Rajeev Manocha Good Quality

www.pdfworld.
in
www.pdfworld.in
www.pdfworld.in
MATHEMATICS
OLYMPIAD
Conducted by Homi Bhabha Centre for Science Education
With
SOLVED PAPERS
RMO & INMO
2016-2019
RAJEEV MANOCHA
ARIHANT PRAKASHAN (SERIES), MEERUT
www.pdfworld.in
www.pdfworld.in
Arihant Prakashan (Series), Meerut

All Rights Reserved
© AUTHOR
No part of this publication may be re-produced, stored in a retrieval system or
distributed in any form or by any means, electronic, mechanical, photocopying,
recording, scanning, web or otherwise without the written permission of the
publisher. Arihant has obtained all the information in this book from the sources
believed to be reliable and true. However, Arihant or its editors or authors or
illustrators don’t take any responsibility for the absolute accuracy of any
information published and the damages or loss suffered there upon.
All disputes subject to Meerut (UP) jurisdiction only.
ADMINISTRATIVE & PRODUCTION OFFICES

Regd. Office
‘Ramchhaya’ 4577/15, Agarwal Road, Darya Ganj, New Delhi -110002
Tele: 011- 47630600, 43518550; Fax: 011- 23280316
Head Office
Kalindi, TP Nagar, Meerut (UP) - 250002
Tele: 0121-2401479, 2512970, 4004199; Fax: 0121-2401648
SALES & SUPPORT OFFICES

Agra, Ahmedabad, Bengaluru, Bareilly, Chennai, Delhi, Guwahati,
Hyderabad, Jaipur, Jhansi, Kolkata, Lucknow, Meerut, Nagpur & Pune
ISBN : 978-93-13193-59-3
Published by Arihant Publications (India) Ltd.

For further information about the books published by Arihant
log on to www.arihantbooks.com or email to info@arihantbooks.com
/arihantpub /@arihantpub Arihant Publications /arihantpub
www.pdfworld.in
www.pdfworld.in
PREFACE
Since the year 1989 when India started participating in the Mathematical Olympiads, the
interest of the school students in the country has tremendously increased in India National
Mathematics Olympiad (INMO). Now, we find a number of really talented students in almost
all the prestigious schools in India who are excited about this mega event and sincerely
desire to participate in it, win a handful of medals and make the country proud. We just need
to educate them about the competition, and provide them the relevant study material.
The Mathematical Olympiad tests the participant's level of mastery of the methods of
Mathematics and the strategizing and tactical skills in plenty. The Olympiad is an open
challenge to all those who love the problem solving.
This book has been written, keeping in mind the orientation required on the parts of
students to face the Olympiads at national or regional level. This book has designed to give
the student an insight and proficiency into almost all the areas of Mathematics. Exhaustive
theory has been provided of selected and relevant chapters to clarify the basic concepts.
Problems from recently held Olympiads have been given to increase awareness of what to
expect in the event.
Revised Edition of This Book Has
1. Complete theory with support of good number of solved examples and exactly on the
pattern and level of Indian National Mathematics Olympiads.
2. Each chapter has two level exercises divided according to RMO Regional Mathematics
Olympiad and INMO (Indian National Mathematical Olympiad)
3. Solutions have been provided for selected questions.
First of all, I would like to thank Mr Deepesh Jain, Director Arihant Group the man with a
distinct vision, for the idea to write this book, and then bringing it to reality. I am also
thankful to my colleagues and students for the moral support they provided. I take this an
opportunity to thank Sunil Chugh, Director, HMA, for the inspiration to write the book of this
nature, and Sumit Malviya for the assistance he provided in the preparation of the
manuscript.
It is hoped, this book will charge you up for the Olympiad juggernaut. I have tried my best to
keep this book error-free. However, if any error or whatsoever is left I request the readers to
bring forward to my notice. Suggestions for the further improvement of the book are
welcome.
With best wishes
Rajeev Manocha
www.pdfworld.in
www.pdfworld.in
CONTENTS
1. Theory of Numbers 1-94

2. Theory of Equations 95-216
3. Inequalities 217-356
4. Combinatorics 357-485
5. Geometry 486-583
6. Functions 584-616
— RMO & INMO 2016-2017 1-8

— RMO & INMO 2017-2018 1-8
— RMO & INMO 2018-2019 1-7
Dedicated to My Parents
I.P.F. Manocha, Bimla Manocha
& My sons Robit and Sarthak
www.pdfworld.in
www.pdfworld.in
INTRODUCTION
MATHEMATICS OLYMPIADS
ABOUT THE EXAM

& HOW TO SUCCEED IN IT ?
International Mathematical Olympiad
International Mathematical Olympiads was created by Dr. George Lenchner, a prominent Maths
educator, in 1977 with the aim of stimulating enthusiasm and love for Mathematics. About 1,50,000
students from 50 states of USA and 25 other countries participate in this competitive examination
every year. Through this examination, effort has been made to introduce important mathematical
concepts and teach major strategies for problem solving. The examination also seeks to foster
Mathematical creativity and ingenuity and provide satisfaction, joy and thrill through meeting
challenges.
Every participant team comprises 35 students. Some schools send more than one team in the
contest. However, the rule indicates that only schools or home-school associations may participate
individuals are barred from entering into the contest.
Every team enters into competition in just one division. Those teams which have members from
more than one school are called 'district teams' or 'institute teams’ and they are not eligible for team
awards. The team score is the total of ten highest individual scores taken after the fifth contest in a
series of contests which involve selection.
The International Mathematics Olympiad invites students from 2nd to 12th class to participate and
excel at international level.
Indian National Mathematical Olympiads

In India, National Board for Higher Mathematics (NBHM) started National Mathematics Olympiads in
1986. It worked with Homi Bhabha Centre for Science Education, Mumbai, for this purpose. One aim
of this activity is to support the mathematical talent among the high school students in the country.
NBHM take the responsibility of selecting, training and grooming of Indian team for participation in
the International Mathematical Olympiad every year. There are certain regional bodies which
provide voluntary service and play an important role at various stages. The country has been divided
into 23 regions for conducting Mathematical Olympiads.
Stages
Stage I Regional Mathematical Olympiad (RMO)
A regional coordinator conducts tests in each region. Regional Mathematical Olympiads (RMO) is
held in each region between September and first Sunday of December every year. The regional
coordinator makes sure that at least one centre in each district of the region is provided for the test.
All the students of high school upto Class XII can appear at Regional Mathematical Olympiads. The
qualifying test is of three hours and consists of six to seven problems.
www.pdfworld.in
www.pdfworld.in
The regional coordinator has the liberty of making his own paper or obtaining paper from NBHM.
The regions which choose to go for paper by NBHM conduct this contest on the first Sunday of
December. The performance in RMO is judged and a certain number of students from every region
are chosen to appear for the next round. There is a nominal fee to meet the expenses of the
organising the contest.
Stage II Indian National Mathematical Olympiad
The second stage of selection involves holding of the contest at the national level. Indian National
Mathematical Olympiads is (INMO) organized on the first Sunday of every February at various centres in
different regions. Only those students who have been selected in Regional Mathematical Olympiads can
appear at this contest. At this level, there is a 4-hour test which is common to all regions. Those who rank
among the top 35, receive a certificate of merit.
Stage III International Mathematical Olympiad Training Camp (IMOTC)
The INMO awardees are invited to a month long training camp held in April-May each year at the
Homi Bhabha Centre for Science Education (HBCSE), Mumbai. INMO awardees of the previous year
who have satisfactorily gone through postal tuition throughout the year are invited again to a
second round of training (Senior Batch). The senior batch participants who successfully complete the
camp receive a prize of 5,000/- in the form of books and cash. On the basis of a number of selection
tests through the camp, a team of the best six students is selected from the combined pool of junior
and senior batch participants.
Stage IV Pre-departure Training Camp for IMO
The selected team of six students goes through another round of training and orientation for about
10 days prior to the departure for IMO.
Stage V International Mathematical Olympiad (IMO)
The six-members team selected at the end of the camp accompanied by a leader and a deputy
leader represent the country at the IMO, held in July each year in a different member country of IMO.
IMO consists of two 4 and 1/2-hour written tests held on two days. Travel to IMO venue and return
takes about 2 weeks. India has been participating in IMO since 1989. Students of the Indian team
who receive gold, silver and bronze medals at the IMO receive a cash prize of 5000/-, 4000/- and
3000/- respectively during the following year at a formal ceremony at the end of the training camp.
Ministry of Human Resource Development (MHRD) finances international travel of the 8-members
Indian delegation, while NBHM (DAE) finances the entire in-country programme and other
expenditure connected with international participation.
Awards and Recognition

— Participation Certificate to every student | Merit Certificate for students appearing in second
level School Topper Medal to the class topper in each participating school (with more than 50
students).
— Top three students from each class are awarded Gold, Silver and Bronze medals.
— The top 500 winners (classwise) are endowed with cash prizes and scholarships, courtesy of the
official sponsor.
— Every participant who scores 50% or more is awarded a Certificate of Participation. Toppers are
awarded merit certificate.
— School/College of each medal winning participant is awarded the 'Intellect Trophy' for having
groomed and nurtured the talent.
www.pdfworld.in
www.pdfworld.in
Career Prospects
Mathematical Olympiads were created to hunt out the talented students and to give them an
opportunity to express their talent in creative and intuitive way. It reads well on a student's
curriculum vitae to have participated in Mathematical Olympiads. Those who have participated in
the Olympiads have a good chance of doing well in Engineering.
The top few students from Mathematical Olympiads are selected for admission to premier
institutions. The career of the students takes up an upwards mobility curve after participating in this
contest. The students become confident of their performance and rich with the feeling of self-worth
after selection to represent at Mathematical Olympiads.
Number of Students Appearing

— Approximately 10,000 students appear for this examination at regional level, though the number
is only eight when they represent the country at international level.
— Students from rural and urban areas alike participate in the regional competition. This
competition is open for all the students from Class IX onwards.
Month and Frequency of Examination

International Mathematical Olympiads is held once every year in July. The qualifying rounds of
Indian National Mathematics Olympiads is held in different months of the years starting from first
Sunday of December.
Craze of the Examination

Mathematical Olympiads, whether they are held at national level or international level, attract
interest not only from the mathematicians but also from the student community, the parents and
society at large. Even those students who may not be participating in the contest are interested in it.
The students who participate in the contest start preparing for it at least three months prior to the
date of examination. The craze for the examination is not limited only to a certain region, the verve
engulfs the entire country. Those who are selected are viewed as the promising stars of future who
will contribute to the progress of science and technology in years to come.
General Eligibility
The students must be from Class IX and must be at least 14 years of age to participate in this contest
both at national and international levels. They must pass the different rounds of selection in order to be
eligible.
At the first round, all students who are in Class IX are eligible to appear for the qualifying test. In the
subsequent rounds, however, only those whom make it in the test would be eligible to carry forth.
Skill Sets Required

In Mathematics, the skill most required is ability to think fast, be spatially gifted and compute
complex data without committing error. The students must develop the habit of calculating data
mentally and simplify the methods in order to arrive at the correct answer with minimum effort.
Another very important part is concentration; lack of concentration can cause the candidates to
commit errors, so sufficient effort must be put into develop the power of concentration.
Difficulty Level of Problems

The problems of Olympiads, chosen from various areas of secondary school Mathematics, require
exceptional mathematical ability and mathematical knowledge on the part of the contestants.
Generally, there are six to seven questions asked and they are of extremely high level. To solve these
questions, the students need to be thorough with their basics and have good knowledge of all the
principles of Mathematics. Since two or three principles from different mathematical streams must
be applied to solve the questions, the problems assume a sophistication which requires an agile and
quick thinking mind with plenty of common sense to answer them.
www.pdfworld.in
www.pdfworld.in
How to Prepare for Different Subjects in Stipulated Time

The best way of dealing with this contest is to cultivate scientific thinking and to develop power of
concentration. The students must try to deal with two or three mathematical principles
simultaneously so that they become adept at solving complex problems which require application
of higher mental faculty. There are many books in the market but Arihant books are the best as they
help in development of scientific temper along with providing reading material and solution.
General Mental Set up Required for the Examination

The candidates must have a positive frame of mind and they must be able to deal with stress that
comes as a package with these examinations. They must have the desire to succeed and ability to
work for long lasting success. The candidates must be well versed in all the fundamentals of the
principles of Mathematics. They must be original thinker and must have a scientific temperament.
The power of concentration of the candidates must be of exceptionally high order. The calculation
skill of the candidates must be extremely good since all Maths sums are based on reasoning and
calculation.
Do’s and Don’t on the Day of Examination

On the day of the examination, the candidates must take care about a few things which are listed
below:
— The candidates must reach the venue of the examination at least half an hour before time.
— The candidates must carry their admit cards to the examination hall.
— The candidates must carry their own pens, pencils, erasers, sharpeners and must refrain from
borrowing these articles from the other candidates.
— The candidates must abstain from talking to other candidates in the examination hall while the
examination is being conducted.
— The candidates must hand over their answer sheets to the invigilator as soon as the stipulated
time is over.
How this Book is Useful for You

This book by Arihant is the best in the market for purpose of preparation. It has many complex
problems and tricky questions requiring sophisticated methods of calculation.
This book coaches the students to think in multidimensional way and apply the principles of
Mathematics to solve the most complex problems in the simplest way. It also gives tremendous
exposure to different types of problems which exist in the realm of Mathematics.
This book has matter for study by the students and solved problems which help in understanding of
the subject itself. It also has previous years papers from Regional and International Olympiads along
with their solutions. It is an excellent book to have which will prove to be the best friend to the
contestants.
www.pdfworld.in
www.pdfworld.in
Unit 1
Theory of Numbers
Natural Numbers
The numbers 1, 2, 3, …, which are used in counting are called natural numbers or positive integers.
Basic Properties of Natural Numbers

In the system of natural numbers, we have two ‘operations’ addition and multiplication with the
following properties.
Let x , y , z denote arbitrary natural numbers, then
1. x + y is a natural number i.e., the sum of two natural number is again a natural number.
2. Commutative law of addition x + y = y + x
3. Associative law of addition x + ( y + z ) = ( x + y ) + z
4. x ⋅ y is a natural number i.e., product of two natural numbers is a natural number.
5. Commutative law of multiplication x ⋅ y = y ⋅ x
6. Associative law of multiplication x ⋅ ( y ⋅ z ) = ( x ⋅ y ) ⋅ z
7. Existence of multiplicative identity a ⋅ 1 = a
8. Distributive law x ( y + z ) = xy + xz
Divisibility of Integers
An integer x ≠ 0 divides y, if there exists an integer a such that y = ax and thus we write as x | y (x divides y).
If x does not divides y, we write as x \| y (x does not divides y )
[This can also be stated as y is divisible by x or x is a divisor of y or y is a multiple of x].
Properties of Divisibility
1. x | y and y | z ⇒ x | z
2. x | y and x | z ⇒ x | (ky ± lz ) for all k , l ∈ z. z = set of all integers.
3. x | y and y | x ⇒ x = ± y
4. x | y, where x > 0, y > 0 ⇒ x ≤ y
5. x | y ⇒ x | yz for any integer z.
6. x | y iff nx | ny, where n ≠ 0.
www.pdfworld.in
www.pdfworld.in
2 Indian National Mathematics Olympiad
Test of Divisibility
Divisibility by certain special numbers can be determined without actually carrying out the process of
division. The following theorem summarizes the result :
A positive integer N is divisible by
• 2 if and only if the last digit (unit's digit) is even.
• 4 if and only if the number formed by last two digits is divisible by 4.
• 8 if and only if the number formed by the last three digits is divisible by 8.
• 3 if and only if the sum of all the digits is divisible by 3.
• 9 if and only if the sum of all the digits is divisible by 9.
• 5 if and only if the last digit is either 0 or 5.
• 25 if and only if the number formed by the last two digits is divisible by 25.
• 125 if and only if the number formed by the last three digits is divisible by 125.
• 11 if and only if the difference between the sum of digits in the odd places (starting from right) and
sum of the digits in the even places (starting from the right) is a multiple of 11.
Division Algorithm
For any two natural numbers a and b, there exists unique numbers q and r called respectively quotient
and remainder, a = bq + r , where 0 ≤ r < b.
Common Divisor
If a number ‘c’ divides any two numbers a and b i.e., if c | a and c | b, then c is known as a common divisor
of a and b.
Greatest Common Divisor

If a number d divides a and b and is divisible by all the common divisors of a and b, then d is known as
the greatest common divisor (GCD) of a and b or HCF of a and b.
The GCD of numbers a and b is the unique positive integer d with the following two properties.
(i) d | a and d | b
(ii) If c | a and c | b; then c | d
We write it as (a , b ) = d
For example, (12, 15) = 3 ; (7, 8) = 1
Note 1. (a, b ) = (b, a ) 2. If a|b; then (a, b ) = a
Properties of GCD
1. If ( b , c ) = g and d is a common divisor of b and c, then d is a divisor of g.
2. For any m > 0, (mb , mc ) = m (b , c )
3. If d |b and d |c and d > 0, then 

b c  1
,  =   (b , c )
d d  d 
b c 
4. If ( b , c ) = g, then  ,  =1
g g 
5. If ( b , c ) = g, then there exists two integers x and y such that g = xb + yc
6. If (a , b ) = 1 and (a , c ) = 1, then ( a , bc ) = 1
7. If a | bc = 1 and (a , b ) = 1, then a | c = 1
www.pdfworld.in
www.pdfworld.in
Theory of Numbers 3
For example a = 6, b = 21, c = 10

6 | 21 × 10 but (6, 21) = 3
and (6, 10) = 2 and 6 divides neither 21 nor 10
LCM of two integers a, b is the smallest positive integer divisible by both a and b and it is denoted by
[a, b].
The Euclidean algorithm can be used to find the GCD of two integers as well as representing GCD as
in 5th property. Consider 2 numbers 18, 28
28 = 1.8 + 10 ; 18 = 110
. + 8
10 = 1 ⋅ 8 + 2 ; 8 = 4 ⋅ 2 + 0
(18, 28) = 2
(18, 28) = 2 = 10 − 1.8 = 10 − (18 − 110
. )
= 210
. − 118
. = 2 (28 − 118
. ) − 118
.
= 228
. − 318
. = 228
. + (− 3)18
Note The representation in 5th property is not unique. In fact we can represent (a, b ) as xa + yb in infinite
number of ways where x , y ∈ Z
Z = set of all integers.
In above example, 252 is LCM of 18 and 28.
252 = 9.28
252 = 1418
.
(18, 28) = 2 . 28 + ( −3) ⋅18
= 2.28 + 252K + ( −3)18 − 252K
= ( 2 + 9K )28 + ( −3 − 14K )18
where K is any integer.
Unit
1 is called unit in the set of positive integers.
Prime
A positive integer P is said to be prime, if
(i) P > 1
(ii) P has no divisors except 1 and P i.e., A number which has exactly two different factors, itself and
one, is called a prime number.
Thus, 2, 3, 5, 7, 11, ... are primes. 2 is the only even number which is prime. All other primes
being odd.
But the converse is not true i.e., every odd number need not be prime.
Composite
Every number (greater than one) which is not prime is called composite number. i.e., a number which has
more than two different factors is called composite. For example 18 is a composite number because 2, 3,
6, 9 are divisors of 18 other than 1 and 18.
We can also define a composite number as : A natural number n is said to be composite, if there exists
integers l and m such that n = lm, where 1 < l < n and l < m < n.
www.pdfworld.in
www.pdfworld.in
Remark
l A prime number P can be written as a product only in one way namely P.1 .
l A composite number n can also be written as n.1. But composite number can be written in one more way
also as mentioned above.
l A composite number has at least three factors.
Note 1 is neither prime nor composite.
Twin Primes
A pair of numbers is said to be twin primes, if they differ by 2.
e.g., 3, 5 are twin primes.
Perfect Number
A number n is said to be perfect if the sum of all divisors of n (including n) is equal to 2n.
For example 28 is a perfect number because divisors of 28 are 1, 2, 4, 7, 14, 28.
Sum of divisors of n (= 28) = 1 + 2 + 4 + 7 + 14 + 28 = 56 = 2n
Coprime Integers
Two numbers a and b are said to be coprime, if 1 is only common divisors of a and b.
i.e., if GCD of a and b = 1 i.e., if (a , b ) = 1
e.g., (4, 5) = 1, (8, 9) = 1.
Theorem 1 If a = qb + r , then (a , b ) = (b , r ).
Proof Let (a , b ) = d and (b , r ) = e

Q (a , b ) = d ∴ d | a and d | b
∴ d | a and d | qb ∴ d | (a –qb)
i.e., d |r [Qa − qb = r ]
∴ d is common divisor of b and r.
∴ d |e [Qe is the GCD of b and r] …(i)
Again Q (b, r) = e
e e
∴ e |b and e |r ∴ and
bq r
∴ e|bq + r i.e., e|a [Qa = bq + r ]
∴ e is a common divisor of a and b.
∴ e|d [Qd is the GCD of a and b] …(ii)
From Eqs. (i) and (ii), we have d = e
i.e., (a , b ) = (b , r )
Remark
The above result can also be stated as :
GCD of a and b is same as GCD of b and r, where r is remainder obtained on dividing a by b.
Corollary If (a , b ) = 1, then (b , r ) = (a , b ) = 1
i.e., if a is coprime to b, then r is coprime to b. r is remainder obtained on dividing a by b.
www.pdfworld.in
www.pdfworld.in
Theory of Numbers 5
Theorem 2 If d is the greatest common divisor of a and b, then there exists integers x and y such that
d = xa + yb and d is the least positive value of xa + yb.
Proof
Case I By successive application of division algorithm to numbers a and b.
Let r1 , r2 , K , rn be successive remainders.
Therefore, a = bq1 + r1 , 0 < r1 < b (Dividing a by b)
b = r1q 2 + r2 , 0 < r2 < r1 (Dividing b by remainder r1 )
r1 = r2q3 + r3 , 0 < r3 < r2 (Dividing r1 by remainder r2 )
M
M
M
rn −2 = rn − 1qn + rn , 0 < rn < rn −1
rn −1 = rn qn +1 + rn + 1, 0 < rn +1 < rn
Since r1 > r2 > K is a set of decreasing integers, this process must terminate after a finite number of step.
i.e., remainder must be zero after some stage.
So let rn +1 = 0 ∴ rn −1 = rn qn +1 + 0 = rn qn +1
rn | rn −1 ∴ (rn − 1 , rn ) = rn [If a | b, then (a, b ) = a ]
Now, (a, b ) = (b, r1 ) = (r1 , r2 ) = (r2 , r3 ) = K = (rn − 1 , rn ) = rn …(i)
i.e., GCD of a and b is rn .
From first of above equations r1 = a − bq1 = ax 1 + by1, where x 1 = 1, y1 = − q ,
Putting the value of r1 = a − bq1 in r2 = b − r1q 2
r2 = b − r1q 2 = b − (a − bq1 ) q 2 = b − aq 2 + bq1q 2
= − aq 2 + b (1 + q1q 2 ) = ax 2 + by 2 , where x 2 = − q 2
y 2 = 1 + q1q 2
Similarly, r3 = ax 3 + by3 and so on rn = ax n + byn
or rn = ax + by
where x n = x and yn = y
i.e., GCD of a and b = rn can be expressed as (a , b ) = d = ax + by [By Eq. (i)]
Case II (a , b ) = d (Q d | a and d | b)
∴ d | (ax + yb ) for all values of x and y.
∴ ∃ an integer k such that xa + yb = kd …(ii)
But least value of k is 1.
Putting k = 1 in Eq. (ii), least value of xa + yb is d.
Corollary If a and b are coprime integers i.e., if (a , b ) = 1, then there exists integers x and y such that
ax + by = 1
Example 1 If (a, b ) = d , then  ,  = 1.

a b
d d
Solution Q (a, b ) = d
∴ d | a and d | b [By definition of GCD].
∴ There exists integers a1and b1 such that a = da1 …(i)
b = db1 …(ii)
Again Q (a, b ) = d
www.pdfworld.in
www.pdfworld.in
There exists integers x and y such that

ax + by = d [By Theorem 2]
Putting values of a and b from Eqs. (i) and (ii)
da1x + db1y = d .
or a1x + b1y = 1
(a1, b1) = 1 [By corollary theorem 1]
 a b  a b
or  ,  =1 QFrom Eq. (i), a1 = d , from Eq. (ii), b1 = d 
d d
Remark
(a, b ) = d and a = a1d , b = b1d , then a1 and b1 are coprime i.e., (a1, b1) = 1
Example 2 If a|bc and (a, b ) = 1, then a|c.

Solution Q a|bc
∴ There exists an integer d such that bc = ad …(i)
Q (a, b ) = 1
∴ There exist integers m and n such that am + bn = 1 …(ii)
Multiplying both sides of Eq.(ii) by c, acm + bcn = c …(iii)
Putting bc = ad from Eq. (i) in Eq. (iii)
acm + adn = c
a(cm + dn ) = c
∴ a|c
Note If a|bc and (a, b ) = 1, then a|c. This result is also known as Gauss Theorem.
Example 3 Prove that every two consecutive integers are coprime.

Solution Let n and n + 1 be two consecutive integers.
Let (n, n + 1) = d
∴ d |n and d |n + 1
d |(n + 1) − n or d |1
∴ d =1
∴ (n, n + 1) = 1
i.e., n and (n + 1) are relatively prime.
Example 4 Show that GCD of a + b and a − b is either 1 or 2, if (a, b ) = 1.

Solution Let (a + b, a − b ) = d
∴ d | (a + b ) and d | (a − b )
∴ d | (a + b + a − b ) and d | (a + b ) − (a − b )
or d | 2a and d | 2b.
i.e., d is a common divisor of 2a and 2b.
∴ d | ( 2a, 2b ) [By definition of GCD]
www.pdfworld.in
www.pdfworld.in
Theory of Numbers 7
i.e., d | 2(a, b ) [Q(ma, mb ) = m (a, b )]

But (a, b ) = 1 ∴ d|2
∴ d = 1 or d = 2 .
Example 5 Find GCD of 858 and 325 and express it in the form m 858 + n 325.
Solution 858 = 325.2 + 208 …(i)
Dividing 858 by 325
325 = 2081
. + 117 …(ii)
Dividing 325 by 208
208 = 1171
. + 91 …(iii)
Dividing 208 by 117
117 = 911
. + 26 …(iv)
Dividing 117 by 91
91 = 26.3 + 13 …(v)
Dividing 91 by 26
26 = 13.2
∴ GCD of 858 and 325 is d = 13
From Eq. (v), d = 13 = 91 − 26.3
Substituting the value of 26 from Eq. (iv)
⇒ 91 − 3(117 − 911
. ) = 91 − 3117
. + 3.91 = 4.91 − 3117
.
Substituting the value of 91 from Eq. (iii)
= 4( 208 − 117) − 3117
. = 4.208 − 7117
.
Substituting the value of 117 from Eq. (ii)
= 4.208 − 7( 325 − 2081) = 4.208 − 7.325 + 7.208
= 11.208 − 7.325
= 11( 858 − 325.2) − 7.325 [Putting the value of 208 from Eq. (i)]
= 11.858 − 22.325 − 7.325 = 11.858 − 29.325
= m 858 + n.325 where m = 11, n = − 29
Example 6 If a and b are relatively prime, then any common divisor of ac and b is a divisor of c.
Solution a and b are relatively prime
∴ ∃ integers x and y such that
ax + by = 1 …(i)
Let d be any common divisor of ac and b.
Q d | ac, ∴ ∃ an integer m such that
ac = dm …(ii)
Q d | b, ∴ ∃ an integer n such that
b = dn …(iii)
Multiplying both sides of Eq. (i) by c
acx + bcy = c …(iv)
www.pdfworld.in
www.pdfworld.in
Putting the values of ac and b from Eqs. (ii) and (iii) in Eq. (iv).
dmx + dncy = c
or d (mx + ncy ) = c
∴ d|c
Example 7 If a and b are any two odd primes, show that (a 2 − b 2 ) is composite.
Solution a 2 − b 2 = (a − b )(a + b )
Qa.0 and b are odd primes.
So, let a = 2k + 1
b = 2k ′ + 1
∴ a − b = 2k + 1 − 2k ′ − 1 = 2k − 2k ′ = 2(k − k ′ ) is even
a + b = 2k + 1 + 2k ′ + 1 = 2k + 2k ′ + 2 = 2(k + k ′ + 1) is even
∴ Neither (a − b ) nor (a + b ) is equal to 1.
∴ Neither of the two divisors (a − b ) and (a + b ) of (a 2 − b 2 ) is equal to 1.
∴ (a 2 − b 2 ) is composite.
[QOut of the two divisors of a prime number p, one must be equal to 1]
Example 8 If a |c, b |c and (a, b ) = 1, then ab |c.

Solution Q a |c
∴There exists an integers d such that
c = ad …(i)
Q b|c
∴ There exists an integer e such that
c = be
Q (a, b ) = 1, therefore there exist integers m, n such that
am + bn = 1 …(ii)
Multiplying both sides by c
acm + bcn = c …(iii)
Putting c = be from Eq. (ii) in acm and c = ad from Eq. (i) in bcn, Eq. (iii) becomes
abem + band = c
ab (em + dn ) = c
∴ ab | c
Example 9 If a 2 − b 2 is a prime number, show that a 2 − b 2 = a + b, where a, b are natural numbers.

Solution a 2 − b 2 = (a − b )(a + b ) …(i)
Q (a − b ) is a prime number
2 2
∴ One of the two factors = 1

Q a − b =1 [Qa − b < a + b ]
www.pdfworld.in
www.pdfworld.in
Theory of Numbers 9
Q The only divisor of a prime number are 1 and itself.

Eq. (i) becomes a 2 − b 2 = 1 (a + b )
or a 2 − b2 = a + b
e.g., 32 − 22 = 5 (which is prime)
⇒ 32 − 22 = 3 + 2, 3, 2 ∈N.
Example 10 Prove that an integer is divisible by 9 if and only if the sum of its digits is divisible by 9.
Solution Let a = an K a 3a 2a1 be an integer
[Note a is not the product of a1, a 2, a 3,…, an but a1, a 2, a 3,…,an are digits in the value of
a. For example 368 is not the product of 3, 6 and 8 rather 3, 6, 8 are digits in value
of 368
= 8 + 6 × 10 + 3 × (10)2 ]
a = an …a 3a 2a1
= a1 + (10)1a 2 + (10)2a 3 + (10)3a 4 + K + (10)n − 1an
= a1 + 10a 2 + 100a 3 + 1000a 4 + K
= a1 + (a 2 + 9a 2 ) + (a 3 + 99a 3 ) + (a 4 + 999a 4 ) + ...
= (a1 + a 2 + a 3 + a 4 + K ) + ( 9a 2 + 99a 3 + 999a 4 + K ) …(i)
or a = S + 9 (a 2 + 11a 3 + 111a 4 + ... )
where S = a1 + a 2 + a 3 + a 4 + K
is the sum of digits in the value of a
∴ a − S = 9 (a 2 + 11a 3 + 111a 4 + ... )
∴ 9 | (a − S ) …(ii)
Case I a is divisible by 9
i.e., 9 |a …(iii)
∴ 9 | [a − (a − S )] [From Eqs. (ii) and (iii)]
i.e., 9|S i.e., sum of digits is divisible by 9.
Case II S (sum of digits) is divisible by 9
i.e., 9|S …(iv)
From Eqs. (ii) and (iv), 9|[(a − S ) + S ]
i.e., 9|a
i.e., the integer a is divisible by 9.
Theorem 3 Prove that the product of any r consecutive numbers is divisible by r ! .

Proof Let
Pn = n (n + 1)(n + 2) K (n + r − 1) …(i)
be the product of r consecutives integers beginning with n.
We shall prove the theorem by Induction method.
For r = 1, Pn = n is divisible by 1! for all n.
∴ The theorem is true for r = 1 i.e., the product of 1 (consecutive) integer is divisible by 1!.
Let us assume the theorem to be true for the product of (r − 1) consecutive integers.
www.pdfworld.in
www.pdfworld.in
i.e., every product of (r − 1) consecutive integers is divisible by (r − 1)! .

Changing n to n + 1 in Eq. (i)
∴ Pn +1 = (n + 1)(n + 2)K (n + r )
Multiplying both sides by n
nPn +1 = n (n + 1)(n + 2) K (n + r )
= n (n + 1)(n + 2) K (n + r − 1)(n + r )
nPn +1 = (n + r )Pn
= nPn + rPn
or n (Pn +1 − Pn ) = r ⋅ Pn
P
or Pn + 1 − Pn = r ⋅ n
n
n (n + 1)(n + 2) K (n + r − 1)
=r ⋅ [Using value of Pn ]
n
or Pn +1 − Pn = r (n + 1)(n + 2) K (n + r − 1)
or Pn +1 − Pn = r
Product of (r − 1) consecutive integers.
= rP …(i)
Where P denotes the product of (r − 1) consecutive integers.
But the product P of (r − 1) consecutive integers is divisible by (r − 1)! . [By assumption]
∴ P = k (r − 1)!
∴ Eq. (i) becomes Pn +1 − Pn = rk (r − 1)! = kr (r − 1) ! = k (r )!
i.e., r ! | (Pn +1 − Pn ), ∀ n
Put n =1
∴ r !| (P2 − P1 )
But P1 = 1 ⋅ 2 ⋅ 3 K r = r ! is divisible by r !
i.e., r1 !| P1
∴ r !| (P2 − P1 ) + P1 i.e., r !| P2
Put n = 2,
∴ r !| (P3 − P2 )
But r !| P2
∴ r !| (P3 − P2 ) + P2
i.e., r !| P3 and so on.
Generalising we can say that r !| Pn for all n.
n
Corollary Cr is an integer
n! n (n − 1)(n − 2) K (n − r + 1)(n − r )!
n
Cr = =
r !(n − r )! r! (n − r )!
n (n − 1)(n − 2) K (n − r + 1)
=
r!
the product of r consecutive integers
= = An integer
r!
(QThe product of r consecutive integer is divisible by r !) .
Note Therefore the product of two consecutive integers is divisible by 2! = 2 ; the product of any three
consecutive integers is divisible by 3 = 6! and so on.
www.pdfworld.in
www.pdfworld.in
Theory of Numbers 11
Example 1 Prove that product of two odd numbers of the form 4n + 1 is of the form ( 4n + 1) .
Solution Let a = 4k + 1, b = 4k ′ + 1
be two numbers of the form ( 4n + 1)
∴ ab = ( 4k + 1)( 4k ′ + 1)
= 16kk ′ + 4k + 4k ′ + 1
= 4( 4kk ′ + k + k ′ ) + 1 = 4l + 1
(where l = 4kk ′ + k + k ′)
Which is in form ( 4n + 1) .
Example 2 Prove that square of each odd number is of the form 8j + 1 .
Solution Let n = 2m + 1 be an odd number .

n 2 = ( 2m + 1)2 = 4m 2 + 4m + 1
= 4m(m + 1) + 1
Now, m(m + 1) being product of two consecutive integers, is divisible by 2! = 2
∴ m(m + 1) = 2j
⇒ n 2 = 4( 2j ) + 1 = 8j + 1
Example 3(a) Show that sum of an integer and its square is even.
Solution Let n be any integer.
So, we have to prove that n 2 + n is even.
⇒ n 2 + n = n(n + 1) which is product of two consecutive numbers n and n + 1 and
hence divisible by 2! = 2
Hence, n 2 + n is an even number.
Example 3(b) If n is an integer. Prove that product n(n 2 − 1) is multiple of 6.

Solution n(n 2 − 1) = n(n − 1)(n + 1) = (n − 1)n(n + 1)
Which being the product of three consecutive integers is divisible by 3! = 6
∴ n(n 2 − 1) is divisible by 6.
i . e., n(n 2 − 1) is a multiple of 6.
Note If n is a multiple of p, we shall write
n = M ( p) .
Example 4 If r is an integer, show that r (r 2 − 1)( 3r + 2) is divisible by 24.

Solution r (r 2 − 1)( 3r + 2) = r (r − 1)(r + 1)( 3r + 2)
= (r − 1)r (r + 1){3(r + 2) − 4}
= 3(r − 1)r (r + 1)(r + 2) − 4(r − 1)r (r + 1)
(r − 1)r (r + 1)(r + 2) being the product of four consecutive integers is divisible by
4! = 24
∴ 3(r − 1)r (r + 1)(r + 2) is also divisible by 24.
www.pdfworld.in
www.pdfworld.in
Again (r − 1)r (r + 1) is divisible by 3! = 6

∴ 4(r − 1)r (r + 1) is also divisible by 4 ⋅ 6 = 24
∴ r (r 2 − 1)( 3r + 2) = 3(r − 1)r (r + 1)(r + 2) − 4(r − 1) r (r + 1)
is also divisible by 24.
Example 5 If m, n are positive integers, show that (m + n )! is divisible by m ! n ! .

(m + n )! 1⋅ 2 ⋅ 3 K m(m + 1)(m + 2) K (m + n )
Solution =
m!n ! m!n !
m !(m + 1)(m + 2) K (m + n )
=
m!n !
(m + 1)(m + 2) K (m + n )
=
n!
The product of n consecutive integers
=
n!
= An integer
∴ (m + n )! is divisible by m ! n !.
Example 6 If ( 4x − y ) is a multiple of 3, show that 4x 2 + 7xy − 2y 2 is divisible by 9.
Solution Q 4x − y is a multiple of 3.
∴ 4x − y = 3m
∴ y = 4x − 3m
On putting value of y in 4x 2 + 7xy − 2y 2
= 4x 2 + 7x ( 4x − 3m ) − 2 ( 4x − 3m )2
= 4x 2 + 28x 2 − 21xm − 2 (16x 2 + 9m 2 − 24xm )
= 4x 2 + 28x 2 − 21xm − 32x 2 − 18m 2 + 48xm
= 27mx − 18m 2 = 9m ( 3x − 2m )
∴ 4x − 7x − 2y 2 is divisible by 9.
2
Example 7 If n is an integer, prove that n (n + 1) ( 2n + 1) is divisible by 6.

Solution n (n + 1) ( 2n + 1) = n (n + 1) [(n + 2) + (n − 1)]
= n (n + 1) (n + 2) + n (n + 1) (n − 1)
= n (n + 1) (n + 2) + (n − 1) n (n + 1)
Each of the two (n − 1) n (n + 1) and (n + 2) (n + 1) n being the product of three
consecutive integers is divisible by 3! = 6
∴ n (n + 1) ( 2n + 1) is also divisible by 6.
Example 8 Prove that 4 does not divide (m 2 + 2) for any integer m.
Solution Q m is an integer.
∴ Either m is even or m is odd.
Case I m is even
So let m = 2k
www.pdfworld.in
www.pdfworld.in
∴ m 2 + 2 = ( 2k )2 + 2 = 4k 2 + 2 = 2 ( 2k 2 + 1)
= (2 × an odd integer)
Which is not divisible by 4.
Case II m is odd
Let m = 2k + 1
∴ m 2 + 2 = ( 2k + 1)2 + 2 = 4k 2 + 1 + 4k + 2
= ( 2 + 4k + 4k 2 ) + 1.
Which being an odd integer is not divisible by 4.
Note Two important formulae.

1. If n is either even or odd,
x n − y n = ( x − y )( x n − 1 + x n − 2y + x n − 3y 2 + K + y n − 1)
2. If n is odd,
x n + y n = ( x + y )( x n − 1 − x n − 2y + x n − 3y 2 − K + y n − 1)
Example 9 Prove that 8n − 3n is divisible by 5.

Solution 8n − 3n = ( 8 − 3)( 8n − 1 + 8n − 2 . 3 + K + 3n − 1)
or 8n − 3n = 5 ( 8n − 1 + 8n − 2.3 K + 3n − 1)
∴ 8n − 3n is divisible by 5.
Example 10 If p > 1 and 2p − 1 is prime, then prove that p is prime.

Solution If possible, let p be not prime
∴ p is composite. (Qp > 1)
∴ p = mn, where m > 1 and n > 1
∴ 2p − 1 = 2mn − 1 = ( 2m )n − 1n
Putting 2 = a = an − 1n , where a = 2m > 2
m
= (a − 1)(an − 1 + an − 2 + K + 1n − 1)
Now, each of the two factors on RHS is greater than 1.
∴ 2p − 1 is composite.
But it is contrary to given
∴ p must be prime.
Remark
l
But converse is not true i.e., when p is prime, 2p − 1 need not be prime.
l
For example, p = 11 is prime but 211 − 1 is divisible by 23 and hence is not prime.
Theorem 4 The number of primes is infinite.

Proof If possible suppose that the numbers of prime is finite.
∴ ∃ the greatest prime say q.
Let b denote the product of these primes 2, 3, 5, …, q.
i.e., let b = 2⋅ 3⋅ 5Kq …(i)
let a =b + 1 …(ii)
Surely, a ≠1 (Qa = b + 1 > 1)
www.pdfworld.in
www.pdfworld.in
∴ The number a must have a prime say factor p i.e., p | a.

Now, p is one of the primes 2, 3, 5, 7 … q (Because according to our assumption 2, 3, 5, 7, … q are the only
primes).
∴ p |b (Qb = 235
. . K q)
Now p | a and p | b
∴ p |a − b or p |1 [Qfrom Eq. (ii), a − b = 1]
p =1 (which is impossible) (Q1 is not prime)
So our supposition is false.
∴ The number of primes is infinite.
Theorem 5 Fundamental Theorem of Arithmetic each natural number greater than 1 can be
expressed as a product of primes in one and only one way (except for the order of the factors).
Example 1 Every natural number other than 1 admits of a prime factor.

Solution Suppose that n ≠ 1 is a natural number.
If n itself is a prime number, the example is proved in as much as the prime number
n is a factor of itself.
If n is composite, then n must have factors other than 1 and n.
Let l be the least of these factors of n other than 1 and n.
i.e., 1 < l < n and l | n
Now, we have to prove, l is prime.
If possible let l be not prime.
But l >1
∴ l is composite
∴ ∃ integers l1 and l2 such that
l = l1l2 where 1 < l1 < l and 1 < l2 < l
⇒ l1| l but l | n
∴ l1| n where 1 < l1 < l < n
i.e., l1( < l ) is a divisor of n other than 1 and n.
But this is a contradiction because l1 < l and l is the least divisor of n other than 1
and n.
∴ Our supposition is wrong.
∴l is a prime factor of n.
Example 2 Show that every odd prime can be put either in the form 4k + 1 or 4k + 3 (i.e., 4k − 1),
where k is a positive integer.
Solution Let n be any odd prime. If we divide any n by 4, we get
n = 4k + r
where 0 ≤ r < 4 i.e., r = 0, 1, 2, 3
∴ Either n = 4k or n = 4k + 1
or n = 4k + 2 or n = 4k + 3
Clearly, 4n is never prime and 4n + 2 = 2( 2n + 1)
cannot be prime unless n = 0 (Q4 and 2 can not be factors of an odd prime)
www.pdfworld.in
www.pdfworld.in
∴ An odd prime n is either of the form

4k + 1 or 4k + 3.
But 4k + 3 = 4(k + 1) − 4 + 3 = 4k ′ − 1 (where k ′ = k + 1)
∴ An odd prime n is either of the form 4k + 1 or ( 4k + 3) i.e., 4k′ − 1.
Note 1. Every number of the form 4k + 3 is of the form 4k − 1 and conversely.
2. Every number of the form 4k + 1 or 4k − 1 is not necessarily prime.
3. The above result should be committed to memory.
Example 3 Show that there are infinitely many primes of the form 4n + 3.
Solution If possible, let number of primes of form ( 4n + 3) be finite.
These primes are 3, 7, 11,…, q (put n = 0, 1, 2, …)
Let q be the greatest of these primes of the form ( 4n + 3) .
Let a = 3, 7,11, K , q be the product of all primes of the form ( 4n + 3) .
Let b = 4a − 1 …(i)
⇒ b >1 [Qa ≥ 3, ∴b = 4a − 1, b ≥ 11]
∴ By fundamental theorem b can be expressed as a product of primes say
p1. p2. p3 K pr .
i.e., b = p1. p2 K pr …(ii)
Now, b = 4a − 1 is odd and hence 2 can't be a factor of b.
∴ None of the prime factors in RHS of Eq. (ii) is 2.
i.e., Every prime factor in RHS of Eq. (ii) is odd.
∴ Each of p1, p2, K , pr is of the form ( 4n + 1) or ( 4n + 3) .
Again, all p1, p2, K , pr can't be of the form ( 4n + 1).
[Q If it were so, then b (their product) will also be of the form ( 4n + 1)] .
But this is contrary to Eq. (i) as b = 4a − 1 is of the form ( 4n + 3) .
∴ At least one of p1, p2 K pr (say p) is (a prime factor of b) of the form ( 4n + 3) i.e.,
p|b.
Also p|a [Qp is one prime of the form ( 4n + 3) and a is product of all such primes] .
∴ p|4a
∴ p | ( 4a − b )
∴ p |1 [Qfrom Eq. (i), 4a − b = 1]
Which is impossible [Q p being prime > 1 ]
∴ Number of primes of the form ( 4n + 3) is infinite.
Theorem 6 The number of divisors of a composite number n : If n is a composite number of the

α α α
order n = p1 1 . p 2 2 …p k k , then the number of divisors denoted by d (n ) is
(α1 + 1)(α 2 + 1) K (α k + 1)
Proof Let n be any composite number, let d (n ) denote the number of divisors of composite number n
by Fundamental Theorem of Arithmetic. n can be expressed as the product of the powers of primes.
α α α
n = p1 1 . p 2 2 Kp k k , where p1 , p 2 , K , p k are distinct primes and α1 , α 2…, α k are non negative integers.
www.pdfworld.in
www.pdfworld.in
α α
Qp1 is a prime number, therefore, the only divisors of p1 1 are 1, p1, p12 , p13 , K , p1 1
α
The number of these divisors of p1 1 = α1 + 1
α
Similarly, the number of divisors of p 2 2 = α 2 + 1
α
The number of divisors of p k k = α k + 1
α α α
Therefore, the total number of divisors of n = p1 1 ⋅ p 2 2 K p k k = (α1 + 1) (α 2 + 1) K (α k + 1)
αi
[Q Every divisor of pi (1 ≤ i ≤ k ) is a divisor of n]
i.e., d (n ) = (α1 + 1)(α 2 + 1)…(α k + 1).
Note Let n = p1α1 ⋅ p2α 2 ⋅ p3α 3 [ p1, p2, p3 are distinct prime numbers]
Let d (n ) denotes number of divisor .
1. If n is a perfect square then d (n ) is odd
(Qall the di are even)
2. If n is not a perfect square then, d (n ) is even.
[The number of ways of writing n are the product of two factors.]
d (n ) + 1
If n is a perfect square, then number of ways are equal to .
2
d (n )
If n is not a perfect square, then number of ways are equal to .
2
Theorem 7 The sum of the divisors of any composite number n is denoted by σ(n ) which is equal to
 p α1 + 1 − 1   pα2 + 1 − 1  pαk + 1 − 1
 1   2  K k .
 p1 − 1   p2 − 1   pk − 1 
     
Proof Let n be any composite number and let σ(n ) is sum of positive divisors of n. By Fundamental
Theorem of Arithmetic n can be expressed as the product of the powers of primes.
α α α
∴ n = p1 1 . p 2 2 K p k k
[where p1 , p 2 , K , p k are distinct primes and α1 , α 2 , K , α k are non-negative integers]
Q p1 is prime number
α α
∴ divisors of p1 1 are only 1, p1 , p12 , K , p1 1
α α 1 [ p1α + 1 − 1]
Sum of these divisors of p1 1 = 1 + p1 + p12 + K + p1 1 =
p −1
 a (r n − 1)
Q RHS is a Geometric Progression with a = 1, r = p1 , n = α1 + 1 and Sn = r − 1 
 
α +1
α (p2 2 − 1)
Similarly sum of divisors of p 2 2 =
p2 − 1
α +1
α (pk k − 1)
Sum of divisors of p k k =
pk − 1
α α α
∴ σ(n ) = Sum of divisors of n = p1 1 ⋅ p 2 2 ... p k k
α +1 α +1 α +1
( p1 1 − 1) ( p 2 2 − 1) (p k − 1)
= . K k .
( p1 − 1) ( p 2 − 1) ( p k − 1)
αi
[QEvery divisor of pi (1 ≤ i ≤ k ) is divisor of n]
www.pdfworld.in
www.pdfworld.in
Note If dk (n ) denotes the sum of k th power of divisor of n, then

( p1k (α1 + 1) − 1) ( pk2 (α 2 + 1) − 1) ( pm
k (α m + 1)
− 1)
dk (n ) = ⋅ K
( p1 − 1)
k
( p2 − 1)
k
( pmk
− 1)
Example 1 Find the sum of the cubes of divisor of 12.

Solution 12 = 22 × 3
23(2 + 1) − 1 33(1 + 1) − 1
dk (12) = × = 2044
23 − 1 33 − 1
Example 2 Find the number of divisor of 600.

Solution 600 = 23 × 31 × 52
α 1 = 3, α 2 = 1, α 3 = 2
Number of divisors = (α 1 + 1)(α 2 + 1)(α 3 + 1)
= ( 3 + 1)(1 + 1)( 2 + 1) = 4 × 2 × 3 = 24
Greatest Integer Function

Greatest integer function is also known as Bracket Function.
If x is any real number, then the largest integer which does not exceed x is called the integral part of x
and will be denoted by [x ] .
The function which associates with each real number x, the integer [x ] is often called the bracket
function.
For example, [3] = 3, [−4] = − 4, [37
. ]=3
. ] = − 5,   = 1, [− π ] = − 4
5
[−42
 3 
Note 1. [ x ] is the largest integer ≤ x.

2. If a and b are positive integer, such that
a = qb + r , 0 ≤ r < b
a r r
Then, = q + , where 0 < < 1
b b b
a 
∴ b  = q
a 
i.e.,   is the quotient in the division of a by b.
b 
Properties of Greatest Integer Function

(1) [x ] ≤ x < [x ] + 1 and x − 1 < [x ] ≤ x , 0 ≤ x , 0 ≤ x − [x ] < 1
(2) If x ≥ 0, [x ] = Σ 1
1 ≤i ≤ x
(3) [x + m ] = [x ] + m , If m is an integer.
(4) [x ] + [y ] ≤ [x + y ] ≤ [x ] + [y ] + 1
(5) [x ] + [−x ] = 0, If x is an integer = − 1 otherwise.
www.pdfworld.in
www.pdfworld.in
(6)   =   , if m is a positive integer.

[x ] x
m   m 
(7) − [−x ] is the least integer greater than or equal to x.

This is denoted as (x ) (read as ceiling x)
. ) = 3, (−25
For example, ( 25 . )= −2
(8) [x + 05
. ] is the nearest integer to x.
If x is midway between two integers, [x + 05
. ] represents the larger of the two integers.
(9) The number of positive integers less than or equal to n and divisible by m is given by 
n
.
 m 
(10) If p is a prime number and e is the largest exponent of p such that

∞ n 
pe | n ! , then e = Σ  
i =1 p
 i
 [a ]
If a is real number, c is natural number, then   =  
a
Theorem 1
 c   c 
Proof Let [a ] = n i.e., n is largest integer ≤ a
∴ a = n + r, 0 ≤ r < 1 …(i)
 [a ]  n 
Let  c  =  c  = m
 
n
∴ = m + s, where 0 ≤ s < 1
c
∴ n = mc + cs, where 0 ≤ cs < c …(ii)
 [a ]
LHS = −   =   = m
n
Now,

 c   c 
n + r
RHS =   = 
a
[Putting value of a from Eq. (i)]
 c   c 
Putting the value of n from Eq. (ii).

mc + cs + r 
=
 c 
cs + r 
= m +
 c 
From Eq. (ii), cs ≤ (c − 1) and r < 1
Adding cs + r < c − 1 + 1
⇒ cs + r < c
cs + r
∴ <1
c
cs + r 
∴ RHS = m + =m
 c 
∴ LHS = RHS.
www.pdfworld.in
www.pdfworld.in
Theorem 2 For every positive real number

 x  +  x + 1  = [x ]
 2   2 
Proof First suppose that x = 2m + y , where m is an integer and 0 ≤ y < 1.
[x ] = 2m ,   = m
x
 2 
 x + 1  =  2m + 1 + y  = m
 2   2 
1 1+y
Since ≤ <1
2 2
∴  x  +  x + 1  = [x ]
 2   2 
Next, let x = (2m + 1) + y , where m is an integer and 0 ≤ y < 1.
Then,  x  = m ,  x + 1  =  (2m + 2) + y  = m + 1
 2   2   2 
[x ] = 2m + 1
∴  x  +  x + 1  = [x ]
 2   2 
The desired equality holds for all possible values of x.
Example 1 Find the highest power of 3 contained in 1000!.

Solution p = 3, n = 1000
 n  1000   1
 p  =  3  = 333 3  = 333
     
 n   333 
 2  =  3  = [111] = 111
p   
 n  111
 3  =  3  = [ 37] = 37
p   
 n   37   1 
 4  =  3  = 12 3  = 12
p     
 n  12 
 5  =  3  = [ 4] = 4
p   
 n  4  1 
 6  =  3  = 1 3  = 1
p     
 n  1 
 7  = 3 = 0
p   
∴ Highest power of 3 contained in 1000!
n   n   n   n   n   n   n 
=   +  2 +  3 +  4 +  5+  6 +  7
p p  p  p  p  p  p 
= 333 + 111 + 37 + 12 + 4 + 1 + 0 = 498
www.pdfworld.in
www.pdfworld.in
n + 1   2n 
If n and k are positive integers and k is greater than 1, then   + 
n
Theorem 3 ≤
 k   k   k 
Proof Let n = qk + r , q and r are integers and

0≤ r ≤ k − 1
n r n+1 r +1
Then, =q + , =q + ,
k k k k
2n 2r
= 2q +
k k
n + 1
(i) r < k − 1, then   = q ,  = q ,   ≥ 2q
n 2n
 k   k   k 
The desired result is immediate.

(ii) r = k − 1, then
 n  = q ,  n + 1  = q + 1,
 k   k 
 2n  = 2q +  2(k − 1) = 2q + 1
 k   k 
 
∴  n  +  n + 1  =  2n 
 k   k   k 
From which the desired result is immediate.
Theorem 4 If n be any positive integer, then show that

 n + 1  +  n + 2  +  n + 4  +  n + 8 + K = n
 2   4   8   16 
Proof We know that,

x + 1
[x ] =   + 
x
 2   2 
n n n n
Applying above formula to n , , , , ,K
2 4 8 16
n + 1
[n ] =   + 
n
 2   2 
 n  =  n  +  (n / 2) + 1 
 2   4   2 

 n  =  n  +  (n / 4) + 1 
 4   8   2 

 n  =  n  +  (n / 8) + 1 
 8   16   2 

Adding corresponding sides and cancelling out the terms   ,   ,   ,… from both sides, we have
n n n
 2   4   8 
n + 1  n + 2  n + 4 
n= + + +K
 2   4   8 
[n ] = n
www.pdfworld.in
www.pdfworld.in
Theorem 5 For every real number x
[x ] + x +  + x +  + K +
1 2 x + n − 1  = [nx ]
 
n   n   n 
Proof Let x = [x ] + y , where 0 ≤ y < 1
Let p be an integer such that
p − 1 ≤ ny < p
(This is always possible because given a real number, we can always find two consecutive integers
between which the number lies).
k k
Now, x + = [x ] + y +
n n
k p −1+ k p+k
Also, y + lies between and
n n n
p −1+ k
So long as < 1,
n
i.e., k < n − (p − 1)
k
y + is less than 1 and consequently
n
 k
x + n  = [x ]
 
 k
i.e., x + n  = [x ] for k = 0, 1, …, n − p
 
 k
But x +  = [x ] + 1, for k = n − p + 1,…, n − 1
 n
n − 1
[x ] + x +  + K + x +
1
∴
 n   n 
= [x ] + K + [x ](n − p + 1 times) + ([x ] + 1) + ([x ] + 1) + K ( p − 1) times
= n[x ] + ( p − 1) …(i)
Also, [nx ] = [n[x ] + ny ] = n[x ] + ( p − 1)
Since p − 1 ≤ ny < p …(ii)
From Eqs. (i) and (ii)
n − 1
[x ] + x +  + x +  + K + x +
1 2
= [nx ]
 n   n   n 
Theorem 6 The highest power of a prime number p contained in n ! is given by

n   n   n 
k (n !) =   +  2  +  3  + K
p  p  p 
Proof Let k (n !)denote the highest power of p contained in n !n ! is the product of the factors 1, 2, 3, …, n.
The factors in n ! which will be divisible by p are
n 
p , 2p , 3p , K  p
p 
n   n  
∴ k (n !) =   + k    ! …(i)
p   p  
www.pdfworld.in
www.pdfworld.in
n
Changing n to in Eq. (i)
p
 n    n   n  
k   !  =  2  + k   2  ! …(ii)
 p   p   p  
Putting the value from Eq. (ii) in Eq. (i)
n   n   n  
k (n !) =   +  2  + k   2  !
p  p   p  
n   n   n 
Continuing this process, we get k (n !) =   +  2  +  3  + K
p  p  p 
This process must end after a finite number of steps.
Congruences
If a and b are two integers and m is a positive integer, then a is said to be congruent to b modulo m, if m
divides a − b denoted by m | (a − b ).
In notation form we express it as a ≡ b mod m or a − b ≡ 0 mod m.
Note 1. a ≡ b mod m, then m | (a − b ) or (a − b ) is a multiple of m.

2. If m | (a − b ) [m does not divides (a − b )], then a is said to be incongruent to b mod m and this
fact is expressed as a is not congruent to b mod m.
3. If m | a, then a ≡ 0 mod m
For example :
(i) 13 ≡ 1 mod 4 (Q4 | (13 − 1) = 12)
(ii) 4 ≡ − 1 mod 5 (Q5 | ( 4 − ( −1)) = 5)
(iii) 12 ≡ 0 mod 4 (Q4|12)
(iv) 17 is not congruent to 3 mod 5 (Q5 | (17 − 3))
Theorem 7 If a ≡ b mod m, then

(i) a + c = b + c mod m
(ii) ac ≡ bc mod m, where c is any integer.
Proof (i) Qa ≡ b mod m
m | (a − b)
⇒ m | {(a + c ) − (b + c )} [Qa + c − ( b + c ) = a − b ]
a + c ≡ b + c mod m
(ii)Qa ≡ b mod m
m | (a − b )
m | c (a − b )
m | (ac − bc )
∴ ac ≡ bc mod m
Note The converse of theorem 15 (ii) is not true.

Theorem states that if a ≡ b mod m, then ac ≡ bc mod m.
i.e., a congruence can always be multiplied by an integer
But the converse is not true i.e.,. It is not always possible to cancel a common factor from a
congruence.
For example : 16 ≡ 8 mod 4 [Q4|16 – 8]
But if we cancel the common factor 8 from numbers 16 and 8, we get 2 ≡ 1 mod 4 which is a false
result because 4 | ( 2 − 1)
www.pdfworld.in
www.pdfworld.in
Theorem 8 If a ≡ b mod m and c ≡ d mod m, then

(i) a + c ≡ b + d mod m (ii) a − c ≡ b − d mod m (iii) ac ≡ bd mod m
Proof (i) Qa ≡ b mod m and c ≡ d mod m
∴ m | (a − b ) and m| (c − d )
m | ((a − b ) + (c − d ))
or m | ((a + c ) − (b + d ))
∴ a + c ≡ b + d mod m
(ii) a ≡ b mod m and c ≡ d mod m
∴ m | (a − b ) and m | (c − d )
∴ m | (a − b ) − (c − d ) or m | (a − c ) − (b − d )
a − c ≡ b − d mod m
(iii)Qa ≡ b mod m and c ≡ d mod m
m | (a − b ) and m | (c − d )
∴ There exists integer h and k such that
a − b = mh and c − d = mk
a = b + mh and c = d + mk
Multiplying the two equations, we get
ac = (b + mh )(d + mk ) = bd + mbk + mhd + m 2hk
ac − bd = m (bk + hd + mhk )
∴ By definition of divisibility m | (ac − bd )
or ac ≡ bd mod m
Corollary If a ≡ b mod m, then a 2 ≡ b 2 mod m

Q a ≡ b mod m and again
a ≡ b mod m
Multiplying the two congruence a 2 ≡ b 2 mod m
Theorem 9 (i) Prove that a ≡ a mod m i.e., every integer is congruent to itself.
(ii) If a ≡ b mod m, then prove that b ≡ a mod m
(iii) If a ≡ b mod m, b ≡ c mod m prove that a ≡ c mod m
Proof (i) We know that m|0 (m ≠ 0)
m | (a − a )
a ≡ a mod m [by definition of congruence]
(ii) Let a ≡ b mod m
m | (a − b )
∴ m |− (a − b ) or m | (b − a )
∴ b ≡ a mod m
(iii) Let a ≡ b mod m and b ≡ c mod m
∴ m | ( a − b ) and m | (b − c )
∴ m | ( a − b ) + (b − c ) or m | ( a − c )
∴ a ≡ c mod m
www.pdfworld.in
www.pdfworld.in
Theorem 10 If a ≡ b mod m, then a k ≡ b k mod m for every positive integer k.
Proof We know that,

a k − b k = (a − b )(a k − 1 + a k − 2b + a k − 3b 2 + K + b k − 1 ) …(i)
But a ≡ b mod m ⇒ m | (a − b )
∴ There exists an integer t such that
a − b = mt …(ii)
Putting this value of (a − b ) from Eq. (ii) in Eq. (i)
a k − b k = mt (a k − 1 + a k − 2b + K b k − 1 )
∴ m| (a k − b k )
∴ a k ≡ b k mod m
−1 −2
Theorem 11 If a ≡ b mod m and f (x ) = p 0x n + p1x n + p 2x n + K +pn − 1x + pn is an integral
rational function of an indeterminate x with integral coefficients, then f ( a ) ≡ f ( b ) mod m
−1 −2
Proof Q f ( x ) = p 0x n + p1x n + p 2x n + K + pn −1 x + pn
Putting x = a
−1 −2
∴ f ( a ) = p 0an + p1an + p 2an + K + pn −1 a + pn …(i)
Putting x = b
−1 −2
∴ f (b ) = p 0bn + p1bn + p 2bn + K + pn − 1b + pn …(ii)
Subtract Eq. (i) from Eq. (ii), we get

−1 −1
f (a ) − f (b ) = p 0 (an − bn ) + p1 (an − bn ) + K + pn − 1 (a − b)
n −1 n −2 n −1
= p 0 (a − b )(a +a b + K+ b )+
n −2 n −3 n −2
p1 (a − b )(a +a b + K+ b ) + K + pn − 1 (a − b)
n −1 n −2 n −2 −2 −3 −2
or f (a ) − f (b ) = (a − b )[p 0 (a +a b + K+ b ) + p1 (an + an b + K + bn )
+ K + pn − 1] …(iii)
= (a − b ) t (say)
But a ≡ b mod m (given)
m| (a − b )
∴ ∃ an integer k such that a − b = mk
Putting this value of a − b = mk in Eq. (iii)
f (a ) − f (b ) = mkt
∴ m |f (a ) − f (b )
⇒ f (a ) ≡ f (b ) mod m
Theorem 12 Fermat Theorem

If p is prime, then
(a + b ) p = (a p + b p ) mod p.
Proof Expanding by binomial theorem

(a + b ) p = a p + pC1a p − 1b + pC2ap − 2b 2 + K + pCp − 1abp − 1 + bp
p −1
or (a + b ) p = (ap + bp ) + Σ p
Cr ap − r br …(i)
r =1
www.pdfworld.in
www.pdfworld.in
p!
Now, p
Cr = ; 1 ≤ r ≤ (p − 1)
r !(p − r )!
But p ! = 123
. . … p is divisible by p
p is coprime to r !
Q p is coprime to 1, 2, 3, … r (Qr < p , p is prime)
∴ p is co prime to their product = r !
Also for the same reason p is coprime to (p − r )!
p!
∴ p
Cr = is divisible by p.
r !(p − r )!
∴ ∃ an integer k r such that
p
Cr = pk r
p
Putting this value of Cr in Eq. (i)
p −1
(a + b )p − (ap + bp ) = p Σ kr ap − r br
r =1
which is divisible by p.
∴ (a + b )p ≡ (ap + bp ) mod p
Generalization
If p is a prime number, prove that
(a1 + a 2 + a3 + K + an ) p
≡ (a1p + ap2 + a3p + K + anp ) mod p
(a1 + a 2 + a3 + K + an ) p = (a1 + b1 ) p
where b1 = a 2 + a3 + K + an
≡ (a1p + b1p ) mod p.
≡ [a1p + (a 2 + a3 + K + an ) p ] mod p
≡ [a1p + (a 2 + c 2 ) p ] mod p
where c 2 = a3 + a 4 + K + an ≡ (a1p + ap2 + c p2 ) mod p
continuing like this, we get
(a1 + a 2 + a3 + K + an ) p ≡ (a1p + ap2 + K + anp ) mod p
Theorem 13 If p prime number, then

ap = a mod p
Proof We know that,

(a1 + a 2 + a3 + K + an ) p ≡ (a1p + ap2 + a3p K + anp ) mod p …(i)
Putting a1 = a 2 = a3 = K = an = 1 in Eq. (i)
(1 + 1 + 1 + K + 1) p
≡ (1p + 1p + 1p + K + 1p ) mod p
or np ≡ (1 + 1 + 1 + K + 1) mod p
or np ≡ n mod p for every natural number n.
Replacing n by a.
ap = a mod p
www.pdfworld.in
www.pdfworld.in
Theorem 14 Fermat Little Theorem

If p is a prime number and (a , p ) ≡ 1, prove that ap − 1 = 1 mod p.
Proof As p is prime.
∴ ap = a mod p
Cancelling a from both sides. [∴a is coprime to p]
We have ap − 1 ≡ 1 mod p.
−2 n −1 n
Theorem 15 n ! = nn − n C1 (n − 1)n + n C2 (n − 2)n − ... + (−1)n Cn − 22
n
+ (−1)n Cn −1
Proof Expanding by binomial theorem

− 1) x − 2) x n − 2 2x n −1 x
(e x − 1)n = enx − n C1e (n + n C2e (n − K + n Cn − 2 (−1) e + n Cn − 1 (−1) e + (−1)n …(i)
θ θ θ 2 3
We know that eθ = 1 + + + +K
1! 2! 3!
Using this expansion of e θ , Eq. (i) becomes
n
 x x 2 x3 xn 
1 + + + + K+ + K − 1
 1! 2! 3! n! 
 (nx )2 (nx )n  (n − 1)x [(n − 1)x ]2 [(n − 1)x ]n 
+ K − n C1 1 +
nx
= 1 + + + K+ + + K+ + K
 1! 2! n!  
 1! 2! n! 
 (n − 2) x [(n − 2) x ]2 [(n − 2) x ]n 
+ n C2 1 + + +K + + K
 1 ! 2 ! n ! 
n − 2 2x (2x )2 (2x )n 
+ K + n Cn − 2 (−1) 1 + + +K+ + K
  1! 2! n 
− 1n  x x2 xn 
+ (−1)n Cn − 1 1 + + + K+ + K + (−1)n
 1! 2! n! 
Comparing coefficient of x n on both sides

nn n (n − 1)n (n − 2)n
1= − C1 + n C2 −K+
n! n! n!
2n (−1)n − 1 n Cn − 1
(−1)n − 2 n Cn − 2 +
n! n!
Multiplying both sides by n !
−2 n −1 n
n ! = nn − n C1 (n − 1)n + n C2 (n − 2)n − K + (−1)n Cn − 22
n
+ (−1)n Cn −1
Theorem 16 Wilson Theorem

If p is prime, then (p − 1)! + 1 ≡ 0 mod p
Proof
Case I when p = 2
(2 − 1)! + 1 ≡ 0 mod 2 [Putting p = 2 in (p − 1)! + 1 ≡ 0 mod p ]
⇒ 1 ! + 1 ≡ 0 mod 2
⇒ 2 ≡ 0 mod 2
which is true
∴ Wilson theorem is true for p = 2 .
www.pdfworld.in
www.pdfworld.in
Case II If p is an odd prime.

−2 n −1 n
n ! = nn − n C1 (n − 1)n + n C2 (n − 2)n − ... + (−1)n Cn − 22
n
+ (−1) n Cn −1
Put n = p − 1 on both sides

∴ ( p − 1)! = ( p − 1) p − 1 − p −1
C1 ( p − 2) p − 1
p −1
+ C2 ( p − 3) p − 1 + .. + (−1) p − 3 p −1
Cp − 32p − 1 + (−1) p − 2 p −1
Cp − 2 …(i)
Q p is prime.
∴ p is coprime to all numbers < p.
i.e., p is coprime to p − 1, p − 2, p − 3, … 2, 1.
∴ Putting a = p − 1, p − 2, p − 3, … 2 in Fermat Theorem
ap − 1 ≡ mod p
( p − 1) p − 1 ≡ 1 mod p
or ( p − 1) p − 1 − 1 = M(p)
( p − 1) p − 1 = M ( p ) + 1
Similarly, ( p − 2) p − 1 = M ( p ) + 1
( p − 3) p − 1 = M ( p ) + 1
…………………
…………………
…………………
2p − 1 = M ( p ) + 1
Putting these values of ( p − 1) p − 1 , ( p − 2) p − 1, …, 2p − 1 in Eq. (i).
p −1 p −1
∴ ( p − 1)! = [M ( p ) + 1] − C1[M ( p ) + 1] + C2[M ( p ) + 1] + .. + (−1) p − 3 p − 1Cp − 3
[M ( p ) + 1] + (−1) p − 2 p − 1Cp − 2
p −1
p −1 p −1
or ( p − 1)! = M ( p ) + 1 − C1 + C2 − K + (−1) p − 3 Cp − 3 + (−1) p − 2 p − 1Cp − 2
Adding and Subtracting (−1) p − 1 in RHS.

( p − 1)! = M (p ) + [(1) p − 1 − p −1
C1 + p −1
C2 − p −1
C3 + K + (−1) p − 3 p − 3Cp − 3
+ (−1) p − 2 p − 1Cp − 2 + (−1) p − 1 ] − (−1) p − 1
( p − 1)! = M ( p ) + (1 − 1) p − 1 − (−1) p − 1
∴ ( p − 1)! = M ( p ) + 0 − (−1) p − 1 = M ( p ) − 1
Qp is odd. ∴p − 1 is even.
∴ (−1) p − 1 = 1
or ( p − 1)! + 1 = M ( p )
( p − 1)! + 1 is divisible by p.
∴ ( p − 1)! + 1 ≡ 0 mod p.
Theorem 17 Converse of Wilson Theorem

If p > 1 and ( p − 1)! + 1 ≡ 0 mod p, then p is a prime number.
Proof If possible let p be not prime.
∴ p is composite (Qp > 1).
So let p = p1p 2, where (1 < p1 < p , 1 < p 2 < p ) or 1 < p1 ≤ p − 1, 1 < p 2 ≤ p − 1
www.pdfworld.in
www.pdfworld.in
Now, 1 < p1 ≤ p − 1
∴p1 is one of the factors in the value of ( p − 1)! and therefore p1 divides ( p − 1)! .
Also p = p1p 2 …(i)
⇒ p1 | p …(ii)
But ( p − 1)! + 1 ≡ 0 mod p
∴ p | ( p − 1)! + 1 …(iii)
From Eqs. (ii) and (iii)
p1 | ( p − 1)! + 1 …(iv)
From Eqs. (iv) and (i)
p1 | ( p − 1)! + 1 − (p − 1)!
i.e., p1 | 1
But this is impossible. (Qp1 > 1)
∴ p is a prime number.
Euler’s Function
Definition : The number of integers ≤ n and coprime to n is called Euler's function for n and is denoted by
φ (n ).
Examples
φ (1) = 1
[Q1 is the only integer ≤ 1 and coprime to 1].
φ ( 2) = 1
[Q1 is the only integer < 2 and coprime to 2].
φ (8) = 4
[Q1, 3, 5, 7 are the only four integers < 8 and coprime to 8].
Remark
l
If p is a prime number, then 1, 2, 3, … ( p − 1) are all less than p and coprime to p and are ( p − 1) in total.
∴ φ ( p) = p − 1
 1 1  1
Theorem 18 Prove that φ (n ) = n  1 −   1 −  K 1 −  where p1 , p 2 , ... , pr are distinct prime
 p1   p2   pr 
factors of n.
Proof Q p1 , p 2 , K , pr are distinct prime factors of n.
k k
∴ n = p1 1 . p 2 2 K prkr
k k k k
∴ φ (n ) = φ ( p1 1 . p 2 2 ... prkr ) = φ ( p1 1 ) φ ( p 2 2 ) K φ ( prkr )
[Qp1 , p 2 , K , pr are distinct primes and hence are coprime to each other and φ (ab ) = φ (a ) φ (b ), if a and b are
coprime to each other.]
k  1 k  1 k  1
= p1 1  1 −  p 2 2  1 −  K pr r  1 − 
 p1   p2   pr 
k k  1 1  1
= p1 1 . p 2 2 K prkr  1 −   1 −  K 1 − 
 p1   p2   pr 
 1 1  1 k k
= n 1 −  1 −  K 1 −  [Qn = p1 1 . p 2 2 K prkr ]
 p1  p 2  pr 
www.pdfworld.in
www.pdfworld.in
 1
Theorem 19 Prove that φ ( p k ) = p k  1 −  , where p is prime.
 p
Proof Number of integers from 1 to p k which are not coprime to p k are p.1, p.2, p.3,… p. p k − 1.
Total number of such integers, which are not coprime to p k = p k − 1.
∴ φ (p k ) = Number of integers coprime to p k and < p k .
= p k − p k − 1 = p k (1 − 1 / p )
Remark
If a and b are coprime to each other, then φ (ab ) = φ (a ) φ (b ).
Example 1 Find the number of positive integers ≤ 3600 that coprime to 3600.
Solution n = 3600 = 24 × 32 × 52
φ (n ) = φ ( 3600) = φ ( 24 × 32 × 52 )
 1  1  1  1  1  1
= n 1 −  1 −  1 −  = 3600 1 −  1 −  1 − 
 p1  p2   p3   2  3  5
[Here p1 = 2, p2 = 3, p3 = 5]
1 2 4
= 3600 × × ×
2 3 5
∴ φ ( 3600) = 960
Example 2 If m > 2, show that φ (m ) is even.
Solution If (a, m ) = 1, then (m − a, m ) = 1

∴ Integers coprime to m occur in pairs of type a and m − a .
∴ φ (m ) is even.
Example 3 For what values of m is φ (m ) odd.

Solution If m > 2, φ (m ) is even.
φ (1) = 1
φ ( 2) = 1
Only for m = 1, m = 2
φ (m ) is odd.
10n − 1 10n − 1
Concept Let a = =
10 − 1 9
10n − 1
We can express any a of the form in terms of perfect square.
9
10n − 1
a= ⇒ 9a = 10n − 1
9
9a + 1 = 10n
Let b = 9a + 1
c = 8a + 1
Now, consider 4ab + c = 4a (9a + 1) + 8a + 1 = 36a 2 + 12a + 1 = (6a + 1)2
www.pdfworld.in
www.pdfworld.in
Verification (6 × 1 + 1)2 = 72 = 49 = (6 × 11 + 1)2 = 672

(6 × 111 + 1)2 = 6672
Now, consider
(a − 1) b + c = (a − 1)(9a + 1) + 8a + 1
= 9a 2 + a − 9a − 1 + 8a + 1 = 9a 2 = (3a )2
Verification a = 1 ⇒ 32
a = 11 ⇒ (33)2
Now, consider (16ab + c )
16a (9a + 1) + 8a + 1
(12a + 1)2
This is also a perfect square.
Concept Prove that every number of the sequence 49, 4489, 44489, 4448889 is a perfect square.
If there are n fours and (n − 1) eight and one 9.
Let us denote 444889 as 43829.
Consider 667 written as 627
We know 444889 = (667)2.
∴We develop (6nm − 17)
2
= 4n 8n − 19
If this is true then

2 2 2
 6(10n − 1)   6 × 10n + 3   2 . 10n + 1 
− 17) =  + 1 =   = 
2
(6n 
 9   9   9 
. 2n
410 . n + 1 40n − 140n − 1 + 1
+ 410
= = = 4n 8n − 19
9 9
Example 1 Let n be the natural number. If 2n + 1 and 3n + 1 are perfect square. Then prove that n
is divided by 40.
Solution 40 = 23 × 5. It is sufficient to prove that n is divisible by 8 and 5.
Let 2n + 1 = x 2 …(i)
and 3n + 1 = y 2
…(ii)
⇒ x 2 is odd.
⇒ x is odd.
Let x = 2a + 1
( 2n + 1) = ( 2a + 1)2
2n + 1 = 4a 2 + 4a + 1
n = 2a 2 + a
⇒ n is even.
If n is even ⇒ 3n + 1 is odd
⇒ y 2 is odd ⇒ y is odd
Let y = 2b + 1
Subtract Eq. (i) from Eq. (ii), we get
n = y2 − x2 …(iii)
www.pdfworld.in
www.pdfworld.in
⇒ n = ( 2b + 1)2 − ( 2a + 1)2
We know square difference of odd number is always divisible by 8.
∴ n is divisible by 8. …(iv)
If we eliminate n between 1 and 2
3x 2 − 2y 2 = 1
Since square of odd number ends with 1, 5 or 9
∴ 3x 2 ends with 3, 5 or 4, 7
⇒ 2y 2 ends with 2, 0, 8
⇒ x 2 ends with 1 and y 2 ends with 1
⇒ n = y2 − x2 [from Eq. (iii)]
=0
∴ It is divisible by 5.
Example 2 Prove that there are infinitely many squares in the sequence 1, 3, 6, 10, 15, 21, 28,…
Solution Suppose Tn is a square
n(n + 1)
Let Tn of the above sequence be
2
n(n + 1)
⇒ Tn =
2
If it is a square then Tn = (m )2
n(n + 1)
⇒ = (m )2
2
⇒ n(n + 1) = 2(m )2
Also, T4n (n + 1) is also a square.
4n(n + 1)[ 4(n )(n + 1) + 1]
∴ T4n (n + 1) =
2
4( 2m 2 )[ 4n 2 + 4n + 1]
= = 4m 2( 2n + 1)2
2
T4n (n + 1)is also a perfect square.
∴ perfect squares are T1 = 1
T8 = 36 is a perfect square.
T288 is also a perfect square.
Example 3 If N = 123 × 34 × 52, find the total number of even factor of N.

Solution If N = 123 × 34 × 52
Then, N = 26 × 37 × 52
∴ Total Number of factors are = ( 6 + 1)(7 + 1)( 2 + 1) = 7 × 8 × 3 = 168
In above factors, some of these are odd multiple and some are even.
The odd multiples are formed only with combination of 35 and 55.
So total number of odd multiples is
(7 + 1)( 2 + 1) = 24
∴ Even multiples = 168 − 24 = 144
www.pdfworld.in
www.pdfworld.in
Example 4 Show that n 2 − 3n − 19 is not a multiple of 289 for any integer n.

Solution Suppose 172 | n 2 − 3n − 19
Since n 2 − 3n − 19 = (n + 7)(n − 10) + 51
17| (n + 7)(n − 10);
∴ 172 | (n + 7)(n − 10) (Qn + 7 ≡ n − 10 (mod 17))
⇒ 172 | (n 2 − 3n − 19) − (n + 7)(n + 10)
i.e., 172 |51 which is a contradiction
Consequently n 2 − 3n − 19 is not a multiple of 289.
Example 5 Determine all integers n such that n 4 − n 2 + 64 is the square of an integer.

Solution Since n 4 − n 2 + 64 > n 4 − 2n 2 + 1 = (n 2 − 1)2 for some non negative integer k,
(n 4 − n 2 + 64) = (n 2 + k )2 = n 4 + 2n 2k + k 2
64 − k 2
i.e., n 2 = from which we find that the possible values 64, 1, 0 for n 2 are
2k + 1
obtained when k = 0, 7, 8 respectively.
Hence, n ∈ ( 0, ± 1, ± 8)
Example 6 Let a, b, c, d , e be consecutive positive integers such that b + c + d is a perfect square

and a + b + c + d + e is a perfect cube. Find smallest possible value of c.
Solution a, b, c, d, e are consecutive positive integer b + c + d = 3c and a + b + c + d + e = 5c
Now, 3|3c ⇒ 32 | 3c (Q 3c is a square)
⇒ 3|c ⇒ 3|5c ⇒ 3 |5c 3
(Q5c is a cube)
⇒ Also 5 | 5c ⇒ 53 |5c ⇒ 52 | c
∴ 3352|c i.e., 675|c
∴ 675 being a possible value of c is the smallest of such numbers.
Example 7 If 11 + 11 11a 2 + 1 is an odd integer where a is a rational number. Prove that a is

perfect square.
Solution Let λ = 11 + 11 11a 2 + 1
Then, ( λ − 11)2 = 112(11a 2 + 1)
Simplifying, we get λ ( λ − 22) = 113a 2
r
Putting | a | = , r , s ∈ N such that (r , s ) = 1 gives λ ( λ − 22)s 2 = 113r 2.
s
Since 112| s because otherwise 11 would divide r,11|λ. Writing 11µ = λ , we get
µ(µ − 2)s 2 = 11r 2
Since 11| s for otherwise we would have 11|r. It follows that s = 1. Thus we have
µ(µ − 2) = 11r 2. Since µ − 2 and µ are consecutive odd integers they are relatively
prime.
If 11| µ − 2, then µ is a square of form 11n + 2 which is not possible.
∴ 11| µ and hence µ = 11n 2 for some n ∈ N.
Thus, we have λ = 11µ = 112n 2
www.pdfworld.in
www.pdfworld.in
Example 8 Determine all pairs of positive integers (m, n ) for which 2m + 3n is a perfect square.
Solution Let 2m + 3n = k 2
Since ( −1)m ≡ 2m ≡ k 2 ≡ 1 (mod 3) (Q 3 | k )
m is even, say 2p.
Now, (k − 2p ) (k + 2p ) = 3n
⇒ k − 2 = 1 and k + 2 = 3n ⇒ 2p + 1 + 1 = 3n
p p
Since ( −1)n ≡ 3n (mod 4) = 2p + 1 + 1 ≡ 1, n is even, say 2q.

Now ( 3q − 1)( 3q + 1) = 2p + 1 ⇒ 3q − 1 = 2
⇒ 3 = 3 ⇒ q = 1 and hence p = 2
q
So, we have only one solution (4, 2).
Example 9 Determine the set of integers n for which n 2 + 19n + 92 is a square.

Solution Let n 2 + 19n + 92 = m 2, m is a non negative integer. Then, n 2 + 19n + 92 − m 2 = 0
1
Solving for n, we get n = ( −19 ± 4m 2 − 7 )
2
∴ 4m 2 − 7 is a square i.e., 4m 2 − 7 = p 2
Where p ∈ N
∴ ( 2m − p )( 2m + p ) = 7
∴ 2m + p being positive therefore (2m+p) is 7 and 2m − p = 1
Hence, 4m = 8 ⇒ m = 2
Thus, we have n + 19n + 92 = 4
2
⇒ n 2 + 19n + 88 = 0
⇒ (n + 8)(n + 11) = 0
⇒ n = − 8 or –11
Example 10 Find n, if 2200 − 2192 ⋅ 31 + 2n is a perfect square.

Solution 2200 − 2192 31 + 2n = 2192( 28 − 31) + 2n = 2192( 256 − 31) + 2n = 2192 ⋅ 225 + 2n
∴ For some m ∈ N
2n = m 2 − 2192 ⋅ 225 = m 2 − ( 296 ⋅15)2 = (m − 296 ⋅ 15)(m + 296 ⋅15)
So, m = 296 ⋅15 = 2α and m + 296 ⋅15 = 2α + β for some non negative integers α , β .
Hence, 297 ⋅15 = 2α + β − 2α = 2α ( 2β − 1)
⇒ 2α = 297 and 2β − 1 = 15.
i.e., α = 97 and β = 4
∴ n = 2α + β = 198
Example 11 Find the number of values of n for which 211 + 28 + 2n is a perfect square.
Solution We can write 211 + 28 + 2n as
28 ( 23 + 1) + 2n
⇒ 28 ⋅ 9 + 2n
⇒ 2n ( 28−n ⋅ 9 + 1)
www.pdfworld.in
www.pdfworld.in
Note that for any k < 8, 2k ( 28 − k 9 + 1) is not a square, when k is odd, 2k is not a square
and in the other case, the second factor is not a square. Hence n ≥ 8. Now write
211 + 28 + 2n as 28( 9 + 2n − 8 ). Then the problem is to find the number of non negative
integers k such that 9 + 2k is a square.
9 + 2k = t 2 ⇒ 2k = (t − 3)(t + 3)
p+q
⇒ t − 3 = 2 and t + 3 = 2
p
for some non negative integers p and q.
∴2 ( 2 − 1) = 6 implying p = 1 from which it follows that t = 5.
p q
Hence, there is a unique solution.
Example 12 Find all positive integers n for which n 2 + 96 is a perfect square.

Solution Suppose m is a positive integer, such that n 2 + 96 = m 2
Then, m 2 − n 2 = 96 ⇒ (m − n )(m + n ) = 96
since m − n < m + n and m − n,m + n must be both even [as m + n = (m − n ) + 2n.
Therefore m − n, m + n must be both odd or both even; also if both of them are odd,
then the product cannot be even.
∴ Only possibilities are
m − n = 2, m + n = 48 ⇒ m = 25, n = 23
m − n = 4,m + n = 24 ⇒ m = 14, n = 10
m − n = 6, m + n = 16 ⇒ m = 11, n = 5
m − n = 8,m + n = 12 ⇒ m = 10, n = 2
Example 13 Give with justification, a natural number n for which 39 + 312 + 315 + 3n is a perfect
cube .
Solution 39 + 312 + 315 + 3n = 39(1 + 34 + 36 + 3n − 9 )
= ( 33 )3{1 + 3 ⋅ 32 + 3( 32 )2 + ( 32 )3 + 3n − 9 − 3( 32 )2}
= ( 33 )3(1 + 32 )3 provided 3n − 9 − 35 = 0
= ( 270)3 provided 3n − 9 = 35
i.e., provided n = 14
So, given number is a perfect cube when n = 14
Example 14 Prove that 2p + 3p is not a perfect power if p is a prime number.

Solution If p = 2, 2p + 3p = 22 + 32 = 13 (not a perfect power)
Let now p be a prime > 2..
x + a divides x p + a p , whenever p is odd [factor theorem]
∴ 2p + 3p is divisible by 2 + 3 = 5. We shall show that 2p + 3p is not divisible by 52.
x p + 3p = ( x + 3)( x p − 1 − 3x p − 2 + 32 x p − 3 + K + ( −3)p − 1)
When x = − 3, then
x p − 1 − 3 x p − 2 + K + ( − 3 )p − 1
= ( − 3 ) p − 1 − 3 ( − 3 ) p − 2 + K + ( − 3 )p − 1
= p 3p − 1
www.pdfworld.in
www.pdfworld.in
Showing that x + 3 does not divide

x p − 1 − 3 x p − 2 + 3 2 x p − 3 + K + ( − 3 )p − 1
Consequently ( x + 3)2 does not divide x p + 3p. So, ( 2 + 3)2 does not divide
2p + 3p . Since 2p + 3p is a multiple of 5 but is not a multiple of 52.
∴ it cannot be a perfect power.
Example 15 A 4 digit number has the following properties (I) It is a perfect square (II) its first 2 digit
are equal to each other (III) its last two digit are equal to each other.Find all such four
digit number.
Solution We want to find positive integers x and y such that1 ≤ x ≤ 9,0 ≤ y ≤ 9 and xxyy is a
perfect square. Since,102 = 100, 1002 = 10000. It follows that xxyy must be the square
of a 2 digit number. Suppose that (ab )2 = xxyy .
The number xxyy is clearly a multiple of 11.
Since, it is a perfect square it must be a multiple of 112 i.e., 121.
∴It must be of the form
121 × 1, 121 × 4, 121 × 9, 121 × 16, 121 × 25,
121 × 36,121 × 49,121 × 64,121 × 81
Out of these 121 × 64 i.e., 7744 is of the form xxyy, we conclude that 7744 is the
desired number.
Example 16 Show that for any integer n, the number n 4 − 20n 2 + 4 is not a prime number.
Solution n 4 − 20n 2 + 4 = (n 4 − 4n 2 + 4) − 16n 2
= (n 2 − 2)2 − 16n 2
= (n 2 − 4n − 2)(n 2 + 4n − 2) …(i)
Note It can be easily seen that none of the factors n − 4n − 2 ,n + 4n − 2 can have the value ± 1,
2 2
whatever integral value n may have. Here four cases arises.

4 ± 28
(i) n 2 − 4n − 2 = 1 ⇒ n =
2
4 ± 20
(ii) n − 4n − 2 = − 1 ⇒ n =
2
2
− 4 ± 28
(iii) n + 4n − 2 = 1 ⇒ n =
2
2
− 4 ± 20
(iv) n + 4n − 2 = − 1 ⇒ n =
2
2
From the above four cases, we find that whatever integral value n may have, n 4 − 20n 2 + 4 is the
product of the integers n 2 − 4n − 2 and n 2 + 4n − 2 neither of which equals ± 1..
Example 17 Prove that the product of four consecutive natural numbers cannot be a perfect cube.
Solution Consider the product
P = n (n + 1) (n + 2)(n + 3), where n is a natural number.
If possible, that P is a perfect cube = k 3 Two cases arises.
Case I If n is odd. n, (n + 1), (n + 3) are all prime to n + 2
Now, we know that every common divisor of n + p and n + q must divide q − p.
www.pdfworld.in
www.pdfworld.in
∴ n + 2 and n(n + 1)(n + 3) are relatively prime.

Since, their product is a perfect cube, each of them must be a perfect cube.
Since, n 3 < n(n + 1)(n + 3) < (n + 3)3
∴ n(n + 1)(n + 3) = (n + 1)3 or (n + 2)3
As n(n + 1)(n + 3) and (n + 2)3 are relatively prime, so second possibility ruled out.
Also n(n + 1)(n + 3) = (n + 1)3 ⇒ n = 1. Since P = 24, when n = 1 which is not a perfect
cube. So the possibility n = 1 is also ruled out. So n cannot odd.
Case II If n is even.
Then n + 1 is prime to n, n + 2 and n + 3. Consequently n + 1 is relatively prime to
n(n + 2)(n + 3). Since the product of relatively prime numbers n + 1 and
n(n + 2)(n + 3) is a perfect cube, each of them must be a perfect cube.
n 3 < n(n + 2)(n + 3) < (n + 3)3
∴ n(n + 2)(n + 3) = either (n + 1)3
or (n + 2) since n(n + 2)(n + 3) and n + 1 are relatively prime
3
∴ First possibility ruled out.

Also n(n + 2)(n + 3) = (n + 2)3 ⇒ n + 4 = 0 which is out of question. Consequently n
cannot be even.
Thus, we find that the product of 4 consecutive integers cannot be a perfect cube.
Example 18 Find all primes p for which the quotient ( 2p − 1 − 1)| p is a square.
Solution Suppose m (m + 1) = 7n 2, m and n are integers since m and m + 1are relatively prime.
∴ m and m + 1 must be the numbers 7p 2, q 2
(in some order) p and q are relatively prime and pq = n; Since the product of 2
consecutive integer is even.
∴ m(m + 1) is even, which means that one of the numbers m, m + 1 must be even.
Suppose m = q 2 (so that m + 1 = 7p 2 ). Since every square number is of one of the
forms 4k , 4k + 1. Consequently m + 1 must be of one of the forms 4k + 1, 4k + 2.
However this is not possible for if p is even, then 7p 2 is of the form 4k. If p is odd,
then 7p 2 is of the form 4k + 3.
∴ m + 1 ≠ 7p 2. So m = 7p 2 and m + 1 = q 2
Concept of Finding Number of Positive Integral Solutions for

the Equation of the Form x2 + y2 = k
We know that,
(2n )2 ≡ 0 mod 4
and ( 2n + 1)2 ≡ 1 mod 4
Now, if
(a) x and y are both even then,
x 2 + y 2 ≡ 0 mod 4
(b) x and y are both odd then,
x 2 + y 2 ≡ 2 mod 4
www.pdfworld.in
www.pdfworld.in
(c) one is even and other is odd then,

x 2 + y 2 ≡ 1 mod 4
{∴x 2 + y 2 ≡ 0, 1, 2 mod 4 and x 2 + y 2 ≡
/ 3 and 4}
the above discussion implies, if

x 2 + y2 = k
and if k is of the form of (4m + 3), then x 2 + y 2 = k does not have any integral solution.
e.g., suppose, we are asked to find integral solution for equation
x 2 + y 2 = 19, then it will not have any integral solution because 19 is of the form (4m + 3).
Now, if we have x 4 + y 4 = k , then we know that
(2n )4 ≡ 0 mod 16 and (2n + 1)4 ≡ 1 mod 16
Again, if (a) x and y are both even then,
x 4 + y 4 ≡ 0 mod 16
(b) x and y are both odd then,
x 4 + y 4 ≡ 2 mod 16
(c) one is even and other is odd, then
x 4 + y 4 ≡ 1 mod 16
∴ x 4 + y 4 ≡ 0, 1, 2 mod 16
and x 4 + y4 ≡
/ i mod 16,
where i = (3, 4, 5, .. , 15)
So, the above discussion implies, if
x 4 + y4 = k
and k is of the form (16m + i ), where i = 3, 4, 5, K , 15, then x 4 + y 4 = k will not have any integral solution.
e.g., suppose we are asked to find integral solutions for equation.
x 4 + y 4 = 16003, then it will not give any integral solution because 16003 is of the form (16m + 3).
The concept can be extended to more than two variables expression, suppose the equation is
x 14 + x 24 + x 34 + x 44 + ... + x 14
4
= 1599
now, we know that
(2n )4 ≡ 0 mod 16
and (2n + 1)4 ≡ 1 mod 16
14
∴ Σ x i4 ≡ 0, 1, 2 K 14 mod 16
i =1
but our RHS is 1599 ≡ 15 mod 16.

∴ No integral solution can be obtained for the above equation.
Reason
[If all the variables are considered to be odd, then maximum remainder which can come out is “14” and if
any of the variable is an even number then remainder will be less than 14.]
Now, let us consider another discussion.
If we have a 2 + b 2 + c 2 = a 2b 2
we know, ( 2n )2 ≡ 0 mod 4 and ( 2n + 1)2 ≡ 1 mod 4
www.pdfworld.in
www.pdfworld.in
Case I If a, b and c all are odd then,

a 2 + b 2 + c 2 ≡ 3 mod 4
whereas a 2b 2 ≡ 1 mod 4.
It will never give an equality, so the given equation has no integral solution.
Case II If two numbers are odd and one is even, then
a 2 + b 2 + c 2 ≡ 2 mod 4
whereas a 2b 2 ≡ 0, 1 mod 4
Again we get no integral solution.
Case III If two even and one odd.
a 2 + b 2 + c 2 ≡ 1 mod 4
whereas a 2b 2 ≡ 0 mod 4
Again, no integral solution
Case IV If all are even then
a 2 + b 2 + c 2 ≡ 0 mod 4
whereas a 2b 2 ≡ 0 mod 4
Now, there is a possibility to have a solution and the only possible solution is (0, 0, 0) which is the only
trivial solution.
Now, let us come again to the discussion of
x 2 + y2 = k
we have seen, if k = 4m + 3 there is no integral solution now, if k is even then it will be either of the form
k = 4m
or k = 4m + 2
considering k = 4m
If m can be expressed as i + j , where i and j are non-negative integers such that
2 2
(i) i ≠ j , then there will be four integral solutions and 8 ordered pairs.
(ii) if i = j or m can be written as i 2 + 02, then there will be two integral solutions and 4 ordered pairs.
Let us consider some examples.
e.g., x 2 + y 2 = 20
here, 20 is of the form 4(5) and 5 can be expressed as 5 = 12 + 22 which gives i = 1 and j = 2 so, it implies
there are 4 integral solutions, which will be of the form ± 2i and ± 2j also we have 8 ordered pairs.
Therefore in this case we have (2, 4) as one of the solutions and other solutions are
(2, –4), (–2, –4), and (–2, 4) also (4, 2), (4, –2), (–4, –2) and (–4, 2) keep this thing in mind there are only 4
integers used.
e.g., x 2 + y2 = 8
Here, 8 = 4(2)
and 2 = 12 + 12
which gives i = j .
So, it implies there are 2 integral solutions which will be of the form 2i and 2j also we have 4 ordered
pairs therefore in this case we have our solutions are
(2, 2), (–2, –2), (2, –2) and (–2, 2)
keep this thing in mind there are only 2 integers used.
www.pdfworld.in
www.pdfworld.in
e.g., x 2 + y2 = 4
Here, 4 = 4(1)
and 1 = 12 + 02
which gives i = 1 and j = 0
So, it implies there are 2 integral solutions, which are of the form 2i and 2j also we have 4 ordered pairs.
Therefore in this case the solutions are
( 2, 0), (−2, 0), (0, 2) and (0, –2).
e.g., x 2 + y 2 = 24
Here, 24 = 4 × 6
and 6 can't be represented as i + j so it will not have any integral solution.
2 2
e.g., x 2 + y 2 = 12
Here, 12 = 4 × 3
and 3 again can't be expressed as i 2 + j 2 so it will also not have any integral solution.
Now, considering k = 4m + 2 and if m can be written as (i 2 + j 2 + i + j ) and if i ≠ j , then these will be
4 integers and 8 ordered pairs of solutions.
And if i = j , then there will be 2 integers and 4 ordered pairs of solutions.
e.g., x 2 + y 2 = 10
Here, 10 = 4 ( 2) + 2
and 2 = (1)2 + (0)2 + (1) + (0)
Therefore there will be four integral solutions which will be given as ± (2i + 1) and ± (2 j + 1) and in this
case the solutions are ± 3 and ± 1 which will give eight ordered pairs as (3, 1), (3, –1), (–3, 1) and (–3, –1)
also (1, 3), (1, –3), (–1, 3) and (–1, –3).
e.g., x 2 + y2 = 2
Here, 2 = 4 (0) + 2
and 0 = (0)2 + (0)2 + (0) + (0)
Therefore there are only two integral solutions which will again be given as ± ( 2i + 1) and ± ( 2 j + 1)
and in this case the solutions are ± 1 and ± 1, which will give four ordered pairs as (1, 1), (1, –1), (–1, 1) and
(−1, −1).
e.g., x 2 + y 2 = 18
Here, 18 = 4 (4) + 2
and 4 = (1)2 + (1)2 + (1) + (1)
So i = 1 and j = 1
⇒ i = j
Therefore it will have 2 integral solutions which will be given as ± ( 2i + 1) and ± ( 2 j + 1) and in this case
the solutions are ± 3 and ± 3 which gives four ordered pairs as
( 3, 3), ( 3, − 3), (−3, 3) and (−3, − 3)
e.g., x 2 + y 2 = 14
Here, 14 = 4 (3) + 2
and 3 can't be expressed as (i 2 + j 2 + i + j ) as it is always an even number and an even number can't be
equal to an odd number. So it implies if right hand side is (4m + 2) and m is an odd number. So the
equation will never produce any integral solution.
Now, we will extend this concept for an odd number in right hand side of the equation.
www.pdfworld.in
www.pdfworld.in
x 2 + y2 = k
i.e., k = 4m + 1
or k = 4m + 3
As it has been already discussed (k = 4m + 3) will not produce any integral solutions.
So, considering k = 4m + 1, only.
If m = i 2 + j 2 + j,
then there will be an odd integer and an even integer, if i ≠ 0 and j ≠ 0 or i ≠ 0 and j = 0 , then there are
four integers and 8 ordered pairs which will satisfy the equation.
So, one of the integral solution is ± 2i and other is ± ( 2 j + 1).
Now, if i = 0, j ≠ 0, then there are two integers and four ordered pairs which will satisfy the equations.
So, one of the integral solution is 0 and other is ± ( 2 j + 1) .
e.g., x 2 + y 2 = 21
Here, 21 = 4 × 5 + 1
But 5 cannot be written as i + j + j , so it will not give any integral solution.
2 2
Exceptional case :
If x 2 + y2 = k
and k is an odd and a perfect square, then perform the following test always.
Take square root of k, which will come out to be as k, now subtract “1” from it we get ( k − 1) always
double it, so it becomes 2( k − 1), now add “1” to it which becomes 2( k − 1). If this value is a perfect
square say, it is a 2 , then the equation will always have 6 integers and 12 ordered pairs as its solutions
and the integers will be ± a , ± ( k − 1), ± k and 0. Always keep this thing in mind k is an integer.
And if the test fails, then equation will be solved by the method discussed earlier.
e.g., x 2 + y 2 = 169
here 169 = 4(42) + 1, which is of the form (4m + 1) and also it is an odd perfect square so we will have to
perform the mentioned test.
e.g., 169 = 13
13 − 1 = 12
12 × 2 = 24
24 + 1 = 25
and 25 = 52
∴ we will have 4 integers in which ± 5, ± 12, will form 8 ordered pairs (5, 12), (5, –12), (–5, 12), (–5, –12), (12, 5),
(12, –5), (–12, 5), (–12, –5) also there will be three ± 13, 0 which will form four pairs (13, 0),(–13, 0), (0, 13),
(0, –13)
e.g., x 2 + y 2 = 49
here 49 = 4 × 12 + 1, which is of the form (4m + 1) and also an odd perfect, so we will again perform the
mentioned test.
49 = 7
7−1= 6
6 × 2 = 12
12 + 1 = 13
but 13 is not a perfect square therefore the solution will be checked by the earlier method.
12 = (0)2 + (3)2 + (3)
Here, i = 0 and j = 3
so the solutions will be 0 and ± 7. Also the ordered pairs will be (0, 7), (0, –7), (7, 0) and (–7, 0)
www.pdfworld.in
www.pdfworld.in
Concept Solving of the equation of the form xy = n.

If we are asked to find the number of positive integral solution for xy = n, we first write n is the form
α α α
p1 1 . p 2 2 . p3 3 . The number of positive integral solution is same as the number of divisors of n which is
equal to (α1 + 1)(α 2 + 1)(α3 + 1)…
Let us consider example xy = 8 = 23
Number of divisors of 8 are 3 + 1 = 4. So there are 4 integral solution and 4 ordered pair namely (1, 8)
(8, 1) (2, 4) (4, 2)
Now let us consider another example xy = 72(x + y )
we can write it as (x − 72)( y − 72) = 722.
Let x − 72 = X , y − 72 = 722
∴ XY = 722 = 26.34
Number of solutions are 35.
A
Concept The area of a ∆ formed by pythogorean triplet with
integer sides is always divisible by 3.
1
Area of ∆ ABC = BC × AB
2 k2 + 1
1
= 2x (x 2 − 1)
2 k2 – 1
= (x 3 − x )
Let p (x ) = x 3 − x
B C
2k
For x > 1 we use induction
p(2) = 8 − 2 = 6 , p(2) is true
Let p (x ) be true for n = m
p (m ) = m3 − m = 3c
∴ p (m + 1) = (m + 1)3 − (m + 1)
= (m + 1)3 − (m + 1) = m3 + 3m 2 + 2m
= 3c + m + 3m 2 + 2m = 3(c + m + m 2 )
∴ P (m + 1) is true.
Concept The radius of the circumcircle of a ∆ formed by pythogorean A
triplet cannot be integer.
The hypotenuse of the ∆ ABC is the diameter of the circle.
Let us consider the pythogorean triplet x 2 − 1, 2x , x 2 + 1
Here x 2 + 1 is hypotenuse. Since x is an even number its square is also even,
therefore an even number plus one is an odd number.
∴ x 2 + 1 is an odd number B C
x2 + 1
∴ Radius of circumcircle is
2
Concept For any natural number x for
x = 0, 1, 2, ... , n
2n + 1, 2x (22x − 2x − 1), 2x (22n − 2x
+ 1)
2n − 2x
AC = [2 (2
2 x
+ 1)]
2
www.pdfworld.in
www.pdfworld.in
− 4x − 2x + 1
= 22x (24n + 22n + 1) C
− 4 x + 2x
= (24n + 22n − 2x + 1 + 2x + 22x )
− 2x +1
AC 2 = (24n + 22n + 2 2x )
− 2x +1 2
2x(22n–2x + 1) 2x + 1
AB 2 + BC 2 = [22x (22n − 1)2 ] + (2n )
− 4x − 2x + 2
= 22x (24n − 2 ⋅ 22n + 1) + 22n
− 2x
= 2 4n + 22x + 22n ⋅ 22 − 22n ⋅ 2
− 2x A 2x (22n–2x – 1) B
= 2 4n + 22x + 4 ⋅ 22n − 22
. 2n
− 2x
= 2 4n + 22x + 22
. 2n
− 2x +1
= 2 4n + 22x + 22n
Example 1 Prove that there are no natural numbers, which are solutions of15x 2 − 7y 2 = 9.
Solution 15x 2 − 9 = 7y 2
3( 5x 2 − 3) = 7y 2
⇒ 7y is a multiple of 3.
2
⇒ y is a multiple of 3.
Let y = 3z
3( 5x 2 − 3) = 7 × 9z 2
5x 2 − 3 = 21z 2
5x 2 = 21z 2 + 3
5x 2 is a multiple of 3.
⇒ x is a multiple of 3
Let x = 3u
15u 2 = 1 + 7z 2
15u 2 − 6z 2 = 1 + z 2
1 + z 2 is a multiple of 3.
But for any z between 0 to 9, 1 + z 2 is not a multiple of 3.
For any z, the given equation has no integral solution.
Aliter
Since RHS is odd, x and y must be opposite
i.e., one even and one odd.
As 3|15 and 3|9
∴ 3 must divide 7y 2.
Let y = 3y1 so 5x 2 − 21y12 = 3
Again, since 3 divide 21 so 3 must divide 5x 2.

Let x = 3x 1 , we get
15x 12 − 7y12 = 1
⇒ 15x 12 = 7y12 + 1
www.pdfworld.in
www.pdfworld.in
Last digit of perfect square y12 may be one of these values 0, 1, 4, 9, 6, 5.

Hence, last digit of 7y12 + 1 will be 1, 8, 9, 4, 3, 6 respectively.
But 15x 12 ends in 0 or 5.
∴ 15x 12 = 7y12 + 1 has no solutions.
Example 2 Show that x 2 + 1 = 3y has no solutions in integers.

Solution Since LHS cannot be a multiple of 3 for any x between 0 to 9.
RHS is always a multiple of 3.
∴ x 2 + 1 = 3y has no integral solutions.
Example 3 Show that 21x 2 − 10y 2 = 9 has no solution.

Solution 21x 2 − 9 = 10y 2
⇒ 3 (7x 2 − 3) = 10y 2
⇒ 10y 2 is a multiple of 3.
⇒ y is a multiple of 3.
Let y = 3y1
3 (7x − 3) = 10 × 9y12
2
7x 2 − 3 = 30y12
7x 2 = 3 + 30y12 ⇒ 7x 2 = 3 (1 + 10y12 )
⇒ 7x 2 is a multiple of 3
So, x is multiple of 3.
Let x = 3x1
7 × 9x12 = 3 (1 + 10y12 )
21 x12 = 1 + 10y12
21 x12 − 9y12 = 1 + y12
3 (7x12 − 3y12 ) = 1 + y12
⇒ 1 + y12 is a multiple of 3.
But 1 + y12 is not a multiple of 3.
∴ The given equation has no integral solution.
Note Every integer m can be written in the form x 2 + y 2 − 5z 2.
If m = 2n, then
= 2n = (n − 2)2 + ( 2n − 1)2 − 5(n − 1)2
If m = 2n + 1 = (n + 1)2 + ( 2n )2 − 5n 2
Verification,
7 = 2( 3) + 1 = ( 3 + 1)2 + ( 2 × 3)2 − 5( 3)2
= 16 + 36 − 45
= 52 − 45 = 7
Similarly, every integer can be written in the form of
x 2 + y 2 + z 2 − 5u 2
www.pdfworld.in
www.pdfworld.in
1 2n
Example 4 Prove Cn is an integer.
n +1
Solution If a and b are integers and a − b = c, then c is also an integer.

Let a= 2n
Cn ; b =
Cn − 1 2n
2 n ! 2n ! 2n !  n 
2n
Cn − 2nCn − 1 = − = 1−
n ! n ! (n + 1)!(n − 1)! n ! n !  n + 1
 1 
2n !
= n + 1 ⇒ it is an integer.
n !n !
 
(kn )! kn !
kn
Cn − knCn − 1 = −
(kn − n )! n ! (kn − n + 1)!(n − 1)!
(kn )!  (kn − n )! n ! 
= 1−
(kn − n )! n !  (kn − n + 1)!(n − 1)! 
(kn )!  n 
= 1 − kn − n + 1
(kn − n )! n !  
kn − n + 1 − n  kn kn − 2n + 1
= knCn   = Cn  kn − n + 1  .
 kn − n + 1   
It is an integer.
Example 5 If xy = 22 ⋅ 34 ⋅ 57( x + y ) , find the number of integral solution.

Solution Let N = 22 ⋅ 34 ⋅ 57
xy = N( x + y )
xy = Nx + Ny
xy − Nx − Ny = 0
( x − N )( y − N ) = N 2
( x − N )( y − N ) = 24.38.514
Number of integral solution
= ( 4 + 1)( 8 + 1)(14 + 1) = 5 × 9 × 15 = 675
Example 6 Find all positive integers x, y satisfying

1 1 1
+ =
x y 20
Solution Suppose x, y are two positive integers such that
1 1 1
+ = …(i)
x y 20
1 1 1 x − 20
then = − =
y 20 x 20 ⋅ x
1 x + 20 − 4 5x
∴ =
y 20x
Implying that 5x is rational.
www.pdfworld.in
www.pdfworld.in
Now x ∈ N ⇒ 5x ∈N. Hence 5x is the square of an integer which is divisible by 5.

∴ 5x = ( 5a )2 for some a ∈ N i.e., x = 5a 2 similarly y = 5b 2 for some b ∈ N.
Now, Eq. (i) becomes
1 1 1
+ = ⇒ 2(a + b ) = ab
a b 2
(a − 2)(b − 2) = 4
⇒ (a, b ) ∈{( 3, 6), ( 4, 4), ( 6, 3)}.
∴ Solution set is {(45, 180), (80, 80), (180, 45)}.
Example 7 Find the number of solutions in positive integers of the equation 3x + 5y = 1008.
Solution Let x , y ∈ N such that 3x + 5y = 1008 then 3 | 5y ⇒ 3|y ⇒ y = 3k for some k ∈ N
Now, 3x + 15k = 1008
⇒ x + 5k = 336
⇒ 5k ≤ 335
⇒ k ≤ 67
Thus, any solution pair is given by ( x , y ) = ( 336 − 5k , 3k ) where 1 ≤ k ≤ 67.
∴ Number of solutions is 67.
Example 8 Prove that there do not exist positive integrs x , y , z satisfying

2xz = y 2 and x + z = 997.
Solution 2|y 2 ⇒ 4|2xz ⇒ 2|x or z ⇒ 2|x and z [Q 2 | ( x + z )]
Let x = 2x1, y = 2y1 and z = 2z1
Then 2x1z1 = y12 and x1 + z1 = 997
Again y1 and one of x1, z1, say x1, are even writing x1 = 2x 2 and y1 = 2y 2
We have x 2z1 = y 22 and 2x 2 + z1 = 997
Since, 997 is a prime, x 2 and z1 are relatively prime.
∴ Each is a square since their product is square.
Since any square is of the form 8n, 8n + 1 or 8n + 4, 2x 2 ≡ 0 or 2(mod 8) and z1 ≡ 1
(mod 8) (Q z1 is odd).
Hence 2x 2 + z1 ≡ 1 or 3 (mod 8).
A contradiction as 997 ≡ 5 (mod 8).
Example 9 Show that the equation 3x 10 − y 10 = 1991 has no integral solution.

Solution Suppose the existence of x , y ∈ Z such that 3x 10 − y 10 = 1991. Note that 11|1991.
∴ Neither x nor y is divisible by 11 for otherwise 11 would divide both .
⇒ 1110|3x 10 − y 10 = 1110|1991
an impossibility.
Hence x and y are prime to 11.
∴ x 10 ≡ y 10 ≡ 1 (mod 11)
⇒ 1991 = 3x 10 − y 10 ≡ 3 − 1 = 2 a contradiction.
www.pdfworld.in
www.pdfworld.in
Example 10 Find all integral solutions of

x 4 + y 4 + z 4 − t 4 = 1991
Solution Let n be any integer ; when it is odd n 4 − 1 = (n − 1)(n + 1)(n 2 + 1) is divisible by 16 as
n − 1, n + 1, n 2 + 1 are all even and n − 1,n + 1being consecutive even integers one of
them is divisible by 4 . When n is even n 4 ≡ 0 (mod 16).
Thus, n 4 ≡ 0 or 1, Now for any x , y , z , t , ∈ Z
x4 + y4 + z4 − t4 ≡ α
where α ∈ {−1, 0, 1, 2, 3}, since 1991 ≡ 7
x 4 + y 4 + z 4 − t 4 ≠ 1991
Example 11 For n ∈ N, let s(n) denote the number of ordered pairs ( x , y ) of positive integers for
1 1 1
+ = . Determine the set of positive integers n for which s(n ) = 5 .
which
x y n
1 1 1
Solution + = ⇒ x, y > n
x y n
∴ x = n + a and n + b, a, b ∈ N.
1 1 1
Now, + =
n+a n+b n
⇒ (n + b + n + a )n = (n + a )(n + b )
⇒ n 2 = ab
∴ s(n ) is the number of divisors of n 2
α
Let n = p1 1K pmαm be prime factorization of n where α 1 ≥ K ≥ α m .
Then s(n ) = (1 + 2α 1) K (1 + 2α m )
∴ s(n ) = 5
⇒ 1 + 2α 1 = 5
and m =1
∴ n = p12
Required set is {p 2 : p is prime}.
1 1 1
Example 12 If + = ;a, b, c are positive integers with no common factors. Prove that (a + b ) is a
a b c
square.
a+b 1
Solution By the hypothesis =
ab c
i.e., (a + b )c = ab
Let p be any prime which divides (a + b ); then p divides one of a, b and therefore
both.
Since gcd (a, b, c ) = 1; p does not divide c.
∴ for any k ∈ N, pk | a ⇔ pk | b
Hence the maximum power of p which divides a + b is the square of the maximum
power of p which divides a.
∴ a + b is a square.
www.pdfworld.in
www.pdfworld.in
3n − 5
Example 13 Find all integers n such that is also an integer.
n +1
3n − 5 8
Solution Since, = 3− is an integer.
n +1 n +1
8
∈ {± 1, ± 2, ± 4, ± 8}
n +1
8
and 3− is a square.
n +1
8
Consequently = − 1 or 2 ;
n +1
i.e., n = − 9 or 3
Example 14 Find the number of pairs of integers (a, b ) such that

a 3 + a 2b + ab 2 + b 3 + 1 = 2002.
Solution a 3 + a 2b + ab 2 + b 3
= (a + b ) (a 2 + b 2 ) = 2001
= 3 × 23 × 29
a + b is therefore one of the three numbers 3, 23, or 29.
If a + b = 3, then a = 1, b = 2
(or a = 2, b = 1) so that a 2 + b 2 = 5.
But in this case a 2 + b 2 = 23 × 29
∴a + b is not 3.
If a + b = 23, then a 2 + b 2 = 87
so that both a and b will be less than 10 and a + b < 20, a contradiction.
If a + b = 29, then a 2 + b 2 = 69 so that both a and b will be less than 10 and
a + b < 20, a contradication.
Thus, the number of pairs (a, b ) satisfying the given condition is zero.
www.pdfworld.in
www.pdfworld.in
Additional Solved Examples

Example 1. How many zeros does 100, ends with ?
Solution If e is the maximum power of 5 in 100!, then
∞ 100 !
e = Σ  i =
100   100   100 
+ + = 20 + 4 + 0 = 24
 5   5   52   53 
i = 1
Hence, 100! ends with 24 zeros.
Example 2. If n! has exactly 20 zeros at the end, find n. How many such n are there ?
Solution Let e be the maximum power of 5 in n! then
∞ ∞ n ∞
e = Σ  i  < Σ  i  =
n n 1
Σ
 5  i = 1  5  5 i = 1 5i − 1
i =1 
n
n  a 
⇒ e= 5 = ∴ S ∞ = 
1−
1 4  1−r
5
∴ n > 4e
Here e is given to be 20.
∴ n ≥ 80 for 80, e = 19
∴ 85 is the required answer.
86, 87, 88, 89 are also valid values of n. If solution exist for this type of problem there will be
5 solutions.
Example 3. Find number of zeros at the end of (5n − 1)!.

Note First student should remember [ x + k ] = x
[ x − k ] = x − 1, 0 ≤ k < 1 .
Solution We will find highest power of 5 in (5n − 1)!

∞  5n − 1 
e= Σ  i 
i =1
 5 
 5n − 1   5n − 1   5n − 1   5n − 1 
= +  2  +  3  + K +  n −1 
 5   5   5  5 
1  n 1  n 1 5 − 1 
= 5n −1
− + 5 −2
− + 5 −3
− + K+
 5   52   53   5n − 1 
−1 −2 −3
= ( 5n − 1) + ( 5n − 1) + ( 5n − 1) + K + ( 5 − 1) [Q[x − h ] = x − 1]
n −1 n −2
= (5 +5 + K + 5) − (n − 1)
n −1
5 (5 − 1)
= − (n − 1)
5−1
( 5n − 4n − 1)
=
4
5n − 4n − 1
∴ Total number of zeros =
4
www.pdfworld.in
www.pdfworld.in
Example 4. Find highest power of 15 in 100!.

Solution Here, 15 = 3 × 5 we first find highest power of 3.
E3 (100 !) = 48 (QE3 means exponent of 3.)
and E5 (100 !) = 24
∴ E15 (100!) = min ( 24, 48) = 24
Example 5. Find the exponent of 6 in 33!.

Solution Here, 6=2×3
E 2 (33 !) = 31
E3 (33 !) = 15
E 6 (33 !) = min (31, 15) = 15
Example 6. Prove that 33! is divisible by 215.

Solution We first find highest power of 2 in 33!
=   +  2  + K +  5  = 16 + 8 + 4 + 2 + 1 = 31
33 33 33
 2   2   2 
Hence, exponent of 2 in 33! is 31.
i.e., 33! is divisible by 231.
31 15
But 2 is also divisible by 2 .
∴ 33! is divisible by 215.
−1
Exmaple 7. Let N = 2n (2n − 1) and (2n − 1) is a prime number.
1 1 1
1 < d1 < d 2 < K < d k = N are divisor of N. Show that 1 + + + K+ =2
d1 d 2 dk
Solution Let 2n − 1 = q
∴ Divisor of N are
1, 2, 22 , K , 2n − 1 , q , 2q , 22q ,... , 2n − 1q
1 1 1
Let S =1+ + K+
d1 d 2 dk
=  1 + + 2 + K + n − 1  +
1 1 1 1
 2 2 2  q
 1 1 1 
1 + + 2 + K + n − 1 
 2 2 2 
1 − 1  1 − 1 
 n   
2  1  2n 
⇒ S =1 +
1 q 1 − 1 
1−  
2  2
2n − 1 1 2n − 1
⇒ S = n −1
+
2 q 2n − 1
q ( 2n − 1) + ( 2n − 1) ( 2n − 1)( q + 1)
= n −1
= −1
q2 q ⋅2n
( 2n − 1) ( 2n ) 2n
= n −1
= n −1
=2
( 2 − 1) 2
n
2
www.pdfworld.in
www.pdfworld.in
−1
Example 8. Show that n = 2m ( 2m − 1)is a perfect number, if ( 2m − 1) is a prime number.
Solution As we know by the definition of perfect numbers that if the sum of the divisors of a number n,
other than itself, is equal to n, then n is called a perfect number.
−1 −1
Let n = 2m × p where p = 2m − 1 is prime number. Divisors of 2m × p are 1, 2, 22 , 23 , K,
m −1 2 m −2 m −1
2 , p , 2p , 2 p , ... , 2 p, 2 p.
−1
Now, we should sum all these divisors excepting the last one, i.e., 2m p
m −1 m −2
S = (1 + 2 + 2 + K + 2 2
) + p (1 + 2 + 2 + K + 2
2
)
m −1
1 ( 2 − 1) p [1 ( 2
m
− 1)] −1
= + = 2m − 1 + p(2m − 1)
2−1 2−1
−1 −1
= 2m + p2m − p − 1 = 2m ( 2 + p ) − ( p + 1)
m −1
=2 (1 + 2 ) − 2
m m
[Qp = 2m − 1]
−1
= 2m ( 2m − 1) = n
Example 9. Prove that product of four consecutive positive integers increased by 1 is a perfect square.
Solution Let the consecutive positive integers be
n , n + 1, n + 2, and n + 3.
Consider the expression, N = n (n + 1)(n + 2)(n + 3) + 1 = (n 2 + 3n )(n 2 + 3n + 2) + 1
= (n 2 + 3n )2 + 2(n 2 + 3n ) + 1 = [(n 2 + 3n ) + 1]2 = (n 2 + 3n + 1)2
Example 10. Three consecutive positive integers raised to the first, second and third powers respectively,
when added make a perfect square, the square root of which is equal to the sum of three consecutive
integers, find these integers.
Solution Let (n − 1), n , n + 1 be the three consecutive integers.
Then, (n − 1)1 + n 2 + (n + 1)3 = (3n )2 = 9n 2
⇒ n − 1 + n 2 + n3 + 3n 2 + 3n + 1 = 9n 2
⇒ n3 − 5n 2 + 4n = 0 ⇒ n (n − 1)(n − 4) = 0
⇒ n = 0 or n = 1 or n = 4
But n = 0 and n = 1 will make the consecutive integers −1, 0, 1 and 0, 1, 2 which contradicts the
hypothesis that the consecutive integers are all greater than zero.
Hence n = 4, corresponding to which the consecutive integers are 3, 4 and 5.
Example 11. Determine the sum of all the divisors d of 1988 − 1 which are of the form d = 2a ⋅3b
with a , b > 0.
Solution 1988 − 1 = (1 + 18)88 − 1 = (1 − 20)88 − 1
In the binomial expansion of (1 + 18)88, first term is 1 and second term is 18 all other terms are divisible
by 182.
∴ Maximum power of 3 divides (1 + 18)88 − 1 is 32. In the expansion of (1 − 20)88, first term is 1 and
88.87
second term is – 20.88 = − 25.55 and third term is 202 × = 26.52.11.87 and therefore (120
. )88 − 1
2
divisible by 26. Hence maximum power of 2 which divides (1 − 20)88 − 1 is 25.
Hence the sum of factors of 1988 − 1 which are of the form a 2a ⋅ 3b where a , b > 0 is
(2 + 22 + 23 + 24 + 25 )(3 + 32 ) = 744
www.pdfworld.in
www.pdfworld.in
Example 12. Determine with proof all the arithmetic progression with integer terms, with the property
that for each positive integer n, the sum of first n terms is a perfect square.
Solution When n = 1, the first term itself is a perfect square. Which is k 2(say)
n
The sum to n terms of the AP is Sn = [2a + (n − 1) d ], where a = k
2
Since, Sn is a perfect square for every n the nth term 2a + (n − 1) d > 0 for every n and hence d > 0 .
p
If n is an odd prime, say p, then Sp = [2a + ( p − 1) d ] .
2
Since, Sp is a perfect square p | [2a + ( p − 1) d ] i.e., p | [(2a − d ) + pd ]
But p | pd , so p | (2a − d ). This is possible for all prime p, if and only if, 2a − d = 0 or 2a = d i.e., d = 2k 2.
So, the required AP is
k 2 , 3k 2 , 5k 2 , K , (2n − 1) k 2
k is any natural number.
Example 13. Show that 13 divides 270 + 370.

Solution We can write 270 + 370 as 435 + 935 and 35 is odd. Now an + bn is divisible by a + b when n is odd.
∴ It follows 435 + 935 is divisible by 13.
Example 14. Determine all positive integer n for which 2n + 1 is divisible by 3.

Solution 2n + 1 is divisible by 3.
⇒ (3 − 1)n + 1 is divisible by 3.
⇒ (−1)n + 1 is divisible by 3.
⇒ n is odd.
∴ Set of all odd natural number in the set of all those positive integer for which 2n + 1 is divisible by 3.
Example 15. Prove that for every positive integer 1n + 8n − 3n − 6n is divisible by 10.
Solution Since, 10 is the product of two prime 2 and 5. It is sufficient to show that given expression is
divisible by both 2 and 5.
So we use the fact an − bn is always divisible by a − b, when n is an integer.
Let A = 1n + 8n − 3n − 6n or A = (8n − 3n ) − (6n − 1n )
We find that 8n − 3n and 6n − 1n is divisible by (8 − 3) = (6 − 1) = 5
Again writing A = (8n − 6n ) − (3n − 1n ) is divisible by 2
∴ A is divisible by 10.
Example 16. Show that 11997 + 21997 + K + 19961997 is divisible by 1997.

Solution We shall make groups of the terms
(11997 + 19961997 ) + (21997 + 19951997 ) + K + (9981997 + 9991997 )
+1 +1
Here each bracket is of the form ai2n + bi2n . It is divisible by (ai + bi )
But ai + bi = 1997 for all i
∴ Each bracket and hence their sum is divisible by 1997.
www.pdfworld.in
www.pdfworld.in
Example 17. Find the remainder of 2100 when divided by 3.

Solution We use the concept of congruent modulo.
2 ≡ 2 mod 3.
∴ a ≡ a mod m
Raising the power to 5
25 ≡ 32 mod 3
or 32 ≡ 2 mod 3 ∴2 is the remainder when 32 is divided by 3.

⇒ 25 ≡ 2 mod 3
Raising the power to 4.

220 ≡ 16 mod 3
or 16 ≡ 1 and 3 ∴1 is the remainder when 16 is divided by 3.

⇒ 220 ≡ 1 mod 23
Raising the power to 5.

2100 ≡ 1 mod 3
∴ 1 is the remainder when 2100 is divided by 3.
Example 18. Find the remainder when 2100 + 3100 + 4100 + 5100 is divided by 7.
Solution 2100 ≡ 2 mod 7
3100 ≡ 4 mod 7, mod m
4100 ≡ 4 mod 7,
5100 ≡ 2 mod 7,
∴ a ≡ b mod m 
∴ 2100 + 3100 + 4100 + 5100 ≡ 12 mod 7  c ≡ d mod m 
 
 a + c ≡ b + d mod m 
 
But 12 ≡ 5 mod 7
∴ 2100 + 3100 + 4100 + 5100 ≡ 5 mod 7
∴ Remainder is 5.
Example 19. Find remainder when 1 ! + 2 ! + 3 ! + K + 100! is divided by 24.

Solution 4 ! ≡ 0 mod 24 [Q 4 ! = 4 × 3 × 2 × 1]
5 ! ≡ 0 mod 24 = 24
100 ! ≡ 0 mod 24
∴ 4 ! + 5 ! + K + 100 ! ≡ 0 mod 24
Now, 1 ! + 2 ! + 3 ! ≡ 9 mod 24
∴ 1 ! + 2 ! + 3 ! + 4 ! + 5 ! + ... + 100 ! ≡ 9 mod 24
∴ 9 is the remainder when
1 ! + 2 ! + 3 ! + K + 100 ! is divided by 24.
www.pdfworld.in
www.pdfworld.in
Example 20. Show that 6n ≡ 6 mod 10 ∀ n ∈ N

Solution Let P (n ) : 6n ≡ 6 mod 10
P (1) : 6 ≡ 6 mod 10
Let us assume P (n ) is true for n = k
6k ≡ 6 mod 10
∴ 6 − 6 ≡ 10 m for some m ∈ Z
k
∴ 6k ≡ 10m + 6
Now, consider 6k + 1 − 6 = 6k ⋅ 6 − 6 = 6(10m + 6) − 6 = 60m + 30 or 10(6m + 3)
Thus, 10 divides 6k + 1 − 6
∴ 6k + 1 ≡ 6 mod 10
Example 21. What is the remainder when

15 + 25 + 35 + K + 995 + 1005 is divided by 4 ?
Solution We observe (2n )5 ≡ 0 mod 4

and (2n − 1)5 ≡ (2n − 1) mod 4
100 50 50
Now Σ r 5 = Σ (2r )5 + Σ (2r − 1)5
r =1 r =1 r =1
100 50
Σ r 5 ≡ 0 + Σ (2r − 1) (mod 4)
r =1 r =1
 n 2
≡ 502 mod 4 ≡ 0 mod 4 Qr Σ= 1 ( 2r − 1) = n 
 
∴ Remainder is 0.
Example 22. Find the remainder when (2222)5555 is divided by 7.

Solution 2222 ≡ 3 mod 7
( 2222)3 ≡ 27 mod 7
and 27 ≡ − 1 mod 7
∴ ( 2222)3 ≡ − 1 mod 7
( 2222)5553 ≡ (−1)1851 mod 7
( 2222)5553 ≡ − 1 mod 7
( 2222)2 ≡ 9 mod 7
(2222)5555 ≡ − 9 mod 7 −9 ≡ 5 mod 7
∴ (2222)5555 = 5 mod 7
Example 23. Show that, if the sum the square of two whole numbers is divisible by 3, then each of them is
divisible by 3.
Solution Let x and y be any two integers.
x ≡ 0 mod 3
x ≡ 1 mod 3
x ≡ 2 mod 3
www.pdfworld.in
www.pdfworld.in
x 2 ≡ 0 mod 3
x 2 ≡ 1 mod 3
x ≡ 2 mod 3
x 2 ≡ 4 mod 3
x 2 ≡ 1 mod 3
x ≡ 1 mod 3
x 2 ≡ 1 mod 3
y 2 ≡ 0 mod 3
y 2 ≡ 1 mod 3
x 2 + y 2 ≡ 0 mod 3 …(i)
x + y ≡ 1 + 1 = 2 mod 3
2 2
…(ii)
x + y = 1 mod 3
2 2
…(iii)
In Eqs. (ii) and (iii), x + y is not a multiple of 3. In Eq. (i) x + y is multiplying of 3. But Eq. (i) is the
2 2 2 2
result of adding x 2 ≡ 0 (mod 3) and y 2 ≡ 0 (mod 3) implying both x 2 and y 2 and hence both are divisible
by 3.
Example 24. Find the last digit of 4317 .

Solution For finding last digit always use congruence with 10.
43 ≡ 3 mod 10
(43)17 ≡ 317 mod 10
i.e., last digit of (43)17 is same as 317
Now, 34 = 81 ≡ 1 mod 10
317 = 34 × 4 + 1 ≡ 1 × 1 × 1 × 1 × 3 mod 10
317 ≡ 3 mod 10
∴ Number in unit place is 3.
Example 25. Find the last two digit of 31997 .

Solution This is same as asking what is remainder when 31997 ÷ 100
34 ≡ 81 mod 100
38 ≡ 61 mod 100
312 ≡ 41 mod 100
316 ≡ 21 mod 100
320 ≡ 1 mod 100
Now, 340 , 360 , 380 , 3100 , K , 31980 all are ≡ 1 mod 100
We know, 316 ≡ 21 mod (100)
317 ≡ 21 × 3 mod 100
317 ≡ 63 mod 100
∴ 31997 = 31980 × 317
Q 31980 ≡ 1 mod 100
www.pdfworld.in
www.pdfworld.in
and 317 ≡ 63 mod 100

∴ 31997 ≡ 63 mod 100
∴ Last two digit is 63.
Example 26. Find the unit digit and ten's digit of 1 ! + 2 ! + 3 ! + 4 ! + K + 1997 !.
Solution Let S = 1 ! + 2 ! + 3 ! + 4 ! + K + 1997 !
From 5! all the numbers will have unit digit 0 and also1 ! + 2 ! + 3 ! + 4 ! = 33
∴ Unit digit of S is 3
Now, from 10! all the unit and ten's digit will be zero and also
1 ! + 2 ! + 3 ! + 4 ! + 5 ! + 6 ! + 7 ! + 8! + 9 ! = 33 + 120 + 720 + 5040 + 40320 + 362880
So, to get the ten's digit of S add only the tens digit of 33 + 120 + K + 362880 which is
3 + 2 + 2 + 4 + 2 + 8 = 21
∴ Tens digit is 1.
Example 27. Find last two digit of (1 ! + 2 ! + 3 ! + K+ 100 !)2

Solution We know that,
1 ! + 2 ! + 3 ! + K + 100 ! ≡ 13 mod 100
⇒ (1 ! + 2 ! + 3 ! + K + 100 !)2 ≡ 169 mod 100
⇒ 169 ≡ 69 mod 100
⇒ (1 ! + 2 ! + 3 ! + .. + 100 !)2 ≡ 69 mod 100
∴ Last two digit is 69.
Example 28. Find the remainder when P = 11 + 22 + 33 + 44 + K + 5050 is divided by 8.

Solution All odd number are going to be of the type 4n ± 1.
Now, (4n ± 1)2 = 16n 2 ± 8n + 1
Thus, any even power of 4n ± 1 will leave remainder 1. Thus, odd power of 4n ± 1 will leave remainder
4n ± 1 i.e., 1, 3, 5, 7 when we reduce the remainder less than 8.
Now in this problem all terms with odd base will leave remainder 1,3, 5, 7, … and so on. There will be 6
sets of (1, 3, 5, 7) as remainder and one extra 1 as remainder from 4949.
Since in the expression all the terms are added we will have the sets of (1 + 3 + 5 + 7) leaving zero as the
remainder. Thus, the remainder from all terms with odd base is 1.
Terms with even base 22 will leave remainder 4 and all terms from 44 will be fully divisible by 8 as they
would be multiple of 23.
Thus, net remainder is, 1 + 4 = 5 .
Example 29. Show that (1 ! + 2 ! + 3 ! + 4 !)5 is of the form 5k + 3.

Solution We know if p is prime, then
(a + b + c ) p = ap + bp + c p + m (p ) [Fermat theorem]
∴ (1 ! + 2 ! + 3 ! + 4 !) = (1 !) + (2 !) + (3 !) + (4 !) + m (5)
5 5 5 5 5
Now 15 ≡ 1 mod 5
(2 !) 5 ≡ 2 mod 5
(3 !) 5 ≡ 1 mod 5
www.pdfworld.in
www.pdfworld.in
(4 !)5 ≡ 4 mod 5
∴ (1 !)5 + (2 !)5 + (3 !)5 + (4 !)5 ≡ 8 mod 5
or (1 !)5 + (2 !)5 + (3 !)5 + (4 !)5 ≡ 3 mod 5
∴ (1 ! + 2 ! + 3 ! + 4 !)5 = 3 mod 5
∴ (1 ! + 2 ! + 3 ! + 4 !)5 is of the form 5k + 3
100
Example 30. Prove that ∑n! is divisible by 1001 but not by 1000.
n = 14
Solution 1001 = 7 ⋅ 11 ⋅ 13
100
∴ Every term of ∑ n ! is divisible by 1001
n = 14
  14   14  + ... = 2
The highest power of 5 which divides the first term is 52 Q  +  52  
 5 
∴ 1000 does not divide 14 ! but every other term divisible by 1000.
100
∴ ∑ n ! is not divisible by 1000.
n = 14
Example 31. (1 + 10 !) (1 + (10 !)2 ) (1 + (10 !)3 ) K (1 + (10 !)100 )]100 is divided by 10!. what is the remainder ?
Solution (1 + x ) n ≡ 1 mod x
(1 + x 2 ) n ≡ 1 mod x 2
M
(1 + x k ) n ≡ 1 mod x k
………………………
………………………
………………………
∴ [(1 + x )(1 + x 2 )... (1 + x k )] n ≡ 1 mod x
Where x is HCF of x , x 2 , K , x k .
If we put x = 10! , then [(1 + 10 !)(1 + (10 !)2 )(1 + (10 !)3 ) K (1 + (10 !)100 )]100 ≡ 1 mod (10 !)
Therefore, 1 is the remainder.
Example 32. Let n = 640640640643, without actually computing n 2. Prove that n 2 leave a remainder 1
when divided by 8.
Solution Since, 640640640000 is a multiple of 8 643 ≡ 3 (mod 8)
∴ n = 8k + 3 for some positive integer k.
∴ n 2 = 64k 2 + 48k + 9 = 1 + a multiple of 8.
Example 33. Prove that n16 − 1 is divisible by 17, if (n , 17) = 1.

Solution (n , 17) = 1 and 17 is a prime number .
∴ By Fermat theorem n17 − 1 ≡ 1 mod 17
[If p is prime and (a , p ) = 1, then ap − 1 ≡ 1 mod p] or n16 ≡ 1 mod 17.
or 17 | (n16 − 1) i.e., (n16 − 1) is divisible by 17.
www.pdfworld.in
www.pdfworld.in
Example 34. If a and b are coprime to the prime number p, then show that ap − 1 − bp − 1 = M ( p ).
Solution Q (a , p ) = 1 and p is prime.
∴ By Fermat theorem, ap − 1 ≡ 1 mod p …(i)
Similarly bp − 1 ≡ 1 mod p …(ii)
Subtracting congruence (i) and (ii)

ap − 1 − bp − 1 ≡ 0 mod p
∴ p divides ap − 1 − bp − 1
∴ ap − 1 − bp − 1 is a multiple of p.
Example 35. If p is a prime number, show that the difference of the pth powers of any two numbers
exceeds the difference of the numbers by a multiple of p.
Solution Let x, y be two numbers
Difference of the pth powers of these numbers
= x p − yp
Difference of these number = x − y
We have to show that (x p − y p ) − (x − y ) is a multiple of p.
i.e., x p − y p ≡ (x − y ) mod p.
Q p is a prime.
x p ≡ x mod p …(i)
y ≡ y mod p
p
…(ii)
Subtracting Eq. (ii) from Eq. (i), we get
x p − y p ≡ (x − y ) mod p
Example 36. Show that n7 ≡ n mod 42.

Solution ap ≡ a mod p, n7 ≡ n mod 7 [Q7 is prime number]
i.e., n − n is divisible by 7
7
…(i)
Again, n − n = n (n − 1)
7 6
= n (n3 − 1)(n3 + 1)
= n (n − 1)(n 2 + n + 1)(n + 1)(n 2 − n + 1)
= (n − 1)n (n + 1)(n 2 + n + 1)(n 2 − n + 1)
Now, (n − 1)n (n + 1) being the product of three consecutive integers is divisible by 3 ! = 6 and hence
n 7 − n = (n − 1)n(n + 1)(n 2 + n + 1)(n 2 − n + 1) is divisible by 6 …(ii)
From Eqs. (i) and (ii), n7 − n is divisible by 42(= 6 × 7)
and (6, 7) = 1
{We know that if a|c, b|c and (a, b) = 1, then ab|c}
www.pdfworld.in
www.pdfworld.in
Example 37. Show that 4th power of every number is of the form 5k or 5k + 1 where k is any positive
integer.
Solution Let a be any number.
Q5 is prime and a is any number
∴ either (a , 5) = 1 or 5|a
Case I (a , 5) = 1
By Fermat theorem, a 4 ≡ 1 mod 5
∴ 5 |a4 − 1
∴ a 4 − 1 = 5k or a 4 = 5k + 1
Case II 5|a
∴ 5|a4
∴ ∃ an integer k such that a 4 = 5k
Hence the result.
Example 38. If m is a prime number and a, b are two numbers less than m, prove that
−2 −3 −4 −2
am + am b + am b 2 + K + bm
is a multiple of m.
Solution Qm is prime and a , b are two numbers both less than m.
∴ (a , m ) = 1 and (b , m ) = 1
[QA prime number is coprime to every number less than it]
∴ By Fermat theorem,
−1
am ≡ 1 mod m …(i)
m −1
b ≡ 1 mod m …(ii)
−1 −1
am − bm ≡ 0 mod m
m −1 m −1
∴ m | (a −b )
m −2 −3 −4 −2
or m | (a − b )[a + am b + am b 2 + K + bm ]
But (m , a − b ) = 1 [Qa < m , b < m ⇒ a − b < m and m is prime]
∴ By Gauss Theorem
−2 −3 −2
m | [am + am b + K + bm ] [QIf a | bc and (a , b ) = 1; then a | c ]
m −2 m −3 m −2
∴ (a +a b + K+ b )
is a multiple of m.
Example 39. If p is prime, prove that ap ( p − 1)! + a is divisible by p.

Solution Qp is prime
By Wilson Theorem, ( p − 1)! ≡ − 1 mod p …(i)
Again, Qp is prime
Now ap ≡ a mod p …(ii)
Multiplying Eqs. (i) and (ii)
ap ( p − 1)! ≡ − a mod p
⇒ ap (p − 1)! + a is divisible by p.
www.pdfworld.in
www.pdfworld.in
Example 40. If p is a prime number, show that

1p − 1 + 2p − 1 + 3p − 1 + K + ( p − 1) p − 1 + 1 ≡ 0 mod p.
Solution Q p is a prime number and 1, 2, 3, ……, are all less than p.

∴ (1, p ) = 1 ; ( 2, p ) = 1 ; ( 3, p ) = 1 ; K ( p − 1, p ) = 1
[QA prime number is coprime to every number less than it]
∴ Putting a = 1, 2, 3, K ( p − 1) in Fermat theorem
ap − 1 ≡ 1 mod p
1p − 1 ≡ 1 mod p
2p − 1 ≡ 1 mod p
3p − 1 ≡ 1 mod p
M M M
M M M
( p − 1) p − 1 ≡ 1 mod p
Adding up all these ( p − 1) congruence, we get
1p − 1 + 2p − 1 + ( p − 1)p − 1 + K+ ≡ ( p − 1) mod p
Also 0 ≡ p mod p
Subtracting these congruence
1p − 1 + 2p − 1 + 3p − 1 + K + ( p − 1) p − 1 ≡ − 1 mod p
or 1p − 1 + 2p − 1 + 3p − 1 + K + ( p − 1) p − 1 + 1 ≡ 0 mod p
Example 41. Show that no square number is of the form 3n − 1 .

Solution Let us suppose if possible N 2 = 3n − 1, N 2 represents any square number.
Then, N 2 + 1 = 3n
i.e., N 2 + 1 is divisible by 3.
But N 2 + 1 = (N 2 − 1) + 2
Now N 2 − 1 is divisible by 3 when N is prime to 3 (Fermat theorem)
Thus, N 2 + 1 exceeds a multiple of 3 by 2 and as such N 2must be of the form 3n + 1 if N is not prime to 3
∴ It must be of the form n 2 = 3n
Thus, no square number is of the form 3n − 1.
Example 42. Prove that every even power of every odd number is of the form 8r + 1.
Solution Any odd number is of the form 2p + 1 and every even number is of the form 2n
∴ ( 2p + 1)2n = [( 2p + 1)2 ] n = (4p 2 + 4p + 1) n
= [4p ( p + 1) + 1)] n … (i)
Since p ( p + 1) is always even.
∴ p ( p + 1) is always even
Also, p ( p + 1) is of the form 2k.
∴ A becomes (8k + 1)n .
Now, (8k + 1)n is always of the form 8k + 1.
www.pdfworld.in
www.pdfworld.in
Example 43. If m and n are positive integer. If (mn − 1) is divided by n, then remainder is always n − 1.
Solution We can write m n − 1 = n (m − 1) + n − 1
∴ Remainder is always n − 1
Example 44. Prove that p ! and ( p − 1)! − 1 are relatively prime if p is an odd prime.
Solution Let ( p !, ( p − 1)! − 1) = d
∴ d | p ! and d | ( p − 1)! − 1
∴ There exist integers k and k such that
p ! = kd or p ( p − 1)! = kd …(i)
and ( p − 1)! − 1 = k ′ d
or ( p − 1)! = k ′ d + 1 …(ii)
Putting the value of ( p − 1)! from Eq. (ii) in Eq. (i)
p (k ′ d + 1) = kd or pk ′ d + p = kd
or d (k − pk ′ ) = p
∴ d |p
∴ d = 1 or d = p [Qp is a prime number.]
If possible, let d = p
∴ d | ( p − 1)! − 1 ⇒ p | ( p − 1)! − 1 …(iii)
But by Wilson theorem (p − 1)! + 1 ≡ 0 mod p
i.e., p | ( p − 1)! + 1 …(iv)
From Eqs. (iv) and (iii)
p | ( p − 1)! + 1 − ( p − 1)! + 1
i.e., p |2 which is impossible. (Qp is an odd prime and hence ≥ 3)
∴ d =1
i.e., ( p !, ( p − 1)! − 1) = 1
i.e., p ! and ( p − 1)! − 1 are relatively prime.
Example 45. Find the number of positive integers n ≤ 1991 such that 6 is a factor of n 2 + 3n + 2.
Solution n 2 + 3n + 2 ≡ 0 (mod 6)
⇔ (n + 1)(n + 2) ≡ 0 mod 2
and (n + 1)(n + 2) ≡ 0 mod 3
⇔ (n + 1)(n + 2) ≡ 0 mod 3
Note : (n + 1)(n + 2) is even for all n ⇔ 3| n
∴ Required number = 1991 − 
1991 
= 1991 − 663 = 1328
 3 
Example 46. Let a, b be odd integers and n a natural number. Prove that a − b is divisible by 2n if and only
if a3 − b3 is divisible by 2n .
Solution a3 − b3 = (a − b )(a 2 + ab + b 2 ) = (a − b ){(a − b )2 + 3ab )}

Since a, b are odd integers therefore 3ab is an odd integer. Also a − b is an even integer and
consequently (a − b )2 + 3ab is an odd integer.
∴ a3 − b3 is divisible by 2n if and only if a − b is divisible by 2n .
www.pdfworld.in
www.pdfworld.in
Example 47. Let a , b be odd integers. If 4 does not divide a − b, then prove that 4 cannot divide a3 − b3.
Solution Suppose that 4 does not divide a − b.
Now, a3 − b3 = (a − b )(a 2 + ab + b 2 )
= (a − b )[(a − b )2 + 3ab ] …(i)
Since, a and b are both odd, 3ab is odd .
Also, a − b being even, (a − b )2 is a multiple of 4 (a − b )2 + 3ab is odd .
Since, (a − b ) is not a multiple of 4, it follows from Eq. (i) thata3 − b3 is not a multiple of 4.
Example 48. Prove that the sum of first n natural numbers (n ≥ 3) is never a prime.
Solution Sum of first n natural number is 1 n (n + 1), which is obviously a composite number.
2
 1  1
(If n is even, it can be factorized as  n  (n + 1); if n is odd, it can be factorized as n (n + 1), which
2  2
cannot be zero).
Example 49. Consider any 3 consecutive natural numbers the smallest of which is greater than 3. Then
prove that the square of the largest cannot be the sum of the squares of the other two.
Solution Let n , n + 1, n + 2 be three consecutive natural numbers with n > 3
(n + 2)2 − [n 2 + (n + 1)2 ] = 3 + 2n − n 2 = − (n + 1)(n − 3)
Which cannot be zero since n > 3.
Example 50. Prove that every natural numbern ≥ 12 is a sum of two composite numbers.
Solution If n is odd, write n = (n − 9) + 9
Since n is odd,∴n − 9 is even.
Also n ≥ 12 ⇒ n − 9 ≥ 3, so that n − 9 is an even number greater than 2 and consequently n − 9 is a
composite number. Also 9 is a composite number. Thus, n is sum of two composite numbers. If n is
even, n = (n − 4) + 4
Since n ≥ 12, n − 4 are both composite even numbers. Thus, n is the sum of two composite even
numbers.
Example 51. Prove that n 4 + 4 is a composite number for each natural number n is greater than 1.
Solution n 4 + 4 = (n 4 + 4n 2 + 4) − 4n 2
= (n 2 + 2)2 − (2n )2
= (n 2 + 2 + 2n )(n 2 + 2 − 2n )
Showing that n 4 + 4 is a composite number.
Example 52. Prove that n 4 + 4n is a composite number for all integer values of n > 1.
Solution If n is even, n 4 + 4n is divisible by 4
∴ It is composite number
If n is odd, suppose n = 2p + 1, where p is positive integer
Then, n 4 + 4n = n 4 + 4.42p = n 4 + 4(2p )4
which is of the form n 4 + 4b 4 , where b is a positive integer (= 2p )
www.pdfworld.in
www.pdfworld.in
n 4 + 4b 4 = (n 4 + 4b 2 + 4b 4 ) − 4b 2
= (n 2 − 2b 2 )2 − (2b )2
= (n 2 + 2b + 2b 2 )(n 2 − 2b + 2b 2 )
We find that n 4 + 4b 4 is a composite number consequently n 4 + 4n is composite when n is odd.

Hence, n 4 + 4n is composite for all integer values of n > 1.
Example 53. Prove that there is an infinitely of numbers m with the property that n 4 + m is composite for
every natural number n.
Solution If m = 4k 4 , where k is any natural number whatever, then n 4 + m = n 4 + 4k 4
= n 4 + 2. n 2 (2k 2 ) + (2k 2 )2 − (2nk )2
= (n 2 + 2k 2 )2 − (2nk )2
= (n 2 + 2k 2 + 2nk )(n 2 + 2k 2 − 2nk )
Showing that n 4 + m is a perfect square.
Example 54. Find the sum of all integers n that satisfy the properties
(a) 1 ≤ n ≤ 1994 (b) 30 divides n3 + 15n 2 + 50n
Solution Let P − n (n 2 + 15n + 590)

= n (n 2 + 3n + 2 + a multiple of 6)
= n (n + 1)(n + 2) + a multiple of 6
Since the product of any three consecutive natural numbers is divisible by 6, it follows that P is a
multiple of 6. Also P is a multiple of 5 iff n3 is a multiple of 5, which is true iff n is a multiple of 5 (as 5 is
prime). Thus, we find that P is a multiple of 30 iff P is a multiple of 5.
We have to find the sum of all multiple of 5 not exceeding 1995.
There are 398 such numbers, namely
5, 10, 16, … , 1990
398 5398399
. .
Sum of all these numbers is . + 3975
(25 . )= = 397005
2 2
Example 55. Show that, if the difference of two consecutive cubes is a square, then it is the square of the
sum of two successive squares.
Solution Suppose a ∈ Z such that
a 2 = (n + 1)3 − n3
Then, a 2 = 3n (n + 1) + 1 …(i)
Now, 4a − 1 = 12n (n + 1) + 3 = 3(4n + 4n + 1)
2 2
⇒ (2a − 1)(2a + 1) = 3(2n + 1)2

∴ 3 |2a − 1 or 2a + 1
If 3 |2a − 1, then 2a − 1 = 3m for some m ∈ Z , then
3m (2a + 1) = 3(2n + 1)2
⇒ n (2a + 1) = (2n + 1)2
Since, m and 2a + 1 are relatively prime.
www.pdfworld.in
www.pdfworld.in
[Q (2a + 1, 2a − 1) = 1], 2a + 1 is an odd square, say (2r + 1)2. Then a = 2r 2 + 2r is even, a contradiction
because by Eq. (i), a 2 is odd.
∴ 3 |2a + 1.
Now proceeding as above we see that 2a − 1 = (2r + 1)2 for some r ∈ Z
i.e., a = r 2 + (r + 1)2
Example 56. Show that the sum of any two consecutive odd primes is the product of at least three primes
which need not be distinct.
Solution Let p1 and p2 are two consecutive odd primes, since p1 + p2 is even, 1 (p1 + p2 ) ∈ Z . Further
2
1
(p1 + p 2 ) being in between the consecutive primes p1 and p 2 , is a composite number.
2
p + p2
p1 + p 2 = 2 1 has at least three prime factors.
2
Example 57. Prove that if p and 8p − 1 are prime numbers then 8p + 1 is a compsite number.
Solution When p = 3, 8p + 1 = 25, a composite number otherwise 3 | (8p − 1) 8p.
⇒ 3 | 8p + 1 [Q3 | (8p − 1)8p (8p + 1)]
Hence, 8p + 1 is not prime.
Example 58. Let p (x ) = x 2 + 40. Show that for any two integers a, b either p (a ) + p (b ) or p (a ) − p (b ) is
composite.
Solution Since, p (x ) = p (−x ) for any real x, we can assume that a and b are non negative and also that
a > b. Since, p (a ) − p (b ) = (a − b )(a + b ) if a − b > 1, then p (a ) − p (b ) is composite. So, assume a − b = 1.
Now, p (a ) − p (b ) = (b + 1) + b = 2b + 1
and p (a ) + p (b ) = (b + 1)2 + 40 + b 2 + 40
= 2b 2 + 2b + 81 = 2b (b + 1) + 81
If 3 | b (b + 1), then p (a ) + p (b ) is composite otherwise 3 | b − 1 and hence 3 |2(b − 1) + 3
i.e., 3| p (a ) − p (b )
∴ p (a ) − p (b ) is composite if b ≠ 1; when b = 1
p (a ) + p (b ) = 85 and so it is composite.
Example 59. Find the smallest integer > 1 which is simultaneously a square, a cube, a fourth power and a
fifth power (of certain integers).
Solution If a positive integer N > 1 is to be simultaneously a square a cube, a fourth power and a fifth
power (of certain integers) it must be a kth power of some integer where k = LCM of 2, 3, 4, 5 = 60. If N is to
be the smallest such integer, then N must be 260.
Example 60. Prove that ten's digit of any power of 3 is even.

Solution For any n ∈ N , let
3 if n ≡ 1 mod 4
9 if n ≡ 2 mod 4
υn =  

 7 if n ≡ 3 mod 4 
1 if n ≡ 4 mod 4 
www.pdfworld.in
www.pdfworld.in
Note that 3un = un +1 mod 20

Since, 20 | ( 3 − u1 ) and for any n ∈ N
+1
20 | (3n − un ) ⇒ 20 | (3n − 3un )
n +1
⇒ 20 | (3 − un + 1)
It follows by induction, 20 |3 − un for all n ∈ N . Hence, ten's digit of 3n is even.

n
(Note that unit's digit of 3n is un )
Example 61. The number of positive integral values of n of which n3 − 8n 2 + 20n − 13 represents a prime
number.
Solution If y = n3 − 8n 2 + 20n − 13
Then, y = (n 2 − 7n + 13)(n − 1)
for n = 1, y = 0 not admissible
for n = 2, y = 3, a prime number
for n = 4, y = 3, a prime number
for n > 4, n − 1 > 3 is a factor of y
Hence, y is not prime.
Number of integral values of n for which y is prime is 3.
Example 62. Find the sum of all the digits of the result of the subtraction1099 − 99.
Solution 1000000........0000
– 99
999999..........9901
(1 followed by 99 zeros)
The result is a 99 digit number having 97 nines followed by a zero and a 1.
∴ The sum of the digits = (97 × 9) + 0 + 1 = 874
Example 63. Find all integers n for which n + 13n + 2 also is an integer.
2
3n + 5
Solution On dividing n + 13n + 2 by 3n + 5, we get n | 3 + 34 | 9 as quotient and remainder −152 | 9. So if

2
(n 2 + 13n + 2)
=m
3n + 5
Then, 9m (3n + 5) = 9(n 2 + 13n + 2) = (3n + 34)(3n + 5) − 152
⇒ 152 = (3n + 5)(3n + 34 − 9m )
If m is an integer, then 3n + 5 is a divisor of 152 = 8 × 19.
The divisors of 152 are ± 1, ± 2 , ± 4, ± 8, ± 19, ± 38, ± 76 and ± 152 .
3n + 5 = ± 1, ± 2, ± 4, ± 8, ± 19, ± 38, ± 76, ± 152
3n + 34 − 9m = ± 152, ± 76, ± 38, ± 19, ± 8 ± 4, ± 2, ± 1
Considering the ‘+’ signs first.
9m − 29 = − 151, −74, − 34, − 11, 11, 34, 74, 151
9m = − 122, − 45, − 5, 18, 40, 63, 103, 180
So the possible values of m are –5, 2, 7 and 20; these correspond to 3n + 5 = 2, 8, 38 and 152.
www.pdfworld.in
www.pdfworld.in
Thus, n = − 1, 1, 11, 49.

Next consider ‘–’ signs.
9m − 29 = 151, 74, 34, 11, –11, –34, –74, –151
These give the same possible values of m, namely 20, 7, 2, and –5 these corresponds to 3n + 5 = − 1,
−4, − 19 and −76. Thus, n = − 2, − 3, − 8 and −27.
Thus, desired values of n are 1, 11, 49, –1, –2, – 3, – 8 and – 27. Possible values of m are –5, 2, 7 and 20.
Example 64. Each of the numbers x 1 , x 2 , ... , x n equal 1 or –1

and x 1x 2x 3x 4 + x 2x 3x 4x 5 + x 3x 4x 5x 6 + K +
xn − 1x n x 1x 2 + x n x 1x 2x 3 = 0.
Prove that n is divisible by 4.
Solution yi = x i + 1x i + 2x i + 3 for i = 1, 2, K , n. where x n +1 = x1 , xn + 2 = x 2 and x n +3 = x3
Then, y1 + y 2 + K + yn = 0
yi = ± 1 for each i.
Suppose that yi = 1 for n1 values of i and yi = − 1 for n2 values of i.
Then, n1 + n2 = n
0 = y1 + y 2 + K + yn = n1 (1) + n2 (−1) = n1 − n2
Hence, n1 = n2 , n = 2n2
n n2 n2
and y1y 2 K yn = (1) 1 (−1) = (−1)
But y1y 2 K yn = x 14x 24x 34 K x n4 = 1
This shows that n2 is even.
Hence, n = 2n2 is divisible by 4.
Example 65. Which of the numbers 101, 10101, 1010101, … with alternating 0's and 1's beginning and
ending with 1 can be primes ?
Solution Let N 2 = 101, N3 = 10101 and in general
100k − 1
N k = 1010 K 01 = 1 + 100 + K + 100k − 1 =
100 − 1
Here N k involves k 1's.
 10k − 1   10k + 1 
If k is odd, then Nk =   
 10 − 1   10 + 1 
This is composite since 10k + 1 is divisible by 10 + 1 if k is odd.
1002r − 1  100r − 1 
If k is even and k = 2r , then N k = =  (100 + 1)
r
99  99 
This is composite if r > 1. Thus,only one number in the list, namely N 2 = 101, is prime, the remaining are
composite.
Example 66. Find all integers n (positive, negative or zero) such that n 2 + 73 is divisible by n + 73.
Solution Let n + 73 = m.
Then, n 2 + 73 = (m − 73)2 + 73 = m 2 − 146m + 732 + 73
This is divisible by m if and only if 732 + 73 is divisible by m. Since 732 + 73 = 73 × 37 × 2.

We must have ± m = 1, 2, 37, 237
. , 73, 2.73, 37.73 or 2.37.73.
www.pdfworld.in
www.pdfworld.in
These give 16 values for m and hence 16 values for n given below.
–72, –71, –36, 1, 0, 73, 2628, 5329
–74, –75, –110, –147, –146, –219, –2774, –5475
Example 67. Find all positive integers which can be expressed as the sum of three distinct composite
numbers.
Solution If n is the sum of three distinct composite number, then n ≥ 4 + 6 + 8 = 18
Let n ≥ 18
If n is even then n − 10 is an even number > 6 and n = 4 + 6 + (n − 10). If n is odd then n − 13 is an even
number greater than 4 and n = 4 + 9 + (n − 13). So the desired positive integers are all n ≥ 18 .
Example 68. If p is a prime other than 2 and 5, prove that p divides an infinite number of the numbers 1,
11, 111, 1111, …
Solution Let N k the kth number in the sequence.
Then, N k = (10k − 1)|9 .
Given that 10 is not divisible by p.
Hence, by Fermat's theorem 10p − 1 ≡ 1 (mod p)
Thus, 10(p − 1)r ≡ 1 mod p for r = 1, 2, 3,...
Let 10(p − 1) r − 1 = mp. If p ≠ 3, then (9, p ) = 1
10k − 1  m 
Since 10k − 1 is divisible by 9 for all k, we see that 9|m. Therefore N k = = p
9  9
when k = ( p − 1) r , r = 1, 2, K
Since the sum of digit in N k is k we see that N k is divisible by 3 when k = 3r , r = 1, 2, …
2 10
Example 69. Find the remainder obtained, when the number 1010 + 10(10 ) + K + 10(10 )
is divided by 7.
Solution By Fermat's theorem 10 ≡ 1 mod 76
Hence, 106m ≡ 1 mod 7 for all m.

Now 10 ≡ 4 mod 6, 102 ≡ 40 ≡ 4 mod 6
By induction 10n ≡ 4 mod 6 for all n.
Thus, 10n = 6m + 4
and 10 (10n )
= 106m . 104 ≡ 104 mod 7 ≡ 4 mod (7)
2 10
consequently 1010 + 10(10 ) + ... + 10(10 )
≡ 4 × 10 mod 7 ≡ 5 mod 7
The remainder is 5.
Example 70. Find a positive integer n such that 7n 25 − 10 is divisible by 83.

Solution Since, 7 × 37 = 259 ≡ 10 mod 83
We have to find a value of n such that 7n 25 ≡ 7 × 37 mod 83
This is equivalent to n 25 ≡ 37 ≡ 220 mod 83
By Fermat's theorem 282k ≡ 1 mod 83 for all k. So it is enough, if we choose n such that n 25 ≡ 282k + 20
mod
83 If k = 15, this will be satisfied if n 25
≡2 1250
mod 83 and so if n = 2 . 50
This gives one value of n.
www.pdfworld.in
www.pdfworld.in
(2a )!(2b + 1)!

Example 71. If a and b are positive integers, then prove that is an integer.
(a !)(b !)(a + b + 1)!
Solution Let p be any prime, we will show that highest power of p that divides the numerator is ≥ the
highest power of p that divides the denominator thus showing that entire denominator divides numerator
exactly .
Let D = (a )!(b )! (a + b + 1)!
N = (2a )!(2b + 1)!
p a the highest power of prime p which divides D pb the highest power of p which divides N. Then, D
divides N if for every prime p , a ≤ b.
∞  k 
Now the greatest exponent α such that p α divides k′ is α = Σ  j  [x ] = integral part of x.
j =1 p
 
Suppose during factorization of D and N, the prime p occurs a and b times respectively.
∞  a   b  a + b + 1  
a ≡ Σ  j  +  j  +  
j =1  p
  p   pj 
∞   2a   2b + 1  
b = Σ  j  +  j 
j =1  p
   p 
In order to prove a ≤ b it is sufficient to prove
 2a   2b   a   b  a + b + 1 
 j +  j≥  j+  j+  
p  p  p  p   pj 
let p j = c , then
 2a  +  2b  ≥  a  +  b  +  b  +  a + b + 1 
 c   c   c   c   c   c 
(2a )!(2b + 1)!

we have proved earlier is an integer.
(a !)(b !)(a + b + 1)!
2000
Example 72. Determine the largest 3 digit prime factor of the integer C1000.
Solution If p is any 3 digit prime, then p > 2000. Thus, the highest power of p that divides 2000!, is
2
 2000   1000  2000 !

 p . Highest power of p that divides 1000! is  p . Since
2000
C1000 =
    (1000 !)2
 2000   1000 
 − 2  p . If p > 666, then
2000
The highest power of p that divides C1000 is 
 p   
 2000   1000 
 p  − 2 p  = 2 − 2(1) = 0
   
2000
i.e., p does not divide C1000.
The required prime is the largest one such that is it less than 666.
∴ p = 661, when p = 661
 2000   1000 
 p  −  p  = 3 − 2(1) = 1
   
∴ 661 is the largest 3 digit prime that divide 2000C1000.
www.pdfworld.in
www.pdfworld.in
Example 73. Prove that inradius of a right angled triangle with integer sides is an integer.
Solution Let ABC be right ∆ with ∠B = 90°, O be its incentre. L, M, N the points of contact of the in circle
with the sides a, b, c respectively.
Suppose that the inradius is r. Now as ∠ABC = 90°, the quadrilateral NBLO is a square. So NB = BL = r .
Also as the two tangents drawn from an external point to a circle are of equal length, we have
AM = AN = AB − NB = c − r
A
and CM = CL = BC − BL = a − r
so, b = AC = AM + CM = c − r + a − r
b − (c + a )
= c + a − 2r ⇒ r =
2
M
As ∠B = 90° ⇒ b 2 = c 2 + a 2 we have (i) if c and a are both odd or both even,
O
c 2 + a 2 is even ⇒b 2 is given ⇒ b is even ⇒b − (c + a ) is even (ii) if one of c and a N
is even and the other odd, c 2 + a 2 is odd ⇒b 2 is odd ⇒ b is odd ⇒b − (c + a ) = an

even number. B L C
b − (c + a )
So, in any case if a, b, c are integers, we have r = = an integer.
2
Example 74. Find the number of positive integers which divide 10999 but not 10998.
Solution All the positive divisors of pr (p is a positive prime, r ∈ N ) are 1, p , p 2 , p3 , K , pr . That is the
number of positive divisors of pr is r + 1. Similarly, the number of the positive divisors of q s (q is a positive
prime, s ∈ N ) is s + 1, so the number of positive divisors of pr q s is (r + 1)(s + 1).
Now, 10999 = 2999 × 5999 where 2 and 5 are positive primes. So the number of positive divisors of 10999 is
(999 + 1)(999 + 1) = 10002. For similar reasons, the number of positive divisors of 10998 is 9992. So the
number of positive numbers which divides 10999 but not 10998 is 10002 − 9992
= (1000 + 999)(1000 − 999)

= 1999 × 1 = 1999
Example 75. Let n = 1983, find the least positive integer k such that
k (n 2 )(n 2 − 12 )(n 2 − 22 ) (n 2 − 32 ) … (n 2 − (n − 1)2 ) = r !
for some positive integer r.

Solution Factorising and rearranging the resulting factors of the left side gives
kn (2n − 1)(2n − 2)…(n + 3) (n + 2)
(n + 1)(n )(n − 1) (n − 2)(n − 3) … (2)(1) ,
which equals kn (2n − 1)!. Thus, k = 2 is the smallest k that makes this a factorial [(2n )!].
Example 76. Find all positive integers n less than 17 for which n ! + (n + 1)! + (n + 2)! is an integral multiple
of 49.
Solution The given expression equals to
n ! {1 + (n + 1) + (n + 1)(n + 2)} = n !(n + 2)2.
Either 7 divides n + 2 or 49 divides n !

∴ n = 5, 12, 14, 15, and 16.
www.pdfworld.in
www.pdfworld.in
Example 77. Prove that n = 5 25 − 1 is a composite number.

125
5 −1
Solution Let x = 525, then
5125 − 1 = x 5 − 1 = (x − 1)(x 4 + x 3 + x 2 + x + 1)
= (x 4 + 9x 2 + 1 + 6x 3 + 6x + 2x 2 − 5x 3 − 10x 2 − 5x )(x − 1)
= ((x 2 + 3x + 1)2 − 5x (x + 1)2 )(x − 1)
= ((x 2 + 3x + 1)2 − (513 (x + 1))2 (x − 1)
= {x 2 + 3x + 1 + 513 (x + 1)}
{x 2 + 3x + 1 − 513 (x + 1)}(x − 1)
Thus, the given number is composite.
Example 78. Find all numbers p such that the numberp 2 + 11 has exactly 6 divisors.
Solution p = 2 does not work as 22 + 11 = 15 = 3 × 5 has 4 divisors but not 6 divisors. So p must be an odd
prime.
⇒ p 2 + 11 is even
∴ p 2 + 11 contains the prime 2 as factor. We now use the formula for d (n ), the divisor function. Since
6 = 3 × 2 there are precisely two categories of numbers with 6 divisors, those of the kind q5 and those of
the kind q 2r (with q , r unequal primes)
If p 2 + 11 has 6 divisors then p 2 + 11 = q5 or q 2r where q , r are primes, q ≠ r .
1st case is ruled out due to earlier observation so p 2 + 11 = q 2r . Here p = 3 works. Which has 6 divisor we
need to only consider the case when p > 3.
Since p is prime, it is indivisible by 3, so p ≡ ± 1 . which means p 2 ≡ 1 (mod 3)
∴ p 2 + 11 ≡ 0 (mod 3).
3 is a divisor of p 2 + 11.
2 is also divisor of p 2 + 11 .
It means that q1 , r are 2, 3 in some order.
For neither 223 = 12 nor 322 = 18 is of the form p 2 + 11 for any prime p.
Neither possibility works.
Example 79. Show that the quantity ( n + 1) ( n + 2)… ( 2n − 1)( 2n ) is divisible by 2n .

Solution We can write an = (n + 1)(n + 2)…(2n − 1)(2n ), then a1 = 2, we have 21|2.
a 2 = 3 × 4 = 12, we have 22 |12
So, assertion does hold good, when n = 1 and 2.
an (n + 1)(n + 2) K ( 2n − 1)( 2n )
Now = = 2(2n − 1)
an − 1 n (n + 1) K ( 2n − 3)( 2n − 2)
So an = 2( 2n − 1) an − 1 , an contains precisely one more 2 in its prime factorization thanan − 1.

