
x  1
1 = x|0
x
2 ={ x| x

2
} 2}=  2, 2 .
5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

4
27. By Theorem 5, the polynomial t  1 is continuous on (   , ) . By Theorem 7, ln x is continuous
) . By Theorem 9, ln ( t  1 ) is continuous on its domain, which is
4
on its domain, ( 0,
{ t |t 4 1>0} = { t |t 4>1} ={t | t >1} = (   , 1 ) ( 1, ) .
x
28. By Theorem 7, x is continuous on 0, ) . By Theorems 7 and 9, e is continuous on 0, ).
Also by Theorems 7 and 9, cos e ( ) is continuous on
x
0, ).
1
29. The function y= is discontinuous at x=0 because the left and right hand limits at x=0 are
1/x
1+e
different.

2 
30. The function y=tan x is discontinuous at x= + k , where k is any integer. The function
2
( 2
) 2
y=ln tan x is also discontinuous where tan x is 0 , that is, at x= k . So y=ln tan x is ( 2
)

discontinuous at x= n , n any integer.
2

31. Because we are dealing with root functions, 5+ x is continuous on 0, ) , x+5 is continuous
5+ x
on  5, ) , so the quotient f (x)= is continuous on 0, ) . Since f is continuous at x=4 ,
5+x
7
lim f (x)= f (4)= .
x 4 3

32. Because x is continuous on R , sin x is continuous on R , and x+sin x is continuous on R , the

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

composite function f (x)=sin (x+sin x) is continuous on R , so lim f (x)= f ( )=sin ( +sin  )=sin  =0 .
x 

2
2 x x
33. Because x  x is continuous on R , the composite function f (x)=e is continuous on R , so
1 1 0
lim f (x)= f (1)=e =e =1 .
x 1

34. Because arctan is a continuous function, we can apply Theorem 8.

2
x 4 (x+2)(x 2) x+2 2
lim arctan =arctan lim =arctan lim =arctan  0.588
2 3x(x 2) 3x 3
x 2 3x  6x x 2 x 2

35. f (x)=
{ 2
x
x
if x<1
if x 1
2
By Theorem 5, since f (x) equals the polynomial x on (  ,1) , f is continuous on (  ,1) . By
Theorem 7, since f (x) equals the root function x on (1, ), f is continuous on (1, ) . At x=1 ,
2
lim f (x)=lim x =1 and lim f (x)=lim x =1 . Thus, lim f (x) exists and equals 1 . Also, f (1)= 1 =1
  + + x 1
x 1 x 1 x 1 x 1
. Thus, f is continuous at x=1 . We conclude that f is continuous on (  , ).

36. f (x)= {
sin x if x< /4
cos x if x  /4
By Theorem 7, the trigonometric functions are continuous. Since f (x)=sin x on (  , /4) and
f (x)=cos x on ( /4, ) , f is continuous on (  , /4) ( /4, ).

lim f (x)= lim sin x=sin =1/ 2 since the sine function is continuous at  /4. Similarly,
  4
x ( /4 ) x ( /4 )
lim f (x)= lim cos x=1/ 2 by continuity of the cosine function at  /4 . Thus, lim f (x)
+ + x ( /4 )
x ( /4 ) x ( /4 )
exists and equals 1/ 2 , which agrees with the value f ( /4) . Therefore, f is continuous at  /4 , so f
is continuous on (  , ) .

{
2
1+x if x
0
37. f (x)= 2 x if 0<x
2
2 if x>2
(x 2)
f is continuous on (  ,0) , (0,2) , and (2, ) since it is a polynomial on each of these intervals. Now

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

2
lim f (x)=lim (1+x )=1 and
 
x 0 x 0

lim f (x)=lim (2 x)=2, so f is discontinuous at 0 . Since f (0)=1 , f is continuous from the left at 0.
+ +
x 0 x 0
2
Also, lim f (x)=lim (2 x)=0,lim f (x)=lim (x 2) =0 , and f (2)=0 , so f is continuous at 2. The only
  + +
x 2 x 2 x 2 x 2
number at which f is discontinuous is 0 .

38. f (x)=
{
x+1
1/x
if x
1
if 1<x<3
x 3 if x 3
f is continuous on (  ,1) , (1,3) , and (3, ) , where it is a polynomial, a rational function, and a
composite of a root function with a polynomial, respectively. Now lim f (x)=lim (x+1)=2 and
 
x 1 x 1
lim f (x)=lim ( 1/x ) =1 , so f is discontinuous at 1 .
+ +
x 1 x 1

Since f (1)=2 , f is continuous from the left at 1 . Also, lim f (x)=lim ( 1/x ) =1/3 , and
 
x 3 x 3
lim f (x)=lim x 3 =0= f (3) , so f is discontinuous at 3 , but it is continuous from the right at 3.
+ +
x 3 x 3

39. f (x)=
{
e
x+2 if x<0
x
if 0
x
1
2 x if x>1
f is continuous on (   ,0 ) and ( 1, ) since on each of these intervals it is a polynomial; it is
continuous on ( 0,1 ) since it is an exponential. Now
8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

x
lim f (x)=lim (x+2)=2 and lim f (x)=lim e =1 , so f is discontinuous at 0 . Since f (0)=1 , f is
  + +
x 0 x 0 x 0 x 0
x
continuous from the right at 0 . Also lim f (x)=lim e =e and lim f (x)=lim (2 x)=1 , so f is
  + +
x 1 x 1 x 1 x 1
discontinuous at 1 . Since f (1)=e , f is continuous from the left at 1 .

40. By Theorem 5, each piece of F is continuous on its domain. We need to check for continuity at
r=R .
GMr GM GM GM GM GM
lim F(r)=lim = and lim F(r)=lim = , so lim F(r)= . Since F(R)= ,
  3 2 + + 2 2 2 2
r R
r R r R R R r R r R r R R R
F is continuous at R . Therefore, F is a continuous function of r .

41. f is continuous on (   ,3) and ( 3, ) . Now lim f (x)=lim ( cx+1 ) =3c+1 and
 
x 3 x 3
1
lim f (x)=lim
+ +
( cx2 1) =9c 1 . So f is continuous  3c+1=9c 1 6c=2 c=
3
. Thus, for f to be
x 3 x 3
1
continuous on (   , ) , c= .
3

2 2
42. The functions x  c and cx+20 , considered on the intervals (   ,4 ) and 4, ) respectively, are
continuous for any value of c . So the only possible discontinuity is at x=4 . For the function to be
continuous at x=4 , the left hand and right hand limits must be the same. Now
lim g(x)=lim
 
( x2 c2) =16 c2 and lim g(x)=lim (cx+20)=4c+20=g(4) . Thus, 16 c2=4c+20
+ +
x 4 x 4 x 4 x 4
2
c +4c+4=0 c= 2 .

2
x  2x 8 (x 4)(x+2)
43. (a) f (x)= = has a removable discontinuity at  2 because g(x)=x 4 is
x+2 x+2
continuous on R and f (x)=g(x) for x  2 . [The discontinuity is removed by defining f ( 2)= 6 .]
(b)

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

x 7
f (x)= lim f (x)= 1 and lim f (x)=1 . Thus, lim f (x) does not exist, so the discontinuity is
x 7  + x 7
x 7 x 7
not removable. (It is a jump discontinuity.)
3 2
x +64 (x+4)(x  4x+16) 2
(c) f (x)= = has a removable discontinuity at  4 because g(x)=x  4x+16 is
x+4 x+4
continuous on R and f (x)=g(x) for x  4 .[The discontinuity is removed by defining f ( 4)=48 .]
3 x 3 x 1
(d) f (x)= = has a removable discontinuity at 9 because g(x)= is
9 x ( 3 x ) ( 3+ x ) 3+ x
1
continuous on R and f (x)=g(x) for x 9 . [The discontinuity is removed by defining f (9)= .]
6

44.

f does not satisfy the conclusion of the Intermediate Value Theorem.

f does satisfy the conclusion of the Intermediate Value Theorem.

3 2
45. f (x)=x  x +x is continuous on the interval 2,3 , f (2)=6 , and f (3)=21 . Since 6<10<21 , there is
a number c in ( 2,3) such that f (c)=10 by the Intermediate Value Theorem.

2
46. f (x)=x is continuous on the interval 1,2 , f (1)=1 , and f (2)=4 . Since 1<2<4 , there is a
2
number c in ( 1,2 ) such that f (c)=c =2 by the Intermediate Value Theorem.

4
47. f (x)=x +x 3 is continuous on the interval [1,2], f (1)= 1 , and f (2)=15 . Since  1<0<15 , there is a
number c in (1,2) such that f (c)=0 by the Intermediate Value Theorem. Thus, there is a root of the
4
equation x +x 3=0 in the interval (1,2).

3
48. f (x)= x +x 1 is continuous on the interval [0,1], f (0)= 1 , and f (1)=1 . Since  1<0<1 , there is a
number c in (0,1) such that f (c)=0 by the Intermediate Value Theorem. Thus, there is a root of the
10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

3 3
equation x +x 1=0 , or x =1 x , in the interval (0,1).

49. f (x)=cos x x is continuous on the interval 0,1 , f (0)=1 , and f (1)=cos 1 1  0.46 . Since
 0.46<0<1 , there is a number c in ( 0,1 ) such that f (c)=0 by the Intermediate Value Theorem. Thus,
there is a root of the equation cos x x=0 , or cos x=x , in the interval ( 0,1 ) .

 x  1  2
50. f (x)=ln x e is continuous on the interval 1,2 , f (1)= e   0.37 , and f (2)=ln 2 e  0.56 .
Since  0.37<0<0.56 , there is a number c in ( 1,2 ) such that f (c)=0 by the Intermediate Value
 x  x
Theorem. Thus, there is a root of the equation ln x e =0 , or ln x=e , in the interval ( 1,2 ) .

x
51. (a) f (x)=e +x 2 is continuous on the interval 0,1 , f (0)= 1<0 , and f (1)=e 1 1.72>0 . Since
 1<0<1.72 , there is a number c in ( 0,1 ) such that f (c)=0 by the Intermediate Value Theorem. Thus,
x x
there is a root of the equation e +x 2=0 , or e =2 x , in the interval ( 0,1 ) .
(b) f (0.44)  0.007<0 and f (0.45) 0.018>0 , so there is a root between 0.44 and 0.45 .

52. (a) f (x)=sin x 2+x is continuous on 0,2 , f (0)= 2 , and f (2)=sin 2 0.91 . Since  2<0<0.91 ,
there is a number c in ( 0,2 ) such that f (c)=0 by the Intermediate Value Theorem. Thus, there is a root
of the equation sin x 2+x=0 , or sin x=2 x , in the interval ( 0,2 ) .
(b) f (1.10)  0.009<0 and f (1.11) 0.006>0 , so there is a root between 1.10 and 1.11 .

5 2 5 2 5 2
53. (a) Let f (x)=x  x  4 . Then f (1)=1  1  4= 4<0 and f (2)=2  2  4=24>0 . So by the Intermediate
5 2
Value Theorem, there is a number c in ( 1,2 ) such that f (c)=c  c  4=0 .
(b) We can see from the graphs that, correct to three decimal places, the root is x 1.434 .

1 1 8
54. (a) Let f (x)= x 5  . Then f (5)= <0 and f (6)= >0 , and f is continuous on 5, ) . So
x+3 8 9
11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

by the Intermediate Value Theorem, there is a number c in ( 5,6 ) such that f (c)=0 . This implies that
1
= c 5 .
c+3
(b) Using the intersect feature of the graphing device, we find that the root of the equation is x=5.016
, correct to three decimal places.

55. ( ) If f is continuous at a , then by Theorem 8 with g(h)=a+h , we have

h 0 (
lim f (a+h)= f lim ( a+h ) = f (a) .
h 0 )
( ) Let  >0 . Since lim f (a+h)= f (a) , there exists  >0 such that 0< h < f (a+h) f (a) < . So
h 0
if 0< x a < , then f (x) f (a) = f (a+(x a)) f (a) < . Thus, lim f (x)= f (a) and so f is continuous
x a
at a .

56.
lim sin (a+h) =lim ( sin acos h+cos asin h ) =lim ( sin acos h ) +lim ( cos asin h )
h 0 h 0 h 0 h 0

(
= lim sin a
h 0 ) ( lim cos h) +( lim cos a) ( lim sin h)
h 0 h 0 h 0
=(sin a)(1)+(cos a)(0)=sin a

57. As in the previous exercise, we must show that lim cos (a+h)=cos a to prove that the cosine
h 0
function is continuous.
lim cos (a+h) =lim ( cos acos h sin asin h )
h 0 h 0
=lim ( cos acos h )  lim ( sin asin h )
h 0 h 0

(
= lim cos a
h 0 ) ( lim cos h)  ( lim sin a) ( lim sin h)
h 0 h 0 h 0
=(cos a)(1) (sin a)(0)=cos a

58. (a) Since f is continuous at a , lim f (x)= f (a) .Thus, using the Constant Multiple Law of Limits,
x a

12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

we have lim (cf )(x)=lim cf (x)=clim f (x)=cf (a)=(cf )(a) . Therefore, cf is continuous at a .
x a x a x a
(b) Since f and g are continuous at a , lim f (x)= f (a) and lim g(x)=g(a) . Since g(a) 0 , we can use
x a x a
lim f (x)
f f (x) x a f (a) f f
the Quotient Law of Limits: lim (x)=lim = = = (a) . Thus, is
x a g x a g(x) lim g(x) g(a) g g
x a
continuous at a .

59. f (x)= {(
0 if x is rational
1 if x is irrational
is continuous nowhere. For, given any number a and any  >0 ,
the interval a  ,a+ ) contains both infinitely many rational and infinitely many irrational numbers.
Since f (a)=0 or 1 , there are infinitely many numbers x with 0< x a < and f (x) f (a) =1 . Thus,
lim f (x) f (a) . [In fact lim f (x) does not even exist.]
x a x a

60. g(x)= {
0 if x is rational
x if x is irrational
is continuous at 0 . To see why, note that  x
g(x)
x , so by
the Squeeze Theorem lim g(x)=0=g(0) . But g is continuous nowhere else. For if a 0 and  >0 , the
x 0
interval ( a  ,a+ ) contains both infinitely many rational and infinitely many irrational numbers.
Since g(a)=0 or a , there are infinitely many numbers x with 0< x a < and g(x) g(a) > a /2 .
Thus, lim g(x) g(a) .
x a

3 3
61. If there is such a number, it satisfies the equation x +1=x x  x+1=0 . Let the left hand side of
this equation be called f (x) . Now f ( 2)= 5<0 , and f ( 1)=1>0 . Note also that f (x) is a polynomial,
and thus continuous. So by the Intermediate Value Theorem, there is a number c between  2 and  1
3
such that f (c)=0 , so that c=c +1 .

62. (a) lim F(x)=0 and lim F(x)=0 , so lim F(x)=0 , which is F(0) , and hence F is continuous at
+  x 0
x 0 x 0
x=a if a=0 . For a>0 , lim F(x)=lim x=a=F(a) . For a<0 , lim F(x)=lim ( x)= a=F(a) . Thus, F is
x a x a x a x a
continuous at x=a ; that is, continuous everywhere.
(b) Assume that f is continuous on the interval I . Then for a I , lim f (x) = lim f (x) = f (a) by
x a x a
Theorem 8. (If a is an endpoint of I , use the appropriate one sided limit.) So f is continuous on I .
(c) No, the converse is false. For example, the function f (x)=
x=0 , but f (x) =1 is continuous on R .
1 if x 0
 1 if x<0
is not continuous at {
13
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.5 Continuity

63. Define u(t) to be the monk’s distance from the monastery, as a function of time, on the first day,
and define d(t) to be his distance from the monastery, as a function of time, on the second day. Let D
be the distance from the monastery to the top of the mountain. From the given information we know
that u(0)=0 , u(12)=D , d(0)=D and d(12)=0 . Now consider the function u d , which is clearly
continuous. We calculate that (u d)(0)= D and (u d)(12)=D . So by the Intermediate Value Theorem,
there must be some time t between 0 and 12 such that (u d)(t )=0 u(t )=d(t ) . So at time t after
0 0 0 0 0
7:00 A.M., the monk will be at the same place on both days.

14
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

1. (a) As x becomes large, the values of f (x) approach 5 .

(b) As x becomes large negative, the values of f (x) approach 3 .

2. (a) The graph of a function can intersect a vertical asymptote in the sense that it can meet but not
cross it.

The graph of a function can intersect a horizontal asymptote. It can even intersect its horizontal
asymptote an infinite number of times.

(b) The graph of a function can have 0 , 1 , or 2 horizontal asymptotes. Representative examples are
shown.

No horizontal One horizontal Two horizontal

asymptote asymptote asymptotes

3. (a) lim f (x)=

x 2

(b) lim f (x)=


x  1

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

(c) lim f (x)= 

+
x  1

(d) lim f (x)=1

x 

(e) lim f (x)=2

x  

(f) Vertical: x= 1 , x=2 ; Horizontal: y=1 , y=2

4. (a) lim g(x)=2

x 

(b) lim g(x)= 2

x  

(c) lim g(x)=

x 3

(d) lim g(x)= 

x 0

(e) lim g(x)= 

+
x  2

(f) Vertical: x= 2 , x=0 , x=3 ; Horizontal: y= 2 , y=2

5. f (0)=0 , f (1)=1 , lim f (x)=0 ,
x 
f is odd

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

6. lim f (x)= , lim f (x)=  ,

+ 
x 0 x 0
lim f (x)=1 , lim f (x)=1
x  x  

7. lim f (x)=  , lim f (x)= ,

x 2 x 
lim f (x)=0 , lim f (x)= ,
x   +
x 0
lim f (x)= 

x 0

8. lim f (x)= , lim f (x)=3 ,

x  2 x  
lim f (x)= 3
x 

2 x
9. If f (x)=x /2 , then a calculator gives f (0)=0 , f (1)=0.5 , f (2)=1 , f (3)=1.125 , f (4)=1 ,
f (5)=0.78125 , f (6)=0.5625 , f (7)=0.3828125 , f (8)=0.25 , f (9)=0.158203125 , f (10)=0.09765625 ,
 12  27
f (20) 0.00038147 , f (50) 2.2204 10 , f (100) 7.8886 10 .
It appears that lim ( x /2 ) =0 .
2 x

x 

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

x
10. (a) From a graph of f (x)= ( 1 2/x ) in a window of 0,10,000 by 0,0.2 , we estimate that
lim f (x)=0.14 (to two decimal places.)
x 
(b)
x f (x)
10,000 0.135308
100,000 0.135333
1,000,000 0.135335

From the table, we estimate that lim f (x)=0.1353 (to four decimal places.)
x 

11.
[ divide both the numerator and
2 2 2
3x  x+4 (3x  x+4)/x denominator by x
2
lim =lim
2 2 2 (the highest power of x that
x  2x +5x 8 x  (2x +5x 8)/x
appears in the denominator)]
2
lim (3 1/x+4/x )
x 
= [ Limit Law 5]
2
lim (2+5/x 8/x )
x 
2
lim 3 lim (1/x)+lim (4/x )
x  x  x 
= [ Limit Laws 1 and 2]
2
lim 2+lim (5/x) lim (8/x )
x  x  x 
2
3 lim (1/x)+4lim (1/x )
x  x 
= [ Limit Laws 7 and 3]
2
2+5lim (1/x) 8lim (1/x )
x  x 
3 0+4(0)
= [Theorem 5 of Section 2.5]
2+5(0) 8(0)
3
=
2

12.

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

3 3
12x  5x+2 12x  5x+2
lim = lim [ Limit Law 11]
x  2 3 x  2 3
1+4x +3x 1+4x +3x
2 3
12 5/x +2/x 3
= lim 3 [ divide by x ]
x  1/x +4/x+3

2 3
lim (12 5/x +2/x )
x 
= [ Limit Law 5]
3
lim (1/x +4/x+3)
x 

2 3
lim 12 lim (5/x )+lim (2/x )
x  x  x 
= [ Limit Laws 1 and 2]
3
lim (1/x )+lim (4/x)+lim 3
x  x  x 

2 3
12 5lim (1/x )+2lim (1/x )
x  x 
= [ Limit Laws 7 and 3]
3
lim (1/x )+4lim (1/x)+3
x  x 

12 5(0)+2(0)
= [Theorem 5 of Section 2.5]
0+4(0)+3
12
= = 4 =2
3

lim (1/x) lim (1/x)

1 1/x x  x  0 0
13. lim =lim = = = = =0
x  2x+3 x  (2x+3)/x lim (2+3/x) lim 2+3lim (1/x) 2+3(0) 2
x  x  x 

1
lim 3+5lim
3x+5 (3x+5)/x 3+5/x x  x  3+5(0) x
14. lim =lim =lim = = =3
x  x 4 x  (x 4)/x x  1 4/x 1 1 4(0)
lim 1 4lim
x  x  x

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

15.
2
2 2 2 lim (1/x  1/x 1)
1 x x (1 x x )/x x  
lim 2
= lim = 2
2 2
x   2x  7 x   (2x  7)/x lim (2 7/x )
x  
2
lim (1/x ) lim (1/x) lim 1
x   x   x   0 0 1 1
= = =
2 2 7(0) 2
lim 2 7 lim (1/x )
x   x  

2 2
2 2 2 lim (2/y  3) 2lim (1/y ) lim 3
2 3y (2 3y )/y y  y  y  2(0) 3 3
16. lim =lim = = = =
y 
2
y  2 2 lim (5+4/y) lim 5+4lim (1/y) 5+4(0) 5
5y +4y (5y +4y)/y y  y  y 

3
17. Divide both the numerator and denominator by x (the highest power of x that occurs in the
denominator).
3
x +5x 5 5
1+ lim 1+
3 3 2 2
x +5x x x x  x
lim =lim =lim =
3 2 3 2 1 4 1 4
x  2x  x +4 x  2x  x +4 x 
2 + lim 2 +
3 x 3 x  x 3
x x x
1
lim 1+5lim
2
x  x  x 1+5(0) 1
= = =
1 1 2 0+4(0) 2
lim 2 lim +4lim
x  x  x x  3
x

18. lim
t +2
2
= lim
( t2+2) /t 3 = lim 1/t+2/t
3
=
0+0
=0
t  
3
t +t  1
2
t   ( t 3+t2 1) /t 3 t   1+1/t 1/t
3 1+0 0

19. First, multiply the factors in the denominator. Then divide both the numerator and denominator by
4
u .

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

4
4u +5 5
4
4+
4 4 4
4u +5 4u +5 u u
lim =lim 4 2
=lim =lim
2 2 4 2 5 2
u  (u  2)(2u  1) u  2u  5u +2 u  2u  5u +2 u  2 +
2 4
4 u u
u
5 1
lim 4+ lim 4+5lim
4 4
u  u u  u  u 4+5(0)
= = =
5 2 1 1 2 5(0)+2(0)
lim 2 + lim 2 5lim +2lim
2 4 2 4
u  u u u  u  u u  u
4
= =2
2

x+2 ( x+2 ) /x 1+2/x 1+0 1

20. lim =lim =lim = =
x  2 x  2 2 x  2 9+0 3
9x +1 9x +1 / x 9+1/x

21.
6 6
6 6 3 lim (9x  x)/x
9x  x 9x  x /x x  3 6
lim 3
=lim = 3 [ since x = x for x>0 ]
3 3
x  x +1 x  (x +1)/x lim (1+1/x )
x 

5 5
lim 9 1/x lim 9 lim (1/x )
x  x  x 
= =
3 1+0
lim 1+lim (1/x )
x  x 

= 9 0 =3

22.
6 6
6 6 3 lim  (9x  x)/x
9x  x 9x  x /x x   3 6
lim 3
= lim = 3 [ since x = x for x<0 ]
3 3
x   x +1 x   (x +1)/x lim (1+1/x )
x  

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

5 5
lim  9 1/x  lim 9 lim (1/x )
x   x   x  
= =
3 1+0
lim 1+ lim (1/x )
x   x  

= 9 0 = 3

23.
2

lim ( 9x +x  3x ) =lim
2 ( 9x +x  3x ) (
2
9x +x +3x )
2
=lim
( 9x +x )  ( 3x )
2 2

x  x  2 x  2
9x +x +3x 9x +x +3x

=lim
( 9x2+x)  9x2 =lim x

1/x
x  2 x  2 1/x
9x +x +3x 9x +x +3x
x/x 1 1 1 1
=lim =lim = = =
x  2 2 2 x  9+1/x +3 9 +3 3+3 6
9x /x +x/x +3x/x

24.

lim ( x+ 2
x +2x ) = lim ( x+ 2
x +2x ) x
2
x +2x
= lim
2
x ( x2+2x)
x   x   x  
2 2
x x +2x x x +2x
 2x  2  2
= lim = lim = = 1
x   2 x   1+ 1+2/x 1+ 1+2 ( 0 )
x x +2x

2
Note: In dividing numerator and denominator by x , we used the fact that for x<0 , x= x .

25.

lim ( 2
x +ax  x +bx
2
) =lim ( 2
x +ax  x +bx
2
)( 2
x +ax + x +bx
2
)
x  x  2 2
x +ax + x +bx
2 2
(x +ax) (x +bx) [(a b)x]/x
=lim =lim
x  2
x +ax + x +bx
2 x 
( 2
x +ax + x +bx / x
2
) 2

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

a b a b a b
=lim = =
x  1+a/x + 1+b/x 1+0 + 1+0 2

26. lim cos x does not exist because as x increases cos x does not approach any one value, but
x 
oscillates between 1 and  1 .

27. x is large when x is large, so lim x = .

x 

3 3
28. x is large negative when x is large negative, so lim x =  .
x  
29. lim ( x x ) =lim x ( x  1 ) = since x  and x  1  as x  .
x  x 

3 3 2
x  2x+3 (x  2x+3)/x
30. lim =lim [divide by the highest power of x in the denominator]
2 2 2
x  5 2x x  (5 2x )/x
2
x 2/x+3/x 2 2
=lim 2
=  because x 2/x+3/x   and 5/x  2  2 as x  .
x  5/x  2

4 5 5 1 5
31. lim (x +x )= lim x ( +1)=  because x    and 1/x+1 1 as x   .
x   x   x

 1
32. lim tan ( x2 x4) =lim tan  1 ( x2 ( 1 x2) ) . If we let t=x2 ( 1 x2) , we know that t   as
x  x 
 1  1 
x  , since x 
2
 and 1 x 
2
  . So lim tan ( x2 ( 1 x2) ) = lim tan t=
2
.
x  t  

33.
3 5 3 5 4
x+x +x (x+x +x )/x
lim =lim [ divide by the highest power of x in the denominator ]
2 4 2 4 4
x  1 x +x x  (1 x +x )/x
3
1/x +1/x+x
=lim 4 2
=
x  1/x  1/x +1
3 4 2
because (1/x +1/x+x)  and (1/x  1/x +1) 1 as x  .

+
34. If we let t=tan x , then as x ( /2) , t   . Thus,
9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

tan x t
lim e = lim e =0 .
+ t  
x ( /2)

35. (a)

2
From the graph of f (x)= x +x+1 +x , we estimate the value of lim f (x) to be  0.5 .
x  

(b)
x f (x)
 10,000  0.4999625
 100,000  0.4999962
 1,000,000  0.4999996

From the table, we estimate the limit to be  0.5 .

(c)

lim ( 2
x +x+1 +x = lim ) ( 2
x +x+1 +x )
2
x +x+1  x
= lim
( x2+x+1)  x2
x   x   2 x   2
x +x+1  x x +x+1  x
(x+1)(1/x) 1+(1/x)
= lim = lim
x   ( 2
x +x+1  x (1/x) ) x  
 1+(1/x)+ 1/x ( 2)  1
1+0 1
= =
 1+0+0  1 2

2
Note that for x<0 , we have x = x = x , so when we divide the radical by x , with x<0 , we get
1
x
2
x +x+1 =
1 2 2
x +x+1 = 1+(1/x)+ 1/x . ( )
2
x

36. (a)

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2 2
From the graph of f (x)= 3x +8x+6  3x +3x+1 , we estimate (to one decimal place) the value of
lim f (x) to be 1.4 .
x 
(b)
x f (x)
10,000 1.44339
100,000 1.44338
1,000,000 1.44338

From the table, we estimate (to four decimal places) the limit to be 1.4434 .
(c)

lim f (x) =lim ( 2

3x +8x+6 
2
3x +3x+1 )( 2
3x +8x+6 + 3x +3x+1
2
)
x 
x  2 2
3x +8x+6 + 3x +3x+1

=lim
( 3x2+8x+6)  ( 3x2+3x+1)
x  2 2
3x +8x+6 + 3x +3x+1
(5x+5)(1/x)
=lim
x  ( 2
3x +8x+6 + 3x +3x+1
2
) (1/x)
5+5/x 5 5 5 3
=lim = = =  1.443376
x  2 2 3+ 3 2 3 6
3+8/x+6/x + 3+3/x+1/x

x 1 1 x
37. lim = lim = =1 , so y=1 is a horizontal asymptote. lim = and
x  x+4 x  1+4/x 1+0  x+4
x  4
x
lim =  , so x= 4 is a vertical asymptote. The graph confirms these calculations.
+ x+4
x  4

11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2
38. Since x  1 0 as x 1 and y<0 for  1<x<1 and y>0 for x< 1 and x>1 , we have
2 2 2 2
x +4 x +4 x +4 x +4
lim 2
=  , lim 2
= , lim 2
= , and lim 2
=  , so x=1 and x= 1 are
 
x 1 x 1 x 1
+
x 1 x  1 x 1 x
+
 1 x 1
2 2
x +4 1+4/x 1+0
vertical asymptotes. Also lim = lim = =1 , so y=1 is a horizontal asymptote.
2 2 1 0
x 1 x  x  1 1/x
The graph confirms these calculations.

3
x x
39. lim = lim =  , so there is no horizontal asymptote.
x 
2
x +3x 10 x  1+(3/x) ( 10/x2)
3 3 3 3
x x x x
lim =lim = , since >0 for x>2 . Similarly, lim = 
+ 2 + (x+5)(x 2) (x+5)(x 2)  2
x 2 x +3x 10 x 2 x 2 x +3x 10
3 3
x x
, lim 2
=  , and lim 2
= , so x=2 and x= 5 are vertical asymptotes. The
 +
xx +3x 10
 5 x  5 x +3x 10
graph confirms these calculations.

40.

12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

3 3 3 3
x +1 1+1/x x +1 x +1
lim = lim =1 , so y=1 is a horizontal asymptote. Since y= = >0 for
x ( x +1 )
3 2 3 2
x  x +x x  1+1/x x +x
3 3
x +1 x +1
x>0 and y<0 for  1<x<0 , lim 3
= and lim 3
=  , so x=0 is a vertical asymptote.
+ 
x 0 x +x x 0 x +x

x 1/x 1 1
41. lim 4
 =lim = 4
=1 and
 4 4 4 
x
x +1 x 1 1+0
1/ x 4 1+
4
x
x 1/x 1 1
lim 4
 = lim = 4
= 1 , so y= 1 are horizontal asymptotes.
4
x   4
 1/
4 x   1  1+0
x +1 x  4 1+
4
x
There is no vertical asymptote.

x 9 1 9/x 1 0 1
42. lim =lim = = .
4+0+0 2
x  2
4x +3x+2 x 
4+ ( 3/x ) + 2/x ( 2)
2
Using the fact that x = x = x for x<0 , we divide the numerator by  x and the denominator by
2
x .
x 9  1+9/x  1+0 1
Thus, lim = lim = = .
4+0+0 2
x   2
4+ ( 3/x ) + 2/x
4x +3x+2 x  
( 2)
1 2
The horizontal asymptotes are y= . The polynomial 4x +3x+2 is positive for all x , so the
2
denominator never approaches zero, and thus there is no vertical asymptote.
13
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

43. Let’s look for a rational function.

(1) lim f (x)=0
degree of numerator < degree of denominator
x 
(2) 2
lim f (x)= 
there is a factor of x in the denominator (not just x , since that would
x 0
produce a sign change at x=0 ), and the function is negative near x=0 .
(3) lim f (x)= and lim f (x)= 
vertical asymptote at x=3 ; there is a factor of ( x 3) in the
 +
x 3 x 3
denominator.
(4) f (2)=0
2 is an x  intercept; there is at least one factor of ( x 2 ) in the numerator.

Combining all of this information and putting in a negative sign to give us the desired left and right
2 x
hand limits gives us f (x)= 2 as one possibility.
x (x 3)

44. Since the function has vertical asymptotes x=1 and x=3 , the denominator of the rational function
we are looking for must have factors ( x 1 ) and ( x 3) . Because the horizontal asymptote is y=1 , the
degree of the numerator must equal the degree of the denominator, and the ratio of the leading
2
x
coefficients must be 1 . One possibility is f (x)= .
(x 1)(x 3)

2
45. y= f (x)=x (x 2)(1 x) . The y intercept is f (0)=0 , and the x intercepts occur when y=0
x=0 , 1
2
, and 2 . Notice that, since x is always positive, the graph does not cross the x axis at 0 , but does
2
cross the x axis at 1 and 2 . lim x (x 2)(1 x)=  , since the first two factors are large positive and
x 
the third large negative when x is large positive.

14
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2
lim x (x 2)(1 x)=  because the first and third factors are large positive and the second large
x  
negative as x   .

3
46. y=(2+x) (1 x)(3 x) . As x  , the first factor is large positive, and the second and third factors
are large negative. Therefore, lim f (x)= . As x   , the first factor is large negative, and the
x 
second and third factors are large positive. Therefore, lim f (x)=  . Now the y intercept is
x  
3
f (0)=(2) (1)(3)=24 and the x intercepts are the solutions to f (x)=0
x= 2 , 1 and 3 , and the graph
crosses the x axis at all of these points.

5 4 5 4
47. y= f (x)=(x+4) (x 3) . The y intercept is f (0)=4 ( 3) =82 , 944 . The x intercepts occur when
4
y=0
x= 4 , 3 . Notice that the graph does not cross the x axis at 3 because (x 3) is always
5 4
positive, but does cross the x axis at  4 . lim (x+4) (x 3) = since both factors are large positive
x 
5 4
when x is large positive. lim (x+4) (x 3) =  since the first factor is large negative and the second
x  
factor is large positive when x is large negative.

15
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2 2
48. y=(1 x)(x 3) (x 5) . As x  , the first factor approaches   while the second and third factors
approach  . Therefore, lim (x)=  . As x   , the factors all approach  . Therefore,
x 
2 2
lim (x)= . Now the y intercept is f (0)=(1)( 3) ( 5) =225 and the x intercepts are the solutions
x  
to f (x)=0
x=1 , 3 , and 5 . Notice that f (x) does not change sign at x=3 or x=5 because the factors
2 2
(x 3) and (x 5) are always positive, so the graph does not cross the x axis at x=3 or x=5 , but does
cross the x axis at x=1 .

1 1 sin x
49. (a) Since  1 sin x 1 for all x,  
for x>0 . As x  ,  1/x 0 and 1/x 0 , so by
x x x
sin x
the Squeeze Theorem, (sin x)/x 0 . Thus, lim =0 .
x  x
(b) From part (a), the horizontal asymptote is y=0 . The function y=(sin x)/x crosses the horizontal
asymptote whenever sin x=0 ; that is, at x= n for every integer n . Thus, the graph crosses the
asymptote an infinite number of times.

50. (a) In both viewing rectangles, lim P(x)=lim Q(x)= and lim P(x)= lim Q(x)=  . In the
x  x  x   x  
larger viewing rectangle, P and Q become less distinguishable.

16
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

5 3
P(x) 3x  5x +2x 5 1 2 1 5 2
(b) lim =lim =lim 1  +  =1 (0)+ (0)=1

x  Q(x) x  3 2 3 4 3 3
5
3x x  x x
P and Q have the same end behavior.

51. Divide the numerator and the denominator by the highest power of x in Q(x) .
(a) If deg P<deg Q , then the numerator  0 but the denominator doesn’t. So lim P(x)/Q(x) =0 .
x 
(b) If deg P>deg Q , then the numerator   but the denominator doesn’t, so
lim P(x)/Q(x) =  (depending on the ratio of the leading coefficients of P and Q ).
x 

52.

(i) n=0 (ii) n>0 (n odd) (iii) n>0 (n even) (iv) n<0 (n odd) (v) n<0 (n even)

(a) lim x =
x 0
+
n
{ 1
0

if n=0
if n>0
if n<0

{
1 if n=0
n 0 if n>0
(b) lim x =
   if n<0, n odd
x 0  if n<0, n even

(c) lim x =
x 
n
{ 1

0
if n=0
if n>0
if n<0

{
1 if n=0
n   if n>0, n odd
(d) lim x =
x    if n>0, n even
0 if n<0

53.
17
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2
4x 1 1 4x +3x 3
lim =lim 4 =4 , and lim =lim 4+ =4 . Therefore, by the Squeeze
x  x x  x x  2 x  x
x
Theorem, lim f (x)=4 .
x 

54. (a) After t minutes, 25t liters of brine with 30 g of salt per liter has been pumped into the tank, so
it contains ( 5000+25t ) liters of water and 25t 30=750t grams of salt. Therefore, the salt
750t 30t g
concentration at time t will be C(t)= = .
5000+25t 200+t L
30t 30t/t 30
(b) lim C(t)=lim =lim = =30 . So the salt concentration approaches that of the
t  t  200+t t  200/t+t/t 0+1
brine being pumped into the tank.

55. (a) lim v(t)=lim v

t  t 
*
( 1 e ) =v (1 0)=v
 gt/v
*
* *

 9.8t *
(b) We graph v(t)=1 e and v(t)=0.99v , or in this case, v(t)=0.99 . Using an intersect feature or
zooming in on the point of intersection, we find that t 0.47 s.

 x/10
56. (a) y=e and y=0.1 intersect at x  23.03 .
1
 x/10
If x>x , then e <0.1 .
1

 x/10
(b) e <0.1
 x/10<ln 0.1

1  1
x> 10ln = 10ln 10 =10ln 10
10

18
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2 2
6x +5x 3 6x +5x 3
57. 2
 3 <0.2 2.8< 2
<3.2 . So we graph the three parts of this inequality on
2x  1 2x  1
2
6x +5x 3
the same screen, and find that the curve y= 2
seems to lie between the lines y=2.8 and y=3.2
2x  1
whenever x>12.8 . So we can choose N=13 (or any larger number) so that the inequality holds
whenever x N .

2
4x +1
58. For  =0.5 , we must find N such that whenever x N , we have  2 <0.5
x+1
2
4x +1
1.5< <2.5 . We graph the three parts of this inequality on the same screen, and find that it
x+1
holds whenever x 3 . So we choose N=3 (or any larger number). For  =0.1 , we must have
2
4x +1
1.9< <2.1 , and the graphs show that this holds whenever x 19 . So we choose N=19 (or
x+1
any larger number).

19
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2 2
4x +1 4x +1
59. For  =0.5 , we need to find N such that  ( 2) <0.5  2.5< < 1.5
x+1 x+1
whenever x N . We graph the three parts of this inequality on the same screen, and see that the
inequality holds for x  6. So we choose N= 6 (or any smaller number).
2
4x +1
For  =0.1 , we need  2.1< < 1.9 whenever x N. From the graph, it seems that this
x+1
inequality holds for x  22 . So we choose N= 22 (or any smaller number).

2x+1
60. We need N such that >100 whenever x N. From the graph, we see that this inequality
x+1
holds for x 2500 . So we choose N=2500 (or any larger number).

2 2
61. (a) 1/x <0.0001 x >1/0.0001=10 , 000 x>100 ( x>0 )
20
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

2 2
(b) If  >0 is given, then 1/x <  x >1/  x>1/  . Let N=1/  .
1 1 1 1
Then x>N
x>
 0 = < , so lim =0 .
 x
2
x
2 x  x
2

4 8
62. (a) 1/ x <0.0001 x >1/0.0001=10  x>10
2 2
(b) If  >0 is given, then 1/ x <  x >1/  x>1/ . Let N=1/ .
1 1 1 1
Then x>N
x>
 0 = < , so lim =0 .
2 x x x
 x 

63. For x<0 , 1/x 0 = 1/x . If  >0 is given, then  1/x<  x< 1/ .
Take N= 1/ . Then x<N
x< 1/
( 1/x )  0 = 1/x< , so lim ( 1/x ) =0 .
x  

3 3 3 3
64. Given M>0 , we need N>0 such that x>N
x >M . Now x >M x> M , so take N= M . Then
3 3 3
x>N= M
x >M , so lim x = .
x 

x x
65. Given M>0 , we need N>0 such that x>N
e >M . Now e >M  x>ln M , so take
x
N=max ( 1,ln M ) . (This ensures that N>0 .) Then x>N=max ( 1,ln M )
e >max ( e,M ) M , so
x
lim e = .
x 

66. Definition Let f be a function defined on some interval (   ,a ) . Then lim f (x)=  means
x  
that for every negative number M there is a corresponding negative number N such that f (x)<M
whenever x<N . Now we use the definition to prove that lim ( 1+x3) =  . Given a negative
x  
3 3 3
number M , we need a negative number N such that x<N
1+x <M . Now 1+x <M x <M 1
3 3 3
x< M 1 . Thus, we take N= M 1 and find that x<N
1+x <M . This proves that
lim ( 1+x3) =  .
x  

67. Suppose that lim f (x)=L . Then for every  >0 there is a corresponding positive number N such
x 
that f (x) L < whenever x>N . If t=1/x , then x>N  0<1/x<1/N  0<t<1/N . Thus, for every  >0
there is a corresponding  >0 (namely 1/N ) such that f (1/t) L < whenever 0<t< . This proves
that
21
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.6 Limits at Infinity; Horizontal Asymptotes

lim f (1/t)=L=lim f (x) .

+ x 
t 0
Now suppose that lim f (x)=L . Then for every  >0 there is a corresponding negative number N
x  
such that f (x) L < whenever x<N . If t=1/x , then x<N  1/N<1/x<0 1/N<t<0 . Thus, for every
 >0 there is a corresponding  >0 (namely  1/N ) such that f (1/t) L < whenever   <t<0 . This
proves that lim f (1/t)=L= lim f (x) .
 x  
t 0

22
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

 y f (x) f (3)
1. (a) This is just the slope of the line through two points: m = = .
PQ  x x 3
f (x) f (3)
(b) This is the limit of the slope of the secant line PQ as Q approaches P : m=lim .
x 3 x 3

 s f (a+h) f (a) f (a+h) f (a)

2. (a) Average velocity = = =
 t (a+h) a h
f (a+h) f (a)
(b) Instantaneous velocity =lim
h 0 h

3. The slope at D is the largest positive slope, followed by the positive slope at E . The slope at C is
zero. The slope at B is steeper than at A (both are negative). In decreasing order, we have the slopes
at: D , E , C , A , and B .

4. The curve looks more like a line as the viewing rectangle gets smaller.

5. (a)
(i) Using Definition 1,

m =lim
f (x) f (a)
lim
f (x) f ( 3)
=lim
x +2x  (3)
=lim
( 2
)
(x+3)(x 1)
x a x a x  3 x ( 3) x  3 x ( 3) x  3 x+3
=lim (x 1)= 4
x  3

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

(ii) Using Equation 2,

2
f (a+h) f (a) f ( 3+h) f ( 3) ( 3+h) +2( 3+h)  (3)
m =lim =lim =lim
h 0 h h 0 h h 0 h
2
9 6h+h  6+2h 3 h(h 4)
=lim =lim =lim (h 4)= 4
h 0 h h 0 h h 0

(b) Using the point slope form of the equation of a line, an equation of the tangent line is
y 3= 4(x+3) . Solving for y gives us y= 4x 9 , which is the slope intercept form of the equation of
the tangent line.

(c)

6. (a)
(i)
m =lim
f (x) f ( 1)
=lim
x  ( 1)
3
=lim
( x+1 ) x  x+1 ( 2
)
x  1 x ( 1) x  1 x+1 x  1 x+1
=lim ( x2 x+1) =3
x  1
(ii) 3 3 2
f ( 1+h) f ( 1) ( 1+h)  ( 1) h  3h +3h 1+1
m =lim =lim =lim
h 0 h h 0 h h 0 h
=lim ( h2 3h+3) =3
h 0

(b) y ( 1)=3 x ( 1)  y+1=3x+3 y=3x+2

(c)

3
7. Using (2) with f (x)=1+2x x and P(1,2) ,

2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

3
f (a+h) f (a) f (1+h) f (1) 1+2(1+h) (1+h)  2
m =lim =lim =lim
h 0 h h 0 h h 0 h
2 3 3 2
1+2+2h (1+3h+3h +h ) 2  h  3h  h
=lim =lim
h 0 h h 0 h
2
h( h  3h 1) 2
=lim =lim ( h  3h 1)= 1
h 0 h h 0

Tangent line: y 2= 1(x 1) y 2= x+1 y= x+3

8. Using (1),
2x+1  2(4)+1 2x+1  3 2x+1 +3
m=lim =lim 
x 4 x 4 x 4 x 4 2x+1 +3
2
(2x+1) 3 2(x 4)
=lim =lim
x 4 (x 4)( 2x+1 +3) x 4 (x 4)( 2x+1 +3)
2 2 1
=lim = = .
x 4 ( 2x+1 +3) 3+3 3

1 1 4 1 5
Tangent line: y 3= (x 4)  y 3= x  y= x+
3 3 3 3 3

x 1
9. Using (1) with f (x)= and P ( 3,2 ) ,
x 2
x 1 x 1 2(x 2)
 2
m f (x) f (a) x 2 x 2 3 x
=lim =lim =lim =lim
x a x a x 3 x 3 x 3 x 3 x 3 (x 2)(x 3)
 1  1
=lim = = 1 .
x 3 x 2 1

Tangent line: y 2= 1(x 3) y 2= x+3  y= x+5

2x
 0
2
(x+1) 2x 2 2
10. Using (1), m=lim =lim =lim = =2 .
x 0 x 0 x 0 2 x 0 2 2
x(x+1) (x+1) 1
Tangent line: y 0=2(x 0)  y=2x

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

11. (a)
f (x) f (a) 2/(x+3) 2/(a+3) 2(a+3) 2(x+3)
m =lim =lim =lim
x a x a x a x a x a (x a)(x+3)(a+3)
2(a x)  2  2
=lim =lim =
x a (x a)(x+3)(a+3) x a (x+3)(a+3) 2
(a+3)

(b)
(i)  2 1
a= 1 m= =
2 2
( 1+3)
(ii)  2 2
a=0 m= =
2 9
(0+3)
(iii)  2 1
a=1 m= =
2 8
(1+3)

12. (a) Using (1),

m = lim
( 1+x+x2)  ( 1+a+a2) =lim 2
x+x  a a
2
=lim
x a+(x a)(x+a)
x a x a x a x a x a x a
(x a)(1+x+a)
= lim =lim ( 1+x+a ) =1+2a
x a x a x a

(b)
(i) x= 1 m=1+2( 1)= 1
1 1
(ii) x=  m=1+2  =0
2 2
(iii) x=1 m=1+2(1)=3

(c)

13. (a) Using (1),

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

m = lim
( x3 4x+1)  ( a3 4a+1) =lim ( x3 a3)  4(x a)
x a x a x a x a

= lim
( 2
(x a) x +ax+a  4(x a)
2
2) 2 2
=lim x +ax+a  4 =3a  4 ( )
x a x a x a

2
(b) At ( 1, 2 ) : m=3(1)  4= 1 , so an equation of the tangent line is y ( 2)= 1(x 1) y= x 1 . At
2
( 2,1 ) : m=3(2)  4=8 , so an equation of the tangent line is y 1=8(x 2) y=8x 15 .

(c)

14. (a) Using (1),

1 1 a x

x a ax ( a x ) ( a+ x )
m=lim =lim =lim
x a x a x a x a x a ax ( x a ) ( a + x )
a x  1  1 1 1  3/2
=lim =lim = = or  a
x a ax ( x a ) ( a + x ) x a ax ( a + x ) 2
2 3/2
a (2 a ) 2a

1 1 1 3
(b) At ( 1,1 ) : m= , so an equation of the tangent line is y 1= (x 1) y= x+ .
2 2 2 2
1 1 1 1 1 3
At 4, : m= , so an equation of the tangent line is y = (x 4) y= x+ .
2 16 2 16 16 4

(c)

15. (a) Since the slope of the tangent at t=0 is 0 , the car’s initial velocity was 0 .
(b) The slope of the tangent is greater at C than at B , so the car was going faster at C .
(c) Near A , the tangent lines are becoming steeper as x increases, so the velocity was increasing, so
the car was speeding up. Near B , the tangent lines are becoming less steep, so the car was slowing
down. The steepest tangent near C is the one at C , so at C the car had just finished speeding up, and
was about to start slowing down.
5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

(d) Between D and E , the slope of the tangent is 0 , so the car did not move during that time.

16. Let a denote the distance traveled from 1:00 to 1:02 ,b from 1:28 to 1:30, and c from 3:30 to 3:33,
where all the times are relative to t=0 at the beginning of the trip.

2
17. Let s(t)=40t 16t .

v(2) = lim
s(t) s(2)
=lim
( 40t 16t2)  16 =lim 2
 16t +40t 16
=lim
(
 8 2t  5t+2
2
)
t 2 t 2 t 2 t 2 t 2 t 2 t 2 t 2
 8(t 2)(2t 1)
= lim = 8lim (2t 1)= 8(3)= 24
t 2 t 2 t 2
Thus, the instantaneous velocity when t=2 is  24 ft / s.

18. (a)
H(1+h) H(1)
v(1) =lim
h 0 h

=lim
( 58+58h 0.83 1.66h 0.83h2)  57.17 =lim (56.34 0.83h)=56.34 m / s
h 0 h h 0

(b)
H(a+h) H(a)
v(a) =lim
h 0 h

=lim
( 58a+58h 0.83a2 1.66ah 0.83h2)  ( 58a 0.83a2)
h 0 h
=lim ( 58 1.66a 0.83h ) =58 1.66a m / s
h 0

(c) The arrow strikes the moon when the height is 0 , that is,
6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

2 58
58t 0.83t =0 t(58 0.83t)=0 t=  69.9 s (since t can’t be 0 ).
0.83
58 58
(d) Using the time from part (c), v =58 1.66 = 58 m / s. Thus, the arrow will
0.83 0.83
have a velocity of  58 m / s.

19.

v(a) =lim
s(a+h) s(a)
=lim
4(a+h) +6(a+h)+2
3
( 4a3+6a+2)
h 0 h h 0 h
3 2 2 3 3
4a +12a h+12ah +4h +6a+6h+2 4a  6a 2
=lim
h 0 h
2 2 3
12a h+12ah +4h +6h
=lim
h
2 2 2
(
=lim 12a +12ah+4h +6 = 12a +6 m / s ) ( )
h 0 h 0
2 2 2
So v(1)=12(1) +6=18 m / s, v(2)=12(2) +6=54 m / s, and v(3)=12(3) +6=114 m / s.

20. (a) The average velocity between times t and t+h is

s(t+h) s(t) (t+h)2 8(t+h)+18 t 2 8t+18 ( )
( t+h )  t = h
2 2 2 2
t +2th+h  8t 8h+18 t +8t 18 2th+h  8h
= =
h h
= ( 2t+h 8 ) m/s

(i)[3,4] : t=3 , h=4 3=1 , so the average (ii)[3.5,4] : t=3.5 , h=0.5 , so the average velocity
velocity is 2(3)+1 8= 1 m / s. is 2(3.5)+0.5 8= 0.5 m / s.
(iii)[4,5] : t=4 , h=1 , so the average velocity is (iv)[4,4.5] : t=4 , h=0.5 , so the average velocity is
2(4)+1 8=1 m / s. 2(4)+0.5 8=0.5 m / s.

s(t+h) s(t)
(b) v(t)=lim =lim ( 2t+h 8 ) =2t 8 , so v ( 4 ) =0 .
h 0 h h 0
(c)

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

21. The sketch shows the graph for a room temperature of 72 and a refrigerator temperature of 38 .
The initial rate of change is greater in magnitude than the rate of change after an hour.

22. The slope of the tangent (that is, the rate of change of temperature with respect to time) at t=1 h
75 168
seems to be about   0.7 F / min.
132 0

7.9 11.5
23. (a) (i) 20,23 : = 1.2 C / h
23 20

9.0 11.5
(ii) 20,22 : = 1.25 C / h
22 20

10.2 11.5
(iii) 20,21 : = 1.3 C / h
21 20

6 15.5
(b) In the figure, we estimate A to be ( 18,15.5) and B as ( 23,6 ) . So the slope is = 1.9 C / h
23 18
at 8:00 P.M.

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

24. (a)
P(1996) P(1992) 10,152 10,036 116
(i) 1992,1996 : = = =29 thousand / year
1996 1992 4 4
P(1996) P(1994) 10,152 10,109 43
(ii) 1994,1996 : = = =21.5 thousand / year
1996 1994 2 2
P(1998) P(1996) 10,175 10,152 23
(iii) 1996,1998 : = = =11.5 thousand / year
1998 1996 2 2

21.5+11.5
(b) Using the values from (ii) and (iii), we have =16.5 thousand / year.
2
(c) Estimating A as (1994,10 , 125) and B as ( 1998,10,182 ) , the slope at 1996 is
10,182 10,125 57
= =14.25 thousand / year.
1998 1994 4

25. (a)
N(1997) N(1995) 2461 873 1588
(i) 1995,1997 : = = =794 thousand / year
1997 1995 2 2
N(1996) N(1995) 1513 873
(ii) 1995,1996 : = =640 thousand / year
1996 1995 1
N(1995) N(1994) 873 572
(iii) 1994,1995 : = =301 thousand / year
1995 1994 1

(b) Using the values from (ii) and (iii), we have

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

640+301 941
= =470.5 thousand / year.
2 2
1275 420 855
(c) A as ( 1994,420 ) and B as ( 1996,1275) , the slope at 1995 is = =427.5 thousand
1996 1994 2
/ year

26. (a)
N(1998) N(1996) 1886 1015 871
(i) 1996,1998 : = = =435.5 locations / year
1998 1996 2 2
N(1998) N(1997) 1886 1412
(ii) 1997,1998 : = =474 locations / year
1998 1997 1
N(1999) N(1998) 2135 1886
(iii) 1998,1999 : = =249 locations / year
1999 1998 1

474+249 723
(b) Using the values from (ii) and (iii), we have = =361.5 362 locations / year.
2 2
(c) Estimating A as ( 1997,1525) and B as ( 1999,2250 ) , the slope at 1998 is
2250 1525 725
= =362.5 locations / year.
1999 1997 2

27. (a)

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.7 Tangents, Velocities, and Other Rates of Change

(i)  C C(105) C(100) 6601.25 6500

= = =$20.25/ unit.
 x 105 100 5
(ii)  C C(101) C(100) 6520.05 6500
= = =$20.05/ unit.
 x 101 100 1

(b)
2 2
C(100+h) C(100) 5000+10(100+h)+0.05(100+h)  6500 20h+0.05h
h = =
h h
=20+0.05h , h
0
C(100+h) C(100)
So the instantaneous rate of change is lim =lim (20+0.05h)=$20/ unit.
h 0 h h 0

28.
t+h 2 t 2
 V =V (t+h) V (t)=100 , 000 1  100 , 000 1
60 60
2 2 2
t+h ( t+h ) t t h 2th h
=100 , 000 1 +  1 + =100 , 000  + +
30 3600 30 3600 30 3600 3600
100,000 250
= h (  120+2t+h ) = h (  120+2t+h )
3600 9

500
Dividing  V by h and then letting h 0 , we see that the instantaneous rate of change is ( t 60 )
9
gal / min.
t Flow rate (gal/min) Water remaining V (t) (gal)
0  3333.3 100,000
10  2777.7 69,444.4
20  2222.2 44,444.4
30  1666.6 25,000
40  1111.1 11,111.1
50  555.5 2,777.7
60 0 0
The magnitude of the flow rate is greatest at the beginning and gradually decreases to 0.

11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

1.

f (2+h) f (2)
The line from P(2, f (2)) to Q(2+h, f (2+h)) is the line that has slope
h

2. As h decreases, the line PQ becomes steeper, so its slope increases. So

f (4) f (2) f (3) f (2) f (x) f (2) 1 /
0< < <lim . Thus, 0< [ f (4) f (2)]< f (3) f (2)< f (2) .
4 2 3 2 x 2 x 2 2

/
3. g (0) is the only negative value. The slope at x=4 is smaller than the slope at x=2 and both are
/ / / /
smaller than the slope at x= 2 . Thus, g (0)<0<g (4)<g (2)<g ( 2) .

1
4. Since (4,3) is on y= f (x) , f (4)=3 . The slope of the tangent line between (0,2) and (4,3) is , so
4
/ 1
f (4)= .
4

5.
We begin by drawing a curve through the origin at a slope of
/ /
3 to satisfy f (0)=0 and f (0)=3 . Since f (1)=0 , we will
round off our figure so that there is a horizontal tangent
directly over x=1 . Lastly, we make sure that the curve has a
slope of  1 as we pass over x=2 . Two of the many
possibilities are shown.

6.

7. Using Definition 2 with

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

2
f (x)=3x  5x and the point (2,2) , we have
2
/ f (2+h) f (2) [3(2+h)  5(2+h)] 2
f (2) =lim =lim
h 0 h h 0 h
2 2
(12+12h+3h  10 5h) 2 3h +7h
=lim =lim =lim (3h+7)=7 .
h 0 h h 0 h h 0

So an equation of the tangent line at (2,2) is y 2=7(x 2) or y=7x 12 .

3
8. Using Definition 2 with g(x)=1 x and the point (0,1) , we have
3 3
/ g(0+h) g(0) [1 (0+h) ] 1 (1 h ) 1 2
g (0)=lim =lim =lim =lim ( h )=0 .
h 0 h h 0 h h 0 h h 0
So an equation of the tangent line is y 1=0(x 0) or y=1 .

3
9. (a) Using Definition 2 with F(x)=x  5x+1 and the point (1, 3) , we have

3
/ F(1+h) F(1) [(1+h)  5(1+h)+1] ( 3)
F (1) =lim =lim
h 0 h h 0 h
2 3 3 2
(1+3h+3h +h  5 5h+1)+3 h +3h  2h
=lim =lim
h 0 h h 0 h
2
h(h +3h 2) 2
=lim =lim (h +3h 2)= 2
h 0 h h 0

So an equation of the tangent line at (1, 3) is y ( 3)= 2(x 1) y= 2x 1 .

(b)

10. (a)
a+h a
/ 
G (a) =lim G(a+h) G(a) =lim 1+2(a+h) 1+2a
h 0 h h 0 h
2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

2 2
a+2a +h+2ah a 2a  2ah 1  2
=lim =lim =(1+2a)
h 0 h(1+2a+2h)(1+2a) h 0 (1+2a+2h)(1+2a)

1 1 1  2
So the slope of the tangent at the point ( , ) is m= 1+2( ) =4 , and thus an equation is
4 2 4
1 1 1
y+ =4(x+ ) or y=4x+ .
2 4 2

(b)

1+h 1
/ f (1+h) f (1) 3  3
11. (a) f (1)=lim =lim .
h 0 h h 0 h
1+h
3  3
So let F(h)= . We calculate:
h

h F(h) h F(h)
0.1 3.484  0.1 3.121
0.01 3.314  0.01 3.278
0.001 3.298  0.001 3.294
0.0001 3.296  0.0001 3.296

/
We estimate that f (1) 3.296 .

(b)
3.2 2.8 0.4
From the graph, we estimate that the slope of the tangent is about =  3.3 .
1.06 0.94 0.12

12. (a)
3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

   
g( +h) g( ) tan( +h) tan( )
/  4 4 4 4
g ( ) =lim =lim .
4 h 0 h h 0 h

tan( +h) 1
4
So let G(h)= . We calculate:
h

h G(h) h G(h)
0.1 2.2305  0.1 1.8237
0.01 2.0203  0.01 1.9803
0.001 2.0020  0.001 1.9980
0.0001 2.0002  0.0001 1.9998

(b)
1.07 0.91 0.16
From the graph, we estimate that the slope of the tangent is about = =2 .
0.82 0.74 0.08

2
13. Use Definition 2 with f (x)=3 2x+4x .
2 2
/ f (a+h) f (a) [3 2(a+h)+4(a+h) ] (3 2a+4a )
f (a) =lim =lim
h 0 h h 0 h
2 2 2 2
[3 2a 2h+4a +8ah+4h ] (3 2a+4a )  2h+8ah+4h
=lim =lim
h 0 h h 0 h
h( 2+8a+4h)
=lim =lim ( 2+8a+4h)= 2+8a
h 0 h h 0

14.
4 4
/ f (a+h) f (a) [(a+h)  5(a+h)] (a  5a)
f (a) =lim =lim
h 0 h h 0 h

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

4 3 2 2 3 4 4
(a +4a h+6a h +4ah +h  5a 5h) (a  5a)
=lim
h 0 h
3 2 2 3 4 3 2 2 3
4a h+6a h +4ah +h  5h h(4a +6a h+4ah +h  5)
=lim =lim
h 0 h h 0 h
3 2 2 3 3
=lim (4a +6a h+4ah +h  5)=4a  5
h 0

15. Use Definition 2 with f (t)=(2t+1)/(t+3) .

2(a+h)+1 2a+1
/ 
f (a) =lim f (a+h) f (a) (a+h)+3 a+3
=lim
h 0 h h 0 h
(2a+2h+1)(a+3) (2a+1)(a+h+3)
=lim
h 0 h(a+h+3)(a+3)
2 2
(2a +6a+2ah+6h+a+3) (2a +2ah+6a+a+h+3)
=lim
h 0 h(a+h+3)(a+3)
5h 5 5
=lim =lim =
h 0 h(a+h+3)(a+3) h 0 (a+h+3)(a+3) 2
(a+3)

16.
2 2
(a+h) +1 a +1
/ 
f (a) f (a+h) f (a) (a+h) 2 a 2
=lim =lim
h 0 h h 0 h
2 2 2
(a +2ah+h +1))(a 2) (a +1)(a+h 2)
=lim
h 0 h(a+h 2)(a 2)
3 2 2 2 2 3 2 2
(a  2a +2a h 4ah+ah  2h +a 2) (a +a h 2a +a+h 2)
=lim
h 0 h(a+h 2)(a 2)
2 2 2 2
a h 4ah+ah  2h  h h(a  4a+ah 2h 1)
=lim =lim
h 0 h(a+h 2)(a 2) h 0 h(a+h 2)(a 2)
2 2
a  4a+ah 2h 1 a  4a 1
=lim =
h 0 (a+h 2)(a 2) 2
(a 2)

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

17. Use Definition 2 with f (x)=1/ x+2 .

1 1
/

f (a) =lim f (a+h) f (a) (a+h)+2 a+2
=lim
h 0 h h 0 h
a+2  a+h+2
a+h+2 a+2 a+2  a+h+2 a+2 + a+h+2
=lim =lim 
h 0 h h 0 h a+h+2 a+2 a+2 + a+h+2
(a+2) (a+h+2)
=lim
h 0 h a+h+2 a+2 ( a+2 + a+h+2 )
 h
=lim
h 0 h a+h+2 a+2 ( a+2 + a+h+2 )
 1
=lim
h 0 a+h+2 a+2 ( a+2 + a+h+2 )
 1 1
= 2
=
3/2
( a+2 ) (2 a+2 ) 2(a+2)

18.
/ f (a+h) f (a) 3(a+h)+1  3a+1
f (a) =lim h
=lim
h
h 0 h 0

( 3a+3h+1  3a+1 )( 3a+3h+1 + 3a+1 )

=lim
h 0 h( 3a+3h+1 + 3a+1 )
(3a+3h+1) (3a+1) 3h
=lim =lim
h 0 h( 3a+3h+1 + 3a+1 ) h 0 h( 3a+3h+1 + 3a+1 )
3 3
=lim =
h 0 3a+3h+1 + 3a+1 2 3a+1

10
(1+h)  1 / 10
19. By Definition 2, lim = f (1) , where f (x)=x and a=1 . Or: By Definition 2,
h 0 h
10
(1+h)  1 / 10
lim = f (0) , where f (x)=(1+x) and a=0 .
h 0 h

20. By Definition 2,

6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

4 4
16+h  2 / 4 16+h  2 /
lim = f (16) , where f (x)= x and a=16 . Or : By Definition 2, lim = f (0) ,
h 0 h h 0 h
4
where f (x)= 16+x and a=0 .

x
2  32 / x
21. By Equation 3, lim = f (5) , where f (x)=2 and a=5 .
x 5 x 5

tan x 1 /
22. By Equation 3, lim = f ( /4) , where f (x)=tan x and a= /4 .
x  /4 x  /4

cos ( +h)+1 /
23. By Definition 2, lim = f ( ) , where f (x)=cos x and a= . Or : By Definition 2,
h 0 h
cos ( +h)+1 /
lim = f (0) , where f (x)=cos ( +x) and a=0 .
h 0 h

4
t +t 2 / 4
24. By Equation 3, lim = f (1) , where f (t)=t +t and a=1 .
t 1 t 1

25.
2 2
/ f (2+h) f (2) [(2+h)  6(2+h) 5] [2  6(2) 5]
v(2) = f (2)=lim =lim
h 0 h h 0 h
2 2
(4+4h+h  12 6h 5) ( 13) h  2h
=lim =lim =lim (h 2)= 2m/s
h 0 h h 0 h h 0

26.
/ f (2+h) f (2)
v(2) = f (2)=lim
h 0 h
3 3
[2(2+h)  (2+h)+1] [2(2)  2+1]
=lim
h 0 h
3 2
(2h +12h +24h+16 2 h+1) 15
=lim
h 0 h

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

3 2
2h +12h +23h 2
=lim =lim (2h +12h+23)=23m/s
h 0 h h 0

/
27. (a) f (x) is the rate of change of the production cost with respect to the number of ounces of gold
produced. Its units are dollars per ounce.
(b) After 800 ounces of gold have been produced, the rate at which the production cost is increasing
is $17/ ounce. So the cost of producing the 800 th (or 801 st) ounce is about $17 .
/
(c) In the short term, the values of f (x) will decrease because more efficient use is made of start up
/
costs as x increases. But eventually f (x) might increase due to large scale operations.

/
28. (a) f (5) is the rate of growth of the bacteria population when t=5 hours. Its units are bacteria
per hour.
/ / /
(b) With unlimited space and nutrients, f should increase as t increases; so f (5)< f (10) . If the
supply of nutrients is limited, the growth rate slows down at some point in time, and the opposite may
be true.

/
29. (a) f (v) is the rate at which the fuel consumption is changing with respect to the speed. Its units
are (gal / h) / (mi / h) .
(b) The fuel consumption is decreasing by 0.05( gal / h) / (mi / h) as the car’s speed reaches 20 mi / h
. So if you increase your speed to 21 mi / h , you could expect to decrease your fuel consumption by
about 0.05( gal / h) / (mi / h) .

/
30. (a) f (8) is the rate of change of the quantity of coffee sold with respect to the price per pound
/
when the price is $8 per pound. The units for f (8) are pounds /( dollars / pound ) .
/
(b) f (8) is negative since the quantity of coffee sold will decrease as the price charged for it
increases. People are generally less willing to buy a product when its price increases.

/
31. T (10) is the rate at which the temperature is changing at 10:00 A.M. To estimate the value of
/
T (10) , we will average the difference quotients obtained using the times t=8 and t=12 . Let
T (8) T (10) 72 81 T (12) T (10) 88 81
A= = =4.5 and B= = =3.5 . Then
8 10  2 12 10 2
/ T (t) T (10) A+B 4.5+3.5 
T (10)= lim  = =4 F / h .
t 10 t 10 2 2

32. For 1910: We will average the difference quotients obtained using the years 1900 and 1920.

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

E(1900) E(1910) 48.3 51.1

Let A= = =0.28 and
1900 1910  10
E(1920) E(1910) 55.2 51.1
B= = =0.41 . Then
1920 1910 10
/ E(t) E(1910) A+B
E (1910)= lim  =0.345 . This means that life expectancy at birth was
t 1910 t 1910 2
increasing at about 0.345 year / year in 1910.
For 1950: Using data for 1940 and 1960 in a similar fashion, we obtain
/
E (1950) [0.31+0.10]/2=0.205 . So life expectancy at birth was increasing at about 0.205 year /
year in 1950.

/
33. (a) S (T ) is the rate at which the oxygen solubility changes with respect to the water temperature.
Its units are ( mg / L )/ C .
(b) For T =16 C , it appears that the tangent line to the curve goes through the points (0,14) and (32,6)
/ 6 14 8
. So S (16) = = 0.25( mg / L )/ C . This means that as the temperature increases past
32 0 32
16 C , the oxygen solubility is decreasing at a rate of 0.25( mg / L )/ C .

/
34. (a) S (T ) is the rate of change of the maximum sustainable speed of Coho salmon with respect to
the temperature. Its units are ( cm / s )/ C .

(b) For T =15 C, it appears the tangent line to the curve goes through the points (10,25) and (20,32) .
/ 32 25
So S (15) =0.7( cm / s )/ C . This tells us that at T =15 C, the maximum sustainable speed
20 10

of Coho salmon is changing at a rate of 0.7 ( cm / s )/ C . In a similar fashion for T =25 C, we can
/ 25 35
use the points (20,35) and (25,25) to obtain S (25) = 2 (cm / s) / C . As it gets warmer than
25 20

20 C, the maximum sustainable speed decreases rapidly.

35. Since f (x)=xsin (1/x) when x

0 and f (0)=0 , we have
/ f (0+h) f (0) hsin (1/h) 0
f (0)=lim =lim =lim (sin(1/h)) . This limit does not exist since sin(1/h)
h 0 h h 0 h h 0
takes the values  1 and 1 on any interval containing 0 . (Compare with Example 4 in Section 2.2.)

2
36. Since f (x)=x sin(1/x) when x
0 and f (0)=0 , we have
2
/ f (0+h) f (0) h sin (1/h) 0 1
f (0)=lim =lim =lim (h sin(1/h)) . Since  1 sin  1 , we have
h 0 h h 0 h h 0 h

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.8 Derivatives

1 1
 h h sin  h   h  hsin  h . Because lim ( h )=0 and lim h =0 , we know that
h h h 0 h 0
1 /
lim (h sin )=0 by the Squeeze Theorem. Thus, f (0)=0 .
h 0 h

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

1. Note: Your answers may vary depending on your estimates. By estimating the slopes of tangent
lines on the graph of f , it appears that
/ /
(a) f (1)  2 (b) f (2) 0.8
/ /
(c) f (3)  1 (d) f (4)  0.5

2. Note: Your answers may vary depending on your estimates. By estimating the slopes of tangent
lines on the graph of f , it appears that
/ /
(a) f (0)  3 (b) f (1) 0
/ /
(c) f (2) 1.5 (d) f (3) 2
/ /
(e) f (4) 0 (f) f (5)  1.2

/ / /
3. It appears that f is an odd function, so f will be an even function ??? that fis, ( a)= f (a)
/ /
(a) f ( 3) 1.5 (b) f ( 2) 1
/ /
(c) f ( 1)  0 (d) f (0)  4
/ /
(e) f (1)  0 (f) f (2)  1
/
(g) f (3) 1.5

1
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

/
4. (a) (a) = II, since from left to right, the slopes of the tangents to graph (a) start out negative,
become 0 , then positive, then 0 , then negative again. The actual function values in graph II follow
the same pattern.
/
(b) (b) = IV, since from left to right, the slopes of the tangents to graph (b) start out at a fixed
positive quantity, then suddenly become negative, then positive again. The discontinuities in graph IV
indicate sudden changes in the slopes of the tangents.
/
(c) (c) = I, since the slopes of the tangents to graph (c) are negative for x<0 and positive for x>0
, as are the function values of graph I.
/
(d) (d) = III, since from left to right, the slopes of the tangents to graph (d) are positive, then 0 ,
then negative, then 0 , then positive, then 0 , then negative again, and the function values in graph III
follow the same pattern.

5.

6.
2
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

7.

8.

9.

10.

3
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

11.

12.

13.

4
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

14. The slopes of the tangent lines on the graph of y=P(t) are always positive, so the y  values of
/
y=P (t) are always positive. These values start out relatively small and keep increasing, reaching a
/
maximum at about t=6 . Then the y  values of y=P (t) decrease and get close to zero. The graph of
/
P tells us that the yeast culture grows most rapidly after 6 hours and then the growth rate declines.

15. It appears that there are horizontal tangents on the graph of M for t=1963 and t=1971 . Thus, there
/
are zeros for those values of t on the graph of M . The derivative is negative for the years 1963 to
1971.

16. See Figure 1 in Section 3.4.

17.

5
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

The slope at 0 appears to be 1 and the slope at 1 appears to be 2.7 . As x decreases, the slope gets
/ x
closer to 0 . Since the graphs are so similar, we might guess that f (x)=e .

18.

/ /
As x increases toward 1 , f (x) decreases from very large numbers to 1 . As x becomes large, f (x)
/ 2 /
gets closer to 0 . As a guess, f (x)=1/x or f (x)=1/x make sense.

/ / 1 / /
19. (a) By zooming in, we estimate that f (0)=0 , f =1 , f (1)=2 , and f (2)=4 .
2
/ / / 1
(b) By symmetry, f ( x)= f (x) . So f  = 1 ,
2
/ /
f ( 1)= 2 , and f ( 2)= 4 .
/ /
(c) It appears that f (x) is twice the value of x , so we guess that f (x)=2x .
(d)
6
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

2 2
/ f (x+h) f (x) (x+h)  x
f (x)=lim =lim
h 0 h h 0 h
2 2 2 2
(x +2hx+h ) x 2hx+h h(2x+h)
=lim =lim =lim =lim (2x+h)=2x
h 0 h h 0 h h 0 h h 0

/ / 1 / /
20. (a) By zooming in, we estimate that f (0)=0 , f  0.75 , f (1) 3 , f (2) 12 , and
2
/
f (3) 27 .
/ / / 1 / /
(b) By symmetry, f ( x)= f (x) . So f   0.75 , f ( 1) 3 , f ( 2) 12 , and
2
/
f ( 3) 27 .

(c)
/ / / 2 /
(d) Since f (0)=0 , it appears that f may have the form f (x)=ax . Using f (1)=3 , we have
/ 2
a=3 , so f (x)=3x .
3 3 3 2 2 3 3
/ f (x+h) f (x) (x+h)  x (x +3x h+3xh +h ) x
(e) f (x)=lim =lim =lim
h 0 h h 0 h h 0 h
2 2 3 2 2
3x h+3xh +h h(3x +3xh+h ) 2 2 2
=lim =lim =lim (3x +3xh+h )=3x
h 0 h h 0 h h 0

/ f (x+h) f (x) 37 37 0

21. f (x)=lim =lim =lim =lim 0=0
h 0 h h 0 h h 0 h h 0
/
Domain of f = domain of f =R .

7
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

22.
/ f (x+h) f (x) [12+7(x+h)] (12+7x)
f (x) =lim h
=lim
h
h 0 h 0
12+7x+7h 12 7x 7h
=lim =lim =lim 7=7
h 0 h h 0 h h 0

/
Domain of f = domain of f =R .

23.
2 2
/ f (x+h) f (x) [1 3(x+h) ] (1 3x )
f (x) =lim =lim
h 0 h h 0 h
2 2 2 2 2 2
[1 3(x +2xh+h )] (1 3x ) 1 3x  6xh 3h  1+3x
=lim =lim
h 0 h h 0 h
2
 6xh 3h h( 6x 3h)
=lim =lim =lim ( 6x 3h)= 6x
h 0 h h 0 h h 0

/
Domain of f = domain of f =R .

24.
2 2
/ f (x+h) f (x) [5(x+h) +3(x+h) 2] (5x +3x 2)
f (x) =lim =lim
h 0 h h 0 h
2 2 2 2
5x +10xh+5h +3x+3h 2 5x  3x+2 10xh+5h +3h
=lim =lim
h 0 h h 0 h
h(10x+5h+3)
=lim =lim (10x+5h+3)=10x+3
h 0 h h 0

/
Domain of f = domain of f =R .

25.
3 3
/ f (x+h) f (x) [(x+h)  3(x+h)+5] (x  3x+5)
f (x) =lim =lim
h 0 h h 0 h

8
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

3 2 2 3 3
(x +3x h+3xh +h  3x 3h+5) (x  3x+5)
=lim
h 0 h
2 2 3 2 2
3x h+3xh +h  3h h(3x +3xh+h  3)
=lim =lim
h 0 h h 0 h
2 2 2
=lim (3x +3xh+h  3)=3x  3
h 0

/
Domain of f = domain of f =R .

26.
/ f (x+h) f (x) (x+h+ x+h ) (x+ x )
f (x) =lim h
=lim
h
h 0 h 0

h x+h  x x+h + x (x+h) x

=lim +  =lim 1+
h 0 h h x+h + x h 0 h( x+h + x )
1 1 1
=lim 1+ =1+ =1+
h 0 x+h + x x+ x 2 x

/
Domain of f = 0, ) , domain of f = ( 0, ) .
27.
/ g(x+h) g(x) 1+2(x+h)  1+2x 1+2(x+h) + 1+2x
g (x) =lim h
=lim
h
h 0 h 0 1+2(x+h) + 1+2x
(1+2x+2h) (1+2x) 2 2 1
=lim =lim = =
h 0 h[ 1+2(x+h) + 1+2x ] h 0 1+2x+2h + 1+2x 2 1+2x 1+2x

1 / 1
Domain of g=  , , domain of g =  , .
2 2

28.
3+(x+h) 3+x
/ 
f (x) =lim f (x+h) f (x) 1 3(x+h) 1 3x
=lim
h 0 h h 0 h

9
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

(3+x+h)(1 3x) (3+x)(1 3x 3h)

=lim
h 0 h(1 3x 3h)(1 3x)
2 2
(3 9x+x 3x +h 3hx) (3 9x 9h+x 3x  3hx)
=lim
h 0 h(1 3x 3h)(1 3x)
10h 10 10
=lim =lim =
h 0 h(1 3x 3h)(1 3x) h 0 (1 3x 3h)(1 3x) 2
(1 3x)

/ 1 1
Domain of f = domain of f =   ,  , .
3 3

29.
4(t+h) 4t 4(t+h)(t+1) 4t(t+h+1)
/ 
G (t) =lim G(t+h) G(t) =lim (t+h)+1 t+1 =lim (t+h+1)(t+1)
h 0 h h 0 h h 0 h
2 2
(4t +4ht+4t+4h) (4t +4ht+4t)
=lim
h 0 h(t+h+1)(t+1)
4h 4 4
=lim =lim =
h 0 h(t+h+1)(t+1) h 0 (t+h+1)(t+1) 2
(t+1)

/
Domain of G= domain of G =(  , 1) ( 1, ).

30.
1 1

/ 2 2 2 2
g (x) =lim g(x+h) g(x) =lim (x+h) x x  (x+h)
=lim
h 0 h h 0 h h 0 2 2
h(x+h) x
2 2 2 2
x  (x +2xh+h )  2xh h  2x h  2x  3
=lim =lim =lim = = 2x
h 0 h h 0 h(x+h)2 x2 h 0 (x+h)2 x2 4
x

/
Domain of g= domain of g = { x| x 0} .

31.

10
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

4 4 4 3 2 2 3 4 4
/ f (x+h) f (x) (x+h)  x (x +4x h+6x h +4xh +h ) x
f (x) =lim =lim =lim
h 0 h h 0 h h 0 h
3 2 2 3 4
4x h+6x h +4xh +h 3 2 2 3 3
=lim =lim (4x +6x h+4xh +h )=4x
h 0 h h 0

/
Domain of f = domain of f =R .

32. (a)

/ 
(b) Note that the third graph in part (a) has small negative values for its slope, f ; but as x 6 ,
/
f    .
See the graph in part (d).
(c)
/ f (x+h) f (x) 6 (x+h)  6 x 6 (x+h) + 6 x
f (x) =lim h
=lim
h
h 0 h 0 6 (x+h) + 6 x
[6 (x+h)] (6 x)  h
=lim =lim
h 0 h[ 6 (x+h) + 6 x ] h 0 h( 6 x h + 6 x )
 1  1
=lim =
h 0 6 x h + 6 x 2 6 x

Domain of f = (   ,6 , domain of
11
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

/
f =(  ,6 ) .

(d)

33. (a)
2 2
/ x+h  x
f (x) =lim f (x+h) f (x) =lim x+h x
h 0 h h 0 h

2 2
h +
(x+h) x  2x+2(x+h) 2h
=lim =lim 1+ =lim 1+
h 0 h h 0 hx(x+h) h 0 hx(x+h)

2 2
=lim 1+ =1+
h 0 x(x+h) 2
x

/ /
(b) Notice that when f has steep tangent lines, f (x) is very large. When f is flatter, f (x) is smaller.

34. (a)
6 6

2 2 2 2
/
f (t) =lim f (t+h) f (t) =lim 1+(t+h) 1+t 6+6t  6 6(t+h)
=lim
h 0 h h 0 h h 0
2 2
h[1+(t+h) ](1+t )
2
 12th 6h  12t 6h  12t
=lim 2 2
=lim 2 2
=
22
h 0 h[1+(t+h) ](1+t ) h 0 [1+(t+h) ](1+t ) (1+t )

/ /
(b) Notice that f has a horizontal tangent when t=0 . This corresponds to f (0)=0 . f is positive
when f is increasing and negative when f is decreasing.
12
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

/
35. (a) U (t) is the rate at which the unemployment rate is changing with respect to time. Its units are
percent per year.
/ U (t+h) U (t) U (t+h) U (t)
(b) To find U (t) , we use lim  for small values of h .
h 0 h h

/ U (1992) U (1991) 7.5 6.8

For 1991: U ( 1991 ) = = =0.70
1992 1991 1

/
For 1992: We estimate U (1992) by using h= 1 and h=1 , and then average the two results to obtain
a final estimate.
/ U (1991) U (1992) 6.8 7.5
h= 1
U (1992) = =0.70 ;
1991 1992  1
/ U (1993) U (1992) 6.9 7.5
h=1
U (1992) = = 0.60 .
1993 1992 1
/ 1
So we estimate that U ( 1992 )  [0.70+( 0.60)]=0.05 .
2

t 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
/
U (t) 0.70 0.05  0.70  0.65  0.35  0.35  0.45  0.35  0.25  0.20

/
36. (a) P (t) is the rate at which the percentage of Americans under the age of 18 is changing with
respect to time. Its units are percent per year (%/yr).
/ P(t+h) P(t) P(t+h) P(t)
(b) To find P (t) , we use lim  for small values of h .
h 0 h h
/ P(1960) P(1950) 35.7 31.1
For 1950: P (1950)= = =0.46
1960 1950 10
/
For 1960: We estimate P (1960) by using h= 10 and h=10 , and then average the two results to
obtain a final estimate.
/ P(1950) P(1960) 31.1 35.7
h= 10
P (1960) = =0.46
1950 1960  10
13
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

/ P(1970) P(1960) 34.0 35.7

h=10
P (1960) = = 0.17
1970 1960 10
/ 1
So we estimate that P (1960) [0.46+( 0.17)]=0.145 .
2

t 1950 1960 1970 1980 1990 2000

P (t) 0.460 0.145  0.385  0.415  0.115 0.000

/

(c)
/
(d) We could get more accurate values for P (t) by obtaining data for the mid decade years 1955,
1965, 1975, 1985, and 1995.

37. f is not differentiable at x= 1 or at x=11 because the graph has vertical tangents at those points;
at x=4 , because there is a discontinuity there; and at x=8 , because the graph has a corner there.

38. (a) g is discontinuous at x= 2 (a removable discontinuity), at x=0 ( g is not defined there), and at
x=5 (a jump discontinuity).
(b) g is not differentiable at the points mentioned in part (a) (by Theorem 4), nor is it differentiable at
x= 1 (corner), x=2 (vertical tangent), or x=4 (vertical tangent).

39. As we zoom in toward ( 1,0) , the curve appears more and more like a straight line, so
f (x)=x+ x is differentiable at x= 1 . But no matter how much we zoom in toward the origin, the
curve doesn’t straighten out   we can’t eliminate the sharp point (a cusp). So f is not differentiable
at x=0 .

14
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

40. As we zoom in toward (0,1) , the curve appears more and more like a straight line, so f is
differentiable at x=0 . But no matter how much we zoom in toward (1,0) or ( 1,0) , the curve doesn’t
straighten out   we can’t eliminate the sharp point (a cusp). So f is not differentiable at x= 1 .

41. (a) Note that we have factored x a as the difference of two cubes in the third step.
1/3 1/3 1/3 1/3
/ f (x) f (a) x  a x  a
f (a) =lim x a
=lim
x a
=lim 1/3 1/3 2/3 1/3 1/3 2/3
x a x a x a (x  a )(x +x a +a )
1 1 1  2/3
=lim = or a
x a 2/3 1/3 1/3 2/3 2/3 3
x +x a +a 3a

3
/ f (0+h) f (0) h 0 1
(b) f (0)=lim =lim =lim . This function increases without bound, so the
h 0 h h 0 h h 0 h2/3
/
limit does not exist, and therefore f (0) does not exist.
/ 1
(c) lim f (x) =lim = and f is continuous at x=0 (root function), so f has a vertical tangent
x 0 x 0 3x2/3
at x=0 .

2/3
/ g(x) g(0) x  0 1
42. (a) g (0)=lim =lim =lim , which does not exist.
x 0 x 0 x 0 x x 0 x1/3
(b)
2/3 2/3 1/3 1/3 1/3 1/3
/ g(x) g(a) x  a (x  a )(x +a )
g (a) =lim x a
=lim
x a
=lim 1/3 1/3 2/3 1/3 1/3 2/3
x a x a x a (x  a )(x +x a +a )
1/3 1/3 1/3
x +a 2a 2 2  1/3
=lim = = or a
x a 2/3 1/3 1/3 2/3 2/3 1/3 3
x +x a +a 3a 3a

2/3 / 2
(c) g(x)=x is continuous at x=0 and lim g (x) =lim = . This shows that
1/3
x 0 x 0 3 x
g has a vertical tangent line at x=0 .
15
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

(d)

43. f (x)=|x 6| =
{ x(x6 6) if x<6
if x 6 {
= 6 x if x<6
x 6 if x 6
f (x) f (6) x 6  0 x 6
lim = lim = lim = lim 1=1 .
+ x 6 + x 6 + x 6 +
x 6 x 6 x 6 x 6
But
f (x) f (6) x 6  0 6 x
lim = lim = lim
 x 6  x 6  x 6
x 6 x 6 x 6

= lim ( 1)= 1

x 6
/ /
So, f (6) does not exist. However, f (x)=
{ 1 1 if x<6
if x 6
Another way of writing the answer is
/ x 6
f (x) = .
|x 6|

44. f (x)=[[x]] is not continuous at any integer n , so f is not differentiable at n by the contrapositive
/
of Theorem 4. If a is not an integer, then f is constant on an open interval containing a , so f (a)=0 .
/
Thus, f (x)=0 , x not an integer.

45. (a) f (x)=x x =

{ x
2

 x
2
if x 0
if x<0

16
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

2 /
(b) Since f (x)=x for x 0 , we have f (x)=2x for x>0 . [See Exercise 2.9.19(d).] Similarly, since
2 /
f (x)= x for x<0 , we have f (x)= 2x for x<0 . At x=0 , we have
/ f (x) f (0) x x
f (0)=lim =lim =lim x =0.
x 0 x 0 x 0 x x 0
So f is differentiable at 0 . Thus, f is differentiable for all x .
/
(c) From part (b), we have f (x)=
2x if x 0
 2x if x<0 {=2 x . }
/ f (4+h) f (4) 5 (4+h) 1  h
46. (a) f (4)= lim = lim = lim = 1 and
  h  h  h
h 0 h 0 h 0
1
 1
/ f (4+h) f (4) 5 (4+h) 1 (1 h) 1
f (4)= lim = lim = lim = lim =1 .
+ + h + h + h(1 h) + 1 h
h 0 h 0 h 0 h 0

(b)

(c) f (x)=
0
{
5 x
if x 0
if 0<x<4 These expressions show that f is continuous on the intervals
1/(5 x) if x 4
(  ,0) , (0,4) , (4,5) and (5, ) . Since lim f (x)= lim (5 x)=5 0= lim f (x) , lim f (x) does not exist,
+ +  x 0
x 0 x 0 x 0
so f is discontinuous (and therefore not differentiable) at 0 .
1
At 4 we have lim f (x)= lim (5 x)=1 and lim f (x)= lim =1 , so lim f (x)=1= f (4) and f is
  + + 5 x x 4
x 4 x 4 x 4 x 4
continuous at 4 . Since f (5) is not defined, f is discontinuous at 5 .
/ /
(d) From (a), f is not differentiable at 4 since f (4) f (4) , and from (c), f is not differentiable at 0
 +
or 5 .

17
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

47. (a) If f is even, then

/ f ( x+h) f ( x) f [ (x h)] f ( x) f (x h) f (x)
f ( x) = lim h
=lim
h
=lim
h
h 0 h 0 h 0

f (x h) f (x) f (x+ x) f (x) /

=  lim [let x= h] = lim = f (x)
h 0  h  x 0  x
/
Therefore, f is odd.
(b) If f is odd, then
/ f ( x+h) f ( x) f [ (x h)] f ( x)  f (x h)+ f (x)
f ( x) = lim =lim =lim
h 0 h h 0 h h 0 h
f (x h) f (x) f (x+ x) f (x) /
= lim [let x= h] = lim = f (x)
h 0  h  x 0  x
/
Therefore, f is even.

48. (a)
(b) The initial temperature of the water is close to room temperature because of the water that was in
the pipes. When the water from the hot water tank starts coming out, dT /dt is large and positive as T
increases to the temperature of the water in the tank. In the next phase, dT /dt=0 as the water comes
out at a constant, high temperature. After some time, dT /dt becomes small and negative as the
contents of the hot water tank are exhausted. Finally, when the hot water has run out, dT /dt is once
again 0 as the water maintains its (cold) temperature.

(c)

49.
In the right triangle in the diagram, let  y be the side opposite angle  and  x the side adjacent
angle  . Then the slope of the tangent line  is m= y/ x=tan  . Note that
18
Stewart Calculus ET 5e 0534393217;2. Limits and Derivatives; 2.9 The Derivative as a Function

 2 /
0< < . We know (see Exercise 19) that the derivative of f (x)=x is f (x)=2x . So the slope of the
2

tangent to the curve at the point (1,1) is 2 . Thus,  is the angle between 0 and whose tangent is 2
2
 1 
; that is,  =tan 2 63 .

19
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

h
e 1
1. (a) e is the number such that lim =1 .
h 0 h
(b)
x x
x (2.7  1)/x x (2.8  1)/x
 0.001 0.9928  0.001 1.0291
 0.0001 0.9932  0.0001 1.0296
0.001 0.9937 0.001 1.0301
0.0001 0.9933 0.0001 1.0297

h h
2.7  1 2.8  1
From the tables (to two decimal places), lim =0.99 and lim =1.03 . Since
h 0 h h 0 h
0.99<1<1.03 , 2.7<e<2.8 .

2. (a)
The function value at x=0 is 1 and the slope at x=0 is 1 .
x e d x x d e e 1
(b) f (x)=e is an exponential function and g(x)=x is a power function. (e )=e and (x )=ex
dx dx
.
x e
(c) f (x)=e grows more rapidly than g(x)=x when x is large.

/
3. f (x)=186.5 is a constant function, so its derivative is 0 , that is, f (x)=0 .

/
4. f (x)= 30 is a constant function, so its derivative is 0 , that is, f (x)=0 .

/
5. f (x)=5x 1 f (x)=5 0=5

10 / 10 1 9
6. F(x)= 4x  F (x)= 4(10x )= 40x

2 / 2 1
7. f (x)=x +3x 4 f (x)=2x +3 0=2x+3
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

8 5 / 8 1 5 1 7 4
8. g(x)=5x  2x +6 g (x)=5(8x ) 2(5x )+0=40x  10x

8 5 / 8 1 5 1 7 4
9. g(x)=5x  2x +6 g (x)=5(8x ) 2(5x )+0=40x  10x

1 6 4 / 1 5 3 5 3
10. f (t)= t  3t +t  f (t)= (6t ) 3(4t )+1=3t  12t +1
2 2

 2/5 / 2 ( 2/5) 1 2  7/5 2

11. y=x  y = x = x =
5 5 7/5
5x

x / x x
12. y=5e +3 y =5(e )+0=5e

4 3 / 4 2 2
13. V (r)=  r  V (r)=  (3r )=4 r
3 3

 3/5 / 3 ( 3/5) 1  8/5

14. R(t)=5t  R (t)=5  t = 3t
5

 9 /  10  10
15. Y (t)=6t  Y (t)=6( 9)t = 54t

10  7 /  8 7 10
16. R(x)= = 10 x  R (x)= 7 10 x =
7 8
x x

x 1/2 x / 1  1/2 x 1 x
17. G(x)= x  2e =x  2e  G (x)= x  2e =  2e
2 2 x
3 1/3 / 1  2/3 1
18. y= x =x  y = x =
3 2/3
3x

1 5 1 5 5 1 5 / 1 4 5 4
19. F(x)= x = x= x F (x)= (5x )= x
2 2 32 32 32

1 1/2  1/2 / 1  1/2 1  3/2 1 1

20. f (t)= t  =t  t  f (t)= t   t = +
t 2 2 2 t 2t t

2 1 2  2 /  3 2
21. g(x)=x + =x +x  g (x)=2x+( 2)x =2x
2 3
x x

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

3/2 1/2 / 3 1/2 1  1/2 1  1/2 1  1/2

22. y= x (x 1)=x  x  y = x  x = x (3x 1) [ factor out x ]
2 2 2 2
/ 3x 1
or y = .
2 x

2
x +4x+3 3/2 1/2  1/2
23. y= =x +4x +3x 
x
/ 3 1/2 1  1/2 1  3/2 3 2 3
y = x +4 x +3  x = x+ 
2 2 2 2 x 2x x
3/2 2/2 1/2
note that x =x  x =x x

2
x 2 x  1/2 / 1  3/2
24. y= =x 2x  y =1 2  x =1+1/(x x )
x 2

2 / 2
25. y=4  y =0 since 4 is a constant.

/ 1  1/2
26. g(u)= 2 u+ 3u = 2 u+ 3 u  g (u)= 2 (1)+ 3 u = 2 + 3 /(2 u )
2

2 /
27. y=ax +bx+c y =2ax+b

v b c v  1  2 / v  2  3 v b 2c
28. y=ae + + =ae +bv +cv  y =ae  bv  2cv =ae  
v 2 2 3
v v v

2 1 2  3/4 / 3  7/4 3 3
29. v=t  =t  t  v =2t  t =2t+ =2t+
4 3 4 7/4 4 3
t 4t 4t t

3 2 3 2/3 3/2 / 2  1/3 3 1/2 2

30. u= t +2 t =t +2t  u = t +2 t = +3 t
3 2 3
3 t

A y  10 y /  11 y 10A y
31. z= +Be =Ay +Be  z = 10Ay +Be = +Be
10 11
y y

x+1 x 1 x / x x+1
32. y=e +1=e e +1=e e +1 y =e e =e

x / x
33. f (x)=e  5x f (x)=e  5 .
3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

/ / /
Notice that f (x)=0 when f has a horizontal tangent, f is positive when f is increasing, and f is
negative when f is decreasing.

5 3 / 4 2
34. f (x)=3x  20x +50x f (x)=15x  60x +50 .

/ /
Notice that f (x)=0 when f has a horizontal tangent and that f is an even function while f is an
odd function.

15 3 / 14 2
35. f (x)=3x  5x +3 f (x)=45x  15x .

/ / /
Notice that f (x)=0 when f has a horizontal tangent, f is positive when f is increasing, and f is
negative when f is decreasing.

 1 /  2 2
36. f (x)=x+1/x=x+x  f (x)=1 x =1 1/x .

/ / /
Notice that f (x)=0 when f has a horizontal tangent, f is positive when f is increasing, and f is
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

negative when f is decreasing.

/ 2 3
37. To graphically estimate the value of f (1) for f (x)=3x  x , we’ll graph f in the viewing
rectangle [1 0.1,1+0.1] by [ f (0.9),f(1.1)] , as shown in the figure. If we have sufficiently zoomed in
on the graph of f , we should obtain a graph that looks like a diagonal line; if not, graph again with
1 0.01 and 1+0.01 , etc.
/ 2.299 1.701 0.589
Estimated value: f (1) = =2.99 .
1.1 0.9 0.2
2 3 / 2 /
Exact value: f (x)=3x  x  f (x)=6x 3x , so f (1)=6 3=3 .

38. See the previous exercise. Since f is a decreasing function, assign Y (3.9) to Y and Y (4.1) to
1 1
Y .
min
/ 0.49386 0.50637  0.01251
Estimated value: f (4) = = 0.06255 .
4.1 3.9 0.2
 1/2 / 1  3/2 / 1  3/2 1 1 1
Exact value: f (x)=x  f (x)= x , so f (4)= (4 )= = = 0.0625 .
2 2 2 8 16

4 x / 3 x /
39. y=x +2e  y =4x +2e . At (0, 2) , y =2 and an equation of the tangent line is y 2=2(x 0) or
y=2x+2.

2 2 / /
40. y=(1+2x) =1+4x+4x  y =4+8x. At (1, 9), y =12 and an equation of the tangent line is
y 9=12(x 1) or y=12x 3.

2 3 / 2 /
41. y=3x  x  y =6x 3x . At (1, 2) , y =6 3=3 , so an equation of the tangent line is y 2=3(x 1) , or
y=3x 1 .

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

3/2 3 1/2 // 3
42. y=x x =x  y =
x . At (4, 8) , y = (2)=3, so an equation of the tangent line is
2 2
y 8=3(x 4), or y=3x 4 .

43. (a)

(b)

/
From the graph in part (a), it appears that f is zero at x   1.25 , x  0.5 , and x  3 . The slopes are
1 2 3
/ /
negative (so f is negative) on (  , x ) and (x , x ). The slopes are positive (so f is positive) on
1 2 3
(x , x ) and (x ,  ).
1 2 3
4 3 2 / 3 2
(c) f (x)=x  3x  6x +7x+30 f (x)=4x  9x  12x+7

44. (a)

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

(b)

/
From the graph in part (a), it appears that f is zero at x  0.2 and x  2.8 . The slopes are positive
1 2
/ /
(so f is positive) on (  , x ) and (x ,  ) . The slopes are negative (so f is negative) on (x , x ) .
1 2 1 2
x 2 / x
(c) g(x)=e  3x  g (x)=e  6x

3 2 / 2
45. The curve y=2x +3x  12x+1 has a horizontal tangent when y =6x +6x 12=0
2
6(x +x 2)=0 6(x+2)(x 1)=0 x= 2 or x=1 . The points on the curve are ( 2, 21)
and (1,  6) .

3 2 / 2
46. f (x)=x +3x +x+3 has a horizontal tangent when f (x)=3x +6x+1=0
 6
36 12 1
x= = 1
6 .
6 3

3 / 2 2
47. y=6x +5x 3 m=y =18x +5 , but x  0 for all x , so m 5 for all x .

x / x
48. The slope of y=1+2e  3x is given by m=y =2e  3 .
The slope of 3x y=5 y=3x 5 is 3 .
x x
m=3 2e  3=3 e =3 x=ln 3 . This occurs at the point (ln 3, 7 3ln 3) (1.1, 3.7) .

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

49.

2
Let (a, a ) be a point on the parabola at which the tangent line passes through the point (0,  4). The
2
tangent line has slope 2a and equation y ( 4)=2a(x 0) y=2ax 4 . Since (a, a ) also lies on the line,
2 2
a =2a(a) 4, or a =4 . So a=
2 and the points are (2, 4) and ( 2, 4).

2 / 2
50. If y=x +x , then y =2x+1 . If the point at which a tangent meets the parabola is (a, a +a), then
the slope of the tangent is 2a+1 . But since it passes through (2,  3), the slope must also be
2
 y a +a+3
= .
 x a 2
2
a +a+3 2 2
Therefore, 2a+1= . Solving this equation for a we get a +a+3=2a  3a 2
a 2
2
a  4a 5=(a 5)(a+1)=0 a=5 or  1 . If a= 1 , the point is ( 1, 0) and the slope is  1, so the equation
is y 0=( 1)(x+1) or y= x 1. If a=5, the point is (5, 30) and the slope is 11, so the equation is
y 30=11(x 5) or y=11x 25.

2 / /
51. y= f (x)=1 x  f (x)= 2x, so the tangent line at (2,  3) has slope f (2)= 4. The normal line has
1 1 1 1 7
slope  = and equation y+3= (x 2) or y= x .
 4 4 4 4 2

2 / /
52. y= f (x)=x x  f (x)=1 2x. So f (1)= 1, and the slope of the normal line is the negative
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

reciprocal of that of the tangent line, that is,  1/( 1)=1. So the equation of the normal line at (1, 0) is
2
y 0=1(x 1) y=x 1. Substituting this into the equation of the parabola, we obtain x 1=x x
x=
1. The solution x= 1 is the one we require. Substituting x= 1 into the equation of the parabola
to find the y  coordinate, we have y= 2. So the point of intersection is ( 1,  2), as shown in the
sketch.

53.
1 1
/ 
f (x) =lim f (x+h) f (x) =lim x+h x
=lim
x (x+h)
h 0 h h 0 h h 0 hx(x+h)
 h  1 1
=lim =lim =
h 0 hx(x+h) h 0 x(x+h) 2
x

2
54. Substituting x=1 and y=1 into y=ax +bx gives us a+b=1 (1) . The slope of the tangent line
/ /
y=3x 2 is 3 and the slope of the tangent to the parabola at (x, y) is y =2ax+b . At x=1 , y =3
3=2a+b (2) . Subtracting (1) from (2) gives us 2=a and it follows that b= 1 . The parabola has
2
equation y=2x  x .

2
55. f (x)=2 x if x 1 and f (x)=x  2x+2 if x>1 . Now we compute the right and left hand
derivatives defined in Exercise :
/ f (1+h) f (1) 2 (1+h) 1  h
f (1)=lim =lim =lim =lim  1= 1 and
  h  h  h 
h 0 h 0 h 0 h 0
2 2
/ f (1+h) f (1) (1+h)  2(1+h)+2 1 h
f (1)=lim =lim =lim =lim h=0 .
+ + h + h + h +
h 0 h 0 h 0 h 0

/ / / /
Thus, f (1) does not exist since f (1) f (1) , so f is not differentiable at 1 . But f (x)= 1 for x<1
 +
/
and f (x)=2x 2 if x>1 .

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

 1 2x if x< 1
56. g(x)=
{
g( 1+h) g( 1)
x
x
2
if  1 x 1
if x>1
 1 2( 1+h)  1  2h
lim =lim =lim =lim (  2 ) = 2 and
 h  h  h 
h 0 h 0 h 0 h 0
2 2
g( 1+h) g( 1) ( 1+h)  1  2h+h
lim =lim =lim =lim ( 2+h)= 2,
+ h + h + h +
h 0 h 0 h 0 h 0
/
so g is differentiable at  1 and g ( 1)= 2.
2 2
g(1) g(1) (1)  1 2h+h
lim =lim =lim =lim (2+h)=2 and
 h  h  h 
h 0 h 0 h 0 h 0
g(1) g(1) (1) 1 h /
lim =lim =lim =lim 1=1 , so g (1) does not exist.
+ h + h + h +
h 0 h 0 h 0 h 0
 2 if x< 1
Thus, g is differentiable except when x=1 , and g (x)=
/
{ 2x if  1 x<1
1 if x>1

2 2
57. (a) Note that x  9<0 for x <9 | x|<3  3<x<3 . So

{
2
x 9 if x  3
f (x) =  x +9
2
x 9
2
if  3<x<3
ifx 3

/
f (x) =
{ 2x

2x
if x< 3
 2x if  3<x<3
if x>3
= { 2x if | x|>3
 2x if | x|<3

To show that
10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

/ f (3+h) f (3)
f (3) does not exist we investigate lim by computing the left and right hand
h 0 h
derivatives defined in Exercise .
2
/ f (3+h) f (3) [ (3+h) +9] 0
f (3)=lim =lim =lim ( 6 h)= 6 and
  h  h 
h 0 h 0 h 0
2 2
/ f (3+h) f (3) (3+h)  9  0 6h+h
f (3)=lim =lim =lim =lim (6+h)=6 .
+ + h + h + h +
h 0 h 0 h 0 h 0
f (3+h) f (3) /
Since the left and right limits are different, lim does not exist, that is, f (3) does not
h 0 h
/
exist. Similarly, f ( 3) does not exist. Therefore, f is not differentiable at 3 or at  3 .

(b)

58. If x 1 , then h(x)=| x 1|+| x+2|=x 1+x+2=2x+1.

If  2<x<1 , then h(x)= (x 1)+x+2=3.
If x  2, then h(x)= (x 1) (x+2)= 2x 1. Therefore,

 2x 1 if x  2  2 if x< 2
h(x)=
{ 3 if  2<x<1
2x+1 if x 1

/
h (x)=
{ 0 if  2<x<1
2 if x>1

/ h(x) h(1) h(x) h(1) 3 3

To see that h (1)=lim does not exist, observe that lim =lim =0 but
x 1 x 1  x 1  3 1
x 1 x 1
h(x) h(1) 2x 2 /
lim =lim =2 . Similarly, h ( 2) does not exist.
+ x 1 + x 1
x 1 x 1

11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

2 /
59. y= f (x)=ax  f (x)=2ax. So the slope of the tangent to the parabola at x=2 is m=2a(2)=4a. The
1
slope of the given line, 2x+y=b y= 2x+b, is seen to be  2, so we must have 4a= 2 a= . So
2
1 2
when x=2, the point in question has y  coordinate   2 = 2. Now we simply require that the given
2
line, whose equation is 2x+y=b, pass through the point (2,  2): 2(2)+( 2)=b b=2. So we must have
1
a= and b=2.
2

/ /
60. f is clearly differentiable for x<2 and for x>2. For x<2, f (x)=2x, so f (2)=4. For x>2,

/ / / /
f (x)=m, so f (2)=m. For f to be differentiable at x=2, we need 4= f (2)= f (2)=m. So
+  +
f (x)=4x+b. We must also have continuity at x=2, so 4= f (2)=lim f (x)=lim (4x+b)=8+b. Hence,
+ +
x 2 x 2
b= 4.

3 2 / 2
61. y= f (x)=ax +bx +cx+d f (x)=3ax +2bx+c. The point ( 2, 6) is on f , so f ( 2)=6
 8a+4b 2c+d=6 (1) . The point (2, 0) is on f , so f (2)=0 8a+4b+2c+d=0 (2) . Since there are
/ /
horizontal tangents at ( 2, 6) and (2, 0), f (
2)=0. f ( 2)=0 12a 4b+c=0 (3) and
/
f (2)=0 12a+4b+c=0 (4) . Subtracting equation (3) from (4) gives 8b=0 b=0. Adding (1) and (2)
gives 8b+2d=6, so d=3 since b=0. From (3) we have c= 12a, so (2) becomes 8a+4(0)+2( 12a)+3=0
3 3 9
 3=16a a= . Now c= 12a= 12 = and the desired cubic function is
16 16 4
3 3 9
y= x x+3.
16 4

c c / c
62. (a) xy=c y= . Let P= a, . The slope of the tangent line at x=a is y (a)= . Its
x a 2
a
c c c 2c 2c
equation is y = (x a) or y= x+ . so its y  intercept is . Setting y=0 gives x=2a, so
a 2 2 a a
a a
2c
the x  intercept is 2a. The midpoint of the line segment joining 0, and (2a, 0) is
a
c
a, =P.
a
(b) We know the x  and y  intercepts of the tangent line from part (a), so the area of the triangle
1 1 1
bounded by the axes and the tangent is (base)(height)= xy= (2a)(2c/a)=2c, a constant.
2 2 2

12
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.1 Derivatives of Polynomials and Exponential Functions

1000
63. Solution 1: Let f (x)=x . Then, by the definition of a derivative,
1000
/ f (x) f (1) x  1
f (1)=lim =lim . But this is just the limit we want to find, and we know (from
x 1 x 1 x 1 x 1
1000
/ 999 / 999 x  1
the Power Rule) that f (x)=1000x , so f (1)=1000(1) =1000 . So lim =1000 .
x 1 x 1
1000 999 998 997 2
Solution 2: Note that (x  1)=(x 1)(x +x +x +   +x +x+1). So
1000 999 998 997 2
x  1 (x 1)(x +x +x +   +x +x+1)
lim = lim
x 1 x 1 x 1 x 1
999 998 997 2
lim (x +x +x +   +x +x+1) = 1+1+1+...+1+1+1
= x 1
1000 ones
= 1000, as above.

64.
In order for the two tangents to intersect on the y  axis, the points of tangency must be at equal
2
distances from the y  axis, since the parabola y=x is symmetric about the y  axis. Say the points of
2 2 2
tangency are (a, a ) and ( a, a ), for some a>0. Then since the derivative of y=x is dy/dx=2x, the
2 2
left hand tangent has slope  2a and equation y a = 2a(x+a), or y= 2ax a , and similarly the right
hand tangent line has equation
2 2 2
y a =2a(x a), or y=2ax a . So the two lines intersect at (0,  a ). Now if the lines are perpendicular,
2 1 1
then the product of their slopes is  1, so ( 2a)(2a)= 1 a = a= . So the lines intersect at
4 2
1
0,  .
4

13
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

3 dV dV dx 2 dx
1. V =x  = =3x
dt dx dt dt

2 dA dA dr dr
2. (a) A= r  = =2 r
dt dr dt dt
dA dr 2
(b) =2 r =2 (30m)(1m/s)=60 m / s
dt dt

3 dy dy dx 2 2 dy
3. y=x +2x = =(3x +2)(5)=5(3x +2) . When x=2 , =5(14)=70 .
dt dx dt dt

2 2 dx dy dx dy dx y dy
4. x +y =25 2x +2y =0 x = y  = .
dt dt dt dt dt x dt
2 2 dy dx 4
When y=4 , x +4 =25 x= 3 . For =6 , = (6)= 8 .
dt dt  3

2 2 2 dz dx dy dz 1 dx dy 2 2 2
5. z =x +y  2z
=2x +2y  = x +y . When x=5 and y=12 , z =5 +12 
dt dt dt dt z dt dt
2 dx dy dz 1 46
z =169 z= 13 . For =2 and =3, = (5 2+12 3)= .
dt dt dt  13 13

2
3 dy dy dx 1 3 1/2 2dx 3x dx dy
6. y= 1+x  = = (1+x ) (3x ) = . With =4 when x=2 and y=3 ,
dt dx dt 2 dt 3 dt dt
2 1+x
3(4) dx dx
we have 4=  =2 cm / s.
2(3) dt dt

7. (a) Given: a plane flying horizontally at an altitude of 1 mi and a speed of 500 mi / h passes
directly over a radar station. If we let t be time (in hours) and x be the horizontal distance traveled by
the plane (in mi), then we are given that dx/dt=500 mi / h.
(b) Unknown: the rate at which the distance from the plane to the station is increasing when it is 2 mi
from the station. If we let y be the distance from the plane to the station, then we want to find dy/dt
when y=2 mi.

(c)
2 2
(d) By the Pythagorean Theorem, y =x +1 2y(dy/dt)=2x(dx/dt) .
dy x dx x 2 2 dy 3
(e) = = (500) . Since y =x +1 , when y=2 , x= 3 , so = (500)=250 3 433 mi / h.
dt y dt y dt 2

8. (a) Given: the rate of decrease of the surface area is 1 cm

1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

2 2
/ min. If we let t be time (in minutes) and S be the surface area (in cm ), then we are given that
2
dS/dt= 1 cm / s.
(b) Unknown: the rate of decrease of the diameter when the diameter is 10 cm. If we let x be the
diameter, then we want to find dx/dt when x=10 cm.

(c)
1 2 1 2 2
(d) If the radius is r and the diameter x=2r , then r= x and S=4 r =4 x = x 
2 2
dS dS dx dx
= =2 x .
dt dx dt dt
dS dx dx 1 dx 1 1
(e)  1= =2 x  = . When x=10 , = . So the rate of decrease is cm /
dt dt dt 2 x dt 20 20
min.

9. (a) Given: a man 6 ft tall walks away from a street light mounted on a 15  ft tall pole at a rate of 5
ft / s. If we let t be time (in s) and x be the distance from the pole to the man (in ft), then we are given
that dx/dt=5 ft / s.
(b) Unknown: the rate at which the tip of his shadow is moving when he is 40 ft from the pole. If we
let y be the distance from the man to the tip of
d
his shadow (in ft), then we want to find (x+y) when x=40 ft.
dt

(c)
15 x+y 2
(d) By similar triangles, =  15y=6x+6y 9y=6x y= x.
6 y 3
d d 2 5 dx 5 25
(e) The tip of the shadow moves at a rate of (x+y)= x+ x = = (5)= ft / s.
dt dt 3 3 dt 3 3

10. (a) Given: at noon, ship A is 150 km west of ship B; ship A is sailing east at 35 km / h, and ship B
is sailing north at 25 km / h. If we let t be time (in hours), x be the distance traveled by ship A (in
km), and y be the distance traveled by ship B (in km), then we are given that dx/dt=35 km / h and
dy/dt=25 km / h.
(b) Unknown: the rate at which the distance between the ships is changing at 4:00 P.M. If we let z be
the distance between the ships, then we want to find dz/dt when t=4 h.
2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

(c)
2 2 2 dz dx dy
(d) z =(150 x) +y  2z =2(150 x)  +2y
dt dt dt
2 2
(e) At 4:00 P.M., x=4(35)=140 and y=4(25)=100 z= (150 140) +100 = 10,100 . So
dz 1 dx dy  10(35)+100(25) 215
= (x 150) +y = =  21.4 km / h.
dt z dt dt 10,100 101

11.
dx dy 2 2 2 dz dx dy
We are given that =60 mi / h and =25 mi / h. z =x +y  2z =2x +2y 
dt dt dt dt dt
dz dx dy dz 1 dx dy
z =x +y  = x +y .
dt dt dt dt z dt dt
2 2
After 2 hours, x=2(60)=120 and y=2(25)=50 z= 120 +50 =130 , so
dz 1 dx dy 120(60)+50(25)
= x +y = =65 mi / h.
dt z dt dt 130

12.
dx y 2 24 dy 24 dx 24
We are given that =1.6 m / s. By similar triangles, =  y=  = = (1.6) .
dt 12 x x dt 2 dt 2
x x
dy 24(1.6)
When x=8 , = = 0.6 m / s, so the shadow is decreasing at a rate of 0.6 m / s.
dt 64

13.
dx dy 2 2 2 dz dx dy
We are given that =4 ft / s and =5 ft / s. z =(x+y) +500  2z =2(x+y) + . 15
dt dt dt dt dt

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

minutes after the woman starts, we have x=(4ft/s)(20min)(60s/min)=4800 ft and y=5 15 60=4500
2 2
z= (4800+4500) +500 = 86,740,000 , so
dz x+y dx dy 4800+4500 837
= + = (4+5)=  8.99 ft / s .
dt z dt dt 86,740,000 8674

dx
14. We are given that =24 ft / s.
dt

(a)
2 2 2 dy
dx
y =(90 x) +90  2y =2(90 x)
. 
dt
dt
2 2 dy 90 x dx 45 24
When x=45 , y= 45 +90 =45 5 , so =  = ( 24)= ,
dt y dt 45 5 5
24
so the distance from second base is decreasing at a rate of  10.7 ft / s.
5

(b) Due to the symmetric nature of the problem in part (a), we expect to get the same answer and we
2 2 2 dz dx dz 45 24
do. z =x +90  2z =2x . When x=45 , z=45 5 , so = (24)=  10.7 ft / s.
dt dt dt 45 5 5

1 dh dA
15. A= bh , where b is the base and h is the altitude. We are given that =1 cm / min and =2
2 dt dt
2 dA 1 dh db
cm / min. Using the Product Rule, we have = b +h . When h=10 and A=100 , we
dt 2 dt dt
1 1 1 db db
have 100= b(10)  b=10  b=20 , so 2= 20 1+10  4=20+10 
2 2 2 dt dt
db 4 20
= = 1.6 cm / min.
dt 10

16.
dy dx 2 2 dy dx dx y dy y
Given = 1 m / s, find when x=8 m. y =x +1 2y =2x  = = . When x=8
dt dt dt dt dt x dt x
, y= 65 , so

4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

dx 65 65
= . Thus, the boat approaches the dock at  1.01 m / s.
dt 8 8

17.
dx dy 2 2 2 dz dx dy
We are given that =35 km / h and =25 km / h. z = ( x+y ) +100  2z =2(x+y) + .
dt dt dt dt dt
2 2
At 4:00 P.M., x=4(35)=140 and y=4(25)=100 z= (140+100) +100 = 67,600 =260 , so
dz x+y dx dy 140+100 720
= + = (35+25)=  55.4 km / h.
dt z dt dt 260 13

18. Let D denote the distance from the origin (0,0) to the point on the curve y= x .
2 2 2 2 2 dD 1 2  1/2 dx 2x+1 dx
D= (x 0) +(y 0) = x +( x ) = x +x  = (x +x) (2x+1) = . With
dt 2 dt 2 dt
2 x +x
dx dD 9 27
=3 when x=4 , = (3)=  3.02 cm / s.
dt dt 2 20 4 5

19.
dV 1 2
If C= the rate at which water is pumped in, then =C 10 , 000 , where V =  r h is the volume at
dt 3
r h 1 1 1 2  3 dV  2 dh
time t . By similar triangles, =  r= h V =  h h= h = h .
2 6 3 3 3 27 dt 9 dt
dh  2 800,000
When h=200cm , =20cm/min , so C 10 , 000= (200) (20) C=10 , 000+   289 ,
dt 9 9
3
253 cm / min.

20.
By similar triangles,

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

3 b 1 2 dV dh dh 2
= , so b=3h . The trough has volume V = bh(10)=5(3h)h=15h  12= =30h  =
1 h 2 dt dt dt 5h
1 dh 2 4
. When h= , = = ft / min.
2 dt 1 5
5
2

21.
1
The figure is labeled in meters. The area A of a trapezoid is ( base + base )( height ) , and the
2 1 2
volume V of the 10 meter long trough is 10A . Thus, the volume of the trapezoid with height h is
1 a 0.25 1
V =(10) [0.3+(0.3+2a)]h . By similar triangles, = = , so
2 h 0.5 2
2 dV dV dh dh dh 0.2
2a=h V =5(0.6+h)h=3h+5h . Now =  0.2=(3+10h)  = . When h=0.3
dt dh dt dt dt 3+10h
dh 0.2 0.2 1 10
, = = m / min = m / min or cm / min.
dt 3+10(0.3) 6 30 3

22.
1
The figure is drawn without the top 3 feet. V =
(b+12)h(20)=10(b+12)h and, from similar triangles,
2
x 6 y 16 8 8h 11h
= and = = , so b=x+12+y=h+12+ =12+ . Thus,
h 6 h 6 3 3 3
2
11h 110h dV 220 dh
V =10 24+ h=240h+ and so 0.8= = 240+ h . When h=5 ,
3 3 dt 3 dt
dh 0.8 3
= =  0.00132 ft / min.
dt 240+5(220/3) 2275

23.
3
dV 3 1 2 1 h 2  h
We are given that =30 ft / min. V =  r h=  h= 
dt 3 3 2 12

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

2
dV dV dh  h dh dh 120 dh 120 6
=  30=  = . When h=10 ft, = 2 =  0.38 ft / min.
dt dh dt 4 dt dt 2 dt 5
 h 10 

24.
2
x dx 2 d d sin 
We are given dx/dt=8 ft / s. cot  =  x=100cot   = 100csc   =  8.
100 dt dt dt 100
2
100 1 d (1/2) 1
When y=200 , sin  = =  =  8= rad / s. The angle is decreasing at a rate of
200 2 dt 100 50
1
rad / s.
50

25.
1 h 1
A= bh , but b=5 m and sin  =  h=4sin  , so A= (5)(4sin  )=10sin  . We are given
2 4 2
d dA dA d 
=0.06 rad / s, so = =(10cos  )(0.06)=0.6cos  . When  = ,
dt dt d dt 3
dA  1 2
=0.6 cos =(0.6) =0.3 m / s.
dt 3 2

26.

We are given d /dt=2 / min = rad / min. By the Law of Cosines,
90
2 2 2 dx d dx 180sin  d
x =12 +15  2(12)(15)cos  =369 360cos   2x =360sin   = . When
dt dt dt x dt


dx 180sin 60   3 7
 =60 , x= 369 360cos 60 = 189 =3 21 , so = = =  0.396 m /
dt 3 21 90 3 21 21
min.

27. Differentiating both sides of PV =C with respect to t and using the Product Rule gives us
dV dP dV V dP dP
P +V =0 = . When V =600 , P=150 and =20 , so we have
dt dt dt P dt dt
7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

dV 600 3
= (20)= 80 . Thus, the volume is decreasing at a rate of 80 cm / min.
dt 150

1.4
1.4 0.4 dV 1.4 dP dV V dP V dP
28. PV =C  P 1.4V +V =0 = = . When V =400 ,
dt dt dt 0.4 dt 1.4P dt
P 1.4V
dP dV 400 250
P=80 and = 10 , so we have = ( 10)= . Thus, the volume is increasing at a
dt dt 1.4(80) 7
250 3
rate of  36 cm / min.
7

1 1 1 1 1 180 9 400
29. With R =80 and R =100 , = + = + = = , so R= . Differentiating
1 2 R R R 80 100 8000 400 9
1 2
dR dR
1 1 1 1 dR 1 1 1 2
= + with respect to t , we have  =  
R R R 2 dt 2 dt 2 dt
1 2 R R R
1 2
dR dR
dR 2 1 1 1 2
=R + . When R =80 and R =100 ,
dt 2 dt 2 dt 1 2
R R
1 2
2
dR 400 1 1 107
= (0.3)+ (0.2) =  0.132
/ s.
dt 2 2 2 810
9 80 100

dB 2/3 2.53
30. We want to find when L=18 using B=0.007W and W =0.12L .
dt
dB dB dW dL 2  1/3 1.53 20 15
= = 0.007 W (0.12 2.53 L )
dt dW dL dt 3 10,000,000
2 2.53  1/3 1.53 5  8
= 0.007 3 (0.12 18 ) (0.12 2.53 18 )  1.045 10 g/yr
7
10

31.
dx x dx d 
We are given that =2 ft / s. sin  =  x=10sin   =10cos  . When  = ,
dt 10 dt dt 4
 d d 2 2
2=10cos  = = rad / s.
4 dt dt 10(1/ 2 ) 5
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

32.
dx dy
Using Q for the origin, we are given = 2 ft / s and need to find when x= 5 . Using the
dt dt
2 2 2 2
Pythagorean Theorem twice, we have x +12 + y +12 =39 , the total length of the rope.
x dx y dy
Differentiating with respect to t , we get + =0 , so
2 2 dt 2 2 dt
x +12 y +12
2 2
dy x y +12 dx 2 2 2 2 2 2 2 2
= . Now when x= 5 , 39= ( 5) +12 + y +12 =13+ y +12  y +12 =26
dt 2 2 dt
y x +12
2 2 dy ( 5)(26) 10
, and y= 26  12 = 532 . So when x= 5 , = ( 2)=   0.87 ft / s. So cart B is
dt 532 (13) 133
moving towards Q at about 0.87 ft / s.

33. (a)
2 2 2
By the Pythagorean Theorem, 4000 +y = . Differentiating with respect to t , we obtain
dy d dy
2y =2 . We know that =600 ft / s, so when y=3000 ft,
dt dt dt
2 2 d y dy 3000 1800
= 4000 +3000 = 25,000,000 =5000 ft and = = (600)= =360 ft / s.
dt dt 5000 5

2
y d d y 2 d 1 dy d cos  dy
(b) Here tan  =  (tan  )=  sec  =  = .
4000 dt dt 4000 dt 4000 dt dt 4000 dt
2
dy 4000 4000 4 d (4/5)
When y=3000 ft, =600 ft / s, =5000 and cos  = = = , so = (600)=0.096
dt 5000 5 dt 4000
rad / s.

34.

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

d dx 2 d 1
We are given that =4(2 )=8 rad / min. x=3tan   =3sec  . When x=1 , tan  = , so
dt dt dt 3
2 1 2 10 dx 10 80
sec  =1+ = and =3 (8 )=  83.8 km / min.
3 9 dt 9 3

35.
dx
We are given that =300 km / h. By the Law of Cosines,
dt
2 2 2 2 1 2 dy dx dx dy 2x+1 dx
y =x +1  2(1)(x)cos 120 =x +1 2x  =x +x+1 , so 2y =2x +  = . After
2 dt dt dt dt 2y dt
1 minute,
300 2 dy 2(5)+1 1650
x= =5 km  y= 5 +5+1 = 31 km  = (300)=  296 km / h.
60 dt 2 31 31

36.
dx dy
We are given that =3 mi / h and =2 mi / h. By the Law of Cosines,
dt dt
2 2 2 2 2 dz dx dy dy dx
z =x +y  2xycos 45 =x +y  2 xy 2z =2x +2y  2x  2y . After 15 minutes
dt dt dt dt dt
1
= h ,
4
3 2 1 2 3 2 2 2 3 2 13 6 2
we have x= and y= =  z = +  2  z= and
4 4 2 4 4 4 4 4
dz 2 3 1 3 1 2 13 6 2
= 2 3+2 2 2 2 2 3 = = 13 6 2
dt 13 6 2 4 2 4 2 13 6 2 2
 2.125mi/h.

37.
Let the distance between the runner and the friend be . Then by the Law of Cosines,
10
 Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.10 Related Rates

2 2 2
=200 +100  2 200 100 cos  =50,000 40,000cos  (*). Differentiating implicitly with respect to t
d d
, we obtain 2 = 40,000( sin  ) . Now if D is the distance run when
dt dt
the angle is  radians, then by the formula for the length of an arc on a circle, s=r , we have
1 d 1 dD 7 d
D=100 , so  = D = = . To substitute into the expression for , we must
100 dt 100 dt 100 dt
2 1
know sin  at the time when =200 , which we find from (*): 200 =50,000 40,000cos   cos  =
4
1 2 15 d 15 7
 sin  = 1 = . Substituting, we get 2(200) =40,000 
4 4 dt 4 100
7 15
d /dt=  6.78 m / s. Whether the distance between them is increasing or decreasing depends on
4
the direction in which the runner is running.

38.
2 
The hour hand of a clock goes around once every 12 hours or, in radians per hour, = rad / h.
12 6
The minute hand goes around once an hour, or at the rate of 2 rad / h. So the angle  between them
(measuring clockwise from the minute hand to the hour hand) is changing at the rate of
 11
d /dt=  2 = rad / h. Now, to relate  to , we use the Law of Cosines:
6 6
2 2 2
=4 +8  2 4 8 cos  =80 64cos  (*).
d d
Differentiating implicitly with respect to t , we get 2 = 64( sin  ) . At 1:00, the angle
dt dt
2 
between the two hands is one twelfth of the circle, that is, = radians. We use (*) to find at
12 6
 d  11
1:00: = 80 64cos = 80 32 3 . Substituting, we get 2 =64sin  
6 dt 6 6
1 11
64 
d 2 6 88
= =   18.6 . So at 1:00, the distance between the tips of the
dt 2 80 32 3 3 80 32 3
hands is decreasing at a rate of 18.6 mm / h  0.005 mm / s.

11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

1. As in Example 1, T (0)=185 , T (10)=172 , T (20)=160 , and

/ T (10) T (20) 172 160  / 
T (20) = = 1.2 F / min. T (30) T (20)+T (20)(30 20) 160 1.2(10)=148
10 20  10
F.

We would expect the temperature of the turkey to get closer to 75 F
 
as time increases. Since the temperature decreased 13 F in the first 10 minutes and 12 F in the
second 10 minutes, we can assume that the slopes of the tangent line are increasing through negative

values:  1.3, 1.2,... . Hence, the tangent lines are under the curve and 148 F

is an underestimate. From the figure, we estimate the slope of the tangent line at t=20 to be
184 147 37
= .
0 30 30
/ 37 2
Then the linear approximation becomes T (30) T (20)+T (20) 10 160 (10)=147  147.7 .
30 3

/ P(1) P(2) 87.1 74.9

2. P (2) = = 12.2 kilopascals / km.
1 2  1
/
P(3) P(2)+P (2)(3 2) 74.9 12.2(1)=62.7 kPa.
98 63 35
From the figure, we estimate the slope of the tangent line at h=2 to be = . Then the linear
0 3 3
/ 35
approximation becomes P(3) P(2)+P (2) 1 74.9  63.23 kPa.
3

3. Extend the tangent line at the point (2030,21) to the t  axis. Answers will vary based on this
approximation   we’ll use t=1900 as our t  intercept. The linearization is then
/
P(t)  P(2030)+P (2030)(t 2030)
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

21
 21+(t 2030)
130
21
P(2040)=21+ (2040 2030) 22.6%
130
21
P(2050)=21+ (2050 2030) 24.2%
130

These predictions are probably too high since the tangent line lies above the graph at t=2030.

N(1980) N(1985) 15.0 17.0 N(1990) N(1985) 19.3 17.0

4. Let A= = =0.4 and B= = =0.46 . Then
1980 1985  5 1990 1985 5
/ N(t) N(1985) A+B
N (1985)= lim  =0.43 million / year. So
t 1985 t 1985 2
/
N(1984) N(1985)+N (1985)(1984 1985) 17.0+0.43( 1)=16.57 million.
/ N(1995) N(2000) 22.0 24.9
N (2000) = =0.58 million / year.
1995 2000  5
/
N(2006) N(2000)+N (2000)(2006 2000) 24.9+0.58(6)=28.38 million.

3 / 2 / /
5. f (x)=x  f (x)=3x , so f (1)=1 and f (1)=3 . With a=1 , L(x)= f (a)+ f (a)(x a) becomes
/
L(x)= f (1)+ f (1)(x 1)=1+3(x 1)=3x 2 .

/ / /
6. f (x)=ln x  f (x)=1/x , so f (1)=0 and f (1)=1 . Thus, L(x)= f (1)+ f (1)(x 1)=0+1(x 1)=x 1.

/  / 
7. f (x)=cos x  f (x)= sin x , so f =0 and f = 1 . Thus,
2 2
 /    
L(x)= f +f x =0 1 x = x+ .
2 2 2 2 2
3 1/3 / 1  2/3 / 1
8. f (x)= x =x  f (x)= x , so f ( 8)= 2 and f ( 8)= . Thus,
3 12
/ 1 1 4
L(x)= f ( 8)+ f ( 8)(x+8)= 2+ (x+8)= x .
12 12 3

9. f (x)= 1 x 

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

/  1 / 1
f (x)= , so f (0)=1 and f (0)= . Therefore,
2 1 x 2
1 x = f (x) f (0)+ f /(0)(x 0)
1 1
= 1+  (x 0)=1 x
2 2
1 1
So 0.9 = 1 0.1  1 (0.1)=0.95 and 0.99 = 1 0.01  1 (0.01)=0.995 .
2 2

3 1/3 / 1  2/3
10. g(x)= 1+x =(1+x)  g (x)= (1+x) ,
3
1/ 3 / 1
so g(0)=1 and g (0)= . Therefore, 1+x =g(x) g(0)+g (0)(x 0)=1+ x . So
3 3
3 3 1 3 3 1
0.95= 1+ (  0.05)  1+ (  0.05) =0.983 , and 1.1 = 1+0.1  1+ (0.1)=1.03 .
3 3

3 1/3 / 1  2/3 / 1
11. f (x)= 1 x =(1 x)  f (x)= (1 x) , so f (0)=1 and f (0)= . Thus,
3 3
/ 1 3 1 3
f (x) f (0)+ f (0)(x 0)=1 x . We need 1 x  0.1<1 x< 1 x +0.1 , which is true when
3 3
 1.204<x<0.706 .

/ 2 / /
12. f (x)=tan x f (x)=sec x , so f (0)=0 and f (0)=1 . Thus, f (x) f (0)+ f (0)(x 0)=0+1(x 0)=x .
We need tan x 0.1<x<tan x+0.1 , which is true when  0.63<x<0.63 .

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

1  4 /  5  8 /
13. f (x)= =(1+2x)  f (x)= 4(1+2x) (2)= , so f (0)=1 and f (0)= 8 . Thus,
4 5
(1+2x) (1+2x)
/
f (x) f (0)+ f (0)(x 0)=1+( 8)(x 0)=1 8x .
4 4
We need 1/(1+2x)  0.1<1 8x<1/(1+2x) +0.1 , which is true when  0.045<x<0.055 .

x / x / /
14. f (x)=e  f (x)=e , so f (0)=1 and f (0)=1 . Thus, f (x) f (0)+ f (0)(x 0)=1+1(x 0)=1+x .
x x
We need e  0.1<1+x<e +0.1 , which is true when  0.483<x<0.416.

/ 4 3
15. If y= f (x) , then the differential dy is equal to f (x)dx . y=x +5x dy=(4x +5)dx .

16. y=cos  x dy= sin  x  dx=  sin  xdx

1
17. y=xln x dy= x +ln x 1 dx=(1+ln x)dx
x

2 1 2 1/2 t
18. y= 1+t  dy= (1+t ) (2t)dt= dt
2 2
1+t

u+1 (u 1)(1) (u+1)(1)  2

19. y=  dy= du= du
u 1 2 2
(u 1) (u 1)

 4  5  5
20. y=(1+2r)  dy= 4(1+2r)  2dr= 8(1+2r) dr

2
21. (a) y=x +2x dy=(2x+2)dx
(b) When x=3 and
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

1 1
dx= , dy=[2(3)+2] =4 .
2 2

x/4 1 x/4
22. (a) y=e  dy= e dx
4
1 0
(b) When x=0 and dx=0.1 , dy= e (0.1)=0.025 .
4

1 5  1/2
23. (a) y= 4+5x  dy= (4+5x)
dx  5dx=
2
2 4+5x
5 5 1 1
(b) When x=0 and dx=0.04 , dy= (0.04)=  = =0.05 .
2 4 4 25 20

1
24. (a) y=1/(x+1)  dy= dx
2
(x+1)
1 1 1 1
(b) When x=1 and dx= 0.01 , dy= ( 0.01)=  = =0.0025 .
2 4 100 400
2

2
25. (a) y=tan x  dy=sec xdx
2 2
(b) When x= /4 and dx= 0.1 , dy=[sec ( /4)] ( 0.1)=( 2 ) ( 0.1)= 0.2 .

26. (a) y=cos x  dy= sin xdx

(b) When x= /3 and dx=0.05 , dy= sin ( /3)(0.05)= 0.5 3 (0.05)= 0.025 3  0.043 .

2 2 2
27. y=x , x=1 ,  x=0.5  y=(1.5)  1 =1.25 . dy=2xdx=2(1)(0.5)=1

28. y= x , x=1 ,  x=1  y= 2  1 = 2  1 0.414

1 1
dy= dx= (1)=0.5
2 x 2

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

2 2 2
29. y=6 x , x= 2 ,  x=0.4  y=(6 ( 1.6) ) (6 ( 2) )=1.44
dy= 2xdx= 2( 2)(0.4)=1.6

16 16 16 4 16 16
30. y= , x=4 ,  x= 1  y=  = . dy= dx= ( 1)=1
x 3 4 3 2 2
x 4

5 4 4
31. y= f (x)=x  dy=5x dx . When x=2 and dx=0.001 , dy=5(2) (0.001)=0.08 , so
5
(2.001) = f (2.001) f (2)+dy=32+0.08=32.08 .

1 1
32. y= f (x)= x  dy= dx . When x=100 and dx= 0.2 , dy= ( 0.2)= 0.01 , so
2 x 2 100
99.8 = f (99.8) f (100)+dy=10 0.01=9.99 .
2/3 2 2
33. y= f (x)=x  dy= 3 dx . When x=8 and dx=0.06 , dy= 3 (0.06)=0.02 , so
3 x 3 8
2/3
(8.06) = f (8.06) f (8)+dy=4+0.02=4.02 .

2 2
34. y= f (x)=1/x dy=( 1/x )dx . When x=1000 and dx=2,dy=[ 1/(1000) ](2)= 0.000002 , so
1/1002= f (1002) f (1000)+dy=1/1000 0.000002=0.000998

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

2  
35. y= f (x)=tan x dy=sec xdx . When x=45 and dx= 1 ,
2  2   
dy=sec 45 (  /180)=( 2 ) (  /180)=  /90 , so tan 44 = f (44 ) f (45 )+dy=1  /90 0.965 .

1 1
36. y= f (x)=ln x dy= dx . When x=1 and dx=0.07 , dy= (0.07)=0.07 , so
x 1
ln 1.07= f (1.07) f (1)+dy=0+0.07=0.07.

/ /
37. y= f (x)=sec x f (x)=sec xtan x , so f (0)=1 and f (0)=1 0=0 . The linear approximation of f at
/
0 is f (0)+ f (0)(x 0)=1+0(x)=1 . Since 0.08 is close to 0 , approximating sec 0.08 with 1 is
reasonable.

6 / 5
38. If y=x , y =6x and the tangent line approximation at (1,1) has slope 6 . If the change in x is
6
0.01 , the change in y on the tangent line is 0.06 , and approximating (1.01) with 1.06 is reasonable.

/ /
39. y= f (x)=ln x f (x)=1/x , so f (1)=0 and f (1)=1 . The linear approximation of f at 1 is
/
f (1)+ f (1)(x 1)=0+1(x 1)=x 1 . Now f (1.05)=ln 1.05 1.05 1=0.05 , so the approximation is
reasonable.

2 / /
40. (a) f (x)=(x 1)  f (x)=2(x 1) , so f (0)=1 and f (0)= 2 .
/
Thus, f (x) L (x)= f (0)+ f (0)(x 0)=1 2x .
f
 2x /  2x /
g(x)=e  g (x)= 2e , so g(0)=1 and g (0)= 2 .
/
Thus, g(x) L (x)=g(0)+g (0)(x 0)=1 2x .
g
/  2 /
h(x)=1+ln (1 2x) h (x)= , so h(0)=1 and h (0)= 2 .
1 2x
/
Thus, h(x) L (x)=h(0)+h (0)(x 0)=1 2x .
h
Notice that L =L =L . This happens because f , g , and h have the same function values and the
f g h
same derivative values at a=0 .

(b)
The linear approximation appears to be the best for the function f since it is closer to f for a larger
domain than it is to g and h . The approximation looks worst for h since h moves away from L faster
7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

than f and g do.

3 2 2
41. (a) If x is the edge length, then V =x  dV =3x dx . When x=30 and dx=0.1 , dV =3(30) (0.1)=270
3
, so the maximum possible error in computing the volume of the cube is about 270 cm . The relative
error is calculated by dividing the change in V ,  V , by V . We approximate  V with dV .
2
 V dV 3x dx dx 0.1
Relative error =  = =3 =3 =0.01 .
V V 3 x 30
x
Percentage error = relative error 100%=0.01 100%=1% .
2
(b) S=6x  dS=12xdx . When x=30 and dx=0.1 , dS=12(30)(0.1)=36 , so the maximum possible
2
error in computing the surface area of the cube is about 36 cm .
 S dS 12xdx dx 0.1
Relative error =  = =2 =2 =0.006 .
S S 2 x 30
6x
Percentage error = relative error 100%=0.006 100%=0.6% .

2
42. (a) A= r  dA=2 r dr . When r=24 and dr=0.2 , dA=2 (24)(0.2)=9.6 , so the maximum
2
possible error in the calculated area of the disk is about 9.6  30 cm .
 A dA 2 rdr 2dr 2(0.2) 0.2 1
(b) Relative error =  = = = = = =0.016 .
A A 2 r 24 12 60
 r
Percentage error = relative error 100%=0.016 100%=1.6% .

2
43. (a) For a sphere of radius r , the circumference is C=2 r and the surface area is S=4 r , so
2 2 2 84
r=C/(2 ) S=4 (C/2 ) =C /  dS=(2/ )C dC . When C=84 and dC=0.5 , dS= (84)(0.5)= ,
 
84 2 dS 84/ 1
so the maximum error is about  27 cm . Relative error  = 2 =  0.012
 S 84
84 /
3 3
4 3 4 C C 1 2
(b) V =  r =  =  dV = C dC . When C=84 and dC=0.5 ,
3 3 2 2 2
6 2
1 2 1764 1764 3
dV = (84) (0.5)= , so the maximum error is about  179 cm . The relative error is
2 2 2
2  
2
dV 1764/ 1
approximately = =  0.018 .
V 3 2 56
(84) /(6 )

44. For a hemispherical dome,

8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

2 3 21
V=  r  dV =2 r dr . When r=
(50)=25 m and
3 2
2 5 5
dr=0.05 cm =0.0005 m, dV =2 (25) (0.0005)= , so the amount of paint needed is about  2
8 8
3
m .

2
45. (a) V = r h  V  dV =2 rhdr=2 rh r
(b) The error is
 V  dV =[ (r+ r)2h  r 2h] 2 rh r= r 2h+2 rh r+ ( r)2h  r 2h 2 rh r
2
= ( r) h

3
4 dF 4kR dR
3 dR
46. F=kR  dF=4kR dR = =4 . Thus, the relative change in F is about 4
F 4 R
kR
times the relative change in R . So a 5% increase in the radius corresponds to a 20% increase in blood
flow.

dc
47. (a) dc= dx=0dx=0
dx
d du
(b) d(cu)= (cu)dx=c dx=cdu
dx dx
d du dv du dv
(c) d(u+v)= (u+v)dx= + dx= dx+ dx=du+dv
dx dx dx dx dx
d dv du dv du
(d) d(uv)= (uv)dx= u +v dx=u dx+v dx=udv+vdu
dx dx dx dx dx
du dv du dv
v  u v dx u dx
u d u dx dx dx dx vdu udv
(e) d = dx= dx= =
v dx v 2 2 2
v v v
n d n n 1
(f) d(x )= (x )dx=nx dx
dx

/ /
48. (a) f (x)=sin x f (x)=cos x , so f (0)=0 and f (0)=1 . Thus,
/
f (x) f (0)+ f (0)(x 0)=0+1(x 0)=x .
(b)

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

We want to know the values of x for which y=x approximates y=sin x with less than a 2% difference;
that is, the values of x for which
x sin x x sin x
<0.02
 0.02< <0.02

sin x sin x

 0.02sin x<x sin x<0.02sin x

{  0.02sin x>x sin x>0.02sin x
if sin x>0
if sin x<0

{ 0.98sin x<x<1.02sin x
1.02sin x<x<0.98sin x
if sin x>0
if sin x<0

In the first figure, we see that the graphs are very close to each other near x=0 . Changing the viewing
rectangle and using an intersect feature (see the second figure) we find that y=x intersects y=1.02sin x
at x 0.344 . By symmetry, they also intersect at x  0.344 (see the third figure.). Converting 0.344

180  
radians to degrees, we get 0.344  19.7  20 , which verifies the statement.


/ /
49. (a) The graph shows that f (1)=2 , so L(x)= f (1)+ f (1)(x 1)=5+2(x 1)=2x+3 .
f (0.9) L(0.9)=4.8 and f (1.1) L(1.1)=5.2 .
/
(b) From the graph, we see that f (x) is positive and decreasing. This means that the slopes of the
tangent lines are positive, but the tangents are becoming less steep. So the tangent lines lie above the
curve. Thus, the estimates in part (a) are too large.
10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.11 Linear Approximations and Differentials

/ 2 / /
50. (a) g (x)= x +5  g (2)= 9 =3 . g(1.95) g(2)+g (2)(1.95 2)= 4+3( 0.05)= 4.15 .
/
g(2.05) g(2)+g (2)(2.05 2)= 4+3(0.05)= 3.85 .
/ 2 /
(b) The formula g (x)= x +5 shows that g (x) is positive and increasing. This means that the slopes
of the tangent lines are positive and the tangents are getting steeper. So the tangent lines lie below the
graph of g . Hence, the estimates in part (a) are too small.

11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

1. Product Rule:
2 3
y=(x +1)(x +1)

/ 2 2 3 4 2 4 4 2
y =(x +1)(3x )+(x +1)(2x)=3x +3x +2x +2x=5x +3x +2x .
2 3 5 3 2 / 4 2
Multiplying first: y=(x +1)(x +1)=x +x +x +1 y =5x +3x +2x (equivalent).

3/2
x 3x x x 3x
2. Quotient Rule: F(x)= = 
x 1/2
x
9 1/2 1 
/
x
1/2
1
2
x  ( x 3x3/2) 2
x
1/2

F (x) =
( x1/2) 2
9 1 1/2 3
1/2 1 1/2
x 
x x + x x  3x
= 2 2 2 2 1  1/2
= = x  3
x x 2
x 3x x 1/2 / 1  1/2
Simplifying first: F(x)= = x  3x=x  3x F (x)= x  3 (equivalent).
x 2
For this problem, simplifying first seems to be the better method.

2 x / 2 d x x d 2 2 x x x
3. By the Product Rule, f (x)=x e  f (x)=x (e )+e (x )=x e +e (2x)=xe (x+2) .
dx dx

x 1/2 x / 1/2 x x 1  1/2 1  1/2 x

4. By the Product Rule, g(x)= x e =x e  g (x)=x (e )+e x = x e (2x+1) .
2 2

2 d x x d 2
x x (e ) e (x ) 2 x x x x
e / dx dx x (e ) e (2x) xe (x 2) e (x 2)
5. By the Quotient Rule, y=  y = = = =
2 22 4 4 3
x (x ) x x x

x x x x x x x
e / (1+x)e  e (1) e +xe  e xe
6. By the Quotient Rule, y=  y = = = .
1+x 2 2 2
(1+x) (x+1) (x+1)
QR
3x 1 / (2x+1)(3) (3x 1)(2) 6x+3 6x+2 5
7. g(x)=  g (x)= = =
2x+1 2 2 2
(2x+1) (2x+1) (2x+1)
QR 2 2 2 2
2t / (4+t )(2) (2t)(2t) 8+2t  4t 8 2t
8. f (t)=  f (t)= = =
2 22 22 22
4+t (4+t ) (4+t ) (4+t )
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

PR
3 4
9. V (x)=(2x +3)(x  2x) 
/ 3 3 4 2 6 3 6 3 6 3
V (x)=(2x +3)(4x  2)+(x  2x)(6x )=(8x +8x  6)+(6x  12x )=14x  4x  6

PR
 2  3 5 2
10. Y (u)=(u +u )(u  2u )
/  2  3 4 5 2  3  4
Y (u) =(u +u )(5u  4u)+(u  2u )( 2u  3u )
2  1  2 2  1  2 2  2
=(5u  4u +5u 4u )+( 2u  3u+4u +6u )=3u +2u+2u

PR
 2  4
11. F(y)=
1
2

3
4
(y+5y )= y  3y
3
( ) ( y+5y3) 
y y
/
F (y) = y  3y (  2  4) ( 1+15y2) + ( y+5y3) (  2y 3+12y 5)
 2  4  2  2  4  2
= ( y +15 3y  45y ) +( 2y +12y  10+60y )
 2  4 2 4
=5+14y +9y or 5+14/y +9/y

t
12. R(t)=(t+e )(3 t )=
/ t 1  1/2 t
R (t) =(t+e )( 2 t )+(3 t )(1+e )
1 1/2 1  1/2 t t t t 3 t t
=  t  t e +(3+3e  t  t e )=3+3e  t t e  e /(2 t )
2 2 2
2 QR
t
13. y= 2

3t  2t+1
2 2 2
/ (3t  2t+1)(2t) t (6t 2) 2t[3t  2t+1 t(3t 1)]
y = 2 2
=
2 2
(3t  2t+1) (3t  2t+1)
2 2
2t (3t  2t+1 3t +t) 2t(1 t)
= =
2 2 2 2
(3t  2t+1) (3t  2t+1)

3 QR 4 2 3 3 6 4 2 6 4
t +t / (t  2)(3t +1) (t +t)(4t ) (3t +t  6t  2) (4t +4t )
14. y= 4
 y = =
4 2 4 2
t  2 (t  2) (t  2)
6 4 2 6 4 2
 t  3t  6t  2 t +3t +6t +2
= =
4 2 4 2
(t  2) (t  2)
2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

2 r / 2 r r r 2 r 2
15. y=(r  2r)e = y =(r  2r)(e )+e (2r 2)=e (r  2r+2r 2)=e (r  2)

s s s
1 / (s+ke )(0) (1)(1+ke ) 1+ke
16. y= =y = =
s s2 s2
s+ke (s+ke ) (s+ke )

3
v  2v v 2 2 1/2 / 1  1/2  1/2
17. y= =v  2 v =v  2v  y =2v 2 v =2v v .
v 2
3/2
 1/2 1 2v v  1 2v  1
We can change the form of the answer as follows: 2v v =2v = =
v v v

3/2 w 5/2 3/2 w / 5 3/2 3/2 w w 3 1/2 5 3/2 1 1/2 w

18. z=w (w+ce )=w +cw e  z = w +c w  e +e  w = w + cw e (2w+3)
2 2 2 2

4 2 3 2
1 / (x +x +1)(0) 1(4x +2x) 2x(2x +1)
19. y= 4 2
 y = =
4 2 2 4 2 2
x +x +1 (x +x +1) (x +x +1)

1 1 1 1 1 1
( x +1)  ( x  1) +  +
x 1 / 2 x 2 x 2 2 x 2 2 x 1
20. y=  y = = =
x +1 2 2 2
( x +1) ( x +1) x ( x +1)

2 2
x (x+c/x)(1) x(1 c/x ) x+c/x x+c/x
/ 2c/x x 2cx
21. f (x)=  f (x)= = =  =
x+c/x c 2 2 2 2 2 2 2 2
x+ x +c (x +c) x (x +c)
x 2
x x

ax+b / (cx+d)(a) (ax+b)(c) acx+ad acx bc ad bc

22. f (x)=  f (x)= = =
cx+d 2 2 2
(cx+d) (cx+d) (cx+d)

2x / (x+1)(2) (2x)(1) 2 / 1
23. y=  y = = . At (1,1) , y = , and an equation of the tangent line
x+1 2 2 2
(x+1) (x+1)
1 1 1
is y 1= (x 1) , or y= x+ .
2 2 2

x
24. y= 
x+1

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

1
(x+1)  x (1)
/ 2 x (x+1) (2x) 1 x /  3
y = = = . At (4,0.4) , y = = 0.03 , and an
2 2 2 100
(x+1) 2 x (x+1) 2 x (x+1)
equation of the tangent line is y 0.4= 0.03(x 4) , or y= 0.03x+0.52 .

x / x x x / 0
25. y=2xe  y =2(x e +e  1)=2e (x+1) . At (0, 0) , y =2e (0+1)=2 1 1=2 , and an equation of the
tangent line is y 0=2(x 0) , or y=2x .

x x x x
e / x e  e  1 e (x 1) /
26. y=  y = = . At (1, e) , y =0 , and an equation of the tangent line is
x 2 2
x x
y e=0(x 1) , or y=e .

2
1 / (1+x )(0) 1(2x)  2x
27. (a) y= f (x)=  f (x)= = . So the slope of the tangent line at the
2 22 22
1+x (1+x ) (1+x )
1 / 2 1 1 1 1
point  1, is f ( 1)= = and its equation is y = (x+1) or y= x+1 .
2 2 2 2 2 2
2

(b)

2 2
x / (1+x )1 x(2x) 1 x
28. (a) y= f (x)=  f (x)= = . So the slope of the tangent line at the
2 22 22
1+x (1+x ) (1+x )
/  8
point (3,0.3) is f (3)= and its equation is y 0.3= 0.08(x 3) or y= 0.08x+0.54 .
100

(b)

x 3 x x 2 2 x x
e / x (e ) e (3x ) x e (x 3) e (x 3)
29. (a) f (x)=  f (x)= = =
3 32 6 4
x (x ) x x
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

(b)
/ / /
f =0 when f has a horizontal tangent line, f is negative when f is decreasing, and f is positive
when f is increasing.

2 2 2
x / (x  1)1 x(2x)  x 1 x +1
30. f (x)= 2
 f (x)= = = Notice that the slopes of all tangents to
2 2 2 2 2 2
x 1 (x  1) (x  1) (x  1)
/
f are negative and f (x)<0 always.

/ /
31. We are given that f (5)=1 , f (5)=6 , g(5)= 3 , and g (5)=2 .
/ / /
(a) ( fg) (5)= f (5)g (5)+g(5) f (5)=(1)(2)+( 3)(6)=2 18= 16
/ / /
f g(5) f (5) f (5)g (5) ( 3)(6) (1)(2) 20
(b) (5)= = =
g 2 2 9
[g(5)] ( 3)
/ / /
g f (5)g (5) g(5) f (5) (1)(2) ( 3)(6)
(c) (5)= = =20
f 2 2
[ f (5)] (1)

/ /
32. We are given that f (3)=4 , g(3)=2 , f (3)= 6 , and g (3)=5 .
/ / /
(a) ( f +g ) (3)= f (3)+g (3)= 6+5= 1
/ / /
(b) ( fg ) (3)= f (3)g (3)+g(3) f (3)=(4)(5)+(2)( 6)=20 12=8
/ /
f / g(3) f (3) f (3)g (3) (2)( 6) (4)(5)  32
(c) (3)= = = = 8
g 2 2 4
[g(3)] (2)
(d)

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

/ / /
f / [ f (3) g(3)] f (3) f (3)[ f (3) g (3)]
(3) =
f g [ f (3) g(3)]
2

(4 2)( 6) 4( 6 5)  12+44

= = =8
2 2
(4 2) 2

x / x / x x /
33. f (x)=e g(x) f (x)=e g (x)+g(x)e =e g (x)+g(x) .
/ 0 /
f (0)=e g (0)+g(0) =1(5+2)=7

/ /
d h(x) xh (x) h(x) 1 d h(x) 2h (2) h(2) 2( 3) (4)  10
34. =  = = = = 2.5
dx x 2 dx x x=2 2 4 4
x 2

35. (a) From the graphs of f and g , we obtain the following values: f (1)=2 since the point (1,2) is on
/
the graph of f ; g(1)=1 since the point (1,1) is on the graph of g ; f (1)=2 since the slope of the line
4 0 /
segment between (0,0) and (2,4) is =2 ; g (1)= 1 since the slope of the line segment between
2 0
0 4 / / /
( 2,4) and (2,0) is = 1 . Now u(x)= f (x)g(x) , so u (1)= f (1)g (1)+g(1) f (1)=2 ( 1)+1 2=0 .
2 ( 2)
1 2 8
/ / 2   3 
/ g(5) f (5) f (5)g (5) 3 3 3 2
(b) v(x)= f (x)/g(x) , so v (5)= = = =
2 2 4 3
[g(5)] 2

/ / / 2 3
36. (a) P(x)=F(x)G(x) , so P (2)=F(2)G (2)+G(2)F (2)=3 +2 0= .
4 2
1 2
/ / 1  5 
/ G(7)F (7) F(7)G (7) 4 3 1 10 43
(b) Q(x)=F(x)/G(x) , so Q (7)= = = + =
2 2 4 3 12
[G(7)] 1

/ / /
37. (a) y=xg(x)  y =xg (x)+g(x) 1=xg (x)+g(x)
/ /
x / g(x) 1 xg (x) g(x) xg (x)
(b) y=  y = =
g(x) 2 2
[g(x)] [g(x)]
/ /
g(x) / xg (x) g(x) 1 xg (x) g(x)
(c) y=  y = =
x 2 2
(x) x

38. (a)
6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

2 / 2 /
y=x f (x) y =x f (x)+ f (x)(2x)
2 / /
f (x) / x f (x) f (x)(2x) xf (x) 2 f (x)
(b) y=  y = =
2 22 3
x (x ) x
2 2 /
x / f (x)(2x) x f (x)
(c) y=  y =
f (x) 2
[ f (x)]
1+xf (x)
(d) y= 
x
/ 1
x[xf (x)+ f (x)] [1+xf (x)]
/ 2 x
y =
2
( x)
3/2 / 1/2 1  1/2 1 1/2
x f (x)+x f (x) x  x f (x) 1/2 2 /
2 2 2x xf (x)+2x f (x) 1
=  =
x 1/2 3/2
2x 2x

39. If P(t) denotes the population at time t and A(t) the average annual income, then T (t)=P(t)A(t) is
/ / /
the total personal income. The rate at which T (t) is rising is given by T (t)=P(t)A (t)+A(t)P (t)
/ / /
T (1999) =P(1999)A (1999)+A(1999)P (1999)=( 961,400 )( $1400 / yr )+( $30,593 )( 9200 / yr )
=$ 1,345,960,000 / yr +$ 281,455,600 / yr =$ 1,627,415,600 / yr

So the total personal income was rising by about $ 1.627 billion per year in 1999.
/
The term P(t)A (t) $ 1.346 billion represents the portion of the rate of change of total income due to
/
the existing population’s increasing income. The term A(t)P (t) $ 281 million represents the portion
of the rate of change of total income due to increasing population.

40. (a) f (20)=10 , 000 means that when the price of the fabric is $20/ yard, 10 , 000 yards will be
sold.
/
f (20)= 350 means that as the price of the fabric increases past $20/ yard, the amount of fabric
which will be sold is decreasing at a rate of 350 yards per (dollar per yard).
/ / / /
(b) R( p)= pf ( p)  R ( p)= pf ( p)+ f ( p) 1 R (20)=20 f (20)+ f (20) 1=20( 350)+10,000=3000.
This means that as the price of the fabric increases past $20/ yard, the total revenue is increasing at
$3000/($/yard). Note that the Product Rule indicates that we will lose $7000/($/yard) due to selling
less fabric, but that that loss is more than made up for by the additional revenue due to the increase in
price.

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

x / (x+1)(1) x(1) 1
41. If y= f (x)= , then f (x)= = . When x=a , the equation of the tangent
x+1 2 2
(x+1) (x+1)
a 1 a 1
line is y = (x a) . This line passes through (1,2) when 2 = (1 a)
a+1 2 a+1 2
(a+1) (a+1)
2 2 2 2
2(a+1)  a(a+1)=1 a  2a +4a+2 a  a 1+a=0 a +4a+1=0 .
2
 4 4  4(1)(1)  4 12
The quadratic formula gives the roots of this equation as a= = = 2 3 ,
2(1) 2
so there are two such tangent lines. Since
 2 3  2 3  1 3
f ( 2 3) = = 
 2 3 +1  1 3  1 3
2 2 3 3  3  1 3 1 3
= = = ,
1 3  2 2
1 3
the lines touch the curve at A  2+ 3,  ( 0.27, 0.37) and
2
1+ 3
B  2 3,  ( 3.73,1.37) .
2
x 1 / (x+1)(1) (x 1)(1) 2
42. y=  y = = . If the tangent intersects the curve when x=a ,
x+1 2 2
(x+1) (x+1)
2 1
then its slope is 2/(a+1) . But if the tangent is parallel to x 2y=2 , that is, y= x 1 , then its slope is
2
1 2 1 2
. Thus, =  (a+1) =4 a+1= 2 a=1 or  3 . When a=1 , y=0 and the equation of the
2 2 2
(a+1)
1 1 1
tangent is y 0= (x 1) or y= x .
2 2 2
1 1 7
When a= 3 , y=2 and the equation of the tangent is y 2= (x+3) or y= x+ .
2 2 2

/ / / / / / / / / /
43. (a) ( fgh) =[( fg)h] =( fg) h+( fg)h =( f g+ fg )h+( fg)h = f gh+ fg h+ fgh
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.2 The Product and Quotient Rules

(b) Putting f =g=h in part (a) , we have

d 3 / / / / / 2 /
[ f (x)] = ( fff ) = f ff + ff f + fff =3 fff =3[ f (x)] f (x).
dx

d 3x d x 3 x2 x 2x x 3x
(c) (e )= (e ) =3(e ) e =3e e =3e
dx dx

44. (a)
d d
d 1 g(x) (1) 1 [g(x)]
dx dx
dx g(x) = [Quotient Rule]
2
[g(x)]
/ / /
g(x) 0 1 g (x) 0 g (x) g (x)
= = =
2 2 2
[g(x)] [g(x)] [g(x)]

3 2
1 / 4x +2x  2x(2x +1)
(b) y= 4 2
 y = or
4 2 2 4 2 2
x +x +1 (x +x +1) (x +x +1)
n / n 1
d n d 1 (x ) nx n 1 2n  n 1
(c) (x )= = [by the Reciprocal Rule] = = nx = nx
dx dx n n2 2n
x (x ) x

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

2 /
1. (a) s= f (t)=t  10t+12 v(t)= f (t)=2t 10
(b) v(3)=2(3) 10= 4 ft / s
(c) The particle is at rest when v(t)=0 2t 10=0 t=5 s.
(d) The particle is moving in the positive direction when v(t)>0 2t 10>0 2t>10 t>5 .
(e) Since the particle is moving in the positive direction and in the negative direction, we need to
calculate the distance traveled in the intervals [0,5] and [5,8] separately. | f (5) f (0)|=| 13 12|=25 ft
and | f (8) f (5)|=|{ 4 ( 13)|=9 ft. The total distance traveled during the first 8 s is 25+9=34 ft.

(f)

3 2 / 2
2. (a) s= f (t)=t  9t +15t+10 v(t)= f (t)=3t  18t+15=3(t 1)(t 5)

(b) v(3)=3(2)( 2)= 12 ft / s

(c) v(t)=0 t=1 s or 5 s
(d) v(t)>0 0 t<1 or t>5
(e) | f (1) f (0)|=|17 10|=7 , | f (5) f (1)|=| 15 17|=32 , and | f (8) f (5)|=|66 ( 15)|=81 . Total distance
=7+32+81=120 ft.

(f)

3 2 / 2
3. (a) s= f (t)=t  12t +36t  v(t)= f (t)=3t  24t+36
(b) v(3)=27 72+36= 9 ft / s
2
(c) The particle is at rest when v(t)=0 . 3t  24t+36=0 3(t 2)(t 6)=0 t=2 s or 6 s.
(d) The particle is moving in the positive direction when v(t)>0 . 3(t 2)(t 6)>0 0 t<2 or t>6 .
(e) Since the particle is moving in the positive direction and in the negative direction, we need to
calculate the distance traveled in the intervals (0,2) , (2,6) , and [6,8] separately.
| f (2) f (0)|=|32 0|=32 .
| f (6) f (2)|=|0 32|=32 .
| f (8) f (6)|=|32 0|=32 .
The total distance is 32+32+32=96 ft.
(f)

1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

4 / 3
4. (a) s= f (t)=t  4t+1 v(t)= f (t)=4t  4
3
(b) v(3)=4(3)  4=104 ft / s
3 2
(c) It is at rest when v(t)=4(t  1)=4(t 1)(t +t+1)=0 t=1 s.
3
(d) It moves in the positive direction when 4(t  1)>0 t>1 .
(e) Distance in positive direction =| f (8) f (1)|=|4065 ( 2)|=4067 ft
Distance in negative direction =| f (1) f (0)|=| 2 1|=3 ft
Total distance traveled =4067+3=4070 ft

(f)

2 2
t / (t +1)(1) t(2t) 1 t
5. (a) s= 2
 v(t)=s (t)= =
2 2 2 2
t +1 (t +1) (t +1)
2
1 (3) 1 9  8 2
(b) v(3)= = = = ft / s
2 2 2 100 25
(3 +1) 10
2
(c) It is at rest when v=0 1 t =0 t=1 s [ t  1 since t 0 ].
2 2
(d) It moves in the positive direction when v>0 1 t >0 t <1 0 t<1 .
1 1
(e) Distance in positive direction =|s(1) s(0)|=  0 = ft
2 2
8 1 49
Distance in negative direction =|s(8) s(1)|=  = ft
65 2 130
1 49 57
Total distance traveled = + = ft
2 130 65

(f)

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

2 5/2 3/2 1/2

6. (a) s= t (3t  35t+90)=3t  35t +90t 
/ 15 3/2 105 1/2  1/2 15  1/2 2 15
v(t)=s (t)= t  t +45t = t (t  7t+6)= (t 1)(t 6)
2 2 2 2 t
15
(b) v(3)= (2)( 3)= 15 3 ft / s
2 3
(c) It is at rest when v=0 t=1 s or 6 s.
(d) It moves in the positive direction when v>0 (t 1)(t 6)>0 0 t<1 or t>6.
(e)
Distance in positive direction =|s(1) s(0)|+|s(8) s(6)|=|58 0|+|4 2  ( 12 6 )

=58+4 2 +12 6 93.05 ft

Distance in negative direction =|s(6) s(1)|=| 12 6  58|=58+12 6 87.39 ft

Total distance traveled =58+4 2 +12 6 +58+12 6 =116+4 2 +24 6 180.44 ft

(f)

3 2 / 2 2
7. s(t)=t  4.5t  7t  v(t)=s (t)=3t  9t 7=5  3t  9t 12=0  3(t 4)(t+1)=0 t=4 or  1 . Since t 0
, the particle reaches a velocity of 5 m / s at t=4 s.

2 ds
8. (a) s=5t+3t  v(t)==5+6t , so v(2)=5+6(2)=17 m / s.
dt
(b) v(t)=35 5+6t=35 6t=30 t=5 s.

dh2
9. (a) h=10t 0.83t  v(t)=
=10 1.66t , so v(3)=10 1.66(3)=5.02 m / s.
dt
2 2 10
17
(b) h=25 10t 0.83t =25 0.83t  10t+25=0 t= 3.54 or 8.51 .
1.66
The value t =(10 17 )/1.66 corresponds to the time it takes for the stone to rise 25 m and
1
t =(10+ 17 )/1.66 corresponds to the time when the stone is 25 m high on the way down. Thus,
2
v(t )=10 1.66[(10 17 )/1.66]= 17 4.12 m / s.
1

/ 5
10. (a) At maximum height the velocity of the ball is 0 ft / s. v(t)=s (t)=80 32t=0 32t=80 t= .
2
So the maximum height is
3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

5 5 5 2
s =80  16 =200 100=100 ft.
2 2 2
2 2 2
(b) s(t)=80t 16t =96  16t  80t+96=0  16(t  5t+6)=0  16(t 3)(t 2)=0 .
So the ball has a height of 96 ft on the way up at t=2 and on the way down at t=3 . At these times the
velocities are v(2)=80 32(2)=16 ft / s and v(3)=80 32(3)= 16 ft / s, respectively.

2 / / 2
11. (a) A(x)=x  A (x)=2x . A (15)=30 mm / mm is the rate at which the area is increasing with
respect to the side length as x reaches 15 mm.

1 1 /
(b) The perimeter is P(x)=4x , so A (x)=2x= (4x)= P(x) . The figure suggests that if  x is small,
2 2
then the change in the area of the square is approximately half of its perimeter ( 2 of the 4 sides)
2
times  x . From the figure,  A=2x( x)+( x) . If  x is small, then  A 2x( x) and so
 A/ x 2x .

dV 3 2 dV 2 3
12. (a) V (x)=x  =3x . =3(3) =27 mm / mm is the rate at which the volume is
dx dx x=3
increasing as x increases past 3 mm.
2 1 2 1 / 2
(b) The surface area is S(x)=6x , so V (x)=3x = (6x )= S(x) . The figure suggests that if  x is
2 2
small, then the change in the volume of the cube is approximately half of its surface area (the area of
2 2 3
3 of the 6 faces) times  x . From the figure,  V =3x ( x)+3x( x) +( x) . If  x is small, then
2 2
 V 3x ( x) and so  V / x 3x .

4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

13. (a)
(i) A(3) A(2) 9  4
= =5
3 2 1
(ii) A(2.5) A(2) 6.25  4
= =4.5
2.5 2 0.5
(iii) A(2.1) A(2) 4.41  4
= =4.1
2.1 2 0.1

2 / /
(b) A(r)= r  A (r)=2 r , so A (2)=4 .
/
(c) The circumference is C(r)=2 r=A (r) . The figure suggests that if  r is small, then the change in
the area of the circle (a ring around the outside) is approximately equal to its circumference times  r
. Straightening out this ring gives us a shape that is approximately rectangular with length 2 r and
2 2 2
width  r , so  A 2 r( r) . Algebraically,  A=A(r+ r) A(r)= (r+ r)   r =2 r( r)+ ( r) .
So we see that if  r is small, then  A 2 r( r) and therefore,  A/ r 2 r .

/ 2
14. (a) A (1)=7200 cm / s
/ 2
(b) A (3)=21 , 600 cm / s
/ 2
(c) A (5)=36 , 000 cm / s

/ 2
15. (a) S (1)=8 ft / ft
/ 2
(b) S (2)=16 ft / ft
/ 2
(c) S (3)=24 ft / ft
5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

16. (a)
(a) 4 4
 (512)  (125)
V (8) V (5) 3 3 3
= =172  m / m
8 5 3
(b) 4 4
 (512)  (125)
V (8) V (5) 3 3 3
= =172  m / m
8 5 3
(c) 4 4
 (216)  (125)
V (6) V (5) 3 3 3
= =121.3  m / m
6 5 1
(d) 4 4
 (216)  (125)
V (6) V (5) 3 3 3
= =121.3  m / m
6 5 1
(e) 4 3 4 3
 (5.1)   (5)
V (5.1) V (5) 3 3 3
= =102.013  m / m
5.1 5 0.1
(f) 4 3 4 3
 (5.1)   (5)
V (5.1) V (5) 3 3 3
= =102.013  m / m
5.1 5 0.1

/ 2 / 3
(b) V (r)=4 r , so V (5)=100  m / m.
4 3 / 2
(c) V (r)=  r  V (r)=4 r =S(r) . By analogy with Exercise 13(c) , we can say that the change in
3
the volume of the spherical shell,  V , is approximately equal to its thickness,  r , times the surface
2 2
area of the inner sphere. Thus,  V 4 r ( r) and so  V / r 4 r .

17. (a)  (1)=6 kg / m

(b)  (2)=12 kg / m
(c)  (3)=18 kg / m

/ 5
18. (a) V (5)= 250 1
= 218.75 gal / min
40
/ 10
(b) V (10)= 250 1 = 187.5 gal / min
40
/ 20
(c) V (20)= 250 1 = 125 gal / min
40
(d)
6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

/ 40
V (40)= 250 1 =0 gal / min
40

/ 2
19. (a) Q (0.5)=3(0.5)  4(0.5)+6=4.75 A
/ 2
(b) Q (1)=3(1)  4(1)+6=5 A

GmM dF  2  3 2GmM
20. (a) F= =(GmM)r  = 2(GmM)r = , which is the rate of change of the
2 dr 3
r r
force with respect to the distance between the bodies. The minus sign indicates that as the distance r
between the bodies increases, the magnitude of the force F exerted by the body of mass m on the
body of mass M is decreasing.
/ / 2GmM 3
(b) Given F (20 , 000)= 2 , find F (10 , 000) .  2=  GmM=20 , 000 .
3
20,000
/
F (10,000)=
(
2 20,000
3
) = 2 23= 16 N / km
3
10,000

21. (a) To find the rate of change of volume with respect to pressure, we first solve for V in terms of
P.
C dV C
PV =C  V =  = .
P dP 2
P
(b) From the formula for dV /dP in part (a), we see that as P increases, the absolute value of dV /dP
decreases. Thus, the volume is decreasing more rapidly at the beginning.
1 dV 1 C C C 1
(c)  = =  = = =
V dP V 2 (PV )P CP P
P

.
22. (a)
C(6) C(2) 0.0295 0.0570
=
6 2 4

.
= 0.006875 ( moles/L ) / min
(i)
C(4) C(2) 0.0408 0.0570
=
4 2 2
= 0.008 ( moles/L ) / min

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

(iii)

(b) Slope =
C(2) C(0)
2 0

 C
 t
=


0.0570 0.0800

0.077
7.8
2
.
= 0.0115 ( moles/L ) / min

 0.01 ( moles/L ) / min

1860 1750 110 2070 1860 210

23. (a) 1920: m = = =11 , m = = =21 ,
1 1920 1910 10 2 1930 1920 10
(m +m )2=(11+21)/2=16 million / year
1 2

4450 3710 740 5280 4450 830

1980: m = = =74 , m = = =83 ,
1 1980 1970 10 2 1990 1980 10
(m +m )2=(74+83)/2=78.5 million / year
1 2

3 2
(b) P(t)=at +bt +ct+d (in millions of people), where a 0.0012937063 , b  7.061421911 ,
c 12,822.97902 , and d  7,743,770.396 .
3 2 / 2
(c) P(t)=at +bt +ct+d P (t)=3at +2bt+c (in millions of people per year)
(d)
/ 2
P (1920) =3(0.0012937063)(1920) +2( 7.061421911)(1920)+12,822.97902
14.48 million / year

P (1980) 75.29 million / year (smaller, but close)

/

/
(e) P (1985) 81.62 million / year, so the rate of growth in 1985 was about 81.62 million / year.

4 3 2  6
24. (a) A(t)=at +bt +ct +dt+e , where a= 5.8275058275396 10 , b=0.0460458430461 ,
c= 136.43277039706 , d=179 , 661.02676871 , and e= 88 , 717 , 597.060767 .
4 3 2 / 3 2
(b) A(t)=at +bt +ct +dt+e A (t)=4at +3bt +2ct+d
(c)
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

/
A (1990) 0.0833 years of age per year

(d)

2
a kt
25. (a) C =  rate of reaction
akt+1
2 2 2 2
d C (akt+1)(a k) (a kt)(ak) a k(akt+1 akt) ak
= = = =
dt 2 2 2
(akt+1) (akt+1) (akt+1)
2 2 2
a kt a kt+a a kt a
(b) If x= C , then a x=a = = .
akt+1 akt+1 akt+1
2 2
2 a ak d C dx
So k(a x) =k = = = .
akt+1 2 dt dt
(akt+1)

2
26. (a) After an hour the population is n(1)=3 500 ; after two hours it is n(2)=3(3 500)=3  500 ; after
2 3 4
three hours, n(3)=3(3  500)=3  500 ; after four hours, n(4)=3  500 . From this pattern, we see that the
t t
population after t hours is n(t)=3  500=500 3 .
d x x t
(b) From (5) in Section 3.1, we have (3 ) (1.10)3 . Thus, for n(t)=500 3 ,
dx
dn d t t dn 6
=500 (3 ) 500(1.10)3  500(1.10)3 400 , 950 bacteria / hour.
dt dt dt t=6

P 2 2
27. (a) Using v= (R  r with R=0.01 , l=3 , P=3000 , and  =0.027 , we have v as a function of r :
4 l
3000 2 2
v(r)= (0.01  r ) . v(0)=0.925 cm / s, v(0.005)=0.694 cm / s, v(0.01)=0 .
4(0.027)3
P 2 2 / P Pr
(b) v(r)= (R  r ) v (r)= ( 2r)= . When l=3 , P=3000 , and  =0.027 , we have
4 l 4 l 2 l

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

/ 3000r / / /
v (r)= . v (0)=0 , v (0.005)= 92.592(cm/s)/ cm, and v (0.01)= 185.185(cm/s)/ cm.
2(0.027)3
(c) The velocity is greatest where r=0 (at the center) and the velocity is changing most where
r=R=0.01 cm (at the edge).

28. (a)
(a) 1 T 1 T  1 df 1 T  2 1 T
f= = L  = L =
2L  2  dL 2  2 
2L
(b) 1 T 1 T  1 df 1 T  2 1 T
f= = L  = L =
2L  2  dL 2  2 
2L
(c) 1 T 1 1/2 df 1 1  1/2 1
f= = T  = T =
2L  2L  dT 2 2L  4L T
(d) 1 T 1 1/2 df 1 1  1/2 1
f= = T  = T =
2L  2L  dT 2 2L  4L T
(e) 1 T T  1/2 df 1 T  3/2 T
f= =   =  =
2L  2L d 2 2L 3/2
4L
(f) 1 T T  1/2 df 1 T  3/2 T
f= =   =  =
2L  2L d 2 2L 3/2
4L

(b)
(i) df
<0 and L is decreasing  f is increasing  higher note
dL
(ii) df
>0 and T is increasing  f is increasing  higher note
dT
(iii) df
<0 and  is increasing  f is decreasing  lower note
d

2 3 / 2
29. (a) C(x)=2000+3x+0.01x +0.0002x  C (x)=3+0.02x+0.0006x
/ /
(b) C (100)=3+0.02(100)+0.0006(10 , 000)=3+2+6=$11/ pair. C (100) is the rate at which the cost
is increasing as the 100 th pair of jeans is produced. It predicts the cost of the 101 st pair.
(c) The cost of manufacturing the 101 st pair of jeans is

10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

C(101) C(100) = (2000+303+102.01+206.0602) (2000+300+100+200)

= 11.0702 $11.07

2 3 / 2 /
30. (a) C(x)=84+0.16x 0.0006x +0.000003x  C (x)=0.16 0.0012x+0.000009x  C (100)=0.13 .
This is the rate at which the cost is increasing as the 100 th item is produced.
(b) C(101) C(100)=97.13030299 97 $0.13 .

/ /
p(x) / xp (x) p(x) 1 xp (x) p(x) /
31. (a) A(x)=  A (x)= = . A (x)>0 A(x) is increasing; that is,
x 2 2
x x
the average productivity increases as the size of the workforce increases.
/ / / p(x)
(b) p (x) is greater than the average productivity  p (x)>A(x) p (x)> 
x
/
/ / xp (x) p(x) /
xp (x)> p(x) xp (x) p(x)>0 >0 A (x)>0 .
2
x

32. (a)

S=
dR
=
1+4x ( 0.4
) ( 9.6x 0.6)  ( 40+24x0.4) ( 1.6x 0.6)
dx ( 1+4x0.4) 2
 0.6  0.2  0.6  0.2  0.6
9.6x +38.4x  64x  38.4x 54.4x
= =
0.4 2
( 1+4x ) ( 1+4x0.4) 2

(b)
At low levels of brightness, R is quite large and is quickly decreasing, that is, S is negative with large
absolute value. This is to be expected: at low levels of brightness, the eye is more sensitive to slight
changes than it is at higher levels of brightness.

PV PV 1
33. PV =nRT  = T= = (PV ) . Using the Product Rule, we have
nR (10)(0.0821) 0.821
dT 1 / / 1
= [P(t)V (t)+V (t)P (t)]= [(8)( 0.15)+(10)(0.10)]  0.2436 K / min.
dt 0.821 0.821

11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.3 Rates of Change in the Natural and Social Sciences

34. (a) If dP/dt=0 , the population is stable (it is constant).

dP P  P P  
(b) =0  P=r 1 P =1  =1  P=P 1 .
dt 0 P r P P r c r
c 0 c c 0 0
4
If P =10 , 000 , r =5%=0.05 , and  =4%=0.04 , then P=10 , 000 1 =2000 .
c 0 5
5
(c) If  =0.05 , then P=10 , 000 1 =0 . There is no stable population.
5

35. (a) If the populations are stable, then the growth rates are neither positive nor negative; that is,
dC dW
=0 and =0 .
dt dt
(b) ‘‘The caribou go extinct’’ means that the population is zero, or mathematically, C=0 .
dC dW
(c) We have the equations =aC bCW and = cW +dCW . Let dC/dt=dW /dt=0 , a=0.05 ,
dt dt
b=0.001 , c=0.05 , and d=0.0001 to obtain 0.05C 0.001CW =0 (1) and  0 .05W +0.0001CW =0 (2) .
Adding 10 times (2) to (1) eliminates the CW  terms and gives us 0.05C 0.5W =0 C=10W .
Substituting C=10W into (1) results in
2 2
0.05(10W ) 0.001(10W )W =0  0.5W  0.01W =0  50W  W =0  W (50 W )=0  W =0 or 50 . Since
C=10W , C=0 or 500 . Thus, the population pairs (C,W) that lead to stable populations are (0,0) and
(500,50) . So it is possible for the two species to live in harmony.

12
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

/
1. f (x)=x 3sin x f (x)=1 3cos x

/
2. f (x)=xsin x f (x)=x cos x+(sin x) 1=xcos x+sin x

/ 2
3. y=sin x+10tan x y =cos x+10sec x

/
4. y=2 x+5cos x y = 2 xcot x 5sin x

3 / 3 2 2 3 2
5. g(t)=t cos t  g (t)=t ( sin t)+(cos t) 3t =3t cos t t sin t or t (3cos t tsin t)

/ 2
6. g(t)=4sec t+tan t  g (t)=4sec ttan t+sec t


7. h( )=csc  +e cot  
/  2   2
h ( )= csc  cot  +e ( csc  )+(cot  )e = csc  cot  +e (cot   csc  )

u / u u u
8. y=e (cos u+cu) y =e ( sin u+c)+(cos u+cu)e =e (cos u sin u+cu+c)

x / (cos x)(1) (x)( sin x) cos x+xsin x

9. y=  y = =
cos x 2 2
(cos x) cos x

1+sin x
10. y= 
x+cos x
2 2
/ (x+cos x)(cos x) (1+sin x)(1 sin x) xcos x+cos x (1 sin x)
y = 2
=
2
(x+cos x) (x+cos x)
2 2
xcos x+cos x (cos x) xcos x
= =
2 2
(x+cos x) (x+cos x)

sec 
11. f ( )= 
1+sec 
(1+sec  )(sec  tan  ) (sec  )(sec  tan  ) (sec  tan  )[(1+sec  ) sec  ] sec  tan 
/ = = =
f ( ) 2 2 2
(1+sec  ) (1+sec  ) (1+sec  )

tan x 1
12. y= 
sec x
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

2
dy sec xsec x (tan x 1)sec xtan x sec x sec x tan x+tan x
= = =
1+tan x ( 2 2
)
dx 2 2 sec x
sec x sec x
/
Another method: Simplify y first: y=sin x cos x y =cos x+sin x .

2
sin x / x cos x (sin x)(2x) x(xcos x 2sin x) xcos x 2sin x
13. y=  y = = =
2
x ( x2) 2 x
4
x
3

14. y= ( +cot  )

/ 2 2 2
y = (1 csc  )+( +cot  )(  cot  )= (1 csc    cot   cot  )
2 2 2 2
= ( cot    cot   cot  ) {1+cot  =csc  }
2
= (  cot   2cot  )=  cot  ( +2cot  )

/ 2 2 2
15. y=sec  tan   y =sec  (sec  )+tan  (sec  tan  )=sec  (sec  +tan  )
2 2
Using the identity 1+tan  =sec  , we can write alternative forms of the answer as
2 2
sec  (1+2tan  ) or sec  (2sec   1)

/ / / /
16. Recall that if y= fgh , then y = f gh+ fg h+ fgh . y=xsin xcos x
dy 2 2
=sin xcos x+xcos xcos x+xsin x( sin x)=sin xcos x+xcos x xsin x
dx

d d 1 (sin x)(0) 1(cos x)  cos x 1 cos x

17. ( csc(x) ) = = = =  = xcot x
dx dx sin x 2 2 sin x sin x
sin x sin x

d d 1 (cos x)(0) 1( sin x) sin x 1 sin x

18. ( sec x ) = = = =  =sec xtan x
dx dx cos x 2 2 cos x cos x
cos x cos x

2 2
d d cos x (sin x)( sin x) (cos x)(cos x) sin x+cos x 1 2
19. ( cot x ) = = = = = x
dx dx sin x 2 2 2
sin x sin x sin x

20. f (x)=cos x
/ f (x+h) f (x) cos (x+h) cos x cos xcos h sin xsin h cos x
f (x) = lim = lim = lim
h 0 h h 0 h h 0 h

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

cos h 1 sin h cos h 1 sin h

= lim cos x  sin x =cos x lim  sin x lim
h 0 h h h 0 h h 0 h
=(cos x)(0) (sin x)(1)= sin x

/ 2  2 2
21. y=tan x y =sec x the slope of the tangent line at ,1 is sec =( 2 ) =2 and an
4 4
 
equation of the tangent line is y 1=2 x or y=2x+1 .
4 2

x / x x x
22. y=e cos x y =e ( sin x)+(cos x)e =e (cos x sin x)  the slope of the tangent line at (0, 1) is
0
e (cos 0 sin 0)=1(1 0)=1 and an equation is y 1=1(x 0) or y=x+1 .

/ /
23. y=x+cos x y =1 sin x . At ( 0,1 ) , y =1 , and an equation of the tangent line is y 1=1(x 0) , or
y=x+1 .

1 / cos x sin x / 1 0
24. y=  y = [ Reciprocal Rule]. At ( 0,1 ) , y = = 1 , and an
sin x+cos x 2 2
(sin x+cos x) (0+1)
equation of the tangent line is y 1= 1(x 0) , or y= x+1 .

/
25. (a) y=xcos x y =x( sin x)+cos x(1)=cos x xsin x . So the slope of the tangent at the point
( ,  ) is cos    sin  = 1  (0)= 1 , and an equation is y+ = (x  ) or y= x .

(b)

/
26. (a) y=sec x 2cos x y =sec xtan x+2sin x
    3
the slope of the tangent line at ,1 is sec tan +2sin =2 3 +2 =3 3 and an
3 3 3 3 2

equation is y 1=3 3 x or y=3 3 x+1  3 .
3
(b)

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

/ 2
27. (a) f (x)=2x+cot x f (x)=2 csc x

(b)
/
Notice that f (x)=0 when f has a horizontal tangent.
/ / /
f is positive when f is increasing and f is negative when f is decreasing. Also, f (x) is large
negative when the graph of f is steep.

/ 1  1/2 sin x
28. (a) f (x)= x sin x f (x)= x cos x+(sin x) x = x cos x+
2 2 x

(b)
/
Notice that f (x)=0 when f has a horizontal tangent.
/ /
f is positive when f is increasing and f is negative when f is decreasing.

/ 1
29. f (x)=x+2sin x has a horizontal tangent when f (x)=0  1+2cos x=0 cos x= 
2
2 4 4 2 
x= +2 n or +2 n , where n is an integer. Note that and are  units from  .
3 3 3 3 3

This allows us to write the solutions in the more compact equivalent form (2n+1)  , n an
3
integer.

2 2
cos x / (2+sin x)( sin x) cos xcos x  2sin x sin x cos x  2sin x 1
30. y=  y = = = =0 when
2+sin x 2 2 2
(2+sin x) (2+sin x) (2+sin x)
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

11 1 7 1 1
 2sin x 1=0 sin x=  x=
+2 n or x= +2 n , n an integer. So y= or y= and
6 2 6 3 3
11 1 7 1
the points on the curve with horizontal tangents are: +2 n, , +2 n, ,
6 3 6 3
n an integer.

/
31. (a) x(t)=8sin t v(t)=x (t)=8cos t
2 2 2 3
(b) The mass at time t= has position x =8sin =8 =4 3 and velocity
3 3 3 2
2 2 1 2
v =8cos =8  = 4 . Since v <0 , the particle is moving to the left.
3 3 2 3

32. (a) s(t)=2cos t+3sin t v(t)= 2sin t+3cos t

(b)
(c) s=0 t 2.55 . So the mass passes through the equilibrium position for the first time when
2
t 2.55 s.
(d) v=0 t 0.98 , s(t ) 3.61 cm. So the mass travels a maximum of about 3.6 cm (upward and
1 1
downward) from its equilibrium position.
(e) The speed |v| is greatest when s=0 ; that is, when t=t +n , n a positive integer.
2

33.
From the diagram we can see that sin  =x/10 x=10sin  . We want to find the rate of change of x
with respect to  ; that is, dx/d . Taking the derivative of the above expression, dx/d =10(cos  ) .
 dx
1
So when  = , =10cos =10 =5 ft/rad.
3 d 3 2

 W dF ( sin  +cos  )(0)  W ( cos   sin  )  W (sin    cos  )

34. (a) F=  = =
 sin  +cos  d 2 2
( sin  +cos  ) ( sin  +cos  )
(b)
5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

dF  1
=0  W (sin    cos  )=0 sin  = cos   tan  =   =tan 
d

(c)
0.6(50) dF
From the graph of F= for 0   1 , we see that =0  0.54 . Checking this
0.6sin  +cos  d
 1
with part (b) and  =0.6 , we calculate  =tan 0.6 0.54 . So the value from the graph is consistent
with the value in part (b).

35.
sin 3x 3sin 3x
lim =lim [ multiply numerator and denominator by 3 ]
x 0 x x 0 3x
sin 3x
=3 lim [ as x 0 , 3x 0]
3x 0 3x
sin 
=3 lim [ let  =3x ]
  0 
=3(1) [ Equation 2]
=3

36.
sin 4x sin 4x x 4sin 4x 6x
lim =lim  =lim  lim
x 0 sin 6x x 0 x sin 6x x 0 4x x 0 6sin 6x
sin 4x 1 6x 1 2
=4lim  lim =4(1) (1)=
x 0 4x 6 x 0 sin 6x 6 3

37.
tan 6t sin 6t 1 t 6sin 6t 1 2t
lim =lim   =lim  lim  lim
t 0 sin 2t t 0 t cos 6t sin 2t t 0 6t t 0 cos 6t t 0 2sin 2t
sin 6t 1 1 2t 1 1
=6lim  lim  lim =6(1)  (1)=3
t 0 6t t 0 cos 6t 2 t 0 sin 2t 1 2

38.

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

cos   1 cos   1
lim
cos   1    0  0
lim = lim = = =0
  0 sin    0 sin  sin  1
lim
   0 

39. lim
sin (cos  )
=
sin
( lim (cos  )) = sin 1 =sin 1
  0

  0 sec  lim (sec  ) 1

  0

40.
2
sin 3t sin 3t sin 3t sin 3t sin 3t
lim =lim  =lim  lim
t 0 2 t 0 t t t 0 t t 0 t
t
sin 3t 2 sin 3t 2 2
= lim = 3lim =(3 1) =9
t 0 t t 0 3t

41.
lim [(sin x)/x]
cot 2x cos2x sinx (sin x)/x x 0
lim =lim =lim cos 2x =lim cos2x
x 0 x x 0 sin2x x 0 (sin 2x)/x x 0 2 lim [(sin 2x)/2x]
x 0
1 1
=1 =
2 1 2

42.
sin x cos x sin x cos x sin x cos x
lim = lim = lim
x  /4 cos x sin x x  /4 (cos x+sin x)(cos x sin x)
cos 2x 2 2
x  /4
 1  1  1
= lim = =
x  /4 cos x+sin x cos  +sin  2
4 4

43. Divide numerator and denominator by  . ( sin( ) also works.)

sin  sin 
lim
sin     0  1 1
lim = lim = = =
  0  +tan    0 1+ sin   1 sin  1 1+1 1 2
1+ lim lim
 cos    0    0 cos 

44.
7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

sin (x 1) sin (x 1) 1 sin (x 1) 1 1

lim =lim =lim lim =  1=
1 (x+2)(x 1) x 1 x+2 x 1 x 1 3 3
2
x 1 x +x 2 x

2 2
d d sin x 2 cos xcos x sin x( sin x) cos x+sin x 2 1
45. (a) tan x=  sec x= = . So sec x= .
dx dx cos x 2 2 2
cos x cos x cos x
d d 1 (cos x)(0) 1( sin x) sin x
(b) sec x=  sec xtan x= . So sec xtan x= .
dx dx cos x 2 2
cos x cos x
d d 1+cot x
(c) (sin x+cos x)= 
dx dx x
2 2
x( csc x) (1+cot x)( xcot x) x[ csc x+(1+cot x)cot x]
cos x sin x = 2
= 2
csc x csc x
2 2
 csc x+cot x+cot x  1+cot x
= =
x x

cot x 1
So cos x sin x= .
x

 r
46. Let PR =x . Then we get the following formulas for r and h in terms of  and x : sin = 
2 x
 h   1 2 1
r=xsin and cos =  h=xcos . Now A ( ) =  r and B ( ) = (2r)h=rh . So
2 2 x 2 2 2
1 2
A ( )  r
lim 2 1 r 1 xsin ( /2 )
+ B ( ) = lim =  lim =  lim
  0 + rh 2 + h 2 + xcos ( /2 )
  0   0   0
1
=  lim tan ( /2 ) =0 .
2 +
  0

47. By the definition of radian measure, s=r , where r is the radius of the circle.
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.4 Derivatives of Trigonometric Functions

 d/2 
By drawing the bisector of the angle  , we can see that sin =  d=2rsin .
2 r 2
s r 2 ( /2 )  /2
So lim = lim = lim = lim =1 .
+ d + 2rsin ( /2 ) + 2sin ( /2 )   0 sin ( /2 )
  0   0   0

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

dy dy du
1. Let u=g(x)=4x and y= f (u)=sin u . Then = =(cos u)(4)=4cos 4x .
dx du dx

1/2 dy dy du 1  1/2 3 3
2. Let u=g(x)=4+3x and y= f (u)= u =u . Then = = u (3)= = .
dx du dx 2 2 u 2 4+3x

2 10 dy dy du 9 29
3. Let u=g(x)=1 x and y= f (u)=u . Then = =(10u )( 2x)= 20x(1 x ) .
dx du dx

dy dy du 2 2
4. Let u=g(x)=sin x and y= f (u)=tan u . Then = =(sec u)(cos x)=(sec u)(sin x) cos x , or
dx du dx
2
equivalently, sec (sin x) cos x .

dy dy du 1  1/2 cos x cos x

5. Let u=g(x)=sin x and y= f (u)= u . Then = = u cos x= = .
dx du dx 2 2 u 2 sin x

x dy dy du x x x
6. Let u=g(x)=e and y= f (u)=sin u . Then = =(cos u)(e )=e cos e .
dx du dx

3 7 / 3 6 2 6 2 6 2
7. F(x)=(x +4x)  F (x)=7(x +4x) (3x +4) [ or 7x (x +4) (3x +4) ]

2 3 / 2 2
8. F(x)=(x  x+1)  F (x)=3(x  x+1) (2x 1)

4 3 3 1/4
9. F(x)=1+2x+x =(1+2x+x ) 
1 3  3/4 d 3 1 2
/ = (1+2x+x )  (1+2x+x )=  (2+3x )
F (x) 4 dx 3 3/4
4(1+2x+x )
2 2
2+3x 2+3x
= =
3 3/4 4 33
4(1+2x+x ) 4 (1+2x+x )
3
4 2/3 / 2 4 1/3 3 8x
10. f (x)=(1+x )  f (x)= (1+x ) (4x )=
3 3 4
3 1+x

3
1 4  3 / 4  4 3 3 4  4  12t
11. g(t)= =(t +1)  g (t)= 3(t +1) (4t )= 12t (t +1) =
4 3 4 4
(t +1) (t +1)

12.
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

2
3 1/3 / 1  2/3 2 sec t
f (t)= 1+tan t =(1+tan t)  f (t)= (1+tan t) sec t=
3 3 2
3 (1+tan t)

3 3 / 3 3 2 3 2 3 3
13. y=cos (a +x ) y = sin (a +x ) 3x [ a is just a constant] = 3x sin (a +x )

3 3 / 2 3 2
14. y=a +cos x y =3(cos x) ( sin x) [ a is just a constant] = 3sin xcos x

 mx /  mx d  mx  mx
15. y=e  y =e ( mx)=e ( m)= me
dx

/
16. y=4sec 5x y =4sec 5xtan 5x(5)=20sec 5xtan 5x

5 28
17. g(x)=(1+4x) (3+x x ) 
/ 5 27 28 4
g (x) =(1+4x)  8(3+x x ) (1 2x)+(3+x x )  5(1+4x)  4
4 27 2
=4(1+4x) (3+x x ) [2(1+4x)(1 2x)+5(3+x x )]
4 27 2 2
=4(1+4x) (3+x x ) [(2+4x 16x )+(15+5x 5x )]
4 27 2
=4(1+4x) (3+x x ) (17+9x 21x )
4 3 3 4
18. h(t)=(t  1) (t +1) 
/ 4 3 3 3 2 3 4 4 2 3
h (t) =(t  1)  4(t +1) (3t )+(t +1)  3(t  1) (4t )
2 4 2 3 3 4 3 2 4 2 3 3 4
=12t (t  1) (t +1) [(t  1)+t(t +1)]=12t (t  1) (t +1) (2t +t 1)

4 2  3
19. y=(2x 5) (8x  5) 
/ 3 2  3 4 2  4
y =4(2x 5) (2)(8x  5) +(2x 5) ( 3)(8x  5) (16x)
3 2  3 4 2  4
=8(2x 5) (8x  5)  48x(2x 5) (8x  5)

2
2 2 1/3 / 2 1/3 2 1 2  2/3 2 1/3 x +1
20. y=(x +1)(x +2)  y =2x(x +2) +(x +1) (x +2) (2x)=2x(x +2) 1+
3 2
3(x +2)
2 2 2 2 2
 x /  x  x  x 2  x 2
21. y=xe  y =xe ( 2x)+e  1=e ( 2x +1)=e (1 2x )

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

 5x /  5x  5x  5x
22. y=e cos 3x y =e ( 3sin 3x)+(cos 3x)( 5e )= e (3sin 3x+5cos 3x)

xcos x / xcos x d xcos x xcos x

23. y=e  y =e  (xcos x)=e x( sin x)+(cos x) 1 =e (cos x xsin x)
dx
2 2 2
1 x / 1 x d 2 1 x
24. Using Formula 5 and the Chain Rule, y=10  y =10 (ln 10) (1 x )= 2x(ln 10)10 .
dx

z 1 z 1 1/2
25. F(z)= = 
z+1 z+1
1 z 1  1/2 d z 1 1 z+1 1/2 (z+1)(1) (z 1)(1)
/
=
F (z) 2  = 
z+1 dz z+1 2 z 1 2
(z+1)
1/2 1/2
1 (z+1) z+1 z+1 1 (z+1) 2 1
=  =  =
2 1/2 2 2 1/2 2 1/2 3/2
(z 1) (z+1) (z 1) (z+1) (z 1) ( z+1 )

4
(y 1)
26. G(y)= 
2 5
(y +2y)
2 5 3 4 2 4
/ (y +2y)  4(y 1)  1 (y 1)  5(y +2y) (2y+2)
G (y) = 2 52
[(y +2y) ]
2 4 3 2
2(y +2y) (y 1) [2(y +2y) 5(y 1)(y+1)]
=
2 10
(y +2y)
3 2 2 3 2
2(y 1) [(2y +4y)+( 5y +5)] 2(y 1) ( 3y +4y+5)
= =
2 6 2 6
(y +2y) (y +2y)

r
27. y= 
2
r +1
2 2 2 2
2 r r +1 r +1  r
2 1 2  1/2 r +1 
/ r +1 (1) r (r +1) (2r) 2 2
y 2 r +1 r +1
= = =
2 2 2
( r +1 )
2
( r +1 )
2
( r +1 )
2

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

2 2
(r +1) r 1 2  3/2
= = or (r +1)
3 2 3/2
( r +1 )
2 (r +1)
2  1/2
Another solution: Write y as a product and make use of the Product Rule. y=r (r +1) 
/ 1 2  3/2 2  1/2
y =r  2 (r +1) (2r)+r +1)  1
2  3/2 2 2 1 2  3/2 2  3/2
=(r +1) [ r +(r +1) ]=(r +1) (1)=(r +1)

2  3/2
The step that students usually have trouble with is factoring out (r +1) . But this is no different
2 2 5
than factoring out x from x +x ; that is, we are just factoring out a factor with the smallest exponent
3 1
that appears on it. In this case,  is smaller than  .
2 2

2u
e
28. y= 
u  u
e +e
u  u 2u 2u u  u 2u u  u u  u 2u u  u
/ (e +e )(e  2) e (e  e ) e (2e +2e  e +e ) e (e +3e )
y = = =
u  u2 u  u2 u  u2
(e +e ) (e +e ) (e +e )
Another solution: Eliminate negative exponents by first changing the form of y .
2u u 3u
e e e
y=  = 
u  u u 2u
e +e e e +1
2u 3u 3u 2u 3u 2u 2u 3u 2u
/ (e +1)(3e ) e (2e ) e (3e +3 2e ) e (e +3)
y = = =
2u 2 2u 2 2u 2
(e +1) (e +1) (e +1)

/ 2 2
29. y=tan (cos x) y =sec (cos x) ( sin x)= sin x sec (cos x)

2
sin x
30. y= 
cos x
2 2 2 2
/ cos x(2sin x cos x) sin x( sin x) sin x(2cos x+sin x) sin x(1+cos x)
y = 2
= 2
= 2
cos x cos x cos x
2
=sin x(1+sec x)

Another method: y=tan x sin x

4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

/ 2 2
y =sec x sin x+tan x cos x=sec x sin x+sin x

sin  x
31. Using Formula 5 and the Chain Rule, y=2 
/ sin  x d sin  x sin  x
y =2 (ln 2) (sin  x)=2 (ln 2) cos  x  =2 ( ln 2)cos  x
dx

2 2 / d 2 2
32. y=tan (3 )=(tan 3 )  y =2(tan 3 ) (tan 3 )=2tan 3  sec 3  3=6tan 3 sec 3
d

2 6 / 2 5 2 5
33. y=(1+cos x)  y =6(1+cos x)  2cos x( sin x)= 12cos xsin x(1+cos x)

1 / 1 1 1 1 1 1
34. y=xsin  y =sin +xcos  =sin  cos
x x x 2 x x x
x

2 2 2 2
35. y=sec x+tan x=(sec x) +(tan x) 
/ 2 2 2 2
y =2(sec x) (sec xtan x)+2(tan x)(sec x)=2sec x tan x+2sec x tan x=4sec x tan x

2
ktan x / ktan x d ktan x 2 1  1/2 ksec x ktan x
36. y=e  y =e  (ktan x )=e ksec x x = e
dx 2 2 x

2 2
37. y=cot (sin  )=[cot (sin  )] 
/ d 2 2
y =2[cot (sin  )] [cot (sin  )]=2cot (sin  ) [ csc (sin  ) cos  ]= 2cos   cot (sin  ) csc (sin  )
d
/ d
38. y=sin (sin (sin x)) y =cos (sin (sin x)) (sin (sin x))=cos (sin (sin x)) cos (sin x) cos x
dx

/ 1  1/2 1  1/2 1 1
39. y= x+ x  y = (x+ x ) 1+ x = 1+
2 2 2 x+ x 2 x

40. y= x+ x+ x  y =
/ 1
2
(
x+ x+ x )  1/2 1
1+ (x+ x )
2
 1/2 1 
1+ x
2
1/2

41. y=sin (tan sin x )

/ d 2 d 1/2
y =cos (tan sin x ) dx (tan sin x )=cos (tan sin x ) sec sin x dx (sin x)
2 1  1/2
=cos (tan sin x )sec sin x (sin x)  cos x
2

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

1
=cos (tan sin x ) sec ( 2
sin x ) 2 sin x
(cos x)

2 2 2

42. y=2 
3
x
/
y =2 (ln 2)
d
dx
x
3
x 3 x
3 =2 (ln 2)3 (ln 3)(2x) ( ) 2 x 2

10 / 9 9 / 9
43. y=(1+2x)  y =10(1+2x)  2=20(1+2x) . At (0,1) , y =20(1+0) =20 , and an equation of the
tangent line is y 1=20(x 0) , or y=20x+1 .

2 / /
44. y=sin x+sin x y =cos x+2sin xcos x . At (0,0) , y =1 , and an equation of the tangent line is
y 0=1(x 0) , or y=x .

/ /
45. y=sin (sin x) y =cos (sin x) cos x . At ( ,0) , y =cos (sin  ) cos  =cos (0) ( 1)=1( 1)= 1 ,
and an equation of the tangent line is y 0= 1(x  ) , or y= x+ .

2  x / 2  x  x  x 2  x 1 /  1  1 1
46. y=x e  y =x ( e )+e (2x)=2xe  x e . At 1, , y =2e  e = . So an equation of
e e
1 1 1
the tangent line is y = (x 1) or y= x .
e e e

 x  x  x
2 / (1+e )(0) 2( e ) 2e
47. (a) y=  y = = .
 x  x2  x2
1+e (1+e ) (1+e )
0
/ 2e 2(1) 2 1
At (0, 1) , y = = = = .
02 2 2 2
(1+e ) (1+1) 2
1 1
So an equation of the tangent line is y 1= (x 0) or y= x+1 .
2 2

(b)

x
48. (a) For x>0 , |x|=x , and y= f (x)= 
2
2 x

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

2 1 2  1/2
2 x (1) x (2 x ) ( 2x) 2 1/2
/ 2 (2 x )
f (x) = 
2 2 1/2
( 2 x )
2 (2 x )
2 2
(2 x )+x 2
= =
2 3/2 2 3/2
(2 x ) (2 x )

/
So at (1,1) , the slope of the tangent line is f (1)=2 and its equation is y 1=2(x 1) or y=2x 1 .

(b)

2
1 x
49. (a) f (x)= 
x
1  1/2 2
x (1 x ( 2x) 1 x (1) 2
/ 2 1 x
f (x) = 
2 2
x 1 x
2 2
 x  (1 x )  1
= =
2 2 2 2
x 1 x x 1 x

(b)
/
Notice that all tangents to the graph of f have negative slopes and f (x)<0 always.

50. (a)

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

From the graph of f , we see that there are 5 horizontal tangents, so there must be 5 zeros on the
/ /
graph of f . From the symmetry of the graph of f , we must have the graph of f as high at x=0 as
/
it is low at x= . The intervals of increase and decrease as well as the signs of f are indicated in the
figure.

(b)
f (x) =sin (x+sin 2x)
/ d
f (x) =cos (x+sin 2x) dx (x+sin 2x)
=cos (x+sin 2x)(1+2cos 2x)

/ 2 /
51. For the tangent line to be horizontal, f (x)=0 . f (x)=2sin x+sin x f (x)=2cos x+2sin xcos x=0
 3
 2cos x ( 1+sin x ) =0 cos x=0 or sin x= 1 , so x= +2n or +2n , where n is any integer.
2 2
 3
Now f =3 and f = 1 , so the points on the curve with a horizontal tangent are
2 2
 3
+2n ,3 and +2n , 1 , where n is any integer.
2 2

52. f (x)=sin 2x 2sin x

8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

/ 2 2
f (x)=2cos 2x 2cos x=4cos x 2cos x 2 , and 4cos x 2cos x 2=0 (cos x 1)(4cos x+2)=0
1 
cos x=1 or cos x= . So x=2n or (2n+1)  , n any integer.
2 3

/ / /
53. F(x)= f (g(x)) F (x)= f (g(x)) g (x) ,
/ / / / / /
so F (3)= f (g(3)) g (3)= f (6) g (3)=7 4=28 . Notice that we did not use f (3)=2 .

/ / /
54. w=u v w(x)=u(v(x))  w ( x ) =u (v(x)) v (x) , so
/ / / / / /
w (0)=u (v(0)) v (0)=u (2) v (0)=4 5=20 . The other pieces of information, u(0)=1 , u (0)=3 ,
/
and v (2)=6 , were not needed.

/ / / / / / /
55. (a) h(x)= f (g(x)) h (x)= f (g(x)) g (x) , so h (1)= f (g(1)) g (1)= f (2) 6=5 6=30 .
/ / / / / / /
(b) H(x)=g( f (x)) H (x)=g ( f (x)) f (x) , so H (1)=g ( f (1)) f (1)=g (3) 4=9 4=36 .

/ / / / / / /
56. (a) F(x)= f ( f (x)) F (x)= f ( f (x)) f (x) , so F (2)= f ( f (2)) f (2)= f (1) 5=4 5=20 .
/ / / / / / /
(b) G(x)=g(g(x)) G (x)=g (g(x)) g (x) , so G (3)=g (g(3)) g (3)=g (2) 9=7 9=63 .

/ / / / / / / / /
57. (a) u(x)= f (g(x)) u (x)= f (g(x))g (x) . So u (1)= f (g(1))g (1)= f (3)g (1) . To find f (3) ,
3 4 1 /
note that f is linear from (2,4) to (6,3) , so its slope is = . To find g (1) , note that g is linear
6 2 4
0 6 / / 1 3
from (0,6) to (2,0) , so its slope is = 3 . Thus, f (3)g (1)=  ( 3)= .
2 0 4 4
/ / / / / / / /
(b) v(x)=g( f (x)) v (x)=g ( f (x)) f (x) . So v (1)=g ( f (1)) f (1)=g (2) f (1) , which does not
/
exist since g (2) does not exist.
/ / / / / / / / /
(c) w(x)=g(g(x)) w (x)=g (g(x))g (x) . So w (1)=g (g(1))g (1)=g (3)g (1) . To find g (3) ,
2 0 2 / / 2
note that g is linear from (2,0) to (5,2) , so its slope is = . Thus, g (3) g (1)= ( 3)= 2
5 2 3 3
.

