Mathematics

ALLEN
CHAPTER - 1 ALGEBRA
Polynomial
Polynomial
An expression of the form a0xn + a1xn–1 + a2xn – 2 + ... + an – 1x + an , where n is a whole number and a0, a1,
a2.....an belong to some number system F, is called a polynomial in the variable x over the number system F.
A polynomial is denoted by f(x) or g(x) etc.
Real Polynomial
A polynomial is called a real polynomial if all the coefficients are real numbers.
Leading Coefficient and Leading Term
If a0 ¹ 0, then a0 (the coefficient of highest degree term) is called the leading coefficient and a0xn is called the
leading term.
Degree of Polynomial
The highest index of the variable x occuring in the polynomial f(x) is called the degree of the polynomial.
Zero Degree Polynomial
The constant c = cx0 is called a polynomial of degree zero.
Linear Polynomial
The polynomial f(x) = ax + b, a ¹ 0 is of degree one and is called a linear polynomial.
Quadratic Polynomial
The polynomial f(x) = ax2 + bx + c, a ¹ 0 is of degree two and is called a quadratic polynomial.
Cubic Polynomial
The polynomial f(x) = ax3 + bx2 + cx + d, a ¹ 0 is of degree three and is called a cubic polynomial.
Biquadratic Polynomial
The polynomial f(x)= ax4 + bx3 + cx2 + dx + e, a ¹ 0 is of degree four and is called a biquadratic polynomial
Zero Polynomial
A polynomial, all of whose coefficients are zero, is called a zero polynomial
Equality of Two Polynomials
Two polynomials
f(x) = a0 + a1x + a2x2 + ... + anxn
g(x) = b0 + b1x + b2x2 + ... + bmxm
are said to be equal if the coefficients of like powers of x in the two polynomials are equal.
When m ³ n, if f(x) = g(x), then a0 = b0, a1 = b1,..., an = bn , bn + 1 = bn + 1 = ... = bm = 0
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

When m £ n, if f(x) = g(x), then a0 = b0, a1 = b1,..., am = bm, am +1 = am +2 .......= an = 0.

Division Algorithm
If f(x) and g(x) are two non-zero polynomials, then there exist unique polynomials q(x)and r(x) such that
f(x) = q(x) × g(x) + r(x)
where either r(x) = 0
deg r(x) < deg g(x)
The polynomial q(x) is called the quotient and r(x) the remainder.
When f(x) is divided by g(x), then degree of q(x) = deg f(x) – deg g(x)
Particular Case : When g(x) = ax + b. a linear polynomial, then either r(x) = 0 or
deg r(x) < deg g(x) =1
i.e,deg r(x) = 0
So that r(x) is a constant.
1
RMO/INMO
ALLEN
Root of an Equation
A number ‘a’ is called a root of equation
f(x) = 0, iff f(a) = 0
a is a root of f(x) = 0
Þ f(a) = 0
Conversely f(a) = 0
Þ a is a root of f(x) = 0
Theorem 1
Remainder Theorem: If f(x) is a polynomial, then f(h) is the remainder when f(x) is divided by x – h.
Proof
Let Q(x) be the quotient R the remainder.
When f(x) is divided by x – h.
Then, f(x) = (x – h)Q(x) + R
Putting x = h, f(h) = R,
i.e., R = f(h)
Theorem 2
Factor Theorem : If h is a root of equation f(x) = 0, then (x – h) is a factor of f(x) and conversely.
Proof
Let Q(x) be the quotient and R, the remainder when f(x) is divided by x – h.
Then, f(x) = (x – h) Q(x) + R
Putting x = h, f(h) = R, But f(h) = 0
Q h is a root of f(x)= 0
\ R=0
\ f(x) = (x – h)Q
which shows that x – h is a factor of f(x).
Conversely, if x – h is a factor of f(x) h must be a root of f(x) = 0.
Divide f(x) by x – h and let Q(x) be the quotient.
Then, f(x) = (x – h) Q(x)
Putting x = h, f(h) = 0,
which shows that h is a root of f(x) = 0
Theorem 3
In an equation with real coefficients, non-real complex roots occur in conjugate pairs.
Proof
Let the given equation be
f(x) = 0 ...(i)
Let a + ib (where b ¹ 0) be a root of f(x) = 0,
then f(a + ib) =0 ...(ii)
We have, to prove that a – ib is also a root of (1)
Divide f(x)by [x – (a + ib)][x – (a – ib)]
Let Q(x) be the quotient and if there is any remainder, since it must be a linear polynomial in x, take it as Rx + R'.
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

\ f(x) = [x – (a + ib)][x – (a – ib)] Q(x) + Rx + R' ...(iii)

Putting x = a + ib,
f(a + ib) = R(a + ib) + R'
But f(a + ib) = 0 [Q Eq. (ii)]
\ R(a + ib) + R' = 0 or (Ra + R') + iR b = 0
Equating the real and imaginary parts on both sides.
Ra + R' = 0 ...(iv)
Rb = 0 ...(v)
from Eq. (v), either R = 0 or b = 0
Now, b¹0
[Q in case b = 0 even the complex root a + ib becomes real]
\ R=0

2
 Mathematics
ALLEN
Putting this in Eq. (iv), we have R' = 0
\ from Eq. (iii)
f(x) = [x – (a + ib)][x – (a – ib)] Q(x)
i.e., [x – (a + ib)][x – (a – ib)]
divides f(x) exactly.
Hece, a – ib is also a root of the give equatio f(x) = 0.
Corollary
Every equation of an odd degree having real coefficients, has atleast one real root, because complex roots occur
in pairs.
Theorem 4
In an equation with rational coefficients irrational roots occur in conjugate pairs.
Proof
Let the given equation be
f(x) = 0 ...(i)
Let a + b be a root of Eq. (i) [where a and b are rational, b is + ve but not a perfect square]

\ f( a + b ) = 0 ...(ii)

We have to prove that a - b is also a root of Eq. (i)

Divide
f(x) by [x – ( a + b )][x – ( a - b )]
Lot Q(x) be the quotient and if there is remainder, since it must be a linear polynomial in x, take it as Rx + R'
\ f(x) = [x – ( a + b )][x – ( a - b )] Q(x) + Rx + R' ...(iii)

Putting x = a+ b

f( a + b ) = R( a + b ) + R'

But f( a + b ) = 0 [Q Eq. (ii)]

\ R( a + b ) + R' = 0

i.e., (Ra + R') + R b = 0

Now, rational and irrational number cannot destroy one another.
\ If there sum vanishes, each must vanish separately.
Þ Ra + R' = 0

R b =0
Since, b = 0, \ from Eq. (v), R = 0
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

Putting R = 0 in Eq. (iv), R' = 0

\ Eq. (iii) becomes
f(x) = [x – ( a + b )][x – ( a - b )] Q(x)
This shows that
[x – ( a + b )][x – ( a - b )] is a factor of f(x)

Hence, a - b is also root of f(x).

Theorem 5
Two nonzero polynomials f(x), g(x) over k are coprime, iff there exist some polynomials a(x), b(x) over k such
that
a(x)f(x) + b(x)g(x) = 1

3
RMO/INMO
ALLEN
Proof
Let f(x) be coprime
\ f(x), g(x)) = 1
\ There exist some polynomials a(x), b(x) over k such that
a(x)f(x) + b(x)g(x) = 1
Conversely
Let there exist polynomials a(x), b(x) over k such that
a(x)f(x) + b(x)g(x) = 1
Well shall prove that (f(x), g(x)) = 1
Let (f(x), g(x)) = d(x)
\ d(x)|f(x) and d(x)|g(x)
\ d(x)|a(x)f(x) + d(x)|b(x)g(x)
\ d(x)|a(x) f(x) + b(x)g(x)
\ d(x)|1
Hence, d|(x) = 1
\ f(x), g(x) are coprime.
Concept 1
To transform an equation into another whose roots are the reciprocals of the roots of the given equation.
Let f(x) = a0xn + a1xn – 1 + a2xn – 2 + ... + an – 1 x + an = 0
be the given equation. If x be a root of the given equation and y that of the transformed equation, then

1 1
y= or x =
x y

1 æ1ö
Hence, the transformed equation is obtained by putting x = in f (x) = 0 and is therefore f ç y ÷ = 0.
y è ø

