MAT 137

Tutorial #5– Continuity and Limit Proofs

October 14-17, 2024
1. Let x ∈ R. Let f be a function with domain R.
(a) Write the “ ε-δ definition of f is continuous at a.”
The standard definition is
∀ε > 0, ∃δ > 0, s.t. |x − 1| < δ =⇒ |f (x) − f (a)| < ε
Remark:
– You can add “∀x ∈ R” right before the implication. but it is not needed:
it is implicit.
– You can write “0 < |x − a| < δ” instead of “|x − a| < δ”. Inside this
statement, it does not change the meaning.
(b) Assume f is continuous. Assume that f (a) > 1. Prove that there exists an
open interval I, centered at a, such that
∀I ∈ I, f (x) > 1.
Write a proof directly from the above definition of continuity. Do not use any
of the limit laws or any other theorem.

Proof. 1. Take ε = f (a) − 1. By hypothesis, ε > 0. Use this value of “ε” in

the definition of “f is continuous at a”, and I have
∃δ > 0 such that |x − a| < δ =⇒ |f (x) − f (a)| < ε
I keep this value of δ.
2. Then I define the interval I to be I = (a − δ, a + δ).I will show that
∀x ∈ I, f (x) > 1.
3. Let x ∈ I. This is equivalent to |x − a| < δ. Therefore, with ε = f (a) − 1,
it follows that
|f (x)−f (a)| < ε =⇒ f (x)−f (a) > −ε =⇒ f (x) > f (a)−ε = f (a)−(f (a)−1) = 1
which is what I wanted to show.

Remark: we only need to find ε such that f (a) − ε ≥ 1. Hence, ε = f (a) − 1

works. We can also choose ε = f (a)−1
2
which means f (a) − ε is the middle point
of 1 and f (a). This also works.
2. (a) Let f and g be two functions defined on R. Let L ∈ R. Assume lim f (x) = L
x→∞
and lim g(x) = ∞. Prove that lim f (g(x)) = L.
x→∞ x→∞
WTS:
∀ε > 0 ∃M > 0 s.t x > M =⇒ |f (g(x)) − L| < ε

Proof. Let ε > 0.

Use ε as the epsilon in the definition of lim f (x) = L and we have
x→∞

∃M1 > 0, s.t. x > M1 =⇒ |f (x) − L| < ε · · · · · · (1)

Use M1 as the cut-off of g(x) in the definition of lim g(x) = ∞ and we have
x→∞

∃M2 > 0, s.t. x > M2 =⇒ g(x) > M1 · · · · · · (2)

Take M = M2 .
Let x ∈ R. Assume x > M .
Since x > M = M2 , we get g(x) > M1 by (2).
Treating g(x) as x in (1), now we have |f (g(x)) − L| < ε by (1).
This completes the proof.

(b) Use part (a) to prove lim cos(x) does not exist.
x→∞
Hint: prove by contradiction.

Proof. Assume lim cos(x) exists. Let lim cos(x) = L. Let g1 (x) = 2πbxc
x→∞ x→∞
which is defined on R. We have lim g1 (x) = ∞. By part (a), we have
x→∞

L = lim f (g1 (x)) = lim cos(2πbxc) = lim 1 = 1.

x→∞ x→∞ x→∞
π
However, let g2 (x) = + 2πbxc which is defined on R. We have lim g2 (x) = ∞.By
2 x→∞
part (a), we have
π
L = lim f (g2 (x)) = lim cos( + 2πbxc) = lim 0 = 0.
x→∞ x→∞ 2 x→∞