−1
∴ if 2n | an −1 then, 2n | an
we have 2 | a1 which leads to 22 | a 2 which leads to 23| a3 which leads to 24 | a 4 and so on. All the way to
1
infinity.
∴ 2n | an for all integers n > 0.
www.pdfworld.in
www.pdfworld.in
Example 80. Find all ordered triples (x , y , z ) such that x , y and z are positive primes and x y + 1 = 2.
Solution If y is odd, then x odd + 1 would be divisible by x + 1 and would not be prime.
So y is even and must be 2.
Also since z must be odd, x must be even.
So, x = 2 which make z = 5
so only answer (2, 2, 5).
Example 81. The integers x , y and z each are perfect square and x > y > z > 0. If x , y and z form an AP
find smallest possible value of x.
Solution Let x = a2, y = b2, z = c 2
c 2 − b2 = b2 − a2
So 2b 2 = a 2 + c 2
Since b is at least 2, consider values of b from 2 onwards and find 2b 2.
The first such value i.e., the sum of two squares occurs when b = 5 (a = 7, c = 1), then x = 49
Example 82. Prove that the expression 21n + 4 is irreducible for every +ve integer n.
14n + 3
Solution If d divides two integers, then it divides their multiples, their sum and difference,
Q 3(14n + 3) − 2(21n + 4) = 1.
A common divisor of numerator and the denominator also divides 1; for this reason the fraction cannot
be multiplied.
a c
Alternative Solution If = e + (a , b , c , e integers) then c = a − be and a = c + be.
b b
Consequently the common divisor of a and b also divides c; common divisors of c and b divide a.
a c a b
∴ can be simplified if and only if can be simplified. Same holds for and .
b b b a
21n + 4 7n + 1 14n + 3 1
Q =1+ and =2+
14n + 3 14n + 3 7n + 1 7n + 1
and the last term cannot be simplified, the same is true for our original expressions.
Example 83. Determine all 3 digit numbers which are equal to 11 times the sum of the squares of their
digits.
Solution Denote the digits by a, b and c (a ≠ 0) According to the assumption
100a + 10b + c = 11(a 2 + b 2 + c 2 ) …(i)
⇒ (99a + 11b ) + (a − b + c ) = 11(a 2 + b 2 + c 2 )
RHS and the first term on LHS is divisible by 11.
Hence, so is a − b + c
Q −8 ≤ a − b + c ≤ 18, we conclude that a − b + c is equal either to 0 or 11.
Now b = a + c substituting this expression in Eq. (i), we get
100a + 10(a + c ) + c = 11(a 2 + (a + c )2 + c 2 )
After ordering the quadratic equation
2a 2 + (2c − 10) a + (2c 2 − c ) = 0 …(ii)
Q The first two terms of this expression are even third term should be even as well.
⇒ c is even.
www.pdfworld.in
www.pdfworld.in
Eq. (ii) admits integer solution if and only if its discriminant 4(−3c 2 − 8c + 25) is a square. This is
possible only in case c = 0, substitute c = 0 in Eq. (ii) we get 2a 2 − 10a = 0
Q a ≠ 0, we have a = 5 ⇒ b = a + c = 5. Hence we get 550 for the integer we are seeking for.
Now, when b = a + c − 11. After substituting and ordering Eq. (i) we get the expression
2a 2 + (2c − 32) a + (2c 2 − 23c + 131) = 0 …(iii)
Now c cannot be even. The discriminant admits the form 4(−3c + 14c − 16) ; for odd c it is a square only
2
in case c = 3. Substituting it in Eq. (iii), we get 2a 2 − 26a + 80 = 0, on solving this quadratic equation, we
get a = 5 or a = 8 ⇒ b = − 3 and b = 0. Only the latter figure might serve as a solution of the original
problem. This provides the second solution 803 and an easy check shows that this number satisfies the
assumption of the problem.
Two solutions are 550 and 803.
Example 84. Prove that there are infinitely many +ve integers a such that z = n 4 + a is not a prime for any
positive integer n.
Solution Let a = 4b 4 where b > 1 is an arbitrary integer, we show that z is not prime.
z = n 4 + 4b 4 + 4n 2b 2 − 4n 2b 2 = (n 2 + 2b 2 )2 − (2nb )2
= (n 2 + 2b 2 + 2nb )(n 2 + 2b 2 − 2nb ) = ((n + b )2 + b 2 ) ((n − b )2 + b 2 )

Q b > 1, both factors are > 1
So, z is not prime.
Example 85. Let p and q be +ve integers such that p = 1 − 1 + 1 − 1 + K − 1 +

1
. Prove that 1979
q 2 3 4 1318 1319
divides p .
Solution Applying the following transformation
p 1 1 1 1 1 1 1 1
=1− + − + K− + =1+ + + K+
q 2 3 4 1318 1319 2 3 1319
 1 + 1 + 1 + K + 1  = 1 + 1 + 1 + K + 1 − 1 + 1 + 1 + K + 1 
   
2 4 6 1318 2 3 1319  2 3 659 
1 1 1
= + + K+
660 661 1319
= 
1 1   1 1   1 + 1 
+  +  +  +K +  
 660 1319   661 1318  989 990 
= 1979  
1 1 1
+ + K+ 
 6601319
. 6611318
. 989.990 
a
The sum in parentheses is , where a is a positive integer and b = 660.661 … 1319
b
As 1979 is a prime and every factor of b is smaller than 1979, b and 1979 are coprime.
p 1979a
Thus, = implies that pb = 1979 aq
q b
So, 1979 divides p.
Remark : The problem can be generalised for primes of the form 3k + 2. If
p 1 1 1 1
=1− + −K− +
q 2 3 2k 2k + 1
then 3k + 2 divides p.
www.pdfworld.in
www.pdfworld.in
Example 86. Find one pair of positive integers a , b such that ab (a + b ) is not divisible by 7 and
(a + b )7 − a7 − b7 is divisible by 77 .
Solution Using the binomial theorem and observing that both (a + b )7 and a7 + b7 is divisible by (a + b ).
a (a + b ) …(i)
and (a + b ) − a − b
7 7 7
…(ii)
= 7ab (a + b )(a + ab + b )
2 2 2
…(iii)
Q ab (a + b ) is not divisible by 7 we must choose a and b such that 76 divides (a 2 + ab + b 2 )2 or

a + ab + b 2 is divisible by 73 = 343.
2
We try to choose a pair a, b such that b = 1 i.e., for some integer k

a 2 + a + 1 = 343k
i.e., a 2 + a + (1 − 343k ) = 0 …(iv)
This is possible, when discriminant of this equation of degree 2 in a is a perfect square. i.e.,
1 − 4(1 − 343k ) = 1372k − 3, is a perfect square.
It is true for k = 1 as 1369 = 372.
Hence Eq. (iv) becomes a 2 + a − 342 = 0 ; a = 18,b = 1 and the pair (a , b ) = (18, 1) satisfies the condition 7
does not divide 18.1.19.
Example 87. Let p be a prime number. Prove that there exist a prime number q such that for every integer
n the numbernp − p is not divisible by q.
pp − 1
Solution Since = 1 + p + p2 + K + p p − 1
p −1
≡ (p + 1) mod p 2 ,
(pp − 1)
We can get at least one prime divisor of which is not congruent to 1 modulo p 2. Denote such a
p −1
prime divisor by q. This q is what we wanted.
Assume that there exist an integer n such that np ≡ p mod (q ).
2
Then, we have np ≡ pp ≡ 1 mod q.
−1
On the other hand from Fermat's little theorem nq ≡ 1 mod q because q is prime. Since,( p 2, (q − 1)), we
have (p 2 , q − 1)| p which leads to np ≡ 1 mod q.
Hence, we have p ≡ 1 (mod q)
However this implies 1 + p + p 2 + ... + pp − 1 ≡ p mod q
From the definition of q, this leads to p ≡ 0 mod q which is a contradiction.
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1. Find GCD of 595 and 252 and express n in the 21. Find the highest power of 7, which is
form 252 m + 595n. contained in 50!.
2. Find the greatest common divisor d of the 22. Find the highest power of 3 which is contained
numbers 275 and 200 and then find integers in 500!.
m and n such that d = m 275 + n 200
23. Find the number of zeros at the end of (5n )! .
3. If a|c and b|c, then is it true that ab|c.
4. If a , m , n are non zero integers, then (a , mn ) = 1 24. Find the number of zeros at the end of (5n + 1)!
if and only if (a , m ) = 1 and (a , n ) = 1. 25. Find all n such that n! has 1998 zeros at the
5. If (a , b ) = 1 and c|a, then (c , b ) = 1 end of n!.
6. If (a , b ) = 1; then (ac , b ) = (c , b ). 26. Find exponent of 12 in 100!.
7. If a|b, c|d and (b , d ) = 1, then prove that 27. Show that 100! is divisible by 1520.
(a , c ) = 1. 28. Show that 33! is not divisible by 240.
8. Show that (a , b ) = (a + b , b ). 29. Find highest power of 3 in 200
C100.
9. If (a , m ) = 1, prove that (m − a , m ) = 1 30. What is the largest integer n such that 33! is
10. Prove that the product of two odd numbers is divisible by 2n .
always an odd number.
31. Find the number of zeros at the end of 200 C100.
11. Show that product of two numbers of the form
(6n + 1) is of the same form. 32. Find the number of zeros at the end of
12. If n is an integer, prove that n (n − 1)(2n − 1) is 2 ⋅ 10n C10n .
divisible by 6. n
33. Find the number of zeros at the end of 10 C5n .
13. If n is odd, show that n (n − 1) is divisible by 24.
2
34. If x + 2 = (18)(18)(18) K n digits, find the

14. If x and y are positive integers and if (x − y ) is number of zeros at the end of x c x/ 2.
even, show that x 2 − y 2 is divisible by 4.
35. Show that there is a positive integer n such
15. Prove that 24n − 1 is divisible by 15. that n! when written in decimal notation ends
with exactly 1993 zeros. (INMO 1993)
16. Prove that 32n + 7 is a multiple of 8.
36. Find the number of zeros at the end of
17. Show that 22n + 1 is divisible by 5 for a positive
 103 − 1  20
odd integer n. (i)  ! (ii) (520 − 1)! (iii) Σ (5n ) !
 9  n =1
18. Prove that an integer is divisible by 3 if and

(iv) Product of 1, 2, 3, ..., 1994.
only if the sum of its digits is divisible by 3.
(RMO 1994 Delhi)
19. Show that there are infinitely many primes of
the form 37. If x = 6 (111 K 1) − (22 K 2) + 2 , find number
2n n
(i) 6n + 5 (ii) 6n − 1 (iii) 4n − 1 of zeros at end of x Cx/ 2.
20. If each of the two primes p and q is a factor
38. Show that the highest power of 2 contained in
of ‘a’, show that the product pq is also a factor
of ‘a’. (2r − 1)! is 2r − r − 1 .
www.pdfworld.in
www.pdfworld.in
. 800 + 73
39. Show that 37 . 800 K 210 is divisible positive) such that both quadratic equations
by 10. x 2 + 2ax + b = 0 and x 2 + ax + b = 0 have
40. Show that integral roots. (INMO 1995)
3(7204 + 7202 + 7200 ) + 7(3204 + 3200 ) − (210) 60. (i) Consider two positive integers a and b
is divisible by 10. which are such that a abb is divisible by
41. Prove that A = (2903)n − (803)n − (464)n 2000. What is the least possible value of
the product ab?
+ (261) is divisible by 1897.
n
(ii) Consider two positive integers a and b
42. Show that 199 + 299 + 399 + 599 is divisible by 5. which are such that abb a is divisible by
2000. What is the least possible value of
43. Show that (6 ! + 1) is divisible by 7. the product ab? (RMO 2000)
44. Find the remainder when 61. Find all primes p and q such that p 2 + 7pq + q 2
(2222)5555 + (5555)2222 is divided by 7.
is the square of an integer. (RMO 2001)
4 44 444
45. Which is the largest 444 , 44 ,4 ? 62. Consider an n × n array of numbers
22  a11 a12 a13 K a1n 
46. Find number of digits in 22 .  
7
 a 21 a 22 a 23 K a 2n 
 100   M M 
47. Show  Σ n ! is of the form of 7k + 4.  
n = 1   an 1 an 2 an 3 K ann 
48. When 31994 + 2 is divided by 11. Find the Suppose each row consists of the n numbers
remainder. 1, 2, 3, . . . , n in some order and aij = a ji for
i = 1, 2, . . . , n and j = 1, 2, . . . , n. If n is odd,
49. Show that 255 + 1 is divisible by 11.
prove that the numbers a11 , a 22 , a33 , . . . , ann
50. Show that the difference of the squares of any are 1, 2, 3, . . . , n in some order. (RMO 2001)
two odd primes greater than 3 is divisible 63. Prove that the product of the first 200 positive
by 24. even integers differs from the product of the
51. Prove that 8th power of any number is of the first 200 positive odd integers by a multiple of
form 17n or 17n ± 1. 401. (RMO 2001)
52. Show that a18 − b18 is divisible by 133 if a and 64. Let a , b , c be positive integers such that a
b are coprime to 133. divides b 2 , b divides c 2 and c divides a 2. Prove
that abc divides (a + b + c )7 . (RMO 2002)
53. Show that n16 − a16 is divisible by 85 if n and a
are coprime to 85. 65. Consider the set X = {1, 2, 3, . . . , 9 , 10}. Find
two disjoint non-empty subsets A and B of X
54. Show that n − n is divisible by 30.
5
such that
55. Prove that a12 − b12 is divisible by 13 if a and b (a) A ∪ B = X ;
(b) prod(A ) is divisible by prod(B ), where for
are coprime to 13.
any finite set of numbers C, prod(C )
56. Prove that n11 − n is divisible by 11 for any denotes the product of all numbers in C;
integer n. (c) the quotient prod(A )/prod(B ) is as small as
possible. (RMO 2003)
57. If n > 1 is always odd, then n3 − 2n 2 + n is
divisible by 4. Prove. 66. Positive integers are written on all the faces of
a cube, one on each. At each corner (vertex) of
58. Let m and n be two integers such that the cube, the product of the numbers on the
m = n 2 − n. Show that m 2 − 2m is divisible faces that meet at the corner is written. The
by 24. sum of the numbers written at all the corners
is 2004. If T denotes the sum of the numbers
59. Show that there are infinitely many pairs (a, b)
on all the faces, find all the possible values
of relatively prime integers (not necessarily
of T . (RMO 2004)
www.pdfworld.in
www.pdfworld.in
67. Let 〈 p1 , p 2 , p3 , . . , pn , . . . 〉 be a sequence of rectangles have integer sides. Prove that there
primes defined by p1 = 2 and for n ≥ 1, pn + 1 is are always two congruent rectangles.
the largest prime factor of p1p 2 . . . pn + 1. (Thus (RMO 2006)
p 2 = 3, p3 = 7). Prove that pn ≠ 5 for any n. 72. Let a , b , c be three natural numbers such that
(RMO 2004)
a < b < c and GCD(c − a , c − b ) = 1. Suppose
68. If x , y are integers and 17 divides both the there exists an integer d such that
expressions x 2 − 2xy + y 2 − 5x + 7y and a + d , b + d , c + d form the sides of a
x 2 − 3xy + 2y 2 + x − y , then prove that 17 right-angled triangle. Prove that there exist
divides xy − 12x + 15y . (RMO 2005) integers l , m such that c + d = l 2 + m 2.
(RMO 2007)
69. Prove that there are infinitely many positive
integers n such that n (n + 1) can be expressed 73. In a book with page numbers from 1 to 100,
as a sum of two positive squares in at least two some pages are torn off. The sum of the
different ways. (Here a 2 + b 2 and b 2 + a 2 are numbers on the remaining pages is 4949. How
considered as the same representation.) many pages are torn off ? (RMO 2009)
(RMO 2006) 74. Show that there is no integer a such that
70. Find the least possible value of a + b , where a 2 − 3a − 19 is divisible by 289. (RMO 2009)
a , b are positive integers such that 11 divides
a + 13b and 13 divides a + 11b. (RMO 2006) 75. Show that 32008 + 42009 can be written as
71. A 6 × 6 square is dissected in to 9 rectangles product of two positive integers each of which
by lines parallel to its sides such that all these is larger than 2009182. (RMO 2009)
Level 2
1. Find all positive integers n for which 2n − 1 is 8. Show that positive integers that have an odd
divisible by 7. number of divisors are the squares.
2. Compute the smallest positive integer N such 9. Prove that there is no positive integer n for
N! which 2n + 1 is divisible by 7.
that 12 is an integer.
12
10. Let a, b, c, d be integers with a > b > c > d > 0.
3. Find one pair of positive integers a and b such Suppose that ac + bd = (b + d + a − c )
that (b + d − a + c ). Prove that ab + cd is not prime.
(i) ab (a + b ) is not divisible by 7.
11. For any positive integer n, let d (n ) denote the
(ii) (a + b )7 − a7 − b7 is divisible by 77 . number of positive divisors of n (including 1
Justify your answer and n itself). Determine all positive integers k
d (n 2 )
4. Factor the number 51985 − 1 into a product of such that = k for some n.
d (n )
three integers each of which is larger than 5100.
12. Determine all pairs (a, b) of positive integers
5. If (1 + x + x 2 + x 3 + x 4 )496 = a 0 + a1x + K
such that ab 2 + b + 7 divides, a 2b + a + b.
+ a1984x 1984 .
13. Find all pairs (n , p ) of positive integers such
(i) Determine GCD of the coefficients that p is prime, n ≤ 2p and ( p − 1)n + 1 is
a3 , a 8 , a13 , .. , a1983. divisible by np −1.
(ii) Show that 10347 > a 992 > 10340. 14. Compute the unique positive integer n such that
6. Find the largest positive integer n with the 2 ⋅ 22 + 3 ⋅ 23 + 4 ⋅ 24 + 5 ⋅ 25 + . . . + n ⋅ 2n
property that n + 10 | n3 + 100. = 2(n + 10)
.
7. Find the number of positive integers n less 15. Given two odd integers a and b. Prove that
than 1991 for which 6/n 2 + 3n + 2 . a 2 − b 2 is divisible by 2n , if and only if a − b is
(INMO 1991) divisible by 2n .
www.pdfworld.in
www.pdfworld.in
16. Prove that [5x ] + [5y ] ≥ [3x + y ] + [3y + x ] where (b) the sequence has infinitely many even
x , y ≥ 0 and [u ] denotes the greatest integer ≤ u. numbers. (INMO 2005)
17. Find all 3 digit numbers equal to the sum of 28. Let n be a natural number such that
the factorials of their digits. n = a 2 + b 2 + c 2 , for some natural numbers
18. Find all whole numbers equal to the sum of the a, b, c .
squares of their digits. Prove that
9n = (p1a + q1b + r1c )2 + (p 2a + q 2b + r2c )2
19. Prove that the expressions 2x + 3y and 9x + 5y
are divisible by 17 for the same set of integral + (p3a + q3b + r3c )2,
values of x and y. where p j ' s , q j ' s , r j ' s are all non-zero integers.
Further, if 3 does not divide at least one of
20. Prove that there are no positive integral
a , b , c , prove that 9n can be expressed in the
numbers which increase twice when their
form x 2 + y 2 + z 2 , where x , y , z are natural
initial digits are carried to the end of the
numbers. numbers none of which is divisible by 3.
(INMO 2007)
21. Find all integers n (positive, negative or zero)
29. Let A be a set of real numbers such that A has
for which n 2 + n + 41 is a perfect square.
at least four elements. Suppose A has the
22. Given any nine integers show that it is possible property that a 2 + bc is a rational number for
to choose, from among them, four integers all distinct numbers a , b , c in A. Prove that
a , b , c , d such that a + b − c − d is divisible by there exists a positive integer M such that
20. Further show that such a selection is not a M is a rational number for every a in A.
possible, if we start with eight integers instead (INMO 2008)
of nine. (INMO 2001) 30. Define a sequence 〈an 〉n∞ = 1 as follows :
23. Do there exist three distinct positive real 0, if the number of positive
numbers a , b , c such that the numbers 
an = 
divisors of n is odd .
a , b , c , b + c − a , c + a − b , a + b − c and
1, if the number of positive
a + b + c form a 7-term arithmetic progression 
in some order ? (INMO 2002)  divisors of n is even.
24. Determine the least positive value taken by (The positive divisors of n include 1 as well as
the expression a3 + b3 + c 3 − 3abc as a , b , c n). Let x = 0. a1a 2a3 . . . be the real number
whose decimal expansion contains an in the
vary over all positive integers. Find also all
nth place, n ≥ 1. Determine, with proof,
triples (a , b , c ) for which this least value is
whether x is rational or irrational. (INMO 2009)
attained. (INMO 2002)
31. Define a sequence 〈an 〉n ≥ 0 by a 0 = 0, a1 = 1 and
25. Find all primes p and q , and even numbers
an = 2an − 1 + an − 2, for n ≥ 2.
n > 2, satisfying the equation
−1
(a) For every m > 0 and 0 ≤ j ≤ m , prove that
pn + pn + . . . + p + 1 = q 2 + q + 1.
2am divides am + j + (−1) j am − j .
(INMO 2003)
(b) Suppose 2k divides n for some natural
26. Let S denote the set of all 6-tuples
numbers n and k. Prove that 2k divides an .
(a , b , c , d , e , f ) of positive integers such that (INMO 2010)
a 2 + b 2 + c 2 + d 2 + e 2 = f 2. Consider the set
32. Find all natural numbers n > 1 such that n 2
T = {abcdef : (a , b , c , d , e , f ) ∈ S }.
does not divide (n − 2)!. (INMO 2010)
Find the greatest common divisor of all the
33. Call a natural number n faithful, if there exist
members of T . (INMO 2004)
natural numbers a < b < c such that a divides
27. Let x 1 be a given positive integer. A sequence b , b divides c and n = a + b + c .
〈x n 〉n∞ = 1 = 〈x 1 , x 2 , x 3 , . . . 〉 of positive integers is (i) Show that all but a finite number of
such that x n , for n ≥ 2, is obtained from x n − 1 by natural numbers are faithful.
adding some non-zero digit of x n − 1. Prove that (ii) Find the sum of all natural numbers which
(a) the sequence has an even number; are not faithful. (INMO 2011)
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1. 595 = 2252
. + 91 …(i) ar + nt (ah + mk ) = 1
[Dividing 595 by 252] ⇒ ar + ah nt + mn tk = 1
252 = 2.91 + 70 …(ii) ⇒ a (r + hnt ) + mn (tk ) = 1
[Dividing 252 by 91] ⇒ (a , mn ) = 1
91 = 170
. + 21 …(iii) Now, let (a , mn ) = 1
[Dividing 91 by 70] ∴ ∃ integers u and v such that
70 = 321
. +7 …(iv) au + mnv = 1
[Dividing 70 by 21] a (u ) + m (nv ) = 1
∴ GCD of 595 and 252 is d = 7 (a , m ) = 1
From Eq. (iv),d = 7 = 70 − 321
. Similarly, (a , n ) = 1
= 70 − 3(91 − 170
. ) 5. Q (a , b ) = 1
[Putting value of 21 from Eq. (iii)] Q ∃ integers x and y such that
= 70 − 3.91 + 370
. = 470
. − 3.91 ax + by = 1 …(i)
= 4(252 − 2.91) − 3.91 Q c|a
[Putting value of 70 from Eq. (ii)] ∴ ∃ an integer m such that a = cm
= 4252
. − 11.91 Putting a = cm in Eq. (i), we get
= 4252
. − 11(595 − 2252
. ) cmx + by = 1
[Putting value of 91 from Eq. (i)] or c (mx ) + b (y ) = 1
= 4252
. − 11595
. + 22252
. ∴ (c , b ) = 1
= 26252
. − 11595
.
6. Let (ac , b ) = d and (c , b ) = e
⇒ d = 252m + 595n
(ac , b ) = d
Here, m = 26, n = − 11
d is a common divisior of ac and b .
2. d = 25 = 3(275) + (−4)(200) ∴ d |c
3. It is not true. i.e., d is a common divisor of b and c.
For example, a = 3 | c = 12 ∴ d |c …(i)
b = 6 | c = 12 Q (c, b) = e
But a . b = 3.6 = 18 × c = 12 Again (c , b ) = e
4. Let (a , m ) = 1 and (a , n ) = 1 e|c and e|b
Q (a , m ) = 1 e|ac and e|b
∴ ∃ integers h and k such that i.e., e is a common divisor of ac and b.
ah + mk = 1 …(i) ∴ e|d [Q(ac , b ) = d ] …(ii)
Q (a , n ) = 1 From Eqs. (i) and (ii) d = e i.e., (ac , b ) = (c , b )
∴ ∃ integers r and t such that 7. Q a|b, so there exists an integer m, such that
ar + nt = 1 ⇒ ar + nt .1 = 1 …(ii) b = am …(i)
Putting the value of 1 from Eq. (i) in LHS of Q c | d , so there exits an integer n, such that
Eq. (ii), we have d = cn …(ii)
www.pdfworld.in
www.pdfworld.in
Q(b , d ) = 1, so there exist integers x and y such ∴ ab = (6k + 1)(6k ′ + 1)

that = 36kk ′ + 6k + 6k ′ + 1
bx + dy = 1 …(iii) = 6(6kk ′ + k + k ′ ) + 1 = 6l + 1
Putting the values of b and d in Eq. (iii)
where l = 6kk ′ + k + k ′, which is of the form.
amx + cny = 1
12. n (n − 1)(2n − 1) = n (n − 1){(n + 1) + (n − 2)}
a (mx ) + c (ny ) = 1
= n (n − 1)(n + 1) + n (n − 1)(n − 2)
∴ (a , c ) = 1.
= (n − 1) n (n + 1) + (n − 2)(n − 1) n
8. Let (a , b ) = d and (a + b , b ) = e
Each of the two (n − 1) n (n + 1) and
Q (a , b ) = d
(n − 2)(n − 1) n being the product of three
∴ d | a and d | b consecutive integers is divisible by 3 ! = 6
∴ d |a + b ∴ Their sum = n (n − 1)(2n − 1) is also divisible
Now, d | (a + b ) and d | b by 6.
∴ d | (a + b ) , b 13. Qn is odd, so let n = 2m + 1
[Since d is a common divisor of a + b and b Putting n = 2m + 1 in n (n 2 − 1), we have
so d also divides gcd of a + b , b].
n (n 2 − 1) = n (n − 1)(n + 1)
or d |e …(i)
= (2m + 1)(2m + 1 − 1)(2m + 1 + 1)
Now Q (a + b , b ) = e
= (2m + 1) 2m (2m + 1)
∴ e | (a + b ) and e | b.
= 4m (m + 1)(2m + 1)
∴ e | (a + b ) − b or e | a.
Now proceed yourself.
Now e | a and e | b
∴ e (a , b ) 14. As (x − y ) is even.
∴ x − y = 2m or x = 2m + y
i.e., e |d …(ii)
From Eqs. (i) and (ii), d = e Putting x = 2m + y in x 2 − y 2 , we have
i.e., (a , b ) = (a + b , b ). x 2 − y 2 = (2m + y )2 − y 2
9. Let (m − a , m ) = d = 4m 2 + 4my + y 2 − y 2
∴ d | m − a and d | m = 4m 2 + 4my
∴ d | [m − (m − a )] or d | a = 4(m 2 + my )
Now, d | m and d | a
so (x 2 − y 2 ) is divisible by 4.
∴ d | (a , m ) [As d is a common divisor of a and
m and d | GCD of a, m] 15. 24n − 1 = (24 )n − 1 = 16n − 1n
−1 −2 −1
But (a , m ) = 1 = (16 − 1)[(16)n + (16)n .1 + K + 1n ]
n −1 n −2
∴ d |1 (Qx n
−y n
= (x − y )(x +x y +K
∴ d = 1 [only divisor of 1 is 1] + yn −1
) ∀ all n
so (m − a , m ) = 1 or 2 4n
− 1 = 15 [(16) n −1
+ (16) n −2
+ K + 1]
10. Let a = 2k + 1 and b = 2k ′ + 1 be two odd ∴ 24n − 1 is divisible by 15.
numbers.
[by definition of divisibility]
∴ ab = (2k + 1)(2k ′ + 1)
= 4kk ′ + 2k + 2k ′ + 1 16. 3 2n
+ 7 = (3 ) 2 n
+ 7 = 9n + 7 = (9n − 1) + 8
= 2(2kk ′ + k + k ′ ) + 1 = 2q + 1 = (9n − 1n ) + 8
−1 −2
where q = 2kk ′ + k + k ′ = (9 − 1)(9n + 9n + K + 1) + 8
∴ ab is an odd number = 8 (9 n −1
+ 9 n −2
+ K + 1) + 8
−1 −2
11. Let a = 6k + 1 and b = 6k ′ + 1 be two numbers = 8 (9n + 9n + K + 1 + 1)
of the form (6n + 1)
∴ 3 2n
+ 7 is a multiple of 8.
www.pdfworld.in
www.pdfworld.in
17. 22n + 1 = (22 ) n + 1 = 4n + 1 = 4n + 1n (ii) Try yourself with the help of part (i)
= (4 + 1)(4n −1
−4 n −2
+ 4 n −3
− K + 1) (iii) Try yourself
[Qfor odd] 20. Qp/a ∴ ∃ an integer m such that

−1 −2 −3 2 a = pm …(i)
x n + y n = (x + y )(x n − xn y + xn y
−1
Q q/a ∴∃ an integer n such that
− K + yn )
a = qn …(ii)
−1 −2 −3
= 5(4n − 4n + 4n K + 1) As p and q are both primes
∴ 22n + 1 is divisible by 5. ∴ They are coprime i.e., (p , q ) = 1
19. (i) If possible let the number of primes of the ∴ ∃ integers x and y such that
form (6n + 5) be finite. px + qy = 1 …(iii)
These primes are 5, 11, 17, ... (putting n = 0, Multiplying both sides of Eq. (iii) by a
1, 2, 3…) let q be the greatest prime of the ∴ apx + aqy = a
form (6n + 5). Let a = 51117
. . …. q be the Putting a = qn from Eq. (ii) in apx and a = pm
product of all primes of the form (6n + 5) .
from Eq. (i) in aqy , we have
Let b = 6a − 1 …(i)
qnpx + pmqy = a
∴ b>1 [Qa ≥ 5 ⇒ b = 6a − 1 ≥ 29]
pq (nx + my ) = a
∴ By Fundamental Theorem, b can be
∴ pq|a.
expressed as a product of primes say
p1 , p 2 , p3 K pr . 21. 8
i.e., b = p1. p 2. p3 K pr …(ii) 22. 247
Now, b = 6a − 1 is odd. 5 n −1
Hence 2 can't be a factor of b.
23. (5 − 1) zeros
4
Again 3 is also not a factor of b. 5 n −1
24. (5 − 1) zeros
∴None of the prime factors in RHS of Eq. (ii) 4
is 2 or 3.
25. The greatest power of a > 1, a ∈ N
i.e., Every prime factor in RHS of (2) is an ∞
dividing n is given by Σ  i 
odd number > 3. n
…(i)
 a 
i = 1
∴ Each of p1, p 2, K pr is of the form (6n + 1)
or (6n + 5). ∞ ∞ n  1 
Σ  i  < Σ i = n
n
but  …(ii)
 a  i = 1 a
i =1   a − 1
∴ Again all of f p1 , p 2 , K , pr can't be primes
of the form (6n + 1).
We want to find n such that
∴ At least one of p1 , p 2 , p3 K pr say p is (a ∞
Σ  i  = 1998
n
prime factor of b) of the form (6n + 5) i.e., i =1 
 5 
p/b
∞  1  n
from Eq. (ii) Σ  i  < n 
n
Also p|a [Q p is one prime of the form  =
i =1 
 5   5 − 1 4
(6n + 5) and a is the product of all such
primes.] n
So, > 1998 so n > 7992
∴ p|6a 4
∴ p|6a − b i.e., p|1 By trial and error, we take n = 7995 and then
[Q from Eq. (i), 6a − b = 1] find the correct value.
Which is impossible. If n = 7995, then number of zeroes at the end
[Qp being a prime of the form 6n + 5 is ≥ 5] of 7995 is from Eq. (i)
∴ Our supposition is wrong. 7995 7995
+ + ...
5 52
∴ Number of primes of the form (6n + 5) is
infinite. = 1599 + 319 + 63 + 12 + 2 = 1995
www.pdfworld.in
www.pdfworld.in
so true for n = 8000, we get the number of Also 3 is prime

zeroes at the end of 8000! ∴ By Fermat theorem, x 2 ≡ 1 mod 3 …(i)
= 1600 + 320 + 64 + 12 + 2 = 1998.
similarly y 2 ≡ 1 mod 3 …(ii)
26. 48
Subtracting congruences (i) and (ii), we have
29. 3 x 2 − y 2 ≡ 0 mod 3
30. 31 ∴ x 2 − y 2 is divisible by 3 …(iii)
31. 1 zero Again, as x and y are odd primes > 3
32. 1 zero ∴ x and y are odd numbers > 3
33. No zeros Let x = 2k + 1; y = 2k ′ + 1, where k > 1, k ′ > 1
34. 1 zero ∴ x 2 − y 2 = (x − y )(x + y )
35. For any k , n ∈ N the highest power of k which = (2k + 1 − 2k ′ − 1)(2k + 2k ′ + 2)

divides n is given by = (2k − 2k ′ ) (2k + 2k ′ + 2)
ψk (n ) =   +  2  +  3  + …
n n n = 4(k − k ′ )(k + k ′ )
 k   k   k 
= 4[(k + k ′ ) − 2k ′ ](k + k ′ + 1)
The number of zeros with which n ends when = 4(k + k ′ )(k + k ′ + 1) − 8k ′ (k + k ′ + 1)
written in decimal system is
= 8l − 8k ′ (k + k ′ + 1)
min {ψ2 (n ), ψ5 (n )} = ψ5 (n )
Hence we have to find n such that [where = (k + k ′ )(k + k ′ + 1) being the product
of two consecutive integers is divisible by
ψ5 (n ) = 1993 2 ! = 2]
Since, ψ5 (n ) ≤ + 2 + K =  1 + + K
n n n 1
= 8[l − k ′ (k + k ′ + 1)]
5 5 5 5 
∴ x 2 − y 2 is divisible by 8. …(iv)
n 5 n
= . = .
5 4 4 From Eqs. (iii) and (iv)
n
We have ≥ 1993 ⇒ n > 7970 x 2 − y 2 is divisible by 24 [Q(3, 8) = 1]
4
51. Let a be any integer
We find that ψ5 (7975) = 1991
Q 17 is prime and a is any number.
∴ ψ(7985) = ψ5 (1975) + 2 = 1993
∴ Either (a , 17) = 1 or 17|a
Hence, 7985 is one solution.
Case I (a, 17) = 1
Other numbers are 7986, 7987, 7988, 7989.
Also 17 is prime.
36. (i) 26 zeros.
∴ By Fermat theorem
520 − 420 − 1 a16 ≡ 1 mod 17
(ii) zeros.
4
∴ 17 | (a16 − 1) or 17 | (a 8 − 1)(a 8 + 1)
(iii) 300 zeros.
∴ Either 17 | (a 8 − 1) or 17 | (a 8 + 1)
(iv) 495 zeros.
[Q17 is prime]
37. 1
∴ a 8 − 1 = 17n or a 8 + 1 = 17n
44. 0
i.e., a 8 = 17n + 1 or a 8 − 17n − 1 …(i)
45. 4444
Case II 17| a
46. 14
∴ 17 | a 8
48. 6
∴ a 8 = 17n …(ii)
50. Let x and y be two odd prime > 3
From Eqs. (i) and (ii) we can say that 8th power
Q x is a prime greater than 3,
of any number is of the form 17n or 17n ± 1
∴ (x , 3) = 1
www.pdfworld.in
www.pdfworld.in
52. Qa and b are both coprime to 133 Subtracting these congruences

∴a and b are coprime to 19. n16 − a16 ≡ 0 mod 5
(Q19 is a factor of 133)
∴ (n16 − a16 ) is divisible by 5 …(ii)
Also 19 is a prime.
∴ By Fermat theorem
a18 ≡ 1 mod 19 n16 − a16 is divisible by 85(= 17 × 5)
b18 ≡ 1 mod 19 [Q(17, 5) = 1]
Subtracting these congruences 54. Q (a ≡ a mod p)

p
a 18
−b 18
≡ 0 mod 19 ∴ n5 ≡ n mod 5
∴ a18 − b18 is divisible by 19 …(i) n5 − n is divisible by 5 …(i)
Again if a and b are both coprime to 133. Q 5 is a prime number)

∴ a and b are coprime to 7 Again, n5 − n = n (n 4 − 1)
(Q7 is a factor of 133) = n (n 2 − 1)(n 2 + 1)
Also 7 is prime = n (n − 1)(n + 1)(n 2 + 1)
∴ a 6 ≡ 1 mod 7
Now (n − 1)n (n + 1) being the product of three
b 6 ≡ 1 mod 7. consecutive integers is divisible by 3 ! = 6 and
Raising both congruences to power 3. hence
a18 ≡ 1 mod 7 n5 − n = n (n − 1)(n + 1)(n 2 + 1) is divisible by 6.
b18 ≡ 1 mod 7 …(ii)
Subtracting these congruences ∴ From Eqs. (i) and (ii)
a18 − b18 ≡ 0 mod 7 n5 − n is divisible by 30 (= 6 × 5)
∴ (a 18
− b ) is divisible by 7.
18
…(ii) [Q(6, 5) = 1]
From Eqs. (i) and (ii) 55. Q13 is prime and (a , 13) = 1
a18 − b18 is divisible by 133(= 19 × 7) By Fermat theorem, a13 − 1 ≡ 1 mod 13
[Q(7, 19) = 1]
i.e., a12 ≡ 1 mod 13 …(i)
53. Q (n , 85) = 1 and (a , 85) = 1
b 12
≡ 1 mod 13 …(ii)
∴ (n , 17) = 1 and (a , 17) = 1
Subtracting congruences Eqs. (i) and (ii)
(Q17 is a divisor of 85)
a12 − b12 ≡ 0 mod 13
But 17 is a prime number.
∴ By Fermat theorem, n16 ≡ 1 mod 17 ∴ (a12 − b12 ) is divisible by 13.
a16 ≡ 1mod 17 56. Q11 is a prime number.
Subtracting these congruences n16 − a16is n11 ≡ n mod 11 (ap ≡ a mod p)
divisible by 17 …(i) ∴ n11 − n is divisible by 11.
Again, as n and a are both coprime to 85
∴ n and a are both coprime to 5. 57. Qn (n 2 − 2n + 1) ⇒ n (n − 1)2
(Q5 is a factor of 85) ⇒ n (n − 1)(n − 1)
∴ By Fermat theorem n ≡ 1 mod 5 4 Since, n is odd, (n − 1) is even.
and a 4 ≡ 1 mod 5. We know (n − 1) n = 2k
Raising both congruences to power 4. (n − 1) is even = 2q
(QIf a ≡ b mod m, then an ≡ bn mod m) ∴ (n − 1) n (n − 1) = 4qk = 4m
n 16
≡ 1 mod 5 ∴ It is of the form 4m
a 16
≡ 1 mod 5 ∴ n3 − 2n 2 + n is divisible by 4.
www.pdfworld.in
www.pdfworld.in
58. m 2 − 2m = m (m − 2) diagonal, there is a corresponding occurrence

= (n − n )(n − n − 2)
2 2 above it, by the symmetry of the array. This
accounts for an even number of occurrences
= n (n − 1)(n − 2)(n + 1) of 1 off the diagonal. But 1 occurs exactly n
= (n − 2)(n − 1)(n )(n + 1) times and n is odd.
which is the product of four consecutive Thus 1 must occur at least once on the main
integers since the product of four consecutive diagonal. This is true of each of the numbers
integers is divisible by 4 ! (= 24). 1, 2, 3, . . . , n. But there are only n numbers on
the diagonal. Thus each of 1, 2, 3, . . . , n occurs
∴ m 2 − 2m is divisible by 24. exactly once on the main diagonal. This
60. We have 2000 = 2453. implies that a11 , a 22 , a33 , . . . , ann is a
permutation of 1, 2, 3, . . . , n.
(i) Since 2000 divides a abb , it follows that 2
63. We have to prove that
divides a or b and similarly 5 divides a or b.
In any case 10 divides ab. Thus the least 401 divides 2 ⋅ 4 ⋅ 6 ⋅ . . . ⋅ 400 − 1 ⋅ 3 ⋅ 5 ⋅ . . . ⋅ 399.
possible value of ab for which 2000 | a abb Write x = 401. Then, this difference is equal to
must be a multiple of 10. Since 2000 (x − 1)(x − 3) . . . (x − 399) − 1 ⋅ 3 ⋅ 5 ⋅ . . . ⋅ 399.
divides 101011 , we can take a = 10, b = 1 to If we expand this as a polynomial in x , the
get the least value of ab equal to 10. constant terms get canceled as there are even
number of odd factors ((−1)200 = 1). The
(ii) As in (i) we conclude that 10 divides ab.
Thus the least value of ab for which remaining terms are integral multiples of x
2000 | abb a is again a multiple of 10. If and hence the difference is a multiple of x .
Thus 401 divides the above difference.
ab = 10, then the possibilities are
(a , b ) = (1, 10), (2, 5), (5, 2), (10, 1). But in all 64. Consider the expansion of (a + b + c )7 . We
these cases it is easy to verify that 2000 show that each term here is divisible by abc. It
does not divide abb a . The next multiple of contains terms of the form rklm a kbl c m , where
10 is 20. In this case we can take rklm is a constant (some binomial coefficient)
(a , b ) = (4, 5) and verify that 2000 divides and k , l , m are non-negative integers such that
4554 . Thus the least value here is 20. k + l + m = 7. If k ≥ 1, l ≥ 1, m ≥ 1, then abc
61. Let p , q be primes such that p 2 + 7pq + q 2 = m 2 divides a kbl c m .
for some positive integer m. We write Hence we have to consider terms in which
5pq = m 2 − (p + q )2 = (m + p + q )(m − p − q ). one or two of k , l , m are zero. Suppose for
example k = l = 0 and consider c 7 . Since b
We can immediately rule out the possibilities
divides c 2 and a divides c 4 , it follows that abc
m + p + q = p , m + p + q = q and m + p + q = 5
(In the last case m > p , m > q and p , q are at divides c 7 .
least 2). A similar argument gives the result for a7 or
Consider the case m + p + q = 5p and b7 . Consider the case in which two indices are
m − p − q = q. Eliminating m , we obtain non-zero, say for example, bc 6. Since a divides
2(p + q ) = 5p − q. It follows that p = q. Similarly, c 4 , here again abc divides bc 6. If we take b 2c 5,
m + p + q = 5q and m − p − q = p leads to p = q. then also using a divides c 4 we obtain the
Finally taking m + p + q = pq , m − p − q = 5 result. For b3c 4 , we use the fact that a divides
and eliminating m , we obtain 2(p + q ) = pq − 5. b 2. Similar argument works for b 4c 3 , b5c 2 and
This can be reduced to (p − 2)(q − 2) = 9. Thus
b 6c . Thus each of the terms in the expansion of
p = q = 5 or (p , q ) = (3, 11), (11, 3). Thus the set
(a + b + c )7 is divisible by abc.
of solutions is
{( p , p ) : p is a prime} ∪ {(3, 11), (11, 3)}. 65. The prime factors of the numbers in set
{1, 2, 3, . . . , 9, 10} are 2, 3, 5, 7. Also only 7 ∈ X
62. Let us see how many times a specific term, say
has the prime factor 7. Hence, it cannot appear
1, occurs in the matrix. Since 1 occurs once in
in B. For otherwise, 7 in the denominator
each row, it occurs n times in the matrix.
would not get canceled. Thus 7 ∈ A.
Now consider its occurrence off the main
diagonal. For each occurrence of 1 below the Hence, prod(A )/prod(B ) ≥ 7.
www.pdfworld.in
www.pdfworld.in
The numbers having prime factor 3 are 3, 6, 9. 68. Observe that

So, 3 and 6 should belong to one of A and B, x 2 − 3xy + 2y 2 + x − y = (x − y )(x − 2y + 1)
and 9 belongs to the other. We may take
Thus, 17 divides either x − y or x − 2y + 1.
3, 6 ∈ A , 9 ∈ B.
Suppose that 17 dividesx − y . In this case x ≡ y
Also 5 divides 5 and 10. We take 5 ∈ A, 10 ∈ B. (mod 17) and hence
Finally we take 1, 2, 4 ∈ A , 8∈ B. Thus,
x 2 − 2xy + y 2 − 5x + 7y ≡ y 2 − 2y 2 + y 2
A = {1, 2, 3, 4, 5, 6, 7}, B = {8, 9, 10}, − 5y + 7y ≡ 2y (mod 17).
prod (A ) 1 ⋅ 2 ⋅ 3 ⋅ 4 ⋅ 5 ⋅ 6 ⋅ 7
so that = =7 Thus, the given condition that 17 divides
prod (B ) 8 ⋅ 9 ⋅ 10 x 2 − 2xy + y 2 − 5x + 7y implies that 17 also
prod (A ) divides 2y and hence y itself. But then x ≡ y
Thus, 7 is the minimum value of .
prod (B ) (mod 17) implies that 17 divides x also. Hence,
There are other possibilities for A and B : e.g., in this case 17 divides xy − 12x + 15y .
1 may belong to either A or B. We may take Suppose on the other hand that 17 divides
A = {3, 5, 6, 7, 8}, B = {1, 2, 4, 9, 10}. x − 2y + 1. Thus, x ≡ 2y − 1 (mod 17) and hence
x 2 − 2xy + y 2 − 5x + 7y ≡ y 2 − 5y + 6 (mod 17).
66. Let ABCDPQRS be a cube, and the numbers
a , b , c , d , e , f be written on the faces Thus, 17 divides y 2 − 5y + 6. But x ≡ 2y − 1
ABCD, BQRC , PQRS , APSD, ABQP , CRSD (mod 17) also implies that
respectively. Then, the products written at the xy − 12x + 15y ≡ 2(y 2 − 5y + 6) (mod 17).
corners A , B , C , D, P , Q , R , S are respectively
Since, 17 divides y 2 − 5y + 6, it follows that 17
ade , abe , abf , aaf , cde , bce , bcf , caf . The sum
of these 8 numbers is divides xy − 12x + 15y .
S R 69. Let Q = n (n + 1). It is convenient to choose

n = m 2 , for then Q is already a sum of two
D C squares : Q = m 2 (m 2 + 1) = (m 2 )2 + m 2. If
further m 2 itself is a sum of two squares, say
m 2 = p 2 + q 2 , then
Q
P Q = (p 2 + q 2 )(m 2 + 1) = (pm + q )2 + (p − qm )2.
A B Note that the two representation for Q are
= (e + f )(ab + bc + cd + ad ) distinct. Thus, for example, we may take
m = 5k , p = 3k , q = 4k , where k varies over
= (e + f )(a + c )(b + d ). natural numbers. In this case n = m 2 = 25k 2 ,
This is given to be equal to 2004 = 22 ⋅ 3 ⋅ 167. are
Observe that none of the factors Q = (25k 2 )2 + (5k )2 = (15k 2 + 4k )2 + (20k 2 − 3k )2.
a + c , b + d , e + f is equal to 1. Thus
As we vary k over natural numbers, we get
(a + c )(b + d )(e + f ) is equal to 4 ⋅ 3 ⋅ 167,
infinitely many number of the from n (n + 1)
2 ⋅ 6 ⋅ 167, 2 ⋅ 3 ⋅ 334 or 2 ⋅ 2 ⋅ 501. Hence the
each of which can be expressed as a sum of
possible values ofT = a + b + c + d + e + f are
two squares in two distinct ways.
4 + 3 + 167 = 174, 2 + 6 + 167 = 175,
2 + 3 + 334 = 339, or 2 + 2 + 501 = 505. 70. Since 13 dividesa + 11b , we see that 13 divides
Thus, there are 4 possible values ofT and they a − 2b and hence it also divides 6a − 12b. This
in turn implies that 13 | (6a + b ).
are 174, 175, 339, 505.
Similarly 11 | (a + 13b ) ⇒ 11 | (a + 2b )
67. By data p1 = 2, p 2 = 3, p3 = 7. It follows by
induction that pn , n ≥ 2 is odd. [For if ⇒ 11 | (6a + 12b ) ⇒ 11 | (6a + b ).
p 2 , p3 , . . . , pn − 1 are odd, then p1p 2 . . . pn − 1 + 1 Since GCD (11, 13) = 1, we conclude that
is also odd and nor 3. This also follows by 143 | (6a + b ).
induction. For if p3 = 7 and if p3 , p 4 , . . . , pn − 1
Thus, we may write 6a + b = 143k for some
are neither 2 nor 3, then p1p 2p3 . . . pn − 1 + 1 are
natural number k. Hence,
neither by 2 nor by 3. So, p a is neither 2 nor 3.
6a + 6b = 143k + 5b = 144k + 6b − (k + b ).
www.pdfworld.in
www.pdfworld.in
This shows that 6 divides k + b and hence or c − a = u 2 , c − b = 2v 2 ,

k + b ≥ 6. We therefore obtain
where u > 0 and v > 0 and GCD (u , v ) = 1. In
6(a + b ) = 143k + 5b = 138k + 5(k + b ) either of the cases d = − (a + b − c ) ± 2uv . In
≥ 138 + 5 × 6 = 168 the first case
If follows that a + b ≥ 28. Taking a = 23 and c + d = 2c − a − b ± 2uv
b = 5, we see that the conditions of the = 2u 2 + v 2 ± 2uv
problem are satisfied. Thus the minimum
value of a + b is 28. = (u ± v )2 + u 2
71. Consider the dissection of the given 6 × 6 We observe that u = v implies that u = v = 1
square in to non-congruent rectangles with and hence c − a = 2, c − b = 1. Hence a , b , c are
least possible areas. The only rectangle with three consecutive integers. We also see that
area 1 is an 1 × 1 rectangle. Similarly, we get c + d = 1. forcing b + d = 0, contradicting that
1 × 2, 1 × 3 rectangles for areas 2, 3 units. In b + d is a side of a triangle. Thus u ≠ v and
the case of 4 units we may have either a 1 × 4 hence c + d is the sum of two non-zero integer
rectangle or a 2 × 2 square. squares.
Similarly, there can be a 1 × 5 rectangle for area Similarly, in the second case we get
5 units and 1 × 6 or 2 × 3 rectangle for 6 units. c + d = v 2 + (u ± v )2. Thus, c + d is the sum of
Any rectangle with area 7 units must be 1 × 7 two squares.
rectangle, which is not possible since the Aliter
largest side could be 6 units. And any One may use characterisation of primitive
rectangle with area 8 units must be a 2 × 4 Pythagorean triples. Observe that GCD
rectangle. If there is any dissection of the (c − a , c − b ) = 1 implies that c + d , a + d , b + d
given 6 × 6 square in to 9 non-congruent are relatively prime. Hence, there exist
rectangles with areas integers m > n such that
a1 ≤ a 2 ≤ a3 ≤ a 4 ≤ a5 ≤ a 6 ≤ a7 ≤ a 8 ≤ a 9, then we
a + d = m 2 − n 2 , b + d + 2mn , c + d = m 2 + n 2.
observe that
a1 ≥ 1, a 2 ≥ 2, a3 ≥ 3, a 4 ≥ 4, a5 ≥ 4, a 6 ≥ 5, 73. Suppose r pages of the book are torn off.
a7 ≥ 6, a 8 ≥ 6, a 9 ≥ 8, Note that the page numbers on both the sides
of a page are of the form 2k − 1 and 2k, and
and hence the total area of all the rectangles is
their sum is 4k − 1. The sum of the numbers on
a1 + a 2 + . . . + a 9 ≥ 1 + 2 + 3 + 4 + 4 + 5 the torn pages must be of the form
+ 6 + 6 + 8 = 39 > 36 4k1 − 1 + 4k2 − 1 + . . . + 4kr − 1
which is the area of the given square. Hence, if = 4(k1 + k2 + . . . + kr ) − r .
a 6 × 6 square is dissected in to 9 rectangles as The sum of the numbers of all the pages in the
stipulated in the problem, there must be two untron book is
congruent rectangles. 1 + 2 + 3 + . . + 100 = 5050
72. We have Hence, the sum of the numbers on the torn
(c + d )2 = (a + d )2 + (b + d )2. pages is 5050 − 4949 = 101.
This reduces to We therefore have
d 2 + 2d (a + b − c ) + a 2 + b 2 − c 2 = 0. 4(k1 + k2 + . . . + kr ) − r = 101
This shows that r ≡ 3 (mod 4). Thus r = 4l + 3
Solving the quadratic equation for d , we
for some l ≥ 0.
obtain
Suppose r ≥ 7, and suppose
d = − (a + b − c ) ± (a + b − c )2 − (a 2 + b 2 − c 2 ) k 1 < k2 < k 3 < . . < k r . Then we see that
= − (a + b − c ) ± 2(c − a )(c − b ). 4(k 1 + k2 + . . . + k r ) − r ≥ 4(k1 + k 2 + . . + k 7 ) − 7
Since d is an integer, 2(c − a )(c − b ) must be a ≥ 4(1 + 2 + . . . + 7) − 7
perfect square; say 2(c − a )(c − b ) = x 2 , But GCD = 4 × 28 − 7 = 105 > 101
(c − a , c − b ) = 1. Hence we have Hence, r = 3. This leads to k 1 + k2 + k 3 = 26
c − a = 2u 2 , c − b = v 2 and one can choose distinct positive integers
k 1 , k2 , k 3 in several ways.
www.pdfworld.in
www.pdfworld.in
74. We write We observe that for any integers x , y ,

a 2 − 3a − 19 = a 2 − 3a − 70 + 51 x 2 + 2xy + 2y 2 = (x + y )2 + y 2 ≥ y 2,
= (a − 10) (a + 7) + 51
and x 2 − 2xy + 2y 2 = (x − y )2 + y 2 ≥ y 2.
Suppose 289 divides a 2 − 3a − 19 for some
We write
integer a. Then, 17 divides it and hence 17
divides (a − 10) (a + 7). Since 17 is a prime, it 32008 + 42009 = 32008 + 4(42008 )
must divide (a − 10) or (a + 7). But = (3502 )4 + 4(4502 )4 .
(a + 7) − (a − 10) = 17. Hence, whenever 17
divides one of (a − 10) and (a + 7), it must Taking x = 3502 and y = 4502, we see that
divide the other also. Thus, 172 = 289 divides 32008 + 42009 = ab, where
(a − 10) (a + 7). It follows that 289 divides 51,
a ≥ (4502 )2 , b ≥ (4502 )2.
which is impossible. Thus, there is no integer a
for which 289 divides a 2 − 3a − 19. But we have
75. We use the standard factorisation : (4502 )2 = 22008 > 22002 = (211 )182 > (2009)182,
x 4 + 4 y 4 = (x 2 + 2xy + 2y 2 ) (x 2 − 2xy + 2y 2 ) since 211 = 2048 > 2009
Level 2
1. We look for powers of 2 congruent to 1  N  = 3 and  N  = 1
modulo 7 and find that 21 ≡ 2 (mod 7)  32   33 
22 ≡ 4 (mod 7); 23 ≡ 1 (mod 7) ∴  N  ≥ 9 and N ≥ 27

 3 
It follows that every natural numbers k
 27  +  27  +  27  +  27 
23k = (23 | k ≡ 1k ≡ 1 (mod 6) Now,
 2   22   23   24 
Hence, every number of the form 23k − 1
= 13 + 6 + 3 + 1 = 23
divisible by 7, when n is not a multiple of 3. 23
So, 2 | N ! (not enough)
It is of the form 3k + 1 or 3k + 2.
But N = 28 works for both 2 and 3.
Q 23k ≡ 1 (mod 7)
3. We have
we have (a + b )7 − a7 − b7 = 7ab (a + b )(a 2 + ab + b 2 )2
23k + 1 = 2 ⋅ 23k ≡ 2 (mod 7)
Since, 7 does not divide ab (a + b ), we must
23k + 2
= 4 ⋅ 23k ≡ 4 (mod 7) choose a, b so that 73 divides a 2 + ab + b 2
From which it follows that multiples of 3 are i.e., a 2 + ab + b 2 ≡ 0 (mod 73 ) …(i)

the only exponents n such that 2n − 1 is Q a − b = (a − b )(a + ab + b )
3 3 2 2
…(ii)
divisible by 7. From relation it follows that
Eq. (ii) is equivalent to
23k + 1 + 1 ≡ 3 (mod 7) a3 ≡ b3 (mod 73 ) …(iii)
3k + 2
2 + 1 ≡ 5 (mod 7) For any number c relatively prime to n, we
φ
Moreover 23k + 1 ≡ 2 (mod 7) have e (n ) ≡ 1 (mod n);
So, 2n + 1 leaves a remainder of 2, 3 or 5 when Now, φ (73 ) = (7 − 1)72 = 3.98

divided by 7 and hence is not divisible by 7. so c 3. 08 ≡ 1 (mod 73 ) for any c ∉0 (mod 7)
2. It is necessary that 234 and 312 each divide N ! . for e.g., c = 2 , set b = 1, a = 298
If   = 8, then  2  = 2 and  3  = 0
N N N
Then, (298 )3 ≡ 1 (mod 73 ).
 3   3   3 
Q 298 ≡ 4 (mod 7), a + b = 298 + 1 ≡ 5 (mod 7)
So, 310| N ! (not enough). If   = 9, then
N
 3  and a − b = 298 − 1 ≡ 3 (mod 7).
www.pdfworld.in
www.pdfworld.in
7 does not divide ab (a + b ) nor (a − b ). (ii) A → setting x = 1, we get 5496 > q 992.
Now, 298 is terribly large and can be reduced to Since ai ≥ 0, to show that 10347 > a 992 it
size [mod 73 = 343] suffices to show that
for e.g., 10347 > 5496 or 347 > 496 (1 − log 2)
2 10
= 1024 = 37
. − 5, so 2
3 10
≡ − 5 (mod 7 )
3
An easy way to verify the last inequality is
so 2 20
≡ 25 (mod 7 ), 2
3 40
≡ 61, 280 ≡ − 52 to recall that
03010
. < log 2 < 03011
.
290 ≡ − 83, 28 ≡ − 87, 298 ≡ 18 (mod 73 )
Then,
so a = 18, b = 1 is a solution. 496(1 − log 2) < 496(1 − 03010
. ) = 346704
.
4. First consider Now, 210 = 1024 > 103, multiply our
x − 1 = (x − 1)(x
5 4
+ x + x + x + 1)
3 2
inequality by 2496 and raise it to the tenth
where x = 5397 . We can verify that power.
It reduces to
x 4 + x 3 + x 2 + x + 1 = (x 2 + 3x + 1)2 24960 > 101490 or 1.024496 > 100
−5x (x + 1)2 …(i) or (1.024)124 > 10
Since, x = 5397 the RHS of Eq. (i) is the In binomial expansion.
difference of two squares and can be factored. (1 + 0.024)124 = 1 + 124(0.024) + K
It is easy to verify that each of the three the sum of first two term is already > 10.
factors of 51985 − 1 exceeds 5100.
(iii) B → It follows inductively that the
5. (i) Denote the five distinct roots of x 5 − 1 = 0 coefficients a 0 , a1 , K are unimodal and
by 1, ω, ω2 , ω3 , ω4 . symmetric a 992 being the largest.
Thus, 1985 a 992 > 5496. To show that
If k is +ve integer, then
a 992 > 10340, it suffices to show that
1+ω +ωk 2k
+ω 3k
+ω 4k
5496 > 2000. 10340 or 497 − 343 > 497 log 2

5 if k is multiple of 5
= 
0 otherwise or 153 > 149.647 …(ii)
Alternatively we can use the fact that
The first result is obvious and second
54 > 29 to conclude first that (54 )39 > (29 )39 or
follows by substitution of x = ωk in the
5156 > 2351
identity
x 5 − 1 = (x − 1)(1 + x + x 2 + x 3 + x 4 ) Q 2351 = 211 ⋅ 2340 > 2 ⋅ 103 ⋅ 2340
Now, replace x successively by 1, ω, ω2 , ω3 , We have that 5156 > 20002

. 340 which is
ω in the identity
4 equivalent to Eq. (ii).
x 2 (1 + x + x 2 + x 3 + x 4 )496 6. Let m = n + 10, then n = m − 10
= x 2 (a 0 + a1x + a 2x 2 + a3x 3 + K ). n3 + 100 = (m − 10)3 + 100
(We multiplied the given identity by x 2 in = m3 − 30m 2 + 300m − 900.

order that a3 , a 8 , K , a5n + 3 be coefficients Condition (n + 10) / (n3 + 100) now translates
of fifth powers of x). Add the resulting five into
equations and divide by 5 to get
m | m3 − 30m 2 + 300m − 900
5495 = a3 + a 8 + a13 + K + a1983
Since, m is a divisor of each of the quantities
It shows that any common divisor of m3 , 30m 2 , 300m.
a3 , a 8 , K , a1983 is a power of 5.
It follows that m |900. The largest m for which
Now, a1983 = 496 which is not divisible by 5. this is true is m = 900, so it follows that the
It follows that gcd we seek is 1. largest n with the given property is n = 890.
www.pdfworld.in
www.pdfworld.in
7. n 2 + 3n + 2 = (n + 1)(n + 2) and 6 = 2 ⋅ 3 = (a + d )(c − 1) + (b − 1)(a + 1)

So if 6 is to be divisor of n + 3n + 2 then either
2 ≥ GCD (a + d , b − c )
(a) 6 is a divisor of n + 1 . which is false.
(b) 6 is a divisor of n + 2 . Case II GCD (a + d , b − c ) = 1

(c) 3 is a divisor of n + 1 and 2 is a divisor of substituting ac + bd = (a + d ) b − (b − c ) a
n + 2 or for LHS of ac + bd = (b + d + a − c )
(d) 2 is a divisor of n + 1 and 3 is a divisor of (b + d − a + c )
n+2
we get (a + d )(a − c − d ) = (b – c ) (b + c + d )
Possibility (a) holds for n = 5, 11, 17,…, 1991 or There exist a +ve integer k such that
332 values in all. Possibility (b) holds for
a − c − d = k (b − c );
n = 4,10, 16,…,1990, another 332 values.
Possibility (c) holds for n = 2, 8, 14, ……, 1981 b + c + d = k (a + d )
another 332 values and possibility (d) holds Adding these we obtain
for n = 1, 7, 13, ……, 1987 yet another 332 a + b = k (a + b − c + d ) and thus,
values. So, there are 4 × 332 = 1328 values of n
k (c − d ) = (k − 1)(a + b )
between 1 and 1991 for which n 2 + 3n + 2 is
Recall that a > b > c > d
divisible by 6.
If k = 1, then c = d , a contradiction.
8. Let n be a given +ve integer, let d be any divisor k a+b
of n. If k ≥ 2, then 2 ≥ = >2
k −1 c −d
Then, n/d is an integer. It is also a divisor of n a contradiction
(as n = d × n / d ). If n is not a square, then
n ≠ d 2, so d and n/d are unequal. If we pair up So, ab + cd is not prime.
a
d and n/d then each divisor acquires one and 11. Let n = p1 1 K prar
precisely one match.
Then, d (n ) = (a1 + 1)(a 2 + 1)…(ar + 1)
The divisors now get grouped into pairs and and d (n 2 ) = (2a1 + 1)(2a 2 + 1)…(2ar + 1)
this tell us that the number of divisor is even
(for it is twice the number of pairs). so, ai must be chosen so that
(2a1 + 1)(2a 2 + 1) K (2ar + 1)
9. Q 23 ≡ 1 mod 7
= k (a1 + 1)(a 2 + 1)K(ar + 1)
23n ≡ 1 mod 7 for any +ve integer n.
Q (2ai + 1) are odd
+1
∴ 23n ≡ 2 mod 7
⇒ k must be odd we show that conversely
+ 2
23n ≡ 4 mod 7 given any odd k, we can find ai .
So, remainder (2n ÷ 7) cycles through the We use a form of induction on k.

values 1, 2 and 4 in that order. It is true for k = 1 (Take n = 1).
If it is not true for 2m k − 1.
We never have 2n ≡ − 1, this means that 2n + 1
is never a multiple of 7. That is sufficient since any odd number has
the form 2m k − 1 for some smaller odd number
10. Suppose ab + cd is prime.
k. Take ai − 2i (2m − 1)k − 1 for i = 0, 1, …, m − 1
ab + cd = (a + d ) c + (b − c ) a
+1 +1
= m GCD (a + d , b − c ) Then, 2ai + 1 = 2i (2m − 1)k − (2i − 1)
for some +ve integer m. By assumption either and ai + 1 = 2 (2

i m
− 1)k − (2 − 1)
i
m = 1 or gcd (a + d , b − c ) = 1 So, the product of the (2ai + 1), divided by the

Case I m = 1, then product of the (ai + 1)'s is 2m (2m − 1)k
− (2m − 1) divided by (2m − 1)k
GCD (a + d , b − c )
= ab + cd > ab + cd − (a − b + c + d ) or (2m k − 1) / k.
www.pdfworld.in
www.pdfworld.in
Thus, if we take these ai ' s together with those Certainly y exists and indeed y < q.
giving k, we get 2m k − 1 which completes Since ( p − 1) q −1
= 1 mod q
induction.
We know that ( p − 1) n = − 1 mod q
12. (a , b ) = (11, 1), (49, 1) or (7k 2 , 7k ).
So, x exists also.
If a ab + b ≥ (a + 1)(ab + 1)
2 2
So, ( p − 1) r = − 1 mod q
= a 2b + a + ab ≥ a 2b + a + b Hence, x ≤ r < y (r cannot be zero).
So, there can be no solutions with a < b. [Q1 is not –1 mod q]
Assume that a ≥ b let k = the integer write n = hx + k with 0 ≤ k < x
(a 2b + a + b ) + (ab 2 + b + 7). We have Then, − 1 = ( p − 1)n = (−1) h ( p − 1)k mod q h
(a / b + 1 / b )(ab + b + 7) = ab + a
2 2
cannot be even, then
+ ab + 7a / b + 7 / b + 1 > ab 2 + a + b ( p − 1)k = − 1 (mod q), contradicting the
So, k < a / b + 1 / b. Now, if b ≥ 3, minimality of x, so h is odd.
Then, b (b − 7 / b ) > 0 Hence, ( p − 1)k = 1 mod q with 0 ≤ k < x < y.
∴ (a / b − 1 / b )(ab + b + 7) = ab + a − a
2 2
This contradicts the minimality of y unless
k=0
(b − 7 / b ) − 1 − 7 / b < ab 2 + a < ab 2 + a + b
So, n = hx but x < q so x = 1.
Either b = 1 or 2 or k > a / b − 1 / b
So, ( p − 1) = − 1 mod q p and q are primes.
If a / b − 1/ b < k< a / b + 1/ b
∴ q =p
Then, a − 1 < kb < a + 1
So, p is the smallest prime divisor of n.
Hence, a = kb. It gives solution (a , b ) = (7k 2 , 7k ) We are also given that n ≤ 2p so either.
It remains to consider b = 1 and 2. p = n or p = 2, n = 4. The latter does not work
If b = 1, then a + 8 divides a 2 + a + 1 so we have shown that n = p.
Evidently n = p = 2 and n = p = 3 work
Also, a (a + 8) − (a 2 + a + 1) = 7a − 1
Now, p > 3 we show that there are no solutions
Also, 7(a + 8) − (7a − 1) = 57
of this type.
Only factors bigger than 8 are 19 and 57.
Expand ( p − 1) p + 1 by the binomial theorem
So a = 11 or 49. It is easy to check
to get
(a , b ) = (11, 1) and (49, 1) are indeed solution. 1
Q (−1) p = (−1) 1 + (−1) + p 2 − p ( p − 1)p 2
If b = 2, then 4a + 9 divides 2a 2 + a + 2 2
p ( p − 1)( p − 2)
Also a (4a + 9) − 2(2a 2 + a + 2) = 7a − 4 +
6p3
Also 7(4a + 9) − 4(7a − 4) = 79
Only factor greater than 9 is 79. The terms of the form pi with i ≥ 3 are
obviously divisible by p3.
But that gives a = 35 / 2 which is not integral.
Hence, there are no solutions for b = 2. QBinomial coefficients are all integral. Hence,
sum is
13. (1, p ) is a solution for every prime p.
p2 + (a multiple of p3 ). so the sum is not
Assume n > 1 and take q to be the smallest
prime divisor of n. divisible by p3. But for p > 3, pp − 1 is divisible by
p3 so it cannot divide ( p − ) p + 1
We first show that q = p let x be the smallest
+ve integer for which ( p − 1)x = − 1 mod q. There are no more solutions.
y the smallest +ve integer for which 14. (1 ⋅ 21 ) + 2 ⋅ 22 + 3 ⋅ 23 + K + n ⋅ 2n

+ 10)
( p − 1)y = 1 mod q. = 2(n + (2)
www.pdfworld.in
www.pdfworld.in
LHS can be summed as 0 ≤ b < 1, 0 ≤ d < 1

n +1
2 + 2 + 2 + K+ 2 =2
1 2 3 n
−2 Then, Eq. (ii) can be written as
2 + 2 + K+ 2 =2
2 3 n n +1
−2 2 a + c ≥ [(3a + c + 3b + d ) / 5]
n +1 + [3c + a + 3d + b ) / 5] …(iii)
2 + K+ 2 =2
3 n
−2 3
………………… Now, 1 > u > 2v , so that 5 > 5u > 10v or 5 > a + b

> 2c + 2d. The first inequality here gives. 5 > a
…………………
or 4 ≥ a.
…………………
2n +1
− 2n IInd inequality gives a ≥ 2c as a < 2c would
+ 2n = n +1 n +1 imply that a ≤ 2c − 1, a + 1 − 2c ≤ 0 and
n(2 ) − (2 − 2)
a + b − 2c < 0
n +1
=2 (n − 1) + 2
Thus, we have 4 ≥ a ≥ 2c and hence only cases
QThis equals we need to consider are
+ 10
2n + 2, a 4 4 4 3 3 2 2 1 0
+ 10 c 2 1 0 1 0 1 0 0 0
2n
n −1= n +1
= 27
2 It is easy to verify Eq. (iii) in these 9 cases
So, n = 2 + 1 = 513
9 as 3a + d and 3d + b < 4
15. Set A = a 2 + ab + b 2 17. Let us denote the digits in tens, hundreds and
ones place of the sought for number N. as
B =a −b x , y , z respectively
Then, a − b3 = AB. If B is divisible by 2n , then
3
N = 100x + 10y + z
product AB is divisible by 2n 100x + 10y + z = x !+ y ! + z !
Q a and b are odd. Now, 7 ! = 5040
A = a 2 + ab + b 2 is a sum of 3 odd numbers So none of digits of N exceeds 6.
N itself does not exceed 700.
and is therefore also odd.
Its digit can exceed 5 (as 6 ! = 720 > 700)
So, A is relatively prime to 2n .
At least one digit of N = 5. N is a 3 digit
Thus, 2n divides AB only if it divides B.
number and 3·4!= 72 < 100.
16. Let x = x1 + u ∴ x cannot be equal to 5
y = y1 + v Q 3 ⋅ 5 ! = 360 < 500
where x 1 and y are non negative integers and x cannot exceed 3.
0 ≤ u < 1, 0 ≤ v < 1 x does not exceed 2 since 3 ! + 2 ⋅ 5 ! = 246 < 300
Now, x 1 + y1 + [5u ] + [5v ] But 255 does not satisfy.
≥ [3u + v ] + [3v + u ] …(i) x cannot exceed 1 as 2 ! + 5 ! + 4 ! = 146 < 200
We prove that, Now, 1 ! + 5 ! + 4 ! = 145 < 150
[5u ] + [5v ] ≥ [3u + v ] + [3v + u ] …(ii) y cannot exceed 4.
which implies Eq. (i). z = 5 [at least one value of N must be 5]
We have x = 1, 4 ≥ y ≥ 0 and z = 5.
In view of the symmetric roles of u and v we
assume u ≥ v which gives inequality So, N = 145
[5u ] ≥ [3u + v ]. 18. N = 1
If u ≤ 2v we also have
19. 2x + 3y = k …(i)
[5v ] ≥ [3v + u ]
k − 3y k −y
So Eq. (ii) is established in this case finally, we We obtain x = = −y + …(ii)
2 2
prove that Eq. (ii) hold if u > 2v
Let 5u = a + b and 5v = c + d . where a and c are From Eq. (ii), we find that x is an integer only if
non –ve integers. (k − y )/2 is an integer.
www.pdfworld.in
www.pdfworld.in
Hence, (k − y / 2 = s We can write by analogy with first of equality

Hence, y = k − 2s (2 ⋅ 10m + x ) ⋅ 2 = 10x + 2
equation 2 yields 4 ⋅ 10m − 2 2 ⋅ 10m − 1
x = =
x = − y + s = − (k − 2s ) + 3s − k. 8 2
If integers x and y satisfy Eq. (i) they are of or (4 ⋅ 10m + x ) ⋅ 2 = 10 × + 4
form
8 ⋅ 10m − 4 2 ⋅ 10m − 1
x = − k + 3s , y = k − 2s …(iii) x = =
8 2
where s is some integer.
Neither of the formulas for no. x hold as a
for an arbitrary integer s, Eq. (iii) defines whole no. can not be equal to a fraction whose
integers x and y which satisfied Eq. (i). numerator is odd and denominator is even.
In a same way we can determine integral 21. Let n 2 + n + 41 = m 2 , then
solution of 9x + 5y = l …(iv)
where l is a given integer. 4m 2 = 4n 2 + 4n + 164 = (2n + 1)2 + 163
l − 9x l +x (2m + 2n + 1)(2m − 2n − 1) = 163
y = = − 2x + …(v)
5 5 Q 163 is prime we must have
y is an integer only if (l + x ) / 5 is an integer say 2m + 2n + 1 = ± 1, ± 163.
l +x Correspondingly 2m − 2n − 1 = ± 163, ± 1.
= t.
5 Subtracting 4n + 2 = + 162
Hence, x = 5t − l and substitute in Eq. (v) Hence, n = − 41 or 40.
y = − 2x + t = − 2(5t − l ) + t = − 9t + 2l 22. Suppose there are four numbers a , b , c , d
If x and y satisfy Eq. (iv) they are of form among the given nine numbers which leave
x = 5t − l , y = − 9t + 2l …(vi) the same remainder modulo 20. Then,
where t is an integer. a + b ≡ c + d (mod 20) and we are done.
If not, there are two possibilities :
Conversely, for an arbitrary integer t
Eq. (vi) defines integers x and y which satisfy 1. We may have two disjoint pairs {a , c } and
Eq. (iv) {b , d } obtained from the given nine
numbers such that a ≡ c (mod 20) and
If 2x + 3y = some multiple of 17 say 17.n.
b ≡ d (mod 20). In this case we get
then x = − 17n + 33
. s (x , y) a + b ≡ c + d (mod 20).
y = 17n − 3s (n , s ) are integers 2. Or else there are at most three numbers
then 9x + 5 = 9(−17n + 35) + 5(17n − 2s ) having the same remainder modulo 20 and
the remaining six numbers leave distinct
= 17(−4n + s ) remainders which are also different from
i.e., if 2x + 3y is divisible by 17 so 9x + 5y . the first remainder (i.e., the remainder of
the three numbers). Thus there are at least
For all integers x and y for which 9x + 5y is
7 distinct remainders modulo 20 that can
divisible by 17, 2x + 3y is also divisible by 17.
be obtained from the given set of nine
20. Q Product of the sought for number by 2 has numbers. These 7 remainders give rise to
same number of digits as the original number.  7
  = 21 pairs of numbers. By pigeonhole
Initial digit of that number cannot exceed 4.  2
When initial digit carried to end the resultant
principle, there must be two pairs
number must be even (it is equal to the
(r1 , r2 ), (r3 , r4 ) such that
duplicated original number). So, initial digit of
r1 + r2 ≡ r3 + r4 (mod 20). Going back we get
the sought for number is equal to 2 or 4. four numbers a , b , c , d such that
Let us suppose that the initial digit of the a + b ≡ c + d (mod 20).
sought for number is equal to 2 or 4 on If we take the numbers 0, 0, 0, 1, 2, 4, 7, 12, we
denoting x the number obtained from sought check that the result is not true for these eight
for number by discading its initial digit. numbers.
www.pdfworld.in
www.pdfworld.in
23. We show that the answer is NO. Suppose, if Thus we see that Q ≥ 4. Taking a = 1, b = 1 and
possible, let a , b , c be three distinct positive c = 2, we get Q = 4. Therefore, the least value
real numbers such that a , b , c , b + c − a , of Q is 4 and this is achieved only by
c + a − b , a + b − c and a + b + c form a 7-term a+b+c=4 and
arithmetic progression in some order. We may (a − b )2 + (b − c )2 + (c − a )2 = 2. The triples for
assume that a < b < c . Then, there are only two which Q = 4 are therefore given by
cases we need to check:
(a , b , c ) = (1, 1, 2), (1, 2, 1), (2, 1, 1).
(I) a + b − c < a < c + a − b < b < c < b + c − a
25. Obviously p ≠ q. We write this in the form
<a + b + c −1 −2
p ( pn + pn + . . . + 1) = q (q + 1).
and (II) a + b − c < a < b < c + a − b < c
< b + c − a < a + b + c. If q ≤ p n/2
− 1, then q < pn / 2 and hence we see
Case I Suppose the chain of inequalities that q 2 < pn . Thus, we obtain

a + b −c<a<c + a −b<c 2. It follows that q ≥ pn / 2. Since, n > 2
difference. Thus we see that
and is an even number, n / 2 is a natural
c = a + b + c − 2d , b = a + b + c − 3d ,
number larger than 1. This implies that
a = a + b + c − 5d .
q ≠ pn / 2 by the given condition that q is a
Adding these, we see that a + b + c = 5d . But
then a = 0 contradicting the positivily of a. prime. We conclude that q ≥ pn / 2 + 1. We may
also write the above relation in the form
Case II Suppose the inequalities
p ( pn / 2 − 1)( pn / 2 + 1) = ( p − 1) q (q + 1).
a + b −c<a<b<c + a −b<c
<b + c −a<a + b + c This shows that q divides ( pn / 2 − 1)( pn / 2 + 1).
are true. Again we see that But q ≥ pn / 2 + 1 and q is a prime. Hence, the
c = a + b + c − 2d , b = a + b + c − 4d , only possibility is q = pn / 2 + 1. This gives
a = a + b + c − 5d . p ( pn / 2 − 1) = ( p − 1)(q + 1) = ( p − 1)( pn / 2 + 2).
We thus obtain a + b + c = (11/ 2) d . This gives Simplification leads to 3p = pn / 2 + 2. This
1 3 7 shows that p divides 2. Thus p = 2 and hence
a = d , b = d , c = d.
2 2 2 q = 5, n = 4. It is easy to verify that these
Note that a + b − c = a + b + c − 6d = − (1/ 2) d . indeed satisfy the given equation.
However we also get
26. We show that the required GCD is 24. Consider
a + b − c = [(1/ 2) + (3/ 2) − (7/ 2)] d an element (a , b , c , d , e , f ) ∈ S . We have
= − (3/ 2) d. a 2 + b 2 + c 2 + d 2 + e 2 = f 2.
It follows that 3e = e giving d = 0. But this is We first observe that not all a , b , c , d , e can be
impossible. odd. Otherwise, we have
Thus there are no three distinct positive real a2 ≡ b 2 ≡ c 2 ≡ d2 ≡ e 2 ≡ 1 (mod 8)
numbers a , b , c such that a, b, c, b + c − a, and hence f 2
≡ 5 (mod 8), which is impossible
c + a − b, a + b − c and a + b + c form a 7-term
because to square can be congruent to 5
arithmetic progression in some order.
modulo 8. Thus, at least one of a , b , c , d , e is
24. We observe that even.
Q = a3 + b3 + c 3 − 3abc Similarly, if none of a , b , c , d , e , is divisible by
1 3, then a 2 ≡ b 2 ≡ c 2 ≡ d 2 ≡ e 2 ≡ 1 (mod 3) and
= (a + b + c ) [(a − b )2 + (b − c )2 + (c − a )2 ]
2 hence f 2 ≡ 2 (mod 2) which again is impossible
Since, we are looking for the least positive because no square is congruent to 2 modulo 3.
value taken by Q, it follows that a , b , c are not Thus, 3 divides abcdef .
all equal. There are several possibilities for a , b , c , d , e.
Thus, a + b + c ≥ 1 + 1 + 2 = 4 and
Case I Suppose one of them is even and the
(a − b )2 + (b − c )2 + (c − a )2 ≥ 1 + 1 + 0 = 2. other four are odd; say a is even, b , c , d , e are
www.pdfworld.in
www.pdfworld.in
odd. Then, b 2 + c 2 + d 2 + e 2 ≡ 4 (mod 8). If (b) If there are only finitely many even terms
a ≡ 4 (mod 8), then f
2 2
≡ 0 (mod 8) and hence and x t is the last term, then the sequence
〈x n 〉n∞ = t + 1 = 〈x t + 1 , x t + 2 , . . . 〉 is obtained in
2 | a , 4 | f giving 8| af . If a 2 ≡ 0 (mod 8), then
f 2 ≡ 4 (mod 8) which again gives that 4 | a and a similar manner and hence must have an
even term by (a), a contradiction. Thus,
2 | f so that 8| af . It follows that 8| abcdef and 〈x n 〉n∞ = 1, has infinitely many even terms.
hence 24 | abcdef .
Case II Suppose a , b are even and c , d , e are 28. It can be easily seen that
odd. Then, c 2 + d 2 + e 2 ≡ 3 (mod 8). Since 9n = (2b + 2c − a )2 + (2c + 2a − b )2

a 2 + b 2 ≡ 0 or 4 modulo 8, it follows that f 2 ≡ 3 + (2a + 2b − c )2.
or 7(mod 8) which is impossible. Hence, this Thus, we can take p1 = p 2 = p3 = 2,
case does not arise. q1 = q 2 = q3 = 2 and r1 = r2 = r3 = − 1. Suppose 3
does not divide GCD (a , b , c ). Then, 3 does
Case III If three of a , b , c , d , e are even and
divide at least one of a , b , c , say 3 does not
two odd, then 8| abcdef and hence 24 | abcdef . divide a. Note that each of 2b + 2c − a ,
Case IV If four of a , b , c , d , e are even, then 2c + 2a − b and 2a + 2b − c is either divisible by
again 8| abcdef and 24 | abcdef . Here, again 3 or none of them is divisible by 3, as the
for any six tuple (a , b , c , d , e , f ) in S, we difference of any two sums is always divisible
observe that 24 | abcdef . Since by 3. If 3 does not divide 2b + 2c − a , then we
have the required representation. If 3 divides
12 + 12 + 12 + 22 + 32 = 42 . 2b + 2c − a , then 3 does not divide 2b + 2c + a.
We see that (1, 1, 1, 2, 3, 4) ∈S and hence 24 ∈T . On the other hand, we also note that
Thus, 24 is the GCD of T . 9n = (2b + 2c + a )2 + (2c − 2a − b )2
27. (a) Let us assume that there are no even + (−2a + 2b − c )2 = x 2 + y 2 + z 2 ,
numbers in the sequence. This means that
x n + 1 is obtained from x n , by adding a where x = 2b + 2c + a , y = 2c − 2a − b and
non-zero even digit of x n to x n , for each z = − 2a + 2b − c . Since, x − y = 3(b + a ) and 3
n ≥ 1. does not divide x , it follows that 3 does not
Let E be the left most even digit in x 1 which divide y as well. Similarly, we conclude that 3
may be taken in the form does not divide z.
x 1 = O1O2 . . . Ok ED1D2 . . . Dl 29. Suppose 0 ∈ A. Then, a 2 = a 2 + 0 × b is rational
where O1 , O2 , . . . , Ok are odd digits (k ≥ 0); and ab = 02 + ab is also rational for all a , b in
D1 , D2 , . . . , Dl − 1 are even or odd; and Dl odd, A , a ≠ 0, b ≠ 0,a ≠ b. Hence, a = a1 M for some
l ≥ 1. rational a1 and natural number M . For any
Since, each time we are adding at least 2 to b ≠ 0, we have
a term of the sequence to get the next term, ab
b M = ,
at some stage, we will have a term of the a1
form which is a rational number.
x r = O1O2 . . . Ok E 999 . . .9 F Hence, we may assume 0 is not in A. If there is
where F = 3, 5, 7 or 9. Now we are forced to a number a in A such that −a is also in A, then
add E to x r to get x r + 1, as it is the only even again we can get the conclusion as follows.
digit available. After at most four steps of Consider two other elements c , d in A. Then,
addition, we see that some next term is of c 2 + da is rational and c 2 − da is also rational.
the form It follows that c 2 is rational and da is rational.
x s = O1O2 . . . OkG000 . . . M Similarly, d 2 and ca are also rationals. Thus,
where G replaces E of x r , G = E + 1, d/c = (da )/(ca ) is rational. Note that we can vary
M = 1, 3, 5 or 7. But x s has no non-zero even d over A with d ≠ c and d ≠ a. Again c 2 is
digit contradicting our assumption. Hence, rational implies that c = c1 M for some
the sequence has some even number as its rational c1 and natural number M . We observe
term. that c M = c1M is rational, and
www.pdfworld.in
www.pdfworld.in
ca Aliter
a M = ,
c1 As earlier, assume that x is rational and
so that a M is a rational number. Similarly is choose natural numbers k , l such that
the case with −a M . For any other element d , an = an + l for all n ≥ k. Consider the numbers
d am + 1 , am + 2 , . . . , am + l , where m ≥ k is any
b M = Mc1 number. This must contain at least one 0.
c
Otherwise an = 1 for all n ≥ k. But ar = 0 if and
is a rational number. only if r is a square. Hence, it follows that
Thus, we may now assume that 0 is not in A there are no squares for n > k, which is absurd.
and a + b ≠ 0 for any a , b in A. Let a , b , c , d be Thus every l consecutive terms of the
four distinct elements of A. We may assume sequence 〈an 〉 must contain a 0 after certain
| a | > | b. Then, d 2 + ab and d 2 + bc are rational stage. Let t = max{k , l }, and consider t 2 and
numbers and so is their difference ab − bc . (t + 1)2. Since, there are no squares between t 2
Writting a 2 + ab = a 2 + bc + (ab − bc ), and and (t + 1)2, we conclude that a = 1 for
t2 + j
using the facts a + bc , ab − bc are rationals,
2
1 ≤ j ≤ 2t. But then, we have 2t (> l ) consecutive
we conclude that a 2 + ab is also a rational terms of the sequence 〈an 〉 which miss 0,
number. Similarly, b 2 + ab is also a rational contradicting our earlier observation.
number. 31. (a) Consider f ( j ) = am + j + (−1) j am − j,
a a 2 + ab
Consider q= = 2 0 ≤ j ≤ m, where m is a natural number. We
b b + ab observe that f (0) = 2am is divisible by 2am .
Note that a 2 + ab > 0. Thus, q is a rational Similarly,
number and a = bq. This gives f (1) = am +1 − am −1 = 2am
a 2 + ab = b 2 (q 2 + q ). Let us take b 2 (q 2 + q ) = l .
is also divisible by 2am . Assume that 2am
Then, |b | =
l
=
x
, divides f ( j ) for all 0 ≤ j < l , where l ≤ m. We
q2 + q y prove that 2am divides f (l ), Observe
−1
where x and y are natural number. Take f ( l − 1) = am + l −1 + (−1) l am − l + 1.
M = xy . Then, | b | M = x is a rational number. f ( l − 2) = am + (−1) l −2
+l −2 am − l + 2.
Finally, for any c in A , we have
c Thus, we have
c M =b M ,
b am +l = 2am + l −1 + am +l −2
−1
is also a rational number. = 2f ( l − 1) − 2(−1) l am −l + 1
−2
30. We show that x is irrational. Suppose that x is + f ( l − 2) − (−1) l am −l + 2
rational. Then, the sequence 〈an 〉n∞ = 1 is
= 2f ( l − 1) + f ( l − 2)
periodic after some stage; there exist natural + (−1) l − 1 (am + 2 − 2am − l + 1 )
−l
numbers k , l such that an = an + l for all n ≥ k.
−1
Choose m such that ml ≥ k and ml is a perfect = 2f ( l − 1) + f ( l − 2) + (−1) l am −l.
square. Let This gives
α α β β
m = p1 1 p 2 2 . . . prαr , l = p1 1 p 2 2 .. . prβr , f (l ) = 2f (l − 1) + f (l − 2).
be the prime decompositions of m , l so that By induction hypothesis 2am divides f (l − 1)
α j + β j is even for 1 ≤ j ≤ r. Now take a prime p and f (l − 2). Hence, 2am divides f (l ). We
different from p1 , p 2 , . . . , pr . Consider ml and conclude that 2am dividesf ( j )for 0 ≤ j ≤ m.
pml. Since pml − ml is divisible by l , we have (b) We see that f (m ) = a 2m . Hence, 2am divides
apml = aml . Hence, d (pml ) and d (ml ) have same a 2m for all natural number m. Let n = 2k l for
parity. But d (pml ) = 2d (ml ), since GCD(p , ml ) = 1 some l ≥ 1. Taking m = 2k − 1l , we see that
and p is a prime. Since ml is a square, d (ml ) is
2am divides an . Using an easy induction, we
odd. It follows that d (pml ) is even and hence
conclude that 2k al divides an . In particular
apml ≠ aml . This contradiction implies that x is
irrational. 2k divides an .
www.pdfworld.in
www.pdfworld.in
32. Suppose n = pqr , where p < q are primes and above observation shows that n is faithful.
r > 1. Then, p ≥ 2, q ≥ 3 and r ≥ 2, not This shows that a number which is not
necessarily a prime. Thus we have faithful must be of the form 2α3β5γ . We also
n − 2 ≥ n − p = pqr − p ≥ 5p > p, observe that 24 = 16 = 12 + 3 + 1,
n − 2 ≥ n − q = q (pr − 1) ≥ 3q > q , 32 = 9 = 6 + 2 + 1 and 52 = 25 = 22 + 2 + 1,
n − 2 ≥ n − pr = pr (q − 1) ≥ 2pr > pr , so that 24 , 32 and 52 are faithful. Hence,
n − 2 ≥ n − qr = qr (p − 1) ≥ qr . n ∈ N is also faithful if it contains a factor
Observe that p , q , pr , qr are all distinct. Hence, of the form 2α where α ≥ 4; a factor of the
their product divides (n − 2)!. Thus, n 2 = p 2q 2r 2 form 3β where β ≥ 2; or a factor of the form
divides (n − 2)! in this case. We conclude that 5γ where γ ≥ 2. Thus the numbers which are
either n = pq where p , q are distinct primes or
not faithful are of the form 2α3β5γ , where
n = p k for some prime p.
α ≤ 3, β ≤ 1 and γ ≤ 1. We may enumerate all
Case I Suppose n = pq for some primes p , q such numbers.
where 2 n − p = p (q − 1) ≥ 4p, Among these 120 = 112 + 7 + 1,
n − 2 > n − q = q (p − 1) ≥ 2q 60 = 48 + 8 + 4, 40 = 36 + 3 + 1,
30 = 18 + 9 + 3, 20 = 12 + 6 + 2,
Thus p , q , 2p , 2q are all distinct numbers in the
15 = 12 + 2 + 1, and 10 = 6 + 3 + 1. It is easy
set (1, 2, 3, . . . . , n − 2). We see that n 2 = p 2q 2
to check that the other numbers cannot be
divides (n − 2)!. We conclude that n = 2q for written in the required form. Hence, the
some prime q ≥ 3. Note that n − 2 = 2q − 2 < 2q only numbers which are not faithful are
in this case so that n 2 does not divide (n − 2)!. 1, 2, 3, 4, 5, 6, 8, 12, 24.
Case II Suppose n = p k for some prime p. We Their sum is 65.
observe that p , 2p , 3p , . . . , ( p k − 1 − 1)p all lie in (ii) If n = a + b + c with a < b < c is faithful, we
the set {1, 2, 3, . . . , n − 2}. If p k − 1 − 1 ≥ 2k, then see that a ≥ 1, b ≥ 2 and c ≥ 4. Hence, n ≥ 7.
Thus 1, 2, 3, 4, 5, 6 are not faithful. As
there are at least 2k multiples of p in the set
observed earlier, kn is faithful whenever n
{1, 2, 3, . . . , n − 2}. Hence, n 2 = p 2k divides
is. We also notice that for odd n ≥ 7, we can
(n − 2)! . This p k − 1 − 1 < 2k. write n = 1 + 2 + (n − 3) so that all odd n ≥ 7
If k ≥ 5, then p k − 1 − 1 ≥ 2k − 1 − 1 ≥ 2k, which are faithful. Consider 2n , 4n ,8 n, where
may be proved by an easy induction. Hence, n ≥ 7 is odd. By observation, they are all
k ≤ 4. If k = 1, we get n = p, a prime. If k = 2, then faithful. Let us list a few of them :
p − 1 < 4 so that p = 2 or 3, we get n = 22 = 4 or 2n : 14, 18, 22, 26, 30, 34, 38, 42, 46, 50, 54,
n = 3 = 9. For k = 3, we have p − 1 < 6 giving
2 2 58, 62, …
4n : 28, 36, 44, 52, 60, 68, …
p = 2; n = 23 = 8 in this case. Finally, k = 4 gives
8n : 56, 72, …
p3 − 1 < 8. Again p = 2 and n = 24 = 16. However
We observe that 16 = 12 + 3 + 1 and hence it
n 2 = 28 divides 14! and hence is not a solution.
is faithful. Thus all multiples of 16 are also
Thus, n = p , 2p for some primep or n = 8, 9. It is faithful. Thus we see that 16, 32, 48, 64, …
easy to verify that these satisfy the conditions are faithful. Any even number which is not
of the problem. a multiple of 16 must be either an odd
multiple of 2 or that of 4, or that of 8.
33. (i) Suppose n ∈ N is faithful. Let k ∈ N and Hence, the only numbers not covered by
consider kn. Since n = a + b + c , with this process are 8, 10, 12, 20, 24, 40. Of
a > b > c , c | b and b | a, we see that these, we see that
kn = ka + kb + kc which shows that kn is 10 = 1 + 3 + 6, 20 = 2 × 10, 40 = 4 × 10,
faithful.
so that 10, 20, 40 are faithful. Thus, the
Let p > 5 be a prime. Then, p is odd and only numbers which are not faithful are
p = ( p − 3) + 2 + 1 shows that p is faithful. If 1, 2, 3, 4, 5, 6, 8, 12, 24.
n ∈ N contains a prime factor p > 5, then the
Their sum is 65.
www.pdfworld.in
www.pdfworld.in
Unit 2
Theory of Equations
Polynomial
An expression of the form
−1 −2
a 0x n + a1x n + a 2x n + K + an − 1x + an
where n is a whole number and a 0 , a1 , a 2 ,... , 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 a 0 ≠ 0, then a 0 (the coefficient of highest degree term) is called the leading coefficient and a 0x n is called
the leading term.
Degree of Polynomial
The highest index of the variablex occuring in the polynomialf (x )is called the degree of the polynomial.
Zero Degree Polynomial

The constant c = cx 0 is called a polynomial of degree zero.
Linear Polynomial
The polynomial f (x ) = ax + b , a ≠ 0 is of degree one and is called a linear polynomial.
Quadratic Polynomial
The polynomial f (x ) = ax 2 + bx + c , a ≠ 0 is of degree two and is called a quadratic polynomial.
Cubic Polynomial
The polynomial f (x ) = ax 3 + bx 2 + cx + d , a ≠ 0 is of degree three and is called a cubic polynomial.
www.pdfworld.in
www.pdfworld.in
Biquadratic Polynomial
The polynomial f (x ) = ax 4 + bx 3 + cx 2+ dx + e, a ≠ 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 ) = a 0 + a1x + a 2x 2 + K + an x n
g (x ) = b0 + b1x + b2x 2 + K + bm x m
are said to be equal if the coefficients of like powers of x in the two polynomials are equal.
When m ≥ n , if f (x ) = g (x ), then a 0 = b0 , a1 = b1 , …, an = bn , bn + 1 = bn + 2 = K = bm = 0
When m ≤ n, if f (x ) = g (x ), then a 0 = b0 , a1 = b1 , K , am = bm , am +1 = am + 2 = K = an = 0
Complete and Incomplete Polynomials

A polynomial of degree n, which contains all powers of the variable from 0 to n, is called a complete
polynomial. Otherwise it is said to be incomplete polynomial.
For example, the incomplete polynomial 2x 5 + 7x 3 − 5 can be made complete by writing it as
2x 5 + 0x 4 + 7x 3 + 0x 2 + 0x − 5.
Theorem 1 If (atleast) one of the two polynomials f (x ) and g (x ) is the zero polynomial, then the
product f (x ) ⋅ g (x ) is also the zero polynomial.
Proof Let
f (x ) = a 0 + a1x + a 2x 2 + K + ai x i + K
g (x ) = b0 + b1x + b2x 2 + K + bi x i + K
be any two given polynomials.
Since, atleast one of the two polynomials f (x ), g (x ) is zero polynomial.
∴ Without loss of generality
Let f (x ) = zero polynomial
Now, coefficient of x i in f (x ) ⋅ g (x )
= a 0bi + a1bi − 1 + K + ai b0 = 0 ⋅ bi + 0 ⋅ bi −1 + K + 0 ⋅ b0 = 0 ∀ i ≥ 0
f (x ) ⋅ g (x ) = 0
Thus, the product f (x ) ⋅ g (x ) is a zero polynomial.
Theorem 2 Let f (x ), g (x ) be two non-zero polynomials, then

(i) f (x ) ⋅ g (x is a non-zero polynomials.
(ii) deg {f (x ) ⋅ g (x )} = deg f (x ) + deg g (x )
Proof Let f (x ) = a 0 + a1x + a 2x 2 + K + ai x i + K
and g (x ) = b0 + b1x + b2x 2 + K + bi x i + K
be any two non-zero polynomials.
Let deg f (x ) = m
∴ am ≠ 0 and ai = 0
∀ i > m and deg g (x ) = n
∴ bn ≠ 0 and bi = 0, ∀ i > n
www.pdfworld.in
www.pdfworld.in
Theory of Equations 97
+n
(i) Coefficient of x m in f (x ) ⋅ g (x )
= a 0bm +n + a1bm + n −1 + K + am bn + K + am +n b0
= a 0 ⋅ 0 + a1 ⋅ 0 + K + am bn + K + 0 ⋅ b0 [Qai = 0 ∀ i > m, bi = 0 ∀ i > n ]
= am bn ≠ 0 [Qam ≠ 0, bn ≠ 0 ∴am bn ≠ 0]
∴ f (x ) ⋅ g (x ) is not a zero polynomial.
+n +1
(ii) Coefficient of x m in f (x ) ⋅ g (x )
= a 0bm +n +1 + a1bm +n + K + am bn +1 + am + 1bn + K + am +n +1 b0
= a 0 ⋅ 0 + a1 ⋅ 0 + K + am ⋅ 0 + 0 ⋅ bn + K + 0 ⋅ b0 [Qbi = 0 ∀ i > n and ai = 0 ∀ i > m ]
=0
+n +1
∴ Coefficient of x m in f (x ) ⋅ g (x ) = 0
m +n + 2
Similarly, coefficient of x in f (x ) ⋅ g (x ) = 0 and so on
m +n
Also coefficient of x in f (x ) ⋅ g (x ) ≠ 0
∴ deg { f (x ) ⋅ g (x )} = m + n
= deg f (x ) + deg g (x )
Division Algorithm
If f (x )and g (x )are two non-zero polynomials, then there exist unique polynomialsq (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.
Root of an Equation
A number ‘ α ’ is called a root of equation
f (x ) = 0, iff f (α ) = 0
α is a root of f (x ) = 0
⇒ f (α ) = 0
Conversely f (α ) = 0
⇒ α is a root of f (x ) = 0
Theorem 3 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 )
www.pdfworld.in
www.pdfworld.in
Theorem 4 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
Qh 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 5 Fundamental Theorem of Algebra : Every polynomial function of degree ≥ 1 has atleast
one zero in the complex number.
n −1
Proof Let f (x ) = an x n + an − 1x + K + a1x + a 0 with n ≥ 1, then there exists atleast one h ∈ C such
n −1
that an hn + an − 1h + K + a1h + a 0 = 0
from this it is easy to deduce that a polynomial function of degree ‘n’ has exactly n zeros.
Theorem 6 Every equation of the nth degree has n roots and no more.
Proof Let the given equation be
−1 −2
f (x ) = a 0x n + a1x n + a 2x n + K + an = 0
Since, this equation must have a root, real or imaginary (by fundamental theorem of algebra).
Let us denote it by α1 so that x − α1 is a factor of f (x ).
Thus, f (x ) = (x − α1 ) φ 1(x ) …(i)
where φ1 (x ) is a polynomial of degree (n − 1).
Again, the equation φ 1(x ) = 0 must also have a root. Let us denote it by α 2 so that x − α 2 is a factor of φ1 (x ).
Thus, φ 1(x ) = (x − α 2 ) φ 2(x ) …(ii)
where φ 2(x ) is a polynomial of degree (n − 2).
Putting this value of φ 1(x ) in Eq. (i), we get
f (x ) = (x − α1 )(x − α 2 ) φ 2 (x )
Proceeding in this way, we shall obtain n factors of f (x ).
∴ f (x ) = (x − α1 )(x − α 2 )K (x − αn ) φ n (x )
φ n (x ) is a polynomial of degree (n − n ) i . e. , 0 so that φ n (x ) is a constant polynomial.
Let φ n (x ) = K
−1 −2
Thus, a 0x n + a1x n + a 2x n + K + an
= K (x − α1 )(x − α 2 ) K (x − αn )
Comparing coefficients of x n on both sides
a0 = K
∴ f (x ) = a 0 (x − α1 )(x − α 2 ) K (x − αn )
www.pdfworld.in
www.pdfworld.in
so that the given equation may be written as

f (x ) = a 0 (x − α1 )(x − α 2 ) K (x − an ) = 0 …(iii)
This is satisfied by n values of x viz., α1 , α 2 ,…, αn .
Hence, equation f (x ) = 0 has n roots.
If possible let x = β be any other root distinct from α1 , α 2 , K , αn .
Then, f ( β ) = a 0 ( β − α1 )( β − α 2 ) K ( β − αn )
Since, a 0 ≠ 0 and β is distinct from α1 , α 2 , K , αn
∴ β − α1 ≠ 0, β − α 2 ≠ 0, K , β − αn ≠ 0
∴ f ( β ) can never vanish.
∴ β is not a root of f (x ) = 0
Hence, f (x ) = 0 has exactly n roots.
Theorem 7 In an equation with real coefficients, non-real complex roots occur in conjugate pairs.
f (x ) = 0 …(i)
Let α + i β (where β ≠ 0) be a root of f (x ) = 0,
then f (α + i β ) = 0 …(ii)
We have to prove that α − iβ is also a root of (1)
Divide f (x ) by [x − (α + i β )][x − (α − i β )]
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 ′.
∴ f (x ) = [x − (α + i β )][x − (α − i β )] Q (x ) + Rx + R ′ …(iii)
Putting x = α + iβ,
f (α + i β ) = R (α + i β ) + R ′
But f (α + i β ) = 0 [ QEq. (ii )]
∴ R (α + i β ) + R ′ = 0 or (Rα + R ′ ) + iR β = 0
Equating the real and imaginary parts on both sides.
Rα + R ′ = 0 …(iv)
Rβ = 0 …(v)
from Eq. (v), either R = 0 or β = 0
Now, β≠0
[Qin case β = 0 even the complex root α + i β becomes real]
∴ R=0
Putting this in Eq. (iv), we have R′ = 0
∴ from Eq. (iii)
f (x ) = [x − (α + i β )][x − (α − i β )] Q (x )
i . e. , [x − (α + i β )][x − (α − i β )]
divides f (x ) exactly.
Hence, α − i β is also a root of the given equation f (x ) = 0.
Corollary Every equation of an odd degree having real coefficients, has atleast one real root, because
complex roots occur in pairs.
www.pdfworld.in
www.pdfworld.in
Theorem 8 In an equation with rational coefficients irrational roots occur in conjugate pairs.
f (x ) = 0 …(i)
Let α + β be a root of Eq. (i) [where α and β are rational, β is +ve but not a perfect square]
∴ f (α + β) = 0 …(ii)
We have to prove that α − β is also a root of Eq. (i)
Divide
f (x ) by [x − (α + β )] [x − (α − β )]
Let Q (x ) be the quotient and if there is a remainder, since it must be a linear polynomial inx , take it
as Rx + R ′
∴ f (x ) = [x − (α + β )][x − (α − β )] Q (x ) + Rx + R ′ …(iii)
Putting x =α + β
f (α + β ) = R (α + β ) + R′
But f (α + β) = 0 [Q Eq. (ii )]
∴ R (α + β ) + R′ = 0
i . e. , (Rα + R ′ ) + R β = 0
Now, rational and irrational number cannot destroy one another.
∴ If there sum vanishes, each must vanish separately.
⇒ Rα + R ′ = 0 …(iv)
R β =0 …(v)
Since, β ≠ 0 , ∴ from Eq. (v), R = 0
Putting R = 0 in Eq. (iv), R′ = 0
∴ Eq. (iii) becomes
f (x ) = [x − (α + β )][x − (α − β )] Q (x )
This shows that
[x − (α + β )][x − (α − β )] is a factor of f (x ).
Hence, α − β is also root of f (x ).
p
Theorem 9 If the rational number ,q ≠ 0, (p , q ) = 1 (i . e. , p and q are relatively prime) is a root of the
q
n −1
equation an x n + an − 1x + K + a1x + a 0 = 0, where a 0 , a1 , a 2 , K , an are integers and an ≠ 0, then p is a
divisor of a 0 and q that of an .
Proof Since, p is a root of given equation, so we have
q
n n −1
p  p  p
an   + an − 1  + K + a1 + a0 = 0
q  q  q
n −1 −1
⇒ an pn + an − 1qp + K + a1qn p + a 0qn = 0 …(i)
n
n −1 n −2 −2 −1 an p
⇒ an − 1p + an − 2p q + K + a1qn p + a 0qn =− …(ii)
q
Since, the coefficients an − 1 , an − 2 , K , a 0 and p , q are all integers and hence the left hand side is an
integer, so that right hand side is also an integer but p and q are relatively prime to each other.
∴ q should divide an .
www.pdfworld.in
www.pdfworld.in
Again
n −1 −1
an pn + an − 1qp + K + a1qn p = − a 0qn
−1 n −2 −1
⇒ an pn + an − 1qp + K + a1qn
This shows that P should divide a 0 since (p , q ) = 1
a qn
=− 0 …(iii)
p
As a consequence of the above theorem we have the following corollary.
Corollary : Every rational root of the equation x n + a1x n − 1 + K + an = 0 where each ai (i = 1, 2, K , n ) is an
integer, must be an integer. More over every such root must be a divisor of the constant an .
Theorem 10 If f (x ) be a polynomial of degree n and α is any complex number, show that there exist
unique numbers b0 , b1 , b2 , K , bn such that
f (x ) = b0 + b1 (x − α ) + b2 (x − α )2 + K + bn (x − α )n (bn ≠ 0)
Proof Let us prove by induction

if n=0
∴ deg f (x ) = 0
⇒ f (x ) is a non-zero constant polynomial, let
f (x ) = b0 ⇒ f (x ) = b0 (x − α )0
∴ Result is true for n = 0
Assume that result is true for n = k − 1 (k > 1) i.e., f (x ) = b0 + b1 (x − α ) + b2 (x − α )2 + ... + bk − 1 (x − α )k − 1
Now, we shall show that result is true for n = k, i.e., to prove that
f (x ) = b0 + b1 (x − α ) + b2 (x − α )2 + K + bk (x − α )k …(i)
Let degree of f (x ) be k.
∴By Euclidean algorithm there exist a polynomialg (x )of degree k − 1 and unique remainderb0 such that
f (x ) = (x − α ) g (x ) + b0 …(ii)
where g (x ) is a polynomial of degree k − 1 [using Eq. (i)]
f (x ) = (x − α )[b1 + b2 (x − α ) + b3 (x − α )2 + K + bk (x − α )k − 1 ] + b0
f (x ) = b0 + b1 (x − α ) + b2 (x − α )2 + K + bk (x − α )k
∴ Result is true for n = k
∴ By induction result is true.
Common Divisor
Let f (x ), g (x ) be two polynomials over any number system F such that atleast one of them is non-zero.
A non-zero polynomial h (x ) is said to be a common divisor of f (x ), g (x ), if h (x ) | f (x ) and h (x )| g (x ).
Theorem 11 Let f (x ), g (x ) be any two non-zero polynomials over k . Then their gcd exists and is
unique. Further, if
d (x ) = ( f (x ), g (x ))
then d (x ) = a (x ) f (x ) + b (x ) g (x )
for some polynomials a (x ) and b (x ) over k.
Concept Two non-zero polynomials f (x ), g (x ) over k are said to be relatively prime or coprime, if
( f (x ), g (x )) = 1
or ( f (x ), g (x )) = c
where c ≠ 0 over k.
www.pdfworld.in
www.pdfworld.in
Theorem 12 Two non-zero polynomialsf (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
Proof Let f (x ), g (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
We 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 ) and 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 Let f (x ) be a non-constant polynomial. Then, a complex number α is called a root of
multiplicity m > 0 of the polynomial equation f (x ) = 0, if there exist a polynomial g (x ) such that
f (x ) = (x − α )m g (x ) and g(α ) ≠ 0
Note 1. If m = 0, then α is not a root of f ( x ) = 0. If m = 1, then α is called a simple root of f ( x ) = 0 α is

called a repeated root of f ( x ) = 0, iff m > 1.
2. Let deg f ( x ) = n, where n ≥ 1. Let α be a root of multiplicity m > 0. Then, there exist a polynomial
g ( x ) such that
f ( x ) = ( x − α )m g ( x ),
where g(α ) ≠ 0
∴ deg f ( x ) = deg ( x − α )m + deg g ( x )
n = m + deg g ( x )
∴ deg g ( x ) = n − m
Since, degree of a polynomial is always a non-negative integer.
∴ n −m ≥ 0
∴ n ≥ m or m ≤ n
Hence, if α is a root of multiplicity m of the equation f ( x ) = 0 of degree n,
then m ≤n
Theorem 13 A complex number α is a repeated root of a polynomial equation f (x ) = 0, if and only if it

is a root of both f (x ) = 0 and f ′ (x ) = 0, where f ′ (x ) denotes the derivative of f (x ).
Proof Let α be a repeated root of f (x ) = 0
∴ (x − α )m | f (x ), where m ≥ 2
∴ f (x ) = (x − α ) g (x )
m
for some polynomial g (x )

Differentiating both sides, we get
www.pdfworld.in
www.pdfworld.in
−1
f ′ (x ) = (x − α ) m g ′ (x ) + m (x − α ) m g (x )
Putting x = α on both sides, we get f ′ (α ) = 0
∴ α is a root of f (x ) = 0 as well as f ′ (x ) = 0
Conversely let α be a root of f (x ) = 0 as well as f ′ (x ) = 0
∴ f (α ) = 0 and f ′ (α ) = 0
Since, α is a root of f (x ) = 0
∴ (x − α )| f (x )
f (x ) = (x − α ) g (x ) …(i)
For some polynomial g (x )
Differentiating both sides, we get
f ′ (x ) = (x − α ) g ′ (x ) + g (x ) ⋅ 1
Putting x = α on both sides
f ′ (α ) = g (α ) ⇒ g(α ) = 0 [Qf ′ (α ) = 0]
⇒ (x − a )| g (x )
⇒ g (x ) = (x − α ) h (x ) …(ii)
For some polynomial h (x )
From Eqs. (i) and (ii), we get
f (x ) = (x − α )2h (x )
which shows that α is a repeated root of f (x ) = 0. Hence, the result is true.
Corollary (1) α is a repeated roots of f (x ) = 0 iff α is a common root of f (x ) = 0 and f ′ (x ) = 0

(2) A complex number α is a repeated root of the polynomial equation f (x ) = 0, iff α is a root of d (x ) = 0,
where d (x ) = (f (x ), f ′ (x ))
Proof Let α be a repeated roots of f (x ) = 0
∴ f (x ) = (x − α )2g (x ), where g(α ) ≠ 0
f ′ (x ) = (x − α )2g ′ (x ) + 2(x − α ) g (x )
= (x − α )[(x − α ) g ′ (x ) + 2g (x )] = (x − α ) G(x )
where G(α ) = 0 + 2g (α ) = 2g (α ) ≠ 0
∴ α is a root of f ′ (x ) = 0
∴ x − α | f (x ) and (x − α )| f ′ (x )
x − α | (f (x ), f ′ (x ))
⇒ x − α | d (x )
⇒ α is a root of d (x ) = 0
Conversely
Let α be a root of d (x ) = 0
∴ x − α | d (x )
But d (x )| ( f (x ), f ′ (x )) ⇒ d (x )| f (x ), d (x )| f ′ (x )
∴ x − α | f (x ) and x − α | f ′ (x )
∴ α is a root of f (x ) = 0 as well as of f ′ (x ) = 0
∴ α is a repeated root of f (x ) = 0
www.pdfworld.in
www.pdfworld.in
Note In order to test whether a given polynomial equation f ( x ) = 0 has repeated roots we find gcd of
f ( x ), f ′ ( x ).
Let d ( x ) = (f ( x ), f ′ ( x )) . If d ( x ) = 1, then d ( x ) is a constant polynomial.
∴ It has no roots.
∴ f ( x ) = 0 has no repeated roots.
If d ( x ) is a non-constant polynomial, then the roots of d ( x ) = 0 are repeated roots of
f ( x ) = 0.
Theorem 14 Prove that α is a root of multiplicity m of the polynomial equation f (x ) = 0, iff

+1
(i) (x − α )m | f (x ) and (ii) (x − α )m | f (x )
Proof α is a root of f (x ) = 0 of multiplicity m, iff there exist a polynomial g (x ) such that
f (x ) = (x − α )m g (x ),
where g(α ) ≠ 0 …(i)
∴ (x − α ) m | f (x )
Next we have to prove that
+1
(x − α ) m \| f (x )
Let if possible
+1
(x − α ) m | f (x )
Then, ∃ a polynomial h (x ) such that
+1
f (x ) = (x − α ) m h (x ) …(ii)
+1
(x − α ) m g (x ) = (x − α ) m h (x )
⇒ (x − α ) [g (x ) − (x − α )h (x )] = 0
m
Since, (x − α )m is not a zero polynomial.

∴ g (x ) − (x − α )h (x ) must be a zero polynomial.
∴ g (x ) − (x − α ) h (x ) = 0
∴ g (x ) = (x − α ) h (x )
g(α ) = 0
which is a contradiction.
∴ Our supposition is wrong. [Qg (α ) ≠ 0]
+1
∴ (x − α )m \| f (x )
Example 1 Prove that the sum of two constant polynomials is a constant polynomials.
Solution Let f ( x ) = a 0 + a1x + a 2x 2 + K + ai x i + K
g ( x ) = b0 + b1x + b2x 2 + K + bi x i + K
be two constant polynomials.
∴ ai = 0 for i > 0 and bi = 0 for i > 0
∴ f ( x ) = a 0 and g ( x ) = b0
∴ f ( x ) + g ( x ) = a 0 + b0 = c 0
where c 0 = a 0 + b0 = a constant
∴ f ( x ) + g ( x ) is a constant polynomial.
[coefficient of x, coefficient of x 2,K are all zero]
www.pdfworld.in
www.pdfworld.in
Example 2 Let f ( x ) = a 0 + a1x + a 2x 2 + K + an x n , find a polynomial g ( x ) such that

f ( x ) + g( x ) = 0
Also, show that this polynomial g ( x ) is unique.
Solution f ( x ) = a 0 + a1x + a 2x 2 + K + an x n
∴ ai = 0, ∀ i > n
Let g ( x ) = b0 + b1x + b2x 2 + K + bi x i + K be a polynomial such that
f ( x ) + g ( x ) = 0 = zero polynomial
∴Coefficient of x i in [f ( x ) + g ( x )] = 0
[QAll coefficients are zero in zero polynomial]
∴ ai + bi = 0 bi = − ai ∀i ≠ 0
bi = − ai for i = 0, 1, 2, 3, K , n
bi = 0, ∀ i > n
∴ g ( x ) = − a 0 − a1x − a 2x 2 − K − an x n
For uniqueness,
Let g ( x ) and h( x ) be two polynomials such that
f ( x ) + g ( x ) = 0 and g ( x ) + h( x ) = 0
f ( x ) + g ( x ) = f ( x ) + h( x )
g ( x ) = h( x )
Hence, g ( x ) is unique.
Example 3 Let f ( x ) = x 5 + x 4 + 1. Find a polynomial g ( x ) such that the degree of f ( x ) + g ( x ) is

zero. Is it possible to find more than one such polynomial g ( x )? If so how many.
Solution f (x ) = x 5 + x 4 + 1
Let g ( x ) = b0 + b1x + b2x 2 + K + bi x i + K be a polynomial such that degree of
{f ( x ) + g ( x )} = 0
∴ Constant term of f ( x ) + g ( x ) ≠ 0
and coefficient of x i in [f ( x ) + g ( x )] = 0, ∀ i ≥ 1
∴ 1 + b0 ≠ 0,i . e.,b0 ≠ − 1
and b2 = 0, b3 = 0,b4 = − 1, b5 = − 1 b1 = 0,
and bi = 0, ∀ i > 5
Hence, g ( x ) = b0 − x 4 − x 5,
where b0 ≠ − 1
Since, we can find infinite number of b0 ≠ − 1 in the given number system.
Hence, there are infinitely many such polynomials g ( x ).
Example 4 Prove that if degree of f ( x ) ⋅ g ( x ) is 1, then one of f ( x ), g ( x ) is a constant and the other
is a linear polynomial.
Solution Since, deg {f ( x ) ⋅ g ( x )} = 1
∴ deg f ( x ) + deg g ( x ) = 1
∴ Either deg f ( x ) = 1 and deg g ( x ) = 0
www.pdfworld.in
www.pdfworld.in
or deg f ( x ) = 0 and deg g ( x ) = 1

∴ Either f ( x ) is a linear polynomial and g ( x ) is a constant polynomial or f ( x ) is a
constant polynomial and g ( x ) is a linear polynomial.
Example 5 If ( x − h )k divides a polynomial f ( x ). Prove that ( x − h )k − 1 divides f ′ ( x ).

Solution Q ( x − h )k divides f ( x )
⇒ There exist a polynomial q ( x ) such that
f ( x ) = ( x − h )k q ( x )
−1
∴ f ′ ( x ) = ( x − h )k q ′ ( x ) + q ( x )k ( x − h )k
k −1
= (x − h ) [( x − h )q ′ ( x ) + kq ( x )]
k −1
= (x − h ) r(x )
[where r ( x ) = ( x − h )q ′ ( x ) + kq ( x )]
k −1
⇒ (x − h ) divides f ′ ( x )
Example 6 Find the GCD of f ( x ) = x 7 − 1 and g ( x ) = x 5 − 1 and express it in the form

a ( x ) f ( x ) + b ( x ) g ( x ), where a( x ) and b ( x ) are polynomials with integral coefficients.
Solution First we find GCD of f ( x ) and g ( x )
x2
x5 − 1 x7 − 1
x7 − x2
x 2 − 1) x 5 − 1 (x 3 + x
x5 − x3
x3 − 1
x3 − x
x − 1) x 2 − 1 ( x + 1
x2 − x
x −1
x −1
×
∴ (f (x), g (x)) = (x − 1) = d (x)
The above division can be written as
f ( x ) = x 2g ( x ) + x 2 − 1
⇒ x 2 − 1 = f ( x ) − x 2g ( x ) …(i)
and g ( x ) = ( x 2 − 1)( x 3 + x ) + x − 1 …(ii)
⇒ x − 1 = g ( x ) − ( x 2 − 1)( x 3 + x )
= g ( x ) − (f ( x ) − x 2g ( x ))( x 3 + x ) [using Eq. (i)]
∴ d ( x ) = g ( x ) − ( x 3 + x ) f ( x ) + x 2( x 3 + x ) g ( x )
= − ( x 3 + x ) f ( x ) + ( x 5 + x 3 + 1) g ( x )
Hence, d ( x ) = a( x ) f ( x ) + b ( x ) g ( x ),
where a( x ) = − ( x 3 + x ), b ( x ) = x 5 + x 3 + 1
www.pdfworld.in
www.pdfworld.in
Example 7 Find c if GCD of f ( x ) = x 3 + cx 2− x + 2c and g ( x ) = x 2 + cx − 2 is a linear polynomial.

x
Solution x 2 + cx − 2 x 3 + cx 2 − x + 2c
x 3 + cx 2 − 2x
x + 2c ) x 2 + cx − 2 ( x − c
x 2 + 2cx
− cx − 2
− cx − 2c 2
2c 2 − 2
GCD of f ( x ) and g ( x ) is a linear polynomial, if g ( x ) is exactly divisible by x + 2c i . e., if
remainder
2c 2 − 2 = 0
⇒ c2 =1
⇒ c = ±1
Example 8 If f ( x ), g ( x ) and h( x ) are three polynomials such that (f ( x ), g ( x )) = 1 and

f ( x )| g ( x ) h( x ),prove that f ( x )| h( x ). Is the result true, if (f ( x ), g ( x )) ≠ 1? Give reasons.
Solution Since, f ( x )| g ( x ) h( x )
∴ There exist a polynomial q ( x ) such that
g ( x ) h( x ) = f ( x ) q ( x )
Again, Q (f ( x ), g ( x )) = 1
∴ ∃ two polynomials a( x ), b( x ) such that
a( x ) f ( x ) + b ( x ) g ( x ) = 1
Multiplying both sides by h( x ), we get
a( x ) f ( x ) h( x ) + b ( x ) g ( x ) h( x ) = h( x )
or a( x ) f ( x ) h( x ) + b ( x ) f ( x ) q ( x ) = h( x )
or f ( x )[a( x ) h( x ) + b ( x ) q ( x )] = h( x )
or f ( x ) Q ( x ) = h( x )
where Q ( x ) = a( x ) h( x ) + b ( x ) q ( x )
This shows that f ( x ) | h( x )
If (f ( x ), g ( x )) ≠ 1. The above result may or may not be true.
For example, take f ( x ) = x 2 − 1
g( x ) = x + 1
h( x ) = x − 1
GCD of f ( x ), g ( x ) = x + 1
g (x )h (x ) = x 2 − 1
Clearly, f ( x ) | g ( x ) h( x )
But f ( x ) does not divide h( x ).
www.pdfworld.in
www.pdfworld.in
Example 9 If f ( x ), g ( x ) are two polynomials (not both zero). Show that

(f ( x ), g ( x )) = (f ( x ) + g ( x ), f ( x ) − g ( x ))
Solution Let (f ( x ), g ( x )) = d ( x )
∴ d ( x ) | f ( x ) and d ( x ) | g ( x ), where d ( x ) is a monic polynomial.
Again d ( x )| f ( x ) and d ( x ) | g ( x )
∴ ∃ two polynomials a( x ) and b( x ) such that
f ( x ) = a( x ) d ( x ) …(i)
g ( x ) = b( x ) d ( x ) …(ii)
Adding Eqs. (i) and (ii), we have
f ( x ) + g ( x ) = [a( x ) + b( x )]d ( x )
i . e., d( x ) | f ( x ) + g( x ) [Qa( x ) + b( x ) is a polynomial]
Similarly, d ( x )| f ( x ) − g ( x )
∴ d( x ) | f ( x ) + g( x )
and d ( x )| f ( x ) − g ( x ) …(iii)
Let d1( x ) be any other common divisor of f ( x ) + g ( x ) and f ( x ) − g ( x )
∴ d1( x )| f ( x ) + g ( x )
∴ ∃ a polynomial p( x ) such that
f ( x ) + g ( x ) = p( x )d1( x ) …(iv)
Again Q d1( x )| f ( x ) − g ( x )
∴ ∃ a polynomial q ( x ) such that
f ( x ) − g ( x ) = q ( x )d1( x ) …(v)
Adding Eqs. (iv) and (v), we have
2f ( x ) = d1( x )[ p ( x ) + q ( x )]
p ( x ) + q( x )
or f ( x ) = d1( x )
2
 p ( x ) + q( x ) 
∴ d1( x )| f ( x ) Q is a polynomial
 2 
Subtracting Eq. (v) from Eq. (iv)
2g ( x ) = d1( x )[ p ( x ) − q ( x )]
p ( x ) − q( x )
g ( x ) = d1( x )
2
∴ d1( x )| g ( x )
Now, d1( x )| f ( x ) and d1( x )| g ( x )
∴ d1( x ) divides the GCD of f ( x ) and g ( x )
i . e., d1( x ) | d ( x ) ...(vi)
Also, f ( x ) + g ( x ) and f ( x ) − g ( x ) are not both zero.
∴ f ( x ) and g ( x ) are not both zero.
∴ (f ( x ) + g ( x ), f ( x ) − g ( x )) exists.
From Eqs. (iii) and (vi), we have
(f ( x ) + g ( x ), f ( x ) − g ( x )) = (f ( x ), g ( x ))
www.pdfworld.in
www.pdfworld.in
Example 10 Show that the polynomial f ( x ) = x 3+ px + q has a repeated zero, if

4p 3 + 27q 2 = 0
Solution f ( x ) = x + px + q and f ′ ( x ) = 3x 2 + p
3
zeros of f ′ ( x ) are given by

−p −p
3x 2 + p = 0 ⇒ x2 = ⇒ x =±
3 3
−p  −p 
If is a zero of f ( x ), then f   =0
3  3 
3
 −p   −p 
⇒  3  + p 3  + q = 0
   
3/ 2 1/ 2
 −p   −p 
or   + p  = −q
 3  3
Squaring both sides
3 3/ 2 1/ 2
 − p 2  − p  − p  − p
  +p   + 2p   .  = q2
 3   3   3   3 
p3 p3 p3
⇒ − − +2 = q2
27 3 9
⇒ − p 3 − 9p 3 + 6 p 3 = 27q 2
⇒ 4p 3 + 27q 2 = 0
which is the required condition.
Example 11 If 1 is a twice repeated root of the equation ax 3 + bx 2 + bx + d = 0. Show that

a = d = −b
Solution The given equation is f ( x ) = 0,
where f ( x ) = ax 3 + bx 2 + bx + d
Since, 1 is a twice repeated root of f ( x ) = 0
∴ 1 is a simple root of
f ′ ( x ) = 3ax 2 + 2bx + b = 0
⇒ f (1) = 0 and f ′ (1) = 0
⇒ a(1) + b (1)2 + b (1) + d = 0
3
i . e., a + 2b + d = 0 …(i)
and 3a(1) + 2b (1) + b = 0
2
i . e., 3a + 3b = 0
⇒ 3a = − 3b
⇒ a = −b …(ii)
From Eq. (i), we get
−b + 2b + d = 0 ⇒ b + d = 0
d = −b …(iii)
From Eqs. (ii) and (iii), we have a = d = − b
www.pdfworld.in
www.pdfworld.in
Example 12 Solve the equation

x 4 + 4x 3 + 6x 2 + 4x + 5 = 0.
Given that its one root is i .
Solution One root of
x 4 + 4x 3 + 6x 2 + 4x + 5 = 0 …(i)
is i , the other root is − i [coefficient of Eq. (i) are real]
∴ ( x − i )( x + i ) is a factor of
x 4 + 4x 3 + 6x 2 + 4x + 5
∴ x 2 + 1 is a factor of x 4 + 4x 3 + 6x 2 + 4x + 5
So, x 4 + 4x 3 + 6x 2 + 4x + 5
= ( x 2 + 1)( x 2 + 4x + 5)
∴ Other two roots are given by
x 2 + 4x + 5 = 0
−4 ± 16 − 20 −4 ± −4
i . e., x = =
2 2
−4 ± 2i
= = −2 ± i
2
Hence, all roots are ± i , − 2 ± i .
Example 13 Construct a real polynomial equation of degree 4 having 2 + 3i as a root with

multiplicity 2.
Solution Since, 2 + 3i is a root of multiplicity 2 of a real polynomial equation.
∴ 2 − 3i is also a root of multiplicity 2.
∴ The required equation is
( x − 2 + 3i )2 ( x − 2 − 3i )2 = 0
⇒ [( x − 2)2 − 9i 2 ] = 0
or ( x 2 − 4 x + 4 + 9 )2 = 0
or ( x 2 − 4x + 13)2 = 0
i . e., x 4 + 16x 2 + 169 − 8x 3 + 26x 2 − 104x = 0
i . e., x 4 − 8x 3 + 42x 2 − 104x + 169 = 0
Example 14 Find the rational roots of 2x 3 − 3x 2 − 11x + 6 = 0

p
Solution Let the roots be of the form , where ( p, q ) = 1 and q > 0
q
Since, q | 2, q must be 1 or 2 and p | 6.
⇒ p = ± 1, ± 2, ± 3, ± 6
By remainder theorem
 1  −2   3
f  =f  =f  = 0
 2  1  1
1
So, the 3 roots of equations are , − 2 and 3.
2
www.pdfworld.in
www.pdfworld.in
Example 15 (a) Form an equation of the lowest degree with real and rational coefficients, one
of its roots being 2 + 3 + i.
(b) Find an equation with rational coefficients which shall have for the roots the
values of the expression θ1 p + θ 2 q + θ 3 r , where θ12 = 1,θ 22 = θ 23 = 1.
(c) Use the fact that a cubic equation has atleast one real root to find the real value
of [ 27 + 756 ]1/ 3 + [ 27 − 756 ]1/ 3.
Solution (a) x = 2 + 3 + i,
∴ x −i = 2 + 3
On squaring both sides, we get
x 2 + i 2 − 2xi = 5 + 2 6
x 2 − 6 = 2xi + 2 6
On again squaring both sides, we get
x 4 − 12x 2 + 36 = − 4x 2 + 24 + 8xi 6
x 4 − 8x 2 + 12 = 8xi 6
( x 4 − 8x 2 )2 + 144 + 24( x 4 − 8x 2 ) = − 64x 2( 6)
or x 8 − 16x 6 + 88x 4 + 192x 2 + 144 = 0
(b) Let us suppose that x = θ1 p + θ 2 q + θ 3 r square and put θ12 = θ 22 = θ 23 = 1
∴ x 2 − ( p + q + r ) = 2θ1 θ 2 pq + 2θ 2 θ 3 qr + 2θ 3 θ1 rp
squaring again and putting
θ12 = θ 22 = θ 23 = 1, we get
x 4 − 2 x 2 ( p + q + r ) + ( p + q + r )2
= 4( pq + qr + rp ) + 8θ1 θ 2 θ 3 pqr [θ1 p + θ 2 q + θ 3 r ]
[ x 4 − 2x 2( p + q + r ) + p 2 + q 2 + r 2 − 2pq − 2qr − 2rp]2
= 64pqrx 2
as θ1 p + θ 2 q + θ 3 r = x and θ12 = θ 22 = θ 23 = 1
(c) Let x = a 1/ 3
+b 1/ 3
cube both sides
x = a + b + 3a1/ 3b1/ 3(a1/ 3 + b1/ 3 )
3
x 3 = 54 + 3[ 272 − (7562 )]1/ 3 x

x 3 = 54 + 3( −27)1/ 3 x
Now, real value of root of ( −27)1/ 3 is – 3
∴ x 3 = 54 + 3( −3)x
x 3 + 9x − 54 = 0
Evidently x − 3 satisfies. It showing there by the real value of x is 3, which is
therefore the required value of the expression.
Example 16 If the roots of the equation

xn − 1 = 0
are 1, α , β, γ, K , show that
(1 − α )(1 − β )(1 − γ )K = n
Solution Q 1, α , β, γ,... are roots of x n − 1 = 0
∴ x n − 1 = ( x − 1)( x − α )( x − β )( x − γ ) …(i)
www.pdfworld.in
www.pdfworld.in
Dividing both sides by ( x − 1)

xn − 1
= ( x − α )( x − β )( x − γ )
x −1
or x n − 1 + x n − 2 + K + x 2 + x + 1 = ( x − α )( x − β )( x − γ )K
Putting x = 1,
1 + 1 + K n times = (1 − α )(1 − β )(1 − γ ) …
Hence, (1 − α )(1 − β )(1 − γ ) K = n
Example 17 Show that x 4 + qx 2 + s = 0 cannot have three equal roots.

Solution Let f ( x ) = x 4 + qx 2 + s
f ′ ( x ) = 4x 3 + 2qx
Now, f ( x ) = 0 has three equal roots, if
f ′ ( x ) = 0 has two equal roots.
But f ′ ( x ) = 0 gives 2x ( 2x 2 + q ) = 0 or x = 0, ± −q / 2
Thus, no two roots of f ′ ( x ) = 0 are equal.
Hence, the given equation cannot have three equal roots.
Example 18 If the equation x 4 + ax 3 + bx 2 + cx + d = 0 has three equal roots. Show that each of
6c − ab
them is equal to .
3a 2 − 8b
Solution f ( x ) = x 4 + ax 3 + bx 2 + cx + d , then
f ′ ( x ) = 4x 3 + 3ax 2 + 2bx + c …(i)
f ′ ′ ( x ) = 12x 2 + 6ax + 2b …(ii)
If f ( x ) = 0 has three equal roots, then
f ′ ( x ) = 0 and f ′ ′ ( x ) = 0 must have a common root
Multiply Eq. (i) by 3 and Eq. (ii) by x and subtracting
3ax 2 + 4bx + 3c = 0 …(iii)
Multiplying Eq. (ii) by a and Eq. (iii) by 4 and subtracting
( 6a 2 − 16b )x + ( 2ab − 12c ) = 0
6c − ab
x =
3a 2 − 8b
is the common root of f ′ ( x ) = 0 and f ′ ′ ( x ) = 0
6c − ab
Hence, required triple root of f ( x ) = 0 is .
3a 2 − 8b
Concept Relation between roots and coefficients
1. If α , β be the roots of the equation ax 2 + bx + c = 0,
then σ1 ≡ α + β = − b / a and σ 2 ≡ αβ = c / a
2. If α , β , γ be the roots of the equation
ax 3 + bx 2 + cx + d = 0,
then σ1 ≡ α + β + γ = − b / a
σ 2 ≡ αβ + βγ + γα = c / a
and σ3 ≡ αβγ = − d / a
www.pdfworld.in
www.pdfworld.in
Remark
l It is convenient to write σ 2 as α (β + γ ) + βγ.
If α , β, γ and δ be the roots of the equation
ax 4 + bx 3 + cx 2 + dx + e = 0,
then σ1 ≡ α + β + γ + δ = − b /a
σ 2 ≡ αβ + αγ + αδ + βγ + βδ + γδ = c / a
σ 3 ≡ αβγ + αβδ + αγδ + βγδ = − d / a
and σ 4 ≡ αβγδ = e / a
l It is convenient to write
σ 2 ≡ (α + β )( γ + δ ) + αβ + γδ
σ 3 ≡ αβ( γ + δ ) + γδ(α + β )
Example 1 Find the condition which must be satisfied by the coefficients of the equation
x 3 − px 2 + qx − r = 0 when two of its roots α , β are connected by the relation
α+β=0
Solution Let the roots of the equation be α , β, γ.
∴ Σα = α + β + γ = p
But α+β=0
∴ γ=p
Now, γ is a root of the given equation and as such γ 3 − pγ 2 + qγ − γ = 0
Put the value of γ
∴ p 3 − p ⋅ p 2 + qp − r = 0 or pq = r is the required condition.
Example 2 If α, β, γ are the roots of the equation (a − x )(b − x ) (c − x ) + 1 = 0, then prove that
a, b, c will be the roots of the equation ( x − α )( x − β )( x − γ ) + 1 = 0
Solution Q α , β, γ are the roots of
(a − x )(b − x )(c − x ) + 1 = 0
∴(a − x )(b − x )(c − x ) + 1 = λ ( x − α )( x − β )( x − γ )
Equating the leading coefficients, i . e., coefficients of x 3 on both sides
− 1 = λ or λ = −1
∴ (a − x )(b − x )(c − x ) + 1 = − ( x − α )( x − β )( x − γ )
⇒ − ( x − a )( x − b )( x − c ) + 1 = − ( x − α )( x − β )( x − γ )
⇒ ( x − a )( x − b )( x − c ) − 1
= ( x − α )( x − β )( x − γ )
⇒ ( x − α )( x − β )( x − γ ) + 1
= ( x − a )( x − b )( x − c )
∴ ( x − α )( x − β )( x − γ ) + 1 = 0 …(i)
⇒ ( x − a )( x − b )( x − c ) = 0
Hence, the roots of Eq. (i) are a, b, c.
Example 3 Find the condition that the roots of the equation x 5 + px 4 + qx 3 + rx 2 + sx + t = 0 be

in AP.
Solution Let the roots be a − 2d , a − d , a, a + d , a + 2d
Sum of roots = 5a = − p, ∴ a = − p / 5
www.pdfworld.in
www.pdfworld.in
But a is a root of the given equation which will satisfy it. Hence, the required
condition is
5 4 3 2
 −p   −p   −p   −p   −p 
  + p  +q   +r   +s  +t =0
 5  5  5  5  5
or 4p 5 − 25qp 3 + 125rp 2 − 625 sp + 3125t = 0
Example 4 Find the condition that the roots of the equation x 3 − px 2 + qx − r = 0 may be in AP
and hence solve the equation
x 3 − 12x 2 + 39x − 28 = 0.
Solution Let the roots of the given equation be a − d , a, a + d being in AP.
∴ Σα = a − d + a + a + d = p
p
or 3a = p or a =
3
But a is a root of the given equation and as such it will satisfy. The given equation
3 2
 p  p  p
∴   − p   + q   −r = 0
 3  3  3
or 2p 3 − 9pq + 27r = 0 is the required condition. Proceeding as above we find that
a = 4 is a root of the given equation which when divided by x − 4 gives quadratic
equation ( x 2 − 8x + 7) = 0 other roots are 1 and 7 so, 1, 4, 7 are in AP.
Example 5 Find the condition that the roots of the equation x 3 − px 2 + qx − r = 0 be in GP.
a
Solution Let the roots of the given equation be ap,a, being in GP.
p
a
∴ Σα β γ = α β γ = ap ⋅ a ⋅ = r or a 3 = r
p
But a is a root of the given equation and as such
a 3 − pa 2 + qa − r = 0
or − pa 2 + qa = 0 [Qa 3 = r ]
or pa = q or p 3a 3 = q 3
or p 3r = q 3
or p 3r − q 3 = 0 is the required condition.
Example 6 Find the condition that the roots of the equation x 3 − px 2 + qx − r = 0 be in HP. Show
that mean root is 3r / q . Hence, solve the equation 6x 3 − 11x 2 − 3x + 2 = 0.
Solution Let the roots be α , β, γ which are in HP.
1 1 1
Hence, , , are in AP.
α β γ
1 1 2
∴ + = or βγ + αβ = 2γα
α γ β
Adding γα to both sides, we get
αβ + βγ + γα = 3γα or q = 3γα or γα = q / 3
www.pdfworld.in
www.pdfworld.in
q
Again αβγ = r or   β = r
 3
3r 3r
∴ β= i . e., is mean root.
q q
But β is a root of the given equation and as such
3 2
 3r   3r   3r 
  − p   + q   −r = 0
q q q
or 27r 2 − 9pqr + 2q 3 = 0
which is the required condition.
Proceeding as above, we get
3 1
3γα = Σαβ = − =−
6 2
1
∴ γα = −
6
2
Also, αβγ =
6
 1 2
or −  β = −
 6 6
∴ β = 2 is a root of the given equation which when divided by ( x − 2), by synthetic
1 1
division gives the quadratic ( 6x 2 + x − 1) = 0 giving the roots as − and . Hence,
2 3
1 1 1
roots are − , 2, which are clearly in HP, for their reciprocals −2, , 3 are in AP.
2 3 2
Example 7 The distances of three points A, B, C on a straight line from a fixed origin O on the line
are the roots of the equation ax 3 + 3bx 2 + 3cx + d = 0. Find
(a) the condition that one of the points A, B, C should bisect the distance between
the other two.
(b) the condition that the four points O , A, B, C should form a Harmonic division.
Solution (a) Let OA = α , OB = β, OC = γ, so that α , β, γ are the roots of the given equation.
By the given condition, B bisects AC
∴ AB = BC or OB − OA = OC − OB
2OB = OA + OC
or 2β = α + γ
i . e., α , β, γ are in AP.
(b) The required condition is 2b 3 − 3abc + a 2d = 0. The four points, O , A, B, C will
1 1 2
form a harmonic division, if + =
OA OC OB
1 1 1
i . e., , , are in AP.
OA OB OC
OA, OB, OC are in HP.
i . e.,α , β, γ the roots of given equation be in HP and ad 2 − 3bcd + 2c 3 = 0.
www.pdfworld.in
www.pdfworld.in
Example 8 Find the condition that the equation x 4 + px 3 + qx 2 + rx + s = 0, should have two
roots connected by the relation α + β = 0 and determine in that case the two quadratic
which shall have for roots (α , β ) and ( γ, δ ). Solve completely the equation
x 4 − 2x 3 + 4x 2 + 6x − 21 = 0.
Solution α+β=0 (given)
Σα = α + β + γ + δ = − p
γ + δ = −p [Q α + β = 0] …(i)
Σαβ = (α + β )( γ + δ ) + αβ + γδ = q
or αβ + γδ = q …(ii)
Σαβγ = αβ( γ + δ ) + γδ(α + β ) = − r
∴ αβ( − p ) = − r [from Eq. (i)]
r
∴ αβ = …(iii)
p
r
Σαβγδ = αβγδ = s or   γδ = s [from Eq. (iii)]
 p
ps
∴ γδ = …(iv)
r
Substituting the values of αβ and γδ in Eq. (ii), we get the required condition as
r ps r
+ = q or pqr − p 2s + r 2 = 0. Again, since α + β = 0 and αβ = . ∴α and β are the
p r p
r ps
roots of the quadratic x 2 − 0x + = 0 or px 2 + r = 0. Also, γ + δ = − p and γδ =
p r
ps
∴γ and δ are the roots of the equation x 2 − ( − p )x + =0
r
or rx 2 + prx + ps = 0
f ( x ) = x 4 − 2x 3 + 4x 2 + 6x − 21 = 0
Since, α + β + γ + δ = 2 and α + β = 0
∴ γ+δ=2
Let αβ = l and γδ = m
∴ f ( x ) = ( x 2 + l )( x 2 − 2x + m )
Comparing the coefficients
l + m = 4, −2l = 6, ⇒ l = − 3 and m = 7
∴ f ( x ) = ( x 2 − 3)( x 2 − 2x + 7) = 0
∴ x = 3, − 3, 1 ± i 6
Example 9 Show that the roots of the biquadratic ax 4 + 4bx 3 + 4dx + e = 0, have only two
ad 2 3bd
distinct values, if = = 1.
b 2e bd − ae
Solution Since, the roots are of two distinct kinds, it means that the biquadratic has two pairs
of equal roots α , α , β, β.
4b 2b
Σα = 2(α + β ) = − ∴ α+β=−
a a
www.pdfworld.in
www.pdfworld.in
e e
and α 2β 2 = , ∴ αβ =
a a
2
4b 3 4d e  2b e
∴ x4 + x + x + = x 2 + x + 
a a a  a a
Comparing the coefficients of like powers of x, we get
4b 2 e
+2 =0 …(i)
a2 a
4b e d ad 2
= 4 ; ∴ 2 = 1,
a a a be
e d
∴ = …(ii)
a b
b2 d
By Eq. (i), 2 + = 0 or 2b 2b + a 2d = 0
a2 b
ad 2 ad 2
Put b2 = ; ∴ 2b + a 2d = 0
e e
or 2bd + ae =0
or 3db = bd − ae
3bd
=1
bd − ae
ad 2
and =1
b 2e
Example 10 If z 4 + 6Hz 2 + 4Gz + (a 02I − 3H 2 ) = 0 has two pairs of equal roots. Prove that G = 0
and a 02I = 12H 2.
Solution Let the roots be α , α , β, β
Σα = 2(α + β ) = 0, ∴ α + β = 0
Let αβ = λ ; ∴z quadratic having roots α , β is z 2 + 0z + λ = 0
∴ f ( z ) = z 4 + 6Hz 2 + 4Gz + a 02I − 3H 2 = ( z 2 + λ )2
Comparing 2λ = 6H, 4G = 0, λ2 = a 02I − 3H 2
∴ λ = 3H and 9H 2 = a 02I − 3H 2
∴ a 02I = 12H 2
Hence, required conditions are that
G = 0 and a 02I = 12H 2
Example 11 If x + ay + a 2z = a 3, x + by + b 2z = b 3, x + cy + c 2z = c 3. Find the values of x , y and

z.
Solution a, b, c satisfy the cubict 3 = t 2z + ty + x ort 3 − t 2z − ty − x = 0 hasa, b andc as its roots
Σa = a + b + c = z
Σab = ab + bc + ca = − y
Σabc = abc = x
Hence the values of x , y and z are abc,− (ab + bc + ca ) and (a + b + c )
respectively.
www.pdfworld.in
www.pdfworld.in
Example 12 If α 1, α 2,…, α n are the roots of the equation

x n − p1x n − 1 + p2x n − 2 − p3x n − 3 + K + ( −1)n pn = 0,
find the value of (1 + α 1)(1 + α 2 )…(1 + α n ).
Solution Now, (1 + α 1)(1 + α 2 )K (1 + α n ) = 1 + s1 + s2 + s3 + K + sn
sn = sum of roots taken n at a time
= ( −1)r [( −1)r pr ] = ( −1)2r pr = pr
Hence, the required value is1 + p1 + p2 + K + pn
Example 13 If α 1, α 2, K , α n are the roots of the equation x n + nax − b = 0, show that

(α 1 − α 2 )(α 1 − α 3 )K (α 1 − α n ) = n (α n1 − 1 + a )
Solution f ( x ) = ( x − α 1 )( x − α 2 )K ( x − α n )
log f ( x ) = log( x − α 1) + log( x − α 2 ) + K + log( x − α n )
f (x ) f (x ) f (x )
f ′ (x ) = + +K+ or nx n − 1 + na = [( x − α 2 )( x − α 3 )K
x − α1 x − α 2 x − αn
( x − α n )] + [other (n − 1) factors of the above type each of which will contain x − α 1
as a factor].
Putting x = α 1 in both sides, we get
n(α n1 − 1 + a ) = {(α 1 − α 2 )(α 1 − α 3 ) K (α 1 − α n )}
Q All other factors vanish because of ( x − α 1) in them.
Example 14 Show that all the roots of the equation

x n + p1x n − 1 + p2x n − 2 + K + pn − 1x + pn = 0
can be obtained when they are in AP.
Solution The roots of the given equation being in AP. may be taken to be a, a + d ,
a + 2d … a + (n − 1)d .
We have to find values of a and d
Σα = na + {1 + 2 + 3 + 4 + K + (n − 1)}d = − p1
(n − 1)n
or na + d = − p1 …(i)
2
α 2 + β 2 + γ 2 = (α + β + γ )2 − 2(αβ + βγ + γα )
i . e., Σα 2 = ( Σα )2 − 2( Σαβ ) = p12 − 2p2
The above relation holds good for any number of roots.
or a 2 + (a + d )2 + (a + 2d )2 + K + {a + (n − 1)d }2 = p12 − 2p2
or na − 2ad {1 + 2 + 3 + K + (n − 1)} + d 2 {12 + 22 + K + (n − 1)2} = p12 − 2p2
2
n(n + 1) n(n + 1)( 2n + 1)

Σn = and Σn 2 =
2 6
Putting n = n − 1 in above formulae,
n(n − 1) d 2n(n − 1)( 2n − 1)
na 2 + 2ad + = p12 − 2p2 …(ii)
2 6
Now, if we square (i) and subtract it from n times (ii) a will be eliminated and we can
find d 2. Flowing found d we can find a from Eq. (i). When a and d are known, all the
roots are known.
www.pdfworld.in
www.pdfworld.in
Example 15 If f ( x ) = 0 is a cubic equation whose roots are α, β, γ and α is the harmonic mean of the
roots of f ′ ( x ) = 0. Show that α 2 = βγ.
Solution Let α , β, γ be roots of
x 3 − px 2 + qx − r = 0
∴ Σα = p, Σαβ = q ,αβγ = r and f ′ ( x ) = 3x 2 − 2px + q = 0
If its roots be a, b, then a + b = 2p / 3 and ab = q / 3
Now, α being the HM of a, b.
∴a, α , b are in HP.
2ab  q  2p q Σαβ
∴ α= = 2  ÷ = =
a+b  3 3 p Σα
∴ α (α + β + γ ) = αβ + βγ + γα
∴ α 2 = βγ
Example 16 Solve for x , y , z the equations

x + y + z = 5, yz + zx + xy = 7 and
xyz = 3.
Solution The given equations are symmetric with respect to x , y , z .
The equation whose roots are x , y , z is
t 3 − ( x + y + z ) t 2 + ( yz + zx + xy ) t − xyz = 0
or t 3 − 5t 2 + 7t − 3 = 0 …(i)
By trial, t = 1 satisfies it.
Dividing LHS of Eq. (i) by t − 1
The depressed equation is t 2 − 4t + 3 = 0 or (t − 1)(t − 3) = 0 ∴ t = 1, 3
values of x , y , z are 1, 1, 3.
Concept To transform an equation into another whose roots are the reciprocals of the roots of the
given equation.
−1 −2
Let f (x ) ≡ a 0x n + a1x n + a 2x n + K + an − 1x + 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  1
Hence, the transformed equation is obtained by putting x = in f (x ) = 0 and is therefore f   = 0.
y y 
1 1 1 1
i . e. , a0 + a1 n − 1 + a 2 n − 2 + K + an −1 + an = 0
yn y y y
n −1
or an y n + an − 1y + K + a 2 y 2 + a1y + a 0 = 0
Rule If the equation given be complete (if not, it may be made to take that form), then the above
transformation is effected, if we take the last coefficient to be the Ist, last but one to be the IInd and so
on.
Concept All those equations which remain unchanged when x is replaced by 1/x are called reciprocal
equation. These are of two types
www.pdfworld.in
www.pdfworld.in
(i) those in which the coefficient of terms, equidistant from the begining and the end, are equal and of
the same sign. e. g. ,
x 4 + 6x 3 + 8x 2 + 6x + 1 = 0
(ii) those in which these coefficients are equal but of opposite sign e. g. ,
x 5 − 4x 4 + 7x 3 − 7x 2 + 4x − 1 = 0
Thus, we conclude that if α is a root of the reciprocal equation, then 1 / α must be its root. Hence, the
root of a reciprocal equation occur in pairs ofα, 1/α, β, 1/β, γ , 1/γ 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 IInd type, then it will be seen that x 2 − 1 will always be its
factor e. g. ,
6 x 6 + 5 x 5 − 44 x 4 + 44 x 2 − 5 x − 6 = 0
6 (x 6 − 1) + 5 x (x 4 − 1) − 44 x 2 (x 2 − 1) = 0
(x 2 − 1)[6(x 4 + x 2 + 1) + 5 x (x 2 + 1) − 44 x 2 ] = 0
(x 2 − 1)(6 x 4 + 5 x 3 − 38x 2 + 5 x + 6) = 0
Above shows that ( x 2 − 1) is a factor of the given equation. The equation 6x 4 + 5x 3 − 38x 2 + 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 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
−1 −2
a 0 y n − a1 y n + a2 y n − K + (−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 IInd, IVth, VIth, i . e. , all even terms, i . e. , by changing
the sign of every alternate term beginning from the IInd.
Concept 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 yn − 1 yn − 2
i . e. , a0 n
+ a1 n − 1 + a 2 n − 2 + K + an = 0
m m m
−1 −2
or a 0y n + a1my n + a 2m 2y n + K + mn an = 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 IInd by m, m 2,
m3 , K , 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.
www.pdfworld.in
www.pdfworld.in
Concept To transform a given equation into another, whose roots are the roots of the given equation
diminished (or increased) by a constant h.
−1 −2
Let f (x ) ≡ a 0x n + a1x n + a 2x n + K + an − 1x + 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
−1 −2
or a 0 ( y + h )n + a1 ( y + h )n + a 2 ( y + h )n + .... + 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.
−1 −2
A 0y n + A1y n + A 2y n + K + An − 1y + An = 0 …(ii)
The problem is to find A 0 , A1 , A 2 , K , An

Now, y =x −h
−1 −2
∴ A 0 (x − h ) + A1 (x − h )n
n
+ A 2 (x − h )n + K + An − 1 (x − h ) + An = 0
n −1 n −2 n −3
or [A 0 (x − h ) + A1 (x − h ) + A 2 (x − h ) + K + 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 A 0 (x − h )n − 1 + A1 (x − h )n − 2 + K + 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 An , An − 1 , …, A 2 , A1 and the last quotient A 0 is clearly equal to a 0.
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.
Example 1 Solve the equation x4 − 10 x3 + 26 x2 −10 x + 1 = 0.

Solution Dividing throughout by x 2 it can be written as
 2 1  1
 x + 2  − 10  x +  + 26 = 0
 x   x
1 1
Put x + = y, ∴ x 2 + 2 = y 2 − 2
x x
∴ ( y 2 − 2) − 10y + 26 = 0
y 2 − 10y + 24 = 0
⇒ ( y − 6)( y − 4) = 0
1
∴ y = 6, 4 or x + =6
x
6± 32
∴ x 2 − 6x + 1 = 0, x = =3± 2 2
2
1
and x + =4
x
∴ x 2 − 4x + 1 = 0
4 ± 12
x = =2± 3
2
www.pdfworld.in
www.pdfworld.in
Example 2 Form equation, whose roots are the roots of the equation x 5 − x 2 + x − 3 = 0 with their
sign changed.
Solution The given equation being not complete, may be written as
x 5 + 0x 4 + 0x 3 − x 2 + x − 3 = 0
and the transformed equation is
x 5 − 0x 4 + 0x 3 + x 2 + x + 3 = 0
∴ x5 + x2 + x + 3 = 0
Example 3 If the reciprocal of every root of x 3 + x 2 + ax + b = 0 is also a root. Prove that a = b = 1

or a = b = −1. In each case find the roots.
Solution Let f ( x ) = x 3 + x 2 + ax + b = 0 …(i)
1
If α is a root, then is also a root.
α
∴ It is a reciprocal equation.
It remains unchanged by the transformation
1
y =
x
1
On putting x = , we get
y
 1 1 1 1
f  = 3 + 2 + a⋅ + b = 0
y y y y
or 1 + y + ay 2 + by 3 = 0
or by 3 + ay 2 + y + 1 = 0
Writing it in x,
∴ bx 3 + ax 2 + x + 1 = 0 …(ii)
Eqs. (i) and (ii) represent the same equation, i . e., they have same roots.
On comparing coefficients of Eqs. (i) and (ii), we get
1 1 a b
= = = ⇒ a 2 = 1 and a = b
b a 1 1
∴ a =b = ±1
(i) If a = b = 1, Eq. (i) becomes
x3 + x2 + x +1= 0
or n x ( x + 1) + ( x + 1) = 0
2
( x 2 + 1)( x + 1) = 0
x = − 1, ± i
(ii) If a = b = −1, Eq. (i) becomes
x 3 + x 2 − x −1 = 0
⇒ x 2( x + 1) − 1( x + 1) = 0
⇒ t( x + 1)( x 2 − 1) = 0
⇒ ( x + 1)( x + 1)( x − 1) = 0
∴ Roots are −1, − 1, 1
www.pdfworld.in
www.pdfworld.in
Example 4 If α, β and γ are roots of the equation x 3 − 6x 2 + 12x − 8 = 0, find an equation whose
roots are α − 2, β − 2, γ − 2. Hence, find roots of the given equation.
Solution Roots of equation
x 3 − 6x 2 + 12x − 8 = 0 …(i)
and α,β, γ. Diminishing its roots by 2
2 1 –6 –12 –8
2 –8 8
1 –4 4 0
2 –4
1 –2 0
2
1 0
1
∴ x3 = 0 …(ii)
Roots of Eq. (ii) are 0, 0, 0
i . e., α − 2 =β − 2 = γ − 2 = 0
∴ α =β = γ = 2
Roots of Eq. (i) are 2, 2, 2.
Example 5 Find the equation whose roots exceed by 2, the roots of the equation
4x 4 + 32x 3 + 83x 2 + 76x + 21 = 0.. Hence, solve the equation.
Solution Equation is
4x 4 + 32x 3 + 83x 2 + 76x + 21 = 0 …(i)
Dividing 4x + 32x + 83x + 76x + 21 successively by x + 2
4 3 2
–2 4 32 83 76 21
–8 –48 –70 –12
4 24 35 6 9
–8 –32 –6
4 16 3 0
–8 –16
4 8 –13
–8
4 0
4
Transformed equation is
4x 4 − 13x 2 + 9 = 0 …(ii)
i . e., ( 4x 2 − 9)( x 2 − 1) = 0
∴ x = ± 3 / 2, ± 1
Roots of Eq. (ii) are the roots of Eq. (i) increased by 2
Roots of Eq. (i) are the roots of Eq. (ii) decreased by 2
Roots of Eq. (i) are 3/2 − 2, − 3/2 − 2, 1 − 2, −1 −2.
i . e., 1/2, –7/2, –1, –3
www.pdfworld.in
www.pdfworld.in
Example 6 The difference between two roots of the equation 2x 3 + x 2 − 7x − 6 = 0, is 3. Solve it

by diminishing the roots by 3.
Solution Let α , α + 3, β be the roots of given equation
2x 3 + x 2 − 7x − 6 = 0 …(i)
Diminishing its roots by 3
3 2 1 –7 –6
6 21 42
2 7 14 36
6 39
2 13 53
6
2 19
2
2x 3 + 19x 2 + 53x + 36 = 0 …(ii)
Roots of this equation are α − 3, α , β − 3.
Thus, Eqs. (i) and (ii) have a common root α, finding the HCF of the LHS of Eqs. (i)
and (ii)
2 1 –7 –6 2 19 53 36 1
×3 2 1 –7 –6
6 3 –21 –18 6 18 60 42
6 20 14 3 10 7 3
–1 –17 –35 –18 3 3
17 35 18 7 7 7
×3 1 1 1
51 105 54 1 1
51 170 119 ×
−65 −65 −65
HCF = x + 1, equating it to zero, x = −1
∴ −1 is a root of given equation.
Dividing Eq. (i) by x + 1
–1 2 1 –7 –6
–2 1 6
2 –1 –6 0
Depressed equation is 2x − x − 6 = 0
2
1 ± 1 + 48
∴ x =
4
1± 7 −3
= = 2,
4 2
Hence, the roots of the given equation are –1, 2, –3/2.
www.pdfworld.in
www.pdfworld.in
Example 7 If α , β, γ are the roots of 8x 3 − 4x 2 + 6x − 1 = 0, find the equations whose roots are
α + 1 / 2, β + 1 / 2 and γ + 1 / 2.
Solution 8x 3 − 4x 2 + 6x − 1 = 0 …(i)
Roots of required equation are the roots of given equation increased by 1/2 (or
decreased by –1/2) Diminishing the roots of Eq. (i) by –1/2
1 8 –4 6 –1
−
2
–4 4 –5
8 –8 1 –6
0
–4 6
8 –12 1
6
–4
8 –16
8
Hence, required equation is
8x 3 − 16x 2 + 16x − 6 = 0
or 4x 3 − 8x 2 + 8x − 3 = 0
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 αβ + βγ + γα is a symmetric function of the roots α , β , γ of a cubic. For the sake of
brevity, we generally denote such symmetric functions by attaching Σ to one of its terms.
Thus, Σαβ = αβ + βγ + γα
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 α , β and γ are the roots of a cubic equation, then the total number of terms in
3!
Σα 2β is =6
( 3 − 3)! 0 !
αβ 3!
Σ = Σαβγ −2 is =3
γ 2
( 3 − 2)! 2 !
2. If α , β , γ and δ are the roots of a biquadratic equation, then the total number of terms in
4!
Σα 2βγ is =2
(4 − 1)! 2 !
α 4!
Σ = Σαβ −1 is = 12
β (4 − 2)! 0 !
www.pdfworld.in
www.pdfworld.in
Example 1 Calculate the values of the following symmetric functions for the cubic
x 3 + px 2 + qx + r = 0, whose roots are α , β, γ
(i) Σα 2β (ii) Σα 2 (iii) Σα 3 (iv) Σα 2βγ
(v) Σα β
2 2
(vi) Σα β 2
(vii) Σα 4
(viii) Σα 3β 2
Solution (i) Σα 2β = α 2β + α 2γ + β 2γ + β 2α + γ 2α + γ 2β [Number of terms 6]
Now, ΣαΣαβ = (α + β + γ )(αβ + βγ + λα ) [Number of terms 9]
= Σα β + 3 αβγ2
Σα β = Σα Σ αβ − 3 αβγ = ( − p )(q ) − 3 ( − r ) = 3 r − pq
2
(ii) Σα = α + β + γ
2 2 2 2
Now, ( Σα )2 = (α + β + γ )2 = Σα 2 + 2Σαβ [Number of terms 9]

∴ Σα = ( Σα ) − 2Σ(αβ ) = p − 2q
2 2 2
Note Above relation is true for equations of all degrees and will be used frequently.
(iii) Σα 3 = α 3 + β 3 + γ 3
Now, Σα Σα 2 = (α + β + γ )(α 2 + β 2 + γ 2 )
= Σα 3 + Σα 2β
∴ Σα 3 = Σα Σα 2 − Σα 2β
The values of Σα 2 and Σα 2β have already been calculated in relations (i) and (ii)
∴ Σα 3 = ( − p )( p 2 − 2q ) − ( 3r − pq ) = 3pq − p 3 − 3r
(iv) Σα βγ = α 2βγ + β 2γα + γ 2αβ = αβγ ⋅ Σα
2
Hence, Σα 2βγ = αβγ Σα = pr

(v) Σα 2β 2 = α 2β 2 + β 2γ 2 + γ 2α 2
Now, ( Σαβ )2 = (αβ + βγ + γα )2 = Σα 2β 2 + 2Σα 2βγ
Σα 2β 2 = ( Σαβ )2 − 2Σα 2βγ = q 2 − 2pr
(vi) Σ a 3β = α 3β + α 3γ + β 3α + β 3γ + γ 3α + γ 3β
Now, Σα 2 Σ αβ = (α 2 + β 2 + γ 2 )(αβ + βγ + γα )
= Σα 3β + Σα 2βγ
∴ Σα 3β = Σα 2Σαβ − αβγ ⋅ Σα
= q ( p 2 − 2q ) − pr
= p 2q − 2q 2 − pr
(vii) Σ α 4 = ( Σα 2 )2 − 2 ( Σα 2β 2 )
= ( p 2 − 2q )2 − 2 (q 2 − 2pr ) [by relations (ii) and (v)]
= p − 4p q + 2q + 4pr
4 2 2
(viii) Σα β
3 2
Now, Σα Σα 2β 2 = Σα 3β 2 + Σα 2β 2 γ
∴ Σα 3β 2 = Σα Σα 2β 2 − αβγ Σαβ
= − p [q 2 − 2pr ] − ( − r )q
= − pq 2 + 2p 2r + qr
www.pdfworld.in
www.pdfworld.in
Example 2 Find the equation, whose roots are the squares of the roots of x 3 + qx + r = 0.
Solution Given equation is
x 3 + qx + r = 0 …(i)
If y be a root of the transformed equation, then
y = x2 …(ii)
Now, we have to eliminate x from Eqs. (i) and (ii)
[Method Collect terms with odd powers of x on one side and with even power of x
on the other side. Square and put x 2 = y ]
From Eq. (i) x3 + qx = − r
Squaring, we get x6 + 2qx 4 + q 2x 2 = r 2
Putting x2 =y
y3 + 2qy 2 + q 2y − r 2 = 0
which is required equation.
Example 3 Find the equation, whose roots are the squares of the roots of
x 3 + px 2 + qx + r = 0
Solution x + px 2 + qx + r = 0
3
…(i)
If y be a root of the transformed equation, then y = x 2
…(ii)
we have, now to eliminate x from Eqs. (i) and (ii)
From Eq. (i), x 3 + qx = − ( px 2 + r )
On squaring both sides
x 6 + 2qx 4 + q 2x 2 = p 2x 4 + 2prx 2 + r 2
Putting x = y , we get
2
y 3 + ( 2q − p 2 )y 2 + (q 2 − 2pr )y − r 2 = 0
Example 4 Find the equation whose roots are the squares of the roots of
x 4 + ax 3 + bx 2 + cx + d = 0
and if α , β, γ and δ are the roots, find the values of
(i) Σα 2 (ii) Σα 2β 2 (iii) Σα 2β 2γ 2
Solution (a) x + ax + bx + cx + d = 0
4 3 2
…(i)
If y be a root of the transformed equation, then
y = x2 ...(ii)
We have to now eliminate x between Eqs. (i) and (ii).
From Eq. (i) x 4 + bx 2 + d = − (ax 3 + cx )
x 8 + b 2x 4 + d 2 + 2bx 6 + 2bdx 2 + 2dx 4 = a 2x 6 + c 2x 2 + 2acx 4
On putting x 2 = y , we get
y 4 + b 2y 2 + d 2 + 2by 3 + 2bdy + 2dy 2 = a 2y 3 + c 2y + 2acy 2
⇒ y 4 + ( 2b − a 2 )y 3 + (b 2 + 2d `− 2ac )y 2 + ( 2bd − c 2 )y − d 2 = 0 …(iii)
(b) If α , β, γ and δ are the roots of Eq. (i), then the roots of Eq. (iii) are α 2, β 2, γ 2, δ 2
(i) Σα 2 = − ( 2b − a 2 ) = a 2 − 2b
(ii) Σα 2β 2 = b 2 + 2d − 2ac
(iii) Σα 2β 2γ 2 = − ( 2bd − c 2 ) = c 2 − 2bd
www.pdfworld.in
www.pdfworld.in
Example 5 Form an equation whose roots are β 2 + γ 2, γ 2 + α 2, α 2 + β 2, where α, β, γ are roots of

z 3 + pz + q = 0
Solution Qα,β, γ are the roots of
z 3 + pz + q = 0 …(i)
α + β + γ = 0
∴  …(ii)
αβγ = − q 
If the transformed equation is in terms of y, then
y = β 2 + γ 2 = (β + γ )2 − 2βγ
2αβγ 2q
= ( −α )2 − = α2 + [ QEq. (i)]
α α
2q
∴ y =z + 2
or z − zy + 2q = 0
3
…(iii)
z
Subtracting Eq. (iii) from Eq. (i)
q
( p + y )z − q = 0, ∴ z =
p+y
Putting this value of z in Eq. (i), we get
q3 q
+ p⋅ +q =0
( p + y )3 p+y
q 2 + p ( p + y )2 + ( p + y )3 = 0
or y + 4py 2 + 5p 2y + ( 2p 3 + q 2 ) = 0
3
which is the required equation.

Concept To find the sum of the integral power of the roots of an equation.
−1 −2
a 0x n + a1x n + a 2x n + K + an = 0 …(A)
If α1 , α 2 , α3, …, are the roots of the polynomial equation, then let us represent sr = Σαr1
i . e. , s1 = Σα1 = α1 + α 2 + α3 + K
s2 = Σα12 = α12 + α 22 + α33 + K
s3 = Σα13 = α31 + α32 + α33 + K and so on
The following results will help us to find out the value of s1 , s2 , s3 , K , sr .
a 0s1 + a1 = 0 …(i)
a 0s2 + a1s1 + 2a 2 = 0 …(ii)
a 0s3 + a1s2 + a 2s1 + 3a3 = 0 …(iii)
a 0sr + a1sr −1 + a 2sr −2 + K + 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 , providedr < n. If however r ≥ n, then multiply Eq. (A) by x r − n
−1 −2 r −n + 1 −n
a 0x r + a1x r + a 2x r + K + an − 1x + anr =0 …(v)
Putting x = α1 , α 2 , K , αn in succession,
a 0αr1 + a1αr1 − 1 + a 2αr1 − 2 + K + an r −n + 1
− 1α1 + an αr −n
=0
a 0αr2 + a1αr2 − 1 + a 2αr2 − 2 + K + an r −n + 1

− 1α 2 + an αr2 − n = 0
www.pdfworld.in
www.pdfworld.in
K K K
K K K
K K K
a 0αrn + a1αrn − 1 + a 2αrn − 2 + K + an r −n + 1
− 1αn + an αrn − n = 0
Adding, we have
a 0sr + a1sr −1 + a 2sr −2 + K + an − 1sr − n + 1 + an sr −n =0
Putting r = n , n + 1, n + 2 etc, we get
a 0sn + a1sn −1 + a 2sn −2 + K + an − 1s1 + nan = 0
[Qs0 = α10 + α 02 + K + αn0 = n]
a 0sn +1 + a1sn + a 2sn −1 + K + an − 1s 2 + an s1 = 0
a 0sn + 2 + a1sn +1 + a 2sn + K + an − 1s3 + an s2 = 0
and so on. These results gives values of
sn , sn + 1, sn + 2 , K .
Example 1 Find the sum of the cubes of the roots of the equation x 4 + ax 3 + bx 2 + cx + d = 0.
Solution Let α 1, α 2, α 3 and α 4 be the roots of the equation.
Here, a 0 = 1, a1 = a,a 2 = b,a 3 = c,a 4 = d
∴ a 0s1 + a1 = 0 ⇒ s1 + a = 0
∴ s1 = − a
⇒ α1 + α 2 + α 3 + α 4 = − a
Now, a 0s2 + a1s1 + 2a 2 = 0
⇒ s2 + a( −a ) + 2b = 0
∴ s2 = a 2 − 2b
⇒ α 12 + α 22 + α 23 + α 24 = a 2 − 2b
Now, a 0s3 + a1s2 + a 2s1 + 3a 3 = 0
⇒ s3 + a (a 2 − 2b ) + b ( −a ) + 3c = 0
∴ s3 = − a 3 + 3ab − 3c
⇒ α 13 + α 32 + α 33 + α 34 = 3ab − a 3 − 3c
Example 2 If α, β and γ be the roots of the equation x 3 + px + q = 0 , find value of Σx 5.

Solution a 0 = 1, a1 = 0, a 2 = p, a 3 = q
∴ a 0s1 + a1 = 0 ⇒ s1 = 0
a 0s2 + a1s1 + 2a 2 = 0 ⇒ s2 = − 2p
a 0s3 + a1s2 + a 2s1 + 3a 3 = 0 ⇒ s3 = − 3q
Multiplying both sides of given equation by x 2
x 5 + px 3 + qx 2 = 0
Putting x = α , β, γ in succession and adding
s5 + ps3 + qs2 = 0
or s5 + p ( −3q ) + q ( −2p ) = 0
∴ s5 = Σx 5 = 5pq
www.pdfworld.in
www.pdfworld.in
Example 3 If α, β, γ be the roots of the equation x 3 − 7x + 7 = 0. Find α − 4 + β − 4 + γ − 4.

Solution Roots of the equation x 3 − 7x + 7 = 0 are α , β, γ. Changing x to 1/x and multiplying by
x 3, we get
7x 3 − 7x 2 + 1 = 0 …(i)
Its roots being the reciprocals of the roots of given equation are1/α , 1/β, 1/ γ.
Let us denote them by α ′ , β′ , γ′.
We have to find
α −4 + β −4 + γ −4 = (α ′ )4 + (β′ )4 + ( γ′ )4 = s′4
Here, a 0 = 7,a1 = − 7,a 2 = 0, a 3 = 1
∴ a 0s1′ + a1 = 0 ⇒ 7s′1 − 7 = 0
∴ s′1 = 1
a 0s′ 2 + a1s′1 + 2a 2 = 0 ⇒ 7s − 7 = 0
∴ s′ 2 = 1
a 0s′ 3 + a1s′ 2 + a 2s′1 + 3a 3 = 0
⇒ 7s′ 3 − 7 + 3 = 0
∴ s′ 3 = 4 / 7
Multiplying Eq. (i) by x, we get
7x 4 − 7x 3 + x = 0
Putting x = α ′ , β′ , γ′ in succession and adding
7s′ 4 − 7s′3 + s1′ = 0 ⇒ 7s′4 − 4 + 1 = 0
Hence, α −4 + β −4 + γ −4 = 3 / 7
∴ s4′ = 3 / 5
Example 4 If α + β + γ = 6; α 2 + β 2 + γ 2 = 14 and α 3 + β 3 + γ 3 = 36. Prove that

α 4 + β 4 + γ 4 = 98.
Solution Let α , β, γ be the roots of the equation
a 0x 3 + a1x 2 + a 2x + a 3 = 0 (a 0 ≠ 0) …(i)
We are given that s1 = 6, s2 = 14 and s3 = 36
∴ a 0s1 + a1 = 0 ⇒ 6a 0 + a1 = 0
∴ a1 = − 6a 0
∴ a 0s2 + a1s1 + 2a 2 = 0
⇒ 14a 0 − 6a 0( 6) + 2a 2 = 0
∴ a 2 = 11a 0
Now, a 0s3 + a1s2 + a 2s1 + 3a 3 = 0
⇒ 36a 0 − 6a 0(14) + 11a 0( 6) + 3a 3 = 0
∴ a 3 = − 6a 0
Putting these values of a1, a 2, a 3 in Eq. (i) and dividing by a 0, we get
x 3 − 6x 2 + 11x − 6 = 0 …(ii)
Multiplying both sides of Eq. (ii) by x, we get
x 4 − 6x 3 + 11x 2 − 6x = 0
www.pdfworld.in
www.pdfworld.in
Putting x = α , β and γ in succession and adding

s4 − 6s3 + 11s2 − 6s1 = 0
s4 − 6 ( 36) + 11(14) − 6 ( 6) = 0
∴ s4 = 216 − 154 + 36 = 98
Hence, α 4 + β 4 + γ 4 = 98
Example 5 Find x 5 + y 5 + z 5, it being given that

x + y + z = 1; x 2 + y 2 + z 2 = 2 ;
x 3 + y 3 + z 3 = 3.
Solution Let x , y , z be the roots of the equation
a0 t 3
+ a1t 2
+ a2 t + a3 = 0 (a 0 ≠ 0) …(i)
We are given that s1 = 1, s2 = 2, s3 = 3
a 0s1 + a1 = 0 ⇒ a 0 + a1 = 0
∴ a1 = − a 0
a 0s2 + a1s1 + 2a 2 = 0 ⇒ 2a 0 − a 0 + 2a 2 = 0
a
∴ a2 = − 0
2
a 0s3 + a1s2 + a 2s1 + 3a 3 = 0
a a0
⇒ 3a 0 − a 0( 2) − 0 ⋅1 + 3a 3 = 0 ⇒ 3a 3 + =0
2 2
a
∴ a3 = − 0
6
Putting these values of a1, a 2, a 3 in Eq. (i) and dividing by a 0, we get
1 1
t3 − t2 − t − = 0
2 6
or 6t 3 − 6t 2 − 3t − 1 = 0 …(ii)
Multiplying both sides of Eq. (ii) by t
6t 4 − 6t 3 − 3t 2 − t = 0
Putting t = x , y , z in succession and adding
6s4 − 6s3 − 3s2 − s1 = 0
or 6s4 − 6 ( 3) − 3 ( 2) − 1 = 0
25
∴ s4 =
6
Multiplying both sides of Eq. (ii) by t 2
6t 5 − 6t 4 − 3t 3 − t 2 = 0
Putting t = x , y , z in succession and adding
6s5 − 6s4 − 3s3 − s2 = 0
 25
6s5 − 6   − 3 ( 3) − 2 = 0
 6
∴ s5 = 6
Hence, x + y + z5 = 6
5 5
www.pdfworld.in
www.pdfworld.in
Example 6 If α, β and γ are the roots of x 3 + px + q = 0. Prove that

α 5 + β5 + γ 5 α 3 + β3 + γ 3 α 2 + β2 + γ 2
(i) = ⋅
5 3 2
(ii) 3 (α 2 + β 2 + γ 2 )(α 5 + β 5 + γ 5 ) = 5 (α 3 + β 3 + γ 3 )(α 4 + β 4 + γ 4 )
Solution Here, a 0 = 1, a1 = 0, a 2 = p, a 3 = q
a 0s1 + a1 = 0 ⇒ s1 = 0
a 0s2 + a1s1 + 2a 2 = 0 ⇒ s2 = − 2p
a 0s3 + a1s2 + a 2s1 + 3a 3 = 0
⇒ s3 = − 3q
(i) Multiplying both sides of given equation by x 2.
x 5 + px 3 + qx 2 = 0
s5 + ps3 + qs2 = 0
or s5 − 3pq − 2pq = 0
∴ s5 = 5pq …(i)
s5
Now, = pq ;
5
s3 s2
⋅ = ( −q )( − p ) = pq
3 2
s5 s3 s2
∴ = ⋅
5 3 2
α 5 + β5 + γ 5 α 3 + β3 + γ 3 α 2 + β2 + γ 2
Hence, = ⋅
5 3 2
(ii) Multiplying both sides of the given equation by x,
x 4 + px 2 + qx = 0
s4 + ps2 + qs1 = 0 or s4 − 2p 2 = 0
∴ s4 = 2p 2
Also, s5 = 5pq [from Eq. (i)]
3 (α + β + γ )(α + β + γ ) = 3s2s5
2 2 2 5 5 5
= 3( −2p )( 5pq ) = − 30p 2q

5 (α 3 + β 3 + γ 3 )(α 4 + β 4 + γ 4 ) = 5s3s4
= 5( −3q )( 2p 2 ) = − 30p 2q
Hence, 3 (α 2 + β 2 + γ 2 )(α 5 + β 5 + γ 5 )
= 5 (α 3 + β 3 + γ 3 )(α 4 + β 4 + γ 4 )
Example 7 If a + b + c = 0, show that a 5 + b 5 + c 5 = − 5 abc(bc + ca + ab )

Solution Let a, b, c be the roots of the equation
a 0x 3 + a 2x + a 3 = 0 …(i)
[IInd term is missing Q a + b + c = 0]
We are given that s1 = 0 ;
www.pdfworld.in
www.pdfworld.in
∴ a1 = 0
2a 2
a 0s2 + a1s1 + 2a 2 = 0 ⇒ s2 =
a0
a 0s3 + a1s2 + a 2s1 + 3a 3 = 0
−3a 3
⇒ s3 =
a0
Multiplying both sides of the Eq. (i) by x 2,
a 0x 5 + a 2x 3 + a 3x 2 = 0
Putting x = a, b, c in succession and adding
a 0s5 + a 2s3 + a 3s2 = 0
3a a 2a a
or a 0s5 − 2 3 − 2 3 = 0
a0 a0
5a 2a 3 5a 2a 3
∴ s5 = or a 5 + b 5 + c 5 =
a 02 a 02
a2 a
Also, bc + ca + ab = and abc = − 3
a0 a0
 a   a  5a a
∴ − 5abc(bc + ca + ab ) = − 5  − 3   2  = 22 3
 a0   a0  a0
Hence, a 5 + b 5 + c 5 = − 5abc (bc + ca + ab )
Example 8 Show that sum of products of the first integers taken three at a time is
n 2(n + 1)2(n − 2) (n − 1) / 48.
Solution The equation whose roots are first n integers is
f ( x ) = ( x − 1)( x − 2)( x − 3) K ( x − n ) [say]
n −1 n −2 n −3
= x + p1x
n
+ p2 x + p3 x +K= 0
We have to find the symmetric function of the type.
Σαβγ = − p3. Here, α , β, γ, K means 1, 2, 3 … respectively. i.e., first n integers.
Now, in usual notation
n (n + 1)
s1 = 1 + 2 + 3 + 4 + K + n =
2
n (n + 1)( 2n + 1)
s2 = 12 + 22 + 32 + 42 + K + n 2 =
6
2
n (n + 1) 
s3 = 13 + 23 + 33 + 43 + K + n 3 =  
 2 
Now, s1 + p1 = 0 ;
− n (n + 1)
∴ p1 = − s1 =
2
Q a0 = 1
s2 + p1s1 + 2p2 = 0
n (n + 1)( 2n + 1) n 2(n + 1)2
or − + 2 p2 = 0
6 4
www.pdfworld.in
www.pdfworld.in
1 n 2(n + 1)2 n (n + 1)( 2n + 1) 

p2 =  − 
2  4 6 
s3 + p1s2 + p2s1 + 3p3 = 0
n 2(n + 1)2 n (n + 1) n (n + 1)( 2n + 1)
or − ⋅
4 2 6
n (n + 1) n 2(n + 1)2 n (n + 1)( 2n + 1) 
+ × −  + 3 p3 = 0
4  4 6 
n 2(n + 1)2  n (n + 1)  n 2(n + 1)2( 2n + 1)  1
or 1+ − × 1 +  + 3p3 = 0
4  4  12  2
n 2(n + 1)2  4 + n 2 + n ( 2n + 1) 3 
or  − ⋅  + 3P3 = 0
4  4 3 2
n 2(n + 1)2  4 + n 2 + n − 4n − 2 
or
4  4  + 3 p3 = 0
 
n 2(n + 1)2 2
∴ − p3 = (n − 3n + 2)
48
n 2(n + 1)2(n − 1)(n − 2)
or Σαβγ = − p3 =
48
f ′ (x )
Example 9 If α, β, γ,K be the roots of the equation f ( x ) = 0. Prove that = ΣSm − 1x −m where Sm
f (x )
is sum of the nth powers of the roots.
Solution f ( x ) = ( x − α )( x − β )( x − γ ) .. . n factors. Taking log on both sides,
log f ( x ) = log( x − α ) + log( x − β ) + log( x − γ ) + K
Differentiating w.r.t. x, we get
1 1 1 1
f ′ (x ) = + + + ... …(i)
f (x ) x −α x −β x − γ
−1
1 1  α
Now, = 1 − 
x −α x  x
1  α α2 α3  1 α α2 α3
= 1 + + 2 + 3 + K  = + 2 + 3 + 4 + K
x  x x x  x x x x
f ′ (x ) 1 α α2 α3 1 β β2 β3
∴ = + 2 + 3 + 4 + K + + 2 + 3 + 4 + K for n roots
f (x ) x x x x x x x x
1 s s s
=n + 1 + 2 + 3 +K
x x2 x3 x4
Now, S0 = 1 + 1 + 1 + K = n
f ′ (x )
= s0x −1 + s1x −2 + s2x −3 + s3x −4 + K
f (x )
= ΣSm − 1x −m
Hence proved.
www.pdfworld.in
www.pdfworld.in
−1
Concept Let f (x ) = a0x n + a1x n + K , an −1 x + an polynomial of degree n ≥ 1 with integral
coefficients and an ≠ 0, then every integral zero of f (x ) is a divisor of the constant terms an . But the
converse is not always true. i.e., every integral divisor of the constant term need not be a zero of the
polynomial with integral coefficients.
For example let f (x ) = x 2 + 6.
Here, 3 divides 6. The constant term of the polynomial f (x ) with integral coefficients.
But f (3) = 32 + 6 = 15 ≠ 0
∴ 3 is not a zero of the polynomial f (x ).
Hence, the converse may or may not be true.
Concept Upper and lower bounds for the real roots of an equation.
Two real numbers M and m [m ≤ M ] are called the upper and the lower bounds respectively of the real
roots of the real equation f (x ) = 0, if the real roots of f (x ) = 0 (if any) lie between m and M .
Method of grouping for finding an upper bound of the real roots.
(1) Combine each −ve term with a + ve term. (the power of positive term must be greater than −ve term).
No −ve term is to be left alone.
(2) Choose a value of x say M which makes every group + ve or zero.
Then, M is called an upper bound of the real roots of the real equationf (x ) = 0.
Method of grouping for finding a lower bound of the real roots.
Let f (− x ) = g (x ), if m is the upper bound of the real roots of g (x ) = 0, then − m is the lower bound of real
roots of f (x ) = 0.
Example 1 Find all the integral roots of 5x 3 − 11x 2 + 12x − 2 = 0.

Solution f ( x ) = 5x 3 − 11x 2 + 12x − 2 = 0
Here, constant term = − 2

Divisors of constant term are ± 1, ± 2.
So, possible values for integral roots are ± 1, ± 2
Now, f (1) = 5 − 11 + 12 − 2 ≠ 0
f ( − 1) = − 5 − 11 − 12 − 2 ≠ 0
f ( 2) = 40 − 44 + 24 − 2 ≠ 0
f ( − 2) = − 40 − 44 − 24 − 2 ≠ 0
Hence, there is no integral root of this equation.
Example 2 Find the integral roots, if any of the equation x 4 + x 3 − 2x 2 + 4x − 24 = 0. Also solve
the equation.
Solution Let f ( x ) = x 4 + x 3 − 2x 2 + 4x − 24 = 0
f ( x ) = x 4 − 2x 2 + x 3 − 24 + 4x
= x 2 ( x 2 − 2) + ( x 3 − 24) + 4x
Clearly, x = 3 makes every groups on RHS +ve or zero.
M (an upper bound) of real roots of f ( x ) = 0 is 3.
f ( − x ) = 4x 4 − 4x 3 − 8x 2 − 16x − 96
= x 4 − 4x 3 + x 4 − 8x 2 + x 4 − 16x + x 4 − 96
= x 3 ( x − 4) + x 2 ( x 2 − 8) + x ( x 3 − 16) + x 4 − 96
www.pdfworld.in
www.pdfworld.in
Clearly, x = 4 makes RHS + ve or zero.

m (lower bound) of f ( x ) = 0 is 4.
Divisors of − 24 between − 4 and 3 are − 4, − 3, − 2, − 1, 1, 2, 3, so 0 is not a root.
Now, f (1) = 20 ≠ 0
∴ 1 is not a root.
Divisors decreased by 1 are – 5, – 4, – 3, – 2, 1, 2, 3 does not divide f (1) = − 20
∴ − 2 rejected
f ( − 1) = − 30 ≠ 0. So, −1, is not a root.
Divisors increased by 1 are −3, − 2, − 1, 3, 4.
4 does not divide 30
∴ 3 rejected
Divisors left − 4, − 3, 2
Hence, integrals roots are −3, 2.
Concept Continuation and variation of signs.
Continuation (or Permanence) or of sign
In any polynomial f (x ), whose terms are arranged in order, when a + ve sign follows a + ve sign and a − ve
sign follows a − ve sign, a continuation of sign is said to be occur.
Variation or Change of sign
In any polynomial f (x ), whose terms are arranged in order, when a + ve sign follows a − ve sign or a − ve
sign follows a + ve sign, a variation or a change of sign is said to occur.
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
2x 5 + 7x 4 − 5x 3 − 4x 2 + 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 x 4 − 3x 2 + 5
The signs are + − +
Number of continuation in sign = 0
Number of variation in sign = 2
Number of non-zero terms = 3
∴ 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 ) = x 3 − 3x 2 + 4x − 5 be a complete polynomial.
The sign are + − + −
There are 3 changes of signs and no continuation in sign.
Again f (−x ) = −_ x 3 −_ 3x 2 −_ 4x −_ 5 . There are 3 continuation of sign.
www.pdfworld.in
www.pdfworld.in
[Ist, IInd, IIIrd, IVth] and no changes of sign.

Let f (x ) = x 4 + x 2 + 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.
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
 a a1 a0 
f (x ) = an x n + an − 1x
n −1
+ K + a1x + a 0 = an x n  1 + n − 1 + K + + 
 an x an x n − 1 an x n 
For every large value of | x | , the quantity in the bracket is very near to 1.
∴ f (x )behaves like an x n . Hence, for every large values of | x | , the graph of y = f (x )is similar to the graph of
y = an x n .
It moves up (or down) as x → ± ∞
Thus, for an > 0, we have
(i) an x n → ∞ as x → ± ∞
(ii) an x n → ∞ as x → ∞, if n is odd.
(iii) an x n → − ∞ as x → ∞, if n is odd.
Same is true for f (x ).
Theorem Let f (x )be a real polynomial of degree n ( ≥ 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 polynomialf (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 polynomialf (x )has either one real zero or an even number
of real zeros in (a , b ) .
Corollary 1 Every equation of an odd degree (having + ve leading coefficients) has atleast one real root
of a sign opposite to that of its last term.
Proof Let f (x ) = 0, where
n −1 n −2
f (x ) = an x n + an − 1x + an − 2x + K + a1x + a 0
 a a0 
= an x n 1 + n − 1 + ...... + n 
; an > 0 ;
 an x anx 
n being odd. We have, f (+ ∞ ) = ∞ ; f (0) = a 0 ; f (−∞ ) = − ∞

If a 0 > 0, then there must be one root of f (x ) = 0 in (−∞ , 0) i.e., a real – ve root.
If a 0 < 0, then there must be a root between 0 and ∞ i.e., a real positive root.
www.pdfworld.in
www.pdfworld.in
Corollary 2 Every equation of an even degree whose constant term is – ve has atleast 2 real roots, one
+ ve and other – ve (leading coefficient being + ve).
Proof Let f (x ) = 0, where
n −1
f (x ) = an x n + an − 1x + K + a1x + a 0
 a a1 a0 
= an x n  1 + n − 1 + K + + 
 an x n − 1 an x n 
 an x
Q n is even and an > 0
f (+ ∞ ) = + ∞ ; f (0) = a 0 ; f (−∞ ) = − ∞
∴ a0 < 0
∴ Atleast one root lies between −∞ and 0 and another between 0 and + ∞
There must be atleast one + ve and atleast one –ve root.
Rolle’s Theorem
Between two consecutive real roots of a polynomial equation f (x ) = 0 with real coefficients there lies
atleast one and always an odd number of real roots of the equation f ′ (x ) = 0.
Corollary 1 Between two consecutive roots of f ′ (x ) = 0 there lie almost one real root of f (x ) = 0
Corollary 2 Between any two consecutive real roots a and b of f ′ (x ) = 0 there lies,
(a) no real root of f (x ) = 0, if f (a ) f (b ) > 0
(b) a unique real root of f (x ) = 0, iff f (a ) f (b ) < 0
Method to determine the number of distinct real roots and to locate them.
Let f (x ) be a given real polynomial equation.
1. Solve f ′ (x ) = 0 for real roots.
2. Arrange the real roots of f ′ (x ) = 0 in ascending order say α1 , α 2 , K , αm .
3. Determine signs of f (x ) at −∞ , α1 , α 2, K , αn , ∞.
4. The number of changes of sign in the above sequence of signs determine the number of real roots of
f (x ) = 0 .
Note It follows from above that if f ( x ) = 0 has n real roots, then f ( x ) = 0, can have at most m + 1 real roots.
Q If deg f (x ) = n, then equation f (x ) = 0 has atleast n − (m + 1) = n − m − 1 non-real roots. Also, f ′ (x ) = 0
has (n − 1) − m = n − m − 1 non-real roots.
Hence, the number of non real roots of f (x ) = 0 cannot be less than the number of non-real roots of
f ′ (x ) = 0.
Example 1 Consider the polynomial

+
2x 5 ++ 7x 4 −_ 5x 3 −_ 4x 2 ++ x ++ 5
There are two changes or variation of sign.
[IInd, IIIrd place + − ; IVth, Vth place − +]
Ambiguity When any term of a polynomial f ( x ) has double sign ± or m, an
ambiguity is said to be occur.
Example 2 Let f ( x ) = a 0 + a1 x + a 2x 2 + ... + an x n be a real polynomial of degree n and a 0 ≠ 0. Let

r and s denote the number of variations in signs in f ( x ) and f ( − x ) respectively. Show
that n − r − s is even.
Solution Let f ( x ) = a 0 + a1x + a 2x 2+ ... + an x n
is a real polynomial of degree n.
∴ f ( 0) = a 0 ≠ 0 (given)
www.pdfworld.in
www.pdfworld.in
∴ 0 is not a root of the equation f ( x ) = 0.

∴ Every real root of f ( x ) = 0 is either +ve or –ve.
Let p be the number of + ve roots of f ( x ) = 0 and q be the number of –ve roots
Qf ( x ) is a real polynomial.
∴ Non-real roots of f ( x ) = 0 occur in conjugate pairs
Q Number of non-real roots of f ( x ) = 0 is even say 2k where k is a +ve integer.
∴ p + q + 2k = n ...(i)
[Q degree of f ( x ) = n ∴ f ( x ) = 0 has n roots]
By Descarte's rule of sign p which is the number of +ve roots of f ( x ) = 0 is ≤ r, the
numbers of variations in sign f ( x ) and falls short of it by an even number.
Let p = r − 2t where t is an integer …(ii)
Similarly, q = s − 2l …(iii)
From Eqs. (i), (ii) and (iii), we have
r − 2t + s − 2l + 2k = n
⇒ n − r − s = 2k − 2t − 2l
= 2 [k − t − l ] = 2k1
where k1 = k − t − l = is an integer
Here, n − r − s is even.
Example 3 If a > 0, prove that x 3 + ax + b = 0 has two complex roots.

Solution Case I b>0
Let f ( x ) = x 3 + ax + b
Signs are + , + , + .
Since, there is no change of sign in f ( x ).
∴ f ( x ) = 0 has no + ve root.
Again, f ( − x ) = − x 3 − ax + b
Signs are −, −, −.
Since, there is one change of sign in f ( − x )
∴ f ( x ) = 0 has one – ve root
f ( x ) = 0 has 2 complex roots
Case II b<0
f ( x ) = x 3 + ax + b
Signs are +, +, –.
Q There is one change of sign in f ( x )
∴ f ( x ) = 0 has one + ve root
Again, f ( − x ) = − x 3 − ax + b
Signs are –, –, –.
Since, there is no change of sign in f ( − x )
∴ There is no − ve root of f ( x ) = 0
∴ f ( x ) = 0 has 2 complex roots
www.pdfworld.in
www.pdfworld.in
Case III b = 0
x 3 + ax + b = 0 becomes ax + x 3 = 0
⇒ x (x 2 + a ) = 0
⇒ x =0 or x 2 + a = 0
∴ x + a = 0 gives x = ± a i
2
[Q a > 0]
f ( x ) = 0 has 2 complex roots
Example 4 If f ( x ) = 0 is a complete equation and has all its roots real, then the number of + ve
roots equals the number of variation in signs and the number of −ve roots equals the
number of continuation in sign of f ( x ).
Solution Let p be number of variation in signs and
q be number of continuation in sign of f ( x ).
Let P be number of + ve roots and
N be number of – ve roots of f ( x ) = 0
Let f ( x ) = 0 be the equation of degree n
Q f ( x ) = 0 is a complete equation
Now, p + q =n …(i)
and equation does not have zero as a root.
Q f ( x ) = 0 has all its roots real
So, P + N =n …(ii)
p + q =P + N …(iii)
We shall prove that P = p and N = q
Let it possible P ≠ p. Either P > p or P p, then, number of + ve roots of
[f ( − x ) = 0] > Number of variation in sign of f ( x ).
Which is contrary to Descarte's Rule.
If P q .
⇒ Number of – ve roots of [f ( x ) = 0] > Number of continuations of sign of f ( x ).
⇒ Number of – ve roots of [f ( x ) = 0] > Number of variations in sign of f ( − x ).
Which is contrary to Descarte's Rule.
Hence P = p
Now, from Eq. (iii), p + q = P + N ⇒ q = N
Example 5 Separate the real roots of the equation. 3x 4 + 4x 3 − 6x 2 − 12x + 2 = 0.

Solution f ( x ) = 3x 4 + 4x 3 − 6x 2 − 12x + 2
f ′ ( x ) = 12x 3 + 12x 2 − 12x − 12 = 12 ( x 3 + x 2 − x − 1)

= 12 ( x + 1)( x 2 − 1) = 12 ( x − 1)( x + 1)2
∴ Roots of f ′ ( x ) = 0 are given by 12 ( x − 1)( x + 1)2 = 0
www.pdfworld.in
www.pdfworld.in
⇒ –1, 1 are the only roots of f ′ ( x ) = 0.
x −∞ −1 1 ∞
f (x ) +∞ 7 −9 +∞
(+) (+) (−) (+)
4
Note that for large | x |, f ( x ) behaves as 3x .
∴ f ( x ) = 0 has a unique real root in each of the intervals ( −1, 1) and (1, ∞ ) .
It also shows that f ( x ) = 0 has 2 real roots.
Example 6 Find the interval in which k should lie so that the roots of equation
2x 3 − 9x 2 + 12x − k = 0 are real.
Solution f ( x ) = 2x 3 − 9x 2 + 12x − k
f ′ ( x ) = 6x 2 − 18x + 12 = 6 ( x 2 − 3x + 2) = 6 ( x − 1)( x − 2)
′
Roots of f ( x ) = 0 are given by
6 ( x − 1)( x − 2) = 0
⇒ x = 1, 2 which are real roots of f ′ ( x ) = 0
x −∞ 1 2 +∞
f (x ) −∞ 5 −k 4 −k +∞
(–) (+) (−) (+)
For large values of | x |, f ( x ) behaves like 2x . Since, all roots of f ( x ) = 0.

3
∴ 5 − k > 0 and 4 − k < 0 i.e., k < 5 and 4 < k

i.e., 4 < k < 5 ∴ k ∈( 4, 5)
Example 7 Show that x 5 − 5ax + 4b = 0 has 3 real roots or only one real root, according as
a 5 > or < b 4, a and b are both + ve.
Solution f ( x ) = x 5 − 5ax + 4b
f ′ ( x ) = 5x 4 − 5a = 5 ( x 4 − a ) = 5 ( x 2 − a )( x 2 + a )
∴ The real roots of f ′ ( x ) = 0 are given by
x2 − a = 0
⇒ x = ± a1/ 4
∴ Only real roots of f ( x ) = 0 are −a1/ 4 and a1/ 4
x −∞ − a1/ 4 a1/ 4 +∞
f (x ) −∞ 4a 5/ 4
+ 4b − 4a 5/ 4
+ 4b +∞
(−) (+) (−) +
5
For large | x | , f ( x ) behaves as x .
(i) If −4a 5/ 4 + 4b < 0 i.e., a 5/ 4 > b or a 5 > b 4
then f ( x ) = 0 has 3 real roots one each in
( −∞, − a1/ 4 ); ( −a1/ 4, a1/ 4 ) and (a1/ 4, + ∞ )
(ii) If −4a 5/ 4 + 4b > 0 i.e., a 5 < b 4, then f ( x ) = 0
has only one real root in ( −∞, − a1/ 4 ).
www.pdfworld.in
www.pdfworld.in

Example 1. Solve x (y + z ) = 44 ; y (z + x ) = 50 ; z (x + y ) = 54.
Solution On adding the given equations
2(xy + yz + zx ) = 148
⇒ xy + yz + zx = 74 …(i)
On subtracting given equations one by one from 1, we get
yz = 30, zx = 24, xy = 20
On multiplying the three, given above together.
x 2y 2z 2 = 30 × 24 × 20 ⇒ xyz = ± 120 …(ii)
On dividing Eq. (ii) by the values of xy , yz and zx respectively, we get
x = ± 4, y = ± 5, z = ± 6
Example 2. Solve the system

x (x + y + z ) = a 2 ; y (x + y + z ) = b 2;
z (x + y + z ) = c 2.
Solution On adding the given equations termwise, we get

(x + y + z )2 = a 2 + b 2 + c 2
⇒ x +y + z=± a2 + b2 + c 2
Consequently,
a2 b2
x = ,y =
± a +b +c
2 2 2
± a + b2 + c2
2
c2
z=
± a + b2 + c2
2

x (x + y + z ) = a − yz , y = (x + y + z ) = b − xz
z (x + y + z ) = c − xy .
Solution Given system can be rewritten as (x + z )(x + y ) = a ;
( y + z )( y + x ) = b ; (z + x ) (z + y ) = c
Multiplying these equations and extracting a square root from both members of the obtained
equality, we have
( x + z )( x + y )( y + z ) = ± abc
abc abc
Hence, y + z=± ;x + z=± ;
a b
abc
x +y =±
c
On adding these equalities termwise, we get
www.pdfworld.in
www.pdfworld.in
abc  1 1 1  abc  1 1 1 
x +y + z=±  + +  ∴ x =±  + − 
2 a b c  2 b c a 
abc  1 1 1 
y =±  + − ;
2 a c b 
abc  1 1 1 
z=±  + − 
2 b a c 
Example 4. Solve the system x + y + z = a; x + y + v = b ; x + z + v = c ; y + z + v = d .

Solution Adding all the given equations, we get x + y + z + v = a + b + c + d consequently,
3
v = (x + y + z + v ) − (x + y + z )
a+b+c+d b + c + d − 2a
= −a =
3 3
On likewise, we get
a + c + d − 2b a + b + d − 2c
z= ,y = ,
3 3
a + b + c − 2d
x =
3
Example 5. Solve the system ay + bx = c ; cx + az = b and bz + cy = a.

Solution Dividing the first equation by ab, the second by ac and third by bc (assuming abc ≠ 0), we get
y x c x z b z y a
+ = ; + = ; + =
b a ab a c ac c b bc
Adding all these equations termwise, we find
+ + = 
x y z 1 c b a 
+ + 
a b c 2  ab ac bc 
=  + +  −  +  = 
z x y z x y 1 c b a  c
Hence, + +  −
c a b c  a b  2  ab ac bc  ab
z a2 + b2 − c 2
⇒ =
c 2abc
a2 + b2 − c 2
⇒ z=
2ab
a2 + c 2 − b2 b2 + c 2 − a2
Similarly, y = ,x =
2ac 2bc
Example 6. Solve the system cy + bz = 2dyz ; ax + cx = 2d ′ zx ; bx + ay = 2d ′ ′ xy .