1 1 1 1
i.e., a0 + a1 n -1 + a2 n -2 + ... + a n-1 + a n = 0
yn y y y

or anyn + an – 1yn – 1 + ... + a2y2 + a1y + a0 = 0

Rule If the equation given be complete (if not, it may be made to take that form), then the above transfor-
mation is effected, if we take the last coefficient to be the 1st, last but one to be the llnd and so on.
Concept 2
All those equations which remain unchanged when x is replaced by 1/x are called reciprocal
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

equation. These are of two types

(i) those in which the coefficient of terms, equidistant from the begining and the end, are equal and of the same
sign. e.g.,
x4 + 6x3 + 8x2 + 6x + 1 = 0
(ii) those in which these coefficients are equal but of opposite sign e.g.,
x5 – 4x4 + 7x3 – 7x2 + 4x – 1 = 0
Thus, we conclude that if a is a root of the reciprocal equation, then 1/a must be its root. Hence, the root of
a reciprocal equation occur in pairs of a, 1/a, b, 1/b, g, 1/g and so on.
In case, the equation be of odd degree then it will be seen that one of its roots must be either +1 or –1. In
case, the equation be of even degree and of Ilnd type, then it will be seen that x2 – 1 will always be its factor
e.g.,

4
 Mathematics
ALLEN
6x6 + 5x5 – 44x4 + 44x2 – 5x – 6 = 0
6(x6 – 1) + 5x(x4 – 1) – 44x2(x2 – 1) = 0
(x2 – 1) [6(x4 + x2 + 1) + 5x(x2 + 1) – 44x2] = 0
(x2 – 1) (6x4 + 5x3 – 38x2 + 5x + 6) = 0
Above shows that (x2 – 1) is a factor of the given equation. The equation 6x4 + 5x3 – 38x2 + 5x + 6 = 0, is
of even degree and of the first type i. e., the coefficient of terms equidistant from the beginning and the end
are equal and is called in the standard form to which all the reciprocal equations can be reduced.
Concept 3
To transform an equation into another, whose roots are the roots of the given equation with sign changed.
If y be the root of the transformed equation, then y = – x or x = – y. Hence, transformed equation is obtained
by putting x = – y in f(x) = 0,
\ f(–y) = 0, which takes the form
a0yn – a1yn – 1 + a2yn – 2 – ... + (–1)n an = 0
Rule If the given equation be complete (if not, it may be made to take that form), then the transformation is
effected by changing the sign of Ilnd, IVth, Vlth, i. e., all even terms, i. e., by changing the sign of every alternate
term beginning from the llnd.
Concept 4
To transform a given equation into another, whose roots are the roots of the given equation multiplied by a given
number m.
If y be a root of the transformed equation, then y = mx or x = y/m. Hence, the transformed equation is obtained
by putting x = y / m in f (x) = 0 and \ f (y/m) = 0
yn y n -1 y n -2
i.e. a0 + a1 + a 2 + ... + a n = 0
mn m n -1 m n -2
or a0yn + a1myn – 1 + a2m2yn – 2 + ... + mnan = 0
Rule If the given equation be complete (if not, it may be made complete), then the above transformation is
effected by multiplying the successive terms beginning from the llnd by m, m2, m3,..., mn respectively.
Note
1. The above transformation is very useful when we are dealing with equations with fractional coefficients. We
can get rid of fractional coefficients by multiplying the roots of the given equation by the LCM of the
denominators of the fractional coefficients. Similarly, if the coefficient of leading term be not unity but k and
we want to make it unity, then it can be done so by multiplying the roots of the given equation by k.
2. If we have to divide the roots of the given equation by m, we say that we have to multiply its roots by
1/m.
Concept 5
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

To transform a given equation into another, whose roots are the roots of the given equation diminished (or increased)
by a constant h.
Let f(x) = a0xn + a1xn – 1 +azxn – 2 + ... + an –1
x + an = 0 ...(i)
If y be a root of the transformed equation, then y = x – h or x = y + h. Hence, the transformed equation is
obtained by putting x = y + h in f (x) = 0 and is therefore,
f (y + h) = 0
or a0(y + h)n + a1 (y + h)n – 1 + a2(y + h)n – 2 +....+ an = 0
The simplification of the above equation will be difficult and let us suppose that this equation, when simplified
and arranged in descending powers of y takes the form.
A0yn + A1yn – 1 + A2yn – 2 + ... + An – 1y + An = 0 (ii)

5
RMO/INMO
ALLEN
The problem is to find A0, A1, A2,...,An
Now, y=x–h
\ A0(x – h)n + A1(x – h)n – 1 + A2 (x – h)n – 2 + ... + An – 1 (x – h) + An = 0
or [A0 (x – h)n – 1 + A1(x – h)n – 2 + A2(x – h)n – 3 + ... + An – 1] × (x – h) +An = 0 ...(iii)
The LHS of the above is identical with LHS of line (i) and hence, if f(x) be divided by (x – h), then the remainder
is the value of An and the quotient is A0 (x – h)n – 1 + A1(x – h)n – 2 + ... + An – 1 and the quotient when again
divided by (x – h) leaves the remainder An – 1. If we continue the above process, then we shall find A n, An – 1,....,
A2, A1 and the last quotient A0 is clearly equal to a0.
Rule In order to find the successive coefficients of the transformed equation, we have to divide the given complete
equation by (x – h), the quotient again by x – h and so on. The successive remainders that are left in the above
procedure of division are the required coefficients, the first coefficient being the same as that of the given equation.
Symmetric Functions of the Roots
A symmetric function of the roots of an equation is a function in which all the roots are involved alike, so that
the expressions remains unaltered when two of the roots are interchanged. e.g., The expression ab + bg + ga
is a symmetric function of the roots a, b, g of a cubic. For the sake of brevity, we generally denote such symmetric
functions by attaching å to one of its terms.
Thus, åab = ab + bg + ga
Rule
To find the number of terms in a symmetric function.
If n = the total number of roots of the equation.
r = the number of roots occuring in the symmetric function.
k = the number of roots having the same index (degree), then the total number of terms in the symmetric
function.
n!
=
(n - r)!k!
For example
1. If a, b and g are the roots of a cubic equation, then the total number of terms in

3!
å a2b is =6
(3 - 3)!0!

ab 3!
å 2
= å abg -2 is =3
g (3 - 2)!2!
2. If a, b, g and d are the roots of a biquadratic equation, then the total number of terms in
4!
å a2b is =2
(4 - 1)!2!
a 4!
å = å ab-1 is = 12
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

b (4 - 2)!0!
Concept 6
To find The sum of the integral power of the roots of an equation.
a0xn + a1xn–1 + a2xn–2 + ... + an = 0 ....(A)
If a1, a2, a3,..., are the roots of the polynomial equation, then let us represent sr = åa1r
i.e.,
s1 = åa1 = a1+ a2 + a3 + ....
s2 = åa12 = a12 + a22 + a32 + ....
s3 = åa13 = a13 + a23 + a33 + .... and so on
The following results will help us to find out the value of s1, s2, s3,..., sr.
a0s1 + a1 = 0 ....(i)
a0s2 + a1s1 + 2a2 = 0 ....(ii)

6
 Mathematics
ALLEN
a0s3 + a1s2 + a2s1 + 3a3 = 0 ....(iii)
a0sr + a1sr – 1 + a2sr – 2 + ... + ar – 1s + rar = 0 ....(iv)
From result (i), we can find the value of s1 and by putting its value in result (ii) we can find s2. Proceeding in the same
way we can find the value of sr, provided r < n, if however r ³ n, then multiply Eq. (A) by xr – n
a0xr + a1xr – 1 + a2xr – 2 + ... + an – 1xr – n + 1 + anr – n = 0 ...(v)
Putting x = a1, a2, ..., an in succession,

a0 a1r + a1a1r -1 + a2a1r -2 + ... + a n -1a1r -n +1 + an ar -n = 0

a0 a2r + a1a2r -1 + a2ar2-2 + ... + a n -1a2r -n +1 + an a2r -n = 0

... ... ...
... ... ...
... ... ...

a0 anr + a1anr-1 + a2 anr-2 + ... + an -1a nr -n +1 + a na nr -n = 0

Adding we have
a0sr + a1sr –1 + a2sr – 2 + ... + an – 1sr – n + 1 + ansr – n = 0
Putting r = n, n + 1, n + 2 etc, we get
a0sn + a1sn –1 + a2sn –2 + ... + an –1s1 + nan=0
éëQ s 0 a10 + a20 + ... + a 0n = n ùû

a0sn + 1 + a1sn + a2sn – 1 + ... + an – 1s2 + ans1 = 0

a0sn + 2 + a1sn + 1 + a2sn + ... + an – 1s3 + ans2 = 0
and so on. These results results gives values of
sn, sn + 1, sn + 2,....
Some Important Observations
1. If an equation of degree n is complete, then it has (n + 1) terms. In this case (the number of continuation
in sign) + (the number of variation in sign) = n.
e.g. In the polynomial
2x5 + 7x4 – 5x3 – 4x2 + x + 5
Number of continuation in signs = 3
Number of variations in signs = 2
So, n=5
2. If the equation is incomplete, then (the number of continuation in sign) + (number of changes in sign) = total
number of non-zero terms in the polynomial –1.
e.g., Consider the polynomial x4 – 3x2 + 5
The signs are + – +
Number of continuation in sign = 0
Number of variation in sign = 2
Number of non-zero terms = 3
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

0+2=3–1
3. If f (x) = 0 is a complete equation, then a continuation in sign f(x) becomes a variation in sign in f(–x) and
vice-versa. But if the equation is not complete, the result may or may not be true.
e.g.. Let f(x) = x3 – 3x2 + 4x – 5 be a complete polynomial.
The sign are + – + –
There are 3 changes of signs and no continuation in sign.
Again f(–x) = –x3 – 3x2 – 4x – 5. There are 3 continuation of sign.
[Ist, IInd, IIIrd, IVth] and no changes of sign.
Let f(x) = x4 + x2 + 5
Signs are + + +
There are 2 continuations of sign and no change of sign.
So, result need not be true, if the polynomial is incomplete.