/ / /
58. (a) h(x)= f ( f (x)) h (x)= f ( f (x)) f (x) .
/ / / / /
So h (2)= f ( f (2)) f (2)= f (1) f (2) ( 1)( 1)=1 .
2 / / 2 d 2 / 2 / / 2 /
(b) g(x)= f (x ) g (x)= f (x ) (x )= f (x )(2x) . So g (2)= f (2 )(2 2)=4 f (4) 4(1.5)=6 .
dx

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

/ / / / / / / /
59. h(x)= f (g(x)) h (x)= f (g(x))g (x) . So h (0.5)= f (g(0.5))g (0.5)= f (0.1)g (0.5) . We can
estimate the derivatives by taking the average of two secant slopes.
m +m
/ 14.8 12.6 18.4 14.8 / 1 2 22+36
For f ( 0.1 ) : m = =22 , m = =36 . So f (0.1) = =29 .
1 0.1 0 2 0.2 0.1 2 2
m +m
/ 0.10 0.17 0.05 0.10 / 1 2
For g (0.5) : m = = 0.7 , m = = 0.5 . So g (0.5) = 0.6 .
1 0.5 0.4 2 0.6 0.5 2
/ / /
Hence, h (0.5)= f (0.1)g (0.5) (29)( 0.6)= 17.4 .

/ / / / / / / /
60. g(x)= f ( f (x))
g (x)= f ( f (x)) f (x) . So g (1)= f ( f (1)) f (1)= f (2) f (1) .
m +m
/ 3.1 2.4 4.4 3.1 / 1 2
For f (2) : m = =1.4 , m = =2.6 . So f (2) =2 .
1 2.0 1.5 2 2.5 2.0 2
m +m
/ 2.0 1.8 2.4 2.0 / 1 2
For f (1) : m = =0.4 , m = =0.8 . So f (1) =0.6 .
1 1.0 0.5 2 1.5 1.0 2
/ / /
Hence, g (1)= f (2) f (1) (2)(0.6)=1.2 .

x / / x d x / x x
61. (a) F(x)= f (e ) F (x)= f (e ) (e )= f (e )e
dx
f (x) / f (x) d f (x) /
(b) G(x)=e  G (x)=e f (x)=e f (x)
dx

 / /  d  /    1
62. (a) F(x)= f (x ) F (x)= f (x ) (x )= f (x ) x
dx
 /   1 /
(b) G(x)= f (x)  G (x)= f (x) f (x)

4 / / 4 3 4 3
63. (a) f (x)=L(x ) f (x)=L (x ) 4x =(1/x ) 4x =4/x for x>0 .
/ /
(b) g(x)=L(4x) g (x)=L (4x) 4=(1/(4x)) 4=1/x for x>0 .
4 / 3 / 3 3
(c) F(x)=[L(x)]  F (x)=4[L(x)]  L (x)=4[L(x)]  (1/x)=4[L(x)] /x
/ / 2 2 2
(d) G(x)=L(1/x) G (x)=L (1/x) ( 1/x )=(1/(1/x)) ( 1/x )=x ( 1/x )= 1/x for x>0 .

/ / / /
64. r(x)= f (g(h(x))) r (x)= f (g(h(x))) g (h(x)) h (x) , so
/ / / / / / /
r (1)= f (g(h(1))) g (h(1)) h (1)= f (g(2)) g (2) 4= f (3) 5 4=6 5 4=120

1
65. s(t)=10+ sin (10 t) the velocity after t seconds is
4
10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

/ 1 5
v(t)=s (t)= cos (10 t)(10 )= cos (10 t) cm / s.
4 2

/
66. (a) s=Acos ( t+ ) velocity =s =  Asin ( t+ ) .
/ n  
(b) If A
0 and 
0 , then s =0 sin ( t+ )=0  t+ =n  t= , n an integer.


2 t dB 2 t 2 0.7 2 t 7 2 t
67. (a) B(t)=4.0+0.35sin  = 0.35cos = cos = cos
5.4 dt 5.4 5.4 5.4 5.4 54 5.4
dB 7 2
(b) At t=1 , = cos 0.16 .
dt 54 5.4

2 / 2 2
68. L(t)=12+2.8sin (t 80)  L (t)=2.8cos (t 80) .
365 365 365
/ /
On March 21, t=80 , and L (80) 0.0482 hours per day. On May 21, t=141 , and L (141) 0.02398 ,
/
which is approximately one half of L (80) .
 1.5t
69. s(t)=2e sin 2 t
/  1.5t  1.5t  1.5t
v(t)=s (t)=2 e (cos 2 t)(2 )+(sin 2 t)e ( 1.5) =2e (2 cos 2 t 1.5sin 2 t)

1 1  kt
70. (a) lim p(t)=lim = =1 , since k>0  kt    e  0.
 kt 1+a 0
t  t  1+ae
 kt
 kt  1 dp  kt  2  kt kae
(b) p(t)=(1+ae )  = (1+ae ) ( kae )=
dt  kt 2
(1+ae )
(c)

11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

 0.5t  1
From the graph of p(t)=(1+10e ) , it seems that p(t)=0.8 (indicating that 80% of the population
has heard the rumor) when t 7.4 hours.

t
71. (a) Using a calculator or CAS, we obtain the model Q=ab with a=100.0124369 and
t t lnb
b=0.000045145933 . We can change this model to one with base e and exponent ln b [ b =e from
tln b  10.005531t
precalculus mathematics or from Section 7.3]: Q=ae =100.012437e .
/ t x
(b) Use Q (t)=ab ln b or the calculator command nDeriv(Y , X, .04) with Y =ab to get
1 1
/
Q (0.04)  670.63 A. The result of Example 2 in Section 2.1 was  670 A.
t  20
72. (a) P=ab with a=4.502714 10 and b=1.029953851 ,
where P is measured in thousands of people. The fit appears to be very good.

5308 3929 7240 5308

(b) For 1800: m = =137.9 , m = =193.2 .
1 1800 1790 2 1810 1800
/
So P (1800) (m +m )/2=165.55 thousand people / year.
1 2
23,192 17,063 31,443 23,192
For 1850: m = =612.9 , m = =825.1 .
1 1850 1840 2 1860 1850
/
So P (1850) (m +m )/2=719 thousand people / year.
1 2
x
(c) Use the calculator command nDeriv(Y , X, .04) with Y =ab to get
1 1
/ /
P (1800) 156.85 and P (1850) 686.07. These estimates are somewhat less than the ones in part
(b).
(d) P(1870) 41,946.56. The difference of 3.4 million people is most likely due to the Civil War
(1861  1865).

12
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

8
/ 45(t 2)
73. (a) Derive gives g (t)= without simplifying. With either Maple or Mathematica, we first
10
(2t+1)
8 9
/ (t 2) (t 2)
get g (t)=9  18 , and the simplification command results in the above expression.
9 10
(2t+1) (2t+1)
/ 3 3 4 3 2
(b) Derive gives y =2(x  x+1) (2x+1) (17x +6x  9x+3) without simplifying.
/ 4 3 4 5 3 3 2
With either Maple or Mathematica, we first get y =10(2x+1) (x  x+1) +4(2x+1) (x  x+1) (3x  1) .
If we use Mathematica’s Factor or Simplify , or Maple’s factor , we get the above expression, but
Maple’s simplify gives the polynomial expansion instead. For locating horizontal tangents, the
factored form is the most helpful.

4
4 x  x+1
(3x  1) 4
4 1/2
x  x+1 / x +x+1
74. (a) f (x)= 4
. Derive gives f (x)= 4 4
whereas either Maple or
x +x+1 (x +x+1)(x  x+1)
4
/ 3x  1
Mathematica give f ( x ) = after simplification.
4
x  x+1 4 2
4
(x +x+1)
x +x+1
/ 4 4 1
(b) f (x)=0 3x  1=0 x=  0.7598 .
3
/ /
(c) f (x)=0 where f has horizontal tangents. f has two maxima and one minimum where f has
inflection points.

75. (a) If f is even, then f (x)= f ( x) . Using the Chain Rule to differentiate this equation, we get
/ / d / / / /
f (x)= f ( x) ( x)= f ( x) . Thus, f ( x)= f (x) , so f is odd.
dx
/ / /
(b) If f is odd, then f (x)= f ( x) . Differentiating this equation, we get f (x)= f ( x)( 1)= f ( x) ,
/
so f is even.

13
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

76.
f (x) /

g(x) = { f (x) g(x)  1} /= f /(x) g(x)  1+( 1) g(x)  2g /(x) f (x)

/ / / /
f (x) f (x)g (x) f (x)g(x) f (x)g (x)
=  =
g(x) 2 2
g(x) g(x)

77. (a)
d
dx
( n
)
sin xcos nx =nsin n 1 xcos xcos nx+sin nx( nsin nx) [Product Rule]
n 1 n 1
=nsin x(cos nxcos x sin nxsin x) [factor out nsin x]
n 1 [Addition Formula for cosine]
=nsin xcos (nx+x)
n 1 [factor out x ]
=nsin xcos (n+1)x

(b)
d
dx
(
n
)
cos xcos nx =ncos n 1 x( sin x)cos nx+cos nx( nsin nx) [Product Rule]
n 1 n 1
= ncos x(cos nxsin x+sin nxcos x) [factor out ncos x]
n 1 [Addition Formula for sine]
= ncos xsin (nx+x)
n 1 [factor out x ]
= ncos xsin (n+1)x

5
78. ‘‘The rate of change of y with respect to x is eighty times the rate of change of y with respect to
d 5 dy 4 dy dy 4 dy
x ’’  y =80  5y =80  5y =80 (Note that
0 since the curve never has a
dx dx dx dx dx
4
horizontal tangent)  y =16 y=2 (since y>0 for all x )

 
79. Since  =  rad, we have
180
   
d
d
( sin   ) =
d
d
sin
180
 =
180
cos
180
 =
180
cos 

.

2 2 1/2 / 1 2 1/2 x x
80. (a) f (x)=|x|= x =(x )  f (x)= (x ) (2x)= = for x
0 .
2 2 |x|
x
f is not differentiable at x=0 .
14
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.5 The Chain Rule

2
(b) f (x)=|sin x|= sin x 
/ 1 2  1/2
f (x)= (sin x)  2sin x cos x=
2
sin x
|sin x|
cos x= { cos x if sin x>0
 cos x if sin x<0

f is not differentiable when x=n , n an integer.

(c) g(x)=sin |x|=sin x 
2 /
g (x)=cos |x|
x x
= cos x=
|x| |x| { cos x if x>0
 cos x if x<0

g is not differentiable at 0 .

81. First note that products and differences of polynomials are polynomials and that the derivative of
a polynomial is also a polynomial. When n=1,
(1) P(x) / Q(x)P /(x) P(x)Q /(x) A (x)
1 / /
f (x)= = = , where A (x)=Q(x)P (x) P(x)Q (x).
Q(x) 2 1+1 1
[Q(x)] [Q(x)]
(k)
A (x)
k
Suppose the result is true for n=k, where k 1. Then f (x)= , so
k+1
[Q(x)]
/ k+1 / k /
(k+1)
A (x) [Q(x)] A (x) A (x) (k+1)[Q(x)]  Q (x)
k k k
f (x) = =
k+1 k+1 2
[Q(x)] {[Q(x)] }
k+1 / k /
[Q(x)] A (x) (k+1)A (x)[Q(x)] Q (x)
k k
=
2k+2
[Q(x)]
k1 / / / /
[Q(x)] { A (x) (k+1)A (x)Q (x)} Q(x)A (x) (k+1)A (x)Q (x)
k k k k
= =
k k+2 k+2
[Q(x)] [Q(x)] [Q(x)]
k+2 / /
=A (x)/[Q(x)] , where A (x)=Q(x)A (x) (k+1)A (x)Q (x) .
k+1 k+1 k k
We have shown that the formula holds for n=1, and that when it holds for n=k it also holds for
n=k+1. Thus, by mathematical induction, the formula holds for all positive integers n.

15
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

d 2 d / / /  y 2 6x
1. (a) (xy+2x+3x )= (4) (x y +y 1)+2+6x=0 xy = y 2 6x y = or
dx dx x
/ y+2
y = 6 .
x
2
2 4 2x 3x 4 2 / 4
(b) xy+2x+3x =4 xy=4 2x 3x  y= =  2 3x , so y =  3.
x x 2
x
/  y 2 6x  (4/x 2 3x) 2 6x  4/x 3x 4
(c) From part (a), y = = = =  3.
x x x 2
x

d 2 2 d / / 8x 4x
2. (a) (4x +9y )= (36) 8x+18y y =0 y = =
dx dx 18y 9y
2 2 2 2 2 4 2 2 2
(b) 4x +9y =36 9y =36 4x  y= (9 x ) y= 9 x , so
9 3
/ 2 1 2 1/2 2x
y =  (9 x ) ( 2x)=
3 2 2
3 9 x
/ 4x 4x 2x
(c) From part (a), y = = = .
9y 2 2 2
9  9 x 3 9 x
3

2
d 1 1 d 1 1 / 1 / 1 / y
3. (a) + = (1)   y =0  y =  y =
dx x y dx 2 2 2 2 2
x y y x x
1 1 1 1 x 1 x / (x 1)(1) (x)(1)  1
(b) + =1 =1 =  y= , so y = = .
x y y x x x 1 2 2
(x 1) (x 1)
2 2 2
/ y [x/(x 1)] x 1
(c) y = = = =
2 2 2 2 2
x x x (x 1) (x 1)

d d 1 1 / / y
4. (a) ( x + y )= (4) + y =0 y =
dx dx 2 x 2 y x
2 / 4
(b) y =4 x y=(4 x ) =16 8 x +x y = +1
x
/ y 4 x 4
(c) y = = = +1
x x x

d 2 2 d / / / x
5. (x +y )= (1) 2x+2yy =0 2yy = 2x y =
dx dx y
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

d 2 2 d / / / x
6. (x  y )= (1) 2x 2yy =0 2x=2yy  y =
dx dx y

d 3 2 2 d 2 2 / / 2 / / 2
7. (x +x y+4y )= (6) 3x +(x y +y 2x)+8yy =0 x y +8yy = 3x  2xy
dx dx
2
2 / 2 / 3x +2xy x(3x+2y)
(x +8y) y = 3x  2xy y = 2
= 2
x +8y x +8y

d 2 3 d / 2 / / 2 / / 2
8. (x  2xy+y )= (c) 2x 2(xy +y 1)+3y y =0 2x 2y=2xy  3y y  2x 2y=y (2x 3y )
dx dx
/ 2x 2y
y =
2
2x 3y

d 2 2 d 2 / / 2 2 / / 2
9. (x y+xy )= (3x) (x y +y 2x)+(x 2yy +y  1)=3 x y +2xyy =3 2xy y 
dx dx
2
/ 2 2 / 3 2xy y
y (x +2xy)=3 2xy y  y = 2
x +2xy

d 5 2 3 d 4 4 / 2 2 / 3 4 / 3
10. (y +x y )= (1+x y) 5y y +x  3y y +y  2x=0+x y +y 4x 
dx dx
3 3
/ 4 2 2 4 3 3 / 4x y 2xy
y (5y +3x y  x )=4x y 2xy  y =
4 2 2 4
5y +3x y  x

d 2 2 d 2 / 2 /
11. (x y +xsin y)= (4)  x  2yy +y  2x+xcos y y +sin y 1=0
dx dx
2
2 / / 2 2 / 2 /  2xy  sin y
2x yy +xcos y y = 2xy  sin y (2x y+xcos y)y = 2xy  sin y y = 2
2x y+xcos y

d d 2 2 / 2 2 / 2 2
12. (1+x)= [sin (xy )] 1=[cos (xy )](x 2yy +y  1) 1=2xycos (xy )y +y cos (xy )
dx dx
2 2
2 2 2 / / 1 y cos (xy )
1 y cos (xy )=2xycos (xy )y  y = 2
2xycos (xy )

d d /
13. (4cos xsin y)= (1) 4[cos x cos y y +sin y ( sin x)]=0
dx dx

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

/ / 4sin xsin y
y (4cos xcos y)=4sin xsin y  y = =tan xtan y
4cos xcos y

d 2 d 2 2 2 / 2 / 2
14. [ysin (x )]= [xsin (y )] ycos (x ) 2x+sin (x ) y =xcos (y ) 2yy +sin (y ) 1
dx dx
2 2
/ 2 2 2 2 / sin (y ) 2xycos (x )
y [sin (x ) 2xycos (y )]=sin (y ) 2xycos (x ) y = 2 2
sin (x ) 2xycos (y )

15.
d
dx
e
x
( ) 2
y
=
d
dx
(x+y)
2
x y 2
e (x y +y 2x)=1+y 
/ /
xe
2 x y
2
/
y +2xye
2
x y
=1+y 
/

2
2 2 2 2 x y
2 x y / / x y / 2 x y x y / 1 2xye
xe y  y =1 2xye  y (x e  1)=1 2xye  y = 2
2 x y
xe  1

d d 2 2 1  1/2 / 2 / 2
16. ( x+y )= (1+x y ) (x+y) (1+y )=x  2yy +y  2x
dx dx 2
/
1 y 2 / 2 / 2 / 2
+ =2x yy +2xy  1+y =4x y x+y y +4xy x+y 
2 x+y 2 x+y
2
/ 2 / 2 / 2 2 / 4xy x+y  1
y  4x y x+y y =4xy x+y  1 y (1 4x y x+y )=4xy x+y  1 y = 2
1 4x y x+y

2 1  1/2 / 2 / x / y 2 /
17. xy =1+x y (xy) (xy +y 1)=0+x y +y 2x y + =x y +2xy
2 2 xy 2 xy
2
/ x 2 y / x 2x xy 4xy xy  y / 4xy xy  y
y  x =2xy  y =  y = 2
2 xy 2 xy 2 xy 2 xy x 2x xy

y 2 2 2 / /
18. tan (x y)=  (1+x )tan (x y)=y (1+x )sec (x y) (1 y )+tan (x y) 2x=y 
2
1+x
2 2 2 2 / /
(1+x )sec (x y) (1+x )sec (x y) y +2xtan (x y)=y 
2 2
2 2 2 2 / / (1+x )sec (x y)+2xtan (x y)
(1+x )sec (x y)+2xtan (x y)=[1+(1+x )]sec (x y) y  y = 2 2
1+(1+x )sec (x y)

/ 2 / / 2 / /
19. xy=cot (xy) y+xy = csc (xy)(y+xy ) (y+xy )[1+csc (xy)]=0 y+xy =0 y = y/x

20. sin x+cos y=sin xcos y

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

/ / /
cos x sin y y =sin x( sin y y )+cos ycos x (sin xsin y sin y) y =cos xcos y cos x
/ cos x(cos y 1)
y =
sin y(sin x 1)

d 2 3 d / 2 2 / 3
21. {1+ f (x)+x [ f (x)] }= (0) f (x)+x  3[ f (x)]  f (x)+[ f (x)]  2x=0 . If x=1 , we have
dx dx
/ 2 2 / 3 / 2 / 3 / /
f (1)+1  3[ f (1)]  f (1)+[ f (1)]  2(1)=0 f (1)+1 3 2  f (1)+2  2=0 f (1)+12 f (1)= 16
/ / 16
13 f (1)= 16 f (1)= .
13

d d 2 / /
22. = (x ) g (x)+xcos g(x) g (x)+sin g(x) 1=2x . If x=1 , we have
dx dx
/ / / / / / /
g (1)+1cos g(1) g (1)+sin g(1)=2(1) g (1)+cos 0 g (1)+sin 0=2 g (1)+g (1)=2 2g (1)=2
/
g (1)=1 .

4 2 2 4 3 2 2 dx 3 dx 4 2
23. y +x y +yx =y+1 4y + x  2y+y  2x + y 4x +x  1 =1 2xy
dy dy
3 2 4
dx 3 dx 3 2 4 dx 1 4y  2x y x
+4x y =1 4y  2x y x  =
dy dy dy 2 3
2xy +4x y

2 2 2
2 22 2 2 2 dx dx 2 dx ax  4y(x +y )
24. (x +y ) =ax y 2(x +y ) 2x +2y =2ayx +ax  =
dy dy dy 2 2
4x(x +y ) 2axy

2 2 / / / / / /  2x y
25. x +xy+y =3 2x+xy +y 1+2yy =0 xy +2yy = 2x y y (x+2y)= 2x y y = . When
x+2y
/  2 1  3
x=1 and y=1 , we have y = = = 1 , so an equation of the tangent line is y 1= 1(x 1) or
1+2 1 3
y= x+2 .

2 2 / / / / /
26. x +2xy y +x=2 2x+2(xy +y 1) 2yy +1=0 2xy  2yy = 2x 2y 1 y (2x 2y)= 2x 2y 1
/  2x 2y 1 /  2 4 1  7 7
y = . When x=1 and y=2 , we have y = = = , so an equation of the tangent
2x 2y 2 4  2 2
7 7 3
line is y 2= (x 1) or y= x .
2 2 2

2 2 2 2 2 / 2 2 / 1
27. x +y =(2x +2y  x)  2x+2yy =2(2x +2y  x)(4x+4yy  1) . When x=0 and y= , we have
2
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

/ 1 / / / / 1
0+y =2 (2y  1) y =2y  1 y =1 , so an equation of the tangent line is y =1(x 0) or
2 2
1
y=x+ .
2

/ 3
2/3 2/3 2  1/3 2  1/3 / 1 y / y
28. x +y =4 x + y y =0 + =0 y = . When x= 3 3 and y=1 , we
3 3 3
x
3
y
3
x
2/3
/ 1 ( 3 3 ) 3 1
have y = = = = , so an equation of the tangent line is
( 3 3 )
1/3  3 3 3 3 3
1 1
y 1= (x+3 3 ) or y= x+4 .
3 3

2 22 2 2 2 2 / /
29. 2(x +y ) =25(x  y ) 4(x +y )(2x+2yy )=25(2x 2yy )
2 2
/ 2 2 / / 2 2 / 2 2 / 25x 4x(x +y )
4(x+yy )(x +y )=25(x yy )  4yy (x +y )+25yy =25x 4x(x +y ) y = 2 2
. When
25y+4y(x +y )
75 120 45 9 /
x=3 and y=1 , we have y = = = , so an equation of the tangent line is
25+40 65 13
9 9 40
y 1= (x 3) or y= x+ .
13 13 13
2 2 2 2 4 2 4 2 3 / / 3
30. y (y  4)=x (x  5) y  4y =x  5x  4y y  8yy =4x  10x . When x=0 and y= 2 , we have
/ / / /
 32y +16y =0  16y =0 y =0 , so an equation of the tangent line is y+2=0(x 0) or y= 2 .

3
2 4 2 / 3 10x  x
/
31. (a) y =5x  x  2yy =5(4x ) 2x y = . So at the point (1,2) we have
y
3
/10(1)  1 9 9 9 5
y = = , and an equation of the tangent line is y 2= (x 1) or y= x .
2 2 2 2 2

(b)

2
2 3 2 / 2 3x +6x
/
32. (a) y =x +3x  2yy =3x +3(2x) y = . So at the point (1, 2) we have
2y

5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

2
/3(1) +6(1) 9 9 9 1
y = = , and an equation of the tangent line is y+2= (x 1) or y= x+ .
2( 2) 4 4 4 4
/ 2
(b) The curve has a horizontal tangent where y =0 3x +6x=0 3x(x+2)=0 x=0 or x= 2 . But note
2 3 2
that at x=0 , y=0 also, so the derivative does not exist. At x= 2 , y =( 2) +3( 2) = 8+12=4 , so y= 2
. So the two points at which the curve has a horizontal tangent are ( 2, 2) and ( 2,2) .

(c)

33. (a)
There are eight points with horizontal tangents: four at x 1.57735 and four at x 0.42265 .
2
/ 3x  6x+2 / / 1
(b) y =  y = 1 at (0, 1) and y = at (0, 2)
3 2 3
2(2y  3y  y+1)
1
Equations of the tangent lines are y= x+1 and y= x+2 .
3
/ 2 1
(c) y =0 3x  6x+2=0 x=1 3
3
(d)
By multiplying the right side of the equation by x 3 , we obtain the
first graph.
By modifying the equation in other ways, we can generate the other
graphs.

2
y(y  1)(y 2)=x(x 1)(x 2)(x 3)

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

2
y(y  4)(y 2)=x(x 1)(x 2)

2
y(y+1)(y  1)(y 2)=(x 1)(x 2)

2
(y+1)(y  1)(y 2)=x(x 1)(x 2)

2
x(y+1)(y  1)(y 2)=y(x 1)(x 2)

2 2
y(y +1)(y 2)=x(x  1)(x 2)

2 2
y(y+1)(y  2)=x(x 1)(x  2)

34. (a)

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

1
(b) There are 9 points with horizontal tangents: 3 at x=0 , 3 at x= , and 3 at x=1 . The three
2
horizontal tangents along the top of the wagon are hard to find, but by limiting the y  range of the
graph (to [1.6,1.7] , for example) they are distinguishable.

/ 2 2
35. From Exercise 29, a tangent to the lemniscate will be horizontal if y =0 25x 4x(x +y )=0
2 2 2 2 25
x[25 4(x +y )]=0  x +y = ( 1 ). (Note that when x is 0 , y is also 0 , and there is no horizontal
4
25 2 2
tangent at the origin.) Substituting for x +y in the equation of the lemniscate,
4
2 22 2 2 2 2 25 2 75 2 25
2(x +y ) =25(x  y ) , we get x  y = ( 2 ). Solving ( 1 ) and ( 2 ), we have x = and y = , so
8 16 16
5 3 5
the four points are  , .
4 4
2 2 / 2
x y 2x 2yy / b x
36. + =1 + =0 y = 2
 an equation of the tangent line at (x ,y ) is
2 2 2 2 0 0
a b a b a y
2 2 2
 b x y y y y x x x
0 0 0 0 0 0
y y = (x x ) . Multiplying both sides by gives  = + . Since (x ,y ) lies on
0 2 0 2 2 2 2 2 0 0
a y b b b a a
0
2 2
x x y y x y
0 0 0 0
the ellipse, we have + = + =1 .
2 2 2 2
a b a b

2 2 / 2
x y 2x 2yy / b x
37.  =1  =0 y = 2
 an equation of the tangent line at (x ,y ) is
2 2 2 2 0 0
a b a b a y
2 2 2
b x y y y y x x x
0 0 0 0 0 0
y y = (x x ) . Multiplying both sides by gives  =  . Since (x ,y ) lies on
0 2 0 2 2 2 2 2 0 0
a y b b b a a
0
the hyperbola, we have

8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

2 2
x x y y x y
0 0 0 0
 =  =1 .
2 2 2 2
a b a b

/
y 1 / y
38. x + y = c  + =0 y =  an equation of the tangent line at (x ,y ) is
2 x 2 y x 0 0

y y
0 0
y y = (x x ) . Now x=0 y=y  ( x )=y + x y , so the y  intercept is
0 0 0 0 0 0 0
x x
0 0

y y x
0 0 0
y+ x y . And y=0  y = (x x ) x x =  x=x + x y , so the x  intercept
0 0 0 0 0 0 0 0 0
x y
0 0

is x + x y . The sum of the intercepts is

0 0 0
2 2
y+ x y + x+ x y =x +2 x y +y = x + y =( c ) =c .
0 0 0 0 0 0 0 0 0 0 0 0

x 2 2 2 / /
39. If the circle has radius r , its equation is x +y =r  , so the slope of the 2x+2yy =0 y =
y
x y
0  1 0
tangent line at P(x ,y ) is  . The negative reciprocal of that slope is = , which is the
0 0 y  x /y x
0 0 0 0
slope of OP , so the tangent line at P is perpendicular to the radius OP .

p 1 p 1 p 1 p/q
q p q 1 / p 1 / px px y px x p ( p/q )  1
40. y =x  qy y = px  y = = = = x
q 1 q p q
qy qy qx

 1 / 1 d 1 1  1/2 1
41. y=tan x y =  ( x )= x =
2 dx 1+x 2 2 x (1+x)
1+( x )

 1  1 1/2
42. y= tan x =(tan x) 
/ 1  1  1/2 d  1 1 1 1
y = (tan x)  (tan x)=  =
2 dx  1 2  1 2
2 tan x 1+x 2 tan x (1+x )

 1
43. y=sin (2x+1)

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

/ 1 d 1 2 1
y =  (2x+1)=  2= =
2 dx 2 2 2
1 (2x+1) 1 (4x +4x+1)  4x  4x  x x

2 / 2 1 1 2 1/2 xarcsinx
44. h(x)= 1 x arcsinx h (x)= 1 x  +arcsinx (1 x ) ( 2x) =1
2 2 2
1 x 1 x

2 / 2 1
45. H(x)=(1+x )arctanx H (x)=(1+x ) +(arctanx)(2x)=1+2xarctanx
2
1+x

46. y=tan
 1
( x 2
x +1  )
2
/ 1 x 1 x +1  x
y = 2
1 =
1+ x( 2
x +1 ) x +1
2 2
1+x  2x x +1 +x +1
2 2
x +1
2

2 2
x +1  x x +1  x
= =
(
2 1+x  x x +1
2 2
) 2
x +1 2
2
x +1 (1+x ) x(x +1)
2 2

2
x +1  x 1
= =
2 (1+x )
2
( 2
x +1  x ) 2(1+x )
2

 1  1
47. h(t)=cot (t)+cot (1/t)
2
/ 1 d 1 1 t1 1 1 1
h (t)=   =  2   = + 2 =0 .
2 2 dt t 2 2 2
1+t 1+(1/t) 1+t t +1 t 1+t t +1
 
Note that this makes sense because h(t)= for t>0 and h(t)= for t<0 .
2 2

 1 2 /  1 x x  1
48. y=xcos x 1 x  y =cos x + =cos x
2 2
1 x 1 x

2x
 1 2x / 1 d 2x 2e
49. y=cos (e ) y =  (e )=
2x 2 dx 4x
1 (e ) 1 e

50. y=arctan(cos  )
10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

/ 1 sin 
y = ( sin  )= 2
2
1+(cos  ) 1+cos 

x 2
51. f (x)=e  x arctanx
x 2 1
/ =e  x +(arctanx)(2x)
f (x) 2
1+x
2
x x
=e   2xarctanx
2
1+x

/ /
This is reasonable because the graphs show that f is increasing when f is positive, and f is zero
when f has a minimum.

2
52. f (x)=xarcsin(1 x )
 2x 2
/ =x +arcsin(1 x ) 1
f (x) 22
1 (1 x )
2
2 2x
=arcsin(1 x )
2 4
2x  x

/
This is reasonable because the graphs show that f is increasing when f is positive, and that f has an
/
inflection point when f changes from increasing to decreasing.

 1 dy
53. Let y=cos x . Then cos y=x and 0
y
   sin y =1
dx
dy 1 1 1
= = = . (Note that sin y 0 for 0
y
 ).
dx sin y 2 2
1 cos y 1 x
11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

 1  3
54. (a) Let y=sec x . Then sec y=x and y 0,  , . Differentiate with respect to x:
2 2
dy dy 1 1 1 2 2
sec ytan y =1 = = = . Note that tan y=sec y 1
dx dx sec ytan y 2 2
sec y sec y 1 x x  1
2  3
tan y= sec y 1 since tan y>0 when 0<y< or  <y< .
2 2
 1 dy dy 1 2 2 2
(b) y=sec x sec y=x sec ytan y =1 = . Now tan y=sec y 1=x  1 , so
dx dx sec ytan y
2 
tan y= x  1 . For y 0, , x 1 , so sec y=x=| x| and tan y 0
2
dy 1 1  2
= = . For y , , x
 1, so | x|= x and tan y= x  1 
dx 2 2 2
x x  1 | x| x  1
dy 1 1 1 1
= = = = .
dx sec ytan y
x  x 1
2
( 2
)
( x) x  1 | x| x  1
2

2 2 2 2 3 3
55. 2x +y =3 and x=y intersect when 2x +x 3=0  (2x+3)(x 1)=0 x= or 1 , but  is
2 2
2 2
extraneous since x=y is nonnegative. When x=1 , 1=y  y= 1 , so there are two points of
2 2 / / 2 / /
intersection: (1, 1) . 2x +y =3 y = 2x/y , and x=y  1=2yy  y =1/(2y) . At (1,1)
4x+2yy =0
1
, the slopes are m = 2(1)/1= 2 and m =1/(2 1)= , so the curves are orthogonal (since m and m
1 2 2 1 2
are negative reciprocals of each other). By symmetry, the curves are also orthogonal at (1, 1) .

2 2 2 2 2 2 2
56. x  y =5 and 4x +9y =72 intersect when 4x +9(x  5)=72 13x =117 x= 3 , so there are four
2 2 / / 2 2 /
points of intersection: ( 3, 2) . x  y =5 2x 2yy =0 y =x/y , and 4x +9y =72 8x+18yy =0
/ 3 2
y = 4x/9y . At (3,2) , the slopes are m = and m = , so the curves are orthogonal. By symmetry,
1 2 2 3
the curves are also orthogonal at (3, 2) , ( 3,2) and ( 3, 2) .

57.
12
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

58. The orthogonal family represents the direction of the wind.

2 2 2 2 2 2 /
59. x +y =r is a circle with center O and ax+by=0 is a line through O . x +y =r  2x+2yy =0
/
y = x/y , so the slope of the tangent line at P (x ,y ) is  x /y . The slope of the line OP is y /x ,
0 0 0 0 0 0 0 0
which is the negative reciprocal of  x /y . Hence, the curves are orthogonal, and the families
0 0
of curves are orthogonal trajectories of each other.

2 2 2 2
60. The circles x +y =ax and x +y =by intersect at the origin where the tangents are vertical and
2 2 2 2
horizontal. If (x ,y ) is the other point of intersection, then x +y =ax ( 1 ) and x +y =by ( 2 ). Now
0 0 0 0 0 0 0 0
2 2 / / a 2x 2 2 / / / 2x
x +y =ax 2x+2yy =a y = and x +y =by  2x+2yy =by  y = . Thus, the curves
2y b 2y
are orthogonal at
a 2x b 2y 2 2 2 2
0 0
(x ,y ) =  2ax  4x =4y  2by  ax +by =2(x +y ) , which is true by ( 1 ) and ( 2 ).
0 0 2y 2x 0 0 0 0 0 0 0 0
0 0

2 / 2 2 / / / x x 1
61. y=cx  y =2cx and x +2y =k 2x+4yy =0 2yy = x  y = = = , so the
2(y) 2 2cx
2(cx )
curves are orthogonal.

13
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

3 / 2 2 2 / / / x x 1
62. y=ax  y =3ax and x +3y =b 2x+6yy =0 3yy = x  y = = = , so
3(y) 3 2
3(ax ) 3ax
the curves are orthogonal.

2 2
63. To find the points at which the ellipse x  xy+y =3 crosses the x  axis, let y=0 and solve for x .
2 2
y=0 x  x(0)+0 =3 x= 3 . So the graph of the ellipse crosses the x  axis at the points ( 3,0) .
/ / / / / y 2x
Using implicit differentiation to find y , we get 2x xy  y+2yy =0 y (2y x)=y 2x y = .
2y x
/ 0 2 3 / 0+2 3
So y at ( 3,0) is =2 and y at ( 3,0) is =2 . Thus, the tangent lines at these
2(0) 3 2(0)+ 3
points are parallel.

y 2x /
64. (a) We use implicit differentiation to find y = as in Exercise 49. The slope of the tangent
2y x
1 2( 1) 3 1
line at ( 1,1) is m= = =1 , so the slope of the normal line is  = 1 , and its equation is
2(1) ( 1) 3 m
2 2
y 1= 1(x+1) y= x . Substituting  x for y in the equation of the ellipse, we get x  x( x)+( x) =3
2
3x =3 x= 1 . So the normal line must intersect the ellipse again at x=1 , and since the equation of
the line is y= x , the other point of intersection must be (1, 1) .

(b)

2 2 2 / 2 / / 2 2
65. x y +xy=2 x  2yy +y  2x+x y +y 1=0 y (2x y+x)= 2xy  y
14
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

2 2
/ 2xy +y 2xy +y 2 2
y = 2
. So  2
= 1 2xy +y=2x y+x y(2xy+1)=x(2xy+1)
2x y+x 2x y+x
1 1 2 2 1 1
y(2xy+1) x(2xy+1)=0  (2xy+1)(y x)=0 xy= or y=x . But xy=  x y +xy=   2 , so we
2 2 4 2
2 2 4 2 4 2 2 2 2
must have x=y . Then x y +xy=2 x +x =2 x +x  2=0 (x +2)(x  1)=0 . So x = 2 , which is
2
impossible, or x =1 x= 1 . Since x=y , the points on the curve where the tangent line has a slope of
 1 are ( 1, 1) and (1,1) .

2 2 x / / 2 2
66. x +4y =36 2x+8yy =0 y =
. Let (a,b) be a point on x +4y =36 whose tangent line
4y
a a
passes through (12,3) . The tangent line is then y 3= (x 12) , so b 3= (a 12) . Multiplying
4b 4b
2 2 2 2 2 2
both sides by 4b gives 4b  12b= a +12a , so 4b +a =12(a+b) . But 4b +a =36 , so 36=12(a+b)
2 2 2 2 2 2
a+b=3 b=3 a . Substituting 3 a for b into a +4b =36 gives a +4(3 a) =36 a +36 24a+4a =36
2 24 24 24 9
5a  24a=0 a(5a 24)=0 , so a=0 or a= . If a=0 , b=3 0=3 , and if a= , b=3 = . So
5 5 5 5
24 9
the two points on the ellipse are (0,3) and , .
5 5
a
Using y 3= (x 12) with (a,b)=(0,3) gives us the tangent line y 3=0 or y=3 . With
4b
24 9 24/5 2 2
(a,b)= , , we have y 3= (x 12) y 3= (x 12) y= x 5 . A graph of the
5 5 4( 9/5) 3 3
ellipse and the tangent lines confirms our results.

 1
67. (a) If y= f (x) , then f (y)=x . Differentiating implicitly with respect to x and remembering that y
/ dy dy 1  1 / 1
is a function of x, we get f (y) =1, so = /  ( f ) (x)= /  1 .
dx dx
f (y) f ( f (x))
 1  1 / /  1 / 2 3
(b) f (4)=5 f (5)=4. By part (a), ( f ) (5)=1/ f ( f (5))=1/ f (4)=1/ = .
3 2

/
68. (a) f (x)=2x+cos x f (x)=2 sin x>0 for all x . Thus, f is increasing for all x and is therefore
one to one.
15
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.6 Implicit Differentiation

 1
(b) Since f is one to one, f (1)=k f (k)=1 . By inspection, we see that f (0)=2(0)+cos 0=1 , so
 1
k= f (1)=0 .
 1 / /  1 / 1
(c) ( f ) (1)=1/ f ( f (1))=1/ f (0)=1/(2 sin 0)=
2

2 2 x
/ /
69. x +4y =5 2x+4(2yy )=0 y =
. Now let h be the height of the lamp, and let (a,b) be the
4y
point of tangency of the line passing through the points (3,h) and ( 5,0) . This line has slope
h 0 1 / a
= h . But the slope of the tangent line through the point (a,b) can be expressed as y = ,
3 ( 5) 8 4b
b 0 b a b 2 2 2 2 2 2
or as = , so  =  4b = a  5a a +4b = 5a . But a +4b =5 , so 5= 5a a= 1 .
a ( 5) a+5 4b a+5
2 2
Then 4b = a  5a= 1 5( 1)=4 b=1 , since the point is on the top half of the ellipse. So
h b 1 1
= = =  h=2 . So the lamp is located 2 units above the x  axis.
8 a+5  1+5 4

16
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

/ //
1. a= f , b= f , c= f . We can see this because where a has a horizontal tangent, b=0 , and where b
/
has a horizontal tangent, c=0 . We can immediately see that c can be neither f nor f , since at the
points where c has a horizontal tangent, neither a nor b is equal to 0 .

/
2. Where d has horizontal tangents, only c is 0 , so d =c . c has negative tangents for x<0 and b is the
/
only graph that is negative for x<0 , so c =b . b has positive tangents on R (except at x=0 ), and the
/ / //
only graph that is positive on the same domain is a , so b =a . We conclude that d= f , c= f , b= f
///
and a= f .

3. We can immediately see that a is the graph of the acceleration function, since at the points where a
has a horizontal tangent, neither c nor b is equal to 0 . Next, we note that a=0 at the point where b has
/
a horizontal tangent, so b must be the graph of the velocity function, and hence, b =a . We conclude
that c is the graph of the position function.

4. a must be the jerk since none of the graphs are 0 at its high and low points. a is 0 where b has a
/ /
maximum, so b =a . b is 0 where c has a maximum, so c =b . We conclude that d is the position
function, c is the velocity, b is the acceleration, and a is the jerk.

5 2 / 4 // 3
5. f (x)=x +6x  7x f (x)=5x +12x 7 f (x)=20x +12

8 6 4 / 7 5 3 // 6 4 2
6. f (t)=t  7t +2t  f (t)=8t  42t +8t  f (t)=56t  210t +24t
/ //
7. y=cos 2  y = 2sin 2  y = 4cos 2

/ //
8. y= sin   y = cos  +sin   y = ( sin  )+cos   1+cos  =2cos    sin 

6 / 5 5 // 4 4
9. F(t)=(1 7t)  F (t)=6(1 7t) ( 7)= 42(1 7t)  F (t)= 42 5(1 7t) ( 7)=1470(1 7t)
2x+1 / (x 1)(2) (2x+1)(1) 2x 2 2x 1  3  2
10. g(x)=  g (x)= = = or  3(x 1)
x 1 2 2 2
(x 1) (x 1) (x 1)
//  3  3 6
 g (x)= 3( 2)(x 1) =6(x 1) or
3
(x 1)

1 4u / (1+3u)( 4) (1 4u)(3)  4 12u 3+12u  7  2

11. h(u)=  h (u)= = = or  7(1+3u) 
1+3u 2 2 2
(1+3u) (1+3u) (1+3u)
//  3  3
h (u)= 7( 2)(1+3u) (3)=42(1+3u) or
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

42
3
(1+3u)

12.
b 1/2  1/2
H(s) =as +bs 
=a s +
s
/ 1  1/2 1  3/2 1  1/2 1  3/2
H (s) =a 2 s +b  2 s = as 
2 2
bs 

// 1 1  3/2 1 3  5/2 1  3/2 3  5/2

H (s) = 2 a  2 s 
2
b 
2
s =
4
as + bs
4

2 / 1 2  1/2 x
13. h(x)= x +1  h (x)= (x +1) (2x)= 
2 2
x +1
2 1 2  1/2
x +1  1 x (x +1) (2x) 2  1/2 2 2
// 2 (x +1) [(x +1) x ] 1
h (x)= = =
2 2 1 2 3/2

cx /
( cx
2
x +1 )
cx cx //
(x +1)
cx cx
(x +1)
cx cx
14. y=xe  y =x e  c+e  1=e (cx+1) y =e (c)+(cx+1)e  c=ce (1+cx+1)=ce (cx+2)

3 2/3 / 2 3  1/3 2 2 3  1/3

15. y=(x +1)  y = (x +1) (3x )=2x (x +1) 
3
// 2 1 3  4/3 2 3  1/3 3  1/3 4 3  4/3
y =2x  (x +1) (3x )+(x +1) (4x)=4x(x +1)  2x (x +1)
3

16.
4x
y = 
x+1
1  1/2
x+1 4 4x (x+1) 4 x+1  2x/ x+1 4(x+1) 2x
/ 2 2x+4
y = = = = 
2 x+1 3/2 3/2
( x+1 ) (x+1) (x+1)
3/2 3 1/2
(x+1)  2 (2x+4) (x+1) 1/2
// 2 (x+1) [2(x+1) 3(x+2)]
y = =
3/2 2 3
[(x+1) ] (x+1)
2x+2 3x 6  x 4
= =
5/2 5/2
(x+1) (x+1)

17. H(t)=tan 3t 
2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

/ 2 // d 2
H (t)=3sec 3t H (t)=2 3sec 3t (sec 3t)=6sec 3t (3sec 3t tan 3t)=18sec 3t tan 3t
dt

2 / 2
18. g(s)=s cos s g (s)=2s cos s s sin s
// 2 2
g (s)=2cos s 2s sin s 2s sin s s cos s=(2 s )cos s 4s sin s

3 5t / 3 5t 5t 2 2 5t
19. g(t)=t e  g (t)=t e  5+e  3t =t e (5t+3)
// 5t 2 5t 2 5t
g (t) =(2t)e (5t+3)+t (5e )(5t+3)+t e (5)
5t 5t 2
=te 2(5t+3)+5t(5t+3)+5t =te (25t +30t+6)

4 3 4
 1 2 / 1 2x // (1+x )(2) (2x)(4x ) 2 6x
20. h(x)=tan (x ) h (x)=  2x=  h (x)= =
22 4 42 42
1+(x ) 1+x (1+x ) (1+x )

2 /
21. (a) f (x)=2cos x+sin x f (x)=2( sin x)+2sin x(cos x)=sin 2x 2sin x
//
f (x)=2cos 2x 2cos x=2(cos 2x cos x)

(b)
/ /
We can see that our answers are plausible, since f has horizontal tangents where f (x)=0 , and f
//
has horizontal tangents where f (x)=0 .

x 3 / x 2 // x
22. (a) f (x)=e  x  f (x)=e  3x  f (x)=e  6x

(b)
/ /
The graphs seem reasonable since f has horizontal tangents where f is zero, f is positive where f
/ /
is increasing, and f is negative where f is decreasing; and the same relationships exist between f
//
and f .

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

1/2 / 1  1/2  1/2 // 1  3/2  3/2

23. y= 2x+3 =(2x+3)  y = (2x+3)  2=(2x+3)  y = (2x+3)  2= (2x+3) 
2 2
/// 3  5/2  5/2
y = (2x+3)  2=3(2x+3)
2

x / (2x 1)(1) x(2)  1  2

24. y=  y = = or  1(2x 1) 
2x 1 2 2
(2x 1) (2x 1)
//  3  3
y = 1( 2)(2x 1) (2)=4(2x 1) 
///  4  4 4
y =4( 3)(2x 1) (2)= 24(2x 1) or  24/(2x 1)

/ //
25. f (t)=tcos t  f (t)=t( sin t)+cos t 1 f (t)=t( cos t) sin t 1 sin t
/// ///
f (t)=tsin t cos t 1 cos t cos t=tsin t 3cos t , so f (0)=0 3= 3 .

/ 1  1/2  1/2 // 1  3/2  3/2

26. g(x)= 5 2x  g (x)= (5 2x) ( 2)= (5 2x)  g (x)= (5 2x) ( 2)= (5 2x) 
2 2
/// 3  5/2  5/2 ///  5/2
g (x)= (5 2x) ( 2)= 3(5 2x) , so g (2)= 3(1) = 3 .
2

/ 2 // 2
27. f ( )=cot   f ( )= csc   f ( )= 2csc ( csc  cot  )=2csc   cot  
/// 2 2 2 2 2 2
f ( )=2( 2csc   cot  )cot  +2csc  ( csc  )= 2csc  (2cot  +csc  )
///  2 2 2
f = 2(2) [2( 3 ) +(2) ]= 80
6

/
28. g(x)=sec x g (x)=sec x tan x
// 2 3 2 3 2 3
g (x)=sec x sec x+tan x(sec x tan x)=sec x+sec x tan x=sec x+sec x(sec x 1)=2sec x sec x
/// 2 2 /// 
g (x)=6sec x(sec x tan x) sec x tan x=sec x tan x(6sec x 1) g = 2 (1)(6 2 1)=11 2
4

2 2 / / /
29. 9x +y =9 18x+2yy =0 2yy = 18x y = 9x/y
/ 2 2
// y 1 x y y x( 9x/y) y +9x 9
y = 9 = 9 = 9 = 9 [ since x and y must satisfy the
2 2 3 3
y y y y
2 2 // 3
original equation, 9x +y =9] . Thus, y = 81/y .

30.

4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

/
1y / y
x + y =1 + =0 y = 
2 x 2 y x
1 / 1 1 y 1 y
// x y  y x   y 1+
y (2 y ) (2 x ) y x x x
= = =
x 2x 2x
x+ y 1
= = since x and y must satisfy the original equation, x + y =1 .
2x x 2x x

2
3 3 2 2 / / x
31. x +y =1 3x +3y y =0 y = 
2
y
2
2 2 x
2xy  2x y 
2 2 / 2 4 4 3 3
// y (2x) x  2yy y 2xy +2x y 2xy(y +x ) 2x
y = = = = = ,
22 4 6 6 5
(y ) y y y y
3 3
since x and y must satisfy the original equation, x +y =1 .

4 4 4 3 3 / 3 / 3 / 3 3
32. x +y =a  4x +4y y =0 4y y = 4x  y = x /y 
3
x
y x 
3 2 3 2 / 3 4 4 4 4 2
// y  3x  x  3y y 2 2 y 2 y +x 2 a  3a x
y = = 3x y  = 3x  = 3x  =
32 6 4 3 7 7
(y ) y y y y y

n / n 1 // n 2 ( n) n n
33. f (x)=x  f (x)=nx  f (x)=n(n 1)x      f (x)=n(n 1)(n 2)   2 1x =n!

1  1 /  2 //  3 2
34. f (x)= =(5x 1)  f (x)= 1(5x 1)  5 f (x)=( 1)( 2)(5x 1)  5 
5x 1
///  4 3 (n) n n  (n+1)
f (x)=( 1)( 2)( 3)(5x 1)  5      f (x)=( 1) n!5 (5x 1)

2x / 2x // 2x 2 2x
35. f (x)=e  f (x)=2e  f (x)=2 2e =2 e 
/// 2 2x 3 2x (n) n 2x
f (x)=2  2e =2 e      f (x)=2 e

1/2 / 1  1/2
36. f (x)= x =x  f (x)= x 
2
// 1 1  3/2 /// 1 1 3  5/2
f (x)=  x  f (x)=   x 
2 2 2 2 2
5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

( 4) 1 1 3 5  7/2 1 3 5  7/2
f (x)=    x = x 
2 2 2 2 4
2
( 5) 1 1 3 5 7  9/2 1 3 5 7  9/2
f (x)=     x = x     
2 2 2 2 2 5
2
( n) 1 1 3 1  ( 2n 1 ) /2 n 1 1 3 5     (2n 3)  ( 2n 1 ) /2
f (x)=       n+1 x =( 1) x
2 2 2 2 n
2

3 1 3 / 1  4 // 1  5
37. f (x)=1/(3x )= x  f (x)= ( 3)x  f (x)= ( 3)( 4)x 
3 3 3
/// 1  6
f (x)= ( 3)( 4)( 5)x     
3
n n
( n) 1  ( n+3 ) ( 1)  3 4 5     (n+2) 2 ( 1) (n+2)!
f (x)= ( 3)( 4)   [ (n+2)]x =  =
3 n+3 2 n+3
3x 6x

2 3 4
38. Dsin x=cos x D sin x= sin x D sin x= cos x D sin x=sin x . The derivatives of sin x occur in
74 2
a cycle of four. Since 74=4(18)+2 , we have D sin x=D sin x= sin x .

/ 2 2 // 3 3 ///
39. Let f (x)=cos x . Then Df (2x)=2 f (2x) , D f (2x)=2 f (2x) , D f (2x)=2 f (2x) , ... ,
(n) n (n)
D f (2x)=2 f (2x) . Since the derivatives of cos x occur in a cycle of four, and since 103=4(25)+3 ,
(103) (3) 103 103 (103) 103
we have f (x)= f (x)=sin x and D cos 2x=2 f (2x)=2 sin 2x .

 x /  x  x  x //  x  x  x
40. f (x)=xe  f (x)=x( e )+e =(1 x)e  f (x)=(1 x)( e )+e ( 1)=(x 2)e 
///  x  x  x (4)  x  x  x
f (x)=(x 2)( e )+e =(3 x)e  f (x)=(3 x)( e )+e ( 1)=(x 4)e     
(n) n  x
f (x)=( 1) (x n)e .
1000  x  x
So D xe =(x 1000)e .

41. By measuring the slope of the graph of s= f (t) at t=0 , 1 , 2 , 3 , 4 , and 5 , and using the method
/
of Example 1 in Section 3.2, we plot the graph of the velocity function v= f (t) in the first figure. The
// /
acceleration when t=2 s is a= f (2) , the slope of the tangent line to the graph of f when t=2 . We
// / 27 2
estimate the slope of this tangent line to be a(2)= f (2)=v (2) =9 ft / s . Similar measurements
3
enable us to graph the acceleration function in the second figure.

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

42. (a) Since we estimate the velocity to be a maximum at t=10 , the acceleration is 0 at t=10 .
(b) Drawing a tangent line at t=10 on the graph of a , a appears to decrease by 10 ft / s over a
period of 20 s. So at t=10 s, the jerk is approximately  10/20= 0.5 (ft/s ) s or ft / s .

43. (a) s=2t  15t +36t+2

3

(b) a(1)=12 1 30= 18 m / s

2
2
v(t)=s (t)=6t  30t+36
2
/

(c) v(t)=6(t  5t+6)=6(t 2)(t 3)=0 when t=2 or 3 and a(2)=24 30= 6 m / s , a(3)=36 30=6 m / s .
2
a(t)=v (t)=12t 30
/
. 2
/

2
3
2

3 2 / 2 /
44. (a) s=2t  3t  12t v(t)=s (t)=6t  6t 12 a(t)=v (t)=12t 6
2
(b) a(1)=12 1 6=6 m / s
2
(c) v(t)=6(t  t 2)=6(t+1)(t 2)=0 when t= 1 or 2 . Since t 0 , t  1 and
2
a(2)=24 6=18 m / s .

 
45. (a) s=sin t +cos t , 0 t 2 .
6 6
/       
v(t)=s (t)=cos t   sin t  = cos t  sin t 
6 6 6 6 6 6 6
2
/        
a(t)=v (t)=  sin t   cos t  = sin t +cos t
6 6 6 6 6 36 6 6
2 2 2
    1 3  2
(b) a(1)= sin  1 +cos  1 = + = (1+ 3 )  0.3745 m / s
36 6 6 36 2 2 72
 
(c) v(t)=0 for 0 t 2 cos t =sin t 
6 6

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives


sin t
6
1= 

cos t
6
   1 6  3
tan t =1 t=tan 1 t=  = =1.5 s. Thus,
6 6  4 2
2 2 2
3   3  3  2 2  2
a( )= sin  +cos  = + = 2   0.3877 m / s .
2 36 6 2 6 2 36 2 2 36

3 2 / 2 /
46. (a) s=2t  7t +4t+1 v(t)=s (t)=6t  14t+4 a(t)=v (t)=12t 14
2
(b) a(1)=12 14= 2 m / s
1 1 1
( 2
(c) v(t)=2 3t  7t+2 =2(3t 1)(t 2)=0 when t= ) 3
or 2 and a
3
=12
3
2
 14= 10 m / s ,
2
a(2)=12(2) 14=10 m / s .

4 3 / 3 2 / 2
47. (a) s(t)=t  4t +2 v(t)=s (t)=4t  12t  a(t)=v (t)=12t  24t=12t(t 2)=0 when t=0 or 2 .
(b) s(0)=2 m, v(0)=0 m / s, s(2)= 14 m, v(2)= 16 m / s

3 2 / 2 /
48. (a) s(t)=2t  9t  v(t)=s (t)=6t  18t  a(t)=v ( t ) =12t 18=0 when t=1.5 .
(b) s(1.5)= 13.5 m, v(1.5)= 13.5 m / s

3 2 / 2
49. (a) s= f (t)=t  12t +36t , t 0 v(t)= f (t)=3t  24t+36 .
/ 2
a(t)=v (t)=6t 24 . a(3)=6(3) 24= 6 ( m/s ) / s or m / s .

(b)
(c) The particle is speeding up when v and a have the same sign. This occurs when 2<t<4 and when
t>6 . It is slowing down when v and a have opposite signs; that is, when 0 t<2 and when 4<t<6 .

2 2
t / (1+t )(1) t(2t) 1 t
50. (a) x(t)=  v(t)=x (t)= = .
2 22 22
1+t (1+t ) (1+t )
2
/ 2t (t  3) 2
a(t)=v (t)= . a(t)=0 2t (t  3)=0 t=0 or 3
23
(1+t )
(b)
8
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

(c) v and a have the same sign and the particle is speeding up when 1<t< 3 . The particle is slowing
down and v and a have opposite signs when 0<t<1 and when t> 3 .

/ / 2
51. (a) y(t)=Asin  t  v(t)=y (t)=A cos  t  a(t)=v (t)= A sin  t
2 2 2
(b) a(t)= A sin  t=  (Asin  t)=  y(t) , so a(t) is proportional to y(t) .
(c) speed = v(t) =A cos  t is a maximum when cos  t=
1 . But when cos  t=
1 , we have
2 2
sin  t=0 , and a(t)= A sin  t= A (0)=0 .

dv dv ds dv dv
52. By the Chain Rule, a(t)= = = v(t)=v(t) . The derivative dv/dt is the rate of change
dt ds dt ds ds
of the velocity with respect to time (in other words, the acceleration) whereas the derivative dv/ds is
the rate of change of the velocity with respect to the displacement.

2 / // //
53. Let P(x)=ax +bx+c . Then P (x)=2ax+b and P (x)=2a . P (2)=2 2a=2 a=1 .
/
P (2)=3 2(1)(2)+b=3 4+b=3 b= 1 .
2 2
P(2)=5 1(2) +( 1)(2)+c=5 2+c=5 c=3 . So P(x)=x  x+3 .

3 2 / 2 // ///
54. Let Q(x)=ax +bx +cx+d . Then Q (x)=3ax +2bx+c , Q (x)=6ax+2b and Q (x)=6a . Thus,
/ // ///
Q ( 1 ) =a+b+c+d=1 , Q ( 1 ) =3a+2b+c=3 , Q (1)=6a+2b=6 and Q (1)=6a=12 . Solving these four
3 2
equations in four unknowns a , b , c and d we get a=2 , b= 3 , c=3 and d= 1 , so Q(x)=2x  3x +3x 1
.

/ // // /
55. y=Asin x+Bcos x y =Acos x Bsin x y = Asin x Bcos x . Substituting into y +y  2y=sin x
gives us ( 3A B)sin x+(A 3B)cos x=1sin x , so we must have  3A B=1 and A 3B=0 . Solving for A
1 3
and B , we add the first equation to three times the second to get B= and A= .
10 10
2 / //
56. y=Ax +Bx+C  y =2Ax+B y =2A . We substitute these expressions into the equation
// / 2
y +y  2y=x to get
2 2
(2A)+(2Ax+B) 2(Ax +Bx+C) = x
2
2A+2Ax+B 2Ax  2Bx 2C = x2

9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

2 2
( 2A)x +(2A 2B)x+(2A+B 2C) = (1)x +(0)x+(0)
2 1
The coefficients of x on each side must be equal, so  2A=1 A= . Similarly, 2A 2B=0
2
1 1 3
A=B= and 2A+B 2C=0  1  2C=0 C= .
2 2 4

rx / rx // 2 rx // / 2 rx rx rx rx 2 rx
57. y=e  y =re  y =r e , so y +5y  6y=r e +5re  6e =e (r +5r 6)=e (r+6)(r 1)=0
(r+6)(r 1)=0 r=1 or  6.

 x /  x // 2  x / //  x  x 2  x  x 2
58. y=e  y = e  y = e . Thus, y+y =y  e + e = e  e (    1)=0
1
5  x
 = , since e  0.
2

2 / / 2 2 2 2 / 2
59. f (x)=xg(x ) f (x)=x g (x ) 2x+g(x ) 1=g(x )+2x g (x )
// / 2 2 // 2 / 2 / 2 3 // 2
f (x)=g (x ) 2x+2x  g (x ) 2x+g (x ) 4x=6xg (x )+4x g (x )
/
g(x) / xg (x) g(x)
60. f (x)=  f (x)= 
x 2
x
2 / // / / 2 // /
// x [g (x)+xg (x) g (x)] 2x[xg (x) g(x)] x g (x) 2xg (x)+2g(x)
f (x)= =
4 3
x x

61.
/
g ( x)
f (x) =g( x ) f (x)=g ( x ) 1 x 1/2= /

/
2 2 x
// 1  1/2 / 1  1/2
2 x g ( x ) x  g ( x ) 2 x x
 1/2 //
[ x g ( x ) g ( x )]
/
// 2 2
f (x) = =
2 4x
(2 x )
// /
x g ( x ) g ( x )
=
4x x

62.
10
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

5 3 / 4 2 // 3 2
f (x)=3x  10x +5 f (x)=15x  30x  f (x)=60x  60x=60x(x  1)=60x(x+1)(x 1)
//
So f (x)>0 when  1<x<0 or x>1 , and on these intervals the graph of f lies above its tangent lines;
//
and f (x)<0 when x< 1 or 0<x<1 , and on these intervals the graph of f lies below its tangent lines.

63. (a)
1 /  (2x+1)
f (x) = 2
 f (x)= 
2 2
x +x (x +x)
2 2 2 2
(x +x) ( 2)+(2x+1)(2)(x +x)(2x+1) 2(3x +3x+1)
//
f (x) = = 
2 4 2 3
(x +x) (x +x)
2 3 2 2 2
/// (x +x) (2)(6x+3) 2(3x +3x+1)(3)(x +x) (2x+1)
f (x) = 2 6
(x +x)
3 2
 6(4x +6x +4x+1)
= 
2 4
(x +x)
2 4 2 3 2 2 3
(4) (x +x) ( 6)(12x +12x+4)+6(4x +6x +4x+1)(4)(x +x) (2x+1)
f (x) = 2 8
(x +x)
4 3 2
24(5x +10x +10x +5x+1)
=
2 5
(x +x)
(5)
f (x) =?

1 1 1 /  2  2 //  3  3
(b) f (x)= =   f (x)= x +(x+1)  f (x)=2x  2(x+1) 
x(x+1) x x+1
///  4  4 ( n) n  ( n+1 )  ( n+1 )
f (x)=( 3)(2)x +(3)(2)(x+1)      f (x)=( 1) n! x  (x+1)

4 3 2
7x+17 ///  6(56x +544x  2184x +6184x 6139)
64. (a) For f (x)= 2
, a CAS gives us f (x)=
2 4
2x  7x 4 (2x  7x 4)
7x+17  3 5
(b) Using a CAS we get f (x)= = + . Now we differentiate three times to obtain
2 2x+1 x 4
2x  7x 4
/// 144 30
f (x)=  .
4 4
(2x+1) (x 4)

65.
2 x / 2 x x x 2
For f (x)=x e , f (x)=x e +e (2x) =e (x +2x). Similarly, we have
11
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

// x 2
f (x) =e (x +4x+2)
/// x 2
f (x) =e (x +6x+6)
(4) x 2
f (x) =e (x +8x+12)
(5) x 2
f (x) =e (x +10x+20)

It appears that the coefficient of x in the quadratic term increases by 2 with each differentiation. The
pattern for the constant terms seems to be 0=1 0, 2=2 1, 6=3 2, 12=4 3, 20=5 4. So a reasonable
(n) x 2
guess is that f (x)=e x +2nx+n(n 1) .
(n) x 2
Proof: Let S be the statement that f (x)=e x +2nx+n(n 1) . 1. S is true because
n 1
/ x 2
f (x)=e (x +2x).
(k) x 2
2. Assume that S is true; that is, f (x)=e x +2kx+k(k 1) . Then
k
(k+1) d (k) x 2 x
f (x) = dx f (x) =e (2x+2k)+ x +2kx+k(k 1) e
x 2 2 x 2
=e x +(2k+2)x+(k +k) =e x +2(k+1)x+(k+1)k
This shows that S is true.
k+1
(n) x 2
3. Therefore, by mathematical induction, S is true for all n; that is, f (x)=e x +2nx+n(n 1) for
n
every positive integer n.

/ / /
66. (a) Use the Product Rule repeatedly: F= fg F = f g+ fg 
// // / / / / // // / / //
F =( f g+ f g )+( f g + fg )= f g+2 f g + fg .
/// /// // / // / / // / // /// /// // / / // ///
(b) F = f g+ f g +2( f g + f g )+ f g + fg = f g+3 f g +3 f g + fg 
( 4) ( 4) /// / /// / // // // // / /// / /// ( 4)
F = f g+ f g +3( f g + f g )+3( f g + f g )+ f g + fg
( 4) /// / // // / /// ( 4)
= f g+4 f g +6 f g +4 f g + fg

(c) By analogy with the Binomial Theorem, we make the guess:

( n) ( n) ( n 1 ) / ( n 2 ) // ( n k ) ( k ) / ( n 1 ) ( n)
F =f g+nf g + n f g +   + n f g +   +nf g + fg
2 k

n! n(n 1)(n 2)   (n k+1)

where n = = .
k k!(n k)! k!

67. The Chain Rule says that

12
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.7 Higher Derivatives

dy dy du
= , so
dx du dx
2
d y d dy d dy du d dy du dy d du
2
= = = + [Product Rule]
dx dx dx du dx dx du dx du dx dx
dx
2 2 2 2
d dy du du dy d u d y du dy d u
= + = +
du du dx dx du 2 2 dx du 2
dx du dx

2 2 2 2
d y d y du dy d u
68. From Exercise 65, = + 
2 2 dx du 2
dx du dx
3 2 2 2 2
d y d d y d d y du d dy d u
= = +
3 dx 2 dx 2 dx dx du 2
dx dx du dx
2 2 2 2 2 2
d d y du d du d y d dy d u d d u dy
= + + +
dx 2 dx dx dx 2 dx du 2 dx 2 du
du du dx dx
2 2 2 2 2 3
d d y du du du d u d y d dy du d u d u dy
= +2 + +
du 2 dx dx dx 2 2 du du dx 2 3 du
du dx du dx dx
3 3 2 2 3
d y du du d u d y dy d u
= +3 +
3 dx dx 2 2 du 3
du dx du dx

(n)
69. We will show that for each positive integer n , the n th derivative f exists and equals one of f ,
/ // /// ( p 1) ( p) / // /// ( p 1)
f ,f ,f , ... , f . Since f = f , the first p derivatives of f are f , f , f , ... , f ,
and f. In particular, our statement is true for n=1 . Suppose that k is an integer, k 1 , for which f is k
(k)
 times differentiable with f in the set
/ // ( p 1)
S={ f , f ,f , ... , f } . Since f is p  times differentiable, every member of S is differentiable,
(k+1) (k+1) / //
so f exists and equals the derivative of some member of S . Thus, f is in the set { f ,f ,
/// ( p) ( p)
f , ... , f } , which equals S since f = f . We have shown that the statement is true for n=1 and
that its truth for n=k implies its truth for n=k+1 . By mathematical induction, the statement is true for
all positive integers n.

13
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

d 1
1. The differentiation formula for logarithmic functions, (log x)= , is simplest when a=e
dx a xln a
because ln e=1.

2 / 1
d 2 2x
2. f (x)=ln (x +10) f (x)= (x +10)=
2 dx 2
x +10 x +10

/ 1 d  sin 
3. f ( )=ln (cos  ) f ( )= (cos  )= = tan 
cos  d cos 

/ 1  sin (ln x)
4. f (x)=cos (ln x) f (x)= sin (ln x) =
x x

/ 1 d  3 3
5. f (x)=log (1 3x) f (x)= (1 3x)= or
2 (1 3x)ln 2 dx (1 3x)ln 2 (3x 1)ln 2

x / 1 1 1
6. f (x)=log =log x log (x 1) f (x)=  or 
10 x 1 10 10 xln 10 (x 1)ln 10 x(x 1)ln 10

5 1/5 / 1  4/5 d 1 1 1
7. f (x)= ln x =(ln x)  f (x)= (ln x) (ln x)=  =
5 dx 4/5 x 5 4
5(ln x) 5x (ln x)

5 1/5 1 / 1 1 1
8. f (x)=ln x =ln x = ln x f (x)=  =
5 5 x 5x

/ 1 1 1 ln x 2+ln x
9. f (x)= x ln x f (x)= x +(ln x) = + =
x 2 x x 2 x 2 x

1+ln t
10. f (t)= 
1 ln t
/ (1 ln t)(1/t) (1+ln t)( 1/t) (1/t) (1 ln t)+(1+ln t) 2
f (t)= = =
2 2 2
(1 ln t) (1 ln t) t(1 ln t)

3
(2t+1) 3 4
11. F(t)=ln =ln (2t+1)  ln (3t 1) =3ln (2t+1) 4ln (3t 1)
4
(3t 1)
/ 1 1 6 12  6(t+3)
F (t)=3  2 4  3=  , or combined, .
2t+1 3t 1 2t+1 3t 1 (2t+1)(3t 1)

12.
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

2
x  1 +x
(
h(x)=ln x+ x  1 
2
) /
h (x)=
1
2
1+
x
2
=
1
2

2
=
1
2
x+ x  1 x 1 x+ x  1 x 1 x 1

a x
13. g(x)=ln =ln (a x) ln (a+x)
a+x
/ 1 1  (a+x) (a x)  2a
g (x)= ( 1) = =
a x a+x (a x)(a+x) 2 2
a x

y
y / 1 y y ye y
14. F(y)=yln (1+e ) F (y)=y  e +ln (1+e ) 1= +ln (1+e )
y y
1+e 1+e

ln u
15. f (u)= 
1+ln (2u)
1 1 1
1+ln (2u)   ln u  2 1+ln (2u) ln u
/ u 2u u
f (u) = =
2 2
1+ln (2u) 1+ln (2u)
1+(ln 2+ln u) ln u 1+ln 2
= =
2 2
u 1+ln (2u) u 1+ln (2u)

4 2 4 2 / 1 1 4
16. y=ln (x sin x)=ln x +ln (sin x) =4ln x+2ln sin x y =4 +2  cos x= +2cot x
x sin x x

2 / 1  10x 1 10x+1
17. y=ln |2 x 5x | y =  ( 1 10x)= or 2
2 2
2 x 5x 2 x 5x 5x +x 2

3u+2 1 / 1 3 3  6
18. G(u)=ln = ln (3u+2) ln (3u 2)  G (u)=  =
3u 2 2 2 3u+2 3u 2 2
9u  4

 x  x  x  x / 1  1 x+1 x
19. y=ln (e +xe )=ln (e (1+x))=ln (e )+ln (1+x)= x+ln (1+x) y = 1+ = =
1+x 1+x 1+x

x x
x 2 / x 1 x 2e ln (1+e )
20. y=[ln (1+e )]  y =2[ln (1+e )]  e=
x x
1+e 1+e

/ //
21. y=xln x y =x(1/x)+(ln x) 1=1+ln x y =1/x
2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

2
ln x / x (1/x) (ln x)(2x) x(1 2ln x) 1 2ln x
22. y=  y = = = 
2 2 2 4 3
x (x ) x x
3 2 2
// x ( 2/x) (1 2ln x)(3x ) x ( 2 3+6ln x) 6ln x 5
y = = =
32 6 4
(x ) x x

/ 1 1 1 // 1 1 1
23. y=log x y = =  y =  =
10 xln 10 ln 10 x ln 10 2 2
x x ln 10

2
/ sec xtan x+sec x //
24. y=ln (sec x+tan x) y = =sec x y =sec xtan x
sec x+tan x

x
25. f (x)= 
1 ln (x 1)
 1 (x 1)[1 ln (x 1)]+x
[1 ln (x 1)] 1 x
/ x 1 x 1 x 1 (x 1)ln (x 1)+x
f (x) = = =
2 2 2
[1 ln (x 1)] [1 ln (x 1)] (x 1)
2x 1 (x 1)ln (x 1)
=
2
(x 1)[1 ln (x 1)]

Dom( f ) ={ x| x 1>0 and 1 ln (x 1) 0}={ x| x>1 and ln (x 1) 1}
1
={ x| x>1 and x 1 e }={ x| x>1 and x 1+e}=(1, 1+e) (1+e,  )

1 / 1/x 1
26. f (x) =  f (x) =  = . Dom( f ) = { x| x>0 and ln x  1} =
1+ln x 2 2
(1+ln x) x(1+ln x)
{ x| x>0 and x 1/e} = (0, 1/e) (1/e,  ).

2 3
2 2 / 2 x ( 2x) 2 2x
27. f (x) = x ln (1 x ) f (x) = 2xln (1 x )+ = 2xln (1 x ) .
2 2
1 x 1 x
2
Dom( f ) = { x|1 x >0} = { x | | x|<1} = ( 1, 1).
/ 1 1 1
28. f (x) = ln ln ln x f (x) =   .
ln ln x ln x x
Dom( f ) = { x|ln ln x>0} = { x|ln x>1} = { x| x>e} = (e,  ).

29. f (x) =
3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

x / ln x x(1/x) ln x 1 / 1 1
 f (x) = =  f (e) = =0
ln x 2 2 2
(ln x) (ln x) 1

2 / 2 1 /
30. f (x) = x ln x f (x) = 2xln x+x = 2xln x+x f (1) = 2ln 1+1 = 1
x

/ 1 1 / 1
31. y = f (x) = ln ln x f (x) =  f (e) = , so an equation of the tangent line at
ln x x e
1 1
(e, 0) is y 0 = (x e), or y = x 1, or x ey = e.
e e

3 / 1 2 / 12
32. y=ln (x  7) y =  3x  y (2)= =12, so an equation of a tangent line at (2, 0) is
3 8 7
x 7
y 0=12(x 2) or y=12x 24.
/
33. f (x)=sin x+ln x f (x)=cos x+1/x. This is reasonable, because the graph shows that f increases
/ /
when f is positive, and f (x)=0 when f has a horizontal tangent.

ln x / x(1/x) ln x 1 ln x / 1 0 / 1 1
34. y=  y = = . y (1)= =1 and y (e)= =0 equations of tangent
x 2 2 2 2
x x 1 e
lines are y 0=1(x 1) or y=x 1 and y 1/e=0(x e) or y=1/e.

5 4 6 5 4 6
35. y=(2x+1) (x  3)  ln y=ln ((2x+1) (x  3) )
4 1 / 1 1 3
ln y=5ln (2x+1)+6ln (x  3) y =5  2+6  4x 
y 2x+1 4
x 3
3 3
/ 10 24x 5 4 6 10 24x
y =y + 4 =(2x+1) (x  3) + 4 .
2x+1 2x+1
x 3 x 3
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

2 2
x 2 10 x 2 10 1 2 2
36. y= x e (x +1)  ln y=ln x +ln e +ln (x +1)  ln y= ln x+x +10ln (x +1)
2
2
1 / 1 1 1 / x 2 10 1 20x
y =  +2x+10  2x y = x e (x +1) +2x+ 2
y 2 x 2 2x
x +1 x +1

2 4
sin xtan x 2 4 2 2
37. y=  ln y=ln (sin xtan x) ln (x +1) 
2 2
(x +1)
2 4 2 2 2
ln y=ln (sin x) +ln (tan x)  ln (x +1)  ln y=2ln |sin x|+4ln |tan x| 2ln (x +1)
1 / 1 1 2 1
y =2  cos x+4  sec x 2 2  2x
y sin x tan x
x +1
2 4 2
/ sin xtan x 4sec x 4x
y = 2cot x+ 
2 2 tan x 2
(x +1) x +1

2
4 x +1 1 2 1 2 1 / 1 1 1 1
38. y=  ln y= ln (x +1) ln (x  1) y =   2x   2x
2 4 4 y 4 2 4 2
x 1 x +1 x 1
2 2 2
/ 4 x +1 1 x x 1 4 x +1  2x x 4 x +1
y =   = =
2 2 2 2 2 2 4 4 2
x 1 x +1 x 1 x 1 x 1 1 x x 1

x x / / / x
39. y=x  ln y=ln x  ln y=xln x y /y=x(1/x)+(ln x) 1  y =y(1+ln x)  y =x (1+ln x)

/
1/x 1 y 1 1 1 / 1/x 1 ln x
40. y=x  ln y= ln x = +(ln x)   y =x
x y x x 2 2
x x

/
sin x sin x y 1
41. y=x  ln y=ln x  ln y=sin xln x =(sin x) +(ln x)(cos x)
y x
/ sin x / sin x sin x
y =y +ln xcos x  y =x +ln xcos x
x x
/
x y 1 / x
42. y=(sin x)  ln y=xln (sin x) =x  cos x+ ln (sin x)  1 y =(sin x) xcot x+ln (sin x)
y sin x

/
x x y 1 1
43. y=(ln x)  ln y=ln (ln x)  ln y=xln ln x =x  +(ln ln x) 1
y ln x x
5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

/ x / x 1
y =y +ln ln x  y =(ln x) +ln ln x
xln x ln x

/
ln x 2 y 1 / ln x 2ln x
44. y=x  ln y=ln xln x=(ln x)  =2ln x  y =x
y x x

x / x
e x y x 1 x / e x 1
45. y=x  ln y=e ln x =e  +(ln x) e  y =x e ln x+
y x x

/
cos x y 1 1
46. y=(ln x)  ln y=cos xln (ln x) =cos x  +(ln ln x)( sin x)
y ln x x
/ cos x cos x
y =(ln x)  sin xln ln x
xln x

/
2 2 / 1 d 2 2 / 2x+2yy 2 / 2 / /
47. y=ln (x +y ) y = (x +y )  y =  x y +y y =2x+2yy 
2 2 dx 2 2
x +y x +y
2 / 2 / / 2 2 / / 2x
x y +y y  2yy =2x  (x +y  2y) y =2x  y = 2 2
x +y  2y

y x 1 / 1 / / x / y / ln y y/x
48. x =y  yln x=xln y y +(ln x) y =x  y +ln y y ln x y =ln y  y =
x y y x ln x x/y
/  1 //  2
49. f (x)=ln (x 1) f (x)=1/(x 1)=(x 1)  f (x)= (x 1) 
///  3 (4)  4
f (x)=2(x 1)  f (x)= 2 3(x 1)     
(n) n 1  n n 1 (n 1)!
f (x)=( 1)  2 3 4 ... (n 1)(x 1) =( 1)
n
(x 1)

8 9 8 / 8 7 7 7
50. y=x ln x , so D y=D y =D (8x ln x+x ). But the eighth derivative of x is 0 , so we now have
8 7 7 6 6 7 6
D (8x ln x) = D (8 7x ln x+8x )=D (8 7x ln x)
= D6(8 7 6x5ln x)=...=D(8! x0ln x)=8!/x.

1 / /
51. If f (x)=ln (1+x), then f (x)= , so f (0)=1. Thus,
1+x
ln (1+x) f (x) f (x) f (0) /
lim =lim =lim = f (0)=1.
x 0 x x 0 x x 0 x 0

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.8 Derivatives of Logarithmic Functions

52. Let m=n/x. Then n=xm, and as n  , m  . Therefore,

x n 1 mx 1 m x x
lim 1+ = lim 1+ = lim 1+ =e by Equation 6.
n  n m  m m  m

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

1 0 0
1. (a) sinh 0= (e  e )=0
2
1 0 0 1
(b) cosh 0= (e +e )= (1+1)=1
2 2

0  0
(e  e )/2
2. (a) tanh 0= 0  0
=0
(e +e )/2
1  1 2
e e e 1
(b) tanh 1= =  0.76159
1  1 2
e +e e +1

1
ln 2  ln 2 ln 2 ln 2  1  1 2
e  e e  (e ) 2 2 2 3
3. (a) sinh (ln 2)= = = = =
2 2 2 2 4
1 2 2
(b) sinh 2= (e  e ) 3.62686
2

1 3 3
4. (a) cosh 3= (e +e ) 10.06766
2
1
ln 3  ln 3 3+
e +e 3 5
(b) cosh (ln 3)= = =
2 2 3

1 1
5. (a) sech 0= = =1
cosh 0 1
 1
(b) cosh 1=0 because cosh 0=1 .

1 1 1
6. (a) sinh 1= (e  e ) 1.17520
2
(b) Using Equation 3, we have sinh 1=ln 1+ 1 +1 =ln (1+ 2 ) 0.88137 .
 1
( 2
)
1 x  ( x) 1 x x 1 x x
7. sinh ( x)= e  e = (e  e )= (e  e )= sinh x
2 2 2

1 x  ( x) 1 x x 1 x x
8. cosh ( x)= e +e = (e +e )= (e +e )=cosh x
2 2 2

1 x x 1 x x 1 x x
9. cosh x+sinh x= (e +e )+ (e  e )= (2e )=e
2 2 2
1
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

1 x x 1 x x 1  x  x
10. cosh x sinh x= (e +e ) (e  e )= (2e )=e
2 2 2

1 x x 1 y y 1 x x 1 y y
11. sinh xcosh y+cosh xsinh y= (e  e ) (e +e ) + (e +e ) (e  e )
2 2 2 2
1 x+y x y  x+y  x y x+y x y  x+y  x y
= (e +e  e  e )+(e  e +e  e )
4
1 x+y  x y 1 x+y  (x+y)
= (2e  2e )= e  e =sinh (x+y)
4 2

1 x x 1 y y 1 x x 1 y y
12. cosh xcosh y+sinh xsinh y= (e +e ) (e +e ) + (e  e ) (e  e )
2 2 2 2
1 x+y x y  x+y  x y x+y x y  x+y  x y
= (e +e +e +e )+(e  e  e +e )
4
1 x+y  x y 1 x+y  (x+y)
= (2e +2e )= e +e =cosh (x+y)
4 2

2 2 2
13. Divide both sides of the identity cosh x sinh x=1 by sinh x :
2 2
cosh x sinh x 1 2 2
2
 2
= 2
 coth x 1=csch x.
sinh x sinh x sinh x
14.
sinh xcosh y cosh xsinh y
+
sinh (x+y) sinh xcosh y+cosh xsinh y cosh xcosh y cosh xcosh y
tanh (x+y) = = =
cosh (x+y) cosh xcosh y+sinh xsinh y cosh xcosh y sinh xsinh y
+
cosh xcosh y cosh xcosh y
tanh x+tanh y
=
1+tanh xtanh y

15. Putting y=x in the result from Exercise 11, we have

sinh 2x=sinh (x+x)=sinh xcosh x+cosh xsinh x=2sinh xcosh x.

16. Putting y=x in the result from Exercise 12, we have

2 2
cosh 2x=cosh (x+x)=cosh xcosh x+sinh xsinh x=cosh x+sinh x .

17.
ln x  ln x ln x  1  1
sinh (ln x) (e  e )/2 x (e ) x x
tanh (ln x) = = ln x  = =
cosh (ln x) ln x ln x  1  1
(e +e )/2 x+(e ) x+x

2
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

2 2
x 1/x (x  1)/x x  1
= = 2 = 2
x+1/x
(x +1)/x x +1

18.
1 x x 1 x x
(e +e )+ (e  e )
1+tanh x 1+(sinh x)/cosh x cosh x+sinh x 2 2
= = =
1 tanh x 1 (sinh x)/cosh x cosh x sinh x 1 x  x 1 x  x
(e +e ) (e  e )
2 2
x  x x  x x
e +e +e  e 2e 2x
= = =e
x  x x  x  x
e +e  e +e 2e

x
cosh x+sinh x e 2x
Or: Using the results of Exercises 9 and 10, = =e
cosh x sinh x  x
e

n xn nx
19. By Exercise 9, (cosh x+sinh x) =(e ) =e =cosh nx+sinh nx .

3 4 2 2 9 25 5
20. sinh x=  csch x=1/sinh x=
. cosh x=sinh x+1= +1=  cosh x= (since cosh x>0 ).
4 3 16 16 4
4 3/4 3 5
coth x=1/cosh x= , tanh x=sinh x/cosh x= = , and cothx=1/tanh x= .
5 5/4 5 3

4 5 2 2 4 2 9 3
21. tanh x= >0 , so x>0 . cothx=1/tanh x= , sech x=1 tanh x=1 =  sech x=
5 4 5 25 5
5 4 5 4 3
(since sech x>0 ), cosh x=1/sech x= , sinh x=tanh xcosh x=  = , and csch x=1/sinh x= .
3 5 3 3 4

22.

23. (a)

3
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

x  x  x  2x
e e e 1 e 1 0
lim tanh x=lim  =lim = =1
x  x  x  2x 1+0
x x e +e e x 1+e
x  x x 2x
e e e e  1 0 1
(b) lim tanh x= lim  = lim = = 1
x  x x 2x 0+1
x  x  e +e e x  e +1
x  x
e e
(c) lim sinh x=lim =
x x 2
x  x
e e
(d) lim sinh x= lim =
x  x  2
2
(e) lim sech x=lim =0
x e +e x
x
x
x  x  x  2x
e +e e 1+e 1+0
(f) lim cothx=lim  =lim = =1
x  x  x  2x 1 0
x e e e x 1 e x
cosh x
(g) lim cothx=lim = , since sinh x 0 through positive values and cosh x 1.
+ + sinh x
x 0 x 0
cosh x
(h) lim cothx=lim = , since sinh x 0 through negative values and cosh x 1.
  sinh x
x 0 x 0
2
(i) lim csch x= lim =0
x  x
x  x  e e

d d 1 x x 1 x x
24. (a) cosh x= (e +e ) = (e  e )=sinh x
dx dx 2 2
2 2
d d sinh x cosh xcosh x sinh xsinh x cosh x sinh x 1 2
(b) tanh x= = = = =sech x
dx dx cosh x 2 2 2
cosh x cosh x cosh x
d d 1 cosh x 1 cosh x
(c) csch x= = =  = csch xcoth x
dx dx sinh x 2 sinh x sinh x
sinh x
d d 1 sinh x 1 sinh x
(d) sech x= = =  = sech xtanh x
dx dx cosh x 2 cosh x cosh x
cosh x
(e)
2 2
d d cosh x sinh xsinh x cosh xcosh x sinh x cosh x 1
coth x = = = =
dx dx sinh x 2 2 2
sinh x sinh x sinh x
2
= csch x
4
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

 1 2 2
25. Let y=sinh x . Then sinh y=x and, by Example 1(a), cosh y sinh y=1 [ with cosh y>0 ]
2 2
cosh y= 1+sinh y = 1+x . So by Exercise 9, e =sinh y+cosh y=x+ 1+x 
y 2
(
y=ln x+ 1+x
2
).
 1 2 2
26. Let y=cosh x . Then cosh y=x and y
0 , so sinh y= cosh y 1 = x  1 . So, by Exercise 9,
y
e =cosh y+sinh y=x+ x  1  y=ln x+ x  1 .
2
( 2
)
1 y y
Another method: Write x=cosh y= (e +e ) and solve a quadratic, as in Example 3.
2

y  y y 2y
sinh y (e  e )/2 e
 1 e  1
27. (a) Let y=tanh x . Then x=tanh y= = y y  = 2y 
cosh y y
(e +e )/2 e e +1
2y 2y 2y 2y 2y
xe +x=e  1 1+x=e  xe  1+x=e (1 x)
2y 1+x 1+x 1 1+x
e =  2y=ln  y= ln .
1 x 1 x 2 1 x
 1
(b) Let y=tanh x . Then x=tanh y , so from Exercise 18 we have
2y 1+tanh y 1+x 1+x 1 1+x
e = =  2y=ln  y= ln .
1 tanh y 1 x 1 x 2 1 x

28. (a)
 1
(i) y=csch x csch y=x ( x 0 )
 1
(ii) We sketch the graph of csch by reflecting the graph of csch (see Exercise 22) about the line
y=x.
2
 1 2 y  y y2 y y 1 x +1
(iii) Let y=csch x . Then x=csch y=  xe  xe =2 x(e )  2e  x=0 e= .
y  y x
e e

2 2
y 1+ x +1 y y 1 x +1
But e >0 , so for x>0 , e = and for x<0 , e = . Thus,
x x
2
 1 1 x +1
csch x=ln + .
x | x|
5
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

(b)
 1
(i) y=sech x sech y=x and y>0.
 1
(ii) We sketch the graph of coth by reflecting the graph of coth (see Exercise 22) about the line
y=x.
2
 1 2 y  y y2 y y 1 1 x
(iii) Let y=coth x , so x=sech y=  xe +xe =2 x(e )  2e +x=0 e= .
y  y x
e +e

2
y 1 1 x 2
But y>0 e >1 . This rules out the minus sign because >1 1 1 x >x
x
2
2 2 2 2 1+ 1 x y
1 x> 1 x  1 2x+x >1 x  x >x x>1 , but x=sech y 1 . Thus, e = 
x
2
 1 1+ 1 x
sech x=ln .
x

(c)
 1
(i) y=coth x coth y=x
 1
(ii) We sketch the graph of coth by reflecting the graph of coth (see Exercise 22) about the line
y=x.
y  y
 1 e +e y  y y  y y  y 2y x+1
Let y=coth x . Then x=coth y=  xe  xe =e +e  (x 1)e =(x+1)e  e = 
y  y x 1
(iii) e e
x+1  1 1 x+1
2y=ln  coth x= ln
x 1 2 x 1

6
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

 1 dy
29. (a) Let y=cosh x . Then cosh y=x and y
0 sinh y =1
dx
dy 1 1 1
= = = (since sinh y
0 for y
0 ). Or: Use Formula 4.
dx sinh y 2 2
cosh y 1 x 1
 1 2 dy dy 1 1 1
(b) Let y=tanh x . Then tanh y=x sech y =1 = = = .
dx dx 2 2 2
sech y 1 tanh y 1 x
Or: Use Formula 5.
 1 dy dy 1
(c) Let y=csch x. Then csch y=x  csch ycoth y =1 = . By Exercise 13,
dx dx csch ycoth y
2 2 2
coth y= csch y+1 =
x +1 . If x>0 , then coth y>0 , so coth y= x +1 . If x<0 , then coth y<0 , so
2 dy 1 1
coth y= x +1 . In either case we have = = .
dx csch ycoth y 2
| x| x +1
 1 dy
(d) Let y=sech x. Then sech y=x  sech ytanh y =1
dx
dy 1 1 1
= = = . (Note that y>0 and so tanh y>0 .)
dx sech ytanh y 2 2
sech y 1 sech y x 1 x
 1 2 dy dy 1 1 1
(e) Let y=coth x. Then coth y=x  csch y =1 = = = by Exercise
dx dx 2 2 2
csch y 1 coth y 1 x
13.

/ 2
30. f (x)=tanh 4x f (x)=4sech 4x

/ / /
31. f (x)=xcosh x f (x)=x(cosh x) +(cosh x)(x) =xsinh x+cosh x

2 /
32. g(x)=sinh x g (x)=2sinh xcosh x

2 / 2 2
33. h(x)=sinh (x ) h (x)=cosh (x ) 2x=2xcosh (x )

/ 2
34. F(x)=sinh xtanh x F (x)=sinh xsech x+tanh xcosh x

1 cosh x
35. G(x)= 
1+cosh x
(1+cosh x)( sinh x) (1 cosh x)(sinh x)
/ =
G (x) 2
(1+cosh x)

7
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

 sinh x sinh xcosh x sinh x+sinh xcosh x  2sinh x

= =
2 2
(1+cosh x) (1+cosh x)

t / t t t
36. f (t)=e sech t  f (t)=e ( sech ttanh t)+(sech t)e =e sech t (1 tanh t)

2 2
2 / 2 1
2 2 1/2 tcsch 1+t
37. h(t)=coth 1+t  h (t)= csch 1+t  (1+t ) (2t)=
2 2
1+t

/ 1
38. f (t)=ln (sinh t) f (t)= cosh t=coth t
sinh t

t / 2 t t t 2 t
39. H(t)=tanh (e ) H (t)=sech (e ) e =e sech (e )

/
40. y=sinh (cosh x) y =cosh (cosh x) sinh x

cosh 3x / cosh 3x cosh 3x

41. y=e  y =e  sinh 3x 3=3e sinh 3x

2  1 / 2 1  1 x  1
42. y=x sinh (2x) y =x   2+sinh (2x) 2x=2x +sinh (2x)
2 2
1+(2x) 1+4x

 1 / 1 1  1/2 1
43. y=tanh x y =  x =
2 2 2 x (1 x)
1 ( x )

 1 2  1 1 2
44. y=xtanh x+ln 1 x =xtanh x+ ln (1 x )
2
/  1 x 1 1  1
y =tanh x+ + ( 2x)=tanh x
2 2 2
1 x 1 x

 1 2
45. y=xsinh (x/3) 9+x 
/  1 x 1/3 2x  1 x x x  1 x
y =sinh +x  =sinh +  =sinh
3 2 2 3 2 2 3
1+(x/3) 2 9+x 9+x 9+x

 1 2
46. y=sech 1 x 

8
 Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

/ 1  2x x
y = = 2
2 2 2
1 x 1 (1 x ) 2 1 x (1 x )| x|

 1 2 / 1 2x 1
47. y=coth x +1  y = 2
=
2 2
1 (x +1) 2 x +1 x x +1

48.
For y=acosh (x/a) with a>0 , we have the y  intercept equal to a. As a increases, the graph flattens.

/ 1
49. (a) y=20cosh (x/20) 15 y =20sinh (x/20) =sinh (x/20) . Since the right pole is positioned at
20
/ 7
x=7, we have y (7)=sinh  0.3572 .
20
7
(b) If  is the angle between the tangent line and the x  axis, then tan  = slope of the line =sinh
20
 1 7 
, so  =tan sinh  0.343 rad  19.66 . Thus, the angle between the line and the pole is
20
 
 =90    70.34 .

T  gx
50. We differentiate the function twice, then substitute into the differential equation: y= cosh
 g T
2
dy T  gx  g  gx d y  gx  g  g  gx
 = sinh =sinh  =cosh = cosh .
dx  g T T T 2 T T T T
dx
2
d y  g  gx
We evaluate the two sides separately: LHS= = cosh ,
2 T T
dx
 g dy 2  g 2  gx  g  gx
RHS= 1+ = 1+sinh = by the identity proved in Example 1(a).
T dx T T T T

/
51. (a) y=Asinh mx+Bcosh mx y =mAcosh mx+mBsinh mx
// 2 2 2 2
y =m Asinh mx+m Bcosh mx=m (Asinh mx+Bcosh mx)=m y
//
(b) From part (a), a solution of y =9y is y(x)=Asinh 3x+Bcosh 3x. So  4=y(0)=Asinh 0+Bcosh 0=B,
9
Stewart Calculus ET 5e 0534393217;3. Differentiation Rules; 3.9 Hyperbolic Functions

/ /
so B= 4. Now y (x)=3Acosh 3x 12sinh 3x 6=y (0)=3A A=2, so y=2sinh 3x 4cosh 3x .

x  x  2x
sinh x e e 1 e 1 0 1
52. lim =lim =lim = =
x x x x x 2 2 2
e 2e

/  1
53. The tangent to y=cosh x has slope 1 when y =sinh x=1 x=sinh 1=ln (1+ 2 ) , by Equation 3.
2
Since sinh x=1 and y=cosh x= 1+sinh x , we have cosh x= 2. The point is (ln (1+ 2 ), 2 ).

54.
1 ln (sec  +tan  )  ln (sec  +tan  )
cosh x =cosh [ln (sec  +tan  )]= e +e
2
1 1 1 sec   tan 
= sec  +tan  + = sec  +tan  +
2 sec  +tan  2 (sec  +tan  )(sec   tan  )
1 sec   tan  1
= sec  +tan  + = (sec  +tan  +sec   tan  )=sec 
2 2 2 2
sec   tan 

x  x
55. If ae +be = cosh (x+ ) [ or  sinh (x+ ) ], then
x  x  x+  x   x   x     x     x
ae +be = (e  e )= (e e  e e )= e e e e . Comparing coefficients
2 2 2 2
x  x     
of e and e , we have a= e (1) and b= e (2) . We need to find  and  . Dividing
2 2
equation (1) by
a 2 a 1 a
equation (2) gives us = e  (*)2 =ln ( )  = ln  . Solving equations (1) and
b b 2 b
  2a   2a  2
(2) for e gives us e = and e = , so =   = 4ab  =2  ab .
 2b  2b
a a
(*) If >0 , we use the + sign and obtain a cosh function, whereas if <0 , we use the  sign and
b b
obtain a sinh function.
x  x 1 a
In summary, if a and b have the same sign, we have ae +be =2 ab cosh x+ ln , whereas, if
2 b
x  x 1 a
a and b have the opposite sign, then ae +be =2  ab sinh x+ ln  .
2 b

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

1. A function f has an absolute minimum at x=c if f (c) is the smallest function value on the entire
domain of f , whereas f has a local minimum at c if f (c) is the smallest function value when x is
near c .

2. (a) The Extreme Value Theorem

(b) See the Closed Interval Method.

3. Absolute maximum at b ; absolute minimum at d ; local maxima at b and e ; local minima at d and
s;
neither a maximum nor a minimum at a , c , r , and t .

4. Absolute maximum at e ; absolute minimum at t ; local maxima at c , e , and s ; local minima at b ,

c , d , and r ;
neither a maximum nor a minimum at a .

5. Absolute maximum value is f (4)=4 ; absolute minimum value is f (7)=0 ; local maximum values
are f (4)=4 and f (6)=3 ; local minimum values are f (2)=1 and f (5)=2 .

6. Absolute maximum value is f (8)=5 ; absolute minimum value is f (2)=0 ; local maximum values
are f (1)=2 , f (4)=4 , and f (6)=3 ; local minimum values are f (2)=0 , f (5)=2 , and f (7)=1 .

7. Absolute minimum at 2, absolute maximum at 3, local minimum at 4

8. Absolute minimum at 1, absolute maximum at 5, local maximum at 2, local minimum at 4

9. Absolute maximum at 5, absolute minimum at 2, local maximum at 3, local minima at 2 and 4

1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

10. f has no local maximum or minimum, but 2 and 4 are critical numbers

11. (a)

(b)

(c)

12. (a) Note that a local maximum cannot occur at an endpoint.

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

(b)

Note: By the Extreme Value Theorem, f must not be continuous.

13. (a) Note: By the Extreme Value Theorem, f must not be continuous; because if it were, it would
attain an absolute minimum.

(b)

14. (a)

3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

(b)

15. f (x)=8 3x , x 1 . Absolute maximum f (1)=5 ; no local maximum. No absolute or local

minimum.

16. f (x)=3 2x , x 5 . Absolute minimum f (5)= 7 ; no local minimum. No absolute or local

maximum.

2
17. f (x)=x , 0<x<2 . No absolute or local maximum or minimum value.

4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

2
18. f (x)=x , 0<x 2 . Absolute maximum f (2)=4 ; no local maximum. No absolute or local
minimum.

2
19. f (x)=x , 0 x<2 . Absolute minimum f (0)=0 ; no local minimum. No absolute or local
maximum.

2
20. f (x)=x , 0 x 2 . Absolute maximum f (2)=4 . Absolute minimum f (0)=0 . No local maximum
or minimum.

2
21. f (x)=x ,  3 x 2 . Absolute maximum f ( 3)=9 . No local maximum. Absolute and local
minimum f (0)=0 .

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

2
22. f (x)=1+(x+1) ,  2 x<5 . No absolute or local maximum. Absolute and local minimum f ( 1)=1
.

23. f (t)=1/t , 0<t<1 . No maximum or minimum.

24. f (t)=1/t , 0<t 1 . Absolute minimum f (1)=1 ; no local minimum. No local or absolute
maximum.

3 
25. f ( )=sin  ,  2    2 . Absolute and local maxima f  =f =1 . Absolute and
2 2
 3
local minima f  =f = 1 .
2 2

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

  
26. f ( )=tan  ,   < . Absolute minimum f  = 1 ; no local minimum. No absolute
4 2 4
or local maximum.

27. f (x)=1 x . Absolute maximum f (0)=1 ; no local maximum. No absolute or local minimum.

x
28. f (x)=e . No absolute or local maximum or minimum value.

29. f (x)= {
1 x if 0 x<2
2x 4 if 2 x 3
Absolute maximum f (3)=2 ; no local maximum. No absolute or local minimum.

30. f (x)=
{ x
2

2 x
2
if 1 x<0
if0 x 1
7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

Absolute and local maximum f (0)=2 .

No absolute or local minimum.

2 / / 2 2
31. f (x)=5x +4x f (x)=10x+4 . f (x)=0 x= , so  is the only critical number.
5 5

3 2 / 2 / 1
32. f (x)=x +x  x f (x)=3x +2x 1 . f (x)=0 (x+1)(3x 1)=0 x= 1 , . These are the only
3
critical numbers.

3
33. f (x)=x +3x  24x
2 / 2
f (x)=3x +6x 24=3 x +2x 8 . ( 2
)
/
f (x)=0 3(x+4)(x 2)=0
x= 4 , 2 . These are the only critical numbers.
3 2 / 2 / 2  2 4 12
34. f (x)=x +x +x  f (x)=3x +2x+1 . f (x)=0 3x +2x+1=0 x= . Neither of these is
6
a real number. Thus, there are no critical numbers.

4
35. s(t)=3t +4t  6t 
3 2 / 3 2
s (t)=12t +12t  12t . s (t)=0
/
(2
12t t +t 1  ) 2
t=0 or t +t 1=0 . Using the
2
 1 1  4(1)( 1)  1 5
quadratic formula to solve the latter equation gives us t= =  0.618 ,
2(1) 2
 1 5
 1.618 . The three critical numbers are 0 , .
2

36. f (z)=
z+1
 f
/
(z)=
( z +z+1 )1 (z+1)(2z+1)
2
=
2
 z  2z
=0 z(z+2)=0 z=0 ,  2 are the
2 2 2
z +z+1 ( z +z+1)
2
( z +z+1)
2

2
critical numbers. (Note that z +z+1
0 since the discriminant <0 .)

{
3
2 if x>
37. g(x)= 2x+3 = { 2x+3 if 2x+3 0
 (2x+3) if 2x+3<0

/
g (x)=
 2 if x<
2
3
/
g (x) is never 0 , but
2
/
g (x) does not exist for
8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

3 3
x= , so  is the only critical number.
2 2

1/3  2/3 / 1  2/3 2  5/3 1  5/3 x+2

38. g(x)=x  x  g (x)= x + x = x (x+2)= .
3 3 3 5/3
3x
/ /
g ( 2)=0 and g (0) does not exist, but 0 is not in the domain of g , so the only critical number is  2 .

2/3 5/3 / 10  1/3 5 2/3 /

39. g(t)=5t +t  g (t)= t + t . g (0) does not exist, so t=0 is a critical number.
3 3
/ 5  1/3
g (t)= t (2+t)=0 t= 2 , so t= 2 is also a critical number.
3

1/2 3/2 / 1 3 /
40. g(t)= t (1 t)=t  t  g (t)=  t . g (0) does not exist, so t=0 is a critical number.
2 t 2
/ 1 3t 1 1
0=g (t)=  t= , so t= is also a critical number.
2 t 3 3

4/5 2
41. F(x)=x (x 4) 
/ 4/5 2 4  1/5 1  1/5
F (x) =x  2(x 4)+(x 4)  5 x = 5 x (x 4) 5 x 2+(x 4)  4
(x 4)(14x 16) 2(x 4)(7x 8) 8 /
= = =0 when x=4 , ; and F (0) does not exist.
1/5 1/5 7
5x 5x

8
Critical numbers are 0 , ,4.
7

3 1 2  2/3
42. G(x)=
2
x  x
3
x x
/ /
(2
)
(2x 1) . G (x) does not exist when x  x=0 , that is, when
G (x)=
/ 1 1
x=0 or 1 . G (x)=0 2x 1=0 x= . So the critical numbers are x=0 , ,1.
2 2

2 / /
43. f ( )=2cos  +sin   f ( )= 2sin  +2sin  cos  . f ( )=0 2sin  (cos   1)=0 sin  =0 or
cos  =1  =n ( n an integer) or  =2n . The solutions  =n include the solutions  =2n , so the
critical numbers are  =n .

/ 2 / 2 1 
44. g( )=4  tan   g ( )=4 sec  . g ( )=0 sec  =4 sec  = 2 cos  =   = +2n ,
2 3

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

5 2 4
+2n , +2n , and +2n are critical numbers.
3 3 3
/
Note: The values of  that make g ( ) undefined are not in the domain of g .

/ /  1
45. f (x)=xln x f (x)=x(1/x)+(ln x) 1=ln x+1 . f (x)=0 ln x= 1 x=e =1/e . Therefore, the only
critical number is x=1/e .

2x / 2x 2x 2x 2x /
46. f (x)=xe  f (x)=x(2e )+e =e (2x+1) . Since e is never 0 , we have f (x)=0 only when
1 1
2x+1=0 x= . So  is the only critical number.
2 2

2 /
47. f (x)=3x  12x+5 , 0,3 . f (x)=6x 12=0 x=2 . Applying the Closed Interval Method, we find
that f (0)=5 , f (2)= 7 , and f (3)= 4 . So f (0)=5 is the absolute maximum value and f (2)= 7 is the
absolute minimum value.

3 / 2
48. f (x)=x  3x+1 , 0,3 . f (x)=3x  3=0 x= 1 , but  1 is not in 0,3 . f (0)=1 , f (1)= 1 , and
f (3)=19 . So f (3)=19 is the absolute maximum value and f (1)= 1 is the absolute minimum value.

3 2 / 2
49. f (x)=2x  3x  12x+1 , [ 2,3] . f (x)=6x  6x 12=6 x  x 2 =6 ( x 2 ) ( x+1 ) =0 x=2, 1 . ( 2
)
f ( 2)= 3 , f ( 1)=8 , f (2)= 19 , and f (3)= 8 . So f ( 1)=8 is the absolute maximum value and
f (2)= 19 is the absolute minimum value.

3 2 / 2
50. f (x)=x  6x +9x+2 , [ 1,4] . f (x)=3x  12x+9=3 x  4x+3 =3(x 1)(x 3)=0 x=1,3 . f ( 1)= 14 ,( 2
)
f (1)=6 , f (3)=2 , and f (4)=6 . So f (1)= f (4)=6 is the absolute maximum value and f ( 1)= 14 is the
absolute minimum value.
4 2 / 3
( 2
)
51. f (x)=x  2x +3 , [ 2,3]. f (x)=4x  4x=4x x  1 =4x(x+1)(x 1)=0 x= 1 , 0 , 1 . f ( 2)=11 ,
f ( 1)=2 , f (0)=3 , f (1)=2 , f (3)=66 . So f (3)=66 is the absolute maximum value and f ( 1)=2 is the
absolute minimum value.
( 2
)3 /
( 2
)2
52. f (x)= x  1 , [ 1,2] . f (x)=3 x  1 ( 2x ) =6x ( x+1 ) ( x 1 ) =0 x= 1,0,1 . f ( 1)=0 , f (0)= 1
2 2

, and f (2)=27 . So f (2)=27 is the absolute maximum value and f (0)= 1 is the absolute minimum
value.

53. f (x)=
x
, 0,2 . f
/
(x)=
( x +1 )  x(2x)
2
=
1 x
2
=0 x= 1 , but  1 is not in 0,2 . f (0)=0
2 2 2
x +1 ( x +1)
2
( x +1)
2

1 2 1
, f (1)= , f (2)= . So f (1)= is the absolute maximum value and f (0)=0 is the absolute
2 5 2
10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

minimum value.

54. f (x)=
2
x 4
, [ 4,4] . f
/
(x)=
( x +4) (2x) ( x  4) (2x)
2 2
=
16x
=0 x=0 . f ( 4)=
12 3
= and
2 2 2 20 5
x +4 ( x +4)
2
( x +4)
2

3
f (0)= 1 . So f ( 4)= is the absolute maximum value and f (0)= 1 is the absolute minimum value.
5

2
55. f (t)=t 4 t ,  1,2 .
/
f (t)=t
1
4 t
2
( )  1/2
(  2t ) + ( 4 t
2 1/2
)  1=
 t
2
+ 4 t =
2
2
(
 t + 4 t
2
)= 4 2t
2
/
. f (t)=0
2 2 2 2
4 t 4 t 4 t
2 2 /
4 2t =0  t =2 t= 2 , but t= 2 is not in the given interval,  1,2 . f (t) does not exist if
2
4 t =0 t= 2 , but  2 is not in the given interval. f ( 1)= 3 , f ( 2 ) =2 , and f (2)=0 . So
f ( 2 ) =2 is the absolute maximum value and f ( 1)= 3 is the absolute minimum value.

3 1/3 4/3 / 8  2/3 4 1/3 4  2/3 4(2 t) /

56. f (t)= t ( 8 t ) , [0,8] . f (t)=8t  t  f (t)= t  t = t (2 t)= . f (t)=0 t=2
3 3 3 3 2
3 t
/ 3
. f (t) does not exist if t=0 . f (0)=0 , f (2)=6 2  7.56 , and f (8)=0 .
3
So f (2)=6 2 is the absolute maximum value and f (0)= f (8)=0 is the absolute minimum value.

 / sin x
57. f (x)=sin x+cos x , 0, . f (x)=cos x sin x=0 sin x=cos x =1 tan x=1
3 cos x
 3 +1  
x= . f (0)=1 , f = 2  1.41 , f
 1.37 . So f = = 2 is the absolute
4 4 3 2 4
maximum value and f (0)=1 is the absolute minimum value.

/ 1 5 
58. f (x)=x 2cos x ,   , . f (x)=1+2sin x=0 sin x= x= , . f (  )=2    1.14 ,
2 6 6
5 5  
f  = 3   0.886 , f  =  3  2.26 , f ( )= +2 5.14 . So f ( )= +2 is the
6 6 6 6
 
absolute maximum value and f  =  3 is the absolute minimum value.
6 6

 x /  x  x  x
59. f (x)=xe , 0,2 . f (x)=x( e )+e =e (1 x)=0 x=1 .
 1 2
f (0)=0 , f (1)=e =1/e 0.37 , f (2)=2/e  0.27 . So f (1)=1/e is the absolute maximum value and
f (0)=0 is the absolute minimum value.
11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

ln x / x(1/x) ln x 1 ln x
60. f (x)= , 1,3 . f (x)= = =0 1 ln x=0 ln x=1 x=e . f (1)=0/1=0 ,
x 2 2
x x
f (e)=1/e 0.368 , f (3)=(ln 3)/3 0.366 . So f (e)=1/e is the absolute maximum value and f (1)=0 is
the absolute minimum value.

3 x 3 / /
61. f (x)=x 3ln x , 1,4 . f (x)=1 = =0 x=3 . f does not exist for x=0 , but 0 is not in the
x x
domain of f . f (1)=1 , f (3)=3 3ln 3  0.296 , f (4)=4 3ln 4  0.159 . So f (1)=1 is the absolute
maximum value and f (3)=3 3ln 3  0.296 is the absolute minimum value.

x
 x  2x /  x  2x 2 1 2 e x
62. f (x)=e  e , 0,1 . f (x)=e ( 1) e ( 2)=  = =0 e =2 x=ln 2 0.69 . f (0)=0
2x x 2x
e e e
 ln 2  2ln 2 ln 2  1
1 1 1 ln 2  2
 1  2  1  2 1
, f (ln 2)=e  e = e  e ( ) ( )
=2  2 = 
2 4 4
= , f (1)=e  e  0.233 . So f (ln 2)=
4
is the absolute maximum value and f (0)=0 is the absolute minimum value.

a b
63. f (x)=x (1 x) , 0 x 1 , a>0 , b>0 .
/ a b 1 b a 1 a 1 b 1
f (x) =x  b(1 x) ( 1)+(1 x)  ax =x (1 x) x b( 1)+(1 x) a
a 1 b 1
=x (1 x) (a ax bx)
/
At the endpoints, we have f (0)= f (1)=0 [ the minimum value of f ]. In the interval (0,1) , f (x)=0
a
x= .
a+b
a b a b a b a b
a a a a a+b a a b ab
f = 1 = =  = .
a+b a+b a+b a a+b a b a+b
(a+b) (a+b) (a+b) ( a+b )
a b
a ab
So f = is the absolute maximum value.
a+b a+b
( a+b )
64.

/ /
We see that f (x)=0 at about x=0.0 and 2.0 , and that f (x) does not exist at about x= 0.7 , 1.0 , and
2.7 , so the critical numbers of f are about  0.7 , 0.0 , 1.0 , 2.0 , and 2.7 .

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

65. (a)
From the graph, it appears that the absolute maximum value is about f ( 1.63)=9.71 , and the absolute
minimum value is about f (1.63)= 7.71 . These values make sense because the graph is symmetric
3
about the point ( 0,1 ) . ( y=x  8x is symmetric about the origin.)

3 / 2 / 8
(b) f (x)=x  8x+1 f (x)=3x  8 . So f (x)=0 x= .
3
3
8 8 8 8 8 8
f  =   8  +1=  8 +1
3 3 3 3 3 3
16 8 32 6 16 8 32 6
= +1=1 or +1=1+
3 3 9 3 3 9

(From the graph, we see that the extreme values do not occur at the endpoints.)

66. (a)
From the graph, it appears that the absolute maximum value is about f ( 0.58)=1.47 , and the absolute
minimum value is about f ( 1)= f (0)=1.00 ; that is, at both endpoints.

3 3
x  x x  x
(b) f (x)=e 
/
f (x)=e ( 3x2 1) . So f /(x)=0 on  1,0  x= 1/3 . f ( 1)= f (0)=1 (minima)
 3 /9+ 3 /3 2 3 /9
and f (  1/3 ) =e =e (maximum).

67. (a)
From the graph, it appears that the absolute maximum value is about f (0.75)=0.32 , and the absolute
minimum value is f (0)= f (1)=0 ; that is, at both endpoints.

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

(b) f (x)=x x x 
2 /
f (x)=x
1 2x
+ x x
2
=
( x 2x ) + ( 2x 2x )
2 2
=
3x 4x
2
/
. So f (x)=0
2 2 2
2 x x 2 x x 2 x x
2 3
3x 4x =0 x(3 4x)=0  x=0 or . f (0)= f (1)=0 ,
4
3 3 3 3 2 3 3
and f =  = .
4 4 4 4 16

68. (a)
From the graph, it appears that the absolute maximum value is about f (5.76)=0.58 , and the absolute
minimum value is about f (3.67)= 0.58 .

cos x / (2+sin x)( sin x) (cos x)(cos x)  1 2sin x

(b) f (x)=  f (x)= = .
2+sin x 2 2
( 2+sin x ) ( 2+sin x )
/ 1 7 11 7  3 /2 1
So f (x)=0 sin x=  x= or . Now f = = ,
2 6 6 6 3/2 3
11 3 /2 1
and f = = .
6 3/2 3

mass 1000 3
69. The density is defined as  = = (in g / cm ). But a critical point of  will also be a
volume V (T )
d  2 dV
critical point of V since [ = 1000V and V is never 0 ], and V is easier to differentiate than
dT dT
 .
2 3
V (T )=999.87 0.06426T +0.0085043T  0.0000679T 
/ 2
V (T )= 0.06426+0.0170086T  0.0002037T . Setting this equal to 0 and using the quadratic formula
to
2
 0.0170086 0.0170086  4 0.0002037 0.06426
find T , we get T =  3.9665 C or 79.5318 C.
2( 0.0002037)

Since we are only interested in the region 0 C  T  30 C, we check the density  at the endpoints
1000 1000
and at 3.9665 C:  (0)  1.00013 ;  (30)  0.99625 ;
999.87 1003.7628

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

1000
 (3.9665)  1.000255 . So water has its maximum density at about 3.9665 C.
999.7447

 W dF ( sin  +cos  )(0)  W ( cos   sin  )   W ( cos   sin  )

70. F=  = = .
 sin  +cos  d 2 2
( sin  +cos  ) ( sin  +cos  )
dF sin 
So =0  cos   sin  =0   = =tan  . Substituting tan  for  in F gives us
d cos 
(tan  )W W tan  W tan  cos  W sin 
F= = = = =W sin  .
(tan  )sin  +cos  2 2 2 1
sin  sin  +cos 
+cos 
cos 
 
If tan  = , then sin  = (see the figure), so F= W . We compare this with the
2 2
 +1  +1
 
value of F at the endpoints: F(0)= W and F =W . Now because  1 and
2 2
 +1
 
  , we have that W
2 2
 +1  +1

 
is less than or equal to each of F(0) and F . Hence, W is the absolute minimum
2 2
 +1
value of F( ) , and it occurs when tan  = .

71. We apply the Closed Interval Method to the continuous function

5 4 3 2
I(t)=0.00009045t +0.001438t  0.06561t +0.4598t  0.6270t+99.33 on 0,10 . Its derivative is
/ 4 3 2 /
I (t)=0.00045225t +0.005752t  0.19683t +0.9196t 0.6270 . Since I exists for all t , the only
/
critical numbers of I occur when I (t)=0 . We use a root finder on a computer algebra system (or a
/
graphing device) to find that I (t)=0 when t  29.7186 , 0.8231 , 5.1309 , or 11.0459 , but only the
second and third roots lie in the interval 0,10 . The values of I at these critical numbers are
I(0.8231) 99.09 and I(5.1309) 100.67 . The values of I at the endpoints of the interval are
I(0)=99.33 and I(10) 96.86 . Comparing these four numbers, we see that food was most expensive
at t 5.1309 (corresponding roughly to August, 1989) and cheapest at t=10 (midyear 1994).

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

72. (a)
The equation of the graph in the figure is
3 2
v(t)=0.00146t  0.11553t +24.98169t 21.26872 .

/ 2 / /
(b) a(t)=v (t)=0.00438t  0.23106t+24.98169 a (t)=0.00876t 0.23106 . a (t)=0
0.23106
t =  26.4 . a(0) 24.98 , a(t ) 21.93 , and a(125) 64.54 . The maximum acceleration is
1 0.00876 1
2 2
about 64.5 ft / s and the minimum acceleration is about 21.93 ft / s .

2 2 3 / 2 / 2
73. (a) v(r)=k(r  r)r =kr r  kr  v (r)=2kr r 3kr . v (r)=0 kr(2r  3r)=0 r=0 or r (but 0
0 0 0 0 3 0
1 2 1 1 3
is not in the interval). Evaluating v at r , r , and r , we get v r = kr ,
2 0 3 0 0 2 0 8 0
2 4 3 4 1 2
v r = kr , and v(r )=0 . Since > , v attains its maximum value at r= r . This
3 0 27 0 0 27 8 3 0
supports the statement in the text.
4 3
(b) From part (a), the maximum value of v is kr .
27 0

(c)

3 / 2 /
74. g(x)=2+(x 5)  g (x)=3(x 5)  g (5)=0 , so 5 is a critical number. But g(5)=2 and g takes on
values >2 and values <2 in any open interval containing 5 , so g does not have a local maximum or
minimum at 5 .

101 51 / 100 50 /
75. f (x)=x +x +x+1 f (x)=101x +51x +1 1 for all x , so f (x)=0 has no solution. Thus,
f (x) has no critical number, so f (x) can have no local maximum or minimum.

76. Suppose that f has a minimum value at c , so f (x) f (c) for all x near c . Then
g(x)= f (x)  f (c)=g(c) for all x near c , so g(x) has a maximum value at c .

/
77. If f has a local minimum at c , then g(x)= f (x) has a local maximum at c , so g (c)=0 by the case
16
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.1 Maximum and Minimum Values

/ /
of Fermat’s Theorem proved in the text. Thus, f (c)= g (c)=0 .

78.
(a) 3 2 / 2
f (x)=ax +bx +cx+d , a
0 . So f (x)=3ax +2bx+c is a quadratic and hence has either 2 , 1 ,
or 0 real roots, so f (x) has either 2 , 1 or 0 critical numbers.

Case (ii) (1 critical

Case (i) (2 critical numbers): Case (iii) (no critical number):
number):
3
f (x)=x  3x f(x)=x  3
f (x)=x +3x
/ 2 / / 2
f (x)=3x  3 , so x= 1,1 f (x)=3x,so x=0 f (x)=3x +3 ,
are critical numbers. is the only critical number. so there are no real roots.

(b) Since there are at most two critical numbers, it can have at most two local extreme values and
by (i) this can occur. By (iii) it can have no local extreme value. However, if there is only one
critical number, then there is no local extreme value.

17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

2+1 1+1
2 x x 3 2
1. f (x)=6x  8x+3 F(x)=6  8 +3x+C=2x  4x +3x+C
2+1 1+1
/ 2 2
Check: F (x)=2 3x  4 2x+3+0=6x  8x+3= f (x)

2 3 1 3 5 4
2. f (x)=4+x  5x  F(x)=4x+ x x +C
3 4
3+1 5+1 7+1
3 5 x x 7 x 1 4 5 6 3 8
3. f (x)=1 x +5x  3x  F(x)=x +5  3 +C=x x+ x x +C
3+1 5+1 7+1 4 6 8
20 10 1 21 4 11
4. f (x)=x +4x +8 F(x)= x + x +8x+C
21 11

1/4+1 3/4+1 5/4 7/4

1/4 3/4 x x x x 5/4 7/4
5. f (x)=5x  7x  F(x)=5  7 +C=5  7 +C=4x  4x +C
1 3 5/4 7/4
+1 +1
4 4

1.7 2 3 2.7 2 10 2.7

6. f (x)=2x+3x  F(x)=x + x +C=x + x +C
2.7 9

6 1/2 1/6
7. f (x)=6 x  x =6x  x 
1/2+1 1/6+1 3/2 7/6
x x x x 3/2 6 7/6
F(x)=6  +C=6  +C=4x  x +C
1 1 3/2 7/6 7
+1 +1
2 6

7/4 7/3
4 3 3 4 3/4 4/3 x x 4 7/4 3 7/3
8. f (x)= x + x =x +x  F(x)= + +C= x + x +C
7/4 7/3 7 7

10  9
9. f (x)= =10x has domain (   ,0 )  ( 0, ) , so
9
x

{
 8
10x 5 if x<0
+C = +C
 8 1 8 1
F(x)= 4x
5
 +C if x>0
8 2
4x
See Example 1 for a similar problem.

10.

1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

3 6
5 4x +2x  6  3
g(x)= =5x  4x +2 has domain (   ,0 )  ( 0, ) , so
6
x

{
 5  2
x x 1 2 if x<0
5  4 +2x+C = + +2x+C
 5  2 1 5 2 1
G(x)= x x
1 2
 + +2x+C if x>0
5 2 2
x x

4 4 1/2
u +3 u u 3u 2  3/2
11. f (u)= = + =u +3u 
2 2 2
u u u
3  3/2+1  1/2
u u 1 3 u 1 3 6
F(u)= +3 +C= u +3 +C= u  +C
3  3/2+1 3  1/2 3 u

x 2 x  
12. f (x)=3e +7sec x F(x)=3e +7tan x+C on the interval n  ,n + .
n 2 2

13. g( )=cos   5sin   G( )=sin   5( cos  )+C=sin  +5cos  +C

sin  1 sin   
14. h( )= =  =sec  tan   H( )=sec  +C on the interval n  ,n + .
2 cos  cos  n 2 2
cos 

2  1/2  1
15. f (x)=2x+5 1 x ( ) =2x+
5

2
F(x)=x +5sin x+C
2
1 x

{
2
1 2
x +x+1 1 x +x+ln x +C if x<0
16. f (x)= =x+1+  F(x)= 2 1
x x 1 2
x +x+ln x +C if x>0
2 2

5 6
4 5 x x 5 1 6 5 1 6
17. f (x)=5x  2x  F(x)=5  2 +C=x  x +C . F(0)=4 0  0 +C=4  C=4 , so
5 6 3 3
1 6
5
F(x)=x  x +4 . The graph confirms our answer since f (x)=0 when F has a local maximum, f is
3
positive when F is increasing, and f is negative when F is decreasing.

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

2  1  1  3
18. f (x)=4 3 1+x ( ) =4
3
2
 F(x)=4x 3tan x+C . F(1)=0 4 3
4
+C=0 C=
4
 4 , so
1+x
3  1
F(x)=4x 3tan x+  4 . Note that f is positive and F is increasing on R . Also, f has smaller
4
values where the slopes of the tangent lines of F are smaller.

2 3
/ / 2 / x x 2 3
19. f (x)=6x+12x  f (x)=6 +12 +C=3x +4x +C
2 3
3 4
x x 3 4
f (x)=3 +4 +Cx+D=x +x +Cx+D
3 4

/ / 3 6 / 1 4 1 7 2 1 5 1 8
20. f (x)=2+x +x  f (x)=2x+ x + x +C  f (x)=x + x+ x +Cx+D
4 7 20 56

/ / 4/5 / 5 9/5 1 2 5 5 14/5 1 2 25 14/5

21. f (x)=1+x  f (x)=x+ x +C  f (x)= x+  x +Cx+D= x + x +Cx+D
9 2 9 14 2 126

/ / /
22. f (x)=cos x f (x)=sin x+C f (x)= cos x+Cx+D

/ / / 2 / / 3 / 4 5 1 2
23. f (t)=60t  f (t)=20t +C  f (t)=5t +Ct+D f (t)=t + Ct +Dt+E
2

/ / / / / 1 2 2 3/2 / 1 3 4 5/2
24. f (t)=t t f (t)= t  t +C  f (t)= t  t +Ct+D
2 3 6 15
1 4 8 7/2 1 2
f (t)= t  t + Ct +Dt+E
24 105 2
/ 2 2
25. f (x)=1 6x f (x)=x 3x +C . f (0)=C and f (0)=8 C=8 , so f (x)=x 3x +8 .
3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

/ 3 4 2
26. f (x)=8x +12x+3 f (x)=2x +6x +3x+C . f (1)=11+C and f (1)=6
4 2
11+C=6 C= 5 , so f (x)=2x +6x +3x 5 .
/ 1/2 3/2 3/2 5/2
27. f (x)= x (6+5x)=6x +5x  f (x)=4x +2x +C .
3/2 5/2
f (1)=6+C and f (1)=10 C=4 , so f (x)=4x +2x +4 .

/ 4  4 2  3
28. f (x)=2x 3/x =2x 3x  f (x)=x +x +C because we’re given that x>0 .
2 3
f (1)=2+C and f (1)=3 C=1 , so f (x)=x +1/x +1 .

/ 2
29. f (t)=2cos t+sec t f (t)=2sin t+tan t+C because   /2<t< /2 .
 
f =2 ( 3 /2 ) + 3 +C=2 3 +C and f =4 C=4 2 3 , so f (t)=2sin t+tan t+4 2 3 .
3 3

/
30. f (x)=3x 
 2
f (x)=
{  3/x+C
 3/x+C
1

2
if x>0
if x<0
 3/x+3 if x>0
f (1)= 3+C =0
1
C =3 ,
1

f ( 1)=3+C =0
/
2
C = 3 . So f (x)=
2 {  3/x 3 if x<0
31. f (x)=2/x f (x)=2ln x +C=2ln ( x)+C (since x<0 ). Now f ( 1)=2ln 1+C=2(0)+C=7 C=7 .
Therefore, f (x)=2ln ( x)+7 , x<0 .

/ 2 1   1 1 1  1
32. f (x)=4/ 1 x  f (x)=4sin x+C . f
=4sin +C=4 +C and f =1
2 2 6 2
2 2  1 2
+C=1 C=1 , so f (x)=4sin x+1 .
3 3 3

/ / 2 / 3 2 / /
33. f (x)=24x +2x+10
f (x)=8x +x +10x+C . f (1)=8+1+10+C and f (1)= 3 19+C= 3
/ 3 2 4 1 3 2 1
C= 22 , so f (x)=8x +x +10x 22 and hence, f (x)=2x + x +5x  22x+D . f (1)=2+ +5 22+D and
3 3
7 59 4 1 3 2 59
f (1)=5 D=22 = , so f (x)=2x + x +5x  22x+ .
3 3 3 3

/ / 3 / 2 4 / /
34. f (x)=4 6x 40x  f (x)=4x 3x  10x +C . f (0)=C and f (0)=1 C=1 , so
/ 2 4 2 3 5
f (x)=4x 3x  10x +1 and hence, f (x)=2x  x  2x +x+D . f (0)=D and f (0)=2 D=2 , so
2 3 5
f (x)=2x  x  2x +x+2 .

35.
4
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

/ / / / /
f ( )=sin  +cos   f ( )= cos  +sin  +C . f (0)= 1+C and f (0)=4 C=5 , so
/
f ( )= cos  +sin  +5 and hence, f ( )= sin   cos  +5 +D. f (0)= 1+D and f (0)=3 D=4 , so
f ( )= sin   cos  +5 +4 .

/ /  1/2 / 1/2 / / / 1/2

36. f (t)=3/ t =3t  f (t)=6t +C . f (4)=12+C and f (4)=7 C= 5 , so f (t)=6t  5 and
3/2 3/2
hence, f (t)=4t  5t+D . f (4)=32 20+D and f (4)=20 D=8 , so f (t)=4t  5t+8 .
/ / / 2 2 3
37. f (x)=2 12x f (x)=2x 6x +C  f (x)=x  2x +Cx+D .
f (0)=D and f (0)=9 D=9 . f (2)=4 16+2C+9=2C 3 and f (2)=15 2C=18 C=9 , so
2 3
f (x)=x  2x +9x+9 .

/ / 3 2 / 4 3 5 4 2
38. f (x)=20x +12x +4 f (x)=5x +4x +4x+C  f (x)=x +x +2x +Cx+D . f (0)=D and f (0)=8
5 4 2
D=8 . f (1)=1+1+2+C+8=C+12 and f (1)=5 C= 7 , so f (x)=x +x +2x  7x+8 .

/ / / 2
39. f (x)=2+cos x f (x)=2x+sin x+C f (x)=x  cos x+Cx+D . f (0)= 1+D and f (0)= 1 D=0 .
 2    2 
f = /4+ C and f =0 C=  /4 C= , so
2 2 2 2 2
2 
f (x)=x  cos x x.
2

/ / t / t t
40. f (t)=2e +3sin t f (t)=2e  3cos t+C  f (t)=2e  3sin t+Ct+D . f (0)=2+D and f (0)=0 D= 2
 
  2 2e t 2 2e
. f ( )=2e + C 2 and f ( )=0  C=2 2e  C= , so f (t)=2e  3sin t+ t 2 .
 

/ /  2 /
41. f (x)=x , x>0 f (x)= 1/x+C  f (x)= ln x +Cx+D= ln x+Cx+D
(since x>0 ). f (1)=0 C+D=0 and f (2)=0  ln 2+2C+D=0
 ln 2+2C C=0 [ since D= C ]   ln 2+C=0 C=ln 2 and D= ln 2 .
So f (x)= ln x+(ln 2)x ln 2 .

/ / / / / / /
42. f (x)=sin x f (x)= cos x+C  1= f (0)= 1+C  C=2 , so
/ / / / / 2
f (x)= cos x+2 f (x)= sin x+2x+D 1= f (0)=D f (x)= sin x+2x+1 f (x)=cos x+x +x+E
2
 1= f (0)=1+E E=0 , so f (x)=cos x+x +x .

/ 2 2
43. Given f (x)=2x+1 , we have f (x)=x +x+C . Since f passes through ( 1,6 ) , f (1)=6  1 +1+C=6
2 2
 C=4 . Therefore, f (x)=x +x+4 and f (2)=2 +2+4=10 .
5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

/ 3 1 4 / 3
44. f (x)=x  f (x)= x +C . x+y=0 y= x m= 1 . Now m= f (x)  1=x  x= 1 y=1 (from
4
1 4
the equation of the tangent line), so (  1,1 ) is a point on the graph of f . From f , 1= ( 1) +C 
4
3 1 4 3
C= . Therefore, the function is f (x)= x + .
4 4 4

45. b is the antiderivative of f . For small x , f is negative, so the graph of its antiderivative must be
decreasing. But both a and c are increasing for small x , so only b can be f ’s antiderivative. Also, f
is positive where b is increasing, which supports our conclusion.

46. We know right away that c cannot be f ’s antiderivative, since the slope of c is not zero at the x
value where f =0 . Now f is positive when a is increasing and negative when a is decreasing, so a is
the antiderivative of f .

/
47. The graph of F will have a minimum at 0 and a maximum at 2 , since f =F goes from negative
to positive at x=0 , and from positive to negative at x=2 .

48. The position function is the antiderivative of the velocity function, so its graph has to be
horizontal where the velocity function is equal to 0 .

49.

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

/
f (x)=
{ 2 if0 x<1
1 if1<x<2
 1 if2<x 3
 f (x)=
{
2x+C if0 x<1
x+D if1<x<2
 x+E if2<x 3
f (0)= 1 2(0)+C= 1 C= 1 .

Starting at the point ( 0, 1 ) and moving to the right on a line with slope 2 gets us to the point ( 1,1 ) .
The slope for 1<x<2 is 1 , so we get to the point ( 2,2 ) . Here we have used the fact that f is
continuous. We can include the point x=1 on either the first or the second part of f . The line
connecting ( 1,1 ) to ( 2,2 ) is y=x , so D=0 . The slope for 2<x 3 is  1 , so we get to ( 3,1 ) . f (3)=1
 3+E=1 E=4 . Thus,

f (x)=
{ 2x 1 if 0 x 1
x if 1<x<2
 x+4 if 2 x 3

/
Note that f (x) does not exist at x=1 or at x=2 .

50. (a)

(b) Since F(0)=1 , we can start our graph at ( 0,1 ) . f has a minimum at about x=0.5 , so its derivative
is zero there. f is decreasing on ( 0,0.5) , so its derivative is negative and hence, F is CD on ( 0,0.5)
/
and has an IP at x 0.5 . On ( 0.5,2.2 ) , f is negative and increasing ( f is positive), so F is
decreasing and CU. On ( 2.2, ) , f is positive and increasing, so F is increasing and CU.

2 2 3/2 2 3/2
(c) f (x)=2x 3 x  F(x)=x  3 x +C . F(0)=C and F(0)=1 C=1 , so F(x)=x  2x +1 .
3

(d)
7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

2
51. f (x)=sin (x ) , 0 x 4

4
52. f (x)=1/(x +1)

53.

54.

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

55.
x f (x) x f (x)
0 1 3.5  0.100
0.5 0.959 4.0  0.189
1.0 0.841 4.5  0.217
1.5 0.665 5.0  0.192
2.0 0.455 5.5  0.128
2.5 0.239 6.0  0.047
3.0 0.047

We compute slopes [values of f (x)=(sin x) / x for 0 < x < 2 ] as in the table lim f (x)=1 and
+
x
0
draw a direction field as in Example 6. Then we use the direction field to graph F starting at ( 0,0 )

56.
x f (x)
0 0
 0.2 0.041
 0.4 0.169
 0.6 0.410
 0.8 0.824
 1.0 1.557
 1.2 3.087

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

 1.4 8.117
 1.5 21.152

We compute slopes [values of f (x) = x tan x for   / 2 < x  / 2 ] as in the table and draw a
direction field as in Example 6. Then we use the direction field to graph F starting at ( 0,0 ) and
extending in both directions. Note that if f is an even function, then the antiderivative F that passes
through the origin is an odd function.]

57.

Remember that the given table values of f are the slopes of F at any x . For example, at x=1.4 , the
slope of F is f (1.4)=0 .

58. (a)

(b) The general antiderivative of f (x)=x

 2

2
is F(x)=
if x>0 {  1/x+C
1
if x<0
since f (x) is not defined
 1/x+C
at x=0 . The graph of the general antiderivatives of f (x) looks like the graph in part (a), as expected.

/
59. v(t)=s (t)=sin t cos t  s(t)= cos t sin t+C . s(0)= 1+C and s(0)=0 C=1 , so s(t)= cos t sin t+1
.

/ 3/2 3/2
60. v(t)=s (t)=1.5 t  s(t)=t +C . s(4)=8+C and s(4)=10 C=2 , so s(t)=t +2 .
10
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

/ 1 2 1 2
61. a(t)=v (t)=t 2 v(t)= t  2t+C . v(0)=C and v(0)=3 C=3 , so v(t)= t  2t+3 and
2 2
1 3 2 1 3 2
s(t)= t  t +3t+D . s(0)=D and s(0)=1 D=1 , and s(t)= t  t +3t+1 .
6 6

/
62. a(t)=v (t)=cos t+sin t  v(t)=sin t cos t+C  5=v(0)= 1+C  C=6 , so v(t)=sin t cos t+6
s(t)= cos t sin t+6t+D 0=s(0)= 1+D D=1 , so s(t)= cos t sin t+6t+1 .

/
63. a(t)=v (t)=10sin t+3cos t v(t)= 10cos t+3sin t+C s(t)= 10sin t 3cos t+Ct+D . s(0)= 3+D=0
6 6
and s(2 )= 3+2 C+D=12 D=3 and C= . Thus, s(t)= 10sin t 3cos t+ t+3 .
 

/ 2 3 2 3 2 1 3 1 4
64. a(t)=v (t)=10+3t 3t  v(t)=10t+ t  t +C  s(t)=5t + t  t +Ct+D 0=s(0)=D and
2 2 4
2 1 3 1 4
10=s(2)=20+4 4+2C C= 5 , so s(t)= 5t+5t + t  t .
2 4

65. (a) We first observe that since the stone is dropped 450 m above the ground, v(0)=0 and s(0)=450
.
/ 2
v (t)=a(t)= 9.8 v(t)= 9.8t+C . Now v(0)=0 C=0 , so v(t)= 9.8t s(t)= 4.9t +D . Last, s(0)=450
2
 D=450 s(t)=450 4.9t .
2 2
(b) The stone reaches the ground when s(t)=0 . 450 4.9t =0 t =450/4.9 t = 450/4.9 9.58 s.
1
(c) The velocity with which the stone strikes the ground is v t = 9.8 450/4.9  93.9 m / s. ( ) 1
(d) This is just reworking parts (a) and (b) with v(0)= 5 . Using v(t)= 9.8t+C , v(0)= 5 0+C= 5
2 2
v(t)= 9.8t 5 . So s(t)= 4.9t  5t+D and s(0)=450 D=450  s(t)= 4.9t  5t+450 . Solving s(t)=0 by
using the quadratic formula gives us t= ( 5 8845 ) / (  9.8 )  t 9.09 s.
1

/
66. v (t)=a(t)=a v(t)=at+C and v =v(0)=C v(t)=at+v 
0 0
1 2 1 2
s(t)= at +v t+D s =s(0)=D s(t)= at +v t+s
2 0 0 2 0 0

2 /
67. By Exercise with a= 9.8 , s(t)= 4.9t +v t+s and v(t)=s ( t ) = 9.8t+v . So
0 0 0
2 2 22 2 2 2 2 2 2
v ( t ) =  9.8t+v
( ) = ( 9.8) t  19.6v t+v =v +96.04t  19.6v t=v  19.6  4.9t +v t . But  4.9t +v t
0 0 0 0 0 0 0 0
2 2
is just s(t) without the s term; that is, s(t) s . Thus, v ( t ) =v  19.6 s ( t )  s .
0 0 0 0

11
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

2
68. For the first ball, s (t)= 16t +48t+432 from Example 8. For the second ball, a(t)= 32
1
2
v(t)= 32t+C , but v(1)= 32(1)+C=24 C=56 , so v(t)= 32t+56 s(t)= 16t +56t+D , but
2 2
s(1)= 16(1) +56(1)+D=432 D=392 , and s (t)= 16t +56t+392 . The balls pass each other when
2
2 2
s (t)=s (t)  16t +48t+432= 16t +56t+392 8t=40 t=5 s.
1 2
2 2
Another solution: From Exercise , we have s (t)= 16t +48t+432 and s (t)= 16t +24t+432 . We now
1 2
2 2
want to solve s (t)=s (t 1)  16t +48t+432= 16(t 1) +24(t 1)+432 48t=32t 16+24t 24 40=8t
1 2
 t=5 s.

69. Using Exercise with a= 32 , v =0 , and s =h (the height of the cliff ), we know that the height at
0 0
2 /
time t is s(t)= 16t +h . v(t)=s (t)= 32t and v(t)= 120  32t= 120 t=3.75 , so
2 2
0=s(3.75)= 16(3.75) +h h=16(3.75) =225 ft.

/ / 1 2 / 1 2 1 3
70. (a) EIy =mg(L x)+  g(L x)  EIy = mg(L x)   g(L x) +C 
2 2 6
1 3 1 4 /
EIy= mg(L x) +  g(L x) +Cx+D . Since the left end of the board is fixed, we must have y=y =0
6 24
1 2 1 3 1 3 1 4
when x=0 . Thus, 0= mgL   gL +C and 0= mgL +  gL +D . It follows that
2 6 6 24
1 3 1 4 1 2 1 3 1 3 1 4
EIy= mg(L x) +  g(L x) + mgL +  gL x mgL +  gL and
6 24 2 6 6 24
1 1 3 1 4 1 2 1 3 1 3 1 4
f (x)=y= mg(L x) +  g(L x) + mgL +  gL x mgL +  gL
EI 6 24 2 6 6 24
(b) f (L)<0 , so the end of the board is a distance approximately  f (L) below the horizontal. From
our result in (a), we calculate

 1 1 3 1 4 1 3 1 4
 f (L) = mgL +  gL  mgL   gL
EI 2 6 6 24
3
 1 1 3 1 4 gL m  L
= mgL +  gL = +
EI 3 8 EI 3 8

Note: This is positive because g is negative.

/ 2
71. Marginal cost =1.92 0.002x=C (x) C(x)=1.92x 0.001x +K . But C(1)=1.92 0.001+K=562
2
K=560.081 . Therefore, C(x)=1.92x 0.001x +560.081 C(100)=742.081 , so the cost of producing
12
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

100 items is $742.08 .

dm  1/2 1/2
72. Let the mass, measured from one end, be m(x) . Then m(0)=0 and  = =x  m(x)=2x +C
dx
and m(0)=C=0 , so m(x)=2 x . Thus, the mass of the 100 centimeter rod is m(100)=2 100 =20 g.

73. Taking the upward direction to be positive we have that for 0 t 10 (using the subscript 1 to
/ 2
refer to 0 t 10 ), a (t)= ( 9 0.9t ) =v (t) v (t)= 9t+0.45t +v , but v (0)=v = 10
1 1 1 0 1 0
2 / 9 2 3
v (t)= 9t+0.45t  10=s (t) s (t)= t +0.15t  10t+s . But s (0)=500=s 
1 1 1 2 0 1 0
9 2 3
s (t)= t +0.15t  10t+500 . s (10)= 450+150 100+500=100 , so it takes more than 10 seconds for
1 2 1
/ 2
the raindrop to fall. Now for t>10 , a(t)=0=v (t) v(t)= constant =v (10)= 9(10)+0.45(10)  10= 55
1
 v(t)= 55 . At 55 ft / s, it will take 100/55 1.8 s to fall the last 100 ft. Hence, the total time is
100 130
10+ = 11.8 s.
55 11

/ 50 5280 220 220

74. v (t)=a(t)= 22 . The initial velocity is 50 mi / h = = ft / s, so v(t)= 22t+ . The
3600 3 3
220 10 2 220
car stops when v(t)=0 t= = . Since s(t)= 11t + t , the distance covered is
3 22 3 3
10 10 2 220 10 1100
s = 11 +  = =122.2 ft.
3 3 3 3 9

5280
75. a(t)=k , the initial velocity is 30 mi / h =30  =44 ft / s, and the final velocity (after 5
3600
5280 220
seconds) is 50 mi / h =50  = ft / s . So v(t)=kt+C and v(0)=44 C=44 . Thus, v(t)=kt+44
3600 3
220 220 88 88 2
 v(5)=5k+44 . But v(5)= , so 5k+44=  5k=  k= 5.87 ft / s .
3 3 3 15

76. a(t)= 16 v(t)= 16t+v where v is the car’s speed (in ft / s) when the brakes were applied. The
0 0
1 1 2 2
car stops when  16t+v =0 t= v . Now s(t)= ( 16)t +v t= 8t +v t . The car travels 200 ft in the
0 16 0 2 0 0
1 1 2 1 1 2
time that it takes to stop, so s v =200 200= 8 v +v v = v 
16 0 16 0 0 16 0 32 0
2
v =32 200=6400 v =80 ft / s ( 54.54 mi / h).
0 0

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

2
77. Let the acceleration be a(t)=k km / h . We have v(0)=100 km / h and we can take the initial
position s(0) to be 0 . We want the time t for which v(t)=0 to satisfy s(t)<0.08 km. In general,
f
/ / 1 2
v (t)=a(t)=k , so v(t)=kt+C , where C=v(0)=100 . Now s (t)=v(t)=kt+100 , so s(t)= kt +100t+D ,
2
1 2
where D=s(0)=0 . Thus, s(t)= kt +100t . Since v(t )=0 , we have kt +100=0 or t = 100/k , so
2 f f f
1 100 2 100 1 1 5,000
s(t )= k  +100  =10 , 000  = . The condition s(t ) must
f 2 k k 2k k k f
5,000 5,000 2
satisfy is  <0.08  >k [ k is negative]  k< 62 , 500 km / h , or equivalently,
k 0.08
3125 2
k<  4.82 m / s .
648

2 2 3
78. (a) For 0 t 3 we have a(t)=60t  v(t)=30t +C  v(0)=0=C  v(t)=30t , so s(t)=10t +C 
3
s(0)=0=C  s(t)=10t . Note that v(3)=270 and s(3)=270 .
For 3<t 17 : a(t)= g= 32 ft / s  v(t)= 32(t 3)+C  v(3)=270=C  v(t)= 32(t 3)+270
2 2
s(t)= 16(t 3) +270(t 3)+C s(3)=270=C  s(t)= 16(t 3) +270(t 3)+270 . Note that v(17)= 178 and
s(17)=914 .
For 17<t 22 : The velocity increases linearly from  178 ft / s to  18 ft / s during this period, so
 v  18 (  178 ) 160 2
= = =32 . Thus, v(t)=32(t 17) 178 s(t)=16(t 17)  178(t 17)+914 and
 t 22 17 5
s(22)=424 ft.
For t>22 : v(t)= 18 s(t)= 18(t 22)+C . But s(22)=424=C  s(t)= 18(t 22)+424 .
Therefore, until the rocket lands, we have

{
2
30t if 0 t 3
v(t)=  32 ( t 3) +270 if 3<t 17
32(t 17) 178 if 17<t 22
 18 if t>22
and

{
3
10t if 0 t 3
2
s(t)=  16(t 3) +270(t 3)+270 if 3<t 17
2
16(t 17)  178 ( t 17) +914 if 17<t 22
 18(t 22)+424 if t>22

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.10 Antiderivatives

(b) To find the maximum height, set v(t) on 3<t 17 equal to 0 .  32(t 3)+270=0 t =11.4375 s and
1
2
the maximum height is s t = 16 t  3 +270 t  3 +270=1409.0625 ft.
( ) ( ) ( )
1 1 1
424
(c) To find the time to land, set s(t)= 18(t 22)+424=0 . Then t 22= =23.5 , so t 45.6 s.
18

5280 2
79. (a) First note that 90 mi / h =90 ft / s =132 ft / s. Then a(t)=4 ft / s  v(t)=4t+C , but
3600
132
v(0)=0 C=0 . Now 4t=132 when t= =33 s, so it takes 33 s to reach 132 ft / s. Therefore, taking
4
2
s(0)=0 , we have s(t)=2t , 0 t 33 . So s(33)=2178 ft. 15 minutes =15(60)=900 s, so for 33<t 933
we have v(t)=132 ft / s  s(933)=132(900)+2178=120 , 978 ft =22.9125 mi.
(b) As in part (a), the train accelerates for 33 s and travels 2178 ft while doing so. Similarly, it
decelerates for 33 s and travels 2178 ft at the end of its trip. During the remaining 900 66=834 s it
travels at 132 ft / s, so the distance traveled is 132 834=110 , 088 ft. Thus, the total distance is
2178+110 , 088+2178=114 , 444 ft =21.675 mi.
(c) 45 mi =45(5280)=237 , 600 ft. Subtract 2(2178) to take care of the speeding up and slowing
down, and we have 233 , 244 ft at 132 ft / s for a trip of 233 , 244/132=1767 s at 90 mi / h. The total
time is 1767+2(33)=1833s=30 min 33s=30.55 min.
(d) 37.5(60)=2250 s. 2250 2(33)=2184 s at maximum speed. 2184(132)+2(2178)=292 , 644 total
feet or 292 , 644/5280=55.425 mi.

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

2
1. f (x)=x  4x+1 , 0,4 . Since f is a polynomial, it is continuous and differentiable on R , so it is
/
continuous on 0,4 and differentiable on ( 0,4 ) . Also, f (0)=1= f (4) . f (c)=0 2c 4=0 c=2 ,
which is in the open interval ( 0,4 ) , so c=2 satisfies the conclusion of Rolle’s Theorem.
3 2
2. f (x)=x  3x +2x+5 , 0,2 . f is continuous on 0,2 and differentiable on ( 0,2 ) . Also,
/ 2 6 36 24 1
f (0)=5= f (2) . f (c)=0 3c  6c+2=0 c= =1 3 , both in ( 0,2 ) .
6 3

3. f (x)=sin 2 x ,  1,1 . f , being the composite of the sine function and the polynomial 2 x , is
continuous and differentiable on R , so it is continuous on  1,1 and differentiable on (  1,1 ) . Also,
/  1
f ( 1)=0= f (1) . f (c)=0 2 cos 2 c=0 cos 2 c=0 2 c= +2 n c= +n . If n=0 or  1 ,
2 4
1 3
then c= , is in (  1,1 ) .
4 4

4. f (x)=x x+6 ,  6,0 . f is continuous on its domain,  6, ) , and differentiable on (  6, ) , so it
/ 3c+12
is continuous on  6,0 and differentiable on (  6,0 ) . Also, f ( 6)=0= f (0) . f (c)=0 =0
2 c+6
c= 4 , which is in (  6,0 ) .
2/3 2/3 / 2  1/3 /
5. f (x)=1 x . f ( 1)=1 ( 1) =1 1=0= f (1) . f (x)= x , so f (c)=0 has no solution. This
3
/
does not contradict Rolle’s Theorem, since f (0) does not exist, and so f is not differentiable on
(  1,1 ) .
 2  2  2 /  3 /
6. f (x)=(x 1) . f (0)=(0 1) =1=(2 1) = f (2) . f (x)= 2(x 1)  f (x) is never 0 . This does not
/
contradict Rolle’s Theorem since f (1) does not exist.

f (8) f (0) 6 4 1 / 1
7. = = . The values of c which satisfy f (c)= seem to be about c=0.8 , 3.2 , 4.4 ,
8 0 8 4 4
and 6.1 .

f (7) f (1) 2 5 1 / 1
8. = = . The values of c which satisfy f (c)= seem to be about c=1.1 , 2.8 ,
7 1 6 2 2
4.6 , and 5.8 .

1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

9.
(a), 8.5 5 1 9
(b) The equation of the secant line is y 5= 8 1 (x 1) y=
2
x+ .
2

(c) / 2
f (x)=x+4/x f (x)=1 4/x .
/ 1 2 4
So f (c)=  c =8 c=2 2 , and f (c)=2 2 + =3 2 . Thus, an equation of the tangent
2 2 2
1 1
line is y 3 2 = ( x 2 2 )  y= x+2 2 .
2 2

10. (a)
It seems that the tangent lines are parallel to the secant at x  1.2 .
3 / 2
(b) The slope of the secant line is 2 , and its equation is y=2x . f (x)=x  2x f (x)=3x  2 ,
/ 2 2 3
so we solve f (c)=2 3c =4 c=  1.155 . Our estimates were off by about 0.045 in each
3
case.

2
11. f (x)=3x +2x+5 ,  1,1 . f is continuous on  1,1 and differentiable on (  1,1 ) since
/ f (b) f (a)
polynomials are continuous and differentiable on R . f (c)= 
b a
2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

f (1) f ( 1) 10 6
6c+2= = =2 6c=0 c=0 , which is in (  1,1 ) .
1 ( 1) 2

3 / f (2) f (0)
12. f (x)=x +x 1 , 0,2 . f is continuous on 0,2 and differentiable on ( 0,2 ) . f (c)= 
2 0
2 9 ( 1) 2 2 4 2 2
3c +1=  3c =5 1 c =  c= , but only is in ( 0,2 ) .
2 3 3 3

 2x
13. f (x)=e , 0,3 . f is continuous and differentiable on R , so it is continuous on 0,3 and
 6 0  6  6
f (b) f (a)
/  2c e  e  2c 1 e 1 e
differentiable on ( 0,3) . f (c)=   2e =  e =   2c=ln 
b a 3 0 6 6
 6
1 1 e
c= ln  0.897 , which is in ( 0,3) .
2 6

x / f (b) f (a)
14. f (x)= , 1,4 . f is continuous on 1,4 and differentiable on ( 1,4 ) . f (c)= 
x+2 b a
2 1

2 3 3 2
=  (c+2) =18 c= 2 3 2 .  2+3 2  2.24 is in ( 1,4 ) .
2 4 1
(c+2)

/ /
15. f (x)= x 1 . f (3) f (0)= 3 1  0 1 =1 . Since f (c)= 1 if c<1 and f (c)=1 if c>1 ,
/
f (c)(3 0)= 3 and so is never equal to 1 . This does not contradict the Mean Value Theorem since
/
f (1) does not exist.
x+1 / 1(x 1) 1(x+1)  2 /
16. f (x)= . f (2) f (0)=3 ( 1)=4 . f (x)= = . Since f (x)<0 for all x
x 1 2 2
(x 1) (x 1)
/
(except x=1 ), f (c)(2 0) is always <0 and hence cannot equal 4 . This does not contradict the Mean
Value Theorem since f is not continuous at x=1 .

3 5
17. Let f (x)=1+2x+x +4x . Then f ( 1)= 6<0 and f (0)=1>0. Since f is a polynomial, it is
continuous, so the Intermediate Value Theorem says that there is a number c between  1 and 0 such
that f (c)=0. Thus, the given equation has a real root. Suppose the equation has distinct real roots a
and b with a<b . Then f (a)= f (b)=0 . Since f is a polynomial, it is differentiable on ( a,b) and
/
continuous on a,b . By Rolle’s Theorem, there is a number r in ( a,b) such that f (r)=0 . But
/ 2 4 /
f (x)=2+3x +20x 2 for all x , so f (x) can never be 0 . This contradiction shows that the equation
can’t have two distinct real roots. Hence, it has exactly one real root.

3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

18. Let f (x)=2x 1 sin x . Then f (0)= 1<0 and f ( /2)=  2>0 . f is the sum of the polynomial 2x 1
and the scalar multiple (  1 )
sin x of the trigonometric function sin x , so f is continuous (and
differentiable) for all x . By the Intermediate Value Theorem, there is a number c in ( 0, /2 ) such that
f (c)=0 . Thus, the given equation has at least one real root. If the equation has distinct real roots a
and b with a<b , then f (a)= f (b)=0 . Since f is continuous on a,b and differentiable on ( a,b) ,
/ /
Rolle’s Theorem implies that there is a number r in ( a,b) such that f (r)=0 . But f (r)=2 cos r >0
since cos r  1 . This contradiction shows that the given equation can’t have two distinct real roots,
so it has exactly one real root.

3
19. Let f (x)=x  15x+c for x in  2,2 . If f has two real roots a and b in  2,2 , with a<b , then
f (a)= f (b)=0 . Since the polynomial f is continuous on a,b and differentiable on ( a,b) , Rolle’s
/ / 2
Theorem implies that there is a number r in ( a,b) such that f (r)=0 . Now f (r)=3r  15 . Since r is
2
in ( a,b) , which is contained in  2,2 , we have r <2 , so r <4 . It follows that
2 /
3r  15<3
4 15= 3<0 . This contradicts f (r)=0 , so the given equation can’t have two real roots in
 2,2 . Hence, it has at most one real root in  2,2 .

4
20. f (x)=x +4x+c . Suppose that f (x)=0 has three distinct real roots a , b , d where a<b<d . Then
f (a)= f (b)= f (d)=0 . By Rolle’s Theorem there are numbers c and c with a<c <b and b<c <d and
1 2 1 2
/ / /
0= f ( c ) = f ( c ) , so f
1 2
(x)=0 must have at least two real solutions. However
/ 3
( 3
) ( 2
)
0= f (x)=4x +4=4 x +1 =4 ( x+1 ) x  x+1 has as its only real solution x= 1 . Thus, f (x) can have
at most two real roots.

21. (a) Suppose that a cubic polynomial P(x) has roots a <a <a <a , so
1 2 3 4

( ) ( ) ( ) ( ) . By Rolle’s Theorem there are numbers c , c , c with a <c <a

P a =P a =P a =P a
1 2 3 4 1 2 3 1 1 2
,
/ / /
a <c <a and a <c <a and P ( c ) =P ( c ) =P ( c ) =0 . Thus, the second degree polynomial
2 2 3 3 3 4 1 2 3
/
P (x) has three distinct real roots, which is impossible.
(b) We prove by induction that a polynomial of degree n has at most n real roots. This is certainly
true for n=1 . Suppose that the result is true for all polynomials of degree n and let P(x) be a
polynomial of degree n+1 . Suppose that P(x) has more than n+1 real roots, say
a <a <a <


<a <a . Then P a =P a =


=P a ( ) ( )
=0 . By Rolle’s Theorem ( )
1 2 3 n+1 n+2 1 2 n+2
there are real numbers c ,


,c with a <c <a ,


,a <c <a and
1 n+1 1 1 2 n+1 n+1 n+2
/ / /
P c =


=P
( ) ( c ) =0 Thus, the n th degree polynomial P (x) has at least n+1 roots. This
1 n+1
contradiction shows that P(x) has at most n+1 real roots.
4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

22. (a) Suppose that f (a)= f (b)=0 where a<b . By Rolle’s Theorem applied to f on a,b there is a
/
number c such that a<c<b and f (c)=0 .
(b) Suppose that f (a)= f (b)= f (c)=0 where a<b<c . By Rolle’s Theorem applied to f (x) on a,b and
/ /
b,c there are numbers a<d<b and b<e<c with f (d)=0 and f (e)=0 . By Rolle’s Theorem applied
/ / /
to f (x) on d,e there is a number g with d<g<e such that f (g)=0 .
( n)
(c) Suppose that f is n times differentiable on R and has n+1 distinct real roots. Then f has at
least one real root.

/
23. By the Mean Value Theorem, f (4) f (1)= f (c)(4 1) for some c ( 1,4 ) . But for every c ( 1,4 )
/ /
we have f (c) 2 . Putting f (c) 2 into the above equation and substituting f (1)=10 , we get
/ /
f (4)= f (1)+ f (c)(4 1)=10+3 f (c) 10+3
2=16 . So the smallest possible value of f (4) is 16 .

/ /
24. If 3 f (x) 5 for all x , then by the Mean Value Theorem, f (8) f (2)= f (c)
( 8 2 ) for some c
in 2,8 .
( f is differentiable for all x , so, in particular, f is differentiable on ( 2,8 ) and continuous on 2,8 .
/
Thus, the hypotheses of the Mean Value Theorem are satisfied.) Since f (8) f (2)=6 f (c) and
/ /
3 f (c) 5 , it follows that 6
3 6 f (c) 6
5 18 f (8) f (2) 30.

25. Suppose that such a function f exists. By the Mean Value Theorem there is a number 0<c<2 with
/ f (2) f (0) 5 / 5
f (c)= = . But this is impossible since f (x) 2< for all x , so no such function can
2 0 2 2
exist.

26. Let h= f  g . Then since f and g are continuous on a,b and differentiable on ( a,b) , so is h , and
thus h satisfies the assumptions of the Mean Value Theorem. Therefore, there is a number c with
/ / /
a<c<b such that h(b)=h(b) h(a)=h (c)(b a) . Since h (c)<0 , h (c)(b a)<0 , so f (b) g(b)=h(b)<0
and hence f (b)<g(b) .

1
27. We use Exercise 26 with f (x)= 1+x , g(x)=1+ x , and a=0 . Notice that f (0)=1=g(0) and
2
/ 1 1 / 1
f (x)= < =g (x) for x>0 . So by Exercise 26, f (b)<g(b) 1+b <1+ b for b>0 .
2 1+x 2 2
1
Another method: Apply the Mean Value Theorem directly to either f (x)=1+ x 1+x or g(x)= 1+x
2
on 0,b .

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

28. f satisfies the conditions for the Mean Value Theorem, so we use this theorem on the interval
f (b) f ( b) /
 b,b : = f (c) for some c  (  b,b) . But since f is odd, f ( b)= f (b) . Substituting this
b ( b)
f (b)+ f (b) / f (b) /
into the above equation, we get = f (c) = f (c) .
2b b

29. Let f (x)=sin x and let b<a . Then f (x) is continuous on b,a and differentiable on ( b,a ) . By the
Mean Value Theorem, there is a number c  ( b,a ) with
/
sin a sin b= f (a) f (b)= f (c)(a b)=(cos c)(a b) . Thus, sin a sin b  cos c b a  a b . If
a<b , then sin a sin b = sin b sin a  b a = a b . If a=b , both sides of the inequality are 0 .

/ /
30. Suppose that f (x)=c . Let g(x)=cx , so g (x)=c . Then, by Corollary 7, f (x)=g(x)+d , where d is
a constant, so f (x)=cx+d .

/ / / / 2 / / 2
31. For x>0 , f (x)=g(x) , so f (x)=g (x) . For x<0 , f (x)=(1/x) = 1/x and g (x)=(1+1/x) = 1/x ,
/ /
so again f (x)=g (x) . However, the domain of g(x) is not an interval so we cannot conclude that
f  g is constant (in fact it is not).

 1  1
32. Let f (x)=2sin x cos 1 2x . Then ( 2
)
/ 2 4x 2 4x /
f (x)=  =  =0 (since x 0 ). Thus, f (x)=0 for
2 2 2 2 2
1 x 1 1 2x ( 1 x )
2x 1 x
all x ( 0,1 ) . Thus, f (x)=C on ( 0,1 ) . To find C , let x=0.5 . Thus,
 1  1  
2sin (0.5) cos (0.5)=2  =0=C . We conclude that f (x)=0 for x in ( 0,1 ) . By continuity
6 3
 1  1
of f , f (x)=0 on 0,1 . Therefore, we see that f (x)=2sin x cos ( 1 2x2) =0
 1  1
2sin x=cos ( 1 2x2) .
x 1 
33. Let f (x)=arcsin  2arctan x + . Note that the domain of f is 0, ) . Thus,
x+1 2
/ 1 ( x+1 )  ( x 1 ) 2 1 1 1
f (x)= 
=  =0 . Then
2 1+x 2 x x (x+1) x (x+1)
x 1 2 ( x+1 )
1
x+1
f (x)=C on ( 0, ) by Theorem 5. By continuity of f , f (x)=C on 0, ) . To find C , we let x=0
   x 1 
arcsin( 1) 2arctan(0)+ =C    0+ =0=C . Thus, f (x)=0 arcsin =2arctan x  .
2 2 2 x+1 2

34. Let v(t) be the velocity of the car t hours after 2:00 P.M. Then
6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.2 The Mean Value Theorem

v(1/6) v(0) 50 30 1

= =120 . By the Mean Value Theorem, there is a number c such that 0<c< with
1/6 0 1/6 6
/ /
v (c)=120 . Since v (t) is the acceleration at time t , the acceleration c hours after 2:00 P.M. is
2
exactly 120 mi / h .

35. Let g(t) and h(t) be the position functions of the two runners and let f (t)=g(t) h(t) . By
hypothesis, f (0)=g(0) h(0)=0 and f (b)=g(b) h(b)=0 , where b is the finishing time. Then by the
/ f (b) f (0)
Mean Value Theorem, there is a time c , with 0<c<b , such that f (c)= . But f (b)= f (0)=0
b 0
/ / / / / /
, so f (c)=0 . Since f (c)=g (c) h (c)=0 , we have g (c)=h (c) . So at time c , both runners have
/ /
the same speed g (c)=h (c) .

/
36. Assume that f is differentiable (and hence continuous) on R and that f (x) 1 for all x . Suppose
f has more than one fixed point. Then there are numbers a and b such that a<b , f (a)=a , and f (b)=b
. Applying the Mean Value Theorem to the function f on a,b , we find that there is a number c in
/ f (b) f (a) / b a
( a,b) such that f (c)= . But then f (c)= =1 , contradicting our assumption that
b a b a
/
f (x) 1 for every real number x . This shows that our supposition was wrong, that is, that f cannot
have more than one fixed point.

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

1. (a) f is increasing on ( 0,6 ) and ( 8,9 ) .

(b) f is decreasing on ( 6,8 ) .
(c) f is concave upward on ( 2,4 ) and ( 7,9 ) .
(d) f is concave downward on ( 0,2 ) and ( 4,7) .
(e) The points of inflection are ( 2,3) , ( 4,4.5) and ( 7,4 ) (where the concavity changes).

2. (a) f is increasing on ( 1, 3.8 ) and ( 5, 6.5) .

(b) f is decreasing on ( 0,1 ) , (  3.8,5) , (  6.5,8 ) , and ( 8,9 ) .
(c) f is concave upward on ( 0,3) and ( 8,9 ) .
(d) f is concave downward on ( 3,5) and ( 5,8 ) .
(e) The point of inflection is ( 3, 1.8 ) (where the concavity changes).

3. (a) Use the Increasing/Decreasing (I/D) Test.

(b) Use the Concavity Test.
(c) At any value of x where the concavity changes, we have an inflection point at ( x,f(x) ) .

4. (a) See the First Derivative Test.

(b) See the Second Derivative Test and the note that precedes Example 7.

/ /
5. (a) Since f (x)>0 on ( 1,5) , f is increasing on this interval. Since f (x)<0 on ( 0,1 ) and ( 5,6 ) , f
is decreasing on these intervals.
/ /
(b) Since f (x)=0 at x=1 and f changes from negative to positive there, f changes from decreasing
/ /
to increasing and has a local minimum at x=1 . Since f (x)=0 at x=5 and f changes from positive
to negative there, f changes from increasing to decreasing and has a local maximum at x=5 .

/ /
6. (a) f (x)>0 and f is increasing on ( 0,1 ) and ( 3,5) . f (x)<0 and f is decreasing on ( 1,3) and
( 5,6 ) .
/ /
(b) Since f (x)=0 at x=1 and x=5 and f changes from positive to negative at both values, f
/
changes from increasing to decreasing and has local maxima at x=1 and x=5 . Since f (x)=0 at x=3
/
and f changes from negative to positive there, f changes from decreasing to increasing and has a
local minimum at x=3 .

/ /
7. There is an inflection point at x=1 because f (x) changes from negative to positive there, and so
the graph of f changes from concave downward to concave upward. There is an inflection point at
/ /
x=7 because f (x) changes from positive to negative there, and so the graph of f changes from
concave upward to concave downward.

8. (a) f is increasing on the intervals where

1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/
f (x)>0 , namely, ( 2,4 ) and ( 6,9 ) .
/
(b) f has a local maximum where it changes from increasing to decreasing, that is, where f
/
changes from positive to negative (at x=4 ). Similarly, where f changes from negative to positive, f
has a local minimum (at x=2 and at x=6 ).
/ / /
(c) When f is increasing, its derivative f is positive and hence, f is concave upward. This
/
happens on ( 1,3) , ( 5,7) , and ( 8,9 ) . Similarly, f is concave downward when f is decreasing   
that is, on ( 0,1 ) , ( 3,5) , and ( 7,8 ) .
(d) f has inflection points at x=1 , 3 , 5 , 7 , and 8 , since the direction of concavity changes at each
of these values.

9. The function must be always decreasing and concave downward.

10. (a) The rate of increase of the population is initially very small, then gets larger until it reaches a
maximum at about t=8 hours, and decreases toward 0 as the population begins to level off.
(b) The rate of increase has its maximum value at t=8 hours.
(c) The population function is concave upward on ( 0,8 ) and concave downward on ( 8,18 ) .
(d) At t=8 , the population is about 350 , so the inflection point is about ( 8,350 ) .

3 / 2
11. (a) f (x)=x  12x+1 f (x)=3x  12=3(x+2)(x 2) .
/
We don’t need to include "3" in the chart to determine the sign of f (x) .
/
Interval x+2 x 2 f (x) f
x< 2   + increasing on (  , 2)
 2<x<2 +   decreasing on ( 2,2)
x>2 + + + increasing on (2, )

So f is increasing on (   , 2 ) and ( 2, ) and f is decreasing on (  2,2 ) .

(b) f changes from increasing to decreasing at x= 2 and from decreasing to increasing at x=2 . Thus,
f ( 2)=17 is a local maximum value and f (2)= 15 is a local minimum value.

/ / / / / /
(c) f (x)=6x . f (x)>0 x>0 and f (x)<0 x<0 . Thus, f is concave upward on ( 0, ) and
2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

concave downward on (   ,0 ) . There is an inflection point where the concavity changes, at

( 0,f(0) ) = ( 0,1 ) .
2 3 / 2 / /
12. (a) f (x)=5 3x +x  f (x)= 6x+3x =3x(x 2) . Thus, f (x)>0 x<0 or x>2 and f (x)<0
0<x<2 . So f is increasing on (   ,0 ) and ( 2, ) and f is decreasing on ( 0,2 ) .
(b) f changes from increasing to decreasing at x=0 and from decreasing to increasing at x=2 . Thus,
f (0)=5 is a local maximum value and f (2)=1 is a local minimum value.
/ / / / / /
(c) f (x)= 6+6x=6(x 1) . f (x)>0 x>1 and f (x)<0 x<1 . Thus, f is concave upward on
( 1, ) and concave downward on (   ,1 ) . There is an inflection point at ( 1,3) .

13. (a) f (x)=x  2x +3

4 2 / 3
(
f (x)=4x  4x=4x x  1 =4x ( x+1 ) ( x 1 ) .
2
)
/
Interval x+1 x x 1 f (x) f
x< 1     decreasing on (  , 1)
 1<x<0 +   + increasing on ( 1,0)
0<x<1 + +   decreasing on (0,1)
x>1 + + + + increasing on (1, )

So f is increasing on ( 1,0) and (1, ) and f is decreasing on (  , 1) and (0,1) .

(b) f changes from increasing to decreasing at x=0 and from decreasing to increasing at x= 1 and
x=1 . Thus, f (0)=3 is a local maximum value and f ( 1)=2 are local minimum values.
/ / 2 2 1 / /
(c) f (x)=12x  4=12 x  =12 ( x+1/ 3 ) ( x 1/ 3 ) . f (x)>0 x< 1/ 3 or x>1/ 3 and
3
/ /
f (x)<0  1/ 3 <x<1/ 3 . Thus, f is concave upward on (   , 3 /3) and ( 3 /3, ) and
22
concave downward on (  3 /3, 3 /3) . There are inflection points at  3 /3, .
9

14. (a) f (x)=

x
2
 f
/
(x)=
( x +3) (2x) x (2x)
2 2
=
6x
. The denominator is positive so the sign
2 2
x +3 ( x2+3) ( x2+3) 2
/ / /
of f (x) is determined by the sign of x . Thus, f (x)>0 x>0 and f (x)<0 x<0 . So f is increasing
on ( 0, ) and f is decreasing on (   ,0 ) .
(b) f changes from decreasing to increasing at x=0 . Thus, f (0)=0 is a local minimum value.
(c)
2
/ /
2
x +3 (6) 6x 2 x +3 (2x) ( 2
) (
6 x +3
2
) 2
x +3 4x
2

f (x) = 2 2
=
( x +3)
2
( x2+3) 4
3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

=
( 2) =  18(x+1)(x 1)
6 3 3x
.
( x2+3) 3 ( x2+3) 3
/ / / /
f (x)>0  1<x<1 and f (x)<0x< 1 or x>1 . Thus, f is concave upward on (  1,1 ) and
1
concave downward on (   , 1 ) and ( 1, ) . There are inflection points at  1, .
4

/ / 1  5
15. (a) f (x)=x 2sin x on ( 0,3 )  f (x)=1 2cos x . f (x)>0 1 2cos x>0 cos x<  <x<
2 3 3
7 / 1  5 7  5
or <x<3 . f (x)<0 cos x>  0<x< or <x< . So f is increasing on ,
3 2 3 3 3 3 3
7  5 7
and ,3 , and f is decreasing on 0, and , .
3 3 3 3
5 
(b) f changes from increasing to decreasing at x= , and from decreasing to increasing at x=
3 3
7 5 5
and at x= . Thus, f = + 3 6.97 is a local maximum value and
3 3 3
  7 7
f =  3  0.68 and f =  3 5.60 are local minimum values.
3 3 3 3
/ / / /
(c) f (x)=2sin x >0 0<x< and 2 <x<3 , f (x)<0  <x<2 . Thus, f is concave upward on
( 0, ) and ( 2 ,3 ) , and f is concave downward on ( ,2 ) . There are inflection points at ( , ) and
( 2 ,2 ) .
2 /
16. (a) f (x)=cos x 2sin x , 0
x
2 . f (x)= 2cos x sin x 2cos x= 2cos x ( 1+sin x ) . Note that
1+sin x  0 [ since sin x  1 ], with equality  sin x= 1 x=3 /2 [ since 0
x
2 ]  cos x=0 .
/ /
Thus, f (x)>0 cos x <0  /2<x<3 /2 and f (x)<0 cos x>0 0<x< /2 or 3 /2<x<2 . Thus, f
is increasing on ( /2,3 /2 ) and f is decreasing on ( 0, /2 ) and ( 3 /2,2 ) .
(b) f changes from decreasing to increasing at x= /2 and from increasing to decreasing at x=3 /2 .
Thus, f ( /2)= 2 is a local minimum value and f (3 /2)=2 is a local maximum value.
(c)
f
/ / 2
(x) =2sin x ( 1+sin x )  2cos x=2sin x+2sin x 2 1 sin x
2
( 2
)
2
=4sin x+2sin x 2=2(2sin x 1)(sin x+1)

/ / 1  5 / / 1 
so f (x)>0 sin x >  <x< , and f (x)<0 sin x < and sin x  1 0<x< or
6 2 6 2 6
5 3 3  5
<x< or <x<2 . Thus, f is concave upward on , and concave downward on
6 2 2 6 6
4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

 5 3 3  1
0, , , , and ,2 . There are inflection points at , and
6 6 2 2 6 4
5 1
, .
6 4

x / x x x /
17. (a) y= f (x)=xe  f (x)=xe +e =e (x+1) . So f (x)>0 x+1>0 x> 1 . Thus, f is increasing on
(  1, ) and decreasing on (   , 1 ) .
 1
(b) f changes from decreasing to increasing at its only critical number, x= 1 . Thus, f ( 1)= e is a
local minimum value.
/ x / / x x x / /
(c) f (x)=e (x+1) f (x)=e (1)+(x+1)e =e (x+2) . So f (x)>0 x+2>0 x> 2 . Thus, f is
concave upward on (  2, ) and concave downward on (   , 2 ) . Since the concavity changes
 2
direction at x= 2 , the point  2, 2e ( ) is an inflection point.
2 x / 2 x x x /
18. (a) y= f (x)=x e  f (x)=x e +2xe =x(x+2)e . So f (x)>0 x(x+2)>0 either x< 2 or x>0 .
Therefore f is increasing on (   , 2 ) and ( 0, ) , and decreasing on (  2,0 ) .
 2
(b) f changes from increasing to decreasing at x= 2 , so f ( 2)=4e is a local maximum value. f
changes from decreasing to increasing at x=0 , so f (0)=0 is a local minimum value.
/
(c) f (x)= x +2x e  ( 2
) x
f
/ /
( 2
(x)= x +2x e +e (2x+2)=e ) x x x
( x2+4x+2) . f / /(x)=0 2
x +4x+2=0
/ /
x= 2 2 . f (x)<0  2 2 <x< 2+ 2 , so f is concave downward on (  2 2 , 2+ 2 ) and
concave upward on (   , 2 2 ) and (  2+ 2 , ) . There are inflection points at
(  2 2 ,f (  2 2 ) )  (  3.41,0.38 ) and (  2+ 2 ,f (  2+ 2 ) )  (  0.59,0.19 ) .
ln x
19. (a) y= f (x)= . (Note that f is only defined for x>0 .)
x
1  1/2 1 ln x
x (1/x) ln x x 
/ 2 x 2 x 2 x 2 ln x 2
f (x)= = = >0 2 ln x>0  ln x<2 x<e .
x x 2 x 3/2
2x
Therefore f is increasing on 0,e ( 2) and decreasing on ( e2, ) .
2
2
f changes from increasing to decreasing at x=e , so f ( e ) =
2 2 ln e
(b) = is a local maximum
e 2
e
value.
(c)

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ /
3/2
2x ( 1/x) (2 ln x) 3x ( 1/2
) =  2x1/2+3x1/2(ln x 2)
f (x) =
( 2x3/2) 2 4x
3

1/2
x ( 2+3ln x 6) 3ln x 8
= =
3 5/2
4x 4x

/ / 8 8/3 / / 8/3 8/3

f (x)=0 ln x=  x=e .f (x)>0 x>e , so f is concave upward on (e , ) and concave
3
8/3 8/3 8  4/3
downward on (0,e ) . There is an inflection point at e , e  ( 14.39,0.70 ) .
3

20. (a) y= f (x)=xln x . (Note that f is only defined for x>0 .)

/ /  1
f (x)=x(1/x)+ln x=1+ln x . f (x)>0 ln x+1>0 ln x> 1 x>e . Therefore f is increasing on
( 1/e, ) and decreasing on ( 0,1/e ) .
(b) f changes from decreasing to increasing at x=1/e , so f (1/e)= 1/e is a local minimum value.
/ /
(c) f (x)=1/x>0 for x>0 . So f is concave upward on its entire domain, and has no inflection point.

21. f (x)=x  5x+3

5 / 4
(
f (x)=5x  5=5 x +1 (x+1)(x 1) .
2
)
/ / /
First Derivative Test: f (x)<0  1<x<1 and f (x)>0 x>1 or x< 1 . Since f changes from
/
positive to negative at x= 1 , f ( 1)=7 is a local maximum value; and since f changes from negative
to positive at x=1 , f (1)= 1 is a local minimum value.
/ / 3 / / /
Second Derivative Test: f (x)=20x . f (x)=0 x= 1 . f ( 1)= 20<0 f ( 1)=7 is a local
/ /
maximum value. f (1)=20>0 f (1)= 1 is a local minimum value.
Preference: For this function, the two tests are equally easy.

22. f (x)=
x
 f
/
(x)=
( x +4) 1 x(2x)
2
=
4 x
2
=
(2+x)(2 x)
.
2 2 2 2
x +4 2
x +4
2
() ( x +4)
2

x +4
/ / /
First Derivative Test: f (x)>0  2<x<2 and f (x)<0 x>2 or x< 2 . Since f changes from
1 /
positive to negative at x=2 , f (2)= is a local maximum value; and since f changes from negative
4
1
to positive at x= 2 , f ( 2)= is a local minimum value. Second Derivative Test:
4

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

2
/ /
2
x +4 ( 2x) ( 4 x2) 2 ( x2+4) (2x)
f (x) = 2
( x2+4) 2
 2x ( x +4) ( x +4) +2 ( 4 x )  2x ( 12 x )
2 2 2 2
= =
4
( x +4)
2
( x2+4) 3
/ / / 1 1
f (x)=0 x= 2 . f ( 2)= >0 f ( 2)= is a local minimum value.
16 4
/ / 1 1
f (2)=
<0 f (2)= is a local maximum value.
16 4
Preference: Since calculating the second derivative is fairly difficult, the First Derivative Test is
easier to use for this function.

/ 1 1  1/2
23. f (x)=x+ 1 x  f (x)=1+ (1 x) ( 1)=1
. Note that f is defined for 1 x 0 ; that is,
2 1 x 2
/ 1 1 3 /
for x
1 . f (x)=0 2 1 x =1 1 x =  1 x=  x= . f does not exist at x=1 , but we
2 4 4
can’t have a local maximum or minimum at an endpoint.
/ 3 / 3 /
First Derivative Test: f (x)>0 x< and f (x)<0 <x<1 . Since f changes from positive to
4 4
3 3 5
negative at x= , f = is a local maximum value.
4 4 4
/ / 1 1  3/2 1 / / 3
Second Derivative Test: f (x)=  (1 x) ( 1)= .f = 2<0
2 2 3 4
4 ( 1 x )
3 5
f = is a local maximum value.
4 4
Preference: The First Derivative Test may be slightly easier to apply in this case.

4 3 / 4 2 3 3 3 2 3 2
24. (a) f (x)=x (x 1)  f (x)=x 3(x 1) +(x 1) 4x =x (x 1) 3x+4(x 1) =x (x 1) (7x 4)
4
The critical numbers are 0 , 1 , and .
7
(b)
/ / 2 2 3 3 2
f (x) =3x (x 1) (7x 4)+x 2(x 1)(7x 4)+x (x 1) 7
2
=x (x 1) 3(x 1)(7x 4)+2x(7x 4)+7x(x 1)

/ / / /
Now f (0)= f (1)=0 , so the Second Derivative Test gives no information for x=0 or x=1 .

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ / 4 4 2 4 4 4 4 2 3 3
f =  1 0+0+7  1 =  (4)  >0 , so
7 7 7 7 7 7 7 7
4
there is a local minimum at x= .
7
/ 4 4
(c) f is positive on (   ,0 ) , negative on 0, , positive on ,1 , and positive on ( 1, ).
7 7
4
So f has a local maximum at x=0 , a local minimum at x= , and no local maximum or minimum at
7
x=1 .

/ / /
25. (a) By the Second Derivative Test, if f (2)=0 and f (2)= 5<0 , f has a local maximum at x=2
.
/ / /
(b) If f (6)=0 , we know that f has a horizontal tangent at x=6 . Knowing that f (6)=0 does not
provide any additional information since the Second Derivative Test fails. For example, the first and
4 4 3
second derivatives of y= ( x 6 ) , y= ( x 6 ) , and y= ( x 6 ) all equal zero for x=6 , but the first has a
local minimum at x=6 , the second has a local maximum at x=6 , and the third has an inflection point
at x=6 .

/
26. f (x)>0 for all x 1 with vertical asymptote x=1 , so f is increasing on (   ,1 ) and ( 1, ).
/ / / /
f (x)>0 if x<1 or x>3 , and f (x)<0 if 1<x<3 , so f is concave upward on (   ,1 ) and ( 3, ),
and concave downward on ( 1,3) . There is an inflection point when x=3 .

/ / / /
27. f (0)= f (2)= f (4)=0 horizontal tangents at x=0 , 2 , 4 . f (x)>0 if x<0 or 2<x<4 f is
/
increasing on (   ,0 ) and ( 2,4 ) . f (x)<0 if 0<x<2 or x>4 f is decreasing on ( 0,2 ) and ( 4, ).
/ / / /
f (x)>0 if 1<x<3 f is concave upward on ( 1,3) . f (x)<0 if x<1 or x>3 f is concave
downward on (   ,1 ) and ( 3, ) . There are inflection points when x=1 and 3 .

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

28.

/ / /
f (1)= f ( 1)=0 horizontal tangents at x= 1 . f (x)<0 if x <1 f is decreasing on (  1,1 ) .
/ /
f (x)>0 if 1< x <2 f is increasing on (  2, 1 ) and ( 1,2 ) . f (x)= 1 if x >2 the graph of f has
/ /
constant slope  1 on (   , 2 ) and ( 2, ).f (x)<0 if  2<x<0 f is concave downward on (  2,0 )
. Inflection point ( 0,1 ) .
/ /
29. f (x)>0 if x <2 f is increasing on (  2,2 ) . f (x)<0 if x >2 f is decreasing on (   , 2 )
/ /
and ( 2, ) . f ( 2)=0 horizontal tangent at x= 2 . lim f (x) =  there is a vertical
x 2
/ /
asymptote or vertical tangent (cusp) at x=2 . f (x)>0 if x 2 f is concave upward on (   ,2 ) and
( 2, ) .

30.

/ /
f (x)>0 if x <2 f is increasing on (  2,2 ) . f (x)<0 if x >2 f is decreasing on (   , 2 ) and
/
( 2, ) . f (2)=0 , so f has a horizontal tangent (and local maximum) at x=2 . lim f (x)=1 y=1 is a
x 
horizontal asymptote. f ( x)= f (x) f is an odd function (its graph is symmetric about the origin).
/ / / /
Finally, f (x)<0 if 0<x<3 and f (x)>0 if x>3 , so f is CD on ( 0,3) and CU on ( 3, ).
/
31. (a) f is increasing where f is positive, that is, on ( 0,2 ) , ( 4,6 ) , and ( 8, ) ; and decreasing
/
where f is negative, that is, on ( 2,4 ) and ( 6,8 ) .
/
(b) f has local maxima where f changes from positive to negative, at x=2 and at x=6 , and local
/
minima where f changes from negative to positive, at x=4 and at x=8 .
9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/
(c) f is concave upward (CU) where f is increasing, that is, on ( 3,6 ) and ( 6, ) , and concave
/
downward (CD) where f is decreasing, that is, on ( 0,3) .
(d) There is a point of in.ection where f changes from being CD to being CU, that is, at x = 3.

(e)

/ /
32. (a) f is increasing where f is positive, on ( 1,6 ) and ( 8, ) , and decreasing where f is
negative, on ( 0,1 ) and ( 6,8 ) .
/
(b) f has a local maximum where f changes from positive to negative, at x=6 , and local minima
/
where f changes from negative to positive, at x=1 and at x=8 .
/
(c) f is concave upward where f is increasing, that is, on ( 0 2 ) , ( 3, 5) ; and ( 7,  ) and
/
concave downward where f is decreasing, that is, on ( 2 3) , ( 5, 7) .
(d) There are points of inflection where f changes its direction of concavity, at x=2 , x=3 , x=5 and
x=7 .

(e)

3
33. (a) f (x)=2x  3x  12x
2 / 2
( 2
)
f (x)=6x  6x 12=6 x  x 2 =6(x 2)(x+1) . f (x)>0
/
x< 1 or x>2
/
and f (x)<0  1<x<2 . So f is increasing on (   , 1 ) and ( 2, ) , and f is decreasing on (  1,2 ) .
(b) Since f changes from increasing to decreasing at x= 1 , f ( 1)=7 is a local maximum value. Since
f changes from decreasing to increasing at x=2 , f (2)= 20 is a local minimum value.
/ / / / 1 / / 1
(c) f (x)=6(2x 1) f (x)>0 on , and f (x)<0 on   , . So f is concave
2 2
1 1
upward on , and concave downward on   , . There is a change in concavity at
2 2
1 1 13
x= , and we have an inflection point at , .
2 2 2
(d)
10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

3 / 2
( 2
34. (a) f (x)=2+3x x  f (x)=3 3x = 3 x  1 = 3(x+1)(x 1) . f (x)>0  1<x<1 and f (x)<0 ) / /

x< 1 or x>1 . So f is increasing on (  1,1 ) and f is decreasing on (   , 1 ) and ( 1, ) .

(b) f ( 1)=0 is a local minimum value and f (1)=4 is a local maximum value.
(c) / / / / / /
f (x)= 6x f (x)>0 on (   ,0 ) and f (x)<0 on ( 0, ) . So f is concave upward on
(   ,0 ) and concave downward on ( 0, ) . There is an inflection point at ( 0,2 ) .
(d)

35. (a) f (x)=x  6x 

4 2 / 3
f (x)=4x  12x=4x x  3 =0 when x=0 ,  ( 2
) 3 .

Interval 2 /
4x x 3 f (x) f
| x< 3  +  decreasing on (   , 3)
 3 <x<0   + increasing on (  3,0 )
0<x< 3 +   decreasing on ( 0, 3 )
x> 3 + + + increasing on ( 3, )

(b) Local minimum values f (  3 ) = 9 , local maximum value f (0)=0

/ / 2
(c) f (x)=12x  12=12 x  1 >0 x >1 x >1 ( 2
) 2
x>1 or x< 1 , so f is CU on (   , 1 ) , ( 1, )
and CD on (  1,1 ) . Inflection points at (  1, 5)

(d)

11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

3 4 / 2 3 2 /
36. (a) g(x)=200+8x +x  g (x)=24x +4x =4x (6+x)=0 when x= 6 and when x=0 . g (x)>0 x> 6
/
( x 0 ) and g (x)<0 x< 6 , so g is decreasing on (   , 6 ) and g is increasing on (  6, ) , with a
horizontal tangent at x=0 .
(b) g( 6)= 232 is a local minimum value. There is no local maximum value.
/ / 2 / /
(c) g (x)=48x+12x =12x(4+x)=0 when x= 4 and when x=0 . g (x)>0 x< 4 or x>0 and
/ /
g (x)<0  4<x<0 , so g is CU on (   , 4 ) and ( 0, ) , and g is CD on (  4,0 ) . Inflection points at
(  4, 56 ) and ( 0,200 )

(d)

37. (a) h(x)=3x  5x +3

5 3 / 4
h (x)=15x  15x =15x
2 2
( x2 1) =0 when x=0 ,  1 . Since 15x is
2

/ 2
nonnegative, h (x)>0 x >1  x >1  x>1 or x< 1 , so h is increasing on (   , 1 ) and ( 1, ) and
decreasing on (  1,1 ) , with a horizontal tangent at x=0 .
(b) Local maximum value h( 1)=5 , local minimum value h(1)=1
(c)
h
/ / 3
(x) =60x  30x=30x 2x  1 ( 2
)
1 1
=60x x+ x 
2 2

/ / 1 1 1 1
h (x)>0 when x> or  <x<0 , so h is CU on  ,0 and , and CD on
2 2 2 2
1 1 1 7
  , and 0, . Inflection points at ( 0,3) and  ,3 2 .
2 2 2 8
(d)

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

( 2
)3
38. (a) h(x)= x  1  h (x)=6x x  1  0
/
( 2
)2 x>0 ( x 1 ), so h is increasing on ( 0, ) and
decreasing on (   ,0 ) .
(b) h(0)= 1 is a local minimum value.
1
(c) h
/ /
( 2
(x)=6 x  1 +24x )2 2
( x2 1) =6 ( x2 1) ( 5x2 1) . The roots  1 and 
5
divide R into five
intervals.

Interval 2 2 / / Concavity
x 5x  1 h ( x)
x< 1 + + + upward
1
 1<x<  +  downward
5
1 1
 <x<   + upward
5 5
1
<x<1  +  downward
5
x>1 + + + upward

1 1
From the table, we see that h is CU on (   , 1 ) ,  , and ( 1, ) , and CD on
5 5
1 1 1 64
 1, and ,1 . Inflection points at (  1,0 ) and  ,
5 5 5 125

(d)

/ 1  1/2 x x+2(x+3) 3x+6

39. (a) A(x)=x x+3  A (x)=x (x+3) + x+3 1= + x+3 = = .
2 2 x+3 2 x+3 2 x+3
/ /
The domain of A is  3, ) . A (x)>0 for x> 2 and A (x)<0 for  3<x< 2 , so A is increasing on
13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(  2, ) and decreasing on (  3, 2 ) .

(b) A( 2)= 2 is a local minimum value.
1
2 x+3 3 (3x+6)
/ / x+3 6(x+3) (3x+6) 3x+12 3(x+4) / /
(c) A (x)= = = = A (x)>0 for all
2 3/2 3/2 3/2
( 2 x+3 ) 4(x+3) 4(x+3) 4(x+3)
x> 3 , so A is concave upward on (  3, ) . There is no inflection point.

(d)

3
2/3 /  1/3 2 2 x / /
40. (a) B(x)=3x  x B (x)=2x  1= 3
 1= 3
. B (x)>0 if 0<x<8 and B (x)<0 if x<0 or
x x
x>8 , so B is decreasing on (   ,0 ) and ( 8, ) , and B is increasing on ( 0,8 ) .
(b) B(0)=0 is a local minimum value.
B(8)=4 is a local maximum value.
(c) / / 2  4/3  2 / /
B (x)= x = , so B (x)<0 for all x 0 . B is concave downward on (   ,0 ) and
3 4/3
3x
( 0, ) . There is no inflection point.
(d)

1/3 4/3 1/3 / 4 1/3 4  2/3 4  2/3 4(x+1) /

41. (a) C(x)=x (x+4)=x +4x  C (x)= x + x = x (x+1)= . C (x)>0 if
3 3 3 3 2
3 x
/
 1<x<0 or x>0 and C (x)<0 for x< 1 , so C is increasing on (  1, ) and C is decreasing on (   , 1 )
.
(b) C( 1)= 3 is a local minimum value.
/ / 4  2/3 8  5/3 4  5/3 4(x 2) / / / /
(c) C (x)= x  x = x ( x 2 ) = . C (x)<0 for 0<x<2 and C (x)>0 for x<0
9 9 9 3 5
9 x
and x>2 , so C is concave downward on ( 0,2 ) and concave upward on (   ,0 ) and ( 2, ) . There are
(
inflection points at ( 0,0 ) and 2,6 2  ( 2,7.56 ) .
3
)
(d)

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

3
4x
42. (a) f (x)=ln x +27  ( 4
) /
f (x)= 4
/ /
. f (x)>0 if x>0 and f (x)<0 if x<0 , so f is increasing on
x +27
( 0, ) and f is decreasing on (   ,0 ).
(b) f (0)=ln 27 3.3 is a local minimum value.
(c)

/ /
( x +27) ( 12x )  4x ( 4x ) 4x 3 ( x +27)  4x
4 2 3 3 2 4 4

f (x) = 2
=
( x +27)
4
( x4+27) 2
4x ( 81 x )  4x ( x +9 ) (x+3)(x 3)
2 4 2 2
= =
2
( x +27)
4
( x4+27) 2
/ / / /
f (x)>0 if  3<x<0 and 0<x<3 , and f (x)<0 if x< 3 or x>3 . Thus, f is concave upward on
(  3,0 ) and ( 0,3) and f is concave downward on (   , 3) and ( 3, ) . There are inflection points at
(  3,ln 108 )  (  3,4.68 ) .

(d)

43. (a) f ( )=2cos   cos 2 , 0


2 .
/
f ( )= 2sin  +2sin 2 = 2sin  +2(2sin  cos  )=2sin  (2cos   1) .

/
Interval sin  2cos   1 f ( ) f
 
0< < + + + increasing on (0, )
3 3
 
< < +   decreasing on ( , )
3 3

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

5 5
 < <   + increasing on ( , )
3 3
5 5
< <2  +  decreasing on ( ,2 )
3 3

 3 5 3
(b) f = and f = are local maximum values and f ( )= 3 is a local minimum
3 2 3 2
value.
/
(c) f ( )= 2sin  +2sin 2 
/ / 2
f ( ) = 2cos  +4cos 2 = 2cos  +4(2cos   1)
2
=2(4cos   cos   2)
/ / 1 33  1 1 33  1 1+ 33
f ( )=0 cos  =   =cos   =cos  0.5678 ,
8 8 8
 1 1+ 33  1 1 33  1 1 33
2  cos  5.7154 , cos  2.2057 , or 2  cos  4.0775 .
8 8 8
Denote these four values of  by  ,  ,  , and  , respectively. Then f is CU on 0, , CD ( )
1 4 2 3 1
on  ( , ),
1 2
CU on  ( , ) , CD on ( , ) , and CU on ( ,2 ) . To find the exact y  coordinate for  = ,
2 3 3 4 4 1
2
2 1+ 33 1+ 33
we have f ( )=2cos   cos 2 =2cos   2cos   1 =2  2 +1
1 1 1 1 1 8 8
1 1 1 1 33 3 3 3
= +
4 4
33  
32 16
33 
32
+1= +
16 16
33 =
16
( 1+ 33 ) =y1 1.26 . Similarly,
3
f ( )=
2 16
( 1 33 ) =y2  0.89 . So f has inflection points at  1,y1 ,  2,y2 ,  3,y2 , and ( ) ( ) ( )
( ,y . )
4 1

(d)

/ / 3
44. (a) f (t)=t+cos t ,  2
t
2  f (t)=1 sin t 0 for all t and f (t)=0 when sin t=1 t= or
2

, so f is increasing on (  2 ,2 ) .
2
16
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(b) No maximum or minimum

/ / 3   3
(c) f (t)= cos t > 0  t  ,   , , so f is CU on these intervals and
2 2 2 2
3   3
CD 0n  2 ,  ,  , , and , 2 . Points of inflection at
2 2 2 2
3 3  
 ,  and  , 
2 2 2 2

(d)

2 2
x x
45. f (x)= = has domain (   , 1 )  (  1,1 )  ( 1, ).
2 (x+1)(x 1)
x 1
2 2
x /x 1 1
(a) lim f (x)= lim = lim = =1 , so y=1 is a HA.
x   x ( x2 1) /x2 x  
  1 1/x
2 1 0
2
x 
since x 1 and ( x  1 )
2 2 +
lim 2
= 0 as x  1 , so x= 1 is a VA.

x  1 x 1
2
x
lim 2
= since x
2
1 and x  1 ( 2
) +
0 as x
+
1 , so x=1 is a VA.
x 1
+
x 1

(b) f (x)=
x
2
 f
/
(x)=
( x  1 ) (2x) x (2x) 2x ( x  1 )  x
2 2
=
2 2
=
 2x
(
. Since x  1
2
) 2 is
2
x 1 ( x2 1) 2 ( x2 1) 2 ( x2 1) 2
positive for all x in the domain of f , the sign of the derivative is determined by the sign of  2x .
/ /
Thus, f (x)>0 if x<0 ( x  1 ) and f (x)<0 if x>0 ( x 1 ) . So f is increasing on (   , 1 ) and (  1,0 ) ,
and f is decreasing on ( 0,1 ) and ( 1, ) .
(c) /
f (x)=0 x=0 and f (0)=0 is a local maximum value.

17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(d) 2
/ / =
( x
2
 1 ) ( 2) ( 2x) 2 ( x
2
 1 ) (2x)
f (x)
( ( x2 1) 2) 2
2 ( x  1 ) (  ( x  1 ) +4x ) 2 ( 3x +1 )
2 2 2 2
= = .
4
( x  1)
2
( x2 1) 3
/ / / /
The sign of f (x) is determined by the denominator; that is, f (x)>0 if x >1 and
/ /
f (x)<0 if x <1 . Thus, f is CU on (   , 1 ) and ( 1, ) , and f is CD on (  1,1 ) . There are
no inflection points.
(e)

2
x
46. f (x)= has domain (   ,2 )  ( 2, ).
2
(x 2)
2 2 2
x x /x 1 1
(a) lim = lim = lim = =1 ,
x  
2
x  4x+4 x   ( x2 4x+4) /x2 x   1 4/x+4/x
2 1 0+0
2
x 2 2 + +
so y=1 is a HA. lim = since x 4 and (x 2) 0 as x 2 , so x=2 is a VA.
+ 2
x 2 (x 2)
2 2 2
x / ( x 2 ) (2x) x 2(x 2) 2x(x 2)[(x 2) x]  4x /
(b) f (x)=  f (x)= = = . f (x)>0 if 0<x<2
2 2 2 4 3
(x 2) ( x 2 ) (x 2) (x 2)
/
and f (x)<0 if x<0 or x>2 , so f is increasing on ( 0,2 ) and f is decreasing on (   ,0 ) and ( 2, ).

(c) f (0)=0 is a local minimum value.

18
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(d) 3 2
/ / (x 2) ( 4) ( 4x) 3(x 2)
f (x) = 3 2
(x 2)
2
4(x 2) [ (x 2)+3x] 8(x+1)
= =
6 4
(x 2) (x 2)

/ / / /
f (x)>0 if x> 1 ( x 2 ) and f (x)<0 if x< 1 . Thus, f is CU on (  1,2 ) and ( 2, ) , and f
1
is CD on (   , 1 ) . There is an inflection point at  1, .
9
(e)

47. (a) lim

x  
( 2
x +1  x = ) and

2
lim
x 
( 2
x +1  x =lim ) x 
( 2
x +1  x ) x +1 +x
2
=lim
x 
1
2
=0 , so y=0 is a HA.
x +1 +x x +1 +x
2 / x x /
(b) f (x)= x +1  x f (x)=  1 . Since <1 for all x , f (x)<0 , so f is decreasing
2 2
x +1 x +1
on R .
(c) No minimum or maximum

19
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(d)
/ /
( x2+1) 1/2 (1) x 12 ( x2+1)  1/2 (2x)
f (x) = 2
( 2
x +1 )
2
1/2
( x +1)
2

x

=
( x2+1) 1/2 = ( x2+1)  x2
2
x +1 ( x2+1) 3/2
1
= >0 , so f is CU on R . No IP
3/2
( x +1)
2

(e)

 
48. (a) lim xtan x= and lim xtan x= , so x= and x= are VA.
 + 2 2
x  /2 x   /2

  / 2  
f (x)=xtan x ,  <x< . f (x)=xsec x+tan x>0 0<x< , so f increases on 0, and
(b) 2 2 2 2

decreases on ,0 . 
2
(c) f (0)=0 is a local minimum value.
(d) / / 2 2    
f (x)=2sec x+2xtan xsec x>0 for  <x< , so f is CU on  , . No IP
2 2 2 2
(e)

49. f (x)=ln(1 ln x) is defined when x>0 (so that ln\:\italic{x} is defined) and 1 ln x>0 [so that
ln(1 ln x) is defined]. The second condition is equivalent to 1>ln x  x<e , so f has domain (0,e) .
+   +
(a) As x 0 ,ln x   , so 1 ln x  and f (x)  . As x e ,ln x 1 , so 1 ln x 0 and
20
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

f (x)  
. Thus, x=0 and x=e are vertical asymptotes. There is no horizontal asymptote.
/ 1 1 1
(b) f (x)=  = <0 on ( 0,e ) . Thus, f is decreasing on its domain, ( 0,e ) .
1 ln x x x(1 ln x)
/
(c) f (x) 0 on ( 0,e ) , so f has no local maximum or minimum value.
(d)
/
/ /  x(1 ln x) x( 1/x)+(1 ln x)
f (x) = 2
=
2 2
x(1 ln x) x (1 ln x)
ln x
=
2 2
x (1 ln x)
/ /
so f (x)>0 ln x<0 0<x<1. Thus, f is CU on ( 0,1 ) and CD on ( 1,e ) . There is an inflection point
at ( 1,0 ) .

(e)

x
e
50. f (x)= has domain R .
x
1+e
x x
e /e 1 1
(a) lim f (x)=lim =lim  x
= =1 , so y=1 is a HA.
x  x  ( 1+ex) /ex x  e +1
0+1
x
e 0
lim f (x)= lim = =0 , so y=0 is a HA. No VA.
x   x   x 1+0
1+e

(b) f
/
(x)=
( 1+e ) e  e e
x x x x
=
e
x
>0 for all x . Thus, f is increasing on R .
x 2 x 2
( 1+e ) ( 1+e )
(c) There is no local maximum or minimum.

21
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(d) x 2 x x
/ / =
( 1+e ) e  e 2 ( 1+e )e
x x

f (x) 2
( 1+ex) 2
e ( 1+e ) ( 1+e )  2e e ( 1 e )
x x x x x x
= =
x 4
( 1+e ) ( 1+ex) 3
/ / x
f (x)>0 1 e >0 x<0 , so f is CU on (   ,0 ) and CD on ( 0, ) . There is an inflection
1
point at 0, .
2
(e)

 1/ ( x+1 )  1/ ( x+1 )
51. (a) lim e =1 since  1/(x+1) 0 , so y=1 is a HA. lim e =0 since  1/(x+1)   ,
x   +
x  1
 1/ ( x+1 )
lim e = since  1/(x+1)  , so x= 1 is a VA.

x  1
 1/(x+1) /  1/(x+1) 1  1/(x+1) 2 /
(b) f (x)=e  f (x)=e  ( 1) [Reciprocal Rule] =e /(x+1)  f ( x ) >0
2
(x+1)
for all x except  1 , so f is increasing on (   , 1 ) and (  1, ).
(c) No local maximum or minimum
(d) 2  1/(x+1) 2  1/(x+1)
/ / (x+1) e 1/(x+1)  e 2(x+1)
f (x) = 2 2
(x+1)
 1/(x+1)  1/(x+1)
e 1 (2x+2) e (2x+1)
= = 
4 4
(x+1) (x+1)

/ / 1 1
f (x)>0 2x+1<0 , so f is CU on (  
x< , 1 ) and  1, , and CD on
2 2
1 1 2
 , . f has an IP at  ,e .
2 2
(e)

22
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

 
52. (a) f is periodic with period  , so we consider only 
2
<x<
2
. lim ln tan x =  ( 2
) ,
x 0


(
lim ln tan x =
2
) , and lim ln tan x =
+
( 2
) , so x=0 , x=
2
are VA.
x ( /2 ) x (   /2 )
2 2
2tan xsec x sec x 
(b) f (x)=ln tan x  ( 2
) /
f (x)= 2
=2
tan x
>0 tan x>0 0<x<
2
, so f is increasing on
tan x
 
0, and decreasing on  ,0 .
2 2
(c) No maximum or minimum
/ 2 4 / /  8cos 2x
(d) f (x)= =  f (x)= <0
sin xcos x sin 2x 2
sin 2x
  
 cos 2x>0  <x< , so f is CD on  ,0
4 4 4
     
and 0, , and CU on  , and , . IP are  ,0 .
4 2 4 4 2 4

(e)

53. (a)

From the graph, we get an estimate of f (1) 1.41 as a local maximum value, and no local minimum
value.
x+1 / 1 x
f (x)=  f (x)= .
3/2
2
x +1
2
x +1 ( )
23
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ 2
f (x)=0 x=1 . f (1)= = 2 is the exact value.
2

1 1
(b) From the graph in part (a), f increases most rapidly somewhere between x= and x= . To
2 4
/
find the exact value, we need to find the maximum value of f , which we can do by finding the
2
/ / / 2x  3x 1 3 17 3+ 17
critical numbers of f .f ( x) = =0 x= . x= corresponds to the
4 4
( x2+1) 5/2
/ / 3 17 7 17
minimum value of f . The maximum value of f is at ,   (  0.28,0.69 ) .
4 6 6

54. (a)
Tracing the graph gives us estimates of f (0)=0 for a local minimum value and f (2)=0.54 for a local
maximum value.
2  x /  x /  2
f (x)=x e  f (x)=xe (2 x) . f (x)=0 x=0 or 2 . f (0)=0 and f (2)=4e are the exact values.

3
(b) From the graph in part (a), f increases most rapidly around x= . To find the exact value, we
4
/ /
need to find the maximum value of f , which we can do by finding the critical numbers of f .
 x
f
/ /
(x)=e ( x2 4x+2) =0 x=2 2 . x=2+ 2 corresponds to the minimum value of f
/
. The
maximum value of f
/
(
is at 2 2 , ( 2
2  2+ 2
2) e ) ( 0.59,0.19 ) .
1 / / /
55. f (x)=cos x+ cos 2x f (x)= sin x sin 2x f (x)= cos x 2cos 2x
2

(a)
From the graph of f , it seems that f is CD on ( 0,1 ) , CU on ( 1,2.5) , CD on ( 2.5,3.7) , CU on
( 3.7,5.3) , and CD on ( 5.3,2 ) . The points of inflection appear to be at ( 1,0.4 ) , ( 2.5, 0.6 ) ,
24
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

( 3.7, 0.6 ) , and ( 5.3,0.4 ) .

(b)
/ /
From the graph of f (and zooming in near the zeros), it seems that f is CD on ( 0,0.94 ) , CU on
( 0.94,2.57) , CD on ( 2.57,3.71 ) , CU on ( 3.71,5.35) , and CD on ( 5.35,2 ) . Refined estimates of
the inflection points are ( 0.94,0.44 ) , ( 2.57, 0.63) , ( 3.71, 0.63) , and ( 5.35,0.44 ) .

3 4 / 3 3 4 2 2 3 2 3
56. f (x)=x (x 2)  f (x)=x 4(x 2) +(x 2) 3x =x (x 2) [4x+3(x 2)]=x (x 2) (7x 6)
/ / 3 2 2 2 3
f (x) =(2x)(x 2) (7x 6)+x 3(x 2) (7x 6)+x (x 2) (7)
2
=x(x 2) [2(x 2)(7x 6)+3x(7x 6)+7x(x 2)]
2 2
=x(x 2) 42x  72x+24 =6x(x 2) 7x  12x+4
2
( 2
)

(a)
From the graph of f , it seems that f is CD on (   ,0 ) , CU on ( 0,0.5) , CD on ( 0.5,1.3) , and CU on
( 1.3, ) . The points of inflection appear to be at ( 0,0 ) , ( 0.5,0.5) , and ( 1.3,0.6 ) .

(b)
/ /
From the graph of f (and zooming in near the zeros), it seems that f is CD on (   ,0 ) , CU on
( 0,0.45) , CD on ( 0.45,1.26 ) , and CU on ( 1.26, ) . Refined estimates of the inflection points are
( 0,0 ) , ( 0.45,0.53) , and ( 1.26,0.60 ) .
57. In Maple, we define f and then use the command plot (diff (diff (f, x), x), x=  3..3);. In
/ /
Mathematica, we define f and then use Plot [Dt [Dt [f, x], x], {x,  3, 3}]. We see that f >0 for

25
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ /
x>0.1 and f <0 for x<0.1 . So f is concave up on ( 0.1, ) and concave down on (   ,0.1 ) .