Solution First of all we have an obvious solution x = y = z = 0. Let us now look for non-zero solutions.i . e. ,
for such in which x , y , z are not equal to zero. Dividing the first of the given equations by yz, the second by
zx and third by xy.
c b a c b a
+ = 2d ; + = 2d ′ ; + = 2d ′′
z y x z y x
a b c
Hence, + + = d + d ′ + d ′′
x y z
www.pdfworld.in
www.pdfworld.in
a b
∴ = d ′ + d ′′ − d ; = d + d ′′ − d ′ ;
x y
c
= d + d ′ − d ′′
z
a b
Finally, x = ;y = ;
d′ + d′′ − d d + d′′ − d′
c
z=
d + d ′ − d ′′

xy xz yz
= c; = b; = a.
ay + bx az + cx bz + cy
Solution Rewrite the system in the following way

ay + bx 1 az + cx 1 bz + cy 1
= ; = ; =
xy c xz b yz a
a b 1 a c 1 b c 1
Hence, + = ; + = ; + = .
x y c x z b y z a
Proceeding in the same way as in example 6
2a 2bc 2ab 2c
x = ;y = ;
ac + ab − bc bc + ab − ac
2abc 2
z=
bc + ac − ab
Example 8. Solve the system y + z − x = xyz

2
;
a
xyz xyz
z + x −y = ;x + y − z = .
b2 c2
Solution Obvious solution is x = y = z = 0.
Dividing both members of each equation of our system by xyz, we get
1 1 1 1 1 1 1 1
+ − = ; + − = ;
xz xy yz a 2 xy yz xz b 2
1 1 1 1
+ − =
yz xz xy c 2
On adding pairwise, we find
2 1 1 2 1 1 2 1 1
= + ; = + ; = +
xy a 2 b 2 yz b 2 c 2 xz a 2 c 2
2a 2b 2 2b 2c 2
Consequently, xy = ; yz = ,
a +b
2 2
b2 + c2
2a 2c 2
xz =
a2 + c 2
On multiplying the equalities, we get
8a 4 b 4 c 4
x 2y 2z 2 =
(a + b )(b 2 + c 2 )(a 2 + c 2 )
2 2
www.pdfworld.in
www.pdfworld.in
2 2 a 2b 2c 2
⇒ xyz = ±
(a 2 + b 2 )(b 2 + c 2 )(a 2 + c 2 )
2a 2b 2
Using the equality xy =
a2 + b2
We find for z two values which differ in the sign. By the obtained value of z we find corresponding
values of y and x from the equalities. Thus, we get 2 sets of values for x , y and z satisfying our
equation.

z + ay + a 2x + a3 = 0
z + by + b 2x + b3 = 0
z + cy + c 2x + c 3 = 0.
Solution The given equations show that the polynomial α3 + xα 2 + yα + z vanishes at three different
values of α namely at α = a, α = b and at α = c (assuming that a , b , c are not equal to one another).
Set up a difference
α3 + xα 2 + yα + z − (α − a )(α − b )(α − c )
This difference also becomes zero at α equal to a , b , c .
Expanding this expression in powers of α, we get
(x + a + b + c ) α 2 + ( y − ab − ac − bc ) α + z + abc
This second degree trinomial vanishes at three different values of α and therefore it equals zero
identically and consequently, all its coefficients are equal to zero. i . e. ,
x + a + b + c = 0 ; y − ab − ac − bc = 0 ;
z + abc = 0
Hence, x = − (a + b + c )
y = ab + ac + bc
z = − abc
is the solution of our system.

x1 + x 2 + x3 + K + xn = 1
x1 + x3 + K + xn = 2
x1 + x 2 + x 4 + K + xn = 3
…………………………
x1 + x 2 + K + xn −1 =n
Solution Let x 1 + x 2 + x 3 + K + x n = s = 1
Then s − x 2 = 2 ;
s − x 3 = 3, K , s − x n −1 = n − 1,
s − xn = n
Consequently, (since s = 1)
x 2 = − 1, x 3 = − 2, K , x n = − (n − 1)
Hence,
n (n − 1)
x 2 + x 3 + K + x n = − [1 + 2 + K + (n − 1)] = −
2
n (n − 1)
Finally, x 1 = 1 − (x 2 + x 3 + K + x n ) = 1 +
2
Similarly, we can find other values.
www.pdfworld.in
www.pdfworld.in

x sin a + y sin 2a + z sin 3a = sin 4a
x sin b + y sin 2b + z sin 3b = sin 4b
x sin c + y sin 2c + z sin 3c = sin 4c
Solution We have, sin 2a = 2 sin a cos a
sin 3a = sin a (4 cos2 a − 1)
sin 4a = 4 sin a (2 cos3 a − cos a )
∴ The first equation of our system is rewritten in the following way
x + 2y cos a + z (4 cos2 a − 1) = 4(2 cos3 a − cos a )
The remaining two are similar. Expand this equation in powers of cos a, we have
8cos3 a − 4z cos2 a − (2y + 4)cos a + z − x = 0
Putting cos a = t and dividing both members by 8, we get
z y +2 z −x
t3 − t 2 − t + =0
2 4 8
Our system of equations is equivalent to the statement that the equations has three roots : t = cos a ;
t = cos b and t = cos c , which follows :
z
= cos a + cos b + cos c
2
y +2
= − (cos a cos b + cos a cos c + cos b cos c )
4
x −z
= cos a cos b cos c
8
∴ The solution of our system will be
x = 2(cos a + cos b + cos c ) + 8(cos a cos b cos c )
y = − 2 − 4(cos a cos b + cos a cos c + cos b cos c )
z = 2(cos a + cos b + cos c )

x 2 = a + ( y − z )2
y 2 = b + ( x − z )2
z 2 = c + ( x − y )2
Solution Reduce the system to the following form

(x + y − z ) (x + z − y ) = a
(y + z − x ) (y + x − z ) = b
(x + z − y ) (z + y − x ) = c
Multiplying and taking a square root, we get
( x + y − z )( x + z − y )( y + z − x ) = ± abc
bc ac ab
Further, y + z −x = ± x + z −y = ± x +y −z=±
a b c
1  ac ab  1  bc ab  1  bc ac 
Consequently, x =±  + , y = ±  + , z = ±  + 
2 b c  2 a c  2 a b 
www.pdfworld.in
www.pdfworld.in

x 2 + y 2 + xy = c 2
z 2 + x 2 + xz = b 2
y 2 + z 2 + yz = a 2
Solution Subtracting the equations term-by-term, we have

(x − y )(x + y + z ) = b 2 − a 2
(x − z )(x + y + z ) = c 2 − a 2
Put x + y + z = t , then
(x − y ) t = b 2 − a 2 ; (x − z ) t = c 2 − a 2
On adding these two equations termwise, we have
[3x − (x + y + z )] t = b 2 + c 2 − 2a 2
t 2 + b 2 + c 2 − 2a 2
Hence, x =
3t
t 2 + a 2 + c 2 − 2b 2
y =
3t
t 2 + a 2 + b 2 − 2c 2
z=
3t
Substituting these values of x , y , z in one of the equations, we find
t 4 − (a 2 + b 2 + c 2 ) t 2 + a 4 + b 4 + c 4 − a 2b 2 − a 2c 2 − b 2c 2 = 0
Hence,
(a 2 + b 2 + c 2 ) ± 3(a + b + c )(−a + b + c )
(a − b + c )(a + b − c )
t =
2
2
Knowing t, we obtain the values of x , y , z.
Example 14. Solve x 2 + y 2 + z 2 = 14

xy + yz + zx = 11
x + y + 2z = 9
Solution We have, x 2 + y 2 + z 2 = 14 …(i)
xy + yz + zx = 11 …(ii)
x + y + 2z = 9 …(iii)
Adding double of Eq. (ii) in Eq. (i), we get
x 2 + y 2 + z 2 + 2xy + 2yz + 2zx = 36
or (x + y + z )2 = 36
or x +y + z=± 6 …(iv)
Subtracting Eq. (iv) from Eq. (iii)
z = 9 ± 6 = 3, 15
z = 3, x + y = 3 [from Eq. (iii)]
z = 15, x + y = − 21 [from Eq. (iii)]
Also from Eq. (ii), xy + z (x + y ) = 11
xy + 3 × 3 = 11
www.pdfworld.in
www.pdfworld.in
or xy = 2
when (x + y ) = 3, xy = 2
∴ x = 1, 2 ; y = 2, 1
From Eq. (ii), when x + y = − 21
xy + 15(−21) = 11
xy = 326
− 21 ± − 863 − 21 ± − 863
∴ x = ; y =
2 2
− 21 ± − 863
Hence, x = 1,2,
2
− 21 ± − 863
y = 2,1,
2
z = 3, 3, 15
Example 15. Solve the equations

x + y + z = 14
x 2 + y 2 + z 2 = 91
y 2 = zx .
Solution We have x + y + z = 14 …(i)

x + y + z = 91
2 2 2
…(ii)
y = zx
2
…(iii)
Squaring the Eq. (i), we get
x 2 + y 2 + z 2 + 2xy + 2yz + 2zx = 196
Putting the values of x 2 + y 2 + z 2 from Eq. (ii) and of zx from Eq. (iii), we get
91 + 2xy + 2yz + 2y 2 = 196
2xy + 2yz + 2y 2 = 105
2y (x + y + z ) = 105
105 15
2y (14) = 105 or y = =
28 4
15
Hence, x + z = 14 − [from Eq. (i)]
4
41
or x + z=
4
2
zx =  
15
Also, [from Eq. (ii)]
 4
225
or zx =
16
Hence, x and z are the roots of the equation.
41 225
t 2 − (x + z ) t + (xz ) = 0 or t 2 − t + =0
4 16
164 ± (164)2 − 4 × 16 × 225
or 16t 2 − 164t + 225 = 0 ⇒ t =
32
www.pdfworld.in
www.pdfworld.in
41 ± 412 − 900 41 ± (41 + 30)(41 − 30)

⇒ l= =
8 8
41 ± 71 × 11 41 ± 781
= =
8 8
41 ± 781
Hence, x =
8
15
y =±
4
41 ± 781
z=
8
Example 16. Solve the equations x +y + z=9

x + y 2 + z 2 = 29
2
x 3 + y 3 + z3 = 99
Solution We have x +y + z=9 …(i)

x + y + z = 29
2 2 2
…(ii)
x + y + z = 99
3 3 3
…(iii)
On squaring both sides of Eq. (i),
x 2 + y 2 + z 2 + 2(xy + yz + zx ) = 81
or 2(xy + yz + zx ) = 52 [from Eq. (ii)]
or xy + yz + zx = 26
Also, x 3 + y 3 + z3 − 3xyz = (x + y + z )(x 2 + y 2 + z 2 − xy − yz − zx )
∴ 99 − 3xyz = 9(29 − 26)
or 3xyz = 72 or xyz = 24 …(iv)
Now, xy + yz + zx = 26
24
or + z (x + y ) = 26 [from Eq. (iv)]
z
24
or + z (9 − z ) = 26 [from Eq. (i)]
z
or z3 − 9z 2 + 26z − 24 = 0
or z = 2, 3, 4
Since, the equation is symmetrical.
Hence, x = 2, 3, 4
y = 3, 4, 2
z = 4, 2, 3.
Example 17. Solve the equations 4x (x + 2y + 4z ) = 21

21
y (x + 2y + 4z ) =
2
z (x + 2y + 4z ) = 21
Solution 4x (x + 2y + 4z ) = 21 …(i)
21
y (x + 2y + 4z ) = …(ii)
2
www.pdfworld.in
www.pdfworld.in
z (x + 2y + 4z ) = 21 …(iii)
4x
From Eqs. (i) and (ii) = 2 or y = 2x
y
4x
From Eqs. (i) and (iii) = 1 or z = 4x
z
Substituting the values of y and z in Eq. (i), we get
4x (x + 4x + 16x ) = 21
21
or 84x 2 = 21 or x 2 =
84
x = ± 1/ 2
y =±1
z=±2
Example 18. Determine the values x 1 , x 2 , x 3 , x 4 , x 5 satisfying x 5 + x 2 = yx 1 ; x 1 + x 3 = yx 2 ; x 2 + x 4 = yx 3 ;

x 3 + x 5 = yx 4 ; x 4 + x 1 = yx 5 where y is a given parameter.
Solution x 5 + x 2 = yx 1 …(i)
x 1 + x 3 = yx 2 …(ii)
x 2 + x 4 = yx 3 …(iii)
x 3 + x 5 = yx 4 …(iv)
x 4 + x 1 = yx 5 …(v)
Express x 5 and x 3 from Eqs. (i) and (ii)
x 5 = yx 1 − x 2 …(vi)
x 3 = yx 2 − x 1 …(vii)
Substitute Eq. (vi) into Eq. (v), we get
x 4 = (y 2 − 1)x 1 − yx 2 …(viii)
After substituting Eqs. (viii) and (vii) into Eq. (iii) and ordering the equation
(y 2 + y − 1) (x 1 − x 2 ) = 0 …(ix)
Substituting Eqs. (vi), (vii) and (viii) into Eq. (iv), we get
( y 2 + y − 1)(( y − 1) x 1 − x 2 ) = 0 …(x)
If y + y − 1 = 0,
2
−1 ± 5
i . e. , y = , then Eqs. (ix) and (x) is satisfied for arbitrary x 1 and x 2 ; since all our previous
2
transformations are reversible, these values uniquely determine x 3 , x 4 , x 5. If
y 2 + y − 1 ≠ 0,
Then Eqs. (ix) and (x) gives x 1 − x 2 = 0
( y − 1) x 1 − x 2 = 0 … (xi)
Implying ( y − 2) x 1 = 0 …(xii)
For y = 2 the value of x 1 = x 2 can be arbitrary, if x 1 = x 2 = c , then x 3 = x 4 = x 5 = c
Finally, if y ≠ 2, then x 1 = 0 from Eq. (xii) and x 2 = 0 from Eq. (xi) implying that all unknowns in the
original system are equal to zero.
www.pdfworld.in
www.pdfworld.in
Example 19. Solve log 2 x + log 4 y + log 4 z = 2

log3 y + log 9 z + log 9 x = 2
log 4 z + log16 x + log16 y = 2
Solution log 2 x + log 4 y + log 4 z = 2
⇒ (log 4 x ) log 2 4 + log 4 y + log 4 z = 2
⇒ x 2yz = 16 [Q log 2 4 = 2]
Similarly, the remaining equations reduce to
y 2zx = 81 and z 2xy = 256
Solving x 2yz = 16; xy 2z = 81 and xyz 2 = 256,
we get
x = 2 / 3; y = 27 / 8 ; z = 32 / 3
Example 20. If a + b + c = 1 ; a 2 + b 2 + c 2 = 9 and a3 + b3 + c 3 = 1 , find 1 + 1 + 1.

a b c
Solution Since,
(a + b + c )2 = a 2 + b 2 + c 2+ 2(bc + ca + ab )
∴ bc + ca + ab = [12 − 9] / 2 = − 4 …(i)
Also, a + b + c − 3abc
3 3 3
= (a + b + c )(a 2 + b 2 + c 2 − bc − ca − ab ) = 1(9 − (−4))

a3 + b3 + c 3 − 3abc = 13 …(ii)
So that abc = (1 − 13) / 3 = − 4 …(iii)
1 1 1 (bc + ca + ab )
Now, + + = = (−4) / (−4) = 1
a b c abc
Example 21. Solve the following system of equations for real x , y , z

x + y − z = 4; x 2 − y 2 + z 2 = − 4; xyz = 6.
Solution Given equations can be written as

x − z = 4 − y ; x 2 + z 2 = y 2 − 4; xz = 6 / y
By using the identity (x − z )2 + 2xz = x 2 + z 2, we eliminate x and z from the above equations, so as to
get
12
(4 − y )2 + = y2 − 4
y
⇒ y (4 − y )2 + 12 = y ( y 2 − 4)
⇒ 2y 2 − 5y − 3 = 0
⇒ y = − 1/ 2 or 3
When y = − 1 / 2 ; x + z = − 15 / 4 which is not possible for any real values ofx and z.
2 2
When y = 3,x − z = 1,xz = 2, so that x and −z are the roots of t 2 − t − 2 = 0, giving t = 2 or −1i . e. , either
x = 2, z = 1
or x = − 1, z = − 2
www.pdfworld.in
www.pdfworld.in

x + y + z = ab
x −1 + y −1 + z −1 = a −1b
xyz = a3.
Solution x + y + z = ab …(i)
1 1 1 b
+ + = …(ii)
x y z a
xyz = a3 …(iii)
By Eqs. (ii) and (iii), we have
xy + yz + zx = a 2b …(iv)
Now, by Eqs. (i), (iii) and (iv), it is clear thatx , y , z are roots of
t 3 − abt 2 + a 2bt − a3 = 0 …(v)
We see that the above equations vanishes for t = a, i . e. , (t − a ) is a factor of Eq. (v).
So, by remainder theorem, Eq. (v) is
t 2 (t − a ) + at (t − a ) + a 2 (t − a ) − abt (t − a ) = 0
or (t − a )(t 2 + at + a 2 − abt ) = 0
or (t − a ){t 2 + t (a − ab ) + a 2} = 0
i . e. , either t = a,
− (a − ab ) ± [(a − ab )2 − 4a 2 ]
or t =
2
i . e. , x , y , z are
1
a, a [b − 1 + b 2 − 2b − 3 ],
2
1
a [b − 1 − b 2 − 2b − 3 ] respectively.
2

x 3 + y 3 + z3 = 495; x + y + z = 15; xyz = 105.
Solution x 3 + y 3 + z3 = 495 …(i)

x + y + z = 15 …(ii)
xyz = 105 …(iii)
From Eqs. (i) and (iii)
x 3 + y 3 + z3 − 3xyz = (x + y + z )(x 2 + y 2 + z 2 − xy − yz − zx )
or 495 − 315 = 15 [x 2 + y 2 + z 2 − xy − yz − zx ]
180
or = [(x + y + z )2 − 3(xy + yz + zx )]
15
12 = [225 − 3(xy + yz + zx )]
or xy + yz + zx = 71
xy + yz + zx = 71 …(iv)
From Eq. (iv), x ( y + z ) + yz = 71
105
or x (15 − x ) + = 71 [using Eq. (ii)]
x
www.pdfworld.in
www.pdfworld.in
x 2 (15 − x ) + 105 − 71x = 0

x 3 − 15x 2 + 71x − 105 = 0
This equation is satisfied when x = 3, 5, 7
∴ By symmetry y = 5, 7, 3
z = 7, 3, 5
Example 24. Eliminate x , y , z , u from the equations

x = by + cz + du ; y = cz + du + ax
z = du + ax + by ; u = ax + by + cz.
Solution Now, x + ax = ax + by + cz + du
y + by = ax + by + cz + du
z + cz = ax + by + cz + du
and u + du = ax + by + cz + du
If ax + by + cz + du = k (say)
Then, x + ax = k
k
⇒ x =
1+a
k
Similarly, y = ;
1+b
k k
z= and u =
1+c 1+d
Substituting these values in
ax + by + cz + du = k , we get
ak bk ck dk
+ + + =k
1+a 1+b 1+c 1+d
a b c d
or + + + =1
1+a 1+b 1+c 1+d
Example 25. Eliminate x , y , z from the equations

x + y + z = 0 ; x 2 + y 2 + z 2 = a 2,
x 3 + y 3 + z3 = b3 and x 5 + y 5 + z5 = c 5.
Solution Given, x +y + z=0 …(i)

x + y + z =a
2 2 2 2
…(ii)
x + y + z =b
3 3 3 3
…(iii)
x + y + z =c
5 5 5 5
…(iv)
From Eq. (i) we can conclude
x 3 + y 3 + z3 = 3xyz ⇒ b3 = 3xyz, [using Eq. (iii)]
Squaring first equation
x 2 + y 2 + z 2 + 2(xy + yz + zx ) = 0
www.pdfworld.in
www.pdfworld.in
or a 2 + 2(xy + yz + zx ) = 0
a2
∴ xy + yz + zx = − …(ii)
2
Now, multiplying Eqs. (ii) and (iii), we get
(x 2 + y 2 + z 2 )(x 3 + y 3 + z3 ) = a 2b3
x 5 + y 5 + z5 + x 2y 3 + x 2z3 + y 2z3 + y 2x 3 + z 2x 3 + z 2x 3 = a 2b3
or c 5 + x 2y 2 (x + y ) + y 2z 2 ( y + z ) + z 2x 2 (z + x ) = a 2b3
or c 5 + x 2y 2 (− z ) + y 2z 2 (−x ) + z 2x 2 (−y ) = a 2b3 [Qx + y + z = 0]
or c − x y z − xy z − yx z = a b
5 2 2 2 2 2 2 2 3
or c 5 − xyz (xy + yz + xy ) = a 2b3

From Eqs. (i) and (ii) it becomes
b3  a 2 
c5 − −  =a b
2 3
3  2
or 6c 5 + a 2b 2 = 6a 2b3 or 6c 5 = 5a 2b3
Example 26. Eliminate x , y from the equations

x + y = a ; x 2 + y 2 = b2; x 4 + y 4 = c 4.
Solution Q x 2 + y 2 = (x + y )2 − 2xy
∴ b 2 = a 2 − 2xy ∴ 2xy = a 2 − b 2
Now, x 4 + y 4 = (x 2 + y 2 )2 − 2x 2y 2
2
a2 − b2 
or c4 = b4 − 2  
 2 
⇒ 2c 4 = 2b 4 − (a 4 + b 4 − 2a 2b 2 )
or b 4 − a 4 − 2c 4 + 2a 2b 2 = 0

x y  y z z x
+ + = a ; + + = b ;  +   +   +  = c .
x y z x y z
y z x z x y y z  z x  x y 
Solution Multiplying the given first two equations

x y z x y z x2 xy
ab =  + +   + +  = 3 +
y z x z x y
∑ yz + ∑ z 2 …(i)
x y  y z z x
Also, c =  +   +   + 
y z  z x  x y 
x z y2 y   z x 
⇒ c= + + 2 +   +  ⇒ c =1
z y z x  x y 
x 2 z 2 zx y 2 xy yz
c =1+ + + 2 + + 2 + 2 +1
yz xy y zx z x
www.pdfworld.in
www.pdfworld.in
x2 xy
⇒ c =2+ Σ + Σ 2 …(ii)
yz z
Hence, from Eqs. (i) and (ii)
ab − 3 = c − 2
ab = c + 1

x 2 (y + z ) y 2( z + x ) z 2 (x + y ) xyz
3
= 3
= = = 1.
a b c3 abc
Solution Given equations can be written as
x 2 ( y + z ) = a3 …(i)
y (z + x ) = b
2 3
…(ii)
z (x + y ) = c
2 3
…(iii)
xyz = abc …(iv)
From Eqs. (i), (ii) and (iii)
x 2y 2z 2 ( y + z ) ( z + x ) ( x + y ) = a3 × b3 × c 3
a 2b 2c 2 ( y + z ) ( z + x ) ( x + y ) = a3b3c 3
Using Eq. (iv),
∴ abc = (y + z ) (z + x ) (x + y )
= ( yz + yx + z 2 + zx ) (x + y )
= xyz + x 2y + xz 2 + x 2z + y 2z + y 2x + yz 2 + xyz
= 2xyz + x 2 ( y + z ) + y 2 ( z + x ) + z 2 ( x + y ) = 2abc + a3 + b3 + c 3
Hence, a3 + b3 + c 3 + abc = 0

( x + y − z )( x − y + z ) = ayz ;
( y + z − x )( y − z + x ) = bzx ;
(z + x − y )(z − x + y ) = cxy .
Solution Given equations are
( x + y − z ) ( x − y + z ) = ayz …(i)
( y + z − x ) ( y − z + x ) = bzx …(ii)
( z + x − y ) ( z − x + y ) = cxy …(iii)
Multiplying Eqs. (i), (ii), (iii), we get
(x + y − z )2 ( x − y + z )2 ( y + z − x )2 = abcx 2y 2z 2
or (−x 3 − y 3 − z3 + y 2z + yz 2 + z 2x + zx 2 + x 2y + xy 2 − 2xyz )2= abcx 2y 2z 2 …(iv)
2 2 2
On dividing both sides of Eq. (iv) by x y z
 x 2 y2 z2 y
+ + + − 2
z z x x y
or abc =  − − − + + + …(v)
 yz zx xy x x y y z z 
But Eq. (i) may be written as

x 2 − y 2 − z 2 + 2yz x 2 y z
a= = − − +2
yz yz z y
www.pdfworld.in
www.pdfworld.in
x2 y z
or a −2= − −
yz z y
y2 z x
Similarly, b −2= − −
zx x z
z2 x y
and c −2= − −
xy y x
Now, from Eq. (v)
abc = [(2 − a ) + (2 − b ) + (2 − c ) − 2]2
abc = (4 − a − b − c )2
Example 30. Show that

(a + b + c )3 − 4(b + c )(c + a )(a + b ) + 5abc = 0
is the eliminant of ax 2 + by 2 + cz 2
= ax + by + cz = yz + zx + xy = 0.
ax 2 + by 2 + cz 2 = 0 …(i)
ax + by + cz = 0 …(ii)
yz + zx + xy = 0 …(iii)
Multiplying Eq. (ii) by (x + y + z ), we have
(ax + by + cz )(x + y + z ) = 0
or ax + by + cz + xy (a + b ) + yz (b + c ) + zx (c + a ) = 0
2 2 2
But ax 2 + by 2 + cz 2 = 0
∴ xy (a + b ) + yz (b + c ) + zx (c + a ) = 0
Also, xy + yz + zx = 0
xy yz zx 1
Hence, = = = (say)
b −a c −b a −c k
Dividing each ratio by xyz

1 1 1 1
= = =
z (b − a ) x (c − b ) y (a − c ) k
k k k
∴ x = ,y = ,z =
(c − b ) (a − c ) (b − a )
Substituting these values in Eq. (ii)
k k k
a +b +c =0
c −b a −c b −a
a (b − a )(a − c ) + b (c − b )(b − a ) + c (c − b )(a − c ) = 0
or a3 + b3 + c 3 − (a + b )(b + c )(c + a ) + 5abc = 0
or a3 + b3 + c 3 + 3(a + b )(b + c )(c + a ) −4(a + b )(b + c )(c + a ) + 5abc = 0
or (a + b + c )3 − 4(a + b )(b + c )(c + a ) + 5abc = 0
www.pdfworld.in
www.pdfworld.in
Example 31. Eliminate a , b , c from the system

x y z 2
= = ; a + b 2 + c 2 = 1;
a b c
a + b + c = 1.
x y z 1
Solution Put = = =
a b c λ
Then, a = xλ ; b = yλ ; c = zλ …(i)
Now, (a + b + c ) = a + b + c + 2ab + 2ac + 2bc
2 2 2 2
…(ii)
Q a + b + c = 1 and a + b + c = 1 2 2 2
(given)
On putting these values in Eq. (ii), we get
ab + bc + ac = 0
Taking into the consideration equalities Eq. (i), we get
xy + xz + yz = 0
Example 32. Eliminate x , y , z from the system

y 2 + z 2 − 2ayz = 0 ;
z 2 + x 2 − 2bxz = 0 ;
x 2 + y 2 − 2cxy = 0.
y z z x
Solution We have, + = 2a ; + = 2b ;
z y x z
x y
and + = 2c
y x
Squaring these equalities and adding them, we get
y2 z2 z2 x2 x2 y2
2
+ 2
+ 2
+ 2
+ 2
+ + 6 = 4a 2 + 4b 2 + 4c 2
z y x z y x2
On multiplying these equalities, we get
y2 z2 z2 x2 x2 y2
2
+ 2
+ 2
+ 2
+ 2
+ + 2 = 8abc
z y x z y x2
Consequently the result of eliminating x , y , z from the given system is
a 2 + b 2 + c 2 − 2abc = 1
Example 33. Eliminate x , y between the equations

x 2 − y 2 = px − qy ; 4xy = qx + py ; x 2 + y 2 = 1.

x 2 − y 2 = px − qy …(i)
4xy = qx + py …(ii)
x 2 + y2 = 1 …(iii)
Multiplying the Eq. (i) by x and Eq. (ii) by y , we get
x 3 − xy 2 = px 2 − qxy …(iv)
and 4xy = py + qxy
2 2
…(v)
www.pdfworld.in
www.pdfworld.in
From Eqs. (iv) and (v)

x 3 + 3xy 2 = p (x 2 + y 2 )
Hence, by Eq. (iii)
p = x 3 + 3xy 2 ; q = 3x 2y + y 3
p + q = (x + y )3 ; p − q = (x − y )3
2 2
∴ (p + q )3 + (p − q )3 = (x + y )2 + (x − y )2 = 2(x 2 + y 2 )
∴ (p + q )2/3 + (p − q )2/3 = 2
Example 34. Eliminate x , y , z between the equations

y 2 + z 2 = ayz ; z 2 + x 2 = bzx ; x 2 + y 2 = cxy .
y z z x x y
Solution We have, + =a; + =b; + =c
z y x z y x
Multiplying together these three equations
y2 z2 z2 x2 x2 y2
2+ 2
+ 2
+ 2
+ 2
+ 2
+ = abc
z y x z y x2
2 + (a 2 − 2) + (b 2 − 2) + (c 2 − 2) = abc
∴ a 2 + b 2 + c 2 − 4 = abc
Example 35. Eliminate x , y , z between the equations

y z z x x y
− = a ; − = b ; − = c.
z y x z y x
Solution We have,
x (y 2 − z 2 ) + y ( z 2 − x 2 ) + z (x 2 − y 2 )
a+b+c=
xyz
(y − z ) ( z − x ) (x − y )
=
xyz
If we change the sign of x , the sign of b and c are changed, while the sign of a remains unaltered.
(y − z ) ( z + x ) (x + y )
Hence, a −b −c =
xyz
( y + z ) (z − x ) ( x − y )
b −c −a =
xyz
(y + z ) ( z + x ) (x − y )
c −a −b =
xyz
∴ (a + b + c )(b + c − a )(c + a − b )(a + b + c )
− ( y 2 − z 2 )2 (z 2 − x 2 )2 (x 2 − y 2 )2
=
x 4y 4 z 4
2 2 2
y z z x x y 
= −  −   −   − 
 z y  x z y x 
= − a 2b 2c 2
∴ 2b 2c 2 + 2c 2a 2 + 2a 2b 2 − a 4 − b 4 − c 4 + a 2b 2c 2 = 0
www.pdfworld.in
www.pdfworld.in
Example 36. If α , β , γ be the roots of the cubic ax 3+ 3bx 2+ 3cx + d = 0. Prove that the equation in y whose
βγ − α 2 γα − β 2 αβ − γ 2
roots are ; ; ; is obtained by the transformation axy + b (x + y ) + c = 0. Hence,
β + γ − 2α γ + α − 2β α + β − 2γ
form the equation with above roots.
Solution Qα , β , γ are the roots of the equation
ax 3 + 3bx 2 + 3cx + d = 0 …(i)
3b 3c
∴ α+β+ γ=− ; αβ + βγ + γα = ;
a a
d
αβγ = −
a
αβγ d
− α2 − − α2
βγ − α 2 α
Now, y = = = aα
β + γ − 2α (α + β + γ ) − 3α − 3b − 3α
a
d + aα3 d + ax 3
= =
3α (b + aα ) 3x (b + ax )
∴ 3xy (b + ax ) = d + ax 3
or ax 3 − 3ayx 2 − 3byx + d = 0 …(ii)
3(b + ay ) x 2 + 3(c + by ) x = 0
(b + ay ) x + c + by = 0 [Qx ≠ 0]
axy + b (x + y ) + c = 0
Which is the required transformation.
Now, (ay + b ) x = − (by + c )
by + c
∴ x =−
ay + b
Putting this value of x in Eq. (i), we get
3 2
 by + c   by + c   by + c 
−a   + 3b   − 3c  +d =0
 ay + b   ay + b   ay + b 
or a (by + c )3 − 3b (by + c )2 (ay + b ) + 3c (by + c )(ay + b )2 − d (ay + b )3 = 0
Which is the required equation.
Example 37. If α , β , γ are the roots of the equation x 3 + 2x 2 + 3x + 1 = 0. Form an equation whose roots
1 1 1 1 1 1 1 1 1
are + − ; + − ; + − .
β3 γ3 α3 α3 γ 3 β3 α3 β3 γ 3
Solution Roots of the equation
x 3 + 2x 2 + 3x + 1 = 0 …(i)
are α , β , γ. Let us form an equation, whose roots are α , β , γ . If y is a root of the transformed
3 3 3
equation, then
y = x3 …(ii)
To eliminate x between Eqs. (i) and (ii)
Eq. (i) can be written as x 3 + 1 = − (2x 2 + 3x )
www.pdfworld.in
www.pdfworld.in
On cubing both sides, we get

x 9 + 3x 6 + 3x 3 + 1 = − [8x 6+ 27x 3 + 18x 3(2x 2 + 3x )]
x 9 + 3x 6 + 3x 3 + 1 = − [8x 6 + 27x 3 + 18x 3 (−x 3 − 1)]
Putting x 3 = y in this equation
y 3 + 3x 2 + 3y + 1 = − 8y 2 − 27y + 18y 2 + 18y

or y 3 − 7y 2 + 12y + 1 = 0 …(iii)
Its roots are α3 , β3 , γ 3

Changing y to 1/y, Eq. (iii) becomes
1 7 12
3
− 2
+ +1=0
y y y
or y + 12y 2 − 7y + 1 = 0
3
…(iv)
1 1 1
Its roots are , , .
α3 β3 γ 3
Let us denote them by a , b , c .
∴ a + b + c = − 12
We have to form an equations whose roots are
1 1 1 1 1 1 1 1 1
+ − ; + − ; + −
β3 γ 3 α3 γ 3 α3 β3 α3 β3 γ 3
i . e. , whose roots are b + c − a , c + a − b, a + b − c where a , b , c are the roots of Eq. (iv). If the new
equation is in terms of z, then
z = b + c − a = (a + b + c ) − 2a = − 12 − 2y
12 + z
∴ y =−
2
Putting this value of y in Eq. (iv), we have
(12 + z )3 (12 + z )2 (12 + z )
− + 12 ⋅ + 7⋅ + 11 = 0
8 4 2
(12 + z )3 (12 + z )2 (12 + z )
− 12 −7 − 11 = 0
8 4 2
(12 + z )3 − 24 (12 + z )2 − 28 (12 + z ) − 88 = 0
or (12 + z )3 − 24 (12 + z )2 − 28 (12 + z ) − 88 = 0
or z3 + 12z 2 − 172z − 2152 = 0
Which is required equation.
Example 38. If x 1 , x 2 , x 3 are the roots of x 3 − x 2+ 4 = 0, form the equation whose roots are x 1 + x 22 + x 32 ;
x 2 + x 32 + x 12 ; x 3 + x 12 + x 22.
Solution Q x 1 , x 2 , x 3 are the roots of equation

x3 − x 2 + 4 = 0 …(i)
∴ x 1 + x 2 + x 3 = 1 ; x 1x 2 + x 2x 3 + x 3x 1 = 0 ;
x 1x 2x 3 = − 4
www.pdfworld.in
www.pdfworld.in
If the transformed equation is in terms of y, then

y = x 1 + x 22 + x 32 = x 1 + (x 2 + x 3 )2 − 2x 2x 3
2x 1x 2x 3 8
= x 1 + (1 − x 1 )2 − = x 1 + (1 − x 1 )2 +
x1 x1
8 8
∴ y = x + (1 − x )2 + =x2 −x + 1 +
x x
or x 3 − x 2 + x − xy + 8 = 0 …(ii)
4
xy − x − 4 = 0 ⇒ x =
y −1
64 16
− + 4=0
( y − 1)3 ( y − 1)2
⇒ ( y − 1)3 − 4( y − 1) + 16 = 0
or y 3 − 3y 2 − y + 19 = 0
Example 39. If α , β , γ are the roots of the cubic x 3 + px 2 + qx + r = 0, find the value of ( β + γ ) ( γ + α )
(α + β ) in terms of p , q , r .
Solution Roots of the equation
x 3 + px 2 + qx + r = 0 …(i)
are α , β , γ.
Let us first form an equation whose roots are
β + γ , y + α , α + β.
Let y be a root of the transformed equation, then
y =β + γ =α + β + γ −α
= −p −α (Qα + β + γ = − p)
= −p −x
∴ x = − (y + p )
Putting this value of x in Eq. (i), we have
− ( y + p )3 + p ( y + p )2 − q ( y + p ) + r = 0
or y 3 + 2py 2 + ( p 2 − q ) y + ( pq − r ) = 0 …(ii)
Its roots are β + γ , γ + α , α + β.
∴ ( β + γ ) (γ + α ) ( α + β )
= product of the roots of Eq. (ii) = − (pq − r ) = r − pq
Example 40. If α is a root of equation x 4 + ax 3 − 6x 2− ax + 1 = 0, then show that 1 + α is also a root.

1−α
Hence, show that the other two roots are
−1 α − 1
, .
α α+1
Solution Since, α is a root.

∴ α 4 + aα3 − 6 α 2 − aα + 1 = 0 …(i)
www.pdfworld.in
www.pdfworld.in
1+α
Putting x = , we get
1−α
[(1 + α )4 + (1 − α )4 ] + a [1 + α ][1 − α ]
[1 + α ]2 − [1 − α ]2 − 6 [(1 + α )2 (1 − α )2 ] = 0
1+α
On simplification it reduces to Eq. (i). Hence, is also a root. Again if we replace x by −1 / x , we
1−α
get the same equation.
1 1 + (−1 / α ) α −1
∴ If α is a root, then –1/α is also a root and if − is a root, then i . e. , is also a root.
α 1 − (−1 / α ) α+1
Example 41. If α , β , γ be the roots of x 3 + 2x 2 − 3x − 1 = 0.

1 1 1
Find the value of + + .
α3 β3 γ3
Solution Roots of the equation,
x 3 + 2x 2 − 3x − 1 = 0 …(i)
are α , β , γ.
Let us first form an equation whose roots are α3 , β3 , γ 3.
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
x 3 − 1 = − (2x 2 − 3x )
On cubing both sides of above equations

⇒ x 9 − 3x 6 + 3x 3 − 1 = − [ 8x 6 − 27x 3− 18x 3 (2x 2 − 3x )]
⇒ x 9 − 3x 6 + 3x 3 − 1 = − [8x 6 − 27x 3 − 18x 3(1 − x 3 )]
⇒ x 9 − 3x 6 + 3x 3 − 1 = − 8x 6 + 27x 3 + 18x 3 − 18x 6
⇒ x 9 + 23x 6 − 42x 3 − 1 = 0
Putting x 3 = y in this equation, we get
y 3 + 23y 2 − 42y − 1 = 0 …(iii)
Its roots are α , β , γ .

3 3 3
1
Changing y to , Eq. (iii) becomes
y
1 23 42
+ − −1= 0
y3 y2 y
y 3 + 42y 2 − 23y − 1 = 0 …(iv)
1 1 1
Its roots are , , .
α3 β3 γ 3
1 1 1
∴ + + = sum of roots of Eq. (iv) = − 42
α3 β3 γ3
www.pdfworld.in
www.pdfworld.in
Example 42. Find the value of k for which the roots of the kx 3 + 2x 2 − 3x + 1 = 0 are in HP.
Solution Let α , β , γ be the roots of
kx 3 + 2x 2 − 3x + 1 = 0 …(i)
which are in HP.
1
Transform the given equation whose roots are in AP. replace x by .
x
x 3 − 3x 2 + 2x + k = 0 …(ii)
Let the roots of Eq. (ii) be
a − d , a, a + d
Sum of roots, 3a = 3
⇒ a =1
Since, a = 1 is one of the root of Eq. (ii)
∴ It will satisfy the equation.
∴ We get, 1−3+ 2+ k = 0
⇒ k=0
Which is a contradiction as the given equation is a degree three polynomial.
∴There is no value of k for which the roots are in HP.
Example 43. If α , β , γ be the roots of the cubic equation x 3 + 3x + 2 = 0, find the equation whose roots are
(α − β ) (α − γ ), ( β − γ ) ( β − α ), (γ − α ) (γ − β ). Hence, show that the above cubic has two imaginary roots.
Solution Let z = ( α − β )( α − γ ) = α 2 − αβ − αγ + βγ
2αβγ
= α 2 − Σαβ +
α
or αz = α3 − 3α + 2(−2) (Q Σαβ = 3 ; αβγ = − 2) …(i)
Also, α + 3α + 2 = 0, we get
3
α 2 − 3α = − 6 α − 2
On putting this value in Eq. (i)
αz = − 6α − 6 or α(z + 6) = − 6
6
∴ α=−
z+ 6
But α is a root of x 3 + 3x + 2 = 0
3
 6   6 
∴ −  + 3 −  +2=0
 z + 6  z + 6
(z + 6)3 − 9(z + 6)2 − 108 = 0
z3 + 9z 2 − 216 = 0
Let z1 , z2 , z3 be the roots of above equation, then
z1z2z3 = ( α − β ) ( α − γ ) ( β − γ ) ( β − α ) (γ − α )
( γ − β ) = 216
or − ( α − β ) ( β − γ ) ( γ − α )2 = 216
2 2
( α − β )2 ( β − γ )2 ( γ − α )2 = 216
www.pdfworld.in
www.pdfworld.in
Hence, one of the factors in RHS must be –ve say ( α − β )2 is –ve. i.e., α − β = pure imaginary,
showing that α and β are conjugate complex.
Hence, the given equation has two imaginary roots.
Example 44. Find the remainder when (x + 1)n is divided by (x − 1)3.

Solution Let x − 1 = y (so that x = y + 1)
We find that ( x + 1) = ( y + 2)
n n
−1 n (n − 1) 2 n −2
= 2n + ny 2n + y ⋅2 +K
2
The remainder when RHS is divided by y 3
n (n − 1) 2 n − 2
= 2n + ny ⋅ 2n − 1 + y 2
2
n (n − 1)
= 2n + n (x − 1)2n − 1 + (x − 1)2 ⋅ 2n −2
2
−3 −2 −1 −3 −1
= n (n − 1) ⋅ 2n x 2 + x [− n (n − 1) ⋅ 2n + n ⋅ 2n ] + n (n − 1) ⋅ 2n − n ⋅ 2n + 2n
−3 −2 −3
= n (n − 1) ⋅ 2n x 2 − 2n x [n 2 − 3n ] + 2n [n 2 − 5n + 8]
Example 45. Show that (x − 1)2 is a factor of x n − nx + n − 1.

Solution Let f (x ) ≡ x n − nx + n − 1
Since, f (x ) = 0
∴ f (x ) is divisible by x − 1 [by factor theorem]
so, f (x ) = (x n − 1) − n (x − 1)
−1 −2
= (x − 1) {x n + xn + K + 1 − n} = (x − 1) g (x )
n −1 n −2
where g (x ) = x +x + K+ 1−n
Since, g(1) = 0
∴ x − 1 is a factor of g (x ). [by factor theorem]
∴ f (x ) is divisible by (x − 1)2
+1
Example 46. Show that (x − 1)2 is a factor of x n − x n − x + 1.
+1
Solution xn − x n − x + 1 = (x n − 1)(x − 1)
−1 −2
= (x − 1)2 (x n + xn + K + 1)
It is clear that
+1
(x − 1)2 is a factor of x n − xn − x + 1
Example 47. f (x ) is a polynomial of degree atleast two with integer coefficients. Show that which it is
 f (a ) − f (b ) af (b ) − bf (a )
divided by (x − a ) (x − b ) where a ≠ b , the remainder is x  + .
 a −b  a −b
Solution Since, the divisor is a polynomial of degree 2 the remainder will be of the form Ax + b.
By division algorithm, we have
f (x ) = q (x ) (x − a ) (x − b ) + Ax + B …(i)
where q (x ) is the quotient.
www.pdfworld.in
www.pdfworld.in
Substituting x = a , b successively in Eq. (i), we get f (a ) = Aa + B ; f (b ) = Ab + B.

Solving for A and B, we have
A = [f (a ) − f (b )] / (a − b )
B = [af (b ) − bf (a )] / (a − b )
Example 48. Let p (x ) = x 2 + ax + b be a quadratic polynomial in which a and b are integers. Given any
integer n, show that there is an integer M such that p (n ) p (n + 1) = p (M ).
Solution Let the zeros of p (x ) be α , β so that p (x ) = (x − α ) (x − β ).
Then, p (n ) = (n − α ) (n − β ),
p (n + 1) = (n + 1 − α ) (n + 1 − β )
We have to show that p (n )p (n + 1) can be written as (t − α ) (t − β ) for some integer t (which will depend
upon n)
p (n ) p (n + 1) = (n − α ) (n − β ) (n + 1 − α ) (n + 1 − β )
= {(n − α ) (n + 1 − β )} {(n − β ) (n + 1 − α )}
= {n (n + 1) − n (α + β ) − α + α β} × {n (n + 1) − n (α + β ) − β + α β}
= {n (n + 1) + na + b − α} {n (n + 1) + na + b − β}
= (t − α ) (t − β ); t = n (n + 1) + an + b = p (t )
Thus, p (n ) p (n + 1) can be written as p (M ) for
M = n (n + 1) + an + b
Example 49. A polynomial f (x ) with rational coefficients leaves remainder 15, when divided by x − 3 and
remainder 2x + 1, when divided by (x − 1)2. Find the remainder when f (x ) is divided by (x − 3) (x − 1)2.
Solution Let quotient be q (x ) and remainder be r (x ) when f (x ) is divided by (x − 3) (x − 1)2.

Now, as divisor is a polynomial of degree 3 the remainder must be a polynomial of degree at most 2
i.e., it must be of the form ax 2 + bx + c ; a , b , c are some rational numbers.
ax 2 + bx + c = a [(x − 1) + 1]2 + b [(x − 1) + 1] + c

= a (x − 1)2 + (2a + b ) (x − 1) + a + b + c
By division algorithm,
f (x ) ≡ q (x ) (x − 3) (x − 1)2 + a (x − 1)2 + (2a + b ) (x − 1) + a + b + c ...(i)
Now, according to given condition f (x ) leaves a remainder 15 when divided by x − 3
∴ f (3) = 15
Now, putting x = 3 in Eq. (i), we have
15 = 9a + 3b + c …(ii)
Also, from Eq. (i), we find that remainder when f (x ) is divided by (x − 1) is (2a + b ) (x − 1) + (a + b + c ).
2
Since, this is given to be 2x + 1, we have

(2a + b ) (x − 1) + (a + b + c ) = 2x + 1
Putting x = 1, we get
a + b + c =3 …(iii)
Putting x = 0 throughout, we get
−a + c =1 …(iv)
www.pdfworld.in
www.pdfworld.in
From Eqs. (ii), (iii) and (iv), we get

a = 2, b = − 2, c = 3
∴ Remainder = ax 2 + bx + c
= 2x 2 − 2x + 3
Example 50. For every pair p , q of + ve integers, whose HCF is 1. Show that
(x p − 1) (x q − 1) divides (x pq − 1) (x − 1)
Solution Since, y − 1 divide y n −1 for every + ve integers n.

∴ By writing x pq − 1 as (x p )q − 1 we find that x p − 1 divides x pq − 1. Similarly, x q − 1 also divides
x pq − 1, where p and q are prime to each other
∴ GCD of x p − 1 and x q − 1 is x − 1.
Consequently, x p − 1 = (x − 1) f (x )
and x q − 1 = (x − 1) g (x )
where f (x ), g (x ) have no common factor.
Now, f (x ), g (x ), (x − 1) have no factor in common and each of them divides x pq − 1
∴ (x − 1) f (x ) g (x ) divides x pq − 1.
Consequently, (x − 1)2 f (x ) g (x ) divides (x pq − 1) (x − 1).
i.e., (x p − 1) (x q − 1) divides (x pq − 1) (x − 1).
Example 51. Prove that the polynomial

x 9999 + x 8888 + x 7777 + ...... + x 1111 + 1
is divisible by x + 1.
Solution Let
M = x 9999 + x 8888 + x 7777 + ... + x 1111 + 1
and N = x 9 + x 8 + x 7 + ... + x 1 + 1
M − N = x 9 (x 9990 − 1) + x 8 (x 8880 − 1) + x 7 (x 7770 − 1) + ... + x (x 1110 − 1)
= x 9 [(x 10 )999 − 1] + x 8 [(x 10 )888 − 1] + x 7 [(x 10 )777 − 1] + ... + x [(x 10 )111 − 1] …(i)
Now, (x 10 )n is divisible by x 10 − 1, ∀ n ≥ 1
RHS of Eq. (i) is divisible by x 10 − 1
∴ M − N is divisible by x 10 − 1
and hence divisible by x 9 + x 8 + ... + 1.
Example 52. If n is an odd + ve integers not divisible by 3. Show that xy (x + y ) (x 2 + xy + y 2 ) is a factor of

(x + y )n − x n − y n .
Solution We have, xy (x + y ) (x 2 + y 2 + xy )
= xy (x + y ) (x − ωy ) (x − ω2y )
[ω, ω2 are non-real cube roots of unity]
It is enough to show that (x + y )n − x n − y n vanishes for x = 0 ; y = 0.
Now, x = − y ; x = ω y and x = ω2y
www.pdfworld.in
www.pdfworld.in
The polynomial obviously vanishes for x = ωy

( ω y + y )n − (ω y )n − y n = y n [( ω + 1)n − ωn − 1]
= y n [(− ω2 )n − ωn − 1] = − y n [( ω + 1)n − ωn − 1]
= y n [(− ω2 )n − ωn − 1] = − y n [ ω2n + ωn + 1] (Q n is odd)
3p + 1
Let n = 3p + 1, then ω = ω
n
and ω2n = ω6p + 2 = ω2

∴ Above expression = − y n [ω2 + ω + 1] = 0
If n = 3p + 2, then ωn = ω2 , ω2n = ω and the above expression is zero. We can similarly prove that the
given polynomial vanishes for x = ω2 y . If n is an odd +ve integer not divisible by 3, then
(x + y )n − x n − y n is divisible by xy (x 2 + xy + y 2 ).
Example 53. If g (x ) and h (x ) are polynomials with real coefficients and f (x ) = g (x 3 ) + x h (x 3 ) is divisible by
x 2 + x + 1. Show that g (x ) and h (x ) are both divisible by x − 1.
Solution If f (x ) = g (x 3 ) + xh (x 3 ) is divisible by x 2 + x + 1. We can write

f (x ) = q (x )(x 2 + x + 1)
Let ω be a non-real cube root of unity. Then,
f ( ω) = q ( ω) ( ω2 + ω + 1) = 0
f ( ω2 ) = q ( ω2 ) ( ω + ω2 + 1) = 0
∴ g (1) + ωh (1) = 0 and g (1) + ω2 h (1) = 0
So, g (1) = h (1) = 0
Both g (x ) and h (x ) are divisible by x − 1.
Example 54. Let z be a root of x 5 − 1 = 0 with z ≠ 1. Find the value of z15 + z16 + z17 + K + z50
Solution z15 + z16 + ... + z50
 z36 − 1   z − 1
= z15 (z1 + .... + z35 ) = 1   = 1  =1
 z − 1   z − 1
Example 55. Let f (x ) be a polynomial leaving the remainder A when divided by (x − a ) and remainder B
when divided by x − b (a ≠ b ). Find remainder left by this polynomial when divided by (x − a ) (x − b ).
Solution Since, the products (x − a ) (x − b ) is a second degree trinomial when divided by it, the polynomial
f (x ) will necessary leave a remainder which is a first degree polynomial inx , a α + β.
Thus, there exists the following identity.
f (x ) = (x − a ) (x − b ) Q (x ) + α x + β
It only remains to determine α + β. Putting in this identity first x = a and then x = b , we get
f (a ) = αa + β ; f (b ) = αβ + β
Remainder from dividing f (x ) by x − a is equal to f (a ) , so f (a ) = A ; f (b ) = B
Thus, for determining α and β, we get the following system of two equations in two unknowns
αa + β = A ; αβ + β = B
1 aB − bA
Hence, α= (A − B ) and β =
a −b a −b
www.pdfworld.in
www.pdfworld.in
Example 56. Find out at what values of p and q the binomial x 4 + 1 is divisible by x 2 + px + q ?
Solution Let us suppose
x 4 + 1 = (x 2 + px + q ) (x 2 + p ′ x + q ′ )
= x 4 + (p + p ′ ) x 3 + (q + q ′ + pp ′ ) x 2 (pq ′ + qp ′ ) x + qq ′
For determining p , q , p ′ and q′ we have four equations

p + p′ = 0 …(i)
pp ′ + q + q ′ = 0 …(ii)
pq ′ + qp ′ = 0 …(iii)
qq′ = 1 …(iv)
From Eqs. (i) and (iii), we find p ′ = − p (q ′ − q ) = 0
Assume
Case I p = 0, p ′ = 0, q + q ′ = 0, qq ′ = 1, q 2 = − 1
q = ± i ;q ′ = ± i
The corresponding factorization has the form
x 4 + 1 = (x 2 + i ) (x 2 − i )
Case II q ′ = q , q 2 = 1, q = ± 1
Suppose first q ′ = q = 1, then pp′ = − 2, p + p ′ = 0, p 2 = 2, p = ± 2 , p′ = ± 2
The corresponding factorization is
x 4 + 1 = (x 2 − 2x + 1) (x 2 + 2x + 1)
Assume, then
q = q ′ = − 1 ; p + p ′ = 0 ; pp ′ = 2 ; p = ± 2 i
p′ = ± 2 i
Factorization will be
x 4 + 1 = (x 2 + 2x i − 1) (x 2 − 2x i − 1)
Example 57. Prove that

(i) the polynomial x (x n −1 − nan − 1 ) + an (n − 1) is divisible by (x − a )2.
(ii) the polynomial (1 − x n ) (1 + x ) − 2nx n (1 − x ) − n 2x n (1 − x )2 is divisible by (1 − x )3.
Solution (i) Rewrite the polynomial as

−1 −1
x n − an − nxan + nan = (x n − an ) − nan (x − a )
−1 −2 −1 −1
= (x − a ) (x n + ax n + K + an − nan )
At x = a, the second factor of the last product vanishes and consequently is divisible byx − a.
∴ Given polynomial is divisible by (x − a )2
(ii) Let us denote polynomial by Pn and set up the difference Pn − Pn − 1. Transforming this
difference, we easily prove that it is divisible by (1 − x )3.
Since, it is true for any + ve integer n we obtain a number of equalities
Pn − Pn −1 = (1 − x )3 φ n (x )
www.pdfworld.in
www.pdfworld.in
Pn − 1 − Pn −2 = (1 − x )3 φ n − 1(x )
………………………………
………………………………
………………………………
………………………………
P3 − P2 = (1 − x )3 φ 2 (x )
P2 − P1 = (1 − x )3 φ 1 (x )
Where φ 1 (x ) are polynomials with respect to x .
Hence, Pn − P1 = (1 − x )3 φ (x )
But since P1 = (1 − x )3
It follows that Pn is divisible by (1 − x )3 and our proposition is proved.
Example 58. Find out whether the polynomial

x 4 a + x 4b + 1 + x 4c + 2
+ x 4d + 3
(a , b , c , d are + ve integers) is divisible by x 3 + x 2 + x + 1.
Solution Put f (x ) = x 4 a + x 4b + 1 + x 4c + 2
+ x 4d + 3
x 3 + x 2 + x + 1 = (x + 1) (x 2 + 1) = (x + 1) (x + i ) (x − i )
It only remains to show that

f (− 1) = f (i ) = f (− i ) = 0
Try yourself
Example 59. Show that the expression.

(x + y + z )m − x m − y m − zm (m is odd) is divisible by (x + y + z )3 − x 3 − y 3 − z3. .
Solution It is known that

(x + y + z )3 − x 3 − y 3 − z3 = 3 (x + y ) (x + z ) (y + z ).
Let us prove that (x + y + z )m − x m − y m − zm is divisible by x + y . Considering our polynomial

rearranged in powers of x . We put in it x = − y .
We have, (− y + y + z )m − (− y )m − y m − zm = 0 [Q m is odd]
Consequently, our polynomial is divisible by (x + y ). Likewise we make sure that it is divisible by

(x + z ) and (y + z ).
Example 60. Prove that the polynomial (cos φ + x sin φ)n − cos n φ − x sin n φ is divisible by x 2 + 1.
Solution Put
f (x ) = (cos φ + x sin φ)n − cos n φ − x sin φ
But x 2 + 1 = (x + i ) (x − i )
f (i ) = (cos φ + i sin φ)n − (cos n φ + i sin n φ) = 0
Likewise we make sure that f (− i ) = 0
Our supposition is proved.
www.pdfworld.in
www.pdfworld.in
−2
Example 61. Find out at what n the polynomial 1 + x 2 + x 4 + K + x 2n is divisible by the polynomial
n −1
1+x +x +x 2
.
−2 x 2n − 1
Solution 1 + x 2 + x 4 + K + x 2n =
x2 −1
−1 xn − 1
1 + x + x 2 + K + xn =
x −1
It is required to find out at what n
 x 2n − 1   x n − 1
 2    will be a polynomial in x .
x −1  x −1 
 x 2n − 1   x n − 1 x n + 1
we find,  2    =
x −1  x −1  x +1
For x n + 1 to be divisible by x + 1, it is necessary and sufficient that (− 1)n + 1 = 0 i.e., n is odd.
−2 −1
Thus, 1 + x 2 + K + x 2n is divisible by 1 + x + x 2 + K + x n , if n is odd.
Example 62. Find the condition necessary and sufficient for x 3 + y 3 + z3 + kxyz to be divisible by
x + y + z.
Solution The condition necessary and sufficient for a polynomial f (x ) to be divisible by x − a, consists in
that f (a ) = 0
Let f (x ) = x 3 + kxyz + y 3 + z3
For this polynomial to be divisible byx + y + z, it is necessary and sufficient that
f (− y − z ) = 0
However,
f (−y − z ) = − ( y + z )3 − kyz ( y + z ) + y 3 + z3
= − ( k + 3) yz ( y + z )
Simplifying, we get k = − 3
Thus, for x 3 + y 3 + z3 + kxyz to be divisible by x + y + z it is necessary and sufficient that k = − 3.
Example 63. Solve the system of equation for real x and y

 1   1 
5x  1 + 2  = 12 ; 5y  1 − 2  = 4.
 x + y2  x + y2
122 42
Solution (5x )2 + (5y )2 = 2
+ 2
 1   1 
1 + 2  1 − 2 
 x + y 2
  x + y 2

1
Put x 2 + y 2 = , we have
t
25 144 16
= +
t (1 + t )2 (1 − t )2
So that 25(1 − t 2 )2 = 144t (1 − t )2 + 16t (1 + t )2
or 25t 4 − 160t 3 − 206t 2 − 160t + 25 = 0
www.pdfworld.in
www.pdfworld.in
Dividing by t 2 , we get
160 25
25t 2 − 160t + 206 − + 2 =0
t t
i . e. ,  2 1  1
25  t + 2  − 160  t +  + 206 = 0
 t   t
1
Let t + = u , we have
t
25(u 2 − 2) − 160u + 206 = 0
i . e. , 25u 2 − 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 x + y =
2 2
or 5
5 5
1
When x 2 + y 2 = , we get
5
5x (1 + 5) = 12, 5y (1 − 5) = 4,
2 1
so that x = ,y = −
5 5
When x 2 + y 2 = 5, we get
5x  1 +  = 12 ; 5y  1 −  = 4,
1 1
 5  5
so that x = 2, y = – 1
2 1
Thus, x = , y = − and x = 2, y = − 1
5 5
Example 64. Prove that the equation

5x 2 + 5y 2 − 8xy − 2x − 4y + 5 = 0
is not satisfied by any pair of real numbers x and y.
Solution Rewriting the given equation as a quadratic in y
5y 2 − y (8x + 4) + 5x 2 − 2x + 5 = 0 …(i)
Solving Eq. (i), we have
(8x + 4) ± (8x + 4)2 − 20(5x 2 − 2x + 5)
y =
10
(8x + 4) ± 2 (−9x 2 + 26x − 21)
= …(ii)
10
Now, the expression under the radical sign in Eq. (ii) can never be +ve and therefore y cannot take a
real value for any real value of x. In fact
 2
20 
−9x 2 + 26x − 21 = − 9   x −
13 
 − 
  9  81 
Which is –ve, whatever real value x may have.
Given equation is not satisfied by any pair of real numbers x and y.
www.pdfworld.in
www.pdfworld.in
Example 65. Prove that if the coefficients of the quadratic equation ax 2 + bx + c = 0 are odd integers,
then the roots of the equation cannot be rational numbers.
Solution Suppose p is a root of quadratic equation ax 2 + bx + c = 0. Then,
q
2
p  p 
a   + b   + c = 0, i . e. , ap 2 + bpq + cq 2 = 0.
q  q 
If p and q are both odd, then all the terms of the expression ap 2 + bpq + cq 2 are odd and so the
expression cannot be equal to zero. If one of p and q is odd and other is even, then two of the terms
are even. One term is odd.
∴ The expression cannot be equal to zero.
Also, p and q cannot be both even as p and q are prime to each other.
Hence proved.
Example 66. Given that x 4 + px 3 + qx 2+ rx + s = 0 has four real +ve roots. Prove that
(i) pr − 16s ≥ 0 (ii) q 2 − 36s ≥ 0 with equality in each case holds if and only if four roots are equal.
Solution Let the roots of the equation.

x 4 + px 3 + qx 2 + rx + s = 0 be α , β , γ , δ so that
α > 0, β > 0, γ > 0, δ > 0
Now, Σα = − p
Σαβ = q
Σαβγ = − r
αβγδ = s
(i) pr = Σα Σαβγ
By the inequality of the means
1
Σα ≥ (αβγδ)1/ 4 …(i)
4
1
Σαβγ ≥ (αβγδ)3/ 4 …(ii)
4
1
Σα Σαβγ ≥ αβγδ
16
i . e. , pr − 16s ≥ 0 …(iii)
Equality hold in Eq. (iii) ⇔ Equalities hold in both Eqs. (i) and (ii) ⇔ α , β , γ , δ are all equal and αβγ ,
αβδ, αγδ, βγδ are all equal ⇔ α = β = γ = δ .
(ii) q 2 = (Σαβ )2 ≥ [6(αβ ⋅ αγ ⋅ αδ ⋅ βγ ⋅ βδ . γδ)1/ 6 ]2 = 36αβγδ = 36s
∴ q 2 ≥ 36s
Equality holds if and only if αβ , αγ , αδ, βγ , βδ, γδ are all equal, i . e. , if and only if α = β = γ = δ .
Example 67. Let a , b , c , d be four real numbers, not all equal to zero. Prove that the zeros of the
polynomial f (x ) = x 6 + ax 3 + bx 2 + cx + d cannot be all real.
Solution Let us suppose that all the 6 roots of the given equation are real.
Let us denote the roots by α , β , γ , δ, λ and µ Σα = 0, Σαβ = 0
∴ Σα 2 = (Σα 2 ) − 2Σαβ = 0
www.pdfworld.in
www.pdfworld.in
Q α 2 + β 2 + γ 2 + δ 2 + λ2 + µ 2 = 0 and α , β… are all real.

∴ α =β = γ =δ= λ =µ = 0
Consequently we must have a = b = c = d = 0 which is impossible, since we are given a , b , c , d are all
not zero.
Hence, roots of given equation cannot be all real.
Example 68. Prove that if a1 , a2 , K , an are all distinct, then the polynomial (x − a1 )2 (x − a2 )2… (x − an )2 + 1
can never be written as the product of two polynomials with integer coefficients.
Solution Suppose that there exists polynomial f (x ) ⋅ g (x ), with integer coefficients such that
f (x ) g (x ) = (x − a1 )2 (x − a 2 )2 K (x − an )2 + 1 …(i)
Q RHS is always +ve.
∴ f (x ) can never vanish
So, its sign never changes.
Similarly, g (x ) can never vanish and its sign never changes
Q f (x ), g (x ) are always + ve, so f (x ) and g (x ) are both always +ve,
Substituting x = a1 , a 2, … , an in Eq. (i), we get
f (a1 ) g (a1 ) = 1, f (a 2 ) g (a 2 ) = 1 , ... , f (an ) g (an ) = 1
Q f (a1 ), K , f (an ) are all +ve integers. It follows that
f (a1 ) = f (a 2 ) = K = f (an ) = 1
Similarly, g (a1 ) = g (a 2 ) = K = g (an ) = 1
Q f (x ) − 1, g (x ) − 1 vanish when x = a1 , a 2 , K , an
∴ f (x ) − 1 = p (x )(x − a1 )(x − a 2 ) K (x − an )
By Factor theorem
g (x ) − 1 = q (x )(x − a1 )(x − a 2 )K (x − an )
p (x ), q (x ) are polynomials with integer coefficients.
Q f (x ) g (x ) is of degree 2n , p (x ) and q (x ) must be both constants. Suppose p (x ) = a , q (x ) = b.
Then, f (x ) = a (x − a1 ) (x − a 2 )... (x − an ) + 1
g (x ) = b (x − a1 ) (x − a 2 )... (x − an ) + 1
Substituting f (x ) and g (x ) in Eq. (i), we get
ab = 1, a + b = 0
Q There do not exist any real numbers a , b satisfying these conditions (these conditions imply
a 2 = − 1, b 2 = − 1).
∴ There is a contradiction and given polynomial cannot be expressed as the product of two
polynomials with integer coefficients.
Example 69. If p (x ) is a polynomial with integer coefficients and a , b , c are three distinct integers, then
show that it is impossible to have p (a ) = b , p (b ) = c , p (c ) = a.
Solution Suppose it is possible that p (a ) = b , p (b ) = c and p (c ) = a.
Q p (x ) − p (b ) vanishes, when x = b
∴ (x − b ) must be a factor of p (x ) − p (b ) .
a − b must divide p (a ) − p (b ) i . e. , b − c
Similarly, b − c must divide a − b.
∴ b − c = ± (a − b )
Q a , b , c are all distinct,
www.pdfworld.in
www.pdfworld.in
So, we cannot have

b − c = − (a − b )
Consequently, b −c =a −b
1
This means that b = (a + c ) so that b must lie between a and c. Similarly, we conclude that, c lies
2
between a and b. It is obvious that both the conclusion cannot be simultaneous true.
Hence, our supposition must be wrong.
It cannot be possible that p (a ) = b , p (b ) = c and p (c ) = a.
Example 70. Let f (x ) be a polynomial with integer coefficients and suppose that for five distinct integers
a1 , a 2 , a3 ,a 4 , a5 , one has f (a1 ) = f (a 2 ) f (a3 ) = f (a 4 ) = f (a5 ) = 2 . Show that there does not exist an integer b such
that f (b ) = 9.
Solution Suppose there exist an integer b such that f (b ) = 9
Q f (x ) − f (b ) = 0, when x = b
∴ x − b must divide f (x ) − f (b )
a1 − b must divide f (a1 ) − f (b )
We are given that f (a1 ) − f (b ) = − 7
∴ (a1 − b )| (−7)
Q Only divisors of –7 are −7, − 1, + 1, + 7
∴ a1 − b must have one of these values.
Similarly, a 2 − b , a3 − b , a 4 − b , a5 − b must take one of the values ± 1, ± 7.
At least some two of the five numbers ai − b , i = 1, 2, K , 5 must be equal.
This is not possibleQai are all distinct.
f (x ) ≠ 9 for any integer x = b.
Example 71. Solve in R the equation

2x 99 + 3x 98 + 2x 97 + 3x 96 + K + 2x + 3 = 0.
Solution 2x 99 + 3x 98 + 2x 97 + 3x 96 + K + 2x + 3
= (2x + 3)(x 98 + x 96 + x 95 + K + 1)
 x 98 − 1 
= (2x + 3)  
 x −1 
The equation x 98 − 1 = 0 has only two real roots i.e., ± 1.

3
∴ The given equation has only 2 real roots i.e., − and –1.
2
Example 72. If x , y , z are three real numbers such that x + y + z = 4 and x 2 + y 2 + z 2 = 6. Show that each
of x , y , z lies in the closed interval  , 2 i.e., ≤ x ≤ 2, ≤ y ≤ 2 , ≤ z ≤ 2. Can x attain extreme value or 2 ?
2 2 2 2 2
 3  3 3 3 3
Solution Rewriting the given equation in the form
y + z = 4 − x , y 2 + z 2 = 6 − x 2, we get
1
yz = [( y + z )2 − ( y 2 + z 2 )] = x 2 − 4x + 5
2
www.pdfworld.in
www.pdfworld.in
∴ y , z must be the roots of equation.

t 2 − (4 − x ) t + x 2 − 4x + 5 = 0 …(i)
Q y , z are real.
Discriminant of Eq. (i) must be non-negative, i.e.,
(4 − x )2 − 4(x 2 − 4x + 5) ≥ 0
⇔ 3x 2 − 8x + 4 ≤ 0
⇔ (3x − 2)(x − 2) ≤ 0
2
⇔ ≤x ≤2
3
Given equations are symmetrical in x , y , z.
2 2
∴ We must have ≤ y ≤ 2, ≤ z ≤ 2
3 3
2 10 25
when x = , Eq. (i) becomes t 2 − t + =0
3 3 9
So that t = , . We have solution  , ,  . Consequently x take the value .

5 5 2 5 5 2
3 3  3 3 3 3
Similarly, we find that when x = 2, Eq. (i) reduces to t 2 − 2t + 1 = 0 so that t = 1, 1.
We have, solution (2, 1, 1) and we conclude that x can take the extreme value 2 as well.
Example 73. Determine x , y , z ∈ R such that

2x 2 + y 2 + 2z 2 − 8x + 2y − 2xy + 2xz − 16z + 35 = 0
Solution 2x 2 + y 2 + 2z 2 − 8x + 2y − 2xy + 2xz − 16z + 35 = 0
⇒ (x − y )2 + (x + z )2 + z 2 − 16z − 8x + 2y + 35 = 0
⇒ (x − y − 1)2 + (x + z − 3)2 + z 2 − 10z + 25 = 0
⇒ (x − y − 1)2 + (x + z − 3)2 + (z − 5)2 = 0
Thus, x − y = 1, x + z = 3, z = 5
Hence, x = − 2, y = − 3
Solution is x = − 2, y = − 3 and z = 5
Example 74. Find all real x , y that satisfying x 3 + y 3 = 7 and x 2 + y 2 + x + y + xy = 4.

Solution Let x + y = α, xy = β and hence x 2 + y 2 = α 2 − 2β
(x 3 + y 3 ) = (x + y )(x 2 − xy + y 2 )
⇒ α (α 2 − 3 β ) = 7
⇒ α3 − 3αβ = 7 …(i)
and x + y + x + y + xy = 4
2 2
⇒ α 2 − 2β + α + β = 4
⇒ α2 − β + α = 4
⇒ β = α2 + α − 4 …(ii)
www.pdfworld.in
www.pdfworld.in
From Eqs. (i) and (ii), we have

α3 − 3α (α 2 + α − 4) = 7
f (α ) = 2α3 + 3α 2 − 12α + 7 = 0
⇒ f (1) = 2 + 3 − 12 + 7 = 0
Hence, (α − 1) is a factor.
So, f (α ) = 2α3 + 3α 2 − 12α + 7 = 0
= (α − 1)(2α 2 + 5α − 7) = 0
= (α − 1)(α − 1)(2α + 7) = 0
7
So, α = 1 or −
2
7 19
When α = 1, β = − 2 and when α = − , β =
2 4
If we take α = 1, β = − 2, then x and y are roots of
t2 + t − 2 = 0
⇒ (t + 2) (t − 1) = 0
⇒ t = − 2 and 1
i.e., x = − 2 and y = 1 or x = 1 and y = − 2
7 19
If we take α = − and β = , then x , y are roots of 4t 2 + 14t + 19 = 0
2 4
Here, discriminant 142 − 4 × 4 × 19 < 0. There are no real roots.
Real values of x , y satisfying the given equation are (2, − 1) or (−1, 2).
Example 75. If x 1 and x 2 are non-zero roots of the equation ax 2 + bx + c = 0 and − ax 2 + bx + c = 0

a 2
respectively. Prove that x + bx + c = 0 has a root between x 1 and x 2.
2
Solution If x 1 and x 2 are roots of
ax 2 + bx + c = 0 …(i)
− ax + bx + c = 0
2
…(ii)
We have,
ax 12 + bx 1 + c = 0 −ax 22 + bx 2 + c = 0
a 2
Let f (x ) = x + bx + c
2
a
Thus, f (x 1 ) = x 12 + bx 1 + c …(iii)
2
a
f (x 2 ) = x 22 + bx 2 + c …(iv)
2
1
Adding ax 12 in Eq. (iii), we get
2
1
f (x 1 ) + ax 12 = ax 12 + bx 1 + c = 0
2
1
⇒ f (x 1 ) = − ax 12 …(v)
2
www.pdfworld.in
www.pdfworld.in
3
Subtracting ax 22 from Eq. (iv), we get
2
3
f (x 2 ) − ax 22 = − ax 22 + bx 2 + c = 0
2
3
⇒ f (x 2 ) = ax 22
2
Thus, f (x 1 ) and f (x 2 ) have opposite signs.
Hence, f (x ) must have a root between x 1 and x 2.
z x y
Example 76. Find all +ve integers x , y , z satisfying x y ⋅ y z ⋅ z x = 5xyz.
z x y
Solution x , y , z are integers and 5 is a prime number and given equation is x y ⋅ y z ⋅ z x = 5xyz
Dividing both sides of the equation by xyz .
z x y
−1 −1 −1
xy ⋅y z ⋅ zx
So, different possibilities are
z z z
−1 −1 −1
xy =5 xy =1 xy =1
x x x
−1 −1 −1
yz =1 or yz =5 or yz =1
y y y
−1 −1 −1
zx =1 zx =1 zx =1
Taking Ist column

x = 5, y z − 1 = 1 ; y z = 2, y = 2 and z = 1
and these values are satisfying the other expressions in first column.
Similarly, from IInd column, we get
y = 5, z = 2 and x = 1
From IIIrd column, we get z = 5, x = 2 and y = 1

f (x ) = x 1000 − x 500 + x 100 + x + 1 = 0 has no rational roots.
Solution If there a rational root, let it be p , where (p , q ) = 1, q ≠ 0. Then, q should divide the coefficient of
q
the leading term and p should divide the constant term.
Thus, q |1 ⇒q = ± 1
and p |1 = p = ± 1
p
Thus, =±1
q
p
If the root = 1, then
q
f (1) = 1 − 1 + 1 + 1 + 1 = 3 ≠ 0
So, 1 is not a root.
p
If = − 1, then f (−1) = 1 ≠ 0
q
Hence, –1 is not a root.
Thus, there exists no rational roots for given polynomial.
www.pdfworld.in
www.pdfworld.in
Example 78. Find all integers x for which

x 4 + x 3 + x 2 + x + 1 is a perfect square.
Solution If x 4 + x 3 + x 2 + x + 1 is a perfect square, then

Let y 2 = x 4 + x3 + x 2 + x + 1
2 2
x 2 + x  = x 4 + x 3 + x
Consider  
 2  4
= x 4 + x 3 + x 2 + x + 1 −  x 2 + x + 1
3
4 
1
= y2 − (3x 2 + 4x + 4)
4
As discriminant of (3x 2 + 4x + 4) is negative.
Therefore, 3x 2 + 4x + 4 is always greater than zero.
2
x 2 + x  < y 2
Thus,  
 2
or x 2 + x  < | y |
 2
x 
=  x +  x is non-negative, ∀ x ∈ Z
1
But x 2 +
2  2
∴ x 2 + x  = x 2 + x < | y |
 2 2
x
If x is even, then | y | ≥ x 2 + +1
2
5 2
y 2 ≥ x 4 + x3 + x 2 + x + 1 + x
4
5 2
⇒ y2 ≥ y2 + x
4
Which is not possible,
If x ≠ 0, then x = 0 is the only solution when x is even.
x 1
If x is odd, then x 2 + + is an integer.
2 2
| y | ≥  x 2 +  +
x 1
So,
 2 2
x 2 x 3
In this case y 2 ≥ x 4 + x3 + x 2 + x + 1 +  − − 
 4 2 4
x 2 x 3
i.e., y2 ≥ y2 +  − − 
 4 2 4
1 2
= y2 + (x − 2x − 3)
4
1 2
and hence (x − 2x − 3) ≤ 0
4
⇒ x 2 − 2x − 3 ≤ 0
⇒ (x − 3)(x + 1) ≤ 0
∴ −1 ≤ x ≤ 3
www.pdfworld.in
www.pdfworld.in
Odd integral values of x are –1, 1 and 3 of which 1 does not give a perfect square.
There are exactly 3 integral values of x namely. 0, –1 and 3 for which the expression is a perfect
square.
Example 79. Let p (x ) be a real polynomial function

p (x ) = ax 3 + bx 2 + cx + d .
Prove, if | p (x )| ≤ 1 for all x such that | x | ≤ 1
then | a | + | b | + | c | + | d | ≤ 7.
Solution Considering the polynomials ± p (± x ). We may assume without loss of generality that a , b ≥ 0.
Case I If c , d ≥ 0, then
p (1) = a + b + c + d ≤ 1 < 7
Case II If d ≤ 0 and c ≥ 0, then
| a | + | b | + | c | + | d | = a + b + c + d = (a + b + c + d ) − 2d
= p (1) – 2p (0) ≤ 1 + 2 = 3 < 7
Case III If d ≥ 0, c > 0, then
|a | + |b | + |c | + |d | = a + b − c + d
−1
p (1) − p (−1) − p   + p   ≤ + + + =
4 1 8 1 8 4 1 8 8 21
= =7
3 3 3  2 3  2  3 3 3 3 3
Case IV d < 0, c < 0, then

|a | + |b | + |c | + |d | = a + b − c − d
p (1) − 4p   + p  −  ≤ + 4 + =
5 1 4 1 5 4 21
= =7
3  2 3  2 3 3 3
Example 80. If x 5 − x 3 + x = a. Prove that x 6 ≥ 2a − 1.

Solution x 6 + 1 = (x 2 + 1)(x 4 − x 2 + 1)≥ 2x (x 4 − x 2 + 1) [Qx 2 + 1 ≥ 2x and x 4 − x 2 + 1 = (x 2 − 1)2 + x 2 > 0]
x 6 + 1 ≥ 2a
Hence, x 6 ≥ 2a − 1
Example 81. Find the real points (x , y ) satisfying

3x 2 + 3y 2 − 4xy + 10x − 10y + 10 = 0.
Solution It can be considered as a quadratic in x

3x 2 + (10 − 4y ) x + (3y 2 − 10y + 10) = 0
Solving for x we get

1
x = {(4y − 10)} ± (10 − 4y )2 − 12(3y 2 − 10y + 10)
6
Since x , y are real
(10 − 4y )2 − 12(3y 2 − 10y + 10) ≥ 0
Which on simplification gives ( y − 1)2 ≤ 0

i.e., y = 1 and so x = − 1
www.pdfworld.in
www.pdfworld.in
Example 82. Solve

(12x − 1) (6x − 1) (4x − 1) ( 3x − 1) = 5.
Solution We can write the equation in the form
x − 1  x − 1  x − 1  x − 1  = 5
     …(i)
 12   6  4  3  12 ⋅ 6 ⋅ 4 ⋅ 3
1 1 1 1 1 1 1 1
Q < < < and − = −
12 6 4 3 6 12 3 4
We can introduce a new variable
y =   x −
1 1  1  1  1 
 + x −  + x −  + x −  
4   12   6  4  3 
5
=x −
24
5
Substitute x = y + in Eq. (i), we get
24
y + 3  y + 1  y − 1  y − 3  = 5
    
 24   24   24   24  12 ⋅ 6 ⋅ 4 ⋅ 3
 2  1  2  2  3  2 5
y −    y −    =
  24     24   12 ⋅ 6 ⋅ 4 ⋅ 3
 
49
So, y2 =
242
7
i.e., y1 =
24
7
and y2 = −
24
1 1
Corresponding roots are − and .
12 2
Example 83. Find the value of a , b , c which will make each of the expressions x 4 + ax 3 + bx 2 + cx + 1 and
x 4 + 2ax 3 + 2bx 2 + 2cx + 1 a perfect square.
Solution Suppose
2
x 4 + ax 3 + bx 2 + cx + 1 ≡  x 2 + + 1
ax
…(i)
 2 
and x 4 + 2ax 3 + 2bx 2 + 2cx + 1 ≡ (x 2 + ax + 1)2 …(ii)
Equating like coefficients from Eqs. (i) and (ii), we get
a2
b= + 2 and c = a
4
2b = a 2 + 2, 2c = 2a
a2  a2
∴ a2 + 2 = 2  + 2 = + 4
 4  2
2a 2 + 4 = a 2 + 8 or a 2 = 4
∴ a = ± 2, c = ± 2, b = 3
www.pdfworld.in
www.pdfworld.in
Example 84. Show that the expression

( x 2 − yz )3 + ( y 2 − zx )3 + ( z 2 − xy )3 − 3( x 2 − yz )( y 2 − zx )( z 2 − xy )
is a perfect square. Find its square root.
Solution ( x 2 − yz )3 + ( y 2 − zx )3 + ( z 2 − xy )3 −3( x 2 − yz )( y 2 − zx )( z 2 − xy )
Putting x 2 − yz = a , y 2 − zx = b and z 2 − xy = c ,
we get a3 + b3 + c 3 − 3abc
= (a + b + c ) (a 2 + b 2 + c 2 − bc − ca − ab )
= (a + b + c ) (a + ωb + ω2c ) (a + ω2b + ωc )
Now, a + b + c = x 2 − yz + z 2 − zx + z 2 − xy
= (x + ωy 2 + ω2z ) ( x + ω2y + ωx ) …(i)
(a + ωb + ω c ) = x − yz + ω ( y − zx ) + ω ( z − xy )
2 2 2 2 2
= x 2 + ωy 2 + ω2z 2 − yz − ωzx − ω2xy

= (x + y + z ) (x + ωy + ω2z ) …(ii)
(a + ω b + ωc ) = y − zx + ω (z − xy ) + ω ( x − yz )
2 2 2 2 2
= y 2 + ω2z 2 + ωx 2 − zx − ω2xy − ωyz

= (x + y + z ) (x + ω2y + ωz ) …(iii)
From Eqs. (i), (ii) and (iii) it is clear that given expression is a perfect square.
Required square root
= (x + y + z ) (x + ωy − ω2z ) (x + ω2y + ωz )
= x 3 + y 3 + z3 − 3xyz
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1. Prove that the equation ax 5 + bx 3 + cx 2 + d = 0 f (α ) ≠ 0, g(α ) ≠ 0. Prove that m =n and
c b 5bd 2 f (x ) = g (x ).
will have three equal roots, if −
= = 2
b 5ac c 13. Solve x 4 + 2x 3 − 2x − 1 = 0, given it has
each of the quantities being equal to the repeated roots.
repeated roots.
14. Solve the equation
2. Find k so that the equation 2x 4 − 3x 2 6x 4 − 13x 3 − 35x 2 − x + 3 = 0
− 2x + k = 0 may have a double root and solve
which has 2 − 3 as a root.
the equation.
15. Solve the equation x 5 − 5x 4 − 5x 3 + 25x 2
3. Show that the equation x 4 + x 2 + 6 = 0 cannot
+ 4x − 20 = 0, whose roots are given to be of
have three equal roots.
the form ± α, ± β, γ.
4. Given that the equation x 5 − x 4 + 2x 3 + 4 = 0
16. Solve x 3 + x 2 − 16x + 20 = 0, the difference
has one root of the form 1 + αi , find all the
between two of its roots being 7.
roots.
17. Solve x 3 − 13x 2 + 15x + 189 = 0, having given
5. For what values of a and b is the polynomial
f (x ) = x 2 + (a 2 + b ) x + b − 9 negative of the that one root exceeds the other by 2.
polynomial g (x ) = − x − 5ax + b − 3.
2
18. Solve x 4 − 8x 3 + 7x 2 + 36x − 36 = 0 given that
product of two of the roots is negative of the
6. If f (x ) = x 3 + 2x 2 + 5 and
product of the remaining two.
g (x ) = x 4 − 3x 2 + 2x + 1, calculate their gcd.
19. Solve the equation 18x 3 + 81x 2 + 121x + 60
Find polynomials a (x ), b (x ) such that
(f (x ), g (x )) = a (x )f (x ) + b (x )g (x ). = 0 given that, one of its roots is equals to half
the sum of the other two.
7. Give an example of two (non-constant)
polynomials f (x ), g (x ) such that (f (x ), g (x )) = 1. 20. The sum of two roots of
x 4 − 8x 3 + 19x 2 + 4λx + 2 = 0,
8. Let f (x ), g (x )and h (x )be three polynomials such
that f (x ) / h (x ), g (x ) / h (x ) and (f (x ), g (x )) = 1. is equal to the sum of the other two roots, find
Prove that f (x ) g (x ) / h (x ). λ. Solve the equation.
9. If − 2 is a root of x 4 − (2a + 3) x 2− 2(a − 1)x 21. Find the condition that the roots of the
equation x 4 + px 3 + qx 2 + rx + s = 0, be
+ 12 = 0. Prove that it is a repeated root. Find
its multiplicity. Solve the equation. connected by the relation αβ + γδ = 0.
1 22. Given that, two of the roots of
10. Verify that is a root of 4x 3 + 20x 2 − 23x
2 45x 4 − 54x 3 − 98x 2 + 150x − 75 = 0 are equal
+ 6 = 0. Find its multiplicity. Solve the in absolute value but opposite in sign. Solve
equation. the equation completely.
11. Find number a, b, c such that 23. Solve 2x 3 − x 2 − 22x − 24 = 0 two of the roots
(x 2 + x + 5)(ax + b ) + c = x 3 + 7x 2 + 3x + 5. being in the ratio 3 : 4.
12. If m , n are integers ≥ 0.f (x ),g (x ) are polynomial 24. The roots of 2x 3 − 15x 2 + 37x − 30 = 0 are in
such that (x − α )m f (x ) = (x − α )n g (x ) with AP. Find them.
www.pdfworld.in
www.pdfworld.in
25. Solve the equation x 3 − 13x 2 + 15x + 189 = 0, 38. If α , β , γ are the roots of the equation
having given that one root exceeds the other x 3 + px 2 + qx + r = 0 for the equation whose
by 2. roots are
26. Solve the equation x 3 − 7x 2 + 36 = 0 given that 1 1 1
(a) α − ,β − ,γ −
(i) one root is double of another βγ γα αβ
(ii) difference of two roots is 5. (b) α ( β + γ ), β (γ + α ), γ (α + β )
27. Solve the equation 4x 3 + 20x 2 − 23x + 6 = 0 1 1 1

(c) βγ + , γα + , αβ +
α β γ
two of its roots being equal.
28. Solve the equation 39. If α , β , γ are the roots of the equation
x 3 + x 2 + 2x + 3 = 0, from the equation whose
x 5 − 5x 4 + 9x 3 − 9x 2 + 5x − 1 = 0
roots are β + γ − α , γ + α − β , α + β + γ .
29. Form equations, whose roots are the roots of
the following equations with their sign 40. If α , β , γ be the roots of x 3 − ax 2 + bx − c = 0,
changed. find an equation whose roots are 2α − β − γ ,
(i) x − 5x − 7x + 3 = 0
3 2 2β − γ − α , 2γ − α − β.
(ii) −4x 3 + 2x 2 − 3x − 5 = 0 Hence, evaluate (2α − β − γ )(2β − γ − α )

(2γ − α − β ).
30. Find the equation whose roots are the roots of
x 5 − 4x 4 + 3x 2 − 4x + 6 = 0 each diminished 41. Solve the system
by 3. x 1 + x 2 + x 3 + x 4 = 2a1
x 1 + x 2 − x 3 − x 4 = 2a 2
31. Diminish the roots of equation
x 1 − x 2 + x 3 − x 4 = 2a3
5x 3 − 13x 2 − 12x + 7 = 0 by 2.
x 1 − x 2 − x 3 + x 4 = 2a 4 .
32. Find the equation whose roots are the squares
42. Solve the system
of the roots of the equation
ax + m ( y + z + v ) = k
x 5 + x 3 + x 2 + 2x + 3 = 0
by + m ( x + z + v ) = l
33. Form the equation whose roots are the cubes
cz + m ( x + y + v ) = p
of the roots of x 3 + 3x 2 + 2 = 0.
dv + m ( x + y + z ) = q.
34. Find the equation whose roots are the cubes of
the roots of 2x 3 − x 2 + 2x − 3 = 0. Obtain value
of Σα3β3. (b + c )( y + z ) − ax = b − c
(c + a )( x + z ) − by = c − a
35. Show that the roots of the equation
1 2 3 (a + b ) ( x + y ) − cz = a − b
x 3 − q 2x − q − r2 = 0
3 27 if a + b + c ≠ 0.
differ by a constant from the squares of roots 44. Solve the system
of z + ay + a 2x + a3t + a 4 = 0
x 3 + qx + r = 0.
z + by + b 2x + b3t + b 4 = 0
36. Show that the cubes of the roots of
z + cy + c 2x + c 3t + c 4 = 0
x 3 + ax 2 + bx + ab = 0 are given by
z + dy + d 2x + d 3t + d 4 = 0.
x 3 + a3x 2 + b3x + a3b3 = 0.
37. If α , β , γ are the roots of z3 + 3Hz + G = 0, find
yz = ax
the equation whose roots are
α+1 β+1 γ+1 zx = by (a > 0, b > 0, c > 0)
, , xy = cz.
β + γ −α γ + α −β α + β − γ
www.pdfworld.in
www.pdfworld.in
46. Solve 55. Solve

x (y + z ) = a 2 x (y + z − x ) = a
y (z + x − y) = b
y (x + z ) = b 2
z (x + y − z ) = c.
z ( x + y ) = c 2.
56. Solve the equations
47. Solve the system (a) ax + by + z = zx + ay + b = yz + bz + a = 0
x 2 + y 2 = cxyz
(b) x + y + z − u = 12 ; x 2 + y 2 − z 2 − u 2 = 6;
x + z = bxyz
2 2
x 3 + y 3 − z3 + u3 = 218 ; xy + zu = 6.
y + z = axyz.
2 2
57. Eliminate p and q from the equations
48. Solve the system x (p + q ) = y ; p − q = k (1 + pq ); xpq = a.
b (x + y ) c (z + x ) 58. Eliminate x,y from the equations
+ =a
x + y + cxy x + z + bxz 4(x 2 + y 2 ) = ax + by ; 2(x 2 − y 2 ) = ax − by and
c (y + z ) a (x + y ) xy = c 2.
+ =b
y + z + ayz x + y + cxy
59. Eliminate x , y , z from the equations
a (x + z ) b (y + z ) ( y + z )2 = 4a 2yz ; (z + x )2 = 4b 2zx ;
+ = c.
x + z + bxz y + z + ayz (x + y )2 = 4c 2xy .
49. Solve the system 60. Eliminate x , y from the equations
x 2 − yz = a x 2y = a ; x (x + y ) = b ; 2x + y = c .
y 2 − xz = b 61. Eliminate x , y from the equations
xy
z − xy = c .
2
ax 2 + by 2 = ax + by = = c.
x +y
50. Solve
62. Show that b3c 3 + c 3a3 + a3b3 = 5a 2b 2c 2 is the
x 2 + y 2 + z 2 = 84
eliminant of ax + yz = bc ; by + zx = ca;
x + y + z = 14 cz + xy = ab and xyz = abc .
xz = y 2. 63. Solve the equation y 3 = x 3 + 8x 2 − 6x + 8, for
51. Solve positive integers x and y . (RMO 2000)
x + y + z = 13 64. Find all real values of a for which the equation
x 2 + y 2 + z 2 = 65 x 4 − 2ax 2 + x + a 2 − a = 0 has all its roots real.
(RMO 2000)
xy = 10.
65. Solve the following equation for real x
52. Solve the equation for all possible values of
(x 2 + x − 2)3 + (2x 2 − x − 1)3 = 27(x 2 − 1)3.
x , y and z.
(RMO 2002)
53. Solve x + y + z = a ; x 2 + y 2 + z 2 = b 2, a and b
66. Find all real numbers a for which the equation
are real numbers and xy = z 2. x 2 + (a − 2) x + 1 = 3 | x | has exactly three
What conditions a and b satisfy for x , y , z to be distinct real solutions in x . (RMO 2003)
all +ve and distinct?
67. Let α and β be the roots of the quadratic
54. Solve equation x 2 + mx − 1 = 0, where m is an odd
y 2 + yz + z 2 = ax integer. Let λn = αn + βm , for n ≥ 0. Prove that
for n ≥ 0.
z 2 + zx + x 2 = ay
(a) λn is an integer.
x 2 + xy + y 2 = az.
(b) GCD (λn , λn + 1) = 1. (RMO 2004)
www.pdfworld.in
www.pdfworld.in
68. Find all pairs (a , b ) of real numbers such that 70. Let P1 (x ) = ax 2 − bx − c , P2 (x ) = bx 2 − cx − a ,
whenever α is a root of x 2 + ax + b = 0,α 2 − 2 is
P3 (x ) = cx 2 − ax − b be three quadratic
also a root of the equation. (RMO 2007)
polynomials where a , b , c are non-zero real
69. Suppose a and b are real numbers such that numbers. Suppose there exists a real number α
the roots of the cubic equation
such that P1 (α ) = P2 (α ) = P3 (α ). Prove that
ax 3 − x 2 + bx − 1 = 0 are all positive real
a = b = c. (RMO 2010)
numbers. Prove that
(a) 0 < 3ab ≤ 1 (b) b ≥ 3. (RMO 2008)
Level 2
1. Eliminate x , y , z from 9. If the roots of the equation
x 2 + y 2 + z2 = x + y + z = 1 x 3 + px 2 + qx + r = 0
a b c
( x − p ) = ( y − q ) = ( z − r ). be α , β , γ . Find the equation whose roots are
x y z
(a) β 2 + γ 2 − α 2 , γ 2 + α 2 − β 2, α 2 + β 2 − γ 2
2. Eliminate x and y from the equations
(b) β 2 + γ 2 , γ 2 + α 2 , α 2 + β 2.
x + y = a ; x 2 + y 2 = b ; x 3 + y 3 = c.
3. If the roots of the equation 10. Find the sum of the fifth powers of the roots of
the following equation.
2x 3 + x 2 + x + 1 = 0
(a) x 4 − 3x 3 + 5x 2 − 12x + 4 = 0
be α , β , γ. Find the equation whose roots are
(b) x 3 + x 2 + x + 5 = 0
1 1 1 1 1 1 1 1 1
+ − ; + − ; + − .
β2 γ 2 α2 α2 γ 2 β2 α2 β2 γ 2 11. Find the value of α5 + β5 + γ 5, where α , β , γ are
4. Find the equation whose roots are the ratios of the roots of x 3 + 3x + 3 = 0.
the roots α, β , γ of the cubic x 3 + qx + r = 0. 12. If α , β , γ are the roots of the cubic
5. If the roots of the equation x 3 + 2x + 6 = 0, prove that s7 = 2(s4 − s6 ).
x 3 − 6x 2 + 11x − 6 = 0 13. Show that the set of polynomials
be α , β , γ. Find the equation whose roots are P = {pk | pk (x ) = x 5k + 4
+ x 3 + x 2 + x + 1, k ∈ N }
β 2 + γ 2 , γ 2 + α 2 , α 2 + β 2. has a common non-trivial polynomial divisor.
6. If α , β , γ are the roots of the equation 14. Show that the polynomial
x 3 + px 2 + qx + r = 0, find the equation whose x (x n − 1 − nan − 1 + an (n − 1) is divisible by
(x − a )2.
roots are βγ − α 2 , γα − β 2, αβ − γ 2.
15. If x 3 + 3px + q has a factor of the form (x − a ) 2.
7. The roots of the equation
Show that q 2 + 4p3 = 0.
x 3 − ax 2 + bx − c = 0
16. Prove that
are α , β , γ form the equation whose roots are −1 −1
1 1 (a) The polynomial x (x n − nan ) + an (n − 1)
α + β , β + γ , γ + α. Also, express +
α+β β+ γ is divisible by (x − a )2.
1
+ in terms of a , b , c . (b) The polynomial
γ+α
(1 − x n ) (1 + x ) − 2n x n (1 − x ) − n 2x n (1 − x )2
8. If α , β , γ are the roots of the cubic is divisible by (1 − x ) 3.
x 3 + 3x + 2 = 0, form an equation whose roots
17. Prove that the polynomial
are ( β − γ )2 , (γ − α )2 , (α − β )2 and hence show
x n sin ψ − en − 1 x sin nψ + en sin (n − 1) ψ
that x 3 + 3x + 2 = 0 has imaginary roots.
is divisible by x 2 − 2ex cos ψ + e 2.
www.pdfworld.in
www.pdfworld.in
18. Prove that if x 1 , x 2 , K , x m are m different ax n2 − 1 + bx n −1 + c = xn

arbitrary quantities. f (x ) is a polynomial of
ax n2 + bx n + c = x 1
degree less than m, then there exists the
identity. is n unknowns x 1 , x 2 , K , x n , then
(x − x 2 ) (x − x 3 ) K (x − x m ) (a) (b − 1) 2 < 4ac (b) (b − 1) 2 = 4ac
f (x ) = f (x 1 ) + ...+ (c) (b − 1) > 4ac
2
(x 1 − x 2 ) (x 1 − x 3 ) K (x 1 − x m )
(x − x 1 ) (x − x 3 ) K (x − x m ) 28. Let p (x ) = 0 be a fifth degree polynomial
f (x 2 ) + ... +
(x 2 − x 1 ) (x 2 − x 3 ) K (x 2 − x m ) equation with integer coefficients that has at
least one integral root. If p (2) = 13 and
(x − x 1 ) (x − x 2 ) K (x − x m −1 ) p (10) = 5. Compute a value of x that must
f (x m ) .
(x m − x 1 ) (x m − x 2 ) K (x m − x m −1 ) satisfy p (x ) = 0.
19. If x and y are real, solve the inequality. 29. If a , b , c , x are real numbers such that abc ≠ 0
log 2 x + log x 2 + 2 cos y ≤ 0 . and
xb + (1 − x ) c xc + (1 − x ) a xa + (1 − x ) b
= = ,
20. Show that (a − b )2 + (a − c )2 = (b − c )2 is not a b c
solvable, when a , b , c are all distinct. then prove that either a + b + c = 0 or a = b = c .
21. Find all pairs of integers x , y such that (INMO 2000)
(xy − 1)2 = (x + 1)2 + ( y + 1)2. 30. Solve for integers x , y , z

x + y = 1 − z , x 3 + y 3 = 1 − z 2. (INMO 2000)
22. Find all the solutions of the system of
equations 31. Show that the equation
y = 4x 3 − 3x ; z = 4y 3 − 3y and x 2 + y 2 + z 2 = (x − y )( y − z )(z − x )
x = 4z3 − 3z. has infinitely many solutions in integers
x , y , z. (INMO 2001)
23. The roots x 1 , x 2 , x 3 of the equation
32. Show that for every real number a the
x 3 + ax + a = 0, where a is a non-zero real,
equation
satisfy 8x 4 − 16 x 3 + 16 x 2 − 8x + a = 0
x 12 x 22 x 32
+ + = − 8 , find x 1 , x 2 , x 3. has atleast one non-real root and find the sum
x2 x3 x1 of all the non-real roots of the equation.
24. Find all values of a for which the system of (INMO 2003)
equation 2| x | + | x | = y + x 2 + a, x 2 + y2 = 1 33. If α is a real root of the equation
has only one solution (a , x , y are real x 5 − x 3 + x − 2 = 0, prove that [α 6 ] = 3. (For any
numbers). real number a, we denote by [a ] the greatest
25. If p (x ) = ax 2 + bx + c and q (x ) = − ax 2 integer not exceeding a.) (INMO 2004)
+ bx + c , ac ≠ 0. Show that p (x ) ⋅ q (x ) = 0 has at 34. Suppose p is a prime greater than 3. Find all
least two real roots. the pairs of integers (a , b ) satisfying the
equation a 2 + 3ab + 2p (a + b ) + p 2 = 0.
26. Find the real roots of the equation
(INMO 2004)
= − a +  a 2 + x −
1 1
x 2 + 2ax + .
16  16  35. Let p , q , r be positive real numbers, not all
equal, such that some two of the equations
27. If a , b , c ∈ R, a ≠ 0. Solve the system of px 2 + 2qx + r = 0
equations. qx 2 + 2rx + p = 0
ax 12 + bx 1 + c = x 2 rx 2 + 2px + q = 0
ax 22 + bx 2 + c = x 3 have a common root, say α. Prove that
K K K (a) α is real and negative,
K K K (b) the third equation has non-real roots.
K K K (INMO 2005)
www.pdfworld.in
www.pdfworld.in
36. Let m and n be positive integers such that the (Here, [x ] denotes the largest integer not
equation x 2 − mx + n = 0 has real roots exceeding x ) (INMO 2009)
α and β. Prove that α and β are integers if and 40. Find all non-zero real numbers x , y , z which
only if [m α ] + [m β ] is the square of an integer. satisfy the system of equations
(Here, [x ] denotes the largest integer not
(x 2 + xy + y 2 ) ( y 2 + yz + z 2 )
exceeding x .) (INMO 2007)
( z 2 + zx + x 2 ) = xyz ,
37. Find all triples ( p , x , y ) such that p x = y 4 + 4,
where p is a prime and x , y are natural (x 4 + x 2y 2 + y 4 ) ( y 4 + y 2z 2 + z 4 )
numbers. (INMO 2008) (z 4 + z 2x 2 + x 4 ) = x 3y 3z3. (INMO 2010)
38. Let P (x ) be a given polynomial with integer
41. Consider two polynomials
coefficients. Prove that there exist two n −1
P (x ) = an x n + an −1x + . . . + a1 + a0
x and
polynomials Q (x ) and R (x ), again with integer
n −1
coefficients, such that (a) P (x ) Q (x ) is a Q (x ) = bn x n + bn − 1x + . . . + b1x + b0 with
polynomial in x 2 (b) P (x ) R (x ) is a polynomial in integer coefficients such that an − bn is a
x 3. (INMO 2008) prime, an − 1 = bn − 1 and an b0 − a 0bn ≠ 0.
Suppose there exists a rational number r such
39. Find all the real numbers x such that
that P (r ) = Q (r ) = 0. Prove that r is an integer.
[x 2 + 2x ] = [x ]2 + 2[x ] . (INMO 2011)
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1. Let f (x ) = ax 5 + bx 3 + cx 2 + d c3 c 5bd
or − =d or − = 2
f ′ (x ) = 5ax 4
+ 3bx + 2cx
2 5b 2 b c
5bd
f ′ ′ (x ) = 20ax + 6bx + 2c
3 or α=
c2
f (x ) = 0 will have three equal roots α , α , α (say) c b2 5bd
Hence, − = = 2 , each = α
If f ′ (x ) = 0 has two equal roots α , α b 5ac c
and f ′ ′ (x ) = 0 has a root α. 2. Let f (x ) = 2x 4 − 3x 2 − 2x + k

∴ aα5 + bα3 + cα 2 + d = 0 …(i) f ′ (x ) = 8x 3 − 6x − 2 = 2(4x 3 − 3x − 1)
5aα 4 + 3bα 2 + 2cα = 0 Since, f (x ) = 0 has a double root f (x ) = 0
or 5aα + 3bα + 2c = 0
3
(Qα ≠ 0) …(ii) and f ′ (x ) = 0 have a common root.
∴ HCF of f (x ) and f ′ (x ) is a linear polynomial.
20aα3 + 6bα + 2c = 0
HCF of f (x ) and f ′ (x ) will be the linear
⇒ 10aα3 + 3bα + c = 0 …(iii)
polynomial (x − 1) only, when 6 − 2k = 0 i . e. ,
Multiply Eq. (ii) by 2 and subtract from Eq. (iii), k = 3.
we get Now, HCF φ (x ) = x − 1
−3bα − 3c = 0 φ (x ) = 0 gives x = 1
c
⇒ α=− ∴ Two roots of the equation
b
c 2x 4 − 3x 2 − 2x + 3 = 0 (Qk = 3)
Putting − α = in Eq. (iii), we have
b are 1, 1.
10ac 3 4. If we put x = − 1 in the given equation, it is
− − 3c + c = 0
b3 satisfied and as such x + 1 is a factor and the
equation when divided by x + 1 by synthetic
10ac 3 5ac 2
or − 3
= 2c or − =1 division gives the quotient as under
b b3
c b2 b2 –1 1 –1 2 0 0 4
or −
= ⇒α =
b 5ac 5ac –1 2 –4 4 –4
c 1 –2 4 –4 4 0
Putting α = − in Eq. (ii), we have
b Quotient is x 4
− 2x + 4x − 4x + 4 = 0
3 2
…(i)
5ac 3
− − 3c + 2c = 0 It has a root 1 + iα and hence, 1 − iα must also
b3
be its root and product of factors
5ac 3
or − = c or b3 = − 5ac 2 corresponding to these is
b3
(x − 1)2 + α 2 or (x 2 − 2x + 1 + α 2 )
c
Putting α = − in Eq. (i), we get
b Let the remaining factor be (x 2 + px + q )
ac 5 c 3 c3 Hence, we have
− 5 − 2 + 2 +d =0
b b b (x 4 − 2x 3 + 4x 2 − 4x + 4)
ac 5 ac 5 = (x 2 − 2x + 1 + α 2 ) (x 2 + px + q )
or = d or =d
b 2 ⋅ b3 b 2 (−5ac 2 )
Comparing the coefficients of like powers of x
[Qb3 = − 5ac 2 ] in both sides, we get
www.pdfworld.in
www.pdfworld.in
−2 = − 2 + p, ∴p = 0, x + 19 / 2
(1 + α ) + q − 2p = 4; p (1 + α ) − 2q = − 4;
2 2 4x − 50 x 2 − 3x + 11
q (1 + α 2 ) = 4; ×4
Putting p = 0, we get q = 2 4x − 12x + 44
2
∴ 1 + α 2 = 2 or α = ± 1 4x 2 − 50x
Hence, the factors are
38x + 44
(x + 1)(x 2 − 2x + 2)(x 2 + 2) = 0 38x − 475
∴ Roots are −1, 1 ± i , ± i 2. 519 / 519
5. f (x ) = − g (x ) 4x − 50
∴x 2 + (a 2 + b ) x + b − 9 = − (− x 2 − 5ax + b − 3) 1 4x − 50
⇒ x 2 + (a 2 + b ) x + b − a = x 2 + 5ax − b + 3 4x − 50
×
On comparing coefficient of like powers of x
In the first stage of above division, we have
on both sides
g (x ) = f (x )(x − 2) + x 2 − 3x + 11
a 2 + b = 5a …(i)
∴ x 2 − 3x + 11 = g (x ) − (x − 2)f (x ) …(i)
and b − 9= −b + 3 …(ii)
In the second stage of above division
2b = 12 ⇒ b = 6
f (x ) = (x 2 − 3x + 11) (x + 5) + 4x − 50
Putting in Eq. (i), we get
= {g (x ) − (x − 2) f (x )} (x + 5) + 4x − 50
a 2 + 6 = 5a
4x − 50 = f (x ) + (x − 2) (x + 5) f (x )
a 2 − 5a + 6 = 0 − (x + 5) g (x ) …(ii)
∴ (a − 2)(a − 3) = 0 ⇒ a = 2, 3 In the third stage of division
4(x 2 − 3x + 11) = (4x − 50)  x +
19 
Hence, a = 2 or 3 and b = 6.  + 519
 2
6. We first, find GCD of f (x ), g (x ) 19 
519 = 4(x 2 − 3x + 11) − (4x − 50)  x + 
GCD of f (x ), g (x ) = 1  2
∴ f (x ), g (x ) are coprime polynomials. = 4 [g (x ) − (x − 2) f (x )] − [(x 2 + 3x − 9) f (x )
x −2
− (x + 5) g (x )] x +
19 
x 3 + 2x 2 + 5 x 4 + 0x 3 − 3x 2 + 2x + 1  2 
∴ 519 = 4g (x ) − (4x − 8) f (x )
x 4 + 2x 3 + 0x 2 + 5x
− (x 2 + 3x − 9)  x +
19 
 f (x )
 2
− 2x 3 − 3x 2 − 3x + 1
+ (x + 5)  x +
19 
− 2x 3 − 4x 2 − 0x − 10  g (x )
 2
x 2 − 3x + 11  19x 2
= −  4x − 8 + x 3 + + 3x 2
x +5  2
x − 3x + 11 x + 2x + 0x + 5
2 3 2 57x 171 
+ − 9x − f (x )
2 2 
x 3 − 3x 2 + 11x
+ x 2 + + 4 g (x )
19x 95
+ 5x +
 2 2 
5x 2 − 11x + 5
f (x )
=− [2x 3 + 25x 2 + 47x − 187]
5x 2 − 15x + 55 2
g (x )
4x − 50 + [2x 2 + 29x + 103]
2
www.pdfworld.in
www.pdfworld.in
 2x 3 + 25x 2 + 47x − 187  Now, it is given that − 2 is a root of f (x ) = 0

∴ 1 = −  f (x )
 1038  ∴ f (−2) = 0 ⇒ 12 − 4a = 0 ⇒ a = 3
[2x + 29x + 103]

2 Putting a = 3 in Eq. (i), we get
+ g (x )
1038 f (x ) = x 4 − 9x 2 − 4x + 12 = 0
Hence, 1 = a (x ) f (x ) + b (x ) g (x ), where By Horner’s method
1
a (x ) = − (2x 3 + 25x 2 + 47x − 187)
1038 −2 1 0 −9 −4 12
1 −2 4 10 −12
and b (x ) = (2x 2 + 29x + 103)
1038 −2 1 −2 −5 6 0
7. Let f (x ) = 3x + 1; g (x ) = x ; −2 8 −6 (Remainder)
To find GCD, we proceed as follows −2 1 −4 3 0
3
x 3x + 1 −2 12
3x −2 1 −6 15
1 −2
x
1x 1 −8
x 1
×
∴ Transformed equation is
GCD = 1 i . e. , (f (x ), g (x )) = 1
y 4 − 8y 3 + 15y 2 = 0
8. Since, f (x ) | h (x )
where y = x − h = x − (−2)
∴ ∃ s a polynomial p (x ) such that
⇒ y 2 ( y 2 − 8y + 15) = 0
h (x ) = g (x ) p (x ) …(i)
Again g (x )| h (x ) ⇒ y 2 = 0, ( y − 3)( y − 5) = 0
∴ ∃s a polynomial q (x ) such that y = 0, 0, 3, 5

h (x ) = q (x ) ⋅ g (x ) …(ii) y = x + 2 ⇒x = y − 2
Further Q (f (x ), g (x )) = 1 x = 0 − 2, 0 − 2, 3 − 2, 5 − 2
∴ ∃ polynomial a (x ) and b (x ) = − 2, − 2, 1, 3
Such that f (x ) a (x ) + g (x ) b (x ) = 1 …(iii) ∴ Roots of f (x ) = 0 are − 2, − 2, 1, 3
Multiplying by h (x ) on both sides, we get 10. Let f (x ) = 4x 3 + 20x 2 − 23x + 6

f (x ) h (x ) a (x ) + g (x ) h (x ) b (x ) = h (x ) By Horner’s method
Using Eqs. (ii) and (i), we get 1 4 20 −23 6
⇒ f (x )[g (x ) q (x )] a (x ) + g (x )[f (x ) p (x )] 2 2 11 −6
b (x ) = h (x ) 4 22 −12 0
⇒ f (x ) g (x )[q (x ) a (x ) + p (x ) b (x )] = h (x ) 2 12
4 24 0
⇒ f (x ) g (x )| h (x )
2
Hence, the result.
4 26
9. Let f (x ) = x 4 − (2a + 3) x 2 − 2(a − 1) x + 12 …(i)
4
f (−2) = (−2)4 − (2a + 3)(−2)2 − 2(a − 1) (−2) + 12 ⇒ Remainder = 0
1
= 16 − 8a − 12 + 4a − 4 + 12 = 12 − 4a ∴ is a root of 4x 3 + 20x 2 − 23x + 6 = 0
2
www.pdfworld.in
www.pdfworld.in
1 ∴ Our supposition is wrong.

Also, it is clear, when f (x ) is divided by x − .
2 Hence, m = n
Quotient is 4x 2 + 22x − 12
and (x − α )m f (x ) = (x − α )n g (x )
f (x ) =  x −  (4x 2 + 22x − 12)
1
∴ ⇒ f (x ) = g (x )
 2
13. f (x ) = x 4
+ 2x − 2x − 1 = 0
3
…(i)
=  x −  4  x 2 + x − 3
1 11
 2  2  f ′ (x ) = 4x + 6x − 2 = 0
3 2
= 4  x −   x −  (x + 6) 2x 3 + 3x 2 − 1 = 0
1 1 i.e., …(ii)
 2  2
Since, f (x ) = 0 has repeated roots (i) and (ii) has
2
= 4  x −  (x + 6)
1 common roots which can be obtained by
 2 solving d (x ) = 0 where d (x ) is HCF of f (x ) and
1 f ′ (x ).
⇒ x = is a root of f (x ) with multiplicity 2 and
2 x 1
2x 3 + 3x 2 − 1) x 4 + 2x 3 − 2x − 1( +
third root is given by factor x + 6 i . e. , −6 2 4
∴ Roots are 1/2, 1/2, –6. 3x 3 x
x4 + −
11. Since, (x 2 + x + 5)(ax + b ) + c 2 2
= x 3 + 7x 2 + 3x + 5 x 3 3x
− −1
2 2
∴ ax 3 + (a + b ) x 2 + (5a + b ) x + 5b + c
x 3 3x 2 1
= x 3 + 7x 2 + 3x + 5 + −
2 4 4
∴ By equality of two polynomials, we have 3 2 3x 3
a =1 − x − −
4 2 4
a + b =7
x 2 + 2x + 1) 2x 3 + 3x 2 − 1 ( 2x − 1
⇒ b = 7 − a = 7 − 1 = 6, 5b + c = 5
5a + b = 3 2x 3 + 4x 2 + 2x
⇒ 5+ 6=3 − x 2 − 2x − 1
which is not true.
− x 2 − 2x − 1
Hence, there do not exist a , b , c for which the
two polynomials are equal. ×
12. Given, (x − α ) f (x ) = (x − α ) g (x ),
m n ∴ HCF of f (x ) = 0 and f ′ (x ) = 0 is
d (x ) = x 2 + 2x + 1 d (x ) = 0 ⇒ x 2 + 2x + 1 = 0
f (α ) ≠ 0, g(α ) ≠ 0
we want to prove that ⇒ (x + 1)2 = 0 ⇒ x = − 1, − 1.
m = n and f (x ) = g (x ) ∴ Roots of f (x ) = 0 are −1, − 1.
If possible let m ≠ n Let us use synthetic division method.
Without any loss of generality, let m < n
–1 1 2 0 –2 –1
∴ n − m is a + ve integer, so that
–1 –1 1 1
(x − α )m f (x ) = (x − α )n g (x )
1 1 –1 –1 0
−m
i.e., f (x ) = (x − α )n g (x ) –1 0 1
i.e., (x − α )| f (x ) 1 0 –1 0
⇒ α is a root of f (x ) = 0.
∴ Reduced equation is x 2 + 0x − 1 = 0, x = ± 1
⇒ f (α ) = 0 which is contrary to the given
∴ Roots of f (x ) = 0 are −1, − 1, − 1, 1.
hypothesis.
www.pdfworld.in
www.pdfworld.in
14. f (x ) = 6x 4 − 13x 3 − 35x 2 − x + 3 …(i) −23 ± 7 8

∴ α= = − 5, −
6 3
All its coefficients are rational since 2 − 3 is a
8
root of Eq. (i) But α = − does not satisfy the given equation.
3
∴ 2+ 3 is also a root of Eq. (i)
∴ α = − 5, β = − 8 − 2α = − 8 + 10 = 2
∴ (x − 2 + 3 )(x − 2 − 3 ) divides f (x )
Roots of the given equation are α , α + 7, β
i . e. , (x − 2)2 − 3 divides f (x ) i.e., x 2 − 4x + 1
i.e., − 5, − 5 + 7, 2 or − 5, 2, 2.
divides f (x )
∴We have, f (x ) = 6x 4 − 13x 3 − 35x 2 − x + 3 17. Let roots be α , α + 2, β then
2α + β = 11 …(i)
= (x 2 − 4x + 1)(6x 2 + 11x + 3)
Also, α (α + 2) + (α + 2) β + αβ = 15
∴The other two roots are given by or α (α + 2) + β (2α + 2) = 15 …(ii)
6x 2 + 11x + 3 = 0 Eliminating β from Eqs. (i) and (ii)
−11 ± 121 − 72 − 11 ± 49 α (α + 2) + (11 − 2α )(2α + 2) = 15
i.e., x = =
12 12 α 2 + 2α + 22 + 18α − 4α 2 = 5
18 4
=− or − 3α 2 − 20 α − 7 = 0
12 12
3 1 20 ± 400 + 84
=− or − ∴ α=
2 3 6
20 ± 22 −1
Hence, all the roots of Eq. (i) are = = 7,
6 3
2± 3 , − 3 / 2, −1 / 3
−1
But α = does not satisfy equation.
15. Sum of the roots = α − α + β − β + γ = 5 3
∴ γ =5 ∴ α = 7, β = 11 − 2α = 11 − 14 = − 3
But γ is a root of the given equation. Roots of given equation are α , α + 2, β
∴ x − 5 is a factor of i.e., 7, 9, –3.
x 5 − 5x 4 − 5x 3 + 25x 2 + 4x − 20
18. Let roots be α , β , γ , δ such that αβ = − γδ
5 1 –5 –5 25 4 – 20 Now, αβγδ = − 36 ⇒ (− γδ) γδ = − 36
5 0 – 25 0 20 ⇒ γ 2δ 2 = 36 ⇒ γδ = 6 or −6
1 0 –5 0 4 0 and αβ = − 6 or 6
Depressed equation is x 4 − 5x 2 + 4 = 0 The factors corresponding to these roots are
(x 2 − 1)(x 2 − 4) = 0 , ∴ x = ± 1, ± 2 of the types x 2 − px − 6 and x 2 − qx + 6
Roots are ± 1, ± 2, 5. We have,

16. Let the roots be α , α + 7, β, then x 4 − 8x 3 + 7x 2 + 36x − 36
α + (α + 7) + β = − 1 ≡ (x 2 − px − 6)(x 2 − qx + 6)
or 2α + β = − 8 …(i) Equating coefficients of like powers of x
Also, α (α + 7) + (α + 7) β + αβ = − 16 Coefficients of x 3 ; − 8 = − p – q
α (α + 7) + β (2α + 7) = − 16 …(ii)
∴ p+q=8 …(i)
Eliminating β from Eqs. (i) and (ii)
Coefficients of x ; 36 = − 6p + 6q
α (α + 7) + (−8 − 2α )(2α + 7) = − 16
∴ p −q= − 6 …(ii)
α + 7α − 30 α − 4 α 2 − 56 + 16 = 0
2
3 α 2 + 23 α + 40 = 0 p = 1 and q = 7
www.pdfworld.in
www.pdfworld.in
∴ Given equation may be written as Coefficient of x , 4λ = − 4( p + q )

(x − x − 6) (x − 7x + 6) = 0
2 2 ⇒ p+q=−λ …(ii)
Constant terms 2 = pq …(iii)
or (x − 3) (x + 2) (x − 1) (x − 6) = 0
∴ x = 3, − 2, 1, 6 From Eqs. (i) and (iii), it is evident that p , q are
the roots of t 2 − 3t + 2 = 0
19. Since, one root is equal to half the sum of the
other two. or (t − 1)(t − 2) = 0; ∴ t = 1, 2
Take, p = 1, q = 2
∴ The roots are in AP.
Given equation can be written as
Let the roots be α − δ, α , α + δ
(x 2 − 4x + 1)(x 2 − 4x + 2) = 0
1
∴ α = (α − δ + α + δ) 4± 16 − 4 4 ± 16 − 8
2 ∴ x = ,
2 2
81
Sum of roots = α − δ + α + α + δ = − or 2± 3, 2 ± 2
18
9 21. αβ + γδ = 0 …(i)
⇒ 3α = −
2 Let αβ = k, then γδ = − k
∴ α=−
3 If α + β = l and γ + δ = m, then
2 (x 4
+ px + qx + rx + s ) ≡ (x − lx + k )
3 2 2
3
∴ − is a root of given equation. (x 2 − mx − k )
2
Comparing the coefficients of like powers of x,
– 3/2 18 81 121 60
we get
– 27 – 81 –0
l + m = − p ; lm = q ; k (l − m ) = r
18 54 40 0 r
∴ l − m = and − k 2 = s
k
Depressed equation is 18x 2 + 54x + 40 = 0
In order to find the condition, we have to
⇒ 9x 2 + 27x + 20 = 0 eliminate l , m and k between the above four
Its roots are relations.
−27 ± 729 − 720 −27 ± 3 4 5 (l − m )2 = (l + m )2 − 4lm
= = − or −
18 18 3 3 r2
∴ = p 2 − 4q or r 2 = − s (p 2 − 4q )
Roots of the given equation are – 4/3, – 3/2, k2
– 5/3. or p 2s + r 2 = 4qs is the required condition.
20. Sum of all the four roots = 8
22. α = − β, i.e., α + β = 0
Q Sum of two roots = Sum of other two roots
∴ α + γ = 54 / 45
∴ Each sum = 4

∴f (x ) =  x −
54 3 98 2 150
x − x + x −
75 
 =0
4
Factors corresponding to these roots are of  45 45 45 45 
the types
≡ (x 2 + l )  x 2 − x + m 
54
x 2 − 4x + p
 45 
and x 2 − 4x + q
98
On comparing, we get l + m = −
Thus, we have 45
x 5 − 8x 3 + 19x 2 + 4λx + 2 54 150
and − l =
≡ (x 2 − 4x + p )(x 2 − 4x + q ) 45 45
25 98
Equating the coefficients of like powers of x ∴ l =− ; m = −l −
9 45
Coefficient of x , 19 = p + q + 16
2
25 98 27 3
m= − = =
⇒ p + q =3 …(i) 9 45 45 5
www.pdfworld.in
www.pdfworld.in
f (x ) = 0 is  x 2 −
25   x 2 − 54 x + 3  = 0 We shall choose that set of values which
∴   
 9  45 5 satisfies third relation
Solving the above, we get αβγ = α (α + 2) β = − 189
5 3±5 6 The value α = 7, β = − 3, satisfies it.
x =± ,
3 5 Roots are 7, 9, −3.
23. Let the roots be 3α , 4α and β. 26. (a) Let roots be α , 2α and β.
1 Σα = α + 2α + β = 7
∴ Σα = 7α + β =
2 ∴ β = 7 − 3α …(i)
β =  − 7α 
1
∴ …(i) Σαβ = α2α + 2αβ + βα = 0
2 
as the coefficient of x is missing.
Σαβ = 12 α 2 + 4 αβ + 3 αβ = − (22) / 2 = − 11 or 2α 2 + 3α (7 − 3α ) = 0 [by Eq. (i)]
12α + 7α  − 7α  = − 11
2 1
or [by Eq. (i)] or α (21 − 7α ) = 0 ; α = 0 or 3
2 
when α = 3, β = 7 − 3 ⋅ 3 = − 2
74 α 2 − 7α − 22 = 0 when α = 0, β = 7
or (2α + 1)(37α − 22) = 0 Also, αβγ = α ⇒ 2αβ = − 36
1 22 ⇒ α 2β = = − 18
∴ α = − or and corresponding values of β
2 37 Roots are α , 2α , β i.e., 3, 6, − 2.
From Eq. (i) are 4 and −171 / 37
Roots for (b) part are 6, 3, − 2.
αβγ = 12α 2β = − 12
27. Let the root be α , α , β.
The set of values α = − 1 / 2 and β = 4 satisfy
the given equation. Σα = 2α + β = − 5
Roots are 3α , 4α , β i . e. , ∴ β = − 5 − 2α …(i)

−3 / 2, − 2, 4. Σαβ = α 2 + αβ + αβ = − 23 / 4
24. Let the roots be a − d , a , a + d or α 2 + 2α (−5 − 2α ) = − 23 / 4 [by Eq. (i)]

15 12α 2 + 40 α − 23 = 0
∴ Σα = 3a = , or
2
5 (2α − 1)(16 α + 23) = 0
∴ a = , which is a root.
2 α = 1 / 2 and −23 / 6
Dividing the given equation by 2x − 5 by α = 1 / 2 satisfies above.
synthetic division, we get quotient Hence, α = − 23 / 6 will be rejected.
2x 2 − 10x + 12 = 0 Putting α = 1 / 2 in Eq. (i), we get
or 2(x − 2)(x − 3) = 0
β=−6
Other roots are 2 and 3.
Roots are 1/2, 1/2, −6.
∴ Roots are 3, 5 / 2, 2.
25. Let roots be α , α + 2, β 28. x = 1 is a root of this equation
∴ Σα = α + α + 2 + β = 13 It can be written as
∴ β = 11 − 2α …(i) (x − 1)(x 4 − 4x 3 + 5x 2 − 4x + 1) = 0
Σαβ = α (α + 2) + (α + 2) β + αβ = 15 Solve the second factor, we get answer
or α + 2α + (2α + 2)(11 − 2α ) = 15 [by Eq. (i)]
2
3± 5 1±i 3
1, ,
or 3α − 20 α − 7 = 0
2 2 2
(3α + 1)(α − 7) = 0 ; α = 7, 1 / 3 29. (a) Answer x 3 + 5x 2 − 7x − 3 = 0

1 2 (b) Answer −4x 3 − 2x 2 − 3x + 5 = 0
when α = 7, β = − 3 and α = , β = 11
3 3
or 4x 3 + 2x 2 + 3x − 5 = 0
[by Eq. (i)]
www.pdfworld.in
www.pdfworld.in
30. x 5 + 11x 4 + 42x 3 + 57x 2 − 13x − 60 = 0 or 8x 9 − 36x 6 + 54x 3 − 27 = x 6 − 8x 3
31. 5x 3 + 17x 2 − 4x − 29 = 0 − 6x 3 (2x 3 − 3)

Putting x 3 = y ,
32. x 5 + x 3 + x 2 + 2x + 3 = 0 …(i)
8y 3 − 36y 2 + 54y − 27 = y 2 − 8y − 6y (2y − 3)
If y be a root of the transformed equation,
then or 8y 3 − 25y 2 + 44y − 27 = 0 …(iii)
y =x 2
…(ii) which is required transformed equation
We have to eliminate x between Eqs. (i) and (ii), Q Roots of Eq. (i) are α , β , γ
from Eq. (i) ∴ Roots of Eq. (iii) are α3 , β3 , γ 3
x 5 + x 3 + 2x = − (x 2 + 3) Σα3β3 =
44 11
=
8 2
On squaring both sides
x 10 + x 6 + 4x 2 + 2x 8 + 4x 4 + 4x 6 35. Let α , β , γ are roots of x 3 + qx + r = 0, equation
in y whose roots are α 2 , β 2 , γ 2 is obtained by
= x 4 + 6x 2 + 9
putting y = x 2,
or y 5 + y 3 + 4y + 2y 4 + 4y 2 + 4y 3
x (x 2 + q ) = − r or y (y + q ) = − r
= y 2 + 6y + 9
or y 5 + 2y 4 + 5y 3 + 3y 2 − 2y − 9 = 0
y ( y 2 + 2qy + q 2 ) = r 2
or y 3 + 2qy 2 + q 2y − r 2 = 0
33. x 3 + 3x 2 + 2 = 0 …(i)
Comparing this with given equation, we
If y be a root of transformed equation, then
conclude that its second term should be
y = x3 …(ii) removed by h where na 0 h + a1 = 0
we have to eliminate x between Eqs. (i) and (ii) or 3h + 2q = 0
from Eq. (i), ∴ h = − 2q / 3
x 3 + 2 = − 3x 2 −2q / 3 1 2q q2 −r 2
Cubing both sides −2q / 3 −8q 2 / 9 −2q3 / 27
x 9 + 8 + 3x 32(x 3 + 2) = − 27x 6 −4q / 3 q2 / 9

(−2q3 / 27 − r 2 )
or x 9 + 33x 6 + 12x 3 + 8 = 0 −2q / 3 −4q / 9
2
Putting x 3 = y , 2q / 3 −q 2 / 3
0
y 2 + 33y 2 + 12y + 8 = 0
which is the required equation.
1 2 2 3
34. 2x 3 − x 2 + 2x − 3 = 0 …(i) z3 − q z− q − r2 = 0
3 27
If y be a root of transformed equation, then Whose roots are α 2 − h , β 2 − h , γ 2 − h
y =x 3
…(ii)
or α 2 + 2q / 3, β 2 + 2q / 3, γ 2 + 2q / 3
We have to eliminate x between Eqs. (i) and (ii),
which differ from the squares of the roots of
from Eq. (i)
x 3 + qx + r = 0 by a constant −2q / 3.
2x 3 − 3 = x 2 − 2x
36. Put x 3 = y ,
(2x 3 − 3)3 = x 6 − 8x 3 − 3x 2⋅ 2x (x 2 − 2x ) ∴ y + ab = − (ay 2/3 + by 1/3 ) …(i)
www.pdfworld.in
www.pdfworld.in
On cubing both sides, we get or r 2y 3 + pry 2 (r + 1) + qy (r + 1)2 + (r + 1) 3 = 0

y 3 + 3y 2ab + 3ya 2b 2 + a3b3 or r 2y 3 + pr (r + 1) y 2 + q (r + 1)2y + (r + 1) 3 = 0
= − [a3y 2 + b3y + 3aby (ay 2/3 + by 1/3 )] which is the required equation.
y + 3aby 2 + 3a 2b 2y + a3b3
3
(b) If the transformed equation is in terms of
y.
= − a y − b y − 3aby {− ( y + ab )}, [by Eq. (i)]
3 2 3
y = α ( β + γ ) = ( αβ + βγ + γα ) − βγ
∴ y 3 + a3y 2 + b3y + a3b3 = 0 αβγ r
=q − =q + [QEq. (ii)]
37. Since, α , β , γ are the roots of α α
r r
z3 + 3Hz + G = 0 …(i) ∴ y =q + or y − q =
x x
∴ α+β+ γ=0 r
or x =
If the transformed equation is in terms of y, y −q
then Putting this value of x in Eq. (i), we get
α+1 α+1 α+1
y = = = r3 r2 r
β + γ − α α + β + γ − 2α 0 − 2α +p +q +r =0
(y − q ) 3
( y − q )2 y −q
z+1
∴ y = or −2yz = z + 1 or r 2 + pr ( y − q ) + q ( y − q )2 + ( y − q )3 = 0
−2z
1 or y 3 − 2qy 2 + ( q 2 + pr ) y + ( r 2 − pqr ) = 0
or (1 + 2y )z = − 1, ∴ z = −
1 + 2y which is the required equation.
Putting the value of z in Eq. (i), we get (c) If the transformed equation is in terms of y.
−
1
−
3H
+G=0 1 αβγ + 1 r +1
y = βγ + = =− [QEq. (ii)]
(1 + 2y )3 (1 + 2y ) α α α
G(1 + 2y )3 − 3H (1 + 2y )2 − 1 = 0 1−r 1−r
∴ y = or x =
x y
⇒ G(1 + 6y + 12y 2 + 8y 3 )
−3H (1 + 4y + 4y 2 ) − 1 = 0 (1 − r )3 (1 − r )2 1−r
+p +q +r =0
⇒ 8Gy + 12(G − H ) y + 6(G − 2H ) y
3 2
y 2
y2 y
+ (G − 3H + 1) = 0 or ry 3 + q (1 − r ) y 2 + p (1 − r )2y + (1 − r )3 = 0
which is the required equation. which is the required equation.
38. x + px + qx + r = 0
3 2
…(i) 39. Qα , β , γ are the roots of equation
∴ α + β + γ = − p , αβ + βγ + γα = q, x 3 + x 2 + 2x + 3 = 0
αβγ = − r …(ii) ∴ α + β + γ = −1 …(i)
(a) If the transformed equation is in terms of y. If y is a root of the transformed equation.
1 α α α y = β + γ − α = (α + β + γ ) − 2α = − 1 − 2α
y =α− =α − =α − =α +
βα αβγ −r r 1+y
∴ y = − 1 − 2x or x = −
[QEq. (ii)] 2
x (r + 1) x Putting this value of x in Eq. (i), we get
∴ y =x + =
r r (1 + y )2 (1 + y )2
ry − + − (1 + y ) + 3 = 0
∴ x = 8 4
r +1
( y + 1)3 − 2( y + 1)2 + 8( y + 1) − 24 = 0
or y 3 + y 2 + 7y − 17 = 0
r 3y 3 r 2y 2
ry
+p⋅ +q⋅ +r =0
(r + 1)3 (r + 1)2 r +1 which is the required equation.
www.pdfworld.in
www.pdfworld.in
40. Hint  1 1 1 1 
⇒ s 1 + m + + + 
y = 2α − β − γ = 3α − (α + β + γ ) = 3α − a  a − m b − m c − m d − m 
y +a
∴ y = 3x − a or x = k l p q
3 = + + +
a −m b −m c −m d −m
Answer
y 3 + 3(3b − a 2 ) y + (9ab − 2a3 − 27) = 0 where from we find s and then from the
equalities (i) we obtain the required values of
and also the product of the roots is
x , y , z.
2a3 + 27c − ab
43. Adding all the three equations, we get
41. Adding all the four equations, we get (x + y + z )(a + b + c ) = 0
4x 1 = 2a1 + 2a 2 + 2a3 + 2a 4
Hence, x + y + z = 0
a + a 2 + a3 + a 4
x1 = 1 a −b a −c b −a
2 ∴ x = ;y = ;z =
a+b+c a+b+c a+b+c
Multiplying the last two equations by –1 and
then adding all the four equations, we get 44. We find similarity,
a + a 2 − a3 − a 4 t = − (a + b + c + d )
x2 = 1
2 x = ab + ac + ad + bc + bd + cd
a1 − a 2 + a3 − a 4
Similarly, x 3 = , y = − (abc + abd + acd + bcd )
2
z = abcd
a − a 2 − a3 + a 4
x4 = 1
2 45. Multiplying the given equations, we get
42. Put x + y + z + v = s, then the following (xyz )2 = abcxyz

system can be rewritten as First of all we have an obvious solution
ax + m (s − x ) = k, x = y = z = 0, then xyz = abc
by + m (s − y ) = l From the original equations, we find
cz + m (s − z ) = p xyz = ax 2 ; xyz = by 2 ; xyz = cz 2
dv + m (s − v ) = q
Hence, ax 2 = abc ; by 2 = abc ; cz 2 = abc
So that, ms + x (a − m ) = k ; ms + y (b − m ) = l
x 2 = bc ; y 2 = ac ; z 2 = ab
ms + z (c − m ) = p;
ms + v (d − m ) = q Thus, we have following solution set
k m x = bc ; y = ac ; z = ab
Hence, x = − s;
a −m a −m
x = − bc ; y = − ac ; z = ab
l m
y = − s;
b −m b −m x = bc ; y = − ac ; z = − ab
p m x = − bc ; y = ac ; z = − ab
z= − s;
c −m c −m
46. The system is reduced to the form
q m
v = − s …(i) xy + xz = a 2
d −m d −m
yz + yx = b 2
Adding these equalities termwise, we get
k l p q zx + zy = c 2
s= + + +
a −m b −m c −m d −m Adding these equations term by term, we get
 1 1 1 1  1
xy + xz + yz = (a 2 + b 2 + c 2 )
− ms  + + + 
a − m b − m c − m d − m  2
www.pdfworld.in
www.pdfworld.in
Taking into consideration the first three and consequently

equations b2 + c 2 − a2 a2 + c 2 − b2
α= ;β = ;
b2 + c 2 − a2 a2 + c 2 − b2 2bc 2ac
yz = , zx =
2 2
a2 + b2 − c 2
a2 + b2 − c 2 γ=
xy = 2ab
2
x + y + cxy 1
Multiplying them, we have Further =
x +y γ
(b 2 + c 2 − a 2 )(a 2 + c 2 − b 2 )
(a 2 + b 2 − c 2 ) cxy 1
(xyz )2 = or = −1
8 x +y γ
(b 2 + c 2 − a 2 )(a 2 + c 2 − b 2 ) x +y γ
or =
(a 2 + b 2 − c 2 ) cxy 1−γ
i.e., xyz = ±
8 1 1 cγ
Finally + =
Now, we easily find x y 1−γ
(a 2 + c 2 − b 2 ) (a 2 + b 2 − c 2 ) Analogously, we get
x =±
8 (b 2 + c 2 − a 2 ) 1 1 bβ 1 1 aα
+ = ; + =
x z 1−β y z 1−α
(a 2 + b 2 − c 2 ) (b 2 + c 2 − a 2 )
y =±
8 (a 2 + c 2 − b 2 ) Where from we find x , y , z.
(a + c − b )(b + c − a )
2 2 2 2 2 2 49. Multiplying the first, second and third
z=± equations respectively by y , z and x , we get
8(a 2 + b 2 − c 2 )
cx + ay + bz = 0
47. Adding the first two equations and
subtracting the third one, we get Likewise multiplying these equations by z , x
2x 2 = (c + b − a ) xyz and y, we get
Likewise, we find bx + cy + az = 0
x y z
2y = (c + a − b ) xyz ; 2z = (a + b − c ) xyz
2 2
Now, = 2 = 2 =λ
a − bc b − ac c − ab
2
Solving out the solution, we get

x =y = z = 0 i.e., x = (a 2 − bc ) λ
We have, y = (b 2 − ac )λ
2x = (c + b − a )yz ; 2y = (c + a − b ) xz; z = (c 2 − ab )λ
2z = (a + b − c ) xy
Substituting these expressions into the third
Now, proceed as in Q. 6.
equation, we find
x +y y + z
48. Put =γ; =α c
x + y + cxy y + z + ayz λ2 = 2
(c − ab )2 − (a 2 − bc )(b 2 − ac )
x + z
and =β 1
x + z + bxz =
a + b + c 3 − 3abc
3 3
Then, the system takes the form

Then, we can easily find x , y , z.
bγ + cβ = a ; cα + aγ = b ; aβ + bα = c
50. x 2 + y 2 + z 2 = 84 …(i)
γ β a α γ b
or + = ; + = ;
c b bc a c ac x + y + z = 14 …(ii)
β α c xz = y 2 …(iii)
+ =
b a ab
Squaring Eq. (ii), we have
α β γ 1 a2 + b2 + c 2
∴ + + = x 2 + y 2 + z 2 + 2xy + 2yz + 2zx = 196
a b c 2 abc
www.pdfworld.in
www.pdfworld.in
Putting the values of x 2 + y 2 + z 2 and xz from y = 2, 5

Eqs. (i) and (iii), we get z = 6, 6
84 + 2xy + 2yz + 2y 2 = 196 When z = 7 ⇒x + y = 6 from Eq. (i) and xy = 10
2xy + 2yz + 2y 2 = 112 x and y are now roots of t 2 − 6t + 10 = 0
xy + yz + y 2 = 56 6± 36 − 40
or t =
y (x + y + z ) = 56 2
t =3± i
Putting the values of x + y + z from Eq. (ii), we
get Hence, x =3± i
14y = 56 ⇒ y = 4 y =3± i
When y = 4 ⇒ xz = 16 from Eq. (iii) ∴ x = 5, 2, 3 ± i
y = 4 ⇒ x + z = 10 from Eq. (ii) y = 2, 5, 3 ± i
Hence, x and z are the roots of the quadratic x = 6, 6, 7, 7

equation p 2 − 10p + 16 = 0 . 52. We have, xz + y = 7z …(i)
Roots of this equations are 8 and 2. yz + x = 8z …(ii)
Hence, x = 8 or 2 x + y + z = 12 …(iii)
z = 2 or 8 Adding Eqs. (i) and (ii), we get
∴ x = 8 or 2
y =4 z (x + y ) + (x + y ) = 15z
z = 2 or 1 (z + 1)(x + y ) = 15z …(iv)
51. x + y + z = 13 …(i) From Eq. (iii) x + y = 12 − z
x + y + z = 65
2 2 2
…(ii) Putting the value of x + y in Eq. (iv), we get
(z + 1)(12 − z ) = 15z
xy = 10 …(iii)
Squaring Eq. (i) or 12z + 12 − z 2 − z = 15z
x 2 + y 2 + z 2 + 2xy + 2yz + 2zx = 169 or z 2 + 4z − 12 = 0

Putting the values of x 2 + y 2 + z 2 and 2xy, or (z + 6)(z − 2) = 0
we get ∴ z = − 6 or z = 2
65 + 20 + 2yz + 2zx = 169 When z = − 6,
or 2yz + 2zx = 84 x + y = 18 from Eq. (iii) and −6x + y = − 42
yz + zx = 42 from Eq. (i).
z (x + y ) = 42 Multiplying first of these equations by 6 and
But x + y = 13 − z [from Eq. (i)] adding to the later, we get
So, z (13 − z ) = 42 6x + 6y = 108
or z − 13z + 42 = 0
2
− 6x + y = − 42
⇒ z = 6 or 7 7y = 66
When z = 6, x + y = 7 [from Eq. (i)] or y =
66
Now, x + y = 7 and xy = 10 7
But x + y = 18
Hence, x and y are the roots of t 2 − 7t + 10 = 0 .
66 60
Roots of this equation are 5, 2. ∴ x + = 18 or x =
7 7
Hence x = 5, y = 2
When we put z = 2 and solve for x and y in a
x = 2, y = 5
similar way, we get x = 4, y = 6
Hence, the first set of values is 60 66
Hence, x = ,y = , z = − 6, 2
x = 5, 2 7 7
www.pdfworld.in
www.pdfworld.in
53. x + y + z =a …(i) x = y which is impossible. The system admits

x + y + z =b
2 2 2 2
…(ii) solutions with the desired properties, if
|b | < a < |b | 3 .
xy = z 2
…(iii)
Remark
Subtract Eq. (ii) from the square of Eq. (i) and
substitute xy by z 2 Let us sketch the geometric background of the
problem (1) is simply the equation of a plane
2xy + 2xz + 2yz = a 2 − b 2 intersecting the axis in points of distance a from
the origin, consequently its distance from the
2z 2 + 2z (x + y ) = a 2 − b 2 a
origin is . (2) is the equation of the sphere of
Now, substitute x + y with a − z 3
2z 2 + 2az − 2z 2 = a 2 + b 2 radius | b | around the origin. The two objects have
common point if and only if
a = 0 implies b = 0, hence from Eq. (ii), we get a
|b | ≥ i.e., a ≤ | b | 3
x = y = z = 0 which according to additional 3
hypothesis is not a solution. (3)is the equation of a cone with the origin at its
Consequently we may assume that a ≠ 0 vertex.
a2 − b2 (a 2 − b 2 )2 54. y 2 + yz + z 2 = ax …(i)
z= , z2 =
2a 4a 2 z + zx + x = ay
2 2
…(ii)
Now, x + y and xy can be expressed in terms of x + xy + y = az
2 2
…(iii)
a and b
Multiplying Eq. (iii) by y and Eq. (ii) by z and
a2 + b2
x + y =a − z = then subtracting, we get
2a
(z3 − y 3 ) + x (z 2 − y 2 ) + x 2 (z − y ) = 0
(a 2 − b 2 )2
and xy = z = 2
…(iv) (z − y )(z 2 + y 2 + zy + xz + xy + x 2 ) = 0 …(iv)
4a 2
Similarly, (x − y )(x + y + z + yz
2 2 2
This yield a quadratic equation in u
+ zx + xy ) = 0 …(v)
4a 2u 2 − 2a (a 2 + b 2 ) u + (a 2 − b 2 )2 = 0 …(v)
i.e., by Eqs. (iv) and (v) we have either
Which is solved by x and y, solutions are x =y = z
1 x 2 + y 2 + z 2 + zx + xy + yz = 0
u1, 2 = {2a (a 2 + b 2 )} or
8a 2
Now, if x = y = z then by Eq. (i), 3x 2 = ax
± 4a (a + b ) − 16a (a − b )
2 2 2 2 2 2 2 2
a
i.e., x = 0 or
1 3
= (a 2 + b 2 ± (3a 2 − b 2 )(3b 2 − a 2 )
4a If x 2 + y 2 + z 2 + xy + yz + zx = 3, then by Eq. (i)
Hence, the solution of the system (when a ≠ 0) x 2 + xy + xz + ax = 0
a2 − b2 i . e. , x = 0 or (x + y + z ) = 0
x = u1 ; y = u2 ; z =
2a In this case the solution is indeterminate, for
a2 − b2 the given equations hold if the relation
x = u2 ; y = u1 ; z =
2a x + y + z = −a
Sufficient conditions for the solutions to exist and x 2 + y 2 + z 2 + zx + xy + yz = 0
Eq. (i) implies a > 0 ; for z > 0, we also need
are satisfied.
a > | b | Eq. (iv) implies that if the solutions
exist, then both x and y are +ve, x and y are 55. Given equations can be rewritten as
a b
distinct, if the discriminant in Eq. (v) is +ve ; = y + z − x ; = z + x − y;
since a > | b | , this positively implies | b | 3 > a. x y
In this case neither x nor y is equal to z. If, for c
=x + y − z
example, x = z from Eq. (iii) we have y = z so z
www.pdfworld.in
www.pdfworld.in
a b c ax 2 − b  a 2x − b 2 
  = bx + a
i . e. , x = y = z a −b  a − bx 
y + z −x z + x −y z + y − z
or (ax 2 − b )(a 2x − b 2 ) = (bx + a )(a − bx )2
b c 
 +  (a3 − b3 )(x 3 − 1) = 0 i.e., x 3 − 1 = 0
y z  bz + cy
= =
2z 2xyz 1±i 3
which gives three values of x as 1,
2
Hence, bz + cy = cx + azx + ay + bx
ax 2 − b
i . e. , cx − cy + (a − b )z = 0 ∴ By y = , we have
a − bx
and bx + (a − c ) y − bz = 0
For x = 1,
x y z
∴ = = a −b
a (−a + b + c ) b (a − b + c ) c (a + b − c ) y = =1
a −b
= k (say) −1 − i 3
For x =
Putting the values of x , y , z from above in 2
x (y + z − x ) = a a (1 − 3 − 2i 3 )
−b
We have, 2 −1 − i 3
y = =
k 2a (−a + b + c )(a − b + c )(a + b − c ) = a i + i 3 2
a −b 
 2 
1
i . e. , k = ± −1 − i 3
(−a + b + c )(a − b + c )(a + b − c ) For x = ,
2
…(i)
−1 − i 3
y =
x = ka (−a + b + c ) 2
y = kb (a − b + c ) −1 − i 3 −1 − i 3
If ω= , then ω2 =
z = kc (a + b − ac ) 2 2
while value of k is as in Eq. (i) i . e. , x = 1, ω, ω2
56. (a) ax + by + z = 0 …(i) y = 1, ω2 , ω
zx + ay + b = 0 …(ii) Then, by z = − (ax + by )
yz + bx + a = 0 …(iii) We have,
From Eq. (i), z = − (ax + by ) z = − (a + b ), − (aω + bω2 ), − (aω2 + bω)
Hence Eqs. (ii) and (iii) after removing z gives
57. x (p + q ) = y …(i)
x (ax + by ) = ay + b
p − q = k (1 + pq ) …(ii)
y (ax + by ) = bx + a
xpq = a …(iii)
x (ax + by ) = ay + b
y
From Eq. (i) p + q =
ax 2 − b
y = x
a − bx
y2 Qpq = a 
∴ p 2 + q 2 + 2pq =
ax − b 2
x2  x 
∴ ax + by = ax + b ⋅
a − bx 2a y 2
p2 + q2 + = 2
a 2x − abx 2 + bax 2 − b 2 x x
=
a − bx y2 2a
p2 + q2 = − …(iv)
a x −b
2 2 x2 x
= Squaring and adding Eqs. (i) and (ii), we have
a − bx
2
y2
+ k 2  1 + 
a
∴ y (ax + by ) = bx + a gives 2( p 2 + q 2 ) =
x 2  x
www.pdfworld.in
www.pdfworld.in
Hence, from Eq. (iv) expression becomes y z

or + + 2 = 4a 2
y z y
2a  y
2 2 2
2 a
2 2 −  = 2 + k  1 +  z x
 x x  x x Similarly, + + 2 = 4b 2
x z
2( y 2x − 2ax ) = y 2 + k 2 (x + a )2 x y
and + + 2 = 4c 2
y 2 − 4ax = k 2 (a + x )2 y x
x
58. 4(x 2 + y 2 ) = ax + by …(i) 4a 2 + 4b 2 + 4c 2 = 6 + Σ
y
2(x 2 − y 2 ) = ax − by …(ii)
Hence, from Eq. (i)
xy = c 2
…(iii)
4a 2 + 4b 2 + 4c 2 = 6 ± 8abc − 2
a 2 + b 2 + c 2 = 1 ± 2abc
ax = 3x 2 + y 2 …(iv)
a 2 + b 2 + c 2 ± 2abc = 1
by = x 2 + 3y 2 …(v)
60. x 2y = a ; x (x + y ) = b ; y = c − 2x
Multiplying Eq. (iv) by y and Eq. (v) by x , we get
axy = 3x 2y + y 3 Putting the value of y in first two equations
x 2 (c − 2x ) = 0
ac 2 = 3x 2y + y 3 …(vi)
or 2x 3 − cx 2 + a = 0 …(i)
bxy = 3y 2x + x 3 (Qxy = c 2 )
x + x (c − 2x ) = b
2
bc 2 = 3y 2x + x 3 …(vii)
x 2 − cx + b = 0 …(ii)
Adding and subtracting Eqs. (vi) and (vii)
Multiplying Eq. (ii) by 2x and subtracting
c 2 (a + b ) = (x + y )3
from Eq. (i)
(x + y ) = (a + b )1/3c 2/3 …(viii)
cx 2 − 2bx + a = 0 …(iii)
c (a − b ) = ( y − x )
2 3
From Eqs. (i) and (iii) by cross multiplication
( y − x ) = (a − b )1/3c 2/3 …(ix)
x2 x 1
= =
From Eqs. (viii) and (ix) ac − 2b 2 a − bc 2b − c 2
( y + x )2 − ( y − x )2 = (a + b )2/3c 4/3
a − bc ac − 2b 2
x = ;x2 =
− (a − b ) 2/3 4/3
c 2b − c 2 2b − c 2
4xy = c 4/3
[(a + b ) 2/3
− (a − b ) 2/3
] 2
 a − bc  ac − 2b 2
∴   =
2
[Qx 2 = (x )2 ]
4c = c
2 4/3
[(a + b )
2/3
− (a − b ) 2/3
]  2b − c  2b − c 2
4c 2/3 = (a + b )2/3 − (a − b )2/3
(a − bc )2 = (ac − 2b 2 )(2b − c 2 )
59. Multiplying these equations
a 2 + b 2c 2 − 2abc = (2abc − 4b3 − ac 3 + 2b 2c 2 )
( y + z )2 (z + x )2 (x + y )2 = 64a 2b 2c 2 (xyz )2
a 2 − 4abc + ac 3 + 4b3 − b 2c 2 = 0
∴ ( y + z )(z + x )(x + y ) = ± 8abcxyz
( yz + xy + zx + z 2 )(x + y ) = ± 8abcxyz 61. Given equations are
ax 2 + by 2 = c …(i)
2xyz + y 2z + x 2y + xy 2 + zx 2 + z 2x + x 2y
ax + by = c …(ii)
= ± 8abcxyz
xy
y x y x z z
2 + + + + + + = ± 8abc …(i) =c …(iii)
x z z y y x x +y
On dividing by xyz, we get From Eqs. (i) and (ii), we get

y + z + 2yz = 4a yz
2 2 2 c (ax 2 + by 2 ) = c 2 = (ax + by )2
www.pdfworld.in
www.pdfworld.in
cax 2 + cby 2 = a 2x 2 + 2abxy + b 2y 2 62. ax + yz = bc …(i)

by + zx = ca …(ii)
a (a − c ) x 2 + 2abxy + b (b − c ) y 2 = 0 …(iv)
cz + xy = ab …(iii)
From Eqs. (ii) and (iii), we get
xyz = abc …(iv)
xy
ax + by = Squaring Eq. (i) and with the help of Eq. (iv), we
x +y
get
ax 2 + bxy + axy + by 2 = xy a 2x 2 + y 2z 2 + 2a 2bc = b 2c 2 …(v)
ax + xy (a + b − 1) + by = 0
2 2
… (v) [Q2axyz = 2a 2bc ]
From Eqs. (iv) and (v) by the method of cross Similarly squaring Eqs. (ii) and (iii), and applying
multiplication Eq. (iv)
x2 xy b 2y 2 + z 2y 2 + 2b 2ca = c 2a 2 …(vi)
=
b {2ab − (b − c )(a + b − 1)} −ab (a − b ) c z + x y + 2c ab = a b
2 2 2 2 2 2 2
…(vii)
y2 Multiplying Eqs. (i), (ii) and (iii), we get

= =k (say)
a {(a − c )(a + b − 1) − 2ab} (ax + yz )(by + zx )(cz + xy ) = a 2b 2c 2
∴ x 2y 2 = k 2 ⋅ ab {2ab − (b − c )(a + b − 1)} or a 2b 2c 2 = (abxy + ax 2z + by 2z + xyz 2 )
{(a − c )(a + b − 1) − 2ab} (cz + xy )

Also xy = − kab (a − b ) = abcxyz + cax 2z 2 + bcy 2z 2 + cxyz3 + abx 2y 2
∴ ab {2ab − (b − c )(a + b − 1)} + ax 2yz + bxy 3z + x 2y 2z 2

{(a − c )(a + b − 1) − 2ab} = (ab )2 (a − b )2 = a 2b 2c 2 + xyz (ax 2 + by 2 + cz 2 ) + abx 2y 2
or {2ab − (b − c )(a + b − 1)}{2ab − (a − c ) + bcy 2z 2 + cax 2z 2 + x 2y 2z 2
(a + b − 1)} = − ab (a − b )2 With the help of Eqs. (iv), (v), (vi), (vii)
or 4a b − 2ab (a + b − 1)(a − c )(b − c )
2 2 = 2a 2b 2c 2 + abc (ax 2 + by 2 + cz 2 ) + abx 2y 2
+ (b − c )(a − c )(a + b − 1)2 + bcy 2z 2 + cax 2z 2

= 2a b c + bc (a x + y z ) + ca (b 2y 2 + z 2x 2 )
2 2 2 2 2 2 2
= − ab (a − b )2
+ ab (c 2z 2 + x 2y 2 )
or 4a 2b 2 − 2ab (a + b − 1)(ab − ab − bc + c 2 )
= 2a 2b 2c 2 + bc (b 2c 2 − 2a 2cb )
+ (ab − bc − ac + c 2 ) (a + b − 1)2
+ ca (c 2a 2 − 2b 2ca ) + ab (a 2b 2 − 2abc )2
= − ab (a − b )2
or a 2b 2c 2 = 2a 2b 2c 2 + b3c 3 − 2a 2b 2c 2 + c 3a3
or 2ab (a + b − 1)(ab − ac − bc + c 2 )
− 2a 2b 2c 2 + a3b3 − 2a 2b 2c
− (a + b − 1) (ab − bc − ac + c )
2 2
or b3c 3 + c 3a3 + a3b3 = 5a 2b 2c 2
−4a 2b 2 = ab (a − b )2
63. We have, y 3 − (x + 1)3 = x 3 + 8x 2 − 6x + 8
or (a + b − 1)(ab − ac − bc + c )
− (x 3 + 3x 2 + 3x + 1) = 5x 2 − 9x + 7
(2ab − a − b + 1) − 4a 2b 2 = ab (a − b )2
Consider the quadratic equation
or (a + b − c )(ab − ac − bc + c 2 ) 5x 2 − 9x + 7 = 0. The discriminant of this
(2ab − a − b + 1) = ab (a + b )2 equation is D = 92 − 4 × 5 × 7 = − 59 < 0 and
hence the expression 5x 2 − 9x + 7 is positive
or c 2 (a + b − c )2 − c (a + b − 1)
for all real values of x . We conclude that
(a 2 − 2ab + b 2 − a − b ) + ab = 0 (x + 1)3 < y 3 and hence x + 1 < y .
www.pdfworld.in
www.pdfworld.in
On the other hand we have (x + 2)3 + (2x + 1)3 = 27(x + 1)3

(x + 3)3 − y 3 = x 2 + 9x 2 + 27x This can be solved as earlier or expanding
+ 27 − (x 3 + 8x 2 − 6x + 8) every thing and simplifying the relation.
= x 2 + 33x + 19 > 0 66. If x ≥ 0, then the given equation assumes the
form,
for all positive x . We conclude that y < x + 3.
Thus, we must have y = x + 2. Putting this x 2 + (a − 5) x + 1 = 0 …(i)
value of y, we get If x < 0, then it takes the form
0 = y 3 − (x + 2)3 = x 3 + 8x 2 − 6x + 8 x 2 + (a + 1) x + 1 = 0 …(ii)
− (x + 6x + 12x + 8) = 2x − 18x
3 2 2
For these two equations to have exactly three
We conclude that x = 0 and y = 2 or x = 9 and distinct real solutions we should have,
y = 11. (I) either (a − 5)2 > 4 and (a + 1)2 = 4
64. Let us consider x 4 − 2ax 2 + x + a 2 − a = 0 as a (II) or (a − 5)2 = 4 and (a + 1)2 > 4
quadratic equation in a. We see that the roots Case I From (a + 1)2 = 4, we have a = 1 or −3.
are
But only a = 1 satisfies (a − 5)2 > 4. Thus, a = 1
a = x 2 + x ,a = x 2 − x + 1
Also, when a = 1, Eq. (i) has solutions
Thus, we get a factorisation x = 2 + 3, and Eq. (ii) has solutionsx = − 1, − 1.
(a − x 2 − x )(a − x 2 + x − 1) = 0 As 2 ± 3 > 0 and − 1 < 0, we see that a = 1 is
indeed a solution.
It follows that x 2 + x = a or x 2 − x + 1 = a
Case II From (a − 5)2 = 4, we have a = 3 or 7
Solving these, we get
Both these values of a satisfy the inequality
−1 ± 1 + 4a
x = (a + 1)2 > 4. When a = 3, Eq. (i) has solutions
2
x = 1, 1 and Eq. (ii) has the solutions
−1 ± 4a − 3
or x = x = − 2 ± 3. As 1 > 0 and −2 ± 3 < 0, we see
2 that a = 3 is in fact a solution.
Thus, all the four roots are real if and only if When a = 7, Eq. (i) has solutions x = − 1, − 1,
3
a≥ . which are negative contradicting x ≥ 0.
4
Thus, a = 1, a = 3 are the two desired values.
65. By setting u = x 2 + x − 2 and v = 2x 2 − x − 1, we
67. Since, α and β are the roots of the equation
observe that the equation reduces to x 2 + mx − 1 = 0, we have α 2 + m α − 1 = 0,
u3 + v 3 = (u + v )3. Since −2 −2
β 2 + m β − 1 = 0. Multiplying by αn , βn
(u + v )3 = u3 + v 3 + 3uv (u + v ), it follows that n −1 n −2
respectively we have α + m α n
−α =0
uv (u + v ) = 0. Hence, u = 0 or v = 0 or u + v = 0. n −1 n −2
and β + m β
n
−β = 0.
Thus, we obtain x2 + x −2 = 0 or
Adding we obtain
2x − x − 1 = 0 or x − 1 = 0. Solving each of
2 2
αn + βn = − m (αn − 1 + βn −1
) + (αn −2
+ βn −2
).
them, we get x = 1, − 2 or x = 1, − 1/2 or
x = 1, − 1. Thus, x = 1 is a root of multiplicity 3 This gives a recurrence relation for n ≥ 2
and the other roots are −1, − 2, − 1/2. λn = − λn −1 + λn − 2, n ≤2 …(i)
Aliter It can be seen that x − 1 is a factor of (a) Now, λ 0 = 1 + 1 = 2 and λ1 = α + β = − m.
Thus, λ 0 and λ1 are integers. By induction,
x 2 + x − 2, 2x 2 − x − 1 and x 2 − 1. Thus, we can
in follows from Eq. (i) that λn is an integer
write the equation in the form
for each n ≥ 0.
(x − 1)3 (x + 2)3 + (x − 1)3 (2x + 1)3
(b) We again use Eq. (i) to prove by induction
= 27(x − 1)3 (x + 1)3 that GCD (λn , λn + 1 ) = 1. This is clearly true
Thus, it is sufficient to solve the cubic for n = 0, as GCD (2, − m ) = 1, by the given
equation condition that m is odd.
www.pdfworld.in
www.pdfworld.in
Let GCD (λn − 2 , λn − 1 = 1, n ≥ 2. If it were to 69. Let α , β , γ be the roots of the given equation.
happen that GCD (λn − 1 , λn ) > 1, take a prime p We have,
that divides both λn − 1 and λn . Then, from 1 b 1
α + β + γ = , αβ + βγ + γα = , αβγ =
Eq. (i), we get that p divides λn − 2 also. Thus, p a a a
is a factor of GCD (λn − 2 , λn − 1 ), a It follows that a , b are positive. We thus obtain,
contradiction. So, GCD (λn − 1 , λn ) = 1. Hence, 3b 1
we have GCD (λn , λn + 1 ) = 1, ∀ n ≥ 0. = 3(αβ + βγ + γα ) ≤ (α + β + γ )2 = 2 ,
a a
68. Consider the equation x 2 + ax + b = 0. It has which gives 0 < 3ab ≤ 1. Moreover,
two roots (not necessarily real), say α and β. b2
= (αβ + βγ + γα )2
Either α = β or α ≠ β. a2
Case I Suppose α = β , so that α is a double = α 2β 2 + β 2γ 2 + γ 2α 2 + 2αβγ (α + β + γ )
root. Since, α 2 − 2 is also a root, the only 2
= α 2β 2 + β 2γ 2 + γ 2α 2 +
possibility is α = α − 2. This reduces to
2 a2
(α + 1)(α − 2) = 0. Hence, α = − 1 or α = 2. Thus,
Observe that a = − 2α and b = α 2. Thus, b2 − 2
= α 2β 2 + β 2γ 2 + γ 2α 2
a2
(a , b ) = ( 2, 1) or (−4, 4).
1 b2
≥
(αβ + βγ + γα )2 =
Case II Suppose α ≠ β. There are four 3 3a 2
possibilities; This implies that 3(b − 2) ≥ b or b 2 ≥ 3. Hence,
2 2
(I) α = α 2 − 2 and β = β 2 − 2;
b ≥ 3, the conclusion follows.
(II) α = β 2 − 2 and β = α 2 − 2;
70. We have three relations
(III) α = α 2 − 2 = β 2 − 2 and α ≠ β;
aα 2 − bα − c = λ ,
(IV) β = α − 2 = β − 2 and α ≠ β
2 2
bα 2 − cα − a = λ ,
(I) Here, (α , β ) = (2, − 1) or (−1, 2). Hence, cα 2 − aα − b = λ ,
(a , b ) = (− (α + β ), αβ ) = (−1, − 2).
where λ is the common value. Eliminating α 2
(II) Suppose α = β 2 − 2 and β = α 2 − 2. Then,
from these, taking these equations pairwise,
α − β = β 2 − α 2 = ( β − α )( β + α ). we get three relations
Since, α ≠ β we get β + α = − 1. However, we (ca − b 2 ) α − (bc − a 2 ) = λ (b − a ),
also have (ab − c 2 ) α − (ca − b 2 ) = λ (c − b ),
α + β = β + α − 4 = ( α + β ) − 2αβ − 4
2 2 2
(bc − a 2 ) − (ab − c 2 ) = λ (a − c )
Thus, −1 = 1 − 2αβ − 4, which implies that Adding these three, we get
αβ = − 1.
(ab + bc + ca − a 2 − b 2 − c 2 )(α − 1) = 0
Therefore, (a , b ) = (− (α + β ), αβ ) = (1, − 1)
(Alternatively, multiplying above relations
(III) If α = α 2 − 2 = β 2 − 2 and α ≠ β , then α = − β.
respectively by b − c , c − a and a − b and
Thus, α = 2, β = − 2 or α = − 1, β = 1. In this adding also leads to this.) Thus, either
case (a , b ) = (0, − 4) and (0, − 1). ab + bc + ca − a 2 − b 2 − c 2 = 0 or α = 1. In the
(IV) Note that β = α 2 − 2 = β 2 − 2 and α ≠ β is first case
identical to (III), so that we get exactly 0 = ab + bc + ca − a 2 − b 2 − c 2
same pairs (a , b ). 1
= [(a − b 2 ) + (b − c )2 + (c − a )2 ]
Thus, we get 6 pairs; 2
(a , b ) = (−4, 4), (2, 1), (−1, − 2), Shows that a = b = c . If α = 1, then we obtain
a − b − c = b − c − a = c − a − b,
(1, − 1), (0, − 4), (0, − 1).
and once again we obtain a = b = c .
www.pdfworld.in
www.pdfworld.in
Level 2
1. Given that, x + y + z = 1 = x 2 + y 2 + z 2 1 1
+
a (bq + cr ) − bc (q + r ) b (cr + ap ) − ca (r + p )
a b c
(x − p ) = ( y − q ) = (z − r ) = k (say) 1 1
x y z + =
c (ap + bq ) − ab ( p + a ) bcqr + carp + abpq
a
(x − p ) = k
x 2. We have,
or ax − ap = kx (x + y )3 = x 3 + y 3 + 3xy (x + y )
ap 3
x (a − k ) = ap ∴ x = = x 3 + y 3 + (x + y )[(x + y )2 − (x 2 + y 2 )]
a −k 2
bq cr And so (x + y )3 = 3(x + y )(x 2 + y 2 ) −2(x 3 + y 3 )
Similarly, y = ;z =
b −k c −k But x 3 + y3 = c
Now, 1 = (x + y + z ) Consequently the result of elimination
a3 = 3ba − 2c
∴ 1 = (x + y + z ) 2
1
= x 2 + y 2 + z 2 + 2( yz + zx + xy ) 3. Let us find the equation whose roots are ,
α2
But x + y + z =1
2 2 2
1 1 1
for which we put y = 2 or x = in the
∴ 2 ( yz + zx + xy ) = 0 γ2 x y
or yz + zx + xy = 0 given equation,
bq cr cr ap x (2x 2 + 1) = − (x 2 + 1)
× + ×
q −k c −k c −k a −k 1 2  1 
or  + 1 = −  + 1
ap bq y y  y 
+ × =0
a −k b −k Squaring, we get
(2 + y )2 = y ( y + 1)2
bcqr (a − k ) + carp (b − k ) + abpq (c − k ) = 0
or 4 + 4y + y 2 = y 3 + 2y 2 + y
k (bcqr + carp + abpq ) = abc (qr + rp + pq )
or y 3 + y 2 − 3y − 4 = 0 …(i)
abc (rp + pq + qr )
∴ k= 1 1 1
bcqr + carp + abpq Its roots are say A , B , C i.e., 2 , 2 , 2 .
α β γ
ap
∴ k= 1 1 1
a −k Let Z = 2 + 2 − 2 = B + C − A = ΣA − 2A
β γ α
ap
∴ x = Z +1
abc (rp + pq + ar ) or Z = − 1 − 2A ; ∴ A = −
a− 2
bcqr + carp + abpq
But A is root of Eq. (i)
 cr  − (Z + 1)3 (Z + 1)2 3(Z + 1)
Qx =  ∴ + + −4=0
 c − k 8 4 2
ap (bcqr + carp + abpq ) or (Z 3 + 3Z 2 + 3Z + 1) − 2(Z 2 + 2Z + 1)
=
a 2p (cr + bq ) − abc (rp + pq ) − 12(Z + 1) + 32 = 0
bcqr + carp + abpq
or Z 3 + Z 2 − 13Z + 19 = 0
=
a (bq + cr ) − bc (q + r ) α
4. Let y =
bcqr + carp + abpq β
Similarly, y =
b (cr + ap ) − ca (r + p ) α β β
[There will be six such ratios as , , ,
bcqr + carp + abpq β α γ
z= γ γ α
c (ap + bq ) − ab (p + q ) , , ]
β α γ
Substituting these values in x + y + z = 1 ∴ α = yβ
www.pdfworld.in
www.pdfworld.in
Now, f ( β ) = 0 and f (α ) = 0 ∴ α + β + γ = − p , αβ + βγ + γα = q ,
∴ f ( βy ) = 0 αβγ = − 4
f ( β ) = β3 + qβ + r = 0 If the transformed equation is in terms of y,
f ( βy ) = β3γ 3 + qβy + r = 0 then
αβγ
Eliminating β between these two y = βγ − α 2 = − α2
α
β3 β 1 −r
= = r
qr (1 − y ) r ( y 3 − 1) qy (1 − y 2 ) = − α2 = − − x 2
α x
− r (1 + y + y 2 ) or x 3 + xy + r = 0 …(ii)
∴ β=
qy (1 + y )
Subtracting Eq. (ii) from Eq. (i), we have
r 1
and β3 = ⋅ px 2 + (q − y ) x = 0
y 1+y
px + (q − y ) = 0 (Qx ≠ 0)
Now, β3 = ( β )3
y −q
r 1 − r 3 (1 + y + y 2 )3 ∴ x =
∴ ⋅ = p
y 1+y q3y 3 (1 + y )3
or r 2 (1 + y + y 2 )3 + q3y 2 (1 + y )2 = 0
( y − q )3 ( y − q )2 y −q
+p⋅ +q⋅ +r =0
is the required equation of 6th degree. p 3
p2 p
5. Let us find the equation whose roots are or ( y − q )3 + p 2 ( y − q )2 + p 2q ( y − q )
squares of the roots of the given equation. Put
+ p3r = 0
y = x 2,
∴ x (x 2 + 11) = 6(x 2 + 1) or y 3 + ( y 2 − 3q ) y 2 + (3q 2 − p 2q ) y
or y ( y + 11) = 6( y + 1) + p3r − q3 = 0
Squaring, we get Which is required equation.
y ( y + 22y + 121) = 36( y + 2y + 1)
2 2
7. Roots of the equation
or y − 14y + 49y − 36 = 0
3 2
…(i) x 3 − ax 2 + bx − c = 0 …(i)
Let its roots be A , B , C which stand for are α , β , γ . If y is a root of the transformed
α 2 , β 2 , γ 2. equation, then
Let Z = β 2 + γ 2 = B + C = ΣA − A = 14 − A y = α + β = (α + β + γ ) − γ = a − γ
∴ A = − (Z − 14). But A is a root of Eq. (i). [Qα + β + γ = a]
∴ − (Z − 14)3 − 14(Z − 14)2 − 49(Z − 14) ∴ y = a − x or x = a − y

Putting this value of x in Eq. (i), we have
− 36 = 0
(a − y )3 − a (a − y )2 + b (a − y ) − c = 0
or Z 3 − 3Z 2 (14) + 3Z (14)2
or y 3 − 2ay 2 + (a 2 + b ) y + (c − ab ) = 0 …(ii)
− (14)3 + 14(z 2 − 28z + 142 )
Which is required equation.
+ 49Z − 49 (14) + 36 = 0
Its roots are α + β , β + γ , γ + α
Z − 28Z + 142Z  3 − 2 + 
3 2 1
or
 4 Changing y into
1
and multiplying by y 3, we
y
− 686 + 36 = 0
get
or Z 3 − 28Z 2 + 245Z − 650 = 0 .
(c − ab ) y 3 + (a 2 + b ) y 2 − 2ay + 1 = 0 …(iii)
6. Q α , β , γ are the roots of
Roots of this equation are the reciprocals of
x 3 + px 2 + qx + r = 0 …(i) the roots of Eq. (ii).
www.pdfworld.in
www.pdfworld.in
1 1 1 9. (a) z3 − ( p 2 − 2q )z 2 − ( p 4 − 4p 2q + 8pr ) z
∴ Roots of Eq. (iii) are , , .
α+β β+ γ γ+α
+ p 4 − 6p 4q + 8p3r + 8p 2q 2
1 1 1
Now, + + = sum of roots of −16pqr + 8r 2 = 0
α+β β+ γ γ+α
Eq. (iii) (b) z3 − 2( p 2 − 2q )z 2 + ( p 4 − 4p 2q + 5q 2 − 2pr )z
−a 2 + b a 2 + b − (p 2q 2 − 2p3r + 4pqr − 2q3 − r 2 ) = 0
= =
c − ab ab − c
10. (a) Here, a 0 = 1, a1 = − 3,a 2 = 5,a3 = − 12, a 4 = 4
8. x 2 + 3x + 2 = 0 …(i) ∴ a 0s1 + a1 = 0 ⇒ s1 − 3 = 0
Q Its roots are α , β , γ . ∴ s1 = 3
∴ α + β + γ = 0, αβ + βγ + γα = 3, αβγ = − 2 a 0s2 + a1s1 + 2a 2 = 0
Let y be a root of the transformed equation ⇒ s2 − 3(3) + 10 = 0
∴ y = ( β − γ ) = ( β + γ ) − 4βγ
2 2 ∴ s2 = − 1
4αβγ a 0s3 + a1s2 + a 2s1 + 3a3 = 0
= (−α )2 − [Qα + β + γ = 0]
α ⇒ s3 − 3(−1) + 5(3) − 36 = 0
8
= α2 + [Q αβγ = − 2] or s3 = − 3 − 15 + 36 = 18
α
Replacing α by x, If α , β , γ , δ are the roots of the given equation,
8 then putting x = α , β , γ , δ in succession and
∴ y =x2 +
x adding.
or x − xy + 8 = 0
3
…(ii) s4 − 3s3 + 5s2 − 12s1 + 16 = 0
∴ s4 − 54 − 5 − 36 + 16 = 0
Subtracting Eq. (ii) from Eq. (i)
s4 = 79
(3 + y )x − 6 = 0
6 Multiplying both sides of given equation by x
∴ x =
3+y x 5 − 3x 4 + 5x 3 − 12x 2 + 4x = 0
Putting this value of x in Eq. (i), we get Putting x = α , β , γ , δ in succession and adding
 6 
3 s5 − 3s4 + 5s3 − 12s2 + 4s1 = 0
6
  + 3⋅ +2=0
3 + y  3+y or s5 − 237 + 90 + 12 + 12 = 0
216 + 18( 3 + y )2 + 2( 3 + y )3 = 0 ∴ s5 = 123

(b) S5 = − 1
y 3 + 18y 2 + 81y + 216 = 0
11. 45
Which is the required equation.
12. Here, a 0 = 1, a1 = 0, a 2 = 2, a3 = 6
Product of all its roots = − 216
∴ ( α − β ) ( β − γ ) ( γ − α ) = − 216
2 2 2 a 0S1 + a1 = 0 ⇒ S1 = 0
a 0S 2 + a1S1 + 2a 2 = 0 ⇒ S 2 = − 4
RHS being –ve, one of the factors.
a 0S3 + a1S 2 + a 2S1 + 3a3 = 0
Q That will make all the three roots imaginary
which is not possible. Every odd degree ⇒ S3 = − 18
equation with real coefficients has at least one Now, given equation can be written as
real root on the LHS (α − β )2 is –ve. x 3 = − 2x 2 − 6x …(i)
∴ α − β is purely imaginary. Multiplying both sides by x, we get

∴ α and β are conjugate complex roots x 4 = − 2x 2 − 6x
Hence, two roots of Eq. (i) are imaginary. Putting x = α , β , γ in sucession and adding
www.pdfworld.in
www.pdfworld.in
S 4 = − 2S 2 − 6S1 = 8 20. We have, (a − b )2 + (a − c )2 = (b − c )2

∴ S4 = 8 ⇒ 2a 2 − 2ab − 2ac + 2bc = 0
Squaring Eq. (i), we get
⇒ a 2 − a (b + c ) + bc = 0
x = 4x + 24x + 36
6 2
…(ii)
⇒ (a − b )(a − c ) = 0
Putting x = α , β , γ in sucession and adding ⇒ a = b or a = c
S 6 = 4S 2 + 24S1 + 108 The equation has no solution if a , b , c are all
= − 16 + 108 distinct.
∴ S 6 = 92
21. We have, ( xy − 1)2 = ( x + 1)2 + ( y + 1)2
Multiplying both sides of Eq. (ii) by x
⇒ ( xy − 1)2 − ( x + 1)2 = ( y + 1)2
x 7 = 4x 3 + 24x 2 + 36x
⇒ ( xy − x − 2)( xy + x ) = ( y + 1)2
Put x = α , β , γ and adding, we get
⇒ x ( xy − x − 2)( y + 1) = ( y + 1)2 …(i)
S 7 = 4S3 + 24S 2 + 36S1
∴ S 7 = − 72 − 96 = − 168 ⇒ ( y + 1)[x ( xy − x − 2) − ( y + 1)] = 0 …(ii)
Also, 2(S 4 − S 6 ) = 2(8 − 92) If y = − 1, then x takes all values from the set of
= 2(−84) = − 168 integers.
S7 = 2(S 4 − S 6 ) Similarly, we also get
19. Here, x > 0 and x ≠ 1 ( x + 1)[y ( xy − y − 2) − ( x + 1)] = 0 …(iii)
Let log 2 x = p as x ≠ 1, p ≠ 0 If x = − 1, then y takes all values from the set of
1
Given inequality becomes p + + 2 cos y ≤ 0 integers.
p
If x ≠ − 1 and y ≠ − 1, then from Eq. (i)
p 2 + 1 + 2p cos y
i . e. , ≤0 x ( xy − x − 2)( y + 1) = ( y + 1)2
p
x ( xy − x − 2) = y + 1 (Qy ≠ − 1)
Case I When p > 0
⇒ x 2y − x 2 − 2x − y − 1 = 0
p 2 + 1 + 2p cos y ≤ 0, ∀ y and p > 0
⇒ y ( x − 1)( x + 1) = ( x + 1)2
( p − 1)2 + 2p (1 + cos y ) ≤ 0
Since, x ≠ − 1, we have
Q p>0 …(i)
y (x − 1) = (x + 1)
∴ 1 + cos y ≥ 0, ( p − 1) ≥ 0
2
x +1
( p − 1)2 + 2p (1 + cos y ) ≥ 0 …(ii) ⇒ y = is an integer.
x −1
Eqs. (i) and (ii) satisfied when
If x = 0, then y = − 1 and when x = 2, y = 3.
( p − 1)2 + 2p (1 + cos y ) = 0
When x = 3, y = 2 for all other values of x , y is
Q( p − 1)2 ≥ 0 and 2p (1 + cos y ) ≥ 0, we get
not an integer.
( p − 1)2 = 0 and 2p (1 + cos y ) = 0 Hence, solution set is
p = 1 and cos y = − 1 ( 3, 2), ( 2, 3), ( x , − 1), (1, y ).
y = (2n + 1)π 22. If x > 1, then y = x 3 + 3x (x 2 − 1)
Solution set is x = 2, y = (2n + 1)π
y > x3 > x > 1
Case II When p < 0
z = 4y 3 − 3y = y 3 + 3y ( y 2 − 1)> y 3 > y > 1
p 2 + 1 + 2p cos y ≥ 0
x = 4z3 − 3z = z3 + 3z (z 2 − 1)> z3 > z > 1
( p + 1)2 − 2p (1 − cos y ) ≥ 0
( p + 1)2 ≥ 0 and −p (1 − cos y ) ≥ 0, ∀ y Thus, x > y > z > x which is impossible and
again x < − 1 and lead to
∴ Solution set is 0 < x < 1 and all y ∈ R.
www.pdfworld.in
www.pdfworld.in
x < z < y < x so x < − 1 ⇒ a = 2 or 0, when a = 2, (1, 0) and (−1, 0) are

so | x | ≤ 1, | y | ≤ 1, | z | ≤ 1 solutions.
We can write x = cos θ, where 0 ≤ θ ≤ π ∴ We have to consider only the possibility that
Now, y = 4 cos3 θ − 3 cos θ = cos 3θ a = 0, we have
z = 4y 3 − 3y = 4 cos3 3 θ − 3 cos 3 θ 2| x | + | x | = y + x 2 and x 2 + y 2 = 1
= cos 3 × 3θ = cos 9 θ ∴ x 2 , y 2 ≤ 1 ⇒ x 2 ≤ | x | and y ≤ 1
x = 4z − 3z = 4 cos 9 θ − 3 cos 9 θ
3 3
Now, 2|x | + | x | ≥ 1 + | x | ≥ 1 + x 2 ≥ y + x 2
= cos 3 × 9 θ = cos 27θ ⇒ y =1
Since, trigonometric functions are periodic, it ∴ x =0
is possible.
Thus, given system has a unique solution if
Thus, cos θ = cos 27θ ⇒ cos θ − cos 27θ = 0 and only if a = 0.
⇒ 2 sin 14θ sin 13θ = 0 25. Roots of the equation p (x ) q (x ) = 0
⇒ sin 14θ = 0 or sin 13θ = 0 i . e. , (ax 2 + bx + c )(−ax 2 + bx + c ) = 0
kπ
So, θ= , where k = 0, 1, 2, K , 12 will be roots of the equations.
13
ax 2 + bx + c = 0 …(i)
kπ
θ= , where k = 0, 1, 2, K , 13
14 −ax + bx + c = 0
2
…(ii)
Solution is (x , y , z ) = (cos θ, cos 3θ, cos 9 θ) If D1 and D2 be discriminants of Eqs. (i) and (ii),
θ takes all above values. then
23. We are given, D1 = b 2 − 4ac and D2 = b 2 + 4ac
x 13x 3 + x 32x 1 + x 33x 2 = − 8x 1x 2x 3 Now, D1 + D2 = 2b 2 ≥ 0 (Qb may zero)
x 1 + x 2 + x 3 = 0 ; x 1x 2 + x 2x 3 + x 3x 1 = a; i . e. , D1 + D2 ≥ 0
x 1x 2x 3 = − a At least one of D1 and D2 ≥ 0 .
and for i = 1, 2, 3, …
i . e. , atleast one of the Eqs. (i) and (ii) has real
x i3 + ax i + a = 0 roots.
Now, x 13 + ax 1 + a = 0 ∴ p (x ) q (x ) = 0 has at least two real roots.
x 32 + ax 2 + a = 0 26. We have,
x 33 + ax 3 + a = 0 x 2 + 2ax +
1
= −a + a 2 + x − 1 
 
16  16 
⇒ (x 13x 3 + x 32x 1 + x 33x 2 ) + a (x 1x 3 + x 2x 1
1
+ x 3x 2 ) + a (x 3 + x 1 + x 2 ) = 0 Let y = x 2 + 2ax + …(i)
16
i . e. , 8a + a = 0 ⇒ a = − 8
2
a 2 + x − 1 
and y1 = − a +   …(ii)
So, given equation is x 3 − 8x − 8 = 0,  16 
One root is −2, other roots are given by ∴ y = y1 …(iii)
x 2 − 2x − 4 = 0 From Eq. (ii) we have
i . e. , x =1± 5 a 2 + x − 1 
y1 + a =  
So, {x 1 , x 2 , x 3} = {−2, 1 − 5 , 1 + 5}  16 
24. If (x , y ) is a solution, then (−x , y ) is also a 1
⇒ x = y12 + 2ay1 +
solution 16
Q There should be only one solution, x = 0 x = y 2 + 2ay +
1
…(iv)
∴ y + a = 1 and y 2 = 1 16
so y =±1 [from Eq. (iii)]
www.pdfworld.in
www.pdfworld.in
Eqs. (i) and (ii) represents parabolas. Both If a > 0, then f (x 1 ) + f (x 2 ) + K + f (x n ) > 0
parabolas are symmetrical about the line If a < 0, then f (x 1 ) + f (x 2 ) + K + f (x n ) ≠ 0
y =x …(v) ∴ No solution.
From Eqs. (i) and (v), we get Case II When (b − 1)2 = 4ac
1
x = x 2 + 2ax + In case 1 and 2 all of
16
f (x 1 ), f (x 2 ), K , f (x n ) ≥ 0
1
⇒ x 2 − (1 − 2a )x + =0 f (x 1 ), f (x 2 ), K , f (x n ) ≤ 0
16
From Eq. (i),
1
(1 − 2a ) ± (1 − 2a )2 −
4 f (x 1 ) + f (x 2 ) + K + f (x n ) = 0
x =
2 f (x 1 ) = f (x 2 ) = K = f (x n ) = 0
1 But f (x i ) = 0 ⇒ ax i2 + (b − 1)x i + c = 0
For real roots, (1 − 2a ) − ≥ 0 2
4
− (b − 1) ± 0
1 1 xi = [Q (b − 1)2 = 4ac ]
1 − 2a > ⇒ a > 2a
2 4
1−b
2 =
(1 − 2a )  1 − 2a  − 1 2a
Hence, x 1 = +  
2  2  16 1−b
Hence, x1 = x 2 = K = xn =
1 − 2a 
2 2a
x 2 =   1 − 2a  − 1
 −   Case III When (b − 1)2 > 4ac
 2   2  16
are real roots and given fact satisfy original Roots of ax i2 + (b − 1) x i + c = 0

equation. are real and unequal.
1
y x2 + 2ax + =y
16
y=x
α β
a<0
Let α and β be roots .
If a < 0, ∀ x 1 ∈[α , β ]
O ax 12 + (b − 1)x 1 + c ≥ 0
x
1
y2 + 2ay + = x i . e. , f (x 1 ) ≥ 0
16
Similarly for all x i ∈[α,β ], (i = 1, 2, 3 , K , n )
27. Given system of equation can be written as
i . e. , f (x i ) ≥ 0
ax 12 + (b − 1)x 1 + c = x 2 − x 1 = f (x 1 ) (say)
But f (x 1 ) + f (x 2 ) + K + f (x n ) = 0
ax 22 + (b − 1)x 2 + c = x 3 − x 2 = f (x 2 ) (say) f (x 1 ) = f (x 2 ) = K = f (x n ) = 0
M ∴ x1 = x 2 = K = xn
M
M then each ax i2 + (b − 1)x i + c = 0
ax n2 − 1 + (b − 1)x n −1 + c = xn − xn −1 So, x 1 = x 2 = K = x n = − (b − 1)
= f (x n − 1) (say) (b − 1)2 − 4ac
2
ax n + (b − 1)x n + c = x 1 − x n = f (x n ) (say) ± .
2a
f (x 1 ) + f (x 2 ) + K + f (x n ) = 0 …(i) Also, ∀ x 1 ∉(α , β ) (i = 1, 2, 3 , K , n )
Case I When (b − 1)2 < 4ac i . e. , f (x i ) ≤ 0
Each roots of ax 12 + (b − 1)x 1 + c = 0 are but f (x 1 ) + f (x 2 ) + K + f (x n ) = 0
imaginary. So, f (x 1 ) = f (x 2 ) = K = f (x n )
www.pdfworld.in
www.pdfworld.in
∴ x1 = x 2 = K = xn Case I Suppose x + y = 0. Then, z = 1 and

then each ax i2 + (b − 1)x i + c = 0 (x , y , z ) = (m , − m , 1), where m is an integer give
one family of solutions.
So, x1 = x 2 = K = xn
Case II Suppose x + y ≠ 0. Then, we must
− (+ b − 1) ± (b − 1)2 − 4ac
= have
2a
x 2 − xy + y 2 + x + y − 2 = 0
when a > 0, we get
This can be written in the form
x1 = x 2 = K = xn
(2x − y + 1)2 + 3( y + 1)2 = 12
(1 − b ) ± (b − 1)2 − 4ac
= Here, there are two possibilities
2a
2x − y + 1 = 0, y + 1 = ± 2;
28. Let p (x ) ≡ (x − 2) q (x ) + p (2), q (x ) would have 2x − y + 1 = ± 3, y + 1 = ± 1
integer coefficients.
Analysing all these cases, we get
Let r be an integer such that p (r ) = 0
(x , y , z ) = (0, 1, 0), (−2, − 3, 6), (1, 0, 0),
Then, p (r ) = (r − 2) ⋅ q (r ) + 13 = 0 (0, − 2, 3), (−2, 0, 3), (−3, − 2, 6).
2
so r − 31. We seek solutions (x , y , z ) which are in
13
arithmetic progression. Let us put
Thus, r − 2 can only equal ± or ± 13 . y − x = z − y = d > 0 so that the equation
Leading to r = 3, 1, 15, or − 11. reduces to the form
Let P (x ) ≡ (x − 10) ⋅ F (x ) + P (10) 3y 2 + 2d 2 = 2d 3.
10 Thus, we get 3y 2 = (d − 1) d 2. We conclude
Leads to r − . So, r can only be 11, 9, 15 or
5 that 2(d − 1) is 3 times a square. This is
5. satisfied if d − 1 = 6n 2 for some n. Thus
Thus, r = 15. d = 6n 2 + 1 and 3y 2 = d 2 ⋅ 2(6n 2 ) giving us
29. Suppose a + b + c ≠ 0 and let the common y 2 = 4d 2n 2. Thus, we can take
value be λ. Then, y = 2dn = 2n (6n 2 + 1). From this we obtain
xb + (1 − x ) c + xc + (1 − x ) a + xa + (1 − x ) b x = y − d = (2n − 1)(6n 2 + 1),
λ=
a+b+c
z = y + d = (2n + 1)(6n 2 + 1).
=1 It is easily verified that
We get two equations (x , y , z ) = [(2n − 1)(6n 2 + 1),
−a + xb + (1 − x ) c = 0,
2n (6n 2 + 1), (2n + 1)(6n 2 + 1)],
(1 − x ) a − b + xc = 0
is indeed a solution for a fixed n and this gives
(The other equation is a linear combination of an infinite set of solutions as n varies over
these two.) Using these two equations, we get natural numbers.
the relations
32. Substituting x = y + (1/2) …(i)
a b c
= = in the equation, we obtain the equation in y
1 −x + x2 x2 −x + 1 (1 − x )2 + x 3
8y 4 + 4y 2 + a − = 0 …(ii)
Since, 1 − x + x 2 ≠ 0, we get a = b = c 2
Using the transformation z = y 2 , we get a
30. Eliminating z from the given set of equations,
quadratic equation in z
we get
3
x 3 + y 3 + {1 − (x + y )} 2 = 1 8z 2 + 4z + a − =0 …(iii)
2
This factors to The discriminant of this equation is 32(2 − a )
(x + y )(x 2 − xy + y 2 + x + y − 2) = 0 which is non-negative if and only if a ≤ 2. For
a ≤ 2, we obtain the roots
www.pdfworld.in
www.pdfworld.in
−1 + 2(2 − a ) Consider α 4 − α 2 + 2α < 4. Since, α > 0, this is

z1 =
4 equivalent to α5 − α3 + 2α 2 < 4α. Using the
−1 − 2(2 − a ) relation (i), we can write 2α 2 − α + 2 < 4α or
z2 =
4 2α 2 − 5α + 2 < 0. Treating this as a quadratic,
For getting real y we need z ≥ 0. Obviously 1
we get this is equivalent to < α < 2. Now,
z2 < 0 and hence it gives only non-real values 2
3
of y. But z1 ≥ 0 if and only if a ≤ . In this case observe that if α ≥ 2, then
2
1 = (α − 1)(α 4 + α3 + 1) ≥ 25 which is
we obtain two real values for y and hence two
1
real roots for the original Eq. (i). Thus, we impossible. If 0< α ≤ , then
conclude that there are two real roots and two 2
3 1 = (α − 1)(α 4 + α3 + 1) < 0 which again is
non-real roots for a ≤ and four non-real roots
2 1
impossible. We conclude that < α < 2.
3 2
for a > . Obviously the sum of all the roots of
2 Similarly, α − α + 2α ≥ 3 is equivalent to
4 2
3
the equation is 2. For a ≤ , two real roots of α5 − α3 + 2α 2 − 3α ≥ 0 which is equivalent to
2
Eq. (ii) are given by y1 = + z1 and y 2 = − z1 . 2α 2 − 4α + 2 ≥ 0. But this is 2(α − 1)2 ≥ 0 which
Hence, the sum of real roots of Eq. (i) is given is valid. Hence, 3 ≤ α 6 < 4 and we get [α 6 ] = 3.
1 1
by y1 + + y 2 + which reduces to 1. It
2 2 34. We write the equation in the form
follows the sum of the non-real roots of Eq. (i) a 2 + 2ap + p 2 + b (3a + 2p ) = 0
3
for a ≤ is also 1. Thus, Hence,
2
1, for a ≤ 3 − (a + p )2
b=
 2 3a + 2p
The sum of non-real roots = 
3
2, for a > is an integer. This shows that 3a + 2p divides
 2
(a + p )2 and hence also divides (3a + 3p )2. But,
33. Suppose α is a real root of the given equation. we have
Then,
(3a + 3p )2 = (3a + 2p + p )2
α5 − α3 + α − 2 = 0 …(i)
= (3a + 2p )2 + 2p (3a + 2p ) + p 2
This gives α5 − α3 + α − 1 = 1 and hence
It follows that 3a + 2p divides p 2. Since, p is a
(α − 1)(α 4 + α3 + 1) = 1. Observe that
α 4 + α3 + 1 ≥ 2α 2 + α3 = α 2 (α + 2). If −1 ≤ α < 0, prime, the only divisors of p 2 are ±1, ± p and
then α + 2 > 0, giving α 2 (α + 2) > 0 and hence ±p 2.