7
RMO/INMO
ALLEN
4. If between two like signs (either both + ve or both – ve), we introduce any number of + ve or – ve signs,
then the total number of resulting variations will be an even number.
5. If between two unlike signs (one + ve and other is – ve). we introduce any number of + ve or – ve signs,
then the total number of resulting variations will be an odd number.
Descarte's Rule of Signs
The polynomial equations f (x) = 0 with real coefficient cannot have more.
1. + ve roots than the number of changes of sign in f(x)
2. – ve roots than the number of changes of signs in f(x)
Concept Location of Zeros
Consider the polynomial

æ a n -1 a1 a0 ö
f(x) = anxn + an – 1xn – 1 + ... + a1x + a0 = anxn ç 1 + a x + a x n -1 + a x n ÷
è n n n ø
For every large value of |x|, the quantity in the bracket is very near to 1.
\ f(x) behaves like anxn. Hence, for every large values of |x|, the graph of y = f(x) is similar to the graph of
y = anxn.
It moves up (or down) as x ® ± ¥
Thus, for an > 0, we have
(i) anxn ¾® ¥ as x ¾® ± ¥
(ii) anxn ¾® ¥ as x ¾® ¥, if n is odd.
(iii) anxn ¾® – ¥ as x ¾® ¥, if n is odd.
Same is true for f(x)
Theorem Let f(x) be a real polynomial of degreen (³ 1) and a, b be two real numbers such that a < b.
(i) If f(a) and f(b) are of opposite signs, then the polynomial f(x) has atleast one and always an odd number of
real zeros in (a, b).
(ii) If f(a) and f(b) are of the same sign, then the polynomial f(x) has either one real zero or an even number
of real zeros in (a,b).

SOLVED EXAMPLES
Ex.1 Solve x(x + 1)(x + 2)(x + 3) + 1 = y2 for integer values of x and y.
Sol. Let P = x(x + 1)(x + 2)(x + 3). Since the product of four consecutive integers is divisible by 24, we may
write P = 24m where m = 0, 1, 2,... Since P + 1 = 24m + 1 is the square of an integer for selected values
of m, we set
x(x + 1)(x + 2)(x + 3) + 1
= x4 + 6x3 + 11x2 + 6x + 1 = (x2 + ax + 1)2.
By comparing the coefficients of like powers of x on both sides of this identity, we find a = 3.
Therefore, y2 = (x2 + 3x + 1), so that y = x2 + 3x + 1 or y = – (x2 + 3x + 1) and, hence, there are infinitely
many solutions in integers since we may assign to x any arbitrary integer value.
Ex.2 Solve in real numbers the system of equations
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

ìx + y + z = 4
ï 2 2 2
í x + y + z = 14
ï 3 3 3
î x + y + z = 34
Sol. Consider the monic polynomial
p(t) = t3 + at2 + bt + c,
with roots x, y, z.
Because x + y + z = 4, it follows that a = – 4, hence
p(t) = t3 – 4t2 + bt + c.
We have
x2 + y2 + z2 = (x + y + z)2 – 2(xy + xz + yz)
It follows that

8
 Mathematics
ALLEN
b = xy + xz + yz = 1
The number x, y, z are the roots of P, thus
x3 – 4x2 + x + c = 0,
y3 – 4y2 + y + c = 0,
z3 – 4z2 + z + c = 0.
Adding these equalities and using the equations of the system, we obtain c = 6, hence
p(t) = t3 – 4t2 + t + 6,
We observe that t1 = –1 is a root, so P factors as
P(t) = (t + 1)(t2 – 5t + 6), the other two roots being t2 = 2 and t3 = 3. It follows that the solutions of the
system are the triple (–1, 2, 3) and all of its permutations.
Ex.3 Find all solutions in integers of the equation m2 + 2n2 = m2n2 – 2000.
Sol. m2n2 – 2n2 – m2 = 2000.
From the first two terms we can take n 2 common. But this process stops there. We cannot proceed further.
If we want to factorize the LHS some mainipulations should be done.
n2(m2 – 2) – m2 + 2 = 2000 + 2.
(i.e.,) n2(m2 – 2) – (m2 – 2) = 2002
Þ (m2 – 2)(n2 – 1) = 2002.
As usual we employ the factor method.
2002 = 1 × 2002 = 2 × 1001 = 7 × 286 = 11 × 182
= 13 × 154 = 14 × 143 = 22 × 91 = 26 × 77
The student now can do the problem fully to get.
m = ± 4, n = ± 12.
Ex.4 If P1 = 2x4 + 3x3 – 4x2 + 5x + 3,
P2 = x3 + 2x2 – 3x + 1,
P3 = x4 + 2x3 – x2 + x + 2,
and aP1 + bP2 + cP3 = 0, find the value of a + b + c where abc ¹ 0.
Sol. aP1 + bP2 + cP3 = (2a + c)x4 + (3a + b + 2c)x3 +
(– 4a + 2b – c)x2 + (5a – 3b + c)x + (3a + b + 2c) = 0.
Therefore,
(1) 2a + c = 0 (2) 3a + b + 2c = 0 (3) –4a + 2b – c = 0
(4) 5a – 3b + c = 0 (5) 3a + b + 2c = 0.
By addition of the five equations, we obtain (6) 9a + b + 5c = 0.
From equation (1), c = –2a or –8a – 4c = 0. Adding –8a – 4c = 0 to equation (6),
we obtain a + b + c = 0.
Comment 1: Taking a = 1, b = 1, c = –2, verify the statement P1 + P2 – 2p3 = 0.
Comment 2: Since aP1 + bP2 + cP3 = 0 and, at least, one of a, b, and c is not equal to zero,
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

and P1, P2, P3 are said to be linearly dependent.

Ex.5 Let a and b be two of the roots of the polynomial x4 + x3 – 1.
Prove that ab is a root of the polynomial x6 + x4 + x3 – x2 – 1.
Sol. Let c and d be the other two roots of x4 + x3 – 1. The Viete’s relations yield.
a + b + c + d = –1,
ab + ac + ad + bc + bd + cd = 0,
abc + abd + acd + bcd = 0,
abcd = –1.
Write these equalities in terms of s = a + b, s' = c + d, p = ab and p' = cd (this is often useful) to obtain.
s + s' = –1,
p + p' + ss' = 0,
9
RMO/INMO
ALLEN
ps' + p's = 0,
pp' = –1.
1
Substituting p ' = - and s' = –1 – s in the second and in the third equalities yields.
p
1 2
p- -s -s =0
p

s
and p(-1 - s) - =0
p

It follows from the second equality that

p2
s=- .
2
p +1
Plugging this into the first equality gives.

1 p4 p2
p- - 2 2
+ 2 = 0,
p (p + 1) p +1

A short computation shows that this is equivalent to

p6 + p4 + p3 – p2 – 1 = 0,
hence p = ab is a root of the polynomial
x6 + x4 + x3 – x2 – 1.
(n + 1)2
Ex.6 Determine all natural numbers ‘n’ such that is an integer
n+7
(n + 1)2
Sol. is an integer..
n+7
(n + 7 - 6)2
Which Þ must be an integer
n+7
(n + 7)2 - 12(n + 7) + 36
(i.e.,)
n+7

36
(i.e.,) n + 7 – 12 + must be an integer..
n+7
\ n + 7 = 36, 18, 12, 9
n = 29, 11, 5, 2
Ex.7 Two different real numbers a and b are such that the equation (x2 + 20ax + 10b)(x2 + 20bx + 10a) = 0
has no real solutions. Prove that 20(b – a) cannot be an integer number.
Sol. Suppose that 20(b – a) is an integer number. By symmetry we can also suppose that b > a, and then
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

20(b – a) ³ 1. Since there are no real solutions, the discriminant of the polynomial x2 + 20bx + 10a is negative,
1 1
therefore 10b2 – a < 0. Then, we have 10b2 < a < b and b < . Hence 0 < b – a < b < and, then
10 10

1
20(b – a) < 2, but if 20(b – a) is an integer number we have 20(b – a) = 1 and then b = a + . Thus,
20

2
æ 1 ö 1
10b2 – a = 10 ç a + ÷ – a = 10a2 + > 0, which is a contradiction. Therefore, 20(b – a) can never
è 20 ø 40
be an integer number.