/ /
58. It appears that f is positive (and thus f is concave up) on (  1.8,0.3) and ( 1.5, ) and negative
(so f is concave down) on (   , 1.8 ) and ( 0.3,1.5) .

59. Most students learn more in the third hour of studying than in the eighth hour, so K(3) K(2) is
larger than K(8) K(7) . In other words, as you begin studying for a test, the rate of knowledge gain is
/
large and then starts to taper off, so K (t) decreases and the graph of K is concave downward.

60. At first the depth increases slowly because the base of the mug is wide. But as the mug narrows,
the coffee rises more quickly. Thus, the depth d increases at an increasing rate and its graph is
concave upward. The rate of increase of d has a maximum where the mug is narrowest; that is, when
the mug is half full. It is there that the inflection point (IP) occurs. Then the rate of increase of d starts
to decrease as the mug widens and the graph becomes concave down.

26
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

61.

From the graph, we estimate that the most rapid increase in the percentage of households in the
United States with at least one VCR occurs at about t=8. To maximize the first derivative, we need to
a
determine the values for which the second derivative is 0. We’ll use V (t)= , and substitute
ct
1+be
a=85 , b=53 , and c= 0.5 later.

/
V (t)=
a bce(
) [ by the Reciprocal Rule] and
ct

( 1+bect ) 2
ct 2
/ /
( 1+be ) ce  e 2 ( 1+be ) bce
ct ct ct ct

V (t) = abc 2
( 1+bect ) 2
 abc ce ( 1+be ) ( 1+be )  2be  abc e ( 1 be )
ct ct ct ct 2 ct ct
= =
ct 4
( 1+be ) ( 1+bect ) 3
/ / ct ct  0.5t 1
So V (t)=0 1=be  e =1/b . Now graph y=e and y= . These graphs intersect at t 7.94
53
years, which corresponds to roughly midyear 1988.

62. (a) As x  , t= x / 2

2
( 2)   , and e
t
0 . The HA is y=0 . Since t takes on its maximum
t
value at x=0 , so does e . Showing this result using derivatives, we have f (x)=e
2
( 2) 
 x / 2

/
f (x)=e
2
( 2) ( 
 x / 2
x/
2
) /
. f (x)=0 x=0 . Because f changes from positive to negative at x=0 ,
/

f (0)=1 is a local maximum. For inflection points, we find

 x / ( 2 )  x / ( 2 )  1  x / ( 2 )
2 2 2 2 2 2
/ /
f (x)=
1
2
e 1+xe  x/
2
=
2
e
2 2
1 x /(. ) ( )
 
/ / 2 2 / / 2 2
f (x)=0 x =  x=  .f (x)<0 x <    <x< . So f is CD on (   , ) and CU on
 1/2
(   ,  ) and ( , ) . IP at (   ,e ).
(b) Since we have IP at x=  , the inflection points move away from the y  axis as  increases.
27
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

(c)
From the graph, we see that as  increases, the graph tends to spread out and there is more area
between the curve and the x  axis.

3 2 / 2
63. f (x)=ax +bx +cx+d f (x)=3ax +2bx+c . We are given that f (1)=0 and f ( 2)=3 , so
/ /
f (1)=a+b+c+d=0 and f ( 2)= 8a+4b 2c+d=3 . Also f (1)=3a+2b+c=0 and f ( 2)=12a 4b+c=0 by
2 1 4 7
Fermat’s Theorem. Solving these four equations, we get a= , b= , c= , d= , so the function
9 3 3 9
1
is f (x)=
9
3
(2
2x +3x  12x+7 . )

64. f (x)=axe

now 1=2ae
 1/2
bx


2

we must have f (2)=0 . f (2)=1


/

a= .
/

/
e 2.
bx
2
f (x)=a xe 2bx+e 1 =ae
1=2ae
2
bx

4b
bx

/
and f (2)=0
2
( 2bx2+1) . For f (2)=1 to be a maximum value,

65. Suppose that f is differentiable on an interval I and f (x)>0 for all x in I except x=c . To show
that f is increasing on I , let x , x be two numbers in I with x <x .
1 2
/
0=(8b+1)ae

1 2
4b
. So 8b+1=0 b=
1
8
and

Case x <x <c . Let J be the interval { x I | x<c} . By applying the Increasing/Decreasing Test to f
2 1 2
on J , we see that f is increasing on J , so f (x )< f (x ) .
1 2
Case c<x <x . Apply the Increasing/Decreasing Test to f on K= { x I | x>c} .
2 1 2

Case x <x =c . Apply the proof of the Increasing/Decreasing Test, using the Mean Value Theorem
3 1 2
(MVT) on the interval x ,x and noting that the MVT does not require f to be
1 2
differentiable at the endpoints of x ,x .
1 2

28
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

Case c=x <x . Same proof as in Case 3.

4 1 2

Case x <c<x . By Cases 3 and 4, f is increasing on x ,c and on c,x , so f (x )< f (c)< f (x ) .

5 1 2 1 2 1 2
In all cases, we have shown that f (x )< f (x ) . Since x , x were any numbers in I with x <x
1 2 1 2 1 2
, we have shown that f is increasing on I .

66. (a) We will make use of the converse of the Concavity Test (along with the stated assumptions);
/ / / /
that is, if f is concave upward on I , then f >0 on I . If f and g are CU on I , then f >0 and
/ / / / / / / /
g >0 on I , so ( f +g ) =f +g >0 on I  f +g is CU on I .
/ / 2 / /
(b) Since f is positive and CU on I , f >0 and f >0 on I . So g ( x ) =[ f ( x ) ]  g =2 ff 
g
/ / /
=2 f f +2 ff
/ / /
=2 f( / ) 2+2 ff / />0 g is CU on I .

/ / / /
67. (a) Since f and g are positive, increasing, and CU on I with f and g never equal to 0 , we
/ / / / / / / / /
have f >0 , f  0 , f >0 , g>0 , g  0 , g >0 on I . Then ( fg) = f g+ fg 
/ / / / / / / / / / / /
( fg) =f g+2 f g + fg  f g+ fg >0 on I  fg is CU on I .
/ /
(b) In part (a), if f and g are both decreasing instead of increasing, then f
0 and g
0 on I , so
/ / / / / / / / / / / / / /
we still have 2 f g  0 on I . Thus, ( fg) =f g+2 f g + fg  f g+ fg >0 on I  fg is CU
on I as in part (a).
/ / / /
(c) Suppose f is increasing and g is decreasing . Then f  0 and g
0 on I , so 2 f g
0 on I and
the argument in parts (a) and (b) fails.
Example 1.
3 2 / / /
I= ( 0, ) , f (x)=x , g(x)=1/x . Then ( fg)(x)=x , so ( fg) (x)=2x and ( fg) (x)=2>0 on I . Thus, fg is
CU on I .

Example 2.
/ / / 3
I= ( 0, ) , f (x)=4x x , g(x)=1/x . Then ( fg)(x)=4 x , so ( fg) (x)=2/ x and ( fg) (x)= 1/ x <0
on I . Thus, fg is CD on I .

Example 3.
2
I= ( 0, ) , f (x)=x , g(x)=1/x . Thus, ( fg)(x)=x , so fg is linear on I .
/ / / /
68. Since f and g are CU on (   , ),f >0 and g >0 on (   , ).
/ / /
h(x)= f (g(x)) h (x)= f (g(x))g (x)
29
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ / / / / / / / / / / / 2 / / / /
h (x)= f (g(x))g (x)g (x)+ f (g(x))g (x)= f (g(x)) g (x) + f (g(x))g (x)>0 if f >0 .
So h is CU if f is increasing.

/ 2  2 
69. f (x)=tan x x f (x)=sec x 1>0 for 0<x< since sec x>1 for 0<x< . So f is increasing on
2 2
  
0, . Thus, f (x)> f (0)=0 for 0<x<  tan x x>0 tan x>x for 0<x< .
2 2 2

x 0 / x
70. (a) Let f (x)=e  1 x . Now f (0)=e  1=0 , and for x 0 , we have f (x)=e  1 0 . Now, since
x x
f (0)=0 and f is increasing on 0, ) , f (x) 0 for x 0 e  1 x 0 e  1+x .
x 1 2 / x
(b) Let f (x)=e  1 x x . Thus, f (x)=e  1 x , which is positive for x 0 by part (a). Thus, f (x) is
2
x 1 2 x 1 2
increasing on ( 0, ) , so on that interval, 0= f (0)
f (x)=e  1 x x  e  1+x+ x .
2 2
2 k
x x x
(c) By part (a), the result holds for n=1 . Suppose that e  1+x+ + + for x 0 . Let
2! k!
2 k k+1 k
x x x x / x x
f (x)=e  1 x    . Then f (x)=e  1 x   0 by assumption. Hence, f (x)
2! k! (k+1)! k!
k k+1
x x x
is increasing on ( 0, ) . So 0
x implies that 0= f (0)
f (x)=e  1 x   , and hence
k! (k+1)!
k k+1 2 n
x x x x x x
e  1+x+ + + for x 0 . Therefore, for x 0 , e  1+x+ + + for every
k! (k+1)! 2! n!
positive integer n , by mathematical induction.

3 2 / 2 / /
71. Let the cubic function be f (x)=ax +bx +cx+d f (x)=3ax +2bx+c f (x)=6ax+2b . So f is
CU when 6ax+2b>0 x> b/(3a) , CD when x< b/(3a) , and so the only point of inflection occurs
when x= b/(3a) . If the graph has three x  intercepts x , x and x , then the expression for f (x) must
1 2 3

( ) ( x x ) ( x x ) . Multiplying these factors together gives us

factor as f (x)=a x x
1 2 3
3 2 2
f (x)=a x  ( x +x +x ) x + ( x x +x x +x x ) x x x x . \ Equating the coefficients of the x  terms
1 2 3 1 2 1 3 2 3 1 2 3
for the two forms of f gives us b= a ( x +x +x ) . Hence, the x coordinate of the point of inflection
1 2 3
 a ( x +x +x ) x +x +x
b 1 2 3 1 2 3
is  = = .
3a 3a 3
4 3 2 / 3 2 / / 2 / /
72. P(x)=x +cx +x  P (x)=4x +3cx +2x P (x)=12x +6cx+2 . The graph of P (x) is a
/ /
parabola. \ If P (x) has two roots, then it changes sign twice and so has two inflection points. This
30
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ / 2 2
happens when the discriminant of P (x) is positive, that is, ( 6c )  4 12 2>0 36c  96>0
2 6 2 2 6 / /
c >  1.63. If 36c  96=0 c= , P (x) is 0 at one point, but there is still no inflection
3 3
/ / 2 2 6 / /
point since P (x) never changes sign, and if 36c  96<0 c < , then P (x) never changes
3
sign, and so there is no inflection point.

c=6 c=3 c=1.8

2 6
c= c=0 c= 2
3
For large positive c , the graph of f has two inflection points and a large dip to the left of the y axis.
As c decreases, the graph of f becomes flatter for x<0 , and eventually the dip rises above the x axis,
and then disappears entirely, along with the inflection points. As c continues to decrease, the dip and
the inflection points reappear, to the right of the origin.

/
73. By hypothesis g= f is differentiable on an open interval containing c . Since ( c,f(c) ) is a point of
/ /
inflection, the concavity changes at x=c , so f (x) changes signs at x=c . Hence, by the First
/ / /
Derivative Test, f has a local extremum at x=c . Thus, by Fermat’s Theorem f (c)=0 .

4 / 3 / / 2 / / / /
74. f (x)=x  f (x)=4x  f (x)=12x  f (0)=0 . For x<0 , f (x)>0 , so f is CU on (   ,0 ) ;
/ /
for x>0 , f (x)>0 , so f is also CU on ( 0, ) . Since f does not change concavity at 0 , ( 0,0 ) is not
an inflection point.

2 2 / 2 2 2
75. Using the fact that x = x , we have that g(x)=x x  g (x)= x + x =2 x =2 x 
2  1/2 2x
/ /
g (x)=2x x ( )=
x
/ /
<0 for x<0 and g (x)>0 for x>0 , so ( 0,0 ) is an inflection point. But

31
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.3 How Derivatives Affect the Shape of a Graph

/ /
g (0) does not exist.

/ / / / / /
76. There must exist some interval containing c on which f is positive, since f (c) is positive
/ / / / / / / / / / / /
and f is continuous. On this interval, f is increasing (since f is positive), so f =( f )
/
changes from negative to positive at c . So by the First Derivative Test, f has a local minimum at
/ /
x=c and thus cannot change sign there, so f has no maximum or minimum at c . But since f
changes from negative to positive at c , f has a point of inflection at c (it changes from concave down
to concave up).

32
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

f (x) 0
1. (a) lim is an indeterminate form of type .
x a g(x) 0
f (x)
(b) lim =0 because the numerator approaches 0 while the denominator becomes large.
x a p(x)
h(x)
(c) lim =0 because the numerator approaches a finite number while the denominator becomes
x a p(x)
large.
p(x)
(d) If lim p(x)= and f (x) 0 through positive values, then lim = . If f (x) 0 through
x a x a f (x)
p(x)
negative values, then lim =  . If f (x) 0 through both positive and negative values, then the
x a f (x)
limit might not exist.
p(x) 
(e) lim is an indeterminate form of type .
x a q(x) 

2. (a) lim f (x) p(x) is an indeterminate form of type 0  .

x a
(b) When x is near a , p(x) is large and h(x) is near 1 , so h(x) p(x) is large. Thus, lim h(x) p(x) = .
x a
(c) When x is near a , p(x) and q(x) are both large, so p(x)q(x) is large. Thus, lim p(x)q(x) = .
x a

3. (a) When x is near a , f (x) is near 0 and p(x) is large, so f (x) p(x) is large negative. Thus,
lim f (x) p(x) =  .
x a
(b) lim p(x) q(x) is an indeterminate form of type    .
x a
(c) When x is near a , p(x) and q(x) are both large, so p(x)+q(x) is large. Thus, lim p(x)+q(x) = .
x a

g ( x) 0
4. (a) lim f (x) is an indeterminate form of type 0 .
x a
p ( x)
(b) If y= f (x) , then ln y= p(x)ln f (x) . When x is near a , p(x)  and ln f (x)   , so
p ( x) ln y p
ln y   . Therefore, lim f (x) =lim y=lim e =0 , provided f is defined.
x a x a x a
p ( x) 
(c) lim h(x) is an indeterminate form of type 1 .
x a
f ( x) 0
(d) lim p(x) is an indeterminate form of type  .
x a
q ( x)
(e) If y= p(x) , then ln y=q(x)ln p(x) . When x is near a , q(x)  and ln p(x)  , so
1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

q ( x) ln y
ln y  . Therefore, lim p(x) =lim y=lim e = .
x a x a x a
q(x) 1/q ( x ) 0
(f) lim p(x) =lim p(x) is an indeterminate form of type  .
x a x a

0
5. This limit has the form . We can simply factor the numerator to evaluate this limit.
0
2
x 1 (x+1)(x 1)
lim =lim =lim (x 1)= 2
x  1 x+1 x  1 x+1 x  1

x+2 x+2 1
6. lim =lim =lim = 1
2 (x+1)(x+2) x  2 x+1
2
x  2 x +3x+2 x 

9 8
0 x 1 9x 9 4 9 9
7. This limit has the form . lim 5 =lim = lim x = (1)=
0 x 1 4 5 x 1 5 5
x  1 x 1 5x

a a 1
x 1 ax a
8. lim =lim =
b b 1 b
x 1 x 1 x 1 bx

0 cos x  sin x
9. This limit has the form . lim = lim = lim tan x=  .
0 + 1 sin x +  cos x +
x ( /2) x ( /2) x ( /2)

2 2
x+tan x 1+sec x 1+1
10. lim =lim = =2
x 0 sin x x 0 cos x 1

t t
0 e 1 e t 2 +
11. This limit has the form . lim =lim = since e  1 and 3t  0 as t 0.
0 t 0 3 t 0 2
t 3t

3t 3t
e  1 3e
12. lim =lim =3
t 0 t t 0 1

2 2
0 tan px psec px p(1) p
13. This limit has the form . lim =lim = =
0 x 0 tan qx x 0 2 2 q
qsec qx q(1)

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

1 sin  0
14. lim = =0 . L’Hospital’s Rule does not apply.
   /2 csc  1

 ln x 1/x
15. This limit has the form . lim =lim =0
 x  x x  1

x x
e e x
16. lim =lim =lim e =
x  x x  1 x 

+
17. lim (ln x)/x =  since ln x   as x 0 and dividing by small values of x just increases the
+
x 0
magnitude of the quotient (ln x)/x . L’Hospital’s Rule does not apply.

1 1

ln ln x ln x x 1
18. lim =lim =lim =0
x  x x  1 x  xln x

t t t t
0 5 3 5 ln 5 3 ln 3 5
19. This limit has the form . lim =lim =ln 5 ln 3=ln
0 t 0 t t 0 1 3

ln x 1/x 1 1
20. lim =lim = =
x 1 sin  x x 1  cos  x  (  1 ) 

x x x
0 e  1 x e 1 e 1
21. This limit has the form . lim =lim =lim =
0 x 0 2 x 0 2x x 0 2 2
x

x 2 x x x
e  1 x x /2 e  1 x e 1 e 1
22. lim =lim =lim =lim =
x 0 6x x 0 6 6
3 2
x 0 x x 0 3x

x x x x
 e e e e
23. This limit has the form . lim =lim =lim =lim =
 x  x3 x  3x2 x  6x x  6

sin x cos x 1
24. lim =lim = =1
x 0 sinh x x 0 cosh x 1

3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

 1 2
0 sin x 1/ 1 x 1 1
25. This limit has the form . lim =lim =lim = =1
0 x 0 x x 0 1 x 0 2 1
1 x

sin x x cos x 1  sin x  cos x 1

26. lim =lim =lim =lim =
x 0 3 x 0 2 x 0 6x x 0 6 6
x 3x

0 1 cos x sin x cos x 1

27. This limit has the form . lim =lim =lim =
0 x 0 2 x 0 2x x 0 2 2
x

2
(ln x) 2(ln x)(1/x) ln x 1/x
28. lim =lim =2lim =2lim =2(0)=0
x  x x  1 x  x x  1

x+sin x 0+0 0
29. lim = = =0 . L’Hospital’s Rule does not apply.
x 0 x+cos x 0+1 1

2 2
 msin mx+nsin nx  m cos mx+n cos nx 1 2 2
30. lim
cos mx cos nx
2
=lim
2x
=lim
2
=
2
n m ( )
x 0 x x 0 x 0

x x
 x 1 1+2e 2e
31. This limit has the form . lim =lim =lim =lim =1
 x  (
ln 1+2e
x
) x  1
x
 2e
x x  2e
x x  2e
x

1+2e

2
x 1 1+16x 1
32. lim  1
=lim =lim =
1 4 4
x 0 tan (4x) x 0  4 x 0
2
1+(4x)

2
0 1 x+ln x  1+1/x  1/x  1 1
33. This limit has the form . lim =lim =lim = =
0 x 1 1+cos  x x 1   sin  x x 1 2 2 2
 cos  x  (  1) 

2 2 2 2
x +2 x +2 x +2 1+2/x 1
34. lim =lim = lim = lim =
x  2 x 
2
x 
2
x  2 2
2x +1 2x +1 2x +1 2+1/x

4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

a a 1 a 2
0 x  ax+a 1 ax  a a(a 1)x a(a 1)
35. This limit has the form . lim =lim =lim =
0 x 1 2 x 1 2(x 1) x 1 2 2
(x 1)

 2x
1 e 1 1
36. lim = =0 . L’Hospital’s Rule does not apply.
x 0 sec x 1

37. This limit has the form 0 (  ) . We need to write this product as a quotient, but keep in mind
1
that we will have to differentiate both the numerator and the denominator. If we differentiate ,
ln x
we get a complicated expression that results in a more difficult limit. Instead we write the quotient as
ln x
.
 1/2
x
3/2
ln x
1/x  2x
lim x ln x=lim =lim  =lim (  2 x ) =0
+  1/2 1  3/2 3/2
x
+
0 x 0 x
+
x 0  x  2x x 0
+

2
2 x x 2x 2 x
38. lim x e = lim = lim = lim = lim 2e =0
 x  x  x
x   x   e x    e x   e x  

0
39. This limit has the form   0 . We’ll change it to the form .
0
sin 6x 6cos 6x 6(1)
lim cot 2xsin 6x=lim =lim = =3
x 0 x 0 tan 2x x 0 2 2
2sec 2x 2(1)

40.
lim sin xln x =lim ln x 1/x sin x
=lim = lim  tan x
csc x +  csc xcot x x
+
x 0 + +
x 0 x 0 x 0
sin x
= lim lim tan x = 1 0=0
+ x +
x 0 x 0

2 3 2
3  x x 3x 3x 3
41. This limit has the form   0 . lim x e =lim 2
=lim 2
=lim 2
=lim 2
=0
x  x  x x  x x  x x  x
e 2xe 2e 4xe

42.
5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

lim (1 tan x)sec x= ( 1 1 ) 2 =0 . L’Hospital’s Rule does not apply.

x  /4

43. This limit has the form 0 (  ) .

ln x 1/x 1 2
lim ln xtan ( x/2)=lim =lim = =
+ cot ( x/2) 2 2 
x
+
1
+
x 1 x 1 (   /2 ) csc ( x/2) (  /2)(1)

44. lim x tan (1/x)=lim

tan (1/x)
=lim
sec (1/x)  1/x
2
( 2
) =lim sec 2(1/x)=12=1
x  x  1/x x  2 
 1/x x

45.
1 1 1 sin x x
lim  csc x =lim  =lim
x 0 x x 0 x sin x x 0 xsin x
cos x 1  sin x 0
=lim =lim = =0
x 0 xcos x+sin x x 0 2cos x xsin x 2

1 cos x 1 cos x sin x

46. lim (csc x cot x)=lim  =lim =lim =0
x 0 x 0 sin x sin x x 0 sin x x 0 cos x

47. We will multiply and divide by the conjugate of the expression to change the form of the
expression.

lim ( 2
)
x +x  x =lim
2
x +x  x

2
x +x +x
=lim
( x2+x)  x2
x  x  1 2 x  2
x +x +x x +x +x
x 1 1 1
=lim =lim = = .
x  2 x  1+1/x +1 1 +1 2
x +x +x

2 2 2 2
As an alternate solution, write x +x  x as x +x  x , factor out x , rewrite as
( 1+1/x  1 ) /(1/x) , and apply l’Hospital’s Rule.
48.
1 1 x 1 ln x 1 1/x x
lim  =lim =lim 
x 1 ln x x 1 x 1 (x 1)ln x x 1 (x 1)(1/x)+ln x x
x 1 1 1 1
=lim =lim = =
x 1 x 1+xln x x 1 1+1+ln x 2+0 2
6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

49. The limit has the form    and we will change the form to a product by factoring out x .
ln x ln x 1/x
lim (x ln x)=lim x 1 = since lim =lim =0 .
x  x  x x  x x  1

1/x
50. As x  , 1/x 0 , and e  1 . So the limit has the form    and we will change the form to
a product by factoring out x .

lim ( 1/x
)
xe  x =lim x e  1 =lim
e  1
=lim
e
( 1/x
)
1/x 1/x
(  1/x2) =lim 1/x
e =e =1
0

 x  x  1/x x  2 
x  1/x x

2
x 2 2 ln x 1/x 1 2
51. y=x  ln y=x ln x , so lim ln y=lim x ln x=lim =lim =lim  x =0 
2 3 2
x
+
0 x
+
0 x 0
+
1/x x 0
+
 2/x x
+
0
2
x ln y 0
lim x =lim e =e =1 .
+ +
x 0 x 0

x
52. y=(tan 2x) ln y=x ln tan 2x , so
lim ln y lim x ln tan 2x=lim ln tan 2x
+ = + + 1/x
x 0 x 0 x 0
2 2
(1/tan 2x)(2sec 2x)  2x cos 2x 2x  x
= lim =lim =lim  lim =1 0=0
+ sin 2x + cos 2x
2 2
x 0
+
 1/x x
+
0 sin 2xcos 2x x 0 x 0
x ln y 0
lim (tan 2x) =lim e =e =1 .
+ +
x 0 x 0

1/x 1 ln (1 2x)  2/(1 2x)

53. y=(1 2x)  ln y= ln (1 2x) , so lim ln y=lim =lim = 2
x x 0 x 0 x x 0 1
1/x ln y  2
lim (1 2x) =lim e =e .
x 0 x 0

a bx a
54. y= 1+  ln y=bxln 1+ , so
x x
1 a
b 
1+a/x 2
bln (1+a/x) x ab
lim ln y=lim =lim =lim =ab
x  x  1/x x  2  1+a/x
 1/x x

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

lim
x 

55. y=

lim ln y=lim
x 

so lim
x

56. y=x


lim ln y=lim
x  x 
1+
a
x

3 5
1+ +
x
x

x
2

1+


(ln 2)(ln x)
1+ln x
3 5
x

( ln 2 ) / ( 1+ln x )
+
bx

ln 2
1+ln x
=lim
x 
=lim e

ln

2
x

ln x 
(ln 2)(1/x)
1/x


1+
x


3 5
x
1/x

x
+

ln y=

=lim ln 2=ln 2 ,
x 
ln y

x
2

=lim e
x
=e

ln y=xln


ab

ln y
.

=lim

=e .
x
1+

3

x

.
3 5
+
x
2


x


3
2

10
x
3

 1/x
/ 2
1+
3 5
x
+
x
2
=lim
x 
3+
3 5
1+ +
x
x
2
10
x
=3 ,

( ln 2 ) / ( 1+ln x ) ln y ln 2
so lim x =lim e =e =2 .
x  x 

1/x ln x 1/x 1/x ln y 0

57. y=x  ln y=(1/x)ln x lim ln y=lim =lim =0 lim x =lim e =e =1
x  x  x x  1 x  x 

1
58. y= e +x ( x
) 1/x ln y=
x
x
ln e +x , so( )
lim ln y=lim
ln e +x e +1
=lim x =lim x =lim
( e
x
e
=1 lim
) x x x
( ex+x) 1/x=lim e
ln y 1
=e =e .
x  x  x x  e +x x  e +1 x  ex x  x 

x x x
59. y=  ln y=xln 
x+1 x+1
x ln x ln (x+1) 1/x 1/(x+1)
lim ln y =lim xln =lim =lim
x  x  x+1 x  1/x x  2
 1/x
2
x  x
=lim  x+ =lim = 1
x  x+1 x  x+1

so

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

x x ln y  1
lim =lim e =e
x  x+1 x 
x  1 x x  1
x x+1 1  1
Or: lim =lim = lim 1+ =e
x  x+1 x  x x  x

5/x 5 ln (cos 3x)  3tan 3x

60. y=(cos 3x)  ln y= ln (cos 3x) lim ln y=5lim =5lim =0 ,
x x 0 x 0 x x 0 1
5/x 0
so lim (cos 3x) =e =1 .
x 0

2 2
1/x 1 ln cos x  tan x  sec x 1
61. y=(cos x)  ln y= ln cos x  lim ln y=lim =lim =lim = 
2 + + 2 + 2x + 2 2
x x 0 x 0 x x 0 x 0
2
1/x ln y  1/2
lim (cos x) =lim e =e =1/ e
+ +
x 0 x 0

2x 3 2x+1 2x 3

62. y=  ln y=(2x+1)ln 
2x+5 2x+5
2
lim ln y =lim ln (2x 3) ln (2x+5) =lim 2/(2x 3) 2/(2x+5) =lim  8(2x+1)
x  1/(2x+1) x  (2x 3)(2x+5)
2
x  x   2/(2x+1)
2
 8(2+1/x) 2x 3 2x+1  8
=lim = 8 lim =e
x  (2 3/x)(2+5/x) x  2x+5

63.

From the graph, it appears that lim x ln (x+5) ln x =5 .

x 
To prove this, we first note that
x+5 5
ln (x+5) ln x=ln =ln 1+  ln 1=0 as x  . Thus,
x x

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

1 1
lim x ln (x+5) ln x 
ln (x+5) ln x x+5 x
x  =lim =lim
 1/x x  2
x  1/x
2 2
x (x+5)  x 5x
=lim  =lim =5
x  x(x+5) 1 x 
2
x +5x

64.

tan 2x
From the graph, it appears that lim (tan x)  0.368 .
x  /4
 tan 2x
The limit has the form 1 . Now y=(tan x) 
ln y=tan 2xln (tan x) , so
2
ln (tan x) sec x/tan x 2/1
lim ln y= lim = lim = = 1
x  /4 x  /4 cot 2x
2
x  /4  2csc 2x  2(1)
tan 2x ln y  1
lim (tan x) = lim e =e =1/e 0.3679 .
x  /4 x  /4

65.

/
f (x) f (x)
From the graph, it appears that lim =lim =0.25. We calculate
x 0 g(x) x 0 g (x)
/

x x
f (x) e 1 e 1
lim =lim 3 =lim = .
x 0 g(x) x 0 x +4x x 0 3x +4 4
2

66.

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

/
f (x) f (x)
From the graph, it appears that lim =lim =4 . We calculate
x 0 g(x) x 0 g (x)
/

f (x) 2xsin x 2(xcos x+sin x)

lim =lim =lim
x 0 g(x) x 0 sec x 1 x 0 sec xtan x
2( xsin x+cos x+cos x) 4
=lim = =4
(
x 0 sec x sec x +tan x(sec xtan x) 1
2
)
x x x x
e e e e
67. lim =lim =lim =   =lim =
x  n!
n n 1 n 2
x  x x  nx x  n(n 1)x

ln x 1/x 1
68. lim =lim =lim =0 since p>0 .
p p 1 p
x  x x  px x  px

i nt  i nt
69. First we will find lim 1+ , which is of the form 1 . y= 1+ 
n  n n
i
ln y=ntln 1+ , so
n

lim ln y=lim ntln 1+

i
=t lim
ln ( 1+i/n )
=t lim
(  i/n2) =t lim i
=ti lim y=e
it

( 1+i/n ) (  1/n ) n 
n  n  n n  1/n n 
2 1+i/n n 

i nt it
. Thus, as n  , A=A 1+  Ae .
0 n 0

70. (a)
lim v =lim mg 1 e ( ct/m
) = mgc lim ( 1 e ct/m)
t  t  c t 
mg mg
= (1 0)= ,
c c

which is the speed the object approaches as time goes on, the so called limiting velocity.
(b) lim v
m 
11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

 ct/m  ct/m
= lim
mg
1 e

( ct/m
) g
= lim
1 e g
= lim
 e ( ct/m2)
m  c c m  1/m c m  2
 1/m
g  ct/m
= (ct) lim e =gt(1)
c m 
The speed of a very heavy falling object is approximately proportional to the elapsed time t , provided
it can fall for time t in an environment where the given model continues to hold. .

71. We see that both numerator and denominator approach 0 , so we can use l’Hospital’s Rule:
1 4  1/2 1  2/3 2
3 4 3
2a x x  a aax 2
3
2a x x ( 3 3
2a  4x  a) ( 3
( )
aax ) a
lim = lim
4 1 3  3/4
x a
a ax
3 x a

4
ax 3ax( ) (
2
)
1 4  1/2 1 3 2  2/3
2
3
2a a a ( 3 3
2a  4a ) ( 3
) ( )
a aa
=
1 3  3/4

4
aa ( ) ( )
3aa
2

4  1/2 1 3 3  2/3 1
a  a ( ) ( )
3
3
a a ( )  a
3
a
4 4 16
= = = a = a
3 3 4  3/4 3 3 3 9

4
a a ( ) 
4

72. Let the radius of the circle be r . We see that A( ) is the area of the whole figure (a sector of the
circle with
1 2
radius 1 ), minus the area of OPR . But the area of the sector of the circle is r  (see Reference
2
1 1 1 2
Page 1), and the area of the triangle is r PQ = r(rsin  )= r sin  . So we have
2 2 2
1 2 1 2 1 2
A( )= r   r sin  = r (  sin  ) . Now by elementary trigonometry,
2 2 2
1 1 1 1 2
B( )= QR PQ = ( r OQ ) PQ = (r rcos  )(rsin  )= r (1 cos  )sin  .
2 2 2 2
So the limit we want is
1 2
A( ) r (  sin  )
lim 2 1 cos 
B( ) = lim =lim
+ 1 2 + (1 cos  )cos  +sin  (sin  )
+
  0   0 r (1 cos  )sin    0
2

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

1 cos  sin 
= lim + 2 2
=lim
 sin   2cos  ( sin  )+2sin  (cos  )
  0 cos   cos  +sin    0
+

sin  1 1 1
= lim +  sin  +4sin  cos  =lim
+  1+4cos 
= =
 1+4cos 0 3
  0   0

0
73. Since f (2)=0 , the given limit has the form .
0
/ /
f (2+3x)+ f (2+5x) f (2+3x) 3+ f (2+5x) 5 / / /
lim =lim = f (2) 3+ f (2) 5=8 f (2)=8 7=56
x 0 x x 0 1

3 2
sin 2x b sin 2x+ax +bx 2cos 2x+3ax +b 2
74. L=lim +a+ =lim =lim . As x 0 , 3x  0 , and
3 2 3 2
x 0 x x x 0 x x 0 3x
( 2cos 2x+3ax +b)  2
b+2 , so the last limit exists only if b+2=0 , that is, b= 2 . Thus,
2
2cos 2x+3ax  2  4sin 2x+6ax  8cos 2x+6a 6a 8
lim =lim =lim = , which is equal to 0 if and
x 0 2
x 0 6x x 0 6 6
3x
4 4
only if a= . Hence, L=0 if and only if b= 2 and a= .
3 3

75. Since lim [ f (x+h) f (x h)]= f (x) f (x)=0 ( f is differentiable and hence continuous) and lim 2h=0
h 0 h 0
, we use l’Hospital’s Rule:
/ / / / /
f (x+h) f (x h) f (x+h)(1) f (x h)( 1) f (x)+ f (x) 2 f (x) /
lim =lim = = = f (x)
h 0 2h h 0 2 2 2
f (x+h) f (x h)
is the slope of the secant line between ( x h,f(x h) ) and ( x+h,f(x+h) ) . As h 0 , this
2h
/
line gets closer to the tangent line and its slope approaches f (x) .

76. Since lim f (x+h) 2 f (x)+ f (x h) = f (x) 2 f (x)+ f (x)=0 ( f is differentiable and hence continuous)
h 0
and

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

2
lim h =0 , we can apply l’Hospital’s Rule:
h 0

/ /
f (x+h) 2 f (x)+ f (x h) f (x+h) f (x h) / /
lim =lim =f (x)
h 0 2 h 0 2h
h

/
At the last step, we have applied the result of Exercise to f (x) .

f (x) 1
77. (a) We show that lim =0 for every integer n
0 . Let y= . Then
n 2
x 0 x x
2
 1/x n n 1
f (x) e y ny n!
lim =lim =lim =lim =   =lim =0
2n 2 n y y y
x 0 x x 0 (x ) y  e y  e y  e
f (x) n f (x) n f (x) / f (x) f (0) f (x)
lim =lim x =lim x lim =0 . Thus, f (0)=lim =lim =0 .
x 0 n x 0 2n x 0 x 0 2n x 0 x 0 x 0 x
x x x
( n)
(b) Using the Chain Rule and the Quotient Rule we see that f (x) exists for x 0 . In fact, we prove
by induction that for each n
0 , there is a polynomial p and a non negative integer k with
n n
( n) k
n
f (x)= p (x) f (x)/x for x 0 . This is true for n=0 ; suppose it is true for the n th derivative. Then
n
/ 3
f (x)= f (x)(2/x ) , so
( n+1 ) k / / k  1  2k
n n n
f (x) = x p (x) f (x)+ p (x) f (x)  k x p (x) f (x) x
n n n n
k k  1  2k
= n /
x p (x)+ p (x) 2/x  k x
n n
( 3) n
n
p (x)
n
f (x)x
n

= x
k +3
n /
p (x)+2 p (x) k x
k +2
n
p (x) f (x)x
 ( 2kn+3)
n n n n
which has the desired form.
( n) / ( n)
Now we show by induction that f ( 0 ) =0 for all n . By part (a), f (0)=0 . Suppose that f (0)=0 .
Then
k
n
( n) ( n) ( n) p (x) f (x)/x p (x) f (x)
( n+1 ) f (x) f (0) f (x) n n
f (0) = lim =lim =lim =lim
x 0 x 0 x 0 x x 0 x x 0 k +1
n
x

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.4 Indeterminate Forms and L’Hospital’s Rule

f (x)
lim p (x)lim = p (0) 0=0
= x 0
n
x 0 k +1 n
n
x

78. (a) For f to be continuous, we need lim f (x)= f (0)=1 . We note that for x 0 ,
x 0
x ln x 1/x
ln f (x)=ln x =xln x . So lim ln f (x)=lim xln x =lim =lim =0 . Therefore,
x 0 x 0 x 0 1/x x 0  1/x2
ln f ( x ) 0
lim f (x)=lim e =e =1 . So f is continuous at 0 .
x 0 x 0
(b) From the graphs, it appears that f is differentiable at 0 .

/
/ f (x) 1
(c) To find f , we use logarithmic differentiation: ln f (x)=xln x  =x +ln x 
f (x) x
/ x /
f (x)= f (x) ( 1+ln x ) = x ( 1+ln x ) , x 0 . Now f (x)   as x 0 , so the curve has a vertical
tangent at ( 0,1 ) and is therefore not differentiable there. The fact cannot be seen in the graphs in part
(b) because ln x    very slowly as x 0 .

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3
( 2
)
1. y= f (x)=x +x=x x +1 A. f is a polynomial, so D=R . B. x  intercept =0 , y intercept = f (0)=0
C. f ( x)= f (x) , so f is odd; the curve is symmetric about the origin. D. f is a polynomial, so there is
/ 2
no asymptote. E. f (x)=3x +1>0 , so f is increasing on (   , ) . F. There is no critical number and
/ / / /
hence, no local maximum or minimum value. G. f (x)=6x>0 on ( 0, ) and f (x)<0 on (   ,0 ) ,
so f is CU on ( 0, ) and CD on (   ,0 ) . Since the concavity changes at x=0 , there is an inflection
point at ( 0,0 ) .
H.

3 2 2
2. y= f (x)=x +6x +9x=x(x+3) A. D=R B. x  intercepts are  3 and 0 , y  intercept =0 C. No
/ 2
symmetry D. No asymptote E. f (x)=3x +12x+9=3(x+1)(x+3)<0  3<x< 1 , so f is decreasing on
(  3, 1 ) and increasing on (   , 3) and (  1, ) . F. Local maximum value f ( 3)=0 , local minimum
/ /
value f ( 1)= 4 G. f (x)=6x+12=6(x+2)>0 x> 2 , so f is CU on (  2, ) and CD on (   , 2 ) .
IP at (  2, 2 )
H.

2 3
( 2
)
3. y= f (x)=2 15x+9x  x = (x 2) x  7x+1 A. D=R B. y  intercept: f (0)=2 ; x intercepts: f (x)=0
7 45
 x=2 or (by the quadratic formula) x=  0.15 , 6.85
2
/
C. No symmetry D. No asymptote E. f (x)= 15+18x 3x = 3 x  6x+5 = 3(x 1)(x 5)>0 1<x<5
2
( 2
)
so f is increasing on ( 1,5) and decreasing on (   ,1 ) and ( 5, ) . F. Local maximum value f (5)=27
/ /
, local minimum value f (1)= 5 G. f (x)=18 6x= 6(x 3)>0 x<3 , so f is CU on (   ,3) and CD
on ( 3, ) . IP at ( 3,11 )
H.
1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2 4 2
( 2
)
4. y= f (x)=8x  x =x 8 x A. D=R B. y  intercept: f (0)=0 ; x intercepts: f (x)=0 x=0 ,  2 2 (
  2.83 ) C. f ( x)= f (x) , so f is even and symmetric about the y  axis. D. No asymptote E.
/ 3
( 2
)
f (x)=16x 4x =4x 4 x =4x(2+x)(2 x)>0 x< 2 or 0<x<2 , so f is increasing on (   , 2 ) and
( 0,2 ) and decreasing on (  2,0 ) and ( 2, ) . F. Local maximum value f ( 2)=16 , local minimum
2 2 2
/ / 2
value f (0)=0 G. f (x)=16 12x =4 4 3x =0 x=
2
( 3
)/ /
. f (x)>0 
3
<x<
3
, so f is CU
2 2 2 2 2 80
on  , and CD on   , and , . IP at  , H.
3 3 3 3 3 9

4 3 3
5. y= f (x)=x +4x =x ( x+4 ) A. D=R B. y  intercept: f (0)=0 ; x  intercepts: f (x)=0 x= 4,0 C. No
/ 3 2 2
symmetry D. No asymptote E. f (x)=4x +12x =4x ( x+3) >0 x> 3 , so f is increasing on (  3, )
and decreasing on (   , 3) . F. Local minimum value f ( 3)= 27 , no local maximum G.
/ / 2
f (x)=12x +24x=12x(x+2)<0  2<x<0 ,
so f is CD on (  2,0 ) and CU on (   , 2 ) and ( 0, ).
IP at ( 0,0 ) and (  2, 16 )
H.

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3
6. y= f (x)=x(x+2) A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0
x= 2,0 C. No symmetry
/ 2 3 2 2 / 1
D. No asymptote E. f (x)=3x(x+2) + ( x+2 ) = ( x+2 ) 3x+ ( x+2 ) =(x+2) (4x+2) . f (x)>0 x> ,
2
/ 1 1
and f (x)<0 x< 2 or  2<x< , so f is increasing on  , and decreasing on (   , 2 ) F.
2 2
1 27
Local minimum value f  = , no local maximum
2 16
G.
/ / 2
f (x) =(x+2) (4)+(4x+2)(2)(x+2)
=2(x+2) (x+2)(2)+4x+2
=2(x+2)(6x+6)=12(x+1)(x+2)

/ /
f (x)<0  2<x< 1 , so f is CD on (  2, 1 ) and CU on (   , 2 ) and (  1, ) . IP at (  2,0 ) and
(  1, 1 )
H.

5 2
7. y= f (x)=2x  5x +1 A. D=R B. y intercept: f (0)=1 C. No symmetry D. No asymptote E.
/ 4 3 2 / /
f (x)=10x  10x=10x(x  1)=10x(x 1)(x +x+1) , so f (x)<0 0<x<1 and f (x)>0 x<0 or x>1 .
Thus, f is increasing on (   ,0 ) and ( 1, ) and decreasing on ( 0,1 ) .
/ / 3 3
F. Local maximum value f (0)=1 , local minimum value f (1)= 2 G. f (x)=40x  10=10(4x  1) so
f
/ /
(x)=0
3
x=1/ 4 . f
/ /
(x)>0
3
x>1/ 4 and f
/ /
(x)<0
3
x<1/ 4 , < so f is CD on   ( 3
,1/ 4 )
1 9
and CU on 1/ 4, ( 3
) . IP at 3
,1
2
 ( 0.630, 0.786 )
4 2( )
3
4
H.

8.
3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3 5 3 2 20
y= f (x)=20x  3x A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0  3x x =0 x=0
3
or  20/3  2.582 C. f ( x)= f (x), so f is odd;
the curve is symmetric about the origin. D. No asymptote E.
/ 2 4 2 2 2 / /
f (x)=60x  15x = 15x (x  4)= 15x (x+2)(x 2) , so f (x)>0  2<x<0 or 0<x<2 and f (x)<0
x< 2 or x>2 . Thus, f is increasing on (  2,0 ) and ( 0,2 ) and f is decreasing on (   , 2 ) and ( 2, ).
F. Local minimum value f ( 2)= 64 , local maximum value f (2)=64 G.
/ / 3 2 / / / /
f (x)=120x 60x = 60x(x  2) . f (x)>0 x< 2 or 0<x< 2 ; f (x)<0  2 <x<0 or x> 2 .
Thus, f is CU on (   , 2 )
and ( 0, 2 ) , and f is CD on (  2 ,0 ) and ( 2 , ) . IP at (  2 , 28 2 )  (  1.414, 39.598 ) , ( 0,0 ) ,
and ( 2 ,28 2 )
H.

9. y= f (x)=x/(x 1) A. D= { x| x 1} = (   ,1 )
( 1, ) B. x  intercept =0 , y  intercept = f (0)=0 C. No
x x x
symmetry D. lim =1 , so y=1 is a HA. lim =  , lim = , so x=1 is a VA. E.
x   x 1  x 1 + x 1
x 1 x 1
/ (x 1) x  1
f (x)= = <0 for x 1 , so f is decreasing on (   ,1 ) and ( 1, ) . F. No extreme values
2 2
(x 1) (x 1)
/ / 2
G. f (x)=
3
>0 x>1 , so f is CU on ( 1, ) and CD on (   ,1 ) . No IP
(x 1)
H.

2
10. y=x/(x 1) A. D= { x| x 1} = (   ,1 )
( 1, ) B. x intercept =0 , y  intercept = f (0)=0 C. No
x x
symmetry D. lim =0 , so y=0 is a HA. lim = , so x=1 is a VA. E.
x   (x 1)2 x 1 (x 1)2

4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2
/ (x 1) (1) x(2)(x 1)  x 1
f (x)= = . This is negative on (   , 1 ) and ( 1, ) and positive on (  1,1 )
4 3
(x 1) (x 1)
, so f (x) is decreasing on (   , 1 ) and ( 1, ) and increasing on (  1,1 ) .
1
F. Local minimum value f ( 1)= , no local maximum. G.
4
3 2
/ / (x 1) ( 1)+(x+1)(3)(x 1) 2(x+2)
f (x)= = . This is negative on (   , 2 ) , and positive on (  2,1 )
6 4
(x 1) (x 1)
2
and ( 1, ) . So f is CD on (   , 2 ) and CU on (  2,1 ) and ( 1, ) . IP at  2,
9
H.

1
( 2
)
11. y= f (x)=1/ x  9 A. D= { x| x  3} = (   , 3)
(  3,3)
( 3, ) B. y  intercept = f (0)=
9
, no x
1
 intercept C. f ( x)= f (x) f is even; the curve is symmetric about the y axis. D. lim 2
=0 , so
x   x 9
1 1
y=0 is a HA. lim 2
=  , lim 2
= ,

x 3 x 9 x
+
3 x 9
1 1 / 2x
lim 2
= , lim 2
=  >0 x<0 ( x  3 )
, so x=3 and x= 3 are VA. E. f (x)=
2
x  3 x 
9 x  3
+
x  9 2
x 9 ( )
so f is increasing on (   , 3) and (  3,0 ) and decreasing on ( 0,3) and ( 3, ) . F. Local maximum

value f (0)=
1
. G. y
/ /
=
( 2
)2
 2 x  9 +(2x)2 x  9 ( 2x ) ( 2
) =
(
6 x +3
2
) >0 2
x >9 x>3 or x< 3 , so f
9 4 3
( x  9)
2
( x  9)
2

is CU on (   , 3) and ( 3, ) and CD on (  3,3) . No IP

H.

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

( 2
)
12. y= f (x)=x/ x  9 A. D= { x| x  3} = (   , 3)
(  3,3)
( 3, ) B. x  intercept =0 , y intercept
x
= f (0)=0. C. f ( x)= f (x) , so f is odd; the curve is symmetric about the origin. D. lim 2
=0 , so
x   x 9
x x x x
y=0 is a HA. lim 2
= , lim 2
=  , lim 2
= , lim 2
=  , so x=3 and x= 3
 
x
+
3 x 9 x 3 x 9 x
+
 3 x 9 x  3 x 9

are VA. E. f (x)=

/ ( x  9)  x(2x) = x +9 <0 ( x  3 ) so f is decreasing on (   , 3) , (  3,3) ,
2 2

( x2 9) 2 ( x2 9) 2
2
2x ( x  9 )  ( x +9 )  2 ( x  9 ) ( 2x ) 2x ( x +27)
2 2 2 2
/ /
and ( 3, ) . F. No extreme values G. f (x)= = >0
4 3
( x  9)
2
( x  9)
2

when  3<x<0 or x>3 , so f is CU on (  3,0 ) and ( 3, ) ; CD on (   , 3) and ( 0,3) > IP at ( 0,0 )
H.

( 2
)
13. y= f (x)=x/ x +9 A. D=R B. y  intercept: f (0)=0 ; x  intercept: f (x)=0 x=0 C. f ( x)= f (x) ,
so f is odd and the curve is symmetric about the origin. D. lim
x  
x / ( x +9)
2
=0 , so y=0 is a HA;

no VA E. f
/
(x)=
( x +9 ) ( 1 )  x(2x)
2
=
9 x
2
=
(3+x)(3 x)
>0  3<x<3 , so f is increasing on
2 2 2
( x +9)
2
( x +9) ( x +9)
2 2

(  3,3) and decreasing on (   , 3) and ( 3, ) .

1 1 / /
F. Local minimum value f ( 3)= , local maximum value f (3)= G. f (x)
6 6
2
=
( x +9 ) ( 2x) ( 9 x )  2 ( x +9 ) (2x) (2x) ( x +9 )  ( x +9 )  2 ( 9 x )
2 2 2
=
2 2 2
=
(
2x x  27
2
) =0 x=0 ,
2 2
( x +9)
2
( x2+9) 4 ( x +9)
2 3

/ /
 27= 3 3 f (x)>0  3 3 <x<0 or x>3 3 , so f is CU on (  3 3,0 ) and ( 3 3, ) , and CD
1
on (   , 3 3 ) and ( 0,3 3 ) . There are three inflection points: ( 0,0 ) and  3 3, 3 .
12
H.

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

14. y= f (x)=x
2
/ ( x +9) 2
A. D=R B. y  intercept: f (0)=0 ; x  intercept: f (x)=0 x=0 C. f ( x)= f (x) ,
so f is even and symmetric about the y  axis. D. lim
x  
x
2
/ ( x +9) 2
=1 , so y=1 is a HA; no VA E.

f
/
(x)=
( x +9 ) ( 2x )  x (2x)
2 2
=
18x
>0 x>0 , so f is increasing on ( 0, ) and decreasing on
2 2
( x +9)
2
( x +9)
2

/ /
(   ,0 ) . F. Local minimum value f (0)=0 ; no local maximum G. f (x)
2
=
( x +9 ) (18) 18x 2 ( x +9 )  2x 18 ( x +9 ) ( x +9 )  4x
2 2
=
2 2 2
=
(
18 9 3x
2
)
2 2
( x +9)
2
( x2+9) 4 ( x2+9) 3
 54 ( x+ 3 ) ( x 3)
= >0  3 <x< 3 so f is CU on (  3, 3 ) and CD on (   , 3 ) and
3
( x +9)
2

1
( 3, ) . There are two inflection points:  3,
4
.
H.

x 1
15. y= f (x)= A. D= { x| x 0} = (   ,0 )
( 0, ) B. No y  intercept; x  intercept: f (x)=0 x=1 C.
2
x
x 1 x 1
No symmetry D. lim =0 , so y=0 is a HA. lim =  , so x=0 is a VA. E.
2 2
x   x x 0 x
2 2
/ x  1 (x 1) 2x  x +2x  (x 2) / /
f (x)= = = , so f (x)>0 0<x<2 and f (x)<0 x<0 or x>2 . Thus, f
2 2 4 3
(x )x x
is increasing on ( 0,2 ) and decreasing on (   ,0 ) and ( 2, ) . F. No local minimum, local maximum
3 2 3 2
1 / / x  (  1 )   (x 2)  3x 2x  6x 2(x 3) / /
value f (2)= . G. f (x)= = = .f (x) is negative on
4 3 2 6 4
(x ) x x

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

(   ,0 ) and ( 0,3) and positive on ( 3, ) , so f is CD on (   ,0 ) and ( 0,3) and CU on ( 3, ) . IP at

2
3,
9
H.

2
x 2
16. y= f (x)= A. D= { x| x 0} = (   ,0 )
( 0, ) B. No y  intercept; x  intercepts: f (x)=0
4
x
2
x 2
x= 2 C. f ( x)= f (x) , so f is even; the curve is symmetric about the y  axis. D. lim =0 , so
4
x   x
2
x 2
y=0 is a HA. lim =  , so x=0 is a VA. E.
4
x 0 x
/
f (x)=
4
x  2x ( x  2) (4x3) =  2x5+8x3 =  2(x2 4) =  2(x+2)(x 2)
2
/
. f (x) is negative on (  2,0 ) and
( x4 ) 2 x
8
x
5
x
5

( 2, ) and positive on (   , 2 ) and ( 0,2 ) , so f is decreasing on (  2,0 ) and ( 2, ) and increasing
1
on (   , 2 ) and ( 0,2 ) . F. Local maximum value f ( 2)= , no local minimum. G.
8

f
/ /
(x)=
5
x  ( 4x)+2 x  4  5x (2 ) 4
=
2x
4 2
 2x +5 x  4( 2
) =
(
2 3x  20
2
) f
/ /
(x) is positive on
( x5) 2 x
10
x
6

20 20 20 20
  , and , and negative on  ,0 and 0, , so f is
3 3 3 3
20 20 20 20
CU on   , and , and CD on  ,0 and 0, . IP at
3 3 3 3
20 21
 ,  (  2.5820,0.105)
3 200
H.

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

17. y= f (x)=
x
2
=
( x +3)  3
2
=1
3
A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0
2 2 2
x +3 x +3 x +3
2
x
x=0 C. f ( x)= f (x) , so f is even; the graph is symmetric about the y  axis. D. lim 2
=1 , so
x   x +3
/  2x 6x /
y=1 is a HA. No VA. E. Using the Reciprocal Rule, f (x)= 3 = . f (x)>0 x>0
2 2
( x +3) ( x +3)
2 2

/
and f (x)<0 x<0 , so f is decreasing on (   ,0 ) and increasing on ( 0, ) . F. Local minimum
2
value f (0)=0 , no local maximum. G. f
/ /
(x)=
( x +3)  6 6x 2 ( x +3)  2x
2 2

2
( x2+3) 2
=
(
6 x +3
2
) ( x2+3)  4x2 =
( 2) =  18(x+1)(x 1)
6 3 3x
( x2+3) 4 ( x2+3) 3 ( x2+3) 3
/ /
f (x) is negative on (  
, 1 ) and ( 1, ) and positive on (  1,1 ) , so f is CD on (   , 1 ) and
1
( 1, ) and CU on (  1,1 ) . IP at  1,
4
H.

3
x 1
18. y= f (x)= 3
A. D= { x| x  1} = (   , 1 )
(  1, ) B. x  intercept =1 , y  intercept = f (0)= 1 C.
x +1
3 3
x 1 1 1/x
No symmetry D. lim 3
= lim =1 , so y=1 is a HA.
3
x   x +1 x   1+1/x

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3 3
x 1 x 1
lim 3
= and lim 3
=  , so x= 1 is a VA. E.
 +
x  1 x +1 x  1 x +1
/
f (x)=
( x +1 ) ( 3x )  ( x  1 ) ( 3x )
3 2
=
2
6x
2 3
>0 ( x  1 ) so f is increasing on (   , 1 ) and
2 2
( x +1)
3
( x +1)
3

2
/ / 12x ( x +1 )  6x  2 ( x +1 )  3x 12x ( 1 2x )
3 2 3 2 3
(  1, ) . F. No extreme values G. y = 4
=
3
>0 x< 1 or
( x +1)
3
( x +1)
3

1 1 1
0<x< 3
, <so f is CU on (   , 1 ) and 0, 3
and CD on (  1,0 ) and 3
, . IP at
2 2 2
1 1
( 0, 1 ) , ,
3
2 3
H.

19. y= f (x)=x 5 x A. The domain is { x|5 x 0} = (   ,5 B. y intercept: f (0)=0 ; x  intercepts:

f (x)=0 x=0 , 5 C. No symmetry D. No asymptote E.
/ 1  1/2 1/2 1  1/2 10 3x 10
f (x)=x (5 x) (  1 ) +(5 x)  1= (5 x)  x+2(5 x) = >0 x< , so f is
2 2 2 5 x 3
10 10
increasing on   , and decreasing on ,5 . F. Local maximum value
3 3
10 10 / /
f = 15 4.3 ; no local minimum G. f (x)
3 9
1/2 1  1/2
2(5 x) ( 3) (10 3x) 2 (5 x) ( 1)  1/2
2 (5 x)  6(5 x)+(10 3x) 3x 20 / /
= = = f (x)<0
2 4(5 x) 3/2
( 2 5 x ) 4(5 x)
for x<5 , so f is CD on (   ,5) . No IP
H.

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2
20. y= f (x)=2 x  x A. D=[0, )] B. y intercept: f (0)=0 ; x  intercepts: f (x)=0 2 x =x 4x=x 
2 / 1 1
4x x =0 x(4 x)=0 x=0 , 4 C. No symmetry D. No asymptote E. f (x)=  1= ( 1 x ) .
x x
This is positive for x<1 and negative for x>1 , so f is increasing on ( 0,1 ) and decreasing on ( 1, ) .
 1/2 / 1  3/2  1
/ /
F. Local maximum value f (1)=1 , no local minimum. G. f (x)= x  1 =
2
x =
3/2
(
<0 for )
2x
x>0 , so f is CD on ( 0, ) . No IP
H.

2
21. y= f (x)= x +1  x A. D=R B. No x  intercept, y  intercept =1 C. No symmetry
2
D. lim
x  
( 2
x +1  x =) and lim
x 
( 2
x +1  x =lim) x 
( 2
x +1  x ) x +1 +x
2
=lim
x  2
1
=0 ,
x +1 +x x +1 +x
2
/ x x x +1 /
so y=0 is a HA. E. f (x)=  1=  f (x)<0 , so f is decreasing on R. F. No
2 2
x +1 x +1
/ / 1
extreme values G. f (x)= >0 , so f is CU on R . No IP
3/2
( x +1)
2

H.

22. y= f (x)= x/(x 5) A. D= { x| x/(x 5) 0} = (   ,0

( 5, ) . B. Intercepts are 0 . C. No symmetry
x 1 x
D. lim = lim =1 , so y=1 is a HA. lim = , so x=5 is a VA. E.
x   x 5 x   1 5/x + x 5
x 5
1 x  1/2
5 3  1/2
/ ( 5)
f ( x) = = x(x 5) <0 , so f is decreasing on (   ,0 ) and ( 5, ) . F.
2 x 5 2 2
(x 5)
/ / 5 3  3/2 2 / /
No extreme values G. f (x)= x ( x 5) (x 5) (4x 5)>0 for x>5 , and f (x)<0 for x<0 , so f
4
11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

is CU on ( 5, ) and CD on (   ,0 ) . No IP
H.

2
23. y= f (x)=x/ x +1 A. D=R B. y  intercept: f (0)=0 ; x  intercepts: f (x)=0 x=0 C. f ( x)= f (x) ,
so f is odd; the graph is symmetric about the origin. D. lim f (x)
x 
x x/x x/x 1
=lim =lim =lim =lim lim f (x)
x  2 x  2 x  2 2 x  2 x  
x +1 x +1 /x x +1 / x 1+1/x
x x/x x/x 1 1
= lim = lim = lim = lim = = 1 so
x  
x +1
2 x   2
x +1 /x x  

2
2
x +1 /( )
2x
 x
2 x  
 1+1/x
2  1+0

x +1  x
2 2 2
/ 2 x +1 x +1 x 1
y= 1 are HA. No VA. E. f (x)= = = >0 for all x , so f is
2 1/2 2 3/2 3/2
[(x +1) ] x +1 x +1 ( 2
) ( 2
)
3 2  5/2  3x
/ /
increasing on R . f (x)=
2
x +1  2x=
2
(
5/2
/ /
)
, so f (x)>0 for x<0 and f
/ /
(x)<0 for
(x +1)
x>0 . Thus, f is CU on (   ,0 ) and CD on ( 0, ) . IP at ( 0,0 ) H.

2
24. y= f (x)=x 2 x A. D=  2 , 2 B. y  intercept: f (0)=0 ; x  intercepts: f (x)=0
x=0 ,  2 . C. f ( x)= f (x) , so f is odd; the graph is symmetric about the origin. D. No asymptote
2 2
/  x 2  x +2 x 2(1+x)(1 x) /
E. f (x)=x + 2 x = = . f (x) is negative for  2 <x< 1 and
2 2 2
2 x 2 x 2 x
1<x< 2 , and positive for  1<x<1 , so f is decreasing on (  2 , 1 ) and ( 1, 2 ) and increasing on
/ /
(  1,1 ) . F. Local minimum value f ( 1)= 1 , local maximum value f (1)=1 . G f (x)

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2 2  x
2 x ( 4x) (2 2x )
2 2 2 3 2
2 x (2 x )( 4x)+(2 2x )x 2x  6x 2x(x  3) 2
= = = = . Since x  3<0 for x
2 1/2 2 2 3/2 2 3/2 2 3/2
(2 x ) (2 x ) (2 x ) (2 x )
/ / / /
in  2 , 2 , f (x)>0 for  2 <x<0 and f (x)<0 for 0<x< 2 . Thus, f is CU on (  2 ,0 ) and
CD on ( 0, 2 ) . The only IP is ( 0,0 ) .
H.

2
25. y= f (x)= 1 x /x A. D= { x |x| 1,x 0 } =  1,0 )
( 0,1 B. x intercepts  1 , no y  intercept C.
2 2
1 x 1 x
f ( x)= f (x) , so the curve is symmetric about ( 0,0 ) . D. lim = , lim =  , so
+ x  x
x 0 x 0

x=0 is a VA. E. f
/
(x)=
( x / 2
1 x
2
) 1 x
2
=
1
<0 , so f is decreasing on (  1,0 ) and
2 2 2
x x 1 x
2
/ / 2 3x 2 2
( 0,1 ) . F. No extreme values G. f (x)= >0  1<x< or 0<x< , so f is
3 3
x
3
( 1 x2) 3/2
2 2 2 2
CU on  1, and 0, and CD on  ,0 and ,1 . IP at
3 3 3 3
2 1
 , H.
3 2

2
26. y= f (x)=x/ x  1 A. D= (   , 1 )
( 1, ) B. No intercepts C. f ( x)= f (x) , so f is odd; the graph
is symmetric about the origin. D.

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

x x
lim =1 and lim = 1 , so y= 1 are HA. lim f (x)=+ and lim f (x)=  , so
x  2 x   2 + 
x 1 x 1 x 1 x  1
2 x
x  1  x
2 2 2
/ x 1 x  1 x  1
x= 1 are VA. E. f (x)= = = <0 , so f is decreasing on
1/2 2 3/2 3/2
( x  1)
2
( x  1)
2
( x  1)
2

(   , 1 ) and ( 1, ) . F. No extreme values G. f

/ /
(x)= (  1 ) 
3
2
( x2 1)  5/2 2x= 3x
( x2 1) 5/2
/ / / /
f (x)<0 on (   , 1 ) and f (x)>0 on ( 1, ) , so f is CD on (   , 1 ) and CU on ( 1, ) . No IP
H.

1/3 1/3 3
27. y= f (x)=x 3x A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0 x=3x  x =27x
3 2
x  27x=0 x(x  27)=0 x=0 ,  3 3 C. f ( x)= f (x) , so f is odd; the graph is symmetric about the
2/3
/  2/3 1 x  1 / /
origin. D. No asymptote E. f (x)=1 x =1 = . f (x)>0 when x >1 and f (x)<0 when
2/3 2/3
x x
0< x <1 , so f is increasing on (   ) , and decreasing on (  1,0 ) and ( 0,1 ) . F. Local
, 1 ) and ( 1,
/ / 2  5/3
maximum value f ( 1)=2 , local minimum value f (1)= 2 G. f (x)= x <0 when x<0 and
3
/ /
f (x)>0 when x>0 , so f is CD on (   ,0 ) and CU on ( 0, ) . IP at ( 0,0 )
H.

5/3 2/3 2/3

28. y= f (x)=x  5x =x (x 5) A. D=R B. x  intercepts 0 , 5 ; y  intercept 0 C. No symmetry D.
2/3 / 5 2/3 10  1/3 5  1/3
lim x (x 5)=  , so there is no asymptote E. f (x)= x  x = x (x 2)>0 x<0 or
x   3 3 3
x>2 , so f is increasing on (   ,0 ) , ( 2, ) and decreasing on ( 0,2 ) . F. Local maximum value
3
f (0)=0 , local minimum value f (2)= 3 4 G.
14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

/ / 10  1/3 10  4/3 10  4/3

f (x)= x + x = x (x+1)>0 x> 1 , so f is CU on (  1,0 ) and ( 0, ) , CD on
9 9 9
(  , 1 ) . IP at (  1, 6 ) H.

29. y= f (x)=x+ x A. D=R B. x  intercepts 0 ,  1 ; y  intercept 0 C. No symmetry D.

lim ( x+ x ) = , lim ( x+ x ) =  . No asymptote E. For x>0 , f (x)=x+ x 
x  x  
/ 1 / 1
f (x)=1+ >0 , so f increases on ( 0, ) . For x<0,f(x)=x+  x  f (x)=1 >0 2  x >1
2 x 2  x
1 1 1 1
  x>  , so f increases on   ,
x< and decreases on  ,0 . F. Local maximum
4 4 4 4
1 1 / / 1  3/2 / /
value f  = , local minimum value f (0)=0 G. For x>0 , f (x)= x  f (x)<0 , so f
4 4 4
/ / 1  3/2 / /
is CD on ( 0, ) . For x<0 , f (x)= ( x)  f (x)<0 , so f is CD on (   ,0 ) . No IP
4
H.

3
30. y= f (x)= ( x2 1) 2 = ( x2 1) 2/3 A. D=R B. x  intercepts  1 , y  intercept 1 C. f ( x)= f (x) , so
2/3
the curve is symmetric about the y axis. D. lim ( x  1 ) = , no asymptote E.
2

x  
4  1/3
/
f (x)= x x  1
3
2
( )/ /
 f (x)>0 x>1 or  1<x<0 , f (x)<0 x< 1 or 0<x<1 . So f is increasing on
(  1,0 ) , ( 1, ) and decreasing on (   , 1 ) , ( 0,1 ) . F. Local minimum values f ( 1)= f (1)=0 , local
4 2  1/3 4 1  4/3
maximum value f (0)=1 G. f (x)=
/ /
3
x 1 (
+ x 
3 3
) 2
x 1 (2x) ( )
4 2  4/3
=
9
(
x 3 x 1
2
)( )
>0 x > 3 so f is CU on (   , 3 ) , ( 3, ) and CD on (  3, 1 ) ,

(  1,1 ) , ( 1, 3 ) . IPs at (  ) H.
3
3, 4

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3 2
31. y= f (x)=3sin x sin x A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0 sin x(3 sin x)=0
2
sin x=0 [ since sin x 1<3 ]  x=n , n an integer.
C. f ( x)= f (x) , so f is odd; the graph (shown for  2  x 2 ) is symmetric about the origin and
/
periodic with period 2 . D. No asymptote E. f (x)=3cos x 3sin x cos x=3cos x 1 sin x =3cos x .
2
( 2
) 3

/   /
f (x)>0 cos x>0 x 2n  ,2n + for each integer n , and f (x)<0
2 2
 3
cos x<0 x 2n + ,2n + for each integer n . Thus, f is increasing on
2 2
   3
2n  ,2n + for each integer n , and f is decreasing on 2n + ,2n + for each
2 2 2 2

integer n . F. f has local maximum values f (2n + )=2 and local minimum values
2
3
f (2n + )= 2 .
2
/ / 2 2 / /
G. f (x)= 9sin x cos x= 9sin x(1 sin x)= 9sin x(1 sin x)(1+sin x) . f (x)<0 sin x>0 and
  / /
sin x  1 x 2n ,2n +
2n + ,2n + for some integer n . f (x)>0 sin x<0 and
2 2
 
sin x  1 x (2n 1) ,(2n 1) +
(2n 1) + ,2n for some integer n . Thus, f is CD
2 2
1 1
on the intervals 2n , 2n+  and 2n+  , ( 2n+1 )  for each integer n , and f is
2 2
1 1
CU on the intervals (2n 1) , 2n  and 2n  ,2n for each integer n . f has
2 2
inflection points at ( n ,0 ) for each integer n .
H.


32. y= f (x)=sin x tan x A. D= { x| x (2n+1)
2 } B. y=0 sin x=tan x=
sin x
cos x
 sin x=0 or cos x=1

16
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

 x=n ( x intercepts), y  intercept = f (0)=0 C. f ( x)= f (x) , so the curve is symmetric about ( 0,0 )
. Also periodic with period 2 D. lim (sin x tan x)=  and lim (sin x tan x)= , so
 +
x ( /2) x ( /2)
 / 2
x=n + are VA. E. f (x)=cos x sec x 0 , so f decreases on each interval in its domain, that is, on
2
 
(2n 1)
2
,(2n+1)
2
. F. No extreme values G. f
/ / 2
(x)= sin x 2sec x tan x=sin x 1+2sec x . ( 3
)
3 3 1 / /  3
Note that 1+2sec x 0 since sec x  .f (x)>0 for  <x<0 and <x<2 , so f is CU on
2 2 2
1 1 /
n  ,n and CD on n , n+  . f has IPs at ( n ,0 ) . Note also that f (0)=0 ,
2 2
/
but f ( )= 2 .
H.

   
33. y= f (x)=x tan x ,  <x< A. D=  , B. Intercepts are 0 C. f ( x)= f (x) , so the curve
2 2 2 2
 
is symmetric about the y axis. D. lim x tan x= and lim x tan x= , so x= and x=
 + 2 2
x ( /2) x  ( /2)
/ 2  
are VA. E. f (x)=tan x+x sec x>0 0<x< , so f increases on 0, and decreases on
2 2

 ,0 .
2
/ / 2 2  
F. Absolute and local minimum value f (0)=0 . G. y =2sec x+2x tan x sec x>0 for  <x< , so
2 2
 
f is CU on  , . No IP
2 2
H.

   
34. y= f (x)=2x tan x ,  <x< A. D=  , B. y intercept: f (0)=0 ; x  intercepts: f (0)=0
2 2 2 2
17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

 2x=tan x x=0 or x  1.17 C. f ( x)= f (x) , so f is odd; the graph is symmetric about the origin.

D. lim ( 2x tan x ) = and lim ( 2x tan x ) =  , so x= are VA. No HA. E.
+  2
x (   /2 ) x ( /2 )
/ 2 /  
f (x)=2 sec x<0 sec x > 2 and f (x)>0 sec x < 2 , so f is decreasing on  , ,
2 4
   
increasing on  , , and decreasing again on , F. Local maximum value
4 4 4 2
    / /
f =  1 , local minimum value f  = +1 G. f (x)
4 2 4 2
2 2 / / 
= 2sec x sec x tan x= 2tan x sec x= 2tan x(tan x+1) so f (x)>0 tan x<0  <x<0 , and
2
/ /   
f (x)<0 tan x>0 0<x< . Thus, f is CU on  ,0 and CD on 0, . f has an IP at
2 2 2
( 0,0 ) . H.

1
35. y= f (x)= x sin x , 0<x<3 A. D= ( 0,3 ) B. No y  intercept. The x  intercept, approximately 1.9
2
/ 1
, can be found using Newton’s Method. C. No symmetry D. No asymptote E. f (x)=  cos\:,x>0
2
1 
5 7  5 7
cos x<  <x<
or <x<3 , so f is increasing on , and ,3 and
23 3 3 3 3 3
 5 7   3
decreasing on 0, and , . F. Local minimum value f =  , local
3 3 3 3 6 2
5 5 3 7 7 3
maximum value f = + , local minimum value f =  G.
3 6 2 3 6 2
/ /
f (x)=sin x>0 0<x< or 2 <x<3 , so f is CU on ( 0, ) and ( 2 ,3 ) and CD on ( ,2 ) .

IPs at  , and ( 2 , ) . H.
2

18
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2
36. y= f (x)=cos x 2sin x A. D=R B. y intercept: f (0)=1 C. No symmetry, but f has period 2 . D.
/ /
No asymptote E. y =2cos x( sin x) 2cos x= 2cos x(sin x+1) . y =0 cos x=0 or sin x= 1
 / /
x=(2n+1) . y >0 when cos x<0 since sin x+1 0 for all x . So y >0 and f is increasing on
2
  /  
(4n+1) ,(4n+3) ; y <0 and f is decreasing on (4n 1) ,(4n+1) . F. Local maximum
2 2 2 2
 
values f (4n+3) =2 , local minimum values f (4n+1) = 2 G.
2 2
/
y = 2cos x(sin x+1)= sin 2x 2cos x y
/ /
= 2cos 2x+2sin x= 2 1 2sin x +2sin x ( 2
)
2 / / 1  5
=4sin x+2sin x 2=2(2sin x 1)(sin x+1)y =0 sin x= or  1 x= +2n , +2n , or
2 6 6
3 / /  5 / /
+2n . y >0 and f is CU on +2n , +2n ; y  0 and f is CD on
2 6 6
5   1 5 1
+2n , +2(n+1) .IPs at +2n , and +2n ,
6 6 6 4 6 4
. H.

37. y= f (x)=sin 2x 2sin x A. D=R B. y  intercept = f (0)=0 . y=0 2sin x=sin 2x=2sin xcos x
sin x=0 or cos x=1 x=n ( x intercepts) C. f ( x)= f (x) , so the curve is symmetric about ( 0,0 ) .
Note: f is periodic with period 2 , so we determine E   G for    x  . D. No asymptotes E.
1 2
/ 2
( )
f (x)=2cos 2x 2cos x=2 2cos x 1 cos x =2(2cos x+1)(cos x 1)>0 cos x<
2
   <x<
3
or
2 2 2 2 2
<x< , so f is increasing on   , , , and decreasing on  , . F.
3 3 3 3 3
2 3 3 2 3 3
Local maximum value f  = , local minimum value f = G.
3 2 3 2
/ / 1  11
f (x)= 4sin 2x+2sin x=2sin x(1 4cos x)=0 when x=0 ,   or cos x= . If  =cos , then f is
4 4
3 15
CU on (   ,0 ) and ( , ) and CD on (   ,  ) and ( 0, ) . IPs at ( 0,0 ) , (   ,0 ) ,  , ,
8
3 15
 , .
8
H.

19
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

38. f (x)=sin x x A. D=R B. x intercept =0=y intercept C. f ( x)=sin ( x) ( x)= (sin x x)= f (x) ,
/
so f is odd. D. No asymptote E. f (x)=cos x 1 0 for all x , so f is decreasing on (   , ) . F. No
/ / / /
extreme values G. f (x)= sin x f (x)>0 sin x<0 (2n 1) <x<2n , so f is CU on
( (2n 1) ,2n ) and CD on ( 2n ,(2n+1) ) , n an integer. Points of inflection occur when x=n .
H.

when
sin x cos x 1 sin x 1 cos x sin x(1 cos x) 1 cos x
39. y= f (x)= =  = = =csc x cot x A.
1+cos x 1+cos x 1 cos x 2 sin x
sin x

The domain of f is the set of all real numbers except odd integer multiples of  . B. y  intercept:
f (0)=0 ; x  intercepts: x=n , n an even integer. C. f ( x)= f (x) , so f is an odd function; the graph
is symmetric about the origin and has period 2 . D. When n is an odd integer, lim f (x)= and

x ( n )
lim f (x)=  , so x=n is a VA for each odd integer n . No HA. E.
+
x ( n )
/ (1+cos x) cos x sin x( sin x) 1+cos x 1 /
f (x)= = = . f (x)>0 for all x except odd multiples
2 2 1+cos x
(1+cos x) (1+cos x)
of  , so f is increasing on ( (2k 1) ,(2k+1) ) for each integer k . F. No extreme values G.
/ / sin x / /
f (x)= >0 sin x>0 x ( 2k ,(2k+1) ) and f (x)<0 on ( (2k 1) ,2k ) for each
2
(1+cos x)
integer k . f is CU on ( 2k ,(2k+1) ) and CD on ( (2k 1) ,2k ) for each integer k . f has IPs at
( 2k ,0 ) for each integer k .
H

20
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

40. y= f (x)=cos x/(2+sin x) A. D=R Note: f is periodic with period 2 , so we determine B  G on

 3 1
[0,2 ]. B. x  intercepts , , y  intercept = f (0)= C. No symmetry other than periodicity D.
2 2 2
/ (2+sin x)( sin x) cos x(cos x) 2sin x+1 /
No asymptote E. f (x)= = . f (x)>0 2sin x+1<0
2 2
(2+sin x) (2+sin x)
1 7 11 7 11 7
sin x<  <x< , so f is increasing on , and decreasing on 0, ,
2 6 6 6 6 6
11 7 1 11 1
,2 . F. Local minimum value f = , local maximum value f =
6 6 3 6 3
2
/ / (2+sin x) (2cos x) (2sin x+1)2(2+sin x)cos x 2cos x(1 sin x)
G. f (x)= = >0 cos x<0
4 3
(2+sin x) (2+sin x)
 3  3  3 
<x< , so f is CU on , and CD on 0, and ,2 . IP at ,0 ,
2 2 2 2 2 2 2
3
,0
2
H.

 x
41. y=1/ 1+e ) A. D=R B. No x  intercept; y  intercept = f (0)= 12 . C. No symmetry D.
(
 x 1  x  x
lim 1/ ( 1+e ) = =1 and lim 1/ ( 1+e ) =0 (since lim e = ), so f has horizontal asymptotes
x  1+0 x   x  
 x  2  x  x  x 2
y=0 and y=1 . E. f (x)= 1+e
/
 e ( ) ( ) =e / ( 1+e ) . This is positive for all x , so f is
increasing on R . F. No extreme values
G.
 x 2  x  x  x  x
/ / ( 1+e ) (  e )  e ( 2 ) ( 1+e ) (  e )
f (x) =
( 1+e x) 4
21
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

 x
=
e ( e x 1)
( 1+e x) 3
The second factor in the numerator is negative for x>0 and positive for x<0 , and the other factors are
1
always positive, so f is CU on (   ,0 ) and CD on ( 0, ) . f has an inflection point at 0, .
2
H.

2x x 2x x x
42. y= f (x)=e  e A. D=R B. y intercept: f (0)=0 ; x  intercepts: f (x)=0 e =e  e =1 x=0. C.
2x x
No symmetry D. lim e  e =0 , so y=0 is a HA. No VA. E. f (x)=2e  e =e 2e  1 , so f (x)>0
/ 2x x x
( x
) /

x  
1
x 1 / x 1 1 1
 e>  x>ln = ln 2 and f (x)<0 e <  x<ln , so f is decreasing on   ,ln and
2 2 2 2 2
1 1 2ln (1/2) ln (1/2) 1 2 1 1
increasing on ln , . F. Local minimum value f ln =e  e =  =
2 2 2 2 4
/ / 2x
G. f (x)=4e  e =e 4e  1 , so f (x)>0
x x
( x
) / /

x 1 1 / / 1
e >  x>ln and f (x)<0 x<ln .
4 4 4
H.

1 1
Thus, f is CD on   ,ln and CU on ln , . f has an IP at
4 4
1 1 2 1 1 3
ln ,  = ln , .
4 4 4 4 16

43. y= f ( x ) =xln x A. D= ( 0, ) B. x intercept when ln x=0 x=1 , no y intercept

22
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

C. No symmetry D. lim xln x= ,

x 
ln x 1/x /
lim xln x=lim =lim =lim (  x ) =0 , no asymptote. E. f ( x ) =ln x+1=0 when ln x= 1
1/x 2
x 0  1/x
+ + + +
x 0 x 0 x 0
 1 /  1
x=e . f ( x ) >0 ln x> 1 x>e , so f is increasing on ( 1/e, ) and decreasing on ( 0,1/e ) . F.
/ /
f (1/e)= 1/e is an absolute and local minimum value. G. f ( x ) =1/x>0 , so f is CU on ( 0, ) . No
IP
H.

x x
x e e
44. y= f (x)=e /x A. D= { x| x 0} B. No intercept C. No symmetry D. lim =lim = ,
x  x x  1
x x x
e e e
lim =0 , so y=0 is a HA. \ lim = , lim =  , so x=0 is a VA.
x   x + x  x
x 0 x 0
x x
/ xe  e x
E. f (x)= >0 (x 1)e >0 x>1 ,
2
x
so f is increasing on ( 1, ) , and decreasing on (   ,0 ) and ( 0,1 ) . F. f (1)=e is a local minimum
value.

G. f
/ /
(x)=
2
( x) (
x xe  2x xe  e
x x
) = ex ( x2 2x+2) >0
4 3
x x
2
 x>0 since x  2x+2>0 for all x . So f is CU on ( 0, ) and CD on (   ,0 ) . No IP
H.

 x  x x 1
45. y= f (x)=xe A. D=R B. Intercepts are 0 C. No symmetry D. lim xe =lim =lim =0 , so
x x
x  x  e x  e
y=0 is a HA.
23
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

 x /  x  x  x
lim xe =  E. f (x)=e  xe =e (1 x)>0 x<1 , so f is increasing on (   ,1 ) and decreasing
x  
on ( 1, ) . F. Absolute and local maximum value f (1)=1/e.
H.

 x
) and CD on (   ,2 ) . IP at ( 2,2/e )
/ / 2
G. f (x)=e ( x 2 ) >0 x>2 , so f is CU on ( 2,

) (
46. y= f (x)=ln x  3x+2 =ln (x 1)(x 2)
2

A. D= { xinR:x  3x+2>0} = (  
2
,1 )
( 2, ).
2 0 2 3 5
B. y  intercept: f (0)=ln 2 ; x  intercepts: f (x)=0 x  3x+2=e  x  3x+1=0 x=  x 0.38 ,
2
2.62 C. No symmetry D. lim f (x)=lim f (x)=  , so x=1 and x=2 are VAs. No HA. E.
 +
x 1 x 2
/ 2x 3 2(x 3/2) / /
f (x)= = , so f (x)<0 for x<1 and f (x)>0
2 (x 1)(x 2)
x  3x+2
for x>2 . Thus, f is decreasing on (   ,1 ) and increasing on ( 2, ) . F. No extreme values
G.

/ /
( x  3x+2 )  2 (2x 3)
2 2

f (x) =
( x2 3x+2) 2
2 2
2x  6x+4 4x +12x 9
=
( x2 3x+2) 2
2
 2x +6x 5
=
( x2 3x+2) 2
/ /
The numerator is negative for all x and the denominator is positive, so f (x)<0 for all x in the
domain of f . Thus, f is CD on (   ,1 ) and ( 2, ) . No IP
H.

24
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

47. y= f (x)=ln (sin x)

A.

D= { x in R|sin x>0} =
( 2n , ( 2n+1 )  )
n= 
=  
(  4 , 3 )
(  2 ,  )
( 0, )
( 2 ,3 )
  
0 
B. No y  intercept; x  intercepts: f (x)=0 ln (sin x)=0 sin x=e =1 x=2n + for each integer n .
2
C. f is periodic with period 2 . D. lim f (x)=  and lim f (x)=  , so the lines x=n
+ 
x (2n ) x [(2n+1) ]
cos x / / 
are VAs for all integers n . E. f (x)= =cot x , so f (x)>0 when 2n <x<2n + for each
sin x 2
/  
integer n , and f (x)<0 when 2n + <x<(2n+1) . Thus, f is increasing on 2n ,2n + and
2 2
 
decreasing on 2n + ,(2n+1) for each integer n . F. Local maximum values f 2n + =0
2 2
/ / 2
, no local minimum. G. f (x)= csc x<0 , so f is CD on ( 2n ,(2n+1) ) for each integer n. No IP
H.

2
48. y= f (x)=x(ln x) A. D= ( 0, ) B. x  intercept =1 , no y  intercept C. No symmetry
2
2 (ln x)
2 2(ln x)(1/x) 2ln x 2/x
D. lim x(ln x) = , lim x(ln x) =lim =lim =lim =lim =lim 2x=0 ,
 + 1/x 2 +  1/x 2
x
x 0
+
x 0 x 0
+
 1/x x 0
+
x 0 1/x x 0
+

/ 2  2
no asymptote E. f (x)=(ln x) +2ln x=(ln x)(ln x+2)=0 when ln x=0 x=1 and when ln x= 2 x=e .
/  2
f (x)>0 when 0<x<e and when x>1 , so
 2
f is increasing on 0,e) and ( 1, ) and decreasing on ( e 2,1) .
(
 2  2
F. Local maximum value f ( e ) =4e ,
25
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

local minimum value f (1)=0

/ /  1 / /
G. f (x)=2(ln x)(1/x)+2/x=(2/x)(ln x+1)=0 when ln x= 1 x=e . f (x)>0 x>1/e , so f is CU
on ( 1/e, ) , CD on ( 0,1/e ) .
IP at ( 1/e,1/e )
H.

2
 x
49. y= f (x)=xe A. D=R B. Intercepts are 0 C. f ( x)= f (x) , so the curve is symmetric
2
 x x 1
about the origin. D. lim xe = lim 2
= lim 2
=0 , so y=0 is a HA. E.
x   x   x x   x
e 2xe
2 2 2
 x 2  x  x 1 1 1 1
/
f (x)=e  2x e =e ( 1 2x2) >0 2
x<
2
 x <
2
, so f is increasing on 
2
,
2
and
1 1 1
decreasing on   , and , . F. Local maximum value f =1/ 2e , local
2 2 2
2 2 2
 x
minimum value f 
1
2
= 1/ 2e G. f
/ /
(x)= 2xe ( 1 2x2)  4xe x =2xe x ( 2x2 3) >0
3 3 3
 x> or  <x<0 , so f is CU on ,
2 2 2
3 3 3
and  ,0 and CD on   , and 0, .
2 2 2
3 3  3/2
IP are ( 0,0 ) and  , e .
2 2
H.

x  x
50. y= f (x)=e  3e  4x A. D=R B. y intercept = 2 ; x  intercept  2.22 C. No symmetry
x  x x x
 x e e e e
D. lim ( e  3e
x
 4x =lim x ) x
 3
x
 4 = , since lim =lim
x x  1
= .
x  x  x 
Similarly,

26
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

lim ( ex 3e x 4x) =  . No HA; no VA

x  
 x  x  x
/ x
( 2x
E. f (x)=e +3e  4=e e  4e +3 =e e  3 e  1 >0 e >3 or e <1
x
) ( x
)( x
) x x

x>ln 3 or x<0 . So f is increasing on (   ,0 ) and ( ln 3, ) and decreasing on ( 0,ln 3) . F. Local

maximum value f (0)= 2 ,
 x  x 2x 1
/ / x 2x
local minimum value f (ln 3)=2 4ln 3 G. f (x)=e  3e =e e  3 >0 e >3 x> ln 3 , so f is
2
( )
1 1 1
CU on ln 3, and CD on   , ln 3 . IP at ln 3, 2ln 3 .
2 2 2
H.

3x  2x
51. y= f (x)=e +e A. D=R B. y intercept = f (0)=2 ; no x  intercept C. No symmetry D. No
asymptotes
/ 3x  2x / 3x  2x 2x 5x 2 2
E. f (x)=3e  2e , so f (x)>0 3e >2e [ multiply by e ] e >  5x>ln 
3 3
1 2 / 1 2
x> ln   0.081 . Similarly, f (x)<0 x< ln .
5 3 5 3
1 2 1 2
f is decreasing on   , ln and increasing on ln , .
5 3 5 3
1 2 2 3/5 2  2/5
F. Local minimum value f ln = +  1.96 ; no local maximum.
5 3 3 3
/ / 3x  2x / /
G. f (x)=9e +4e , so f (x)>0 for all x , and f is CU on (   , ) . No IP
H.

x 1  1
52. y= f (x)=tan A. D= { x| x  1}
x+1
B. x  intercept =1 , y  intercept
27
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

 1 
= f (0)=tan ( 1)= C. No symmetry D.
4
 1 x 1  1 1 1/x  1  
lim tan = lim tan =tan 1= , so y= is a HA. Also
x   x+1 x   1+1/x 4 4
 1 x 1   1 x 1 
lim tan = and lim tan = . E.
+ x+1 2  x+1 2
x  1 x  1
1 (x+1) (x 1)
/ =
f (x) 2 2
1+ (x 1)/(x+1) (x+1)
2 1
= = 2 >0
2 2
(x+1) +(x 1) x +1

so f is increasing on (   , 1 ) and (  1, ) . F. No extreme values

H.

2
( x ) = 2x/ ( x +1 ) >0
/ / 2
G. f x<0 , so f is CU on (   , 1 ) and (  1,0 ) , and CD on ( 0, ).

IP at 0,
4
2
W 4 WL 3 WL 2 W 2 2 W 2
53. y=
24EI
x+
12EI
x
24EI
x =
24EI
2
x x  2Lx+L =
24EI
2 2
(
2
x (x L) =cx (x L) where )
W 2 2
c= is a negative constant and 0 x L . We sketch f (x)=cx (x L) for c= 1 . f (0)= f (L)=0
24EI
/ 2 2 /
f (x)=cx 2(x L) +(x L) (2cx)=2cx(x L) x+(x L) =2cx(x L)(2x L) . So for 0<x<L , f (x)>0
/
x(x L)(2x L)<0 (since c<0 )  L/2<x<L and f (x)<0 0<x<L/2 . So f is increasing on ( L/2,L ) and
decreasing on ( 0,L/2 ) , and there is a local and absolute minimum at ( L/2,f(L/2) ) = L/2,cL /16 . ( ) 4

f (x)=2c 1(x L)(2x L)+x(1)(2x L)+x(x L)(2) =2c ( 6x  6Lx+L ) =0

/ / / 2 2
f (x)=2c x(x L)(2x L) 
2
6L 12L 1 3
x= = L L , and these are the x  coordinates of the two inflection points.
12 2 6

28
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

k k / 2k 2k
54. F(x)= + , where k>0 and 0<x<2 . For 0<x<2 , x 2<0 , so F (x)=  >0 and
2 2 3 3
x ( x 2 ) x ( x 2 )
F is increasing. lim F(x)=  and lim F(x)= , so x=0 and x=2 are vertical asymptotes. Notice that
+ 
x 0 x 2
when the middle particle is at x=1 , the net force acting on it is 0 . When x>1 , the net force is
positive, meaning that it acts to the right. And if the particle approaches x=2 , the force on it rapidly
becomes very large. When x<1 , the net force is negative, so it acts to the left. If the particle
approaches 0 , the force becomes very large to the left.

55.
2
x +1 x 1
y= . Long division gives us: 2
x+1 2 x + 1
x + 2x
2
x +x
 x+1
 x+1
2
2
2
x +1 2 2 x
Thus, y= f (x)= =x 1+ and f (x) (x 1)= =  0 as x   . So the line y=x 1 is
x+1 x+1 x+1 1
1+
x
a slant asymptote (SA).

56.

29
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3
2x +x +x+3
2 2x  3
y= 2
. Long division gives us: 2 3 2
x +2x x + 2x 2x + x + x + 3
3 2
2x + 4x
2
 3x + x
2
 3x  6x
7x + 3
7 3
+
3 2 x 2
2x +x +x+3 7x+3 7x+3 x
Thus, y= f (x)= =2x 3+ 2 and f (x) (2x 3)= 2 =  0 as x   . So
2 2
x +2x x +2x x +2x 1+
x
the line y=2x 3 is a SA.

57.
3
4x  2x +5
2 2x  2
y= . Long division gives us: 3 2
2
2x +x 3
2
2x + x  3 4x  2x +3
3 2
4x + 2x  6x
2
 4x + 6x + 5
2
 4x  2x + 6
8x  3
3 2
4x  2x +5 8x 1
Thus, y= f (x)= 2
=2x 2+ 2
and
2x +x 3 2x +x 3
8 1

2
8x 1 x x
f (x) (2x 2)= 2
=  0 as x   . So the line y=2x 2 is a SA.
1 3
2x +x 3 2+ 
2
x x

58.

30
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

4 2 5x + 2
5x +x +x
y= . Long division gives us: 3 2 4 2
3 2
x  x +2 x  x +2 5x +x +x
4 3
5x  5x + 10x
3 2
 5x + x  9x
3 2
5x  5x + 10
2
6x  9x  10
6 9 10
4 2 2 2
 
2 3
5x +x +x 6x  9x 10 6x  9x 10 x x x
Thus, y= f (x)= 3 2 =5x+5+ 3 2 and f (x) (5x+5)= 3 2 =  0 as
x  x +2 x  x +2 x  x +2 1 2
1 +
3
x x
x   . So the line y=5x+5 is a SA.

2
 2x +5x 1
59. y= f (x)=
2x 1
= x+2+
1
2x 1
A. D= { x R| x
1
2 } =   ,
1
2

1
2
,
2  5 17
B. y  intercept: f (0)=1 ; x  intercepts: f (x)=0  2x +5x 1=0 x=  x 0.22 , 2.28 . C.
 4
1
No symmetry D. lim f (x)=  and lim f (x)= , so x= is a VA.
 + 2
x (1/2) x (1/2)
1 / 2
lim f (x) ( x+2) = lim =0 , so the line y= x+2 is a SA. E. f (x)= 1 <0 for
x   x   2x 1 2
( 2x 1 )
1 1 1
x , so f is decreasing on   , and , . F. No extreme values G.
2 2 2
/  2 / /  3 8 / / 1 / /
f (x)= 1 2(2x 1)  f (x)= 2( 2)(2x 1) (2)= , so f (x)>0 when x> and f (x)<0
3 2
( 2x 1 )
1 1 1
when x< . Thus, f is CU on , and CD on   , . No IP
2 2 2
H.

31
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2
x +12 16
60. y= f (x)= =x+2+ A. D= { x R| x 2} = (   ,2 )
( 2, ) B. y  intercept: f (0)= 6 ; no x 
x 2 x 2
intercepts. C. No symmetry D. lim f (x)=  and lim f (x)= , so x=2 is a VA.
 +
x 2 x 2
16
lim f (x) (x+2) = lim =0 , so the line y=x+2 is a slant asymptote. E.
x   x   x 2
2
/ 16 x  4x 12 (x 6)(x+2) / /
f (x)=1 = = , so f (x)>0 when x< 2 or x>6 and f (x)<0 when
2 2 2
( x 2 ) ( x 2 ) ( x 2 )
 2<x<2 or 2<x<6. Thus, f is increasing on (   , 2 ) and ( 6, ) and decreasing on (  2,2 ) and ( 2,6 ) .
F. Local maximum value f ( 2)= 4 , local minimum value f (6)=12 G.
/ /  3 32 / / / /
f (x)=16( 2)(x 2) = , so f (x)>0 for x>2 and f (x)<0 for x<2 . f is CU on ( 2, ) and
3
(x 2)
CD on (   ,2 ) . No IP
H.

( 2
)
61. y= f (x)= x +4 /x=x+4/x A. D= { x| x 0} = (   ,0 )
( 0, ) B. No intercept C. f ( x)= f (x)
symmetry about the origin D. lim ( x+4/x ) = but f (x) x=4/x 0 as x   , so y=x is a slant
x 
/ 2 2
asymptote. lim ( x+4/x ) = and lim ( x+4/x ) =  , so x=0 is a VA. E. f (x)=1 4/x >0 x >4 x>2
+ 
x 0 x 0
or x< 2 , so f is increasing on (   , 2 ) and ( 2, ) and decreasing on (  2,0 ) and ( 0,2 ) . F. Local
/ / 3
maximum value f ( 2)= 4 , local minimum value f (2)=4 G. f (x)=8/x >0 x>0 so f is CU on
( 0, ) and CD on (   ,0 ) . No IP
H.

62.
32
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

x
y= f (x)=e  x A. D=R B. No x intercept; y  intercept =1 C. No symmetry D. lim ( ex x) = ,
x  
x x x
e e e
lim ( e  x) =lim
x
x
x
 1 = since lim =lim
x x  1
= .
x  x  x 

y= x is a slant asymptote since e  x  ( x)=e  0 as x ( x

) x
 
/ x
. E. f (x)=e  1>0
x
e >1 x>0 ,
so f is increasing on ( 0, ) and decreasing on (   ,0 ) .
F. f (0)=1 is a local and absolute minimum value.
/ / x
G. f (x)=e >0 for all x , so f is CU on R . No IP
H.

3 2
2x +x +1  2x
63. y= f (x)= 2
=2x+1+ 2
A. D=R B. y  intercept: f (0)=1 ; x  intercept: f (x)=0
x +1 x +1
3 2 2
0=2x +x +1=(x+1)(2x  x+1) x= 1 C. No symmetry D. No VA
 2x  2/x
lim f (x) (2x+1) = lim 2
= lim =0 , so the line y=2x+1 is a slant asymptote. E.
2
x   x   x +1 x   1+1/x
/
(x)=2+
( x +1 ) ( 2) ( 2x)(2x) 2 ( x +2x +1 )  2x  2+4x
2
=
4 2 2 2
=
4
2x +6x
2
=
2 2
2x (x +3) /
so f (x)>0 if
2 2 2 2 2 2
( x +1)
2
( x +1)
2
(x +1) (x +1)
x 0 . Thus, f is increasing on (   ,0 ) and ( 0, ) . Since f is continuous at 0 , f is increasing on R .
2
F. No extreme values f (x)=
( x +1 )  ( 8x +12x )  ( 2x +6x )  2 ( x +1 ) (2x)
2 3
/ /
4 2 2

2
( x2+1) 2
4x ( x +1 ) ( x +1 ) ( 2x +3)  2x  6x ] 4x (  x +3)
2 2 2 4 2 2
/ /
= = so f (x)>0 for x< 3 and 0<x< 3 , and
( x2+1) 4 ( x2+1) 3
/ /
f (x)<0 for  3 <x<0 and x> 3 . f is CU on (   , 3 ) and ( 0, 3 ) , and CD on (  3,0 ) and
3
( 3, ) . There are three IPs: ( 0,1 ) ,  3,
2
3 +1  (  1.73, 1.60 ) , and
3
3, 3 +1  ( 1.73,3.60 ) .
2
H.

33
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

3 3 2
(x+1) x +3x +3x+1 12x 4
64. y= f (x)= = 2
=x+5+ A. D= { x R| x 1} = (   ,1 )
( 1, ) B. y 
2 2
(x 1) x  2x+1 (x 1)
intercept: f (0)=1 ; x  intercept: f (x)=0 x= 1 C. No symmetry D. lim f (x)= , so x=1 is a VA.
x 1
12 4

2
12x 4 x x /
lim f (x) (x+5) = lim 2
= lim =0 , so the line y=x+5 is a SA. f (x)
x   x   x  2x+1 x   22 1
1 +
x 2
x
2 2 3 2 2
(x 1)  3(x+1)  (x+1)  2(x 1) (x 1)(x+1) [3(x 1) 2(x+1)] (x+1) (x 5) /
= = = so f (x)>0 when x< 1 ,
2 2 4 3
(x 1) (x 1) (x 1)
/
 1<x<1 , or x>5 , and f (x)<0 when 1<x<5 . f is increasing on (   ,1 ) and ( 5, ) and decreasing
on ( 1,5) .
216 27 / /
F. Local minimum value f (5)= = , no local maximum G. f (x)
16 2
3 2 2 2
(x 1) (x 1) +(x 5) 2(x+1)  (x+1) (x 5) 3(x 1)
=
3 2
(x 1)

=
2
(x 1) (x+1){ (x 1)[(x+1)+2(x 5)] 3(x+1)(x 5)}
=
{ (
(x+1) (x 1)[3x 9] 3 x  4x 5
2
)} =
(x+1)(24)
6 4 4
(x 1) (x 1) (x 1)
/ / / /
</ so f (x)>0 if  1<x<1 or x>1 , and f (x)<0 if x< 1 . Thus, f is CU on (  1,1 ) and ( 1, ) and
CD on (   , 1 ) . IP at (  1,0 )
H.

34
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

2 2
 1 / 1 1+x  1 x
65. y= f (x)=x tan x , f (x)=1 = = ,
2 2 2
1+x 1+x 1+x

f
/ /
(x)=
( 1+x ) (2x) x (2x) = 2x ( 1+x  x ) = 2x
2 2 2 2
.
( 1+x2) 2 ( 1+x2) 2 ( 1+x2) 2
   1   
lim f (x) x =lim  tan x =  =0 , so y=x is a SA. Also,
x  2 x  2 2 2 2
   1
lim f (x) x+ = lim   tan x
x   2 x   2
 
=   =0
2 2

 / / /
so y=x+ is also a SA. f (x) 0 for all x , with equality  (x) x=0 , so f is increasing on R . f
2
has the same sign as x , so f is CD on (   ,0 ) and CU on ( 0, ) . f ( x)= f (x) , so f is an odd
function; its graph is symmetric about the origin. f has no local extreme values. Its only IP is at ( 0,0 )
.

2
66. y= f (x)= x +4x = x(x+4) . x(x+4) 0 x  4 or x 0 , so D= (   , 4
0, ) . y  intercept:
2
f (0)=0 ; x  intercepts: f (x)=0 x= 4 , 0 . x +4x  ( x+2 )
2 2 2 2
x +4x  ( x+2 ) x +4x  ( x+2 ) (x +4x) (x +4x+4)  4
=  = = so
1 2 2 2
x +4x  ( x+2 ) x +4x  ( x+2 ) x +4x  ( x+2 )
lim f (x) (x+2) =0 . Thus, the graph of f approaches the slant asymptote y=x+2 as x  and it
x  
/ x+2 /
approaches the slant asymptote y= (x+2) as x   . f (x)= , so f (x)<0 for x< 4 and
2
x +4x
/
f (x)>0 for x>0 ; that is, f is decreasing on (   , 4 ) and increasing on ( 0, ) . There are no local
/ 2  1/2
extreme values. f (x)=(x+2)(x +4x) 
35
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

f
/ / 1
(x)= ( x+2 ) 
2
2  3/2

2
(x +4x)  (2x+4)+(x +4x)
 1/2 2
=(x +4x)
 3/2 2 2
 (x+2) +(x +4x) = 4 x +4x ( 2
)  3/2<0
on D so f is CD on (   , 4 ) and ( 0, ) . No IP

2 2
x y b 2 2
67.  =1 y= x  a . Now
2 2 a
a b
2 2 2
x  a +x  a
lim
x 
b
a
2
x a 
b
a
b
2
x =  lim
a x 
( 2 2
x a  x ) 2 2
b
=  lim
a x  2 2
=0 ,
x  a +x x  a +x
b
which shows that y= x is a slant asymptote. Similarly,
a
2
b 2 2 b b  a b
lim  x a   x =  lim =0 , so y= x is a slant asymptote.
x  a a a x  2 2 a
x  a +x

3 3 3
x +1 2 x +1 x
2 1 1 2
68. f (x) x =  x= = , and lim =0. Therefore, lim f (x) x =0 , and so the
x x x x   x x  
2 2
graph of f is asymptotic to that of y=x . For purposes of differentiation, we will use f (x)=x +1/x.
A. D= { x| x 0} B. No y intercept; to find the x  intercept, we set y=0 x= 1.
3 3
x +1 x +1
C. No symmetry D. lim = and lim =  , so x=0 is a vertical asymptote. Also, the
+ x  x
x 0 x 0
2 / 2 1
graph is asymptotic to the parabola y=x , as shown above. E. f (x)=2x 1/x >0 x> 3
, so f is
2
1 1
increasing on 3
, and decreasing on (   ,0 ) and 0, 3
. F. Local minimum value
2 2
3
1 3 3 / / 3
f = , no local maximum G. f (x)=2+2/x >0 x< 1 or x>0 , so f is CU on
3
2 2
(   , 1 ) and ( 0, ) , and CD on (  1,0 ) . IP at (  1,0 )
H.

36
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

4 4
3 1 x +1 x 3
69. lim f (x) x = lim  = lim
=0 , so the graph of f is asymptotic to that of y=x .
x   x   x x x x  
3 1
A. D= { x| x 0} B. No intercept C. f is symmetric about the origin. D. lim x + =  and
 x
x 0
1 3
lim x+ = , so x=0 is a vertical asymptote, and as shown above, the graph of f is asymptotic
+ x
x 0
3 / 2 2 4 1 1 1
to that of y=x . E. f (x)=3x  1/x >0 x>  x > , so f is increasing on   , and
3 4
3
4
3
1 1 1
4
, and decreasing on  4
,0 and 0, 4
. F. Local maximum value
3 3 3
1  5/4 1  5/4
f  4
= 4 3 , local minimum value f 4
=4 3
3 3
/ / 3
G. f (x)=6x+2/x >0 x>0 , so f is CU on ( 0, ) and CD on (   ,0 ) . No IP
H.

2
70. lim f (x) cos x = lim 1/x =0 , so the graph of f is asymptotic to that of cos x . The
x   x  
1
intercepts can only be found approximately. f (x)= f ( x) , so f is even. lim cos x+ = , so
2
x x 0
x=0 is a vertical asymptote. We don’t need to calculate the derivatives, since we know the asymptotic
behavior of the curve.

37
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.5 Summary of Curve Sketching

38
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

4 3 2 / 3 2
1. f (x)=4x  32x +89x  95x+29  f (x)=16x  96x +178x 95
/ / 2 /
f (x)=48x  192x+178 . f (x)=0 x 0.5 , 1.60 ; f (x)=0 x 0.92 , 2.5 , 2.58 and
/ /
f (x)=0  x 1.46 , 2.54 .

/ /
From the graphs of f , we estimate that f <0 and that f is decreasing on (   ,0.92 ) and ( 2.5,2.58 )
/
, and that f >0 and f is increasing on ( 0.92,2.5) and ( 2.58, ) with local minimum values
/
f (0.92)  5.12 and f (2.58) 3.998 and local maximum value f (2.5)=4 . The graphs of f make it
clear that f has a maximum and a minimum near x=2.5 , shown more clearly in the fourth graph.
/ / / /
From the graph of f , we estimate that f >0 and that f is CU on (   ,1.46 ) and ( 2.54, ) , and
/ /
that f <0 and f is CD on ( 1.46,2.54 ) . There are inflection points at about ( 1.46, 1.40 ) and
( 2.54,3.999 ) .

6 5 4 3 / 5 4 3 2
2. f (x)=x  15x +75x  125x  x f (x)=6x  75x +300x  375x  1
/ / 4 3 2 /
f (x)=30x  300x +900x  750x . f (x)=0 x=0 or x 5.33 ; f (x)=0 x 2.50 , 4.95 , or 5.05;
/ /
f (x)=0 x=0 , 5 or x 1.38 , 3.62 .

1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/
From the graphs of f , we estimate that f is decreasing on (   ,2.50 ) , increasing on ( 2.50,4.95) ,
decreasing on ( 4.95,5.05) , and increasing on ( 5.05, ) , with local minimum values
f (2.50)  246.6 and f (5.05)  5.03 , and local maximum value f (4.95)  4.965 (notice the second
/ /
graph of f ). From the graph of f , we estimate that f is CU on (   ,0 ) , CD on ( 0,1.38 ) , CU on
( 1.38,3.62 ) , CD on ( 3.62,5) , and CU on ( 5, ) . There are inflection points at ( 0,0 ) and ( 5, 5) ,
and at about ( 1.38, 126.38 ) and ( 3.62, 128.62 ) .

2
3 2 / 1 2x 3 / / 2 x  3x+24
3. f (x)= x  3x 5  f (x)=  f (x)=
3 2/3 9
( x  3x 5)
2
( x2 3x 5) 5/3

2
x  3x 5 2 1/3
Note: With some CAS’s, including Maple, it is necessary to define f (x)= 2
x  3x 5 ,
x  3x 5
since the CAS does not compute real cube roots of negative numbers. We estimate from the graph of
/
f that f is increasing on ( 1.5, ) , and decreasing on (   ,1.5) . f has no maximum. Minimum
value: f (1.5)  1.9 .
/ /
From the graph of f , we estimate that f is CU on (  1.2,4.2 ) and CD on (   , 1.2 ) and ( 4.2, ).
IP at (  1.2,0 ) and ( 4.2,0 ) .

4 3 2 5 4 3 2
x +x  2x +2 / x +2x  3x  4x +2x 1
4. f (x)= 2
 f (x)=2 
2
x +x 2 ( x +x 2)
2

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

6 5 4 3 2
/ / x +3x  3x  11x +12x +18x 2
f (x)=2
( x2+x 2) 3

/
We estimate from the graph of f that f is increasing on (  2.4, 2 ) , (  2, 1.5) and ( 1.5, ) and
decreasing on (   , 2.4 ) , (  1.5,1 ) and ( 1,1.5) . Local maximum value: f ( 1.5) 0.7 .
/ /
Local minimum values: f ( 2.4) 7.2 , f (1.5) 3.4 . From the graph of f , we estimate that f is
CU on (  , 2) , (  1.1,0.1 ) and ( 1, ) and CD on (  2, 1.1 ) and ( 0.1,1 ) .
f has IP at (  1.1,0.2 ) and ( 0.1, 1.1 ) .

5. f (x)=
x

/
f (x)=
3 2
 2x +x +1
 f
/ /
(x)=
( 5 4 3
2 3x  3x +5x  6x +3x+4
2
)
3 2
x  x  4x+1 ( x3 x2 4x+1) 2 ( x3 x2 4x+1) 3

We estimate from the graph of f that y=0 is a horizontal asymptote, and that there are vertical
/
asymptotes at x= 1.7 , x=0.24 , and x=2.46 . From the graph of f , we estimate that f is increasing
on (   , 1.7) , (  1.7,0.24 ) , and ( 0.24,1 ) , and that f is decreasing on ( 1,2.46 ) and ( 2.46, ) .
1 / /
There is a local maximum value at f (1)= . From the graph of f , we estimate that f is CU on
3
(   , 1.7) , (  0.506,0.24 ) , and ( 2.46, ) , and that f is CD on (  1.7, 0.506 ) and ( 0.24,2.46 ) .
3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

There is an inflection point at (  0.506, 0.192 ) .

/ 2 / / 2
6. f (x)=tan x+5cos x (x)=2sec x tan x 5cos x . Since f is periodic with
f (x)=sec x 5sin x f

period 2 , and defined for all x except odd multiples of , we graph f and its derivatives on
2
 3
 , .
2 2

/   
We estimate from the graph of f that f is increasing on  ,0.21 , 1.07, , ,2.07
2 2 2
3
, and 2.93, , and decreasing on ( 0.21,1.07) and ( 2.07,2.93) . Local minimum values:
2
f (1.07) 4.23 , f ( 2.93)   5.10 . Local maximum values: f (0.21) 5.10 , f (2.07)  4.23 .
/ /  3
From the graph of f , we estimate that f is CU on 0.76, and 2.38, , and CD on
2 2
 
 ,0.76 and ,2.38 . f has IP at ( 0.76,4.57) and ( 2.38, 4.57) .
2 2
2 / / /
7. f (x)=x  4x+7cos x ,  4 x 4 . f (x)=2x 4 7sin x  f (x)=2 7cos x . f (x)=0 x 1.10 ;
/ / /  1 2
f (x)=0 x  1.49 ,  1.07 , or 2.89 ; f (x)=0 x= cos  1.28
7

/ /
From the graphs of f , we estimate that f is decreasing ( f <0) on (  4, 1.49 ) , increasing on
(  1.49, 1.07) , decreasing on (  1.07,2.89 ) , and increasing on ( 2.89,4 ) , with local minimum values
f ( 1.49) 8.75 and f (2.89)  9.99 and local maximum value f ( 1.07) 8.79 (notice the second
/ / / /
graph of f ). From the graph of f , we estimate that f is CU ( f >0 ) on (  4, 1.28 ) , CD on
(  1.28,1.28 ) , and CU on ( 1.28,4 ) . There are inflection points at about (  1.28,8.77) and
4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

( 1.28, 1.48 ) .
x x 2 x 4 3 2
e / e (x  2x 9) / / e (x  4x  12x +36x+99)
8. f (x)= 2
 f (x)=  f (x)=
2
x 9 x 9 ( 2
) 2
x 9 ( )3
There are vertical asymptotes at x= 3 . It is difficult to show all the pertinent features in one viewing
/ / /
rectangle, so we’ll show f , f , and f for x<3 and also for x>3 .

For x<3

For x>3

/
We estimate from the graphs of f and f that f is increasing on (   , 3) , (  3, 2.16 ) , and
( 4.16, ) and decreasing on (  2.16,3) and ( 3,4.16 ) . There is a local maximum value of
/ /
f ( 2.16)  0.03 and a local minimum value of f (4.16) 7.71 . From the graphs of f , we see that
f is CU on (   , 3) and ( 3, ) and CD on (  3,3) . There is no inflection point.

3 2 / 2 / /
9. f (x)=8x  3x  10 f (x)=24x  6x f (x)=48x 6

3 2
From the graphs, it appears that f (x)=8x  3x  10 increases on (   ,0 ) and ( 0.25, ) and decreases
on ( 0,0.25) ; that f has a local maximum value of f (0)= 10.0 and a local minimum value of
f (0.25)  10.1 ; that f is CU on ( 0.1, ) and CD on (   ,0.1 ) ; and that f has an IP at ( 0.1, 10 ) .
/ 2
To find the exact values, note that f (x)=24x  6x=6x(4x 1) , which is positive ( f is increasing) for
5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

1 1
(   ,0 ) and , , and negative ( f is decreasing) on 0, . By the FDT, f has a local
4 4
1 1 1 3 161
maximum at x=0 : f (0)= 10 ; and f has a local minimum at :f =   10= .
4 4 8 16 16
/ / 1
f (x)=48x 6=6(8x 1) , which is positive ( f is CU) on , and negative ( f is CD) on
8
1 1 1 1 321
  , . f has an IP at ,f = , .
8 8 8 8 32

10.

From the graphs, it appears that f increases on ( 0,3.6 ) and decreases on (   ,0 ) and ( 3.6, ) ; that f
has a local maximum of f (3.6) 2.5 and no local minima; that f is CU on ( 5.5, ) and CD on
2
11 20 x +11x 20
(   ,0 ) and ( 0,5.5) ; and that f has an IP at ( 5.5,2.3) . f (x)=   =1+
2 x 2
x x
/  2  3  3 40
f (x)= 11x +40x = x (11x 40) , which is positive ( f is increasing) on 0, , and negative (
11
40 40
f is decreasing) on (   ,0 ) and on , . By the FDT, f has a local maximum at x= :
11 11
40 2 40
+11  20
40 11 11 1600+11
11
40 20
121 201
f = = = ; and f has no local
11 40 2 1600 80
11
/  2  3 / /  3  4  4
minimum. f (x)= 11x +40x  f (x)=22x  120x =2x (11x 60) , which is positive ( f is CU)
60 60
on , , and negative ( f is CD) on (   ,0 ) and 0, . f has an IP at
11 11
60 60 60 211
,f = , .
11 11 11 90

11. From the graph, it appears that f increases on (  2.1,2.1 ) and decreases on (  3, 2.1 ) and ( 2.1,3) ;
that f has a local maximum of f (2.1) 4.5 and a local minimum of f ( 2.1)  4.5 ; that f is CU on
2
(  3.0,0 ) and CD on ( 0,3.0 ) , and that f has an IP at ( 0,0 ) . f (x)=x 9 x 
6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

2 2
/  x 2 9 2x  3 2 3 2
f (x)= + 9 x = , which is positive ( f is increasing) on , and
2 2 2 2
9 x 9 x
 3 2 3 2
negative ( f is decreasing) on  3, and ,3 . By the FDT, f has a local maximum
2 2
2
3 2 3 2 3 2 9
value of f = 9 = ; and f has a local minimum value of
2 2 2 2
2
 3 2 9 /  x 2 / /
f = (since f is an odd function). f (x)= + 9 x  f (x)
2 2 2
9 x
2
9 x ( 2x)+x
2 1
( 9 x2)  1/2 ( 2x) 2  1/2  2x x
3
( 9 x2)  1 x
=
2
2
(
 x 9 x ) =
2
9 x 9 x

=
 3x

x
3
=
( 2 )
x 2x  27
which is positive ( f is CU) on (  3,0 ) and negative ( f is CD)
2 3/2
9 x
2
9 x ( ) ( 9 x2) 3/2
on ( 0,3) . f has an IP at ( 0,0 ) .

12. From the graph, it appears that f increases on (  5.2, 1.0 ) and ( 1.0,5.2 ) and decreases on
(  2 , 5.2 ) , (  1.0,1.0 ) , and ( 5.2,2 ) ; that f has local maximum values of f ( 1.0) 0.7 and
f (5.2) 7.0 and local minimum values of f ( 5.2)  7.0 and f (1.0)  0.7 ; that f is CU on
(  2 , 3.1 ) and ( 0,3.1 ) and CD on (  3.1,0 ) and ( 3.1,2 ) , and that f has IP at ( 0,0 ) , (  3.1, 3.1 )
/ 1
and ( 3.1,3.1 ) . f (x)=x 2sin x f (x)=1 2cos x , which is positive ( f is increasing) when cos x< ,
2
5   5 5
that is, on 
, and , , and negative ( f is decreasing) on  2 , ,
3 3 3 3 3
  5  
 , , and ,2 . By the FDT, f has local maximum values of f  = + 3
3 3 3 3 3
5 5 5 5
and f = + 3 , and local minimum values of f  =  3 and
3 3 3 3
  / / /
f =  3 . f (x)=1 2cos x f (x)=2sin x , which is positive ( f is CU) on (  2 ,  )
3 3
7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

and ( 0, ) and negative ( f is CD) on (   ,0 ) and ( ,2 ) . f has IP at ( 0,0 ) , (   ,  ) and ( , ) .

2
13. (a) f (x)=x ln x . The domain of f is ( 0, ).

2
2 ln x 1/x x
(b) lim x ln x=lim =lim =lim  =0 . There is a hole at ( 0,0 ) .
2 3 2
x 0
+
x 0
+
1/x x
+
0  2/x x
+
0
2
(c) It appears that there is an IP at about ( 0.2, 0.06 ) and a local minimum at ( 0.6, 0.18 ) . f (x)=x ln x
/ 2 1  1/2
 f (x)=x (1/x)+(ln x)(2x)=x(2ln x+1)>0 ln x>  x>e , so f is increasing on ( 1/ e , ) ,
2
decreasing on ( 0,1/ e ) . By the FDT, f ( 1/ e ) = 1/(2e) is a local minimum value. This point is
approximately ( 0.6065, 0.1839 ) , which agrees with our estimate.
3  3/2  3/2
/ /
f (x)=x(2/x)+(2ln x+1)=2ln x+3>0 ln x>
2
 x>e , so f is CU on e , and CD on ( )
( 0,e 3/2) . IP is ( e 3/2, 3/ ( 2e3) )  ( 0.2231, 0.0747) .
1/x
14. (a) f (x)=xe . The domain of f is (   ,0 )  ( 0, ).

(b) lim xe =lim

e 1/x
=lim
e
1/x 1/x
(  1/x2) =lim e1/x= , so x=0 is a VA.
+ 1/x 2
+
x 0
+
x 0 x 0  1/x x 0
+

Also

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

1/x 1/x
lim xe =0 since 1/x    e  0.

x 0
(c) It appears that there is a local minimum at ( 1,2.7) . There are no IP and f is CD on (   ,0 ) and
CU on ( 0, ) .
1/x / 1/x 1 1/x 1/x 1 1
f (x)=xe  f (x)=xe  +e =e 1 >0 <1 x<0 or x>1 , so f is increasing on
2 x x
x
(   ,0 ) and ( 1, ) , and decreasing on ( 0,1 ) . By the FDT, f (1)=e is a local minimum value, which
agrees with our estimate.
/ /
f (x)=e
1/x
( 2)
1/x +(1 1/x)e
1/x
(  1/x2) = ( e1/x/x2) (1 1+1/x)=e1/x/x3>0  x>0 , so f is CU on ( 0, )
and CD on (   ,0 ) . No IP.

15.

2
(x+4)(x 3)
f (x)= 4
has VA at x=0 and at x=1 since lim f (x)=  , lim f (x)=  and lim f (x)= .
x 0  +
x (x 1) x 1 x 1
2
x+4 (x 3)

x 2
x 3
f (x) = [dividing numerator and denominator by x ]
4
x

(x 1)
3
x
2
(1+4/x)(1 3/x)
=  0 as x  , so f is asymptotic to the x  axis. Since f is undefined at x=0 , it
x(x 1)
2
has no y  intercept. f (x)=0 (x+4)(x 3) =0 x= 4 or x=3 , so f has x  intercepts  4 and 3 . Note,
however, that the graph of f is only tangent to the x axis and does not cross it at x=3 , since f is
 +
positive as x 3 and as x 3 .

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

From these graphs, it appears that f has three maximum values and one minimum value. The
maximum values are approximately f ( 5.6)=0.0182 , f (0.82)= 281.5 and f (5.2)=0.0145 and we
know (since the graph is tangent to the x axis at x=3 ) that the minimum value is f (3)=0 .

16.

4
10x(x 1)
f (x)= has VA at x= 1 and at x=2 since lim f (x)= , lim f (x)=  and lim f (x)= .
3 2 
(x 2) (x+1) x  1 +
x 2 x 2
4
10(1 1/x)
f (x)=  10 as x  , so f is asymptotic to the line y=10 . f (0)=0 , so f has a y 
3 2
(1 2/x) (1+1/x)
4
intercept at 0 . f (x)=0 10x(x 1) =0 x=0 or x=1 . So f has x  intercepts 0 and 1 . Note, however,
that f does not change sign at x=1 , so the graph is tangent to the x axis and does not cross it. We
know (since the graph is tangent to the x axis at x=1 ) that the maximum value is f (1)=0 . From the
graphs it appears that the minimum value is about f (0.2)= 0.1 .

17. f (x)=
2
x (x+1)
3

/
f (x)=
x(x+1)
2
( x3+18x2 44x 16) (from CAS).
2 4 3 5
(x 2) (x 4) (x 2) (x 4)

/
From the graphs of f , it seems that the critical points which indicate extrema occur at x  20 ,  0.3

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/
, and 2.5 , as estimated in Example 3. (There is another critical point at x= 1 , but the sign of f does
not change there.) We differentiate again, obtaining

f
/ /
(x)=2
( 6 5
(x+1) x +36x +6x  628x +684x +672x+64
4 3 2
) .
4 6
(x 2) (x 4)

/ /
From the graphs of f , it appears that f is CU on (  35.3, 5.0 ) , (  1, 0.5) , (  0.1,2 ) , ( 2,4 ) and
( 4, ) and CD on (   , 35.3) , (  5.0, 1 ) and (  0.5, 0.1 ) . We check back on the graphs of f to find
the y coordinates of the inflection points, and find that these points are approximately
(  35.3, 0.015) , (  5.0, 0.005) , (  1,0 ) , (  0.5,0.00001 ) , and (  0.1,0.0000066 ) .
4 3
10x(x 1) / (x 1) (5x 1)
18. f (x)=  f (x)= 20 (from CAS).
3 2 4 3
(x 2) (x+1) (x 2) (x+1)

/
From the graphs of f , we estimate that f is increasing on (   , 1 ) and ( 0.2,1 ) and decreasing on

(  1,0.2 ) , ( 1,2 ) and ( 2, ) . Differentiating f (x) , we get f

/ / /
(x)=60
2
(x 1) ( 5x3 8x2+17x 6) .
5 4
(x 2) (x+1)

11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/ /
From the graphs of f , it seems that f is CU on (   , 1.0 ) , (  1.0,0.4 ) and ( 2.0, ) , and CD on
( 0.4,2 ) . There is an inflection point at about ( 0.4, 0.06 ) .

19. y= f (x)=
sin x
2
with 0 x 3 . From a CAS, y =
/ ( 2
)
sin x 2 x +1 cos x xsin x
and
3/2
2
x +1 ( x +1)
2

y
/ /
=
( 4x +6x +5) cos x 4x ( x +1 ) sin xcos x 2x  2x  3
4 2 2 2 4 2
.
( x2+1) 5/2

/ / /
From the graph of f and the formula for y , we determine that y =0 when x= , 2 , 3 , or
x 1.3 , 4.6 , or 7.8 . So f is increasing on ( 0,1.3) , ( ,4.6 ) , and ( 2 ,7.8 ) . f is decreasing on
( 1.3, ) , ( 4.6,2 ) , and ( 7.8,3 ) . Local maximum values: f (1.3) 0.6 , f (4.6) 0.21 , and
/ / / /
f (7.8) 0.13 . Local minimum values: f ( )= f (2 )=0 . From the graph of f , we see that y =0
 x 0.6 , 2.1 , 3.8 , 5.4 , 7.0 , or 8.6 . So f is CU on ( 0,0.6 ) , ( 2.1,3.8 ) , ( 5.4,7.0 ) , and ( 8.6,3 ) .
f is CD on ( 0.6,2.1 ) , ( 3.8,5.4 ) , and ( 7.0,8.6 ) . There are IP at ( 0.6,0.25) , ( 2.1,0.31 ) , ( 3.8,0.10 ) ,
( 5.4,0.11 ) , ( 7.0,0.061 ) , and ( 8.6,0.065) .
2x 1
20. f (x)= 4

4
x +x+1
3
/ 4x +6x+9
f (x)= 
5/4
( 4
4 x +x+1 )

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

6 4 3 2
/ / 32x +96x +152x  48x +6x+21
f (x)=
(
16 x +x+1
4
) 9/4

/
From the graph of f , f appears to be decreasing on (   , 0.94 ) and increasing on (  0.94, ).
/ /
There is a local minimum value of f ( 0.94)  3.01 . From the graph of f , f appears to be CU on
(  1.25, 0.44 ) and CD on (   , 1.25) and (  0.44, ) . There are inflection points at (  1.25, 2.87)
and (  0.44, 2.14 ) .

21. y= f (x)=
1 e
1/x
. From a CAS, y =
/
1/x
2e
and y
/ /
=
1/x
 2e ( 1 e1/x+2x+2xe1/x) .
1/x 1/x 2 1/x 3
1+e
2
x 1+e ( ) x ( 1+e )
4

f is an odd function defined on (   ,0 )  ( 0,

) . Its graph has no x  or y  intercepts. Since
/
lim f (x)=0 , the x  axis is a HA. f (x)>0 for x 0 , so f is increasing on (   ,0 ) and ( 0, ) . It
x 
/ /
has no local extreme values. f (x)=0 for x 0.417 , so f is CU on (   , 0.417) , CD on
(  0.417,0 ) , CU on ( 0,0.417) , and CD on ( 0.417, ) . f has IPs at (  0.417,0.834 ) and
( 0.417, 0.834 ) .
1
22. y= f (x)= . From a CAS,
tan x
1+e
13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

tan x tan x tan x

/ e / / e e (2sin xcos x 1)+2sin xcos x+1
y = and y = . f is a periodic
tan x 2 tan x 3
cos x 1+e
2
( ) cos x 1+e
4
( )
function with period  that has positive values throughout its domain, which consists of all real
  3 5
numbers except odd multiples of (that is, , , , and so on). f has y  intercept
2 2 2 2
1 / / /  
, but no x  intercepts. We graph f , f , and f on one period,  , .
2 2 2

/
Since f (x)<0 for all x in the domain of f , f is decreasing on the intervals between odd multiples of
 / / 
. f (x)=0 for x=0+n and for x 1.124+n , so f is CD on  , 1.124 , CU on
2 2

(  1.124,0 ) , CD on ( 0,1.124 ) , and CU on 1.124, . Since f is periodic, this behavior repeats
2
1
on every interval of length  . f has IPs at (  1.124+n ,0.890 ) , n , , and ( 1.124+n ,0.110 ) .
2

1/x
23. (a) f (x)=x

b bln a 1/x ( 1/x ) ln x + ln x 1/x (1/x)ln x

(b) Recall that a =e . lim x =lim e . As x 0 ,    , so x =e  0 . This
+ + x
x 0 x 0
0
indicates that there is a hole at ( 0,0 ) . As x  , we have the indeterminate form  .

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

1/x ( 1/x ) ln x ln x 1/x 1/x 0

lim x =lim e , but lim =lim =0 , so lim x =e =1 . This indicates that y=1 is a
x  x  x  x x  1 x 
HA.
(c) Estimated maximum: ( 2.72,1.45) . No estimated minimum. We use logarithmic differentiation to
/
1/x 1 y 1 1 1
find any critical numbers. y=x  ln y= ln x  =
+ ( ln x )  
x y x x 2
x
/ 1/x 1 ln x / / 1/e
y =x =0  ln x=1  x=e . For 0<x<e , y >0 and for x>e , y <0 , so f (e)=e is a
2
x
local maximum value. This point is approximately ( 2.7183,1.4447) , which agrees with our estimate.

(d)

/ / / /
From the graph, we see that f (x)=0 at x 0.58 and x 4.37 . Since f changes sign at these
values, they are x  coordinates of inflection points.

sin x
24. (a) f (x)=(sin x) is continuous where sin x>0 , that is, on intervals of the form ( 2n ,(2n+1) )
, so we have graphed f on ( 0, ) .

sin x
(b) y=(sin x) 
ln y=sin xln sin x , so
ln sin x cot x 0
lim ln y=lim sin xln sin x=lim =lim =lim ( sin x)=0 lim y=e =1 .
+ + + csc x +  csc xcot x + +
x 0 x 0 x 0 x 0 x 0 x 0
(c) It appears that we have a local maximum at ( 1.57,1 ) and local minima at ( 0.38,0.69 ) and
sin x
( 2.76,0.69 ) . y=(sin x)  ln y=sin xln sin x

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/
y cos x / sin x /
=(sin x) +(ln sin x)cos x=cos x(1+ln sin x) y =(sin x) (cos x)(1+ln sin x) . y =0
y sin x
  1  1  1
cos x=0 or ln sin x= 1 x=
2 2
or sin x=e . On ( 0, ) , sin x=e  x =sin
1
( e 1) and
 1
x =  sin
3
( e 1) . Approximating these points gives us
( x ,f ( x ) ) 
1 1
( 0.3767,0.6922 ) , x2,f x2  ( 1.5708,1 ) , and x3,f x3  ( 2.7649,0.6922 ) . The
( ( )) ( ( ))
approximations confirm our estimates.

(d)

/ / / /
From the graph, we see that f (x)=0 at x 0.94 and x 2.20 . Since f changes sign at these
values, they are x  coordinates of inflection points.

25.

From the graph of f (x)=sin (x+sin 3x) in the viewing rectangle 0, by  1.2,1.2 , it looks like f
/
has two maxima and two minima. If we calculate and graph f (x)= cos (x+sin 3x) (1+3cos 3x) on
0,2 ,
/
we see that the graph of f appears to be almost tangent to the x axis at about x=0.7 . The graph of
/ / 2
f = sin (x+sin 3x) (1+3cos 3x) +cos (x+sin 3x)( 9sin 3x) is even more interesting near this x
value: it seems to just touch the x axis.

16
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/ / / /
If we zoom in on this place on the graph of f , we see that f actually does cross the axis twice
/
near x=0.65 , indicating a change in concavity for a very short interval. If we look at the graph of f
on the same interval, we see that it changes sign three times near x=0.65 , indicating that what we had
thought was a broad extremum at about x=0.7 actually consists of three extrema (two maxima and a
minimum). These maximum values are roughly f (0.59)=1 and f (0.68)=1 , and the minimum value is
roughly f (0.64)=0.99996 . There are also a maximum value of about f (1.96)=1 and minimum values
of about f (1.46)=0.49 and f (2.73)= 0.51 . The points of inflection on ( 0, ) are about ( 0.61,0.99998 )
, ( 0.66,0.99998 ) , ( 1.17,0.72 ) , ( 1.75,0.77) , and ( 2.28,0.34 ) . On ( ,2 ) , they are about
( 4.01, 0.34 ) , ( 4.54, 0.77) , ( 5.11, 0.72 ) , ( 5.62, 0.99998 ) , and ( 5.67, 0.99998 ) . There are also
IP at ( 0,0 ) and ( ,0 ) . Note that the function is odd and periodic with period 2 , and it is also
rotationally symmetric about all points of the form ( (2n+1) ,0 ) , n an integer.

3
26. f (x)=x +cx=x x +c  ( 2
) /
f (x)=3x +c
2
f
/ /
(x)=6x

c= 6 c=0 c=6

x  intercepts: When c 0 , 0 is the only x  intercept. When c<0 , the x  intercepts are 0 and  c .
/ /
y  intercept = f (0)=0 . f is odd, so the graph is symmetric with respect to the origin. f (x)<0 for
/ /
x<0 and f (x)>0 for x>0 , so f is CD on (   ,0 ) and CU on ( 0, ) . The origin is the only
inflection point.
/
If c>0 , then f (x)>0 for all x , so f is increasing and has no local maximum or minimum.
/
If c=0 , then f (x) 0 with equality at x=0 , so again f is increasing and has no local maximum or
minimum.
/ 2 /
If c<0 , then f (x)=3 x  ( c / 3) =3 ( x+  c / 3 ) ( x  c / 3 ) , so f ( x ) >0 on (   ,  c / 3 ) and
/ 2
(  c / 3, ) ; f ( x ) <0 on (   c / 3,  c / 3 ) . It follows that f (   c / 3 ) =
3
c  c / 3 is a local
2
maximum value and f
3
(  c / 3 )=
c  c / 3 is a local minimum value. As c decreases (toward more
negative values), the local maximum and minimum move further apart. There is no absolute
maximum or minimum value. The only transitional value of c corresponding to a change in character
of the graph is c=0 .
17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

4
27. f (x)=x +cx =x
2 2
( x2+c ) . Note that f is an even function. For c 0 , the only x  intercept is the
/ 13 / / 2 2
point ( 0,0 ) . We calculate f (x)=4x +2cx=4x c  f (x)=12x +2c . If c 0 , x=0 is the
x+
2
only critical point and there is no inflection point. As we can see from the examples, there is no
change in the basic shape of the graph for c 0 ; it merely becomes steeper as c increases. For c=0 ,
the graph is the simple curve
4
y=x . For c<0 , there are x intercepts at 0 and at  c . Also, there is a maximum at ( 0,0 ) , and
1 1 2
there are minima at  c , c . As c   , the x coordinates of these minima get
2 4
larger in absolute value, and the minimum points move downward. There are inflection points at
1 5 2
 c , c , which also move away from the origin as c   .
6 36

2 2 2
28. We need only consider the function f (x)=x c  x for c 0 , because if c is replaced by  c , the
function is unchanged. For c=0 , the graph consists of the single point ( 0,0 ) . The domain of f is
 c,c , and the graph of f is symmetric about the y axis.
2 2 2
3x x 
/ 2 2
f (x)=2x c  x +x
2  2x 2 2
=2x c  x 
x
3
=
2 2
2x c  x  x
3
=
3
c
(. So we
)
2 2 2 2 2 2 2 2
2 c x c x c x c x
/
see that all members of the family of curves have horizontal tangents at x=0 , since f (0)=0 for all
c>0 .
Also, the tangents to all the curves become very steep as x c , since
18
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

/ / / 2 2 2
lim f (x)= and lim f (x)=  . We set f (x)=0 x=0 or x  c =0 , so the absolute maximum
+  3
x  c x c
2 2 3
values are f c = c .
3 3 3

f
/ /
(x)=
(  9x +2c )
2 2
c x 
2 2
(  3x3+2c2x) (  x/ c x
2 2
) = 6x  9c x +2c
4 2 2 4
.
2 2 2 3/2
c x (c  x )
2

2 2
/ / 2 9c c 33
Using the quadratic formula, we find that f (x)=0 x= . Since  c<x<c , we take
12
2 9
33 2 9 33 ( 9 33 ) ( 33  3) 3
x= c , so the inflection points are c, c .
12 12 144
From these calculations we can see that the maxima and the points of inflection get both horizontally
and vertically further from the origin as c increases. Since all of the functions have two maxima and
two inflection points, we see that the basic shape of the curve does not change as c changes.

29.

c=0 is a transitional value    we get the graph of y=1 . For c>0 , we see that there is a HA at y=1 ,
and that the graph spreads out as c increases. At first glance there appears to be a minimum at ( 0,0 ) ,
but f ( 0 ) is undefined, so there is no minimum or maximum. For c<0 , we still have the HA at y=1 ,

19
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

2
 c/x
but the range is ( 1, ) rather than ( 0,1 ) . We also have a VA at x=0 . f (x)=e 
/
f (x)=e
 c/x
2
(  2c/x3)  f
/ /
(x)=
(
2c 2c 3x
2
) / /
. f (x) 0 and f (x) exists for all x 0 (and 0 is not
2
6 c/x
xe
/ /
in the domain of f ), so there are no maxima or minima. f (x)=0 x= 2c/3 , so if c>0 , the
inflection points spread out as c increases,
 3/2
and if c<0 , there are no IP. For c>0, there are IP at ( 2c/3,e ) . Note that the y  coordinate of
the IP is constant.

30. We see that if c 0 , f (x)=ln x +c is only defined for x > c ( 2

)
x >  c , and
2

lim f (x)= lim f (x)=  , since ln y   as y 0 . Thus, for c<0 , there are vertical
+ 
x  c x   c
asymptotes at x= c , and as c decreases (that is, c increases), the asymptotes get further apart.
For c=0 , lim f (x)=  , so there is a vertical asymptote at x=0 . If c>0 , there are no asymptotes. To
x 0
find the extrema
1
and inflection points, we differentiate: f (x)=ln x +c  ( 2
) /
f (x)= 2
(2x) , so by the First Derivative
x +c
Test there is a local and absolute minimum at x=0 . Differentiating again, we get

f
/ /
(x)=
1
(2)+2x  ( x +c )
2  2
(2x) =
(
2 c x
2
) . Now if c 0 , f
/ /
is always negative, so f is
2 2
x +c ( x +c )
2

/ /
concave down on both of the intervals on which it is defined. If c>0 , then f changes sign when
2
c=x  x= c . So for c>0 there are inflection points at x= c , and as c increases, the inflection
points get further apart.

31. Note that c=0 is a transitional value at which the graph consists of the x axis. Also, we can see
cx
that if we substitute  c for c , the function f (x)= will be reflected in the x axis, so we
2 2
1+c x
investigate only positive
values of c (except c= 1 , as a demonstration of this reflective property). Also, f is an odd function.

20
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

lim f (x)=0 , so y=0 is a horizontal asymptote for all c . We calculate

x 

f
/
(x)=
( 1+c x ) c cx ( 2c x )
2 2 2
=
c ( c x  1)
2 2
/ 2 2
. f (x)=0 c x  1=0 x= 1/c . So there
2 2 2
( 1+c x ) ( 1+c2x2) 2
1 1
is an absolute maximum value of f (1/c)= and an absolute minimum value of f ( 1/c)= . These
2 2
extrema have the same value regardless of c , but the maximum points move closer to the y axis as
c increases.
2 2 2
/ / (  2c x ) ( 1+c x )  (  c x +c )
3 3 2

f (x) =
( 1+c2x2) 4
=
(  2c x ) ( 1+c x ) + ( c x  c ) ( 4c x ) 2c x ( c x  3)
3 2 2 3 2 2
=
3 2 2

2 2 3
( 1+c x ) ( 1+c2x2) 3
/ /
f (x)=0 x=0 or 3 /c , so there are inflection points at ( 0,0 ) and at ( 3 /c, 3 /4 ) .

Again, the y coordinate of the inflection points does not depend on c , but as c increases, both
inflection points approach the y axis.
1
32. Note that f (x)= is an even function, and also that lim f (x)=0 for any value of c ,
2 2
1 x +cx
2
( ) x 

so y=0 is a horizontal asymptote. We calculate the derivatives:

2 1
/
f (x)=
2
 4 1 x x+2cx (=
)
4x x +
2
c 1
, and
2 2
( 1 x2) 2+cx2
( 1 x2) 2+cx2
10x +(9c 18)x + ( 3c  12c+6 ) x +2 c
6 4 2 2
/ /
f (x)=2 .
4 2 3
x +(c 2)x +1

21
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

2 2
/
We first consider the case c>0 . Then the denominator of f is positive, that is, 1 x +cx >0 for all ( ) 2

1
x , so f has domain R and also f >0 . If c 1 0 ; that is, c 2 , then the only critical point is f (0)=1
2
, a maximum. Graphing a few examples for c 2 shows that there are two IP which approach the y
axis as c  . c=2 and c=0 are transitional values of c at which the shape of the curve changes. For
1 1
0<c<2 , there are three critical points: f (0)=1 , a minimum value, and f 1 c = ,
2 c(1 c/4)
both maximum values. As c decreases from 2 to 0 , the maximum values get larger and larger, and the
x values at which they occur go from 0 to 1 . Graphs show that there are four inflection points for
+
0<c<2 , and that they get farther away from the origin, both vertically and horizontally, as c 0 . For
c=0 , the function is simply asymptotic to the x axis and to the lines x= 1 , approaching + from

. .
both sides of each. The y  intercept is 1 , and ( 0,1 ) is a local minimum. There are no inflection
points. Now if c<0 , we can write
1 1 1
f (x)= = = 2 . So f has vertical
( 2 2
1 x +cx )
2 2 2
(
1 x  (  c x )
2
) 2
(
x   c x 1 x +  c x 1 )( )
2
/
asymptotes where x  c x 1=0 x= (   c 4 c ) 2 or x= (  c 4 c ) 2 . As c decreases, the /
two exterior asymptotes move away from the origin, while the two interior ones move toward it. We
graph a few examples to see the behavior of the graph near the asymptotes, and the nature of the
1
critical points x=0 and x= 1 c :
2

c= 1 c= 1 c= 1

We see that there is one local minimum value, f (0)=1 , and there are two local maximum values,

22
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

1 1
f c = 1 as before. As c decreases, the x values at which these maxima occur
2 c(1 c/4)
get larger, and the maximum values themselves approach 0 , though they are always negative.

/ / /
33. f (x)=cx+sin x f (x)=c+cos x f (x)= sin x
f ( x)= f (x) , so f is an odd function and its graph is symmetric with respect to the origin.
f (x)=0 sin x= cx , so 0 is always an x  intercept.
/
f (x)=0 cos x= c , so there is no critical number when c >1 . If c  1 , then there are infinitely
many critical numbers. If x is the unique solution of cos x= c in the interval 0, , then the critical
1
numbers are 2n x , where n ranges over the integers. (Special cases: When c=1 , x =0 ; when c=0
1 1

, x= ; and when c= 1 , x = .)
2 1
/ /
f (x)<0 sin x>0 , so f is CD on intervals of the form ( 2n ,(2n+1) ) . f is CU on intervals of the
form ( (2n 1) ,2n ) . The inflection points of f are the points ( 2n ,2n c ) , where n is an integer.
/ /
If c 1 , then f (x) 0 for all x , so f is increasing and has no extremum. If c  1 , then f (x) 0
/
for all x , so f is decreasing and has no extremum. If c <1 , then f (x)>0 cos x> c x is in an
interval of the form 2n  x ,2n +x for some integer n . These are the intervals on which f is
( )
1 1
increasing. Similarly, we
find that f is decreasing on the intervals of the form 2n +x ,2(n+1)  x ( ) . Thus, f has local
1 1
2
maxima at the points 2n +x , where f has the values c(2n +x )+sin x =c(2n +x )+ 1 c , and f
1 1 1 1
has local minima at the points 2n  x , where we have
1
2
f (2n  x )=c(2n  x ) sin x =c(2n  x ) 1 c .The transitional values of c are  1 and 1 . The
1 1 1 1
inflection points move vertically, but not horizontally, when c changes. When c  1 , there is no
extremum. For c <1 , the maxima are spaced 2 apart horizontally, as are the minima. The
horizontal spacing between maxima and adjacent minima is regular (and equals  ) when c=0 , but
the horizontal space between a local maximum and the nearest local minimum shrinks as c
approaches 1 .

23
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

 at  bt
34. For f (t)=C e  e ( )
, C affects only vertical stretching, so we let C=1 . From the first figure,
we notice that the graphs all pass through the origin, approach the t  axis as t increases, and approach
  as t   . Next we let a=2 and produce the second figure.

Here, as b increases, the slope of the tangent at the origin increases and the local maximum value
 2t  bt /  bt  2t /
increases. f (t)=e  e  f (t)=be  2e . f (0)=b 2 , which increases as b increases.
/  bt  2t b (b 2)t b ln b ln 2
f (t)=0 be =2e  =e  ln =(b 2)t  t=t = , which decreases as b increases
2 2 1 b 2
2/(b 2)
(b 2)2
(the maximum is getting closer to the y  axis). f (t )= . We can show that this value
1 1+2/(b 2)
b
increases as b increases by considering it to be a function of b and graphing its derivative with respect
to b , which is always positive.

x 1
35. If c<0 , then lim f (x)= lim = lim =0 , and lim f (x)= .
cx cx
x   x   e x   ce x 
1
If c>0 , then lim f (x)=  , and lim f (x)=lim =0 .
cx
x   x  x
ce 
If c=0 , then f (x)=x , so lim f (x)=  respectively.
x 
So we see that c=0 is a transitional value. We now exclude the case c=0 , since we know how the
 cx
function behaves in that case. To find the maxima and minima of f , we differentiate: f (x)=xe 
 cx  cx  cx
/
f (x)=x  ce ( ) +e = ( 1 cx ) e . This is 0 when 1 cx=0 x=1/c . If c<0 then this
/
represents a minimum value of f ( 1/c ) =1/ ( ce ) , since f (x) changes from negative to positive at
x=1/c ;
24
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

and if c>0 , it represents a maximum value. As c increases, the maximum or minimum point gets
/  cx
closer to the origin. To find the inflection points, we differentiate again: f (x)=e (1 cx) 
 cx  cx  cx
/ /
f (x)=e ( c)+(1 cx)  ce ( )
=(cx 2)ce . This changes sign when cx 2=0  x=2/c . So as c
increases, the points of inflection get closer to the origin.

36. For c=0 , there is no inflection point; the curve is CU everywhere. If c increases, the curve simply
becomes steeper, and there are still no inflection points. If c starts at 0 and decreases, a slight upward
bulge appears near x=0 , so that there are two inflection points for any c<0 . This can be seen
4 2 / 3
algebraically by calculating the second derivative: f (x)=x +cx +x f (x)=4x +2cx+1
/ / 2 / /
f (x)=12x +2c . Thus, f (x)>0 when c>0 . For c<0 , there are inflection points when
1
x=  c . For c=0 , the graph has one critical number, at the absolute minimum somewhere
6
around x= 0.6 . As c increases, the number of critical points does not change. If c instead decreases
from 0 , we see that the graph eventually sprouts another local minimum, to the right of the origin,
somewhere between x=1 and x=2 . Consequently, there is also a maximum near x=0 .
After a bit of experimentation, we find that at c= 1.5 , there appear to be two critical numbers: the
absolute minimum at about x= 1 , and a horizontal tangent with no extremum at about x=0.5 . For
any c smaller than this there will be 3 critical points, as shown in the graphs with c= 3 and with c= 5
/ 3
. To prove this algebraically, we calculate f (x)=4x +2cx+1 . Now if we substitute our value of
/ 3 2 1
c= 1.5 , the formula for f (x) becomes 4x  3x+1=(x+1)(2x 1) . This has a double root at x= ,
2
1
indicating that the function has two critical points: x= 1 and x= , just as we had guessed from the
2
graph.

25
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

4 2 2
37. (a) f (x)=cx  2x +1 . For c=0 , f (x)= 2x +1 , a parabola whose vertex, ( 0,1 ) , is the absolute
4 2
maximum. For c>0 , f (x)=cx  2x +1 opens upward with two minimum points. As c 0 , the
minimum points spread apart and move downward; they are below the x  axis for 0<c<1 and above
for c>1 . For c<0 , the graph opens downward, and has an absolute maximum at x=0 and no local
minimum.
/ 3
( 2
)
(b) f (x)=4cx  4x=4cx x  1/c ( c 0 ). If c 0 , 0 is the only critical number. f
/ / 2
(x)=12cx  4 ,
/ / 2
so f (0)= 4 and there is a local maximum at ( 0,f(0) ) = ( 0,1 ) , which lies on y=1 x . If c>0 , the
critical numbers are 0 and 1/ c . As before, there is a local maximum at ( 0,f(0) ) = ( 0,1 ) , which lies
2 / /
on y=1 x . f ( 1/ c ) =12 4=8>0 , so there is a local minimum at x= 1/ c . Here
( 2)
f ( 1/ c ) =c 1/c  2/c+1= 1/c+1 . But ( 1/ c , 1/c+1 ) lies on y=1 x since 1 ( 1/ c ) =1 1/c .
2 2

2
 c c  12
38. (a) f (x)=2x +cx +2x
3 2 / 2
( 2
f (x)=6x +2cx+2=2 3x +cx+1 . f (x)=0 ) /
x=
6
. So f has
2 / /
critical points  c  12 0 c  2 3 . For c= 2 3 , f (x) 0 on (   , ) , so f does not
2 /
change signs at  c/6 , and there is no extremum. If c  12>0 , then f changes from positive to
2 2
 c c  12  c+ c  12
negative at x= and from negative to positive at x= . So f has a local
6 6
maximum at

26
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.6 Graphing with Calculus and Calculators

2 2
 c c  12  c+ c  12
x= and a local minimum at x= .
6 6
2
/ 2
 1 3x
0
(b) Let x be a critical number for f (x) . Then f (x )=0 3x +cx +1=0 c= . Now
0 0 0 0 x
0
2
 1 3x 3 3 3
f (x ) = 2x3+cx2+2x =2x3+x2 0
+2x = 2x0 x0 3x0+2x0=x0 x0
0 0 0 0 0 0 0
x
0
3
So the point is x ,y = x ,x  x
( ) ; that is, the point lies
0 0 0 0 0
3
on the curve y=x x .

27
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

1. (a)
First Number Second Number Product
1 22 22
2 21 42
3 20 60
4 19 76
5 18 90
6 17 102
7 16 112
8 15 120
9 14 126
10 13 130
11 12 132

We needn’t consider pairs where the first number is larger than the second, since we can just
interchange the numbers in such cases. The answer appears to be 11 and 12 , but we have considered
only integers in the table.

(b) Call the two numbers x and y . Then x+y=23 , so y=23 x . Call the product P . Then
2 2 /
P=xy=x(23 x)=23x x , so we wish to maximize the function P(x)=23x x . Since P (x)=23 2x , we
/ 23 2
see that P (x)=0 x= =11.5 . Thus, the maximum value of P is P(11.5)= ( 11.5) =132.25 and it
2
occurs when x=y=11.5 .
/ /
Or: Note that P (x)= 2<0 for all x , so P is everywhere concave downward and the local maximum
at x=11.5 must be an absolute maximum.

2 /
2. The two numbers are x+100 and x . Minimize f (x)=(x+100)x=x +100x . f (x)=2x+100=0 x= 50
/ /
. Since f (x)=2>0 , there is an absolute minimum at x= 50 . The two numbers are 50 and  50 .

2
100 100 / 100 x  100
3. The two numbers are x and , where x>0 . Minimize f (x)=x+ . f (x)=1 = .
x x 2 2
x x
/ /
The critical number is x=10 . Since f (x)<0 for 0<x<10 and f (x)>0 for x>10 , there is an absolute
minimum at x=10 . The numbers are 10 and 10 .

4. Let x>0 and let f (x)=x+1/x . We wish to minimize f (x) . Now

/
f (x)=1
1 1
2
=
2
2 1
( )
x  1 = (x+1)(x 1) , so the only critical number in ( 0,
2
) is 1 .
x x x
1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

/ /
f (x)<0 for 0<x<1 and f (x)>0 for x>1 , so f has an absolute minimum at x=1 , and f (1)=2 .
/ / 3
Or: f (x)=2/x >0 for all x>0 , so f is concave upward everywhere and the critical point ( 1,2 ) must
correspond to a local minimum for f .

5. If the rectangle has dimensions x and y , then its perimeter is 2x+2y=100 m, so y=50 x . Thus, the
2
area is A=xy=x(50 x) . We wish to maximize the function A(x)=x(50 x)=50x x , where 0<x<50 .
/ / /
Since A (x)=50 2x= 2(x 25) , A (x)>0 for 0<x<25 and A (x)<0 for 25<x<50 . Thus, A has an
2 2
absolute maximum at x=25 , and A(25)=25 =625 m . The dimensions of the rectangle that maximize
its area are x=y=25 m. (The rectangle is a square.)

2
6. If the rectangle has dimensions x and y , then its area is xy=1000 m , so y=1000/x . The perimeter
P=2x+2y=2x+2000/x . We wish to minimize the function P(x)=2x+2000/x for x>0 .
/
P (x)=2 2000/x = 2/x
2
( 2) ( x2 1000) , so the only critical number in the domain of P is x= 1000 .
/ / 3
P (x)=4000/x >0 , so P is concave upward throughout its domain and P ( 1000 ) =4 1000 is an
absolute minimum value. The dimensions of the rectangle with minimal perimeter are
x=y= 1000 =10 10 m.
(The rectangle is a square.)

7. (a)

2
The areas of the three figures are 12 , 500 , 12 , 500 , and 9000 ft . There appears to be a maximum
2
area of at least 12 , 500 ft .
(b) Let x denote the length of each of two sides and three dividers.
Let y denote the length of the other two sides.

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

(c) Area A= length  width =y x

(d) Length of fencing =750 5x+2y=750
5 5 5 2
(e) 5x+2y=750 y=375 x A(x)= 375 x x=375x x
2 2 2
/ / /
(f) A (x)=375 5x=0 x=75 . Since A (x)= 5<0 there is an absolute maximum when x=75 . Then
375 375 2
y= =187.5 . The largest area is 75 =14 , 062.5 ft . These values of x and y are between
2 2
the values in the first and second figures in part (a). Our original estimate was low.

8. (a)

3
The volumes of the resulting boxes are 1 , 1.6875 , and 2 ft . There appears to be a maximum
3
volume of at least 2 ft .

(b) Let x denote the length of the side of the square being cut out. Let y denote the length of the
base.

2
(c) Volume V = length  width  height  V =y y x=xy
3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

(d) Length of cardboard =3 x+y+x=3 y+2x=3

(e) 2
y+2x=3 y=3 2x V ( x ) =x(3 2x)
(f) 2
V (x)=x(3 2x) 
/ 2
V (x)=x 2(3 2x)( 2)+(3 2x)  1=(3 2x)[ 4x+(3 2x)]=(3 2x)( 6x+3) ,
3 1 3 3
so the critical numbers are x= and x= . Now 0 x and V (0)=V =0 , so the
2 2 2 2
1 1 2 3
maximum is V = ( 2 ) =2 ft , which is the value found from our third figure in
2 2
part (a).
9.

6 6
xy=1.5 10 , so y=1.5 10 /x . Minimize the amount of fencing, which is
( 6
) 6
3x+2y=3x+2 1.5 10 /x =3x+3 10 /x=F(x) . F (x)=3 3 10 /x =3 x  10 /x . The critical
/ 6 2
( 2 6
) 2

3 / 3 / 3
number is x=10 and F (x)<0 for 0<x<10 and F (x)>0 if x>10 , so the absolute minimum occurs
3 3
when x=10 and y=1.5 10 . The field should be 1000 feet by 1500 feet with the middle fence
parallel to the short side of the field.

2
10. Let b be the length of the base of the box and h the height. The volume is 32 , 000=b h  h=32 ,
2 2 2 2 2
000/b . The surface area of the open box is S=b +4hb=b +4(32 , 000/b )b=b +4 ( 32,000 ) /b . So
/ 2
S (b)=2b 4(32 , 000)/b =2 b  64,000 /b =0 ( 3
) 2 3
b= 64,000 =40 . This gives an absolute minimum
/ /
since S (b)<0 if 0<b<40 and S (b)>0 if b>40 . The box should be 40 40 20 .

2
11. Let b be the length of the base of the box and h the height. The surface area is 1200=b +4hb
3 2
( 2
) 2 2 2 3
h= 1200 b /(4b) . The volume is V =b h=b 1200 b /4b=300b b /4 V (b)=300
/
4
(b . )
/ 3 2 2 / /
V (b)=0 300= b  b =400  b= 400 =20 . Since V (b)>0 for 0<b<20 and V (b)<0 for b>20 ,
4
there is an absolute maximum when b=20 by the First Derivative Test for Absolute Extreme Values
(see page 280 ). If b=20 , then h= 1200 20 ( 2
) / ( 4 20 ) =10 , so the largest possible volume is
2 2 3
b h=(20) (10)=4000 cm .

4
 .
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

12.

V =lwh 
2 2
10=(2w)(w)h=2w h , so h=5/w . The cost is 10 2w +6 2(2wh)+2(hw) =20w +36wh , so ( 2) 2

2 2 2
( )/
C(w)=20w +36w 5/w =20w +180/w . C (w)=40w 180/w =40
2
w
3 9
2
2
w  w=
3 9
2
is

3 9 /
/
the critical number. There is an absolute minimum for C when w= since C (w)<0 for
2
2
3 9 / 3 9 3 9 3 9 180
0<w< and C (w)>0 for w> .C =20 +
$163.54 .
2 2 2 2 3
9/2

.
13.

2 2
10=(2w)(w)h=2w h , so h=5/w . The cost is C(w)=10 2w +6 2(2wh)+2hw +6 2w = ( 2) ( 2)
45
2 2 / 2 3
32w +36wh=32w +180/wC (w)=64w 180/w =4 16w  45 w  w=
2 3
16
(
is the critical )/
/ 3 45 / 3 45
number. C (w)<0 for 0<w< and C (w)>0 for w> . The minimum cost is
16 16
3 45 2/3
C
16
=32(2.8125) +180 / 2.8125
$191.28 .

14. (a) Let the rectangle have sides x and y and area A , so A=xy or y=A/x . The problem is to
/
minimize the perimeter =2x+2y=2x+2A/x=P(x) . Now P (x)=2 2A/x =2 x  A /x . So the critical
2
( 2
) 2

/ /
number is x= A . Since P (x)<0 for 0<x< A and P (x)>0 for x> A , there is an absolute minimum
at x= A . The sides of the rectangle are A and A/ A= A , so the rectangle is a square.
1
(b) Let p be the perimeter and x and y the lengths of the sides, so p=2x+2y 2y= p 2x y= p x .
2
1 1 2 / 1 1 1
The area is A(x)=x p x = px x . Now A (x)=0 p 2x=0 2x= p x= p . Since
2 2 2 2 4
/ / 1
A (x)= 2<0 , there is an absolute maximum for A when x= p by the Second Derivative Test. The
4
sides of the rectangle are
5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

1 1 1 1
p and p p= p , so the rectangle is a square.
4 2 4 4

2 2 2 2
15. The distance from a point ( x,y ) on the line y=4x+7 to the origin is (x 0) +(y 0) = x +y .
However, it is easier to work with the square of the distance; that is,
2
D(x)= ( x +y ) =x +y =x + ( 4x+7)
2 2 2 2 2 2
. Because the distance is positive, its minimum value will
occur at the same point as the minimum value of D .
/ / 28
D (x)=2x+2(4x+7)(4)=34x+56 , so D (x)=0 x= .
17
/ / 28
D (x)=34>0 , so D is concave upward for all x . Thus, D has an absolute minimum at x= . The
17
28 28 28 7
point closest to the origin is ( x,y ) =  ,4  +7 =  , .
17 17 17 17

16. The square of the distance from a point ( x,y ) on the line y= 6x+9 to the point (  3,1 ) is
2 2 2 2 2 / / 45
D(x)=(x+3) +(y 1) =(x+3) +( 6x+8) =37x  90x+73 . D (x)=74x 90 , so D (x)=0 x= .
37
/ / 45
D (x)=74>0 , so D is concave upward for all x . Thus, D has an absolute minimum at x= . The
37
45 63
point on the line closest to (  3,1 ) is , .
37 37

17.

From the figure, we see that there are two points that are farthest away from A(1,0) . The distance d
2 2
from A to an arbitrary point P(x,y) on the ellipse is d= (x 1) +(y 0) and the square of the distance
1
2 2 2 2 2 2
( /
)
is S=d =x  2x+1+y =x  2x+1+ 4 4x = 3x  2x+5 . S = 6x 2 and S =0 x=
/
3
. Now
/ / 1 1 16
S = 6<0 , so we know that S has a maximum at x= . Since  1 x 1 , S( 1)=4 , S  =
3 3 3
16
, and S(1)=0 , we see that the maximum distance is . The corresponding y  values are
3

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

1 2 32 4 1 4
y= 4 4  = = 2
 1.89 . The points are  , 2 .
3 9 3 3 3

18.

2 2
The distance d from ( 1,1 ) to an arbitrary point P(x,y) on the curve y=tan x is d= (x 1) +(y 1) and
2 2 2 / 2 /
the square of the distance is S=d =(x 1) +(tan x 1) . S =2(x 1)+2(tan x 1)sec x . Graphing S on
 
 , gives us a zero at x
0.82 , and so tan x
1.08 . The point on y=tan x that is closest to
2 2
( 1,1 ) is approximately ( 0.82,1.08 ) .
19.

2 2 2 2 2
The area of the rectangle is (2x)(2y)=4xy . Also r =x +y so y= r  x , so the area is
2 2 2
2 2 / 2 2 x r  2x
A(x)=4x r  x . Now A (x)=4 r  x  =4 . The critical number is
2 2 2 2
r  x r  x
1 2 1 2 1 2 1
x= r . Clearly this gives a maximum. y= r  r = r = r=x , which tells us
2 2 2 2
that the rectangle is a square. The dimensions are 2x= 2 r and 2y= 2 r .
20.

The area of the rectangle is (2x)(2y)=4xy . Now

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2
x y b 2 2 b 2 2 /
+ =1 gives y= a  x , so we maximize A(x)=4 x a  x A (x)=
2 2 a a
a b
4b
a
x
1 2 2
2
a x ( )  1/2 ( 2x)+ ( a2 x2) 1/2 1 =
4b 2 2
a
(
a x )  1/2 2 2 2
 x +a  x =
4b 2
a  2x
2
So
2 2
a a x
1 1
the critical number is x= a , and this clearly gives a maximum. Then y= b , so the maximum
2 2
1 1
area is 4 a b =2ab .
2 2

21.

3 2 2 2
The height h of the equilateral triangle with sides of length L is L , since h +(L/2) =L 
2
3 3
L y L
2 2 1 2 3 2 3 2 2 3
h =L  L= L h= L . Using similar triangles, = = 3 3 x= L y
4 4 2 x L/2 2
3 3
y= L 3 x y= (L 2x) . The area of the inscribed rectangle is
2 2
2 /
A(x)=(2x)y= 3 x(L 2x)= 3 Lx 2 3 x , where 0 x L/2 . Now 0=A (x)= 3 L 4 3 x
/
x= 3 L ( 4 3 ) =L/4 . Since A(0)=A(L/2)=0 , the maximum occurs when x=L/4 , and
3 3 3 3
y= L L= L , so the dimensions are L/2 and L.
2 4 4 4

22.

( 2
) 3
The rectangle has area A(x)=2xy=2x 8 x =16x 2x , where 0 x 2 2 . Now A (x)=16 6x =0
/ 2

8
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2 16
x=2 . Since A ( 0 ) =A ( 2 2 ) =0 , there is a maximum when x=2 . Then y= , so the
3 3 3
2 16
rectangle has dimensions 4 and .
3 3

23.

1 2 2
The area of the triangle is A(x)= (2t)(r+x)=t(r+x)= r  x (r+x) . 0=
2
2 2
/  2x 2 2  2x x +rx 2 2 x +rx 2 2 2 2 2
A (x)=r + r  x +x = + r  x  = r  x  x +rx=r  x
2 2 2 2 2 2 2 2
2 r  x 2 r  x r  x r  x
2 2 1
 0=2x +rx r =(2x r)(x+r) x= r or x= r . Now A(r)=0=A( r) the maximum occurs where
2
1 1 3 2 1 2 3 2
x= r , so the triangle has height r+ r= r and base 2 r  r =2 r = 3r .
2 2 2 2 4

24.

3 y 3 3
The rectangle has area xy . By similar triangles =   4y+12=3x or y= x+3 . So the area is
x 4 4
3 3 2 / 3 3
A(x)=x 
x+3 = x +3x where 0 x 4 . Now 0=A (x)= x+3 x=2 and y= . Since
4 4 2 2
3 2
A(0)=A(4)=0 , the maximum area is A(2)=2 =3 cm .
2

25.

9
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2 2 2 2 2 2
The cylinder has volume V = y (2x) . Also x +y =r  y =r  x , so
( 2 2) (2 3
)
V (x)= r  x (2x)=2 r x x , where 0 x r . V (x)=2
/
(r 2 3x2) =0 x=r / 3 . Now
V (0)=V (r)=0 , so there is a maximum when x=r / 3 and
V ( r / 3 ) = (r 2 r 2/3) ( 2r / 3 ) =4 r
3
/ (3 3 ) .

26.

By similar triangles, y/x=h/r , so y=hx/r . The volume of the cylinder is

2 2 3 / 2 /
 x ( h y ) = hx  ( h/r ) x =V (x) . Now V (x)=2 hx ( 3 h/r ) x = hx ( 2 3x/r ) . So V (x)=0 x=0 or
2 2
x= r . The maximum clearly occurs when x= r and then the volume is
3 3
2 3 2 2 2 2 4 2
 hx  ( h/r ) x = hx ( 1 x/r ) = r h 1 =  r h.
3 3 27

27.

2
The cylinder has surface area 2( area of the base )+( lateral surface area )=2 ( radius ) +2 ( radius )(
2 2 2 2 2 2 2 2 2
height )=2 y +2 y(2x) . Now x +y =r  y =r  x  y= r  x , so the surface area is S(x)=
2 (r 2 x2) +4 2 2 2
x r  x , 0 x r =2 r  2 x +4 x r  x
2
( 2 2
) Thus, S (x) /

1 2 2  1/2 2 2 1/2
=0 4 x+4 x
2
(
r  x )
(  2x ) + r  x (
 1 )

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2 2 2 2 2
x 2 2  x r  x  x +r  x
=4  x + r  x =4 
2 2 2 2
r  x r  x
2
S (x)=0
/ 2 2 2
x r  x =r  2x ( * ) 
2
(x 2
r  x
2
) = (r  2x )  x (r  x ) =r  4r x +4x 
2 2 2 2 2 2 4 2 2 4

2 2 4 4 2 2 4 4 2 2 4 2
r x  x =r  4r x +4x  5x  5r x +r =0 . This is a quadratic equation in x . By the quadratic
2 5 5 2
formula, x = r , but we reject the root with the + sign since it doesn’t satisfy ( * ). So
10
5 5
x= r . Since S(0)=S(r)=0 , the maximum surface area occurs at the critical number and
10
2 5 5 2 2 2 5 5 2 5+ 5 2
x= r  y =r  r = r  the surface area is
10 10 10
5+ 5 2 5 5 5+ 5 2 2 5+ 5 ( 5 5 ) ( 5+ 5 )
2 r +4 r = r 2 +4
10 10 10 10 10
2 5+ 5 2 20 2 5+ 5+2 2 5 2 5+5 5 2
= r + = r = r = r ( 1+ 5 ) .
5 5 5 5

28.

x 1  x x  x
Perimeter=30 2y+x+ =30 y= 30 x =15  . The area is the area of the
2 2 2 2 4
1 2 x
rectangle plus the area of the semicircle, or xy+  , so
2 2
x  x 1 2 1 2  2 /  15 60
A(x)=x 15  +  x =15x x x . A (x)=15 1+ x=0 x= = .
2 4 8 2 8 4 1+ /4 4+
/ /  60
A (x)= 1+ <0 , so this gives a maximum. The dimensions are x= ft and
4 4+
30 15 60+15  30 15 30
y=15  = = ft, so the height of the rectangle is half the base.
4+ 4+ 4+ 4+

29.

11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

xy=384 y=384/x . Total area is A(x)=(8+x)(12+384/x)=12(40+x+256/x) , so

/
(
A (x)=12 1 256/x =0
2
) /
x=16 . There is an absolute minimum when x=16 since A (x)<0 for
/
0<x<16 and A (x)>0 for x>16 . When x=16 , y=384/16=24 , so the dimensions are 24 cm and 36 cm.

30.

xy=180 , so y=180/x . The printed area is (x 2)(y 3)=(x 2)(180/x 3)=186 3x 360/x=A(x) .
/ 2 2 /
A (x)= 3+360/x =0 when x =120 x=2 30 . This gives an absolute maximum since A (x)>0 for
/
0<x<2 30 and A (x)<0 for x>2 30 . When x=2 30 , y=180/(2 30 ) , so the dimensions are
2 30 in. and 90/ 30 in.

31.

Let x be the length of the wire used for the square. The total area is
x 2 1 10 x 3 10 x 1 2 3 2
A(x)= + = x+ (10 x) , 0 x 10
4 2 3 2 3 16 36
/ 1 3 9 4 3 40 3 40 3 3
A (x)= x (10 x)=0 x+ x =0 x= . Now A(0)= 100
4.81
8 18 72 72 72 9+4 3 36
100 40 3
, A(10)= =6.25 and A
2.72 , so
16 9+4 3
(a) The maximum area occurs when x=10 m, and all the wire is used for the square.
40 3
(b) The minimum area occurs when x=
4.35 m.
9+4 3

32.

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2
x 2 10 x 2 x (10 x)
Total area is A(x)= + = + , 0 x 10 .
4 2 16 4
/ x 10 x 1 1 5
A (x)=  = + x =0 x=40/(4+ ) . A(0)=25/
7.96 , A(10)=6.25 , and
8 2 2 8 
A(40/(4+ ))
3.5 , so the maximum occurs when x=0 m and the minimum occurs when x=40/(4+ )
m.

33.

2 2 2 V 2 2V
The volume is V = r h and the surface area is S(r)= r +2 rh= r +2 r = r + .
2 r
 r
/ 2V 3 3 V
S (r)=2 r =0 2 r =2V  r= cm.
2 
r

/ 3 V / 3 V
This gives an absolute minimum since S (r)<0 for 0<r< and S (r)>0 for r> . When
 
3 V V V 3 V
r= , h= = = cm.
 2 2/3 
 r  (V / )

34.

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

 dL 3
L=8 +4sec  , 0< < , = 8 cot  +4sec  tan  =0 when sec  tan  =2 cot   tan  =2
2 d
3  13
tan  = 2   =tan 2 .
 13  13 
dL/d <0 when 0< <tan 2 , dL/d >0 when tan 2 < < , so L has an absolute minimum
2
2/3
 13 1+2 2/3
when  =tan 2 , and the shortest ladder has length L=8 +4 1+2
16.65 ft.
1/3
2
x 8 2 2 2
Another method: Minimize L =x + ( 4+y ) , where = .
4+y y

35.

 
2
h +r =R 
2 2
V=
3
r h=
2
3
( R2 h2) h=  3 ( R2h h3) . V / ( h) =  3 ( R2 3h2) =0 when h= 1
3
R . This
/ 1 / 1
gives an absolute maximum, since V (h)>0 for 0<h< R and V (h)<0 for h> R . The
3 3
1  1 3 1 3 2 3
maximum volume is V R = R R =  R .
3 3 3 3 3 9 3

1 2
36. The volume and surface area of a cone with radius r and height h are given by V =  r h and
3
2 2 2 1 2 2 81
S= r r +h . We’ll minimize A=S subject to V =27 . V =27  r h=27 r = (1) .
3  h
2 2 2
2 2 2 2 2 81 81 2 81 /  2 81 2 81
A= r (r +h )= +h = +81 h , so A =0 +81 =0 81 = 
 h  h 2 3 3
h h h
3 162 3 162 3 6 2 81 81 27
h=  h= =3
3.722 . From (1) , r = = = 
    h   3 6/
3 3 2
6
3 3 / / 2 4
r=
2.632 . A =6 81 /h >0 , so A and hence S has an absolute minimum at these values of
6 2
6
r and h .

37.
14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

H H h 1 2
By similar triangles, = (1) . The volume of the inner cone is V =  r h , so we’ll solve (1)
R r 3
Hr Hr HR Hr H
for h . =H h h=H = = (R r) (2) .
R R R R
 2 H  H 2 3 /  H 2  H
Thus, V (r)= r  (R r)= (Rr  r ) V (r)= (2Rr 3r )= r(2R 3r) .
3 R 3R 3R 3R
/ 2 H 2 H 1 1
V (r)=0 r=0 or 2R=3r r= R and from (2) , h= R R = R = H.
3 R 3 R 3 3
/ 2
V (r) changes from positive to negative at r= R , so the inner cone has a maximum volume of
3
1 2 1 2 2 1 4 1 2
V =  r h=  R H =   R H , which is approximately 15% of the volume of
3 3 3 3 27 3
the larger cone.

3 2 3
aLv / (v u)3v  v 3 2 3
38. (a) E(v)=  E (v)=aL =0 when 2v =3uv  2v=3u v= u.
v u 2 2
(v u)
The First Derivative Test shows that this value of v gives the minimum value of E .

(b)

3 2 2 3
39. S=6sh s cot  +3s 
2 2
dS 3 2 2 2 3 3 2
(a) = s   3s  cot  s  (   3 cot  ) .
or
d 2 2 2
dS 1 cos  1
(b) =0 when   3 cot  =0  3 =0 cos  = . The First Derivative Test
d sin  sin  3
 1 1 
shows that the minimum surface area occurs when  =cos
55 .
3
(c)
15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

1 1 3
If cos  = , then cot  = and  = , so the surface area is S < =
3 2 2
3 2 1 2 3 3 3 2 9 2 6 2 1
6sh s +3s =6sh s+ s =6sh+ s =6s h+ s
2 2 2 2 2 2 2 2 2 2 2 2

40.

Let t be the time, in hours, after 2:00 P.M. The position of the boat heading south at time t is ( 0, 20t )
. The position of the boat heading east at time t is (  15+15t,0 ) . If D(t) is the distance between the
2 22 2 2 /
boats at time t , we minimize f (t)=[D(t)] =20 t +15 (t 1) . f (t)=800t+450(t 1)=1250t 450=0 when
450 60min / /
t= =0.36 h. 0.36 h  =21.6 min =21 min 36 s. Since f (t)>0 , this gives a minimum, so
1250 h
the boats are closest together at 2:21:36 P.M.

2
x +25 5 x / x 1 2
41. Here T (x)= + , 0 x 5 T (x)=  =0 8x=6 x +25 
6 8 2 8
6 x +25
15 15
2
(
16x =9 x +25 
2
)7
x=
. But
7
>5 , so T has no critical number. Since T (0)
1.46 and
T (5)
1.18 , he should row directly to B .

42.


In isosceles triangle AOB , O=180     , so BOC=2 . The distance rowed is 4cos  while the
distance walked is the length of arc BC=2(2 )=4 . The time taken is given by
16
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

4cos  4  / 1 
T ( )= + =2cos  + , 0  . T ( )= 2sin  +1=0 sin  =   = .
2 4 2 2 6

 
Check the value of T at  = and at the endpoints of the domain of T ; that is,  =0 and  = .
6 2
   
T (0)=2 , T = 3+
2.26 , and T =
1.57 . Therefore, the minimum value of T is
6 6 2 2
  / /
when  = ; that is, the woman should walk all the way. Note that T ( )= 2cos  <0 for
2 2
 
0  < , so  = gives a maximum time.
2 6

43.

3k k /  6k 2k
The total illumination is I(x)= + , 0<x<10 . Then I (x)= + =0
2 2 3 3
x (10 x) x (10 x)
3 3 3 3 3 3 3 3 3
6k(10 x) =2kx  3(10 x) =x  3(10 x)=x 10 3  3 x=x 10 3 =x+ 3 x
3
10 3
3
10 3 = 1+ 3 x ( 3
) x= 3

5.9 ft. This gives a minimum since I
/ /
(x)>0 for 0<x<10 .
1+ 3

44.

The line with slope m (where m<0 ) through ( 3,5) has equation y 5=m(x 3) or y=mx+ ( 5 3m ) . The
y intercept is 5 3m and the x intercept is  5/m+3 . So the triangle has area
1 9 / 25 9 2 25 5
A(m)= (5 3m)( 5/m+3)=15 25/(2m) m . Now A (m)=  =0 m =  m= (since
2 2 2 2 9 3
2m
/ / 25 5
m<0 ). A (m)= >0 , so there is an absolute minimum when m= . Thus, an equation of the
3 3
m
5 5
line is y 5= (x 3) or y= x+10 .
3 3

45.

17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

Every line segment in the first quadrant passing through ( a,b) with endpoints on the x  and y  axes
satisfies an equation of the form y b=m(x a) , where m<0 . By setting x=0 and then y=0 , we find its
b
endpoints, A(0,b am) and B a ,0 . The distance d from A to B is given by
m
b 2 2
d= a  0 + 0 (b am) .
m
It follows that the square of the length of the line segment, as a function of m , is given by
2 2
b 2ab b
2 2 2 2 2
S(m)= a +(am b) =a  + +a m  2abm+b . Thus,
m m 2
m
2
/
S (m) =
2ab
2

2b
3
2
+2a m 2ab=
2
3
( abm b +a m  abm )
2 2 4 3

m m m

=
2
3
b(am b)+am (am b) =
3 2
3
(am b) b+am( 3
)
m m

/ 3 b 3 b
Thus, S (m)=0 m=b/a or m= . Since b/a>0 and m<0 , m must equal  . Since
a a
2 / 3 b / 3 b
<0 , we see that S (m)<0 for m< and S (m)>0 for m> . Thus, S has its absolute
3 a a
m
3 b
minimum value when m= . That value is
a
2 2 2 2
= ( a+ ab ) + ( a b +b )
3 b 3 a 3 b 3 2 3 2
S  = a+b +  a  b
a b a
2 4/3 2/3 2/3 4/3 4/3 2/3 2/3 4/3 2 2 4/3 2/3 2/3 4/3 2
=a +2a b +a b +a b +2a b +b =a +3a b +3a b +b
3 2 2 3 3 2/3 2/3
The last expression is of the form x +3x y+3xy +y =(x+y) with x=a and y=b ,
2/3 2/3 3 2/3 2/3 3/2
so we can write it as (a +b ) and the shortest such line segment has length S =(a +b ) .

3 5 / 2 4
46. y=1+40x  3x  y =120x  15x , so the tangent line to the curve at x=a has slope
2 4 / 3 2 /
m(a)=120a  15a . Now m (a)=240a 60a = 60a(a  4)= 60a(a+2)(a 2) , so m (a)>0 for a< 2 ,
18
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

/
and 0<a<2 , and m (a)<0 for  2<a<0 and a>2 . Thus, m is increasing on (   , 2 ) , decreasing on
(  2,0 ) , increasing on ( 0,2 ) , and decreasing on ( 2, ) . Clearly, m(a)   as a   , so the
maximum value of m(a) must be one of the two local maxima, m( 2) or m(2) . But both m( 2) and
2 4
m(2) equal 120 2  15 2 =480 240=240 . So 240 is the largest slope, and it occurs at the points
(  2, 223) and ( 2,225) . Note: a=0 corresponds to a local minimum of m .
47.

2 2 2 2 2 2 1 2 2
Here s =h +b /4 , so h =s  b /4 . The area is A= b s  b /4 . Let the perimeter be p , so 2s+b= p
2
1 2 2 2
or s=( p b)/2 A(b)= b ( p b ) /4 b /4 =b p  2 pb /4 . Now
2
2 2
/ p  2 pb bp/4  3 pb+ p / 2
A (b)=  = . Therefore, A (b)=0  3 pb+ p =0 b= p/3 . Since
4 2 2
p  2 pb 4 p  2 pb
/ /
A (b)>0 for b< p/3 and A (b)<0 for b> p/3 , there is an absolute maximum when b= p/3 . But then
2s+ p/3= p , so s= p/3 s=b the triangle is equilateral.

48.
See the figure. The area is given by
A(x)=
1
2
2 2
(
2 a  x x+
1
2
2 2
2 a x) 2 2 2 2 2
( 2 2 2
x +b  a = a  x x+ x +b  a )( ) ( ) for 0 x a . Now
/ 2 2
A (x)= a  x 1+
x
2 2 2
2 2 2
+ x+ x +b  a
 x
2 2
=0 ( )
x +b  a a x
2 2 2
x+ x +b  a
x
2 2
( x+ 2
x +b  a
2 2
)= 2
a x
2
2 2 2
.
a x x +b  a
Except for the trivial case where x=0 , a=b and A(x)=0 , we have
19
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

2 2
2 2 2 x a x 2 2 2 2 2
x+ x +b  a >0 . Hence, cancelling this factor gives =  x x +b  a =a  x
2 2 2 2 2
a x x +b  a
2
 x
2
( x +b  a ) =a  2a x +x 
2 2 2 4 2 2 4 2
x ( b  a ) =a  2a x 
2 2 4 2 2
x
2
( b +a ) =a 
2 2 4
x=
a
.
2 2
a +b
Now we must check the value of A at this point as well as at the endpoints of the domain to see which
2 2
gives the maximum value. A(0)=a b  a , A(a)=0 and
2 2 2 2 2 2
a 2 a a a 2 2
A = a + +b  a =
2 2 2 2 2 2 2 2
a +b a +b a +b a +b

ab a
2
+
b
2
=
ab a +b ( 2 2
) =ab=0
2 2 2 2 2 2 2 2
a +b a +b a +b a +b
2
Since b
2
b a ,A a /
2
( 2
a +b
2 2
) A ( 0 ) . So there is an absolute maximum when x=
a
2 2
.
a +b
2ab
In this case the horizontal piece should be and the vertical piece should be
2 2
a +b
2 2
a +b 2 2
= a +b .
2 2
a +b

49. Note that AD = AP + PD  5=x+ PD  PD =5 x . Using the Pythagorean Theorem for
 PDB and  PDC gives us
L(x) = 2 2 2 2
AP + BP + CP =x+ (5 x) +2 + (5 x) +3
= x+ x2 10x+29 + x2 10x+34
/ x 5 x 5 /
 L (x)=1+ + From the graphs of L and L , it seems that the
2 2
x  10x+29 x  10x+34
minimum value of L is about L(3.59)=9.35 m.

50.

20
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

We note that since c is the consumption in gallons per hour, and v is the velocity in miles per hour,
c gallons/hour gallons
then = = gives us the consumption in gallons per mile, that is, the quantity G
v miles/hour mile
dc dv dc
v  c v  c
dG d c dv dv dv
. To find the minimum, we calculate = = = . This is 0 when
dv dv v 2 2
v v
dc dc c
v  c=0 = . This implies that the tangent line of c(v) passes through the origin, and this
dv dv v
occurs when v
53 mi / h. Note that the slope of the secant line through the origin and a point
( v,c(v) ) on the graph is equal to G(v) , and it is intuitively clear that G is minimized in the case where
the secant is in fact a tangent.

51.

The total time is

2 2 2 2
a +x b +(d x)
T (x) </ = ( time from A to C ) + ( time from C to B ) = + , 0<x<d
v v
1 2
sin  sin 
/ x d x 1 2
T (x)=  = 
2 2 2 2 v v
v a +x v b +(d x) 1 2
1 2

/
sin  sin 
1 2
The minimum occurs when T (x)=0 = .
v v
1 2

52.

21
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

If d= QT , we minimize f  ( ) = PR + RS =a  +b  . Differentiating with respect to  , and

1 1 2 1
d
df df 2
setting equal to 0 , we get =0= a  cot   b cot  .
d d 1 1 2 2 d
1 1 1

d
2
So we need to find an expression for . We can do this by observing that QT = constant
d
1
=acot  +bcot  .
1 2

2 2
d
2
Differentiating this equation implicitly with respect to  , we get  a   b  =0
1 1 2 d
1
2
d a 
2 1 df
= . We substitute this into the expression for to get
d 2 d
1 b  1
2
2 2
a  csc  cot 
1 1 2
 a cot   b cot   =0  a csc  cot  +a =0
1 1 2 2
b 
2 1 1 csc 
2 2

cot  cot 
1 2
cot  csc  =csc  cot   =  cos  =cos  . Since  and  are both acute, we
1 2 1 2 csc  csc  1 2 1 2
1 2
have  = .
1 2

53.

2 2 2
y =x +z , but triangles CDE and BCA are similar, so z/8=x
2 2 2 3
/ (4 x 4 )  z=2x / x 4 . Thus, we
minimize f ( x ) =y =x +4x /(x 4)=x /(x 4) , 4<x 8 .
22
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

/
f ( x) =
( 2)
(x 4) 3x  x
3
=
2
x 3(x 4) x
=
2
2x ( x 6 ) / /
=0 when x=6 . f (x)<0 when x<6 , f (x)>0
2 2 2
(x 4) (x 4) ( x 4 )
when x>6 , so the minimum occurs when x=6 in.

54.

Paradoxically, we solve this maximum problem by solving a minimum problem. Let L be the length

of the line ACB going from wall to wall touching the inner corner C . As   0 or   , we have
2
L  and there will be an angle that makes L a minimum. A pipe of this length will just fit around
the corner.
From the diagram, L=L +L =9csc  +6sec   dL/d = 9csc  cot  +6sec  tan  =0 when
1 2
3 9 2 3 2/3
6sec  tan  =9csc  cot   tan  = =1.5 tan  = [3]1.5 . Then sec  =1+ and
6 2
2 3  2/3
csc  =1+ , so the longest pipe has length
2
3  2/3 1/2
3 2/3 1/2
L=9 1+ +6 1+
21.07 ft.
2 2
 1 3
Or, use  =tan ( )
1.5
0.852  L=9  +6sec 
21.07 ft.

55.

3t tan  +tan  t+tan 

It suffices to maximize tan  . Now =tan ( + ) = = . So
1 1 tan  tan  1 ttan 
3t ( 1 ttan  ) =t+tan  
2
( )
2t= 1+3t tan   tan  =
2t
2
. Let f (t)=tan  =
2t
2

1+3t 1+3t
/
f (t)=
( 2
2 1+3t  2t(6t) ) =
(
2 1 3t
2
) =0 2
1 3t =0 t=
1
since t 0 .
( 1+3t2) 2 ( 1+3t2) 2 3
Now
23
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

/ 1 / 1 1
f (t)>0 for 0 t< and f (t)<0 for t> , so f has an absolute maximum when t=
3 3 3
2 ( 1/ 3 ) 1 
and tan  = =   = . Substituting for t and  in 3t=tan ( + ) gives us
2 3 6
1+3 ( 1/ 3 )
 
3 =tan  +   = .
6 6

56.

We maximize the cross sectional area A ( ) =

1 
10h+2 dh =10h+dh=10(10sin  )+(10cos  )(10sin  ) > =100(sin  +sin  cos  ), 0 
2 2
/
( 2 2
)
A ( ) =100 cos  +cos   sin  =100 cos  +2cos   1 =100(2cos   1)(cos  +1)=0 when ( 2
)
1  
cos  =   = . (cos    1 since 0  .)
2 3 2
  
Now A(0)=0 , A =100 and A =75 3
129.9 , so the maximum occurs when  = .
2 3 3

57.

5 2 
From the figure, tan  = and tan  = . Since  + + =180 = ,
x 3 x
5  1  1 2
 =  tan  tan 
x 3 x
1 5 1 2
d =  
5 2 2 2 2 2
dx 1+ x 1+ ( 3 x )
x 3 x
2 2
x 5 ( 3 x ) 2
= 2
  2
 .
2 2
x +25 x ( 3 x ) +4 ( 3 x )
24
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

d 5 2 2 2
Now =0  =  2x +50=5x  30x+65
dx 2 2
x +25 x  6x+13
2 2
3x  30x+15=0  x  10x+5=0 x=5 2 5 . We reject the root with the + sign,
since it is larger than 3 . d /dx>0 for x<5 2 5 and d /dx<0 for x>5 2 5 , so  is maximized when
AP =x=5 2 5
0.53 .

58. Let x be the distance from the observer to the wall. Then, from the given figure, <
 1 h+d  1 d
 =tan  tan , x>0
x x
d 1 h+d 1 d h+d d
=    = +
2 2 2 2 2 2 2 2
dx 1+ (h+d)/x x 1+ ( d/x ) x x + ( h+d ) x +d

=
2
d x + ( h+d )
2
 ( h+d ) x +d ( 2 2
)= 2
h d+hd  hx
2 2
=0 
2
x + ( h+d )
2
( x +d )
2 2 2
x + ( h+d )
2
( x +d )
2 2

2 2 2 2 2
hx =h d+hd  x =hd+d  x= d ( h+d ) . Since d /dx>0 for all x< d ( h+d ) and d /dx<0 for all
x> d ( h+d ) , the absolute maximum occurs when x= d ( h+d ) .

59.

a c
In the small triangle with sides a and c and hypotenuse W , sin  = and cos  = . In the triangle
W W
d b
with sides b and d and hypotenuse L , sin  = and cos  = . Thus, a=W sin  , c=W cos  ,
L L
d=Lsin  , and b=Lcos  , so the area of the circumscribed rectangle is
A( ) = (a+b)(c+d)=(W sin  +Lcos  )(W cos  +Lsin  )=1 12pt
= W 2sin  cos  +WLsin 2 +LW cos 2 +L2sin  cos  =1 12pt
= LW sin 2 +LW cos 2 + L2+W 2 sin  cos  =1 12pt ( )
2 1
2
( 2 2
) (
= LW sin  +cos  + L +W   2sin  cos  =1 12pt
2
)
1 
= LW +
2
2 2
(
L +W sin 2 , 0 
2
)
25
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

This expression shows, without calculus, that the maximum value of A( ) occurs when sin 2 =1

2 =
2
  1 1 1
  =
4
. So the maximum area is A
4
=LW +
2
2 2
L +W =
2
2 2
L +2LW +W = ( L+W ) .
2
2
( ) ( )
b
60. (a) Let D be the point such that a= AD . From the figure, sin  =  BC =b  and
BC
BD a AB
cos  = =  BC = ( a AB ) sec  . Eliminating BC gives ( a AB ) sec  =b  
BC BC
bcot  =a AB  AB =a bcot  . The total resistance is
AB BC a bcot  b csc 
R( )=C +C =C + .
4 4 4 4
r r r r
1 2 1 2
2
/ b csc  b csc  cot  csc  cot 
(b) R ( )=C  =bC csc   .
4 4 4 4
r r r r
1 2 1 2
4
r cot 
/ csc  cot  2
R ( )=0 =  = =cos  .
4 4 4 csc 
r r r
1 2 1
4 4
r r
/ csc  cot  2 / 2
R ( )>0 >  cos  < and R ( )<0 when cos  > , so there is an absolute
4 4 4 4
r r r r
1 2 1 1
4 4
minimum when cos  =r / r .
2 1
(c) 2 2 4  1 2 4 
When r = r , we have cos  = , so  =cos
79 .
2 3 1 3 3

61. (a)

If k= energy / km over land, then energy / km over water =1.4k . So the total energy is
2 dE 1.4kx
E=1.4k 25+x +k(13 x) , 0 x 13 , and so =  k.
dx 2 1/2
25+x ( )
dE 1/2 5
Set
dx
=0 : 1.4kx=k 25+x
2
( 2 2 2
)
 1.96x =x +25 0.96x =25 x=
0.96

5.1 . Testing against the

26
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

value of E at the endpoints: E(0)=1.4k(5)+13k=20k , E(5.1)

17.9k , E(13)
19.5k . Thus, to
minimize energy, the bird should fly to a point about 5.1 km from B .
(b) If W / L is large, the bird would fly to a point C that is closer to B than to D to minimize the
energy used flying over water. If W / L is small, the bird would fly to a point C that is closer to D than
2 dE Wx
to B to minimize the distance of the flight. E=W 25+x +L(13 x) =  L=0 when
dx 2
25+x
2
W 25+x
= . By the same sort of argument as in part (a), this ratio will give the minimal
L x
expenditure of energy if the bird heads for the point x km from B .
2
25+13
(c) For flight direct to D , x=13 , so from part (b), W / L=
1.07 . There is no value of
13
W / L for which the bird should fly directly to B . But note that lim (W / L)= , so if the point at
+
x 0
which E is a minimum is close to B , then W / L is large.
(d) Assuming that the birds instinctively choose the path that minimizes the energy expenditure, we
2 1/2
can use the equation for dE/dx=0 from part (a) with 1.4k=c , x=4 , and k=1 : (c)(4)=1 25+4 ( ) 
c= 41 /4
1.6 .

62. (a) I(x) 

strength of source
. Adding the intensities from the left and right lightbulbs,
2
( distance from source )
k k k k
I(x)= + = + .
2 2 2 2 2 2 2 2
x +d ( 10 x ) +d x +d x  20x+100+d
(b) The magnitude of the constant k won’t affect the location of the point of maximum intensity, so
/ 2x 2 ( x 10 )
for convenience we take k=1 . I (x)=  .
2 2 2
2 2
x +d
2
(
x  20x+100+d ) ( )
/
Substituting d=5 into the equations for I(x) and I (x) , we get
1 1 / 2x 2(x 10)
I (x)= 2 + 2 and I (x)= 
5 5 2
x +25 x  20x+125 2
x +25
2
(
x  20x+125 ) ( )2

27
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

From the graphs, it appears that I (x) has a minimum at x=5 m.

5

/ 1 1
(c) Substituting d=10 into the equations for I(x) and I (x) gives I (x)= 2
+ 2
and
10
x +100 x  20x+200
/ 2x 2 ( x 10 )
I (x)=  .
10 2 2
( x +100) ( x  20x+200)
2 2

From the graphs, it seems that for d=10 , the intensity is minimized at the endpoints, that is, x=0 and
x=10 . The midpoint is now the most brightly lit point!

(d) From the first figures in parts (b) and (c), we see that the minimal illumination changes from the
midpoint ( x=5 with d=5 ) to the endpoints ( x=0 and x=10 with d=10 ).

28
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.7 Optimization Problems

So we try d=6 (see the first figure) and we see that the minimum value still occurs at x=5 . Next, we
let d=8 (see the second figure) and we see that the minimum value occurs at the endpoints. It appears
that for some value of d between 6 and 8 , we must have minima at both the midpoint and the
endpoints, that is, I(5) must equal I(0) . To find this value of d , we solve I(0)=I(5) (with k=1 ):
1
2
+
1
2
=
1
2
+
1
2
=
2
2
 25+d
2 2
( 2 2
)(
100+d +d 25+d =2d 100+d 
2 2
) ( ) ( )
d 100+d 25+d 25+d 25+d
2 4 2 4 2 4 2 2
2500+125d +d +25d +d =200d +2d  2500=50d  d =50 d=5 2
7.071 (for 0 d 10 ). The
third figure, a graph of I(0) I(5) with d independent, confirms that I(0) I(5)=0 , that is, I(0)=I(5) ,
when d=5 2 . Thus, the point of minimal illumination changes abruptly from the midpoint to the
endpoints when d=5 2 .

29
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

1. (a) C(0) represents the fixed costs of production, such as rent, utilities, machinery etc., which are
incurred even when nothing is produced.
/ /
(b) The inflection point is the point at which C (x) changes from negative to positive; that is, the
/
marginal cost C (x) changes from decreasing to increasing. Thus, the marginal cost is minimized at
the inflection point.
/
(c) The marginal cost function is C (x) . We graph it as in Example 1 in Section.

/
2. (a) We graph C as in Example 1 in Section.

(b) By reading values of C(x) from its graph, we can plot c(x)=C(x)/x .

(c) Since the graph in part (b) is decreasing, we estimate that the minimum value of c(x) occurs at x=7
. The average cost and the marginal cost are equal at that value. See the box preceding Example 1.

2 /
3. c(x)=21.4 0.002x and c(x)=C(x)/x C(x)=21.4x 0.002x . C (x)=21.4 0.004x and
/
C (1000)=17.4 . This means that the cost of producing the 1001 st unit is about $17.40 .

4. (a) Profit is maximized when the marginal revenue is equal to the marginal cost; that is, when R
and C have equal slopes. See the box preceding Example 2.
1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

(b) P(x)=R(x) C(x) is sketched.

/
(c) The marginal profit function is defined as P (x) .

2
5. (a) The cost function is C(x)=40 , 000+300x+x , so the cost at a production level of 1000 is
C(x) 40,000
C(1000)=$1 , 340 , 000 . The average cost function is c(x)= = +300+x and
x x
/ /
c(1000)=$1340/ unit. The marginal cost function is C (x)=300+2x and C (1000)=$2300/ unit.
/ 40,000
(b) See the box preceding Example 1. We must have C ( x ) =c ( x )  300+2x= +300+x
x
40,000 2
x=  x =40 , 000 x= 40,000 =200 . This gives a minimum value of the average cost
x
/ / 80,000
function c ( x ) since c ( x ) = >0 .
3
x
(c) The minimum average cost is c(200)=$700/ unit.

2 C(x) 25,000
6. (a) C(x)=25 , 000+120x+0.1x , C(1000)=$245 , 000 . c(x)= = +120+0.1x ,
x x
/ /
c(1000)=$245/ unit. C (x)=120+0.2x , C (1000)=$320/ unit.
/ 25,000 25,000 2
(b) We must have C (x)=c(x) 120+0.2x= +120+0.1x 0.1x=  0.1x =25 , 000
x x
2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

/ / 50,000
x= 250,000 =500 . This gives a minimum since c (x)= >0 .
3
x
(c) The minimum average cost is c(500)=$220.00/ unit.

3/2
7. (a) C(x)=16 , 000+200x+4x , C(1000)=16 , 000+200 , 000+40 , 000 10  216 , 000+126 , 491
16,000 1/2
, so C(1000) $342 , 491 . c(x)=C(x)/x= +200+4x , c(1000) $342.49/unit .
x
/ 1/2 /
C (x)=200+6x , C (1000)=200+60 10  $389.74/unit .
/ 1/2 16,000 1/2 3/2 2/3
(b) We must have C (x)=c(x) 200+6x = +200+4x  2x =16 , 000 x=(8 , 000) =400
x
/  16,000 2 2 3/2
units. To check that this is a minimum, we calculate c (x)= + = (x  8000) . This is
2 x x2
x
2/3
negative for x<(8000) =400 , zero at x=400 , and positive for x>400 , so c is decreasing on ( 0,400 )
and increasing on ( 400, ) . Thus, c has an absolute minimum at x=400 .
(c) The minimum average cost is c(400)=40+200+80=$320/ unit.

2 3
8. (a) C(x)=10 , 000+340x 0.3x +0.0001x , C(1000)=$150 , 000 .
10,000 2 / 2
c(x)=C(x)/x= +340 0.3x+0.0001x , c(1000)=$150/unit . C (x)=340 0.6x+0.0003x ,
x
/
C (1000)=$40/unit .
/ 2 10,000 2
(b) We must have C (x)=c(x) 340 0.6x+0.0003x = +340 0.3x+0.0001x 
x
2 10,000 3 2 3 2
0.0002x = +0.3x 0.0002x  0.3x  10 , 000=0 x  1500x  50 , 000 , 000=0
x
/ / 20,000
x 1521.60 1522 units. This gives a minimum since c (x)= +0.0002>0.
3
x
(c) The minimum average cost is about c(1521.60) $121.62/ unit.

2 3 / 2
9. (a) C(x)=3700+5x 0.04x +0.0003x  C (x)=5 0.08x+0.0009x (marginal cost).
C(x) 3700 2
c(x)= = +5 0.04x+0.0003x (average cost).
x x

(b)
The graphs intersect at ( 208.51,27.45) , so the production level that minimizes average cost is about
3
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

209 units.
/ 3700 2 3
(c) c (x)=  0.04+0.0006x=0 3700+0.04x  0.0006x =0 x  208.51 . c x  $27.45/ unit.
( )
2 1 1
x
/
(d) The marginal cost is given by C (x) , so to find its minimum value we’ll find the derivative of
/ / / / / 800 /
C ; that is, C . C (x)= 0.08+0.0018x=0 x =
1 18
=44.44 . C x =$3.22/ unit.
1 ( )
/ / / / / /
C (x)=0.0018>0 for all x , so this is the minimum marginal cost. C is the second derivative
/ /
of C . cost is given by C (x) .

2 3 / 2
10. (a) C(x)=339+25x 0.09x +0.0004x  C (x)=25 0.18x+0.0012x (marginal cost).
C(x) 339 2
c(x)= = +25 0.09x+0.0004x (average cost).
x x

(b)
The graphs intersect at ( 135.56,22.65) , so the production level that minimizes average cost is about
136 units.

/ 339
(c) c (x)=  0.09+0.0008x=0 x  135.56 . c x  $22.65/ unit.
( )
2 1 1
x
/ / 1800 /
(d) C (x)= 0.18+0.0024x=0 x= =75 . C (75)=$18.25/ unit.
24
/ / /
C (x)=0.0024>0 for all x , so this is the minimum marginal cost.

2 / /
11. C(x)=680+4x+0.01x , p(x)=12 R(x)=xp(x)=12x . If the profit is maximum, then R (x)=C (x)
/ /
 12=4+0.02x 0.02x=8 x=400 . The profit is maximized if P (x)<0 , but since
/ / / / / / / / / /
P (x)=R (x) C (x) , we can just check the condition R (x)<C (x) . Now
/ / / /
R (x)=0<0.02=C (x) , so x=400 gives a maximum.

2 2
12. C(x)=680+4x+0.01x , p(x)=12 x/500 . Then R(x)=xp(x)=12x x /500 . If the profit is maximum,
/ / 1000
then R (x)=C (x) 12 x/250=4+0.02x 8=0.024x x=8/0.024= . The profit is maximized if
3
/ / / / / / / / / / / /
P (x)<0 , but since P (x)=R (x) C (x) , we can just check the condition R (x)<C (x) .
4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

/ / 1 / / 1000
Now R (x)= <0.02=C (x) , so x= gives a maximum.
250 3

2 3 2
13. C(x)=1450+36x x +0.001x , p(x)=60 0.01x . Then R(x)=xp(x)=60x 0.01x . If the profit is
/ / 2 2
maximum, then R (x)=C (x) 60 0.02x=36 2x+0.003x  0.003x  1.98x 24=0 . By the quadratic
2
1.98 ( 1.98) +4(0.003)(24) 1.98 4.2084
formula, x= = . Since x>0 ,
2(0.003) 0.006
/ / / / / /
x (1.98+2.05)/0.006 672 . Now R (x)= 0.02 and C (x)= 2+0.006x C (672)=2.032
/ / / /
R (672)<C (672) there is a maximum at x=672 .

2 3 2
14. C(x)=16 , 000+500x 1.6x +0.004x , p(x)=1700 7x . Then R(x)=xp(x)=1700x 7x . If the profit
/ / 2 2
is maximum, then R (x)=C (x) 1700 14x=500 3.2x+0.012x  0.012x +10.8x 1200=0
2
x +900x 100 , 000=0 (x+1000)(x 100)=0 x=100 (since x>0 ). The profit is maximized if
/ / / / / / / / / / / /
P (x)<0 , but since P (x)=R (x) C (x) , we can just check the condition R (x)<C (x) .
/ / / /
Now R (x)= 14< 3.2+0.024x=C (x) for x>0 , so there is a maximum at x=100 .

3 2 / 2
15. C(x)=0.001x  0.3x +6x+900 . The marginal cost is C (x)=0.003x  0.6x+6 .
/ / / /
C (x) is increasing when C (x)>0 0.006x 0.6>0 x>0.6/0.006=100 . So C (x) starts to increase
when x=100 .

3 2 / 2
16. C(x)=0.0002x  0.25x +4x+1500 . The marginal cost is C (x)=0.0006x  0.50x+4 .
/ / / /
C (x) is increasing when C (x)>0 0.0012x 0.5>0 x>0.5/0.0012 417 . So C (x) starts to
increase when x=417 .

2 3 2
17. (a) C(x)=1200+12x 0.1x +0.0005x . R(x)=xp(x)=29x 0.00021x . Since the profit is maximized
/ /
when R (x)=C (x) , we examine the curves R and C in the figure, looking for x values at which the
slopes of the tangent lines are equal. It appears that x=200 is a good estimate.

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

/ / 2 2
(b) R (x)=C (x) 29 0.00042x=12 0.2x+0.0015x  0.0015x  0.19958x 17=0 x 192.06 (for
/ / / /
x>0 ). As in Exercise 11, R (x)<C (x)  0.00042< 0.2+0.003x 0.003x>0.19958 x>66.5 . Our
value of 192 is in this range, so we have a maximum profit when we produce 192 yards of fabric.

3/4 2
18. (a) Cost=setupcost+manufacturingcost C(x)=500+m(x)=500+20x 5x +0.01x . We can solve
x( p)=320 7.7 p for p in terms of x to find the demand (or price) function. x=320 7.7 p 7.7 p=320 x
2
320 x 320x x
 p(x)= . R(x)=xp(x)= .
7.7 7.7
/ / 15  1/4 320 2x
(b) C (x)=R (x) 20 x +0.02x=  x 81.53 planes, and p(x)=$30.97 million. The
4 7.7
maximum profit associated with these values is about $463.59 million.

19. (a) We are given that the demand function p is linear and p(27 , 000)=10 , p(33 , 000)=8 , so the
10 8 1 1
slope is = and an equation of the line is y 10=  ( x 27,000 ) 
27,000 33,000 3000 3000
1
y= p(x)= x+19=19 (x/3000) .
3000
2 /
(b) The revenue is R(x)=xp(x)=19x (x /3000) R (x)=19 (x/1500)=0 when x=28 , 500 . Since
/ /
R (x)= 1/1500<0 , the maximum revenue occurs when x=28 , 500 the price is p(28 , 500)=$9.50
.

20. (a) Let p(x) be the demand function. Then p(x) is linear and y= p(x) passes through ( 20,10 ) and
1 1 1
( 18,11 ) , so the slope is  and an equation of the line is y 10= (x 20) y= x+20 . Thus, the
2 2 2
1 1 2
demand is p(x)= x+20 and the revenue is R(x)=xp(x)= x +20x .
2 2
1 2 /
(b) The cost is C(x)=6x , so the profit is P(x)=R(x) C(x)= x +14x . Then 0=P (x)= x+14 x=14 .
2
/ / 1
Since P (x)= 1<0 , the selling price for maximum profit is p(14)= (14)+20=$13 .
2

21. (a) As in Example 3, we see that the demand function p is linear. We are given that p(1000)=450
and deduce that p(1100)=440 , since a $10 reduction in price increases sales by 100 per week. The
440 450 1 1 1
slope for p is = , so an equation is p 450= (x 1000) or p(x)= x+550 .
1100 1000 10 10 10
1 2 / 1
(b) R(x)=xp(x)= x +550x . R (x)= x+550=0 when x=5(550)=2750 .
10 5
p(2750)=275 , so the rebate should be 450 275=$175 .
(c) C(x)=68 , 000+150x
6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.8 Applications to Business and Economics

1 2 1 2 / 1
P(x)=R(x) C(x)= x +550x 68 , 000 150x= x +400x 68 , 000 , P (x)= x+400=0 when
10 10 5
x=2000 . p(2000)=350 . Therefore, the rebate to maximize profits should be 450 350=$100 .

22. Let x denote the number of $10 increases in rent. Then the price is p(x)=800+10x , and the
number of units occupied is 100 x . Now the revenue is
R(x) = ( rental price per unit ) ( number of unit srented )
= (800+10x)(100 x)= 10x2+200x+80,000for0
x
100 
/ / /
R (x)= 20x+200=0 x=10 . This is a maximum since R (x)= 20<0 for all x . Now we must check
the value of R(x)= ( 800+10x ) ( 100 x ) at x=10 and at the endpoints of the domain to see which value
of x gives the maximum value of R . R(0)=80 , 000 , R(10)=(900)(90)=81 , 000 , and
R(100)=(1800)(0)=0 . Thus, the maximum revenue of $81 , 000/ week occurs when 90 units are
occupied at a rent of $900/ week.

800
23. If the reorder quantity is x , then the manager places orders per year. Storage costs for the
x
1 800 80,000
year are x 4=2x dollars. Handling costs are $100 per delivery, for a total of  100=
2 x x
80,000
dollars. The total costs for the year are C(x)=2x+ . To minimize C(x) , we calculate
x
/ 80,000 2 2
C (x)=2 = (x  40 , 000) . This is negative when x<200 , zero when x=200 , and positive
2 2
x x
when x>200 , so C is decreasing on ( 0,200 ) and increasing on ( 200, ) , reaching its absolute
minimum when x=200 . Thus, the optimal reorder quantity is 200 cases. The manager will place 4
orders per year for a total cost of C(200)=$800 .