(α − 1)(α 4 + α3 + 1) < 0. Since, p > 3, we also have p = 3k + 1 or 3k + 2.
If α < − 1, then α 4 + α3 = α3 (α + 1) > 0 and hence Case I Suppose p = 2k + 1. Obviously

α + α + 1 > 0.
4 3 3a + 2p = 1 is not possible. Infact, we get
1 = 3a + 2p = 3a + 2(3k + 1) ⇒ 3a + 6k = − 1
This again gives (α − 1)(α 4 + α3 + 1) < 0. which is impossible. On the other hand
The above resoning shows that for α < 0, we 3a + 2p = − 1 gives 3a = − 2p − 1 = − 6k − 3 ⇒
have α5 − α3 + α − 1 < 0 and hence cannot be a = − 2k − 1 and a + p = − 2k − 1 + 3k + 1 = k.
equal to 1. We conclude that a real root α of − (a + p )2
Thus, b= = k 2. Thus,
x 5 − x 3 + x − 2 = 0 is positive (obviously α ≠ 0). (3k + 2p )
Now, using α5 − α3 + α − 2 = 0, we get (a , b ) = (−2k − 1, k 2 ) when p = 3k + 1. Similarly,
α 6 = α 4 − α 2 + 2α 3a + 2p = p ⇒ 3a = − p which is not possible.
Considering 3a + 2p = − p, we get 3a = − 3p or
The statement [ α ] = 3 is
6
equivalent to
a = − p ⇒ b = 0. Hence, (a , b ) = (−3k − 1, 0)
3 ≤ α < 4.
6
where p = 3k + 1.
www.pdfworld.in
www.pdfworld.in
Let us consider 3a + 2p = p 2. Hence, then α is the other root of both Eqs. (ii) and
3a = p − 2p = p (p − 2)and neither p nor p − 2 is
2 (iii). Hence, it follows that Eqs. (ii) and (iii) have
same set of roots. This implies that
divisible by 3. 3a + 2p = − p 2 , then
If
q r p
3a = − p (p + 2) ⇒ a = − (3k + 1)(k + 1). = =
r p q
Hence, a + p = (3k + 1)(−k − 1 + 1) = − (3k + 1)k
This gives b = k 2. Again Thus p = q = r contradicting the given
condition. Hence, both D2 and D3 cannot be
(a , b ) = (− (k + 1)(3k + 1), k ) when p = 3k + 1.
2
negative. We may assume D2 ≥ 0. Thus, we
Case II Suppose p = 3k − 1. If 3a + 2p = 1, have,
then 3a = − 6k + 3 or a = − 2k + 1. We also get q 2 − rp > 0, r 2 − pq ≥ 0.
− (a + p )2 These two give
b=
1 q 2r 2 > p 2qr
− (−2k + 1 + 3k − 1) 2
since p , q , r are all positive. Hence, we obtain
= = − k2
1 qr > p 2 or D3 < 0. We conclude that the
and we get the solution (a , b ) = (−2k + 1, k 2 ). On common root must be between Eqs. (i) and (ii).
the other hand 3a + 2p = − 1 does not have any Thus,
solution integral solution for a. Similarly, pα 2 + qα + r = 0
there is no solution in the case 3a + 2p = p. qα 2 + rα + p = 0
Taking 3a + 2p = − p , we get a = − p and hence
b = 0. We get the solution (a , b ) = (−3k + 1, 0). If Eliminating α 2 , we obtain
3a + 2p = p 2, then 3a = p (p − 2) = (3k − 1)(3k − 3) 2(q 2 − pr ) α = p 2 − qr
giving a = (3k − 1)(k − 1) and hence Since, q 2 − pr > 0 and p 2 − qr < 0, we conclude
a + p = (3k − 1)(1 + k − 1) = k (3k − 1). This gives that α < 0.
b = − k 2 and hence (a , b ) = (3k − 1, − k 2 ). Finally
The condition p 2 − qr < 0 implies that the
3a + 2p = − p 2 does not have any solution.
Eq. (iii) has only non-real roots.
35. Consider the discriminants of the three Aliter One can argue as follows. Suppose α is a
equations common root of two equations, say, Eqs. (i)
and (ii). If α is non-real, then α is also a root of
px 2 + qx + r = 0 …(i)
both Eqs. (i) and (ii). Hence, the coefficients of
qx + rx + p = 0
2
…(ii) Eqs. (i) and (ii) are proportional. This forces
p = q = r , a contradiction. Hence, the common
rx + px + q = 0
2
…(iii) root between any two equations cannot be
Let us denote them by D1 , D2 , D3 respectively. non-real. Looking at the coefficients, we
Then, we have conclude that the common root α must be
negative. If Eqs. (i) and (ii) have common root
D1 = 4(q 2 − rp )
α, then q 2 ≥ rp and r 2 ≥ pq. Here, atleast one
D2 = 4(r 2 − pq ) inequality is strict for q 2 = pr and r 2 = pq
D3 = 4(p 2 − qr ) forces p = q = r . Hence, q 2r 2 > p 2qr . This gives
p 2 < qr and hence Eq. (iii) has non-real roots.
We observe that
D1 + D2 + D3 = 4( p 2 + q 2 + r 2 − pq − qr − rp ) 36. If α and β are both integers, then
= 2{( p − q ) + (q − r ) + (r − p ) } > 0
2 2 2 [m α ] + [m β ] = m α + m β = m (α + β ) = m 2
This proves one implication.
since p , q , r are not all equal. Hence, atleast
one of D1 , D2 , D3 must be positive. We may Observe that α + β = m and αβ = n. We use the
assume D1 > 0. property of integer function
Suppose D2 < 0 and D3 < 0. In this case both the x − 1 < [x ] ≤ x for any real number x . Thus,
Eqs. (ii) and (iii) have only non-real roots and m 2 − 2 = m (α + β ) − 2
Eq. (i) has only real roots. Hence, the common = m α − 1 + m β − 1 < [m α ] + [m β ] ≤ m (α + β )
root α must be between Eqs. (ii) and (iii). But = m2
www.pdfworld.in
www.pdfworld.in
Since, m and n are positive integers, both y 2 − 2y = 0 giving y = 2. But then y 4 + 4 = 20 is

α and β must be positive. If m ≥ 2, we observe not the power of a prime. The equations
that there is no square between y 2 − 2y + 2 = 4 and y 2 − 2y + 2 = 8 have no
m 2 − 2 and m 2. Hence, either m = 1 or
integer solutions. We conclude that
[m α ] + [m β ] = m 2. If m = 1, then α + β = 1 (p , x , y ) = (5, 1, 1) is the only solution.
implies that both α and β are positive reals Aliter Using y 2 − 2y + 2 = pm and
smaller than 1. Hence, n = αβ cannot be a
y + 2y + 2 = p , we may get
2 n
positive integer. We conclude that
−m
[m α ] + [m β ] = m 2. Putting m = α + β in this 4y = pm (pn − 1)
relation, we get If m > 0, then p divides 4 or y. If p divides 4,
[α 2 + n ] + [β 2 + n ] = (α + β )2 then p = 2. If p divides y , then y 2 − 2y + 2 = pm
Using [x + k ] = [x ] + k for any real number x shows that p divides 2 and hence p = 2. But
and integer k, this reduces to then 2x = y 4 + 4, which shows that y is even.
[α 2 ] + [β 2 ] = α 2 + β 2 Taking y = 2z , we get 2x − 2 = 4x 4 + 1. This

implies that z = 0 and hence y = 0, which is a
This shows that α 2 and β 2 are both integers.
contradiction. Thus, m = 0 and y 2 − 2y + 2 = 1.
On the other hand,
This gives y = 1 and hence p = 5, x = 1.
α 2 − β 2 = (α + β )(α − β ) = m (α − β )
38. Let P (x ) = a 0 + a1x + a 2x 2 + . . . + an x n be a
α −β
2 2
polynomial with integer coefficients.
Thus, (α − β ) = ,
m
Case I We may write
is a rational number. Since, α + β = m is a P (x ) = a 0 + a 2x 2 + a 4x 4 + . . .
rational number, it follows that both α and β
are rational numbers. However, both α 2 and β 2 + x (a1 + a3x 2 + a 0x 5 + . . . ).
are integers. Hence, each of α and β is an Define
integer. Q (x ) = a 0 + a 2x 2 + a 4x 4 + . . .
37. We begin with the standard factorisation − x (a1 + a3x 2 + a5x 5 + . . . ).
y 4
+ 4 = ( y − 2y + 2)( y + 2y + 2)
2 2
Then, Q (x ) is also a polynomial with integer
Thus, we have y 2 − 2y + 2 = pm and coefficients and
y 2 + 2y + 2 = pn for some positive integers P (x ) Q (x ) = (a 0 + a 2x 2 + a 4x 4 + . . . )2
m and n such that m + n = x. Since, − x 2 (a1 + a3x 2 + a5x 5 + . . . )2
y 2 − 2y + 2 < y 2 + 2y + 2, we have m < n so that
is a polynomial in x 2.
pm divides pn . Thus, y 2 − 2y + 2 divides
Case II We write again
y 2 + 2y + 2. Writing
P (x ) = A (x ) + xB (x ) + x 2C (x ),
y 2 + 2y + 2 = y 2 − 2y + 2 + 4y ,
where
we infer that y 2 − 2y + 2 divides 4y and hence A (x ) = a 0 + a3x 3 + a 6x 6 + . . . ,
y − 2y + 2 divides 4y . But
2 2
B (x ) = a1 + a 4x 3 + a7x 6 + . . . ,
4y = 4( y − 2y + 2) + 8 ( y − 1)
2 2
C (x ) = a 2 + a5x 3 + a 8x 6 + . . . ,
Thus, y − 2y + 2 divides
2
8 ( y − 1). Since, Note that A (x ), B (x ) and C (x ) are polynomials
y − 2y + 2 divides both 4y and 8 (y − 1), we
2 with integer coefficients and each of these is a
conclude that it also divides 8. This gives polynomial in x 3. We may introduce
y 2 − 2y + 2 = 1, 2, 4 or 8. S (x ) = A (x ) + ωxB (x ) + ω2x 2C (x ),
If y 2 − 2y + 2 = 1, then y = 1 and y 4 + 4 = 5, T (x ) = A (x ) + ω2xB (x ) + ωx 2C (x )
giving p = 5 and x = 1. If y − 2y + 2 = 2, then
2
where ω is an imaginary cube root of unity.
www.pdfworld.in
www.pdfworld.in
Then, However, for any real numbers x , y , we have

S (x )T (x ) = (A (x ))2 + x 2 (B (x ))2 + x 4 (C (x ))2 x 2 − xy + y 2 ≥ | xy |
− xA (x )B (x ) − x 3B (x ) C (x ) − x 2C (x )A (x ) Since, x 2y 2z 2 = | xy | | yz | | zx |, we get
Since, ω = 1 and ω + ω = − 1
3 2
| xy | | yz | | zx | = (x 2 − xy + y 2 )
Taking R (x ) = S (x ) T (x ), we obtain ( y 2 − yz + z 2 )(z 2 − zx + x 2 )
P (x )R (x ) = (A (x )) + x (B (x )) + x (C (x ))
3 3 3 6 3
≥ | xy | | yz | | zx |
− 3x 3A (x ) B (x ) C (x ), This is possible only if
which is a polynomial in x 3. This follows from x 2 − xy + y 2 = | xy | , y 2 − yz + z 2 = | yz | ,
the identity z 2 − zx + x 2 = | zx | ,
(a + b + c )(a 2 + b 2 + c 2 − ab − bc − ca ) hold simultaneously. However | xy | = ± xy . If
= a3 + b3 + c 3 − 3abc x 2 − xy + y 2 = − xy , then x 2 + y 2 = 0 giving
Alternately, R (x ) may be directly defined by x = y = 0. Since, we are looking for non-zero
R (x ) = (A (x ))2 + x 2 (B (x ))2 + x 4 (C (x ))2 x , y , z, we conclude that x 2 − xy + y 2 = xy
− xA (x )B (x ) − x 3B (x ) C (x ) − x 2C (x )A (x ). which is same as x = y . Using the other two
relations, we also get y = z and z = x . The first
39. Adding 1 both sides, the equation reduces to
equation now gives 27x 6 = x 3. This gives
[(x + 1)2 ] = ([x + 1])2 x 3 = 1/27 (since x ≠ 0) or x = 1/3. We thus have,
We have, used [x ] + m = [x + m ] for every x = y = z = 1/3. These also satisfy the second
integer m. Suppose x + 1 ≤ 0. Then, relation, as may be verified.
[x + 1] ≤ x + 1 ≤ 0. Thus
41. Let r = u/v where GCD (u , v ) = 1. Then, we get
([x + 1])2 ≥ (x + 1)2 ≥ [(x + 1)2 ] = ([x + 1])2
an un + an − 1un − 1v + . .. + a1uv n − 1 + a 0v n = 0,
Thus equality holds everywhere. This gives n −1 −1
bn un + bn − 1u v + . . . + b1uv n + b0v n = 0
[x + 1] = x + 1 and thus x + 1 is an integer.
Using x + 1 ≤ 0, we conclude that Subtraction gives
n −2 2
x ∈ {−1, − 2, − 3, . . .}. (an − bn ) un + (an −2 − bn− 2 )u v
n −1
Suppose x + 1 > 0. We have, + . . . + (a1 − b1 ) uv + (a 0 − b0 ) v n =0
(x + 1)2 ≥ [(x + 1)2 ] = ([x + 1])2 Since an − 1 = bn − 1. This shows that v divides
Moreover, we also have (an − bn ) un and hence it divides an − bn . Since
(x + 1)2 ≤ 1 + [(x + 1)2 ] = 1 + ([x + 1])2 an − bn is a prime, either v = 1 or v = an − bn .
Thus, we obtain Suppose the latter holds. The relation takes
[x ] + 1 = [x + 1] ≤ (x + 1) the form
n −2
< 1 + ([x + 1])2 = 1 + ([x ] + 1)2 un + (an −2 − bn − 2 )u v + ... +
n −2 −1
This shows that (a1 − b1 )uv + (a 0 − b0 )v n = 0.
x ∈ [n , 1 + (n + 1)2 − 1], (Here we have divided throughout by v.) If
n > 1, this forces v | u, which is impossible
where n ≥ − 1 is an integer. Thus, the solution
since GCD (v , u ) = 1(v > 1 since it is equal to the
set is
prime (an − bn ). If n = 1, then we get two
{−1, − 2, − 3, . . .} ∪ {∪n∞ = −1 [n , 1 + (n + 1)2 − 1]}
equations
It is easy verify that all the real numbers in this a1u + a 0v = 0, b1u + b0v = 0.
set indeed satisfy the given equation. This forces a1b0 − a 0b1 = 0 contradicting
40. Since, xyz ≠ 0, we can divide the second an b0 − a 0bn ≠ 0.
relation by the first. Observe that Note The condition an b0 − a0bn ≠ 0 is extraneous.
x 4 + x 2y 2 + y 4 = (x 2 + xy + y 2 )(x 2 − xy + y 2 ), The condition an − 1 = bn − 1 forces that for
holds for any x , y . Thus, we get n = 1, we have a0 = b0 . Thus, we obtain, after
(x 2 − xy + y 2 )( y 2 − yz + z 2 )(z 2 − zx + x 2 ) subtraction (a1 − b1)u = 0. This implies that
u = 0 and hence r = 0 is an integer).
= x 2y 2z 2
www.pdfworld.in
www.pdfworld.in
Unit 3
Inequalities
Inequality
The relation between two unequal numbers (real numbers) is called an inequality.
A quantity x is said to be greater than the quantity y, if x − y is +ve.
The quantity x is said to be less than quantity y, if x − y is –ve.
The symbols > and < are used for greater than and less than respectively.
Some Important Properties

Property 1. If a > b , b > c , then a > c
We have a − c = (a − b ) + (b − c ) = + ve
Since, a > b and b > c i.e., (a − b ) and (b − c ) are +ve.
∴ a>c
In same way, we can prove that if, we are given a < b, b < c , then a < c .
Property 2. If a > b, then a + c > b + c
We have (a + c ) − (b + c ) = a − b = + ve [Q a > b]
∴ (a + c ) > (b + c )
Similarly, we can prove that
a b
a − c > b − c, > and a × c > b × c
c c
provided c is +ve.
Similarly, we can also prove that, if a < b, then
a + c < b + c and a − c < b − c
Hence, if both sides of an inequality are increased or diminished by the same number (+ve, –ve or zero),
then the sign of inequality remains unaltered i.e., remains the same.
Property 3. In an equality any term may be transposed from one side to the other provided its sign is
changed.
If a + b > c + d , then a + b − c > d
or −c − d > − a − b etc.
Property 4. If the sides of an inequality be changed the sign of inequality is reversed, if a > b, then
b < a.
Property 5. If both sides of an inequality are multiplied or divided by same +ve number, the sign of
inequality remains unaltered.
If a > b, then a − b is +ve i.e., a − b > 0 multiplying both sides by x where x > 0 we have x (a − b ) > 0 or
xa − xb > 0. Hence proved.
www.pdfworld.in
www.pdfworld.in
1
Again dividing both sides of a − b > 0 by x, where x > 0 i.e., multiplying both sides of a − b > 0 by , where
x
x > 0, we get
1 a b
(a − b ) > 0 or − >0
x x x
a b
or >
x x
Hence proved.
a b
Similarly, we can prove that if a 0, then xa < xb, < .
x x
Property 6. If both sides of an inequality be multiplied by a –ve quantity, the sign of inequality is
reversed.
If a > b, then a − b is +ve multiplying both sides by –x, we get
−x (a − b ) = –ve or −xa + xb = –ve
or −xa + xb < 0 or −xa < − xb
i.e., −x (a ) < − x (b )
Hence proved.
Similarly, we can prove that if a 0,
then −x (a ) > − x (b )
Property 7. If a and b are two +ve numbers and a > b, then 1 < 1 .
a b
Q a > 0 and b > 0 , so ab > 0
Also, it is given that a > b
Dividing both sides of this inequality by ab, where ab > 0. So, by property 5
a b 1 1
> or >
ab ab b a
1 1
or <
a b
Property 8. If a1 > b1 ,a2 > b2,…, an > bn then a1 + a 2 + K + an > b1 + b2 + … + bn for all +ve a's and +ve
b's.
If a1 > b1, then a1 − b1 > 0
Similarly, a 2 − b2 > 0, a3 − b3 > 0, …,
an − bn > 0
Adding these, we get
(a1 − b1 ) + (a 2 − b2 ) + K + (an − bn ) > 0
or (a1 + a 2 + K + an ) − (b1 + b2 + K + bn ) > 0
or a1 + a 2 + K + an > b1 + b2 + K + bn
Property 9. If a1 > b1 , a2 > b2 , K , an > bn , then a1a 2a3 K an > b1b2b3Kbn for all +ve numbers a's and b's.
From property 5, we know that if a1 > b1, then a1x > b1x where x > 0.
Substituting a 2a3 K an for x, we get
a1a 2a3 K an > b1a 2a3 K an …(i)
Again as a 2 > b2 so a 2y > b2y where y > 0
Substituting b1a3 K an for y, we get
a 2b1a3 K an > b2b1a3 K an
b1a 2a3 K an > b1b2 a3 K an …(ii)
www.pdfworld.in
www.pdfworld.in
Inequalities 219
From (i) and (ii)

a1a 2a3 K an > b1b2a3 K an
Applying this method successively, we get
a1a 2a3 K an > b1b2b3 K bn
Hence proved.
Property 10. If a > b and n is a +ve integer, then an > bn and a1/n > b1/n provided a and b are both +ve
and only real +ve values of the nth roots are taken into account.
From property 9, if a1 > b1 , a 2 > b2,…, an > bn we have a1a 2a3 .. . an > b1b2b3 K bn
Put a1 = a 2 = a3 = K = a
and b1 = b2 = b3 = K = b,
we get a ⋅ a ⋅ a K n times > b ⋅ b ⋅ b. .. n times or an > bn
Similarly, we can prove second result.
Concept We know that given any 2 distinct real numbers a , b.
Either a b
But to decide whether it is the former inequality or the latter for any given pair of real number often
requires use of certain facts in clever way.
For example 25 , 33
if 0 ≤ m < n, we know that 2m < 2n or 3m < 3n
Again for m > 0, 2m < 3m (2 < 3)
But these inequalities as such do not help to compare 25 and 33.
On the other hand
33 = (2 + 1)3 = 23 + 3 × 22 + 3(2) + 1 < 23 + 4(2)2 + 4 × 2
= 23 + 24 + 23
= 2 × 23 + 24 = 2 ⋅ 24 = 25
of course 25 = 32 and 33 = 27 are easily computable and is evident that 25 > 33.
If however the indices m , n of 2m , 3n are very large, it is necessary to think of smaller indices from which
we could derive the inequality of larger indices.
e.g., We know that 25 > 33
We immediately deduce
25 × 121 = 2605 > 33 × 121 = 3363
Also, 27 > 34
Q 2 ⋅ 25 > 22 ⋅ 33 = 4 × 33 > 3 × 33 = 34
2
Hence, 2700 > 3400 (raising power to 100).

Like that another simple deduction in the routine
9 = 32 > 23 = 8
⇒ 3200 > 2300
Example 1 Which number is greater ( 31)12 or (17)17?

Solution Now, 31 < 32
( 31)12 < ( 32)12
⇒ ( 31)12
< ( 25 )12 = 260 …(i)
www.pdfworld.in
www.pdfworld.in
Now, 260 < 268 = 24 × 17 …(ii)

⇒ ( 24 )17 < (16)17
⇒ (16)17 < (17)17 …(iii)
From (i), (ii) and (iii), we get
( 31)12 < 260 < 268 < (17)17
⇒ ( 31)12 < (17)17
∴ (17)17 > ( 31)12
Example 2 Which number is greater ( 30)100 or 2567?

Solution ( 30)100 < ( 32)100
So, ( 25 )100 = ( 210 )50 = (1024)50
Now, (1024)50 < (1024)54 = ((1024)2 )27 = ( 220 )27
Now, 220 < 221
(2 ) 20 27
< ( 221)27 = 2567
∴ ( 30)100 < 2567
∴ 2567 > ( 30)100
Example 3 Which is greater 792 or 891?

Solution If a, b are +ve and n is a natural number, then
(a + b )n = an + nan − 1 b + …
(term involving powers of a and b).
Also, (a + b )n ≥ an + nan − 1b (equality for n = 1)
Now, ( 8) 91
= (7 + 1) 91
>7 91
+ 91⋅ 7 90
= 7 (7 + 91) = 790( 98)

90
Now, 7 ( 98) > 7

90 90
⋅ 4 9 = 790 ⋅ 72 = 792
Hence, ( 8)91 > 792
Example 4 Which is greater

(150)300 or ( 20000)100 × (100)100?
Solution We know (150)3 = 150 × 150 × 150

= 30 × 30 × 30 × 125 = 27000 × 125
Now, 27000 × 125 > 20000 × 100
Hence, (150)3 > 20000 × 100
(150)300 > ( 20000)100 × (100)100
2+ 3
Example 5 Show that 2< < 2.
2+ 3
Solution Q 3 > 2 and 3 < 2
Q 3> 2
www.pdfworld.in
www.pdfworld.in
Inequalities 221
2+ 3>2 2 …(i)
2> 3
2+ 2> 3+ 2
⇒ 4> 3+2 …(ii)
⇒ 3+2<2 …(iii)
1 1
⇒ > …(iv)
3+2 2
Multiplying (i) and (iv), we get

2+ 3
> 2 …(v)
3+2
Again, 2< 3
2+ 3<2 3 …(vi)
1< 3
1+ 2 < 3 + 2
3< 3+ 2
3< 3+2
1 1
or > …(vii)
3 3+2
1 1
⇒ < …(viii)
3+2 3
Multiplying (vi) and (viii), we get

2+ 3 1
<2 3× =2 …(ix)
3+2 3
2+ 3
From (ix) 2< <2
3+2
Hence proved.
Example 6 Show that (1.01)1000 > 1000 .

Solution We can write 1.01 as (1 + 0.01)
Now, (1 + 0.01)8 > 1 + 0.08 [Q (1 + x )n > 1 + nx ]
⇒ (1.01)1000 > (1 + 0.08)125 = ((1 + 0.08)5 )25 > (1 + 0.4)25
⇒ (1 + 0.4)25 > (1 + 0.4)24 = ((1.4)3 )8 [Q (1 + 0.08)5 = (1 + 0.4)]
Now, ((1.4)3 )8 > ( 2.7)8 > 74 = 2401 > 1000
∴ (1.01)1000 > 1000
www.pdfworld.in
www.pdfworld.in
Author’s Observation
101 10001
Consider the numbers ,
1001 100001
Note that the pattern of the numbers in both the numerator and denominator of each fraction is same
i.e., 0 is flanked by 1. The only difference is that the denominator has one more zero than the
numerator. If x is the numerator of any one of the two fractions, then the corresponding denominator is
10x − 9.
For example if x = 101
10x − 9 = (101)10 − 9 = 1001
x
∴ Let be denoted by a
10x − 9
1 10x − 9
then =
a x
9
which is equal to 10 − .
x
9 −9 1
Thus, if x increases, decreases and so increases. Hence, increases where ‘x’ increases or ‘a’
x x a
decreases, when x increases.
101 10001
In other words >
1001 100001
K zero L zero
100............. 01 10 ................ 01
>
1000 .............. 01 100 ................. 01
K + 1 zero L + 1 zero
∴ In general
If L> K
Concept If we have to prove f (x ) > g (x ) .

We consider difference of f (x ) and g (x ) say f (x ) − g (x ) .
By proving the difference f (x ) − g (x ) > 0 by some method we get the required result.
Example 1 Prove that m 3 + 1 > m 2 + m, m ≠ 1 ⇒ m > −1.

Solution We consider difference of
(m 3 + 1) − (m 2 + m ) = (m 3 − m 2 ) − (m − 1)
= m 2(m − 1) − (m − 1) = (m − 1)(m 2 − 1)
= (m − 1)(m − 1)(m + 1) = (m − 1)2(m + 1)
⇒ The expression is positive.
Hence, m3 + 1 > m2 + m
Example 2 Prove that

a 3b + ab 3 < a 4 + b 4.
Solution Consider difference of
(a 4 + b 4 ) − (a 3b + ab 3 ) = a 3(a − b ) − b 3(a − b ) = (a − b ) (a 3 − b 3 )
= (a − b )(a − b )(a 2 + ab + b 2 ) = (a − b )2(a 2 + ab + b 2 )
www.pdfworld.in
www.pdfworld.in
Inequalities 223
Q (a − b )2 is +ve
and a 2 + ab + b 2 is +ve
∴ (a − b )2(a 2 + ab + b 2 ) is +ve.
∴ a 3b + ab 3 < a 4 + b 4
Hence proved.
Example 3 If x , y > 0, then prove x 5 + y 5 > x 4y + xy 4 unless x = y .

Solution Consider x 5 + y 5 − x 4y − xy 4
= ( x 5 − x 4y ) + ( y 5 − xy 4 )
= x 4( x − y ) + y 4( y − x ) = ( x 4 − y 4 ) ( x − y )
= ( x 2 + y 2 )( x 2 − y 2 )( x − y )
= ( x 2 + y 2 )( x − y )2( x + y ) = +ve, if x ≠ y
a+x a
Example 4 If a > b and x is +ve, then prove < .
b+x b
Solution Consider the difference
a + x a b(a + x ) − a(b + x ) bx − ax − x (a − b ) − x (a − b )
− = = = = is −ve.
b+x b b(b + x ) b(b + x ) b(b + x ) b (b + x )
a+x a
∴ <
b+x b
Hence proved.
1
Example 5 If x > 0, then prove x + ≥2.
x
1
Solution Consider difference of x + − 2
x
x 2 − 2x + 1 ( x − 1)2
⇒ ⇒
x x
Q x > 0 given ( x − 1)2 is > 0.
1 1
∴ x + − 2 ≥ 0 for x > 0 ⇒ x + ≥ 2
x x
1
So, x + = 2 for x = 1
x
≤ − 2 combining the above inequalities, we get

x + 1 
1 ≥ 2 ∀ x .
Note Similarly, if x < 0, then x +
x  x
Example 6 Show that a(a − b )(a − c ) + b(b − c )(b − a ) + c(c − a ) (c − b ) cannot be –ve.
Solution Whenever we have symmetrical expression in a, b, c then without loss of generality we
can assume
a ≥b ≥c
a(a − b )(a − c ) + b(b − c )(b − a ) + c(c − a ) (c − b )
= (a − b )[a(a − c ) − b(b − c )] + c(c − a )(c − b )
= (a − b )[(a 2 − b 2 ) + (bc − ac )] + c(c − a )(c − b )
www.pdfworld.in
www.pdfworld.in
= (a − b )[a 2 − b 2 − c(a − b )] + c(c − a )(c − b )

= (a − b )[(a − b )(a + b − c )] + c(c − a )(c − b )
= (a − b )2(a + b − c ) + c(c − a )(c − b )
= (a − b )2(a + b − c ) + c(a − c )(b − c ) = +ve
As a >b >c
∴ b − c is + ve
and a − c is +ve
If a = b = c, then it is zero.
∴ a(a − b )(a − c ) + b(b − c )(b − a ) + c(c − a )(c − b )
It cannot be –ve.
Example 7 For real a, b, c show that

a 2 + b 2 + c 2 ≥ ab + bc + ca.
Solution a + b + c ≥ ab + bc + ca
2 2 2
Multiply 2 on both sides, we get

2a 2 + 2b 2 + 2c 2 ≥ 2ab + 2bc + 2ca
⇒ 2a 2 + 2b 2 + 2c 2 − 2ab − 2bc − 2ca ≥ 0
⇒ (a 2 − 2ab + b 2 ) + (b 2 − bc + c 2 ) + (c 2 − ac + c 2 ) ≥ 0
⇒ (a − b )2 + (b − c )2 + (c − a )2 ≥ 0
Hence proved.
Note Egives us a new concept by use of which we can prove several inequalities i.e., without using
Arithmetic mean (AM) and Geometric mean (GM). Let us solve some examples by using above
concept.

a 4 + b 4 + c 4 ≥ a 2b 2 + b 2c 2 + c 2a 2.
Solution We know that
a 2 + b 2 + c 2 ≥ ab + bc + ca …(i)
Now, we can write a + b + c as
4 4 4
(a 2 )2 + (b 2 )2 + (c 2 )2
Let a2 = u …(ii)
b2 = v …(iii)
c2 = w …(iv)
Using (i), we get
u 2 + v 2 + w 2 ≥ uv + vw + uw
Substitute the values of u, v , w , we get
a 4 + b 4 + c 4 ≥ a 2b 2 + b 2c 2 + a 2c 2

a 2b 2 + b 2c 2 + c 2a 2 ≥ abc(a + b + c ).
Solution Let ab = u
www.pdfworld.in
www.pdfworld.in
Inequalities 225
bc = v
ca = w
Now, we have
u 2 + v 2 + w 2 ≥ uv + vw + uw
Substitute the value, we get
a 2b 2 + b 2c 2 + c 2a 2 ≥ (ab )(bc ) + (bc )(ca ) + (ab )(ca )
⇒ a 2b 2 + b 2c 2 + c 2a 2 ≥ abc(a + b + c )
Hence proved.
Example 10 Without using AM and GM, prove that

ab bc ac
+ + > a + b + c.
c a b
ab
Solution Let = u2
c
bc ac
= v 2 and =w2
a b
As we know that
u 2 + v 2 + w 2 ≥ uv + vw + uw
ab bc
Now, u 2v 2 = ×
c a
u 2v 2 = b 2
u 2v 2 = b [Qb > 0]
uv = b
bc ac
Similarly, v 2w 2 = ×
a b
v 2w 2 = c 2
vw = c [Q c > 0]
∴ Like wise uw = a [∴ a > 0]
∴ We get
ab bc ac
+ + > a + b+ c
c a b
Example 11 For real a, b, c > 0. Without using AM or GM, prove that

a 2b 2 b 2c 2 c 2a 2
+ + > ab + bc + ca .
c2 a2 b2
a 2b 2 b 2c 2 c 2a 2  ab   bc   bc   ca   ab   ca 
Solution + + >   +   +  
c2 a2 b2 c a   a b cb
a 2b 2 b 2c 2 c 2a 2
∴ + + > b2 + a 2 + c 2 …(i)
c2 a2 b2
Now, we know that
a 2 + b 2 + c 2 > ab + bc + ca …(ii)
www.pdfworld.in
www.pdfworld.in
∴ From (i) and (ii)

a 2b 2 b 2c 2 c 2a 2
+ + > ab + bc + ac
c2 a2 b2
Concept
Maximization Principle
(a) Whenever we write x ≥ a, this is analysed as ‘a’ is the minimum value of x.
(b) Whenever we write x ≤ a, this is analysed as ‘a’ is maximum value of x.
(c) If we have a ≤ x ≤ b, this is analysed as ‘a’ is minimum value of x and ‘b’ is maximum value of x.
(I) Now, suppose we have
S =k −x
where k is constant.
Now, ‘maximum value’ of S
= k − (minimum value of x)
i.e., S ≤k −a
∴ We read as k − a is maximum value of S.
Now, ‘minimum value’ of S = k − (maximum value of x)
i.e., S ≥ (k − b )
∴ We read as k − b is minimum values of S.
(II) Now, suppose we have
S =k + x
Maximum value of S = k + maximum (value of x)
=k + b
Minimum value of S = k + minimum (value of x)
=k + a
Note Student should keep the above concept for proving various inequalities.
1
Example 1 Let y = ,x ≠0.
1
x + +5
x
Find maximum and minimum value of y.
1
Solution Now, ymax =
 1 
min x + + 2
 x 
 1   1  1 
Now, min  x + + 2 = min  x +  + 2 = 2 + 5 = 7 Q x + ≥ 2 ∀ x > 0
 x   x  x 
1 1
∴ ymax = i.e., y ≤
7 7
1
Now, ymin =
 1 
max  x + + 5
 x 
www.pdfworld.in
www.pdfworld.in
Inequalities 227
 1   1
Now, max  x + + 5 ⇒ max  x +  + 5
 x   x
⇒ −2 + 5 = 3
1
∴ x + ≤ − 2, ∀ x < 0
x
1 1
ymin = i . e., y ≥
3 3
Example 2 If x + y = 1, x > 0, y > 0, then without using AM − GM prove

1 1
(i ) 0 < xy ≤ (ii ) x 2 + y 2 ≥
4 2
2 2
1  1  1 25
(iii ) x 4 + y 4 ≥ (iv)  x +  +  y +  ≥ .
8  x  y 2
Solution (i) Consider ( x − y )2 ≥ 0
⇒ x 2 + y 2 − 2xy ≥ 0
Add 4xy on both sides,
⇒ x 2 + y 2 + 2xy ≥ 4xy ⇒ ( x + y )2 ≥ 4xy
⇒ 1 ≥ 4xy [Qx + y = 1]
1
⇒ xy ≤ …(i)
4
Also, x > 0, y > 0 ⇒ xy > 0 …(ii)
From (i) and (ii)
1
0 < xy ≤ …(iii)
4
(ii) Let S = ( x 2 + y 2 )
S = ( x + y )2 − 2xy
S = 1 − 2xy
Q We have to find minimum of S.
∴ min(S ) = 1 − 2 max ( xy )
 1  1
⇒ min (S) = 1 − 2    from (iii) xy ≤ 4 
 4  
1 1
i.e., S ≥1 − =
2 2
1
⇒ x2 + y2 ≥ …(iv)
2
(iii) Let S = x 4 + y 4
S = ( x 2 + y 2 )2 − 2 x 2 y 2
S = S1 − S 2 = S1 + ( −S 2 )
Let S1 = ( x 2 + y 2 )2
1
We have, x2 + y2 ≥
2
www.pdfworld.in
www.pdfworld.in
1
Squaring, ( x 2 + y 2 )2 ≥ …(v)
4
1
∴ S1 ≥ …(vi)
4
Let S 2 = − 2x 2y 2
1
We have, xy ≤ [from (iii)]
4
1 1
x y ≤
2 2
− 2x y ≥ −
2 2
16 8
1
∴ S2 ≥ − …(vii)
8
Q a1 > b1 and a 2 > b2
⇒ a1 + a 2 > b1 + b2
1 1 1
⇒ S1 + S 2 ≥ − =
4 8 8
2 2
 1  1 1 1
(iv) Let S =  x +  +  y +  = x 2 + y 2 + 2 + 2 + 4
 x  y x y
(x 2 + y 2 )
= x2 + y2 + +4
x 2y 2
S = S1 + S 2 + S 3
Let S1 = x 2 + y 2
1
S1 ≥ [from (iv)] …(viii)
2
x2 + y2
Let S2 =
x 2y 2
1
x2 + y2 ≥ …(ix)
2
1
x 2y 2 ≤
16
1
2 2
≥ 16 …(x)
x y
Multiplying (ix) and (x)
x2 + y2
≥8 …(xi)
x 2y 2
⇒ S2 ≥ 8 …(xii)
Adding (viii) and (xii)
1
S1 + S 2 ≥ +8
2
On adding S 3 on both sides
1
⇒ S1 + S 2 + S 3 ≥ + 8 + S3
2
1
⇒ S1 + S 2 + S 3 ≥ + 8 + 4
2
www.pdfworld.in
www.pdfworld.in
Inequalities 229
25
⇒ S1 + S 2 + S 3 ≥ [Q S 3 = 4]
2
25
⇒ S≥
2
2 2
 1  1 25
So, x +  + y +  ≥
 x  y 2
Example 3 If a > 0 and n ∈ N, then prove

an 1
< .
1 + a + a + K + a 2n 2n
2
an
Solution Let S =
1 + a + a + K + a 2n
2
Dividing numerator and denominator by an

1
S=
1 1 1 1
n
+ n − 1 + n − 2 + K + + 1 + a + a 2 + ...+ an
a a a a
1
S=
 1  1  1
1 + a +  + a 2 + 2  + K + an + n 
 a  a   a 
QWe have to find maximum value of S.
∴ Maximum S
1
=
  1  2 1   1 
min 1 + a +  + a + 2  + ...+ an + n  
  a   a   a 
1
=
1 + ( 2 + 2 + K n times)
1
Q ak + k ≥ 2, ∀ k = 1, 2, 3, K , n
a
1
Max (S ) =
1 + 2n
1
∴ S≤ …(i)
1 + 2n
Now, 1 + 2n > 2n, ∀ n ∈ N
1 1
⇒ < …(ii)
1 + 2n 2n
From (i) and (ii), we get
1 1
S≤ <
1 + 2n 2n
1
⇒ S<
2n
an 1
⇒ <
1 + a + a + K + a 2n 2n
2
www.pdfworld.in
www.pdfworld.in
First Fundamental Special Concept

We know (a − b )2 ≥ 0
⇒ a 2 + b 2 − 2ab ≥ 0
⇒ a 2 + b 2 ≥ 2ab
⇒ 2a 2 + 2b 2 ≥ a 2 + b 2 + 2ab
⇒ 2(a 2 + b 2 ) ≥ (a + b )2
1
⇒ (a 2 + b 2 ) ≥ (a + b )2
2
a2 + b2 1
⇒ ≥ (a + b )
a+b 2
Second Special Fundamental Concept

(a 2 − b 2 )2 ≥ 0
a 4 + b 4 − 2a 2b 2 ≥ 0
a 4 + b 4 ≥ 2a 2b 2
2a 4 + 2b 4 ≥ a 4 + b 4 + 2a 2b 2
2(a 4 + b 4 ) ≥ (a 2 + b 2 )2
a4 + b4 1 2
≥ (a + b 2 )
a2 + b2 2
Third Special Fundamental Concept

x 2 + y2 + a2 + b2 ≥ (x + a )2 + (y + b )2 (Qboth sides are +ve)
QSquaring, we get
x 2 + y 2 + a 2 + b 2 + 2 x 2 + y 2 a 2 + b 2 ≥ (x + a )2 + (y + b )2
⇒ 2 x 2 + y2 a 2 + b 2 ≥ 2ax + 2by
⇒ x 2 + y 2 a 2 + b 2 ≥ ax + by …(i)
If ax + by < 0, then (i) is true and given inequality is true.
If ax + by ≥ 0, then squaring (i)
(x 2 + y 2 )(a 2 + b 2 ) ≥ (ax + by )2
which on arranging becomes
(ax − by )2 ≥ 0
which is true.
∴ Result is proved.
Fourth Special Fundamental Concept

We have, (a − b )2 ≥ 0, a , b are +ve
a 2 + b 2 − 2ab ≥ 0
a 2 + b 2 ≥ 2ab
www.pdfworld.in
www.pdfworld.in
Inequalities 231
Note a 2 + b 2 − ab > ab
Multiplying (a + b ) on both sides
(a + b )(a 2 − ab + b 2 ) > ab(a + b )
a 3 + b 3 > ab(a + b )
a 3 + b3
> ab
a+b
Fifth Special Fundamental Concept

We have proved earlier
a 2 + b 2 + c 2 ≥ ab + bc + ca
Now, we can write
(a + b + c )2 − 2ab − 2bc − 2ac > ab + bc + ca
or (a + b + c )2 > 3(ab + bc + ca )
2
a + b + c  1
or   > (ab + bc + ca )
 3  3
b2 + c 2 c 2 + a 2 a 2 + b2
Example 1 Without using AM and GM, prove + + ≥ a + b + c.
b+c c+a a+b
Solution From above fundamental concept
a 2 + b2 1
We have, ≥ (a + b ) …(i)
a+b 2
b2 + c 2 1
≥ (b + c ) …(ii)
b+c 2
c2 + a2 1
and ≥ (c + a ) …(iii)
c+a 2
Adding (i), (ii) and (iii), we get
a 2 + b2 b2 + c 2 c 2 + a 2
+ + ≥a + b + c
a+b b+c c+a
Example 2 Without using AM and GM, prove

a 4 + b4 b4 + c 4 c 4 + a 4
+ + ≥ ab + bc + ca.
a 2 + b2 b2 + c 2 c 2 + a 2
Solution From second fundamental concept
a 4 + b4 1 2
≥ (a + b 2 ) …(i)
a 2 + b2 2
b4 + c 4 1 2
≥ (b + c 2 ) …(ii)
b2 + c 2 2
c4 + a4 1 2
and ≥ (c + a 2 ) …(iii)
c2 + a2 2
www.pdfworld.in
www.pdfworld.in

a 4 + b4 b4 + c 4 c 4 + a 4
+ + ≥ a 2 + b2 + c 2 …(i)
a 2 + b2 b2 + c 2 c 2 + a 2
Also, a 2 + b 2 + c 2 ≥ ab + bc + ca …(ii)
a +b
4 4
b + c4 c4 + a4
4
∴ + + ≥ ab + bc + ca
a 2 + b2 b2 + c 2 c 2 + a 2
Example 3 If A + B + C = π , without using AM and GM, show that

A B C A B C
sin4 A + sin4 B + sin4 C ≥ 32 sin sin sin cos 2 cos 2 cos 2 .
2 2 2 2 2 2
Solution We have, a 4 + b 4 + c 4 ≥ abc(a + b + c )
∴ sin4 A + sin4 B + sin4 C ≥ sin A sin B sin C (sin A + sin B + sin C )

A A B
⇒ sin4 A + sin4 B + sin4 C ≥ sin cos sin
2 2 2
B C C A B C
cos sin cos  4 cos cos cos  [QA + B + C = π]
2 2 3 2 2 2
A B C
∴ sin A + sin B + sin C = 4 cos cos cos
2 2 2
A B C A B C
⇒ sin4 A + sin4 B + sin4 C ≥ 32 sin sin sin cos 4 cos 4 cos 4
2 2 2 2 2 2
Example 4 a, b, c are +ve, prove that

1
a 3 + b3 + c 3 ≥ [ab(a + b ) + bc(b + c ) + ca(c + a )]
2
Solution From fourth fundamental concept, we have
a 3 + b 3 > ab(a + b ) …(i)
b 3 + c 3 > bc(b + c ) …(ii)
c 3 + a 3 > ca(c + a ) …(iii)
Adding (i), (ii), (iii), we get
⇒ 2(a 3 + b 3 + c 3 ) > ab(a + b ) + bc(b + c ) + ca(c + a )
1
a 3 + b 3 + c 3 ≥ [ab(a + b ) + bc(b + c ) + ca(c + a )]
2
Example 5 Without using AM − GM , prove that

tan2 A tan2 B tan2 C > 3(tan A tan B + tan B tan C + tan A tan C ).
Solution From fifth fundamental concept
2
a + b + c 1
  > (ab + bc + ca )
 3  3
Let a = tan A
b = tan B
c = tan C
www.pdfworld.in
www.pdfworld.in
Inequalities 233
We have,
2
 tan A + tan B + tan C  1
  > (tan A tan B + tan B tan C + tan C tan A )
 3  3
If A+B+C=π
tan A + tan B + tan C = tan A tan B tan C
⇒ tan2 A tan2 B tan2 C > 3(tan A tan B + tan B tan C + tan A tan C )
Arithmetic Mean (AM)

AM of a set of n +ve real numbers a1 , a 2 , ... , an is defined to be the average value of the n numbers.
a1 + a 2 + K + an
i.e.,
n
Geometric Mean (GM)

GM of the same set of numbers is defined to be n a1 × a 2 × K × an .
a+b
If a , b > 0, then a, , b are such that the difference between consecutive terms, viz.,
2
a+b a+b b −a a+b
− a, b − are each i.e., the same. The three terms form an AP is thus, an AM
2 2 2 2
inserted between a and b to get an AP. Again a, ab , b are such that the quotient of consecutive terms
ab b b
, are each i.e., the same a , ab , b is said to form a GP and ab has been inserted (as a GM)
a ab a
between a and b.
a+b
ab ≤ …(i)
2
GM ≤ AM
Now, (i) is equivalent to 2 ab ≤ a + b
(Multiplying an inequality by a +ve real number does not alter the inequality).
i.e., 0 ≤ a + b − 2 ab
Subtracting the same real number from both sides of an inequality does not alter the inequality.
i.e., a + b − 2 ab = ( a − b )2 ≥ 0
a +b
Note The proof shows that ab = , if and only if a = b or a = b.
2
a+b
To sum up ab ≤ i.e., GM ≤ AM equality holds, if and only if both the quantities are equal.
2
The GM − AM
inequality is true in the general form
a + ... + an
n a ... a ≤ 1
1 n
n
If equality holds and only if a1 = a 2 = K = an
It is easy to deduce it for a set of 4 +ve numbers from the case of 2 numbers for
4 a1a 2a3a 4 = 4 a1a 2 ⋅ 4 a3a 4
= a1a 2 ⋅ a3a 4
1
≤ ( a1a 2 + a3a 4 )
2
www.pdfworld.in
www.pdfworld.in
1 1
(a1 + a 2 ) + (a3 + a 4 )
1
≤
2  2 2 
a + a 2 + a3 + a 4
= 1
4
This method can be extended to the case n = 2k for some k = 2, 3, 4,…
Now, we have to prove
a1 + a 2 + a3
3 a1a 2a3 ≤ …(ii)
3
for ai > 0, i = 1, 2, 3
Let a1 = x 13
a 2 = x 32
a3 = x 33
x i > 0, i = 1, 2, 3
Now, (ii) becomes
3x 1x 2x 3 ≤ x 13 + x 32 + x 33
or x 13 + x 32 + x 33 − 3x 1x 2x 3 ≥ 0 …(iii)
Rewrite LHS of (iii) as
(x 1 + x 2 + x 3 )(x 12 + x 22 + x 32 − x 1x 2 − x 2x 3 − x 3x 1 )
[ (x 1 − x 2 )2 + (x 2 − x 3 )2 + (x 3 − x 1 )2 ]
= (x 1 + x 2 + x 3 )
2
Q The sum of three squares is always non–negative.
The equality can occur if and only if
x 1 = x 2 = x 3 or a1 = a 2 = a3.
Geometrical Interpretation of the AM–GM Inequality

We have segments of length a , b on a straight line putting them side by side along PR and RQ
O is the mid-point of PQ.
Draw a perpendicular to PQ at R to cut the semicircle on PQ at A and A ′.
By the secant theorem
PR ⋅ RQ = a ⋅ b = A ′ R ⋅ AR = AR 2
or AR = a ⋅ b
AR represents the GM of a and b.

AM being PO = radius of circle
QIn any circle, half of a chord is less than the radius.
(a + b ) A
ab ≤
2
equality hold if and only if
a b
P Q
Cauchy's Inequality or Cauchy Schwarz Inequality O R
If a , b , x , y are real, then
| ax + by | ≤ a 2 + b 2 x 2 + y2 A'
www.pdfworld.in
www.pdfworld.in
Inequalities 235
Proof To prove
(ax + by )2 ≤ (a 2 + b 2 ) (x 2 + y 2 ) …(A)
i.e., a x + 2axby + b y ≤ a x + a y + b x + b y
2 2 2 2 2 2 2 2 2 2 2 2
⇒ 2axby ≤ a 2y 2 + b 2x 2
⇒ a 2y 2 − 2axby + b 2y 2 ≥ 0
⇒ (ay − bx )2 ≥ 0
which is true.
Similarly, if a1 , a 2 , K , an ; b1 , b2 , K , bn are all real .
Then | a1b1 + a 2b2 + K + an bn | ≤ a12 + a 22 + K + an2 . b12 + b22 + K + bn2
Note Let P ( x , y ) be a point in the plane having origin O.

Let ax + by = 0 be a given straight line through the origin.
| ax + by |
Then is the perpendicular distance PM of P (x , y )from the given y
a2 + b2 ax + by = 0 P (x, y)
straight line ax + by = 0, x 2 + y 2 is the distance of x , y from origin.
−a y
The slope of the line is and of the line joining origin (0, 0) to (x , y ) is . x' x
b x O
y  −a  a b
If these two lines are perpendicular and only then,   = − 1 or = M
x  b  x y
This is the only case in which equality occur.
y'
Harmonic Mean (HM)
The +ve number
−1
 1 1 1 
 + + K+ 
  a1 a 2 an  
 n 
 
 
is called the HM of a1, a 2, K an and is denoted by H.
Example 1 If a > 0, b > 0, a + b = 1, show that

1 1
8+ + 8+ ≥ 24.
a2 b2
Solution Q AM ≥ GM
a+b
⇒ ≥ ab
2
a + b ≥ 2 ab
1 ≥ 2 ab
4ab ≤ 1
1
ab ≤ …(i)
4
www.pdfworld.in
www.pdfworld.in
1 1
8+ 2
+ 8+ 2
Now, a b ≥ 8+ 1 8+
1
[Q AM ≥ GM]
2 a2 b2
1 1  1 1 1
⇒ 8+ + 8+ ≥ 2 64 + 8  2 + 2  + 2 2
a2 b 2 a b  a b
 b2 + a 2  1
≥ 2 64 + 8  2 2  + 2 2
 a b  a b
16 1
≥ 2 64 + +
ab a 2b 2
 
 Q a 2 + b 2 ≥ 2ab 
 
≥ 2 64 + 16 × 4 + 16  8(a + b ) ≥ 16ab 
2 2
 
 8 (a + b ) ≥ 16 
2 2
 a 2b 2 ab 
≥ 2 144
≥ 2 × 12 ≥ 24
Example 2 For any +ve a, b. Prove that

2 2
 1  1
a +  + b +  ≥ 8.
 a  b
Solution Using AM-GM inequality
2 2 2 2
 1  1  1  1
a +  + b +  ≥ 2 a +  b + 
 a  b  a  b
2 2
 1  1  1 a b
a +  + b +  ≥ 2 ab + + + 
 a  b  ab b a 
 1 
 1 
2
1
2 Q ab + ab ≥ 2
a +  + b +  ≥ 2( 2 + 2)  
 a  b  a b
+ ≥ 2
 b a 
2 2
 1  1
Hence, a +  + b +  ≥ 8
 a  b
Example 3 If a1, a 2, a 3, K , an are non –ve and a1a 2a 3 K an = 1, then show that
(1 + a1)(1 + a 2 )(1 + a 3 ) K (1 + an ) ≥ 2n .
Solution By AM-GM
1 + ai 
  ≥ ai , i = 1, 2, K , n
 2 
Multiplying the inequalities,
(1 + a1)(1 + a 2 ) K (1 + an ) ≥ 2n a1a 2 K an
(1 + a1)(1 + a 2 ) K (1 + an ) ≥ 2n [Q a1a 2 ... an = 1]
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 237
Example 4 If a, b, c > 0, then show that

a 2(b + c ) + b 2(c + a ) + c 2(a + b ) ≥ 6abc.
Solution By applying AM ≥ GM for 6 numbers
a 2b, a 2c, b 2c, b 2a, c 2a, c 2b
1
a 2b + a 2c + b 2c + b 2a + c 2a + c 2b
We get, ≥ [a 2b ⋅ a 2c ⋅ b 2c ⋅ b 2a ⋅ c 2a ⋅ c 2b ] 6
6
⇒ a 2(b + c ) + b 2(c + a ) + c 2(a + b ) ≥ 6abc
Example 5 If a, b, c are greater than zero. Prove that

b+c c+a a+b
+ + >6
a b c
or bc(b + c ) + ca(c + a ) + ab(a + b ) > 6abc.
Solution From example 4, we have
a 2(b + c ) + b 2(c + a ) + c 2(a + b ) ≥ 6abc
⇒ (a 2b + b 2a ) + (a 2c + c 2a ) + (b 2c + c 2b ) ≥ 6abc
⇒ ab(a + b ) + ac(a + c ) + bc(b + c ) ≥ 6abc
 a + b  a + c   b + c 
⇒   +  +  ≥6
 c   b   a 
Aliter
a b
+
b a > a ×b
2 b a
a b
⇒ + >2 …(i)
b a
b c
Similarly, + >2 …(ii)
c b
a c
and + >2 …(iii)
c a
 a b b c c a
 +  +  +  + +  >6
b a  c b  a c 
 a c   b c   a b
or  +  + +  + +  >6
 b b  a a   c c 
 a + c   b + c   a + b
⇒   +  +  >6
 b   a   c 
Hence proved.
Example 6 If ai > 0, ∀ i = 1, 2, K , n. Prove that

1 1 1
(a1 + a 2 + K + an )  + + K +  > n2
 a1 a 2 an 
Solution QAM > GM, we have
a1 + a 2 + K + an
> (a1a 2 K an )1/n
n
www.pdfworld.in
www.pdfworld.in
1 1 1
+ +K+ 1/n
a1 a 2 an  1 1 1
and > ⋅ K 
n  a1 a 2 an 
⇒ (a1 + a 2 + K + an ) > n(a1a 2 K an )1/n

1/n
1 1 1  1 
and  + + K +  > n 
 1
a a 2 an  1 2
a a K an
1 1 1  n 
⇒ (a1 + a 2 K + an )  + + K +  > n[a1a 2 K an ](1/n )  1/n 
 1
a a 2 an (a1a 2 K an ) 
1 1 1
⇒ (a1 + a 2 + K + an )  + + K +  > n2
 1
a a 2 an
Example 7 Show that

1 1 1 1 1 1
+ + > + +
a b c ab bc ac
Solution Q AM > GM
1  1 1 1 1
∴  +  > ⋅
2  a b a b
1 1  1 
+ > 2 
a b  ab 
1  1 1  1
Similarly,  +  >  
2 b c  bc 
1 1  1 
i . e., + > 2 
b c  bc 
1  1 1 1 1 1  1 
and  +  > i.e., + > 2  
2 c a ca c a  ca 
Adding these three results
 1 1  1 1  1 1 2 2 2
 +  + +  + + > + +
 a b  b c   c a  ab bc ca
 1 1 1  1 1 1 
or 2 + +  > 2 + + 
a b c  ab bc ca 
1 1 1 1 1 1
or + + > + +
a b c ab bc ca
Hence proved.
Example 8 If x , y , z are three +ve integers. Prove that

 1 1 1
(x + y + z )  + +  > 9 (RMO 1994)
x y z
Solution Q AM > GM
x +y +z
∴ > ( xyz )1/ 3
3
www.pdfworld.in
www.pdfworld.in
Inequalities 239
or x + y + z > 3( xyz )1/ 3 …(i)

1/ 3
 1 1 1  1 1 1
Similarly,  + +  > 3 ⋅ ⋅  …(ii)
x y z x y z
Multiplying (i) and (ii), we get
1/ 3
 1 1 1  1 
( x + y + z )  + +  > 9 ( xyz )1/ 3  
x y z  xyz 
 1 1 1
⇒ (x + y + z )  + +  > 9
x y z
Hence proved.

a1 a 2 a 3 an − 1 an
+ + +K+ + >n (RMO Punjab 1993)
a2 a3 a4 an a1
Solution Q AM > GM
1  a1 a 2 a 3 an − 1 an   a1 a 2 a 3 an − 1 an 
∴  + + +K+ +  > ⋅ ⋅ K ⋅ 
n  a2 a3 a4 an a1   a 2 a 3 a 4 an a1 
 a1 a 2 a 3 an − 1 an 
or  + + +K+ +  > n(1)n
 a2 a3 a4 an a1 
 a1 a 2 a 3 an − 1
or  + + +K+  >n
 a2 a3 a4 an 
Hence proved.
Example 10 If a, b, c > 0, show that

a b c 3
+ + ≥ (IMP)
b+c c+a a+b 2
a a+b+c
Solution = −1
b+c b+c
b a+b+c
= −1
c+a c+a
c a+b+c
and = −1
a+b a+b
Now,
 1 1 1 
LHS = (a + b + c )  + +  −3
 b + c c + a a + b
1 1 1 1 1 1
But + + ≥3 3 + +
b+c c+a a+b b+c c+a a+b
Also, 2 (a + b + c ) = (b + c ) + (c + a ) + (a + b ) ≥ 3 (b + c ) (c + a ) (a + b )
 1 1 1  3×3
So, (a + b + c )  + +  ≥
 b + c c + a a + b 2
a b c 9 3
+ + ≥ −3=
a+c c+a a+b 2 2
www.pdfworld.in
www.pdfworld.in
Example 11 Find the minimum value of a 2 + b 2 , when 3a + 4b = 15.

Solution |3a + 4b| ≤ a 2 + b 2 32 + 42
= 5 a 2 + b2 [By Cauchy's inequality]
So, under given condition

15
a 2 + b2 ≥ =3
5
Weighted AM–GM Inequality
If a1 , a 2 , K , an are n +ve real numbers and m1 , m 2 , K , mn are n +ve rational numbers, then
1
m1a1 + m 2a 2 + K + mn an m m m + m 2 + K + mn
> (a1 1 ⋅ a 2 2 K anm n ) 1
m1 + m 2 + K + mn
Some Important Inequalities

(A) If a1 , a 2 , K , an are n +ve distinct real numbers, then
m
a1m + am
2 + K + an >  a1 + a 2 + K + an 
m
(i)  
n  n 
if m < 0 or m > 1
m
am + am
2 + K + an <  a1 + a 2 + K + an 
m
(ii) 1  
n  n 
if 0< m < 1
i.e., the AM of mth powers of n +ve quantities is greater than the mth power of their AM except
when
m is a +ve proper fraction.
(iii) If a1 , a2 , K , an and b1 , b2 , K , bn are rational numbers and m is a rational number, then
m
b1a1m + b2am2 + K + bn an >  b1a1 + b2a 2 + K + bn an 
m
 
b1 + b2 + K + bn  b1 + b2 + K + bn 
if m < 0 or m > 1
m
b1a1m + b2am 
2 + K + bn an < b1a1 + b2a 2 + K + bn an
m

and  
b1 + b2 + K + bn  b1 + b2 + K + bn 
if 0< m < 1
(B) If a1 , a 2 , a3 , K , an are distinct +ve real numbers and p , q , r are natural numbers, then
a1p + q +r
+ ap2 + q +r
+ K + anp + q +r
a1p + ap2 + K + anp a1q + aq2 + K + anq a1r + ar2 + K + anr
> ⋅ ⋅
n n n n
(C) Weierstrass Inequality

(i) If a1 , a2 , K , an are n +ve real numbers, then for n ≥ 2
(1 + a1 )(1 + a 2 ) ... (1 + an ) > 1 + a1 + a 2 + K + an
(ii) If a1 , a2 , K , an are +ve real numbers less than unity, then
(1 − a1 )(1 − a 2 ) K (1 − an ) > 1 − a1 − a 2 − K − an
www.pdfworld.in
www.pdfworld.in
Inequalities 241
(D) Tchebychef's Inequality

If a1 , a 2 , K , an and b1 , b2 , K , bn are real numbers such that
a1 ≤ a 2 ≤ a3 ≤ K ≤ an
and b1 ≤ b2 ≤ b3 ≤ K ≤ bn
Then n (a1b1 + a 2b2 + K + an bn ) ≥ (a1 + a 2 + K + an )(b1 + b2 + K + bn )
a1b1 + a 2b2 + K + an bn  a1 + a 2 + K + an   b1 + b2 + K + bn 
or ≥  
n  n  n 
(E) Holder's Inequality

(a1b1 + a 2b2 + K an bn )pq ≤ (a1p + ap2 + K + anp ) q (b1q + b2q + K + bn q ) p
1 1
where + = 1, ai and bi are non-negative real numbers.
p q
Example 1 If a, b are +ve real numbers. Prove that

a+b a+b
 a 2 + b2   a + b
(i )   > aabb (ii ) abaa <   < aabb
 a+b   2 
Solution (i) Q Weighted AM > Weighted GM

1
 a ⋅ a + b ⋅ b
∴   > (aa ⋅ bb )a + b
 a +b 
a+b
 a 2 + b2 
⇒   > aa bb
 a+b 
(ii) QWeighted AM > Weighted GM
1
b ⋅a + a ⋅b
> (ab ⋅ ba )a + b
b+a
1
2ab
⇒ > (ab ⋅ ba )a + b
a+b
1
a+b 2ab  a+b 2ab 
⇒ > > (ab ⋅ ba )a + b  QAM > HM, ∴ > 
2 a+b  2 a + b
1
a+b
⇒ > (ab ⋅ ba )a + b
2
a+b
 a + b
⇒   > abba …(i)
 2 
Again, Qweighted AM > weighted GM
1 1 1
a ⋅ + b⋅   1  a  1  b a + b
⇒ a b >   ⋅   
a+b  a   b  
www.pdfworld.in
www.pdfworld.in
1
2  1 a+b
⇒ > 
a + b  aabb 
a+b
 2  1
⇒   >
 a + b aabb
a+b
 a + b
⇒ aabb >   …(ii)
 2 
From (i) and (ii), we have
a+b
 a + b
ab ⋅ ba <   < aa ⋅ bb
 2 

y2 + z2 z2 + x2 x2 + y2
+ + > x + y + z.
y +z z +x x +y
Solution AM of 2nd power > 2nd power of AM

2
y2 + z2 y + z
> 
2  2 
2
y + z
⇒ y2 + z2 >2  
 2 
y2 + z2 y + z
⇒ > …(i)
y +z 2
x2 + z2 z + x
Similarly, > …(ii)
x +z 2
x2 + y2 x + y
and > …(iii)
x +y 2
y2 + z2 z2 + x2 x2 + y2
+ + >x +y +z
y +z z +x x +y
Example 3 If a, b > 0 such that a 3 + b 3 = 8, then show that a + b ≤ 2.

 1  1
Solution AM of   th power ≤   th power of AM
 3  3
1/ 3
(a 3 )1/ 3 + (b 3 )1/ 3  a 3 + b3 
∴ ≤  
2  2 
a+b
⇒ ≤1
2
⇒ a+b≤2
www.pdfworld.in
www.pdfworld.in
Inequalities 243
Example 4 If m > 1, n ∈ N show that

1m + 2m + 22m + 23m + K+ 2nm − m > n1 − m ( 2n − 1)m .
Solution Q m>0
AM of mth power > mth power of AM
m
1m + 2m + 4m + 8m + K + ( 2n − 1)m 1 + 2 + 4 + K + 2n − 1)
> 

n  n 
m
 2n − 1
⇒ 1m + 2m + 4m + K+ 2(n − 1)m > n   > n1 − m ( 2n − 1)m
 n 
Hence proved.
Example 5 If m > 1, then show that 2m + 4m + 6m + K + ( 2n )m > n(n + 1)m

Solution AM of mth power > mth power of AM
m
2m + 4m + 6m + K + ( 2n )m ( 2 + 4 + 6 + K + 2n ) 
> 
n  n
m
n(n + 1) 
⇒ 2m + 4m + 6m + ...+ ( 2n )m > n 
 n 
⇒ 2m + 4m + 6m + K + ( 2n )m > n(n + 1)m
Hence proved.
Example 6 If n is a +ve integer > 1, show that

n +1
1+ 2+ 3 +K+ n <n
2
 1  1
Solution AM of   th power <   th power of AM
 2  2
Since, 0 < m <1
1/ 2
1+ 2+ 3 + K + n 1 + 2 + 3 K + n 
∴ < 
n  n 
 n + 1
1+ 2+ 3 + K + n < n  
 2 
24n
Example 7 Show that C04 + C14 + C24 + K + Cn4 > ,
n3
n!
where n
Cr = .
r !(n − r )!
Solution AM of 4th power > 4th power of AM
4
C04 + C14 + K + Cn4  C0 + C1 + C2 + K + Cn 
> 
n  n 
4
 2n 
⇒ C04 + C14 + C24 + K + Cn4 > n  
n
24n
⇒ C04 + C14 + C24 +K + Cn4 >
n3
www.pdfworld.in
www.pdfworld.in
Example 8 If a1, a 2, K , an are +ve real numbers less than unity and Sn = a1 + a 2 + K + an , then
1
(i ) 1 − Sn < (1 − a1)(1 − a 2 ) .. (1 − an ) <
1 + Sn
1
(ii ) 1 + Sn < (1 + a1)(1 + a 2 ) K (1 + an ) <
1 − Sn
provided Sn < 1.
Solution We have
0 < ai < 1 ∀ i = 1, 2, K , n
⇒ 0 < 1 − ai2 < 1 ∀ i = 1, 2, K , n
⇒ 0 < (1 − ai )(1 + ai ) < 1 ∀ i = 1, 2, K , n
1 1
⇒ 0 < 1 − ai < and 0 < 1 + ai < ∀ i = 1, 2, K , n
1 + ai 1 − ai
1
⇒ 0 < (1 + a1)(1 + a 2 ) K (1 + an ) < …(i)
(1 − a1)(1 + a 2 ) K (1 + an )
By Weierstrass inequality
(1 + a1)(1 + a 2 ) K (1 + an ) > 1 + (a1 + a 2 + K + an )
and (1 − a1)(1 − a 2 ) K (1 − an ) > 1 − (a1 + a 2 + K + an )
⇒ (1 + a1)(1 + a 2 ) K (1 + an ) > 1 + Sn …(ii)
and (1 − a1)(1 − a 2 ) K (1 − an ) > 1 − Sn
1 1
⇒ < …(iii)
(1 + a1)(1 + a 2 ) K (1 + an ) 1 + Sn
1 1
and <
(1 − a1)(1 − a 2 ) K (1 − an ) 1 − Sn
1
1 − Sn < (1 − a1)(1 − a 2 ) K (1 − an ) <
1 + Sn
1
and 1 + Sn < (1 + a1)(1 + a 2 ) K (1 + an ) <
1 − Sn
provided Sn < 1
Example 9 If a1, a 2, K , an be +ve real numbers. Prove that

(a1 + a 2 + K + an )3 ≤ n 2(a13 + a 23 + K + an3 )
Also, show that equality sign holds iff
a1 = a 2 = K = an .
Solution Let two sets of numbers are
a13/ 2, a 23/ 2, K , an3/ 2 and a11/ 2, a12/ 2, K , an1/ 2
By Cauchy schwartz inequality, we have
(a13/ 2a11/ 2 + a 23/ 2a12/ 2 + K + an3/ 2an1/ 2 )2 ≤ (a13 + a 23 + K + an3 ) × (a1 + a 2 + ... + an )
or (a12 + a 22 + K + an2 )2 ≤ (a13 + a 23 + K + an3 ) (a1 + a 2 + K + an ) …(i)
Again taking two sets of numbers a1, a 2, K , an and 1, 1, K , 1.
Applying Cauchy Schwartz inequality
www.pdfworld.in
www.pdfworld.in
Inequalities 245
(a1 ⋅1 + a 2 ⋅1 + K + an ⋅1)2 ≤ (a12 + a 22 + K + an2 ) (12 + 12 + 12 + K + 12 )

or (a1 + a 2 + K + an )2 ≤ n(a12 + a 22 + K + an2 )2 …(ii)
(a1 + a 2 + K + an )4 ≤ n 2(a13 + a 23 + K + an3 ) (a1 + a 2 + K + an )
or (a1 + a 2 + K + an )3 ≤ n 2(a13 + a 23 + K + an3 )
Equality sign hold if and only if
a13/ 2 a 23/ 2 a 3/ 2
1/ 2
= 1/ 2 = K = n1/ 2
a1 a2 an
i.e., if and only if a1 = a 2 = K = an

1 2 3 n n + 1 1 1 1
+ + +K+ ≥  + +K+ 
n n −1 n − 2 1 2 n n −1 1
where n is any +ve integer.
Solution Let n is a +ve integer such that 1 < 2 < 3 < … < (n − 2) < (n − 1) < n
1 1 1 1 1 1
> > >K > > >
1 2 3 n − 2 n −1 n
1 1 1 1 1 1
or < < <K < < <
n n −1 n − 2 3 2 1
For two sets of numbers 1, 2, 3,…, n
1 1 1 1 1
and , , ,K ,
n n −1 n − 2 2 1
By Tchebychef's inequality, we have
 1 1 1 1
n 1⋅ + 2 ⋅ + 3⋅ +K+ n ⋅ 
 n n −1 n−2 1
1 1 1 1
≥ (1 + 2 + 3 + K + n ) ×  + + +K+ 
n n −1 n − 2 1
1 2 3 n n (n + 1)  1 1 1
⇒ n + + +K+  ≥  + +K+ 
 n n − 1 n − 2 1  2  n n − 1 1
1 2 3 n n + 1  1 1 1
∴  + + +K+  ≥  + +K+ 
n n −1 n − 2 1 2 n n −1 1
Example 11 If a, b, c, d are +ve real numbers. Prove that

(a 5 + b 5 + c 5 + d 5 ) ≥ abcd (a + b + c + d ).
Solution Assume a < b < c < d
Then a 4 < b4 < c 4 < d 4
For two sets i.e., a, b, c, d
and a 4, b 4, c 4, d 4
www.pdfworld.in
www.pdfworld.in
Apply Tchebychef’s inequality, we have

(a + b + c + d )(a 4 + b 4 + c 4 + d 4 ) ≤ 4(a ⋅ a 4 + b ⋅ b 4 + c ⋅ c 4 + d ⋅ d 4 )
(a 4 + b 4 + c 4 + d 4 )
or (a 5 + b 5 + c 5 + d 5 ) ≥ (a + b + c + d ) …(i)
4
a 4 + b4 + c 4 + d 4
∴ ≥ (a 4b 4c 4d 4 )1/ 4
4
a 4 + b4 + c 4 + d 4
⇒ ≥ abcd
4
(a 4 + b 4 + c 4 + d 4 )
or (a + b + c + d ) ≥ abcd (a + b + c + d )
4
Thus, (a 5 + b 5 + c 5 + d 5 ) ≥ abcd (a + b + c + d )
Jensen's Inequality
Suppose f (x ) is a twice differentiable function on an interval [a , b ] and f ′ ′ (x ) < 0 ∀ a < x < b.
Then for every +ve integer m and for all points.
x 1 , x 2 , x 3 , K , x m in [a , b ] , we have
x + x 2 + x 3 + K + x m  f (x 1 ) + f (x 2 ) + .. + f (x m )
f  1 ≥
 m  m
Moreover equality holds if and only if x 1 = x 2 = x 3 = K = x n
Note In this case, graph of f ( x ) is concave down.
Corollary Let f ′ (x ) > 0 and f ′ ′ (x ) < 0

and let x 1 , x 2 ∈ [a , b ] , then
x + x2  f (x 1 ) + f (x 2 )
f  1  >
 2  2
x + x 2 f (x 1 ) + f (x 2 )
Coordinates of M are  1 , 
 2 2 
Thus, (a5 + b5 + c 5 + d 5) ≥ abcd (a + b + c + d )
x + x 2 x + x2 
Coordinates of R are  1 , f  1 
 2  2 
y (x2, f(x2))
Now, from the figure R
Q
RC > MC
(x1, f(x1)) M
x + x2 
RC = f  1  P
 2 
f (x 1 ) + f (x 2 )
MC =
2 A C B x
+ + x1 x2
f  1 2 ≥
x x f (x ) f (x 2 ) O x1+x2
∴ 
1
2
 2  2
Concept If f ′ ′ (x ) > 0, ∀ x ∈ ] a , b [, then

x + x 2 + K + x m  f (x 1 ) + f (x 2 ) + K + f (x m )
f  1 ≤
 m  m
Note In this case, graph of f ( x ) is concave up.
www.pdfworld.in
www.pdfworld.in
Inequalities 247
It is easy to derive the AM–GM inequality as a special case of the Jensen's inequality.
Suppose y1 , y 2 , K , ym are +ve real numbers.
Let x i = log yi for i = 1, 2, K , m
i.e., logarithm w.r.t. the base e (so that for i = 1, 2, K , m, we have yi = e xi ) .
Take [a , b ] to be any interval containing all the numbers x 1 , x 2 , K , x m
The function h (x ) = e x satisfies the condition in the inequality.
Q h ′ ′ (x ) = e x is +ve.
∴ By Jensen's inequality, we have
x1 + x 2 + K + xm x1 x2
e +e + K + e xm
e m ≤
m
xi
Q e = yi for i = 1, 2, K , m
1
y1 + y 2 + K + ym
i.e., (y1 y 2 K ym )m ≤
m
which is truly by AM–GM inequality. Also, equality holds if and only if all x i ′ are equal which is same
thing as saying that all yi ′ s are equal.
Example 1 ABC is an acute angled triangle, show that

cos A + cos B + cos C ≤ 3 / 2.
Solution QABC is an acute angled triangle.
∴ A, B, C would lie in the interval [ 0, π / 2]
Let h( x ) = cos x
h′ ( x ) = − sin x
 π
h′ ′ ( x ) = − cos x < 0 ∀ x ∈ 0, 
 2
By using Jensen's inequality
 x + x 2 + ... + xm  f ( x1) + f ( x 2 ) K f ( xm )
f 1  ≥
 m  m
 A + B + C  cos A + cos B + cos C
⇒ cos   ≥
 3  3
Now, A+B+C=π
π cos A + cos B + cos C
⇒ cos ≥
3 3
⇒ cos A + cos B + cos C ≤ 3 / 2
Equality holds if and only if A = B = C
Example 2 In a ∆ ABC; A, B, C ∈( 0, π ), show that

cos A + cos B + cos C ≤ 3 / 2 .
Solution Here, it is not given that the ∆ABC is acute angled triangle.
So, we need to modify the solution a little.
For this, we first note that in any ∆ABC (whether it is acute angled or not).
www.pdfworld.in
www.pdfworld.in
A B C
sin , sin , sin are always +ve.
2 2 2
Hence,
 A + B  A − B
cos A + cos B = 2 cos   cos  
 2   2 
C  A − B C
= 2 sin cos   ≤ 2 sin
2  2  2
A
Similarly, cos B + cos C ≤ 2 sin
2
B
cos C + cos A ≤ 2 sin
2
Adding these we have
A B C
cos A + cos B + cos C ≤ sin + sin + sin …(i)
2 2 2
Instead of applying Jensen's inequality to the function f ( x ) = cos x over [ 0, π ], as we
do not know about second derivative of cos x which is neither +ve throughout nor –ve
throughout in 0 < x < π.
∴ By applying Jensen's inequality
x
for g ( x ) = sin  
 2
1 x
g′ ( x ) = cos
2 2
1 x
g ′ ′ ( x ) = − sin < 0 ∀ x in ] 0, π [
4 2
∴ Applying the inequality
 A + B + C  g ( A ) + g (B ) + g (C )
g  ≥
 3  3
A B C
sin + sin + sin
⇒ 2 2 2 ≤ sin  π 
3  6
A B C 3
⇒ sin + sin + sin ≤ …(ii)
2 2 2 2
3
cos A + cos B + cos C ≤
2
Example 3 Show that

 π
tan A + tan B + tan C ≥ 3 3, A, B, C ∈ 0, .
 2
Solution Let f ( x ) = tan x
f ′ ( x ) = sec2 x
f ′ ′ ( x ) = 2 sec2 x tan x > 0 ∀ x ∈[ 0, π / 2[
 A + B + C  f ( A ) + f (B ) + f (C )
f  ≤
 3  3
www.pdfworld.in
www.pdfworld.in
Inequalities 249
 A + B + C
tan A + tan B + tan C ≥ 3 tan  
 3 
π
≥ 3 tan
3
∴ tan A + tan B + tan C ≥ 3 3
Example 4 If 0 < Ai < π, ∀ i = 1, 2, K , n, show that

 A + A2 + K + An 
sin A1 + sin A2 + K + sin An ≤ n sin  1 
 n 
Solution Let f ( x ) = sin x
f ′ ( x ) = cos x
f ′ ′ ( x ) = − sin x < 0, ∀ x ∈ ] 0, π [
∴ Using Jensen's inequality
 A + A2 + K + An  f ( A1) + f ( A2 ) + K + f ( An )
f 1  ≥
 n  n
 A + A2 + K + An  sin A1 + sin A2 + K + sin An
⇒ sin  1 ≥
 n  n
( A1 + A2 + K + An )
⇒ sin A1 + sin A2 + K + sin An ≤ n sin
n
Finding the greatest value of ambnc p K, when m, n, p,K being +ve integers, a + b + c
is constant.
Let Z denotes the product ambnc p …
Then Z = ambnc p
  a     b   c  
m n p
= mm    nn   p p   …
  m     n    p  
  
m n p
 a   b  c 
= mm ⋅ n n ⋅ p p K       …(i)
 m   n   p
Q m, n, p are constants.
mm , nn , p p are also constants.
Hence, Z will be maximum when
m n p
 a   b  c 
      … is maximum.
 m   n   p
m n p
 a   b  c  a
But       … is the product of m factors such that each equal to   , n
 m   n   p  m
 b c
factors each equal to   , p factors each equal to   etc.
n  p
a  b c
The sum of these factors is equal to m   + n   + p   + K or
 m n  p
m n p
 a   b  c 
a + b + c K, which is given to be constant. Hence, the product       …
 m   n   p
www.pdfworld.in
www.pdfworld.in
a b c
will be maximum when all the factors , , , … are equal i.e., when
m n p
a b c a + b + c +K
= = =K =
m n p m + n + p +K
Thus, the greatest value of the product Z from Eq. (i)
m
 a + b + c +K
= mm n n p p K  
 m + n + p + K
n m + n + p + ...
 a + b + c +K  a + b + c +K
  = mm n n p p K  
 m + n + p + K  m + n + p + K
Corollary Let x 1 , x 2 , x 3 , K , x n are n +ve variables and Z is constant, then
If x 1 + x 2 + x 3 + K + x n = Z (constant)
n
The value of x 1x 2x 3 K x n is greatest when x 1 = x 2 = x 3 = K = x n and is given by   .
Z
n
Example 1 Find the greatest value of a 2b 3c 4 subject to the condition a + b + c = 18.

Solution Let Z = a 2b 3c 4
2 3 4
 a   b  c 
Z = 223344       …(i)
 2  3  4
∴ Z will have maximum value when
2 3 4
 a   b  c 
      is maximum.
 2  3  4
2 3 4
 a   b  c 
But       is product of 2 + 3 + 4
 2  3  4
i.e., 9 factors whose sum
a  b c
= 2  + 3  + 4  
 2  3  4
= a + b + c = 18 (constant)
2 3 4
 a   b  c 
∴       will be maximum if all the factors are equal.
 2  3  4
a b c a + b + c 18
i . e., If = = = = =2
2 3 4 2+ 3+ 4 9
∴ Maximum value of Z from Eq. (i)
 a b c 
= 2 2 ⋅ 3 3 ⋅ 4 4 ( 2 )2 ( 2 )3 ( 2 )4 Q 2 = 3 = 4 = 2
= 219 × 33
Example 2 If 2x + 3y = 7 and x ≥ 0, y ≥ 0, then find the greatest value of x 3y 4.

3 4 3
 3  4  2x   3y 
Solution Let Z = x 3y 4 =         …(i)
 2  3  3   4 
3 4
 2x   3y 
∴Z will have maximum value when     is maximum.
 3  4
www.pdfworld.in
www.pdfworld.in
Inequalities 251
3 4
 2x   3y 
But     is the product of 3 + 4 = 7 factors, the sum of which
 3  4
 2x   3y 
= 3   + 4   = 2x + 3y = 7(constant)
 3  4
3 4
 2x   3y 
∴     will be maximum if all the factors are equal.
 3  4
2x 3y 2x + 3y 7
i . e., If = = = =1 [Q 2x + 3y = 7]
3 4 3+ 4 7
∴ From Eq. (i) the maximum value of Z i.e., x 3y 4
3 4
 3  4  2x 3y 
=     (1)3(1)4 Q 3 = 4 = 1
 2  3
27 256 32
= × =
8 81 3
Example 3 Find the greatest value of x 2y 3z 4. If x 2 + y 2 + z 2 = 1, where x , y , z are +ve.

Solution Let A = x 2y 3z 4
then A 2 = x 4y 6z 8
2 3 4
 x 2  y 2  z 2
= 22 ⋅ 33 ⋅ 44       …(i)
 2  3  4
2 3 4
 x 2  y 2  z 2
∴ A will have maximum value when A 2 is maximum i.e., when       is
 2  3  4
maximum.
2 3 4
 x 2  y 2  z 2
But       is the product of 2 + 3 + 4 i.e., 9 factors.
 2  3  4
 x 2  y 2  z 2
The sum of which = 2   + 3   + 4  
 2  3  4
= x 2 + y 2 + z 2 = 1 (constant)
2 3 4
 x 2  y 2  z 2
∴       will be maximum if all the factors are equal i.e.,
 2  3  4
x2 y2 z2 x2 + y2 + z2 1
if = = = = [Q x 2 + y 2 + z 2 = 1]
2 3 4 2+ 3+ 4 9
From Eq. (i) maximum value of A 2 is
2 3 4
 1  1  1 22 × 33 × 44
22 ⋅ 33 ⋅ 44       =
 9  9  9 99
22 × 33 × 28
=
318
210
=
315
∴ Maximum value of x 2y 3z 4 is 25 × 3−15/ 2.
www.pdfworld.in
www.pdfworld.in
Example 4 If yz + zx + xy = 12 , where x , y , z are +ve values, find the greatest value of xyz.
Solution Let A = xyz
A 2 = x 2y 2z 2 = ( xy )( yz )( zx ) …(i)
∴ A will have maximum value when A 2 is maximum i.e., when ( xy )( yz )( zx ) is
maximum.
But ( xy )( yz )( zx ) is product of 3 factors. The sum of which xy + yz + zx = 12
(constant)
xy yz zx
∴ ( xy )( yz )( zx ) will be maximum if all the factors are equal i.e., if = =
1 1 1
xy + yz + zx 12
= = =4
1+1+1 3
From Eq. (i), maximum value of A 2 = 64
∴ Maximum value of xyz is 8.
x2 y2 z2
Example 5 Find the maximum value of xyz when + + = 1.
a 2 b2 c 2
Solution Let A = xyz
 x 2  y 2  z 2
∴ A 2 = x 2y 2z 2 = a 2b 2c 2  2   2  2 …(i)
a  b  c 
 x 2  y 2  z 2
∴ A will be maximum when A 2 is maximum i.e., when  2   2   2  is maximum.
a  b  c 
 x 2  y 2  z 2
But  2   2  2  is the product of 3 factors.
a  b  c 
x2 y2 z2
The sum of which = + + = 1 (constant).
a 2 b2 c 2
 x 2  y 2  z 2
∴  2  2  2  will be maximum if all the factors are equal.
a  b  c 
x2 y2 z2
i.e., If a = b = c2
2 2
1 1 1
x2 y2 z2
+ 2 + 2  x2 y2 z2 
=a
2
b c =1
Q 2 + 2 + 2 = 1
1+1+1 3  a b c 
 1  1  1
From Eq. (i) maximum value of A 2 = a 2 ⋅ b 2 ⋅ c 2      
 3  3  3
abc
∴ Maximum value of A =
27
Example 6 Prove that the greatest value of xy is c 3 / 2ab , if a 2x 4 + b 2y 4 = c 6.

Solution Let z = xy
1
z 4 = ( xy )4 = (a 2x 4 )(b 2y 4 ) …(i)
(ab )2
www.pdfworld.in
www.pdfworld.in
Inequalities 253
∴ z will maximum when z 4 is maximum i.e., when (a 2x 4 )(b 2y 4 ) is maximum.

But (a 2x 4 ) (b 2y 4 ) is the product of 2 factors whose sum is a 2x 4 + b 2y 4 = c 6
∴ (a 2x 4 )(b 2y 4 ) will be maximum if both the factors are equal.
a 2x 4 b 2y 4
i.e., If =
1 1
a 2x 4 + b 2y 4 c 6
= =
1+1 2
1 c6 c6
∴Maximum value of z 4 =   
(ab )2  2   2 
c3
Maximum value of z is .
2ab
Hence proved.
1 1
Example 7 If x 2 + y 2 = c 2, find the least value of 2
+ 2.
x y
1 1 y2 + x2 c2
Solution Let z ′ = 2
+ 2 = 2 2
= 2 2
x y x y x y
x 2y 2
∴z′ will be minimum when will be maximum.
c2
x 2y 2 1
Now, Let z = = 2 ( x 2 )( y 2 ) …(i)
c2 c
∴z will maximum when x 2y 2 is maximum but ( x 2 )( y 2 ) is the product of two factors
whose sum is x 2 + y 2 = c 2
∴ x 2y 2 will be maximum when both these factors are equal i.e., when
x 2 y 2 x 2 + y 2 c2
= = =
1 1 1 1
c2
From Eq. (i) maximum value of z =
4
1 1 4
∴ Least value of 2 + 2 = 2
x y c
Example 8 Find the greatest value of (a + x )3(a − x )4 for any real value of x numerically less
than a.
Solution Let z = (a + x )3(a − x )4
3 4
a + x  a − x 
= 33 ⋅ 44     …(i)
 3   4 
3 4 3 4
a + x  a − x  a + x  a − x 
z will be maximum, when     is maximum but     is
 3   4   3   4 
product of 3 + 4 = 7 factors.
The sum of which
a + x  a − x 
= 3  + 4  = (a + x ) + (a − x ) = 2a
 3   4 
www.pdfworld.in
www.pdfworld.in
3 4
a + x  a − x  a+x a−x
∴    will be maximum if all the factors are equal i.e., if =
 3   4  3 4
a
or 4a + 4x = 3a − 3x or x = −
7
So, from Eq. (i) maximum value of z
3 4
a − (a / 7)  a + (a / 7) 
= 33 ⋅ 44    
 3 4
3 4
 6a   8a 
= 33 ⋅ 44 ⋅    
 3 × 7  7 × 4
63 ⋅ 84 7
= a
77
Example 9 Find the greatest value of (a − x )(b − y )(c − z ) (ax + by + cz ), where a, b, c are +ve
quantities and (a − x ), (b − y ), (c − z ) are also +ve.
Solution Let z = (a − x )(b − y )(c − z )(ax + by + cz )
1
= (a 2 − ax )(b 2 − by )(c 2 − cz )(ax + by + cz ) …(i)
abc
∴ z will be maximum when (a 2 − ax )(b 2 − by ) (c 2 − cz ) (ax + by + cz ) is maximum.
But (a 2 − ax )(b 2 − by ) (c 2 − cz )(ax + by + cz ) is the product of 4 factors, the sum
of which
= (a 2 − ax ) + (b 2 − by ) + (c 2 − cz ) + (ax + by + cz )
= a 2 + b 2 + c 2 (constant)
∴ (a 2 − ax )(b 2 − by )(c 2 − cz ) (ax + by + cz ) will be maximum, if all factors are
equal, i.e., if
a 2 − ax b 2 − by c 2 − cz
= =
1 1 1
ax + by + cz
=
1
(a 2 − ax ) + (b 2 − by ) + (c 2 − cz ) + (ax + by + cz )
=
1+1+1+1
a 2 + b2 + c 2
=
4
4
1  a 2 + b2 + c 2 
∴ From Eq. (i), the maximum value of z =  
abc  4 
(a + x )(b + x )
Example 10 Find the value of .
(c + x )
Solution Let c + x = y or x = y − c
Then, the given expression
(a + y − c )(b + y − c ) (a − c + y )(b − c + y )
= =
y y
www.pdfworld.in
www.pdfworld.in
Inequalities 255
(a − c )(b − c ) + y (b − c ) + y (a − c ) + y 2
=
y
(a − c )(b − c )
= + (b − c ) + (a − c ) + y
y
2
 (a − c )(b − c ) 
= − y  + (b − c ) + (a − c ) + 2 (a − c )(b − c )
 y 
We find that as the value of the given expression varies as y varies. It is minimum
when
2
 (a − c )(b − c ) 
 − y  is zero.
 y 
(a − c )(b − c )
i.e., − y =0
y
or (a − c )(b − c ) = y
∴ The minimum value of the given expression
= (b − c ) + (a − c ) + 2 (a − c )(b − c )
Concept If the product of any number of +ve quantities is given, then their sum will be minimum if
the quantities are all equal.
Proof Let us consider first two quantities x and y. Their sum is denoted byY ′ and product by Z.
∴ Y ′ = x + y and Z = xy
We know that (x + y )2 − (x − y )2 = 4xy
or (Y ′ )2 − (x − y )2 = 4Z
or (Y ′ )2 = 4Z + (x − y )2
It is evident that (Y ′ )2 will be minimum and Z remaining constant i.e., when (x − y )2 is zero or when
x − y = 0 or x = y .
Let us suppose that there are more than two quantities. If we replace any two quantities x and y by two
equal quantities xy and xy , their product xy remains unaltered. Their sum is xy + xy i.e., 2 xy
which is less than x + y . [Q AM > GM]
So, the sum of any two quantities can be diminished by making the quantities equal whereas their
product remains unchanged.
∴ When all the quantities are equal their sum will have minimum value.
Corollary
1. If x 1x 2x 3 K x n = z (constant), then the value of x 1 + x 2 + x 3 + K + x n is least when x 1 = x 2 = K = x n ,
then n (z )1/n is least value of x 1 + x 2 + x 3 + K + x n .
2. If x 1 + x 2 + x 3 + K + x n = z (constant) and if r may or may not lie between 0 and 1, then the least
or the greatest value of x 1r + x r2 + x 3r + .. + x nr occurs when x 1 = x 2 = x 3 = K = x n and is given by
n1 − r ⋅ zr .
3. If x 1r + x r2 + x 3r + K + x nr and if r may or may not lie between 0 and 1, then the least or the greatest
1 1
1−
value of x 1 + x 2 + x 3 + K + x n occurs when x 1 = x 2 = x 3 = K = x n and is given by n r ⋅ zr .
www.pdfworld.in
www.pdfworld.in
Example 1 If x 2y 3 = 6, find the least value of 3x + 4y for +ve value of x and y.

Solution We have x 2y 3 = 6
2 3 2 3
 3x   4y   3  4
or     =6×   
 2  3  2  3
2 3
 3x   4y 
⇒     = 32 …(i)
 2  3
2 3
 3x   4y 
Now,     is product of 5 ( 3 + 2) factors and product = 32, which is
 2  3
constant. Hence, the sum of these factors will be minimum when all of them are
equal.
4y 3x
i.e., = …(ii)
3 2
2 3 5
 3x   3x   3x 
∴     = 32 or   = 32 = 2
5
[from Eq. (i)]
 2  2  2
3x
or =2
2
∴ From Eq. (ii), we have
3x 4y
= =2
2 3
∴ Minimum value of the sum of the factors of
2 3
 3x   4y   3x   4y 
    = 2   + 3   = 2( 2) + 3( 2) = 10
 2  3  2  3
Example 2 Find the minimum value of bcx + cay + abz when xyz = abc.
Solution We have abc = xyz
or (bcx )(cay )(abz ) = (abc )3 …(i)
Now, (bcx )(cay )(abz ) is product of 3 factors. Its product is constant.
∴ Sum of these factors will be minimum when all of them are equal.
i.e., abz = cay = bcx …(ii)
Now, (bcx )(bcx )(bcx ) = (abc ) 3
or (bcx )3 = (abc )3
Hence, abc = bcx
From Eq. (ii), we get bcx = acy = abc = abz
∴ Required minimum value is 3abc.
Example 3 Prove that the cube is the rectangular parallelopiped of maximum volume for given
surface and of minimum surface for given volume.
Solution We have a, b, c as the edges of rectangular parallelopiped.
Let V be its volume and S be its surface.
Then S = 2(ab + bc + ca ) and V = abc
www.pdfworld.in
www.pdfworld.in
Inequalities 257
(i) Surface constant

Now, we have to find a, b, c when V is maximum.
i.e., V = abc
1
V 2 = (abc )2 = ( 2ab )( 2bc )( 2ca )
8
Now, V 2 is maximum when ( 2ab )( 2bc )( 2ca ) is maximum.
But ( 2ab )( 2bc )( 2ca ) is the product of 3 factors sum of which 2ab + 2bc + 2ca is
constant.
∴ The product ( 2ab )( 2bc )( 2ca ) will be maximum when all factors are equal i.e.,
when
2ab = 2bc = 2ca
1 1 1
i.e., when = = [Dividing by 2abc ]
a b c
a =b =c
∴When all the edges of the parallelopiped are equal i.e., when it is cube.
(ii) Volume constant
abc = constant
∴ 8(abc )2 = constant
i.e., ( 2ab )( 2bc )( 2ca ) = constant
Product of these factors ( 2ab ), ( 2bc ) and ( 2ca ) being constant, their sum
( 2ab + 2bc + 2ca ) will be minimum when all of these factors are equal.
i.e., when 2ab = 2bc = 2ca
1 1 1
i.e., when = =
c b a
i.e., c =b =a
i.e., when parallelopiped is a cube.
Example 4 Prove that the equilateral triangle has maximum area for given perimeter and minimum
perimeter for given area.
Solution A and P are the area and the perimeter of a triangle respectively.
∴ A = s(s − a )(s − b )(s − c ) and P = 2s
where 2s = a + b + c
Let s −a = x
s −b = y
s −c = z
Then A = (sxyz ) or A 2 = sxyz
x + y + z = (s − a ) + (s − b ) + (s − c )
= 3s − (a + b + c ) = 3s − 2s = s
(i) If perimeter of triangle is constant.
P = 2s = 2( x + y + z ) = constant
A 2 = (sxyz )
A 2 will be maximum, if xyz is maximum.
www.pdfworld.in
www.pdfworld.in
Now, xyz is the product of 3 factors whose sum is x + y + z (a constant)

∴ xyz will be maximum when all the factors are equal
i.e., when x = y = z i.e.,
when (s − a ) = (s − b ) = (s − c )
i.e., when a = b = c i.e., when triangle is equilateral.
(ii) When area is constant.
A 2 = s( xyz ) = ( x + y + z )( xyz ) = x 2yz + y 2zx + z 2xy
Let x 2yz = D, y 2zx = E , and z 2xy = F
Then A 2 = D + E + F = constant
 D E F 
P = 2s = 2( x + y + z ) = 2  + + 
 xyz xyz xyz 
2(D + E + F ) 2A 2 2A 2
= = = [Q DEF = x 4y 4z 4]
xyz xyz (DEF )1/ 4
2A 2
or P=
(DEF )1/ 4
2A 2 (DEF )1/ 4
So, perimeter P will be minimum when is minimum i.e., when is
(DEF )1/ 4 2A 2
maximum or (DEF )1/ 4 is maximum or DEF is maximum.
But D + E + F = constant (given)
∴ DEF is maximum only when all the factors are equal i.e., when D = E = F
i.e., x 2yz = y 2zx = z 2xy
or x = y = z or (s − a ) = (s − b ) = (s − c )
or a = b = c i.e., triangle is equilateral.
Example 5 If x 2y 3 = 6 , find least value of 3x + 4y .

Solution We have x 2y 3 = 6
Now, ( x )( x )( y )( y )( y ) = 6
 3x   4y  3x 3x 4y 4y 4y
∴ 3x + 4y = 2   + 3   = + + + +
 2  3 2 2 3 3 3
(Multiply and divide coefficient of x and y by 2 and 3 respectively).
2
 3x   3x   4y   4y   4y  3 43
∴           = 2 × 3 × 6 = 32
 2 2 3 3 3 2 3
Here, n=5
z = 32 ∴ 5( 32)1/ 5 = 10
Example 6 Find the minimum value of bcx + cay + abz when xyz = abc
Solution We have xyz = abc
(bcx )(cay )(abz ) = a 3b 3c 3 = z
n=3
Minimum value of bcx + cay + abz
= n( z )1/n = 3(a 3b 3c 3 )1/ 3 = 3abc
www.pdfworld.in
www.pdfworld.in
Inequalities 259
1 1 1
Example 7 Find the minimum value of + + for possible values of x , y , z which satisfy the
x y z
condition x + y + z = 9.
Solution We have r = −1
n=3
1 1 1
∴ Minimum value of + + is 31 − (−1) ⋅ 9−1
x y z
= 9 / 9 =1
Example 8 Find the greatest value of 2( x )1/ 2 + 3( y )1/ 2 + 4( z )1/ 2 + 5(a )1/ 2 for possible values of
x , y , z , a which satisfy the condition
4x + 9y + 16z + 25a = 720.
Solution z = 720
Now, 2( x )1/ 2 + 3( y )1/ 2 + 4( z )1/ 2 + 5(a )1/ 2
= 4x + 9y + 16z + 25a
1
Now, we have r = and n = 4
2
1
1−
Hence, greatest value is 4 2 (720 )1/ 2
= 24 5
Example 9 Find the maximum value of a + b + c for possible values of a, b, c satisfying condition
a 3 + b 3 + c 3 = 1.
Solution Here, r = 3 and n = 3
1
1−
∴ Maximum value of x + y + z = 3 3 11/ 3
= 32/ 3
Concept 1. If a and b are +ve quantities and a > b. If x be a +ve quantity, then
a b
1 + x  > 1 + x 
   
 a  b
Concept 2. If a and b are +ve proper fractions and a > b, then
1/ a 1/b
1 + a  1 + b 
  > 
1 − a 1 − b 
a b
a + c b + c
Example 1 If a, b, c are descending order of magnitude, show that   < 
a − c b − c
Solution Q a >b >c
∴ c/a and c/b are +ve fraction.
Also, c/a < c/b
www.pdfworld.in
www.pdfworld.in
a b
a + c b + c
Now, we have to prove that   < 
a − c b − c
a b
a b
 c  c  cc  cc
1 +  1 +  1 +  1 + 
or  a  < b  or  a  < b
1 − c  1 − c  1 − c  1 − c 
 a  b  a  b
1 1
1 + y  y 1 + x  x
or   < 
1 − y  1 − x 
1 1
c c 1 + x  x 1 + y  y
where x = ,y = or   > 
b a 1 − x  1 − y 
x >y [Qc / b > c / a ]
a b
a + c b + c
∴   < 
a − c b − c
Example 2 If x is a +ve proper fraction. Prove that

(1 + x )1 − x (1 − x )1 + x < 1.
a+b
 a + b
Also, show abba <   .
 2 
Solution Let z = (1 + x )1 − x (1 − x )1 + x
∴ log z = (1 − x )log(1 + x ) + (1 + x )log(1 − x )
= [log(1 + x ) + log(1 − x )] − x [log(1 + x ) − log(1 − x )]
 x2 x3   x2 x3 
=  x − + − K +  − x − − −K  
 2 3   2 3  
 x2 x3 x4   x2 x3 x4 
− x  x − + − + K −  − x − − − − K 
 2 3 4   2 3 4  
x2 x4 x6   x3 x5 
= −2 + + + K − 2x  x + + + K
 2 4 6   3 5 
 1   1 1  1 1 
= − 2 x 2  + 1 + x 4  +  + x 6  +  + K
 2   4 3  6 5 
 x 2 7x 4 11x 6 
= − 2 3 + + +K
 1 ⋅ 2 3 ⋅ 4 5 ⋅ 6 
= –ve as x is a +ve proper fraction.
∴ log z = − ve or z < 1
or (1 + x )1 − x (1 − x )1 + x < 1 …(i)
Hence proved.
a −b 2a
Let x = , then 1 + x =
a+b a+b
www.pdfworld.in
www.pdfworld.in
Inequalities 261
2b
and 1− x =
a+b
Substituting these values in (i), we get
2b 2a
b a
 2a  a + b  2b  a + b  2a   2b 
    < 1 or     <1
 a + b  a + b  a + b  a + b
b a
 2  b 2  a
or   a   b <1
 a + b  a + b
a+b
 2 
or abba   <1
 a + b
a+b
 a + b
Hence, abba <  
 2 
Example 3 If x < 1 , prove that (1 + x )1 + x (1 − x )1− x < 1 . Also, show that

a+b
 a + b
aabb >  
 2 
Solution Let z = (1 + x )1 + x (1 − x )1 − x , then
log z = (1 + x )log(1 + x ) + (1 − x )log(1 − x )
= [log(1 + x ) + log(1 − x )] + x [log(1 + x ) − log(1 − x )]
 x2 x3   x2 x3 
=  x − + − K  +  −x − − − K 
 2 3   2 3  
 x2 x3   x2 x3 
+ x  x − + − K −  − x − − −K  [Q x < 1]
 2 3   2 3  
x2 x4 x6   x3 x5 
= −2 + + + K + 2x  x + + + K
 2 4 6   3 5 
1   1 1  1 1
= − 2x 2  − 1 − 2x 4  −  − 2x 6  −  + …
2   4 3  6 5
1 4 1 6
= x2 + x + x + K = + ve
6 15
∴ log z > 0 or z > 1
or (1 + x )1 + x (1 − x )1 − x > 1 …(i)
Hence proved.
a −b
Now, let x =
a+b
2a 2b
1+ x = and 1 − x =
a +b a+b
Substituting these values in (i), we get
2a 2b
 2a  a + b  2b  a + b
    >1
 a + b  a + b
www.pdfworld.in
www.pdfworld.in
a+b
Raising both sides to power , we get
2
aabb (2 /a + b )a + b > 1
a+b
 a + b
or aabb >  
 2 
Hence proved.
x
Example 4 If x is +ve, then show that log(1 + x ) < x and log(1 + x ) > . Also, show that
1+ x
1 1 1 1
log(1 + n ) < 1 + + + +K+ .
2 3 4 n
Solution If log(1 + x ) < x, then (1 + x ) < e x
x2 x3
or 1 + x <1 + x + + + K which is true. [Qx is +ve]
2! 3!
∴ log(1 + x ) < x …(i)
x
x 1+ x
If log(1 + x ) > , then (1 + x ) > e
1+ x
x −1 x
1 1 + x  x 
or >e or 1 −  > e1 + x
1−
x  1+ x
1+ x
2 3
x  x   x 
or 1+ +  +  +K
1 + x 1 + x  1 + x 
2 3
 x  1  x  1  x 
>1 +   +   +   +K
1 + x  2! 1 + x  3! 1 + x 
It is clear that above relation is true.
x
∴ log(1 + x ) >
1+ x
Hence proved.
Now, log (1 + x ) < x
 1 1  1
∴ log 1 +  < Put x = y 
 y y  
1 + y  1 1
or log   < or log (1 + y ) − log y <
 y  y y
Put y = 1, 2, 3, K , n, we get
1
log 2 − log1 <
1
1
log 3 − log 2 <
2
1
log 4 − log 3 <
3
www.pdfworld.in
www.pdfworld.in
Inequalities 263
1
log 5 − log 4 <
4
…………
…………
…………
1
log(n + 1) − log n <
n
Adding the above terms, we have
1 1 1 1
log(n + 1) − log1 < 1 + + + +K+
2 3 4 n
1 1 1
or log(1 + n ) < 1 + + + .... + [Q log1 = 0]
2 3 n
(n + 1)n
Example 5 Show that en > , n being an integer.
n!
n2 n3 nn
Solution en = 1 + n + + + ...+ +K
2! 3! n!
n2 n3 nn
en > 1 + n + + +K+
2! 3! n!
nn nn − 1 nn − 2 n3 n2
en > + + + .... + + + n +1
n ! (n − 1)! (n − 2)! 3! 2!
1 1 1 1 1 1
or en > nn  + + + + K n −1 + n 
n ! n(n − 1)! n 2(n − 2)! n 3(n − 2)! n n 
nn  1 1 1 n! 
i.e., en > 1 + n ⋅ n + n 2 ⋅ n(n − 1) + n 3 ⋅ n(n − 1) (n − 2) + K + nn 
n!
 n(n − 1) 
Q n(n − 1) > 2! 
 
n n
 1 n (n − 1) 1 1  n(n − 1)(n − 2) > n (n − 1)(n − 2 ) 
en > 1 + n ⋅ + ⋅ + +
n !  nn  
K
n 2! n2 3! 
n ! > 1 (Q n is an integer)
 
 
nn  1
∴ en > 1 + 
n!  n
nn (n + 1)n
i . e., en > ⋅
n! nn
(n + 1)n
en >
n!
Hence proved.
www.pdfworld.in
www.pdfworld.in

Example 1. If a1 , a2 , a3,…, an be non-negative real numbers such that a1 + a2 + a3 + …+ an = m, then
m2
prove that Σ ai a j ≤ .
i < j 2
Solution Now, m = a1 + a 2 + a3 + … + an
∴ m 2 = (a1 + a 2 + a3 + K + an )2
m 2 = a12 + a 22 + K + an2 + 2Σ ai a j
i < j
m − 2Σai a j =
2
a12 + a 22 + K + an2
Q a12 + a 22 + K + an2 ≥ 0
∴ m 2 − 2Σai a j ≥ 0
m2
⇒ Σ ai a j ≤
i < j 2
Hence proved.
2
Example 2. Prove Σ i ⋅ j Ci2C j2 < n (2n −1
Cn )2, where Ci , C j are binomial coefficient.
i < j 2
Solution From Binomial theorem, we know that
−1
1C12 + 2C22 + 3C32 + K + nCn2 = n 2n Cn
Squaring both sides, we get
−1
(1C12 + 2C22 + 3C32 + K + nCn2 )2 = n 2 (2n Cn )2
12C14 + 22C24 + 32C34 + K + n 2Cn4 + 2Σi ⋅ j ⋅ Ci2C j2
−1
= n 2 (2n Cn )2
Q 12C14 + 22C24 + K + n 2Cn4 > 0
−1
∴ n 2 (2n Cn )2 − 2Σ i ⋅ j Ci2C 2j > 0
n 2 2n − 1
⇒ Σ i ⋅ j Ci2 C j2 < ( Cn )2
i < j 2
Hence proved.
+1
Example 3. Given n 4 < 10n for n ≥ 2. Prove that (n + 1)4 < 10n .
4 4
Solution Consider  n + 1  =  1 + 1 
  n n
We are given n ≥ 2
1 1
⇒ ≤
n 2
1 1
so 1+ ≤1+
n 2
4 4
1 + 1  ≤ 1 + 1 
or    
 n  2
www.pdfworld.in
www.pdfworld.in
Inequalities 265
4 4 4
1 + 1  ≤  3  ⇒  n + 1  < 10
or      
 n  2  n 
Since, the logic is that if n ≤ 2 it is also less than 3, 6, 7, 8 etc. So, we can consider any one.
n 4 ⋅ 10 > (n + 1)4 or (n + 1)4 < n 4 ⋅ 10 …(i)
So, n 4 < 10n
+1
∴ 10n 4 < 10n …(ii)
+1
(n + 1)4 < 10n
Hence proved.
1 1 1
Example 4. If θi ∈ (− π , π ) for i = 2, 3, K , n. Prove ⋅ ... ≥ n!
sin 2 θ 2 sin 2 θ 3 sin 2 θn
2 3 n
2
θ 2. cosec2 θ 3 2
θn
or 2cosec 3 K n cosec ≥ n!
Solution ∀ i = 2, 3, K , n
sin 2 θi ≤ 1
1 1
⇒ ≥1 ⇒ ≥ i1
sin 2 θi i
sin 2 θi
i = 2, 3, K , n
1 1 1
⇒ ≥ 2, ≥ 3, K , ≥n
sin 2θ 2 sin 2 θ 3 sin 2 θn
2 3 n
Therefore,
1 1 1
⋅ K ≥ 2⋅3 K n ≥ n !
sin 2 θ 2 sin 2 θ 3 2
θn
2 3 n sin
cosec 2θ 2 cosec 2θ 3 2
θn
or 2 ⋅3 K n cosec ≥ n!
Hence proved.
n!
Example 5. If n Cr = , then prove
r !(n − r )!
−1 n −1
C1 + C2 + K + Cn ≤ 2n +
2
Solution Consider
( C1 − 1)2 + ( C2 − 1)2 + ( C3 − 1)2 + K + ( Cn − 1)2 ≥ 0
⇒ (C1 + C2 + C3 + K + Cn ) − 2( C1 + C2 + K + Cn ) + n ≥ 0
⇒ 2 − 1 + n ≥ 2( C1 +
n
C2 K + Cn ) (Q C0 + C1 + C2 + K+ Cn = 2n )
2n + n − 1
⇒ ( C1 + C2 + K + Cn ) ≤
2
−1 (n − 1)
( C1 + C2 + K+ Cn ) ≤ 2n +
2
Hence proved.
www.pdfworld.in
www.pdfworld.in
Example 6. If a > b > 0, then without using AM–GM prove

−1 −1
1 + a + a 2 + K + an 1 + b + b 2 + K + bn
< .
1+ a + a + K+ a
2 n
1+ b + b + K+ b
2 n
−1
1 + a + a 2 + K + an
Solution Let A = n −1
1+ a + a + K+ a
2
+ an
−1
1 + b + b 2 + K + bn
and B= n −1
1+ b + b + K+ b
2
+ bn
1 1 + a + a 2 + K + an − 1 + an
∴ =
A 1 + a + a 2 + K + an − 1
1 an
=1+ −1
…(i)
A 1 + a + a + K + an
2
1 1
=1+ …(ii)
A 1 1 1
+ n −1 + K +
an a a
1 bn
Similarly, =1+ −1
…(iii)
B 1 + b + b 2 + K + bn
1 1
=1+ …(iv)
B 1 1 1
+ n −1 + K +
bn b b
1 1 1 1
It is given a > b > 0 ⇒ < ⇒ n < n , ∀n ∈ N
a b a b
⇒  1 + 1 + K + 1  <  1 + 1 + K+ 1 
   
a a2 an   b b 2 bn 
1 1
⇒ >
1 + 1 + K + 1  1 + 1 + K + 1 
   
a a2 an   b b 2 bn 
Using Eqs. (ii) and (iii), we get
1 1 1 1
− 1> − 1⇒ > ⇒A < B
A B A B
−1 −1
1 + a + a 2 + K + an 1 + b + b 2 + K + bn
⇒ <
1+ a + a + K+ a
2 n
1+ b + b + K+ b
2 n
Hence proved.
Example 7. If a1 , a2 , a3 , K , an and b1 , b2 , b3 , K , bn be +ve such that k the largest of fraction

ai
, i = 1, 2, K , n. Prove
bi
a1 + a 22 + a33 + K + ann
−1 n
≤ k.
b1 + kb22 + k 2b33 + K + kn bn
ai
Solution Qk is maximum , ∀ i = 1, 2, K , n
bi
ai
⇒ ≤ k, ∀ i = 1, 2, K , n
bi
a1 ≤ kb1 , a 22 ≤ k 2b22 , a33 = k3b33 , K , ann ≤ kn bnn
www.pdfworld.in
www.pdfworld.in
Inequalities 267

a1 + a 22 + a33 + K + ann ≤ kb1 + k 2b22 + k3b33 + K + kn bnn
−1 n
⇒ a1 + a 22 + a33 + K + ann ≤ k (b1 + b22k + k3b33 + K + kn bn )
a1 + a 22 + a33 + K + ann
⇒ −1 n
≤k
b1 + kb22 + k 2b33 + K + kn bn
Example 8. If a1 , a2 , a3 , K , an are unequal real number, then prove that

(1 + a1 + a12 )(1 + a 2 + a 22 ) K (1 + an + an2 )
≥ 3n .
a1a 2a3 K an
Solution For any a > 0, we have

1 + a + a2 1
=a + + 1
a a
1
Now, let a + =S
a
S ≥2
S + 1≥ 3
1
∴ a + + 1≥ 3
a
1 + ai + ai2
Thus, ≥ 3, ∀ i = 1, 2, 3, K , n
ai
 1 + a1 + a12   1 + a 2 + a 22   1 + a3 + a32   1 + an + an2 

⇒       …   ≥ 3 × 3 K n times = 3n
 a1  a 2  a3   an 
 1 + a1 + a12   1 + a 2 + a 22   1 + an + an2 
∴     K   ≥ 3
n
 a1  a 2   an 
Example 9. If a , b , c are +ve real numbers, then prove that the expression
1 bc ca ab
(a + b + c ) − − −
2 b+c c+a a+b
is always non –ve. Find the condition that this expression is zero. (RMO 1993)
Solution Let
1 bc ca ab
S = (a + b + c ) − − −
2 b+c c+a a+b
1 4  bc ca ab 
= (2a + 2b + 2c ) −  + + 
4 4 b + c c + a a + b 
1 4bc 4ca 4ab 

=  (b + c ) − +c+a− +a+b− 
4 b+c c+a a + b
1  (b − c )2 (c − a )2 (a − b )2 
=  + + ≥ 0 [Qa , b , c are all +ve].
4 b+c c+a a+b 
Now, S = 0 iff each of the three terms in the expression are zero i.e.,
a =b =c
www.pdfworld.in
www.pdfworld.in
Example 10. If a , b , c , d , e, f are real numbers, then show that

a2 + b2 + c2 + d 2 + e2 + f 2
≥ 1/ 2.
(a + c )2 + (b + d )2 + (c + e )2 + (d + f )2
+ (e + a )2 + (f + b )2
Solution From Third fundamental concept

a2 + b2 + c2 + d 2 ≥ (a + c )2 + (b + d )2 …(i)
c2 + d 2 + e2 + f 2 ≥ (c + e )2 + (d + f )2 …(ii)
e +f
2 2
+ a +b ≥
2 2
(e + a ) + (f + b )
2 2
…(iii)
2 ( a2 + b2 + c2 + d 2 + e2 + f 2 )
≥ (a + c )2 + (b + d )2 + (c + e )2 + (d + f )2 + (e + a )2 + (f + b )2
a2 + b2 + c2 + d 2 + e2 + f 2 1
∴ ≥
(a + c ) + (b + d ) +
2 2
(c + e ) + (d + f )
2 2 2
+ (e + a )2 + (f + b )2
Hence proved.
Example 11. If a , b , c are real +ve, then prove that

1 3
a6 + b6 + c 6 > [b ac (a + e ) + c 3ab (a + b ) + a3bc (b + c ) .
2
Without using AM–GM.
Solution We have, (a 2 − b 2 )2 ≥ 0
a 4 + b 4 − 2a 2b 2 ≥ 0
a 4 + b 4 − a 2b 2 > a 2b 2
(a 2 + b 2 )(a 4 − a 2b 2 + b 4 ) > a 2b 2 (a 2 + b 2 )
a 6 + b 6 > a 2b 2 (a 2 + b 2 ) …(i)
Also, a + b > 2ab

2 2
a b (a 2 + b 2 ) > 2a3b3
2 2
…(ii)
Using (i) and (ii)
⇒ a 6 + b 6 > 2a3b3 …(iii)
Similarly, b + c > 2b c
6 6 3 3
…(iv)
and c + a > 2c a
6 6 3 3
…(v)
Adding (iii), (iv), (v), we get
2(a 6 + b 6 + c 6 ) > 2(a3b3 + b3c 3 + c 3a3 )
a 6 + b 6 + c 6 > (a3b3 + b3c 3 + c 3a3 ) …(vi)
Now, we know that
1
x 3 + y 3 + z3 > [xy (x + y ) + yz (y + z ) + xz (x + z )]
2
www.pdfworld.in
www.pdfworld.in
Inequalities 269
1 3
∴ a3b3 + b3c 3 + c 3a3 > [b ac (a + c ) + c 3ab (a + b ) + a3bc (b + c )] …(vii)
2
From (vi) and (vii), we get
1 3
a6 + b6 + c 6 > [b ac (a + c ) + c 3ab (a + b ) + a3bc (b + c )]
2
Hence proved.
Example 12. If a , b , c are +ve real numbers, then show that

a 2b + ab 2 + c 2a + ca 2 + b 2c + bc 2 + 2abc ≥ 8abc (RMO 1992)
Solution Let S = a 2b + ab 2 + c 2a + ca 2 + b 2c + bc 2 + 2abc

Factorizing, we get
S = (b + c )(c + a )(a + b )
We have a + b ≥ 2 ab
b + c ≥ 2 bc and a + c ≥ 2 ac
Multiplying these, we get
(a + b )(b + c )(a + c ) ≥ 8abc
S ≥ 8abc
Hence proved.
Example 13. Show that in any triangle with sides a , b , c , we have

(a + b + c )2 < 4(ab + bc + ca ). (RMO 1992)
Solution Without loss of generality, let us assume a > b > c > 0, since the sum of any two sides of a
triangle is always greater than the third side. So, the difference between any two sides can never exceeds
the third.
∴ 0 2(bc + ca + ab )
⇒ (a + b + c )2 − 2ab − 2bc − 2ac < 2(ab + bc + ca )
⇒ (a + b + c )2 < 4(ab + bc + ca )
Example 14. If a < c , (b − a )(b − c ) < 0. Show that

3  + + 1 > (a + b + c )  + +  .
a c 1 1 1
c a  a b c 
Solution Q (b − a )(b − c ) < 0

⇒ b lies between a and c.
∴ a<b<c [Qa < c ]
1 1 1
⇒ < <
c b a
www.pdfworld.in
www.pdfworld.in
(a − b )  −  > 0
1 1
Now,
c b 
(b − c )  −  > 0 , (a − c )  −  > 0
1 1 1 1
b a  c a 
On adding
(a − b )  −  + (b − c )  −  + (a − c )  −  > 0
1 1 1 1 1 1
c b  b a  c a 
Simplifying, we get
3  + 1 +  > (a + b + c )  + + 
a c 1 1 1
c a a b c 
Hence proved.
Example 15. If a , b , c > 0 are such that a + b + c = 1 , prove that ab + bc + ca ≤ 1 / 3. (RMO 1988)
Solution For all real numbers a , b , c

(b − c )2 + (c − a )2 + (a − b )2 ≥ 0
so that a 2 + b 2 + c 2 ≥ ab + bc + ca
⇔ (a + b + c )2 ≥ 3(ab + bc + ca )
Q a + b + c = 1, we get 1 ≥ 3(ab + bc + ca )
1
i.e., ab + bc + ca ≤
3
Hence proved.
Note If a1, a 2, a 3, K , an are +ve real numbers whose sum is A, then
1
a1a 2 + a 2a 3 + a 3a 4 + K + an − 1an ≤ A 2
4
To prove it, we put a1 + a 2 + a 3 + K + an = A
a1 − a 2 + a 3 − a 4 + K = B
so that 2(a1 + a 3 + K ) = A + B,
2(a 2 + a 4 + a 6 + K ) = A − B
Multiplying corresponding sides of the above relations
We have
4(a1 + a 3 + K )(a 2 + a 4 + K ) = A 2 − B 2 ≤ A 2 (QB 2 ≥ 0)
1
Now, a1a 2 + a 2a 3 + a 3a 4 +K ≤ (a1 + a 3 + K )(a 2 + a 4 + K ) ≤ A 2
4
Example 16. If a , b , c , d are four non –ve real numbers and a + b + c + d = 1 , show that
ab + bc + cd ≤ 1 / 4. (INMO 1993)
Solution Let a + b + c + d = A
a −b + c −d = B
Then, 2(a + c ) = A + B , 2(b + d ) = A − B, so that
4(a + c )(b + d ) = A 2 − B 2 ≤ A 2 (QB 2 ≥ 0)
Now, we have A = 1
1
So, (a + c )(b + d ) ≤
4
Now, ab + bc + cd ≤ (a + c )(b + d ) ≤ 1 / 4
www.pdfworld.in
www.pdfworld.in
Inequalities 271
Example 17. If x , y , z > 0 and x + y + z = 1.

Prove x 2 + y 2 + z 2 ≥ 1 / 3.
Solution ∀ x , y , z we have
x 2 + y 2 + z 2 ≥ xy + yz + zx …(i)
Now, (x + y + z ) = x + y + z + 2(xy + yz + zx )
2 2 2 2
(x + y + z )2 − (x 2 + y 2 + z 2 )
⇒ xy + yz + zx =
2
1 − (x 2 + y 2 + z 2 )
⇒ xy + yz + zx =
2
Substitute the value of xy + yz + zx in (i), we get
1 − (x 2 + y 2 + z 2 )
x 2 + y 2 + z2 ≥
2
⇒ x 2 + y 2 + z2 ≥ 1 / 3
Example 18. For given real numbers, x 1 , x 2 , y1 , y 2 , z1 , z2 satisfying x 1 > 0, x 2 > 0, x 1y1 − z12 > 0 and
x 2y 2 − z22 > 0.
8 1 1
Prove that ≤ + , give necessary and sufficient conditions
(x 1 + x 2 )(y1 + y 2 ) − (z1 + z2 )2 x 1y1 − z12 x 2y 2 − z22
for equality.
Solution We have to prove
8 1 1
≤ + …(i)
(x 1 + x 2 )(y1 + y 2 ) − (z1 + z2 )2 x 1y1 − z12 x 2y 2 − z22
Let us denote the denominators appearing in (i) by A , A1 , A 2 respectively.

A = A1 + A 2 + x 1y 2 + x 2y1 − 2z1z2 …(ii)
Applying x + y ≥ 2 xy (x > 0, y > 0)
x1 x
⇒ A = A1 + A 2 + (A 2 + z22 ) + 2
x2 x1
(A1 + z12 ) − 2z1z2

2
x1 x  x x 
A = A1 + A 2 + A 2 + 2 A1 +  z1 2 − z2 1 
x2 x1  x1 x2 
∴ A>0 …(iii)
Q x + y ≥ 2 xy
x1 x
∴ A ≥ A1 + A 2 + A 2 + 2 A1 ≥ A1 + A 2 + 2 A1A 2
x2 x1
= ( A1 + A 2 )2 …(iv)
Equality holds only in case

x2 x x x
z1 = z2 1 and 1 A 2 = 2 A1 …(v)
x1 x2 x2 x1
www.pdfworld.in
www.pdfworld.in
Now, (i) can be written as

8 1 1 8A1A 2
≤ + or A ≥
A A1 A 2 A1 + A 2
Now, it is enough to show that
( A1 + A 2 )2 ≥ 8A1A 2
2
 A1 + A 2  A1 + A 2
i.e.,   ⋅ ≥ A1A 2
 2  2
Based on inequality between AM and GM, we have

2
 A1 + A 2 
  ≥ A1A 2
 2 
A1 + A 2
and ≥ A1A 2
2
Equality holds, if A1 = A 2
Together with (v) it gives x 12 = x 22
Consequently x 1 = x 2 is necessary for equality. Based on (v) these conditions imply z1 = z2
Now, A1 = A 2 forces y1 = y 2
But x 1 = x 2 , y1 = y 2 , z1 = z2 obviously imply equality in (i).
Example 19. Let a , b , c be real numbers such that a + b + c = 1. Prove that

a 2 + b 2 + c 2 ≥ 4 (ab + bc + ca ) − 1
when does equality hold?

Solution a 2 + b 2 + c 2 = (a + b + c )2 − 2 (ab + bc + ca ) = 1 − 2(ab + bc + ca )
So, it is enough to prove that
1 − 2(ab − bc − ca ) ≥ 4(ab + bc + ca ) − 1
i.e., 6(ab + bc + ca ) ≤ 2 or 3(ab + bc + ca ) ≤ 1
It happens only if
0 ≤ 1 − 3(ab + bc + ca )
= (a + b + c )2 − 3(ab + bc + ca )
(a − b )2 + (b − c )2 + (c − a )2
=
2
which is always true.
Clearly, equality holds when a = b = c = 1 / 3
Example 20. Let a , b , c be real numbers such that

0 < a < 1, 0 < b < 1, 0 < c < 1. Prove that
(i) ab + (1 − a )(1 − b ) ≤ 1 (ii) abc + (1 − a )(1 − b )(1 − c ) ≤ 1
Solution Taking a = sin 2 θ

b = sin 2 φ [Q0 < a < 1 and 0 < b < 1]
www.pdfworld.in
www.pdfworld.in
Inequalities 273
Then, ab + (1 − a )(1 − b )
= sin θ sin φ + cos θ cos φ

= cos (θ − φ) ≤ 1
Q 0 < c < 1, 0 < 1 − c < 1
We have
abc + (1 − a )(1 − b )(1 − c ) ≤ ab + (1 − a )(1 − b ) ≤ 1
Hence proved.
Example 21. Let a , b , c be +ve real numbers such that abc = 1. Prove that
ab bc ca
+ + ≤1
a5 + b5 + ab b5 + c 5 + bc c 5 + a5 + ca
when does equality hold?

Solution Now, a5 + b5 = (a + b )(a 4 − a3b + a 2b 2 − ab3 + b 4 )
= (a + b )[(a − b )2 (a 2 + ab + b 2 ) + a 2b 2 ] ≥ (a + b ) a 2b 2
if and only if a = b
ab ab
So, ≤
a5 + b5 + ab a 2b 2 (a + b ) + ab
1 1
= =
ab (a + b ) + 1 ab (a + b + c )
c
=
a+b+c
Now, writing Z for the sum Σ ab / (a5 + b5 + ab ), we get

c a b
Z≤ + + =1
a+b+c a+b+c a+b+c
Equality holds, if a = b = c = 1.
Example 22. Let a , b , c be sides of triangle while t is its area, show that a 2 + b 2 + c 2 ≥ 4t 3, when does
equality holds?
C
Solution Suppose that the largest angle of the ∆ ABC is at C. The foot of the
altitude m at C is T which is an inner point of the interval AB. Let x denote AT.
Now, apply formula cm = 2t and express a 2 and b 2 using the Pythagoras
a m b
theorem.
a 2 + b 2 + c 2 − 4t 3
= [m 2 + (c − x )2 ] + (m 2 + x 2 ) + c 2 − 2 3 cm
B c–x T x A
= 2c 2 + 2m 2 + 2x 2 − 2cx − 2 3 cm
1
= [(c − 2x )2 + (c 3 − 2m )2 ] ≥ 0
2
c c 3
Equality holds if x = and m = i.e., triangle is equilateral.
2 2
www.pdfworld.in
www.pdfworld.in
Aliter
Now, a 2 + b 2 + c 2 ≥ 4t 3 …(i)
(i) is equivalent to
(a 2 + b 2 + c 2 )2 ≥ 48t 2 = 3 ⋅ 16t 2
Apply an appropriate version of formula of Heron

16t 2 = − a 4 − b 4 − c 4 + 2a 2b 2 + 2b 2c 2 + 2c 2a 2
We only need to verify that

a 4 + b 4 + c 4 + 2a 2b 2 + 2b 2c 2 + 2c 2a 2 ≥ − 3a 4 − 3b 4 − 3c 4 + 6a 2b 2 + 6b 2c 2 + 6c 2a 2
which is equivalent to
(a 2 − b 2 )2 + (b 2 − c 2 )2 + (c 2 − a 2 )2 ≥ 0
equality holds if and only if a = b = c .
Aliter
Apply Hero’s formula again by AM–GM inequality, we have
3
 (s − a ) + (s − b ) + (s − c ) s3
(s − a )(s − b )(s − c ) ≤   =
 3  27
s4 s2
Hence, t = s (s − a ) (s − b ) (s − c ) ≤ =
27 3 3
Equality holds only for equilateral triangle.

This last inequality implies
2 2
a + b + c 
4t 3 ≤ 3   = 3 
2s

3  3 
a2 + b2 + c 2
≤3 = a2 + b2 + c 2
3
Equality holds only if a = b = c .
Aliter
Apply area formula
It is equal to ab sin θ and c 2 = a 2 + b 2 − 2ab cos θ
which is a direct consequence of the law of cosines
 3 1 
a 2 + b 2 + c 2 − 4t 3 = 2a 2 + 2b 2 − 4ab  sin θ + cos θ
 2 2 
= 2a 2 + 2b 2 − 4ab sin (θ + 30° ) ≥ 2a 2 + 2b 2 − 4ab
= 2(a − b )2 ≥ 0
Equality holds if a = b , θ = 60°.
Example 23. If a > 0, b > 0, then prove that for any x and y the following inequality holds true.
a ⋅ 2x + b ⋅ 3y + 1 ≤ (4x + 9y + 1) ⋅ a 2 + b 2 + 1
Solution By hypothesis both sides of this inequality is +ve.

On squaring, we get
⇒ (a ⋅ 2x + b ⋅ 3y + 1)2 ≤ (4x + 9y + 1) (a 2 + b 2 + 1)
www.pdfworld.in
www.pdfworld.in
Inequalities 275
⇒ a 2 4x + b 2 9y + 1 + 2ab2x3y + 2b ⋅ 3y + 2a ⋅ 2x
≤ 4x a 2 + 4x b 2 + 4x + a 2 9y + b 2 9y + 9y + a 2 + b 2 + 1
⇒ (a 2 9y − 2ab2x3y + 4x b 2 ) + (4x − 2a2x + a 2 ) + (9y − 2b3y + b 2 ) ≥ 0
⇒ (a3y − b2x )2 + (2x − a )2 + (3y − b )2 ≥ 0
So, original inequality is true.
Example 24. There are real numbers a , b , c such that a ≥ b ≥ c . Prove that
a2 − b2 c 2 − b2 a2 − c 2
+ + ≥ 3a − 4b + c.
c a b
Solution From a ≥ b ≥ c > 0, we have
a+b b+c
≥ 2, 0 < ≤2
c a
a+c
and ≥1
b
a2 − b2
Now, we get ≥ 2(a − b ) [Qa ≥ b ]
c
c2 − b2
≥ 2(c − b ) [Qc ≤ b ]
a
a2 − c 2
≥ (a − c ) [Qa ≥ c ]
a
After addition of these inequalities, we get
a2 − b2 c 2 − b2 a2 − c 2
+ + ≥ 2 (a − b ) + 2 (c − b ) + (a − c )
c a b
a2 − b2 c 2 − b2 a2 − c 2
i.e., + + ≥ 3a − 4b + c
c a b
Equality holds if a = b = c > 0.
Example 25. Prove that if the number x 1 and x 2 does not exceed 1 in absolute value, then
2
x + x2 
1 − x 12 + 1 − x 22 ≤ 2 1 −  1  .
 2 
For what number x 1 and x 2 does the equality holds?
Solution Both numbers of the inequality are +ve. On squaring both sides of given inequation, we get
1 − x 12 + 1 − x 22 + 2 (1 − x 12 )(1 − x 22 ) ≤ 4 − (x 12 + 2x 1x 2 + x 22 )
i.e., 2 (1 − x 12 )(1 − x 22 ) ≤ 2 − 2x 1x 2
⇒ (1 − x 12 )(1 − x 22 ) ≤ 1 − x 1x 2
Again squaring both sides

1 − x 12 − x 22 + x 12x 22 ≤ 1 − 2x 1x 2 + x 12x 22
If all transposed to RHS, we have
0 ≤ (x 1 − x 2 )2
Equality holds if x 1 = x 2.
www.pdfworld.in
www.pdfworld.in
Example 26. If a , b , c are real numbers such that a 2 + b 2 + c 2 = 1.

Prove that −1 / 2 ≤ ab + bc + ca ≤ 1.
Solution Q a2 + b2 + c 2 = 1
1 2
then − (a + b 2 + c 2 ) ≤ ab + bc + ca ≤ a 2 + b 2 + c 2
2
or − (a 2 + b 2 + c 2 ) ≤ 2(ab + bc + ca ) ≤ 2(a 2 + b 2 + c 2 )
These inequalities are indeed true.
Q 2 (ab + bc + ca ) + (a 2 + b 2 + c 2 )
= (a + b + c )2 ≥ 0
and 2 (a 2 + b 2 + c 2 ) − 2(ab + bc + ca )
= (a − b )2 + (a − c )2 + (b − c )2 ≥ 0
Example 27. Show that (n !)2 > nn , if n is a +ve integer.

Solution If r is a +ve integer lying between 1 and n, then we know that
r (n − r ) > (n − r )
or r (n − r ) + r > n or r (n − r + 1) > n
Putting r = 1, 2, 3, K n, we have
1(n ) = n
2(n − 1) > n
3(n − 2) > n
4(n − 3) > n
…………
…………
…………
(n − 1)(2) > n
n (1) = n
Multiplying the above n rows, we get
[1 ⋅ 2 ⋅ 3 K (n − 1) n ][n (n − 1)(n − 2) K 2] > nn
i . e. , n ! × n ! > nn or (n !)2 > nn
Hence proved.

1 1 1 1
+ + K+ >
n+1 n+2 2n 2
where n is a +ve integer.
Solution We have
1 1 1 1 1 1 1 1
= , > ,K , > , >
2n 2n 2n − 1 2n n + 2 2n n + 1 2n
Adding these inequalities termwise, we find
1 1 1 1 1 1
+ + K+ > + + K+
n+1 n+2 2n 2n 2n 2n
n 1
= =
2n 2
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 277
Example 29. Prove that the sum of any number of fractions taken from among the sequence
1 1 1
, , , K is always less than unity.
22 32 42
Solution Let us have n fractions (n ≥ 1)
1 1 1 1 1 1
, , , ,K, ,
a b c d k l
Let us assume 2≤ a < b < c < d < …… < k < l
Then b ≥ a + 1, c ≥ b + 1, d ≥ c + 1, K , l ≥ k + 1
Consequently
b ≥ a + 1, c ≥ a + 2, d ≥ a + 3,…, l ≥ a + n − 1
1 1 1 1 1 1 1 1
∴ + 2 + K+ 2 ≤ 2 + + K+ < −
a 2
b l a (a + 1)2
(a + n − 1)2
a − 1 a + n −1
1 1 1 n
Hence, + + K+ <
a2 b2 l2 (a − 1)(a + n − 1)
But a − 1 ≥ 1, a + n − 1 ≥ n + 1
1 1 1 n
(a − 1)(a + n − 1) ≥ n + 1 and + + K+ 2 ≤ <1
a2 b2 l n+1
Hence proved.

n 1 2 3 n
< + + + K+ <n
2 2 3 4 n+1
1 1
Solution We have, =
2 2
1 2
<
2 3
1 3
<
2 4
……
……
……
1 n
<
2 n+1
Adding these inequalities we get
n 1 2 3 n
< + + + K+ …(i)
2 2 3 4 n+1
1 2 3 n
Again, < 1, < 1, < 1, K <1
2 3 4 n+1
Adding all these inequalities, we get
1 2 3 n
+ + + K+ <n …(ii)
2 3 4 n+1
n 1 2 3 n
< + + + K+ <n
2 2 3 4 n+1
www.pdfworld.in
www.pdfworld.in
Example 31. If n ∈ N , prove that

n 1 1 1 1
<1+ + + + K+ n < n.
2 2 3 4 2 −1
Solution We have,
1 1 1 1
S =1+ + + + K+ n
2 3 4 2 −1
=  1 +  +  +  +  + + +  +  +
1 1 1 1 1 1 1 1 1 1
+ K+ 
 2   3 4   5 6 7 8  9 10 16 
 1 1 1 
+ K +  n − 1 + n −1 + K+ n 
2 +1 2 +2 2 − 1 
 1 1  1 1  1 1
=  0 + 1  +  1
1 1
+  +  + K + 3  +  3 + K + 4 
2 2   2 + 1 22   22 + 1 2  2 + 1 2 
 1 1 1 
+ K +  n − 1 + n −1 + K+ n 
2 +1 2 +2 2 − 1 
1 1 1
Now, + >
20 21 2
1 1 2
+ >
21 + 1 22 22
1 1 1 22
+ + K+ >
2 +1
2
2 +22
23
23
………………………………
………………………………
………………………………
1 1 2n − 1
−1
+ K+ >
2n +1 2n − 1 2n
n
Adding these inequalities, we get S >
2
1 1 1 1
Again, S =1+ + + + K+ n
2 3 4 2 −1
= 1 +  +  +  + + +  +  + + K +
1 1 1 1 1 1 1 1 1  1 1 
 + K+  n −1 + K + n −1
 2 3  4 5 6 7  8 9 15  2 2 
1 1   1 1 
= 1 +  1 + 2  +  + K+ 3 
2 2 − 1   22 2 − 1 
1 1   1 1 
+  3 + K + 4  + K +  n −1 + K + n 
2 2 − 1  2 2 − 1
Now, 1=1
1 1
+ K+ <1
21 22 − 1
1 1
+ K+ <1
22 23 − 1
1 1
+ K+ <1
23 24 − 1
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 279
Example 32. Let m and n be +ve integers. Prove that

1 1 1 1 1 1 1
− < + +K + < − .
n + 1 n + m + 1 (n + 1)2
(n + 2)2
(n + m )2
n n+m
Solution For any +ve integer, r satisfying 0 < r < m , we have n + r − 1 < n + r < n + r + 1
⇒ (n + r )(n + r − 1) < (n + r )2 < (n + r )(n + r + 1)
1 1 1
⇒ > >
(n + r )(n + r − 1) (n + r )2 (n + r )(n + r + 1)
1 1 1 1 1
⇒ − > > −
(n + r − 1) (n + r ) (n + r )2 (n + r ) (n + r + 1)
1 1 1 1 1
⇒ − < < −
n + r n + r + 1 (n + r )2 n + r − 1 n + r
Putting r = 1, 2, 3, K , m, we get
1 1 1 1 1
− < < −
n + 1 n + 2 (n + 1)2 n n + 1
1 1 1 1 1
− < < −
n + 2 n + 3 (n + 2)2 (n + 1) (n + 2)
……………………………………………
……………………………………………
……………………………………………
1 1 1 1 1
Similarly, − < < −
n + m n + m + 1 (n + m )2 n + m − 1 n + m
Adding the above inequalities, we get
1 1 1 1 1 1 1
− < + + K+ < −
n + 1 n + m + 1 (n + 1)2 (n + 2)2 (n + m )2 n n + m
Hence proved.
Example 33. Determine the largest number in the infinite sequence 1, 2 , 3 3 , 4 4 ,... , n n .
Solution We find that 31/3 to be the largest.
So, we will prove that (n1/n ), n ≥ 3 is a decreasing sequence.
1
+1
n1/n > (n + 1)n
+1
nn > (n + 1)n
n
n >  1 + 
1
 n
n
1 + 1  1 n (n − 1) 1 n (n − 1)(n − 2) 1
Now,   =1+ n⋅ + ⋅ 2 + ⋅ 3 +K
 n n 2 n 6 n
1 1 1  1  2 1 1 1
=1+ 1+  1 −  +  1 −   1 −  + K< 1 + 1 + + + + K < 3
2 n 6  n  n 2 4 8
n
3 >  1 + 
1
or
 n
1
+1
∴ If n ≥ 3, n1/n > (n + 1)n
www.pdfworld.in
www.pdfworld.in
i.e., (n1/n ) is decreasing for n ≥ 3

But 31/3 is also greater than 1 and 21/ 2.
1/3
Therefore, 3 is the largest.
Example 34. For all n ∈ N . Prove that

1 1 1
2 n − 2< 1 + + + K+ <2 n −1
2 3 n
Solution Now, first we will prove that
1
2 p + 1 −2 p < < 2 p − 2 p − 1, ∀ p ∈ N
p
We have 2 p + 1 − 2 p = 2( p + 1 − p )
2 ( p + 1 − p )( p + 1 + p)
=
( p+1+ p)
 
Q 1 p + 1 > p 
2 2  
= < ⇒ p + 1 + p > p + p 
p+1+ p p + p  
⇒ 1 1 
<
 p+1+ p p + p 
1
∴ 2 p + 1 −2 p < …(i)
p
2( p − p − 1 )( p + p − 1)
and 2 p −2 p −1=
( p + p − 1)
Q p − 1 > p ∴ p + p − 1 ⇒ 1 1 
p + p −1 p + p >
 p + p − 1 p + p 
 
1
∴ 2 p −2 p −1> …(ii)
p
1
2 p + 1 −2 p < <2 p −2 p −1 …(iii)
p
∀p ∈ N
Putting p = 2, 3, K , n in (iii), we have
1
2 3 −2 2< <2 2 −2 1
2
1
2 4 −2 3< <2 3 −2 2
3
1
2 5 −2 4< <2 4 −2 3
4
………………………………
………………………………
………………………………
1
2 n −2 n −1< <2 n −1 −2 n −2
n −1
www.pdfworld.in
www.pdfworld.in
Inequalities 281
1
2 n + 1 −2 n < <2 n −2 n −1
n
Adding all inequalities, we get
1 1 1 1
2 n + 1 −2 2< + + + K+ <2 n −2 1
2 3 4 n
Adding 1 throughout, we have
1 1 1
2 n + 1 − 2 2 + 1< 1 + + + K+ <2 n −1 …(iv)
2 3 n
But we have
[2( n ) − 2] − (2 n + 1 − 2 2 + 1)
= 2( n − n + 1 ) − 3 + 2 2
= 2( n − n + 1 ) + (2 2 − 3) < 0 [Q n < n + 1 and 2 2 − 3 < 0]
∴ 2 n − 2< 2 n + 1 − 2 2 + 1 …(v)
From (iv) and (v), we get
1 1 1 1
2 n − 2< 1 + + + + K+ <2 n −1
2 3 4 n
Example 35. Let a > 1 and n be a +ve integer.

 n +1 n −1
Prove that an − 1 ≥ n  a 2 −a 2 .
 
 
Solution Let a = α 2
It is required to prove that
+1 −1
α 2n − 1 ≥ n (αn − αn )
or which is the same.
−1
α 2n − 1 ≥ n αn (α 2 − 1)
α 2n − 1 −1
≥ n αn
α2 − 1
α 2n − 1 − 1) − 2)
But = α 2(n + α 2(n + K + α2 + 1
α2 − 1
−2
≥ n n α 2 ⋅ α 4 K α 2n
(Using theorem on AM and GM of several numbers)

Q 2 + 4 + K + (2n − 2) = n (n − 1)
α 2n − 1 −1
we have ≥ n αn
α2 − 1
 n +1 n −1
∴ an − 1 ≥ n  a 2 −a 2 
 
 
Hence proved.
www.pdfworld.in
www.pdfworld.in
Example 36. Let s > 1 and n be a +ve integer. Prove that

1 1 1 1
+ s + K+ s < .
1s 2 n 1 − 21 − s
Solution 1s + 1s + K + 1s < 1 + 1s + K + 1s + 1
2 n s 2 n

= s +  s + s  +  s + s + s + s  + K< s + s + s + s + K
1 1 1  1 1 1 1 1 2 4 8
1 2 3  4 5 6 7  1 2 4 8
1 1 1 1 1
= s −1 + s −1 + s −1 + s −1 + K=
1 2 4 8 1 − 21 − s
Example 37. If a , b , c are +ve real numbers, prove that

6abc ≤ a 2 (b + c ) + b 2 (c + a ) + c 2 (a + b )≤ 2(a3 + b3 + c 3 )
Solution We have a 2 + b 2 > 2ab

⇒ a 2 + b 2 − ab > ab
⇒ (a + b )(a 2 + b 2 − ab ) > ab (a + b )
⇒ a3 + b3 > ab (a + b )
Similarly, b3 + c 3 > bc (b + c )
c 3 + a3 > ca (c + a )
Adding all these, we get
2(a3 + b3 + c 3 ) > ab (a + b ) + bc (b + c ) + ca (c + a )
Again, AM > GM
ab (a + b ) + bc (b + c ) + ca (c + a )
⇒ > (a 2b ⋅ ab 2 ⋅ b 2c ⋅ bc 2 ⋅ c 2a ⋅ ca 2 )1/ 6
6
⇒ ab (a + b ) + bc (b + c ) + ca (c + a ) > 6abc
Hence,
6abc ≤ a 2 (b + c ) + b 2 (c + a ) + c 2 (a + b )≤ 2(a3 + b3 + c 3 )
Example 38. If x + y + z = 1 and x , y , z > 0, then show that

 1 + 1   1 + 1   1 + 1  ≥ 64.
    
 x  y  z
Solution Consider LHS

Open the brackets
Let
 1 1 1  1 1 1  1
S =1+  + +  +  + +  +
x y z   xy yz zx  xyz
 x + y + z  ≥ (xyz )1/3
We know  
 3 
x + y + z ≥ 3(xyz )1/3
1 ≥ 3(xyz )1/3
1
⇒ (xyz )1/3 ≤ …(i)
3
www.pdfworld.in
www.pdfworld.in
Inequalities 283
1
⇒ xyz ≤ …(ii)
27
1
or ≥ 27 …(iii)
xyz
2
or ≥ 54 …(iv)
xyz
1 1 1
+ + 1/3
x y z ≥ 1 
Also,  
3  xyz 
1/3
1 1 1  1 
+ + ≥ 3 
x y z  xyz 
1 1 1
+ + ≥9 [from (iii)] …(v)
x y z
Now, (v) becomes
 1 1 1 x + y + z 1
1+  + +  +   +
x y z   xyz  xyz
 1 1 1 2
=1 +  + +  + [Qx + y + z = 1]
x y z  xyz
Adding (iv) and (v), we get
1 1 1 2
+ + + ≥ 63
x y z xyz
Adding 1 on both sides, we get
1 1 1 2
1+ + + + ≥ 64
x y z xyz
 1 + 1   1 + 1   1 + 1  ≥ 64
or    
 x y  z

xyz > ( y + z − x ) (z + x − y ) (x + y − z ).
Solution Consider two quantities ( y + z − x ) and (z + x − y ).
Q AM > GM
( y + z − x ) + (z + x − y )
∴ > ( y + z − x )(z + x − y )
2
or z> ( y + z − x )(z + x − y )
Similarly, y> ( y + z − x )(x + y − z )
and x> ( x + y − z )(x + z − y )
Multiplying these three results, we get
xyz > ( y + z − x )2 (z + x − y )2 (x + y − z )2
xyz > ( y + z − x )(z + x − y )(x + y + z )

Hence proved.
www.pdfworld.in
www.pdfworld.in
Example 40. If a + b + c + d = s , show that (s − a )(s − b ) (s − c ) (s – d ) > 81 abcd.

Solution Qa + b + c + d = s
∴ s −a =b + c + d
b+c+d
As AM > GM ∴ > (bcd )1/3
3
or (b + c + d ) > 3(bcd )1/3 or (s − a ) > 3(bcd )1/3 …(i)
(s − b ) > 3(cad )1/3
…(ii)
(s − c ) > 3(abd )1/3 …(iii)
(s − d ) > 3(abc )1/3 …(iv)
Multiplying (i), (ii), (iii) and (iv), we get
(s − a )(s − b )(s − c )(s − d ) > 81 (bcd )1/3 (cad )1/3 (abd )1/3 (abc )1/3
or (s − a )(s − b )(s − c ) > 81(a3b3c 3d 3 )1/3
or (s − a )(s − b )(s − c )(s − d ) > 81abcd
Hence proved.

2 2 2 9
+ + > .
b+c c+a a+b a+b+c
Solution Q AM > GM
1  2 2 2   2 2 2 
∴  + + > × ×  …(i)
3 b + c c + a a + b  b + c c + a a + b 
1 b + c c + a a + b  b + c c + a a + b 
Also,  + + > × ×  …(ii)
3 2 2 2   2 2 2 
Multiplying (i) and (ii), we have
1 2 2 2 
 + + 
9 b + c c + a a + b 
 (b + c ) + (c + a ) + (a + b )
 >1
 2 
 2 2 2   2a + 2b + 2c 
or  + +   >9
b + c c + a a + b   2 
 2 2 2 
or  + +  (a + b + c ) > 9
b + c c + a a + b 
2 2 2 9
or + + >
b+c c+a a+b a+b+c
Hence proved.
Example 42. If ai > 0, ∀ i = 1, 2,…, n and s = a1 + a2 + K + an , then prove that

s s s s n2
+ + + K+ > .
s − a1 s − a 2 s − a3 s − an n − 1
Solution QAM > GM we have

(s − a1 ) + (s − a 2 ) + K + (s − an )
> [(s − a1 )(s − a 2 ) K (s − an )]1/n
n
www.pdfworld.in
www.pdfworld.in
Inequalities 285
1 1 1
+ + K+ 1/n
s − a1 s − a 2 s − an  1 
and > 
n  (s − a1 )(s − a 2 ) .. (s − an )
ns − (a1 + a 2 + K + an )
⇒ > [(s − a1 )(s − a 2 ) K (s − an )]1/n
n
1 1 1 1  1
and  + + K+ >
n  s − a1 s − a 2 s − an  [(s − a1 )(s − a 2 ) K (s − an )]1/n
(n − 1)s
⇒ > [(s − a1 )(s − a 2 ) K (s − an )]1/n
n
1 1 1 1  1
and  + + K+ >
n  s − a1 s − a 2 s − an  [(s − a1 )(s − a 2 ) K (s − an )]1/n
⇒  n − 1 s × 1  1 + 1
+ K+
1 
    >1
 n  n  s − a1 s − a 2 s − an 
⇒  n − 1  s +
s
+ K+
s 
 2   >1
 n   s − a1 s − a 2 s − an 
s s s n2
⇒ + + K+ >
s − a1 s − a 2 s − an n − 1
Example 43. If a , b , c are +ve real numbers representing the sides of a triangle, prove that
ab + bc + ca < a 2 + b 2 + c 2 < 2(ab + bc + ca )
a2 + b2 + c 2
or 1< <2
ab + bc + ca
Hence prove that
3(ab + bc + ca ) < (a + b + c )2 < 4(ab + bc + ca )
(a + b + c )2
or 3< < 4.
ab + bc + ca
a2 + b2 b2 + c2
Solution > ab , > bc
2 2
c 2 + a2
and > ca [Q AM > GM]
2
⇒ a 2 + b 2 > 2ab , b 2 + c 2 > 2bc and c 2 + a 2 > 2ca
⇒ a 2 + b 2 + b 2 + c 2 + c 2 + a 2 > 2(ab + bc + ca )
⇒ a 2 + b 2 + c 2 > ab + bc + ca
⇒ ab + bc + ca < a 2 + b 2 + c 2 …(i)
In a ∆ ABC with sides BC = a , CA = b , AB = c , we have b 2 + c 2 − a 2 = 2bc cos A
⇒ b 2 + c 2 − a 2 < 2bc [Qcos A < 1]
Similarly, we have c + a − b < 2ca
2 2 2
and a + b − c < 2ab
2 2 2
Adding these three, we get

a 2 + b 2 + c 2 < 2(ab + bc + ca ) …(ii)
www.pdfworld.in
www.pdfworld.in

ab + bc + ca < a 2 + b 2 + c 2 < 2(ab + bc + ca ) …(iii)
a +b +c
2 2 2
⇒ 1< <2
ab + bc + ca
Adding 2(ab + bc + ca ) throughout in (iii)
3(ab + bc + ca ) < (a + b + c )2 < 4(ab + bc + ca )
(a + b + c )2
3< <4
ab + bc + ca
Hence proved.
Example 44. If a , b , c are +ve real numbers, prove

a3 + b 3 + c 3  a + b + c   a 2 + b 2 + c 2 
>    .
9  3  3 
a2 + b2
Solution > ab [Q AM > GM]
2
⇒ a 2 + b 2 > 2ab
⇒ a 2 + b 2 − ab > ab
⇒ (a 2 + b 2 − ab )(a + b ) > ab (a + b )
⇒ a3 + b3 > ab (a + b )
Similarly, we have b3 + c 3 > bc (b + c )
c 3 + a3 > ca (c + a )
Adding all these, we get
2(a3 + b3 + c 3 ) > ab (a + b ) + bc (b + c ) + ca (c + a )
Adding a3 + b3 + c 3 on both sides, we get
3(a3 + b3 + c 3 ) > (a3 + b3 + c 3 ) + ab (a + b ) + bc (b + c ) + ca (c + a )

⇒ 3(a3 + b3 + c 3 ) > (a + b + c )(a 2 + b 2 + c 2 )
a3 + b 3 + c 3  a + b + c   a 2 + b 2 + c 2 
⇒ >   
9  3  3 
Hence proved.
Example 45. If a , b , c , d are four +ve real numbers such that abcd = 1, prove that
(1 + a )(1 + b )(1 + c )(1 + d ) ≥ 16.
Solution Q AM > GM
1+a 1+b
∴ ≥ 1⋅a, ≥ 1⋅b
2 2
1+c 1+d
≥ 1⋅c, ≥ 1⋅d
2 2
Multiplying corresponding sides of the above inequalities, we have
(1 + a )(1 + b )(1 + c )(1 + d ) ≥ 16 abcd ≥ 16 [Qabcd = 1]
www.pdfworld.in
www.pdfworld.in
Inequalities 287
Example 46. If a , b , c are +ve real numbers such that a + b + c = 1. Prove that
(1 + a )(1 + b )(1 + c ) ≥ 81
( − a )(1 − b ) (1 − c ).
Solution Q a + b + c =1
The given inequality is equivalent to
(2a + b + c )(a + 2b + c )(a + b + 2c ) ≥ 8(b + c )(c + a )(a + b )
b+c c+a a+b
If we write = x, = y, =z
2 2 2
b+c c+a a+b
Then x +y + z= + +
2 2 2
=a + b + c =1
x ,y, z > 0
and given inequality becomes
(y + z )(z + x )(x + y ) ≥ 8xyz …(i)
In order to prove the given inequality, it is enough to prove (i)
Q x , y , z are all +ve.
y + z z+x
∴ ≥ (yz )1/ 2, ≥ (zx )1/ 2,
2 2
x +y
≥ (xy )1/ 2
2
By the inequality of the means (AM ≥ GM)
Now, (y + z )(z + x )(x + y ) ≥ [2(yz )1/ 2 ] ⋅ [2(zx )1/ 2 ] ⋅ [2(xy )1/ 2 ]
= 8xyz
Example 47. If a , b , c are real numbers such that 0 < a < 1, 0 < b < 1, 0 < c < 1, a + b + c = 2.
a b c
Prove that ⋅ ⋅ ≥ 8.
1 − a 1−b 1 − c
Solution Put a = y + z , b = z + x , c = x + y
so that x + y + z =1
Q x = 1 − a, y = 1 − b, z = 1 − c
and 0 < a < 1, 0 0
a b c
Now, ⋅ ⋅
1−a 1−b 1−c
[(y + z )(z + x )(x + y )]
= …(i)
xyz
By the inequality of the means
y + z z+x x +y
≥ yz ; ≥ zx ; ≥ xy
2 2 2
It follows that
[(y + z )(z + x )(x + y )]
≥8 …(ii)
(xyz )
a b c
⋅ ⋅ ≥8
1−a 1−b 1−c
www.pdfworld.in
www.pdfworld.in
a b c
Example 48. Let a , b , c be the sides of a triangle, show that + + must lie between the
b+c c+a a+b
limits 3/2 and 2. Can equality hold at either limit?
Solution Q a , b , c denotes the sides of a triangle
∴ 0 < a < b + c , 0 < b < c + a, 0 < c < a + b
a b c
∴ 0< < 1, 0 < < 1, 0 < <1
b+c c+a a+b
x
Now, we know that if x , y are +ve real numbers such that < 1 and z be any +ve real number, then
y
x x + z  x + z x z (y − x ) 
<  for − = > 0
y y + z  y + z y y (y + z ) 
Using the fact, we find that
a a+a a 2a
< so that <
b+c b+c+a b+c a+b+c
b 2b
Similarly, <
c+a a+b+c
c 2c
and <
a+b a+b+c
Adding the above inequalities, we have

a b c 2a + 2b + 2c
+ + <
b+c c+a a+b a+b+c
a b c
so that + + <2 …(i)
b + c c+a a + b
QAM ≥ GM to the +ve numbers b + c , c + a , a + b, we have

1
[(b + c ) + (c + a ) + (a + b )] ≥ [(b + c )(c + a )(a + b )]1/3 …(ii)
3
1 1 1
Q AM ≥ GM to the +ve numbers , , we have
b+c c+a a+b
1/3
1 1 1 1   1 1 1 
 + + ≥ ⋅ ⋅  …(iii)
3 b + c c + a a + b  b + c c + a a + b 
Multiplying (ii) and (iii), we get

2  1 1 1 
(a + b + c )  + + ≥1
9 b + c c + a a + b 
 1 1 1  9
or (a + b + c )  + + ≥
b + c c + a a + b  2
a b c 3
or + + ≥ …(iv)
b+c c+a a+b 2
QThe inequalities in (ii) and (iii) are both equalities when a = b = c .∴Inequality in (iv) is an quality when
a , b , c are all equal.
www.pdfworld.in
www.pdfworld.in
Inequalities 289
Example 49. If a , b , c are +ve distinct integers, show that (a + b + c )2 > abc
[(ab )1/ 4 + (bc )1/ 4 + (ca )1/ 4 ]
Solution Q a 2 + b 2 > 2ab [AM > GM]

b + c > 2bc , c + a > 2ca
2 2 2 2
(unequal numbers)
Adding these three inequation, we get
a 2 + b 2 + c 2 > ab + bc + ca …(i)
We have, also ab + bc > 2b ac
Similarly, bc + ca > 2c ab
and ac + ab > 2a bc
Adding these three inequation
ab + bc + ca > a bc + b ca + c ab
= abc [ a + b + c] …(ii)
Also, a + b > 2(ab ) 1/ 4
b + c > 2(bc )1/ 4 , c + a > 2(ca )1/ 4

Adding these three relations
a + b + c > (ab )1/ 4 + (bc )1/ 4 + (ca )1/ 4 …(iii)
a 2 + b 2 + c 2 > abc [(ab )1/ 4 + (bc )1/ 4 + (ca )1/ 4 ]
Q (a + b + c )2 > a 2 + b 2 + c 2
∴ (a + b + c )2 > abc
[(ab )1/. 4 + (bc )1/ 4 + (ca )1/ 4 ]
Hence proved.
Example 50. If a , b , c are the sides of a triangle. Prove that

a b c
+ + ≥ 3.
c + a −b a + b −c b + c −a
Solution a , b , c are sides of triangle.

We have a , b , c are all > 0
Also, c + a − b , a + b − c , b + c − a are all > 0.
By AM–GM inequality
1/3
a b c  abc 
+ + ≥3  …(i)
c + a −b a + b −c b + c −a  (c + a − b )(a + b − c )(b + c − a ) 
We have a 2 ≥ a 2 − (b − c )2
⇒ a 2 ≥ (a + b − c )(a − b + c ) …(ii)
Similarly, b ≥ (b + c − a )(b − c + a )
2
…(iii)
and c ≥ (c + a − b )(c − a + b )
2
…(iv)
Multiplying (ii), (iii) and (iv), we get
a 2b 2c 2 ≥ (a + b − c )2(b + c − a )2 (c + a − b )2 …(v)
www.pdfworld.in
www.pdfworld.in
Taking +ve square root of (v), we get

abc ≥ (a + b − c )(b + c − a )(c + a − b )
abc
⇒ ≥1
(a + b − c )(b + c − a )(c + a − b )
1/3
 abc 
⇒   ≥1
 (a + b − c )(b + c − a )(c + a − b ) 
So, (i) implies that
a b c
+ + ≥3
c + a −b a + b −c b + c −a
kabc
Example 51. Find the largest constant k such that ≤ (a + b )2 + (a + b + 4c )2, ∀ a , b , c > 0.
a+b+c
Solution By AM–GM inequality

(a + b )2 + (a + b + 4c )2
= (a + b )2 + (a + 2c + b + 2c )2 ≥ (2 ab 2 ) + (2 2ac + 2 2bc 2 )
= 4ab + 8ac + 8bc + 16c ab
(a + b )2 + (a + b + 4c )2 4ab + 8ac + 8bc + 16c ab
∴ (a + b + c ) ≥ (a + b + c )
abc abc
4 8 8 16 
= + + +  (a + b + c )
c b a ab 
1 1 1 1 1 
= 8 + + + + 
 2c b a ab ab 
a a b b   1   2 2 
 + + + + c  ≥ 8 5 5   5 5 a b c  = 100
2 2 2 2   2a b c 
2 2  24 

Again, by AM – GM inequality.
Hence, largest constant k is 100.
For k = 100 equality holds, if a = b = 2c > 0.
Example 52. If a , b , c , K , k are +ve quantities.

a +b +c +K+ k
a + b + c + K + k
Prove that   < a ab b c c K k k .
 n 
Solution Consider a quantities each equal to (1/a), b quantities each equal to (1/b) ………… and k
quantities each equal to (1/k).
As AM > GM
+  + + K to k terms
1 1
 1 + 1 + K to a terms +  1 + 1 + K to b terms k k 
   
a a  b b  a + b + K+ k
1/ ( a + b + K + k )
 1 1 
>   ⋅ K to a factors  ⋅ K to b factors … ⋅ to k factors 
1 1 1 1
  a a   b b   k k  
www.pdfworld.in
www.pdfworld.in
Inequalities 291
1 1  1  1
 ⋅ a ) +  ⋅ b  + K +  ⋅ k 
1  (a + b + K + k)
a b k
a b  k 
>    ⋅   K   
1 1
or
a + b + c + K+ k   a  b  k 

1
1+ 1+ K+ 1  1  (a + b + K + k)
or > 
a + b + K + k  a ab b K k k 
1
n  1  (a + b + K + k)
or > 
a + b + c + K + k  a abb K k k 
a +b +c +K+ k
 n   1 
or   >  a b 
k
a + b + c + K + k  a b K k 
[Raising both sides to power (a + b + c + K + k )]
a +b +K+ k
a + b + c + K + k
or   < (a abb K k k )
 n 
p +q
 
Example 53. Prove that apbq >  ap + bq  .
 p+q 
Solution Consider p quantities each equal to a and q quantities each equal to b.

Q AM > GM
(a + a + K to p terms) + (b + b + K to q terms)
∴ > [(a × a .. . to p factors)
p+q
1
p +q
(b × b… to q factors) ]
1
ap + bq
or > (apbq )p + q
p+q
Raising both sides to power p + q, we get
p +q
 ap + bq 
  > (apbq )
 p+q 
Hence proved.
Example 54. Prove that unless p = q = r or x = 1,

−r −p
px q + qx r + rx p − q > p + q + r.
−r
Solution Consider p quantities each equal to x q
r −p
q quantities each equal to x
r quantities each equal to x p − q
Q AM > GM
q −r q −r
(x +x ... to p terms) + (x r − p + x r − p ...
to q terms) + (x p –q + x p –q K to r terms) −r −r −p −p
∴ > [(x q ⋅ xq … to p factors) (x r ⋅ xr …
p+q+r
1
p −q p −q p +q +r
to q factors) (x x K to r factors)]
−r −p
or px q + qx r + rx p − q > [x p (q −r )
× x q (r − p)
× x r (p − q ) ]1/ (p + q +r)
www.pdfworld.in
www.pdfworld.in
1
−r
rx q + qx r − p + rx p − q − pr − qp + rp − rq p + q + r
or > [x pq ]
p+q+r
1
−r
px q + qx r − p + rx p − q
or > (x 0 )p + q +r
p+q+r
−r
px q + qx r − p + rx p − q
or >1
p+q+r
−r −p
or px q + qx r + rx p − q > p + q + r
2n
Example 55. Prove that (n !)3 < nn  n + 1  .
 2 
Solution Q AM > GM
1 + 2 + 3 + K + n3
3 3 3
∴ > (13 ⋅ 23 ⋅ 33 K n3 )1/n
n
n
 13 + 23 + 33 + K + n3 
or   > (1 ⋅ 2 ⋅ 3 K n )3 …(i)
 n 
2
n (n + 1)
Now, 13 + 23 + 33 + K + n3 = 
 2 
n 2 (n + 1)2
=
4
Substituting this value of 13 + 23 + K + n3 in (i), we get
n n
 n 2 (n + 1)2  
n  n + 1
2
 > (n !) or n     > (n !)
3 3
 
 4n  
 2 

2n n
n + 1 n + 1
or nn   > (n !)3 or (n !)3 < nn  
 2   2 
−1
Example 56. Prove that 2n > 1 + n sn
−1
Solution Consider n quantities 1, 2, 22 , K , 2n . Then as AM > GM
n −1
1 + 2 + 2 + ...... + 2
2
− 1 1/n
> (1 ⋅ 2 ⋅ 22 ...... 2n )
n
1  1(2n − 1) 1+ 2 + 3 + (n − 1) 1/n
∴   > [2 ] (QLHS is sum of GP)
n  (2 − 1) 
2n − 1 − 1)n / 2 1/n Q1 + 2 + 3 + K + n = 1 n (n + 1)

or > [2(n ]
n  2 
2n − 1 − 1)/ 2
or > 2(n
n
2n − 1 −1
or > 2n
n
−1
or 2n − 1 > n 2n
−1
or 2n > 1 + n 2n
www.pdfworld.in
www.pdfworld.in
Inequalities 293
Example 57. If n > 1. Prove that

−1 −1
1 + 2 + 22 + 23 + K + 2n > n 2n .
Solution Consider n quantities,
−1
1, 2, 22 , 23 , K , 2n
Then as AM > GM
 1 + 2 + 22 + 23 + K + 2n −1
− 1 1/n
∴   > (1 ⋅ 2 ⋅ 22 ⋅ 23 K 2n )
 n 
−1
or (1 + 2 + 22 + 23 + K + 2n > n [21 + 2 + 3 + K + (n − 1) 1/n
] ) …(i)
1
Now, 1 + 2 + 3 + K + (n − 1) = (n − 1)
2
1
[2(1) + (n − 2)] = (n − 1)n
2
∴From (i), we get
−1 − 1)/ 2 1/n
1 + 2 + 22 + K + 2n > n[2n (n ]
n −1 (n − 1)/ 2
⇒ 1+ 2+ 2 + K+ 2
2
> n[2 ]
n −1 n −1
⇒ 1+ 2+ 2 + K+ 2
2
>n 2
Hence proved.

(n + 1)
2n + 1 
1 ⋅ 2 ⋅ 3 ⋅ 4 K n < 
2 3 4 n 2
 .
 3 
Solution Q 1 ⋅ 22 ⋅ 33 ⋅ 44 K nn
= 1 ⋅ (2 ⋅ 2) ⋅ (3 ⋅ 3 ⋅ 3) ⋅ (4 ⋅ 4 ⋅ 4 ⋅ 4) K (n ⋅ n ⋅ n ⋅ K n )
Total number = 1 + 2 + 3 + 4 + K + n
n (n + 1)
= Σn =
2
Q AM > GM
1 + (2 + 2) + (3 + 3 + 3) + (4 + 4 + 4 + 4) 1
K + (n + n + n ... n times) + 1)/ 2
∴ > (1 ⋅ 22 ⋅ 33 ⋅ 44 K nn )n (n
n (n + 1)
2
2
12 + 22 + 32 + 42 + K + n 2 + 1)
⇒ > (1 ⋅ 22 ⋅ 33 ⋅ 44 K nn )n (n
n (n + 1)
2
2
n (n + 1)(2n + 1) + 1)
⇒ > (1 ⋅ 22 ⋅ 33 ⋅ 44 K nn )n (n
6n (n + 1)
2
2
⇒  2n + 1  > (1 ⋅ 22 ⋅ 33 ⋅ 44 K nn )n (n + 1)
 
 3 
n (n + 1)
⇒  2n + 1  2
> 1 ⋅ 22 ⋅ 33 ⋅ 44 K nn
 
 3 
n (n + 1)
2n + 1 
1 ⋅ 2 ⋅ 3 ⋅ 4 K n < 
2 3 4 n 2
Hence, 
 3 
Hence proved.
www.pdfworld.in
www.pdfworld.in
Example 59. For any n ∈ N . Prove that

+1 n
[(n + 1)!]1/n <1+ (n !)1/n .
n+1
Solution Q AM > GM we have

[ (n + 1) + (n !)1/n + (n !)1/n + K + (n !)1/n ]
+1
> (n + 1)[(n + 1)(n !)1/n (n !)1/n (n !)1/n K (n !)1/n ]1/n
⇒ [(n + 1) + n (n !)1/n ] > (n + 1)
+1
[(n + 1) ⋅ n !]1/n
+1
⇒ [(n + 1) + n (n !)1/n ] > (n + 1)[(n + 1)!]1/n
n +1
⇒ 1+ (n !)1/n > [(n + 1)!]1/n
n+1
+1 n
⇒ [(n + 1)!]1/n <1+ (n !)1/n
n+1
Hence proved.
Example 60. Prove that for each natural number n

1 1 1
+ + K+ > 1.
n+1 n+2 3n + 1
1 1 1 1
Solution + + + K+
n+1 n+2 n+3 3n + 1
 1 1   1 1  1 1  1
= +  +  +  +… +  +  +
 n + 1 3n + 1   n + 2 3n   2n 2n + 2  2n + 1
Since, the AM of two unequal +ve numbers exceeds their HM.
−1
1  1 1   (n + 1) + (3n + 1)
∴  + >
2  n + 1 3n + 1   2 
1
=
2n + 1
−1
1  1 1   (n + 2) + 3n  1
 + > =
2  n + 2 3n   2 
 2n +1
−1
1  1 1   2n + (2n + 2) 1
 + > =
2  2n 2n + 2   2 
 2n + 1
Adding corresponding sides of inequalities, we get
1 1 1 1 1 1  n
 + + K+ + + K+ >
2 n + 1 n + 2 2n 2n + 2 3n + 1  2n + 1
1
Multiplying throughout by 2 and adding to both sides, we get
2n + 1
1 1 1
+ + K+ >1
n+1 n+2 3n + 1
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 295

1 1 1 1 1
1< + + + K+ <1 .
1001 1002 1003 3001 3
Solution AM of any n distinct +ve numbers always exceeds their HM. We shall apply this result to the
2001 numbers.
1001, 1002, … , 3000, 3001
For these numbers
1001 + 1002 + K + 3001
AM =
2001
2001 1
= (1001 + 3001) ⋅
2 2001
= 2001
–1
 1 + 1 + K+ 1 + 1 
 3000 3001 
HM =  1001 1002 
 2001 
 
= 2001 / 
1 1 1 
+ + K+ 
 1001 1002 3001 
Q AM > HM
2001 > 2001 / 

1 1 1 
∴ + + K+ 
 1001 1002 3001 
1 1 1
i.e., + + K+ >1
1001 1002 3001
Now, let us observe that
1 1 1 250 1
+ + K+ < =
1001 1002 1250 1000 4
1 1 1 250 1
+ + K+ < =
1251 1252 1500 1000 5
1 1 500 1
+ ... + < =
1501 2000 1500 3
1 1 1 1001
+ + K+ <
2001 2002 3001 2000
Adding throughout, we have
1 1 1 1 1 1 1001 7703 1
+ + K+ < + + + = <1
1001 1002 3001 4 5 3 2000 6000 3
Example 62. If n be a +ve integer. Prove

n n +1 n n +1
 1   1 
(i )  1 +  <  1 + (ii )  1 −  <  1 −
1 1
 
 n  n + 1  n  n + 1
n +1 n + 2
 1 
(iii )  1 + 
1
> 1 + 
 n  n + 1
Solution (i) If x 1 , K , x n +1 be +ve numbers (not all equal), then by the inequality of the means, we have
1
+1 x1 + K + xn +1
(x 1 K x n + 1)
n
< …(i)
n+1
www.pdfworld.in
www.pdfworld.in
1
Putting x1 = x 2 = K = xn = 1 + ; xn +1 =1
n
we get
n / (n + 1)
1 + 1  n+2
  <
 n n+1
n / (n + 1)
1 + 1  <1+
1
 
 n n+1
Raising both sides to power n + 1, we get
n n +1
1 + 1  < 1 + 1 
   
 n  n + 1
1
(ii) Putting x 1 = x 2 = K = x n = 1 − ; xn +1 = 1 in (i), we get
n
1/ (n + 1)
 1 
n
n
1 −   <
  n   n+1
n n +1
1 − 1  < 1 − 1 
or     …(ii)
 n  n + 1
n +1 n +1
n + 1
(iii)  1 +  = 
1 1 1
 = = …(iii)
 n  n   n 
n +1 n +1
 1 
  1 − 
 n + 1  n + 1 
Replacing n by n + 1 in (iii), we get
n + 2
 1  1
1 +  = n + 2
…(iv)
 n + 1  1 
1 − 
 n + 2
Replacing n by n + 1 in (ii), we get
n +1 n + 2
 1   1 
1 −  < 1 − 
 n + 1  n + 2
1 1
or n +1
> n + 2
 1   1 
1 −  1 − 
 n + 1  n + 2
n +1 n + 2
1 + 1   1 
or   > 1 +  [using (iii) and (iv)]
 n  n + 1

1 ! 3 ! 5 ! K (2n − 1)! > (n !)n .
Solution First let us prove the following

r !(2n − r )!
> n !, n > r
n
r !(2n − r )!
Q
n!
(1 ⋅ 2 ⋅ 3 K r )(2n − r )(2n − r − 1) K (n + 1)n !
=
n!
www.pdfworld.in
www.pdfworld.in
Inequalities 297
= (1 ⋅ 2 ⋅ 3 K r )(2n − r )(2n − r − 1) K (n + 1)
> (1 ⋅ 2 ⋅ 3 K r ) [ (r + 1)(r + 2) K n ] [Qn > r ]
Also, (n + 1) > r + 1
n −r > 0
2n − r > n
∴ 1 ⋅ 2 ⋅ 3 K r (r + 1)(r + 2) ... > n !
r !(2n − r )!
∴ > n ! or r !(2n − r )! > (n !)2
n!
Now, giving r successively the value 1, 3, 5, …, (2n − 1)
We get 1 !(2n − 1)! > (n !)2
3 !(2n − 3)! > (n !)2
5 !(2n − 5)! > (n !)2
…………………
…………………
…………………
(2n − 1)! 1 ! > (n !)2
Multiplying both sides, we have
[1 ! 3 ! 5 ! 7 ! K (2n − 1)!]2 > (n !)2n
Taking square root of both sides
1 ! 3 ! 5 ! K (2n − 1) > (n !)n
Hence proved.
Example 64. If a , b , c , d , … are p +ve integers, whose sum is equal to n. Show that the least value of
a ! b ! c ! d !…is (q !)p − r (q ! + 1)!)r , where q is the quotient and r the remainder when n is divided by p.
Solution Take the expression a ! b ! c ! d !…

Then, if any two of the quantities a and b say, are not equal, we can diminish a ! b ! by taking a, b equal
and keeping sum of a and b to be the same.
Hence, the value of a ! b ! c ! d !… is least when all the quantities a , b , c , d , … are equal. If however n is not
exactly divisible by p, this will not be the case.
Suppose r is the remainder, q is quotient when n is divided by p.
Thus, n = pq + r
= (p − r )q + r (q + 1)
∴ p − r of the quantities a , b , c , d , … will be equal and the remainder r will be equal to (q + 1).
So, the least value of the expression is (q !)p − r ((q ! + 1)!)r .
Example 65. For any natural number n, prove the following inequality.
1  1 1 1  1 1 1 1
1 + + + K + ≥  + + K+  (RMO 1998)
n+1 3 5 2n − 1  n  2 4 2n 
Solution We will prove it by induction.

Given inequality is true for n = 1. Assume it to be true as such and show that when n is replaced by (n + 1)
too it is true.
www.pdfworld.in
www.pdfworld.in
Rewriting, the inequality is

1  1 1  1 1 1
1 + + K + ≥  + K+ 
n+1 3 2n − 1  n  2 2n 
n+1 1 1 + 1 + K + 1 
= ⋅  
n n+1 2 4 2n 
=  1 + 
1 1 1 1 1
 + + K+ 
 n n + 1 2 4 2n 
1  1 1 1 1 1 1 1 + 1 + K + 1 
1 − + − + K + − ≥  
n+1 2 3 5 2n − 1 2n  n (n + 1)  2 4 2n 
 1 1 1 1 1  1 1 1 1
i.e., 1 − + − + K + − ≥  + + K+ 
 2 3 5 2n − 1 2n  n  2 4 2n 
1 log n
(As n → ∞. LHS converges to log 2, while RHS converges to lim ⋅ = 0). Assuming the last
n→ ∞ 2 n
inequality we have to prove.
1 1 1 1 1 1 1 1 1 
1− + −K+ − + − ≥  + K+ 
2 3 2n − 1 2n 2n + 1 2n + 2 n + 1 2 2n + 2 
LHS (By induction hypothesis)
≥  + K +
1 1 1 1 1
 + −
n 2 2n  2n + 1 2n + 2
n + 1 1 1 1 1
=  + K+  + −
n (n + 1)  2 2n  2n + 1 2n + 2
=  1 + 
1 1 1 1 1 1 1 1 1  1 1
 + K+  + − ≥  + K+  +  + K+ 
 n n + 1 2 2n  2n + 1 2n + 2 n + 1  2 2n   2 2n 
1 1 1
+ −
n (n + 1) 2n + 1 2n + 2
It suffices to show that
 1 + ... + 1  1
+
1
−
1
≥
1
⋅
1
 
2 2n  n (n + 1) 2n + 1 2n + 2 n + 1 2n + 2
1 1 1
LHS ≥ + −
2n (2n + 1) 2n + 1 2n + 2
1 1
= +
2n (n + 1) (2n + 1)(2n + 2)
1 1
= +
2n (n + 1) 2(n + 1)(2n + 2)
1 1 1 
=  + 
2(n + 1)  n 2n + 1 
3n + 1
=
2(n + 1)n (2n + 1)
3n + 1 1 1
= ≥ ⋅
2n (n + 1)(2n + 1) n + 1 2n + 2
Q 3n + 1 > 2n and 2n + 2 > 2n + 1
∴ Given inequality is true by induction.
www.pdfworld.in
www.pdfworld.in
Inequalities 299
n
Example 66. Prove that for any +ve integer n we have 2 ≤  1 + 1  < 3. (IMO)
  n
Solution Let us prove that
k
k  k k2
≤  1 +  < 1 + + 2
1
1+
n  n n n
for any +ve integer k such that k ≤ n
We shall use the method of mathematical induction for k = 1, the required relation obviously holds.
Now, let us assume that it holds for k + 1 as well. We have
k+1 k
1 + 1   k 
=  1 +   1 +  ≥
1 1 1
  1 +  1 + 
 n  n  n  n  n
k+1 k k+1
=1+ + 2 >1+
n n n
It should be noted that here we have not used the relation k ≤ n.
Consequently, this inequality holds for any +ve integer k.
Let us put k ≤ n. Then, we get
k+1
1 + 1 
k
 k k2 
=  1 +   1 +  <  1 + + 2  1 + 1 
1 1
   
 n  n  n  n n   n
k + 1 k 2 + 2k + 1 k + 1 k 2
=1+ + − + 3
n n2 n2 n
k + 1 (k + 1)2 n (k + 1) − k 2 k + 1 (k + 1)2
=1+ + 2
− 3
<1+ +
n n n n n2
as n (k + 1) > k 2 for n ≥ k
On substituting the value k = n into the inequalities we have derived we get
n
n  n n2
≤  1 +  < 1 + + 2 = 3
1
2=1+
n  n  n n
Example 67. If abc = 1 a , b , c > 0. Show that

1 1 1 3
+ + ≥ . (IMO 1995)
a3 (b + c ) b3 (c + a ) c 3 (a + b ) 2
1 1 1
Solution Let x = ,y = , z =
a b c
1
xyz = = 1, ∴ xyz = 1
abc
1 1 1
Now, + +
a3 (b + c ) b3 (c + a ) c 3 (a + b )
x2 y2 z2
= + +
y + z z+x x +y
x2 y2 z2
Let S = + +
y + z z+x x +y
Consider (x + y + z )2
2
x y + z y z + x z x +y
=  + + 
 y + z z+x x + y 
www.pdfworld.in
www.pdfworld.in
Using Cauchy's inequality

(a1b1 + a 2b2 + a3b3 )3 ≤ (a12 + a 22 + a32 )
(b12 + b22 + b32 )
2
 x y z 
 y + z + z+x + x + y 
 y + z z+x x +y 
 x2 y2 z2 
≤  + +  ⋅ (y + z + z + x + x + y )
y + z z + x x + y 
⇒ (x + y + z )2 ≤ S (y + z + z + x + x + y )
x +y + z
or S≥
2
x +y + z
We know, ≥ (xyz )1/3
3
or x + y + z ≥ 3(xyz )1/3
x + y + z 3(xyz )1/3
or ≥
2 2
3
∴ We get S ≥ (xyz )1/3
2
3
S≥ [Qxyz = 1]
2
1 1 1 3
∴ + + ≥
a3 (b + c ) b3 (c + a ) c 3 (a + b ) 2
n!
Example 68. If Cr = . Prove that
r !(n − r )!
C1 + C2 + K + Cn < [n (2n − 1)]
Solution AM of (1/2)th powers < (1/2)th power of AM

1 1 1
1/ 2
(C1 )2 + (C2 )2 + K + (Cn )2  C1 + C2 + K + Cn 
∴ < 
n  n 
1/ 2
C1 + C2 + K + Cn  2n − 1 
⇒ < 
n  n 
n (2n − 1)
⇒ C1 + C2 + K + Cn <
n
Hence, C1 + C2 + K + Cn < [n (2n − 1)]
Example 69. If x + y + z = 1, x , y , z are +ve, show that

2 2 2
 x + 1  +  y + 1  +  z + 1  > 100
     
 x  y  z 3
Solution AM of (2)th power >(2)th power of AM
2 2 2 2
x + 1  + y + 1  +  z + 1   1  1  1 




  


   x +  +  y +  +  z +  
x  y z x  y z 
>
3  3 
 
 
www.pdfworld.in
www.pdfworld.in
Inequalities 301
2 2 2
x + 1  + y + 1  +  z + 1 
      2
 x  y  z 1 1 1 1
⇒ > x + y + z + + + 
3 3 x y z
2 2 2
x + 1  + y + 1  +  z + 1 
      2
 x  y  z 1 1 1 1
⇒ > 1 + + +  [Qx + y + z = 1]
3 9 x y z
−1
x −1 + y −1 + z −1  x + y + z 
Again, > 
3  3 
or x −1 + y −1 + z −1 > 9
2 2 2
 x + 1  +  y + 1  +  z + 1  (1 + 9)2 > 100
     
 x  y  z 3
Example 70. If x , y , z are unequal +ve quantities such that sum of any two is greater than the third. Show
that
1 1 1 9
+ + > .
y + z −x z + x −y x + y − z x + y + z
Solution Consider three quantities (y + z − x ), (z + x − y ) and (x + y − z ) and taking m = − 1

QAM of mth power > mth power of AM
−1
(z + x − y )−1 + (y + z − x )−1 + (x + y − z )−1  (y + z − x ) + (z + x − y ) + (x + y − z )
∴ >
3  3 
−1
1  1 1 1  x + y + z 3
or  + + >   =
3  y + z − x z + x − y x + y − z   3  x +y + z
1 1 1 9
or + + >
y + z −x z + x −y x + y − z x + y + z
Hence proved.

(b + c − a )2 + (c + a − b )2 + (a + b − c )2 ≥ (bc + ca + ab ).
Solution Taking m = 2 and considering three quantities b + c − a, c + a − b and a + b − c .

QAM of mth power > mth power of AM
2
(b + c − a )2 + (c + a − b )2 + (a + b − c )2  (b + c − a ) + (c + a − b ) + (a + b − c )
∴ > 
3  3 
2
(b + c − a )2 + (c + a − b )2 + (a + b − c )2  a + b + c 
or >  …(i)
3  3 
Now, considering a 2 and b 2
Q AM > GM
1 2
∴ (a + b 2 ) > a 2b 2
2
or a 2 + b 2 > 2ab
Similarly, b 2 + c 2 > 2bc
and c 2 + a 2 > 2ac
www.pdfworld.in
www.pdfworld.in
Adding these results, we get

(a 2 + b 2 ) + (b 2 + c 2 ) + (c 2 + a 2 ) > (2ab + 2bc + 2ca )
or (a 2 + b 2 + c 2 ) > ab + bc + ca
Now, add 2ab + 2bc + 2ca to both sides, we have
a 2 + b 2 + c 2 + 2ab + 2bc + 2ca > (3ab + 3bc + 3ca )
or (a + b + c )2 > 3(ab + bc + ca )
2
a + b + c  3(ab + bc + ca )
or   > [on dividing by 9]
 3  9
2
a + b + c  ab + bc + ca
or   > …(i)
 3  3
(b + c − a )2 + (c + a − b )2 + (a + b − c )2 > ab + bc + ca
If a = b = c , then
(b + c − a )2 + (c + a − b )2 + (a + b − c )2
= ab + bc + ca
Example 72. Prove that a +3b 3 +3 c > 1 + 1 + 1.