10
 Mathematics
ALLEN
Ex.8 Solve the equation

1 + mx = x + 1 - mx, where m is a real parameter..

Sol. The equation is equivalent to 1 + mx - 1 - mx = x. Multiplying and dividing by 1 + mx + 1 - mx, which

is the conjugate of 1 + mx - 1 - mx, it follows that:

2mx
= x.
1 + mx + 1 - mx
A solution is x = 0 and, if x ¹ 0, then 2m = 1 + mx + 1 - mx, hence m is positive. Squaring and simplifying,

it follows that 2m2 - 1 = 1 - m2 x2 , hence 2m2 – 1 ³ 0. Squaring again and solving for x we get

é 1 ù
x = ±2 1 - m2 and, since 1 – m2 ³ 0, it follows that m Î ê ,1ú .
ë 2 û
Ex.9 If a, b and g be the roots of the equation x3 + px + q = 0, find value of Sx5.
Sol. a0 = 1, a1 = 0, a2 = p, a3 = q
\ a0s1 + a1 = 0 Þ s1 = 0
a0s2 + a1s1 + 2a2 = 0 Þ s2 = –2p
a0s3 + a1s2 + a2s1 + 3a3 = 0 Þ s3 = –3q
Multiplying both sides of given equation by x2
x5 + px3 + qx2 = 0
Putting x = a, b, g in succession and adding
s5 + ps3 + qs2 = 0
or S5 + p(–3q) + q(–2p) = 0
\ S5 = Sx5 = 5pq
Ex.10 Fill all values of x satisfying the pair of equations x2 – px + 20 = 0, x2 – 20x + p = 0.
Sol. Case I: When p = 20, the equations are identical and satisfied by two values of x, 10 + 4 5 and 10 - 4 5 .
Case II: When p ¹ 20, then x2 – px + 20 = x2 – 20x + p, (20 – p)x = p – 20. Since p ¹ 20, we obtain
by division x –1.
To satisfy the given equations with the value x = –1, the value of p must be – 21.
Generally, the pair of equations x2 – px – p – 1 = 0 and x2 + (p + 1)x + p = 0 is satisfied by x = –1.
Ex.11 (x2 + 2x + 5) is a factor of (x4 + px2 + q) find p, q.
Sol. It is given that x2 + 2x + 5 is a factor of x4 + px2 + q. The other factor must also be a quadratic factor.
Let it be ax2 + bx + c.
\ (x4 + px2 + q) º (x2 + 2x + 5)(ax2 + bx + c).
Equating the coefficients of x4 we have a = 1.
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

\ x4 + px2 + q º (x2 + 2x + 4)(x2 + bx + c).

x4 + px2 + q º (x4 + 2x3 + bx3 + cx2 + 5x2 + 2bx2 + 2cx + 5bx + 5c).
Equating the like coefficients, we get
2 + b = 0 Þ b = –2 (1)
c + 5 + 2b = p
or
c+5–4= p
c + 1 = p. (2)
2c + 5b = 0. Using (1) 2c – 10 = 0 Þ c = 5
(2) Þ p = 6.
5c = q Þ q = 25.

11
RMO/INMO
ALLEN
Ex.12 If x = cy + bz, y = az + cx, z = bx + ay where x ¹ 0, y ¹ 0, z ¹ 0. Prove that a2 + b2 + c2 + 2abc = 1.
Sol. The student can brush up his memory of the rule of cross multiplication to solve set of simultaneous equation
a1x + b1y + c1z = 0
a2x + b2y + c2z = 0

x -y z
= = .
Then b1 c2 - b2 c1 a1 c2 - a 2 c1 a1 b2 - a 2 b1

The two equations in the given problem can be written as

x – cy – bz = 0
– cx + y – az = 0.
Using the rule of cross multiplication

x -y z
= = = k(say)
-ac - b -bc - a -1 + c2
Þ x = k(ac – b)
y = k(–bc – a)
z = k(–1 + c2).
The third equation is bx + ay = z substituting in this we get
bk(–ac – b) + ak(–bc – a) = k(–1 + c2)
Þ c2 – 1 = –b2 – abc – a2 – abc
Þ a2 + b2 + c2 + 2abc = 1.
Ex.13 It a, b, g be the roots of the equation x3 – 7x + 7 = 0. Find a–4 + b–4 + y–4.
Sol. Roots of the equation x3 – 7x + 7 = 0 are a, b, g. Changing x to 1/x and multiplying by x3, we get
7x3 – 7x2 + 1 = 0 ...(i)
Its roots being the reciprocals of the roots of given equation are 1/a, 1/b, 1/g.
Let us denote them by a', b', g'.
We have to find
a–4 + b–4 + g–4 = (a')4 + (b')4 + (g')4 = s'4
Here, a0 = 7, a1 = – 7, a2 = 0, a3 = 1
\ a0S1' + a1 = 0 Þ 7S'1 – 7 = 0
\ S’1 = 1
a0S'2 + a1S'1 + 2a2 = 0 Þ 7s – 7 = 0
S'2 = 1
a0S'3 + a1S'2 + a2S'1 + 3a3 = 0
Þ 7S'3 – 7 + 3 = 0
\ S'3 = 4/7
Multiplying Eq. (i) by x, we get
7x4 – 7x3 + x = 0
Putting x = a', b', g' in succession and adding
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

7S'4 – 7S'3 + S'1 = 0 Þ 7S'4 – 4 + 1 = 0

Hence, a–4 + b–4 + g–4 = 3/7
\ S4' = 3/7
Ex.14 If a, b and g are the roots of x3 + px + q = 0. Prove that

a 5 + b5 + g 5 a 3 + b3 + g 3 a 2 + b 2 + g 2
(i) = ×
5 3 2
(ii) 3(a2 + b2 + g2) (a5 + b5 + g5) = 5(a3 + b3 + g3) (a4 + b4 + g4)
Sol. Here, a0 = 1, a1 = 0, a2 = p, a3 = q
a0 s1 + a1 = 0 Þ s1 = 0
a0s2 + a1s1 + 2a2 = 0 Þ s2 = –2p
a0s3 + a1s2 + a2s1 + 3a3 = 0
12
 Mathematics
ALLEN
Þ s3 = –3q
(i) Multiplying both sides of given equation by x2.
x5 + px3 + qx2 = 0
Putting x = a, b, g in succession and adding
s5 + ps3 + qs2 = 0
or s5 – 3pq – 2pq = 0
\ s5 = 5pq ...(i)
s5
Now, = pq;
5
s 3 s2
× = (–q) (–p) = pq
3 2
s5 s3 s2
\ × ×
5 3 2

a 5 + b5 + g 5 a 3 + b3 + g 3 a 2 + b 2 + g 2
Hence, = ×
5 3 2
(ii) Multiplying both sides of the given equation by x,
x4 + px2 + qx = 0
Putting x = a, b, g in succession and adding
s4 + ps2 + qs1 = 0 or s4 – 2p2 = 0
\ s4 = 2p2
Also, s5 = 5pq [from Eq. (i)]
3(a2 + b2 + g2)(a5 + b5 + g5) = 3s2s5
= 3(–2p)(5pq) = –30p2q
5 (a3 + b3 + g3) (a4 + b4 + g4) = 5s3s4
= 5(–3q)(2p2) = –30p2q
Hence, 3(a2 + b2 + g2) (a5 + b5 + g5)
= 5(a3 + b3 + g3) (a4 + b4 + g4)
Ex.15 If the coefficients a, b, c of the equation ax2 + bx + c = 0 are odd integers, find a relation between a, b,
c for which the roots are rational.
Sol. We prove that if a, b, c are odd integers, there are no rational roots.
Method I: For rational roots, the discriminant must be the square of an integer. Let b2 – 4ac = t2, and
b2 – t2 = 4 ac = (b – t)(b + t) where bot b and t are odd. (Why is t odd?)
Let b = 2b1 + 1 and let t = 2t1 + 1.
Therefore, (2b1 + 1 – 2t1– 1)(2b1 + 1 + 2t1 + 1) = 4ac,
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