24. She will have A/n dollars after each withdrawal and 0 dollars just before the next withdrawal, so
1
her average cash balance at any given time is (A/n+0)=A/(2n) . The transaction costs for n
2
withdrawals are nT . The lost interest cost on the average cash balance is A/(2n)  R . Thus, the total
AR / AR / AR 2 AR
cost for n transactions is C(n)=nT + . Now C (n)=T  and C (n)=0 =T  n = 
2n 2 2 2T
2n 2n
AR / / AR
n= , the value of n that minimizes total costs since C (n)= <0 . Thus, the optimal
2T 3
n
A A 2T AT AT
average cash balance is = = = .
2n 2 AR 2R 2R

7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

1. (a)

The tangent line at x=1 intersects the x  axis at x 2.3 , so x  2.3 . The tangent line at x=2.3
2
intersects the x  axis at x 3 , so x  3.0 .
3

(b) x =5 would not be a better first approximation than x =1 since the tangent line is nearly
1 1
horizontal. In fact, the second approximation for x =5 appears to be to the left of x=1 .
1

2.

The tangent line at x=9 intersects the x  axis at x 6.0 , so x  6.0 . The tangent line at x=6.0
2
intersects the x  axis at x 8.0 , so x  8.0 .
3

3. Since x =3 and y=5x 4 is tangent to y= f (x) at x=3 , we simply need to find where the tangent line
1
4
intersects the x axis. y=0 5x  4=0 x= .
2 2 5

4. (a)
If x =0 , then x is negative, and x is even more negative. The sequence of approximations does not
1 2 3
converge, that is, Newton’s method fails.

(b)
1
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

If x =1 , the tangent line is horizontal and Newton’s method fails.

1

(c)
If x =3 , then x =1 and we have the same situation as in part (b). Newton’s method fails again.
1 2

(d)
If x =4 , the tangent line is horizontal and Newton’s method fails.
1

3
x +2x  4
3 / 2 n n 1+2 4  1
5. f (x)=x +2x 4 f (x)=3x +2 , so x =x  . Now x =1 x =1 =1 =1.2
n+1 n 2 1 2 2 5
3x +2 3 1 +2
n
3
(1.2) +2(1.2) 4
x =1.2 2
 1.1797 .
3
3(1.2) +2

3 2
3 2 / 2 ( ) =x 
f x x x 1
n n n
6. f (x)=x  x  1 f (x)=3x  2x , so x =x  .
n+1 n / n 2
f (x ) 3x  2x
n n n
3 2
1 1 1 2 2 1
Now x =1 x =1 =2 x =2 =1.625 .
1 2 3 2 3 2
3 2  2 2

4
4 / 3 ( ) =x 
f x x  20
n n
7. f (x)=x  20 f (x)=4x , so x =x  .
n+1 n / n 3
f (x ) n
4x
n
4 4
2  20 (2.125)  20
Now x =2 x =2 =2.125 x =2.125  2.1148 .
1 2 3 3 3
4(2) 4(2.125)

5 / 4
8. f (x)=x +2 f (x)=5x , so

2
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

5
x +2 5 5
n ( 1) +2 1 ( 1.2) +2
x =x  . Now x = 1 x = 1 = 1 = 1.2  x = 1.2   1.1529 .
n+1 n 4 1 2 4 5 3 4
5x 5 ( 1) 5( 1.2)
n

3
x +x +3
3 / 2 n n
9. f (x)=x +x+3  f (x)=3x +1 , so x =x  2
. Now x = 1
n+1 n 1
3x +1
n
3
( 1) +( 1)+3  1 1+3 1
x = 1 = 1 = 1 = 1.25 . Newton’s method follows the tangent line at
2 2 3+1 4
3( 1) +1
(  1,1 ) down to its intersection with the x  axis at (  1.25,0 ) , giving the second approximation
x = 1.25 .
2

4
x x 1 4
4 / 3 n n 1  1 1  1 4
10. f (x)=x  x 1 f (x)=4x  1 , so x =x  . Now x =1 x =1 =1 = .
n+1 n 3 1 2 3 3 3
4x  1 4 1  1
n
4
Newton’s method follows the tangent line at ( 1, 1 ) up to its intersection with the x  axis at ,0
3
4
, giving the second approximation x = .
2 3

3 3 3 / 2
11. To approximate x= 30 (so that x =30 ), we can take f (x)=x  30 . So f (x)=3x , and thus,

3
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

3
x  30
n 3
x =x  . Since 27=3 and 27 is close to 30 , we’ll use x =3 . We need to find
n+1 n 2 1
3x
n
approximations until they agree to eight decimal places. x =3 x  3.11111111 , x  3.10723734 ,
1 2 3
3
x  3.10723251 x . So 30  3.10723251 , to eight decimal places. Here is a quick and easy
4 5
method for finding the iterations for Newton’s method on a programmable calculator. (The screens
3
shown are from the TI 83 Plus, but the method is similar on other calculators.) Assign f (x)=x  30 to
/ 2
Y , and f (x)=3x to Y . Now store x =3 in X and then enter X Y /Y  X to get x =3.1 . By
1 2 1 1 2 2
successively pressing the ENTER key, you get the approximations x , x , ... .
3 4

In Derive, load the utility file NEWTON ( x^3 30, x,3) and then APPROXIMATE to get . You can
request a specific iteration by adding a fourth argument. For example, NEWTON ( x^3 30, x,3, 2 )
gives [3,3.11111111,3.10723733] .
In Maple, make the assignments f:=x x^3 30; , g:=x x f (x)/D( f )(x); , and x:=3.; . Repeatedly
execute the command x:=g(x); to generate successive approximations.
/
In Mathematica, make the assignments f [x]:=x^3 30 , g[x]:=x f [x]/ f [x] , and x=3. Repeatedly
execute the command x=g[x] to generate successive approximations.

7
7 / 6
x  1000
n
12. f (x)=x  1000 f (x)=7x , so x =x  . We need to find approximations until they
n+1 n 6
7x
n
agree to eight decimal places. x =3 x  2.76739173 , x  2.69008741 , x  2.68275645 ,
1 2 3 4
7
x  2.68269580 x . So 1000  2.68269580 , to eight decimal places.
5 6

3 2 / 2
13. f (x)=2x  6x +3x+1 f (x)=6x  12x+3

4
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

3 2
2x  6x +3x +1
n n n
x =x  2
. We need to find approximations until they agree to six decimal places.
n+1 n
6x  12x +3
n n
x =2.5 x  2.285714 , x  2.228824 , x  2.224765 , x  2.224745 x . So the root is 2.224745 ,
1 2 3 4 5 6
to six decimal places.

4
x +x  4
4 / 3 n n
14. f (x)=x +x 4 f (x)=4x +1 x =x  3
. x =1.5 x  1.323276 , x  1.285346 ,
n+1 n 1 2 3
4x +1
n
x  1.283784 , x  1.283782 x . So the root is 1.283782 , to six decimal places.
4 5 6

2
2 2 /
sin x  x
n n
15. sin x=x , so f (x)=sin x x  f (x)=cos x 2x x =x  . From the figure, the
n+1 n cos x  2x
n n
2
positive root of sin x=x is near 1 . x =1 x  0.891396 , x  0.876985 , x  0.876726 x . So the
1 2 3 4 5
positive root is 0.876726 , to six decimal places.

4
4 4 / 3
2cos x  x
n n
16. 2cos x=x , so f (x)=2cos x x  f (x)= 2sin x 4x  x =x  . From the figure,
n+1 n 3
 2sin x  4x
n n
4
the positive root of 2cos x=x is near 1.x =1 x  1.014184 , x  1.013958 x . So the positive root
1 2 3 4
is 1.013958 , to six decimal places.

17.

5
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

From the graph, we see that there appear to be points of intersection near x= 0.7 and x=1.2 . Solving
4
x x 1
4 4 4 / 3 n n
x =1+x is the same as solving f (x)=x  x 1=0 . f (x)=x  x 1 f (x)=4x  1 , so x =x  3
.
n+1 n
4x  1
n

x = 0.7 x =1.2
1 1
x   0.725253 x  1.221380
2 2
x   0.724493 x  1.220745
3 3
x   0.724492 x x  1.220744 x
4 5 4 5

To six decimal places, the roots of the equation are  0.724492 and 1.220744 .

18.

5
From the graph, we see that reasonable first approximations are x=0.5 and x= 1.5 . f (x)=x  5x+2
5
x  5x +2
/ 4 n n
f (x)=5x  5 , so x =x  4
.
n+1 n
5x  5
n

6
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

x = 1.5 x =1.5
1 x =0.5 1
1
x   1.593846 x  1.396923
2 x =0.4 2
2
x   1.582241 x  1.373078
3 x  0.402102 x 3
3 4
x   1.582036 x x  1.371885
4 5 4
x  1.371882 x
5 6

To six decimal places, the roots are  1.582036 , 0.402102 , and 1.371882 .

19.

From the graph, we see that there appear to be points of intersection near x= 0.5 and x=1.5 . Solving
3 2 3 2 3 2 / 1  2/3
x =x  1 is the same as solving f (x)= x  x +1=0 . f (x)= x  x +1 f (x)= x  2x , so
3
2
3
x  x +1
n n
x =x  .
n+1 n 1  2/3
x  2x
3 n n

x = 0.5 x =1.5
1 1
x   0.471421 x  1.461653
2 2
x   0.471074 x x  1.461070 x
3 4 3 4

To six decimal places, the roots are  0.471074 and 1.461070 .

20.

From the graph, we see that there appear to be points of intersection near x= 1.2 and x=1.5 . Solving
7
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

2 2 2 / 1
x+3 =x is the same as solving f (x)=x  x+3 =0 . f (x)=x  x+3  f (x)=2x , so
2 x+3
2
x x +3
n n
x =x  .
n+1 n
2x  1
n / 2 x +3
n

x = 1.2 x =1.5
1 1
x   1.164526 x  1.453449
2 2
x   1.164035 x x  1.452627 x
3 4 3 4

To six decimal places, the roots of the equation are  1.164035 and 1.452627 .

21. From the graph, there appears to be a point of intersection near x=0.6 . Solving cos x= x is the
/
same as solving f (x)=cos x x =0 . f (x)=cos x x f (x)= sin x 1/ ( 2 x ) , so
cos x  x
n n
x =x  . Now x =0.6
n+1 n  sin x  1/ ( 2 x ) 1
n
x  0.641928 , x  0.641714 x . To six decimal places, the root of the equation is 0.641714 .
2 3 4

2
22. From the graph, there appears to be a point of intersection near x=0.7 . Solving tan x= 1 x is
2 2 / 2 2
the same as solving f (x)=tan x 1 x =0 . f (x)=tan x 1 x  f (x)=sec x+x/ 1 x , so
2
tan x  1 x
n n
x =x  . x =0.7 x  0.652356 , x  0.649895 , x  0.649889 x . To six
n+1 n 2 2 1 2 3 4 5
sec x +x / 1 x
n n n
decimal places, the root of the equation is 0.649889 .

8
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

23.

5 4 3 2
x  x  5x  x +4x +3
5 4 3 2 / 4 3 2 n n n n n
f (x)=x  x  5x  x +4x+3 f (x)=5x  4x  15x  2x+4 x =x  4 3 2
. From the
n+1 n
5x  4x  15x  2x +4
n n n n
graph of f , there appear to be roots near  1.4 , 1.1 , and 2.7 .

x = 1.4 x =1.1 x =2.7

1 1 1
x   1.39210970 x  1.07780402 x  2.72046250
2 2 2
x   1.39194698 x  1.07739442 x  2.71987870
3 3 3
x   1.39194691 x x  1.07739428 x x  2.71987822 x
4 5 4 5 4 5

To eight decimal places, the roots of the equation are  1.39194691 , 1.07739428 , and 2.71987822 .

24.

4 4
Solving x
2
( 4 x2) = 2
is the same as solving f (x)=4x  x 
2 4
2
/
=0 . f (x)=8x 4x +
3 8x
2

x +1 x +1 ( x +1)
2

2 4 2
4x  x  4 / x +1
n n n
x =x  . From the graph of f (x) , there appear to be roots near x= 1.9 and
n+1 n 3 2 2
8x  4x +8x / x +1
n n n n

9
 Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

x= 0.8 . Since f is even, we only need to find the positive roots.

x =0.8 x =1.9
1 1
x  0.84287645 x  1.94689103
2 2
x  0.84310820 x  1.94383891
3 3
x  0.84310821 x x  1.94382538 x
4 5 4 5

To eight decimal places, the roots of the equation are  0.84310821 and  1.94382538 .

25.

2 2 2 2
From the graph, y=x 2 x x and y=1 intersect twice, at x  2 and at x  1 . f (x)=x 2 x x  1

2 1 2  1/2 2 1/2
/
f (x) =x 
2
2 x x ( ( 1 2x)+)2 x x  2x ( )
1 2  1/2
= x 2 x x
2
( )
x( 1 2x)+4 2 x x
2
( )
x ( 8 5x 6x )
2
= ,
2 (2+x)(1 x)

2 2
x 2  x  x  1
n n /
so x =x  2
. Trying x = 2 won’t work because f ( 2) is undefined, so
n+1 n 1
x 8  5 x  6x
n n n

2 2(2 + x )(1  x )
n n
we’ll try x = 1.95 .
1

10
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

x = 1.95 x = 0.8
1 1
x   1.98580357 x   0.82674444
2 2
x   1.97899778 x   0.82646236
3 3
x   1.97807848 x   0.82646233 x
4 4 5
x   1.97806682
5
x   1.97806681 x
6 7

To eight decimal places, the roots of the equation are  1.97806681 and  0.82646233 .

26.

2
From the equations y=3sin (x ) and y=2x and the graph, we deduce that one root of the equation
2
3sin (x )=2x is x=0 . We also see that the graphs intersect at approximately x=0.7 and x=1.4 .
2
3sin (x ) 2x
2 / 2 n n
f (x)=3sin (x ) 2x f (x)=3cos (x ) 2x 2 , so x =x  2
.
n+1 n
6x cos (x ) 2
n n

x =0.7 x =1.4
1 1
x  0.69303689 x  1.39530295
2 2
x  0.69299996 x x  1.39525078
3 4 3
x  1.39525077 x
4 5

To eight decimal places, the roots of the equation are 0.69299996 and 1.39525077 .

27.

11
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

2
 x 2
From the graph, we see that y=e and y=x  x intersect twice. Good first approximations are x= 0.5
2
 x 2
n
2 2 e  x +x
 x 2 /  x n n
and x=1.1 . f (x)=e  x +x f (x)= 2xe  2x+1 , so x =x  2 .
n+1 n  x
n
 2x e  2x +1
n n

x = 0.5 x =1.1
1 1
x   0.51036446 x  1.20139754
2 2
x   0.51031156 x x  1.19844118
3 4 3
x  1.19843871 x
4 5

To eight decimal places, the roots of the equation are  0.51031156 and 1.19843871 .

28.

From the graph, y=ln 4 x ( 2

) and y=x intersect twice, at x  2 and at x 1 . f (x)=ln 4 x  x ( 2
)
2
ln 4 x  x
/  2x n n
f (x)=  1 , so x =x  . Trying x = 2 won’t work because it’s not in the
4 x
2

domain of y=ln 4 x (
n+1

2
n
 2x
n / 4 x
2
n
 1
1

) . Trying x1= 1.9 also fails after one iteration because the approximation x2 is
less than  2 . We try x = 1.99 .
1
x = 1.99 x =1.1
1 1
x   1.97753026 x  1.05864851
2 2
x   1.96741777 x  1.05800655
3 3
x   1.96475281 x  1.05800640 x
4 4 5
x   1.96463580
5
x   1.96463560 x
6 7

12
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

To eight decimal places, the roots of the equation are  1.96463560 and 1.05800640 .

2 /
29. (a) f (x)=x  a f (x)=2x , so Newton’s method gives
2
x a 1 1 1
n a a a
x =x  =x  x+ = x+ = x+ .
n+1 n 2x n 2 n 2x 2 n 2x 2 n x
n n n n
(b) Using (a) with a=1000 and x = 900 =30 , we get x  31.666667 , x  31.622807 , and
1 2 3
x  31.622777 x . So 1000  31.622777 .
4 5

1 1/x  a
/ 1 n 2 2
30. (a) f (x)=  a f (x)= , so x =x  =x +x  ax =2x  ax .
x 2 n+1 n 2 n n n n n
x  1/x
n
1
(b) Using (a) with a=1.6894 and x = =0.5 , we get x =0.5754 , x  0.588485 , and
1 2 2 3
x  0.588789 x . So 1/1.6984 0.588789 .
4 5

3 / 2 /
31. f (x)=x  3x+6 f (x)=3x  3 . If x =1 , then f ( x ) =0 and the tangent line used for
1 1
approximating x is horizontal. Attempting to find x results in trying to divide by zero.
2 2

3
x x 1
3 3 3 / 2 n n
32. x  x=1 x  x 1=0 . f (x)=x  x 1 f (x)=3x  1 , so x =x  2
.
n+1 n
3x  1
n
(a) x =1 , x =1.5 , x  1.347826 , x  1.325200 , x  1.324718 x
1 2 3 4 5 6
(b) x =0.6 , x =17.9 , x  11.946802 , x  7.985520 , x  5.356909 , x  3.624996 , x  2.505589 ,
1 2 3 4 5 6 7
x  1.820129 , x  1.461044 , x  1.339323 , x  1.324913 , x  1.324718 x
8 9 10 11 12 13
(c) x =0.57 , x   54.165455 , x   36.114293 , x   24.082094 , x   16.063387 , x   10.721483
1 2 3 4 5 6
, x   7.165534 , x   4.801704 , x   3.233425 , x   2.193674 , x   1.496867 ,
7 8 9 10 11
x   0.997546 , x   0.496305 , x   2.894162 , x   1.967962 , x   1.341355 ,
12 13 14 15 16
x   0.870187 , x   0.249949 , x   1.192219 , x   0.731952 , x  0.355213 ,
17 18 19 20 21
x   1.753322 , x   1.189420 , x   0.729123 , x  0.377844 , x   1.937872 ,
22 23 24 25 26
x   1.320350 , x   0.851919 , x   0.200959 , x   1.119386 , x   0.654291 ,
27 28 29 30 31

13
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

x  1.547010 , x  1.360051 , x  1.325828 , x  1.324719 , x  1.324718 x .

32 33 34 35 36 37

(d)
From the figure, we see that the tangent line corresponding to x =1 results in a sequence of
1
approximations that converges quite quickly ( x  x ). The tangent line corresponding to x =0.6 is
5 6 1
close to being horizontal, so x is quite far from the root. But the sequence still converges    just a
2
little more slowly ( x  x ). Lastly, the tangent line corresponding to x =0.57 is very nearly
12 13 1
horizontal, x is farther away from the root, and the sequence takes more iterations to converge (
2
x  x ).
36 37

1/3

33. For f (x)=x

1/3 1 
/
, f (x)= x
2/3
=x 
( ) =x 
f x
=x  3x = 2x . Therefore, each
and x
n
x
n
3 1  2/3 n n n+1 n
n / n
f
n x
3 n (x )
successive approximation becomes twice as large as the previous one in absolute value, so the
sequence of approximations fails to converge to the root, which is 0 . In the figure, we have x =0.5 ,
1
x = 2(0.5)= 1 , and x = 2( 1)=2 .
2 3

x
n
34. According to Newton’s Method, for x >0 , x =x  =x  2x = x and for x <0 ,
n n+1 n n n n n
1/ 2 x
n

  x
n
x =x  =x   2  x
( ) = x . So we can see that after choosing any value x the
n+1 n n n n 1
1/ 2  x
n

14
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

subsequent values will alternate between  x and x and never approach the root.
1 1

4 3 2 / 3 2 / / 2
35. (a) f (x)=3x  28x +6x +24x f (x)=12x  84x +12x+24 f (x)=36x  168x+12 . Now to
/
/ 1
solve f (x)=0 , try x =  x =x 
f (x ) = 2 
1
x  0.6455 x  0.6452 x  0.6452 . Now
1 2 2 1 / / 3 3 4 5
f (x )1
try x =6 x =7.12
1 2
x  6.8353 x  6.8102 x  6.8100 . Finally try x = 0.5 x   0.4571 x   0.4552
3 4 5 1 2 3
x   0.4552 . Therefore, x= 0.455 , 6.810 and 0.645 are all critical numbers correct to three decimal
4
places.
(b) f ( 1)=13 , f (7)= 1939 , f (6.810)  1949.07 , f ( 0.455)  6.912 , f (0.645) 10.982 .
Therefore, f (6.810)  1949.07 is the absolute minimum correct to two decimal places.

2 / /
36. f (x)=x +sin x f (x)=2x+cos x . f (x) exists for all x , so to find the minimum of f , we can
/ /
examine the zeros of f . From the graph of f , we see that a good choice for x is x = 0.5 . Use
1 1
/
g(x)=2x+cos x and g (x)=2 sin x to obtain x   0.450627 , x   0.450184 x . Since
2 3 4
/ /
f (x)=2 sin x>0 for all x , f ( 0.450184)  0.232466 is the absolute minimum.

37.

15
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

cos x
From the figure, we see that y= f (x)=e is periodic with period 2 . To find the x  coordinates of
/ / / cos x
the IP, we only need to approximate the zeros of y on 0, . f (x)= e sin x
f
/ /
(x)=e
cos x
( sin 2x cos x) . Since ecos x
2
0 , we will use Newton’s method with g(x)=sin x cos x ,
/
g (x)=2sin xcos x+sin x , and x =1 . x  0.904173 , x  0.904557 x . Thus, ( 0.904557,1.855277)
1 2 3 4
is the IP.

38.

/ /
f (x)= sin x f (x)= cos x . At x=a , the slope of the tangent line is f (a)= cos a . The line through
 sin a 0
the origin and ( a,f(a) ) is y= x . If this line is to be tangent to f at x=a , then its slope must
a 0
/  sin a
equal f (a) . Thus, = cos a tan a=a . To solve this equation using Newton’s method, let
a
g(x)=tan x x ,

tan x  x
/ 2 n n
g (x)=sec x 1 , and x =x  2
with x =4.5 (estimated from the figure). x  4.493614 ,
n+1 n 1 2
sec x  1
n
x  4.493410 , x  4.493409 x . Thus, the slope of the line that has the largest slope is
3 4 5
/
f ( x )
5
0.217234 .

39.

1 4 3 2 2 3 2
The volume of the silo, in terms of its radius, is V (r)= r (30)+  r =30 r +  r .
2 3 3
From a graph of V , we see that V (r)=15 , 000 at r 11 ft. Now we use Newton’s method to solve the
dV 2
equation V (r) 15 , 000=0 . =60 r+2 r , so
dr
16
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

2 2 3
30 r +  r  15,000
n 3 n
r =r  . Taking r =11 , we get r  11.2853 , r  11.2807 r . So in order
n+1 n 2 1 2 3 4
60 r +2 r
n n
3
for the silo to hold 15 , 000 ft of grain, its radius must be about 11.2807 ft.

40. Let the radius of the circle be r . Using s=r , we have 5=r and so r=5/ . From the Law of
2 2 2 2 2
Cosines we get 4 =r +r  2  r  r  cos  16=2r (1 cos  )=2 ( 5/ ) (1 cos  ) .
2 2 2
Multiplying by  gives 16 =50(1 cos  ) , so we take f ( )=16 +50cos   50 and
2
/
16 +50cos   50
n n
f ( )=32  50sin  . The formula for Newton’s method is  =  . From the
n+1 n 32  50sin 
n n
graph of f , we can use  =2.2 , giving us   2.2662 ,   2.2622  . So correct to four decimal
1 2 3 4

places, the angle is 2.2622 radians  130 .

R  n
41. In this case, A=18 , 000 , R=375 , and n=5(12)=60 . So the formula A= 1 (1+i) becomes
i
375  60  60 60
18 , 000= 1 (1+x) 48x=1 (1+x) [ multiply each term by (1+x) ]
x
60 60
48x(1+x)  (1+x) +1=0 . Let the LHS be called f (x) , so that
/ 59 60 59
f (x) = 48x(60)(1+x) +48(1+x)  60(1+x)
= 12(1+x)59 4x(60)+4(1+x) 5 =12(1+x)59(244x 1)
60 60

x =x 
48x
n ( 1+x )  ( 1+x ) +1
n n
. An interest rate of 1% per month seems like a reasonable
n+1 n 59
12 ( 1+x ) ( 244x  1 )
n n
estimate for x=i . So let x =1%=0.01 , and we get x  0.0082202 , x  0.0076802 , x  0.0076291 ,
1 2 3 4
x  0.0076286 x . Thus, the dealer is charging a monthly interest rate of 0.76286% (or 9.55% per
5 6
year, compounded monthly).
17
Stewart Calculus ET 5e 0534393217;4. Applications of Differentiation; 4.9 Newton’s Method

5 4 3 2
42. (a) p(x)=x  (2+r)x +(1+2r)x  (1 r)x +2(1 r)x+r 1
/ 4 3 2
p (x)=5x  4(2+r)x +3(1+2r)x  2(1 r)x+2(1 r) . So we use
5 4 3 2
x  (2+r)x +(1+2r)x  (1 r)x +2(1 r)x +r 1
n n n n n  6
x =x  4 3 2
. We substitute in the value r 3.04042 10
n+1 n
5x  4(2+r)x +3(1+2r)x  2(1 r)x +2(1 r)
n n n n
in order to evaluate the approximations numerically. The libration point L is slightly less than 1 AU
1
from the Sun, so we take x =0.95 as our first approximation, and get x  0.96682 , x  0.97770 ,
1 2 3
x  0.98451 , x  0.98830 , x  0.98976 , x  0.98998 , x  0.98999 x .
4 5 6 7 8 9
So, to five decimal places, L is located 0.98999 AU from the Sun (or 0.01001 AU from Earth).
1
(b) In this case we use Newton’s method with the function
2 5 4 3 2
p(x) 2rx =x  (2+r)x +(1+2r)x  (1+r)x +2(1 r)x+r 1
2 / 4 3 2
p(x) 2rx =5x  4(2+r)x +3(1+2r)x  2(1+r)x+2(1 r) . So
5 4 3 2
x  (2+r)x +(1+2r)x  (1+r)x +2(1 r)x +r 1
n n n n n  6
x =x  4 3 2
. Again, we substitute r 3.04042 10 . L is
n+1 n 2
5x  4(2+r)x +3(1+2r)x  2(1+r)x +2(1 r)
n n n n
slightly more than 1 AU from the Sun and, judging from the result of part (a), probably less than 0.02
AU from Earth. So we take x =1.02 and get x  1.01422 , x  1.01118 , x  1.01018 ,
1 2 3 4
x  1.01008 x . So, to five decimal places, L is located 1.01008 AU from the Sun (or 0.01008 AU
5 6 2
from Earth).

18
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

1. (a) Since f is increasing , we can obtain a lower estimate by using left endpoints. We are instructed
to use five rectangles, so n=5 .
L = 5
5 i=1( ) x
f x
i 1

= f ( x )  2+ f ( x )  2+ f ( x )  2+ f ( x )  2+ f ( x )  2
0 1 2 3 4
=2 f (0)+ f (2)+ f (4)+ f (6)+ f (8)
 2(1+3+4.3+5.4+6.3)=2(20)=40

Since f is increasing , we can obtain an upper estimate by using right endpoints.

R = 5
5 ( )f x  x
i=1 i

=2 f (x )+ f (x )+ f (x )+ f (x )+ f (x )
1 2 3 4 5
=2 f (2)+ f (4)+ f (6)+ f (8)+ f (10)
 2(3+4.3+5.4+6.3+7)=2(26)=52

Comparing R to L , we see that we have added the area of the rightmost upper rectangle, f (10) 2 ,
5 5
to the sum and subtracted the area of the leftmost lower rectangle, f (0) 2 , from the sum. aaaaa
(b)
=
10
L
10 i=1 ( )
f x
i 1
x

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

=1 f x + f x +   + f x
( ) ( ) ( )
0 1 9
= f (0)+ f (1)+   + f ( 9 )
 1+2.1+3+3.7+4.3+4.9+5.4+5.8+6.3+6.7
=43.2

=
10
R
10 f x  x= f (1)+ f (2)+   + f (10)
( )
i=1 i

=L +1 f (10) 1 f (0) [add rightmost upper rectangle, subtract leftmost lower rectangle ]
10
=43.2+7 1=49.2

2. (a)
(i)
L = 6
6 i=1 ( ) x
f x
i 1

=2 f (x )+ f (x )+ f (x )+ f (x )+ f (x )+ f (x )
0 1 2 3 4 5
=2[ f (0)+ f (2)+ f (4)+ f (6)+ f (8)+ f (10)]
 2(9+8.8+8.2+7.3+5.9+4.1)
=2(43.3)=86.6

2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

(ii) R =L +2 f (12) 2 f (0)

6 6
 86.6+2(1) 2(9)=70.6

(iii)
M = 6 *
6 f x  x
i=1 i
=2[ f (1)+ f (3)+ f (5)+ f (7)+ f (9)+ f (11)]
 2(8.9+8.5+7.8+6.6+5.1+2.8)
=2(39.7)=79.4

(b) Since f is decreasing , we obtain an overestimate by using left endpoints; that is, L .
6
(c) Since f is decreasing , we obtain an underestimate by using right endpoints; that is, R .
6
(d) M gives the best estimate, since the area of each rectangle appears to be closer to the true area
6
than the overestimates and underestimates in L and R .
6 6

3. (a)
3
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

R = 4
4 f x  x
( )
i=1 i

= f x  1+ f x  1+ f x  1+ f x  1
( ) ( ) ( ) ( )
1 2 3 4
= f (2)+ f (3)+ f (4)+ f (5)
1 1 1 1 77
= + + + = =1.283
2 3 4 5 60

Since f is decreasing on 1,5 , an underestimate is obtained by using the right endpoint

approximation, R .
4

(b)
L = 4
4 i=1
f x( )
i 1
x
= f (1)+ f (2)+ f (3)+ f (4)
1 1 1 25
=1+ + + = =2.083
2 3 4 12

L is an overestimate. Alternatively, we could just add the area of the leftmost upper rectangle and
4
subtract the area of the rightmost lower rectangle; that is, L =R + f (1) 1 f (5) 1 .
4 4

4. (a)
R = 5
5 f x  x
( )
i=1 i

4
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

= f x  1+ f x  1+ f x  1+ f x  1+ f x  1
( ) ( ) ( ) ( ) ( )
1 2 3 4 5
= f (1)+ f (2)+ f (3)+ f (4)+ f (5)
=24+21+16+9+0=70

Since f is decreasing on 0,5 , R is an underestimate.

5

(b)
L = 5
5 i=1 ( )
f x
i 1
x
= f (0)+ f (1)+ f (2)+ f (3)+ f (4)
=25+24+21+16+9=95

L is an overestimate.
5

2 2 ( 1)
5. (a) f (x)=1+x and  x= =1 R =1 f (0)+1 f (1)+1 f (2)=1 1+1 2+1 5=8 .
3 3
2 ( 1)
 x= =0.5
6
R =0.5[ f ( 0.5)+ f (0)+ f (0.5)+ f (1)+ f (1.5)+ f (2)]
6
=0.5(1.25+1+1.25+2+3.25+5)
=0.5(13.75)=6.875

5
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

(b)
L =1 f ( 1)+1 f (0)+1 f (1)=1 2+1 1+1 2=5
3

L =0.5[ f ( 1)+ f ( 0.5)+ f (0)+ f (0.5)+ f (1)+ f (1.5)]

6
=0.5(2+1.25+1+1.25+2+3.25)
=0.5(10.75)=5.375

6
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

(c)
M =1 f ( 0.5)+1 f (0.5)+1 f (1.5)
3
=1 1.25+1 1.25+1 3.25=5.75
M =0.5[ f ( 0.75)+ f ( 0.25)+ f (0.25)]
6
+ f (0.75)+ f (1.25)+ f (1.75)]
=0.5(1.5625+1.0625+1.0625+1.5625+2.5625+4.0625)
=0.5(11.875)=5.9375

(d) M appears to be the best estimate.

6

6. (a)

2
 x 2 (  2 )
(b) f (x)=e and  x= =1
4
(i)  1  1  4
R =1 f ( 1)+1 f (0)+1 f (1)+1 f (2)=e +1+e +e  1.754
4

7
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

(ii)  2.25  0.25  0.25  2.25

M =1 f ( 1.5)+1 f ( 0.5)+1 f (0.5)+1 f (1.5)=e +e +e +e  1.768
4

(c)
(i)

+e
8
 0.25  1
+e +e
 2.25  4
+e  1.761
.
R =0.5 f ( 1.5)+ f ( 1)+ f ( 0.5)+ f (0) + f (0.5)+ f (1)+ f (1.5)+ f (2) =e
 2.25  1
+e +e
 0.25
+1

(ii) Due to the symmetry of the figure, we see that M =(0.5)(2)[ f (0.25)+ f (0.75)+ f (1.25)+ f (1.75)]
8
 0.0625  0.5625  1.5625 3.0625
=e +e +e +e  1.766

7. Here is one possible algorithm (ordered sequence of operations) for calculating the sums:
(a) Let SUM =0 , X_MIN =0 , X_MAX = , N =10 (or 30 or 50 , depending on which sum we are
calculating), DELTA_X = ( XMAX XMIN ) / N, and RIGHT_ENDPOINT = X_MIN + DELTA_X.
(b) Repeat steps 2a, 2b in sequence until RIGHT_ENDPOINT > X_MAX.
(c) Add sin ( RIGHTENDPOINT ) to SUM.
(d) Add DELTA_X to RIGHT_ENDPOINT.
At the end of this procedure, ( DELTAX )  ( SUM ) is equal to the answer we are looking for. We find
 i  i
 sin 10  1.9835 , R30= 30  i=1sin 30  1.9982 , and
10 30
that R =
10 10 i=1
 i

50
R = sin  1.9993 . It appears that the exact area is 2 .
50 50 i=1 50
Shown below is program SUMRIGHT and its output from a TI 83 Plus calculator. To generalize the
8
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

program, we have input (rather than assigned) values for Xmin, Xmax, and N. Also, the function,
sin x , is assigned to Y , enabling us to evaluate any right sum merely by changing Y and running
1 1
the program.

8. We can use the algorithm from Exercise 7 with X_MIN =1 , X_MAX =2 , and
/ 1 10 1
2
1 ( RIGHTENDPOINT ) instead of sin ( RIGHTENDPOINT ) in step 2a. We find that
1 30 1
R = 
10 10 i=1 2
 0.4640 , R = 
30 30 i=1 2
 0.4877 , and
( 1+i/10 ) ( 1+i/30 )
1 50 1 1
R = 
50 50 i=1 2
 0.4926 . It appears that the exact area is
2
.
( 1+i/50 )
9. In Maple, we have to perform a number of steps before getting a numerical answer. After loading
the student package we use the command sum: =leftsum(x^(1 / 2) , x=1..4,10 [ or 30, or 50]) ;
which gives us the expression in summation notation. To get a numerical approximation to the sum,
we use evalf(left_sum);. Mathematica does not have a special command for these sums, so we must
type them in manually. For example, the first left sum is given by
(3 / 10)*Sum[Sqrt[1 + 3 (i  1) / 10] , {i, 1, 10}] , ], and we use the N command on the resulting
output to get a numerical approximation.
In Derive, we use the LEFT_RIEMANN command to get the left sums, but must define the right
sums ourselves. (We can define a new function using LEFT_RIEMANN with k ranging from 1 to n
instead of from 0 to n 1 .)
3 n 3(i 1)
(a) With f (x)= x , 1 x 4 , the left sums are of the form L =  1+ . Specifically,
n n i=1 n
3 3i

n
L  4.5148 , L  4.6165 , and L  4.6366 . The right sums are of the form R = 1+ .
10 30 50 n n i=1 n
Specifically, R  4.8148 , R  4.7165 , and R  4.6966 .
10 30 50
(b) In Maple, we use the leftbox and rightbox commands (with the same arguments as leftsum and
rightsum above) to generate the graphs.
9
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

left endpoints, n=10

left endpoints, n=30

left endpoints, n=50

right endpoints, n=10

10
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

right endpoints, n=30

right endpoints, n=50

(c) We know that since x is an increasing function on ( 1,4 ) , all of the left sums are smaller than the
actual area, and all of the right sums are larger than the actual area. Since the left sum with n=50 is
about 4.637>4.6 and the right sum with n=50 is about 4.697<4.7 , we conclude that 4.6<L < exact
50
area <R <4.7 , so the exact area is between 4.6 and 4.7 .
50

10.
   ( i 1 )

n
(a) With f (x)=sin (sin x) , 0 x , the left sums are of the form L =
sin sin .
2 n 2n i=1 2n
In particular, L  0.8251 , L  0.8710 , and L  0.8799 . The right sums are of the form
10 30 50
  i

n
R= sin sin . In particular, R  0.9573 , R  0.9150 , and R  0.9064 .
n 2n i=1 2n 10 30 50
(b) In Maple, we use the leftbox and rightbox commands (with the same arguments as leftsum and
rightsum above) to generate the graphs.

left endpoints, n=10

left endpoints, n=30

11
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

left endpoints, n=50

right endpoints, n=10

right endpoints, n=30

righnt endpoints, n=10


(c) We know that since sin (sin x) is an increasing function on 0,
, all of the left sums are
2
smaller than the actual area, and all of the right sums are larger than the actual area. Since the left sum
with n=50 is about 0.8799>0.87 and the right sum with n=50 is about 0.9064<0.91 , we conclude that
0.87<L < exact area <R <0.91 , so the exact area is between 0.87 and 0.91 .
50 50

11. Since v is an increasing function, L will give us a lower estimate and R will give us an upper
6 6
estimate.
L = ( 0ft/s ) ( 0.5s ) + ( 6.2 ) ( 0.5) + ( 10.8 ) ( 0.5) + ( 14.9 ) ( 0.5) + ( 18.1 ) ( 0.5) + ( 19.4 ) ( 0.5)
6
=0.5 ( 69.4 ) =34.7 ft

R =0.5 ( 6.2+10.8+14.9+18.1+19.4+20.2 ) =0.5 ( 89.6 ) =44.8 ft

6

12
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

12.
(a)
d L =(30 ft / s )(12 s )+28 12+25 12+22 12+24 12
5
=(30+28+25+22+24) 12=129 12=1548 ft
(b) d R = ( 28+25+22+24+27)  12=126 12=1512 ft
5
(c) The estimates are neither lower nor upper estimates since v is neither an increasing nor a
decreasing function of t .

13. Lower estimate for oil leakage: R =(7.6+6.8+6.2+5.7+5.3)(2)=(31.6)(2)=63.2 L.

5
Upper estimate for oil leakage: L =(8.7+7.6+6.8+6.2+5.7)(2)=(35)(2)=70 L.
5

14. We can find an upper estimate by using the final velocity for each time interval. Thus, the
distance d traveled after 62 seconds can be approximated by
d=
6
v t  t = ( 185ft/s ) ( 10s ) +319 5+447 5+742 12+1325 27+1445 3=54 , 694 ft
()
i=1 i i

15. For a decreasing function, using left endpoints gives us an overestimate and using right endpoints
results in an underestimate. We will use M to get an estimate.  t=1 , so
6
M = 1[v(0.5)+v(1.5)+v(2.5)+v(3.5)+v(4.5)+v(5.5)]
6
 55+40+28+18+10+4=155ft
For a very rough check on the above calculation, we can draw a line from ( 0,70 ) to ( 6,0 ) and
1
calculate the area of the triangle: (70)(6)=210 . This is clearly an overestimate, so our midpoint
2
estimate of 155 is reasonable.

16. For an increasing function, using left endpoints gives us an underestimate and using right
30 0 5
endpoints results in an overestimate. We will use M to get an estimate.  t= =5 s = h
6 6 3600
1
= h.
720
M = 1 [v(2.5)+v(7.5)+v(12.5)+v(17.5)+v(22.5)+v(27.5)]
6 720
1 1
= (31.25+66+88+103.5+113.75+119.25)= (521.75) 0.725km
720 720
For a very rough check on the above calculation, we can draw a line from ( 0,0 ) to ( 30,120 ) and
calculate the area
of the triangle:
13
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

1
(30)(120)=1800 . Divide by 3600 to get 0.5 , which is clearly an underestimate, making our
2
midpoint estimate of 0.725 seem reasonable. Of course, answers will vary due to different readings of
the graph.

4
17. f (x)= x , 1 x 16 .  x=(16 1)/n=15/n and x =1+i  x=1+15i/n .
i
15i 15
 
n n 4
A=lim R =lim f x  x=lim
( ) 1+  .
n

n
n

i=1 i
n

i=1 n n

ln x
18. f (x)= , 3 x 10 .  x=(10 3)/n=7/n and x =3+i  x=3+7i/n .
x i
n ln (3+7i/n) 7
A=lim R =lim  f x  x=lim 
n
 ( ).
n i=1 i i=1 3+7i/n n
n
n
n

   
19. f (x)=xcos x , 0 x .  x=(  0)/n= /n and x =0+i  x= i/n .
2 2 2 i 2
i i 
 
n n
A=lim R =lim f x  x=lim
( ) cos  .
n

n
n

i=1 i
n

i=1 2n 2n 2n

2 2i 10

n
20. lim 5+ can be interpreted as the area of the region lying under the graph of
n

i=1 n n
10 10 2 0 2 2i
y=(5+x) on the interval 0,2 , since for y=(5+x) on 0,2 with  x= = , x =0+i  x= ,
n n i n
2i 10 2
 
* n * n
and x =x , the expression for the area is A=lim f (x ) x=lim 5+ . Note that
i i
n

i=1 i
n

i=1 n n
10 10
the answer is not unique. We could use y=x on 5,7 or, in general, y=((5 n)+x) on n,n+2 .

 i

n
21. lim tan can be interpreted as the area of the region lying under the graph of y=tan x
i=1 4n 4n
n

   /4 0  i
on the interval 0, , since for y=tan x on 0, with  x= = , x =0+i  x= ,
4 4 n 4n i 4n
i 
and x =x , the expression for the area is A=lim  f x  x=lim 
* n * n
tan . Note
i i
n

i=1 i
n

i=1 4n 4n
that this answer is not unique, since the expression for the area is the same for the function

y=tan (x k ) on the interval k ,k + , where k is any integer.
4

22. (a)
14
 Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

1 0 1 i i 3 1
 
n n
 x= = and x =0+i  x= . A=lim R =lim f x  x=lim
( )  .
n n i n n

n
n

i=1 i
n

i=1 n n
3 2 2 2
i 1 1 1 n(n+1) ( n+1 ) 1 1 1
 
n n 3
(b) lim  =lim i =lim =lim = lim 1+ =

i=1 3 n n
4 i=1 4 2 2 4 n
n 4
n n n n
n n
4n

5 2 0 2 2i
23. (a) y= f (x)=x .  x= = and x =0+i  x= .
n n i n
5 5
2i 2 32i 2 64
   
n n n n 5
A=lim R =lim f x  x=lim
( )  =lim  =lim i .
n

n
n

i=1 i
n

i=1 n n n
i=1 5 n n
6 i=1
n n

(b) 
n
i=
n (n+1) 2n +2n 1
5
2 2
( 2
)
i=1 12

(c) lim
64 n (n+1) 2n +2n 1
 =
64
lim
n +2n+1 2n +2n 1
2 2
( 2
) ( 2
)( 2
)
n
n 6 12 12 n
2 2
n n
16 2 1 2 1 16 32
= lim 1+ + 2+  =  1 2=
3 n
n 2 n 2 3 3
n n

 2
24. From Example 3(a), we have A=lim
2

n
e
 2i/n
. Using a CAS, 
n  2i/n
e =
e ( e2 1) and
n
n
i=1 i=1 2/n
e  1
 2
lim
2 e

( e2 1) =e 2 ( e2 1)  0.8647 , whereas the estimate from Example 3(b) using M was
n
n
2/n 10
e  1
0.8632 .

b 0 b bi
25. y= f (x)=cos x .  x= = and x =0+i  x= .
n n i n

1
bsin b +1
bi b 2n b
 
n n
A=lim R =lim f x  x=lim
( ) cos  =lim  =sin b
n

n
n

i=1 i
n

i=1 n n n
b 2n
2nsin
2n

 
If b= , then A=sin =1 .
2 2

26. (a)

15
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.1 Areas and Distances

The diagram shows one of the n congruent triangles,  AOB , with central angle 2 /n . O is the
center of the circle and AB is one of the sides of the polygon. Radius OC is drawn so as to bisect
 AOB . It follows that OC intersects AB at right angles and bisects AB . Thus,  AOB is divided into
1
two right triangles with legs of length (AB)=rsin ( /n ) and rcos ( /n ) .
2

1 2 1 2
 AOB has area 2 [rsin ( /n ) ][rcos ( /n ) ]=r sin ( /n)cos ( /n)= r sin (2 /n) , so A =n area
2 2 n
1 2
(  AOB) = nr sin (2 /n) .
2
sin 
(b) To use Equation 3.4.2, lim =1 , we need to have the same expression in the denominator as
 0 
we have in the argument of the sine function    in this case, 2 /n .
1 2 1 2 sin (2 /n) 2 sin (2 /n) 2 2
lim A =lim nr sin ( 2 /n ) =lim nr  =lim  r . Let  = .
n
2 n
2 2 /n n n
2 /n n
n
n

sin (2 /n) 2 sin  2 2 2

Then as n
,  0 , so lim  r =lim  r = ( 1 )  r = r .
n
2 /n  0 

16
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

1.
R = 4
4 f x  x
( )
i=1 i
=0.5 f (0.5)+ f (1)+ f (1.5)+ f (2)
=0.5 1.75+1+( 0.25)+( 2)
=0.5(0.5)=0.25

The Riemann sum represents the sum of the areas of the two rectangles above the x  axis minus the
sum of the areas of the two rectangles below the x  axis; that is, the net area of the rectangles with
respect to the x  axis.

2. L =
6
6 i=1 ( )
f x
i 1
x=0.5 f (1)+ f (1.5)+ f (2)+ f (2.5)+ f (3)+ f (3.5)
 0.5( 1 0.5945349 0.3068528 0.0837093+0.0986123+0.2527630)=0.5( 1.6337217)  0.816861
The Riemann sum represents the sum of the areas of the two rectangles above the x  axis minus the
sum of the areas of the four rectangles below the x  axis; that is, the net area of the rectangles with
respect to the x  axis.

.
3.
M = 5
5 i=1 ( )
f x  x
i
=1 f (1.5)+ f (2.5)+ f (3.5)
+ f (4.5)+ f (5.5)
  0.856759

The Riemann sum represents the sum of the areas of the two rectangles above the x  axis minus the
sum of the areas of the three rectangles below the x  axis.

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

.
4. (a)
R = 6
6 f x  x
( )
i=1 i
=0.5 f (0.5)+ f (1)+ f (1.5)+ f (2)
+ f (2.5)+ f (3)
 5.353254

The Riemann sum represents the sum of the areas of the four rectangles above the x  axis minus the
sum of the areas of the two rectangles below the x  axis.

.
(b)
M = 6
6 i=1 ( )
f x  x
i
=0.5 f (0.25)+ f (0.75)+ f (1.25)+ f (1.75)
+ f (2.25)+ f (2.75)
 4.458461

The Riemann sum represents the sum of the areas of the four rectangles above the x  axis minus the
sum of the areas of the two rectangles below the x  axis.

2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

5.  x=(b a)/n=(8 0)/4=8/4=2 .

(a) Using the right endpoints to approximate 
8
f (x)dx , we have
0


4
f x  x=2[ f (2)+ f (4)+ f (6)+ f (8)] 2[1+2+( 2)+1]=4 .
( )
i=1 i

(b) Using the left endpoints to approximate 

8
f (x)dx , we have
0


4
i=1 ( )
f x
i 1
x=2[ f (0)+ f (2)+ f (4)+ f (6)] 2[2+1+2+( 2)]=6 .

(c) Using the midpoint of each subinterval to approximate 

8
f (x)dx , we have
0


4
i=1 ( )
f x  x=2[ f (1)+ f (3)+ f (5)+ f (7)] 2[3+2+1+( 1)]=10 .
i

6. (a) Using the right endpoints to approximate 

3
g(x)dx , we have
 3


6
g x  x = 1[g( 2)+g( 1)+g(0)+g(1)+g(2)+g(3)]
( )
i=1 i
 1 0.5 1.5 1.5 0.5+2.5= 0.5

(b) Using the left endpoints to approximate 

3
g(x)dx , we have
 3


6
i=1 ( )
g x
i 1
x = 1[g( 3)+g( 2)+g( 1)+g(0)+g(1)+g(2)]
 2+1 0.5 1.5 1.5 0.5= 1

(c) Using the midpoint of each subinterval to approximate 

3
g(x)dx , we have
 3


6
i=1 ( )
g x  x = 1[g( 2.5)+g( 1.5)+g( 0.5)+g(0.5)+g(1.5)+g(2.5)]
i
 1.5+0 1 1.75 1+0.5= 1.75


25
7. Since f is increasing, L  f (x)dx R .
5 0 5
3
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

Lower estimate = L =
5
5 i=1
f x ( )
i 1
x=5[ f (0)+ f (5)+ f (10)+ f (15)+ f (20)]
= 5( 42 37 25 6+15)=5( 95)= 475

Upper estimate = R =
5
f x  x=5[ f (5)+ f (10)+ f (15)+ f (20)+ f (25)]
( )
5 i=1 i
= 5( 37 25 6+15+36)=5( 17)= 85

8. (a) Using the right endpoints to approximate 

6
f (x)dx , we have
0


3
f x  x=2[ f (2)+ f (4)+ f (6)]=2(8.3+2.3 10.5)=0.2
( )
i=1 i

(b) Using the left endpoints to approximate 

6
f (x)dx , we have
0


3
i=1 ( )
f x
i 1
x=2[ f (0)+ f (2)+ f (4)]=2(9.3+8.3+2.3)=39.8

(c) Using the midpoint of each interval to approximate 

6
f (x)dx , we have
0


3
i=1 ( )
f x  x=2 f (1)+ f (3)+ f (5) =2(9.0+6.5 7.6)=15.8 .
i

The estimate using the right endpoints must be less than 

6 *
f (x)dx , since if we take x to be the right
0 i
endpoint x of each interval, then f x  ( ) f (x) for all x on x ,x , which implies that
i i i 1 i
x x
 f (x)dx , and so the sum    =
i 3 3 i 6
f x  x
( ) f x  x 
( ) f (x)dx f (x)dx .
i x i=1 i i=1 x 0
i 1 i 1
*
Similarly, if we take x to be the left endpoint x
i i 1
of each interval, then f x ( )
i 1
f (x) for all x on

, and so  
3 6
x ,x
i 1 i i=1
f x ( )i 1
x 
0
f (x)dx . We cannot say anything about the midpoint
estimate.

9.  x=(10 2)/4=2 , so the endpoints are 2 , 4 , 6 , 8 , and 10 , and the midpoints are 3 , 5 , 7 , and 9 .
The Midpoint Rule gives

10
2
3
x +1 dx 
4
i=1 ( )
f x  x=2
i
( 3 3
3 +1 + 5 +1 + 7 +1 + 9 +1  124.1644 .
3 3
)
 2 3 4 5 6 
10.  x=(  0)/6= , so the endpoints are 0 ,
, , , , and , and the ,
6 6 6 6 6 6 6
 3 5 7 9 11
midpoints are , , , , , and . The Midpoint Rule gives
12 12 12 12 12 12
   3 5 7 9 11
 0 sec ( x/3) dx  i=1 f xi  x= 6 sec 36 +sec 36 +sec 36 +sec 36 +sec 36 +sec 36  3.9379
6
( )
4
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

11.  x=(1 0)/5=0.2 , so the endpoints are 0, 0.2, 0.4, 0.6, 0.8, and 1, and the midpoints are 0.1, 0.3,
0.5, 0.7, and 0.9. The Midpoint Rule gives
 
1 2 5 2 2 2 2 2
0
sin (x )dx
i=1 ( )
f x  x=0.2 sin (0.1) +sin (0.3) +sin (0.5) +sin (0.7) +sin (0.9)  0.3084 .
i

12.  x=(5 1)/4=1 , so the endpoints are 1, 2, 3, 4, and 5, and the midpoints are 1.5, 2.5, 3.5, and 4.5.
The Midpoint Rule gives
5 2  x 2  1.5 2  2.5 2  3.5 2  4.5
 
4
1
x e dx
n=1 ( )
f x  x=1 (1.5) e
i
+(2.5) e +(3.5) e +(4.5) e  1.6099 .
13. In Maple, we use the command with(student) to load the sum and box commands, then
m:=middlesum(sin(x^),x=0..1,5); which gives us the sum in summation notation, then M:=evalf(m);
which gives M  0.30843908 , confirming the result of Exercise 11. The command
5
middlebox(sin(x^),x=0..1,5) generates the graph. Repeating for n=10 and n=20 gives
M  0.30981629 and M  0.31015563 .
10 20

14. See the solution to Exercise 5.1.7 for a possible algorithm to calculate the sums. With
 x=(1 0)/100=0.01 and subinterval endpoints 1, 1.01, 1.02, ... , 1.99, 2, we calculate that the left
=
100
Riemann sum is L sin (x ) x 0.30607 , and the right Riemann sum is
100 i=1 i 1

=
100 2
R sin x  x 0.31448 .
100 i=1 i

Since f (x)=sin x ( 2) is an increasing function, we must have L100  10sin ( x2) dx R100 , so
 sin ( x ) dx R <0.315 . Therefore, the approximate value 0.3084 0.31 in Exercise
1 2
0.306<L 
100 0 100
11 must be accurate to two decimal places.


15. We’ll create the table of values to approximate  sin xdx by using the program in the solution to
0
Exercise 5.1.7 with Y =sin x , Xmin=0 , X = , and n=5 , 10 , 50 , and 100 .
1
The values of R appear to be approaching 2 .
n

n R
n

5
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

5 1.933766
10 1.983524
50 1.999342
100 1.999836

2
2  x
16.  e dx with n=5 , 10 , 50 , and 100 .
0

n L R
n n
5 1.077467 0.684794
10 0.980007 0.783670
50 0.901705 0.862438
100 0.891896 0.872262
2
 x
The value of the integral lies between 0.872 and 0.892 . Note that f (x)=e is decreasing on ( 0,2 ) .
2
 x
We cannot make a similar statement for 
2
e dx since f is increasing on (  1,0 ) .
 1


 x sin x  x=
n
17. On 0, , lim xsin xdx .
i=1 i i 0
n

x x
i
e e
  x=
n 5
18. On 1,5 , lim dx .
n

i=1 1+x 1 1+x

i

* 2
  x=
n * 8 2
19. On 1,8 , lim 2x + x 2x+x dx .
i=1 i i 1
n

  x= (4 3x +6x )dx .

n 2 2 5
20. On [0,2] , lim
i=1 0
n

5 (  1 ) 6 6i
21. Note that  x= = and x = 1+i  x= 1+ .
n n i n
6i 6
 
n n

5
(1+3x)dx = lim f x  x=lim
( ) 1+3  1+
 1 n

i=1 i
n

i=1 n n
6 18i 6 18i
= lim n   ( 2)+
n n n
 2+ =lim
n

i=1 n n
n i=1 i=1 n

6
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

6 18 6 18 n(n+1)

n
= lim n  2n+ i =lim  2n+ 
n
n i=1
n
n n 2
108 n(n+1) n+1
= lim  12+  =lim  12+54
n
2 2 n
n
n
1
= lim  12+54 1+ = 12+54 1=42
n
n

22.

n

4 2
(x +2x 5)dx =lim f (x ) x [  x=3/n and x =1+3i/n ]
1 i=1 i i
n

3i 2 3i 3

n
=lim 1+ +2 1+  5
n

i=1 n n n
2
3 6i 6i
 i=1 1+ n + 9i2 +2+ n  5
n
=lim
n
n n
3 2 12
 i=1 92  i + n  i 2
n
=lim
n
n
n
3 9 n 2 12 n
2 i=1i + n  i=1i  i=12
n
=lim
n
n
n
27 n(n+1)(2n+1) 36 n(n+1) 6
=lim  +    n
n
3 6 2 2 n
n n
9 n+1 2n+1 n+1
=lim   +18  6
n
2 n n n
9 1 1 1 9
=lim 1+ 2+ +18 1+  6 =  1 2+18 1 6=21
n
2 n n n 2

2 0 2 2i
23. Note that  x= = and x =0+i x= .
n n i n
2
4i 2

2
0
( 2 x ) dx =lim 
2 n
i=1
f x  x=lim
( ) i

n
i=1
2
2 n
n
n
n
2 4 2 4
  
n n 2 n 2
=lim 2 i =lim 2n i
n
n i=1 2 i=1
n
n 2 i=1
n n

7
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

8 n(n+1)(2n+1) 4 n+1 2n+1

=lim 4  =lim 4  
n
3 6 n
3 n n
n
4 1 1 4 4
=lim 4 1+ 2+ =4  1 2=
n
3 n n 3 3

24.

1+2x ) dx =lim 
n

5
0
( 3
n

i=1
f x  x
( ) i

3
125i 5 5 250
  
n n n 3
=lim 1+2 =lim 1+ i
n

i=1 3 n n
n i=1 3 i=1
n n
2 2
5 250 1250 n (n+1)

n 3
=lim 1 n+ i =lim 5+ 
n
n 4
3 i=1 4
n n
n
2 2
( n+1 ) 1
=lim 5+312.5 =lim 5+312.5 1+
n
2 n
n
n
=5+312.5=317.5
2 1 1
25. Note that  x= = and x =1+i x=1+i ( 1/n ) =1+i/n .
n n i

i 3 1 1 n+i 3
 1x dx = lim  i=1 f xi  x=lim  i=1 1+ n 
2 3 n n n

n
n

( )n
=lim
n
n
i=1 n

= lim
1
4

n
i=1
( n +3n i+3ni +i ) =lim
3 2 2 3 1
4

n
i=1
n +
3
i=1
n
3n i+
2 n
i=1
3ni +
2 n
i=1
i
3

n
n n
n
1
 i+3n i +
3 2 n n 2 n 3
= lim 4
n n +3n
i=1 i=1 i=1
i
n
n
2 2
n(n+1) 3 n(n+1)(2n+1) 1 n (n+1)
3
= lim 1+  +  + 
n
2 2 3 6 4 4
n n n
2
3 n+1 1 n+1 2n+1 1 ( n+1 )
= lim 1+  +   + 
n
2 n 2 n n 4 2
n
3 1 1 1 1 1 1 2 3 1 1
= lim 1+ 1+ + 1+ 2+ + 1+ =1+ +  2+ =3.75
n
2 n 2 n n 4 n 2 2 4

8
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

=
.
26. (a)  x=(4 0)/8=0.5 and x =x =0.5i .

( x  3x) dx  i=1 f xi  x

4
0

1
2
2

=0.5 { 0.5  3(0.5) + 1.0  3(1.0) +

2

2
2

+ 3.5  3(3.5) + 4.0  3(4.0) }


5
4
 2
9
4
8

 2
5
4
7
*

+0+ +4 = 1.5
4
*
i i

(b)

4i 2 4i 4
(c) 
4
0
( x  3x ) dx=lim 
2 n
i=1 n
 3
n n
n

4 16 12
 
n 2 n
=lim i  i
n
n 2 i=1 n i=1
n
64 n(n+1)(2n+1) 48 n(n+1)
=lim   
n
3 6 2 2
n n
32 1 1 1
=lim 1+ 2+  24 1+
n
3 n n n
32 8
=  2 24=
3 3

(d) 
4
0
( x  3x ) dx=A 
2
1
A , where A is the area marked + and A is the area marked  .
2 1 2

27.

9
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

2
b a b a a(b a) (b a)
  
n n n

b
xdx =lim a+ i =lim 1+ i
a n
n i=1 n n
n i=1 2 i=1
n
2 2
a(b a) (b a) n(n+1) ( b a ) 1
=lim n+  =a ( b a ) +lim 1+
n
n 2 2 n
2 n
n
1 1 1 1 1 2 2
2
2
=a(b a)+ (b a) =(b a) a+ b
2 2
a =(b a) (b+a)=
2 2
b a ( )
28.
2 2
b a b a b a b a ( b a ) 2
 
n n 2

b 2
x dx =lim a+ i =lim a +2a i+ i
a n
n i=1 n n
n i=1 n 2
n
3 2 2
( b a ) 2a ( b a ) a ( b a )
  
n 2 n n
=lim i+ i+ 1
n
3 i=1 2 i=1 n i=1
n n
3 2 2
( b a ) n ( n+1 ) ( 2n+1 ) 2a ( b a ) n ( n+1 ) a ( b a )
=lim + + n
n
3 6 2 2 n
n n
3
( b a ) 1 1 2 1 2
=lim  1 1+ 2+ +a ( b a )  1 1+ +a ( b a )
n
6 n n n
3 3 2 2 3
( b a ) b  3ab +3a b a
2 2 2 2 3 2 3
= +a ( b a ) +a ( b a ) = +ab  2a b+a +a b a
3 3
3 3 3 3
b a 2 2 2 2 b a
=   ab +a b+ab  a b=
3 3 3

x 6 2 4 * 4i
29. f (x)= , a=2 , b=6 , and  x= = . Using Equation 3, we get x =x =2+i  x=2+ , so
5 n n i i n
1+x
4i
2+
x n 4
 
6 n
dx=lim R =lim  .
2 5 n

n
n

i=1 4i 5 n
1+x 1+ 2+
n

10 1 9 9i
30.  x= = and x =1+i  x=1+ , so
n n i n

10
 Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

9i 9i 9
 
10 n
(x 4ln x)dx=lim R =lim 1+  4ln 1+  .
1
n

n
n

i=1 n n n

*
31.  x=(  0)/n= /n and x =x = i/n .
i i
  5 i  1 5 2 2
  
n n
0
sin 5xdx=lim
n

i=1 ( sin 5x
i ) n
=lim
n

i=1
sin
n n
= lim
n
n
cot
2n
=
5
=
5

*
32.  x=(10 2)/n=8/n and x =x =2+8i/n .
i i

8i 6 8 1 8i 6
 
n n

10 6
x dx =lim 2+ =8lim 2+
2 n

i=1 n n n
n
i=1 n

=8 lim
(
1 64 58,593n +164,052n +131,208n  27,776n +2048

6 5 4 2
)
n
n 5
21n
1,249,984 9,999,872
=8 =  1,428,553.1
7 7

33. (a) Think of 

2
f (x)dx as the area of a trapezoid with bases 1 and 3 and height 2 . The area of a
0
1 1
(b+B)h , so 
2
trapezoid is A= f (x)dx= (1+3)2=4 .
2 0 2
(b)
   
5 2 3 5
f (x)dx = f (x)dx + f (x)dx + f (x)dx
0 0 2 3
trapezoid rectangle triangle
1 1
= (1+3)2 + 3  1 +  2 3 =4+3+3=10
2 2


7
(c) f (x)dx is the negative of the area of the triangle with base 2 and height 3 .
5
1

7
f (x)dx=  2 3= 3 .
5 2

9
(d) f (x)dx is the negative of the area of a trapezoid with bases 3 and 2 and height 2 , so it equals
7
1 1
(3+2)2= 5 . Thus,  f (x)dx= f (x)dx+ f (x)dx+
9 5 7 9
 (B+b)h= f (x)dx=10+( 3)+( 5)=2 .
2 2 0 0 5 7

34. (a)
11
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

1

2
g(x)dx=  4 2=4 (area of a triangle)
0 2
1
(b)  g(x)dx=  (2) = 2 (negative of the area of a semicircle)
6 2
2 2
1 1
(c)  g(x)dx=  1 1= (area of a triangle)
7
6 2 2
1
 0g(x)dx= 0g(x)dx+ 2g(x)dx+ 6g(x)dx=4 2 + 2 =4.5 2
7 2 6 7

1
35. 
3
x 1 dx can be interpreted as the area of the triangle above the x  axis minus the area of
0 2
1 1 1 1 3
the triangle below the x  axis; that is, (1)  (2)(1)=  1= .
2 2 2 4 4

36. 
2 2 2
4 x dx can be interpreted as the area under the graph of f (x)= 4 x between x= 2 and
 2
1
x=2 . This is equal to half the area of the circle with radius 2 , so 
2 2 2
4 x dx=   2 =2 .
 2 2

37. 
0
 3
( 1+ 9 x
2
) dx can be interpreted as the area under the graph of f (x)=1+ 2
9 x between x= 3
and x=0 . This is equal to one quarter the area of the circle with radius 3 , plus the area of the
rectangle, so 
0
 3
( 2 1
4
2
)
1+ 9 x dx=   3 +1 3=3+  .
9
4

12
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

38. 
3
(3 2x)dx can be interpreted as the area of the triangle above the x  axis minus the area of the
 1
1 5 1 3 25 9
triangle below the x  axis; that is, (5) (3)=  =4 .
2 2 2 2 4 4

39. 
2
x dx can be interpreted as the sum of the areas of the two shaded triangles; that is,
 1
1 1 1 4 5
(1)(1)+ (2)(2)= + = .
2 2 2 2 2

40. 
10
x 5 dx can be interpreted as the sum of the areas of the two shaded triangles; that is,
0
1
2 (5)(5)=25 .
2

41.
 
4 9
t dt = t dt
9 4


9
= x dx
4
38
=
3

13
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

42.  x cos xdx=0 since the limits of integration are equal.

1 2
1

1
43. 
1
0
( 5 6x ) dx=
2 1
0
5dx 6 x dx=5(1 0) 6
1 2
0 3
=5 2=3

44.  (1 2ex 1) dx=2

3 3 x
1
e dx 
3
1
( 3
)
1dx=2 e  e  1( 3 1 ) =2e  2e 2
3

45.  e
3 x+2
1
dx= e  e dx=e
3 x
1
2

2 3 x
1
e dx=e
2
( e3 e) =e5 e3
46.
 /2  /2  /2  /2  /2
 0
(2cos x 5x)dx =
0
2cos xdx  0
5xdx=2
0
cos xdx 5
0
xdx
2 2 2
( /2)  0 5
=2(1) 5 =2
2 8

47.
 1  2
 f (x)dx+  f (x)dx = f (x)dx+
2 5 5
f (x)dx f (x)dx
 2 2  2  2  1

=
5
f (x)dx
 1

48.  f (x)dx= 
4 5 5
f (x)dx f (x)dx=12 3.6=8.4
1 1 4

49.  [2 f (x)+3g(x)]dx=2 f (x)dx+3 g(x)dx=2(37)+3(16)=122

9 9 9
0 0 0

50. If f (x)= {
3 for x<3
x for x 3
, then  f (x)dx can be interpreted as the area of the shaded region,
5
0
which consists of a 5 by 3 rectangle surmounted by an isosceles right triangle whose legs have
1
length 2. Thus,  f (x)dx=5(3)+ (2)(2)=17 .
5
0 2

14
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

   /4  /4
. Hence,  
3 2 3 2
51. 0  sin x < 1 on 0, , so sin x  sin x on 0, sin xdx sin xdx
4 4 0 0
(Property 7).

x+1 and  
2 2
52. 5 x 3 x+1 on 1,2 , so 5 x 5 x dx x+1 dx .
1 1

2 2 2
53. If  1 x 1 , then 0 x  1 and 1 1+x  2 , so 1 1+x  2 and
 
1 2 1 2
1 1 ( 1)  1+x dx 2 1 ( 1) ; that is, 2 1+x dx 2 2 .
 1  1

1   1    /2  
54.
2
 sin x  1 for
6
 x
2
, so
2 2

6
   /6
sin xdx 1
2

6
; that is,
  /2 
6
   /6
sin xdx
3
.

1 1 1 1 1 1
 
2 2
55. If 1 x 2 , then   1 , so (2 1) dx 1(2 1) or  dx 1 .
2 x 2 1 x 2 1 x

3 3 3
56. If 0 x 2 , then 0 x  8 , so 1 x +1 9 and 1 x +1  3 . Thus,
 
2 3 2 3
1(2 0) x +1 dx 3(2 0) that is, 2 x +1 dx 6 .
0 0

     /3  
57. If
4
 x
3
, then 1  tan x 3 , so 1
3

4
   /4
tan xdx 3
3

4
or
  /3 
12
   /4
tan xdx
12
3 .

3 / 2
58. Let f (x)=x  3x+3 for 0 x 2 . Then f (x)=3x  3=3(x+1)(x 1) , so f is decreasing on ( 0,1 ) and
increasing on ( 1,2 ) . f has the absolute minimum value f (1)=1 . Since f (0)=3 and f (2)=5 , the
3
absolute maximum value of f is f (2)=5 . Thus, 1 x  3x+3 5 for x in 0,2 . It follows from
Property 8 that 1 (2 0)  (0 x3 3x+3) dx
2
5 (2 0) ; that is, 2  (0 x3 3x+3) dx
2
10 .

 x  1
59. The only critical number of f (x)=xe on 0,2 is x=1 . Since f (0)=0 , f (1)=e  0.368 , and
 2
f (2)=2e  0.271 , we know that the absolute minimum value of f on 0,2 is 0 , and the absolute
 1  x  1  x  1

2
maximum is e . By Property 8 ,0 xe  e for 0 x 2 0(2 0) xe dx e (2 0)
0

15
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

 x

2
0 xe dx 2/e .
0

1 3 2 1 2
60. If   x  , then  sin x 1 and  sin x 1 , so
4 4 2 2
1 3 1 3 1 1 1
 
3 /4 2 3 /4 2
    sin xdx 1    ; that is,   sin xdx  .
2 4 4  /4 4 4 4  /4 2

1 3 3 26
61.
4
x +1  x =x , so 
4 2 3
1
4
x +1 dx 
3 2
1
x dx=
3
(
3 1 =
3
. )
2
  /2  /2 1  2 
 
2
62. 0 sin x 1 for 0 x , so xsin x x xsin xdx xdx=  0 = .
2 0 0 2 2 8

63. Using a regular partition and right endpoints as in the proof of Property 2, we calculate
  cf x  x =lim c  f x  x =c
b n n n b
cf (x)dx=lim ( ) f x  x =clim
( ) ( ) f (x)dx .
a i=1 i i i=1 i i i=1 i i a
n
n
n

64. As in the proof of Property 2, we write  

b n
f (x)dx= f x  x . Now f x  0 and  x 0
( ) ( )
a n
i=1 i i

, so f x  x 0 and therefore 
n
( ) f x  x 0 . But the limit of nonnegative quantities is
( )
i i=1 i

nonnegative by Theorem 2.3.2, so 

b
f (x)dx 0 .
a

65. Since  f (x)  f (x) f (x) , it follows from Property 7 that

    
b b b b b
 f (x) dx f (x)dx f (x) dx f (x)dx  f (x) dx
a a a a a

Note that the definite integral is a real number, and so the following property applies:  a b a
b  a for all real numbers b and nonnegative numbers a .

  f (x)sin 2x dx= 
2 2 2 2
66. f (x)sin 2xdx  f (x) sin 2x dx f (x) dx by Property 7 , since
0 0 0 0
sin 2x  1 f (x) sin 2x  f (x) .

4 4
i 1 i
  = x dx
n n 1 4
67. lim =lim
n

i=1 5 n
n i=1 n 0
n

68.
16
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.2 The Definite Integral

1 1 dx
 =
n 1
lim
n
n i=1 2 0 2
1+(i/n) 1+x

i * i 1 i
69. Choose x =1+ and x = x x = 1+ 1+ . Then
i n i i 1 i n n
1 n 1

2  2 lim 
x dx = n
n i=1 i 1 i
1 1+ 1+
n n
1
= lim n i=1 (n+i 1)(n+i)
n

1 1
= lim n
n
 [by the hint]
n

i=1 n+i 1 n+i

1 1
 
n 1 n
= lim n 
n

i=0 n+i i=1 n+i

1 1 1 1 1 1
= lim n + +   +  +   + +
n
n n+1 2n 1 n+1 2n 1 2n
1 1 1 1
= lim n  =lim 1 =
n
n 2n n
2 2

17
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

1. The precise version of this statement is given by the Fundamental Theorem of Calculus. See the
statement of this theorem and the paragraph that follows it at the end of Section 5.3.

2. (a)
g(x) = f (t)dt , so g(0)=
x 0
f (t)dt=0.
0 0
1 1
g(1) =
1
f (t)dt=  1 1 [area of triangle] = .
0 2 2
g(2) = f (t)dt= f (t)dt+
2 1 2
f (t)dt [below the x  axis]
0 0 1
1 1
=   1 1=0.
2 2
1 1
g(3) =g(2)+
3
 1 1= .
f (t)dt=0
2 2 2
1 1
g(4) =g(3)+ f (t)dt=
4
+  1 1=0.
3 2 2
g(5) =g(4)+
5
f (t)dt=0+1.5=1.5.
4

g(6) =g(5)+
6
f (t)dt=1.5+2.5=4.
5

(b) g(7)=g(6)+
7
f (t)dt 4+2.2 [estimate from the graph] =6.2 .
6
(c) The answers from part (a) and part (b) indicate that g has a minimum at x=3 and a maximum at
x=7 . This makes sense from the graph of f since we are subtracting area on 1<x<3 and adding area
on 3<x<7 .

(d)

3. (a)
g(x) =
x
f (t)dt .
0

g(0) =
0
f (t)dt=0
0

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

g(1) =
1
f (t)dt=1 2=2 ,
0

g(2) = f (t)dt= f (t)dt+ f (t)dt=g(1)+

2 1 2 2
f (t)dt
0 0 1 1
1
=2+1 2+  1 2=5 ,
2
1
g(3) = f (t)dt=g(2)+
3 3
f (t)dt=5+  1 4=7 ,
0 2 2
g(6) =g(3)+
6
f (t)dt
3
1
=7+   2 2+1 2 =7 4=3
2

(b)

(c) g is increasing on ( 0,3) because as x increases from 0 to 3 , we keep adding more area.
(d) g has a maximum value when we start subtracting area; that is, at x=3 .

 3
4. (a) g( 3)= f (t)dt=0 , g(3)= f (t)dt= f (t)dt+
3 0 3
f (t)dt=0 by symmetry, since the area above
 3  3  3 0
the x axis is the same as the area below the axis.
1  2  1 1
(b) From the graph, it appears that to the nearest , g( 2)= f (t)dt 1 , g( 1)= f (t)dt 3 ,
2  3  3 2
1
and g(0)=
0
f (t)dt 5
.
 3 2
(c) g is increasing on (  3,0 ) because as x increases from  3 to 0 , we keep adding more area.
(d) g has a maximum value when we start subtracting area; that is, at x=0 .

(e)

/
(f) The graph of g (x) is the same as that of f (x) , as indicated by FTC1.

2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

5.

(a) By FTC1 with f (t)=t and a=1 , g(x)= t dt

2 x 2 / 2
g (x)= f (x)=x .
1

1 3 x 1 3 1
(b) Using FTC2, g(x)= t dt=
x 2 / 2
t = x  g (x)=x .
1 3 1 3 3

6.

(a) By FTC1 with f (t)=1+ t and a=0 , g(x)=

x /
0
( 1+ t ) dt  g (x)= f (x)=1+ x .

2 3/2 x 2 3/2
(b) Using FTC2, g(x)=
x / 1/2
0
( 1+ t ) dt= t+
3
t =x+
0 3
x  g (x)=1+x =1+ x .

7. f (t)= 1+2t and g(x)=

x /
1+2t dt , so by FTC1, g (x)= f (x)= 1+2x .
0

8. f (t)=ln t and g(x)= ln t dt , so by FTC1, g (x)= f (x)=ln x .

x /
1

9. f (t)=t sin t and g(y)= t sin t dt , so by FTC1, g (y)= f (y)=y sin y .

2 y2 / 2
2

1 1 1
and g(u)=
u /
10. f (x)= dx , so g (u)= f (u)= .
2 3 2 2
x+x x+x u+u

11. F(x)= cos t dt=

2
x
( 2) 
x
2
cos t dt ( 2) /
F (x)= cos x ( 2)
and F(x)= 
10 x /
12. f ( )=tan  tan  d = tan  d , so by FTC1, F (x)= f (x)= tan x .
x 10

3
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

1 du 1 dh dh du
13. Let u= . Then = . Also, = , so
x dx 2 dx du dx
x
d 1/x d u du du arctan ( 1/x )
h (x)=  arctant dt= 
/
arctant dt =arctanu = .
dx 2 du 2 dx dx 2
x

2 du dh dh du
14. Let u=x . Then =2x . Also, = , so
dx dx du dx
2
d d u du
h (x)=  
/ x 3 3 3 23 6
1+r dr= 1+r dr = 1+u (2x)=2x 1+(x ) =2x 1+x .
dx 0 du 0 dx

du 1 dy dy du
15. Let u= x . Then = . Also, = , so
dx 2 x dx du dx
d x cos t d cos t du cos u 1 cos x 1 cos x
 
/ u
y = dt= dt =  =  = .
dx 3 t du 3 t dx u 2 x x 2 x 2x

du dy dy du
16. Let u=cos x . Then = sin x . Also, = , so
dx dx du dx
d cos x d u du
y = dx  1 (t+sin t)dt= du  1(t+sin t)dt dx
/

= (u+sin u) ( sin x)= sin x cos x + sin ( cos x )

dw dy dy dw
17. Let w=1 3x . Then = 3 . Also, = , so
dx dx dw dx
3 3
d u d u dw
y = dx  
/ 1 1
du= du
1 3x 2 dw w 2 dx
1+u 1+u
3 3 3
d u dw w 3(1 3x)

w
=  du = ( 3)=
dw 1 2 dx 2 2
1+u 1+w 1+(1 3x)

du x x dy dy du
18. Let u=e . Then =e . Also, = , so
dx dx du dx
/ d 0 d 0 3 du d u 3 du
y =  xsin t dt=
dx e
3
du u
 sin t dt
dx
= 
du 0
sin t dt
dx
3 x x 3 x
= sin u e = e sin e . ( )
6 3 6 6
x 3 ( 1) 729 1 364
19. 
3 5
x dx= =  = =
 1 6  1 6 6 6 3

4
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

20. 
5 5
6dx= 6x =6[5 ( 2)]=6(7)=42
 2  2

4 2 8
21.  (4x+3)dx=
8
2 2
x +3x
2
( 2
)(
= 2 8 +3 8  2 2 +3 2 =152 14=138
2
)
3 2 1 3 4 3 1 20
22. 
4
0
( 1+3y y ) dy=
2
y+
2
y
3
y
0
= 4+  16  64  ( 0 ) =
2 3 3

5 9/5 1 5 5
23.  x
1 4/5
dx= x =  0=
0 9 0 9 9

3 4/3 8 3 4/3 4/3 3 4 3 3 45

24.  x dx= x
83 8 1/3
dx= x = (8  1 )= (2  1)= (16 1)= (15)=
1 1 4 1 4 4 4 4 4

 3 2
3 2 4 t 3 1 2 1 7
25.  dt=3 t dt=3
2
= = 1  1 =
1 4 1  3 1  3 3 8 8
t t 1

 5  5
26. 
3
x dx does not exist because the function f (x)=x has an infinite discontinuity at x=0 ; that is,
 2
f is discontinuous on the interval  2,3 .

2 2
27. 
5
dx does not exist because the function f (x)= has an infinite discontinuity at x=0 ; that
 5 3 3
x x
is, f is discontinuous on the interval  5,5 .

28. 
2 2
cos  d = sin  =sin 2  sin  =0 0=0
 

1 7 2 128 156
29.  x(2+x )dx= (2x+x )dx= x + x
2 5 2 6 2
= 4+  ( 0+0 ) =
0 0 7 0 7 7

1/2 4
1 4  x 1/2 4
30.  dx= x
4 1/2
dx= = 2x =2 4  2 1 =4 2=2
1 x 1 1/2 1 1

 /4  /4 
31. 
2
sec t dt= tan t =tan  tan 0=1 0=1
0 0 4

32.
5
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

2 5/2 1 2 17
 0 ( 3+x
1
x ) dx= (0 3+x3/2) dx=
1
3x+ x
5
=
0
3+
5
 0 =
5

33. 
2 2 2
csc  d does not exist because the function f ( )=csc  has infinite discontinuities at  =

and  =2 ; that is, f is discontinuous on the interval  ,2 .

 /6
34.  csc  cot  d does not exist because the function f ( )=csc  cot  has an infinite
0
discontinuity at  =0 ;

that is, f is discontinuous on the interval 0, .
6

1 1 1 1 1 1
35.  dx= 
9 9 9 1/2
dx= ln x = ( ln 9 ln 1 ) = ln 9 0=ln 9 =ln 3
1 2x 2 1 x 2 1 2 2

x 1
10 10 1 9
36.  10 dx=
1 x
=  =
0 ln 10 0 ln 10 ln 10 ln 10

37.
3 /2 6 1  1 3 /2  1 3  1 1

3 /2
dt =6 dt=6 sin t =6 sin  sin
1/2 2 1/2 2 1/2 2 2
1 t 1 t
  
=6  =6 =
3 6 6

4  1 1  1  1 
38. 
1
0 2
dt=4
1
0
1
2
dt=4 tan t =4 tan 1 tan 0 =4
0
( ) 4
 0 =
t +1 1+t

u+1 1
39. 
1 u+1 2 0 2 u+1 u 1
e du= e =e  e =e  1 [ or start with e =e e ]
 1  1

40.
2
 3  1 4 2 2  2 2

2
1
4+u
3
du =
2
1
( 4u +u ) du=
 2
u +ln u
1
=
2
+ln u
u u 1

1 3
=  +ln 2  ( 2+ln 1)= +ln 2
2 2

6
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

1 5 1 1 6 2 1 64 1
41.  f (x)dx= x dx+ x dx=
2 1 4 2 5
x + x =  0 +  =10.7
0 0 1 5 0 6 1 5 6 6

42.
2
 0 
 
f (x)dx = 0 
xdx+ sin xdx=
1 2
x

 cos x = 0  (cos   cos 0)
 0 2  0 2
2 2
 
=  ( 1 1)=2
2 2

43. From the graph, it appears that the area is about 60 . The actual area is
3 4/3 27 3 243 3
 0 x dx= 4 x 0 = 4  81 0= 4 =60.75 . This is 4 of the area of the viewing rectangle.
27 1/3

1
44. From the graph, it appears that the area is about . The actual area is
3
 3 6
6  4 x  1 6 1 1 215
 1
x dx=
 3
=
1 3
= + =
3 216 3 648
 0.3318 .
3x 1

2
45. It appears that the area under the graph is about of the area of the viewing rectangle, or about
3
2  
  2.1 . The actual area is  sin xdx=  cos x = (  cos  )  (  cos 0 ) = (  1 ) +1=2 .
3 0 0

7
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

46. Splitting up the region as shown, we estimate that the area under the graph is
 1   /3  /3
 1.8 . The actual area is 
2
+ 3 sec xdx= tan x = 3  0= 3 1.73 .
3 4 3 0 0

1 4 2 1 15
47. 
2 3
x dx= x =4 = =3.75
 1 4  1 4 4

2 2
48. 
5 /2 5 /2
sin xdx=  cos x =0+ =
 /4  /4 2 2

2 2 2 2 2
u 1 u 1 u 1 u 1 u 1
49. g(x)= du= du+  du+
3x 0 3x 2x 3x
2 2 2
du= 2 2
du
2x 2x 0 0 0
u +1 u +1 u +1 u +1 u +1
2 2 2 2
/ (2x)  1
d (3x)  1 d 4x  1 9x  1
g (x)= (2x)+   (3x)= 2 +3
2 dx 2 dx 2 2
(2x) +1 (3x) +1 4x +1 9x +1
2 2 2
1 dt dt dt dt
50. g(x)= dt= +  +
x 1 x tan x x
= 
tan x 4 tan x 4 1 4 1 4 1 4
2+t 2+t 2+t 2+t 2+t
2
 1 d d sec x
/
g (x)=
4 dx
(tan x)+
1
dx
2
x = ( ) 4
+
2x
8 8
2+tan x 2+x 2+tan x 2+x
8
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

3 3 3
x
51. y= t sin t dt= t sin t dt+  t sin t dt+
x 1 x x
t sin t dt= t sin t dt
x x 1 1 1
4
d d x sin x
y =
/ 4
x ( sin x ) 
dx
( 3/2
x ) +x sin x 
dx
3
x = ( )
3
( ) 2 x
3/2
+x sin x ( 3) ( 3x2)
sin x
7/2
=3x sin x  ( 3) 4
2 x

52. y=
cos x
( 2
)
5x 5x
0
cos ( u ) du 
cos u du=
2 cos x
0
cos ( u ) du
2

d d
y =cos ( 25x )  dx (5x) cos ( cos x )  dx (cos x)=cos ( 25x )  5 cos ( cos x )  (  sin x )
/ 2 2 2 2

=5cos ( 25x ) +sin xcos ( cos x )

2 2

2 4 2 4
1+u 1+u
53. F(x)= F (x)= f (x)= since f (t)=
x / x t
f (t)dt du du 
1 1 u 1 u
24 8 8
1+(x ) d 1+x 2 1+x
F
/ /
(x)= f (x)=
/
2

dx
2
x = ( ) 2
 2x=
x
. So F
/ /
(2)= 1+2 = 257 .
8

x x

1 1
>0 . y=
/ / x /
54. For the curve to be concave upward, we must have y dt  y = 
0 2 2
1+t+t 1+x+x
 ( 1+2x ) 2 2
y
/ /
=
2 2
. For this expression to be positive, we must have ( 1+2x ) <0 , since 1+x+x ( ) >0
( 1+x+x )
1 1
for all x . ( 1+2x ) <0 x< . Thus, the curve is concave upward on  , .
2 2
55. By FTC2, 
4 /
f (x)dx= f (4) f (1) , so 17= f (4) 12 f (4)=17+12=29 .
1

2 x  t
2
x  t
2

56. (a) erf(x)=  e dt   e dt= erf(x). By Property 5 of definite integrals in
 0 0 2
2 2 2
b  t a  t b  t
Section 5.2,  e dt= e dt+ e dt , so
0 0 a
b  t
2
b  t
2
a  t
2
  1
 a
e dt=
0
e dt  0
e dt=
2
erf(b)
2
erf(a)=
2
 erf(b) erf(a) .
2
x
(b) y=e erf(x)
9
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

2 2 2 2
/ x x / x 2  x
y =2xe erf(x)+e erf (x)=2xy+e  e [ by FTC1] =2xy+2/  .



57. (a) The Fresnel function S(x)= sin (
x 2
t )dt has local maximum values where
0 2
/  2 /
0=S (x)=sin ( x ) and S changes from positive to negative. For x>0 , this happens when
2
 2 2 /
x =(2n 1)  x =2(2n 1) x= 4n 2 , n any positive integer. For x<0 , S changes from positive
2
 2 2 /
to negative where x =2n  x =4n  x= 2 n . S does not change sign at x=0 .
2
/ / /
(b) S is concave upward on those intervals where S (x)>0 . Differentiating our expression for S (x)
/ /  2   2 / /  2
, we get S (x)=cos ( x ) 2 x = xcos ( x ) . For x>0 , S (x)>0 where cos ( x )>0
2 2 2 2
 2  1  2 1
0< x < or 2n  < x < 2n+  , n any integer  0<x<1 or 4n 1 <x< 4n+1 , n
2 2 2 2 2
/ /  2 3  2 1
any positive integer. For x<0 , S (x)>0 where cos ( x )<0 2n  < x < 2n  ,n
2 2 2 2
2
any integer  4n 3<x <4n 1 4n 3 < x < 4n 1  4n 3 < x< 4n 1   4n 3 >x> 4n 1 , so
the intervals of upward concavity for x<0 are (  4n 1 , 4n 3 ) , n any positive integer. To
summarize: S is concave upward on the intervals ( 0,1 ) , (  3, 1 ) , ( 3, 5 ) , (  7, 5 ) , ( 7,3)
, ... .
(c) In Maple, we use plot({int(sin(Pi*t^2/2),t=0..x),0.2},x=0..2); . Note that Maple recognizes the
Fresnel function, calling it FresnelS(x) . In Mathematica, we use
Plot[{Integrate[Sin[Pi*t^2/2],{t,0,x}],0.2},{x,0,2}] . In Derive, we load the utility file FRESNEL and
 2
plot FRESNEL_SIN(x) . From the graphs, we see that  sin ( t )dt=0.2 at x 0.74 .
x
0 2

58. (a) In Maple, we should start by setting si:=int(sin(t)/t,t=0..x); . In Mathematica, the command is
10
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

si=Integrate[Sin[t]/t,{t,0,x}] . Note that both systems recognize this function; Maple calls it Si(x) and
Mathematica calls it SinIntegral[x] . In Maple, the command to generate the graph is
plot(si,x= 4*Pi..4*Pi); . In Mathematica, it is plot[si,{x, 4*Pi.,4*Pi}] . In Derive, we load the utility
file EXP_INT and plot SI(x) .

/
(b) SI(x) has local maximum values where Si (x) changes from positive to negative, passing through
d x sin t sin x
0 . From the Fundamental Theorem we know that Si (x)= 
/
dt= , so we must have
dx 0 t x
/
sin x=0 for a maximum, and for x>0 we must have x=(2n 1) , n any positive integer, for Si to be
changing from positive to negative at x . For x<0 , we must have x=2n , n any positive integer, for a
/
maximum, since the denominator of Si (x) is negative for x<0 . Thus, the local maxima occur at
x= , 2 ,3 , 4 ,5 , 6 ,... .
/ / cos x sin x
(c) To find the first inflection point, we solve Si (x)=  =0 . We can see from the graph
x 2
x
that the first inflection point lies somewhere between x=3 and x=5 . Using a root finder gives the
value x 4.4934 . To find the y coordinate of the inflection point, we evaluate Si(4.4934) 1.6556 .
So the coordinates of the first inflection point to the right of the origin are about ( 4.4934,1.6556 ) .
/ /
Alternatively, we could graph S (x) and estimate the first positive x value at which it changes sign.
(d) It seems from the graph that the function has horizontal asymptotes at y 1.5 , with

lim Si(x)  1.5 respectively. Using the limit command, we get lim Si(x)= . Since Si(x) is an
x
 x
2
 
odd function, lim Si(x)= . So Si(x) has the horizontal asymptotes y= .
x
 2 2
(e) We use the fsolve command in Maple (or FindRoot in Mathematica) to find that the solution is
x 1.1 . Or, as in Exercise (c), we graph y=Si(x) and y=1 on the same screen to see where they
intersect.

/ /
59. (a) By FTC1, g (x)= f (x) . So g (x)= f (x)=0 at x=1,3,5,7 , and 9 . g has local maxima at x=1 and
/
5 (since f =g changes from positive to negative there) and local minima at x=3 and 7 . There is no
local maximum or minimum at x=9 , since f is not defined for x>9 .
    
1 3 5 7 9
(b) We can see from the graph that f dt < f dt < f dt < f dt < f dt . So
0 1 3 5 7
11
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

 , g(5)=   , and g(9)=  

1 5 3 5 9 7 9
g(1)= f dt f dt=g(1) f dt + f dt f dt=g(5) f dt + f dt .
0 0 1 3 0 5 7
Thus, g(1)<g(5)<g(9) , and so the absolute maximum of g(x) occurs at x=9 .
/ / / / / /
(c) g is concave downward on those intervals where g <0 . But g (x)= f (x) , so g (x)= f (x) ,
1
which is negative on (approximately) ,2 , ( 4,6 ) and ( 8,9 ) . So g is concave downward on
2
these intervals.

(d)

/ /
60. (a) By FTC1, g (x)= f (x) . So g (x)= f (x)=0 at x=2 , 4 , 6 , 8 , and 10 . g has local maxima at x=2
/
and 6 (since f =g changes from positive to negative there) and local minima at x=4 and 8 . There is
no local maximum or minimum at x=10 , since f is not defined for x>10 .
    
2 4 6 8 10
(b) We can see from the graph that f dt > f dt > f dt > f dt > f dt . So
0 2 4 6 8

 , g(6)=  
2 6 4 6
g(2)= f dt f dt=g(2) f dt + f dt , and
0 0 2 4

g(10)=  
10 8 10
f dt=g(6) f dt + f dt . Thus, g(2)>g(6)>g(10) , and so the absolute maximum of
0 6 8
g(x) occurs at x=2 .
/ / / / / /
(c) g is concave downward on those intervals where g <0 . But g (x)= f (x) , so g (x)= f (x) ,
which is negative on ( 1,3) , ( 5,7) and ( 9,10 ) . So g is concave downward on these intervals.

(d)

3 4 1
1 0 i 3 x 1
i
  = x dx=
n n 1 3
61. lim =lim =
n

i=1 4 n
n i=1 n 0 4 0 4
n

3/2 1
1 1 2 n 1 0 i 2x 2 2
 =
n 1
62. lim + +   + =lim x dx= =  0=
n
n n n n n n 3 3 3
i=1 0 0
n

63. Suppose h<0 . Since f is continuous on x+h,x , the Extreme Value Theorem says that there are

12
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

numbers u and v in x+h,x such that f (u)=m and f (v)=M , where m and M are the absolute
minimum and maximum values of f on x+h,x . By Property 8 of integrals,
 
x x+h
m( h) f (t)dt M( h) ; that is, f (u)( h)  f (t)dt f (v)( h) . Since  h>0 , we can divide
x+h x
1 x+h g(x+h) g(x) 1 x+h
this inequality by  h : f (u)
h x
 f ( t ) dt f (v) . By Equation 2,
h
= 
h x
f (t)dt for
g(x+h) g(x)
h 0 , and hence f (u) f (v) , which is Equation 3 in the case where h<0 .
h

64.
d d
  f (t)dt+
h(x) a h(x)
f (t)dt = f (t)dt (where a is in the domain of f )
dx g(x) dx g(x) a
d d
 
g(x) h(x) / /
=  f (t)dt + f (t)dt = f ( g(x) ) g (x)+ f ( h(x) ) h (x)
dx a dx a
/ /
= f ( h(x) ) h ( x )  f ( g(x) ) g (x)

/ 3
65. (a) Let f (x)= x  f (x)=1/(2 x )>0 for x>0 f is increasing on ( 0, ) . If x 0 , then x  0 ,
3
so 1+x  1 and since f is increasing, this means that f 1+x  ( 3
) f (1)
3
1+x  1 for x 0 . Next let
2 / /
g(t)=t  t g (t)=2t 1 g (t)>0 when t 1 . Thus, g is increasing on ( 1, ) . And since g(1)=0 ,
3 3
g(t) 0 when t 1 . Now let t= 1+x , where x 0 . 1+x  1 (from above)  t 1 g(t) 0
( 1+x3)  3
1+x  0 for x 0 . Therefore, 1
3
1+x 1+x for x 0 .
3

(b) From part (a) and Property 7:  1dx

1
0

1
0
1+x dx
3
 (0 1+x3) dx
1

1 4 1 1
 
1 1 3 1 3
x 1+x dx x+ x  1 1+x dx 1+ =1.25 .
0 0 4 0 0 4

66. (a) If x<0 , then g(x)= f (t)dt=

x x
0dt=0 .
0 0
1 2 x 1 2
If 0 x 1 , then g(x)= f (t)dt= t dt=
x x
t = x .
0 0 2 0 2
If 1<x 2 , then
g(x) = f (t)dt= f (t)dt+
x 1 x
f (t)dt
0 0 1

1 2 1 2 x
=g(1)+ (2 t)dt=
x
(1) + 2t t
1 2 2 1

13
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

1 1 2 1 1 2
= + 2x x  2 =2x x 1.
2 2 2 2

If x>2 , then g(x)= f (t)dt=g(2)+ 0dt=1+0=1 . So

x x
0 2

{
0 if x<0
1 2
x if 0 x 1
g(x)= 2
1 2
2x x  1 if 1<x 2
2
1 if x>2

(b)

(c) f is not differentiable at its corners at x=0 , 1 , and 2 . f is differentiable on (  ,0 ) , ( 0,1 ) ,

( 1,2 ) and ( 2, ) . g is differentiable on (  , ) .

f (t) f (x) 1
67. Using FTC1, we differentiate both sides of 6+
x 3/2
dt=2 x to get =2  f (x)=x .
a 2 2 2 x
t x
f (t)
To find a , we substitute x=a in the original equation to obtain 6+
a
dt=2 a  6+0=2 a 
a 2
t
3= a  a=9 .

x b x a
68. B=3A 
b x
0
e dx=3 e dx
a x
0
e =3 e
0 0

b
e  1=3 e  1  ( a
)
b a
e =3e  2 (
b=ln 3e  2
a
)
69. (a) Let F(t)=
t /
f (s)ds . Then, by FTC1, F (t)= f (t)= rate of depreciation, so F(t) represents the
0
loss in value over the interval 0,t .

14
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

1 A+F(t)
A+ f (s)ds =
t
(b) C(t)= represents the average expenditure per unit of t during the
t 0 t
interval 0,t , assuming that there has been only one overhaul during that time period. The company
wants to minimize average expenditure.
1 1 1
A+ f (s)ds . Using FTC1, we have C (t)= A+ f (s)ds + f (t) .
t / t
(c) C(t)=
t 0 2 0 t
t
1
C (t)=0 t f (t)=A+ f (s)ds f (t)= A+ f (s)ds =C(t) .
/ t t
0 t 0

1

t
70. (a) C(t)= f (s)+g(s) ds . Using FTC1 and the Product Rule, we have
t 0
1 1 t 1 1 t
  f (s)+g(s) ds=0
/ /
C (t)= f (t)+g(t)  f (s)+g(s) ds . Set C (t)=0 : f (t)+g(t) 
t 2 0 t 2 0
t t
1 t
f (t)+g(t)   f (s)+g(s) ds=0 f (t)+g(t)  C(t)=0  C(t)= f (t)+g(t) .
t 0
V V V V 2 t V V 2
(b) For 0 t 30 , we have D(t)=
t
 s ds= s s = t t .
0 15 450 15 900 0 15 900
V V 2 2 2
So D(t)=V  t t =V  60t t =900 t  60t+900=0
15 900

2
( t 30 ) =0 t=30 . So the length of time T is 30 months.
(c)
1 V V V 1 V V 2 V t

t 2 3
C(t) =  s+ s ds= s s+ s
t 0 15 450 12,900 t 15 900 38,700 0
1 V V 2 V 3 V V V 2
= t t + t =  t+ t 
t 15 900 38,700 15 900 38,700

/ V V 1 1
C (t)= + t=0 when t=  t=21.5 .
900 19,350 19,350 900
V V V 2 V
C(21.5)=  (21.5)+ (21.5)  0.05472V , C(0)=  0.06667V , and
15 900 38,700 15
V V V 2
C(30)=  (30)+ (30)  0.05659V , so the absolute minimum is C(21.5) 0.05472V .
15 900 38,700
V V V 2
(d) As in part (c), we have C(t)=  t+ t , so C(t)= f (t)+g(t)
15 900 38,700
V V V 2 V V V 2 2 1 1 1 1
 t+ t =  t+ t  t  =t  
15 900 38,700 15 450 12,900 12,900 38,700 450 900
1/900 43
t= = =21.5 . This is the value of t that we obtained as the critical number of C in part (c),
2/38,700 2
15
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.3 The Fundamental Theorem of Calculus

so we have verified the result of (a) in this case.

16
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

1.
d
dx
2
x +1 +C =
d
dx
( x2+1) 1/2+C =
1 2
2
x +1 ( )  1/2 2x= x
2
x +1

d
2. xsin x+cos x+C =xcos x+ ( sin x )  1 sin x=xcos x
dx

3.
d
dx
x
+C =
1
2
2 2
a  x  x( x)
2
/
2
2
a x
2
=
1
2
( a2 x2) +x2 = 1
a
2
a x
2 2
a a x a ( a2 x2) 3/2 ( a2 x2) 3
4.

d

x +a
2 2
+C =
1 d x +a
2 2
=
(
x x/ x +a
2 2
) 2 2
x +a  1
dx 2 2 dx x 2 2
a x a a x

=
x
2
( x2+a2) =
1
2 2 2 2 2 2 2
a x x +a x x +a

 3/4+1 1/4
 3/4 x x
5.  x
1/4
dx= +C= +C=4x +C
 3/4+1 1/4

4/3
x 3 4/3
6.  x dx= x
3 1/3
dx= +C= x +C
4/3 4

4 2
x x 1 4 2
7.  ( 3
)
x +6x+1 dx= +6 +x+C= x +3x +x+C
4 2 4

2 6
x x 1 2 1 6
(
8.  x 1+2x dx=
4
) ( 5
)
x+2x dx= +2 +C= x + x +C
2 6 2 3

2 3 4
t t t 2 1 3 1 4
9.  (1 t) 2+t dt= ( 2
) ( 2
2 2t+t  t dt=2t 2 + 
2 3 4
3
)
+C=2t t + t 
3 4
t +C

3
1 x  1
10. 
2
x +1+ dx= +x+tan x+C
2 3
x +1

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

3/2 2
x x 8 3/2 1 2
11.  dx= ( 4 4 x +x ) dx=4x 4
2
( 2 x ) + +C=4x
3/2 2 3
x + x +C
2

12.  ( 3e u + sec2 u ) du = 3e u + tan u + C

sin x sin x 1 sin x
13.  dx= dx=  dx= sec x tan xdx=sec x+C
2 2 cos x cos x
1 sin x cos x

sin 2x 2sin xcos x

14.  dx= dx= 2cos xdx=2sin x+C
sin x sin x

2 5/2
15.  x x dx= x
3/2
x +C . dx=
5
The members of the family in the figure correspond to C=5 , 3 , 0 ,  2 , and  4 .

16.  (cos x 2sin x)dx=sin x+2cos x+C .

The members of the family in the figure correspond to C=5 , 3 , 0 ,  2 , and  4 .

1 3 1 2 2 2
17.  (6x  4x+5)dx= 6
2 2 3 2
x  4 x +5x = 2x  2x +5x =(16 8+10) 0=18
0 3 2 0 0

18.
1 2 1 4 3 3

3 3 2 4
1
(1+2x 4x )dx = x+2 x  4 x = x+x  x
2 4 1 1
=(3+9 81) (1+1 1)= 69 1= 70

x 0  1
19. 
0
 1
( 2x e ) dx=
x 2
x e
 1
(
= ( 0 1 )  1 e ) = 2+1/e
2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

1 6 1 4 1 3 0 32 8
20. 
0 5 3 2
(u  u +u )du= u u+ u =0  4 = 4
 2 6 4 3  2 3 3

21.
1 3 1 2 2 2

2 2
 2
(3u+1) du =
2
 2
( 9u +6u+1 ) du=
2
3
u +6 u +u
2
9
 2
= 3u +3u +u
3 2
 2
=(24+12+2) ( 24+12 2)=38 ( 14)=52

22.
1 3 1 2 4 13 2 4

4
(2v+5)(3v 1)dv = (6v +13v 5)dv= 6
4 2 3
0
v +13 v  5v = 2v + v  5v
0 3 2 0 2 0
=(128+104 20) 0=212

2 3/2 2 5/2 4 16 64 2 2 14 62 256

23.  t (1+t)dt= (t
4 4 1/2 3/2
+t )dt= t + t = +  + = + =
1 1 3 5 1 3 5 3 5 3 5 15

2 3/2 9 2
24.  2t dt=
9 9 1/2
2 t dt= 2 t = 2  27 0=18 2
0 0 3 0 3

 1 1 4 1   1  1 1 63
2 1
25. 
3 2 4
4y + dy= 4 y +2 y = y =(1 1) 16 =
 2 3 4  2  2 2 4 4
y y  2

7 2 2
y+5y 2  2  1 1 5 1 5 1 63
26.  dy= (y +5y )dy=  y +5
2 4
y =  +y =  +32  (  1+1 ) =
1 3 1 5 1 y 1 2 2
y

3 7/3 4 9/4 1 3 4 55
27.  x ( 3 x + 4 x ) dx=
1 1 4/3 5/4
(x +x )dx= x + x = +  0=
0 0 7 9 0 7 9 63

5
28. 
5
0
( 2e +4cos x ) dx= 2e +4sin x = ( 2e +4sin 5)  ( 2e +4sin 0 ) =2e +4sin 5 2
x x
0
5 0 5
290.99

4  1/2
29.  5/x dx= 5  x
4 4
dx= 5 2 x = 5 (2 2 2 1)=2 5
1 1 1

30.
3x 2 9

9
dx = (3x1/2 2x
9 1/2 2 3/2 1/2 3/2 1/2 9
1 )dx= 3 x  2 2x = 2x  4x
x 1 3 1 1

3
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

=(54 12) (2 4)=44

 
31.  (4sin   3cos  )d =  4cos   3sin  = ( 4 0 )  (  4 0 ) =8
0 0

 /3  /3  
32.  sec  tan  d = sec  =sec  sec =2 2
 /4  /4 3 4

33.
2 2
 /4 1+cos   /4 1 cos   /4
 0 2
d =
0 2
+ 2
d =
0
( sec 2 +1) d
cos  cos  cos 
 /4   
= tan  + = tan +  ( 0+0 ) =1+
0 4 4 4

34.
2 2 2
 /3 sin  +sin  tan   /3 sin  (1+tan  )  /3 sin  sec   /3
 0 2
d =
0 2
d =
0 2
d =
0
sin  d
sec  sec  sec 
 /3 1 1
=  cos  =  ( 1)=
0 2 2

35.
3 1/3
1 x 64  1/2 64  1/2  1/6
dx = dx= )dx=
1+ x 64 (1/3) (1/2)

64
+ (x +x (x +x )dx
1 1 1/2 1/2 1 1
x x x
1/2 6 5/6 64 192 6 186 256
= 2x + x = 16+  2+ =14+ =
5 1 5 5 5 5

3 5 1 7 1 3 1 96
36.  (1+x ) dx= (1+3x +3x +x )dx= x+x +
1 23 1 2 4 6 3
x+ x = 1+1+ +  0=
0 0 5 7 0 5 7 35

37.
2 e
x +x+1 1 1 2 1 2 1 1 2 1
 dx=
e e
x+1+ dx= x +x+ln x = e +e+ln e  +1+ln 1 = e +e
1 x 1 x 2 1 2 2 2 2

38.
1 2 1 1 2 9 81 85 9
 dx=
9 9
x+ x+2+ dx= x +2x+ln x = +18+ln 9 (8+8+ln 4)= +ln
4 x 4 x 2 4 2 2 4
4
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

39.
1 2 0 1 2 2

2
( x 2 x ) dx = [x 2( x)]dx+ [x 2(x)]dx= 3xdx+ ( x)dx=3
0 2 0 2

x  x
1  1 0  1 0 2  1 2 0
1 7
=3 0  (2 0)= = 3.5
2 2

40.
 
 sin x dx = sin xdx+
3 /2 3 /2 3 /2
0 0 
(  sin x ) dx=  cos x 0 + cos x 
= 1 (  1 ) + 0 (  1 ) =2+1=3

2 4
41. The graph shows that y=x+x  x has x intercepts at x=0 and at x=a 1.32 . So the area of the
region that lies under the curve and above the x axis is

( 1 5 a
x+x  x ) dx =
1 2 1 3

a 2 4
0
x+ x x
2 3 5 0
1 2 1 3 1 5
= a+ a a  0
2 3 5
 0.84

4 6
42. The graph shows that y=2x+3x  2x has x intercepts at x=0 and at x=a 1.37 . So the area of the
region that lies under the curve and above the x axis is

( 2x+3x4 2x6) dx = 3 5 2 7 a

a 2
0
x+x x
5 7 0
2 3 5 2 7
= a+ a a  0
5 7
 2.18

5
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

1 3 2 8 4
43. A=
2
0
( 2y y ) dy=
2 2
y
3
y = 4
0 3
 0=
3

1 5 1 1
x=y , so A= y dy=
4 4 1 4
44. y= x  y = .
0 5 0 5

/
45. If w (t) is the rate of change of weight in pounds per year, then w(t) represents the weight in
pounds of the child at age t . We know from the Net Change Theorem that 
10 /
w (t)dt=w(10) w(5) ,
5
so the integral represents the increase in the child’s weight (in pounds) between the ages of 5 and 10 .

46.  I(t)dt= Q (t)dt=Q(b) Q(a) by the Net Change Theorem, so it represents the change in the
b b /
a a
charge Q from time t=a to t=b .

/
47. Since r(t) is the rate at which oil leaks, we can write r(t)= V (t) , where V (t) is the volume of oil
at time t . Thus, by the Net Change Theorem,
 
120 120 /
r(t)dt= V (t)dt= V (120) V (0) =V (0) V (120) , which is the number of gallons of oil that
0 0
leaked from the tank in the first two hours (120 minutes).

48. By the Net Change Theorem, 

15 /
n (t)dt=n(15) n(0)=n(15) 100 represents the increase in the bee
0

population in 15 weeks. So 100+

15 /
n (t)dt=n(15) represents the total bee population after 15 weeks.
0

49. By the Net Change Theorem, 

5000 /
R (x)dx=R(5000) R(1000) , so it represents the increase in
1000
revenue when production is increased from 1000 units to 5000 units.

/
50. The slope of the trail is the rate of change of the elevation E , so f (x)=E (x) . By the Net Change
Theorem,  f (x)dx= E (x)dx=E(5) E(3) is the change in the elevation E between x=3 miles and
5 5 /
3 3
x=5 miles from the start of the trail.

6
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

51. In general, the unit of measurement for 

b
f (x)dx is the product of the unit for f (x) and the unit for
a
x . Since
f (x) is measured in newtons and x is measured in meters, the units for 
100
f (x)dx are newton meters.
0
(A newton meter is abbreviated N m and is called a joule.)

52. The units for a(x) are pounds per foot and the units for x are feet, so the units for da/dx are
pounds per foot per foot, denoted (lb / ft) / ft. The unit of measurement for  a(x)dx is the product of
8
2
pounds per foot and feet; that is, pounds.

3 2 3 27 3
53. (a) displacement = (3t 5)dt=
3
t  5t =  15= m
0 2 0 2 2
(b)
distance traveled = |3t 5|dt= (5 3t)dt+
3 5/3 3
(3t 5)dt
0 0 5/3

3 2 5/3 3 2 3 25 3 25 27 3 25 25 41
= 5t t + t  5t =   +  15   = m
2 0 2 5/3 3 2 9 2 2 9 3 6

1 3 2 6 1 10
54. (a) displacement =
6
1
( t  2t 8 ) dt=
2
3
t  t  8t
1
= ( 72 36 48 ) 
3
 1 8 =
3
m
(b)
distance traveled = |t  2t 8|dt= |(t 4)(t+2)|dt
6 2 6
1 1

1 3 2 4 1 3 2 6
=
4
1
(  t +2t+8 ) dt+ ( t  2t 8 ) dt=
2 6 2
4

3
t +t +8t
1
+
3
t  t  8t
4
64 1 64 98
=  +16+32   +1+8 + ( 72 36 48 )   16 32 = m
3 3 3 3

/ 1 2 1 2
55. (a) v (t)=a(t)=t+4 v(t)= t +4t+C  v(0)=C=5 v(t)= t +4t+5 m / s
2 2
(b)
1 2 1 2
distance traveled = |v(t)|dt= t +4t+5 dt=
10 10 10
t +4t+5 dt
0 0 2 0 2
1 3 2 10 500 2
= t +2t +5t = +200+50=416 m
6 0 3 3

/ 2 2
56. (a) v (t)=a(t)=2t+3 v(t)=t +3t+C  v(0)=C= 4 v(t)=t +3t 4
7
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

(b)
distance traveled = t +3t 4 dt=
3 2 3
(t+4)(t 1) dt
0 0

=
1
0
(  t  3t+4) dt+ ( t +3t 4) dt
2
1
3 2

1 3 3 2 1 1 3 3 2 3
=  t  t +4t + t + t  4t
3 2 0 3 2 1
1 3 27 1 3 89
=   +4 + 9+  12  +  4 = m
3 2 2 3 2 6

4 3/2 4 32 140 2
57. Since m (x)= (x) , m=  (x)dx= ( 9+2 x ) dx= 9x+ x
/ 4 4
=36+  0= =46 kg.
0 0 3 0 3 3 3

58. By the Net Change Theorem, the amount of water that flows from the tank is
2 10
 r(t)dt=
10 10
(200 4t)dt= 200t 2t = ( 2000 200 )  0=1800 liters.
0 0 0

59. Let s be the position of the car. We know from Equation 2 that s(100) s(0)=
100
v(t)dt . We use
0
the Midpoint Rule for 0 t 100 with n=5 . Note that the length of each of the five time intervals is
20 1
20 seconds = hour= hour. So the distance traveled is
3600 180
1
 0 v(t)dt  180 [v(10)+v(30)+v(50)+v(70)+v(90)]
100

1
= (38+58+51+53+47)
180
247
=  1.4miles
180

60. (a) By the Net Change Theorem, the total amount spewed into the atmosphere is
Q(6) Q(0)= r(t)dt=Q(6) since Q(0)=0 . The rate r(t) is positive, so Q is an increasing function.
6
0
b a 6 0
Thus, an upper estimate for Q(6) is R and a lower estimate for Q(6) is L .  t= = =1 .
6 6 n 6
R =
6
r t  t=10+24+36+46+54+60=230 tonnes.
()
6 i=1 i

L =
6
r (t )  t=R +r(0) r(6)=230+2 60=172 tonnes.
6 i=1 i 1 6
b a 6 0
(b)  t= = =2 . Q(6) M =2[r(1)+r(3)+r(5)]=2(10+36+54)=2(100)=200 tonnes.
n 3 3

8
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.4 Indefinite Integrals and the Net Change Theorem

61. From the Net Change Theorem, the increase in cost if the production level is raised from 2000
yards to 4000 yards is C(4000) C(2000)=
4000 /
C (x)dx .
2000


4000 /
C (x)dx =
2000

4000
2000
( 3 0.01x+0.000006x ) dx
2

2 3 4000
= 3x 0.005x +0.000002x =60,000 2,000=$58,000
2000

62. By the Net Change Theorem, the amount of water after four days is
25 , 000+ r(t)dt  25 , 000+M 4
4
0
4 0
=25 , 000+ r(0.5)+r(1.5)+r(2.5)+r(3.5)
4
 25 , 000+ 1500+1770+740+ (  690 ) =28 , 320 liters

63. (a) We can find the area between the Lorenz curve and the line y=x by subtracting the area under
y=L(x) from the area under y=x . Thus,

1
x L(x) dx
area between Lorenz curve and line y=x 0
coefficient of inequality = =
area under line y=x

1
xdx
0


1

1
x L(x) dx x L(x) dx
=2
0 0 1
= = x L(x) dx
2 1 1/2 0
x /2
0

5 2 7 1 5 7 19
(b) L(x)= x+ x L(50%)=L = + = =0.39583 , so the bottom 50% of the
12 12 2 48 24 48
households receive at most about 40% of the income. Using the result in part (a),
5 2 7
coefficientofinequality = 2 x L(x) dx=2
1 1
x x x dx
0 0 12 12
5 5 2 1 5
= 2
1
0 12
x
12
x dx=2
0 12
2
x x dx ( )
5 1 2 1 3 1 5 1 1 5 1 5
= x x =  = =
6 2 3 0 6 2 3 6 6 36

3 2
64. (a) From Exercise 4.1. (a), v(t)=0.00146t  0.11553t +24.98169t 21.26872 .
125
(b) h(125) h(0)=
125 4 3 2
v(t)dt= 0.000365t  0.03851t +12.490845t  21.26872t  206 , 407 ft
0 0
9
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

1
1. Let u=3x . Then du=3dx , so dx= du . Thus,
3
1 1 1 1
 cos 3xdx= cos u 3 du = 3  cos udu= 3 sin u+C= 3 sin 3x+C . Don’t forget that it is often very
easy to check an indefinite integration by differentiating your answer. In this case,
d 1 1
sin 3x+C = (cos 3x) 3=cos 3x , the desired result.
dx 3 3

2 1
2. Let u=4+x . Then du=2xdx and xdx= du , so
2
2 10 1 1 1 11 1
 (
x 4+x ) dx= u
10
2
du = 
2 11
u +C=
22
4+x
2
( ) 11+C .
3 2 2 1
3. Let u=x +1 . Then du=3x dx and x dx= du , so
3
3/2
1 1 u 1 2 3/2 2 3 3/2
 x +1 dx=
2 3
x u du = +C=  u +C= (x +1) +C .
3 3 3/2 3 3 9

1 1
4. Let u= x . Then du= dx and dx=2du , so
2 x x
sin x
 dx= sin u (2du)=2( cos u)+C= 2cos x +C .
x

1
5. Let u=1+2x . Then du=2dx and dx= du , so
2
 2
4  3 1 u 1 1
 3
dx=4 u
2
du =2
 2
+C=
2
+C=
2
+C .
(1+2x) u (1+2x)

sin  sin 
. Then du=cos  d , so  e cos  d = e du=e +C=e
u u
6. Let u=sin  +C .

1 5 1 2 5
7. Let u=x +3 . Then du=2xdx , so  2x(x +3) dx= u du=
2 2 4 4
u +C= (x +3) +C .
5 5

3 2 1 2
8. Let u=x +5 . Then du=3x dx and x dx= du , so
3
1 1 1 10 1 3 10
 x (x +5) dx= u
2 3 9 9
du =  u +C= (x +5) +C .
3 3 10 30

9. Let u=3x 2 . Then du=3dx and

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

1 1 1 1 21 1
du , so  (3x 2) dx= u
20 20 21
dx= du =  u +C= (3x 2) +C .
3 3 3 21 63

1 7 1
10. Let u=2 x . Then du= dx and dx= du , so  (2 x) dx= u ( du)=
6 6 7
u +C= (2 x) +C .
7 7

2
11. Let u=1+x+2x . Then du=(1+4x)dx , so
1/2
1+4x du  u
 dx= = u
1/2 2
du= +C=2 1+x+2x +C .
2 u 1/2
1+x+2x

2 1
12. Let u=x +1 . Then du=2xdx and xdx= du , so
2
x  2 1 1  1  1  1
 2 2
dx= u
2
du = 
2 u
+C=
2u
+C= 2
+C .
(x +1) 2(x +1)

1
13. Let u=5 3x . Then du= 3dx and dx= du , so
3
dx 1 1 1 1
 5 3x
=
u

3
du =
3
ln u +C=
3
ln 5 3x +C .

2 1
14. Let u=x +1 . Then du=2xdx and xdx= du , so
2
1
du
x 2 1 1 1
 2
dx=
u 2 2
2
2
2 2
= ln u +C= ln x +1 +C= ln x +1 +C [ since x +1>0 ] ( )
x +1
2 1/2 2
or ln (x +1) +C=ln x +1 +C .

1
15. Let u=2y+1 . Then du=2dy and dy= du , so
2
3  5 1 3 1 4  3
 5
dy= 3u
2
du = 
2  4
u +C=
4
+C .
(2y+1) 8(2y+1)

1
16. Let u=5t+4 . Then du=5dt and dt= du , so
5
1  2.7 1 1 1  1.7  1   2
 dt= u
1.7
du =  u +C= u +C= +C .
2.7 5 5  1.7 8.5 1.7
(5t+4) 17(5t+4)

2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

2 3/2 2
17. Let u=4 t . Then du= dt and dt= du , so  4 t dt= u
1/2 3/2
( du)= u +C= (4 t) +C .
3 3

4 3 1 3
18. Let u=2y  1 . Then du=8y dy and y dy= du , so
8
1 1 2 3/2 1
 2y  1 dy= u
3 4 1/2 4 3/2
y du =  u +C= (2y  1) +C .
8 8 3 12

1
19. Let u= t . Then du= dt and dt= du , so

1 1 1
 sin  t dt= sin u

du =

( cos u)+C=

cos  t+C .

1
20. Let u=2 . Then du=2d and d = du , so  sec 2 tan 2 d = sec u
2
1 1 1
tan u du = sec u+C= sec 2 +C .
2 2 2

2
dx ( ln x ) 1 3 1
, so  dx= u du= u +C= ( ln x ) +C .
2 3
21. Let u=ln x . Then du=
x x 3 3

 1
 1
22. Let u=tan x . Then du=
dx
, so 
 1
tan x u
dx= udu= +C=
tan x
2
( ) 2 +C .
2 2 2 2
1+x 1+x

dt 1 cos t
23. Let u= t . Then du= and dt=2du , so  dt= cos u (2du)=2sin u+C=2sin t +C
2 t t t
.

3/2 3 1/2 2
24. Let u=1+x . Then du= x dx and x dx= du , so
2 3
2 2 2
 x sin (1+x )dx= sin u
3/2 3/2
du =  ( cos u)+C= cos (1+x )+C .
3 3 3

1 7 1
. Then du=cos  d , so  cos  sin  d = u du=
6 6 7
25. Let u=sin  u +C= sin  +C .
7 7

1 6 1
, so  (1+tan  ) sec  d = u du=
2 5 2 5 6
26. Let u=1+tan  . Then du=sec  d u +C= (1+tan  ) +C .
6 6

3
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

2 3/2 2
27. Let u=1+e . Then du=e dx , so  e 1+e dx=
x x x x x 3/2
u du= u +C= (1+e ) +C .
3 3
x 2 x x
Or: Let u= 1+e . Then u =1+e and 2udu=e dx , so
2 3 2
 e 1+e dx= u 2udu= 3 u +C= 3 (1+e ) +C .
x x x 3/2

28. Let u=cos t . Then du= sin t dt and sin t dt= du , so  e sin t dt= e ( du)= e +C= e
cos t u u cos t
+C .

3 2 2 1
29. Let u=1+z . Then du=3z dz and z dz= du , so
3
2
z  1/3 1 1 3 2/3 1
 dz= u
3 2/3
du =  u +C= (1+z ) +C .
3 3 3 3 2 2
1+z

2 1
30. Let u=ax +2bx+c . Then du=2(ax+b)dx and (ax+b)dx= du , so
2
1
du
(ax+b)dx 2 1 
 = =  u
1/2 1/2 2
du=u +C= ax +2bx+c +C .
2 u 2
ax +2bx+c

dx dx du
31. Let u=ln x . Then du= , so  = =ln u +C=ln ln x +C .
x xln x u

x
e du
32. Let u=e +1 . Then du=e dx , so 
x x
x
dx=
u
x
(
=ln u +C=ln e +1 +C . )
e +1

2 2
33. Let u=cot x . Then du= csc xdx and csc xdx= du , so
3/2
u 2
 cot x csc xdx=
2 3/2
u ( du)= +C= (cot x) +C .
3/2 3

  1 1
34. Let u= . Then du= dx and dx= du , so
x 2 2 
x x
cos ( /x ) 1 1 1 
 2
dx= cos u 

du =

sin u+C=

sin +C .
x
x

35.
4
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

cos x 1
 cot xdx=
sin x
dx . Let u=sin x . Then du=cos xdx , so  cot xdx=
u
du=ln u +C=ln sin x +C .

36. Let u=cos x . Then du= sin xdx and sin xdx= du , so
sin x  du  1  1
 2
dx=  2
= tan u+C= tan (cos x)+C .
1+cos x 1+u

37. Let u=sec x . Then du=sec x tan xdx , so

1 3 1
 sec x tan xdx= sec x(sec x tan x)dx= u du=
3 2 2 3
u +C= sec x+C .
3 3

3 3 2
38. Let u=x +1 . Then x =u 1 and du=3x dx , so
3 1 1
 x +1  x  x dx= u (u 1) du =  (u  u )du
3 3 3 2 1/3 4/3 1/3

3 5
x +1 x dx = 3 3
1 3 7/3 3 4/3 1 3 7/3 1 3 4/3
= u  u +C= (x +1)  (x +1) +C
3 7 4 7 4

a+1 a
39. Let u=b+cx . Then du=(a+1)cx dx , so
1 1 2 3/2 2
 x
a
b+cx dx= u
a+1 1/2
(a+1)c
du=
(a+1)c 3
u +C=
3c(a+1)
b+cx
a+1
( ) 3/2+C .
40. Let u=cos t. Then du= sin t dt and sin t dt= du , so
 sin t sec (cos t)dt= sec u (  du ) = tan u+C= tan (cos t)+C .
2 2

2
41. Let u=1+x . Then du=2xdx , so
1
du
1+x 1 x  1 2  1 1
 2
dx=
2
dx+
2
dx=tan x+
u
=tan x+ ln u +C
2
1+x 1+x 1+x
 1 1  1 1
=tan x+
2
2 2 2
ln 1+x +C=tan x+ ln 1+x +C [ since 1+x >0 ].
2
( )
1
du
1  1 1 
42. Let u=x . Then du=2xdx , so 
2 x
4
dx=
2
2
=
2
tan u+C= tan
2
1
( x2) +C .
1+x 1+u

43. Let u=x+2 . Then du=dx , so

5
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

 4
x
dx = 
u 2
4
du= ( u3/4 2u 1/4) du= 47 u7/4 2 4 3/4
3
u +C
x+2 u
4 7/4 8 3/4
= (x+2)  (x+2) +C
7 3

44. Let u=1 x . Then x=1 u and dx= du , so

2 2 2
x (1 u) 1 2u+u  1/2

1 x
dx = 
u
(  du ) = 
u
du=  u  2u +u
1/2 3/2
du ( )
1/2 2 3/2 2 5/2 4 3/2 2 5/2
=  2u  2 u + u +C= 2 1 x + (1 x)  (1 x) +C
3 5 3 5

3x 1
45. f (x)= .
4
( 3x  2x+1)
2

2
u=3x  2x+1  du=(6x 2)dx=2(3x 1)dx , so
3x 1 1 1 4
 4
dx =  1 du =  u du
2 2
( 2
3x  2x+1 ) u
4

1 3 1
=  6 u +C= 3
+C
2
6 3x  2x+1 ( )

1
Notice that at x= , f changes from negative to positive, and F has a local minimum.
3

x 2
46. f (x)= . u=x +1 du=2xdx , so
2
x +1
x
 dx = 
1 1 1 
du =  u
1/2
du
2 u 2 2
x +1

6
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

= u1/2+C= x2+1 +C.

Note that at x=0 , f changes from negative to positive and F has a local minimum.

1 4 1
 sin x cos xdx= u du=
3 3 3 4
47. f (x)=sin x cos x . u=sin x du=cos xdx,so u +C= sin x+C
4 4

Note that at x= , f changes from positive to negative and F has a local maximum. Also, both f
2
and F are periodic with period  , so at x=0 and at x= , f changes from negative to positive and F
has local minima.

2 2 2
48. f ( )=tan  sec  . u=tan   du=sec  d , so

1 3 1
 tan  sec  d = u du=
2 2 2 3
u +C= tan  +C
3 3

Note that f is positive and F is increasing. At x=0 , f =0 and F has a horizontal tangent.

49. Let u=x 1 , so du=dx . When x=0 , u= 1 ; when x=2 , u=1 . Thus,  (x 1) dx=
2 25 1 25
u du=0 by
0  1
25
Theorem 7(b), since f (u)=u is an odd function.
7
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

50. Let u=4+3x , so du=3dx . When x=0 , u=4 ; when x=7 , u=25 . Thus,

3/2 25
1 1 u 2 3/2 3/2 2 234
 4+3x dx=
7 25
u du = = (25  4 )= (125 8)= =26
0 4 3 3 3/2 4 9 9 9

3 2
51. Let u=1+2x , so du=6x dx . When x=0 , u=1 ; when x=1 , u=3 . Thus,

1 1 1 6 3 1 6 6 1 728 182

1 2
0
x ( 1+2x3) 5 dx= 1
3 5
u
6
du =
6 6
u
1
=
36
(3  1 )=
36
(729 1)=
36
=
9

2
52. Let u=x , so du=2xdx . When x=0 , u=0 ; when x=  , u= . Thus,
  1 1  1 1
( )
 0 xcos x dx= 0 cos u 2 du = 2 sin u 0 = 2 (sin   sin 0)= 2 (0 0)=0 .
2

1
53. Let u=t/4 , so du= dt . When t=0 , u=0 ; when t= , u= /4 . Thus,
4
  /4  /4 
 sec (t/4)dt=
2 2
sec u (4du)=4 tan u =4 tan  tan 0 =4(1 0)=4 .
0 0 0 4

1  1 
54. Let u= t , so du= dt . When t= , u= ; when t= , u= . Thus,
6 6 2 2
 1 1  /2 1 1
 csc  t cot  t dt=
1/2 /2
csc u cot u du =  csc u = (1 2)= .
1/6  /6    /6  

 /6
55. 
3 3
tan  d =0 by Theorem (b), since f ( )=tan  is an odd function.
  /6

dx 1 3
56. 
2
does not exist since f (x)= has an infinite discontinuity at x= .
0 2 2 2
(2x 3) (2x 3)

2 1
57. Let u=1/x , so du= 1/x dx . When x=1 , u=1 ; when x=2 , u= . Thus,
2
1/x
e u 1/2
 dx=
2 1/2 u 1/2
e ( du)= e = (e  e)=e e .
1 2 1 1
x

2
58. Let u= x , so du= 2xdx . When x=0 , u=0 ; when x=1 , u= 1 . Thus,
2
1  x  1 u 1 1 u 1 1 1 0 1
 0xe dx= 0 e  2 du = 2 e 0 = 2 e  e = 2 (1 1/e) . ( )
8
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

 1
59. Let u=cos  , so du= sin  d . When  =0 , u=1 ; when  = , u= . Thus,
3 2
 /3 sin   du  2 1 1
 d = =
1/2 1
u du=  = 1 ( 2)=1 .
0 2 1 2 1/2 u 1/2
cos  u

2 2
 /2 x sin x x sin x
60.  dx=0 by Theorem (b), since f (x)= is an odd function.
  /2 6 6
1+x 1+x

61. Let u=1+2x , so du=2dx . When x=0 , u=1 ; when x=13 , u=27 . Thus,
27  2/3 1 1 1/3 27 3
 0 3 dx = u
13
du =  3u = (3 1)=3 .
2 1 2 2 1 2
( 1+2x )

62. Let u=sin x , so du=cos xdx . When x=0 , u=0 ; when x= , u=1 . Thus,
2
 /2
 cos x sin (sin x)dx= sin udu=  cos u = (cos 1 1)=1 cos 1 .
1 1
0 0 0

63. Let u=x 1 , so u+1=x and du=dx . When x=1 , u=0 ; when x=2 , u=1 . Thus,
2 5/2 2 3/2 1 2 2 16
 1x x 1 dx= 0(u+1) u du= 0(u +u )du= 5 u + 3 u
2 1 1 3/2 1/2
= + = .
0 5 3 15

1
64. Let u=1+2x , so x= (u 1) and du=2dx . When x=0 , u=1 ; when x=4 , u=9 . Thus,
2
1
xdx (u 1)

4 9
= 9 2 du 1 9 1/2  1/2 1 2 3/2 1/2
0 1+2x  1 =  (u  u )du=
2 4 1 4 3
u  2u
1
u
1 2 3/2 1/2 9 1 20 10
=  u  3u = (27 9) (1 3) = =
4 3 1 6 6 3

dx 4
65. Let u=ln x , so du= . When x=e , u=1 ; when x=e ; u=4 . Thus,
x
4
dx 4  1/2 4
 = u
e 1/2
du=2 u =2(2 1)=2 .
e x ln x 1 1

 1 dx 1 
66. Let u=sin x , so du= . When x=0 , u=0 ; when x= , u= . Thus,
2 2 6
1 x

9
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

 1 2  /6 2
sin x  /6 u 
 dx=
1/2
udu= = .
0 2 0 2 0 72
1 x

dx 1
67. 
4
does not exist since f (x)= has an infinite discontinuity at x=2 .
0 3 3
(x 2) (x 2)

2 2 2
68. Assume a>0 . Let u=a  x , so du= 2xdx . When x=0 , u=a ; when x=a , u=0 . Thus,
2 2
1 1 a 1/2 1 2 3/2 a 1 3
 0x a  x dx= 2u du =  u du= 
a 2 2 0 1/2
 u = a .
a 2 2 0 2 3 0 3

2 2 1 2 2
69. Let u=x +a , so du=2xdx and xdx= du . When x=0 , u=a ; when x=a , u=2a . Thus,
2
2 2 2
1 1 2 3/2 2a 1 3/2 2a
x +a dx =
2a 1/2

a 2 2
x u du = u = u
0 2
a 2 2 3 a
2 3 2
a
1 2 3/2 2 3/2 1 3
= (2a )  (a ) = ( 2 2  1 ) a
3 3

70. 
a 2 2 2 2
x x +a dx=0 by Theorem 7(b), since f (x)=x x +a is an odd function.
 a

71. From the graph, it appears that the area under the curve is about
1
1+ a little more than  1 0.7 , or about 1.4 . The exact area is given by A=
1
2x+1 dx . Let
2 0
u=2x+1 , so du=2dx . The limits change to 2 0+1=1 and 2 1+1=3 , and
1 1 2 3/2 3 1 1
A=
3
u du = u = ( 3 3  1) = 3   1.399 .
1 2 2 3 1 3 3

1
72. From the graph, it appears that the area under the curve is almost    2.6 , or about 4 . The
2

10
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

exact area is given by

  
A = (2sin x sin 2x)dx= 2 cos x   sin 2xdx
0 0 0
= 2( 1 1) 0=4


Note:  sin 2xdx=0 since it is clear from the graph of y=sin 2x
0
  /2
that  sin 2xdx=  sin 2xdx .
 /2 0

73. First write the integral as a sum of two integrals: I= (x+3) 4 x d x=I +I =
2 2 2 2
x 4 x d
 2 1 2  2

x+
2 2 2
3 4 x d x . I =0 by Theorem 7(b), since f (x)=x 4 x is an odd function and we are
 2 1
integrating from x= 2 to x=2 . We interpret I as three times the area of a semicircle with radius 2 ,
2
1
so I=0+3
2
( 2
  2 =6 . )
2 2 2
74. Let u=x . Then du=2xdx and the limits are unchanged ( 0 =0 and 1 =1 ), so
1 1
I= x 1 x dx= 
1 4 2
1 u du . But this integral can be interpreted as the area of a quarter circle
0 2 0
1 1 1
with radius 1 . So I= 
2 4
2
  1 =  .
8
( )
75. First Figure
2
Let u= x , so x=u and dx=2udu . When x=0 , u=0 ; when x=1 , u=1 . Thus,
x
A = e dx= e (2udu)=2 ue du .
1 1 u 1 u
1 0 0 0

Second Figure
A = 2xe dx=2 ue du .
1 x 1 u
2 0 0

Third Figure
Let u=sin x , so du=cos xdx . When x=0 , u=0 ; when
11
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

  /2 sin x  /2 sin x
, u=1 . Thus, A = sin 2xdx= (2sin xcos x)dx= e (2udu)=2 ue du .
1 u 1 u
x= e e
2 3 0 0 0 0

Since A =A =A , all three areas are equal.

1 2 3

bt /
76. Let r(t)=ae with a=450.268 and b=1.12567 , and n(t)= population after t hours. Since r(t)=n (t)
,  r(t)dt=n(3) n(0) is the total change in the population after three hours. Since we start with 400
3
0
bacteria, the population will be
a bt 3 a 3b
n(3)=400+ r(t)dt=400+ ae dt=400+
3
0
3
0
bt
b
e
0
=400+
b
( )
e  1  400+11,313=11,713 bacteria

77. The volume of inhaled air in the lungs at time t is

t 1 2 2 t/5 1 5 2 2
V (t) =  f (u)du= u du=
t
sin sin v dv v= u,dv= du]
0 02 5 0 2 2 5 5
5 2 t/5 5 2 5 2
=  cos v =  cos t +1 = 1 cos t liters
4 0 4 5 4 5

78.
 2
Number of calculators =x(4) x(2)= 5000 1 100(t+10)
4
dt
2
 1 4 100 100
=5000 t+100(t+10) =5000 4+  2+  4048
2 14 12

1 1 1
79. Let u=2x . Then du=2dx , so  f (2x)dx= du = 
2 4 4
f (u) f (u)du= (10)=5 .
0 0 2 2 0 2

1 1 1
80. Let u=x . Then du=2xdx , so  xf x dx=
2 3
0
( 2) 9
0
f (u)
2
du = 
2
9
0
f (u)du=
2
(4)=2 .

81. Let u= x . Then du= dx , so

 b  a  a
 f ( x)dx= f (u)( du)= f (u)du=
b
f (x)dx
a  a  b  b

From the diagram, we see that the equality follows from the fact that we are reflecting the graph of f ,
and the limits of integration, about the y axis.

12
 Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

82. Let u=x+c . Then du=dx , so

 f (x+c)dx= f (u)du=
b b+c b+c
f (x)dx
a a+c a+c

From the diagram, we see that the equality follows from the fact that
we are translating the graph of f , and the limits of integration, by a distance c .

83. Let u=1 x . Then x=1 u and dx= du , so

 x (1 x) dx= ( 1 u ) u ( du)= 0u (1 u) du= 0x (1 x) dx .
1 a b 0 a b 1 b a 1 b a
0 1

84. Let u=  x . Then du= dx . When x= , u=0 and when x=0 , u= . So
 
  (  u) f (sin (  u))du=
0
xf (sin x)dx =  (  u) f (sin u)du
0  0
   
=   0
f (sin u)du  0
uf (sin u)du=  0
f (sin x)dx  0
xf (sin x)dx
    
 2 xf (sin x)dx=  f (sin x)dx  xf (sin x)dx=  f (sin x)dx .
0 0 0 2 0

xsin x sin x t
85. 2
=x 2
=xf (sin x) , where f (t)=
. By Exercise 84,
2
1+cos x 2 sin x 2 t
 xsin x      sin x
 0 2
dx= 0
xf (sin x)dx= 
2 0
f (sin x)dx= 
2 0 2
dx
1+cos x 1+cos x
Let u=cos x . Then du= sin xdx . When x= , u= 1 and when x=0 , u=1 . So
  sin x   1 du  1 du   1 1

2 0 2
dx= 
2 1 2
= 
2 1 2
=
2
tan u =
 1
1+cos x 1+u 1+u

13
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.5 The Substitution Rule

2
  1  1    
tan 1 tan ( 1) =   =
2 2 4 4 4

14
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.6 The Logarithm Defined as an Integral

1. (a)
We interpret ln 1.5 as the area under the curve y=1/x from x=1 to x=1.5 . The area of the rectangle
1 2 1 1 1 2 5
BCDE is  = . The area of the trapezoid ABCD is  1+ = . Thus, by comparing
2 3 3 2 2 3 12
1 5
areas, we observe that <ln 1.5< .
3 12

(b) With f (t)=1/t , n=10 , and  x=0.05 , we have


1.5
ln 1.5 = (1/t)dt (0.05)[ f (1.025)+ f (1.075)+   + f (1.475)]
1
1 1 1
= (0.05) + +   +  0.4054
1.025 1.075 1.475

1 / 1 1/2 1 1
2. (a) y= , y = . The slope of AD is = . Let c be the t coordinate of the point on
t 2 2 1 2
t
1 1 1 1 2
y= with slope  . Then  =  c =2 c= 2 since c>0 . Therefore, the tangent line is given
t 2 2 2
c
1 1 1
by y =
2
( t 2 )  y=
2
t+ 2 .
2

(b)
Since the graph of y=1/t is concave upward, the graph lies above the tangent line, that is, above the
1
line segment BC . Now AB = + 2 and CD = 1+ 2 . So the area of the trapezoid ABCD is
2
1 1 3
 + 2 + (  1+ 2 ) 1 = + 2  0.6642 . So ln 2> area of trapezoid ABCD>0.66 .
2 2 4

3.

1
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.6 The Logarithm Defined as an Integral

1 1 1 1 1
and so + +   + <
n
The area of R is dt=ln n .
i i+1 2 3 n 1 t

1 1 1 n1
The area of S isand so 1+ +   + > dt=ln n .
i i 2 n 1 1 t
1 1 1 1 1
Thus, + +   + <ln n<1+ +   + .
2 3 n 2 n 1

4. (a) From the diagram, we see that the area under the graph of y=1/x between x=1 and x=2 is less
than the area of the square, which is 1 . So ln 2= (1/x)dx<1 . To show the other side of the
2
1
inequality, we must find an area larger than 1 which lies under the graph of y=1/x between x=1 and
x=3 . One way to do this is to partition the interval 1,3 into 8 intervals of equal length and calculate
the resulting Riemann sum, using the right endpoints:

1 1 1 1 1 1 1 1 1 28,271
+ + + + + + + = >1
4 5/4 3/2 7/4 2 9/4 5/2 11/4 3 27,720

and therefore 1< (1/x)dx=ln 3 .

3
1
A slightly easier method uses the fact that since y=1/x is concave upward, it lies above all its tangent
3 2 5 2
lines. Drawing two such tangent lines at the points , and , , we see that the area
2 3 2 5
under the curve from x=1 to x=3 is more than the sum of the areas of the two trapezoids, that is,
2 2 16 16 3
+ = . Thus, 1< < (1/x)dx=ln 3 .
3 5 15 15 1
(b) By part (a), ln 2<1<ln 3 . But e is defined such that ln e=1 , and because the natural logarithm
function is increasing, we have ln 2<ln e<ln 3 2<e<3 .

( r ) , then f /(x)= ( 1/xr ) (rxr 1) =r/x . But if g(x)=rln x , then g /(x)=r/x . So f and g
5. If f (x)=ln x
must differ by a constant: ln ( x ) =rln x+C . Put x=1 : ln ( 1 ) =rln 1+C  C=0 , so ln ( x ) =rln x .
r r r

2
Stewart Calculus ET 5e 0534393217;5. Integrals; 5.6 The Logarithm Defined as an Integral

6. Using the second law of logarithms and Equation 10, we have

( x y
)
ln e /e =ln e  ln e =x y=ln e
x y
( x y) . Since ln x y
is a one to one function, it follows that e /e =e
x y
.

7. Using the third law of logarithms and Equation 10, we have ln e =rx=rln e =ln e
rx x
( x) r . Since ln is
x r
a one to one function, it follows that e = e
rx
( ) .

x
8. Using Definition 13 and the second law of exponents for e , we have
xln a x
x y ( x y ) ln a xln a yln a e a
a =e =e = = .
yln a y
e a

x
9. Using Definition 13, the first law of logarithms, and the first law of exponents for e , we have
x xln (ab) x(ln a+ln b) xln a+xln b xln a xln b x x
(ab) =e =e =e =e e =a b .

r s
10. Let log x=r and log y=s . Then a =x and a =y .
a a
r s r+s
(a) xy=a a =a  log (xy)=r+s=log x+log y
a a a
r
x a r s x
(b) = =a  log =r s=log x log y
y s a y a a
a
(c) x = a
y
( r ) y=ary  log
a
( x ) =ry=ylog x
y
a

3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

1.
x=4 4 4
A= ( y  y ) dx= ( 5x x )  x dx= ( 4x x ) dx
2 2
T B
x=0 0 0

1 3 2 4 64 32
= 2x  x = 32  (0)=
3 0 3 3

2.
1 2 2
A =
2 3/2
x+2  dx= (x+2)  ln (x+1)
0 x+1 3 0
2 3/2 2 3/2 16 4
= (4)  ln 3  (2)  ln 1 =  ln 3 2
3 3 3 3

3.
A =
y=1
y= 1 ( x  x ) dy=
R L
1
 1
y
e ( y2 2) dy

1 3 1 1  1 1 1 10
=
1
 1
( e  y +2 ) dy=
y 2
e
y
3
y +2y
 1
1
= e
3
+2  e +
3
 2 =e +
e 3

4.
3 3
A = ( 2y y )  ( y  4y)
2 2
dy= (  2y2+6y) dy
0 0

2 3 2 3
=  y +3y =( 18+27) 0=9
3 0

5.
2
A = ( 9 x2)  ( x+1 ) dx
 1
2
= ( 8 x x2) dx
 1
2 3 2
x x
= 8x 
2 3  1
8 1 1
= 16 2   8 +
3 2 3

1
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

1 39
=22 3+ =
2 2

6.
 /2
A =
0
( ex sin x) dx
x  /2
= e +cos x
0

= e( /2
)
+0  ( 1+1 )
 /2
=e  2

2
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2 2
7. The curves intersect when x=x  x  x=0 x(x 1)=0 x=0 , 1 .
1
A = ( x x2) dx
0

1 2 1 3 1
= x x
2 3 0
1 1
= 
2 3
1
=
6

8.
1
A = ( x2 x4) dx
 1
1
=2 ( x2 x4) dx
0

1 3 1 5 1
=2 x x
3 5 0
1 1 4
=2  =
3 5 15

3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

9.
1 1 1 2
A =
2
 dx= ln x+
1 x 2 x 1
x
1
= ln 2+  ( ln 1+1 )
2
1
=ln 2  0.19
2

2
3+x x x x 2
10. 1+ x = =1+  x=  x=  9x x =0 x(9 x)=0 x=0 or 9 , so
3 3 3 9
9 9
3+x x
A=  ( 1+ x )
3
dx= ( 1+ x ) 1+
3
dx=
0 0
9 9
1 2 3/2 1 2 27 9
 x
3
x dx=
3
x 
6
x
0
=18 =
2 2
0

4
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

11.
1
A = ( 2
)
x  x dx
0

2 3/2 1 3 1
= x  x
3 3 0
2 1
= 
3 3
1
=
3

3 3 3 2
12. x= x  x =x x  x=0 x(x  1)=0 x(x+1)(x 1)=0
x= 1 , 0 , or 1 , so
1 0 1 1
A= 
3
x  x dx=  ( x 3
x dx+
) ( 3
x  x dx=2
) ( x1/3 x) dx
 1  1 0 0
[by symmetry]
3 4/3 1 2 1 3 1 1
=2 x  x =2  =
4 2 0 4 2 2

5
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2 2 2 2 2
13. 12 x =x  6 2x =18 x =9 x = 3 , so
3 3
A = ( 12 x )  ( x  6)
2 2
dx=2 ( 18 2x2) dx [ by symmetry]
 3 0

2 3 3
=2 18x x =2 ( 54 18 )  0 =2(36)=72
3 0

3 3 2
14. x  x=3x x  4x=0 x(x  4)=0 x(x+2)(x 2)=0 x=0 ,  2 , or 2 .
By symmetry,
2 2 2 2
1 4
A=  3x ( x  x)3
dx=2 3x ( x  x)
3
dx=2 ( 4x x3) dx=2 2
2x 
4
x
0
 2 0 0
= 2(8 4)=8 ]

1 1 2 2
15. x= x  x =x x  4x=0 x(x 4)=0 x=0 or 4 , so
2 4

6
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

4 9 4 9
1 1 2 3/2 1 2 1 2 2 3/2
A = x x dx+ x x dx= x  x + x x
0 2 4 2 3 4 0 4 3 4

16 81 16 81 32 59
=  4  0 +  18  4 = +  26=
3 4 3 4 3 12

16.
3 3 2 3
A = ( 8 x )  x
2 2
dx=2 8 2x dx=2
2
( 8 2x ) dx+2 ( 2x2 8) dx
2

 3 0 0 2

2 32 3 32 16 16
=2 8x x x  8x =2
+2 16  0 +2 ( 18 24 )   16
0 3 3 2 3 3
32 32 64 92
=32 +20 =52 =
3 3 3 3

2 2 1 1
17. 2y =1 y 2y +y 1=0 (2y 1)(y+1)=0 y= or  1 , so x= or 2 and
2 2
1/2 1/2 1/2
1 2 2 3
A = ( 1 y )  2y dy= 
2
( 1 y 2y2) dy= y
2
y
3
y
 1
 1  1
1 1 1 1 2 7 5 7 20 27 9
=     1 + =   = + = =
2 8 12 2 3 24 6 24 24 24 8
7
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2
18. 4x+x =12 (x+6)(x 2)=0 x= 6 or x=2 , so y= 6 or y=2 and
2 2
1 2 1 3 1 2 2 2 64
A=   y +3  y dy=  y y +3y =   2+6  (18 18 18)=22 = .
4 12 2  6 3 3 3
 6

2 2 2 2
19. The curves intersect when 1 y =y  1 2=2y  y =1 y= 1 .
1
A = ( 1 y2)  ( y2 1) dy
 1
1
=  2 1 y dy( 2
)
 1
1
=2 2 ( 1 y2) dy
0
1 3 1 1 8
=4 y y =4 1 =
3 0 3 3

8
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

1 2 1
 x 2  x x 1 2 4
20. A=2 sin  x dx=2  cos  =2 0    0 =  1
0 2  2 2 0 2  

21. Notice that cos x=sin 2x=2sin xcos x

2sin xcos x cos x=0 cos x(2sin x 1)=0
 
2sin x=1 or cos x=0 x= or .
6 2
 /6  /2
A=  ( cos x sin 2x ) dx+  ( sin 2x cos x ) dx
0  /6

1  /6 1  /2
= sin x+
cos 2x +  cos 2x sin x
2 0 2  /6
1 1 1 1 1 1 1 1 1
= +  0+ 1 +  1    =
2 2 2 2 2 2 2 2 2

1
22. sin x=sin 2x=2sin xcos x when sin x=0 and when cos x= ;
2

that is, when x=0 or .
3
 /3  /2
A=  ( sin 2x sin x ) dx+  ( sin x sin 2x ) dx
0  /3

1  /3 1  /2
=  cos 2x+cos x + cos 2x cos x
2 0 2  /3

9
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

1 1 1 1
=   +   +1
2 2 2 2
1 1 1 1 1
+   0    =
2 2 2 2 2


23. From the graph, we see that the curves intersect at x=0 , x= , and x= . By symmetry,
2
  /2  /2
2x 2x 2x
A = cos x 1 dx=2  cos x 1 dx=2  cos x 1+ dx
0  0  0 
 /2 2
1 2  1    
=2 sin x x+ x =2 1 +  0 =2 1 + =2
 0 2  4 2 4 2

2 2
24. For x>0 , x=x  2 0=x  x 2 0=(x 2)(x+1) x=2 . By symmetry,
2 2 2 2
1 2 1 3
 x  ( x  2)
2
dx = 2 x ( x  2)
2
dx=2 ( x x2+2) dx=2 2
x
3
x +2x
0
=
 2 0 0
8 20
2 2 +4 =
3 3

10
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2
25. The curves intersect when x =
2
2

4 2
x +x =2
4 2
x +x  2=0 ( x2+2) ( x2 1) =0 2
x =1 x= 1 .
x +1
2 2  1 1 3 1  1 2
A= dx=2
1 2 1 2
 x  x dx=2 2tan x x =2 2  =   2.47
 1 2 0 2 3 0 4 3 3
x +1 x +1

26.
1 2
A = sin  x ( x  x)
2
dx+ ( x2 x)  sin  x dx
0 1

11
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

1 1 3 1 2 1 1 3 1 2 1 2
=  cos  x x+ x + x x + cos  x
 3 2 0 3 2  1
1 1 1 1 8 1 1 1 1
=  +   +  2+   
 3 2  3  3 2 
4
= +1


27.

From the graph, we see that the curves intersect at x= a  1.02 , with
2
2cos x>x on (  a,a ) . So the area of the region bounded by the
curves is
a a
A=  ( 2cos x x ) dx=2 ( 2cos x x2) dx
2

 a 0

1 3 a
=2 2sin x x  2.70
3 0

28.

12
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2 5
4 7 4
A =  x+5   x+5 dx+  x+5  ( x 4 ) dx
0 5 2 2 5
2 5
27 9
= xdx+  x+9 dx
0 10 2 5
27 2 2 9 2 5
= x +  x +9x
20 0 10 2
27 45 18 27
=  0 +  +45   +18 =
5 2 5 2

29.
1
A =
3
x  x dx
 1
1
=2 ( x x3) dx [ by symmetry]
0
1 2 1 4 1
=2 x x
2 4 0
1 1 1
=2  =
2 4 2

13
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2 2
30. The curves intersect when x+2 =x x+2=x  x  x 2=0
(x 2)(x+1)=0 x= 1 or 2 .
4
A = x+2  x dx
0
2 4
= ( x+2  x ) dx+ ( x x+2 ) dx
0 2

2 3/2 1 2 2 1 2 2 3/2 4
= (x+2)  x + x (x+2)
3 2 0 2 3 2
16 2 2 16
=
3
 2 
3
(2 2 )  0 + 8
3
(6 6 )  2
3
32 4 44 4
=4+  2 4 6=  4 6 2
3 3 3 3

 x 2 2  x 1 0
31. Let f (x)=cos  sin and  x= .
4 4 4
The shaded area is given by
1
1 1 3 5 7
A= f(x)dx M = f +f +f +f  0.6407
0
4 4 8 8 8 8

3 3
32. The curves intersect when 16 x =x
14
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

3 3 3 3
16 x =x  2x =16
x=2. x =8
3 3 2 0
Let f (x)= 16 x  x and  x= .
4
The shaded area is given by
2
A= f (x)dx  M 4
0
2 1 3 5 7
= f +f +f +f
4 4 4 4 4
 2.8144

33.

From the graph, we see that the curves intersect at x= a  1.02 , with
2
2cos x>x on (  a,a ) . So the area of the region bounded by the
curves is
a a
A=  ( 2cos x x ) dx=2 ( 2cos x x2) dx
2

 a 0

1 3 a
=2 2sin x x  2.70
3 0

34.

15
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

3 4
From the graph, we see that the curves intersect at x=0 and at x=a 1.17 , with 3x x >x on ( 0,a ) . So
the area of the region bounded by the curves is
a a
3 2 1 4 1 5
A=  ( 3x x3)  x4 dx= 2
x
4
x
5
x
0
0
 1.15

35.

From the graph, we see that the curves intersect at x= a  0.86 .
So the area of the region bounded by the curves is

1 2 1 4 a
A=2
a 2 3
xcos (x ) x dx=2 sin (x ) x  0.40
0 2 4 0

36.

2 x
From the graph, we see that the curves intersect at x=a  1.32 and x=b 0.54 , with 2 x >e on ( a,b )
. So the area of the region bounded by the curves is

1 3 x b
A=
b
a
( 2 x2)  ex dx= 2x
3
x e  1.45
a

37. As the figure illustrates, the curves y=x and

16
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

5 3 5 3
y=x  6x +4x enclose a four part region symmetric about the origin (since x  6x +4x and x are odd
5 3 4 2
functions of x ). The curves intersect at values of x where x  6x +4x=x ; that is, where x(x  6x +3)=0
2 6 36 12
. That happens at x=0 and where x = =3 6 ; that is, at x= 3+ 6 ,  3 6 , 0 ,
2
3 6 , and 3+ 6 .
The exact area is
3+ 6 3+ 6
2  ( x  6x +4x)  x
5 3
dx =2 
5 3
x  6x +3x dx
0 0
3 6 3+ 6
=2  ( x  6x +3x) dx+2
5 3
 (  x5+6x3 3x) dx
0 3 6

=12 6  9

2 2
38. The inequality x
2y describes the region that lies on, or to the right of, the parabola x=2y . The
inequality x 1 y describes the region
that lies on, or to the left of, the curve x=1 y =
1 y if y
0
1+y if y<0
.
So the given region is the shaded region that lies between the curves.
{
2 2
The graphs of x=1 y and x=2y intersect when 1 y=2y 
2 1
2y +y 1=0 (2y 1)(y+1)=0 y= (for y
0 ).
2

17
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

By symmetry,
1/2 1/2
2 3 1 2 1 1 1 7 7
A=2 
2
(1 y) 2y dy=2  y y +y =2   +  0 =2 = .
0 3 2 0 12 8 2 24 12

1 1
39. 1 second = hour, so 10 s = h. With the given data, we can take n=5 to use the
3600 360
1/360 0 1

.
Midpoint Rule.  t= = , so
5 1800
1/360 1/360 1/360
distance
Kelly
 distance
Chris =  v dt
K
 v dt=
C
 ( v  v ) dt
K C
0 0 0
1
 M= ( v  v ) (1)+ ( v  v ) (3)+ ( v  v ) (5)
5 1800 K C K C K C

+ ( v  v ) (7)+ ( v  v ) (9)
K C K C
1
= ( 22 20 ) + ( 52 46 ) + ( 71 62 ) + ( 86 75) + ( 98 86 )
1800
1 1 1 1
= ( 2+6+9+11+12 ) = (40)= mile,or117 feet
1800 1800 45 3

40. If x= distance from left end of pool and w=w(x)= width at x , then the Midpoint Rule with n=4
16
b a 8 2 0
=4 gives Area =  wdx 4(6.2+6.8+5.0+4.8)=4(22.8)=91.2 m .
2
and  x= =
n 4 0

x
41. We know that the area under curve A between t=0 and t=x is  v (t)dt=s (x) , where v (t) is the
A A A
0
velocity of car A and s is its displacement. Similarly, the area under curve B between t=0 and t=x is
A
x
 v (t)dt=s (x) .
B B
0
(a) After one minute, the area under curve A is greater than the area under curve B . So car A is ahead
after one minute.
(b) The area of the shaded region has numerical value s (1) s (1) , which is the distance by which A
A B
is ahead of B after 1 minute.
(c) After two minutes, car B is traveling faster than car A and has gained some ground, but the area
under curve A from t=0 to t=2 is still greater than the corresponding area for curve B , so car A is still
ahead.
(d) From the graph, it appears that the area between curves A and B for 0 t 1 (when car A is going
18
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

faster), which corresponds to the distance by which car A is ahead, seems to be about 3 squares.
Therefore, the cars will be side by side at the time x where the area between the curves for 1 t x
(when car B is going faster) is the same as the area for 0 t 1 . From the graph, it appears that this
time is x 2.2 . So the cars are side by side when t 2.2 minutes.

/
42. The area under R (x) from x=50 to x=100 represents the change in revenue, and the area under
/
C (x) from x=50 to x=100 represents the change in cost. The shaded region represents the difference
between these two values; that is, the increase in profit as the production level increases from 50 units
to 100 units. We use the
Midpoint Rule with n=5 and  x=10 :
M =  x{+[R /(65) C /(65)]+[R /(75) C /(75)]
5
/ / / /
+[R (85) C (85)]+[R (95) C (95)]}
 10(2.40 0.85+2.20 0.90+2.00 1.00+1.80 1.10+1.70 1.20)
= 10(5.05)=50.5 thousand dollars
Using M would give us 50(2 1)=50 thousand dollars.
1

43.

To graph this function, we must first express it as a combination of explicit functions of y ; namely,
y= x x+3 . We can see from the graph that the loop extends from x= 3 to x=0 , and that by
symmetry, the area we seek is just twice the area under the top half of the curve on this interval, the
0
equation of the top half being y= x x+3 . So the area is A=2  ( x x+3 ) dx . We substitute u=x+3 ,
 3
so du=dx and the limits change to 0 and 3 , and we get
3 3
A =  2 (u 3) u du= 2 ( u3/2 3u1/2) du
0 0

2 5/2 3/2 3 2 2 24
= 2
5
u  2u = 2
0 5
( )
3 3  2 (3 3 ) =
5
3

19
 Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

44.

2 /
We start by finding the equation of the tangent line to y=x at the point ( 1,1 ) : y =2x , so the slope of
the tangent is 2(1)=2 , and its equation is y 1=2(x 1) , or y=2x 1 . We would need two integrals to
integrate with respect to x , but only one to integrate with respect to y .
1 1
1 1 2 1 2 3/2
A=  2
(y+1) y dy=
4
y + y
2 3
y
0
0
1 1 2 1
= +  =
4 2 3 12

45.

2
By the symmetry of the problem, we consider only the first quadrant, where y=x  x= y . We are
b 4
2 3/2 b 2 3/2 4
looking for a number b such that  y dy=
3/2 3/2 3/2 3/2
y dy y = y  b =4  b  2b =8
0 b 3 0 3 b
3/2 2/3
 b =4 b=4  2.52 .

a 4

46. (a) We want to choose a so that  1

2
dx=  1
2
dx
 1
x
a
=
1
 1
x
4

a
 
1
a
+1=
1 1
+ 
4 a
1 x a x
5 2 8
=  a= .
4 a 5
2 3
(b) The area under the curve y=1/x from x=1 to x=4 is . Now the line y=b must intersect the
4
2 3
curve x=1/ y and not the line x=4 , since the area under the line y=1/4 from x=1 to x=4 is only
16
, which is less than half of
20
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

3
. We want to choose b so that the upper area in the diagram is half of the total area under the curve
4
1
y= from x=1 to x=4 . This implies that
2
x
1
1 3 3 3 5
 ( 1/
1
y  1 ) dy=  2 y y =  1 2 b +b=  b 2 b + =0 . Letting c= b , we get
b 2 4 b 8 8 8
2 5 2 16 256 160 6 6
c  2c+ =0 8c  16c+5=0 . Thus, c= =1 . But c= b <1 c=1 
8 16 4 4
2 3 6 1
b=c =1+  = ( 11 4 6 )  0.1503 .
8 2 8

47. We first assume that c>0 , since c can be replaced by  c in both equations without changing the
graphs, and if c=0 the curves do not enclose a region. We see from the graph that the enclosed area A
lies between x= c and x=c , and by symmetry, it is equal to four times the area in the first quadrant.
The enclosed area is
c c
1 3
A = 4 ( c2 x2) dx=4 2
c x
3
x
0
0
3 1 3 2 3 8 3
=4 c c =4 c = c
3 3 3
8 3 3 3
So A=576 c =576 c =216 c= 216 =6 . Note that c= 6 is another solution, since the graphs
3
are the same.

48.

21
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

It appears from the diagram that the curves y=cos x and y=cos (x c) intersect halfway between 0 and
c , namely, when x=c/2 . We can verify that this is indeed true by noting that
cos (c/2 c)=cos ( c/2)=cos (c/2) .

The point where cos (x c) crosses the x axis is x= +c . So we require that
2
c/2 
 cos x cos (x c) dx=  cos (x c)dx (the negative sign on the RHS is needed since the second
0  /2+c
area is beneath the x axis) 

c/2 
sin x sin ( x c ) = sin ( x c ) 
0  /2+c

sin (c/2) sin ( c/2)   sin ( c) = sin (  c)+sin +c  c  2sin (c/2) sin c= sin c+1 . . So
2
1  
2sin (c/2)=1 sin (c/2)=  c/2=  c= .
2 6 3

49. The curve and the line will determine a region when they intersect at two or more points. So we
solve the equation x/ x +1 =mx (2 ) x=x ( mx +m ) 
2

( 2
x mx +m  x=0  ) x ( mx +m 1 ) =0 
2

2 2 1 m
x=0 or mx +m 1=0 x=0 or x = 
m

1
x=0 or x=  1 . Note that if m=1 , this has only the solution x=0 , and no region is determined.
m
But if 1/m 1>0 1/m>1 0<m<1 , then there are two solutions. [Another way of seeing this is to
2 /
observe that the slope of the tangent to y=x/(x +1) at the origin is y =1 and therefore we must have
0<m<1 .] Note that we cannot just integrate between the positive and negative roots, since the curve
22
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.1 Areas between Curves

2
and the line cross at the origin. Since mx and x/(x +1) are both odd functions, the total area is twice
the area between the curves on the interval 0, 1/m 1 . So the total area enclosed is

x 1 1 2 1/m 1
2
0
1/m 1
2
 mx dx=2
2
2
(
ln x +1 
2
mx ) 0
= ln (1/m 1+1) m(1/m 1)  (ln 1 0) =
x +1
ln (1/m) 1+m=m ln m 1

23
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

2
1. A cross section is circular with radius x , so its area is A(x)= ( x2) 2 .
1 1 1
2 2 1 5 1 
V = A(x)dx=  ( x ) dx= 
4
x dx=
5
x
0
=
5
0 0 0

x
2. A cross section is a disk with radius e , so its area is A(x)= ( ex ) 2 .
1 
V =
1
0
A(x)dx= 
1
0
( ex) 2 dx= 
1 2x
0
e dx= 
2
e
2x 1
0
=
2
( e2 1)

2
3. A cross section is a disk with radius 1/x , so its area is A(x)= (1/x) .

1
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes


2 2
1 2 1 1 2 1 
V = A(x)dx=  dx= dx=  =   (  1) =
1 1 x 2 x 1 2 2
1 x

2
4. A cross section is circular with radius x 1 , so its area is A(x)= ( x 1 ) = (x 1) .

5 5 5
1 2 25 4 15
V = A(x)dx=  (x 1)dx= x x =  5 +2 = 
2 2 2 2 2 2 2

2
5. A cross section is a disk with radius y , so its area is A(y)= ( y) .

2
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

4 4 4 4
1 2
V = A(y)dy=  
2
( y ) dy= ydy=
2
y
0
=8
0 0 0

6. A cross section is a disk with radius y y , so its area is A(y)=

2
( y y2) 2 .
1 1 1 1
2 2 1 5 1 4 1 3
V =  A(y)dy=  ( y y ) dy=  ( y4 2y3+y2) dy= 5
y
2
y+ y
3 0
0 0 0
1 1 1 
  + =
5 2 3 30

2
7. A cross section is a washer (annulus) with inner radius x and outer radius x , so its area is
A(x)= ( x )
2
 ( x2) 2= ( x x4) .
3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

1 1 1
1 2 1 5 1 1 3
V = A(x)dx=  ( x x4) dx= 2
x
5
x =
0

2 5
=
10
0 0

8. A cross section is a washer with inner radius 1 and outer radius sec x , so its area is
2
A(x)= (sec x)   (1) =
2
( sec 2x 1) .
1 1 1
V =  A(x)dx=   ( sec x 1) dx=2  ( sec 2x 1) dx=2
2 1
tan x x =2 (tan 1 1)
0
 1  1 0
3.5023

2
9. A cross section is a washer with inner radius y and outer radius 2y , so its area is
A(y)= (2y)  
2
( y2) 2= ( 4y2 y4) .
4
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

2 2 2
4 3 1 5 32 32 64
V = A(y)dy=  ( 4y2 y4) dy= 3
y
5
y =
0 3

5
=
15
0 0

2/3 3/2 3/2

10. y=x  x=y , so a cross section is a washer with inner radius y and outer radius 1 , and its
3/2 2
area is A(y)= (1)  
2
( y ) = ( 1 y ) . 3

1 1 1
1 4 3
V = A(y)dy=  ( 1 y3) dy= y
4
y = 
0 4
0 0

11. A cross section is a washer with inner radius 1 x and outer radius 1 x , so its area is
A(x)= (1 x)  
2
( 1 x ( 1 2x+x ) 
)
2
=
2
( 1 2 x +x ) = (  3x+x2+2 x ) .
1 1
  (  3x+x +2 x ) dx=
2
V = A(x)dx=
0 0

5
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

3 2 1 3 4 3/2 1 3 5 
  x+ x+ x =  + =
2 3 3 0 2 3 6

12. A cross section is circular with radius 4 x , so its area is A(x)=

2
( 4 x2) 2= ( 16 8x2+x4) .
2 2 2 2
8 3 1 5
V =  A(x)dx=2 A(x)dx=2  ( 16 8x2+x4) dx=2 16x
3
x+ x
5 0
=
 2 0 0
64 32 2 1 8 512
2 32 + =64 1 + =64  =
3 5 3 5 15 15

4
13. A cross section is an annulus with inner radius 2 1 and outer radius 2 x , so its area is

6
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

A(x)= ( 2 x4) 2  2
( 2 1 ) = ( 3 4x4+x8) .
1 1 1 1
4 5 1 9
 A(x)dx=2 A(x)dx=2  ( 3 4x +x ) dx=2
4 8
V = 3x x+ x =
 1 0 0 5 9 0

4 1 208
2 3 + = 
5 9 45

14.

V = 
3
1 { 1
x
 (  1)
2
 0 (  1 ) }
2
dx= 
3
1
1
x
+1
2
 1
2
dx

2 1 3
1

3
= + dx=  +2ln x
1 2 x x 1
x
1 2 1
=  +2ln 3  (  1+0 ) = 2ln 3+ =2 ln 3+
3 3 3

7
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

15.
1 1 1
2 2 2 2
V =  ( 1 y ) dy=2  ( 1 y ) dy=2  ( 1 2y2+y4) dy
 1 0 0

2 3 1 5 1 8 16
=2 y y+ y =2  = 
3 5 0 15 15

2 2
16. y= x  x=y , so the outer radius is 2 y .
1 1
2 2
V =  ( 2 y )  ( 2 y ) dy= 
2
( 4 4y2+y4)  ( 4 4y+y2) dy
0 0
1 1
1 5 5 3 2 1 5 8
=  ( y4 5y2+4y) dy= 5
y
3
y +2y =
0

5 3
+2 = 
15
0

8
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

2
17. y=x  x= y for x 0 . The outer radius is the distance from x= 1 to x= y and the inner radius
2
is the distance from x= 1 to x=y .
1 1
V =  { 2
y  ( 1) 
2
y  ( 1)
2
} dy=  (
2
y +1 )  ( y2+1) 2 dy
0 0
1 1
=  ( y+2 4 2
)
y +1 y  2y  1 dy=  ( y+2 4 2
y  y  2y dy)
0 0

1 2 4 3/2 1 5 2 3 1 1 4 1 2 29
= y+ y  y y = +   = 
2 3 5 3 0 2 3 5 3 30

18. For 0
y<2 , a cross section is an annulus with inner radius 2 1 and outer radius 4 1 , the area of
2 2
which is A (y)= ( 4 1 )   ( 2 1 ) . For 2
y
4 , a cross section is an annulus with inner radius y 1
1
2 2
and outer radius 4 1 , the area of which is A (y)= ( 4 1 )   ( y 1 ) .
2

9
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

4 2 4
  ( 4 1 )  ( 2 1 )  ( 4 1 )  ( y 1 )
2 2 2 2
V = A(y)dy= dy+ dy
0 0 2
4 4
1 3
= 8y +
2
0
 ( 8+2y y2) dy=16 + 8y+y 
2
3
y
2
2
64 8 76
= 16 + 32+16  16+4 = 
3 3 3

1 1 1
1 7 1 
19. R about OA (the line y=0 ): V = A(x)dx=  (x ) dx= 
32 6
x dx= x =
1
0 0 0 7 0 7

20. R about OC (the line x=0 ):

1
1 1 1 1
2 3 5/3 3 2
V = A(y)dy= ( ) 
2 3 2/3
 (1)   y dy= (1 y )dy= y y = 1 =
0 0 0 5 0 5 5

21. R about AB (the line x=1 ):

1
1 1
2
V = A(y)dy=  ( 1 ) 
1 3 1/3 2/3
y dy= (1 2y +y )dy
0
0 0

3 4/3 3 5/3 1 3 3 
= y y + y = 1 + =
2 5 0 2 5 10

22. R about BC (the line y=1 ):

1
1 1 1
V = A(x)dx= 
2 32 3 6
 (1)   (1 x ) dx= 1 (1 2x +x ) dx
0 0 0

10
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

1 1
1 4 1 7 1 1 5

3 6
= (2x  x )dx= x x =  =
0 2 7 0 2 7 14

23. R about OA (the line y=0 ):

2
1 1 1
1 2 1 1 
V = A(x)dx= 
2 2
 (1)   ( x ) dx= (1 x)dx= x
2
x
0
= 1
2
=
2
0 0 0

1 1 1
1 5 1 
24. R about OC (the line x=0 ): V = A(y)dy=  (y ) dy= 
22 4
y dy= y =
2
0 0 0 5 0 5

25. R about AB (the line x=1 ):

2
1 1 1
V = A(y)dy= 
2 22 2 4
 (1)   (1 y ) dy= 1 (1 2y +y ) dy
0 0 0
1 1
2 3 1 5 2 1 7

2 4
= (2y  y )dy= y y =  =
0 3 5 0 3 5 15

26. R about BC (the line y=1 ):

2
1 1 1
V = A(x)dx=  
2 1/2
( 1 x ) dx= (1 2x +x)dx
0 0 0

4 3/2 1 2 1 4 1 
= x x + x = 1 + =
3 2 0 3 2 6

27. R about OA (the line y=0 ):

3
1 1 1 1
1 2 1 7 1 1 5
V = A(x)dx= 
2 32 6
 ( x )   (x ) dx= (x x )dx=
2
x
7
x =
0

2 7
=
14
.
0 0 0
Note: Let = + mathpi5 +R . If we rotate about any of the segments OA , OC , AB , or BC , we
1 2 3
2 2
obtain a right circular cylinder of height 1 and radius 1 . Its volume is  r h= (1)  1= . As a check
for Exercises 19, 23, and 27, we can add the answers, and that sum must equal  . Thus,
  5 2+7+5
+ + =  = .
7 2 14 14

28.
11
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

R about OC (the line x=0 ):

3
1 1 1
2
V = A(y)dy= ( )  
3 22 2/3 4
 y (y ) dy= (y  y )dy
0 0 0

3 5/3 1 5 1 3 1 2
= y  y =  =
5 5 0 5 5 5

2  2
Note: See the note in Exercise 27. For Exercises 20, 24, and 28, we have + + = .
5 5 5

29. R about AB (the line x=1 ):

3
1 1 1
2
V = A(y)dy= ( 1 ) 
22 3 2 4 1/3 2/3
 (1 y )   y dy= (1 2y +y ) (1 2y +y ) dy
0 0 0
1 1
2 3 1 5 3 4/3 3 5/3

2 4 1/3 2/3
= ( 2y +y +2y  y )dy=  y+ y+ y  y
0 3 5 2 5 0

2 1 3 3 13
=  + +  =
3 5 2 5 30

Note: See the note in Exercise 27. For Exercises 21, 25, and 29, we have
 7 13 3+14+13
+ + =  = .
10 15 30 30

30. R about BC (the line y=1 ):

3
1 1
V = A(x)dx=
32 2
 (1 x )   ( 1 x ) dx
0 0
1 1
 
3 6 1/2 3 6 1/2
= (1 2x +x ) (1 2x +x) dx= ( 2x +x +2x  x)dx
0 0

1 4 1 7 4 3/2 1 2 1 1 1 4 1 10
=  x+ x+ x  x =  + +  =
2 7 3 2 0 2 7 3 2 21

Note: See the note in Exercise 27. For Exercises 22, 26, and 30, we have
5  10 15+7+20
+ + =  = .
14 6 21 42
31.

12
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

 /4

3 2
V = (1 tan x) dx
0

4 4 4
32. y=(x 2) and 8x y=16 intersect when (x 2) =8x 16=8(x 2) (x 2)  8(x 2)=0
3 4 4 4
(x 2) (x 2)  8 =0 x 2=0 or x 2=2 x=2 or 4 . y=(x 2)  x 2= y x=2+ y [ since x 2 ].
1
8x y=16 8x=y+16 x= y+2 .
8
16
V = 
0
{ 10
1
8
y+2
2
( 4
 10 2+ y ) }
2
dy

33.

 ( 1 0 )  ( 1
2 2
V = sin x ) dx
0


2 2
= 1  ( 1 sin x ) dx
0

34.
13
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes


 ( sin x+2 )
2 2
V =  2 dx
0

35.
8
V =  { 3 ( 2) 2

2
y +1  ( 2)
2
} dy
 8
2 2 2
 5 ( 1+y +2 )
2 2
= dy
 2 2

2 2 3
36. Solve the equations for x : ( y 1 ) =4 x x=4 ( y 1 ) and 2x+3y=6 x=3 y.
2
9 7
The points of intersection of the two curves are ( 3,0 ) and  , . Therefore,
4 2
7/2
V = 
0
{ 2
4 ( y 1 )  ( 5) 
2
3
3
2
y ( 5) }
2
dy

7/2
= 
0
{ 9 ( y 1 )
2 2
 8
3
2
y
2
} dy

14
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

37.
2
y=x and y=ln (x+1) intersect at x=0 and at x=a 0.747 .

V = 
a
0
{ ln (x+1)  ( x ) } dx
2 2 2
0.132

38.

2 x/2  2x
y=3sin (x ) and y=e +e intersect at x=a 0.772 and at x=b 1.524 .

V = 
b
a
{ 2
3sin (x ) 
2
( ex/2+e 2x) 2} dx  7.519

39.

V =  { 2
sin x (  1 )
2
 0 ( 1)
2
} dx
0
11 2
= 
8

40. V
15
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

CAS 142

2 2 1 x/2 2 2
= (3 x)  (3 xe ) dx =   2e +24e
0 3

 /2

2
41.  cos xdx describes the volume of the solid obtained by rotating the region
0

R= { ( x,y ) |0
x

2
,0
y
cos x } of the xy  plane about the x  axis.

5 5
  (
2
42.  ydy= y ) dy describes the volume of the solid obtained by rotating the region
2 2
R={ ( x,y ) |2
y
5,0
x
y } of the xy  plane about the y  axis.

1 1
43.  ( y  y ) dy=  ( y2) 2 ( y4) 2 dy describes the volume of the solid obtained by rotating the

4 8

0 0

region R= { ( x,y ) |0
y
1,y
x
y } of the xy  plane about the y  axis.
4 2

 /2

2 2
44.  ( 1+cos x )  1 dx describes the volume of the solid obtained by rotating the region
0

= {( x,y ) |0
x

2
,1
y
1+cos x } of the xy  plane about the x  axis.

Or: The solid could be obtained by rotating the region =
line y= 1 .
/
{ ( x,y ) |0
x

2
,0
y }
x about the

45. There are 10 subintervals over the 15 cm length, so we’ll use n=10/2=5 for the Midpoint Rule.
15
15 0
V =  A(x)dx M = [A(1.5)+A(4.5)+A(7.5)+A(10.5)+A(13.5)]
0
5 5
3
=3(18+79+106+128+39)=3 370=1110 cm

16
 Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

46.
10
10 0
V =  A(x)dx M = [A(1)+A(3)+A(5)+A(7)+A(9)]
0
5 5
3
=2(0.65+0.61+0.59+0.55+0.50)=2(2.90)=5.80m

r
47. We’ll form a right circular cone with height h and base radius r by revolving the line y= x about
h
the x  axis.
h


h 2 2 2 h
r r r 1 3

V = 2
x dx= x dx= x
0 h 2 2 3 0
0 h h
2
r 1 3 1 2
= h =  r h
2 3 3
h

r
Another solution: Revolve x= y+r about the y  axis.
h
h


h 2 2 2
r r 2r

V = 2 2
 y+r dy= y y+r dy
0 h 2 h
0 h
2 2 h
r 3 r 2 2 1 2 2 2 1 2
= y y +r y = r h r h+r h =  r h
2 h 3 3
3h 0

* r r
Or use substitution with u=r y and du= dy to get
h h
0 0
h h 1 3 h 1 3 1 2

2
 u  du=  u =   r =  r h.
r r r 3 r r 3 3

17
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

48.
h 2
R r
V =  R
h
y dy
0
h 2 2
2R(R r) R r

2
= R y+ y dy
0 h h

R(R r) 2 1
2 R r 2 3 h
= R y y+ y
h 3 h 0
2 1 2
= R h R(R r)h+ (R r) h
3
1 2 2 1 2 2
=  h 3Rr+(R  2Rr+r ) =  h(R +Rr+r )
3 3

H H h
Another solution: = by similar triangles. Therefore,
R r
hR
Hr=HR hR hR=H(R r) H= . Now
R r
1 2 1 2
V =  R H  r (H h) [by Exercise 47]
3 3
1 2 hR 1 2 rh rH rhR
=  R   r H h= =
3 R r 3 R r R R(R r)
3 3
1 R r 1
=  h
3 R r
2
=  h R +Rr+r
3
2
( )
18
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

1 1
=
3
2 2
 R + r + ( R
2
) ( r 2 ) h=
3
A +A +
1 2
A A
1 2
h

where A and A are the areas of the bases of the frustum. (See Exercise 50 for a related result.)
1 2

2 2 2 2 2 2
49. x +y =r  x =r  y
r 3 r
y
V =  (r  y ) dy=
2 2
r y
2
3 r h
r h

= { r 
r
3
3
2
3
 r (r h)
( r h )
3
3

}
= { 2 3 1
3
r 
3
2
(r h) 3r  ( r h )
2
}
1
= 
3
{ 2r 3 (r h) 2
(2
3r  r  2rh+h
2
)}
1
= 
3
{ 2r 3 ( r h ) 2r +2rh h
2 2
}
1
= 
3
( 2r 3 2r 3 2r 2h+rh2+2r 2h+2rh2 h3)
1
= 
3
( 3rh2 h3) = 13  2
h (3r h) , or, equivalently,  h
2
r
h
3

50. An equation of the line is

19
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

 x a/2 b/2 b a b b
x= y+(x  intercept )= y+ = y+ .
 y h 0 2 2h 2
h h
V = A(y)dy= (2x) dy
2

0 0
h 2 h 2
a b b a b
= 2 y+ dy= y+b dy
0 2h 2 0 h
h


2
( a b ) 2 2b(a b) 2
= y+ y+b dy
2 h
0 h
2 h
( a b ) 3 b(a b) 2 2
= y+ y +b y
2 h
3h 0
1 1 2
=
3
2 2
(a b) h+b(a b)h+b h=
3
2
a  2ab+b +3ab h ( )
1 2
=
3
( 2
a +ab+b h )

1
Note that this can be written as A +A + A A h , as in Exercise 48.
3 1 2 1 2
2
If a=b , we get a rectangular solid with volume b h . If a=0 , we get a square pyramid with volume
1 2
b h.
3

 /2 h y y
51. For a cross section at height y , we see from similar triangles that = , so  =b 1
b/2 h h
.
y
Similarly, for cross sections having 2b as their base and  replacing  ,  =2b 1 . So
h

20
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

h h
y y
V = A(y)dy= b 1 2b 1 dy
0 0 h h
h


h 2 2
y 2y y
= 2b
2 2
1 dy=2b 1 + dy
0 h h 2
0 h
2 3 h
y 2 y 2 1
=2b y + =2b h h+ h
h 2 3
3h 0
2 2 1
= b h [= Bh where B is the area of the base, as with any pyramid.]
3 3

52. Consider the triangle consisting of two vertices of the base and the center of the base. This
triangle is similar to the corresponding triangle at a height y , so a/b= /   =a /b . Also by similar
triangles, b/h= /(h y)
  =b(h y)/h . These two equations imply that  =a ( 1 y/h ) , and since the cross section is an
2 2
1 3 a ( 1 y/h )
equilateral triangle, it has area A(y)=     = 3 , so
2 2 4
h 2 h
a 3 y 2
V =  A(y)dy=
4
 1
h
dy
0 0
2 h
3
= a 3

h
1
y
=
3 2
a h( 1)=
3 2
ah
4 3 h 0 12 12

21
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

53. A cross section at height z is a triangle similar to the base, so we’ll multiply the legs of the base
triangle, 3 and 4, by a proportionality factor of (5 z)/5 . Thus, the triangle at height z has area
1 5 z 5 z z 2
A(z)=  3  4 =6 1 , so
2 5 5 5
5 5
 A(z)dz=6 ( 1 z/5) dz
2
V =
0 0
0
1
= 6 u ( 5du)
2
u=1 z/5,du= dz
1 5
1 3 0 1 3
=  30 u = 30  =10cm
3 1 3

54. A cross section is shaded in the diagram.

2
A(x)=(2y) = ( 2 )
2 2 2
r  x , so
r r
V =  A(x)dx=2 4 r  x dx ( 2 2)
 r 0

2 1 3 r
= 8 r x x
3 0
2 3 16 3
=8 r = r
3 3

22
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

55. If l is a leg of the isosceles right triangle and 2y is the hypotenuse,

2 2 2 2 2 2 2
then l +l = ( 2y )  2l =4y  l =2y .
2 2 2 2
1
V =  A(x)dx=2 A(x)dx=2 (l)(l)dx=2 y dx
2

 2 0 0 2 0
2 2
1 9
=2 (
36 9x dx=  4 x dx
2 2
) ( )
0 4 2 0
9 1 3 2 9 8
= 4x x = 8 =24
2 3 0 2 3

56. The cross section of the base corresponding to the coordinate y has length 2x=2 y . The
corresponding equilateral triangle with side s has area
2 3 2 3 2 3
A(y)=s = ( 2x ) =(2 y ) =y 3 . Therefore,
4 4 4
1 1 1
1 2 3
V = A(y)dy= y 3 dy= 3 y = .
0 0 2 0 2

57. The cross section of the base corresponding to the coordinate y has length 2x=2 y . The square
1 1
2 1
has area A(y)= ( 2 y ) =4y , so V = A(y) dy= 4ydy= 2y
2
=2 .
0
0 0

23
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

3
58. A typical cross section perpendicular to the y  axis in the base has length  (y)=3 y. This
2
length is the diameter of a cross sectional semicircle in S , so
2 2 2
  ( y) 2  3 2
V =  A(y)dy=
2 2
dy=
8
 3
2
y dy
0 0 0
0
 2 3 3

2
= u  du u=3 y,du= dy
8 3 3 2 2
 1 3 0  3
= u = ( 9)=
12 3 3 12 4

3
59. A typical cross section perpendicular to the y  axis in the base has length  (y)=3 y. This
2
length is the leg of an isosceles right triangle, so
1 2 1
A(y) =  (y) bh with base = height
2 2
1 1 2 9 1 2
= 3 1 y = 1 y
2 2 2 2
Thus,

2 0
9 2 1 1
V =  A(y)dy=  u ( 2du) u=1 y,du= dy
0 2 1 2 2
1 3 0 1
= 9 u = 9  =3
3 1 3

24
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

r r r r
60. (a) V = A(x)dx=2 A(x)dx=2
1
h 2 r  x dx=2h r  x dx
2 2 2 2
( )
 r 0 0 2 0
(b) Observe that the integral represents one quarter of the area of a circle of radius r , so
1 2 1 2
V =2h  r =  hr .
4 2

2 2 2
61. (a) The torus is obtained by rotating the circle ( x R ) +y =r
about the y axis. Solving for x , we see that the right half of the
2 2 2 2
circle is given by x=R+ r  y = f (y) and the left half by x=R r  y =g(y) . So
r
V =  { f (y)  g(y)
2 2
} dy
 r
r
= 2  ( R +2R 2 2 2
r  y +r  y 
2 2
) ( R  2R
2 2 2
r  y +r  y
2 2
) dy
0
r r
 4R r  y dy=8 R
2 2 2 2
= 2 r  y dy
0 0

(b) Observe that the integral represents a quarter of the area of a circle with radius r , so
r
1 2
8 R
2 2 2 2
r  y dy=8 R  r =2 r R.
0 4

62. The cross sections perpendicular to the y axis in Figure 17 are rectangles. The rectangle
2
corresponding to the coordinate y has a base of length 2 16 y in the xy plane and a height of
1  1 2 2
y , since BAC=30 and BC = AB . Thus, A(y)= y 16 y and
3 3 3
4 4
2
V =  A(y)dy= 
2
16 y ydy
0 3 0

25
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

0
2 1

1/2 2
= u  du[Putu=16 y ,sodu= 2ydy]
3 16 2
16
1 1 2 3/2 16 2 128

1/2
= u du= u = ( 64 ) =
3 0 3 3 0 3 3 3 3

h
63. (a) Volume S = A(z)dz= Volume S ( ) ( ) since the cross sectional area A(z) at height z is the
1 2
0
same for both solids.
(b) By Cavalieri’s Principle, the volume of the cylinder in the figure is the same as that of a right
2
circular cylinder with radius r and height h , that is,  r h .
64. Each cross section of the solid S in a plane perpendicular to the x axis is a square (since the
edges of the cut lie on the cylinders, which are perpendicular). One quarter of this square and one 
eighth of S are shown. The area of this quarter  square is PQ =r  x . Therefore, A(x)=4 r  x
2 2 2
( 2 2)
and the volume of S is
r r
V =  A(x)dx=4 (r 2 x2) dx
 r  r
r r
1 3 16 3
= 8 (r 2 x2) dx=8 2
r x
3
x =
0 3
r
0

65. The volume is obtained by rotating the area common to two circles of radius r , as shown. The
volume of the right half is
r/2 r/2 2
1
 
2 2
V =  y dx= r  r+x dx
right 2
0 0

26
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

3 r/2
2 1 1 1 3 1 3 1 3 5 3
=  r x r+x r  r  = 0 r =  r
3 0 2 2 3 24 24
5 3
So by symmetry, the total volume is twice this, or  r .
12

Another solution: We observe that the volume is the twice the volume of a cap of a sphere, so we can
1
use the formula from Exercise 49 with h= r :
2
1 2 2 1 2 1 5 3
V =2  h (3r h)=  r 3r r =  r .
3 3 2 2 12

66. We consider two cases: one in which the ball is not completely submerged and the other in which
it is.
Case 1: 0
h
10 The ball will not be completely submerged, and so a cross section of the water
parallel to the surface will be the shaded area shown in the first diagram. We can find the area of the
cross section at height x above the bottom of the bowl by using the Pythagorean Theorem:
2 2 2 2 2
R =15  ( 15 x ) and r =5  ( x 5) , so A(x)=
2
( R2 r 2) =20 x . The volume of water when it has
h h
2 h
depth h is then V (h)= A(x)dx= 20 xdx= 10 x
2 3
=10 h cm , 0
h
10 .
0
0 0
Case 2: 10<h
15 In this case we can find the volume by simply subtracting the volume displaced by
the ball from the total volume inside the bowl underneath the surface of the water. The total volume
underneath the
surface is just the volume of a cap of the bowl, so we use the formula from Exercise 49:
1 2
V (h)=  h (45 h) . The volume of
cap 3
4 3 500
the small sphere is V =  (5) =  ,
ball 3 3
1 2 3
so the total volume is V  V =  45h  h  500 cm .
cap ball 3
3
( )

27
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

67. Take the x axis to be the axis of the cylindrical hole of radius r . A quarter of the cross section
through y, perpendicular to the y axis, is the rectangle shown. Using the Pythagorean Theorem
2 2 2 2
twice, we see that the dimensions of this rectangle are x= R  y and z= r  y , so
1 2 2 2 2
A(y)=xz= r  y R  y , and
4
r r
 A(y)dy= 4 r  y
2 2 2 2
V = R  y dy
 r  r
r
= 8
2 2 2 2
r  y R  y dy
0

2 2 2 2 2 2 2
68. The line y=r intersects the semicircle y= R  x when r= R  x  r =R  x 
2 2 2 2 2
x =R  r  x= R  r . Rotating the shaded region about the x  axis gives us
2 2
R r 2
V =   ( R x
2 2
)r 2
dx
2 2
 R r

28
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

2 2
R r
= 2  ( R2 x2 r 2) dx [by symmetry]
0
2 2
R r 2 2
1 3 R r
= 2  (R  r ) x
2 2 2
dx=2 (R  r )
2 2
x
3
x
0
0

= 2 ( R2 r 2) 3/2 13 ( R2 r 2) 3/2

2
= 2 
3
2 2
R r ( ) 3/2= 43 ( R2 r 2) 3/2
4 3
Our answer makes sense in limiting cases. As r 0 ,V
 R , which is the volume of the full
3
sphere. As r R ,V 0 , which makes sense because the hole’s radius is approaching that of the
sphere.

2
69. (a) The radius of the barrel is the same at each end by symmetry, since the function y=R cx is
even. Since the barrel is obtained by rotating the graph of the function y about the x axis, this radius
1 1 2
is equal to the value of y at x= h , which is R c h =R d=r .
2 2

29
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.2 Volumes

(b) The barrel is symmetric about the y axis, so its volume is twice the volume of that part of the
barrel for x>0 . Also, the barrel is a volume of rotation, so
h/2 h/2 h/2
2 3 1 2 5
V = 2   y dx=2
2
 ( R cx2) 2dx=2 2
R x
3
Rcx + c x
5 0
0 0
1 2 1 3 1 2 5
= 2 R h Rch + ch
2 12 160
Trying to make this look more like the expression we want, we rewrite it as
1 2 2 1 2 3 2 4
V =  h 2R + R  Rch + ch . But
3 2 80
2 1 2 3 2 4 1 2 2 1 2 4 2 2 1 2 2 2 2 2
R Rch + c h = R ch  c h = ( R d )  ch =r  d .
2 80 4 40 5 4 5
1 2 2 2 2
Substituting this back into V , we see that V =  h 2R +r  d , as required.
3 5

70. It suffices to consider the case where is bounded by the curves y= f (x) and y=g(x) for a
x
b ,
where g(x)
f (x) for all x in a,b , since other regions can be decomposed into subregions of this
type. We are concerned with the volume obtained when is rotated about the line y= k , which is equal
to
b b b
V = 
2  ( 2
f (x)+k  g(x)+k
2
) dx=  ( 2
f (x)  g(x)
2
) dx+2 k f (x) g(x) dx
a a a

= V 1+2 kA

30
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

1.

If we were to use the "washer’’ method, we would first have to locate the local maximum point (a,b)
2 2
of y=x ( x 1 ) using the methods of Chapter 4. Then we would have to solve the equation y=x ( x 1 )
for x in terms of y to obtain the functions x=g (y) and x=g (y) shown in the first figure. This step
1 2
would be difficult because it involves the cubic formula. Finally we would find the volume using
b
V = 
0
{ g (y)
1
2
 g (y)
2
2
} dy .
Using shells, we find that a typical approximating shell has radius x , so its circumference is 2 x . Its
2
height is y , that is, x(x 1) . So the total volume is

1 1 5 4 3 1
x x x 
V = 2 x x(x 1) dx=2
2
 ( x  2x +x ) dx=2
4 3 2
5
 2 +
4 3 0
=
15
0 0

2.


A typical cylindrical shell has circumference 2 x and height sin x ( ) .V= 
2
( 2)
2 xsin x dx . Let
0
1
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells




2
u=x . Then du=2xdx , so V = sin udu=  cos u = [1 (  1 ) ]=2 .
0
0

For slicing, we would first have to locate the local maximum point ( a,b) of y=sin x ( 2) using the
methods of Chapter 4. Then we would have to solve the equation y=sin x for x in terms of y to ( 2)
obtain the functions x=g (y) and x=g (y) shown in the second figure. Finally we would find the
1 2
volume using . Using shells is definitely preferable to slicing.

3.
2 2
1
V = 2 x dx=2  1dx
1 x 1
2
=2 x =2 (2 1)=2
1

4.
1 1
V = 2 x x dx=2 
2 3
x dx
0 0
1 4 1 1 
=2 x =2  =
4 0 4 2

2
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

2
 x
5. V = 2 xe
1 2
dx . Let u=x . Thus, du=2xdx , so
0

1  u  u 1
V =  0
e du=  e
0
= (1 1/e)

6.
3 3
V =2  { x[(3+2x x ) (3 x)]} dx=2
2
 x(3x x ) dx
2

0 0
3 3
1 4 81 27 27

2 3 3
=2 (3x  x )dx=2 x x =2 27 =2 =
0 4 0 4 4 2

2 2 2 2 2
7. The curves intersect when 4(x 2) =x  4x+7 4x  16x+16=x  4x+7 3x  12x+9=0
2
3(x  4x+3)=0 3(x 1)(x 3)=0 , so x=1 or 3 .
3 3
V =2  { x ( x  4x+7)  4(x 2) } dx=2
2 2

2 2
x(x  4x+7 4x +16x 16) dx
1 1

3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

3 3 3
1 4 4 3 3 2
 x( 3x +12x 9) dx=2 ( 3) (x  4x +3x)dx= 6
2 3 2
=2 x x+ x
1 1 4 3 2 1

81 27 1 4 3 4 8
= 6  36+   + = 6 20 36+12+ = 6  =16
4 2 4 3 2 3 3

8. By slicing:
1 1 1
2 2 1 2 1 5 1 1 3
V =  ( y) 
2
(y ) dy=  ( y y4) dy= 2
y
5
y =
0

2 5
=
10
0 0

By cylindrical shells:
1 1
V = 2 x ( 2
x  x dx=2 )  (x
3/2 3
 x )dx
0 0

2 5/2 1 4 1 2 1
=2 x  x =2 
5 4 0 5 4

4
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

3 3
=2 =
20 10

9.
2 2 2
1 2 1 4
(
V = 2 y 1+y dy=2
2
)  ( y+y3) dy=2 2
y+ y
4 1
1 1
1 1 21 21
=2 ( 2+4 )  + =2 =
2 4 4 2

10.
1 1
V = 2 y y dy=2 
3/2
y dy
0 0

2 5/2 1 4
=2 y =
5 0 5

5
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

11.
8
V =2  3
y( y  0) dy
0
8 8
3 7/3

4/3
=2 y dy=2 y
0 7 0

6 7/3 6 7 768
= (8 )= (2 )=
7 7 7

12.
4

2 3
V =2 y(4y  y ) dy
0
4

3 4
=2 (4y  y )dy
0

6
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

4 1 5 4 1024
=2 y y =2 256
5 0 5
256 512
=2 =
5 5

2 2 3
13. The curves intersect when 4x =6 2x 2x +x 3=0 (2x+3)(x 1)=0 x= or 1 . Solving the
2
2 1 1
equations for x gives us y=4x  x= y and 2x+y=6 x= y+3 .
2 2

=2 
4
0 { y
1
2
y  
1
2
y } dy+2 
9
4 { y 
1
2
y+3  
1
2
y } dy
1 2 1 3/2 2 5/2 4 1 3 3 2 1 5/2 9
 0 (y 
V 4 9
=2 y ) dy+2 
y +3y+ y dy=2 y +2  y+ y+ y
4 2 2 5 0 6 2 5 4
2 243 243 243 32 32
=2  32 +2  + +   +24+
5 2 2 5 3 5
128 433 1250 250
=  +2 =  = 
5 15 15 3

14.

7
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

3
V = 2 y 4 ( y 1 )  ( 3 y ) dy
2

0
3
=2  (
y  y +3y dy
2
)
0
3 3
1 4 3
=2  (  y3+3y2) dy=2 
4
y +y
0
0
81 27 27
=2  +27 =2 =
4 4 2

15.
2 2
1 4 1 3
V = 2 (x 1)x dx=2
2
x x
1 4 3 1

8 1 1 17
=2 4   = 
3 4 3 6

16.

8
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

 1  1
1 4
V =  2 ( x) x dx=2
2
 x
4  2
 2
1 15
=2   (  4) = 
4 2

17.
2 2
4 3 1 4
V = 2 (4 x)x dx=2
2
x x
1 3 4 1

32 4 1 67
=2  4   = 
3 3 4 6

18.
4
V = 2 [x ( 2)][(8x 2x ) (4x x )]dx
2 2

0
4
= 2 (2+x)(4x x )dx
2

0
4

2 3
=2 (8x+2x  x )dx
0

2 3 1 4 4
2
=2 4x + x x
3 4 0
128 256
=2 64+  64 = 
3 3
9
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

19.
2
V = 2 (3 y)(5 x)dy
0
2
(
= 2 ( 3 y ) 5 y  1 dy
2
)
0
2
= 2 ( 12 4y 3y2+y3) dy
0

2 3 1 4 2
=2 12y 2y  y + y
4 0
=2 (24 8 8+4)=24

20.
1 1
V = 2 ( y+1 ) ( 2
)
y  y dy=2 
3/2 1/2
(y +y  y  y )dy
3 2

0 0

2 5/2 2 3/2 1 4 1 3 1 2 2 1 1 29 29

=2 y + y  y y =2 +   =2 =
5 3 4 3 0 5 3 4 3 60 30

10
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

21. V = 2 xln xdx

2
1

3
22. V = 2 ( 7 x ) ( 4x x2)  x dx
0

1

23. V = 2
4
x ( 1) sin x x dx
0 2


24. V = 2 (2 x)
1
2
dx
0 1+x

11
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells


25. V = 2 (4 y) sin y dy
0

3
26. V = 2 (5 y) 4 ( 2
)
y +7 dy
 3

 /4 0 
27.  x= = .
4 16
 /4
   3 3 5 5 7 7
V =  2 xtan xdx 2  tan + tan + tan + tan 1.142
0 16 32 32 32 32 32 32 32 32

12 2 *
28.  x= =2 , n=5 and x =2+(2i+1) , where i=0 , 1 , 2 , 3 , 4 . The values of f (x) are taken
5 i
directly from the diagram.
12
V =  2 xf (x)dx 2 3 f (3)+5 f (5)+7 f (7)+9 f (9)+11 f (11)  2
2
2 3(2)+5(4)+7(4)+9(2)+11(1) 2=332
12
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

3 3
29.  2 x dx=2 
5 4 4
x(x )dx . The solid is obtained by rotating the region 0
y
x , 0
x
3 about the
0 0
y  axis using cylindrical shells.

2 2

30. 2  y
2
dy=2  y
1
2
dy . The solid is obtained by rotating the region 0
x

1
2
,
1+y 0 0 1+y 1+y
0
y
2 about the x  axis using cylindrical shells.

1
31.  2 (3 y)(1 y )dy . The solid is obtained by rotating the region bounded by (i) x=1 y , x=0 , and
2 2

0
2
y=0 or (ii) x=y , x=1 , and y=0 about the line y=3 using cylindrical shells.

 /4
32.  2 (  x)(cos x sin x)dx . The solid is obtained by rotating the region bounded by
0
 
(i) 0
y
cos x sin x , 0
x
or (ii) sin x
y
cos x , 0
x
about the line x= using
4 4
cylindrical shells.

33.

2 4
From the graph, the curves intersect at x=0 and at x=a 1.32 , with x+x  x >0 on the interval ( 0,a ) .
So the volume of the solid obtained by rotating the region about the y axis is

a a
 
2 4 2 3 5
V = 2 x(x+x  x ) dx=2 (x +x  x )dx
0 0

1 3 1 4 1 6 a
= 2 x+ x x 4.05
3 4 6 0

34.

13
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

3 4
From the graph, the curves intersect at x=0 and at x=a 1.17 , with 3x x >x on the interval ( 0,a ) .
So the volume of the solid obtained by rotating the region about the y axis is
a a
V = 2  { x[(3x x ) x ]} dx=23 4

2 4
(3x  x  x )dx
5

0 0

3 1 5 1 6 a
= 2 x x x 4.62
5 6 0

35.
 /2

V =2  2
 x ( sin 2x sin 4x) dx
0
1 3
= 
32

36.

V =2  { [x ( 1)](x3sin x)} dx=2 (
4
+
3
 12
2
 6 +48)
0
5 4 3 2
=2 +2  24  12 +96

14
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

37. Use disks:

1 1
2
V =   ( x +x 2)
2
dx=  ( x4+2x3 3x2 4x+4) dx
 2  2

1 5 1 4 3 2 1 1 1 32
= x + x  x  2x +4x = +  1 2+4   +8+8 8 8
5 2  2 5 2 5
33 3 81
= + = 
5 2 10

38. Use shells:

2 2
V =  2 x  x +3x 2 dx=2 ( 2
)  (  x3+3x2 2x) dx
1 1

1 4 3 2 2 1 
= 2  x +x  x =2 (  4+8 4 )   +1 1 =
4 1 4 2

39. Use shells:

V =  2
4
1
x ( 1) 5 (x+4/x) dx =2 
4
(x+1)(5 x 4/x)dx =2
1
 (1 5x x2 4+5 x 4/x) dx =
4

1 3 2 4
2  (1  x2+4x+1 4/x) dx=2
4

3
x +2x +x 4ln x
1
=
64 1
2  +32+4 4ln 4   +2+1 0 =2 (12 4ln 4)=8 (3 ln 4)
3 3

15
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

40. Use washers:

1
V =   { 2 0 2
 2 1 y ( 4
) 2} dy
 1
1
4 2
= 2  (
4 1+y ) dy [by symmetry]
0
1 1
= 2  (
4 1+2y +y
4 8
) dy=2  ( 3 2y4 y8) dy
0 0

2 5 1 9 1 2 1 112 224
= 2 3y y y =2 3  =2 =
5 9 0 5 9 45 45

2 2 2 2
1 3 8 4
41. Use disks: V =  1 ( y 1 )
2
dy=  ( 2y y2) dy= 2
y
3
y =
0
4
3
= 
3
0 0

42. Using shells, we have

2
V = 2 y 1 (y 1) 
2
( 1 (y 1)
2
) dy
0
2 1
 
2 2
=2 y 2 1 ( y 1 ) dy=4 (u+1) 1 u du [let u=y 1]
0  1
1 1
 
2 2
=4 u 1 u du+4 1 u du
 1  1

The first definite integral equals zero because its integrand is an odd function. The second is the area
  2
of a semicircle of radius 1 , that is, . Thus, V =4  0+4  =2 .
2 2

43.
16
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

r r r
2 2 2
V =2 2 x r  x dx= 2
2 2
 (r 2 x2) 1/2 (  2x ) dx=  2 
3
(r  x ) 3/2 0
0 0
4
=
3
 ( 0 r 3) = 43  r
3

44.
R+r
V =  2 x 2 r  ( x R ) dx
2 2

R r
r
= 4 (u+R) r  u du [u=x R ]
2 2

 r
r r
=4 R 
2 2 2 2
r  u du+4 u r  u du
 r  r

1 2
The first integral is the area of a semicircle of radius r , that is,  r ,
2
1 2 2
and the second is zero since the integrand is an odd function. Thus, V =4 R  r +4  0=2 Rr
2
.

r r 2 3 2 r 2 2
h x x x r  r h
45. V =2  x  x+h dx=2 h  +x dx=2 h  + =2 h =
0 r 0 r 3r 2 0 6 3

46. By symmetry, the volume of a napkin ring obtained by drilling a hole of radius r through a sphere
with radius R is twice the volume obtained by rotating the area above the x axis and below the curve
2 2
y= R  x (the equation of the top half of the cross section of the sphere), between x=r and x=R ,
about the y axis.
This volume is equal to

17
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.3 Volumes by Cylindrical Shells

outerradius R R
1 2 2 4
2  2 rhdx=2 2  x
2 2
R  x dx=4 
3
(R x ) 3/2 = 
r 3
( R2 r 2) 3/2
innerradius r

2 2 1 2
But by the Pythagorean Theorem, R  r = h , so the volume of the napkin ring is
2
4 1 3 1 3
 h =  h , which is independent of both R and r ; that is, the amount of wood in a
3 2 6
napkin ring of height h is the same regardless of the size of the sphere used. Note that most of this
calculation has been done already, but with more difficulty, in Exercise 6.2.68.
Another solution: The height of the missing cap is the radius of the sphere minus half the height of the
1
cut out cylinder, that is, R h . Using Exercise 6.2.49,
2

4 3 2  1 2 1 1 3
V =V V  2V =  R   r h 2 R h 3R R h =  h
napkinring sphere cylinder cap 3 3 2 2 6

18
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

1. By Equation 2, W =Fd=(900)(8)=7200 J.
2. F=mg=(60)(9.8)=588 N; W =Fd=588 2=1176 J

3.
9 10

 
b
W = f (x)dx= 10 1
dx=10 du [u=1+x,du=dx]
2 2
a 0 ( 1+x ) 1 u
1 10 1
=10  =10  +1 =9ft lb
u 1 10

2 2
1 3 1 3 3 3
4. W = cos  x dx= sin  x =  =0 N m =0 J.
1 3  3 1  2 2
3/2
3 1 3 3
Interpretation: From x=1 to x= , the force does work equal to
2
 cos
3
 x dx=

1
2
1
3
J in accelerating the particle and increasing its kinetic energy. From x= to x=2 , the force opposes
2
the motion of the particle, decreasing its kinetic energy. This is negative work, equal in magnitude but
3
opposite in sign to the work done from x=1 to x= .
2

5. The force function is given by F(x) (in newtons) and the work (in joules) is the area under the
8 4 8
1
curve, given by  F(x)dx= F(x)dx+ F(x)dx= (4)(30)+(4)(30)=180 J.
0 0 4 2

20
20 4
6. W =  f (x)dx M = x[ f (6)+ f (10)+ f (14)+ f (18)]= [5.8+8.8+8.2+5.2]=4(28)=112 J
4
4 4

1 1 1
7. 10= f (x)=kx= k [ 4 inches = foot], so k=30 lb / ft and f (x)=30x . Now 6 inches = foot, so
3 3 2
1/2
2 1/2 15
W =  30xdx= 15x = ft lb.
0
0 4

8. 25= f (x)=kx=k(0.1) [10 cm =0.1 m], so k=250 N / m and f (x)=250x . Now 5 cm =0.05 m, so
0.05
2 0.05
W=  250xdx= 125x
0
=125(0.0025)=0.3125 0.31 J.
0

1
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

0.12
1 2 0.12 1 2 2500
9. If  kxdx=2 J, then 2= kx = k(0.0144)=0.0072k and k= =  277.78 N /
0 2 0 2 0.0072 9
m. Thus, the work needed to stretch the spring from 35 cm to 40 cm is
0.10 1/10
2500 1250 2 1250 1 1 25
 xdx= x =  =  1.04 J.
0.05 9 9 1/20 9 100 400 24

1 1
1 2 1
10. If 12= kxdx= kx = k , then k=24 lb / ft and the work required is
0 2 0 2
3/4
2 3/4 9 27
 24xdx= 12x
0
=12 =
16 4
=6.75 ft lb.
0

2500 270
11. f (x)=kx , so 30= x and x= m =10.8 cm
9 2500

12. Let L be the natural length of the spring in meters. Then

0.12 L 0.12 L
1 2 1

2 2
6= kxdx= kx = k ( 0.12 L )  ( 0.10 L ) and
0.10 L 2 0.10 L 2
0.14 L
1 2 0.14 L 1

2 2
10= kx kxdx= = k ( 0.14 L )  ( 0.12 L ) . Simplifying gives us
0.12 L 2 0.12 L 2
12=k(0.0044 0.04L) and 20=k(0.0052 0.04L) . Subtracting the first equation from the second gives
32
8=0.0008k , so k=10 , 000 . Now the second equation becomes 20=52 400L , so L= m =8 cm.
400

1
13. (a) The portion of the rope from x ft to ( x+ x ) ft below the top of the building weighs  x lb
2
* 1 *
and must be lifted x ft, so its contribution to the total work is x  x ft lb. The total work is
i 2 i
50
1 * 1 1 2 50 2500
W =lim x  x= 
n
xdx= x = =625 ft lb Notice that the exact height of the
i=1 2 i
0 2
n  4 0 4
building does not matter (as long as it is more than 50 ft).
(b) When half the rope is pulled to the top of the building, the work to lift the top half of the rope is
25 25
1 1 2 625
W = xdx= x = ft lb. The bottom half of the rope is lifted 25 ft and the work needed
1
0 2 4 0 4
50
1 25 625
to accomplish that is W = 
50
 25dx= x = ft lb. The total work done in pulling half the
2
25 2 2 25 2
2
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

625 625 3 1875

rope to the top of the building is W =W +W = + =  625= ft lb.
1 2 2 4 4 4

14. Assumptions : 1. After lifting, the chain is L shaped, with 4 m of the chain lying along the ground.
2. The chain slides effortlessly and without friction along the ground while its end is lifted.
3. The weight density of the chain is constant throughout its length and therefore equals
2
(8kg/m)(9.8m/s )=78.4N/m .
The part of the chain xm from the lifted end is raised 6 xm if 0
x
6 m, and it is lifted 0 m if x>6m
.
Thus, the work needed is
6 6
1 2
(6 x ) 78.4 x= (6 x)78.4dx=78.4
n *
W =lim 6x x =(78.4)(18)=1411.2 J.
n 
i=1 i
0 2 0

500
2 500
2x  x=  2xdx= x
n *
15. The work needed to lift the cable is lim =250 , 000 ft lb. The work
i=1 i 0
n  0
needed to lift the coal is 800 lb  500 ft =400 , 000 ft lb. Thus, the total work required is 250 ,
000+400 , 000=650 , 000 ft lb.