8 8 8
abc a b c
Solution Take m = 8 and consider a , b , c
AM of mth power > mth power of AM
8
a8 + b8 + c 8  a + b + c 
∴   >  
 3   3 
6 2
a8 + b8 + c 8  a + b + c  a + b + c 
  >   
 


…(i)
 3  3 3
3
a + b + c 
We know that   > abc …(ii)
 3 
2
a + b + c   ab + bc + ca 
  >  …(iii)
 3   3 
Squaring both sides of (ii), we get
6
a + b + c 
 >a b c
2 2 2
 …(iv)
 3 
Putting the value of (iii) and (iv) in (i), we have
a8 + b8 + c 8  ab + bc + ca 
> a 2b 2c 2  
3  3 
Dividing both sides by a3b3c 3
a8 + b8 + c 8 ab + bc + ca
>
a3b3c 3 abc
a8 + b8 + c 8 1 1 1
or > + +
a3b3c 3 c a b
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 303

a5 + b5 + c 5 > abc (ab + bc + ca )
Solution QAM of mth power > mth power of AM

5
a5 + b 5 + c 5  a + b + c 
Q > 
3  3 
3 2
a5 + b 5 + c 5  a + b + c   a + b + c 
>    …(i)
3  3   3 
Q AM > GM
3
a+b+c a + b + c 
∴ > (abc )1/3 or   > abc
3  3 
From (i), we have
2
a5 + b 5 + c 5 a + b + c 
> abc   …(ii)
3  3 
Q AM > GM
a + b2
2
∴ > a 2b 2 or a 2 + b 2 > 2ab
2
Similarly, b 2 + c 2 > 2bc and c 2 + a 2 > 2ac
Adding these results, we have
(a 2 + b 2 ) + (b 2 + c 2 ) + (c 2 + a 2 ) > 2ab + 2bc + 2ac
2(a 2 + b 2 + c 2 ) > 2(ab + bc + ca )
or a 2 + b 2 + c 2 > ab + bc + ca
Adding 2ab + 2bc + 2ca to both sides, we get
a 2 + b 2 + c 2 + 2ab + 2bc + 2ca > 3(ab + bc + ca )
(a + b + c )2 > 3(ab + bc + ca )
(a + b + c )2 3(ab + bc + ca )
> [on dividing by 9]
9 9
2
a + b + c   ab + bc + ca 
  > 
 3   3 
a5 + b 5 + c 5
From (ii), we get > abc
3
 ab + bc + ca 
 
 3 
∴ a5 + b5 + c 5 > abc (ab + bc + ca )
Example 74. By Cauchy's inequality. Prove that

b+c c+a a+b
+ + ≥6
a b c
where a , b , c are all +ve.
1 1 1
Solution Let two sets of numbers are a , b , c and , , . Now, from Cauchy-Schwartz
a b c
inequality, we have
www.pdfworld.in
www.pdfworld.in
[( a )2 + ( b )2 + ( c )2 ]
 1  2  1  2  1  ≥  a ⋅ 1 +
2
1 1 
2
  +   +     b⋅ + c⋅ 
  a   b  c  
 a b c
(a + b + c )  +
1 1 1
or +  ≥ (1 + 1 + 1)2
a b c
a+b+c a+b+c a+b+c
or + + ≥9
a b c
b+c c+a a+b
or 1+ +1+ +1+ ≥9
a b c
b+c c+a a+b
or + + ≥ 9−3
a b c
b+c c+a a+b
Hence, + + ≥6
a b c
Example 75. If x 1 , x 2 , …, x n and y1 , y 2 , …, yn be any two sets of +ve numbers, then by using Tchebychef 's
inequality show that
 x1 x 2 x  y y y 
 + + K + n   1 + 2 + K + n  ≥ n 2.
 y1 y 2 yn   x 1 x 2 xn 
Solution Assume that

x1 x 2 x
≤ ≤ K ≤ n , then
y1 y 2 yn
y1 y 2 y
≥ ≥ K ≥ n for two sets
x1 x 2 xn
x1 x 2 x y y y
i.e., , , K , n and 1 , 2 , K , n
y1 y 2 yn x1 x 2 xn
By applying Tchebychef 's inequality, we get
 x1 x 2 x  y y y  x y x y x y 
 + + K+ n   1 + 2 + K+ n ≥n 1 ⋅ 1 + 2 ⋅ 2 + K+ n ⋅ n 
 y1 y 2 yn   x 1 x 2 xn   y1 x 1 y 2 x 2 yn x n 
 x1 x 2 x  y y y 
or  + + K + n   1 + 2 + K + n  ≥ n(1 + 1 + K + 1)
 y1 y 2 yn   x 1 x 2 xn 
 x1 x 2 x  y y y 
∴  + + K + n   1 + 2 + K + n  ≥ n2
 1
y y 2 yn   1
x x 2 xn 

1 1
1+ + K+
2 n < (2n − 1)1/ 4
n
where n is any +ve integer.
Solution Q n is +ve integer such that 1 > 1 > K > 1 for two sets of number
2 n
1 1
1, ,K,
2 n
1 1
and 1, ,K,
2 n
www.pdfworld.in
www.pdfworld.in
Inequalities 305
Applying Tchebychef 's inequality

 1 1  1 1  1 1 1 1
 1+ + K+  1+ + K+ <n 1⋅ 1 + . + K+ . 
 2 n  2 n  2 2 n n
2
 1 1  1 1
⇒  1+ + K+  < n 1 + + K +  …(i)
 2 n  2 n
Again applying Tchebychef ’s inequality for two sets.
1 1 1 1
1, , K , and 1, , K ,
2 n 2 n
1 1
Q 1> > K>
2 n
∴ 1 + 1 + K + 1  1 + 1 + K + 1  < n 1 ⋅ 1 + 1 ⋅ 1 + K 1 ⋅ 1 
    
 2 n  2 n  2 2 n n
2
∴  1 + 1 + K + 1  < n  12 + 1 + K + 1 
    …(ii)
 2 n  22 n2 
Q n > 0 and − n < 0
or n2 − n < n2
n (n − 1) < n 2
1 1
or <
n (n − 1) n 2
Putting n = 2, 3, 4, …, n
1 1
>
1 ⋅ 2 22
1 1
>
2 ⋅ 3 32
1 1
>
3 ⋅ 4 42
1 1
> 2
(n − 1) ⋅ n n
…………
…………
…………
Adding all corresponding sides, we get
1 1 1 1 1 1 1 1
+ + + K+ > + + + K+ 2
1⋅2 2⋅3 3⋅ 4 (n − 1)n 22 32 42 n
1 1 1 1 1 1 1 1 1 1 1
or 1− + − + − +K+ − > + + + K+ 2
2 2 3 3 4 n − 1 n 22 32 42 n
1 1 1 1 1
⇒ 1− > 2 + 2 + 2 + K+ 2
n 2 3 4 n
1 1 1 1
or 1+ 2 + 2 + K+ 2 <2−
2 3 n n
n  1 + 2 + 2 + K + 2  < (2n − 1)
1 1 1
…(iii)
 2 3 n 
www.pdfworld.in
www.pdfworld.in
From (ii) and (iii), we get

2
 1 + 1 + K + 1  < (2n − 1)
 
 2 n
 1 + 1 + K + 1  < (2n − 1)1/ 2
  …(iv)
 2 n
From (i) and (iv), we have
2
 1 1
 1+ + K+  < n (2n − 1)
1/ 2
 2 n
 1 1
 1+ + K+  < n (2n − 1)
1/ 4
 2 n
1 1
1+ + K+
Hence, 2 n < (2n − 1)1/ 4
n

1 ⋅ 3 ⋅ 5 ⋅ 7 K (2n − 1) 1
>
2 ⋅ 4 ⋅ 6 K 2n (n + 1)
where n is +ve integer.
1 1 1 1
Solution Let a1 = , a 2 = , a3 = , …, an =
2 4 6 2n
1 1 1 1
then a1 + a 2 + a3 + K + an = Sn = + + + K+ <1
2 4 6 2n
Now, apply second inequality of Weierstress's inequality, then
(1 + Sn ) < (1 + a1 )(1 + a 2 )(1 + a3 ) K (1 + an )
(1 + a1 )(1 + a 2 )(1 + a3 ) K (1 + an ) > 1 + Sn
⇒ 1 + 1  1 + 1  1 + 1  K 1 + 1  > 1 + 1 + 1 + 1 + K + 1
     
 2  4  6  2n  2 4 6 2n
3 5 7 (2n + 1)
⇒ ⋅ ⋅ K > 1 + 2−1 + 4−1 + 6−1 + K + (2n )−1 …(i)
2 4 6 2n
Q AM of (−1) th power > (−1) th power of AM
−1
2−1 + 4−1 + 6−1 + K + (2n )−1  2 + 4 + 6 + K + 2n 
i.e., > 
n  n 
 n 
= 
 2 + 4 + 6 + K + 2n 
n2
2−1 + 4−1 + 6−1 + K + (2n )−1 >
n
(2 + 2n )
2
n
2 −1 + 4 −1 + 6 −1 + K + (2n )−1 >
n+1
n
1 + 2−1 + 4−1 + 6−1 + K + (2n )−1 > 1 +
n+1
2n + 1
1 + 2−1 + 4−1 + 6−1 + K + (2n )−1 > …(ii)
n+1
www.pdfworld.in
www.pdfworld.in
Inequalities 307

3 5 7 (2n + 1) (2n + 1)
⋅ ⋅ K >
2 4 6 2n n+1
3 ⋅ 5 ⋅ 7 K (2n − 1)(2n + 1) (2n + 1)
⇒ >
2 ⋅ 4 ⋅ 6 K 2n n+1
3 ⋅ 5 ⋅ 7 K (2n − 1) 1
⇒ >
2 ⋅ 4 ⋅ 6 K 2n n+1
Example 78. If x , y , z are each +ve and x + y + z = 6. Show that

2 2 2
 1  1  1 75
x +  + y +  +  z +  ≥ . (IMO)
 y  z  x 4
Solution By using HM inequality

 1 1 1
 + + 
x y z 3
≥
3 x +y + z
3 1
= =
6 2
1 1 1 3
So, + + ≥
x y z 2
2 2 2
 1  1  1 1 1 1
x +  + y +  +  z +  x + +y + + z+
 y  z  x y z x
Now, ≥
3 3
3
6+
≥ 2 =5
3 2
Now, squaring both sides and multiplying by 3, we get
2 2 2
 1  1  1 75
x +  + y +  +  z +  ≥
 y  z  x 4
Example 79. If x , y , z > 0, prove that

(xy + yz + zx )
 1 1 1  9
 + + 2
≥ . (IMO)
 (x + y ) (y + z ) (z + x )  4
2 2
Solution Let y + z = a,
z + x = b and x + y = c
a
We have x =b + c −
2
a 2 − (b − c )2
Thus, yz =
4
a 2 − b 2 − c 2 + 2bc
=
4
1
Σ yz = Σ (2bc − a 2 )
4
www.pdfworld.in
www.pdfworld.in
Now, assume a ≥ b ≥ c without loss of generality, then

2bc − a 2 ≤ 2ca − b 2 ≤ 2ab − c 2
∴By Tchebychef ’s inequality and AM–GM inequality
4  1  1 1 1
(Σyz )  Σ 2
= Σ(2bc − a 2 ) Σ 2
9  (y + z )  3 3 a
1/3
1   2bc   2bc 
Σ (2bc − a 2 ) ⋅ 2  =
1 1
≥ Σ 2  − 1≥ π 2  −1=2−1=1
3  a  3  a   a 
Example 80. Determine all real numbers satisfying the inequality

1
log(2x − 1) + log x − 9 > 1 .
2
[log denotes logarithm to base 10]
Solution 1 log(2x − 1) + log x − 9 > 1.
2
1 1
⇒ log(2x − 1) + log(x − 9) > 1
2 2
⇒ log [(2x − 1)(x − 9)] > 2
⇒ (2x − 1)(x − 9) > 102
⇒ 2x 2 − 19x − 91 > 0
⇒ (x − 13)(2x + 7) > 0
−7
⇒ Either x < or x > 13
2
Qx cannot be less than –7/2 (as if x < 9, then log x − 9 has no meaning)
∴ x must be > 13
The desired set is {x : x > 13 and x ∈ R}
Example 81. Without evaluating square roots, find the larger of two surds : 12 − 11 , 7 − 6
Solution Q ( 12 − 11 )( 12 + 11 ) = 1
( 7− 6 )( 7 + 6) = 1
∴ ( 12 − 11 )( 12 + 11 ) = ( 7 − 6 )( 7 + 6)
( 12 − 11 ) ( 7+ 6)
i.e., < …(i)
( 7 − 6) ( 12 + 11)
∴ 7+ 6 < 6, 12 + 11 > 6
7+ 6 1
∴ < 6⋅ = 1
12 + 11 6
12 − 11
Consequently, <1
7− 6
so that 7− 6 > 12 − 11
Example 82. Let a1 , a2 , K , an be real numbers all greater than 1 and such that | ak − ak + 1| < 1 for
1 ≤ k ≤ n − 1. Show that
a1 a 2 a a
+ + K + n − 1 + n < 2n − 1.
a 2 a3 an a1
www.pdfworld.in
www.pdfworld.in
Inequalities 309
Solution By using mathematical induction, we will prove the result.

a1 a 2
(i) We will prove that + <3
a 2 a1
Assume a1 < a 2
Q | a1 − a 2| < 1, we have a 2 − a1 < 1
i.e., a1 < a 2 < a1 + 1
a1 a 2 a +1 1
So + <1+ 1 =2+ <3 [Qa1 > 1]
a 2 a1 a1 a1
Assume a 2 < a1
Q | a1 − a 2| < 1, we have a 2 < a1 < a 2 + 1
a1 a 2 a 2 + 1 1
+ < + 1=2+ <3 [Qa 2 > 1]
a 2 a1 a2 a2
a1 a 2
Assume a1 = a 2, then + = 2< 3
a 2 a1
(ii) Now assume
a1 a 2 a
+ + K + k < 2k − 1
a 2 a3 a1
a1 a 2 ak + 1
+ + K+
a 2 a3 a1
a a a  ak a a  a a a 
=  1 + K + k −1 + k  + + k + 1 − k < (2k − 1) +  k + k+1 − k
 2 a a1 
a a k a1 a k+1 a1 a1  k+1 a1
Now it is enough to show that

ak ak + 1 ak
+ − is less than 2.
ak + 1 a1 a1
If a k < a k + 1, then a k + 1 − a k < 1 and we have
ak a a 1
+ k+1 − k <1+ <2 [Qa1 > 1]
ak + 1 a1 a1 a1
If a k > a k + 1, then a k − a k + 1 < 1, we have

ak a a
< 2, k + 1 − k < 0
ak + 1 a1 a1
 a a a 
If a k = a k + 1, then  k + k + 1 − k  = 1
a a a1 
 k+1 1
So in all the cases

ak ak + 1 ak
+ − <2
ak + 1 a1 a1
Example 83. If a , b are +ve real numbers. Prove that

+1 +1 n
[(a + b )(a 2 + b 2 ) K (an + bn )]2 > (an + bn ) ,
for every +ve integer n.
Solution [(a + b )(a 2 + b 2 ) K (an + bn )]2
= [(a + b )(a 2 + b 2 ) K (an + bn )]
−1 −1
[(an + bn )(an + bn ) K (a + b )]
−1 −1
= (a + b )(a + b )(a + b 2 )(an
n n 2
+ bn ) K (an + bn )(a + b )
www.pdfworld.in
www.pdfworld.in
If k be any +ve integer such that

1 ≤ k ≤ n, then
−k+1 −k+1
(a k + b k )(an + bn )
+1 +1 −k+1 −k+1 +1 +1
= an + bn + a kbn + b kan > an + bn
Applying this result to each of the n products we find
[(a + b )(a 2 + b 2 ) K (an + bn )2 ]2
+1 +1 +1 +1
> (an + bn )(an + bn ) K to n factors
+1 +1 n
= (an + bn )
Hence,
+1 +1 n
[(a + b )(a 2 + b 2 ) K (an + bn )]2 > (an + bn )
Hence proved.
Example 84. If a1 ≥ a2 ≥ K ≥ an be real numbers such that a1k + a2k + K + ank ≥ 0 for all integers k > 0 and
p = max {| a1| , | a 2| , K , | an | }. Prove that p = | a1| = a1 and that (x − a1 )(x − a 2 )…
(x − an ) ≤ x n − a1n
Solution Take k = 1
Q a1k + a 2k + K + ank ≥ 0
and for k = 1, we have
a1 + a 2 + K + an ≥ 0 …(i)
Q a1 ≥ a 2 ≥ a3 ≥ K ≥ an , a1 ≥ 0 …(ii)
and if all ai , i = 1, 2, K n are +ve
a1 is the maximum of all ai ' s.
∴ p = | a1| = a1
Suppose some of the ai ' s are –ve and p ≠ a , then an < 0
Hence, p = | an |
Let r be an index such that
an = an −1 = K = ar +1 < ar ≤ ar −1 ≤ K ≤ a1
Then a1k + a 2k + K + ar −1 k + ar k + K + ank
 a  k  a  k a 
k
a 
k 
= ank   1  +  2  + K +  r − 1  +  r  + (n − r )
  an   an   an   an  
 
= ank X
 a  k  a  k  ar − 1 
k
a 
k 
where X =  1  +  2  + K +   +  r  + (n − r )
  an   an   an   an  
 

 1
a a 2
, , K ,
ar 
< 1
Q
an  an  an 
So their kth powers < these fractions and by taking k sufficiently large which would make X > 0 and
Xank < 0 for k odd.
www.pdfworld.in
www.pdfworld.in
Inequalities 311
A contradiction and hence p = a1

Let x > a1
(x − a 2 )(x − a3 )(x − a 4 ) K (x − an )
n −1
 n 
i Σ (x − ai ) n −1
=2  (n − 1) x + a1   b 
≤  ≤   Qi Σ ai ≥ 0
 n −1   n −1   =1 
 
n −1
 a1 
= x +
 n − 1 
≤ xn −1
+ xn −2
a1 + x n −2 2
+ K + a1n −1 Q  n − 1  ≤ (n − 1)r , r ≥ 1
  r 
a1

 
Multiplying both sides by (x − a1 ), we get
(x − a1 ) (x − a 2 ) (x − a3 ) K (x − an )
−1 −2 −1
≤ (x − a1 )(x n + xn a1 + K + a1n ) = x n − a1n
Example 85. If x , y , z be +ve real numbers and xyz = 1, prove that

xy yz xz
+ + ≤ 1,
x 5 + y 5 + xy y 5 + z5 + yz z5 + x 5 + zx
when does equality hold. (IMO 1996)
Solution x 5 + y 5 = (x + y )(x 4 − x 3y + x 2y 2 − xy 3 + y 4 )
= (x + y )[x 4 + x 2y 2 + y 4 − xy (x 2 + y 2 )]
= (x + y )[(x 2 + xy + y 2 ) (x 2 − xy + y 2 ) − xy (x 2 + xy + y 2 ) + x 2y 2 ]
= (x + y )[(x 2 + xy + y 2 ) (x 2 − 2xy + y 2 ) + x 2y 2 ]
= (x + y )[(x 2 + xy + y 2 )(x − y )2 + x 2y 2 ]
≥ (x + y ) × x 2y 2 [Q (x − y )2 (x 2 + xy + y 2 ) ≥ 0]
i.e., x 5 + y 5 ≥ x 2y 2 (x + y )
Equality holds if a = b
1 1
≤
x 5 + y5 x 2y 2 (x + y )
x 5 + y5
⇒ ≥ xy (x + y )
xy
x 5 + y5 + 1
⇒ ≥ xy (x + y ) + 1
xy
x 5 + y 5 + xy
⇒ ≥ xy (x + y ) + 1
xy
xy 1
⇒ ≤
x 5 + y 5 + xy 1 + xy (x + y )
www.pdfworld.in
www.pdfworld.in
Example 86. If x , y , z are sides of a triangle and x + y + z = 2, show that

x 2 + y 2 + z 2 + 2xyz < 2. (INMO 1993)
Solution It is given x + y + z = 2
On squaring both sides of above equation, we get
⇒ 4 = (x + y + z )2 = x 2 + y 2 + z 2 + 2(xy + yz + zx )
⇒ x 2 + y 2 + z 2 = 2 (2 − xy − yz − zx )
Adding 2xyz on both sides, we get
x 2 + y 2 + z 2 + 2xyz = 2(2 − xy − yz − zx + xyz)
Now, to show that x 2 + y 2 + z 2 + 2xyz < 2 it is enough to show that
2(2 − xy − yz − zx + xyz ) < 2
or 2 + xyz − xy − yz − zx < 1
or xy + yz + zx − xyz − 1 > 0
Q x + y + z = 2s = 2
So s =1
Now, 1(1 − x )(1 − y )(1 − z ) > 0 as the expression in the LHS is square of the area of triangle with sides
x , y , z.
⇒ 13 − (x + y + z )12 + (xy + yz + zx )1 − xyz > 0
or 1 − 2 + xy + yz + zx − xyz > 0
or xy + yz + zx − xyz − 1 > 0 as desired.
Example 87. Let a , b , c be +ve real numbers such that abc = 1. Prove that
a − 1 + 1  b − 1 + 1  c − 1 + 1  ≤ 1
   
 b  c  a
Solution Now, b − 1 + 1  = b 1 − 1 + 1 
   
 c  b bc 
= b  1 + a −  Qa = 1 
1
 b  bc 
⇒  a − 1 + 1   b − 1 + 1  = b a 2 −  1 − 1  
     
 b  c   b 2  
≤ ba 2
So, the square of product of all three
≤ ba 2cb 2ac 2 = 1
Actually that is not quite true. The last sentence would not follow if we had some –ve LHS, as then we
couldn't multiply inequalities.
But it is easy to deal separately with the case where  a − 1 +  ,  b − 1 +  ,  c − 1 +  are not all +ve. If
1 1 1
 b  c  a
one of the three terms is –ve then other two must be +ve. For example
1 1
if a − 1 + < 0, then a < 1 so c − 1 + > 0
b a
1
b > 1 so b − 1 + > 0
c
But if one term is –ve and two are +ve, then their product is –ve and hence less than 1.
www.pdfworld.in
www.pdfworld.in
Inequalities 313

k+1
n k + 1 < 1k + 2k + 3k + .. + n k <  1 + 
1 1 1
nk + 1
k+1  n k+1
(n and k are arbitrary integers).

Solution Now, in
S = x k + x k −1 + x k − 2 + K + x + 1
if x > 1, then the first term is numerically the greatest but if x < 1, then the last term is greatest. It follows
that
(k + 1) x k > s > k + 1, if x > 1
(k + 1) x k < s < k + 1, if x < 1
If both sides of these inequalities are multiplied by x − 1, it is found that for x ≠ 1

(k + 1)x k ( x − 1) > x k + 1 − 1 > (k + 1)(x − 1)
p
Assume that x = , then we find
( p − 1)
(k + 1) p k p k + 1 − ( p − 1)k + 1 (k + 1) ( p − 1)k
> >
( p − 1)k + 1 ( p − 1)k + 1 ( p − 1)k + 1
p+1
If we assume x = , we get
p
(k + 1)( p + 1)k ( p + 1)k + 1 − p k + 1 (k + 1) p k

k+1
> k+1
>
p p pk + 1
If follows that
( p + 1)k + 1 − p k + 1 > (k + 1) p k > p k + 1 − ( p − 1)k + 1
Letting p successively have the values 1, 2, 3, ... , n.

2k + 1 − 1k + 1 > (k + 1)1k > 1k + 1 − 0
3k + 1 − 2k + 1 > (k + 1)2k > 2k + 1 − 1k + 1
4k + 1 − 3k + 1 > (k + 1)3k > 3k + 1 − 2k + 1
…………
…………
…………
k+1
(n + 1) − n k + 1 > (k + 1)n k > n k + 1 − (n − 1)k + 1
If these inequalities are added together, we get

(n + 1)k + 1 − 1 > (k + 1)(1k + 2k + 3k + K + n k ) > n k + 1
Dividing through these inequalities by k + 1

 k+1
1 1  1 1
1 +  − k + 1 n k + 1 > 1k + 2k + 3k + K + n k > nk + 1
  n  n  k + 1 k + 1
www.pdfworld.in
www.pdfworld.in

2 (a + b )  + 
a b 1 1
3 + 3 ≤ 3
b a a b 
for all +ve real numbers a and b.
Determine when equality occurs. (IMO)
Solution Multiplying both sides of desired inequality by 3 ab gives

3 3
a2 + b 2 ≤ 3 2(a + b )2
Put 3 a = x and 3 b = y we find that it is sufficient to prove
x 2 + y 2 ≤ 3 2(x 3 + y 3 )2 for x , y > 0

By AM–GM inequality, we have
x 6 + x 3y 3 + x 3y 3
≥ (x 12y 6 )1/3
3
or 3x 4y 2 ≤ x 6 + x 3y 3 + x 3y 3
and 3x 2y 4 ≤ y 6 + x 3y 3 + x 3y 3
with equality if and only if x = x y = y or equivalently if and only if x = y . Adding these inequalities
6 3 3 6
and adding x 6 + y 6 to both sides yields

x 6 + y 6 + 3x 2y 2 (x 2 + y 2 )≤ 2(x 6 + y 6 + 2x 3y 3 )
Taking cube roots of both sides yields.
Equality occurs when x = y or a = b.

a3 b3 c 3 (a + b + c )3
+ + ≥
x y z 3(x + y + z )
a , b , c , x , y , z are +ve real numbers.
Solution By Holder's inequality
3
Π ( pi3 + qi3 + ri3 )1/3 ≥ p1p 2p3 + q1q 2q3 + r1r2 r3
i =1
for all +ve reals pi , q i , ri
1/3
 a1/3 b3 c 3 
Hence,  + +  (1 + 1 + 1)1/3
 x y z
(x + y + z )1/3 > a + b + c
Cubing both sides and then dividing both sides by 3(x + y + z ), we get
a3 b3 c 3 (a + b + c )3
+ + ≥
x y z 3(x + y + z )
Example 91. Let m and n be +ve integers. Let a1 , a2 , K , am be distinct elements of {1, 2, K , n} such that
whenever ai + a j ≤ n for some i , j (possibly the same) we have ai + a j = a k for some k. Prove that
a1 + a 2 + K + am n + 1
≥ .
m 2
Solution A = {a1 , a 2 , K , an }
Assume, a1 > a 2 > K > am
www.pdfworld.in
www.pdfworld.in
Inequalities 315
Assign to any ai element of A, the element am − i + 1 as its pair. This matching is symmetric, the pair of
am − i + 1 is ai and if m is odd, then median am + 1 is equal to its own pair.
2
First we show that sum of any pair is at least n + 1. If to the contrary ai + am −i + 1 < n + 1 for some
(1 ≤ i ≤ m ) then ordering of A gives
ai < ai + am < ai + am −1 < K< ai + am −i + 1 ≤n …(i)
Hence, by condition the i sums in (i) produce elements of A. All of them are greater than ai a
contradiction since there are only i − 1 elements in A beyond ai .
∴ a1 + am ≥ n + 1
a 2 + am −1 ≥n + 1
…………
…………
…………
am + a1 ≥ n + 1
2(a1 + a 2 + K + am ) ≥ m (n + 1)
Example 92. Let {ak } be a sequence of distinct +ve integers (k = 1, 2, K , n , K ). Prove that for every +ve
n ak n 1
integer n, Σ 2
≥ Σ . (INMO 1978)
k =1 k k =1 k
Solution Let x 1 , x 2 , K , x n ; y1 , y 2 , K , yn be real numbers
where x 1 ≥ x 2 ≥ K ≥ x n , y1 ≤ y 2 ≤ ... ≤ yn
Moreover z1 , z2 , K , zn a permutations of the yi ’s, then
n k
Σ x i yi ≤ Σ x i zi …(i)
i =1 i =1
In (i) equality holds if the order of zi ’ s agrees with the order of the yi ’ s.
If to the contrary, we assume that they first differ at the kth place i.e., y1 = z1 , y 2 = z2 , K , y k − 1 = zk − 1
But y k ≠ zk
Let zk = yr and y k = zs
where r and s are greater than k , r > k and s > k using this notation x k zk + x s zs = x kyr + x s y k .
Now, change the LHS by substituting
x k zk + x s zs = x kyr + x s y k by x ky k + x s yr . This way we increased the sum on the left.
Q (x ky k + x s yr ) − (x kyr + x s y k ) = (x k − x s ) ( y k − yr ) ≥ 0
This implies that the sum Σ x i yi , where x i − s are in decreasing order is the smallest if yi − s are in
increasing order. There is a minimal among the above sums as there are finitely many of them and that
can only occur when the yi − s are in increasing order otherwise we could decrease the sum using our
arguments.
So, for a given n let ai 1 ≤ ai 2 ≤ K ≤ ain be the first k elements in increasing order.
1 1 1 1
But > > > K> 2
12 22 32 n
and hence by (i)
n a n a
Σ 2k ≥ Σ ik2 …(ii)
k =1 k k =1 k
ai 1 ≥ 1, ai 2 ≥ 2, K , aik ≥ k , K , ain ≥ n and so (3) implies that

n a n k n 1
Σ k≥ Σ 2 = Σ
k =1 2 k =1 k k =1 k
www.pdfworld.in
www.pdfworld.in
Example 93. Consider the infinite non-increasing sequence {x i } of +ve reals such that x 0 = 1. Prove that for
every such sequence there is an n ≥ 1 such that
x 02 x 12 x2
Sn = + + K + n − 1 ≥ 3999.
x1 x2 xn
Also find such a sequence for which
x 02 x 12 x2
Sn = + + K + n − 1 < 4. (INMO 1982)
x1 x2 xn
Solution It is given
x 02 x 12 x2
Sn = + + K + n − 1 ≤ 3999 …(i)
x1 x2 xn
Q (x i − 2x i + 1)
2
≥ 0 ; x i2 ≥ 4x i x i +1 − 4x i2+ 1
x i2
follows and so ≥ 4(x i − x i + 1)
xi + 1
Applying this to sum in (i)
x 02 x 12 x2
Sn = + + K + n − 1 ≥ 4(x 0 − x 1 + x 1 − x 2 + K + x n −1 − xn )
x1 x2 xn
= 4(1 − x n )
1
If the limit of the sequence x n is zero, then there exists an n such that x n ≤ and for this n,
4000
Sn ≥ 4  1 −
1 
 = 3999
 4000 
Now, let the limit c be different from 0, c > b > 0 so every element of the sequence is greater than b.
Let n ≥ 4 / b, then
n x i2 n x x n
Sn = Σ > Σ i i + 1 = Σ x i > nb ≥ 4
i =0x i =0 x i =0
i +1 i+1
Now, let x i = 2−i (i = 0, 1, ... ). Now

x i2 2−2i
= − (i + 1) = 21 − i
xi + 1 2
1 1
and so Sn = 2 + 1 + + K + n −1
2 2
Extending it to an infinite geometric sequence the sum is 4.
Hence, the finite sum Sn never reaches 4.
Example 94. If a , b , c be the length of the sides of a triangle. Prove that

a 2b (a − b ) + b 2c (b − c ) + c 2a (c − a ) ≥ 0. (INMO 1983)
Solution Let
a 2b (a − b ) + b 2c (b − c ) + c 2a (c − a ) = 0 …(i)
Let x = −a + b + c ;y =a −b + c
and z =a + b −c
www.pdfworld.in
www.pdfworld.in
Inequalities 317
Thus, x , y , z are twice the length of the segments between the vertices and the touching point of the
incircles. So
y + z z+x x +y
a= ,b = ,c =
2 2 2
Substitute them to (i) and multiply the inequality by 16.
(y + z )2 (z + x )(y − x ) + (z + x )2 (x + y )(z − y ) + (x + y )2 (y + z )(x − z ) ≥ 0
⇒ x 3z + y 3x + z 2y ≥ x 2yz + y 2zx + z 2xy
and so x 3z + y 3x + z3y − x 2yz − y 2zx − z 2xy
= zx (x − y )2 + xy (y − z )2 + yz (z − x )2 ≥ 0
As x > 0, y > 0, z > 0, equality holds if and only if x = y = z.
i.e., if a = b = c , then triangle is equilateral.
Example 95. Prove that 0 ≤ xy + yz + zx − 2xyz ≤ 7 , where x , y , z are non-negative real numbers for
27
which x + y + z = 1. (INMO 1984)
Solution We have to prove
7
0 ≤ xy + yz + zx − 2xyz ≤ …(i)
27
x + y + z =1 …(ii)
To prove LHS of inequality observe that (ii) implies
0 ≤ x , y , z ≤ 1 and so
xy + yz + zx − 2xyz = xy (1 − z ) + yz (1 − x ) + zx ≥ 0
(QAll terms of LHS are non –ve.)
In order to show the RHS of inequality.
1 1 1
Let x =a + ,y = b + ,z = c +
3 3 3
Now, (ii) gives
a+b+c=0 …(iii)
1 2
and − ≤ a, b, c ≤
3 3
After substituting b + c = − a , we get
7 1
xy + yz + zx − 2xyz = +
27 3
7 1
(ab + bc + ca − 6abc ) = + (bc − a 2 − 6abc ) …(iv)
27 3
Our original expression is symmetric in a , b , c .
Hence we may assume that a ≤ b ≤ c .
According to (iii), a , b , c cannot be all +ve or –ve at the same time so there are 2 possibilities.
1
(a) a ≤ b ≤ 0 ≤ c or (b) − ≤ a ≤ 0 ≤ b ≤ c
3
Now, bc − a 2 − 6abc is not +ve. In case (a) every expression is +ve, so 2 holds.
Now to prove (b), let us arrange (iv)
bc − a 2 − 6abc = bc − (b + c )2 − 6abc
= − (b − c )2 − 3bc (1 + 2a )
Now, as 1 + 2a > 0 and bc ≥ 0, the expression cannot be +ve, so (ii) holds.
www.pdfworld.in
www.pdfworld.in
Aliter
(ii) implies that at least one of x , y , z, let us say x is not greater than 1/3. Thus, 1 − 2x > 0 and so
xy + yz + zx − 2xyz = xy + yz (1 − 2x ) + zx > 0
Hence we proved LHS of (i)
To prove RHS. We consider 2 cases.
(a) One of x , y , z, let us say z is at least 1/2; z ≥ 1 / 2
(b) 0 ≤ x , y , z ≤ 1 / 2
In (a) case x + y = 1 − z we get
xy + yz + zx − 2xyz = z (x + y ) + xy (1 − 2z )
= z (1 − z ) + xy (1 − 2z )
1 7
≤ z (1 − z ) ≤ ≤
4 27
In (b) case, apply the following substitution
a = 1 − 2x , b = 1 − 2y , x = 1 − 2z
These values gives a , b , c ≥ 0, a + b + c = 1 …(v)
Substituting and applying (v), we get
1 + abc
xy + yz + zx − 2xyz = …(vi)
4
Now, AM-GM inequality gives
3
a + b + c  1
abc ≤   =
 3  27
Hence, (vi) implies
28 7
xy + yz + zx − 2xyz ≤ =
27 27
4
So, we proved (i) in every case.
Aliter
The problem involves the so called elementary symmetric polynomials of x , y , z that lead us the
following idea. Consider the polynomialf (t ) of degree (3)
f (t ) = (t − x )(t − y )(t − z ) = t 3 − t 2 (x + y + z ) + t (xy + yz + zx ) − xyz …(vii)
From (ii), we have
2f   = − + (xy + yz + zx ) − 2xyz
1 1 1
 2 4 2
xy + yz + zx − 2xyz = 2f   +
1 1
 2 4
0 ≤ 2f   + ≤
1 1 7
Now,
 2  4 27
≤ f   ≤
1 1 1
i.e., − …(viii)
8  2  216
Assume x ≥y ≥ z≥ 0
Choose x , y , z such that x ≥ 1 / 2, y ≤ 1 / 2, z ≤ 1 / 2 holds.
(vii) implies that f   ≤ 0.
1
 2
− f   =  x − 
1 1  1 − y   1 − z
Hence,   
 2  2 2  2 
www.pdfworld.in
www.pdfworld.in
Inequalities 319
Applying AM–GM inequality and using x ≤ 1

3 3
x − y − z + 1  2x − 1 
 1  2
  
2 ≤1
−f   ≤   = 
 2  3   3  8
   
f   ≥ −
1 1
⇒
 2 8
If we choose the values of the variables such that x ≤ 1 / 2 and so y ≤ 1 / 2, z ≤ 1 / 2 holds.
The previous method gives
3
 3 − (x + y + z )
0 ≤ f   ≤  − x 
1 1 1 − y  1 − z ≤  2 
     
 2  2  2  2  3
 
 
1
=
216
Hence proved.
Example 96. Let x 1 , x 2 , K , x n be real numbers satisfying x 12 + x 22 + K + x n2 = 1. Prove that for every
integer k ≥ 2 there are integers ai (i = 1, 2, K , n ) not all zero such that | ai | ≤ k − 1 for all i and
(k − 1) n
| a1x 1 + a 2x 2 + K + an x n | ≤ (INMO 1987)
kn − 1
Solution x 12 + x 22 + K + x n2 = 1 …(i)
(k − 1) n
| a1x 1 + a 2x 2 + .. + an x n | ≤ …(ii)
kn − 1
Applying AM–GM inequality for (i), yields
1 x 12 + x 22 + K + x n2 | x 1| + | x 2| + K + | x n |
= ≥
n n n
i . e. , | x 1| + | x 2| + K + | x n | ≤ n
Consider, now the k − 1 sequence of length n, none of them is constantly zero that are composed from
n
the numbers (0, 1, 2, K , k − 1).

Let one of them be (a1 , a 2 , K , an ). Set now the sign of ai in such a way that the product ai x i is not –ve
(i = 1, 2, K , n ), clearly
a1x 1 + K + an x n = | a1| | x 1| + | a 2| | x 2| + K + | an | x k |
≤ (k − 1)(| x 1| + | x 2| + K + | x n | )
≤ (k − 1) n …(iii)
(k − 1) n
Split the interval [0 , (k − 1) n ] into kn − 1 equal parts the length of each part is n . If any one of the
k −1
sums above is inside the first interval, then (ii) clearly holds for this sum and we are done. If there is no
such a sum, then there are at least two of them in some of the remaining kn − 2 intervals. If these sums
are
b1x 1 + b2x 2 + K + bn x n
and c1x 1 + c 2x 2 + K + cn x n [| bi | , | ci | ≤ k − 1]
www.pdfworld.in
www.pdfworld.in
Then obviously their difference cannot exceed the length of the interval
| b1x 1 + b2x 2 + K + bn x n − c1x 1 − c 2x 2 −cn x n |
= | (b1 − c1 )x 1 + (b2 − c 2 )x 2 + K + (bn − cn )x n |
(k − 1) n
≤
kn − 1
Q bi and ci are of same sign, | bi − ci | ≤ k − 1
Clearly, ai = bi − ci are integer of (ii).
Example 97. Let ABC be a triangle with sides a, b, c. Consider a ∆A1B1C1 with sides equal to a + b ,b + c ,
2 2
a
c+ . Show that
2
[A1B1C1 ] ≥ 9 / 4 [ABC ]
where [XYZ ] denotes area of ∆ XYZ. (INMO 2003)
b c a
Solution It is readily seen that a + , b + , c + can serve as the sides of a ∆. Heron's formula gives
2 2 2
16 [ABC ]2 = (a + b + c )(a + b − c )(b + c − a ) (c + a − b )
3
16 [A1B1C1 ]2 = (a + b + c )(−a + b + 3c ) (−b + c + 3a )(−c + a + 3b )
16
Now, a , b , c are the sides of a triangle, we may assume that
a = q + r ; b = r + p; c = p + q
p , q , r are +ve real numbers.
So, we have
[ABC ]2 16pqr
=
[A1B1C1 ]2 3(2p + q )(2q + r )(2r + p )
∴ It is enough to prove that
(2p + q )(2q + r )(2r + p ) ≥ 27pqr
for +ve real numbers p , q , r .
Now, AM–GM inequality gives
2p + q ≥ 3(p 2q )1/3,
2q + r ≥ 3(q 2r )1/3
and 2r + p ≥ 3(r 2p )1/3
If we multiply all these inequalities, we get
(2q + r )(2r + p )(2p + q ) ≥ 27pqr
Equality holds if p = q = r
i.e., when ∆ ABC is equilateral.
Example 98. The diagonal connecting two opposite vertices of a rectangular parallelopiped is 73. Prove
that if the square of the edges of the parallelopiped are integers, its volume does not exceed 120.
Solution Let a , b , c be the lengths of the edges of the parallelopiped.
We have a 2 + b 2 + c 2 = 73
By AM–GM inequality,
3
a2 + b2 + c 2   73 
3
a 2b 2c 2 ≤   = 
 3   3
www.pdfworld.in
www.pdfworld.in
Inequalities 321
Q a 2 , b 2 , c 2 are integers.
 73 3 
∴ a 2b 2c 2 ≤     = 14408
  3  
⇒ a 2b 2[73 − (a 2 + b 2 )] ≤ 14408
We need the largest integer N such that N ≤ 14408.
N has 3 factors whose sum is 73.
14401 and 14407 are primes.
14402 = 7201 × 2 ⇒ 7201 is a prime.
14403 = 4801 × 3 ⇒ 4801 is a prime.
14404 = 4 × 13 × 277
14405 = 43 × 67 × 5
14406 = 2 × 3 × 74
14408 = 1801 × 8
None of these numbers have 3 factors with sum = 73
Now, 14400 = 24 × 24 × 25
and 24 + 24 + 25 = 73
∴ Maximum volume is 14400 = 120
Example 99. Let x 1 , x 2 be the roots of the equation x 2 + px − 1 2 = 0 ,where x is unknown and p is a real
2p
parameter. Prove that x 14 + x 24 ≥ 2 + 2.
Solution We have x 1 + x 2 = − p
1
x 1x 2 = −
2p 2
1
Thus, x 12 + x 22 = (x 1 + x 2 )2 − 2x 1x 2 = p 2 +
p2
x 14 + x 24 = (x 12 + x 22 )2 − 2x 12x 22
2
 1 1 1
= p 2 + 2  − = p4 + +2
 p  2p 4 2p 4
For α ≥ 0, β ≥ 0, we have α + β ≥ 2 αβ
1  1 
So x 14 + x 24 = p 4 + + 2 ≥ 2 p4  4  + 2 = 2 + 2
2p 4  2p 
Hence proved.
Example 100. Prove that for every three non –ve numbers a , b , c the inequality a3 + b3 + c 3 + 6abc
1
≥ (a + b + c )3 holds and that equality occurs only when two of these numbers are equal and the third is
4
zero.
Solution Without loss of generality, assume a ≥ b ≥ c
1
a3 + b3 + c 3 + 6abc ≥ (a + b + c )3
4
⇔ 4(a3 + b3 + c 3 + 6abc ) ≥ (a3 + b3 + c 3 )
+ 3ab (a + b ) + 3bc (b + c ) + 3ca (c + a ) + 6abc
www.pdfworld.in
www.pdfworld.in
⇔ (a3 + b3 + c 3 + 6abc ) ≥ ab (a + b ) + bc (b + c ) + ca (c + a )
We have a −c≥b −c
⇒ a (a − c ) ≥ b (b − c )
⇒ a (a − c )(a − b ) ≥ b (b − c )(a − b )
⇒ a (a − b )(a − c ) + b (b − c )(b − a ) ≥ 0
Also, c − a ≤ 0 and c − b ≤ 0
and c (c − a )(c − b ) ≥ 0
a (a − b )(a − c ) + b (b − c )(b − a ) + c (c − a )
(c − b ) ≥ 0
Expanding LHS, we get
a3 + b3 + c 3 − ab (a + b ) − bc (b + c ) − ca (c + a ) + 3abc ≥ 0
Thus, ab (a + b ) + bc (b + c ) + ca (c + a )
≤ a3 + b3 + c 3 + 3abc ≤ a3 + b3 + c 3 + 6abc
Equality holds if and only if
a (a − b )(a − c ) + (b − c )b (b − a ) = 0
and c (c − a )(c − b ) = 0
Second equality holds if either c = 0 or c − a = 0
or c −b = 0
If c = 0, from first equality, we get
a 2 (a − b ) + b 2 (b − a ) = 0
⇒ (a − b )(a 2 + b 2 ) = 0
⇒ a =b
If c − a = 0 Q a ≥ b ≥ c it follows that a = b = c
Substituting in the equality we see that a = b = c = 0
If c = b then a (a − b ) (a − b ) = 0 ⇒ a = 0 or a = b
If a = b then a = b = c = 0.
This completes the proof.
Example 101. Find the largest integer n for which n 200 < 5300.
Solution n 200 < 5300

200
 3
⇒ n 200 <  5 2 
 
 
⇒ n < 53/ 2 = ( 5 )3 = 5 5
To find n such that n < 5 5 ≤ n + 1
⇔ n 2 ≤ 125 ≤ (n + 1)2
By inspection we get 11.
∴ Answer is 11.
www.pdfworld.in
www.pdfworld.in
Inequalities 323
Example 102. Let P be a point inside a ∆ABC. Let r1 , r2 , r3 denote the distances from P to the sides
BC , CA , AB respectively. If R is the circumradius of ∆ABC, show that
1
r1 + r2 + r3 < (a 2 + b 2 + c 2 )1/ 2
2R
where BC = a , CA = b and AB = C.
1 1 1
Solution r1 + r2 + r3 = ar1 ⋅ + br2 ⋅ + cr3 ⋅
a b c
Applying Cauchy-Schwartz inequality for the sets
A
 1 1 1 
{ ar1 , br2 , cr3 } and  , , ,
 a b c
F r3 E
we have r1 + r2 + r3 r2
P
1/ 2
≤ (ar1 + br2 + cr3 ) 1/ 2  1 + 1 + 1 r1
  …(i)
a b c 
B D C
Now, ar1 + br2 + cr3 = 2(∆PBC + ∆PCA + ∆PAB )
abc  abc 
= 2∆ABC =  for ∆ABC Area =  …(ii)
2R  4R 
Now, (ii) reduces (i) to

1/ 2 1/ 2
 abc   1 + 1 + 1
r1 + r2 + r3 ≤     …(iii)
 2R  a b c 
Again by Cauchy-Schwartz inequality, we get

(ab + bc + ca ) ≤ (a 2 + b 2 + c 2 )1/ 2
(b 2 + c 2 + a 2 )1/ 2 = a 2 + b 2 + c 2 …(iv)
Applying (iv) to (iii)

1/ 2 1/ 2
 abc   bc + ca + ab 
r1 + r2 + r3 ≤    
 2R   abc 
1/ 2 1/ 2
 abc  a2 + b2 + c 2 
≤   
 2R   abc 
1
∴ r1 + r2 + r3 ≤ (a 2 + b 2 + c 2 )1/ 2
2R
Example 103. Let a , b , c be +ve real numbers such that a + b + c ≥ abc . Prove that
a 2 + b 2 + c 2 ≥ 3 abc .
Solution Assume that a 2 + b 2 + c 2 < 3 abc

3
3 a 2b 2c 2 ≤ a 2 + b 2 + c 2 < 3 abc
It follows that abc > 3 3 …(i)
www.pdfworld.in
www.pdfworld.in
Also, by given condition on a , b , c

a 2b 2c 2 (a + b + c )2
≤ ≤ a 2 + b 2 + c 2< 3 abc
3 3
⇒ abc < 3 3 …(ii)
A contradiction to (i)
∴ a 2 + b 2 + c 2 ≥ 3 abc .
Example 104. Let k and n1 < n2 < K < nk be odd +ve integers, show that
n12 − n22 + n32 − n42 + K + nk2 ≥ 2k 2 − 1.
Solution Case k = 1 is clearly true.

We prove the result by induction. Assume that
n12 − n22 + n32 − n42 + K + nk2 ≥ 2k 2 − 1 …(i)
Now, we would like to prove
n12 − n22 + n32 − n42 + K + nk2
nk2 + 1 + nk2 + 2 ≥ 2(k + 2)2 − 1 = 2k 2 − 1 + 8k + 8
It is enough to prove that

nk2 + 2 − nk2 + 1 ≥ 8k + 8 …(ii)
Now, (nk + 2 − nk + 1 ) ≥ (nk + 2 + nk + 1 )

and nk ≥ 2k − 1
∴ We get (nk + 2 − nk + 1 )(nk + 2 + nk + 1 )
= nk2 + 2 − nk2 + 1 ≥ 2 [(2k + 1) + (2k + 3)] = 8k + 8
This proves (ii) and hence result is true for k + 1.
Example 105. If a , b be +ve numbers and p , q be rational numbers ( p > 1) such that 1 + 1 = 1. Show that
p q
p q
a b
ab ≤ + .
p q
1 1
Solution Q p > 1 and + =1
p q
∴ q>1
α α′
Let p = ,q =
β β′
where α , β , α ′ , β ′ are +ve integers.
ap b q aα / β b α ′ / β′
Then + = +
p q (α / β ) (α ′/β ′ )
β a α/ β β ′ b α ′ / β ′
= +
α α′
α ′ βa α/ β + αβ ′ b α ′/ β ′
= ≥ (a αα ′b αα ′ )1/ αα ′ > ab
αα ′
ap b q
∴ ab ≤ +
p q
www.pdfworld.in
www.pdfworld.in
Inequalities 325
Example 106. If a + b + c = n, where a , b , c are +ve integers. Prove that

1 1 1
(a a ⋅ bb ⋅ c c )n + (ab ⋅ bc ⋅ c a )n + (ac ⋅ b a ⋅ c b )n ≤ n.
Solution Let a a = (a ⋅ a K a times),

bb = (b ⋅ b K b times)
and c c = (c ⋅ c K c times)
Now, using AM–GM inequality
(a + a K a times) + (b + b K b times) 1
+ (c + c ... c times)
≥ (a a ⋅ bb ⋅ c c )a + b + c
a+b+c
But a + b + c =n
1
a2 + b2 + c 2
∴ ≥ (a abbc c )n …(i)
n
Similarly, ab = (a ⋅ a ⋅ a K b times) , bc = (b ⋅ b ⋅ b K c times)
and c a = (c ⋅ c ⋅ c K a times)
Again, using AM–GM inequality
(a + a K b times) + (b + b K c times) 1
+ (c ⋅ c ⋅ c K a times)
≥ (ab ⋅ bc ⋅ c a )a + b + c
b+c+a
1
(ab + bc + ca )
≥ (ab ⋅ bc ⋅ c a )n …(ii)
n
Similarly, we prove
1
ab + bc + ca
≥ (ac ⋅ b a ⋅ c b )n …(iii)
n
1 1 1
(a 2 + b 2 + c 2 + 2ab + 2bc + 2ca )
≥ (a a ⋅ bb ⋅ c c )n + (ab ⋅ bc ⋅ c a )n + (ac ⋅ b a ⋅ c b )n
n
1 1 1
(a + b + c )2
⇒ ≥ (a a ⋅ bb ⋅ c c )n + (ab ⋅ bc ⋅ c a )n + (ac ⋅ b a ⋅ c b )n
n
1 1 1
n2
⇒ ≥ (a a ⋅ bb ⋅ c c )n + (ab ⋅ bc ⋅ c a )n + (ac ⋅ b a ⋅ c b )n
n
1 1 1
or (a a ⋅ bb ⋅ c c )n + (ab ⋅ bc ⋅ c a )n + (ac ⋅ b a ⋅ c b )n ≤ n
Hence proved.
Example 107. A triangle has sides of length a , b , c and altitudes of length ha , hb , hc respectively. If
a ≥ b ≥ c , show that a + ha ≥ b + hb ≥ c + hc .
1 1 1 A
Solution Area of ∆ = a ⋅ ha = b ⋅ hb = c ⋅ hc
2 2 2
2∆
Now, ha =
a c b
ha
2∆
hb =
b
2∆
hc =
c B a C
www.pdfworld.in
www.pdfworld.in
Let us assume
a + ha ≥ b + hb
a − b ≥ hb − ha
2∆ 2∆
a −b≥ −
b a
a − b ≥ 2∆  − 
1 1
b a 
a − b 
a − b ≥ 2∆  
 ab 
2∆ 
(a − b )  1 − ≥0
 ab 
(a − b )(ab − 2∆ ) ≥ 0
1
Area of ∆ = ab sin C
2
2∆ = ab sin C
2∆ = ab sin C ≤ ab ∴ 2∆ ≤ ab
ab − 2∆ ≥ 0 ∴ (a − b ) ≥ 0
which is true as it is given a ≥ b
Similarly, we can prove other.
Example 108. If p , q , r , s are the sides of a quadrilateral. Find the minimum value of
p2 + q2 + r 2
.
s2
Solution We have D
AB = p r
C
BC = q
CD = r s
AD = s q
We know that
( p − q )2 + (q − r )2 + (r − p )2 ≥ 0 A p B
⇒ 2( p + q + r ) ≥ 2( pq + qr + rp )
2 2 2
⇒ 3( p 2 + q 2 + r 2 ) ≥ ( p 2 + q 2 + r 2 ) + 2( pq + qr + rp )
[on adding p 2 + q 2 + r 2 to both sides]
⇒ 3( p 2 + q 2 + r 2 ) ≥ ( p + q + r )2
[Qsum of any three sides of a quadrilateral is greater than fourth one]

⇒ 3( p 2 + q 2 + r 2 ) ≥ ( p + q + r )2 > s 2
p2 + q2 + r 2 1
⇒ 2
>
s 3
p2 + q2 + r 2 1
∴ Minimum value of is .
s2 3
www.pdfworld.in
www.pdfworld.in
Inequalities 327
Example 109. If n is a +ve integer > 1. Prove that

n −1
2n + n ⋅ 6 2 < 3n .
−1 −2 −3 2 −1
Solution Q an − bn = (a − b )(an + an b + an b + K+ bn )
n −1 n −2 n −3 n −1
∴ 3 −2 =3
n n
+3 −2+ 3 ⋅2 + K+ 2
2
…(i)
QAM > GM for distinct +ve numbers
−1 −2 −1 1
3n + 3n ⋅ 2 + K + 2n −1 −1 n
∴ > [(3 ⋅ 32 K 3n )(2 ⋅ 22 K 2n )]
n
n −1 n −1 n −1
=3 2 ⋅2 2 = 6 2 …(ii)
n −1
3n > 2n + n ⋅ 6 2
Example 110. If A > 0, B > 0 and A + B = π / 3, find maximum value of tan A ⋅ tan B.
Solution Let y = tan A tan B
Q AM ≥ GM
tan A + tan B
∴ ≥ tan A tan B
2
tan A + tan B
Q tan (A + B ) =
1 − tan A tan B
π tan A + tan B
tan =
3 1 − tan A tan B
⇒ 3 (1 − tan A tan B ) = tan A + tan B
3 (1 − tan A tan B )
∴ ≥ tan A tan B
2
∴ 3 (1 − y ) ≥ 2 y (Q tan A tan B = y )
∴ 3(1 − y ) ≥ 4y
2
⇒ 3y − 10y + 3 ≥ 0
2
∴ y ≤ 1 / 3 or y ≥ 3
But A + B = π / 3 and A > 0, B > 0
∴ y ≤ 1/ 3
1
∴ Maximum value of tan A ⋅ tan B =
3
Example 111. Let a , b , c be the length of the sides of a triangle and r be its radius. Show that
a2 + b2 + c 2
r< .
3(a + b + c )
Solution We know for a ∆ABC

Area ∆
r = =
Semi-perimeter a + b+c
2
www.pdfworld.in
www.pdfworld.in
⇒ r (a + b + c ) = 2∆ = bc sin A = ca sin B = ab sin C

⇒ 3r (a + b + c ) = 6∆ = bc sin A + ca sin B + ab sin C …(A)
Now, bc sin A < bc …(i)
ca sin B < ca …(ii)
ab sin C < ab …(iii)
From (i), (ii) and (iii), we have
bc sin A + ca sin B + ab sin C < bc + ca + ab
⇒ 3r (a + b + c ) < bc + ca + ab …(B)
b2 + c2
bc ≤
2
c 2 + a2
ca ≤
2
a2 + b2
ab ≤
2
2 2
⇒ bc + ca + ab ≤ (a + b 2 + c 2 ) …(C)
2
From (B) and (C), we get
a2 + b2 + c 2
3r (a + b + c ) < (a 2 + b 2 + c 2 ) ⇒ r<
3(a + b + c )
Example 112. In any acute angle ∆ABC, show that

A B C 3 3 3 3
(i) cos cos cos ≤ or sin A + sin B + sin C ≤ (ii) tan A + tan B + tan C ≥ 3 3.
2 2 2 8 2
Solution (i) We have to show
A B C 3 3
cos cos cos ≤ …(i)
2 2 2 8
If A + B + C = π, we know that
A B C
sin A + sin B + sin C = 4 cos cos cos …(ii)
2 2 2
Using (ii) in (i), it is sufficient to show that
1 3 3
(sin A + sin B + sin C ) ≤
4 8
3 3
sin A + sin B + sin C ≤
2
3 3
or sin A + sin B + sin C ≤
2
(ii) If A + B + C = π, we know that
tan A + tan B + tan C = tan A ⋅ tan B ⋅ tan C …(i)
Consider three non –ve quantities
tan A , tan B , tan C
Q AM ≥ GM
1
∴ (tan A + tan B + tan C ) ≥ (tan A + tan B + tan C )1/3
3
∴ tan A + tan B + tan C ≥ 3(tan A + tan B + tan C )1 /3
www.pdfworld.in
www.pdfworld.in
Inequalities 329
Cubing both sides, we get

(tan A + tan B + tan C ) 3 ≥ 27
∴ tan A + tan B + tan C ≥ 3 3
Example 113. In a rectangular parallelopiped, the length of the edges are a1 , a2 , a3 and length of the
diagonal is d. Prove that
d < a1 + a 2 + a3 < 2d .
Solution From
d < a12 + a 22 + a3 < a1 + a 2 + a3 a3
Also, (a1 + a 2 + a3 )2 − (a1 + a 2 − a3 )2 = 4(a1 + a 2 ) a3 d
so that (a1 + a 2 + a3 )2 ≥ 4(a1 + a 2 ) a3 a2

a1
Similarly, (a1 + a 2 + a3 )2 ≥ 4(a 2 + a3 ) a1
a12 + a22
and (a1 + a 2 + a3 )2 ≥ 4(a3 + a1 ) a 2
Adding these inequalities, we get
3(a1 + a 2 + a3 )2 ≥ 8 (a1a 2 + a 2a3 + a3a1 )
8
(a1 + a 2 + a3 )2 ≥ (a1a 2 + a 2a3 + a3a1 )
3
4d 2
Adding to both sides, where
3
d 2 = a12 + a 22 + a32, we get
4d 2 4
+ (a1 + a 2 + a3 )2 ≥ [(a12 + a 22 + a32 ) + 2(a1a 2 + a 2a3 + a3a1 )
3 3
4
= (a1 + a 2 + a3 ) 2
3
4d 2 1
or ≥ (a1 + a 2 + a3 )2
3 3
2d ≥ a1 + a 2 + a3 or a1 + a 2 + a3 < 2d
Equality sign hold if and only if
a1 + a 2 − a3 = 0 ; a 2 + a3 − a1 = 0
or a3 + a1 − a 2 = 0
i.e., if a1 + a 2 + a3 = 0
so we get d < a1 + a 2 + a3 < 2d
Example 114. In any ∆ABC. Prove that

sin A
Σ ≥ 3.
A, B, C sin B + sin C − sin A
a
Solution Let sin A = ,
2R
b C
sin B = and sin C =
2R 2R
Using the above concept, we get the desired result.
www.pdfworld.in
www.pdfworld.in
Example 115. Show that if A1 , A 2 , K , An are the angles of a convex polygon, then
2π
sin A1 + sin A 2 + K + sin An ≤ n sin
n
A1 A2 An π
and tan + tan + K + tan ≥ n cot .
2 2 2 n
Solution QThe sum of exterior angles of a convex polygon is 2π, the sum of the interior angles is (n − 2)π.
If 0 < x < π, then sin x a concave function of x so that sin A1 + sin A 2 + K + sin An is a concave function
(n − 2)π
of A1 , A 2 , K , An whose maximum value occurs when A1 = A 2 = K = An =
n
so that maximum value of sin A1 + K + sin An is
(n − 2)π
n sin
n
(n − 2)π 
∴ sin A1 + sin A 2 + K + sin An ≤ n sin  
 n 
2π
= n sin
n
, tan 2 , K , tan  n  are all convex functions so that tan 1
A1 A A A
SimilarlyQA1 , A 2 , K , An are all < π, tan
2 2  2  2
A A
+ tan 2 + K + tan n is a convex function.
2 2
Its maximum value occurs when
(n − 2)π
A1 = A 2 = K = An = so that
2n
A A A (n − 2)π
tan 1 + tan 2 + K + tan n ≥ n tan
2 2 2 2n
 π π
= n tan  − 
 2 2n 
π
= n cot
2n
Example 116. Prove that if p , q , k are all> 1,a1 , a2 , a3 , K , an are all +ve numbers whose sum is unity, then
n (np + nq )k
Σ (aip + aiq )k ≥ .
i =1 (n )kp + kq − 1
Solution x p is convex, if p > 1

x q is convex, if q > 1
x p + x q is convex, if p , q > 1
Let f (x ) = (x p + x q ) k
f ′ (x ) = k (x p + x q )k − 1 ( px p − 1 + qx q −1
)
q k−2
f ′ ′ (x ) = k (k − 1)(x + x )
p
[ p ( p − 1)x p − 2 + q (q − 1)x q −2
]
If p , q , k are all >1, then
f ′ ′ (x ) > 0 and f (x ) is a convex function.
f (a1 ) + f (a 2 ) + K + f (an ) (a + a 2 + K + an )
≥f 1
n  n 
www.pdfworld.in
www.pdfworld.in
Inequalities 331
but a1 + a 2 + K + an = 1 so that
f (a1 ) + f (a 2 ) + K + f (an )
≥ f  
1
n x 
Put f (x ) = (x p + x q )k , we get
k
1 n
Σ (ap + aiq )k ≥  p + q 
1 1
n i =1 i n n 
n (np + nq )k
or Σ (aip + aiq )k ≥
i =1 n (p + q ) k − 1
Equality sign holds only when
1
a1 = a 2 = K = an =
n

Example 117. For a acute angle ∆ABC  0, π  . Prove that
 2 
3 3
sin A sin B sin C ≤ .
8
Solution Let f (x ) = log sin x
f ′ (x ) = cot x
f ′ ′ (x ) = − cosec 2x
f ′ ′ (x ) < 0
+ + C  f (A ) + f (B ) + f (C )
f 
A B
≥
 3  3
π log sin A + log sin B + log sin C
log sin   ≥
3 3
 3
3 log   ≥ log(sin A sin B sin C )
 2 
3 3
log   ≥ log(sin A sin B sin C )
 8 
3 3
∴ sin A sin B sin C ≤
8
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1. If x > a, then prove that 1 + 9 + 92 + K + 999
16. If A =
x 3 + 13a 2x > 5ax 2 + 9a3. 1 + 9 + 92 + K + 9100
2. If x , y > 0, then show that 1 + 7 + 72 + K + 799
and B= .
x n + y n > x n − 1y + xy n −1
. 1 + 7 + 72 + K + 7100
3. Prove that a 2 (a − b )(a − c ) + b 2 (b − c )(b − a ) Prove A < B.
+ c 2 (c − a ) (c − b ) ≥ 0
17. Without using AM–GM, prove
4. If a , b , c > 0, then prove that
a+b+c 3
a + b + c > ab + bc + ac . ≥ abc .
3
5. Without using AM–GM show that 18. Let a , b , c , …, l be n real +ve numbers, p and q
a 8 + b 8 + c 8 ≥ a 2b 2c 2 (a 2 + b 2 + c 2 ). be also two real numbers. Prove that if p and q
a 4b 4 b 4c 4 c 4a 4 are of same sign, then
6. Prove + + > (abc ) (a + b + c )
c 4
a b4 4 n (ap + q + bp + q + K + l p + q )
without using AM–GM ≥ (ap + bp + K + l p ) (aq + bq + K + l q )
x2 1
7. If x is any real number show ≤ . If p and q are of different signs, then
1 + x4 2
n (ap + q + bp + q + K + l p + q )
8. IfA + B + C = π , show that
≤ (ap + bp + K + l p ) (aq + bq + K + l q )
tan 2 A + tan 2 B tan 2 B + tan 2 C
+
tan A + tan B tan B + tan C 19. Prove that
tan C + tan A
2 2 a+b
+ ≥ tan A tan B tan C. (i) ≥ ab (a , b > 0)
tan C + tan A 2
9. If x , y , z are positive, show that 1 (a − b )2 a + b 1 (a − b )2
(ii) ⋅ ≤ − ab ≤ ⋅
x 3 + y 3 y 3 + z3 z3 + x 3 8 a 2 8 b
+ + ≥ (xy + yz + zx ).
x +y y + z z+x if a ≥ b.
10. If m > 1 and n ∈ N . Prove that 20. If x + y = a, then without AM–GM, show that
m
n + 1
> n  2 2 2
1m
+2 m
+3 m
+ K+ n m
 . x + 1  + y + 1  ≥ 1 a + 4  .
 2       
 x  y 2 a
11. If m > 1 and n ∈ N . Prove that
+1 21. Prove that
1m + 3m + 5m + K + (2n − 1)m > nm .
3
a3 + b 3  a + b 
12. If a , b , c ≥ 0, then show that ≥  , a > 0, b > 0.
2  2 
ab + bc + ca ≥ a bc + b ca + c ab.
13. Prove that if a , b , c , d > 0, then 22. If a1 , a 2 , .. , an are real numbers show that
(cos a1 + cos a 2 + K + cos an )2
(a + c )(b + d ) ≥ ab + cd .
14. Find, which is greater x 3 + 2y 3 or 3xy 2 + (sin a1 + sin a 2 + K + sin an )2 ≤ n 2.
15. Let a , b , c and a + b − c , a + c − b , b + c − a be 23. For n ∈ N , n > 1 show that

+ve. Prove that 1 1 1 1
+ + + K + 2 > 1.
abc ≥ (a + b − c )(a + c − b )(b + c − a ). n n+1 n+2 n
www.pdfworld.in
www.pdfworld.in
Inequalities 333
1 1 1 37. Show that

24. Prove that + + .. + >1
n+1 n+2 3n + 1 bm + c m c m + am am + b m
m −1
+ m −1
+
25. If a , b , c are unequal and +ve. Prove that (b + c ) (c + a ) (a + b )m − 1
bc ca ab 1 a+b+c
+ + < (a + b + c ). >
b+c c+a a+b 2 2m − 2
where m does not lie between 0 and 1.
26. Show that (a + b + c + d )
38. Show that
(a3 + b3 + c 3 + d 3 ) > (a 2 + b 2 + c 2 + d 2 )2.
b2 + c 2 c 2 + a2 a2 + b2
+ + >a + b + c
27. Prove that (a1b1 + a 2b2 + a3b3 ) b+c c+a a+b
 a1 a 2 a3  39. Find the maximum value of (8 − x )3 (x + 6)4 , if x
 + +  > (a1 + a 2 + a3 ) .
2
 b1 b2 b3  lies between −6 and 8.
28. Show that 40. Find the maximum value of (7 − x )4 (2 + x )5 ,
3 3 3 3 when x lies between −2 and 7.
+ + +
b+c+d c+d +a d +a+b a+b+c 1 1
>
16
. 41. Find the greatest value of x 2 (1 − x )3 when x lies
a+b+c+d between 0 and 1.
29. If a , b , c , d are +ve real numbers, prove that (5 + x )(2 + x )
42. Find minimum value of .
bcd cda dab abc 1+x
+ 2 + 2 + 2 > a + b + c + d.
a2 b c d 43. If 5x + 12y = 60, find the minimum value of
30. Prove that x 2 + y 2.
(a + b )(b + c )(a + c ) ≥ 8abc (a , b , c > 0).
9x 2 sin 2 x + 4
44. Find the minimum value of .
31. Let a , b , c be +ve integers, prove that x sin x
a b c
a +b +c a +b +c a +b +c 1
a ⋅b ⋅c ≥ (a + b + c ). 45. If a , b are +ve real numbers such that
3 a 2 + b 2 = 1 show that
32. Prove that (1 + α ) n > 1 + α λ (α is any the 2
a + b + 
1 
number and λ > 1 is rational).  ≥2+ 2.
 ab 
1
33. Show that (1 + α ) n < real, λ rational and 46. For a > b > 0.n is a +ve integer > 1, show that for
1 − αλ
k ≥ 0 n an + kn − n bn + kn ≤ a − b.
+ve, αλ < 1.
34. If x , y , z are unequal +ve quantities. Prove that 47. If a , b , c are unequal +ve rational number,
a +b +c
x+ y + z a + b + c
 x 2 + y 2 + z2  show that   < a abb c c
  > x xy y z z  3 
 x +y + z  a +b +c
x+ y + z a2 + b2 + c 2 
x + y + z <  .
>   .  a+b+c 
 3 
a +b +c 48. If a triangle having base a and ratio of other
 bc + ca + ab 
35. Prove that   two sides is r < 1, show that altitude of triangle
 a+b+c  ar
is .
> (bc )a (ca )b (ab ) c . 1 − r2
36. If x , y , z are +ve real numbers such that 49. Suppose (x 1 , x 2 , K , x n ,... ) is a sequence of
x 3y 2z 4 = 7, show that positive real numbers such that
1 x 1 ≥ x 2 ≥ x 3 ≥ K ≥ x n ... , and for all n
2x + 5y + 3z ≥ 9  7  .
525 9 x1 x 4 x 9 x2
 2  + + + K + n ≤ 1.
1 2 3 n
www.pdfworld.in
www.pdfworld.in
Show that for all k the following inequality is 55. Let x and y be positive real numbers such that
satisfied y 3 + y ≤ x − x 3. Prove that
x1 x 2 x3
+ +
x
+ .... + k ≤ 3. (i) y < x < 1;
1 2 3 k (ii) x 2 + y 2 < 1. (RMO 2004)
(RMO 2000)
56. Let a , b , c be three positive real numbers such
50. If x , y , z are the sides of a triangle, then prove
that a + b + c = 1. Let
that
λ = min {a3 + a 2bc , b3 + ab 2c , c 3 + abc 2}.
| x 2 ( y − z ) + y 2 (z − x ) + z 2 (x − y )| < xyz.
(RMO 2001)
Prove that the roots of the equation
x 2 + x + 4λ = 0 are real. (RMO 2005)
51. Find all integers a , b , c , d satisfying the
following relations. 57. If a , b , c are three real numbers such that
(i) 1 ≤ a ≤ b ≤ c ≤ d ;
| a − b| ≥ | c | ,| b − c | ≥ | a| , | c − a| ≥ | b| ,
(ii) ab + cd = a + b + c + d + 3. (RMO 2002) then prove that one of a , b , c is the sum of the
52. For any natural number n > 1, prove the other two. (RMO 2005)
inequality (RMO 2002)
58. If a , b , c are three positive real numbers, prove
1 1 2 3
< + + that
2 n2 + 1 n2 + 2 n2 + 3 a2 + 1 b2 + 1 c 2 + 1
+ + ≥ 3. (RMO 2006)
n 1 1 b+c c+a a+b
+ K+ 2 < + .
n + n 2 2n
59. Prove that
53. Suppose the integers 1, 2, 3, ... , 10 are split (i) 5 < 5 + 3
5+ 4
5;
into two disjoint collections a1 , a 2 , a3 , a 4 , a5
(ii) 8 > 8+ 3
8+ 4
8;
and b1 , b2 , b3 , b4 , b5 such that
(iii) n > n + 3
n + 4
n for all integers n ≥ 9.
a1 < a 2 < a3 < a 4 < a5, b1 > b2 > b3 > b4 > b5.
(RMO 2007)
(i) Show that the larger number in any pair
{a j , b j }, 1 ≤ j ≤ 5, is at least 6. 60. Prove that there exist two infinite sequences
(ii) Show that an n ≥ 1 and bn n ≥ 1 of positive integers such
| a1 − b1| + | a1 − b2| + | a3 − b3| + | a 4 − b4 | that the following conditions hold
+ | a5 − b5| = 25 simultaneously
for every such partition. (RMO 2002) (i) 1 < a1 < a 2 < a3 < ... ;
54. Let a , b , c be three positive real numbers such (ii) an < bn < an2 , for all n ≥ 1;
that a + b + c = 1. Prove that among the three
numbers a − ab , b − bc , c − ca there is one (iii) an − 1 divides bn − 1, for all n ≥ 1;
which is at most 1/4 and there is one which is (iv) an2 − 1 divides bn2 − 1, for all n ≥ 1.
at least 2/9. (RMO 2003) (RMO 2008)
Level 2
1. If a , b , c are real numbers such that 3. Prove that for every three non-negative
numbers a, b, c , the inequality
a + b + c =1
2 2 2
1
a3 + b3 + c 3 + 6abc ≥ (a + b + c ) 3 holds and
prove the inequalities 4
that equality occurs only when two of these
1
− ≤ ab + bc + ca ≤ 1 numbers are equal and the third is zero.
2
4. Show that for all natural numbers n > 1 the
2. If a , b and c are integers with c > 0, then inequality
(n − 1)
 2a  +  2b + 1  ≥  a  +  b  +  a + b + 1  .
n
+1
 1 + (n + 1) n   1 + nn 
 c   c   c   c   c    >  is valid.
 n+2   n+1
www.pdfworld.in
www.pdfworld.in
Inequalities 335
5. Show that if x , y , z > 0, 14. Prove that

1 1 1 1 1
 1 1 1  1+ + + + + K+ 2 <2
(xy + yz + zx )  + +  4 9 16 25 n
 (x + y )2 ( y + z )2 (z + x )2 
for any n.
9
≥ 15. Let a , b , c be three real numbers such that
4 1 ≥ a ≥ b ≥ c ≥ 0. Prove that if λ is a root of the
6. Find the largest constant k such that cubic equation x 3 + ax 2 + bx + c = 0 (real or
kabc complex), then |λ | ≤ 1. (INMO 2000)
≤ (a + b )2 + (a + b + 4c )2
a+b+c 16. If a , b , c are positive real numbers such that
abc = 1, prove that
for all a , b , c > 0
ab + c bc + a c a + b ≤ 1. (INMO 2001)
7. Given positive real numbers a , b and c such
that a + b + c = 1, show that 17. Let x , y be positive reals such that x + y = 2.
a abbc c + abbc c a + ac b ac b ≤ 1. Prove that
x 3y 3 (x 3 + y 3 ) ≤ 2. (INMO 2002)
8. If a , b , c , x , y and z are real and
a 2 + b 2 + c 2 = 25, x 2 + y 2 + z 2 = 36, and 18. Let ABC be a triangle with sides a , b , c .
a+b+c Consider a ∆A1B1C1 with sides equal to
ax + by + cz = 30, compute . b c a
x +y + z a + ,b + , c + . Show that
2 2 2
9. Prove that 9
[A1B1C1 ] ≥ [ABC ],
1 4
n k + 1 < 1k + 2k + 3k
k+1 where [XYZ ] denotes the area of the ∆XYZ.
k +1
(INMO 2003)
+ K + n <  1 + 
1 1
k
nk + 1
 n k+1 19. Let α and β be positive integers such that
43 α 17
(n and k are arbitrary integers). < < .
197 β 77
10. There are real numbers a , b , c such that Find the minimum possible value of β.
a ≥ b ≥ c . Prove that
(INMO 2005)
a2 − b2 c 2 − b2 a2 − c 2
+ + ≥ 3a − 4b + c . 20. (i) Prove that if n is a positive integer such
c a b
that n ≥ 40112 , then there exists an integer
1 1 1 1
11. Prove that 1 < + + K+ <1 l such that n < l 2 <  1 +
1 
 n.
1001 1002 3001 3  2005 
12. Let a1 , a 2 , a3 , K , an be real numbers all greater (ii) Find the smallest positive integer M for
than 1 and such that | a k − a k +1| < 1 for which whenever an integer n is such that
1 ≤ k ≤ n − 1. n ≥ M , there exists an integer l , such that
n < l 2 <  1 +
Show that 1 
 n. (INMO 2006)
a1 a 2 a3 a a  2005 
+ + + K + n −1 + n < 2n − 1.
a 2 a3 a 4 an a1
21. If x , y , z are positive real numbers, prove that
13. For any natural number n, prove the following (x + y + z )2 ( yz + zx + xy )2 ≤ 3 ( y 2 + yz + z 2 )
inequality (z 2 + zx + x 2 ) (x 2 + xy + y 2 ). (INMO 2007)
1  1 1 1 
1 + + + K +  22. Let a , b , c be positive real numbers such that
n+1 3 5 2n − 1 
a3 + b3 = c 3. Prove that
≥  + + ... +
1 1 1 1
. a 2 + b 2 − c 2 > 6 (c − a ) (c − b ). (INMO 2009)
n 2 4 2a 
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1. (x 3 + 13a 2x ) − (5ax 2 + 9a3 ) which is non –ve as each term of RHS is non
–ve.
= (x − a )(x − 4ax + 9a )
2 2
Q a ≥ b ≥ c and a , b , c be non –ve
= (x − a )[(x − 2a )2 + 5a 2 ]
Thus, a 2 (a − b )(a − c ) + b 2 (b − c )(b − a )
= −ve
+ c 2 (c − a )(c − b ) ≥ 0
if x > a
Hence, (x 3 + 13a 2x ) > (5ax 2 + 9a3 ) 4. a = ( a )2
−1 −1 b = ( b )2
2. (x n + y n ) − (x n y + xy n )
−1 −1 c = ( c )2
⇒ (x n − x n y ) + ( y n − xy n )
−1 −1 ∴ Now, let a = u
⇒ xn (x − y ) − y n (x − y )
b =v
−1 −1
⇒ (x n − yn )(x − y )
c =w
−2 −3 −2
(x − y )2 (x n + xn y + K + yn )
u 2 + v 2 + w 2 > uv + vw + uw
(x − y )2 > 0
a + b + c > ab + bc + ac
Q x and y are +ve.
(x n − 2 + x n − 3y + K + y n −2
)> 0 5. a 8 + b 8 + c 8 ≥ a 2b 2c 2(a 2 + b 2 + c 2 )
−1 −1
∴ x n + yn > x n y + xy n (a 4 )2 + (b 4 )2 + (c 4 )2 ≥ a 4b 4 + b 4c 4 + a 4c 4
3. QThe given inequality is symmetrical in a , b , c . …(i)

(a b ) + (b c ) + (a c ) ≥ (a b )(b c )
2 2 2 2 2 2 2 2 2 2 2 2 2
Without loss of generality assume a ≥ b ≥ c we
have + (b 2c 2 )(a 2c 2 ) + (a 2b 2 )(a 2c 2 )
a 2 (a − b )(a − c ) + b 2 (b − c )(b − a ) ∴ a 4b 4 + b 4c 4 + a 4c 4 ≥ a 2b 2c 2 (a 2 + b 2 + c 2 )
+ c (c − a )(c − b )
2
…(ii)
= (a − b ){a 2 (a − c ) − b 2 (b − c )} From (i) and (ii), we get
+ c 2 (c − a )(c − b ) a 8 + b 8 + c 8 ≥ a 2b 2c 2 (a 2 + b 2 + c 2 )
= (a − b ){(a3 − b3 ) − c (a 2 − b 2 )} 6. Let
ab
= u,
bc
= v,
ca
=w
+ c (a − c )(b − c )
2 c a b
We know that
= (a − b ){(a − b )(a 2 + ab + b 2 ) − c (a − b )
u 4 + v 4 + w 4 > uvw (u + v + w ) …(i)
(a + b )} + c 2 (a − c )(b − c )
+ w 4 > abc 
ab bc ca 
u4 + v 4 + + 
= (a − b )2 {a 2 + ab + b 2 − c (a + b )}  c a b 
+ c 2 (a − c )(b − c )  a 2b 2 + b 2c 2 + c 2a 2 
u 4 + v 4 + w 4 > abc  
= (a − b )2 {a 2 + ab + b 2 − ca − bc }  abc 
+ c 2 (a − c )(b − c ) u 4 + v 4 + w 4 > a 2b 2 + b 2c 2 + c 2a 2 …(ii)
= (a − b )2 {a (a − c ) + b (b − c ) + ab} Now, we know that
+ c 2 (a − c )(b − c ) a 2b 2 + b 2c 2 + c 2a 2 > abc (a + b + c ) …(iii)
www.pdfworld.in
www.pdfworld.in
Inequalities 337
m
From (ii) and (iii), we have 1m + 3m + 5m + K + (2n − 1)m  n 2 
⇒ > 
u +v
4 4
+ w > abc (a + b + c ).
4
n n 
+1
Substitute the value of u , v , w we get the ⇒ 1m + 3m + 5m + K + (2n − 1)m > nm
desired result.
12. ∀ x , y , z ≥ 0
0 1 1
7. If x = 0, then result ≤ i.e., 0 ≤ is true. x 2 + y 2 y 2 + z2 z2 + x 2
1+ 0 2 2 x 2 + y 2 + z2 = + +
2 2 2
For x ≠ 0, x 2 is +ve.
≥ xy + yz + zx
x2
Let S = for x ≠ 0 Let x = ab , y = bc , z = ca
1 + x4
1 Put the value of x , y , z, we get
We can write S =
1
+ x2 ab + bc + ca ≥ a bc + b ca + c ab
x2
13. a , b , c , d > 0 (given)
Max S =
1 Qx 2 + 1 ≥ 2
  Then (a + c )(b + d ) = ab + ad + bc + cd

min  2 + x 2  x2
1
x  ≥ ab + 2 adbc + cd
1 = ( ab + cd )2
S≤
2
x2 1 or (a + c )(b + d ) ≥ ab + cd
∴ ≤
1+x 4
2 14. 3xy 2 − (x 3 + 2y 3 ) = xy 2 + 2xy 2 − x 3 − 2y 3
8. Let tan A = u , tan B = v , tan C = w , we have = (xy 2 − x 3 ) + (2xy 2 − 2y 3 )
u2 + v 2 v 2 + w 2 w 2 + u2 = x ( y 2 − x 2 ) + 2y 2 (x − y )
+ + ≥u + v + w
u+v v +w u+w = (x − y )[−x ( y + x ) + 2y 2 ]
⇒ u + v + w = tan A + tan B + tan C = (x − y )[2y 2 − yx − x 2 ]
Now, when A + B + C = π
= (x − y )(2y + x )(y − x )
tan A + tan B + tan C = tan A tan B tan C
= − (x − y )2 (x + 2y ) < 0
Hence proved.
9. Use fourth fundamental concept. If x ≠y
Hence, 3xy 2 < (x 3 + 2y 3 )
10. For m > 1
AM of mth power > mth power of AM i.e., x 3 + 2y 3 > 3xy 2
1m + 2m + 3m + K + nm 15. We have
∴
n a 2 ≥ a 2 − (b − c )2 …(i)
m
1 + 2 + 3 K + n
>   b ≥ b − (c − a )
2 2 2
…(ii)
 n 
m c ≥ c − (a − b )
2 2 2
…(iii)
1m + 2m + 3m + K + nm  n + 1 
⇒ > 
n  2  Multiplying (i), (ii) and (iii), we get
n + 1
m
a 2b 2c 2 ≥ (a + b − c )2 (a + c − b )2 (b + c − a )2
⇒ 1m + 2m + K + nm > n  
 2  or abc ≥ (a + b − c )(a + c − b )(b + c − a )
11. Now, AM of mth power > mth power of AM 17. Let a = x 3, b = y 3, c = z3
1m + 3m + 5m + K + (2n − 1)m
It is sufficient to prove that
n
m x 3 + y 3 + z3 − 3xyz ≥ 0
1 + 3 + 5 + K + (2n − 1)
>  
 n  for any non –ve x , y , z
www.pdfworld.in
www.pdfworld.in
We know that It is necessary to prove following

2
x + y + z − 3xyz = (x + y + z ) (x + y + z
3 3 3 2 2 2
 a + b 1
  ≥
− xy − yz − zx )  8b  2
2 2
we also know that  a + b 1 a 1
We have,   = 1 +  ≥
x 2 + y 2 + z 2 − xy − yz − zx ≥ 0  8b  8 b 2
a
∴ x 3 + y 3 + z3 − 3xyz ≥ 0 Q >1
b
Hence proved. Similarly, we can prove second inequality.
18. Let a and b be two real + ve numbers. If p > 0, 
2
1
2
20. Let S =  x +  +  y + 
1
then ap − bp > 0 for a > b.  x  y
If p < 0, then ap − bp < 0 for a > b Opening the brackets, we have
∴We may assert (ap − bp )(aq − bq ) ≥ 0, if p and  1 1 
S =  x 2 + 2 + 2 + y 2 + 2 + 2
q are of same sign.  x y 
We have ap + q + bp + q ≥ apbq + aq bp  
p +q p +q
Q if x + y = a 
a +c ≥a c + a c
p p q p
 2
y2 + x 2  x 2 + y2 ≥ a 
S = x 2 + y2 + + 4
ap + q + l p + q ≥ ap l q + aq l p x 2y 2  2
 2
b p + q + c p + q ≥ b pc q + b q c p  xy ≤
a 
 4
Adding these inequalities termwise, we get
a2
(n − 1)(ap + q + bp + q + K + l p + q ) ≥ Σapbq a 2
∴ S≥ + 24 + 4
2 a
where a and b attain all the values from the
series a , b , c , K , l . 16
a2 a2
Adding Σ ap + q to both number of this S≥ + 4 × 8+ 4
2 a
inequality, we get a2 8
p +q p +q p +q S≥ + 2 + 4
n (a +b + K+ l ) 2 a
≥ (ap + bp + K + l p ) (aq + bq + K + l q ) a 4 + 16 + 8 a 2 (a 2 + 4)2
S≥ 2
=
2a 2a 2
19. (i) AM of two +ve numbers is not less than 1 4 
2
their GM S ≥   a +  
 2  a  
a+b 
Indeed − ab
2 Hence proved.
1 21. We have a 2 + b 2 − 2ab = (a − b )2 ≥ 0
= (a + b − 2 ab )
2
Hence, a 2 − ab + b 2 ≥ ab
1
= ( a − b )2 ≥ 0
8 a3 + b3 ≥ ab (a + b )
(ii) To prove that Consequently
a+b 1 (a − b )2 3a3 + 3b3 ≥ 3a 2b + 3ab 2
− ab ≤ ⋅ (a > b )
2 8 2 Adding a3 + b3, we get
It is sufficient to prove 4a3 + 4b3 ≥ (a + b )3
2
 b 1 (a − b )2 a3 + b 3  a + b 
3
 a −  ≤ ⋅ and so, ≥ 
 2  8 b 2  2 
www.pdfworld.in
www.pdfworld.in
Inequalities 339
22. Let Z = (cos a1 + cos a 2 + K + cos an )2 4n + 2 4n + 2 4n + 2

= + + ... +
(n + 1) (3n + 1) (n + 2)3n 2(2n + 1)2
+ (sin a1 + sin a 2 + K + sin an )2
 n 1 
= (cos2 a1 + cos2 a 2 + K + cos2 an ) > (4n + 2)  + 2
=1
 (2n + 1)2
2(2n + 1) 
+ (sin 2 a1 + sin 2 a 2 + K + sin 2 an )
25. Q AM > GM
+ 2(cos a1 cos a 2 + cos a1 cos a3 + K 2
b+c b + c
∴ > bc or   > bc
+ cos an − 1 cos an ) + 2(sin a1 sin a 2 2  2 
+ sin a1 sin a3 + K + sin an −1 sin an ) b+c bc
or >
= (1 + 1 + .. n times) + 2 [(cos(a1 − a 2 ) 4 b+c
+ cos(a1 − a3 ) + K + cos(an − an ) ] c+a ca a+b ab
−1 Similarly, > and >
Q cos(a1 − a 2 ) ≤ 1 4 c+a 4 a+b
cos(a1 − a3 ) ≤ 1 Adding these results, we get

b+c c+a a+b bc ca ab
………………… + + > + +
4 4 4 b+c c+a a+b
…………………
1 bc ca ab
………………… (a + b + c ) > + +
2 b+c c+a a+b
cos(an −1 − an ) ≤ 1
bc ca ab 1
∴ cos(a1 − a 2 ) + cos(a1 − a3 ) or + + < (a + b + c )
b+c c+a a+b 2
+ K + cos(an −1 − an )
n (n − 1) 26. Now,
≤1+ 1+ 1+ K times
2 (a 4 + b 4 + c 4 + d 4 ) + a (b3 + c 3 + d 3 )
n (n − 1) + b (a3 + c 3 + d 3 ) + c (a3 + b3 + d 3 )
=
2
+ d (a3 + b3 + c 3 ) > a 4 + b 4 + c 4 + d 4
n (n − 1)
∴ Z ≤ n + 2⋅ = n2 + 2(a 2b 2 + a 2c 2 + a 2d 2 + b 2c 2 + b 2d 2 + c 2d 2 )
2
∴ Z ≤ n2 or (a 4 + b 4 + c 4 + d 4 ) + (ab3 + a3b )
1 1 1 1 + (ac 3 + ca3 ) + (ad 3 + da3 ) + (bc 3 + b3c )

23. + + + K+ 2
n n+1 n+2 n + (bd 3 + b3d ) + (cd 3 + c 3d )
1  1 > (a 4 + b 4 + c 4 + d 4 ) + 2a 2b 2
+ K + 2 
1 1 1
> +  + +
n  n2 n2 n2 n  + 2a 2c 2 + 2a 2d 2 + 2b 2c 2 + 2b 2d 2 + 2c 2d 2
(n − n times)
2
…(i)
1 1 1 1
⇒ + + + K+ 2 Q AM > GM
n n+1 n+2 n
ab3 + a3b
∴ > ab3 × a3b
1 (n 2 − n ) 1 1 2
> + 2
= + 1− =1
n n n n
or (ab3 + a3b ) > 2a 2b 2
24. We have,
Similarly, (ac 3 + a3c ) > 2a 2c 2
1 1 1 1
+ + K+ +
n+1 n+2 3n 3n + 1 (ad 3 + a3d ) > 2a 2d 2
 1 1   1 1 and (bc 3 + b3c ) > 2b 2c 2

= +  +  + 
 n + 1 3n + 1   n + 2 3n and (bd 3 + b3d ) > 2b 2d 2
1
+ K+ and (cd 3 + c 3d ) > 2c 2d 2
2n + 1
www.pdfworld.in
www.pdfworld.in
Adding these results, we get 28. Q AM > GM

(ab + a b ) + (ac + a c ) + (ad + a d )
3 3 3 3 3 3 1  3 3 3
∴ b + c + d + c + d + a + d + a + b
4 
+ (bc + b c ) + (bd + b d ) + (cd + c d )
3 3 3 3 3 3
3 
> 2a 2b 2 + 2a 2c 2 + 2a 2d 2 + 2b 2c 2 + 2b 2d 2 +
a + b + c 
+ 2c 2d 2
 3 3 3
> × ×
Adding (a 4 + b 4 + c 4 + d 4 ) to both sides, we  b + c + d c + d + a d + a +b
get (i) 1/ 4
3 
× …(i)
a + b + c 
27. We are to prove that
a a a 
(a1b1 + a 2b2 + a3b3 )  1 + 2 + 3  1 b + c + d c + d + a d + a + b
 1
b b b3  Also, + +
4 
2
3 3 3
> (a1 + a 2 + a3 )2 1/ 4
a + b + c
+
or a12 +
a1a 2b a1b1a3 a1a 2b2
+ + + a 22 3 
b2 b3 b1
1  3(a + b + c + d )  b + c + d
or >
+
a 2b2a3 a1a3b3 a 2a3b3
+ + + a32 4  3   3
b3 b1 b2 1/ 4
c+d +a d +a+b a + b + c
> a12 + d 22 + a32 + 2a1a 2 + 2a1a3 + 2a 2a3 × × ×
3 3 3 
a a b aab  a + b + c + d b + c + d c + d + a
or (a12 + a 22 + a32 ) +  1 2 1 + 1 2 2  or > ×
 b2 b1  4  3 3
a a b a a b  a a b aab  1/ 4
+  1 3 1 + 1 3 3 +  2 3 2 + 2 3 3 d + a + b a + b + c
 b3 b1   b3 b2  × × …(ii)
3 3 
> (a12 + a 22 + a32 ) + 2a1a 2 + 2a1a3 + 2a 2a3
Multiplying (i) and (ii), we get
…(i)
1  3 3 3
Q AM > GM b + c + d + c + d + a + d + a + b
16 
1  a1a 2b1 a1a 2b2   a1a 2b1 a1a 2b2 
∴  + >  ×  3 
2  b2 b1   b2 b1  +
a + b + c 
a1a 2b1 a1a 2b2
or + > 2a1a 2 …(ii) (a + b + c + d ) > 1
b2 b1
3 3 3
or + +
Similarly, b+c+d c+d +a d +a+b
a1a3b1 a1a3b3
+ > 2a1a3 …(iii) +
3
>
16
b3 b1 a+b+c a+b+c+d
a 2a3b2 a 2a3b3
+ > 2a 2a3 …(iv) 29. We have to prove that
b3 b2
bcd cda dab abc
Adding (ii), (iii) and (iv) + + + >a + b + c + d
a2 b2 c2 d2
 a1a 2b1 a1a 2b2   a1a3b3  1 1 1 1
 +  +   or + + +
 b2 b1   b2  a 3
b 3
c 3
d3
a a b aab  > 
1
+
1
+
1
+
1 
+  2 3 2 + 2 3 3  > 2a1a 2 + 2a1a3 + 2a 2a3 
 b3 b2   bcd cda abd abc 
Adding (a12 + a 22 + a32 ) to both sides we get (i). (Dividing both sides by abcd )
www.pdfworld.in
www.pdfworld.in
Inequalities 341
Q AM > GM we have m
32. Let λ = ,m > n
1/3 n
> 3  3 3 3 
1 1 1 1
+ + We have
a3 b3 c3 a b c 
1/3 1 + α m   1 + α m  K  1 + α m  ⋅ 1 ⋅ 1...1
m
     
> 3  3 3 3 
1 1 1 1
+ +  n  n  n
a3 b3 d 3  abd 
1 + α m  + 1 + α m  + K + 1 + α m 
1/3
     
+ 3 + 3 > 3  3 3 3 
1 1 1 1  n  n  n
a 3
c d  acd  + m −n
1/3 <
> 3  3 3 3 
1 1 1 1 m
+ +
b3 c3 d 3  bcd  m
The factor 1 + α of the radicand is taken n
Adding these inequalities, we get n
3  3 + 3 + 3 + 3 
1 1 1 1 times, the factor 1 is taken as m − n times.
a b c d  n
1 + α m  m < 1 + α
Hence,  
> 3 
1 1 1 1   n
+ + + 
 abc abd acd bcd  m
or (1 + α )m /n > 1 + α
1 1 1 1 n
⇒ 3
+ 3 + 3 + 3
a b c d m
1 1 1 1 33. Let λ =
> + + + n
abc abd acd bcd
Assume that m > n i.e., λ > 1
30. Make use of the following identity we have
(a + b )(b + c )(a + c ) = (ab + ac + bc ) 1 − α m  1 − α m  ⋅ 1 − α m  ⋅ 1 ⋅ 1 ⋅ 1
m     
(a + b + c ) − abc  n n  n
a+b+c 3 ab + ac + bc
But ≥ abc ;
3 3 1 − α m  n + m − n
 
3  n
≥ a b c
2 2 2
<
m
∴ (a + b + c )(ab + ac + bc ) ≥ 9 abc m
The factor 1 − α of the radicand taken
Consequently, (a + b )(a + c )(b + c ) ≥ 8 abc n
1
31. Consider a quantities equal to . n times and factor 1 is taken m − n times.
a n
Hence,  1 − α  < 1 − α <

m m 1
1
b quantities equal to .  n 1+α
b
m 1
1
c quantities equal to . 1−α < m
c n
(1 + α ) n
AM of these quantities will be m
1
1 1 1 (1 + α ) n <
a⋅ +b⋅ +c⋅ m
a b c = 3 1−α
a+b+c a+b+c n
Assume that m < n we have
GM is equal to
n (1 + α ) m = n (1 + α )(1 + α ) K (1 + α ) ⋅ 1 ⋅ 1 ⋅ 1
1 1 1
a +b +c ⋅ ⋅ (1 + α ) m + n − m m 1
a a bb c c < =1+ α⋅ <
n n 1 − αm
3 1 1 1
Consequently, ≥ a +b +c ⋅ ⋅ n
a+b+c a a bb c c m
1
a b c So in this case, also (1 + α ) n <
1 αm
i.e., a
a +b +c
b
a +b +c
c
a +b +c
≥ (a + b + c ) 1−
3 n
www.pdfworld.in
www.pdfworld.in
x+ y + z
34. Consider x quantities each equal to x ; y x + y + z < x xy y z z
or  
quantities each equal to y and z quantities  3 
each equal to z. x+ y + z
x + y + z
Q AM > GM or x xy y z z >  
 3 
(x + x + K to x terms ) + ( y + y + K y terms)
+ (z + z + K z terms) 35. We have to prove that
a +b +c
x +y + z  bc + ca + ab 
  > b a c a ⋅ c b ab ⋅ ac b c
 a+b+c 
> [(x ⋅ x K to x factors) (y ⋅ y K to y factors)  bc + ca + ab 
a +b +c
1 or  
(z ⋅ z K z factors)]  a+b+c 
x+y + z
1
> ab + c × b c + a
× ca + b …(i)
(x ⋅ x ) + ( y ⋅ y ) + (z. z ) x+ y + z
or > [(x x )( y y )(z z )] Consider, (b + c ) quantities each equal to a,
x +y+z
(c + a ) quantities each equal to b and (a + b )
x+ y + z
x 2 + y 2 + z 2  quantities each equal to c.
or   >x ⋅y ⋅ z
x y z
 x +y + z  Q AM > GM
1 ∴ [to (b + c ) terms] × [b + b + Kto (c + a ) terms]
Again, consider x quantities each equal to ,y
x × [c + c + K to (a + b ) terms]
quantities each equal to
1
and z quantities (b + c ) + (c + a ) + (a + b )
y
1 > [{a ⋅ a ... to (b + c ) factors} {b ⋅ b K to (c + a )
each equal to .
z factors}
Q AM > GM 1
(b + c ) + (c + a ) + ( a + b )
∴ × {c ⋅ c K to (a + b ) factors]
 1 + 1 + K to x terms  +  1 + 1 + . . to y terms  [(b + c ) a + (c + a ) b + (a + b ) c ]
    or
x x  y y  (b + c ) + (c + a ) + (a + b )
1 1 
+  + + K to z terms  1

z z 
> [ab + c × bc + a
× c a + b ](b + c ) + (c + a ) + ( a + b )
xyz
1 1
2(ab + bc + ca )
 1 1 
>   ⋅ . . . to x terms   ⋅ K to y terms 
or
2(a + b + c )
 x x  y y 
1
1
 1 ⋅ 1 K to z terms  x + y + z > [ab + c × bc + a
× c a + b ]2 ( a + b + c )
 
z z  a +b +c
 ab + bc + ca 
x ⋅ 1  + y ⋅ 1  +  z ⋅ 1  or  
     a+b+c 
 x   y   z
or
x +y + z > [ab + c × bc + a
× c a + b ]1/ 2
1 a +b +c
 1 x  1 y 1 z x + y + z  ab + bc + ca 
>         or  
 a+b+c 
  x   y   z  
1
> ab + c × b c + a
× ca + b
1+1+1  1 x + y + z
or >  x y z
x + y + z x y z  36. Apply AM–GM inequality to the 9 numbers
1 2x 2x 2x 5y 5y 3z 3z 3z 3z
x +y + z z x+ y + z , , , , , , , ,
or < (x xy y z ) 3 3 3 2 2 4 4 4 4
3
www.pdfworld.in
www.pdfworld.in
Inequalities 343
we get Adding (i), (ii) and (iii), we get

 1 b2 + c 2 c 2 + a2 a2 + b2
  2x  3  5y  2  3z  4  9 
 + +
1 b+c c+a a+b
(2x + 5y + 3z ) ≥        
9  3 2 4 
b+c c+a a+b
> + + = (a + b + c )
1 2 2 2
 2 3 5 2 3 4 9
=        x 3y 2z 4  39. Let z = (8 − x )3 (x + 6)4
  3   2   4   3 4
8 − x   x + 6
1 1 = 3344     …(i)
5 ⋅ 3 2
9  3   4 
=  7 ⋅ 7 =  525  9
 7 
 2   2 3 4
8 − x   x + 6
∴ z will be maximum when    
 3   4 
37. If m does not lie between 0 and 1. Now,
AM of mth power > mth power of AM is maximum.
m 3 4
m
+c m
b + c 8 − x   x + 6
> 
b
∴  But     is product of 7 factors,
2  2   3   4 
Dividing both sides by (b + c )m −1 sum of which = 14
3 4
8 − x   x + 6
bm + c m
>
(b + c )m ∴     will be maximum if all the
2(b + c )m −1
2m (b + c )m −1  3   4 
factors are equal i.e.,
bm + c m b+c
or > …(i) 8− x x + 6
(b + c )m −1
2m −1 = =2
3 4
c m + am c+a ∴ Maximum value of z = 63 ⋅ 84
Similarly, m −1
> m −1
…(ii)
(c + a ) 2
40. Let z = (7 − x )4 (2 + x )5
a m
+b m
a+b
and > …(iii) 4
7 − x  2 + x 
5
(a + b )m −1
2m −1 = 4455     …(i)
 4   5 
Adding (i), (ii) and (iii), we get 4 5
7 − x  2 + x 
bm + c m c m + am am + b m ∴ z is maximum when     is
−1
+ −1
+ −1
 4   5 
(b + c )m (c + a )m (a + b )m
maximum.
1
> −1
[(b + c ) + (c + a ) + (a + b )] But it is the product of 9 factors, sum of which
2m
is 9.
1 a+b+c
= [2(a + b + c )] = 4 5
2m −1
2m −2
∴  7 − x   2 + x  will be maximum, if all
   
 4   5 
Hence proved.
factors are equal.
38. AM of mth power > mth power of AM 7−x 2+ x
i.e., If =
2 4 5
b2 + c2 b + c 
∴ >  (7 − x ) + (2 + x ) 9
2  2  = = =1
4+5 9
b2 + c2 b + c
or > …(i) Now, do yourself.
b+c 2
41. Let z = x 1/ 2 ⋅ (1 − x )1/3
c 2 + a2 c + a
Similarly, > …(ii)
c+a 2 z 6 = x 3 (1 − x )2
3 2
1−x
z 6 = 3322   
a2 + b2 a + b x
> …(iii) or  …(i)
 3  2 
a+b 2
www.pdfworld.in
www.pdfworld.in
3 2
∴ z is maximum when x  1 − x  44. Dividing out gives
    is
 3  2  f (x ) = 9x sin x +
4
maximum. x sin x
3 2
1−x Qx sin x is non –ve in 0 < x < π. Using AM–GM
Sum of the factors of   
x

 3  2  inequality, we get
1−x
= 3   + 2 
x 4
 f (x ) ≥ 2 9x sin x ⋅ = 12
 3  2  x sin x
= x + (1 − x ) = 1 ∴ Minimum possible value is 12.
3 2 4
1−x It is attained when x 2 sin x 2 =
∴   
x
 will be maximum, if all factors 9
 3  2 
are equal. π
Q > 1, x sin x being continuous does attain
x 1 − x x + (1 − x ) 1 2
i.e., = = = π
the value for some x in the interval  0, .
2
3 2 3+2 5
3  2 
From Eq. (i), maximum value of
1
3 2 45. Let u =
z 6 = 3322    
1 1
ab
 5  5
1 1
Then, a 2 + b 2 = 1 ⇒ 2ab ≤ 1 ⇒ u ≥ 2
∴ Maximum value is     .  3 2  2 3  2
 5  5 ∴ u + 2− >0
 u 
42. Let y =1+ x  2
or x =y −1 Hence, (u − 2 )  u + 2− ≥ 0
 u 
(5 + x ) (2 + x ) (5 + y − 1)(2 + y − 1)
∴ = 2
1+x y ⇒ u2 − 2 − 2 + ≥0
u
(4 + y )( y + 1)
= ⇒
2
u2 +
≥2+ 2
y u
(4y + 4 + y 2 + y ) 4 1 1
= =y + 5+ ∴ a+b+ ≥ 2 ab + ≥2+ 2
y y ab ab
2 Equality holds if and only if a = b = 2
 2 
= y −  +5+ 4 −1 −2
 y 46. Now, x n − y n = (x − y )(x n + xn y
n −2 −1
 2 
2 + K + xy + yn )
= y −  + 9 …(i)
 y Let x = (a + k )
n n 1/n
The given expression will have minimum value y = (bn + kn )1/n

2
 2  We get
when  y −  =0 −1 −2
 y n (an + kn ) − n (bn + kn ) (an + an b+K
n −1
∴ Minimum value of given expression = 9 +b ) ≤ a − bn
n
…(i)
43. For real number a , b , x y , we have inequality [Qx ≥ a , y ≥ b , k ≥ 0 and a > b > 0]
−1 −2 −2 −1
ax + by ≤ a + b 2 2
x +y
2 2 But an + an b + K + abn + bn
an − b n
Hence, 60 = 5x + 12y ≤ 52 + 122 x 2 + y2 = is +ve
a −b
= 13 x 2 + y 2 ∴ From (i)
60 n an + kn − n bn + kn ≤ a − b
The minimum value of x + y is . 2 2
13
Hence proved.
www.pdfworld.in
www.pdfworld.in
Inequalities 345
47. Qa , b , c are +ve rational numbers. abc sin (B − C ) abc sin (B − C )

= =
Using Weighted mean theorem for a , a ; b , b ; 2 sin 2 C 
sin 2 B (b 2 − c 2 )
a  2
− 2 
c , c we get  sin A sin A 
1
a  
(a ⋅ a + b ⋅ b + c ⋅ c ) c
> (a abbc c )a + b + c abc b  ar
a+b+c ≤ = =
b2 − c2 2
1 − r2
1 −  
a +b +c c
a2 + b2 + c 2  b 
or   > a ab b c c …(i)
 a+b+c  ar
∴ p≤
1 1 1 1 − r2
Using same result for a , ; b , ; c , we get
a b c
49. Let k be a natural number and n be the unique
1 1 1 1
a⋅ +b⋅ +c⋅ integer such that (n − 1)2 ≤ k < n 2. Then, we see
c >   1   1   1  
a b c a +b +c
a b
       that
a+b+c  a b c  k x
Σ r ≤  1 + 2 + 3  +  4 + 5 + K + 8 
x x x x x x
1 r =1 r 1 2 3  4 5 8
 3   1 a +b +c
or  < a b c  x 2 x x 2 
a + b + c  a ⋅ b ⋅ c  + K +  (n − 1) 2 + K + k + K + 2n −1 
 (n − 1) k n − 1 
a +b +c 
 3  1
>
≤  1 + 1 + 1  +  4 + 4 + K + 4 
or   x x x x x x
a + b + c  a ab b c c 1 1 1  4 4 4
a +b +c
a + b + c   x 2 x 2 
or   < a ab b c c …(ii) + K +  (n − 1) 2 + K + (n −1) 2 
 3   (n − 1) (n − 1) 

From (i) and (ii), we get (2n − 1)x
3x 1 5x 2 (n − 1) 2
a +b +c = + + ... +
a + b + c  <a b c a b c 1 4 (n − 1)2
 
 3 
n −1 (2r + 1)x
a +b +c = Σ r2
a + b + c 
2 2 2
r =1 r2
< 
 a+b+c  n −1 3r
≤ Σ x
r =1 r2 r2
1 1
48. bc sin A = a ⋅ p n −1 x
2 2 =3 Σ r2 ≤ 3,
bc sin A r =1 r
⇒ p=
a where the last inequality follows from the
A given hypothesis.
50. The given inequality may be written in the
form
| (x − y ) ( y − z )(z − x )| < xyz.
p
Since x , y , z are the sides of a triangle, we
know that
| x − y | < z ,| y − z | < x and | z − x | < y .
Multiplying these, we obtain the required
B M C inequality.
abc (sin 2 B − sin 2 C )sin A 51. We may write (ii) in the form
= ab − a − b + 1 + cd − c − d + 1 = 5
a (sin B − sin C )
2 2 2
Thus, we obtain the equation

abc sin (B − C ) sin 2 A (a − 1)(b − 1) + (c − 1)(d − 1) = 5. If a − 1 ≥ 2,
=
a 2 (sin 2 B − sin 2 C ) then (i) shows that
www.pdfworld.in
www.pdfworld.in
b − 1 ≥ 2, c − 1 ≥ 2 and d − 1 ≥ 2 6, 7, 8, 9, 10 and the smaller numbers are

so that (a − 1) (b − 1) + (c − 1)(d − 1) ≥ 8. 1, 2, 3, 4, 5. This implies that
If follows that a − 1 = 0 or 1. | a1 − b1| + | a 2 − b2| + | a3 − b3| + | a 4 − b4 |
If a − 1 = 0, then the contribution from + | a5 − b5|
(a − 1)(b − 1) to the sum is zero for any choice = 10 + 9 + 8 + 7 + 6 − (1 + 2 + 3 + 4 + 5) = 25
of b. But then (c − 1) (d − 1) = 5 implies that
54. By AM-GM inequality, we have
c − 1 = 1 and d − 1 = 5 by (i). Again (i) shows 2
a + 1 − a
that b − 1 = 0 or 1 since b ≤ c . Taking a (1 − a ) ≤ 
1
 =
b − 1 = 0, c − 1 = 1 and d − 1 = 5 we get the  2  4
solution (a , b , c , d ) = (1, 1, 2, 6). Similarly, we also have
Similarly,b − 1 = 1, c − 1 = 1 and d − 1 = 5 gives 1 1
(a , b , c , d ) = (1, 2, 2, 6). b (1 − b ) ≤ and c (1 − c ) ≤
4 4
In the other case a − 1 = 1, we see that b − 1 = 2
Multiplying these we obtain
is not possible for then c − 1 ≥ 2 and d − 1 ≥ 2.
1
Thus b − 1 = 1 and this gives (c − 1) (d − 1) = 4. It abc (1 − a ) (1 − b ) (1 − c ) ≤
follows that c − 1 = 1, d −1= 4 or 43
c − 1 = 2, d − 1 = 2. Considering each of these, We may rewrite this in the form
we get two more solutions 1
(a , b , c , d ) = (2, 2, 2, 5), (2, 2, 3, 3). a (1 − b ) ⋅ b (1 − c ) ⋅ c (1 − a ) ≤
43
It is easy to verify all these four quadruples are
indeed solutions to our problem. Hence, one factor at least (among a (1 − b ),
b (1 − c ), c (1 − a )) has to be less than or equal to
52. We have 1 1
; otherwise lhs would exceed 3 .
n 2 < n 2 + 1 < n 2 + 2 < n 2 + 3 < .... < n 2 + n 4 4
Hence, we see that Again, consider the sum
1 2 n 1 a (1 − b ) + b (1 − c ) + c (1 − a ).
+ + K+ 2 >
n2 + 1 n2 + 2 n + n n2 + n This is equal to a + b + c − ab − bc − ca
We observe that
2 n
+ + K+ 3(ab + bc + ca ) ≤ (a + b + c )2
n2 + n n2 + n
which, in fact, is equivalent to
1 1
= (1 + 2 + 3 + K + n ) = (a − b )2 + (b − c )2 + (c − a )2 ≥ 0
n2 + n 2
This leads to the inequality
Similarly, we see that
1 2 n a + b + c − ab − bc − ca ≥ (a + b + c )
+ 2 + K+ 2 1 1 2
n +1 n +2
2
n +n − (a + b + c )2 = 1 − =
3 3 3
1 2 n 1
< + + K + 2 = 2 (1 + 2 + 3 + K + n ) Hence, one summand at least (among
n2 n2 n n
a (1 − b ), b (1 − c ), c (1 − a )) has to be greater than
1 1
= + 2
or equal to ; (otherwise lhs would be less
2 2n 9
53. (i) Fix any pair {a j , b j }. We have 2
than . )
a1 < a 2 < K < a j −1 < a j and 3
b j > b j +1 > K > b5. Thus there are j − 1 55. (i) Since, x and y are positive, we have
numbers smaller than a j and 5 − j y ≤ x − x 3 − y 3 < x . Also, x − x 3 ≥ y + y 3 > 0. So,
numbers smaller than b j . Together they x (1 − x 2 ) > 0
account for j − 1 + 5 − j = 4 distinct
numbers smaller than a j as well as b j . Hence, x < 1. Thus, y < x < 1, proving part (i).
Hence, the larger of a j and b j is at least 6. (ii) Again, x 3 + y3 ≤ x − y
(ii) The first part shows that the larger x −y
So, x 2 − xy + y 2 ≤
numbers in the pairs {a j , b j }, 1 ≤ j ≤ 5, are x +y
www.pdfworld.in
www.pdfworld.in
Inequalities 347
That is 58. We use the trivial inequalities

x −y x − y + xy (x + y ) a 2 + 1 ≥ 2a , b 2 + 1 ≥ 2b and c 2 + 1 ≥ 2c . Hence,
x +y ≤
2 2
+ xy =
x +y x +y we obtain
Here, xy (x + y ) < 1 ⋅ y ⋅ (1 + 1) = 2y a2 + 1 b2 + 1 c 2 + 1
+ +
x − y + 2y b+c c+a a+b
So, x 2 + y2 <
x +y 2a 2b 2c
≥ + +
x +y b+c c+a a+b
= =1
x +y 2a 2b 2c
+ + ≥3
This proves (ii) b+c c+a a+b
56. Suppose the equation x 2 + x + 4λ = 0 has no Adding 6 both sides, this is equivalent to
real roots. Then 1 − 16λ < 0. This implies that  1 1 1 
(2a + 2b + 2c )  + + ≥9
1 − 16(a3 + a 2bc ) < 0, 1 − 16(b3 + ab 2c ) < 0, b + c c + a a + b 
1 − 16(c 3 + abc 2 ) < 0 Taking x = b + c , y = c + a , z = a + b , this is
Observe that equivalent to
1 − 16 (a3 + a 2bc ) < 0  1 1 1
(x + y + z )  + +  ≥ 9
x y z
⇒ 1 − 16a 2 (a + bc ) < 0
This is a consequence of AM-GM inequality.
⇒ 1 − 16a 2 (1 − b − c + bc ) < 0
Aliter The substitutions
⇒ 1 − 16a 2 (1 − b ) (1 − c ) < 0 b + c = x , c + a = y , a + b = z leads to
2a y + z −x x y 
⇒
1
< a 2 (1 − b ) (1 − c ) Σ =Σ =Σ + 
b+c x y x 
16
Similarly, we may obtain − 3≥ 6 − 3 = 3
1
 22
59. We gave (22 2
.
16 16 Hence, 4
5 > 22
. > 1.4, as (1.4)2 = 1.96 < 22
. ,
Multiplying these three inequalities, we get therefore 5 > 5 > 1.4
3 4
1
a 2b 2c 2 (1 − a )2 (1 − b )2 (1 − c )2 > 3 Adding, we get
16
5+ 3
5+ 4
5 > 22
. + 1.4 + 1.4 = 5
However, 0 < a < 1 implies that a (1 − a ) ≤ 1 / 4.
We observe that
Hence,
3 < 3, 3 8 = 2 and 4
8 < 3 8 = 2. Thus,
a 2b 2c 2 (1 − a )2 (1 − b )2 (1 − c )2
8+ 3
8+ 4
8< 3 + 2 + 2 = 7< 8
1
= (a (1 − a ))2 (b (1 − b ))2 (c (1 − c ))2 ≤ 3 Suppose n ≥ 9. Then n 2 ≥ 9n , so that n ≥ 3 n .
16
This gives n ≤ n / 3.
a contradiction. We conclude that the given
Therefore, 4 n < 3 n < n ≤ n / 3.
equation has real roots.
We thus obtain
57. Using | a − b | ≥ | c | , we obtain (a − b )2 ≥ c 2 which n + 3
n + 4
n < (n / 3) + (n / 3) + (n / 3) = n
is equivalent to (a − b − c ) (a − b + c ) ≥ 0.
60. Let us look at the problem of finding two
Similarly, (b − c − a )(b − c + a ) ≥ 0 and positive integers a, b such that
(c − a − b )(c − a + b ) ≥ 0. Multiplying these 1 < a < b < a 2 , a − 1 divides b − 1 and a 2 − 1
inequalities, we get divides b 2 − 1.
− (a + b − c ) (b + c − a ) (c + a − b ) ≥ 0.
2 2 2
Thus, we have
This forces that lhs is equal to zero. Hence, it b − 1 = k (a − 1) and b 2 − 1 = l (a 2 − 1)
follows that either a + b = c or b + c = a or Eliminating b from these equations, we get
c = a = b. (k 2 − l ) a = k 2 − 2k + l .
www.pdfworld.in
www.pdfworld.in
Thus, it follows that bn = (n + 2)3 − (n + 2)2 − 2(n + 2) + 1

k − 2k + l
2
2(k − l ) = n3 + 5n 2 + 6n + 1,
a= =1−
k2 − l k2 − l
for all n ≥ 1. Then we see that
We need a to be an integer. Choose k 2 − l = 2 so
1 < an < bn < bn2 ,
that a = 1 + l − k = k2 − k − 1 and
b = k (a − 1) + 1 = k3 − k 2 − 2k + 1. We want a > 1 for all n ≥ 1. Moreover
which is assured if we choose k ≥ 3. Now, a 0 which again is and an2 − 1 = n (n + 3)(n + 1)(n + 2),
assured once k ≥ 3. It is easy to see that b < a 2 is
bn2 − 1 = n (n + 3)(n + 2)(n + 1) (n 2 + 4n + 2)
equivalent to k (k3 − 3k 2 + 4) > 0 and this is
also true for all k ≥ 3. Thus we define Thus, we have a pair of desired sequences
an = (n + 2)2 − (n + 2) − 1 = n 2 + 3n + 1, an and bn .
Level 2
1. Since, a 2 + b 2 + c 2 = 1, the inequalities to be Now, the proof
proved may be written in the form  2a  +  2b + 1  = 2m + 2r  + 2n + 2s + 1 
1  c   c   c   c 
− (a 2 + b 2 + c 2 ) ≤ ab + bc + ca ≤ a 2 + b 2 + c 2
2 2s + 1 
= 2m + 2n +   + 
2r
or − (a 2 + b 2 + c 2 ) ≤ 2(ab + bc + ca )
 c   c 
≤ 2(a 2 + b 2 + c 2 ). r + s + 1
≥m +   + n +   + m + n + 
r s
These inequalities are indeed true since  c   c   c 
2(ab + bc + ca ) + (a 2 + b 2 + c 2 ) r + s + 1
= m +  + n +  + m + n +
r s
= (a + b + c )2 ≥ 0  c   c   c 
and 2(a 2 + b 2 + c 2 ) − 2 (ab + bc + ca ) a + b + 1
= +  + 
a b
= (a − b ) + (a − c ) + (b − c ) ≥ 0
2 2 2  c   c   c 
2. Let a = mc + r and b = nc + s , where m , n , r and 3. Without loss of generality, we may assume

s are integers and 0 ≤ r < c and 0 ≤ s < c. As the that a ≥ b ≥ c .
coming proof will indicate, a certain 1
a3 + b3 + c 3 + 6abc ≥ (a + b + c )3
relationship must be established; for 4
conciseness of proof, we will establish that ⇔ 4(a3 + b3 + c 3 + 6abc )
first
≥ (a3 + b3 + c 3 ) + 3ab (a + b ) + 3bc (b + c )
If s ≥ r , then 2s + 1 ≥ r + s + 1
+ 3ca (c + a ) + 6abc
So,  2s + 1  ≥  r + s + 1  ;
 c   c  ⇔ (a3 + b3 + c 3 + 6abc ) ≥ ab (a + b ) + bc (b + c )
if s < r , then 2r ≥ r + s + 1 and + ca (c + a )

We have, a − c ≥ b − c
 2r  ≥  r + s + 1 ;
 c   c  ⇒ a (a − c ) ≥ (b − c )
in either case, ⇒ a (a − c )(a − b ) ≥ b (b − c )(a − b )
 2r  +  2s + 1  ≥  r + s + 1  ⇒ a (a − b )(a − c ) + b (b − c ) (b − a ) ≥ 0
 c   c   c 
Also, c −a≤ 0
Furthermore, since   =   = 0, we can also
r s and c −b≤ 0
 c   c  Thus, c (c − a ) (c − b ) ≥ 0
say that
Adding these two inequalities, we get
 2r  +  2s + 1  ≥  r  +  s  +  r + s + 1 
a (a − b ) (a − c ) + b (b − c )(b − a )
 c   c   c   c   c 
+ (c − a )(c − b ) ≥ 0
www.pdfworld.in
www.pdfworld.in
Inequalities 349
Expanding the left hand side of this inequality, and hence, in cyclic sum notation,
we get 1
Σyz = Σ (2bc − a 2 )
a3 + b3 + c 3 − ab (a + b ) − bc (b + c ) − ca (c + a ) 4
Now, assume a ≥ b ≥ c without loss of
+ 3abc ≥ 0
generality; then
Thus, ab (a + b ) + bc (b + c ) + ca (c + a )
2bc − a 2 ≤ 2ca − b 2 ≤ 2ab − c 2
≤ a3 + b3 + c 3 + 3abc ≤ a 2 + b3 + c 3 + 6abc
and therefore, by Chebyshev’s inequality and
The equality holds if and only if the AM–GM inequality.
a (a − b ) (a − c ) + b (b − c )(b − a ) = 0 and 4  1  1 2 1 1
c (c − a ) (c − b ) = 0. The second equality holds if (Σyz )  Σ  = Σ (2bc − a ) Σ 2
9  ( y + z )2  3 3 a
either c = 0 or c − a = 0 or c − b = 0 . If c = 0,
from the first equality, we get ≥
1 
Σ  (2bc − a 2 ) ⋅ 2  =
1 1  Σ 2bc  − 1
 
a 2 (a − b ) + b 2 (b − a ) = 0 ⇒ (a − b )(a 2 + b 2 ) = 0 3  a  3  a2 
⇒ a = b. If c − a = 0, then since a ≥ b ≥ c , it 1/3
≥  Π 2 
2bc
follows that a = b = c and substituting in the −1=2−1=1
 a 
equality we see that a = b = c = 0. If c = b , then
a (a − b )(a − b ) = 0 ⇒ a = 0 or a = b. If a = b , we The inequality follows.
see as before that a = b = c = 0.
6. By the AM–GM inequality,
This completes the proof.
(a + b )2 + (a + b + 4c )2
4. By the power means inequality we have
= (a + b )2 + (a + 2c + b + 2c )2
1 + (n + 1)n + 1
n ≥ (2 ab 2 ) + (2 2ac + 2 2bc 2 )
n+2
= 4ab + 8 ac + 8 bc + 16c ab
1 + (n + 1)n + K + (n + 1)n + 1
= n Therefore,
n+2
(a + b )2 + (a + b + 4c )2
n −1 n −1 ⋅ (a + b + c )
1 + (n + 1) + K + (n + 1) abc
> n −1
n+2 4ab + 8ac + 8bc + 16c ab
≥ × (a + b + c )
If we take away one of the (n + 1)n −1 terms in abc
=  + + +
this average, the average will obviously 4 8 8 16 
 (a + b + c )
decrease, so c b a ab 
1 + (n + 1)n + 1
= 8 
1 1 1 1 1 
n + + + + 
n+2  2c b a ab ab 
1 + (n + 1) n −1
+ K + (n + 1) n −1 a + a + b + b + c
> n −1  
n+1 2 2 2 2 
1 + (n + 1)n −1 + K + nn −1  5 1   5 a 2b 2c 
> n −1 ≥ 8 5  5  = 100
n+1  2a 2b 2c   24 
1 + n ⋅ nn −1 1 + nn Again, by the AM–GM inequality. Hence, the

= n −1 = n −1
n+1 n+1 largest constant k is 100. For k = 100, equality
holds if and only if a = b = 2c > 0
and the proof is complete.
7. Using the weighted AM–GM inequality three
5. Writing y + z = a , z + x = b and x + y = c , we times, we have the following
have 1
c ⋅a + a ⋅b + b ⋅c
x = b + c − a / 2, etc., ≥ (ac b ac b )a + b + c
c+a+b
thus 1
a 2 − (b − c )2 a 2 − b 2 − c 2 + 2bc b ⋅a + c ⋅b + a ⋅c
yz = = , etc. ≥ (abbc c a )a + b + c
4 4 b+c+a
www.pdfworld.in
www.pdfworld.in
1
a ⋅a + b ⋅b + c ⋅c If follows that
≥ (a abbc c )a + b + c ( p + 1)k + 1 − p k + 1 > (k + 1) p k > p k + 1 − ( p − 1)k + 1
a+b+c
Adding these inequalities together gives or letting p successively have the values
(a + b + c )2 1, 2, 3, K , n
1=a + b + c =
a+b+c 2k + 1 − 1k + 1 > (k + 1)1k > 1k + 1 − 0,
a 2 + b 2 + c 2 + 2ab + 2ac + 2bc 3k + 1 − 2k + 1 > (k + 1)2k > 2k + 1 − 1k + 1,

=
a+b+c 4k + 1 − 3k + 1 > (k + 1)3k > 3k + 1 − 2k + 1 ,
= a a b b c c + ab b c c a + ac b a c b ....................................................
2 2 2 (n + 1)k + 1 − n k + 1 > (k + 1)n k > n k + 1 − (n − 1)k + 1
8. We have,   +   +   − 2
a b c
 5  5  5 If these inequalities are added together, the
2 2 2
following inequalities result;
 ax + by + cz  +  x  +  y  +  z 
      (n + 1)k + 1 − 1 > (k + 1) (1k + 2k + 3k
 30 30 30   6   6  6
+ K + n k ) > n k + 1,
=1−2+ 1= 0
2 2 2 or dividing through these inequalities by
=  −  =  −  +  −  ;
a x b y c z
k + 1,
 5 6  5 6  5 6
k +1
 1 1  1
  1 +  − k+1  nk +1
= , so a = kx  where k = 
a x 5
thus n k + 1
5 6  6   n 
and b = ky and c = kz. 1
> 1k + 2k + 3k + K + n k > n k + 1.
5 k+1
The answer is then k, or . In a geometric
6 This is essentially the set of inequalities
setting, if the given information is applied to sought.
triangles ABC and XYZ (even replacing the 25,
36 and 30 by r 2 , s 2 and r , s respectively), the 10. From a ≥ b ≥ c > 0, we have
a+b b+c a+c
triangles must then be similar. ≥ 2, 0 < ≤ 2 and ≥1
c a b
9. We note that in the
a2 − b2
S = x k + x k −1 + x k − 2 + K + x + 1 if x > 1, Now, we get ≥ 2 (a − b ), because a ≥ b ;
c
then the first term is numerically the greatest, c2 − b2
but if x < 1, then the last term is greatest. It ≥ 2(c − b ), because c ≤ b ; and
a
follows that
a2 − c 2
(k + 1)x k > S > k + 1, if x > 1; ≥ a − c , because a ≥ c .
b
(k + 1)x k < S < k + 1, if x < 1.
After addition of these inequalities, we have
If both sides of these inequalities are
a2 − b2 c 2 − b2 a2 − c 2
multiplied by x − 1, it is found that for x ≠ 1 + +
a a b
(k + 1)x k (x − 1) > x k + 1 − 1 > (k + 1) (x − 1).
p ≥ 2(a − b ) + 2 (c − b ) + (a − c )
Assume now that x = ; then we find
( p − 1) that is,
(k + 1)p k
p k +1
− ( p − 1) k +1
k + 1 − ( p − 1) k a2 − b2 c 2 − b2 a2 − c 2
> > + + ≥ 3a − 4b + c
( p − 1)k + 1 ( p − 1)k + 1 ( p − 1)k + 1 c a b
p+1 The equality holds if and only if a = b = c > 0
Analogously, if we assume that x = , we
= 
p 1 1 1 1 
11. S = + K+ + 
obtain 1001 3001  1001 3001 
(k + 1) (p + 1)k ( p + 1)k + 1 − p k + 1 (k + 1)p k + K + 
1
+
1 
 +
1
k +1
> k +1
> k +1  2000 2002  2001
p p p
www.pdfworld.in
www.pdfworld.in
Inequalities 351
For any n, we have Hence, the given sum

n 2 − 4002 n + 20012 ≥ 0 and a1 a 2 a a
+ + K + n −1 + n
hence n (4002 − n ) ≤ 20012 a 2 a3 an a1
  < n − 1 + k + n − k = 2n − 1
 1 1  This completes the proof.
∴S > 4002  2
+ K+ 2
 2001
1444 42444 2001
3
4 13. Proof by induction. Given inequality is true for
 1000 terms  n = 1. Assume it to be true as such and show
+
1
= 4002 ⋅
1000
+
1 that when n is replaced by (n + 1) too it is true.
2 Rewriting, the inequality is
2001 2001 2001
2000 + 1 1  1 1 
= =1 1 + + K + 
2001 n+1 3 2n − 1 
Again, taking terms hundred at a time 1 + K + 1 
1
1 1 ≥  
S = + K+ < 100 2
n 2n 
1001 3001 n+1 1 1 1 1
 1 + 1 + K+ 1  = ⋅  + + ... + 
  n n + 1 2 4 2n 
 1001 1101 2901 
=  1 +  ⋅ 1 + 1 + . . . + 1 
1 1
< 100 
1 1 1 
  
+ + ... +  n n + 1 2 4 2n 
 1000 1100 2900 
< 5 
1 1 1 1 1 1 1
< + + ... + + + + 
10 11 29  10 15 20 25  1  1 1 1 1 1
1 − + − + K + − 
1 1 1 1 77 17 1 n+1 2 3 5 2n − 1 2n 
= + + + = =1 <1
2 3 4 5 60 60 3
≥
1 1 + 1 + K + 1 
 
Hence, 1< S < 1
1 n (n + 1) 2 4 2n 
3  1 1 1 1 1
i.e., 1 − + − + K + − 
12. From what has been given we have, for  2 3 5 2n − 1 2n
1 ≤ i ≤ n − 1,
≥  + + K +
1 1 1 1
ai − ai ≤ | ai − ai + 1| < 1 < ai + 1 
+1 n 2 4 2n 
⇒ ai < 2ai + 1 (As n → ∞, note that LHS converges to log 2
Hence, each term in the sum 1 log n
while RHS converges to lim ⋅ = 0.)
a a a n→ ∞ 2 n
S = 1 + 2 + K + n −1
a 2 a3 an Assuming the last inequality we are to prove
1 1 1 1
is less than 2. Suppose that k terms in this sum 1 − + − ... + −
2 3 2n − 1 2n
are ≥ 1 and the remaining n − 1 − k terms are
< 1. Then 1 1 1 1 1 
+ − ≥  + K+ 
S ≤ 2k + (n − 1 − k ) ⋅ 1 = n − 1 + k 2n + 1 2n + 2 n + 1 2 2n + 2 
Next, an − a1 = (a 2 − a1 ) + (a3 − a 2 ) lhs (by induction hypothesis)
≥  + K +
+ K + (an − an −1 ) 1 1 1 1 1
+ −
each term on the right hand side is less than 1. n 2 2n  2n + 1 2n + 2
If ai + 1 − ai > 0, then ai + 1 > ai , or ai + 1 < 1. Hence, n + 1 1 1 1 1
=  + K+ + −
there are only n − 1 − k positive terms in this n (n + 1) 2 2n  2n + 1 2n + 2
sum for an − a1. Thus
1  1 
an − a1 ≤ (n − 1 − k ) × 1 + k × 0 =  1 +   
1 + . . . + 1 
 
 n   n + 1 2 2n 
= n − 1 − k < (n − 1 − k ) a1
1 1
a + −
i . e. , an < (n − k ) a1 , n < n − k 2n + 1 2n + 2
a1
www.pdfworld.in
www.pdfworld.in
≥
1 1 + K + 1  + 1 + K + 1  Thus, for any α > 1the sum
   
n+1 2 2n   2 2n  1 1 1
1+ α + α + K + α
1 1 1 2 3 n
+ −
n (n + 1) 2n + 1 2n + 2 remains bounded for arbitrarily large n it follows
If suffices to show that that for α ≤ 1the sum 1 + 1 / 2 α + 1 / 3 α + K + 1 / n α
can be made arbitrarily large by taking a
1 + K + 1  1
+
1
−
1
  sufficiently large value of n.
2 2n  n (n + 1) 2n + 1 2n + 2
1 1 15. Since, λ is a root of the equation
≥ ⋅ x 3 + ax 2 + bx + c = 0, we have
n + 1 2n + 2
lhs ≥
1
+
1
−
1 λ3 = − aλ2 − bλ − c
2n (n + 1) 2n + 1 2n + 2 This implies that
1 1
= + λ4 = − aλ3 − bλ2 − cλ
2n (n + 1) (2n + 1)(2n + 2)
1 1 = (1 − a )λ3 + (a − b )λ2 + (b − c ) λ + c
= +
2n (n + 1) 2(n + 1)(2n + 2) where we have used again
1 1 1  − λ3 − aλ2 − bλ − c = 0
= +
2(n + 1)  n 2n + 1  Suppose | λ | ≥ 1. Then we obtain
3n + 1 3n + 1 | λ|4 ≤ (1 − a )| λ|3 + (a − b )| λ|2 + (b − c ) | λ| + c
= =
2(n + 1)n (2n + 1) 2n (n + 1)(2n + 1)
≤ (1 − a ) | λ|3 + (a − b ) | λ|3 + (b − c )| λ|3 + c | λ|3
1 1
≥ ⋅ ≤ | λ|3
n + 1 2n + 2
This shows that | λ| ≤ 1. Hence, the only
14. Let us consider the sum possibility in this case is | λ| = 1. We conclude
1 + 1 / 4 + 1 / 9 + K + 1 / n 2 with n smaller that | λ| ≤ 1 is always true.
than 2k + 1 and also the sum
16. Note that the inequality is symmetric in a , b , c
1 + 1 / 2 + 1 / 32 + K + 1 / (2k + 1 − 1)2.
2
so that we may assume that a ≥ b ≥ c . Since
On grouping the terms abc = 1, it follows that a ≥ 1 and c ≤ 1. Using
b = 1/ac , we get
1 +  2 + 2  +  2 + 2 + 2 + 2  + K +
1 1 1 1 1 1
2 3   4 5 6 7  ab + c c a + b
ab + c b c + a c a + b = c + a
 1 1 1  a + cc + a
 k2 + k + K + k +1 2
(2 + 1) − 1)  c b −c
2
 (2 ) (2
= ≤1
a a −b
< 1 +  2 + 2  +  2 + 2 + 2 + 2  + K +
1 1 1 1 1 1
2 2  4 4 4 4  because c ≤ 1, b ≥ c , a ≥ 1 and a ≥ b.
 1 + 1 + K+ 1  17. We have, from the AM–GM inequality, that
 k2 
 (2 ) (2k )2 (2k )2  2
x + y
1 xy ≤   =1
1 − k +1  2 
1 1 1 2 1
=1+ + + K+ k = =2− k <2
2 4 2 1 2 Thus, we obtain 0 < xy ≤ 1. We write
1−
2 x 3y 3 (x 3 + y 3 ) = (xy )3 (x + y )(x 2 − xy + y 2 )
which is what we intended to prove.
= 2(xy )3 ((x + y )2 − 3xy )
Remark = 2(xy )3 (4 − 3xy )
In a completely similar manner we can show that if
Thus, we need to prove that
α is a number greater than 1, then
(xy )3 (4 − 3xy ) ≤ 1
1 1 1 2 α −1
1 + α + α + K + α < α −1
2 3 n 2 −1 Putting z = xy , this inequality reduces to
z3 (4 − 3z ) ≤ 1
for any n
www.pdfworld.in
www.pdfworld.in
Inequalities 353
for 0 < z ≤ 1. We can prove this in different Now, we get

ways. We can put the inequality in the form 9 x 25
4+ <4+ <4+
3z 4 − 4z3 + 1 ≥ 0 17 α 43
9 x 25 43x 17x
Here, the expression in the LHS factors to So, < < ; that is, <α<
17 α 43 25 9
(z − 1)2 (3z 2 + 2z + 1) and (3z 2 + 2z + 1) is
positive since its discriminant D = − 8 < 0. Or We find the smallest value of x for which α
applying the AM–GM inequality to the positive becomes a well-defined integer. For x = 1, 2, 3
reals 4 − 3z , z , z , z we obtain the bounds of α are respectively
4 − 3z + 3z 
4  1 18 , 1 8  ,  3 11 , 3 7  ,  5 4 , 5 2 
     
z3 (4 − 3z ) ≤   ≤1  25 9   25 9  9 3
 4 
None of these pairs contain an integer
18. It is easy to observe that there is a triangle between them.
b c a
with sides a + , b + , c + . Using Heron’s For x = 4, we have
2 2 2 43x 12 17x 5
formula, we get =6 and =7 .
25 25 9 9
16[ABC ]2 = (a + b + c )(a + b − c ) (b + c − a ) Hence, in this case
(c + a − b ) α = 7 and β = 4α + x = 28 + 4 = 32
3
and 16[A1B1C1 ] = 2
(a + b + c ) (− a + b + 3c ) This is also the least possible value, because, if
16 43x 43
x ≥ 5, then α > ≥ > 8 and so β > 37.
(− b + c + 3a )(−c + a + 3b ) 25 5
Since a , b , c are the sides of a triangle, there Hence, the minimum possible value of β is 32.
are positive real numbers p , q , r such that
20. (i) Let n ≥ 40112 and m ∈ N be such that
a = q + r , b = r + p , c = p + q. Using these
relations, we obtain m ≤ n < (m + 1) . Then
2 2
2  1 + 1  n − (m + 1)2
[ABC ]
=
16pqr  
 2005 
[A1B1C1 ]2 3(2p + q )(2q + r )(2r + p )
≥  1 +
1  2
 m − (m + 1)
2
Thus, it is sufficient to prove that  2005 
(2p + q )(2q + r )(2r + p ) ≥ 27pqr m2
= − 2m − 1
for positive real numbers p , q , r . Using 2005
AM–GM inequality, we get 1
= (m 2 − 4010m − 2005)
2p + q ≥ 3 ( p 2q )1/3 , 2q + r ≥ 3 (q 2r )1/3 , 2005
1
2r + p ≥ 3(r 2p )1/3 = [(m − 2005)2 − 20052 − 2005]
2005
Multiplying these, we obtain the desired 1
≥ [(4011 − 2005)2 − 20052 − 2005]
result. We also observe that equality holds if 2005
and only if p = q = r . This is equivalent to the 1
= (20062 − 20052 − 2005)
statement that ABC is equilateral. 2005
1 2006
19. We have = (4011 − 2005) = >0
77 β 197 2005 2005
< <
Thus, we get n < (m + 1)2 <  1 +
1 
17 α 43 n
 2005 
That is,
and l 2 = (m + 1)2 is the desired square.
9 β 25
4+ < <4+
17 α 43 (ii) We show that M = 40102 + 1 is the
β required least number. Suppose n ≥ M .
Thus, 4 < < 5. Since, α and β are positive
α Write n = 40102 + k , where k is a positive
integers, we may write β = 4α + x , where integer. Note that we may assume
0 < x < α. n < 40112 by part (i). Now,
www.pdfworld.in
www.pdfworld.in
 1 + 1  n − 40112 This follows from AM–GM inequalities

 
 2005  x + y ≥ 2 xy , y + z ≥ 2 yx , z + x ≥ zx
=  1 +
1  Aliter
 (4010 + k ) − 4011
2 2
 2005  Let us introduce x + y = c , y + z = a and
k z + x = b. Then a , b , c are the sides of a
= 40102 + 2 ⋅ 4010 + k + − 40112
2005 triangle. If s = (a + b + c ) / 2, then it is easy to
k calculate x = s − a , y = s − b , z = s − c and
= (4010 + 1)2 + (k − 1) + − 40112
2005 x + y + z = s. We also observe that
= (k − 1) +
k
>0 x 2 + xy + y 2 = (x + y )2 − xy
2005 1
= c2 − (c + a − b )(c + b − a )
Thus, we obtain 4
40102 < n < 40112 <  1 +
1  3 1 3
 n. = c 2 + (a − b )2 ≥ c 2
 2005  4 4 4
We check that M = 40102 will not work. Moreover,
For suppose n = 40102. . Then xy + yz + zx = (s − a )(s − b ) + (s − b ) (s − c )
+ (s − c ) (s − a )
 1 + 1  40102 = 40102 + 2 ⋅ 4010
 
 2005  Thus, it is sufficient to prove that
= 40112 − 1 < 40112 9
s Σ (s − a ) (s − b ) ≤ abc
8
Thus, there is no square integer between n and
 1 + 1  n. But Σ(s − a )(s − b ) = r (4R + r ), where r , R are
 
 2005  respectively the inradius, the circumradius of
This proves (ii). the triangle whose sides are a , b , c and
abc = 4Rrs. Thus, the inequality reduces to
21. We begin with the observation that
9
x 2 + xy + y 2 r (4R + r ) ≤ Rr
2
3 1 3
= (x + y )2 + (x − y )2 ≤ (x + y )2 , This is simply 2r ≤ R. This follows from
4 4 4
IO 2 = R (R − 2r ), where I is the incentre and O
and similar bounds for
the circumcentre.
y 2 + yz + z 2 , z 2 + zx + x 2
Aliter
Thus,
3(x 2 + xy + y 2 )( y 2 + yz + z 2 ) (z 2 + zx + x 2 ) If we set x = λa , y = λb , z = λc , then the
81 inequality changes to
≥ (x + y )2 ( y + z )2 (z + x )2
64 (a + b + c )2 (ab + bc + ca )2
Thus, it is sufficient to prove that
≤ 3 (a 2 + ab + b 2 )(b 2 + bc + c 2 )
(x + y + z )(xy + yz + zx )
(c 2 + ca + a 2 )
9
≤ (x + y )( y + z ) (z + x ) This shows that we may assume x + y + z = 1.
8
Let α = xy + yz + zx. We see that
Equivalently, we need to prove that
x 2 + xy + y 2 = (x + y )2 − xy
8(x + y + z )(xy + yz + zx )
≤ 9 (x + y )( y + z ) (z + x ) = (x + y ) (1 − z ) − xy
However, we note that =x + y −α =1− z −α
(x + y )( y + z )(z + x ) Thus, Π(x 2 + xy + y 2 )
= (x + y + z ) ( yz + zx + xy ) − xyz = (1 − α − z ) (1 − α − x ) (1 − α − y )
Thus, the required inequality takes the form = (1 − α )3 − (1 − α )2 + (1 − α ) α − xyz
(x + y )( y + z )(z + x ) ≥ 8 xyz = α 2 − α3 − xyz
www.pdfworld.in
www.pdfworld.in
Inequalities 355
Thus, we need to prove that Remark

α 2 ≤ 3(α 2 − α3 − xyz ). This reduces to The best constant θ in the inequality
3xyz ≤ α 2 (2 − 3α ) a 2 + b 2 − c 2 ≥ θ (c − a) (c − b), where a, b, c are
positive reals such that a 3 + b 3 = c 3 , is
However, θ = 2 (1 + 21/ 3 + 2 −1/ 3 ). Here again, there were
3α = 3 (xy + yz + zx ) ≤ (x + y + z )2 = 1, some beautiful solutions given by students.
1. We have
so that 2 − 3α ≥ 1. Thus, it suffices to prove that
a3 = c 3 − b3 = (c − b )(c 2 + cb + b 2 ),
3xyz ≤ α 2. But
a2 c 2 + cb + b 2
α 2 − 3xyz = (xy + yz + zx )2 − 3xyz (x + y + z ) which is same as =
c −b a
= Σ x 2y 2 − xyz (x + y + z )
cyclic Similarly, we get
1 b2 c 2 + ca + a 2
= Σ (xy − yz )2 ≥ 0 =
2 cyclic c −a b
22. The given inequality may be written in the We observe that
form a2 b2 c (a 2 + b 2 ) − a3 − b3
+ =
7c − 6(a + b ) c − (a + b − 6ab ) < 0.
2 2 2
c −b c −a (c − a )(c − b )
Putting x = 7c 2 , y = − 6 (a + b ) c , c (a 2 + b 2 − c 2 )
=
(c − a )(c − b )
z = − (a 2 + b 2 − 6ab ),
This shows that
we have to prove that x + y + z < 0. Observe
that x , y , z are not all equal (x > 0, y < 0). Using a 2 + b 2 − c 2 c 2 + cb + b 2 c 2 + ca + a 2
= +
the identity (c − a )(c − b ) ca cb
1 Thus, it is sufficient to prove that
x 3 + y 3 + z3 − 3xyz = (x + y + z )
2 c 2 + cb + b 2 c 2 + ca + a 2
+ ≥6
[(x − y )2 + ( y − z )2 + (z − x )2 ]. ca cb
we infer that it is sufficient to prove However, we have
x + y + z − 3xyz < 0
3 3 3
c 2 + b 2 ≥ 2cb and c 2 + a 2 ≥ 2ca.
Substituting the values of x , y , z we see that c 2 + cb + b 2 c 2 + ca + a 2
Hence, +
this is equivalent to ca cb
343c 6 − 216(a + b )3c 3 − (a 2 + b 2 − 6ab )3
≥ 3  +  ≥ 3 × 2 = 6
b a
− 126c (a + b ) (a + b − 6ab ) < 0
3 2 2 a b 
We have used AM–GM inequality.
Using c = a + b3 , this reduces to
3 3
2. Let us set x = a/c and y = b/c . Then

343(a3 + b3 )2 − 216(a + b )3 (a 2 + b3 ) x 3 + y 3 = 1 and the inequality to be proved is
− (a 2 + b 2 − 6ab )3 − 126 [(a3 + b3 ) x 2 + y 2 − 1 > 6 (1 − x ) (1 − y )
(a + b ) (a 2 + b 2 − 6ab )] < 0 This reduces to
This may be simplified (after some tedious (x + y )2 + 6 (x + y ) − 8 xy − 7 > 0 ...(i)
calculations) to But 1 = x + y = (x + y )(x − xy + y )
3 3 2 2
− a 2b 2 (129a 2 − 254ab + 129b 2 ) < 0

which gives xy = [(x + y )3 − 1] / 3(x + y ).
But 129a − 254ab + 129b
2 2
Substituting this in (i) and introducing
x + y = t , the inequality takes the form
= 129(a − b ) + 4ab > 0
2
8 (t 3 − 1)
Hence, the result follows. t 2 + 6t − ⋅ − 7> 0 ...(ii)
3 t
www.pdfworld.in
www.pdfworld.in
This may be simplified to − c 6 − (a + b ) c 5 − abc 4 + (a3 + b3 ) c 3

− 5t + 18 t − 2t + 8 > 0
3 2
+ (a 4 + a3b + a 2b 2 + ab3 + b 4 ) c 2
Equivalently (a 2b + ab 2 + a3 + b3 ) abc + (a 4b 2 − 6a3b3
− (5t − 8)(t − 1)2 > 0
+ a 2b 4 ) > 0
Thus, we need to prove that 5t < 8 . Observe
Substituting
that (x + y )3 > x 3 + y 3 = 1, so that t > 1. We also
have c 3 = a3 + b3 , c 4 = c (a3 + b3 ),
x + y  ≤ x + y = 1
3 3
c 5 = c 2 (a3 + b3 ), c 6 = (a3 + b3 )2
 
 2  2 2
the inequality further reduces to
This shows that t 3 ≤ 4. Thus,
a 2b 2 (a 2 + b 2 + c 2 + ac + bc − 6ab ) > 0
3
 5t  ≤ 125 × 4 = 500 < 1.
  Thus, we need to prove that
 8 512 512
a 2 + b 2 + c 2 + ac + bc − 6ab > 0
Hence, 5t < 8, which proves the given
inequality. Since, a 2 + b 2 ≥ 2ab , it is enough to prove that
3. We write b3 = c 3 − a3 and a3 = c 3 − b3 c 2 + c (a + b ) − 4ab > 0. Multiplying this by c
so that and using a3 + b3 = c 3 , we need to prove that
b3 a3 a3 + b3 + c 2a + c 2b > 4abc
c −a = ,c − b =
c − ca + a
2 2
c − cb + b 2
2
Using AM–GM inequality to these 4 terms and

Thus, the inequality reduces to
using c > a , c > b, we get
a3b3
a2 + b2 − c 2 > 6 ⋅ a3 + b3 + c 2a + c 2 b > 4 (a3b3c 2ac 2b )1/ 4
(c 2 − ca + a 2 )(c 2 − cb + b 2 )
= 4abc
This simplifies (after some lengthy
calculations) to which proves the inequality.
www.pdfworld.in
www.pdfworld.in
Unit 4
Combinatorics
Basic Rule of Counting
In discrete mathematics and everyday walk of life, we always face the problem of counting the things or
finding the number of ways in which we can perform a certain activity.
Let us have some examples
1. Suppose there are 3 junior colleges attached to senior colleges 5 junior colleges attached to schools
and 6 separate junior colleges. Then, how many options are there to Mr. X for taking admission to his
son in a junior college ?
2. Suppose there are 10 locks and corresponding 10 keys. A child insert the keys into the locks at
random. In how many different ways can a child do this?
In how many cases all the key will go wrong ?
3. If we write 1 to 100000, how many times the digit 7 will be written?
4. In how many different ways, the letters of ‘WWW DOT COM’ can be arranged in a row? Out of these
arrangements, how many of them have vowels in symmetric positions?
5. How many different 6 digits numbers can be formed by using six digits 4, 4, 2, 2, 2, 5?
These examples varies from very simple childish problems to harder one.
Combinatorics (Combinatorial Analysis) is the branch of mathematics which develops the rules and find
solutions to such problems.
There are two basic rules.
1. The Rule of Addition
2. The Rule of Multiplication
Before discussing the rule, let us take the above example 1 of Mr. X
As there are 3 junior colleges attached to senior college, 5 junior colleges attached to school and 6
separate junior colleges.
So, Mr. X has 3 + 5 + 6 = 14 options while seeking the admission to his son in a junior college. This is the
rule of addition.
The Rule of Addition

If a collection of objects consists of r1 distinct objects of type 1, r2 distinct objects of type 2, …, rn
distinct objects of type n, then the total number of options to pick an object from the collection is
r1 + r2 + r3 + … + rn .
This rule can be put in set notation also.
Let us | S | denote the number of elements in the set S, then
www.pdfworld.in
www.pdfworld.in
If S1 , S 2 , K , Sn are pair-wise disjoint sets, then

| S 1 ∪ S 2 ∪ K ∪ Sn | = | S1| + | S 2| + K + | Sn |
n sets must be pair-wise disjoint.
If this condition is violated, then there are chances of misuse of rule.
Let us take the example 2 in which a child is playing with 10 locks and 10 keys. His experiment of
inserting the keys into the locks consists of 10 successive activities.
First activity To insert the first key into one of the available locks.
Second activity To insert the second key into one of the available locks.
Now, while performing the first activity, i.e., to insert the first key k1 he has 10 available options viz., 10
locks L1 , L2,…, L10. First activity can be performed in 10 different ways. When second activity of the
experiment starts, there are 9 locks available to insert the second key. Hence, second activity can be
performed in 9 different ways.
The total number of ways of performing the first and second activity is 10 × 9 = 90 continuing, in the
same way the 10 successive activities can be performed in 10 × 9 × 8 × 7 × 6 × 5 × 4 × 3 × 2 × 1 = 3628800
different ways.
This is the rule of multiplication.
Proof of the Rule of Addition (By Principle of Induction)

Corresponding to a +ve integer ‘n’, consider the statement
P (n ) : If S1 , S 2 , K , Sn are pair-wise disjoint sets then
| S1 ∪ S 2 ∪ K ∪ Sn | = | S1| + | S 2| + K+ | Sn |
If n = 1, then there is only one set
S1 = {a1 , a 2 , K , a k }
and one object can be chosen in k ways, so P(1) is true.
If S1 = {a1 , a 2 , K , a k }
S 2 = {b1 , b2 , K , bl }
are two disjoint sets
i.e., ai ≠ b j , then
S1 ∪ S 2 = {a1 , a 2 , K a k , b1 , b2 , K , bl }
| S1 ∪ S 2| = k + l = | S1| + | S 2|
So, P (2) is also true.
Assume that P (n ) is true.
Now, consider n + 1 sets S1 , S 2 , K , Sn , Sn + 1, so that any element of Sn +1 is not previously listed.
Then, | S1 ∪ S 2 ∪ K ∪ Sn ∪ Sn + 1|
= | ( S1 ∪ S 2 ∪ K ∪ Sn ) ∪ Sn + 1|
= | S1 ∪ S 2 ∪ K ∪ Sn | + | Sn + 1|
= | S1| + | S 2| K + | Sn | + | Sn + 1| (since by hypothesis)
Thus, P (n + 1) is true whenever P (n ) is true.
Hence, by the principle of induction P (n ) is true for every +ve integer ‘n’.
The Rule of Multiplication

If an experiment consists of n successive activities, where the first activity can be performed in r1
different ways, second activity performed in r2 different ways, … and nth activity can be performed in rn
different ways.
Then, the total number of ways in which experiment can be performed is r1 , r2 , K , rn
www.pdfworld.in
www.pdfworld.in
Combinatorics 359
In set notation, this rule is stated as.

If S1 , S 2 , K , Sn are n sets, then
| S1 × S 2 × K × Sn | = | S1| | S 2| K | Sn |
Here again it is to be remembered that the n successive activities must be different from each other,
otherwise the application of the rule results in wrong answer.
Proof of the Rule of Multiplication (By Induction Method)

Corresponding to +ve integer n ≥ 2
Consider the statement P (n ) : | S1 × S 2 × K × Sn | = | S1| | S 2| ...| Sn |
If S1 = {a1 , a 2 , K , a k }
S 2 = {b1 , b2 , K , bl }
Then, S1 × S 2 consists of kl ordered pairs
(a1 , b1 ), K , (a1 , bl ), K , (a k , b1 ),…, (a kbl ).
| S1 × S 2| = kl = | S1| | S 2|
This shows that P (2) is true.
Assume that P (n ) is true i.e.,
| S1 × S 2 × K × Sn | = | S1| | S 2| K | Sn |
Then, for next +ve integer n + 1 , we have
| S1 × S 2 × K × Sn × Sn + 1| = | (S1 × S 2 × K × Sn ) × Sn + 1|
= | S1 × S 2 × K × Sn | | Sn + 1|
= | S1| | S 2| K | Sn | | Sn + 1| (since by hypothesis)
Thus, P (n + 1) is true whenever P (n ) is true.
Hence, by the principle of induction, P (n ) is true for every integer n ≥ 2.
Example 1 Given 8 different Physics books, 7 different Chemistry books and 5 different
Mathematics books. How many ways are there to select one book ?
Solution We have to choose one book.
It can be
A Physics book = 8 choices
A Chemistry book = 7 choices
A Mathematics book = 5 choices
Also, all the books are different.
∴ By addition principle
Total number of choices = 8 + 7 + 5 = 20
Example 2 How many 4 digit numbers (with possible repetition) can be formed with no digit less
than 4 ?
Solution The digits used in forming a 4 digit number are 4, 5, 6, 7, 8, 9 i.e., 6 digits.
∴ Out of 4 places the left most place can be filled in 6 ways.
Second left place can be filled in 6 ways.
Third left place can be filled in 6 ways.
Right most place can be filled in 6 ways.
So, by using multiplication rule, the required number of 4 digit number is
6 × 6 × 6 × 6 = 1296
www.pdfworld.in
www.pdfworld.in
Example 3 How many non-empty collections are possible by using 5 P’s and 6 Q’s?
Solution We have 5 P ’s and 6 Q’s, so any non-empty collection will contain 1 to 11 items.
Let p denotes the number of P ’s and q denotes the number of Q’s in the collection,
then
p = 0, 1, 2, 3, 4, 5
and q = 0, 1, 2, 3, 4, 5, 6
The number of pairs of the type ( p, q ) is, then 6 × 7 = 42
Out of these 42 pairs, the one pair viz., (0, 0) gives empty collection.
Hence, the number of non-empty collections is 42 − 1 = 41
Example 4 Four digit numbers are formed by using the digits 1, 2, 3, 4, 5 with possible repetitions.
Find how many of them are divisible by 4?
Solution Q4 digit number is divisible by 4, the unit’s place must be either 2 or 4.
We consider the following two cases.
Case I Unit’s place is 2
Q The number is divisible by 4. Ten’s place must be 1 or 3 or 5.
∴ There are 3 choices for ten’s place. Now there are 5 choices for hundred’s place and
5 for thousand’s place.
In this case, the number of 4 digit number is
1 × 3 × 5 × 5 = 75
Case II Suppose unit’s place is 4
Q The number is divisible by 4
∴ The ten’s place must be 2 or 4. Thus, there are 2 choices for ten’s place, then there
are 5 choices for hundred’s place and 5 for thousand’s place. Hence, in this case, the
number of 4 digit numbers is1 × 2 × 5 × 5 = 50. By addition rule, the required number
of 4 digit number is
75 + 50 = 125
Example 5 Three digits numbers are formed by using the digits 1, 2, 3, 4, find how many of them
are there when
(i ) repetition of digits is permitted
(ii ) repetition of digits is not permitted
(iii ) repetition of digits is not permitted but contain the digit 3
(iv ) repetition of digits is permitted and contain the digit 3.
Solution We shall read the digits from left to right.
(i) In this case, there are 4 choices for each of three places.
By multiplication rule, the number of 3 digit number is 4 × 4 × 4 × = 64
(ii) In this case, there are 4 choices for the first place, 3 for second and 2 for third. By
multiplication rule, the number of 3 digit number is 4 × 3 × 2 = 24
(iii) Here, digit 3 will appear either in the first, second or third place but not elsewhere
more.
If 3 appears in the first place, then second place has 3 choices and third place has
2 choices.
www.pdfworld.in
www.pdfworld.in
Combinatorics 361
By multiplication rule, the number is 3 × 2 = 6

If 3 appears in second place, then first place has 3 choices and third place has 2
choices. By multiplication rule, the number is 3 × 2 = 6.
If 3 appears in third place, then first place has 3 choices and second place has 2
choices. By multiplication rule, the number is 3 × 2 = 6.
Now, the above 3 cases are pair wise disjoint
∴ By addition rule, the required number of 3 digit numbers is 6 + 6 + 6 = 18.
(d) Here the digit 3 will appear once, twice or thrice. We have 3 cases.
Case I First appearance of 3 is in the first place. Then, second place has 4 choices and 3rd place has 4
choices. By multiplication rule, the number of 3 digit numbers is 4 × 4 = 16
Case II First appearance of 3 is in the second place.
Then first place has 3 choices and third place has 4 choices.
By multiplication rule, the number of 3 digits numbers is3 × 4 = 12
Case III First appearance of 3 is in the third place. First place has 3 choices and second place has 3
choices. By multiplication rule, the number of 3 digit numbers is3 × 3 = 9
By addition rule, the total number of 3 digit numbers is16 + 12 + 9 = 37
Example 6 How many ways are there to pick a sequence of two different letters of the alphabets
from the word BOAT and from MATHEMATICS. How many ways are there to pick first a
vowel and then a consonant from each of these words ?
Solution We are to choose 2 different letters from the 4 letters B, O, A, T. First letter can be
chosen in 4 ways since it can be any one of B, O , A, T .
The second letter can be chosen in 3 ways, since it has to be different from the first, so
it can be any one of the remaining 3 letters.
By multiplication principle, total number of way to pick a 2 letter sequence is
4 × 3 = 12
Similarly, from the 8 different letters M , A, T , H, E , I , C and S, a 2 letters sequence can
be picked in 8 × 7 = 56 ways.
Now, in the word BOAT , there are 2 vowels O and A and B, T as consonants.
Hence, first a vowel and then a consonant can be chosen in 4 ways.
Similarly, number of ways in the second case is 3 × 5 = 15 (3 vowels, 5 consonants).
Example 7 A new flag has to be designed with 6 vertical stripes using some or all of the colours
yellow, green, blue and red. In how many ways can this be done so that no two
adjacent stripes have same colour.
Solution Let a, b, c, d , e, f denote the 6 vertical stripes in order from the left.
Then, a can be of any one of the 4 colours, b can be any one of the other 3 colours, c
can be any one of the 3 colours.(as colour used in a can be used in c)
Similarly, there are 3 possible colours for each of d , e, f .
Hence, by multiplication principle, there are 4 × 35 ways of designing the flag.
www.pdfworld.in
www.pdfworld.in
Principle of Inclusion-Exclusion
Use of Venn Diagrams
In some counting problems, it is required to count the subsets of outcomes that possess or do not
possess the combination of various properties.
The ‘Venn Diagrams' are very much useful in such cases to describe the situation pictorially.
In the forthcoming discussions throughout U stands for the universal set which contains all the sets
under discussion. Also we assume that U contains N elements, i.e., |U | = N .
Let us consider the first and the simplest case when the problems of counting takes care of only one
property.
For example : Suppose out of 50 students in a class, 20 of them offer Mathematics as a major subject.
Here, universal set consists of 50 students and the set M consists of 20 students, who have offered
Mathematics as a major subject, so that N = |U | = 50 and | M | = 20 U 30
20
It can be represented by a venn diagram also.
From the above venn diagram, we can easily conclude that there are 30 students
who have not offered Mathematics as a major subject.
If M ′ denotes the complement of M in the universal set U, then | M ′ | = 30
M
i.e., | M ′| = N − | M|
If a problem involves 2 properties sayx 1 and x 2 , then we consider two sets A and B
A = {x : x satisfy property x 1}
B = {y : y satisfy property x 2}
It can happen that some elements in the universe will satisfy both the properties and some will satisfy
none of the two properties. This situation can be represented by venn diagram.
To find the number of elements which satisfy either propertyx 1 or x 2 or both. U A'∩B'
We add | A | and | B |
A B
But in this sum of the elements of A ∩ B are counted twice, so we make the
correction by subtracting | A ∩ B | A∩B
Thus, |A ∪ B| = |A| + |B| − |A ∩ B|

Now, we find the number of elements not satisfying both the properties.
i.e., | A ′ ∩ B′|
From De-Morgan’s law, we have
(A ∪ B )′ = A ′ ∩ B ′
∴ | A ′ ∪ B ′ | = | (A ∪ B )′ |
∴ | A ′ ∩ B′| = N − | A ∪ B |
∴ | A ′ ∩ B ′ | = N − (| A | + | B | − | A ∩ B | )
∴ | A ′ ∩ B′| = N − | A | − | B | + | A ∩ B |
Inclusion–Exclusion Principle for Three Sets

U
If A , B , C are three sets of elements satisfying the properties x 1 , x 2 , x 3
A B
respectively. By using Venn Diagram, it can be shown as below.
Now, we find | A ∪ B ∪ C | A∩B∩C
| A ∪ B ∪ C | = | (A ∪ B ) ∪ C |
0 = | A ∪ B | + | C | − | (A ∪ B ) ∩ C | C
= | A | + | B | − | A ∩ B | + | C | − | (A ∩ C ) ∪ (B ∩ C )|
∴ | A ∪ B ∪ C | = | A | + | B | + | C | − | A ∪ B | − (| A ∩ C | + | B ∩ C | − | A ∩ C ∩ B ∩ C | )
www.pdfworld.in
www.pdfworld.in
Combinatorics 363
∴ | A ∪ B ∪ C | = | A | + | B | + |C | − | A ∩ B | – | A ∩ C | − | B ∩ C | + | A ∩ B ∩ C |
From this it follows that | A ′ ∩ B′ ∩ C′ | = N − | A ∪ B ∪ C |
∴ | A ′ ∩ B′ ∩ C′ | = N − | A | − | B | − | C | + | A ∩ B | + | A ∩ C | + | B ∩ C | + | B ∩ C | − | A ∩ B ∩ C |
Example 1 How many integers between 1 and 567 are divisible by either 3 or 5 ?
Solution Let Z = {1, 2, 3, K , 567}
P = {x ∈ Z / 3 divides x}
Q = {x ∈ Z / 5 divides x}
We have to find | P ∪ Q |
Now since 567 = 189 × 3 ,
The set P of multiples of 3 in Z contains 189 numbers i.e.,
| P | = 189
Similarly, since 567 = 113 × 5 + 2
The set Q of multiples of 5 in Z contains 113 number. i.e.,
| Q | = 113
567 = 37 × 15 + 12
The set P ∩ Q of multiples of both 3 and 5 i.e., multiples of 15 in Z contains 37
numbers i.e.,,
| P ∩ Q | = 37
Hence, | P ∪ Q | = 189 + 113 − 37 = 265
Example 2 105 students appeared in an examination. Out of which 80 students pass in English, 75
students pass in Mathematics and 60 students pass in both subjects. How many
students fail in both subjects ?
Solution Let Z = the set of students appeared in the examination.
E = the set of students pass in English
and M = the set of students pass in Mathematics
We are given
n( Z ) = 105
n(E ) = 80
n(M ) = 75 and n(E ∩ M ) = 60
Q n(E ∪ M ) = n(E ) + (M ) − n(E ∩ M )
∴ n(E ∪ M ) = 80 + 75 − 60 = 95
Required number = n( Z ) − n(E ∪ M ) = 105 − 95 = 10
So, 10 students fail in both subjects.
Example 3 How many integers between 999 and 9999 either begin or end with 3 ?
Solution Let Z = the set of 4 digit numbers.
P = {x ∈ Z / x begins with 3}
Q = {x ∈ Z / x ends with 3}
We have to find | P ∪ Q |.
www.pdfworld.in
www.pdfworld.in
If a 4 digit number begins with 3, then each of its remaining 3 digits can be chosen
in 10 ways and so by multiplication principle
| P | = 103
Similarly if a 4 digit number ends with 3, then its leading digit being non-zero can be
chosen is 9 ways and each of its remaining two digits can be chosen in 10 ways, so by
multiplication principle
| Q | = 9 × 10 × 10 = 900
Finally, if a 4 digit number begins and ends with 3, then each of its remaining 2
digits can be chosen in 10 ways, so by multiplication principle.
| P ∩ Q | = 10 × 10 =100
∴ | P ∪ Q | = 1000 + 900 − 100 = 1800
Example 4 How many arrangements of the digits 0, 1, 2, …, 9 are there that do not end with 8 and
do not begin with 3 ?
Solution The total number of arrangements of 10 digits 0, 1, 2, … ,9 in a row are 10!. This is
universal set.
Let A = {arrangement ending with 8}
B = {arrangements beginning with 3}
We have to find | A′ ∩ B′ |
If an arrangement ends with 8, they are 9! in number.
If an arrangement begins with 3, they are also 9! in number, so | A | = 9!, | B | = 9!
and | A ∩ B | = 8!
Now, | A′ ∩ B′ | = N − | A| − | B| + | A ∩ B|
= 10! − 9! − 9! + 8!
= 8!( 90 − 9 − 9 + 1) = (73)( 8!) = 2943360
Example 5 How many integers between 1 and 567 are divisible by either 3 or 5 or 7?
Solution Let Z = {1, 2, K ..., 567}
A = {x ∈ Z / 3 divides x}
B = {x ∈ Z / 5 divides x}
C = {x ∈ Z / 7 divides x}
Now, | Z | = 567
If [ x ] denotes the integral part of x, then
| A | = [ 567 / 3] = 189
| B | = [ 567 / 5] = 113
| C | = [ 567 / 7] = 81
| A ∩ B | = [ 567 / 15] = 37
| B ∩ A | = [ 567 / 35] = 16
| C ∩ A | = [ 567 / 21] = 27
| A ∩ B ∩ C | = [ 567 / 105] = 5
Hence,
| A′ ∩ B′ ∩ C′ | = | Z | − | A| − | B| − | C| + | A ∩ B| + | B ∩ C| + | C ∩ A| − | A ∩ B ∩ C|
= 567 − 189 − 113 − 81 + 37 + 16 + 27 – 5 = 308
www.pdfworld.in
www.pdfworld.in
Combinatorics 365
Example 6 Three identical blue balls, four identical red balls and 5 identical white balls are to be
arranged in a row, find the number of ways that this can be done, if all the balls with the
same colour do not form a single block.
Solution Total number of arrangements is n = 12!/ 3! 4! 5!.
Let A, B, C denote the set of arrangements with blue balls together, red ball together
and white balls together. Then, by Inclusion-Exclusion principle, required number of
arrangements is
 10! 9! 8!   7! 5! 6! 
| A′ ∩ B′ ∩ C′ | = n −  + + + + + − 3!
 4! 5! 3! 4! 3! 4!   5! 3! 4! 
Example 7 Find the number of numbers from 1 to 1000 which are neither divisible by 2 nor by 3 nor by
5.
Solution From 1 to N, the number of integers which are divisible by a fixed integer k.
N 
1 < 1 ≤ N is equal to  
k 
where [x] denotes greatest integer ≤ x
Let us define the sets
A : Numbers which are divisible by 2.
B : Number which are divisible by 3.
C : Numbers which are divisible by 5.
1000  1000 
Then, n( A ) =  ; n(B ) =  ;
 2   3 
1000 
n(C ) = 
 5 
n( A ∩ B ) = number of numbers which are divisible by both 2 and 3 (i.e., by 6)
1000 
=
 6 
1000  1000 
n(B ∩ C ) =  ; n(C ∩ A ) = 
 15   10 
1000 
n( A ∩ B ∩ C ) = number of numbers which are divisible by 2, 3 and 5 = 
 30 
Now, n( A ∪ B ∪ C ) = number of numbers which are divisible at least one of the
three numbers 2, 3 or 5.
Then,
n( A ) + n(B ) + n(C ) − n( A ∩ B ) − n(B ∩ C ) −n(C ∩ A ) + n( A ∩ B ∩ C)
1000  1000  1000  1000  1000  1000  1000 
= + + − − − +
 2   3   5   6   15   10   30 
= 500 + 333 + 200 − 166 − 66 − 100 + 33

= 734
∴ The required number = 1000 − 734
= 266
www.pdfworld.in
www.pdfworld.in
General Form of Principle of Inclusion and Exclusion

Theorem Let A1 , A 2 , ... , Am be subsets of a finite set U.
Let Sr = Σ | Ai 1 ∩ Ai 2 ∩ ... ∩ Ai r |
 
where the sum is taken over the  m  choices of sets of r integers i1 , ... ir such that
r 
1 ≤ i1 < i 2 < ... < ir ≤ m
Then
(i) | A1 ∪ A 2 ∪ .. ∪ Am | = S1 − S 2 + ... + (−1) m − 1Sm
(ii) | A ′1 ∩ A ′2 ∩ .. ∩ A ′m | = |U | − S1 + S 2 ... + (−1) m Sm
Dearrangements
Consider n distinct objects ai , 1 ≤ i ≤ n, arranged in a row in the following order a1 , a 2 , ... , an . Then, a
dearrangement of these objects is a permutation in which no object is in its original position i.e, a1 is not
in the first place, a 2 is not in the second place, ..., an is not in the nth place.
Theorem The number Dn of dearrangements of n distinct objects is given by
Dn = n ! 1 − + K + (−1)n 
1 1 1 1
+ −
 1! 2! 3! n ! 
Proof Let the given objects be denoted by the integers 1, 2, …, n.