2(b1 – t1)(2)(b1 + t1 + 1) = 4ac, and (b1 – t1) (b1 + t1 + 1) = ac

The product ac is odd. If b1 and t1 are each even, then b1 – t1 is even. If b1 and t1 are each odd, then b1
+ t1 + 1 is even. Hence, for all possibilites, (b1 – t1) (b1 + t1 + 1) is even; we have a contradiction.
Method II : Let b = 2b1 + 1, a = 2a1 + 1, c = 2c1 + 1.
é b1 (b1 + 1) ù
Then D = b2 – 4ac = (2b1 + 1)2 – 4(2a1 + 1)(2c1 + 1) = 8 ê - 2a1 c1 - a1 - c1 - 1ú + 5. Since
ë 2 û
b1 (b1 + 1)
is integral, then D = 8k + 5. If D = N2, with N odd, then, N2 = 8k + 5. However, the square
e
2
of an odd number leaves a remainder of 1 when divided by 8; that is, N2 º 1(mod 8).
Now Q 8k + 5 ¹ 1(mod 8)
Therefore, rational roots are impossible.
13
RMO/INMO
ALLEN
Ex.16 Solve in real numbers the system
ì x + y + z = 0,
ï 3 3 3
í x + y + z = 18,
ï x7 + y7 + z7 = 2058.
î
Sol. Consider the polynomial
P(t) = t3 + at2 + bt + c,
with roots x, y, z.
Since x + y + z = 0, it follows that a = 0, hence
P(t) = t3 + bt + c.
Because x, y, z are the roots of P, we have
x3 + bx + c = 0,
y3 + by + c = 0,
z3 + bz + c = 0.
Adding these equalities and using the fact that x3 + y3 + z3 = 18, we obtain c = – 6.
Therefore
P(t) = t3 + bt – 6.
Now, use the last equation of the system to find b. Multiply the previous equalities
by xn, yn, zn respectively, and add them to obtain
xn + 3 + yn + 3 + zn + 3 + b(xn + 1 + yn + 1 + zn + 1) – 6(xn + yn + zn) = 0.
Denoting Sn = xn + yn + zn for all n ³ I, this equality becomes
Sn + 3 + bSn + 1 – 6Sn = 0 (*)
for all positive integers n. We have S7 = 2058. On the other hand, using (*), we obtain
S7 = –bS5 + 6S4 = –b(–bS3 + 6S2) + 6(–bS2 + 6S1)
= b2S3 – 12bS2 + 36S1.
Since S3 = 18, S2 = (x + y + z)2 – 2(xy + xz + yz) = –2b and S1 = 0, it follows that
S7 = 42b2.
so b = ±7.
The equation t3 + 7t – 6 = 0 has only one real root (observe that the function f(t) = t3 + 7t – 6 is strictly
increasing). The equation t3 – 7t – 6 = 0 has roots ti = – 1, t2 = –2 and t3 = 3, and thus the solutions of
the system are (– 1, –2, 3) and all of its permutations.
Ex.17 Solve x(y + z) = 44; y(z + x) = 50; z(x + y) = 54.
Sol. On adding the given equations
2(xy + yz + zx) = 148
Þ xy + yz + zx = 74 ...(i)
On subtracting given equations one by one from 1, we get
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

yz = 30, zx = 24, xy = 20
On multiplying the three, given above together.
x2y2z2 = 30 × 24 × 20 Þ xyz = ± 120 ...(ii)
On dividing Eq. (ii) by the values of xy, yz and zx respectively, we get
x = ±4, y = ±5, z = ± 6
Ex.18 If a, b, g be the roots of x3 + 2x2 – 3x – 1 = 0.
1 1 1
Find the value of 3
+ 3+ 3.
a b g
Sol. Roots of the equation,
x3 + 2x2 – 3x – 1 = 0 ...(i)
are a, b, g.

14
 Mathematics
ALLEN
Let us first form an equation whose roots are a3, b3, g3.
If y is a root of the transformed equation, then
y = x3 ...(ii)
To eliminate x between Eqs. (i) and (ii), Eq. (i) can be written as
x3 – 1 = – (2x2 – 3x)
On cubing both sides of above equations
Þ x9 – 3x6 + 3x3 – 1 = – [8x6 – 27x3 – 18x3 (2x2 – 3x)]
Þ x9 – 3x6 + 3x3 – 1 = – [8x6 – 27x3 – 18x3 (1 – x3)]
Þ x9 – 3x6 + 3x3 – 1 = – 8x6 + 27x3 + 18x3 – 18x6
Þ x9 + 23x6 – 42x3 – 1 = 0
3
Putting x = y in this equation, we get
y3 + 23y2 – 42y – 1 = 0 ...(iii)
Its roots are a3, b3, g3.
1
Changing y to , Eq. (iii) becomes
y

1 23 42
+ - –1=0
y3 y2 y
y3 + 42y2 – 23y – 1 = 0 ...(iv)
1 1 1
Its roots are , , .
a 3 b3 g 3
1 1 1
\ 3
+ 3 + 3 = sum of roots of Eq. (iv) = – 42
a b g
Ex.19 Let a, b and c be real numbers different from zero, with a + b + c = 0 and a3 + b3 + c3 = a5 + b5 + c5.
6
Prove that a2 + b2 + c2 = .
5
Sol. Let Tn = an + bn + cn for each integer number n, then T0 = 3, T1 = 0 and T2 = (a + b + c)2 – 2(ab + bc
+ ea) = –2(ab + bc + ca). Now, define A = ab + bc + ca and B = abc; then, by Vieta's formulas, it follows
that a, b and c are the roots of the equation x3 + Ax – B = 0 and T2 = –2A. For n ³ 0, we substitute a,
b and c in xn + 3 = –Axn + 1 + Bxn and adding we obtain Tn + 3 = – ATn + 1 + BTn. Then
T3 = –AT1 + BT0 = 3B,
T4 = -AT2 + BT1 = 2A2,
T5 = –AT3 + BT2 = –5AB.

T5 T T 6
Hence, = – AB = 3 × 2 . Since T3 = T5, the last equality implies that T2 = .
5 3 2 5
Ex.20 If a, b, c, x are real numbers such that abc ¹ 0 and
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

xb + (1 - x)c xc + (1 - x)a xa + (1 - x)b

= =
a b c
Prove that
a + b + c = 0 or a = b = c.
Sol. Let

xb + (1 - x)c xc + (1 - x)a xa + (1 - x)b

= = =k
a b c
Þ ak = bx + c – cx
bk – cx + a – ax
ck = ax + b – bx.

15
RMO/INMO
ALLEN
Add
Þ (a + b + c)k = a + b + c.
Þ k = 1 Q a + b + c ¹ 0.
\ bx + c – cx = a
Þ a2 = abx + ac – acx
b2 = bcx + ab – abx
c2 = cax + bc – bcx
Add
a2 + b2 + c2 = ab + bc + ca.
Þ a2 + b2 + c2 – ab – bc – ca = 0
Þ [2a2 + 2b2 + 2c2 – 2ab – 2bc – 2ca] = 0
Þ (a – b)2 + (b – c)2 + (c – a)2 = 0
Þ a=b=c
Ex.21 Let a, b and c be real numbers such that the equation ax2 +bx + c = 0 has two different real solutions p1,
p2 and the equation cx2 + bx + a = 0 has two different real solutions q1, q2. Also the numbers p1, q1, p2,
q2, in this order, form an arithmetic progression. Prove that a + c = 0.
Sol. Each of the equations ax2 + bx + c = 0 and cx2 + bx + a = 0 have two different real solutions, a ¹ 0 and
1
c ¹ 0. Moreover, r is the root of ax2 + bx + c = 0 if and only if is a root of cx2 + bx + a = 0. Therefore,
r

ì1 1 ü
{q1, q2} = í , ý .
î p1 p2 þ

If p1, q1, p2, q2 are in arithmetic progression,

1 1 |p1 - p2 |
|p1 – p2|=|q1 – q2|= - =
p1 p2 |p1p2 |
from where |p1p2| = 1.

c
Using Vieta's formula (4.1), we have p1p2 = , so |c| = |a| and then a = ±c.
a
If a = c, the two given quadratic equations are equal, and then p1 = q1, p2 = q2, which tell us that the difference
of the progression is 0. Then, p1 = q1 = p2 = q2 which is a contradiction. Therefore, a = –c or a + c = 0.
Ex.22 Find the positive solutions of the following system of equations:
ì a 2 b2 4 4
ï 2 - 2 = 8(y - x ),
í x y
ïax - by = x 4 - y 4
î
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

where a, b > 0 are parameters.

Sol. We solve the system for a, b instead of x, y. Multiplying the first equation by x4 yields
b2 x 4
a2x2
= + 8x4(y4 – x4),
y2
and from the second equation we deduce
a2x2 = (by + x4 – y4)2 = b2y2 + 2by(x4 – y4) + (x4 – y4)2.
It follows that

b2 (x 4 - y 4 )
– 2by(x4 – y4) – (x4 – y4) (9x4 – y4) = 0.
y2

If x = y, from the original system we obtain a = b.