16. The work needed to lift the bucket itself is 4 lb  80 ft =320 ft lb. At time t (in seconds) the bucket
*
is x =2t ft above its original 80 ft depth, but it now holds only ( 40 0.2t ) lb of water. In terms of
i
1 * *
distance, the bucket holds 40 0.2 x lb of water when it is x ft above its original 80 ft
2 i i
1 *
depth. Moving this amount of water a distance  x requires 40 x  x ft lb of work. Thus, the
10 i
work needed to lift the water is
80 80
1 * 1 1 2
W =lim x  x= 
n
40 40 x dx= 40x x = ( 3200 320 ) ft lb Adding the
n 
i=1 10 i 0 10 20 0
work of lifting the bucket gives a total of 3200 ft lb of work.

17. At a height of x meters ( 0

x
12 ), the mass of the rope is (0.8 kg / m )(12 x m )=(9.6 0.8x) kg
36
and the mass of the water is kg/m (12 xm)=(36 3x) kg. The mass of the bucket is 10 kg, so
12
the total mass is (9.6 0.8x)+(36 3x)+10=(55.6 3.8x) kg, and hence, the total force is 9.8(55.6 3.8x)
N.
*
The work needed to lift the bucket  x m through the i th subinterval of 0,12 is 9.8(55.6 3.8x ) x ,
i
so the total work is

3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

12
2 12
9.8(55.6 3.8x ) x=  (9.8)(55.6 3.8x)dx=9.8 55.6x 1.9x
n *
W =lim
i=1 i 0
n  0
=9.8(393.6) 3857 J

25lb
18. The chain’s weight density is =2.5 lb / ft. The part of the chain x ft below the ceiling (for
10ft
5
x
10 ) has to be lifted 2(x 5) ft, so the work needed to lift the i th subinterval of the chain is 2 (
*
x  5)(2.5 x) . The total work needed is
i
10 10
2(x  5)(2.5) x=  [2(x 5)(2.5)]dx=5 (x 5)dx
n *
W =lim
i=1 i
n  5 5

1 2 10 25 25
=5 x  5x =5 (50 50)  25 =5 =62.5 ft lb
2 5 2 2

* * 1
19. A ‘‘slice’’ of water  x m thick and lying at a depth of x m (where 0
x
) has volume
i i 2
3
( 2 1  x ) m , a mass of 2000 x kg, weighs about (9.8)(2000 x)=19 , 600 x N, and thus
19 , 600x  x=
* n * 1/2
requires about 19 , 600x  x J of work for its removal. So W =lim 19 ,
i i=1 i 0
n 
2 1/2
600xdx= 9800x =2450 J.
0

2 2 3
20. A horizontal cylindrical slice of water  x ft thick has a volume of  r h=  12   x ft and
weighs about 62.5lb/ft ( 3
) ( 144 3
) *
 xft =9000  x lb. If the slice lies x ft below the edge of the
i
* *
pool (where 1
x
5 ), then the work needed to pump it out is about 9000 x  x . Thus,
i i

5
2 5
9000 x  x= 9000 xdx= 4500 x
n *
W =lim =4500 (25 1)=108,000 ft lb
i=1 i 1
n  1

21. A rectangular ‘‘slice’’ of water  x m thick and lying x ft above the bottom has width x ft and
3
volume 8x x m . It weighs about ( 9.8 1000 ) ( 8x x ) N, and must be lifted ( 5 x ) m by the pump,
so the work needed is about 9.8 10 ( 3
) (5 x)(8x x) J. The total work required is

4
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

3 3 3
8 3
W   ( 9.8 10
3
) (5 x)8xdx= ( 9.8 3
10 )  ( 40x 8x2) dx= ( 9.8 10
3
) 20x 
2
3
x
0
0 0

= ( 9.8 10
3
) (180 72)= ( 9.8 3
10 ) (108)=1058.4 3
10  1.06 10 J
6

22. For convenience, measure depth x from the middle of the tank, so that  1.5
x
1.5 m. Lifting a
slice of water of thickness  x at depth x requires a work contribution of
 W ( 9.8 103) ( 2 ( 1.5) 2 x2 ) (6 x)(2.5+x) , so
1.5
 ( 9.8 10 ) 12 2.25 x (2.5+x)dx
3 2
W 
 1.5
3/2 3/2
9 2 9 2
= ( 9.8 10
3
) 60 
4
 x dx+12  x
4
 x dx
0  3/2
The second integral is 0 because its integrand is an odd function, and the first integral represents the
area of a quarter circle of radius
3/2 2
3 9 2 1 3
2
.Therefore,[W  ( 9.8 3
10 ) 60  4
3
(
 x dx= 9.8 10 ( 60 ) ) 4

2
6
=330,750  1.04 10 J]
0

23. Measure depth x downward from the flat top of the tank, so that 0
x
2 ft. Then
 W = ( 62.5) 2 4 x ( 2
) (8 x)(x+1) ft lb, so
2 2 2
(62.5)(16) (x+1)  4 x dx+
2 2 2
W  4 x dx=1000 x 4 x dx
0 0 0
4
1 1
= 1000  u
1/2
2
du+ 
4
( 22) 2
[Put u=4 x ,so du= 2xdx]
0
1 2 3/2 4 8 3
= 1000  u + =1000 +  5.8 10 ft lb
2 3 0 3
Note: The second integral represents the area of a quarter circle of radius 2 .

2
24. Let x be depth in feet, so that 0
x
5 . Then  W =(62.5) ( 2
5 x
2
) x x ft lb and
5 5
25 2 1 4 625 625 625
 x ( 25 x ) dx=62.5
2
W  62.5 x x =62.5  =62.5
0 2 4 0 2 4 4
4
 3.07 10 ft lb

5
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

5
25. If only 4.7 10 J of work is done, then only the water above a certain level (call it h ) will be
pumped out. So we use the same formula as in Exercise 21 , except that the work is fixed, and we are
trying to find the lower limit of integration:
3 3
8 3
4.7 10 
5
 ( 9.8 10 ) (5 x)8xdx= ( 9.8
3
10 )
3
20x 
3
2
x
h

h
4.7 2 2 8 3 2 8 3 3 2
 10  48= 20 3   3  20h  h  2h  15h +45=0 . To find the solution of this
9.8 3 3
3 2
equation, we plot 2h  15h +45 between h=0 and h=3 . We see that the equation is satisfied for h 2.0
. So the depth of water remaining in the tank is about 2.0 m.

3/2 3/2 3/2

9 2 5 9 2 9 2
26. W  ( 9.8 920 )  12
4
 x
2
+x dx=9016 30 
4
 x dx+12  x
4
 x dx .
0 0 0
3/2 2 3/2 9/4
9 2 1 3 9 9 2 1 1/2 9 2
Here  4
 x dx= 
4 2
=
16
and  x
4
 x dx=  u du [ where u=  x , so
0 2
0 0 4
1 3/2 9/4 1 27 9
du= 2xdx ] = u = = , so
3 0 3 8 8
9 9 135 27 5
W  9016 30  +12 =9016  +  6.00 10 J.
16 8 8 2

2 2
27. V = r x , so V is a function of x and P can also be regarded as a function of x . If V = r x and
1 1
2
V = r x , then
2 2
x x
2 2

 F(x)dx=  r P(V (x))dx

2
W =
x x
1 1

x
2


2 2
= P(V (x))dV (x)[LetV (x)= r x,sodV (x)= r dx.]
x
1

6
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.4 Work

V
2
=  P(V )dV by the Substitution Rule.
V
1

2 2 100
3 3 3 800 3
28. 160 lb / in =160 144 lb / ft , 100 in = ft , and 800 in = ft .
1728 1728
1.4 100 1.4 25 1.4  1.4
k=PV = ( 160 144 ) =23 , 040  426.5 . Therefore, P 426.5V and
1728 432
800/1728 25/54
 1.4 1 

0.4
W = 426.5V dV =426.5 V
100/1728  0.4 25/432

432 0.4 54 0.4

= (426.5)(2.5) 
25 25
3
 1.88 10 ft lb


b mm  1 b 1 1
29. W = F(r)dr= G
1 2
dr=Gm m =Gm m 
a
2 1 2 r a 1 2 a b
a r

1 1
30. By Exercise 29 , W =GMm  where M= mass of Earth in kg, R= radius of
R R+1,000,000
Earth in m, and m= mass of satellite in kg. (Note that 1000 km =1 , 000 , 000 m.) Thus,

 11
(
W = 6.67 10 ) ( 5.98 10
24
) (1000) 1
6

1
6
9
 8.50 10 J
6.37 10 7.37 10

7
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

b 1 1 1
1 1 1 1 3 1
 f (x)dx= 1 ( 1)  x dx= 2  2 x dx=
2 2
1. f = x =
ave b a 3 0 3
a  1 0

1 4 1 4 1
2. f
ave
= 
4 1 1
(1/x)dx=
3
ln x = ln 4 0.46
1 3

 /2
1 2  /2 2 2
3. g =
ave 
 cos xdx=

sin x
0
=

(1 0)=

 0 0
2

4.
2 9
1 1 1
 
2 3 3 2
g = x 1+x dx= u du [ u=1+x , du=3x dx ]
ave 2 0 0 2 3
1
1 2 3/2 9 1 26
= u = ( 27 1 ) =
6 3 1 9 9
2
1 5 t 1  25 u 1 1
 te dt= 
2
5. f = e  du [ u= t , du= 2t dt , t dt= du ]
ave 5 0 0 5 0 2 2
1 u  25 1  25 1  25
=
10
e
0
=
10
e  1 =
10
1 e ( ) ( )
1  /4 4  /4 4
6. f
ave
=

 0
sec  tan  d =

sec 
0
=

( 2  1)
 0
4

7.
  1
1 1
 
4 4
h = cos xsin xdx= u ( du) [ u=cos x , du= sin xdx ]
ave   0 0  1
1 1 1
1 1 2 1 5 2
  2 u du=
4 4
= u du= u =
  1  0  5 0 5

8.
6


7
1 3 1  2
dr=  3u du [ u=1+r , du=dr ]
h
ave =
6 1 2 52
1 ( 1+r )

1
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

3  1 7 3 1 1 3 1 1 3 5 3
= u =  =  =  =
5 2 5 7 2 5 2 7 5 14 14

9.
(a) 1
5
1 1 5

2 3
f = (x 3) dx= (x 3)
ave 5 2 2 3 3 2

1 3 3 1
= 2  ( 1) = (8+1)=1
9 9
(b) 2
f (c)= f  (c 3) =1 c 3= 1
ave
 c=2 or 4
(c)

10.
(a) 4 4
1 1 2 3/2
f
ave =
4 0
 x dx=
4 3
x
0 (b) 4 16
0 f (c)= f  c=  c=
1 3/2 4 1 4
ave 3 9
= x = [8 0]=
6 0 6 3
(c)

11.

2
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

(a) 
1
f
ave =  (2sin x sin 2x)dx
  0 0
(b) 4
 f (c)= f  2sin c sin 2c= 
1 1 ave 
=  2cos x+ cos 2x c  1.238 or c  2.808
 2 0 1 2
1 1 1 4
= 2+   2+ =
 2 2 
(c)

12.
(a) 2

f
ave =
1
2 0
 2x
22
dx
0 (1+x ) (b) 2c 2 22
5 f (c)= f  =  5c=(1+c )
ave 22 5

1 (1+c )
1 2
= du [ u=1+x , du=2xdx ]  c  0.220 or c  1.207
2 2 1 2
1 u
1 1 5 1 1 2
=  =  1 =
2 u 1 2 5 5
(c)

13. f is continuous on 1,3 , so by the Mean Value Theorem for Integrals there exists a number c in
3
8
1,3 such that  f (x)dx= f (c)(3 1) 8=2 f (c) ; that is, there is a number c such that f (c)= =4 .
1 2

14. The requirement is that

3
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

b
1
b 0
 f (x)dx=3 . The LHS of this equation is equal to
0
b
1
b
 ( 2+6x 3x2) dx= 1b 2x+3x  x
2 3 b
0
2
=2+3b b , so we solve the equation 2+3b b =3
2 2
b  3b+1=0
0
2
3 ( 3)  4 1 1 3 5
b= = . Both roots are valid since they are positive.
2 1 2

15.
1 1 1 50 20

50
f
ave = f (x)dx M=  [ f (25)+ f (35)+ f (45)]
50 20 20 30 3 30 3
1 115 1
= (38+29+48)= =38
3 3 3

12
1 1 12 0
16. (a) v =
ave 12 0 12

I . Use the Midpoint Rule with n=3 and  t=
v(t)dt=
3
=4 to estimate I .
0
1 2
I M =4[v(2)+v(6)+v(10)]=4[21+50+66]=4(137)=548 . Thus, v  (548)=45 km / h.
3 ave 12 3
2
(b) Estimating from the graph, v(t)=45 when t 5.2 s.
3

17. Let t=0 and t=12 correspond to 9 A.M. and 9 P.M., respectively.
12 12
1 1 1 12 1
T
ave =
12 0
 50+14sin
12
 t dt=
12
50t 14

cos
12
 t
0
0
1 12 12 28

= 50 12+14 +14 = 50+ F  59 F

12   

18.
 t/50  t/50
T
ave =
1

30 0 0
30
20+75e ( dt=
1
30
)
20t 50 75e
30
0
=
1
30
( 600 3750e 3/5)  (  3750 )
1  3/5  3/5

=
30
(
4350 3750e )
=145 125e  76.4 C

8 8
1 12 3 
 dx=  ( x+1 )
1/2 8
19.  = dx= 3 x+1 =9 3=6 kg / m
ave 8 x+1 2 0 0
0

4
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

2
1 2 ds 2 v
20. s= gt t= 2s/g [ since t 0 ]. Now v= =gt=g 2s/g = 2gs v =2gs s= . We see that
2 dt 2g
v can be regarded as a function of t or of s : v=F(t)=gt and v=G(s)= 2gs . Note that v =F(T )=gT .
T
2 2
1 2 v [F(t)]
Displacement can be viewed as a function of t : s=s(t)= gt ; also s(t)= = .
2 2g 2g
When t=T , these two formulas for s(t) imply that
1 2
2gs(T ) =F(T )=v =gT =2 gT /T =2s(T )/T (*)
T 2
The average of the velocities with respect to time t during the interval 0,T is
T
1 1
v
t ave
=F
ave = 
T 0 0
F(t)dt=
T
s(T ) s(0) [ by FTC]

s(T ) 1
= [ since s(0)=0 ] = v [ by (*) ]
T 2 T

But the average of the velocities with respect to displacement s during the corresponding
displacement interval s(0),s(T ) = 0,s(T ) is
s(T ) s(T ) s(T )
1 1 2g
  
1/2
v =G = G(s)ds= 2gs ds= s ds
s ave ave s(T ) 0 s(T ) s(T )
0 0 0

2g 2 3/2 s(T )2 2g 3/2 2 2

=  s =   s(T ) = 2gs(T ) = v [by(*)]
s(T ) 3 0 3 s(T ) 3 3 T

21.
5 5 5
1 1 5 2 1 2
V
ave =  V (t)dt=  1 cos  t dt=  1 cos  t dt
50 5 0 4 5 4 0 5
1 5 2 5 1 5
= t sin  t = ( 5 0 )  0 =  0.4 L
4 2 5 0 4 4

R R R 2
1 1 P P 1 3 P 2 PR
22. v =
ave R 0
 v(r)dr= 
R 0 4 l
( 2 2
R  r dr= )
4 lR
2
R r
3
r =
0 4 lR 3
3
R=
6 l
.
0
2
PR 2
Since v(r) is decreasing on ( 0,R , v =v(0)= . Thus, v = v .
max 4 l ave 3 max

23. Let

5
Stewart Calculus ET 5e 0534393217;6. Application of Integration; 6.5 Average Value of a Function

x
F(x)= f (t)dt for x in a,b . Then F is continuous on a,b and differentiable on ( a,b) , so by the
a
/ /
Mean Value Theorem there is a number c in ( a,b) such that F(b) F(a)=F (c)(b a) . But F (x)= f (x)
b
by the Fundamental Theorem of Calculus. Therefore,  f (t)dt 0= f (c)(b a) .
a

24.
b c b
f a,b = 1 1
 f (x)dx= b a
1
 f (x)dx+ b a  f (x)dx
ave b a a a c
c b
c a 1 b c 1 c a b c
=
b a c a
 f (x)dx +
b a b c
 f (x)dx = f
b a ave
a,c + f
b a ave
c,b
a c

6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

1 2
1. Let u=ln x , dv=xdx du=dx/x , v= x . Then by Equation 2,  udv=uv  vdu ,
2
1 2 1 2 1 2 1 1 2 1 1 2
 xln xdx = x ln x  x ( dx/x ) = x ln x  xdx= x ln x  x +C
2 2 2 2 2 2 2
1 2 1 2
= x ln x x +C
2 4

2
2. Let u= , dv=sec  d  du=d , v=tan  . Then
 
2
 sec  d = tan   tan  d = tan   ln sec  +C .

1
3. Let u=x , dv=cos 5xdx du=dx , v=
sin 5x . Then by Equation 2,
5
1 1 1 1
 xcos 5xdx=
5
xsin 5x  5
sin 5xdx= xsin 5x+
5 25
cos 5x+C .

 x  x  x  x  x  x  x
4. Let u=x , dv=e dx du=dx , v= e . Then  xe dx= xe + e dx= xe  e +C .

. Then  re dr=2re  
r/2 r/2 r/2 r/2 r/2 r/2 r/2
5. Let u=r , dv=e dr du=dr , v=2e 2e dr=2re  4e +C .

1 1 1
6. Let u=t , dv=sin 2t dt du=dt , v= cos 2t . Then  tsin 2t dt= tcos 2t+  cos 2t
2 2 2
1 1
dt= tcos 2t+ sin 2t+C .
2 4

2 1
7. Let u=x , dv=sin  xdx du=2xdx and v= cos  x . Then

1 2
I= x sin  xdx= 
2 2
x cos  x+ xcos  xdx ( * ).
 
1
Next let U =x , dV =cos  xdx dU =dx , V = sin  x , so

1 1 1 1
 xcos  xdx=

xsin  x

 sin  xdx=

xsin  x+
2
cos  x+C . Substituting for  xcos  xdx in (
1

* ), we get
1 2 2 1 1 1 2 2 2
I= x cos  x+ xsin  x+ cos  x+C = x cos  x+ xsin  x+ cos  x+C ,
   2 1  2 3
  
2
where C= C .
 1

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2 1
8. Let u=x , dv=cos mxdx du=2xdx , v= sin mx . Then
m
1 2 2
I= x cos mxdx=  xsin mxdx ( * ). Next let U =x , dV =sin mxdx dU =dx ,
2
x sin mx
m m
1 1 1 1 1
V = cos mx , so  xsin mxdx= xcos mx+  cos mxdx= xcos mx+ sin mx+C .
m m m m 2 1
m
Substituting for  xsin mxdx in ( * ), we get
1 2 2 1 1 1 2 2 2
I= x sin mx  xcos mx+ sin mx+C = x sin mx+ xcos mx sin mx+C ,
m m m 2 1 m 2 3
m m m
2
where C= C .
m 1

2
9. Let u=ln (2x+1) , dv=dx du= dx , v=x . Then
2x+1
2x (2x+1) 1
 ln (2x+1)dx =xln (2x+1)  2x+1
dx=xln (2x+1) 
2x+1
dx

1 1
=xln (2x+1)  1
2x+1
dx=xln (2x+1) x+ ln (2x+1)+C
2
1
= (2x+1)ln (2x+1) x+C
2

 1 dx  1  1 x
10. Let u=sin x , dv=dx du= , v=x . Then  sin xdx=xsin x  dx . Setting
2 2
1 x 1 x
 1/2 1 1
2
t=1 x , we get dt= 2xdx , so   xdx
=  t 
2
dt =
2
2t
1/2
(1/2
) 2
+C=t +C= 1 x +C .
2
1 x
 1  1
Hence,  sin xdx=xsin x+ 1 x +C .
2

4 4
11. Let u=arctan 4t , dv=dt  du= dt= dt , v=t . Then
2 2
1+(4t) 1+16t
4t 1 32t
 arctan 4t dt =t arctan 4t  2
dt=t arctan 4t
8
 2
dt
1+16t 1+16t
1 2
=t arctan 4t ln (1+16t )+C
8

5
12. Let u=ln p , dv= p dp
2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

1 1 6 1 6 1 5 1 6 1 6
dp , v= p . Then  p ln pdp= p ln p 
5
du= p dp= p ln p p +C .
p 6 6 6 6 36

2 1
13. First let u= ( ln x ) , dv=dx du=2ln x dx , v=x . Then by Equation 2,
x
1
I= (ln x) dx=x(ln x)  2 xln x dx=x(ln x)  2 ln xdx . Next let U =ln x , dV =dx
2 2 2
dU =1/xdx , V =x
x
to get  ln xdx=xln x  x ( 1/x ) dx=xln x  dx=xln x x+C . Thus,
1
2 2
I=x(ln x)  2 xln x x+C
( ) =x(ln x)  2xln x+2x+C , where C= 2C .
1 1

du=3t dt , v=e . Then I= t e dt=t e  

3 t 2 t 3 t 3 t 2 t
14. Let u=t , dv=e dt  3t e dt . Integrate by parts twice
t
more with dv=e dt .
I = t 3et  ( 3t2et  t
) 3 t
6te dt =t e  3t e +6te 
2 t t

t
6e dt
(
= t 3et  3t 2et+6tet  6et+C= t 3 3t 2+6t 6 et+C )
More generally, if p(t) is a polynomial of degree n in t , then repeated integration by parts shows that
 p ( t ) e dt= p(t) p (t)+ p (t) p (t)+   + (  1 ) p ( ) (t) e +C .
t / / / / / / n n t

2 1 2
15. First let u=sin 3 , dv=e d  du=3cos 3 d , v= e . Then
2
1 2 3 2
I= e sin 3 d = 
2 2
e sin 3  e cos 3 d . Next let U =cos 3 , dV =e d
2 2
1 2
 dU = 3sin 3 d , V = e to get
2
1 2 3 2
 e cos 3 d = 2 e cos 3 + 2  e sin 3 d . Substituting in the previous formula gives
2

1 2 3 2 9 2 1 2 3 2 9
I= e sin 3 
2 4
e cos 3 
4
 e sin 3 d = e sin 3 
2 4
e cos 3 
4
I
13 1 2 3 2 1 2 4
I= e sin 3  e cos 3 +C . Hence, I = e ( 2sin 3  3cos 3 ) +C , where C= C .
4 2 4 1 13 13 1

 1 
16. First let u=e , dv=cos 2 d  sin 2 . Then du= e d , v=
2
 1  1  1  1 
I= e cos 2 d = e sin 2  
2 2
sin 2  e d = e sin 2 +  e sin 2 d .
2 2
( )
  1
Next let U =e , dV =sin 2 d  dU = e d , V = cos 2 , so
2

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

 1  1  1  1 
 e cos 2  
e sin 2 d =
2

2
cos 2  e d =
2
e cos 2 
2
 e cos 2 d . ( )
1  1 1  1 1  1  1
So I= e sin 2 +  e cos 2  I = e sin 2  e cos 2  I
2 2 2 2 2 4 4
5 1  1 
I= e sin 2  e cos 2 +C 
4 2 4 1
4 1  1  2  1 
I= e sin 2  e cos 2 +C = e sin 2  e cos 2 +C .
5 2 4 1 5 5

17. Let u=y , dv=sinh ydy du=dy , v=cosh y . Then

 ysinh ydy=ycosh y  cosh ydy=ycosh y sinh y+C .
sinh ay
18. Let u=y , dv=cosh aydy . Then du=dy , v=
a
ysinh ay 1 ysinh ay cosh ay
 ycosh aydy=
a

a
 sinh aydy=
a

2
+C .
a

1
19. Let u=t , dv=sin 3t dt du=dt , v=
cos 3t . Then
3
 1  1  1 1  
 0
t sin 3t dt= 
3
t cos 3t +  cos 3t dt=
0 3 0 3
  0 +
9
sin 3t = .
0 3

2  x  x
20. First let u=x +1 , dv=e dx du=2xdx , v= e . By (6),
 x  x 1  x  1  x  x

1
0
( x +1) e
2
dx=  ( x +1) e
2
+ 2xe dx= 2e +1+2 xe dx . Next let U =x , dV =e dx
0
1
0 0
1
dU =dx
 x  x  x 1 1  x  1  x 1  1  1  1
. By (6) again,  xe dx=  xe + e dx= e +  e
1
, V = e = e  e +1= 2e +1 . So
0 0 0 0
 x  1  1  1  1  1

1
0
( x +1) e
2
(
dx= 2e +1+2  2e +1 = 2e +1 4e +2= 6e +3 .)
 2 1  1
21. Let u=ln x , dv=x dx du= dx , v= x . By (6),
x
2

 ln x
2
dx= 
ln x
x
2 2  2
+ x dx=
1 1
1
2
ln 2+ln 1+ 
1
x
2

1
=
1
2
ln 2+0
1
2
1 1
+1= 
2 2
ln 2 .
1 x

2 3/2
22. Let u=ln t , dv= t dt du=dt/t , v= t . By Formula 6,
3

4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2 3/2 4 2 2 2 2 3/2 4 16 4 16 28
 
4 4
t ln t dt= t ln t  t dt=  8 ln 4 0  t = ln 4 ( 8 1 ) = ln 4
1 3 1 3 1 3 3 3 1 3 9 3 9
.

dy  2y 1  2y
23. Let u=y , dv= =e dy du=dy , v= e . Then
2y 2
e
1  1 1 1  1 2 1  2y 1 1 2 1 2 1 1 3 2
y
  e
1 2y 2y
dy=  ye + dy=  e +0  e = e  e + =  e .
0 2y 2 0 2 0 2 4 0 2 4 4 4 4
e

2
24. Let u=x , dv=csc xdx du=dx , v= cot x . Then
  /2  /2    /2

/2 2
 /4
xcsc xdx =  xcot x + 
 /4  /4
cot xdx=
2
 0+
4
 1+ ln sin x
 /4
[see Exercise 5.5.]

 1   1/2  1
= +ln 1 ln = +0 ln 2 = + ln 2
4 2 4 4 2

 1 dx
25. Let u=cos x , dv=dx du= , v=x . Then
2
1 x
1/2

1/2

0
 1
I=  cos xdx= xcos x
 1 1/2
0
+  0
xdx

1 x
2
1 
=  + t
2 3 1

3/4
1/2

1
2
dt , where t=1 x 
2
dt= 2xdx .

1
 1  1/2   3 1

1
Thus, I= + t dt= + t = +1 = ( +6 3 3 ) .
6 2 3/4 6 3/4 6 2 6

26. Let u=x , dv=5 dx

x
du=dx , v= 5 /ln 5 . Then ( x
)
1 x 1 1 x x 1
x5 5 5 1 5 5

x
x5 dx =
ln 5

0 0
 ln 5
dx=
ln 5
 0
ln 5 ln 5 0
=
ln 5

5
2
+
1
2
0 ( ln 5) ( ln 5)
5 4
= ln 5  2
( ln 5)

cos x
27. Let u= ln ( sin x ) , dv=cos xdx du= dx , v=sin x . Then
sin x
I= cos x ln ( sin x ) dx=sin x ln ( sin x )   cos xdx=sin x ln ( sin x )  sin x+C .

5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

Another method: Substitute t=sin x , so dt=cos xdx . Then I= ln t dt=t ln t t+C (see Example 2) and
so I=sin x ( ln sin x 1 ) +C .

1  1  dx
28. Let u=arctan(1/x) , dv=dx du=  dx= 2
, v=x . Then
2 2
1+(1/x) x x +1
3 3
1 3 x 2   1 3
 +
2
arctan(1/x)dx = xarctan x +1= 3  1 + ln (x +1)
1 x 1 1 dx 6 4 2 1

 3  1  3  1 4  3  1
=  + (ln 4 ln 2)=  + ln =  + ln 2
6 4 2 6 2 2 2 6 2 2

w w
29. Let w=ln x dw=dx/x . Then x=e and dx=e dw , so
1 w
cos ( ln x ) dx =  e cos wdw= e ( sin w+cos w ) +C [by the method of Example 4]
w
 2
1
= x sin ( ln x ) +cos ( ln x ) +C
2

2 r 2
30. Let u=r , dv= dr  du=2r dr , v= 4+r . By (6),
2
4+r
1

 1 1 1
2
r
3
dr
= r
2
4+r
2
0
 2 r 4+r dr= 5 
3
2
( 4+r 2) 3/2 0
0
2
0 4+r
2 3/2 2 10 16 16 7
= 5 (5) + (8)= 5 1 + =  5
3 3 3 3 3 3

5
2 4 ln x x
31. Let u=(ln x) , dv=x dx du=2 dx , v= . By (6),
x 5
2 5 2 2 4 2 4
x x 32 x
  2 ( ln 2 )  0 2
4 2 2 2
x (ln x) dx= ( ln x ) ln xdx= ln xdx .
1 5 1 1 5 5 1 5
4 5
x 1 x
Let U =ln x , dV = dx dU = dx , V = .
5 x 25
2 4 5 2 2 4 5 2
x x x 32 x 32 32 1
Then  ln xdx= ln x   dx= ln 2 0 = ln 2  .
1 5 25 1 1 25 25 125 1 25 125 125
6
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2
32 32 31 32 2 64 62
So  x (ln x) dx=
4 2 2
(ln 2)  2 ln 2 = (ln 2)  ln 2+ .
1 5 25 125 5 25 125

s s
32. Let u=sin (t s) , dv=e ds du= cos (t s)ds , v=e . Then
t
I= e sin (t s)ds= e sin (t s) + e cos (t s)ds=e sin 0 e sin t+I . For I , let U =cos (t s) , dV =e ds
t s s t s t 0 s
0 0 0 1 1
t

s s t s t 0
 dU =sin (t s)ds , V =e . So I = e cos (t s)  e sin (t s)ds=e cos 0 e cos t I . Thus,
1 0 0
1 t
I= sin t+e  cos t I
t t
2I=e  cos t sin t I=
2
(
e  cos t sin t . )
33. Let w= x , so that x=w and dx=2wdw . Thus,  sin x dx= 2wsin wdw . Now use parts with
2

u=2w , dv=sin wdw , du=2dw , v= cos w to get

 2wsin wdw =  2wcos w+ 2cos wdw= 2wcos w+2sin w+C
=  2 x cos x +2sin x +C=2 ( sin x  x cos x ) +C

x
34. Let w= x , so that x=w and dx=2wdw . Thus,  e dx= e 2wdw . Now use parts with u=2w ,
2 4 2 w
1 1
w 2
dv=e dw , du=2dw , v=e to get  e 2wdw= 2we
w w 2 w
1
 2 e dw=4e  2e 2 e  e =2e .
1
2 w
1
2
( 2
) 2

2
35. Let x= , so that dx=2 d . Thus,

 1 1 
 
3
cos  ( 2) d =
 /2

2
cos  ( 2) 2
(2 d )= 
2  /2
xcos x}{dx} . Now use parts with u=x ,
 /2
dv=cos xdx , du=dx , v=sin x to get
 
1 1  1 
2
 xcos xdx = 2 xsin x
 /2
  sin xdx =
2
xsin x+cos x
 /2
/2  /2
1 1    1 1  1 
= ( sin  +cos  ) sin +cos = (  0 1)  1+0 = 
2 2 2 2 2 2 2 2 2 4

36.
2 2
2 2 x 1 1
 x e dx =
5 x (x ) e xdx= t e
2 t
2
2
dt [ where t=x 
2
dt=xdx ]
2
1 2 1 4 2
=
2
( t
)
t  2t+2 e +C [ by Example 3] =
2
x
x  2x +2 e +C ( )
7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

37. Let u=x , dv=cos  xdx du=dx , v= ( sin  x ) / . Then

sin  x sin  x xsin  x cos  x
 xcos  xdx=x   

dx=

+
2
+C . We see from the graph that this is

reasonable, since F has extreme values where f is 0 .

1
3/2 2 5/2
38. Let u=ln x , dv=x dx du=
dx , v= x . Then
x 5
2 5/2 2 3/2 2 5/2 2 2 5/2
 x ln xdx 5
3/2
= x ln x
5
 x dx=
5
x ln x
5
x +C

2 5/2 4 5/2
= x ln x x +C.
5 25
We see from the graph that this is reasonable, since F has a minimum where f changes from negative
to positive.

x x
39. Let u=2x+3 , dv=e dx du=2dx , v=e . Then
 (2x+3)e dx=(2x+3)e  2 e dx=(2x+3)e  2e +C= ( 2x+1 ) e +C . We see from the graph that this is
x x x x x x

reasonable, since F has a minimum where f changes from negative to positive.

8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2 2
40.  x e dx= x  xe dx=I .
3 x 2 x

2 2
2 x 1 x
Let u=x , dv=xe dx du=2xdx , v= e . Then
2
2 2 2 2 2
1 2 x 1 2 x 1 x 1 x 2
I= x e   xe dx= x e 
2
x
2 2
(
e +C= e x  1 +C . We see from the graph that this is
2
)
reasonable, since F has a minimum where f changes from negative to positive.

1 1 x sin 2x
41. (a) Take n=2 in Example 6 to get  sin xdx= cos xsin x+  1dx= 
2
+C .
2 2 2 4
1 3 1 3 3
 cos xsin x+  sin xdx=
4 3 2 3
(b) sin xdx= cos xsin x+ x sin 2x+C .
4 4 4 8 16

n 1 n 2
42. (a) Let u=cos x , dv=cos xdx du= (n 1)cos xsin xdx , v=sin x in (2):
 cos xdx = cos xsin x+(n 1) cos xsin xdx
n n 1 n 2 2

= cos n 1 xsin x+(n 1) cos n 2 x 1 cos 2 x dx ( )

= cos n 1 xsin x+(n 1) cos n 2 xdx (n 1) cos nxdx

Rearranging terms gives n cos xdx=cos xsin x+(n 1) cos

n n 1 n 2
xdx or

1 n 1
 
n n 1 n 2
cos xdx= cos xsin x+ cos xdx
n n

1 1 x sin 2x
(b) Take n=2 in part (a) to get  cos xdx= cos xsin x+  1dx= +
2
+C .
2 2 2 4
1 3 1 3 3
 cos xdx= cos xsin x+  cos xdx= cos xsin x+ x+
4 3 2 3
(c) sin 2x+C
4 4 4 8 16

1 n 1
43. (a) From Example 6,  sin xdx= 
n n 1 n 2
cos xsin x+ sin xdx . Using (6),
n n

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

n 1  /2
 /2

n cos xsin x n 1  /2 n 2
0
sin xdx = 
n 0
+
n 0
 sin xdx
n 1  /2 n 2 n 1  /2 n 2
= (0 0)+
n 0
 sin xdx=
n 0
 sin xdx

2  /2  /22  /2 2
(b) Using n=3 in part (a), we have  sin xdx= 
3
sin xdx=  cos x = .
0 3 0 3 0 3
 /2 4  /2 3 4 2 8
Using n=5 in part (a), we have  sin xdx= 
5
sin xdx=  = .
0 5 0 5 3 15
(c) The formula holds for n=1 (that is, 2n+1=3 ) by (b). Assume it holds for some k 1 . Then
 /2 2 4 6     ( 2k )
 0 sin xdx= 3 5 7     ( 2k+1 ) . By Example 6,
2k+1

 /2 2k+2  /2 2k+2 2 4 6     ( 2k )
 0 sin xdx = 2k+3  0 sin 2k+1xdx= 2k+3  3 5 7     2k+1
2k+3
( )
2 4 6     (2k)[2 ( k+1 ) ]
= ,
3 5 7     (2k+1)[2 ( k+1 ) +1]
so the formula holds for n=k+1 . By induction, the formula holds for all n 1 .

44. Using Exercise 43 (a), we see that the formula holds for n=1 , because
 /2 1  /2 1  /2 1 
 0 sin xdx= 2  0 1dx= 2 x 0 = 2  2 .
2

 /2 1 3 5     (2k 1) 
Now assume it holds for some k 1 . Then 
2k
sin xdx= . By Exercise 43
0 2 4 6     ( 2k ) 2
(a),
 /2 2k+1  /2 2k+1 1 3 5     (2k 1) 
 0 sin ( ) xdx = 2k+2  0 sin 2k xdx= 2k+2  2 4 6     2k 2
2 k+1
( )
1 3 5     (2k 1)(2k+1) 
=  ,
2 4 6     (2k)(2k+2) 2
so the formula holds for n=k+1 . By induction, the formula holds for all n 1 .

n n 1
45. Let u=(ln x) , dv=dx du=n(ln x) ( dx/x ) , v=x . By Equation 2,
  ( dx/x ) =x(ln x)  n (ln x) dx .
n n n 1 n n 1
(ln x) dx=x(ln x)  nx(ln x)

dx , v=e . By Equation 2,  x e dx=x e  n x

n x n 1 x n x n x n 1 x
46. Let u=x , dv=e dx du=nx e dx .

2 n 2 n 1
47. Let u= x +a ( 2
) , dv=dx du=n x +a ( 2
) 2xdx , v=x . Then

10
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

 ( x2+a2) ndx = x ( x2+a2) n 2n x2 ( x2+a2) n 1 dx

2 2 n 2 2 n 2 2 n 1
= x ( x +a )  2n  ( x +a ) dx a  ( x +a ) dx since x = ( x +a )  a
2 2 2 2 2

2 2 n 2 2 n 2 2 n 1
( 2n+1 )  ( x +a ) dx=x ( x +a ) +2na  ( x +a ) dx , and
2

2 2 n
x ( x +a )
2
2na
 ( x +a ) dx= 2n+1 + 2n+1  ( x2+a2) n 1 dx [ provided 2n+1 0 ].
2 2 n

n 2 2 n 3
48. Let u=sec x , dv=sec xdx du= ( n 2 ) sec xsec xtan xdx , v=tan x . Then by Equation 2,
 sec xdx = tan xsec x ( n 2 )  sec xtan xdx
n n 2 n 2 2

= tan xsec n 2 x ( n 2 )  sec n 2 x sec 2 x 1 dx ( )

= tan xsec n 2 x ( n 2 )  sec nxdx+ ( n 2 )  sec n 2 xdx

so ( n 1 )  sec xdx=tan xsec x+ ( n 2 )  sec

n n 2 n 2
xdx . If n 1 0 , then
n 2
tan xsec x n 2
 
n n 2
sec xdx= + sec xdx .
n 1 n 1

49. Take n=3 in Exercise 45 to get  (ln x) dx=x ( ln x )  3 (ln x) dx=x(ln x)  3x(ln x) +6xln x 6x+C [
3 3 2 3 2

by Exercise 13 ].
Or : Instead of using Exercise 13 , apply Exercise 45 again with n=2 .

50. Take n=4 in Exercise 46 to get

 x e dx = x e  4 x e dx=x e  4 x  3x +6x 6 e +C
4 x 4 x 3 x 4 x 3 2 x
( ) [by Exercise 14 ]
(
= ex x4 4x3+12x2 24x+24 +C )
Or: Instead of using Exercise 14 , apply Exercise 46 with n=3 , then n=2 , then n=1 .

 0.4x  0.4x
51. Area = xe
5
dx . Let u=x , dv=e dx
0
 0.4x
du=dx , v= 2.5e . Then
area =  2.5xe 0.4x 5
+2.5 e
5  0.4x
dx
0 0
 2  0.4x 5
= 12.5e +0+2.5  2.5e
0
 2  2  2 25 75  2
= 12.5e  6.25(e  1)=6.25 18.75e or  e
4 4
11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

52. The curves y=xln x and y=5ln x intersect when xln x=5ln x xln x 5ln x=0 ( x 5) ln x=0 ; that is,
when x=1 or x=5 . For 1<x<5 , we have 5ln x>xln x since ln x>0 . Thus, area
1 2
= ( 5ln x xln x ) dx=
5 5
(5 x)ln x dx . Let u=ln x , dv= ( 5 x ) dx du=dx/x , v=5x x . Then
1 1 2
1 2 5 1 2 1 25 1
 
5 5
area = (ln x) 5x x  5x x dx=(ln 5)  0 5 x dx
2 1 1 2 x 2 1 2
25 1 2 5 25 25 1 25
= ln 5 5x x = ln 5 25  5 = ln 5 14
2 4 1 2 4 4 2

2
53. The curves y=xsin x and y=(x 2) intersect at a
1.04748 and b
2.87307 , so
area =
b 2
xsin x (x 2) dx
a

1 3 b
=  xcos x+sin x (x 2) [ by Example 1]
3 a

2.81358 0.63075=2.18283

54. The curves y=arctan3x and y=x/2 intersect at x= a

 2.91379 , so
area = a arctan3x 1 x dx=2 a arctan3x 1 x dx
 a 2 0 2
1 2 1 2 a
=2 xarctan3x ln (1+9x ) x [ see Example 5]
6 4 0

2(1.39768)=2.79536 .

12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2
55. V = 2 xcos ( x/2)dx . Let u=x , dv=cos ( x/2)dx
1
du=dx , v= sin ( x/2) .
0 
2  x 1 2  x 2 2  x 1

1
V =2 xsin  2  sin dx=2  0  4  cos
 2 0  0 2   2 0
8 8
=4+ (0 1)=4 .
 

56.
 x
Volume = 2 x e  e
1
0
( x
) dx=2  10 ( xex xe x) dx
 x
 
1 x 1
=2 xe dx xe dx [ both integrals by parts]
0 0

=2 ( xex ex)  (  xe x e x) 1

=2 [2/e 0]=4 /e
0

 x  x  x
57. Volume =
0
2 (1 x)e dx . Let u=1 x , dv=e dx du= dx , v= e .
 1
 x 0  x  x  x 0

0
V =2 (1 x)( e )  2 e dx=2 (x 1)(e )+e
 1  1  1
 x 0
=2 xe =2 (0+e)=2 e
 1

58.
 1 2 1 2 
=
Volume
2 y ln ydy=2 y ln y y
1 2 4 1

 2 3
1 2  (2ln   1) (0 1) 3  
=2 y (2ln y 1) =2  = ln   +
4 1 4 4 2 2

1 3 2 1

2
59. The average value of f (x)=x ln x on the interval 1,3 is f = x ln xdx= I .
ave 3 1 1 2
2 1 3
Let u=ln x , dv=x dx du=(1/x)dx , v= x . So
3

13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

1 3 3 31 2 1 3 3 1 26
I= x ln x   x dx= ( 9ln 3 0 )  x =9ln 3 3 =9ln 3 .
3 1 1 3 9 1 9 9
1 1 26 9 13
Thus, f = I= 9ln 3 = ln 3 .
ave 2 2 9 2 9

60. The rocket will have height H=

60
v(t)dt after 60 seconds.
0
m rt 1 2 60
  
60 60 60
H =  gt v ln dt= g t  v ln (m rt)dt ln mdt
0 e m 2 0 e 0 0

=  g(1800)+v (ln m)(60) v 

60
ln (m rt)dt
e e 0

1
Let u=ln (m rt) , dv=dt  du= ( r)dt , v=t . Then
m rt
60 60 rt m

60
tln (m rt) + dt=60ln (m 60r)+
60
ln (m rt)dt =  1+ dt
0 0 0 m rt 0 m rt
m 60
= 60ln (m 60r)+  t ln (m rt)
r 0
m m
= 60ln (m 60r) 60
ln (m 60r)+ ln m
r r
m m
So H= 1800g+60v ln m 60v ln (m 60r)+60v + v ln (m 60r) v ln m . Substituting g=9.8 ,
e e e r e r e
m=30 , 000 , r=160 , and v =3000 gives us H
14 , 844 m.
e

2  w
61. Since v(t)>0 for all t , the desired distance is s(t)= v(w)dw= w e dw .
t t
0 0
 w  w 2  w t  w
+2 we dw .
2 t
First let u=w , dv=e dw du=2wdw , v= e . Then s(t)=  w e
0 0
 w  w
Next let U =w , dV =e dw dU =dw , V = e . Then
2  t  w t t  w 2  t  t  w t
s(t) =  t e +2  we + e dw = t e +2  te +0+  e
0 0 0

=  t 2e t+2  te t  e t+1 = t 2e t  2te t  2e t+2

( )
= 2 e t
( t2+2t+2) meters
/ / / /
62. Suppose f (0)=g(0)=0 and let u= f (x) , dv=g (x)dx du= f (x)dx , v=g (x) . Then
a
  
a / / / a / / / a / / /
f (x)g (x)dx= f (x)g (x)  f (x)g (x)dx= f (a)g (a) f (x)g (x)dx . Now let U = f (x) ,
0 0 0 0
14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

/ / /
dV =g (x)dx dU = f (x)dx and V =g(x) , so
a
  
a / / / a / / / a / /
f (x)g (x)dx= f (x)g(x)  f (x)g(x)dx= f (a)g(a) f (x)g(x)dx .
0 0 0 0

Combining the two results, we get  (x)dx= f (a)g (a) f (a)g(a)+

a / / / / a / /
f (x)g f (x)g(x)dx .
0 0

63. For I= xf

4 / / / / /
(x)dx , let u=x , dv= f (x)dx du=dx , v= f (x) . Then
1
4

/ 4 / / /
I= xf (x)  f (x)dx=4 f (4) 1 f (1) [ f (4) f (1)]=4 3 1 5 (7 2)=12 5 5=2 .
1 1
/ /
We used the fact that f is continuous to guarantee that I exists.

/
64. (a) Take g(x)=x and g (x)=1 in Equation 1.
(b) By part (a),  
b b /
f (x)dx=bf (b) a f (a) x f (x)dx . Now let y= f (x) , so that x=g(y) and
a a
f ( b)
dy= f (x)dx . Then  x f (x)dx=
/ b /
g(y)dy . The result follows.
a f ( a)
(c) Part (b) says that the area of region ABFC is
f ( b)
= bf (b)  af (a)   f ( a)
g(y)dy
= ( area of rectangle OBFE )  ( area of rectangle OACD)  ( area of region DCFE )

 1 x  1 y
(d) We have f (x)=ln x , so f (x)=e , and since g= f , we have g(y)=e . By part (b),
y 1
  
e ln e y 1 y
ln xdx=eln e 1ln 1 e dy=e e dy=e e =e (e 1)=1 .
1 ln 1 0 0

65. Using the formula for volumes of rotation and the figure, we see that Volume
=  b dy   
d 2 c 2 d 2 2 2 d 2 /
 a dy  g(y) dy= b d  a c  g(y) dy . Let y= f (x) , which gives dy= f (x)dx
0 0 c c


2 2 b 2 / 2
and g(y)=x , so that V = b d  a c  x f (x)dx . Now integrate by parts with u=x , and
a
/
dv= f (x)dx du=2xdx , v= f (x) , and
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

b
  
b 2 / 2 b 2 2 b
x f (x)dx= x f (x)  2x f (x)dx=b f (b) a f (a) 2x f (x)dx , but f (a)=c and f (b)=d
a a a a

 2xf (x)dx = 2 xf (x)dx .

2 2 2 2 b b
V = b d  a c  b d a c
a a

 2n+2 2n+1 2n
66. (a) We note that for 0 x , 0 sin x 1 , so sin x sin x sin x . So by the second
2
Comparison Property of the Integral, I  I  I .
2n+2 2n+1 2n
(b) Substituting directly into the result from Exercise 44 , we get

1 3 5     2(n+1) 1 
I 2 4 6     2(n+1) 2 2(n+1) 1 2n+1
2n+2
= = =
I 1 3 5     (2n 1)  2(n+1) 2n+2
2n
2 4 6     ( 2n ) 2

(c) We divide the result from part (a) by I . The inequalities are preserved since I is positive:
2n 2n
I I I 2n+1
2n+2 2n+1 2n
  . Now from part (b), the left term is equal to , so the expression becomes
I I I 2n+2
2n 2n 2n

2n+1 I 2n+1 I
2n+1 2n+1
  1 . Now lim =lim 1=1 , so by the Squeeze Theorem, lim =1 .
2n+2 I n  2n+2 n  n  I
2n 2n
(d) We substitute the results from Exercises 43 and 44 into the result from part (c):
2 4 6     ( 2n )
I 3 5 7     ( 2n+1 )
2n+1
1 = lim =lim
n  I n  1 3 5     ( 2n 1 ) 
2n
2 4 6     ( 2n ) 2
2 4 6     ( 2n ) 2 4 6     ( 2n ) 2
= lim
n  3 5 7     ( 2n+1 ) 1 3 5     ( 2n 1 ) 
2 2 4 4 6 6 2n 2n 2
= lim 1  3  3  5  5  7      2n 1  2n+1   [rearrange terms]
n 

Multiplying both sides by gives us the Wallis product :
2

2 2 4 4 6 6 
=         
2 1 3 3 5 5 7
(e) The area of the k th rectangle is k . At the 2n th step, the area is increased from 2n 1 to 2n by

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.1 Integration by Parts

2n
multiplying the width by , and at the ( 2n+1 ) th step, the area is increased from 2n to 2n+1 by
2n 1
2n+1 2n
multiplying the height by . These two steps multiply the ratio of width to height by and
2n 2n 1
1 2n 2 2 4 4 6 6 
= respectively. So, by part (d), the limiting ratio is          = .
(2n+1)/(2n) 2n+1 1 3 3 5 5 7 2

17
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1.
 sin xcos xdx = sin xcos xsin xdx=
3 2 2 2
( 1 cos 2x) cos 2xsin xdx= ( 1 u2) u2( du)
= ( u2 1) u2 du= ( u4 u2) du= 15 u5 13 u3+C= 15 cos 5x 13 cos 3x+C
2.
 sin xcos xdx = sin xcos xcos xdx= sin x 1 sin x cos xdx= u 1 u du
6 3 6 2 6
( 2
) 6
( 2
)
1 7 1 9 1 1
= u  u du= u 
6 8
7
( 9
u +C= sin x
7
7
9
)
9
sin x+C

3.
3 /4 3 /4 3 /4

5
sin xcos xdx =
3

5
sin xcos xcos xdx=
2

5
(
sin x 1 sin x cos xdx
2
)
 /2  /2  /2
2 /2 2 /2
1 6 1 8 2 /2
s
=  u
5
( 1 u ) du=  ( u5 u7) du=
2
6
u
8
u
1
1 1
1/8 1/16 1 1 11
=    =
6 8 6 8 384

4.
 /2  /2  /2 1
2
 cos xdx = 
5
( cos x) 2
cos xdx=  ( 1 sin x) cos xdx= ( 1 u2) 2 du
2 2

0 0 0 0
1 1
2 3 1 5 2 1 8
= ( 1 2u2+u4) du= u
3
u+ u
5 0
= 1 +
3 5
 0=
15
0

5.


5 4
cos xsin xdx = cos xsin xcos xdx=
4 4
( 1 sin 2x) 2sin 4xcos xdx= ( 1 u2) 2u4 du
= ( 1 2u2+u4) u4 du= ( u4 2u6+u8) du= 15 u5 27 u7+ 19 u9+C
1 5 2 7 1 9
= sin x sin x+ sin x+C
5 7 9

6.

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1
 sin (mx)dx =
3
( 1 cos 2mx) sin mxdx= m
 ( 1 u2) du [ u=cos mx , du= msin mxdx ]
1 1 3 1 1 3
= u u +C= cos mx cos mx +C
m 3 m 3
1 3 1
= cos mx cos mx+C
3m m

7.
 /2  /2 1

2
cos  d = (1+cos 2 )d [ half angle identity]
0 0 2
1 1  /2 1  
=  + sin 2 = +0  ( 0+0 ) =
2 2 0 2 2 4

 /2  /2 1 1 1  /2 1  
8.  sin (2 )d =
2
(1 cos 4 )d =   sin 4 =  0  ( 0 0 ) =
0 0 2 2 4 0 2 2 4

9.
  2  1 2 1 

4
sin (3t)dt = sin (3t) dt= 
2 2
0
(1 cos 6t) dt= (1 2cos 6t+cos 6t)dt
0 0 2 4 0
1  1 1  3 1
=
4
 0
1 2cos 6t+
2
(1+cos 12t) dt= 
4 0 2
 2cos 6t+ cos 12t dt
2
1 3 1 1  1 3 3
= t sin 6t+ sin 12t =  0+0  ( 0 0+0 ) =
4 2 3 24 0 4 2 8

10.
   1 3 1 

6
= (cos  ) d = 
2 3 2 3
cos  d (1+cos 2 ) d = (1+3cos 2 +3cos 2 +cos 2 )d
0 0 0 2 8 0

1 3  1  3 1 
 (1+cos 4 ) d + 
2
=  + sin 2 + (1 sin 2 )cos 2 d
8 2 0 8 0 2 8 0

1 3 1  1 0 1

2
=  +  + sin 4 + (1 u ) du [ u=sin 2 , du=2cos 2 d ]
8 16 4 0 8 0 2
 3 5
= + +0=
8 16 16

11.

2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1
= (1+2cos  +cos  )d = +2sin  +
 (1+cos 2 )d
2

2
(1+cos  ) d 2
1 1 3 1
= +2sin  +  + sin 2 +C=  +2sin  + sin 2 +C
2 4 2 4

1 1 1
du=dx , v= cos xdx=
2 2
12. Let u=x , dv=cos xdx (1+cos 2x)dx= x+ sin 2x , so
2 2 4
1 1 1 1 1 2 1 1 2 1
 x+ sin 2x  
2
xcos xdx =x 2 4 2
x+ sin 2x dx= x + xsin 2x
4 2 4 4
x + cos 2x+C
8
1 2 1 1
= x + xsin 2x+ cos 2x+C
4 4 8

13.
 /4  /4 1 2
 /4 1

4 2
sin xcos xdx = sin x(sin xcos x) dx=
2 2
0
sin 2x dx (1 cos 2x)
0 0 2 2
1  /4 1  /4 2 1  /4 2
=  (1 cos 2x)sin 2xdx=   sin 2xcos 2xdx
2
sin 2xdx
8 0 8 0 8 0
1  /4 1 1  /4 1 1 1  /4

3 3
= (1 cos 4x)dx sin 2x = x sin 4x sin 2x
16 0 16 3 0 16 4 3 0
1  1 1
=  0 = (3  4)
16 4 3 192

14.
 /2  /2 1  /2 1 1  /2
 0
2
sin xcos xdx =
2
0 4
(
4sin xcos x dx=
2 2
) 0 4
(2sin xcos x) dx= 
2
4 0
2
sin 2xdx

1  /2 1 1  /2 1 1  /2
=
4
 0 2
(1 cos 4x)dx= 
8 0
(1 cos 4x)dx=
8
x
4
sin 4x
0
1  
= =
8 2 16

15.
 sin x cos x dx =
3
( 1 cos 2x)
cos x sin xdx= 1 u u (  du ) = ( 2
) 1/2
( u5/2 u1/2) du
2 7/2 2 3/2 2 7/2 2 3/2
= u  u +C= ( cos x )  ( cos x ) +C
7 3 7 3
2 3 2
= cos x cos x cos x +C
7 3
3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

16. Let u=sin  . Then du=cos  d and

= cos u du= (cos u) cos udu= (1 sin u) cos u du

5 2 2 2 2

5
cos  cos (sin  )d
= (1 2sin u+sin u)cos u du=I
2 4

Now let x=sin u . Then dx=cos udu and

2 3 1 5 2 1
I = (1 2x +x )dx=x 3 x + 5 x +C=sin u 3 sin u+ 5 sin u+C
2 4 3 5

2 3 1 5
=sin (sin  ) sin (sin  )+ sin (sin  )+C
3 5

17.


2 3
cos xtan xdx =
sin x
dx=
3
1 u (  du )
=
 1 (
+u du
2
)
cos x u u
1 2 1 2
= ln u + u +C= cos x ln cos x +C
2 2

18.

=
cos 
5
sin  d =
4 cos 
5
d =
cos 
4
cos  d =
( 1 sin 2 ) 2 cos 

5 4
cot  sin  d d
5 sin  sin  sin 
sin 

=
( 1 u2) 2 du= 1 2u +u
2
du=
4
1 3
 2u+u du
u u u
2 1 4 2 1 4
=ln u  u + u +C=ln sin   sin  + sin  +C
4 4

19.

 cos x dx = ( sec x tan x ) dx=ln sec x+tan x  ln sec x +C

1 sin x by (1) and the boxed
formula above it

=ln ( sec x+tan x ) cos x +C=ln 1+sin x +C

=ln ( 1+sin x ) +C since 1+sin x 0

Or:


1 sin x
dx = 1 sin x

1+sin x
dx= 
1 sin
2
x dx
= 
(
cos xdx )
cos x cos x 1+sin x cos x ( 1+sin x ) 1+sin x
4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

dw
= [ where w=1+sin x , dw=cos xdx ]
w
=ln w +C=ln 1+sin x +C=ln ( 1+sin x ) +C

1 4 1
20.  cos xsin 2xdx=2 cos xsin x dx= 2 u du=
2 3 3 4
u +C= cos x+C
2 2

1 2 1
21. Let u=tan x , du=sec xdx . Then  sec xtan xdx= udu=
2 2 2
u +C= tan x+C .
2 2
1 2 1
Or: Let v=sec x , dv=sec xtan xdx . Then  sec xtan xdx= vdv= v +C= sec x+C .
2 2
2 2

22.
 /2  /4 1  /4

4
sec (t/2)dt = dt ] =2
4 2 2
sec x(2dx) [ x=t/2 , dx= sec x(1+tan x)dx
0 0 2 0

1 3 1 1 8
=2 (1+u )du [ u=tan x , du=sec xdx ] =2
1 2 2
u+ u =2 1+ =
0 3 0 3 3

23.  tan xdx=

2
( sec 2x 1) dx=tan x x+C
1
4 2
(
24.  tan xdx= tan x sec x 1 dx= tan xsec xdx
2
)
3
3 2
tan x tan x+x+C
2
 tan xdx=
2

(Set u=tan x in the first integral and use Exercise 23 for the second.)

25.
 sec t dt = sec t sec t dt= (tan t+1) sec t dt= (u +1) du
6 4 2 2 2 2 2 2

1 5 2 3 1 5 2
= (u +2u +1)du= u + u +u+C= tan t+ tan t+tan t+C
4 2 3
5 3 5 3

26.
 /4  /4
 = (tan  +1)tan  sec  d = (u +1)u du
4 4 2 4 2 1 2 4
sec  tan  d
0 0 0

1 7 1 5 1 1 1 12
= (u +u )du=
1 6 4
u+ u = + =
0 7 5 0 7 5 35

27.
 /3  /3
 tan xsec xdx =
5 4 5 2 2
tan x(tan x+1)sec xdx
0 0
5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

3 5 2
= u (u +1)du
0
3 1 8 1 6 3 81 27 81 9 81 36 117
=
7 5
(u +u )du= u+ u = + = + = + =
0 8 6 0 8 6 8 2 8 8 8

Alternate solution:
 /3  /3  /3
 tan xsec xdx = tan xsec xsec xtan xdx=
5 4 4 3 2 2 3
(sec x 1) sec xsec xtan xdx
0 0 0

= (u  1) u du
2 2 2 3
1

= (u  2u +1)u du= (u  2u +u )du

2 4 2 3 2 7 5 3
1 1

1 8 1 6 1 4 2 64 1 1 1 117
= u u+ u = 32 +4   + =
8 3 4 1 3 8 3 4 8

28.
 tan (2x)sec (2x)dx = tan (2x)sec (2x) sec (2x)tan (2x)dx
3 5 2 4

4 1
= (u  1)u ( du) [ u=sec (2x),du=2sec (2x)tan (2x)dx ]
2
2
1 1 7 1 5 1 1
=  (u  u )du=
6 4 7 5
u u +C= sec (2x) sec (2x)+C
2 14 10 14 10

29.
 tan xsec xdx = tan xsec xtan xdx=
3 2
( sec 2x 1) sec xtan xdx
= (u  1)du
2

1 3 1 3
= u  u+C= sec x sec x+C
3 3

30.
 /3  /3  /3
 0
5 6
tan xsec xdx =
0
tan xsec xsec xdx=
5 4 2
0
5
(
tan x 1+tan x sec xdx
2
)2 2

2 2
=
0
3 5
(
u 1+u ) du [ u=tan x , du=sec xdx ] =
2
0
3 5
( 2
u 1+2u +u du
4
)
1 6 1 8 1 10 3 27 81 243 981
=
0
3
( u5+2u7+u9) du= 6
u+ u+
4 10
u
0
=
6
+
4
+
10
=
20

6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

Alternate solution:
 /3  /3  /3
 0
5
tan xsec xdx =
6
0
tan xsec xsec xtan xdx=
4 5
0
( sec 2x 1) 2sec 5xsec xtan xdx
= (1 u2 1) 2u5 du [ u=sec x , du=sec xtan xdx ]
2

= (1 u  2u2+1) u5 du= 21 ( u9 2u7+u5) du

2 4

1 10 1 8 1 6 2 512 32 1 1 1 981
= u  u+ u =  64+   + =
10 4 6 1 5 3 10 4 6 20

31.


5
tan xdx = ( sec 2x 1) 2tan xdx= sec xtan xdx 2 sec xtan xdx+ tan xdx
4 2

= sec xsec xtan xdx 2 tan xsec xdx+ tan xdx

3 2

1 4 2 1 4 2
= sec x tan x+ln sec x +C [ or sec x sec x+ln sec x +C ]
4 4

32.

4
( 2
)
tan aydy = tan ay sec ay 1 dy= tan aysec aydy  tan aydy
6 4 2 4

1
=
5a
tan ay  tan ay sec ay 1 dy
5 2 2
( )
1
=
5a
tan ay  tan aysec aydy+ sec ay 1 dy
5 2 2 2
( )
1 5 1 3 1
= tan ay tan ay+ tan ay y+C
5a 3a a

33.
3
tan 
 4
d = tan  sec  d = tan   (tan  +1) sec  d
3 4 3 2 2

cos 
= u (u +1)duu=tan  , du=sec  d ]
3 2 2

1 6 1 4 1 1
= (u +u )du= u + u +C= tan  + tan  +C
5 3 6 4
6 4 6 4

34.
 tan xsec xdx =
2
( sec 2x 1) sec xdx= 3
sec xdx  sec xdx
7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1
= ( sec xtan x+ln sec x+tan x )  ln sec x+tan x +C [ by Example 8 and (1)]
2
1
= ( sec xtan x ln sec x+tan x ) +C
2

 /2  /2    
35. 
 /6
cot xdx=
2
 /6
( csc x 1 ) dx=  cot x x
2

/2
/6
= 0
2
  3
6
= 3
3

36.
 /2  /2  /2  /2 cos x
  /4
cot xdx =
3
 /4
(
cot x csc x 1 dx=
2
)  /4
2
cot xcsc xdx   /4 sin x
dx

1 2  /2 1 1 1 1 1
=  cot x ln sin x = ( 0 ln 1 )    ln = +ln = (1 ln 2)
2  /4 2 2 2 2 2

37.
 = cot  csc   csc  cot  d = (csc   1)csc   csc  cot  d
3 3 2 2 2 2
cot  csc  d

= (u  1)u  ( du) [ u=csc  , du= csc  cot  d ]

2 2

1 3 1 5 1 1
= (u  u )du= u 
2 4 3 5
u +C= csc   csc  +C
3 5 3 5

38.
 csc xcot xdx = cot x(cot x+1)csc xdx
4 6 6 2 2

= u (u +1) ( du) [ u=cot x , du= csc xdx ]

6 2 2

1 9 1 7 1 1
= ( u  u )du=
8 6 9 7
u u +C= cot x cot x+C
9 7 9 7

2
csc x ( csc x cot x )  csc xcot x+csc x
39. I= csc xdx= dx= dx . Let u=csc x cot x
csc x cot x csc x cot x
( )
du=  csc xcot x+csc x dx . Then I= du/u=ln u =ln csc x cot x +C .
2

40.
 csc xcot xdx = cot x(cot x+1)csc xdx
4 6 6 2 2

= u (u +1) ( du) [ u=cot x , du= csc xdx ]

6 2 2

8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1 9 1 7 1 1
= ( u  u )du=
8 6 9 7
u u +C= cot x cot x+C
9 7 9 7

41. Use Equation 2(b):

1 1
 sin 5xsin 2xdx = 2 cos (5x 2x) cos (5x+2x) dx= 2  ( cos 3x cos 7x ) dx
1 1
= sin 3x sin 7x+C
6 14

42. Use Equation 2(a):

1 1
 sin 3xcos xdx = 2 sin (3x+x)+sin (3x x) dx= 2  ( sin 4x+sin 2x ) dx
1 1
= cos 4x cos 2x+C
8 4

43. Use Equation 2(c):

1 1
 cos 7 cos 5 d = cos (7  5 )+cos (7 +5 ) d =  ( cos 2 +cos 12 ) d
2 2
1 1 1 1 1
= sin 2 + sin 12 +C= sin 2 + sin 12 +C
2 2 12 4 24

44.
cos x+sin x 1 cos x+sin x 1
 sin 2x dx = 2  sin xcos x dx= 2  (csc x+sec x)dx
1
= ( ln csc x cot x +ln sec x+tan x ) +C [ by Exercise 39 and (1)]
2

2
1 tan x 1
45.  2
dx= ( cos 2x sin 2x) dx= cos 2xdx=
2
sin 2x+C
sec x

46.
1 cos x+1 cos x+1 cos x+1

dx =  dx= dx= dx
cos x 1 cos x+1 2 2
cos x 1 cos x 1  sin x
= (  cot xcsc x csc 2x) dx=csc x+cot x+C
2 2 2
47. Let u=tan (t ) du=2tsec (t )dt . Then
9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1 1 5 1
 tsec
2
( t2) tan 4 ( t2) dt= u
4
2
du =
10
u +C=
10
5 2
tan (t )+C .

7 6 2
48. Let u=tan x , dv=sec xtan xdx du=7tan xsec xdx , v=sec x . Then
 tan xsec xdx = tan x sec xtan xdx=tan xsec x 
8 7 7 6 2
7tan xsec xsec xdx

=tan xsec x 7 tan x tan x+1 sec xdx

7 6
( 2
)
=tan xsec x 7 tan xsec xdx 7 tan xsec xdx .
7 8 6

 /4
Thus, 8 tan xsec xdx=tan xsec x 7 tan xsec xdx and 
8 7 6 8
tan xsec x
0
1  /4 7  /4 2 7

7 6
dx= tan xsec x  tan xsec xdx=  I.
8 0 8 0 8 8

49. Let u=cos x du= sin xdx . Then


5
sin xdx =  ( 1 cos 2x) 2sin xdx= ( 1 u2) 2 (  du )
=  (  1+2u2 u4) du= 15 u5+ 23 u3 u+C
1 5 2 3
= 
cos x+ cos x cos x+C
5 3
Notice that F is increasing when f ( x ) >0 , so the graphs serve as a check on our work.

50.
1 4 1 1 1 2
sin xcos xdx =  
4

4 4
sin 2x dx= sin 2xdx= ( 1 cos 4x ) dx
2 16 16 2
1
=
64
 ( 1 2cos 4x+cos 24x) dx
1 1 1
=
64
x
2
sin 4x +
128
 ( 1+cos 8x ) dx

10
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1 1 1 1
= x sin 4x + x+ sin 8x +C
64 2 128 8
3 1 1
= x sin 4x+ sin 8x+C
128 128 1024

Notice that f (x)=0 whenever F has a horizontal tangent.

51.
1
 sin 3xsin 6xdx = cos (3x 6x) cos (3x+6x) dx
2
1
=
2
 (cos 3x cos 9x)dx
1 1
= sin 3x sin 9x+C
6 18

Notice that f (x)=0 whenever F has a horizontal tangent.

52.
x x 2 x
 dx =
4 2
sec tan
+1 sec dx
2 2 2
x 1 2 x
( )
= u +1 2du [ u=tan
2
2
, du= sec
2 2
dx ]

2 3 2 3 x x
= u +2u+C= tan +2tan +C
3 3 2 2

Notice that F is increasing and f is positive on the intervals on which they are defined. Also, F has
11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

no horizontal tangent and f is never zero.

53.
1  1 
f
ave =
2
 
2
sin xcos xdx=
3
2
 
2
( 2
sin x 1 sin x cos xdx )
1
=
2

0 2
0
( 2
)
u 1 u du [ where u=sin x ]

=0

1 2 1
 sin xcos xdx= u( du)=
2
54. (a) Let u=cos x . Then du= sin xdx u +C= cos x+C .
2 2 1
1 2 1
 sin xcos xdx= udu= u +C= sin x+C .
2
(b) Let u=sin x . Then du=cos xdx
2 2 2
1 1
(c)  sin xcos xdx=
2
sin 2xdx=
4
cos 2x+C
3
(d) Let u=sin x , dv=cos xdx . Then du=cos xdx , v=sin x ,
1
so  sin xcos xdx=sin x  sin xcos xdx , by Equation .1.2, so  sin xcos xdx=
2 2
sin x+C .
2 4

The answers differ from one another by constants. Since

2 2 1 1 2 1 1 2 1
cos 2x=1 2sin x=2cos x 1 , we find that  cos 2x= sin x = cos x+ .
4 2 4 2 4

 3
55. For 0<x< , we have 0<sin x<1 , so sin x<sin x . Hence the area is
2
 /2
 0
( sin x sin 3x) dx=  0 /2sin x ( 1 sin 2x) dx=  0 /2cos 2xsin xdx . Now let u=cos x du= sin xdx .
1 3 1 1
Then area = u (  du ) = u du=
0 2 1 2
u = .
1 0 3 0 3
 2 1
56. sin x>0 for 0<x< , so the sign of 2sin x sin x [ which equals 2sin x sin x ] is the same
2 2
1 2   
as that of sin x . Thus 2sin x sin x is positive on , and negative on 0, . The
2 6 2 6
desired area is
12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

 /6
 0
( sin x 2sin 2x) dx+  /2/6 ( 2sin 2x sin x) dx
 /6  /2
=  0
( sin x 1+cos 2x ) dx+  /6
( 1 cos 2x sin x ) dx
1  /6 1  /2
=  cos x x+ sin 2x + x sin 2x+cos x
2 0 2  /6

3  3   3 3
=  +  (  1) +   +
2 6 4 2 6 4 2
 3
= 1+ 
6 2

57.

It seems from the graph that 

2 3
cos xdx=0 , since the area below the x  axis and above the graph
0
looks about equal to the area above the axis and below the graph. By Example 1, the integral is
1 3 2
sin x sin x =0 . Note that due to symmetry, the integral of any odd power of sin x or cos x
3 0
between limits which differ by 2n ( n any integer) is 0 .

58.

It seems from the graph that  sin 2 xcos 5 xdx=0 , since each bulge above the x axis seems to
2
0
have a corresponding depression below the x axis. To evaluate the integral, we use a trigonometric
identity:
1 2
 0sin 2 xcos 5 xdx = 2  0 sin (2 x 5 x)+sin (2 x+5 x) dx
1

1 2
=
2 0
 sin ( 3 x)+sin 7 x dx

13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1 1 1 2
= cos ( 3 x) cos 7 x
2 3 7 0
1 1 1
= (1 1) (1 1) =0
2 3 7

 2
  1 1 1   
59. V = 
2
 sin xdx= (1 cos 2x)dx= x sin 2x =  0 +0 =
 /2  /2 2 2 4  /2 2 4 4

60.
 /4 2 2  /4  /4  /4
Volume =
0
( tan x ) dx= 
0
tan x ( sec x 1 ) dx=
2 2
 0
2
tan xsec xdx 
2
 0
2
tan xdx
 /4 2  /4
 0 u du   0 ( sec x 1 ) dx
2
=

1 3  /4  /4 1 3  /4 1   2
= u  tan x x = tan x tan x+x =  1+ = 
3 x=0 0 3 0 3 4 4 3

61.
 /2  /2
Volume =  0
( 1+cos x )  1 dx= 
2 2
0
( 2cos x+cos 2x) dx
 /2 2
1 1  
= 2sin x+ x+ sin 2x = 2+ =2 +
2 4 0 4 4

62.
 /2  /2
Volume =  0
2
1  ( 1 cos x )
2
dx=  0
( 2cos x cos 2x) dx
 /2 2
1 1  
= 2sin x x sin 2x = 2  0  0 =2 
2 4 0 4 4

63. s= f (t)= sin  ucos  udu . Let y=cos  u

t 2
dy=  sin  udu . Then
0
1 cos  t 2 1 1 3 cos  t 1
s=

 1
y dy=
 3
y
1
=
3
( 1 cos 3 t ) .
64. (a) We want to calculate the square root of the average value of
2 2 2 2
E(t) = 155sin (120 t) =155 sin (120 t) . First, we calculate the average value itself, by
2 1
integrating E(t) over one cycle (between t=0 and t= , since there are 60 cycles per second) and
60
14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

1
dividing by  0 :
60
1 1/60 1
 155 sin (120 t) dt=60 155 
2 2 2 2 1/60
E(t) = 1 cos (240 t) dt
ave 1/60 0 0 2
1/60 2
= 60 1552 1 1 2 1 1 155
t sin (240 t) =60 155  0  ( 0 0 ) =
2 240 0 2 60 2
155
The RMS value is just the square root of this quantity, which is  110 V.
2
2
(b) 220= E(t) 
ave
1 1
 A sin (120 t)dt=60A 
2 2 1/60 2 2 2 1/60
220 = E(t) ave
=
1/60 0 0 2
[1 cos (240 t)]dt

2 1 1/60 2 1 1 2
= 30A t sin (240 t) =30A  0  ( 0 0 ) = A
240 0 60 2
2 1 2
Thus, 220 = A A=220 2  311 V.
2

65. Just note that the integrand is odd .

Or: If m n , calculate
  1
   sin mxcos nxdx =    2 sin (m n)x+sin (m+n)x dx
1 cos (m n)x cos (m+n)x 
 =  =0
2 m n m+n 
If m=n , then the first term in each set of brackets is zero.

 1 
66.  sin mxsin nxdx= [cos (m n)x cos (m+n)x]dx . If m n , this is equal to
  2
1 sin (m n)x sin (m+n)x 
 =0 . If m=n , we get
2 m n m+n 
 1 1  sin (m+n)x 
  2
[1 cos (m+n)x]dx=
2
x


2(m+n) 
=  0= .

 1 
67.  cos mxcos nxdx= cos (m n)x+cos (m+n)x dx . If m n , this is equal to
  2
1 sin (m n)x sin (m+n)x 
+ =0 . If m=n , we get
2 m n m+n 
 1 1  sin (m+n)x 
  2
[1+cos (m+n)x]dx=
2
x

+
2(m+n) 
= +0= .
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.2 Trigonometric Integrals

a
1  1  n 
 
a sin nx sin mx dx=

m m
68. f (x)sin mxdx= sin mxsin nxdx . By
    n=1 n n=1  
a
m
Exercise 66 , every term is zero except the m th one, and that term is   =a .
 m

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

 3
1. Let x=3sec  , where 0  < or    < . Then dx=3sec  tan  d and
2 2
2
x 9 =
2 2
9sec   9 = 9 sec   1 = 9tan 
2
( )
= 3 tan  =3tan  for the relevant values of 

2
1 1 1 1 1 x 9
 dx= 2
3sec  tan  d =  cos  d = sin  +C=
9 9 9 x
+C
2
x x 9
2
9sec   3tan 

3
Note that  sec ( + )=sec  , so the figure is sufficient for the case    < .
2

 
2. Let x=3sin  , where     . Then dx=3cos  d and
2 2
2
9 x =
2 2
9 9sin  = 9 1 sin  = 9cos 
2
( )
= 3 cos  =3cos  for the relevant values of \italic{\theta}\,\,.

 3 sin   3cos   3cos  d =3  sin  cos  d

3 3 5 3 2

3 2
x 9 x dx =
= 35 sin 2 cos 2 sin  d =35 ( 1 cos 2 ) cos 2 sin  d
= 35 ( 1 u2) u2 ( du)[u=cos  ,du= sin  d ]

= 3
5
( u4 u2) du=35 15 u5 13 u3 +C=35 1
5
5
cos  
1
3
3
cos  +C

2 5/2 2 3/2
5 1 ( 9 x ) 1 ( 9 x )
= 3  +C
5 5 3 3
3 3

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

1 1 2
=
5
(
9 x
2
) 5/2 3 ( 9 x2) 3/2+Cor 5
( ) 3/2
x +6 ( 9 x ) +C

  2
3. Let x=3tan  , where  < < . Then dx=3sec  d and
2 2
2
x +9 =
2 2
9tan  +9 = 9 tan  +1 = 9sec 
2
( )
= 3 sec  =3sec  for the relevant values of  .

3
x 3 3
 dx = 
3 tan 
3sec  d =3  tan  sec  d =3  tan  tan  sec  d
2 3 3 3 2
2
x +9 3sec 

= 33 ( sec 2 )
 1 tan  sec  d =3 
3
( u2 1) du[u=sec  ,du=sec  tan  d ]

= 3
3 1 3
u  u +C=3
3 1 3
sec   sec  +C=3
3 1 ( x2+9) 3/2  2
x +9
+C
3 3 3 3 3
3
1 2 1 2
=
3
(
x +9 ) 3/2 9 2
x +9 +C or
3
(
x  18 ) 2
x +9 +C

4. Let x=4sin  , where   /2    /2 . Then dx=4cos  d and

2 2 2
16 x = 16 16sin  = 16cos  =4 cos  =4cos  . When x=0 , 4sin  =0  =0 , and when
3 
x=2 3 , 4sin  =2 3  sin  =   = . Thus, substitution gives
2 3

2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 3
 /3


3 3  /3
4 sin 
= 
3 3
3 4cos  d =4 sin  d
x 0 4cos  0
dx  /3
2
0 16 x =4
3
 ( 1 cos 2 ) sin  d
0
1/2
1 3 1/2
= 4
3
 ( 1 u2) du= 64 u
3
u
1
1
1 1 1 5 40
= 64   1 = 64  =
2 24 3 24 3

2 2
Or: Let u=16 x , x =16 u , du= 2xdx .

 
5. Let t=sec  , so dt=sec  tan  d , t= 2   = , and t=2  = . Then
4 3
2
 /3  /3


 /3
1
dt =  3
1
sec  tan  d =  1
2
d =  cos  d
2

3 2  /4 sec  tan   /4 sec   /4

2 t t  1
 /3  /3
1 1 1
=  (1+cos 2 )d =  + sin 2
 /4
/4 2
 2 2
1  1 3  1 1  3 1  3 1
= +  +  1 = +  = + 
2 3 2 2 4 2 2 12 4 2 24 8 4

2 
6. Let x=2tan  , so dx=2sec  d , x=0  =0 , and x=2  = . Then
4
 /4 3 5  /4
x +4 dx = 
3 2 2 2

2 3 2
x 2 tan   2sec   2sec  d =2 tan  sec  sec  tan  d
0 0 0
5  /4

2 2
=2 (sec   1)sec  sec  tan  d
0
2

5 2 2
=2 (u  1)u du [u=sec  ,du=sec  tan  d ]
1
2 1 5 1 3 2 1 1 1 1

5 4 2 5 5
=2 (u  u )du=2 u u =2  4 2  2 2  
1 5 3 1 5 3 5 3
2 2 64
=32
15
2+
15
=
15
( 2 +1 )
3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 2
Or: Let u=x +4 , x =u 4 , du=2xdx .

7. Let x=5sin  , so dx=5cos  d . Then

1
 dx = 
1
5cos  d
2 2 2 2
x 25 x 5 sin   5cos 
1 1

2
= csc  d = cot  +C
25 25
2
= 1 25 x
+C
25 x

 3 2 2
8. Let x=asec  , where 0  < or    < . Then dx=asec  tan  d and x  a =atan  ,
2 2
so
2
x a
2 atan 
 dx =  4 4
asec  tan  d
x
4 a sec 
1

2
= sin  cos  d
2
a
2 2 3/2
=
1
sin 
3
+C=
( x a )
+C
2 2 3
3a 3a x

  2
9. Let x=4tan  , where  < < . Then dx=4sec  d and
2 2
4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 2 2
x +16 = 16tan  +16 = 16(tan  +1)
2
= 16sec  =4 sec 
=4sec  for the relevant values of  .

dx

2
4sec  d
2 = = sec  d =ln sec  +tan  +C
x +16 4sec  1

2
x +16 x 2
=ln + +C =ln x +16 +x  ln 4 +C
4 4 1 1

=ln ( 2
)
x +16 +x +C , where C=C  ln 4 .
1

2
(Since x +16 +x>0 , we don’t need the absolute value.)

  2
10. Let t= 2 tan  , where  < < . Then dt= 2 sec  d and
2 2
2 2 2 2
t +2 = 2tan  +2 = 2(tan  +1) = 2sec 
= 2 sec  = 2 sec  for the relevant values of  .

5 5
t 4 2 tan 
 dt = 2 sec  d =4 2  tan  sec  d =4 2 
2 5
( sec 2 )2
 1 sec  tan  d
2
t +2 2 sec 

=4 2  ( u2 1) 2du [ u=sec  , du=sec  tan  d ] =4 2  ( u4 2u2+1) du

5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

1 5 2 3 4 2 4 2
=4 2 u u +u +C= u(3u  10u +15)+C
5 3 15
2 2 2 2
4 2 t +2 (t +2) t +2
=  3  10 +15 +C
15 2 2 2
2
4 2 1 4 2 2
= t +2  3(t +4t +4) 20(t +2)+60 +C
15 4
1 2 4 2
= t +2 (3t  8t +32)+C
15

  1 1
11. Let 2x=sin  , where     . Then x= sin  , dx= cos  d , and
2 2 2 2
2 2
1 4x = 1 ( 2x ) =cos  .
1 1
 1 4x dx =  cos  cos   ( 1+cos 2 ) d
2 d =
2 4
1 1 1
=  + sin 2 +C= ( +sin  cos  )+C
4 2 4
1  1 2
= sin (2x)+2x 1 4x +C
4

1 1 2 3/2 5 1
12.  x x +4 dx=
1 2 5 2
u du [ u=x +4 , du=2xdx ] =  u = ( 5 5  8)
0 4 2 2 3 4 3

 3 2
13. Let x=3sec  , where 0  < or    < . Then dx=3sec  tan  d and x  9 =3tan  ,
2 2
so
2 2
x 9 3tan  1 tan 
 dx = 3sec  tan  d =  d
3 3 3 2
x 27sec  sec 

6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

1 1 1 1 1 1 1
 sin  d = 
2
= (1 cos 2 )d =   sin 2 +C=   sin  cos  +C
3 3 2 6 12 6 6
2 2
1  1 x 1 x 9 3 1  1 x x 9
= sec  +C= sec  +C
6 3 6 x x 6 3 2
2x

14. Let u= 5 sin  , so du= 5 cos  d . Then

du
 
1
5 cos  d =
1
 csc  d
2 = 5 sin   5 cos  5
u 5 u
1
= ln csc   cot  +C [byExercise2.39
5
2
= 1 5 5 u
ln  +C
5 u u
2
= 1 5 5 u
ln +C
5 u

 
15. Let x=asin  , where     . Then dx=acos  d and
2 2
2 2 2
x dx a sin  acos  d
 = tan  d
2
=
( a2 x2) 3/2 3
a cos 
3

=  ( sec 2  1) d =tan    +C

7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

x  1 x
=  sin +C
2 2 a
a x

 3 3
16. Let 4x=3sec  , where 0  < or    < . Then dx= sec  tan  d and
2 2 4
2
16x  9 =3tan  , so
3
sec  tan  d
dx 4
 =
2 2 3 2 2
x 16x  9 sec  3tan 
4

2 2
4 4 4 16x  9 16x  9
=  cos  d = sin  +C= +C= +C
9 9 9 4x 9x

x 1 1 1
17. Let u=x  7 , so du=2xdx . Then  
2 2
dx= du=  2 u +C= x  7 +C .
2 2 u 2
x 7

18. Let ax=bsec 

2 2 2
, so ( ax ) =b sec   ( ax )  b =b sec   b =b
2 2 2 2 2 2
( sec 2 ) 2
 1 =b tan 
2
. So
b
( ) 2 2
ax  b =btan  , dx= sec  tan  d , and
a

8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

b
dx sec  tan 
 a 1 sec  1 cos  1
2
( ax )  b
2 3/2 =  3 3
d =
2
 2
d =
2
 2
d =
2
 csc  cot  d
b tan  ab tan  ab sin  ab
1 1 ax x
= 2
csc  +C=
2 2 2
+C=
2 2 2
+C
ab ab ( ax )  b b ( ax )  b

  2 2
19. Let x=tan  , where  < < . Then dx=sec  d and 1+x =sec  , so
2 2
2 sec  sec 
 sec  d =
1+x 2 2
 = (1+tan  )d
x
dx tan  tan 

= (csc  +sec  tan  )d

=ln csc   cot  +sec  +C [ by Exercise 8.2.39]
2 2 2
1+x 1 1+x 1+x  1 2
=ln  + +C=ln + 1+x +C
x x 1 x

  2
20. Let t=5sin  , where     . Then dt=5cos  d and 25 t =5cos  , so
2 2
t
 dt = 5sin  5cos  d =5 sin  d
25 t
2 5cos 
2
25 t 2
= 5cos  +C= 5 +C= 25 t +C
5

2
Or: Let u=25 t , so du= 2t dt .

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 12
21. Let u=4 9x  du= 18xdx . Then x = ( 4 u ) and
9
1 1 1

2/3 3
0
x
2
4 9x dx = 
0
49
(4 u)u
1/2

18
du=
162
 (0 4u1/2 u3/2) du
4

1 8 3/2 2 5/2 4 1 64 64 64
= u  u =  =
162 3 5 0 162 3 5 1215

 
Or: Let 3x=2sin  , where     .
2 2

  2 2
22. Let x=tan  , where  < < . Then dx=sec  d , x +1 =sec  and x=0  =0 , x=1
2 2

 = , so
4
 /4  /4
  sec  sec  d =
1 2 2 3
x +1 dx = sec  d
0 0 0
1  /4
= sec  tan  +ln sec  +tan  [by Example 8.2.8]
2 0
1 1
= 2 1+ln ( 1+ 2 )  0 ln (1+0) = 2 +ln ( 1+ 2 )
2 2

2 2 2  
23. 5+4x x = (x  4x+4)+9= (x 2) +9 . Let x 2=3sin  ,    , so dx=3cos  d . Then
2 2

 5+4x x dx = 9 (x 2) dx=

2 2 2
9 9sin  3cos  d

= 3cos  d = 9cos  d
2 2
9cos 

10
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

9 9 1
=
2
 (1+cos 2 )d =
2
 + sin 2
2
+C

9 9 9 9
=  + sin 2 +C=  + (2sin  cos  )+C
2 4 2 4
2
9  1 x 2 9 x 2 5+4x x
= sin +   +C
2 3 2 3 3
9  1 x 2 1 2
= sin + (x 2) 5+4x x +C
2 3 2

2
(2 ) 2
24. t  6t+13= t  6t+9 +4= ( t 3) +2 . Let t 3=2tan 
2
, so dt=2sec  d
2
. Then
2

dt 1 2sec 
 2sec  d =
2
= d
2 2sec 
t  6t+13 2
( 2tan  ) +2
2

=  sec  d =ln sec  +tan  +C

1
[by Formula 8.2.1]

2
= ln t  6t+13 t 3
+ +C
2 2 1

2
= ln t  6t+13 +t 3 +CwhereC=C  ln 2
1

1
du
dx 3
25. 9x +6x 8= ( 3x+1 )  9 , so let u=3x+1 , du=3dx . Then  =
2 2
. Now let
2 2
9x +6x 8 u 9
u=3sec  , where

11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

 3 2
0  < or    < . Then du=3sec  tan  d and u  9 =3tan  , so
2 2
1
du 2
3 sec  tan  d u+ u  9
 = 
1 1
=  sec  d = ln sec  +tan 
1
+C = ln +C
2 3tan  3 3 1 3 3 1
u 9
1 2 1 2
= ln u+ u  9 +C= ln 3x+1+ 9x +6x 8 +C
3 3

26. 4x x =
2
( x2 4x+4) +4=4 2
( x 2 ) , so let u=x 2 . Then x=u+2 and dx=du , so
2 2 2
x dx ( u+2 ) du ( 2sin  +2 )
 = = 2cos  d
2 2 2cos 
4x x 4 u
(
=4 sin  +2sin  +1 d
2
)
=2 ( 1 cos 2 ) d +8 sin  d +4 d
=2  sin 2  8cos  +4 +C
=6  8cos   2sin  cos  +C
 1 1 2 1 2
=6sin u  4 4 u  u 4 u +C
2 2
 1 x 2 2 x 2 2
=6sin  4 4x x  4x x +C
2 2

2 2
27. x +2x+2=(x+1) +1 . Let u=x+1 , du=dx . Then
2
 dx du sec  d 2
=  = where u=tan  ,du=sec  d ,
( x2+2x+2) 2 ( u2+1) 2 sec 
4 2
and u +1=sec  x
2

1 1
 
2
= cos  d = ( 1+cos 2 ) d = ( +sin  cos  )+C
2 2

12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

1  1 u 1  1 x+1
= 2 tan u+ +C= tan (x+1)+ +C
2 2 2
1+u x +2x+2

28. 5 4x x =
2
( x2+4x+4) +9=9 2
( x+2 ) . Let u=x+2 du=dx . Then
dx du 3cos  d where u=3sin  , du=3cos  d ,
 =  =
( 5 4x x2) 5/2 ( 9 u2) 5/2 (3cos  )
5
and
2
9 u =3cos 
1 1 1 1
=
81

4
sec  d =
81
 ( tan 2 +1) sec 2 d =
81 3
3
tan  +tan  +C
3 3
1 u 3u 1 (x+2) 3(x+2)
= 243 + +C= + +C
2 3/2 243 2 3/2
( 9 u ) 9 u
2
( 5 4x x ) 5 4x x
2

2
29. Let u=x , du=2xdx . Then
1 1 whereu=sin  ,du=cos  d ,
1 x dx = du =  cos   cos  d
2

4 1 u
x 2 2 2
and 1 u =cos 
1 1 1 1 1 1
= 
2 2
(1+cos 2 )d =  + sin 2 +C=  + sin  cos  +C
4 8 4 4
1  1 1 2 1  1 2 1 2 4
= sin u+ u 1 u +C= sin (x )+ x 1 x +C
4 4 4 4

30. Let u=sin t , du=cos tdt . Then

 /2 1

 0
cos t

1+sin t
2
dt =  0
1

1+u
2
dt= 
 /4

0
1
sec 
2
sec  d
where u=tan  , du=sec  d ,
and 1+u =sec 
2
2

 /4  /4
= sec  d = ln sec  +tan  [ by (1) in Section 8.2]
0 0

=ln ( 2 +1 )  ln (1+0)=ln ( 2 +1 )

  2 2
31. (a) Let x=atan  , where  < < . Then x +a =asec  and
2 2
dx
 asec  d
2
x +a
2
x
2
2
x +a
2 =  asec 
= sec  d =ln sec  +tan  +C =ln
1 a
+
a
+C
1

13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

(
= ln x+ x +a
2 2
) +C whereC=C  ln 1
a

2 2
(b) Let x=asinh t , so that dx=acosh t dt and x +a =acosh t . Then
dx acosh t dt  1 x
 =
acosh t
=t+C=sinh
a
+C .
2 2
x +a

 
32. (a) Let x=atan  , < < . Then
2 2
2 2 2 2 2
x a tan  tan  sec   1
 dx= asec  d = d =
2
I = d
2 3/2 3 3 sec  sec 
( x +a )2
a sec 
=  ( sec   cos  ) d =ln sec  +tan   sin  +C
2 2
= ln
x +a
a
+
x
a

x
2 2
(
+C=ln x+ x +a
2 2
) 2
x
2
+C
1
x +a x +a

(b) Let x=asinh t . Then

2 2
a sinh t
I =  3 3
acosh t dt= tanh t dt=
2
( 1 sech2t ) dt=t tanh t+C
a cosh t
 1 x x
= sinh a  2 2
+C
a +x

2
33. The average value of f (x)= x  1 /x on the interval 1,7 is
7 2  where x=sec  ,dx=sec  tan  d ,
x 1 tan 

1
1
dx = 6  sec 
 sec  tan  d 2  1
7 1 1
x 0 x  1 =tan  ,and  =sec 7
1  1 
 tan  d =  (sec   1)d
2 2
=
6 0 6 0
1  1
= tan    = (tan    )
6 0 6
1  1
=
6
48  sec 7 ( )
14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 2 3 2
34. 9x  4y =36 y= x  4
2
3
area = 2 x  4 dx=3
3 2 3 2
x  4 dx
2 2 2

where x=2sec  ,
 dx=2sec  tan  d ,
= 3 2tan  2sec  tan  d  1 3
0
 =sec
2
 
= 12
0
( sec   1 ) sec 
2
d =12
0
( 3
sec   sec  ) d
1 
= 12 ( sec  tan  +ln sec  +tan  )  ln sec  +tan 
2 0

= 6 sec  tan   ln sec  +tan 
0
3 5 3 5 9 5 3+ 5
=6  ln + =  6ln
4 2 2 2 2

1 1 2
(rcos  )(rsin  )= r sin  cos  . Area of region PQR=
r 2 2
35. Area of POQ= r  x dx . Let
2 2 rcos 

x=rcos u dx= rsin udu for   u . Then we obtain
2
1 2
 
2 2
 r  x dx
2 2 = rsin u ( rsin u)du= r sin udu= r (u sin ucos u)+C
2
1 2  1 1 2 2
=  r cos (x/r)+ x r  x +C
2 2
so
r
area of region = 1  r 2cos  1(x/r)+x r 2 x2
2 rcos 
1
=
2
2
0  r  +rcos  rsin  ( )

15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

1 2 1 2
= r  r sin  cos 
2 2
1 2
and thus, ( area of sector POR ) = ( area of POQ ) + ( area of region PQR ) = r .
2

 3
36. Let x= 2 sec  , where 0  < or    < , so dx= 2 sec  tan  d . Then
2 2
dx 2 sec  tan  d
 = 4
4 2
x x 2 4sec  2 tan 
1 1
=  cos  d =  1 sin  cos  d
4
3
4
2
( )
1 1 3
= sin   sin  +C [substitute u=sin  ]
4 3

=
1
2
x 2

( x2 2) 3/2 +C
4 x 3
3x

From the graph, it appears that our answer is reasonable.

2 2
37. From the graph, it appears that the curve y=x 4 x and the line y=2 x intersect at about x=0.81
2 2
and x=2 , with x 4 x >2 x on ( 0.81,2 ) . So the area bounded by the curve and the line is
1 2 2
 x 4 x  (2 x) dx=
2 2 2 2 2 2
A
x 4 x dx 2x x . To evaluate the integral, we put
0.81 0.81 2 0.81
 
x=2sin  , where     . Then
2 2

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

 1   1
dx=2cos  d , x=2  =sin 1= , and x=0.81  =sin 0.405
0.417 . So
2
 /2  /2  /2 1
 (2cos  )(2cos  d )=4 sin 2 d =4
2 2

2 2 2
x 4 x dx
0.417
4sin 
0.417 0.417 2
(1 cos 4 )d
0.81
1  /2  1
= 2   sin 4 =2  0  0.417 (0.995)
2.81
4 0.417 2 4
1 2 1 2
Thus, A
2.81 2 2  2  2 0.81  0.81
2.10 .
2 2

2 2 2
38. Let x=btan  , so that dx=bsec  d and x +b =bsec  .
L a

E ( P) =
  a
4 
0
(
 b
2 2 3/2
x +b )
dx=
 b
4 
0



2

1
1
( bsec  )
3
bsec  d
2

 

2
1 
2
 
=  sec 
d =  cos  d = sin 
2

4  b 4  b 4  b 
0  1 0  1 0 1

 x L a  L a a
= 4  = +
b 2 2 4  b 2 2 2 2
0 x +b  a 0 (L a) +b a +b

2 2 2
39. Let the equation of the large circle be x +y =R . Then the equation of the small circle is
2 2 2 2 2
x +(y b) =r , where b= R  r is the distance between the centers of the circles. The desired area is
17
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

A= 
r
 r
( b+ 2
r  x
2
) R x
2 2
dx=2
r
0
( b+ 2
r  x 
2
R x
2 2
) dx
= 2 bdx+2 r  x dx 2
r r 2 2 r 2 2
R  x dx
0 0 0
2 2
The first integral is just 2br=2r R  r . To evaluate the other two integrals, note that
1 2
a  x dx =  a cos  d [x=asin  ,dx=acos  d ] = 2 a  ( 1+cos 2 ) d
2 2

2 2

1 2 1 1 2
= a  + sin 2 +C= a ( +sin  cos  )+C
2 2 2
2 2 2 2 2
= a arcsin x a x a x a x x 2 2
+ +C= arcsin + a  x +C
2 a 2 a a 2 a 2
so the desired area is
r r
2 2 2 2 2 2 2 2
A = 2r R  r + r arcsin(x/r)+x r  x  R arcsin(x/R)+x R x
0 0
2 2 2  2 2 2 2 2  2 2
= 2r R  r +r  R arcsin(r/R)+r R r =r R r + r  R arcsin(r/R)
2 2

40. Note that the circular cross sections of the tank are the same everywhere, so the percentage of the
total capacity that is being used is equal to the percentage of any cross section that is under water.
The underwater area is

A = 2
2 2
25 y dy
 5
2
2
= 25 arcsin(y/5)+y 25 y y=5sin  ]
 5
2 25 2
= 25 arcsin +2 21 + 
58.72ft
5 2
A 58.72
so the fraction of the total capacity in use is

0.748 or 74.8% .
2 25
 (5)

18
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.3 Trigonometric Substitution

2 2 2 2 2
41. We use cylindrical shells and assume that R>r . x =r  (y R)  x= r  (y R) , so
2 2
g(y)=2 r  (y R) and

 2 y 2 r  (y R) dy=
R+r 2 2 r 2 2
V = 4 (u+R) r  u du [where u=y R]
R r  r

 u r  u du+4 R
r 2 2 r 2 2
= 4 r  u du where u=rsin  ,du=rcos  d
 r  r in the second tegral
1 2 2 r  /2 2 4  /2
= 4 
3
(
r  u ) 3/2  r
+4 R
  /2
2
r cos  d =
3
(0 0)+4 Rr 
2
  /2
2
cos  d

2  /2 1  /2

2 2 2
= 2 Rr (1+cos 2 )d =2 Rr  + sin 2 =2 Rr
  /2 2   /2

Another method: Use washers instead of shells, so V =8 R

r 2 2
r  y dy as in Exercise 6.2.61(a) , but
0
evaluate the integral using y=rsin  .

19
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

2x A B
1. (a) = +
(x+3)(3x+1) x+3 3x+1
1 1 1 A B C
(b) 3 2 = = = + +
2 2 x x+1 2
x +2x +x x(x +2x+1) x(x+1) (x+1)

x 1 A B C x 1
2. (a) + + = =
3 2 2 x 2 x+1
x +x x (x+1) x
x 1 x 1 A Bx+C
(b) 3 = = + 2
2
x +x x x +1
x
x +1 ( )
2 2 A B
3. (a) = = +
2 (x+4)(x 1) x+4 x 1
x +3x 4
2
2 x A Bx+C
(b) x +x+1 is irreducible, so = + 2 .
(x 1) x +x+1 ( 2
) x 1
x +x+1

3
x 13x+12 13x+12 A B
4. (a) =x 4+ =x 4+ =x 4+ +
2 2 (x+1)(x+3) x+1 x+3
x +4x+3
x +4x+3
A
2x+1 B C Dx+E Fx+G
(b) = + + + 2 +
3 2 2 x+1 2 3 2 2
(x+1) (x +4) (x+1) (x+1) x +4 (x +4)

5. (a)
4 4
x (x  1)+1 1 1
4
= 4
=1+ 4
[ or use long division] =1+ 2 2
x 1 x 1 x 1 (x  1)(x +1)
1 A B Cx+D
=1+ =1+ + + 2
2 x 1 x+1
(x 1)(x+1)(x +1) x +1

4 2
t +t +1 At+B Ct+D Et+F
(b) = 2
+ 2
+
( t2+1) ( t2+4) 2 t +1 t +4 ( t2+4) 2
4 4 3
x x x Ax+B Cx+D
6. (a) 3 2
= 2 2
= 2 2
= 2
+ 2
(x +x)(x  x+3) x(x +1)(x  x+3) (x +1)(x  x+3) x +1 x  x+3
(b)

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

1 1 1 A B C D Ex+F
= = = + + + + 2
6
x x
3
x
3
( x3 1) 3
( 2
x (x 1) x +x+1 ) x
x
2
x
3 x 1
x +x+1

x (x 6)+6 6
7.  dx= dx= 1+ dx=x+6ln x 6 +C
x 6 x 6 x 6

8.
2 2
r r  16 16 16
 r+4
dr = +
r+4 r+4
dr= r 4+
r+4
dr [ or use long division]

1 2
= r  4r+16ln r+4 +C
2

x 9 A B
9. = + . Multiply both sides by (x+5)(x 2) to get x 9=A(x 2)+B(x+5) .
(x+5)(x 2) x+5 x 2
Substituting 2 for x gives  7=7B B= 1 . Substituting  5 for x gives  14= 7A A=2 . Thus,
x 9 2  1
 (x+5)(x 2) dx= x+5 + x 2 dx=2ln x+5  ln x 2 +C

1 A B
10. = +  1=A(t 1)+B(t+4) .
(t+4)(t 1) t+4 t 1
1 1
t=1 1=5B B= . t= 4 1= 5A A= . Thus,
5 5
1  1/5 1/5 1 1 1 t 1
 (t+4)(t 1) dt= t+4 + t 1 dt= 5 ln t+4 + 5 ln t 1 +C or 5 ln t+4
+C

1 1 A B
11. = = + . Multiply both sides by (x+1)(x 1) to get 1=A(x 1)+B(x+1) .
2 (x+1)(x 1) x+1 x 1
x 1
1 1
Substituting 1 for x gives 1=2B B= . Substituting  1 for x gives 1= 2A A= . Thus,
2 2
3 3
 1/2 1/2
 1 1 3
dx =
1
+ dx=  ln x+1 + ln x 1
2 x+1 x 1 2 2 2
2 x 1 2

1 1 1 1 1 1 3
=  ln 4+ ln 2   ln 3+ ln 1 = (ln 2+ln 3 ln 4) or ln
2 2 2 2 2 2 2

x 1 A B
12. = + . Multiply both sides by (x+1)(x+2) to get x 1=A(x+2)+B(x+1) .
2 x+1 x+2
x +3x+2
Substituting  2 for x gives  3= B  B=3 . Substituting  1 for x gives  2=A . Thus,
2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

1 1
 2

3
dx =
x 1 1
+ dx=  2ln x+1 +3ln x+2
2 0 x+1 x+2 0
0 x +3x+2
27
=( 2ln 2+3ln 3) ( 2ln 1+3ln 2)=3ln 3 5ln 2 orln
32

ax ax a
13.  dx= dx= dx=aln x b +C
2 x(x b) x b
x  bx

1 1 1 1
14. If a b , =  , so if a b , then
(x+a)(x+b) b a x+a x+b

dx 1 1 x+a
 =
(x+a)(x+b) b a
( ln x+a  ln x+b ) +C=
b a
ln
x+b
+C

dx 1
If a=b , then  = +C .
2 x+a
(x+a)

2x+3 A B
15. = +  2x+3=A(x+1)+B . Take x= 1 to get B=1 , and equate coefficients of x
2 x+1 2
(x+1) ( x+1 )
to get A=2 . Now
1 1

 2x+3
2
dx =  2
x+1
+
1
2
dx= 2ln (x+1)
1
x+1
1

0
0 (x+1) 0 ( x+1 )
1 1
= 2ln 2  ( 2ln 1 1 ) =2ln 2+
2 2

3
x  4x 10 3x 4 3x 4 A B
16. =x+1+ . Write = + . Then 3x 4=A(x+2)+B(x 3) .
2 (x 3)(x+2) (x 3)(x+2) x 3 x+2
x  x 6
Taking x=3 and x= 2 , we get 5=5A A=1 and  10= 5B B=2 , so
1 1
3

1
x  4x 10 1 2 1 2
2
dx = x+1+ +
x 3 x+2
dx=
2
x +x+ln x 3 +2ln (x+2)
0
0 x  x 6 0

1 3 3 3
= +1+ln 2+2ln 3  ( 0+0+ln 3+2ln 2 ) = +ln 3 ln 2= +ln
2 2 2 2

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

2
4y  7y 12 A B C 2
17. = + +  4y  7y 12=A(y+2)(y 3)+By(y 3)+Cy(y+2) . Setting y=0 gives
y(y+2)(y 3) y y+2 y 3
9 1
 12= 6A , so A=2 . Setting y= 2 gives 18=10B , so B= . Setting y=3 gives 3=15C , so C= .
5 5
Now
2 2 2
4y  7y 12 2 9/5 1/5 9 1 2
 y(y+2)(y 3)
dy =  + +
y y+2 y 3
dy= 2ln y + ln y+2 + ln y 3
5 5 1
1 1
9 1 9 1
= 2ln 2+ ln 4+ ln 1 2ln 1 ln 3 ln 2
5 5 5 5
18 1 9 27 9 9 9 8
= 2ln 2+ ln 2 ln 2 ln 3= ln 2 ln 3= (3ln 2 ln 3)= ln
5 5 5 5 5 5 5 3

2 2
x +2x 1 x +2x 1 A B C
18. = = + + . Multiply both sides by x(x+1)(x 1) to get
3 x(x+1)(x 1) x x+1 x 1
x x
2
x +2x 1=A(x+1)(x 1)+Bx(x 1)+Cx(x+1) . Substituting 0 for x gives  1= A A=1 . Substituting  1
for x gives  2=2B B= 1 . Substituting 1 for x gives 2=2C  C=1 . Thus,
2
x +2x 1 1 1 1 x(x 1)
 3
dx=  +
x x+1 x 1
dx=ln x  ln x+1 +ln x 1 +C=ln
x+1
+C .
x x

1 A B C 2
19. = + +  1=A(x+5)(x 1)+B(x 1)+C(x+5) . Setting x= 5 gives
2 x+5 2 x 1
( x+5) (x 1) ( x+5)
1 1
1= 6B , so B= . Setting x=1 gives 1=36C , so C= . Setting x= 2 gives
6 36
1 1 3 1 1
2
( )
1=A(3)( 3)+B( 3)+C 3 = 9A 3B+9C= 9A+ + = 9A+ , so 9A=
2 4 4 4
and A=
36
. Now
1  1/36 1/6 1/36
 2
dx =  x+5

2
+
x 1
dx
( x+5) (x 1) ( x+5)
1 1 1
=  ln x+5 + + ln x 1 +C
36 6(x+5) 36

2
x A B C 2 2
20. = + +  x =A(x+2) +B(x 3)(x+2)+C(x 3) .
2 x 3 x+2 2
(x 3)(x+2) (x+2)
9 4 2
Setting x=3 gives A= . Take x= 2 to get C= , and equate the coefficients of x to get 1=A+B
25 5
4
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

16
 B= . Then
25
2 9/25 16/25 4/5
 x
dx =  x 3
+
x+2

2
dx
2
(x 3)(x+2) (x+2)
9 16 4
= ln x 3 + ln x+2 + +C
25 25 5(x+2)

2 2
5x +3x 2 5x +3x 2 A B C 2
21. = = + + . Multiply by x (x+2) to get
3 2 2 x 2 x+2
x +2x x ( x+2 ) x
2 2
5x +3x 2=Ax(x+2)+B(x+2)+Cx . Set x= 2 to get C=3 , and take x=0 to get B= 1 . Equating the
2
coefficients of x gives 5=A+C  A=2 . So
2
5x +3x 2 2 1 3 1
 3 2
dx=
x
 +
2 x+2
dx=2ln x +
x
+3ln x+2 +C .
x +2x x

1 A B C D 2 2 2 2
22. = + + +  1=As(s 1) +B(s 1) +Cs (s 1)+Ds . Set s=0 , giving B=1 .
2 2 s 2 s 1 2
s (s 1) s (s 1)
3
Then set s=1 to get D=1 . Equate the coefficients of s to get 0=A+C or A= C , and finally set s=2 to
get 1=2A+1 4A+4 or A=2 . Now
2 1 2 1 1
 2 ds 2 = s
+ 
2 s 1
+
1
2
ds=2ln s 
s
 2ln s 1 
s 1
+C .
s (s 1) s (s 1)

2
x A B C 3 2 2
23. = + + . Multiply by ( x+1 ) to get x =A(x+1) +B(x+1)+C . Setting
3 x+1 2 3
(x+1) (x+1) (x+1)
2
x= 1 gives C=1 . Equating the coefficients of x gives A=1 , and setting x=0 gives B= 2 . Now
2
x dx 1 2 1 2 1
 3
=
x+1

2
+
3
dx=ln x+1 +
x+1

2
+C .
(x+1) (x+1) (x+1) 2(x+1)

3 3
x (x+1) 1 1 x 1 3 3 1
24. = =1 , so = 1 =1 +  . Thus,
x+1 x+1 x+1 3 x+1 x+1 2 3
(x+1) (x+1) (x+1)
3
x 3 3 1 3 1
 3
dx= 1
x+1
+
2

3
dx=x 3ln x+1 
x+1
+
2
+C .
(x+1) (x+1) (x+1) 2(x+1)
5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

10 A Bx+C
25. = + 2
2
. Multiply both sides by (x 1) x +9 to get ( )
(x 1) x +9 ( 2
) x 1
x +9
( 2
)
10=A x +9 +(Bx+C)(x 1) ( * ). Substituting 1 for x gives 10=10A A=1 . Substituting 0 for x gives
2
10=9A C  C=9(1) 10= 1 . The coefficients of the x  terms in ( * ) must be equal, so 0=A+B
 B= 1. Thus,
10 1  x 1 1 x 1
 2
dx =  x 1
+ 2
dx=  x 1
 2
 2
dx
(x 1)(x +9) x +9 x +9 x +9
1 1  1 x
=ln x 1 
2
2 2
ln x +9 [let u=x +9] 
3
tan
3
( )
[Formula10] +C

2 2
x  x+6 x  x+6 A Bx+C
26. = = + 2
2 2 2
(
. Multiply by x x +3 to get x  x+6=A x +3 +(Bx+C)x . ) ( )
x ( x +3)
3 2 x
x +3x x +3
2
Substituting 0 for x gives 6=3A A=2 . The coefficients of the x  terms must be equal, so 1=A+B
B=1 2= 1 . The coefficients of the x  terms must be equal, so  1=C . Thus,
2
x  x+6 2  x 1 2 x 1
 3
dx = x
+ 2 dx=
x
 2
 2
dx
x +3x x +3 x +3 x +3
1 1  x
=2ln x 
2
2
(
ln x +3  ) 3
tan
1
3
+C

3 2
x +x +2x+1 Ax+B Cx+D
27. = + . Multiply both sides by x +1 ( 2
) ( x2+2) to get
( x2+1) ( x2+2) x +1
2
x +2
2

3 2
x +x +2x+1=(Ax+B) x +2 +(Cx+D) x +1  ( 2
) ( 2
)
3 2
(
x +x +2x+1= Ax +Bx +2Ax+2B + Cx +Dx +Cx+D 
3 2
) ( 3 2
)
3 2 3 2
x +x +2x+1=(A+C)x +(B+D)x +(2A+C)x+(2B+D) . Comparing coefficients gives us the following
system of equations:

A+C=1 (1) B+D=1 (2)

2A+C=2 (3) 2B+D=1 (4)

Subtracting equation (1) from equation (3) gives us A=1 , so C=0 . Subtracting equation (2) from
3 2
x +x +2x+1 x 1
equation (4) gives us B=0 , so D=1 . Thus, I= dx= + dx . For
( x +1) ( x +2)
2 2 2
x +1
2
x +2
6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

x x 1 1 1 1
 2
dx , let u=x +1 so du=2xdx and then 
2
2
dx= 
2 u
2
du= ln u +C= ln x +1 +C . For
2 2
( )
x +1 x +1
1 1 1 1  x
 dx , use Formula 10 with a= 2 . So  dx=
1
2 2
dx= tan +C . Thus,
2 2 2 2
x +2 x +2 x +( 2 )
1 1  x
2
2
I= ln x +1 +
2
tan ( ) 1
2
+C .

2
x  2x 1 A Cx+D
28. = +
B
+ 
2
( 2
x  2x 1=A(x 1) x +1 +B x +1 +(Cx+D) ( x 1 ) . ) ( 2
) 2

( x 1 )
2
( x2+1) x 1
(x 1)
2
x +1
2

3
Setting x=1 gives B= 1 . Equating the coefficients of x gives A= C . Equating the constant terms
gives  1= A 1+D , so D=A ,
and setting x=2 gives  1=5A 5 2A+A or A=1 . We have
2
x  2x 1 1 1 x 1
 dx =  x 1

2
 2
dx
( x 1 )
2
( x +1) 2
(x 1) x +1
1 1  1
= ln x 1 + 
x 1 2
2
ln x +1 +tan x+C ( )
29.
x+4 x+1 3 1 (2x+2)dx 3dx
 2
dx = 2
dx+ 2
dx=
2
 2
+ 2
x +2x+5 x +2x+5 x +2x+5 x +2x+5 (x+1) +4
1 2du
= ln x +2x+5 +3
2 where x+1=2u,
2 2 and dx=2du
4(u +1)
1 2 3  1 1 2 3  1 x+1
= ln (x +2x+5)+ tan u+C= ln (x +2x+5)+ tan +C
2 2 2 2 2

3 2 3 2
x  2x +x+1 x  2x +x+1 Ax+B Cx+D
30. = = + 
3 2
x  2x +x+1= ( Ax+B ) x +4 + (Cx+D ) x +1 ( 2
) ( 2
)
x +5x +4
4
x +1 x +4
2
x +1 x +4 ( 2
)( 2
) 2 2

. Equating coefficients gives A+C=1 , B+D= 2 , 4A+C=1 , 4B+D=1 A=0 , C=1 , B=1 , D= 3 . Now
3 2
x  2x +x+1 dx x 3  1 1 3  1
 4 2
dx= 2
+ 2
dx=tan x+
2
2
ln x +4  (
2
)
tan (x/2)+C .
x +5x +4 x +1 x +4

1 1 A Bx+C
31. = = + 2  ( 2
1=A x +x+1 +(Bx+C)(x 1) . )
x 1
3
(
(x 1) x +x+1
2
) x 1
x +x+1
7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

1 2
Take x=1 to get A= . Equating coefficients of x and then comparing the constant terms, we get
3
1 1 1 2
0= +B , 1=  C , so B= , C= 
3 3 3 3
1 1 2
1  x
 3 dx =  3 dx+ 3 3 dx= 1 ln x 1  1  x+2 dx
x 1 x 1 2 3 3 2
x +x+1 x +x+1

.
1 1 x+1/2 1 ( 3/2 ) dx
= 3 ln x 1  3  2 dx
3
 2
x +x+1 ( x+1/2 ) +3/4
1
x+
= 1 1 1 2  1 2
ln x 1 
2
ln x +x+1  tan ( +K )
3 6 2 3 3 2 /
1 1 1  1 1
=
3
ln x 1 
6
2
ln x +x+1 
3
tan
3
(
(2x+1) +K )
32.
1 1 1


1
(2x+4)
 2
x
dx = 2
2
dx 2  dx
2
0 x +4x+13 0 x +4x+13 0 (x+2) +9
18 1 2
1 dy 3
 y  2  du 9u +9
2
= where y=x +4x+13,dy=(2x+4)dx,
2 13 2/3 x+2=3u, and dx=3du
1 18 2  1 1 1 18 2   1 2
= ln y  tan u = ln   tan
2 13 3 2/3 2 13 3 4 3
1 18  2  1 2
= ln  + tan
2 13 6 3 3

3 2
33. Let u=x +3x +4 . Then du=3 x +2x dx ( 2
)
5
2 204
 3
x +2x
2
dx=
1
3
 du
u=
1
3
ln u
204
24
=
1
3
1
(ln 204 ln 24)= ln
3
204 1
= ln
24 3
17
2
.
2 x +3x +4 24

34.
x
3
=
( x +1 )  1
3
A
=1
+
Bx+C
 1=A x
1
2
 x+1 +(Bx+C)(x+1) . Equate the
=1 ( )
3 3 3 x+1 2
x +1 x +1 x +1 x  x+1
terms of degree 2 , 1 and 0 to get 0=A+B , 0= A+B+C , 1=A+C . Solve the three equations to get
8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

1 1 2
A= , B= , and C= . So
3 3 3
3 1 1 2
x x
 3
dx =  1
3 3
+ 2
3
dx
x +1 x+1
x  x+1
1 1 2x 1 1 dx
x ln x+1 +  2 dx 
= 3 6 2 1 2 3
x  x+1 x +
2 4
1 1 1  1 1
= x
3 6
2
ln x+1 + ln x  x+1  ( ) 3
tan
3
( 2x 1 ) +K

1 1 A B C D 2
35. = = + + + . Multiply by x (x 1)(x+1) to get
4 2 2 2
x x x (x 1)(x+1) x x x 1 x+1
2 2 1
1=Ax(x 1)(x+1)+B(x 1)(x+1)+Cx (x+1)+Dx (x 1) . Setting x=1 gives C= , taking x= 1 gives
2
1 3
D= . Equating the coefficients of x gives 0=A+C+D=A . Finally, setting x=0 yields B= 1 . Now
2
dx  1 1/2 1/2 1 1 x 1
 4 2
=
2
+  dx= + ln
2
+C .
x x x x 1 x+1 x x+1

36. Let u=x +5x +4

4 2
( 3
)
du= 4x +10x dx=2 2x +5x dx , so ( 3
)
1
3 10
 2x +5x
4 2
dx=
1
2

du 1
=
u 2
ln u
10
4
=
1
2
1
(ln 10 ln 4)= ln
2
5
2
.
0 x +5x +4 4

x 3 x 3 u 4
37.  dx= dx= du [ with u=x+1 ]
2 2 2
( x +2x+4)
2
( x+1 ) +3
2
( u +3)
2

2
udu 1 dv 3sec  d
du 2
=  4 =   4 v=u +3 in the first integral;
2 2 2 2 4
2
u +3 ( 2
)
u +3 v( 9sec  ) u= 3 tan  in the second
 1 4 3  1 2 3

2
=  cos  d =  ( +sin  cos  ) +C
( 2v ) 9
2 u +3
2 9 ( )
 1 2 3  1 x+1 3 (x+1)
=  tan + +C
2
2 x +2x+4 ( 9 ) 3 2
x +2x+4

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

 1 2 3  1 x+1 2(x+1)
=  tan  +C
2 ( x +2x+4) 3 ( x +2x+4)
2 9 3 2

4
A Bx+C
38.
x +1
2
=
x
+ 2 +
Dx+E
2

4
( 2
)2
x +1=A x +1 +(Bx+C)x x +1 +(Dx+E)x . Setting x=0 ( 2
)
(
x x +1
2
) x +1 2
x +1 ( )
4
gives A=1 , and equating the coefficients of x gives 1=A+B , so B=0 . Now
C
+
Dx+E
=
4
x +1

1
=
1
4
x +1 ( x4+2x2+1) =
 2x
, so we can take C=0 ,
2 2 2
x +1 ( x +1)
2
(
x x +1
2
) x x ( x2+1) 2 ( x +1)
2 2

4
x +1 1 2x 1
D= 2 , and E=0 . Hence,  dx=  dx=ln x + 2
+C .
2 2
x x +1 ( 2
) x ( x +1) 2
x +1

2
39. Let u= x+1 . Then x=u  1 , dx=2udu
dx 2udu du u 1 x+1  1
 = =2 2 =ln +C=ln +C .
x x+1 2
u 1 u u 1 (u+1 ) x+1 +1

dx 2udu udu
= 2 =2 2 I=
2
40. Let u= x+2 . Then x=u  2 , dx=2udu and
x x+2 u  2 u u  u 2
u A B 1
= +  u=A(u+1)+B(u 2) . Substituting  1 for u gives  1= 3B B= and
2 u 2 u+1 3
u  u 2
2
substituting 2 for u gives 2=3A A= . Thus,
3
2 2 1 2
I =
3
 +
u 2 u+1
du= ( 2ln u 2 +ln u+1 ) +C
3
2
= 2ln x+2  2 +ln ( x+2 +1 ) +C
3

2
41. Let u= x , so u =x and dx=2udu . Thus,
16 4 4 4
2

x
x 4
dx =  2
u
2udu=2  u
2
du=2  1+ 2
4
du [by long division]
9 3 u 4 3 u 4 3 u 4
4
=2+8 (u+2)(u 2). ( * )
3 du

10
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

1 A B
Multiply = + by (u+2)(u 2) to get 1=A(u 2)+B(u+2) . Equating coefficients we
(u+2)(u 2) u+2 u 2
1 1 1  1/4 1/4
get A+B=0 and  2A+2B=1 . Solving gives us B= and A= , so = + and ( *
4 4 (u+2)(u 2) u+2 u 2
) is
4
 1/4 1/4 1 1 4
2+8 + du = 2+8  ln u+2 + ln u 2
3 u+2 u 2 4 4 3

4 u 2 4
= 2+ 2ln u 2  2ln u+2 =2+2 ln
3 u+2 3
2 1 2/6
= 2+2 ln  ln =2+2ln
6 5 1/5
5 5 2 25
= 2+2ln or2+ln =2+ln
3 3 9

3 3 2
42. Let u= x . Then x=u , dx=3u du
1 1 2 1
 3u 3 3 2 1
1
3
dx =  du
1+u= 3u 3+
1+u
du=
2
u  3u+3ln (1+u)
0
0 1+ x 0 0

1
= 3 ln 2
2

3 2 2 3 2
43. Let u= x +1 . Then x =u  1 , 2xdx=3u du
3 3 2

x dx
=
3
u 1
2
u du ( 3
) 3 5 3 2
3 2
x +1
 u
=  u  u du=
2
4
10
u
4
u +C ( )
3 5/3 3 2/3
=
10
2
x +1 
4
2
(
x +1 +C ) ( )
3 3 3
44. Let u= x . Then x=u , dx=2udu
2
 2
x
dx=  u 2udu
4 2
=2 
du
2
1/3 x +x 1/ 3 u +u 1/ 3 u +1
 1 3   
=2 tan u =2  =
1/ 3 3 6 3

45. If we were to substitute u= x , then the square root would disappear but a cube root would
remain. On the other hand, the substitution
11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

3
u= x would eliminate the cube root but leave a square root. We can eliminate both roots by means
6
of the substitution u= x . (Note that 6 is the least common multiple of 2 and 3 .)
6 6 5 3 3 2
Let u= x . Then x=u , so dx=6u du and x =u , x =u . Thus,
5 5 3
dx
 3 =  6u du
=6
u
du=6
u
du
x x 3 2 2 u 1
u u u (u 1)
1
= 6
2
u +u+1+ du [by long division]
u 1
1 3 1 2 3 6 6
= 6 u + u +u+ln u 1 +C=2 x +3 x +6 x +6ln x  1 +C
3 2

12 12 11
46. Let u= x . Then x=u , dx=12u du
11 8
dx
 3 4 =  12u du
=12
u du
=12 du
7 6 5 4 3
u  u +u  u +u  u +u 1+
2 1
x+ x 4 3 u+1 u+1
u +u
3 8 12 7 6 12 5 4 3 2
= u u +2u  u +3u  4u +6u  12u+12ln u+1 +C
2 7 5
3 2/3 12 7/12 12 5/12 3
=
2
x 
7
x +2 x 
5
4 6 12
x +3 x  4 x +6 x  12 x +12ln ( 12 x +1) +C
x du
47. Let u=e . Then x=ln u , dx= 
u
2x 2
e dx u ( du/u ) udu  1 2
 2x x
=  2
=
(u+1)(u+2)
= +
u+1 u+2
du
e +3e +2 u +3u+2
= 2ln u+2  ln u+1 +C=ln (ex+2)2 /(ex+1) +C

48. Let u=sin x . Then du=cos xdx

cos xdx du du 1 1
 2 =  2
=  u(u+1)
= 
u u+1
du
sin x+sin x u +u
u sin x
= ln +C=ln +C
u+1 1+sin x

2x 1
( 2
49. Let u=ln x  x+2 , dv=dx . Then du= ) 2
dx , v=x , and (by integration by parts)
x  x+2
12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

2
2x  x x 4
 ( 2
)
ln x  x+2 dx =xln x  x+2  ( 2
)  2
dx=xln x  x+2 
2
 2+ 2
dx
x  x+2 x  x+2
1
(2x 1)
2 7 dx
2
(
=xln x  x+2  2x ) 2
dx+
2

1 2 7
x  x+2 x +
2 4
7
du
1 7 2
( 2
=xln x  x+2  2x ) 2
ln x  x+2 + 
2
2
( ) 7 2
(u +1)
4
1 7
where x = u,
2 2
7
dx= du,
2
1 2 7 7 2
x + = (u +1)
2 4 4
1  1
= x
2
( 2
ln x  x+2  2x+ 7 tan u+C )
1  1 2x 1
= x
2
( 2
ln x  x+2  2x+ 7 tan ) 7
+C

 1
50. Let u=tan x , dv=xdx du=dx/ 1+x ( 2
) , v= 12 x2 .
2
 1
1 2 1 1 x
Then  xtan xdx= x tan x  dx . To evaluate the last integral, use long division or
2 2 2
1+x

observe that 
2
x
dx=
( 1+x )  1 dx=
2
1dx  1  1
dx=x tan x+C . So
2 2 2 1
1+x 1+x 1+x
 1 1 2 1 1  1 1 2 1  1
 xtan xdx= x tan x
2 2
x tan x+C =
1 2
x tan x+tan x x +C . ( )
51.

13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

From the graph, we see that the integral will be negative, and we guess that the area is about the same
as that of a rectangle with width 2 and height 0.3 , so we estimate the integral to be  ( 2 0.3) = 0.6 .
1 1 A B 1
Now 2 = = +  1=(A+B)x+A 3B , so A= B and A 3B=1 A= and
(x 3)(x+1) x 3 x+1 4
x  2x 3
1
B= , so the integral becomes
4
2 2 2
1 1 1
  
2 2
x  2x 3 = x 3 x+1= ln x 3  ln x+1
0 dx 4 0 dx 4 0 dx 4 0

1 x 3 2 1 1 1
= ln = ln  ln 3 = ln 3  0.55
4 x+1 0 4 3 2

1 1 A B C 2
52. = + = +  1=(A+C)x +(B 2A)x 2B , so A+C=B 2A=0 and
3 2 2 x 2 x 2
x  2x x (x 2) x
1 1 1 1
 2B=1  B= , A= , and C= . So the general antiderivative of is
2 4 4 3 2
x  2x
 3dx 2 = 14  dx 1 dx 1 dx

x 2
 2 + 4  x 2
x  2x x
1 1 1
= ln x  (  1/x ) + ln x 2 +C
4 2 4
1 x 2 1
= ln + +C
4 x 2x

1
We plot this function with C=0 on the same screen as y= .
3 2
x  2x

14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

53.
dx dx du
 2
= 2
= 2
x  2x (x 1)  1 u 1
1 u 1 1 x 2
= ln +C [ by Equation 6] = ln +C
2 u+1 2 x

54.
(2x+1)dx 1 (8x+12)dx 2dx
 2 4
 2
=   2
4x +12x 7 4x +12x 7 (2x+3)  16
1 du
= ln 4x +12x 7   2
2
4
u  16
1 2 1
= ln 4x +12x 7  ln (u 4)/(u+4) +C
4 8
1 2 1
= ln 4x +12x 7  ln (2x 1)/(2x+7) +C
4 8

x x  1 x 1
55. (a) If t=tan , then =tan t . The figure gives cos = and
2 2 2 2
1+t
x t
sin = .
2 2
1+t

x 2 x
(b) cos x=cos 2 =2cos  1
2 2
2 2
1 2 1 t
=2  1=  1=
2 2 2
1+t 1+t 1+t
x x x t 1 2t
sin x=sin 2 =2sin cos =2 =
2 2 2 2 2 2
1+t 1+t 1+t
(c)
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

x 2
=arctant  x=2arctant dx= dt
2 2
1+t

2 2t
56. Let t=tan (x/2) . Then, using Exercise 55 , dx= dt , sin x= 
2 2
1+t 1+t


dx
=  2dt/ 1+t( 2
) =
2dt
=2
dt
3 5sin x
3 10t/ 1+t ( 2
) (
3 1+t  10t
2
) 3t  10t+3
2

1 1 3 1 1 tan (x/2) 3
=
4
 
t 3 3t 1
dt= ( ln t 3  ln 3t 1 ) +C= ln
4 4 3tan (x/2) 1
+C

57. Let t=tan (x/2) . Then, the expressions in Exercise 55 , we have

1 2dt dt dt
1
 =2 =
 3sin x 4cos x dx = 3 2t  4 1 t
2
1+t
2
3(2t) 4 1 t
2 2
2t +3t 2 ( )
2 2
1+t 1+t
dt 2 1 1 1
=  =  dt [using partial fractions]
(2t 1)(t+2) 5 2t 1 5 t+2
1 1 2t 1 1 2tan ( x/2 )  1
= ln 2t 1  ln t+2 +C= ln +C= ln +C
5 5 t+2 5 tan ( x/2 ) +2

58. Let t=tan (x/2) . Then, by Exercise 55 ,

 /2 1 1
dx
 1+sin x cos x =  ( 2
) 2dt/ 1+t
= 
2dt
1/ 3 1+2t/ ( 1+t )  ( 1 t ) / ( 1+t ) 1/ 3 1+t +2t 1+t
2 2 2 2 2
 /3
1
1 1 1 1 3 +1

1
=  dt= ln t ln ( t+1 ) =ln  ln =ln
t t+1 1/ 3 2 3 +1 2
1/ 3

59. Let t=tan (x/2) . Then, by Exercise 55 ,

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions


dx 1 dx 1
= 2  sin x+sin xcos x = 2 
( 2) 2dt/ 1+t
2 2
2t/ ( 1+t ) +2t ( 1 t ) / ( 1+t )
2sin x+sin 2x 2 2

( 1+t2) dt 1 ( 1+t ) dt 1
2
1 1
=  =  =  +t dt
t ( 1+t ) +t ( 1 t )
2 2 2 4 t 4 t

1 1 2 1 1 1 2 1
= ln t + t +C= ln tan x + tan x +C
4 8 4 2 8 2

2 2
60. x  6x+8=(x 3)  1 is positive for 5
x
10 , so
10 7 7
1 u 1
 (x 3)  1=
2 2
area = u  1 [putu=x 3] = ln
5 dx 2 du 2 u+1 2

1 3 1 1 1 3
= ln  ln = (ln 3 2ln 2+ln 3)=ln 3 ln 2=ln
2 4 2 3 2 2

x+1 2
61. =1+ >0 for 2
x
3 , so area
x 1 x 1
3
2
=
3
1+ dx= x+2ln x 1 = ( 3+2ln 2 )  ( 2+2ln 1 ) =1+2ln 2 .
2 x 1 2

1 2 dx
 
1 1
62. (a) We use disks, so the volume is V = 2
dx= . To evaluate the
0 0 2 2
x +3x+2 (x+1) (x+2)
1 A B C D
integral, we use partial fractions: = + + + 
2 2 x+1 2 x+2 2
(x+1) (x+2) (x+1) (x+2)
2 2 2 2
1=A(x+1)(x+2) +B(x+2) +C(x+1) (x+2)+D(x+1) . We set x= 1 , giving B=1 , then set x= 2 , giving
3
D=1 . Now equating coefficients of x gives A= C , and then equating constants gives
1=4A+4+2( A)+1 A= 2 C=2 . So the expression becomes
 2 1 2 1 x+2 1 1 1

1
V = 0 x+1
+
2
+
( x+2 )
+
2
dx= 2ln
x+1
 
x+1 x+2 0
( x+1 ) ( x+2 )
3 1 1 1 3/2 2 2 9
= 2ln    2ln 2 1 = 2ln + = +ln
2 2 3 2 2 3 3 16

xdx xdx
 
1 1
(b) In this case, we use cylindrical shells, so the volume is V =2 =2 . We
0 2 0 (x+1)(x+2)
x +3x+2
17
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

x A B
use partial fractions to simplify the integrand: = +  x=(A+B)x+2A+B . So
(x+1)(x+2) x+1 x+2
A+B=1 and 2A+B=0 A= 1 and B=2 . So the volume is
 1 2
2 
1 1
+ dx = 2  ln x+1 +2ln x+2
0 x+1 x+2 0
9
= 2 ( ln 2+2ln 3+ln 1 2ln 2)=2 (2ln 3 3ln 2)=2 ln
8

P+S A B
63. = +  P+S=A (r 1)P S +BP= (r 1)A+B P AS  (r 1)A+B=1 ,  A=1
P (r 1)P S P (r 1)P S
 A= 1 , B=r . Now

P+S  1 r dP r r 1
t= dP= + dP=  +  dP
P (r 1)P S P (r 1)P S P r 1 (r 1)P S

r
so t= ln P+ ln (r 1)P S +C . Here r=0.10 and S=900 , so
r 1
0.1 1
t = ln P+ ln  0.9P 900 +C= ln P ln (  1 0.9P+900 )
 0.9 9
1
= ln P ln (0.9P+900)+C
9

1 1
When t=0 , P=10 , 000 , so 0= ln 10 , 000 ln (9900)+C . Thus, C=ln 10 , 000+ ln 9900 , so our
9 9
equation becomes
1 1 10,000 1 9900
t = ln 10,000 ln P+ln 9900 ln (0.9P+900)=ln + ln
9 9 P 9 0.9P+900
10,000 1 1100 10,000 1 11,000
= ln + ln =ln + ln
P 9 0.1P+100 P 9 P+1000

2
64. If we subtract and add 2x , we get
2 2
4 4 2 2 2 2 2
(
x +1 = x +2x +1 2x = x +1  2x = x +1  ) ( ) ( 2 x)
2

= ( x2+1)  ( x2+1) +
2 x = x  2 x+1 x + 2 x+1
2x ( 2
)( 2
)
1 Ax+B Cx+D
So we can decompose 4 = 2 + 2 
x +1 x + 2 x+1 x  2 x+1
1= ( Ax+B ) x  ( 2
) ( 2
)
2 x+1 + (Cx+D ) x + 2 x+1 . Setting the constant terms equal gives B+D=1 , then

18
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

3
from the coefficients of x we get A+C=0 . Now from the coefficients of x we get A+C+ ( B D ) 2 =0
1 1 2
 (1 D) D 2 =0 D=  B= , and finally, from the coefficients of x we get
2 2
1 2 2
2 (C A) +B+D=0 C A=  C= and A= .
2 4 4
So we rewrite the integrand, splitting the terms into forms which we know how to integrate:
2 1 2 1
1 x+  x+
= 4 2 4 2 1 2x+2 2 2x 2 2
4
x +1 2
+ 2
= 2
 2
x + 2 x+1 x 2 x+1 4 2 x + 2 x+1 x 2 x+1
2 2x+ 2 2x 2 1 1 1
 + +
= 8 2 2 4 1 2 1 1 2 1
x + 2 x+1 x 2 x+1 x+ + x +
2 2 2 2
2
dx 2 x + 2 x+1 2  1  1
Now we integrate:  = ln + tan ( 2 x+1 ) +tan ( 2 x 1 ) +C .
4 8 2 4
x +1 x 2 x+1

65. (a) In Maple, we define f (x) , and then use convert(f,parfrac,x) ; to obtain
24,110/4879 668/323 9438/80,155 (22,098x+48,935)/260,015
f (x)=   + .
5x+2 2x+1 3x 7 2
x +x+5
In Mathematica, we use the command Apart, and in Derive, we use Expand.
(b)
24,110 1 668 1 9438 1
 f (x)dx = 4879  5 ln 5x+2  323  2 ln 2x+1  80,155  3 ln 3x 7
1
22,098 x+ +37,886
1 2
+
260,015
 dx+C
1 2 19
x+ +
2 4
24,110 1 668 1 9438 1
=  ln 5x+2   ln 2x+1   ln 3x 7
4879 5 323 2 80,155 3
1 1 4  1 1
+
260,015
22,098
2
2
(
ln x +x+5 +37,886 ) 19
tan
1
19/4
x+
2
+C

4822 334 3146 11,049

=
4879
ln 5x+2 
323
ln 2x+1 
80,155
ln 3x 7 +
260,015
2
ln x +x+5 ( )
75,772  1 1
+ tan ( 2x+1 ) +C
260,015 19 19

19
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

Using a CAS, we get

4822ln (5x+2) 334ln (2x+1) 3146ln (3x 7)

 
4879 323 80,155

+
11,049ln x +x+5
+
( 2
3988 19
tan
)
 1 19
(2x+1)
260,015 260,015 19

The main difference in this answer is that the absolute value signs and the constant of integration have
been omitted. Also, the fractions have been reduced and the denominators rationalized.

66. (a) In Maple, we define f (x) , and then use convert(f,parfrac ,x) ; to get
5828/1815 59,096/19,965 2(2843x+816)/3993 (313x 251)/363
f (x)=  + + .
2 5x 2 2 2
( 5x 2 ) 2x +1 2
2x +1 ( )
In Mathematica, we use the command Apart, and in Derive, we use Expand.
(b) As we saw in Exercise 65, computer algebra systems omit the absolute value signs in
 ( 1/y) dy=ln y . So we use the CAS to integrate the expression in part (a) and add the necessary
absolute value signs and constant of integration to get

 f (x)dx =  5828

59,096ln 5x 2
+
2843ln 2x +1 ( 2
)
9075(5x 2) 99,825 7986
503  1 1 1004x+626
+ 2 tan ( 2 x )  +C
15,972 2904 2
2x +1

(c) From the graph, we see that f goes from negative to positive at x  0.78 , then back to negative at
x 0.8 , and finally back to positive at x=1 . Also, f (x)= . So we see (by the First Derivative
x 0.4
Test) that  f (x)dx has minima at x  0.78 and x=1 , and a maximum at x 0.80 , and that  f (x)dx is
unbounded as x 0.4 . Note also that just to the right of x=0.4 , f has large values, so  f (x)dx
increases rapidly, but slows down as f drops toward 0 .  f (x)dx decreases from about 0.8 to 1 , then
increases slowly since f stays small and positive.

67. There are only finitely many values of x where Q(x)=0 (assuming that Q is not the zero
20
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.4 Integration of Rational Functions by Partial Fractions

polynomial). At all other values of x , F(x) / Q(x) =G(x) / Q(x) , so F(x)=G(x) . In other words, the
values of F and G agree at all except perhaps finitely many values of x . By continuity of F and G ,
the polynomials F and G must agree at those values of x too.
More explicitly: if a is a value of x such that Q(a)=0 , then Q(x) 0 for all x sufficiently close to a .
Thus,
F(a)=lim F(x)[by continuity of F] = lim G(x) [whenever Q(x) 0]
x a x a
= G(a) [by continuity of G]

2 f (x)
68. Let f (x)=ax +bx+c . We calculate the partial fraction decomposition of . Since f (0)=1 ,
2 3
x (x+1)
2
f (x) A B C
ax +bx+1 D E
we must have c=1 , so =
+ + + = + . Now in order for the
2 3 2 3 x 2 x+1 2 3
x (x+1) x (x+1) x (x+1) (x+1)
integral not to contain any logarithms (that is, in order for it to be a rational function), we must have
2 3 2 2
A=C=0 , so ax +bx+1=B(x+1) +Dx (x+1)+Ex . Equating constant terms gives B=1 , then equating
/
coefficients of x gives 3B=b b=3 . This is the quantity we are looking for, since f (0)=b.

21
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

sin x+sec x sin x sec x

1.  dx= + dx= (cos x+csc x)dx=sin x+ln csc x cot x +C
tan x tan x tan x

2.
sin 
= (sec   1)tan  d = tan  sec  d  
2 2

3
tan  d cos 
d

dv u=tan  , v=cos  ,
= udu+ 2
v du=sec  d dv= sin  d
1 2 1 2
= u +ln v +C= tan  +ln cos  +C
2 2

3.
2  1  1

 2t
2
dt =  2 ( u+3)
2
du=
2 6
u
+ 2
du= 2ln u 
6
u
 1

 3
0 ( t 3)  3 u  3 u
= ( 2ln 1+6 )  ( 2ln 3+2 ) =4 2ln 3 or 4 ln 9

2
xdx 1 du 1  u 1  x
 = 
2 1 1
4. Let u=x . Then du=2xdx = sin +C= sin +C .
4 2 2 2 3 2 3
3 x 3 u


arctany  /4
dy e u  /4  /4   /4

u
5. Let u=arctany . Then du=  dy= e du= e =e  e .
2 2   /4
1+y  1 1+y   /4

6.

 x csc x cot xdx

u=x, dv=csc x cot xdx,
du=dx v= csc x = x csc x  ( csc x)dx

= xcsc x+ln csc x cot x +C

7.
3 4
u=ln r, dv=r dr 3 3 3
 1 5 1 4 243 1 5
4
r ln r dr
du=
dr 1 5 =
5
r ln r   5
r dr=
5
ln 3 0
25
r
1 r v= r 1 1 1
5
243 243 1 243 242
= ln 3  = ln 3
5 25 25 5 25
1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

x 1 x 1 A B 1
8. = = +  x 1=A(x+1)+B(x 5) . Setting x= 1 gives  2= 6B , so B=
2 (x 5)(x+1) x 5 x+1 3
x  4x 5
2
. Setting x=5 gives 4=6A , so A= . Now
3
4 4
 2/3 1/3 2 1 4

2
x 1
dx =  +
x 5 x+1
dx=
3
ln x 5 + ln x+1
3 0
0 x  4x 5 0

2 1 2 1 1
= ln 1+ ln 5 ln 5 ln 1= ln 5
3 3 3 3 3

9.
x 1 ( x 2 ) +1 u 1
 2
dx = 2
dx= 2
+ 2
du [ u=x 2 , du=dx ]
x  4x+5 ( x 2 ) +1 u +1 u +1
1  1 1  1
=
2
2
( 2
2
)
ln u +1 +tan u+C= ln x  4x+5 +tan (x 2)+C ( )
10.
1
x du
 4 2
dx = 2 2
[ u=x , du=2xdx ] =
1

du
x +x +1 2 2 1 2 3
u +u+1 u+ +
2 4
3
dv
1 2 1 3 3 3 4 dv
=
2
 3 2
[ u+
2
=
2
v , du=
2
dv ] =  
4 3 2
4
(
v +1 ) v +1

1  1 1  1 2 2 1
= tan v+C= tan x+ +C
3 3 3 2

11.
 = cos  sin  sin  d = 
3 5 5 2 5 2
sin  cos  d cos  (1 cos  )( sin  )d
u=cos  ,

5 2
= u (1 u )du
du= sin  d
1 8 1 6 1 1
= (u  u )du=
7 5 8 6
u u +C= cos   cos  +C
8 6 8 6

Another solution:
2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

 = sin  d = sin 
3 5 3 2 2 3 2 2
sin  cos  d (cos  ) cos  (1 sin  ) cos  d
u=sin  ,
= u (1 u ) du = u (1 2u +u )du
3 22 3 2 4
du=cos  d
1 4 1 6 1 8 1 1 1
= (u  2u +u )du=
3 5 7 4 6 8
u u + u +C= sin   sin  + sin  +C
4 3 8 4 3 8

12. Let u=cos x . Then du= sin xdx  sin xcos (cos x)dx=  cos udu= sin u+C= sin (cos x)+C .

  2 1/2
13. Let x=sin  , where     . Then dx=cos  d and (1 x ) =cos  , so
2 2
dx cos  d
 = = sec  d
2
2 3/2 3
(1 x ) (cos  )
x
=tan  +C= +C
2
1 x

14. Let u=ln x . Then du=dx/x

1+u v
 du=
1+ln x 2
 dx = u 2
2vdv [put v= 1+u ,u=v  1,du=2vdv]
xln x v 1
1 v 1 1+ln x  1
= 2 1+ 2
dv=2v+ln
v+1
+C=2 1+ln x +ln +C
v 1 1+ln x +1

2
15. Let u=1 x  du= 2xdx . Then

1/2

 0
x

1 x
2
dx=
1
2
3/4

1
1
u
du=
1
2

3/4
1
u
 1/2
du=
1
2
2u
1/2 1
=
3/4
u
1
3/4
=1
2
3

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

16.
2 /2


 /4 2
sin 
x
2
= cos  d [ x=sin  , }{dx}=cos  d ]
dx 0 cos 
2
0 1 x
 /4 1 1 1  /4 1  1  1
= (1 cos 2 )d =   sin 2 =   (0 0) = 
0 2 2 2 0 2 4 2 8 4

17.
2
u=x, dv=sin xdx

2
xsin xdx 1 1 1
du=dx v= sin xdx=
2
(1 cos 2x)dx= x sin xcos x
2 2 2
1 2 1 1 1
= x x sin x cos x  x sin x cos x dx
2 2 2 2
1 2 1 1 2 1 2 1 2 1 1 2
= x x sin x cos x x + sin x+C= x  x sin x cos x+ sin x+C
2 2 4 4 4 2 4

1 2
Note:  sin x cos xdx= sds=
2
s +C where s=sin x,ds=cos xdx .
1 1 1
A slightly different method is to write  x sin xdx= x (1 cos 2x)dx=  xdx  xcos 2xdx . If
2
2 2 2
we evaluate the second integral by parts, we arrive at the equivalent answer
1 2 1 1
x xsin 2x cos 2x+C .
4 4 8

2t 2t
18. Let u=e , du=2e dt . Then
1 2t 1
2t (2e )dt du
e 2 2 1  1 1  1 2t
 4t
dt =
2t 2
=
2
= tan u+C= tan (e )+C .
2 2
1+e 1+(e ) 1+u
x x x
19. Let u=e . Then  e dx= e e dx= e du=e +C=e +C .
x x+e e x u u e

[3]x
 dx= e  3u du . Now use parts: let w=u , dv=e du
3 3 u 2 2 u
20. Let u= x . Then x=u  e dw=2udu ,
v=e 
u u 2
( 2 u u
)
3 e u du=3 u e  2 ue du . Now use parts again with W =u , dV =e du to get
u

4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

3

u 2
e 3u du=e
u
( 3u2 6u+6) +C=3e x ( x2/3 2 3 x +2) +C .
3  2t
21. Integrate by parts three times, first with u=t , dv=e dt :
1 3  2t 1 2  2t 1 3  2t 3 2  2t 1  2t
 t e dt =  2 t e + 2  3t e dt= 2 t e  4 t e + 2  3te dt
3  2t

 2t 1 3 3 2 3  2t 3  2t  2t 1 3 3 2 3 3
=  e t + t  te +  e dt= e t + t + t+ +C
2 4 4 4 2 4 4 8
1  2t
= 
8
e
3 2
4t +6t +6t+3 +C ( )

22. Integrate by parts: u=sin x , dv=xdx

 1
du= 1 ( / 1 x
2
) dx , v= 12 x 2
, so
2 2
1 2 1 1 x dx 1 2  1 1 sin  cos  d
2
x sin x
2
 = x sin x
2 2
 cos 
where x=sin  for
 1 2
 x sin xdx = 1 x
 
   
2 2
1 2 1 1 1 2 1 1
=
2
x sin x
4
 ( 1 cos 2 ) d = x sin x (  sin  cos  )+C
2 4
1 2 1 1  1 1  1
=
2
x sin x
4
sin x x 1 x +C=
2
4
2 2
2x  1 sin x+x 1 x +C ( )
2
23. Let u=1+ x . Then x=(u 1) , dx=2(u 1)du
1 10 1 9 2
 0 ( 1+
1 8
x ) dx =  u  2(u 1)du=2
2 8
1
(1 u9 u8) du=
2
5
u  2 u
9 1
1024 1024 1 2 4097
=   + =
5 9 5 9 45

2x
24. Let u=ln x  1 , dv=dx du= ( 2
) 2
, v=x . Then
x 1
2
2x 2
 ( 2
)
ln x  1 dx =x ln x  1  ( 2
)  2
(
dx=x ln x  1 
2
)  2+
(x 1)(x+1)
dx
x 1
1 1
=x ln x  1  ( 2
)  2+ 
x 1 x+1
dx

5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

( 2
)
=x ln x  1  2x ln x 1 +ln x+1 +C

2
3x  2 6x+22 A B
25. =3+ =3+ +  6x+22=A(x+2)+B(x 4) . Setting x=4 gives 46=6A , so
2 (x 4)(x+2) x 4 x+2
x  2x 8
23 5
A= . Setting x= 2 gives 10= 6B , so B= . Now
3 3
2
3x  2 23/3 5/3 23 5
 2
dx= 3+ 
x 4 x+2
dx=3x+
3
ln x 4 
3
ln x+2 +C .
x  2x 8

2 3
3x  2 du u=x  2x 8,
26.  dx=
3
=ln u +C=ln x  2x 8 +C
3
x  2x 8
u ( 2
)
du= 3x  2 dx

1 2 1
 cot xln (sin x)dx= udu=
2
27. Let u=ln (sin x) . Then du=cot xdx u +C= ln (sin x) +C .
2 2

28.
2 2
sin at dt = sin u udu [ u= at , u =at , 2udu=adt ] =  usin udu
2
 a a
2 2 2
=  ucos u+sin u +C [ integration by parts] = at cos at + sin at +C
a a a
t 2
= 2 cos at + sin at +C
a a

29.
5 5
3w 1 7
 dw =
5
3 dw= 3w 7ln w+2
0 w+2 0 w+2 0

2
=15 7ln 7+7ln 2=15+7(ln 2 ln 7)=15+7ln
7

2
30. x  4x<0 on 0,4 , so
1 3 2 0 1 3 2

2 2
 2
x  4x dx =
0
 2
(x  4x)dx+
2
(0 4x x2) dx=
2
3
x  2x
 2
+ 2x 
2
3
x
0
8 8
=0   8 + 8  0=16
3 3
6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

31. As in Example 5,
1+x 1+x 1+x dx xdx

1+x
dx =   dx= dx= +
1 x 1 x 1+x 2 2 2
1 x 1 x 1 x
= sin  1 x 2
1 x +C
Another method: Substitute u= (1+x)/(1 x) .

32.
u udu 4
dx = u= 2x 1 ,2x+3=u +4,u =2x 1,udu=dx = 
2x 1 2 2
 2x+3
2
1 2 du
u +4 u +4
1  1 1  1 1
=u 4 tan u +C= 2x 1  2tan 2x 1 +C
2 2 2

2 2 2  
33. 3 2x x = (x +2x+1)+4=4 (x+1) . Let x+1=2sin  , where     . Then dx=2cos  d
2 2
and

 3 2x x dx = 4 (x+1) dx=

2 2 2
4 4sin  2cos  d

=4 cos  d =2 (1+cos 2 )d

2

=2 +sin 2 +C=2 +2sin  cos  +C

2
 1 x+1 x+1 3 2x x
=2sin +2  +C
2 2 2
 1 x+1 x+1 2
=2sin + 3 2x x +C
2 2

34.
 /2  /2  /2
1+4cot x (1+4cos x/sin x) sin x sin x+4cos x
 4 cot x dx =  
(4 cos x/sin x) sin x
dx= 
4sin x cos x
dx
 /4  /4  /4

7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

4
1
 u=4sin x cos x, 4
= du = ln u
u du=(4cos x+sin x)dx 3/ 2
3/ 2
3 4 4
=ln 4 ln =ln =ln 2
2 3/ 2 3

8
35. Because f (x)=x sin x is the product of an even function and an odd function, it is odd. Therefore,

1 8
x sin xdx=0 [ by (5.5. )(b)].
 1

1
36. sin 4xcos 3x= (sin x+sin 7x) by Formula .2.2(a), so
2
1 1 1 1 1
 sin 4xcos 3xdx=  (sin x+sin 7x)dx=
2 2
 cos x
7
cos 7x +C=
2
cos x
14
cos 7x+C .

37.
 /4  /4  /41 1 1  /4

2 2
= d =
2
cos  tan  d sin  ( 1 cos 2 ) d =   sin 2
0 0 0 2 2 4 0
 1  1
=   ( 0 0 ) = 
8 4 8 4

38.
 /4
 /4
= ( tan 2 ) 2sec 2  sec  tan 

5 3 0
tan  sec  d 2 u=sec 
d =
0 2 2 2
(u  1) u du
1 du=sec  tan  d
2 1 7 2 5 1 3 2
=
6 4 2
(u  2u +u )du=
u u+ u
1 7 5 3 1
8 8 2 1 2 1 22 8 2
= 2 2+ 2   + = 2 = (11 2  4)
7 5 3 7 5 3 105 105 105

2
39. Let u=1 x . Then du= 2xdx
xdx
 = 
1

du
= 
vdv 2
[v= u ,u=v ,du=2vdv]
2 2 2 u+ u 2
1 x + 1 x v +v

=  
dv
v+1
= ln v+1 +C= ln ( 2
1 x +1 +C )
8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

2 2 2
40. 4y  4y 3=(2y 1)  2 , so let u=2y 1 du=2dy . Thus,
dy dy 1 du
 =  =
2

2 2 2 2 2
4y  4y 3 (2y 1)  2 u 2
1 2 2
= ln u+ u  2
2
1 2
= ln 2y 1+ 4y  4y 3 +C
2

41. Let u= , dv=tan 

2
(
d = sec   1 d 
2
) du=d and v=tan    . So
1

2 2

2
 tan  d =  (tan    ) (tan    )d = tan     ln sec  +  +C
2
1 2
=  tan     ln sec  +C
2

 1
42. Integrate by parts with u=tan x , dv=x dx
2
(
du=dx/ 1+x
2
) , v= 13 x3 :
3
1 3 1 x dx 1 3 1 1 x

2  1
x tan xdx = 3 x tan x  3 2
=
3
x tan x
3
 x 2
dx
1+x x +1
1 3 1 1 2 1
=
3
x tan x
6
2
x + ln x +1 +C
6
( )
x x
43. Let u=1+e , so that du=e dx . Then
2 3/2 2

1/2 x 3/2
 e 1+e dx
x x = u du= u +C= (1+e ) +C .
3 3

x 2 x x
Or: Let u= 1+e , so that u =1+e and 2udu=e dx . Then
2 3 2
 e 1+e dx = u 2udu= 2u du= 3 u +C= 3 (1+e ) +C .
x x 2 x 3/2

x 2 x x 2 2u
44. Let u= 1+e . Then u =1+e , 2udu=e dx=(u  1)dx , and dx= 2
du , so
u 1

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

2
2u 2u 2 1 1
 1+e dx = u du= du= du= 
x 2+ 2+ du
2 2 2 u 1 u+1
u 1 u 1 u 1

=2u+ln u 1  ln u+1 +C=2 1+e +ln

x
( x
1+e  1  ln ) ( x
1+e +1 +C )
3
5  x 1  t  t
I= x e 
3 2
45. Let t=x . Then dt=3x dx dx= te dt . Now integrate by parts with u=t , dv=e dt :
3
3
1 t 1 t 1 t 1 t 1 
I=
3
te +  e dt=
3 3
te 
3
e +C=
3
e
x
( x3+1) +C .
x
46. Let u=e . Then x=ln u , dx=du/u
x
1+e (1+u)du (u+1)du 2 1
 x
dx =  =  =   du
(1 u)u (u 1)u u 1 u
1 e
= ln u  2ln u 1 +C=ln ex 2ln ex 1 +C=x 2ln ex 1 +C

47.
x+a 1 2xdx 1 1  x
 2 2
dx = 
2 2 2
+a dx
=
2 2 2
ln x
2 2
+a +a
a
tan ( ) 1
a
+C
x +a x +a x +a
2 2  1
=ln x +a +tan ( x/a ) +C
2
48. Let u=x . Then du=2xdx
1
du 2 2 2
xdx 2 1 u a 1 x a
 4 4 = = ln +C= ln +C .
2 2 2 2 2 2 2
x a
2
u a 4a ( ) u+a 4a x +a

2 1
49. Let u= 4x+1  u =4x+1 2udu=4dx dx= udu . So
2
1
udu
1 2 du 1 u 1
 dx =  =2 =2 ln +C
x 4x+1 1 2 2 2 u+1
4
(
u 1 u ) u 1

4x+1  1
= ln +C
4x+1 +1

10
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

1
udu
dx 2 du
50. As in Exercise 49, let u= 4x+1 . Then  2
= 2
=8 . Now
2
x 4x+1 1 2
4
(
u 1 ) u ( u  1)
2

1 1 A B C D
= = + + + 
2 2 2 u+1 2 u 1 2
( u  1)
2
( u+1 ) ( u 1 ) (u+1) (u 1)
2 2 2 2 1 1
1=A(u+1)(u 1) +B(u 1) +C(u 1)(u+1) +D(u+1) . u=1 D= , u= 1 B= . Equating
4 4
3 1 1
coefficients of u gives A+C=0 , and equating coefficients of 1 gives 1=A+B C+D 1=A+  C+
4 4
1 1 1
 =A C . So A= and C= . Therefore,
2 4 4
dx 1/4 1/4  1/4 1/4
 2 = 8 u+1
+
2
+
u 1
+
2
du
x 4x+1 (u+1) (u 1)
2  2 2  2
=  +2(u+1)  +2(u 1) du
u+1 u 1
2 2
= 2ln u+1   2ln u 1  +C
u+1 u 1
2 2
= 2ln ( 4x+1 +1 )   2ln 4x+1  1  +C
4x+1 +1 4x+1  1

1 1 2 2
51. Let 2x=tan   x= tan  , dx= sec  d , 4x +1 =sec  , so
2 2

1 2
sec  d
dx 2 sec 
 =
1
=
tan 
d = csc  d
2
x 4x +1 tan  sec 
2
= ln csc  +cot  +C
2 2
4x +1 1 4x +1 1
= ln + +C orln  +C
2x 2x 2x 2x

11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

2
52. Let u=x . Then du=2xdx
dx xdx 1 du 1 1 u 1 1
 =  =  =   du= ln u  (2
ln u +1 +C )
(4
x x +1 ) 2 4
x x +1
2 2
u u +1 (
2 ) ( ) u 2
u +1
2 4
4
1 1 1 1 x
=
2
2
ln x 
4
4
( )
ln x +1 +C=
4
4 4
(
ln x  ln x +1 ) ( ) ( ) +C= ln
4 4
+C
x +1
3
x dx
Or: Write I=
4
and let u=x .
x
4
( x4+1)
2 dv=sinh (mx)dx
1 2 2 u=x ,
53.  x sinh (mx)dx= x cosh (mx)  xcosh (mx)dx
2
1
m m du=2xdx v= m cosh (mx)

U =x, dV =cosh (mx)dx

1 2 2 1 1
= x cosh (mx)
m m m
xsinh (mx)
m
 sinh (mx)dx 1
V = sinh (mx)
dU =dx m
1 2 2 2
= x cosh (mx) xsinh (mx)+ cosh (mx)+C
m 2 3
m m

54.

 ( x+sin x )
2
dx = ( x2+2xsin x+sin 2x) dx= 13 x3+2(sin x xcos x)+ 12 (x sin xcos x)+C
1 3 1 1
= x + x+2sin x sin xcos x 2xcos x+C
3 2 2

2
55. Let u= x+1 . Then x=u  1
2udu  1 3
=  2 =
dx
 +
u+1 u+3
du
x+4+4 x+1 u +3+4u
= 3ln u+3  ln u+1 +C=3ln ( x+1 +3)  ln ( x+1 +1 ) +C

2
56. Let t= x  1 . Then dt= x
xln x 1
( /
x  1 dx , x  1=t , x= t +1 , so
2
) 2 2 2

I=
2
(
dx= ln t +1 dt=  ln t +1 dt . Now use parts with u=ln t +1 , dv=dt :
2
2
2 2
) ( )
x 1

12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

2
1 t 1 1
2
I = tln t +1 
2
( )  2
dt=
2
(
2
tln t +1  )  1 2
dt
t +1 t +1
1  1  1
=
2
2
( )
2
tln t +1  t+tan t+C= x  1 ln x
2
x  1 +tan
2
x  1 +C

Another method: First integrate by parts with u=ln x , dv= x ( / 2

x 1 ) dx and then use substitution
( x=sec  or u= x  1
2
).
3 3
57. Let u= x+c . Then x=u  c

 x
3
x+c dx =  ( u3 c ) u 3u2 du=3 ( u6 cu3) du= 37 u7 3 4
4
cu +C
3 7/3 3 4/3
= (x+c)  c(x+c) +C
7 4

2 1 3
58. Integrate by parts with u=ln (1+x) , dv=x dx du=dx/(1+x) , v= x :
3
3
1 3 x dx 1 3 1 1

2
x ln (1+x)  
2
x ln (1+x)dx = = x ln (1+x) x  x+1 dx
3 3(1+x) 3 3 x+1
1 3 1 3 1 2 1 1
= x ln (1+x) x+ x x+ ln (1+x)+C
3 9 6 3 3

x
59. Let u=e . Then x=ln u , dx=du/u

 3xdx x =  du/u du 1/2 1 1/2

3
=  2
=
u 1
 
2 u+1
du
e  e u u (u 1)u (u+1) u
x
1 1 u 1 1  x e 1
= + ln +C=e + ln +C
u 2 u+1 2 x
e +1

3 3 2
60. Let u= x . Then x=u , dx=3u du
2
dx 3u du 3 2udu 3 3
 3
= 3
=
2
 2
=
2
ln u
2
+1 +C=
2
(
ln x
2/3
+1 +C . ) ( )
x+ x u +u u +1

5 4
61. Let u=x . Then du=5x dx

13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

1
4 du
x dx 5 1 1  1 1  1 5
 =
1 1
=  tan u +C= tan x +C .
10 2 5 4 4 20 4
x +16 u +16

62. Let u=x+1 . Then du=dx

3 3
 x
10
dx =  (u 1)
10
du= ( u 7 3u 8+3u 9 u 10) du
(x+1) u
1 6 3 7 3 8 1 9
=  u + u  u + u +C
6 7 8 9
 9 1 3 3 2 3 1
= (x+1)  (x+1) + (x+1)  (x+1)+ +C
6 7 8 9

1
63. Let y= x so that dy= dx dx=2 x dy=2ydy . Then
2 x

2 y
dx = ye (2ydy)= 2y e dy
x y 2 y u=2y , dv=e dy,
 xe y
du=4ydy v=e
y

2 y y U=4y, dV=e dy,
=2y e  4ye dy y
dU=4dy V=e
=2y e 
2 y
( 4yey  y
) 2 y
4e dy =2y e  4ye +4e +C
y y

2 y x
=2(y  2y+2)e +C=2 ( x 2 x +2 ) e +C

64. Let u=tan x . Then

 /3  /3 3
ln (tan x) ln (tan x) ln u
 dx
sin xcos x =  sec xdx= 
2
du
 /4  /4 tan x 1 u
1 2 3 1 2 1 2
=
2
(ln u)
1
=
2
( ln 3 ) = (ln 3)
8

65.
dx 1 x+1  x
 =  dx= ( x+1  x ) dx
x+1 + x x+1 + x x+1  x
2 3/2 3/2
= (x+1)  x +C
3
14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

3 2
u +1 u +1 2 1 1 1
66.  du= 1+ du=u+   du=u+2ln u 1  ln u + +C . Thus,
3 2 2 u 1 u 2 u
u u (u 1)u u
3


3 1 3 1 1
u +1 
du = u+2ln (u 1) ln u+
u
= 3+2ln 2 ln 3+
2 3
2+2ln 1 ln 2+
2
3 2
2 u u
1 5 8
= 1+3ln 2 ln 3 = +ln
6 6 3

67. Let u= t . Then du=dt/ ( 2 t ) 

3 3
arctan t  1  1 1 3
 dt =  tan u (2du)=2 utan u
2
ln (1+u ) [ Example 5 in Section 8.1]
1 t 1 2 1

 1 1  1 1
=2 3 tan 3
ln 4  tan 1 ln 2
2 2
  1 2 1
=2 3  ln 2   ln 2 = 3     ln 2
3 4 2 3 2

x
68. Let u=e . Then x=ln u , dx=du/u
dx du/u du 2/3 1/3
 x  x =  =  =   du
1+2u 1/u 2 2u 1 u+1
1+2e  e 2u +u 1
1 1 1
=
3
ln 2u 1 
3
x
ln u+1 +C= ln 2e  1 / e +1
3
x
( )( ) +C

x
69. Let u=e . Then x=ln u , dx=du/u
2x 2
e
 x
dx = 
u du
=
u
du= 1
1
du
1+e 1+u u 1+u 1+u
= u ln 1+u +C=ex ln 1+ex +C ( )
2
70. Use parts with u=ln (x+1) , dv=dx/x :

 ln (x+1)
2
dx =  1 ln (x+1)+ dx
=
1
ln (x+1)+
1

1
dx
x x x(x+1) x x x+1
1 1
=  ln (x+1)+ln x  ln (x+1)+C= 1+ ln (x+1)+ln x +C
x x
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

x x Ax+B Cx+D
71. = = + 
4
x +4x +3 ( x +3) ( x +1) x +3 x +1
2 2 2 2 2

x = ( Ax+B ) ( x2+1 ) + (Cx+D ) ( x2+3) = ( Ax3+Bx2+Ax+B) + ( Cx3+Dx2+3Cx+3D)

= ( A+C ) x3+ ( B+D ) x2+ ( A+3C ) x+ ( B+3D ) 
1 1
A+C=0 , B+D=0 , A+3C=1 , B+3D=0 A= , C= , B=0 , D=0.Thus,
2 2
1 1


x  x x
 4 2
dx = 2 2
x +4x +3 2
+ 2
dx
x +3 x +1
2
1 1 1 x +1
= 
4
2 2
(
ln x +3 + ln x +1 +C or
4 4
ln ) ( ) 2
+C
x +3

6 6 5
72. Let u= t . Then t=u , dt=6u du
3 5 8
t dt u  6u du u 1
 3 =  2
=6 2
du=6
6
u  u +u  1+
4 2
2
du
1+ t 1+u u +1 u +1
1 7 1 5 1 3  1
= 6 u u + u  u+tan u +C
7 5 3
1 7/6 1 5/6 1 1/2 1/6  1 1/6
= 6 t  t + t  t +tan t +C
7 5 3
1 A Bx+C
73. = +  1=A x
2
+4 + ( Bx+C ) ( x 2 ) = ( A+B ) x
2
( )
+ (C 2B ) x+ ( 4A 2C ) . So
2
( x 2 ) x +4 x 2( 2
x +4 )
1 1 1
0=A+B=C 2B , 1=4A 2C . Setting x=2 gives A=  B= and C= . So
8 8 4
1 1 1
1  x
 dx = 8 8 4 1 dx 1 2xdx 1 dx
2
( x 2 ) x +4 (  x 2
+ ) 2
dx= 
8 x 2 16
  2

4
 2
x +4 x +4 x +4
1 1 1  1
=
8
ln x 2 
16
2
ln x +4 
8
tan (x/2)+C ( )
x
74. Let u=e . Then x=ln u , dx=du/u
x x
dx e dx u du du 1 u 1 1 e 1
 x  x
= 2x
= 2 u
= 2
= ln
2 u+1
+C= ln
2 x
+C .
e e e  1 u 1 u 1 e +1
16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

75.
1 1
 sin xsin 2xsin 3xdx = sin x cos (2x 3x) cos (2x+3x) dx=  ( sin xcos x sin xcos 5x ) dx
2 2
1 1 1
=  sin 2xdx  sin (x+5x)+sin (x 5x) dx
4 2 2
1 1 1 1 1
=
8
cos 2x
4
 (sin 6x sin 4x)dx=
8
cos 2x+
24
cos 6x
16
cos 4x+C

76.

 ( x2 bx) sin 2xdx = 12 ( x2 bx) cos 2x+ 12  (2x b)cos 2xdx

1 2 1 1
=
2
( )
x  bx cos 2x+
2 2
(2x b)sin 2x  sin 2xdx

1 2 1 1
=
2
( )
x  bx cos 2x+ (2x b)sin 2x+ cos 2x+C
4 4

3/2 2 3 3 1/2 2
77. Let u=x so that u =x and du= x dx x dx= du . Then
2 3
2
x 3 2  1 2  1 3/2
 3
dx=
2
du=
3
tan u+C= tan (x )+C .
3
1+x 1+u

78.
sec xcos 2x sec xcos 2x 2cos x 2cos 2x
 sin x+sec x dx = sin x+sec x  2cos x dx= 2sin xcos x+2 dx
2cos 2x 1
= dx=
u=sin 2x+2,
du
sin 2x+2 u du=2cos 2xdx
=ln u +C=ln sin 2x+2 +C=ln (sin 2x+2)+C

2 1 3
79. Let u=x , dv=sin xcos xdx du=dx , v= sin x . Then
3
1 1 1 1
xsin x  
3 3 3 2

2
xsin xcos xdx = 3 3
sin xdx= xsin x
3 3
(1 cos x)sin xdx

1 1
xsin x+  (1 y )dy
3 2 y=cos x,
=
3 3 dy= sin xdx
17
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.5 Strategy for Integration

1 3 1 1 3 1 3 1 1 3
= xsin x+ y y +C= xsin x+ cos x cos x+C
3 3 9 3 3 9

80.
sin xcos x sin xcos x sin xcos x
 4 4
dx = 2 2 2 2
dx=
2 2 2 2
dx
sin x+cos x (sin x) +(cos x) (sin x) +(1 sin x)
1 1 2
= du u=sin x,
2 2 2 du=2sin xcos xdx
u +(1 u)
1 1
= 2
du du= 2
4u  4u+2 (4u  4u+1)+1
1 1 1
=  y=2u 1,
du= dy
2 2 2 dy=2du
(2u 1) +1 y +1
1  1 1  1 1  1 2
= tan y+C= tan (2u 1)+C= tan (2sin x 1)+C
2 2 2

Another solution:

4 2
sin xcos x (sin xcos x)/cos x tan xsec x
 4 4
dx = 4 4 4
dx= 4
dx
sin x+cos x (sin x+cos x)/cos x tan x+1
2
1 1
=
u=tan x,
du 2
2 2 du=2tan x sec xdx
u +1
1  1 1  1 2
= tan u+C= tan (tan x)+C
2 2

2
x
81. The function y=2xe does have an elementary antiderivative, so we’ll use this fact to help
evaluate the integral.
2 2 2 2 2
 (2x +1)e dx = 2x e dx+ e dx= x(2xe )dx+ e dx
2 x 2 x x x x

2
2 2 2 x 2
 e dx+ e dx
x x x u=x, dv=2xe dx, x
=xe  2 =xe +C
x
du=dx v=e

18
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2
1. We could make the substitution u = 2 x to obtain the radical and then use Formula 33 7 u
7
with a= 7 . Alternatively, we will factor 2 out of the radical and use a= .
2
7 2
2  x
7 2x 2 1 7 2  x
2
1
 2
dx = 2
dx= 2 
x 2
 x  sin +C
x x 7
2
1 2  1 2
=  7 2x  2 sin x +C
x 7

2.
x 2
dx =3
3x
 3+( 2)x
dx=3
2
( 2x 2 3) 3+( 2)x +C
3 2x 3( 2)
1
= ( 2x 6) 3 2x +C= (x+3) 3 2x +C
2

3. Let u = x du= dx , so
1 1 1 1
sec ( x)dx =   sec u du= 
3

3
sec u tan u + ln sec u +tan u +C
2 2
1 1
= sec  xtan  x+ ln sec  x+tan  x +C
2 2

2
e 2 2 3 2
4.  e sin 3
2
d = (2sin 3  3cos 3 )+C= e sin 3  e cos 3 +C
2 2 13 13
2 +3

1
1 2 2
 1 2x  1  1 x 1 x 1 1   
5.  2xcos xdx=2 cos x =2  0 0     0 =2 =
0
4 4 0 4 4 2 8 4

6.

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems


3

 2 x
2
1
2
4x  7
dx
=
1 1
2
du [u=2x, du=2dx]
1 2 2
u u 7
4 2
6
6 2
u 7
=2
2 2
u u  7 =2
4 du 7u 4

29 3 29 3
=2  = 
42 28 21 14

7. By Formula 99 with a= 3 and b=4 ,

 3x  3x
 3x e e
 e cos 4xdx=
2 2
(  3cos 4x+4sin 4x ) +C=
25
(  3cos 4x+4sin 4x ) +C .
(  3) +4
8. Let u =x/2 , so dx=2du , and we use Formula 72:
 csc (x/2)dx = 2 csc u du= csc u cot u +ln csc u  cot u +C
3 3

=  csc (x/2)cot (x/2)+ln csc (x/2) cot (x/2) +C

9. Let u =2x and a=3 . Then du=2dx and

1
du 2 2
dx a +u
 =
2
2
=2
du
= 2 2
+C
2 2 2 2 2
x 4x +9 u 2 2 u a +u au
u +a
4
2 2
4x +9 4x +9
= 2 +C= +C
9 2x 9x

10. Let u = 2 y and a= 3 . Then du= 2 dy and

2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2 2 2 2
2 u  a du u  a
dy = = 2
2y  3
 2 1 2 2 2
du
y u u
2
2 2
u  a 2 2
= 2  +ln u + u  a +C
u
2
2y  3 2
= 2  +ln 2 y+ 2y  3 +C
2y
2
2y  3 2
= + 2 ln 2 y+ 2y  3 +C
y

11.
1 2  0 2  t  t  t 0
1
 te dt=e+2
0 2  t t 0 0
 t e dt =
 1  1
t e
 1

 1  1  1
te dt=e+2
2
(  t 1 ) e
(  1)  1
0
=e+2  e +0 =e 2

12. Let u =3x . Then du=3dx , so

1 1
 x cos 3xdx = 27  u cos u du= 27 u sin u  2 u sin u du
2 2 2
( )
1 2 2
= x sin 3x (sin 3x 3xcos 3x)+C
3 27
1
=
27
2
(
9x  2 sin 3x+6xcos 3x +C )
 2 1  1 6 2
Thus,  x cos 3xdx=
0 27
2
9x  2 sin 3x+6xcos 3x =
0 27
(
( )
0+6 ( 1) )  ( 0+0 ) =
27
=
9
.

13.
3
tan (1/z) u=1/z, 1
 2
dz 2 = 
3
tan u du=
2
tan u  ln cos u +C
du= dz/z 2
z
1 2 1 1
= tan  ln cos +C
2 z z

2
14. Let u = x . Then u =x and 2u du=dx , so

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2 2
 1  1 2u  1  1 u 1 u
 sin x dx =2 usin udu=
2
sin u+
2
+C
2x 1  1 x(1 x)
= sin x+ +C
2 2

x x
15. Let u =e . Then du=e dx , so
 1  1

x
( x)
e sech e dx= sechu du
107
= tan sinh u +C=tan sinh e ( x) +C

2
16. Let u =x , so that du=2xdx . Then
1 1 cos (1 3)u 1 cos (1+3)u
 xsin x cos 3x dx = 2  sin u cos 3u du= 2 2(1 3)  2 2(1+3) +C
( ) ( )
2 2

1 1 1 2 1 2
= cos 2u  cos 4u +C= cos (2x ) cos (4x )+C
8 16 8 16

2 2 2 2 2
17. Let z=6+4y 4y =6 (4y  4y+1)+1=7 (2y 1) , u =2y 1 , and a= 7 . Then z=a  u , du=2dy , and

1 2 1
= y z dy=
2
(u+1) a  u du

2
y 6+4y 4y dy 2 2
1 1
=  u a  u du+  a  u du
2 2 2 2
4 4
1 1
=  a  u du 
2 2 2 2
( 2u) a  u du
4 8
2
u a  u 1 2 2
 w=a  u ,
2 2 1
= a  u + sin  w dw
8 8 a 8 dw= 2udu
2y 1 72  1 2y 1 1 2 3/2
= 6+4y 4y + sin   w +C
8 8 7 8 3
2y 1 2 7  1 2y 1 1 2 3/2
= 6+4y 4y + sin  (6+4y 4y ) +C.
8 8 7 12

This can be rewritten as

4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2 1 1 2 7  1 2y 1
6+4y 4y (2y 1) (6+4y 4y ) + sin +C
8 12 8 7
1 2 1 5 2 7  1 2y 1
= y y 6+4y 4y + sin +C
3 12 8 8 7
1 2 2 7  1 2y 1
= (8y  2y 15) 6+4y 4y + sin +C
24 8 7

2
18. Let u =x . Then du=2xdx , so by Formula 48,
5 2
x dx
 2
=
1

u
du=
1 1
 ( u + 2 )
2
 4 2 ( u + 2 ) +4ln u + 2 +C
x+ 2 2 u+ 2 2 2
1 2
=
4
2 2
( )
x + 2  4 2 x + 2 +4ln x + 2 +C
2
( ) ( )
1 4 1 2
=
4
x
2
2
x +ln x + 2 +K ( )
2
Or: Let u =x + 2 .

19. Let u =sin x . Then du=cos xdx , so

2+1
u 1 3
sin xcos xln (sin x)dx = u ln udu=
2

2
(2+1)ln u  1 +C= u (3ln u 1)+C
2 9
(2+1)
1 3
= sin x 3ln (sin x) 1 +C
9

x
20. Let u =e . Then x=ln u , dx=du/u , so

 x dx x =  du/u = 2 du = 1 +2ln 1+2u

+C
e 1+2e ( u (1+2u ) )
u (1+2u )
u u

=  e x+2ln e x+2 +C
( )
x x
21. Let u =e and a= 3 . Then du=e dx and
x x
e du 1 u +a 1 e+ 3
 2x
dx=
2
=
2 2a
ln
u a
+C= ln x
+C .
3 e a u 2 3 e 3

22. Let
5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2
u =x and a=2 . Then du=2xdx and

1 2 2 1 4
 x 2 2 x  (x )  2xdx=  u 2au u du
2 22 2
=

2 3 2 4
x 4x  x dx 2 0 2 0
0
2 2 3 4
2u  au 3a a 
2 1 a u
= 2au u + cos
12 4 a 0
2 4
2u  2u  12 2 8  1 2 u
= 4u  u + cos
12 4 2 0
2 4
u  u 6 2  1 2 u
= 4u  u +2cos
6 2 0
 1
= (0+2cos 1)=2   2 0=2
23.
1 3 1 3 1 1
sec xdx = 4 tan xsec x+ 4  sec xdx= 4 tan xsec x+ 4 tan xsec x+  sec xdx
3 3 3

5
2 2
1 3 3 3
= tan xsec x+ tan xsec x+ ln sec x+tan x +C
4 8 8

24. Let u =2x . Then du=2dx , so

1 1 1 5
 sin 2xdx = 2  sin u du= 2  6 sin u cos u + 6  sin u du
6 6 5 4

1 5 1 3
sin u cos u +  sin u du
5 3 2
= sin u cos u + 
12 12 4 4
1 5 5 3 5 1 1
= sin u cos u  sin u cos u + u sin 2u +C
12 48 16 2 4
1 5 5 3 5 5
=  sin 2xcos 2x sin 2xcos 2x sin 4x+ x+C
12 48 64 16

dx
25. Let u =ln x and a=2 . Then du= and
x

2 2


4+(ln x)
x
2 2 u 2 2 a
dx = a +u du= 2 a +u + 2 ln u + a +u
2 2
+C ( )
1 2 2
= (ln x) 4+(ln x) +2ln ln x+ 4+(ln x) +C
2
26.

6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

4  x 4  x 3  x 4  x 3  x 2  x
 x e dx = x e +4 x e dx= x e +4  x e +3 x e dx ( )
= ( x4+4x3) e x+12 (  x2e x+2 xe x dx)
= ( x4+4x3+12x2) e x+24 ( x 1)e x +C
= ( x4+4x3+12x2+24x+24) e x+C
1 4  x
So  x e dx = 
0
( x4+4x3+12x2+24x+24) e x 1
0
 1 0  1
= (1+4+12+24+24)e +24e =24 65e .

x
27. Let u =e . Then x=ln u , dx=du/u , so
2
u  1  1 
 e  1 dx=
2x
u
2 2x
du= u  1  cos ( 1/u ) +C= e  1  cos
1
( e x) +C .
28. Let u= t 3 and assume that   0 . Then du= dt and
1 ( u+3 ) / 1 3/
 e sin ( t 3)dt =   e sin udu= e  e
t (1/ )u
sin udu

( 1/ ) u
1 3/ e 1
= e sin u cos u +C
 2 2 
(1/ ) +1
2
1 3/ (1/ )u  1
= e e sin u cos u +C
 2 
1+
1 ( u+3 ) /
= e (sin u  cos u)+C
2
1+
1 t
= e sin ( t 3)  cos ( t 3) +C
2
1+

29.
4 4
x dx x dx 1 du
 = 
5 4
= [ u =x , du=5x dx ]
10 52 5 2
x  2 (x )  2 u  2
1 2 1 5 10
= ln u + u  2 +C= ln x + x  2 +C
5 5

7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2
30. Let u =tan  and a=3 . Then du=sec  d and

2 2 2 2
sec  tan  =
u u 2 a 
2 1 u
 d
2 2
du=
2
a  u + sin
2 a
+C
2
9 tan  a u
1 2 9  1 tan 
= tan  9 tan  + sin +C
2 2 3

31. Using cylindrical shells, we get

2 2 2
x 2 16  x
 
2 2 2 2 1
V = 2 x x 4 x dx=2 x 4 x dx=2 (2x  4) 4 x + sin
0 0 8 8 2 0

 1  1  2
= 2 (0+2sin 1) (0+2sin 0 =2 2 =2
2

32. Using disks, we get

 /4 1  /4  /4 1  /4
Volume =  tan xdx=
4 3
tan x   tan xdx =
2 3
tan x tan x+x
0 3 0 0 3 0
1   2
=  1+ = 
3 4 4 3

2 2
d 1 a 1 ba 2ab
33. (a) a+bu  2aln a+bu +C = b+ 
du 3 a+bu 3 2 (a+bu)
b b (a+bu)
2 2 3 2 2
1 b(a+bu) +ba  (a+bu)2ab 1 bu u
= = =
3 2 3 2 2
b ( a+bu ) b (a+bu) (a+bu)

t a 1
(b) Let t=a+bu dt=bdu . Note that u = and du= dt .
b b
2 2 2 2
u du 1 (t a) 1 t  2at+a
 2
= 3
 2
dt=
3
 2
dt
(a+bu) b t b t
2 2
1 2a a 1 a
=
3
 1
t
+
2
dt=
3
t 2aln t 
t
+C
b t b

8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

2
1 a
= a+bu  2aln a+bu +C
3 a+bu
b

4
d u a  u
34. (a)
du 8
( 2u 2
 a
2
) 2
a  u + sin
8
2 1
a
+C
4
u  u u 2 2 1 a
=
8
( 2u 2
 a
2
) + a u
2 2
8
(4u )+ 2u  a
8
( ) +
8
1/a
2 2 2 2
a u 1 u /a

=
u
2
( 2u 2 a2) + 2
a u
2 u
+
2
2u  a
2 2
+
a
4

2 2 2 8 2 2
8 a u 8 a u
2 4
1 2 2  1/2 u 1 2 2 a
=
2
a u ( ) 
4
( 2u
(a  u ) ( ) (
2
 a +u
2
) 2 2 2
+
4
a u
2
2u  a +
4
2
)
u (a  u )
2 2 2
=
1 2 2
2
(
a u )  1/2 2 2 4
2u a  2u = =u
2 2
a u
2
2 2
a u

(b) Let u =asin   du=acos  d . Then

 a  u du =  a sin  a 1 sin  d =a  sin  cos  d

2 2 2 2 2 2 4 2 2
u acos 
4 1 1 1 4
=a
2
(1+cos 2 ) (1
2
cos 2 )d = a  1 cos 2
4
2
d ( )
1 4 1 1 4 1 1
= a  1 (1+cos 4 ) d = a   sin 4 +C
4 2 4 2 8
1 4 1 1 1 4 1 1
= a
4 2
 
8
2sin 2 cos 2 +C= a
4 2
 
2
sin  cos  ( 1 2sin 2 ) +C

4 2 2 2
a  1 u u a u 2u
= sin  1 +C
8 a a a 2
a
4 2 2 2 2
a  1 u u a u a  2u
= sin  +C
8 a a a 2
a
4

= u ( 2u 2
 a
2
) 2 a
a  u + sin
2 1 u
+C
8 8 a

35. Maple, Mathematica and Derive all give

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

1 2 3/2 5 2 25  1 1

2
x
2
5 x dx=
4
x 5 x ( )
+ x 5 x + sin
8 8 5
x . Using Formula 31, we get

1 2 1  1 1
2 2
(
 x 5 x dx= 8 x 2x  5 5 x + 8 5 sin
2 2
) 5
( )
x +C . But

1 2 3/2 5 2 1 1  1

4
(
x 5 x ) 2 2
+ x 5 x = x 5 x 5 2 5 x = x 2x  5
8 8 8
2
( 2
) (
5 x , and the sin terms are )
the same in each expression, so the answers are equivalent.

36. Maple and Mathematica both give  x

2
( 1+x3) 4 dx= 151 x15+ 13 x12+ 23 x9+ 23 x6+ 13 x3 , while
Derive gives  x ( 1+x3) 4 dx= 151 ( x3+1) 5 . Using the substitution u =1+x3 du=3x2 dx , we get
2

3 4 1 1 5 1 3 5
 x ( 1+x ) dx= u ( 1+x ) +C . We can use the Binomial Theorem or a
2 4
du = u +C=
3 15 15
1 3 5 1 1 3 2 6 2 9 1 12 1 15
CAS to expand this expression, and we get
15
( 1+x ) +C=
15 3
+ x+ x+ x+ x +
3 3 3 15
x +C
.

1 2
37. Maple and Derive both give  sin xcos xdx=
2 3 3 2 3
sin xcos x cos x (although Derive factors
5 15
1 1 1
the expression), and Mathematica gives  sin xcos xdx=
3 2
cos x cos 3x+ cos 5x . We can
8 48 80
1 3 1 5
use a CAS to show that both of these expressions are equal to  cos x+ cos x . Using Formula
3 5
86, we write
1 2 1 2 1
 sin xcos xdx =  5 sin xcos x+ 5  sin xcos xdx= 5 sin xcos x+ 5  3 cos x +C
3 2 2 3 2 2 3 3

1 2 3 2 3
=  sin xcos x cos x+C
5 15
3 3
1 sin x 2 sin x
38. Maple gives  tan xsec dx=
2 4
+ , Mathematica gives
5 5 15 3
cos x cos x
1
 tan xsec dx= 120 sec x( 20sin x+5sin 3x+sin 5x) , and Derive gives
2 4 5

2 sin x sin x
 tan xsec dx= 15 tan x
2 4
3
+ 5
. All of these expressions can be ‘‘simplified’’ to
15cos x 5cos x


( 2
1 sin x cos x 2cos x 3
4
) 2
using Maple. Using the identity 1+tan x=sec x , we write
2
15 5
cos x

10
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

 tan xsec xdx= tan x 1+tan x sec xdx=

2 4 2
( 2
) 2
( tan 2x+tan 4x) sec 2xdx . Now we substitute u =tan x
 du=sec xdx , and the integral becomes 
2
( u 2+u 4) du= 13 u 3+ 15 u 5+C= 13 tan 3x+ 15 tan 5x+C . If
5 3
( 2
)
we write sin x=sin x 1 cos x and substitute into the numerator of the tan x term, this becomes
5

+
3
1 sin x 1 sin x 1 cos x
3
+C=
(1 sin x
+
2
) 3
1 1

sin x
3
+C=
1 sin x 2 sin x
+
3
+C ,
3

3 3 5 5 5 5 3 5 3 5 5 15 3
cos x cos x cos x cos x cos x cos x
which is the same as Maple’s expression.

1 5/2 1
39. Maple gives  x 1+2x dx=
3/2
(1+2x)  (1+2x) , Mathematica gives
10 6
2 2 1 1 1 3/2
1+2x x+ x , and Derive gives (1+2x) (3x 1) . The first two expressions can be
5 15 15 15
simplified to Derive’s result. If we use Formula 54, we get
2 3/2 1 3/2
 x 1+2x dx = 2
(3 2x 2 1)(1+2x) +C=
30
(6x 2)(1+2x) +C
15 ( 2 )
1 3/2
= (3x 1)(1+2x)
15

1 3 3
40. Maple and Derive both give  sin xdx=
4 3
sin xcos x cos xsin x+ x , while Mathematica
4 8 8
1
gives (12x 8sin 2x+sin 4x) , which can be expanded and simplified to give the other expression.
32
Now

1 3 1 3 1 1
sin xdx = 4 sin xcos x+ 4  sin xdx= 4 sin xcos x+ 4
3 2 3

4 x sin 2x +C
2 4
1 3 3 3
= sin xcos x sin xcos x+ x+C since sin 2x=2sin xcos x
4 8 8

1 1 1
41. Maple gives  tan xdx=
5
4
4
tan x
2
2
2
2
(
tan x+ ln 1+tan x , Mathematica gives)
1

5 4
tan xdx= [ 1 2cos (2x)]sec x ln (cos x) , and Derive gives
4
1 1
 tan xdx= 4 tan x 2 tan x ln (cos x) . These expressions are equivalent, and none includes
5 4 2

absolute value bars or a constant of integration. Note that Mathematica’s and Derive’s expressions
suggest that the integral is undefined where cos x<0 , which is not the case.
Using Formula 75,
11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

1 1
 tan x  tan xdx= tan x  tan xdx . Using Formula 69,
5 5 1 5 2 4 3
tan xdx=
5 1 4
1 1 1
 tan xdx= 2 tan x+ln cos x +C , so  tan xdx= 4 tan x 2 tan x ln cos x +C .
3 2 5 4 2

1 4 4 2 8 2 1 6
42. Maple gives  x x +1 dx=
5 2 2 2 2
x 1+x  x 1+x + 1+x + x 1+x . When we
35 105 105 7
1 2 3/2
use the factor command on this expression, it becomes
105
1+x
4 2
15x  12x +8 . Mathematica ( ) ( )
2 8 4 2 1 4 1 6
gives 1+x  x+ x+ x , which again factors to give the above expression, and
105 105 35 7
Derive gives the factored form immediately. If we substitute u = x +1 
2 4
(
x= u  1
2
) 2 , xdx=u du ,
then the integral
becomes
1 7 2 5 1 3
( 2 2
) (
 u  1 u (u du) =  u  2u +1 u du= 7 u  5 u + 3 u +C
4 2 2
)
3/2 1 2 2 2 1
2
= x +1 (
7
x +1  )5
2
x +1 + (
3
+C ) ( )
1 3/2 2
=
105
2
x +1
2
( ) 2
15 x +1  42 x +1 +35 +C ( ) ( )
1 3/2
=
105
2
x +1
4
(2
) (
15x  12x +8 +C )
43. Derive gives I= 2
x x
4  1 dx=
2
x 1 2x
2  1 ln

( 2x
2  1 +2
x
) immediately. Neither Maple nor
ln 2 2ln 2
Mathematica is able to evaluate I in its given form. However, if we instead write I as  2 (2 )  1 dx
x x2

, both systems give the same answer as Derive (after minor simplification). Our trick works because
x
the CAS now recognizes 2 as a promising substitution.

44. None of Maple, Mathematica and Derive is able to evaluate  (1+ln x) 1+(xln x) dx . However,
2

if we let u =xln x , then du=(1+ln x)dx and the integral is simply  1+u du , which any CAS can
2

1
(2 1
evaluate. The antiderivative is ln xln x+ 1+ ( xln x ) + xln x 1+(xln x) +C .
2 2
2
)
2
x 1 1 1
45. Maple gives the antiderivative F(x)= 4 2
dx=
2
2
( 2
2
)
ln x +x+1 + ln x  x+1 . We can ( )
x +x +1
see that at 0 , this antiderivative is 0 . From the graphs, it appears that F has a maximum at x= 1 and
12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

a minimum at x=1 , and that F has inflection points at x  1.7 , x=0 , and x 1.7 .

46. Maple gives the antiderivative which, after we use the simplify command, becomes
 x 1 x 1
 xe sin xdx= 2 e (cos x+xcos x+xsin x) . At x=0 , this antiderivative has the value  2 , so we use
1 x 1
F(x)= e (cos x+xcos x+xsin x)+ to make F(0)=0 .
2 2

From the graphs, it appears that F has a minimum at x  3.1 and a maximum at x 3.1 , and that F
/
has inflection points where f changes sign, at x  2.5 , x=0 , x 1.3 and x 4.1 .

4 6
47. Since f (x)=sin xcos x is everywhere positive, we know that its antiderivative F is increasing.
Maple gives
1 3 1 1 3 3
 f (x)dx= 10 sin xcos x 80 sin xcos x+ 160 cos xsin x+ 128 cos xsin x+ 256 cos xsin x+ 256 x
3 7 7 5 3

and this expression is 0 at x=0 .

/
F has a minimum at x=0 and a maximum at x= . F has inflection points where f changes sign, that
13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.6 Integration Using Tables and Computer Algebra Systems

is, at x 0.7 , x= /2 , and x 2.5 .

3
x x
48. From the graph of f (x)= 6
, we can see that F has a maximum at x=0 , and minima at x 1 .
x +1
1 1
The antiderivative given by Maple is F(x)=
3
2
( 6
)
4 2
( )
ln x +1 + ln x  x +1 , and F(0)=0 . Note that f
is odd, and its antiderivative F is even.

/
F has inflection points where f changes sign, that is, at x 0.5 and x 1.4 .

14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1. (a)  x=(b a)/n=(4 0)/2=2

L =
2
2 i=1 ( )  x= f ( x ) 2+ f ( x ) 2=2 f (0)+ f (2) =2(0.5+2.5)=6
f x
i 1 0 1

R =
2
f ( x )  x= f ( x )  2+ f ( x )  2=2 f (2)+ f (4) =2(2.5+3.5)=12
2 i=1 i 1 2

M =
2
2 i=1 ( ) i ( ) 1 ( )
f x  x= f x  2+ f x  2=2 f (1)+ f (3)  2(1.6+3.2)=9.6
2

(b)
L is an underestimate, since the area under the small rectangles is less than the area under the curve,
2
and R is an overestimate, since the area under the large rectangles is greater than the area under the
2
curve. It appears that M is an overestimate, though it is fairly close to I . See the solution to Exercise
2
45 for a proof of the fact that if f is concave down on a,b , then the Midpoint Rule is an
overestimate of 
b
f (x)dx .
a

1 2
(c) T =  x f x +2 f x + f x ( ) ( ) ( )
= f (0)+2 f (2)+ f (4) =0.5+2(2.5)+3.5=9 .
2 2 0 1 2 2
This approximation is an underestimate, since the graph is concave down. Thus, T =9<I . See the
2
solution to Exercise 45 for a general proof of this conclusion.
(d) For any n , we will have L <T <I<M <R .
n n n n

2.

The diagram shows that L >T > f (x)dx>R , and it appears that M is a bit less than 
2 2
f (x)dx . In
4 4 0 4 4 0

fact, for any function that is concave upward, it can be shown that L >T >
2
f (x)dx>M >R .
n n 0 n n
(a) Since 0.9540>0.8675>0.8632>0.7811 , it follows that L =0.9540T =0.8675 , M =0.8632 , and
n n n

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

R =0.7811 .
n

(b) Since M < f (x)dx<T , we have 0.8632<

2 2
f (x)dx<0.8675 .
n 0 n 0

1 0 1
3. f (x)=cos x ( 2) , 
4
=
4
x=
1 1 2 3
(a) T = f (0)+2 f +2 f +2 f + f (1)  0.895759
4 4 2 4 4 4
1 1 3 5 7
(b) M = f +f +f +f  0.908907
4 4 8 8 8 8

The graph shows that f is concave down on 0,1 . So T is an underestimate and M is an

4 4

overestimate. We can conclude that 0.895759<

1
0
( )
2
cos x dx<0.908907 .

4.

(a) Since f is increasing on 0,1 , L will underestimate I (since the area of the darkest rectangle is
2
less than the area under the curve), and R will overestimate I . Since f is concave upward on 0,1 ,
2
M will underestimate I and T will overestimate I (the area under the straight line segments is
2 2
greater than the area under the curve).
(b) For any n , we will have L <M <I<T <R .
n n n n
1
(c) L =
5
f x ( )  x= [ f (0.0)+ f (0.2)+ f (0.4)+ f (0.6)+ f (0.8)] 0.1187
5 i=1 i 1 5
1
R =
5
f x  x= [ f (0.2)+ f (0.4)+ f (0.6)+ f (0.8)+ f (1)] 0.2146
( )
5 i=1 i 5
1
M =
5
5 i=1 i ( )
f x  x= [ f (0.1)+ f (0.3)+ f (0.5)+ f (0.7)+ f (0.9)] 0.1622
5
2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1
T =  x  0.1666
5 2
From the graph, it appears that the Midpoint Rule gives the best approximation. (This is in fact the
case, since I 0.16371405 .)

2 b a   0 
5. f (x)=x sin x ,  x= = =
n 8 8
  3 5 15
(a) M = f +f +f +   + f  5.932957
8 8 16 16 16 16
(b)
S =  f (0)+4 f

+2 f
2
+4 f
3
+2 f
4
+4 f
5
+2 f
6
+4 f
7
+
8 8 3 8 8 8 8 8 8 8
 5.869247

Actual:
   

2
+2
2 2
x sin xdx =  x cos x xcos xdx=   (  1 )  0 +2 cos x+xsin x
0 0 0 0
2 2
= +2[ (  1+0 )  ( 1+0 ) ]=  4 5.869604


Errors: E = actual  M =
2
x sin xdx M   0.063353
M 8 0 8

E = actual  S =
2
x sin xdx S  0.000357
S 8 0 8

b a 1 0 1 x
6. f (x)=e ,  x= = =
n 6 6
1 1 3 5 7 9 11
(a) M = f +f +f +f +f +f  0.525100
6 6 12 12 12 12 12 12
1 1 2 3 4 5
(b) S = f (0)+4 f +2 f +4 f +2 f +4 f + f (1)  0.533979
6 6 3 6 6 6 6 6
Actual:
1   1 u
dx =
x 2
 0
e 0
e 2uduu= x , u =x , 2udu=dx ]
u  1  1  1
=2 (u 1)e
0
=2  2e  (  1e0) =2 4e  0.528482

1  x
Errors: E = actual  M = e dx M  0.003382
M 6 0 6
1  x
E = actual  S = e dx S   0.005497
S 6 0 6

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

4 2 2 0 1
7. f (x)= 1+x ,  x= =
8 4
1 1 1 3 7
(a) T = f (0)+2 f +2 f +   +2 f +2 f + f (2)  2.413790
8 4 2 4 2 2 4
1 1 3 13 15
(b) M = f +f +   + f +f  2.411453
8 4 8 8 8 8
(c)
1 1 1 3 5 3 7
S= f (0)+4 f +2 f +4 f +2 f (1)+4 f +2 f +4 f + f (2)  2.4122
8 4 3 4 2 4 4 2 4

1
 0
2 1
2
( )
8. f (x)=sin x ,  x=
4
=
8
1 1 2 3 1
(a) T = f (0)+2 f +2 f +2 f +f  0.042743
4 8 2 8 8 8 2
1 1 3 5 7
(b) M = f +f +f +f  0.040850
4 8 16 16 16 16
1 1 2 3 1
(c) S = f (0)+4 f +2 f +4 f +f  0.041478
4 8 3 8 8 8 2

ln x 2 1 1
9. f (x)= ,  x= =
1+x 10 10
1
(a) T = f (1)+2 f (1.1)+2 f (1.2)+   +2 f (1.8)+2 f (1.9)+ f (2)  0.146879
10 10 2
1
(b) M = f (1.05)+ f (1.15)+   + f (1.85)+ f (1.95)  0.147391
10 10
(c)
S = 1 [ f (1)+4 f (1.1)+2 f (1.2)+4 f (1.3)+2 f (1.4)+4 f (1.5)+2 f (1.6)+4 f (1.7)]
10 10 3
+ 2 f (1.8)+4 f (1.9)+ f (2)]
 0.147219
1 3 0 1
10. f (t)= ,  t= =
2 4 6 2
1+t +t
1 1 3 5
(a) T = f (0)+2 f +2 f (1)+2 f +2 f (2)+2 f + f (3)  0.895122
6 2 2 2 2 2
1 1 3 5 7 9 11
(b) M = f +f +f +f +f +f  0.895478
6 2 4 4 4 4 4 4
1 1 3 5
(c) S = f (0)+4 f +2 f (1)+4 f +2 f (2)+4 f + f (3)  0.898014
6 2 3 2 2 2
11.
4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1
 0
t /2 2 1
f (t)=sin (e ) ,  t= =
8 16
1 1 2 7 1
(a) T = f (0)+2 f +2 f +   +2 f +f  0.451948
8 16 2 16 16 16 2
1 1 3 5 13 15
(b) M = f +f +f +   + f +f  0.451991
8 16 32 32 32 32 32
1 1 2 7 1
(c) S = f (0)+4 f +2 f +   +4 f +f  0.451976
8 16 3 16 16 16 2

4 0 1
12. f (x)= 1+ x ,  x= =
8 2
1 1 7
(a) T = f (0)+2 f +2 f (1)+   +2 f (3)+2 f + f (4)  6.042985
8 2 2 2 2
1 1 3 13 15
(b) M = f +f +   + f +f  6.084778
8 2 4 4 4 4
(c)
1 1 3 5 7
S= f (0)+4 f +2 f (1)+4 f +2 f (2)+4 f +2 f (3)+4 f + f (4)  6.061678
8 2 3 2 2 2 2

1/x 2 1 1
13. f (x)=e ,  x= =
4 4
1
(a) T = [ f (1)+2 f (1.25)+2 f (1.5)+2 f (1.75)+ f (2)] 2.031893
4 4 2
1
(b) M = [ f (1.125)+ f (1.375)+ f (1.625)+ f (1.875)] 2.014207
4 4
1
(c) S = [ f (1)+4 f (1.25)+2 f (1.5)+4 f (1.75)+ f (2)] 2.020651
4 4 3

4 0 1
14. f (x)= x sin x ,  x= =
8 2
(a)
T =
8
1
2 2
1
{ f (0)+2
1
f
1
2
3
+ f (1)+ f
5
3
2
+ f (2)+ f
7
5
2
+ f (3)+ f
13
7
2
15
}
+ f (4)  1.732865

(b) M = f +f +f +f +   + f +f  1.787427
8 2 4 4 4 4 4 4
(c)
1 1 3 5 7
S= f (0)+4 f +2 f (1)+4 f +2 f (2)+4 f +2 f (3)+4 f + f (4)  1.772142
8 2 3 2 2 2 2
cos x 5 1 1
15. f (x)= ,  x= =
x 8 2
5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1 3 9
(a) T = f (1)+2 f +2 f (2)+   +2 f (4)+2 f + f (5)   0.495333
8 2 2 2 2
(b)
1 5 7 9 11 13 15 17 19
M= f +f +f +f +f +f +f +f   0.5433
8 2 4 4 4 4 4 4 4 4
(c)
S = 1 f (1)+4 f
3
+2 f (2)+4 f
5
+2 f (3)+4 f
7
+2 f (4)+4 f
9
+ f (5)
8 2 3 2 2 2 2
  0.526123

6 4 1
(
16. f (x)=ln x +2 ,  x=
3
) 10 5
=
1
(a) T = f (4)+2 f (4.2)+2 f (4.4)+   +2 f (5.6)+2 f (5.8)+ f (6)  9.649753
10 5 2
1
(b) M = f (4.1)+ f (4.3)+   + f (5.7)+ f (5.9)  9.650912
10 5
(c)
S = 1 [ f (4)+4 f (4.2)+2 f (4.4)+4 f (4.6)+2 f (4.8)+4 f (5)]
10 5 3
+ 2 f (5.2)+4 f (5.4)+2 f (5.6)+4 f (5.8)+ f (6)]
 9.650526

1 3 0 1
17. f (y)= ,  y= =
5 6 2
1+y
1 1 2 3 4 5
(a) T = f (0)+2 f +2 f +2 f +2 f +2 f + f (3)  1.064275
6 2 2 2 2 2 2 2
1 1 3 5 7 9 11
(b) M = f +f +f +f +f +f  1.067416
6 2 4 4 4 4 4 4
1 1 2 3 4 5
(c) S = f (0)+4 f +2 f +4 f +2 f +4 f + f (3)  1.074915
6 2 3 2 2 2 2 2

x
e 4 2 1
18. f (x)= ,  x= =
x 10 5
1
(a) T =
10 5 2
{ f (2)+2[ f (2.2)+ f (2.4)+ f (2.6)+   + f (3.8)]+ f (4)}  14.704592
1
(b) M = [ f (2.1)+ f (2.3)+ f (2.5)+ f (2.7)+   + f (3.7)+ f (3.9)] 14.662669
10 5
1
(c) S = [ f (2)+4 f (2.2)+2 f (2.4)+4 f (2.6)+   +2 f (3.6)+4 f (3.8)+ f (4)] 14.676696
10 5 3
6
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

2
 x 2 0 1
19. f (x)=e ,  x= =
10 5
1
(a) T = { f (0)+2[ f (0.2)+ f (0.4)+   + f (1.8)]+ f (2)}  0.881839
10 5 2
1
M = [ f (0.1)+ f (0.3)+ f (0.5)+   + f (1.7)+ f (1.9)] 0.882202
10 5
2 2 2 2
 x  x  x  x
(b) f (x)=e
/
, f (x)= 2xe
/ /
, f (x)= 4x  2 e , f ( 2
)
(x)=4x 3 2x e .
/ / /
( 2
)

.
/ / / 3 / /
f (x)=0 x=0 or x= . So to find the maximum value of f (x) on 0,2 , we need
2
3 / / / /
only consider its values at x=0 , x=2 , and x= . f (0) =2 , f (2)  0.2564 and
2
/ / 3
f  0.8925 . Thus, taking K=2 , a=0 , b=2 , and n=10 in Theorem 3, we get
2
E
T

2 2
3
/ ( 12 10 ) = 751 =0.013 , and
2
E
M

E
T / 2
0.006 .
3 3
K ( b a )  5 2(2 0)  5 3 2 5
(c) Take K=2 in Theorem 3. E

10 
10  n  10  n 365.1 ... 
T 2 2 4
12n 12n
 5 3 2 5
n 366 . Take n=366 for T . For E , again take K=2 in Theorem 3 to get E
10  n  10
n M M 2
 n 258.2 n 259 . Take n=259 for M .
n

20. (a) T =
1
1
8 8 2
1
{ f (0)+2
3
f
1
8
5
+f
2
8
+   + f
15
7
8
+ f (1) }  0.902333

M= f +f +f +   + f =0.905620
8 8 16 16 16 16
(b) f (x)=cos x ( 2) , f /(x)= 2xsin ( x2) , f / /(x)= 2sin ( x2)  4x2cos ( x2) . For 0
x
1 , sin and
cos are positive, so f (x) =2sin ( x ) +4x cos ( x )
2 1+4 1 1=6 since sin ( x )
1 and
/ / 2 2 2 2

cos ( x )
1 for all x , and x
1 for 0
x
1 . So for n=8 , we take K=6 , a=0 , and b=1 in Theorem
2 2

1 1
3, to get E
6 1 / ( 12 8 ) =
3 2
=0.0078125 and E
=0.00390625 .
T 128 M 256
 5
(c) Using K=6 as in part (b), we have E
6 1 / ( 12n ) =1/ ( 2n )
10  2n  10 
3 2 2 2 5
T
1  5
 10 or n 224 . To guarantee that E
0.00001 , we need 6 1 / ( 24n )
10 
5 3 2
n
2 M

2 5 1 5
4n  10  n  10 or n 159 .
4
7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1
21. (a) T = { f (0)+2[ f (0.1)+ f (0.2)+   + f (0.9)]+ f (1)}  1.71971349
10 10 2
1
S = [ f (0)+4 f (0.1)+2 f (0.2)+4 f (0.3)+   +4 f (0.9)+ f (1)] 1.71828278
10 10 3
x 1
Since I=
1 x
e dx= e =e 1 1.71828183 , E =I T   0.00143166 and E =I S   0.00000095 .
0 0 T 10 S 10
x / / x
(b) f (x)=e  f ( x ) =e
e for 0
x
1 . Taking K=e , a=0 , b=1 , and n=10 in Theorem 3, we get
E
T

e(1)
3
/ ( 12 10 )  0.002265>0.00143166 . f ( ) (x)=e <e for 0
x
1 . Using Theorem 4, we
2 4 x

e(1) / ( 180 10 )  0.0000015>0.00000095 . We see that the actual errors are about two
5 4
have E

S
 thirds the size of the error estimates.