Suppose that these are arranged in their natural order.
Let U be the set of all permutations of these integers.
Let Ai denote the set of those permutation in each of which the integer i is in the ith place. Then, it is
clear that
Dn = | A1 ′ ∩ A 2 ′ ∩ K ∩ An ′ |
Now, for each i = 1, 2, K n , | Ai | = (n − 1)! because after putting the integer i in the ith place, the remaining
n − 1 integers can be arranged the remaining (n − 1) places in (n − 1)! ways.
So, S1 = Σ | Ai | = n × (n − 1)!
For 1 ≤ i < j ≤ n, we have
| Ai ∩ A j | = (n − 2)!
because after putting the integers i , j in their respective original places, the remaining n − 2 integers can
be arranged in the remaining (n − 2) places in (n − 2)! ways.
n 
QThere are   pairs Ai A j , we have
 2
n 
S 2 = Σ | Ai ∩ A j | =   (n − 2)!
 2
Similarly, for any setT = {i1 , K , ir } of r integers such that 1 ≤ i1 < i 2 < K< ir ≤ n.
We have | Ai ∩ Ai ∩ K ∩ Air | = (n − r ) !
1 2
There are  n  different r sets T.

r 
Hence, Sr = Σ | Ai ∩ Ai ∩ K ∩ Ai r |
1 2
n 
=   (n − r ) !, Sn = 1
r 
www.pdfworld.in
www.pdfworld.in
Combinatorics 367
Remark
n n
l
  = Cr
 r
Hence, by principle of inclusion-exclusion, the number of dearrangements is
    n
Dn = |U | − S1 + S 2 − S 3 + K + ( −1)n Sn = n ! − n (n − 1)! + n (n − 2)! −   (n − 3)! + K + ( −1)n
 1  2  3
n! n! n!  1 1 1 n 1 
= n! − + − + K + ( −1) = n ! 1 − +
n
− + K + ( −1)
1! 2! 3!  1! 2! 3! n ! 
l If r things goes to wrong place out of n things, then (n − r ) things goes to original place (here r < n)
If Dn = Number of ways, if all n things goes to wrong place.
Dr = Number of ways, if r things goes to wrong place.
If r goes to wrong places out of n, then (n − r ) goes to correct places.
Then, Dn = nCn − r Dr
If at least p of them are in wrong places.
n
Then, Dn = Σ n
Cn − r Dr
r =p
 1 1 1 1
where Dr = r ! 1 − + − + K + ( −1)r 
 1! 2! 3! r !
Example 1 A new employee checks the hats of n people visiting a restaurant, forgetting to put
claim check numbers on the hats, when customer return for their hats, the checker
gives them back hats chosen at random from the remaining hats. What is the
probability that no one receives the correct hat ?
D
Solution The probability that no one receives the correct hat is n
n!
1 1 ( −1)n
=1 − + −K +
1! 2! n!
1 1 1 ( −1)n
Q e −1 = 1 −+ − +K+ +K
1! 2! 3! n!
D
e −1 is a good approximation to n , when n is large i.e.,
n!
Dn n ( −1)k
is approximately equal to Σ
n! k =0 k!
Example 2 While at racetrack Jayesh bets on each of the ten horses in a race to come according
to how they are favoured. In how many ways can they reach the finish line so that he
loses all his bets? What is the probability that he wins at least one bet ?
Solution We actually want to know in how many ways we can arrange the numbers 1, 2, ..., 10,
so that 1 is not in first place (its natural position, 2 is not in second place and so on)
 1 1 1  1
Number is D10 = 10! 1 − + −K +  = 10!
 1! 2! 10! e
The probability that he wins at least 1 bet
10! 1 / e
=1 −
10!
1
= 1 − = 0.632
e
www.pdfworld.in
www.pdfworld.in
Example 3 A person writes letters to 6 friends and addresses the corresponding envelopes. In
how many ways can the letters be placed in the envelopes so that
(i ) at least 2 of them are in the wrong envelopes.
(ii ) all the letters in wrong envelopes.
Solution (i) The number of ways in which at least 2 of them are in wrong places
6
= Σ n
Cn − r Dr
r =2
= nCn − 2D2 + nCn − 3D3 + nCn − 4D4 + nCn − 5 D5 + nCn − 6D6

Here, n = 6
 1 1  1 1 1
= 6C4 2! 1 − +  + 6C3 ⋅ 3! 1 − + − 
 1! 2!  1! 2! 3!
 1 1 1 1  1 1 1 1 1
+ 6C2 4! 1 − + − +  + 6C15! 1 − + − + − 
 1! 2! 3! 4!  1! 2! 3! 4! 5!
 1 1 1 1 1 1
+ 6C0 6! 1 − + − + − + 
 1! 2! 3! 4! 5! 6!
= 15 + 40 + 135 + 264 + 265 = 719
(ii) The number of ways in which all letters be placed in wrong envelopes
 1 1 1 1 1 1
= 6! 1 − + − + − + 
 1! 2! 3! 4! 5! 6!
1 1 1 1 
= 720  − + − 120 +  = 360 − 120 + 30 − 6 + 1
 2 6 24 720
= 265
Example 4 (i) For n ≥ 2, show that Dn = (n − 1)(Dn − 1 + Dn − 2 )

(ii) show that Dn = nDn − 1 + ( −1)n , n ≥ 2
(iii) find D4, K , D9 .
Solution (i) The set of the Dn dearrangement of 1, 2, …, n is the disjoint union of the following
n − 1 sets Ai . Fix an integer i with 2 ≤ i ≤ n. Let Ai be the set of those
dearrangements x = a1a 2 K an for which a1 = i
Now, the dearrangements in Ai are of 2 types
(a) a1 = i and ai = 1 (b) a1 = i and ai ≠ 1
For example, for n = 4, A3 contains 1 dearrangement 3, 4, 1, 2 of type (a) in which
1 and 3 have been interchanged and 2 dearrangements 3, 1, 4, 2 and 3, 4, 2, 1 of
type (b) in which the first element is 3 but 1 is not in the third place.
The number of dearrangements of type (a) is Dn − 2, since in this case 1 and i have
changed places and the remaining n − 2 integers are deranged. Also, the
number of dearrangements of type (b) is Dn − 1, since in this case i is put in the first
place and 1 is not allowed to occupy the place of i and the remaining n − 1
integers including 1 are deranged. Thus
| Ai | = Dn − 2 + Dn − 1, for i = 2, K , n
Hence,
Dn = (n − 1)| A2| = (n − 1)[Dn − 2 + Dn − 1]
www.pdfworld.in
www.pdfworld.in
Combinatorics 369
(ii) Rewrite the formula in (i), thus

Dn − nDn − 1 = − Dn − 1 + (n − 1)Dn − 2
Then, successively, we get
Dn − Dn − 1 = ( −1) [Dn − 1 − (n − 1)Dn − 2 ]
= ( −1)2 [Dn − 2 − (n − 2)(Dn − 3 ]
= ( −1)n − 2[D2 − ( 2)D1]
Thus, Dn − nDn −1 = ( −1)n − 2 [1 − ( 2)( 0)] = [ −1]n
Q D1 = 0, D2 = 1
(iii) QD2 = 1, D3 = 3, using (ii) we get
D4 = 9,D5 = 44, D6 = 265,
D7 = 1854, D8 = 14833, D9 = 133496
Example 5 On a rainy day n people go to a party in a hotel. Each of them leaves his umbrella at the
property counter. Find the number of ways in which the umbrellas are handed back to
them after the party in such a manner that no person recieves his own umbrella.
Solution Let us name the persons as A1, A2, K , An and label their umbrellas as a1, a 2, K , an
respectively.
Let Dn denote the number of ways in which the umbrellas can be returned so that no
person recieves his own umbrella.
To find Dn , we shall obtain a recurrence relation connecting Dn , Dn − 1, Dn − 2.
Step 1. There are n − 1 possible choices for A1 to recieve a wrong umbrella.
Suppose A1 is given the umbrella a 2
Step 2. There are 2 different strategies for disposing off umbrella a1.
(i) Give umbrella a1 to A2. We are left with n − 2 persons A3, A4, K , An and their
umbrellas a 3, a 4, K , an . The desired task of handing over umbrellas can be
accomplished in Dn − 2 ways.
(ii) Do not give umbrella a1 to A2. We have n − 1 persons A2, A3, K , An and n − 1
umbrellas a1, a 3, K , an . The number of ways of giving away the umbrellas so that
A2 does not recieve a1, A3 does not recieved a 3, K , An does not recieve an is Dn − 1.
From the two mutually exclusive and exhaustive cases (i) and (ii), the total number
of ways Dn − 1 + Dn − 2.
Step 3. Since for each of the n − 1 ways of giving a wrong umbrella to A1, there are
Dn − 1 + Dn − 2 ways of giving the remaining umbrellas to the remaining persons so
that no one recieves his own umbrella.
∴ Dn = (n − 1)(Dn − 1 + Dn − 2 )
Step 4. In order to obtain an explicit expression for Dn , we proceed as follows
Dn − nDn −n = (n − 1)Dn − 2 − Dn − 1
= ( −1)[Dn − 1 − (n − 1)Dn − 2 ] = ( −1)2[Dn − 2 − (n − 2)Dn − 3 ]
……………
……………
……………
……………
= ( −1)n − 2(D2 − 2D1)
www.pdfworld.in
www.pdfworld.in
Now, D1 = 0, D2 = 1
n −2
∴ Dn − nDn − 1 = ( −1) = ( −1)n
Dividing throughout by n !, we have
Dn Dn − 1 ( −1)n
− =
n ! (n − 1)! n!
Replacing n successively by n − 1, n − 2, …, 2 and adding corresponding sides, we
have
Dn D1 ( −1)n ( −1)n − 1 ( −1)n − 2 ( −1)2
− = + + +…...
n ! 1! n! (n − 1)! (n − 2)! 2!
Q D1 = 0, we can rearrange the above relation as.
 1 1 1 ( −1)n 
Dn = (n !) 1 − + − + ... + 
 1! 2! 3! n! 
Permutations-Combinations
Theorem 1 The number of r permutations of a set S containing n different objects is denoted by
P (n , r ) or n Pr and is given by
n!
n
Pr = n (n − 1)(n − 2) ... (n − r + 1) =
(n − r )!
At times the word permutation is used for arrangement. P (n , r ) is also called the number of
permutations of n distinct objects taken r at a time. Note that P (n , r ) is also equal to
(a) Number of r digits numbers, which can be made by using n distinct digits and repetition of a digit is
not allowed.
(b) Number of ways in which r persons can stay {n ≥ r } in r hotels (each in a different hotel).
(c) Number of ordered samples of size r without repetitions taken from the set {a1 , a 2 , ... an }.
(d) Number of ways of making r person stand in a line taken from n persons.
(e) Number of one-one functions, which can be defined from a set containing r elements to a another set
containing n elements (n ≥ r ) .
n!
(f) P (n , n ) = = n!
0!
= Number of ways arranging n distinct objects taken all at a time.
Theorem 2 (a) P (n , r ) = n P (n − 1, r − 1)
(b) P (n , r ) = P (n − 1, r ) + rP[(n − 1), (r − 1)]
n! n (n − 1)! n!
Proof (a) LHS = ; RHS = =
(n − r )! (n − 1 − r + 1)! (n − r )!
To prove it physically, we note that LHS is number of r digit numbers which can be made by using n
distinct digits d1 , d 2 , ... , dn , when repetition of a digit is not allowed, fill up the first position by d1 and
fill up remaining ( r − 1) places by the digits d 2 , d3 , ... , dn also (n ways).
Thus, it is evident that
P (n , r ) = n ⋅ P (n − 1, r − 1)
(b) RHS = P (n − 1, r ) + r ⋅ P (n − 1, r − 1)
(n − 1)! r (n − 1)!
= +
(n − 1 − r )! (n − 1 − r + 1)!
www.pdfworld.in
www.pdfworld.in
Combinatorics 371
(n − 1)! (n − 1)!
= +r⋅
(n − 1 − r )! (n − r )!
(n − 1)!  r  n!
= 1 +  = = P (n , r )
(n − 1 − r )!  n − r  (n − r )!
To prove it physically, the LHS again has the same meaning as in (a). The RHS can be explained as
follows.
In forming r digits numbers, the digit d1 is either used or not used. If it is not used, then the number of
numbers which can be formed = P [(n − 1), r ]
If the digit d1 is used, it can be used at r possible positions and in every position of d1, the number of
numbers which can be made must be P (n − 1, r − 1). Number of r digit numbers which can be made from
n distinct digits = Number of numbers which can be used without d1 + Number of such numbers which
can be used with d1
⇒ P (n , r ) = P (n − 1, r ) + r ⋅ P (n − 1, r − 1)
n 
Theorem 3 The number of r combinations of an element set S is denoted by   or n Cr or C (n , r ) and
r 
is given by
n
Pr n!
n
Cr = =
r! r !(n − r )!
Permutations with repetitions
Concept So far we have considered permutations of distinct or unrepeated objects. The effect of
allowing repetition is to decrease the number of permutations.
For example : 2 distinct letters a1 and a 2 can be arranged in one way as a1a 2. Before deriving the general
formula for the number of permutations with repetitions, consider the problem of arranging the 7
letters a , a , b , b , b , c , c in a row.
Here, we are to fill 7 places −, −, −, −, −, −, −
with the given letters. First for placing the a's, 2 of these places can be chosen in 7 C2 ways.
Note that for each such choice we can place the letters a , a in the chosen places in one way only. Hence,
the a’s can be placed in 7 C2 ways.
Then, there remain 7 − 2 = 5 places and the b’s can be placed in 3 of these 5 places in 5C3 ways.
Then, there remain 5 − 3 = 2 places, in which the c’s can be put in 2C2 ways, so by the multiplication
principle, the number of permutations is
7! 5! 7!
7
C2 × 5C3 × 2C2 = × ×1=
2!5! 3!2! 2!3!2!
Alternatively, we can find the number, say m of the 7 permutations of the given letters as follows.
Replace the 2 identical a’s by the distinct letters a1 , a 2. Similarly, replace the b’s by b1 , b2 , b3 and c’s by
c1 , c 2. Then, we get 7 different letters which can be arranged in a row in 7! ways.
Now, these 7! permutations arise out of the required m permutations in the following ways.
Consider any one of the required permutations say t = bacabbc .
On replacing the letters as above, we get t1 = b1a1c1a 2b2b3c 2, as one possible permutation of the distinct
latters.
Now, we can arrange a1 , a 2 in their positions in 2! ways. This gives the following 2 permutations.
t1 = b1a1c1a 2b2b3c 2
t 2 = b1a 2c1a1b2b3c 2
Similarly, b1 , b2 , b3 can be arranged in their positions in 3! ways.
www.pdfworld.in
www.pdfworld.in
c1 , c 2 can be arranged in their positions in 2! ways, so the permutation t give rise to in all 2 ! × 3 ! × 2 !
different permutations of the 7 different letters.
Hence, the required m permutations give m × 2 ! × 3 ! × 2 ! permutations.
This happens because every permutation of the 7 different letters can be obtained by first choosing
places to put the a’s, b’s and c’s and then rearranging the letters in those places in a particular order.
Hence, 7! = m × 2! × 3! × 2!
i.e., as before
7!
m=
2!3!2!
Theorem 4 Suppose there are n objects of which n1 are identical of first type, n2 are identical of
second type, ..., nk are identical of kth type so that
n = n1 + n2 + ... + nk
Then, the number of permutations of these n objects, taken all at a time is denoted by P (n ; n1 , n2 , ... , nk )
and is given by.
 n − n − n − ... − n 
 n   n − n1   1 2 k − 1
P (n , n1 , .. , nk ) =    … …  
 n1   n2   nk 
 
n!
=
n1 ! n2 ! ... nk !
Circular Permutations
A A B B
1 1 2 2
C1 2 B C2 3 C C3 1 A C4 1 A
3 2 3 3
C B C C
C C
3 3
C5 1 A C6 2 B
2 1
B A
Suppose in a circular order, there are n numbered seats numbered 1 through n and we want to make n
persons sit at these seats, then the number of ways must be same as number of ways in a line i.e., the
number of ways must be n !.
We will illustrate this for n = 3. (The persons are denoted by A , B , C ).
Observe carefully that all the 6 cases C1 , C2 , ... , C6 are different.
Now, if the number 1, 2, 3 are removed we can easily note that C1 , C3 and C5 denote the same cases, since
all the cases are equivalent to A , B , C , C2 , C4 and C6 are equivalent to only one case.
A A
C B
B C
www.pdfworld.in
www.pdfworld.in
Combinatorics 373
Thus, 3 persons on a circular order can sit in 2 ways only namely which can be proved theoretically as
follows. Keep A fixed and arranged remaining 2 persons in a line which can be done in 2! ways.
In the general case, we easily conclude that n persons in a circular order can stand in (n − 1)! ways.
We further observe that corresponding to any clockwise arrangement of distinct objects in a circular
order, there is an anticlockwise arrangement.
By the principle of homogeniety, the number of anticlockwise arrangements must be same as number of
clockwise arrangements. Now as there are (n − 1)! arrangements in all, the number of clockwise or
anticlockwise arrangement
1
= (n − 1)!
2
Theorem 5 The number of distinct circular permutation of n different objects is (n − 1)!
Corollary Let n ≥ 3 be a +ve integer. The number of different circular necklaces that can be made from
1
n different beads is (n − 1)!.
2
Example 1 How many numbers each lying between 1000 and 10000 can be formed with the digits
0, 1, 2, 3, 4, 5, no digit being repeated?
Solution Here, number of digits (objects = n = 6).
Numbers lying between 1000 and 10000 are of 4 digits.
Hence, number of places to be filled up = r = 4. Hence, number of different
arrangements of 6 given
But in these 6P4 numbers, some number begin with 0 and hence these numbers are
actually of 3 digits not of our purpose.
Now, we find number of numbers beginning with zero. In this case, number digits
remain to be utilised in (1, 2, 3, 4, 5) = 5 and number of places to be filled up = r = 3
Hence, number of such numbers
5! 5!
= 5P3 = =
( 5 − 3)! 2!
1⋅ 2 ⋅ 3 ⋅ 4 ⋅ 5
= = 60
1⋅ 2
Number of numbers of 4 digits formed by 0, 1, 2, 3, 4, 5 is 360 − 60 = 300
Example 2 Find the number of numbers between 300 and 3000 that can be formed with the digits
0, 1, 2, 3, 4 and 5, no digit being repeated.
Solution We first notice that numbers between 300 and 3000 are of 3 and 4 digits. We first find
the number of number’s of 3 digits.
But we require only these numbers of 3 digits which begin with 3, 4 or 5 for those
numbers. Number of places to be filled up = r = 2 and number of digits remain to be
used = n = 5.
5!
Hence, number of numbers of 3 digits begining with 3 = 5P2 = = 20
3!
Number of numbers of 3 digit begining with 5 = 5P2 = 20
Hence, number of number’s of 3 digits begining with 3, 4 or 5 = 20 + 20 + 20
= 60 …(i)
www.pdfworld.in
www.pdfworld.in
Also, for the number of 4 digits less than 3000, we require only these numbers which
begin with 1 or 2.
In this case, number of places to be filled up = r = 3.
Number of digits remain to be used = n = 5.
Hence, number of numbers of 4 digit begining with 2 = 5P3 = 60
Hence, number of numbers of 4 digits begining with 2 = 60 + 60 = 120 …(ii)
Total number of numbers lying between 300 and 3000 and formed with the digits 0,
1, 2, 3, 4,
5 = 60 + 120 = 180 [from Eqs. (i) and (ii)]
Example 3 How many numbers between 400 and 1000 can be formed with the digits 2, 3, 4, 5, 6,
0? Number lying between 400 and 1000 are of 3 digits only.
Solution Now, we first find the number of numbers of 3 digits and greater than 400. These
number begins with 4, 5 or 6. For these numbers of places to be filled up r = 2 and
number of digits remain to be used = n = 5
5!
Hence, number of numbers of 3 digits beginning with 4 = 5P2 = = 20
3!
Number of numbers of 3 digit beginning with 5 = 5P2 = 20
Number of numbers of 3 digit beginning with 6 = 5P2 = 20
Hence, number of numbers 3 digits beginning with 4, 5 or 6
= 20 + 20 + 20 = 60
∴ 60 numbers are lying between 400 and 1000 formed with digits 2, 3, 4, 5, 6, 0.
Example 4 How many different numbers of 6 digits can be formed with the digits 4, 5, 6, 7, 8, 9?
How many of them are divisible by 6. How many of them are not divisible by 5?
Solution Number of digits = 6 = n
For the numbers of 6 digits number of places to be filled up = r = 6. Hence, total
number of numbers of 6 digits formed with the digits 4, 5, 6, 7, 8, 9. is
6!
6
P6 = = 1⋅ 2 ⋅ 3 ⋅ 4 ⋅ 5 ⋅ 6 = 720 …(i)
0!
Also of these numbers of numbers which end with 5 must be divisible by 5. Now, for
the number of numbers of 6 digits divisible by 5.
Number of digits remain to be utilised = n = 5.
Hence, number of number’s of 6 digits divisible by 5 and formed with the digits 4, 5,
6, 7, 8, 9 is
5!
5
P5 = = 120 …(ii)
0!
Now, number of numbers not divisible by 5 = total number of numbers – number of
numbers divisible by 5
= 720 − 120 = 600
www.pdfworld.in
www.pdfworld.in
Combinatorics 375
Example 5 How many numbers of 6 digits can be formed with the digits 1, 2, 3, 4, 5, 6, in which 5
always occur in ten’s place?
Solution Number of digits = 6
Number of places to be filled up = r = 6
But as 5 always occur in ten place, so that number of places to be filled up = r = 5.
Number of (objects) digits remain to be utilised = n = 5
Hence, number of numbers of 6 digits formed with 1, 2, 3, 4, 5, 6, such that 5
always occurs in the ten’s place is
5!
= 5P5 = = 120
0!
Example 6 Find the sum of all the numbers greater than 10000 formed with the digits 0, 2, 4, 6 and
8. No digit being repeated.
Solution Numbers formed with 0, 2, 4, 6 and 8 and greater than 1000 are of 5 digits.
Keeping 0 at units (or tens or hundreds or thousands place).
Number of places remain to be filled up = r = 4 Number of digits remain to be utilised
= n = 4. Hence, number of numbers in which 0 comes at unit (tens or hundred or
thousands place)
4!
= 4P4 = = 24
0!
i.e., 0 may come at unit (or tens or hundred or thousands place) = 24 number of times.
Also, keeping 2 (or 4 or 6 or 8) at unit (or tens or hundreds or thousands) place.
Number of digits remain (0, 4, 6, 8) to be utilised = n = 4
Hence, number of numbers of 4 digits such that 2 comes at unit place = 4P4 = 24. But
out of these 24 numbers, some numbers begins with 0.
To find the number of such numbers, let us keep 0 at 10000 place and 2 at unit place,
then number of places remain to be filled up r = 3
Number of digits remain to be utilised
=n = 3
Hence, number of numbers of 5 digits in which 2(or 4 or 6 or 8) comes at unit (or tens or
hundred or thousands) place = 24 − 6 = 18
Also, keeping 2 (or 4 or 8) at 10000 place, number of places remain to filled up = r = 4
Number of digits remain to be utilised = 4. Hence, number of numbers in which 2(or 4
4!
or 6 or 8) comes at 10000 place = 4P4 = = 24. Hence, 2(4,6 or 8) comes at unit (or
0!
tens or hundred or thousands) place in 18 number of times.
But 2(4, 6 or 8) comes at 10000 place in 24 number of times.
Hence, sum of the numbers at unit place
= [ 0 × 24 + ( 2 + 4 + 6 + 8) × 18] × 1 …(i)
Sum of numbers at tens place
= [ 0 × 24 + ( 2 + 4 + 6 + 8) × 18] × 10 …(ii)
www.pdfworld.in
www.pdfworld.in
Sum of numbers at hundreds place

= [ 0 × 24 × ( 2 + 4 + 6 + 8) × 18] × 100 …(iii)
sum of numbers at thousands place
= [ 0 × 24 + ( 2 + 4 + 6 + 8) × 18] × 100 …(iv)
sum of numbers at ten thousands place
= [ 0 × 24 + ( 2 + 4 + 6 + 8) × 24] × 10000 …(v)
Adding Eqs. (i), (ii), (iii), (iv) and (v) altogether, we get 5199960.
Hence, sum of all numbers greater than 10000 formed with digits 0, 2, 4, 6, 8 is
5199960.
Example 7 How many number of 4 digits can be formed with the digits 1, 2, 3, 4, 5, when repetition
of digits is allowed.
Solution In the number of 4 digits 4 places are to be filled up by 5 digits.
Now, 1st place can be filled up in 5 ways (out of any one of 1, 2, 3, 4, 5).
Also, 2nd place can be filled up in 5 ways.
Hence, first two places can be filled up in 5 × 5 = 52 ways.
Also, 3rd place can be filled up in 5 ways .
Hence, first three places can be filled up in 5 2 × 5 = 53 ways.
Lastly 4th place can be filled up in 5 ways.
Hence, all the 4 places can be filled up in 5 × 53 = 54 ways.
Hence, number of numbers of 4 digits formed with the digit 1, 2, 3, 4, 5, when
repetitions of digits is allowed = 54 = 625
Example 8 In how many ways can a ten question multiple choice exam be answered if there are 4
choices a, b, c and d , to each question? If no two consecutive questions can be
answered the same. In how many ways can 10 questions can be answered ?
Solution In the first case, when there is no restriction, number of questions = 10
Number of choices = 4
1st question can be answered in 4 ways a or b or c or d. 2nd question can be
answered can be in 4 ways.
Hence, first 2 questions can be answered in 4 × 4 = 42 ways.
Also 3rd question can be answered in 4 ways, so first three questions can be
answered in 43 ways.
Finally, all ten questions can be answered in 410 ways.
In the second case, it is given that no two consecutive question can be answered the
same.
First question can be answered in 4 ways and second in 3 ways, third question also in
3 ways.
So, first 3 questions can be answered in
4 × 3 × 3 = 4 × 32 ways.
4th question can also be answered in 3 ways and so on.
Hence, total number of ways to answer all the ten questions is 4 × 39.
www.pdfworld.in
www.pdfworld.in
Combinatorics 377
Example 9 In how many ways can n things be given to p persons, when, each person can get any
number of things ? (n > p ).
Solution First thing can be given to p person in p ways.
Similarly, 2nd thing and 3rd thing can be given to p persons in p ways respectively.
So, all the n things can be given to p persons in pn ways.
Concept I Any number of things out of the given things occur together.
For example, we have to find the number of arrangements of 9 persons including 4 boys, when it is given
that 4 boys will sit together.
Concept II When any number of things (or no two of a certain given things) comes together, to find
the number of permutations, when any number of a kind of things do not come altogether. We find the
total number of permutations without any restriction ...(i)
Then, we find the number of permutations when all the things of that kind are kept together ...(ii)
Subtracting (ii) from (i), we get the number of permutations, such that all the given number of things of a
kind do not come together.
For example, there are 9 students of which 5 are boys and 4 are girls and we have to find the number of
arrangements of all the 9 students in such a way that all the 4 girls do not come together (2 may come
together, 3 may come together but not all the 4 girls come together).
Then, we first find the total number of arrangements without any restriction for number of students
(things) to be arranged = n = 9
Hence, number of arrangements of all the 9 students without restriction is
9!
= 9P3 = = 9!
0!
Then, keeping all the 4 girls together.
i < B1 , B2 , B3 , B4 , B5 , (C1 , C2 , C3 , C4 , )and counting then as one (i) we see that number of students = 6 (5 + 1)
and number of places to be filled up = r = 6.
( 5 places for boys and 1 place for girls)
Hence, number of arrangements of these 6 is 6
P6 = 6 ! . But all the 4 girls among themselves can be
arranged in P4 = 4! ways.
4
Hence, number of arrangements of 9 students when all the 4 girls are together is 9 ! − 6 ! × 4 !
Concept III GAP METHOD

But to find the number of permutations of some number of things such that no two of the things of a
kind come together.
We put all the things on which there is no restriction in a line. Then, we count the places between every
pair of things including the places to the left of the Ist thing and to the right of the 2nd thing if l is the
number of things on which there is no restriction and m is the number of things such that no two of m
things come together, then we place things in a line in l Pl = l ! ways.
For m things there are (l + 1) places and hence required number of arrangements
+1
=l Pm l !
For example, there are 9 students of which 5 are boys and 4 are girls.
We have to find the number of arrangements of all the 9 students that no 2 girls sit together.
Then, we first place 5 boys in a line in 5 P5 = 5 ! ways for 4 girls such that no 2 girls sit together there 6
places.
www.pdfworld.in
www.pdfworld.in
6
Hence, we made the arrangements of 4 girls in 6 places in P4 ways. So that required number of
6
arrangements is 5 ! P4 .
Concept IV Number of permutations of n things taken all at a time out of which p things all alike (of
one kind), q things all alike (of 2nd kind and rest are of different is
n
Pn n!
=
p !q ! p !q !
As for example, we have to find the number of numbers of 8 digits formed with the digits 1, 2, 3, 3, 4, 5,
5, 5.
Number of things = n = 8.
Out of which two are alike (3, 3) and three are alike (5, 5, 5).
Number of places to be fixed up = 8.
Number of numbers of 8 digits formed with the digits 1, 2, 3, 4, 5, 5, 5 is
8!
=
2!3!
Example 1 There are 2 books each of 3 volumes and 2 books each of 2 volumes. In how many
ways can the 10 books be arranged on a long table so that volume of the volume of
same books are not separated?
Solution Let A and B two books having volumes A1, A2, A3 and B1, B2, B3 respectively.
Let C and D be the other two books C1, C2 and D1, D2 be their volumes.
There are 4 books. These 4 books, on 4 places can be arranged in 4P4 ways, i.e., 4!
ways.
Also three volumes ( A1, A2, A3 ) can be arranged among themselves in 3P3, i.e., in 3!
ways.
Three volumes B1, B2, B3 of the books B can be arranged among themselves in 3!
ways.
Again, (C1, C2 ) two volumes of the book C can be arranged in 2! ways and (D1, D2 )
two volumes of the book D can be arranged in 2! ways.
Total number of arrangements of book = 4! 3! 3! 2! 2! = 3456
Example 2 In a dinner party there, are 10 Indians, 5 Americans and 5 English men. In how many
ways can they be arranged in row so that all persons of same rationality sit together ?
Solution Let I1, I 2,..., I10 be 10 Indians.
A1, A2, ..., A5 be 5 Americans.
E 1, E 2,..., E 5 be 5 English men.
Now, there are 20 persons.
For the arrangement of these 3 sections.
No. of places to be filled up = r = 3.
Number of sections (thing) = n = 3.
Arrangement of these 20 persons is 3P3 = 3!
Arrangement of 10 Indians among themselves = 10P10 =10!
Arrangement of 5 American = 5P5 = 5!
Arrangement of 5 English men = 5!
Total number of arrangements = 3!10! 5! 5!
www.pdfworld.in
www.pdfworld.in
Combinatorics 379
Example 3 There are 6 balls of different colours (black, white, red, green, violet, yellow). In how
many ways can the 6 balls be arranged in a row so that black and white balls may
never come together?
Solution There is no restriction on red, green, violet and yellow balls. We first place the four
balls in a line for the arrangements of there 4 balls.
Number of place to be filled up = r = 4
3R 3G 3V 3 Y
Number of balls n = 4
Hence, four balls red, green, violet and yellow can be arranged in a line in
4!
4
P4= = 4! ways.
0!
Now, for black and white balls, these are five (5) places and hence two balls (black
5! 1 ⋅ 2 ⋅ 3 ⋅ 4 ⋅ 5
and white) can be arranged in 5P2 = = = 20
3! 1⋅ 2 ⋅ 3
Required number of arrangements of the 6 balls such that white and black balls do
not come together
= 4! × 20 = 24 × 20 = 480
Example 4 In a class of students, there are 4 girls and 6 boys. In how many ways can they sit in a
row that no 2 girls sit together?
6!
Solution 6 boys can be arranged in a line in 6P6 = = 6! ways.
0!
üB1 üB2 üB3, üB4 üB5 üB6
Also, number of places for 4 girls when no 2 girls sit together.
= 7(marked with ü)
Number of arrangements of the 4 girls is
7!
= 7P4 =
3!
7! 6! 7!
Required number of arrangements in this case = 6! × =
3! 3!
Example 5 3 women and 5 men are to sit in a row. Find, in how many ways they can be arranged
so that no 2 women sit next to each other?
Solution There is no restriction on men, so we first place 5 men in a line.
Here, n = 5 (number of men)
Number of place to be filled up = r = 5.
5!
5 men can be arranged in 5P5 = = 5! ways
0!
Also for the 3 women, there are 6 places and 3 women can take their sits in 6P3
6!
= ways.
3!
6!
Required number of arrangements = 5! × = 14400
3!
www.pdfworld.in
www.pdfworld.in
Example 6 In how many ways 16 rupees and 12 paise coins be arranged in a line, so that no
2 paise coins may occupy consecutive positions?
Solution We first arrange 16 rupees coin in a line.
For n =16 and r =16
All the 16 coin are identical.
16
P16 16!
Number of arrangements of 16 coins = =
16! 16!
Also since no two paise coins come together i.e., no two paise coin occupy
consecutive places so that number of places for 12 coins are = 17
17
P12
Hence, 12 paise coins can be arranged in ways.
12!
(Since all the 12 paise coins are also identical)
16! 17P12 17!
Required number of arrangements = × =1 ×
16! 12! 5!12!
12! ⋅13 ⋅ 14 ⋅ 15 ⋅ 16 ⋅ 17
= = 6188
12! ⋅ 2 ⋅ 3 ⋅ 4 ⋅ 5
Example 7 Show that the number of ways in which n books may be arranged on a shelf so that 2
particular books shall not be together is (n − 2)(n − 1)! .
Solution Total number of arrangements of the n books = n Pn = n !
Keeping 2 particular books together, number of book is (n −1)
These (n −1) books can be arranged in n −1Pn −1 = (n − 1)! ways.
But two particular books which are kept together can be arranged among themselves
in 2P2 = 2! ways.
Number of arrangements of n books when 2 particular books are together is
(n − 1)! 2! = 2 (n − 1)!
Required number of arrangements of n books when 2 particular books are not kept
together
= n ! − 2(n − 1)!
= n(n − 1)! − 2 (n − 1)!
= (n − 2)(n − 1)!
Example 8 Find the number of ways of arranging the letters a, a, a, a, b, b, b, c, c, c, d, e, c, f in a

row, if the letter c are separated from one another.
Solution Total number of letters is = 15 ( 5 a' s + 3 b' s + 3 c' s + 1d + 2 e' s + 1f ) Number of
arrangements of 12 letters excluding 3 c' s on which, there is no restriction is
12
P12 12!
=
5! 3! 2! 5! 3! 2!
Now since no 2 c’s come together
so that number of places for c is 13.
13
P3 1 13! 13!
Hence, arrangements of 3 c’s = = × =
3! 3! 10! 3! 10!
www.pdfworld.in
www.pdfworld.in
Combinatorics 381
Required number of arrangements of these 15 letters, when no 2 c’s are together

13! 12!
= ×
3!10! 3! 10!
Example 9 In how many ways can the letters of the word ‘SALOON’ be arranged, if the consonant
and vowels must occupy alternate place?
Solution Number of places, r = 6 .
Case I When vowels comes at first place. In this case 3 vowels (A, O, O) occupy 3
places 1st, 3rd and 5th.
3
P3 3!
Number of arrangements of vowels = =
2! 2!
Since consonants also come alternately so that number of places provided for the 3
consonants (S , L, N ) = r = 3
Number of arrangements of consonants = 3P3 = 3!
3! 6.6
Required number of arrangements = × 3! = = 18
2! 2
Case II When consonant comes at 1st place. If a consonant comes at the first place,
then (R , L, N ) is r = 3 (1st, 3rd, 5th)
Number of arrangements of consonants alternately is 3P3 (n = 3, r = 3).
3!
= = 3!
0!
Since vowels also come alternately, so that number of places for the (n = 3) vowels
( A, 0, 0), 2 alike ( 0,0) is r = 3.
3
P3 3!
Number of arrangements of vowels on the alternate place is = .
2! 2!
3!
Required number of arrangement = 3! × = 18
2!
Required number of ways of arrangement of the letters of the word ‘SALOON’ so that
consonants and vowels must occupy alternate places
= 18 + 18 = 36
Example 10 How many different words can be formed with the letters of the word
'MATHEMATICS' ?
Solution In the word 'MATHEMATICS'
Number of letters = n = 11
2 A’s are identical.
2 T ’s are identical.
2 M’s are identical.
Number of places to be filled up = r = 11.
Number of different arrangements or different words that can be formed with the
11
P11 11!
letters of the word 'MATHEMATICS' is =
2! 2! 2! 2! 2! 2!
www.pdfworld.in
www.pdfworld.in
Example 11 How many different signals can be made by hoisting 6 differently coloured flags one
above the other, when any number of them may be hoisted at once ?
Solution There are 6 differently coloured flags and any number of them may be taken at a
time.
Hence, number of things, n = 6
But number of places to be filled up vary from r =1 to 6.i.e., r =1 or 2 or 3 or 4 or 5 or
6.
Hence, required number of signals (arrangements)
= 6P1 + 6P2 + 6P3 + 6P4 + 6P5 + 6P6
6! 6! 6! 6! 6! 6!
= + + + + +
5! 4! 3! 2! 1! 0!
= 6 + 30 + 120 + 360 + 720 + 720 =1956
Example 12 (i) In how many ways 7 men can sit round a table ?
(ii) In how many ways we can place 7 apples in a circle ?
Solution (i) Let M1, M 2, M 3,..., M 7 be 7 persons keeping one person M1
fixed, remaining 6 persons can take their places in 6P6 = 6!
ways.
M
Here n = 6 , r = 6 .
(Also, there is a difference in clockwise and anticlockwise arrangements).
Hence, number of ways in which 7 men can sit round a table in 6! ways.
(ii) Keeping one apple fixed (n = 6) 6 apples can be arranged in 6P6 ways
= 6! = 720 ways.
But there is no difference in clockwise and anticlockwise arrangements so that
required number of arrangements
6!
= = 360
2
Example 13 4 gentleman and 4 ladies are invited to a certain party, find the number of ways of
seating them around a table so that only ladies are seated on the 2 sides of each
gentleman.
Solution Let G1,..., G 4be 4 gentleman
and L1,..., L4 be 4 ladies
Keeping one gentleman G1 fixed, remaining 3 gentleman G 2, G 3, G 4 can take their
seats in 3P3 = 3! ways
Then number of seats for ladies (one lady between gentleman or only ladies are
seated on either side of each gentleman) is 4.
Hence, 4 ladies can take their seats in 4P4 = 4! ways
Required number of arrangements = 3! × 4! =144
www.pdfworld.in
www.pdfworld.in
Combinatorics 383
Example 14 5 boys and 5 girls form a line with the boys and girls alternating. In how many different
ways could they form a circle such that the boys and girls are alternate ?
Solution Keeping any one say boy fixed, then remaining 4 boys can take their seats in 4!
ways.
Now, boys and girls are alternating number of places for 5 girls is 5.
Hence, 5 ways can be arranged in 5P5 ways
= 5! ways.
Required number of arrangements of 5 boys and 5 girls in a circle so that boys and
girls are alternating
= 4! 5! = 2880
Example 15 A gentleman invites a party of m + n friends to a dinner and places m at one round
table and n at another. Find the number of arrangements.
Solution We first divide m + n friends into 2 parts, one of part consist m and other n in
m +n (m + n )!
Cm . nCn = way
m!n !
Number of ways in which 2 groups (of m and n respectively) can go at 2 round tables
in 2! = 2 ways.
For the number of arrangements of m friends at a round table keeping one fixed and
there remaining (m −1) friends can take this seats in (m −1)! .
Other n persons can take their seats at other round table in (n −1) ! ways.
Required number of ways
(m + n )!
=2 . (m − 1)!(n − 1)!
m! n !
(m + n )!
=2
mn
Problem Based on the Use of

n!
(a) n Cr = (b) n − x Cr − x = n − x Cr
r ! (n − r ) !
(i) Number of different selections of n things taken r at a time such that any thing comes once in a
n!
combination is given by n Cr =
r ! (n − r )!
(ii) Number of different selection of n things taken r at a time such that x particular thing always occur,
is given by n − x Cr .
Setting x particular things aside, number of remaining things = n − x .
Now to get all the combinations in which x particular things always occur, we have to select r − x
things in every possible way from n − x things.
Hence, number of different combinations of n things taken r at a time such that x particular things
are included
−x
=n Cr −x
(iii) Number of different combinations of n things taken r at a time, such that x particular things always
occur (included) is given by n − x Cr − x .
(iv) Now, setting x particular things aside, number of remaining things = n − x .
www.pdfworld.in
www.pdfworld.in
Now to get all the combinations in which x particular things never occur, we have to select r things
from the remaining n − x things.
Number of combination of n things taken r at a time in which x particular things are excluded is
n −x
Cr .
Example 1 A man has 7 friends. He invites 3 of them to party. Find, how many parties to each of 3
different friends he can give and how many times any particular friend will attend the
parties?
Solution Number of friends n = 7
Number of places to be filled up in a party, r = 3
Number of different parties to each of 3 different friends can be given, is 7C3
7!
= = 35
3! 4!
If a particular friend always attend the parties, then keeping one particular friend
aside, number of remaining friends is n − x = 7 − 1 = 6.
We select only ( 3 – 1 = 2) friends from the 6 friends, since any particular friend always
attend the parties and hence
6! 6.5 × 4!
6
C2 = = = 15
2! 4! 2! 4!
Example 2 A delegation of 6 members is to be sent abroad out of 12 members. In how many ways
can the selection be made so that a particular member is included ?
Solution Number of members, n =12
Number of places to be filled up for a delegation = r = 6
If a member (particular) is included, then keeping the particular member aside,
number of remaining member = n − x = 12 − 1 = 11 and number of places to be filled up
= r = 6 −1 = 5
Number of ways to send a delegation of 6 members out of 12 members if a
particular number is included = 11C5
11!
= = 22 × 21 = 462
6! 5!
Example 3 At an election 3 wards of a town are canvassed by 4, 5 and 8 men respectively. If there
are 20 volunteer. In how many ways can they be alloted to different wards ?
20
Solution We have to select 4 men out of 20 for a ward which we can select in C4 ways.
Now, only 20 − 4 = 16 men remain.
Now, we have to select 5 men from 16 men which we can select in 16C5 ways.
Now,16 − 5 = 11 men remain and we have to select 8 men from these 11 which we can
select in 11C8 ways.
Number of ways in which 20 volunteers be alloted to 3 different wards
= 20C4 ⋅ 16C5 ⋅ 11C8
www.pdfworld.in
www.pdfworld.in
Combinatorics 385
Example 4 A candidate is required to answer 6 out of 10 questions which are divided into
2 groups each containing 5 questions and he is not permitted to attempt more than
4 from each group. In how many ways can he make up his choice?
Solution Let there be 2 groups A and B. Let group A contains 5 questions and group B contains
5 questions. Since a candidate is required to answer 6 questions and not more than 4
questions from any group.
Hence, a candidate can make up his choice in the following ways
(i) 4 questions from group A and 2 from group B.
(ii) 3 questions from group A and 3 from group B.
(iii) 2 questions from group A and 4 from group B.
Number of ways to answer the questions is
5
C4 ⋅ 5C2 + 5C3 ⋅ 5C3 + 5C2 ⋅ 5C4 = 5 × 10 + 10 × 10 + 10 × 5
= 50 + 100 + 50 = 200
Example 5 A person has 12 friends of whom 8 are relatives in how many ways can he invite 7
friends such that at least 5 of them may be relatives?
Solution Since out of 12 friends, 8 are relatives .
There are 12 − 8 = 4 which are only friend not relative.
He can invite 7 friends such that at least 5 of them may be relatives in following ways.
He can invite 5 relatives and 2 friends in 8C5 ⋅ 4C2 ways.
He can invite 6 relatives and 1 friends in 8C6 ⋅ 4C1 ways.
He can invite 7 relatives and no friends in 8C7 ⋅ 4C0 ways.
Total number of ways to invite 7 friends
= 8C5 ⋅ 4C2 + 8C6 ⋅ 4C1 + 8C7 ⋅ 4C0
8! 4! 8! 4! 8!
= ⋅ + ⋅ + ⋅1
5! 3! 2! 2! 6! 2! 1! 3! 7! 1!
= 56 × 6 + 28 × 4 + 8 = 336 + 112 + 8 = 456
Example 6 There are 3 sections in a paper each having 3 questions. A candidate has to solve any
5 questions choosing at least one question from each section. In how many ways can
he make up his choice?
Solution Let A,B,C be 3 sections each containing 5 questions.
He can answer any 5 questions selecting 1 from A, 1 from B and 3 from C in
5
C1 ⋅ 5C1 ⋅ 5C3 = 250 ways
He can answer any 5 questions selecting 1 from A, 2 from B, and 2 from C in
5
C1 ⋅ 5 C2 ⋅ 5 C2 = 500 ways
He can answer any 5 questions selecting 1 from A, 3 from B, 1 from C in
5
C1 ⋅ 5 C3 ⋅ 5 C1 = 250 ways
5
C2 ⋅ 5 C2 ⋅ 5 C1 = 500 ways
5
C3 ⋅ 5C1 ⋅ 5C1 = 250 ways
Total number of ways to answer any 5 questions
= 250 + 500 + 250 + 500 + 500 + 250 = 2250
www.pdfworld.in
www.pdfworld.in
Example 7 An 8 oared boat is to be moved by a crew chosen from 11 men of whom 3 can steer but
can't row and rest can't steer. In how many ways can the crew be arranged ?
Solution We can select one persons to steer the boat from the 3 who can steer in 3C1 ways.
We now require 2 particular men who can row on the bow side.
We now require 2 more persons on the bow side and these can be selected in, from
the remaining 6 men in 6C2ways.
The remaining 4 can be placed on the other side in 4C4ways.
Number of selections = 3C1 ⋅ 6 C2 ⋅ 4 C4.
We have 4 persons on the side A and 4 persons on the side B.
Now, number of arrangements of 4 persons in the side A = 4!
Number of arrangements of 4 persons in the side B = 4!
Required number of arrangements
= 3C1 ⋅ 6C2 = 4! 4! = 25920
Example 8 A table has 7 seats, 4 being on one side facing the window and 3 being on opposite
side. In how many ways can 7 people be seated on table, if 3 people X ,Y , Z must sit on
the side facing the window?
Solution Since the three people X ,Y , Z must sit on the side facing the window (4 seats) so that
we have to select 1 people for the side facing the window but of 7 − 3 = 4 in 4C1 ways.
Now, 3 people remain and we have to select all the remaining 3 people for other side in
3
C3 ways.
Number of selections = 4C1 ⋅ 3C3

Now, we have 4 persons (including X ,Y , Z ) on the side facing the window and 3
persons on other side.
Now, number of arrangements of 4 people in the side facing the window = 4! and
number of arrangements of 3 people in the other side is 3!.
Number of ways in which 7 people can be seated is
4
C1 ⋅ 3C3 ⋅ 4! 3! = 576
Example 9 How many words can be formed out of 10 consonants and 4 vowels such that each
contains 3 consonants and 2 vowels ?
Solution We have to select 3 consonants out of 10 in 10C3 ways.
2 vowels out of 4 in 4C2 ways.
In this we have 5 letters ( 3 + 2)
which can be arranged in 5! ways. Required number of words
= 10C3 × 4C2 5! = 720 × 120
= 86400
www.pdfworld.in
www.pdfworld.in
Combinatorics 387
Example 10 Four visitor A, B, C, D arrised at a town which has 5 hotels. In how many ways can they
disperse themselves among hotels ?
(i ) If 4 hotels are used to accommodate them
(ii ) If 3 hotels are used to accommodate them in such a way that A and B stay at
the same hotel.
Solution (i) Now, 4 hotels out of 5 can be selected in 5C4ways.
Now, 4 visitor can stay in one set of 4 hotels (i.e, one visitor in one hotel) is 4P4 = 4!
Hence, total number of way of accommodating the visitor is 5C4 4! = 120
(ii) When A and B stay at the same hotel, then we require 3 hotels, one for ( A, B ), one
for C and one for D. In this case, number of ways of accommodating the visitor is
5
C3 × 3! = 60
Example 11 10 persons amongst whom A, B, C are to speak at function. Find the number of ways in
which it can the done, if A wants to speak before B and B wants the speak before C.
Solution Here, places for A,B and C can be chosen in 10C3 ways. Now, 7 persons remain to
speak and these 7 persons can speak in 7! ways. Hence, number of way in which
they can speak is 10C3 × 7! .
Example 12 How many different word of 6 letter can be formed from the letters of the word
‘ALLAHABAD’?
Solution Words having 6 letters from the letters of the word ALLAHABAD can be made in the
following case :
(i) 2 alike, 4 distinct (ii) 2 alike, 2 alike, 2 distinct
(iii) 3 alike, 2 alike, 1 other (iv) 3 alike, 3 distinct
(v) 4 alike, 2 distinct (vi) 4 alike, 2 alike.
In the word ALLAHABAD, the letters are AAAA, LL,H,B,D.
(i) Now, two alike can be chosen from (AAAA) (LL) in 2C1 = 2 ways. After this one pair
and 3 (H, B, D) i.e., 4 different letters from these 4 letters we have to choose 4
distinct which is selected in 4C4 = 1.
Hence, number of selection of 2 alike and 4 distinct is 2 × 1 = 2
Now, number of word of these 6 letters consisting of 2 alike and 4 distinct is
6!
=2×
2!
= 2 × 360 = 720
(ii) Two alike and two alike can be chosen from (AAAA), (LL) in 2C2 = 1 way. After this
H,B,D remain, from these 3 we have to select 2 different which we can select in
3
C2 = 3 ways.
Hence, number of selection in this case = 1 × 3 = 3
Now, number of words of 6 letters consisting of 2 alike, 2 alike and 2 different
6!
=3 × = 3 × 180 = 540
2! ⋅ 2!
(iii) 3 alike i.e., 3A from 4 A’s can be chosen in 1C1 = 1 way, 2 alike i.e., 2 L’s from (LL)
can be chosen in 1C1 = 1 way. Now 1 distinct from the remaining 3 can be selected
in 3C1 = 3 ways.
Hence, number of selection in this case = 1 × 1 × 3 = 3
www.pdfworld.in
www.pdfworld.in
Now, number of words of these 6 letters consisting of 3 alike and 1 other is

6!
= 3× = 3 × 60 =180
3! 2!
(iv) Now, 3 alike from (AAAA) can be chosen in 1C1 = 1 way. Also 3 distinct from
4 (LI + BD ) can be chosen in 4C3 = 4 ways.
Hence, number of selections in this case
=1 × 4 = 4
Now, number of words of these 6 letters
6!
=4 × = 480
3!
(v) 4 alike can be chosen from (AAAA) in 1C1 = 1 way.
Now, 2 distinct can be chosen from the remaining 4 (L, H, B, D ) in 4C2 = 6 ways.
Hence, number of selection in this case
=1 × 6 = 6
Now, number of words of this 6
6!
=6 × = 6 × 30 =180
4!
(vi) 4 alike and 2 alike can be chosen from (A, A, A, A), (L,L) in 2C2 = 1 way.
6!
Now, number of words of these 6 = 1 × = 15
2! ⋅ 4!
Now, total number of words of 6 letters formed with the letters of the word
ALLAHABAD
= 720 + 540 + 180 + 480 + 180 + 15 = 2115
Example 13 How many quadrilateral can be formed by joining the vertices of a polygon of n sides ?
Solution Four sides are to be selected for a quadrilateral. Also, there are n vertices of a polygon
of n sides and any 4 points out of n can be selected in n C4 ways.
Hence, required number of quadrilateral = nC4.
Example 14 Find the number of diagonals that can be formed in a polygon of n sides.
Solution We know there are n vertices in a polygon of n sides, we also know joining any two
vertices of a polygon we have either a side or a diagonal of the polygon.
Now, number of ways in which any 2 points out of n points can be selected is n C2.
Hence, number of sides + number of diagonals = nC2.
But number of sides is equal to n and hence number of diagonals
n(n − 1)
= nC2 − n = −n
2
n(n − 3)
=
2
Example 15 If m parallel lines are intersected by n other parallel lines, find the number of
parallelogram then formed.
Solution Now, forming a parallelogram, is actually selecting of 2 lines from the set of m parallel
lines and 2 lines from the set of n parallel lines.
www.pdfworld.in
www.pdfworld.in
Combinatorics 389
But number of selections of 4 lines and 2 lines from the set of m parallel lines and 2
lines from the set of n parallel lines is
= mC 2 ⋅ n C 2
Hence, number of parallelogram thus, formed is m C2 ⋅ nC2 = mn(m − 1)(n − 1)
Example 16 There are 15 points in a plane, no two of them are collinear with the exception of 6,
which are all the same straight line. Find the number of
(i) straight lines formed (ii) number of triangles
Solution Supposing number three points of the 15 points are collinear.
Now, number of straight lines formed by the 15 points = 15C2.
Also number of straight lines formed by the 6 points is = 6C2.
But as 6 points are collinear and hence 6C2 straight lines reduce to a single straight
lines. Hence, number of straight lines formed by the 15 points such that 6 points are
collinear = 15C2 − 6C2 + 1
= 105 − 15 + 1
= 91
Also, number of triangles formed by 15 points = 15C3 if the 15 points be such that no
three of the 15 points are collinear. But it is given that 6 points are collinear and no
triangle is formed by taking any point out of these 6 points.
Hence, required number of triangles
= 15C3 − 6C3
15 6!
= −
3 !12! 3! 3!
15 ⋅14 ⋅13 ⋅12! 6 ⋅ 5 ⋅ 4
= −
3 ⋅ 2 ⋅12! 3 ⋅ 2 ⋅1
= 455 − 20
Concept Total number of combination or selections of different things taken some or all at a time
= n C1 + n C2 + n C3 + n Cn = 2n − 1
In the problem of this type, all the things are different and also things to be selected are not fixed in
number.
Example 1 Given 5 different green dyes, 4 different blue dyes and 3 different red dyes. How many
combination of dyes can be chosen taking at least one green and one blue dyes?
Solution Here, it is given that at least one green dyes must be selected. Hence one or more
or all the 5 green dyes can be selected in
5
C1 + 5C2 + 5C3 + 5C4 + 5C5 = 25 − 1 = 31 ways.
Similarly, one or more or all the 4 different blue dyes can be selected in
4
C1 + 4C2 + 4C3 + 4C4
= 24 − 1 = 15 ways
But it is not given that at least one red dye must be selected so that case of selection of
zero (0) dye is also to be included.
And number of ways of selection of 0, 1, 2, 3, red dyes is
= 3C0 + 3C1 + 3C2 + 3C3 = 23 = 8
www.pdfworld.in
www.pdfworld.in
Hence, total number of combination of taking at least one green and at least one blue
dye out of 5 green, 4 blue and 3 red dyes
= 31 × 15 × 8 = 3720
Example 2 A student is allowed to select at most n books from a collection of ( 2n + 1) books. If the
total number of ways in which he can select at least one book is 63. Find the value of n.
Solution Since the students is allowed to select at most n books out of ( 2n + 1) books so that
number of selections
2n + 1 2n + 1 2n + 1
= C1 + C2 + C3 + ...+ 2n − 1Cn = 63 ...(i)
2n + 1 2n + 1 2n + 1 2n + 1 2n + 1 2n + 1
Now, C0 + C1 + C2 …+ Cn + Cn + 1 + Cn + 2
2n + 1
+ ... C2n + 1 = 22n + 1
or 1 + ( 2n + 1C1 + 2n + 1
C2 + ... + 2n + 1
Cn ) + ( 2n + 1C2n + 1 − (n + 1) + 2n + 1
C2n + 1 − (n + 2)
2n + 1 2n + 1
+ C2n + 1 − (n + 2) + ... + C2n + 1 − 2n ) = 22n + 1
or 2 + 2 ( 2n + 1C1 + 2n + 1
C2 + ... + 2n + 1
Cn ) = 22n + 1
or 2 + 2 ⋅ 63 = 22n + 1
or 2 + 126 = 22n + 1
or 22n + 1 = 128
22n = 64 = 26 or 2n = 6
n=3
Example 3 We like to form a bouquet from 11 different flowers so that it should contain not less
than 3 flowers. Find the number of different ways of forming such a bouquet.
Solution Since the bouquet should not contain less than 3 flowers so that to form a bouquet we
are to select at least 3 flowers and at most 11 flowers out of 11 and hence, required
number of ways
11
C3 + C4 +
11 11
C5 + ... + 11
C11
= ( C0 +
11
C1 +
11 11
C2 ) + 11
C3 + 11
C4 + ... 11C11 − 11C0 − 11C1 − 11C2
= 211 − (11C0 + C1 +
11 11
C2 )
=2 11
− (1 + 11 + 55) = 1981
Distribution of Things into Persons

or Groups, Sets, Lots, Packet, Parcels
Let total number of things be n. Now regarding distribution of these n things into groups or persons, we
discuss some cases and each case discusses a type of problems.
Case I Number of ways of dividing the given different things into these groups containing
respectively p, q, r things, is given by
n!
n
Cp ⋅ n − pCq ⋅ n −p − q Cr =
p! q! r!
Here, the things p, q, r are distinct and
p + q + r =n
www.pdfworld.in
www.pdfworld.in
Combinatorics 391
Case II If n = m + m + m = 3m . Then number of ways of dividing n = 3m things into three persons, each
containing m things or number of ways of dividing n = 3m things equally among three persons or
distributing n = 3m things into 3 equal sets with permutation of sets is given by
(3m )!
3m
Cm 2m Cm m Cm =
(m !)3
Case III If n = m + m + m = 3m , then number of ways of spliting or dividing n = 3m things into three
groups (equally) each containing m things is given by
3m
Cm ⋅ 2m
Cm ⋅ m Cm
=
3!
3m !
=
3 ! (m !)3
Case IV The number of ways of distributing n different things into r different groups or lots is = r n
Here, blank lots have been taken into account.
Case V The number of ways of distributing n different things into r different parcels no lots being
blank is
r n − r C1 (r − 1)n + r C2 (r − 2)n + ... + (−1)r − 1 r Cr −1
+ r −1
Case VI The number of ways of distributing n things all alike into r different groups is n Cr −1 lots
may be blank. No lots being blank and is
n −1
Cr −1 or coefficient of x n in (1 + x + x 2 + ... ∞ )r
Case VII Number of ways of arranging n different things into r different groups is
r (r + 1) (r + 2)... (r + n − 1) −1
or n !⋅ n Cr −1
r (r + 1) (r + 2)... (r + n − 1)
according groups may or may not be blank.
Example 1 In how many ways can 52 cards be distributed equally among four players ?
Solution If we distribute 52 cards equally among 4 players, then (13 × 4 = 52) each player will
receive 13 cards.
n = 13 + 13 + 13 + 13 = 52
Hence, number of ways of distribution
52 − 13 52 − 13 −13 52 − 13 − 13 − 13
= 52C13 ⋅ C13 ⋅ C13 ⋅ C13
= C13 ⋅
52 39
C13 ⋅ C13 ⋅ C13
26 13
52!
=
(13!)4
Example 2 In how many ways can a pack of 52 playing cards be divided in 4 groups, three of
them having 17 cards each and the fourth first one card ?
Solution If we divide 52 cards in 4 groups such that first three groups contain 17 card each and
fourth first contains one card i.e., 17 + 17 + 17 + 1 = 52
52
C17 ⋅ C17 ⋅ 18C17 ⋅ 1C1
35
Then, the number of ways of required distribution =
3!
1 52!
= ⋅
3! (17!)3
www.pdfworld.in
www.pdfworld.in
Example 3 In how many ways 12 different books can be distributed equally among 4 persons ?
Solution If we distribute 12 different books equally among 4 persons then each person will get
3 books. Hence, required number of distribution is
12!
= 12C3 ⋅ 9C3 ⋅ 6C3 ⋅ 3C3 =
( 3!)4
Example 4 In how many ways 12 things be divided equally among four groups?
Solution If we divide 12 different things equally among 4 groups then each groups contains 3
things. Hence required number of ways of distribution of 12 different things equally
among 4 groups is given by
C3 ⋅ 9C3 ⋅ 6C3 ⋅ 3C3
12
12!
= =
4! 4!( 3!)4
Example 5 In how many ways can 10 balls be divided between two boys, one receiving two and
other eight balls ?
Solution Let B1 and B2 be two boys. Then the number of ways of distributing 10 balls such that
one boy, say B1 receives two balls and other boy B2 receives eight balls or one boy B2
receives 2 balls and other boy B1 receives 8 balls is
= 10C2 ⋅ 10 − 2 C8 + 10C8 ⋅ 10 − 8C2
= 10C2 ⋅ 8C8 + 10C8 ⋅ 2C2
= 10C2 + C2 = 2 ⋅ 10C2 = 90
10
Example 6 In how many ways 10 mangoes can be distributed among 4 persons if any person can
get number of mangoes ?
Solution A person can get 0,1, 2, 3,... mangoes and there are four persons.
Here, r =4
Hence, number of distribution of 10 mangoes among 4 persons if any person can get
any number of mangoes [including zero mangoes] is
= coefficient of x 10 in (1 + x + x 2 + x 3 + ... )4
4
 1 
= coefficient of x 10 in   = coefficient of x 10 in (1 − x )−4
1 − x 
[Q (1 − x )−n = 1 + nC1x + n +1
C2x 2 + K + n+ 2
C3x 3 + K ]
= coefficient of x 10 in (1 + 4C1x + 5C2x 2 + ... )
= 13C10 = 13C13 − 10 13
C3
Example 7 In how many ways can r flags be displaced on n poles in a row, disregarding the
limitation on the number of flags on a pole ?
Solution Number of ways in which r flags be displaced on n poles (such that a pole may have
0,1, 2, 3 and so on flags)
= coefficient of x r in (1 + x + x 2 + ... )n
www.pdfworld.in
www.pdfworld.in
Combinatorics 393
n
 1 
= coefficient of x r in  
1 − x 
= coefficient of x r in (1 − x )−n
n +1
= 1 + nC1x + C2x 2 + ... = n − 1 + rCr = n + r − 1Cr
Example 8 (i ) In how many ways can five different books be distributed among three student if
each student is to have at least one book?
(ii ) In how many ways can five different books be tied up in three bundles?
Solution (i) We know that number of ways in which n different things can be distributed into r
different parcels, no lots being blank is
r n − rC1(r − 1)n + rC2 (r − 1)n − ...
Here, n = 5 and r = 3, since 5 different books are to be distributed among 3
students. Hence, number of ways of distributing 5 books among 3 students so
that each student receives at least one book is
= 35 − 3C1( 3 − 1)5 + 3C2( 3 − 2)5
= 35 − 3 × 25 + 3 ⋅ 1 = 150
(ii) Since books are to be tied up in a bundle so that books are to be kept in a group
or set and hence required number of ways
1 5 3
= [ 3 − C1 ( 3 − 1)5 + 3C2( 3 − 2)5 ]
3!
1 5
= ( 3 − 3 × 25 + 3 ⋅ 1)
6
150
= = 25
6
Example 9 In how many ways can 15 identical Mathematics books be distributed among six
students ?
Solution We know that number of ways in which n things all alike can be distributed into r
different parcels is n + r − 1Cr − 1 when lots may be blank.
Hence, number of ways in which 15 identical books be distributed among six
students is
20!
= 15 + 6 − 1C6 − 1 = 20C5 =
5!15!
20 × 19 × 18 × 17 × 16 × 15!
=
5 × 4 × 3 × 2 × 1 × 15!
=15504
Concepts
(i) If p1 + p 2 + p3 things are such that p1 things are alike, p 2 things are alike and p3 things all different,
then number of selections of any r things out of p1 + p 2 + p3 things
p1
= coefficient of x r in (x 0 + x + x 2 + ... + x )
p2 p3
× (x 0 + x + x 2 + ... + x ) (x 0 + x 1 ) = coefficient of x r in
p1 p2 p3
(1 + x + x + x 2 + ... + x ) × (1 + x + x 2 + ... + x ) (1 + x )
www.pdfworld.in
www.pdfworld.in
(ii) If p1 + p 2 + p3 things are such that p1 things are alike of 1st kind, p 2 things alike of 2nd kind and p3
things alike of 3rd kind, then we have
(a) Number of selection of any r things out of p1 + p 2 + p3 things containing at least one thing from
p1 alike things
p
= coefficient of x r in (x + x 2 + x 3 + ...+ x 1 )
p2 p3
(x 0 + x + x 2 + ... + x ) × (x 0 + x + x 2 + ... + x )
p1 p2
= coefficient of x in (x + x + ... + x
r 2
) (1 + x + x 2 + ... + x )
p3
(1 + x + x + ... + x
2
)
(b) Number of selection of any r things containing at least one thing from p1 alike and one things
from p 2 alike things is given by
p1
= coefficient of x r in (x + x 2 + ...+ x )
p2 p3
(x + x + ... + x
2
) (x + x + x + ...+ x
0 2
)
p1
= coefficient of x in (x + x + x + ... + x
r 2 3
)
p2 p3
(x + x + ...+ x
2
) (1 + x + x + ...+ x
2
)
(c) Number of selection of any r things containing at least one things from p1 alike and two things
from p 2 alike things is given by
p1
= coefficient of x r in (x + x 2 + ... + x )
p2 p3
(x + x + ...+ x
2 3
) (1 + x + x + ... + x 2
)
and so on.
Example 1 Prove that the number of ways in which we can select four letters out of the word
EXAMINATION is 136.
Solution Letters of the word EXAMINATION are A,A,I,I,N,N,E,X,M,T,O. i.e., 2 (A's) are alike, 2
(I 's) are alike and 2 (N ′ s) are alike and other 5 are all different.
Now, number of selections of any four letters out of the letters of the word
EXAMINATION i.e., out of the letters ( A, A ),(I , I ),(N, N ), E , X , M ,T , O
= coefficient of x 4 in ( x 0 + x 1 + x 2 ) ( x 0 + x 1 + x 2 )( x 0 + x 1 + x 2 ) (1 + x )5
= coefficient of x 4in (1 + x 1 + x 2 )3 (1 + x )5
3
1 − x 3 
= coefficient of x 4 in   (1 + x )5
 1− x 
=coefficient of x 4 in (1 − x 3 )3 (1 + x )5(1 − x )−3
= coefficient of x 4 in (1 − 3C1x 3 + 3C2x 6 − x 9 ) (1 + 5C1x + 5C2x 2
+ 5C3x 3 + 5C4x 4 + x 5 )(1 − x )−3
= coefficient of x 4 in (1 − 3x 3 + 3x 6 − x 9 )
(1 + 5x + 10x 2 + 10x 3 + 5 x 4 + x 5 )(1 − x )−3
= coefficient of x 4in (1 − 3x 3 + 5x + 10x 2 + 10x 3 + 5x 4
− 15 x 4 + ... )(1 − x )−3
= coefficient of x 4 (1 + 5x + 10x 2 + 7x 3 − 10x 4 ) (1 + 3C1x + ... )
= 6C4 + 5 5C3 + 10 4C2 + 7 3C1 − 10
= 15 + 50 + 60 + 21 − 10
= 146 − 10
= 136
www.pdfworld.in
www.pdfworld.in
Combinatorics 395
Example 2 A box contains two white, three black and four red balls. Find the number of ways in
which we can select three balls from the box, if at least one black ball is to be included
in the selection.
Solution Considering balls of the same colour alike we have 2 (white) alike balls of one kind, 3
(black) alike balls of 2nd kind and 4 (red) alike balls of 3rd kind. Hence, number of
selections of any 3 balls containing at least one black balls is
= coefficient of x 3 in ( x 0 + x + x 2 )( x + x 2 + x 3 )
white black
(x 0 + x + x 2 + x 3 + x 4 )
= coefficient of x 3 in (1 + x + x 2 ) x (1 + x + x 2 )
(1 + x + x 2 + x 3 + x 4 )
1 − x 5 
= coefficient of x 3 in x (1 + x + x 2 )2  
 1− x 
2
1 − x 3  1 − x 5 
= coefficient of x 2 in    
 1− x   1− x 
= coefficient of x 2 in (1 − 2x 3 + x 6 ) (1 − x 5 ) (1 − x )−3
= coefficient of x 2 in (1 − 2x 3 + x 6 − x 5 + 2x 8 − x 11) (1 − x )−3
= coefficient of x 2 in (1 − 2x 3 − x 5 + x 6 + ... ) (1 + 3C1x + 4C2x 2 + ... )
= 4C2 = 6
Example 3 In how many ways can an examiner assign 30 marks to 8 questions giving not less
than 2 marks to any question ?
Solution Since number of question = 8
No question has mark less than 2 and no question has marks greater than 16 i.e.,
minimum marks = 2 and maximum marks for a question is 16.
Hence, number of ways in which 30 marks can be assigned to 8 questions
= coefficient of x 30 in ( x 2 + x 3 + x 4 + ... + x 16 )8
= coefficient of x 30 in x 16 (1 + x + x 2 + ... + x 14 )8
8
1 − x 15 
= coefficient of x 14 in  
 1− x 
= coefficient of x 14 in (1 − x 15 )8 (1 − x )−8
= coefficient of x 14 in (1 − x )−8
=coefficient of x 14 in (1 + 8C1x + 9C2x 2 + ... )
= 21C14
21!
=
14! 7!
21⋅ 20 ⋅19 ⋅18 ⋅17 ⋅16 ⋅15
=
7⋅ 6⋅ 5⋅ 4⋅ 3⋅ 2
= 116280
www.pdfworld.in
www.pdfworld.in
Example 4 Show that the number of ways of selecting n things out of 3n things of which n are of
one kind and alike and n are of 2nd kind and rest are unike is (n + 2)2n − 1 .
Solution Here, n things are alike, out of which 0, 1, 2, 3, ...,n may be selected.
Also, n other things are alike out of which 0, 1, 2, 3, ......, n may be selected and rest n
things are all different.
Now , the number of selection of n things = coefficient of x n in
(x 0 + x + x 2 + xn ) (x 0 + x + x 2 + xn )
(1 + x ) (1 + x )...(1 + x )
n times
= coefficient of x n in (1 + x + x 2 + x 3 ...+ x n )2 (1 + x )n
2
1 − x n + 1 
= coefficient of x n in   (1 + x )n
 1− x 
= coefficient of x n in (1 − x )−2 (1 + x )n
= coefficient of x n in (1 − x )−2 [ 2 − (1 − x )]n
= coefficient of x n in (1 − x )−2 [ 2n − n 2n − 1
n (n − 1) n − 2
(1 − x ) + 2 + ... + ( −1)n (1 − x )n ]
2!
= coefficient of x n in 2n (1 − x )−2 − n ⋅ 2n − 1 (1 − x )−1
= 2n ⋅ n + 1Cn − n ⋅ 2n − 1 = 2n (n + 1) − n ⋅ 2n − 1
= [ 2 (n + 1) − n ] 2n − 1 = (n + 2) ⋅ 2n − 1
Example 5 Suppose that there are pile of red , blue and green balls and that each pile contains at
least eight balls .
(i ) In how many ways can eight ball be selected ?
(ii ) In how many ways can eight balls be selected if at least one ball of each colour
is to be selected ?
Solution There are at least 8 (red) alike balls, 8 (blue) alike balls and 8 (green) alike balls. Balls
of each colour may be unlimited in numbers.
To select 8 balls in all we select
0,1, 2,... balls from each colours of balls. Hence, number of selection of any eight
balls
( x 0 + x + x 2 + ... )
= coefficient of x 8 in
red
( x 0 + x + x 2 + ... ) ( x 0 + x + x 2... )
×
blue green
3
 1  −3
= coefficient of x 8 in   = (1 − x )
1 − x 
10!
= 10C8 =
8! 2!
= 5 × 9 = 45
www.pdfworld.in
www.pdfworld.in
Combinatorics 397
Example 6 Show that the number of different selection of 5 letters from 5 a's, 4 b's, 3c's, 2d's and
one e is 71.
Solution Here, 5 letters (a's) are alike, 4 letters (b's) are alike, 3 letters (c's) are alike and 2
letters (d's) are alike and one other letters is e.
Hence, number of different selections of 5 letters
= coefficient of x 5 in ( x 0 + x + x 2 + ... + x 5 ) ( x 0 + x + x 2 + ... + x 4 )
(x 0 + x + x 2 + x 3 ) (x 0 + x + x 2 ) (x 0 + x )
= coefficient of x in (1 + x + x 2 + K + x 5 ) (1 + x + x 2 + K + x 4 )
5
(1 + x + x 2 + x 3 ) (1 + x + x 2 ) (1 + x )
1− x6 1− x5 1− x4 1− x3
= coefficient of x 5 in ⋅ ⋅ ⋅ (1 + x )
1− x 1− x 1− x 1− x
= coefficient of x 5 in (1 − x 6 ) (1 − x 5 ) (1 − x 4 ) (1 − x 3 ) (1 + x ) (1 − x )−4
= coefficient of x 5 in (1 + x − x 3 − 2x 4 − 2x 5... ) (1 + 4C1x + 5C2x 2.... )
= 8C5 + 7C4 − 5C2 − 24C1 − 2
= 56 + 35 − 10 − 8 − 2
= 91 − 20 = 71
Example 7 If n objects are arranged in a row, then find the number of ways of selecting three of
them objects so that no two of them are next to each other.
Solution Let the three persons P1, P2, and P3 selected n1 P1 n 2 P2 n 3 P3 n 4
Let n1 and n 4 be the number of persons to the left of P1 and to the right of P3. Let n 2 and
n 3 be respectively the number of persons between P1, P2; P2, P3
The n1 + n 2 + n 3 + n 4 = n − 3 ( 3 are P1, P2, P3 ) where n1, n 4 ≥ 0 and n 2, n 3 ≥ 1
Hence, required number of selections
= coefficient of x n − 3 in ( x 0 + x + x 2 + ... )
( x 1 + x 2 + x 3 ) ( x + x 2 + x 3 + .. ) ( x 0 + x + x 3 + ... )
n −3
= coefficient of x in (1 + x + x 2 + x 3 + ... )2 ( x + x 2 + x 3 + ... )2
= coefficient of x n − 3 in x 2 (1 + x + x 2 + x 3 + ... )2 (1 + x + x 2 + ... )2
n −5
= coefficient of x in (1 + x + x 2 + x 3 + ... )4
4
 1 
= coefficient of x n − 5 in   = coefficient of x
x −5
in (1 − x )−4
1 − x 
= (1 + 4C1x + 5C2x 2 + ... ) = n − 2Cn − 5 = n − 2C3
Example 8 A box contains 2 white balls, 3 black balls and 4 red balls. In how many ways can three
balls be drawn from the box if at least one black ball is to be included in the draw?
(i ) If balls of the same colour are considered to be identical.
(ii ) All balls are considered to be different.
Solution (i) In this case, there are 2 (white) alike balls, 3 (black) alike balls and 4 (red) alike
balls, we have to select 3 balls including at least one black balls i.e., we may
select, 0 or 1or 2 balls from white balls, 1or 2 or 3 from 3 black balls and 0 or 1 or 2
or 3 or 4 from 4 red balls. Hence, number of selection of any 3 balls containing at
least one black balls
www.pdfworld.in
www.pdfworld.in
= coefficient of x 3 in ( x 0 + x + x 2 )
(x + x 2 + x 3 ) ( x 0 + x + x 2 + x 3 + x 4 )
= coefficient of x in (1 + x + x ) (1 + x + x 2 ) (1 + x + x 2 + x 3 + x 4 )
3 2
= coefficient of x 2 in (1 + x + x 2 )2 (1 + x + x 2 + x 3 + x 4 )
2
1 − x 3  1 − x 5 
= coefficient of x in 
2
  
1− x  1− x 
(1 − 3x 2 + 3x 6 − x 9 )(1 − x 5 )
(1 − x )3
= coefficient of x 2 in (1 − 3x 3 + ... ) (1 − x )−3
= 1 + 3C1x + 4C2x 2 + ...
= 4C2
=6
(ii) In this case, all the balls are different and total number of balls = 2 (white) + 3
(black) + 4 (red) = 9 Hence, number of selections of any 3 balls from 9 = 9C3 = 84
Also, number of selection of 3 balls containing no black balls = 6C3 = 20
Hence, number of selection of 3 balls containing at least one black balls = 84 − 20
= 64
Example 9 If there are three different kinds of mangoes for sale in a market. In how many ways
can you purchase 25 mangoes?
Solution Here, we have unlimited number of 3 kinds of mangoes and we can select 25
mangoes in all. Hence, we can select 0 or1or 2 or 3… mangoes from each kind of
mangoes.
Hence, number of selection of 25 mangoes
= coefficient of x 25 in ( x 0 + x + x 2 + x 3... )3
3
 1 
1 − x 
= 1 + 3C1x + 4C2x 2 + ...
27 ⋅ 26
= 27C25 = 27C2 = = 351
2
Example 10 If n ( ≥ 3) is a +ve integer, then find the number of positive integral solution of
x + y + z = n.
Solution The maximum value of x , y , z satisfying x + y + z = n is n − 2.
Hence, the number of positive integral solution of x + y + z = n is given by the
= coefficient of x n in ( x + x 2 + x 3 + ... + x n − 2 )3
= coefficient of x n in x 3 (1 + x + x 2 + ... + x n − 3 )3
3
1 − x n − 2 
= coefficient of x n − 3 in  
 1− x 
www.pdfworld.in
www.pdfworld.in
Combinatorics 399
= coefficient of x n − 3 in (1 − x n − 2 )3 (1 − x )−3
= coefficient of x n − 3 in (1 − x )−3
= 1 + 3C1x + 4C2x 2 + ...
= n − 1Cn − 3 = n − 1C2
Example 11 Find the number of integral solution of x1 + x 2 + x 3 = 0 with x 2 ≥ − 5.

Solution Here, we have to choose any three integers out of −5, − 4, − 3, − 2, − 1,0, 1, 2...such that
sum of the three chosen integral is 0. Hence, number of integral solutions of
x1 + x 2 + x 3 = 0 …(i)
is given by
= coefficient of x 0 in ( x −5 + x −4 + x −3 + ... )3
3
 1 1 1 
= coefficient of x 0 in  5 + 4 + 3 + K 
x x x 
3
1 + x + x 2 + x 3 + ...
= coefficient x 0 in  
 x5 
(1 + x + x 2 + ... )3
x 15
= coefficient of x 15 in (1 + x + x 2 + ... )3
3
 1 
1 − x 
= coefficient of x 15 in (1 − x )−3 = 1 + 3C1x + 4C2x + ...
17! 17 ⋅16 ⋅15!
= 17C15 = =
2! 15! 2 ⋅15!
= 17 × 8 = 136
Hence, number of integral solutions of x1 + x 2 + x 3 = 0 with x 2 ≥ − 5 is 136.
Example 12 Find the number of integral solutions of 2x + y + z = 20 with x , y , z ≥ 0.

Solution Here setting y = z = 0, we have 2x = 20 or x = 10.
Hence, maximum value of x = 10
i.e., x may be taken from 0, 1, 2, 3, … to 10
such that 2x + y + z = 20
For x = 0, y + z = 20. But integral solutions of y + z = 20
2
 1 
1 − x 
= coefficient of x 20 in (1 + 2C1x + 3C2x 2 + ...)
= 21C20 = 21
www.pdfworld.in
www.pdfworld.in
For x = 1, y + z = 18 and number of integral solutions of y + z = 18 is given by the

coefficient of x 18 in (1 + x + x 2 + ... )2(1 − x )−2
= 1 + 2C1x + ... = 19C18 = 19
For x = 2, y + z = 16 and number of integral solution of y + z = 16 is 17C16 = 17
For x =3
y + z = 14 and number of integral solutions of this equation is 15C14 = 15 and so on.
For x =9
y + z = 2 and number of integral solutions of this equation is 3C2 = 3.
For x = 10
y + z = 0 and number of integral solution of this equation is = 1
For y = 0, z = 0
In this way, the number of integral solution of 2x + y + z = 20 with x , y , z ≥ 0 is
= 21 + 19 + 17 + 15 + 13 + 11 + 9 + 7 + 5 + 3 + 1
11
= [ 2.21 + (11 − 1) ( −2)]
2
11
= ( 42 − 20)
2
11
= × 22
2
= 121
We know from the greatest coefficient in the binomial theorem that for n is even, n Cr
is greatest, if
n
r =
2
for n is odd, n Cr is greatest, if
n + 1 n −1
r = or
2 2
Example 13 Out of 15 balls, of which some are white and rest are black. How many should be white
so that the number of ways in which the balls can be arranged in a row may be the
greatest possible? It is given that the balls of the same colour are alike.
Solution Let number of white balls be n. Then number of black balls is15 − n. Now number of
arrangements of these 15 balls consisting of n white (alike) and15 − n black (alike)
balls
15!
= = 15Cn
15! − n n !
15 15
It is given that number of arrangement is greatest i.e., Cn is greatest. But Cn is
greatest, if
15 − 1 15 + 1
n= or
2 2
i.e., n = 7 or 8
Hence, number of white balls = 7 or 8
www.pdfworld.in
www.pdfworld.in
Combinatorics 401
Example 14 A person wishes to make up as many different parties as he can out of 20 friends. Each
party consists of the same number. How many should be invited at a time? In how
many of these parties would the same be found?
Solution Let the number of friends be invited be n, then number of parties = 20
Cn .
20
But it is given that Cn is maximum and hence
20
n= = 10.
2
Let a particular friend attends P parties. Then, number of combination of 20 taken 10 at
a time such that the particular friend must be included = P.
But number of combination of 20 taken 10 at a time such that the particular member is
included is
20 − 1
= C10 − 1 = 19C9
Hence, P= 19
C9
Combinatorial Identities and Binomial Coefficients

In the counting process, the counting of things in two different ways gives equal number of ways of
counting and this leads to an equality called identity. for example, if we choose r objects out of a
n 
collection of n distinct objects, this can be done in   ways. But to choose r objects is equivalent to
r 
n   n 
reject the remaining n − r objects. Hence, we have an identity   =  .
r  n − r 
Such identities are called combinatorial identities and the process which leads to the formulation of
combinatorial identities is called combinatorial argument.
n 
The number of the type   which we are using are called Binomial coefficients because they appear in
r 
the Binomial expansion (x + y )n ; where n is a positive integer. The proof of Binomial theorem by using
the principle of inductions is already known to the students. Below we present the combinatorial proof
of Binomial theorem.
Combinatorial Proof of Binomial Theorem

Consider the expansion of (x + y )n , where n is a positive integer.
(x + y )n = (x + y ) (x + y ) (x + y )... ( x + y )
This expansion consists of three steps.
Step 1. Select one term from each of n factors.
Step 2. Multiply the selections together.
Step 3. Sum the products.
For example
(x + y )3 = (x + y ) (x + y ) (x + y ) = xxx + xxy + xyx + xyy + yxx + yxy + yyx + yyy
= x 3 + x 2y + x 2y + xy 2 + x 2y + x y 2 + y 2x + y 3 = x 3 + 3x 2y + 3xy 2 + y 3
The term of the type x r y n − r arises by selecting x's from r factors and y's from n − r factors. But this can
n  n 
be done in   ways. Hence, the coefficient of x r y n − r is   .
r  r 
www.pdfworld.in
www.pdfworld.in
n  n  −1 n  −2 n 
Therefore, (x + y )n =   x 0y n +   x 1y n +   x 2y n + ... +   x n y 0,
 0 1  2  n 
It is the same things as
n  n  −1 n  −2 2 n 
(x + y )n =   x n y 0 +   x n y +   xn y + ... +   x 0 y n .
 
0 1  2  n 
This is Binomial theorem.
If we put x = y = 1 in the Binomial theorem,
we get
n  n  −1 n  −2 n 
(1 + 1)n =   1n 10 +   1n 11 +   1n 12 + ... +   10 1n
 
0  1  2 n 
n  n  n 
∴ 2n =   +   + ... +  
 0  1  n 
We now give combinatorial argument to prove this identity.
Consider the set S with n elements
S = {x 1, x 2 , ... , x n }
n 
In the right hand side of the identity, the general term   counts the number of r elements subset of S.
r 
Hence, the right hand side counts the number of subsets of the set S. Now if A is any subset of S, then
x 1 ∈ A or x 1 ∉A . Similarlyx 2 ∈ A or x 2 ∉A ,... and finally x n ∈ A or x n ∉A . Thus the subset A is formed in 2n
different ways. This is equivalent to saying that there are 2n different subsets of the set S.
n  n  n  n 
∴ 2n =   +   +   + .. .+  
 0  1   2  n 
n n − 1 n − 1
Example 1 Give combinatorial argument to prove that   =   + .
k   k  k − 1
Solution Suppose S is a set with n − 1elements. Choosing an element 'a' not belonging to S,
n
consider the setT = S ∪ {a} of n elements. Now   denotes the number of k element
k 
subsets of the set T. This is left hand side of the identity.
We partition the k element subsets of T in two classes :
(i) The k element subsets of T not containing element 'a'. Here we have to choose all
n − 1
k elements from the set S. These subset are   in number.
 k 
a
S1 S2
S Sn
T
(ii) The k element subsets of T containing element 'a'. Here element 'a' is already with
n − 1
us. Hence to form k element subsets are   in number.
k − 1
www.pdfworld.in
www.pdfworld.in
Combinatorics 403
n − 1 n − 1
Hence, the total number of k element subsets of T are   + .
 k  k − 1
 n   n − 1  n − 1
∴   = + 
 k   k   k − 1
 2n n
Example 2 Give combinatorial proof of   = 2   + n 2.
 2  2
Solution Let X = {x1, x 2, ..., xn } andY = {y1, y 2, ..., yn } has 2n elements. The two element
 2n
subsets of X ∪ Y are   in number. This is left hand side of the identity, we partition
 2
the 2 element subsets of X ∪ Y into 3 classes.
1. Both the elements are from the set X . This gives 2 elements subsets of X. These
n
are again   in number.
 2
2. Both the elements are from the set Y. This gives 2 elements subsets of Y. These
n
are again   in number.
 2
3. One element is from X and one element is from Y . In this case the two elements
n n
can be chosen in     = n. n = n 2 ways.
1  1 
So these subsets are n 2 in number. Hence, the total number of 2 elements
subsets of X ∪ Y are
n n n 2  2n  n
  +   = 2   + n ,   = 2  + n .
2
 2  2  2  2  2
Example 3 Give combinatorial argument to prove that

2 2 2 2
n n n n  2n
  +   +   + ... +   =  .
 0 1   2 n n
Solution Consider two disjoint sets X = {x1, x 2,..., xn } andY = {y1, y 2, ..., yn } of n elements
each.
Then, X ∪ Y = {x1, x 2,..., xn , y1, y 2 ,..., yn } is a set with 2n elements. The n element
 2n
subsets of X ∪ Y are   in number. This is right hand side of the identity. Now any n
n
element subset of X ∪ Y is formed by choosing k elements from X and n − k elements
n  n 
form Y : k = 0,1, 2, 3,... , n this can be done in     ways.
k  n − k 
n  n 
But we know   =  
k  n − k 
2
n  n  n n n
∴    =    =  ,
k  n − k  k  k  k 
k = 0,1, 2,..., n
www.pdfworld.in
www.pdfworld.in
Therefore, the total number of n element subsets of X ∪ Y in

2 2 2 2
n n n n
  +   +   + ....+  
 0 1   2 n
This is left hand side of the identity.
Example 4 Give combinatorial argument to prove that

n n
Σ k   = n ⋅ 2n − 1.
k =1 k 
n
Solution Consider a set of n person. There are   different ways of forming a committee of k
k 
person k = 1, 2, 3, ..., n
After the committee is formed, those k members decide to have a dinner party at each
n
member's house. Thus, there will be k   parties of k member's each. Putting
k 
k =1, 2, 3,..., n, the total number of parties will be
n  n  n  n  n n 
1   + 2   + 3   + ... + n   = Σ k  
1   2  3 n k = 1 k 
This is left hand side of the identity. Now consider the first person P1 and the number of
parties at his house. For each of remaining n −1 persons, there are two possibilities,
being or not being included in the party at P1' s house. So there are 2n − 1 parties at P1' s
house. Similarly there are 2n − 1 parties at P2' s house, ... pn ' s house, Hence, the total
number of parties is n ⋅ 2n − 1. This is the right hand side of the identity
n  n  n  n 
∴ 1   + 2   + 3   + ... + n   = n ⋅ 2n − 1
1   2  3 n
Example 5 Give combinatorial proof of the identity

n n 
Σ 2k   = 3n
k =0 k 
Solution Consider three letters A, B, C. We form different strings of length n formed by these 3
letters with repetition.
Since each of n position can be filled in 3 ways, there are 3n strings.
This is right hand side of the identity. Now, we divide the strings of length n into disjoint
classes depending upon how many times the letters A appears in the string.
n 
Suppose A appears k times. Then, letter A can choose k positions in   ways. The
k 
remaining n − k places are to be filled by using two letters B and C . This can be done
n 
in 2n −k different ways. Thus, there are 2n − k   strings of length n containing k A's. Put
k 
k = 0,1, 2,... n and add, the total number of strings is
www.pdfworld.in
www.pdfworld.in
Combinatorics 405
n  n  n  n 
2n   + 2n − 1  + 2n − 2   + ... + 20  
 0 1
   2 n
n   n  n −2  n  0 n 
= 2n   + 2n − 1 +2   + ... + 2  
n n − 1 n − 2  0
This is the left hand side of the identity
n n 
∴ Σ 2k   = 3n
n =1 k 

r  r + 1 r + 2 n n + 1
 + +  + ...   =  .
r   r   r  r  r + 1
Use this result to find the sum
1 + 2 + 3 + ... + n.
Solution We have
m  m  m + 1
 +  = 
r  r + 1  r + 1
m m + 1  m 
  =  −  
r  r + 1  r + 1
Put m = r , r + 1, r + 2 ..., n and add
r  r + 1 n r + 1  r  r + 3 r + 2
∴  +  + ... +   =   −  +  − 
r   r  r  r + 1 r + 1 r + 1  r + 1 
r + 2 r + 1 r + 4 r + 3 n + 1  n 
+  −  +  −  + .....+   − 
r + 1  r + 1 r + 1  r + 1  r + 1 r + 1
On the right side, the terms cancel diagonally and also
r  r + 1 r + 2 n n + 1
∴  + +  + ... +   =  
r   r   r  r  r + 1
Deduction: Put r = 1 in the above result
1  2  3 n n + 1
∴   +   +   + ... +   =  
1 1  1  1   2 
n (n + 1)
∴ 1 + 2 + 3 + ... + n =
2
Example 7 Give combinatorial argument and prove that

n n + 1 n + 2 n + r  n + r + 1
 + +  + ... +   = 
 0  1   2   r   r 
Solution Consider the equation x1 + x 2 + ... + xn + xn + 1 + xn + 2 = r
n + 2 + r − 1 n + r + 1
It has   =  solutions in non-negative integers. This gives right
 r   r 
hand side of the identity.
www.pdfworld.in
www.pdfworld.in
Now, depending upon the value of xn + 2, we divide these solution into various
disjoints classes as below.
Put xn + 2 = 0. Then, equation is
x1 + x 2 + ... + xn + 1 = r
n + 1 + r − 1 − 1  n + r − 1
The number of solution is   = 
 r −1   r −1 
Put xn + 2 = 2. Then, equation is
x1 + x 2 + ... + xn + 1 = r − 2
n + 1 + r − 2 − 1 n + r − 2
The number of solution is   = 
 r −2   r −2 
Put xn + 2 = r − 1. Then, equation is
x1 + x 2 + K + xn + 1 =1
The number of solution is
n + 1 + 1 − 1 n + 1
  = 
 1   1 
Finally put xn + 2 = r . Then, equation is x1 + x 2 + ... + xn + 1 = 0
The number of solution is
n + 1 + 0 − 1 n
  = 
 0   0
Addition of these gives the total number of solutions
 n + r  n + r − 1 n + r − 2 n + 1 n
  +  +  + ... +   + 
 r   r −1   r − 2   1   0
n n + 1 n + 2 n + r 
=  +  +  + ... +  
 0  1   2   r 
This is left hand side of the identity
n n + 1 n + 2 n + r  n + r + 1
∴   +  +  + ... +   = 
  
0 1   2   r   r 
The Pigeonhole Principle

Introduction The truth of the following statement is obvious :
If 4 pigeons fly into 3 pigeonholes, some pigeonhole must receive at least two pigeons.
This statement is a sample example of a very basic combinatorial principle, called the pigeonhole
principle. The simplest form of this principle is as follows:
Pigeonhole Principle (PP1) If more than n objects are distributed into n boxes, then at least one box
must receive at least two objects.
This follows by noting that, if each of the n boxes contained at most 1 object only, then the n boxes
together would contain at most n objects contrary to the assumption that more than n objects were put
in the boxes.
This principle was first stated formally by Dirichlet and is therefore also called Dirichlet’s box principle
or Dirichlet’s drawer principle. Let us apply it to some simple problems.
www.pdfworld.in
www.pdfworld.in
Combinatorics 407
Example 1 A box contains three pair of socks; coloured red, blue and white. Suppose I take out
the socks without looking at them. How many socks must I take out in order to be sure
that they will include a matching pair ?
Solution If I take only 2 or 3 socks, it is possible that they are all different.
For example, they may be one red and one blue or one red, one blue and one white.
But if I takes 4 socks, these must include a matching pair. Here the 4 chosen socks
are the “objects” and the 3 colour are the “boxes” and by PP1, it follows that at least
two of the 4 chosen socks must have the same colour and hence must for a matching
pair. Thus the minimum number of socks to be taken is 4.
Example 2 Show that in a group of 8 people, at least two will have their birthday on the same day
of the week.
Solution Here the 8 people are the “objects” and the 7 days namely,
Monday, Tuesday, ……., Sunday are the “boxes”. Hence, by PP1 at least two of the
people must belong to the same day. Similarly, it is clear that in a group of 13 people,
at least two will have their birthday in the same month.
Example 3 How many student do you need in a school to guarantee that there are at least 2
students, who have the same first two initials ?
Solution There are 26 × 26 = 676 different possible sets of two initials that can be obtained
using the 26 letters A, B, …, Z. So by PP1, the number of students should be greater
than 676.
Example 4 A professor tells 3 jokes in his ethics class each year. How large a set of jokes does the
professor need in order never to repeat the exact same triple of jokes over a period of
12 years ?
Solution Let n be the number of jokes needed. Now for a period of 12 years 12 triples are
n
needed. Also these are to be all distinct. These are   different triples that can be
 3
n
chosen from the n jokes. Hence n must be chosen, so that   ≥ 12
 3
n(n − 1)(n − 2)
or ≥ 12 or n ≥ 6
6
Hence by PP1, the smallest value of n is 6.
Example 5 In a round−robin tournament, show that there must be two players with the same
number of wins, if no player loses all matches.
Solution In a round-robin tournament of n players, each player plays every other player once
and each game results in a win for one of the players. Let ai , denote the number of
games won by player i. Then, by assumption, every player wins at least one game, so
that ai ≥ 1 for all i. Also, a player can win at most n − 1 games.
Hence, 1 ≤ ai ≤ n − 1 for i = 1, ..., n. Thus, the n ‘objects’ i.e., the number a1, ..., an are
put in the (n − 1) ‘boxes’
{1}, {2}, {3}, ..., {n − 1}
Hence, by PP1, two of the numbers, say ar and as , must be the same box i.e., they
must be equal.
www.pdfworld.in
www.pdfworld.in
Example 6 Show that a party of 20 people, there are two people who have the same number of
friends.
Solution Here it is assumed that, if A is a friend of B, then B is also a friend of A. Also, we will
show that the result is true not only for 20 people but for any number n of them (n ≥ 2).
Let P1, ..., Pn denote the n people and let ai be the number of friends of person Pi . Then
since a person has either no friends or at most n − 1friends, we have 0 ≤ ai ≤ (n − 1),for
1 ≤ i ≤ n.
First suppose that everyone at the party has at least one friend. Then, ai ≥ 1for all
i = 1, ..., n and so the n number a1, a 2 K an come from then (n − 1) numbers1, ..., n − 1.
Hence, by PP1, at least two of the a′s must be equal, say ar = as . That is the persons Pr
and Ps have the same number of friends.
Secondly, suppose that one of the persons, say Pn , has no friends. Then, each of the
remaining n − 1persons has at most n − 2, friends so that now the n numbers a1, ..., an
come from the n − 1 numbers 0,1, …, n − 2. Hence, as before, two of the n persons
must have the same number of friends.
Remark
Note that in applying the Pigeonhole Principle, the crucial step is always to decide, what are the ‘objects’
and what are the ‘boxes’. This choice is not always easy to make. Also, PP1 does not tell us how to make
this choice, but a clue is provided by the fact that the arithmetic of the numbers involved in the problem is
to be exploited .
Further note that PP1 claims only the existence of at least one box containing two or more objects; it
does not tell us, how to locate this box.
Second Form of the Pigeonhole Principle (PP2)

For any positive integers n, t , if tn + 1 or more objects are placed in n boxes, then at least one box will
contain more than t objects.
Proof If each of the n boxes contained at most t objects, then the n boxes together would contain at
most tn objects. This contradict the fact that, tn + 1 or more objects were placed in the boxes. Hence at
least one box must contain the more than t objects.
As an illustration of PP2, assume that no human has more than 300000 hairs on his/her head. A city has
a population over 300000. Here the residents are the ‘objects’ and there are n = 300000 ‘boxes’
corresponding to the different possible number of hairs. Thus, by PP1, at least 2 residents of the city
must be in the same box i.e., must have the same number of hair on their heads. But if the city has more
than 15 million residents, then at least 50 of them must have the same number of hairs on their heads.
This follows by PP2, since in this case the number of residents is more than 15000000 = 50 × 300000
and so some box must contain more than t = 50 objects.
Third Form of the Pigeonhole Principle (PP3)

If the average of n positive number is t, then at least one of the numbers is greater than or equal to t.
Further, at least one of the numbers is less than or equal to t.
Proof Let a1 , a 2,…, an be the numbers. Then by data,
a1 + ... + an
=t
n
So that a1 + ... + an = tn …(i)
Hence, if each of the n numbers a1 , ... , an is less than t, then the sum of these numbers would be less
than nt, contradicting (i).
A similar argument shows that at least one of the numbers is less than or equal to t.
www.pdfworld.in
www.pdfworld.in
Combinatorics 409
Remark
If the number a1, a 2, …,an are integers, then PP3 says that at least one of them is ≥ t 0, where t 0 is the
smallest integer not less than t ; and at least one is ≤ [t ], where [t ] is the integral part of t.
Fourth Form of the Pigeonhole Principle (PP4)

Let q1 , q 2 , ... , qn be positive integers. If (q1 + q 2 + ... + qn − n + 1) objects are put into n boxes, then either
the first box contains at least objects or the second box contains at least q 2 objects, … or the nth box
contains at least qn objects.
Proof Suppose we distribute
(q1 + q 2 + …+ qn − n + 1) objects in n boxes and if for each i = 1, 2, ... , n, the ith box contains less than qi
object, then the total number of objects in the n boxes is
≤ (q1 − 1) + (q 2 − 1) + ... + (qn − 1) = q1 + q 2 + ... + qn − n
But this number is less than the number of object placed in the n boxes. Hence, for at least one i, ith box
must contain at least qi objects.
Recurrence Relations
A recurrence relation is a way of defining a sequence inductively.
We may define the sequence a 0 , a1 , a 2 , … (or {an }) inductively in 2 steps as follows.
(i) For a fixed integer m ≥ 0, give the terms
a 0 , …, am explicitly and
(ii) States an equation that relates an , where n > m, to certain of its predecessors a 0 , a1 , ... , an − 1
The values of the terms a 0 , ... , am in (i) are called initial conditions for {an } and equation in (ii) is called
recurrence relation for {an }.
We give 2 examples for recurrence relations.
1. The sequence of Fibonacci Numbers Fn is defined by
(a) F0 = 1, F1 = 1 and
(b) Fn = Fn − 1 + Fn − 2 , for n ≥ 2
This sequence is 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, …
2. Derangements
(a) D1 = 0, D2 = 1
(b) Dn = (n − 1)(Dn − 1 + Dn − 2 ), for n ≥ 3
A counting problem can sometime be modeled with recurrence relation follows :
Suppose that the number of ways of performing an action using n object is an . Then it may be possible to
divide these an ways into mutually exclusive cases, such that these cases involve n − 1 or fewer objects.
This allows us to relate an with an − 1 ,an − 2 , … and the result is a recurrence relation satisfied by an . The
corresponding initial conditions have to be obtained by direct calculation. Then, solving the recurrence
relation we get an .
So, in the following we first illustrate by examples the process of obtaining a recurrence relation.
Example 1 If there are 48 different pairs of people, who know each other at a party of 20 people,
then show that some person has 4 or fewer acquaintances.
Solution Let ai denotes the number of acquaintances of person Pi . Now, if P1 and P2 know each
other, then the pair (P1, P2 ) contributes 1 to the count a1 and 1 to the count a 2. Thus,
each pair of acquaintances contributes 2 to the total count of acquaintances. But there
are 48 pairs of acquaintances.
www.pdfworld.in
www.pdfworld.in
Hence, the total count of acquaintances is 48 × 2 = 96 . Hence, the average number of

acquaintance per person is
a1 + ... + a 20 96
= = 4.8
20 20
Therefore, by PP3, at least one person has 4 or fewer acquaintances.
Example 2 A computer is used for 99 hours over a period of 12 days, an integral number of hours
each day. Show that on some pair of consecutive days, the computer was used at
least 17 hours.
Solution Let ai denote the number of hours the computer is used on the i th day. Consider the 6
pairs of consecutive numbers :
(a1, a 2 ), (a 3, a 4 ), ..., (a11, a12 )
The average of the corresponding 6 sums is
(a1 + a 2 ) + (a 3 + a 4 ) + K+ (a11 + a12 ) 99
= = 16.5
6 6
Hence, by PP3, at least one of the sum is ≥ 16.5 and hence, that sum is ≥ 17, since it
is an integer.
Example 3 Given 10 French books, 20 Spanish books, 8 German books, 15 Russian books and
25 Italian books, how many books must be chosen to guarantee that there are 12
books of the same language ?
Solution Note first that there are less than 12 books in French and in German and so the 12
required books in the same language must come from one of the other languages.
therefore, we must allow the possibility that the chosen books include all of the 10.
French and 8 German books. Thus by PP4, it is enough to chose
(10 + 11 + 8 + 11 + 11) + 1 = 52 books.
Example 4 Find a recurrence relation for the number an of pairs of rabbits after n months, if
(i) initially there is a new born pair (ii) a new born pair will produce their first pair of
offspring (a male and a female) after 2 months and (iii) every pair older than 2 months
will produce a pair every month. Assume that no deaths occur.
Solution Let A denote a new born pair B a one month old pair C a pair 2 or more months old.
Then, starting with A the pairs grow as follows.
(0) Initially A
(1) After one month B (2) After 2 month C, A
(3) After 3 month C, B, A (4) After 4 month C, C, B, A, A and so on
Thus, a 0 = 1,a1 = 1, a 2 = 2,a 3 = 3,a 4 = 5. To see how the number a 4arises, note that at
the end of the fourth month, the a 3 pairs in (3) are still there and they have become C,
C, B. Further the a 2 pairs in (2) have become C, C and they produce a pair each A, A.
Hence, we see that
a4 = a3 + a2
Similarly for n ≥ 2, we have the following situation
(n − 2) After n − 2 months an − 2 pairs
(n − 1) After n − 1 months an − 1 pairs
(n ) After n months an pairs
www.pdfworld.in
www.pdfworld.in
Combinatorics 411
The number an arises thus : in (n), there are an − 1 pairs, which were present in (n − 1)
and each of these an − 2 pairs in (n − 2) produces a pair (since each of them has
become a C pair.) Hence, we obtain
an = an − 1 + an − 2
with initial conditions a 0 = 1, a1 = 1
Example 5 A man has a staircase of n stairs to climb. Each step he takes, can cover either 1 stair
or 2 stairs. Find a recurrence relation for an , the number of different ways for the man to
ascend the n stair staircase.
Solution Let the steps covering first stair and second stairs be denoted respectively by a and
b. Then, ascending an n staircase corresponds to a sequence of a’s and b’s .
Clearly if n = 1, then there is only one stair and so the only possibility is a.
Hence, a1 = 1. If n = 2, then there are 2 stairs and so the only possibilities are aa or b.
Hence, a 2 = 2, if n = 3, then there are 3 stairs and so the only possibilities are aaa or ab
or ba. Hence, a 3 = 3.
Now, suppose that there are n ≥ 4 stairs. Then. there are exactly 2 mutually exclusive
possibilities for the first step : a or b. If the first step is a, then there remain n − 1stairs
to be covered and this can be done in an − 1 ways. If the first step is b, then there remain
n − 2 stairs to be covered and this can be done in an − 2 ways.
Hence, by AP, we get the required recurrence relations as.
an = an − 1 + an − 2
with initial conditions a1 = 1, a 2 = 2.
Example 6 Find a recurrence relation for the number an of binary sequences of length n that do
not contain the pattern 11.
Solution Consider the setSn of all binary sequences of length n that do not contain the pattern 11.
Clearly S1 = {0, 1}, so that a1 = 2
S 2 = {00, 01, 10}, so that a 2 = 3
Now, let n ≥ 3. The every sequence in the set Sn either starts with a zero or with a1. All
such sequence starting with a zero are obtained by appending a zero at the beginning
of every binary sequence of length n − 1 and not containing 11. Hence, there are an − 1
such sequences, Next if a sequence x in Sn starts with a1, then second digit must be
zero. Hence, all the sequences starting with a 1 are obtained by appending the
pattern 01 at the beginning of every binary sequence of length n − 2 and not
containing 11. Hence, there are an − 2 such sequences. Required recurrence relation is
an = an − 1 + an − 2 with initial conditions a1 = 2, a 2 = 3.
Example 7 Find a recurrence relation for the number an of ternary sequences of length n that
contain 2 consecutive digits that are the same. What are the initial conditions ? find a 6.
Solution Clearly, no ternary sequence of length 1 can contain 2 consecutive identical digits
and so a1 = 0. Next the only ternary 2-sequences of the required type are 00, 11, 22
and so a 2 = 3.
Let n ≥ 3. Every n sequence of the required form satisfies exactly one of the following
conditions :
(i) Its first 2 digits are unequal.
(ii) Its first 2 digits are identical.
www.pdfworld.in
www.pdfworld.in
Let (i) hold. Then, the sequence starts with one of 01, 02, 12, 20, 21.
First suppose that it starts with 01. Now the condition that the sequence contains “2
consecutive identical digits” is symmetric w.r.t. all 3 digits 0, 1, 2. Hence, there are an
1
equal number, namely, m = an − 1, of sequences of length n − 1 and starting with 0,1
3
or 2. So by appending 0 as first digit to each (n − 1) sequence starting with 1, we get m
sequences of length n which start with 01. Similarly, there are m sequences of length n
starting with 02, 10,12, 20 or 21. Thus there are 6m = 2an − 1 sequences in this case,
Let (ii) hold, then the sequences starts with 00 or 11 or 22 and its remaining n − 2 digits
can form any (n − 2) ternary sequence.
Hence, there are 3n − 2n sequences starting with 00; and the same holds for 11 and 22.
Thus, there are 3 × 3n − 2 = 3n − 1 sequences in this case.
Required recurrence relation is
an = 2an − 1 + 3n − 1
With initial conditions a1 = 0, a 2 = 3
Hence, a 3 = 15, a 4 = 57
a 5 = 195, a 6 = 633
Example 8 Find a recurrence relation for the number an of ways to distribute n distinct objects into
5 boxes.
Solution Hence, a box can hold any number of objects. Clearly the first object can be put in any
one of the 5 boxes.
So a1 = 5, Let n ≥ 2
Then, again the first object can be placed in 5 ways. Then, the remaining n − 1objects
can be placed in 5 boxes in an − 1 ways. Hence, the n objects can be placed in 5an − 1
ways. This gives the recurrence relation an = 5an − 1 with initial condition a1 = 5.
Example 9 A man has a staircase of n stairs to climb. Find a recurrence relation for the number an
of different ways for the man to ascend the n staircase if each step covers either 1 or 2
or 3 stairs.
Solution Let the steps covering 1 stair, 2 stairs and 3 stairs be denoted by respectively a, b, c.
Then, ascending an n stair staircase corresponds to a sequence of a′ s′ , b′ s, c′ s.
Clearly a1 = 1 and a 2 = 2. If n = 3. Then, the only possibilities are aaa or ab or ba or c.
Hence, a 3 = 4.
Now suppose that there are n ≥ 4 stairs. Then, there are exactly 3 mutually exclusive
possibilities for first step : a or b or c.
If first step is a, then there are an − 1 ways.
If first step is b, then there are an − 2 ways.
If first step is c, then there are an − 3 ways.
Hence, by AP, we get required recurrence relation as
an = an − 1 + an − 2 + an − 3
with initial conditions a1 = 1, a 2 = 2,a 3 = 4
Example 10 A child has ` n. Each day he buys either milk for ` 1 or orange juice for ` 2 or
pineapple juice for ` 2.If Sn denotes the number of ways of spending all the money.
Find the recurrence relation for this sequence. In how many ways can he spend ` 7?
www.pdfworld.in
www.pdfworld.in
Combinatorics 413
Solution If he buys milk on first day, then remaining ` (n − 1) can be spent in Sn − 1 ways. If he
buys orange juice on the first day, then remaining `(n − 2) can be spent in Sn − 2 ways
If he buys pineapple juice on the first day, then remaining `(n − 2) can be spent in
Sn − 2 ways.
The above cases are disjoint
∴ Sn = Sn − 1 + Sn − 2 + Sn − 2
∴ Sn = Sn − 1 + 2Sn − 2
This is recurrence relation.
In the other part, a child can spend ` 7 in S 7 ways.
S 7 = S 6 + 2S 5
= S 5 + 2S 4 + 2(S 4 + 2S 3 )
= S 5 + 4S 4 + 4S 3
= S 4 + 2S 3 + 4(S 3 + 2S 2 ) + 4(S 2 + 2S1)
= S 4 + 6S 3 + 12S 2 + 8S1
= S 3 + 2S 2 + 6(S 2 + 2S1) + ( 2S 2 + 8S1)
= S 2 + 2S1 + 2S 2 + 6S 2 + 12S1 + 12S 2 + 8S1
= 21S 2 + 22S1
But S1 = 1. Also ` 2 can be spent in 2 ways as 1,1 or 2
S 7 = 21( 2) + 22(1) = 64
www.pdfworld.in
www.pdfworld.in

Example 1. Find the number of ordered pairs (x , y ) of +ve integers such that x + y ≤ 4.
Solution We have to find the number of elements in the set
z = {(x , y ) | x , y + ve integers, x + y ≤ 4}
∴ x ≥ 1; y ≥ 1
We have 3 mutually exclusive cases
x + y =2
or x + y = 3 or x +y =4
for i = 2, 3, 4
Let zi = {(x , y ) | x , y + ve integers, x + y = i }
Then z2 = {( 1, 1)}
z3 = {( 1, 2), ( 2, 1)}
z4 = {( 1, 3), ( 2, 2), ( 3, 1)}
These sets are pairwise disjoint sets.
∴ By Addition principle, we get
|z| = 1 + 2 + 3 = 6
Example 2. Eight cards bearing number 1, 2, 3, 4, 5, 6, 7, 8 are well shuffled. Find in how many cases the
top 2 cards will form a pair of twin prime?
Solution Out of 8 integers 1,..., 8 the pairs of twin primes are (3, 5), (5, 3), (5, 7) and (7, 5). We consider the
following 3 cases
Case I Top card bears number 3. Then second by multiplication rule, the number of arrangements of
8 cards is
1 × 1 × 6 × 5 × 4 × 3 × 2 × 1 = 720
Case II Top card bears number 5. Then second card has 2 choices 3 and 7.
By Multiplication rule, the number of arrangements of 8 cards is
1 × 2 × 6 × 5 × 4 × 3 × 2 × 1 = 1440
Case III Top card bears number 7. Then second card has only one choice i.e.,5
By Multiplication rule, the number of arrangements of 8 cards is
1 × 1 × 6 × 5 × 4 × 3 × 2 × 1 = 720
Now, by rule of addition, the required number of arrangements is
720 + 1440 + 720 = 2880
Example 3. How many 8 bit strings can formed by using 0 and 1? How many of them begin with 110 or have
the 4th bit 1?
Solution For each of the 8 places, there are 2 choices viz., 0 and 1. Hence, the number of 8 bit strings is
2 × 2 ×...× 2 = 28 = 256
For the 2nd part, we consider three cases.
Case I String starts with 110 and 4th bit is 1.
Here, first four places are fixed.
For next four places, there are 2 choices each. Thus in this case, the number of string is
2 × 2 × 2 × 2 = 16
www.pdfworld.in
www.pdfworld.in
Combinatorics 415
Case II String starts with 110 but 4th bit is not 1. In fact, 4th bit is zero. Here, first four places are
fixed. Next four places have 2 choices each. Thus in this case, the number of strings is
2 × 2 × 2 × 2 = 16
Case III String does not start with 110 and 4th bit is 1. If the group of first three bits is not 110; then it
is one of the following 7 groups
111, 100, 010, 000, 011, 101, 001
The group of first three bits has 7 choices
4th bit is 1.
Remaining of strings in this case is
7 × 1 × 2 × 2 × 2 × 2 = 112
By Addition rule, the total number of required strings is
16 + 16 + 112 = 144
Example 4. Find, how many different + ve integers can be obtained by finding the sum of two or more
from the list 2, 5, 15, 30, 55?
Solution In the given list 2, 5, 15, 30, 55, we see that any member of this list cannot be expressed as the
sum of two or more of its predecessors.
This fact suggests that all the sums of two or more will give us different + ve integers.
There are 5 elements in the given list
Sum of two elements : Out of 5 elements, we can choose 3 elements for addition in 10 ways.
Sum of three elements : Out of 5 elements, we can choose 3 elements for addition in 10 ways.
Sum of four elements : Out of 5 elements, we can choose 4 elements for addition in 5 ways.
Sum of five elements : 5 out of 5 elements can be choose for addition in only one way.
By the rule of addition, the number of different sums are
10 + 10 + 5 + 1 = 26
Example 5. In how many ways can the letters of the word JUPITER be arranged in a row so that the vowels
will appear in alphabetic order?
Solution JUPITER has 7 letters having 3 vowels U, I, E and 4 consonants J, P, T, R.
Alphabetic order of vowels is E, I, U.
The condition alphabetic order to vowels implies that E can appear in Ist, 2nd, 3rd or 5th place.
In every arrangement of 3 vowels, the 4 consonants can be placed in the remaining 4 places in
4 × 3 × 2 × 1 = 24 ways
We consider following 5 cases.
Case I If E appears in Ist place, then I and U will appear in following 15 ways.
(2, 3), (2, 4), (2, 5), (2, 6), (2, 7), (3, 4), (3, 5), (3, 6), (3, 7),
(4, 5), (4, 6), (4, 7), (5, 6), (5, 7) and (6, 7)
Total number of arrangements is
15 × 24 = 360
Case II If E appears in 2nd place, then I and U will appear in following 10 ways.
(3, 4), (3, 5), (3, 6), (3, 7), (4, 5), (4, 6), (4, 7),
(5, 6), (5, 7), (6, 7)
10 × 24 = 240
www.pdfworld.in
www.pdfworld.in
Case III If E appears in 3rd place, then I and U will appear in following 6 ways.
(4, 5), (4, 6), (4, 7), (5, 6), (5, 7), (6, 7)
6 × 24 = 144
Case IV If E appears in 4th place, then I and U will appear in following 3 ways.
(5, 6), (5, 7) and (6, 7)
3 × 24 = 72
Case V If E appears in 5th place, then I and U will appear in 1 way.
(6, 7)
1 × 24 = 24
By Addition rule, total number of arrangements is
360 + 240 + 144 + 72 + 24 = 840
Example 6. A binary word or a binary sequence is a sequence of length n such that each of its term is 0 or 1.
( i ) How many binary words of length n are there?
( ii ) How many binary words of length 10 begin with three 0’s ? How many ends with two 1’s?
Solution (i) Each of the n terms in the word can be chosen in 2 ways 0 or 1.
By Multiplication principle
There are 2n binary words of length n.
(ii) Let P = set of binary words of length 10 which begins with three 0’s.
Q = set of binary words of length 10 which end with two 1’s.
A word in P is of the form 0 0 0 − − − − − − − − whether each of the 7 dashes is either 0 or 1.
Hence, | P | = 27
Likewise, a word in Q is of the form.
− − − − − − − − 11
where each of the 8 dashes is either 0 or 1.
Hence, | Q | = 28
Example 7. Show that the number of ways of making a non-empty collection by choosing some or all of
n1 + ... + nk objects, where n1 are alike of one kind, n2 alike of second kind, ... , nk alike of kth kind is
(n1 + 1) (n2 + 1) .... (nk + 1) − 1.
Solution To make a collection, we have to select a certain number of objects of each kind.
Now, for each value of r, 1 ≤ r ≤ k, from nr like objects we can choose 0, 1,...., or nr objects i.e., there are
nr + 1 choices.
Hence, by Multiplication principle, there are in all (n1 + 1) (n2 + 1) .... (nk + 1) collections.
So, the number of non-empty collections is (n1 + 1) (n2 + 1)... (nk + 1) − 1.
Example 8. Show that the total number of subsets of a set Z with n elements is2n .
Solution Let Z = {a1 , a 2 , ... , an }
Note that we form a subset y of Z in n stages as follows.
We have two choices for a1, either a1 is included in y or a1 is not included in y.
www.pdfworld.in
www.pdfworld.in
Combinatorics 417
Similarly, we have two choices for a 2, either a 2 is included in y or a 2 is not included in y etc., finally, we
have 2 choices for an either an is included in y or an , is not included in y.
For example
If n = 4, then subset {a 2 , a 4 } corresponds to the sequence of choices no, yes, no, yes.
Hence, by Multiplication principle the total number of subsets is
2 × 2 × . . . × 2 (n factors) i . e. , 2n
Example 9. How many times is digit 0 written when listing all numbers from 1 to 3333?
Solution We have to consider integers t such that
1 ≤ t ≤ 3333
There largest number t having 0 in the units place is 3330.
There are 333 numbers t having 0 in the units place they are 10, 20,..., 3330.
We can describe these numbers as t = x 0
where x is anyone of 1, 2,..., 333.
Similarly, number t = x 0y i.e., number having 0 in the ten’s place are in all 33 × 10 because x can be
anyone of 1, 2,..., 33
y can be anyone of 0, 1, 2, …, 9
There are 33 × 10 = 330 number like x 0yz
In the same way there are 3 × 102 = 300 number with 0 in the hundreds place i.e.,
x 0yz where 1 ≤ x ≤ 3
0 ≤ y , z ≤ 10
Hence, the total number of times 0 is written is 333 + 330 + 300 = 963
Example 10. Find the number of ways in which 20 passengers can be put in 3 rooms, such that no room is
vacant. Assume that there is no restriction on the capacity of accommodation in any room.
Solution The total number of ways of accommodating 20 passengers in 3 rooms is320.
Let Ai be the event that the i th room is vacant.
In this case, the 20 passengers are accommodated in 2 rooms in 220 ways.
∴ | A1| = 220
| A 2| = 220
| A3| = 220
If 2 rooms i and j are vacant, then 20 passengers are accommodated in the remaining one room in only
one way.
∴ | A1 ∩ A 2| = 1
| A1 ∩ A3| = 1
| A 2 ∩ A3| = 1
It is impossible that all the rooms remain vacant
∴ | A1 ∩ A 2 ∪ A3 | = 0
3
Now, | A1 ∪ A 2 ∪ A3 | = Σ | Ai | Σ | Ai ∩ A j | + | A1 ∩ A 2 ∩ A3 |
i =1 1≤i ≤ j ≤3
∴ | A1 ∩ A 2 ∩ A3 | = 3 ⋅ 220 − 3 + 0
If no room is vacant, this happens in
320 − 3 ⋅ 220 + 3 ways
www.pdfworld.in
www.pdfworld.in
Example 11. Find the number of +ve integers less than 30 which are relatively prime with 30.
Solution Prime divisor of 30 are 2, 3, 5.
A multiple of any of these 3 numbers is not relatively prime with 30.
We have
A = {multiples of 2} = {2, 4, 6 ,... , 30}
∴ | A | = 15
B = {multiples of 3} = {3, 6, 9 ,... , 30}
∴ | B | = 10
C = {multiples of 5} = {5, 10 ,... , 30}
∴ |C| = 6
A ∩ B is the set of multiples of 2 and 3.
Hence, A ∩ B = {multiples of 6}
∴ A ∩ B = {6, 12 ,... , 30}
∴ |A ∩B|=5
A ∩ C is set of multiples of 2 and 5
∴ A ∩ C = {multiples of 10}
∴ A ∩ C = {10, 20, 30}
∴ |A ∩C|=3
B ∩ C is the set of multiples of 3 and 5.
∴ B ∩ C = {multiples of 15}
∴ B ∩ C = {15, 30}
∴ | B ∩C| =2
A ∩ B ∩ C is the set of multiples of 2, 3 and 5.
∴ A ∩ B ∩ C = {multiples of 30}
∴ A ∩ B ∩ C = {30}
∴ | A ∩ B ∩C| =1
Now, | A ∪ B ∪ C| = | A |+ | B | + |C|−| A ∩ B |−| A ∩C| −|B ∩C|
+ | A ∩ B ∩C|
= 15 + 10 + 6 − 5 − 3 − 2 + 1 = 22
∴ There are 30 − 22 = 8 integers ≤ 30 which are relatively prime with 30.
Example 12. 60 persons form a syndicate and purchase 1 lottery ticket each. The syndicate includes
4 member from the family of Mr. X. If 3 winning tickets are drawn without replacement, find in how many
cases the family of Mr. X will be happy?
Solution Let Ai be the event that the number i of the family of Mr. X wins the lottery;
i = 1, 2, 3, 4
The family will be happy in
| A1 ∪ A 2 ∪ A3 ∪ A 4 | cases.
If member i of family X wins, then remaining 2 tickets are drawn from 59 tickets.
This happens in  59 = 1711 ways.
 2
Here, i = 1, 2, 3, 4
 58
If 2 members i and j of family X win, then remaining 1 ticket is drawn in   ways i.e., 58 ways.
1
www.pdfworld.in
www.pdfworld.in
Combinatorics 419
 4
The 2 members out of 4 are chosen in   = 6 ways.
 2
If 3 members i, j, k of family X win, then this happen in only one way and 3 members can be chosen in
 4
  = 4 ways.
 3
It cannot happen that all 4 members win as only 3 tickets are drawn.
4
Now, | A1 ∪ A 2 ∪ A3 ∪ A 4 | = Σ | Ai | − Σ | Ai ∩ A j |
i =1 1≤i ≤ j ≤ 4
+ Σ | Ai ∩ A j ∩ A k | − | A1 ∩ A 2 ∩ A3 ∩ A 4 |
1≤i ≤ j ≤ k ≤ 4
∴ | A1 ∪ A 2 ∪ A3 ∪ A 4 |
= 4 (1711) − 6 (58) + 4 (1) − 0
= 6844 − 348 + 4
= 6848 − 348 = 6500
The family X is happy in 6500 cases.
Example 13. How many +ve integers n are there such that n is a divisor of one of the numbers1040 , 2030 ?
Solution We first note that the number of +ve divisors of a +ve integer n is
(a1 + 1) (a 2 + 1). . . (a k + 1)
If n = p1a1 p 2a 2 .. . p k a k
where p1, ..., p k are integers.

Now, a = 1040 = 240 540
b = 2030 = 260 530
GCD of a, b is c = 240 530

Let A, B denote the sets of divisor of a, b respectively.
Then A ∩ B is set of divisors of c.
| A | = 412
| B | = 61 × 31
| A ∩ B | = 41 × 31
Hence, | A ∪ B | = 1681 + 1891 − 1271 = 2301
Example 14. Suppose in a poll from 150 people following information obtained. 70 of them read ‘The
Hindustan Times’, 80 read ‘The Times of India’, 50 read ‘Indian Express’, 30 read both ‘The Hindustan Times’
and the ‘Times of India’. 20 read both ‘The Hindustan Times’ and the ‘Indian Express’. 25 read both ‘The
Times of India’ and the ‘Indian Express’. Find at most, how many of them read all the three?
Solution Let H, I, D respectively the set of those who read ‘The Hindustan Times’, The Times of India and
the ‘Indian Express’.
Then
| H ∪ I ∪ D | ≤ 150
| H | = 70
| I | = 80
| D | = 50
www.pdfworld.in
www.pdfworld.in
| H ∩ I | = 30
| H ∩ D | = 20
| I ∩ D | = 25
We need to find the maximum possible value of | H ∩ I ∩ D |
150 ≥ | H ∪ I ∪ D | = | H | + | I | + | D | − | H ∩ I | − | I ∩ D | − | H ∩ D | + | H ∩ I ∩ D |
⇒ 150 − 70 − 80 − 50 + 30 + 20 + 25 ≥ | H ∩ I ∩ D |
∴ | H ∩ I ∩ D | ≤ 25
At most 25 of them read all the three.
Example 15. Let Z be the set of pensioners. E is the set of those that lost an eye. H those that lost an ear. A
those that lost an arm. L those that lost a leg. Given that n (E ) = 70% , n (H ) = 75% , n (A ) = 80% , n (L ) = 85% , find,
what percentage at least must have lost all the four?
Solution Let n (Z ) be 100,
n (Z ) ≥ n (E ∪ H ) = n (E ) + n (H ) − n (E ∩ H )
100 ≥ 70 + 75 − n (E ∩ H )
n (E ∩ H ) ≥ 45
Similarly,
100 ≥ n (L ∪ A ) = n (L ) + n (A ) − n (L ∩ A ) = 80 + 85 − n (L ∩ A )
n (L ∩ A ) ≥ 65
Now, n (Z ) = 100 ≥ n [(E ∩ H ) ∪ (L ∩ A )] = n[(E ∩ H ) + n (L ∩ A ) − n (E ∩ H ∩ L ∩ A )]
⇒ 100 ≥ 45 + 65 − n (E ∩ H ∩ L ∩ A )
⇒ n (E ∩ H ∩ L ∩ A ) ≥ 110 − 100 = 10
At least 10% of the people must have lost all the four.
Example 16. a, b, c, d be integers ≥ 0 , d ≤ a , d ≤ b and a + b = c + d . Prove that there exists sets A and B
satisfying n (A ) = a and n (B ) = b. n (A ∪ B ) = c , n (A ∩ B ) = d
Solution (A ∩ B ) ⊆ A
⇒ n (A ∩ B ) ≤ n (A ) or d ≤a
(A ∩ B ) ⊆ B
⇒ n (A ∩ B ) ≤ n (B )
d ≤a
n (A ∪ B ) = n (A ) + n ( B ) − n (A ∩ B )
⇒ n (A ∪ B ) + n (A ∩ B ) = n (A ) + n (B )
c + d =a + b
Example 17. Find the number of numbers ≤ 108 which are neither perfect squares, nor perfect cubes, nor
perfect fifth powers.
Solution Let [z ] denotes greatest integer ≤ z
Then number of integers from 1 to 108
which are perfect squares =  108  = 104
 
which are perfect cube =  108  = 464

3
 
www.pdfworld.in
www.pdfworld.in
Combinatorics 421
which are perfect fifth power =  108  = 39

5
 
which are both squares and cubes =  108  = 21

6
 
which are both cubes and fifth power =  108  = 3
15
 
which are both squares and fifth power =  108  = 6

10
 
which are squares, cubes and fifth powers =  108  = 1
30
 
By principle of exclusion and inclusion
108 − [104 + 464 + 39 − 21 − 3 − 6 + 1]
is the required number.
Example 18. Consider the following picture

Z1 Z2 Z3 Z4
How many ways are there to colour each of the squares Z1 , Z 2 , Z3 , Z 4 with n different colours provided
that no two adjacent squares can receive the same colour?
Solution Let U = set of all possible ways to colour the squares.
A = set of ways in which Z1 , Z 2 receive the same colour.
C = set of ways in which Z3 , Z 4 receive the same colour.
Now, we can colour the squares using any of the n colours with repetition allowed.
So, | U | = n4
| A | = n3 = | B | = | C |
| A ∩ B | = n2 = | B ∩ C | = | C ∩ A |
| A ∩ B ∩C| = n
Hence, | A ' ∩ B ' ∩ C ' | = n 4 − 3 ⋅ n3 + 3 ⋅ n 2 − n
Example 19. Find the number of ways to choose an ordered pair (a , b ) of numbers from the set
{1, 2,...,10}, such that | a − b | ≤ 5.
Solution Let A1 = [(a , b ) | a , b ∈ {1, 2, 3,... ,10}
| a − b | = { i }, i = 0, 1, 2, 3, 4, 5
A 0 = {( i , i ) | i = 1, 2,... ,10} and | A 0| = 10
A1 = {( i , i + 1) | i = 1, 2, 3,... , 9} ∪ {( i , i − 1) | i = 2, 3,... , 10}
and | A1 | = 9 + 9 = 18
A 2 = {( i , i + 2) | i = 1, 2, 3,... , 8} ∪ {( i , i − 2) | i = 3, 4,... , 10}
and | A 2 | = 8 + 8 = 16
A3 = {( i , i + 3) | i = 1, 2,... , 7} ∪ {( i , i − 3) | i = 4, 5,... , 10}
and | A3 | = 7 + 7 = 14
A 4 = {( i , i + 4) | i = 1, 2,... ,6} ∪ {( i , i − 4) | i = 5, 6,... , 10}
and | A 4 | = 6 + 6 = 12
A5 = {( i , i + 5) | i = 1, 2,... , 5} ∪ {( i , i − 5) | i = 6, 7,... , 10}
and | A5 | = 5 + 5 = 10
www.pdfworld.in
www.pdfworld.in
5
∴ The required set of pairs (a , b ) = ∪ Ai
i =0
The number of such pairs, (which are disjoint)

5 5
= | ∪ Ai | = Σ | Ai | = 10 + 18 + 16 + 14 + 12 + 10 = 80
i =0 i =0
Example 20. How many numbers from 1 to 1000 are not divisible by 2, 3 and 5?
Solution Let a1 being the property of being divisible by 2, a2 the property of being divisible by 3, a3 the
property of being divisible by 5.
Then
a1 a 2 denote the property of being divisible by 6
a1 a3 denote the property of being divisible by 10
a 2 a3 denote the property of being divisible by 15
a1 a 2 a3 denote the property of being divisible by 30
We have to find n (a1′ a 2′ a3′ ) = n − n (a1 ) − n (a 2 ) − n (a3 ) + n (a1 a 2 ) + n (a1 a3 ) + n (a 2 a3 ) − n (a1a 2 a3 ) ...(i)
We are given that n = 1000
n (a1 ) = 
1000 
= 500
 2 
n (a 2 ) = 
1000   1000  = 200
= 333 ; n (a3 ) =
 3   5 
 1000   1000 
n (a1a 2 ) =   = 166 ; n (a1a3 ) =   = 100
 2 ⋅ 3   2⋅5 
 1000   1000  = 33
n (a 2a3 ) =   = 66 ; n (a1a 2a3 ) = 
 3 ⋅ 5   30 
Substitute these values in Eq. (i), we get
n (a1′ a 2′ a3′ )
= 1000 − 500 − 333 − 200 + 166 + 100 + 66 − 33
= 266
Example 21. A man has 6 friends. At dinner in a certain restaurant, he has met each of them 12 times,
every 2 of them 6 times, every 3 of them 4 times, every 4 of them 3 times, every 5 twice and all 6 only once. He
has dined out 8 times without meeting any of them. How many times has he dined out altogether?
Solution Let U be the set of all days on which the man has dined out.
Let a1 , a 2 ,... , a 6 be the 6 friends.
Let us say that a day in U has property pi , if the man meets ai on that day
Let N be the number of days with at least one of the properties p1 , ... , p 6
Then we are given that
N ( p1 ) = N ( p 2 ) = ... = 12
N ( p1p 2 ) = N ( p1p3 ) = ... = 6
N ( p1p 2p3 ) = N ( p1p 2p 4 ) = ... = 4
N ( p1... p 4 ) = 3
N ( p1... p5 ) = 2
N ( p1... p 6 ) = 1
|U | − N = 8
www.pdfworld.in
www.pdfworld.in
Combinatorics 423
 6  6  6  6  6 6 
∴ N =   12 −   6 +   4 −   3 +   2 −  1
 1  2  3  4  5 6 
N = 28
So, |U | − N = 8
| U | = N + 8 = 36
Example 22. Let A, B be finite sets with | A | = r and | B | = n. Find the number of functions from A onto B.
Solution Let A = {a1 ,... , ar }
B = {b1 , ... , bn }
Let A1 be the set of functions from A into B − {b1}
A 2 be the set of functions from A into B − {b2}, ... An be the set of functions from A into B − {bn }.
Let U be the set of functions from A into B.
Let f ∈ U i.e., let f : A → B
Then f is onto B if and only if the range of f is the whole of B. i. e., if and only if f is not in any of the sets
A1 , ... , An .
For example, if f ∈ A1 then b1 is not in the range of f.
Hence, the number of functions from A onto B is
| A1′ ∩ ... ∩ An′ |
We have | U | = nr
Similarly, | Ai | = (n − 1)r for 1 ≤ i ≤ n
∴ | B − {bi } | = n − 1
Hence, S1 = Σ | Ai | = n (n − 1)r
f ∈ A1 ∩ A 2 if and only if f is from A into B − {b1 , b2}.
Hence, | A1 ∩ A 2| = (n − 2)r
Similarly, | Ai ∩ A j | = (n − 2)r
n 
for each of the   pairs Ai , A j with 1 ≤ i ≤ j ≤ n
 2
n 
Hence, S 2 = Σ | Ai ∩ A j | =   (n − 2)r
 2
n 
∴ S k =   (n − k )r , for 3 ≤ k ≤ n − 1
k
So, by principle of inclusion-exclusion, we have
| A 'i ∩ A '2 ∩ ... ∩ A 'm | = | U | − S1 + S 2 − ... + ( − 1)n −1 Sn −1
n 
= nr − n (n − 1)r +   (n − 2)r − ... + ( − 1)n −1n
 2
Example 23. Let φ (n ) denote Euler’s totient function prove that if p1e1 pe22 ..... pekk is the factorization of the
+ve integers n into distinct primes pi and ei are +ve integers, then φ (n ) is given by
 1  1  1
φ (n ) = n  1 −   1 −  ...  1 − .
 p1   p2   pk 
www.pdfworld.in
www.pdfworld.in
Solution φ (n ) is the number of integers < n and relatively prime to n.

Let U = {1, 2, ... , n}
Let Ai be the set of those integers in U which are divisible by the prime pi .
n
Then | Ai | =
pi
| Ai ∩ A j | = numbers in U divisible by pi and p j
n
=
pi p j
n
| Ai ∩ A j ∩ A k | = and so on.
pi p jp k
Hence, principle of inclusion-exclusion gives the number of numbers in U not divisible by any of the
primes pi i.e., φ (n )
n n n n
φ (n ) = n − Σ + Σ −Σ + ... + ( − 1)k
pi pi p j pi p j p k p1 p 2 ... p k
 1 1 1 1 
= n 1 − Σ + Σ −Σ + ... + ( − 1)k 
 pi pi p j pi p j p k p1 p 2 ... p k 
 1  1  1
= n 1 −   1 −  … 1 − 
 p1   p2   pk 
Example 24. A, B and C are the sets of all +ve divisors of 1060 , 2050 and 3040 respectively.
Find n (A ∪ B ∪ C ).
Solution Let n (A ) = number of positive divisors of 1060
= 260 × 560 is 612
n (B ) = number of +ve divisors of 2050

= 2100 × 550 is 101 × 51
40
n (C ) = number of +ve divisors of 30
= 240 × 340 × 540 = 413
The set of common factors of A and B will be of the form 2m 5n , where 0 ≤ m ≤ 60, 0 ≤ n ≤ 50
So, n (A ∩ B ) = 61 × 51
Similarly, the common factors of B and C, A and C are also of the form 2m × 5n
and in the former case 0 ≤ m ≤ 40, 0 ≤ n ≤ 40
and in the latter case 0 ≤ m ≤ 40, 0 ≤ n ≤ 40
∴ n (B ∩ C ) = 412 also, n (A ∩ C ) = 412
and n (A ∩ B ∩ C ) = 412
∴ n (A ∪ B ∪ C ) = n (A ) + n (B ) + n (C ) − n (A ∩ B ) − n (B ∩ C )
− n (A ∩ C ) + n (A ∩ B ∩ C )
= 612 + 101 × 51 + 413 − 61 × 51 − 412 − 412 + 412
= 61 (61 − 51) + 412 (41 − 1) + 101 × 51
= 610 + 1681 × 40 + 101 × 51
= 73001
www.pdfworld.in
www.pdfworld.in
Combinatorics 425
Example 25. I have 6 friends and during a certain vacation, I met them during several dinners. I found
that I dined with all the 6 exactly on 1 day, with every 5 of them on 2 days, with every 4 of them on 3 days,
with every 3 of them on 4 days and with every 2 of them on 5 days. Further every friends was present at 7
dinners and every friend was absent at 7 dinners. How many dinners did I have alone?
Solution For i =1, 2, 3, . . . ,6. Let Ai be the set of days on which ith friend is present at dinner then given
n (Ai ) or | Ai | = 7 and | A ′i | = 7
So, | Ai ∩ A j | = 5 , | Ai ∩ A j ∩ A k | = 4
| Ai ∩ A j ∩ A k ∩ Al | = 3,
| Ai ∩ A j ∩ A k ∩ Al ∩ Am | = 2
and | A1 ∩ A 2 ∩ A3 ∩ A 4 ∩ A5 ∩ A 6 | = 1
where i, j, k, l, m vary between 1 to 6 and are distinct.
| A1 ∪ A 2 ∪ A3 ∪ ... ∪ A 6 |
6
= Σ | Ai | − Σ | Ai ∩ A j | + Σ | Ai ∩ A j ∩ A k | − Σ | Ai ∩ A j ∩ A k ∩ Al |
i =1
+ Σ| Ai ∩ A j ∩ A k ∩ Al ∩ Am | − | A1 ∩ A 2 ∩ A3 ∩ A 4 ∩ A5 ∩ A 6 |
 6  6  6  6  6  6
=   (7) −   (5) +   (4) −   (3) +   (2) −   (1)
 1  2  4  4  5  6
= 42 − 75 + 80 − 45 + 12 − 1 = 13
Total number of dinners
= | Ai | + | A ′i | = 7 + 7 = 14
The number of dinners in which at least one friend was present.
= | A1 ∪ A 2 ∪ A3 ∪ A 4 ∪ A5 ∪ A 6 | = 13
Number of dinner i dine alone = 14 – 13 = 1
Example 26. A student on vacation for d days observed that

(i) it rained 7 times morning or afternoon
(ii) when it rained in the afternoon it was clear in the morning
(iii) there were 5 clear afternoons, and
(iv) there were 6 clear mornings.
Find d.
Solution Let the set of days it rained in the morning be M2.
Let Ar be the set of days it rained in afternoon.
M ′ be the set of days, when there were clear morning.
Ar′ be the set of days when there were clear afternoon.
By condition (ii), we get Mr ∩ Ar = φ
By (iv), we get Mr′ = 6
By (iii), we get Ar′ = 5
By (i), we get Mr ∪ Ar = 7
Mr and Ar are disjoint sets
n (Mr ) = d − 6
n (Ar ) = d − 5
Applying the principle of inclusion-exclusion
n (Mr ∪ Ar ) = n (Mr ) + n (Ar ) − n (Mr ∩ Ar )
www.pdfworld.in
www.pdfworld.in
⇒ 7 = (d − 6) + (d − 5) − 0
⇒ 2d = 18
⇒ d =9
Example 27. Find, how many number of 4 different digits out of the digits 1, 2, 3, 4, 5, 6, 7 can be made?
Also find, how many of them are greater than 3400?
Solution Number of digits, n = 7 for the numbers of 4 digit number of places to be filled upr = 4
Hence, number of numbers of 4 different digits out of the digits 1, 2, 3, 4, 5, 6, 7 = 7 P4
7!
= = 840
(7 − 4)!
Now, numbers in which thousands place is filled with 3 and hundred with 4 and remaining 2 places with
any must be greater than 3400.
Now, for the number of numbers of 4 digits in which thousands i.e., (Ist from left) places is filled with 3
and hundred (2nd from left) with 4 then for the remaining 2 places there are (1, 2, 5, 6, 7,) 5 digits i.e.,
n = 5 and r = 2
5!
Number of 4 digits such that 3 is at thousands place and 4 at hundreds place 5 P2 = = 20
3!
Similarly, number of numbers of 4 digits such that 3 comes at thousands and 5 at hundred place is
5
P2 = 20
Number of numbers of 4 digits such that 3 comes at thousand and 6 at hundreds place is
5
P2 = 20
Number of numbers of 4 digits such that 3 comes at thousand and 7 at hundred places
= 5P2 = 20
Number of numbers of 4 digits beginning with 3 (and in which 4, 5, 6 or 7 come at 100 place) and greater
than 3400
= 20 + 20 + 20 + 20 = 80
Also, number of 4 digits beginning with 4, 5, 6 or 7 are all greater than 3400.
Number of numbers of 4 digits beginning with 4
Number of places remain to be filled up isr = 3
Number of digits remain to be utilised is n = 6
Number of numbers of 4 digits beginning with 4 is
6!
= 6P3 = = 120
3!
Also, number of numbers of 4 digits beginning with 5 = 6P3 = 120
Number of numbers of 4 digits beginning with
6 = 120
Number of numbers of 4 digits beginning with
7 = 120
Number of numbers of 4 digits and greater than 3400 = 80 + 480 = 560
Example 28. How many +ve numbers can be formed by using any number of the digit 0, 1, 2, 3, and 4 no
digit being repeated?
Solution Number of digits, n = 5
Number of numbers of 1 digit = 4 …(i)
www.pdfworld.in
www.pdfworld.in
Combinatorics 427
To find the number of numbers of 2 digits number of places to be filled upr = 2

Number of digits, n = 5
Number of numbers of 2 digits
5!
= 5P2 = = 20
3!
But out of these 20 numbers of 2 digits, some numbers begins with 0 which are not of our purpose.
To find the number of such numbers of digits beginning with 0, number of places and number of digits
remains to be utilised = n = 4
Hence, number of numbers of 2 digits which begin with 0 = 4
P1 = 4
= 20 − 4 = 16 …(ii)
For the number of numbers of 3 digits, number of places to be filled up = r = 3
Number of objects, n = 5
5!
= 5P3 = = 60
2!
But of these 60 numbers, the number which begin with 0 cannot be regarded as the number of 3 digits.
Now, to find the number of such number of 3 digits which begin with 0.
Number of places = r = 2
Number of things remains to be utilised n = 4
Number of numbers of 3 digits beginning with 0 is 4 P2 i.e.,
4!
= 12
2!
Number of numbers of 3 digits = 60 − 12 = 48
Also, for the number of numbers of 4 digits number of places to be filled up = r = 4. Number of digits to
be utilised, n = 5
5!
Number of numbers of 4 digits = 5P4 = = 120
1!
But out of these 120 numbers, there are some numbers which begin with zero.
For the numbers of 4 digits beginning with 0.
Number of places remain to be filled up,r = 3
Number of digits remain to be utilised n = 4
Number of numbers of 4 digits beginning with 0
4!
= 4 P3 = = 24
1!
For the number of numbers of 5 digits, number of places to be filled up,r = 5
Number of digits, n = 5
5!
= 5P5 = = 120
0!
But out of these 120 numbers, there are some numbers which begin with zero.
Number of places to be filled up r = 4
Number of digits remain to be utilised, n = 4
4!
Number of numbers of 5 digits which begin with 0 = 4 P4 = = 24
0!
www.pdfworld.in
www.pdfworld.in

Number of +ve numbers formed 0, 1, 2, 3, 4, is
4 + 16 + 48 + 96 + 96 = 260
Example 29. How many numbers can be formed with the digits 1, 3, 7, 9, when taken all at a time? Find
their sum.
Solution Number of given digits n = 5
Number of places to be filled = r = 5
Numbers are to be formed taken all at a time
5!
Number of different numbers of 5 digits formed with the digits 1, 3, 5, 7, 9 is 5 P5 == 120
0!
Also, keeping 1 at unit place, number of numbers of 5 digits in which 1 comes at unit place
4!
4
P4 = = 24
0!
Number of places remain to be filled up = r = 4
Number of digits remain to be utilised = n = 4
i.e, in the number of numbers of 5 digits 1 may come at unit place in 24 number of times.
Similarly, 3, 5, 7, 9, each may come at (unit, tens, hundreds, thousands and ten thousands) each place in
24 number of times.
Sum of the digits at unit place
= (1 + 3 + 5 + 7 + 9) × 24 × 1
Sum of the digits at tens place
= (1 + 3 + 5 + 7 + 9) × 24 × 10
Sum of the digits at 100 place
= (1 + 3 + 5 + 7 + 9) × 24 × 100
= (1 + 3 + 5 + 7 + 9) × 24 × 1000
= (1 + 3 + 5 + 7 + 9) × 24 × 10000
Sum of all the number of 5 digits
= (1 + 3 + 5 + 7 + 9) × 24
(1 + 10 + 100 + 1000 + 10000)
= 25 × 24 × 11111 = 600 × 11111 = 6666600
Example 30. Find the sum of all the four digits number that formed with the digits 3, 2, 3, 4.
Solution We first keep one '3' at unit place, then number of places to be filled up, r = 3
Number of digits remain to be utilised, n = 3 ( 3, 2, 4)
Number of numbers of 4 digits formed with ( 3, 2, 3, 4) in which 3 comes at unit place
3!
= 3P3 = =6
0!
Hence, 3 may come at unit place in 6 ways.
Similarly, 3 may come at (tens, hundreds and thousands) each place in 6 number of times.
But keeping 2 at unit place, number of places be filled up,r = 3
Number of digits remain to be utilised = n = 3 ( 3, 3, 4)
www.pdfworld.in
www.pdfworld.in
Combinatorics 429
3
P3 6 !
Number of numbers of 4 digits in which 2 occurs at unit place = =
2! 2
Similarly, 2 may come at (10’s, 100’s, 1000’s) each place 3 number of times.
Similarly, 4 may come at (unit, 10’s, 100’s, 1000’s) each place 3 number of times.
Sum of digits at unit place
= [(2 + 4) × 3 + 3 × 6] × 1 …(i)
Sum of digits at 10’s place
= [(2 + 4) × 3 + 3 × 6] × 10 …(ii)
= [(2 + 4) × 3 + 3 × 6] × 100 …(iii)
= [(2 + 4) × 3 + 3 × 6] × 1000 …(iv)
Adding (1), (2), (3), (4), altogether we get 39996
Sum of all four digits number formed with the digits 3, 2, 3, 4 is 39996.
Example 31. Find the sum of all the 4 digit numbers, that can be formed with the digits 1, 2, 2 and 3.
Solution Keeping 1 or 3 at unit place
Number of places remain to be filled = r = 3
Number, of digits remain to be utilised (2, 2, 3) = n = 3
3
P3 6 !
Hence, number of numbers of 4 digits in which 1 or 3 comes at unit place = = =3
2! 2
Similarly, 1 or 3 may come at (10’s, 100’s, 1000’s) each place in 3 number of times.
i.e., 1 or 3 may come at (unit or 10’s or 100’s or 1000’s) place 3 number of times.
Also, keeping 2 at unit place
Number of places remain, to be filled up = r = 3
Number of digits remain to be utilised = n = 3 (1, 2, 3)
3!
Number of numbers of 4 digits such that 2 comes at unit place = 3P3 = =6
0!
i.e., 2 may come at unit (or 10’s or 100’s or 1000’s) place 6 number of times.
Sum of digits at unit place = [(1 + 3) × 3 + 2 × 6] × 1
Sum of digits at 10’s place = [(1 + 3) × 3 + 2 × 6] × 10
Sum of digits at 100’s place = [(1 + 3) × 3 + 2 × 6] × 100
Sum of at 1000’s place = [(1 + 3) × 3 + 2 × 6] × 1000
Sum of all the 4 digit numbers = [(1 + 3) × 3 + 2 × 6] [1 + 10 + 100 + 1000]
= 24 × 1111 = 26664
Example 32. How many odd numbers greater than 600000 can be formed from the digits 5, 6, 7, 8, 9, 0 if
repetition of digits is allowed?
Solution Numbers greater than 600000 and formed with the digits 5, 6, 7, 8, 9, 0 are of 6 digits but begin
with 6, 7, 8 or 9.
Also, the numbers which end with 5, 7, 9 are odd.
Hence, first place can be filled by 4 ways (out of 6, 7, 8 or 9). Last place can be filled by 3 ways.
Hence, first and last places can be filled by 4 × 3 ways.
Also, 2nd place can be filled by 6 ways.
3rd place can be filled by 6 ways.
www.pdfworld.in
www.pdfworld.in
4th place can be filled by 6 ways.

5th place can be filled by 6 ways.
Hence, all the 6 places can be filled by
4 × 3 × 6 × 6 × 6 × 6 = 15552 ways.
Example 33. In a car plate number containing only 3 or 4 digits not containing the digit 0. What is the
maximum number of cars that can be numbered?
Solution Here, repetition of digits is allowed.
Also, numbers are formed with the digits 1, ..., 9.
Case I When car plate numbers contain 3 digits number of places to be filled upr = 3
Out of the 9 digits first place can be filled by 9 ways.
Similarly, 2nd and 3rd place can be filled in 9 ways respectively.
So, when car plate number contains 3 digit, maximum number of cars = 93
Case II When car plate number contains 4 digit in this case number of cars to be filled up = r = 4
Ist place can be filled by 9 ways,
2nd place can be filled by 9 ways and so on.
Maximum number of cars that can be numbered = 93 + 94 = 7290
Example 34. m men and n women are to be seated in a row so that no two women sit together, if m > n.
Show that the number of ways in which they can be seated is
m ! ( m + 1)!
.
(m − n + 1)!
Solution Number of arrangements of m men in a line

(n = number of thing (men )m )
( r = number of place)
m!
is m
Pm = =m!
0!
Since, no 2 women are to be seated together so that number of women may be m + 1.
Hence, number of arrangements of n women in the (m + 1) place is m +1pn .
Required number of arrangements
m +1 m !(m + 1)!
=m! × Pn =
(m + 1 − n )!
m !(m + 1)!
=
(m − n + 1)!
Example 35. Show that number of ways in which p +ve and n –ve signs may be placed in a row so that no
two –ve signs shall be together is p + 1Cn .
Solution Number of arrangements of all the p +ve signs in a row.

p
Pp
= ( since n = p , r = p )
p!
p!
=
p!
Since, two –ve signs come together so that number of places for the −ve signs is p + 1 .
Number of arrangements of the n –ve signs according to the condition
www.pdfworld.in
www.pdfworld.in
Combinatorics 431
p +1
Pn
is ( since n = p + 1 ; r = p + 1 ) (all the n are identical )
n!
( p + 1)!
=
n ! ( p + 1 − n )!
Finally, required number of arrangements of p +ve and n –ve signs in a row so that no two −ve signs are
together
p! ( p + 1)!
= ×
p ! ( p + 1 − n )! n !
( p + 1)! p +1  n n! 
= = Cn Q Cr = 
n ! ( p + 1 − n )!  (n − r )! r ! 
Example 36. If the letters of the word ‘LATE’ to be permutated and the words so formed be arranged as in
a dictionary what with be the rank of the word LATE.
Solution Arrange alphabetically LATE, we have AELT. Now, for the words beginning with A, number of
places to be filled up = r = 3.
Number of letters to be utilised, n = 3.
Number of words beginning with A = 3P3 = 3! number of words beginning with E = 3! .
Also, for the words beginning with LA, number of places to be filled up = r = 2 .
Number of letters remain to be utilised, n = 2 . Hence, number of word LATE.
Rank of the word LATE in the dictionary = 3 ! + 3 ! + 2 ! = 14
Example 37. On a new year day, every students of a class sends a card to every other student. The
postman delivers 600 cards. How many students are there in the class?
Solution Total number of students = n
Number of pair of students = n C2
Two students out of n can be selected in n C2 ways. Here, for each pair of students, number of cards sent
is 2.
If P sends card to Q , then Q also sends a card to P. Number of cards sent = 2 ⋅ n C2.
According to the problem 2 ⋅ n C2 = 600
n (n − 1)
or 2⋅ = 600
2!
or n 2 − n − 600 = 0
(n − 25) (n + 24) = 0
n = 25 [ Q n ≠ − 24]
Example 38. You are given n ≥ 3 circles. Find the number of radical axis and radical centres of these
circles. Find the value of n to which number of radical axis is equal to number of radical centres.
Solution Radical axis is the locus of the intersection of the two tangents equal in length to the 2 circles.
Hence, radical axis associates with 2 circles. Number of radical axis of n circles is n C2. Also, number of
radical centres of n circles is n C3 .
We are given n
C2 = n C3 ⇒ n C2 = n Cn −3
or n −3=2
or n =5
www.pdfworld.in
www.pdfworld.in
Example 39. In how many ways the letters of the word PERSON can be placed in the squares of the given
figure shown so that no row remain empty?
R1
R2
R3
Solution There are 6 different letters in the word PERSON.

There are 3 rows, we have to select 6 squares taking at least one from each row.
Selection of 6 squars 1 from Ist row, 1 from 2nd row and 4 from 3rd row can be made in
2
C1 ⋅ 2C1 ⋅ 4C4 = 4 ways.
Selection of 6 squares, 1 from 1st row, 2 from 2nd row and 2 from the 3rd row can be made in
2
C1 ⋅ 2C2 ⋅ 4C3 = 8 ways.
Selection of 6 squares 2 from 1st row, 1 from 2nd row and 3 from 3rd row can be made in
2
C2 ⋅ 2C1 ⋅ 4C3 = 8 ways.
Selection of 6 squares, 2 from 1st row, 2 from 2nd row and 2 from the 3rd row can be made in
2
C2 ⋅ 2C2 ⋅ 4C2 = 6 ways.
Total number of selection of 6 squares
= 4 + 8+ 8+ 6
= 26
Now, for each selection of 6 squares, the number of arrangements of 6 letters
= 6 ! = 720
Required number of ways = 26 × 720
= 18720
Example 40. (i) In how many ways can 16 constables be divided into 8 batches of 2 each?
(ii) In how many ways 16 constables can be assigned to petrol 8 villages 2 for each?
Solution (i) In case of dividing 16 constables into 8 batches of 2 each, order of formation of groups does
not matter. Hence, 16 constables can be divided into 8 batches of 2 each in
16 14 12 10
C2 C2 C2 C2 8C2 6C2 4C2 2C2
=
8!
16 !
=
8! (2 !)8
(ii) When 16 constables are to be assigned to petrolling 8 villages 2 for each. Then we first divide
16 !
16 constables into 8 batches as in (i) in ways.
8! (2 !)8
But in this case, a question that which batch petrols the first village and which the 2nd and so on, arises
automatically i.e., order of formation of batches must be taken into account. Hence, number of ways of
assigning 8 batches to petrolling 8 villages
16
= 8!
8! (2 !)8
16
=
(2 !)8
www.pdfworld.in
www.pdfworld.in
Combinatorics 433
Example 41. Find the number of ways of dividing 2n people into n couples.
Solution Here, we have to make a group of 2 people out of 2n people. The number of ways in which we
can divide 2n people into n group of 2.
2
2n 2n − 2 2n − 4 C2
C2 C2 C2 …
n2 !
2n ! 2n ! − 2 2n ! − 4 4! 2!
= ⋅ ⋅ … ⋅
2 ! 2n ! − 2 2 ! 2n ! − 4 2 ! 2n ! − 6 2!2! 2! 0!
2n !
=
2n (2n !)2
Example 42. Find the number of ways in which we can distribute mn students equally among m sections.
Solution When we distribute mn students equally among m sections, then each section has n students.
Hence, number of ways of distribution
mn −n mn − 2n 1
= (mn Cn Cn Cn ...2n Cnm! n
Cn )
m!
mn ! mn ! − n 2n ! n!
= ⋅ ....... ⋅
n ! mn ! − n n ! mn ! − 2n n ! n ! n ! 0!
mn !
=
(n !)m
Example 43. From a given number of 4n books, there are three sets of n identical books each on Physics,
Chemistry and Mathematics. The remaining n books are distincts books of other subjects. Find the number of
ways of choosing n books out of the 4n books.
Solution There are n identical books of one kind, n identical books of 2nd, n identical books of 3rd kind
and rest n books are all different.
Hence, number of selections of any n books out of these 4n books
= coefficient of x n in ( x 0 + x + x 2 + x 3 + ... + x n )
(x 0 + x + x 2 + … + x n ) (x 0 + x + x 2 + x 3 + … + x n ) (1 + x )n
= coefficient of x n in (1 + x + x 2 + x 3 + …+ x n )3 (1 + x )n
3
 1 − x n +1 
= coefficient of x n in   ⋅ (1 + x )
n
 1−x 
= coefficient of x n in (1 − x n + 1 )3 (1 − x )−3 (1 + x )n
= coefficient of x n in (1 − x )−3 [2 − (1 − x )]n
= coefficient of x n in (1 − x )−3 [2n − n C12n −1
(1 − x )
−2
+ n C2nn (1 − x )2 − n
C3 2n −3 ( 1 − x )3 + … + (− 1)n ( 1 − x )n ]
= coefficient of x n in 2n (1 − x )−3 − n C1 2n −1 (1 − x )−2 + n C2 2n − 2 (1 − x )
= coefficient of x n in [2n (1 + 3C1x + 4 C2 x 2 +…) − n ⋅ 2n −1 (1 + 2C1 x + 3C2 x 2 + …)
n (n − 1) n − 2
+ 2 (1 + 1C1x + 2C2 x 2 +…)]
2
n+2 n (n − 1) n − 2
= 2n Cn − n ⋅ 2n −1 n +1
Cn + ⋅2
2
n+2
= 2n C2 − n ⋅ 2n −1 n +1
C1 + 2n − 3 (n 2 − n )
www.pdfworld.in
www.pdfworld.in
(n + 2) (n + 1)
= 2n − n ⋅ 2n − 1 (n + 1) + 2n −3
(n 2 − n )
2
= 2n −1 (n 2 + 3n + 2) − 2n −1 ( n 2 + n ) +2n − 3 (n 2 − n )
= 2n − 3[22 (n 2 + 3n + 2) − 22 (n 2 + n ) + n 2 − n ] = 2n − 3 (n 2 + 7n + 8)
Example 44. In an examination of the maximum marks for each of the three papers are 50 each and
maximum marks for the fourth paper is 100. Find the number of ways in which a candidate secure 60%
marks in aggregate.
Solution Here, aggregate of marks
= 50 + 50 + 50 + 100 = 250
and 60% of aggregate 250
60
= × 250 = 150
100
Now, number of ways in which we can get 150 marks in the examination of 4 papers such that first three
have maximum marks 50 and 4th has maximum marks 100 is
= coefficient of x 150 in (x 0 + x + x 2 + ...+ x 50 )3 (x 0 + x + x 2 + ... +x 100 )
= coefficient of x 150 in (1 + x + x 2 + K + x 50 )3 (1 + x + x 2 + K + x 100 )
3
 1 − x 51   1 − x 101 
= coefficient of x 150 in    
 1−x   1−x 
(1 − 3x 51 + 3x 102 − x 153 ) (1 − x 101 )
(1 − x )4
= coefficient of x 150 in (1 − 3x 51 − x 101 + 3x 102 + ... ) (1 − x )−4
= coefficient of x 150 in (1 − 3x 51 − x 101 + 3x 102 + ... )
(1 + 4C1 x + 5C2 x 2 + ... )
= 153C150 − 3 102C99 − 52C49 + 3 51
C48
153 ⋅ 152 ⋅ 151 102 ⋅ 101 ⋅ 100 52 ⋅ 51 ⋅ 50 51 ⋅ 50 ⋅ 49
= −3⋅ − + 3⋅
6 6 6 6
= 76 ⋅ 51 ⋅ 151 − 100 ⋅ 101 ⋅ 51 − 50 ⋅ 26 ⋅ 17 + 51 ⋅ 49 ⋅ 25
= 51 (76 × 151 − 100 × 101 ) + 17 (75 × 49 − 50 × 26)
= 51 (11476 − 10100) + 17 (3675 − 1300)
= 51 × 1376 + 17 × 2375
= 70176 + 40375
= 110551
Example 45. In how many ways can 16 apples be distributed among 4 persons, each receiving not less
than 3 apples?
Solution The apples are considered to be alike.
Since, a person will not receive less than 3 apples so that one person can get 3, 4, … apples.
Hence, number of ways of distributing 16 apples among 4 persons.
= coefficient of x 16 in (x 3 + x 4 + x 5 + ... )4
= coefficient of x 16 in x 12 (1 + x + x 2 + ... )4
www.pdfworld.in
www.pdfworld.in
Combinatorics 435
4
 1 
1 − x 
= coefficient of x 4 in (1 + 4C1 x + 5C2 x 2 + ... ) = 7C4
7! 7 × 6 × 5 × 4!
= = = 35
4! 3! 4! × 3 × 2
Example 46. There are 12 intermediate stations between two places A and B. In how many ways can a
train be made to stop at 4 of these 12 intermediate stations, no two of which are consecutive?
Solution Let S1 , S 2 , S3 , S 4 be the 4 stopping stations.
Let ... S1 ... S 2 ... S3 ... S 4 ...
n1 n2 n3 n4 n5
Let n1 and n5 be the number of stations respectively before S1 and after S 4 and let n2 , n3 and n4 be the
number of stations between S1 and S 2, S 2 and S3, S3 and S 4 respectively. Then
n1 + n2 + n3 + n4 + n5 = 12 − 4 = 8
where n1 n5 ≥ 0 ; n3 , n4 ≥ 1
Now, the number of ways in which a train stops at 4 of these 12 stations such that no two of four are
consecutive is
= coefficient of x 8 in (x 0 + x + x 2 + ... ) (x + x 2 + x 3 + ... )
(x + x 2 + x 3 + ... ) (x + x 2 + x 3 + ... ) (x 0 + x + x 2 + ... )
= coefficient of x 8 in (1 + x + x 2 + ... )2 (x + x 2 + x 3 + ... )3
= coefficient of x 8 in x 3 (1 + x + x 2 + ... )2 (1 + x + x 2 + ... )3
5
 1 
1 − x 
= 9C5
9!
=
5! 4!
9 ⋅ 8⋅ 7 ⋅ 6 ⋅ 5 !
=
4⋅3⋅2 ⋅ 5!
=9×2×7
= 126
Example 47. How many integral Solutions are there to x + y + z + t = 29, when x > 0, y > 1, z > 2 and
t ≥ 0?
Solution We have to select four integral for x , t , z and t. For x we may choose ( > 0) 1, 2, 3, 4,…
For y we may choose (> 1 ) 2, 3, 4, …
For z we may choose (> 2) 3, 4, 5, …
For t we may choose (≥ 0) 0, 1, 2, …
Such that sum of the four chosen integers for x, y, z and t is = 29 i.e., x + y + z + t = 29
www.pdfworld.in
www.pdfworld.in
Hence, number of ways of choosing four integers for x , y , z and t = number of integral solutions of
x + y + z + t = 29 for x > 0, y > 1, z > 2, t ≥ 0 is given by the coefficient of x 29 in (x + x 2 + x 3 + ... )
(x 2 + x 3 + x 4 + ... ) (x 3 + x 4 + x 5 + ... ) (x 0 + x + x 2 + ... )
= coefficient of x 29 in x (1 + x + x 2 + ... ) x 2 (1 + x + x 2 + ... )
x 3 (1 + x + x 2 + x 3 + ... ) (1 + x + x 2 + ... )
= coefficient of x 29 in x 6 (1 + x + x 2 + ... )4
4
 1 
1 − x 
= coefficient of x 23 in (1 + 4C1 x + 5C2 x 2 + C3 x 3 + ... )
6
26 !
= 26
C23 =
23 ! 3 !
26 ⋅ 25 ⋅ 24
=
3⋅ 2 ⋅ 1
= 26 × 100
= 2600
Example 48. How many integral solutions are there to the system of equations
x 1 + x 2 + x 3 + x 4 + x 5 = 20
and x 1 + x 2 + x 3 = 5, when x 2 ≥ 0 ?
Solution Here, x 1 + x 2 + x 3 + x 4 + x 5 = 20 …(i)
x1 + x 2 + x3 = 5 …(ii)
so that x 3 + x 4 = 15 …(iii)
with x 2 ≥ 0
Now, obviously Eqs. (i) and (ii) hold if Eqs. (ii) and (iii) hold. Hence, we find the integral solutions of
x1 + x 2 + x3 = 5
and x 3 + x 4 = 15 …(iv)
with x 2 ≥ 0
Now, number of integral solutions of Eq. (i) is given by the coefficient of x 5 in (x 0 + x + x 2 + ...... )3
= coefficient of x 5 in (1 + x + x 2 + ...... )3
3
 1 
1 − x 
= 1 + 3C1 x + 4C2 x 2 + ...
7⋅ 6
= 7C5 = = 21
2⋅1
Also, number of integral solutions, if Eq. (ii) is given by the coefficient ofx 15 in (x 0 + x + x 2 + ... )2
2
 1 
1 − x 
www.pdfworld.in
www.pdfworld.in
Combinatorics 437
= 1 + 2C1 + 3C2 + … = 16
C15
Hence, number of solutions = C2 × 16 = 336
3
Example 49. Find the number of non-negative integral solutions of

x 1 + x 2 + x 3 + 4x 4 = 20
Solution We have to find the number of non-negative integral solutions of
x 1 + x 2 + x 3 + 4x 4 = 20 ...(i)
Hence, we have to take x 1 , x 2 , x 3 , x 4 from 0, 1, 2, 3, …, so that Eq. (i) holds
Now, putting x 1 = x 2 = x 3 = 0, we have 4x 4 = 20
∴ x4 = 5
Hence, maximum value of x 4 = 5
i.e., we may take x 4 = 0 or 1 or 3 or 4 or 5
For x 4 = 0, x 1 + x 2 + x 3 = 20 has number of integral solutions
3
 1 
1 − x 
= coefficient of x 20 in (1 − x )–3
= coefficient of x 20 (1 + 3C1 x + 4C2 x 2 + …)
22 ! 22 × 21
= 22
C20 = =
20 ! 2 ! 2
= 231
For x 4 = 1, number of integral solutions of x 1 + x 2 + x 3 = 16 is given by the coefficient of x 16 in
(1 + x + x 3 + ... )3
3
 1  −3
=  = (1 − x )
1 − x 
= 1 + 3C1x + 4C2x 2 + … = 18C16
For x 4 = 2, number of integral solutions of x 1 + x 2 + x 3 = 12 is 14 C12 = 14C12
For x 4 = 3, number of integral solutions of x 1 + x 2 + x 3 = 8 is 10C8.
For x 4 = 4, number of integral solutions of x 1 + x 2 + x 3 = 4 is 16C4 .
For x 4 = 5, number of integral solutions of
x 1 + x 2 + x 3 = 0 is 1 (for x 1 ≥ 0, x 2 ≥ 0, x 3 ≥ 0)
Hence, total number of integral solutions of
x 1 + x 2 + x 3 + 4x 4 = 20 with x 2 ≥ 0
is 22
C20 + 18
C16 + 14
C12 + 10
C8 + 6C4 + 1
22 ⋅ 21 18 ⋅ 17 14 ⋅ 13 10 ⋅ 9 6 ⋅ 5
= + + + + +1
2 2 2 2 2
= 231 + 153 + 91 + 45 + 15 + 1
= 536
www.pdfworld.in
www.pdfworld.in
Example 50. How many integers between 1 and 1000000 have the sum of their digits equal to 18?
Solution Integers between 1 and 1000000 may be of 1 digits or of 2 digits or of 3 digits or of 4 digits or of
5 digits or of 6 digits and
Let the digits be x 1 , x 2 , x 3 , x 4 , x 5 , x 6
Then x 1 + x 2 + x 3 + x 4 + x 5 + x 6 = 18 …(i)
with 0 ≤ x 2 ≤ 9 for i = 1, 2, 3, 4, 5, 6
Hence, we have to find the number of integral solutions of Eq. (i) with the digits 0, 1, 2, …, 9.
= coefficient of x 18 in (1 + x + x 2 + ... + x 9 )6
6
 1 − x 10 
 1−x 
= coefficient of x 18 in (1 − x 10 )6 (1 − x )−6
= coefficient of x 18 in (1 − 6C1x 10 + 6C2x 20 + ... ) (1 + 6C1x + 7C2x 2 + ... )
23 ! 13 !
= 23
C18 − 6C1 × C8 =
13
− 6⋅
18! 5 ! 8! 5 !
23 ⋅ 22 ⋅ 21 ⋅ 20 ⋅ 19 ⋅ 18! 13 ⋅ 12 ⋅ 11 ⋅ 10 ⋅ 9 ⋅ 8!
= − 6⋅
5 ⋅ 4 ⋅ 3 ⋅ 2 ⋅ 1 ⋅ 18! 2 ⋅ 3 ⋅ 4 ⋅ 5 ⋅ 8!
= 23 × 11 × 7 × 19 − 6 × 13 × 11 × 9
= 33649 − 6 × 1287
= 33649 − 7722
= 25927
Example 51. There are m seats in the first row of a theature, of which n are to be occupied. Find the
number of ways of arranging n persons so that
(i) no two person sit side by side (ii) each person has exactly one neighbour.
Solution (i) Let P1 , P2 , ... , Pn be the n persons and S1 , S 2 , Sm be the m seats.
Let the n persons P1 , P2 ,... , Pn be seated as
... P1 ... P2 ... P3 ... Pn ...
n1 n2 n3 nn nn + 1
Let n1 be the number of empty seats to the left of P1 and nn +1 be the number of seats to the right of Pn , let
n2 be the number of seats between P1 and P2, n3 between P2 , P3... nn between Pn and Pn +1, then
n1 , nn + 1 ≥ 0 and n2 , n3 , ... , nn ≥ 0
Then n1 + n2 + ...+ nn = m − n
Now, solution if (i), given by coefficient of
x m −n (1 + x + x 2 + ... ) (x + x 2 + x 3 + ... )n −1 (1 + x + x 2 + ... )
= coefficient of x m − n in (1 + x + x 2 + ... )2 (x + x 2 + x 3 + ... ) n −1
= coefficient of x m − n in x n −1 (1 + x + x 2 + ... )2 (1 + x + x 2 + ... ) n −1
= coefficient of x m − 2n + 1 in (1 + x + x 2 + ... ) n + 1
= coefficient of x m − 2n + 1 in (1 + x + x 2 + ... ) n + 1
n +1
 1 
= coefficient of x m − 2n + 1 in  
1 − x 
= coefficient of x m − 2n + 1 in (1 − x )− (n + 1)
www.pdfworld.in
www.pdfworld.in
Combinatorics 439
= coefficient of x m − 2n + 1 in (1 + n +1
C1x + n+2
C2x 2 + ... )
= m −n + 1Cm − 2n + 1 = m −n + 1
Cn
Now, since n persons can be permuted in Pn = n ! ways, Hence, required number of ways in the case
n
m!− n + 1
= m − n + 1Cn n
Pn = n!
n ! m ! − 2n + 1
= m − n + 1Pn
(ii) It is important to notice that no such arrangement exist when n is odd. Hence, we suppose that n is
even. Let n = 2k , k being a +ve integer. Let x 0 denote the number of empty seats to the left of first pair
and x i (1 ≤ i ≤ k − 1), the number of empty seats to the right of mth pair. Here, x 0 > 0,x 1 ≥ 0 ; x i ≥ 0 , x 2 ≥ 1
(1 ≤ i ≤ k − 1) and n0 + n1 + n2 + ...+ nk = m − 2k …
Now, as in part (i) number of integers solution of (ii) is m − 2k + 1.
Also, n person can be permuted in n Pn = n ! ways. Hence, in the case required number of ways are
m − 2k + 1 m ! − 2k + 1
Cm ⋅ 2k
P2k = ⋅ 2k !
k ! (m ! − 3k + 1)
2k ! (m ! − 2k + 1)
=
k ! (m ! − 3k + 1)
m − 2k + 1
= 2k
Pk Pk = n Pn / 2 m −n + 1Pn / 2
Example 52. ‘A’ is a set consisting of n elements A subset ‘P’ of ‘A’ is chosen. The set A is reconstructed by
replacing the elements of P. A subset Q of A is chosen. Find the number of ways of choosing P and Q solution
so that P ∩ Q contains exactly one element.
Solution We choose one element out of n elements of A is n C1 = n ways. Then (n − 1) element remains and
these (n − 1) elements should not be in P ∩ Q in 3 ways.
(i) ar ∈ P , ar ∉ Q (ii)ar ∉P , ar ∈ Q (iii)ar ∉P , ar ∉ Q
Hence, remaining (n − 1) elements are not in P ∩ Q in 3n −1 ways.
Hence, number of ways of choosing the subset P and Q.
So that P ∩ Q contains exactly one element is n ⋅ 3n −1 ways.
Example 53. A is a set containing n elements. A subset P of A is chosen. Then the set A is reconstructed by
replacing the elements of P1. Then a subset P2 of A is chosen and again the set A is reconstructed by the
elements of P2. In this way m ( >1) subsets P1 , P2 , Pm of A are chosen. Find the number of ways of choosing
P1 , P2 ,... , Pm so that
(i) P1 ∩ P2 ∩ ... ∩ Pm = φ (ii) P1 ∪ P2 ∪ P3 ∪ ... ∪ Pn = A .
Solution Let A ⊃ {a1 , a2 , a3 ,... , an }. In course of choosing P1 , P2 ,... , Pn for any element ar of A’, we have two
cases either ar ∈ Pi or ar ∉Pi (1 ≤ i ≤ m )
Hence, total number of ways in which ar may be in Pi ′ are 2m . Also, out these 2m ways there is only one
ways in which ar ∉Pi for i = 1, 2,... , m which does not just the condition P1 ∩ P2 ∩ Pm = φ
Thus, for any element ar ∈ A ar ∉ P1 ∩ P2 ∩…∩ Pm in 2n − 1 ways. Hence, total number of ways is
(2m − 1)n .
Also, there is only one way i.e., ar ∉Pi which does not suit P1 ∪ P2 ∪...∪Pm = A . Hence, ai P1 ∪ P2 ∪... ∪Pm ,
in 2n − 1 ways. Also, since these n elements in the set A and hence the number of ways in which
P1 ∪ P2 ∪ ... ∪ Pm = A is ( 2m − 1)n .
www.pdfworld.in
www.pdfworld.in
Example 54. If a city has m parallel roads running East-West and n parallel roads running North-South.
How many rectangles are formed with sides along these roads? If the distance between every consecutive
pair of parallel roads is the same, how many shortest possible routes are there to go from one corner of the
city to its diagonally opposite corner?
Solution A rectangle is formed by two lines from E − W and two lines from N − S but two lines out of m
lines running E − W can be selected in m C2 ways and 2 lines out of n running N − W can be selected in n C2
ways. Hence, number of rectangles thus formed = m C2 ⋅ n C2
mn (m − 1)(n − 1)
=
4
1 2 3 4 n –1 n
1
m–1
Now, if distance between two consecutive parallel roads is (= 1 unit). Now, total distance a man has to
travel to go from one corner to a diagonally opposite corner is (m + n − 2) units. Of these (m + n − 2)
units, he has travel (m − 1) units in one direction and (n − 1) units in the other direction.
Hence, the number of ways of arranging the steps
= m + n − 2Cm −1 n −1Cn −1 = m + n − 2Cm −1
Example 55. The sides AB , BC , CA of a ∆ ABC have 3, 4, 5 interior points respectively on them. Find the
number of triangle formed by any three of the 12 points as vertices.
Solution Here, total number of points = 3 + 4 + 5 = 12. If no any three out of 12 points are collinear thus,
total number of triangles formed by taking any 3 points out of 12 points vertices = 12
C3. Thus, 12C3 include
the case of selection of 3 points from the 3 points on a side say AB.
A
B C
But no triangle is formed by taking 3 points lying on the side AB.
Hence, we subtract 3C3 from 12C3. Similarly, no triangle is formed with the 4 points lying on the other
side BC and no triangle is formed with 5 points lying on the side AC. Hence, we also subtract 4 C3 and
12
C3.
www.pdfworld.in
www.pdfworld.in
Combinatorics 441
Finally required number of triangles

= 12C3 − (3C3 + 4C3 + 5C3 ) = 12C3 − (1 + 4 + 10)
12 ! 12 ⋅ 11 ⋅ 10
= − 15 = − 15 = 195
3! 9! 3⋅2
Example 56. If n points in a plane be joined in all possible ways by indefinite straight lines and if no two of
straight lines be coincident or parallel and no three pass through the same point with the exception of the n
points. Find the number of distinct point of intersection.
Solution Out of n points, number of different straight lines formed = x (say)
Now. since no three straight lines pass through the same point and no two straight lines are parallel. We
see that each pair of lines from the n C2 lines intersect at one point.
Hence, number of points of intersection = xC2. Now, let P1 , P2 ,... , Pn be the n given points. It is here
important to notice that through the point P1 (n − 1) lines by joining P1 to other (n − 1) points, pass. But
any two of these (n − 1) lines give P1 as point of intersection.
Hence, we see the point P1 occurs n −1C2 times. Similarly, each point P2 , P3 ,... , Pn occurs n ⋅ n −1C2 times
i.e., n given points occur n ⋅ n −1C2 times as point of intersection.
Now, number of points of intersection except n given points
= xC2 − n ⋅ n − 1C2
1 (n − 1) (n − 2)
= x ⋅ (x − 1) − n ⋅
2 2
1 n (n − 1)  n ⋅ (n − 1) n ⋅ (n − 1) (n − 1)
= ⋅ − 1 −
2 2  2  2
1 n ⋅ (n − 1) (n − 2)
= n ⋅ (n − 1) (n 2 − n − 2) −
8 2
n ⋅ (n − 1) 2
= [n − n − 2 − 4(n − 2)]
8
n ⋅ (n − 1) 2
= (n − 5n + 6)
8
Hence, total number of points of intersection including the n given points
n ⋅ (n − 1) (n 2 − 5n + 6)
=n +
8
1
= n ⋅ (n3 − 6n 2 + 11 n + 2)
8
Example 57. Find the number of rectangles that you can have on a chess-board.
Solution We observe that there are 9 horizontal and 9 vertical lines on a chess-board such that each
horizontal line intersect each vertical line. A rectangle is formed by two horizontal and two vertical lines.
But if we select 2 horizontal lines out of 9 horizontal lines in 9C2 ways and we can select 2 vertical lines
out of 9 vertical lines in 9C2 ways.
Hence, total number of rectangles on a chess-board = 9C2 × 9C2 = 36 × 36 = 1296.
Example 58. The straight lines l1 , l 2 and l3 are parallel and lie on the same plane. A total number m points
are taken on l1 ; n points on l 2, k points on l3. Find the maximum number of triangles formed with the vertices
at these points.
www.pdfworld.in
www.pdfworld.in
Solution Total number of points = m + n + k

m +n + k
Now, number of triangles formed by the points m + n + k = C3
m
But as m points be on the same straight line C3 give no triangle, n points be on the same straight line so
that n C3 give no triangle and k points lie on the same straight line so that k C3 also give no triangle.
m +n + k
Hence, required number of triangles = C3 − m
C3 − n C3 − kC3.
Example 59. In how many ways 4 square are can be chosen on a chess-board, such that all the squares lie
in a diagonal line.
Solution Let us consider the ∆ABC. Number of ways in which 4 selected squares are along the lines
A 4C4 , A3C3 , A 2C2 , A1C1and AC are
A
A1
A2
A3
A4
B C4 C3 C2 C1 C
4 5 6 7 8
C4 , C4 , C4 , C4 and C4 respectively.
Similarly, in ∆ACB, number of ways in which 4 selected squares are along the diagonal line parallel to
AC are 4 C4 ,5 C4 , 6C4 , 7C4 and 8C4 .
But 8C4 triangles occur only once.
Hence, the total number of ways in which the 4 selected squares are in a diagonal line parallel to AC are
2 (4 C4 + 5C4 + 6C4 + 7C4 ) + 8C4
Also, same is the case of selecting 4 squares in a diagonal line parallel to BD.
Hence, the number of ways of selecting 4 squares on a chess-board. Such that the 4 squares are in a
diagonal line
= 2 [2 (4 C4 + 5C4 + 6C4 + 7C4 ) + 8C4 ]
Example 60. Find the number of positive divisor of n = p1k1 ⋅ p2k2 ⋅ ... ⋅ prkr , where p1 , p2 ,... , pr are distinct
prime numbers and k1 , k2 ,... , k r are positive integers.
Solution A divisor d of n is of the form d = p1l1 ⋅ pl22 ⋅...⋅prlr
where 0 ≤ li ≤ ki for i = 1, 2, …, r.
Associate each divisor of n with an r tuple (l1 , l 2 ,... , lr ), such that 0≤ li ≤ ki . Therefore, the numbers of
divisors is the same as the number of r tuples (l1 , l 2 ,... , lr ), 0≤ li ≤ ki , i = 1, 2,... , r .
Since, l1 can have k 1 + 1 possible values 0, 1, 2,... ,k 1 and l 2 can have k 2 + 1 values 0, 1, 2,... ,k 2 and l r can
have k r + 1 values. The number of r triples (l1 , l 2 ,... , lr ) is
(k 1 + 1) × (k2 + 1) × (k3 + 1) × ... × (kr + 1)
www.pdfworld.in
www.pdfworld.in
Combinatorics 443
r
or Π (ki + 1)
i =1
That is total number of divisors of

k k
n = p1 1 , p 2 2 ,... prkr
r
= (k1 + 1) (k2 + 1)... (kr + 1) or Π (ki + 1)
i =1
Example 61. Prove that, there are 2 (2n −1 − 1) ways of dealing n cards to two persons. (The players may
receive unequal number of cards).
Solution Let us number of the cards, for the moment. Let us accept the case, where all the cards, go to
one of the two players, also.
With just two cards we have the possibilities.
AA, AB, BA, BB …(i)
Here, AA means A gets 1 card and also card 2,
AB means A gets card 1 and B gets card 2,
BA means B gets card 1 and A gets card 2,
BB means B gets card 1 and also card 2.
Thus, for two cards we have 4 possibilities. Here, for three cards
AAA, ABA, BAA, AAB, ABB, BAB, BBB …(ii)
i.e., for three cards there are 23 = 8 possibilities. Here, if the third card goes to A, then in Eq. (i) annex A at
the end, thus getting AAA, ABA, BAA, BBA,. If it goes to B then in Eq. (i) annex A at the end, which gives
AAB, ABB, BAB, BBB.
Thus, the possibilities doubled, when a new card (third card) is included.
In fact just with one card it may either go to A or B, giving AB.
By annexing the second card, it may give
AA, BA, giving Eq. (i)
AB, BB
Thus, every new card doubles the existing number of possibilities of distributing the cards.
Hence, the number of possibilities with n cards is 2n . But this include the distribution where one of
them gets all the cards and the other none.
So, total number of possibilities is 2n − 2 = 2 (2n −1 − 1)
Example 62. In a plane, a set of 8 parallel lines intersects a set of n other parallel lines, giving rise to 420
parallelograms (many of them over lap with one another). find the value of n.
Solution If two lines which are parallel to one another (in one direction) intersect another two lines,
which are parallel, we get one parallelograms. Thus, we can choose C (8, 2) pair of parallel line in one
direction and the number of parallel lines intersecting there will be C (n, 2) pair.
So, the number of parallelograms thus obtained is C (n , 2) × C (8, 2) = 420
[With three sets of parallel lines, we have 9 parallelograms (verify) i.e., 3C2 × 3C2 = 9 ]
n (n − 1) 8 × 7
× = 420
2×1 1×2
n (n − 1) = 30
or n=6
or n = − 5 (which is not admissible).
Thus, n = 6 is the solution.
www.pdfworld.in
www.pdfworld.in
Example 63. Let A be an 2n element set, where n ≥ 1. Find the number of pairing of A.
Solution A pair is set of 2 elements. These are thus finally n pairs.
 2n 
Now, the first pair can be selected in   ways
2
(2n ) !
i.e., ways.
(2n − 2)! 21
 2n − 2
The second pair can be selected from the remaining (2n − 2) elements in   ways, i.e.,
 2 
(2n − 2)!  2 2!
and so on up to   =
(2n − 4)! × 2 !  2 0 ! × 2 !
Number of pairing of A is, therefore,
2n × (2n − 1) (2n − 2)(2n − 3) (2n − 4)(2n − 5) 2×1
= × × × .. ×
2 2 2 5
n!
2n × 2(n − 1) × 2(n − 2) × ... × (2n − 1) (2n − 3) × (2n − 5) × ... × 1
=
2n × n !
2n × n (n − 1) × (n − 2) × ... × 1 × (2n − 1) (2n − 3)(2n − 5) × ... × 1
=
2n × n !
2n × n ! × (2n − 1)(2n − 3)(2n − 5) × ... × 1
=
2n × n !
= (2n − 1)(2n − 3)(2n − 5) × ... × 1 ways
Example 64. Find the number of 6 digits natural numbers, where each digit appears at least twice.
Solution We consider number like 222222 or 233200 but not 212222, since the digit 1 occurs only once.
The set of all such 6 digits can be divided into the following classes
S1 = the set of all 6 digit numbers where a single digit is repeated 6 times.
n (S1 ) = 9, since ‘0’ cannot be significant number when all its digits are zero
Let S 2 be the set of all 6 numbers, made up of three distinct digits.
Here, we should have two cases S 2(a) one with the exclusion of zero as a digit and other S 2(b) with the
inclusion of zero as a digit.
S 2 (a) the number of ways, three digits could be chosen from 1, 2,... , 9 is 9C3. Each of these three digits
occur twice. So, the number of 6 digit numbers in the case is
6! 9 × 8 × 7 720
= 9C3 × = ×
2! × 2! × 2! 1 × 2 × 3 8
= 9 × 8 × 7 × 15 = 7560
S 2 (b), the three digits used include one zero, implying, we have to choose the other two digits from the
non-zero digits.
9×8
This could be done in 9C2 = = 36 . Since, zero cannot be in the leading digit. So, let us fix one of the
1×2
digit (non-zero) in the extreme left . Then the other five digits are made up of 2 zeroes, 2 fixed non- zero
number and the another non-zero number, one of which is put in the extreme left. In this case the
5!
number of six digit numbers that could be formed is × 2 ( Since, from either of the pairs of
2! × 2! × 1!
fixed non-zero numbers, one can occupy the extreme digits = 60
www.pdfworld.in
www.pdfworld.in
Combinatorics 445
So, the total numbers in this case = 36 × 60 = 2160

∴ n (S 2 ) = n (S 2 a ) + n (S 2 b ) = 7560 + 2160 = 9720.
Now, let S3 be the set of 6 digit numbers, whose digits are made up of two distinct digit, each of which
occurs thrice. Here again, there are two cases : S3 (a ) excluding the digit S3 (b ) including the digit zero.
S3 (a ) is the set of six digit numbers whose digits are made up of two non-zero digits, each occurring
thrice.
6!
∴ n [S3 (a )] = 9C2 × = 36 × 20 = 720
3!3!
S3 (b ) consist of 6 digits numbers whose digits are made up of three zeroes and one of non-zero digit,
occurring thrice. If you fix one of the nine non-zero digit, use that digit in the extreme left. This digit
should be used thrice. So, in the remaining 5 digits, this fixed non-zero digit is use twice and the digit
zero occurs thrice.
5!
So, the number of 6 digit numbers formed in this case is 9 × = 90
3! × 2!
∴ n (S3 ) = n (S3 (a )) + n (S3 (b )) = 720 + 90 = 810
Now, let us take S 4 , the case where the six digit number consist of exactly two digits, one of which
occurs twice and the other four times.
Here again, there are two cases : S 4 (a ) excluding zero and S 4 (b ) including zero.
6!
If a and b are the two non-zero numbers ‘a’ used twice and ‘b’ four times, then we get and when
2! × 4!
6!
‘a’ used four times, b twice, we again get
2! × 4!
So, when 2 of the nine non-zero digits are used to form the six digit number in this case, the total
number got is
6!
9
C2 × 2 × = 36 × 5 × 6 = 1080
4! × 2!
Thus, n [S 4 (a ) ] = 1080
for counting the numbers in S 4 (b )
In this case we use 4-zeros and a non-zero number twice 2-zeros and a non-zero number four times. In
the former case, assuming the one of the fixed non-zero digit occupying extremely left, we get the other
five digits consisting of 4-zeroes and one non-zero number.
9 × 5!
This result in = 45 six digit number.
4! × 1!
9 × 5!
When we use the fixed non-zero digit 4 times and use zero twice, then we get = 90 six digit
3! × 2!
numbers, as the fixed number occupies the extreme left and for the remaining 3 times it occupies 3 of
the remaining digits, other digits being occupied by the two zeroes.
So, n (S 4 ) = n [ S 4 (a )] + n [ S 4 (b )] = 1080 + 45 + 90 = 1215
Hence, the total number of six digit number satisfying the given condition
= n (S1 ) + n (S 2 ) + n (S3 ) + n (S 4 ) = 9 + 720 + 810 + 1215 = 2754.
Example 65. Let S = {1, 2, 3,…(n + 1) }, where n ≤ 2 and let T = {(x , y , z ) : x , y , z ∈S , x < z , y < z }. By counting
n  n + 1  n + 1
the members of T in two different ways, prove that Σ K 2 =   +2  .
K =1  2   3 
Solution T can be written as T = T ∪T2 ,T1 = { (x , x , z ) : x , z ∈S , x < z} and T2= {(x , y , z ) : x , y , z ∈S , x + y < z}.
The number of elements in T1is the same as choosing two elements from the sets where n (S ) = (n + 1),
www.pdfworld.in
www.pdfworld.in
 n + 1
i.e., n (T1 ) =   , (as every subset of two elements, the large element will be x and the smaller will be x
 2 
and y).
 n + 1
InT2, we have 2   elements, after choosing 3 elements from the set S, two of the smaller elements
 3 
will be x and y and they may be either taken as (x, y, z) or as (y, x, z) or in other words, every three
elements subset of S1say {a, b, c}, the greatest is z, and the other two can be placed in two different ways
in the first two positions.
 n + 1  n + 1
∴ n (T ) or (|T | ) =   + 2 
 2   2 
n +1
T can also be considered as U Si , where Si = {(x , y , i ): x , y < i , x , y ∈S }. All these sets are pairwise disjoint
i =2
as for different i, we get different ordered triplets (x , y , i ).

Now, in Si , the first two components of (x , y , i ), namely (x , y ) can be any element from the set
1, 2, 3,... , i − 1.
x and y can be any member from 1, 2, 3,... , (i − 1) equal or distinct.
∴ The number of ways of selecting (x , y ), x , y ∈ {1, 2, 3,... , (i − 1)} is (i − 1)2.
Thus, n (Si ) for each i is (i − 1)2 , i ≥ 2. For example n (S 2 ) = 1, n (S3 ) = 22 = 4 and so on.
n + 1  n + 1
Now, n (T ) = n  ∪ Si  = Σ n (Si )
i = 2  i = 2
(because all Si ’s are pairwise disjoint)
n +1 n
= Σ (i − 1)2 = Σ i 2
i =2 i =1
 n + 1  n + 1 n
 + 2  = Σ k
2
and hence, 
 2   3  k =1
Example 66. Among the integers 1, 2,…., 200 if any 101 integers are chosen, then show that there are 2
among the chosen integers, such that one is divisible by other.
Solution Let each of 101 integers chosen from 1, 2, …, 200 be factorized in the form mi = 2k (i ) ⋅ ai for
l = 1, 2, …, 101, where ai is an odd number and k (i) is a non-negative integers. Each ai is one of the 100 odd
numbers integers only. Hence by PP1, among the chosen 101 numbers mi at least 2 have equal odd parts.
Thus, let mt = 2k (t ) ⋅ at and m s = 2k ( s ) ⋅ a s with at = a s . Then if k (t ) < k (s ), then mt divides m s .
Example 67. Show that every sequence a1 , a2 ,…,amn +1 of mn + 1 distinct real numbers contains either an
increasing subsequence of length m + 1 or decreasing subsequence of length n + 1.
Solution Let x i denote the length of the longest increasing subsequence beginning at ai and let yi be the
length of longest decreasing subsequence beginning at ai . Consider the mn + 1 ordered pairs (x i , yi ).
1 ≤ i ≤ mn + 1. We will prove that all these mn + 1 ordered pairs are distinct.
For any distinct integers i , j such that 1 ≤ i , j ≤ mn + 1 we are given that ai ≠ a j , since the terms of the
sequence are distinct. So first let ai < a j . Then it follows that x i > x j since we can append ai at the
beginning of every increasing subsequence beginning with a j and obtain a longer subsequence of the
same type. Similarly, if ai > a j then yi > y j . Hence, if i ≠ j , then (x i , yi ) and (x j , y j ) are distinct ordered
pairs.
www.pdfworld.in
www.pdfworld.in
Combinatorics 447
Now, suppose that stated result is not true. Then for each i, we must have 1 ≤ x i ≤ m and 1 ≤ yi ≤ n, thus,
there are only m possible different values of x i and only n possible different value of yi . Hence, by the
multiplication principle only mn of the mn + 1 ordered pair (x i , yi ) can be distinct. Hence, by PP1 at least
2 of the mn + 1 ordered pairs must be equal, a contradiction. Hence, result is true.
Example 68. Suppose the numbers 1 to 10 are randomly positioned around a circle. Show that m 2 of some
set of 3 consecutive numbers be at least 17.
Solution Let the number be a1 ,. .. , a10. Note that these are the numbers1, 2, …, 10 in some order. Now,
there are 10 triples consisting of 3 consecutive numbers namely.
(a1 , a 2 , a3 ), (a 2 , a3 , a 4 ),…, (a10 , a1 , a 2 )
Note that each ai occurs in exactly 3 triples. Hence, average sum per triple is
3(a1 + ... + a10 ) 3(1 + ... + 10)
=
10 10
3(10) (11) / 2
= = 165
.
10
Hence, by PP3, the sum of the numbers in at least one triple must be ≥ 165
. . Hence, that sum must be
≥ 17 since it is an integer.
Example 69. 13 persons have first names, Bapu, Chandru, Damu and last names Kale, Late, Mate and
Natu. Show that at least 2 persons have the same first and last names.
Solution By the multiplication principle, these are 4 × 3 = 12 possible names. Now, regard the 13 persons
as 13 objects and 12 names as 12 boxes. Then by PP1, it follows that at least 2 objects are in the same box
i.e., at least 2 persons have the same name.
Example 70. 18 persons have first name Ekta, Ganesh and Hari and last names Patil and Rathi. Show that
at least 3 persons have the same first and last names.
Solution By multiplication principle, there are 3 × 2 = 6 names, such as Ekta, Ganesh etc. Now, since
there are 18 persons and 18 = 3 × 6, it follows by PP2 that at least 3 persons have the same first and last
names.
Example 71. The members of a class of 27 pupils each go swimming on some of the days from Mon-Fri in a
certain week. If each pupil goes at least twice, show that there must be 2 pupils, who go swimming on exactly
the same days.
Solution The set {Mon, …, Fri} of 5 days has
 5  5   5  5
  +   +   +   = 1 + 5 + 10 + 10 = 26
 5  4  3  2
subsets each containing 2 or more days. Regard the 27 pupils as objects and these 26 subsets as ‘boxes’.
Then by PP1, there must be at least one box containing at least 2 pupils i.e., at least 2 pupils must go
swimming on the same days.
Example 72. Let A be any set of 20 distinct integers chosen from AP 1, 4, 7, …, 100. Prove that there must
be 2 distinct integers in A, whose sum is 104.
Solution The given AP 1, 4, …, 100 has nth term 3n − 2 and hence contains 34 terms. Arrange these terms
in 18 boxes as follows.
1 52 4 100 7 97 49 55
www.pdfworld.in
www.pdfworld.in
Note that the sum of the numbers in each of the last 16 boxes is 104. Then by PP1, if 20 numbers are
taken from these 18 boxes, then at least 2 of the numbers must come from the same box (and that box
must be one of the last 16 boxes since the first 2 contain only 1 number each) and this sum is 104 as
required.
Example 73. Let S = {3, 7, 11, ... , 103}. How many elements must we select from S to ensure that there will
be at least 2 distinct integers among them whose sum is 110.
Solution The set S contains 26 elements which form an AP. These numbers can be put in following 14
3 55 7 103 11 99 51 59 boxes where the sum of the numbers in each of the last 12 boxes
is 110. Hence, by PP1 is enough to select 15 elements from S.
Example 74. If 11 integers are selected from {1, 2, ... , 100} , prove that there are at least 2 say x, y such that
0 < | x − y | < 1.
Solution Consider the integral part [t] of the real number t. Then t − [t ] = f is the fractional part of t and
0 ≤ f < 1. Now, since 1 ≤ x ≤ 100, we have 1 ≤ x ≤ 10, and so 1 ≤ [ x ] ≤ 10, thus for elements 1, 2,…,10. Hence,
if 11 numbers are taken from S, then by PP1 at least 2 of them say x, y must be such that x and y have the
same integral part say i.
So, if x = i + f1 and y = i + f2 ,
0 ≤ f1 , f2 < 1, then 0 < | x − y |
= | f1 − f2| < 1
Example 75. Let n be an odd +ve integer. If i1 , i 2 , …, in is a permutation of 1, 2, … , n. Prove that

(1 − i1 )(2 − i 2 ) … (n − in ) is an even integer.
Solution Since, n is odd, let n = 2m + 1, where m is a non-negative integer. Then set S = {1, 2, ... , n}
contains m + 1 odd numbers namely 1, 3,…, 2m + 1 but only m even numbers, namely 2, 4,…, 2m. The same
is true of the permutation i1 , i 2 ,…, in of S. Consider the m + 1 numbers 1 − i1 , 3 − i3 , …, n − in which are of the
form r − ir , where r is odd. Since, i s is even for only m values of S, by PP1, one of the m + 1 numbers i1 , i 2 , …, in
say it is odd, where t is also odd. Hence, t − it is even and so the product
(1 − i1 ) ⋅ (2 − i 2 ) … (n − in ) is even.
−1
Example 76. Let n ≥ 3 be an odd number. Show that there is a number in the set {21 − 1, 22 − 1, …, 2n − 1},
which is divisible by n.
−1
Solution Consider the n numbers 20 , 21 , …, 2n since n is odd, none of these numbers is divisible by n
and so modulo n, they can leave only n − 1 different remainders 1, 2, …, n − 1. Hence, by PP1, 2 of them say
2r and 2s , 0 ≤ s < r ≤ n − 1 must leave the remainder modulo n, so that n divider2r − 2s = 2s (2r − s − 1). Hence,
−s
n divides2r − 1, since n is odd and so n and 2s are coprime. So, the result follows since 1 ≤ r − s ≤ n − 1.
Example 77. Given m integers a1 , a2 , …, am , show that there exist integers k, s with 0 ≤ k < s ≤ m such that
ak + 1 + ak + 2 + ... + a s is divisible by m.
Solution Consider the sequence a1 , a1 + a2, …, a1 + a2 + … + am . If anyone of these m sums is divisible
by m, then we are through. Otherwise suppose that none of them is divisible by m so each leaves a non-zero
remainder 1, 2,…,(m − 1). Since, there are m sums and (m − 1) possible values of the remainders by the
pigeonhole principle 2 of the sums leave the same remainder after division by m. So, let
a1 + a 2 + ... + a k = bm + r
and a1 + a 2 + ... + a s = cm + r
www.pdfworld.in
www.pdfworld.in
Combinatorics 449
This gives (if k < s ), ak + 1 + ak + 2 + .... + a s = m (c − b )

Thus, m divides ak + 1 + ak + 2 + ... + a s
Example 78. Suppose numbers 1 to 20 are placed in any order around a circle. Show that the sum of some
3 consecutive numbers must be at least 32.
Solution Let a1 , …, a20 be the numbers placed around the circle. Since, the mean of the 20 sums of
3 consecutive numbers namely
a1 + a 2 + a3 , a 2 + a3 + a 4 , ... , a19 + a 20 + a1,
a 20 + a1 + a 2
1
is [3(a1 + a 2 + ... + a 20 )]
20
3(20)(21)
= = 315
.
2(20)
We see by the alternate form of PP that at least one of the sums must be ≥ 32.
Example 79. A storekeeper’s list consist of 115 items each marked “available” or “unavailable”. There are
60 available items. Show that there are at least 2 available items in the list exactly 4 items apart.
Solution Let the position of the available items be a1 , a2 , …, a60. Since a60 ≤ 115, we see that 120 numbers
a1 < a 2 < ... < a 60 and a1 + 4 < a 2 + 4 < … < a 60 + 4 lie between 1 and 119. Hence, by PP1, 2 of these numbers
must be equal. But the numbers in the first are all distinct. Similarly, the numbers in the second row are all
distinct. Hence, the some number in the first row must be equal to a number in the second row i.e., for
some i, j we must have ai = a j + 4, so that ai − a j = 4 as required.
Example 80. Prove that, if in a group of 6 persons, each pair is either of mutual friends or mutual enemies,
then there are either three mutual friends or 3 mutual enemies. Also, show that the result is not true in case of
a group of 5 persons.
Solution Consider a fixed person A of the other five. By PP, there are either 3 who are friends of A or 3
who are not. In the first case, the 3 friends of A are either mutual enemies or 2 of them are friends and form
a triplet of friends with A. The other case is similar.
Now, consider a group of 5 persons and suppose that exactly the following are pairs of friends AB, BC,
CD, DE, EA. Then it is easy to see that no three are mutual friends or mutual enemies.
Example 81. Let (x i , yi ), 1 ≤ i ≤ 5 be set of 5 distinct points with integer coordinates in x-y plane. Show that
the mid-point of the line joining at least one pair of these points has integer coordinates.
Solution Since an integer must be either even or odd. Every point (a, b) with integer coordinates must be
put in one of the 4 pigeonholes (even, even), (even, odd), (odd, even) and (odd, odd). Hence, 2 of the given 5
points [say A (x 1 , y1 ) and B (x 2 , y 2 )] must lie in the same pigeon hole, so that their x coordinate must have
same parity (i.e., they are either both even or both odd) and their y coordinates must have same parity.
Hence, x 1 + x 2 and y1 + y 2 are even. Thus, (x 1 + x 2 ) / 2 and (y1 + y 2 ) / 2 are both integers coordinates.
Example 82. Prove that when a rational numbera / b in lowest term is expressed as a decimal, the decimal
must either terminates or recur.
Solution Divide 10a by b to get, 10a = x 1b + r1, 0 ≤ r1 < b. Next divide 10r by b to get, 10r1 = x 2b + r2 ,
0 ≤ r2 < b. Divide 10r2 by b to get 10r2 = x 3b + r3 , 0 ≤ r3 < b and so on. Then we get a / b = 0, x 1x 2x 3… .
www.pdfworld.in
www.pdfworld.in
Now, if the decimal does not terminate, then we must obtain non-zero remainder at each stage. Since,
there are only b − 1 possible different non-zero remainders. By PP1 some remainder must be repeated
after at most b steps. Hence, the expansion will recur from this point onwards.
Example 83. Prove that, if given a set of any 7 distinct integers, there must exist 2 integers in this set,
whose sum or difference is a multiple of 10.
Solution Consider the 6 boxes labelled as (0, 0), (1, 9), (2, 8), (3, 7), (4, 6), (5, 5). Let A be the set of 7
distinct integers.
We know that every integer is congruent to 0 or 1 or 2 or…or 9 module 10. If integer from A is
congruent to k or 10 − k module 10, then we shall put that integers in the box lablled (k, 10 − k). Since,
there are 7 integers and only 6 boxes. By the Pigeonhole principle, at least one box say (k, 10 − k)
contains 2 integers say x and y. If x and y both are congruent to k modulo 10, then x − y ≡ 0 (mod 10).
∴ 10 divides x − y
If x and y both are congruent to 10 − k mod 10 then again x − y = 0 (mod 10)
∴ 10 divides x − y
If one of x and y is congruent to k and the other is congruent to 10 − k mod 10, then
x + y ≡ 10 (mod 10)
i.e., x + y ≡ 0 (mod 10)
∴ 10 divides x + y
Thus, in any case either x − y or x + y is divisible by 10.
Example 84. Show that any subset of 8 distinct integers between 1 and 14 contains a pair of integers k
and l, such that either k divides l or l divides k.
Solution 8 integers are chosen from 1 to 14. By taking the factor 2 as many times as possible from each
of these 8 integers, we can write them in the form 2α a; ‘a’ is an odd integer. Possible values of ‘a’ are 1, 3, 5,
7, 9, 11, 13 which are 7 in numbers.
Since, there are 8 integers and only 7 values of ‘a’ by Pigeonhole principle, there must be 2 integers k
and l having same value ‘a’.
Suppose k = 2α a and l = 2β a, if α < β, then k divides l. If β < α, then l divides k.
Example 85. Given m consecutive integers. Prove that there is one which is divisible by m.
Solution Consider m consecutive integers k + 1, k + 2, k + 3, ...., k + m. Assume that none of these is
divisible by m. Therefore, after dividing each of them by m, we shall get non-zero remainders 1, 2, 3,…, m – 1
(not necessary in this order).
Since, there are m integers and only m – 1 remainders. By Pigeonhole principle, at least 2 integers will
leave the same remainder.
Suppose, k + i and k + j leave the same remainder r. when divided by m, where i < j
∴ k + i = xm + r and k + j = ym + r
subtraction gives j − i = (y − x )m
This shows that m divides +ve integer j − i which is itself less than m.
But this is impossible. Therefore, our assumption must be wrong. Hence, one ofk + 1, k + 2, …, k + m
must be divide by m.
Example 86. The circumference of a wheel is divided into 36 sectors and the numbers 1, 2,… 36 are
assigned to them in arbitrary manner. Show that there are 3 consecutive sectors such that sum of their
assigned number is at least 56.
www.pdfworld.in
www.pdfworld.in
Combinatorics 451
Solution Let a1 , a2 , …, a36 be arbitrary assignment of numbers 1, 2, 3, …, 36 to the 36 sectors.