16
 Mathematics
ALLEN
If x ¹ y, we can divide by x4 – y4 to obtain a quadratic equation in b :
b2 – 2by3 – y2(9x4 – y4) = 0.
The solutions are y3 + 3x2y and y3 – 3x2y. If b = y3 – 3x2y. the second equation gives a = x3 – 3xy2. Because
a, b > 0, it follows that x2 > 3y2 and y2 > 3x2 > 9y2; this is a contradiction. Thus b = y3 + 3x2y and a =
x3 + 3xy2. We observe that a + b = (x + y)3 and a – b = (x – y)3; hence x + y = 3
a + b and x – y = 3
a-b ,
yielding

æ 3 a+b + 3 a-b 3 a+b - 3 a -b ö

(x, y) = çç , ÷÷ .
è 2 2 ø

Ex.23 Solve the system of equation for real x and y

æ 1 ö æ 1 ö
5x ç 1 + 2 ÷ = 12 ; 5y ç 1 - 2 ÷ = 4.
è x + y2 ø è x + y2 ø

122 42
Sol. (5x)2 + (5y)2 = 2
+ 2
æ 1 ö æ 1 ö
ç1 + 2 ÷ ç1 - 2 ÷
è x + y2 ø è x + y2 ø

1
Put x2 + y2 = , we have
t

25 144 16
= 2
+
t (1 + t) (1 - t)2

So that 25(1 – t2)2 = 144t(1 – t)2 + 16t(1 + t)2

or 25t4 – 160t3 – 206t2 – 160t + 25 = 0
Dividing by t2, we get

160 25
25t2 – 160t + 206 – + 2 =0
t t

æ 1ö æ 1ö
i.e., 25 ç t 2 + 2 ÷ - 160 ç t + ÷ + 206 = 0
è t ø è tø

1
Let t + = u, we have
t
25(u2 – 2) – 160u + 206 = 0
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

i.e., 25u2 – 160u + 156 = 0

160 ± (1602 - 4 × 25 × 156) 160 ± 100 6 26

\ u= = = ,
50 50 5 5

6 1 6
When u = , t + = which does not give any real values of t.
5 t 5

26 1 26 1
If u= ,t+ = , which gives t = or 5
5 t 5 5

1 1
\ t = 5 or so that x2 + y2 = or 5
5 5
17
RMO/INMO
ALLEN
1
When x2 + y2 = , we get
5
5x(1 + 5) = 12, 5y(1 – 5) = 4,

2 1
so that x= ,y=–
5 5
When x2 + y2 = 5, we get

æ 1ö æ 1ö
5x ç 1 + ÷ = 12 ; 5y ç 1 - ÷ = 4,
è 5 ø è 5ø

so that x = 2, y = – 1

2 1
Thus, x= , y = – and x = 2, y = – 1
5 5
Ex.24 Solve the equation :
x3(x + 1) = 2(x + a) (x + 2a),
where a is a real parameter.
Sol. The equation is equivalent to
x4 + x3 – 2x2 – 6ax – 4a2 = 0.
This fourth degree equation is difficult to solve. We might try to factor the left-hand side, but without some
appropriate software, the process would get quite complicated. What if we think of a as the unknown and
x as the parameter? In this case, the equation can be written as a quadratic :
4a2 + 6xa – x4 – x3 + 2x2 = 0,
whose discriminant is
36x2 + 16(x4 + x3 – 2x2) = 4x2(2x + l)2,
1 2
fortunately a square. Solving for a, we obtain the solutions a1 = – x – x and
2
1 2 1
a2 = x – x, yielding the factorization
2 2

æ 1 2 ö æ 1 2 1 ö
4a2 + 6ax – x4 – x3 + 2x2 = 4 ç a + 2 x + x ÷ ça - 2 x + 2 x÷
è ø è ø

= –(x2 + 2x + 2a) (x2 – x – 2a).

1 1
Finally, we obtain the solutions x1,2 = – 1 ± 1 - 2a , x3,4 = 2 ± 2 1 + 8a .
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

Ex.25 Find all pairs (x, y) of positive integers satisfying the equation x3 – y3 = 2xy + 8.
Sol. We have to find positive integral solutions of x3 – y3 = 2xy + 8.
The RHS is even Þ LHS also must be even.
\ Both x, y are even or both are odd.
Now
x3 – y3 = (x – y)(x2 + xy + y2) (1)
Apply AM ³ GM for x2, xy, y2

x 2 + xy + y2 3 2
³ x .xy.y2
3
Þ x2 + xy + y2 ³ 3xy.

18
 Mathematics
ALLEN
(1) Þ \ x3 – y3 ³ (x – y)3xy.
Since both x, y are either even or odd x – y must be even and greater than or equal to 2.
(i.e.,) x – y ³ 2.
Þ x3 – y3 ³ 6xy
\ 2xy + 8 ³ 6xy
Þ 8 ³ 4xy
Þ xy £ 2.
If x, y are both even, then xy cannot be 2. If x, y are both odd then xy cannot be 2.
But xy = 1 is possible with x = 1, y = 1.
But x = 1, y = 1 do not satisfy the equation x3 – y3 = 2xy + 8.
\ There are no positive integral solution.
Ex.26 Determine all triples (x, y, z) of positive integers which are solutions of
2x2y2 + 2y2z2 + 2z2x2 – x4 – y4 – z4 = 576.
Sol. To find the positive integers x, y, z such that
2x2y2 + 2y2z2 + 2z2x2 – x4 – y4 – z4 = 576.
All such problems belong to the Diophantine equations. In such a problem we seek positive integral solutions.
The RHS is 576. If the LHS is expressed as a product of factors then we can proceed. Thus essentially it
is a problem on factorization.
E = 2x2y2 + 2y2z2 + 2z2x2 – x4 – y4 – z4
= 4x2y2 – [x4 + y4 + z4 + 2x2y2 – 2y2z2 – 2z2x2]
= 4x2y2 – [x2 + y2 – z2]2
= (2xy)2 – (x2 + y2 – z2)2
= (2xy + x2 + y2 – z2) (2xy + z2 – x2 – y2)
= [(x + y)2 – z2] [z2 – (x – y)2]
= (x + y + z) (x + y – z) (z + x – y) (z – x + y).
\ The equation becomes.
(x + y + z) (x + y – z) (x – y + z) (–x + y + z) = 576.
x, y, z are positive integers. We find thai x + y + z = (x + y – z) + 2z etc.
Þ All the factors are of same parity.
Þ All of them must be even. Let
x + y + z = 2a, x + y – z = 2b, x – y + z = 2c, – x + y + z = 2d.
Þ abcd = 36.
Without loss of generality assume x ³ y > z.
Þ a > b ³ c ³ d.
We observe a = b + c + d. 36 = 6 × 3 × 2 × 1. This factorization is unique.
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

Þ a = 6, b = 3, c = 2, d = 1.
x = 5, y = 4, z = 3.
Exploiting the symmetry of the equation. x, y, z can be cyclically changed.
\ (x, y, z) º (5, 4, 3), (5, 3, 4),(4, 5, 3),(4, 3, 5),(3, 4, 5), (3, 5, 4) are the solutions.
Ex.27 Let P(x) be a cubic polynomial with integer coefficients such that there exist different integers a, b and c
such that P(a) = P(b) = P(c) = 2. Prove that no integer number d satisfying P(d) = 3 exists.
Sol. Let Q(x) = P(x) – 2, since a, b and c are the roots of Q(x), it is clear that Q(x) = a(x – a) (x – b) (x – c), for
some integer number a. If for some integer number d we have P(d) = 3: then, since 1 = P(d) – 2 = Q(d)
= a(d – a) (d – b) (d – c), the factors on the right-hand side of the equation have to be –1 or 1, then two
of d – a, d – b, d – c are equal, so a, b, c are not different, which is a contradiction. This guarantees that
there does not exist an integer number d with P(d) = 3.

19
RMO/INMO
ALLEN
Ex.28 Solve the equation 5 - x = 5 – x2.

Sol. There is no parameter! Note, however, that we must have x Î [– 5 , 5 ], since the left-hand side is
nonnegative. Squaring both sides, we obtain the equation
x4 – 10x2 + x + 20 = 0
and, if we are lucky, we might observe the factorization
(x2 + x – 5) (x2 – x – 4) = 0.
What if we are not lucky? Well, let us introduce a parameter ourselves: replace 5 by a,
where a > 0: 2
a - x = a – x . Squaring both sides yields the equation
x4 – 2ax2 + x + a2 – a = 0
Consider this as a quadratic in a with x as a parameter :
a2 – (2x2 + 1)a + x4 + x = 0.
The discriminant of the quadratic is
D = (2x – 1)2
and thus the roots are a1 = x2 + x and a2 = x2 – x + 1. It follows that
a2 – (2x2 + 1)a + x4 + x = (a – x2 – x) (a – x2 + x – 1).
Returning to a = 5, we arrive at the desired factorization.
The equations x2 + x – 5 = 0 and x2 – x – 4 = 0 have the solutions

1 1
x1,2 = (–1 ± 2 ) and X3,4 = (1 ± 17 ). respectively.
2 2

Only two of them belong to the interval [– 5 , 5 ], and therefore the solutions of the initial equation are