(c) From part (b), we take K=e to get E

K(b a)
3

0.00001
2
n
e 1 ( 3)
 n 150.5 .
T 2 12(0.00001)
12n
3
K(b a)
Take n=151 for T . Now E

0.00001 n 106.4 . Take n=107 for M . Finally,
n M 2 n
24n

K(b a)
5

0.00001 n
4 ( 5)
e 1
 n 6.23 . Take n=8 for S (since n has to be even
S 4 180(0.00001) n
180n
for Simpson’s Rule).

22. From Example 7(b), we take K=76e to get E

76e(1)
S
5
/ ( 180n )

4
0.00001
4
n  76e/[180(0.00001)] n 18.4 . Take n=20 (since n must be even).

23. (a) Using a CAS, we differentiate f (x)=e

cos x
twice, and find that f
/ /
(x)=e
cos x
( sin 2x cos x) .
/ /
From the graph, we see that the maximum value of f (x) occurs at the endpoints of the
/ /
interval 0,2 . Since f (0)= e , we can use K=e or K=2.8 .

(b) A CAS gives M  7.954926518 . (In Maple, use student[middlesum]. )

10
(c) Using Theorem 3 for the Midpoint Rule, with K=e , we get
8
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

3 3
e(2  0) 2.8(2  0)
E
 0.280945995 . With K=2.8 , we get E
=0.289391916 .
M 2 M 2
24 10 24 10
(d) A CAS gives I 7.954926521 .
 9
(e) The actual error is only about 3 10 , much less than the estimate in part (c).
(f) We use the CAS to differentiate twice more, and then graph
( 4)
f (x)=e
cos x 4
( 2 2
sin x 6sin xcos x+3 7sin x+cos x . )
(4)
From the graph, we see that the maximum value of f (x) occurs
( 4)
at the endpoints of the interval 0,2 . Since f (0)=4e , we can use
K=4e or K=10.9 .

(g) A CAS gives S  7.953789422 . (In Maple, use student[simpson ]. .)

10
5
4e(2  0)
(h) Using Theorem 4 with K=4e , we get E
 0.059153618 . With K=10.9 , we get
S 4
180 10
5
10.9(2  0)
E
 0.059299814 .
S 4
180 10
(i) The actual error is about 7.954926521 7.953789422 0.00114 . This is quite a bit smaller than the
estimate in part (h), though the difference is not nearly as great as it was in the case of the Midpoint
Rule.
5 5
4e(2 ) 4e(2 ) 4
(j) To ensure that E
0.0001 , we use Theorem 4: E

0.0001
n
S S 4 180 0.0001
180 n
4
 n  5 , 915 , 362 n 49.3 . So we must take n 50 to ensure that I S
0.0001 . ( K=10.9
n
leads to the same value of n .)

3
24. (a) Using the CAS, we differentiate f (x)= 4 x twice,
4
/ / 9x 3x
and find that f (x)=  .
3 3/2 3 1/2
(
4 4 x ) ( 4 x )
9
 Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

/ /
From the graph, we see that f (x) <2.2 on  1,1 .

(b) A CAS gives M  3.995804152 . (In Maple, use student[middlesum] .)

10
3
2.2 1 (  1 )
(c) Using Theorem 3 for the Midpoint Rule, with K=2.2 , we get E
 0.00733 .
M 2
24 10
(d) A CAS gives I 3.995487677 .
(e) The actual error is about  0.0003165 , much less than the estimate in part (c).

(f) We use the CAS to differentiate twice more, and then graph f
( 4)
(x)=
9 x
2
( x6 224x3 1280) .
16
( 4 x3) 7/2
(4)
From the graph, we see that f (x) <18.1 on  1,1 .

(g) A CAS gives S  3.995449790 . (In Maple, use student[simpson] .)

10
5
18.1 1 (  1 )
(h) Using Theorem 4 with K=18.1 , we get E
 0.000322 .
S 4
180 10
(i) The actual error is about 3.995487677 3.995449790 0.0000379 . This is quite a bit smaller than
the estimate in part (h).
5 5
18.1(2) 18.1(2) 4
(j) To ensure that E
0.0001 ,we use Theorem 4: E

0.0001
n
S S 4 180 0.0001
180 n
4
 n  32 , 178 n 13.4 . So we must take n 14 to ensure that I S
0.0001 .
n

1 4 1
25. I= x dx=
1 3 3
x =0.25 . f (x)=x .
0 4 0
10
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

n=4 :
1 3 1 3 2 3 3 3
L= 0+ + + =0.140625
4 4 4 4 4
1 1 3 3 2 3 3 3
R= + + +1 =0.390625
4 4 4 4 4
1 3 1 3 2 3 3 3 3
T = 0 +2 +2 +2 +1 =0.265625 ,
4 4 2 4 4 4
1 1 3 3 3 5 3 7 3
M= + + + =0.2421875 ,
4 4 8 8 8 8
1 1
E =I L =  0.140625=0.109375 , E =  0.390625= 0.140625 ,
L 4 4 R 4
1 1
E =  0.265625= 0.015625 , E =  0.2421875=0.0078125
T 4 M 4

n=8 :
1 1 2 7
L= f (0)+ f +f +   + f  0.191406
8 8 8 8 8
1 1 2 7
R= f +f +   + f + f (1)  0.316406
8 8 8 8 8
T =
1
8 8 2
1
{
f (0)+2 f
1
1
8
3
+f
2
8
+   + f
13
7
8
15
}
+ f (1)  0.253906

M= f +f +   +f +f =0.248047
8 8 16 16 16 16
1 1
E   0.191406 0.058594 , E   0.316406  0.066406 ,
L 4 R 4
1 1
E   0.253906  0.003906 , E   0.248047 0.001953 .
T 4 M 4

n=16 :
1 1 2 15
L = f (0)+ f +f +   + f  0.219727
16 16 16 16 16
1 1 2 15
R = f +f +   + f + f (1)  0.282227
16 16 16 16 16
T =
16 16 2
1
1
{
f (0)+2 f
1
1
16
3
+f
2
16
+ 
31
 +f
15
16 }
+ f (1)  0.250977

M = f +f +   + f  0.249512
16 16 32 32 32
1 1
E   0.219727 0.030273 , E   0.282227  0.032227 ,
L 4 R 4

11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1 1
E   0.250977  0.000977 , E   0.249512 0.000488 .
T 4 M 4

n L R T M
n n n n
4 0.140625 0.390625 0.265625 0.242188
8 0.191406 0.316406 0.253906 0.248047
16 0.219727 0.282227 0.250977 0.249512

n E E E E
L R T M
4 0.109375  0.140625  0.015625 0.007813
8 0.058594  0.066406  0.003906 0.001953
16 0.030273  0.032227  0.000977 0.000488

Observations:
(a) E and E are always opposite in sign, as are E and E .
L R T M
(b) As n is doubled, E and E are decreased by about a factor of 2 , and E and E are decreased by
L R T M
a factor of about 4 .
(c) The Midpoint approximation is about twice as accurate as the Trapezoidal approximation.
(d) All the approximations become more accurate as the value of n increases.
(e) The Midpoint and Trapezoidal approximations are much more accurate than the endpoint
approximations.

x 2
26. 
2 x 2 x
e dx= e =e  1 6.389056 . f (x)=e
0 0
n=4 :
1
 x=(2 0)/4=
2
1 0 1/2 1 3/2
L= e +e +e +e  4.924346
4 2
1 1/2 1 3/2 2
R= e +e +e +e  8.118874
4 2
1 0 1/2 1 3/2 2
T = e +2e +2e +2e +e  6.521610
4 2 2
1 1/4 3/4 5/4 7/4
M= e +e +e +e  6.322986 .
4 2
E  6.389056 4.924346 1.464710 , E  6.389056 8.118874= 1.729818 ,
L R

12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

E  6.389056 6.521610  0.132554 , E  6.389056 6.322986=0.0660706 .

T M

n=8 :
1
 x=(2 0)/8=
4
1 0 1/4 1/2 3/4 1 5/4 3/2 7/4
L= e +e +e +e +e +e +e +e  5.623666
8 4
1 1/4 1/2 3/4 1 5/4 3/2 7/4 2
R= e +e +e +e +e +e +e +e  7.220930
8 4
1 0 1/4 1/2 3/4 1 5/4 3/2 7/4 2
T = e +2e +2e +2e +2e +2e +2e +2e +e  6.422298
8 4 2
1 1/8 3/8 5/8 7/8 9/8 11/8 13/8 15/8
M= e +e +e +e +e +e +e +e  6.372448
8 4
E  6.389056 5.623666 0.765390 , E  6.389056 7.220930  0.831874 ,
L R
E  6.389056 6.422298  0.033242 , E  6.389056 6.372448 0.016608 .
T M

n=16 :
1
 x=(2 0)/16=
8
1 1 2 14 15
L = f (0)+ f +f +   + f +f  5.998057
16 8 8 8 8 8
1 1 2 3 15
R = f +f +f +   + f + f (2)  6.796689
16 8 8 8 8 8
1 1 2 3 15
T = f ( 0 ) +2 f +f +f +   + f + f (2)  6.397373
16 8 2 8 8 8 8
1 1 3 5 29 31
M = f +f +f +   + f +f  6.384899
16 8 16 16 16 16 16
E  6.389056 5.998057 0.390999 , E  6.389056 6.796689  0.407633 ,
L R
E  6.389056 6.397373  0.008317 , E  6.389056 6.384899 0.004158 .
T M

n E E E E
L R T M
4 1.464710  1.729818  0.132554 0.066071
8 0.765390  0.831874  0.033242 0.016608
16 0.390999  0.407633  0.008317 0.004158

Observations:
(a) E and E are always opposite in sign, as are E and E .
L R T M
(b) As n is doubled,
13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

E and E are decreased by a factor of about 2 , and E and E are decreased by a factor of about 4 .
L R T M
(c) The Midpoint approximation is about twice as accurate as the Trapezoidal approximation.
(d) All the approximations become more accurate as the value of n increases.
(e) The Midpoint and Trapezoidal approximations are much more accurate than the endpoint
approximations.

2 3/2 4 2 14
27. 
4
x dx= x = ( 8 1 ) =  4.666667
1 3 1 3 3
n=6 :
1
 x= ( 4 1 ) /6=
2
1
T = 1 +2 1.5+2 2 +2 2.5+2 3 +2 3.5+ 4  4.661488
6 2 2
1
M= 1.25+ 1.75+ 2.25+ 2.75+ 3.25+ 3.75  4.669245
6 2
1
S= 1 +4 1.5+2 2 +4 2.5+2 3 +4 3.5+ 4  4.666563
6 2 3
14 14
E   4.661488 0.005178 , E   4.669245  0.002578 ,
T 3 M 3
14
E  4.666563 0.000104 .
S 3

n=12 :
1
 x= ( 4 1 ) /12=
4
1
T =
12 4 2
( f (1)+2 f (1.25)+ f (1.5)+   + f (3.5)+ f (3.75) + f (4) )  4.665367
1
M = f (1.125)+ f (1.375)+ f (1.625)+   + f (3.875)  4.667316
12 4
1
S =  4.666659
12 4 3
14 14
E   4.665367 0.001300 , E   4.667316  0.000649 ,
T 3 M 3
14
E  4.666659 0.000007 .
S 3

Note: These errors were computed more precisely and then rounded to six places. That is, they were
not computed by comparing the rounded values of T , M , and S with the rounded value of the
n n n
actual integral.

14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

n Tn M
n
S
n
6 4.661488 4.669245 4.666563
12 4.665367 4.667316 4.666659

n E E E
T M S
6 0.005178  0.002578 0.000104
12 0.001300  0.000649 0.000007
Observations:
(a) E and E are opposite in sign and decrease by a factor of about 4 as n is doubled.
T M
(b) The Simpson’s approximation is much more accurate than the Midpoint and Trapezoidal
approximations, and seems to decrease by a factor of about 16 as n is doubled.

x 2
28. I=
2 x x 2 x
xe dx= xe  e =e +2/e 8.124815 . f (x)=xe .
 1  1
n=6 :
1
 x=[2 ( 1)]/6=
2
1
T =
6 2 2
{ f ( 1)+2 f ( 0.5)+ f (0)+   + f (1.5) + f (2)}  8.583514
1
M= f ( 0.75)+ f ( 0.25)+   + f (1.75)  7.896632
6 2
1
S= f ( 1)+4 f ( 0.5)+2 f (0)+4 f (0.5)+2 f (1)+4 f (1.5)+ f (2)  8.136885
6 2 3
E  I 8.583514  0.458699 , E  I 7.896632 0.228183 ,
T M
E  I 8.136885  0.012070 .
S

n=12 :
1
 x=[2 ( 1)]/12=
4
1
T =
12 4 2
{ f ( 1)+2 f ( 0.75)+ f ( 0.5)+   + f (1.75) + f (2)}  8.240073
1 7 5 13 15
M = f  +f  +   + f +f  8.067259
12 4 8 8 8 8
1
S = f ( 1)+4 f ( 0.75)+2 f ( 0.5)+   +2 f (1.5)+4 f (1.75)+ f (2)  8.125593
12 4 3

E  I 8.240073  0.115258 , E  I 8.067259 0.057556 ,

T M
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

E  I 8.125593  0.000778
S

n Tn M
n
S
n
6 8.583514 7.896632 8.136885
12 8.240073 8.067259 8.125593

n E E E
T M S
6  0.458699 0.228183  0.012070
12  0.115258 0.057556  0.000778

Observations:
(a) E and E are opposite in sign and decrease by a factor of about 4 as n is doubled.
T M
(b) The Simpson’s approximation is much more accurate than the Midpoint and Trapezoidal
approximations, and seems to decrease by a factor of about 16 as n is doubled.

29.  x= ( 4 0 ) /4=1
1 1
(a) T = f (0)+2 f (1)+2 f (2)+2 f (3)+ f (4)  0+2(3)+2(5)+2(3)+1 =11.5
4 2 2
(b) M =1 f (0.5)+ f (1.5)+ f (2.5)+ f (3.5)  1+4.5+4.5+2=12
4
1 1
(c) S = f (0)+4 f (1)+2 f (2)+4 f (3)+ f (4)  0+4(3)+2(5)+4(3)+1 =11.6
4 3 3

30. If x= distance from left end of pool and w=w(x)= width at x , then Simpson’s Rule with n=8 and
2
 x=2 gives Area = wdx
16 2
0+4(6.2)+2(7.2)+4(6.8)+2(5.6)+4(5.0)+2(4.8)+4(4.8)+0  84 m .
0 3

31. (a) We are given the function values at the endpoints of 8 intervals of length 0.4, so we’ll use the
Midpoint Rule with n=8/2=4 and  x=(3.2 0)/4=0.8 .

3.2
f (x)dx M =0.8[ f (0.4)+ f (1.2)+ f (2.0)+ f (2.8)]
0 4
=0.8[6.5+6.4+7.6+8.8]
=0.8(29.3)=23.44

/ / / /
(b)  4
f (x)
1 f (x)
4 , so use K=4 , a=0 , b=3.2 , and n=4 in Theorem 3. So

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

3
4(3.2 0) 128
E
= =0.3413 .
M 2 375
24(4)

1
32. We use Simpson’s Rule with n=10 and  x= :
2

1
distance=
5
v(t)dt S = f (0)+4 f (0.5)+2 f (1)+   +4 f (4.5)+ f (5)
0 10 2 3
1
=
6
1
= (268.41)=44.735m
6

33. By the Net Change Theorem, the increase in velocity is equal to  a(t)dt . We use Simpson’s
6
0
Rule with n=6 and  t=(6 0)/6=1 to estimate this integral:
1
 0a(t)dt S6 = 3 [a(0)+4a(1)+2a(2)+4a(3)+2a(4)+4a(5)+a(6)]
6

1 1
 [0+4(0.5)+2(4.1)+4(9.8)+2(12.9)+4(9.5)+0]= (113.2)=37.73 ft/s
3 3

34. By the Net Change Theorem, the total amount of water that leaked out during the first six hours is
6 0
equal to  r(t)dt . We use Simpson’s Rule with n=6 and  t=
6
=1 to estimate this integral:
0 6

1

6
r(t)dt S = r(0)+4r(1)+2r(2)+4r(3)+2r(4)+4r(5)+r(6)
0 6 3
1
 4+4(3)+2(2.4)+4(1.9)+2(1.4)+4(1.1)+1
3
1
= (36.6)=12.2liters
3

The function values were obtained from a high resolution graph.

35. By the Net Change Theorem, the energy used is equal to  P(t)dt . We use Simpson’s Rule with
6
0
1
n=12 and  t=(6 0)/12= to estimate this integral:
2

17
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

1/2

6
P(t)dt S = [P(0)+4P(0.5)+2P(1)+4P(1.5)+2P(2)+4P(2.5)]
0 12 3
+2P(3)+4P(3.5)+2P(4)+4P(4.5)+2P(5)+4P(5.5)+P(6)]
1
= [1814+4(1735)+2(1686)+4(1646)+2(1637)+4(1609)+2(1604)]
6
+4(1611)+2(1621)+4(1666)+2(1745)+4(1886)+2052]
1
= (61,064)=10,177.3 megawatt hours.
6

36. By the Net Change Theorem, the total amount of data transmitted is equal to  D(t)dt 3600 . We
8
0
use Simpson’s Rule with n=8 and  t=(8 0)/8=1 to estimate this integral:
1
 0D(t)dt S8 = 3 [D(0)+4D(1)+2D(2)+4D(3)+2D(4)+4D(5)+2D(6)+4D(7)+D(8)]
8

1
 [0.35+4(0.32)+2(0.41)+4(0.50)+2(0.51)+4(0.56)+2(0.56)+4(0.83)+0.88]
3
1
= (13.03)=4.343
3
Now multiply by 3600 to obtain 15 , 636 megabits.

37. Let y= f (x) denote the curve. Using cylindrical shells, V = 

10 10
2 xf (x)dx=2 xf (x)dx=2 I .

.
2 2
Now use Simpson’s Rule to approximate I :
I S = 10 2 2 f (2)+4 3 f (3)+2 4 f (4)+4 5 f (5)+2 6 f (6)
8 3(8)
+4 7 f (7)+2 8 f (8)+4 9 f (9)+10 f (10)
1
 [2(0)+12(1.5)+8(1.9)+20(2.2)+12(3.0)+28(3.8)+16(4.0)+36(3.1)+10(0)]
3
1
= (395.2)
3
1
Thus, V  2  (395.2) 827.7 or 828 cubic units.
3

38.
18 0
Work =
18
f (x)dx S = f (0)+4 f (3)+2 f (6)+4 f (9)+2 f (12)+4 f (15)+ f (18)
0 6 6 3
=1 9.8+4(9.1)+2(8.5)+4(8.0)+2(7.7)+4(7.5)+7.4 =148 joules

18
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

2
39. Volume = 
2 3
0
( 1+x
3
) dx=  ( 1+x )2
0
3 2/3
dx . V    S
10
(
where f (x)= 1+x
3 2/3
) and
1

.
 x=(2 0)/10= . Therefore,
5
1
V    S = f (0)+4 f (0.2)+2 f (0.4)+4 f (0.6)+2 f (0.8)+4 f (1)
10 5 3
+2 f (1.2)+4 f (1.4)+2 f (1.6)+4 f (1.8)+ f (2)  12.325078

 /2 42 2
40. Using Simpson’s Rule with n=10 ,  x= , L=1 ,  = radians, g=9.8 m / s ,
10 0 180
2 2 1 2 2
k =sin  , and f (x)=1/ 1 k sin x , we get
2 0

L  /2 dx L
T =4 g
 0
 4 S
g 10
2 2
1 k sin x
1  /2  2 9 
=4 f (0)+4 f +2 f +   +4 f +f  2.07665
9.8 10 3 20 20 20 2

2 2
N sin k  Ndsin   4  9
41. I( )= , where k= , N=10 , 000 , d=10 , and  =632.8 10 . So
2 
k
 6
I( )=
( 4 2
10 ) sin k
2
, where k=
 ( 104) ( 10 4) sin  . Now n=10 and   =
10  (  10 6) =2  7
10 ,
2  9 10
k 632.8 10
 7
so M =2 10 [I( 0.0000009)+I( 0.0000007)+   +I(0.0000009)] 59.4 .
10

20 0
42. f (x)=cos ( x) ,  x= =2
10
2
T
10 ={ f (0)+2[ f (2)+ f (4)+   + f (18)]+ f (20)}
2
= 1[cos 0+2(cos 2 +cos 4 +   +cos 18 )+cos 20 ]
= 1+2(1+1+1+1+1+1+1+1+1)+1=20
1 20 1
The actual value is  cos ( x)dx=
20
sin  x = (sin 20  sin 0)=0 . The discrepancy is due to
0  0 
the fact that the function is sampled only at points of the form 2n , where its value is
f (2n)=cos (2n )=1 .

43. Consider the function f whose graph is shown. The area

19
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

2 0 1

2
f (x)dx is close to 2 . The Trapezoidal Rule gives T = f (0)+2 f (1)+ f (2) = 1+2 1+1 =2 .
0 2 2 2 2
2 0
The Midpoint Rule gives M = f (0.5)+ f (1.5) =1[0+0]=0 ,
2 2
so the Trapezoidal Rule is more accurate.

44. Consider the function f (x)= x 1 , 0

x
2 . The area 
2
f (x)dx is exactly 1 . So is the right
0
endpoint approximation: R = f (1) x+ f (2) x=0 1+1 1=1 . But Simpson’s Rule approximates f with
2
2  x 1 2
the parabola y=(x 1) , shown dashed, and S = f (0)+4 f (1)+ f (2) = 1+4 0+1 = .
2 3 3 3

45. Since the Trapezoidal and Midpoint approximations on the interval a,b are the sums of the
Trapezoidal and Midpoint approximations on the subintervals x ,x , i=1,2,... ,n , we can focus
i 1 i
/ /
our attention on one such interval. The condition f (x)<0 for a
x
b means that the graph of f is
concave down as in Figure 5. In that figure, T is the area of the trapezoid AQRD , 
b
f (x)dx is the
n a

area of the region AQPRD , and M is the area of the trapezoid ABCD , so T <
b
f (x)dx<M . In
n n a n
/ /
general, the condition f <0 implies that the graph of f on a,b lies above the chord joining the
points ( a,f(a) ) and ( b,f(b) ) . Thus, 
b
f (x)dx>T . Since M is the area under a tangent to the graph,
a n n

<0 implies that the tangent lies above the graph, we also have M >
/ / b
and since f f (x)dx . Thus,
n a

T <
b
f (x)dx<M .
n a n

3 2
46. Let f be a polynomial of degree
3 ; say f (x)=Ax +Bx +Cx+D . It will suffice to show that
Simpson’s estimate is exact when there are two subintervals ( n=2 ), because for a larger even number
20
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

of subintervals the sum of exact estimates is exact. As in the derivation of Simpson’s Rule, we can
assume that x = h , x =0 , and x =h . Then Simpson’s approximation is
0 1 2
1

h
f (x)dx h[ f ( h)+4 f (0)+ f (h)]
 h 3
1
=
3
3
( 2 3 2
h  Ah +Bh  Ch+D +4D+ Ah +Bh +Ch+D ) ( )
1 2 2 3
= h 2Bh +6D = Bh +2Dh
3 3
The exact value of the integral is

h
 h
( Ax +Bx +Cx+D) dx =
3 2
2
h
0
( Bx +D) dx [by Theorem 5.5. (a) and (b)]
2

1 3 h 2 3
= 2 Bx +Dx = Bh +2Dh
3 0 3
Thus, Simpson’s Rule is exact.

.
1
47. T =  x f x +2 f x +   +2 f x
( ) ( ) ( )+ f (x ) and
n 2 0 1 n 1 n

( ) ( )
M = x f x + f x +   + f x
n 1 2 ( ) + f ( x ) , where x = 12 ( x +x ) . Now
n 1 n i i 1 i
1 1
T
2n =  x f ( x ) +2 f ( x ) +2 f ( x ) +2 f ( x ) +2 f ( x ) +  
2 2 0 1 1 2 2

+2 f ( x ) +2 f ( x ) +2 f ( x ) + f ( x )
n 1 n 1 n n

1 1 1
so
2 (
T +M =
n n )
T + M
2 n 2 n
1
=  x f x +2 f x +   +2 f x
( ) ( ) +f x ( ) ( )
4 0 1 n 1 n
1
+  x 2 f x +2 f x +   +2 f x
4 1 2 ( ) ( )
n 1
+2 f x
n ( ) ( )
= T 2n
 x  x
f x +2 and M = x
n 1 n
48. T = ( ) ( ) ( )
f x +f x f x , so
n 2 0 i=1 i n n i=1 i 2

1 2 1  x  x
f x +2 f x + f x +4
n 1 n
T + M= T +2M = ( ) ( ) ( ) ( ) f x
3 n 3 n 3 n n 3 2 0 i=1 i n i=1 i 2

where

21
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.7 Approximate Integration

.
b a b a
 x= . Let  x= . Then  x=2 x , so
n 2n
1 2  x
f x +2 f x + f x +4
n 1 n
T + M = ( ) ( ) ( ) f x  x ( )
3 n 3 n 3 0 i=1 i n i=1 i
1
=  x f x +4 f x   x +2 f x +4 f x   x
( ) ( ) ( ) ( )
3 0 1 1 2

+2 f ( x ) +   +2 f ( x ) +4 f ( x   x ) + f ( x )
2 n 1 n n
b a
Since x ,x   x,x ,x   x,x ,... ,x ,x   x,x are the subinterval endpoints for S , and since  x=
0 1 1 2 2 n 1 n n 2n 2n
1 2
is the width of the subintervals for S , the last expression for T + M is the usual expression for
2n 3 n 3 n
1 2
S . Therefore, T + M =S .
2n 3 n 3 n 2n

22
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 4
4  x
1. (a) Since  xe dx has an infinite interval of integration, it is an improper integral of Type I.
1
 /2

(b) Since y=sec x has an infinite discontinuity at x=
2
,  sec xdx is a Type II improper integral.
0
2

(c) Since y=
x
(x 2)(x 3)
has an infinite discontinuity at x=2 ,  2
x
dx is a Type II improper
0 x  5x+6
integral.
0

(d) Since  1
2
dx has an infinite interval of integration, it is an improper integral of Type I.
  x +5

2. (a) Since y=1/(2x 1) is defined and continuous on 1,2 , the integral is proper.
1
1 1 1
(b) Since y= has an infinite discontinuity at x= ,  dx is a Type II improper integral.
2x 1 2 0 2x 1


(c) Since  sin x

2
dx has an infinite interval of integration, it is an improper integral of Type I.
  1+x
2
(d) Since y=ln (x 1) has an infinite discontinuity at x=1 ,  ln (x 1)dx is a Type II improper integral.
1

3  3
3. The area under the graph of y=1/x =x between x=1 and x=t is
t
 3 1 2 t 1 2 1 1
A(t)= x dx= 
1 2
x
1
=
2
t  
2
=  1 2t
2
2
. So the area for 1 x 10 is /( )
A(10)=0.5 0.005=0.495 , the area for 1 x 100 is A(100)=0.5 0.00005=0.49995 , and the area for
1 x 1000 is A(1000)=0.5 0.0000005=0.4999995 . The total area under the curve for x 1 is
1 1
lim A(t)=lim
t  t  2
 1 2t
2
= .
2 /( )

4. (a)

1
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

(b) The area under the graph of f from x=1 to x=t is

t t t
 1.1 1 
 f (x)dx= x
0.1
F(t) = dx=  x
1 1 0.1 1

=  10 t  0.1 1 =10 1 t  0.1( ) ( )

and the area under the graph of g is
t t t
 0.9 1 0.1
G(t) =  g(x)dx= x dx= x
1 1 0.1 1

= 10 t 0.1 1 ( )
t F(t) G(t)
10 2.06 2.59
100 3.69 5.85
4
10 6.02 15.12
6
10 7.49 29.81
10
10 9 90
20
10 9.9 990

 0.1
(c) The total area under the graph of f is lim F(t)=lim 10 1 t ( ) =10 .
t  t 

The total area under the graph of g does not exist, since lim G(t)=lim 10 t ( 0.1 1) = .
t  t 

 t

5. I=  1
2
dx=lim
t 
 1
2
dx . Now
1 (3x+1) 1 (3x+1)
1 1 1
 dx =  du [ u=3x+1 , du=3dx ]
2 3 2
(3x+1) u
2
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

1 1
= +C= +C ,
3u 3(3x+1)

1 t 1 1 1 1
so I=lim  =lim  + =0+ = . Convergent
t  3(3x+1) 1 t  3(3t+1) 12 12 12

0 0 0
1 1 1 1 1
6.  dx= lim  dx= lim ln 2x 5 =lim ln 5 ln 2t 5 =  .
  2x 5 t   t 2x 5 t   2 t t  2 2
Divergent

7.
 1  1
1 1  1
 dw = lim  dw= lim  2 2 w
t
  2 w t   t 2 w t  

= lim  2 3 +2 2 t = . Divergent
t  

8.
 t
1  1 t 1  1 1

x
x =lim  2 2
dx=lim
2 2
= lim
2 t  2
+
2
dx t  0 (x +2) t  x +2 0 t +2
2 2
0 (x +2)
1 1 1
= 0+ = . Convergent
2 2 4

 t
 y/2  y/2  y/2 t  t/2  2  2  2
9.  e dy=lim  e dy=lim  2e
4
=lim ( 2e +2e )=0+2e =2e .
4 t  4 t  t 

Convergent

 1  1  1
 2t  2t 1  1 2 1 
 
2t 2x
10. e dt= lim e dt= lim  e = lim  e+ e = . Divergent
  x   x x   2 x x   2 2
 0 
x x x
 1+x =  1+x + 
2 2 2
11. 1+x and
  dx   dx 0 dx
0 0
x 1 1
 dx
2
1+x = lim
2
ln 1+x
2
( ) t t
= lim 0
2
ln 1+t
2
( ) =  . Divergent
  t    

3
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 0 
12. I=  ( 2 v ) dv=I1+I2=  ( 2 v ) dv+  ( 2 v4) dv , but
4 4

    0
1 5 0 1 5
I = lim 2v v =  2t+ t =  . Since I is divergent, I is divergent, and there is no
1 5 t t   5 1
t  
need to evaluate I . Divergent
2

 2 0 2  2
 x  x  x
13.  xe dx=  xe dx+  xe dx .
    0
0 0
 xe
 x
2
dx= lim
 

1
2
e
 x
2

t
=lim


1
2
( e  1) = 12 1= 12 , and
 t
2

  t t
 t
 xe
 x
2
dx=lim
t 

1
2
e
 x
2

0
=lim
t 

1
2
( e  1) = 12 ( 1)= 12 .
 t
2

0
 2
 x 1 1
Therefore,  xe dx= + =0 . Convergent
2 2
 

 3 0 3  3
2  x 2  x 2  x
14.  xe dx=  x e dx+  x e dx , and
    0
0 3 3 0 3
2  x 1  1 1  t

x
xe dx= lim  e = + lim e = . Divergent
  t   3 t 3 3 t  


15.  
t t
sin  d =lim sin  d =lim  cos  =lim ( cos t+1) . This limit does not exist, so the
2 2 2
t  t  t 
integral is divergent. Divergent
 t 1 1 1 t 1 1
16.  cos  d =lim 
2
(1+cos 2 )d =lim  + sin 2 =lim t+ sin 2t = since
0 02 2 4 0 t 2 4
t  t  
1 1 1
sin 2t  for all t , but t  as t  . Divergent
4 4 2

17.
 t 1
(2x+2)
 x+1
2
dx =lim  2
2
dx=
1
lim
2 t 
2
ln (x +2x) =
t
1
1
lim
2 t 
2
ln (t +2t) ln 3
1 x +2x t  x +2x 1
= . Divergent

4
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

18.
 t
1 1 z+1 t
 2
dz =lim
t 

0

z+1 z+2
dz=lim
t 
ln
z+2 0
0 z +3z+2
t+1 1
=lim ln  ln =ln 1+ln 2=ln 2 . Convergent
t  t+2 2

19.
 t t
 5s  5s 1  1 
  by integration by
5s 5s
se ds =lim se ds=lim  se  e
0 t  0 t  5 25 0 parts with u=s
1  5t 1  5t 1 1
=lim  te  e + =0 0+ [ by l’Hospital’s Rule]
t  5 25 25 25
1
= . Convergent
25

20.
6 6
r/3 6
  by integration by
r/3 r/3 r/3
re dr = lim re dr= lim 3re  9e
  t   t t  
t parts with u=r
2 2 t/3 t/3 2
= lim (18e  9e  3te +9e )=9e  0+0 [ by l’Hospital’s Rule]
t  
2
=9e . Convergent

 2 t 2
ln x ( ln x ) ( ln t )
21.  dx=lim (by substitution with u=ln x , du=dx/x ) =lim = .
1 x t  2 1 t  2
Divergent

 0  0
  x x 0
22.  e
x
dx=  e dx+  e dx ,
x

x
e dx= lim e
t
= lim ( 1 et ) =1 , and
    0   t   t  
 
 x  x t  t 
 e dx=lim  e
0
=lim ( 1 e ) =1 . Therefore,  e
x
dx=1+1=2 . Convergent
0 t  t   

23.

5
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 0  

   
2 2 2 2
x x x x
dx= dx+ dx=2 dx [ since the integrand is even].
6 6 6 6
  9+x   9+x 0 9+x 0 9+x
2 3 1 1
x dx du (3dv)
Now 
u=x 3 3 1 dv
= = 
u=3v
6 2 =
9+x du=3x dx 2 du=3dv 2 9 2
9+u 9+9v 1+v
3
1  1 1  1 u 1  1 x
= tan v+C= tan +C= tan +C ,
9 9 3 9 3

 t

 
2 2 3 t
x x 1  1 x
so 2 dx =2lim dx=2lim tan
6 t  6 t  9 3 0
0 9+x 0 9+x
3
1  1 t 2  
=2lim tan = = . Convergent
t  9 3 9 2 9

24. Integrate by parts with u=ln x , dv=dx/x

3
du=dx/x , v= 1/ 2x ( 2) .
 t t


1
ln x
x
3
dx =lim
t 
 ln x
3
dx=lim
t 

1
2
ln x
t
+
1
2
 1
3
dx
x 2x 1 1 1 x
1 ln t 1 1 1
=lim  +0 + =
t  2 2 2 4 4
t 4t
ln t 1/t 1
since lim =lim =lim =0 . Convergent
t  2 t  2t t  2
t 2t

2
25. Integrate by parts with u=ln x , dv=dx/x
du=dx/x , v= 1/x .
 t

 ln x
t  2
dx
=
lim
t 
 ln x
2
dx=lim 
ln x 1
x

x
t

1 t
=lim


ln t 1
t

t
+0+1
x
1 x 1
=  0 0+0+1=1
ln t 1/t
since lim =lim =0 . Convergent
t  t t  1
6
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 t

26.  x arctanx
22
dx=lim
t 
 x arctanx
22
dx . Let u=arctanx , dv=
xdx
22
. Then du=
dx
2
,
0 (1+x ) 0 (1+x ) (1+x ) 1+x
1 2xdx  1/2
v=  = , and
2 22 2
(1+x ) 1+x
x arctanx x=tan  ,
 22
dx =
1 arctanx 1
+ 
dx
2
(1+x ) 2
1+x
2 2 22
(1+x ) dx=sec  d
2
1 arctanx 1 sec  d
= + 
2 2 2 2 2
1+x (sec  )
1 arctanx 1
+  cos  d
2
=
2 2 2
1+x
1 arctanx  sin  cos 
= + + +C
2 2 4 4
1+x
1 arctanx 1 1 x
= + arctanx+ +C
2 2 4 4 2
1+x 1+x

It follows that

1 arctanx 1 1 x t

 x arctanx
22
dx
=lim
t 

2
1+x
2
+ arctanx+
4 4
1+x
2
0
0 (1+x )
1 arctant 1 1 t 1  
=lim  + arctant+ =0+ +0= .
t  2 2 4 4 2 4 2 8
1+t 1+t

Convergent.

27. There is an infinite discontinuity at the left endpoint of 0,3 .

7
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

3 3
dx dx
 =lim 
3
=lim 2 x
t
= +
(2 3  2 t ) =2 3 . Convergent
0 x +t x + t 0
t 0 t 0

28. There is an infinite discontinuity at the left endpoint of 0,3 .

3


3

0
dx
x x
=lim
+
t 0
 t
dx

x
3/2
=lim
t 0
+
 2
x
3

t
=
 2
3
+lim
+
2

t
t
=
0
. Divergent

29. There is an infinite discontinuity at the right endpoint of  1,0 .

0 t

 dx
2
=lim

 dx
2
=lim

 1
x
t

 1
=lim


1 1
+
t  1
= . Divergent
 1 x t 0  1 x t 0 t 0

9
t
 dx 3 t 3 2/3 3
 3 2/3
30. =lim x 9 =lim (x 9) = 
(t 9)  (4) =0 6= 6 . Convergent
3
x 9  dx  2 1 t 9 2 2
1 t 9 1 t 9

0 3 0
3

  
 3 t
dx dx dx x 1 1

4
31. x= + , but =lim  =lim   = . Divergent
dx 4 4 4  3  2  3
 2  2 x 0 x  2 x t 0 t 0 3t 24

1 t

32.  0
dx

1 x
2
=lim
t

1
 0
dx

1 x
2
=lim
t

1
 1
sin x =lim sin t=
t
0
t 1

 1 
2
. Convergent

33 1 33
 1/5  1/5  1/5
33. There is an infinite discontinuity at x=1 .  (x 1) dx= (x 1) dx+  (x 1) dx . Here
0 0 1
1 t
 1/5  1/5 5 t 5 4/5 5 5
 
4/5
(x 1) dx=lim (x 1) dx=lim (x 1) =lim (t 1)  = and
0   4 0  4 4 4
t 1 0 t 1 t 1

8
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

33 33 33
 1/5  1/5 5 5 5
 
4/5 4/5
(x 1) dx=lim (x 1) dx=lim (x 1) =lim 16 (t 1) =20 . Thus,
1 + + 4 t + 4 4
t 1 t t 1 t 1
33
 1/5 5 75
 (x 1) dx=
4
+20=
4
. Convergent
0

1
34. f (y)=1/(4y 1) has an infinite discontinuity at y= .
4
1 1 1
1 1
lim 
1
 4y 1
dy =
+ t 4y 1
dy= lim
+ 4
ln 4y 1
t
1/4 t ( 1/4 ) t ( 1/4 )
1 1
= lim ln 3 ln (4t 1) =
+ 4 4
t ( 1/4 )
1 1
1 1
so  4y 1
dy diverges, and hence,  dy diverges. Divergent
1/4 0 4y 1

35.
  /2   /2 t
 sec xdx =  sec xdx+  sec xdx .  sec xdx= lim  sec xdx
0  /2 0 
0 t  /2 0
t
= lim ln sec x+tan x = lim ln sec t+tan t = . Divergent
 0 
t  /2 t  /2

4
4 2 4
36.  dx
=
dx
=
dx
+
dx
, and
0 (x+3)(x 2) 0 (x 2)(x+3) 2 (x 2)(x+3)
2
0 x +x 6
2 t t
1/5 1/5 1 x 2

dx
= lim  
x 2 x+3
dx [partial fractions =lim
5
ln
x+3
(x 2)(x+3)   0
0 t 2 0 t 2
1 t 2 2
= lim ln  ln =  .Divergent
 5 t+3 3
t 2

1 0 1
x x x
37. There is an infinite discontinuity at x=0 .  e
x
dx=  e
x
dx+  e
x
dx .
 1 e 1  1 e 1 0 e 1

9
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

0 t
x x
 e
x
dx=lim

 e
x
dx=lim

ln e  1
x t
 1
=lim

t
ln e  1  ln e  1
 1
=  ,
 1 e 1 t 0  1 e 1 t 0 t 0
1 1
x x
so  e
x
dx is divergent. The integral  e
x
dx also diverges since
 1 e 1 0 e 1
1 1
x x
 e
x
dx=lim  e
x
dx=lim ln e  1
x 1
t
=lim ln e 1  ln e  1
t
= .
e 1 0 t
+
0 t e 1 t
+
0 t
+
0
Divergent

2 3/2 2
x 3 x 3 x 3
38.  dx=  dx+  dx and
0 2x 3 0 2x 3 3/2 2x 3
x 3 1 2x 6 1 3 1 3
 2x 3 dx= 2  2x 3 dx= 2  1 2x 3 dx= 2 x 4 ln 2x 3 +C , so
3/2
x 3 1

t
dx= lim 2x 3ln 2x 3 = . Divergent
0 2x 3  4 0
t 3/2

39.
2 2 3 2
z
I = z ln zdz=lim 
2 2
z ln zdz=lim (3ln z 1)
+t + 2
0 t 0 t 0 3 t
8 1 3 8 8 1 3 8 8 1
=lim (3ln 2 1) t (3ln t 1) = ln 2  lim t (3ln t 1) = ln 2  L.
+ 9 9 3 9 9 + 3 9 9
t 0 t 0

Now L=lim t (3ln t 1) =lim

3 3ln t 1
 3
=lim
3/t
4
=lim (  t 3) =0 . Thus, L=0 and I= 83 ln 2 8
9
.
t
+
0 t
+
0 t t 0
+
 3/t t
+
0
Convergent

40. Integrate by parts with u=ln x , dv=dx / x

du=dx/x , v=2 x .
1 1 1
ln x
dx = lim  2 x ln x  2
ln x 1 1
 x
dx=lim
t dx
x =lim  2 t ln t 4 x
t
0 x t 0
+t
t 0
+ t t 0
+

=
lim ( 2 t ln t 4+4 t ) = 4
+
t 0
10
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

ln t 1/t
since lim t ln t=lim
 1/2
=lim  3/2
=lim ( 2 t ) =0 . Convergent
t 0
+
t 0
+
t t
+
0  t /2 t 0
+

41.

1
x 1
Area =  e dx= lim
x
e
t
  t  
t
=e lim e =e
t  

42.


 x/2  x/2 t
Area =  e dx= 2lim e
 2
 2 t 
 t/2
= 2lim e +2e=2e
t 

43.

 

Area =  2
2
dx=2 2  1
2
dx
  x +9 0 x +9
11
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 1 1  1 x t
=4lim dx=4lim tan
t 
2
t  3 3 0
0 x +9
4  1 t 4  2
= lim tan  0 = =
3 t  3 3 2 3

44.

 t

Area =  x
2
dx=lim  2
x
dx
0 x +9 t  0 x +9
1 t
=lim
2
2
ln x +9 ( ) 0
t 
1
= lim
2 t 
(2 )
ln t +9  ln 9 =

Infinite area

45.

 /2 t
=  sec xdx= lim 
2 2
Area sec xdx
0 
t ( /2) 0
t
= lim tan x = lim (tan t 0)
 0 
t ( /2) t ( /2)
=
Infinite area

12
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

46.

0 0
1 1
Area =  dx= lim  dx
 2 x+2 +t x+2
t  2
0
= lim 2 x+2
t
= lim (2 2  2 t+2 )
+ +
t  2 t  2

=2 2  0=2 2

47. (a)
t
t  g(x)dx
1
2 0.447453
5 0.577101
10 0.621306
100 0.668479
1000 0.672957
10 , 000 0.673407

2
sin x
g(x)= . It appears that the integral is convergent.
2
x



2
2 sin x 1 1
(b)  1 sin x 1
0 sin x 1
0  . Since dx is convergent
2 2 2
x x 1 x



2
sin x
(Equation 2 with p=2>1 ), dx is convergent by the Comparison Theorem.
2
1 x

(c)
13
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals


Since  f (x)dx is finite and the area under g(x) is less than the area under f (x) on any interval 1,t ,
1

 g(x)dx must be finite; that is, the integral is convergent.
1

48. (a)
t
t  g(x)dx
2
5 3.830327
10 6.801200
100 23.328769
1000 69.023361
10 , 000 208.124560

1
g(x)= . It appears that the integral is divergent.
x 1

1 1 1 1
(b) For x 2 , x > x  1
< . Since  dx is divergent (Equation 2 with p=  1 ),
x x 1 2 x 2

1
 dx is divergent by the Comparison Theorem.
2 x 1

(c)

Since  f (x)dx is infinite and the area under g(x) is greater than the area under f (x) on any interval
2

2,t ,  g(x)dx must be infinite; that is, the integral is divergent.
2


2
cos x 1 1 1
49. For x 1 ,  < . dx is convergent by Equation 2 with p=2>1 , so
2 2 2 2
1+x 1+x x 1 x

14
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals


2
cos x
dx is convergent by the Comparison Theorem.
2
1 1+x

 x 
2+e 2  x 1 1
50. For x 1 ,
x
> [ since e >0 ] > .
x x
 x
dx is divergent by Equation 2 with p=1 1 , so
1
  x
2+e
 x
dx is divergent by the Comparison Theorem.
1

2x 2x 1 1  2x
51. For x 1 , x+e >e >0
 =e on 1, ).
2x 2x
x+e e
 t 
 2x 1  1  2t 1  1   2x
 
2x 2 2
e dx=lim  e =lim  e + e = e . Therefore, e dx is convergent, and
1 t  2 1 t  2 2 2 1


2x
by the Comparison Theorem, x+e is also convergent.
1 dx

52. For x 1 , 0<

x
6
<
x
6
=
x
3
=
1
2
.  1
2
dx is convergent by Equation 2 with p=2>1 , so
1+x x x x 1 x


 1 1+x
x
6
dx is convergent by the Comparison Theorem.

 /2  /2
1 1  dx dx  /2
53.
x sin x

x
on 0,
2
since 0 sin x 1 .  x
=lim 
x
=lim ln x
t
.
0 + t +
t 0 t 0
 /2  /2
dx dx
 
+
But ln t   as t 0 , so is divergent, and by the Comparison Theorem, is also
0 x 0 x sin x
divergent.

 x
 x e 1
54. For 0 x 1 , e  1
 .
x x
15
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

1 1 1  x
1 1 e
 dx=lim  2 t ) =2 is convergent. Therefore, 
1
dx=lim 2 x
t
=lim ( 2 dx is
0 x +t x + + 0 x
t 0 t 0 t 0
convergent by the Comparison Theorem.

 1  1 t
dx dx dx dx dx
55.  = + =lim  +lim  . Now
0 x (1+x) 0 x (1+x) 1 x (1+x) +t x (1+x) t  1 x (1+x)
t 0
2udu
=
dx 2
 [ u= x ,x=u ,dx=2udu ]
x (1+x) u 1+u
2
( )
du  1  1
=2 =2tan u+C=2tan x +C ,
2
1+u

so
  1 1  1 t
dx =lim 2tan x +lim 2tan x
 + t
t 
1
0 x (1+x) t 0
  1  1    
=lim 2  2tan t +lim 2tan t 2 =  0+2  =
+ 4 t  4 2 2 2
t 0
.

 3  3 t

56.  2 x
dx
2
x 4
=  2 x
dx

x 4
2
+  3 x
dx

x 4
2
=lim
t 2
+
 t x
dx
2
x 4
+lim
t 
 3 x
dx
2
x 4
. Now

dx
 =
2sec  tan  d
[ x=2sec  , where 0  < /2 or    <3 /2 ]
x
2
x 4 2sec  2tan 
1 1  1 1
=  +C= sec x +C , so
2 2 2



1  1 1 3 1  1 1 t
dx =lim sec x +lim sec x
+ 2 2 t t  2 2 3
t 2
2
2 x x 4

16
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

1  1 3 1  1  1 3 
= sec  0+  sec =
2 2 2 2 2 2 4

1 1
57. If p=1 , then  
p 1
x =lim x=lim ln x = . Divergent.
0 dx +t dx + t
t 0 t 0
If p 1 , then
1 1

p

p
x =lim dx
x (note that the integral is not improper if p<0)
dx +
0 t 0 t
 p+1 1
x 1 1
=lim =lim 1
+  p+1 t + 1 p p 1
t 0 t 0 t

1 +
If p>1 , then p 1>0 , so   as t 0 , and the integral diverges.
p 1
t
1
1 1
If p<1 , then p 1<0 , so
1
p 1
 0 as t 0 and 
+
dx
p
x =
1 p
lim
+
( 1 t1 p) =
1 p
.
t 0 t 0
1
Thus, the integral converges if and only if p<1 , and in that case its value is .
1 p

 

58. Let u=ln x . Then du=dx/x

 dx
p
=  du
p
. By Example 4, this converges to
1
p 1
if p>1
e x ( ln x ) 1 u
and diverges otherwise.

59. First suppose p= 1 . Then

1 1 1 1
ln x ln x 1 1
 x ln xdx= x dx=lim  x dx=lim
p 2 2
(ln x) = lim (ln t) =  , so the integral diverges.
0 0 +t + 2 t 2 +
t 0 t 0 t 0
Now suppose p  1 . Then integration by parts gives
p+1 p p+1 p+1
x x x x
 
p
x ln xdx= ln x dx= ln x +C . If p< 1 , then p+1<0 , so
p+1 p+1 p+1 2
( p+1)
1 p+1 p+1 1
x x  1 1 1

p p+1
x ln xdx=lim ln x =  lim t ln t = .
+ p+1 2 2 p+1 + p+1
0 t 0 ( p+1) t ( p+1 ) t 0
If p> 1 , then p+1>0 and
17
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

1  1 1 ln t 1/( p+1)  1 1 1/t


p
x ln xdx =  lim =  lim
2 p+1  ( p+1 ) 2 p+1  ( p+2 )
0 ( p+1) t 0
+
t ( p+1) t 0
+
 ( p+1)t
 1 1 p+1  1
= + lim t =
2 2 + 2
( p+1) ( p+1) t 0 ( p+1)
1
Thus, the integral converges to  if p> 1 and diverges otherwise.
2
( p+1)

60. (a)
n=0 :
 t
n  x  x  x t
 x e dx =lim  e dx=lim  e
0
0 t  0 t 
 t
=lim  e +1 =0+1=1
t 

n=1 :
 t
n  x  x  x
 x e dx=lim  xe dx . To evaluate  xe dx , we’ll use integration by parts
0 t  0
 x  x
with u=x , dv=e dx
du=dx , v= e .
 x  x  x  x  x  x
So  xe dx= xe    e dx= xe  e +C=( x 1)e +C and
t
 x  x t
lim  xe dx =lim ( x 1)e
0
t  0 t 
 t  t  t
=lim ( t 1)e +1 =lim  te  e +1
t  t 
=0 0+1 [ use l’Hospital’s Rule] =1
n=2 :
 t
n  x 2  x 2  x
 x e dx=lim  x e dx . To evaluate  x e dx , we could use integration by parts
0 t  0
again or Formula 97. Thus,
t t
2  x 2  x t  x
lim  x e dx =lim  xe
0
+2lim  xe dx
t  0 t  t  0
=0+0+2(1) [ use l’Hospital’s Rule and the result for n=1 ] =2
18
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

n=3 :
 t t
n  x 3  x 3  x t 2  x
 x e dx =lim  x e dx=lim  xe
0
+3lim  x e dx
0 t  0 t  t  0
=0+0+3(2) [ use l’Hospital’s Rule and the result for n=2 ] =6


n  x
(b) For n=1 , 2 , and 3 , we have  x e dx=1 , 2 , and 6 . The values for the integral are equal to the
0

n  x
factorials for n , so we guess  x e dx=n! .
0
  t
k  x k+1  x k+1  x
(c) Suppose that  x e dx=k! for some positive integer k . Then  x e dx=lim  x e dx . To
0 0 t  0
k+1  x  x  x
evaluate  x
k+1 k
e dx , we use parts with u=x , dv=e dx
du=(k+1)x dx , v= e . So
k+1  x k+1  x k  x k+1  x k  x
 x e dx= x e    (k+1)x e dx= x e +(k+1) x e dx and
t t
k+1  x k+1  x t k  x
lim  x e dx =lim  x e +(k+1)lim
0
 x e dx
t  0 t  t  0
k+1  t
=lim  t e +0 +(k+1)k!=0+0+(k+1)!=(k+1)! ,
t 

so the formula holds for k+1 . By induction, the formula holds for all positive integers. (Since 0!=1 ,
the formula holds for n=0 , too.)

 0   t t
1 2 1 2
61. (a) I=  xdx=  xdx+  xdx , and  xdx=lim  xdx=lim x =lim t  0 = , so I is
    0 0 t  0 t  2 0 t  2
divergent.
t t t  t
1 2 1 2 1 2
(b)  xdx=
2
x

= t 
t 2 2
t =0 , so lim  xdx=0 . Therefore,  xdx lim  xdx .
 t t   t   t   t

 2
M 4 3/2 3 
k  ve
kv
62. Let k= so that v= dv . Let I denote the integral and use parts to integrate I .
2RT  0
2 2
2  kv 1  kv
Let  =v , d =ve dv
d =2vdv ,  = e :
2k

19
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

t t
I =lim 
t 
1 2  kv
2k
ve + 
0 k
1
2

0
ve
 kv
2
dv=
1
lim
2k t 
(t e ) + 1k lim
2  kt
2

t 

1 
2k
e
2
kv
0
1 1 1
= 0 (0 1)=
2k 2 2
2k 2k

4 3/2 1 2 2 2 2 RT 8RT
Thus, v= k = = = = .
 2 1/2 1/2  M  M
2k (k )  M/ ( 2RT )

 2 t t
1 1 1
63. Volume =   
2
dx= lim x = lim  = lim 1 = < .
1 x t  1 dx t  x 1 t  t


t
GMm GMm  1 t  1 1 GMm
64. Work =
2
dr=lim  2
dr=lim GMm
r R
=GMm
t  t
+
R
=
R
, where
R r t  R r t 
24 3 6
M= mass of Earth =5.98 10 kg, m= mass of satellite =10 kg, R= radius of Earth =6.37 10 m, and
 11 2
G= gravitational constant =6.67 10 N m / kg.
 11 24 3
6.67 10 5.98 10 10 10
Therefore, Work = 6.26 10 J.
6
6.37 10



GmM 1 1 GmM
65. Work =  F dr=lim dr=lim GmM  = . The initial kinetic energy
R t  2 t  R t R
R r
1 2 GmM 2GM
provides the work, so mv =
v= .
2 0 R 0 R

66. y(s)=  s
2r
2
r  s
2
x(r)dr and x(r)=
1
2
2
(R r)

20
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

R R

y(s) =
lim
t s
+
 t
r(R r)
2
r  s
2

2
dr=lim
t
+
s
 t
3
r  2Rr +R r
2
r  s
2

2
2
dr

R 3 R 2 R
r r r
 r  s  2R 
2 2 2 2 2 2 2
= lim r  s +R r  s
+ t dr t dr t dr
t s
2
= lim + I 1 2RI 2+R I 3 =L
t s
2 2 2 2 2 2 2 2
For I : Let u= r  s
u =r  s , r =u +s , 2r dr=2udu , so, omitting limits and constant of
1
integration,

I =
1 
( u2+s2) u du= ( u2+s2) du= 1 u3+s2u= 1 u ( u2+3s2)
u 3 3
1
=
3
2
r  s
2
(r 2 s2+3s2) = 13 2
r  s
2
(r 2+2s2)
2
r s 2 2 2 2
For I : Using Formula 44, I = r  s + ln r+ r  s .
2 2 2 2
1 du 1
du=2r dr . Then I = 
2 2 2 2
For I : Let u=r  s
= 2 u= r  s .
3 3 2 u 2
Thus,
2 R
1 r s
L = lim
+ 3
2
r  s
2
(r +2s )  2R
2 2
2
2 2 2 2
r  s + ln r+ r  s
2
+R
2 2
r  s
2
t
t s
2
1 R s
= lim
+ 3
R s
2 2
( R +2s )  2R
2 2
2
2 2 2 2
R  s + ln R+ R  s
2
+R
2
R s
2 2

t s
2
1 t s
 lim
+ 3
2
t  s
2
( t +2s )  2R
2 2
2
2 2 2 2
t  s + ln t+ t  s
2
2
+R
2
t  s
2

t s
1
=
3
2
R s
2
( R2+2s2)  Rs2ln R+ R  s
2 2

2
 Rs ln s

2 2
= 1 R+ R  s
3
R s
2 2
( R +2s )  Rs ln
2 2 2
s
21
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

67. (a)

/
(b) r(t)=F (t) is the rate at which the fraction F(t) of burnt out bulbs increases as t increases. This
could be interpreted as a fractional burnout rate.

(c)  r(t)dt=lim F(x)=1 , since all of the bulbs will eventually burn out.
0 x 

68.

1 s
I =  te dt=lim
kt kt
2
( kt 1 ) e [ Formula 96, or parts]
0 s  k 0
1 ks 1 ks 1
=lim se  e   .
s  k 2 2
k k

Since k<0 the first two terms approach 0 (you can verify that the first term does so with l’Hospital’s
Rule), so the limit is equal to 1/k . Thus, M= kI= k 1/k = 1/k= 1/( 0.000121) 8264.5 years.
2
( 2)
 t

69. I=  1
2
dx=lim  1
2
dx=lim
 1
tan x =lim
t
a
( tan  1t tan  1a) =  2  tan  1a . I<0.001

a x +1 t  a x +1 t  t 
  1  1  
 tan a<0.001
tan a>  0.001
a>tan  0.001 1000 .
2 2 2
2
 x 4 0 1
70. f (x)=e and  x= = .
8 2
1

4
f (x)dx S =
8 [ f (0)+4 f (0.5)+2 f (1)+ +2 f (3)+4 f (3.5)+ f (4)]
0 2 3
1
(5.31717808) 0.8862
6
2 2
 x  4x   x   4x
Now x>4
 x x< x 4
e <e
 4
e dx<
4
e dx .

22
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

  4x 1  t 1 
 4
e dx=lim 
4
e
4x
4
=
4
0 e( 16
) =1 / ( 4e16) 0.0000000281<0.0000001 , as desired.
t 

   st n  sn
 st  st e e 1 1
71. (a) F(s)=  f (t)e dt=  e dt=lim  =lim + . This converges to only
0 0 n  s 0 n   s s s
1
if s>0 . Therefore F(s)= with domain { s|s>0} .
s
(b)
  n n
 st t  st t ( 1 s ) 1 t ( 1 s )
F(s) =  f (t)e dt=  e e dt=lim  e dt=lim e
0 0 n  0 n  1 s 0

( 1 s ) n
e 1
=lim 
n  1 s 1 s

1
This converges only if 1 s<0
s>1 , in which case F(s)= with domain { s|s>1} .
s 1
 n  st
 st  st  st e
(c) F(s)=  f (t)e dt=lim  te dt . Use integration by parts: let u=t , dv=e dt
du=dt , v= .
0 n  0 s
t  st 1  st n  n 1 1 1
Then F(s)=lim  e  e =lim  +0+ = only if s>0 . Therefore,
2 sn 2 sn 2 2
n  s s 0 n  se se s s
1
F(s)= and the domain of F is { s|s>0} .
2
s

at  st at  st
72. 0 f (t) Me
0 f (t)e  Me e for t 0 . Now use the Comparison Theorem:

 n n
at  st t ( a s ) 1 t ( a s ) 1 n ( a s )
 Me e dt=lim M  e dt=M lim
a s
e
0
=M lim
a s
e  1
0 n  0 n  n 

This is convergent only when a s<0

s>a . Therefore, by the Comparison Theorem,

 st
F(s)=  f ( t ) e dt is also convergent for s>a .
0


 st  st  st
73. G(s)=  f (t)e dt . Integrate by parts with u=e
/ /
, dv= f (t)dt
du= se , v= f (t) :
0

23
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals


 st n  st  sn
G(s)=lim f (t)e +s  f (t)e dt=lim f (n)e  f (0)+sF(s)
0
n  0 n 

at  st at  st t ( a s )
But 0 f (t) Me
0 f (t)e  Me e and lim Me =0 for s>a . So by the Squeeze Theorem,
t 
 st
lim f (t)e =0 for s>a
G(s)=0 f (0)+sF(s)=sF(s) f (0) for s>a .f}(0) for s>a .
t 

74. Assume without loss of generality that a<b . Then

a  a u
 f (x)dx+  f (x)dx = lim  f (x)dx+lim  f (x)dx
  a t   t u  a
a b u
= lim  f (x)dx+lim  f (x)dx+ f (x)dx
t   t u  a b
a b u
= lim  f (x)dx+ f (x)dx+lim  f (x)dx
t   t a u  b
a b 
= lim  f (x)dx+ f (x)dx +  f (x)dx
t   t a b
b 
= lim  f (x)dx+  f (x)dx
t   t b
b 
=  f (x)dx+  f (x)dx
  b

2 2
 x 1  x
75. We use integration by parts: let u=x , dv=xe dx
du=dx , v= e . So
2
 2 2 t  2
2  x 1  1  x
 
x
xe dx = lim  xe + e dx
0 t  2 0 2 0
 
 t / ( 2e )
2 2 2
1  x 1  x
 
t
= lim + e dx= e dx
t  2 0 2 0
(The limit is 0 by l’Hospital’s Rule.)

76.

24
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

 2 2 2
 x  x  x
 e dx is the area under the curve y=e for 0 x< and 0<y 1 . Solving y=e for x , we get
0
2
 x 2 2
y=e
ln y= x
 ln y=x
x=  ln y . Since x is positive, choose x=  ln y , and the area is
1
represented by   ln y dy . Therefore, each integral represents the same area, so the integrals are
0
equal.

2
77. For the first part of the integral, let x=2tan 
dx=2sec  d .
2
1 2sec  x
 dx=
2sec 
d = sec  d =ln sec  +tan  . From the figure, tan  =
2
, and
2
x +4
2
x +4
sec  = . So
2


t
2
1 C = lim x +4 x
I=  dx ln +  Cln x+2
2 x+2 t  2 2 0
0 x +4
2
= lim t +4 +t
ln  Cln (t+2) ( ln 1 Cln 2 )
t  2
2
t +4 +t C
= lim ln +ln 2
C
t  2 ( t+2 )
2
t+ t +4 C 1
= ln lim +ln 2
C
t  ( t+2 )
2 2
t+ t +4 1+t/ t +4 2
Now L=lim =lim = .
C C 1 C 1
t  ( t+2 ) t  C ( t+2 ) Clim ( t+2 )
t 
If C<1 , L= and I diverges. If C=1 , L=2 and I converges to
25
Stewart Calculus ET 5e 0534393217;7. Techniques of Integration; 7.8 Improper Integrals

0
ln 2+ln 2 =ln 2 . If C>1 , L=0 and I diverges to   .

78.

1 1 t
I=  x

C
dx
=lim
2
2
(
ln x +1 
3
)
Cln (3x+1)
0
2 3x+1 t 
0 x +1

=lim ( 2 ) 1/2 ln (3t+1)C/3

ln t +1
t 

=lim ln
( t2+1) 1/2 =ln lim
2
t +1
C/3 C/3
t  ( 3t+1 ) t  ( 3t+1 )

For C 0 , the integral diverges. For C>0 , we have

L=lim
2
t +1
C/3 t
=lim
t / 2
t +1
(C/3) 1
=
1
lim
C t 
1
( C/3 )  1
.
t  (3t+1)  C ( 3t+1 ) ( 3t+1 )
1 1
For C/3<1 C<3 , L= and I diverges. For C=3 , L= and I=ln . For C>3 , L=0 and I diverges
3 3
to   .

26
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

1 1
L=  1+({dy /dx) italic{dx}= 
2 2
1. y=2 3x 1+( 3) dx= 10 1 ( 2) =3 10 .
 2  2
The arc length can be calculated using the distance formula, since the curve is a line segment, so

2 2
L=[ distance from ( 2,8) to (1, 1) ]= [1 ( 2)] +[ (  1 )  8] = 90 =3 10

2 dy x
2. Using the arc length formula with y= 4 x  = , we get
dx 2
4 x
2 2 t

L=

2

0
1+
dy
dx
2
dx=  0
1+
x
4 x
2

2
dx=
4 x
 0t
2dx
2
=2lim

2
 0
dx
2
2 x
2

 1 t  1  1 
= 2lim  sin (x/2) 0=2lim  sin (t/2) sin 0 =2  0 =
2
t 2 t 2
1
The curve is a quarter of a circle with radius 2 , so the length of the arc is (2  2)= , as above.
4

3.

From the figure, the length of the curve is slightly larger than the hypotenuse of the triangle formed
2 2 3/2
by the points ( 1,0 ) , ( 3,0 ) , and ( 3,f(3) )  ( 3,15) , where y= f (x)=
3
x 1 . This length is about ( )
2 2 3/2 1/2
2 2
15 +2  15 , so we might estimate the length to be 15.5 . y=
3
x 1
/ 2
 y = x  1 (2x) ( ) ( )

1+(y ) =1+4x
/ 2 2
( x2 1) =4x4 4x2+1= ( 2x2 1) 2 , so, using the fact that 2x2 1>0 for 1 x 3 ,
3 3 3 3
2 3
 (2x  1) dx= 2x  1 dx= (2x  1)dx=
2 2 2 2
L= x x
1 1 1 3 1

2 46
= (18 3)  1 = =15.3
3 3

1
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

4.

From the figure, the length of the curve is slightly larger than the hypotenuse of the triangle formed
2 3
by the points ( 0.5,f(0.5) 1 ) , ( 1,f(0.5) 1 ) and 1, , where y= f (x)=x /6+1/(2x) . This length is
3
1 2 1 2
about +  0.6 , so we might estimate the length to be 0.65 .
2 3

3 2  2
x 1 x x
/
y= +  y =  
6 2x 2 2
4  4 4  4 2  2 2
/ 2 x 1 x x 1 x x x
1+ y ( ) =1+  +
4 2 4
= + +
4 2 4
=
2
+
2

so, using the fact that the parenthetical expression is positive,

1
 2 2  2 1

2 1 2 3
x x x x x 1
L= + dx=  + dx= 
2 2 1/2 2 2 6 2x 1/2
1/2
1 1 1 31
=    1 = =0.64583
6 2 48 48

3/2 1/2 2
5. y=1+6x  dy/dx=9x  1+(dy/dx) =1+81x . So
1 82
1
L = 1+81x dx=  u
1/2
du [ where u=1+81x and du=81dx ]
0 1 81
1 2 3/2 82 2
= 
81 3
u
1
=
243
( 82 82  1 )

2 3 3/2 1/2
6. y =4(x+4) , y>0 y=2(x+4)  dy/dx=3(x+4) 
2
1+(dy/dx) =1+9(x+4)=9x+37 . So

2
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

2 55
1
L = u=9x+37,
= u
1/2
9x+37 dx du
0
du=9dx 37 9
1 2 3/2 55 2
= 
9 3
u
37
=
27
( 55 55  37 37 )

5
x 1 dy 5 4 3  4
7. y= +  = x x 
6 3 dx 6 10
10x
2 25 8 1 9  8 25 8 1 9 8 5 4 3  4 2
1+ ( dy/dx ) =1+ x + x = x+ + x = x+ x . So
36 2 100 36 2 100 6 10
2 2
5 4 3  2 5 4 3  1 5 1  2
L = dx=
4 4 3
x+ x x+ x dx= x x
1 6 10 1 6 10 6 10 1

32 1 1 1 31 7 1261
=    = + =
6 80 6 10 6 80 240

2 2
x ln x dy 1 dy 2 1 1
8. y=   =x  1+ =x + + . So
2 4 dx 4x dx 2 2
16x
4 2 4
1 x ln x 2ln 2 ln 2 ln 2
L= x+ dx= + = 8+  2+ =6+ .
2 4x 2 4 2 4 4 4

1 1 3/2 1/2 1 1/2 1  1/2

9. x= y  y  dx/dy= y 
y ( y 3) = y 
3 3 2 2
2 1 1 1 1 1 1 1 1 1 1/2 1  1/2 2
1+ ( dx/dy ) =1+ y + y = y+ + y = y + y . So
4 2 4 4 2 4 2 2
9 9
1 1/2 1  1 2 3/2 1 2 2
L =
1/2 1/2
y + y dy= y +2y =  27+2 3   1+2 1
1 2 2 2 3 1 2 3 3
1 8 1 64 32
= 24 = =
2 3 2 3 3

2 2 2
10. y=ln (cos x) dy/dx= tan x 1+(dy/dx) =1+tan x=sec x . So
 /3  /3
 /3
L=  sec x dx=  sec xdx= ln sec x+tan x
2
=ln ( 2+ 3 )  ln (1+0)=ln ( 2+ 3 ) .
0
0 0

11. y=ln (sec x)

3
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

dy sec xtan x dy 2 2 2
= =tan x 1+ =1+tan x=sec x , so
dx sec x dx
 /4  /4  /4
 /4
L = sec x dx=  sec x dx=  sec xdx= ln (sec x+tan x)
2
0
0 0 0
=ln ( 2 +1 )  ln (1+0)=ln ( 2 +1 )

2 3 2
2 2
. So L= 
dy 1 dy 1 1+x 1+x
12. y=ln x =  1+ = 1+ = dx . Now let
dx x dx x x 1
x
2 2 2
v= 1+x , so v =1+x and vdv=xdx. Thus
2 2 2
v 1/2 1/2 1 1
L=  2
vdv=  1+ 
v 1 v+1
dv= v+ ln v 1 
2 2
ln v+1
2
2 v 1 2

1 v+1 2 1 1 2 +1 1
= v ln =2 ln 3 2+ ln =2 2 +ln ( 2 +1 )  ln 3
2 v 1 2 2 2 2 1 2
Or: Use Formula 23 in the table of integrals.

13. y=cosh x
/
y =sinh x 1+ y ( / ) 2=1+sinh 2x=cosh 2x .
1
1
So L= cosh xdx= sinh x =sinh 1=
1
(e 1/e) .
0
0 2

2 1 2 dx 1 dx 2 1 2
14. y =4x , x= y = y 1+ =1+ y . So
4 dy 2 dy 4

1 2
L= y dy=
2 1 2
1+ 1+u  2du
0 4 0
1
2 2
= u 1+u +ln u+ 1+u = 2 +ln ( 1+ 2 )
0

15. y=e 
x /
y =e 
x
1+ y ( / ) 2=1+e2x . So
1 e
du
 dx=
2x 2 x
L= 1+e 1+u u=e ,sox=ln u,dx=du/u
0 1 u

4
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

e 2


2 1+e
1+u v

= 2 2 2
udu= 2
vdv v= 1+u ,sov =1+u ,vdv=udu
2
1 u 2 v 1
2
1+e 2
1/2 1/2 1 v 1 1+e
=  1+ 
v 1 v+1
dv= v+ ln
2 v+1 2
2

2
1 2 1+e  1 1 2 1
= 1+e + ln  2 ln
2 2 2 2 +1
1+e +1
= 1+e 
2
2 +ln ( 2
1+e  1  1 ln ) ( 2  1)

Or: Use Formula 23 for  ( 2

)
1+u /u du , or substitute u=tan  .

x x x x
e +1 e e  2e
16. y=ln x
( x
=ln e +1  ln e  1  ) ( x
) y =
/
x
 x
= 2x

e 1 e +1 e 1 e  1
2  x
1+ ( y ) =1+
/ 2
=
( e +1)  2x
4e
2x
1+ y( ) / 2
=
2x
e +1
=
e +e
x
=
cosh x
.
2x  x sinh x
( e2x 1) 2 ( e2x 1) 2 e  1 e e
x

b b  b
cosh x sinh b e e
So L=
b
dx= ln sinh x =ln sinh b ln sinh a=ln =ln .
 a
a sinh x
a sinh a a
e e

2
1+(dy/dx) =1+sin x . So L= 
2 2 2
17. y=cos x dy/dx= sin x 1+sin x dx .
0

.
18. y=2 
x
dy/dx= 2 ln 2 ( ) x
L= 1+ ( ln 2 ) 2 dx
2 2x

4
1+(dx/dy) =1+(1+3y ) =9y +6y +2 . So L=
3 2 2 22 4 2 4 2
19. x=y+y  dx/dy=1+3y  9y +6y +2 dy .
1
2 2
b
20.
x
a
2
+
y
2
b
=1 , y=
b 1 x
2
/ a =

2
a
2
a x .
2

5
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

2 2
b 2 2 dy  bx dy 2 b x
y= a x  =  = .
a dx
a a x
2 2 dx
a
2
( a2 x2)
a a

 
1/2 1/2
So L=2 1+
2 2
b x
dx=
4 ( b2 a2) x2+a4 dx .
 a a
2
(a  x )
2 2 a
0 a x
2 2

 x  x  x  x 2  2x 2
21. y=xe  dy/dx=e  xe =e (1 x) 1+(dy/dx) =1+e (1 x) . Let
5
 2x 5 0 1
(1 x) . Then L= f (x)dx . Since n=10 ,  x=
2 2
f (x)= 1+(dy/dx) = 1+e = . Now
0 10 2

L S 1/2 1 3 5
10 = [ f (0)+4 f +2 f (1)+4 f +2 f (2)+4 f +2 f (3)]
3 2 2 2
7 9
+4 f +2 f (4)+4 f + f (5)] 5.115840
2 2

The value of the integral produced by a calculator is 5.113568 (to six decimal places).

1 2 1 2 1 1 2
22. x=y+ y  dx/dy=1+  1+ ( dx/dy ) =1+ 1+ =2+ + . Let g(y)= 1+(dx/dy)
2 y 2 y y 4y
2
2 1 1
. Then L= g(y)dy . Since n=10 ,  y= = . Now
1 10 10

L S 1/10
10 = [g(1)+4g(1.1)+2g(1.2)+4g(1.3)+2g(1.4)+4g(1.5)]
3
+2g(1.6)+4g(1.7)+2g(1.8)+4g(1.9)+g(2)] 1.732215 ,

which is the same value of the integral produced by a calculator to six decimal places.

 /3
L=  f (x)dx , where f (x)= 1+sec xtan x .
2 2
23. y=sec x dy/dx=sec xtan x
0
 /3 0 
Since n=10 ,  x= = . Now
10 30
L S =  /30 f (0)+4 f 
+2 f
2
+4 f
3
+2 f
4
+4 f
5
10 3 30 30 30 30 30
6 7 8 9 
+2 f +4 f +2 f +4 f +f  1.569619 .
30 30 30 30 3

6
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

The value of the integral produced by a calculator is 1.569259 (to six decimal places).

2 2
24. y=xln x dy/dx=1+ln x . Let f (x)= 1+(dy/dx) = 1+(1+ln x) .
3
3 1 1
Then L= f (x)dx . Since n=10 ,  x= = . Now
1 10 5

L S 1/5
10 = [ f (1)+4 f (1.2)+2 f (1.4)+4 f (1.6)+2 f (1.8)+4 f (2)]
3
+2 f (2.2)+4 f (2.4)+2 f (2.6)+4 f (2.8)+ f (3)] 3.869618 .
The value of the integral produced by a calculator is 3.869617 (to six decimal places).

25. (a)

(b)
3
Let f (x)=y=x 4 x . The polygon with one side is just the line segment joining the points
( 0,f(0) ) = ( 0,0 ) and ( 4,f(4) ) = ( 4,0 ) , and its length is 4 . The polygon with two sides joins the points
( 0,0 ) , ( 2,f(2) ) = ( 2,2 2 ) and ( 4,0 ) .
3

Its length is

2
(2 0) + 2 ( 3
2 0 )2 + 2
(4 2) + 0 2( 3
2 ) 2 =2 4+2
8/3
 6.43

Similarly, the inscribed polygon with four sides joins the points ( 0,0 ) , 1, 3 , 2,2 2 , ( 3,3) , ( 3
) ( 3
)
and ( 4,0 ) , so its length is

1+ ( 3 3 )2 + 1+ 2( 3
2
3
3 )2 + (
1+ 3 2
3
2 )2 + 1+9 7.50
7
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

dy 1  2/3 3 12 4x
(c) Using the arc length formula with =x (4 x) ( 1) + 4 x = , the length of the
dx 3 2/3
3(4 x)
4

curve is L=
0
4
1+
dy
dx
2

dx=
3(4 x)
0
1+
12 4x
2/3
2
dx .

(d) According to a CAS, the length of the curve is L 7.7988 . The actual value is larger than any of
the approximations in part (b). This is always true, since any approximating straight line between two
points on the curve is shorter than the length of the curve between the two points.

26. (a) Let f (x)=y=x+sin x with 0 x 2 .

(b) The polygon with one side is just the line segment joining the points ( 0,f(0) ) = ( 0,0 ) and
2 2
(2  0) +(2  0) =2 2   8.9 .
( 2 ,f(2 ) ) = ( 2 ,2 ) , and its length is
The polygon with two sides joins the points ( 0,0 ) , ( ,f( ) ) = ( , ) , and ( 2 ,2 ) . Its length is

2 2 2 2
(  0) +(  0) + (2   ) +(2   ) = 2  + 2  =2 2   8.9

Note from the diagram that the two approximations are the same because the sides of the 2  sided
1
polygon are in fact on the same line, since f ( )= = f (2 ) .
2

  3 3
The four sided polygon joins the points ( 0,0 ) , , +1 , ( , ) , ,  1 , and
2 2 2 2
( 2 ,2 ) , so its length is

8
 Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

 2  2  2  2  2  2  2  2
+ +1 + +  1 + +  1 + + +1  9.4
2 2 2 2 2 2 2 2

(c) Using the arc length formula with dy/dx=1+cos x , the length of the curve is

2 2
L=  1+(1+cos x) dx= 
2 2
2+2cos x+cos x dx
0 0

(d) The CAS approximates the integral as 9.5076 . The actual length is larger than the approximations
in part (b).

2 2
27. x=ln 1 y  ( 2
) dx  2y
=  1+
dx 2
=1+ =
( 1+y )
2
4y
. So
dy 2 dy
1 y ( 1 y ) ( 1 y2) 2
2 2

1/2

L=  0
( 1+y )
( 1 y2) 2
2 2
dy= 
1/2

0
1+y
1 y
2

2
dy=ln 3
1
2
 0.599

4/3 4 1/3 2 16 2/3

28. y=x  dy/dx= x  1+(dy/dx) =1+ x 
3 9

16 2/3 81 2 4 1/3 4  9 2/3 9 9 2 81 2

L= x dx=
1 4/3 2 2/3
1+ 1+u u du u= x ,du= x dx,dx= x du=  u du= u du
0 9 0 64 3 9 4 4 16 64
4/3
=
81
64
1
8
u 1+2u
2
( 2 1
1+u 
8
ln u+ 1+u )
2
( ) 0
81 1 32 25 1 4 25 81 1 41 5 1
= 1+  ln + =    ln 3
64 6 9 9 8 3 9 64 6 9 3 8
205 81
=  ln 3 1.4277586
128 512

2/3 3/2 dy 3 2   1/3

2/3
29. y =1 x 
2/3
(
y= 1 x )  =
dx 2
1 x
2/3
( ) 1/2 
3
x
1/3
= x ( 1 x2/3) 1/2 
9
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

2  2/3
dy
dx
=x ( 1 x2/3) =x 2/3 1 . Thus
1 1 1
 2/3  1/3 3 2/3
L=4 1+ x ( )
 1 dx=4 x dx=4lim
+ 2
x
t
=6 .
0 0 t 0

30. (a)

2 2 1
dy 2  4  4 
. So L=
2/3 1/3 2/3 2/3
(b) y=x  1+ =1+ x =1+ x 1+ x dx [ an improper
dx 3 9 0 9
2 2 1
dx 3 1/2 9 9
=1+ y . So L=
3/2
integral]. x=y  1+ =1+ y 1+ y dy .
dy 2 4 0 4
1
4 2 9 3/2 8 13 13 13 13  8
The second integral equals  1+ y =  1 = . The first
9 3 4 0 27 8 27
integral can be evaluated as follows:
1


1 2/3 9
4 
 u+4 
2/3 9x +4

2/3 1/3
1+ x dx = lim dx=lim du u=9x ,du=6x dx
0 9 + 1/3 + 2/3 18
t 0 t 3x t 0 9t
9
u+4 1 2 9 1 3/2 3/2 13 13  8

3/2
= du=  (u+4) = (13  4 )=
0 18 18 3 0 27 27

(c)
L = length of the arc of this curve from (  1,1 ) to ( 8,4 )
1 4
9 9 13 13  8 8 9 3/2 4
= 1+ y dy+ 1+ y dy= + 1+ y
0 4 0 4 27 27 4 0

10
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

13 13  8 8 13 13 +80 10  16
=
27
+
27
( 10 10  1 ) =
27

31. y=2x
3/2

/
y =3x
1/2
 1+ y( / ) 2=1+9x . The arc length function with starting point P0 ( 1,2 ) is
x x
2 2
s ( x ) =
3/2 3/2
1+9t dt= (1+9t) = (1+9x)  10 10
1 27 1 27

32. (a)

dy 2 1
4 1
(b) 1+ =x + + ,
dx 2 4
16x
x
s ( x) = 
2
t +1/ 4t ( 2) dt
1

1 3 x
= t  1/(4t)
3 1
1 3 1 1
= x  1/(4x) 
3 3 4
1 3 1
= x  1/(4x) for x 1
3 12

(c)

1 2 2
33. The prey hits the ground when y=0 180 x =0 x =45 180 x= 8100 =90 , since x must be
45
2
positive. y =
/
45
x 1+ y ( / ) 2=1+ 4
2
2
x , so the distance traveled by the prey is
45

11
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

4 45
L= x dx=
90 2 4 2
1+ 1+u du
0 2 0 2
45
4
=
45
2
1
2
u
1
1+u + ln u+ 1+u
2
2
0
2
( )
45 1 45
= 2 17+ ln ( 4+ 17 ) =45 17+ ln ( 4+ 17 )  209.1m
2 2 4

1 1
34. y=150
40
2
(x 50)  y =
/
20
(x 50) 1+ y ( / ) 2=1+ 1
2
2
(x 50) , so the distance traveled by the
20
kite is

80

L=  0
1+
20
1
2
(x 50) dx=
2
3/2

 5/2
2
1+u (20du) [u=
1
20
(x 50),du=
1
20
dx]

3/2
=20
1
2
u
1
1+u + ln u+ 1+u
2
2
 5/2
2
( )
13 3 3 13 5 29 5 29
=10 +ln + +  ln  +
4 2 2 4 2 4 2 4
15 25 3+ 13
= 13 + 29 +10ln  122.8ft
2 2  5+ 29

35. The sine wave has amplitude 1 and period 14 , since it goes through two periods in a distance of
2 
28 in., so its equation is y=1sin x =sin x . The width w of the flat metal sheet needed
14 7
to make the panel is the arc length of the sine curve from x=0 to x=28 . We set up the integral to
dy  
evaluate w using the arc length formula with = cos x :
dx 7 7
28 14
  2   2
L=  1+ cos x dx=2  1+
dx . This integral would be very cos x
0 7 7 0 7 7
difficult to evaluate exactly, so we use a CAS, and find that L 29.36 inches.

x x x x
36. (a) y=c+acosh
a

/
y =sinh
a
 ( / ) 2=1+sinh 2
1+ y
a
=cosh
2
a
. So

12
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

b b b
x x x b
L=  dx=2 cosh
2
cosh dx=2 asinh =2asinh
 b a 0 a a 0 a

(b) At x=0 , y=c+a , so c+a=20 . The poles are 50 ft apart, so b=25 , and L=51 51=2asinh (b/a) .
From the figure, we see that y=51 intersects y=2xsinh (25/x) at x 72.3843 for x>0 . So a 72.3843
and the wire should be attached at a distance of y=c+acosh (25/a)=20 a+acosh (25/a) 24.36 ft
above the ground.

x 2
37. y=
3
t  1 dt
dy
dx
3
= x  1 [ by FTC1]  1+
dy
dx
2
=1+ ( 3
x 1 ) =x 3

4 4
2 5/2 4 2 62
L= x dx= x
3 3/2
dx= x = (32 1)= =12.4
1 1 5 1 5 5

38. By symmetry, the length of the curve in each quadrant is the same, so we’ll find the length in the
2k 1/ ( 2k )
2k 2k 2k 2k
first quadrant and multiply by 4 . x +y =1 y =1 x  y= 1 x ( )
(in the first quadrant), so
we use the arc length formula with
dy 1 2k 1/ ( 2k )  1
dx
=
2k
1 x( )  2kx
2k 1
( )
=  x2k 1 1 x2k 1/ ( 2k )  1
( )
The total length is therefore

13
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.1 Arc Legth

1 1
2k 1/ ( 2k )  1 2
L =4
2k
1+  x
2k 1
( 1 x ) dx=4 1+x
2 ( 2k 1 )
( 1 x2k ) 1/k 2 dx Now from the graph, we
0 0
see that as k increases, the ‘‘corners’’ of these fat circles get closer to the points (
1,
1 ) and
(
1, 1 ) , and the ‘‘edges’’ of the fat circles approach the lines joining these four points. It seems
plausible that as k  , the total length of the fat circle with n=2k will approach the length of the
perimeter of the square with sides of length 2 . This is supported by taking the limit as k  of the
2k 1/ ( 2k )
equation of the fat circle in the first quadrant: lim 1 x ( )
=1 for 0 x<1 . So we guess that
k 
lim L =4 2=8 .
2k
k 

14
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

3
S= 2 (ln x) 1+(1/x) dx
2 2 2
1. y=ln x ds= 1+(dy/dx) dx= 1+(1/x) dx
1

 /2
S=  2 sin x 1+(2sin xcos x) dx [ by
2 2 2 2 2
2. y=sin x ds= 1+(dy/dx) dx= 1+(2sin xcos x) dx
0
(7)]

 /4
S=  2 x
2 2 2
3. y=sec x ds= 1+(dy/dx) dx= 1+(sec xtan x) dx 1+(sec xtan x) dx [ by (8)]
0

ln 2 2
S=  2 x dx [ by (8)] or  2 (ln y)
x 2 2x 2x 2
4. y=e  ds= 1+(dy/dx) dx= 1+e dx 1+e 1+ ( 1/y ) dy
0 1
[ by (6)]

3 / 2
5. y=x  y =3x . So
2 2
/ 2
S =  2 y ( )
1+ y dx=2  x
3 4
1+9x dx[u=1+9x ,du=36x dx]
4 3

0 0
145
2  2 3/2 145 
=
36
 u du=
18 3
u
1
=
27
( 145 145  1 )
1
2
6. The curve 9x=y +18 is symmetric about the x  axis, so we only use its top half, given by y=3 x 2
3 2 9
. dy/dx= , so 1+(dy/dx) =1+ . Thus,
2 x 2 4(x 2)
6 1/2
9 9 1
S = 2  3 x 2  
6 6
1+ dx=6 x 2+ dx=6 x+ dx
2 4(x 2) 2 4 2 4
3/2 6 3/2 3/2
2 1 25 9 125 27 98
=6  x+ =4  =4  =4  =49
3 4 2 4 4 8 8 8

2 2
7. y= x  1+(dy/dx) =1+ 1/ ( 2 x ) =1+1/(4x) . So
9 9 9
dy 2 1 1
S =  2 y 1+ dx= 2 x 1+ dx=2  x+ dx
4 dx 4 4x 4 4

1
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

9
2 1 3/2 4 1 3/2 9 
= 2
3
x+
4 4
=
3 8
(4x+1) =
4 6
( 37 37  17 17 )

2 2
8. y=cos 2x ds= 1+(dy/dx) dx= 1+( 2sin 2x) dx

 /6 3
1
S=  2 cos 2x 
2 2
1+4sin 2x dx=2 1+u du[u=2sin 2x,du=4cos 2xdx]
0 0 4
   
=
2
1
2
u 1+u +
2 1
2
ln u+ 1+u
2
( ) 0
3
=
2 2
3 1
 2+ ln
2
( 3 +2 ) =
2
3
+
4
ln ( 2+ 3 )

2 2 2
9. y=cosh x 1+(dy/dx) =1+sinh x=cosh x . So
1 1 1
1 1
S = 2  cosh xcosh xdx=2  2
(1+cosh 2x)dx= x+ sinh 2x
2 0
0 0
1 1 2 
= 1+
2
sinh 2 or 1+
4
(
e e
2
)
3 2
x 1 dy x 1
10. y= +  =  
6 2x dx 2 2
2x
4 2 2 2
2 x 1 1 x 1 x 1
1+(dy/dx) = + + = + = + 
4 2 4 2 2 2 2
4x 2x 2x
1 1

 
3 2 5
S = 2 x 1 x 1 x x x 1
+ + dx=2 + + + dx
6 2x 2 2 12 12 4 3
1/2 2x 1/2 4x
1 5  3 6 2  2 1
x x x x x x
=2  + +
12 3 4
dx=2 + 
72 6 8 1/2
1/2
1 1 1 1 1 1 263 263
=2 +   +  =2 = 
72 6 8 64 72 24 2 512 256

1 2 1 2
11. x=
3
y +2 ( ) 3/2 dx/dy=
2
(
y +2 ) 1/2(2y)=y 2
y +2 
2
1+(dx/dy) =1+y
2
( y2+2) = ( y2+1) 2 . So

2
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

2 2
1 4 1 2 1 1 21
 y ( y +1 ) dy=2
2
S=2 y+ y =2 4+2  =
1 4 2 1 4 2 2

2 2 2 2
12. x=1+2y  1+(dx/dy) =1+(4y) =1+16y . So
2 2
  2 2
S = 2  y 1+16y dy=
16
2
 ( 16y2+1) 1/232ydy= 16 3
( 2
16y +1 ) 3/2 1
1 1

=
24
( 65 65  17 17 )

3 3 2 4
13. y= x  x=y  1+(dx/dy) =1+9y . So
2 2 2
2
  
2 3 4 4 3
S = 2 x 1+(dx/dy) dy=2 y 1+9y dy= 1+9y 36y dy
1 1 36 1

 2 2 
=
18 3
1+9y
4
( ) 3/2 1
=
27
( 145 145  10 10 )

2 2 2
14. y=1 x  1+(dy/dx) =1+4x 

1 1
  2 1 
S=2  x 1+4x dx=
2
4
 8x
2
4x +1 dx=
4 3
(
2
4x +1 ) 3/2 0
=
6
(5 5  1)
0 0

2 2 1 2 2 1/2 2 2
15. x= a  y  dx/dy= (a  y ) ( 2y)= y/ a  y 
2
2 2 2 2 2
2 y a y y a
1+(dx/dy) =1+ = + = 
2 2 2 2 2 2 2 2
a y a y a y a y
a/2

S=  0
2 a y
2 2 a
2
a y
2
dy=2
a/2

0
ady=2 a y
a/2
0
=2 a
a
2
2
 0 = a . Note that this is

1 1
the surface area of a sphere of radius a , and the length of the interval y=0 to y=a/2 is the
4 4
length of the interval y= a to y=a .

2 2 2
16. x=acosh (y/a) 1+(dx/dy) =1+sinh (y/a)=cosh (y/a) . So

3
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

a a a
y y y 2y
 dy=4 a cosh dy=2 a
2
S = 2 acosh cosh 1+cosh dy
 a a a 0 a 0 a

= 2 a a
y+ sinh
2y a
=2 a
a
a+ sinh 2 =2 a
2 1
1+ sinh 2 or
 a
2
( e2+4 e 2)
2 a 0 2 2 2

S= 2 ln x
2 2 3 2
17. y=ln x dy/dx=1/x 1+(dy/dx) =1+1/x  1+1/x dx .
1
2 3 1 1
Let f (x)=ln x 1+1/x . Since n=10 ,  x= = . Then
10 5
1/5
S S =2  f (1)+4 f (1.2)+2 f (1.4)+   +2 f (2.6)+4 f (2.8)+ f (3)  9.023754 .
10 3
The value of the integral produced by a calculator is 9.024262 (to six decimal places).

1  1/2 2  1/2 1 1
18. y=x+ x  dy/dx=1+ x  1+(dy/dx) =2+x + x 
2 4
2

S= 2
1 1 1 1
( x+ x ) + 2+dx . Let f (x)=(x+ x ) 2+ + .
1 x 4x x 4x
2 1 1
Since n=10 ,  x= = . Then
10 10
1/10
S S =2  f (1)+4 f (1.1)+2 f (1.2)+   +2 f (1.8)+4 f (1.9)+ f (2)  29.506566 .
10 3
The value of the integral produced by a calculator is 29.506568 (to six decimal places).

2 2 2
19. y=sec x dy/dx=sec xtan x 1+(dy/dx) =1+sec xtan x
 /3
S=  2 sec x
2 2 2 2
1+sec xtan x dx . Let f (x)=sec x 1+sec xtan x .
0
 /3 0 
Since n=10 ,  x= = . Then
10 30
 /30  2 8 9 
S S =2  f (0)+4 f +2 f +   +2 f +4 f +f  13.527296
10 3 30 30 30 30 3
.
The value of the integral produced by a calculator is 13.516987 (to six decimal places).

x
x 1/2 dy 1 x 1/2 x e
20. y=(1+e )  = (1+e )  e= 
dx 2 x 1/2
2(1+e )

4
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

2 2x x 2x x 2
dy e 4+4e +e (e +2)
1+ =1+ = = 
dx x x x
4(1+e ) 4(1+e ) 4(1+e )
1

S=  0
2 1+e
x

2
x
e +2

1+e
x
dx= 
1

0
x
(e +2)dx=
x 1
e +2x = [(e+2) (1+0)]= (e+1) .
0

1 x 1 0 1
Let f (x)= (e +2) . Since n=10 ,  x= = . Then
2 10 10
1/10
S S =2  f (0)+4 f (0.1)+2 f (0.2)+   +2 f (0.8)+4 f (0.9)+ f (1)  11.681330 .
10 3
The value of the integral produced by a calculator is 11.681327 (to six decimal places).

2 2
21. y=1/x ds= 1+(dy/dx) dx= 1+  1/x
2
( ) dx= 1+1/x dx
4


2 4

 
4 2
S= 1 1 x +1 u +1 1 2
2  1+ dx=2 dx=2 du[u=x ,du=2xdx]
x 4 3 2 2
1 x 1 x 1 u
4 4


2 2
= 1+u
2
du= 
1+u
u
+ln u+ 1+u
2
( ) 1
1 u
17 2 17 4+ 17
=  +ln ( 4+ 17 ) +  ln ( 1+ 2 ) = 2 +ln
4 1 4 1+ 2

2 2
2 dy x dy x
22. y= x +1  =  ds= 1+ dx= 1+ dx
dx 2 dx 2
x +1 x +1
3

S
= 0
2 x +1
2
1+
x +1
x
2
2
dx=2 
3

0
2
2x +1 dx=2 2 
3

0
2
x+
1
2
2
dx

3
1 21 1 12
=2 2 x x + + ln x+ x+
2 2 4 2 0

5
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

3 1 1 1 1 1 3 19 1 19 1
=2 2 9+ + ln 3+ 9+  ln =2 2 + ln 3+ +
2 2 4 2 4 2 2 2 4 2 4
3 19 1 
=2 2 + ln ( 3 2 + 19 ) =3 19 + ln ( 3 2 + 19 )
2 2 4 2

3 / 2
23. y=x and 0 y 1 y =3x and 0 x 1 .
1 3
2 2 1
S =  2 x 1+ 3x( ) dx=2  1+u
2
6
2
du[u=3x ,du=6xdx]
0 0
3
  3
=
3

2
1+u du= [or use CAS ]
3
1
2
u 1+u +
2 1
2
ln u+ 1+u
2
( ) 0
0
 3 1 
= 10 + ln ( 3+ 10 ) = 3 10 +ln ( 3+ 10 )
3 2 2 6

dy 2 1 2
24. y=ln (x+1) , 0 x 1 . ds= 1+ dx= 1+ dx , so S
dx x+1
1 2

=  0
2 x 1+
1
( x+1 )
2
dx=  1
2 (u 1) 1+
1
u
2
du[u=x+1,du=dx]

2 2 2 2 2 2
2
=2  u
1+u
u
du 2  1+u
u
du=2 
2
1+u du 2  1+u
u
du
1 1 1 1
2
2 2
= [or use CAS ] 2
1
2
2 1
u 1+u + ln u+ 1+u
2
2
1
 2 ( ) 2
1+u  ln
1+ 1+u
u 1
1 1 1 1+ 5
=2 5+ ln ( 2+ 5 )  2 ln ( 1+ 2 )  2 5  ln  2 +ln ( 1+ 2 )
2 2 2 2
1 1+ 5 2 3
=2 ln ( 2+ 5 ) +ln +  ln ( 1+ 2 )
2 2 2 2





2 4
dy 1 x +1
25. S=2  y 1+
dx
dx=2
x
1+
1
4
dx=2
3
dx . Rather than trying to
1 x 1 x
1

6
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

4 4 2
evaluate this integral, note that x +1 > x =x for x>0 . Thus, if the area is finite,

 
4 2
x +1 x 1
S=2
3
dx>2
3
dx=2  x
dx
1 x 1 x 1

But we know that this integral diverges, so the area S is infinite.

 x  x2
26. S=  2 y 
2
1+(dy/dx) dx=2 e 1+( e ) dx .
0 0
 x  x2  x  x
Evaluate I= e 1+( e ) dx by using the substitution u= e , du=e dx .

I=
2 1 2 1 2
1+u du = 2 u 1+u + 2 ln u+ 1+u +C ( )
1 x  2x 1  x
= ( e ) 1+e + ln  e + 1+e
2 2

( 2x
) +C
Returning to the surface area integral, we have
t
 x  x2
S =2 lim  e 1+( e ) dx
t
0
t
=2 lim
t

1 x
2
( e ) 1+e
 2x
+
1
2
 x
ln  e + 1+e

( 2x
) 0

=2 lim
t

{ 1 t
2
( e ) 1+e
 2t
+
1
2
 t
(
ln  e + 1+e
 2t
) 
1
2
1
( 1) 1+1 + ln (  1+ 1+1
2
) }
=2 { 1
2
1
(0) 1 + ln ( 0+ 1
2
)  
1
2
1
2 + ln (  1+ 2
2
) }
=2 { [0]+
1
2
2  ln ( 2  1 ) } = 2  ln ( 2 1)

2 2 2 2
27. Since a>0 , the curve 3ay =x(a x) only has points with x 0 . (3ay  0 x(a x)  0 x 0 . )
The curve is symmetric about the x  axis (since the equation is unchanged when y is replaced by  y ).
y=0 when x=0 or a , so the curve’s loop extends from x=0 to x=a .

7
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

d 2 d dy 2 dy (a x)[ 2x+a x]

(3ay )=  6ay =x 2(a x)( 1)+(a x)  = 
dx dx dx dx 6ay
2 2 2 2 2 2
dy (a x) (a 3x) (a x) (a 3x) 3a the last fraction (a 3x)
= =  2 = 
dx 2 2 2 2 is 1/y 12ax
36a y 36a x(a x)
2 2 2 2 2 2 2
dy 2 a  6ax+9x 12ax a  6ax+9x a +6ax+9x (a+3x)
1+ =1+ = + = = for x 0 .
dx 12ax 12ax 12ax 12ax 12ax
(a)
a a a
x (a x) a+3x (a x)(a+3x)
S =  2 yds=2   dx=2  dx
x=0 0 3a 12ax 0 6a
a 2
  2 3 a   3  a

2 2 2 3 3 3
= (a +2ax 3x )dx= a x+ax  x = (a +a  a )=  a= .
3a 0 3a 0 3a 3a 3

Note that we have rotated the top half of the loop about the x  axis. This generates the full surface.
(b) We must rotate the full loop about the y  axis, so we get double the area obtained by rotating the
top half of the loop:
a a a
a+3x 4
S =2 2  xds=4  x 
1/2
dx= x (a+3x)dx
x=0 0 12ax 2 3a 0
a a
2 2 2 3/2 6 5/2 2 3 2 5/2 6 5/2

1/2 3/2
= (ax +3x )dx= ax + x = a + a
3a 0 3a 3 5 0 3 a 3 5
2
2 3 2 6 2 2 3 28 2 56 3a
= + a= a=
3 3 5 3 15 45

2
28. In general, if the parabola y=ax ,  c x c , is rotated about the y  axis, the surface area it
generates is
c 2ac
u 1
 
2 2
2 x 1+(2ax) dx = 2 1+u du u=2ax,du=2adx
0 0 2a 2a

8
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

2ac
  2 2ac
= 2
 ( 1+u2) 1/22udu= 2 3
(
1+u
2
) 3/2 0
4a 0 4a

= 2
( 1+4a2c2) 3/2 1
6a
2 2
Here 2c=10 ft and ac =2 ft, so c=5 and a= . Thus, the surface area is
25
625 4 3/2 625 16 3/2 625 41 41
S= 1+4  25  1 = 1+  1 =  1
6 4 625 24 25 24 125
5 2
=
24
( 41 41  125)  90.01ft

2 2 2
x y y(dy/dx) x dy b x
29. + =1 =  = 2 
2 2 2 2 dx
a b b a a y

1+
dy 2
= 1+
4 2
b x
=
4 2
b x +a y
=
( 1 x /a ) a b +b x  a b x
4 2
2 2
=
4 2 4 2 2 2 2
b x +a b
4 2 4 2

dx 4 2
a y
4 2
a y
4 2
( 2 2
a b 1 x /a ) 4 2 2 2 2
ab ab x
a  (a  b ) x
4 2 2 2 2 4 2 2 2
a +b x  a x
= =
a (a  x )
4 2 2 2 2 2
a a x
The ellipsoid’s surface area is twice the area generated by rotating the first quadrant portion of the
ellipse about the x axis. Thus,
a

S=
a
2 2 y
0
1+
dy
dx
2
dx=4  0
b
a
2
a x
2 a
4

a
( a2 b2) x2
2
a x
2
dx

2 2
a a b

=
4 b
2
a

a

0
4
a ( a2 b2) x2 dx= 4 2b
a
 0
a u
4 2 du
2
a b
2
2
[u= a  b x]
2

4 2 2
4 b u 4 a 2 1 u a a b
= a  u + sin
2 2 2 2 2 2
a a b a 0

9
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

2 2
2  1 a b
2 2 4 2 2 a bsin
a a b a  a b
= 4 b
2
a a
4 2
(a  b )2 2
+ sin
2
1
a
=2 b+
2 a
2 2 2 2 2
a a b a b

2 2 2
30. The upper half of the torus is generated by rotating the curve (x R) +y =r , y>0 , about the y
2 2 2 2 2
dy dy (x R) y +(x R) r
axis. y = (x R) 1+ =1+ = = . Thus,
dx dx 2 2 2 2
y y r  ( x R )
R+r

S=
R+r
2  2 x
R r
1+
dy
dx
2
dx=4  R r
2
rx

r  (x R)
2
dx

=
4 r   r
u+R
2
r  u
2
du[u=x R]

r r

=
4 r   r
udu
2
r  u
2
+4 Rr   r
du
2
r  u
2

=
4 r 0+8 Rr  0
du
2
r  u
2
[since the first integrand is odd and the second is even ]

 1 r  2
= 8 Rr sin (u/r) =8 Rr =4 Rr
0 2

* *
31. The analogue of f x in the derivation of (4) is now c f x , so
i i
b
2 2
 x= 2
n * / * /
S=lim 2 c f x 1+ f x c f (x) 1+ f (x) dx .
i=1 i i
n
a

1 
32. y=x
1/2
 y =
/
2
x
1/2
 1+ y ( / ) 2=1+1/4x , so by Exercise 31,
10
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

4

S= 2 ( 4 x ) 1+1/(4x) dx . Using a CAS, we get S=2 ln ( 17+4 ) + ( 31 17+1 )  80.6095 .
0 6

2 2 / 2 2
33. For the upper semicircle, f (x)= r  x , f (x)= x/ r  x . The surface area generated is
r r

S =
1
  r
2 (r 2
r  x
2
) 1+
2
r  x
2
x
2
dx=4  0
(r 2
r  x
2
) r
2
r  x
2
dx

=
4  0
r
2
2

r  x
2
 r dx

For the lower semicircle, f (x)=

2 2
r  x and f (x)=
/ x
2
r  x
2
, so S =4
2
 0
r
2
2

r  x
2
+r dx .

Thus, the total area is

r

S=S +S =8
1 2
 0
r
2
2

r  x
2
dx=8
2
r sin
 1 x
r
r

0
=8 r
2 
2
=4
2 2
r .

1 2 2 2 2
34. Take the sphere x +y +z = d and let the intersecting planes be y=c and y=c+h , where
4
1 1 2 2 1 2
 d c d h . The sphere intersects the xy plane in the circle x +y = d . From this equation,
2 2 4
dx dx y
we get x +y=0 , so = . The desired surface area is
dy dy x

11
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.2 Area of a Surface of Revolution

c+h c+h c+h

   
2 2 2 2 2
S = 2 xds=2 x 1+(dx/dy) dy=2 x 1+y /x dy=2 x +y dy
c c c
c+h c+h
1
= 2  2
d dy= d  dy= dh
c c

35. In the derivation of (4), we computed a typical contribution to the surface area to be
y +y
i 1 i
2 P P , the area of a frustum of a cone. When f (x) is not necessarily positive, the
2 i 1 i
* *
approximations y = f x  ( ) f x and y = f x ( )  f x must be replaced by
i i i i 1 i 1 i
* *
y= f x ( )  f x and y = f x ( )  f x . Thus,
i i i i 1 i 1 i
y +y 2
i 1 i * / *
2 P P  2 f x 1+ f x  x . Continuing with the rest of the derivation
2 i 1 i i i
b
2
as before, we obtain S= 2
/
f (x) 1+ f (x) dx .
a

/ /
36. Since g(x)= f (x)+c , we have g (x)= f (x) . Thus,
b b
2 2
 dx= 2
/ /
S = 2 g(x) 1+ g (x) f (x)+c 1+ f (x) dx
g
a a
b b
2 2
 dx+2 c
/ /
= 2 f (x) 1+ f (x) 1+ f (x) dx=S +2 cL
f
a a

12
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

3
1. The weight density of water is  =62.5 lb / ft .
(a) P= d ( 62.5lb/ft3) ( 3ft ) =187.5 lb / ft 2
(b) F=P A ( 187.5lb/ft2) ( 5ft ) ( 2ft ) =1875 lb. ( A is the area of the bottom of the tank.)
(c) As in Example 1, the area of the i th strip is 2( x) and the pressure is  d= x . Thus,
i

3 3 3
1 2 9
F=  x 2dx (62.5)(2) xdx=125 x =125 =562.5lb
0 0 2 0 2

4
2. (a) P= gd=1030(9.8)(2.5)=25 , 235 2.52 10 Pa =25.2 kPa
(b) F=P A ( 2.52 4
10 N/m
2
) ( 50m2) =1.26 6
10 N
2.5 2.5
2 2.5
(c) F=   gx 5dx=(1030)(9.8)(5)  xdx 2.52 10
4 5
x  1.58 10 N
0
0 0

3. Set up a vertical x  axis as shown, with x=0 at the water’s surface and x increasing in the
downward direction. Then the area of the i th rectangular strip is 6 x and the pressure on the strip is
* 3 *
 x (where   62.5lb/ft ). Thus, the hydrostatic force on the strip is  x  6 x and the total
i i


n *
hydrostatic force   x  6 x . The total force
i=1 i

6 6
  x  6 x=  x 6dx=6 
n *
F = lim xdx
i=1 i
n
2 2

1 2 6
= 6 x =6 (18 2)=96  6000lb
2 2

4 *
4. Set up a vertical x  axis as shown. Then the area of the i th rectangular strip is (4 x ) x . The
3 i
* * 4 *
pressure on the strip is  x , so the hydrostatic force on the strip is  x  (4 x ) x and the total
i i 3 i
force on the plate
1
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

4

n * *
  x (4 x ) x . The total force
i=1 i 3 i

4 4
4 4 4
  x  (4 x ) x=  x (4 x)dx=  
n ** 2
F =lim (4x x )dx
i=1 i 3 i 3 3
n
1 1

4 2 1 3 4 4 64 1 4
=  2x  x =  32  2 =  (9)=12  750 lb
3 3 1 3 3 3 3

2 2 2 2
5. Since an equation for the shape is x +y =10 ( x 0 ), we have y= 100 x . Thus, the area of the i
*2
th strip is 2 100 (x )  x
i
*
and the pressure on the strip is  gx , so the hydrostatic force on the
i
* *2
strip is  gx  2 100 (x )  x and the total force on the
i i


n * *2
plate   gx  2 100 (x )  x . The total force
i=1 i i
10
 100 (x )  x=  2 gx
n * *2 2
F = lim  gx  2 100 x dx
i=1 i i
n
0
10 10
2 2
=   g ( 100 x2) 1/2( 2x)dx=  g
3
2 3/2
(100 x )
0
=
3
 g(0 1000)
0
2000 2000 6 3 2
=  g  1000 9.8 6.5 10 N   1000kg/m andg 9.8m/s .
3 3

2
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

* 4 *
6. By similar triangles, w /4=x /5 , so w = x and the area of the
i i i 5 i
4 * *
i th strip is x  x . The pressure on the strip is  gx , so the hydrostatic
5 i i
* 4 * * 4 *
x  x and the total force on the plate  
n
force on the strip is  gx   gx  x  x . The total
i 5 i i=1 i 5 i
force

4 * 4 4 1 3 5 4 125 100
 x  x=  gx
n * 5
F =lim  gx  xdx=  g x =  g =  g
n

i=1 i 5 i 0 5 5 3 0 5 3 3
100 5
  1000 9.8 3.3 10 N.
3

4 ft wide a ft wide 1 *
7. Using similar triangles, = * , so a= x and the width of the i th rectangular
8 ft high 2 i
x ft high
i
* * *
strip is 12+2a=12+x . The area of the strip is 12+x  x . The pressure on the strip is  x .
i i i
8
  x=  x (12+x)dx
n * *
F = lim  x 12+x
i=1 i i
n
0
8 3 8
x 512
=  ( 12x+x ) dx= 2
6x +
2
3
=
0
384+
3
0
1664 4
= (62.5)  3.47 10 lb
3

8. In the figure, deleting a b h rectangle leaves a triangle with base a b and height h . By similar
triangles,
3
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

( a b ) ft wide d ft wide
= , so the width of the triangle is
h ft high *
h x ft high
i

* * *
h x x x
i i i
d= (a b)= 1 (a b)=a b (a b)
h h h

*
x
i
and the width of the trapezoid is b+d=a (a b) . The area of the i th rectangular strip is
h
*
x
i *
a (a b)  x and the pressure on it is  gx .
h i
*
x h
x
  x=  gx a
F = lim n * i
 gx a (a b) (a b) dx
n

i=1 i h 0 h
h h 2 3
 g(b a) 2 h b a h
=  ga xdx+  x dx= ga + g
0 h 0 2 h 3
2 a b a 2 a+2b 500 2
=  gh + = gh  gh (a+2b)N
2 3 6 3

2 * 2
9. From the figure, the area of the i th rectangular strip is 2 r  x  x and the pressure on it is
i
*
 g x +r .
i

* 2

n * 2
F = lim  g x +r 2 r  x  x
i=1 i i
n

r
=   g(x+r) 2 r  x dx
2 2

 r

4
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

r r
=  g r  x 2xdx+2 gr 
2 2 2 2
r  x dx
 r  r
The first integral is 0 because the integrand is an odd function. The second integral can be interpreted
as the area of a semicircular disk with radius r , or we could make the trigonometric substitution
1 2 3 3
x=rsin  . Continuing: F= g 0+2 gr  r = g r =1000g r N (SI units assumed).
2

* *
10. The area of the i th rectangular strip is 2 2y  y and the pressure on it is  d = 8 y .
i i i
8 8
F =   (8 y)2 2y dy=42 2 2  (8 y)y
1/2
dy
0 0
8 8
2 3/2 2 5/2
= 84 2  ( 8y1/2 y3/2) dy=84 2 8
3
y 
5
y
0
0
2 2
= 84 2  16 2   128 2
8
3 5
1 1 2
= 84 2 256 2  =43,008 =5734.4lb
3 5 15

5
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

* *
w 2x 2x
8 i i i
11. By similar triangles, =  w= . The area of the i th rectangular strip is  x and
4 3 * i 3 3
x
i
*
the pressure on it is  g 4 3  x .
i
4 3 4 3 4 3
2x 2 g
  
2
F=  g ( 4 3  x) dx=8 g xdx x dx
0 3 0 3 0
2 4 3 2 g 3 4 3 2 g
=4 g x  x =192 g 64 3 3
0 3 3 0 3 3
5
=192 g 128 g=64 g 64(840)(9.8) 5.27 10 N

12.
2
F =  g(10 x)2
2
4 x dx
0
2 2
=20 g 4 x dx  g
2 2
4 x 2xdx
0 0
4
1
=20 g 
4
(2 )  
2
g u
1/2
du [ u=4 x , du= 2xdx ]
2

6
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

2 3/2 4 16 16
=20  g  g u =20  g  g= g 20 
3 0 3 3
16 5
=(1000)(9.8) 20   5.63 10 N
3

13. (a) The top of the cube has depth d=1 m  20 cm =80 cm =0.8 m.

2
F= gdA (1000)(9.8)(0.8)(0.2) =313.6 314N

*
(b) The area of a strip is 0.2 x and the pressure on it is  gx .
i
1 1
1 2
F =   gx(0.2)dx=0.2 g x =(0.2 g)(0.18)=0.036 g=0.036(1000)(9.8)
0.8 2 0.8

=352.8 353N

2 2 3
14. The height of the dam is h= 70  25 cos 30 =15 19 . From the solution for Exercise
2
x 50x
8, the width of the trapezoid is 100 (100 50)=100 . From the small triangle in the second
h h
 x
figure, cos 30 =  z= xsec 30 =2 x/ 3 .
z
h h h
50x 2 200 100
  xdx 
2
F=  x 100 dx= x dx
0 h 3 3 0 h 3 0
2 3 2
200 h 100 h 200 h 200(62.5) 12,825
=  = = 
3 2 h 3 3 3 3 3 3 4
6
 7.71 10 lb

7
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

15. (a) The area of a strip is 20 x and the pressure on it is  x .

i
3 3
1 2 9
F =   x20dx=20 x =20  =90
0 2 0 2

= 90(62.5)=5625lb 5.63 103lb

9
1 2 9 81
(b) F=  x20dx=20
4
x =20  =810 =810(62.5)=50 , 625 lb  5.06 10 lb.
0 2 0 2
(c) For the first 3 ft, the length of the side is constant at 40 ft. For 3<x 9 , we can use similar
a 9 x 9 x
triangles to find the length a : =  a=40 .
40 6 6
3 9 3 9
9 x 1 2 20
F =   x40dx+  x(40)
6
dx=40
2
x + 
0 3
 ( 9x x2) dx
0 3 3

20 9 2 1 3 9 20 729 81
= 180 +  x x =180 +   243   9
3 2 3 3 3 2 2
= 180 +600 =780 =780(62.5)=48,750lb 4.88 104lb

8
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

 x
(d) For any right triangle with hypotenuse on the bottom, csc  =  hypotenuse
hypotenuse
2 2
40 +6 409
= xcsc  = x =  x.
6 3
9 9
409 1 1 2
F=   x20
3
dx= ( 20 409 ) 
3 2
x
3
3
1
=  10 409  (81 9)
3
5
 303,356lb 3.03 10 lb

*
16. Partition the interval a,b by points x as usual and choose x  x ,x for each i . The i th
i i i 1 i
*
horizontal strip of the immersed plate is approximated by a rectangle of height  x and width w x
i i
*
, so its area is A  w x  x . For small  x , the pressure P on the i th strip is almost constant and
i i i i i
* * *
P   gx by Equation 1. The hydrostatic force F acting on the i th strip is F =P A   gx w x  x .
i i i i i i i i i
Adding these forces and taking the limit as n
, we obtain the hydrostatic force on the immersed
plate:

b
   x =  gxw(x)dx
n n * *
F=lim F =lim  gx w x
i=1 i i=1 i i i
n
n
a

5
5 2 1
17. F=  gx w(x)dx , where w(x) is the width of the plate at depth x . Since n=6 ,  x= = ,
2 6 2
and
F S = g 1/2 [2 w(2)+4 2.5 w(2.5)+2 3 w(3)+4 3.5 w(3.5)]
6 3
+2 4 w(4)+4 4.5 w(4.5)+5 w(5)]
9
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

1
=  g(2 0+10 0.8+6 1.7+14 2.4+8 2.9+18 3.3+5 3.6)
6
1 5
= (1000)(9.8)(152.4) 2.5 10 N
6

b b b
1
18. (a) From Equation 8, x=  xw(x)dx Ax= xw(x)dx  gAx= g xw(x)dx
Aa a a
b
( gx)A=  gxw(x)dx=F by Exercise 16.
a
(b) The centroid of a circle is its center. In this case, the center is at a depth of r meters, so x=r .
2 3
Thus, F=( gx)A=( gr)( r )= g r .

19. The moment M of the system about the origin is M=

2
m x =m x +m x =40 2+30 5=230 .
i=1 i i 1 1 2 2

The mass m of the system is m=

2
m =m +m =40+30=70 . The center of mass of the system is
i=1 i 1 2
230 23
M/m= = .
70 7

20. M=m x +m x +m x =25( 2)+20(3)+10(7)=80 ;

1 1 2 2 3 3
80 16
x=M/(m +m +m )= = .
1 2 3 55 11
21. m= m =6+5+10=21 . M =
3 3
m y =6(5)+5( 2)+10( 1)=10 ;
i=1 i x i=1 i i
M M
1 x 10
M =
3 y
m x =6(1)+5(3)+10( 2)=1 . x= = and y= = , so the center of mass of the
y i=1 i i m 21 m 21
1 10
system is , .
21 21
22. M = m y =6( 2)+5(4)+1( 7)+4( 1)= 3 , M =
4 4
m x =6(1)+5(3)+1( 3)+4(6)=42 , and
x i=1 i i y i=1 i i
M M
42 21 3 21 3
m=
4 y x
m =16 , so x= = = and y= = ; the center of mass is ( x,y ) = , .
i=1 i m 16 8 m 16 8 16

23. Since the region in the figure is symmetric about the y  axis, we know that x=0 . The region is
‘‘bottom heavy,’’ so we know that y<2 , and we might guess that y=1.5 .

10
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

2 2 2
1 3
A =  (4 x )dx=2 (4 x )dx=2
2 2
4x x
 2 0 3 0

8 32
=2 8 =
3 3
2
1
x=  x(4 x )dx=0 since f (x)=x(4 x ) is an odd function (or since the region is symmetric about the
2 2
A 2
y  axis).

2 2 2
1 1 3 1 3 8 3 1 5
y=    2 (16 8x +x )dx=
22 2 4
(4 x ) dx= 16x x+ x
A 22 32 2 0 32 3 5 0

3 64 32 2 1 8 8
= 32 + =3 1 + =3 =
32 3 5 3 5 15 5

8
Thus, the centroid is ( x,y ) = 0, .
5

24. The region in the figure is ‘‘left heavy’’ and ‘‘bottom heavy,’’ so we know x<1 and y<1.5 , and
we might guess that x=0.7 and y=1.2 .
3
3x+2y=6 2y=6 3x y=3 x.
2
2 2
3 3 2
A= 3 x dx= 3x x =6 3=3 .
0 2 4 0

2 2
1 3 1 3 2
x=  x 3 x dx=  3x x dx
A0 2 30 2
1 3 2 1 3 2 1 2
= x x = (6 4)= ;
3 2 2 0 3 3

11
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

2 2 2 2
1 1 3 1 1 9 2 1 9 2 3 3 1
y=  3 dx=  
x 9 9x+ x dx= 9x x+ x = ( 18 18+6 ) =1 .
A0 2 2 3 2 0 4 6 2 4 0 6
2
Thus, the centroid is ( x,y ) = ,1 .
3

25. The region in the figure is "right heavy’’ and "bottom heavy,’’ so we know x>0.5 and y<1 , and
we might guess that x=0.6 and y=0.9 .
1
x 1
A= e dx= e
x
=e 1 ,
0
0
1
1 1 1
x =  xe dx=
x x x
xe  e [ by parts]
A0 e 1 0

1 1
= 0 (  1 ) = ,
e 1 e 1

1
1 1 x 2 1 1 2x 1 1
y= 
A0 2
e dx= ( ) 
e 1 4
e =
0 4 ( e 1 )
( e2 1) = e+1
4
. Thus, the centroid is

1 e+1
( x,y ) = e 1 , 4  ( 0.58,0.93) .

26. The region in the figure is "left heavy’’ and "bottom heavy,’’ so we know x<1.5 and y<0.5 , and
we might guess that
12
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

x=1.4 and y=0.4 .

2 2
1 1 1 1 2 1 1
A= dx= ln x =ln 2 , x=  x
2
dx= x = = ,
1 x
1 A1 x A 1 A ln 2

2 2 2 2
1 1 1 1  2 1 1
y=  dx=  x dx= 
A1 2 x 2A 1 2A x 1

1 1 1
=  +1 = .
2ln 2 2 4ln 2

1 1
Thus, the centroid is ( x,y ) = ,  ( 1.44,0.36 ) .
ln 2 4ln 2

1 1
2 3/2 1 2 2 1 1
27. A= ( x  x ) dx= x  x =  = .
0 3 2 0 3 2 6
1 1
1
x =  x( x  x)dx=6 (x  x )dx
3/2 2
A0 0

2 5/2 1 3 1 2 1 1 2
=6 x  x =6  =6 = ;
5 3 0 5 3 15 5

1 1
1 1 1
y=   x dx=6  (x x )dx
2 2 2
A0 2
( x ) 2 0
1 2 1 3 1 1 1 1
=3 x x =3  = .
2 3 0 2 3 2

2 1
Thus, the centroid is ( x,y ) = , .
5 2

13
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

28.
2 2
1 2 1 3
A =  (x+2 x )dx=
2
x +2x x
2 3  1
 1
8 1 1 9
= 2+4   2+ = .
3 2 3 2

2 2
1 2
x =  x(x+2 x )dx=  x +2x x dx
A 1
2
91
2 3
( )
2 1 3 2 1 4 2
= x +x  x
9 3 4  1
2 8 1 1 2 9 1
= +4 4   +1 =  = ;
9 3 3 4 9 4 2

2 2 2
1 1 2 1
y= 
A 12
2 22
(x+2)  (x ) dx= 
9 2
 ( x2+4x+4 x4) dx= 19 1 3 2
3
x +2x +4x
1 5
5
x
 1
 1
1 8 32 1 1 1 9 33 1 72 8
= +8+8   +2 4+ = 18+  =  = .
9 3 5 3 5 9 3 5 9 5 5

1 8
Thus, the centroid is ( x,y ) = , .
2 5

 /4
 /4
29. A=  (cos x sin x)dx= sin x+cos x = 2 1 ,
0
0

14
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

 /4
 1
x =A  x(cos x sin x)dx
0
 1  /4
=A x(sin x+cos x)+cos x sin x [integration by parts]
0
1
 2 1
 1  4
=A 2 1 =
4 2 1

 /4  /4
 1 1 1 1  /4 1 1
y=A  2
2 2
(
cos x sin x dx=
2A
)  cos 2xdx=
4A
sin 2x
0
= =
4A 4 ( 2  1 )
0 0
 2 4 1
Thus, the centroid is ( x,y ) = ,  (0.27,0.60) .
4 ( 2  1) 4 ( 2  1)

1 2 1
1 1 2 1
30. A= xdx+
2
dx= x + ln x = +ln 2 ,
0 1 x 2 0 1 2
1 2 1
1 1 1 3
 x dx+ 1dx
2 2
x= = x + x
A 0 1 A 3 0 1

1 1 2 4 8
= +1 =  = ,
A 3 1+2ln 2 3 3(1+2ln 2)


1 1 2
y=1 
1 2
x dx+
1
dx =
1 1 3
x + 
1
A 0 2 2 2A 3 0 x 1
2x
1
1 1 1 5 5
= + = = .
2A 3 2 12A 6+12ln 2

15
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

8 5
Thus, the centroid is ( x,y ) = ,  (1.12,0.35) . The principle used in this
3(1+2ln 2) 6(1+2ln 2)
problem is stated after Example 3: the moment of the union of two nonoverlapping regions is the sum
of the moments of the individual regions.

31. From the figure we see that y=0 . Now

5 5
2
A =  2 5 x dx=2
3/2
 (5 x)
0 3 0

2 3/2 20
=2 0+  5 = 5
3 3
so

5 5
1 1
x=
A0
 x 5 x  (  5 x ) dx=  2x 5 x dx
A0
0
1
=
A
 (
2 5 u
2
) u( 2u)du[u= 2 2
5 x ,x=5 u ,u =5 x,dx= 2udu]
5
5
4 5
=
A
 u
2
( 5 u2) du= 4A 5 3 1 5
3
u
5
u
0
=
3
5 5
25
3
5 5 5 =5 3=2
0

Thus, the centroid is (x,y)=(2,0) .

32. By symmetry, M =0 and x=0 ;

y
16
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

1 2  5
A=   1 +4 , so m= A=5( +4)= ( +8) ;
2 2 2
1 1 1
1 1 3 10 50
M =  2
x 2
[( 1 x )  ( 2) ]dx=5  x  3 dx= 5
2 2 2 2
3
x +3x = 5
0 3
=
3
; ( )
0 0
1 2  50 20  20
y= M =  = . Thus, the centroid is ( x,y ) = 0, .
m x 5( +8) 3 3( +8) 3( +8)

1 1
33. By symmetry, M =0 and x=0 . A= bh=  2 2=2 .
y 2 2
1 1
1 1
 
2 2
M =  (2 2x) dx=2 (2 2x) dx
x 2 2
 1 0
1
1 2
 ( 1 x )
2
= 2 1  2 dx
2 0
0
= 4 u ( du)[u=1 x,du= dx]
2

1 3 0 1 4
=  4 u = 4  =
3 1 3 3
1 1 1 4 2 2
y= M= M=  = . Thus, the centroid is ( x,y ) = 0, .
m x  A x 1 2 3 3 3

1 2 1 2
34. By symmetry about the line y=x , we expect that x=y . A=  r , so m= A=2A=  r .
4 2
r 2 r
M =
x

1
2
( 2
r  x
2
) dx=2
1
2
 (r 2 x2) dx= 2
r x
1 3
3
x
r

0
=
2 3
3
r .
0 0
2
r r r 2
2 1/2 2 3 1 2 3/2 2 3 4
M =
y
 x r  x dx=
2 2
(r 2 x ) 2xdx= u
0 3
1/2
du=
r . x= M =
m y 3
2
u
2
r
=
3
r =
3
r
0 0 0  r
1 2 2 3 4 4 4
; y= M= r = r . Thus, the centroid is ( x,y ) = r, r .
m x 2 3 3 3 3
 r
17
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

35.
2 x 3 2
2 x
A = ( 2  x ) dx=
x 2

ln 2 3 0
0
4 8 1 3 8
=   =   1.661418 .
ln 2 3 ln 2 ln 2 3

2 2
1 1
x =  x(2  x )dx=  (x2  x )dx
x 2 x 3
A0 A0
x x 4 2
1 x2 2 x
=   [use parts]
A ln 2 2 4
( ln 2 ) 0
1 8 4 1
=   4+
A ln 2 2 2
( ln 2 ) ( ln 2 )
1 8 3 1
=   4  (1.297453) 0.781
A ln 2 2 A
( ln 2 )

2 2 2x 5 2
1 1 x 2 2 2 1 1 2x 4 1 1 2 x
y= 
A0 2
(2 )  ( x ) dx= 
A0 2
(2  x )dx= 
A 2

2ln 2 5 0

1 1 16 32 1 1 15 16 1
=    =   (2.210106) 1.330
A 2 2ln 2 5 2ln 2 A 4ln 2 5 A
3
36. The curves y=x+ln x and y=x  x intersect at ( a,c )  ( 0.447141, 0.357742 ) and
( b,d )  ( 1.507397,1.917782 ) .
b b
A = ( x+ln x x +x) dx= ( 2x+ln x x3) dx
3

a a

2 1 4 b
= x +xln x x x  0.709781
4 a

18
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

b b
1 1
Aa
3
( Aa
2
)
x =  x 2x+ln x x dx=  2x +xln x x dx
4
( )
1 2 3 1 2 1 5 b 1
= x + x ( 2ln x 1 )  x  (0.699489) 0.985501
A 3 4 5 a A

b b
1 1 2 1
 ( x+ln x )  ( x  x ) 
2 3 2 6 4
y= dx= 2xln x+ ( ln x )  x +2x dx
Aa 2 2A a
1 2 1 2 2 1 7 2 5 b 1
= x ln x x +x ( ln x )  2xln x+2x x+ x  (0.765092) 0.538964
2A 2 7 5 a 2A

Thus, the centroid is ( x,y )  ( 0.986,0.539 ) .

37. Choose x and y axes so that the base (one side of the triangle) lies along the x axis with the
other vertex along the positive y axis as shown. From geometry, we know the medians intersect at a
2
point of the way from each vertex (along the median) to the opposite side. The median from B
3
1
goes to the midpoint (a+c),0 of side AC , so the point of intersection of the medians is
2
2 1 1 1 1
 (a+c), b = (a+c), b .
3 2 3 3 3

This can also be verified by finding the equations of two medians, and solving them simultaneously to
find their point of intersection. Now let us compute the location of the centroid of the triangle. The
1
area is A= (c a)b .
2

19
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

0 c 0 c
1 b b 1 b b
x=
A
 x a ( a x ) dx+ x c (c x)dx =
A a
 ( ax x dx+
2
c
)  ( cx x2) dx
a 0 a 0

b 1 2 1 3 0 b 1 2 1 3 c b 1 3 1 3 b 1 3 1 3
= ax  x + cx  x =  a+ a + c c
Aa 2 3 a Ac 2 3 0 Aa 2 3 Ac 2 3
3 3
= 2  a 2 c 1 2 2 a+c
 +  = (c  a )=
a ( c  a) 6 c ( c  a) 6 3( c  a) 3
and
0 2 c 2
1 1 b 1 b
y=
A
 2 a
(a x) dx+
2 c
(c x) dx
a 0
2 0 2 c
1
=
A
b
2
 ( a  2ax+x ) dx+
2 2 b
2
 ( c2 2cx+x2) dx
2a a 2c 0
2 0 2 c
1 b 2 1 3 2 b 2 1 3
2
= a x ax + x + c x cx + x
A 2 3 a 2 3 0
2a 2c
2 2 2 2
1 b 1 3 b 3 3 3 1 3
3 1 b 2 ( c  a) b b
=  a +a  a + c c+ c = (  a+c ) =  =
A 2 3 2 3 A 6 ( c  a) b 6 3
2a 2c
a+c b
Thus, the centroid is ( x,y ) = , , as claimed.
3 3
Remarks: Actually the computation of y is all that is needed. By considering each side of the triangle
1
in turn to be the base, we see that the centroid is of the way from each side to the opposite vertex
3
and must therefore be the intersection of the medians.
The computation of y in this problem (and many others) can be simplified by using horizontal rather
than vertical approximating rectangles. If the length of a thin rectangle at coordinate y is  (y) , then its
area is  (y) y , its mass is   (y) y , and its moment about the x axis is  M = y (y) y . Thus,
x

c a
In this problem,  (y)= ( b y ) by similar triangles, so
b

20
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

b b b 3
1 c a 1 2 1 3 b b
y= 
A0 b
y(b y)dy=
2
2
 ( by y ) dy=
2 2
2 2
by 
3
y
0
= 
2
2 6
=
3
b 0 b b

Notice that only one integral is needed when this method is used.

3
38. Divide the lamina into three rectangles with masses 2 , 2 and 6 , with centroids  ,1 ,
2
1 3
0, and 2, , respectively. The total mass of the lamina is 10 . So, using Formulas 5, 6,
2 2
and 7, we have
M
1 3 1 3 1
= 
y
x= mx= 2  +2(0)+6(2) = (9) , and
m m i=1 i i 10 2 10
M
1 3 1 1 3 1
= 
x
y= m y= 2(1)+2 +6 = (12) .
m m i=1 i i 10 2 2 10
9 6
Thus, the centroid is ( x,y ) = , .
10 5

39. Divide the lamina into two triangles and one rectangle with respective masses of 2 , 2 and 4 , so
2
that the total mass is 8 . Using the result of Exercise 37, the triangles have centroids  1, and
3
2 1
1, . The centroid of the rectangle (its center) is 0, . So, using Formulas 5 and 7, we
3 2
M
1 3 1 2 2 1 1 2 1
= 
x
have y= m y= 2 +2 +4  = = , and x=0 , since
m m i=1 i i 8 3 3 2 8 3 12
1
the lamina is symmetric about the line x=0 . Thus, the centroid is ( x,y ) = 0, .
12

40. A sphere can be generated by rotating a semicircle about its diameter. By Example 4, the center of
4r 8r
mass travels a distance 2 y=2 = , so by the Theorem of Pappus, the volume of the
3 3
2
 r 8r 4 3
sphere is V =Ad=  =  r .
2 3 3

41. A cone of height h and radius r can be generated by rotating a right triangle about one of its legs
1
as shown. By Exercise 37, x= r , so by the Theorem of Pappus, the volume of the cone is
3
1 1 1 1 2
V =Ad=  base height  (2 x)= rh 2 r =  r h.
2 2 3 3
21
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

42. From the symmetry in the figure, y=4 . So the distance traveled by the centroid when rotating the
1 1
triangle about the x  axis is d=2  4=8 . The area of the triangle is A= bh= (2)(3)=3 . By the
2 2
Theorem of Pappus, the volume of the resulting solid is Ad=3(8 )=24 .

43. Suppose the region lies between two curves y= f (x) and y=g(x) where f (x) g(x) , as illustrated in
*
Figure 13. Choose points x with a=x <x <   <x =b and choose x to be the midpoint of the i th
i 0 1 n i
1 *
subinterval; that is, x =x =
i i 2 i 1 i ( )
x +x . Then the centroid of the i th approximating rectangle R is its
i
1
center C = x ,
i i 2
f x +g x
i i ( ) ( )( ) ( )
. Its area is f x  g x  x , so its mass is
i i

( ) ( ) ( ) ( ) ( )
 f x  g x  x . Thus, M R = f x  g x  x x = x f x  g x  x and
i i ( ) ( )
y i i i i i i i
1 1 2 2
M ( R ) = f ( x )  g ( x )  x f ( x ) +g ( x ) =  f ( x )  g ( x )  x . Summing over i and
x i 2 i i 2 i i i i
b
taking the limit as n
, we get M =lim   x f ( x )  g ( x )  x=  x f (x) g(x) dx and
y i i i i
n
a
b
1 2 2 1
 
2 2
M =lim
x
n

 
i 2 ( )  g( x )
f x
i i
 x=
a 2
f (x)  g(x) dx . Thus,

M b M M M b
1 1 1
=  x f (x) g(x) dx and y= = 
y y x x 2 2
x= = = f (x)  g(x) dx
m  A Aa m  A Aa 2

n m
44. (a) Let 0 x 1 . If n<m , then x >x ; that is, raising x to a larger power produces a smaller
22
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.3 Applications to Physics and Engineering

number.

(b) Using Formulas 9 and the fact that the area of is

1
1 1 m n
A= ( xn xm) dx= n+1  =
m+1 (n+1)(m+1)
, we get
0
1 1
(n+1)(m+1) (n+1)(m+1)
x=
m n

n m
x x  x dx=
m n

n+1 m+1
x  x dx ( )
0 0
1
(n+1)(m+1) 1 1 (n+1)(m+1) (n+1)(m+1) 1 (n+1
=
m n

n+2 m+2
=
(n+2)(m+2)
and y=
m n
 2 ( xn) 2 ( xm) 2 dx=
2(
0
(n+1)(m+1) 1 1 (n+1)(m+1)
=  =
2(m n) 2n+1 2m+1 (2n+1)(2m+1)

(c) If we take n=3 and m=4 , then

4 5 4 5 2 20
( x,y ) = ,
5 6 7 9
= ,
3 63

2 3 8 20
which lies outside R since = < . This is the simplest of many possibilities.
3 27 63

23
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.4 Applications to Economics and Biology

2000

/
1. By the Net Change Theorem, C(2000) C(0)= C (x)dx 
0
2000
3 2000

2 2
C(2000) =20 , 000+ (5 0.008x+0.000009x )dx=20 , 000+ 5x 0.004x +0.000003x
0
0
=20 , 000+10 , 000 0.004(4 , 000 , 000)+0.000003(8 , 000 , 000 , 000)=30 , 000 16 ,
000+24 , 000
=$38 , 000

5000

/
2. By the Net Change Theorem, R(5000) R(1000)= R (x)dx 
1000

2 5000
R(5000) =12,400+
5000
(12 0.0004x)dx=12,400+ 12x 0.0002x
1000 1000
=12,400+(60,000 5,000) (12,000 200)=$55,600

3. If the production level is raised from 1200 units to 1600 units, then the increase in cost is
1600 1600
C(1600) C(1200) = 
/
C (x)dx=  ( 74+1.1x 0.002x2+0.00004x3) dx
1200 1200

0.002 3 2 4 1600
= 74x+0.55x 
x +0.00001x
3 1200
= 64,331,733.33 20,464,800=$43,866,933.33

4.
30
Consumer surplus =  p(x) p(30) dx
0
30
1 30
= 5 x 5 dx
0 10 10
1 2 30
= 3x x =90 45=$45
20 0

1
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.4 Applications to Economics and Biology

450
5. p(x)=10  =10 x+8=45 x=37 .
x+8
37 37
450
Consumersurplus =  p(x) 10 dx= 
x+8
 10 dx
0 0
37
= 450ln ( x+8 )  10x
0
= (450ln 45 370) 450ln 8
45
= 450ln  370 $407.25
8

2
6. p (x)=3+0.01x . P= p (10)=3+1=4 .
S S
10
Producersurplus =  [P p (x)]dx
S
0
10 10
0.01 3

2
= 4 3 0.01x dx= x x
0 3 0

 10 3.33=$6.67

2
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.4 Applications to Economics and Biology

3/2 3/2 3/2 2/3

7. P= p (x) 400=200+0.2x  200=0.2x  1000=x  x=1000 =100 .
S
100 100 100
1 3/2
Producer surplus =  P p (x) dx= 
S
400 200+0.2x( 3/2
) dx=  200
5
x dx
0 0 0

2 5/2 100
= 200x x =20 , 000 8 , 000=$12 , 000
25 0

1 1
8. p=50 x and p=20+ x intersect at p=40 and x=200 .
20 10
200
1
Consumer surplus=  50
20
x 40 dx
0
1 2 200
= 10x x =$1000
40 0
1

200
Producer surplus= 40 20 x dx
0 10
1 2 200
= 20x x =$2000
20 0

 x/5000
800,000e
9. p(x)= =16 x=x  3727.04 .
x+20,000 1

x
1

Consumersurplus= [ p(x) 16]dx $37,753

0

3
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.4 Applications to Economics and Biology

 0.5 1
10. The demand function is linear with slope = and p(400)=7.5 , so an equation is
35 70
1 1 185 1 185
p 7.5= (x 400) or p= x+ . A selling price of $6 implies that 6= x+ 
70 70 14 70 14
1 185 84 101
x=  =  x=505 .
70 14 14 14

505 505
1 185 1 2 101
Consumer surplus=   x+  6 dx=  x+ x  $1821.61
0 70 14 140 14 0

8 8 8
2 3/2 2
11. f (8) f (4)= f (t)dt=
/
t dt=
3
t
4
=
3
( 16 2  8 )  $9.75 million
4 4

12.
9 0.8t 9
10e 25 0.8t 9
n(9) n(5) = ( 2200+10e ) dt=
0.8t
2200t+
0.8
= 2200t +
5
9
5 2
e
5
5

(
=2200(9 5)+12.5 e  e  24 , 860
7.2 4
)
4 4
 PR  (4000)(0.008)  4 3
13. F= =  1.19 10 cm / s
8 l 8(0.027)(2)

4 4
 P R 4 R
0 0  PR 4 4 P 0
14. If the flux remains constant, then =  P R =PR  = .
8 l 8 l 0 0 P R
0
4
3 R 4 4
P 0
R= R  =  P=P  3.1605P >3P ; that is, the blood pressure is more
4 0 P 3 0 3 0 0
0 R
4 0
than tripled.

15.
4
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.4 Applications to Economics and Biology

12 12 12
1 1 2 3 2 1 3 12
 c(t)dt=  4
t(12 t)dt= 
3t
4
t dt=
2
t 
12
t
0
= ( 216 144 ) =72 mg  s / L.
0 0 0
A 8mg 1 60
Thus, the cardiac output is F= 12 = = L/s= L / min.
72mg s/L 9 9
 c(t)dt
0

.
16. As in Example 2, we will estimate the cardiac output using Simpson’s Rule with  t=2 .
20
2
 c(t)dt 
3
1(0)+4(2.4)+2(5.1)+4(7.8)+2(7.6)
0
+4(5.4)+2(3.9)+4(2.3)+2(1.6)+4(0.7)+1(0)
2
= (110.8) 73.87mg s/L
3
A 8
Therefore, F =  0.1083 L / s or 6.498 L / min.
73.87 73.87

5
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

40,000
1. (a)  f (x)dx is the probability that a randomly chosen tire will have a lifetime between 30 ,
30,000
000 and 40 , 000 miles.

(b)  f (x)dx is the probability that a randomly chosen tire will have a lifetime of at least 25 , 000
25,000
miles.

15
2. (a) The probability that you drive to school in less than 15 minutes is  f (t)dt .
0

(b) The probability that it takes you more than half an hour to get to school is  f (t)dt .
30

3. (a) In general, we must satisfy the two conditions that are mentioned before Example 1   

3

2
namely, (1) f (x) 0 for all x , and (2) f (x)dx=1 . For 0 x 4 , we have f (x)= x 16 x  0 ,
  64
so f (x) 0 for all x . Also,
 4 4 4
3 3 3 2
 f (x)dx = 
2 2 1/2 2 3/2
x 16 x dx= (16 x ) ( 2x)dx= (16 x )
0 64
  128 0 128 3 0

1 4 1
=
64
( 16 x2) 3/2 0
=
64
(0 64)=1 .

Therefore, f is a probability density function.

(b)
2 2
3 3
P(X<2) =  f (x)dx= 
2 2 2 1/2
x 16 x dx= (16 x ) ( 2x)dx
 
0 64 128 0
3 2 2 3/2 2 1 2 3/2 2 1 3/2 3/2
= (16 x ) = (16 x ) = (12  16 )
128 3 0 64 0 64
1 1 3
=
64
( 64 12 12 ) =
64
( 64 24 3 ) =1
8
3 0.350481

2
4. (a) For 0 x 1 , we have f (x)=kx (1 x) , which is nonnegative if and only if k 0 . Also,
 1 1 1
1 3 1 4
 f (x)dx= kx (1 x)dx=k  (x  x )dx=k
2 2 3
x x =k/12 . Now k/12=1 k=12 . Therefore, f is
  0 0 3 4 0

1
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

a probability density function if and only if k=12 .

(b) Let k=12 .
 1 1
1 4 1
=  f (x)dx=  12x (1 x)dx=  (12x  12x )dx= 4x  3x
2 2 3 3
P X
2 1/2 1/2 1/2
1/2

1 3 5 11
=(4 3)  =1 =
2 16 16 16

(c) The mean

 1 1 1
1 4 1 5
=  xf (x)dx= x
12x (1 x)dx=12 (x  x )dx=12
2 3 4
 x x
  0 0 4 5 0

1 1 12 3
=12  = =
4 5 20 5

5. (a) In general, we must satisfy the two conditions that are mentioned before Example 1   

namely, (1) f (x) 0 for all x , and (2)  f (x)dx=1 . Since f (x)=0 or f (x)=0.1 , condition (1) is
 
 10 10
1
satisfied. For condition (2), we see that  f (x)dx=  0.1dx=
10
x
0
=1 . Thus, f (x) is a probability
  0
density function for the spinner’s values.
(b) Since all the numbers between 0 and 10 are equally likely to be selected, we expect the mean to
be halfway between the endpoints of the interval; that is, x=5 .
 10 10
1 2 100
 =  xf (x)dx=  x(0.1)dx= x = =5 , as expected.
  0 20 0 20

 10
1
6. (a) As in the preceding exercise, (1) f (x) 0 and (2)  f (x)dx=  f (x)dx=
2
(10)(0.2)=1 . So f (x)
  0
is a probability density function.
(b)
(a) 3
1 3
P(X<3)= f (x)dx= (3)(0.1)= =0.15
0 2 20
(b) 3
1 3
P(X<3)= f (x)dx= (3)(0.1)= =0.15
0 2 20
(c) We first compute P(X>8) and then subtract that value and our answer in (i) from 1 (the total

2
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

10
1 2
probability). P(X>8)=  f (x)dx= (2)(0.1)= =0.10 . So P(3 X 8)=1 0.15 0.10=0.75 .
8 2 20
(d) We first compute P(X>8) and then subtract that value and our answer in (i) from 1 (the total
10
1 2
probability). P(X>8)=  f (x)dx= (2)(0.1)= =0.10 . So P(3 X 8)=1 0.15 0.10=0.75 .
8 2 20

(c) We find equations of the lines from ( 0,0 ) to ( 6,0.2 ) and from ( 6,0.2 ) to ( 10,0 ) , and find that

{
1
x
30 if 0 x<6
f (x)= 1 1 if 6 x<10
 x+ otherwise
20 2
0

 6 10 6 10
1 1 1 1 3 1 3 1 2
 =  xf (x)dx= x
30
x dx+  x 
20
x+
2
dx=
90
x + 
0 60
x+ x
4 6
  0 6
216 1000 100 216 36 16
= +  +   + = =5.3
90 60 4 60 4 3

 x x
1 1  t/5 1 1  1
7. We need to find m so that  f (t)dt=  lim 
t/5
e dt=  lim ( 5)e = 
m 2 x  m 5 2 x  5 m 2
 m/5 1  m/5 1 1 1
( 1)(0 e )=  e =   m/5=ln  m= 5ln =5ln 2 3.47 min.
2 2 2 2

8. (a)
(a) 200
1   t/1000 200  1/5
P(0 X 200)= 
t/1000
e dt=  e = e +1 0.181
0 1000 0

(b) 200
1   t/1000 200  1/5
P(0 X 200)= 
t/1000
e dt=  e = e +1 0.181
0 1000 0

(c) 
1   t/1000 x  4/5
P(X>800)= 
t/1000
e dt=lim  e =0+e  0.449
800 1000 x 
800

(d) 
1   t/1000 x  4/5
P(X>800)= 
t/1000
e dt=lim  e =0+e  0.449
800 1000 x 
800

3
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

 x
1 1  t/1000 1  t/1000 x 1
(b) We need to find m so that  f (t)dt=  lim  e dt=  lim  e = 
m 2 x  m 1000 2 x 
m 2
 m/1000 1 1 1
0+e =   m/1000=ln  m= 1000ln =1000ln 2 693.1 h.
2 2 2

9. We use an exponential density function with  =2.5 min.

 x
1   t/2.5 x  4/2.5
(a) P(X>4)=  f (t)dt=lim 
t/2.5
e dt=lim  e =0+e  0.202
4 x  4 2.5 x 
4
2
 t/2.5 2  2/2.5
(b) P(0 X 2)= f (t)dt=  e = e +1 0.551
0
0
(c) We need to find a value a so that P(X a)=0.02 , or, equivalently, P(0 X a)=0.98
a
 t/2.5 a  a/2.5  a/2.5
 f (t)dt=0.98  e
0
=0.98  e +1=0.98 e =0.02  a/2.5=ln 0.02
0
1
a= 2.5ln=2.5ln 50 9.78 min  10 min. The ad should say that if you aren’t served within 10
50
minutes, you get a free hamburger.

73


2
1 ( x 69 )
10. (a) With  =69 and  =2.8 , we have P(65 X 73)= exp  dx 0.847
2.8 2 2
65 2
2.8
(using a calculator or computer to estimate the integral).
(b) P(X>6 feet )=P(X>72 inches )=1 P(0 X 72) 1 0.858=0.142 , so 14.2% of the adult male
population is more than 6 feet tall.


2
1 ( x 9.4 )
11. P(X 10)= exp  dx . To avoid the improper integral we approximate it
4.2 2 2
10 2
4.2
100


2
1 ( x 9.4 )
by the integral from 10 to 100 . Thus, P(X 10) exp  dx 0.443 (using a
4.2 2 2
2
4.2 10
calculator or computer to estimate the integral), so about 44 percent of the households throw out at
least 10 lb of paper a week.
Note : We can’t evaluate 1 P(0 X 10) for this problem since a significant amount of area lies to
the left of X=0 .

12. (a)

4
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

480


2
1 ( x 500 )
P(0 X 480)= exp  dx 0.0478 (using a calculator or computer to
12 2 2
2
12
0
estimate the integral), so there is about a 4.78% chance that a particular box contains less than 480 g
of cereal.
(b) We need to find  so that P(0 X<500)=0.05 . Using our calculator or computer to find
P(0 X 500) for various values of  , we find that if  =519.73 , P=0.05007 ; and if  =519.74 ,
P=0.04998 . So a good target weight is at least 519.74 g.

 +2

. .
2
1 ( x  ) x  1
13. P(  2  X  +2 )= exp  dx . Substituting t= and dt= dx
 2 2  
  2 2
gives us

.
2 2

1
e
 t
2
/ 2( dt)=
1
 e
 t
2
/ 2 dt 0.9545
 2 2 2  2

14. Let f (x)=

{ 0
ce
 cx
if x<0
if x 0
where c=1/ . By using parts, tables, or a CAS, we find that

(1):  xe dx=
bx
( / b ) ( bx 1)
e
bx 2

(2):  x e dx= ( e / b ) ( b x  2bx+2 )

2 bx bx 3 2 2

Now
 0 
 (x  ) f (x)dx=  (x  ) f (x)dx+  (x  ) f (x)dx
2 2 2 2
 =
    0
t t
2  cx
= 0+lim c (x  ) e dx=c
lim  ( x2e cx 2x  cx
e +
2  cx
e ) dx
t  0 t  0
Next we use (2) and (1) with b= ctoget

 cx  cx  cx t
e

2
=clim 
e
3
( c x +2cx+2)  2
2 2 e
2
( cx 1)+
2
 c
t  c c 0

Using l’Hospital’s Rule several times, along with the fact that  =1/c , we get

5
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

2 2 2 1 1 1 1 1 1
 =c 0  +
+
=c =   = =
3 c 2 2  c 3 2 c
c c c c c

4 2  2r/a0
15. (a) First p(r)= r e  0 for r 0 . Next,
3
a
0

 
 p(r)dr=  0
a
4
3
0
2  2r/a0
r e dr=
4
3
a t
0
lim


t

0
2  2r/a0
r e dr

By using parts, tables, or a CAS , we find that  x e dx=(e /b )(b x  2bx+2) . ( * )

2 bx bx 3 2 2

3
a
4 0
Next, we use ( * ) (with b= 2/a ) and l’Hospital’s Rule to get ( 2) =1 . This satisfies the
0 3  8
a
0
second condition for a function to be a probability density function.
2
4 r 4 2r 2 2
(b) Using l’Hospital’s Rule, lim = lim = lim =0 .
3 2r/a 3 2r/a 2 2r/a
a r  0 a r  0 a r  0
0 e 0 ( 2/a ) e
0
0 ( 2/a ) e
0
To find the maximum of p , we differentiate:

/ 4 2  2r/a0 2  2r/a
0 4  2r/a
0 r
p (r)= r e  +e (2r) = e (2r)  +1
3 a 3 a
a 0 a 0
0 0

/ r /
p (r)=0 r=0 or 1= r=a . p (r) changes from positive to negative at r=a , so p(r) has its
a 0 0
0
maximum value at r=a .
0
(c) It is fairly difficult to find a viewing rectangle, but knowing the maximum value from part (b)
helps.

4 2  2a0/a0 4 2
p a =( ) ae = e  9,684,098,979
0 3 0 a
a 0
0

6
Stewart Calculus ET 5e 0534393217;8. Further Applications of Integration; 8.5 Probability

With a maximum of nearly 10 billion and a total area under the curve of 1 , we know that the
‘‘hump’’ in the graph must be extremely narrow.

r 4a
0

(d) P(r)=  0
4
a
0
3
2  2s/a0
se ds ( )
P 4a =
0
 0
4
a
0
3
2  2s/a0
se ds . Using ( * ) from part (a)

(with b= 2/a ),

0
 2s/a 4a 3
0 0 a
4 e 4 2 4 4 0  8
( )=
P 4a
0 3 3 2
s+
a
s+2 =
3  8
e (64+16+2) 1(2)
a  8/a a 0 a
0 0 0 0 0
1  8  8
= 
2
(
82e  2 =1 41e  0.986 )
 t
4 3  2r/a0
(e)  =  rp(r)dr= lim  r e dr . Integrating by parts three times or using a CAS, we find that
3
  a t  0
0
bx

3 bx
x e dx=
e
4
( b3x3 3b2x2+6bx 6) . So with b= 2
, we use l’Hospital’s Rule, and get
b a
0
4
a
4 0 3
 =  ( 6) = a .
3 16 2 0
a
0

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.1 Modeling with Differential Equations

 1 /  2
1. y=x x  y =1+x . To show that y is a solution of the differential equation, we will substitute
/
the expressions for y and y in the left hand side of the equation and show that the left hand side is
equal to the right hand side.

 2
LHS=xy +y=x 1+x
/
( ) + ( x x 1) =x+x 1+x x 1=2x=RHS
/ 2 2
2. y=sin xcos x cos x y =sin x( sin x)+cos x(cos x) ( sin x)=cos x sin x+sin x .
LHS = y / +(tan x)y=cos 2 x sin 2 x+sin x+(tan x)(sin xcos x cos x)
= cos 2 x sin 2 x+sin x+sin 2 x sin x=cos 2 x=RHS,
so y is a solution of the differential equation. Also, y(0)=sin 0cos 0 cos 0=0 1 1= 1 , so the initial
condition is satisfied.

/ / / 2 / /
3. (a) y=sin kt y =kcos kt y = k sin kt . y +9y=0
2
 k sin kt+9sin kt=0 for all t ( 9 k2) sin kt=0 for all t 2
9 k =0 k= 3
/ / / 2 2
(b) y=Asin kt+Bcos kt y =Akcos kt Bksin kt y = Ak sin kt Bk cos kt .
/ /
Thus, y +9y=0  Ak sin kt Bk cos kt+9(Asin kt+Bcos kt)=0
2 2
( 9 k2) Asin kt+ ( 9 k2) Bcos kt=0 .
The last equation is true for all values of A and B if k= 3 .

rt
4. y=e  y =re 
/ rt
y
/ / 2 rt
=r e . y
/ / /
+y  6y=0
2 rt
r e +re  6e =0
rt rt
(r 2+r 6) ert=0 (r+3)(r 2)=0
 r= 3 or 2

t / t / / t / / / t t t t t
5. (a) y=e  y =e  y =e . LHS =y +2y +y=e +2e +e =4e  0 , so y=e is not a solution of the
differential equation.
 t /  t / /  t / / /  t  t  t  t
(b) y=e  y = e  y =e . LHS =y +2y +y=e  2e +e =0= RHS, so y=e is a solution.
 t
(c) y=te 
/
y =t  e (  t ) +e t(1)=e t(1 t) y
/ /  t
=e (t 2) .
/ / /  t  t  t
LHS = y +2y +y=e (t 2)+2e (1 t)+te
= e t
( t 2 ) +2(1 t)+t =e (0)=0=RHS,
 t

 t
so y=te is a solution.
2  t  t  t
(d) y=t e 
/
y =te ( 2 t )  y
/ /
=e ( t2 4t+2) .
 t
LHS = y / /
+2y +y=e
/
( t2 4t+2) +2te t(2 t)+t2e t
= e t
( t2 4t+2) +2t(2 t)+t2 =e t(2) 0,

1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.1 Modeling with Differential Equations

2  t
so y=t e is not a solution.

2 2 2
x /2 / x /2 x /2
6. (a) y=Ce  y =Ce ( 2x/2 ) =xCe =xy .

(b)

2
0 x /2
(c) y(0)=5 Ce =5 C=5 , so the solution is y=5e .

(d) y(1)=2 Ce =2

1/2
C=2e
 1/2
, so the solution is y=2e
 1/2 x /2
e
2
=2e
( x  1 ) /2
2
.

/ 2
7. (a) Since the derivative y = y is always negative (or 0 if y=0 ), the function y must be decreasing
(or equal to 0 ) on any interval on which it is defined.
1 1 2
/ 1 / 1 2
(b) y=  y = . LHS =y = = = y = RHS
x+C 2 2 x+C
( x+C ) ( x+C )
/ 2
(c) y=0 is a solution of y = y that is not a member of the family in part (b).
1 1 1 1 1 1
(d) If y(x)= , then y(0)= = . Since y(0)=0.5 , =  C=2 , so y= .
x+C 0+C C C 2 x+2

3 /
8. (a) If x is close to 0 , then xy is close to 0 , and hence, y is close to 0 . Thus, the graph of y must
3
have a tangent line that is nearly horizontal. If x is large, then xy is large, and the graph of y must
have a tangent line that is nearly vertical. (In both cases, we assume reasonable values for y .)
2  1/2 2  3/2
(b) y= c x( ) 
/
y =x c x ( ) .
2  1/2 3 2  3/2
RHS =xy =x
3
( c x ) (
=x c x ) /
=y = LHS

(c)

/ /
When x is close to 0 , y is also close to 0 . As x gets larger, so does y .
2
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.1 Modeling with Differential Equations

 1/2 1 1 1 2  1/2
(d) y(0)= ( c 0 ) =1/ c and y(0)=2 c=  c= , so y=  x .
2 4 4

dP P dP P P
9. (a) =1.2P 1 . Now >0 1 >0 <1 P<4200 the population is
dt 4200 dt 4200 4200
increasing for 0<P<4200 .
dP
(b) <0 P>4200
dt
dP
(c) =0 P=4200 or P=0
dt

dy 4 3 2
10. (a) y=k
/
y =0 , so
dt
=y  6y +5y 
4 3
0=k  6k +5k  k
2 2
( k2 6k+5) =0 2
k (k 1)(k 5)=0 k=0 ,
1 , or 5
dy 2
(b) y is increasing  >0 y (y 1)(y 5)>0 y ( 
,0 )  ( 0,1 )  ( 5,
)
dt
dy
(c) y is decreasing  <0 y ( 1,5)
dt

t 2
11. (a) This function is increasing and also decreasing. But dy/dt=e (y 1)  0 for all t , implying that
the graph of the solution of the differential equation cannot be decreasing on any interval.
(b) When y=1 , dy/dt=0 , but the graph does not have a horizontal tangent line.

12. The graph for this exercise is shown in the figure at the right.
/ /
A. y =1+xy>1 for points in the first quadrant, but we can see that y <0 for some points in the first
quadrant. So equation A is incorrect.
/ /
B. y = 2xy=0 when x=0 , but we can see that y >0 for x=0 . So equation B is incorrect.
/
C. y =1 2xy seems reasonable since:

(1) /
When x=0 , y could be 1 .
(2) /
When x<0 , y could be greater than 1 .

3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.1 Modeling with Differential Equations

(3) /
/ 1 y /
Solving y =1 2xy for y gives us y= . If y takes on small negative values, then as x

2x
+
, y 0 , as shown in the figure. Thus, the correct equation is C.

13. (a) P increases most rapidly at the beginning, since there are usually many simple, easily learned
sub skills associated with learning a skill. As t increases, we would expect dP/dt to remain positive,
but decrease. This is because as time progresses, the only points left to learn are the more difficult
ones.
dP
(b) =k(M P) is always positive, so the level of performance P is increasing. As P gets close to M
dt
, dP/dt gets close to 0 ; that is, the performance levels off, as explained in part (a).

(c)

14. (a) P increases most rapidly at the beginning, since there are usually many simple, easily learned
sub skills associated with learning a skill. As t increases, we would expect dP/dt to remain positive,
but decrease. This is because as time progresses, the only points left to learn are the more difficult
ones.
dP
(b) =k(M P) is always positive, so the level of performance P is increasing. As P gets close to M
dt
, dP/dt gets close to 0 ; that is, the performance levels off, as explained in part (a).

(c)

4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

1. (a)

(b) It appears that the constant functions y=0 , y= 2 , and y=2 are equilibrium solutions. Note that
/ 1 2
these three values of y satisfy the given differential equation y =y 1 y .
4

2. (a)

/
(b) From the figure, it appears that y= is an equilibrium solution. From the equation y =xsin y , we
see that y=n ( n an integer) describes all the equilibrium solutions.

/
3. y =y 1 . The slopes at each point are independent of x , so the slopes are the same along each line
/
parallel to the x axis. Thus, IV is the direction field for this equation. Note that for y=1 , y =0 .

/
4. y =y x=0 on the line y=x , when x=0 the slope is y , and when y=0 the slope is  x . Direction field
II satisfies these conditions.

/ 2 2
5. y =y  x =0 y= x . There are horizontal tangents on these lines only in graph III, so this
equation corresponds to direction field III.

/ 3 3 3 3
6. y =y  x =0 on the line y=x , when x=0 the slope is y , and when y=0 the slope is  x . The graph
is similar to the graph for Exercise 4, but the segments must get steeper very rapidly as they move
away from the origin, because x and y are raised to the third power. This is the case in direction field
I.

1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

7. (a) y(0)=1

(b) y(0)=0

(c) y(0)= 1

8. (a) y(0)=1

(b) y(0)=0

2
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

(c) y(0)= 1

9.

10.
/ 2 2
x y y =x  y
 1  3  8
 3  1 8
 1  0.5 0.75
 0.5  1  0.75

/ /
Note that y =0 for y= x . If x < y , then y <0 ; that is, the slopes are negative for all points in
quadrants I and II above both of the lines y=x and y= x , and all points in quadrants III and IV below
both of the lines y= x and y=x . A similar statement holds for positive slopes.

11.
x y y / =y 2x
 2 2 2
 2 2 6
2 2 2
2 2 6

Note that
3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

/
y =0 for any point on the line y=2x . The slopes are positive to the left of the line and negative to the
right of the line. The solution curve in the graph passes through (1,0) .

12.
/
x y
y =1 xy
 1 1 0
 2 2 3
 2 2 5

/
Note that y =0 for any point on the hyperbola xy=1 (or y=1/x ). The slopes are negative at points
‘‘inside’’ the branches and positive at points everywhere else. The solution curve in the graph passes
through (0,0) .

13.
/
x y y =y+xy
0 2 2
1 2 4
 3 2 4

/
Note that y =y(x+1)=0 for any point on y=0 or on x= 1 . The slopes are positive when the factors y
and x+1 have the same sign and negative when they have opposite signs. The solution curve in the
graph passes through (0,1) .

4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

14.
/
x y y =x xy
 2 0 2
 2 3 4
 2 1 4

/
Note that y =x(1 y)=0 for any point on x=0 or on y=1 . The slopes are positive when the factors x
and 1 y have the same sign and negative when they have opposite signs. The solution curve in the
graph passes through (1,0) .

15. In Maple, we can use either directionfield (in Maple’s share library) or plots [fieldplot] to plot the
direction field. To plot the solution, we can either use the initial value option in directionfield, or
actually solve the equation. In Mathematica, we use PlotVectorField for the direction field, and the
Plot [Evaluate[...]] construction to plot the solution, which

is y=e
.
( 1 cos 2x ) / 2
. In Derive, use Direction_Field (in utility file ODE_APPR) to plot the direction
field. Then use DSOLVE1( y*SIN(2*x),1,x,y,0,1) (in utility file ODE1) to solve the equation.
Simplify each result.
5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

16. See Exercise 15 for specific CAS directions. The exact solution is

17.

L=lim y(t) exists for  2 c 2 ; L= 2 for c= 2 and L=0 for  2<c<2 . For other values of c , L does
t 
not exist.

18.

/
Note that when f (y)=0 on the graph in the text, we have y = f (y)=0 ; so we get horizontal segments
at y= 1 ,  2 . We get segments with negative slopes only for 1< y <2 . All other segments have
positive slope. For the limiting behavior of solutions:

If y(0)>2 , then lim y= and lim y=2 .
t  t  

If 1<y(0)<2 , then tlim y=1 and tlim  y=2 .

If  1<y(0)<1 , then tlim y=1 and tlim  y= 1 .

If  2<y(0)< 1 , then tlim y= 2 and tlim  y= 1 .

6
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

If y< 2 , then tlim y= 2 and tlim  y=  .

19. (a)
( )
(a) h=0.4 and y =y +hF x ,y
1 0 0 0
y =1+0.4 1=1.4 . x =x +h=0+0.4=0.4 , so y =y ( 0.4 ) =1.4 .
1 1 0 1
(b) h=0.4 and y =y +hF x ,y 
( ) 1 0 0 0
y =1+0.4 1=1.4 . x =x +h=0+0.4=0.4 , so y =y ( 0.4 ) =1.4 .
1 1 0 1
(c) h=0.2 x =0.2 and x =0.4 , so we need to find y . y =y +hF x ,y =1+0.2y =1+0.2 1=1.2 ,
1 2 2 1 0 ( 0 0 ) 0

2 1 ( 1 1 )
y =y +hF x ,y =1.2+0.2y =1.2+0.2 1.2=1.44 .
1
(d) h=0.2 x =0.2 and x =0.4 , so we need to find y . y =y +hF x ,y =1+0.2y =1+0.2 1=1.2 ,
1 2 2 1 0 ( 0 0 ) 0

2 1 ( 1 1 )
y =y +hF x ,y =1.2+0.2y =1.2+0.2 1.2=1.44 .
1
(e) h=0.1
4 1 0 (
x =0.4 , so we need to find y . y =y +hF x ,y =1+0.1y =1+0.1 1=1.1 ,
4 0 0 ) 0

1 ( )
y =y +hF x ,y =1.1+0.1y =1.1+0.1 1.1=1.21 ,
2 1 1 1
y =y +hF ( x ,y ) =1.21+0.1y =1.21+0.1 1.21=1.331 ,
3 2 2 2 2
y =y +hF ( x ,y ) =1.331+0.1y =1.331+0.1 1.331=1.4641 .
4 3 3 3 3
(f) h=0.1 x =0.4 , so we need to find y . y =y +hF ( x ,y ) =1+0.1y =1+0.1 1=1.1 ,
4 4 1 0 0 0 0
y =y +hF ( x ,y ) =1.1+0.1y =1.1+0.1 1.1=1.21 ,
2 1 1 1 1
y =y +hF ( x ,y ) =1.21+0.1y =1.21+0.1 1.21=1.331 ,
3 2 2 2 2
y =y +hF ( x ,y ) =1.331+0.1y =1.331+0.1 1.331=1.4641 .
4 3 3 3 3

(b)

x
We see that the estimates are underestimates since they are all below the graph of y=e .

(c)

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

(a) 0.4
For h=0.4 : ( exactvalue )  ( approximatevalue ) =e  1.4 0.0918
(b) 0.4
For h=0.2 : ( exactvalue )  ( approximatevalue ) =e  1.44 0.0518
(c) 0.4
For h=0.1 : ( exactvalue )  ( approximatevalue ) =e  1.4641 0.0277
Each time the step size is halved, the error estimate also appears to be halved (approximately).

20.

As x increases, the slopes decrease and all of the estimates are above the true values. Thus, all of the
estimates are overestimates.

21. h=0.5 , x =1 , y =0 , and F ( x,y ) =y 2x .

0 0
Note that x =x +h=1+0.5=1.5 , x =2 , and x =2.5 .
1 0 2 3

1 0 ( )
y =y +hF x ,y =0+0.5F ( 1,0 ) =0.5[0 2(1)]= 1 .
0 0
y =y +hF ( x ,y ) = 1+0.5F ( 1.5, 1 ) = 1+0.5[ 1 2(1.5)]= 3 .
2 1 1 1
y =y +hF ( x ,y ) = 3+0.5F ( 2, 3) = 3+0.5[ 3 2(2)]= 6.5 .
3 2 2 2
y =y +hF ( x ,y ) = 6.5+0.5F ( 2.5, 6.5) = 6.5+0.5[ 6.5 2(2.5)]= 12.25 .
4 3 3 3
22. h=0.2 , x =0 , y =0 , and F ( x,y ) =1 xy .
0 0
Note that x =x +h=0+0.2=0.2 , x =0.4 , x =0.6 , and x =0.8 .
1 0 2 3 4

1 0 ( )
y =y +hF x ,y =0+0.2F ( 0,0 ) =0.2[1 (0)(0)]=0.2 .
0 0
y =y +hF ( x ,y ) =0.2+0.2F ( 0.2,0.2 ) =0.2+0.2[1 (0.2)(0.2)]=0.392 .
2 1 1 1
y =y +hF ( x ,y ) =0.392+0.2F ( 0.4,0.392 ) =0.392+0.2[1 (0.4)(0.392)]=0.56064 .
3 2 2 2
y =y +hF ( x ,y ) =0.56064+0.2[1 (0.6)(0.56064)]=0.6933632 .
4 3 3 3
y =y +hF ( x ,y ) =0.6933632+0.2[1 (0.8)(0.6933632)]=0.782425088 .
5 4 4 4
Thus, y(1) 0.7824 .

8
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

23. h=0.1 , x =0 , y =1 , and F ( x,y ) =y+xy .

0 0
Note that x =x +h=0+0.1=0.1 , x =0.2 , x =0.3 , and x =0.4 .
1 0 2 3 4

1 0 ( )
y =y +hF x ,y =1+0.1F ( 0,1 ) =1+0.1[1+(0)(1)]=1.1 .
0 0
y =y +hF ( x ,y ) =1.1+0.1F ( 0.1,1.1 ) =1.1+0.1[1.1+(0.1)(1.1)]=1.221 .
2 1 1 1
y =y +hF ( x ,y ) =1.221+0.1F ( 0.2,1.221 ) =1.221+0.1[1.221+(0.2)(1.221)]=1.36752 .
3 2 2 2
y =y +hF ( x ,y ) =1.36752+0.1F ( 0.3,1.36752 ) =1.36752+0.1[1.36752+(0.3)(1.36752)]
4 3 3 3
=1.5452976 .

y =y +hF x ,y
5 4 ( 4 4 ) =1.5452976+0.1F ( 0.4,1.5452976)
=1.5452976+0.1[1.5452976+(0.4)(1.5452976)]=1.761639264 .

Thus, y(0.5) 1.7616 .

24. (a) h=0.2 , x =1 , y =0 , and F ( x,y ) =x xy .

0 0
We need to find y , because x =1.2 and x =1.4 .
2 1 2

1 0 ( )
y =y +hF x ,y =0+0.2F ( 1,0 ) =0.2[1 (1)(0)]=0.2 .
0 0
y =y +hF ( x ,y ) =0.2+0.2F ( 1.2,0.2 ) =0.2+0.2[1.2 (1.2)(0.2)]=0.392 y(1.4) .
2 1 1 1
(b) Now h=0.1 , so we need to find y .
4
y =0+0.1[1 (1)(0)]=0.1 ,
1
y =0.1+0.1[1.1 (1.1)(0.1)]=0.199 ,
2
y =0.199+0.1[1.2 (1.2)(0.199)]=0.29512 , and
3
y =0.29512+0.1[1.3 (1.3)(0.29512)]=0.3867544 y(1.4) .
4

25. (a)
(a) H =1 , N =1 y(1)=3
(b) H =1 , N =1 y(1)=3
(c) H =0.1 , N =10 y(1) 2.3928
(d) H =0.1 , N =10 y(1) 2.3928
(e) H =0.01 , N =100 y(1) 2.3701
(f) H =0.01 , N =100 y(1) 2.3701

9
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

(g) H =0.001, N =1000 y(1) 2.3681

(h) H =0.001, N =1000 y(1) 2.3681

3 3
 x / 2  x
(b) y=2+e  y = 3x e

/ 2
LHS=y +3x y= 3x e +3x
2  x
3
2
( 2+e ) = 3x e
 x
3
2  x
3
2 2  x
+6x +3x e =6x =RHS
3
2

 0
y(0)=2+e =2+1=3
(c)
(a)  1
For h=1 : ( exactvalue )  ( approximatevalue ) =2+e  3  0.6321
(b)  1
For h=1 : ( exactvalue )  ( approximatevalue ) =2+e  3  0.6321
(c)  1
For h=0.1 : ( exactvalue )  ( approximatevalue ) =2+e  2.3928  0.0249
(d)  1
For h=0.1 : ( exactvalue )  ( approximatevalue ) =2+e  2.3928  0.0249
(e)  1
For h=0.01 : ( exactvalue )  ( approximatevalue ) =2+e  2.3701  0.0022
(f)  1
For h=0.01 : ( exactvalue )  ( approximatevalue ) =2+e  2.3701  0.0022
(g)  1
For h=0.001 : ( exactvalue )  ( approximatevalue ) =2+e  2.3681  0.0002
(h)  1
For h=0.001 : ( exactvalue )  ( approximatevalue ) =2+e  2.3681  0.0002

3 3
26. (a) We use the program from the solution to Exercise 25 with Y =x  y , H =0.01 , and N
1
2 0
=
0.01 0 0 ( )
=200 . With x ,y = ( 0,1 ) , we get y(2) 1.9000 .

(b)
Notice from the graph that y(2) 1.9 , which serves as a check on our calculation in part (a).

dQ 1 / 1 /
27. (a) R + Q=E(t) becomes 5Q + Q=60 or Q +4Q=12 .
dt C 0.05

10
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

(b) From the graph, it appears that the limiting value of the charge Q is about 3 .
/
(c) If Q =0 , then 4Q=12 Q=3 is an equilibrium solution.

(d)

/ /
(e) Q +4Q=12 Q =12 4Q . Now Q(0)=0 , so t =0 and Q =0 .
0 0

1 0 0 0 ( )
Q = Q +hF t ,Q =0+0.1(12 4 0)=1.2

Q
2 = Q +hF ( t ,Q ) =1.2+0.1(12 4 1.2)=1.92
1 1 1

Q
3 = Q +hF ( t ,Q ) =1.92+0.1(12 4 1.92)=2.352
2 2 2

Q
4 = Q +hF ( t ,Q ) =2.352+0.1(12 4 2.352)=2.6112
3 3 3

Q
5 = Q +hF ( t ,Q ) =2.6112+0.1(12 4 2.6112)=2.76672
4 4 4
Thus, Q =Q(0.5) 2.77 C.
5

28. (a) From Exercise .1.14, we have dy/dt=k(y R) . We are given that R=20 C and dy/dt= 1 C /
1
min when y=70 C. Thus,  1=k(70 20) k= and the differential equation becomes
50
1
dy/dt= (y 20) .
50
(b)

11
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.2 Direction Fields and Euler’s Method

The limiting value of the temperature is 20 C;
that is, the temperature of the room.

1
(c) From part (a), dy/dt= (y 20) . With t =0 , y =95 , and h=2 min, we get
50 0 0
1
1 0 ( )
y =y +hF t ,y
0 0 =95+2 
50
(95 20) =92

y =y +hF ( t ,y ) 1
2 1 1 1 =92+2  (92 20) =89.12
50
y =y +hF ( t ,y ) 1
3 2 2 2 =89.12+2  (89.12 20) =86.3552
50
y =y +hF ( t ,y ) 1
4 3 3 3 =86.3552+2  (86.3552 20) =83.700992
50
y =y +hF ( t ,y ) 1
5 4 4 4 =83.700992+2  (83.700992 20) =81.15295232
50

Thus, y(10) 81.15 C.

12
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

dy y dy dx dy dx
 =
ln x +C ln x C C
1. =  =   ln y =ln x +C  y =e =e e =e x  y=Kx ,
dx x y x y x
C
where K= e is a constant. (In our derivation, K was nonzero, but we can restore the excluded case
y=0 by allowing K to be zero.)

2x
dy e 1 2x 1 2x
 4y dy= e dx
3 2x 3 2x 4 4
2. =  4y dy=e dx y= e +C  y= e +C
dx 3 2 2
4y

dy xy dy xdx dy xdx 1
 =
2 / 2
3. (x +1)y =xy = 2  = 2 [ y 0 ]   ln y = ln (x +1)+C [
dx y y 2 2
x +1 x +1 x +1
2
u=x +1 , du=2xdx ] =ln (x +1) +ln e =ln e
2 1/2 C
( C 2
x +1  ) y =e
C 2
x +1  y=K
2
x +1 , where
C
K= e is a constant. (In our derivation, K was nonzero, but we can restore the excluded case y=0 by
allowing K to be zero.)

dy 2 dy dy 1 1
 = sin xdx
/ 2
4. y =y sin x =y sin x =sin xdx  = cos x+C  =cos x C
dx 2 2 y y
y y
1
y= , where K= C . y=0 is also a solution.
cos x+K

/ 2 dy 2 sin y 2
5. (1+tan y) y =x +1 (1+tan y) =x +1 1+ dy=(x +1)dx
dx cos y
 sin y 1 3
 dy= (x +1)dx
2
1 y ln cos y = x +x+C . Note: The left side is equivalent to
cos y 3
y+ln sec y .

du 1+ r 2 3/2 2 3/2
 (1+u )du= (1+r
1/2 1/2
6. =
dr 1+ u
 ( 1+ u ) du= ( 1+ r ) dr  )dr  u+
3
u =r+ r +C
3

t
dy 1
7.
dt
=
te
 y
2
1+y dy=te dt
t
 y 1+y dy= te dt
2 t
3
1+y
2
( ) 3/2=tet et+C 
2
y 1+y
2 t t 2/3 t t 2/3
1+y =[3(te  e +C)]  y= [3(te  e +C)]  1

2
xy 2ln y 2ln y x
 dy= xdx
/ 2 2
8. y =  dy=xdx ( ln y ) = +C  ln y= x /2+C 
2ln y y y 2

1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

2
 x /2+C
y=e

du du du 1 2
9.
dt
=2+2u+t+tu
dt
=(1+u)(2+t)  1+u
= (2+t)dt  ln 1+u =
2
t +2t+C 
2 2 2 2
t /2+2t +C t /2+2t C t /2+2t t /2+2t
1+u =e =Ke , where K=e  1+u= Ke  u= 1 Ke where K>0 . u= 1 is
2
t /2+2t
also a solution, so u= 1+Ae , where A is an arbitrary constant.

dz t + z dz  z  z  z 1 1
 
t z t t t t z
10. +e =0 = e e  e dz= e dt   e = e +C  e =e  C  =e  C  e= 
dt dt z t
e e C
1
z=ln t
 (t )
z= ln e  C
e C

dy 2 dy  1  1
11. =y +1 , y(1)=0 .  = dx tan y=x+C . y=0 when x=1 , so 1+C=tan 0=0 C= 1 .
dx 2
y +1
 1
Thus, tan y=x 1 and y=tan (x 1) .
2
dy ycos x 1+y 1
12.
dx
=
2
2
, y(0)=1 . 1+y dy=ycos xdx
y
(dy=cos xdx ) y
+y dy= cos xdx
1+y
1 2 1 1 1 2 1
ln y + y =sin x+C . y(0)=1 ln 1+ =sin 0+C C= , so ln y + y =sin x+ . We cannot
2 2 2 2 2
solve explicitly for y .

13. xcos x=(2y+e ) y  xcos xdx=(2y+e )dy  (2y+e )dy= xcos xdx
3y / 3y 3y

2 1 3y
y + e =xsin x+cos x+C [ where the second integral is evaluated using integration by parts]. Now
3
1 2 2 1 3y 2
y(0)=0 0+ =0+1+C  C= . Thus, a solution is y + e =xsin x+cos x .
3 3 3 3
We cannot solve explicitly for y .

dP  1/22 3/2
 dP= t
1/2 1/2
14. = Pt  dP/ P = t dt P t +C . dt 2P =
dt 3
2 2 1/2 2 3/2 2 1 3/2 1
P(1)=2 2 2= +C  C=2 2  , so 2P = t +2 2   P= t + 2 
3 3 3 3 3 3
1 3/2 1 2
P= t + 2 .
3 3

2
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

2
du 2t+sec t
15.
dt
=
2u
, u(0)= 5 .  2udu= ( 2t+sec 2t ) dt 2 2
u =t +tan t+C , where u(0) =0 +tan 0+C
2 2

2 2 2 2
 C=( 5) =25 . Therefore, u =t +tan t+25 , so u= t +tan t+25 . Since u(0)= 5 , we must have
2
u= t +tan t+25 .

dy  y  y 1 2 0 1 2
=te , y(1)=0 .  e dy= t dt
y
16.  e = t +C . Since y(1)=0 ,  e =  1 +C . Therefore,
dt 2 2
2
1 3  y 1 2 3  y 3 1 2 3 t
C= 1
2
=
2
and  e = t 
2 2
. So e = 
2 2
t =
2

y
e=
2
2
 (
y=ln 2 ln 3 t
2
) for t < 3
3 t
.

/ dy a+y dy
17. y tan x=a+y , 0<x< /2 =  =cot xdx [ a+y 0 ] 
dx tan x a+y
dy cos x
 =
ln sin x +C ln sin x C C
dx ln a+y =ln sin x +C  a+y =e =e  e =e sin x 
a+y sin x
C
a+y=Ksin x , where K= e . (In our derivation, K was nonzero, but we can restore the excluded case
y= a by allowing K to be zero.) y( /3)=a
 3 4a 4a 4a
a+a=Ksin  2a=K  K= . Thus, a+y= sin x and so y= sin x a .
3 2 3 3 3

/ 2 dy 2 2 dy dx
18. xy +y=y  x =y  y xdy=(y  y)dx = 
dx 2 x
y y
dy dx 1 1 dx
 y(y 1) = x [ y 0 , 1 ]   y 1  y dy= x  ln y 1  ln y =ln x +C 
y 1 y 1 y 1 1 1
ln
y
C
(
=ln e x 
y
) C
=e x 
y
C
=Kx , where K= e  1
y
=Kx
y
=1 Kx
1 1
y= . Now y(1)= 1  1=  1 K= 1 K=2 ,
1 Kx 1 K
1
so y= .
1 2x
4 4
dy dy dy
=4x dx  = 4x dx ln y =x +C  e
3 3 3 4 ln y x +C x C
19. =4x y , y(0)=7 . =e  y =e e 
dx y y
4 4
x x
y=Ae ; y(0)=7 A=7 y=7e .

20.

3
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

2
dy y dy dx 1 1 1 1
= , y(1)=1 .  =   = +C . y(1)=1  1= +C  C= . So
dx 3 2 3 y 2 2 2
x y x 2x
2 2
1 1 1 2+2x 2x
= + =  y= .
y 2 2 2 2
2x 2 2x x +1

dy dy dy  1
 = 2xdx
/ 2 2 2
21. (a) y =2x 1 y  =2x 1 y  =2xdx sin y=x +C for
dx 2 2
1 y 1 y
 2 
 x +C .
2 2
 1  1
(b) y(0)=0 sin 0=0 +C 
2
C=0 , so sin
2
y=x and y=sin x ( 2) for   /2 x  /2 .

2
(c) For 1 y to be a real number, we must have  1 y 1 ; that is,  1 y(0) 1 . Thus, the initial
/ 2
value problem y =2x 1 y , y(0)=2 does not have a solution.
 y  y  y
 
/
22. e y +cos x=0
e dy= cos xdx
 e = sin x+C
y= ln (sin x+C) . The solution is periodic,
1
with period 2 . Note that for C>1 , the domain of the solution is R , but for  1<C 1 it is only
defined on the intervals where sin x+C>0 , and it is meaningless for C  1 , since then sin x+C 0 ,
and the logarithm is undefined.

4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

For  1<C<1 , the solution curve consists of concave up pieces separated by intervals on which the
solution is not defined (where sin x+C 0 ). For C=1 , the solution curve consists of concave up
pieces separated by vertical asymptotes at the points where sin x+C=0
sin x= 1 . For C>1 , the
curve is continuous, and as C increases, the graph moves downward, and the amplitude of the
oscillations decreases.

dy sin x 
23. = , y(0)= . So  sin ydy= sin xdx
 cos y= cos x+C
cos y=cos x C . From the
dx sin y 2

initial condition, we need cos =cos 0 C  0=1 C  C=1 , so the solution is cos y=cos x 1 . Note
2
 1
that we cannot take cos of both sides, since that would unnecessarily restrict the solution to the
 1
case where  1 cos x 1
0 cos x , as cos is defined only on  1,1 . Instead we plot the graph
using Maple’s plots [implicitplot] or Mathematica’s Plot [Evaluate[..]].

2
dy x x +1
 ye dy= x
y 2 y
24. =
x +1 dx . We use parts on the LHS with u=y , dv=e dy , and on
dx y
ye
1
 
2 y y
the RHS we use the substitution z=x +1 , so dz=2xdx . The equation becomes ye  e dy= z dz

2
1 2
y
e (y 1)=
3
(
x +1 ) 3/2+C , so we see that the curves are symmetric about the y axis. Every point
1 2
( x,y ) in the plane lies on one of the curves, namely the one for which C=(y 1)e 
y
3
(
x +1 ) 3/2 . For
example,
1 y 4
along the y axis, C=(y 1)e , so the origin lies on the curve with C= . We use Maple’s plots
3 3
command or Plot] in Mathematica to plot the solution curves for various values of C .

5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

4 1 4
It seems that the transitional values of C are  and  . For C< , the graph consists of left and
3 3 3
4
right branches. At C= , the two branches become connected at the origin, and as C increases, the
3
1
graph splits into top and bottom branches. At C= , the bottom half disappears. As C increases
3
further, the graph moves upward, but doesn’t change shape much.

25. (a)
/
x y y =1/y
0 0.5 2
0 0.5  2
0 1 1
0 1  1
0 2 0.5
/
x y y =1/y
0 2  0.5
0 4 0.25

6
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

0 3 0.3
0 0.25 4
0 0.3 3

/
(b) y =1/y  dy/dx=1/y 
1 2
 ydy= dx 
2
ydy=dx  y =x+c y =2(x+c) or y= 2(x+c) .
2

(c)

26. (a)
/ 2
x y y =x /y
1 1 1
 1 1 1
 1  1  1
1  1  1
1 2 0.5
2 1 4
2 2 2
1 0.5 2
0.5 1 0.25
2 0.5 8

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

/ 2 2 1 2 1 3 2 3 1/2
(b) y =x /y ydy=x dx , so y = x +c , or y= x +c .
2 3 1 3

(c)

2
/
2
27. The curves y=kx form a family of parabolas with axis the y axis. Differentiating gives y =2kx ,
2 /
but k=y/x , so y =2y/x . Thus, the slope of the tangent line at any point ( x,y ) on one of the parabolas
is y =2y/x , so the orthogonal trajectories must satisfy y = x / (2y)
2ydy= xdx

2
x +2y =C . This is a family of ellipses.
/ 2
.
y=  x
2
/

/ 2+C
1

2 2 /
28. The curves x  y =k form a family of hyperbolas. Differentiating gives 2x 2y ( dy/dx ) =0 or y =x/y
, the slope of the tangent line at ( x,y ) on one of the hyperbolas. Thus, the orthogonal trajectories must
/
satisfy y = y/x
dy/y= dx/x
ln y = ln x +C
ln x +ln y =C
ln xy =C

1 1 1
C
1
xy =e
xy=C . This is a family of hyperbolas.

8
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

 1 / 1 1 / 1 2
29. Differentiating y= ( x+k ) gives y = , but k=  x , so y = = y . Thus, the
2 y 2
( x+k ) ( 1/y )
3
/ 1 1 2 y 1/3
orthogonal trajectories must satisfy y = =
y dy=dx
=x+C or y= 3(x+C)
2 2 3
 y y

 x /  x x /
30. Differentiating y=ke gives y = ke , but k=ye , so y = y . Thus, the orthogonal trajectories
/ 1 2 1/2
must satisfy y = 1/( y)=1/y
ydy=dx
y =x+C
y= 2(C+x) . This is a family of parabolas
2
with axis the x  axis.

dQ dQ 1
31. From Exercise .2.27,
dt
=12 4Q
 12 4Q
= dt

4
ln 12 4Q =t+C
ln 12 4Q = 4t 4C
 4t 4C  4t  4t  4t

12 4Q =e
12 4Q=Ke
4Q=12 Ke
Q=3 Ae . Q(0)=0
0=3 A
A=3

 4t
Q(t)=3 3e . As t  , Q(t) 3 0=3 (the limiting value).

9
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

dy 1 dy 1 1
32. From Exercise .2.28,
dt
=
50
(y 20)
 y 20
= 
50
dt
ln y 20 =
50
t+C

 t/50  t/50  t/50

y 20=Ke
y(t)=Ke +20 . y(0)=95
95=K+20
K=75
y(t)=75e +20 .

dP dP  kt+C  kt 
=k(M P)
 = ( k)dt
ln P M = kt+C
P M =e
kt
33.
P M=Ae
P=M+Ae
dt P M
. If we assume that performance is at level 0 when t=0 , then P(0)=0
0=M+A
A= M

 kt
P(t)=M Me . lim P(t)=M M 0=M .
t 

dx 1 1/(b a) 1/(b a)
34. (a) =k ( a x ) ( b x ) , a b . Using partial fractions, =  , so
dt (a x)(b x) a x b x
dx 1 b x
 (a x)(b x) = k dt b a (  ln a x +ln b x ) =kt+C  ln a x = ( b a ) ( kt+C ) . The

.
b x b x b x
concentrations A =a x and B =b x cannot be negative, so 0 and = . We now
a x a x a x
b x b
have ln =(b a)(kt+C) . Since x(0)=0 , we get ln =(b a)C . Hence,
a x a
( b a ) kt ( b a ) kt
b x b b x b ( b a ) kt b e  1 ab e  1
ln =(b a)kt+ln  = e  x= = moles /

L.
a x a a x a
be
( b a ) kt
a 1 be /
( b a ) kt
 a

dx dx 1 1
=k(a x) , so  = k dt and
2
(b) If b=a , then =kt+C . Since x(0)=0 , we get C= .
dt 2 a x a
( a x )
2
1 a a kt moles
Thus, a x= and x=a = .
kt+1/a akt+1 akt+1 L
2
a 20a k 2 2 2
Suppose x= C =a/2 when t=20 . Then x(20)=a/2 =  40a k=20a k+a 20a k=a
2 20ak+1
2
1 a t/(20a) at/20 at moles
k= , so x= = = .
20a 1+at/(20a) 1+t/20 t+20 L

dx 1/2 dx 3/2  3/2

35. (a) If a=b , then =k(a x)(b x) becomes =k(a x)  (a x) dx=k dt 
dt dt
 3/2  1/2 2 2 2 4
 (a x) dx= k dt  2(a x) =kt+C 
kt+C
= a x 
kt+C
=a x x(t)=a
2
. The
(kt+C)
4 4 2 4
initial concentration of HBr is 0 , so x(0)=0 0=a  =a C =  C=2/ a ( C is positive
2 2 a
C C
since
10
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

 1/2 4
kt+C=2(a x) >0 ). Thus, x(t)=a .
2
( kt+2/ a)
dx dx dx
 = k dt ( * ) . From the hint, u= b x 
1/2
(b) =k(a x)(b x)  =k dt
dt (a x) b x (a x) b x
2
u =b x 2udu= dx , so
dx  2udu du du 1  1 u
 = = 2 = 2 = 2 tan . So (
(a x) b x a b u
2
u ( )
a b+u
2 2 2
( a b ) +u a b a b
 2  1 b x  2  1 b
* ) becomes tan =kt+C . Now x(0)=0 C= tan and we have
a b a b a b a b
 2  1 b x 2  1 b 2  1 b  1 b x
tan =kt tan  tan  tan =kt 
a b a b a b a b a b a b a b
2  1 b  1 b x
t(x)= tan  tan .
k a b a b a b

2 2
dT dS d T dT 2 dT
36. If S= , then = . The differential equation + =0 can be written as
dr dr 2 2 r dr
dr dr
dS 2 dS  2S dS 2 1 2
+ S=0 . Thus, =  = dr  dS=  dr  ln S = 2ln r +C . Assuming
dr r dr r S r S r
 2
 2ln r+C ln r C  2 1
S=dT /dr>0 and r>0 , we have S=e =e e =r k S= k 
2
r
dT 1 1 1 k
dr
= k
2
dT =
2
k dr   dT =
2
k dr T (r)=
r
+A .
r r r
1
T (1)=15 15= k+A (1) and T (2)=25 25= k+A (2) .
2
Now solve for k and A :  2 (2) + (1)   35= A , so A=35 and k=20 , and T (r)= 20/r+35 .

dC dC dC
37. (a)
dt
=r kC 
dt
= (kC r)  kC r
=  dt  (1/k)ln kC r = t+M 
1
ln kC r = kt+M
2
 kt+M  kt  kt  kt
2
 kC r =e  kC r=M e  kC=M e +r C(t)=M e +r/k . C(0)=C  C =M +r/k
3 3 4 0 0 4
 kt
M =C  r/k C(t)= C  r/k e +r/k .
( )
4 0 0
(b) If C <r/k , then C  r/k<0 and the formula for C(t) shows that C(t) increases and lim C(t)=r/k .
0 0
t 
As t increases, the formula for C(t) shows how the role of C steadily diminishes as that of r/k
0
increases.
11
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

38. (a) Use 1 billion dollars as the x unit and 1 day as the t unit. Initially, there is $10 billion of old
currency in circulation, so all of the $50 million returned to the banks is old. At time t , the amount of
new currency is x(t) billion dollars, so 10 x(t) billion dollars of currency is old. The fraction of
circulating money that is old is 10 x(t) /10 , and the amount of old currency being returned to the
10 x(t)
banks each day is 0.05 billion dollars. This amount of new currency per day is introduced
10
dx 10 x
into circulation, so =  0.05=0.005(10 x) billion dollars per day.
dt 10
dx  dx  0.005t c
(b) =0.005dt  = 0.005dt  ln (10 x)= 0.005t+c 10 x=Ce , where C=e 
10 x 10 x
 0.005t  0.005t
x(t)=10 Ce . From x(0)=0 , we get C=10 , so x(t)=10 1 e . ( )
(c) The new bills make up 90% of the circulating currency when x(t)=0.9 10=9 billion dollars.
 0.005t  0.005t  0.005t
(
9=10 1 e ) 0.9=1 e  e =0.1  0.005t= ln 10 t=200ln 10 460.517 days
 1.26 years.

39. (a) Let y(t) be the amount of salt (in kg) after t minutes. Then y(0)=15 . The amount of liquid in
the tank is 1000 L at all times, so the concentration at time t (in minutes) is y(t)/1000 kg/L and
dy y(t) kg L y(t) kg dy 1 t
= 10 = . =  dt  ln y= +C , and y(0)=15
dt 1000 L min 100 min y 100 100
t y t y  t/100  t/100
ln 15=C , so ln y=ln 15 . It follows that ln = and =e , so y=15e kg.
100 15 100 15
 20/100  0.2
(b) After 20 minutes, y=15e =15e  12.3 kg.

40. (a) If y(t) is the amount of salt (in kg) after t minutes, then y(0)=0 and the total amount of liquid
in the tank remains constant at 1000 L.
dy kg L kg L y(t) kg L
= 0.05 5 + 0.04 10  15
dt L min L min 1000 L min
130 3y kg
= 0.25+0.40 0.015y=0.65 0.015y=
200 min
dy dt 1 1 1
so  = and  ln 130 3y = t+C ; since y(0)=0 , we have  ln 130=C ,
130 3y 200 3 200 3
1 1 1 3  3t/200
so 
3
ln 130 3y =
200 3
t ln 130 ln 130 3y =
200
t+ln 130=ln 130e , and ( )
 3t/200
130 3y =130e . Since y is continuous, y(0)=0 , and the right hand side is never zero, we
 3t/200 130  3t/200
deduce that 130 3y is always positive. Thus, 130 3y=130e and y=
3
1 e kg. ( )
130  3 60/200 130  0.9
(b) After one hour, y=
3
1 e ( =
3
1 e )
 25.7 kg. ( )
Note: As t  ,
12
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

130 1
y(t) =43 kg.
3 3

/
41. Assume that the raindrop begins at rest, so that v(0)=0 . dm/dt=km and ( mv ) =gm
dv
mv +vm =gm mv +v(km)=gm v +vk=g  dv/dt=g kv  = dt  ( 1/k ) ln g kv =t+C 
/ / / /
g kv
 kt  kt  kt
ln g kv = kt kC  g kv=Ae . v(0)=0 A=g . So kv=g ge  v= ( g/k ) 1 e ( ) . Since k>0 , as
 kt
t  ,e  0 and therefore, lim v(t)=g/k .
t 

dv dv k k
42. (a) m = kv = dt  ln v = t+C . Since v(0)=v , ln v =C . Therefore,
dt v m m 0 0
v k v  kt/m  kt/m
ln = t =e  v(t)= v e . The sign is + when t=0 , and we assume v is
v m v 0
0 0
mv
 kt/m  kt/m 0  kt/m /
continuous, so that the sign is + for all t . Thus, v(t)=v e . ds/dt=v e  s(t)= e +C .
0 0 k
mv mv mv
From s(0)=s , we get s =
0 0 k
0
+C
/ /
, so C =s +
0 k
0
and s(t)=s +
0 k
0
( 1 e kt/m) . The distance
mv
0
traveled from time 0 to time t is s(t) s , so the total distance traveled is lim s(t) s = .
0
t 
0 k
 kt/m
Note: In finding the limit, we use the fact that k>0 to conclude that lim e =0 .
t 
dv 2 dv k  1 kt 1 kt 1 1 kt 1
(b) m = kv  = dt  = +C  =  C . Since v(0)=v , C= and = +
dt 2 m v m v m 0 v v m v
v 0 0
mv mv kv dt
1 ds m m
s(t)= 
0 0 0 /
. Therefore, v(t)= = . =  = ln kv t+m +C .
kt/m+1/v kv0t+m dt kv0t+m k kv t+m k 0
0
0
m / / m
Since s(0)=s , we get s = ln m+C  C =s  ln m
0 0 k 0 k
kv t+m
m m 0
s(t)=s + ln kv t+m  ln m =s + ln
( ) . We can rewrite the formulas for v(t) and
0 k 0 0 k m
v kv
0 m 0
s(t) as v(t)= and s(t)=s + ln 1+ t .
1+ kv /m t ( 0
0 k
) m
Remarks: This model of horizontal motion through a resistive medium was designed to handle the
case in which
13
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

2
v >0 . Then the term  kv representing the resisting force causes the object to decelerate. The absolute
0
value in the expression for s(t) is unnecessary (since k , v , and m are all positive), and lim s(t)= .
0
t 
2
In other words, the object travels infinitely far. However, lim v(t)=0 . When v <0 , the term  kv
0
t 
increases the magnitude of the object’s negative velocity. According to the formula for s(t) , the
position of the object approaches   as t approaches m/k  v : lim s(t)=  . Again the
( ) 0
t ( 0)
 m/ kv
object travels infinitely far, but this time the feat is accomplished in a finite amount of time. Notice
also that lim v(t)=  when v <0 , showing that the speed of the object increases without limit.
0
t ( 0)
 m/ kv

43. (a) The rate of growth of the area is jointly proportional to A(t) and M A(t) ; that is, the rate is
proportional to the product of those two quantities. So for some constant k , dA/dt=k A (M A) . We
are interested in the maximum of the function dA/dt (when the tissue grows the fastest), so we
differentiate, using the Chain Rule and then substituting for dA/dt from the differential equation:
d dA dA 1  1/2 dA 1  1/2 dA
= k A ( 1) +(M A) A = kA  2A+ ( M A)
dt dt dt 2 dt 2 dt
1  1/2 1 2
= kA k A(M A) M 3A = k (M A)(M 3A)
2 2
This is 0 when M A=0 and when M 3A=0
A(t)=M/3 . This represents a maximum by the First
d dA
Derivative Test, since goes from positive to negative when A(t)=M/3 .
dt dt
M kt 2
Ce  1
(b) From the CAS, we get A(t)=M . To get C in terms of the initial area A and the
M kt 0
Ce +1
C 1 2
maximum area M , we substitute t=0 and A=A =A(0) : A =M
(C+1) A =(C 1) M

0 0 C+1 0

C A + A =C M  M
M+ A =C M  C A

0 0 0 0

M+ A
0
M+ A =C M A
C= . (Notice that if A =0 , then C=1 .)
0 0 0
M A
0

2
dv dv dx dv mgR
44. (a) According to the hint we use the Chain Rule: m =m  =mv = 
dt dx dt dx 2
( x+R)

14
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.3 Separable Equations

2
2 2 2 v 2
 gR dx v gR gR 1 2
 vdv=
0
 = +C . When x=0 , v=v , so = +C  C= v  gR
2 2 x+R 0 2 0+R 2 0
( x+R)
2
1 2 1 2 gR
v v=  gR . Now at the top of its flight, the rocket’s velocity will be 0 , and its height will
2 2 0 x+R
2
2 v 2
1 2 gR 0 R R ( R+h ) gRh 2gRh
be x=h . Solving for v :  v=  gR =g  + =  v= .
0 2 0 h+R 2 R+h R+h R+h 0 R+h
2gRh 2gR
(b) v =lim v =lim =lim = 2gR
e
h 
0
h  R+h h  ( R/h ) +1
2
(c) v = 2 32ft/s  3960mi 5280ft/mi  36 , 581 ft / s  6.93 mi / s
e

15
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

1 dP dP
1. The relative growth rate is =0.7944 , so =0.7944P and, by Theorem 2,
P dt dt
0.7944t 0.7944t 0.7944 ( 6 )
P(t)=P(0)e =2e . Thus, P(6)=2e  234.99 or about 235 members.

kt kt 1
2. (a) By Theorem 2, P(t)=P(0)e =60e . In 20 minutes ( hour), there are 120 cells, so
3
1
P
3
k/3
=60e =120
k/3
e =2 k/3=ln 2 ( 3)
k=3ln 2=ln 2 =ln 8 .
( ln 8 ) t t
(b) P(t)=60e =60 8
8 24
(c) P(8)=60 8 =60 2 =1 , 006 , 632 , 960
/
(d) dP/dt=kP P (8)=kP(8)=(ln 8)P(8) 2.093 billion cells / h
t t ln ( 1000/3)
(e) P(t)=20 , 000 60 8 =20 , 000 8 =1000/3 tln 8=ln ( 1000/3)  t=  2.79 h
ln 8

kt kt k(3) 3k 8000
3. (a) By Theorem 2, y(t)=y(0)e =500e . Now y(3)=500e =8000 e =  3k=ln 16
500
( ln 16 ) t/3 t/3
k= ( ln 16 ) /3 . So y(t)=500e =500 16
4/3
(b) y(4)=500 16  20 , 159
1
/
(c) dy/dt=ky y (4)=ky(4)= ln 16 500 16
3
4/3
(
 18 , 631 cells / h )
t/3 t/3 1
(d) y(t)=500 16 =30 , 000 16 =60 tln 16=ln 60 t=3(ln 60)/(ln 16) 4.4h
3

kt 2k 8k
4. (a) y(t)=y(0)e y(2)=y(0)e =600 , y(8)=y(0)e =75 , 000 . Dividing these equations, we get
8k 2k 6k 3 3 1
e /e =75 , 000/600 e =125 6k=ln 125=ln 5 =3ln 5 k= ln 5= ln 5 . Thus,
6 2
2k ln 5 600
y(0)=600/e =600/e = =120 .
5
kt (ln 5)t/2 t/2
(b) y(t)=y(0)e =120e or y=120 5
5/2
(c) y(5)=120 5 =120 25 5=3000 5 6708 bacteria.
t/2 / t/2 1 t/2
(d) y(t)=120 5  y (t)=120 5  ln 5 =60 ln 5 5 .
2
/ 5/2
y (5)=60 ln 5 5 =60 ln 5 25 5 5398 bacteria / hour.
(ln 5)t/2 (ln 5)t/2 5000 5000
(e) y(t)=200 , 000  120e =200 , 000 e =  (ln 5)t/2=ln 
3 3
5000
t= 2ln /ln 5 9.2 h.
3
1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

5. (a) Let the population (in millions) in the year t be P(t) . Since the initial time is the year 1750, we
k ( t 1750 )
substitute t 1750 for t in Theorem 2, so the exponential model gives P(t)=P(1750)e . Then
k(1800 1750) 980 k(50) 980 1 980
P(1800)=980=790e  =e  ln =50k k= ln  0.0043104 . So with
790 790 50 790
k(1900 1750) k(1950 1750)
this model, we have P(1900)=790e  1508 million, and P(1950)=790e  1871
million. Both of these estimates are much too low.
k(t 1850)
(b) In this case, the exponential model gives P(t)=P(1850)e 
k(1900 1850) 1650 1 1650
P(1900)=1650=1260e  ln =k(50) k= ln  0.005393 . So with this
1260 50 1260
k(1950 1850)
model, we estimate P(1950)=1260e  2161 million. This is still too low, but closer than the
estimate of P(1950) in part (a).
k(t 1900) k(1950 1900)
(c) The exponential model gives P(t)=P(1900)e  P(1950)=2560=1650e 
2560 1 2560
ln =k(50) k= ln  0.008785 . With this model, we estimate
1650 50 1650
k(2000 1900)
P(2000)=1650e  3972 million. This is much too low. The discrepancy is explained by the
fact that the world birth rate (average yearly number of births per person) is about the same as always,
whereas the mortality rate (especially the infant mortality rate) is much lower, owing mostly to
advances in medical science and to the wars in the first part of the twentieth century. The exponential
model assumes, among other things, that the birth and mortality rates will remain constant.