We group them into the collection of 3 consecutive and find 36 sums as below.
a36
ai a1
a2
S1 = a1 + a 2 + a3 , S 2 = a 2 + a3 + a 4 ,
S3 = a3 + a 4 + a5 ,…, S36 = a36 + a1 + a 2
The sum of these 36 numbers is 3 times the sum 1 + 2 + 3 + 4 +…+ 36 because each ai is counted
3 times.
∴ S1 + S 2 + ... + S36 = 3(1 + 2 + ... + 36) = 3(666) = 1998
1998
Sum of 36 +ve integers is 1998 and = 555
. This implies that at least one Si ≥ 56. Hence, there are 3
36
consecutive sectors, the sum of whose assigned numbers is at least 56.
Example 87. 8 composite integers are chosen from 1 to 360. Prove that the selection includes 2 integers
which are not relatively prime.
Solution Consider the first 7 prime numbers 2, 3, 5, 7, 11, 13, 17. The next prime number is 19
19 × 19 = 361 > 360 from this, we conclude that any composite number from 1– 360 must have one of the
above 7 prime number as a factor.
We write 8 composite numbers in the form xp, where p is smallest prime factor of that composite
number.
Now, p can take 7 values and there are 8 composite numbers. By Pigeonhole principle, there exist 2
composite numbers in the selection such that they have p as smallest prime factor.
GCD of these 2 integers is ≥ p ≥ 2.
So, they are not relatively prime.
Example 88. If 11 distinct integers are chosen from among 1, 2, 3, …, 20. Show that selection includes at
least one pair of integers which are relatively prime.
Solution The selection of 11 distinct integers from 1 − 20 must contain at least one odd integer. If the
selection includes 1, 3, 5, 7, 11, 13, 17, 19 then clearly the conclusion holds because these are prime
numbers.
So, we assume that above primes are absent in the selection. Hence, the selection must contain either 9
or 15 or both 9 and 15. If the selection includes 9 but not 15, then remaining 10 integers in the selection
are all even and 4 is one of them. Now, 4 and 9 are relatively prime and the conclusion holds. Suppose
the selection includes 15 but not 9. Then as before remaining 10 integers in the selection are all even
and 4 is one of them. Now, 4 and 15 are relatively prime and conclusion holds. Finally, suppose
selection includes 9 and 15 both. The remaining 8 integers in selection are all even. The set {2, 4, 6, 8,
10, 12, 14, 16, 18, 20} contains 5 integers which are multiples of 3 or 5 (factors of 9 and 15).
–π/2 –π/4 0 π/4
They are 6, 12, 18, 10, 20. Hence, 8 even integers required for our selection must contain at least
3 integers from 2, 4, 8, 14, 16. Each of them is relatively prime with 9 and 15 and the conclusion holds.
In any case, we see that there is a pair of integers which are relatively prime.
www.pdfworld.in
www.pdfworld.in
Example 89. Given any 5 distinct real numbers. Prove that there are two of them say x and y such that
0 < (x − y ) / (1 + xy ) .
Solution Here, we are using the property of tangent functions of trigonometry.
Given a real number a, we can find a unique real number A lying between − π / 2 and π / 2 i.e., lying in the
real interval (− π / 2, π / 2) such that tan A = a as the tangent function in the open interval (− π / 2, π / 2)
is continuous and strictly increasing and covers R completely. Therefore, corresponding to the five
given real numbers ai (i = 1, 2, ... , 5), we can find 5 distinct real numbers ai (i = 1, 2, .. , 5), lying between
− π / 2 and π / 2 such that tan Ai = ai .
Divide the open interval (− π / 2, π / 2) into 4 equal intervals, each of length π / 4. Now, by Pigeonhole
principle, at least 2 of the Ai ' s must lie in one of the 4 intervals. Suppose A k and Al with A k > Al lie in the
same interval, then
0 < A k − Al < π / 4
so tan 0 < tan (A k − Al ) < tan π / 4
[It is so because tan functions increases in the interval (− π / 2, π / 2)]
tan A k − tan Al
i.e., 0< <1
1 + tan A k tan Al
a k − al
0< <1
1 + a kal
There are 2 real numbers x = a k , y = al such that
x −y
0<
1 + xy
Example 90. Show that given 12 integers, there exists two of them whose difference is divisible by 11.
Solution Possible remainders, when any integer is divided by 11 are 0, 1, 2, …, 10. Treat these
remainders as ‘holes’ and the 12 integers as pigeons. By PHP, two of them should lie in the same hole, i.e.,
should leave the same remainder, when divided by 11. Hence, their difference is divisible by 11.
Example 91. Five points are marked at random in a square plate of length 2 units. Show that a pair of
them are apart by not more than 2 units.
Solution By taking the mid-points of the edges and joining the points on opposite edge, we get a grid of
4 units squares. By PHP two of the points lie in one of these grids and the maximum distance between these
points is 2 units (the length of the diagonal of the unit square).
Example 92. Prove that any set of 55 integers {x 1 , ... , x 55} such that 1 ≤ x 1 ≤ x 2 < .... < x 55 = 100, there will
be some two that differ by 9, some two that differ by 10, a pair that differ by 12 and a pair that differ by 13.
Solution Consider the sets
A1 = {1, 10}, A 2 = {2, 11},…, A 9 = {9, 18}
A10 = {19, 28}, A11 = {20, 29}, …, A18 = {27, 36},
A19 = {37, 46}, A 20 = {38, 47}, …,
A 27 = {45, 54} A 28 = {55, 64}, A 29 = {56, 65}, …, A36 = {63, 72},
A37 = {73, 82}, A38 = {74, 83}, …, A 45 = {81, 90},
A 46 = {91, 100}
A 47 = {92}, A 48 = {93}, A 49 = {94}
A50 = {95}, A51 = {96}, A52 = {97}
A53 = {98}, A54 = {99}
www.pdfworld.in
www.pdfworld.in
Combinatorics 453
When 55 numbers are chosen from {1, 2, ... , 100} some two of them come from the same set. Two such
numbers differ by 9. Thus, any set of 55 numbers will have a pair differing by 9. Proof is similar for 10,
12 and 13.
Example 93. A chess player plays at least one game of chess a day, but in order to avoid overstrain he
plays no more than 12 games a week. Prove that, in period of 77 days, there must be a period of several
consecutive days during which he plays exactly 20 games.
Solution Let us suppose that the chess player plays a1, games on Monday, a2 game during Monday and
Tuesday, a3 games during the first three days etc…., and finally a77 games during 77 days.
Consider a1 , a 2 … a77 , a1 + 20, a 2 + 20, …, a77 + 20. This is a sequence of 154 numbers, each of which
does not exceed 11 × 12 + 20 = 152 (the 77 days, period has 11 weeks and in each week at most 12
games are played). Consequently, at least two of these numbers are equal to each other. Since ai ≠ a j for
i , j , we must have k, l such that a k = ai + 20. Thus, a k − ai = 20 and it follows that during k − l days from
(l + 1)th day to the kth day inclusive, the player plays exactly 20 games.
Example 94. Show that there exist two powers of 1999 whose difference is divisible by 1998.
Solution Any number, when divided by 1998 leaves one of 0, 1, … ,1997 as remainder. If we consider
19991 , ... , 19991999 the first 1999 powers of 1999, they all cannot leave distinct remainders when divided by
1998. Two of them must thus leave the same remainder and their difference is a multiple of 1998.
Example 95. Let ‘a’ be any irrational number. Show that there exist infinitely many rational numbers
1
r = p / q much that | a − r | < .
q2
Solution We can assume that a > 0 let Q be a positive integer and consider fractional parts of 0, a , 2a ,…Qa
 1  1 2 
of the first (Q + 1) multiples of a. By PHP, two of these must fall into one of Q interval 0, , , , …,
 Q  Q Q 
   
Q − 1 
, 1 , whereas usual [A , B ] = {x : A ≤ x < B}. In other words, there exist integers q1 and q 2 such that
 Q 
 
(where {x } denotes the fractional part of x )
 S S + 1
{q1 a}, {q 2 a} ∈ , , q1 ≠ q 2
Q Q 

since {q1a} = q1a − [q1a ] , {q 2a} = q 2a − [q 2a ],
1
we get, | qa − b | = | {q1 , a} − {q 2 , a}| <
Q
where q = q1 − q 2 , p = [q , a ] − [q 2 − a ]
since 0 < | q | ≤ Q , this gives 

q − p 
< 1 ≤ 1
 q  Q| q | q2
We have to show that the number of such pair (p, q) is infinite. Assume on the contrary that only for a
pi 1
finite number of ri = ,i = 1, 2, …, N we have | q − ri | < . Since, none of the differences a − ri exactly 0,
qi qi
1
there exist an integer Q such that | a − ri | > for all i = 1, 2, ... , N apply our starting arguments to this Q
Q
1 1
and produce r = p / q, such that | a − r | < ≤ . Hence, r cannot be one of ri , i = 1, 2, ... , N on other
qQ Q
1
hand | a − r | < 2 contradicting this assumption that the fractions ri , i = 1, 2, …,N where all the fraction
q
with this property.
www.pdfworld.in
www.pdfworld.in
Example 96. Each of the given 9 lines cuts a given square into two quadrilaterals, whose areas are in the
ratio 2 : 3. Prove that at least there of these lines pass through the same point.
Solution If a line cuts a square AB (1) into two parts, it must intersect two of its sides (internally); the two
parts are quadrilaterals only if it intersect a pair of opposite sides. At least five of the nine lines must meet
the pair AB, CD or else, the pair AD, BC. Suppose that five meet AB, CD, and let PQ be one of them, let E, F be
the mid-points of AD, BC and PQ meets EF at G. The quadrilaterals APQD, PBCQ formed by PQ are
trapeziums. So,
Q
D C
E F
G
A B
P
1
AD (AP + DQ )
2 2 3
= or
1 3 3
BC (PB + QC )
2
1 1
but (AP + DQ ) = EG and (PB + QC ) = GF
2 2
EG 2 3
Hence, = or
GF 3 2
If G1 and G2 are the points dividing EF in the ratio 2 : 3 and 3 : 2, then PQ contain G1 or G2 of the five lines
meeting AB, CD, at least there must pass through the same point G1 or G2.
Example 97 Given a set of 25 points in the plane such that among any there of them there exist ‘a’ pair at
the distance less than 1. Prove that there exist a circle of radius 1 that contain at least 13 of the given points.
Solution Let A1 , A 2 , ... , A 25 be the 25 points. If the circle with centre A1 and radius 1 contain 12 more
points other than A1, we are done. So, suppose that there are 13 points A 2 , A3 , ... , A14 outside the circle.
Then A1A 2 > 1 and A1Ai > 1 for 3 ≤ i ≤ 14 . So, the given condition, applied to the three points A1 , A 2 ,...,Ai
implies that A 2Ai ≤ 1 for 3 ≤ i ≤ 14. Hence, the circle with centre A 2 and the radius 1 contain 13 points
Ai , 2 ≤ i ≤ 14.
Example 98. Given 6 points inside a circle of radius 1, some two of the 6 units are with in 1 point of each
other.
Solution Let O be the centre of the circle, P and Q be two of the six points inside the circle. If P or Q is the
same as O, then PQ < 1. So let O , P , Q be distinct, we can choose P , Q such that ∠POQ ≤ 60° , let ∠POQ = θ,
then by cosine formula,
PQ 2 = OP 2 + OQ 2 − 2OP ⋅ OQ cos θ
≤ OP 2 + OQ 2 − 2OP ⋅ OQ (1 / 2)
If OQ ≤ OP , then we get
PQ 2 ≤ OP 2 + OQ 2 − OP ⋅ OQ ≤ OP 2 < 1
Similarly, we can prove OP ≤ OQ .
www.pdfworld.in
www.pdfworld.in
Combinatorics 455
Example 99. Given 14 or more integer from {1, 2, ... , 28} there exist four of the given integers which can
be split into two groups of two numbers each with the same sum.
Solution Let a1 , a2 , ... , a14 be 14 numbers in {1, 2, ... , 28}, clearly, 1 + 2 ≤ ai + a j ≤ 27 + 28 for i ≠ j . There
 14
are   = 91 pairs (ai , a j ) and there are only 53 possible values for the sum ai + a j . So, there exist two pairs
2
(ai , a j ) and (ar , a s ) such that ai + a j = ar + a s . If ar = ai , then a s = a j and the pairs are equal. So, ar ≠ ai
similarly, ar ≠ a j , a s ≠ ai and a s ≠ a j . The four numbers ai , a j , ar , a s have the desired property.
Example 100. A person takes at least one aspirin a day for 30 days. If he takes 45 aspirin all together. Show
that in some sequence of consecutive days he takes exactly 14 aspirins.
Solution Let ai be the number of aspirin taken in the ith day, and let Sn = a1 + ... + an , where S1 = a1.
Then, 1 ≤ Sn ≤ 45 for n = 1, 2, ... , 30. Consider the numbers Sn , Sn + 14 for 1 ≤ n < 30. These 60 numbers must
belong to {1, 2, ... , 59}. So some two of them are equal. Since ai ≥ 1 for all i, we see that Sn ≠ Sm for n ≠ m. It
follows that Sn = Sm + 14 for some n > m. Hence,
am + 1 + .. .+ an = Sn − Sm = 14.
Example 101. Prove that we can choose a subset of a set of ten given integers, such that, their sum is
divisible by 10.
Solution Let a1 , a2 , ... , a10 be 10 integers. Let Sn = a1 + ... + an . The 10 numbers S1 ... S10 are either in
congruent modulo 10, or else, two of them congruent modulo 10. In the first case, some Sn is congruent to
0, and hence a1 + ... + an is divisible by 10. In the other case, let Si ≡ S j (mod 10), when i < j . Then,
ai + 1 + ... + a j = S j − Si is divisible by 10.
Example 102. In a group of 7 people, the sum of the age of the members is 332 years. Prove that these
members can be chosen, so that the sum of their ages is not less than 142 years.
Solution Let a1 , a2 , ... , a7 be the ages. Then
a1 + ... + a7 = 332
 7
Suppose that ai + a j + a k < 142 for distinct i, j, k there are   such inequalities. Add them. Each ai is
 2
 6
repeated   times. Thus, we get
 2
 6  7
  (a1 + a 2 + ... + a7 ) <   142,
 2  3
that is, 15332
. < 35142
. . Since, this is wrong, we must have ai + a j + a k ≥ 142 for some distinct i, j, k.
Example 103. Suppose a coin is flipped until 2 heads appear (2 heads need not be consecutive) and then the
experiment stops. Find a recurrence relation for the number an of experiments that end on the nth flip or
sooner.
Solution Clearly, a1 = 0, a2 = 1. Also a3 = 3 . since the possibilities are HH, THH, HTH.
Let n ≥ 2 . Each of the an experiments starts with T or H. Those which starts with T are obtained by
adding T at the beginning of the an − 1 experiments with at most n − 1 flips.
Those which start with H are n − 1 in number as these are of the form HTT….TH, where there are r, T3
and 0 ≤ r ≤ n − 2.
Hence, an = an −1 + n −1
www.pdfworld.in
www.pdfworld.in
Example 104. Find a recurrence relation for the number an of n digit quaternary sequences (i.e., sequence
with terms 0, 1, 2, or 3) with at least one 1, and the first 1 occurring before the first 0 (possibly no 0's).
Solution a1 = 1 since 1 is only possibility.
Also for n = 2, the possibilities are 10, 11, 12, 13, 21, 31 so that a 2 = 6.
Let n ≥ 2
Let x = x 1x 2 ... x n be anyone of the an sequences of required type, then x cannot start with 0 since the
form 0…1… is not allowed while 1 must appear in x.
So, x 1 = 1, 2 or 3. If x 1 = 2 or 3, then sequences x 2 … x n can be anyone of the an −1 required sequences.
Thus, 2an − 1n sequences start with 2 or 3. If x 1 = 1, then each of x 2 , x 3 … can be 0, 1, 2, 3,. Hence,
there are 4n − 1n sequences starting with 1. Thus, an = 2an − 1 + 4n − 1
Example 105. Find a recurrence relation for the number an : r of ways of selecting r integers from the
ordered set x = {1, 2, ... , n}, so that consecutive integers are not selected.
Solution For n = r = 1, the only possible selection is {1}.
for n = 2, r = 1 the only possible selections are {1} and {2}.
Hence, a1, 1 = 1 ; a 2, 1 = 2
Let s = (i1 , i 2 , ... , ir ) be a selection from x of the required form. Assume that
1 ≤ i1 ≤ i 2 < ... < ir ≤ n
Then, we have 2 mutually exclusive cases.
(i) ir ≠ n (ii) ir = n
In case (i), s is a selection of required form and contains r numbers from x − {n}.
So, there are an − 1, r selections of this type. In case (ii) the element ir in s has already been chosen to
be n.
So, to obtain s, we only have to choose the selection t = (i1 , ... , ir − 1) of required type from
x − {n − 1, n}.
Hence, there are an − 2, rn − 1 selections of this type. Therefore, we get.
an , r = an − 1, r + an − 2, r −1
= x n  x n +  for
1
Example 106. For every real number x 1 , construct the sequence x 1 , x 2 , ... by setting x n +1
 n
each n ≥ 1. Prove that there exists exactly one value of x 1 for which 0 < x n < x n + 1 < 1 for every n.
Solution Let P1 (x ) = x
= Pn (n )  Pn (x ) +  for n = 1, 2
1
Pn + 1 (n ) …(i)
 2 
from this recursive definition, we see inductively that
−1
(i) Pn is a polynomial of degree 2n
(ii) Pn has positive coefficients is therefore an increasing convex function for x ≥ 0.
(iii) Pn (0) = 0, Pn (1) ≥ 1
(iv) Pn (x 1 ) = x n
1
Since the condition x n +1 > x n is equivalent to x n > 1 −
, we can reformulate the problem as follows
n
show that there is unique positive real number t such that.
1
1 − < Pn (t ) < 1 for every n.
n
www.pdfworld.in
www.pdfworld.in
Combinatorics 457
Since Pn is continuous and increases from 0 to a value of ≥ 1 for 0 ≤ x ≤ 1, there is unique values an and bn
such that
1
an < bn , Pn (an ) = 1 − , Pn (bn ) = 1 …(ii)
n
Pn + 1 (an ) =  1 −   1 − +  = 1 −
1 1 1 1
By definition (i),
 n  n n n
1
Pn + 1 (an − 1 ) = 1 −
n+1
We see that an < an +1 …(iii)
1
Also since Pn + 1 (bn ) = 1 + and Pn + 1 (bn + 1 ) =1
n
bn > bn +1 …(iv)
1
Since Pn is convex, the graph of Pn (x ) lies below the chord y = x for 0 ≤ x ≤ bn
bn
1 an
In particular Pn (an ) = 1 − ≤
n bn
from this and the fact that bn ≤ 1,
b
we find that bn − n ≤ an
n
b 1
bn − an ≤ n ≤ for all n.
n n
Thus, we have 2 infinite bounded sequences {an }, {bn } the first is increasing the second decreasing
an < bn and the difference between their nth members approaches 0 as n increases. We conclude that
there is a unique common value t that they approach
an < t < bn ∀ n
Number uniquely satisfies
1
1− < Pn (t ) < 1 ∀ n
n
Example 107. Let P1 , P2 , ... , Pn be distinct 2 element of the set of elements {a1 , a2 , ... , an } such that if
Pi ∩ Pj ≠ φ, then {ai , a j } is one of the P’s. Prove that each of a’s appear in exactly two of the P’s.
Solution Denote by m j the number of P ’ s containing a j , j = 1, 2, ... , n, since each pair consists of two
distinct elements.
m1 + m 2 + ... + mn = 2n …(i)
mk 
The number of pairs {Pi , Pj } both containing a k is   . Now consider the one to one mapping
 2 
(Pi , Pj ) ↔ (ai , a j ). By hypothesis (ai , a j ) is a pair p k , if and only if Pi ∩ Pj ≠ φ. Since there are only n pairs
P1 ,... , Pn , it follows that
m  1
Σ  k  = (Σ m k2 − Σm k ) ≤ n …(ii)
 2  2
By the Power mean or Cauch inequality.
(Σ m k )2 ≤ nΣ m k2
Using Eqs. (i) and (ii) we conclude that
1
n≤ (Σ m k2 − Σ m k ) ≤ n …(iii)
2
www.pdfworld.in
www.pdfworld.in
Thus, equality must hold in the power mean inequality. But this occurs if and only if all the m k are equal.
By Eq. (i), m1 = m 2 = ... = mn = 2
Example 108. Three distinct vertices are chosen at random from the vertices of a given regular polygon
of (2n + 1) sides. If all such choices are equally likely. What is the probability that the center of the given
polygon lies in the interior of the ∆ determined by 3 chosen random points?
Solution Let the vertices in order be V0 , V1 ,... , V2n . We can assume that the first vertex chosen is fixed at
 2n 
V0. Then, the number of ways of picking 2 more vertices is   . Now, if one of the remaining 2 random
2
vertices is Vk , 1 ≤ k ≤ n , there will k ∆s possible that contain the centre. If say k = 3 , then the only possible
triangles with vertices V0 , V3 which. Contain the centre are V0V3Vn + 1 , V0V3Vn + 2 and V0V3Vn + 3. Thus, the
number of favourable cases is
n
Σ k = n (n + 1) / 2
k =1
Finally the desired probability is

n (n + 1) n+1
P= =
 
2n 2(2n − 1)
2 
2
Example 109. There are n people at a party. Prove that, there are 2 people such that of the remaining n − 2
people there are at least (n / 2) − 1 of them, each of whom knows both or else knows neither of two. Assume
that knowing is a symmetrical relation [x ] denotes the greatest integer less than or equal x.
Solution Given 2 people at the party, we describe a third person as “mixed” w.r.t. that pair, if that person
knows exactly one of the 2. Thus a person, who know exactly k people at the party is mixed with respect to
k (n − 1 − k ) pairs. By AM–GM inequality, each person is mixed w.r.t. at most (n − 1)2 / 4 pairs. Thus, there are
at most
n (n − 1)2 n − 1 n 
=  
4 2  2
n 
person mixed pair combinations and so for at least one of the   pairs at most [(n − 1) / 2] of remaining
 2
people are mixed. For this pair there are at least
n − 1 n
n − 2−  = −1
 2  2
other who knows either both or neither of two.
Example 110. (i) Suppose that each square of a 4 × 7 chessboard is coloured either black or white. Prove
that, with any such colouring, the, board must contain a rectangle, whose 4 distinct unit corner squares are
all of same colour.
(ii) Exhibit a black white colouring of a 4 × 6 board in which 4 corner squares of every rectangle are not all of
same colour.
www.pdfworld.in
www.pdfworld.in
Combinatorics 459
Solution (i) The stated result is even valid for a 3 × 7 board. We call any column (consisting of 3 unit
squares) black, if it has more black unit squares than white ones otherwise we call it white. Since there are 7
colours, at least 4 of theam are of the same kind, say black. We now show that there is a rectangle, whose
vertices are these 4 black columns and whose corner squares are all black. We can even assume that each of
these 4 black column has one white unit square, since the white square can be in one of only 3 positions it
follows that 2 of these 4 columns are identically coloured giving desired result.
 4
(ii) For the 4 × 6 chessboard there are   = 6 different arrangements of two B’s and two W’s in a
 2
column. One such coluring is given in the figure below, it contains no rectangle with identically colourd
corner squares
B B B W W W
B W W W B B
W B W B W B
W W B B B W p
Example 111. 9 mathematicians meet at an international conference and discover that among any 3 of
them at least two speak a common language. If each of the mathematician can speak at most 3 language,
prove that there are at least 3 of the mathematician who can speak the same language.
Solution We assume that at most 2 mathematician speak a common language. Each mathematician can
speak to at most 3 others, one for each language he or she knows. Suppose mathematician M1 can only
speak with M 2 , M3 and M 4 . Now M5 can speak with at most 3 of M 2 , M3 and M 4 or at most 3 of M 6 , M7 , M 8.
This leaves one of the last 4 who cannot speak with M1 or M5 giving desired contradiction.
Example 112. In a party with 1982 persons, among any group of 4, there is at last one person, who knows
each of other 3. What is the minimum number of people in the party, who know everyone else?
Solution
Case I We assume ‘Knowing’ is not a symmetrical relation i.e., A may know B but B does not know A.
For this case no one need know everyone else. Just consider all the people arranged in a circle such that
each person knows everyone else except the person next to him clockwise.
Case II We assume ‘Knowing’ is a symmetrical relation.
Let {P1 , P2} be a pair who do not know each other. Then if {P3 , P4 } is a pair disjoint from the first pair, P3
and P4 must know each other, since one of P1 , P2 , P3 , P4 knows the other 3 by hypothesis. So if there is a
third person P3 who does not know everyone, it must be P1 or P2 he (she) does not know. If there were a
fourth person P4 who did not know everyone, it would again be P1 or P2 he (she) did not know, but then
{P1 , P2 , P3 , P4 } would violate hypothesis. Thus all except at most 3 people must know everyone else.
Example 113. Find the number of permutations (p1 , ... , p6 ) of 1, 2, 3, 4, 5, 6 such that for any k , 1 ≤ k ≤ 5,
(p1 , p 2 , ... , p k ) is not a permutation of (1, 2, … , k) i.e., p1 ≠ 1. (p1 , p 2 ) is not a permutation of (1, 2), (p1 , p 2 , p3 ) is
not a permutation of (1, 2, 3) etc.
Solution For each positive integer k , 1 ≤ k ≤ 5. Let N k denote the number of permutations (p1 , ... p6 ) such
that p1 ≠ 1, (p1 , p 2 ), is not a permutation (1, 2), …, (p1 , ... , p5 ) is not a permutation of (1, 2, …,5). We are
required to find N 5.
Since out of the 6! permutations of (1, 2,…, 6) there are 5! permutations having 1 in the first place. We
have to find the number of permutations left after removing from the set of all permutations of
(1, 2, …, 6) the one that begins with 1.
www.pdfworld.in
www.pdfworld.in
Now N 1 = 6 ! − 5 ! = 600. To get N 2 we now remove those permutations in which (p1 , p 2 ) is a permutation
of (1, 2). Since all permutations of the type (1, 2, p3 , ... , p 6 ) have already been removed. We have to
further remove permutations of type (2, 1, p3 ,... , p 6 ). The number of such permutation being 4!.
We get N 2 = 600 − 4 ! = 576.
To get N 3 , we have to subtract from N 2 the no. of those permutation (p1 ,... , p 6 ) in which (p1p 2p3 ) is a
permutation of (1, 2, 3) and which have not already been removed. Since all permutations of the form
(1, p 2 , ... , p 6 ) and (2, 1, p3 ,... , p 6 ) have already been removed, we have to count the number of
permutations of the form (2, 3, 1, p 4 , ... , p 6 ), (3, 1, 2, p 4 , p5 , p 6 ) and (3, 2, 1, p 4 , p5 , p 6 ). The number of all
such permutation is 3 × (3 !) i.e., 18
∴ N 3 = 576 − 18 = 558
To get N 4 , we have to remove those permutations (p1 , p 2 ,... , p 6) in which (p1 , ... , p 4 ) is a permutation of
(1, 2, 3, 4) and which have not already been removed i.e., permutations of the type (1, p 2 ,…,p 6 ), (2, 1,
p3,…,p 6 ). (2, 3, 1, p 4 ,... , p 6 ) (3, 1, 2, p 4 ,... , p 6 ) and (3, 2, 1, p 4 ,... , p 6 ). This means that we have to remove all
permutations of type (4, p 2 ,…,p 6 ), (2, 4, p3,…, p 6) (3, 4, p3 ,…, p 6 ) (2, 3, 4, p 4 ,…, p 6 ), (3, 2, 4, p 4 ,…, p 6 ) in
which the first 4 elements are a permutation of (1, …, 4). There are (3 ! + 2 ! + 2 ! + 1 + 1) × 2 = 24
permutations in all to be removed. N 4 = 558 − 24 = 534. To get N 5, we have to further remove
permutations (p1 ,... , p 6 ) in which (p1,…,p5 ) is a permutation of (1,…,5) and which have not been
removed. There are 24(4 !) such permutations for which p1 = 5, 18 with p 2 = 5, 16 with p3 = 5 and 13 with
p 4 = 5 so that in all there are 71 such permutations.
Thus, N 5 = 534 − 71 = 463, which is desired number of permutations.
Example 114. In a list of 200 number, everyone (except the end ones) is equal to the sum of the two adjacent
number in the list. The sum of all the numbers is equal to the sum of the first 100 of them. Find that sum if the
35th number in the list in 6.
Solution Q a 2 = a1 + a3 ∴ a3 = a 2 − a1
a3 = a 2 + a 4 ∴ a 4 = a3 − a 2 = −a1
a 4 = a3 + a5 ∴ a5 = a 4 − a3 = −a 2
a5 = a3 + a 6 ∴ a 6 = a5 − a 4 = − a3 = a1 − a 2
Also, a1 + a 2 + a3 + a 4 + a5 + a 6 = 0
Again, since an = an −1 + an + 1
an + 1 = an − an −1 = −an − 2 ∀ n > 2.
Also, an + 1 = −an − 2 = (−1)2 an −5 = an −5 ∀ n > 5.
This shows that a7 = a1,a 8 = a 2 , a 9 = a3 , a10 = a 4

a11 = a5 , a12, = a 6 i.e., in the list the first 6 elements are repeated so that the list is
a1 , a 2 , a 2 − a1 ,−a1 ,−a 2 , a1 − a 2 , a1 , a 2…
Clearly a35 = a5 = −a 2, since a35 = −6.
Therefore it follows that a 2 = 6.
Also, the sum of the first hundred elements (s100, say) is equal to the sum of all the elements in the
list.
∴ a101 + a102 +……+ a 200 = 0.
Since, the sum of the first 6 elements is zero.
∴ a k + 1 + a k + 2 + …+ a k + 6 = 0 k = 0, 1, 2, …
This shows that a103 + a104 + …+ a198 = 0.
Also, since a101 = a5, a102 = a 6, a199 = a , a 200 = a 2.
∴ a1 + a 2 + a5 + a 6 = 0
So that a1 + a 2 + (−a 2 ) + (a1 − a 2 ) = 0
www.pdfworld.in
www.pdfworld.in
Combinatorics 461
i . e. , 2a1 = 6
which gives a1 = 3.
S 200 = S100 = a1 + ... + a100
= a 97 + a 98 + a 99 + a100
= a1 + a 2 + a3 + a 4
= 2a 2 − a1 = 9
Sum of all the numbers in the list is 9.
Example 115. A 2 × 2 × 12 hole in a wall is to be filled with twenty four 1 × 2 × 2 bricks. In how many
different ways can this be done, if the bricks are in distinguishable?
Solution Let Tn be the number of ways of filling up a 2 × 2 × n holes with 1 × 1 × 2 bricks. We obtain a
recursion relation for Tn by expressing in terms of the numbers of ways of filling smaller holes. Assume
that, the long axis of hole is vertical. The other two edge directions will be called to the right and forward.
First, we count the number of ways of filling up the hole, if the bottom layer consist of two bricks lying
on their sides. They can lie in two ways, with their long axes in the left - right or backward - forward
position the number of these packings is 2Tn −1.
Next consider the packing, in which some of the bricks of the bottom layer stick out, but the bricks in
layer 1 and 2 fill this sub-box with nothing sticking out. There are 5 such packings; one in which all four
bricks are vertical, one in which the two bricks in front are vertical and the two in the back are
horizontal and the three more packings of this sort in which the two standing bricks are on the left in
the back, and on the right, this gives 5Tn − 2 more packings.
Next let K be an integer, 2< K ≤ n. We count the number of ways to fill one hole in such a manner that
the lowest bottom sub - hole which is completely filled with nothing sticking out has K layers. The
bottom layer of such a packing must consist of two vertical and one horizontal bricks. There is no
further choice until we get above the Kth-layer: the two empty spaces in each layer must be filled with
the two, vertical bricks until we reach layer K, when the two spaces must be filled with a horizontal brick
thus for all, the values of K in the given range, there are 4 ways of filling the K lowest layers. Since for
every packing there is one and only one value of K which satisfies the above conditions we get
Tn = 2Tn −1 + 5Tn − 2 + 4Tn − 2 + ... + 4T0 …(i)
Here, we set
T0 = 1 (Also)
Tn −1 = 2Tn − 2 + 5Tn −3 + 4Tn − 4 + …+ 4T0 …(ii)
Combining Eqs. (i) and (ii), we get
Tn = 3Tn −1 + 3Tn − 2 − Tn −3 …(iii)
The characteristic equation of the latter recurrence relation is x − 3x − 3x + 1 = 0
3 2
its roots are −1, 2 + 3, 2 − 3. Thus, Tn has the form

Tn = C1(−1)n + C2(2 + 3 )n + C3 (2 − 3 )n .
Since the initial values are T0 = 1, T1 = 2 and

T2 = 9
+1 +1
(−1)n (2 + 3 )n (2 − 3 )n
Tn = + +
3 6 6
In particular,
T12 = 1 / 3 + (2 + 3 )13 + (2 − 3 )13
the nearest integer to (2 + 3 )13 = 4, 541, 161
www.pdfworld.in
www.pdfworld.in
Example 116. 29th Feb of year 2000 will fall on a Tuesday show that after this date 29th Feb will fall on
Tuesdays thrice in the whole next century. What are the 3 years when this will happen?
Solution 365 ≡ 1 (Mod 7).
Therefore, 28th Feb (the last day Feb) of 2001 will be a Wednesday (the day next to Tuesday). Let us
agree to express this by saying that there is an excess of one day in any ordinary year. With this
terminology. There will be an excess of two days in a leap year the next leap year after year 2000 will be
year 2004, there will be 1 + 1 + 1 + 2 i.e., 5 excess days upto 29th Feb 2004.
The day of the week of 29 th February will be Tuesday, when the number of excess day is an exact
multiple of 7. Therefore our problem is to find those positive integer K for which the number of excess
days in 4K years after the year 2000 is an exact multiple of 7 and which 4K < 100. The number of excess
days in 4K years is 5K. This is a multiple of 7 when K = 7, 14, 21, …i.e.,
in 4 × 7, 4 × 14, 4 × 21,…years, after the year 2000. Since, only 3 of these numbers are less then 2000,
therefore in 21st century, there are only 3 years namely 2028 2056 and 2084 in which 29th Feb is a
Tuesday.
Example 117. The number 3 can be written as a sum of positive integers in 4 ways viz.,
3, 2 + 1, 1, 1 + 2, 1 + 1 + 1. Show that any positive integer n can be so expressed in 2n −1 ways.
Solution Consider the sum

1+1+1+1+ …+1
There being n terms in all. We can break this sum into one or more parts (n parts at the most) by either
putting or not putting parenthesis after the n − 1 ‘+’ signs. This can be done in 2n −1ways.
Example 118. The 64 squares of an 8 × 8 chess board are filled with positive integers in such a ways that
each integer is the average of the integers on the neighbouring squares. Show that all the 64 integer entries
are in fact equal.
Solution Choose the square (call it A) which is filled with the smallest of all the positive integers (call it K,
say) filled in the 64 squares. Consider all its neighbours. If any one of them is filled with a positive integer
greater than K, then the average of all the integers in the squares occupying the neighbouring positions will
be greater than K. This is a contradiction. Hence, all the neighbouring squares are filled with the same
integer K.
Let us apply this procedure to all the squares in the row in which A lies, first beginning with the squares
immediately preceding A (if A is not in first column) and reaching the first column and then starting
with the square immediately to the right of A (if A is not in last column) reaching the last column. By
this process, we find that all the squares in the row to which A belongs and the row immediately above
and immediately below are filled with K. We now apply this procedure to all the elements in the row
above A (if there be one such row) and then to the row below A (if there be one such row). We find that
all the rows can be exhausted in a finite number of steps and that all the 64 squares are first with the
same positive integer K.
Hence, all 64 integers entries are equal.
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1. How many numbers of 4 digits can be formed 14. The letters of the word ‘OUGHT’ are written in
with the digits 1,2,3,4,5, no digits being all possible orders and these words are
repeated. written out as in a dictionary. Find the rank of
2. How many numbers each lying between 100 the word ‘TOUGH’ in this dictionary.
and 1000 can be formed with the digits 15. In how many ways, 5 boys and 4 girls can be
2,3,4,0, 8,9, no digit being repeated ? seated at a round table in the following case
3. How many numbers, of 9 digit numbers, (i) when there is no restriction.
which have all different digits ? (ii) all the 4 girls sit together.
4. Find the sum of all the 4 digits numbers that (iii) all the 4 girls don’t sit together.
can be formed with the digits 0,2,3 and 5. (iv) no two girsl sit together.
5. There are 20 books, of which 4 are single 16. A man has 8 children to take them to a zoo.
volume and the other are books of 8, 5 and 3 He takes three of them at a time to the zoo as
volumes respectively. In how many ways can often as he can without taking the same 3
all these books be arranged on a shelf, so that children together more than once. How many
volumes of the same book are not separated? times will he have to go to zoo? How many
6. A library has 5 copies of one book, 4 copies times a particular child will go ?
each of 2 books, 6 copies each of 3 books and 17. Out of 7 men and 4 ladies a committee of 5 is
single copies each of 8 books. In how many to be found . In how many ways can this be
ways, can all the books be arranged, so that done so as to include at least 3 ladies ?
copies of the same books are always together?
18. In an examination the question paper contain
7. 6 papers are set in an examination, 2 of them three different sections A, B and C containing
in Mathematics. In how many different orders 4, 5 and 6 questions respectively. In how
can the papers be given, if two Mathematics many ways, a candidate can make a selection
papers are not successive ? of 7 questions selecting at’ least two question
8. In how many ways 18 white and 19 black be from each selection ?
arranged so that all the 18 white balls are not 19. From 8 gentlemen and 4 ladies, a committee
be together. It is given that balls of same of 5 is to be formed. In how many way can this
colour are identical. be done so as to include at least one lady ?
9. In a class of 10 students, there are 3 girls, in 20. In an election for 3 seats there are 6
how many ways can they be arranged in a row candidates. A voter can not vote for more
such that no two of the three girls are than 3 candidates . In how many ways can he
consecutive ? vote ?
10. In how many ways, the letter of the word 21. In an election, number of candidate exceeds
‘DIRECTON’ be arranged, so that their vowels the number to be elected by 2. A man can vote
are never together ? in 56 ways. Find the number of candidates.
11. How many words can be formed with the 22. At an election a voter can vote for any number
letter of the word ‘VICE-CHANCELLOR’ so that of candidates not greater than the number to
the vowels are together ? be chosen. There are 10 candidates and 5
12. Find the number of different permutations of members are to be chosen. Find the numbers
the letters of the word ‘BANANA’. of ways in which a voter may vote.
13. How many number of arrangements of the 23. A bag contain 5 red, 4 green and 3 blue balls
letters of the word ‘BENEVOLENT’ How many of the same are supposed to be distinct (not
of them end in L ? alike). In how many ways
www.pdfworld.in
www.pdfworld.in
(i) some balls can be drawn from the bag ? 32. Find the number of all 5-digit numbers (in
(ii) some balls containing at least one red and base 10) each of which contains the block 15
one green ball can be drawn ? and is divisible by 15. (e.g., 31545, 34155 are
24. Find the number of selection of at least one two such numbers.) (RMO 2005)
red ball from 4 red and 3 green balls, if the 33. How many 6-digit numbers are there such
balls of the same are different. that:
25. There are 3 books of Mathematics, 4 of Science (a) the digits of each number are all from the
and 5 of Literature. How many different set {1, 2, 3, 4, 5}
collections can be made, such that each (b) any digit that appears in the number
collection consist appears at least twice?
(i) one book of each subject (e.g., 225252 is an admissible number,
(ii) at least one book of each subject while 222133 is not.) (RMO 2007)
(iii) at least one book of Literature
34. Three non-zero real numbers a , b , c are said
26. All the 7-digit numbers containing each of the 1 1 2
digits 1, 2, 3, 4, 5, 6, 7 exactly once and not to be in harmonic progression, if + = .
a c b
divisible by 5, are arranged in the increasing
Find all three-term harmonic progressions
order. Find the 2000th number in this list.
(RMO 2000)
a , b , c of strictly increasing positive integers
in which a = 20 and b divides c. (RMO 2008)
27. Find the number of positive integers x which
35. Find the number of all integer-sided isosceles
satisfy the condition   = 
x x 
(Here, [z ]
 99   101  obtuse angled triangles with perimeter 2008.
(RMO 2008)
denotes, for any real z, the largest integer not
exceeding z; e.g., [7/4] = 1.) (RMO 2001) 36. Find the number of all 6-digit natural
numbers such that the sum of their digits is
28. In n is an integer greater than 7, prove that
10 and each of the digits 0, 1, 2, 3 occurs at
 n  n 
  −   is divisible by 7. least once in them. (RMO 2008)
 7  7 
37. Find the sum of all 3-digit natural numbers
n  which contain at least one odd digit and at
[Here,   denotes the number of ways of
 7 least one even digit. (RMO 2009)
choosing 7 objects from among n objects; also,  n 
38. For each integer n ≥ 1, define an =  ,
for any real number x , [x ] denotes the greatest  [ n ]
integer not exceeding x. ] (RMO 2003)
where [x ] denotes the largest integer not
29. Find the number of ordered triples (x , y , z ) of exceeding x , for any real number x. Find the
non-negative integers satisfying the number of all n in the set {1, 2, 3, ..., 2010} for
conditions: which an > an + 1. (RMO 2010)
(i) x ≤ y ≤ z ; (ii) x + y + z ≤ 100. (RMO 2003) 39. Find three distinct positive integers with the
30. Prove that the number of triples (A , B , C ) least possible sum such that the sum of the
where, A , B , C are subsets of {1, 2,... ,n such reciprocals of any two integers among them
that A ∩ B ∩ C = φ, A ∩ B ≠ φ, B ∩ C ≠ 0 is is an integral, multiple of the reciprocal of the
7n – 2 ⋅ 6n + 5n . (RMO 2004) third integer. (RMO 2010)
31. Determine all triples (a , b , c ) of positive 40. Find the number of 4-digit numbers (in base
integers such that a ≤ b ≤ c and 10) having non-zero digits and which are
a + b + c + ab + bc + ca = abc + 1. (RMO 2005) divisible by 4 but not by 8. (RMO 2010)
Level 2
1 1 1 1 1 2. There are 1994 employees in the office. Each
1. The sum + + + + can
1! 9! 3!7! 5!5! 7!3! 9!1! of them knows 1600 others of them. Prove
2a
be written in the form , where a and b are that we can find 6 employees, each of them
b!
positive integers. Find the ordered pair (a, b). knowing all 5 others.
(Note The ! marks are ‘‘factorial’’ symbols.)
www.pdfworld.in
www.pdfworld.in
Combinatorics 465
3. Find the number of permutations 11. In a lottery, tickets are given nine-digit
( p1 , p 2 ,.... , p 6 ) of 1, 2,....,6 such that for any k, numbers using only the digits 1, 2, 3. They
1 ≤ k ≤ 5, ( p1 , p 2 ,.... , p k ) does not form a are also coloured red, blue or green in such a
permutation of 1, 2,..., k. way that two tickets whose numbers differ in
i.e., p1 ≠ 1 ( p1 , p 2 ) is not a permutation of 1, 2 all the nine places get different colours.
etc. Suppose the ticket bearing the number
4. For non-negative integers n , r the binomial 122222222 is red and that bearing the
n  number 222222222 is green. Determine, with
coefficient   denotes the number of
r  proof, the colour of the ticket bearing the
combinations of n objects chosen r at a time, number 123123123. (INMO 2003)
n  n  12. Prove that the number of 5-tuples of positive
with the convention that   = 1 and   = 0,
 0 r  integers (a , b , c , d , e ) satisfying the equation
if n < r. Prove the identity abcde = 5(bcde + acde + abde + abce + abcd )
∞
 n – r + 1  r − 1   n 
∑  d   d − 1 =  r  for all integers n , r is an odd integer. (INMO 2004)
d =1 13. All possible 6-digit numbers, in each of which
with 1 ≤ r ≤ n. the digits occur in non-increasing order (from
5. Show that, where k + n ≤ m , left to right e.g., 877550) are written as a
n
n   m  m + n  sequence in increasing order. Find the
∑  i   k + i  =  n + k  . 2005th number in this sequence. (INMO 2005)
i =0
6. An ‘‘n-society’’ is a group of n girls and m boys. 14. Some 46 squared are randomly chosen from
Show that there exist numbers n0 and m 0 such a 9 × 9 chess board and are coloured red.
that every n0 – m 0 society contains a subgroup Show that there exists a 2 × 2 block of
of 5 boys and 5 girls in which all of the boys 4 squares of which at least three are coloured
know all of the girls or none of the boys knows
red. (INMO 2006)
none of the girls.
15. Let σ = (a1 , a 2 , a3 ,... , an ) be a permutation of
7. (a) Find the three-digit numbers equal to the
sums of the factorials of their digits. (1, 2, 3,...,n). A pair (ai , a j ) is said to
(b) Find all whole numbers equal to the sums correspond to an inversion of σ, if i < j but
of the squares of their digits. ai > a j . (Example In the permutation (2, 4, 5,
8. Suppose the n 2 numbers 1, 2, 3,..., n 2 are 3, 1), there are 6 inversions corresponding to
arranged to form an n by n array consisting of the pairs (2, 1), (4, 3), (4, 1), (5, 3), (5, 1), (3, 1).
n rows and n columns such that the numbers How many permutations of (1, 2, 3,...,n),
in each row (from left to right) and each (n ≥ 3), have exactly two inversions? (INMO 2007)
column (from top to bottom) are in increasing
16. How many 6-tuples (a1 , a 2 , a3 , a 4 , a5 , a 6 ) are
order. Denote by a jk the number in jth row
there such that each of a1 , a 2 , a3 , a 4 , a5 , a 6 is
and kth column. Suppose b j is the maximum
possible number of entries that can occur as from the set {1, 2, 3, 4} and the six
a jj , 1 ≤ j ≤ n. expressions a 2j – a ja j + 1 + a 2j + 1 for j = 1, 2, 3,
Prove that b1 + b2 + b3 + ... + bn 4, 5, 6 (where a7 is to be taken as a1) are all
n
≤ (n 2 – 3n + 5). equal to one another? (INMO 2010)
3 17. All the points in the plane are coloured using
(Example In the case n = 3, the only numbers three colours. Prove that there exists a
which can occur as a 22 are 4, 5 or 6 so that triangle with vertices having the same colour
b2 = 3.) (INMO 2002)
such that either it is isosceles or its angles are
9. Do there exist 100 lines in the plane, no three in geometric progression. (INMO 2010)
of them concurrent, such that they intersect
18. Suppose five of the nine vertices of a regular
exactly in 2002 points? (INMO 2002)
nine-sided polygon are arbitrarily chosen.
10. Find all 7-digit numbers formed by using only Show that one can select four among these
the digits 5 and 7 and divisible by both 5 and 7. five such that they are the vertices of a
(INMO 2003) trapezium. (INMO 2011)
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1. Number of digits, n = 5 But in the 10
P9 numbers, there are some
Number of places to be filled, r = 4. numbers which begin with 0.
Number of numbers of 4 digits out of the 5 For the number of numbers beginning with
digits (1,2,3,4,5) = Number of permutations of zero number of places to be filled up = r = 8.
5 things taken 4 at a time. Number of digit remain to be utilised = n = 9.
5! 5! Number of number of 9 digits beginning with
= 5P4 = = = 1⋅2⋅3⋅ 4⋅5 0 = 9P8.
(5 − 4)! 1 !
= 120
10 ! 9 !
2. Number of different digits, n = 6. Since = 10P9 − 9P8 = −
1! 1!
number lying between 100 and 1000 are of 3
digits. = 10 ⋅ 9 ! − 9 ! = (10 − 1)9 !
Total number of numbers of 3 digits formed = 9 ⋅ 9!

with the digits 2,3,4,0,8,9, is 4. Keeping 0 at unit (or tens or hundreds) place
6! 6! number of places to be filled up = r = 3.
6
P3 = = = 120 …(i)
(6 − 3)! 3 ! Number of digits remain to be utilised = n = 3.
Hence, number of numbers of 4 digits formed
But of these 120 numbers, there are some with the digits 0,2,3 and 5 in which 0 may
numbers which begin with (zero) and which 3!
comes at unit place is 3 P3 = =6
are not our purpose. 0!
Now for the numbers beginning with zero i.e., 0 may come at unit or tens or hundreds
number of digits remain to be used is place in 6 number of times.
(2,3,4,8,9) = 5 = n and number of places Also keeping 2 (3 or 5) at unit (or tens or
remain to be filled up, r = 2. hundred) place. Number of places to be filled
Hence, number of numbers of 3 digits up = r = 3.
beginning with 0 (zero) Number of digits remain (0,3,5) to be utilised
= 5P2 =
5!
=
5!
= 20 … (ii) of 4 digits = n = 3.
(5 − 2)! 3 !
Hence number of numbers of 4 digits, such
Required number of numbers between 100 that 2 comes at unit place = 3P3. But out of
and 1000 formed with the digits 2,3,4,0,8,9 = these 3 P3, numbers there are some number,
number of numbers of 3 digits formed with which begin with zero. To find the number of
the digits 2,3,4,0,8,9 these numbers keeping 0, at thousand place,
= 120 − 20 = 100 2 at unit place then number of places to be
filled up = r = 2.
3. Since no digits are given, so that we consider
all the digits 0,1,2,3,4,5,6,7,8,9. Number of digits remain to be utilised
Number of digit, n = 10 = n = 2.
For the numbers of digits, number of places to Numbers of numbers of 4 digits beginning
be filled up, r = 9. with zero and ending with 2 is
Number of numbers of 9 digits formed with 2!
2
P2 = = 2 ! = 2.
the digits 0,1,2,3,4,5,6,7,8,9, is 10 P9. 0!
www.pdfworld.in
www.pdfworld.in
Combinatorics 467
Number of numbers such that 2(3 or 5) comes Total number of books = 14.
at unit place = 3P3 − 2 = 6 − 2 = 4.
Now, for arrangements of these books
Hence, number of numbers, such that 2(3 or 5) number of places to be filled up = r = 14.
comes at unit place = 4. i.e., 2 (3 or 5) may
Number of books = n = 14.
comes at unit place in 4 number of times, 2 (3
or 5) may come at 10’s place in 4 number of Arrangements of these 14 books is 14!
times 2 (3 or 5) may come at hundred place in
Now for arrangements of (A1 , ... , A5 ) 5 copies
4 number of times. Number of digits remain
of the book A among themselves, we see that
to be filled up, r = 3 . Number of digits remain
number of places to be filled up = r = 5
to be utilised n = 3. Number of numbers of 4
digits such that 2(3 or 5) comes at 1000 place Number of things = n = 5.
= 3P3 = 6.
But all the copies of the same book are
i.e., 2(3 or 5) may come at 1000 place 6 identical. Hence, number of arrangement of 5
number of times. copies of the book A is
5
6 number of times. P5 5 !
= =1
Sum of digits at unit place 5! 5!
= [0 × 6 + (2 + 3 + 5) × 4] × 1 Number of arrangements of the copies of the
4! 4! 6!
Sum of digits at tens place book B = of book C = of books D = of
4! 4! 6!
= [0 × 6 + (2 + 3 + 5) × 4] × 10 6! 6!
book E = of book F = .
Sum of digits at hundred place 6! 6!
= [0 × 6 + (2 + 3 + 5) × 4] × 100 Total number of arrangements of all the 14
Sum of digits at thousand place books is
5! 4! 4! 6! 6! 6!
14 ! + × × × × × = 14 ! + 1.
= [0 × 6 + (2 + 3 + 5) × 4] × 1000 5! 4! 4! 6! 6! 6!
Hence, sum of all number of 4 digits formed 7. Let P1 , P2 , P3 , P4 , P5m and P6m be 6 papers,
with digits 0,2,3,5 is 44440. where P5m and P6m are 2 papers of Maths.
5. Let A, B, C, D be 4 books, each of single Number of arrangements of the 4 papers on
volume. Let the books E, F and G have which there is no restriction in a line is
volumes 4
P4 = 4 ! (n = 4; r = 4)
(E1 ... E 8 ), (F1 ... F5 ), (G1 ... G3 ) respectively
Two Mathematics papers on these 5 places
Number of place to be filled up = 7
can be arranged in 5 P2 ways.
Number of books = n = 7.
Number of arrangements of the 6 papers, if 2
Number of arrangement of these books is
Mathematics papers are not successive is
7!
7
P7 = = 7! 4 ! × 5P2 = 480
0!
Also the 8 volumes (E1 , ... , E 8 ) of book E can be 8. Total number of arrangements of 37 ball
arranged among themselves in 8! ways, 5 without any restriction in a line
volumes (F1 , ... , F5 ) of book, F can be arranged 37
P37 37 !
in 5! ways and 3 volumes (G1 , ... , G3 ) of book G = =
18! 19 ! 18! 19 !
can be arranged among themselves in 3! ways.
Total number of arrangements = 7 ! 8! 5 ! 3 !. Also keeping 18 white balls together
B1 , ... , B19 Number of different balls = 20.
6. Let (A1 , ... , A5 ) be 5 copies of one book A, 20
(B1 , ... , B4 ) and (C1 , ... , C4 ) be 4 copies each of 2 P20
There can be arranged in ways
books B and C. Let (D1 , ... , D6 ), (E1 , ... , E 6 ), 19 !
(F1 , ... , F6 ) be 6 copies each of 3 books, D, Since, 19 are alike.
E,…,N be the single copies of 8 other books.
www.pdfworld.in
www.pdfworld.in
Also, the white balls among themselves can be Hence, the required arrangement of the
18 letters of the word DIRECTOR such that the
P18 18!
arranged in = ways.
18! 18! three vowels never come together is
8! 6 ! 6!
Number of arrangement of these 37 balls, so − × 3! = (8 × 7 − 3 !)
that all the 18 white balls are together 2! 2! 2!
720
20
P20 18! 20 ! = (56 − 6)
= × = 2
19 ! 18! 19 !
= 360 × 50 = 18000
So, required number of arrangements of the
37 balls, so that all 18 white balls are not 11. Keeping the vowels together V, C, C, H, N,C L,
together L, R (I, E, A, E, O)
37 ! 20 ! Number of letters (things) = 10
= −
18! 19 ! 19 number of places to be filled up = 10
9. Since in a class of 10 students, 3 are girls, so Hence, number of arrangements
that number of boys = 7 and number of girls 10
P10 10 !
= 3. = =
3! × 2! 3! × 2!
Now, number of 7 boys in a line on which
there is no restriction = 7 P7 = 7 ! (n = 7; r = 7) Also for the vowels (I , E , A , E , O ), number of
vowels = 5 and number of places for the 5
Also since no two girls sit together, so that vowels is = r = 5
number of places for the 3 girls = 8 . Hence, vowels among themselves can be
Hence, number of arrangements of the 3 girls arranged in
according to the condition = 8P3. 5
P5 5 !
= = = 60
Hence, required number of arrangements of 2! 2!
10 student’s (7 boys +3 girls) so that no two
8! Hence, total number of arrangements or total
girls are together = 7 ! × 8P3 = 7 ! × number of words formed with the letters of
5!
the word VICE-CHANCELLOR so that the
10. Total number of letters in the word vowels are together is
DIRECTOR, is n = 8(D, I , R , E , C , T , O , R )
10 !
× 60 = 5 × 10 !
Number of places = r = 8, two are alike. Hence 3!2!
number of, arrangement of the letters of the
12. There are (r = 6) letters (B , A , N , A , N , A ),
word DIRECTOR without any restriction is
8 since all the letters are to be taken, so that
P8 8!
= number of places to be filled up = r = 6.
2! 2!
Hence 3A’s are alike and 2N's are alike.
Now keeping three vowels (I, E, O) together
number of letters Hence number of different permutations of
= r = 6 {D, R , C , T , R (I , E , O )} and number of the letters of the word BANANA is
places r = 6. 6
P6 6! 6 × 5 × 4 × 3!
= = = = 60
Hence number of arrangements of these 6 is 3!2! 3!2! 3!2!
6
P6 6 !
= 13. Number of letters = n = 10 in the word
2! 2!
BENEVOLENT of which 3 E's alike 2 N's are
But the three vowels among themselves can alike. Also number of places to be filed up
be arranged in 3! ways. = r = 10
Hence, number of arrangements of letters of Hence number of arrangements
the word DIRECTOR, such that the three 10
6! P10 10 !
vowels are together = × 3! = = = 302400
2! 3!2! 3!2!
www.pdfworld.in
www.pdfworld.in
Combinatorics 469
But of these 302400 one word (arrangement is 15. (i) When there is no restriction, we have total
BENEVOLENT itself). numbers of boys and girls = 5 + 4 = 9,
Hence, number of rearrangements keeping one out of 9 fixed, remaining 8
can be seated at 9 round table in 8 P8 = 8!
= 302400 − 1
ways.
= 302399 (ii) In this case we first arrange the boys, for
Now for the 2nd part putting L at the end keeping one boy fixed remaining 4 boys
number of letters remain to be utilised = n = 9 can be seated in 4! ways. Also as all the 4
of which 3E's are alike and 2N's are alike. girls are to sit together we select one
Hence, number of word or arrangements' region and then 4 girls can be seated in
formed out of the letter of the word this region in 4! ways.
BENEVOLENT, ending in L is Hence, number of ways in which 5 boys
9 and 4 girls can be seated at a round table,
P9 9!
= so that all the 4 girls sit together is
3!2! 3! × 2!
4 ! ⋅ 5 C1 ⋅ 4 ! = 5 × 576 = 2880
= 30240
(iii) From case (i) we see that total numbers of
14. Number of letters in the word OUGHT arrangements 5 boys and 4 girls at a round
= n = 5. table is = 8! = 40320 and from case (ii)
number of arrangements of 5 boys and 4
But writing the letters alphabetically girls at a round table, so that all the 4 girls
G, H , O ,T ,U Now for the words beginning with are not together
G number of places to be filled up = r = 4 and
number of letters utilised = n = 4 = 40320 − 2880
Hence, number of words beginning with = 37440
G = 4 P4 = 4! (iv) For, no two girls sit together, we first
Similarly, number of words beginning with arrange 5 boys keeping one boy fixed
H = 4! remaining 4 boys can be arranged in 4!
Similarly, number of words beginning with ways.
O = 4! Since, no two girls are to sit together, so
Again for the words beginning with TG... that number of place for the girls (between
number of places to be filled up = r = 3 and every two boys) is 5 and hence girls can be
number of letters to be utilised = n = 3 arranged according to the condition of the
problem in 5! ways.
Hence, number of words beginning with
TG = 3P3 = 3 ! = 6 Hence required number of arrangements
in this case = 4 ! ⋅ 5 !
Similarly, number of words beginning with
= 2880
TH = 6
Again for the words beginning with TOG.. 16. Number of place = number of children taken
number of places remain to be filled up at a time = r = 3.
= r = 2 and number of letter remain to be Hence, 3 children out of 8 can be selected in
utilised = n = 2.
8!
Hence, number of words beginning with
8
C3 = = 56 ways
3!5!
TOG = 2P2 = 2!
Hence, the man has to go to the zoo 56 times.
Number of words beginning withTOH = 2!
In the 2nd part of the problem, a particular
Now, the words beginning with TOU and
child is to be included. Keeping the particular
TOUGH come.
child aside number of remaining children
Hence, rank of the word TOUGH in the
dictionary = n − 1 = 8− 1 = 7
= 24 + 24 + 24 + 6 + 6 + 2 + 2 + 1 And number of places to be filled up
= 89th r =3−1=2
www.pdfworld.in
www.pdfworld.in
Hence, number of selection of 3 children out ways of committees of 5 consisting of 1 ladies

of 8 children including a particular child and 4 gentelmen can be made in 4 C1 ⋅ 8C4
= 7C2 × 1C1 ways.
7! 7 × 6 × 5! Hence, total number of committees.
= =
5!2! 5!2!
= 4C1 ⋅ 8C4 + 4C2 ⋅ 8C3 + 4C3 ⋅ 8C2 + 4C4 ⋅ 8 C1
= 21
4 × 8! 4! 8! 4! 8! 8!
Hence, a particular child will go to the zoo 21 = + × + × +1×
number of times. 4! 4! 2!2! 3!5! 3!1! 2! 6! 1!7!
4 × 8 × 7 × 6 × 5 × 4! 2 × 3 × 4
17. Committees of 5 consisting of at least 3 ladies = +
4! 4! 4
can be made in the following ways. 8 × 7 × 6 × 5! 4 × 3! 8 × 7 × 6!
Committees of 5 consisting of 3 ladies and × + ×
3! × 5! 3!1! 2! 6!
man can be made in 4 C3 ⋅ 7C2 ways. 1 × 8 × 7!
+
Committees of 5 consisting of 4 ladies and 1 1! × 7!
man can be made in 4 C4 ⋅ 7C1 ways.
= 280 + 336 + 112 + 8 = 736
Hence, committees of 5 can be formed in
20. Since there are 3 seats and a voter can not
= 4C3 ⋅ 7C2 + 4C4 ⋅ 7C1
vote for more than 3 candidates.
= 4 × 21 + 1 × 7 = 84 + 7 = 91 Hence, the voter can vote for 1 candidate out
18. Three groups A, B and C contain respectively of 6 candidates in 6C1 ways. Voter can vote for
4, 5 and 6 question now to make a selection of 2 candidates out of 6 candidates in 6C2 ways.
7 question, selecting at least two question
And the vote can vote for 3 candidates out of
from each section, a candidate will have to
6 candidates in 6C3 ways.
make up his choice in the following ways.
He can answer 2 questions from group A 2 Finally total number of ways to vote
question from group B and 3 questions from = 6C1 + 6C2 + 6C3 = 6 + 15 + 20 = 41
group C in 4 C2 ⋅ 5C3 ⋅ 6C2 ways.
21. Let the number of candidates be n. Therefore
He can answer 2 questions from group A and
number of members to be elected = n − 2 and
3 questions from group B and 2 questions
hence one can vote atmost for n −2
from group C in 4 C2 ⋅ 5C2 ⋅ 6C3 ways.
candidates.
He can answer 3 question from group A 2 Hence, total number of ways in which one can
question from group B and 2 question from vote
group C in 4 C3 ⋅ 5C2 ⋅ 6C2 ways.
= n C1 + n C2 + n C3 + ... + n Cn −2 = 56
Hence, total number of way to answer all the
question or n
CO + (n C1 + n C2 + n C3 + ... + n Cn − 2)
= C2 ⋅ C2 ⋅ C3 + C2 ⋅ C3 ⋅ C2
4 5 6 4 5 6
+ Cn
n
−1 + Cn = 1 + 56 + n + 1
n
+ 4C3 ⋅ 5C2 ⋅ 6C2

(Q n C0 = n Cn = 1 and n Cn −1 = n)
= 6 × 10 × 20 + 6 × 10 × 15 + 4 × 10 × 15
= 1200 + 900 + 600 = 2700 or 2 = n + 58
n
…(i)
19. Committees of 5 consisting of at least one Now only n = 6 satisfies (i) so that number of
lady can be made in the following ways. candidates = n = 6.
Committees of 5 consisting of 2 ladies and 3 22. According to the problem, a voter has to vote
gentelmen can be made in 4 C2 ⋅ 8C3 ways. for at least one candidates, and at most 5
Committees of 5 consisting of 3 ladies and 2 candidates, the number of ways in which the
Gentelmen can be made in 4 C3 ⋅ 8C2. voter may vote is
= 10
C1 + 10
C2 + C3 +
10 10
C4 + 10
C5
Committees of 5 consisting of 4 ladies and
and 1 Gentelmen Can be made in 4 C4 ⋅ 8C1 = 10 + 45 + 120 + 210 + 252 = 637
www.pdfworld.in
www.pdfworld.in
Combinatorics 471
23. (i) Since, balls of same colour are also 25. (i) In this case, we have to select one book
different. So that we have total number of out of 3 books of Mathematics in 3C1 ways,
different balls. = 5 + 4 + 3 = 12 one book out of books of Science in 4
C1
In this case we have to select some balls ways
(i.e., zero or some or all 12 balls). and one book out of 5 books of literature
Now, one ball out of 12 can be selected in in 5C1 ways.
12
C1 ways. Hence, number of different collections of
12
2 bals out of 12 can be selected in C2 one (exactly one) book of each subject
ways. = 3C1 ⋅ 4C1 ⋅ 5C1 = 3 ⋅ 4 ⋅ 5 = 60
Similarly, 12 balls out of 12 can be
(ii) Here we have to select at least one book of
selected in 12C12 ways.
each subject out of 3 books of
Hence, number of selection of some balls Mathematics, 4 books of Science and 5
= C1 +
12 12
C2 + ... + 12
C12 books of Literature. Here it is not given
how many books are to be selected. Hence,
= ( C0 +
12 12
C1 + C2 + ... +
12
C12 ) − 1
12
we have to select some books containing
one book of each subject out of 3 books of
= 212 − 1 Mathematics, 4 book of Science and 5 of
(ii) Some balls containing at least one red and Literature, which can be selected in
one green balls are to be selected. ( 3C1 + 3C2 + 3C3 ) ( 4C1 + 4C2 + 4C3 + 4C4 )
Now 1 or some red balls can be selected in × ( 5C1 + 5C2 + ... + 5C5 )
5
C1 + ... + 5C5 = 25 − 1 = 31 ways = ( 23 − 1) ( 24 − 1) ( 25 − 1)
Also 1 or some green balls out of 4 green (iii) In this case, at least one book of literature
balls can be selected in is to be selected. Hence, we have to select
4
C1 + 4C2 + ... + 4C4 = 24 − 1 = 15 ways. 0 or 1, or 2 or 3 books from 3 books of
mathematics, 0 or 1 or 2 or 3 or 4 books of
Now zero or some blue balls out of 3 blue science; 1 or 2 or 3 or 4 or 5 books from 5
balls can be selected in books of literature which can be selected
3
C0 + 3C1 + ... + 3C3 = 23 ways. in
(3C0 + 3C1 + 3C2 + 3C3 )(4 C0 + 4C1
Hence, total number of selections of balls
containing at least one red and one green + 4C2 + 4C3 + 4C4 ) (5C1 + 5C2 + ... + 5C5 )
balls.
= 23.24 (25 − 1) = 1282
( 5 − 1)
= 31 × 15 × 23 = 31 × 15 × 8
26. The number of 7-digit numbers with 1 in the
= 3720 left most place and containing each of the
24. Since, at least one red ball must be drawn, so digits 1, 2, 3, 4, 5, 6, 7 exactly once is 6! = 720.
that number of selectin of 1 or some or all 4 But 120 of these end in 5 and hence are
red balls (all different) divisible by 5. Thus, the number of 7-digit
numbers with 1 in the left most place and
= 4C1 + 4C2 + K + 4C4
containing each of the digits 1, 2, 3, 4, 5, 6, 7
= 24 − 1 = 15 exactly once but not divisible by 5 is 600.
Similarly, the number of 7-digits numbers
Also the number of selection of zero, some with 2 and 3 in the left most place and
green balls from 3 different green balls containing each of the digits 1, 2, 3, 4, 5, 6, 7
= 3C0 + 3C1 + K + 3C3 = 23 = 8 exactly once but not divisible by 5 is also 600
each. These account for 1800 numbers.
Hence, total number of required selection Hence, 2000th number must have 4 in the left
= 15 × 8 = 120 most place.
www.pdfworld.in
www.pdfworld.in
Again the number of such 7-digit numbers  (7k + 1)... (7k6 + 6) – 6 ! 

=k  1 
beginning with 41, 42 and not divisible by 5 is  6! 
120 – 24 = 96 each and these account for 192
k [7t + 6 ! – 6 !]
numbers. This shows that 2000th number in =
the list must begin with 43. 6!
7tk
The next 8 numbers in the list are 4312567, =
4312576, 4312657, 4312756, 4315267, 6!
4315276, 4315627 and 4315672. Thus, We know that, Q is an integer and so 6! divides
2000th number in the list is 4315672. 7tk. Since, GCD (7, 6 !) = 1, even after
cancellation there is a factor of 7 still left in
27. We observe that   = 
x x 
= 0, if and only the numerator. Hence, 7 dividesQ, as desired.
 99   101 
if x ∈ {1, 2, 3,..., 98} and there are 98 such 29. We count by brute force considering the cases
numbers. If we want   = 
x x 
= 1, then x x = 0, x = 1,... , x = 33. Observe that the least
 99   101  value x can take is zero and its largest value is
should lie in the set {101, 102,...,197}, which 33.
accounts for 97 numbers. In general, if we x = 0. If y = 0, then z ∈ {0, 1, 2,... ,100}; if y = 1,
z ∈ {1, 2,... ,99}; y = 2,
require   = 
x x  then if then
= k , where k ≥ 1, then x
 99   101  z ∈ {2, 3,... , 98} and so on. Finally if y = 50,
must be in the set {101k, 101k + 1,..., then z ∈ {50}. Thus, there are altogether
99(k + 1) − 1} and there are 99 – 2k such 101 + 99 + 97 + ...+ 1 = 512 possibilities
numbers. Observes that this set is not empty x = 1. Observe that y ≥ 1. If y = 1, then
only if 99(k + 1) – 1 ≥ 101k and this z ∈ {1, 2,... ,98}; if y = 2, then z ∈ {2, 3, ... ,97}; if
requirement is met only if k ≤ 49. Thus, the y = 3, then z ∈ {3, 4,... ,96} and so on. Finally if
total number of positive integers x for which y = 49, then z ∈ {49, 50}. Thus, there are
 x  =  x  is given by altogether 98 + 96 + 94 + ... + 2 = 49 ⋅ 50
 99   101  possibilities.
49 General case. Let x be even, say, x = 2k ,
98 + ∑ (99 – 2k ) = 2499 0 ≤ k ≤ 16. If y = 2k , then
k =1
z ∈ {2k , 2k + 1,... , 100 − 4k; if y = 2k + 1, then
Remark z ∈ {2k + 1, 2k + 2,... , 99 – 4k); if y = 2k + 2,
For any m ≥ 2 the number of positive integers x then z ∈ {2k + 2, 2k + 3,... , 99 – 4k} and so on.
2
 x   x  m –4 Finally, if y = 50 – k , then z ∈ {50 – k}. There are
such that = is if m is even and
m –1 m + 1 4 altogether
m2 – 5 (101 – 6k ) + (99 – 6k ) + (97 – 6k ) + ... + 1
, if m is odd. = (51 – 3k )2 possibilities.
4
Let x be odd, say, x = 2k + 1, 0 ≤ k ≤ 16. If
 n  n (n – 1)(n – 2)... (n – 6)
28. We have,   = y = 2k + 1, then z ∈ {2k + 1, 2k + 2, . . . , 98 − 4k};
 7 7! if y = 2k + 2, then
In the numerator, there is a factor divisible by z ∈ {2k + 2, 2k + 3, . . . , 97 − 4k}; if y = 2k + 3,
7 and the other six factors leave the then z ∈ {2k + 3, 2k + 4, . . . , 96 − 4k}; and so
remainders 1, 2, 3, 4, 5, 6 in some order when on.
divided by 7. Finally, if y = 49 − k , then z ∈ {49 − k , 50 − k}.
Hence, the numerator may be written as There are altogether
7k ⋅ (7k1 + 1) ⋅ (7k2 + 2)... (7k6 + 6). (98 − 6k ) + (96 − 6k ) + (94 − 6k ) + . . . + 2
n  = (49 − 3k )(50 − 3k )
Also, we conclude that   = k, as in the set
p  possibilities.
{n , n – 1,... , n – 6}, 7k is the only number which The last two cases would be as follows :
is a multiple of 7. If the given number is called x = 32 : if y = 32, then z ∈ {32, 33, 34, 35, 36}; if
Q , then
y = 33, then z ∈ {33, 34, 35}; if y = 34, then
(7k + 1)(7k2 + 2)... (7k6 + 6)
Q = 7k ⋅ 1 −k z ∈ {34}; altogether 5 + 3 + 1 = 9 = 32
7!
possibilities.
www.pdfworld.in
www.pdfworld.in
Combinatorics 473
x = 33 : if y = 33, then z ∈ {33, 34}; only 2 = 12

. 100 (n + 3)2 115
T = Σ −
possibilities. n =0 12 12
Thus the total number of triples, say T , is (12 + 22 + 32 + . . . + 1032 ) − (12 + 22 ) 115
given by, = −
12 12
16 16
T = Σ (51 − 3k )2 + Σ (49 − 3k )(50 − 3k ). 103 ⋅ 104 ⋅ 207 5 115
k=0 k=0 = − −
6 ⋅ 12 12 12
Writing this in the reverse order, we obtain
17 17
= 30787.
T = Σ (3k )2 + Σ (3k − 2)(3k − 1)
k =1 k=0 30.
17 17 A B
= 18 Σ k 2 − 9 Σ k + 34
k =1 k =1 2
3
17 ⋅ 18 ⋅ 35   17 ⋅ 18 + 34
= 18   −9  1
 6   2 
4 5
= 30, 787.
C
Thus the answer is 30787. 6
Aliter X 7
It is known that the number of ways in which
a given positive integer n ≥ 3 can be expressed Let X = {1, 2, 3,... , n}. We use Venn diagram for
as a sum of three positive integers x , y , z (that sets A, B, C to solve the problem. The regions
is, x + y + z = n), subject to the condition other than A ∩ B ∩ C (which is to be empty)
n 2  are numbered 1, 2, 3, 4, 5, 6, 7 as shown in the
x ≤ y ≤ z is  , where {a} represents the
figure; e.g., 1 corresponds to
12 
A / (B ∪ C ) – A ∩ Bc ∪ C c , corresponds to
integer closest to a. If zero values are allowed
for x , y , z then the corresponding count is A ∩ B / C = A ∩ B ∩ C c , 7 corresponds to
 (n + 3)2  X / (A ∪ B ∪ C ) = A c ∩ Bc ∩ C c , since
 , where now n ≥ 0.
 12  A ∩ B ∩ C = φ.
Since in our problem Firstly the number of ways of assigning

n = x + y + z ∈ {0, 1, 2, . . . , 100}, the desired elements of X to the numbers regions without
answer is any condition is 7n . Among these there are
cases in which 2 or 5 or both are empty. The
100  (n + 3)2 
Σ  . number of distributions in which 2 is empty is
n =0
 12  6n . Likewise the number of distributions in
For n = 0, 1, 2, 3, . . . , 11, the corrections for {} which 5 is empty is also 6n . But then we have
to get the nearest integers are subtracted twice the number of distributions
3 −4 −1 −1 −4 3 −4 −1 −1 −4 in which both the regions 2 and 5 are empty.
, , , 0, , , , , , 0, ,
12 12 12 12 12 12 12 12 12 12 So, to compensate we have to add the number
So, for 12 consecutive integer values of n, the of distributions in which both 2 and 5 are
sum of the corrections is equal to empty. This is 5n . Hence, the desired number
 3 − 4 − 1 − 0 − 1 − 4 − 3  × 2 = −7 . of triples (A , B , C ) in
  7n – 6n – 6n + 5n = 7n – 2 ⋅ 6n + 5n .
 12  6
Since,
101
= 8+
5
, there are 8 sets of 12 31. Putting a – 1 = p , b – 1 = q and c – 1 = r , the
12 12 equation may be written in the form
consecutive integers in {3, 4, 5, . . . , 103} with pqr = 2( p + q + r ) + 4,
99, 100, 101, 102, 103 still remaining. Hence
the total correction is where p , q , r are integers such that
0 ≤ p ≤ q ≤ r. Observe that p = 0 is not possible,
 −7  × 8 + 3 − 4 − 1 − 0 − 1 = −28 − 1 = −115.
  for then 0 = 2( p + q ) + 4 which is impossible
 6 12 3 4 12
in non-negative integers. Thus, we may write
So the desired number T of triples (x , y , z ) is this in the form
equal to
www.pdfworld.in
www.pdfworld.in
 1 1 1 4 (vii) Here again there are 33 numbers; ab must

2 + +  + =1
 pq qr rp  pqr be from the set {01, 04, 07,...,97}.
Adding all these we get
If p ≥ 3, then q ≥ 3 and r ≥ 3. Then, left side is
bounded by 6/9 + 4/27 which is less than 1. 300 + 30 + 30 + 30 + 30 + 34 + 33 = 487
We conclude that p = 1 or 2. numbers.
However this is not the correct figure as there
Case I Suppose p = 1. Then, we have
is over counting. Let us see how much over
qr = 2(q + r ) + 6 or (q – 2)(r − 2) = 10. This gives counting is done by looking at the
q − 2 = 1, r − 2 = 10 or q − 2 = 2 and r − 2 = 5 intersection of each pair of categories. A
(recall q ≤ r ). This implies ( p , q , r ) = (1, 3, 12), number in (i) obviously cannot lie in (ii), (iv) or
(1, 4, 7). (vi) as is evident from the last digit. There
Case II If p = 2, the equation reduces to cannot be a common number in (i) and (iii) as
2qr = 2(2 + q + r ) + 4 or qr = q + r + 4. This any two such numbers differ in the 4th digit.
reduces to (q − 1) (r − 1) = 5. Hence, q − 1 = 1 If a number belongs to both (i) and (v), then
and r − 1 = 5 is the only solution. This gives such a number of the form a1515. This, is
(p , q , r ) = (2, 2, 6). divisible by 3 only for a = 3, 6, 9. Thus there
are 3 common numbers in (i) and (ii). A
Reverting back to a , b , c , we get three triples:
number which is both in (i) and (vii) is of form
(a , b , c ) = (2, 4, 13), (2, 5, 8), (3, 3, 7).
15 c15 and divisibility by 3 gives c = 0, 3, 6, 9;
32. Any such number should be both divisible by thus we have 4 numbers common in (i) and
5 and 3. The last digit of a number divisible by (vii). That exhaust all possibilities with (i).
5 must be either 5 or 0. Hence, any such Now (ii) can have common numbers with only
number falls into one of the following seven categories (iv) and (vi). There are no numbers
categories: common between (ii) and (vi) as evident from
(i) abc15 3rd digit. There is only one number common
(ii) ab150 to (ii) and (vi), namely 15150 and this is
(iii) ab155 divisible by 3. There is nothing common to (iii)
and (v) as can be seen from the 3rd digit. The
(iv) a15b0
only number common to (iii) and (vii) is 15155
(v) a15b5 and this is not divisible by 3. It can easily be
(vi) 15ab 0 inferred that no number is common to (iv)
(vii) 15ab5 and (vi) by looking at the 2nd digit. Similarly
Here, a , b , c are digit. Let us count how many no number is common to (v) and (vii). Thus
numbers of each category are there. there are 3 + 4 + 1 = 8 numbers which are
counted twice.
(i) In this case a ≠ 0 and the 3-digit number
abc is divisible by 3, and hence one of the We conclude that the number of 5-digit
numbers in the set {102, 105, ..., 999}. numbers which contain the block 15 and
This gives 300 numbers. divisible by 15 is 487 – 8 = 479.
(ii) Again a number of the form ab150 is 33. Since each digit occurs at least twice, we have
divisible by 15 if and only if the 2-digit following possibilities
number ab is divisible by 3. Hence, it 1. Three digits occur twice each. We may
must be from the set {12, 15, ..., 99}. There choose three digits from {1, 2, 3, 4, 5} in
are 30 such numbers.  5
(iii) As in (ii), here are again 30 numbers.   = 10 ways. If each occurs exactly
 3
(iv) Similar to (ii); 30 numbers twice, the number of such admissible
(v) Similar to (ii), 30 numbers. 6-digit numbers is
(vi) We can begin the analysis of the number 6!
× 10 = 900
of the form 15 ab0 as in (ii). Here, again ab 2!2!2!
as a 2-digit number must be divisible by 3,
2. Two digits occur three times each. We can
but a = 0 is also permissible. Hence, it  5
must be from the set {00, 03, 06,...,99}. choose 2 digits in   = 10 ways
 2
There are 34 such numbers.
www.pdfworld.in
www.pdfworld.in
Combinatorics 475
Hence, the number of admissible 6-digit But the condition that the triangle is obtuse
6! leads to (1004 – 2k )2 > 2(502 + k )2.
numbers is × 10 = 200
3!3! This simplifies to 5022 + k 2 − 6(502)k > 0
3. One digit occurs four times and the other Solving this quadratic inequality for k, we see
twice. We are choosing two digits again, that
which can be done in 10 ways. The two
digits are interchangeable. Hence, the k < 502(3 – 2 2 ) or k > 502(3 + 2 2 ).
desired number of admissible 6-digit Since, k ≤ 501, we can rule out the second
numbers is possibility. Thus k < 502(3 – 2 2 ), which is
6! approximately 861432
. . We conclude that
2× × 10 = 300
4!2! k ≤ 86. Thus, we get 86 triangles
(y , x , x ) = (1004 – 2k , 502 + k , 502 + k ),
4. Finally all digits are the same. There are 5
such numbers. k = 1, 2, 3, ... ,86.
Thus, the total number of admissible The last obtuse triangle in this list is
numbers is 900 + 200 + 300 + 5 = 1405. (832, 588, 588). (It is easy to check that
34. Since 20, b , c are in harmonic progression, we 8322 – 5882 – 5882 = 736 > 0, where as
have 8302 – 5892 – 5892 = –4942 < 0. )
1 1 2
+ = , 36. We observe that 0 + 1 + 2 + 3 = 6. Hence, the
20 c b remaining two digits must account for the
which reduces to bc + 20b − 40c = 0. This may sum 4. This is possible with 4 = 0 + 4 = 1 + 3
also be written in the form = 2 + 2. Thus we see that the digits in any such
(40 – b )(c + 20) = 800. 6-digit number must be from one of the
Thus, we must have 20 < b < 40 or collections:
equivalently, 0 < 40 − b < 20. Let us consider {0, 1, 2, 3, 0, 4}, {0, 1, 2, 3, 1, 3} or {0, 1, 2, 3,
the factorisation of 800 in which one term is 2, 2}.
less than 20
Consider the case in which the digits are from
(40 − b ) (c + 20) = 800 = 1 × 800 = 2 × 400 the collection {0, 1, 2, 3, 0, 4}. Here 0 occurs
= 4 × 200 twice and the digits 1, 2, 3, 4 occur once each.
But 0 cannot be the first digit. Hence, the first
= 5 × 160 = 8 × 100
digit must be one of 1, 2, 3, 4. Suppose we fix 1
= 10 × 80 = 16 × 50 as the first digit. Then, the number of 6-digit
We thus get the pairs numbers in which the remaining 5 digits are
0, 0, 2, 3, 4 is 5 !/ 2 ! = 60. Same is the case with
(b , c ) = (39, 780), (38, 380), (36, 180), (35, 140), other digits: 2, 3, 4. Thus the number of
(32, 80), (30, 60), (24, 30). 6-digit numbers in which the digits 0, 1, 2, 3,
Among these 7 pairs, we see that only 5 pairs 0, 4 occur is 60 × 4 = 240.
(39, 780), (38, 380), (36, 180), (35, 140), Suppose the digits are from the collection
(30, 60) fulfill the condition of divisibility: b {0, 1, 2, 3, 1, 3}. The number of 6-digit
divides c. Thus, there are 5 triples satisfying numbers beginning with 1 is 5 !/ 2 ! = 60. The
the requirement of the problem. number of those beginning with 2 is
35. Let the sides be x , x , y , where x , y are positive 5 !/ (2 !) (2 !) = 30 and the number of those
integers. Since we are looking for obtuse beginning with 3 is 5 !/ 2 ! = 60. Thus the total
angled triangles, y > x. Moreover, number in this case is 60 + 30 + 60 = 150.
2x + y = 2008 shows that y is even. But Alternately, we can also count it as follows:
y < x + x , by triangle inequality. Thus, the number of 6-digit numbers one can obtain
y < 1004. Thus, the possible triples are ( y , x , x ) from the collection {0, 1, 2, 3, 1, 3} with 0 also
= (1002, 503, 503), (1000, 504, 504), as a possible first digit is 6 !/(2 !)(2 !) = 180; the
(998, 505, 505) and so on. The general form is numbers of 6-digit number one can obtain
(y , x , x ) = (1004 – 2k , 502 + k , 502, + k ), where from the collection {0, 1, 2, 3, 1, 3} in which 0
k = 1, 2, 3, ... , 501. is the first digit is 5 !/(2 !) (2 !) = 30. Thus the
www.pdfworld.in
www.pdfworld.in
number of 6-digit numbers formed by the hundred's place is never 0. Suppose the digit
collection {0, 1, 2, 3, 1, 3} such that no number in unit’s place is 0. There are 4 × 5 = 20 such
has its first digit 0 is 180 – 30 = 150. Finally numbers. Similarly, 20 numbers each having
look at the collection {0, 1, 2, 3, 2, 2}. Here the digits 2, 4, 6, 8 in their unit’s place. Thus, the
number of of 6-digit numbers in which 1 is sum of the digits in unit’s place of all the
first digit is 5 !/ 3 ! = 20; the number of those numbers in E is
having 2 as the first digit is 5!/2! = 60; and the 20 (0 + 2 + 4 + 6 + 8) = 20 × 20 = 400
number of those having 3 as the first digit is
5!/3! = 20. Thus the number of admissible A similar reasoning shows that the sum of the
6-digit numbers here is 20 + 60 + 20 = 100. digits in ten’s place of all the numbers in E is
This may also be obtained using the other 400, but the sum of the digits in hundred’s
method of counting: place of all the numbers in E is 25 × 20 = 500.
6!/3! − 5!/3! = 120 − 20 = 100. Thus the sum of all the numbers in E is
Finally the total number of 6-digit numbers in 500 × 102 + 400 × 10 + 400 = 54400
which each of the digits 0, 1, 2, 3 appears at
The required sum is
least once is
240 + 150 + 100 = 490. 494550 – 69375 – 54400 = 370775
37. Let X denotes the set of all 3-digit natural 38. Let us examine the first few natural numbers:
numbers; let O be those numbers in X having 1, 2, 3, 4, 5, 6, 7, 8, 9. Here we see that an = 1, 2,
only odd digits; and E be those numbers in X 3, 2, 2, 3, 3, 4, 3. We observe that an ≤ an +1 for
having only even digits. Then X /(O ∪ E ) is the all n except when n + 1 is a square in which
set of all 3-digit natural numbers having at case an > an +1. We prove that this observation
least one odd digit and at least one even digit. is valid in general.
The desired sum is therefore Consider the range
∑x − ∑ y − ∑ z. m 2 , m 2 + 1, m 2 + 2,... , m 2 + m ,
x ∈X y ∈O x ∈E m 2 + m + 1,... , m 2 + 2m.
It is easy to compute the first sum
Let n take values in this range so that
999 99
999 × 1000 99 × 100
∑x = ∑j − ∑k =
2
−
2
n = m 2 + r , where 0 ≤ r ≤ 2m. Then we see that
x ∈X j =1 k =1 [ n ] = m and hence
= 50 × 9891 = 494550  n  m 2 + r  r 
 [ n ] =  m  = m +  m 
Consider the set O. Each number in O has its    
digits from the set {1, 3, 5, 7, 9}. Suppose the
digit in unit’s place is 1. We can fill the digit in Thus an takes the value
ten's place in 5 ways and the digit in m , m , m ,... , m ,
1442443
hundred’s place in 5 ways. Thus there are 25 m times
numbers in the set O each of which has 1 in its m + 1, m + 1, m + 1,... , m + 1, m + 2,
144444424444443
unit’s place. Similarly, there are 25 numbers m times
whose digit in unit’s place is 3; 25 having its
digit in unit’s place as 5; 25 with 7 and 25 with in this range. But when n = (m + 1)2 , we see
9. Thus the sum of the digits in unit’s place of that an = m + 1. This shows that an –1 > an
all the numbers in O is whenever n = (m + 1)2. When we take n in the
set {1, 2, 3,...,2010}, we see that the only
25 (1 + 3 + 5 + 7 + 9) = 25 × 25 = 625
squares are 12 , 22 ,.... , 442 (since 442 = 1936
A similar argument shows that the sum of
and 452 = 2025) and n = (m + 1)2 is possible
digits in ten’s place of all the numbers in O is
for only 43 values of m. Thus an > an +1 for 43
625 and that in hundred’s place is also 625.
values of n. (These are 22 – 1, 32 – 1,... ,442 – 1.)
Thus the sum of all the numbers in O is
625 (102 + 10 + 1) = 625 × 111 = 69375 39. Let x , y , z be three distinct positive integers
satisfying the given conditions.
Consider the set E. The digits of numbers in E We may assume that x < y < z. Thus we have
are from the set {0, 2, 4, 6, 8}, but the digit in three relations:
www.pdfworld.in
www.pdfworld.in
Combinatorics 477
1 1 a conditions. (We may observe (2, 4, 5) is also

+ = ,
y z x not a solution.) Since 3 + 4 + 5 = 12 > 11 = 2 + 3
+ 6, it follows that (2, 3, 6) has the required
1 1 b
+ = , minimality.
z x y
40. We divide the even 4-digit numbers having
1 1 c
+ = , non-zero digits into 4 classes those ending in
x y z 2, 4, 6, 8.
for some positive integers a , b , c . Thus (A) Suppose a 4-digit number ends in 2. Then
1 1 1 the second right digit must be odd in
+ + , order to be divisible by 4. Thus the last 2
x y z
digits must be of the form 12, 32, 52, 72 or
a+1 b+1 c+1
= = = r, 92. If a number ends in 12, 52 or 92, then
x y z the previous digits must be even in order
say. Since x < y < z , we observe that a < b < c . not to be divisible by 8 and we have 4
We also get admissible even digits. Now the left most
digit of such a 4-digit number can be any
1 r 1 r 1 r
= , = , = non-zero digit and there are 9 such ways
x a+1 y b+1 z c+1 and we get 9 × 4 × 3 = 108 such numbers. If
Adding these, we obtain a number ends in 32 or 72, then the
previous digit must be odd in order not to
1 1 1 r r r
r = + + = + + be divisible by 8 and we have 5 admissible
x y z a+1 b+1 c+1 odd digits. Here again the left most digit
1 1 1 of such a 4-digit number can be any
or + + =1 …(i)
a+1 b+1 c+1 non-zero digit and there are 9 such ways,
and we get 9 × 5 × 2= 90 such numbers.
Using a < b < c , we get
Thus the number of 4-digit numbers
1 1 1 3
1= + + < having non-zero digits, ending in 2,
a+1 b+1 c+1 a+1 divisible by 4 but not by 8 is 108 + 90 =
Thus, a < 2. We conclude that a = 1. Put this 198.
in the relation (i), we get (B) If the number ends in 4, then the previous
1 1 1 1 digit must be even for divisibility by 4.
+ =1− = Thus the last two digits must be of the
b+1 c+1 2 2
form 24, 44, 54, 84. If we take numbers
1 2
Hence, b < c gives < ending with 24 and 64, then the previous
2 b+1 digit must be odd for non-divisibility by 8
Thus, b + 1 < 4 or b < 3. Since b > a = 1, we and the left most digit can be any
must have b = 2. This gives non-zero digit. Here we get 9 × 5 × 2 = 90
1 1 1 1 such numbers. If the last two digits are of
= − = , the form 44 and 84, then previous digit
c+1 2 3 6
must be even for non-divisibility by 8. And
or c = 5. Thus x : y : z = a + 1 : b + 1 : c + 1 the left most digit can take 9 possible
= 2 : 3 : 6. Thus the required numbers with the values. We thus get 9 × 4 × 2 = 72 numbers.
least sum are 2, 3, 6. Thus the admissible numbers ending in 4
Aliter We first observe that (1, a , b) is not a is 90 + 72 = 162.
solution whenever 1 < a < b. Otherwise we (C) If a number ends with 6, then the last two
1 1 l digits must be of the form 16, 36, 56, 76,
should have + = = l for some integer l.
a b 1 96. For numbers ending with 16, 56, 76,
a+b the previous digit must be odd. For
Hence, we obtain = l showing that a| b
ab numbers ending with 36, 76, the
and b| a. Thus a = b contradicting a ≠ b. Thus previous digit must be even. Thus we get
the least number should be 2. It is easy to her (9 × 5 × 3) + (9 × 4 × 2) = 135 + 72 = 207
verify that (2, 3, 4) and (2, 3, 5) are not numbers.
solutions and (2, 3, 6) satisfies all the
www.pdfworld.in
www.pdfworld.in
(D) If a number ends with 8, then the last two 1000a + 100b + 10c + 2,
digits must be of the form 28, 48, 68, 88.
1000a + 100b + 10c + 4,
For numbers ending with 28, 68, the
previous digit must be even. For numbers 1000a + 100b + 10c + 6,
ending with 48, 88, the previous digit
1000a + 100b + 10c + 8,
must be odd. Thus we get (9 × 4 × 2) + (9 ×
5 × 2) = 72 + 90 = 162 numbers there is exactly one among these four
Thus the number of 4-digit numbers, which is divisible by 4 but not by 8. Now
having non-zero digits, and divisible by 4 we can divide the set of all 4-digits even
but not by 8 is 198 + 162 + 207 + 162 = numbers with non-zero digits into groups
729. of 4 such consecutive even numbers with
Aliter If we take any four consecutive a, b, c non-zero. And is each group, there
even numbers and divide them by 8, we is exactly one number which is divisible by
get remainders 0, 2, 4, 6 in some order. 4 but not by 8. The number of such groups
Thus there is only one number of the form is precisely equal to 9 × 9 × 9 = 729, since
8k + 4 among them which is divisible by 4
we can vary a , b , c in the set {1, 2, 3, 4, 5, 6,
but not by 8. Hence, if we take four even
7, 8, 9}.
consecutive numbers
Level 2
1. If the sum is x , then Finally, since
x (10 !) = + + | S (a ) ∪ S (b ) ∪ S (c ) ∪ S (d ) ∪ S (e )| ≤ 5 × 393
10C1 10C3 10C3
+ + + = 1965 × 1989,
10C5 10C7 10C9.
∃f ∈ E such that a , b , c , d , e and f from a
Since the sum of the odd-positioned binomial
sextuple of mutual acquaintances.
coefficients is equal to the sum of the
even-positioned ones in any line of Pascal’s 3. Let M k denote the number of permutations
triangle (this should be known to students, (p1 ,.... , p k ) of 1, 2, ...., k such that for any
and is established by expanding (1 − 1)n ), the i < k , (p1 ,... , p1 ) is not a permutation of 1, 2..., i.
Then, (n − k ) M k is the number of
above sum is half of 10C0 + 10 C1 + ... + 10 C10
permutations (p1 ,.... , pn ) of 1, 2, ... , n in which
(which equals 210). Thus this sum is 29, and
k is the least integer such that (p1 ,.... , p k ) is a
x = 29/10 ! . We have (a , b ) = (9, 10). permutation of 1, 2,..., k. Hence,
n −1
2. Let E denote the set of these 1994 employees.
For each x ∈ E , let S (x ) denote the set of all ∑ M k (n − k )!
k =1
employees whom x does not know. Then, by
assumption, | S (x )| = 393 for all x ∈ E . Let a and is the total number of permutations of
b be any two employees who know each other. (1, 2, ..., n) in which there is a k < n such that
Since (p1 ,... , p k ) is a permutation of 1, 2,..., k.
Hence,
| S (a ) ∪ S (b )| ≤ 2 × 393 = 786 < 1992,
n −1
∃c ∈ E such that a , b and c form a triple of Mn = n ! ∑ M k (n − k )!
mutual acquaintances, since k =1
| S (a ) ∪ S (b ) ∪ S (c )| ≤ 3 × 393 = 1179 < 1991, We need M 6 clearly

∃d ∈ E such that, a , b , c and d form a M1 = 1
quadruple of mutual acquaintances.
M2 = 2! − 1 = 1
Since
M3 = 3 ! − (2 ⋅ 1 + 1 ⋅ 1) = 3
| S (a ) ∪ S (b ) ∪ S (c ) ∪ S (d )| ≤ 4 × 393
M 4 = 4 ! − (3 + 2 + 6) = 13
= 1572 < 1990,
M5 = 5 ! − (13 + 2 ⋅ 3 + 6 ⋅ 1 + 24)
∃ e ∈ E such that, a , b , c , d and e form a
quintuple of mutual acquaintances. = 71
www.pdfworld.in
www.pdfworld.in
Combinatorics 479
M 6 = 6 ! – (71 + 2 ⋅ 13 + 6 ⋅ 3 + 24 ⋅ 1 + 120 ⋅ 1) m 0 ≥ 4 × 252 + 1 = 1009, at least 5 girls must

= 461 be assigned the same ordered pair, producing
5 girls and 5 boys for which each girl knows
4. We use a combinatorial argument to establish each boy, or no girl knows any of the boys.
the obviously equivalent identity
∞
7. (a) Let us denote the digits in the hundreds,
 n – r + 1  r − 1   n 
∑  d   r − d  =  r  …(*) tens and ones places of the sought-for
number N as x , y , and z respectively; then
d =1
we have N = 100x + 10y + z. The condition
where k = min (r , n − r + 1). It clearly suffices
of the problem yields the relation
to demonstrate that the left hand side of (i)
counts the number of ways of selecting r 100x + 10y + z = x ! + y ! + z !
objects from n distinct object (without Since, 7 ! = 5040 is a four-digit number,
replacements). Let | S 2| = r − 1. For each fixed none of the digits of the number N can
d = 1, 2,... , k , any selection of d object from exceed 6. Consequently, the number N
S1 (S/S 2 ) together with any selection of r − d itself does not exceed 700, whence it
objects from S 2 would yield a selection of r follows that none of its digits can exceed 5
objects from S. The total number of such (because 6! = 720 > 700) . Further, at least
 n − r + 1  r − 1  one digit of the number N is equal to 5
electrons is    . Conversely,
 d  r − d  because N is a three-digit number and
each selection of r objects from S clearly 3 ⋅ 4 ! = 72 × 100. It is clear that x cannot be
much arise in this manner. equal to 5 since we have 3 ⋅ 5 ! = 360 < 500. It
Summing over d = 1, 2,.... (*) follows. also follows that x cannot exceed 3.
Further, we can assert that x does not
m + n  exceed 2 since 3! + 2 ⋅ 5! = 246 < 300.
5. With k + n ≤ m , we represent   by
n+k Further, the number 255 does not satisfy
taking j element from n, and the remaining the condition of the problem, and if only
(n + k ) − j elements from m , for every j one digit of the sought-for number is
0 ≤ j ≤ n. Thus equal to 5, then x cannot exceed 1 because
2! +5! + 4! = 146 < 200. Moreover, since
m + n  n
n   m 

n+k
 = ∑  j   n + k − j  1! + 5! + 4! = 145 < 150 we conclude that y
j =0 cannot exceed 4; consequently z is equal to
5 because at least one of the digits of the
n
 n  m 
= ∑  n − i   k + i  number N must be equal to 5. Thus, we
i =0 have x = 1, 4 ≥ y > 0 and z = 5, which allows
n
n   m  us to easily find the single solution of the
= ∑  i   k + i  , problem: n = 145.
i =0
(b) The sought-for number N cannot consist
where we substituted i = n − j , and used of more three digits because even
n   n  4 ⋅ 92 = 324 is a three-digit number. This
  = 
 i  n − a  allows us to write N = 100x + 10y + z
where x , y and z are the digits of the
6. We will show that we can take n0 = 9. For
number N ; here x can be equal to 0 and it is
n0 ≥ 9, observe that for each girl there must be
even possible that x and y are
at least 5 boys whom she does not know. We
associate to each girl an ordered pair, the first simultaneously equal to 0.
element of which is a subset of 5 of, the boys The condition of the problem implies
all of whom she knows or all of whom she 100x + 10y + z = x 2 + y 2 + z 2 whence
does not know, and the second element of (100 − x )x + (10 − y )y = z (z − 1) …(*)
which is 0 or 1 according as she knows the From the last equality if follows that x = 0
boys or not. because, if otherwise, the number on the
There are   × 2 = 252 such pairs. Invoking
9 left-hand side of the equality would not be
 5 less than 90 (in case x ≥ 1 we have
the Pigeonhole Principle, if 100 − x ≥ 90 and (10 − y ) y ≥ 0) whereas the
www.pdfworld.in
www.pdfworld.in
k
number on the right-hand side would not
be greater than 9 ⋅ 8 = 72 (since z ≤ 9). ∑mj = 100, ∑ m j ml = 2002,
j =1
Consequently, Eq. (*) has the form
(10 − y )y = z (z − 1). It can easily be verified for an affirmative answer to the given
that the last equality cannot be fulfilled for question.
any positive integers z and y not exceeding We observe that
9 unless y ≠ 0. If y = 0 we have a single 2
solution: it is obvious that in this case z = 1. k  k 
∑ m j =  ∑ m j  − 2(∑ m jml )
2
Thus, the only number satisfying the
j =1  j =1 
condition of the problem is N = 1.
8. Since a jj had to exceed all the numbers in the = 1002 − 2(2002) = 5996
top left j × j submatrix (excluding itself), and Thus we have to choose m1 , m 2 ,... , m k
since there are j 2 − 1 entries, we must have such that
a jj ≥ j 2. Similarly, a jj must not exceed each of k k
∑mj = 100, ∑mj = 5996
2
the numbers in the bottom right
j =1 j =1
(n − j + 1) × (n − j + 1) submatrix (other than
itself) and there are (n − j + 1)2 − 1 such We observe that [ 5996 ] = 77. So, we may
entries giving a jj ≤ n 2 − (n − j + 1)2 + 1. Thus take m1 = 77, so that
k
we see that
a jj ∈ { j 2 , j 2 + 1, j 2 + 2,... , n 2 − (n − j + 1)2 + 1}. ∑mj = 23, ∑ j = 2k m 2j = 67
j =2
The number of elements in this set is

Now, we may choose
n 2 − (n − j + 1)2 − j 2 + 2. This implies that
m 2 = 5, m3 = m 4 = 4,
b j ≤ n 2 − (n − j + 1)2 − j 2 + 2
m5 = m 6 = ... = m14 = 1
= (2n + 2)j − 2 j 2 − (2n − 1)
Finally, we can take
It follows that
n n n k = 14, (m1 , m 2 ,... , m14 )
∑ b j ≤ (2n + 2)∑ j − 2 ∑ j 2 − n (2n − 1)
= (77, 5, 4, 4, 1, 1, 1, 1, 1, 1, 1, 1,1, 1),
j =1 j =1 j =1
n (n + 1)  n (n + 1)(2n + 1) proving the existence of 100 lines with exactly
= (2n + 2)  – 2  2002 points of intersection.
 2   6 
−n (2n − 1) 10. Clearly, the last digit must be 5 and we have to
n determine the remaining 6 digits. For
= (n 2 − 3n + 5)
3 divisibility by 7, it is sufficient to consider the
which is the required bound. number obtained by replacing 7 by 0; e.g.,
5775755 is divisible by 7, if and only if
9. Any set of 100 lines in the plane can be 5005055 is divisible by7. Each such number is
partitioned into a finite number of disjoint obtained by adding some of the numbers
sets, say A1 , A 2 , A3 ,... , A k , such that from the set {50, 500, 5000, 50000, 500000,
(i) Any two lines is each A J are parallel to 5000000} along with 5. We look at the
each other, for 1 ≤ j ≤ k (provided, of remainders of these when divided by 7; they
course, | A j | ≥ 2) are {1, 3, 2, 6, 4, 5}. Thus it is sufficient to
(ii) for j ≠ l , the lines in A j and Ai are not check for those combinations of remainders
parallel. which add up to a number of the from 2 + 7k,
If | A j | = m j , 1 ≤ j < k , then the total since the last digit is already 5. These are
number of points of intersection is given {2}, {3, 6}, {4, 5}, {2, 3, 4}, {1, 3, 5}, {1, 2, 6}, {2, 3,
5, 6}, {1, 4, 5, 6} and {1, 2, 3, 4, 6}. These
by ∑ m jml , as no three lines are
correspond to the numbers 7775775,
1≤ j ≤l < k
concurrent,. Thus we have to find positive 7757575, 5577775, 7575575, 5777555,

integers m1 , m 2 ,... , m k such that 7755755, 5755575, 5557755, 755555.
www.pdfworld.in
www.pdfworld.in
Combinatorics 481
11. The following sequence of moves lead to the 2 2 1 1

+ + =
colour of the ticket bearing the number a c e 5
123123123 Here again the tuple (a , a , c , c , e ) for which
Line Ticket a ≠ c is even because we can associate
Colour Reason different solution (c , c , a , a , e ) to this five
No. No.
tuple. Thus it suffices to consider the
1 122222222 red Given
equation
2 222222222 green Given 4 1 1
+ =
3 313113113 blue Lines 1 & 2 a e 5
4 231331331 green Lines 1 & 3 and show that the number of pairs (a, e)
satisfying this equation is odd.
5 331331331 blue Lines 1 & 2 This reduces to
6 123123123 red Lines 4 & 5 ae = 20e + 5a
or (a – 20) (e – 5) = 100
If 123123123 is reached by some other root,
red colour must be obtained even along that But observe that
root. For if for example 123123123 gets blue 100 = 1 × 100 = 2 × 50
from some other root, then the following
sequence leads to a contradiction = 4 × 25 = 5 × 20
= 10 × 10 = 20 × 5 = 25 × 4
List Ticket
Colour Reason = 50 × 2 = 100 × 1
No. No.
1 122222222 red Given Note that no factorisation of 100 as product
of two negative numbers yield a positive tuple
2 123123123 blue Given (a , e ). Hence we get these 9 solutions. This
3 231311311 green Lines 1 & 2 proves that the total number of five tuples
(a , b , c , d , e ) satisfying the given equation is
4 211331311 green Lines 1 & 2 odd.
5 332212212 red Lines 4 & 2 13. I. Consider a 6-digit number whose digits
6 113133133 blue Lines 3 & 5 from left to right are in non-increasing order.
If 1 is the first digit of such a number, then the
7 331331331 green Lines 1 & 2 subsequent digits exceed 1. The set of all such
8 222222222 red Lines 6 & 7 numbers with initial digit equal to 1 is
{100000, 110000, 111000, 111100, 111110,
Thus the colour of 22222222 is red 111111}
contradicting that it is green. There are elements in this set.
12. We write the equation in the form Let us consider 6-digit number with initial
digit 2. Starting form 200000, we an up to
1 1 1 1 1 1
+ + + + = 222222. We count these numbers as follows
a b c d e 5
200000 - 211111 : 6
The number of five tuple (a , b , c , d , e ) which
220000 - 221111 : 5
satisfy the given relation and for which a ≠ b
is even, because for if (a , b , c , d , e ) is a 2220000 - 222111 : 4
solution, then so is (b , a , c , d , e ) which is 222200 - 222211 : 3
distinct from (a , b , c , d , e ). Similarly the 222220 - 222221 : 2
number of five tuples which satisfy the
222222 - 222222 : 1
equation and for which c ≠ d is also even.
Hence it suffices to count only those five The number of such numbers is 21. Similarly,
tuples (a , b , c , d , e ) for which a = b , c = d . Thus we count numbers with initial digit 3; the
the equation reduces to sequence starts from 300000 and with
333333. We have
www.pdfworld.in
www.pdfworld.in
300000 - 322222 : 21 777700 - 777766 : 28

330000 - 332222 : 15 777770 - 777776 :7
333000 - 333222 : 10 777777 - 777777 :1
333300 - 333322 :6 792
333330 - 333332 :3 Thus, the number of 6-digit numbers where
digits are non-digits are non-increasing
333333 - 333333 :1
starting from 100000 and ending with
we obtain the total number of numbers 777777 is
starting from 3 equal to 56. Similarly,
792 + 462 + 252 + 126 + 56 + 21 + 6 = 1715
400000 - 433333 : 56
Since 2005 − 1715 = 290, we have to consider
440000 - 443333 : 35 only 290 numbers in the sequence with initial
444000 - 444333 : 20 digit 8. We have
444400 - 444433 : 10 800000 - 855555 : 252
444440 - 444443 :1 860000 - 863333 : 35
444444 - 444444 : 1 864000 - 864110 : 3
126 II. It is known that the number of ways of
500000 - 544444 : 126 choosing r objects from n different types of
550000 - 554444 : 70 objects (with repetitions allowed) is
 n + r − 1
555000 - 555444 : 35   . In particular, if we want to write
 r 
555500 - 555544 : 15
r-digit numbers using n digits allowing for
555550 - 555554 : 5
repetitions with the additional condition
555555 - 555555 : 1 that the digits appear in non-increasing order,
252  n + r − 1
we see that this can be done in  
600000 - 655555 : 252  r 
660000 - 665555 : 126 ways.
666000 - 666555 : 56 Now we group the given numbers into
666600 - 666655 : 21 different classes and write the number of
666660 - 666665 : 6 ways in which each class can be obtained. To
keep track we also write the cumulative sums
666666 - 666666 : 1
of the number of numbers so obtained.
142 Observe that the numbers themselves are
700000 - 766666 : 462 written in ascending order. So, we exhaust
770000 - 776666 : 210 numbers beginning with 1, then beginning
777000 - 777666 : 84 with 2 and so on.
Numbers Digits used  n + r − 1 Cumulative

other than n r  
 r  sum
the fixed part
 6
beginning with 1 1, 0 2 5   =6 6
 5
 7
2 2, 1, 0 3 5   = 21 27
 5
 8
3 3, 2, 1, 0 4 5   = 56 83
 5
 9
4 4, 3, 2, 1, 0 5 5   = 126 209
 5
www.pdfworld.in
www.pdfworld.in
Combinatorics 483
Numbers Digits used  n + r − 1 Cumulative

other than n r  
 r  sum
the fixed part
 10
5 5, 4, 3, 2, 1, 0 6 5   = 252 461
5
6, 5, 4, 3, 2, 1,  11
6 7 5   = 462 923
0 5
7, 6, 5, 4, 3, 2,  12
7 8 5   = 792 1715
1, 0 5
 10
form 800000 to 855555 5, 4, 3, 2, 1, 0 6 5   = 252 1967
5
 7
form 860000 to 863333 3, 2, 1, 0 4 4   = 35 2002
 5
The next three 6-digit numbers are 864000,

864100, 864110.
Hence, the 2005th number in the sequence is
864110.
14. Consider a partition of 9 × 9 chessboard using
sixteen 2 × 2 block of 4 squares each and
remaining seventeen single squares as shown
in the figure below.
1 2 3 4
7 6 5
16
8 9
15 14
10
11 12 13
We split this again in to several mirror images

If any one of these 16 big squares contain 3 of L-shaped figures as shown above. There
red squares then we are done. On the are four such forks. If all the five units
contrary, each may contain at most 2 red squares of the first fork are red, then we can
squares and these account for at most 16 ⋅ 2 get a 2 × 2 square having three red squares.
Hence, there can be at most four unit squares
= 32 red squares. Then there are 17 single
having red colour. Similarly, there can be at
squares connected in zig-zag fashion. It looks
most three red squares from each of the
as follows remaining three forks. Together we get
www.pdfworld.in
www.pdfworld.in
4 + 3 ⋅ 3 = 13 red squares. These together with Using this and the relation
32 from the big squares account for only 45
a32 − a3a 4 + a 42 = a 42 − a 4a5 + a52 ,
red squares. But we know that 46 squares
have red colour. The conclusion follows. We conclude that a3 = a5 as above. Thus, we
have
15. In a permutation of (1, 2, 3,...,n), two
inversions can occur in only one of the a1 = a3 = a5 = max {a1 , a 2 , a3 , a 4 , a5 , a 6}.
following two ways Let us consider the other relations. Using
(A) Two disjoint consecutive pairs are
a 22 − a 2a3 + a32 = a32 − a3a 4 + a 42 ,
interchanged
(1, 2, 3, j − 1, j , j + 1, j + 2,... , k − 1, we get a 2 = a 4 or a 2 + a 4 = a3 = a1. Similarly,
two more relations give either a 4 = a 6 or
k , k + 1, k + 2,... , n ) a 4 + a 6 = a5 = a1; and either a 6 = a 2 or
→ (1, 2, ... , j − 1, j + 1, j , j + 2,... , a 6 + a 2 = a1. Let us give values to a1 and count
k − 1, k + 1, k , k + 2,... , n ) the number of six-tuples in each case.
(A) Suppose a1 = 1. In this case all a j ’ s
(B) Each block of three consecutive integers
are equal and we get only one six-tuple
can be permuted in any of the following 2
ways; (1, 1, 1, 1, 1, 1).
(B) If a1 = 2, we have a3 = a5 = 2. We observe
(1, 2, 3, ... k , k + 1, k + 2,... , n )
that a 2 = a 4 = a 6 = 1 or a 2 = a 4 = a 6 = 2.
→ (1, 2, ... , k + 2, k , k + 1, ... , n ); We get two more six-tuples (2, 1, 2, 1, 2, 1),
(1, 2, 3, ... , k , k + 1, k + 2, .... , n ) (2, 2, 2, 2, 2, 2).
(C) Taking a1 = 3, we see that a3 = a5 = 3. In
→ (1, 2, ... , k + 1, k + 2, k ,... , n ).
this case we get nine possibilities for
Consider case (A). For j = 1, there are n − 3 (a 2 , a 4 , a 6 ) ;
possible values of k; for j = 2, there are n − 4 (1, 1, 1), (2, 2, 2), (3, 3, 3), (1, 1, 2), (1, 2, 1),
possibilities for k and so on. Thus, the number (2, 1, 1), (1, 2, 2), (2, 1, 2), (2, 2, 1).
of permutations with two inversions of this
(D) In the case a1 = 4, we have a3 = a5 = 4 and
type is
(n − 3)(n − 2) (a 2 , a 4 , a 6 ) = (2, 2, 2), (4, 4, 4),
1 + 2+ ...+ (n − 3) =
2 (1, 1, 1), (3, 3, 3), (1, 1, 3),
In case (B), we see that there are n − 2 (1, 3, 1), (3, 1, 1), (1, 3, 3),
permutations of each type, since k can take (3, 1, 3), (3, 3, 1).
values from 1 to n − 2. Hence, we get 2(n − 2)
permutations of this type. Thus, we get 1 + 2 + 9 + 10 = 22 solutions.
Finally, the number of permutations with two Since (a1 , a3 , a5 ) and (a 2 , a 4 , a 6 ) may be
inversions is interchanged we get 22 more six-tuples.
However, there are 4 common among these,
(n − 3)(n − 2) (n + 1)(n − 2)
+ 2(n − 2) = . namely, (1, 1, 1, 1, 1, 1), (2, 2, 2, 2, 2, 2), (3, 3, 3,
2 2 3, 3, 3) and (4, 4, 4, 4, 4, 4). Hence, the total
16. Without loss of generality we may assume number of six-tuples is 22 + 22 − 4 = 40.
that a1 is the largest among 17. Consider a circle of positive radius in the
a1 , a 2 , a3 , a 4 , a5 , a 6. Consider the relation plane and inscribe a regular heptagon
a12 − a1a 2 + a 22 = a 22 − a 2a3 + a32. ABCDEFG in it. Since the seven vertices of this
heptagon are coloured by three colours, some
This leads to three vertices have the same colour, by
(a1 − a3 )(a1 + a3 − a 2 ) = 0 pigeonhole principle. Consider the triangle
formed by these three vertices. Let us call the
Observe that a1 ≥ a 2 and a3 > 0 together imply
part of the circumference separated by any
that the second factor on the left side is
two consecutive vertices of the heptagon an
positive. Thus,
arc. The three vertices of the same colour are
a1 = a3 = max {a1 , a 2 , a3 , a 4 , a5 , a 6}. separated by arcs of length l, m, n as we move
www.pdfworld.in
www.pdfworld.in
Combinatorics 485
say, counter-clockwise, along the circle, cases, we can pair them with A , C to get
starting from a fixed vertex among these equal arcs. In the last case, we observe that
three, where l + m + n = 7. Since, the order of HA = CE and AHEC is an isosceles
l , m , n does not matter for a triangle, there are trapezium.
four possibilities: 1 + 1 + 5 = 7; 1 + 2 + 4 =7; 1 +
3 + 3 = 7; 2 + 2 + 3 = 7. In the first, third and B B
A A
fourth cases, we have isosceles triangles. In
C C
the second case, we have a triangle whose
angles are in geometric progression. The four
corresponding figures are shown below. D
D H
E
C G F
Fig. 1 Fig. 2
B
A
B B
A A
(i) (ii) D
E E H
E
C G F
Fig. 3
l
Suppose only two among the five are
B adjacent, say A , B. Then, the remaining
three are among D, E , F , G, H . (See Fig 3.) If
A A any two of these are adjacent, we can
(iii) (iv)
combine them with A , B to get equal arcs. If
In (i), AB = BC; in (iii), AE = BE ; in (iv), no two among these three vertices are
AC = CE ; and in (ii), we see that ∠D = π/7, adjacent, then they must be D, F , H . In this
∠A = 2 π/7, and ∠B = 4 π/7, which are in case HA = BD and AHDB is an isosceles
geometric progression. trapezium.
18. Suppose four distinct points P , Q , R , S (in that Finally, if we choose 5 among the 9 vertices
order on the circle) among these five are such of a regular nine-sides polygon, then some
that PQ = RS . Then, PQRS is an isosceles two must be adjacent. Thus, any choice of 5
trapezium, with PS | | QR. We use this in our among 9 must fall in to one of the above
argument. three possibilities.
l
If four of the five points chosen are Aliter Here, is another solution used by
adjacent, then we are through as observed many students. Suppose you join the vertices
earlier. (In this case four points A , B , C , D of the nine-sides regular polygon. You get
are such that AB = BC = CD. ) See Fig 1.  9  = 36 line segments. All these fall in to 9
 
 2
l
Suppose only three of the vertices are
adjacent, say A , B , C (see Fig 2.). Then, the sets of parallel lines. Now, using any 5 points,
remaining two must be among E , F , G, H . If  5
you get   = 10 line segments. By pigeonhole
these two are adjacent vertices, we can pair  2
them with A , B or B , C to get equal arcs. If principle, two of these must be parallel. But,
they are not adjacent, then they must be these parallel lines determine a trapezium.
either E , G or F , H or E , H . In the first two
www.pdfworld.in
www.pdfworld.in
Unit 5
Geometry
Congruent Triangles
Two triangles are congruent if and only if one of them can be made to superpose on the other so as to
cover it exactly.
Some Elementary Theorem of Congruency (Without proof) which are often used
1. Side Angle Side (SAS) Congruence Theorem
Two triangles are congruent, if two side and the included angle of one are equal to the corresponding
sides and the included angle of the other triangle.
2. Angle Side Angle (ASA) Congruence Theorem
Two triangles are congruent, if two angles and the included side of one triangle are equal to the
corresponding two angles and the included side of the other triangle.
3. Angle Angle Side (AAS) Congruence Theorem
If any two angles and a non-included side of one triangle are equal to the corresponding angles side
of another triangle, then the two triangles are congruent.
4. Side Side Side (SSS) Congruence Theorem
Two triangles are congruent, if the three side of one triangle are equal to the corresponding three
side of the other triangle.
5. Right Angle Hypotenuse Side (RHS) Congruence Theorem
Two right triangles are congruent, if the hypotenuse and one side of one triangle are respectively
equal to the hypotenuse and one side of the other triangle.
Note Angles opposite to equal sides of a triangle are equal.
Example 1 In the given figure, AB = CF , EF = BD ; ∠ AFE = ∠ DBC.

Prove that ∆ AFE ≅ ∆ CBD.
D
B
A C
F
www.pdfworld.in
www.pdfworld.in
Geometry 487
Solution ∴ AB = CF
∴ AB + BF = BF + FC
⇒ AF = CB …(i)
In ∆ AFE and ∆ CBD
AF = CB [from Eq. (i)]
EF = BD [given]
∠ AFE = ∠ DBC [given]
∴ ∆ AFE ≅ ∆CBD [By SAA congruence rule]
Example 2 P and Q are two points on equal sides AB and AC of an isosceles ∆ ABC such that
AP = AQ .
Prove that PC = QB.
A
Q P
O
C B
Solution Q AP = AQ and AB = AC
∴ AB − AP = AC − AQ
⇒ PB = QC …(i)
In ∆s PBC and QBC, we have
PB = QC [by Eq. (i)]
BC = BC [Common]
∠ PBC = ∠ PCB [ Q AB = AC ]
∴ ∆ PBC ≅ ∆QBC [By SAS congruence rule]
∴ PC = QB
Example 3 In ∆ ABC and ∆ PQR, AB = PQ , BC = QR ; CB and RQ are extended to X and Y

respectively. ∠ ABX = ∠ PQY . Prove that ∆ ABC ≅ ∆ PQR .
A P
X B C Y Q R
Solution Q ∠ABX = ∠PQY

180° − ∠ABC = 180° − ∠PQR [Q ∠ ABX + ∠ ABC = 180° ;
∠ PQY + ∠ PQR = 180° (linear pair)]
∠ ABC = ∠ PQR …(i)
www.pdfworld.in
www.pdfworld.in
In ∆ ABC and ∆ PQR

AB = PQ (given)
∠ ABC = ∠ PQR [from Eq. (i)]
BC = QR (given)
∆ ABC ≅ ∆ PQR [from Eq. (i)]
[By SAS congruence rule]
Example 4 Hari wishes to determine the distance between two objects A

and B, but there is an obstacle between these two objects A B
which prevent him from making a direct measurement. He
devises an ingenious way to overcome this difficulty. First he
fixes a pole at a convenient point O so that from O, both A and O
B are visible. Then he fixes another pole at the point D on the
line AO (produced) such that AO = DO . In a similar way he
fixes a third pole at the point C on the line BO (produced) such C D
that BO = CO . Then he measures CD which is equal to 170 cm.
Prove that the distance between the objects A and B is also 170 cm.
Solution In ∆s AOB and ∆ COD
OA = OD (given)
∠ AOB = ∠ COD (vertically opposite angles)
OB = OC (given)
∴ ∆ AOB ≅ ∆COD
⇒ AB = CD (c.p.c.t.)
⇒ AB = 170 cm [ Q CD = 170 cm]
Hence proved.
Example 5 The line segment joining the mid point M and N of opposite sides AB and DC of
quadrilateral ABCD is perpendicular to both these sides.
A M
B
900
900
D C
N
Prove that the other sides of the quadrilateral are equal.

Solution Join M and D and also M and C
in ∆ CMN and ∆ DMN
DN = NC [N is mid point of CD]
∠ DNM = ∠ CNM = 90°
MN = MN (common)
∆ CMN ≅ ∆ DMN [By SAS congruence rule]
www.pdfworld.in
www.pdfworld.in
Geometry 489
⇒ CM = DM …(i)
∠ DMN = ∠ CMN
and ∠ AMN = ∠ BMN [ 90° each (given)]
⇒ ∠ AMN − ∠ DMN = ∠ BMN − ∠ CMN
⇒ ∠ AMD = ∠ CMB …(ii)
In ∆ AMD and ∆ CMB
AM = MB [ Q M is mid point of AB ]
∠ AMD = ∠ CMB [from Eq. (ii)]
DM = MC [from Eq. (i)]
∴ ∆ AMD ≅ ∆ CMB ⇒ AD = BC
Example 6 In the given figure,

AC = AE ; AB = AD, ∠ BAD = ∠ EAC. Prove that BC = DE
E
A
B C
D
Solution Join DE
∠ BAD = ∠ EAC
⇒ ∠ BAD + ∠DAC = ∠ EAC + ∠ DAC ⇒ ∠ BAC = ∠ DAE ...(i)
In ∆ ABC and ∆ ADE
AB = AD (given)
∠ BAC = ∠ DAE [ from Eq. (i) ]
AC = AE
∆ ABC ≅ ∆ ADE [By SAS congruence rule ]
⇒ BC = DE [ c.p.c.t ]
Example 7 AB is a line segment. AX and BY are two equal line segments drawn opposite sides of
line AB such that AX | | BY .
3 B
P
A
4
www.pdfworld.in
www.pdfworld.in
If line segments AB and XY intersect each other at point P . Prove that

(a) ∆ APX ≅ ∆ BPY
(b) line segments AB and XY bisect each other at P.
Solution (a) QAX | | BY and XY is a transversal
∠1 = ∠ 2 (Alternate interior angles) …(i)
Q AX | | BY and AB is a transversal
∠3=∠4 (Alternate interior angles) …(ii)
In ∆ APX and ∆ PBY
∠1 = ∠ 2 [from Eq. (i)]
AX = BY (given)
∠3=∠4 [By Eq. (ii)]
∴ ∆ APX ≅ ∆ BPY (ASA rule)
(b) Q ∆ APX ≅ ∆ BPY
AP = PB and PX = PY [c.p.c.t.]
∴ AB and XY bisect each other at P.
Example 8 In the figure, C is the mid point of AB, ∠ BAD = ∠ CBE , ∠ ECA = ∠ DCB.
Prove that
E D
A B
C
(a) ∆ DAC ≅ ∆ EBC (b) DA = EB
Solution In ∆ DAC and ∆ EBC
AC = CB [ Q C is mid point of AB]
∠ CAD = ∠ CBE (given)
∠ ACD = ∠ BCE
[ Q∠ ACD = 180° − ∠ DCB, ∠ BCE = 180° − ∠ ECA and ∠ DCB = ∠ ECA]
∆ DAC ≅ ∆ EBC [By AAS rule]
DA = EB (c.p.c.t.)
Example 9 In the figure, ∠ CPD = ∠ BPD, AD is bisector of ∠ BAC.

Prove that ∆ CAP ≅ ∆ BAP and hence CP = BP .
C
D
A
P
www.pdfworld.in
www.pdfworld.in
Geometry 491
Solution ∠BPD = ∠CPD

⇒ 180° − ∠BPA = 180° − ∠CPA
⇒ ∠BPA = ∠CPA …(i)
In ∆s BAP and ∆CAP, we have
∠ BAP = ∠CAP [QAD is bisector of ∠BAC]
AP = AP (common)
∠BPA = ∠CPA [from Eq. (i)]
∴ ∆ BAP ≅ ∆CAP [By ASA congruence rule]
∴ BP = CP (By c.p.c.t.)
Example 10 In ∆ ABC, AB = AC. Bisectors of angles B and C intersect at point O. Prove that
BO = CO and the ray AO is bisector of ∠ BAC.
A
B C
Solution In ∆ ABC,
AB = AC
⇒ ∠B =∠C [QAngle opposite to equal sides are equal]
1 1
⇒ ∠B = ∠ C
2 2
⇒ ∠ ABO = ∠ ACO …(i)
[QOB and OC are bisectors of ∠ B and C respectively]
1
∴ ∠ OBC = ∠ B
2
1
and ∠ OCB = ∠ C
2
⇒ OB = OC ...(ii)
[Q sides opposite to equal angles are equal]
In ∆ ABO and ∆ ACO
AB = AC (given)
∠ ABO = ∠ ACO [from Eq. (i)]
OB = OC [ from Eq. (ii)]
∆ ABO ≅ ∆ ACO [By SAS rule]
⇒ ∠ BAO = ∠ CAO [ c.p.c.t.]
⇒ AO is the bisector of ∠ BAC.
www.pdfworld.in
www.pdfworld.in
Example 11 In the given figure, AD = AE .D and E are points on BC such that BD = EC. Prove that
AB = AC.
A
B D E C
Solution In ∆ ADE
AD = AE
∠ ADE = ∠ AED
⇒ 180° − ∠ ADE = 180° − ∠ AED
⇒ ∠ ADB = ∠ AEC …(i)
In ∆ ABD and ∆ ACE
AD = AE (given)
∠ ADB = ∠ AEC [from Eq. (i)]
BD = EC
∆ ABD ≅ ∆ ACE [By ASA rule]
⇒ AB = AC [c.p.c.t.]
Example 12 In figure,
PS = PR ,
∠ TPS = ∠ QPR
Prove that PT = PQ
T S R Q
Solution In ∆ PRS
PS = PR
⇒ ∠ PRS = ∠ PSR [QAngle opposite to equal sides are equal]
⇒ 180° − ∠ PRS = 180° − ∠ PSR
⇒ ∠ PRQ = ∠ PST …(i)
In ∆ PST and ∆ PRQ
∠ TPS = ∠ QPR (given)
PS = PR (given)
www.pdfworld.in
www.pdfworld.in
Geometry 493
∠ PST = ∠ PRQ [from Eq. (i)]

∆ PST ≅ ∆ PRQ [By ASA rule]
⇒ PT = PQ
Example 13 PQRS is a quadrilateral. T and U are points on PS and RS respectively such that
PQ = RQ . ∠ PQT = ∠ RQU and ∠ TQS = ∠ UQS.
Prove that QT = QU.
P
T
1
3
Q S
4
2
U
R
Solution In ∆ PQS and RQS,

QS = QS (common)
∠ PQS = ∠ RQS [Q∠ PQS = ∠ 1 + ∠ 3 = ∠ 2 + ∠ 4 = ∠ RQS ]
PQ = QR
∴ ∆ PQS ≅ ∆ RQS [By SAS rule]
∴ ∠ QPS = ∠QRS
i.e., ∠ QPT = ∠QRU
In ∆ PQT and ∆ QRU,
∠1 = ∠ 2
∠ QPT = ∠ QRU
PQ = QR
∴ ∆ PQT ≅ ∆ QRU (By c.p.c.t.)
∴ QT = QU
Concept
Some Properties of An Isosceles Triangle (without proof)
1. If two angles of a triangle are equal, then sides opposite to them are also equal.
2. If the altitude from one vertex of a triangle bisects the opposite side, then the triangle is isosceles.
3. In an isosceles triangle , altitude from the vertex bisects the base.
4. If the bisector of the vertical angle of a triangle bisects the base of the triangle , then the triangle is
isosceles.
Example 1 The altitudes of ∆ ABC, AD, BE and CF are equal. Prove that ∆ ABC is an equilateral
triangle .
A
F E
B C
D
www.pdfworld.in
www.pdfworld.in
Solution In right angled triangles BCE and BCF

BC = BC (common)
BE = CF (given)
∴ ∆ BCE ≅ ∆ BCF [ By RHS rule]
∴ ∠B =∠C [c.p.c.t]
⇒ AC = AB ...(i)
[Qside opposite to equal angles are equal]
Similarly, ∆ ABD ≅ ∆ ABE
∴ ∠B =∠ A
⇒ AC = BC …(ii)
[Qside opposite to equal angles are equal]
From Eqs. (i) and (ii),
AB = BC = AC
∴ ∆ ABC is an equilateral triangle.
Example 2 In figure, AB = AC. D and E are points on BC such that BD = EC (or BE = CD ). Prove
that AD = AE .
A
B D E C
Solution Q AB = AC
∴ ∠ ABC = ∠ ACB
∠B =∠C
In ∆ ABD and ∆ AEC
AB = AC (given)
BD = EC (given)
∠B =∠C (Proved)
∴ ∆ ABD ≅ ∆ AEC
∴ AD = AE [c.p.c.t.]
Example 3 ∆ ABC is isosceles with AB = AC. Prove that altitudes BD and CE of triangle are equal.
A
E D
B C
www.pdfworld.in
www.pdfworld.in
Geometry 495
Solution In ∆ ABC
∠ ABC = ∠ ACB [QAB = AC]
i.e., ∠ CBE = ∠ BCD
In ∆ s BCE and ∆ BCD
∠ CBE = ∠ BCD [from Eq. (i)]
∠ CEB = ∠ CDB (90° each)
∠ BCE = ∠ CBD
[Q∠ BCE = 90° − ∠ B ; ∠ CBD = 90° − ∠ C, ∠ B = ∠ C]
BC = BC (common)
∴ ∆ BCD ≅ ∆ BCE [By AAS rule]
∴ BD = CE [c.p.c.t.]
Example 4 In the figure, PQ = PT , ∠ TPS = ∠ QPR. Prove that ∆ PRS is isosceles

T S R Q
Solution In ∆ PQT
PQ = PT
∴ ∠ TQP = ∠ QTP
In ∆ PQR and ∆ PST
PQ = PT
∠ TQP = ∠ QTP
∠ QPR = ∠ TPS (given)
∴ ∆ PQR ≅ ∆ PST (By AAS rule)
∴ PS = PR [c.p.c.t.]
∴ ∆ PRS is isosceles.
Example 5 ∆ ABC is an isosceles triangle with AB = ACBD and CE are two medians of triangle.
Prove that BD = CE
A
E D
B C
www.pdfworld.in
www.pdfworld.in
Solution In ∆ ABD and ∆ AEC

AB = AC (given)
AD = AE
1
[Q D is mid point of AC ∴AD = DC i.e., AD = AC]
2
∠ BAC = ∠ CAB (common)
[QE is mid point of AB, ∴ AE = EB
i.e., AE = 1 / 2 AB,
Q AB = AC ∴ AD = AE ]
∴ ∆ ABD ≅ ∆ AEC (By SAS rule)
∴ BD = CE [c.p.c.t.]
Some Inequality Relations in a Triangle (Without Proof)

1. If two sides of a triangle are unequal, the longer side has greater angle opposite to it.
2. In a triangle, the greater angle has longer side opposite to it.
3. The sum of any two sides of a triangle is greater than the third side.
4. Of all the line segments that can be drawn to a given line from a point not lying on it the
perpendicular line segment is the shortest.
Example 1 In figure PQ > PR . QS and RS are the bisectors of ∠ Q and ∠ R respectively. Prove
that SQ > SR .
P
Q R
Solution In ∆ PQR
PQ > PR (given)
⇒ ∠ PRQ > ∠ PQR
[Angle opposite to greater side of a triangle is greater]
1
⇒ ∠ PRQ
2
1
> ∠ PQR
2
⇒ ∠ SRQ > ∠ SQR
[QRS and QS are bisectors of ∠ PRQ and ∠ PQR respectively]
⇒ SQ > SR
(Side opposite to greater angle is greater)
www.pdfworld.in
www.pdfworld.in
Geometry 497
Example 2 In figure PQ and PR are produced and ∠ SQR < ∠ TRQ.

P
R
Q
S T
Prove that PR > PQ .

∠ PQR + ∠ SQR = 180°
Solution  linear pair
∠ PRQ + ∠ TRQ = 180°
∴ ∠ PQR + ∠ SQR = ∠ PRQ + ∠ TRQ
But ∠ SQR < ∠ TRQ
∴ ∠ PQR < ∠ PRQ
∴ PR > PQ [Side opposite to greater angle is greater]
Example 3 Q is a point on side RS of ∆ PSR such that PQ = PR . Show that PS > PQ

P
S Q R
Solution In ∆ PQR,PQ = PR
∴ ∠ PRQ = ∠ PQR [Angles opposite to equal sides are equal] …(i)
In ∆ PSQ,SQ is produced to R
ext. ∠ PQR > ∠ PSQ …(ii)
[exterior angle of a triangle > each of interior opposite angles]
∠ PRQ = ext. ∠ PQR > ∠ PSQ
i.e., ∠ PRQ > ∠ PSQ
i.e., ∠ PRS > ∠ PSR [Q∠ PRQ = ∠ PRS and ∠ PSQ = ∠ PSR ]
In ∆ PSR,
∠ PRS > ∠ PSR
∴ PS > PR
[Qside opposite to greater angle is greater]
PR = PQ
∴ PS > PQ
www.pdfworld.in
www.pdfworld.in
Example 4 AD is the bisector of ∠ A of ∆ ABC, D lies on BC. Show that AB > BD and AC > CD.
A
1 2
4 3
B D C
Solution In ∆ ABC
AD is bisector of ∠ A
∴ ∠1 = ∠ 2 ...(i)
But in ∆ ABD
ext. ∠ ADC > ∠ 1
[exterior ∠ of a triangle is greater than each of interior opposite angles]
i.e., ∠ 3 > ∠1
∴ ∠3>∠2 [ from Eq. (i) ]
∴ from ∆ ADC
AC > AD
[Q Side opposite to greater angle is greater ]
Also ∠4>∠2 [In ∆ ADC, ∠ 4 is exterior angle]
∠ 4 > ∠1
AB > BD [QSide opposite to greater angle is greater]
Example 5 In figure, AB > AC. D is any point on side BC of ∆ ABC. Show that AB > AD.
A
B D C
Solution QAB > AC

∴ ∠ ACB > ∠ ABC …(i)
[QAngle opposite to greater side is greater]
In ∆ ACD, CD is produced to B forming an exterior angle ADB
∴ ∠ ADB > ∠ ACD
[Qexterior angle of a triangle is greater than each of interior opposite angles ]
⇒ ∠ ADB > ∠ ACB [Q∠ ACD = ∠ ACB] …(ii)
www.pdfworld.in
www.pdfworld.in
Geometry 499
from Eqs. (i) and (ii)

∠ ADB > ∠ ABC
⇒ ∠ ADB > ∠ ABD [ Q ∠ ABC = ∠ ABD]
⇒ AB > AD [Qside opposite to greater angle is greater]
Example 6 S is any point in the interior of ∆ PQR. Show that SQ + SR < PQ + RP

P
T
S
Q R
Solution Produce QS to meet PR at T .

In ∆ PQT
PQ + PT > QT
[QSum of two sides of a triangle is greater than third side ]
⇒ PQ + PT > QS + ST [ Q QT = QS + ST ] ...(i)
In ∆ RST
ST + TR > SR …(ii)
On adding Egs. (i) and (ii) gives
PQ + PT + ST + TR > SQ + ST + SR
⇒ PQ + (PT + TR ) > SQ + SR
⇒ PQ + PR > SQ + SR
∴ SQ + SR < PQ + PR
Example 7 S is any point on the side QR of ∆ PQR. Show that PQ + QR + RP > 2PS .
P
Q S R
Solution In ∆ PQS
PQ + QS > PS …(i)
[Qsum of two sides of a triangle is greater than third side]
In ∆ PSR
RP + RS > PS …(ii)
On adding Eqs. (i) and (ii) , we get
PQ + QS + RP + RS > 2PS
PQ + (QS + RS ) + RP > 2PS
PQ + QR + RP > 2PS [ Q QS + RS = QR ]
www.pdfworld.in
www.pdfworld.in
Example 8 In figure, AP ⊥ l i.e., AP is the shortest line segment that can be drawn from A to the
line l. If PR > PQ . Show that AR > AQ
A
1 3 2
Q P S R l
Solution Cut off PS = PQ from PR join AS.

In ∆ APQ and ∆ APS
∠APQ = ∠ APS (each 90°)
AP = AP (common side)
PQ = PS
∴ ∆ APQ ≅ ∆ APS
AP = AS and ∠ 1 = ∠ 3 …(i)
In ∆ARS, ∠3>∠2 [exterior angle property ] …(ii)
∠1 > ∠ 2
from ∆ AQR ∠1 > ∠ 2
∴ AR > QR
∴ AR > AQ
Example 9 Prove that perimeter of a triangle is greater than sum of its three medians.
A
F E
B C
D
Solution In a triangle ABC let, perimeters is equal to AB + BC + AC and sum of medians is

equal to AD + BE + CF .Now we know that sum of any two sides of a triangle is greater
than the third side.
∴ AB + AC > 2AD … (i)
[QAD is median of ∆ ABC through the point A]
Similarly,
BC + BA > 2BE …(ii)
and CA + CB > 2CF ...(iii)
www.pdfworld.in
www.pdfworld.in
Geometry 501
Adding Eqs. (i), (ii) and (iii), we get

( AB + AC ) + (BC + BA ) + (CA + CB ) > 2
( AD + BE + CF ) 2 ( AB + BC + AC ) > 2 ( AD + BE + CF )
Hence, AB + BC + AC > AD + BE + CF
Example 10 Show that the sum of the three altitudes of a triangle is less than the sum of the three
sides of a triangle.
Solution In a ∆ ABC let perimeter equal to AB + BC + CA and sum of altitudes is equal to
AD + BE + CF . Now since AD ⊥ BC
A
F E
B C
D
∴ AD < AB , AD < AC
∴ AD + AD < AB + AC
2AD < AB + AC …(i)
∴ BE⊥ CA
⇒ BE < BC , BE < BA
⇒ BE + BE < BC + BA
2BE < BC + BA …(ii)
∴ CF ⊥ AB
∴ CF < CA,CF < CB
∴ CF + CF < CA + CB
2 CF < CA + CB …(iii)
On adding Eqs. (i), (ii) and (iii), we get
2 ( AD + BE + CF ) < AB + AC + BC + BA + CA + CB
= 2 AB + 2 BC + 2 CA = 2 ( AB + BC + CA )
Hence, AD + BE + CF < AB + BC + CA
Example 11 PR > PQ , PS is the bisector of ∠ QPR.

Show that x > y .
Solution In ∆ PQR , PR > PQ ∠ PQR > ∠ PRQ
P
∠1 = ∠ 2 [Q PS is bisector of ∠ P]
∴ ∠ PQR + ∠ 1 > ∠ PRQ + ∠ 2 1 2
⇒ 180° − y ° > 180° − x °
[Q∠ PQS + ∠ 1 + y ° = 180°, ∠ PRS + ∠ 2 + x ° = 180°]
⇒ − y° > − x° y° x°
⇒ y ° < x ° i.e., x ° > y °
Q S R
www.pdfworld.in
www.pdfworld.in
Example 12 T is a point on side QR of ∆ PQR . S is a point such that RT = ST . Prove that

PQ + PR > QS
S T
R P
Solution In ∆ RPQ
PQ + PR > RQ …(i)
RQ = TR + TQ
= ST + TQ > QS [from ∆ STQ] …(ii)
PQ + PR > QS
Example 13 PQRS is a quadrilateral. PQ is its longest side. RS is the shortest side, prove that
∠ R > ∠ P and ∠ S > ∠ Q.
S R
8 6
7 5
1 4
2 3
P Q
Solution Join PR and QS,

Q PQ is longest side of quadrilateral PQRS
∴ In ∆ PQR
PQ > QR
∴ ∠ 5> ∠ 2 … (i)
QRS is smallest side of quadrilateral. PQRS
∴ PS > RS
∴ ∠6 > ∠1 …(ii)
On adding Eqs. (i) and (ii)
∠ 5 + ∠ 6 > ∠ 2 + ∠1
⇒ ∠R >∠P
Hence proved.
www.pdfworld.in
www.pdfworld.in
Geometry 503
In ∆ PQS, PQ > PS [ Q PQ is longest side]

∴ ∠8>∠3 …(iii)
In ∆ SRQ, RQ > RS [QRS is shortest side]
∴ ∠7>∠ 4 …(iv)
On adding Eqs. (iii) and (iv), we get
∠ 8+ ∠7>∠ 3+ ∠ 4
⇒ ∠S >∠Q
Hence proved.
Example 14 In ∆ ABC, AD⊥ BC; BE ⊥ AC;CF ⊥ AB. Prove that AD + BE + CF > AB + BC + CA.
F
E
B D C
Solution In ∆ ABD, AD < AB …(i)

(Perpendicular line segment is shortest)
In ∆ ADC, AD < AC …(ii)
(Perpendicular line segment is shortest)
2 AD < AB + AC …(iii)
[On adding Eqs. (i) and (ii)]
Similarly 2 BE < AB + BC … (iv)
2 CF < BC + AC …(v)
On adding Eqs. (iii), (iv) and (v), we have
2 ( AD + BE + CF ) < 2 ( AB + BC + CA )
⇒ AD + BE + CF < AB + BC + CA
Example 15 In ∆ ABC, AC > AB
Prove that AC − AB < BC, AC − BC < AB,
BC − AB < AC.
Solution From AC, cut AD = AB,
A
2 D
1 3
4
B C
join AB = AD [construction]
www.pdfworld.in
www.pdfworld.in
∠ 2 = ∠1 …(i)
[Base angle of an isosceles triangle]
∠ 3 > ∠1 …(ii)
[ext. angle of ∆ ABD]
∠2>∠4 …(iii)
[ext. angle of ∆ BDC]
∠3 > ∠2
[from Eqs. (i) and (ii)]
∠3>∠4 [from Eqs. (ii) and (iii)]
In ∆ BDC
Q ∠3>∠4
∴ BC > DC
⇒ DC < BC ⇒ AC − AD < BC
⇒ AC − AB < BC [Q AD = AB]
Similarly, AC − BC < AB, BC − AB < AC
Hence proved.
Properties of a Parallelogram
1. A diagonal of a parallelogram divides it into two congruent triangles.
2. In a parallelogram opposite sides are equal.
3. In a parallelogram the opposite angles are equal.
4. In a parallelogram the diagonals bisect each other.
5. In a quadrilateral, if opposite sides are equal, it is a parallelogram.
6. In a quadrilateral, if opposite angles are equal, it is parallelogram.
7. If the diagonals of a quadrilateral bisect each other, then the quadrilateral is a parallelogram.
8. Each of the four sides of a rhombus is of same length.
9. Each of the angles of a square is a right angle and each of the four sides is of same length.
10. If the two diagonals of a parallelogram are equal, it is a rectangle.
11. The diagonal of a rhombus are perpendicular to each other.
12. If the diagonals of a parallelogram are perpendicular then it is a rhombus.
13. The diagonals of a square are equal and perpendicular to each other.
Example 1 In a parallelogram, show that angle bisector of two adjacent angles intersect at right
angles
D C
A B
Solution In a parallelogram ABCD ∠A and ∠B are adjacent angles. B sectors of ∠ A and ∠ B
meet at the point P. Now
∠ A + ∠ B = 90° [sum of adjacent angles of a parallelogram]
www.pdfworld.in
www.pdfworld.in
Geometry 505
1 1
⇒ ∠ A + ∠ B = 90°
2 2
⇒ ∠ PAB + ∠ PBA = 90°
But ∠ APB + ∠ PBA + ∠ PAB = 180° [sum of angles of a triangle]
∴ 90° + ∠ APB = 180°
∠ APB = 90°
Example 2 Let ABCD be a parallelogram. Let AP , CQ be the perpendicular from A and C on its
diagonal BD. Prove that AP = CQ
D C
A B
Solution From ∆ APD and ∆ BQC

AD = BC ( Q ABCD is a parallelogram ∴ AD = BC )
∠ ADP = ∠ CBQ [Q AD | | BC and BD is transversal]
∠ APD = ∠ CQB [each 90°]
∆ APD ≅ ∆ BQC [By AAS rule]
∴ AP = CQ
Example 3 AB | | PQ , AB = PQ , AC | | PR
AC = PR . Prove that
BC | | QR and BC = QR .
A P
Q
B
C R
Solution Q AB | | PQ , AB = PQ
∴ ABQP is a parallelogram
∴ BQ = AP and BQ | | AP …(i)
Q AC | | PR , AB = PR
∴ ACRP is parallelogram
∴ CR = AP and CR | | AP …(ii)
www.pdfworld.in
www.pdfworld.in

BQ = CR and BQ | | CR
∴ BQRC is a parallelogram
∴ BC | | QR and BC = QR
Example 4 In a parallelogram ABCD, P and Q are two points taken on its diagonal BD such that
DP = BQ . Prove that APCQ is a parallelogram.
A D
P
Q
B C
Solution In ∆ APD and ∆ CQB

DP = BQ (given)
AD = BC [opposite sides of a parallelogram]
∠ ADP = ∠ CBQ (Alternate angles)
[QAD | | BC and transversal BD meets them at D, P respectively]
∴ ∆ APD ≅ ∆ CQB [By SAS]
AP = CQ [c.p.c.t.]
Similarly,
CP = AQ
∴ APCQ is a parallelogram.
Example 5 Prove that the angle bisector of a parallelogram forms a rectangle.

Solution Let ABCD is a parallelogram in which angle bisectors from a quadrilateral PQRS.
QAP is bisector of ∠ A and BR is bisector of ∠ B meeting at each other at point S.
A D
R
S Q
P
B C
∴ ∠ BAS + ∠ ABS
1 1 1
= ∠ A + ∠ B = × 180°
2 2 2
[QAD | | BC and AB intersects them]
∠ BAS + ∠ ABS = 90°
But ∠ BAS + ∠ ABC + ∠ ASB = 180°
∴ ∠ ASB = 90°
www.pdfworld.in
www.pdfworld.in
Geometry 507
Thus ∠ RSP = 90°

[Q∠ RSP and ∠ ASB are vertically opposite angles]
Similarly, we prove that
∠ SRQ = ∠ RQA = 90°
Hence, PQRS is a rectangle.
Example 6 Prove that the figure formed by joining the mid points of the pairs of adjacent sides of
a quadrilateral is a parallelogram.
D
G
C
H
F
A B
E
Solution ABCD is a quadrilateral E , F , G , H are mid points of sides AB, BC, CD, DA
respectively.
PQRS is a parallelogram
Join A and C.
In ∆ ABC,E and F are the mid points of sides AB and BC respectively
1
∴ EF | | AC and EF = AC …(i)
2
In ∆ ADC,G and H are mid points of CD and AD respectively
1
HG | | AC and RS = AC …(ii)
2
from Eqs. (i) and (ii), we get
EF | | HG and EF | | RS
∴ EFGH is a parallelogram.
Example 7 The diagonals of a quadrilateral ABCD are perpendicular. Show that the quadrilateral
formed by joining the mid points of its sides, is a rectangle.
Solution A quadrilateral ABCD. AC and BD are its diagonals such that AC ⊥ BD
D R
C
S Q
O
N M
A P B
P , Q , R , S are mid points of sides AB, BC, CD, DA respectively.
www.pdfworld.in
www.pdfworld.in
Join PQ , QR , RS and SP
In ∆ ABC, P and Q are mid points of AB and BC respectively.
1
∴ PQ | | AC and AC and PQ = AC …(i)
2
In ∆ ADC, R and S are mid points of CD and AD respectively.
1
∴ RS | | AC and RS = AC …(ii)
2
∴ PQ | | RS and PQ = RS [from Eqs. (i) and (ii)]
∴ PQRS is a parallelogram
Let the diagonals AC and BD meet at point O.
Now, in ∆ ABD, P is mid point of AB and S is mid point of AD.
∴ PS | | BD i.e., PN | | BD
Also from Eq. (i), QP | | AC
∴ PM | | NO
In quadrilateral PMON,PN | | MO and PM | | NO
PMON is a parallelogram ⇒ ∠ MPN = ∠ MON
[Qopposite angles of parallelogram are equal]
⇒ ∠ MPN = ∠ BOA ( Q ∠ BOA = ∠ MON )
⇒ ∠ MPN = 90° [QAC ⊥ BC,∴ ∠ BOA = 90°]
⇒ ∠ QPS = 90°
Hence PQRS is a parallelogram with one angle
∠ QPS = 90°
∴ PQRS is a rectangle.
Example 8 ABCD is a rhombus. P , Q , R , S are mid points of AB, BC, CD, DA respectively. Prove
that PQRS is a rectangle.
Solution Join AC
R
D C
4
S Q
1
3 2
A B
P
In ∆ ABC,P and Q are mid points of AB and BC respectively

1
∴ PQ | | AC and PQ = AC …(i)
2
In ∆ ADC,R and S are mid points of CD and AD respectively
1
∴ SR | | AC and SR = AC …(ii)
2
www.pdfworld.in
www.pdfworld.in
Geometry 509

PQ | | SR and PQ = SR
∴ PQRS is a parallelogram.
Again ABCD is a rhombus.
Q AB = BC
[QAll sides of a rhombus are equal]
1 1
∴ AB = BC ⇒ PB = BQ …(iii)
2 2
[QP and Q are mid points of AB, BC respectively]
In ∆ PBQ,PB = PQ
∴ ∠1 = ∠ 2 …(iv)
[QAngle opposite to equal sides are equal]
Q ABCD is a rhombus.
∴ AB = BC = CD = AD
⇒ AB = BC, CD = AD
1 1 1 1
⇒ AB = BC ; CD = AD
2 2 2 2
∴ AD = CQ ,CR = AS
In ∆APS and ∆CQR
AP = CQ [By Eqs. (iii)]
AS = CR [By Eqs. (iv)]
PS = QR [QPQRS is parallelogram]
∆ APS ≅ ∆ CQR [By SSS rule]
∠3=∠4 [c.p.c.t.] …(v)
Now, ∠ 3 + ∠ SPQ + ∠ 2 = 180°
and ∠ 1 + ∠ PQR + ∠ 4 = 180°
∴ ∠ 3 + ∠ SPQ + ∠ 2 = ∠ 1 + ∠ PQR + ∠ 4
⇒ ∠ SPQ = ∠ PQR [Q∠ 1 = ∠ 2 , ∠ 3 = ∠ 4] …(vi)
Now transversal PQ cuts parallel lines SP and RQ at P , Q respectively.
∴ ∠ SPQ + ∠ PQR = 180°
⇒ ∠ SPQ + ∠ SPQ = 180°
2 ∠ SPQ = 180°
∠ SPQ = 90°
Hence, PQRS is a parallelogram such that its one angle ∠ SPQ = 90°.
Hence PQRS is rectangle.
www.pdfworld.in
www.pdfworld.in
Similar Triangles (Ratio and Proportion)

Two figures are said to be similar, if they are of same shape. For two polygons to be similar this requires
that each angle of the one must be equal to each angle of the other and corresponding sides are
proportional. In case of a triangle equality of angles turns out to be equivalent to proportionality of
corresponding sides. Here are some facts regarding similar triangles.
1. In any triangle, a straight line drawn parallel to one of the sides divides the remaining sides
proportionally and conversely.
In the figure, if DE | | BC, then A
AD AE
=
DB EC
D E
AD AE
and equivalently, =
AB EC
B C
Conversely, if D and E are points in the sides AB and AC respectively, such that
AD AE
= , then DE | | BC
DB EC
2. If in the two triangles, the sides of one triangle are proportional to those of the other then the
corresponding angle of the two triangle s are equal.
AB AC BC
In figure, if = =
DE DF EF
D
A
B C E F
then ∠ A = ∠ D, ∠ B = ∠ E, ∠ C = ∠ F
3. If in two triangles the angles of one triangle are equal to those of the other, then side opposite to
these angles are proportional. In the previous figure
if ∠ A = ∠ D, ∠ B = ∠ E, ∠ C = ∠ F
AB AC BC
then = =
DE DF EF
4. If in two triangles, one angle of one triangle is equal to one angle of the other triangle and the side
containing these angles are proportional, then two triangle are similar.
AB AC
In the figure, if ∠ A = ∠ D and = , then ∆ ABC and ∆ DEF are similar.
DE DF
5. The internal (respectively external) bisector of an angle of a triangle divides the opposite side
internally (respectively externally) in the ratio of the sides containing the angle.
E
A A
B C B D
D C
www.pdfworld.in
www.pdfworld.in
Geometry 511
AB BD
In figure (a), AD is the internal bisector of ∠ A Consequently =
AC DC
AB BD
In figure (b), AD is the external bisector of ∠ A Consequently =
AC DC
6. The areas of similar triangles are in the ratio of squares of the corresponding sides.
∆ ABC ~ ∆ DEF
Correspondence being
A ↔ D, B ↔ E , C ↔ F
D
A
B C E F
[ABC ] AB 2
Consequently =
[DEF ] DE 2
AC 2 BC 2
= 2
=
DF EF 2
[ABC ] and [DEF ] denote the area of ∆ABC and ∆DEF respectively.
Definitions
1. A line segment joining a vertex of a triangle to any point on the opposite side (the point may be on
the extension of the opposite side also) is now called a Cevian.
Altitudes, medians, angle bisectors are all Cevians.
2. Three points A , B , C are said to be collinear, if they lie on a straight line.
3. Three straight lines are said to be concurrent, if all three passes through a point.
Theorem 1 If A , B , C and A ′ , B ′ , C ′ are point an two parallel lines such that AB/A ′ B ′ = BC/B ′ C ′, then
AA ′ , BB ′ , CC ′ are concurrent, if they are not parallel.
O
Proof Let AA ′ and BB ′ meet at O when AA ′ and BB ′ not parallel. Join OC and
let it cut A ′ B ′ at C1. By Basic proportionality theorem
BC AB BC
= = A' C1
B ′ C1 A ′ B ′ B ′ C ′ B' C'
⇒ B ′ C1 = B ′ C ′
⇒ C1 and C ′ coincide. Thus, CC′ passes through O.
A B C
Theorem 2 From a point O ; OD, OE , OF are drawn perpendicular to the sides BC , CA and AB
respectively of a ∆ABC, then
BD2 − DC 2 + CE 2 − EA 2 + AF 2 − FB 2 = 0
A
F E
O
B D C
www.pdfworld.in
www.pdfworld.in
Proof BD2 = OB 2 − OD2

and DC 2 = OC 2 − OD2
⇒ BD2 − DC 2 = OB 2 − OC 2 …(i)
Similarly,
CE 2 − EA 2 = OC 2 − OA 2 …(ii)
AF 2 − FB 2 = OA 2 − OB 2 …(iii)
On adding Eqs. (i), (ii) and (iii), we get
BD2 − DC 2 + CE 2 − EA 2+ AF 2 − FB 2 = 0
Theorem 3 If D, E , F be points on sides BC , CA , AB of a ∆ ABC such that

BD2 − DC 2 + CE 2 − EA 2 + AF 2 − FB 2 = 0,
then perpendiculars at D, E , F to the respective sides are concurrent.
Proof Let the perpendiculars at D, E to BC , CA respectively meet at O. Let OF ′ be the perpendicular
from O to AB. Using theorem 2 we have
BD2 − DC 2 + CE 2 − EA 2 + AF ′2 − F ′ B 2 = 0 … (i) A
But BD2 − DC 2 + CE 2 − EA 2 +AF 2 − FB 2 = 0 …(ii) E

⇒ AF ′ − F ′ B = AF − FB
2 2 2 2 F’
or (AF ′ + F ′ B ) (AF ′− FB ′ ) F O
= (AF + FB ) (AF − FB )
⇒ AB (AF ′ − F ′ B ) = AB (AF − FB ) B D C
⇒ AF ′ − F ′ B = AF − FB [ QAB ≠ 0 ]
AF − AF ′ = FB − F ′ B
FF ′ = − FF ′
2FF ′ = 0 ⇒ FF ′ = 0
is F , F ′ coincides
Theorem 4 CEVAS' S THEOREM

ABC is a triangle and AX , BY and CZ are three Concurrent cevians. Then,
A
Y
Z
B X C
BX CY AZ
. . =1
XC YA ZB
Proof Let the cevians meet at P. Let [ ABC ] denote the area of ∆ ABC
1
∴ [ BPX ] = BX ⋅ h
2
1
and [ CPX ] = XC ⋅ h
2
www.pdfworld.in
www.pdfworld.in
Geometry 513
h = length of perpendicular from P to BC, then

1
BX ⋅ h
[BPX ] 2 BX
= = …(i)
[CPX ] 1 XC ⋅ h XC
2
If h ′ = length of perpendicular from A to BC, then
1
BX ⋅ h ′
[BAX ] 2 BX
= = …(ii)
[CAX ] 1 XC ⋅ h ′ XC
2
BX [BAX ] [BPX ]
= = …(iii)
XC [CAX ] [CPX ]
a c a c a −c
from elementary algebra, if = , then each of the ratios , =
b d b d b −d
BX [BAX ] − [BPX ] [APB ]
So, from Eq. (iii) = =
XC [CAX ] − [CPX ] [APC ]
BX [APB ]
∴ =
XC [APC ]
CY [BPC ] AZ [APC ]
Similarly, = and =
YA [APB ] ZB [BPC ]
BX CY AZ
Multiplying, we get . . =1
XC YA ZB
Theorem 5 CONVERSE OF THEOREM 4.
BX CY AZ
If 3 Cevians AX , BY , CZ satisfy . . = 1, then they are concurrent.
XC YA ZB
A
Y
Z
B X’ X C
Proof Let BY and CZ meet at P. Let AP meet BC at X ′. By ceva's theorem

BX ′ CY AZ
⋅ ⋅ = +1 …(i)
X ′ C YA ZB
BX CY AZ
⋅ ⋅ = +1 …(ii)
XC YA ZB
BX ′ BX
We have =
X ′ C XC
Adding 1 to both sides
BX ′ BX
+1 = +1
X′C XC
www.pdfworld.in
www.pdfworld.in
BX ′ + X ′ C BX + XC
=
X′C XC
BC BC
⇒ =
X ′ C XC
X ′ C = XC ⇒ X ′ C − XC = 0
⇒ XX ′ = 0
∴ X and X ′ Coincides. Thus, the 3 cevians are concurrent.
Theorem 6 MENELAUS THEOREM

If a transversal cuts the sides BC , CA , AB of a ∆ ABC at X , Y , Z respectively.
Then,
BX CY AZ
⋅ ⋅ = −1
XC YA ZB
Proof Let h1 , h2 , h3 be lengths of ⊥s AP , BQ , CR respective from A , B , C A
on the transversal.
Q
∆ BQX and ∆ CRX are similar. Z h1
BX h2
∴ = …(i) h2 P
XC h3 YR
Similarly, h3
CY h3 B C X
= …(ii)
YA h1
AZ h1
and = …(iii)
ZB h2
BX CY AZ h2 h3 h1
Hence, ⋅ ⋅ = ⋅ ⋅ =1
XC YA ZB h3 h1 h2
BX
As per the directed lengths we have is negative. The other two ratios are positive.
XC
BX CY AZ
⋅ ⋅ = −1
XC YA ZB
Theorem 7 (CONVERSE OF THEOREM 6.)

If X , Y , Z are three points on each of the sides BC , CA , AB of ∆ ABC or on their extensions such that
BX CY AZ
⋅ ⋅ = −1, then X , Y , Z are collinear.
XC YA ZB
Proof Let us give the proof of this theorem for a transversal which cuts all the 3 sides externally.
X , Y , Z are points on the extensions of BC , CA and AB respectivey. Such that
Y
Z
A
B C X X'
BX CY AZ
⋅ ⋅ = −1 …(i)
XC YA ZB
ProduceYZ to meet BC extended at X ′. As in the case of converse of ceva's theorem we shall prove that
X , X ′ coincide. We have MENELAUS THEOREM.
www.pdfworld.in
www.pdfworld.in
Geometry 515
BX ′ CY AZ
⋅ ⋅ = −1 …(ii)
X ′ C YA ZB
BX BX ′
=
XC X ′ C
Subtract 1 from both sides.
BX BX ′
−1= −1
XC X ′C
BX − XC BX ′ − X ′
i.e., =
XC X ′C
BC BC
i.e., =
XC X ′ C
⇒ X ′ C = XC
or X ′ C − XC = 0
i.e., XX ′ = 0 or X , X ′ Coincides.
Theorem 8 PAPPUS THEOREM

If A , C , E are three points on one straight line. B , D, F on another and if the three lines AB , CD, EF meet
respectively DE , FA and BC at L , M , N , then these three points L , M , N are collinear.
E C A
L M
N
B F D
Proof Try yourself

For students convenience the adjacent diagram is given.
U
C
E A
N W
L M
V
B F D
Use Menelaus theorem for ∆ UVW.

Considering the transversals LDE , AMF , BCN , ACE , BDF
Concept The square of the length of the internal bisector of ∠ A of ∆ ABC is 4bcs (s −2 a ).
(b + c )
Proof Q D divides BC internally in AB : AC
ac bc
∴ BD = ; DC =
b+c b+c
∴ AD2 = AB ⋅ AC − BD ⋅ DC
ac bc bc 4bcs (s − a )
= bc − ⋅ = {(b + c )2 − a 2} =
b + c b + c (b + c )2 (b + c )2
www.pdfworld.in
www.pdfworld.in
Theorem 9 STEINER LEHMUS PROBLEM

Prove that, if two angle bisector of a triangle are equal, then triangle is isosceles.
Proof 1 Suppose BE and CF are internal bisectors of ∠ B and ∠ C, are equal.
We know that,
4ca (s − b ) 4abs (s − c )
BE = , CF =
(c + a )2 (a + b )2
4acs (s − b ) 4abs (s − c )
∴ =
(c + a )2 (a + b )2
⇒ c (a + b )2(s − b ) = b (c + a )2 (s − c )
⇒ s {c (a + b )2 − b (c + a )2} + bc { (c + a )2 − (a + b )2} = 0
⇒ s {c (a 2 + b 2 ) − b (c 2 + a 2 )} + bc (c − b ) (2a + b + c ) = 0
⇒ s (c − b ) (a 2 − bc ) + bc (c − b ) (2s + a ) = 0
⇒ (c − b ) {bcs + a 2s + abc } = 0
⇒ c =b
∴ bcs + a 2s + abs > 0
Proof 2 AD and BE are bisectors of ∠ A and ∠ B in ∆ ABC G
F
They, intersect at K C
AD = BE (given) D
Draw ∠ BEF and ∠ EBF are equal to ∠ BAD and ∠ ADB respectively. Draw AH and
FG ⊥ AC and FB E K
B
(a) ∆ ADB ≅ ∆ EBF [by ASA]
AB = EF
900
[QAD = BE ; ∠ DAB = ∠ FEB, ∠ ADB = ∠ EBF] F
H
A
DB = BF
(b) ∠ AEF = ∠ AEK + ∠ KEF
= ∠ AEK + ∠ EAK
= ∠ AKB = ∠ KDB + ∠KBD
= ∠ EBF + ∠ EBA = ∠ ABF
∴ ∠ FEG = ∠ ABH
(c) ∆ ABH ≅ ∆ FEG [By (a) and (b) and construction]
∴ AH = FG
BH = EG
(d) ∆ AFG ≅ ∆ FAH (hypotenuse and length)
AG = FH
(e) AE = AG − GE = FH − HB
FB = BD [By (a)]
(f ) ∆ ABE ≅ ∆ BAD (SSS) ⇒ ∠ EAB = ∠ DBA
⇒ ∠A =∠B
⇒ CB = CA
www.pdfworld.in
www.pdfworld.in
Geometry 517
Basic Rules for a Triangle

In a ∆ ABC, the angles are denoted by capital letters A , B and C; and the A
lengths of the sides opposite to these angles are denoted by small letters
a , b and c respectively.
Simiperimeter of the triangle is c b
a+b+c
s=
2
and its are is denoted by ∆.
B a C
I. Sine rule In any ∆ABC
a b c
= = = 2R
sin A sin B sin C
where R is the radius of the circumcircle of the ∆ABC.
II. Cosine rule In any ∆ABC
a 2 = b 2 + c 2 − 2bc cos A
b2 + c 2 − a2
or cos A =
2bc
a2 + c 2 − b2
Similarly, cos B =
2ac
a2 + b2 − c 2
and cos C =
2ab
III. Projection rule In any ∆ABC
a = b cos C + c cos B
b = c cos A + a cos C
c = a cos B + b cos A
IV. Area of a triangle
1 1
∆= bc sin A = ca sin B
2 2
1 abc
= ab sin C = s (s − a )(s − b )(s − c ) = = rs
2 4R
where R and r are the radii of the circumcircle and the incircle of ∆ABC respectively.
Theorem 1 If D be a point on the side BC of a ∆ABC such that BD : DC = m : n and ∠ADC = θ,

∠BAD = α and ∠DAC = β, then prove that
(i) (m + n ) cot θ = m cot α − n cot β (ii) (m + n ) cot θ = n cot B − m cot C
Proof (i) Given, BD = m and ∠ADC = θ
DC n
∴ ∠ADB = (180° − θ), ∠BAD = α and ∠DAC = β
∴ ∠ABD = 180° − (α + 180° − θ) = θ − α
and ∠ACD = 180° − (θ + β )
BD AD A
From ∆ABD, =
sin α sin (θ − α ) α
β
DC AD
From ∆ADC, =
sin β sin [180° − (θ + β )]
DC AD 180° _ θ
or = θ
sin β sin (θ + β ) B m D n C
www.pdfworld.in
www.pdfworld.in
Dividing Eq. (i) by Eq. (ii), we get

BD sin β sin (θ + β )
=
DC sin α sin (θ − α )
m sin β sin θ cos β + cos θ sin β
or =
n sin α sin θ cos α − cos θ sin α
or m sin θ sin β cos α − m cos θ sin α sin β
= n sin α sin θ cos β + n sin α cos θ sin β
or m cot α − m cot θ = n cot β + n cot θ
[dividing both sides by sin α sin θ sin β]
or (m + n ) cot θ = m cot α − n cot β
BD m
(ii) Given, = and ∠ADC = θ
DC n
∴ ∠ADB = 180° − θ
∠ABD = B and ∠ACD = C
∴ ∠BAD = 180° − (180° − θ + B ) = θ − B
and ∠DAC = 180° − (θ + C )
BD AD
Now from ∆ABD, = …(i)
sin (θ − B ) sin B
and from ∆ADC,
DC AD
=
sin [180° − (θ + C )] sin C
DC AD
or = …(ii)
sin (θ + C ) sin C
Dividing Eq. (i) by Eq. (ii), we get
BD sin (θ + C ) sin C
⋅ =
DC sin (θ − B ) sin B
m (sin θ cos C + cos θ sin C ) sin C
or =
n (sin θ cos B − cos θ sin B ) sin B
or m sin θ cos C sin B + m cos θ sin C sin B
= n sin θ sin C cos B − n cos θ sin B sin C
or m cot C + m cot θ = n cot B − n cot θ
[dividing both sides by sin B sin C sin θ]
or (m + n ) cot θ = n cot B − m cot C
Theorem 2 STEWART’S THEOREM

Let AX be a cevian of length p divide BC into segments BX = m , XC = n , then A
a (p 2 + mn ) = b 2m + c 2n
Proof Using cosine rule for ∆ABX and ∆ACX
p2 + m2 − c 2
cos θ = …(i)
2pm
θ 180° _ θ
p 2 + n2 − b 2
cos (180 − θ) = = − cos θ …(ii) B m X n C
2pn
Where ∠AXD = θ and ∠AXC = 180° − θ
www.pdfworld.in
www.pdfworld.in
Geometry 519
p2 + m2 − c 2 p 2 + n 2 − b 2 
So, = − 
2pm  2pm 
m (p 2 + n 2 − b 2 ) = − n (p 2 + m 2 − c 2 )
or p 2m + nm 2 − b 2m = − p 2n − nm 2 + c 2n
ie, b 2m + c 2n = p 2m + p 2n + mn 2 + nm 2
= p 2 (m + n ) + mn (m + n )
⇒ b m + c n = (m + n )(p 2 + mn )
2 2
⇒ b 2m + c 2n = a (p 2 + mn )
Example 1 Find the length of the medians of a triangle.

Solution Using Stewart’s theorem A
a a  a a
b 2
+ c 2 = a  AD 2 + ⋅ 
2 2  2 2
b2 + c 2 a 2 c b
i.e., = AD 2 +
2 4
b2 + c 2 a 2
i.e., AD =2
−
2 4 B a D a C
1 2 2
i.e., AD = 2b + 2c 2 − a 2
2
2
Example 2 If AD be the median through vertex A of a ∆ABC, then prove that

AB 2 + AC 2 = 2( AD 2 + BD 2 )
Solution Using Stewart’s theorem, we have
1
AD = 2b 2 + 2c 2 − a 2 A
2
1
⇒ AD 2 = ( 2b 2 + 2c 2 − a 2 )
4 c b
a2
⇒ 2AD 2 = b 2 + c 2 −
2
= AC 2 + AB 2 − 2BD 2 B a D a
2
C
2
⇒ 2( AD + BD ) = AC + AB
2 2 2 2
Construction of Triangles
A triangle can be constructed in each of the following cases
⇒ When all the three sides are given
⇒ When one side and two angles are given
⇒ When two sides and the included angle are given
But there are many other cases when it is possible to construct the triangle.
Let us take a very special case when two sides and one anlge (other than the included anlge) are given.
To construct a ∆ABC, when sides a and b and ∠A are given.
There may exist no one or two triangles depending on the relation between the given parts.
Now, there is possibility of having two triangles This case is called the Ambigous case.
We construct ∠A and cut off AC = b this fixes the vertex C.
With C as centre we draw an arc in order to locate (if possible) B.
www.pdfworld.in
www.pdfworld.in
(I) If A < 90°, then several cases arise

(i) if a < p (where p = b sin A is perpendicular from C on AX ), then the arc does not cut AX and no
triangle is possible.
C
a
b
A X
(i)
(ii) If a = p , then the arc touches AX . Therefore one triangle is possible and it is right angled.
C
b a
A (ii) X
(iii) If a > p, then the arc cuts AX, at two points both these points lie to the right of A, if a < b. One of them
lies to the right of A and other coincides with A if a = b [See fig. (iii) b] and one of them lies to the
right of A and other to left of A, if a > b [See fig. (iii) c].
C
b a a
A X
B1 B2
(iii) a
C C
b a a a
b
A A
B1 B2 B1 B2
(iii) b (iii) c
Thus, two triangles are possible if a b sin A .
A acute is called the Ambiguous case
(II) If A > 90°, then following cases arise
C a
a C
b
b
X A X
B2 B1 A
(i) a (i) b
www.pdfworld.in
www.pdfworld.in
Geometry 521
(i) If a ≤ b, the arc does not cut AX at any point to the right of A and no triangle is possible.
(ii) If a > b, then the arc cuts AX at two points, only one triangle is possible.
a a
b
X
B1 A B2
Now we will discuss the case by using sine formula

(I) If A < 90°, then following cases arise
(i) If a 1 and consequently no solution is possible
(ii) If a > b sin A , then Eq. (i) gives two values of B one of which is acute and the other obtuse if a ≤ b , then
A ≤ B , so that only the acute value of B is permissible and consequently there is one solution.
If a < b, then A 90°, then following cases arise
(i) If a ≤ b , then A ≤ B , so that B must also be an obtuse angle which is not possible.
Hence, no solution is possible.
(ii) If a > b, then only the acute value of B is permissible. Only one triangle is possible.
Having determined B (whenever there exists a permissible value of B), We determine C by the formula
C = 180° − (A + B )
The remaining side c is then found in SAS case. In the Ambiguous case the value of C and c
corresponding to the two values of B have to be found separately.
By use of logarithms
log sin B > 0 ⇒ sin B > 1 ⇒ No solution
log sin B = 0 ⇒ sin B = 1 ⇒ ∠B = 90°
log sin B < 0 ⇒ sin B < 1 . There are two values of B , B1 and B2
Remark
Next case can be determined by using Cosine formula.
If a , b , A are given, then cosine formula for a gives
a 2 = b 2 + c 2 − 2bc cos A
or c 2 − 2bc cos A + b 2 − a 2 = 0 …(i)
Solving Eq. (i) as a quadratic in c, we get
2b cos A ± {4b 2 cos2 A − 4 (b 2 − a 2 )}
c=
2
= b cos A ± a − b sin 2 A
2 2
…(ii)
Since, c is the length of the side of a triangle therefore it must be +ve.
Two different possibilites arise
www.pdfworld.in
www.pdfworld.in
(i) A < 90°. If A < 90° , cos A is +ve so that b cos A is +ve.
Three cases arise.
(a) If a < b sin A , then a 2 < b 2 sin 2 A so that a 2 − b 2 sin 2 A < 0. The two values of c are imaginary and no
triangle is possible.
(b) If a = b sin A , then a 2 = b 2 sin 2 A so that a 2 − b 2 sin 2 A = 0. There is only one value of c (= b cos A )
from Eq. (ii) which is +ve. Only one triangle is possible.
(c) If a > b sin A , then a 2 > b 2 sin 2 A so that a 2 − b 2 sin 2 A > 0. In this case, Eq. (ii) gives two real and
distinct values of c.
i.e., b cos A + a 2 − b 2 sin 2 A is +ve
and b cos A − a 2 − b 2 sin 2 A is +ve

If b 2 cos2 A > a 2 − b 2 sin 2 A
b2 > a2
b>a
Two triangles are possible if b > a only one triangle is possible, if b ≤ a
(ii) A > 90°. If A > 90°
cos A is − ve so that b cos A is − ve
The value of b cos A − a 2 − b 2 sin 2 A is – ve
The value of b cos A + a 2 − b 2 sin 2 A is +ve if
a 2 − b 2 sin 2
A > − b cos A
a − b sin A > b 2 cos2 A
2 2 2
a2 > b2
a>b
We find that when A > 90°
No triangle is possible when a ≤ b.
Only one triangle is possible when a > b.
Circumcircle
The perpendicular bisectors of the sides of a triangle are concurrent. The point of concurrence is the
centre of the circumcircle of the triangle. It is called the circumcentre of the triangle and its radius is
called the circumradius of the triangle and is denoted by
abc
R=
4∆
Where a , b , c are length of the sides of triangle
and ∆ = Area
Incircle
The internal bisectors of the angles of a triangle are concurrent. The point of concurrence is called the
incentre of the triangle.
The circle with this point as centre and the length of the perpendicular from this point on any one of the
sides of the triangle touches all the three sides of the triangle internally and is called the incircle of the
triangle.
www.pdfworld.in
www.pdfworld.in
Geometry 523
Its radius is called inradius of the triangle and is denoted by r.

∆
r =
s
where s = semiperimeter of the triangle
∆ = its area
Excircles
Given a ∆ ABC, there are four circles which touch all the three sides of the triangle. One of them is the
incircle, which touches all the sides internally. The other three touch the side externally and are called
excircles.
The excircle opposite A (respectively B , C ) is the one whose centre lies on the internal bisector of ∠ A
(respectively ∠ B , ∠ C ).
The centre of the excircle opposite A (respectively B , C) is usually denoted by I1 (respectively I 2 , I3)
Its radius is r1 ( respectively r2 r3).
∆ ∆ ∆
r1 = , r2 = , r3 =
s −a s −b s −c
Nine Point on Circle

Prove that in any triangle the mid point of the sides, the feet of perpendiculars from the vertices on the
opposite sides and the mid points of the joins of the orthocentre to the vertices all lie on a circle.
Let X , Y be the feet of perpendiculars drawn from A and B on BC and CA respectively.
Let O be the orthocentre of the triangle. Let P be the mid point of AO.
A
P
Y
F O E
B C
D X
In ∆ ABO, F is the mid point of AB.

P is the mid point of AO.
∴ FB | | BO
In ∆ ABC,
F is mid point of AB.
D is mid point of BC.
∴ FD | | AC
Now, FP | | BY , FD | | AC and BY ⊥ AC
∴ FP⊥ FD i.e., ∠ DFP is a rigtht angle.
Similarly, ∠ PED is a right angle. Also ∠ PXD is a right angle.
∴F , E and X all lie on a circle with PD as diameter so that P and X both lie on the circle through the
points D , E , F .
Similarly, we can show that, if Z be the foot of perpendicular. from C on AB. Q and R be the mid points of
BO and CO respectively, then Y , Z , Q , R also lie on this circle. Thus the nine points
D, E , F , P , Q , R , X , Y , Z all lie on a circle.
www.pdfworld.in
www.pdfworld.in
Theorem 1 Prove that the nine point centre of a triangle is collinear with the circumcentre and the
orthocentre and bisects the segment joining them. Also prove that radius of the nine point circle of a
triangle is half the radius of the circumcircle.
Proof Let S be the circumcentre of ∆ ABC.
A
P
E
F S N
Y
O
B C
D U X
QD and X lie on nine point circle.
∴ Its centre lies on the perpendicular bisector of DX . Let U be mid point of DX . Let the perpendicular
from U on BC meet SO at N . Since SD, NU and OX are parallel and DU = UX .
∴ SN = NO i.e., N is mid point of SO.
Now, to show that N is centre of the nine point of circle, we have to observe that nine point centre must
also lie on the perpendicular bisector of EY . This perpendicular will also meet SO in N . N is nine point
centre.
It follows that the circumcentre, nine points centre and orthocentre are all collinear.
The nine point centre is the mid point of the segment joining the circumcentre and orthocentre.
Now, to show that the radius of nine point circle is half the circumradius.
Since PD is a diameter of the nine point circle so N is mid point of PD.
Q SO and PD bisect each other at N
∴ S , D, O , P are vertices of of a parallelogram.
⇒ SD = PO = AP [ Q P is mid point of AO ]
Now, SD | | AP and SD = AP
∴ S , D, P , A are the vertices of a parallelogram
Consequently DP = SA
But SA is the circumradius of ∆ ABC.
∴ Radius of the nine point circle is half the radius of circumcircle.
Theorem 2 Prove that in any triangle the circumcentre, the centroid, the nine point centre and
the orthocentre are all collinear.
Proof Through P draw PG ′ || SO so as to meet AD in G′.
Let AD meet SO in G.
We will show that
AG′ = G′ G = GD
So as to conclude that G divides AD in 2 : 1.
Consequently it is the centroid of ∆ ABC. A
In ∆ AGO , P is mid point of AO and PG′ | | GO.
∴ G′ is mid point of AG.
G'
i.e., AG′ = G′ G. P
In ∆ PDG′ , N is the mid point of PD and NG | | PG′. S G
N
∴ G is mid point of G′ D i.e., G′ G = GD. O
2
Now, AG′ = G′ G = GD so that AG = AD B C
3 D U X
www.pdfworld.in
www.pdfworld.in
Geometry 525
Simson's Line F
A P
Prove that the feet of perpendiculars drawn from a point on the circumcircle of a
triangle on the sides are collinear.
E
Let D, E , F be the feet of perpendiculars drawn from a point P on the
circumcircle of triangle ABC on the sides BC , CA , AB respectively.
We shall prove that the points D, E , F are collinear by showing that
∠ PED + ∠ PEF = 180°
∠ PEA + ∠PFA = 180° B D C
∴ The points P , E , A , F are concyclic.
Consequently
∠ PEF = ∠ PAF …(i)
[Angles in the same segment]
Q ∠ PEC = ∠ PDC = 90°
∴ P , E , D, C are concyclic.
∴ ∠ PED + ∠ PCD = 180° …(ii)
Q P , A , B , C are concyclic.
∴ ∠ PAF = ∠ PCB …(iii)
from Eqs. (i) and (iii)
∠ PEF = ∠ PCD …(iv)
from Eqs. (ii) and (iv)
∠ PED + ∠ PEF = 2 right angles
Hence, D, E , F are collinear.
Some Important Properties Related to Circle

1. A chord subtends an angle of the same magnitude at a point at the circumference of each of the two
segments into which it divides the circle.
2. A chord subtends at the centre an angle twice the magnitude of the angle it subtends at the
circumferences.
3. A diameter subtends a right angle at any point of the circumference.
4. The line joining the centre of a circle to the midpoint of a chord is perpendicular to the chord.
5. The perpendicular from the centre of a circle to a chord bisects the chord.
6. If a line segment AB subtends a right angle at C, then C lies on the circle on AB as diameter.
7. The tangent to a circle at a point theorem is perpendicular to the radius through that point.
8. Alternate segment theorem If AB is a chord of a circle. BC is the tangent to the circle at B, then
∠ ABC = ∠ ADC, where D lies on the segment which is not included between AB and BC.
D
C
B
9. Tangents to a circle which are parallelogram touch the circle at the ends of a diameter which is
perpendicular to them.
10. The two tangents to a circle drawn from a point outside it are of equal length.
www.pdfworld.in
www.pdfworld.in
Some Definitions
1. A quadrilateral which has a circle passing through all its four vertices is called a cyclic quadrilateral.
The centre of the circle is called the centre of the quadrilateral.
2. A quadrilateral which has a circle touching all its four sides is called a circumscribed quadrilateral.
3. A quadrilateral which has both a circumcircle and an incircle is called a bicentric quadrilateral.
Note
1. Any rectanlge is a cyclic quadrilateral. The diagonals are equal and bisect each other so that the circle
with centre at the point of intersection of the diagonals and radius equal to half the length of a diagonal
passes through all the vertices of the rectangle.
2. Assume that a parallelogram is a cyclic quadrilateral and O is the centre of the D C
circumscribing circle. Let E be the point of intersection of the diagonals. As E is
the midpoint of both the diagonals, OE is perpendicular to both AC and BD which
is impossible unless O and E coincides. If O and E coincides BD and AC should E
be equal, so that ABCD is a rectangle. A B
i.e., a parallelogram is a cyclic quadrilateral if and only if it is a rectangle.
3. A cyclic quadrilateral need not always be a rectangle. Take a circle with centre O. AB is its diameter.
Take two points C and D one on each of the two semicircles determined by AB such that AC ≠ BD.
Thus, ACBD is a cyclic quadrilateral which is not a rectangle.
D
A B
4. A square is a circumscribed quadrilateral. If O is the point of intersection of its

diagonals, then O is equidistant from the sides (i.e., the perpendicular distance
of each side from O is the same i.e., half the length of the side). So the circle with
centre at O and radius equal to half the length of the side touches the side at their O
respective midpoints. Thus, a square is bicentric as it also have a circumcircle.
5. A rectangle which is not a square, is not a circumscribed

quadrilateral since the opposite sides being parallel tangents of a
circle, the line joining the points of contact is a diameter of the circle
perpendicular to them so the diameter is equal to either of the sides
of the rectangle. This is impossible unless the length and breadth of
the rectangle are the same. So, a rectangle is a circumscribed
quadrilateral if and only if it is a square.
D B
6. There are non square quadrilaterals which are bicentric. To construct one
such, take any circle. Draw pairs of perpendicular tangents from points A, B
not lying on the same diameter of the circle. So, ACBD is a resultant
quadrilateral which is bicentric.
A C
www.pdfworld.in
www.pdfworld.in
Geometry 527
Cyclic Quadrilateral
Theorem 1 If a quadrilateral is cyclic, then sum of each pair of opposite angles is180°.
A
D B
C
Proof ABCD is a cyclic quadrilateral with O is its centre.
Now, one of the angles subtended by BD at the centre is reflex.
1
∠ DAB = reflex∠ DOB
2
1
∠ DCB = ∠ DOB
2
1
Hence, ∠ DAB + ∠ DCB = (reflex ∠ DOB + ∠ DOB)
2
1
= × 360° = 180°
2
i.e., ∠ A + ∠ C = 180°
Q ∠ A + ∠ B + ∠ C + ∠ D = 360°
∴ ∠ B + ∠ D = 180°
Corollary The exterior angle of a cyclic quadrilateral is equal to the interior opposite angles.
Proof Let ABCD be cyclic.
A B
X
D
C
Extend AB to X
∠ ABC + ∠ XBC = 180° (linear pair)
∠ ABC + ∠ ADC = 180°
∠ XBC = ∠ ADC
Theorem 2 If in a quadrilateral the sum of a pair of opposite angles is180°, then it is cyclic.
Proof Let ABCD be a quadrilateral with ∠ B + ∠ D = 180°.Let Q be the circle passing through the
vertices A , B and C.
B D
D'
C
www.pdfworld.in
www.pdfworld.in
If it passes through D, then ABCD is a cyclic quadrilateral, if Q does not pass through D. Let CD meet Q at
D′. ABCD′ is cyclic.
By theorem 1, ∠ B + ∠ D ′ = 180°
But ∠B + ∠D = 180°
So that ∠ D = ∠ D ′. But ∠ AD ′ C = ∠ D ′ is the exterior angle of ∆ ADD ′
∠ AD ′ C = ∠ ADC + ∠ DAD ′
∠ DAD ′ = 0 i.e., D and D′ coincides.
ABCD is cyclic.
Theorem 3 Ptolemy's Theorem

In a cyclic quadrilateral the product of the diagonals is equal to the sum of the products of the pairs of
opposite sides.
Proof We need to prove that
AC ⋅ BD = AB ⋅ CD + AD ⋅ BC
Now, ABCD is cyclic quadrilateral choose a point on H on BD such that ∠ DAH = ∠ BAC
∠ ADH = ∠ ADB = ∠ ACB …(i) A
So, ∆ HAD ~ ∆ BAC
D B
DH AD H
Hence, = or AD ⋅ BC = AC ⋅ DH …(ii)
BC AC
By (1), ∠ AHD = ∠ ABC
Hence, ∠ AHB = ∠ ADC (supplementary angles)
Also ∠ ABH = ∠ ABD = ∠ ACD C
∆ HAB ~ ∆ DAC
HB AB
Hence, =
DC AC
i.e., HB ⋅ AC = DC ⋅ AB …(iii)
From Eqs. (ii) and (iii)
AB ⋅ DC + AD ⋅ BC = AC ⋅ (HB + HD)
= AC ⋅ BD
Note If the quadrilateral is a rectangle (whose diagonals are equal), then it is

cyclic.
In this case Ptolemy's theorem reduces to Pythagoras theorem. A B
AC = BD (diagonals)
AB = CD and AD = BC
D C
AC ⋅ BD = AB ⋅ CD + AD ⋅ BC
AC 2 = AB 2 + BC 2
Example 1 ABCD is a trapezium. AB | | CD and AD = BC

It is given that AB = a,CD = b, AD = BC = c
Find the lengths of the diagonals.
Solution ∠ ABC + ∠ BCD = 180° [QAB | | CD]
∠ ADC = ∠ BCD
www.pdfworld.in
www.pdfworld.in
Geometry 529
∠ ABC + ∠ ADC = 180°

∴ ABCD is cyclic [By theorem 2]
AB ⋅ CD + AD ⋅ BC = AC ⋅ BD [Ptolemy's theorem]
A a B
c c
D C
b
i.e., ab + c 2 = AC ⋅ BD
But AC = BD
So that AC 2 = ab + c 2
or AC = ab + c 2 = BD
Example 2 E is a point on the side AD of rectangle ABCD. So that DE = 6. Also DA = 8 ; DC = 6. If

CE extended meets the circumcircle of the rectangle at F . Find the lengths of DF and
FB.
F
D A
E
0
45
0
45
C B
Solution AC = BD = 82 + 62 = 10
∠ DCE = ∠ DEC = 45° [QCD = DE ]
CE = DE + CD = 6 2 2 2
Also ∠ FDB = ∠ FCB = 45°

∠DBF = ∠DCF = 45°
In ∆ BFD, ∠ FDB = ∠ FBD = 45°
So FD = FB
DE ⋅ EA = CE ⋅ EF
6 × 2 = 6 2 ⋅ EF
EF = 2 and CF = 7 2
CDFB is a cyclic quadrilateral
CD ⋅ FB + BC ⋅ FD = CF ⋅ BD [By Ptolemy's theorem]
(CD + BC ) BF = 7 2 ⋅10
14 × BF = 7 2 ⋅10
BF = 5 2 = FD
www.pdfworld.in
www.pdfworld.in
Example 3 A line drawn from vertex A of an equilateral ∆ ABC meets BC at D and the circumcircle
at P. Prove that
1 1 1
(a) PA = PB + PC (b) = +
PD PB PC
A
D
B C
Solution ABPC is cyclic quadrilateral

AB ⋅ PC + AC ⋅ PB = AP ⋅ BC [from Theorem 3]
But AB = BC = CA
So, PA = PB + PC …(i)
Dividing Eq. (i) By PB ⋅ PC
1 1 PA
We have + =
PC PB PB ⋅ PC
It is enough to prove that
PA PC
= …(ii)
PB PD
In ∆ ABP and ∆ CDP
∠ BAP = ∠ DCP
∠ APB = ∠ ACB = 60°
∠ APC = ∠ ABC = 60°
Hence, ∠ APB = ∠ APC = ∠ DPC
∴ ∆ ABP ~ ∆ CDP
PA PB
So =
PC PD
PA PC
or =
PB PD
Non-cyclic Quadrilaterals
Theorem 1 ABCD is a quadrilateral with AB = a , BC = b , CD = c , DA = d , AC = m and BD = n.
Then, m 2 n 2 = a 2 c 2 + b 2 d 2 − 2abcd cos (A + C )
c
D C
m
d b
n
A B
a
www.pdfworld.in
www.pdfworld.in
Geometry 531
Proof Construct a ∆ ABE ~ ∆ CAD on side AB so that ∠ ABE = ∠ CAD and ∠ BAE = ∠ ACB.
AB AE BE
Now, = =
CA CD AD
ac ad
Hence, AE = and BE = …(i)
m m
Construct a ∆ ADF ~ ∆ CAB so that
∠ ADF = ∠ CAB and ∠ DAF = ∠ ACD
AD AF DF
We have = =
CA CB AB
bd
Hence, AF =
m
da
and DF = ...(ii)
m
From Eqs. (i) and (ii), we get, BE = DF
Now, ∠ EBD + ∠ BDF = 3 + ∠ ABD + ∠ BDA + 4
= ∠ ABD + ∠ BDA + ∠ BAD = 180°
So, BE | | DF and BE = DF
Hence, BEFD is a parallelogram
So EF = BD = n
∠ EAF = 1 + 2 + 3 + 4 = ∠ A + ∠ C
Using cosine rule in ∆ EAF, we get
EF 2 = AE 2 + AF 2 − 2AE ⋅ AF ⋅ cos ∠ EAF
a2 c 2 b2 d 2 2ac bd
i.e., n2 = + − . cos (∠ A + ∠ C )
m2 m2 m2 m
i.e., m 2n 2 = a 2c 2 + b 2d 2 −2abcd cos (∠ A + ∠ C )
Circumscribed Quadrilateral
Theorem 2 Let ABCD be a quadrilateral with an incircle. Then, the sum D
of the opposite sides are equal
S R
i.e., AB + CD = AD + BC
A C
Proof Let the incircle touch the sides AB , BC , CD, DA at P , Q , R and S
respectively,
P Q
Now, AP = AS
BP = BQ
CR = CQ B
DR = DS …(i)
So AB + CD = AP + BP + CR + DR
= AS + BQ + CQ + DS = AS + DS + BQ + CQ
= AD + BC
Hence AB + CD = AD + BC
www.pdfworld.in
www.pdfworld.in
Example 1 ABCD be a circumscribed quadrilateral, then prove that the circles inscribed in two
∆ s ABC and ∆ ADC touch each other.
C
B
Z1
P Z2
Q
A D
Solution Let the incircle of ∆ ABC be Z 1 and of ∆ ADC be Z 2.
Since, Z 1 and Z 2 lie on either side of AC, if they touch each other, they touch a point
on AC.
Let Z 1 touch AC at P and Z 2 touch AC at Q.
PQ = AQ − AP
1 1
= ( AD + AC − DC ) − ( AB + AC − BC )
2 2
1
= ( AD + BC − AB − CD ) = 0
2
Since, the quadrilateral is circumscribed.
Example 2 Let the incircle of ∆ ABC touch AB at D and let E be a point on the side AC. Prove that
the incircles of ∆ ADE , ∆ BCE and ∆ BDE have common tangents.
A
P
U E
D
R
N
V
B C
F M
t
Solution Let the incircle Q of ∆ ABC touch AB at D,BC at F and AC at G respectively. Let the
incircle Q1 of ∆ ADE touch the sides EA, AD and DE at P , Q and R respectively. Let the
incircle Z 2 of ∆ BCE touch the sides BC, CE , EB at M , N and L respectively Let t be the
common tangent of circles Z 1 and Z 2 meeting the lines DE , BE at S and T and
touching Z 1 at U and Q 2 at V respectively.
We have to prove that t touches the incircle of ∆ BED. It is enough to prove that the
quadrilateral BDST is a circumscribed quadrilateral since the incircle of BDST is in
circle of ∆ BDE
BD + ST = BF + UV − SU − TV
= BF + PN − SU − TV
[QUV , PN lengths of direct common tangents between Q1 and Q 2]
= BF + GP + GN − SR − TL
= BF + DQ + FM − SR − TL
= BM + DR − SR − TL
= BL − TL + DS = BT + DS
www.pdfworld.in
www.pdfworld.in
Geometry 533
Area Axiom
Every polygonal regoin has an area, measured in square units
There is a standard square region of side one metre called a square metre. Which is unit of area denoted
as m 2 or sq m and is a +ve real number.
Congruent Area Axiom

If ∆ ABC and ∆ PQR are two congruent angles, then area (region ∆ ABC) = area (region ∆ PQR).
i.e., two congruent regions have equal area.
Area Monotone Axiom
If R1 , R2 are two polygonal regions such that R1 ⊂ R2, then ar (R1 ) ≤ ar (R2 )
Area Addition Axiom

If R1 and R2 are two polygonal regions, whose intersection is a finite number of points and line segments
and R = R1 ∪ R2, then
ar (R ) = ar (R1 ) + ar (R2 )
Let us take some facts about the area of parallelogram.
1. A diagonal of a parallelogram divides it into two triangles of equal area.
2. Parallelogram on the same base and between the same parallel are equal in area.
3. The area of a parallelogram is the product of its base and the corresponding altitude.
Now, let us take some examples.
Example 1 The diagonals of a | | gm ABCD intersects at O. A line through O meets AB in X and CD

in Y. Show that
1
ar (quad. AXYD ) = ar ( | | gm ABCD )
2
D Y C
A X B
1
Solution ar ( ∆ ACD ) = ar (parallelogramABCD ) …(i)
2
[diagonal of a parallelogram divides it into two triangles of equal area]
In ∆ AOX and ∆ COY , we have
∠ AOX = ∠ COY (vertically opposite angles)
AO = CO [Qdiagonals of parallelogram bisect each other
∴ O is mid point of AC ]
∠ OAX = ∠ OCY [QAB | | DC and AC cuts them
∴∠ CAB = ∠ ACD ⇒ ∠ OAX = ∠ OCY ]
∆ AOX ≅ ∆ COY [ASA]
⇒ ar ( ∆ AOX ) = ar ( ∆COY )
www.pdfworld.in
www.pdfworld.in
⇒ ar ( ∆ AOX ) + ar (quad. AOYD )

= ar ( ∆ COY ) + ar (quad. AOYD )
⇒ ar (quad. AXYD ) = ar ( ∆ ACD ) …(ii)
1
ar (quad. AXYD ) = ar (parallelogram ABCD)
2
Example 2 Prove that of all parallelogram of which the sides are given , the parallelogram which is
rectangle has greatest area.
D C D C
b h
A B A B
a a
Solution Let ABCD be parallelogram in which AB = a and AD = b. Let h be its altitude

corresponding to base AB, then
ar ( | | gm ABCD ) = AB × h = ah
Q Sides a and b are given.
∴ With the same sides a and b we can construct infinitely many parallelogram with
different heights.
Now, ar ( | | gm ABCD ) = ah
⇒ ar ( | | gm ABCD ) is greatest when h is maximum. [Qa is constant]
But the maximum value which h can attains is AD = b and this is possible when
AD ⊥ AB,
i.e., | | gm ABCD becomes a rectangle.
Thus ar ( | | gm ABCD ) is greatest when AD⊥ AB i.e., when | | gm ABCD is rectangle.
Example 3 Show that a median of a triangle divides it into two triangles of equal area.
A
B L D C
Solution In ∆ ABC, AD is median. Draw AL ⊥ BC. D is mid point of BC.

⇒ BD = DC
⇒ BD × AL = DC × AL (Multiplying by AL)
1 1
⇒ (BD × AL ) = (DC × AL )
2 2
⇒ ar ( ∆ ABD ) = ar ( ∆ ADC )
www.pdfworld.in
www.pdfworld.in
Geometry 535
Example 4 AD is one of the medians of a ∆ ABC. X is any point on AD. Show that
A
B D C
ar ( ∆ ABX ) = ar ( ∆ ACX )
Solution AD is a median in ∆ ABC
∴ ar ( ∆ ABD ) = ar ( ∆ ACD ) ...(i)
In ∆ XBC, XD is a median
∴ ar ( ∆ XBD ) = ar ( ∆ XCD ) …(ii)
ar ( ∆ ABD ) − ar ( ∆ XBD ) = ar ( ∆ ACD ) − ar ( ∆ XCD )
⇒ ar ( ∆ ABX ) = ar ( ∆ ACX )
Example 5 Show that the diagonals of a parallelogram divide it into four triangles of equal area.
D C
A B
Solution AC and BD intersect at O in | | gm ABCD.