1 1
(– 1 + 21 )and (1 – 17 ).
2 2
Ex.29 Express P(c) = c6 + 10c4 + 25c2 as a polynomial of least positive degree when c is a root of
x3 + 3x2 + 4 = 0.
Sol. Method I : Since c is a root of x3 + 3x2 + 4 = 0, we have c3 + 3c2 + 4 = 0 (I). We multiply (I) successively
by c, c2, and c3 to obtain c4 + 3c3 + 4c = 0 (II), c5 + 3c4 + 4c2 = 0 (III), and c6 + 3c5 + 4c3 = 0 (IV). From
(I) we have c3 = –3c2 – 4 (V), and from (II) we have c4 = –3c3 – 4c (VI). Substituting (V) into (VI), we get
c4 = 9c2 – 4c + 12 (VII). From (III) we get c5 = –3c4 – 4c2 (VIII) so that c5 = –31c2 + 12c – 36 [after substituting
(VII) into (VIII)]. Finally, by similar operations, we get c6 = –3c5 – 4c3 = 105c2 – 36c + 124 (IX).
The sum of (IX) and 10 times (VII) and 25c2 is, therefore,
Method II : P(c) = Q(c) (c3 + 3c2 + 4) + R(c) where Q(c) is the quotient obtained when P(c) is divided by
c3 + 3c2 + 4 and R(c), of degree less than 3, is the remainder. But c3 + 3c2 + 4 = 0. Therefore,
P(c) = R(c), which is found by actual division to be 220c2 – 76c + 244.
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

Ex.30 Find real values of x, y, z satisfying the equations

4x2 4y2 4z 2
= y, = z, = x.
1 + 4x 2 1 + 4y 2 1 + 4z2

Sol. The values of x, y, z are real.

Clearly x = 0, y = 0, z = 0 do not satisfy the equation
Þ x ¹ 0, y ¹ 0, z ¹ 0.
Multiply all the equations.

64x2 y 2 z 2
= xyz
(1 + 4x )(1 + 4y 2 )(1 + 4z 2 )
2

Þ 64xyz = (1 + 4x2) (l + 4y2) (1 + 4z2) (1)

20
 Mathematics
ALLEN
Clearly
(1 – 2x)2 ³ 0
Þ 1 + 4x2 ³ 4x
1 + 4y2 ³ 4y
1 + 4z2 ³ 4z
Þ (1 + 4x2) (1 + 4y2) (l + 4z2) ³ 64xyz
The equality holds when x = y = z. As per (1) the equality holds.
Þ x = y = z.

4x 2
Now = y.
1 + 4x 2
Þ 4x2 = y(1 + 4x2)
Þ 4x2 = x(1 + 4x2)
x ¹ 0 Þ 4x = 1 + 4x2

1 1 1
or (1 – 2x)2 = 0 or x = ,y= ,z = .
2 2 2
Ex.31 Let the roots of the cubic equation x2 + ax2 + bx + c = 0 be real. Show that the difference between the

greatest and the least of them is not less than a2 - 3b nor greater than 2 (a - 3b) .
2

Sol. The cubic equation is t3 + at2 + bt + c = 0.

Let a, b, g, be the real roots. Let a ³ b ³ g
Þ a is the greatest root and g is the smallest. To prove
(a2 – 3b)1/2 £ a – g £ 2(a2 – 3b)1/2
Let a – g = l, a – b = m Þ b – g = l – m. (*)
We have a+ b+ g= –a (1)
ab + bg + ga = b (2)
abg = – c (3)
2(a2 – 3b) = 2[(a + b + g)2 – 3(ab + bg + ga)]
= 2a2 + 2b2 + 2g2 – 2ab – 2bg – 2ga
= (a – b)2 + (b – g)2 + (g – a)2
= m2 + l2 + (l – m)2 ³ 0
2(a2 – 3b) = l2 + m2 + l2 + m2 – 2lm
= 2(l2 + m2) – 2lm

l 2 3l 2
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

= 2(m – ) + ³0
2 2
= m2 + l2 + (l – m)2 ³ 0
l
Minimum of the expression occurs when m =
2
maximum will occur at m = 0 or m = l.

3 2
Thus l £ 2(a2 – 3b) £ 2l2
2
2
Þ (a2 – 3b)1/2 £ l £ (a2 – 3b)1/2 < 2(a2 – 3b)1/2
3
Þ (a2 – 3b)1/2 £ a – g < 2(a2 – 3b)1/2

21
RMO/INMO
ALLEN
Ex.32 Let P(x) be a polynomial with integer coefficients. Prove that if P(0) and P(1) are odd, then P(x) = 0 has no
integer solutions.
Sol. If a is an integer with P(a) = 0, then P(x) = (x – a) Q(x), for some polynomial Q(x) with integer coefficients,
and P(0) = –aQ(0) and P(1) = (1 – a) Q(1). But if a is an integer, then either a or 1 – a is even, and so one
of P(0) and P(1) is even, which is a contradiction.
Ex.33 a,b are the roots of the quadratic equation x2 + px + q = 0 and for the equation x2n + pnxn + qn = 0 When

a b
n is an even integer. Prove that , are the roots of xn + 1 + (x + l)n = 0.
b a

Sol. a,b are the roots of the quadratic equation x2 + px + q = 0.

Þ a + b = –p (1)
ab = q (2)
a,b are also the roots of x2n + pnxn + qn = 0.
This equation has 2n roots out of which a, b are two roots.
The student should note that we cannot write a + b = –pn ab = qn.
a,b satisfy the equation because they are roots
a2n + pnan + qn = 0, b2n + pnbn + qn = 0
Subtract a2n – b2n + pn (an – bn) = 0
(an – bn) (an + bn) + pn = 0
Þ (an – bn) (an + bn + pn) = 0 (3)
From (3) a + b = –p.
(a + b)n = (–P)n = pn. Q n is an even integer. Þ (a + b)n = pn
(1) Þ an + bn + (a + b)n = 0
Dividing through out by bn , we get
n n
æaö æa ö
ç ÷ + 1 + ç + 1÷ = 0
èbø èb ø

a
Þ is a root of
b

xn + 1 + (x + l)n = 0

b
Similarly is also a root.
a
Ex.34 Prove that the polynomial x4 – x3 – 3x2 + 5x + 1 is irreducible over ¤ [x].
Sol. Since the coefficients of the polynomial are integers, it is enough to see that it is irreducible over ¢ [x]. The
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

polynomial has no integer roots, since a root of x4 – x3 – 3x2 + 5x + 1 must divide 1, and then it must be
1 or –1, but P(1) = 3 and P(–1) = –5. Or, because P(0) and P(1) are odd, P(x) = 0 has no integer solutions.
Hence, if P(x) can be factored, it must be into two monic quadratic polynomials, as follows: x4 – x3 – 3x2
+ 5x + 1 = (x2 + bx + c) (x2 + dx + e), with b, c, d, e integers.
Equating coefficients, it follows that :
b + d = –1
c + e + bd = –3
be + cd = 5
ce = 1.
The last identity implies c = e = 1 or c = e = –1. Now, the third identity takes the form b + d = 5 or b
+ d = – 5. In any case, these equalities are in contradiction with the first identity.

22
 Mathematics
ALLEN
Ex.35 Solve the system of equations :
a3 + 3ab2 + 3ac2 – 6abc = 1,
b3 + 3ba2 + 3bc2 – 6abc = 1,
c3 + 3cb2 + 3ca2 – 6abc = 1.
Sol. If A, B, C are the expressions on the left-hand sides of the equations, it follows that – A + B + C = (– a
+ b + c)3, A – B + C = (a – b + c)3, A + B – C = (a + b – c)3 and –A + B + C = A – B + C = A + B
– C = 1. The system is equivalent to –a + b + c = 1, a – b + c = 1, a + b – c = 1, which has the unique
solution (a, b, c) = (1,1,1).