6. (a) Let P(t) be the population (in millions) in the year t . Since the initial time is the year 1900, we
k(t 1900)
substitute t 1900 for t in Theorem 2, and find that the exponential model gives P(t)=P(1900)e
k(1910 1900) 1 92
 P(1910)=92=76e  k= ln  0.0191 . With this model, we estimate
10 76
k(2000 1900)
P(2000)=76e  514 million. This estimate is much too high. The discrepancy is explained
by the fact that, between the years 1900 and 1910, an enormous number of immigrants (compared to
the total population) came to the United States. Since that time, immigration (as a proportion of total
population) has been much lower. Also, the birth rate in the United States has declined since the turn
of the century. So our calculation of the constant k was based partly on factors which no longer exist.
(b) Substituting t 1980 for t in Theorem 2, we find that the exponential model gives
k(t 1980) k(1990 1980) 1 250
P(t)=P(1980)e  P(1990)=250=227e  k= ln  0.00965 . With this model,
10 227
k(2000 1980)
we estimate P(2000)=227e  275.3 million. This is quite accurate. The further estimates are
30k 40k
P(2010)=227e  303 million and P(2020)=227e  334 million.
(c)

2
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

The model in part (a) is quite inaccurate after 1910 (off by 5 million in 1920 and 12 million in 1930).
The model in part (b) is more accurate (which is not surprising, since it is based on more recent
information).

dy  0.0005t  0.0005t
7. (a) If y= N O then by Theorem 2, = 0.0005y y(t)=y(0)e =Ce .
2 5 dt
 0.0005t  0.0005t
(b) y(t)=Ce =0.9C  e =0.9  0.0005t=ln 0.9 t= 2000ln 0.9 211 s

kt kt
8. (a) The mass remaining after t days is y(t)=y(0)e =800e . Since the half life is 5.0 days,
5k 5k 1
y(5)=800e =400 e = 
2
1  (ln 2)t/5  t/5
5k=ln  k= (ln 2)/5 , so y(t)=800e =800 2 .
2

 30/5
(b) y(30)=800 2 =12.5 mg

 (ln 2)t/5 t 1 ln 800

(c) 800e =1  (ln 2) =ln = ln 800 t=5  48 days
5 800 ln 2

(d)

kt kt
9. (a) If y(t) is the mass (in mg) remaining after t years, then y(t)=y(0)e =100e .
30k 1 30k 1  (ln 2)t/30  t/30
y(30)=100e = (100) e =  k= (ln 2)/30 y(t)=100e =100 2
2 2
 100/30
(b) y(100)=100 2  9.92 mg
 (ln 2)t/30
(c) 100e =1
3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

1 ln 0.01
 (ln 2)t/30=ln  t= 30  199.3 years
100 ln 2

kt 3k
10. (a) If y(t) is the mass after t days and y(0)=A , then y(t)=Ae . y(3)=Ae =0.58A
3k 1 ( ln 0.58 ) t/3 1
e =0.58 3k=ln 0.58 k= ln 0.58 . Then Ae = A
3 2
( ln 0.58 ) t/3 1 (ln 0.58)t 1 3ln 2
ln e =ln  =ln , so the half life is t=  3.82 days.
2 3 2 ln 0.58
( ln 0.58 ) t/3 (ln 0.58)t 1 3ln 10
(b) Ae =0.10A =ln  t=  12.68 days
3 10 ln 0.58

 kt
and k is determined by using the half
11. Let y(t) be the level of radioactivity. Thus, y(t)=y(0)e
1
ln
1  k(5730) 1  5730k 1 1 2 ln 2
life: y(5730)= y(0) y(0)e = y(0)  e =   5730k=ln  k= = . If
2 2 2 2 5730 5730
14  t ( ln 2 ) /5730 tln 2
74% of the C remains, then we know that y(t)=0.74y(0) 0.74=e  ln 0.74= 
5730
5730(ln 0.74)
t=  2489 2500 years.
ln 2

dy 2x
12. From the information given, we know that =2y y=Ce by Theorem 2. To calculate C we
dx
2(0) 2x
use the point ( 0,5) : 5=Ce  C=5 . Thus, the equation of the curve is y=5e .

dT dT
13. (a) Using Newton’s Law of Cooling, =k(T  T ) , we have =k(T  75) .
dt s dt
Now let y=T  75 , so y(0)=T (0) 75=185 75=110 , so y is a solution of the initial value
kt kt
problem dy/dt=ky with y(0)=110 and by Theorem 2 we have y(t)=y(0)e =110e .
30k 30k 75 15 1 15
y(30)=110e =150 75 e = =  k= ln ,
110 22 30 22
1 15 45 15
tln ln
30 22 30 22  
so y(t)=110e and y(45)=110e  62 F. Thus, T (45) 62+75=137 F.
1 15 1 15
tln tln
30 22 30 22 25 1 15 25
(b) T (t)=100 y(t)=25 . y(t)=110e =25 e =  tln =ln 
110 30 22 110
25
30ln
110
t=  116 min.
15
ln
22
4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

dT
14. (a) Let T (t)= temperature after t minutes. Newton’s Law of Cooling implies that =k(T  5) .
dt
dy kt kt kt
Let y(t)=T (t) 5 . Then =ky , so y(t)=y(0)e =15e  T (t)=5+15e 
dt
k k 7 7 ln ( 7/15 ) t 2ln ( 7/15 ) 
T (1)=5+15e =12 e =  k=ln , so T (t)=5+15e and T (2)=5+15e  8.3 C.
15 15
1
ln
ln ( 7/15 ) t ln ( 7/15 ) t 1 7 1 15
(b) 5+15e =6 when e =  ln t=ln  t=  3.6 min.
15 15 15 7
ln
15

dT dy kt
15. =k(T  20) . Letting y=T  20 , we get =ky , so y(t)=y(0)e . y(0)=T (0) 20=5 20= 15 , so
dt dt
25k 25k 25k 25k 2
y(25)=y(0)e = 15e , and y(25)=T (25) 20=10 20= 10 , so  15e = 10 e = . Thus,
3
2 1 2 kt (1/25)ln (2/3)t 25k 2
25k=ln and k= ln , so y(t)=y(0)e = 15e . More simply, e = 
3 25 3 3
k 2 1/25 kt 2 t/25 2 t/25
e=  e =  y(t)= 15 .
3 3 3
2 50/25 2 2 20 
(a) T (50)=20+y(50)=20 15 =20 15 =20 =13.3 C
3 3 3
2 t/25 2 t/25 2 t/25 1
(b) 15=T (t)=20+y(t)=20 15  15 =5 = 
3 3 3 3
(t/25)ln
2
3
=ln
1
3
 t=25ln
1
3
ln
2
3 /
 67.74 min.

dT dy kt
16. =k(T  20) . Let y=T  20 . Then =ky , so y(t)=y(0)e . y(0)=T (0) 20=95 20=75 , so
dt dt
kt dT  dy
y(t)=75e . When T (t)=70 , = 1 C / min. Equivalently, = 1 when y(t)=50 . Thus,
dt dt
dy kt
 1= =ky(t)=50k and 50=y(t)=75e . The first relation implies k= 1/50 , so the second relation says
dt
 t/50  t/50 2 2 2
50=75e . Thus, e =   t/50=ln  t= 50ln  20.27 min.
3 3 3

kh kh
17. (a) Let P(h) be the pressure at altitude h . Then dP/dh=kP P(h)=P(0)e =101.3e .
1000k 87.14 1 87.14
P(1000)=101.3e =87.14 1000k=ln  k= ln  P(h)=101.3
101.3 1000 101.3
1 87.14 87.14
hln 3ln
1000 101.3 101.3
e , so P(3000)=101.3e  64.5 kPa.
5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

6187 87.14
ln
1000 101.3
(b) P(6187)=101.3e  39.9 kPa

r nt
18. (a) Using A=A 1+ with A =500 , r=0.14 , and t=2 , we have:
0 n 0

0.14 1 2
(i) Annually: n=1 ; A=500 1+ =$649.80
1
0.14 4 2
(ii) Quarterly: n=4 ; A=500 1+ =$658.40
4
0.14 12 2
(iii) Monthly: n=12 ; A=500 1+ =$660.49
12
0.14 365 2
(iv) Daily: n=365 ; A=500 1+ =$661.53
365
0.14 365 24 2
(v) Hourly: n=365 24 ; A=500 1+ =$661.56
365 24
(vi) Continuously: (0.14)2
A=500e =$661.56

(b)
A (2)=$661.56 , A (2)=$610.70 , and A (2)=$563.75 .
0.14 0.10 0.06

r nt
19. (a) Using A=A 1+ with A =3000 , r=0.05 , and t=5 , we have:
0 n 0

0.05 1 5
(i) Annually: n=1 ; A=3000 1+ =$3828.84
1
0.05 2 5
(ii) Semiannually: n=2 ; A=3000 1+ =$3840.25
2
0.05 12 5
(iii) Monthly: n=12 ; A=3000 1+ =$3850.08
12
0.05 52 5
(iv) Weekly: n=52 ; A=3000 1+ =$3851.61
52

6
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

0.05 365 5
(v) Daily: n=365 ; A=3000 1+ =$3852.01
365
(vi) Continuously: (0.05)5
A=3000e =$3852.08

(b) dA/dt=0.05A and A(0)=3000 .

0.06t 0.06t 50
20. (a) A e =2A  e =2 0.06t=ln 2 t= ln 2 11.55 , so the investment will double in
0 0 3
about 11.55 years.
t 0.06t
(b) The annual interest rate in A=A (1+r) is r . From part (a), we have A=A e . These amounts
0 0
t 0.06t 0.06 0.06
must be equal, so (1+r) =e  1+r=e  r=e  1 0.0618=6.18% , which is the equivalent
annual interest rate.

dP m m
dy dP
21. (a) =kP m=k P, so = . Let y=P
and the differential equation becomes
dt k k
dt dt
dy kt m m kt m m kt
=ky . The solution is y=y e  P = P e  P(t)= + P  e .
dt 0 k 0 k k 0 k
m
(b) Since k>0 , there will be an exponential expansion  P  >0 m<kP .
0 k 0
m m
(c) The population will be constant if P  =0 m=kP . It will decline if P  <0 m>kP .
0 k 0 0 k 0
(d) P =8 , 000 , 000 , k=   =0.016 , m=210 , 000 m>kP ( =128 , 000 ), so by part (c), the
0 0
population was declining.

 c  c
 c y  c y
dy 1+c  1 c y 0 y 0
22. (a) =ky  y dy=k dt =kt+C . Since y(0)=y , we have C= . Thus, =kt+
dt  c 0  c  c  c
c
y y
 c  c c 1 0 0
, or y =y  ckt . So y =  c
= c
and y(t)= .
0 c 1/c
y  ckt 1 cy kt 1 cy kt
0 0 0
c 1 1
(b) y(t)
as 1 cy kt 0 , that is, as t c
. Define T = c
. Then lim y(t)=
.
0 
cy k cy k t T
0 0
(c) According to the data given, we have c=0.01 , y(0)=2 , and y(3)=16 , where the time t is given in
y 1 c
0
months. Thus, y =2 and 16=y(3)= . Since T = c , we will solve for cy k .
0 c 1/c 0
1 cy k 3 cy k
0 0

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.4 Exponential Growth and Decay

1 0.01  0.01 1  0.01

16=
2
c 100
 1 3cy k=
c
0 8
=8 
c
cy k=
0 3
(
1 8 ) . Thus, doomsday occurs
1 3cy k
0
1 3
when t=T = =  145.77 months or 12.15 years.
c  0.01
cy k 1 8
0

8
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

2
1. (a) dP/dt=0.05P 0.0005P =0.05P(1 0.01P)=0.05P ( 1 P/100 ) . Comparing to Equation 1,
dP/dt=kP ( 1 P/K ) , we see that the carrying capacity is K=100 and the value of k is 0.05 .
(b) The slopes close to 0 occur where P is near 0 or 100 . The largest slopes appear to be on the line
P=50 . The solutions are increasing for 0<P <100 and decreasing for P >100 .
0 0

(c)
All of the solutions approach P=100 as t increases. As in part (b), the solutions differ since for
0<P <100 they are increasing, and for P >100 they are decreasing. Also, some have an IP and some
0 0
don’t. It appears that the solutions which have P =20 and P =40 have inflection points at P=50 .
0 0

(d) The equilibrium solutions are P=0 (trivial solution) and P=100 . The increasing solutions move
away from P=0 and all nonzero solutions approach P=100 as t  .

2. (a) K=6000 and k=0.0015 dP/dt=0.0015P ( 1 P/6000 ) .

(b)
All of the solution curves approach 6000 as t  .

(c)
The curves with P =1000 and P =2000 appear to be concave upward at first and then concave
0 0
downward. The curve with P =4000 appears to be concave downward everywhere. The curve with
0
P =8000 appears to be concave upward everywhere. The inflection points are where the population
0
grows the fastest.

1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

(d) See the solution to Exercise .2.25 for a possible program to calculate P(50) . We find that
P(50) 1064 .
K 6000
(e) Using Equation 4 with K=6000 , k=0.0015 , and P =1000 , we have P(t)= =
0  kt  0.0015t
1+Ae 1+Ae
K P 6000 1000
0 6000
, where A= = =5 . Thus, P ( 50 ) =  1064.1 , which is extremely
P 1000  0.0015 ( 50 )
0 1+5e
close to the estimate obtained in part (d).

(f)

The curves are very similar.

dy y K K y(0) 7
3. (a) =ky 1  y(t)= with A= . With K=8 10 , k=0.71 , and
dt K  kt y(0)
1+Ae
7 7
7 8 10 8 10 7
y(0)=2 10 , we get the model y(t)= , so y(1)=  3.23 10 kg.
 0.71t  0.71
1+3e 1+3e
7
7 8 10 7  0.71t  0.71t 1 1 ln 3
(b) y(t)=4 10  =4 10  2=1+3e  e =   0.71t=ln  t=  1.55
 0.71t 3 3 0.71
1+3e
years

4. (a)
From the graph, we estimate the carrying capacity K for the yeast population to be 680 .
1 dP 1 39 18 7
(b) An estimate of the initial relative growth rate is = = =0.583 .
P dt 18 2 0 12
0
7t/12 680
(c) An exponential model is P(t)=18e . A logistic model is P(t)= , where
 7t/12
1+Ae
680 18 331
A= = .
18 9
2
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

(d)
Time in Observed Exponential Logistic
Hours Values Model Model
0 18 18 18
2 39 58 55
4 80 186 149
6 171 596 322
8 336 1914 505
10 509 6147 614
12 597 19 , 739 658
14 640 63 , 389 673
16 664 203 , 558 678
18 672 653 , 679 679

The exponential model is a poor fit for anything beyond the first two observed values. The logistic
model varies more for the middle values than it does for the values at either end, but provides a good
general fit, as shown in the figure.
680
(e) P(7)=  420 yeast cells
331  7 ( 7/12 )
1+ e
9

5. (a) We will assume that the difference in the birth and death rates is 20 million / year. Let t=0
correspond to the year 1990 and use a unit of 1 billion for all calculations.
1 dP 1 1
k = (0.02)= , so
P dt 5.3 265

dP P 1 P
=kP 1 = P 1 ,Pinbillions
dt K 265 100

K P 100 5.3 947

0 K 100
(b) A= = =  17.8679 . P(t)= = , so P(10) 5.49
P 5.3 53  kt 947  ( 1/265 ) t
0 1+Ae 1+ e
53 3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

billion.
(c) P(110) 7.81 , and P(510) 27.72 . The predictions are 7.81 billion in the year 2100 and 27.72
billion in 2500.
50
(d) If K=50 , then P(t)= . So P(10) 5.48 , P(110) 7.61 , and P(510) 22.41 . The
447  ( 1/265 ) t
1+ e
53
predictions become 5.48 billion in the year 2000, 7.61 billion in 2100, and 22.41 billion in the year
2500.

6. (a) If we assume that the carrying capacity for the world population is 100 billion, it would seem
reasonable that the carrying capacity for the U.S. is 3   5 billion by using current populations and
simple proportions. We will use K=4 billion or 4000 million. With t=0 corresponding to 1980, we
have

4000 4000
P(t)= =
4000 250  kt  kt
1+ e 1+15e
250

160
 1
4000  10k 4000  10k 11 149
(b) P(10)=275 =275 1+15e =  e =   10k=ln 
 10k 275 15 165
1+15e
1 149
k= ln  0.01019992 .
10 165
(c) 2100 1990=110 and P(110) 680 million.
2200 1990=210 and P(210) 1449 million, or about 1.4 billion.
4000  kt 40  kt 37 1 37
(d) P(t)=300 =300 1+15e =  e =   kt=ln 
 kt 3 3 15 45
1+15e
37
ln
45
t=10  19.19 19 . So we predict that the U.S. population will exceed 300 million in the
149
ln
165
year 1990+19=2009 .

dy
7. (a) Our assumption is that =ky(1 y) , where y is the fraction of the population that has heard the
dt
rumor.
dP P P dP dy
(b) Using the logistic equation (1), =kP 1 , we substitute y= , P=Ky , and =K ,
dt K K dt dt
to obtain
4
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

dy dy K
K =k(Ky)(1 y)
=ky(1 y) , our equation in part (a). Now the solution to (1) is P(t)= ,
dt dt  kt
1+Ae
K P
0 K
where A= . We use the same substitution to obtain Ky=
P K Ky
0 0  kt
1+ e
Ky
0
y
0
 y= .
 kt
0 ( )e
y + 1 y
0
Alternatively, we could use the same steps as outlined in ‘‘The Analytic Solution,’’ following
Example 2.
80 1
(c) Let t be the number of hours since 8 A.M. Then y =y(0)= =0.08 and y(4)= , so
0 1000 2
1 0.08  4k  4k 0.08 2  k 2 1/4
=y(4)= . Thus, 0.08+0.92e =0.16 , e = = , and e = , so
2  4k 0.92 23 23
0.08+0.92e
0.08 2
y= = . Solving this equation for t , we get
t/4 t/4
0.08+0.92 ( 2/23) 2+23 ( 2/23)
2 t/4 2 t/4 2 2y 2 t/4 2 1 y 2 t/4 1 1 y
2y+23y =2 =  =  = . It follows
23 23 23y 23 23 y 23 y
t ln (1 y)/y ln ( (1 y)/y )
that  1= , so t=4 1+ .
4 2 2
ln ln
23 23
1 y 1 ln 9
When y=0.9 , = , so t=4 1  7.6 h or 7 h 36 min. Thus, 90% of the population
y 9 2
ln
23
will have heard the rumor by 3:36 P.M.

8. (a) P(0)=P =400 , P(1)=1200 and K=10 , 000 . From the solution to the logistic differential
0
PK
0 400 ( 10,000 ) 10,000
equation P(t)= , we get P= = . P(1)=1200
 kt  kt  kt
P + K P
0 ( 0 )e 400+(9600)e 1+24e
 k 100 k 288 36 10,000 10,000
1+24e =  e=  k=ln . So P= = .
12 88 11  tln ( 36/11 ) t
1+24e 1+24 ( 11/36 )
11 t 11 1
10,000
(b) 5000=  24 =1 tln =ln  t 2.68 years.
t 36 36 24
1+24 ( 11/36 )

5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

dP P
9. (a) =kP 1 
dt K
2
d P 1 dP P dP dP P P
2
= k P  + 1 =k  +1
K dt K dt dt K K
dt
P 2P 2 P 2P
= k kP 1 1 =k P 1 1
K K K K

/ / / / /
(b) P grows fastest when P has a maximum, that is, when P =0 . From part (a), P =0
P=0 ,
/ /
P=K , or P=K/2 . Since 0<P<K , we see that P =0
P=K/2 .

10.

10P
0
First we keep k constant (at 0.1 , say) and change P in the function P= . (Notice
0  0.1t
(
P + 10 P
0 0 )e
that P is the P intercept.) If P =0 , the function is 0 everywhere. For 0<P <5 , the curve has an
0 0 0
inflection point, which moves to the right as P decreases. If 5<P <10 , the graph is concave down
0 0
everywhere. (We are

considering only t 0 .) If P =10 , the function is the constant function P=10 , and if P >10 , the
0 0
function decreases. For all P  0 , lim P=10 .
0
t 

10
Now we instead keep P constant (at P =1 ) and change k in the function P= . It seems that
0 0  kt
1+9e
as k increases, the graph approaches the line P=10 more and more quickly. (Note that the only
difference in the shape of the curves is in the horizontal scaling; if we choose suitable x  scales, the
6
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

graphs all look the same.)

11. (a) The term  15 represents a harvesting of fish at a constant rate    in this case, 15 fish / week.
This is the rate at which fish are caught.

(b)

(c) From the graph in part (b), it appears that P(t)=250 and P(t)=750 are the equilibrium solutions.
We confirm this analytically by solving the equation dP/dt=0 as follows: 0.08P ( 1 P/1000 )  15=0
0.08P 0.00008P  15=0
2
( 2
 0.00008 P  1000P+187,500 =0 ) (P 250)(P 750)=0 P=250 or 750 .

(d)
For 0<P <250 , P(t) decreases to 0 . For P =250 , P(t) remains constant. For 250<P <750 , P(t)
0 0 0
increases and approaches 750 . For P =750 , P(t) remains constant. For P >750 , P(t) decreases and
0 0
approaches 750 .

dP P 100,000 dP 100,000
(e)
dt
=0.08P 1
1000
 15

8

dt
2
= 0.08P 0.00008P  15  ( 8
)
 12 ,
dP 2 dP 1
500 =P  1000P+187 , 500
= dt

dt (P 250)(P 750) 12,500

 1/500 1/500 1 1 1 1
+
P 250 P 750
dP=
12,500
dt

P 250 P 750
dP=
25
dt

1 P 250 1 P 250 t/25+C t/25

ln P 250  ln P 750 = t+C
ln = t+C
=e =ke

25 P 750 25 P 750
t/25
P 250 t/25 t/25 t/25 t/25 t/25 250 750ke
=ke
P 250=Pke  750ke
P Pke =250 750ke
P(t)= . If t=0
P 750 t/25
1 ke
and P=200 , then
250 750k 1
200=
200 200k=250 750k
550k=50
k= . Similarly, if t=0 and P=300 , then
1 k 11

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

k=
1
. Simplifying P with these two values of k gives us P(t)=
250 3e  11 ( t/25
) and
9 t/25
e  11

P(t)=
750 e ( t/25+3) .
t/25
e +9

12. (a)

c=0

c=10

c=20

8
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

c=21

c=25

c=30

(b) For 0 c 20 , there is at least one equilibrium solution. For c>20 , the population always dies
out.
2
dP 2 dP  0.08 (0.08)  4( 0.00008)( c)
(c) =0.08P 0.00008P  c . =0
P= , which has at least
dt dt 2( 0.00008)
one solution when the discriminant is nonnegative  0.0064 0.00032c 0
c 20 . For 0 c 20 ,
there is at least one value of P such that dP/dt=0 and hence, at least one equilibrium solution. For
c>20 , dP/dt<0 and the population always dies out.
(d) The weekly catch should be less than 20 fish per week.

dP P m
13. (a) =(kP) 1 1 . If m<P<K , then dP/dt=(+)(+)(+)=+ P is increasing.
dt K P
If 0<P<m , then dP/dt=(+)(+)( )=  P is decreasing.
(b)

9
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

k=0.08 , K=1000 , and m=200

dP P 200
=0.08P 1 1
dt 1000 P

For 0<P <200 , the population dies out. For P =200 , the population is steady. For 200<P <1000 , the
0 0 0
population increases and approaches 1000 . For P >1000 , the population decreases and approaches
0
1000 .
The equilibrium solutions are P(t)=200 and P(t)=1000 .
dP P m K P P m k dP k
(c) =kP 1 1 =kP = (K P)(P m)
= dt
dt K P K P K (K P)(P m) K
.
1 A B
By partial fractions, = + , so A(P m)+B(K P)=1 .
(K P)(P m) K P P m
1 1 1 1 1 k
If P=m , B=
K m
; if P=K , A=
K m
, so
K m
+
K P P m
dP=
K
dt
1 k 1 P m k P m k
(  ln K P +ln P m ) = t+M ln = t+M ln =(K m) t+M
K m K K m K P K K P K 1
P m ( K m ) ( k/K ) t

=De .
K P
P m P m P0 m ( K m ) ( k/K ) t
0
Let t=0 : =D . So = e . Solving for P , we get
K P K P K P
0 0
( K m ) ( )t
k/K
m K P +K P  m e
( ) ( )
0 0
P(t)= .
( K m ) ( k/K ) t
K P + P  m e
( )
0 0
(d) If P <m , then P  m<0 . Let N(t) be the numerator of the expression for P(t) in part (c). Then
0 0
( K m ) ( k/K ) t
N(0)=P ( K m ) >0 , and P  m<0
lim K P  m e
( ) =   lim N(t)=  . Since N is
0 0 0
t  t 
continuous, there is a number t such that N(t)=0 and thus P(t)=0 . So the species will become extinct.

14. (a)
10
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

dP K dP K dP
=cln P = cdt . Let u=ln =ln K ln P du=
dt P Pln ( K/P ) P P
du  ( ct+D )  ( ct+D )
  =ct+D ln u = ct D u =e  ln ( K/P ) =e 
u
 ( ct+D )  D
ln ( K/P ) = e . Letting t=0 , we get ln K/P = e , so ( 0 )  ct  ct
 ct D  ct  D  ct

(
ln K/P
0 )e 
(
 ln K/P
0 )e
ln ( K/P ) = e = e e =ln K/P e
 ct
( 0 ) K/P=e P(t)=Ke , c 0 .

(b) lim P(t)=lim Ke

(
 ln K/P
0 )e =Ke
(
 ln K/P 0
0 ) =Ke =K
0

t  t 

(c)
The graphs look very similar. For the Gompertz function, P(40) 732 , nearly the same as the logistic
function. The Gompertz function reaches P=900 at t 61.7 and its value at t=80 is about 959 , so it
doesn’t increase quite as fast as the logistic curve.

dP K
(d) =cln P=cP(ln K ln P)
dt P
2
d P 1 dP dP dP K
2
= c P  +(ln K ln P) =c  1+ln
P dt dt dt P
dt
= c cln ( K/P ) P ln ( K/P )  1 =c2Pln ( K/P ) ln ( K/P )  1
/ / / / / /
Since 0<P<K , P =0
ln ( K/P ) =1
K/P=e
P=K/e . P >0 for 0<P<K/e and P <0 for
/
K/e<P<K , so P is a maximum (and P grows fastest) when P=K/e .
/ /
Note: If P>K , then ln (K/P)<0 , so P (t)>0 .

15. (a) dP/dt=kPcos (rt  ) (dP)/P=kcos (rt  )dt  (dP)/P=k (rt  )dt 
ln P=(k/r)sin (rt  )+C . (Since this is a growth model, P>0 and we can write ln P instead of ln P .)
Since P(0)=P , we obtain ln P =(k/r)sin (  )+C= ( k/r ) sin  +C C=ln P +(k/r)sin  . Thus,
0 0 0
ln P=(k/r)sin (rt  )+ln P +(k/r)sin  , which we can rewrite as ln P/P =(k/r) sin (rt  )+sin  ( )
0 0
(k/r) sin ( rt  ) +sin 
or, after exponentiation, P(t)=P e .
0
(b)

11
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation

As k increases, the amplitude A change in  produces slight

As r increases, the amplitude and
increases, but the minimum adjustments in the phase shift
the period decrease.
value stays the same. and amplitude.

( k/r ) ( 1+sin  ) ( k/r ) (  1+sin  )

P(t) oscillates between P e and P e (the extreme values are attained when
0 0

rt  is an odd multiple of ), so lim P(t) does not exist.
2 t 

16. (a) dP/dt=kPcos (rt  ) (dP)/P=kcos (rt  )dt (dP)/P=k (rt  )dt
2 2 2

1+cos (2(rt  )) k k
ln P=k dt= t+ (2(rt  ))+C . From P(0)=P , we get
2 2 4r 0
k k k
ln P = ( 2 )+C=C sin 2 , so C=ln P + sin 2 and
0 4r 4r 0 4r
k k k
ln P= t+ sin (2(rt  ))+ln P + sin 2 . Simplifying, we get
2 4r 0 4r
P k k f (t)
ln = t+ sin (2(rt  ))+sin 2 = f (t) , or P(t)=P e .
P 2 4r 0
0
(b)
An increase in k stretches the An increase in r compresses the As in Exercise 15, a change in 
graph of P vertically while graph of P horizontally    only makes slight adjustments in
maintaining P(0)=P . similar to changing the period in the growth of P , as shown in the
0 Exercise 15. figure.

/ f (t)
f (t)=k/2+ k/(4r) 2rcos (2(rt  )) = ( k/2 ) 1+cos (2(rt  ))  0 . Since P(t)=P e , we have
0
/ / f (t)
P (t)=P f (t)e  0 , with equality only when cos (2(rt  ))= 1 ; that is, when rt  is an odd
0
12
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.5 The Logistic Equation


multiple of . Therefore, P(t) is an increasing function on ( 0, ) . P can also be written as
2
kt/2 ( k/4r ) sin ( 2 ( rt  ) ) +sin 2 ( k/4r ) ( 1+sin 2 )
P(t)=P e e . The second exponential oscillates between e
0
( k/4r ) (  1+sin 2 ) kt/2
and e , while the first one, e , grows without bound. So lim P(t)= .
t 

K 1
17. By Equation (4), P(t)= . By comparison, if c=(ln A)/k and u= k(t c) , then
 kt 2
1+Ae
u  u u  u u  u u  u
e e e +e e e 2e e 2
1+tanh u=1+ = + = = and
u  u u  u u  u u  u  u  2u
e +e e +e e +e e +e e 1+e
 2u  k(t c) kc  kt ln A  kt  kt
e =e =e e =e e =Ae , so
1 1 K K 2 K K
K 1+tanh k(t c) = 1+tanh u = = = =P(t) .
2 2 2 2  2u  2u  kt
1+e 1+e 1+Ae

13
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

/ x 2 2 /
1. y +e y=x y is not linear since it cannot be put into the standard linear form (1), y +P(x) y=Q(x) .

3 / 3 / / 1 sin x
2. y+sin x=x y  x y  y=sin x y +  y= . This equation is in the standard linear
3 3
x x
form (1), so it is linear.

/ 2 / 2 / ln x
3. xy +ln x x y=0 xy  x y= ln x y + (  x ) y= , which is in the standard linear form (1), so
x
this equation is linear.

/ /
4. y +cos y=tan x is not linear since it cannot be put into the standard linear form (1), y +P(x) y=Q(x)
.

/ x /
5. Comparing the given equation, y +2y=2e , with the general form, y +P(x)y=Q(x) , we see that
 P ( x ) dx  2dx 2x
P(x)=2 and the integrating factor is I(x)=e =e =e . Multiplying the differential equation by
/ 2 3x 2 x  2x
2x / 2x
I(x) gives e y +2e y=2e 
3x
e y =2e  e y= 2e dx e y= e +C  y= e +Ce .
2x 3x 2x 3x 2x
3
( 3
)
/ /  P ( x ) dx  (  5 ) dx  5x
6. y =x+5y y  5y=x . I(x)=e =e =e . Multiplying the differential equation by I(x)
 5x  5x  5x
gives e
/
y  5e y=xe  ( e 5xy) /
=xe
 5x

 5x  5x
e y= xe dx=
1  5x 1  5x
5
xe 
25
e +C [ by parts] 
1 1 5x
y= x +Ce
5 25

/ 2 / 2
7. xy  2y=x [ divide by x ]  y +  y=x ( * ).
x
 2 2
 P(x)dx  ( 2/x)dx  2ln x ln x ln (1/x ) 2
I(x)=e =e =e =e =e =1/x . Multiplying the differential equation ( * ) by
1 / 2 1 1 / 1 1 2 2 2
I(x) gives y  y=  y =  y=ln x +C  y=x ( ln x +C ) =x ln x +Cx .
2 3 x 2 x 2
x x x x

2 2
2 / 2 cos x 2  P(x)dx  2/xdx 2ln x ln (x ) 2
/
8. x y +2xy=cos x y + y= . I(x)=e =e =e =e =x . Multiplying by
x 2
x
I(x) gives us our original equation back. You may have noticed this immediately, since P(x) is the
/ 2 / 2
derivative of the coefficient of y . We rewrite it as ( x y) =cos x . Thus,
1 1 1
x y= cos xdx=
2 2
(1+cos 2x)dx= x+ sin 2x+C
2 2 4
1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

1 1 C 1 1 C
y= + sin 2x+ or y= + sin xcos x+ .
2x 2 2 2x 2 2
4x x 2x x

/
9. Since P(x) is the derivative of the coefficient of y , we can write the differential equation
/ / 2 3/2 2
xy +y= x in the easily integrable form ( xy) = x  xy= x +C  y= x +C/x .
3 3

/  (  1 ) dx  x
10. y  y=1/x , so I(x)=e =e . Multiplying the differential equation by I(x) gives
 x  x  x
/
e y  e y=e /x ( e xy) /  x
=e /x y=e
x
 ( e x /x) dx+C .

2
 2xdx x / 2
11. I(x)=e =e . Multiplying the differential equation y +2xy=x by I(x) gives
/
x
2
/
e y +2xe y=x e 
2
x
2

2
2 x
2
( e y)
x

2
2

2
2 x
2
=x e . Thus
2 2 2 2
 x  x 1 x 1 x 1   x 1 x
 xe   
2 x x
y=e x e dx+C =e e dx+C = x+Ce  e e dx .
2 2 2 2

  tan xdx ln cos x  

12. I(x)=e =e =cos x (since  <x< ). Multiplying the differential equation by I(x)
2 2
/ /
gives y cos x ytan xcos x=xcos xsin 2x ( ycos x ) =xcos xsin 2x . So
1 1
 
2
y= xcos xsin 2xdx+C = 2xcos xsin xdx+C
cos x cos x
3 3 2 2
= 1  2xcos x 2 sin x  2xcos x C 3 sin x
+ sin x +C = + +2tan x
cos x 3 3 3 3 cos x 9

du du 1
13. ( 1+t ) +u=1+t , t>0 [ divide by 1+t ] 
+ u=1 ( * ), which has the form
dt dt 1+t
/  P(t)dt  1/ ( 1+t ) dt ln ( 1+t )
u +P(t)u=Q(t) . The integrating factor is I(t)=e =e =e =1+t .
/
Multiplying ( * ) by I(t) gives us our original equation back. We rewrite it as ( 1+t ) u =1+t . Thus,
1 2
t+ t +C 2
1 2 2 t +2t+2C
( 1+t ) u= ( 1+t ) dt=t+ t +C  u= or u= .
2 1+t 2 ( t+1 )

t
dr t dr 1 e  dt/(tln t) ln (ln t)
14. tln t +r=te  + r= . I(t)=e =e =ln t . Multiplying by ln t gives
dt dt tln t ln t

2
 Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

t
dr 1 t / t t e +C
ln t + r=e  [(ln t)r] =e  (ln t)r=e +C  r= .
dt t ln t

/ /  ( 1)dx  x  x  x /  x  x
15. y =x+y y +( 1)y=x . I(x)=e =e . Multiplying by e gives e y  e y=xe 
 x  x  x  x  x  x  x
e y= xe dx= xe  e +C [ integration by parts with u=x , dv=e dx ] 
/ x
(e y) =xe  y= x 1+Ce
x
. y(0)=2  1+C=2 C=3 , so y= x 1+3e .

dy 3 dy 2 2  ( 2/t ) dt 2ln t 2
16. t +2y=t , t>0 , y ( 1 ) =0 . Divide by t to get + y=t , which is linear. I(t)=e =e =t
dt dt t
3
dy / 1 5 t C 1
. Multiplying by t gives t
dt
4
+2ty=t 
2 2
( t y)2 4
=t 
2
t y= t +C 
5
y= +
5 2
. Thus, 0=y(1)= +C
5
t
3
1 t 1
 C= , so y=  .
5 5 2
5t
2 2
dv 2 t  (  2t ) dt  t
17.  2tv=3t e , v ( 0 ) =5 . I(t)=e =e . Multiply the differential equation by I(t) to get
dt
/
e
2
 t dv
dt
 t
2
 2te v=3t 
2  t
2
2
( )
 t
2 2 2
e v =3t  e v= 3t dt=t +C  v=t e +Ce . 5=v(0)=0 1+C 1=C , so
2 3 3 t t

2 2
3 t t
v=t e +5e .
1/2
/ 1 /  1/(2x)dx (1/2)ln x ln x
18. 2xy +y=6x , x>0  y + y=3 . I(x)=e =e =e = x . Multiplying by x gives
2x
1 C
x y= 3 x dx=2x +C 
/ / 3/2
xy + y=3 x  ( x y ) =3 x  y=2x+ . y(4)=20
2 x x
C 24
8+ =20 C=24 , so y=2x+ .
2 x
 1
/ 1 2  ( 1/x)dx  ln x ln x 1
/
19. xy =y+x sin x y  y=xsin x . I(x)=e =e =e = .
x x
1 1 / 1 1 / 1
Multiplying by gives y  y=sin x y =sin x y= cos x+C y= xcos x+Cx .
x x 2 x x
x
y( )=0    ( 1)+C =0 C= 1 , so y= xcos x x .

dy y / y   1/ x ( x+1 ) dx  ( ln x  ln x+1 ) x+1

20. x  =x y  =1 ( x>0 ) , so I(x)=e =e = .
dx x+1 x ( x+1 ) x
Multiplying the differential equation by I(x) gives
3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

x+1 / y x+1 x+1 x+1 / x+1

y  =  y = . Then
x x ( x+1 ) x x x x
x 1 x 1
y=
x+1
 1+
x
dx+C =
x+1
( x+ln x+C ) . But 0=y ( 1 ) = 1+C so C= 1 and the solution to
2
x
the initial value problem is y= ( x 1+ln x ) .
x+1

/ 1  ( 1/x ) dx ln x
21. y + y=cos x ( x 0 ), so I(x)=e =e =x (for x>0 ). Multiplying the differential
x
/ /
equation by I(x) gives xy +y=xcos x ( xy ) =xcos x . Thus,
1 1
y=
x
 xcos xdx+C =
x
xsin x+cos x+C

cos x C
= sin x+ +
x x

+
The solutions are asymptotic to the y axis (except for C= 1 ). In fact, for C> 1 , y as x 0 ,
+
whereas for C< 1 , y  as x 0 . As x gets larger, the solutions approximate y=sin x more
closely. The graphs for larger C lie above those for smaller C . The distance between the graphs
lessens as x increases.

 cos xdx sin x

22. I(x)=e =e . Multiplying the differential equation by I(x) gives
 sin x  sin x
sin x / sin x
e y +cos x e y=cos x e 
sin x
( sin x ) /
e y =cos x e  y=e  cos x e dx+C =1+Ce
sin x sin x

. The graphs for C= 3 , 0 , 1 , and 3 are shown. As the values of C get further from zero the graph is
stretched away from the line y=1 , which is the value for C=0 . The graphs are all periodic in x , with
a period of 2 .

4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

n n/ ( 1 n )
1 n du dy 
dy y du u
n du
23. Setting u=y , = ( 1 n ) y
or = = . Then the Bernoulli
dx dx dx 1 n dx 1 n dx
n/ ( 1 n )
u du 1/ ( 1 n ) n/ ( 1 n ) du
differential equation becomes +P(x)u =Q(x)u or +(1 n)P(x)u=Q(x)(1 n)
1 n dx dx
.

y
/ 2 1  1 / 1
24. Here y + = y , so n=2 , P(x)= and Q(x)= 1 . Setting u=y , u satisfies u  u=1 . Then
x x x
 (  1/x ) dx 1 1 1
I(x)=e = (for x>0 ) and u=x  dx+C =x ( ln x +C ) . Thus, y= .
x x x (C+ln x )

3
2
/ y 2 1  2 / 4u 2
25. y + y= . Here n=3 , P(x)= , Q(x)= and setting u=y , u satisfies u  = .
x 2 x 2 x 2
x x x
 (  4/x ) dx  4 2 2 4 2
=x and u=x  
4 4
Then I(x)=e dx+C =x +C =Cx + .
6 5 5x
x 5x
4 2  1/2
Thus, y=
Cx + .
5x

 2 /  (  2 ) dx  2x
26. Here n=3 , P(x)=1 , Q(x)=x and setting u=y , u satisfies u  2u= 2x . Then I(x)=e =e
 2x  2x 1  2x 1
and u=e   2xe dx+C =e
2x 2x 2x
xe + e +C =x+ +Ce .
2 2
 2 1 2x 1 2x  1/2
So y =x+ +Ce  y=
x+ +Ce .
2 2

dI dI  5dt 5t
27. (a) 2 +10I=40 or +5I=20 . Then the integrating factor is e =e . Multiplying the
dt dt
5t dI /
differential equation by the integrating factor gives e
dt
5t 5t
+5Ie =20e 
5t
e I =20e 
5t
( )
 5t  5t  5t

5t
I(t)=e 20e dt+C =4+Ce . But 0=I(0)=4+C , so I(t)=4 4e .
 0.5
(b) I(0.1)=4 4e  1.57 A

5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

dI 20t
28. (a) +20I=40sin 60t , so the integrating factor is e . Multiplying the differential equation by
dt
20t dI /
the integrating factor gives e
dt
20t 20t
+20Ie =40e sin 60t 
20t 20t
e I =40e sin 60t  ( )
I(t) =e 20t

40e sin 60t dt+C
20t

 20t 20t 1  20t

=e 40e ( 20sin 60t 60cos 60t ) +Ce
4000
sin 60t 3cos 60t  20t
= +Ce
5

 20t
3 sin 60t 3cos 60t+8e
But 1=I(0)= +C , so I(t)= .
5 5

(b)
 2
I(0.1) = sin 6 3cos 6+8e
5
  0.42 A

(c)

dQ  4dt 4t
29. 5 +20Q=60 with Q(0)=0 C. Then the integrating factor is e =e , and multiplying the
dt
4t dQ /
differential equation by the integrating factor gives e
dt
4t
+4e Q=12e 
4t 4t
e Q =12e 
4t
( )
 4t  4t  4t
Q(t)=e 
4t
12e dt+C =3+Ce . But 0=Q(0)=3+C so Q(t)=3 1 e ( ) is the charge at time t and
 4t
I=dQ/dt=12e is the current at time t .

dQ dQ  50dt 50t
30. 2 +100Q=10sin 60t or +50Q=5sin 60t . Then the integrating factor is e =e , and
dt dt
50t dQ 50t 50t
multiplying the differential equation by the integrating factor gives e +50e Q=5e sin 60t
dt
( e50tQ ) /=5e50tsin 60t
6
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

 50t  50t 1  50t


50t 50t
Q(t) = e 5e sin 60tdt+C =e 5e ( 50sin 60t 60cos 60t ) +Ce
6100
1  50t
= (5sin 60t 6cos 60t)+Ce
122
 50t
6 3 5sin 60t 6cos 60t 3e
But 0=Q(0)= +C so C= and Q(t)= + is the charge at time t , while
122 61 122 61
 50t
dQ 150cos 60t+180sin 60t 150e
the current is I(t)= = .
dt 61

dP  k dt kt
31. +kP=kM , so I(t)=e =e . Multiplying the differential equation by I(t) gives
dt
kt dP /  kt  kt
e
dt
kt
+kPe =kMe 
kt kt kt
e P =kMe  P(t)=e( ) 
kt
(
kMe dt+C =M+Ce , k>0 . Furthermore, it is )
reasonable to assume that 0 P(0) M , so  M C 0 .

 kt
32. Since P(0)=0 , we have P(t)=M(1 e ) . If P (t) is Jim’s learning curve, then P (1)=25 and
1 1
 k  2k  k
P (2)=45 . Hence, 25=M (1 e ) and 45=M (1 e ) , so 1 25/M =e or
1 1 1 1
2
M
 2k
M  25
25 1 1
k= ln 1 =ln . But 45=M (1 e ) so 45=M 1 or
M M  25 1 1 M
1 1 1
50M  625
1
45= . Thus, M =125 is the maximum number of units per hour Jim is capable of
M 1
1
processing. Similarly, if P (t) is Mark’s learning curve, then P (1)=35 and P (2)=50 . So
2 2 2
2
M
2
M  35
2
k=ln and 50=M 1 or M =61.25 . Hence the maximum number
M  35 2 M 2
2 2
of units per hour for Mark is approximately 61 . Another approach would be to use the midpoints of
the intervals so that P (0.5)=25 and P (1.5)=45 . Doing so gives us M  52.6 and M  51.8 .
1 1 1 2

7
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

kg L kg
33. y(0)=0 kg. Salt is added at a rate of 0.4
5 =2 . Since solution is drained from
L min min
the tank at a rate of 3 L / min, but salt solution is added at a rate of 5 L / min, the tank, which starts
out with 100 L of water, contains ( 100+2t ) L of liquid after t min. Thus, the salt concentration at time
y(t) kg
t is . Salt therefore leaves the tank at a rate of
100+2t L
y(t) kg L 3y kg
3 = . Combining the rates at which salt enters and leaves the
100+2t L min 100+2t min
dy 3y dy 3
tank, we get =2 . Rewriting this equation as + y=2 , we see that it is
dt 100+2t dt 100+2t
3dt 3
linear. I(t)=exp 
3/2
=exp ln (100+2t) = ( 100+2t ) . Multiplying the differential
100+2t 2
3/2 dy 1/2 3/2 3/2 / 3/2
equation by I(t) gives (100+2t) +3(100+2t) y=2(100+2t)  (100+2t) y =2(100+2t) 
dt
3/2 2 5/2 2  3/2
(100+2t) y= (100+2t) +C  y= (100+2t)+C(100+2t) . Now
5 5
2  3/2 1 2  3/2
0=y(0)= (100)+C 100 =40+ C  C= 40 , 000 , so y= (100+2t) 40,000(100+2t)
5 1000 5
kg. From this solution (no pun intended), we calculate the salt concentration at time t to be
y(t)  40,000 2 kg  40,000 2 kg
C(t)= = + . In particular, C(20)= +  0.2275 and
100+2t 5/2 5 L 5/2 5 L
( 100+2t ) 140
2  3/2
y(20)= (140) 40 , 000(140)  31.85 kg.
5

34. Let y(t) denote the amount of chlorine in the tank at time t (in seconds).
y(0)= ( 0.05g/L ) ( 400L ) =20 g. The amount of liquid in the tank at time t is ( 400 6t ) L since 4 L of
water enters the tank each second and 10 L of liquid leaves the tank each second. Thus, the
y(t) g
concentration of chlorine at time t is . Chlorine doesn’t enter the tank, but it leaves at a
400 6t L
y(t) g L 10 y(t) g 5 y(t) g
rate of 10 = = . Therefore,
400 6t L s 400 6t s 200 3t s
dy 5y dy  5dt 5
=   =  ln y= ln (200 3t)+C 
dt 200 3t y 200 3t 3
5 C 5/3 C 5/3 C 20
y=exp ln (200 3t)+C =e (200 3t) . Now 20=y(0)=e  200  e = , so
3 5/3
200
5/3
(200 3t) 5/3 2
y(t)=20 =20(1 0.015t) g for 0 t 66 s, at which time the tank is empty.
5/3 3
200

35. (a)
8
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.6 Linear Equations

dv c  ( c/m ) dt ( c/m ) t
+ v=g and I(t)=e =e , and multiplying the differential equation by I(t) gives
dt m
( c/m ) t
( c/m ) t dv vce ( c/m ) t ( c/m ) t / ( c/m ) t
e + =ge  e v =ge . Hence,
dt m
 ( c/m ) t  c/m t
 ge ( ) dt+K =mg/c+Ke ( ) . But the object is dropped from rest, so v(0)=0 and
c/m t
v(t)=e
 ( c/m ) t
K= mg/c . Thus, the velocity at time t is v(t)=(mg/c) 1 e .
(b) lim v(t)=mg/c
t
 ( c/m ) t
(c) s(t)= v(t)dt=(mg/c) t+(m/c)e
2 2
+c where c =s(0) m g/c . s(0) is the initial position, so
1 1
 ( c/m ) t 2 2
s(0)=0 and s(t)=(mg/c) t+(m/c)e  m g/c .

 ct/m
36. v=(mg/c)(1 e )
dv mg  ct/m ct g  ct/m gt  ct/m g g  ct/m g  ct/m ct  ct/m
dm c
= 0 e 
2
+
c
1 e (  1=
m
e )
+ 
c c
e =
c
1 e 
m
e 
m
c dv ct  ct/m 1+ct/m 1+Q ct Q
=1 1+ e =1 =1 , where Q= 0 . Since e >1+Q for all Q>0 , it
g dm m ct/m Q m
e e
follows that dv/dm>0 for t>0 . In other words, for all t>0 , v increases as m increases.

9
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

1. (a) dx/dt= 0.05x+0.0001xy . If y=0 , we have dx/dt= 0.05x , which indicates that in the absence of
y , x declines at a rate proportional to itself. So x represents the predator population and y represents
the prey population. The growth of the prey population, 0.1y (from dy/dt=0.1y 0.005xy ), is restricted
only by encounters with predators (the term  0.005xy ). The predator population increases only
through the term 0.0001xy ; that is, by encounters with the prey and not through additional food
sources.
(b) dy/dt= 0.015y+0.00008xy . If x=0 , we have dy/dt= 0.015y , which indicates that in the absence
of x , y would decline at a rate proportional to itself. So y represents the predator population and x
represents the prey population. The growth of the prey population, 0.2x (from
2
dx/dt=0.2x 0.0002x  0.006xy=0.2x(1 0.001x) 0.006xy ), is restricted by a carrying capacity of 1000
and by encounters with predators (the term  0.006xy ). The predator population increases only
through the term 0.00008xy ; that is, by encounters with the prey and not through additional food
sources.

2
2. (a) dx/dt=0.12x 0.0006x +0.00001xy . dy/dt=0.08y+0.00004xy .
The xy terms represent encounters between the two species x and y . An increase in y makes dx/dt
(the growth rate of x ) larger due to the positive term 0.00001xy . An increase in x makes dy/dt (the
growth rate of y ) larger due to the positive term 0.00004xy . Hence, the system describes a
cooperation model.
2
(b) dx/dt=0.15x 0.0002x  0.0006xy=0.15x ( 1 x/750 )  0.0006xy .
2
dy/dt=0.2y 0.00008y  0.0002xy=0.2y ( 1 y/2500 )  0.0002xy .
The system shows that x and y have carrying capacities of 750 and 2500 . An increase in x reduces
the growth rate of y due to the negative term  0.0002xy . An increase in y reduces the growth rate of
x due to the negative term  0.0006xy . Hence, the system describes a competition model.

3. (a) At t=0 , there are about 300 rabbits and 100 foxes. At t=t , the number of foxes reaches a
1
minimum of about 20 while the number of rabbits is about 1000 . At t=t , the number of rabbits
2
reaches a maximum of about 2400 , while the number of foxes rebounds to 100 . At t=t , the number
3
of rabbits decreases to about 1000 and the number of foxes reaches a maximum of about 315 . As t
increases, the number of foxes decreases greatly to 100 , and the number of rabbits decreases to 300
(the initial populations), and the cycle starts again.

1
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

(b)

4. (a) At t=0 , there are about 600 rabbits and 160 foxes. At t=t , the number of rabbits reaches a
1
minimum of about 80 and the number of foxes is also 80 . At t=t , the number of foxes reaches a
2
minimum of about 25 while the number of rabbits rebounds to 1000 . At t=t , the number of foxes
3
has increased to 40 and the rabbit population has reached a maximum of about 1750 . The curve ends
at t=t , where the number of foxes has increased to 65 and the number of rabbits has decreased to
4
about 950 .

(b)

5.

2
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

6.

dW  0.02W +0.00002RW
7. =  (0.08 0.001W )RdW =( 0.02+0.00002R)W dR
dR 0.08R 0.001RW
0.08 0.001W  0.02+0.00002R 0.08 0.02
dW = dR   0.001 dW =  +0.00002 dR
W R W R
0.08ln W  0.001W = 0.02ln R +0.00002R+K 0.08ln W +0.02ln R=0.001W +0.00002R+K 
ln W ( 0.08 0.02
R ) =0.00002R+0.001W +K W
0.08 0.02
R =e
0.00002R+0.001W +K

0.02 0.08
0.02 0.08 0.00002R 0.001W R W
R W =Ce e  =C .
0.00002R 0.001W
e e
r k
dy  ry+bxy xy
In general, if = , then C= .
dx kx axy bx ay
e e

8. (a) A and L are constant 

/
A =0 and L =0
/
{ 0=2A 0.01AL
0= 0.5L+0.0001AL } 

{ 0=A(2 0.01L)
0=L( 0.5+0.0001A)
2 0.5
So either A=L=0 or L= =200 and A= =5000 . The trivial solution A=L=0 just says that if
0.01 0.0001
there aren’t any aphids or ladybugs, then the populations will not change. The non trivial solution,
L=200 and A=5000 , indicates the population sizes needed so that there are no changes in either the
number of aphids or the number of ladybugs.
dL dL/dt  0.5L+0.0001AL
(b) = =
dA dA/dt 2A 0.01AL

3
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

(c) C100708cC100708c.texC100708c.tex

(d)
At P ( 1000,200 ) , dA/dt=0 and dL/dt= 80<0 , so the number of ladybugs is decreasing and hence, we
0
are proceeding in a counterclockwise direction. At P , there aren’t enough aphids to support the
0
ladybug population, so the number of ladybugs decreases and the number of aphids begins to
increase. The ladybug population reaches a minimum at P ( 5000,100 ) while the aphid population
1
increases in a dramatic way, reaching its maximum at P ( 14,250,200 ) .
2

Meanwhile, the ladybug population is increasing from P to P ( 5000,355) , and as we pass through
1 3
P , the increasing number of ladybugs starts to deplete the aphid population. At P the ladybugs
2 3
reach a maximum population, and start to decrease due to the reduced aphid population. Both
populations then decrease until P , where the cycle starts over again.
0
(e) Both graphs have the same period and the graph of L peaks about a quarter of a cycle after the
graph of A .

9. (a) Letting W =0 gives us dR/dt=0.08R(1 0.0002R) . dR/dt=0 R=0 or 5000 . Since dR/dt>0 for
0<R<5000 , we would expect the rabbit population to increase to 5000 for these values of R . Since
dR/dt<0 for R>5000 , we would expect the rabbit population to decrease to 5000 for these values of
R . Hence, in the absence of wolves, we would expect the rabbit population to stabilize at 5000 .
/ /
(b) R and W are constant  R =0 and W =0

{ 0=0.08R(1 0.0002R) 0.001RW

0= 0.02W+0.00002RW } { 
0=R 0.08(1 0.0002R) 0.001W
0=W( 0.02+0.00002R)

The second equation is true if W =0 or

4
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

0.02 1
R= =1000 . If W =0 in the first equation, then either R=0 or R= =5000 . If R=1000 ,
0.00002 0.0002
then 0=1000 0.08(1 0.0002 1000) 0.001W  0=80(1 0.2) W  W =64 .
Case (i): W =0 , R=0 : both populations are zero
Case (ii): W =0 , R=5000 : see part (a)
Case (iii): R=1000 , W =64 : the predator/prey interaction balances and the populations are stable.
(c) The populations of wolves and rabbits fluctuate around 64 and 1000 , respectively, and eventually
stabilize at those values.

(d)

1
10. (a) If L=0 , dA/dt=2A ( 1 0.0001A) , so dA/dt=0 A=0 or A=
=10 , 000 . Since dA/dt>0
0.0001
for 0<A<10 , 000 , we expect the aphid population to increase to 10 , 000 for these values of A .
Since dA/dt<0 for A>10 , 000 , we expect the aphid population to decrease to 10 , 000 for these
values of A . Hence, in the absence of ladybugs we expect the aphid population to stabilize at 10 , 000
.
/ /
(b) A and L are constant  A =0 and L =0

{ 0=2A(1 0.0001A) 0.01AL

0= 0.5L+0.0001AL } {
0=A 2(1 0.0001A) 0.01L
0=L( 0.5+0.0001A)

0.5
The second equation is true if L=0 or A= =5000 . If L=0 in the first equation, then either A=0
0.0001
1
or A= =10 , 000 . If A=5000 , then 0=5000 2(1 0.0001 5000) 0.01L  0=10 ,
0.0001
000(1 0.5) 50L 50L=5000 L=100 . The equilibrium solutions are: (i) L=0,A=0 (ii) L=0,A=10 ,
000 (iii) A=5000,L=100
dL dL/dt  0.5L+0.0001AL
(c) = =
dA dA/dt 2A(1 0.0001A) 0.01AL

5
Stewart Calculus ET 5e 0534393217;9. Differential Equations; 9.7 Predator-Prey Systems

(d) All of the phase trajectories spiral tightly around the equilibrium solution ( 5000,100 ) .

The graph of A peaks just after the graph of L has a minimum.

(e)

At t=0 , the ladybug population decreases rapidly and the aphid population decreases slightly before
beginning to increase. As the aphid population continues to increase, the ladybug population reaches
a minimum at about ( 5000,75) . The ladybug population starts to increase and quickly stabilizes at
100 , while the aphid population stabilizes at 5000 .

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

1. x=1+t , y=5 2t ,  2 t 3
(a)
t  2  1 0 1 2 3
x  1 0 1 2 3 4
y 9 7 5 3 1  1

(b) x=1+t  t=x 1 y=5 2(x 1) , so y= 2x+7 ,  1 x 4 .

2. x=2cos t , y=t cos t , 0 t 2

t 0  /2  3 /2 2
x 2 0  2 0 2
y  1  /2  +1 3 /2 2  1
1.57 4.14 4.71 5.28

2
3. x=5sin t , y=t ,    t 
t    /2 0  /2 
x 0  5 0 5 0
y 2 2 0 2 2
  /4  /4 
9.87 2.47 2.47 9.87

1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

 t t
4. x=e +t , y=e  t ,  2 t 2
t  2  1 0 1 2
x 2 e 1 1 1  2
e 2 e +1 e +2
1.72
5.39 1.37 2.14
y 2  1 1 e 1 2
e +2 e +1 e 2
1.72
2.14 1.37 5.39

5. x=3t 5 , y=2t+1
(a)
t  2  1 0 1 234
x  11  8  5  2 1 4 7
y  3  1 1 3 579

1 1 2 13
(b) x=3t 5 3t=x+5 t= (x+5) y=2 (x+5)+1 , so y= x+ .
3 3 3 3

6. x=1+t , y=5 2t ,  2 t 3
2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

(a)
t  2  1 0 1 2 3
x  1 0 1 2 3 4
y 9 7 5 3 1  1

(b) x=1+t  t=x 1 y=5 2(x 1) , so y= 2x+7 ,  1 x 4 .

2
7. x=t  2 , y=5 2t ,  3 t 4
(a)
t  3  2  1 0 1 2 3 4
x 7 2  1  2  1 2 7 14
y 11 9 7 5 3 1  1  3

1 1 2 1 2
(b) y=5 2t  2t=5 y t= (5 y) x= (5 y)  2 , so x= (5 y)  2 ,  3 y 11 .
2 2 4

2
8. x=1+3t , y=2 t
(a)
t  3  2  1 0 1 2 3
x  8  5  2 1 4 7 10
y  7  2 1 2 1  2  7

1 1 2 1 2
(b) x=1+3t t= (x 1) y=2 (x 1) , so y= (x 1) +2 .
3 3 9
3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

9. (a) x= t , y=1 t
t 0 1 2 3 4
x 0 1 1.414 1.732 2
y 1 0  1  2  3

2 2
(b) x= t ,  t=x  y=1 t=1 x .
Since t 0 , x 0

2 3
10. (a) x=t , y=t
t  2  1 0 1 2
x 4 1 0 1 4
y  8  1 0 1 8

3 3
(b) y=t  t= y  x=t =
2
( 3 y ) 2=y2/3 .
t R , y R , x 0 .

4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

2 2 2 2
11. (a) x=sin  , y=cos  , 0    . x +y =sin  +cos  =1 . Since 0    , we have sin   0 ,
so x 0 .

(b)

12. (a) x=4cos  , y=5sin  ,   /2    /2 .

x 2 y 2 2 2
+ =cos  +sin  =1 , which is an ellipse with x  intercepts ( 4,0 ) and y  intercepts
4 5
( 0, 5) . We obtain the portion of the ellipse with x 0 since 4cos   0 for   /2    /2 .

(b)

2 2 2 2
13. (a) x=sin  , y=cos  . x+y=sin  +cos  =1 , 0 x 1 . Note that the curve is at ( 0,1 )

whenever  = n and is at ( 1,0 ) whenever  = n for every integer n .
2

(b)

5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

  2 2 2 2
14. (a) x=sec  , y=tan  , < < . x  y =sec   tan  =1 , x 1 ,
2 2
2
or x= y +1 .

(b)

t  t
15. (a) x=e , y=e .
t
y=1/e =1/x , x>0

(b)

16. (a) x=ln t , y= t , t 1 .

x x/2
x=ln t t=e  y= t =e , x 0 .

(b)

17. (a) x=cosh t , y=sinh t,

2 2 2 2
x  y =cosh t sinh t=1 , x 1

(b)

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

18. (a) x=1+cos   cos  =x 1 .

y=2cos   1=2(x 1) 1=2x 3 , 0 x 2.

(b)

2 2 2 2
19. x +y =cos  t+sin  t=1 , 1 t 2 , so the particle moves counterclockwise along the circle
2 2
x +y =1 from (  1,0 ) to ( 1,0 ) , along the lower half of the circle.

2 2 2 2
20. ( x 2 ) + ( y 3) =cos t+sin t=1 , so the motion takes place on a unit circle centered at ( 2,3) . As t
goes from 0 to 2 , the particle makes one complete counterclockwise rotation around the circle,
starting and ending at ( 3,3) .

1 2 1 2 2 2
21. x + y =sin t+cos t=1 , so the particle moves once clockwise along the ellipse
2 3
1 2 1 2
x + y =1 , starting and ending at ( 0,3) .
4 9

2 2 2
22. x=cos t=y , so the particle moves along the parabola x=y . As t goes from 0 to 4 , the particle
moves from ( 1,1 ) down to ( 1, 1 ) (at t= ), back up to ( 1,1 ) again (at t=2 ), and then repeats this
entire cycle between t=2 and t=4 .

23. We must have 1 x 4 and 2 y 3 . So the graph of the curve must be contained in the
rectangle [1,4] by [2,3] .

24. (a) From the first graph, we have 1 x 2 . From the second graph, we have  1 y 1. The only
choice that satisfies either of those conditions is III.
(b) From the first graph, the values of x cycle through the values from  2 to 2 four times. From the
second graph, the values of y cycle through the values from  2 to 2 six times. Choice I satisfies these
conditions.
(c) From the first graph, the values of x cycle through the values from  2 to 2 three times. From the
second graph, we have 0 y 2 . Choice IV satisfies these conditions.
(d) From the first graph, the values of x cycle through the values from  2 to 2 two times. From the
second graph, the values of y do the same thing. Choice II satisfies these conditions.

7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

25. When t= 1 , (x,y)=(0, 1) . As t increases to 0 , x decreases to  1 and y increases to 0 . As t

increases from 0 to 1 , x increases to 0 and y increases to 1 . As t increases beyond 1 , both x and y
increase. For t< 1 , x is positive and decreasing and y is negative and increasing. We could achieve
greater accuracy by estimating x  and y  values for selected values of t from the given graphs and
plotting the corresponding points.

26. For t< 1 , x is positive and decreasing, while y is negative and increasing (these points are in
Quadrant IV). When t= 1 , ( x,y ) = ( 0,0 ) and, as t increases from  1 to 0 , x becomes negative and y
increases from 0 to 1 .
At t=0 , ( x,y ) = ( 0,1 ) and, as t increases from 0 to 1 , y decreases from 1 to 0 and x is positive. At
t=1, ( x,y ) = ( 0,0 ) again, so the loop is completed. For t>1 , x and y both become large negative. This
enables us to draw a rough sketch. We could achieve greater accuracy by estimating x  and y  values
for selected values of t from the given graphs and plotting the corresponding points.

1
27. When t=0 we see that x=0 and y=0 , so the curve starts at the origin. As t increases from 0 to ,
2
the graphs show that y increases from 0 to 1 while x increases from 0 to 1 , decreases to 0 and to  1 ,
1
then increases back to 0 , so we arrive at the point ( 0,1 ) . Similarly, as t increases from to 1 , y
2
decreases from 1 to 0 while x repeats its

pattern, and we arrive back at the origin. We could achieve greater accuracy by estimating x  and y 
8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

values for selected values of t from the given graphs and plotting the corresponding points.

28. (a) Note that as t

  , we have x
  and y
 , whereas when t
 , both x and y
 .
This description fits only IV.
(b) Note that as t
 , y
  . This is only the case with VI.
(c) If t=0 , then ( x,y ) = ( sin 0,sin 0 ) = ( 0,0 ) . Also, x = sin 3t  1 for all t , and y = sin 4t  1 for all
t . The only graph which includes the point ( 0,0 ) and which has x  1 and y  1 , is V.
(d) Note that as t
  , both x and y
  , and as t
 , both x and y
 . This description fits
only III. (Also note that, since sin 2t and sin 3t lie between  1 and 1 , the curve never strays very far
from the line y=x .)
(e) Note that both x(t) and y(t) are periodic with period 2 and satisfy x  1 and y  1 . Now the
only y intercepts occur when x=sin (t+sin t)=0 t=0 or  . So there should be two y intercepts:
y(0)=cos 1 0.54 and y( )=cos (  1)  0.54 . Similarly, there should be two x intercepts:
  3 3
x =sin +1 0.54 and x =sin  1  0.54 . The only curve with these x
2 2 2 2
and y intercepts is I.
(f) Note that x(t) is periodic with period 2 , so the only y intercepts occur when x=cos t=0
 3
t= or . Also, the graph is symmetric about the x axis, since
2 2
y( t)=sin (  t+sin 5( t) ) =sin (  t sin 5t ) = sin ( t+sin 5t ) = y(t) , and x( t)=cos ( t)=cos t=x(t) . The
only graph which has only two y intercepts, and is symmetric about the x axis, is II.

3 5
29. As in Example 5, we let y=t and x=t 3t +t and use a t  interval of  2 ,2 .

5 2
30. We use x =t , y =t and x =t ( t 1 ) , y =t with  2  t 2 . There are 3 points of intersection;
1 1 2 2
( 0,0 ) is fairly obvious. The point in quadrant III is approximately (  0.8, 0.4 ) and the point in
quadrant I is approximately ( 1.1,1.8 ) .

31. (a) x=x +(x  x )t , y=y +(y  y )t , 0 t 1 . Clearly the curve passes through P ( x ,y ) when
1 2 1 1 2 1 1 1 1
9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

t=0 and through P

2 ( x ,y ) when t=1 . For 0<t<1 , x is strictly between x
2 2 1
and x and y is strictly
2
y y
2 1
between y and y . For every value of t , x and y satisfy the relation y y =
1 2 1 x x
2 1
( x x ) , which is
1

the equation of the line through P

1 ( x ,y ) and P ( x ,y ) .
1 1 2 2 2
y y x x
1 1
Finally, any point ( x,y ) on that line satisfies = ; if we call that common value t , then
y y x x
2 1 2 1
the given parametric equations yield the point ( x,y ) ; and any ( x,y ) on the line between P x ,y
1 1 1 ( )
and P ( x ,y ) yields a value of t in 0,1
2 2 2
. So the given parametric equations exactly specify the line
segment from P ( x ,y ) to P ( x ,y ) .
1 1 1 2 2 2
(b) x= 2+[3 ( 2)]t= 2+5t and y=7+( 1 7)t=7 8t for 0 t 1 .

( )
32. For the side of the triangle from A to B , use x ,y = ( 1,1 ) and x ,y = ( 4,2 ) . Hence, the
1 1 ( 2 2 )
equations are x=x + x  x t=1+ ( 4 1 ) t=1+3t , y=y + y  y t=1+ ( 2 1 ) t=1+t . Graphing x=1+3t and
( ) ( )
1 2 1 1 2 1
y=1+t with 0 t 1 gives us the side of the triangle from A to B . Similarly, for the side BC we use
x=4 3t and y=2+3t , and for the side AC we use x=1 and y=1+4t .

2 2
33. The circle x +y =4 can be represented parametrically by x=2cos t , y=2sin t ; 0 t 2 . The
2 2
circle x + ( y 1 ) =4 can be represented by x=2cos t , y=1+2sin t ; 0 t 2 . This representation gives
us the circle with a counterclockwise orientation starting at ( 2,1 ) .
(a) To get a clockwise orientation, we could change the equations to x=2cos t , y=1 2sin t , 0 t 2
.
(b) To get three times around in the counterclockwise direction, we use the original equations
x=2cos t , y=1+2sin t with the domain expanded to 0 t 6 .

(c) To start at ( 0,3) using the original equations, we must have x =0 ; that is, 2cos t=0 . Hence, t=
1 2
 3
. So we use x=2cos t , y=1+2sin t ;  t .
2 2
Alternatively, if we want t to start at 0 , we could change the equations of the curve. For example, we
could use x= 2sin t , y=1+2cos t , 0 t  .

10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

34.

35. (a) Let x

2
/a 2 2
=sin t and y
2
/b2 2
=cos t to obtain x=asin t and y=bcos t with 0 t 2 as
possible parametric equations for the ellipse x
2
/a
2
+y
2
/b
2
=1 .

(b) The equations are x=3sin t and y=bcos t for b { 1,2,4,8} .

(c) As b increases, the ellipse stretches vertically.

3
36. The possible parametrizations of the curve y=x include
3
(1) x=t , y=t , t R
3
(2) x= t , y= t , t R
3
(3) x=t+1 , y= ( t+1 ) , t R


37. The case < < is illustrated. C has coordinates ( r ,r ) as in Example 6, and Q has coordinates
2
( r ,r+rcos (   ) ) = ( r ,r(1 cos  ) ) , so P has coordinates
( r  rsin (   ),r(1 cos  ) ) = ( r(  sin  ),r(1 cos  ) ) . Again we have the parametric equations
x=r(  sin  ) , y=r(1 cos  ) .

11
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

3
38. The first two diagrams depict the case  < < , d<r . As in Example 6, C has coordinates
2
( r ,r ) .
Now Q (in the second diagram) has coordinates ( r ,r+dcos (   ) ) = ( r ,r dcos  ) , so a typical point
P of the trochoid has coordinates ( r +dsin (   ),r dcos  ) . That is, P has coordinates ( x,y ) , where
x=r  dsin  and y=r dcos  . When d=r , these equations agree with those of the cycloid.

39. It is apparent that x= OQ and y= QP = ST . From the diagram, x= OQ =acos  and

y= ST =bsin  . Thus, the parametric equations are x=acos  and y=bsin  . To eliminate  we
2 2
rearrange: sin  =y/b sin  = ( y/b ) and
2 2 2 2 2 2 2 2
cos  =x/a cos  = ( x/a ) . Adding the two equations: sin  +cos  =1=x /a +y /b . Thus, we have
an ellipse.

12
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

40. A has coordinates ( acos  ,asin  ) . Since OA is perpendicular to AB ,  OAB is a right triangle
and B has coordinates ( asec  ,0 ) . It follows that P has coordinates ( asec  ,bsin  ) . Thus, the
parametric equations are x=asec  , y=bsin  .

41. C= ( 2acot  ,2a ) , so the x coordinate of P is x=2acot  . Let B= ( 0,2a ) . Then  OAB is a right
angle and  OBA= , so OA =2asin  and A= ( ( 2asin  ) cos  , ( 2asin  ) sin  ) . Thus, the y
2
coordinate of P is y=2asin  .

2a
42. Let  be the angle of inclination of segment OP . Then OB = . Let C= ( 2a,0 ) . Then by
cos 
use of right triangle OAC we see that OA =2acos  . Now
1
OP = AB = OB  OA =2a  cos 
cos 
2 2
= 2a 1 cos  =2a
sin 
=2asin  tan 
cos  cos 
2 2
So P has coordinates x=2asin  tan   cos  =2asin  and y=2asin  tan   sin  =2asin  tan  .

43. (a)

13
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

There are 2 points of intersection:

(  3,0 ) and approximately (  2.1,1.4 ) .

(b) A collision point occurs when x =x and y =y for the same t . So solve the equations:
1 2 1 2

3sin t = 3+cos t (1)

2cos t =1+sin t (2)

From (2) , sin t=2cos t 1 . Substituting into (1) , we get 3(2cos t 1)= 3+cos t
 3 3 
5cos t=0 ( * )  cos t=0 t= or . We check that t= satisfies (1) and (2) but t= does
2 2 2 2
3
not. So the only collision point occurs when t= , and this gives the point (  3,0 ) .
2
(c) The circle is centered at ( 3,1 ) instead of (  3,1 ) . There are still 2 intersection points: ( 3,0 ) and
6
( 2.1,1.4 ) , but there are no collision points, since ( * ) in part (b) becomes 5cos t=6 cos t= >1 .
5

 
44. (a) If  =30 and v =500 m / s, then the equations become x=(500cos 30 )t=250 3t and
0
1
(
y= 500sin 30
2
) t
2 2
(9.8)t =250t 4.9t . y=0 when t=0 (when the gun is fired) and again when
250 250
t= 51 s. Then x= ( 250 3 ) 22 , 092 m, so the bullet hits the ground about 22 km
4.9 4.9
from the gun.
The formula for y is quadratic in t . To find the maximum y  value, we will complete the square:
2 2 2 2 2
2502 2 250 125 125 125 125 125
y= 4.9 t  t = 4.9 t  t+ + = 4.9 t + 
4.9 4.9 4.9 4.9 4.9 4.9 4.9
2
125 125
with equality when t= s, so the maximum height attained is 3189 m.
4.9 4.9

(b)
14
 Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

  
As  ( 0 < <90 ) increases up to 45 , the projectile attains a greater height and a greater range. As

 increases past 45 , the projectile attains a greater height, but its range decreases.

x
(c) x=(v cos  )t  t= .
0 v cos 
0
1 2 x g x 2 g 2
y=(v sin  )t gt  y=(v sin  )  =(tan  )x x , which
0 2 0 v cos  2 v cos  2
2v cos 
2
0 0 0
is the equation of a parabola (quadratic in x ).

2 3
45. x=t ,y=t  ct . We use a graphing device to produce the graphs for various values of c with
   t  . Note that all the members of the family are symmetric about the x axis. For c<0 , the
graph does not cross itself, but for c=0 it has a cusp at ( 0,0 ) and for c>0 the graph crosses itself at
x=c , so the loop grows larger as c increases.

3 2 4
46. x=2ct 4t ,y= ct +3t . We use a graphing device to produce the graphs for various values of c
with    t  . Note that all the members of the family are symmetric about the y axis. When c<0
, the graph resembles that of a polynomial of even degree, but when c=0 there is a corner at the
origin, and when c>0 , the graph crosses itself at the origin, and has two cusps below the x axis. The
size of the ‘‘swallowtail’’ increases as c increases.

15
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

47. Note that all the Lissajous figures are symmetric about the x axis. The parameters a and b simply
stretch the graph in the x and y directions respectively. For a=b=n=1 the graph is simply a circle
with radius 1 . For n=2 the graph crosses itself at the origin and there are loops above and below the
x axis. In general, the figures have n 1 points of intersection, all of which are on the y axis, and a
total of n closed loops.

a=b=1

n=2

n=3

48. We use    t  in the viewing rectangle  4,2   3,3 . We first observe that for c=0 , we
1 1
obtain a circle with center  ,0 and radius . As the value of c increases, there is a larger
2 2
outer loop and a smaller inner loop until c=1 , when we obtain a curve with a dent (called a cardioid
). As c increases, we get a curve with a dimple (called a limacon ) until c=2 . For c>2 , we have
convex limacons. For negative values of c , we obtain the same graphs as for positive c , but with
different values of t corresponding to the points on the curve.

16
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.1 Curves Defined by Parametric Equations

17
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

3 dy dx 2 dy dy/dt  5 5
1. x=t t , y=2 5t  = 5 , =1 3t , and = = or 2 .
dt dt dx dx/dt 2
1 3t 3t  1

t
t t dy t dx t t dy dy/dt 1+e
2. x=te , y=t+e  =1+e , =te +e , and = = .
dt dt dx dx/dt t t
te +e

2
4 3 dy 2 dx 3 dy dy/dt 3t +1
3. x=t +1 , y=t +t ; t= 1. =3t +1 , =4t , and = = .
dt dt dx dx/dt 3
4t
4
When t= 1 , (x,y)=(2, 2) and dy/dx= = 1 , so an equation of the tangent to the curve at the point
 4
corresponding to t= 1 is y ( 2)=( 1)(x 2) , or y= x .

2
2 1 3 dy 2 dx dy dy/dt t  1
4. x=2t +1 , y= t  t ; t=3 . =t  1 , =4t , and = = . When t=3 , ( x,y ) = ( 19,6 )
3 dt dt dx dx/dt 4t
8 2 2 2 20
and dy/dx= = , so an equation of the tangent line is y 6= (x 19) , or y= x .
12 3 3 3 3

t
t 2 dy 2t 2 dx e dy dy/dt 1 2/t 2t 2t 4
5. x=e , y=t ln t ; t=1 . =1 =1 , = , and = =  =
dt 2 t dt 2 t dx dx/dt t 2t t
t e / (2 t ) te
dy 2 2
. When t=1 , ( x,y ) = ( e,1 ) and = , so an equation of the tangent line is y 1= (x e) , or
dx e e
2
y= x+3 .
e

cos   2sin 2
dy dy/d
6. x=cos  +sin 2 , y=sin  +cos 2 ;  =0 . = . When  =0 , (x,y)=(1,1)
=
 sin  +2cos 2
dx dx/d
1 1 1 1
and dy/dx= , so an equation of the tangent to the curve is y 1= (x 1) , or y= x+ .
2 2 2 2

t 2 dy dx t dy dy/dt 2(t 1)
7. (a) x=e , y= ( t 1 ) ; ( 1,1 ) . =2(t 1) , =e , and = = .
dt dt dx dx/dt t
e
dy
At ( 1,1 ) , t=0 and = 2 , so an equation of the tangent is y 1= 2(x 1) , or y= 2x+3 .
dx
t 2 2 dy 1 dy
(b) x=e  t=ln x , so y= ( t 1 ) = ( ln x 1 ) and =2(ln x 1) . When x=1 , =2( 1)(1)= 2 ,
dx x dx
so an equation of the tangent is y= 2x+3 , as in part (a).

1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

dy dy/d sec  tan  tan 

8. (a) x=tan  , y=sec  ; ( 1, 2 ) . = = = =sin  .
dx dx/d 2 sec 
sec 
  
When (x,y)= ( 1, 2 ) ,  = (or = 2 /2 . Thus, an
+2 n for some integer n ), so dy/dx=sin
4 4 4
equation of the tangent to the curve is y 2 = ( 2 /2 ) (x 1) , or y= ( 2 /2 ) x+ ( 2 /2 ) .
2 2 2 2 d 2 d 2 dy
(b) tan  +1=sec   x +1=y , so (x +1)= (y ) 2x=2y .
dx dx dx
dy x 1 2
When (x,y)= ( 1, 2 ) , = = = , so an equation of the tangent is y 2 = ( 2 /2 ) (x 1) , as
dx y 2 2
in part (a).

dy dy/dt 2cos t cos t

9. x=2sin 2t , y=2sin t ; ( = 3,1 ) .
=
dx dx/dt 2 2cos 2t 2cos 2t
= . The point ( 3,1 ) corresponds to
 3
cos
 6 2 3
t= , so the slope of the tangent at that point is = = . An equation of the tangent
6  1 2
2cos 2
3 2
3 3 1
is therefore ( y 1 ) =
2
( x 3 ) , or y=
2
x
2
.

10. x=sin t , y=sin (t+sin t) ; ( 0,0 ) .

dy dy/dt cos (t+sin t)(1+cos t)

= = = ( sec t+1 ) cos (t+sin t)
dx dx/dt cos t

Note that there are two tangents at the point ( 0,0 ) , since both t=0 and t= correspond to the origin.
The tangent corresponding to t=0 has slope ( sec 0+1 ) cos (0+sin 0)=2cos 0=2 , and its equation is
y=2x . The tangent corresponding to t= has slope ( sec  +1 ) cos ( +sin  )=0 , so it is the x axis.

2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

2
dy dy/dt 2t+3t
2 2 3 3
11. x=4+t , y=t +t  = = =1+ t 
dx dx/dt 2t 2
3
2 (d/dt) 1+ t 2
d y d dy d(dy/dx)/dt 2 3/2 3 d y
= = = = = . The curve is CU when >0 , that
2 dx dx dx/dt 2t 2t 4t 2
dx dx
is, when t>0 .

3 2 dy dy/dt 2t
12. x=t  12t , y=t  1 = = 
dx dx/dt 2
3t  12
2
(3t  12) 2 2t(6t)
d dy
2 2 2 2 2 2
d y dt dx (3t  12)  6t  24  6(t +4)  2(t +4)
= = = = = . Thus, the curve is CU
2 dx/dt 2 2 3 3 2 3 2 3
dx 3t  12 (3t  12) 3 (t  4) 9(t  4)
2
when t  4<0 t <2  2<t<2 .
t  t
13. x=t e , y=t+e 
t
1 e 1 d dy d  t
1
 t t t 2 ( e )  t
dy dy/dt 1 e e e  t d y dt dx dt e
= = = = = e  = = = . The curve is CU
dx dx/dt t t t 2 dx/dt dx/dt t
1 e 1 e 1 e dx 1 e
t
when e <1 t<0 .
dy dy/dt 1 1/t t 1 2
14. x=t+ln t , y=t ln t  = = = =1 
dx dx/dt 1+1/t t+1 t+1
d dy d 2
2 1 2
d y dt dx dt t+1 2/(t+1) 2t
= = = = , so the curve is CU for all t in its domain,
2 dx/dt 1+1/t (t+1)/t 3
dx (t+1)
that is, t>0 .

15. x=2sin t , y=3cos t , 0<t<2 .

3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

d dy 3 2
2 
sec t
dy dy/dt  3sin t 3 d y dt dx 2 3 3
= = = tan t , so = = = sec t .
dx dx/dt 2cos t 2 2 dx/dt 2cos t 4
dx
3  3
The curve is CU when sec t<0 sec t<0 cos t<0 <t< .
2 2

dy dy/dt  sin t sin t 1 1

16. x=cos 2t , y=cos t , 0<t< . = = = = = sec t , so
dx dx/dt  2sin 2t 2 2sin tcos t 4cos t 4
d dy 1
2 sec ttan t
d y dt dx 4 1 3 3
= = = sec t. The curve is CU when sec t<0 sec t<0 cos t<0
2 dx/dt  4sin tcos t 16
dx

<t< .
2

2 3
17. x=10 t , y=t  12t .
2
dy/dt=3t  12=3(t+2)(t 2) , so dy/dt=0 t= 2 ( x,y ) = ( 6, 16 ) . dx/dt= 2t , so dx/dt=0 t=0
( x,y ) = ( 10,0 ) . The curve has horizontal tangents at ( 6, 16 ) and a vertical tangent at ( 10,0 ) .

3 2 3 2
18. x=2t +3t  12t , y=2t +3t +1 .
2 2
dy/dt=6t +6t=6t(t+1) , so dy/dt=0 t=0 or  1 ( x,y ) = ( 0,1 ) or ( 13,2 ) . dx/dt=6t +6t 12=6(t+2)(t 1) ,
so dx/dt=0 t= 2 or 1

( x,y ) = ( 20, 3) or (  7,6 ) . The curve has horizontal tangents at ( 0,1 ) and ( 13,2 ) , and vertical
tangents at ( 20, 3) and (  7,6 ) .
4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

19. x=2cos  , y=sin 2 .

  
dy/d =2cos 2 , so dy/d =0 2 = +n n ( x,y ) = (  2 , 1 ) . Also,
( n an integer)   = +
2 4 2
dx/d = 2sin  , so dx/d =0  =n  ( x,y ) = (  2,0 ) . The curve has horizontal tangents at
(  2 , 1 ) (four points), and vertical tangents at (  2,0 ) .


20. x=cos 3 , y=2sin  . dy/d =2cos  , so dy/d =0  = +n ( n an integer)  ( x,y ) = ( 0, 2 ) .
2

Also, dx/d = 3sin 3 , so dx/d =0 n ( x,y ) = (  1,0 ) or (  1, 3 ) . The curve
3 =n   =
3
has horizontal tangents at ( 0, 2 ) , and vertical tangents at (  1,0 ) , (  1, 3 ) and (  1, 3 ) .

4 2
21. From the graph, it appears that the leftmost point on the curve x=t  t , y=t+ln t is about
(  0.25,0.36 ) . To find the exact coordinates, we find the value of t for which the graph has a vertical
1
3 2
( )
tangent, that is, 0=dx/dt=4t  2t  2t 2t  1 =0 2t ( 2 t+1 ) ( 2 t 1 ) =0 t=0 or 
2
. The
negative and 0 roots are inadmissible since y(t) is only defined for t>0 , so the leftmost point must be
1 1 1 4 1 2 1 1 1 1 1
x ,y =  , +ln =  ,  ln 2
2 2 2 2 2 2 4 2 2

5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

22. The curve is symmetric about the line y= x since replacing t with  t has the effect of replacing
( x,y ) with (  y, x ) , so if we can find the highest point x ,y , then the leftmost point is
( h h )
t
x ,y =  y , x
( )( ) . After carefully zooming in, we estimate that the highest point on the curve x=te ,
l l h h
 t
y=te is about ( 2.7,0.37) .

To find the exact coordinates of the highest point, we find the value of t for which the curve has a
 t  t  t
horizontal tangent, that is, dy/dt=0 t( e )+e =0 (1 t)e =0 t=1 . This corresponds to the point
t t
( x(1),y(1) ) = ( e,1/e ) . To find the leftmost point, we find the value of t for which 0=dx/dt=te +e 
t t 
(1+t)e =0 t= 1 . This corresponds to the point ( x( 1),y( 1) ) = (  1/e, e ) . As t 
, x(t)=te 0 by
 t
l’Hospital’s Rule and y(t)=te 
, so the y axis is an asymptote. As t
, x(t)
and
+
y(t) 0 , so the x axis is the other asymptote. The asymptotes can also be determined from the
graph, if we use a larger t  interval.

4 3 2 3
23. We graph the curve x=t  2t  2t , y=t  t in the viewing rectangle  2,1.1 by  0.5,0.5 . This
rectangle corresponds approximately to t  1,0.8 . We estimate that the curve has horizontal
tangents at about

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

(  1, 0.4 ) and (  0.17,0.39 ) and vertical tangents at about ( 0,0 ) and (  0.19,0.37) . We calculate
2
dy dy/dt 3t  1 2 1
= = . The horizontal tangents occur when dy/dt=3t  1=0 t= , so both
dx dx/dt 3 2 3
4t  6t  4t
2
horizontal tangents are shown in our graph. The vertical tangents occur when dx/dt=2t(2t  3t 2)=0
1
2t(2t+1)(t 2)=0 t=0 ,  or 2 . It seems that we have missed one vertical tangent, and indeed if we
2
plot the curve on the t interval  1.2,2.2 we see that there is another vertical tangent at (  8,6 ) .

4 3 2 2
24. We graph the curve x=t +4t  8t , y=2t  t in the viewing rectangle  3.7,0.2 by  0.2,1.4 . It
appears that there is a horizontal tangent at about (  0.4, 0.1 ) , and vertical tangents at about (  3,1 )
and ( 0,0 ) .

dy dy/dt 4t 1 1
We calculate = = , so there is a horizontal tangent where dy/dt=4t 1=0 t=
dx dx/dt 3 2 4
4t +12t  16t
. This point (the lowest point) is shown in the first graph. There are vertical tangents where
3 2 2
dx/dt=4t +12t  16t=0 4t(t +3t 4)=0 4t(t+4)(t 1)=0 . We have missed one vertical tangent
corresponding to t= 4 , and if we plot the graph for t  5,3 , we see that the curve has another
vertical tangent line at approximately (  128,36 ) .
dx dy 2 2
25. x=cos t , y=sin tcos t . = sin t , = sin t+cos t=cos 2t . ( x,y ) = ( 0,0 )  cos t=0 t is an odd
dt dt
multiple of
7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

 dx  dy dy 3 dx dy dy
. When t= = 1 and , = 1 , so =1 . When t= , =1 and = 1 . So = 1 .
2 2 dt dt dx 2 dt dt dx
Thus, y=x and y= x are both tangent to the curve at ( 0,0 ) .

2 2
26. x=1 2cos t= cos 2t , y=(tan t)(1 2cos t)= (tan t)cos 2t . To find a point where the curve crosses
itself, we look for two values of t that give the same point ( x,y ) . Call these values t and t . Then
1 2
2 2 2 2 1
cos t =cos t (from the equation for x ) and either tan t =tan t or cos t =cos t = (from the
1 2 1 2 1 2 2
2 2
equation for y ). We can satisfy cos t =cos t and tan t =tan t by choosing t arbitrarily and taking
1 2 1 2 1
t =t + , so evidently the whole curve is retraced every time t traverses an interval of length  .
2 1
  2 2
Thus, we can restrict our attention to the interval  , . If t = t , then cos t =cos t , but
2 2 2 1 2 1
2 2 1 
tan t = tan t . This suggests that we try to satisfy the condition cos t =cos t = . Taking t =
2 1 1 2 2 1 4
 dx dy 2
and t = gives ( x,y ) = ( 0,0 ) for both values of t .
=2sin 2t ,and =2sin 2ttan t cos 2tsec t .
2 4 dt dt
 dx dy dy  dx dy dy
When t= , =2 and =2 , so =1 . When t= , = 2 and =2 , so = 1 . Thus,
4 dt dt dx 4 dt dt dx
the equations of the two tangents at ( 0,0 ) are y=x and y= x .

dx dy dy dsin 
27. (a) x=r  dsin  , y=r dcos  ; =r dcos  , =dsin  . So = .
d d dx r dcos 
(b) If 0<d<r , then dcos   d<r , so r dcos  r d>0 . This shows that dx/d never vanishes, so
the trochoid can have no vertical tangent if d<r .

3 3
28. x=acos  , y=asin  .
(a)
8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

dx 2 dy 2 dy sin 
= 3acos  sin  , =3asin  cos  , so = = tan  .
d d dx cos 
(b) The tangent is horizontal  dy/dx=0 tan  =0  =n  ( x,y ) = (  a,0 ) . The tangent is vertical

 cos  =0  is an odd multiple of  ( x,y ) = ( 0, a) .
2
 2 2
(c) dy/dx= 1 tan  = 1  is an odd multiple of  ( x,y ) =  a, a (All sign
4 4 4
choices are valid.)

1 12
29. The line with parametric equations x= 7t , y=12t 5 is y=12  x  5 , which has slope  .
7 7
3 2 dy dy/dt 12t 12 2
The curve x=t +4t , y=6t has slope = = 2 . This equals   3t +4= 7t
dx dx/dt 7
3t +4
4 208 32
(3t+4)(t+1)=0 t= 1 or t=  ( x,y ) = (  5,6 ) or  , .
3 27 3

2
2 dx dy
3 2 dy 6t
30. x=3t +1 , y=2t +1 , =6t , =6t , so = =t (even where t=0 ).
dt dt dx 6t
So at the point corresponding to parameter value t , an equation of the tangent line is
3 2 3 2
y (2t +1)=t x (3t +1) . If this line is to pass through ( 4,3) , we must have 3 (2t +1)=t 4 (3t +1)
3 3 3 2
 2t  2=3t  3t t  3t+2=0 (t 1) (t+2)=0 t=1 or  2 . Hence, the desired equations are y 3=x 4 ,
or y=x 1 , tangent to the curve at ( 4,3) , and y ( 15)= 2(x 13) , or y= 2x+11 , tangent to the curve at
( 13, 15) .
31. By symmetry of the ellipse about the x and y axes,
 /2  /2 1
A = 4 ydx=4 bsin  ( asin  )d =4ab sin  d =4ab
a 0 2
(1 cos 2 )d
0  /2 0 0 2
1  /2 
= 2ab   sin 2 =2ab = ab
2 0 2

1 1 3 1 3
32. t+1/t=2.5 t= or 2 , and for <t<2 , we have t+1/t<2.5 . x= when t= and x= when
2 2 2 2 2
t=2 . \
5
 (2.5 y)dx=
3/2 2 2 2
A=  t 1/t (1+1/t )dt [x=t 1/t,dx=(1+1/t )dt]
 3/2 1/2 2
2 2
5  1 5  2   t 5t 5 1

2 3
=  t+  2t + t  t dt= +  2ln t  +
1/2 2 2 2 2 2t 2
2t 1/2
9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

5 1 1 5 15
=  2+5 2ln 2 +   + +2ln 2 5+2 =  4ln 2
4 8 8 4 4

33.
 /2 t 1 t  /2
A = (y 1)dx= (e  1)( sin t)dt=
1 0 t
(e sin t sin t)dt= e (sin t cos t)+cos t
0  /2 0 2 0
1  /2
= (e  1)
2

2  /2
34. By symmetry, A=4 ydx=4 
a 0 3 2 4 2
asin  ( 3acos  sin  )d =12a sin  cos  d . Now
0  /2 0
1 1
 
2 2 2

4 2
sin  cos  d = sin  sin 2 d =
(1 cos 2 )sin 2 d
4 8
1 1 1 1 1

2 3
= (1 cos 4 ) sin 2 cos 2 d =   sin 4  sin 2 +C
8 2 16 64 48
 /2 1 1 1  /2   3 2
so 
4 2 3 2
sin  cos  d =   sin 4  sin 2 = . Thus, A=12a =  a .
0 16 64 48 0 32 32 8

35.
A = ydx= (r dcos  )(r dcos  )d =
2 r 2 2 2 2 2
(r  2drcos  +d cos  )d
0 0 0

2 1 2 1 2 2 2
= r   2drsin  + d  + sin 2 =2 r + d
2 2 0

36. (a) By symmetry, the area of is twice the area inside above the x axis. The top half of the loop is
2 3 3
described by x=t , y=t  3t ,  3 t 0 , so, using the Substitution Rule with y=t  3t and dx=2t dt ,
we find that
 3 3  3 2 5 3  3
area = 2 ydx=2 (t  3t)2t dt=2
3 4 2
(2t  6t )dt=2 t  2t
0 0 0 5 0
2 1/2 5 1/2 3 2 24
=2
5
( 3 )  2( 3 ) =2
5
(  9 3 )  2 ( 3 3 ) =
5
3

(b) Here we use the formula for disks and use the Substitution Rule as in part (a):
 3 3  3 6
  
3 2 2 4 2
volume =  y dx= (t  3t) 2t dt=2 (t  6t +9t )t dt
0 0 0
1 8 6 9 4  3 1 1/2 8 1/2 6 9 1/2 4
= 2 t  t + t =2 ( 3 )  ( 3 ) + ( 3 )
8 4 0 8 4
10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

81 81 27
= 2  27+ = 
8 4 4

(c) By symmetry, the y coordinate of the centroid is 0 . To find the x coordinate, we note that it is
the same as the x coordinate of the centroid of the top half of , the area of which is
1 24 12 12
 3= 3 . So, using Formula .3.8 with A= 3 , we get
2 5 5 5
5 5  3 2 3 5 1 7 3 5  3
 0xydx=  0 t (t  3t)2t dt=
3
x= t  t
12 3 12 3 6 3 7 5 0

5 1 1/2 7 3 1/2 5 5 27 27 9
= ( 3 )  ( 3 ) =  3+ 3 =
6 3 7 5 6 3 7 5 7
9
So the coordinates of the centroid of are ( x,y ) = ,0 .
7
2 4 3/2 1/2
37. x=t t , y= t , 1 t 2 . dx/dt=1 2t and dy/dt=2t , so
3
2 2 2 1/2 2 2 2
(dx/dt) +(dy/dt) = ( 1 2t ) +(2t ) =1 4t+4t +4t=1+4t . Thus,
L= (dx/dt) +(dy/dt) dt=
b 2 2 2 2
1+4t dt .
a 1

t 2 t 2 2 2t 2
38. x=1+e , y=t ,  3 t 3 . dx/dt=e and dy/dt=2t , so (dx/dt) +(dy/dt) =e +4t . Thus,
L= (dx/dt) +(dy/dt) dt=
b 2 2 3 2t 2
e +4t dt .
a  3
39. x=t+cos t , y=t sin t , 0 t 2 . dx/dt=1 sin t and dy/dt=1 cos t , so
2 2 2 2 2 2
(dx/dt) +(dy/dt) =(1 sin t) +(1 cos t) =(1 2sin t+sin t)+(1 2cos t+cos t)
=3 2sin t 2cos t

Thus, L= (dx/dt) +(dy/dt) dt=

b 2 2 2
3 2sin t 2cos t dt .
a 0

dx 1 dy 1
40. x=ln t , y= t+1 , 1 t 5 . = and = , so
dt t dt 2 t+1
2 2 2
dx dy 1 1 t +4t+4
+ = + = 2 . Thus,
dt dt 2 4 ( t+1 )
t 4t ( t+1 )
2 2 2 2
dx dy t +4t+4 (t+2) t+2
L= dt= dt= dt=
b 5 5 5
+ dt .
a dt dt 1 2 1 2 1 2t t+1
4t ( t+1 ) (2t) ( t+1 )

11
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

2 3 2 2 2 2 4
41. x=1+3t , y=4+2t , 0 t 1 . dx/dt=6t and dy/dt=6t , so (dx/dt) +(dy/dt) =36t +36t .
Thus,
1
L = 36t +36t dt= 6t 1+t dt=6
1 2 4 1 2 2 2
u du [ u=1+t , du=2t dt ]
0 0 1 2
2 3/2 2 3/2
=3 u =2(2  1)=2 ( 2 2  1 )
3 1

42. x=a(cos  + sin  ) , y=a(sin    cos  ) , 0    .

2 2 2 2 2
( dx/d ) + ( dy/d ) =a (  sin  + cos  +sin  ) + ( cos  + sin   cos  )
=a  ( cos  +sin  ) = ( a )
2 2 2 2 2

 1  1
Thus, L=
2 2
a d =a  =  a.
0 2 0 2

t dx ( 1+t )  1 t 1 1 dy 1
43. x= , y=ln ( 1+t ) , 0 t 2 . = = and = , so
1+t dt 2 2 dt 1+t
( 1+t ) ( 1+t )
2 2 2
dx dy 1 1 1 2 t +2t+2
+ = + = 1+ ( 1+t ) = . Thus,
dt dt 4 2 4 4
( 1+t ) ( 1+t ) ( 1+t ) ( 1+t )
3
2 2 2
L= 
2
0
t +2t+2
2
dt=
3
1
u +1
2
du[u=t+1,du=dt]= 
u +1
u
2
+ln u+ u +1 ( ) 1
( 1+t ) u
10
= +ln ( 3+ 10 ) + 2  ln ( 1+ 2 )
3

t  t t  t
44. x=e +e , y=5 2t , 0 t 3 . dx/dt=e  e and dy/dt= 2 , so
2 2 2t  2t 2t  2t t  t2
(dx/dt) +(dy/dt) =e  2+e +4=e +2+e =(e +e ) . Thus,
 t  t 3  3  3
L= (e +e )dt= e  e
3 t t 3 3
=e  e  (1 1)=e  e .
0 0

t t
45. x=e cos t , y=e sin t , 0 t  .
dx 2 dy 2
t 2 t 2
+ = e ( cos t sin t ) + e ( sin t+cos t )
dt dt
= ( et ) 2 ( cos 2t 2cos t sin t+sin 2t )
12
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

( t ) 2 ( sin 2t+2sin t cos t+cos 2t )

+ e
= e2t ( 2cos 2t+2sin 2t ) =2e2t
  t  
Thus, L= 2e dt= 2 e dt= 2 e = 2 ( e  1 ) .
2t t
0 0 0

1
46. x=cos t+ln (tan t) , y=sin t ,  /4 t 3 /4 .
2

1 2
sec (t/2)
dx 2 1 1
= sin t+ = sin t+ = sin t+
dt tan (t/2) 2sin (t/2)cos (t/2) sin t

dy dx 2 dy 2 2 1 2 2 2
and =cos t , so + =sin t 2+ +cos t=1 2+csc t=cot t . Thus,
dt dt dt 2
sin t

 /2  /2 1 1
L=  cot t dt=2
3 /4
cot t dt=2[ln sin t ] =2 ln 1 ln =2 ( 0+ln 2 ) =2 ln 2 =ln 2.
 /4  /4  /4 2 2

t t/2
47. x=e  t , y=4e ,  8 t 3 .

t/2 2
( dx/dt ) + ( dy/dt ) =(e  1) + ( 2e ) =e2t 2et+1+4et =e2t+2et+1=(et+1)2
2 2 t 2

Thus,

L=
13
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

3  8  8
 (e +1) dt=
3 t 2 3 t t 3 3
(e +1)dt= e +t =(e +3) (e  8)=e  e +11.
 8  8  8

3 2 2 dx 2 dy 2 22 2 22
48. x=3t t , y=3t . dx/dt=3 3t and dy/dt=6t , so + =(3 3t ) +(6t) =(3+3t ) and
dt dt
the length of the loop is given by
3 3 3
 (3+3t )dt=2
2 2 3
L= (3+3t )dt=2[3t+t ]
 3 0 0

= 2 ( 3 3 +3 3 ) =12 3.

t t
49. x=t e , y=t+e ,  6 t 6 .
dx 2 dy 2 t 2 t 2
dt
+
dt
( ) ( ) (
= 1 e + 1+e = 1 2e +e + 1+2e +e =2+2e , so L=
t 2t t 2t 2t
) ( ) 6
 6
2t
2+2e dt . Set
2t 6 ( 6)
f (t)= 2+2e . Then by Simpson’s Rule with n=6 and  t= =2 , we get
6
2
L [ f ( 6)+4 f ( 4)+2 f ( 2)+4 f (0)+2 f (2)+4 f (4)+ f (6)] 612.3053 .
3

2 2
50. x=2acot   dx/dt= 2acsc  and y=2asin   dy/dt=4asin  cos  =2asin 2 .
 /2  /2
So L= 4a csc  +4a sin 2 d =2a
2 4 2 2 4 2
csc  +sin 2 d . Using Simpson’s Rule with n=4 ,
 /4  /4

14
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

 /2  /4  4 2
  = = , and f ( )= csc  +sin 2 , we get
4 16
  5 3 7 
L 2a S = ( 2a ) f +4 f +2 f +4 f +f  2.2605a .
4 16 3 4 16 8 16 2

2 2
51. x=sin t , y=cos t , 0 t 3 .
2 2 2 2 2 2 2
( dx/dt ) + ( dy/dt ) = ( 2sin tcos t ) + (  2cos tsin t ) =8sin tcos t=2sin 2t
 /2  /2
 2 sin 2t dt=6 2 
3
Distance = sin 2t dt [bysymmetry]= 3 2 cos 2t
0 0 0
=  3 2 (  1 1 ) =6 2

The full curve is traversed as t goes from 0 to , because the curve is the segment of x+y=1 that
2
lies in the first quadrant (since x , y 0 ), and this segment is completely traversed as t goes from 0 to
  /2
. Thus, L= sin 2t dt= 2 , as above.
2 0
2 dx 2 dy 2 2 2 2 2
52. x=cos t , y=cos t , 0 t 4 . + = (  2cos tsin t ) + (  sin t ) =sin t(4cos t+1)
dt dt

 4cos t+1 dt=4 sin t 4cos t+1 dt
4 2 2
Distance = sin t
0 0
 1
=  4 4u +1 du[u=cos t,du= sin t dt]=4 4u +1 du=8
2 1 2 1 2
4u +1 du
1  1 0
 1
1
= 8
tan 2 2 2
sec   sec  d 2u=tan  ,2du=sec  d ]
0 2
 1  1
= 4 tan 2 3
sec  d = 2sec  tan  +2ln sec  +tan 
tan 2
=4 5+2ln ( 5+2 )
0 0
 1
Thus, L=
2
0
sin t 4cos t+1 dt= 5+
2
ln ( 5+2 ) .

53. x=asin  , y=bcos  , 0   2 .

dx 2 dy 2

d
+
d
= ( acos  ) 2+ (  bsin  ) 2=a2cos 2 +b2sin 2 =a2 1 sin 2 ( ) +b2sin 2
2
= a
2
( a  b ) sin 
2 2 2 2 2
=a  c sin  =a
2 2
1
c
2
sin 
2
=a
2
( 1 e2sin 2 )
a
 /2  /2
So L=4
0
2
(
a 1 e sin 
2 2
) d =4a
0
2
1 e sin  d
2
.

15
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

3 3
54. x=acos  , y=asin  .
( dx/d ) + ( dy/d ) = (  3acos  sin 
2 2 2
) 2+ ( 3asin 2 cos  ) 2=9a2cos 4 2 2 4 2
sin  +9a sin  cos 
2 2 2
( 2
=9a sin  cos  cos  +sin  =9a sin  cos  .
2
) 2 2 2

The graph has four fold symmetry and the curve in the first quadrant corresponds to 0    /2 .
Thus,
 /2
L=4 3asin  cos  d [ since a>0 and sin  and cos  are positive for 0    /2]
0
1 2  /2 1
=12a sin  =12a  0 =6a .
2 0 2

55. (a) Notice that 0 t 2 does not give the complete curve because x(0) x(2 ) . In fact, we must
take t 0,4 in order to obtain the complete curve, since the first term in each of the parametric
2 4
equations has period 2 and the second has period = , and the least common integer
11/2 11
multiple of these two numbers is 4 .

(b) We use the CAS to find the derivatives dx/dt and dy/dt , and then use Formula 1 to find the arc
length. Recent versions of Maple express the integral 
4 2 2
(dx/dt) +(dy/dt) dt as 88E ( 2 2 i ) ,
0
22
1 x t
where E(x) is the elliptic integral 
1
dt and i is the imaginary number  1 . Some earlier
0 2
1 t
versions of Maple (as well as Mathematica) cannot do the integral exactly, so we use the command
evalf(Int(sqrt(diff(x,t)^2+diff(y,t)^2),t=0..4*Pi)); to estimate the length,
and find that the arc length is approximately 294.03 . Derive’s Para_arc_length function in the
utility file Int_apps simplifies the integral to

16
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

11t 11t
11
4
 4cos t cos  4sin t sin +5 dt .
0 2 2

1 1 1 1
56. (a) It appears that as t
, ( x,y ) , , and as t 
, ( x,y )  , .
2 2 2 2

 2  2
(b) By the Fundamental Theorem of Calculus, dx/dt=cos t and dy/dt=sin t , so by
2 2
Formula 6, the length of the curve from the origin to the point with parameter value t is

 
 ( dx/du ) + ( dy/du ) du=
t 2 2 t 2 2 2 2
L= cos u +sin u du
0 0 2 2


t
= 1du=t [or tift<0]
0
We have used u as the dummy variable so as not to confuse it with the upper limit of integration.

4 3/2 2 dx 2 dy 2 2 1/2 2 2 2
57. x=t t , y= t , 1 t 2 . + = ( 1 2t ) +(2t ) =1 4t+4t +4t=1+4t , so
3 dt dt
4 3/2 8 3/2
S= 2 yds= 2  t 1+4t dt=
2 2 2 2 2
t 1+4t dt .
1 1 3 1 3

2 
58. x=sin t , y=sin 3t , 0 t . dx/dt=2sin tcos t=sin 2t and dy/dt=3cos 3t , so
3
 /3
( dx/dt ) + ( dy/dt ) =sin 2t+9cos 3t and S= 2 yds=
2 2 2 2 2 2
2 sin 3t sin 2t+9cos 3t dt .
0

dx 2 dy 2
3
59. x=t , y=t , 0 t 1 .
2
dt
+
dt
= 3t( 2) 2+ ( 2t ) 2=9t 4+4t2 .
dx 2 dy 2
 dt= 2 t 
1 1 2 4 2 12 2 2
S= 2 y + 9t +4t dt=2 t t (9t +4) dt
0 dt dt 0 0

17
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

2 2
u 4 1 u=9t +4t =(u 4)/9 2
 
13 13 3/2 1/2
= 2 u du 1 = (u  4u )du
4 9 18 du=18t dt so, t dt= du 9 18 4
18
 2 5/2 8 3/2 13  2 5/2 3/2 13
= u  u =  [3u  20u ]
81 5 3 4 81 15 4
2
=
1215
2
(
3 13 13  20 13 13  ( 3 32 20 8 ) )
2
=
1215
( 247 13 +64)

dx 2 dy 2 2 2 2
3
60. x=3t t , y=3t , 0 t 1 .
2
dt
+
dt
(
= 3 3t ) + ( 6t ) 2=9(1+2t2+t 4)= 3(1+t )
2
.
1 3 1 5 1 48
S= 2  3t  3(1+t )dt=18
1
0
2 2
 (0 t2+t 4) dt=18
1
3
t + t
5 0
=
5


3 3 
61. x=acos  , y=asin  , 0   .
2
dx 2 dy 2

d
+
d
(
=  3acos  sin 
2
) 2+ ( 3asin 2 cos  ) 2=9a2sin 2 cos 
2
.
 /2 2  /2 6 2  /2 6 2
S= 
3 4 5
2  asin   3asin  cos  d =6 a sin  cos  d =  a sin  =  a
0 0 5 0 5
dx 2 dy 2 2 2
62. + = (  2sin  +2sin 2 ) + ( 2cos   2cos 2 )
d d
( 2
=4 sin   2sin  sin 2 +sin 2 + cos   2cos  cos 2 +cos 2
2
) ( 2 2
)
=4 1+1 2(cos 2 cos  +sin 2 sin  ) =8 1 cos (2   ) =8(1 cos  )
We plot the graph with parameter interval 0,2 , and see that we should only integrate between 0
and  . (If the interval 0,2 were taken, the surface of revolution would be generated twice.) Also

note that y=2sin   sin 2 =2sin  ( 1 cos  ) . So S=  2  2sin  ( 1 cos  ) 2 2 1 cos  d =
0

 0 ( 1 
3/2 2 3
8 2 cos  ) sin  d =8 2 u du [whereu = 1 cos  ,du = sin  d ] =
0
5
2 5/2 2 16 2 128
8 2 u = 2 ( 2 ) = 
5 0 5 5

18
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

3 1 dx 2 dy 2
63. x=t+t , y=t , 1 t 2 .=1+3t and =1+ , so
2 dt dt 3
t t
dx 2 dy 2 2 2

dt
+
dt
(
2
= 1+3t + 1+
2
)
3
and
t
2
S= 2 yds= 2
2
1
t
1
2
( 1+3t2) 2+ 1+
2
3
dt 59.101 .
t t

 /2  /2
64. S= csc  +sin 2 dt=4 a
2 4 2 2 4 2
2  2asin  sin  csc  +sin 2 d .
 /4  /4
 /2  /4  2 4 2
Using Simpson’s Rule with n=4 ,   = = , and f ( )=sin  csc  +sin 2 , we get
4 16
  5 3 7 
S ( 4 a ) f +4 f +2 f +4 f +f  11.0893a .
16 3 4 16 8 16 2
2 3 dx 2 dy 2 2 22 2 2
65. x=3t , y=2t , 0 t 5 + =(6t) +(6t ) =36t (1+t )
dt dt

 (dx/dt) +(dy/dt) dt= 2 (3t )6t 

5 2 2 5 2 2 52 2
S= 2 x 1+t dt=18 t 1+t 2t dt
0 0 0

= 18 
26
1
( u 1 ) u du[whereu=1+t ,du=2t dt]=18 
2 26
1
( u  u ) du
3/2 1/2

2 5/2 2 3/2 26 2 2 2 2
= 18 u  u =18  676 26   26 26  
5 3 1 5 3 5 3
24
=
5
 ( 949 26 +1 )

dx 2 dy 2
t
66. x=e  t , y=4e
t/2
, 0 t 1 .
dt
+
dt
( t )2 (
= e  1 + 2e
t/2 2
) 2t t
=e +2e +1= e +1 ( t )2 .
S= 
1
0
2 ( et t ) ( et 1) 2+ ( 2et/2) 2 dt= 1
0
2 ( et t ) ( et+1) dt

19
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

1 2t t t 1 2 1
= 2
2
e +e  ( t 1 ) e 
2
t =
0
( e2+2e 6)
/ / / /
67. If f is continuous and f (t) 0 for a t b , then either f (t)>0 for all t in a,b or f (t)<0 for
all t in a,b . Thus, f is monotonic (in fact, strictly increasing or strictly decreasing) on a,b . It
 1  1
follows that f has an inverse. Set F=g f , that is, define F by F(x)=g( f (x)) . Then x= f (t)
 1  1
f (x)=t , so y=g(t)=g( f (x))=F(x) .

68. By Formula .2.5 with y=F(x) , S= 2 F(x) 1+[F (x)] dx . But by Formula .2.2,
b / 2
a
2 2 2 2
/ 2 dy dy/dt ( dx/dt ) + ( dy/dt )
1+[F (x)] =1+ =1+ = . Using the Substitution Rule with
dx dx/dt 2
( dx/dt )
dx
x=x(t) , where a=x( ) and b=x( ) , we have (since dx= dt )
dt

2 2 2 2
 ( dx/dt ) + ( dy/dt ) dx  dx dy
S= 2 F(x(t)) dt= 2 y + dt
 2 dt  dt dt
( dx/dt )

which is Formula .2.7.

 1 dy d d  1 dy 1 d dy
69. (a)  =tan  = tan = . But
dx dt dt dx 2 dt dx
1+ ( dy/dx )
                   
y y y x x y y x x y x y x y
dy dy/dt d dy d d 1
= =  = =  = =
dx dx/dt  dt dx dt   2 dt   2   2  2  2
x x x 1+ y/x x x +y
.
dx 2 dy 2
Using the Chain Rule, and the fact that s=
t
+ dt 
0 dt dt
 2  2 1/2
ds dx 2 dy 2
= + = x +y , we have that
dt dt dt

20
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

           
x y x y x y x y
d d /dt 1
= = = . So
ds ds/dt  2  2  2  2 1/2  2  2 3/2
x +y x +y x +y
           
x y x y x y x y
d
= = = .
ds  2  2 3/2  2  2 3/2
x +y x +y

     2
dy   d y
(b) x=x and y= f (x) x=1 , x =0 and y = , y= . So
dx 2
dx

=
( 2 2
)
1 d y/dx  0 ( dy/dx )
=
2
d y/dx
2
.
2 3/2 2 3/2
1+ ( dy/dx ) 1+ ( dy/dx )

2 2 2
2 dy d y d y/dx 2
70. (a) y=x  =2x =2 . So = = , and at ( 1,1 ) ,
dx 2 2 3/2 2 3/2
dx 1+ ( dy/dx ) ( 1+4x )
2 2
= = .
3/2 5 5
5
/ d 2 5/2 / /
(b) = = 3(1+4x ) (8x)=0 x=0 y=0 . This is a maximum since >0 for x<0 and <0
dx
2
for x>0 . So the parabola y=x has maximum curvature at the origin.

     
71. x=  sin   x=1 cos   x =sin  , and y=1 cos   y=sin   y =cos  . Therefore,

=
2
cos   cos   sin 
2
=
cos   ( cos 2 +sin  ) = cos   1
2
. The top of the arch is
3/2 3/2 3/2
2
( 1 cos  ) +sin 
2
+cos  ( 1 2cos  2
+sin  )
2
( 2 2cos  )
characterized by a horizontal tangent, and from Example 2(b) in Section .2, the tangent is horizontal
when  =(2n 1) , so take n=1 and substitute  = into the expression for :
cos   1  1 1 1
= = = .
3/2 3/2 4
( 2 2cos  ) 2 2( 1)

72. (a) Every straight line has parametrizations of the form x=a+vt , y=b+wt , where a , b are
arbitrary and v , w 0 . For example, a straight line passing through distinct points ( a,b) and ( c,d )

21
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves

can be described as the parametrized curve x=a+(c a)t , y=b+(d b)t . Starting with x=a+vt , y=b+wt ,
     
v 0 w 0
we compute x=v , y=w , x = y =0 , and = =0 .
2 3/2
2
v +w ( )
(b) Parametric equations for a circle of radius r are x=rcos  and y=rsin  . We can take the center to
     
be the origin. So x= rsin   x = rcos  and y=rcos   y = rsin  . Therefore,
2 2 2 2 2
r sin  +r cos  r 1 1
= = = . And so for any  (and thus any point), = .
r r
(r 2sin 2 2
+r cos 
2
) 3/2 r
3

73. The coordinates of T are ( rcos  ,rsin  ) . Since TP was unwound from arc TA , TP has length r
1
. Also  PTQ= PTR  QTR=    , so P has coordinates
2
1 1
x=rcos  +r cos   =r ( cos  + sin  ) , y=rsin   r sin   =r ( sin    cos  ) .
2 2

74. If the cow walks with the rope taut, it traces out the portion of the involute in Exercise 73
corresponding to the range 0    , arriving at the point (  r, r ) when  = . With the rope now
fully extended, the cow walks in a semicircle of radius  r , arriving at (  r,  r ) . Finally, the cow
traces out another portion of the involute, namely the reflection about the x axis of the initial
involute path. (This corresponds to the range      0 .) Referring to the figure, we see that the total
grazing area is 2(A +A ) . A is one quarter of the area of a circle of radius  r , so
1 3 3
1 2 1 3 2 1 2
A =  ( r) =  r . We will compute A +A and then subtract A =  r to obtain A .
3 4 4 1 2 2 2 1

22
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.2 Calculus with Parametric Curves


To find A +A , first note that the rightmost point of the involute is r,r . The leftmost point of
1 2 2
 /2  /2
the involute is (  r, r ) . Thus, A +A =  ydx=
0
ydx ydx . Now
1 2  =  =0  =
2 2 2
ydx=r(sin    cos  )r cos  d =r ( sin  cos    cos  )d . Integrate:
1 1 3 1
( )
1/r  ydx=  cos  
2 2
2
2
( )
  1 sin  cos    +  +C . This enables us to compute
6 2
3
1 1 1 0  
A +A = r 2
1 2   cos  
2

2
( 2
)
 1 sin  cos   
6
3
+ 
2 
=r
2
0  
6
+
2
3
= r 2  
+
2 6
1 32
Therefore, A = A +A  A =  r , so the grazing area is
( )
1 1 2 2 6
1 32 1 32 5 32
(
2 A +A =2 )  r +  r =  r .
1 3 6 4 6

23
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 5  3
1. (a) By adding 2 to , we obtain the point 1, . The direction opposite is , so
2 2 2 2
3
 1, is a point that satisfies the r<0 requirement.
2


(b)  2,
4

5 9
2, ,  2,
4 4
(c) ( 3,2 )

( 3,2+2 ) , (  3,2+ )
2. (a) ( 3,0 )

( 3,2 ) , (  3, )

(b) 2,
7

1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

13 6
2, ,  2,
7 7

(c)  1,
2

 3
1, ,  1,
2 2

3. (a)

x=3cos =3(0)=0 and
2

y=3sin =3 ( 1 ) =3 give us the Cartesian coordinates ( 0,3) .
2

(b)
3
x =2 2 cos
4
1
=2 2  = 2 and
2

3 1
y=2 2 sin =2 2 =2 give us (  2,2 ) .
4 2

(c)
2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 1
x= 1cos =
and
3 2
 3
y= 1sin = give
3 2
1 3
us  , .
2 2

4. (a)

2 2
x=2cos = 1 and y=2sin = 3 give us (  1, 3 ) .
3 3

(b)

x=4cos 3 = 4 and y=4sin 3 =0 give

us (  4,0 ) .

(c)

5 5
x= 2cos 
6
= 3 and y= 2sin 
6
=1 give us ( 3,1 ) .

2 2  1 1 
5. (a) x=1 and y=1 r= 1 +1 = 2 and  =tan = . Since ( 1,1 ) is in the first quadrant,
1 4
 5
the polar coordinates are (i) 2, and (ii)  2, .
4 4
2 2
(b) x=2 3 and y= 2 r= ( 2 3 ) +( 2) = 12+4 = 16 =4 and
 1 2  1 1 
 =tan  =tan  = . Since ( 2 3, 2 ) is in the fourth quadrant and
2 3 3 6
3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

11 5
0   2 , the polar coordinates are (i) 4, and (ii)  4, .
6 6

4
6. (a) ( x,y ) = (  1, 3 ) , r= 1+3 =2 , tan  =y/x= 3 and ( x,y ) is in the third quadrant, so  = .
3
4 
The polar coordinates are (i) 2, and (ii)  2, .
3 3
3
(b) ( x,y ) = (  2,3) , r= 4+9 = 13 , tan  =y/x= and ( x,y ) is in the second quadrant, so
2
 1 3
 =tan 
2
+ . The polar coordinates are (i) ( 13, ) and (ii) (  13, + ).
7. The curves r=1 and r=2 represent circles with center O and radii 1 and 2. The region in the plane
satisfying 1 r 2 consists of both circles and the shaded region between them in the figure.

8. r 0 ,  /3   2 /3


9. The region satisfying 0 r<4 and   /2  < /6 does not include the circle r=4 nor the line  = .
6

10. 2<r 5 , 3 /4< <5 /4

4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

5 7
11. 2<r<3 ,   
3 3

 3
12.  1 r 1 ,   
4 4

 
3  1
13. ( r, ) = 1,  x=1cos =
and y=1sin = .
6 6 2 6 2
3 3 3 2 3 3 2
( r, ) = 3,  x=3cos = and y=3sin = . The distance between them is
4 4 2 4 2
2 2 1 2 1 2
3 3 2 1 3 2
  +  =
4
( 3 +3 2 ) +
4
( 1 3 2 )
2 2 2 2
1 1
=
4
( 3+6 6 +18 ) + ( 1 6 2 +18 ) =
2
40+6 6  6 2

14. The points r , ( 1 1 ) and (r , ) in Cartesian coordinates are (r cos 

2 2 1
,r sin 
1 1 1 ) and
(r cos 
2
,r sin 
2 2 2 ) , respectively.The square of the distance between them is
5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

2 2
( r cos   r cos  ) (+ r sin   r sin  ) =
2 2 1 1 2 2 1 1
2 2 2 2 2 2 2 2 2 2 2 2
r cos   2r r cos  cos  +r cos  + r sin   2r r sin  sin  +r sin  =r sin  +cos  +r
2 2 1 2 1 2 1 1 2 2 1 2 1 2 1 1 1 1 1 2
2 2
=r  2r r cos  (  ) +r ,
1 1 2 1 2 2

2 2
so the distance between them is r  2r r cos (  )+r .
1 1 2 1 2 2

2 2 2 2
15. r=2 x +y =2 x +y =4 , a circle of radius 2 centered at the origin.

16. rcos  =1 x=1 , a vertical line.

2 2 2 2 3 2 3 2 3
17. r=3sin   r =3rsin   x +y =3y x + y = , a circle of radius centered at
2 2 2
3 2
0, . The first two equations are actually equivalent since r =3rsin   r(r 3sin  )=0 r=0 or
2
r=3sin  . But r=3sin  gives the point r=0 (the pole) when  =0 . Thus, the single equation r=3sin 
is equivalent to the compound condition ( r=0 or r=3sin  ).

18. r=2sin  +2cos  

2
r =2rsin  +2rcos  
2 2
x +y =2y+2x ( x2 2x+1) + ( y2 2y+1) =2
2 2 2
( x 1 ) + ( y 1 ) =2 . The first implication is reversible since r =2rsin  +2rcos   r=0 or

r=2sin  +2cos  , but the curve r=2sin  +2cos  passes through the pole ( r=0 ) when  =
, so
4
r=2sin  +2cos  includes the single point of r=0 . The curve is a circle of radius 2 , centered at
( 1,1 ) .
1
19. r=csc   r=  rsin  =1 y=1 , a horizontal line 1 unit above the x  axis.
sin 

sin  2 2 2
20. r=tan  sec  = 2
 rcos  =sin   (rcos  ) =rsin   x =y , a parabola with vertex at the
cos 
origin opening upward. The first implication is reversible since cos  =0 would imply
2 2 2
sin  =rcos  =0 , contradicting the fact that cos  +sin  =1 .

21. x=3 rcos  =3 r=3/cos   r=3sec  .

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

2 2 2
22. x +y =9 r =9 r=3 . [r= 3 gives the same curve. ]

2 2 2 2 cos 
23. x= y  rcos  = r sin   cos  = rsin   r= 2
= cot  csc  .
sin 

24. x+y=9 rcos  +rsin  =9 r=9/(cos  +sin  ) .

2 2 2 2
25. x +y =2cx r =2crcos   r  2crcos  =0 r(r 2ccos  )=0 r=0 or r=2ccos  . r=0 is included

in r=2ccos  when  = +n , so the curve is represented by the single equation r=2ccos  .
2

( rcos  )  ( rsin  ) =1  r ( cos   sin  ) =1

2 2 2 2 2 2 2 2 2
26. x  y =1 r cos 2 =1 r =sec 2


27. (a) The description leads immediately to the polar equation  = , and the Cartesian equation
6
1
tan  =y/x y= tan x=
x is slightly more difficult to derive.
6 3
(b) The easier description here is the Cartesian equation x=3 .

28. (a) Because its center is not at the origin, it is more easily described by its Cartesian equation,
2 2 2
( x 2 ) + ( y 3) =5 .
(b) This circle is more easily given in polar coordinates: r=4 . The Cartesian equation is also simple:
2 2
x +y =16 .

29.  =  /6

2
30. r  3r+2=0 (r 1)(r 2)=0 r=1 or r=2

7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

2 2 2 2 1 2 1 2
31. r=sin   r =rsin   x +y =y x + y = . The reasoning here is the same as in
2 2
1 1
Exercise 17. This is a circle of radius centered at 0, .
2 2

2 2 2 3 2 2 3 2
32. r= 3cos   r = 3rcos   x +y = 3x x+ +y = . This curve is a circle of radius
2 2
3 3
centered at  ,0 .
2 2

33. r=2(1 sin  ) . This curve is a cardioid.

34. r=1 3cos  . This is a limacon.

8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

35. r= ,   0

36. r=ln  ,   1

37. r=sin 2

9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

38. r=2cos 3

39. r=2cos 4

40. r=sin 5

2
41. r =4cos 2

2
42. r =sin 2

10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

3
43. r=2cos 
2

2
44. r  =1 r= 1/  for  >0

45. r=1+2cos 2

11
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

46. r=1+2cos ( /2)

 3
47. For  =0 ,  , and 2 , r has its minimum value of about 0.5 . For  = and , r attains its
2 2
maximum value of 2 . We see that the graph has a similar shape for 0    and     2 .

48.

 
49. x=rcos  =(4+2sec  )cos  =4cos  +2 . Now, r
 ( 4+2sec  )
  or
2
3 +
 (since we need only consider 0  <2 ), so lim x= lim (4cos  +2)=2 . Also,
2 r

  /2
 + 3 
r 
 ( 4+2sec  ) 
  or  , so lim x= lim (4cos  +2)=2 .
2 2 r 
+
  /2
12
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

Therefore, lim x=2 x=2 is a vertical asymptote.

r 

+
50. y=rsin  =2sin   csc  sin  =2sin   1 . r
 ( 2 csc  )
 csc  
   (since we
need only consider 0   2 ) and so lim y= lim 2sin   1= 1 . Also r 
 ( 2 csc  ) 

r
+
 
csc 



  and so lim x= lim 2sin   1= 1 . Therefore lim y= 1 y= 1 is a horizontal asymptote.
r 
 r 

 

2
51. To show that x=1 is an asymptote we must prove lim x=1 . x=rcos  = ( sin  tan  ) cos  =sin 
r 

  2
. Now, r
 sin  tan 
  , so lim x= lim sin  =1 . Also, r 

2 r

  /2
 + 2
sin  tan  
  , so lim x= lim sin  =1 .
2 r 
+
  /2

13
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

2
Therefore, lim x=1 x=1 is a vertical asymptote. Also notice that x=sin   0 for all  , and
r 

2 
x=sin   1 for all  . And x 1 , since the curve is not defined at odd multiples of . Therefore,
2
the curve lies entirely within the vertical strip 0 x<1 .

2 3
52. The equation is x +y ( 2
) =4x2y2 , but using polar coordinates we know that x2+y2=r 2 and
6 2 2 2 2
x=rcos  and y=rsin  . Substituting into the given equation: r =4r cos  r sin  
2 2 2
r =4cos  sin   r= 2cos  sin  = sin 2 . r= sin 2 is sketched at right.

53. (a) We see that the curve crosses itself at the origin, where r=0 (in fact the inner loop corresponds
to negative r values), so we solve the equation of the limacon for r=0 csin  = 1 sin  = 1/c .
Now if c <1 , then this equation has no solution and hence there is no inner loop. But if c< 1 , then
 1  1
on the interval ( 0,2 ) the equation has the two solutions  =sin (  1/c ) and  =  sin (  1/c ) , and
 1  1
if c>1 , the solutions are  = +sin ( 1/c ) and  =2  sin ( 1/c ) . In each case, r<0 for  between the
two solutions, indicating a loop.
(b) For 0<c<1 , the dimple (if it exists) is characterized by the fact that y has a local maximum at
2
3 d y 3
 = . So we determine for what c values is negative at  = , since by the Second
2 2 2
d
2
Derivative Test this indicates a maximum: y=rsin  =sin  +csin  
2
dy d y 3
=cos  +2csin  cos  =cos  +csin 2  = sin  +2ccos 2 . At  = , this is equal to
d 2 2
d
1
 ( 1)+2c( 1)=1 2c , which is negative only for c> . A similar argument shows that for  1<c<0 , y
2
only has a local minimum at
14
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 1
 = (indicating a dimple) for c< .
2 2

54. (a) r=sin ( /2 ) . This equation must correspond to one of II, III or VI, since these are the only
graphs which are bounded. In fact it must be VI, since this is the only graph which is completed after
a rotation of exactly 4 .
(b) r=sin ( /2 ) . This equation must correspond to one of II, III or VI, since these are the only
graphs which are bounded. In fact it must be VI, since this is the only graph which is completed after
a rotation of exactly 4 .
(c) r=sin ( /4 ) . This equation must correspond to III, since this is the only graph which is
completed after a rotation of exactly 8 .
(d) r=sin ( /4 ) . This equation must correspond to III, since this is the only graph which is
completed after a rotation of exactly 8 .
  2
(e) r=sec (3 ) . This must correspond to IV, since the graph is unbounded at  = , , , and
6 2 3
so on.
  2
(f) r=sec (3 ) . This must correspond to IV, since the graph is unbounded at  = , , , and
6 2 3
so on.
(g) r= sin  . This must correspond to V. Note that r=0 whenever  is a multiple of  . This graph
is unbounded, and each time  moves through an interval of 2 , the same basic shape is repeated
(because of the periodic sin  factor) but it gets larger each time (since  increases each time we go
around.)
(h) r= sin  . This must correspond to V. Note that r=0 whenever  is a multiple of  . This graph
is unbounded, and each time  moves through an interval of 2 , the same basic shape is repeated
(because of the periodic sin  factor) but it gets larger each time (since  increases each time we go
around.)
(i) r=1+4cos 5 . This corresponds to II, since it is bounded, has fivefold rotational symmetry, and
takes only one takes only one rotation through 2 to be complete.
(j) r=1+4cos 5 . This corresponds to II, since it is bounded, has fivefold rotational symmetry, and
takes only one takes only one rotation through 2 to be complete.
(k) r=1/  . This corresponds to I, since it is unbounded at  =0 , and r decreases as  increases; in
fact r 0 as 
.
(l) r=1/  . This corresponds to I, since it is unbounded at  =0 , and r decreases as  increases; in
fact r 0 as 
.

2
55. r=2sin   x=rcos  =2sin  cos  =sin 2 , y=rsin  =2sin  

dy dy/d 2 2sin  cos  sin 2

= = = =tan 2
dx dx/d cos 2  2 cos 2

15
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 dy  
When  = =tan 2 , =tan = 3 .
6 dx 6 3
56. r=2 sin   x=rcos  =(2 sin  )cos  , y=rsin  =(2 sin  )sin  

dy dy/d (2 sin  )cos  +sin  ( cos  ) 2cos   2sin  cos  2cos   sin 2
= = = =
dx dx/d (2 sin  )( sin  )+cos  ( cos  ) 2 2  2sin   cos 2
 2sin  +sin   cos 
 dy 2(1/2) ( 3 /2 ) 1 3 /2 2 2 3
When  = , = =  = .
3 dx  2 ( 3 /2 )  ( 1/2)  3 +1/2 2 1 2 3

57. r=1/  x=rcos  =(cos  )/ , y=rsin  =(sin  )/ 

2 2
dy dy/d sin  ( 1/ )+ ( 1/ ) cos    sin  + cos 
= =  =
dx dx/d 2 2  cos    sin 
cos  ( 1/ ) ( 1/ ) sin  
dy  0+ ( 1)  
When  = , = = =  .
dx  ( 1)  (0) 1

58. r=ln   x=rcos  =ln  cos  , y=rsin  =ln  sin  

dy dy/d sin  ( 1/ ) +ln  cos   sin  + ln  cos 

= =  =
dx dx/d cos  ( 1/ )  ln  sin   cos    ln  sin 
dy sin e+eln ecos e sin e+ecos e
When  =e , = = .
dx cos e eln esin e cos e esin e

2
59. r=1+cos   x=rcos  =cos  +cos  , y=rsin  =sin  +sin  cos  

2 2
dy dy/d cos  +cos   sin  cos  +cos 2
= = =
dx dx/d  sin   2cos  sin   sin   sin 2
3 1 3 1
+ +
 dy 2 2 2 2
When  = , = = = 1 .
6 dx 1 3 1 3
   +
2 2 2 2

60. r=sin 3  x=rcos  =sin 3 cos  , y=rsin  =sin 3 sin  

dy dy/d 3cos 3 sin  +sin 3 cos 

= =
dx dx/d 3cos 3 cos   sin 3 sin 
 dy 3(0)(1/2)+1 ( 3 /2 ) 3 /2
When  = , = = = 3 .
6 dx 3(0) ( 3 /2 )  1(1/2)  1/2
16
 Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

2 2
61. r=3cos   x=rcos  =3cos  cos  , y=rsin  =3cos  sin   dy/d = 3sin  +3cos  =3cos 2 =0
 3  3 3 
 2 = or   = or . So the tangent is horizontal at , and
2 2 4 4 2 4
3 3 3 
 , sameas , . dx/d = 6sin  cos  = 3sin 2 =0 2 =0 or    =0 or
2 4 2 4
 
. So the tangent is vertical at ( 3,0 ) and 0, .
2 2

2 1 2 3
62. y=rsin  =cos  sin  +sin  = sin 2 +sin   dy/d =cos 2 +sin 2 =0 tan 2 = 1 2 =
2 4
7 3 7 3 3 3
or   = or  horizontal tangents at cos +sin , and
4 8 8 8 8 8
7 7 7 2
cos +sin , . x=rcos  =cos  +cos  sin   dx/d = sin 2 +cos 2 =0 tan 2 =1
8 8 8
 5  5   
2 = or   = or  vertical tangents at cos +sin , and
4 4 8 8 8 8 8
5 5 5
cos +sin , .
8 8 8
Note: These expressions can be simplified using trigonometric identities. For example,
  1
cos +sin = 4+2 2 .
8 8 2

63. r=1+cos   x=rcos  =cos  (1+cos  ) , y=rsin  =sin  (1+cos  )

2 2 1 
dy/d =(1+cos  )cos   sin  =2cos  +cos   1=(2cos   1)(cos  +1)=0 cos  = or  1  = ,
2 3
5 3  3 5
 , or  horizontal tangent at , , ( 0, ) [ the pole ] , and
, .
3 2 3 2 3
1 2
dx/d = (1+cos  )sin   cos  sin  = sin  (1+2cos  )=0 sin  =0 or cos  =   =0 ,  , , or
2 3
4 1 2 1 4
 vertical tangent at ( 2,0 ) , , , and , . Note that the tangent is horizontal,
3 2 3 2 3
dy/d
not vertical when  = , since lim =0 .
  dx/d

dy    1
64. =e sin  +e cos  =e (sin  +cos  )=0 sin  = cos   tan  = 1  =  +n ( n any
d 4
 ( n 1/4 ) 1
integer)  horizontal tangents at e , n .
4
dx    1
=e cos   e sin  =e ( cos   sin  ) =0 sin  =cos   tan  =1  =  +n ( n any integer) 
d 4
17
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 ( n+1/4 ) 1
vertical tangents at e , n+ .
4

65. r=cos 2  x=rcos  =cos 2 cos  , y=rsin  =cos 2 sin  

dy/d = 2sin 2 sin  +cos 2 cos  = 4sin  cos  + cos   sin  cos 
2
( 3 2
)=
cos  ( cos 2  5sin 
2
) =cos  ( 1 6sin 2 ) =0
1  3  1 1
cos  =0 or sin  =   = , ,  ,    ,  + , or 2   (where  =sin ).
6 2 2 6
 3 2 2 2
So the tangent is horizontal at  1, ,  1, , , , ,   , , + , and
2 2 3 3 3
2
,2   .
3

2 2 2
dx/d = 2sin 2 cos   cos 2 sin  = 4sin  cos   (2cos   1)sin  =sin  (1 6cos  )=0

1  1 1
sin  =0 or cos  =   =0 ,  , ,   ,  + , or 2   (where  =cos ).
6 6
2 2 2
So the tangent is vertical at ( 1,0 ) , ( 1, ) ,  , ,  ,   ,  , + , and
3 3 3
2
 ,2   .
3

66.
2
By differentiating implicitly, r =sin 2 
2r (dr/d )=2cos 2  dr/d =(1/r)cos 2 1 1
dy
, so =
r
cos 2 sin  +rcos  =
r
(
2
cos 2 sin  +r cos  )
d
1 1
= ( cos 2 sin  +sin 2 cos  ) = sin 3
r r

This is 0 when sin 3 =0  =0 ,

18
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

 4
or (restricting  to the domain of the lemniscate), so there are horizontal tangents at
3 3
4 3  4 3 4  7
, , , and ( 0,0 ) . Similarly, dx/d =(1/r)cos 3 =0 when  = or ,
4 3 4 3 6 6
4 3  4 3 7
so there are vertical tangents at , and , .
4 6 4 6

2 2 2
67. r=asin  +bcos   r =arsin  +brcos   x +y =ay+bx
1 2 2 1 2 1 2 1 2 1 2 1 2 1 2 2
2
x  bx+
2
b +y  ay+
2
a =
2
b +
2
a  x
2
b + y
2
a =
4
( )
a +b ,
1 1 1 2 2
and this is a circle with center b, a and radius a +b .
2 2 2

1 
68. These curves are circles which intersect at the origin and at a, . At the origin, the first
2 4
circle has a horizontal tangent and the second a vertical one, so the tangents are perpendicular here.

For the first circle ( r=asin  ), dy/d =acos  sin  +asin  cos  =asin 2 =a at  = and
4
2 2 
dx/d =acos   asin  =acos 2 =0 at  = , so the tangent here is vertical. Similarly, for the second
4

circle ( r=acos  ), dy/d =acos 2 =0 and dx/d = asin 2 = a at  = , so the tangent is horizontal,
4
and again the tangents are perpendicular.

69. r=1+2sin ( /2 ) . The parameter interval is 0,4 .

2
70. r= 1 0.8sin  . The parameter interval is 0,2 .

19
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

sin 
71. r=e  2cos (4 ) . The parameter interval is 0,2 .

2
72. r=sin (4 )+cos (4 ) . The parameter interval is 0,2 .

73. r=2 5sin ( /6) . The parameter interval is  6 ,6 .

74. r=cos ( /2)+cos ( /3) . The parameter interval is  6 ,6 .

75.

20
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates


It appears that the graph of r=1+sin   is the same shape as the graph of r=1+sin  , but
6
 
rotated counterclockwise about the origin by . Similarly, the graph of r=1+sin   is
6 3

rotated by . In general, the
3
graph of r= f (   ) is the same shape as that of r= f ( ) , but rotated counterclockwise through 
about the origin. That is, for any point r , ( )
on the curve r= f ( ) , the point r , +
0 0
is on the ( 0 0 )
curve r= f (   ) , since r = f ( ) = f ( ( + )  ) .
0 0 0

76.

From the graph, the highest points seem to have y 0.77 . To find the exact value, we solve dy/d =0 .
y=rsin  =sin  sin 2 
dy/d = 2sin  cos 2 +cos  sin 2
= 2sin  ( 2cos 2  1) +cos  ( 2sin  cos  )
= 2sin  ( 3cos 2  1)
1 2
In the first quadrant, this is 0 when cos  =  sin  = 
3 3
2 2 1 4 3
y=2sin  cos  =2  = 0.77 .
3 3 9

77. (a) r=sin n . From the graphs, it seems that when n is even, the number of loops in the curve
(called a rose) is 2n , and when n is odd, the number of loops is simply n .
This is because in the case of n odd, every point on the graph is traversed twice, due to the fact that
21
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

r( + )=sin n( + ) =sin n cos n +cos n sin n = { sin n

 sin n
if n is even
if n is odd

n=2

n=3

n=4

n=5
(b) The graph of r= sin n has 2n loops whether n is odd or even, since r( + )=r( ) .

n=2

22
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

n=3

n=4

n=5

78. r=1+csin n . We vary n while keeping c constant at 2 . As n changes, the curves change in the
same way as those in Exercise 77: the number of loops increases. Note that if n is even, the smaller
loops are outside the larger ones; if n is odd, they are inside.
c=2

n=2 n=3 n=4 n=5

Now we vary c while keeping n=3 . As c increases toward 0 , the entire graph gets smaller (the graphs
below are not to scale) and the smaller loops shrink in relation to the large ones. At c= 1 , the small
loops disappear entirely, and for  1<c<1 , the graph is a simple, closed curve (at c=0 it is a circle). As
c continues to increase, the same changes are seen, but in reverse order, since
1+( c)sin n =1+csin n( + ) , so the graph for c=c is the same as that for c= c , with a rotation
0 0
through  . As c
, the smaller loops get relatively closer in size to the
large ones. Note that the distance between the outermost points of corresponding inner and outer
loops is always 2 .
23
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

Maple’s animate command (or Mathematica’s Animate ) is very useful for seeing the changes
that occur as c varies.
n=3

c= 4 c= 1.4 c= 1 c= 0.8

c= 0.2 c=0 c=0.5 c=8

1 acos 
79. r= . We start with a=0 , since in this case the curve is simply the circle r=1 .
1+acos 
As a increases, the graph moves to the left, and its right side becomes flattened. As a increases
through about 0.4 , the right side seems to grow a dimple, which upon closer investigation (with
narrower   ranges) seems to appear at a 0.42 (the actual value is 2  1 ). As a 1 , this dimple
becomes more pronounced, and the curve begins to stretch out horizontally, until at a=1 the
denominator vanishes at  = , and the dimple becomes an actual cusp. For a>1 we must choose our
 1
parameter interval carefully, since r
as 1+acos  0   cos (  1/a ) . As a increases from
1 , the curve splits into two parts. The left part has a loop, which grows larger as a increases, and the
right part grows broader vertically, and its left tip develops a dimple when a 2.42 (actually, 2 +1 ).
As a increases, the dimple grows more and more pronounced. If a<0 , we get the same graph as we
do for the corresponding positive a value, but with a rotation through  about the pole, as happened
when c was replaced with  c in Exercise 78.

a=0

24
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

a=0.3

a=0.41,   0.5

a=0.42,   0.5

a=0.9,   0.5

25
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

a=1,   0.1

a=2

a=2.41,     0.2

a=2.42,     0.2

a=4

26
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

80. Most graphing devices cannot plot implicit polar equations, so we must first find an explicit
expression (or expressions) for r in terms of  , a , and c . We note that the given equation is a
2
quadratic in r , so we use the quadratic formula and find that

2
r =
2
2c cos 2 
4 2
4c cos 2  4 c  a ( 4 4
) 2
=c cos 2 
4 4 2
a  c sin 2 so
2
2 4 4 2
r= c cos 2  a  c sin 2 .

So for each graph, we must plot four curves to be sure of plotting all the points which satisfy the
given equation. Note that all four functions have period  .
We start with the case a=c=1 , and the resulting curve resembles the symbol for infinity. If we let a
decrease, the curve splits into two symmetric parts, and as a decreases further, the parts become
smaller, further apart, and rounder. If instead we let a increase from 1 , the two lobes of the curve join
together, and as a increases further they continue to merge, until at a 1.4 , the graph no longer has
2 4
dimples, and has an oval shape. As a
, the oval becomes larger and rounder, since the c and c
terms lose their significance. Note that the shape of the graph seems to depend only on the ratio c/a ,
while the size of the graph varies as c and a jointly increase.

( a,c ) = ( 1,1 )

( a,c ) = ( 0.99,1 )

( a,c ) = ( 0.9,1 )

27
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

( a,c ) = ( 0.6,1 )

( a,c ) = ( 1.01,1 )

( a,c ) = ( 4.04,4 )

( a,c ) = ( 1.3,1 )

( a,c ) = ( 1.5,1 )

( a,c ) = ( 2,1 )

( a,c ) = ( 4,1 )
81.
28
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.3 Polar Coordinates

dy dy/d
 tan   tan 
tan   tan  dx dx/d
tan  =tan (   )= = =
1+tan  tan  dy dy/d
1+ tan  1+ tan 
dx dx/d
dy dx dr dr
 tan  sin  +rcos   tan  cos   rsin 
d d d d
= =
dx dy dr dr
+ tan  cos   rsin  +tan  sin  +rcos 
d d d d
2
sin 
rcos  +r 2 2
cos  rcos  +rsin  r
= = =
2 dr 2 dr 2 dr/d
dr dr sin  cos  + sin 
cos  +  d d
d d cos 

   
82. (a) r=e  dr/d =e , so by Exercise 81, tan  =r/e =1 .  =arctan1=
4
(b) The Cartesian equation of the tangent line at ( 1,0 ) is y=x 1 , and that of the tangent line at
( 0,e /2) is y=e /2 x .

(c) Let a be the tangent of the angle between the tangent and radial lines, that is, a=tan  . Then, by
r dr 1  /a
Exercise 81 , a=  = r  r=Ce (by Theorem 10.4.2).
dr/d d a

29
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

  /4 1 2  /4 1  /4 1 1  /4 1
. A= r d = d =
2 2 2
1. r=  , 0  
4 0 2 0 2
(  ) 0 2
 d =
4

0
=
64


 /2 1  1  1  1 2 
2. r=e ,     2 . A=

2
2
e ( /2) 2d =

2
2
e d =
2
e

2
=
2
e  e ( )
2 
3. r=sin  , .   
3 3
2 /3 1 1 1 1 2 /3
A= sin  d = 
2 2 /3
(1 cos 2 )d =   sin 2
 /3 2 4  /3 4 2  /3

1 2 1 4  1 2 1 2 1 3  1 3 1  3
=  sin  + sin =    + = +
4 3 2 3 3 2 3 4 3 2 2 3 2 2 4 3 2

 1  1 1  1 1
. A= d =
2
4. r= sin  , 0   
0 2
( sin  ) 0 2
sin  d = 
2
cos 
0
= + =1
2 2
 1 1  1
. A=
2 3 3
5. r= , 0     d =  = 
0 2 6 0 6

6. r=1+sin  ,     .
2
 1 1  1  1
A=   /2 2
(1+sin  ) d = 
2
2  /2
( 1+2sin  +sin 
2
) d =
2
  /2
1+2sin  +
2
( 1 cos 2 ) d

1 1 1  1    1 3 3
=   2cos  +   sin 2 =  +2+  0  0+  0 = +2 = +1
2 2 4  /2 2 2 2 4 2 4 8

 
7. r=4+3sin  ,    .
2 2
 /2 1 1  /2
A=    /2 2
(4+3sin  ) d = 
2
2  /2
( 16+24sin  +9sin 
2
) d
1  /2
= 
2   /2
2
(
16+9sin  d [byTheorem5.5.(b) )
1  /2 1
=  2 16+9 (1 cos 2 ) d [byTheorem5.5.(a)
2 0 2
 /2 41 9 41 9  /2 41 41
=  0

2 2
cos 2 d =
2
 
4
sin 2
0
=
4
 0  ( 0 0 ) =
4

  /4 1  /4 1 1 1  /4 
. A= sin 4 d =
2
8. r=sin 4 , 0   (1 cos 8 )d =   sin 8 =
4 0 2 0 4 4 32 0 16

9. The area above the polar axis is bounded by r=3cos  for  =0 to  = /2 ( not  ) . By symmetry,
1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

 /2 1 2  /2
A =2 r d =
2
(3cos  ) d
0 2 0
2  /2  /2 1
=3  cos  d =9
2
(1+cos 2 )d
0 0 2
9 1  /2 9  9
=  + sin 2 = +0  ( 0+0 ) = .
2 2 0 2 2 4

3 3 2 9
Also, note that this is a circle with radius , so its area is  = .
2 2 4

10.
1 2 1
A = r d =
2 2 2
3(1+cos  ) d
0 2 0 2
9
=
2

2
0
( 1+2cos  +cos 
2
) d
9 1
= 
2
1+2cos  + (1+cos 2 ) d
2 0 2
9 3 1 2 27
=  +2sin  + sin 2 = 
2 2 4 0 2

2
11. The curve r =4cos 2 goes through the pole when  = /4 , so we’ll find the area for 0    /4
and multiply it by 4 .

2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

 /4 1 2  /4
A =4 r d =2 ( 4cos 2 ) d
0 2 0
 /4  /4
=8 cos 2 d =4 sin 2 =4(1 0)=4
0 0

2
12. The curve r =sin 2 goes through the pole when  = /2 , so we’ll find the area for 0    /2 and
multiply it by 2 .
 /2 1 2  /2 1  /2
A =2 r d = sin 2 d = [cos 2 ]
0 2 0 2 0
1 1
= ( 1 1)= ( 2)=1
2 2

13. One sixth of the area lies above the polar axis and is bounded by the curve r=2cos 3 for  =0 to
 = /6 .
 /6 1  /6
A =6 (2cos 3 ) d =12
2 2
cos 3 d
0 2 0
12  /6
=
2 0
 (1+cos 6 )d
1  /6 
=6  + sin 6 =6 =
6 0 6

3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

14.
1 1 2
A = (2+cos 2 ) d =  (4+4cos 2 +cos 2 )d
2 2 2
0 2 2 0
1 2 1 1
=  4+4cos 2 + + cos 4 d
2 0 2 2
1 9 1 2
=  +2sin 2 + sin 4
2 2 8 0
1 9
= (9 )=
2 2

15.
1 1 2
A = (1+2sin 6 ) d =  (1+4sin 6 +4sin 6 )d
2 2 2
0 2 2 0
1 2 1
=  1+4sin 6 +4 (1 cos 12 ) d
2 0 2
1 2
=  (3+4sin 6  2cos 12 )d
2 0
1 2 1 2
= 3  cos 6  sin 12
2 3 6 0
1 2 2
= (6   0) 0  0 =3 .
2 3 3

4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

16.
 1  /2 1
A = (2sin  +3sin 9 ) d =2
2 2
(2sin  +3sin 9 ) d
0 2 0 2
 /2
=
2 2
(4sin  +12sin  sin 9 +9sin 9 )d
0
 /2 1 9
= 2(1 cos 2 )+12 (cos (  9 ) cos ( +9 ))+ (1 cos 18 ) d
0 2 2
 /2 9 9
= (2 2cos 2 +6cos 8  6cos 10 +  cos 18 )d
0 2 2
13 3 3 1  /2 13
=   sin 2 + sin 8  sin 10  sin 18 = 
2 4 5 4 0 4

1 2 1  /2 2  /2
17. The shaded loop is traced out from  =0 to  = /2 . A= r d =  sin 2 d
0 2 2 0
1  /2 1 1 1  /2 1  
=  ( 1 cos 4 ) d =   sin 4 = =
2 0 2 4 4 0 4 2 8
 /3 1  /3  /3 1  /3 4
18. A= ( 4sin 3 ) d =8 0 sin 3 d =4 0 ( 1 cos 6 ) d =4   sin 6
2 2
=
0 2 6 0 3

 
19. r=0 3cos 5 =0 5 =   = .
2 10
 /10 1  /10 9  /10 9 1  /10 9
A= (3cos 5 ) d = 
2 2
9cos 5 d = (1+cos 10 )d =  + sin 10 =
  /10 2 0 2 0 2 10 0 20

5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

 /8 1  /8 1  /8 
20. A=2 (2cos 4 ) d =2
2
(1+cos 8 )d =2  + sin 8 =
0 2 0 8 0 4

21.

7 11
This is a limacon, with inner loop traced out between  = and
6 6
.

1
A =2
3 /2
7 /6 2
(1+2sin  ) d =
2 3 /2
7 /6
( 2
1+4sin  +4sin  d )
1
=
3 /2 3 /2
1+4sin  +4 (1 cos 2 ) d =   4cos  +2  sin 2
7 /6 2 7 /6

9 7 3 3 3
=  +2 3  = 
2 2 2 2

22. To determine when the strophoid r=2cos   sec  passes

1
through the pole, we solve r=0 2cos   =0
cos 
2 2 1 1
2cos   1=0 cos  =  cos  = 
2 2
 3 
 = or  = for 0    with  .
4 4 2
 /4 1  /4
A = 2 (2cos   sec  ) d =
2 2 2
(4cos   4+sec  )d
0 2 0
 /4 1  /4
= d =
2 2
4 (1+cos 2 ) 4+sec  ( 2+2cos 2 +sec  )d
0 2 0
6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

 /4  
=  2 +sin 2 +tan  =  +1+1  0=2
0 2 2

1  5
23. 4sin  =2
sin  =
 = or (for 0   2 ). We’ll subtract the unshaded area from the
2 6 6
shaded area for  /6    /2 and double that value.
 /2 1  /2 1  /2 1
A = 2 (4sin  ) d  2 (2) d =2
2 2 2 2
( 4sin  )  2 d
 /6 2  /6 2  /6 2
 /2  /2
=   /6
( 16sin 2  4) d =
 /6
8(1 cos 2 ) 4 d
 /2  /2
=  (4 8cos 2 )d = 4  4sin 2
 /6  /6

2 3 4
= (2  0)  4 =  +2 3
3 2 3

24. 1 sin  =1 sin  =0  =0 or  

2 1 1 2
A=
 2
( 1 sin  )  1 d = 
2
2 
( sin 2  2sin  ) d
1 2 1 1 2
= 
4 
( 1 cos 2  4sin  ) d =
4
 
2
sin 2 +4cos 

1
=  +2
4
7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

2
25. To find the area inside the leminiscate r =8cos 2 and outside the circle r=2 , we first note that
2 1 1
the two curves intersect when r =8cos 2 and r=2; i.e., when cos 2 = . For   <   , cos 2 =
2 2

2 =  /3 or  5 /3
 =  /6 or  5 /6 . The figure shows that the desired area is 4 times the
area between the curves from 0 to  /6 . Thus,
 /6 1 1 2  /6
A = 4 (8cos 2 ) (2) d =8 (2cos 2  1)d
0 2 2 0
 /6
= 8 sin 2   =8 ( 3 /2  /6 ) =4 3  4 /3
0

2
26. To find the area inside the leminiscate r =8cos 2 and outside the circle r=2 , we first note that
2 1 1
the two curves intersect when r =8cos 2 and r=2; i.e., when cos 2 = . For   <   , cos 2 =
2 2

2 =  /3 or  5 /3
 =  /6 or  5 /6 . The figure shows that the desired area is 4 times the
area between the curves from 0 to  /6 . Thus,
 /6 1 1 2  /6
A = 4 (8cos 2 ) (2) d =8 (2cos 2  1)d
0 2 2 0
 /6
= 8 sin 2   =8 ( 3 /2  /6 ) =4 3  4 /3
0

8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

1  
27. 3cos  =1+cos 
cos  =   = or  .
2 3 3
 /3 1
A = 2
2 2
( 3cos  )  ( 1+cos  ) d
0 2
 /3
=  0
( 8cos 2  2cos   1 d )
 /3
=  0
4(1+cos 2 ) 2cos   1 d
 /3
=  0
( 3+4cos 2  2cos  ) d
 /3
= 3 +2sin 2  2sin 
0
=  + 3 3 =

28. Note that r=1+cos  goes through the pole when  = , but r=3cos  goes through the pole when
 = /2 .
 1  /2 1
A =2 (1+cos  ) d  2
2 2
(3cos  ) d
 /3 2  /3 2
 1 9  /2
= 1+2cos  + (1+cos 2 ) d   (1+cos 2 )d
 /3 2 2  /3
1 1  9 1  /2
=  +2sin  +  + sin 2   + sin 2
2 2  /3 2 2  /3
9 9  1 
=   3   3 =
8 2 6 4 4
9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

29.
 /4 1  /4 1
A =2 sin  d =
2
( 1 cos 2 ) d
0 2 0 2
1 1  /4 1  1
=   sin 2 =   1  (0 0)
2 2 0 2 4 2
1 1
=  
8 4

30. r=sin 2 takes on both positive and negative values. sin  = 2 = 2sin  cos  
1
sin  ( 1 2cos  ) =0 . From the figure we can see that the intersections occur where cos  = , or
2
 2
 = and .
3 3
 /3 1  1
A=2  sin  d +
2 /2 2
sin 2 d
0 2  /3 2
 /3 1  /2 1
= (1 cos 2 )d + (1 cos 4 )d
0 2  /3 2

1 1  /3 1 1  /2 4  3 3
=   sin 2 +   sin 4 =
2 2 0 2 4  /3 16

10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

sin 2
31. sin 2 =cos 2  =1 tan 2 =1
cos 2
 
2 =   =

4 8
 /8 1  /8 1
A = 8 2 sin 2 d =8
2
(1 cos 4 )d
0 2 0 2
1  /8  1 1
=4   sin 4 =4   1 =   1
4 0 8 4 2

2 1  5  5
32. 2sin 2 =1  sin 2 =  2 = or   = or .
2 6 6 12 12
 /12 1  /4 1 2
A=4  0 2
 2sin 2 d +
 /12 2
( )
1 d
 /12  /4
=  2cos 2 + 2
0  /12
3 1 1
= 2  1 +2   
2 4 12

= 2 3+
3

11
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

33.
  /6 1  /2 1 2
 ( 3+2sin  ) d +
2
A =2 2 d
  /2 2   /6 2
  /6  /2
=
 /2
( 9+12sin  2
+4sin  ) d + 4
  /6

  /6 8 19 11 3
= 9  12cos  +2  sin 2 + = 
  /2 3 3 2

 1
34. Let  =tan ( b/a ) . Then
 1  /2 1
A = asin  ) d +
2 2
( ( bcos  ) d
0 2  2
1 2 1  1 2 1  /2
= a   sin 2 + b  + sin 2
4 2 0 4 2 
1 1 2 1 2 2
2 2
(
=  a b +  b
4 8 4
) (
a +b ( sin  cos  ) )
1 2 2  1 1 2 1
=
4
( )
a  b tan ( b/a ) +  b 
8 4
ab

12
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

1 1
35. The darker shaded region (from  =0 to  =2 /3 ) represents of the desired area plus of the
2 2
1
area of the inner loop. From this area, we’ll subtract of the area of the inner loop (the lighter
2
shaded region from  =2 /3
to  = ), and then double that difference to obtain the desired area.
1 1 2  1 1 2
 
2 /3
A =2 +cos  d  +cos  d
0 2 2 2 /3 2 2
1  1
= d  
2 /3 2 2
+cos  +cos  +cos  +cos  d
0 4 2 /3 4
1 1
=
2 /3
+cos  + (1+cos 2 ) d
0 4 2
 1 1
  2 /3 4
+cos  + (1+cos 2 ) d
2
  sin 2 2 /3   sin 2 
= +sin  + +  +sin  + +
4 2 4 0 4 2 4 2 /3

3  
3    3  3
= +   + + + + + 
6 2 3 8 4 2 6 2 3 8
 3 1
= + 3 = ( +3 3 )
4 4 4

2 4 2 41
36. r=0 1+2cos 3 =0 , (for 0 3  2 )  =
cos 3 = , . The 3 =
3 3 9 92
1 1
darker shaded region (from  =0 to  =2 /9 ) represents of the desired area plus of the area of
2 2
1
the inner loop. From this area, we’ll subtract of the area of the inner loop (the lighter shaded
2
region from  =2 /9 to  = /3 ), and then double that difference to obtain the desired area.

1  /3 1
 
2 /9 2 2
A=2 (1+2cos 3 ) d  (1+2cos 3 ) d
0 2 2 /9 2

13
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

Now
2 2 2 1
r = (1+2cos 3 ) =1+4cos 3 +4cos 3 =1+4cos 3 +4 2 (1+cos 6 )
= 1+4cos 3 +2+2cos 6 =3+4cos 3 +2cos 6
4 1
and  r d =3 + sin 3 + sin 6 +C , so
2
3 3
4 1 2 /9 4 1  /3
A = 3 + sin 3 + sin 6  3 + sin 3 + sin 6
3 3 0 3 3 2 /9

2 4 3 1  3 2 4 3 1  3
= +  +   0  ( +0+0) +  + 
3 3 2 3 2 3 3 2 3 2
4 4 1 
= + 3 3  = + 3
3 3 3 3


37. The two circles intersect at the pole since ( 0,0 ) satisfies the first equation and 0, the
2
1 
second. The other intersection point , occurs where sin  =cos  .
2 4

 3 
38. 2cos 2 = 2 cos 2 = 1  =0 , , , or , so the points are ( 2,0 ) , 2, , ( 2, ) ,
2 2 2
3
and 2, .
2

14
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates


39. The curves intersect at the pole since 0, satisfies r=cos  and ( 0,0 ) satisfies r=1 cos  .
2
1  5
Now cos  =1 cos   2cos  =1 cos  =   = or  the other intersection points are
2 3 3
1  1 5
, and , .
2 3 2 3


40. Clearly the pole lies on both curves. sin 3 =cos 3  tan 3 =1 3 = +n ( n any integer) 
4
   5 3 1 
 = + n  = , , or , so the three remaining intersection points are , ,
12 3 12 12 4 2 12
1 5 1 3
 , , and , .
2 12 2 4

41. The pole is a point of intersection. sin  =sin 2 =2sin  cos 

sin  ( 1 2cos  ) =0
sin  =0 or
1   3  3 2
cos  =   =0 ,  , ,  , and , (by symmetry) are the other
2 3 3 2 3 2 3
intersection points.

15
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates


42. Clearly the pole is a point of intersection. sin 2 =cos 2  tan 2 =1 2 = +2n (since sin 2
4
  9
and cos 2 must be positive in the equations)   = +n   = or . So the curves also
8 8 8
intersect at
1  1 9
, and , .
2 8 2 8
4 4

43.

From the first graph, we see that the pole is one point of intersection. By zooming in or using the
cursor, we find the   values of the intersection points to be   0.88786 0.89 and     2.25 .
(The first of these values may be more easily estimated by plotting y=1+sin x and y=2x in rectangular
coordinates; see the second graph.)
16
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

By symmetry, the total area contained is twice the area contained in the first quadrant, that is,
 1  /2 1   /2 1
A = 2 2 ) d +2 1+sin  ) d = 4 d +
2 2 2
( ( 1+2sin  + (1 cos 2 ) d
0 2  2 0  2
4 3  1 1  /2
=  +   2cos  +   sin 2
3 0 2 4 
4 3   1 1
=  + +    2cos  +   sin 2  3.4645
3 2 4 2 4

44.

From the first graph, it appears that the   values of the points of intersection are   0.57504 0.58
and     2.57 . (These values may be more easily estimated by plotting y=3+sin 5x and y=6sin x in
rectangular coordinates; see the second graph.) By symmetry, the total area enclosed in both curves is
 1  /2 1   /2
A = 2
0 2
( 6sin  ) d +2
2
 2
( 3+sin 5 ) d = 36sin  d +
2
0
2

2
9+6sin 5 +sin 5 d ( )
 1  /2 1
=  36 (1 cos 2 )d + 9+6sin 5 + (1 cos 10 ) d
0 2  2
1 1  6 1 1  /2
= 36   sin 2 + 9  cos 5 +   sin 10  10.41
2 4 0 5 2 20 

45.
 /3  /3
L = r + ( dr/d ) d = (3sin  ) +(3cos  ) d =
b 2 2 2 2 2 2
9(sin  +cos  ) d
a 0 0
 /3  /3 
=3 d =3  =3 = .
0 0 3
17
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

3 3
As a check, note that the circumference of a circle with radius is 2 =3 , and since  =0 to
2 2
 1 1
 = traces out of the circle (from  =0 to  = ), (3 )= .
3 3 3

46.

L = b
a
r + ( dr/d ) d =
2 2 2
0
( e2 ) 2+ ( 2e2 ) 2 d =
2
0
4
e +4e
4
d =
2
0
5e
4
d

5 2 5
= 5
2
0
2
e d =
2
e
2
0
=
2
( e4  1)
47.

L = b
a
r + ( dr/d ) d =
2 2 2
0
( 2) 2+ ( 2 ) d =
2 2
0

4
+4
2
d

=
2
0

2
( +4 d =
2
) 2
0
 
2
+4 d

2 1
Now let u= +4 , so that du=2 d  d = du and
2

2
1 1 2 3/2 4  ( 2
+1 ) 1 3/2
( ) 3/2 43/2
=
4 +4 2
u du=  u = 4  +1

2 2
  +4 d 4 2 2 3 4 3
0
8 3/2
=
3
( 2
+1 )  1

48.
 2
L =
b
a
r + ( dr/d ) d =
2 2 2
0

2
+1 d =
2

2
+1 +
1
2
ln  +  ( 2
+1 ) 0

= 4
2
+1 +
1
2
ln 2 + 4 ( 2
+1 )
49. L=
b
a
r + ( dr/d ) d =
2 2 2
0
( 2) 2+ ( 2 ) d =
2 2
0

4
+4
2
d

=
2
0

2
( 2
)
+4 d =
2
0
 
2
+4 d
2 1
Now let u= +4 , so that du=2 d  d = du and
2

18
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

2
1 1 2 3/2 4  ( 2
+1 ) 1 3/2
( ) 3/2 43/2 8
( ) 3/2 1
 +4 d =
2 2 4 +4 2 2
  u du=  u = 4  +1 = +1
0 4 2 2 3 4 3 3

50. The curve r=4sin 3 is completely traced with 0    .

2 dr 2 2 2
r + =(4sin 3 ) +(12cos 3 ) 
d

L=
2 2
16sin 3 +144cos 3 d  26.7298
0


51. The curve r=sin is completely traced with 0   4 .
2
dr 2  1  2  1 
L=
2 2 4 2 2
r + =sin + cos  sin + cos d
d 2 2 2 0 2 4 2
 9.6884


52. The curve r=1+cos is completely traced with 0   6 .
3
2 dr 2  2 1  2
r + = 1+cos +  sin 
d 3 3 3
 2 1 
L=
6 2
1+cos + sin d  19.6676
0 3 9 3

19
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

4
53. The curve r=cos ( /4) is completely traced with 0   4 .
4 2 3 1 2
2 2
r +(dr/d ) =[cos ( /4)] + 4cos ( /4) (  sin ( /4) )  4
8 6 2
=cos ( /4 ) +cos ( /4 ) sin ( /4 )
6 2 2
=cos ( /4 ) [cos ( /4 ) +sin ( /4 ) ]
6
=cos ( /4 )

L = cos ( /4) d =
4 6 4 3
cos ( /4) d
0 0
 /2 1
=2 ] =8
2 3 3 3
cos ( /4)d [ since cos ( /4) 0 for 0   2 cos udu u= 
0 0 4
1 2  /2 8 16
=8 (2+cos u)sin u = [(2 1) (3 0)]=
3 0 3 3

2
54. The curve r=cos ( /2) is completely traced with 0   2 .
2 2 1 2
2 2
r +(dr/d ) =[cos ( /2)] + 2cos ( /2) ( sin ( /2)) 2
4 2 2
=cos ( /2 ) +cos ( /2 ) sin ( /2 )
2 2 2
=cos ( /2 ) [cos ( /2 ) +sin ( /2 ) ]
2
=cos ( /2 )

20
 Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates


L = cos ( /2) d = cos ( /2) d =2
2 2 2
cos ( /2)d [ since cos ( /2) 0 for 0    ]
0 0 0
 /2 1  /2
=4 cos udu u=  =4 sin u =4(1 0)=4
0 2 0

55. (a) From ( .2.7),

S = 2 y
b 2 2
a
( dx/d ) + ( dy/d ) d

= 2 y r + ( dr/d ) d [fromthederivationofEquation.4.5
b 2 2
a

= 2 rsin 
b 2 2
r + ( dr/d ) d
a

2  
(b) The curve r =cos 2 goes through the pole when cos 2 =0  2 =   = . We’ll rotate the
2 4
 2
curve from  =0 to  = and double this value to obtain the total surface area generated. r =cos 2
4
2 2 2
dr dr sin 2 sin 2
 2r = 2sin 2  = = .
d d 2 cos 2
r

 /4
S = 2
0
2 cos 2 sin  cos 2 + sin 2 ( 2
) /cos 2 d

2 2
 /4 cos 2 +sin 2  /4 1
= 4  0
cos 2 sin 
cos 2
d =4  0
cos 2 sin  d
cos 2

21
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.4 Areas and Lengths in Polar Coordinates

 /4  /4 2
= 4  0
sin  d =4  cos 
0
= 4
2
 1 =2 ( 2 2 )


is the same as rotation around the y axis, that is, S= 2 xds where
b
56. (a) Rotation around  =
2 a
2 2
ds= (dx/dt) +(dy/dt) dt for a parametric equation, and for the special case of a polar equation,
2 2 2 2
x=rcos  and ds= (dx/d ) +(dy/d ) d = r +(dr/d ) d . Therefore, for a polar equation rotated

, S= 2 rcos 
b 2 2
around  = r +(dr/d ) d .
2 a
(b) As in the solution for Exercise 55(b), we can double the surface area generated by rotating the

curve from  =0 to  = to obtain the total surface area.
4
 /4
S = 2
2
2 cos 2 cos  cos 2 +(sin 2 )/cos 2 d
0

2 2
 /4 cos 2 +sin 2
= 4  0
cos 2 cos 
cos 2
d

 /4 1
= 4  0
cos 2 cos  d
cos 2
 /4  /4 2
= 4  0
cos  d =4 sin 
0
=4
2
 0 =2 2 

22
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2 1 1 1 1
1. x=2y  y= x . 4 p= , so p= . The vertex is ( 0,0 ) , the focus is ,0 , and the directrix
2 2 8 8
1
is x= .
8

2 2
2. 4y+x =0 x = 4y . 4 p= 4 , so p= 1 . The vertex is ( 0,0 ) , the focus is ( 0, 1 ) , and the directrix is
y=1 .

2 2 1 1 1 1
3. 4x = y x = y . 4 p= , so p= . The vertex is ( 0,0 ) , the focus is 0, , and the
4 4 16 16
1
directrix is y= .
16

2
4. y =12x . 4 p=12 , so p=3 . The vertex is ( 0,0 ) , the focus is ( 3,0 ) , and the directrix is x= 3 .

1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2
5. ( x+2 ) =8 ( y 3) . 4 p=8 , so p=2 . The vertex is (  2,3) , the focus is (  2,5) , and the directrix is y=1
.

2 1 5
6. x 1= ( y+5) . 4 p=1 , so p= . The vertex is ( 1, 5) , the focus is , 5 , and the directrix is
4 4
3
x= .
4

2 2 2
7. y +2y+12x+25=0  y +2y+1= 12x 24 (y+1) = 12(x+2) . 4 p= 12 , so p= 3 . The vertex is
(  2, 1 ) , the focus is (  5, 1 ) , and the directrix is x=1 .

2 2 2 2 1 2 1
8. y+12x 2x =16  2x  12x=y 16 2(x  6x+9)=y 16+18 2(x 3) =y+2 (x 3) =
(y+2) . 4 p=
2 2
1 15 17
, so p= . The vertex is ( 3, 2 ) , the focus is 3, , and the directrix is y= .
8 8 8

2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2
9. The equation has the form y =4 px , where p<0 . Since the parabola passes through (  1,1 ) , we
2 2 2 1
have 1 =4 p( 1) , so 4 p= 1 and an equation is y = x or x= y . 4 p= 1 , so p= and the focus is
4
1 1
 ,0 while the directrix is x= .
4 4

2
10. The vertex is ( 2, 2 ) , so the equation is of the form ( x 2 ) =4 p(y+2) , where p>0 . The point
2 1
( 0,0 ) is on the parabola, so 4=4 p(2) and 4 p=2 . Thus, an equation is ( x 2 ) =2(y+2) . 4 p=2 , so p=
2
3 5
and the focus is 2, while the directrix is y= .
2 2

2 2
x y 2 2
11. + =1 a= 9 =3 , b= 5 , c= a  b = 9 5=2 . The ellipse is centered at (0,0) , with
9 5
vertices at ( 3,0) . The foci are ( 2,0) .

2 2
x y 2 2
12. + =1 a= 100 =10 , b= 64 =8 , c= a  b = 100 64 =6 . The ellipse is centered at
64 100
( 0,0 ) , with vertices at ( 0, 10 ) . The foci are ( 0, 6 ) .

3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
2 2 x y 2 2
13. 4x +y =16 + =1 a= 16 =4 , b= 4 =2 , c= a  b = 16 4 =2 3 . The ellipse is
4 16
centered at (0,0) , with vertices at (0, 4) . The foci are ( 0, 2 3 ) .

2 2
2 2 x y 25 5 2 2 25 21 21
14. 4x +25y =25 + =1 = , b= 1 =1 , c= a  b =
a=  1= =
25/4 1 4 2 4 4 2
5 21
. The ellipse is centered at ( 0,0 ) , with vertices at  ,0 . The foci are  ,0 .
2 2

2 2
2 2 2 2 2 2 ( x 1 ) y
15. 9x  18x+4y =27 9(x  2x+1)+4y =27+9 9(x 1) +4y =36 + =1 a=3 , b=2 ,
4 9
c= 5  center ( 1,0 ) , vertices ( 1, 3) , foci ( 1, 5)

4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
16. x  6x+2y +4y= 7
2 2 2 2
x  6x+9+2(y +2y+1)= 7+9+2 (x 3) +2(y+1) =4
2 2
( x 3) ( y+1 )
+ =1 a=2 , b= 2 =c center ( 3, 1 ) , vertices ( 1, 1 ) and ( 5, 1 ) , foci ( 3 2 , 1 )
4 2

2 2
x y 2 2
17. The center is ( 0,0 ) , a=3 , and b=2 , so an equation is + =1 . c= a  b = 5 , so the foci
4 9
are ( 0, 5 ) .

2 2
( x 2 ) ( y 1 )
18. The ellipse is centered at ( 2,1 ) , with a=3 and b=2 . An equation is + =1 .
9 4
2 2
c= a  b = 5 , so the foci are ( 2 5,1 ) .

2 2
x y
19.  =1 a=12 , b=5 , c= 144+25=13 center ( 0,0 ) , vertices (  12,0 ) , foci (  13,0 ) ,
144 25
5
asymptotes y= x.
12
Note: It is helpful to draw a 2a  by 2b rectangle whose center is the center of the hyperbola. The
asymptotes are the extended diagonals of the rectangle.

5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
y x 2 2
20.  =1 a=4 , b=6 , c= a +b = 16+36 = 52 =2 13 . The center is ( 0,0 ) , the vertices are
16 36
a 2
( 0, 4 ) , the foci are ( 0, 2 13 ) , and the asymptotes are the lines y= x= x.
b 3

2 2
2 2y x
21. y  x =4  =1 a= 4 =2=b , c= 4+4 =2 2  center ( 0,0 ) , vertices ( 0, 2 ) , foci
4 4
( 0, 2 2 ) , asymptotes y= x

2 2
2 2 x y
22. 9x  4y =36  =1 a= 4 =2 , b= 9 =3 , c= 4+9 = 13  center ( 0,0 ) , vertices (  2,0 ) ,
4 9
3
foci (  13,0 ) , asymptotes y= x
2

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
23. 2y  4y 3x +12x= 8
2 2
2 2 2 2 ( x 2 ) ( y 1 )
2(y  2y+1) 3(x  4x+4)= 8+2 12  2(y 1)  3(x 2) = 18   =1 a= 6 , b=3 ,
6 9
3
c= 15  center ( 2,1 ) , vertices ( 2 6 ,1 ) , foci ( 2 15,1 ) , asymptotes y 1= ( x 2 ) or
6
6
y 1= (x 2)
2

2 2
24. 16x +64x 9y  90y=305
2 2
2 2 2 2 ( x+2 ) ( y+5)
16(x +4x+4) 9(y +10y+25)=305+64 225  16(x+2)  9(y+5) =144   =1 a=3 ,
9 16
b=4 , c=5 center (  2, 5) , vertices (  5, 5) and ( 1, 5) , foci (  7, 5) and ( 3, 5) , asymptotes
4
y+5= ( x+2 )
3

7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2 1
25. x =y+1 x =1(y+1) . This is an equation of a parabola with 4 p=1 , so p= . The vertex is
4
3
( 0, 1 ) and the focus is 0, .
4

2 2 2 2
26. x =y +1 x  y =1 . This is an equation of a hyperbola with vertices (  1,0 ) . The foci are at
(  1+1 ,0 ) = (  2 ,0 ) .
2 2
2 2 2 2 2 2 2 2 x (y 1)
27. x =4y 2y  x +2y  4y=0 x +2(y  2y+1)=2 x +2(y 1) =2 + =1 . This is an
2 1
equation of an ellipse with vertices at (  2 ,1 ) . The foci are at (  2 1 ,1 ) = (  1,1 ) .
2 2 2 3
28. y  8y=6x 16 y  8y+16=6x (y 4) =6x . This is an equation of a parabola with 4 p=6 , so p=
2
3
. The vertex is ( 0,4 ) and the focus is ,4 .
2

2
2 2 2 2 (y+1) 2 2 2
29. y +2y=4x +3 y +2y+1=4x +4 ( y+1 )  4x =4  x =1 . This is an equation of a
4
hyperbola with vertices ( 0, 1 2 ) = ( 0,1 ) and ( 0, 3) . The foci are at ( 0, 1 4+1 ) = ( 0, 1 5 ) .
1 2
x+
2 2 2 1 2 1 2 2 2 2
30. 4x +4x+y =0 4 x +x+ +y =1 4 x+ +y =1 +y =1 . This is an
4 2 1/4
1 1
equation of an ellipse with vertices  ,0 1 =  , 1 . The foci are at
2 2
1 1 1
 ,0 1 =  , 3 /2 .
2 4 2

31. The parabola with vertex ( 0,0 ) and focus ( 0, 2 ) opens downward and has p= 2 , so its equation
2
is x =4 py= 8y .

32. The parabola with vertex ( 1,0 ) and directrix x= 5 opens to the right and has p=6 , so its equation
2
is y =4 p(x 1)=24(x 1) .

33. The distance from the focus ( 4,0) to the directrix x=2 is 2 ( 4)=6 , so the distance from the focus
1
to the vertex is (6)=3 and the vertex is (  1,0 ) . Since the focus is to the left of the vertex, p= 3 . An
2
2 2
equation is y =4 p(x+1) y = 12(x+1) .

8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

34. The distance from the focus (3,6) to the vertex ( 3,2 ) is 6 2=4 . Since the focus is above the
2 2
vertex, p=4 . An equation is (x 3) =4 p(y 2) (x 3) =16(y 2) .

2 2 2
35. The parabola must have equation y =4 px , so (  4 ) =4 p(1) p=4 y =16x .

2 2
36. Vertical axis  ( x h ) =4 p(y k) . Substituting (  2,3) and ( 0,3) gives (  2 h ) =4 p(3 k) and
2 2 2 2 2
(  h ) =4 p(3 k) (  2 h ) = (  h )  4+4h+h =h  h= 1 1=4 p(3 k) . Substituting ( 1,9 ) gives
2 1 1
1 (  1 ) =4 p(9 k) 4=4 p(9 k) . Solving for p from these equations gives p= = 
4(3 k) 9 k
1 2 1 2
4(3 k)=9 k k=1 p=  ( x+1 ) = (y 1) 2x +4x y+3=0 .
8 2

37. The ellipse with foci (  2,0 ) and vertices (  5,0 ) has center ( 0,0 ) and a horizontal major axis,
2 2
x 2 y 2
with a=5 and c=2 , so b= a  c = 21 . An equation is + =1 .
25 21

38. The ellipse with foci ( 0, 5) and vertices ( 0, 13) has center ( 0,0 ) and a vertical major axis, with
2 2
2 x y 2
c=5 and a=13 , so b= a  c =12 . An equation is + =1 .
144 169

39. Since the vertices are ( 0,0 ) and (0,8) , the ellipse has center ( 0,4 ) with a vertical axis and a=4 .
2 2 2 2
The foci at ( 0,2 ) and ( 0,6 ) are 2 units from the center, so c=2 and b= a  c = 4  2 = 12 . An
2 2 2 2
(x 0) (y 4) x (y 4)
equation is + =1 + =1 .
2 2 12 16
b a

40. Since the foci are (0, 1) and (8, 1) , the ellipse has center ( 4, 1 ) with a horizontal axis and c=4 .
2 2 2 2
The vertex ( 9, 1 ) is 5 units from the center, so a=5 and b= a  c = 5  4 = 9 . An equation is
2 2 2 2
(x 4) (y+1) (x 4) (y+1)
+ =1  + =1 .
2 2 25 9
a b

1 2 1 2
41. Center ( 2,2 ) , c=2 , a=3 b= 5  (x 2) + (y 2) =1
9 5

42. Center ( 0,0 ) , c=2 , major axis horizontal 

9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
x y 2 2 2 2
+ =1 and b =a  c =a  4 . Since the ellipse passes through ( 2,1 ) , we have
2 2
a b
2 9+ 17 2 1+ 17
2a= PF + PF = 17+1 a= and b = , so the ellipse has equation
1 2 2 2
2 2
2x 2y
+ =1 .
9+ 17 1+ 17

2 1 2
43. Center ( 0,0 ) , vertical axis, c=3 , a=1 b= 8 =2 2  y x =1
8

1 2 1 2
44. Center ( 0,0 ) , horizontal axis, c=6 , a=4 b=2 5  x y =1
16 20

1 2 1 2
45. Center ( 4,3) , horizontal axis, c=3 , a=2 b= 5  (x 4)  (y 3) =1
4 5

1 2 1 2
46. Center ( 2,3) , vertical axis, c=5 , a=3 b=4 (y 3)  (x 2) =1
9 16

b 1 2 1 2
47. Center ( 0,0 ) , horizontal axis, a=3 , =2 b=6 x y =1
a 9 36

2 2 2 2
48. Center ( 4,2 ) , horizontal axis, asymptotes y 2= (x 4) c=2 , b/a=1 a=b c =4=a +b =2a 
2 1 2 1 2
a =2 (x 4)  (y 2) =1
2 2

49. In Figure 8, we see that the point on the ellipse closest to a focus is the closer vertex (which is a
distance a c from it) while the farthest point is the other vertex (at a distance of a+c ). So for this
lunar orbit, ( a c ) + ( a+c ) =2a= ( 1728+110 ) + ( 1728+314 ) , or a=1940 ; and ( a+c )  ( a c ) =2c=314 110
2 2
2 2 2 x y
, or c=102 . Thus, b =a  c =3 , 753 , 196 , and the equation is + =1 .
3,763,600 3,753,196

50. (a) Choose V to be the origin, with x axis through V and F . \ Then F is ( p,0 ) , A is ( p,5) , so
2 2 5 2
substituting A into the equation y =4 px gives 25=4 p so p= and y =10x .
2
(b) x=11 y= 110  CD =2 110

51. (a) Set up the coordinate system so that A is (  200,0 ) and B is ( 200,0 ) .
PA  PB = ( 1200 ) ( 980 ) =1 , 176 , 000 ft
10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2
2450 1225 2 2 2 3,339,375 121x 121y
= mi =2a a= , and c=200 so b =c  a =   =1 .
11 11 121 1,500,625 3,339,375
2 2
( 121 ) ( 200 ) 121y 133,575
(b) Due north of B x=200  =1 y=  248 mi
1,500,625 3,339,375 539

2 2 2 2 2 2 2 2
52. PF  PF = 2a ( x+c ) +y  ( x c ) +y = 2a ( x+c ) +y = ( x c ) +y  2a
1 2
2 2 2 2 2 2 2 2 2 2
( x+c ) +y = ( x c ) +y +4a  4a ( x c ) +y  4cx 4a = 4a ( x c ) +y 
2 2
c x  2a cx+a =a
2 4 2
( x2 2cx+c2+y2)  ( c2 a2) x2 a2y2=a2 ( c2 a2)  2 2 2 2 2 2 2
b x  a y =a b (where b =c  a
2 2

2 2
x y
)  =1
2 2
a b

53. The function whose graph is the upper branch of this hyperbola is concave upward. The function
2
is y= f (x)=a 1+
x
2
=
a
b
2
b +x , so y =
2 / a
b
( 2 2
x b +x )  1/2 and
b

y
/ /
=
a
b
( b2+x2)  1/2 x2 ( b2+x2)  3/2 =ab b +x ( 2 2  3/2
) >0 for all x , and so f is concave upward.

54. We can follow exactly the same sequence of steps as in the derivation of Formula 4, except we
use the points ( 1,1 ) and (  1, 1 ) in the distance formula (first equation of that derivation) so
2 2 2 2
(x 1) +(y 1) + (x+1) +(y+1) =4 will lead (after moving the second term to the right, squaring,
2 2
and simplifying) to 2 (x+1) +(y+1) =x+y+4 , which, after squaring and simplifying again, leads to
2 2
3x  2xy+3y =8 .

2 2
x y
55. (a) If k>16 , then k 16>0 , and + =1 is an ellipse since it is the sum of two squares on the
k k 16
left side.
2 2
x y
(b) If 0<k<16 , then k 16<0 , and + =1 is a hyperbola since it is the difference of two
k k 16
squares on the left side.
(c) If k<0 , then k 16<0 , and there is no curve since the left side is the sum of two negative terms,
which cannot equal 1.
2 2 2 2 2
(d) In case (a), a =k , b =k 16 , and c =a  b =16 , so the foci are at (  4,0 ) . In case (b), k 16<0 , so

11
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2 2 2 2 2
a =k , b =16 k , and c =a +b =16 , and so again the foci are at (  4,0 ) .

2 / / 2p 2p 2
56. (a) y =4 px 2yy =4 p y = , so the tangent line is y y = (x x ) yy  y =2 p(x x )
y 0 y 0 0 0 0
0
yy  4 px =2 px 2 px  yy =2 p(x+x ) .
0 0 0 0 0

(b) The x intercept is  x .

0

57. Use the parametrization x=2cos t , y=sin t , 0 t 2 to get

 /2  /2  /2
L=4 (dx/dt) +(dy/dt) dt=4 4sin t+cos t dt=4
2 2 2 2 2
3sin t+1 dt
0 0 0

 /2 0  2
Using Simpson’s Rule with n=10 ,  t= = , and f (t)= 3sin t+1 , we get
10 20

4   2 8 9 
L f (0)+4 f +2 f +


+2 f +4 f +f  9.69
3 20 20 20 20 20 2

1 10 9
58. The length of the major axis is 2a , so a= (1.18 10 )=5.9 10 . The length of the minor axis is
2
2 2
1 10 9 x y
2b , so b= (1.14 10 )=5.7 10 . An equation of the ellipse is + =1 , or converting into
2 2 2
a b
parametric equations, x=acos  and y=bsin  . So

 /2  /2
L=4 (dx/d ) +(dy/d ) d =4
2 2 2 2 2 2
a sin  +b cos  d
0 0

 /2 0  2 2 2 2
Using Simpson’s Rule with n=10 ,   = = , and f ( )= a sin  +b cos  , we get
10 20
L 4
S
10

12
 Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

  2 8 9 
= 4
f (0)+4 f +2 f +


+2 f +4 f +f
20
3 20 20 20 20 2
10
 3.64 10 km

2 2 / 2
x y 2x 2yy / b x
59. + =1 + =0 y = 2
( y 0 ). Thus, the slope of the tangent line at P is
2 2 2 2
a b a b a y
2
b x y y
1 1 1
 . The slope of F P is and of F P is . By the formula from Problems Plus, we
a y
2 1 x +c 2 x c
1 1
1
have
2
y b x
1 1
+
x +c 2 2 2 2 2 2 2 2 2 2 2 2 2
1 a y a y +b x (x +c) a b +b cx using b x +a y =a b
1 1 1 1 1
tan  = 2
= = 1 1
=
2 2 2 2 2 2 2
b x y a y (x +c) b x y c x y +a cy and a  b =c
1 1 1 1 1 1 1 1 1
1 2
a y (x +c)
1 1
2 2
b (cx +a ) 2
b1
2
= and
cy (cx +a ) cy
1 1 1
2
y b x
1 1
 
x c 2 2 2 2 2 2 2 2
b (cx  a )
2
1 a y  a y b x x c ( )=  a b +b cx 2
1 1 1 1 b
1 1
tan  = 2
= = 2
= So  =
2 2 2 2 cy
b x y
1 1
a y (x  c) b x y c x y  a cy cy1(cx1 a ) 1
1 1 1 1 1 1 1
1 2
a y (x  c)
1 1

y y
1 1
60. The slopes of the line segments F P and F P are
1 2 x +c
1
and
x c
1
, where P is x ,y ( 1 1 ).
2
/ 2 b x
2x 2yy / b x 1
Differentiating implicitly,  =0 y = 2
 the slope of the tangent at P is , so by
2 2 2
a b a y a y
1
the formula from Problems Plus,

13
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.5 Conic Sections

2
b x y
1 1

2 x +c 2 2 2 2 2 2 2 2 2
a y 1 b x (x +c) a y b (cx +a ) using x /a  y /b =1 2
1 1 1 1 1 b
tan  = 2
= = 2
1 1
=
2 2 2 2 2 cy
b x y a y (x +c)+b x y cy (cx +a ) and a +b =c 1
1 1 1 1 1 1 1 1
1+ 2
a y (x +c)
1 1
and

2
b x y
1 1
 +
2 x c 2 2 2 2 2
b (cx  a )
a y 1  b x (x  c)+a y 2
1 1 1 1 b 1
tan  = 2
= = 2
=
2 2
b x y
1 1
a y (x  c)+b x y cy1(cx1 a ) cy1
1 1 1 1
1+ 2
a y (x  c)
1 1

So  = .

14
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

1. The directrix y=6 is above the focus at the origin, so we use the form with ‘‘ +esin  ’’ in the
7
 6
ed 4 42
denominator. (See Theorem 6 and Figure 2.) r= = =
1+esin  7 4+7sin 
1+ sin 
4

2. The directrix x=4 is to the right of the focus at the origin, so we use the form with ‘‘ +ecos  ’’ in
ed 1 4 4
the denominator. e=1 for a parabola, so an equation is r= = =
1+ecos  1+1cos  1+cos 
3. The directrix x= 5 is to the left of the focus at the origin, so we use the form with ‘‘  ecos  ’’ in
3
 5
ed 4 15
the denominator. r= = =
1 ecos  3 4 3cos 
1 cos 
4

4. The directrix y= 2 is below the focus at the origin, so we use the form with ‘‘  esin  ’’ in the
ed 2 2 4
denominator. r= = =
1 esin  1 2sin  1 2sin 

5. The vertex ( 4,3 /2 ) is 4 units below the focus at the origin, so the directrix is 8 units below the
focus (d=8) , and we use the form with ‘‘  esin  ’’ in the denominator. e=1 for a parabola, so an
ed 1(8) 8
equation is r= = = .
1 esin  1 1sin  1 sin 
6. The vertex P(1, /2) is 1 unit above the focus F at the origin, so PF =1 and we use the form with
‘‘ +esin  ’’ in the denominator. The distance from the focus to the directrix l is d , so
PF 1
e=  0.8=  0.8d 0.8=1 0.8d=1.8 d=2.25 .
Pl d 1
ed 0.8(2.25) 5 9
An equation is r= =  = .
1+esin  1+0.8sin  5 5+4sin 

7. The directrix r=4sec  (equivalent to rcos  =4 or x=4 ) is to the right of the focus at the origin, so
we will use the form with ‘‘ +ecos  ’’ in the denominator. The distance from the focus to the
ed 0.5(4) 2 4
directrix is d=4 , so an equation is r= =  = .
1+ecos  1+0.5cos  2 2+cos 
8. The directrix r= 6csc  (equivalent to rsin  = 6 or y= 6 ) is below the focus at the origin, so we
will use the form with ‘‘  esin  ’’ in the denominator. The distance from the focus to the directrix is
ed 3(6) 18
d=6 , so an equation is r= = = .
1 esin  1 3sin  1 3sin 

1 ed
9. r= = , where d=e=1 .
1+sin  1+esin 
1
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

(a) Eccentricity =e=1

(b) Since e=1 , the conic is a parabola.
(c) Since " +esin  ’’ appears in the denominator, the directrix is above the focus at the origin.
d= Fl =1 , so an equation of the directrix is y=1 .
1 
(d) The vertex is at , , midway between the focus and the directrix.
2 2

2
 3
6 2 3
10. r= = =
3+2sin  2 2
1+ sin  1+ sin 
3 3
2
(a) e=
3
(b) Ellipse
(c) y=3
6  3 12 3
(d) Vertices , and 6, ; center ,
5 2 2 5 2

12 1/4 3 1
11. r=  = , where e= and ed=3 d=12 .
4 sin  1/4 1 4
1 sin 
4
1
(a) Eccentricity =e=
4
(b) Since

2
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

1
e= <1 , the conic is an ellipse.
4
(c) Since ‘‘  esin  ’’ appears in the denominator, the directrix is below the focus at the origin.
d= Fl =12 , so an equation of the directrix is y= 12 .
 12 3
(d) The vertices are 4, and , , so the center is midway between them, that is,
2 5 2
4 
, .
5 2

3 4

4 2 2 3
12. r= = =
2 3cos  3 3
1 cos  1 cos 
2 2
3
(a) e=
2
(b) Hyperbola
4
(c) x=
3
4 4 12
(d) The vertices are (  4,0 ) and , =  ,0 , so the center is  ,0 . The asymptotes
5 5 5
 1 2  1 2 5
are parallel to  = cos . [Their slopes are  tan cos = ]
3 3 2

13.
3
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

9 1/6 3/2 1 3 9
r=  = , where e= and ed=  d= .
6+2cos  1/6 1 3 2 2
1+ cos 
3
1
(a) Eccentricity =e=
3
1
(b) Since e= <1 , the conic is an ellipse.
3
(c) Since " +ecos  ’’ appears in the denominator, the directrix is to the right of the focus at the
9 9
origin. d= Fl = , so an equation of the directrix is x= .
2 2
9 9
(d) The vertices are ,0 and , , so the center is midway between them, that is,
8 4
9
, .
16

5
5 2
14. r= =
2 2sin  1 sin 
(a) e=1
(b) Parabola
5
(c) y=
2
5 3
(d) The focus is ( 0,0 ) , so the vertex is , and the parabola opens up.
4 2

3 1/4 3/4 3 3
15. r=  = , where e=2 and ed=  d= .
4 8cos  1/4 1 2cos  4 8
4
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

(a) Eccentricity =e=2

(b) Since e=2>1 , the conic is a hyperbola.
(c) Since "  ecos  ’’ appears in the denominator, the directrix is to the left of the focus at the origin.
3 3
d= Fl = , so an equation of the directrix is x= .
8 8
3 1
(d) The vertices are  ,0 and , , so the center is midway between them, that is,
4 4
1
, .
2

1
 4
4 2 2
16. r= = =
2+cos  1 1
1+ cos  1+ cos 
2 2
1
(a) e=
2
(b) Ellipse
(c) x=4
4 4
(d) The vertices are ,0 and ( 4, ) = (  4,0 ) , so the center is  ,0 .
3 3

1 1/4 3 1 1
17. (a) The equation is r= = , so e= and ed=  d= . The conic is an
4 3cos  3 4 4 3
1 cos 
4
1 1
ellipse, and the equation of its directrix is x=rcos  =  r= . We must be careful in our
3 3cos 
5
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

choice of parameter values in this equation (  1   1 works well).


(b) The equation is obtained by replacing  with   in the equation of the original conic (see
3
1
Example 4), so r= .

4 3cos  
3

5 5/2 5 5
18. r= = , so e=1 and d= . The equation of the directrix is y=rsin  = 
2+2sin  1+sin  2 2
5 
r= . If the parabola is rotated about its focus (the origin) through , its equation is the same
2sin  6
 5
as that of the original, with  replaced by   (see Example 4), so r= . In
6 2+2sin (   /6 )
graphing each of these curves, we must be careful to select parameter ranges which prevent the
denominator from vanishing while still showing enough of the curve.

19. For e<1 the curve is an ellipse. It is nearly circular when e is close to 0 . As e increases, the graph
is stretched out to the right, and grows larger (that is, its right hand focus moves to the right while its
left hand focus remains at the origin.) At e=1 , the curve becomes a parabola with focus at the origin.

6
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

20. (a) The value of d does not seem to affect the shape of the conic (a parabola) at all, just its size,
position, and orientation (for d<0 it opens upward, for d>0 it opens downward).

+
(b) We consider only positive values of e . When 0<e<1 , the conic is an ellipse. As e 0 , the graph
approaches perfect roundness and zero size. As e increases, the ellipse becomes more elongated, until
at e=1 it turns into a parabola. For e>1 , the conic is a hyperbola, which moves downward and gets
broader as e continues to increase.

e=0.1

e=0.5

7
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

e=0.9

e=1

e=1.1

e=1.5

8
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

e=10

ed
21. PF =e Pl  r=e[d rcos (   )]=e(d+rcos  ) r(1 ecos  )=ed r=
1 ecos 

ed
22. PF =e Pl  r=e[d rsin  ] r(1+esin  )=ed r=
1+esin 

ed
23. PF =e Pl  r=e[d rsin (   )]=e(d+rsin  ) r(1 esin  )=ed r=
1 esin 

c d c d c+d
24. The parabolas intersect at the two points where =  cos  =  r= .
1+cos  1 cos  c+d 2
dr csin 
For the first parabola, = , so
d 2
( 1+cos  )
2
dy ( dr/d ) sin  +rcos  csin  +ccos  (1+cos  ) 1+cos 
= = =
dx ( dr/d ) cos   rsin  csin  cos   csin  (1+cos  )  sin 

and similarly for the second,

9
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

dy 1 cos  sin 
= = . Since the product of these slopes is  1 , the parabolas intersect at right
dx sin  1+cos 
angles.

2 2
ed 2 ed 2
25. (a) If the directrix is x= d , then r= , and, from (4), a =  ed=a(1 e ) .
1 ecos 
( 1 e2) 2
2
a(1 e )
Therefore, r= .
1 ecos 
8 8
(b) e=0.017 and the major axis =2a=2.99 10  a=1.495 10 . Therefore
8 2 8
1.495 10 1 ( 0.017) 1.49 10
r=
.
1 0.017cos  1 0.017cos 

26. (a) The Sun is at point F in Figure 1 so that perihelion is in the positive x  direction and aphelion

is in the negative x  direction. At perihelion,  =0 , so r=

a 1 e
=
a(1 e)(1+e)
=a(1 e) .
( 2
)
1+ecos 0 1+e

At aphelion,  = , so r=
a 1 e ( 2
) = a(1 e)(1+e) =a(1+e) .
1+ecos  1 e
8 8
(b) At perihelion, r=a(1 e)
(1.495 10 )(1 0.017)
1.47 10 km.
At aphelion, r=a(1+e)
( 1.495 10
8
) (1+0.017)
1.52 10 km.
8

27. Here 2a= length of major axis =36.18 AU  a=18.09 AU and e=0.97 . By Exercise 25(a), the
2
18.09 1 ( 0.97) 1.07
equation of the orbit is r=
. By Exercise 26(a), the maximum
1 0.97cos  1 0.97cos 
distance from the comet to the sun is 18.09(1+0.97)
35.64 AU or about 3.314 billion miles.

28. Here 2a= length of major axis =356.5 AU  a=178.25 AU and e=0.9951 . By Exercise 25(a), the
2
178.25 1 ( 0.9951 ) 1.7426
equation of the orbit is r=
. By Exercise 26(a), the minimum
1 0.9951cos  1 0.9951cos 
distance from the comet to the sun is 178.25 ( 1 0.9951 )
0.8734 AU or about 81 million miles.

29. The minimum distance is at perihelion, where

7 7
4.6 10 =r=a(1 e)=a(1 0.206)=a(0.794)  a=4.6 10 /0.794 . So the maximum distance, which is at
( 7
aphelion, is r=a(1+e)= 4.6 10 /0.794 (1.206)
7.0 10 km. ) 7

9 9
30. At perihelion, r=a(1 e)=4.43 10 , and at aphelion, r=a(1+e)=7.37 10 . Adding, we get

10
Stewart Calculus ET 5e 0534393217;10. Parametric Equations and Polar Coordinates; 10.6 Conic Sections in Polar Coordinates

9 9 7.37
2a=11.80 10 , so a=5.90 10 km. Therefore 1+e=a(1+e)/a=
1.249 and e
0.249 .
5.90

7 7
31. From Exercise 29, we have e=0.206 and a ( 1 e ) =4.6 10 km. Thus, a=4.6 10 /0.794 . From
2
1 e
Exercise 25, we can write the equation of Mercury’s orbit as r=a . So since
1 ecos 
2
dr  a(1 e )esin 
= 
d 2
( 1 ecos  )

2
r +
dr 2
=
a
2
( 1 e2) 2 +
2
a ( 1 e2) 2e2sin 2 =
a
2
( 1 e2) 2 ( 1 2ecos  2
+e )
d 2 4 4
( 1 ecos  ) ( 1 ecos  ) ( 1 ecos  )
the length of the orbit is

2
1+e  2ecos 
L=
2
0
2 2
r + ( dr/d ) d =a 1 e ( 2
) 2
0 2
d
3.6 10 km
8

( 1 ecos  )
This seems reasonable, since Mercury’s orbit is nearly circular, and the circumference of a circle of
8
radius a is 2 a
3.6 10 km.

11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

1. (a) A sequence is an ordered list of numbers. It can also be defined as a function whose domain is
the set of positive integers.
(b) The terms a approach 8 as n becomes large. In fact, we can make a as close to 8 as we like by
n n
taking n sufficiently large.
(c) The terms a become large as n becomes large. In fact, we can make a as large as we like by
n n
taking n sufficiently large.

2. (a) From Definition 1, a convergent sequence is a sequence for which lim a exists. Examples:
n
n 

{ 1/n} , { 1/2n}
(b) A divergent sequence is a sequence for which lim a does not exist. Examples: { n} , { sin n}
n
n 

n
3. a =1 ( 0.2 ) , so the sequence is { 0.8,0.96,0.992,0.9984,0.99968,... } .
n

4. a =
n
n+1
3n 1
, so the sequence is { 2 3 4 5 6
, , , , ,...
2 5 8 11 14 }{
= 1,
3 1 5 3
, , , ,...
5 2 11 7 } .

n
3(  1)  3 3  3 3  3
5. a =
n n!
, so the sequence is { , , , ,
1 2 6 24 120
,... }{ =  3,
3 1 1
, , ,
2 2 8 40
1
,... } .

6. a =2  4  6      ( 2n ) , so the sequence is
n
{ 2,2  4,2  4  6,2  4  6  8,2  4  6  8  10,... } ={ 2,8,48,384,3840,... } .

7. a =3 , a =2a  1 . Each term is defined in terms of the preceding term.

1 n+1 n
a =2a  1=2(3) 1=5 . a =2a  1=2(5) 1=9 . a =2a  1=2(9) 1=17 . a =2a  1=2(17) 1=33 . The
2 1 3 2 4 3 5 4
sequence is { 3,5,9,17,33,... } .

a
n
8. a =4 , a = . Each term is defined in terms of the preceding term.
1 n+1 a 1
n
a a
1 4 4 2 4/3 4/3
a= = = .a = = = =4 . Since a =a , we can see that the terms of the
2 a  1 4 1 3 3 a  1 4 1/3 3 1
1 2  1
3
4 4
sequence will alternately equal 4 and 4/3 , so the sequence is 4, ,4, ,4,...
3 3
. { }
1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

1
9. The numerators are all 1 and the denominators are powers of 2 , so a = .
n n
2

1
10. The numerators are all 1 and the denominators are multiples of 2 , so a = .
n 2n

11. { 2,7,12,17,... } . Each term is larger than the preceding one by 5 , so

a =a +d(n 1)=2+5(n 1)=5n 3 .
n 1

12. { 
1 2
, ,
3 4
, ,...
4 9 16 25 } . The numerator of the n th term is n and its denominator is ( n+1 ) .
n n
2

Including the alternating signs, we get a = (  1 ) .

n 2
( n+1 )
n 1
13. { 1,
2 4 8
, ,
3 9 27
,... } . Each term is 
2
3
times the preceding one, so a = 
n
2
3
.

14. { 5,1,5,1,5,1,... } . The average of 5 and 1 is 3 , so we can think of the sequence as alternately
n+1
adding 2 and  2 to 3 . Thus, a =3+( 1)  2.
n

15. a =n(n 1) . a   as n  , so the sequence diverges.

n n

n+1 1+1/n 1+0 1

16. a = = , so a  = as n  . Converges
n 3n 1 3 1/n n 3 0 3

17. a =
3+5n
2
=
( 3+5n ) /n
2 2
=
5+3/n
2
, so a 
5+0
=5 as n  . Converges
n
n+n
2
( n+n ) /n 1+1/n
2 2 n 1+0

n 1 1
18. a = = , so a  =1 as n  . Converges
n 1+ n 1/ n +1 n 0+1

n n n
2 1 2 1 2 1 2
19. a = = , so lim a = lim =  0=0 by (8) with r= . Converges
n n+1 3 3 n 
n 3 n  3 3 3
3

n n
20. a = = . The numerator approaches  and the denominator approaches 0+1=1 as
n 1+ n 1/ n +1
2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

n  , so a   as n  and the sequence diverges.

n

n 1 n 1
(  1) n (  1) 1 1
21. a = = , so 0 a =   0 as n  , so a  0 by the Squeeze
n 2 n+1/n n n+1/n n n
n +1
Theorem and Theorem 6. Converges

n 3 3
( 1) n 1 n
22. a = 3 2 3 2
 1 as n  , but the terms of the sequence
. Now a = =
n n 2 1
n +2n +1 n +2n +1 1+ +
n 3
n
{a } alternate in sign, so the sequence a ,a ,a ... converges to  1 and the sequence a ,a ,a ,...
n 1 3 5, 2 4 6
converges to +1 . This shows that the given sequence diverges since its terms don’t approach a single
real number.

23. a =cos ( n/2) . This sequence diverges since the terms don’t approach any particular real number
n
as n  . The terms take on values between  1 and 1 .

24. a =cos (2/n) . As n  , 2/n 0 , so cos (2/n) cos 0=1 . Converges

n

(2n 1)! (2n 1)! 1

25. a = = =  0 as n  . Converges
n (2n+1)! (2n+1)(2n)(2n 1)! (2n+1)(2n)

 
26. 2n  as n  , so since lim arctanx= , we have lim arctan2n= . Converges
x  2 n  2

n  n  n  2n
e +e e 1+e 1+0
27. a =  =   0 as n  . Converges
n 2n  n n  n n
e  1 e e e e 0

ln n ln n 1 1
28. a = = =   1 as n  . Converges
n ln 2n ln 2+ln n ln 2 0+1
+1
ln n

2 2
2  n n x 2x 2
29. a =n e = . Since lim =lim =lim =0 , it follows from Theorem 3 that lim a =0 .
n n x x x n
e x  e x  e x  e n 
Converges

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

n
30. a =ncos n =n( 1) . Since a =n  as n  , the given sequence diverges.
n n

{ }
2 2
cos n 1 1 cos n
31. 0  , so since lim =0 , converges to 0 by the Squeeze Theorem.
n n n n
2 2 n  2 2

n+1 1
32. a =ln (n+1) ln n=ln =ln 1+  ln (1)=0 as n  . Converges
n n n

sin (1/n) sin (1/x) sin t

33. a =nsin (1/n)= . Since lim =lim [] where t=1/x]=1 , it follows from
n 1/n x  1/x + t
t 0
Theorem 3 that {a } n
converges to 1.

2 2 1 2 1 1 1
34. a = n  n  1= n   n 1 =n  1  n(0 1)  n as n  ,
n n 2 n 2
n n
so a    as n  . Diverges
n

2 1/n 1 2 1 2
35. a = 1+  ln a = ln 1+ . As n  ,  0 and ln 1+  0 , so ln a  0.
n n n n n n n n
0
Thus, a  e =1 as n  . Converges
n
sin 2n 1 1  1 1
36. a = . a  and lim =0 , so  a  lim a =0 by the
n 1+ n n 1+ n n n
n  1+ n 1+ n 1+ n n 
Squeeze Theorem. Converges

37. { 0,1,0,0,1,0,0,0,1,... } diverges since the sequence takes on only two values, 0 and 1, and never
stays arbitrarily close to either one (or any other value) for n sufficiently large.