Since L cannot be two different values, we have a contradiction. Therefore,

lim cos(x) does not exist.
x→∞
 Q3, Q4 and Q5 are the practice for the students outside of the tutorial. Please don’t
cover anything in Q3-Q5 during the tutorial.
Note: L’Hôpitals’ Rule is not allowed to use for any limits computation in
this worksheet.
3. Calculate the following limits:
x2 − 9 (x + 3)(x − 3)
(a) lim = lim =lim (x + 3) = 3 + 3 = 6
x→3 x − 3 x→3 x−3 x→3
2−x
(b) lim
x→3 x − 3
2−x 2−x 2−x
Since lim+ = −∞ and lim− = ∞, we have lim does not ex-
x→3 x−3 x→3 x−3 x→3 x − 3
ist.
2−x
(c) lim
x→3 (x − 3)2
2−x 2−x 2−x
Since lim+ 2
= −∞ and lim− 2
= −∞, we have lim = −∞.
x→3 (x − 3) x→3 (x − 3) x→3 x − 3

sin x sin 1
(d) lim = 1 = sin 1. Note that sinx x is continuous at 1.
x→1 x
sin x 1 −1
(e) lim Note that −1 x
≤ sinx x ≤ x1 when x > 0. Since lim = lim = 0,
x→∞ x x→∞ x x→∞ x
sin x
we have lim =0 by the Squeeze Theorem
x→∞ x
√
x+1−1
(f) lim
x→0 √x √
( x + 1 − 1)( x + 1 + 1) x+1−1 1 1
=lim √ = lim √ =lim √ =
x→0 x( x + 1 + 1) x→0 x( x + 1 + 1) x→0 x+1+1 2
x3 + 2x + 1
(g) lim
x→∞ 4x3 − x2 + 6
x3 (1 + 2/x2 + 1/x3 ) 1 + 2/x2 + 1/x3 1+0+0 1
= lim 3 3
= lim 3
= =
x→∞ x (4 − 1/x + 6/x ) x→∞ 4 − 1/x + 6/x 4−0+0 4
√
q q
x2 + 1 + 2x x2 (1 + x12 ) + 2x |x| (1 + x12 ) + 2x
(h) lim = lim = lim
x→−∞ 5x x→−∞ 5x x→−∞ 5x
Note that |x| = −x if x q < 0. q q
−x (1 + x12 ) + 2x x(− (1 + x12 ) + 2) − (1 + x12 ) + 2
Thus, we have lim = lim = lim
x→−∞ 5x x→−∞
q 5x x→−∞ 5
1
− (1 + x2 ) + 2 −1 + 2 1
Since x12 → 0 as x → −∞, we have lim = =
x→−∞ 5 5 5
4. Calculate the following limits:

(a) lim x2 + 2x 1 − cos(3t)

x→1 (h) lim
t→0 t2
h3 − 5h2 + 3h + 6 √ √
(b) lim y + 4 − 4y + 1
h→2 h3 − h2 − 3h + 2 (i) lim √
t y→1 y−1
(c) lim
t→0 sin(2t) sin (1 − cos x)
sin(2x) (j) lim
(d) lim x→0 x tan(πx)
x→0 sin(3x) r !
sin(2z 2 ) 1 u+2
(e) lim (k) lim −2
u→2 2 − u u−1
z→0 cos(3z) sin2 (5z)
tan(x − 3) h √ i
(f) lim (l) lim x + x2 − x
x→3 2x − 6 x→∞
2ex h √ i
(g) lim (m) lim x + x2 − x
x→0 sin(2ex ) x→−∞

Hint: The answers to Questions 4l and 4m are different.

−3
(a) 3, (b)−1, (c) 12 ,(d) 23 ,(e) 25
2
, (f) 12 , (g) sin2 2 , (h) 92 , (i) √ 5
1
, (j) 2π , (k) 34 ,
(l)∞(There is no indeterminate form to begin with. It is ∞ + ∞.),
(m) 21 (Multiply and divide by the conjugate first. Make sure you get 1/2 and not
−1/2.)

5. Calculate the following limits (Hint: you may need to use Squeeze Theorem):
1 x2 (2 + sin2 x)
(a) lim sin(x2 ) sin( ) (c) lim
x→0 x x→∞ x + 10
√ 1 5x2 − sin(3x)
(b) lim− 1 − x2 cos( ) (d) lim
x→1 (x − 1)2 x→−∞ x2 + 2

(a)0, (b) 0, (c)∞, (d)5.