EXERCISE
1
1. If x = 2016, y = 2017, z = find the value of (x + y + z)3 – (x + y – z)3 – (y + z – x)3 – (z + x – y)3 – 23xyz.
2017

2. Find all integral solutions of x3 + 5y3 + 25z3 – 15xyz = 0.

3. If f(x) = x4 + 3x3 + 9x2 + 12x + 20, and g(x) = x4 + 3x3 + 4x2 – 3x – 5, find the functions a(x), b(x) of smallest
degree such that a(x)f(x) + b(x)g(x) = 0.
4. Find m and solve the following equation, knowing that its roots form a geometrical sequence:
x4 – 15x3 + 70x2 – 120x + m = 0.
5. When x, y, z are real, solve the equations x + y = 2, xy – z2 = 1.
6. The roots of the polynomial x2 + ax + b + 1 = 0 are natural numbers. Prove that a 2 + b2 is not a prime number.
7. Solve x2 + y(x + 1) = 17, y2 + x (y + 1) = 13.
8. Factor x4 – 6x3 + 9x2 + 100 into quadratic factors with integral coefficients.
9. Find positive integers a, b such that a3 + a2b + ab2 + b3 + 1 = 2002.
10. Find all solutions of m2 – 3m + 1 = n2 + n – 1, where m and n are positive integers.
p
11. If a, b, c, p and q are integers, with q ¹ 0, (p, q) = 1 and is a root of the equation ax2 + bx + c = 0, Prove
q
that p divides c and q divides a.
12. If bd + cd is an odd integer, prove that the cubic polynomial (x3 + bx2 + cx + d) cannot be expressed in the
form (x + r)(x2 + px + q).
(a - b)(c - d) 5 (a - c)(b - d)
13. If = - ,find .
(b - c)(d - a) 3 (a - b)(c - d)
14. Find all solutions of the system of equations.
ax = (x + y + z)y, ay= (x + y + z)z, az = (x + y + z)x.
15. Let a, b, c be real numbers, with a and c non-zero. Let a and b be the roots of the polynomial
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

ax2 + bx + c and let a’ and b’ be the roots of the polynomial cx2 + bx + a. Prove that if a, b, a',b' are positive
numbers then (a + b)(a' + b') ³ 4.
16. Let a be a nonzero real number. Solve the equation: a3x4 + 2a2x2 + x + a + 1 = 0.
17. Solve the equations
x + y + z = 14
x2 + y2 + z2 = 91
y2 = zx.
18. Determine if there exist polynomials x2 – b1x + c1 = 0 and x2 – b2x + c2 = 0, with b1, c1, b2 and c2 different,
such that the four roots are b1, c1, b2 and c2.

23
RMO/INMO
ALLEN
19. Solve:
x2 + y2 + z2 = 14
xy + yz + zx = 11
x + y + 2z = 9
20. A student, required to solve the equation x2 + bx + c = 0, inadvertently solves the equation x2 + cx + b = 0;
b, c integers. One of the roots obtained is the same as a root of the original equation, but the second root is
m less than the second root of the original equation. Find b and c in terms of m.
21. Determine all real numbers a, such that all the solutions of the equation

a2 a2 a2 a2 a2
+ + + + = 1 are real
x(x + 1) (x + 1)(x + 2) (x + 2)(x + 3) (x + 3)(x + 4) (x + 4)(x + 5)

22. The solutions of the equation x3 + bx2 + cx + d =0 are p, q are r. Find a quadratic equation with roots
p2 + q2 + r2 and p + q + r, in terms of b, c, and d.
23. Find x5 + y5 + z5, it being given that
x + y + z = 1; x2 + y2 + z2 = 2;
x3 + y3 + z3 = 3.
24. Prove that, for n ³ 1, (x – 1)2 divides nxn+1 – (n + 1)xn + 1.
25. Find the solutions of the system
x+y+z=w

1 1 1 1
+ + = .
x y z w

26. Find the sum of the cubes of the roots of the equation x4 + ax3 + bx2 +cx + d = 0.
27. Show that if p, q, r, s are real numbers and pr = 2(q + s), then atleast one of the equation x2 + px + q = 0,
x2 + rx + s = 0 has real roots.
28. Find all real numbers a such that the sum of the squares of the roots of P(x) = x 2 – (a – 2)x – a – 1 is a minimum.
29. Solve the equations
x+y+z=9
x2 + y2 + z2 = 29
x3 + y3 + z3 = 99
30. Find all triples (x, y, z) of positive integers satisfying 2x + 2y + 2z = 2336.

æ 2m - 1 ö m2 - 3
For which positive real number m, the roots x1 and x2 of the equation x - ç x+ = 0, satisfy the
2
31. ÷
è 2 ø 2

1
condition x1 = x2 – ?
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

2
32. Let P(x) be a polynomial such that P(x2 + 1) = x4 + 4x2. Find P(x2 – 1).
33. Solve the following system of equations for real x, y, z. x + y – z = 4; x2 – y2 + z2 = – 4; xyz = 6.
34. Find the integral solutions of the equation x2 + y2 + x + y = 3.
35. Let a, b, c, d, e and f be positive integers such that they satisfy the relation ab + ac + bc = de + df + ef, and
let N = a + b + c + d + e + f. Prove that if N divides abc + def, then N is a composite number.
36. Find all pairs (x, y) of positive integers such that x3 – y3 = xy + 61.
37. Solve the equations
x3 + y3 + z3 = 495; x + y + z = 15; xyz = 105.
38. If a3 + b3 + c3 = (a + b + c)3 prove that a5 + b5 + c5 = (a + b + c)5.

24
 Mathematics
ALLEN
39. The natural numbers a, b, c and d satisfy that a3 + b3 = c3 + d3 and a + b = c + d. Prove that two of these
numbers are equal.
40. Let a, b, c be non zero real numbers such that (ab + bc + ca)3 = abc (a + b + c)3. Prove that a, b & c are
terms of a geometric sequence.
1 2
41.
2
Solve x + = 21- y where x, y are integers.
x2
42. Let a, b and c be real numbers. Prove that at least one of the following equations has a real solution.
x2 + (a – b)x + (b – c) = 0,
x2 + (b – c)x + (c – a) = 0,
x2 + (c – a)x + (a – b) = 0.
43. If a, b, c are the roots of x3 – x2 – x – 1 = 0. Prove that
(i) a, b, c are distinct.

a 2016 - b2016 b2016 - c 2016 c2016 - a 2016

(ii) + + is an integer..
a-b b-c c-a
44. If a, b, c are three consecutive odd integers such that a < b < c, find the value of a2 – 2b2 + c2.
45. Solve for x and y, x2 + y2 = (x + y + 1)2, x2 + y2 = (x – y + 2)2.
46. Find all pairs of integers (a, b) such that a + b is a root of the polynomial x2 + ax + b.
47. Find all integer values of x for which the polynomial x2 – 5x – 1 is a perfect square.
48. Let P(x) be a monic quadratic polynomial such that P(x) and P(P(P(x))) have a common root. Prove that
P(0)P(1) = 0.
49. If (x + 1)2 = x find thevalue of 11x3 + 8x2 + 8x – 2.
50. Solve xy + x + y = 23, yz + y + z = 41, zx + z + x = 27.
51. If f1 = 3x – y + 2z + w,
f2 = 2x + 3y – z + 2w,
f3 = 5x – 9y + 8z – w, find numberical values of a, b, c so that af1 + bf2 + cf3 = 0.
52. Find all pairs (x, y) where x, y are integers such that x3 + 113 = y3.
53. Do there exist integers m, n such that m2 + (m + 1)2 = n4 + (n + 1)4?
54. Solve in real numbers the system of equations

ìa + b = 8
ï
ïab + c + d = 23
í
ïad + bc = 28
ïîcd = 12

55. Fill the smallest value of p2 for which the pair of equations, (4 – p2)x + 2y = 0, 2x + (7 – p2)y = 0 has a solution
node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

other than x = y = 0, and find the ratio x : y for this value of p2.
56. Find x, y, z satisfying the equations
(x + y)(x + y + z) = 66
(y + z)(x + y + z) = 99
(z + x)(x + y + z) = 77
57. If p, q, r are the roots of the cubic equation x3 – 3px2 + 3q2x – r3 = 0, prove that p = q = r.
4mn
58. Find all solutions in integers m, n of the equation (m – n)2 = .
m + n -1
bx + (1 - x)c cx + (1 - x)a ax + (1 - x)b
59. If a, b, c, x are real numbers such that abc ¹ 0 and = = then, prove that,
a b c
a + b + c = 0 or a = b = c.

25
RMO/INMO
ALLEN
60. Let Nk = 131313...131 be the (2k + 1) digit number in base 10 formed by k pieces of 13 and appended by
1 at the end. Prove that Nk is not divisible by 31 for any value of k = 1, 2, 3 ...
61. Find all real solutions of the system of equations: x + y = 2; xy – z2 = 1.
62. Find all integers m, n such that 2mn – 5m + n = 55.
63. Find all real numbers a and b such that x2 + ax + b2 = 0 have at least one common root.
64. Prove that the product of the first 1000 positive even integers differs from the product of the first 1000 positive
odd integers by a multiple of 2015.
65. Find all pairs (x, y) where (x, y) are integers such that x3 + 113 = y3.
66. If the quadratic ax2 + bx + c takes rational values for more than two rational values of x, then, show that a,
b, c are all rational numbers.
67. If a ¹ 0, b ¹ 0, c ¹ 0 and if

1 1 1
+ + =0
a b a+x

1 1 1
+ + =0
a c a+y

1 1 1
+ + =0
a x y

Prove that a + b + c = 0.
68. Show that there are no integers a, b, c for which a2 + b2 – 8c = 6.
69. Let A = {a2 + 4ab + b2|a, b are positive integers}. Prove that 2015 ¹ A.
70. Find integers x, y, z such that
x2y + y2z + 4xy = 40
x2 + y2 + xyz = 20

node05\B0AI-B0\CBSE\RMO & INMO Module\Volume - 1\01-Algebra.p65

26