38. { 1 1 1 1 1 1 1 1
, , , , , , , ,...
1 3 2 4 3 5 4 6 2n 1
1
. a = and a =
n 2n
1
n+2 }
for all positive integers n . lim a =0
n 
n
1 1
since lim a =lim =0 and lim a =lim =0 . For n sufficiently large, a can be made as
n  n n  n+2
2n 1 2n n
n  n 
close to 0 as we like. Converges
n! 1 2 3 ( n 1 ) n 1 n n
39. a = =          =   as n  , so a diverges. { }
n n 2 2 2 2 2 2 2 4 n
2

40.
4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

n
3 3 3 3 3 3 3 3 3 27
0< a = =            =  0 as n  , so by the
n n! 1 2 3 (n 1) n 1 2 n 2n
Squeeze Theorem and Theorem 6, { (  3) n/n} converges to 0 .

41.

From the graph, we see that the sequence

and  1 (approximately).
{(  1)
n n+1
n } is divergent, since it oscillates between 1

42.

From the graph, it appears that the sequence converges to 2 .

{ 

2 n
}
converges to 0 by (6), and hence 2+ 

2
{ n
} converges to 2+0=2 .

43.

From the graph, it appears that the sequence converges to about 0.78 .
2n 2 2n 
lim =lim =1 , so lim arctan =arctan1= .
n  2n+1 n  2+1/n n  2n+1 4
5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

44.

From the graph, it appears that the sequence converges (slowly) to 0 .

0
to 0 .
sin n
n

1
n
 0 as n  , so by the Squeeze Theorem and Theorem 6,
{ }
sin n
n
converges

45.

From the graph, it appears that the sequence converges to 0 .

3
0<a = n n n n 1 1 1 1
n =          
n! n (n 1) ( n 2 ) ( n 3) 3 2 1
2
 n
[for n 4]
(n 1)(n 2)(n 3)
1/n
=  0 as n 
( 1 1/n ) ( 1 2/n ) ( 1 3/n )
So by the Squeeze Theorem, { n3/n!} converges to 0 .

46.

From the graph, it appears that the sequence converges to 5.

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

n n n n
5 
n n n n n n n
5= 3 +5  5 +5 = 2 5
n 1/n 0
= 2 5 5asn  lim 2 =2 =1]
n 
Hence, a  5 by the Squeeze Theorem.
n

Alternate Solution: Let y= 3 +5 ( x x) 1/x. Then

ln ( 3 +5 )
x x x x
lim ln y = lim 3 ln 3+5 ln 5

=lim
x
x  x x  x x
3 +5
3 x
ln 3+ln 5
5
= lim =ln 5
3 x
x 
+1
5

so lim y=e
x 
ln 5
=5 , and so { n
3 +5
n n
} converges to 5 .

47.

From the graph, it appears that the sequence approaches 0 .

1 3 5     ( 2n 1 ) 1 3 5 2n 1
0<a = =       
n n 2n 2n 2n 2n
( 2n )
1 1
  (1) (1)     (1)=  0 as n 
2n 2n
1 3 5     ( 2n 1 )
So by the Squeeze Theorem,
( 2n )
n { converges to 0 .
}
48.

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

1 3 5     ( 2n 1 )
From the graphs, it seems that the sequence diverges. a = . We first
n n!
3 n 1
prove by induction that a for all n . This is clearly true for n=1 , so let P(n) be the
n 2
statement that the above is true for n . We must show it is then true for n+1 .
2n+1 3 n 1 2n+1 2n+1 3
a =a   (induction hypothesis). But , and so we get that
n+1 n n+1 2 n+1 n+1 2
3 n 1 3 3 n
a  = which is P(n+1) . Thus, we have proved our first assertion, so since
n+1 2 2 2

{ 3 n 1
2 }
diverges (by (8)), so does the given sequence a
n
n
. { }
49. (a) a =1000(1.06)  a =1060 , a =1123.60 , a =1191.02 , a =1262.48 , and a =1338.23 .
n 1 2 3 4 5
n
(b) lim a =1000lim (1.06) , so the sequence diverges by (8) with r=1.06>1 .
n
n  n 

50. a
n+1
=
{ 1
a
2 n
if a is an even number
n

3a +1 if a is an odd number
n n
When a =11 , the first 40 terms are 11 , 34 , 17
1

, 52 , 26 , 13 , 40 , 20 , 10 , 5 , 16 , 8 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2
, 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 . When a =25 , the first 40 terms are 25 , 76 , 38 , 19 , 58 , 29 , 88 , 44 , 22
1
, 11 , 34 , 17 , 52 , 26 , 13 , 40 , 20 , 10 , 5 , 16 , 8 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4 , 2 , 1 , 4
, 2 , 1 , 4 . The famous Collatz conjecture is that this sequence always reaches 1 , regardless of the
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

starting point a .
1

51. If r 1 , then { r n} diverges by (8), so { nr n} n

diverges also, since nr =n r
n
r
n
. If
x
x 1 r
r <1 then lim xr =lim
x
 x
=lim
 x
=lim
 ln r
n
=0 , so lim nr =0 , and hence { nr n}
r x  x  x  (  ln r ) r x  n 
converges whenever r <1 .

52. (a) Let lim a =L . By Definition 1, this means that for every  >0 there is an integer N such that
n
n 
a  L < whenever n>N . Thus, a  L < whenever n+1>N
n>N 1 . It follows that
n n+1
lim a =L and so lim a =lim a .
n+1 n n+1
n  n  n 
2  1+ 5
(b) If L=lim a then lim a =L also, so L must satisfy L=1/ ( 1+L )  L +L 1=0 L=
n 
n
n 
n+1 2
(since L has to be non negative if it exists).

53. Since { a } is a decreasing sequence, a >a for all n 1 . Because all of its terms lie between
n n n+1
5 and 8 , { a } is a bounded sequence. By the Monotonic Sequence Theorem, { a } is
n n
convergent; that is, { a } has a limit L . L must be less than 8 since { a } is decreasing, so
n n
5 L<8 .

n 1
54. a =1/5 defines a decreasing geometric sequence since a = a <a for each n 1 . The
n n+1 5 n n
1
sequence is bounded since 0<a  for all n 1 .
n 5

1 1 1 1
55. a = is decreasing since a = = < =a for each n 1 . The sequence is
n 2n+3 n+1 2(n+1)+3 2n+5 2n+3 n
1 1
bounded since 0<a  for all n 1 . Note that a = .
n 5 1 5

2n 3 2x 3
56. a = defines an increasing sequence since for f (x)= ,
n 3n+4 3x+4
/ (3x+4)(2) (2x 3)(3) 17 1
f (x)= = >0 . The sequence is bounded since a a = for n 1 ,
2 2 n 1 7
( 3x+4 ) ( 3x+4 )
2n 3 2n 2
and a < < = for n 1 .
n 3n 3n 3
9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

57. a =cos ( n /2 ) is not monotonic. The first few terms are 0 ,  1 , 0 , 1 , 0 ,  1 , 0 , 1 , ... . In fact,
n
the sequence consists of the terms 0 ,  1 , 0 , 1 repeated over and over again in that order. The
sequence is bounded since a  1 for all n 1 .
n

 n  x
58. a =ne defines a positive decreasing sequence since the function f (x)=xe is decreasing for x>1
n
/  x  x  x 1
. [ f (x)=e  xe =e (1 x)<0] for x>1 .] The sequence is bounded above by a = and below by 0.
1 e
n x
59. a = 2
defines a decreasing sequence since for f (x)= 2
,
n
n +1 x +1

f
/
(x)=
( x +1 ) ( 1 )  x(2x)
2
=
1 x
2
 0 for x 1 . The sequence is bounded since 0<a 
1
for all
2 2 n 2
( x +1)
2
( x +1)
2

n 1 .

1 1
60. a =n+ defines an increasing sequence since the function g(x)=x+ is increasing for x>1 .
n n x
/ 2
[g (x)=1 1/x >0] for x>1 .] The sequence is unbounded since a   as n  . (It is, however,
n
bounded below by a =2 .)
1

61. a =2
1/2
, a =2
3/4
, a =2
7/8
, ... , so a =2
( 2  1 ) /2
n n
=2
1 ( 1/2 )
n
. lim a =lim 2
1 ( 1/2n) =21=2 .
1 2 3 n n
n  n 
Alternate solution : Let L=lim a . (We could show the limit exists by showing that
n 
n {a } n
is
2
bounded and increasing.) Then L must satisfy L= 2  L  L =2L L(L 2)=0 . L 0 since the
sequence increases, so L=2 .

62. (a) Let P be the statement that a a and a  3 . P is obviously true. We will assume that P
n n+1 n n 1 n
is true and then show that as a consequence P must also be true. a a
2+a 2+a
n+1 n+2 n+1 n+1 n

2+a 2+a
a a , which is the induction hypothesis. a  3
2+a  3
2+a  9

n+1 n n+1 n n+1 n n

a  7 , which is certainly true because we are assuming that a  3 . So P is true for all n , and so
n n n
a  a  3 (showing that the sequence is bounded), and hence by the Monotonic Sequence Theorem,
1 n
lim a exists.
n
n 
(b) If
10
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

2 2
L=lim a , then lim a =L also, so L= 2+L  L =2+L
L  L 2=0
(L+1)(L 2)=0
L=2 (since
n n+1
n  n 
L can’t be negative).

63. We show by induction that {a } n

is increasing and bounded above by 3 .
Let P be the proposition that a >a and 0<a <3 . Clearly P is true. Assume that P is true. Then
n n+1 n n 1 n
1 1 1 1
a >a  <   > .
n+1 n a a a a
n+1 n n+1 n
1 1
Now a =3 >3 =a
P . This proves that {a } is increasing and bounded above by
n+2 a a n+1 n+1 n
n+1 n
3 , so 1=a <a <3 , that is,
1 n {a } n
is bounded, and hence convergent by the Monotonic Sequence
Theorem.
2 3 5
If L=lim a , then lim a =L also, so L must satisfy L=3 1/L L  3L+1=0 L= . But L>1
n 
n
n 
n+1 2
3+ 5
, so L= .
2

64. We use induction. Let P be the statement that 0<a  a  2 . Clearly P is true, since
n n+1 n 1
a =1/(3 2)=1 . Now assume that P is true. Then a  a  a  a 3 a 3 a 
2 n n+1 n n+1 n n+1 n
1 1
a =  =a . Also a >0 (since 3 a is positive) and a  2 by the induction
n+2 3 a 3 a n+1 n+2 n+1 n+1
n+1 n
hypothesis, so P is true.
n+1
1 2 3 5
To find the limit, we use the fact that lim a =lim a  L=  L  3L+1=0 L= . But
n 
n
n 
n+1 3 L 2
3 5
L 2 , so we must have L= .
2

65. (a) Let a be the number of rabbit pairs in the n th month. Clearly a =1=a . In the n th month,
n 1 2
each pair that is 2 or more months old (that is, a pairs) will produce a new pair to add to the a
n 2 n 1
pairs already present. Thus, a =a
n n 1
+a
n 2
, so that {a } ={ f }
n n
, the Fibonacci sequence.
f f f +f f
n+1 n n 1 n 2 n 2 1 1
(b) a =  a = = =1+ =1+ =1+ . If L=lim a , then
n f
n
n 1 f
n 1
f
n 1
f
n 1
f
n 1 /f
n 2
a
n 2
n 
n

11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

1 2 1+ 5
L=lim a and L=lim a , so L must satisfy L=1+  L  L 1=0 L= (since L must be
n 
n 1
n 
n 2 L 2
positive).

66. (a) If f is continuous, then f (L)= f lim a

n 
n
=lim f a =lim a
n 
( )
n
n 
=L by Exercise 52(a).
n+1
(b) By repeatedly pressing the cosine key on the calculator (that is, taking cosine of the previous
answer) until the displayed value stabilizes, we see that L 0.73909 .

67. (a)

From the graph, it appears that the sequence { } n

n!
5
converges to 0 , that is, lim
n
n  n!
5
=0 .

(b)

From the first graph, it seems that the smallest possible value of N corresponding to  =0.1 is 9 , since
5 5
n /n!<0.1 whenever n 10 , but 9 /9!>0.1 . From the second graph, it seems that for  =0.001 , the
smallest possible value for N is 11 .
12
 Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

68. Let  >0 and let N be any positive integer larger than ln ( ) /ln r . If n>N then n>ln ( ) /ln r
 nln r <ln   ln ( r n) <ln  
n
r < 
n
r  0 < , and so by Definition 1, lim r =0 .
n

n 

69. If lim a =0 then lim  a =0 , and since  a  a a , we have that lim a =0 by the
n n n n n n
n  n  n 
Squeeze Theorem.

70. (a)
n+1 n+1
b  a n n 1 n 2 2 n 3 3 n 1 n
=b +b a+b a +b a+   +ba +a
b a
n n 1 n 2 2 n 3 3 n 1 n n
<b +b b+b b +b b+   +bb +b =(n+1)b

n+1 n+1 n n+1 n n+1 n n+1

(b) Since b a>0 , we have b  a <(n+1)b (b a) b  (n+1)b (b a)<a  b (n+1)a nb <a
.
n 1 n n+1 1 n+1
(c) With this substitution, (n+1)a nb=1 , and so b = 1+ <a = 1+ .
n n+1
1 n 1 1 n 1 2n
(d) With this substitution, we get 1+
<1 1+ <2 1+ <4 .
2n 2 2n 2n
(e) a <a
n 2n
since {a } n
is increasing, so a <a <4 .
n 2n
(f) Since {a } is increasing and bounded above by 4 , a  a  4 , and so {a } is bounded and
n 1 n n
monotonic, and hence has a limit by Theorem 11.

71. (a) First we show that a>a >b >b .

1 1
a+b 1 1 2
a b=  ab = ( a 2 ab +b ) = ( a  b ) >0 (since a>b )  a >b . Also
1 1 2 2 2 1 1
1 1
a a =a (a+b)= (a b)>0 and b b =b ab = b ( b  a ) <0 , so a>a >b >b . In the same way we
1 2 2 1 1 1
can show that a >a >b >b and so the given assertion is true for n=1 . Suppose it is true for n=k , that
1 2 2 1
is, a >a >b >b . Then
k k+1 k+1 k
1 1
 b  a b = a  2 a b +b
a
k+2 k+2 = a +b
2 k+1 k+1 ( k+1 k+1 2 k+1 )
k+1 k+1 k+1

1 2
= a  b >0
2 k+1 k+1

 a 1 1
a
k+1 k+2 = a  (a +b )= a  b >0 ( )
k+1 2 k+1 k+1 2 k+1 k+1
13
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

and b  b =b  a b = b b  a <0 a >a >b >b , so the

k+1 k+2 k+1 k+1 k+1 k+1 k+1 k+1 k+1 k+2 k+2 k+1
assertion is true for n=k+1 . Thus, it is true for all n by mathematical induction.
(b) From part (a) we have a>a >a >b >b >b , which shows that both sequences,
n n+1 n+1 n {a } n
and

{b } n
, are monotonic and bounded. So they are both convergent by the Monotonic Sequence
Theorem.
a +b
n n  +
(c) Let lim a = and lim b = . Then lim a =lim   = 
n 
n
n 
n
n 
n+1
n  2 2
2 = +   = .

72. (a) Let  >0 . Since lim a =L , there exists N such that a  L < for n>N . Since
2n 1 2n 1
n 
lim a =L , there exists N such that a  L < for n>N . Let N=max 2N ,2N +1 { } and let
2n+1 2 2n+1 2 1 2
n 
n>N . If n is even, then n=2m where m>N , so a  L = a  L < . If n is odd, then n=2m+1 ,
1 n 2m
where m>N , so a  L = a  L < . Therefore lim a =L .
2 n 2m+1 n
n 
1 3 1 7 1 17
(b) a =1 , a =1+ = =1.5 , a =1+ = =1.4 , a =1+ = =1.416 ,
1 2 1+1 2 3 5/2 5 4 12/5 12
1 41 1 99 1 239
a =1+ = 1.413793 , a =1+ = 1.414286 , a =1+ = 1.414201 ,
5 29/12 29 6 70/29 70 7 169/70 169
1 577
a =1+ = 1.414216 . Notice that a <a <a <a and a >a >a >a . It appears that the odd
8 408/169 408 1 3 5 7 2 4 6 8
terms are increasing and the even terms are decreasing. Let’s prove that a >a and a <a by
2n 2 2n 2n 1 2n+1
1 1
mathematical induction. Suppose that a >a . Then 1+a >1+a  < 
2k 2 2k 2k 2 2k 1+a 1+a
2k 2 2k
1 1 1 1
1+ <1+  a <a  1+a <1+a  > 
1+a 1+a 2k 1 2k+1 2k 1 2k+1 1+a 1+a
2k 2 2k 2k 1 2k+1
1 1
1+ >1+  a >a . We have thus shown, by induction, that the odd terms are
1+a 1+a 2k 2k+2
2k 1 2k+1
increasing and the even terms are decreasing. Also all terms lie between 1 and 2 , so both {a }n
and

{b } n
are bounded monotonic sequences and are therefore convergent by Theorem 11. Let

1 1 4+3a
n
lim a =L . Then lim a =L also. We have a =1+ =1+ = ,
2n 2n+2 n+2 1+1+1/(1+a ) (3+2a )/(1+a ) 3+2a
n  n  n n n n
so

14
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.1 Sequences

4+3a 4+3L 2 2
2n
a = . Taking limits of both sides, we get L=  3L+2L =4+3L L =2 L= 2
2n+2 3+2a 3+2L
2n
(since L>0 ). Thus, lim a = 2 . Similarly we find that lim a = 2 . So, by part (a), lim a = 2 .
2n 2n+1 n
n  n  n 

73. (a) Suppose {p } n

converges to p . Then

bp blim p
n bp
n n  2
p =  lim p =  p=  p +ap=bp  p( p+a b)=0  p=0 or p=b a .
n+1 a+ p n 
n+1 a+lim p a+ p
n n
n 
b
p
bp p
n a n b n
(b) p = = < p since 1+ >1 .
n+1 a+ p p a n a
n n
1+
a
b b b 2 b b 3
(c) By part (b), p < p ,p< p< p ,p< p< p , etc. In general,
1 a 0 2 a 1 a 0 3 a 2 a 0
b n b n
p< p , so lim p  lim  p =0 since b<a .
n a 0
n 
n
n  a 0

(d) Let a<b . We first show, by induction, that if p <b a , then p <b a and p >p .
0 n n+1 n
bp p (b a p )
0 0 0
For n=0 , we have p  p =  p= >0 since p <b a . So p > p .
1 0 a+ p 0 a+ p 0 1 0
0 0
Now we suppose the assertion is true for n=k , that is, p <b a and p > p . Then
k k+1 k
bp a(b a)+bp  ap  bp a(b a p )
k k k k k
b a p =b a = = >0 because p <b a . So p <b a . And
k+1 a+ p a+ p a+ p k k+1
k k k
bp p (b a p )
k+1 k+1 k+1
p  p =  p = >0 since p <b a . Therefore, p >p . Thus, the
k+2 k+1 a+ p k+1 a+ p k+1 k+2 k+1
k+1 k+1
assertion is true for n=k+1 . It is therefore true for all n by mathematical induction. A similar proof by
induction shows that if p >b a , then p >b a and p
0
is decreasing. In either case the sequence
n { } n

{p } n
is bounded and monotonic, so it is convergent by the Monotonic Sequence Theorem. It then
follows from part (a) that lim p =b a .
n
n 

15
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

( n) ( 8)
 f (a) f (5)
1. Using Theorem 5 with 
n
b (x 5) , b = , so b = .
n=0 n n n! 8 8!

/
2. (a) Using Formula 6, a power series expansion of f at 1 must have the form f (1)+ f (1)(x 1)+  
/
. Comparing to the given series, 1.6 0.8(x 1)+   , we must have f (1)= 0.8 . But from the graph,
/
f (1) is positive. Hence, the given series is not the Taylor series of f centered at 1 .
/ 1 / / 2
(b) A power series expansion of f at 2 must have the form f (2)+ f (2)(x 2)+ f (2)(x 2) +   .
2
2 3 1 / /
Comparing to the given series, 2.8+0.5(x 2)+1.5(x 2)  0.1(x 2) +   , we must have f (2)=1.5
2
/ / / /
; that is, f (2) is positive. But from the graph, f is concave downward near x=2 , so f (2) must
be negative. Hence, the given series is not the Taylor series of f centered at 2 .

3.
n ( n) ( n)
f (x) f ( 0)
0 cos x 1
1  sin x 0
2  cos x  1
3 sin x 0
4 cos x 1
  
  
  
We use Equation 7 with f (x)=cos x .
/ / ( 3) ( 4)
cos x = f (0)+ f /(0)x+ f (0) 2 f (0) 3 f ( 0 ) 4
x+ x+ x +  
2! 3! 4!
2 4 n 2n
x x  (  1) x
=1 +     =
2! 4! n=0 (2n)!

n 2n
(  1) x
If a = , then
n ( 2n )!

a 2n+2
n+1 x (2n)! 2 1
lim =lim  =x lim =0<1 for all x .
n  a n  (2n+2)!
x
2n
n  ( 2n+2 ) ( 2n+1 )
n

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

So R= (Ratio Test).

4.
n ( n) ( n)
f (x) f ( 0)
0 sin 2x 0
1 2cos 2x 2
2
2  2 sin 2x 0
3 3
3  2 cos 2x  2
4
4 2 sin 2x 0
  
  
  
( n) ( 2n+1 ) n 2n+1
f (0)=0 if n is even and f (0)=( 1) 2 , so
( n) ( 2n+1 )
sin 2x =   f (0) n
x =
 f (0) 2n+1
x
n=0 n! n=0 (2n+1)!
n 2n+1 2n+1
 ( 1) 2 x
=  n=0 (2n+1)!
a 2 2
n+1 2 x
lim =lim =0<1 for all x ,
n  a n  (2n+3)(2n+2)
n
so R= (Ratio Test).

5.
n ( n) ( n)
f (x) f ( 0)
 3
0 ( 1+x ) 1
 4
1  3(1+x)  3
 5
2 12(1+x) 12
 6
3  60(1+x)  60
 7
4 360(1+x) 360
  
  
  

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

/ / / / / ( 4)
 3 (0) 2 f
/ f
(0) 3 f (0) 4
( 1+x ) = f (0)+ f (0)x+ x+ x+ x +  
2! 3! 4!
4 3 2 5 4 3 3 6 5 4 3 4
=1 3x+ x x+ x   
2! 3! 4!
4 3 2 2 5 4 3 2 3 6 5 4 3 2 4
=1 3x+ x x+ x   
2 2! 2 3! 2 4!
n n n n

( 1) (n+2)! x  ( 1) (n+2)(n+1)x
= =
n=0 2(n!) n=0 2

a n+1
n+1 (n+3)(n+2)x 2 n+3
lim =lim  = x lim = x <1 for convergence,
n  a n  2 n n  n+1
n (n+2)(n+1)x
so R=1 (Ratio Test).

6.
n ( n) ( n)
f (x) f ( 0)
0 ln (1+x) 0
 1
1 ( 1+x ) 1
 2
2  ( 1+x )  1
 3
3 2(1+x) 2
 4
4  6(1+x)  6
 5
5 24(1+x) 24
  
  
  
/ / / / /
ln (1+x) = f (0)+ f (0)x+ / f (0) 2 f (0) 3
x+ x
2! 3!
( 5)( 4)
f
(0) 4 f (0) 5
+ x+ x +  
4! 5!
1 2 2 3 6 4 24 5
=x x+ x x+ x   
2 6 24 120
2 3 4 5 n 1
x x x x  (  1)
+     =
n
=x +  x
2 3 4 5 n=1 n

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

a n+1
n+1 x n x
lim =lim  =lim = x <1 for convergence, so R=1 .
n  a n  n+1 n n  1+1/n
n x

7.
n ( n) ( n)
f (x) f ( 0)
5x
0 e 1
5x
1 5e 5
2 5x
2 5e 25
3 5x
3 5e 125
4 5x
4 5e 625
  
  
  
(n) n
 f (0) n  5 n
e = x =
5x
x .
n=0 n! n=0 n!
a n+1 n+1
n+1 5 x n!
lim = lim 
n  a n  (n+1)! n n
n 5 x
5 x
= lim n+1 =0<1forallx,soR= .
n 

8.
n ( n) ( n)
f (x) f ( 0)
x
0 xe 0
x
1 (x+1)e 1
x
2 (x+2)e 2
x
3 (x+3)e 3
  
  
  
(n) n
 f (0) n  n n  n n  x
xe = x = x = x =
x
.
n=0 n! n=0 n! n=1 n! n=1 (n 1)!

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

a n+1
n+1 x (n 1)! x
lim =lim  =lim =0<1 for all x , so R= .
n  a n  n! n
n  n
n x

9.
n ( n) ( n)
f (x) f ( 0)
0 sinh x 0
1 cosh x 1
2 sinh x 0
3 cosh x 1
4 sinh x 0
  
  
  
2n+1
f
( n)
(0)= { 0 if n is even
1 if n is odd
so sinh x=
2n+1

n=0
x
( 2n+1 )!
.

x
Use the Ratio Test to find R . If a = , then
n ( 2n+1 )!

a 2n+3
n+1 x (2n+1)!
lim = lim 
n  a n  (2n+3)! 2n+1
n x
2 1
= x  lim (2n+3)(2n+2) =0<1
n 
for all x , so R= .

10.
n ( n) ( n)
f (x) f ( 0)
0 cosh x 1
1 sinh x 0
2 cosh x 1
3 sinh x 0
  
  
  
2n
f (0)=
( n)
{
1 if n is even
0 if n is odd
Use the Ratio Test to find R . If
so cosh x=

n=0
x
( 2n )!
.

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

2n
x
a= , then
n (2n)!

a 2n+2
n+1 x (2n)!
lim = lim 
n  a n  (2n+2)! 2n
n x
2 1
= x  lim (2n+2)(2n+1) =0<1
n 
for all x , so R= .

11.
n ( n) ( n)
f (x) f ( 0)
1 1+2x 5
2 2 2
3 0 0
4 0 0
  
  
  
2 
0
f (x) =7+5(x 2)+ (x 2) +
2 n
(x 2)
2! n=3 n!
2
=7+5(x 2)+ ( x 2 )

Since a =0 for large n , R= .

n

12.
n ( n) ( n)
f (x) f ( 0)
3
0 x  1
2
1 3x 3
2 6x  6
3 6 6
4 0 0
5 0 0
  
  
  

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

6 2 6 3
f (x) = 1+3(x+1) (x+1) + (x+1)
2! 3!
2 3
= 1+3(x+1) 3(x+1) + ( x+1 )

Since a =0 for large n , R= .

n

3 3
( n) ( n)  e e
(3)=e and e =
x 3 x n n
13. Clearly, f (x)=e , so f (x 3) . If a = (x 3) , then
n=0 n! n n!
a 3 n+1
n+1 e (x 3) n! x 3
lim =lim  =lim =0<1 for all x , so R= .
n  a n  (n+1)! 3 n n  n+1
n e (x 3)

14.
n ( n) ( n)
f (x) f ( 0)
0 ln x ln 2
 1 1
1 x 2
 2 1
2  x 
4
 3 2
3 2x 8
 4 3 2
4  3 2x 
16
  
  
  

n 1 n 1 n
( n) ( 1) (n 1)!  ( 1) (x 2)
f (2)= for n 1 , so ln x=ln 2+ .
n n=1 n
2 n 2
a x 2 n x 2
n+1
lim = lim = <1 for convergence, so x 2 <2 R=2 .
n  a 2 n  n+1 2
n

15.
n ( n) ( n)
f (x) f ( 0)
0 cos x  1

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

1  sin x 0
2  cos x 1
3 sin x 0
4 cos x  1
  
  
  

(k) 2 4 6 2n
 f ( ) (x  ) (x  ) (x  )  (x  )
cos x=     =
k n+1
(x  ) = 1+  + ( 1) .
k=0 k! 2! 4! 6! n=0 (2n)!
a 2n+2 2
n+1 x  (2n)! x 
lim =lim  =lim =0<1 for all x , so R= .
 a  (2n+2)! 2n  (2n+2)(2n+1)
n
n
n x  n

16.
n ( n) ( n)
f (x) f ( 0)
0 sin x 1
1 cos x 0
2  sin x  1
3  cos x 0
4 sin x 1
  
  
  
(k)
sin x =  f ( /2)  k
x
k=0 k! 2
2 4 6
(x  /2) (x  /2) (x  /2)
=1 +  +  
2! 4! 6!
2n
 (x  /2)
=
n
( 1)
n=0 (2n)!

a 2n+2 2
n+1 x  /2 (2n)! x  /2
lim =lim  =lim =0<1 for all x , so R= .
 a  (2n+2)! 2n
n  (2n+2)(2n+1)
n
n
n x  /2

17.
n ( n) ( n)
f (x) f ( 0)
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

 1/2 1
0 x 3
1  1 1
1  x
3/2  
2 2 3
3
3  1 3 1
2 x
5/2    
4 2 2 5
3
15  1 3 5 1
3  x
7/2      
8 2 2 2 7
3
  
  
  

2 3
1 1 1 3 (x 9) 3 5 (x 9)
=  (x 9)+  +  
x 3 3 2 5 2! 3 7 3!
2 3 2 3 2 3
 n 1 3 5     (2n 1)
=
n
( 1) n 2n+1
(x 9) .
n=0
2  3  n!

a n+1 n 2n+1
n+1 1 3 5     (2n 1)[2(n+1) 1] x 9 2 3 !
lim =lim n+1 [2(n+1)+1]

 a n
n
n
n  2  3  (n+1)! 1 3 5     (2n 1) x 9
(2n+1) x 9 1
=lim = x 9 <1
n 
2 9
2 3 (n+1)

for convergence, so x 9 <9 and R=9 .

18.
n ( n) ( n)
f (x) f ( 0)
 2
0 x 1
 3
1  2x  2
 4
2 6x 6

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

 5
3  24x  24
 6
4 120x 120
  
  
  
2 3 4
 2 (x 1) (x 1) (x 1)
x =1 2(x 1)+6  24 +120    
2! 3! 4!
2 3 4
=1 2(x 1)+3(x 1)  4(x 1) +5(x 1)    

=
n n
( 1) (n+1)(x 1) .
n=0

a n+1
n+1 (n+2) x 1 n+2
lim =lim =lim  x 1 = x 1 <1 for convergence, so R=1 .
n  a n  n n  n+1
(n+1) x 1
n
( n+1 ) ( n+1 )
19. If f (x)=cos x , then f (x)=
sin x or
cos x . In each case, f (x)  1 , so by Formula 9
1 n+1
with a=0 and M=1 , R (x)  x . Thus, R (x)  0 as n  by Equation 10. So
n (n+1)! n
lim R (x)=0 and, by Theorem 8, the series in Exercise 3 represents cos x for all x .
n
n 

( n+1 ) ( n+1 )
20. If f (x)=sin x , then f(x)=
sin x or
cos x . In each case, f (x)  1 , so by Formula 9
1  n+1
with a=0 and M=1 , R (x)  x . Thus, R (x)  0 as n  by Equation 10. So
n (n+1)! 2 n
lim R (x)=0 and, by Theorem 8, the series in Exercise 16 represents sin x for all x .
n
n 

( n+1 )
21. If f (x)=sinh x , then for all n , f (x)=cosh x or sinh x . Since sinh x < cosh x =cosh x for all
( n+1 )
x , we have f (x)  cosh x for all n . If d is any positive number and x  d , then
( n+1 )
f (x)  cosh x cosh d , so by Formula 9 with a=0 and M=cosh d , we have
cosh d n+1
R (x)  x . It follows that R (x)  0 as n  for x  d (by Equation 10). But d
n ( n+1 )! n
was an arbitrary positive number. So by Theorem 8, the series represents sinh x for all x .

( n+1 )
22. If f (x)=cosh x , then for all n , f (x)=cosh x or sinh x . Since sinh x < cosh x =cosh x for all
( n+1 )
x , we have f (x)  cosh x for all n . If d is any positive number and x  d , then

10
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

( n+1 )
f (x)  cosh x cosh d , so by Formula 9 with a=0 and M=cosh d , we have
cosh d n+1
R (x)  x . It follows that R (x)  0 as n  for x  d (by Equation 10). But d
n ( n+1 )! n
was an arbitrary positive number. So by Theorem 8, the series represents cosh x for all x .

2n n 2n n 2n 2n
x  ( 1) ( x)   ( 1)  x
23. cos x= f (x)=cos ( x)= =
n
( 1) , R=
n=0 (2n)! n=0 (2n)! n=0 (2n)!

n n n
 x  x/2  (  x/2 )  (  1)
24. e = = =
x n
f (x)=e x , R=
n=0 n! n=0 n! n=0 n
2 n!

2n+1 2n+1 2n+2

 1 x  x  1   x
25. tan x= f (x)=xtan x=x =
n n n
( 1) ( 1) ( 1) , R=1
n=0 2n+1 n=0 2n+1 n=0 2n+1

26. sin x= ( 1)

x 

n
2n+1
f (x)=sin x = ( ) 4 
( 1)
n ( x4) 2n+1 =  (  1)
n
x
8n+4
, R=
n=0 (2n+1)! n=0 (2n+1)! n=0 (2n+1)!

n n n n+2
 x 2  x  (  x)  (  1) x
27. e =  =
x 2
f (x)=x e =x , R=
n=0 n! n=0 n! n=0 n!

2n 2n n 2n
x   ( 2x )  (  1) 2
28. cos x= cos 2x= =
n n 2n
(  1) ( 1) x
n=0 (2n)! n=0 (2n)! n=0 (2n)!
n 2n
 (  1) 2
f (x)=xcos 2x=
2n+1
x , R=
n=0 (2n)!

29.
n 2n n 2n
1
2 1  ( 1) (2x) 1  ( 1) (2x)
sin x = ( 1 cos 2x ) =
2 2
1  n=0 (2n)!
=
2
1 1  n=1 (2n)!
n+1 2n 1 2n
 ( 1) 2 x
= , R=
n=1 (2n)!

30.
n 2n n 2n 2n
12 1  (  1 ) (2x) 1  (  1) 2 x
cos x = (1+cos 2x)= 1+ = 1+1+
2 2 n=0 ( 2n )! 2 n=1 ( 2n )!

11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

n 2n 1 2n
 (  1) 2 x
=1+ , R=
n=1 ( 2n )!
2 2
Another method: Use cos x=1 sin x and Exercise 29.

n 2n+1 n 2n
sin x 1  ( 1) x  ( 1) x
31. =  = and this series also gives the required value at x=0
x x n=0 (2n+1)! n=0 (2n+1)!
(namely 1 ); R= .

32.
n 2n+1 n 2n+1 n+1 2n+3
x sin x 1  ( 1) x 1  ( 1) x 1  ( 1) x
3
=
3
x  n=0 (2n+1)!
=
3
x x  n=1 (2n+1)!
=
3
  n=0 (2n+3)!
x x x x
n 2n+3 n 2n
1 
( 1) x  ( 1) x
=
3
 n=0 (2n+3)!
=
n=0 (2n+3)!
and this series also gives the required value at x=0 (namely 1/6
x
); R= .

33.
n ( n) ( n)
f (x) f ( 0)
1/2
0 ( 1+x ) 1
1  1/2 1
1 (1+x)
2 2
1  3/2 1
2  (1+x) 
4 4
3  5/2 3
3 (1+x)
8 8
15  7/2 15
4  (1+x) 
16 16
  
  
  

n 1
( n) (  1 ) 1 3 5     ( 2n 3)
So f (0)= for n 2 , and
n
2

12
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

n 1 n 1
x  ( 1) 1 3 5     (2n 3) ( 1) 1 3 5     (2n 3)
1+x =1+ +
n n
x . If a = x ,
2 n=2 n n n
2 n! 2 n!
a n+1 n
n+1 1 3 5     (2n 3)(2n 1)x 2 n!
then lim =lim n+1

 a n
n n  2 (n+1)! 1 3 5     (2n 3)x
n
x 2n 1 x
= lim =  2= x <1 for convergence, so R=1 .
2 n  n+1 2

Notice that, as n increases, T (x) becomes a better approximation to f (x) for  1<x<1 .
n

n 2 2n 2n 2n
 x  x
( x )   x  n x
34. e = , so e = = . Also, cos x=
x n
( 1) ( 1) , so
n=0 n! n=0 n! n=0 n! n=0 (2n)!
2
 x  1 1 3 2 13 4 121 6
f (x)=e +cos x=
n 2n
( 1) + x =2 x+ x x +   .
n=0 n! (2n)! 2 24 720
x
The series for e and cos x converge for all x , so the same is true of the series for f (x) ; that is, R=
. From the graphs of f and the first few Taylor polynomials, we see that T (x) provides a closer fit to
n
f (x) near 0 as n increases.

13
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

2 2n
(  1) ( x )
2n n n 4n
x 
( )   (  1) x
35. cos x= f (x)=cos x = =
n 2
( 1) , R=
n=0 ( 2n )! n=0 (2n)! n=0 (2n)!

Notice that, as n increases, T (x) becomes a better approximation to f (x) .

n

n n n
( ) =e ln 2 x  ( xln 2 )  ( ln 2 ) x
= =
x xln 2
36. 2 = e , R= .
n! n=0 n=0 n!
Notice that, as n increases, T (x) becomes
n
a better approximation to f (x) .

n
 x
37. e =
x
, so
n=0 n!
n
 0.2 
( 0.2) 1 2 1 3 1 4 1 5 1
=
6
e =1 0.2+ (0.2)  (0.2) + (0.2)  (0.2) + (0.2)     . But
n=0 n! 2! 3! 4! 5! 6!
n
1  8  ( 0.2)

6 0.2 5
(0.2) =8.8 10 , so by the Alternating Series Estimation Theorem, e 0.81873
6! n=0 n!
, correct to five decimal places.

n 2n+1
  (  1) x
38. 3 = radians and sin x= , so
60 n=0 (2n+1)!

14
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

 3  5
3 5
  60 60   
sin =  +     =  +     . But
60 60 3! 5! 60 1,296,000 93,312,000,000
5
  8
<10 , so by the Alternating Series Estimation Theorem,
93,312,000,000
3
  
sin  0.05234 .
60 60 1,296,000
2n 3 2n 6n
x  (x )   x
39. cos x= cos (x )= =
n 3 n n
( 1) ( 1) ( 1)
n=0 (2n)! n=0 (2n)! n=0 (2n)!
6n+1 6n+2
x  x 
xcos (x )=  xcos (x )dx=C+
3 n 3 n
( 1) ( 1) , with R= .
n=0 (2n)! n=0 (6n+2)(2n)!

n 2n+1 n 2n
sin x 1  (  1) x  (  1) x
40. =  = , so
x x n=0 (2n+1)! n=0 (2n+1)!
n 2n n 2n+1
sin x  (  1) x  (  1) x
 x
dx=  n=0 (2n+1)!
dx=C+
n=0 (2n+1)(2n+1)!

3
41. Using the series from Exercise 33 and substituting x for x , we get
3 n 1
x  (  1 ) 1 3 5     ( 2n 3)
 1+ +
3n

3
x +1 dx = 2 n=2 n
x dx
2 n!
4 n 1
x  (  1 ) 1 3 5     ( 2n 3)
= C+x+ +
3n+1
x
8 n=2 n
2 n!(3n+1)
n n x n 1
 x  x e 1  x
42. e = e  1= =
x x

n=0 n! n=1 n! x n=1 n!
x n
e 1  x
 x
dx=C+
n=1 n n!
, with R= .

6n+2 1
x 
43. By Exercise 39,  xcos (x )dx=C+ , so  xcos (x )dx
3 3 n
( 1)
n=0 (6n+2)(2n)! 0
6n+2 1 n
 x  ( 1) 1 1 1 1
=  =
n
( 1) =  +  +   , but
n=0 (6n+2)(2n)! 0 n=0 (6n+2)(2n)! 2 8 2! 14 4! 20 6!
1 1
= 0.000069 , so
20 6! 14,400
15
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

1
1 1 1

3

xcos (x )dx
+ 0.440 (correct to three decimal places) by the Alternating Series
0 2 16 336
Estimation Theorem.

44. From the table of Maclaurin series in Section .10, we see that
2n+1 2n+1
 1 x   x
tan x= for x in  1,1 and sin x=
n n
( 1) ( 1) for all real numbers x , so
n=0 2n+1 n=0 (2n+1)!
6n+3 6n+3
 1 x   x
tan (x )+sin (x )= +
3 3 n n 3
( 1) ( 1) for x in  1,1  x in  1,1 . Thus,
n=0 2n+1 n=0 (2n+1)!
0.2 0.2
 1 1
I = dx=  
n 6n+3
( 1) x + dx
0 0
n=0 2n+1 (2n+1)!
6n+4 0.2
 x 1 1

n
= ( 1) +
n=0 6n+4 2n+1 (2n+1)! 0
6n+4 4 10
 (0.2) 1 1 (0.2) (0.2) 1 1
=
n
( 1) + = (1+1) + +   .
n=0 6n+4 2n+1 (2n+1)! 4 10 3 3!
10 10
(0.2) 1 1 (0.2)  9
But + = =5.12 10 , so by the Alternating Series Estimation Theorem,
10 3 3! 20
4
(0.2)
I =0.00080 (correct to five decimal places).
2

 1/2
45. We first find a series representation for f (x)= ( 1+x ) , and then substitute.
n ( n) ( n)
f (x) f ( 0)
 1/2
0 ( 1+x ) 1
1  3/2 1
1  (1+x) 
2 2
3  5/2 3
2 (1+x)
4 4
15  7/2 15
3  (1+x) 
8 8
  
  
  

16
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

2 3
1 x 3 x 15 x 1 1 3 3 6 5 9
=1 +  +   =1 x+ x x +  
1+x 2 4 2! 8 3! 3 2 8 16
1+x
0.1

 0
dx

1+x
3
= x
1 4 3 7 1 10
8
x+
56
x
32
x +  
0.1

0
( 0.1 ) 
1
8
4
(0.1) , by the Alternating Series

3 7  8
Estimation Theorem, since (0.1) 0.0000000054<10 , which is the maximum desired error.
56
0.1

Therefore,  0
dx

1+x
3
0.09998750 .

0.5 2 0.5 n 2n+2 n 2n+3 1/2 n

2  x  (  1) x  (  1) x  (  1)
46.  xe dx=   dx= = and since the
0 0
n=0 n! n=0 n!(2n+3) 0 n=0 2n+3
n!(2n+3)2
n
1 (  1) 1 1
<0.001 , we use 
1
term with n=2 is =  0.0354 .
1792 n=0 2n+3 24 160
n!(2n+3)2

47.
 1 1 3 1 5 1 7 1 3 1 5 1 7
x tan x x x x+ x x +   x x+ x   
lim 3 5 7 3 5 7
3 =lim =lim
x 0 x 3 3
x 0 x x 0 x
1 1 2 1 4 1
=lim  x+ x    =
x 0 3 5 7 3

since power series are continuous functions.

48.
1 2 1 4 1 6
1 cos x 1 x+ x
1 x +  
lim 2! 4! 6!
x =lim
x 0 1+x e 1 2 1 3 1 4 1 5 1 6
x 0 1+x 1+x+ x + x+ x+ x + x +  
2! 3! 4! 5! 6!

17
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

1 2 1 4 1 6
x x+ x   
2! 4! 6!
=lim
1 2 1 3 1 4 1 5 1 6
x 0  x x x x x   
2! 3! 4! 5! 6!
1 1 2 1 4 1
 x+ x     0
2! 4! 6! 2
=lim = = 1
1 1 1 2 1 3 1 4 1
x 0   x x x x      0
2! 3! 4! 5! 6! 2

since power series are continuous functions.

49.
1 3 1 3 1 5 1 7 1 3
sin x x+ x x x+ x x +    x+ x
6 3! 5! 7! 6
lim =lim
5 5
x 0 x x 0 x
1 5 1 7
x x +   2 4
5! 7! 1 x x 1 1
=lim =lim  +     = =
x 0 5 x 0 5! 7! 9! 5! 120
x

since power series are continuous functions.

50.
1 3 2 5 1 3 2 5
x+ x+ x +    x x+ x +  
tan x x 3 15 3 15 1 2 2 1
lim =lim =lim =lim + x +   =
x 0 3
x 0 3 x 0 3 x 0 3 15 3
x x x
since power series are continuous functions.

2 2 4 6 2 4
x x x  x x x
51. As in Example 8(a), we have e =1 +  +   and we know that cos x=1 +    
1! 2! 3! 2! 4!
2
 x 2 1 4 1 2 1 4
from Equation 16. Therefore, e cos x= 1 x + x     1 x+ x    . Writing only
2 2 24
the terms with degree  4 , we get
2
 x 1 2 1 4 2 1 4 1 4 3 2 25 4
e cos x=1 x+ x  x + x + x +   =1 x+ x +   .
2 24 2 2 2 24
52.

18
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

1 2 1 4
1+ x+ x   
2 24
1 2 5 4 1
1 x+ x    1 2 1 4
2 24
1 x+ x   
2 24
1 2 1 4
x x +  
2 24
1 2 1 4
x x +  
2 4
5 4
x +  
24
5 4
x +  
24
  

1 1 1 2 5 4
sec x= = . From the long division above, sec x=1+ x+ x +   .
cos x 1 2 1 4 2 24
1 x+ x   
2 24
53.

1 2 7 4
1+ x+ x +  
6 360
1 3 1 5 x
x x+ x    1 3 1 5
6 120
x x+ x   
6 120
1 3 1 5
x x +  
6 120
1 3 1 5
x x +  
6 36
7 5
x +  
360
7 5
x +  
360
  

19
 Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

x x x 1 2 7 4
= . From the long division above, =1+ x + x +   .
sin x 1 3 1 5 sin x 6 360
x x+ x   
6 120

1 2 1 3
54. From Example 6 in Section .9, we have ln (1 x)= x x x     , x <1 . Therefore,
2 3
x 1 2 1 2 1 3
e ln (1 x) = 1+x+ x +    x x x   
2 2 3
1 2 1 3 2 1 3 1 3
=  x x x x x x   
2 3 2 2
3 2 4 3
=  x x x     , x <1
2 3

55. 

( 1)
x
=
n
4n
 (  x4) n =e x 4
, by (11).
n=0 n! n=0 n!

 2n
n 2n
 (  1)   6  3
56.  =
n
( 1) =cos = , by (16).
n=0 2n n=0 (2n)! 6 2
6 (2n)!

n  2n+1
n 2n+1 (  1)
 (  1)   4  1
57.  = =sin = , by (15).
n=0 2n+1 n=0 (2n+1)! 4 2
4 (2n+1)!

n n
 3  ( 3/5)
58.  =
3/5
=e , by (11).
n=0 n n=0 n!
5 n!

1 2 3 4 n n
9 27 81 3 3 3 3  3  3
+   = + + + +   = =
3
59. 3+ + +  1=e  1 , by (11).
2! 3! 4! 1! 2! 3! 4! n=1 n! n=0 n!

2 3 n
( ln 2 ) ( ln 2 )  (  ln 2 )  ln 2
( ln 2)  1=2 1= 12
60. 1 ln 2+  +   = =e = e , by (11).
2! 3! n=0 n!
(x) M for a x a+d . Now  
/ / / / / / x / / / x
61. Assume that f (x)  M , so f f (t)dt M dt
a a

(a)+M(x a) . Thus,  
/ / / / / / / / x / / x / /
f (x) f (a) M(x a) f (x) f f (t)dt f (a)+M(t a) dt
a a

20
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.10 Taylor and Maclaurin Series

/ / / / 1 2 / / / / 1 2
f (x) f (a) f M(x a) f (x) f (a)+ f (a)(x a)+ M(x a)
(a)(x a)+
2 2
1
 a f (t)dt  a f (a)+ f (a)(t a)+ 2 M(t a) dt
x / x / / / 2

/ 1 / / 2 1 3
f (x) f (a) f (a)(x a)+ f (a)(x a) + M(x a) . So
2 6
/ 1 / / 2 1 3
f (x) f (a) f (a)(x a) f (a)(x a)  M(x a) . But
2 6
/ 1 / / 2 1 3
R (x)= f (x) T (x)= f (x) f (a) f (a)(x a) f (a)(x a) , so R (x) M(x a) .
2 2 2 2 6
/ / / 1 3
A similar argument using f (x)  M shows that R (x)  M(x a) . So
2 6
1 3
R x 
( ) M x a .
2 2 6
Although we have assumed that x>a , a similar calculation shows that this inequality is also true if
x<a .

{
2
 1/x
62. (a) f (x)= e if x 0 so
0 if x=0
2
 1/x
/ f (x) f (0) e 1/x x
f (0)=lim =lim =lim =lim =0 (using l’Hospital’s Rule and
x 0 x 0 x 0 x x 0
2
1/x x 0 1/x
2
e 2e
simplifying in the penultimate step). Similarly, we can use the definition of the derivative and
/ / ( 3) ( n)
l’Hospital’s Rule to show that f (0)=0 , f (0)=0 , ... , f (0)=0 , so that the Maclaurin series
for f consists entirely of zero terms. But since f ( x )  0 except for x=0 , we see that f cannot equal its
Maclaurin series except at x=0 .

(b)
From the graph, it seems that the function is extremely flat at the origin. In fact, it could be said to be
‘‘infinitely flat’’ at x=0 , since all of its derivatives are 0 there.

21
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

1. The general binomial series in (2) is

 k(k 1) 2 k(k 1)(k 2) 3
( 1+x ) = n=0 k x =1+kx+
k n
x+ x +   .
n 2! 3!
1 1 1 1 3
1   
 1 2 2 2 2 2

1/2 n 2 3
( 1+x ) =
2 x =1+ x+ x+ x +  
n=0 2 2! 3!
n
2 3 4
x x 1 3 x 1 3 5 x
= 1+  +  +  
2 2 3 4
2  2! 2  3! 2  4!
n 1 n
x  (  1 ) 1 3 5     (2n 3)x
= 1+ + for x <1,soR=1
2 n=2 n
2  n!

1  4 
2. = ( 1+x ) =  4 xn . The binomial coefficient is
4 n=0 n
( 1+x )
 4 ( 4)( 5)( 6)     (  4 n+1 ) ( 4)( 5)( 6)      (n+3)
= =
n n! n!
n n
= (  1 )  2 3 4 5 6     (n+1)(n+2)(n+3) = ( 1) (n+1)(n+2)(n+3)
2 3 n! 6
n
1 
( 1) (n+1)(n+2)(n+3) n
Thus, = x for x <1 , so R=1 .
4 n=0 6
( 1+x )

1 x  3 1  x n
1 1
3.
3
=
3
=
8
1+
2
=
8
 n=0
 3
2
. The binomial coefficient is
( 2+x ) 2 ( 1+x/2 ) n
 3 ( 3)( 4)( 5)     (  3 n+1 ) ( 3)( 4)( 5)      (n+2)
= =
n n! n!
n n
= (  1 )  2 3 4 5     (n+1)(n+2) ( 1) (n+1)(n+2)
=
2 n! 2
n n n n
1 1  ( 1) (n+1)(n+2) x  ( 1) (n+1)(n+2)x x
Thus, =  = for <1
3 8 n=0 2 n n=0 n+4 2
( 2+x ) 2 2
x <2 , so R=2 .

4.

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

 2
=
n
(1 x)
2/3
n=0 3 (  x)
n
2 1 2 1 4
  
2 3 3 2 3 3 3 3
=1+ ( x)+ (  x) + (  x ) +  
3 2! 3!
n 1 n
2  (  1) (  1 )  2 [1 4 7     ( 3n 5) ]
x+
n
=1 x
3 n=2 n
3  n!
2  1 4 7     ( 3n 5)
x 2
n
=1 x
3 n=2 n
3  n!

and  x <1 x <1 , so R=1 .

5.
 1
=
1/4 n
4
1 8x
= ( 1 8x )
n=0 4 (  8x )
n
1 3 1 3 7
  
1 4 4 2 4 4 4 3
=1+ (  8x ) + (  8x ) + (  8x ) +  
4 2! 3!
n n 1 n
 (  1) (  1)  3 7     ( 4n 5) 8
=1 2x+
n
n
x
n=2
4  n!
n
 3 7     ( 4n 5) 2 n
=1 2x  n=2 n!
x

1 1
and  8x <1 x < , so R= .
8 8

6.
 1/5 n n
1 1 1 x 1 1 x n 1  1 (  1) x
5
= 5 = 1 =  
5  =  
5
2 1 x/32 2 32 2 32 2 n=0
n=0 5n
32 x n n 2
1 6 1 6 11
  2    3
1 1 x 5 5 x 5 5 5 x
= 1+   + +  +  
2 5 5 2! 10 3! 15
2 2 2
2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

1 1 1 6 1 6 11 3 1  1 6     ( 5n 4 )
x +   = +
2 n
= + x+ x+ x
2 6 2 11 3 16 2 n=1 n 5n+1
5 2 5  2! 2 5  3! 2 52 n!

The radius of convergence is 32 .

7. We must write the binomial in the form ( 1+ expression), so we’ll factor out a 4 .
2  1/2 2 n
x x x x x x  1 x
2
= = = 1+ =  
2
2 4 2 n=0 4
4+x (
4 1+x /4 2 1+x /4
2
) 2
n
1 3 1 3 5
2   2 2    2 3
x 1 x 2 2 x 2 2 2 x
= 1+  + + +  
2 2 4 2! 4 3! 4
x x  n 1 3 5     ( 2n 1 ) 2n
= +  ( 1) x
2 2 n=1 n n
2  4  n!
2
x  1 3 5     ( 2n 1 ) x x
= +
n 2n+1
( 1) x and <1 <1
2 n=1 3n+1 4 2
n! 2

x <2 , so R=2 .

8.
2 2 2  1/2 2
x x x x  1 x n
x = = 1+
2
=  n=0

2 2
2+x 2 ( 1+x/2 ) 2 2
n
1 3 1 3 5
2   2    3
x 1 x 2 2 x 2 2 2 x
= 1+  + + +  
2 2 2 2! 2 3! 2
2 2
x x  1 3 5     ( 2n 1 )

n n
= + ( 1) x
2 2 n=1 2n
n! 2
2
x  1 3 5     ( 2n 1 ) x
+
n n+2
= ( 1) x and <1 x <2 , so R=2 .
2 n=1 2n+1/2 2
n! 2

9.

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

3 1 3 1 5
3/4   
(1+2x) 3 4 4 2 4 4 4 3
=1+ (2x)+ (2x) + (2x) +  
4 2! 3!
3  1 5 9     ( 4n 7)
=1+ x+3
n+1 n n
( 1)  2x
2 n=2 n
4  n!
3  1 5 9     ( 4n 7) 1 1
x+3
n+1 n
=1+ ( 1) x and 2x <1 x < , so R= .
2 n=2 n 2 2
2  n!

3 3 3 2 3 3 2 5 3
The three Taylor polynomials are T (x)=1+ x , T (x)=1+ x x , and T (x)=1+ x x+ x
1 2 2 2 8 3 2 8 16
.

10.
3 1/3
1+4x =(1+4x)
1 2 1 2 5
  
1 3 2 3 3 3 3 3
=1+ (4x)+ (4x) + (4x) +  
3 2! 3!
4  n+1 2 5 8     ( 3n 4 ) 1 1
=1+ x+
n
( 1) (4x) and 4x <1 x < , so R= .
3 n=2 n 4 4
3  n!

4 4 16 2
The three Taylor polynomials are T (x)=1+ x , T (x)=1+ x x , and
1 3 2 3 9
4 16 2 320 3
T (x)=1+ x x+ x .
3 3 9 81

11. (a)

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

 1/2
2
1/ 1 x = 1+  x ( 2)
1 3 1 3 5
    
1 2 2 2 2 2
=1+ 
2
(  x2) + 2!
(  x2) 2+ 3!
(  x2) 3+  
 1 3 5     ( 2n 1 )
=1+
2n
n
x
n=1
2  n!

(b)
1  1 3 5     ( 2n 1 )
= dx=C+x+
2n+1
 1 x
sin x n=1 n
1 x
2 (2n+1)2  n!
 1 3 5     ( 2n 1 )  1
=x+
2n+1
n
x since 0=sin 0=C .
n=1
(2n+1)2  n!

n 2n
( 1+x )2  1/2   (  1 )  1 3 5     ( 2n 1 ) x
12. (a) =  1/2 x2n=1+
n
n=0 n n=1
2  n!
n 2n+1
 1 dx  (  1 )  1 3 5     ( 2n 1 ) x  1
(b) sinh x= =C+x+ n
, but C=0 since sinh 0=0 , so
n=1
1+x
2 2  n! ( 2n+1 )
n 2n+1
 1  (  1 )  1 3 5     ( 2n 1 ) x
sinh x=x+ n
, R=1 .
n=1
2  n! ( 2n+1 )

13. (a)
 1
=(1+x) =
1/3 n
3
3 x
1+x n=0
n
1 2 1 2 5
  
1 3 3 2 3 3 3 3
=1+ x+ x+ x +  
3 2! 3!
x  2 5 8     (3n 4)
=1+ +
n+1 n
( 1) x
3 n=2 n
3  n!

3 1 1 2 5 3 3 3
(b) 1+x =1+ x x+ x    . 1.01 = 1+0.01 , so let x=0.01 . The sum of the first two
3 9 81
terms is then
5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

1 1 2
1+ (0.01) 1.0033 . The third term is (0.01) 0.00001 , which does not affect the fourth decimal
3 9
3
place of the sum, so we have 1.01 1.0033 .

14. (a)
 1/4  1
=
n
=(1+x) 
4 x
4
1/ 1+x n=0
n
1 5 1 5 9
    
1 4 4 2 4 4 4 3
=1 x+ x+ x +  
4 2! 3!
1  1 5 9     (4n 3)
x+
n n
=1 ( 1) x
4 n=2 n
4  n!

4 1 5 2 15 3 195 4 4 4
(b) 1/ 1+x =1 x+ x x+ x     . 1/ 1.1 =1/ 1+0.1 , so let x=0.1 . The sum of
4 32 128 2048
1 5 2 15 3
the first four terms is then 1 (0.1)+ (0.1)  (0.1) 0.976 . The fifth term is
4 32 128
195 4
(0.1) 0.0000095 , which does not affect the third decimal place of the sum, so we have
2048
4
1/ 1.1 0.976 . (Note that the third decimal place of the sum of the first three terms is affected by
the fourth term, so we need to use more than three terms for the sum.)

15. (a)
 2 ( 2)( 3) 2 ( 2)( 3)( 4) 3
1+ (  x ) =1+( 2)( x)+ ( x) + ( x) +  
2! 3!

=1+2x+3x +4x +   =
2 3 n
(n+1)x ,
n=0

x   
=x (n+1)x = =
n n+1 n
so (n+1)x nx .
2 n=0 n=0 n=1
( 1 x )
1 1
1  1 n  2
n 2
(b) With x= in part (a), we have  n = = = =2 .
2 n=1 2 n=1 n 1 2 1
2 1
2 4

16. (a)

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series


1+ (  x )
 3 =  3 (  x) n
n=0 n

( 3)( 4) 2 ( 3)( 4)( 5) 3

=1+( 3)( x)+ ( x) + ( x) +  
2! 3!
 3 4 5     ( n+2 ) n  2 3 4 5     ( n+2 ) n
=1+ x = x
n=1 n! n=0 2 n!
 (n+1)(n+2) n
= x

n=0 2

 3  3  3
( x+x2) 1+ (  x ) =x 1+( x) +x 1+( x)
2

 (n+1)(n+2) n+1  (n+1)(n+2) n+2

= x + x
n=0 2 n=0 2
 n(n+1) n  n(n+1) n+1
= x + x
n=1 2 n=1 2
 n(n+1) n  (n 1)n n  n(n+1) (n 1)n
=x+ x + x =x+
n
+ x
n=2 2 n=2 2 n=2 2 2
 
=x+ n x =
2 n 2 n
n x ,  1<x<1
n=2 n=1

1 1 2 3
2 +
1  n 2 2 4
(b) Setting x= in the last series above gives the required series, so  = = =6
2 n=1 n 1 3 1
2 1
2 8
.

17. (a)
1 1 1 1 3
2 1/2   
( 1+x ) =1+
1 2
x+
2 2
( x2) 2+ 2 2 2
( x2) 3+  
2 2! 3!
2 n 1
x  (  1 ) 1 3 5     ( 2n 3)
=1+ +
2n
x
2 n=2 n
2  n!

( 10 )
10 f (0)
(b) The coefficient of x (corresponding to n=5 ) in the above Maclaurin series is , so
10!

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

( 10 ) 4
f (0) (  1 )  1 3 5 7 ( 10 ) 1 3 5 7
=
f (0)=10! =99 , 225 .
10! 5 5
2  5! 2  5!

18. (a)
 1
( 1+x ) 3  1/2 =
n=0

2
( x3) n
n
1 3 1 3 5
    
1 2 2 3 2 2 2 2
=1+ 
2
( x3) + 2!
(x ) + 3!
( x3) 3+  
n 3n
 (  1 ) 1 3 5     ( 2n 1 ) x
=1+ n
n=1
2  n!

9
(b) The coefficient of x (corresponding to n=3 ) in the preceding series is
( 9) ( 9) 3
f (0) f (0) (  1 ) 1 3 5 ( 9) 9! 5
, so =
f (0)= = 113 , 400 .
9! 9! 3 8 2
2  3!

 
19. (a) g(x)= k xn
g (x)=
/
k nxn 1 , so
n=0 n n=1 n
  
(1+x)g (x) =(1+x)
/ k nxn 1= k nxn 1+ k nxn
n=1 n n=1 n n=1 n

= k (n+1)xn+  k nxn Replace n with n+1
n=0 n n=0 n+1 in the first series
 k(k 1)(k 2)   (k n+1)(k n) n
= (n+1) x
n=0 ( n+1 )!
 k(k 1)(k 2)   (k n+1)
+
n
(n) x
n=0 n!
 (n+1)k(k 1)(k 2)   (k n+1)
=
n
( k n ) +n x
n=0 ( n+1 )!
 k(k 1)(k 2)   (k n+1) n 
=k  x =k  k xn=kg(x)
n=0 n! n=0 n

/ kg(x)
Thus, g (x)= .
1+x
(b)
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.11 The Binomial Series

 k
h(x)= ( 1+x ) g(x)

/  k 1  k /
h (x) = k(1+x) g(x)+ ( 1+x ) g (x) [ProductRule
 k 1  k kg(x)
= k(1+x) g(x)+ ( 1+x )
1+x
 k 1  k 1
= k(1+x) g(x)+k(1+x) g(x)=0

(c) From part (b) we see that h(x) must be constant for x (  1,1 ) , so h(x)=h(0)=1 for x (  1,1 ) .
 k k
Thus, h(x)=1= ( 1+x ) g(x) g(x)= ( 1+x ) for x (  1,1 ) .

n 1 n
x  ( 1) 1 3 5     (2n 3)x
20. By Exercise .11.1, 1+x =1+ + , so
2 n=2 n
2  n!
2 1/2 1 2  1 3 5     (2n 3)
( 1 x ) =1
2
x  n=2 n
x and
2n

2  n!
1 2 2  1 3 5     (2n 3)

2 2 2n 2n
1 e sin  =1 e sin   e sin  . Thus,
2 n=2 n
2  n!
 /2  /2 1 2 2  1 3 5     (2n 3)
L =4a  1 e sin  d =4a  
2 2 2n 2n
1 e sin   e sin  d
0 0 2 n=2 n
2  n!
2 2 n
 e  1 3 5     (2n 3) e
=4a 
2 2 1
S   n=2 n! 2
S
n

 /2 1 3 5     (2n 1) 
where S =
2n
sin  d = by Exercise 44 of 8.1.
n 0 2 4 6     2n 2
2 2 n
 e 1  1 3 5     (2n 3) e 1 3 5     (2n 1)
L = 4a
2
1  
2 2
 n=2 n! 2 2 4 6     2n
2 2n 2 2 2 2
e  e 1  3  5      (2n 3) (2n 1)
= 2 a 1
4
  n=2 n
 n
2 n! 2  n!
2 2n
e  e 1 3     (2n 3) 2
= 2 a 1
4
  n=2 n n!
(2n 1)
4
2 4 6
= 2 a e 3e 5e  a 2 4 6
1       = (256 64e  12e  5e     )
4 64 256 128

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

1. (a)
n ( n) ( n) T (x)
f (x) f (0) n
0 cos x 1 1
1  sin x 0 1
1 2
2  cos x  1 1 x
2
1 2
3 sin x 0 1 x
2
1 2 1 4
4 cos x 1 1 x+ x
2 24
1 2 1 4
5  sin x 0 1 x+ x
2 24
1 2 1 4 1 6
6  cos x  1 1 x+ x x
2 24 720

(b)
x f T =T T =T T =T T
0 1 2 3 4 5 6

0.7071 1 0.6916 0.7074 0.7071
4

0 1  0.2337 0.0200  0.0009
2
  1 1  3.9348 0.1239  1.2114

(c) As n increases, T (x) is a good approximation to f (x) on a larger and larger interval.
n

2. (a)
n ( n) ( n) T (x)
f (x) f (0) n

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

 1
0 x 1 1
 2
1  x  1 1 ( x 1 ) =2 x
 3 2 2
2 2x 2 1 ( x 1 ) + ( x 1 ) =x  3x+3
 4 2 3 3 2
3  6x  6 1 ( x 1 ) + ( x 1 )  ( x 1 ) = x +4x  6x+4

(b)
x f T T T T
0 1 2 3
0.9 1.1 1 1.1 1.11 1.111
1.3 0.7692 1 0.7 0.79 0.763

(c) As n increases, T (x) is a good approximation to f (x) on a larger and larger interval.
n

3.
n ( n) ( n)
f (x) f (1)
0 ln x 0
1 1/x 1
2
2  1/x  1
3
3 2/x 2
4
4  6/x  6

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

( n)
f (1) 1 2 1 3 1
T (x)=
4 n 4
(x 1) =0+ ( x 1 )  (x 1) + (x 1)  (x 1)
4 n=0 n! 2 3 4

4.
n ( n) ( n)
f (x) f (2)
x 2
0 e e
x 2
1 e e
x 2
2 e e
x 2
3 e e

( n) 2 2
f (2) e 2 e
T (x)=
3 n 2 2 3
(x 2) =e +e (x 2)+ (x 2) + (x 2)
3 n=0 n! 2 6

5.
( n) ( n) 
n f (x) f
6
1
0 sin x
2
3
1 cos x
2

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

1
2  sin x 
2
3
3  cos x 
2

( n) 
f
6  n 1 3  1  2 3  3
T (x)=
3
x = + x  x  x
3 n=0 n! 6 2 2 6 4 6 12 6

6.
( n) ( n) 2
n f (x) f
3
1
0 cos x 
2
3
1  sin x 
2
1
2  cos x
2
3
3 sin x
2
1
4 cos x 
2

( n) 2
f n
3 2
T (x)=
4
x
4 n=0 n! 3
4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

1 3 2 1 2 2 3 2 3 1 2 4
=  x + x + x  x
2 2 3 4 3 12 3 48 3

7.
n ( n) ( n)
f (x) f ( 0)
0 arcsinx 0
1 2 1
1/ 1 x
2 3/2
2 x/(1 x ) 0
2 2 5/2
3 (2x +1)/(1 x ) 1

( n) 3
f (0) n x
T (x)=
3
x =x+
3 n=0 n! 6

8.
n ( n) ( n)
f (x) f ( 1)
0 (ln x)/x 0
2
1 (1 ln x)/x 1
3
2 ( 3+2ln x)/x  3
4
3 (11 6ln x)/x 11

( n)
f (1) 3 2 11
T (x)=
3 n 3
(x 1) =(x 1) (x 1) + (x 1)
3 n=0 n! 2 6

9.

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

n ( n) ( n)
f (x) f ( 0)
 2x
0 xe 0
 2x
1 (1 2x)e 1
 2x
2 4(x 1)e  4
 2x
3 4(3 2x)e 12

( n)
f (0) n 0 1 1  4 2 12 3
T (x)=
3 2 3
x =  1+ x + x+ x =x 2x +2x
3 n=0 n! 1 1 2 6

10.
n ( n) ( n)
f (x) f (1)
0 ( 3+x2) 1/2 2
2  1/2 1
1 x ( 3+x ) 2
2  3/2 3
2 3 ( 3+x ) 8

( n)
f (1) 1 3/8 1 3
T (x)=
2 n 2 2
(x 1) =2+ (x 1)+ (x 1) =2+ (x 1)+ (x 1)
2 n=0 n! 2 2 2 16

11. In Maple, we can find the Taylor polynomials by the following method: first define f:=sec(x);
and then set
6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

T2:=convert(taylor(f,x=0,3),polynom);, T4:=convert(taylor(f,x=0,5),polynom);,
etc. (The third argument in the taylor function is one more than the degree of the desired
polynomial). We must convert to the type polynom because the output of the taylor function
contains an error term which we do not want. In Mathematica, we use
Tn:=Normal[Series[f,{x,0,n}]] , with n=2,4, etc. Note that in Mathematica, the "degree"
argument is the same as the degree of the desired polynomial. In Derive, author sec x , then enter
Calculus ,Taylor,8,0; and then simplify the expression. The eighth Taylor polynomial is
1 2 5 4 61 6 277 8
T (x)=1+ x + x+ x+ x .
8 2 24 720 8064

12. See Exercise 11 for the CAS commands used to generate the Taylor polynomials. The ninth
1 3 2 5 17 7 62 9
Taylor polynomial for tan x is T (x)=x+ x + x+ x+ x .
9 3 15 315 2835

13.
n ( n) ( n)
f (x) f (4)
0 x 2
1  1/2 1
1 x
2 4
1  3/2 1
2  x 
4 32
3  5/2
3 x
8

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

1 1/32 2 1 1 2
(a) f (x)= x  T (x)=2+ (x 4) (x 4) =2+ (x 4) (x 4)
2 4 2! 4 64
M 3 / / / 3
(b) R (x)  x 4 , where f (x)  M. Now 4 x 4.2 x 4  0.2 x 4  0.008 .
2 3!
/ / / / / / 3  5/2 3
Since f (x) is decreasing on 4,4.2 , we can take M= f (4) = 4 = , so
8 256
3/256 0.008
R (x)  (0.008)= =0.000015625 .
2 6 512
 5
(c) From the graph of R (x) = x  T (x) , it seems that the error is less than 1.52 10 on 4,4.2
2 2
.

14.
n ( n) ( n)
f (x) f (1)
 2
0 x 1
 3
1  2x  2
 4
2 6x 6
 5
3  24x

(a)
 2
f (x) =x  T (x)
2
6 2
=1 2(x 1)+ (x 1)
2!
2
=1 2(x 1)+3(x 1)

M 3 / / / 3
(b) R (x)  x 1 , where f (x)  M . Now 0.9 x 1.1 x 1  0.1 x 1  0.001 .
2 3!
/ / /
Since f (x) is decreasing on 0.9,1.1 , we can take
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

/ / / 24
M= f (0.9) = , so
5
( 0.9 )
5
R (x)  24/(0.9) 0.004
2 (0.001)=
6 0.59049
 0.00677404

(c)

 2
From the graph of R (x) = x  T (x) , it seems that the error is less than 0.0046 on 0.9,1.1 .
2 2

15.
n ( n) ( n)
f (x) f (1)
2/3
0 x 1
2  1/3 2
1 x
3 3
2  4/3 2
2  x 
9 9
8  7/3 8
3 x
27 27
56  10/3
4  x
81

2/3 2 2/9 2 8/27 3 2 1 2 4 3

(a) f (x)=x  T (x)=1+ (x 1) (x 1) + (x 1) =1+ (x 1) (x 1) + (x 1)
3 3 2! 3! 3 9 81
M 4 (4) 4
(b) R (x)  x 1 , where f (x)  M . Now 0.8 x 1.2 x 1  0.2 x 1  0.0016 .
3 4!
(4) (4) 56  10/3
Since f (x) is decreasing on 0.8,1.2 , we can take M= f (0.8) = (0.8) , so
81
56  10/3
(0.8)
81
R (x)  (0.0016) 0.00009697 .
3 24

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

2/3
(c) From the graph of R (x) = x  T (x) , it seems that the error is less than 0.0000533 on
3 3
0.8,1.2 .

16.
( n) ( n) 
n f (x) f
3
1
0 cos x
2
3
1  sin x 
2
1
2  cos x 
2
3
3 sin x
2
1
4 cos x
2
5  sin x

(a)
f (x)=cos x  T 4(x)
1 3  1  2 3  3 1  4
=  x  x + x + x
2 2 3 4 3 12 3 48 3

M  5 ( 5) 2  5  5
(b) R (x)  x , where f (x)  M . Now 0 x  x  , and
4 5! 3 3 3 3
 1  5
letting x= gives M=1 , so R (x)   0.0105 .
2 4 5! 3
(c)

10
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

2
From the graph of R (x) = cos x T (x) , it seems that the error is less than 0.01 on 0, .
4 4 3

17.
n ( n) ( n)
f (x) f (0)
0 tan x 0
2
1 sec x 1
2
2 2sec x tan x 0
2 2 4
3 4sec x tan x+2sec x 2
2 3 4
4 8sec x tan x+16sec x tan x

1 3
(a) f (x)=tan x T (x)=x+ x
3 3

M 4 ( 4)  4  4 
(b) R (x)  x , where f (x)  M . Now 0 x  x , and letting x=
3 4! 6 6 6
( 4) 
since f isincreasingon 0, gives
6
2 2 1 3 2 4 1
R (x)  8 +16
3 3 3 3 3  4

4! 6
4 3  4
=  0.057859
9 6

(c)

From the graph of

11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

R (x) = tan x T (3) , it seems that the error is less than 0.006 on 0, /6 .
3 3

18.
n ( n) ( n)
f (x) f (1)
0 ln (1+2x) ln 3
2
1 2/(1+2x)
3
2 4
2  4/(1+2x) 
9
3 16
3 16/(1+2x) 27
4
4  96/(1+2x)

2 4/9 2 16/27 3
(a) f (x)=ln (1+2x) T (x)=ln 3+ (x 1) (x 1) + (x 1)
3 3 2! 3!
M 4 (4)
(b) R (x)  x 1 , where f (x)  M . Now 0.5 x 1.5  0.5 x 1 0.5 x 1  0.5
3 4!
4 1 6 1 1
x 1  , and letting x=0.5 gives M=6 , so R (x)   = =0.015625 .
16 3 4! 16 64

(c)
From the graph of R (x) = ln (1+2x) T (x) , it seems that the error is less than 0.005 on 0.5,1.5 .
3 3

19.
n ( n) ( n)
f (x) f (0)
2
0 x 1
e
2
1 x 0
e (2x)
2
2 x 2 2
e (2+4x )
2
3 x 3 0
e (12x+8x )
2
4 x 2 4
e (12+48x +16x )
12
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

2
x 2 2 2
(a) f (x)=e  T (x)=1+ x =1+x
3 2!

M 4 ( 4)
(b) R (x)  x , where f (x)  M .
3 4!
4 4
Now 0 x 0.1 x  ( 0.1 ) , and
letting x=0.1 gives
0.01
e ( 12+0.48+0.0016 ) 4
R (x)  (0.1)  0.00006 .
3 24

(c)

2
From the graph of R (x) = e  1+x
3
x
( 2
) , it appears that the error is less than 0.000051 on 0,0.1 .

20.
n ( n) ( n)
f (x) f (1)
0 x ln x 0
1 ln x+1 1
2 1/x 1
2
3  1/x  1
3
4 2/x

1 2 1 3
(a) f (x)=xln x T (x)=(x 1)+ (x 1)  (x 1)
3 2 6
M 4 (4) 1 4 1
(b) R (x)  x 1 , where f (x)  M . Now 0.5 x 1.5  x 1   x 1  .
3 4! 2 16
(4) (4) 3
Since f (x) is decreasing on 0.5,1.5 , we can take M= f (0.5) =2/(0.5) =16 , so
16 1
R (x)  (1/16)= =0.0416 .
3 24 24
(c) From the graph of R (x) = xln x T (x) , it seems that the error is less than 0.0076 on 0.5,1.5
3 3
.
13
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

21.
n ( n) ( n)
f (x) f (0)
0 xsin x 0
1 sin x+xcos x 0
2 2cos x xsin x 2
3  3sin x xcos x 0
4  4cos x+xsin x  4
5 5sin x+xcos x

2 2  4 4 2 1 4
(a) f (x)=xsin x T (x)= (x 0) + (x 0) =x  x
4 2! 4! 6
M 5 (5) (5)
(b) R (x)  x , where f (x)  M . Now  1 x 1 x  1 , and a graph of f (x) shows
4 5!
(5) 5 5 1
that f (x)  5 for  1 x 1 . Thus, we can take M=5 and get R (x)   1= =0.0416 .
4 5! 24
(c) From the graph of R (x) = xsin x T (x) , it seems that the error is less than 0.0082 on  1,1 .
4 4

22.
n ( n) ( n)
f (x) f (0)
0 sinh 2x 0
1 2cosh 2x 2
2 4sinh 2x 0
3 8cosh 2x 8
4 16sinh 2x 0
5 32cosh 2x 32
14
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

6 64sinh 2x

8 3 32 5 4 3 4 5
(a) f (x)=sinh 2x T (x)=2x+ x+ x =2x+ x + x
5 3! 5! 3 15
M 6 (6) (6)
(b) R (x)  x , where f (x)  M . For x in  1,1 , we have x  1 . Since f (x) is an
5 6!
(6) (6) 2  2
increasing odd function on  1,1 , we see that f (x)  f (1)=64sinh 2=32(e  e ) 232.119 , so
232.12 6
we can take M=232.12 and get R (x)   1  0.3224 .
5 720
(c) From the graph of R (x) = sinh 2x T (x) , it seems that the error is less than 0.027 on  1,1 .
5 5

1 3  1  2 3  3
23. From Exercise 5, sin x= + x  x  x +R (x) , where
2 2 6 4 6 12 6 3
M  4 ( 4)


 
R (x)  x with f (x) = sin x  M=1 . Now x=35 =(30 +5 )= +
3 4! 6 6 36
radians, so the error is
 4

 36
R  <0.000003 . Therefore, to five decimal places,
3 36 4!

1 3  1  2 3  3
sin 35  +    0.57358 .
2 2 36 4 36 12 36

24. From Exercise 16,

1 3  1  2 3  3 1  4
cos x=  x  x + x + x +R (x) . Now since
2 2 3 4 3 12 3 48 3 4
 5

  20  7
x=69 =(60 +9 )= + radians, the error is R (x)  <8 10 . Therefore, to five
3 20 4 5!

1 3  1  2 3  3 1  4
decimal places, cos 69    + +  0.35837 .
2 2 20 4 20 12 20 48 20

15
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

x
x x e n+1
25. All derivatives of e are e , so R (x)  x , where 0<x<0.1 . Letting x=0.1 ,
n (n+1)!
0.1
e n+1
R (0.1) (0.1) <0.00001 , and by trial and error we find that n=3 satisfies this inequality
n (n+1)!
x
since R (0.1)<0.0000046 . Thus, by adding the four terms of the Maclaurin series for e
3
0.1
corresponding to n=0 , 1 , 2 , and 3 , we can estimate e to within 0.00001 . (In fact, this sum is
0.1
1.10516 and e  1.10517 .)

n
 x
26. Example 6 in Section .9 gives the Maclaurin series for ln (1 x) as   n=1 n
for x <1 . Thus,
n n
( 0.4)   n+1 (0.4)
ln 1.4=ln [1 ( 0.4)]=  = ( 1) . Since this is an alternating series, the error
n=1 n n=1 n
is less than the first neglected term by the Alternating Series Estimation Theorem, and we find that
6
a = ( 0.4 ) /6 0.0007<0.001 . So we need the first five (non zero) terms of the Maclaurin series for
6
the desired accuracy. (In fact, this sum is approximately 0.33698 and ln 1.4 0.33647 .)

1 3 1 5
27. sin x=x x+ x     . By the Alternating Series Estimation Theorem, the error in the
3! 5!
1 3 1 5 5 1/5
approximation sin x=x x is less than x <0.01 x <120(0.01) x < ( 1.2 )  1.037 .
3! 5!
1 3
The curves y=x x and y=sin x 0.01 intersect at x 1.043 , so the graph confirms our estimate.
6
Since both the sine function and

the given approximation are odd functions, we need to check the estimate only for x>0 . Thus, the
desired range of values for x is  1.037<x<1.037 .

28.

16
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

1 2 1 4 1 6
cos x=1 x+ x x +   . By the Alternating Series Estimation Theorem, the error is less
2! 4! 6!
1 6 6 1/6 1 2 1 4
than  x <0.005 x <720(0.005) x < ( 3.6 )  1.238 . The curves y=1 x+ x and
6! 2 24
y=cos x+0.005 intersect at x 1.244 , so the graph confirms our estimate. Since both the cosine
function and the given approximation

are even functions, we need to check the estimate only for x>0 . Thus, the desired range of values for
x is  1.238<x<1.238 .

29. Let s(t) be the position function of the car, and for convenience set s(0)=0 . The velocity of the
car is
/ / /
v(t)=s (t) and the acceleration is a(t)=s (t) , so the second degree Taylor polynomial is
a(0) 2 2
T (t)=s(0)+v(0)t+ t =20t+t . We estimate the distance travelled during the next second to be
2 2
s(1) T (1)=20+1=21 m. The function T (t) would not be accurate over a full minute, since the car
2 2
2
could not possibly maintain an acceleration of 2 m / s for that long (if it did, its final speed would be
140 m / s  313 mi / h!)

30. (a)
(n) (n)
n  (t)  (20)
 ( t 20 )
0  e 
20 20
 ( t 20 )
1   e  
20 20
2  ( t 20 ) 2
2   e  
20 20

/
The linear approximation is T (t)= (20)+ (20)(t 20)= 1+ (t 20) . The quadratic
1 20
approximation is

17
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

/ /
/  (20) 2 1 2 2
T (t)= (20)+ (20)(t 20)+ (t 20) = 1+ (t 20)+  (t 20)
2 2 20 2

(b)

(c)

From the graph, it seems that T (t) is within 1% of  (t) , that is, 0.99 (t) T (t) 1.01 (t) , for  14
1 1

C  t 58 C.

d  2
q q q q q
31. E=  =  = 1 1+ .
2 2 2 2 2 2 D
D ( D+d ) D D ( 1+d/D ) D
 2
We use the Binomial Series to expand ( 1+d/D ) :
d 2 3 d 2 2 3 4 d 3
q
E= 1 1 2 +  +  
2 D 2! D 3! D
D
d d 2 d 3 d
q q 1
= 2  3 +4       2 =2qd
2 D D D 2 D 3
D D D
when D is much larger than d ; that is, when P is far away from the dipole.

n n 1 ns ns
1 2 2 i 1 o
32. (a) + =  (Equation 1) where
R
o i i o

18
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

2 2 2 2
= R + s +R  2R s +R cos 
( ) ( ) and = R + s  R +2R s  R cos 
( ) ( ) (2)
o o o i i i

Using cos   1 gives

2 2 2 2 2 2 2
= R + s +R  2R s +R =
( ) ( ) R +s +2Rs +R  2Rs  2R = s =s
o o o o o o o o

and similarly, =s . Thus, Equation 1 becomes

i i

n n ns ns n n n n
1 2 1 2 i 1 o 1 2 2 1
+ =   + =
s s R s s s s R
o i i o o i

1 2
(b) Using cos   1  in (2) gives us
2

2 2 1 2
o = R + s +R  2R s +R
( ) ( ) 1 
o o 2
2 2 2 2 2 2 2 2 2 2 2
= R +s +2Rs +R  2Rs +Rs   2R +R  = s +Rs  +R 
o o o o o o
k
Anticipating that we will use the binomial series expansion ( 1+x )  1+kx , we can write the last
2 2
2 R R 2 R R
expression for as s 1+ + and similarly, =s 1   . Thus,
o o s 2 i i s s2
o s i
o i

n n ns ns n s n s
1 2 1 2 i 1 o  1  1 2 i 1 o
from Equation 1, + =   n +n =    
R 1 o 2 i R R
o i i o i o
n 2  1/2 n 2  1/2
1 2 R R 2 2 R R
1+ + + 1  
s s 2 s s s2
o o s i i
o i
n 2  1/2 n 2  1/2
2 2 R R 1 2 R R
= 1    1+ +
R s s2 R s 2
i o s
i o
 1  1
Approximating the expressions for and by the first two terms in their binomial series, we get
o i

19
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

n 2 n 2
1 1 2 R R 2 1 2 R R
1  + + 1+  
s 2 s 2 s 2 s s2
o o s i i
o i
n 2 n 2
2 1 2 R R 1 1 2 R R
= 1+    1  + 
R 2 s s2 R 2 s 2
i o s
i o
2 2
n n 2 n n 2
1 1 R R 2 2 R R
 + + + 
s 2s s 2 s 2s s s2
o o o s i i i
o i
2 2
n n 2 n n 2
2 2 R R 1 1 R R
= +   + + 
R 2R s s2 R 2R s 2
i o s
i o
n n
1 2
+
s s
o i
2 2 2 2
n n n 2 n 2 n 2 n 2
2 1 1 R R 1 R R 2 R R 2 R R
=  + + + + +   
R R 2s s 2 2R s 2 2R s s2 2s s s2
o o s o s i i i
o o i i
2 2
n n n 2 n 2
2 1 1 R R 1 1 2 R R 1 1
= + + + +  
R 2 s 2 s R 2 s s2 R s
o s o i i
o i
2 2 2 2
n n n R n R
2 1 1 1 1 1 1 2 1 1 1 1
= + + + +  
R 2s R s R s 2s R s R s
o o o i i i
n n n 2 n 2
2 1 2 2 1 1 1 2 1 1
= + R + + 
R 2s R s 2s R s
o o i i

From Figure 8, we see that sin  =h/R . So if we approximate sin  with  , we get h=R and
2 2 2
h = R and hence, Equation 4, as desired.

33. (a) If the water is deep, then 2 d/L is large, and we know that tanh x 1 as x  . So we can
2
approximate tanh (2 d/L) 1 , and so v  gL/(2 ) v gL/(2 ) .
(b) From the table, the first term in the Maclaurin series of tanh x is x , so if the water is shallow, we
can approximate

20
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

2 d 2 d 2 gL 2 d
tanh  , and so v    v gd .
L L 2 L
(n) (n)
n f (x) f (0)
0 tanh x 0
2
1 sech x 1
2
2  2sech x tanh x 0
2 2
3 2sech x(3tanh x 1)  2

(c) Since tanh x is an odd function, its Maclaurin series is alternating, so the error in the
2 d 2 d
approximation tanh  is less than the first neglected term, which is
L L
/ / / 3 3
f (0) 2 d 1 2 d
= .
3! L 3 L
3
1 2 d 3 1 1 3  2
If L>10d , then < 2  = , so the error in the approximation v =gd is
3 L 3 10 375
3
gL 
less than   0.0132gL .
2 375

 /2

34. (a) 4
L
g
 0
dx
2
1 k sin x
2
=4
L
g
 /2

0
(
1+  k sin x
2 2
)  1/2
dx

 /2 1 3 1 3 5


  
L 1 2 2 2 2 2
=4
g
1
2
2 2
(
 k sin x +
2!
) (  k2sin 2x) 2 3!
(  k2sin 2x) 3+   dx
0
 /2
L 1 1 3 1 3 5

2 2 4 4 6 6
=4 1+ k sin x+ k sin x+ k sin x+   dx
g 0 2 2 4 2 4 6
L  1 1  2 1 3 1 3  4
=4 +  k+  k
g 2 2 2 2 2 4 2 4 2
1 3 5 1 3 5  6
+  k +  
2 4 6 2 4 6 2
[split up the integral and use the result from Exercise 8.1.44]
2 2 2 2 2 2
L 1 2 1 3 4 1 3 5 6
=2 1+ k+ k+ k +  
g 2 2 2 2 2 2
2 2 4 2 4 6
21
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

(b) The first of the two inequalities is true because all of the terms in the series are positive. For the
second,
2 2 2 2 2 2 2 2 2 2
L 1 2 1 3 4 1 3 5 6 1 3 5 7 8
T = 2 1+ k+ k+ k+ k +  
g 2 2 2 2 2 2 2 2 2 2
2 2 4 2 4 6 2 4 6 8
L 1 2 1 4 1 6 1 8
 2 1+ k + k + k + k +  
g 4 4 4 4
1 2 2 2 1
The terms in brackets (after the first) form a geometric series with a= k and r=k =sin  <1
4 2 0
2 2
L k /4 L 4 3k
. So T  2 1+ =2 .
g 2 g 2
1 k 4 4k

(
(c) We substitute L=1 , g=9.8 , and k=sin 10 /2  0.08716 , and the inequality from part (b) )
becomes 2.01090 T  2.01093 , so T  2.0109 . The estimate T  2 L/g 2.0071 differs by about
0.2% .

If  =42 , then k 0.35837 and the inequality becomes 2.07153 T  2.08103 , so T  2.0763 . The
0
one term estimate is the same, and the discrepancy between the two estimates increases to about
3.4% .

35. (a) L is the length of the arc subtended by the angle  , so L=R   =L/R . Now sec  =(R+C)/R
 Rsec  =R+C  C=Rsec   R=Rsec (L/R) R .

1 2 5 4
(b) From Exercise 11, sec x T (x)=1+ x+ x . By part (a),
4 2 24
2 4 2 4 2 4
1 L 5 1 L 5 L L L 5L
C R 1+  R=R+ R
+ + R  R= + .
2 R 24 2 2 24 R 4 2R 3
R R 24R
(c) Taking L=100 km and R=6370 km, the formula in part (a) says that C=Rsec (L/R) R=6370 sec
(100/6370) 6370 0.78500996544 km
2 4 2 4
L 5L 100 5 100
The formula in part (b) says that C + = +  0.78500995736 km.
2R 3 2 6370 3
24R 24 6370

22
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.12 Applications of Taylor Polynomials

The difference between these two results is only 0.00000000808 km, or 0.00000808 m!

/ / / (n)
f (a) f (a) 2 f (a) n
36. T (x)= f (a)+ (x a)+ (x a) +   + (x a) . Let 0 m n . Then
n 1! 2! n!
(m) (m+1)
(m) f (a) 0 f (a) 1
T (x)=m! (x a) +(m+1)(m)   (2) (x a) +  
n m! ( m+1 )!
(n)
f (a) n m
+n(n 1)   (n m+1) (x a)
n!
(m)
(m) m! f (a) (m)
For x=a , all terms in this sum except the first one are 0 , so T (a)= = f (a) .
n m!

37. Using f (x)=T (x)+R (x) with n=1 and x=r , we have f (r)=T (r)+R (r) , where T is the first
n n 1 1 1
/
degree Taylor polynomial of f at a . Because a=x , f (r)= f x + f ( ) ( x ) (r x ) +R (r) . But r is a n n n n 1
/
root of f , so f (r)=0 and we have 0= f ( x ) + f ( x ) ( r x ) +R (r) . Taking the first two terms to the
n n n 1
/ /
left side gives us f ( x ) ( x  r )  f ( x ) =R (r) . Dividing by f ( x ) , we get
n n n 1 n
f (x ) R (r)
n 1
x  r = . By the formula for Newton’s method, the left side of the preceding
n / /
f (x ) f (x )
n n
R (r)
1
equation is x  r , so x  r = . Taylor’s Inequality gives us
n+1 n+1 /
f (x )n
/ /
f (r) 2 / /
R (r)  r x . Combining this inequality with the facts f (x)  M and
1 2! n
/ M 2
f (x)  K gives us x  r  x r .
n+1 2K n

23
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

1. (a) A sequence is an ordered list of numbers whereas a series is the sum of a list of numbers.
(b) A series is convergent if the sequence of partial sums is a convergent sequence. A series is
divergent if it is not convergent.


2.  a =5 means that by adding sufficiently many terms of the series we can get as close as we like
n=1 n
to the number 5 .
s =5 , where s is the n th partial sum, that is, 
n
In other words, it means that a .
n  n n i=1 i

3.
n s
n
1  2.40000
2  1.92000
3  2.01600
4  1.99680
5  2.00064
6  1.99987
7  2.00003
8  1.99999
9  2.00000
10  2.00000

From the graph and the table, it seems that the series converges to  2 . In fact, it is a geometric series
1  12  2.4  2.4
with a= 2.4 and r= , so its sum is  = = = 2. Note that the dot
5 n=1 n 1 1.2
( 5) 1 
5
{ } { }
corresponding to n=1 is part of both a and s .
n n
TI 86 Note: To graph { a } and { s } , set your calculator to Param mode and DrawDot mode.
n n
(DrawDot is under GRAPH, MORE, FORMT (F3). ) Now under E(t)= make the assignments:
xt1=t, yt1=12/( 5)^t, xt2=t, yt2=sum seq(yt1,t,1,t,1). (sum and seq are under
1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

LIST, OPS (F5), MORE.) Under WIND use 1,10,1,0,10,1, 3,1,1 to obtain a graph similar
to the one above. Then use TRACE\;(F4) to see the values.

4.
n s
n
1 0.50000
2 1.90000
3 3.60000
4 5.42353
5 7.30814
6 9.22706
7 11.16706
8 13.12091
9 15.08432
10 17.05462

2
 2n  1
The series  2
diverges, since its terms do not approach 0.
n=1
n +1

5.
n s
n
1 1.55741
2  0.62763
3  0.77018
4 0.38764
5  2.99287
6  3.28388
7  2.41243
8  9.21214
9  9.66446
2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

10  9.01610


The series  tan n diverges, since its terms do not approach 0.
n=1

6.
n s
n
1 1.00000
2 1.60000
3 1.96000
4 2.17600
5 2.30560
6 2.38336
7 2.43002
8 2.45801
9 2.47481
10 2.48488

From the graph and the table, it seems that the series converges to 2.5.
 1 1
In fact, it is a geometric series with a=1 and r=0.6 , so its sum is 
n 1
(0.6) = = =2.5 .
n=1 1 0.6 2/5

7.

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

n s
n
1 0.64645
2 0.80755
3 0.87500
4 0.91056
5 0.93196
6 0.94601
7 0.95581
8 0.96296
9 0.96838
10 0.97259

From the graph, it seems that the series converges to 1 . To find the sum, we write
1 1
s = i=1
n

n 1.5 1.5
i ( i+1 )
1 1 1 1 1 1 1 1
= 1 +  +  +   +  =1
1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
2 2 3 3 4 n ( n+1 ) ( n+1 )

So the sum is lim s =1 0=1.

n
n 

8.
n s
n
2 0.50000
3 0.66667
4 0.75000
5 0.80000
6 0.83333

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

7 0.85714
8 0.87500
9 0.88889
10 0.90000
11 0.90909
100 0.99000

From the graph and the table, it seems that the series converges to 1 . To find the sum, we write
s = n 1 1 1
=
n
n  [ partial fractions ]
i=2 i(i 1) i=2 i 1 i
1 1 1 1 1 1 1 1
= 1 +  +  +   +  =1 ,
2 2 3 3 4 n 1 n n
and so the sum is lim s =1 0=1 .
n
n 

2n 2
9. (a) lim a =lim
n 
n
n 
= , so the sequence
3n+1 3 {a }
n
is convergent by ( .1.1).

2 
(b) Since lim a =  0 , the series  a is divergent by the Test for Divergence (7).
n 
n 3 n=1 n

10. (a) Both  a and 

n n
i=1 i
a represent the sum of the first n terms of the sequence
j=1 j {a }n
, that is,
the n th partial sum.
 a =a +a +...+a n terms =na , which, in general, is not the same as 
n n
(b) a =a +a +   +a .
i=1 j j j j j i=1 i 1 2 n

4 8 2
11. 3+2+ + +   is a geometric series with first term a=3 and common ratio r= . Since
3 9 3
2 a 3 3
r = <1 , the series converges to = = =9 .
3 1 r 1 2/3 1/3

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

1 1 1
12.  +  1+   is a geometric series with r= 2 . Since r =2>1 , the series diverges.
8 4 2

5 25 125 5/2 5 5
13.  2+  +     is a geometric series with a= 2 and r= = . Since r = >1 , the
2 8 32  2 4 4
series diverges by (4).

14. 1+0.4+0.16+0.064+   is a geometric series with ratio 0.4 . The series converges to
a 1 5 2
= = since r = <1 .
1 r 1 2/5 3 5

 2 n 1 2 2
15.  5 is a geometric series with a=5 and r= . Since r = <1 , the series converges
n=1 3 3 3
a 5 5
to = = =15 .
1 r 1 2/3 1/3

n 1
 (  6) 6 6
16.  is a geometric series with a=1 and r= . The series diverges since r = >1 .
n=1 n 1 5 5
5

n 1
 (  3) 1  3 n 1 3
17.  =   . The latter series is geometric with a=1 and r= . Since
n=1 n 4 n=1 4 4
4
3 1 4 1 4 1
r = <1 , it converges to = . Thus, the given series converges to = .
4 1 (  3/4 ) 7 4 7 7

 1 1 1
18.  is a geometric series with ratio r= . Since r = <1 , the series converges. Its
n=0 n 2 2
( 2)
1 2 2 2 +1
sum is = =  = 2 ( 2 +1)=2+ 2 .
1 1/ 2 2 1 2 1 2 +1

n
  1   n 
19.  =  is a geometric series with ratio r= . Since r >1 , the series
n=0 n+1 3 n=0 3 3
3
diverges.

n n
 e  e e
20.  =3 is a geometric series with first term 3(e/3)=e and ratio r= . Since
n=1 n 1 n=1 3 3
3
e 3e
r <1 , the series converges. Its sum is = .
1 e/3 3 e
6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

 n n
21.  diverges since lim a =lim =1 0 .
n=1 n+5 n 
n
n  n+5

 1 3 
22.  =3
diverges since each of its partial sums is 3 times the corresponding partial sum
n=1 n=1 n n
 1
of the harmonic series  , which diverges. In general, constant multiples of divergent series are
n=1 n
divergent.

23. Using partial fractions, the partial sums are

s =  n 2 1 1
=
n
n 
i=2 (i 1)(i+1) i=2 i 1 i+1
1 1 1 1 1 1 1 1 1
= 1 +  +  +   +  + 
3 2 4 3 5 n 3 n 1 n 2 n
1 1 1
This sum is a telescoping series and s =1+   .
n 2 n 1 n
 2 1 1 1 3
Thus,  =lim 1+   = .
n=2 2 2 n 1 n 2
n 1 n 

2
 ( n+1 )
24.  diverges by (7), the Test for Divergence, since
n=1 n(n+2)
2
n +2n+1 1
lim a =lim 2
=lim 1+ 2
=1 0 .
n
n  n  n +2n n  n +2n

2 2
 k k
25.  diverges by the Test for Divergence since lim a =lim =1 0 .
k=2 2 k 2
k  1 k  k  k  1

2 1 1
26. Converges. s = =
n n
 (using partial fractions). The latter sum is
n i=1 2 i=1 i+1 i+3
i +4i+3

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
 +  +  +  +   +  +  = +  
2 4 3 5 4 6 5 7 n n+2 n+1 n+3 2 3 n+2 n+3
 2 1 1 1 1 1 1 5
(telescoping series). Thus,  =lim +   = + = .
n=1 2
n  2 3 n+2 n+3 2 3 6
n +4n+3

27. Converges.

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

n n n n n n
 3 +2  3 2  1 1 1/2 1/3 1 3
 n=1 n
=
n=1 n
+
n
=
n=1 2
+
3
= +
1 1/2 1 1/3
=1+ =
2 2
6 6 6

   1 0.3 3 32
28.   (0.3) = 
n 1 n n 1 n
(0.8) (0.8)  (0.3) =  =5 = .
n=1 n=1 n=1 1 0.8 1 0.3 7 7


29. 
n 3 4
2 =2+ 2 + 2 + 2 +   diverges by the Test for Divergence since
n=1
n 1/n 0
lim a =lim 2 =lim 2 =2 =1 0 .
n
n  n  n 

n 1 1
30. lim a =lim ln =lim ln =ln  0 , so the series diverges by the Test for
n 
n
n  2n+5 n  2+5/n 2
Divergence.


31. lim a =lim arctann=  0 , so the series diverges by the Test for Divergence.
n 
n
n  2


32. 
k
(cos 1) is a geometric series with ratio r=cos 1 0.540302 . It converges because r <1 . Its
k=1
cos 1
sum is 1.175343 .
1 cos 1

33. The first series is a telescoping sum:

 3  1 1  1 1 1 1 1 1
 n=1 n(n+3) =  n=1 n  n+3 = n=1
 + 
n n+1 n+1 n+2 n+2 n+3
+ 
 1 1  1 1  1 1
=   +  + 
n=1 n n+1 n=1 n+1 n+2 n=1 n+2 n+3
1 1 11
= 1+ + =
2 3 6
5 1  5 5/4 5
The second series is geometric with first term and ratio : = = . Thus,
4 4 n=1 n 1 1/4 3
4
 3 5  3  5 11 5 7
 n=1 n(n+3)
+
n
=
n=1 n(n+3)
+
n=1 n
= + = .
6 3 2
4 4

 3 2  2  1
34.  + diverges because  =2 diverges. (If it converged, then
n=1 n n n=1 n n=1 n
5

8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

1  1
 2 would also converge by Theorem 8(i), but we know from Example 7 that the harmonic
2 n=1 n
 1
series  diverges.) If the given series converges, then the difference
n=1 n
 2  
 n=1 3n + n   n=1 3n must converge (since  n=1 3n is a convergent geometric series) and
5 5 5
 2  2
equal  , but we have just seen that  diverges, so the given series must also diverge.
n=1 n n=1 n

2 2 2 1
35. 0.2= + +   is a geometric series with a= and r= . It converges to
10 2 10 10
10
a 2/10 2
= = .
1 r 1 1/10 9

2
73 73 73/10 73/100 73
36. 0.73= + +   = = =
2 4 2 99/100 99
10 10 1 1/10

3
417 417 417/10 417 3414 1138
37. 3.417=3+ + +   =3+ =3+ = =
3 6 3 999 999 333
10 10 1 1/10

3
54 54 54/10 62 54 6192 344
38. 6.254=6.2+ + +   =6.2+ = + = =
3 5 2 10 990 990 55
10 10 1 1/10

123 0.000456 123 456 123,333 41,111

39. 0.123456= + = + = =
1000 1 0.001 1000 999,000 999,000 333,000

4
6021 6021 6021/10 6021 56,016 6224
40. 5.6021=5+ + +   =5+ =5+ = =
4 8 4 9999 9999 1111
10 10 1 1/10

n n
 x  x x
41.  = is a geometric series with r= , so the series converges
r <1

n=1 n n=1 3 3
3
x
<1
x <3 ; that is,  3<x<3 . In that case, the sum of the series is
3
a x/3 x/3 3 x
= =  = .
1 r 1 x/3 1 x/3 3 3 x

9
 Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series


42. 
n
n=1
( x 4 ) is a geometric series with r=x 4 , so the series converges
r <1
x 4 <1

x 4 x 4
3<x<5 . In that case, the sum of the series is = .
1 ( x 4 ) 5 x

 
43.  4 x =
n n n
n=0 n=0
( 4x ) is a geometric series with r=4x , so the series converges
r <1
4 x <1
1 1

x < . In that case, the sum of the series is .

4 1 4x

n
 ( x+3) x+3 x+3
44.  is a geometric series with r= , so the series converges
r <1
<1

n=0 n 2 2
2
1 2 2
x+3 <2
 5<x< 1 . For these values of x , the sum of the series is = = .
1 (x+3)/2 2 ( x+3) x+1

n
 cos x cos x
45.  is a geometric series with first term 1 and ratio r= , so it converges
r <1 .
n=0 n 2
2
cos x 1
But r =  for all x . Thus, the series converges for all real values of x and the sum of the
2 2
1 2
series is = .
1 (cos x)/2 2 cos x

1 1
46. Because  0 and ln is continuous, we have lim ln 1+ =ln 1=0 . We now show that the
n n  n
 1  n+1 
series  ln = 1+ ln = [ln (n+1) ln n] diverges.
n=1 n n=1 n n=1
s =(ln 2 ln 1)+(ln 3 ln 2)+   +(ln (n+1) ln n)=ln (n+1) ln 1=ln (n+1). As n  , s =ln (n+1)  ,
n n
so the series diverges.

47. After defining f , We use convert(f,parfrac); in Maple, Apart in Mathematica, or

Expand Rational and Simplify in Derive to find that the general term is
1 1/4 1/4
= + . So the
(4n+1)(4n 3) 4n+1 4n 3
n th partial sum is
s =  n 1/4 1/4 1 n 1 1
n  + =  
k=1 4k+1 4k 3 4 k=1 4k 3 4k+1
1 1 1 1 1 1 1 1 1 1
= 1 +  +  +   +  = 1
4 5 5 9 9 13 4n 3 4n+1 4 4n+1

10
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

1
The series converges to lim s = . This can be confirmed by directly computing the sum using
n 4
n 
sum(f,1..infinity); (in Maple), Sum[f,{n,1,Infinity}] (in Mathematica), or Calculus Sum
(from 1 to  ) and Simplify (in Derive).

2
n +3n+1 1 1 1 1
48. See Exercise 47 for specific CAS commands. = +   . So the n th
n n+1
( n2+n) 2 n
2
( n+1 )
2

partial sum is
1 1 1 1

n
s = +  
n k=1 2 k 2 k+1
k ( k+1 )
1 1 1 1 1 1 1 1 1 1
= 1+1  + +   +   + +  
2 2 2 2 2 3 2 n 2 n+1
2 2 3 n ( n+1 )
1 1
= 1+1 
2 n+1
( n+1 )
The series converges to lim s =2 .
n
n 

49. For n=1 , a =0 since s =0 . For n>1 ,

1 1

n 1 (n 1) 1 (n 1)n (n+1)(n 2) 2

a =s  s =  = =
n n n 1 n+1 ( n 1 ) +1 (n+1)n n(n+1)

 1 1/n
Also,  a =lim s =lim =1 .
n=1 n
n 
n
n  1+1/n

1 5
50. a =s =3 = . For n 1 ,
1 1 2 2

 n  (n 1) n 1 2 2(n 1) n n 2
a =s  s
n n n 1
(
= 3 n2 ) 3 (n 1)2 =
n
n
+  =
n 1 2 n

n
=
n
2 2 2 2 2

 n x 1
Also,  a =lim s =lim 3 =3 because lim =lim x
=0 .
n=1 n n n x
n  n  2 x  2 x  2 ln 2

51. (a) The first step in the chain occurs when the local government spends D dollars. The people
who receive it spend a fraction c of those D dollars, that is, Dc dollars. Those who receive the Dc
11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

2
dollars spend a fraction c of it, that is, Dc dollars. Continuing in this way, we see that the total

spending after n transactions is S =D+Dc+Dc +   +Dc

2 n  1
=
(
D 1 c
n
) by (3).
n 1 c
(b)

lim S
n =lim
D 1 c ( n
)= D
lim ( 1 cn) = 1Dc (since 0<c<1 lim c =0 )
n
n  1 c 1 c n 
n  n 
D
= (since c+s=1 ) =kD (since k=1/s )
s

If c=0.8 , then s=1 c=0.2 and the multiplier is k=1/s=5 .

2
52. (a) Initially, the ball falls a distance H , then rebounds a distance rH , falls rH , rebounds r H ,
2
falls r H , etc. The total distance it travels is

2 3
H+2rH+2r H+2r H+   =H 1+2r+2r +2r +   ( 2 3
)
1 1+r
2
(
=H 1+2r 1+r+r +   =H 1+2r
1 r
) =H
1 r
meters

1 2
(b) From Example 3 in Section 2.1, we know that a ball falls gt meters in t seconds, where g is
2
the gravitational acceleration. Thus, a ball falls h meters in t= 2h/g seconds. The total travel time in
seconds is

2H 2H 2H 2 2H 3 2H 2 3
+2 r +2 r +2 r +   = 1+2 r +2 r +2 r +  
g g g g g
2H 2H 1 2H 1+ r
=
g
( 1+2 r 1+ r + r +  
2
)= g
1+2 r
1 r
=
g 1 r

(c) It will help to make a chart of the time for each descent and each rebound of the ball, together with
the velocity just before and just after each bounce. Recall that the time in seconds needed to fall h
meters is 2h/g . The ball hits the ground with velocity  g 2h/g = 2hg (taking the upward
direction to be positive) and rebounds with velocity kg 2h/g =k 2hg , taking time k 2h/g to reach
2
the top of its bounce, where its velocity is 0 . At that point, its height is k h . All these results follow
from the formulas for vertical motion with gravitational acceleration  g :
12
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

2
d y dy 1 2
= g v= =v  gt  y=y +v t gt .
2 dt 0 0 0 2
dt
number of time of speed before speed after time of peak
descent descent bounce bounce ascent height
2
1 2H/g 2Hg k 2Hg k 2H/g k H
2 2 2 2 4
2 2k H/g 2k Hg k 2k Hg k 2k H/g k H
4 4 4 4 6
3 2k H/g 2k Hg k 2k Hg k 2k H/g k H
                 
otal travel time in seconds is

2H 2H 2H 2H 2H 2H
g
+k
g
+k
g
+k
2
g
+k
2
g
+   =
g
( 1+2k+2k2+2k3+   )
2H 2H 1 2H 1+k
=
g
(
1+2k 1+k+k +  
2
) =
g
1+2k
1 k
=
g 1 k

2
Another method: We could use part (b). At the top of the bounce, the height is k h=rh , so r =k and
the result follows from part (b).

  n  2  1
53.  (1+c) is a geometric series with a= ( 1+c ) and r= ( 1+c ) , so the series converges when
n=2
 1
( 1+c ) <1
1+c >1
1+c>1 or 1+c< 1
c>0 or c< 2 . We calculate
 2 2
( 1+c ) 1 1
the sum of the series and set it equal to 2 : =2
=2 2

 1 1+c 1+c
1 ( 1+c )
2 2  2 12 3  1
1=2(1+c)  2(1+c)
2c +2c 1=0
c= = . However, the negative root is
4 2
 3 1 3 1
inadmissible because  2< <0 . So c= .
2 2

n 1 n
54. The area between y=x and y=x for 0 x 1 is
n n+1 1

1
0
( xn 1 xn) dx = x x

1
= 
1
n n+1 0 n n+1
( n+1 )  n 1
= =
n(n+1) n(n+1)

13
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

We can see from the diagram that as n 

, the sum of the areas between the successive curves
 1
approaches the area of the unit square, that is, 1 . So  =1 .
n=1 n ( n+1 )

55. Let d be the diameter of C . We draw lines from the centers of the C to the center of D (or C ),
n n i
2 1 2 1 2
and using the Pythagorean Theorem, we can write 1 + 1 d = 1+ d

2 1 2 1
1 2 1 2 1
1= 1+ d  1 d =2d (difference of squares)  d= . Similarly,
2 1 2 1 1 1 2
1 2 1 2 2
1= 1+ d  1 d  d =2d +2d  d  d d
2 2 1 2 2 2 1 1 1 2

= (2 d 1)(d 1+d 2)

2 2

d=
1
 d=
( 1 d
1 ) , 1= 1+
1
d
2
 1 d  d 
1
d
2

d=
1 ( ) d +d
1 2
, and in
2 2 d 1 2 d 2 3 1 2 2 3 3 2 ( d +d )
1 1 1 2
2

n
1 d
i=1 i
general, d = . If we actually calculate d and d from the formulas above, we find
n+1 2 3

n
2 d
i=1 i

14
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

1 1 1 1 1
that they are = and = respectively, so we suspect that in general, d = . To
6 2 3 12 3 4 n n(n+1)
1 1 1
prove this, we use induction: Assume that for all k n , d = =  . Then
k k(k+1) k k+1
1 n
 i=1di=1 n+1 = n+1 (telescoping sum). Substituting this into our formula for d n+1 , we get
n

2 1
n
1 2
n+1 ( n+1 ) 1
d = = = , and the induction is complete.
n+1 n n+2 (n+1)(n+2)
2
n+1 n+1
Now, we observe that the partial sums 
n
d of the diameters of the circles approach 1 as n  ;
i=1 i
  1
that is,  a = =1 , which is what we wanted to prove.
n=1 n n=1 n(n+1)
2 3
56. CD =bsin  , DE = CD sin  =bsin 
, EF = DE sin  =bsin  , ... . Therefore,
 sin 
CD + DE + EF + FG +   =b
n
sin  =b since this is a geometric series with
n=1 1 sin 

r=sin  and sin  <1 (because 0< < ).
2
57. The series 1 1+1 1+1 1+   diverges (geometric series with r= 1 ) so we cannot say that
0=1 1+1 1+1 1+   .

 1  1
58. If  a is convergent, then lim a =0 by Theorem 6, so lim  0 , and so  is
n=1 n n a n=1 a
n  n 
n n
divergent by the Test for Divergence.

 
59.   ca =lim c  a =c
n n n
ca =lim a =clim a , which exists by hypothesis.
n=1 n i=1 i i=1 i i=1 i n=1 n
n  n  n 

60. If  ca were convergent, then  (1/c)(ca )= a would be also, by Theorem 8. But this is not the
n n n
case, so  ca must diverge.
n

61. Suppose on the contrary that  ( a +b ) converges. Then  ( a +b ) and  a are convergent
n n n n n
series. So by Theorem 8,  ( a +b )  a would also be convergent. But  ( a +b )  a = b , a
n n n n n n n
contradiction, since  b is given to be divergent.
n

62. No. For example, take

15
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

 a =
n
n and  b = ( n) , which both diverge, yet 
n ( a +b ) =
n n
0 , which converges with sum 0 .

63. The partial sums {s } form an increasing sequence, since s  s =a >0 for all n . Also, the
n n n 1 n
sequence {s } is bounded since s  1000 for all n . So by Theorem .1.11, the sequence of partial
n n
sums converges, that is, the series  a is convergent.
n

64. (a)
f f  f f f  f ( f +f ) f
1 1 n n+1 n n 1 n+1 n 1 n 1 n n 1
RHS =  = = =
f f f f 2 f f f f f f
n 1 n n n+1 f f f n n 1 n+1 n n 1 n+1
n n 1 n+1
1
= = LHS
f f
n 1 n+1

 1  1 1
(b)  n=2
=
n=2

f f f f f f
n 1 n+1 n 1 n n n+1
1 1 1 1 1 1 1 1
=lim  +  +  +   + 
n  f f f f f f f f f f f f f f f f
1 2 2 3 2 3 3 4 3 4 4 5 n 1 n n n+1
1 1 1 1
=lim  =  0= =1 because f   as n  .
n  f f f f f f 1 1 n
1 2 n n+1 1 2


f 
f f
(c)  n=2
n
=
n=2
n

n
(as above)
f f f f f f
n 1 n+1 n 1 n n n+1
 1 1
= 
n=2 f f
n 1 n+1
1 1 1 1 1 1 1 1 1 1
=lim  +  +  +  +   + 
n  f f f f f f f f f f
1 3 2 4 3 5 4 6 n 1 n+1
1 1 1 1
=lim +   =1+1 0 0=2 because f   as n  .
f f f f n
n 
1 2 n n+1

1 2 1
65. (a) At the first step, only the interval , (length ) is removed. At the second step, we
3 3 3
1 2 7 8 1 2
remove the intervals , and , , which have a total length of 2 . At the third
9 9 9 9 3
16
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

2 1 3 n 1
step, we remove 2 intervals, each of length . In general, at the n th step we remove 2
3
1 n n 1 1 n 1 2 n 1
intervals, each of length , for a length of 2  = . Thus, the total length
3 3 3 3
 1 1/3 2 n 1 1 2
of all removed intervals is  =
=1 (geometric series with a= and r= ).
n=1 31 2/3 3 3 3
1 n 2 n
Notice that at the n th step, the leftmost interval that is removed is , , so we
3 3
never remove 0 , and 0 is in the Cantor set. Also, the rightmost interval removed is
2 n 1 n 1 2
1 ,1 , so 1 is never removed. Some other numbers in the Cantor set are ,
3 3 3 3
1 2 7 8
, , , , and .
9 9 9 9
1 1 2
(b) The area removed at the first step is ; at the second step, 8 ; at the third step,
9 9
2 1 3 n 1 1 n 1 8 n 1
( 8)  . In general, the area removed at the n th step is ( 8 ) = , so the
9 9 9 9
 1 8 n 1 1/9
total area of all removed squares is  = =1 .
n=1 9 9 1 8/9

66. (a)
a 1 2 4 1 1 1000
1
a 2 3 1 4 1000 1
2
a 1.5 2.5 2.5 2.5 500.5 500.5
3
a 1.75 2.75 1.75 3.25 750.25 250.75
4
a 1.625 2.625 2.125 2.875 625.375 375.625
5
a 1.6875 2.6875 1.9375 3.0625 687.813 313.188
6
a 1.65625 2.65625 2.03125 2.96875 656.594 344.406
7
a 1.67188 2.67188 1.98438 3.01563 672.203 328.797
8
a 1.66406 2.66406 2.00781 2.99219 664.398 336.602
9
a 1.66797 2.66797 1.99609 3.00391 668.301 332.699
10
a 1.66602 2.66602 2.00195 2.99805 666.350 334.650
11

17
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

a 1.66699 2.66699 1.99902 3.00098 667.325 333.675

12

5 8
The limits seem to be , , 2 , 3 , 667 , and 334 . Note that the limits appear to be "weighted’’
3 3
a +2a
1 2
more toward a . In general, we guess that the limit is .
2 3
(b)
a  a = 1 a +a  a =
1
a a
1 1
 a
n+1 n 2 n n 1 (n 2 n n 1
=
2 ) a +a
2 n 1 n 2 n 1 ( ) ( )
1 1 1 n 1
=  a  a
( ) =   =  (a  a )
2 2 n 1 n 2 2 2 1

1
Note that we have used the formula a =
k 2
a +a
k 1 k ( 2 ) a total of n 1 times in this calculation, once
for each k between 3 and n+1 . Now we can write
a = a + a  a + a  a +   + a  a
( )( ) ( ) +( a  a )
n 1 2 1 3 2 n 1 n 2 n n 1

1 k 1
= a +  a =a +
n 1 n 1
( a )  (a  a )
1 k=1 k+1 k 1 k=1 2 2 1
and so
 1 k 1 1
lim a = a + a  a
n 
n
1 ( 2 1)
 k=1

2
=a + a  a
1 ( 2 1 ) 1 (  1/2 )
a +2a
= a+2 a  a = 1 2
1 3 2 1 ( ) 3

n  1 1 1 2 5 5 3 23
67. (a) For  ,s = = ,s = + = ,s = + = ,
( n+1 )!
n=1 1 1 2 2 2 2 1 2 3 6 3 6 1 2 3 4 24
23 4 119 ( n+1 )! 1
s= + = . The denominators are ( n+1 )! , so a guess would be s = .
4 24 1 2 3 4 5 120 n ( n+1 )!
1 2! 1 (k+1)! 1
(b) For n=1 , s = = , so the formula holds for n=1 . Assume s = . Then
1 2 2! k (k+1)!
s (k+1)! 1 k+1 (k+1)! 1 k+1
k+1 = (k+1)!
+
(k+2)!
=
(k+1)!
+
(k+1)!(k+2)
(k+2)! (k+2)+k+1 (k+2)! 1
= =
(k+2)! (k+2)!
Thus, the formula is true for n=k+1 . So by induction, the guess is correct.
(c)
18
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.2 Series

(n+1)! 1 1  n
lim s =lim =lim 1 =1 and so  =1 .
n 
n
n  (n+1)! n  (n+1)! n=1 (n+1)!

68.

Let r = radius of the large circle, r = radius of next circle, and so on. From the figure we have
1 2
 
 BAC=60 and cos 60 =r / AB , so AB =2r and DB =2r . Therefore, 2r =r +r +2r =r +3r 
1 1 2 1 1 2 2 1 2
1
r =3r . In general, we have r = r , so the total area is
1 2 n+1 3 n
2 2 2
A =  r +3 r +3 r +  
1 2 3
2 2 1 1 1
=  r +3 r
1 2
1+
2
+
4
+
6
+  
3 3 3
2 2 1 2 27 2
=  r +3 r  = r +  r
1 2 1 1/9 1 8 2

r 
1  1 tan 30 1
Since the sides of the triangle have length 1 , BC = and tan 30 = . Thus, r = =
2 1/2 1 2 2 3
1 1 2 27 1 2   11
 r = , so A= + = + = . The area of the triangle is
2 6 3 2 3 8 6 3 12 32 96
3
, so the circles occupy about 83.1% of the area of the triangle.
4

19
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

2 3

1. The picture shows that a =

2
1
1.3
< 1
1.3
dx , a =
3
1
1.3

<
1
1.3
dx , and so on, so
2 1 x 3 2 x




n=2
1
1.3
< 1
1.3
dx . The integral converges by ( .2) with p=1.3>1 , so the series converges.
n 1 x

2. From the first figure, we see that  f (x)dx<

6 5
a . From the second figure, we see that
1 i=1 i

 a < f (x)dx . Thus, we have  a < f (x)dx<

6 6 6 6 5
a .
i=2 i 1 i=2 i 1 i=1 i

4
3. The function f (x)=1/x is continuous, positive, and decreasing on 1, ) , so the Integral Test



t  3 t
1  4 x 1 1 1
applies.
4
dx=lim  x dx=lim
 3 1 t
=lim 
3
+
3
=
3
. Since this improper integral
1 x t  1 t   3t
 1
is convergent, the series  is also convergent by the Integral Test.
n=1 4
n

4  1/4
4. The function f (x)=1/ x =x is continuous, positive, and decreasing on 1, ) , so the Integral
1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

 t t
 1/4  1/4 4 3/4 4 3/4 4 
Test applies.  x dx=lim  x dx=lim
3
x
1 t
=lim
3
t 
3
= , so 
n=1
4
1/ n
1 t  1 t  
diverges.

5. The function f (x)=1/(3x+1) is continuous, positive, and decreasing on 1, ) , so the Integral Test
applies.

 b b
1 1 1
 dx
3x+1=lim  dx
3x+1=lim
3
ln (3x+1)
1 b
=lim
3
ln (3b+1)
3
ln 4 =
1 b  1 b  


so the improper integral diverges, and so does the series  1/(3n+1) .
n=1

 x
6. The function f (x)=e is continuous, positive, and decreasing on 1, ) , so the Integral Test
  x b  x  x b 
applies. 
1
e dx=lim  1
e dx=lim  e
1
=lim (  e b+e 1) =e 1 , so  n=1
 n
e converges. Note:
b  b  b 
 1  1
This is a geometric series, with first term a=e and ratio r=e . Since r <1 , the series converges to
 1  1
e /(1 e )=1/(e 1) .

 x /  x  x  x
7. f (x)=xe is continuous and positive on 1, ) . f (x)= xe +e =e (1 x)<0 for x>1 , so f is
decreasing on 1, ) . Thus, the Integral Test applies.
  x  x  x  x b

b
 1
xe dx = lim 1
xe dx=lim  xe  e
1
(byparts)
b  b 
 b  b  1  1
= lim  be  e +e +e =2/e
b 

 b  b   n
since lim be =lim ( b/eb) =lim ( 1/eb) =0 and lim e =0 . Thus, 
n=1
ne converges.
b  b  b  b 
x+2 1
8. The function f (x)= =1+ is continuous, positive, and decreasing on [1, )] , so the
x+1 x+1
 t
1
 
t
Integral Test applies. f (x)dx=lim 1+ dx=lim [x+ln (x+1)] =lim (t+ln (t+1) 1 ln 2)= ,
1 t  1 x+1 t 
1
t 

x+2  n+2 n+2
so
x+1

dx is divergent and the series 
n=1 n+1
is divergent. NOTE: lim
n+1
=1 , so the given
1 n 
series diverges by the Test for Divergence.

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

 1
9. The series  is a p  series with p=0.85 1 , so it diverges by (1). Therefore, the series
n=1 0.85
n
 2  1
 n=1 0.85
must also diverge, for if it converged, then 
n=1 0.85
would have to converge (by
n n
Theorem 8(i) in Section 11.2).

  1.4   1.2   1.2

10.  n and  n are p  series with p>1 , so they converge by (1). Thus,  3n
n=1 n=1 n=1
converges by Theorem 8(i) in Section 11.2. It follows from Theorem 8(ii) that the given series
  1.4  1.2
 n=1
(n +3n ) also converges.

1 1 1 1  1
11. 1+ + + + + = . This is a p  series with p=3>1 , so it converges by (1).
8 27 64 125 n=1 3
n

1 1 1 1  1  1 3
12. 1+ + + + + = = . This is a p  series with p= >1 ,
2 2 3 3 4 4 5 5 n=1 n n n=1 3/2 2
n
so it converges by (1).

 5 2 n  1  1  1  1
13.  =5  2 by Theorem .2.8, since  and  both
n=1 3 n=1 3 n=1 5/2 n=1 3 n=1 5/2
n n n n n
5  5 2 n
converge by (1) (with p=3>1 and p= >1 ). Thus,  converges.
2 n=1 3
n

5
14. The function f (x)= is continuous, positive, and decreasing on [3, )] , so we can apply the
x 2
 5 t 5
Integral Test.  dx=lim 
t
dx=lim [5ln (x 2)] =lim [5ln (t 2) 0]= , so the series
3 x 2 t 
3 x 2
t 
3
t 
 5
 n=3 n 2
diverges.

1
15. The function f (x)= 2
is continuous, positive, and decreasing on [1, )] , so we can apply the
x +4
Integral Test.
 1 1 1  x t 1  1 t  1 1
 1 2 dx = lim 
t
dx=lim tan
1
= lim tan  tan
x +4 t 
1 2
t  2 2 1 2 t  2 2
x +4

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

1   1 1
=  tan
2 2 2
 1
Therefore, the series  2
converges.
n=1
n +4

3x+2 2 1
16. The function f (x)= = + is continuous, positive, and decreasing on [1, )] since it is
x(x+1) x x+1
the sum of two such functions. Thus, we can apply the Integral Test.
 t
3x+2 2 1
 
t
dx = lim + dx=lim [2ln x+ln (x+1)]
1 x(x+1) t  1 x x+1 t 
1

= tlim [2ln t+ln (t+1) ln 2]=

 3n+2
Thus, the series  diverges.
n=1 n(n+1)

2
x / 1 x
17. f (x)= 2
is continuous and positive on 1, ) , and since f (x)= 2
<0 for x>1 , f is
x +1 ( x +1)
2

also decreasing. Using the Integral Test,

2 t
 x x ln (x +1) 1
 
t 2
dx=lim dx=lim = lim [ln (t +1) ln 2]= , so the series
1 2
t 
1 2
t  2 1 2 t 
x +1 x +1
diverges.
1 1
18. The function f (x)= 2
= 2
is continuous, positive, and decreasing on [2, )] , so the
x  4x+5 (x 2) +1
Integral Test applies.
t
 t
 f (x)dx=lim  f (x)dx=lim  1
2
 1 t
dx=lim [tan (x 2)] =lim [tan (t 2) tan 0]=
2
 1  1 
2
 0=

2
,
2 t  2 t  2 (x 2) +1 t  t 
 1  1
so the series  2
converges. Of course this means that  2
converges too.
n=2 n=1
n  4n+5 n  4n+5
2 2
 x  x
19. f (x)=xe is continuous and positive on 1, ) , and since f (x)=e
/
( 1 2x2) <0 for x>1 , f is
decreasing as well. Thus, we can use the Integral Test.
2 2 t
  x 1  1 

x 1
xe dx=lim  e =0  e =1/(2e) . Since the integral converges, the series
1
t  2 1 2
converges.
4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

ln x / 1 2ln x
20. f (x)= is continuous and positive for x
2 , and f (x)= <0 for x
2 , so f is
2 3
x x


decreasing.  ln x
2
dx=lim
t 

ln x 1
x

x
t

2
=1 . Thus, 

n=1
ln n
2
=

n=2
ln n
2
converges by the
2 x n n
Integral Test.

1 / 1+ln x
21. f (x)= is continuous and positive on 2, ) , and also decreasing since f (x)= <0
xln x 2 2
x (ln x)

1

t
for x>2 , so we can use the Integral Test. dx=lim ln (ln x) =lim ln (ln t) ln (ln 2) = , so
2 xln x t 
2
t 
the series diverges.

x
22. The function f (x)= 4
is positive, continuous, and decreasing on [1, )] . Thus, we can apply
x +1
the Integral Test.
 t 1
 
x (2x) t
2 1  1 2 1  1 2  1
4
dx = lim dx=lim tan (x ) = lim [tan (t ) tan 1]
1 x +1 t t  22  2 1 2 t 
1+(x ) 1
1   
=  =
2 2 4 8
 n
so the series  converges.
n=1 4
n +1

1
23. The function f (x)= 3
is continuous, positive, and decreasing on [1, )] , so the Integral Test
x +x
applies. We use partial fractions to evaluate the integral:
 t


1
1
3
x +x
dx = lim
t 
 1

x
2
dx=lim
t 
ln x
1
2
2
ln (1+x )
t

1
1 x 1+x
x t t 1
= lim ln =lim ln  ln
t  2 t  2 2
1+x 1 1+t

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

1 1 1
= tlim ln
2
+
2
ln 2 =
2
ln 2
1+1/t

 1
so the series  3
converges.
n=1
n +n

1
24. f (x)= is positive and continuous on 3, ) , and is decreasing since x , ln x , and
x ln x ln (ln x)
ln (ln x) are all increasing; so we can apply the Integral Test.

dx  1
 . The integral diverges, so 
t
=lim ln (ln (ln x)) = diverges.
3 xln xln (ln x) t  3 n=3 n ln n ln (ln n)

 1
25. We have already shown (in Exercise 21 ) that when p=1 the series  diverges, so
n=2 p
n(ln n)
1 / p+ln x
assume that p 1 . f (x)=
p
is continuous and positive on 2, ) , and f (x)= 2 p+1
<0
x(ln x) x (ln x)
 p
if x>e , so that f is eventually decreasing and we can use the Integral Test.


1 p t 1 p 1 p
1 ( ln x ) ( ln t ) ( ln 2 )
dx=lim (for p 1)=lim 
p t  1 p 2 t  1 p 1 p
2 x(ln x)

This limit exists whenever 1 p<0 p>1 , so the series converges for p>1 .


 p+1 t
dx ( ln ln x )
26. As in Exercise 24, we can apply the Integral Test. =lim
p t   p+1 3
3 x ln x ( ln ln x )
 p+1
( ln ln t )
(for p 1 ; if p=1 see Exercise 24) and lim exists whenever  p+1<0 p>1 , so the
t   p+1
series converges for p>1 .

1 2 p
27. Clearly the series cannot converge if p
 , because then lim n(1+n )  0 . Also, if p= 1 the
2 n 
1 2 p
series diverges (see Exercise 17). So assume p< , p  1 . Then f (x)=x(1+x ) is continuous,
2

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

positive, and eventually decreasing on 1, ) , and we can use the Integral Test.
 2 p+1 t 2 p+1
 x(1+x ) dx=lim 2
2 p 1 1+x
=lim
1 1+t ( ) ( ) 
2
p
. This limit exists and is finite 
1 t  p+1 1 t  2 p+1 p+1
p+1<0 p< 1 , so the series converges whenever p< 1 .

ln n ln x
28. If p 0 , lim = and the series diverges, so assume p>0 . f (x)= is positive and
p p
n  n x
/ 1/ p
continuous and f (x)<0 for x>e , so f is eventually decreasing and we can use the Integral Test.



1 p t
ln x x ( 1 p) ln x 1
Integration by parts gives dx=lim (for p 1 )
p 2
1 x t  ( 1 p) 1
1 1 p
=
2
lim t ( 1 p) ln t 1 +1 , which exists whenever 1 p<0 p>1 . Since we have already
( 1 p) t 
 ln n
done the case p=1 in Exercise 25 (set p= 1 in that exercise),  converges  p>1 .
n=1 p
n

29. Since this is a p series with p=x ,  (x) is defined when x>1 . Unless specified otherwise, the
domain of a function f is the set of numbers x such that the expression for f (x) makes sense and
defines a real number. So, in the case of a series, it’s the set of numbers x such that the series is
convergent.

4 / 5
30. (a) f (x)=1/x is positive and continuous and f (x)= 4/x is negative for x>0 , and so the Integral
 1 1 1 1 1
Test applies.  s = + + + + 1.082037 .
n=1 4 10 4 4 4 4
n 1 2 3 10


R 
10
 1
4
dx=lim

1
3
t
=lim


1
3
+
1
3
=
1
3000
, so the error is at most 0.0003 .
10 x t  3x 10 t 3t 3 ( 10 )
 

(b) s +
10
 1
4
dx s s +
10
 1
4
dx s +
10
1
3
 s s +
10
1
3

11 x 10 x 3(11) 3(10)
1.082037+0.000250=1.082287 s 1.082037+0.000333=1.082370 , so we get s 1.08233 with error
 0.00005 .


(c) R 
n
 1
4
dx=
1
3
. So R <0.00001
n
1
3
<
1
5

3
3n >10 
5
n>
3 5
(10) /3 32.2 , that is, for
n x 3n 3n 10
7
 Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

n>32 .

1 / 2
31. (a) f (x)= is positive and continuous and f (x)= is negative for x>0 , and so the Integral
2 3
x x
 1 1 1 1 1
Test applies.  s = + + + + 1.549768 .
n=1 2 10 2 2 2 2
n 1 2 3 10
 1  1 t 1 1 1
R 
10
 10 2
dx=lim
x 10 t
=lim  +
t 10
=
10
, so the error is at most 0.1 .
x t  
 1  1 1 1
(b) s + dx s s + dx s +  s s + 
10 11 2 10 10 2 10 11 10 10
x x
1.549768+0.090909=1.640677 s 1.549768+0.1=1.649768 , so we get s 1.64522 (the average of
1.640677 and 1.649768 ) with error  0.005 (the maximum of 1.649768 1.64522 and
1.64522 1.640677 , rounded up).
 1 1 1 1
(c) R   dx= . So R <0.001 if <  n>1000 .
n n 2 n n n 1000
x

5 / 6
32. f (x)=1/x is positive and continuous and f (x)= 5/x is negative for x>0 , and so the Integral
  5  1 t 1
Test applies. Using (3), R   x dx=lim = . If we take n=5 , then s 1.036662 and
n n 4 4 5
t  4x n 4n
R  0.0004 . So s s 1.037 .
5 5

 3/2 3  5/2 /
33. f (x)=x is positive and continuous and f (x)=
x is negative for x>0 , so the Integral Test
2
applies. From the end of Example 6, we see that the error is at most half the length of the interval.


From (3), the interval is s+ f (x)dx,s + f (x)dx , so its length is . Thus, we need n such
n n n
n+1
that

n+1
1  3/2 1  2 n+1 1 1
0.01>
2
 x dx=
2
= 
n x n n n+1

 n>13.08 (use a graphing calculator to solve 1/ x  1/ x+1 <0.01 ). Again from the end of Example
6, we approximate s by the midpoint of this interval. In general, the midpoint is
   
1 1
s+ f (x)dx + s +  f (x)dx =s +  f (x)dx+  f (x)dx . So using n=14 , we
2 n
n+1
n
n
n 2
n+1 n
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

 
1  3/2  3/2 1 1
have s s +
14 2
 x dx+  x dx 2.0872+ + 2.6127 2.61 . Any larger value of
14 15 14 15
1 1
n will also work. For instance, s s + + 2.6124 .
30 30 31

1 / ln x+2
34. f (x)= is positive and continuous and f (x)= is negative for x>1 , so the
2 2 3
x(ln x) x (ln x)

 1 t 1
Integral Test applies. Using (2), we need 0.01> 
2
x(ln x) =lim = . This is true for
n dx t  ln x n ln n
100
n>e , so we would have to take this many terms, which would be problematic because
100 43
e 2.7  10 .

  1.001  1
35.  n = is a convergent p  series with p=1.001>1 . Using (2), we get
n=1 n=1 1.001
n
  0.001 t
 1.001 x 1 t 1 1000
R
n
 x dx=lim
 0.001 n
= 1000lim
0.001
= 1000 
0.001
=
0.001
. We want
n t  t  x n n n
1000  9 0.001 1000
R <0.000000005 <5  10  n > 
n 0.001  9
n 5  10
11 1000
(
n> 2  10 ) =2
1000
 10
11,000
1.07  10
301
 10
11,000
=1.07  10
11,301
.

2
ln x / 2ln x ( 1 ln x )
36. (a) f (x)= is continuous and positive for x>1 , and since f (x)= <0 for
x 3
x
 2
ln x
x>e , we can apply the Integral Test. Using a CAS, we get  x
dx=2 , so the series also
1
converges.
 2 2
ln x ( ln n ) +2ln n+2
(b) Since the Integral Test applies, the error in s s is R 
n n
 x
dx=
n
.
n
2
( ln x ) +2ln x+2
(c) By graphing the functions y = and y =0.05 , we see that y <y for n
1373 .
x 2 1 2 1
(d) Using the CAS to sum the first 1373 terms, we get s 1.94 .
1373

37. (a) From the figure,

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

n n
1 1 1 1 1 1
a +a + +a   f (x)dx , so with f (x)= , + + + +   dx=ln n . Thus,
2 3 n
1 x 2 3 4 n 1 x
1 1 1 1
s =1+ + + + +  1+ln n .
n 2 3 4 n

6 9
(b) By part (a), s 6 1+ln 10 14.82<15 and s 9 1+ln 10 21.72<22 .
10 10

1 1 1
38. (a) The sum of the areas of the n rectangles in the graph to the right is 1+ + + + . Now
2 3 n
n+1
dx
 x
is less than this sum because the rectangles extend above the curve y=1/x , so
1
n+1
1 1 1 1 1 1 1
 x
dx=ln (n+1)<1+ + + + , and since ln n<ln (n+1) , 0<1+ + + +  ln n=t .
2 3 n 2 3 n n
1

n+1
dx
(b) The area under f (x)=1/x between x=n and x=n+1 is
x

=ln (n+1) ln n , and this is clearly
n
1
greater than the area of the inscribed rectangle in the figure to the right which is , so
n+1
1
t  t = ln (n+1) ln n  >0 , and so t >t , so {t } is a decreasing sequence.
n n+1 n+1 n n+1 n

10
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.3 The Integral Test and Estimates of Sums

(c) We have shown that {t } is decreasing and that t >0 for all n . Thus, 0<t  t =1 , so {t } is a
n n n 1 n
bounded monotonic sequence, and hence converges by Theorem .1.11.

39. b
ln n
( ln b) ln n= ( eln n) ln b=nln b=
= e
1
 ln b
. This is a p series, which converges for all b such that
n
 1
 ln b>1 ln b< 1 b<e  b<1/e .

11
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

1. (a) We cannot say anything about  a . If a >b for all n and  b is convergent, then  a could
n n n n n
be convergent or divergent. (See the note after Example 2.)
(b) If a <b for all n , then  a is convergent.
n n n

2. (a) If a >b for all n , then  a is divergent.

n n n
(b) We cannot say anything about  a . If a <b for all n and  b is divergent, then  a could be
n n n n n
convergent or divergent.

1 1  1  1
3. 2
< for all n 1 , so  2
converges by comparison with  , which
2 n=1 n=1 2
n +n+1 n n +n+1 n
converges because it is a p  series with p=2>1 .

2 2  2  2  1
4. 3
< for all n 1 , so  3
converges by comparison with  =2 ,
3 n=1 n=1 3 n=1 3
n +4 n n +4 n n
which converges because it is a constant multiple of a convergent p  series ( p=3>1 ).

5 5  5  5  1
5. < for all n 1 , so  converges by comparison with  =5 ,
n n n=1 n n=1 n n=1 n
2+3 3 2+3 3 3
 1 1
which converges because  is a convergent geometric series with r= ( r <1 ).
n=1 n 3
3

1 1  1
6. for all n 2 , so 
> diverges by comparison with the divergent (partial)
n n n n=2 n n
 1
harmonic series  .
n=2 n

n+1 n 1  n+1
7. > = for all n 1 , so  diverges by comparison with the harmonic series
2 2 n n=1 2
n n n
 1
 n=1 n
.

n n n n
4+3 3 3  4+3
8. > = for all n 1 , so  diverges by comparison with the divergent
n n 2 n=1 n
2 2 2
 3 n
geometric series  .
n=1 2
1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

2 2
cos n 1 1  cos n
9. 2
 2
< , so the series  2
converges by comparison with the p  series
2 n=1
n +1 n +1 n n +1
 1
 n=1 2
( p=2>1) .
n

2 2 2 2
n 1 1 1  n n 1 n  1
10. 4
< 4
< =
4 3 2
.  n=1 4
converges by comparison with  n=1 2
, which
3n +1 3n +1 3n n 3n +1 3n
converges because it is a constant multiple of a convergent p  series ( p=2>1) . The terms of the given
series are positive for n>1 , which is good enough.

2 a 3 2 2
1 n +1 n +n 1+1/n  n +1
=1 , so 
n
11. If a = 3 and b = , then lim =lim 3 =lim diverges by
n n n n=2 3
n 1 n  b n  n  1 n  1 1/n3 n 1
n
 1
the Limit Comparison Test with the divergent (partial) harmonic series  . Or: Since
n=2 n
2 2 2
n +1 n +1 n 1
a= 3
> > = =b , we could use the Comparison Test.
n 3 3 n
n 1 n n n
  1 n
1+sin n 2 2
12.  and  =2 , so the given series converges by comparison
n n n=0 n n=0 10
10 10 10
with a constant multiple of a convergent geometric series.

n 1 n 1 n 1  n 1
13. is positive for n>1 and < = , so  converges by comparison with the
n n n n n=1 n
n4 n4 n4 4 n4
 1 n
convergent geometric series  .
n=1 4

n n 1  n
14. > = , so  diverges by comparison with the divergent (partial) p  series
n 1 n n n=2 n 1
 1 1
 n=2 n
p=  1 .
2

n
2+( 1) 3  3
15.  , and  converges because it is a constant multiple of the convergent p
n n n n n=1 n n
 1 3
 series  p= >1 , so the given series converges by the Comparison Test.
n=1 n n 2
2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

1 1 1  1
16. < = , so  converges by comparison with the convergent p 
3 3 3/2 n=1 3
n +1 n n n +1
 1 3
series  p= >1 .
n=1 3/2 2
n

1 1
17. Use the Limit Comparison Test with a = and b = :
n 2 n n
n +1
a n 1  1
=1>0 . Since the harmonic series 
n
lim =lim =lim diverges, so
n  b n  2 n  2 n=1 n
n n +1 1+(1/n )
 1
does  .
n=1 2
n +1

1 1
18. Use the Limit Comparison Test with a = and b = :
n 2n+3 n n
a n 1 1  1
= >0 . Since the harmonic series 
n
lim =lim =lim diverges, so does
n  b n  2n+3 n  2+(3/n) 2 n=1 n
n
 1
 n=1 2n+3
.

n n n n
2 2 2  2 2
19. < = . is a convergent geometric series ( r = <1 ), so
n n 3 n=1 3 3
1+3 3
n
 2
 n=1 n
converges by the Comparison Test.
1+3

n n a n
1+2 2 n ( 1/2 ) +1
20. Use the Limit Comparison Test with a = and b = : lim =lim n
=1>0 .
n n n n b
1+3 3 n  n  ( 1/3) +1
n
n
 2  1+2
Since  b converges (geometric series with r = <1 ),  also converges.
n=1 n 3 n=1 n
1+3

1 1 a n
n
21. Use the Limit Comparison Test with a = and b = : lim =lim =1>0 .
n 1+ n n n n  b n  1+ n
n

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

 1 1  1
Since  is a divergent p series ( p=  1 ),  also diverges.
n=1 n 2 n=1 1+ n

n+2 1
22. Use the Limit Comparison Test with a = and b = :
n 3 n 2
(n+1) n
2
a 2 1+
n (n+2) n  1
=1>0 . Since 
n
lim =lim =lim is a convergent (partial) p  series
b 3 1 3 n=3 2
n  n  (n+1) n  n
n 1+
n
 n+2
( p=2>1 ) , the series  n=3 3
also converges.
(n+1)

5+2n 1
23. Use the Limit Comparison Test with a = and b = :
n 22 n 3
(1+n ) n
5
a 3 3 4 4 +2
n (5+2n) 5n +2n 1/n n  1
=2>0 . Since 
n
lim =lim =lim =lim is a
b 22 22 22 1 2 n=1 3
n  n  (1+n ) n  (1+n ) 1/(n ) n  n
n +1
2
n
 5+2n
convergent p  series ( p=3>1 ) , the series  also converges.
n=1 22
(1+n )

2a 3 2
1 n  5n
n n  5n 1 5/n
24. If a = 3 and b = , then lim =lim 3 =lim =1>0 , so
n n n b 2 3
n +n+1 n  n  n +n+1 n  1+1/n +1/n
n
2
 n  5n  1
 n=1 3
diverges by the Limit Comparison Test with the divergent harmonic series 
n=1 n
.
n +n+1
(Note that a >0 for n 6 .)
n

2 a 2 3 2
1+n+n 1 n n+n +n 1/n +1/n+1
25. If a = and b = , then lim =lim =lim =1>0 , so
n n n
2 6 n  b n  2 6 n  6 4
1+n +n n 1+n +n 1/n +1/n +1
2
 1+n+n  1
 n=1
diverges by the Limit Comparison Test with the divergent harmonic series 
n=1 n
2 6
1+n +n
.

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

n+5 n n 1
26. If a = and b = = = , then
n 3 7 2 n 3 7 7/3 4/3
n +n n n n
a 7/3 4/3  7/3
n n +5n n 1+5/n
lim = lim =lim
 7/3
n  b
n
n  ( n7+n2) 1/3 n n  ( n7+n2) /n7 1/3
1+5/n 1+0
= lim 5 1/3
=
1/3
=1>0,
n  ( 1+1/n ) ( 1+0 )

 n+5  1
so  converges by the Limit Comparison Test with the convergent p  series 
n=1 3 7 2 n=1 4/3
n +n n
.

1 2  n  n
27. Use the Limit Comparison Test with a = 1+ e and b =e :
n n n
a 1 2   n  1
=1>0 . Since  e =
n
lim =lim 1+ is a convergent geometric series
n  b n  n n=1 n=1 n
n e
1  1 2  n
r = <1 , the series  1+ e also converges.
e n=1 n

2
2n +7n 1
28. Use the Limit Comparison Test with a = and b = .
3 ( n +5n 1 )
n n 2 n n
3
a 2
2n +7n  1
=2>0 , and since 
n
lim =lim b is a convergent geometric series ( r = <1 ),
n  b n 
2 n=1 n 3
n n +5n 1
2
 2n +7n
 n=1 n 2
converges also.
3 (n +5n 1)

1 1  1
.
n 1
29. Clearly n!=n(n 1)(n 2) (3)(2) 2 2 2 2 2=2 , so  is a
n! n 1 n=1 n 1
2 2
1  1
convergent geometric series ( r = <1 ), so  converges by the Comparison Test.
2 n=1 n!

n! 1 2 3 (n 1)n 1 2  2
30. =  1 1 1 for n 2 , so since  converges ( p=2>1 ),
n n n n n n n n n=1 2
n n
5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

 n!
 n=1 n
converges also by the Comparison Test.
n

1 1
31. Use the Limit Comparison Test with a =sin and b = . Then  a and  b are series
n n n n n n
a sin ( 1/n ) sin  
=1>0 . Since 
n
with positive terms and lim =lim =lim b is the divergent
1/n   0 
b n=1 n
n  n 
n

harmonic series,  sin ( 1/n ) also diverges. (Note that we could also use l’Hospital’s Rule to
n=1

evaluate the limit: lim

sin ( 1/x )
=lim
cos ( 1/x )  1/x ( 2
) =lim cos 1 =cos 0=1 .)
x  1/x x  2  x
 1/x x

1 a
n n 1 1
32. Use the Limit Comparison Test with a = and b = . lim =lim =lim =1
n 1+1/n n n b 1+1/n 1/n
n n  n  n n  n
n
 1  1
(since lim x =1 by l’Hospital’s Rule), so  diverges (harmonic series)

1/x
n=1 n n=1 1+1/n
x  n
diverges.

1 1 1 1 1 1 1
33. 
10
= + + + +  0.567975 . Now < , so using the reasoning
n=1 4 2 2 20 90 10,100 4 2 4
n +n n +n n
and notation of Example 5, the error is
  3 t
 1 x 1
R  T = 
4
 x =lim  = =0.0003 .
10 10 n=11 4
10 dx t  3 10 3000
n

1+cos n 1+cos 2 1+cos 3 1+cos 10 1+cos n 2

34. 
10
=1+cos 1+ + + +  1.55972 . Now  ,
n=1 5 32 243 100,000 5 5
n n n


so as in Example 5, R  T 
10 10
 2
5
dx=2lim
t 

1 
4
x
4 t

10
=0.00005 .
10 x

1 1 1 1 1 1 1
35. 
10
= + + + +  0.76352 . Now < , so the error is
n=1 n 3 5 9 1025 n n
1+2 1+2 2

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

11
 1/2
1
R  T = = (geometric series)  0.00098 .
10 10 n=11 n 1 1/2
2

n 1 2 3 10 n n 1
36. 
10
= + + + +  0.283597 . Now < = , so the
n=1 n 6 27 108 649,539 n n n
(n+1)3 (n+1)3 n 3 3
11
 1 1/3
error is R  T = =  0.0000085 .
10 10 n=11 n 1 1/3
3

d  1
9 9
for each n , and since 
n
37. Since  is a convergent geometric series ( r = <1
n n n=1 n 10
10 10 10

d
), 0.d d d ... =
n
will always converge by the Comparison Test.
1 2 3 n=1 n
10

1
38. Clearly, if p<0 then the series diverges, since lim p
= . If 0 p 1 , then
n  n ln n
1 1  1  1
and  diverges (Exercise .3.21), so 
p
n ln n  n ln n

p n ln n n =2 n ln n n =2 p
n ln n n ln n
1 1 
diverges. If p>1 , use the Limit Comparison Test with a = p
and b = . b converges,
n n p n =2 n
n ln n n
a 1  1
=0 , so 
n
and lim = lim also converges. (Or use the Comparison Test, since
n  b n  ln n n =2 p
n n ln n
p p
n ln n >n for n >e .) In summary, the series converges if and only if p>1 .

39. Since  a converges, lim a =0 , so there exists N such that a  0 <1 for all n>N
0 a <1 for
n n n n
n 

0 a  a . Since  a converges, so does 

2 2
all n>N
a by the Comparison Test.
n n n n

40. (a) Since lim ( a /b ) =0 , there is a number N>0 such that a /b  0 <1 for all n>N , and so
n n n n
n 
a <b since a and b are positive. Thus, since  b converges, so does  a by the Comparison Test.
n n n n n n
(b)

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

(i) a
ln n 1 ln n ln x 1/x  ln n
=0 , so 
n
If a = and b = , then lim =lim =lim =lim
n 3 n
n 2  b n  n x  x x  1 n=1 3
n n n n
converges by part (a).
(ii) ln n 1
If a = and b = , then
n n n n
ne e

a ln n ln x 1/x 2
n
lim =lim =lim =lim =lim =0
n  b n  n x  x x  1/(2 x ) x  x
n

Now  b is a convergent geometric series with ratio r=1/e( r <1) , so  a converges by part
n n
(a).

a a
n n
41. (a) Since lim = , there is an integer N such that >1 whenever n>N . (Take M=1 in
n  b b
n n
Definition .1.5.) Then a >b whenever n>N and since  b is divergent,  a is also divergent by the
n n n n
Comparison Test.
(b)
1 1 a n x 1
n
(c) If a = and b = for n 2 , then lim =lim =lim =lim =lim x= , so
n ln n n n
n  b n  ln n x  ln x x  1/x x 
n
1 
by part (a),  is divergent.
n=2 ln n
ln n 1 
(d) If a = and b = , then  b is the divergent harmonic series and
n n n n n=1 n
a 
=lim ln n=lim ln x= , so 
n
lim a diverges by part (a).
n=1 n
n  b n  x 
n

1 a 1
1
=0 , but  b diverges while 
n
42. Let a = and b = . Then lim =lim a converges.
n n
n  n
n 2 n n
n n  b
n

a
n 1
43. lim na =lim , so we apply the Limit Comparison Test with b =
. Since lim na >0 we
n  1/n
n n n n
n  n 
know that either both series converge or both series diverge, and we also know that
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.4 The Comparison Tests

 1
 n=0 n
diverges ( p series with p=1 ). Therefore,  a must be divergent.
n

ln (1+x) 1
44. First we observe that, by l’Hospital’s Rule, lim =lim =1 . Also, if  a converges,
x 0 x x 0 1+x
n

then lim a =0 by Theorem .2.6. Therefore, lim

(
ln 1+a
n ) =lim ln (1+x)
=1>0 . We are given that
n 
n
n  a x 0 x
n
 a is convergent and a >0 . Thus, 
n n
ln 1+a ( n ) is convergent by the Limit Comparison Test.
45. Yes. Since  a is a convergent series with positive terms, lim a =0 by Theorem .2.6, and
n n
n 
 b =
n ( ) is a series with positive terms (for large enough n ). We have
sin a
n
b sin ( a )
=1>0 by Theorem 3. .2. Thus,  b is also convergent by the Limit
n n
lim =lim
a a n
n  n 
n n
Comparison Test.

46. Yes. Since  a converges, its terms approach 0 as n  , so for some integer N , a  1 for all
n n
  
n N . But then  a b = a b +  a b +
N 1 N 1
ab b . The first term is a finite sum,
n=1 n n n=1 n n n=N n n n=1 n n n=N n

and the second term converges since  b converges. So  a b converges by the Comparison
n=1 n n n
Test.

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

1. (a) An alternating series is a series whose terms are alternately positive and negative.

(b) An alternating series 
n 1
( 1) b converges if 0<b  b for all n and lim b =0 . (This is the
n=1 n n+1 n n
n 
Alternating Series Test.)
(c) The error involved in using the partial sum s as an approximation to the total sum s is the
n
remainder R =s s and the size of the error is smaller than b ; that is, R  b . (This is the
n n n+1 n n+1
Alternating Series Estimation Theorem.)

1 2 3 4 5  n n n n
2.  +  +  +   = ( 1) . Here a =( 1) . Since lim a 0 (in fact the limit
3 4 5 6 7 n=1 n+2 n n+2 n 
n
does not exist), the series diverges by the Test for Divergence.

4 4 4 4 4  n 1 4 4
3.  +  +     = ( 1) . Now b = >0 , {b } is decreasing, and
7 8 9 10 11 n=1 n+6 n n+6 n
lim b =0 , so the series converges by the Alternating Series Test.
n
n 

1 1 1
4. 
n
(  1)
.b = is positive and {b } is decreasing; lim =0 , so the series
ln n
n=2 n ln n n
n  ln n
converges by the Alternating Series Test.

n 1
1  ( 1)
5. b = >0 , b is decreasing, and lim b =0 , so the series 
{ } converges by the
n n n n n=1 n
n 
Alternating Series Test.
n 1
1  (  1)
6. b = >0 , b is decreasing, and lim b =0 , so the series 
{ } converges by the
n 3n 1 n n n=1 3n 1
n 
Alternating Series Test.

 3n 1  3 1/n 3
7.  a = =
n n
( 1)
( 1) b . Now lim b =lim = 0 . Since lim a 0
n=1 n n=1 2n+1 n=1 n
n 
n
n  2+1/n 2 n 
n
(in fact the limit does not exist), the series diverges by the Test for Divergence.

2n 2/n  2n
=0 , so the series 
n
8. b =
4n +1
n 2
>0 , { } b
n 
n
is decreasing , and lim b =lim
n  4+1/n
n 2 n=1
( 1)
4n +1
2

converges by the Alternating Series Test. Alternatively, to show that {b } n

is decreasing, we could
d 2x
verify that <0 for x
1 .
dx 2
4x +1

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

n+1
1  (  1)
9. b = 2
>0 , {b } is decreasing, and lim b =0 , so the series  2
converges by the
n n n n=1
4n +1 n  4n +1
Alternating Series Test.

  1 n
1 
10.  a = =
n n
( 1) = 0 . Since lim a 0 ( 1) b . Now lim b =lim
n=1 n n=1 1+2 n n=1 n
n 
n
n  2+1/ n 2 n 
n
(in fact the limit does not exist), the series diverges by the Test for Divergence.

2
n
11. b =
n
n +4
3
>0 for n
1 . {b } n
is decreasing for n
2 since

/
x
2
=
( x +4) ( 2x )  x ( (3x ) x ( 2x +8 3x ) x ( 8 x )
3 2 2
=
3 3
=
3
<0 for x>2 . Also,
3 2 2 2
x +4 ( x +4)
3
( x +4)
3
( x +4)
3

2
1/n  n
=0 . Thus, the series 
n+1
lim b =lim ( 1) 3
converges by the Alternating Series
n 3 n=1
n  n  1+4/n n +4
Test.

1/n
e
12. b =
n n
>0 for n
1 . {b } n
is decreasing since
/
1/x
e
=
x e
1/x
(  1/x2)  e1/x 1 =  e1/x(1+x) <0 for x>0 . Also, lim b =0 since lim e =1 . Thus,
1/n
x 2 3 n 
n
n 
x x
1/n
 e
the series 
n 1
( 1) converges by the Alternating Series Test.
n=1 n
 n n x 1
13. 
n
( 1) . lim =lim =lim = , so the series diverges by the Test for
n=2 ln n n  ln n x  ln x x  1/x
Divergence.
 ln n  ln n ln n ln x
14.  =0+
n 1 n 1
( 1) ( 1) .b = >0 for n
2 , and if f (x)= , then
n=1 n n=2 n n n x
/ 1 ln x
f (x)=
x
2
<0 for x>e , so {b } n
is eventually decreasing. Also,

ln n ln x 1/x
lim b =lim =lim =lim =0 , so the series converges by the Alternating Series Test.
n 
n
n  n x  x x  1

15.

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

n
 cos n  (  1) 1 1
 n=1 3/4
=
n=1 3/4
.b =
n 3/4
is decreasing and positive and lim
3/4
=0 , so the series
n n n n  n
converges by the Alternating Series Test.

n
n  (  1) 1
=0 if n is even and (  1 ) if n=2k+1 , so the series is 
k
16. sin .b = >0
2 n=0 ( 2n+1 )! n ( 2n+1 )!
1
, { } b
n
is decreasing, and lim
n  ( 2n+1 )!
=0 , so the series converges by the Alternating Series Test.

     
17. 
n
( 1) sin . b =sin >0 for n
2 and sin
sin , and lim sin =sin 0=0 , so
n n=1 n n n n+1 n  n
the series converges by the Alternating Series Test.

   
18. 
n n
( 1) cos . lim cos =cos (0)=1 , so lim ( 1) cos does not exist and
n
n=1
n  n n  n
the series diverges by the Test for Divergence.

n n n n
n n n     n n (  1) n
19. =
n lim =  lim does not exist. So the series diverges by the
n! 1 2     n n  n! n  n!
Test for Divergence.

n  n n n n n
20.   diverges by the Test for Divergence since lim =  lim 
n=1 5 n  5 n  5
does not exist.

21.
n a s
n n
1 1 1
2  0.35355 0.64645
3 0.19245 0.83890
4  0.125 0.71390
5 0.08944 0.80334
6  0.06804 0.73530
7 0.05399 0.78929
8  0.04419 0.74510
9 0.03704 0.78214
10  0.03162 0.75051
3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

By the Alternating Series Estimation Theorem, the error in the approximation

n 1
 (  1)

3/2
 0.75051 is s s  b =1/(11)  0.0275 (to four decimal places, rounded up).
n=1 3/2 10 11
n

22.
n a s
n n
1 1 1
2  0.125 0.875
3 0.03704 0.91204
4  0.01563 0.89641
5 0.008 0.90441
6  0.00463 0.89978
7 0.00292 0.90270
8  0.00195 0.90074
9 0.00137 0.90212
10  0.001 0.90112

By the Alternating Series Estimation Theorem, the error in the approximation

n 1
 (  1)

3
 0.90112 is s s  b =1/11  0.0007513 .
n=1 3 10 11
n
 1 1 1
23. The series 
n 1
( 1) satisfies (i) of the Alternating Series Test because < and (ii)
n=1 2 2 2
n (n+1) n
1 1 1 1
lim =0 , so the series is convergent. Now b = =0.01 and b = =  0.008<0.01 , so
n  2 10 2 11 2 121
n 10 11
by the Alternating Series Estimation Theorem, n=10 . (That is, since the 11 th term is less than the
4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

desired error, we need to add the first 10 terms to get the sum to the desired accuracy.)

 1 1 1
24. The series 
n+1
( 1) satisfies (i) of the Alternating Series Test because < and
n=1 4 4 4
n (n+1) n
1 4
(ii) lim =0 , so the series is convergent. Now b =1/5 =0.0016>0.001 and
4 5
n  n
4
b =1/6  0.00077<0.001 , so by the Alternating Series Estimation Theorem, n=5 .
6

n n
 ( 2)  2
25. The series  =
n
( 1) satisfies (i) of the Alternating Series Test because
n=1 n! n=1 n!
n+1 n n n
2 2 2 2 2 2 2 2 2 2 2
b = = =  =  b  b and (ii) lim =        =0 , so the
n+1 (n+1)! (n+1)n! n+1 n! n+1 n n n  n! n n 1 2 1
7 8
series is convergent. Now b =2 /7! 0.025>0.01 and b =2 /8! 0.006<0.01 , so by the Alternating
7 8
Series Estimation Theorem, n=7 . (That is, since the 8 th term is less than the desired error, we need
to add the first 7 terms to get the sum to the desired accuracy.)
n
 ( 1) n  n
26. The series  =
n
( 1) satisfies (i) of the Alternating Series Test because
n=1 n n=1 n
4 4
n+1 n+3n 4n n n
b = < = = =b and (ii) lim =0 , so the series is convergent. Now
n+1 n+1 n 1 n n n n
4 4 4 4 4 4 n  4
5 6
b =5/4  0.0049>0.002 and b =6/4  0.0015<0.002 , so by the Alternating Series Estimation
5 6
Theorem, n=5 .

1 1
27. b = =  0.0000595 , so
7 5 16,807
7
n+1 n+1
 (  1) (  1) 1 1 1 1 1
  s =
6
=1 +  +   0.972080 . Adding b to s
n=1 5 6 n=1 5 32 243 1024 3125 7776 7 6
n n
does not change the fourth decimal place of s , so the sum of the series, correct to four decimal
6
places, is 0.9721 .

6 6
28. b = =  0.000023 , so
6 6 262,144
8

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

n n
 ( 1) n (  1) n 1 2 3 4 5
  s =
5
= +  +    0.098785 . Adding b to s does
n=1 n 5 n=1 n 8 64 512 4096 32,768 6 5
8 8
not change the fourth decimal place of s , so the sum of the series, correct to four decimal places, is
5
 0.0988 .

2
7
29. b = =0.0000049 , so
7 7
10
n 1 2 n 1 2
 ( 1) n ( 1) n 1 4 9 16 25 36
  s =
6
=  +  +  =0.067614 .
n=1 n 6 n=1 n 10 100 1000 10,000 100,000 1,000,000
10 10
Adding b to s does not change the fourth decimal place of s , so the sum of the series, correct to
7 6 6
four decimal places, is 0.0676 .

1 1
30. b = =  0.0000019 , so
6 6 524,880
3  6!
n n
 ( 1) ( 1) 1 1 1 1 1
  s =
5
= +  +    0.283471 . Adding b to s does
n=1 n 5 n=1 n 3 18 162 1944 29,160 6 5
3 n! 3 n!
not change the fourth decimal place of s , so the sum of the series, correct to four decimal places, is
5
 0.2835 .

n 1
 ( 1) 1 1 1 1 1 1 1
31.  =1 +  +   +  +  +   . The 50 th partial sum of this series
n=1 n 2 3 4 49 50 51 52
n 1
( 1) 1 1 1  1
is an underestimate, since  =s +  +  +   , and the terms in
n=1 n 50 51 52 53 54
parentheses are all positive. The result can be seen geometrically in Figure 1.

32. If p>0 ,
1
p

1
p
( { 1/n } )
p
is decreasing) and lim
1
p
=0 , so the series converges by
( n+1 ) n n  n
n 1
(  1)
the Alternating Series Test. If p 0 , lim does not exist, so the series diverges by the Test
p
n  n
n 1
 (  1)
for Divergence. Thus,  converges p>0 .
n=1 p
n

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.5 Alternating Series

1
33. Clearly b = is decreasing and eventually positive and lim b =0 for any p . So the series
n n+ p n 
n

converges (by the Alternating Series Test) for any p for which every b is defined, that is, n+ p 0 for
n
n
1 , or p is not a negative integer.

p p 1
( ln x ) / ( ln x ) ( p ln x ) p
34. Let f ( x ) = . Then f ( x ) = <0 if x>e so f is eventually decreasing for
x 2
x
p
( ln n )
every p . Clearly lim =0 if p 0 , and if p>0 we can apply l’Hospital’s Rule [ p+1] times to
n n 
get a limit of 0 as well. So the series converges for all p (by the Alternating Series Test).

35.  b = 1/ ( 2n ) clearly converges (by comparison with the p series for p=2 ). So suppose that
2
2n

 (  1)
n 1
b converges. Then by Theorem .2.8(ii), so does
n
1 1 1
 =2
n 1
(  1) b +b =2 1+ + +   . But this diverges by comparison with the
n n 3 5 2n 1
harmonic series, a contradiction. Therefore, 
n 1
(  1) b must diverge. The Alternating Series Test
n
does not apply since {b } n
is not decreasing.

36. (a) We will prove this by induction. Let P(n) be the proposition that s =h  h . P(1) is the
2n 2n n
1 1
statement s =h  h , which is true since 1 = 1+  1 . So suppose that P(n) is true. We will
2 2 1 2 2
show that P(n+1) must be true as a consequence.
 h 1 1 1 1 1
h
2n+2 n+1 = h + +  h+ = h  h +  ( )
2n 2n+1 2n+2 n n+1 2n n 2n+1 2n+2
1 1
= s +  =s
2n 2n+1 2n+2 2n+2

which is P(n+1) , and proves that s =h  h for all n .

2n 2n n
(b) We know that h  ln (2n)  and h  ln n  as n  . So
2n n
s =h  h =[h  ln (2n)] ( h  ln n) +[ln (2n) ln n] , and
2n 2n n 2n n
lim s =   +lim [ln (2n) ln n]=lim ( ln 2+ln n ln n ) =ln 2 .
2n
n  n  n 

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

a
=8>1 , part (b) of the Ratio Test tells us that the series 
n+1
1. (a) Since lim a is divergent.
a n
n 
n
a
=0.8<1 , part (a) of the Ratio Test tells us that the series 
n+1
(b) Since lim a is absolutely
a n
n 
n
convergent (and therefore convergent).
a
=1 , the Ratio Test fails and the series 
n+1
(c) Since lim a might converge or it might
a n
n 
n
diverge.
2
 n
2. The series  has positive terms and
n=1 n
2
a 2 n 2
n+1 ( n+1 ) 2 1 1 1
lim =lim  =lim 1+  = <1 , so the series is absolutely
n  a n  n+1 2 n  n 2 2
n2 n
convergent by the Ratio Test.

n

( 10)
3.  . Using the Ratio Test,
n=0 n!
a n+1
n+1 ( 10) n!  10
lim =lim  =lim =0<1 , so the series is absolutely
n  a n  (n+1)! n n  n+1
n ( 10)
convergent.

n n n
 2 2 2
4. 
n 1 n 1
( 1) diverges by the Test for Divergence. lim = , so lim ( 1) does not
n=1 4 4 4
n n  n n  n
exist.

n+1
 ( 1)  1
5.  4
converges by the Alternating Series Test, but  is a divergent p  series
n=1 n=1 [4]n
n
1
p=  1 , so the given series is conditionally convergent.
4

n
 1  ( 1)
6.  is a convergent p  series ( p=4>1 ) , so  is absolutely convergent.
n=1 4 n=1 4
n n

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

n 1
7. lim =lim =1 , so lim a 0 . Thus, the given series is divergent by the
a =lim
n  5+n n  5/n+1
n n
n  n 
Test for Divergence.

 n
8.  2
diverges by the Limit Comparison Test with the harmonic series:
n=1
n +1
2 2
n/(n +1) n  n
=1 . But 
n 1
lim =lim (  1) converges by the Alternating Series Test:
n  1/n n 
2 n=1 2
n +1 n +1
/ 2

{ } n +1
n
2
has positive terms, is decreasing since 2
x
x +1
=
1 x
( x +1)
2 2
 0 for x
1 , and

n  n
=0 . Thus, 
n 1
lim 2
( 1) 2
is conditionally convergent.
n=1
n  n +1 n +1

9.
a 1/(2n+2)! ( 2n )! ( 2n )! 1
n+1
lim =lim =lim =lim =lim =0<1
n  a n  1/(2n)! n  ( 2n+2 )! n  (2n+2)(2n+1)(2n)! n  (2n+2)(2n+1)
n
1 
, so the series  is absolutely convergent by the Ratio Test. Of course, absolute
n=1 (2n)!
convergence is the same as convergence for this series, since all of its terms are positive.

a n+1
( n+1 )!/e 1   n
, so the series 
n+1
10. lim =lim = lim ( n+1 ) = e n! diverges by
n  a n  n e n  n=1
n n!/e
the Ratio Test.

1/n 1/n
e e 1  1  e
11. Since 0  =e and  is a convergent p  series ( p=3>1 ) , 
3 3 3 n=1 3 n=1 3
n n n n n
n 1/n
 ( 1) e
converges, and so  is absolutely convergent.
n=1 3
n

sin 4n 1  sin 4n
12.  , so  converges by comparison with the convergent geometric
n n n=1 n
4 4 4
series

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

 1 1  sin 4n
 n=1 n
r =
4
<1 . Thus, 
n=1 n
is absolutely convergent.
4 4

a n+1 n 1
n+1 ( n+1 ) 3 4 3 n+1 3
13. lim =lim  =lim  = <1 , so the series
n  a n  n n n  4 n 4
n 4 n 3
n
 n( 3)
 n=1 n 1
is absolutely convergent by the Ratio Test.
4

a 2 n+1 2
n+1 ( n+1 ) 2 n! 1 2
14. lim =lim  =lim 1+  =0 , so the series
n  a n  ( n+1 )! 2 n
n2 n  n n+1
n
2 n
 n2

n+1
(  1) is absolutely convergent by the Ratio Test.
n=1 n!

a n+1 2n+1
n+1 10 ( n+1 ) 4 10 n+1 5
15. lim =lim  =lim  = <1 , so the series
a 2n+3 n 2 n+2 8
n  n  ( n+2 ) 4 10 n  4
n
n
 10
 n=1 2n+1
is absolutely convergent by the Ratio Test. Since the terms of this series are
(n+1)4
positive, absolute convergence is the same as convergence.

3 cos n 1 1  1 2
for n
3 . Since 
2/3
16. n  2>0 for n
3 , so > > diverges p=  1
2/3 2/3 2/3 n=1 2/3 3
n  2 n  2 n n
 3 cos n
, so does  2/3
by the Comparison Test.
n=1
n  2

n
17. 
n=2 ln n

( 1)
converges by the Alternating Series Test since lim
1
n  ln n
=0 and
1
ln n
is { }
1 1  1
decreasing. Now ln n<n , so > , and since  is the divergent (partial) harmonic series,
ln n n n=2 n
n
 1  ( 1)
 n=2 ln n
diverges by the Comparison Test. Thus, 
n=2 ln n
is conditionally convergent.

18.

3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

a n+1 n
n+1 ( n+1 )!/ ( n+1 ) n 1 1
lim =lim =lim =lim = <1 , so the series
n  a n  n
n  n
n 
n e
n n!/n ( n+1 ) ( 1+1/n )
 n!
 n=1 n
converges absolutely by the Ratio Test.
n

cos ( n /3) 1  1
19.  and  converges (use the Ratio Test or the result of Exercise .4.29),
n! n! n=1 n!
 cos (n /3)
so the series  converges absolutely by the Comparison Test.
n=1 n!

n
1  ( 1)
20. lim n
a =lim =0<1 , so the series  converges absolutely by the Root Test.
n  ln n
n n=2 n
n  (ln n)

n 1/n n
n n  n
21. lim n
a =lim =lim = , so the series  is divergent by
n 1+3n n 3 n=1 1+3n
n  n  3 n  3 3 3
the Root Test.
Or:
a n+1 1+3n n
n+1 ( n+1 ) 3 1 n+1
lim =lim  =lim  ( n+1 )
n  a n  4+3n n n  3 n
n 3 n 3
1 1 n 1
= lim 1+ lim ( n+1 ) = elim ( n+1 ) = ,
27 n  n n  27 n 

so the series is divergent by the Ratio Test.

n
22. Since { }
is decreasing and lim
11
nln n
n  nln n
=0 , the series 
 ( 1)
n=2 nln n
converges by the
 1
Alternating Series Test. Since  diverges by the Integral Test (Exercise .3.21), the series
n=2 nln n
n
 ( 1)
 n=2 nln n
is conditionally convergent.

2 2 2 n
n +1
1 1+1/n  n +1
23. lim n
a =lim =lim = <1 , so the series  is absolutely
n  2n +1 n  2+1/n2 2
n 2 n=1 2
n  2n +1

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

convergent by the Root Test.

n
1 1 2  ( 1)
24. lim n
a =lim = = <1 , so the series  is absolutely
n  arctan n  /2 
n n=1 n
n  (arctan n)
convergent by the Root Test.

25. Use the Ratio Test with the series

1 3 1 3 5 1 3 5 7 1 3 5     (2n 1)  1 3 5     (2n 1)
+   =
n 1 n 1
1 +  +   +( 1) ( 1)
3! 5! 7! (2n 1)! n=1 (2n 1)!
.
a n
n+1 ( 1)  1 3 5     (2n 1)[2(n+1) 1] (2n 1)!
lim = lim 
 a [2(n+1) 1]! n 1
n
n n  ( 1)  1 3 5     (2n 1)
( 1)(2n+1)(2n 1)!
= lim
n  (2n+1)(2n)(2n 1)!
1
= lim 2n =0<1,
n 
so the given series is absolutely convergent and therefore convergent.

26. Use the Ratio Test with the series

2 2 6 2 6 10 2 6 10 14  2 6 10 14     (4n 2)
+ + + +   = .
5 5 8 5 8 11 5 8 11 14 n=1 5 8 11 14     (3n+2)
a 2 6 10     (4n 2)[4(n+1) 2] 5 8 11     (3n+2)
n+1
lim = lim 
n  a n  5 8 11     (3n+2)[3(n+1)+2] 2 6 10     (4n 2)
n
4n+2 4
= lim 3n+5 = 3 >1,
n 
so the given series is divergent.

n
 2 4 6     ( 2n )  ( 2 1 )  ( 2 2 )  ( 2 3)      ( 2 n )  2 n! 
27.  = = =
n
2 , which
n=1 n! n=1 n! n=1 n! n=1
n
diverges by the Test for Divergence since lim 2 = .
n 

28.

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

n+1
2 ( n+1 )!
a 5 8 11     ( 3n+5) 2(n+1) 2
n+1
lim =lim =lim = <1 , so the series converges
n  a n 
n
n  3n+5 3
n 2 n!
5 8 11     ( 3n+2 )
absolutely by the Ratio Test.

a 5n+1 5
n+1
29. By the recursive definition, lim =lim = >1 , so the series diverges by
n  a n  4n+3 4
n
the Ratio Test.

a 2+cos n
n+1
30. By the recursive definition, lim =lim =0<1 , so the series converges
n  a n  n
n
absolutely by the Ratio Test.

3 3
1/(n+1) n 1
31. (a) lim =lim =lim =1 . Inconclusive.
3 3 3
n  1/n n  ( n+1 ) n  ( 1+1/n )
n
( n+1 ) 2 n+1 1 1 1
(b) lim  =lim =lim + = . Conclusive (convergent).
n  n+1 n n  2n n  2 2n 2
2
n
( 3) n n 1
(c) lim  =3lim =3lim =3 . Conclusive (divergent).
n  n+1 ( 3)n 1 n  n+1 n  1+1/n
2 2
n+1 1+n 1 1/n +1
(d) lim  =lim 1+  =1 . Inconclusive.
n 
2 n n  n 2 2
1+ ( n+1 ) 1/n + ( 1+1/n )

32. We use the Ratio Test:

a 2 2
n+1 (n+1)! (kn)! ( n+1 )
lim =lim  =lim Now if
n  a n  k(n+1) ! 2
n  k(n+1) k(n+1) 1    kn+1
n (n!)
2
( n+1 )
k=1 , then this is equal to lim = , so the series diverges; if k=2 , the limit is
n  ( n+1 )
2
1 ( n+1 )
lim =
<1 , so the series converges, and if k>2 , then the highest power of n in
n  (2n+2)(2n+1) 4
the denominator is larger than 2 , and so the limit is 0 , indicating convergence. So the series
converges for k
2 .
6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

a n+1
n+1 x n! x 1
33. (a) lim =lim  =lim = x lim = x  0=0<1 , so by
n  a n  (n+1)! n n  n+1 n  n+1
n x
n
 x
the Ratio Test the series  converges for all x .
n=0 n!
n
x
(b) Since the series of part (a) always converges, we must have lim =0 by Theorem .2.6.
n  n!

34. (a)
a a a
n+2 n+3 n+4
R =a
n +a +a +a +   =a 1+ + + +  
n+1 n+2 n+3 n+4 n+1 a a a
n+1 n+1 n+1
a a a a a a
n+2 n+3 n+2 n+4 n+3 n+2
=a 1+ + + +  
n+1 a a a a a a
n+1 n+2 n+1 n+3 n+2 n+1

=a ( 1+r +r r +r r r +   )(*)
n+1 n+1 n+2 n+1 n+3 n+2 n+1

2 3
a
n+1
 a 1+r +r +r +   =
n+1 n+1 n+1 n+1 1 r
n+1

(b) Note that since {r } is increasing and r  L as n  , we have r <L for all n . So, starting
n n n

with equation ( ),
*
a
R =a
n n+1 ( 1+r n+1
+r r
n+2 n+1
+r r
n+3 n+2 n+1
r +   ) a
n+1
( 1+L+L +L +   ) =
2 n+1
3
1 L

1 1 1 1 1 1 661
35. (a) s =
5
= + + + + =  0.68854 . Now the ratios
5 n=1 n 2 8 24 64 160 960
n2
a n
n+1 n2 n
r = = = form an increasing sequence, since
n a n+1 2(n+1)
n (n+1)2
2
n+1 n ( n+1 )  n(n+2) 1
r  r =  = = >0 . So by Exercise 34(b), the error in
n+1 n 2(n+2) 2(n+1) 2(n+1)(n+2) 2(n+1)(n+2)
7
 Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests

using s is R  =
a
1/ 6 26 ( 6
)= 1
 0.00521 .
5 5 1 lim r 1 1/2 192
n
n 
a
n+1 2
(b) The error in using s as an approximation to the sum is R = = . We want
n n 1 n+1
1 (n+1)2
2
1 n
R <0.00005 <0.00005 (n+1)2 >20 , 000 . To find such an n we can use trial and error or
n n
(n+1)2
1
a graph. We calculate (11+1)2 =24 , 576 , so s =
11 11
 0.693109 is within 0.00005 of the
11 n=1 n
n2
actual sum.

n 1 2 3 10
36. s =
10
= + + +   +  1.988 . The ratios
10 n=1 n 2 4 8 1024
2
a n
n+1 n+1 2 n+1 1 1 11+1 12 6
r = =  = = 1+ form a decreasing sequence, and r = = = <1 ,
n a n+1 n 2n 2 n 11 2(11) 22 11
n 2
 n
so by Exercise 34(a), the error in using s to approximate the sum of the series  is
10 n=1 n
2
11
a 2048 121
11
R  = =  0.0118 .
10 1 r 6 10,240
11 1
11

  
n n n
37. Summing the inequalities  a  a a for i=1,2,... ,n , we get  a  a a
i i i i=1 i i=1 i i=1 i
  
     
n n n
 lim a  lim a  lim a  a  a a
i=1 i i=1 i i=1 i n=1 n n=1 n n=1 n
n  n  n 
 
 n=1 n
a   n=1
a
n
.


38. (a) Following the hint, we get that a <r for n
N , and so since the geometric series 
n n
r
n n=1

converges ( 0<r<1 ), the series  a converges as well by the Comparison Test, and hence so
n=N n
 
does  a , so  a is absolutely convergent.
n=1 n n=1 n
n n
(b) If lim a =L>1 , then there is an integer N such that a >1 for all n
N , so a >1
n n n
n 
8
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.6 Absolute Convergence and the Ratio and Root Tests


for n
N . Thus, lim a 0 , so  a diverges by the Test for Divergence.
n n=1 n
n 


39. (a) Since 
+ +
a is absolutely convergent, and since a  a and a  a (because a and
n n n n n n
 
a each equal either a or 0 ), we conclude by the Comparison Test that both  a and 
+
a must be
n n n n
absolutely convergent. (Or use Theorem .2.8.)

(b) We will show by contradiction that both  a and  a must diverge. For suppose that 
+ +
a
n n n
1
converged. Then so would 
+
a  a by Theorem .2.8. But
n 2 n
+ 1 1 1 1
 a 
n 2 n
a =
2 n
a+ a
n
 a = 
2 n 2
a , which diverges because 
(
n ) a is only
n

conditionally convergent. Hence,  a can’t converge. Similarly, neither can 
+
a .
n n

40. Let  b be the rearranged series constructed in the hint. This series will have partial sums s that
n n
oscillate in value back and forth across r . Since lim a =0 (by Theorem .2.6), and since the size of
n
n 
the oscillations s  r is always less than a because of the way  b was constructed, we have
n n n
that  b =lim s =r .
n n
n 

9
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

2 2 2
n 1 1 1/n  n 1
1. lim a =lim =lim =1 0 , so the series  diverges by the Test for
n 
n
n 
2
n  1+1/n n=1 2
n +1 n +1
Divergence.

a 2
1 n 1 n n 1 1/n  n 1
=1 , so the series 
n
2. If a = 2 and b = , then lim =lim 2 =lim
n n n b 1+1/n n=1 2
n +n n  n  n +n n  n +n
n
diverges by the Limit Comparison Test with the harmonic series.

1 1  1  1
3. 2
< for all n 1 , so  2
converges by the Comparison Test with  ,a p
2 n=1 n=1 2
n +n n n +n n
series that converges because p=2>1 .

n 1 1
4. Let b = . Then b =0 , and b =b = , but b >b for n 3 since
n 2 1 2 3 6 n n+1
n +n
x 1 /
=
( x +x )  (x 1)(2x+1)  x +2x+1 2 (x 1)
2
=
2
=
2
<0 for x 3 . Thus,
2 2 2 2 2
x +x ( x +x)
2
( x +x) (x +x)
2

 n 1
is decreasing and lim b =0 , so 
n 1
{ b | n 3}
n n n=3
(  1) 2
converges by the Alternating Series
n  n +n
 n 1
Test. Hence, the full series 
n 1
n=1
(  1) 2
also converges.
n +n

a n+2 3n 3n
n+1 ( 3) 2  3 2 3 3
5. lim =lim =lim =lim = <1 , so the series
 a  3(n+1) n+1  3n 3 
3 8
n
n
n 2 ( 3) n 2 2 n 2
n+1
 ( 3)
 n=1 3n
is absolutely convergent by the Ratio Test.
2

3n 3 3  3n n
6. lim n
a =lim =lim = <1 , so  converges by the Root Test.
n 
n
n  1+8n n  1/n+8 8 n=1 1+8n

1
7. Let f (x)= . Then f is positive, continuous, and decreasing on [2, )] , so we can apply the
x ln x
1  1/2
Integral Test. Since
=
u du=2u +C=2 ln x +C , we find
u=ln x, 1/2
dx
x ln x du=dx/x

1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

 t

t
x ln x =lim x ln x =lim 2 ln x =lim ( 2 ln t  2 ln 2 ) = . Since the integral
2 dx t  2 dx t 
2
t 
 1
diverges, the given series  diverges.
n=2 n ln n

k k
2 k!  2 
8.  = . Using the Ratio Test, we get
k=1 (k+2)! k=1 (k+1)(k+2)
a k+1
k+1 2 (k+1)(k+2) k+1
lim =lim =lim 2 =2>1 , so the series diverges.
k  a k  (k+2)(k+3) k k  k+3
2
k
Or: Use the Test for Divergence.

2
 2  k  k
9.  k e = . Using the Ratio Test, we get
k=1 k=1 k
e
a 2 k 2
k+1 (k+1) e k+1 1 2 1 1
lim =lim =lim =1 = <1 , so the series
k  a k  k+1 2 k  k e e e
k e k
converges.

3 3
2  x / x(2 3x )
10. Let f (x)=x e . Then f is continuous and positive on 1, ) , and f (x)= 3
<0 for x 1 ,
x
e
so f is decreasing on 1, ) as well, and we can apply the Integral Test.
3 3 t
 2  x 1  1

x
xe dx=lim  e = , so the integral converges, and hence, the series converges.
1
t  3 1 3e

n+1
1  (  1)
11. b = >0 for n 2 , b is decreasing, and lim b =0 , so the given series  { }
n nln n n n n=2 nln n
n 
converges by the Alternating Series Test.

2
n n n+1 n +n 25
12. Let b = 2
. Then b >0 , lim b =0 , and b  b = 2
 2
= 2 2
,
n n n n n+1
n +25 n  n +25 n +2n+26 (n +25)(n +2n+26)
which is positive for n 5 , so the sequence {b } n
decreases from n=5 on. Hence, the given series
 n

n
( 1) 2
converges by the Alternating Series Test.
n=1
n +25

2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

a n+1 2 2
n+1 3 ( n+1 ) n! 3(n+1) n+1
13. lim =lim =lim =3lim =0<1 , so the
n  a n  ( n+1 )! n 2
3n n  (n+1)n
2
n  n
2
n
n 2
 3n
series  converges by the Ratio Test.
n=1 n!


14. The series  sin n diverges by the Test for Divergence since lim sin n does not exist.
n=1
n 

15.
a (n+1)! 2 5 8 (3n+2)
n+1
lim =lim
n  a n  2 5 8 (3n+2)[3(n+1)+2] n!
n
n+1 1
=lim = <1
n  3n+5 3
 n!
so the series  converges by the Ratio Test.
n=0 2 5 8 (3n+2)

2
n +1 1
16. Using the Limit Comparison Test with a = and b = , we have
n 3 n n
n +1
a 2 3 2
n +1 n n +n 1+1/n 
=1>0 . Since 
n
lim =lim =lim =lim b is the divergent
n  b n 
3 1 n 
3
n  3 n=1 n
n n +1 n +1 1+1/n

harmonic series,  a is also divergent.
n=1 n


does not exist and the series 
1/n 0 n 1/n n 1/n
17. lim 2 =2 =1 , so lim (  1 ) 2 ( 1) 2 diverges by the Test
n=1
n  n 
for Divergence.

1
18. b =
n n 1
for n 2 . {b } n
is a decreasing sequence of positive numbers and lim b =0 , so
n 
n
n 1
 (  1)
 converges by the Alternating Series Test.
n=2 n 1

ln x / 2 ln x 2 ln n 2
19. Let f (x)= . Then f (x)= <0 when ln x>2 or x>e , so is decreasing for n>e .
x 3/2 n
2x
3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

ln n 1/n 2  ln n
=0 , so the series 
n
By l’Hospital’s Rule, lim =lim =lim ( 1)
n  n n  1/ ( 2 n ) n  n n=1 n
converges by the Alternating Series Test.

a k
k+6 5 1 k+6 1  k+5
= <1 , so the series 
k+1
20. lim =lim = lim converges
k  a k  k+1 k+5 5 k  k+5 5 k=1 k
k 5 5
by the Ratio Test.

2n
 (  2)  4 n 4
21.  = . lim n
a =lim =0<1 , so the given series is absolutely
n=1 n n=1 n n 
n
n  n
n
convergent by the Root Test.

2 2
n 1 n n 1  n 1
22. 3 2
< 3 2
< = for n 1 , so  3 2
converges by the Comparison
3 2 n=1
n +2n +5 n +2n +5 n n n +2n +5

Test with the convergent p  series 
2
1/n ( p=2>1 ) .
n=1

1 1
23. Using the Limit Comparison Test with a =tan and b = , we have
n n n n
a 2 2
n tan (1/n) tan (1/x) sec (1/x) ( 1/x ) 2 2
lim =lim =lim =lim =lim sec (1/x)=1 =1>0 . Since
 b  1/n x  1/x x  2 x 
n
n
n  1/x
 
 n=1 n
b is the divergent harmonic series, 
n=1 n
a is also divergent.

cos (n/2) 1 1  1  cos (n/2)

24. 2
< 2
< and since  converges ( p=2>1 ),  2
converges
2 n=1 2 n=1
n +4n n +4n n n n +4n
absolutely by the Comparison Test.

25. Use the Ratio Test.

2 2
a n n
(n+1)! e (n+1)n! e n+1  n!
=0<1 , so 
n+1
lim =lim =lim 2
=lim
n  a n  (n+1)
2 n! n  n +2n+1 n  2n+1 n=1 2
n
n e e n! e e
converges.

26.

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

a a 2 n 2
n+1 n+1 n +2n+2 5 1+2/n+2/n 1 1
lim =lim =lim =lim = <1 , so
n  a n  a n  n+1 2
n  2 5 5
n n 5 n +1 1+1/n
2
 n +1
 n=1 n
converges by the Ratio Test.
5

27.
ln x
2
dx=lim
t 

ln x 1
x

x
t

1
(using integration by parts) =1 . So 

n=1
ln n
2
converges by the
2 x n
kln k kln k ln k  kln k
Integral Test, and since < = , the given series  converges by the
3 3 2 k=1 3
( k+1 ) k k ( k+1 )
Comparison Test.

28. Since { } 1
n
is a decreasing sequence, e  e =e for all n 1 , and 
1/n 1/1 
n=1
e
n
2
converges (

1/n
 e
p=2>1 ), so  converges by the Comparison Test. (Or use the Integral Test.)
n=1 2
n

 1
tan n  /2   /2   1 3
29. 0< < . =  which is a convergent p series ( p= >1 ), so
3/2 3/2 n=1 3/2 2 n=1 3/2 2
n n n n
 1
 tan n
 n=1 3/2
converges by the Comparison Test.
n

x / 5 x
30. Let f (x)= . Then f (x) is continuous and positive on 1, ) , and since f (x)= <0
x+5 2
2 x ( x+5)
for x>5 , f (x) is eventually decreasing, so we can use the Alternating Series Test.
n 1  j
=0 , so the series 
j
lim =lim ( 1) converges.
n  n+5 n  n1/2+5n 1/2 j+5
j=1

k k k
5 k (5/4) 3
31. lim a =lim = divide by 4 lim = since lim =0 and
k 
k
k  k k k 
k
k  4
3 +4 (3/4) +1
k k
5  5
lim = . Thus,  diverges by the Test for Divergence.
k  4 k=1 k k
3 +4

5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

n
2n 2  (2n)
32. lim n
a =lim =lim =0 , so the series  converges by the Root Test.
n  n
n 2 n=1 2n
n  n  n n

sin (1/n) 1 a sin (1/n)  sin (1/n)

=1>0 , so 
n
33. Let a = and b = . Then lim =lim
n n n n n
n  b n  1/n n=1 n
n
 1
converges by limit comparison with the convergent p  series  ( p=3/2>1 ) .
n=1 3/2
n

1 1 1  1
. Thus, 
2
34. 0 ncos n n , so  = diverges by comparison with
2 n+n 2n n=1 2
n+ncos n n+ncos n
 1
 n=1 2n
, which is a constant multiple of the (divergent) harmonic series.

2
n n n /n 1 1 1
35. lim a =lim =lim = = <1 , so the series
n 
n
n  n+1 n  n n e
( n+1 ) /n lim ( 1+1/n )
n 
2
 n n
 n=1 n+1
converges by the Root Test.
ln ln n ln n
36. Note that ( ln n )
ln n
(
= e ) ( ln n) ln ln n=nln ln n and ln ln n
as n  , so ln ln n>2 for
= e 
1 1  1
. Since 
ln n 2
sufficiently large n . For these n we have ( ln n ) >n , so <
ln n 2 n=2 2
( ln n ) n n
 1
converges ( p=2>1 ), so does  by the Comparison Test.
n=2 ln n
( ln n )

37. lim n
a
n
=lim ( 21/n 1) =1 1=0<1 , so the series  n=1
( n n
2  1 ) converges by the Root Test.
n  n 

n
38. Use the Limit Comparison Test with a = 2  1 and b =1/n . Then
n n
a 1/n 1/x 1/x 2
n 2  1 2  1 2 ln 2 ( 1/x ) 1/x
lim =lim =lim =lim =lim (2 ln 2)=1 ln 2=ln 2>0 . So since
 b  1/n x  1/x x  2
x 
n
n
n  1/x
 
 n=1 n
b diverges (harmonic series), so does 
n=1
( n
2  1) .
Alternate Solution:

6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.7 Strategy for Testing Series

n 1 1
2  1=  ,
( n 1 ) /n ( n 2 ) /n ( n 3 ) /n 1/n 2n
2 +2 +2 + +2 +1
 1 1  1 
and since  =  diverges (harmonic series), so does  ( n 2  1) by the Comparison
n=1 2n 2 n=1 n n=1
Test.

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series


1. A power series is a series of the form 
n 2 3
c x =c +c x+c x +c x +   , where x is a variable and
n=0 n 0 1 2 3
the c ’s are constants called the coefficients of the series.
n

More generally, a series of the form 
n 2
c (x a) =c +c (x a)+c (x a) +   is called a power series
n=0 n 0 1 2
in ( x a ) or a power series centered at a or a power series about a , where a is a constant.


2. (a) Given the power series 
n
c (x a) , the radius of convergence is:
n=0 n
(a) 0 if the series converges only when x=a
(b)  if the series converges for all x , or
(c) a positive number R such that the series converges if x a <R and diverges if x a >R .
In most cases, R can be found by using the Ratio Test.
(b) The interval of convergence of a power series is the interval that consists of all values of x for
which the series converges. Corresponding to the cases in part (a), the interval of convergence is: (i)
the single point { a} , (ii) all real numbers; that is, the real number line (   , ) , or (iii) an interval
with endpoints a R and a+R which can contain neither, either, or both of the endpoints. In this case,
we must test the series for convergence at each endpoint to determine the interval of convergence.

n
x
3. If a = , then
n n
a n+1
n+1 x n x x
lim =lim  =lim =lim = x .
n  a n  n+1 x n  n+1 / n n  1+1/n
n
n
x 
By the Ratio Test, the series  converges when x <1 , so the radius of convergence R=1 .
n=1 n
 1
Now we’ll check the endpoints, that is, x= 1 . When x=1 , the series  diverges because it
n=1 n
n
1  ( 1)
is a p  series with p= 1 . When x= 1 , the series  converges by the Alternating Series
2 n=1 n
Test. Thus, the interval of convergence is I=  1,1 ) .

n n a n+1
( 1) x n+1 x n+1 x
4. If a = , then lim =lim  =lim = x . By the Ratio
n n+1 n  a n  n+2 n
n  1+1/(n+1)
n x
n n

( 1) x
Test, the series  converges when x <1 , so R=1 . When x= 1 , the series diverges
n=0 n+1
because it is the harmonic series; when x=1 , it is the alternating harmonic series, which converges by
1
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

the Alternating Series Test. Thus, I= (  1,1 .

n 1 n
( 1) x
5. If a = , then
n 3
n
a n n+1 3 3 3
n+1 ( 1) x n ( 1)xn n 3
lim =lim  =lim =lim x =1  x = x
n  a n  3 n 1 n n  3 n  n+1
n (n+1) ( 1) x (n+1)
n 1 n
 ( 1) x
. By the Ratio Test, the series  converges when x <1 , so the radius of convergence
n=1 3
n
n 1
 ( 1)
R=1 . Now we’ll check the endpoints, that is, x= 1 . When x=1 , the series  converges
n=1 3
n
n 1 n
 ( 1) ( 1)  1
by the Alternating Series Test. When x= 1 , the series  =  converges
n=1 3 n=1 3
n n
because it is a constant multiple of a convergent p  series ( p=3>1 ) . Thus, the interval of
convergence is I=  1,1 .

a n+1
n n+1 n+1 x 1
6. a = n x , so we need lim =lim =lim 1+ x = x <1 for
n
n  a n 
n
n  n
n x n
convergence (by the Ratio Test), so R=1 . When x= 1 , lim a =lim n = , so the series
n
n  n 
diverges by the Test for Divergence. Thus, I= (  1,1 ) .

n
x
7. If a = , then
n n!
a n+1
n+1 x n! x 1
lim =lim  =lim = x lim = x  0=0<1 for all real x .
n  a n  (n+1)! n n  n+1 n  n+1
n x
So, by the Ratio Test, R= , and I= (   , ) .

n n n
8. Here the Root Test is easier. If a =n x then lim a =lim n x = if x
0 , so R=0 and
n n
n  n 
I={0} .

a n+1 n+1
n+1 ( n+1 ) 4 x 1 1
9. lim =lim =lim 1+ 4 x =4 x . Now 4 x <1 x < , so
n  a n 
n n
n  n 4
n n4 x
2
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

1 1  1
. When x= , we get the divergent series 
n
by the Ratio Test, R= ( 1) n , and when x= , we
4 4 n=1 4
 1 1
get the divergent series  n . Thus, I=  , .
n=1 4 4

n
x
10. If a = , then
n n
n3
a n+1 n
n+1 x n3 xn x n x
lim =lim  =lim = lim = . By the Ratio
n  a n  n+1 n n  (n+1)3 3 n  n+1 3
n (n+1)3 x
x
Test, the series converges when <1 x <3 , so R=3 . When x= 3 , the series is the alternating
3
harmonic series, which converges by the Alternating Series Test. When x=3 , it is the harmonic series,
which diverges. Thus, I=  3,3) .

n n a n+1 n+1 4
( 2) x n+1 2 x n 4 n
11. a = , so lim =lim  =lim 2 x =2 x , so by the
n 4
n n  a n 
4
n+1
n n
n  n+1
n 2 x
1 1 1
Ratio Test, the series converges when 2 x <1  x < , so R= . When x= , we get the
2 2 2
divergent
n
 1 1 1  ( 1)
p  series  p= 1 . When x= , we get the series  , which converges by
n=1 4
n 4 2 n=1 4
n
1 1
the Alternating Series Test. Thus, I=  , .
2 2

n a n+1 n 5 5
x n+1 x 5n x n x
12. a = , so lim =lim  =lim = . By
n n 5
n  a n 
n+1 5 n
n  5 n+1 5
5n n 5 (n+1) x
the Ratio Test, the series converges when x /5<1  x <5 , so R=5 . When x= 5 , we get the series
n
 ( 1)
 n=1 5
, which converges by the Alternating Series Test. When x=5 , we get the convergent p 
n
 1
series  ( p=5>1 ) . Thus, I=  5,5 .
n=1 5
n

n
n x
13. If a =( 1) n
, then
n
4 ln n
3
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

a n+1 n
n+1 x 4 ln n x ln n x
lim =lim  = lim =  1 (by l’Hospital’s
n  a n 
n+1 n 4 n  ln (n+1) 4
n 4 ln (n+1) x
x x
Rule) = . By the Ratio Test, the series converges when <1  x <4 , so R=4 . When x= 4 ,
4 4
n n
 x  [( 1)( 4)]  1 1 1  1
 = = > and 
n
( 1) . Since ln n<n for n 2 , is
n=2 n n=2 n n=2 ln n ln n n n=2 n
4 ln n 4 ln n
 1
the divergent harmonic series (without the n=1 term),  is divergent by the Comparison
n=2 ln n
n
 x  1
Test. When x=4 ,  =
n n
( 1) ( 1) , which converges by the Alternating Series Test.
n=2 n n=2 ln n
4 ln n
Thus, I= (  4,4 .

2n a 2n+2 2
x n x (2n)! x n+1
14. a =( 1) , so lim =lim  =lim =0 . Thus, by the
n (2n)! n  a n  (2n+2)!
2n
n  (2n+1)(2n+2)
n x
Ratio Test, the series converges for all real x and we have R= and I= (   , ) .

a n+1
n n+1 n+1 x 1 1
15. If a = n (x 1) , then lim =lim =lim 1+ x 1 = x 1 .
n
n  a n 
n
n  n
n x 1 n
By the Ratio Test, the series converges when x 1 <1  1<x 1<1 0<x<2 . When x=0 , the series

becomes 
n
( 1) n , which diverges by the Test for Divergence. When x=2 , the series becomes
n=0

 n=0
n , which also diverges by the Test for Divergence. Thus, I= ( 0,2 ) .

a 3 n+1 3
3 n n+1 (n+1) (x 5) 1
16. If a =n (x 5) , lim =lim =lim 1+ x 5 = x 5 . By
n
n  a n 
3 n
n  n
n (x 5) n
the Ratio Test, the series converges when x 5 <1  1<x 5<1  4<x<6 . When x=4 ,

the series becomes 
n 3
( 1) n , which diverges by the Test for Divergence. When x=6 , the series
n=0

becomes 
3
n , which also diverges by the Test for Divergence. Thus, R=1 and I= ( 4,6 ) .
n=0

n
n (x+2)
17. If a =( 1) , then
n n
n2

4
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

a n+1 n
n+1 x+2 n2 n x+2 x+2
lim =lim  =lim  = . By the Ratio Test, the
n  a n  n+1 n
n  n+1 2 2
n (n+1)2 x+2
x+2
series converges when <1 x+2 <2  2<x+2<2  4<x<0 . When x= 4 , the series becomes
2
n n
 (  2)  2  1
 = =
n
( 1) , which is the divergent harmonic series. When x=0 , the series
n=1 n n=1 n n=1 n
n2 n2
n
 (  1)
is  , the alternating harmonic series, which converges by the Alternating Series Test.
n n=1
Thus, I= (  4,0 .

n
( 2) n
18. If a = (x+3) , then
n n
a n+1 n+1
n+1 ( 2) (x+3) n 2 x+3 1
lim =lim  =lim =2 x+3 <1 x+3 <
n  a n  n+1 n n n  1+1/n 2
n ( 2) (x+3)
1 7 5 7  1
soR=   <x< . When x= , the series becomes  , which diverges because it
2 2 2 2 n=1 n
n
1 5  (  1)
is a p  series with p= 1 . When x= , the series becomes  , which converges by the
2 2 n=1 n
7 5
Alternating Series Test. Thus, I=  , .
2 2

n
( x 2 ) n x 2
19. If a = , then lim a =lim =0 , so the series converges for all x (by the Root
n n n 
n
n  n
n
Test). R= and I= (   , ).
20.
a n+1 n
n+1 ( 3x 2 ) n3 3x 2 1 3x 2 2
lim =lim  =lim  = = x ,
n  a n 
n+1 n
n  3 1+1/n 3 3
n ( n+1 ) 3 ( 3x 2 )
2 1 5 1
so by the Ratio Test, the series converges when x <1  <x< . R=1 . When x= , the
3 3 3 3
n
 (  1) 5
series is  , the convergent alternating harmonic series. When x= , the series becomes
n=1 n 3
1 5
the divergent harmonic series. Thus, I=  , .
3 3
5
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

n n
21. a = (x a) , where b>0 .
n n
b
a n+1 n
n+1 (n+1) x a b 1 x a x a
lim =lim  =lim 1+ = .
n  a n  n+1 n n  n b b
n b n x a
x a
By the Ratio Test, the series converges when <1 x a <b [ so R=b ] 
b
 b<x a<b a b<x<a+b . When x a =b , lim a =lim n= , so the series diverges. Thus,
n
n  n 
I=(a b,a+b) .

n
n(x 4)
22. a = 3
, so
n
n +1
a n+1 3 3
n+1 (n+1) x 4 n +1 1 n +1
lim =lim  =lim 1+ x 4 = x 4 .
n  a n 
3 n n  n 3 2
(n+1) +1
n n x 4 n +3n +3n+2
By the Ratio Test, the series converges when x 4 <1  1<x 4<1 3<x<5 .
  n
When x 4 =1 ,  a = , which converges by comparison with the convergent p 
n=1 n n=1 3
n +1
 1
series  ( p=2>1 ) . Thus, I= 3,5 .
n=1 2
n

a n+1
n n+1 (n+1)!(2x 1)
23. If a =n!(2x 1) , then lim =lim =lim (n+1) 2x 1   as
n a n
n  n!(2x 1) n  n 
n

n  for all x

1
2
. Since the series diverges for all x

1
2
, R=0 and I= { } 1
2
.

2 n 2 n n
nx nx nx
24. a = = n = n , so
n 2 4 6     (2n)
2 n! 2 (n 1)!
a n+1 n
n+1 (n+1) x 2 (n 1)! n+1 x
lim =lim  =lim =0 . Thus, by the Ratio Test, the series
n  a n 
n+1 n
n 
2 2
n 2 n! n x n
converges for all real x and we have R= and I= (   , ) .

a n+1 2
n+1 4x+1 n 4x+1
25. lim =lim  =lim = 4x+1 , so by the Ratio
a 2 n 2
n  n  ( n+1 ) 4x+1 n  ( 1+1/n )
n
6
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

1 1
Test, the series converges when 4x+1 <1  1<4x+1<1  2<4x<0  <x<0 , so R= . When
2 4
n
1  (  1)
x= , the series becomes  , which converges by the Alternating Series Test. When x=0
2 n=1 2
n
 1 1
, the series becomes  , a convergent p  series ( p=2>1 ). I=  ,0 .
n=1 2 2
n

n n a
(  1 ) ( 2x+3) n+1 nln n
26. If a = , then we need lim = 2x+3 lim = 2x+3 <1 for
n nln n n  a n  ( n+1 ) ln ( n+1 )
n
1   1
convergence, so  2<x< 1 and R= . When x= 2 ,  a = , which diverges (Integral
2 n=2 n n=2 nln n
n
  (  1)
Test), and when x= 1 ,  a = , which converges (Alternating Series Test), so
n=2 n n=2 nln n
I= (  2, 1 .

n
x x
27. If a = , then lim n
a =lim =0<1 for all x , so R= and I= (   , ) by the
n n n 
n
n  ln n
( ln n )
Root Test.

n a
2 4 6     ( 2n ) x n+1 2n+2
28. If a = , then we need lim =lim x = x <1 for
n 1 3 5     ( 2n 1 ) a 2n+1
n  n 
n
2 4 6     ( 2n )
convergence, so R=1 . If x= 1 , a = >1 for all n since each integer in the
n 1 3 5     ( 2n 1 )
numerator is larger than the corresponding one in the denominator, so  a diverges in both cases by
n
the Test for Divergence, and I= (  1,1 ) .


29. (a) We are given that the power series 
n
c x is convergent for x=4 . So by Theorem 3, it must
n=0 n

converge for at least  4<x 4 . In particular, it converges when x= 2 ; that is, 
n
c ( 2) is
n=0 n
convergent.

(b) It does not follow that 
n
c ( 4) is necessarily convergent.
n=0 n


30. We are given that the power series 
n
c x is convergent for x= 4 and divergent when x=6 . So
n=0 n

7
Stewart Calculus ET 5e 0534393217;11. Infinite Sequences and Series; 11.8 Power Series

by Theorem 3 it converges for at least  4 x<4 and diverges for at least x 6 and x< 6 . Therefore:
(a) It converges when x=1 ; that is,  c is convergent.
n

(b) It diverges when x=8 ; that is,  c 8 is divergent.

n
n

(c) It converges when x= 3 ; that is,  c (  3) is convergent.

n
n

(d) It diverges when x= 9 ; that is,  c ( 9) = ( 1) c 9 is divergent.

n n n
n n

k
( n! ) n
31. If a = x , then
n (kn)!
a k k
n+1 (n+1)! (kn)! ( n+1 )
lim = lim x =lim x
n  a 
k
 (kn+k)(kn+k 1)   (kn+2)(kn+1)
n n ( n! ) k(n+1) ! n

( n+1 ) ( n+1 ) ( n+1 )

= lim    x
n  ( kn+1 ) ( kn+2 ) ( kn+k )
n+1 n+1 n+1 1 k
= lim lim    lim