Q The diagonals of a parallelogram bisect each other at the point of intersection.
∴ AO = OC
and OB = OD
Also, the median of a triangle divides it into two equal parts.
In ∆ ABC,BO is the median
∴ ar ( ∆ OAB ) = ar ( ∆ OBC ) …(i)
In ∆ BCD,CO is the median
∴ ar ( ∆ OBC ) = ar ( ∆ OCD ) …(ii)
In ∆ ACD,DO is the median.
∴ ar ( ∆ OCD ) = ar ( ∆ OAD ) …(iii)
from Eqs. (i), (ii) and (iii), we get
ar ( ∆ OAB ) = ar ( ∆ OBC )
= ar ( ∆ OCD )
= ar ( ∆ OAD )
www.pdfworld.in
www.pdfworld.in
Example 6 The diagonals of a ABCD, AC and BD intersect in O. Prove that if BO = OD, the ∆ ABC
and ∆ ADC are equal in area.
D C
A B
Solution In ∆ ABD, we have

BO = OD
⇒ O is mid point of BD
⇒ AO is the median.
⇒ ar ( ∆ AOB ) = ar ( ∆ AOD ) …(i)
[QMedian divides ∆ into two triangles of equal area]
In ∆ CBD, O is the mid point of BD
∴ CO is a median.
⇒ ar ( ∆ COB ) = ar ( ∆ COD ) …(ii)
On adding Eqs. (i) and (ii), we have
ar ( ∆ AOB ) + ar ( ∆ COB ) = ar ( ∆ AOD ) + ar ( ∆ COD )
⇒ ar ( ∆ ABC ) = ar ( ∆ ADC )
Example 7 ABCD is a parallelogram. O is any point in its interior.

Prove that
(a) ar ( ∆ AOB ) + ar ( ∆ COD ) = ar ( ∆ BOC ) + ar ( ∆ AOD )
1
(b) ar ( ∆ AOB ) + ar ( ∆ COD ) = ar ( | | gm ABCD )
2
Solution (a) Q GN | | AD and EF | | DC
∴ OG | | DE and OE | | GD
⇒ EOGD is parallelogram
G
D C
E F
O
A B
H
Similarly, EAHO , HBFO and FOGC are parallelogram
Now, OD is a diagonal of | | gm EOGD
⇒ ar ( ∆EOD ) = ar ( ∆ DOG ) …(i)
OA is a diagonal of | | gm EAHO
⇒ ar ( ∆ EOA ) = ar ( ∆ AOH ) …(ii)
www.pdfworld.in
www.pdfworld.in
Geometry 537
OB is a diagonal of | | gm HBFO
⇒ ar ( ∆ BOF ) = ar ( ∆ BOH ) …(iii)
OC is a diagonal of | | gm FOGC.
⇒ ar ( ∆ FOC ) = ar ( ∆ COG ) …(iv)
Adding Eqs. (i),(ii),(iii),(iv), we get
ar ( ∆ EOD ) + ar ( ∆ EOA ) + ar ( ∆ BOF ) + ar ( ∆ FOC )
= ar ( ∆ DOG ) + ar ( ∆ AOH ) + ar ( ∆ BOH ) + ar ( ∆ COG )
⇒ ar ( ∆ AOD ) + ar ( ∆ BOC ) = ar ( ∆ AOB ) + ar ( ∆COD )
(b) Q ∆ AOB and | | gm ABFE are on same base AB and between the same parallel
lines AB and EF
1
∴ ar ( ∆ AOB ) = ar ( | | gm ABFE ) …(v)
2
Similarly,
1
ar ( ∆ COD ) = ar ( | | gm DEFC ) ...(vi)
2
On adding Eqs. (v) and (vi)
ar ( ∆ AOB ) + ar ( ∆ COD )
1
= ar ( | | gm ABCD )
2
Example 8 ABCD is a parallelogram. P is any point within it. Show that the sum of the areas of
∆PAB, ∆PCD is equal to half the area of parallelogram.
D C
A B
Solution If AB = CD = a
Let x , y be the lengths of the perpendicular from P on AB and CD respectively.
1 1
Then, ∆ PAB + ∆ PCD = ax + ay
2 2
1 1
= a ( x + y ) = ah
2 2
1
= × area of | | gm ABCD
2
Where h is height of parallelogram.
www.pdfworld.in
www.pdfworld.in

Example 1. A ball of diameter 13 cm is floating so that top of the ball is 4 cm above the smooth surface of
the pond. What is the circumference (in cm) of the circle formed by the contact of water surface with the ball ?
Solution We should find the circumference of the circle on AB as diameter.
C
4cm
A D B
m
O 6.5c
CD = 4 cm
13
OC = OB = = 6.5 cm
2
OD = 65
. cm − 4 cm = 25
. cm
So, DB = . )2 − (25
(65 . )2 = 6 cm
So, circumference of circle

= 2 π × 6 = 12 π cm
Example 2. OPQ is a quadrant of a circle and semicircles are drawn on OP and OQ. Show that shaded
areas a and b are equal.
Solution Area of quadrant = areas of two semi circles + b − a.
O P
2 2
π r 2 = π   + π   + b − a
1 1 r 1 r
i.e.,
4 2  2 2  2
1 1
⇒ π r2 = π r2 + b − a
4 4
⇒ b −a = 0
⇒ a =b
www.pdfworld.in
www.pdfworld.in
Geometry 539
Example 3. AB is a line segment of length 48 cm, C is its middle point. On AB , AC , CB semicircles are
described.Determine the radius of the circle inscribed in the space enclosed by three semicircles.
Solution Let X , Y be the centres of the semi circles described on AC , CB respectively as diameters O is
the centre of the circle inscribed in the space enclosed by the three semicircles and r the radius of this circle
AX = XC = 12 cm
OC ⊥ AB
OC = CD − OD = 24 − r
OX = OY = 12 + r
∆ XCO is right angled at C
So, OX 2 = XC 2 + OC 2
⇒ (12 + r )2 = 122 + (24 − r )2
⇒ 72r = 576 ⇒ r = 8
∴ Radius of circle is 8 cm.
Example 4. The radii of two circles in a plane are 13 and 18. Let AB be a diameter of the larger circle and
BC a chord of larger circle tangent to smaller circle at D. Find AD.
Solution Q CB = 13 cm , CD = 8 cm
∴ BD = 132 − 82 = 105 cm
In ∆ ADB, C is mid point of AB.
By Appolonius theorem
AD2 + BD2 = 2 (AC 2 + CD2 )
⇒ AD2 = 2 (132 + 82 ) − 105 = 361
∴ AD = 19 cm
Example 5. Let A be the area of a square inscribed in a circle of radius r. Let B be the area of a hexagon
inscribed in same circle. Obtain formula for A and B.
Compute the ratio B / A .

Solution Each side of a square inscribed in a circle of radiusr = r 2
∴ A = (r 2 )2 = 2r 2
Each side of a regular hexagon inscribed in a circle of radiusr = r
∴ B = 6 × area of an equilateral triangle having each side r
3 2 3 3r 2
=6× r =
4 2
3 3 2 
 r 
3 3 
∴ B/ A = 2 2 = 
 (2r )   4 
 
www.pdfworld.in
www.pdfworld.in
Example 6. ∆ ABC is an isosceles triangle. XY is drawn parallel to the base cutting the sides in X and Y ,
show that B , C , X , Y lies on a circle.
Solution QXY | | BC and AB meets them
∴ ∠ BXY + ∠ XBC = 2 right angles …(i)
Q AB = AC , ∠ B = ∠ C …(ii)
∠ BXY + ∠ BCY = 2 right angles
A
X Y
3
1 2
B C
Since, a pair of opposite angles of the quadrilateral. BCYX is supplementary.

∴Quadrilateral B C Y X is cyclic.
i.e., the points B , C , X , Y lie on circle.
Example 7. The angles of a polygon having no reflex angle are in AP. The smallest angle is 2 π / 3 radians
and common difference is 5°. Find the number of sides.
Solution Let the number of sides be n.
a the first term of AP
d common difference
Then, a = 120° , d = 5
n
∴ {2a + (n − 1) d }
2
= (2n − 4) × 90° = 180° n − 360°
We get n 2 − 25n + 144 = 0
So that n = 16, 9
If n = 16,
the greatest angle (in degrees)
= a + (n − 1) d = 120 + 155
. = 195
Which is not possible since polygon has no reflex angle.
The only other possibility being n = 9
It follows that the polygon has 9 sides.
Example 8. Show that a triangle with sides 3, 4, 5, is the only right angled triangle with integer sides,
whose semiperimeter equal its area.
Solution Let a , b be the legs and c the hypotenuse of a right angle triangle having its area equal to its
semiperimeter
Then, a2 + b2 = c 2 …(i)
1 1
(a + b + c ) = ab …(ii)
2 2
www.pdfworld.in
www.pdfworld.in
Geometry 541

[ab − (a + b )]2 = (a + b )2 − 2ab
ab − 2 (a + b ) + 2 = 0
∴ (a − 2) (b − 2) = 2
Q a , b are +ve integers
We must have either a − 2 = 2
b − 2 = 1 ; or a − 2 = 1, b − 2 = 2
⇒ either a = 4, b = 3 or a = 3, b = 4
In either case c = 5
Thus a triangle with sides 3, 4, 5 is the only right angled triangle with integer sides whose
semiperimeter equals its area.
Example 9. ABCD is a square of which no angle is 60°. Equilateral ∆ ADE and ∆DCF are drawn outwardly
on the sides AD and DC.
Show that ∆ ABE ≅ ∆ CFB
Solution AE = AD [Q∆ ADE is equilateral]
In ∆ ABE and ∆ CFB,
D C
B
A
AB = CF , AE = BC
∠ EAB = ∠ BCF
Therefore, ∆ ABE ≅ ∆ CFB
Example 10. ABCD is a parallelogram P , Q , R and S are points on sides AB , BC , CD, DA respectively, such
that AP = DR. If area of | | gm ABCD is 16 cm 2. Find area of the quadrilateral PQRS.
Solution Q AP = DR and AP | | DR
R C
D
A P B
∴ APRD is a parallelogram
Consequently PR | | AD.
www.pdfworld.in
www.pdfworld.in
Q ∆ PRS and | | gm PRDA have the same base PR and have the same altitude.
1
∴ ∆ PQR = × ar ( | | gm PBCR)
2
ar (quad. PSRQ ) = ar (∆ PSR ) + ar (∆ PQR)
1 1
= × ar ( | | gm PRDA) + × ar ( | | gm PRCB)
2 2
1 1
= × ar ( | | gm ABCD) = × 16 cm 2 = 8 cm 2
2 2
Example 11. XY is a line parallel to side BC of ∆ ABC. BE | | AC and CF | | AB meet XY in E and F

respectively.
Show that ar (∆ ABE ) = ar (∆ ACF )
Solution Q XY | | BC and BE | | CY
A
E F
X Y
B C
∴ BCYE is a parallelogram.
Q∆ ABE and | | gm BCYE are on same base BE and between the same parallel lines BE and AC
1
∴ ar ( ∆ ABE ) = ar ( | | gm BCYE ) …(i)
2
Now, CF | | AB and XY | | BC
⇒ BCFX is a parallelogram.
Q ∆ ACF and | | gm BCFX are on same base CF and between the same parallel lines AB and FC.
1
∴ ar (∆ ACF ) = ar ( | | gm BCFX ) …(ii)
2
But | | gm BCFX and | | gm BCYE are on same base BC and between the same parallel lines BC and EF
∴ ar ( | | gm BCFX ) = ar ( | | gm BCYE ) …(iii)
from Eqs. (i), (ii) and (iii), we get
ar (∆ ABE ) = ar (∆ ACF )
Example 12. If the medians of a ∆ ABC intersect at G. Show that

1
ar (∆ AGB ) = ar (∆ AGC ) = ar (∆BGC ) = ar (∆ ABC )
3
Solution We know that the median of a triangle divides it into two triangles of equal area.
A
F E
B C
D
www.pdfworld.in
www.pdfworld.in
Geometry 543
In ∆ ABC , AD is the median

⇒ ar (∆ ABD) = ar (∆ ACD) …(i)
In ∆ GBC , GD is the median
⇒ ar (∆ GBD) = ar (∆ GCD) …(ii)
ar (∆ ABD) − ar (∆ GBD)
= ar (∆ ACD) − ar (∆ GCD)
⇒ ar (∆ AGB ) = ar (∆ AGC ) …(iii)
Similarly,
ar (∆ AGB ) = ar (∆ BGC ) …(iv)
From Eqs. (iii) and (iv), we get
ar (∆ AGB ) = ar (∆ BGC ) = ar (∆ AGC )
But, ar (∆ AGB ) + ar (∆ ABGC ) + ar (∆ AGC ) = ar (∆ ABC )
∴ 3ar (∆ AGB ) = ar (∆ ABC )
1
⇒ ar (∆ AGB ) = ar (∆ ABC )
3
Hence, ar (∆ AGB ) = ar (∆ AGC )
1
= ar (∆ BGC ) = ar (∆ ABC )
3
Example 13. The side AB of a parallelogram ABCD is produed to any point P. A line through A parallel to
CP meets CB produced in Q and the parallellogram PBQR completed.
Show that ar ( | | gm ABCD) = ar ( | | gm BPRQ )
Solution Join AC and PQ
D C
A P
B
Q R
1
Q ar (∆ ABC ) = ar ( | | gm ABCD) …(i)
2
1
ar (∆ PBQ ) = ar ( | | gm BPRQ ) …(ii)
2
∆ ACQ and ∆ AQP are on the same base AQ and between same parallel AQ and CP.
∴ ar (∆ ACQ ) = ar (∆ AQP )
⇒ ar (∆ ACQ ) − ar (∆ ABQ )
= ar (∆ AQP ) − ar (∆ ABQ )
⇒ ar (∆ ABC ) = ar (∆ BPQ )
1 1
= ar (| | gm ABCD) = ar ( | | gm BPRQ )
2 2
[from Eqs. (i) and (ii)]
⇒ ar ( | | gm ABCD) = ar ( | | gm BPRQ )
www.pdfworld.in
www.pdfworld.in
Example 14. In a | | gm ABCD, E and F are any two points on sides AB and BC respectively.
Show that ar (∆ ADF ) = ar (∆ DCE )
Solution Draw EG | | AD and FH | | BA
Q FH | | AB
∴ ABFH is parallelogram
D G C
H F
A E B
AF is a diagonal of | | gm ABFH
1
∴ ar (∆ AFH ) = ar ( | | gm ABFH ) …(i)
2
In | | gm DCFH , DF is a diagonal
1
∴ ar (∆ DFH ) = ar ( | | gm DCFH ) …(ii)
2
ar (∆ AFH ) + ar (∆ DFH )
1 1
= ar ( | | gm ABFH ) + ar ( | | gm DCFH )
2 2
1
= [ ar ( | | gm ABFH ) + ar ( | | gm DCFH ) ]
2
1
= ar ( | | gm ABCD)
2
1
⇒ ar (∆ ADF ) = ar ( | | gm ABCD) …(iii)
2
In | | gm AEGD; DE is diagonal
1
∴ ar (∆ DEG) = ar ( | | gm AEGD) …(iv)
2
In | | gm CBEG ; CE is diagonal.
1
∴ ar (∆ CEG) = ar ( | | gm CBEG) …(v)
2
From Eqs. (iv) and (v), we get
ar (∆ DEG) + ar (∆ CEG)
1 1
= ar ( | | gm AEGD) + ar ( | | gm CBEG)
2 2
1
= [ ar ( | | gm AEGD) + ar ( | | gm CBEG) ]
2
1
= [ ar ( | | gm ABCD) ]
2
1
⇒ ar (∆ DCE ) = ar ( | | gm ABCD) …(vi)
2
From Eqs. (iii) and (vi), we get
ar (∆ ADF ) = ar (∆ DCE )
www.pdfworld.in
www.pdfworld.in
Geometry 545
Example 15. BC | | XY , BX | | CA and AB | | YC

Prove that ar (∆ ABX ) = ar (∆ ACY )
Solution Join XC and BY
Q∆ BXC and ∆ BCY are on same base BC and between the same parallel lines BC and XY .
∴ ar (∆ BXC ) = ar (∆ BCY ) …(i)
Also, ∆ BXC and ∆ ABX are on the same base BX and between same parallel lines BX and AC
B C
X Y
P Q
A
∴ ar (∆ BXC ) = ar (∆ ABX ) ...(ii)
∆ BCY and ∆ ACY are on the same base CY and between the same parallel AB and CY
∴ ar (∆ BCY ) = ar (∆ ACY ) …(iii)
From Eqs. (i), (ii) and (iii), we get
ar (∆ ABX ) = ar (∆ ACY )
Example 16. ABCD is a parallelogram. X and Y are mid points of BC and CD respectively. Prove that
3
ar (∆ AXY ) = ar ( | | gm ABCD).
8
Solution Join BD
QX and Y are mid points of side BC and CD respectively in ∆ BCD
D Y C
A B
∴ XY | | BD
1
and XY = BD
2
1
⇒ ar (∆ CYX ) = ar (∆ DBC )
4
1
= ar ( | | gm ABCD) …(i)
8
1
ar ( | | gm ABCD) = ar (∆ BCD)]
[Q
2
1
Q| | gm ABCD and ∆ ABX are between same parallel lines AD and BD and BX = BC
2
1
∴ ar (∆ ABX ) = ar ( | | gm ABCD) …(ii)
4
www.pdfworld.in
www.pdfworld.in
Similarly,
1
ar (∆ AYD) = ar ( | | gm ABCD) …(iii)
4
Now, ar (∆ AXY )
= ar ( | | gm ABCD) − [ar (∆ ABX ) + ar (∆ AYD) + ar (∆ CYX )]
= ar ( | | gm ABCD) −  + + 
1 1 1
 4 4 8
ar ( | | gm ABCD)
From Eqs. (i), (ii) and (iii)
 1 − 5  ar ( | | gm ABCD) = 3 ar ( | | gm ABCD)
 
 8 8
Example 17. Given two points A and B and +ve real number x . Find the locus of a point P such that
ar (∆ PAB ) = x
Solution Let the perpendicular distance of P from AB be h.
Then, ar (∆ PAB ) = x
P P
A B
P P
1
⇒ × AB × h = x
2
2x
⇒ h=
AB
Q AB and x are given.
∴ h is a fixed +ve real number.
Thus, the point P moves in such a way that its distance from AB is always same.
i.e., P lies on a line parallel to AB at a distance h from it.
But there are two such lines on either side of AB.
2x
Hence, the locus of P is a pair of lines at a distance h = parallel to AB.
AB
Example 18. Let G be the centroid of the ∆ ABC in which the angle at C is obtuse and let AD and CF be
medians from A and C respectively onto the sides BC and AB. If the four points B , D, G and F are concyclic.
AC
Show that > 2. If, further, P is a point on the line BG extended such that AGCP is a parallelogram, show
BC
that the ∆ ABC and ∆ GAP are similar.
Solution If ∠ADB = θ, then from ∆ ABD and ∆ ADC , we get
AB 2 = AD2 + BD2 − 2AD ⋅ BD ⋅ cos θ
AC 2 = AD2 + DC 2 + 2AD ⋅ DC cos θ
www.pdfworld.in
www.pdfworld.in
Geometry 547
Adding, A
1
a + b = 2AD + a 2
2 2 2
2
P
b2 + c 2 a2 F
i.e., AD2 = −
2 4
E
Similarly, the other medians are given by G
c 2 + a2 b2 θ
BE 2 = − , B C
2 4 D
a2 + b2 c 2
CF 2 = −
2 4
Since B , D, G, F lie on a circle we have
AF ⋅ AB = AG ⋅ AD
2
But AG = AD
3
1 2 2 1 a2 
Hence, c = AD2 =  b 2 + c 2 − 
2 3 3 2
3 2 1
⇒ c = b2 + c 2 − a2
2 2
1 2
⇒ b = (a + a )
2 2
2
For the first part we have
b 2 = c 2 + a 2 − 2ca cos B = 2b 2 −2ca cos B ;
i.e., b 2 = 2ca cos B ,
a = c cos B + b cos C < c cos B
b
Since C > 90°. Hence, 2a < 2c cos B = b 2 ; > 2.
2
a
This proves the first part.
For the second part, let the line passing through C and parallel to AG meet BG produced in P. Given that
AGCP is a parallelogram. So, AC and GP have the same mid point E.
2
Hence, GP = 2GE = BE
3
2 2
AG = AD, AP = CG = CF
3 3
1 2
Since, b = (c + a ), we get
2 2
2
1 1 1 3
AD = b 2 + c 2 − a 2 = c 2
2
2 2 4 4
1 2 1 2 1 2 3 2
BE = c + a − b = b
2
2 2 4 4
1 2 1 2 1 2 3 2
CF = a + b − c = a
2
2 2 4 4
2 3 1
Hence, AG = ⋅ c= c,
3 2 3
2 3 1
GP = ⋅ b= b,
3 2 3
www.pdfworld.in
www.pdfworld.in
2 3 1
PA = ⋅ a= a
3 2 3
AG GP PA 1
Thus, = = =
BA AC CB 3
So, AGP is similar to ABC.
Example 19. The incircle of ABC touches BC , CA and AB at D, E and F respectively. X is a point inside
∆ ABC such that the incircle of ∆ XBC touches BC at D also and touches CX and XB at Y and Z respectively.
Prove that EFZY is a cyclic quadrilateral.
Solution Let P be the intersection of EF with BC. Then, by Menelaus’ Theorem we have
BP CE AF A
⋅ ⋅ =1 …(i)
PC EA FB
Since, CE = CD, EA = AF and FB = BD, we get
BP CD
⋅ =1
PC BD F
so that
BP BD X
= …(ii) Z
Y
E
PC CD
Since, XZ = XY , BZ = BD and CY = CD, we have from
Eq. (ii)
B D C P
PB CY XZ BD CD XY
⋅ ⋅ = ⋅ ⋅ =1
PC YX ZB CD YX BD
Hence, by Menelaus’ Theorem P , Z and Y are collinear.
Since, PF ⋅ PE = PD2 and PZ ⋅ PY = PD2 we have
PF ⋅ PE = PZ ⋅ PY
Hence, EFZY is a cyclic quadrilateral.
Comment If AB = AC, then BD = DC and then it can easily be proved that AD is the perpendicular
bisector of EF and YZ so that EFZY is an isosceles trapezoid and is a cyclic trapezoid.
Example 20. Let A , B and C be non-collinear points. Prove that there is a unique point X in the plane of
ABC such that XA 2 + XB 2 + AB 2 = XB 2 + XC 2 + BC 2 = XC 2 + XA 2 + CA 2.
Solution First Solution :

From the hypothesis we have
AX 2 + AB 2 = CX 2 + CB 2 …(i)
If B1 is the mid point of BX , applying the first theorem of the median in the triangles ∆ABX , ∆CBX , we
get
2AB12 + 2BB12 = 2CB12 + 2BB12 or AB1 = CB1 …(ii)
B
This indicates that the perpendicular bisector of the side AC
passes through the point B1. Let A1 , C1 be the mid point of
AX and CX , respectively. B1
Similarly, we obtain that the perpendicular bisectors of
BC and AB pass through the mid points A1 and C1
respectively. …(iii)
Furthermore we obtain
AB | | A1B1 , AC | | A1C1 and BC | | B1C1. …(iv) A1 C1
A C
www.pdfworld.in
www.pdfworld.in
Geometry 549
From Eqs. (iii) and (iv) we get the circumcentre O of ABC is the orthocentre H1 of A1B1C1. …(v)
1
Also from Eq. (iv) the ∆ ABC and A1B1C1 are similar with X the centre of similarity and ratio . …(vi)
2
So, their orthocentres H and H1 lie in the same straight line with the point X and HH1 = H1X …(vii)
Combining Eqs. (v) and (vii) we get HO = OX ; that is the point X is known (constant), because X is
symmetric to H with respect to the orthocentre O of ABC.
Example 21. A hexagon is inscribed in a circle with radius r. Two of its sides have length 1, two have
length 2 and the last two have length 3. Prove that r is a root of the equation
2r 3 − 7r − 3 = 0
Solution Equal chords subtend equal angles at the centre of a circle, if each of sides of length i subtends
an angle α, (i = 1, 2, 3) at the centre of the given circle, then 2α1 + 2α 2 + 2α3 = 360°,
α1 α 2 α
whence + = 90° − 3 ,
2 2 2
 α1 α 2   α 
and cos  +  = cos  90° − 3 
2 2  2
α3
= sin
2
Now, we apply the addition formula for the cosine :
α α α α α
cos 1 cos 2 − sin 1 cos 2 = sin 3 , …(i)
2 2 2 2 2
where (see figures)
α1 1/2 α 4r 2 − 1
sin = ,cos 1 = ,
2 r 2 2r
α2 1 α r2 − 1
sin = , cos 2 = ,
2 r 2 r
α 3/2
sin 3 =
2 r
α/2 α/2 α/2

r r r
1 3
2 1 2
A B C
We substitute these expressions into Eq. (i) and obtain, after multiplying both sides by2r 2,
4r 2 − 1 ⋅ r 2 − 1 − 1 = 3r
Now, write it in the form
(4r 2 − 1)(r 2 − 1) = 3r + 1
and square, obtaining
(4r 2 − 1)(r 3 − 1) = 9r 2 + 6r + 1
www.pdfworld.in
www.pdfworld.in
r (2r 3 − 7r − 3) = 0
Since r ≠ 0, we have
2r 3 − 7r − 3 = 0
which was to be shown.
Example 22. Let ∆ABC be equilateral. On side AB produced, we choose a point P such that A lies between P
and B. We now denote a as the length of sides of ∆ABC; r1 as the radius of incircle of ∆PAC; and r2 as the
exradius of ∆PBC with respect to side BC. Determine the sum r1 + r2 as a function of a alone.
Solution Looking at the figure, we see that ∠T1O1R = 60° since it is the supplement of ∠T1AR = 120° (as
an exterior angle for ∆ABC). Hence, ∠AO1R = 30°. Similarly, we obtain ∠BO2S = 30°.
T2
O2
T1 r2
O1
r1 S
R
P T1 A B T2
Since, tangents drawn to a circle an external point are equal, we have

T1T2 = T1A + AB + BT2
= RA + AB + SB
r1 + r2
= r1 tan 30° + a + r2 tan 30° = +a
3
and T1 ′T2 ′ = T1 ′ C + CT2 ′
r1 + r2
= CR + CS = (a − RA ) + (a − SB ) = 2a −
3
Since, common external tangents to two circles are equal
T1T2 = T1 ′T2 ′
r1 + r2 r + r2
Hence, + a = 2a − 1 ,
3 3
whence we find that
a 3
r1 + r2 =
2
Example 23. Let ABC be a triangle and a circle Γ′ be drawn lying inside the triangle, touching its incircle
Γ externally and also touching the two sides AB and AC. Show that the ratio of the radii of the circles Γ′ and
π − A
Γ is equal to tan 2  .
 4 
Solution Let r , r ′ be the radii and I , I ′ be the centres of Γ , Γ ′ respectively.

r A r′
Then, = sin =
AI 2 AI ′
www.pdfworld.in
www.pdfworld.in
Geometry 551
A r − r′
Hence, sin = A
2 AI − AI ′
r − r′
= ;
II ′ I'
R'
sin A/2 r − r ′
= ;
1 r + r′
I r
1 − sin A/2 (r + r ′ ) − (r − r ′ ) 2r ′
= =
1 + sin A/2 (r + r ′ ) + (r − r ′ ) 2r
B C
r ′ 1 − cos (π/2 − A/2)
=
r 1 + cos (π/2 − A/2)
π − A
2 sin 2  
 4 
=
π − A
2 cos2  
 4 
π − A
= tan 2  
 4 
This proves the result.
Example 24. LetT be an acute triangle. Inscribe a pair R, S of rectangles in T as shown : Let A (X ) denote
A (R ) + A (S )
the area of polygon X. Find the maximum value, or show that no maximum exists, of , where T
A (T )
ranges over all triangles and R, S over all rectangles as above.
S z
a
R y
b
x
Solution As in the figure

A (R ) + A (S ) ay + bz
= ,
A (T ) hx/2
where h = a + b + c , the altitude of T . By similar triangles
x y z
= = ,
h b+c c
(b + c ) x cx
a +b
A (R ) + A (S ) h h = 2 (ab + ac + bc )
so =
A (T ) hx/2 h2
We need to maximize ab + ac + bc subject to a + b + c = h. One way to do this is first to fix a, so
b + c = h − a. Then,
ab + ac + bc = a (h − a ) + bc ,
and bc is maximized when b = c . We now wish to maximize 2ab + b 2 subject to a + 2b = h. This is a
straight-forward calculus problem giving a = b = c = h/3. Hence, the maximum ratio is 2/3 (independent
of T ).
www.pdfworld.in
www.pdfworld.in
Example 25. Given is a regular 7-gon ABCDEFG. The sides have length 1. Prove for the diagonals AC and
AD.
1 1
+ = 1.
AC AD
F F'
E E'
G
D
D'
A
C
B C'
B'
Solution Reflect the 7-gon with AG as an axis to obtain another 7-gon AB ′ C ′ D′ E ′ F ′ G′.
4π 3π
Since, ∠GAC + ∠GAD′ = + = π,
7 7
C , A and D′ are collinear, besides
π
∠GCA = ∠GD′ A =
7
= ∠CAB = ∠ACB
Therefore, ∆GCD ~ ∆BAC giving
AC CD′ AC + AD′ AC + AD
= = = .
AB CG AD AD
But AB = 1 so
1 1 AC + AD 1
+ = = = 1,
AC AD AC ⋅ AD AB
as required.
Example 26. The line l is tangent to the circle S at the point A; B and C are points on I on opposite sides of A
and the other tangents from B, C to S intersect at a point P. If B, C vary along l in such a way that the product
| AB | ⋅ | AC | is constant, find the locus of P.
Solution Let S be the incircle S (1, 2) of ∆BCP. We denote ∠PBA = β, ∠PCA = γ.
AB = pAC = q with pq = k 2 , a constant.
Let S touch BP and CP and D and E respectively. For ∆PEI
P
t
t E
r q
D r I γ/2
p r γ/2
β/2
β/2 q C
B p H A
1
we have ∠EIP = ( β + γ )
2
1 (p + q )r 2
Thus, t = r tan (β + γ ) =
2 pq − r 2
www.pdfworld.in
www.pdfworld.in
Geometry 553
The semiperimeter of ∆BCP is

(p + q )r 2 pq (p + q )
p + q + t =p + q + =
pq − r 2 pq − r 2
The area F, of ∆BCP is
pq (p + q ) 1
r = (p + q ) PH ,
pq − r 2 2
where PH is the altitude to BC. It follows immediately that
2pqr 2k 2r
PH = =
pq − r 2
k2 − r
So, the locus of P is a line parallel to BC.
Example 27. In a ∆ ABC, ∠A is twice ∠B. Show that a 2 = b (b + c ).

Solution A = 2B ⇒ ∠C = π − 3B, Since
a b c
= = ,
sin A sin B sin C
we have
a 2 = b (b + c ) ⇔ sin 2 A = sin B (sin B + sin C ) ⇔ sin 2 2B = sin B (sin B + sin 3B )
⇔ sin 2 2B = sin B ⋅ 2 sin 2B ⋅ cos B
⇔ sin 2 2B = 2 sin B cos B ⋅ sin 2B
⇔ sin 2 2B = sin 2 2B
Example 28. Let AC and BD be two chords of a circle with centre Q such that they intersect at right-angles
inside the circle at the point M. Suppose K and L are the mid points of the chords AB and CD respectively.
Prove that OKML is a parallelogram.
O L
A M C
K
B
Solution Choose M as the origin, AC as the x -axis and BD as the y-axis. Let the equation of the circle be
x 2 + y 2 + 2gx + 2fy + c = 0. The x -coordinates of A and C given by
x 2 + 2gx + c = 0
If A is (−g − g 2 − c , 0) , then C is (−g + g 2 − c , 0). The y-coordinates of B and D are given by
y 2 + 2fy + c = 0
If is (0, − f − f 2 − x ) , then D is (0, − f + f 2 − c ).
www.pdfworld.in
www.pdfworld.in
 −g − g 2 − c −f − f 2 − c 
Thus, K is  , ;
 2 2 
 
 −g + g 2 − c −f + f 2 − c 
L is  , .
 2 2 
 
 −g −f 
Hence, the mid point of KL is  ,  . Since, O is (−g , − f ) and M is (0, 0) the mid point of OM also is
 2 2
 −g −f 
 ,  . Hence, OM and KL have the same mid point and so OKML is a parallelogram.
 2 2
Aliter Let the radius of the circle be R , AB = a , CD = b ,
∠ADB = θ and∠CAD = φ . Then, θ + φ = 90° from the right angled triangle AMD.
Thus, a = AB = 2R sin θ,
b = CD = 2R sin φ = 2R cos θ,
a 2 + b 2 = 4R 2.
2 2
Hence, OK 2 = R 2 −  a  =  b  . But L is the circumcentre of the right angled ∆ CMD.
1 1
2  2 
1
Hence, b = LC = LM . Thus, OK = ML.
2
Similarly OL = KM . This proves the result. The arguments show that M may be outside the circle also.
Example 29. In ∆ABC, AB = AC and ∠BAC = 30°, A ′ , B ′ , C ′ are reflections of A , B , C respectively on

BC , CA , AB, show that ∆A ′ B ′ C ′ is equilateral.
Solution ∆A ′ BC , ∆ B ′ AC , ∆ C ′ AB are reflections of ∆ABC on BC , CA , AB respectively, ∠BAC = 90°, If
AB = AC = a , B ′ C ′ = 2 a since ∆B ′ AC ′ is a right angled isosceles triangle.
BC = 2a cos 75° = 2a cos ( 30° + 45° )
= 2a(cos 30° cos 45° − sin 30° sin 45° )
 3 1 1 1  A
= 2a  − 
 2 2 2 2
30° 30° 30° a
( 3 − 1)
= a
2 a
75° 75°
By the cosine formula (in ∆A ′ CB′ ). C' B'
(A ′ B ′ )2 = A ′ C 2 + CB ′2 − 2A ′ C ′ ⋅ CB ′ cos ∠A ′ CB ′ 75° 75°

75°
B C
2
 3 − 1  3 − 1 75° 75° 135°
= a2 +   − 2a   a cos 135°
 2   2 
(since CB ′ = BC)
  3 − 1
2
 3 − 1 1 
= a 2 1 +   +2  
  2   2  2
 
A'
= a 2 (1 + 2 − 3 + 3 − 1)
= 2a 2 (since cos 135° = cos (90 + 45° ) = − sin 45°)
Hence, A ′ B ′ = 2a = B ′ C ′. We can similarly show that A ′ C ′ = 2a so that ∆A ′ B ′ C ′ is equilateral.
www.pdfworld.in
www.pdfworld.in
Geometry 555
Example 30. In a ∆ ABC, the incircle touches the sides BC, CA and AB respectively at D, E and F. If the
radius of the incircle is 4 units and if BD, CE and AF are consecutive integers, find the sides of the ∆ ABC.
Solution Suppose that BD = x ,
B
CE = x + 1
and AF = x + 2. F
D
Then, CD = CE = x + 1,
AE = AF = x + 2,
BF = BD = x ,
Hence, a = BC = x + x + 1 = 2x + 1,
b = CA = x + 1 + x + 2 = 2x + 3,
A C
c = AB = x + 2 + x = 2x + 2; E
a+b+c
s= = 3x + 3;
2
s − a = x + 2, s − b = x , s − c = x + 1;
∆ 1
Since, r = = s (s − a )(s − b )(s − c ),
s s
(x + 2)x (x + 1)
we get 4=
3x + 3
x (x + 2) 2
i.e., 16 = ; x + 2x = 48
3
i.e., (x + 8)(x − 6) = 0; x = 6 or − 8
Since, x cannot be negative x = 6 and a = 13, b = 15, c = 14.
Example 31. Let ABC be an acute angled triangle in which D, E, F are points on BC, CA, AB respectively
such that AD ⊥ BC; AE = EC and CF bisects ∠C internally. Suppose CF meets AD and DE in M and N
respectively. If FM = 2, MN = 1, NC = 3, find the perimeter of the ∆ ABC.
Solution FN = 3 = NC. So, N is mid point of FC and E the A
mid point of AC. Hence ND | | AB and so D is the mid point of
BC. Thus AD is altitude as well as median, so that AB = AC. E
AD is also angle bisector of ∠A. ∆AFM and ∆DNM are similar
since
∠FAM = ∠MDN , F C
∠AMF = ∠DMN . M N
AM FM 2
So, = =
MD MN 1
Thus, M is the centroid of ∆ABC. Since, CF passes D
through M , CF is also a median, i.e., angle bisector CR is
also median. Hence, CA = CB; Consequently ∆ABC is
B
equilateral CF = 6 = altitude of an equilateral triangle.
12 12
Side of the equilateral triangle = . Perimeter = .
3 3
Aliter Having proved AB = AC, consider M which is the incentre and MD the inradius. D is the point
where the incircle touches BC. Now BM is the angle bisector of ∆FBC.
FM FB 2 1
= = =
MC BC 4 2
BC
So, FB = = BD ⋅ F is the point where the incircle touches AB. But angle bisector CF meets AB at F.
2
CF⊥ AB. So, CF is also median, altitude and angle bisector. Hence, the triangle is equilateral.
www.pdfworld.in
www.pdfworld.in
Aliter Consider ∆AFC and the fact that AM is the angle bisector of ∠A
AF FM 1
= = ,
AC MC 2
AC AB
AF = =
2 2
Since, AB = AC. So, F is mid point of AB. Hence, CF is the median, angle bisector and altitude, so
AB = CA = CB. Hence, triangle is equilateral.
Note Can also be solved using coordinate geometry with DA as origin BC as x-axis and AD as y-axis.
Example 32. Let A1A 2 . . . An be an n-sided regular polygon such that

1 1 1
= + .
A1A 2 A1A3 A1A 4
Determine n, the number of sides of the polygon.
A4
Solution Let O be the centre of the polygon and OA1 = r . Let θ be the angle
subtended by any side at the centre.
2π A3
Then, θ= .
n
θ 3θ
Also, A1A 2 = 2r sin , A1A3 = 2r sin θ, A1A 4 = 2r sin
2 2
1 1 1 A2
∴ = +
θ 2r sin θ 3θ
2r sin 2r sin
2 2 A1
1 1 1
⇒ = +
θ sin θ 3θ
sin sin
2 2
3θ
sin θ + sin
1 2
⇒ =
θ 3θ
sin sin θ ⋅ sin
2 2
3θ θ 3θ
⇒ 2 sin cos = sin θ + sin
2 2 2
3θ
⇒ sin 2θ + sin θ = sin θ + sin
2
3θ
⇒ sin 2θ − sin =0
2
7θ θ
⇒ 2 cos ⋅ sin = 0
4 4
7θ π 3 π
∴ = , ,...
4 2 2
2π 6π
⇒ θ= , ,...
7 7
2π 2π
Since, θ = , we must have θ = and n = 7.
0 7
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1. Let ABCD be a square and M , N points on sides ∆PAB and ∆ QBC on one side of the segment AC
AB , BC respectively, such that ∠MDN = 45°. If such that ∠APB = ∠BQC = 120° and an
R is the mid point of MN , show that RP = RQ isosceles ∆ RAC on the other side of AC such
where P , Q are the points of intersection of AC that ∠ ARC = 120° Show that PQR is an
with the lines MD, ND. equilateral triangle.
2. Prove that the inradius of a right-angled 10. In a ∆ ABC, D is a point on BC such that AD is
triangle with integer sides is an integer. the internal bisector of ∠A. Suppose ∠B = 2∠C
and CD = AB. Prove that ∠A = 72°.
3. The cyclic octagon A B C D E F G H has sides,
a , a , a , a , b , b , b , b respectively. Find the radius 11. Let BE and CF be the altitudes of an acute
of the circle that circumscribes ∆ABC, with E on AC and F on AB. Let O be the
A B C D E F G H. point of intersection of BE and CF. Take any
4. ABCD is quadrilateral and P , Q are mid points line KL through O with K on AB and L on AC.
of CD, AB , AP , DQ meet at X , and BP , CQ meet Suppose M and N are located on BE and CF
at Y . Prove that area of ∆ADX + area of respectively, such that KM is perpendicular to
∆BCY = area of quadrilateral PXQY . BE and LN is perpendicular to CF. Prove that
FM is parallel to EN .
5. ABCD is a cyclic quadrilateral; x , y , z are the
distances of A from the lines BD, BC , CD 12. The circumference of a circle is divided into
respectively. Prove that eight arcs by a convex quadrilateral ABCD,
BD BC CD with four arcs lying inside the quadrilateral
= + and the remaining four lying outside it. The
x y z
lengths of the arcs lying inside the
6. ABCD is a cyclic quadrilateral with AC ⊥ BD quadrilateral are denoted by p , q , r , s in
and AC meets BD at E. Prove that counter-clockwise direction starting from
EA 2 + EB 2 + EC 2 + ED2 = 4R 2 some arc. Suppose p + r = q + s. Prove that
ABCD is a cyclic quadrilateral.
where R is the radius of the circumscribed
circle. 13. In an acute ∆ ABC, points D, E , F are located on
the sides BC, CA , AB respectively such that
7. Let Γ and Γ′ be concentric circles. Let
CD CA AE AB BF BC
ABC , A ′ B ′ C ′ be any 2 equilateral triangles = , = , =
CE CB AF AC BD BA
inscribed in Γ and Γ′ respectively. If P and P ′
are any two points on Γ and Γ′ respectively, Prove that AD, BE , CF are the altitudes of ABC.
show that 14. Suppose P is an interior point of a ∆ ABC such
P ′ A 2 + P ′ B 2 + P ′ C2 = A ′ P 2 + B′ P 2 + C′ P 2 that the ratios
d (A , BC ) d (B , CA ) d (C , AB )
8. The internal bisector of ∠ A in a ∆ ABC with , ,
d (P , BC ) d (P , CA ) d (P , AB )
AC > AB, meets the circumcircle Γ of the
triangle in D. Join D to the centre O of the circle are all equal. Find the common value of these
Γ and suppose DO meets AC in E, possibly ratios. [Here d (X ,YZ ) denotes the
when extended. Given that BE is perpendicular perpendicular distance from a point X to the
to AD, show that AO is parallel to BD. line YZ.]
9. Let AC be a line segment in the plane and B a 15. Let ABC be a triangle in which AB = AC and
point between A and C. Construct isosceles ∠CAB = 90°. Suppose M and N are points on
www.pdfworld.in
www.pdfworld.in
the hypotenuse BC such that 23. Let ABC be an acute-angled triangle; AD be

BM 2 + CN 2 = MN 2. Prove that ∠MAN = 45°. bisector of ∠BAC with D on BC; and BE be the
altitude from B on AC. Show that ∠CED > 45°.
16. Let ABCD be a quadrilateral; X andY be the mid
points of AC and BD respectively and the lines 24. A trapezium ABCD, in which AB is parallel to
through X and Y respectively parallel to CD, is inscribed in a circle with centre O.
BD, AC meet in O. Let P , Q , R , S be the mid Suppose the diagonals AC and BD of the
points of AB , BC , CD, DA respectively. Prove trapezium intersect at M and OM = 2.
that (a) If ∠AMB is 60°, determine, with proof the
(a) quadrilaterals APOS and APXS have the difference between the lengths of the
same area; parallel sides.
(b) the areas of the quadrilateral (b) If ∠AMD is 60°, find the difference between
APOS , BQOP , CROQ , DSOR are all equal. the lengths of the parallel sides.
17. Consider in the plane a circle Γ with centre O 25. Let ABC be an acute-angled triangle; let D, F be
and a line l not intersecting circle Γ. Prove that the mid-points of BC , AB respectively. Let the
there is a unique point Q on the perpendicular perpendicular from F to AC and the
drawn from O to the line l, such that for any perpendicular at B to BC meet in N . Prove that
point P on the line l , PQ represents the length ND is equal to the circumradius of ABC.
of the tangent from P to the circle Γ.
26. Let ABC be a triangle in which AB = AC and let
18. In ∆ ABC, let D be th mid point of BC. If I be its incentre. Suppose BC = AB + AI . Find
∠ADB = 45° and ∠ACD = 30°, Determine ∠BAC.
∠BAD.
27. A convex polygon Γ is such that the distance
19. Let ABCD be a convex quadrilateral; P , Q , R , S between any two vertices of Γ does not exceed
be the mid points of AB , BC , CD, DA 1.
respectively such that ∆ AQR and ∆ CSP are (i) Prove that the distance between any two
equilateral. Prove that ABCD is a rhombus. points on the boundary of Γ does not
Determine its angles. exceed 1.
20. Let ABCD be a quadrilateral in which AB is (ii) If X and Y are two distinct points inside Γ,
parallel to CD and perpendicular to AD; prove that there exists a point Z on the
AB = 3CD ; and the area of the quadrilateral is boundary of Γ such that XZ + YZ ≤ 1.
4. If a circle can be drawn touching all the sides
28. Let ABCDEF be a convex hexagon in which the
of the quadrilateral, find its radius.
diagonals AD, BE , CF are concurrent at O.
21. A 6 × 6 square is dissected in to 9 rectangles Suppose the area of ∆ OAF is the geometric
by lines parallel to its sides such that all these mean of those of OAB and OEF, and the area of
rectangles have integer sides. Prove that there ∆ OBC is the geometric mean of those of OAB
are always two congruent rectangles. and OCD. Prove that the area of ∆ OED is the
22. Let ABC be an acute-angled triangle and let geometric mean of those of OCD and OEF.
D, E , F be the feet of perpendiculars from 29. Let ABC be a triangle in which ∠A = 60° Let BE
A,B,C respectively to BC,CA,AB. Let the and CF be the bisectors of the angles ∠B and
perpendiculars from F to CB , CA , AD, BE meet ∠C with E on AC and F on AB. Let M be the
them in P , Q , M , N respectively. Prove that reflection of A in the line EF. Prove that M lies
P , Q , M , N are collinear. on BC.
www.pdfworld.in
www.pdfworld.in
Geometry 559
Level 2
1. A circle passes through the vertex C of a A1 , B1 , C1 denote the reflections of P in the
rectangle ABCD and touches its sides sides BC , CA , AB respectively. Prove the
AB and AD at M and N respectively. If the following statements;
distance from C to the line segment MN is (a) If P is the incentre or an excentre of ABC,
equal to 5 units find the area of the rectangle
then P is the circumcentre of A1B1C1 ;
ABCD.
(b) If P is the circumcentre of ABC, then P is
2. In an acute-angled ∆ ABC, ∠A = 30° , H is the the orthocentre of A1B1C1;
orthocentre, and M is the mid point of BC. On
(c) If P is the orthocentre of ABC, then P is
the line HM , take a point T such that
either the incentre or an excentre of
HM = MT . Show that AT = 2BC. A1B1C1.
3. The inscribed circumference in the ∆ ABC is 9. Let ABC be a triangle and D be the mid point of
tangent to BC , CA and AB at D, E and F , side BC. Suppose ∠DAB = ∠BCA and
respectively. Suppose that this circumference ∠DAC = 15°. Show that ∠ADC is obtuse.
meets AD again at its mid-point X ; that is, Further, if O is the circumcentre of ADC, prove
AX = XD. The lines XB and XC meet the that ∆ AOD is equilateral.
inscribed circumference again at Y and Z ,
10. For a convex hexagon ABCDEF in which each
respectively.
pair of opposite sides is unequal, consider the
Show that EY = FZ .
following six statements :
4. Two externally tangent circles of radii (a1 ) AB is parallel to DE ;
R1 and R2 are internally tangent to a semicircle
(a 2 ) AE = BD;
of radius 1, as in the figure.
(b1 ) BC is parallel to EF ;
(b2 ) BF = CE ;
(c1 ) CD is parallel to FA;
(c 2 ) CA = DF .
(a) Show that, if all the six statements are true;
Prove that R1 + R2 ≤ 2( 2 − 1) then the hexagon is cyclic (i.e., it can be
inscribed in a circle).
with equality holds if and only if R1 = R2.
(b) Prove that, in fact, any five of these six
5. T1 is an isosceles triangle with circumcircle K. statements also imply that the hexagon is
Let T2 be another isosceles triangle inscribed cyclic.
in K whose base is one of the equal sides ofT1
and which overlaps the interior ofT1. Similarly 11. Let ABC be a triangle with sides a , b , c .
b
create isosceles trianglesT3 fromT2 , T4 fromT3 Consider a ∆ A1B1C1 with sides equal to a + ,
and so on. Do the triangles Tn approach an 2
c a
equilateral triangle as n → ∞ ? b + ,c + .
2 2
6. The incircle of ∆ABC touches the sides 9
BC , CA and AB in K , L and M respectively. The Show that [A1B1C1 ] ≥ [ABC ] ,
4
line through A and parallel to LK meets MK in
P and the line through A and parallel to MK where [XYZ ] denotes the area of the ∆ XYZ.
meets LK in Q. Show that the line PQ bisects 12. Consider an acute ∆ ABC and let P be an
the sides AB and AC of ∆ ABC. interior point of ABC. Suppose the lines
7. In a convex quadrilateral PQRS , PQ = RS , BP and CP , when produced, meet AC and AB in
( 3 + 1)QR = SP and ∠RSP − ∠SPQ = 30°. Prove E and F respectively. Let D be the point where
that ∠PQR − ∠QRS = 90° AP intersects the line segment EF and K be the
foot of perpendicular from D on to BC. Show
8. Let ABC be a triangle in which no angle is 90°.
that DK bisects ∠EKF.
For any point P in the plane of the triangle, let
www.pdfworld.in
www.pdfworld.in
13. Let R denote the circumradius of a ∆ ABC; 19. In a ∆ ABC right angled at C, the median
a , b , c its sides BC , CA , AB; and ra , rb , rc its through B bisects the angle between BA and
exradii opposite A , B , C. If 2R ≤ ra , prove that the bisector of ∠B. Prove that
(i) a > b and a > c ; 5 AB
< < 3.
(ii) 2R > rb and 2R > rc . 2 BC
14. Consider a convex quadrilateral ABCD, in 20. Let ABC be a triangle, I its incentre;
which K , L , M , N are the mid points of the A1 , B1 , C1 be the reflections of I in BC , CA , AB
sides AB , BC , CD, DA respectively. Suppose respectively. Suppose the circumcircle of
(a) BD bisects KM at Q; ∆A1B1C1 passes through A , Prove that
B1 , C1 , I , I1 are concyclic, where I1 is the
(b) QA = QB = QC = QD ; and
incentre of ∆ A1B1C1.
(c) LK/LM = CD/CB.
21. Let ABC be a triangle; ΓA , ΓB , ΓC be the three
Prove that ABCD is a square.
equal, disjoint circles inside ABC such that ΓA
15. Let M be the mid point of side BC of a ∆ ABC. touches AB and AC ; ΓB touches AB and BC ;
Let the median AM intersect the incircle of and ΓC touches BC and CA . Let Γ be a circle
ABC at K and L , K being nearer to A than L. If touching circles ΓA , ΓB , ΓC externally. Prove
AK = KL = LM , prove that the sides of ∆ ABC that the line joining the circumcentre O and
are in the ratio 5 : 10 : 13 in some order. the incentre I of ∆ABC passes through the
16. In a non-equilateral ∆ ABC, the sides a , b , c centre of Γ.
form an arithmetic progression. Let I and O 22. Let ABC be a triangle and let P be interior point
denote the incentre and circumcentre of the such that ∠BPC = 90° , ∠BAP = ∠BCP. Let M , N
triangle respectively. be the mid points of AC , BC respectively.
(i) Prove that IO is perpendicular to BI . Suppose, BP = 2PM . Prove that A , P , N are
(ii) Suppose BI extended meets AC in collinear.
K and D, E are the mid points of BC , BA 23. Let ABC be an acute angles triangle and let H
respectively. Prove that I is the be its orthocentre. Let h max denote the largest
circumcentre of ∆ DKE. altitude of the ∆ABC. Prove that
17. In a cyclic quadrilateral ABCD, AH + BH + CH ≤ 2h max .
AB = a , BC = b , CD = c , ∠ABC = 120°, and
24. Let ABC be an acute-angled triangle with
∠ABD = 30°. Prove that
altitude AK. Let H be its orthocentre and O be
(i) c ≥ a + b ; its circumcentre. Suppose KOH is an
(ii) | c + a − c + b | = c − a − b . acute-angled triangle and P its circumcentre.
Let Q be the reflection of P in the line HO. Show
18. Let ABC be a triangle in which AB = AC. Let D
that Q lies on the line joining the mid points of
be the mid point of BC and P be a point on AD.
AB and AC.
Suppose E is the foot of perpendicular from P
AP BP BD 25. Let ABC be a triangle with circumcircle Γ. Let
on AC. If = = λ, =m and
PD PE AD M be a point in the interior of ∆ ABC which is
z = m 2 (1 + λ ), prove that also on the bisector of ∠A. Let AM , BM , CM
z 2 − (λ3 − λ2 − 2)z + 1 = 0. meet Γ in A1 , B1 , C1 respectively. Suppose P is
the point of intersection of A1C1 with AB; and
Hence, show that λ ≥ 2 and λ = 2, if and only if Q is the point of intersection of A1B1 with AC.
and only if ABC is equilateral. Prove that PQ is parallel to BC.
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1. As construction join PN and QM . In the figure A
AC is a diagonal of the quadrilateral ABCD
with ∠ACB = 45° and ∠MDN = 45° (by
hypothesis). So, ∠PDN = ∠PCN each angle
being equal to 45°. Now, as ∠PDN and ∠PCN
M
are equal angles subtended by the same line
segment PN at points D and C, the O
quadrilateral PNCD is cyclic. So N
∠DCN + ∠DPN = 180°. But ∠NCD = 90°
which implies that ∠NPD = 90°. So
B C
∠NPM = 90°, as ∠NPD and ∠NPM form a linear L
pair. Now, as ∠NPM = 90°, MN is the diameter
of the circumcircle of the right ∆NPM with R as AM = AN = AB − NB = c − r and
its centre as R is the mid point of MN by CM = CL = BC − CL = a − r .
hypothesis. So, b = AC = AM + CM = c − r + a − r
D C
b − (c + a )
= c + a − 2r ⇒ r =
45° 2
Q
As ∠B = 90° ⇒ b 2 = c 2 + a 2, we have
N (i) if c and a are both odd or both even, c 2 + a 2

P is even ⇒b 2 is given ⇒b is even ⇒b − (c + a ) is
even.
R
(ii) if one of c and a is even and the other odd,
A B c 2 + a 2 is odd ⇒ b 2 is odd ⇒ b is odd ⇒
M
b − (c + a ) = an even number.
Similarly by showing that quadrilateral QMAD So, in any case, if a , b , c are integers, we have
is a cyclic one and arguing as before, we can b − (c + a )
r = = an integer.
conclude that MN is the diameter of the 2
circumcircle of the right ∆NQM with
∠NQM = 90° and R as its centre. 3. Let r be the radius of the circumcircle and θ, ϕ
angle subtended by the sides of lengths a and
The two preceding paragraphs imply that
MNQM is a cyclic quadrilateral with R as its b respectively at the centre of the circle. Then,
π
centre. Hence RP = RQ . 4θ + 4ϕ = 2 π and hence ϕ = − θ. Using cosine
2
2. Let ABC be a right triangle with ∠B = 90°. rule, we have
Let O be its incentre and L , M , N the points of
contact of the incircle with the a , b , c a
respectively. b
a
Suppose that the inradius is r. Now as
∠ABC = 90°, the quadrilateral NBLO is a
b
square. So NB = BL = r . Also, as the two
tangents drawn from an external point to a
circle are of equal length, we have
www.pdfworld.in
www.pdfworld.in
a 2 = 2r 2 − 2r 2 cos θ = 2r 2 (1 − cos θ) z
D
A
b 2 = 2r 2 − 2r 2 sin θ = 2r 2 (1 − sin θ) x
2
 a  = 1 − cos θ = 2t θ
2
  , where t = tan
b  1 − sin θ (1 − t )2 2
y
a 2t a t a
∴ = ⇒ = ⇒ t =
b 1−t 2b 1 − t a + 2b C
2 2
2b a
∴ 1 − cos θ = = B
1+t 2
a + b 2 + ab 2
2
a 2 + b 2 + ab 2
∴ r = ∴
BC CD BD
+ − = cot ∠ABC + cot ∠ADC = 0
2 y z x
4. Let d1, d 2 , d3 be the perpendicular distances of Since, ∠ABC + ∠ADC = 180°
the points D, P , C respectively from AB. Then,
d + d3 6. Let O be the centre of the circle.
d2 = 1 , since P is the mid point of CD.
2 ∠AOB + ∠COD = 2(∠ACD + ∠CBD)
P
D C = 2 × 90° = 180°
C
x y
A B O
Q E
D B
Area of quad. PXQY
= area of ∆DQC − area ∆DXP − area of ∆PYC
1 A
= ⋅ CD ⋅ d 2 − area of ∆DXP − area of ∆PYC
2
So ∠AOB = θ, Then
1
= ⋅ CD ⋅ d 2 − (area of ∆DAP − area of ∆DAX ) AB 2 + CD2 = 2(R 2 − R 2 cos θ)
2
− (area of ∆PBC − area of ∆BYC) + 2(R 2 − R 2 cos(π − θ)) = 4R 2
1 1 1 Similarly,
= CD ⋅ d 2 − ⋅ PD ⋅ d1 − ⋅ PC ⋅ d3
2 2 2 BC 2 + AD2 = 4R 2EA 2 + EB 2 + EC 2 + ED2
+ area of ∆DAX + area of ∆BYC 1
= ∠(AB 2 + BC 2 + CD2 + DA 2 ) = 4R 2
1 1 1 1 1 2
= CD ⋅ d 2 − ⋅ CD ⋅ d1 − ⋅ CD ⋅ d3
2 2 2 2 2
7. Let O be the centre and let r be the radius of
+ area of ∆ADX + area of ∆BCY
the inner circle and R the radius of the outer
1  d + d3  circle. Let ∠POA ′ = θ. Then
= CD ⋅  d 2 − 1  + area of ∆ADX + area
2  2 
(PA ′ )2 = (OP )2 + (OA ′ )2 − 2 ⋅ OA ′ ⋅ OP cos θ
of ∆BCY = area of ∆ADX + area of ∆ BCY
= r 2 + R 2 − 2r R cos θ
BD CB
5. = cot ∠ABD + cot ∠ADB (PB ′ )2 = r 2 + R 2 − 2r R cos(θ − 120° )
x y
CD (PC ′ )2 = r 2 + R 2 − 2r R cos(θ + 120° )
= cot ∠ABC + cot ∠ACB
z (PA ′ )2 + (PB ′ )2 + (PC ′ )2
= cot ∠ACD + cot ∠ADC ⋅ ∠ACB = ∠ADB = 3r 2 + 3R 2 − 2r R(cos θ + cos(θ − 120° )
and ∠ACD = ∠ABD
+ cos(θ + 120° ))
www.pdfworld.in
www.pdfworld.in
Geometry 563
= 3r 2 + 3R 2 − 2rR(cos θ + 2 ⋅ cos θ cos 120° ) ∠BDE. But ∠BDA = ∠BCA = ∠C. Thus
1 ∠ODA = ∠C. Since OD = OA , we get
= 3r 2 + 3R 2 − 2rR(cos θ − 2 ⋅ cos θ) ∠OAD = ∠C. It follows that
2
∠BDA = ∠C = ∠OAD. This implies that OA is
= 3r 2 + 3R 2 parallel to BD.
Similarly, P ′ A 2 + P ′ B 2 + P ′ C 2 = 3r 2 + 3R 2. 9. We give here two different cases.
Thus, they are equal.
Case I Drop perpendiculars from P and Q to
AC and extend them to meet AR , RC in K , L
A respectively. Join KB , PB , QB , LB , KL (Fig.1)
P' C' C
P
O
A'
θ Q
P
B'
B
B A C
Aliter O is the centroid of both the ∆ ABC and L

∆ A ′ B ′ C ′. In a ∆ XYZ , if G is the centroid an P is
any point, then K
PX 2
+ PY 2
+ PZ = 3PG + QX
2 2 2
+ GY 2
+ GZ 2
R
Fig. 1
In ∆ABC using P ′ as point and O as the K
centroid.
PA 2 + P ′ B 2 + P ′ C 2 P
= 3P ′ O + OA + OB + OC = 3R + 3r
2 2 2 2 2 2
Q
Similarly, the other expression too is equal to
3R 2 + 3r 2. Hence the result. A C
B
8. We consider here the case when ABC is an
acute-angled triangle; the cases when ∠A is
obtuse or one of the ∠B and ∠C is obtuse may
be handled similarly.
R
A
Fig. 2
E
Observe that K , B , Q are collinear and so are
N O P , B , L. (This is because
B C ∠QBC = ∠PBA = ∠KBA and similarly
M ∠PBA = ∠CBL.) by symmetry we see that
∠KPQ = ∠PKL and ∠KPB = ∠PKB. If follows
that ∠LPQ = ∠LKQ and hence K , L , Q , P are
D concyclic. We also note that
∠ KPL + ∠ KRL = 60° + 120° = 180° . This
Let M be the point of intersection of DE and implies that P, K, R, L are concyclic. We
BC; let AD intersect BE in N . Since ME is the conclude that P , K , R , L , Q are concyclic. This
perpendicular bisector of BC, we have BE = CE .
gives
Since AN is the internal bisector of ∠A and is
perpendicular to BE, it must bisect BE; i.e., ∠PRQ = ∠PKQ = 60°,
BN = NE . This in turn implies that DN bisects ∠RPQ = ∠RKQ = ∠RAP = 60°
www.pdfworld.in
www.pdfworld.in
Case II Produce AP and CQ to meet at K. ∠ A = 2 ∠C as well and we get ∠C = 36°. In

Observe that AKCR is a rhombus and BQKP is turn ∠A = 72° .
a parallelogram. (See Fig.2.) Put AP = x ,CQ = y . 11. Observe that KMOF and ONLE are cyclic
Then PK = BQ = y , KQ = PB = x and quadrilaterals. Hence, ∠FMO = ∠FKO and
AR = RC = CK = KA = x + y . Using cosine rule ∠OEN = ∠OLN .
in ∆PKQ, we get A
PQ 2 = x 2 + y 2 − 2xy cos 120° = x 2 + y + xy .
Similarly cosine rule in ∆QCR gives
QR 2 = y 2 + (x + y )2 − 2xy cos 60°
= x 2 + y 2 + xy and cosine rule in ∆PAR gives E
F
RP 2 = x 2 + (x + y )2 − 2xy cos 60° O L
= x 2 + y 2 + xy . If follows that PQ = QR = RP . K
10. Draw the angle bisector BE of ∠ABC to meet M N

AC in E. Join ED. Since, ∠B = 2∠C, it follows
that ∠EBC = ∠ECB. We obtain EB = EC. B C
A However we see that

π π
∠OLN = − ∠NOL = − ∠KOF = ∠OKF .
β β 2 2
If follows that ∠FMO = ∠OEN .
E
This forces that FM is parallel to EN .
2β
12. Let the lengths of the arcs XY , UV , EF , GH be
β respectively p , q , r , s. We also the following
α notations : (See figure)
α α
B C D G
D F
C
Consider the ∆ BEA and ∆ CED. We observe
that BA = CD, BE = CE and ∠EBA = ∠ECD. E
Hence, BEA = CED giving EA = ED. If ∠DAC = β, H
then we obtain ∠ADE = β, Let I be the point of
intersection of AD and BE. Now consider the P
triangles AIB and DIE. They are similar since V
∠BAI = β = ∠IDE and ∠AIB = ∠DIE. It follows
Y B
that ∠DEI = ∠ABI = ∠DBI . Thus BDE is U
isosceles and DB = DE = EA . We also observe A X
that ∠CED = ∠EAD + ∠EDA = 2β = ∠A . This
implies that ED is parallel to AB. Since BD = AE , ∠XAY = α1, ∠AYP = α 2, ∠YPX = α3,
we conclude that BC = AC. In particular ∠PXA = α 4 , ∠UBY = β1, ∠BVP = β 2, ∠VPU = β3,
∠A = 2∠C. Thus the total angle of ABC is 5∠C ∠PUB = β 4 , ∠ECF = γ 1, ∠CFP = γ 2, ∠FPE = γ 3,
giving ∠C = 36°. We obtain ∠A = 72°. ∠PEC = γ 4 , ∠GDH = δ1, ∠DHP = δ 2, ∠HPG = δ3,
Aliter We make use of the charactarisation, ∠PGD = δ 4 .
in a ∆ABC , ∠ B = 2 ∠ C if and only if We observe that
b 2 = c (c + a ). Note that CD = c and BD = a − c . Σ α j = Σ β j = Σ γ j = Σ δ j = 2π
Since AD is the angle bisector, we also have It follows that
a −c c Σ (α j + γ j ) = Σ (β j + δ j )
= .
c b On the other hand, we also have α 2 = β 4 since
This gives c 2 = ab − bc and hence PY = PU . Similarly we have other relations;
β 2 = γ 4 , γ 2 = δ 4 and δ 2 = α 4 . It follows
b = ca + ab − bc .
2
It follows that
that
b (b + c ) = a (b + c ) so that a = b. Hence,
α1 + α3 + γ 1 + γ 3 = β1 + β3 + δ1 + δ3
www.pdfworld.in
www.pdfworld.in
Geometry 565
But p + r = q + s implies that α3 + γ 3 = β3 + δ3. However, if L is the foot of perpendicular from

We thus obtain A on to BC, then using Pythagoras theorem in
∆ ALB and ∆ ALC, we get
α1 + γ 1 = β1 + δ1
b 2 − LC 2 = c 2 − (a − LC )2
Since, α1 + γ 1 + β1 + δ1 = 360°, it follows that
Which reduces to LC = (a 2 − c 2 + b 2 )/ 2a . We
ABCD is a cyclic quadrilateral.
conclude that LC = DC proving L = D. Or we
13. Put CD = x . Then, with usual notations, we get can also infer that x = b cos C from cosine rule
CD ⋅ CB ax in ∆ ABC. This implies that CD = CL, since
CE = =
CA b CL = b cos C from right ∆ ALC. Thus, AD is
Since, AE = AC − CE = b − CE , we obtain altitude on to BC . Similar proof works for the
remaining altitudes.
b 2 − ax
AE = ,
b Aliter We see that CD ⋅ CB = CE ⋅ CA , so that
AE ⋅ AC b − ax 2 ABDE is a cyclic quadrilateral. Similarly we
AF = = infer that BCEF and CAFD are also cyclic
AB c
quadrilaterals. (See Fig. 2.) Thus
A
∠AEF = ∠B = ∠CED. Moreover
α ∠BED = ∠DAF = ∠DCF = ∠BCF = ∠BEF. It
follows that ∠BEA = ∠BEC and hence each is a
β E
right angle thus proving that BE is an altitude.
α β Similarly we prove that CF and AD are
α
F altitudes. (Note that the concurrence of the
lines AD, BE, CF are not required.)
14. Let AP , BP , CP when extended, meet the sides
α BC , CA , AB in D, E , F respectively. Draw
B C
D AK , PL perpendicular to BC with K , L on BC.
Fig. 1 (See Fig. 3)
A A
E
E
F
P
F
B K L D C
B D C
Fig. 3
Fig. 2 X
A (h, k)
This in turn gives
c 2 − b 2 + ax
BF = AB − AF =
c
Finally, we obtain
c 2 − b 2 + ax
BD = P (u, v)
a
Using BD = a − x , we get
X'
a2 − c 2 + b2 B (0, 0) C (a, 0)
x =
2a Fig. 4
www.pdfworld.in
www.pdfworld.in
d (A , BC ) AK AD d (C , AB ) ka (ku − hv ) ka
Now, = = = / = .
d (P , BC ) PL PD d (P , AB ) h2 + k2 h 2 + k 2 ku − hv
d (B , CA ) BE
Similarly, = From the equality of these ratios, we get
d (P , CA ) PE
k − ka ka
d (C , AB ) CF = = .
and = v ku − (h − a ) v − ka ku − hv
d (P , AB ) PF
The equality of the first and third ratios gives
So, we obtain ku − (h + a )v = 0. Similarly the equalilty of
AD BE CF AP BP CP second and third ratios gives
= = and hence = = .
PD PE PF PD PE PF 2ku − (2h − a )v = ka. Solving for u and v, we get
AP BP h+a k
From = and ∠APB = ∠DPE, u= ,v = .
PD PE 3 3
it follows that triangles APB and DPE are Thus, P is the centroid of the triangle and each
k
similar. So, ∠ABP = ∠DEP and hence AB is of the ratios is equal to = 3
parallel to DE. v
Similarly, BC is parallel to EF and AC is parallel 15. Draw CP perpendicular to CB and BQ
to DF. Using these we obtain perpendicular to CB such that CP = BM ,
BD AE AF DC BQ = CN . Join PA, PM , PN , QA, QM , QN .
= = =
DC EC FB BD C
2α
whence BD = CD 2 2
or which is same as
N
BD = CD. Thus D is the mid point of BC. P
Similarly E , F are the mid-points of CA and AB
α
respectively. α
We infer that AD, BE , CF are indeed the M
medians of the ∆ ABC and hence P is the 2β
centroid of the triangle. So α
β B
AD BE CF A
= = =3
PD PE PF
α
and consequently each of the given ratios is Q
also equal to 3.
In ∆ CPA and ∆ BMA, we have
Aliter Let ABC, the given triangle be placed
∠PCA = 45° = ∠MBA ; PC = MB, CA = BA . So,
in the xy-plane so that B = (0, 0), C = (a , 0) (on
∆CPA ≡ ∆BMA . Hence ∠PAC = ∠BAM = α, say.
the x -axis). (See Fig. 4.)
Consequently, ∠MAP = ∠BAC = 90°, whence
Let A = (h , k ) and P = (u , v ). Clearly, d (A , BC ) = k PAMC is a cyclic quadrilateral. Therefore
and d (P , BC ) = v , so that ∠PMC = ∠PAC = α, Again
d (A , BC ) k
= PN 2 = PC 2 + CN 2 = BM 2 + CN 2 = MN 2. So,
d (P , BC ) u PN = MN , giving ∠NPM = ∠NMP = α, in ∆PMN .
The equation to CA is kx − (h − a )y − ka = 0. Hence ∠PNC = 2α. Likewise ∠QMB = 2β, where
β = ∠CAN . Also ∆NCP ≡ ∆QBM , as CP = BM ,
So,
NC = BQ and ∠NCP = 90° = ∠QBM . Therefore,
d (B , CA ) − ka (ku − (h − a )v − ka ) ∠CPN = ∠BMQ = 2β, wherce 2α + 2β = 90°;
= /
d (P , CA ) k + (h − a )
2 2
k 2 + (h − a )2 α + β = 45°; finally ∠MAN = 90° − (α + β ) = 45°.
− ka Aliter Let AB = AC = a, so that BC = 2a ; and
= ∠MAB = α, ∠CAN = β.
ku − (h − a )v − ka
By the Sine Law, we have from ∆ABM that
Again the equation to AB is kx − hy = 0.
BM AB
Therefore, =
sin α sin(α + 45° )
www.pdfworld.in
www.pdfworld.in
Geometry 567
a 2 sin α a 2u (b) Now, [APXS ] = [APX ] + [ASX ]

So BM = = ,
cos α + sin α 1 + u 1 1
= [ABX ] + [ADX ]
where u = tan α. 2 2
1 1
= [ABC ] + [ADC ]
C 4 4
π/4 1
N = [ABCD]
4
a 1
Hence by (a), [APOS ] = [ABCD] . Similarly, by
M 4
β
symmetry each of the areas [AQOP ], [CROQ ]
α π/4 1
A a B and [DSOR ] is equal to [ABCD] . Thus, the
4
four given areas are equal. This proves part.
a 2v
Similarly, CN = , where v = tan β. (b). [Note [⋅] denotes area].
1+v
17. Let R be the foot
But BM 2 + CN 2 = MN 2 = (BC − MB − NC )2 Y
of the
= BC 2 + BM 2 + CN 2 perpendicular O
− 2BC ⋅ MB − 2BC ⋅ NC + MB ⋅ NC from O to the line l Q

and u be the X u z y
So, BC 2 − 2BC ⋅ MB − 2BC ⋅ NC + 2MB ⋅ NC = 0 x
length of the
u
This reduces to tangent RX from
a 2u a 2v R to circle Γ. On
2a 2 − 2 2a − 2 2a OR take a point Q
I R P
1+u 1+v
such that QR = u.
4a 2uv We show that Q is the desired point. To this
+ =0
(1 + u ) (1 + v ) end, take any point P on line l and let y be the
length of the tangent PY from P to Γ.
Multiplying by (1 + u ) (1 + v )/ 2a 2, we obtain
Further let r be the radius of the circle Γ and
(1 + u ) (1 + v ) − 2u (1 + v ) − 2v (1 + u ) + 2uv = 0.
let y be the length of the tangent PY from P to
Simplification gives 1 − u − v − uv = 0. So
Γ. Join OP , QP . Let QP = x , OP = z, RP = t . From
u+v
tan (α + β ) = =1 right angled triangles POY , OXP, ORP, PQR we
1 − uv have respectively z 2 = r 2 + y 2, OR 2 = r 2 + u 2,
This gives α + β = 45°, whence ∠MAN = 45°, z 2 = OR 2 + t 2 = r 2 + u 2 + t 2, x 2 = u 2 + t 2. So
as well. we obtain y 2 = z 2 − r 2 = r 2 + u 2 + t 2 − r 2
16. We use the facts : (i) C = u + t 2 = x 2. Hence y = x . This gives PY = PX
2
the line joining the which is what we needed to show.

R
mid points of the
D X 18. Draw BL perpendicular to AC and join L to D.
sides of a triangle
Q
is parallel to the O Since ∠BCL = 30°, we get ∠CBL = 60°. Since
S
third side; (ii) and Y BLC is a right triangle with ∠BCL = 30°, we
any median of a have BL = BC/2 = BD. Thus in ∆ BLD, we
triangle bisects its A P B observe that BL = BD and ∠DBL = 60°. This
area; (iii) two implies that BLD is an equilateral triangle and
triangles having equal bases and bounded by hence LB = LD. Using ∠LDB = 60° and
same parallel lines have equal area. ∠ADB = 45°, we get ∠ADL = 15°. But
(a) Now BD is parallel to PS as well as OX is ∠DAL = 15°. Thus LD = LA . We hence have
parallel to PS. Hence [PXS ] = [POS ]. Adding LD = LA = LB. This implies that L is the
[PAS ] to both sides we get [APXS ] = [APOS ]. circumcentre of the ∆ . Thus,
This proves part (a).
www.pdfworld.in
www.pdfworld.in
A 1
4= AD (AB + CD)
2
15° 1
L = (2r ) (4(r + x )).
2
Thus we obtain r (r + x ) = 1. Using Pythagoras
theorem, we obtain BC 2 = BK 2 + CK 2.
60°45° 30° However BC = y + x , BK = y − x and CK = 2r .
B D C Substituting these and simplifying, we get
1 1 xy = r 2. But r + y = 3(r + x ) gives y = 2r + 3x .
∠BAD = ∠BLD = × 60° = 30°
2 2 Thus r 2 = x (2r + 3x ) and this simplifies to
(r − 3x ) (r + x ) = 0. We conclude that r = 3x .
19. We have QR = BD/2 = PS . Since AQR and CSP
Now the relation r (r + x ) = 1 implies that
are both equilateral and QR = PS , they must be
4r 2 = 3, giving r = 3/ 2.
congruent triangles. This implies that
AQ = QR = RA = CS = SP = PC. 21. Consider the dissection of the given 6 × 6
D R C square in to non-congruent rectangles with
E least possible areas. The only rectangle with
VF area 1 is an 1 × 1 rectangle. Similarly, we get
L W
S M Q 1 × 2, 1 × 3 rectangles for areas 2, 3, units. In
U O the case of 4 units we may have either a 1 × 4
N rectangle or a 2 × 2 square. Similarly, there can
A P B be a 1 × 5 rectangle for area 5 units and 1 × 6 or
2 × 3 rectangle for 6 units. Any rectangle with
Also ∠CEF = 60° = ∠RQA . area 7 units must be 1 × 7 rectangle, which is
Hence, CS is parallel to QA. Now, CS = QA not possible since the largest side could be 6
implies that CSQA is a parallelogram. In units. And any rectangle with area 8 units
particular SA is parallel to CQ and SA = CQ . must be a 2 × 4 rectangle. If there is any
This shows that AD is parallel to BC and dissection of the given 6 × 6 square in to 9
AD = BC. Hence, ABCD is a parallelogram. non-congruent rectangles with areas
Let the diagonal AC and BD bisect each other a1 ≤ a 2 ≤ a3 ≤ a 4 ≤ a5 ≤ a 6 ≤ a7 ≤ a 8 ≤ a 9, then we
at W. Then, DW = DB /2 = QR = CS = AR. Thus, observe that a1 ≥ 1, a 2 ≥ 2, a3 ≥ 3, a 4 ≥ 4, a5 ≥ 4,
in ∆ ADC, the medians AR , DW , CS are all a 6 ≥ 5, a7 ≥ 6, a 8 ≥ 6, a 9 ≥ 8, and hence the total
equal. Thus, ADC is equilateral. This implies area of all the rectangles is a1 + a 2 + . . . + a 9
ABCD is a rhombus. Moreover the angles are ≥ 1 + 2 + 3 + 4 + 4 + 5 + 6 + 6 + 8 = 39 > 36,
60° and 120°.
Which is the area of the given square. Hence if
20. Let P, Q, R, S be the D r R x C a 6 × 6 square is dissected in to 9 rectangles as
points of contact x stipulated in the problem, there must be two
of incircle with the r Q congruent rectangles.
sides AB, BC, CD, y
S 22. Observe that C , Q , F , P are concyclic.
DA respectively.
Since AD is r A Q
perpendicular to _ E
AB and AB is A r P K y x B F
M
parallel to DC, we N
see that AP = AS = SD = DR = r , the radius of
the inscribed circle. Let BP = BQ = y and
CQ = CR = x . Using AB = 3CD, we get
r + y = 3(r + x ). B P D C
Since the area of ABCD is 4, we also get
Hence, ∠CQP = ∠CEP = 90° − ∠FCP = ∠B
www.pdfworld.in
www.pdfworld.in
Geometry 569
Similarly the concyclicity of F , M , Q , A given b sin C sin B

so that tan θ = =
∠AQN = 90° + ∠FQM = 90° + ∠FAM c cos C cos C
= 90° + 90° − ∠B = 180° − ∠B sin B

=
sin (π/2 − C )
Thus we obtain ∠CQP + ∠AQN = 180°. It
follows that Q , N , P lie on the same line. Since, ABC is acute-angled, we have A < π/2.
Hence, B + C > π/2 or B > (π/2) − C. Therefore
We can similarly prove that
sin B > sin (π/2 − C ) . This implies that tan θ > 1
∠CPQ + ∠BPM = 180°. This implies that
and hence θ > π/4.
P , M , Q are collinear. Thus M , N both lie on
the line joining P and Q. 24. Suppose ∠AMB = 60°. Then, AMB and CMD are
equilateral triangle. Draw OK perpendicular to
23. Draw DL A
BD. (see Fig. 1). Note that OM bisects ∠AMB so
perpendicular to that ∠OMK = 30°. Hence, OK = OM/2 = 1. It
AB; DK E follows that KM = OM 2 − OK 2 = 3. We also
perpendicular to
M observe that
AC; and DM K
L AB − CD = BM − MD
perpendicular to
= BK + KM − (DK − KM ) + 2KM ,
BE. Then EM = DK .
SinceAD bisects B
D C since K is the mid-point of BD. Hence
AB − CD = 2 3.
∠A, we observe
D C D C
that ∠BAD = ∠KAD. Thus in ∆ ALD and ∆ AKD, M
we see that ∠ LAD = ∠ KAD ; ∠ AKD = 90° = M
∠ALD and AD is common. Hence ∆ ALD and ∆ A K P B
K
AKD are congruent, giving DL = DK . But O Q
O
DL > DM , since BE lies inside the triangle (by
acuteness property). Thus EM > DM . This A B
implies that ∠EDM > ∠DEM = 90° − ∠EDM .
Fig. 1 Fig. 2
We conclude that ∠EDM > 45°. Since
∠CED = ∠EDM , the result follows.
Suppose ∠AMD = 60° so that ∠AMB = 120°.
Aliter Draw PQ through O parallel to AC (with Q on
Let ∠CED = θ. We have CD = ab/(b + c ) and AB and P on BD). (see Fig. 2). Again OM bisects
CE = a cos C. Using sine rule in ∆CED, we have ∠AMB so that ∠OPM = ∠OMP = 60°. Thus OMP
CD CE is an equilateral triangle. Hence diameter
= .
sin θ sin (C + θ) perpendicular to BD also bisects MP. This gives
DM = PB . In the triangles DMC and BPQ, we
A
have BP = DM , ∠DMC = 120° = ∠BPQ and
∠DCM = ∠PBQ (property of cyclic
E quadrilateral). Hence DMC and BPQ are
c congruent so that DC = BQ . Thus
θ b AB − DC = AQ . Note that AQ = KP since KAQP
is a parallelogram. But KP is twice the altitude
of ∆ OPM . Since OM = 2, the altitude of OPM is
B D ab/(b + c) C 2 × 3/2 = 3. This gives AQ = 2 3.
a
Aliter
This reduces to Using some trigonometry, we can get
(b + c ) sin θ cos C = b sin C cos θ + b cos C sin θ. solutions for both the parts simultaneously.
Simplification gives Let K , L be the mid points of AB and CD
respectively. Then L , M , O , K are collinear (see
c sin θ cos C = b sin C cos θ
Fig. 3 and Fig. 4). Let ∠AMK = θ(= ∠DML ), and
www.pdfworld.in
www.pdfworld.in
OM = d . Since AMB and CMD are similar ∠FNB + ∠BOF = ∠KNB + ∠BCK = 180°
triangles, if MD = MC = x , then MA = MB = kx This implies that B , O , F , N are concyclic.
for some positive constant k.
Hence, ∠BFO = ∠BNO. But observe that
Now MK = kx cos θ, ML = x cos θ, so that ∠BFO = 90° since OF is perpendicular to AB.
OK = [kx cos θ − d ] and OL = x cos θ + d . Also Thus ∠BNO = 90°. Since NB and OD are
AK = kx sin θ and DL = x sin θ. Using perpendicular to BC, it follows that BDON is a
AK 2 + OK 2 = AO 2 = DO 2 = DL2 + OL2, rectangle.
we get k 2x 2 sin 2 θ + (kx cos θ − d )2 Aliter We can also get the conclusion using
trigonometry. Observe that ∠NFB = ∠AFK
= x sin θ + (x cos θ + d ) .
2 2 2
= 90° − ∠A and ∠BNF = 180° − ∠B since BCKN
D L C D L C is a cyclic quadrilateral. Using the sine rule in
x x x x
the ∆ BFN ,
M kx θM kx NB BF
K =
kx θ kx A B sin ∠NFB sin ∠BFN
O O
This reduces to
A K B c cos A
NB = = R cos A
2 sin C
Fig. 3 Fig. 4
But BD = a/2 = R sin A . Thus
ND2 = NB 2 + BD2 = R 2
Simplification gives
(k 2 − 1) x 2 = 2xd (k + 1) cos θ This gives ND = R
Since, k + 1 > 0, we get (k − 1) x = 2d cos θ. 26. We observe that ∠AIB = 90° + (C/2). Extend
Thus CA to D such that AD = AI . Then, CD = CB by
the hypothesis. Hence,
AB − CD = 2(AK − LD) = 2(kx sin θ − x sin θ)
∠CDB = ∠CBD = 90° − (C/2). Thus
= 2(k − 1) x sin θ = 4d cos θ sin θ C
= 2d sin 2θ
If ∠AMB = 60°, then 2θ = 60° . If ∠AMD = 60°,
then 2θ = 120°. In either case sin 2θ = 3/2. If I
d = 2, then AB − CD = 2 3, in both the cases.
25. Let O be the circumcentre of ABC. Join A B
OD, ON and OF . We show that BDON is a
rectangle. It follows that DN = BO = R, the D
circumradius of ABC.
C C
A ∠AIB + ∠ADB = 90° + + 90° − = 180°
2 2
K
Hence, ADBI is a cyclic quadrilateral. This
implies that
F B
∠ADI = ∠ABI =
N O 2
But ADI is isosceles, since AD = AI . This gives
∠DAI = 180° − 2(∠ADI ) = 180° − B
B D C
Thus, ∠CAI = B and this gives A = 2B. Since
Observe that ∠NBC = ∠NKC = 90°. Hence C = B, we obtain 4B = 180° and hence B = 45°.
BCKN is a cyclic quadrilateral. Thus We thus get A = 2B = 90°.
∠KNB = 180° − ∠BCA . But ∠BOA = 2∠BCA
27. (i) Let S and T be two points on the boundary
and OF bisects ∠BOA . Hence, ∠BOF = ∠BCA.
of Γ, with S lying on the side AB and T lying on
We thus obtain
www.pdfworld.in
www.pdfworld.in
Geometry 571
the side PQ of Γ. (see Fig. 1). Join TA , TB , TS . Similarly, we obtain

Now ST lies between TA and TB in ∆ TAB. One v fa w bc
of ∠AST and ∠BST is at least 90°, say = , =
y cd z ef
∠AST ≥ 90°. Hence, AT ≥ TS . But AT lies inside
∆ APQ and one of ∠ATP and ∠ATQ is at least Multiplying, these three equalities, we get
90°, say ∠ATP ≥ 90°. Then AP ≥ AT . Thus, we uvw = xyz.
get TS ≤ AT ≤ AP ≤ 1. Hence, x 2y 2z 2 = u 2v 2w 2 = u 2 (zx )(xy ).
Q F
T Z2 This gives u 2 = yz, as desired.
P E
29. Draw AL ⊥ EF and extend it to meet AB in M .
Y We show that AL = LM . First we show that
A , F , I , E are concyclic. We have
X
∠A
∠BIC = 90° + = 90° + 30° = 120°.
2
A S B C Z1 D
Hence, ∠FIE = ∠BIC = 120°. Since ∠A = 60°, it
Fig. 1 Fig. 2
follows that A , F , I , E are concyclic.
Hence,
(ii) Let X and Y be points in the interior Γ. Join
XY and produce them on either side to meet ∠BEF = ∠IEF = ∠IAF = ∠A/2. This gives
the sides CD and EF of Γ at Z1 and Z 2 ∠B ∠A
∠AFE = ∠ABE + ∠BEF = +
respectively. We have 2 2
(XZ1 + YZ1 ) + (XZ 2 + YZ 2 )
A
= (XZ1 + XZ 2 ) + (YZ1 + YZ 2 ) = 2Z1Z 2 < 2.
by the first part. Therefore, one of the same
XZ1 + YZ1 and XZ 2 + YZ 2 is at most 1. We may L E
choose Z accordingly as Z1 or Z 2. F
28. Let OA = a , OB = b , OC = c , OD = d , OE = e , I
OF = f , [OAB ] = x , [OCD] = y , [OEF ] = z ,
[ODE ] = u , [OFA ] = v and [OBC ] = w . We are B M C
given that v 2 = zx , w 2 = xy and we have to
prove that u 2 = yz. Since ∠ALF = 90°, we see that
D ∠B ∠A
∠FAM = 90° − ∠AFE = 90° − −
E 2 2
u
d ∠C
e y = = ∠FCM .
z 2
f O C
c This implies that F , M , C , A are concyclic. It
F w
v a b follows that
x
B ∠FMA = ∠FCA
A
∠C
= = ∠FAM
Since, ∠AOB = ∠DOE, we have 2
1 Hence, FMA is an isosceles triangle. But
de sin ∠DOE
u 2
= =
de FL ⊥ AM . Hence, L is the mid point of AM or
x 1 ab sin ∠AOB ab AL = LM .
2
www.pdfworld.in
www.pdfworld.in
Level 2
1. Let O be the centre of the circle and P be the Similarly TC ⊥ AC since BH ⊥ AC . Thus the
foot of perpendicular from C to MN . circle on AT as diameter passes through
A
N D
B and C. This is the circumcircle of ∆ ABC. If R
is its radius, then
P AT = 2R , BC = 2R sin A = 2R sin 30° = R
O Hence AT = 2BC.
M
Aliter We can assume that A
the circumcentre of ∆ABC is
at the origin. If R is the
B C circumradius,
BC = 2R sin A = R. Also if H
Z1 , Z 2 , Z3 are complex
Then, OM is perpendicular to AB , ON is
numbers representing B C
perpendicular to AD and OM = ON = the M
A , B , C respectively than
radius of the circle. So, AMON is a square. T
Z1 + Z 2 + Z3 represents H
1 Z 2 + Z3
∠MCN = ∠MON = 45° and M is . If t
2 2
∠CMP + ∠CNP = 135° = ∠CMP + ∠CMB represents T , then
= ∠CNP + ∠CND. t + Z1 + Z 2 + Z3 Z 2 + Z3
=
Hence, ∠CNP = ∠CMB and ∠CMP = ∠CND. 2 2
Thus we see that the right ∆CNP and ∆ CMB ⇒ t = − Z1 ⇒ AT = 2T = 2BC
are similar, and ∠CMP and ∠CND are similar.
3. Since ∠BFY = ∠BXF and A
So
∠FBY = ∠XBF we have
CN CP CM CP
= , = ∆BFY and ∆BXF are
CM CB CN CD similar, so that X
CP 2 FY : FX = BF : BX F E
Multiplying, 1 = . Hence, the area of …(i)
CB ⋅ CD Similarly, we get
the rectangle is Y Z
DY : DX = BD : BX …(ii)
CB ⋅ CD = CP = 5 = 25
2 2
B D C
As BF = BD, we have
Aliter Let AB = a , BC = b , BM = x , DN = y from Eqs. (i) and (ii)
and r = radius of the circle. Producing MO to that
meet the opposite side we can see that FY : FX = DY : DX
x 2 + y 2 = OC 2 = r 2. Since AX = DX , we get
Thus FY : FX = DY : AX …(iii)
CM 2 = b 2 + x 2 = (r + y )2 + x 2 = 2r 2 + 2ry = 2br Since, X , F , Y , D are concyclic we have
CN 2 = a 2 + y 2 = (r + x )2 + y 2 = 2r 2 + 2rx = 2ar ∠FYD = ∠AXF …(iv)

CP Thus, we get from Eqs. (iii) and (iv) that ∆FYD
Also = sin ∠CNP ; is similar to ∆FXA.
CN
Hence, ∠YFD = ∠XFA = ∠XDF so that FY | | XD.
CM = 2r sin ∠CNM .
Similarly we have EZ | | XD. Thus, FY | | EZ .
Hence, CM ⋅ CN = 2r ⋅ CP = 10r .
Therefore, FYZE is an isosceles trapezoid and
Thus (2ar )(2br ) = (10r )2 ⇒ ab = 25 then EY = FZ .
2. The diagonals BC and TH of the quadrilateral 4. Let O1 , O2 and O denote the centres of the
BTCH are bisected at M. Hence, BTCH is a circles and let A1 , A 2 , B1 and B2 denote the
parallelogram. Since CH ⊥ AB and CH | | TB , we points of tangency of these circles with the
have TB ⊥ AB.
www.pdfworld.in
www.pdfworld.in
Geometry 573
semicircle, as shown in the diagram. Then, Suppose now that θ is the base angle forT1.
O1O2 = R1 + R2 , O1A1 = R1 and O2A 2 = R2 , so Then, the base angle for Tn is
B2
B1 O O2
1
θ
A1 O A2
θ
A1A 2 = (O1O2 )2 − (O1A1 − O2A 2 )2
= (R1 + R2 )2 − (R1 − R2 )2 = 2 R1R2 90° 90° 90°

90° − + −
2 4 8
Also OB1 = 1, OB2 = 1, O1B1 = R1 and O2B2 = R2 −2 90° −1 θ
+ . . . + (−1)n + (−1)n .
so that OO1 = 1 − R1 and OO2 = 1 − R2. 2n −2
2n −1
Therefore, A1A 2 = OA1 + OA 2 Note that the limit as n → ∞ of the above is

90°
= (OO1 )2 − (O1A1 )2 + (OO2 )2 − (O2A 2 )2 = 60° by the formula for the sum of an
1 + 1/2
= (1 − R1 )2 − R12 + (1 − R2 )2 − R22 infinite geometric series. Since each Tn is
isosceles, the angles of Tn do approach 60° as
= 1 − 2R1 + 1 − 2R2 n → ∞.
Thus, 1 − 2R1 + 1 − 2R2 = 2 R1R2 6. Let AP , AQ A
Squaring, then dividing each term by 2 and produced meet BC
rearranging the terms, we get in D, E respectively.
L
Since MK is parallel
(1 − 2R1 )(1 − 2R2 ) = 2R1R2 + R1 + R2 − 1. M
Q
to AE, we have P
Square both sides and simplify. ∠AEK = ∠MKB.
8R1R2 = (2R1R2 + R1 + R2 )2 …(i) Since BK = BM ,
both being
So, 2 2R1R2 = 2R1R2 + R1 + R2 tangents to the B D K E C
circle from
Thus, R1 + R2 = 2 R1R2 ( 2 − R1R2 ) …(ii)
B , ∠MKB = ∠BMK . This with the fact that MK is
≤ (R1 + R2 )( 2 − R1R2 ) parallel to AE gives us ∠AEK = ∠MAE. This
and therefore R1R2 ≤ 2 − 1 shows that MAEK is an isosceles trapezoid. We
conclude that MA = KE . Similarly, we can
Now consider the function f (x ) = 2x ( 2 − x ). prove that AL = DK . But AM = AL. We get that
f (x ) is increasing on the interval (0, 1/ 2 ) since DK = KE . Since KP is parallel to AE , we get
f ′ (x ) = 2 2 − 4x > 0 for x in the interval. Since DP = PA and similarly EQ = QA . This implies
0 < R1R2 ≤ 2 − 1 < 1/ 2 and that PQ is parallel to DE and hence bisects
R1 + R2 = f ( R1R2 )from Eq. (ii), R1 + R2 attains AB , AC when produced.
its maximum when R1R2 = 2 − 1. Hence [The same argument holds even if one or both
of P and Q lie outside ∆ABC].
R1 + R2 ≤ 2( 2 − 1)[ 2 − ( 2 − 1)] = 2( 2 − 1).
7. Let denote the area of figure. We have
Equally holds when R1 = R2.
S
5. Note that the base angle of Tn is equal to the
angle opposite the base of Tn + 1 (as the figure
indicates). Therefore if θ is the base angle for
Tn , then the base angle for the next triangle P
(Tn + 1 ) is
180° − θ θ
= 90° − .
2 2 R
Q
www.pdfworld.in
www.pdfworld.in
[PQRS ] = [PQR ] + [RSP ] = [QRS ] + [SPQ ] . If P = I is the incentre of ∆ ABC, and r its
Let us write PQ = p , QR = q , RS = r , SP = s. The inradius, then it is clear that
above relations reduce to A1I = B1I = C1I = 2r . It follows that I is the
circumcentre of A1B1C1. On the otherhand
pq sin ∠PQR + rs sin ∠RSP
if P = I1 is the excentre of ABC opposite
= qr sin ∠QRS + sp sin ∠SPQ A and r1 the corresponding exradius, then
Using p = r and ( 3 + 1) q = s and dividing by again we see that A1I1 = B1I1 = C1I1 = 2r1.
pq, we get Thus I1 is the circumcentre of A1B1C1.
sin ∠PQR + ( 3 + 1) sin ∠RSP A1
A
= sin ∠QRS + ( 3 + 1) sin ∠SPQ .
B1
Therefore, sin ∠PQR − sin ∠QRS
= ( 3 + 1)(sin ∠SPQ − sin ∠RSP ).
B C
This can be written in the form
∠PQR − ∠QRS ∠PQR + ∠QRS
2 sin cos C1
2 2
∠SPQ − ∠RSP I1
= ( 3 + 1)2 sin
2 (b) Let P = O be the circumcentre of ABC. By
∠SPQ + ∠RSP definition, it follows that OA1 bisects and is
cos
2 bisected by BC and so on. Let D, E , F be the
mid points of BC , CA , AB respectively.
Using the relations
Then, FE is parallel to BC. ButE , F are also
∠PQR + ∠QRS ∠SPQ + ∠RSP mid points of OB1 , OC1 and hence FE is
cos = − cos
2 2 parallel to B1C1 as well. We conclude that
∠SPQ − ∠RSP ( 3 − 1) BC is parallel to B1C1. Since, OA1 is
and sin = − sin 15° = − , perpendicular to BC, it follows that OA1 is
2 2 2
perpendicular to B1C1. Similarly OB1 is
we obtain perpendicular to C1A1 and OC1 is
∠PQR − ∠QRS ( 3 − 1) 1 perpendicular to A1B1. These imply that O
sin = ( 3 + 1)[− ]=
2 2 2 2 is the orthocentre of A1B1C1. (This applies
whether O is inside or outside ABC).
This shows that
(c) Let P = H , the orthocentre of ABC. We
∠PQR − ∠QRS π
= consider two possibilities; H falls inside
2 4 ABC and H falls outside ABC.
3π Suppose H is inside ABC; this happens if ABC
or
4 is an acute triangle. It is known that A1 , B1 , C1
Using the convexity of PQRS, we can rule out lie on the circumcircle of ABC. Thus
the latter alternative. We obtain ∠C1A1A = ∠C1CA = 90° − A . Similarly
∠B1A1A = ∠B1BA = 90° − A . These show that
π
∠PQR − ∠QRS = . ∠C1A1A = ∠B1A1A . Thus A1A is an internal
2 bisector of ∠C1A1B1. Similarly we can show
8. (a) A that B1 bisects ∠A1B1C1 and C1C bisects
B1 ∠B1C1A1. Since A1A , B1B , C1C concur at H , we
C1 conclude that H is the incentre of A1B1C1.
OR
I
If D, E , F are the feet of perpendiculars of
A , B , C to the sides BC , CA , AB respectively,
then we see that EF , FD, DE are respectively
C
parallel to B1C1 , C1A1 , A1B1. This implies that
B
∠C1A1H = ∠FDH = ∠ABE = 90° − A , as BDHF
A1 is a cyclic quadrilateral. Similarly, we can show
that ∠B1A1H = 90° − A . It follows that A1H is
www.pdfworld.in
www.pdfworld.in
Geometry 575
the internal bisector of ∠C1A1B1. We can OR

proceed as in the earlier case. Let m a denote the median AD. Then, we can
If H is outside ABC, the same proofs go compute
through again, except that two of c 2 + m a2 − (a 2/4) 2∆
A1H , B1H , C1H are external angle bisectors cos α = , sin α = ,
2cm a cm a
and one of these is an internal angle bisector.
Thus, H becomes an excentre of ∆ A1B1C1. where ∆ denotes the area of ∆ ABC. These
two expressions give
9. Let α denote the equal angles ∠BAD = ∠D CA.
c 2 + m a2 − (a 2/4)
Using sine rule in cot α =
4∆
∆ DAB and ∆DAC, we get Similarly, we obtain
AD BD CD AD b 2 + m a2 − (a 2/4)
= , = . cos ∠CAD =
sin B sin α sin 15° sin α 4∆
Thus, we get
c 2 − a2
cot α − cot 15° =
4∆
Similarly, we can also obtain
A O c 2 − a2
15° cot B − cot α = ,
α 4∆
giving us the relation
F
B α
E cot B = 2 cot α − cot 15°
D
C If B is acute, then
2 cot α > cot 15° = 2 + 3 > 2 3.
Eliminating α (using BD = DC and It follows that cot α > 3. This implies that
2α + B + 15° = π), we obtain α < 30° and hence
1 + cos (B + 15° ) = 2 sin B sin 15°. But we know
that B = 180° − 2α − 15° > 105°
2 sin B sin 15° = cos (B − 15° ) − cos (B + 15° ). This contradiction forces that ∠ B is obtuse
Putting β = B − 15°, we get a relation and consequently ∠ADC is obtuse.
1 + 2 cos ( β + 30) = cos β. We write this in the Since ∠BAD = α = ∠ACD, the line AB is tangent
form to the circumcircle Γ of ADC at A. Hence, OA is
(1 − 3 ) cos β + sin β = 1. perpendicular to AB. Draw DE and BF
perpendicular to AC and join OD. Since
Since sin β ≤ 1, it follows that (1 − 3 ) cos β ≥ 0. ∠DAC = 15°, we see that ∠DOC = 30° and
We conclude that cos β ≤ 0 and hence that β is hence DE = OD/2. But DE is parallel to
obtuse. So is angle B and hence ∠ADC. BF and BD = DC shows that BF = 2DE . We
We have the relation (1 − 3 ) cos β + sin β = 1. conclude that BF = DO. ButDO = AO , both
If we set x = tan ( β/2), then we get, using being radii of Γ. Thus BF = AO. Using right
cos β = (1 − x 2 ) / (1 + x 2 ), sin β = 2x/(1 + x 2 ). ∆BFO and ∆BAO, we infer that AB = OF .We
conclude that ABFO is a rectangle. In
( 3 − 2) x 2 + 2x − 3 = 0. particular ∠AOF = 90°. It follows that
Solving for x , we obtain x = 1 or x = 3 (2 + 3 ). ∠AOD = 90° − ∠DOC = 90° − 30° = 60°
If x = 3 (2 + 3 ), then Since, OA = OD, we conclude that AOD is
tan ( β/2) > 2 + 3 = tan 75° giving us β > 150°. equilateral.
This forces that B > 165° and hence OR
B + A > 165° + 15° = 180°, a contradiction, Note that ∆ABD and ∆CBA are similar. Thus
thus x = 1 giving us β = π/2. This gives B = 105° we have the ratios
and hence α = 30°. Thus ∠DAO = 60°. Since AB CB
OA = OD, the result follows. =
BD BA
www.pdfworld.in
www.pdfworld.in
This reduces to a 2 = 2c 2 giving us a = 2c . 11. It is easy to observe that there is a triangle

This is equivalent to sin 2 (α + 15° ) = 2 sin 2 α. b c a
with sides a + , b + , c + . Using Heron’s
2 2 2
We write this in the form
formula, we get
cos 15° + cot α sin 15° = 2.
16 [ABC ]2 = (a + b + c )(a + b − c )
Solving for cot α, we get cot α = 3. We
conclude that α = 30° and the result follows. (b + c − a )(c + a − b )
3
10. (a) Suppose all the six statements are true. and 16 [A1B1C1 ] = 2
(a + b + c )(−a + b + 3c )
16
Then ABDE , BCEF , CDFA are isosceles
trapeziums; if K , L , M , P , Q , R are the mid (−b + c + 3a )(−c + a + 3b ).
points of AB , BC , CD, DE , EF , FA Since a , b , c are the sides of a triangle, there
respectively, then we see that KP ⊥ AB , ED; are positive real numbers p , q , r such that
LQ ⊥ BC , EF and MR ⊥ CD, FA . a = q + r , b = r + p , c = p + q. Using these
E P D relations we obtain
Q [ABC ]2 16pqr
M = .
Z [A1B1C1 ]2 3(2p + q )(2q + r )(2r + p )
F
X Y C Thus it is sufficient to prove that
R (2p + q )(2q + r )(2r + p ) ≥ 27pqr ,
L
A
for positive real numbers p , q , r . Using
K B
AM-GM inequality, we get
If AD, BE , CF themselves concur at a point 2p + q ≥ 3(p 2q )1/3 ,
O, then OA = OB = OC = OD = OE = OF . (O is 2q + r ≥ 3(q 2r )1/3 , 2r + p ≥ 3(r 2p )1/3.
on the perpendicular bisector of each of
the sides). Hence, A , B , C , D, E , F are Multiplying these, we obtain the desired
concyclic and lie on a circle with centre O. result. We also observe that equality holds if
Otherwise these lines AD, BE , CF form a and only p = q = r . This is equivalent to the
triangle , say XYZ. (See Fig.) Then, statement that ABC is equilateral.
KX , MY , QZ , when extended, become the 12. Produce AP to meet BC in Q. Join KE and KF .
internal angle bisectors of the ∆XYZ and Draw perpendiculars from F and E on to BC to
hence concur at the incentre O′ of XYZ. As meet it in M and L respectively. Let us denote
earlier O′ lies on the perpendicular bisector ∠BKF by α and ∠CKE by β. We show that α = β
of each of the sides. Hence, by proving tan α = tan β. This implies that
O ′ A = O ′ B = O ′ C = O ′ D = O ′ E = O ′ F, ∠DKF = ∠DKE. (See Figure below).
giving the concyclicity of A , B , C , D, E , F . Since the cevians AQ , BE and CF the concur,
(b) Suppose (a1 ), (a 2 ), (b1 ), (b2 ) are true. Then, we may write
we see that AD = BE = CF . Assume that (c1 )
A
is true. Then, CD is parallel to AF. It follows
that ∆YCD and ∆YFA are similar. This gives
FY YC FY + YC FC
= = = =1 F
AY YD AY + YD AD D
We obtain FY = AY and YC = YD. This E
forces that ∆ CYA and ∆DYE are congruent. P
In particular AC = DF so that (c 2 ) is true.
B M K QL C
The conclusion follows from (a). Now
assume that (c 2 ) is true; i.e., AC = FD. We BQ z CE x AF y
= , = , = ,
have seen that AD = BE = CF . It follows QC y EA z FB x
that ∆ FDC and ∆ACD are congruent. In
particular ∠ADC = ∠FCD. Similarly, we can We observe that
show that ∠CFA = ∠DAF. We conclude FD [AFD] [PFD] [AFP ]
= = = .
that CD is parallel to AF giving (c1 ). DE [AED] [PED] [AEP ]
www.pdfworld.in
www.pdfworld.in
Geometry 577
However standard computations involving qr (q + r ) ≤ p 2r + pr 2 < p 2r + pr 2 + p 2q + pq 2

bases give
= p 2 (q + r ) + p (q 2 + r 2 )
y
[AFP ] = [ABP ], which contradicts (*). Similary, a ≤ c is also not
y +x
possible. This proves (i). Suppose 2R ≤ rb . As
z
[AEP ] = [ACP ] above this takes the form
z+x
q 2 (r + p ) + q (r 2 + p 2 ) ≤ pr (p + r ) (**)
z
and [ABP ] = [ABC ],
x +y + z Since, a > b and a > c , we have q > p , r > p. Thus
y q 2r > p 2r and qr 2 > pr 2. Hence,
[ACP ] = [ABC ] .
x +y + z q 2 (r + p ) + q (r 2 + p 2 ) > q 2r + qr 2
Thus, we obtain > p 2r + pr 2 = pr (p + r )
FD x + z which contradicts (**). Hence, 2R > rb .
= .
DE x + y Similarly, we can prove that 2R > rc . This
On the other hand proves (ii).
FM FB sin B EL EC sin C 14. Observe that KLMN is a paralellogram,Q is the
tan α = = , tan β = = .
KM KM KL KL mid point of MK and hence NL also passes
 x   x  through Q. LetT be the point of intersection of
Using FB =   AB , EC =   AC and AC and BD, and let S be the point of
x + y  x + z
intersection of BD and MN .
AB sin B = AC sin C , we obtain M
D C
tan α  x + z   KL   x + z   DE 
=   =   
tan β  x + y   KM   x + y   FD 
x + z x + y  N L
=    =1 Q
x + y  x + z
We conclude that α = β.
abc ∆ A K B
13. We know that, 2R = and ra = , where
2∆ s −a
a+b+c Consider the ∆ MNK. Note that SQ is parallel to
a , b , c are the sides of the ∆ ABC, s = NK and Q is the mid point of MK. Hence S is
2
and ∆ is the area of ABC. Thus, the given the mid point of MN . Since MN is parallel to
condition 2R ≤ ra translates to AC, it follows that T is the mid point of AC.
Now Q is the circumcentre of ∆ABC and the
2∆2
abc ≤ median BT passes through Q. Here, there are
s −a two possibilities.
Putting s − a = p , s − b = q , s − c = r , we get (i) ABC is a right triangle with ∠ABC = 90° and
a = q + r , b = r + p , c = p + q and the condition T = Q ; and
now is (ii) T ≠ Q in which case BT is perpendicular to
p (p + q )(q + r )(r + p ) ≤ 2∆2 AC.
But Heron’s formula gives, Suppose ∠ABC = 90° and T = Q . Observe that
∆2 = s (s − a )(s − b )(s − c ) = pqr (p + q + r ) . We Q is the circumcentre of the ∆DCB and hence
∠DCB = 90°. Similarly DAB = 90°. It follows
obtain (p + q )(q + r )(r + p ) ≤ 2qr (p + q + r ).
that ∠ADC = 90° and ABCD is a rectangle. This
Expanding and effecting some cancellations,
implies that KLMN is a rhombus. Hence,
we get
LK/LM = 1 and this gives CD = CB. Thus, ABCD
p 2 (q + r ) + p (q 2 + r 2 ) ≤ qr (q + r ) (*) is a square.
Suppose a ≤ b. This implies that q + r ≤ r + p In the second case, observe that BD is
and hence q ≤ p. This implies that q 2r ≤ p 2r and perpendicular to AC , KL is parallel to
qr 2 ≤ pr 2 giving AC and LM is parallel to BD. Hence, it follows
www.pdfworld.in
www.pdfworld.in
that ML is perpendicular to LK. Similar 1 2 1

= AI 2 − ma + m a2
reasoning shows that KLMN is a rectangle. 4 36
Using LM/LM = CD/CB , we get that CBD is 2
= r 2 cosec 2 (A/2) − m a2
similar toLMK . In particular, 9
∠LMK = ∠CBD = α (say). Since LM is parallel to 2
DB, , we also get ∠BQK = α. Since KLMN is a Hence, r 2 cot 2 (A/2) = m a2. From the above,
9
cyclic quadrilateral we also get we get
∠LNK = ∠LMK = α. Using the fact that BD is 2
parallel to NK , we get ∠LQB = ∠LNK = α. Since  b − c  = 2 ⋅ 1 (2b 2 + 2c 2 − a 2 )
 
BD bisects ∠CBA, we also have ∠KBQ = α.  2  9 4
Thus Simplification gives 5b 2 + 13c 2 − 18bc = 0. This
QK = KB = BL = LQ can be written as (b − c )(5b − 13c ) = 0. As b ≠ c ,
and BL is parallel to QK. This gives QM is we get 5b − 13c = 0. To conclude,
parallel to LC and a = 2c , 5b = 13c yield
QM = QL = BL = LC a b c
= =
It follows that QLCM is a parallelogram. But 10 13 5
∠LCM = 90°. Hence, ∠MQL = 90°. This implies 16. (i) Extend BI to meet the circumcircle in F.
that KLMN is a square. Also observe that Then, we know that FA = FI = FC (See Figure).
∠LQK = 90° and hence ∠CBA = ∠LQK = 90°. Let BI : IF = λ : µ. Applying Stewart’s theorem
This gives∠CDA = 90° and hence ABCD is a to ∆ BAF, we get
rectangle. Since BA = BC , it follows that ABCD
A
is a square.
15. Let I be the A
F
incentre of ∆ O
ABC and D be its E K
K L
projection on BC.
Observe that r N I
r
AB ≠ AC as I L C
r B D
AB = AC implies
that D = L = M . So, B D M C
assume that λAF 2 + µAB 2 = (λ + µ )(AI 2 + BI ⋅ IF ).
AC > AB. Let N be the projection of I on KL .
Similarly, Stewart’s theorem to ∆BCF gives
Then, the perpendicular IN from I to KL is a
bisector of KL and as AK = LM , it is a bisector λCF 2 + µBC 2 = (λ + µ )(CI 2 + BI ⋅ IF ).
of AM also. Hence, AI = IM . Since, CF = AF , subtraction gives
r µ (AB 2 − BC 2 ) = (λ + µ )(AI 2 − CI 2 )
But AI = = r cosec (A/2)
sin (A/2)
Using the standard notations AB = c , BC = a ,
and IM 2 = ID2 + DM 2 = r + (BM − BD)2 CA = b and s = (a + b + c )/2, we get
2 AI 2 = r 2 + (s − a )2 and CI 2 = r 2 + (s − c )2
= r 2 +  − (s − b ) .
a
where r is the inradius of ∆ABC.
2 
Thus,
Hence, r 2 cosec 2 (A/2) = r 2 + ((a/2) − (s − b )2 )2 µ (c 2 − a 2 ) = (λ + µ )((s − a )2 − (s − c )2 )
giving r cot (A/2) = ((b − c ) /2) . Since b > c ,
2 2 2
= (λ + µ )(c − a )b
we obtain r cot (A/2) = ((b − c )/2). So
It follows that either c = a or µ (c + a ) = (λ + µ )b.
(s − a ) = ((b − c ) /2). This gives a = 2c .
But c = a implies that a = b = c since a , b , c are
As KN = NL and AK = KL = LM , we have in arithmetic progression. However, we have
NL = AM/6. We also have AN = NM . taken a non-equilateral ∆ ABC. Thus c ≠ a and
Now, r 2 = IL2 = IN 2 + NL2 = AI 2 − AN 2 + NL2 we have µ (c + a ) = (λ + µ )b. But c + a = 2b and
we obtain 2bµ = (λ + µ )b. We conclude that
www.pdfworld.in
www.pdfworld.in
Geometry 579
λ = µ . This in turn tells that I is the mid point Thus, ID = IE = IK and I is the circumcentre of
of BF. Since OF = OB, we conclude that OI is DKE.
perpendicular to BF. Aliter
Aliter Observe that AK = bc/(a + c ) = c/2 = AE . Since,
Applying Ptolemy’s theorem to the cyclic AI bisects ∠ A, we see that AIE is congruent
quadrilateral ABCF, we get to AIK. This gives IE = IK . Similarly CID is
AB ⋅ CF + AF ⋅ BC = BF ⋅ CA congruent to CIK giving ID = IK . We conclude
Since CF = AF , we get that
CF (c + a ) = BF ⋅ b = BF (c + a )/2. This gives ID = IK = IE
BF = 2CF = 2IF . Hence, I is the mid point ofBF
17. Applying cosine rule to ∆ ABC, we get
and as earlier we conclude that OI is
perpendicular to BF. B
b C
Aliter
°
30°
30
Join BO. We have to prove that ∠BIO = 90°, a
which is equivalent to BI 2 + IO 2 = BO 2. Draw A
c
IL perpendicular toAB. Let R denote the
circumradius of ABC and let ∆ denote its area.
Observe that BO = R , IO 2 = R 2 − 2Rr ,
BL ca D
BI = = (s − b )
cos (B/2) s (s − b )
Thus, we obtain AC 2 = a 2 + b 2 − 2ab cos 120° = a 2 + b 2 + ab
ac Observe that
BI 2 = ac (s − b )/s =
3 ∠DAC = ∠DBC = 120° − 30° = 90°.
Since a , b , c are in arithmetic progression. Thus, we get
Thus, we need to prove that
AC 2 4 2
ac
+ R 2 − 2Rr = R 2 c2 = = (a + b 2 + ab )
3 cos2 30° 3
This reduces to proving 2Rr = ac/3. 4 2
So, c 2 − (a + b )2 = (a + b 2 + ab )
abc ∆ abc abc ac 3
But 2Rr = 2 ⋅ ⋅ = = = ,
4∆ s 2s a+b+c 3 − (a 2 + b 2 + 2ab )
using a + c = 2b. This proves the claim. (a − b )2
= ≥0
(ii) Join ID. Note that ∠BIO = ∠BDO = 90°. 3
Hence B , D, I , O are concyclic and hence
This proves c ≥ a + b and thus (i) is true.
∠BID = ∠BOD = A. Since, ∠DBI = ∠KBA = B/2,
it follows that ∆ BAK and ∆BID are similar. This For proving (ii), consider the product
gives Q = (α + β + γ )(α − β − γ ) (α + β − γ )(α − β + γ ),
BA BK AK
= = . where α = c + a , β = c + b and
BI BD ID
However, we have seen earlier that BI = ac/3. γ = c − a − b. Expanding the product, we get
Moreover AK = bc/(a + c ). Thus we obtain Q = (c + a )2 + (c + b )2 + (c − a − b )2
BA ⋅ BD 1
BK = = 3ac , −2(c + a )(c + b ) − 2(a + a )(c − a − b )
BI 2
AK ⋅ BI 1 ac − 2(c + b )(c − a − b )
ID = =
BA 2 3 = − 3c 2 + 4a 2 + 4b 2 + 4ab = 0
1 ac Thus at least one of the factors must be equal
By symmetry, we must have IE = . Finally
2 3 to 0. Since, α + β + γ > 0 and α + β − γ > 0, it
b 1 1 ac follows that the product of the remaining two
IK = ⋅ BK = BK = factors is 0. This gives
a+b+c 3 2 3
www.pdfworld.in
www.pdfworld.in
c + a − c + b = c −a −b Substituting this in the equation satisfied by z,

we obtain
or c + a − c + b = − c −a −b
(1 + λ )2 − 3(1 + λ )(λ3 − λ2 − 2) + 9 = 0.
We conclude that
This may be written in the form
| c + a − c + b |= c −a −b
(λ − 2)( 3λ3 + 6λ + 8λ + 8) = 0. Here, the second
18. Let AD = h , PD = y and BD = DC = a. We factor is positive because λ > 0. We conclude
observe that BP 2 = a 2 + y 2. Moreover, that λ = 2.
A
19. (1) Since E is the mid point of AC, we have
A/2 AE = EC = b/2. Since, BD bisects ∠ABC, we also
E know that CD = ab/(a + c ). Since, BE bisects
b h–y P
∠ABD, we also have
y
A
B a D a C b/2
DC a (h − y ) c E
PE = PA sin ∠DAC = (h − y ) = ,
AC b
4
where b = AC = AB. Using AP/PD = (h − y )/y , we B/ 4 D
B/ ab/(a+c)
obtain y = h/(1 + λ ). B/2
Thus, B a C
BP 2
(a + y )b
2 2 2
λ2 = = BD2 DE 2
PE 2 (h − y )2a 2 2
=
BA EA 2
But (h − y ) = λy = λh/(1 + λ ) and b = a + h . 2 2 2
a 2b 2
However, BD2 = BC 2 + CD2 = a 2 + ,
Thus, we obtain (a + c )2
(a 2 (1 + λ )2 + h 2 )(a 2 + h 2 )
λ4 = . b ab 
2
a 2h 2 DE 2 =  − 
a 2 a + c
Using m = and z = m 2 (1 + λ ), this simplifies
h Using these in the above expression and
to simplifying, we get
z 2 − z (λ3 − λ2 − 2) + 1 = 0. a 2 {(a + c )2 + b 2} = c 2 (c − a )2
Dividing by z, this gives Using c 2 = a 2 + b 2 and eliminating b, we obtain
1
z + = λ3 − λ2 − 2 c 3 − 2ac 2 − a 2c − 2a3 = 0
z
However z + (1/z ) ≥ 2 for any positive real Introducing t = c/a, this reduces to a cubic
number z. Thus, λ3 − λ2 − 4 ≥ 0. This may be equation;
written in the form (λ − 2)(λ2 + λ + 2) ≥ 0. But t 3 − 2t 2 − t − 2 = 0.
λ + λ + 2 > 0. (For example, one may check
2
Consider the function f (t ) = t 3 − 2t 2 − t − 2 for
that its discriminant is negative). Hence, λ ≥ 2. t > 0 (as c/a is positive). For 0 < t ≤ 2, we see
If λ = 2, then z + (1/z ) = 2 and hence z = 1. This that f (t ) = t 2 (t − 2) − t − 2 < 0. We also observe
gives m 2 = 1/3 or tan (A/2) = m = 1/ 3. Thus
that f (t ) = (t − 2)(t 2 − 1) − 4 is strictly increasing
A = 60° and hence ∆ ABC is equilateral.
on (2, ∞ ). It is easy to compute
Conversely, if ∆ ABC is equilateral, then
11
m = tan (A/2) = 1/ 3 and hence z = (1 + λ )/3. f (5/2) = − < 0 and f (3) = 4 > 0.
8
www.pdfworld.in
www.pdfworld.in
Geometry 581
Hence, there is a unique value of t in the ∴IB1AC1 is a rhombus and ∠BA1C1 = 120°
interval (5/2, 3) such that f (t ) = 0. We conclude ∠A1 = 60° (by concyclicity)
that
A is mid point of arc B1AC1 (QAB1 = AC1)
5 c
< <3 Hence, it follows that A1A bisects
2 a
∠A1 and I1 lies on line A1A .
(2) Let us take ∠B/4 = θ. Then,
∠A1
∠EBC = ∠DBE = θ and ∠CBD = 2θ. Using sine ⇒ ∠B1I1C1 = 90° + = 90° + 30° = 120°
rule in ∆ BEA and ∆ BEC, we get 2
BE
=
AE
,
BE
=
CE ∴ B1 , I , I1 , C1 are concyclic.
sin A sin θ sin 90° sin 3θ [QB1IC1 = 120° and A is centre]
Since, AE = CE , we obtain sin 3θ sin A = sin θ. 21. Let O1 , O2 and O3 be the centre of the circles
However A = 90° − 4θ. Thus, we get ΓA , ΓB and ΓC respectively.
sin 3θ cos 4θ = sin θ. Note that Let P be the circumcentre of the ∆O1O2O3.
c 1 sin 3θ
= = = 3 − 4 sin 2 θ Let x denote the common radius of three
a cos 4θ sin θ circles ΓA , ΓB , ΓC .
This shows that c/a < 3. Using ca = 3 − 4 sin 2 θ, A
it is easy to compute cos 3θ = ((c/a ) − 1) /2. O1
2
= cos 4θ =  − 1 − 1.
a 1 c
Hence,
c 2 a 
I P O
Suppose c/a ≤ 5/2. Then, ((c/a ) − 1)2 ≤ 9/4 and O2 O3
a/c ≥ 2/5. Thus
2 B L DXK M C
2 a 1
≤ =  c − 1 − 1 ≤ 9 − 1 = 1 ,
 
5 c 2 a  8 8
P is also the centre of circle Γ.
which is absurd. We conclude that c/a > 5/2.
[QO1P , O2P and O3P each exceed the radius of Γ
20. Here, IA1 = IB1 = IC1 = 2r where r is inradius of by x ]
∆ABC. Let D, X , K , L and M be respectively the
C1 A projections of I , P , O , O1 , O2 on BC.
x (s − b )  BL LO2 
Now, BL =  form = 
r  BD DI 
K
B I1 I B1 (where ID = r , BD = s − b)
x (s − c )
Similarly, CM =
r
A1
x a
∴ LM = a − (s − b + s − c ) = (r − x )
C r r
∴I is the circumcentre of ∆A1B1C1. PX is perpendicular bisector of LM .

Let K be the point of intersection of (QO2LMO3 is a rectangle and PX is
IB1 and AC. perpendicular bisector of O2O3)
IK = r 1 a
Then, Thus, LX = LM = (r − x )
IA = 2r and ∠IKA = 90° 2 2r
∴ ∠IAK = 30° BX = BL + LX
and ∠IAB1 = 60° x a
= (s − b ) + (r − x )
Hence, AIB1 is an equilateral triangle. r 2r
Similarly, ∆AIC1 is an equilateral, a x (b − c )
= −
∴We get, 2 2r
AB1 = AC1 = AI = IB1 = IC1 = 2r a b −c
DK = BK − BD = − (s − b ) =
Hence, ∠B1IC1 = 120° 2 2
www.pdfworld.in
www.pdfworld.in
a a x (b − c ) x (b − c ) and DB | | PA
XK = BK − BX = − + =
2 2 2r 2r Hence, we get
XK x
Hence, we get = ∠DPA = ∠BAP = ∠BCP = ∠NPC
DK r
∠BPC = 90°
∴Sides of ∆O1O2O3 are
Q
a and N is mid point of CB.
O2O3 = LM = (r − x )
r ∴NP = NC = NB for right angled ∆BPC.
b Hence, A , P and N are collinear.
O3O1 = (r − x )
r 23. Let ∠C be the smallest angle.
c CA ≥ AB and CB ≥ AB
and O1O2 = (r − x ) So that
r
Here, the altitude through C is the longest one.
QSides of O1O2O3 | | ABC. Let the altitude through C meets AB in D. Let H
And I is incentre of O1O2O3, be the orthocentre of ∆ABC.
IP r − x
∴ = Let CD extended meet the circumcircle of ABC
IO r in K.
PO x
Hence, = We have A
IO r CD = h max
XK PO
We get, = Using, K D
DK IO
CD = CH + HD H
∴I , P , O are collinear.
Which reduced to C
22. Extend CP to D. B
AH + BH ≤ CD + HD
Now, CP = PD
D A But AH = AK
BH = BK (Q∆DEK ≡ ∆DBH )
α
α DH = DK
α Apply Ptolemy’s theorem to the cyclic
P M quadrilateral BCAK.
We get, AB ⋅ CK = AC ⋅ BK + BC ⋅ AK
α
≥ AB ⋅ BK + AB ⋅ AK
α ⇒ CK ≥ AK + BK
B N C
24. Let D be the mid point ofBC ; M that of HK ; and
Let ∠BCP = ∠BAP = α, T that of OH. Then, PM is perpendicular to
BC = BD HK and PT is perpendicular to OH . Since, Q is
(QBP is perpendicular bisector of CD ) the reflection of P in HO, we observe that
∴BCD is an isosceles triangle. P , T , Q are collinear and PT = TQ . Let
Thus, ∠BDP = α QL , TN and OS be the perpendiculars drawn
respectively from Q , T and O on to the altitude
Then, ∠BDP = ∠BAP = α
AK , (See the figure).
Hence, B , P , A , D all lie on a circle.
A
∠DAB = ∠DPB = 90°
(Q P is mid point of CD and M is mid point of L
CA) Q
H
∴ PM | | DA N T
S O
where DA = 2PM = BP M P
Thus, DBPA is an isosceles trapezium B
K D C
www.pdfworld.in
www.pdfworld.in
Geometry 583
We have LN = NM , since T is the mid point of QC BC

= 1
QP : HN = NS , since T is the mid point of OH ; MA1 B1A1
and HM = MK , as P is the circumcentre of
Similarly, ∆ ACM and ∆ C1A1M are similar and
KHO, we obtain
we get
LH + HN = LN = NM = NS + SM
AC C1A1
which gives LH = SM . We know that, =
AM C1M
AH = 2OD
Thus, AL = AH − LH = 2OD − LH = 2SK − SM C1 A
= SK + (SK − SM ) = SK + MK Γ
αα αα B1
= SK + HM = SK + HS + SM αα
= SK + HS + LH = SK + LS = LK P M Q
This shows that L is the mid point of AK and
hence lies on the line joining the mid points of γ γ β
AB and AC. We observe that the line joining β
the mid points of AB and AC is also B γ C
β
perpendicular to AK . Since QL is
A1
perpendicular to AK, we conclude that Q also
lies on the line joining the mid points of
AB and AC. Using the point P , we get similar ratios
PB C B AB A B
Remark = 1 , = 1 1
A MA1 A1C1 AM MB1
It may happen that H is above L as in
the adjoining figure but the result QC A1C1 ⋅ B1C
Thus, =
remains true here as well. We have H PB C1B ⋅ B1A1
HN = NS ,LN = NM L Q
AC MB1 ⋅ C1A1
N and =
and HM = MK as earlier. Thus, M P T
AB A1B1 ⋅ C1M
HN = HL + LN and NS = SM + NM S O
give HL = SM MB1 C1A1
=
Now, AL = AH + HL = 2OD + SM K D C1M A1B1
= 2SK + SM MA1C1B ⋅ QC
=
= SK + (SK + SM ) = SK + MK C1M PB ⋅ B1C
= SK + HM = SK + HL + LM
= SK + SM + LM = LK However, ∆C1BM and ∆B1CM are similar, which
The conclusion that Q lies on the line joining the mid gives
points of AB and AC follows as earlier. B1C MB
=
C1B MC1
25. Let A = 2α. Then ∠A1AC = ∠BAA1 = α. Thus
∠A1B1C = α = ∠BB1A1 = ∠A1C1C = ∠BC1A1 Putting this in the last expression, we get
AC QC
We also have ∠B1CQ = ∠AA1B1 = β, say. If =
follows that ∆ MA1B1 and ∆ QCB1 are similar AB PB
and hence We conclude that PQ is parallel to BC.
www.pdfworld.in
www.pdfworld.in
Unit 6
Functions
Definition of Function
Let A and B be two given non-empty sets, then a function (or mapping or map
or transformation) f from A to B written as f : A → B or A f
→ B (to be read x p
f y
as f is a function from the set A to the set B or A is mapped to B under q
z r
function f ) is a rule which associates every element of A to a unique
element of B. t s
m
Here f (x ) = s, f (y ) = q, f (z ) = p, f (t ) = r , f (u ) = n. The elements f (x ), f (y ), f (z ), f (t ), u n
f (u ) of B are called images of x , y , z , t and u of A respectively and the
A B
elements x , y , z , t and u are called the pre-images of the elements s , q , p , r
and n of B respectively. The element m ∈ B has no pre-image in A.
If f : A → B, then the set A is called the domain of the function f and is generally denoted by D (f) or Df
and the set B is called the codomain of the function f, generally denoted by Cf or C ( f ). The set of images
of the elements of set A is called the range of function, which is generally denoted by Rf or R ( f ).
Hence, Domain of f = Dr = {x : x ∈ A , (x , f (x )) ∈ A },
Range of f = Rf = (f (x ) : x ∈ A , f (x ) ∈ B} ⊆ B
A function f is said to be integral, rational, irrational, non-negative, real or complex valued function
according as f (x ) ∈ Z , Q, R − Q , R + ∪ {0}, R or C. We shall discuss the nature of real valued functions only.
On the basis of images, a function is classified into two groups – one-one and many-one.
A function f from A to B (i.e., f : A → B) is said to be one-one (injective) iff different elements in A have
different images in B.
i.e., f (x 1 ) = f (x 2 ) ⇒ x 1 = x 2 , ∀ x 1 , x 2 ∈ A (0,5)
For example let us consider a function f defined as y = f (x ) = x + 5.
Here, f (x 1 ) = f (x 2 ) ⇒ x 1 + 5 = x 2 + 5 (–5,0)
⇒ x1 = x 2
i.e., images under function f are equal iff the pre-images are equal. In other words, different elements in
the domain of the function have different images. Hence f is one-one or injective. Here we observe that
any straight line parallel to y-axis intersects the graph of the function in at most one point. Also a
straight line parallel to x -axis intersects the graph in at most one point, hence the function is one-one or
equivalently, a straight line parallel to x -axis intersects the graph in not more than one point.
A function f : A → B is said to be many-one iff different element in A have same image in B.
i.e., f (x 1 ) = f (x 2 ) ⇒ x 1 ≠ x 2, for at least two elements x 1 and x 2 of A.
For example let a function f defined as f (x ) = x 2
www.pdfworld.in
www.pdfworld.in
Functions 585
We observe that
f (x 1 ) = f (x 2 ) ⇒ x 12 = x 22
⇒ x 12 − x 22 = 0
⇒ (x 1 + x 2 ) (x − x 2 ) = 0
⇒ x1 ≠ x 2 always
Which shows that images are equal even, if pre-images are not equal. Hence, the function under
consideration is many one.
On the basis of existence of pre-images, a function is also classified into two groups-onto and into.
Let f : A → B. Then, f is said to be onto (surjective) if every element of B has a pre-image in A.
Mathematically, y ∈ B ⇒ ∃ x ∈ A such that y = f (x ), ∀ y ∈ B, then f is called onto function or surjective
function.
For example, Let A = {1, 2, 3, 4, 5, − 4}, B = {1, 4, 9, 16, 25} and f : A → B defined by f (x ) = x 2, then, we
observe that every element of B has its pre-image in A, which is clear from the arrow diagram given
below
f
1 1
2 4
3 9
4 16
–4
5 25
A B
Hence, f is an onto function.
A function f : A → B is said to be into iff there exists at least one element in B having no pre-image in A.
i.e., ∃ at least one y ∈ B which has no pre-imagex in A or equivalently for at least one elements y ∈ B, there
does not exist x ∈ A such that y = f (x ) holds good.
Let A = {1, 2, 3, 4, 5}, B = {2, 1, 9, 16, 25, 36} and f : A → B defined by f (x ) = x 2.
A B
f 1
1 2
2 4
3 9
4 16
5 25
36
Here we observe that 2 ∈ B, 36 ∈ B have no pre-image in A. Hence, f is into.
Note
1. The number of functions from a finite set A containing m elements to a finite set B containing n
elements is equal to nm = (n(B ))n (A )
n P ; n ≥ m
2. The number of one-one (injective) functions from A to B =  m
 0; n < m
 n n− r n
3. The number of surjective functions (onto functions) = rΣ=1( − 1) Crr m ; if 1 ≤ n ≤ m
 0; if n > m
4. Number of bijective (one-one and onto) functions = n ! if m = n. If. m ≠ n the number of bijective,
function is 0. The above results can be proved with the help of permutation and combination).
www.pdfworld.in
www.pdfworld.in
Real Valued Function of a Real Variable

A function whose domain an range are subsets of the set R of real numbers is called a real valued
function.
Value of a Function
If y = f (x )be any function of x and x = a is admissible value of x , then the value of the function is obtained
simply by replacing x by a in y = f (x ) and f (a ).
Equality of Functions
Two functions f and g are said to be equal, i.e., f = g, if (i) Df = Dg , (ii) f (x ) = g (x ) for all x ∈ Df (or Dg ).
Inverse Image of an Element

Let f be a function defined from the set A to the set B, then the inverse image of an element b ∈ B under f
is denoted by f −1 (b ) to be read as ‘f inverse b’.
∴ f −1 (b ) = {a : a ∈ A and f (a ) ∈ B}
i.e., f −1 (b ) is the set of all those elements in A which has b as their pre-image.
Inverse Function
Let f be a function defined from the set A to the set B, i.e., f : A → B and g be a function defined from the
set B to the set A i.e., g : B → A ; then the function g is said to be inverse of f , if
g {f (x )} = x , ∀ x ∈ A and the function g is denoted by f −1.
Example 1 If f ( x + y + 1) = ( f ( x ) + f ( y ))2 and f ( 0) = 1 , for all x, y ∈ R , then find f ( x )

Solution We have, f ( x + y + 1) = ( f ( x ) + f ( y ))2, ∀ x, y ∈ R , and f ( 0) = 1.
Putting x = y = 0, we get
f (1) = ( f ( 0) + f ( 0))2 = ( 1 + 1)2 = 4 = (1 + 1)2
For x = 1, y = 0, f (1 + 0 + 1) = ( f (1) + f ( 0))2
= ( 2 + 1)2 = 32
⇒ f ( 2) = ( 2 + 1)2
and for x = 1, y = 1
f (1 + 1 + 1) = ( f (1) + f (1))2 = ( 2 + 2)2 = 42
⇒ f ( 3) = ( 3 + 1)2
Here we observe that
f ( 0) = 1 = ( 0 + 1)2
f (1) = 22 = (1 + 1)2
f ( 2) = 32 = ( 2 + 1)2
f ( 3) = 42 = ( 3 + 1)2
The above observation suggests that f ( x ) = ( x + 1)2
Let us verify whether f ( x ) = ( x + 1)2 satisfies the given condition
f ( x + y + 1) = ( f ( x ) + f ( y ))2 or not.
We have f (x ) + f (y ) = x + 1 + y + 1 = x + y + 1 + 1
⇒ ( f ( x ) + f ( y ))2 = ( x + y + 1 + 1)2 = f ( x + y + 1)
Hence, f ( x ) = ( x + 1)2
www.pdfworld.in
www.pdfworld.in
Functions 587
Example 2 If mf ( x ) + nf   = x − 5, m ≠ n and x ≠ 0, find f ( x )

1
x
 1
Solution We have, mf ( x ) + nf   = x − 5, m ≠ n and x ≠ 0, x ∈ R …(i)
x
 1 1
⇒ nf ( x ) + mf   = − 5 …(ii)
x x
On solving Eqs. (i) and (ii) by cross-multiplication, we get
 1
f 
f (x ) x 1
= =
1  1  m2 − n 2
− n  − 5 + m( x − 5) − n( x − 5) + m  − 5
x  x 
n
mx − + 5(n − m )
⇒ f (x ) = x
m2 − n 2
n
mx −
5
⇒ f ( x ) = 2 x2 −
m −n m+n
Example 3 If f ( x ) + 2f (1 − x ) = x 2, ∀ x ∈ R, find f ( x )
Solution We have, f ( x ) + 2f (1 − x ) = x 2, ∀ x ∈ R …(i)
Putting 1 − x, for x, we get
f (1 − x ) + 2f ( x ) = (1 − x )2 …(ii)
Thus, given equation reduces to
f ( x ) + 2(1 − x )2 − 4f ( x ) = x 2
⇒ 3f ( x ) = 2 + 2x 2 − 4x − x 2
⇒ 3f ( x ) = x 2 − 4x + 2
1
⇒ f ( x ) = [ x 2 − 4x + 2]
3
Algebra of Real Functions

(i) Addition of two real functions Let f : X → R and g : X → R be any two real functions, where
X ⊂ R. Then, we define (f + g ); X → R by
( f + g ) (x ) = f (x ) + g (x ), for all x ∈ X
(ii) Subtraction of a real function from another Let f : X → R and g : X → R be any two real
functions, where X ⊂ R. Then, we define ( f − g ) ; X → R by ( f − g ) (x ) = f (x ) − g (x ), for all x ∈ X .
(iii) Multiplication by a scalar Let f : X → R be a real valued function and α be a scalar. Here, by scalar,
we mean a real number. Then the product αf is a function from X to R defined by
(α f ) (x ) = α f (x ), x ∈ X .
(iv) Multiplication of two real functions The product (or multiplication) of two real functions
f : X → R and g : X → R is a function fg : X → R defined by (fg )(x ) = f (x )g (x ), ∀x ∈ X . This is also
called pointwise multiplication.
www.pdfworld.in
www.pdfworld.in
(v) Quotient of two real functions Let f and g be two real functions defined from X → R where
f f  f (x )
X ⊂ R. The quotient of f by g denoted by is a function defined by,   (x ) = , provided
g g  g (x )
g (x ) ≠ 0, x ∈ X
Example 1 Let f ( x ) = x 2 and g ( x ) = 2x + 1 be two real functions. Find

f 
(f + g ) ( x ), (f − g ) ( x ), (fg ) ( x ),   ( x ).
g
Solution We have,
(f + g ) ( x ) = x 2 + 2x + 1,
(f − g ) ( x ) = x 2 − 2x − 1
(fg ) ( x ) = x 2( 2x + 1) = 2x 3 + x 2,
f  x2 1
  (x ) = ,x ≠−
g 2x + 1 2
Example 2 Let f ( x ) = x and g ( x ) = x be two functions defined over the set of non-negative
f 
real numbers. Find (f + g ) ( x ), (f − g ) ( x ), (fg ) ( x ) and   ( x ).
g
Solution We have,
(f + g ) ( x ) = x + x , (f − g ) ( x ) = x − x ,
3 1
f  x −
(fg ) x = x ( x ) = x 2 and   ( x ) = = x 2, x ≠ 0
g x
Types of Functions
(i) Identity function The function f defined by f (x ) = x for ∀ x ∈ R is called an identity function of R.
For identity function Df = R, Rf = R.
(ii) Constant function The function f defined by f (x ) = c for all x ∈ R, where c is some real number is
called a constant function, Df = R , Rf = {c }.
1
(iii) Reciprocal function The function f defined by f (x ) = is called reciprocal function.
x
Df = R − {0}, Df = R − {0}.
(iv) Modulus function (or absolute value function) The function f defined by
 x where x ≥ 0
f (x ) = x = 
−x where x < 0
is called the modulus or absolute value function
Df = R ; Rf = [0, ∞ )
Properties of Modulus Function
(a) x ≤ a ⇒ − a ≤ x ≤ a ; (a ≥ 0)
(b) x ≥ a ⇒ x ≤ − a or x ≥ a ; (a ≥ 0)
(c) x + y = x + y ⇔ x ≥ 0 and y ≥ 0 or x ≤ 0 and y ≤ 0
(d) x − y = x − y ⇒ x ≥ 0 and x ≥ y or x ≤ 0 and y ≤ 0 and x ≥ y
(e) x ± y ≤ x + y
(f) x ± y ≥ x − y
www.pdfworld.in
www.pdfworld.in
Functions 589
(v) Greatest Integer or Integral Part of a Real Number : For every x ∈ R, [x ] is the greatest integer ≤ x , i.e.,
[x ] = x if x is and integer and [x] equal to integer immediately to the left of x , if x is not an integer.
Greatest Integer Function The function f defined by f (x ) = [x ] for all x ∈ R is called the greatest integer
function.
Properties of Greatest Integer Function
(a) If f (x ) = [x + n ], where n ∈ I and [ . ] denotes the greatest integer function, then f (x ) = n + [x ]
(b) x = [x ] + {n}, [ . ] and {x } denote the integral and fractional part of x respectively, then x − 1 < [x ] ≤ x
[− x ] = − [x ] , if x ∈ I
[− x ] = − [x ] − 1, if x ∉ I
[x ] − [− x ] = 2n , if x < n , n ∈ I
[x ] − [x ] = 2n + 1, if x = n + {x }, x ∈ I
[x ] ≥ n ⇒ x ≥ n , n ∈ I
[x ] ≤ n ⇒ x < n + 1, n ∈ I
[x ] > n ⇒ x ≥ n + 1, n ∈ I
[x ] < n ⇒ x < n , n ∈ I
(c) Df = R , Rf = I
(vi) Signum function The function f defined by
 x
f (x ) =  x , x ≠ 0
 0, x = 0
 1, x > 0
f (x ) =  0, x = 0
−1, x < 0

is called the signum function
Df = R ; Rf = {− 1, 0, 1}.
(vii) Logarithmic function The function f defined by f (x ) = log a x , where x > 0, a > 0, x , a ∈ R is called
the logarithmic function
D f = (0, ∞ ); R f = R
(viii) Exponential function The function f defined f (x ) = a x , when a > 0 and x ∈ R is called exponential
function
Df = R ; Rf = (0, ∞ )
(ix) Square root function The function f defined by f (x ) = x is called the square root function
Df = [0, ∞ ); Rf = [0, ∞ )
(x) Polynomial function The function f defined by f (x ) = a 0 + a1x + a 2x 2 + . . . + an x x , an ≠ 0, where
a 0 , a1 , a 2 , . . . , an are real numbers and n ∈ N is called a polynomial function of degree n.
Df = R ; Rf = R
g (x )
(xi) Rational function The function f defined by f (x ) = , where g (x ) and h (x ) are polynomial
h (x )
functions and h (x ) ≠ 0 is called a rational function.
The domain of f is the set of all real numbers except those values of x for which h (x ) = 0
(xii) Odd and even functions
(a) A function is an odd function if f (−x ) = − f (x ) for all x .
(b) A function is an even function if f (−x ) = f (x ) for all x .
www.pdfworld.in
www.pdfworld.in
(xiii) Explicit and implicit functions A function which is expressed directly in terms of independent
variable is called an explicit function.
If a function is not expressed directly in terms of independent variable, it is called an implicit
function.
(xiv) Single and many valued functions A function y = f (x )is said to be single valued function, if there
is one and only one value of y corresponding to each value of x .
A function y = f (x ) is said to be many valued, if there are more than one value of y corresponding to
each value of x .
(xv) Periodic functions A function f (x ) is said to be a periodic function, if there exists a positive real
function T such that f (x + T ) = f (x ) for all x ∈ R. If T is the smallest positive real number such that
f (x + T ) = f (x ) for all x ∈ R, then T is called the period of f (x ).
If f (x ) is a periodic function with periodT and a , b ∈ R such that a > 0, then f (ax + b ) is periodic with
T
period .
a
Example 1 If 2f ( x ) + 3f   = x 2 − 1, then prove that f ( x ) is an even function.

1
  x
 1
Solution Given, 2f ( x ) + 3f   = x 2 − 1 …(i)
x
1
Replacing x by , we get
x
 1 1
2f   + 3f ( x ) = 2 − 1 …(ii)
x x
Solving Eqs. (i) and (ii), we have
3 − 2x 4 − x 2
f (x ) =
5x 2
Clearly, f ( − x ) = f ( x )
Example 2 Prove that,

f ( x ) = log ( x + x 2 + 1)
is an odd function.
Solution f ( x ) = log ( x + x 2 + 1)
⇒ f ( − x ) = log ( − x + x 2 + 1)
⇒ f ( x ) + f ( − x ) = log ( x + x 2 + 1) + log ( − x + x 2 + 1) = log1 = 0
⇒ f ( −x ) = − f ( x )
Hence, f ( x ) is an odd function.
Example 3 If f : R → R is a function satisfying f ( 2x + 3) + f ( 2x + 7) = 2 ∀ x ∈ R, then prove that

f ( x ) is a periodic function.
Solution Given, f ( 2x + 3) + f ( 2x + 7) = 2 …(i)
Replacing x by (x + 1)
f ( 2x + 5) + f ( 2x + 9) = 2 …(ii)
Now replace x by (x + 2)
f ( 2x + 7) + f ( 2x + 11) = 2 …(iii)
www.pdfworld.in
www.pdfworld.in
Functions 591
Subtracting Eq. (iii) from Eq. (i), we get

f ( 2x + 3) − f ( 2x + 11) = 0
⇒ f ( 2x + 3) = f ( 2x + 11)
⇒ f ( x ) is a periodic function with period 4.
(xvi) Composite function Let f : A → R and g : B → R, then the composite map of f and g denoted by
gof is the map from A to R defined by (gof )(x ) = g (f (x )).
Example f : R → R defined by f (x ) = 2x
g : R → R defined by g (x ) = x 2
Then, gof : R → R defined by
(gof )(x ) = g (f (x )) = g (2x ) = (2x )2 = 4x 2
Note gof may not be equal to fog in the above example

(fog )( x ) = f (g ( x )) = f ( x 2 ) = 2x 2
1
Example 1 If f ( x ) = x 2 + 1, g ( x ) = , then find (fog )( x ) and (gof )( x ).
x −1
Solution Given, f (x ) = x 2 + 1 …(i)

1
g( x ) = …(ii)
x −1
 1 
Now, (fog )( x ) = f (g ( x )) = f   = f ( z ),
 x − 1
1
where z = = z2 +1 [Qf ( x ) = x 2 + 1]
x −1
2
 1  1
=  +1= +1
 x − 1 ( x − 1)2
(gof )( x ) = g (f ( x )) = g ( x 2 + 1) = g (u ), where u = x 2 + 1
1 1 1
= = =
u −1 x 2 + 1 −1 x 2
Example 2 If f ( x ) = ( 2 − x n )1/n , n > 0, then show that for x > 0

  1
f (f ( x )) + f f    ≥ 2
  x 
Solution Given, f ( x ) = ( 2 − x n )1/n , x > 0

Now, f (f ( x )) = f ( y ), where y = f ( x )
= ( 2 − y n )1/n , where y = ( 2 − x n )1/n
[ 2 − ( 2 − x n )]1/n = ( x n )1/n = x , x > 0
  1 1
f (f ( x )) + f f    = x +
  x  x
2
 1 
= x −  + 2≥2
 x
www.pdfworld.in
www.pdfworld.in
Functional Equations
Equation which involves an unknown function is called a functional equation.
Example 1 If f ( x ) satisfies the relation f ( x + y ) = f ( x ) + f ( y ) for all x , y ∈ R and f (1) = 5, then find
m
Σ f (n ). Also prove that f ( x ) is an odd function.
n =1
Solution f (x + y ) = f (x ) + f (y )
Now, f (r ) = f (r − 1 + 1) = f (r − 1) + f (1)
= f (r − 1) + 5 = {f (r − 2) + 5} + 5
= f (r − 2) + 2 ⋅ 5 = {f (r − 3) + 5} + 2 ⋅ 5
= f (r − 3) + 3 ⋅ 5
………
………
………
………
= f [r − (r − 1)] + (r − 1) ⋅ 5
= f (1) + 5(r − 1) = 5 + 5r − 5 = 5r
m m m(m + 1)
∴ Σ f (n ) = Σ 5x = 5
n =1 n =1 2
Now, putting x = y = 0 in given relation we, get
f ( 0 + 0) = f ( 0) + f ( 0) ⇒ f ( 0) = 0
By putting y = − x , we get
f ( 0) = f ( x ) + f ( −x )
⇒ f ( x ) + f ( −x ) = 0
⇒ f ( −x ) = − f ( x )
∴f ( x ) is an odd function.
Example 2 f ( x ) = x 4 + ax 3 + bx 2 + cx + d , where a, b, c, d ∈ R . If f (1) = 10 , f ( 2) = 20 and

f ( 3) = 30, find the value of f (12) + f ( − 8).
Solution Given, f ( x ) = x 4 + ax 3 + bx 2 + cx + d …(i)
Given, f (1) = 10
⇒ 1 + a + b + c + d = 10
⇒ a+b+c+d =9 …(ii)
f ( 2) = 20
⇒ 8a + 4b + 2c + d = 4 …(iii)
f ( 3) = 30
⇒ 27a + 9b + 3c + d = − 51 …(iv)
Let y = f (12) + f ( − 8)
Then, y = 124 + 84 + a(123 − 83 ) + b(122 + 82 ) + c(12 − 8) + 2d
= 44( 81 + 16) + 4 ⋅ 304a + 208b + 4c + 2d
= 256 ⋅ 97 + 1216a + 208b + 4c + 2d …(v)
www.pdfworld.in
www.pdfworld.in
Functions 593
On adding Eqs. (ii) and (iv), we get

28a + 10b + 4c + 2d = − 42 …(vi)
On multiplying by 2 in Eq. (iii) and subtracting from Eq. (vi)
12a + 2b = − 50
⇒ 6a + b = − 25 …(vii)
From Eq. (v), y = 256 ⋅ 97 + 1200a + 200b + 16a + 8b + 4c + 2d
= 256 ⋅ 97 + 200( 6a + b ) + 2( 8a + 4b + 2c + d )
= 256 ⋅ 97 + 200 ( −25) + 2 ⋅ 4 [from Eq. (iii)]
= 24832 − 5000 + 8 = 19840
f (12) + f ( − 8)
∴ = 1984
10
Example 3 If f ( x ) is a polynomial function such that

 1  1
f ( x ) + f   = f ( x ) ⋅ f   , show that f ( x ) = 1 ± x n .
x x
Solution Let f ( x ) = a 0 + a1x + a 2x 2 + . . . an x n , an ≠ 0 …(i)

 1 a a a
Then, f   = a 0 + 1 + 22 + . . . + nn
x x x x
1 n −1
= n (a 0x + a1xn
+ a 2x n − 2 + .. . .+ an )
x
 1  1
Given, f (x ) + f   = f (x ) ⋅ f  
x x
1
⇒ a 0x + a1x + a 2x 2 + . . .an − 1x n − 1 + an x n + n (a 0x n + a1x n −1 + . . .+ an )
x
1
= (a 0 + a1 x + a 2 x 2 + an x n ) n (a 0x n + a1x n −1 + . . .+ an )
x
⇒ a 0x n + a1x n + 1 + a 2x n + 2+ ...+ an − 1x 2n − 1 + an x 2n + a 0x n + a1x n − 1 + . . . + an
= (a 0 + a1x + a 2x 2 + . . . an x n ) (a 0x n + a1x n − 1 + . . .+ an )
Equating the coefficient of x 2n , we get
an = a 0an ⇒ a 0 = 1 [Qan ≠ 0]
Equating the coefficient of x 2n − 1, we get
an −1 = a 0 an −1 + an a1
⇒ ana1 = 0 [Qa 0 = 1]
⇒ a1 = 0 [Qan ≠ 0]
Similarly, equating the coefficients of other powers of x, we get
a 2 = a 3 = . . . = an − 1 = 0
Equating the coefficient of x n , we get
2a 0 = a 02 + an2
⇒ an2 = 1 [Qa 0 = 1]
⇒ an = ± 1
Thus, f ( x ) = a 0 + a1x + a 2x 2 + . . . an x n = 1 ± x n
www.pdfworld.in
www.pdfworld.in
4x 2001  r 
Example 4 If f ( x ) = , find Σ f  
4 +2
x r = 1  2002
4x
Solution Given, f (x ) = …(i)
4x + 2
41 − x 4 2
∴ f (1) − x ) = 1− x
= = …(ii)
4 + 2 4 + 2⋅ 4 x
2 + 4x
Clearly, f ( x ) + f (1 − x ) = 1
2001  r   1   2   3   2001
Now, Σ f  =f  + f  +f  + ... + f  
r = 1  2002  2002  2002  2002  2002
  1   2001    2   2000 
= f   + f  + f   + f 
  2002  2002   2002  2002 
  1000   1002   1001  
+ . . . + f   + f  + f 
  2002   2002   2002 
 1
= 1 + 1 + 1 + . . . to 1000 terms + f  
 2
4 1
= 1000 + = 1000 + = 1000.5
4+2 2
100
Example 5 If 2f ( xy ) = {f ( x )} y + {f ( y )} x for all x , y , ∈ R , and f (1) = 2, then find the value of Σ f (r ).
r =1
Solution Given, 2f ( xy ) = {f ( x )} + {f ( y )} for all x , y , ∈ R

y x
…(i)
Putting y = 1, we get 2f ( x ) = f ( x ) + {f (1)} x
∴ f ( x ) = 2x ∴f (r ) = 2r
100 100 2( 2100 − 1)
Now, Σ f (r ) = Σ 2r = 2 + 22 + 23 + . . . + 2100 = = 2201 − 2
r =1 r =1 2 −1
1− x
Example 6 If a function satisfies 2f ( x − 1) − f   = x ; ( x ≠ 0), then determine f ( x ).
 x 
1 − x 
Solution We have, 2f ( x − 1) − f   = x, x ≠ 0
 x 
1 
⇒ 2f ( x − 1) − f  − 1 = x …(i)
x 
1 1  1
Replacing x by , we get 2f  − 1 − f ( x − 1) = …(ii)
x x  x
1
Now, the operation 2x(i) + (ii) gives 4f ( x − 1) − f ( x − 1) = 2x +
x
2x 2 + 1
⇒ 3f ( x − 1) =
x
2( x − 1 + 1)2 + 1
⇒ 3f ( x − 1) =
x −1 + 1
2( x + 1)2 + 1 2x 2 + 4x + 3
⇒ f (x ) = =
3( x + 1) 3( x + 1)
www.pdfworld.in
www.pdfworld.in

n
Example 1. Let f be a function satisfying, f (x + y ) = f (x ) f (y ) ∀ x , y ∈ R. If f (1) = 3, find, the value of Σ f (r )
r =1
Solution Q f (x + y ) = f (x ) f (y )
∴ f (x ) = a λx , where λ is constant and f (1) = a λ = 3
n n
Σ (r ) = Σ a λr = a λ + a 2λ + a3λ + . . . + anλ
r =1 r =1
a λ (anλ − 1) 3(3n − 1) 3 n
= λ
= = (3 − 1)
a −1 3−1 2
Example 2. Determine all functions f : R → R such that

f ( x − f ( y )) = f ( f ( y )) + xf ( y ) + f (x ) − 1, ∀ x , y ∈ R,
Solution f ( x − f ( y )) = f ( f ( y )) + xf ( y ) + f (x ) − 1 …(i)
Put x = f ( y ) = 0
Then, f (0) = f (0) + 0 + f (0) − 1
∴ f (0) = 1 …(ii)
Again put x = f ( y ) = λ in Eq. (i), then f (0) = f (λ ) + λ + f (λ ) − 1 2
⇒ 1 = 2f (λ ) + λ2 − 1
2 − λ2 λ2
∴ f (λ ) = =1−
2 2
x2
Hence, f ( x ) = 1 − is the unique function.
2
Example 3. Let n be a positive integer and define f (n ) = 1 ! + 2 ! + 3 ! + . . . + n ! where

n ! = n (n − 1) (n − 2) . . . 3 ⋅ 2 ⋅ 1 find polynomials P (x ), Q (x ) such that
f (n + 2) = P (n ) f (n + 1) + Q (n ) f (n ) ∀ n ≥ 1.
Solution Q f (n ) = 1 ! + 2 ! + 3 ! + . . . + n ! …(i)
∴ f (n + 1) = 1 ! + 2 ! + . . . + n ! + (n + 1)! …(ii)
Subtract Eq. (i) from Eq. (ii), then
f (n + 1) − f (n ) = (n + 1)! …(iii)
Now, f (n + 2) = 1 ! + 2 ! + . . . + (n + 1)! + (n + 2)!
= f (n + 1) + (n + 2)(n + 1)!
= f (n + 1) + (n + 2){ f (n + 1) − f (n )}
∴ f (n + 2) = (n + 3)f (n + 1) + (−n − 2)f (n ) …(iv)
But given
f (n + 2) = P (n )f (n + 1) + Q (n )f (n ) …(v)
Comparing Eqs. (iv) and (v), we get
P (n ) = n + 3, Q (n ) = − n − 2
∴ P (x ) = x + 3, Q (x ) = − x − 2
www.pdfworld.in
www.pdfworld.in
Example 4. A function f defined ∀ x , y ∈ R is such that f (1) = 2; f (2) = 8 and f (x + y ) − kxy = f (x ) + 2y 2

 1 
where k is constant. Find f (x ) and show that f (x + y ) f   = k for x + y ≠ 0.
x + y 
Solution f (x + y ) − kxy = f (x ) + 2y 2
Replace y by – x, we get
f (0) + kx 2 = f (x ) + 2x 2
⇒ f (x ) = f (0) + kx 2 − 2x 2 ...(i)
∴ f (1) = f (0) + k − 2 = 2 ...(ii)
and f (2) = f (0) + 4k − 8 = 8 ...(iii)
Solve Eqs. (ii) and (iii), we get
k = 4 and f (0) = 0 ...(iv)
∴from Eq. (i), we get f (x ) = 2x 2
2
 1   1 
Also f (x + y ) ⋅ f   = 2 (x + y ) ⋅ 2 
2

x + y  x + y 
= 4=k [from Eq. (iv)]
Example 5. Consider a real valued function f (x ) satisfying

2f ( xy ) = ( f ( x ))y + ( f ( y ))x ∀ x , y ∈ R
and f (1) = a , where a ≠ 1. Show that
+1
(a − 1) Σ f (i ) = an −a
i = 1n
Solution 2f (xy ) = ( f ( x ))y + ( f ( y ))x

Replace y by 1, we get
2f (x ) = f (x ) + ( f (1))x
⇒ f (x ) = (f (1))x = a x [Qf (1) = a ]
n n
∴ LHS = (a − 1) Σ f (i ) = (a − 1) Σ ai
i =1 i =1
= (a − 1) (a + a 2 + a3 + K + an )
a (an − 1)
= (a − 1)   (If a > 1)
 (a − 1) 
+1
= a (an − 1) = (an − a ) = RHS
Example 6. If p and q are +ve integers, f is a function defined for +ve numbers and attains only positive
values such that f (xf (y )) = x py q .
Prove that q = p 2.
Solution f (xf (y )) = x py q
or [f ( xf ( y ))]1/p = xy q /p
[ f ( xf ( y ))]1/p
or x = ...(i)
y q /p
www.pdfworld.in
www.pdfworld.in
Functions 597
1
Let xf ( y ) = 1 ⇒ x = ...(ii)
f (y)
y q /p
from Eq. (i) f (y) =
{f (1)}1/p
1
∴ f (1) = ⇒ f (1) = 1
{f (1)}1/p
Then, f ( y ) = y q /p ...(iii)
Hence, f (xy q /p
)=x y p q
put y q /p
=z
Then, f (xz ) = (xz )p
⇒ f (λ ) = λp ...(iv)
from Eqs. (iii) and (iv), we get
y p = y q /p ⇒ p = q / p
Hence, p2 = q
f (x )
Example 7. If f (x + y ) = f (x ) . f ( y )∀ real x , y and f (0) ≠ 0, then prove that the function F (x ) =
1 + [f (x )]2
is an even function.
Solution f (x + y ) = f (x ) ⋅ f (y ) ...(i)
Put x = y = 0, then (f (0)) = f (0) 2
⇒ f (0) {f (0) − 1} = 0 ∴ f (0) = 1

Putting y = − x in Eq. (i), then
f (0) = f (x ) ⋅ f (−x )
⇒ 1 = f (x ) ⋅ f (− x )
1
∴ f (− x ) = ...(ii)
f (x )
f (x )
Also, F (x ) =
1 + [f (x )]2
1
f (− x ) f (x ) f (x )
∴ F (− x ) = = = = F (x )
1 + ( f (− x ))2 1 1 + ( f (x ))2
1+
( f (x ))2
Hence, F (x ) is an even function.
Example 8. If the function f satisfies the relation f (x + y ) + f (x − y ) = 2f (x ) ⋅ f ( y ) ∀ x , y ∈ R and f (0) ≠ 0.

Prove that f (x ) is an even function.
Solution f (x + y ) + f (x − y ) = 2f (x ) ⋅ f (y ) ...(i)
Replace x by y and y by x in Eq. (i)
Then, f ( y + x ) + f ( y − x ) = 2f ( y ) f (x ) ...(ii)
∴From Eqs. (i) and (ii),we get
f ( y − x ) = f (x − y )
www.pdfworld.in
www.pdfworld.in
Put y = 2x , then f (x ) = f (−x )

Hence, f (x ) is an even function.
n
Example 9. If f (x )satisfied the relation f (x + y ) = f (x ) + f (y ) ∀x , y ∈ R and f (1) = 7, find value of Σ f (i ).
i =1
Solution f (x + y ) = f (x ) + f ( y )
Put x = y = 1, then f (2) = 2f (1)
Put x = 1, y = 2, then
f (3) = f (1) + f (2) = 3f (1)
Similarly, f (i ) = i f (1) = 7i [Qf (1) = 7]
n n n 7n (n + 1)
∴ Σ f (i ) = Σ 7i = 7 Σ i =
i =1 i =1 i =1 2
Also f (x + y ) = f (x ) + f (y )
Put y = −x
then f (0) = f (x ) + f (−x ) [for x = y = 0 f (0) = 0]
∴ 0 = f (x ) + f (−x )
∴ f (−x ) = − f (x )
Hence, f (x ) is an odd function.
Example 10. If f (a − x ) = f (a + x ) and f (b − x ) = f (b + x ) ∀ real x where a , b (a > b ) are constants, then

prove that f (x ) is a periodic function.
Solution f (b + x ) = f (b − x ) (given)
Replace x by x + b
Then, f (x + 2b ) = f (b − (x + b )) = f (−x )
Replace x by – x, then
f (x ) = f (2b − x ) = f (2b + x )
[Qf (a − x ) = f (a + x ) and a > b, Here, a = 2b ]
⇒ f (2b + x ) = f (x )
Hence, f (x ) is periodic with period 2b.
Example 11. If f be a function defined on the set of non-negative integers and taking values in the same
set. Given that,
 f (x )
(i) x − f (x ) = 19   − 90 
x
 19   ∀ non-negative integers.
 90 
(ii) 1900 < f (1990) < 2000
Find the possible values of f (1990) can take.
Solution Q 1900 < f (1990) < 2000
1900 f (1990) 2000
∴ < <
90 90 90
 1900  <  f (1990)  2000 
⇒  90  <  90 
 90   
 f (1990)
⇒ 21 <   < 22
 90 
{Q[21.111] = 21 and [22.222] = 22}
www.pdfworld.in
www.pdfworld.in
Functions 599
Case I Suppose f (1990) = 21, then substitute x = 1990 in Eq. (i)
 90 
 f (1990)
1990 − f (1990) = 19 
1990 
− 90  
 19   90 
⇒ 1990 − f (1990) = 19104
. − 9021
.
⇒ f (1990) = 1904
 f (1990)
Case II Suppose   = 22
 90 
then substitute x = 1990 in Eq. (i)
 f (1990)
1990 − f (1990) = 19 
1990 
− 90  
 19   90 
⇒ 1990 − f (1990) = 19104
. − 9022
.
⇒ f (1990) = 1994
Both the value satisfying given conditions.
∴f (1990) can take values 1904 and 1994.
Example 12. Let f be a real valued function with domain R. Now if for some +ve constant a, the equation.
f (x + a ) = 1 + {2 − 3f (x ) + 3 (f (x ))2 − (f (x ))3}1/3 holds good for x ∈ R. Prove that f (x ) is a periodic function
Solution f (x + a ) = 1 + [2 − 3f (x ) + 3 (f (x ))2 − (f (x ))3 ]1/3

= 1 + {1 + (1 − f (x ))3}1/3 ...(i)
Replace x by (x + a ), then
f (x + 2a ) = 1 + {1 + (1 − f (x + a ))3}1/3
= 1 + {1 − (f (x + a ) − 1)3}1/3
= 1 + {1 − (1 + (1 − f (x ))3 )}1/3
= 1 − {(1 − f (x ))3}1/3
= 1 − (1 − f (x )) = f (x )
Hence, f (x ) is periodic with period 2a.
Example 13. Determine all functions f satisfying the functional relation

 1  2 (1 − 2x )
f (x ) + f   =
 1 − x  x (1 − x )
where x is a real number x ≠ 0, x ≠ 1.
 1  2 (1 − 2x ) 2 2
Solution f (x ) + f   = = −
1 − x  x (1 − x ) x 1 − x
1 1
Let y = ⇒x = 1 − …(i)
1−x y
2
from Eq. (i), f (x ) + f ( y ) = − 2y …(ii)
x
 1  2 2
from Eq. (i), f (y) + f   = − …(iii)
1 − y  y 1 − y
1 1
Let z= ⇒ y =1−
1−y z
www.pdfworld.in
www.pdfworld.in
2
from Eq. (iii) we have f (y ) + f (z ) = − 2z …(iv)
y
whenever y , z ≠ 0 or 1.
1 1
y = and z =
1−x 1−y
1
Now, x =
1−z
1
⇒ z =1−
x
2
Similarly, f (x ) + f (z ) = − 2x …(v)
z
Adding corresponding sides of Eqs. (ii), (iv) and (v) and divide throughout by 2, we get
 1 1 1
f (x ) + f (y ) + f (z ) =  + +  − (x + y + z )
x y z
2  1 1 1
⇒ f (x ) + − 2z =  + +  − (x + y + z )
y x y z
 1 1 1
⇒ f (x ) =  − +  − (x + y − z )
x y z
 1
=  − x  −  y +  +  z + 
1 1
x   y  z
 1   x 
=  − x  − 
1 1
+ 1 − x  + 1 − + 
x  1 − x   x x + 1
1 x
=− +
1−x x −1
 x + 1
=  for real values of x except x = 0 or 1.
 x − 1
Aliter
 1  2 (1 − 2x ) 2 2
f (x ) + f   = = − ...(i)
 1 − x  x (1 − x ) x 1−x
1
Replace x by , we get
1−x
 
 1   
1 2
f   +f   = 2 (1 − x ) −
1 − x  1 − 1  1−
1
 1−x  1−x
 
 1   1  1
⇒ f   + f  1 −  = 2 (1 − x ) − 2  1 − 
 1 − x   x   x 
2
= − 2x + ...(ii)
x
1
Replace x by 1 − in Eq. (i), we get
x
www.pdfworld.in
www.pdfworld.in
Functions 601
 
 
 1
f 1 −  + f 
1  = 2
−
2
 x 1 − 1 − 1   1 − 1  1 − 1 − 1 
      
  x  x  x
f  1 −  + f (x ) =
1 2x
⇒ − 2x ...(iii)
 x x −1
Subtract Eq. (ii) from Eq. (i), we get
f (x ) − f  1 −  = 2x −
1 2
...(iv)
 x 1−x
Adding Eqs. (iii) and (iv), we get
2x 2
2f (x ) = −
x −1 1−x
x +1
f (x ) =
x −1
Example 14. If f : R → R is a function satisfying the properties (i) f (−x ) = − f (x )
(ii) f (x + 1) = f (x ) + 1 (iii) f   = 2 , x ≠ 0. Prove that f (x ) = x ∀ x ∈ R.

1 f (x )
x  x
Solution Let us observe that if f (x )is known ∀ x > 0, then f (x )can be found ∀ x < 0 by using (i)
…(i)
By putting x = 0 in (1), we find that f (− 0) = − f (0) i . e. , f (0) = 0 ...(ii)
from Eqs. (i) and (ii) we find that it enough to find f (x ) ∀ x > 0.
 1 
We shall take x > 0 and compute f   in terms of f (x ) in two different ways. By equating
 x + 1
 1 
the two expressions for f   thus obtained, we shall find f (x ) ...(iii)
 x + 1
 1   f (x + 1)
Now, f   =  [from Eq. (iii)]
 x + 1   (x + 1)2 
 f (x ) + 1 
= 2 
[by Eq. (ii)] ...(A)
 (x + 1) 
 1   x   −x 
Again, f   =f 1 −  =f   +1 [by Eq. (i)]
 x + 1  x + 1  x + 1
 x 
= −f +1 [by Eq. (i)]
 x + 1
  x + 1 
f  
 x 
=− +1 [by Eq. (iii)]
  x + 1 
2
  
 x  
−x 2 −x 2  f (x ) 
f  1 +  + 1 =
1
= + 1 + 1 [by Eq. (iii)] ...(B)
(x + 1) 2  x (x + 1)2  x 2 
www.pdfworld.in
www.pdfworld.in
from Eqs. (A) and (B), we have

f (x ) + 1 x2  f (x ) 
=− + 1 + 1
(x + 1) 2
(x + 1)2  x 2 
2f (x ) −x2 1
= +1−
(x + 1) 2
(x + 1)2
(x + 1)2
⇒ f (x ) = x
Thus we find that f (x ) = x ∀x > 0
Let x<0 ...(iv)
Put x = − y so that y > 0
We have f (x ) = f (−y ) = − f ( y ) [by Eq. (i)]
= −y [by Eq. (iii)]
=x
QWe have already seen that f (x ) = x when x = 0, it follows that f (x ) = x ∀ x ∈ R.
Example 15. Let f (x , y ) be a periodic function satisfying f (x , y ) = f (2x + 2y , 2y − 2x ) ∀ x , y . Define

g (x ) = f (2x , 0). Show that g (x ) is a periodic function with period 12.
Solution f (x , y ) = f (2x + 2y , 2y − 2x ) ...(i)
or f (I , II ) = f (2I + 2II , 2II − 2I )

∴ f (2x + 2y , 2y − 2x )
= f (2 (2x + 2y ) + 2 (2y − 2x ), 2 (2y − 2x ) − (2x + 2y ))
= f (8y , − 8x ) ...(ii)
f (x , y ) = f (8y , − 8x ) ...(iii)
or f (I , II ) = f (8II , − 8I )
∴ f (8y , − 8x ) = f (8 (− 8x ), − 8 (8y ))
= f (− 64x , − 64y ) ...(iv)
from Eqs. (iii) and (iv), we get,
f (x , y ) = f (− 64x , − 64y ) ...(v)
or f (I , II ) = f (− 64I , − 64II )
∴ f (− 64x , − 64y ) = f (− 64 (− 64x ), − 64 (− 64y ))
= f (212x , 212y ) ...(vi)
f (x , y ) = f (2 x , 2 y )
12 12
[from Eqs. (v) and (vi)]
⇒ f (x , 0) = f (2 x , 0)
12
Replace x by 2x , then f (2x , 0) = f (2x + 12 , 0)
⇒ g (x ) = g (x + 12) [Qg (x ) = f (2x , 0)]

Hence, g (x ) is periodic with period 12.
www.pdfworld.in
www.pdfworld.in
Functions 603
Example 16. If for all real values of u and v, 2f (u ) cos v = f (u + v ) + f (u − v )

Prove that ∀ real values of x.
(i) f (x ) + f (−x ) = 2a cos x (ii) f (π − x ) + f (−x )
(iii) f (π − x ) + f (x ) = 2b sin x .
Deduce that f (x ) = a cos x + b sin x where, a , b are arbitrary constant.
Solution 2f (u ) cos v = f (u + v ) + f (u − v ) ...(i)
put u = 0 and v = x in Eq. (i)
we get, f (x ) + f (−x ) = 2f (0) cos x = 2a cos x ...(ii)
a is arbitrary constant.
π π
Put u = − x and v = in Eq. (i), we get
2 2
f (π − x ) + f (−x ) = 0 ...(iii)
Again put u = π / 2 and v = (π / 2) − x in Eq. (i), we get
f (π − x ) + f (x ) = 2f (π / 2) sin x = 2b sin x ...(iv)
b is arbitrary constant.
Adding Eqs. (ii) and (iv), then
2f (x ) + f (π − x ) + f (−x ) = 2a cos x + 2b sin x
2f (x ) + 0 = 2a cos x + 2b sin x
∴ f (x ) = a cos x + b sin x
Example 17. Let f and g be real valued functions such that f (x + y ) + f (x − y ) = 2f (x ) ⋅ g ( y ) ∀ x , y ∈ R. Prove
that if f is not identically zero and | f (x )| ≤ 1 ∀ x ∈ R, then | g (y )| ≤ 1 ∀ y ∈ R.
Solution 0 < | f (x )| ≤ 1 (given)
Let max | f (x )| = M ∀ 0 < M ≤ 1 ...(i)

∴ | f (x + y )| ≤ M and | f (x − y )| ≤ M
Then, | f (x + y ) | + | f (x − y )| ≤ 2M ...(ii)
2f (x ) g ( y ) = f (x + y ) + f (x − y )
⇒ |2f (x ) g (y )| = | f (x + y ) + f (x − y )|
≤ | f (x + y )| + | f (x − y )| ≤ 2M [from Eq. (ii)]
⇒ 2 | f (x )| | g ( y )| ≤ 2M
⇒ | f (x )| | g ( y )| ≤ M ≤ 1 [from Eq. (i)]
⇒ | f (x )| | g ( y )| ≤ 1
| f (x )| ≤ 1
∴ | g ( y )| ≤ 1 ∀ y ∈ R
Example 18. Let f be a real valued function defined ∀ real numbers, such that for some a > 0 it satisfies
1
f (x + a ) = + f (x ) − (f (x ))2
2
Prove that f is periodic i.e., there exists a +ve real b such that f (x + b ) = f (x ) holds for every x.
Also gives an example of such a non constant f for a = 1.
www.pdfworld.in
www.pdfworld.in
1
Solution f (x + a ) = + f (x ) − (f (x ))2 ...(i)
2
1
Note that ≤ f (x ) ≤ 1 ...(ii)
2
1
holds for all x Qf (x ) is the sum of and a non negative real and the expression under the square
2
root is non negative only, if f (x ) (1 − f (x )) ≥ 0
1
which according to f (x ) ≥ ⇒ f (x ) ≤ 1
2
First Solution
QEq. (i) provides a relation between f (x + a ) and f (x ) substitute x → x + a in Eq. (i)
1
f (x + 2a ) = + f (x + a ) − (f (x + a ))2
2
2
f (x ) − (f (x ))2 −  − f (x ) − (f (x ))2 
1 1 1
= + +
2 2 2 
1 1 1
= + + f (x ) − (f (x ))2 − − f (x ) + (f (x ))2
2 2 4
− f (x ) − (f (x ))2
1 1
= + − f (x ) + (f (x ))2
2 4
2
=
1
+  f (x ) − 1  = 1 + f (x ) − 1 = f (x )
 
2  2 2 2
It shows that b = 2a is a period of f.
Second Solution
Consider Eq. (i) as a quadratic equation with f (x ) as unknown.
Taking its square root, we get
2
= f (x ) − (f (x ))2 = f (x )2 − f (x ) +  f (x + a ) −  = 0
1 1
f (x + a ) −
2  2
Now use quadratic formula
1
+ f (x + a ) − (f (x + a ))2 ,
2
1
− f (x + a ) − (f (x + a ))2
2
1
Second root is less than hence the unique solution for f (x ) is
2
1
f (x ) = + f (x + a ) − (f (x + a ))2
2
Substitute x → x − a and we get
1
f (x − a ) = + f (x ) − (f (x ))2
2
According to Eq. (i) this shows that
f (x − a ) = f (x + a )
Hence, finally substitute x → x + a
⇒ f (x ) = f (x + 2a )
Consequently b = 2a is a period of f.
www.pdfworld.in
www.pdfworld.in
Functions 605
Example 19. The function f (x , y ) satisfies

f (0, y ) = y + 1, f (x + 1, 0) = f (x , 1)
f (x + 1, y + 1) = f (x , f (x + 1, y ))
for every integer x and y find f (4, 1981).
Solution Let y be an arbitrary +ve integer. We want to find the values f (1, y ), f (2, y ), f (3, y ).
f (1, 0) = f (0, 1) = 2
f (1, y + 1) = f (0, f (1, y )) = f (1, y ) + 1
= f (0, f (1, y − 1)) + 1 = f (1, y − 1) + 2
Thus, f (1, y ) = f (1, y − 1) + 1
Again,
f (1, y ) = f (1, 0) + y = f (0, 1) + y = y + 2
Now, we determine the value of f (2, y )
f (2, y ) = f (1, f (2, y − 1)) = f (2, y − 1) + 2
f (2, y ) = f (2, 0) + 2y = f (1, 1) + 2y = 2y + 3
f (3, y ) = f (2, f (3, y − 1)) = 2f (3, y − 1) + 3
f (3, y ) = 2 (2f (3, y − 2) + 3) + 3
= 22f (3, y − 2) + 3 + 2 . 3
−1
= 3 + 2.3 + 22. 3 + .... + 2y . 3 + 2y f (3, 0)
= 3 (2y − 1) + 2y f (2, 1)
= 3 (2y − 1) + 2y 5
+3
f (3, y ) = 2y −3
( 4 , y − 1) + 3
finally we determine value of f (4, y ) ⋅ f (4, y ) = f (3, f (4, y − 1)) = 2f −3
( 2f ( 4 , y − 2) + 3)
=2 −3
2f ( 4 , y − 2) + 3)
=2 −3
2 ... 2f ( 4 , 0) + 3
= f (4, y ) = 22 −3
and there are y many
f (4, 0) = f (3, 1)
2
= 24 − 3 = 22 − 3
2 .... 2
f (4, y ) = 22 −3
Example 20. The function f (x )is defined on the +ve integers and takes on non – ve integers values ∀ n , m
f (m + n ) − f (m ) − f (n ) = 0 or 1
f (2) = 0, f (3) > 0
f (9999) = 3333
find f (1982) .
Solution f (m + n ) − f (m ) − f (n ) = 0 or 1 ...(i)
f (2) = 0, f (3) > 0 ...(ii)
f (9999) = 3333 ...(iii)
Let m = n = 1, then from Eq. (i) it follows that f (2) − 2f (1) = 0 or Eq. (ii) implies that 2f (1) = 0 or − 1
www.pdfworld.in
www.pdfworld.in
Hence, f (1) = 0.
Similarly, if m = 2, n = 1
f (3) − f (2) − f (1) = f (3) = 0 or 1
and by Eq. (ii), we have f (3) = 1
Using induction we show that for every +ve integer k
f (3k ) ≥ k ...(iv)
Assume that Eq. (iv) holds until some k. Let m = 3k , n = 3 then Eq. (i) implies f (3 (k + 1))
= f (3k + 3) ≥ f (3k ) + f (3) ≥ k + 1
Hence, Eq. (iv) holds for every k. If in the last step f (3k ) > k, then f (3 (k + 1)) > k + 1 holds
i . e. , if for some k0 we have f (3k0 ) > k0 in Eq. (iv), then for every k > k0 , f (3k ) > k holds as well.
This observation implies that in case k < 3333, f (3k ) > k cannot hold as, then
f (33333
. ) > 3333 follows contradicting Eq. (iii)
∴ f (3.1982) = 1982.
Now Eq. (i) implies.
1982 = f (3.1982) = f (2.1982 + 1982)
≥ f (2.1982) + f (1982)
f (1982 + 1982) ≥ 2f (1982)
Combine last two results, we get 1982 ≥ 3f (1982),
1982
f (1982) ≤ < 661 ...(v)
3
Using Eq. (i) again f (1982) = f (1980 + 2)
≥ f (1980) + f (2) = f (3.660) + 0 = 660
from Eqs. (v), 660 ≤ f (1982) < 661
so f (1982) = 660
Example 21. Prove that there is no function f from the set of non – ve integers into itself such that
f ( f (n )) = n + 1987 ∀n.
Solution f ( f ( n )) = n + 1987 ...(i)
Substitute f (n ) for n in Eq. (i), we get
f ( f ( f ( n ))) = f ( n ) + 1987 ...(ii)
Again f ( f ( f ( n ))) = f ( n + 1987) ...(iii)
from Eqs. (ii) and (iii), we get
f ( n + 1987) = f ( n ) + 1987 ...(iv)
Let t be any arbitrary +ve integer let us now use mathematical induction.
f ( n + 1987 t ) = f ( n ) + 1987 t ...(v)
for t = 1 f ( n + 1987 ( t − 1)) = f ( n ) + 1987 ( t − 1)
Substitute here n + 1987 for n
Then from Eq. (iv)
f (n + 1987 t ) = f (n + 1987) + 1987 (t − 1) = f (n ) + 1987 + 1987 (t − 1) = f (n ) + 1987 t
which is equal to Eq. (v)
Let s be an arbitrary non – ve integer less than 1987 and consider the remainder r when f (s ) is
divided by 1987.
f (s ) = 1987 k + r (0 ≤ k and 0 ≤ r ≤ 1986)
By condition f (s ) is not – ve ...(vi)
⇒ f (f (s )) = s + 1987 [By Eq. (i)]
www.pdfworld.in
www.pdfworld.in
Functions 607
from Eqs. (v) and (vi)

f (f (s )) = f (1987 k + r ) = f (r ) + 1987 k
⇒ s + 1987 = f (r ) + 1987 k ...(vii)
Q s < 1987
f (r ) + 1987 k < 2.1987
f (r ) < 1987 (2 − k )
Now, f (r ) ≥ 0
either k = 1 or k = 0
from Eqs. (vi) and (vii) f (s ) = 1987 + r ...(viii)
f (r ) = s …(ix)
⇒ 1987 = 0 which is a contradiction.
In other case k = 0
Again from Eqs. (vi) and (vii) implies
f (s ) = r ...(x)
f (r ) = 1987 + s ...(xi)
So, r = s leads to contradiction.
Eqs. (ix) and (x) together imply when acting on the numbers 0, 1, ....,1986.
When f is arranging them into pairs (a , b ) in such a way that either
f (a ) = b and f (b ) = a + 1987
f (b ) = a and f (a ) = b + 1987
We observe that numbers in each pair are different. Which is a contradiction. Since, the number
of elements of the set 0, 1, .... , 1986 is odd.
Example 22. Find all function f defined on the non –ve reals and taking non –ve real values such that,
f (x ⋅ f ( y )) ⋅ f (y ) = f (x + y )
for every non – ve x and y f (2) = 0,
f (x ) ≠ 0, if 0 ≤ x < 2
Solution Let x ≥ 2 and y = 2 ...(A)
from f (x ⋅ f ( y )) ⋅ f ( y ) = f (x + y ), we have f ((x − 2) f (2)) ⋅ f (2) = f (x − 2 + 2) = f (x )
from f (2) = 0 …(B)
for every x ≥ 2 f (x ) = 0 ...(i)
and this holds only in case x ≥ 2
Now, let 0≤ y < 2
2
LHS of Eq. (A) is 0 if xf (y ) ≥ 2 i . e. , x≥ ...(ii)
f (y )
RHS equals to 0 if x + y ≥ 2
Thus x ≥2−y ...(iii)
⇒ Under the given conditions the equality
2
=2−y ...(iv)
f (y )
If Eq. (iv) does not hold for some y, then there is a smaller number on the LHS of Eq. (iv) than
on RHS.
2
Hence there were an x such that <x <2−y ...(v)
f (y )
But now from Eq. (ii) it implies there is 0 on LHS of Eq. (A)
It is given f (x ) ≠ 0 if 0 ≤ x < 2 ...(C)
www.pdfworld.in
www.pdfworld.in
⇒ RHS ≠ 0 which is a contradiction

∴for every 0 ≤ y < 2 Eq. (iv) holds
2
f (y ) = (0 ≤ y < 2) ...(vi)
2−y
So only function which satisfies A,B,C, is
 2 0≤ x < 2
f (x ) = 2 − x
if
0 if x ≥2
if y ≥ 2, Eqs. (B) and (C) are true
∴ Both sides of Eq. (A) is 0
Considering the case 0≤ y < 2
Now, assume that x + y ≥2
Hence RHS = 0
then x ≥2−y
2x 2 (2 − y )
and x ⋅ f (y ) = ≥ =2
2−y 2−y
2 2 2 (2 − y ) ⋅ 2 2
So, f (x ⋅ f (y )) ⋅ f (y ) = ⋅ = = = f (x + y )
2−
2x 2 − y (4 − 2y − 2x ) (2 − y ) 2 − (x + y)
2−y
So, f (x ) satisfies Eq. (A).
Example 23. Find all functions of defined on the set of all real numbers with real values, such that,
f (x 2 + f ( y )) = y + ( f (x ))2 ∀ x , y
Solution f (x 2 + f ( y )) = y + (f (x ))2 ...(i)

Assume that for some real number y
f ( y ) < y i . e. , y − f ( y ) > 0 …(ii)
Let x be such that x = y − f ( y )
2
i . e. , y = x 2 + f (y )
from Eq. (i)
f ( y ) = f (x 2 + f ( y )) = y + f 2 (x ) ≥ y
which contradicts Eq. (ii)
Hence for every real number y
f (y ) ≥ y ...(iii)
Set now y 0 − t to be smaller than − f 2 (0) denote f (y 0 ) by a
from Eqs. (i) and (iii), we get
a ≤ f (a ) = f (02 + f ( y 0 )) = y 0 + f 2 (0) < 0
i . e. , a ≤ f (a ) < 0
∴ a 2 ≥ f 2 (a ) ...(iv)
Let x be now arbitrary
Using Eqs. (i), (iii) and (iv), we get
x + a 2 ≤ a 2 + f (x ) ≤ f (a 2 + f (x ))
= x + f 2 (a ) ≤ x + a 2 ...(v)
QThe lowest and highest terms are equal there is equality every where in Eq. (v)
Thus, f (x ) = x
for every real number x.
www.pdfworld.in
www.pdfworld.in
Let us Practice
Level 1
1+x f (x − y ) = f (x )f ( y ) − f (a − x )f (a + y )
1. If f (x ) = , prove that
1−x
where a is given constant and f (0) = 1, then
f (x ) ⋅ f (x 2 )
1 prove that
=
1 + [f (x )]2 2 f (2a − x ) = − f (x ) .
2. If f (x + y , x − y ) = xy , then find the arithmetic αx
7. If f (x ) = , x ≠ − 1, then for what value of α
mean of x +1
f (x , y ) and f ( y , x ). is f (f (x )) = x .
1 g (x ) = 1 +
3. If f (x ) = 64x 3 + and a , b are roots of 8. If x
x3
1 and f ( g (x )) = 3 + 2 x + x , then find f (x ) .
4x + = 3, then prove that
x 9. If f is an even function defined on the interval
f (a ) = f (b ). [−5, 5] , then find real values of x satisfying
 x + 1
4. If f (x + y ) = f (x ) + f (y ) − 1 for all x , y ∈ R and f (x ) = f  
f (1) = 1, then find the number of solutions of  x + 2
f (n ) = n , n ∈ N . 10. Determine the function f satisfying
x − 1
5. Find the value of natural number a for which f (x ) + f   = 1 + x , ∀ x ∈ R − {0, 1}
n  x 
Σ f (a + k ) = 16(2n − 1) where the function f
k =1
11. Let X be the set of all positive integers greater
satisfies the relation f (x + y ) = f (x )f (y ) for all
than or equal to 8 and let f : X → X be a
natural numbers x , y and further f (1) = 2.
function such that f (x + y ) = f (xy ) for all
6. A real valued function f (x ) satisfies the x ≥ 4, y ≥ 4. If f (8) = 9, determine f (9).
functional equation (RMO 2006)
Level 2
ax 3. Let f (x ) be a polynomial function of degree n
1. If f (x ) = , a > 0, then prove that
ax + a satisfying the condition
 1  1
f (x ) f   = f (x ) + f   , Find f (x ) .
f (x ) + f (1 − x ) = 1 x  x 
2n − 1
 k 
and Σ 2f   = 2n − 1 4. Let f be a function from the set of natural
k =1  2n 
numbers to the set of real numbers i.e.,
2n  k  f : N → R such that
and Σ 2f   = 2n
k =1  2n + 1  (i) f (1) = 1
(ii) f (1) + 2f (2) + 3f (3) + . . . + nf (n ) = n (n + 1)f (n )
2. Determine a function f : R → R such that
for all n ≥ 2, then find f (2008.)
f (x − f ( y )) = f (f ( y )) + xf ( y ) + f (x ) − 1 for all
x , y ∈ R. 5. Let f be polynomial function such that
f (x ) f (y ) + 2 = f (x ) + f ( y ) + f (xy ) ; ∀ x , y ∈ R.
www.pdfworld.in
www.pdfworld.in
Suppose f is one-one. If f (3) = 82 and f (9) ≠ 2, f (x + y ) = f (x ) f ( y )f (xy ) for all x , y in R.

then show that (INMO 2001)
f (x ) = x 4 + 1, ∀x ∈ R.
9. Find all functions f : R → R such that
6. Determine all functions f satisfying the f (x 2 + yf (z )) = xf (x ) + zf ( y )
functional relation
for all x , y , z in R. (Here R denotes the set of all
 2x + 29  real numbers). (INMO 2005)
2f (x ) + 3f   = 100x + 80,
 x −2 
10. Let X denote the set of all triples (a , b , c ) of
where x is a real number different from 2. integers. Define a function f : X → X by
f (a , b , c ) = (a + b + c , ab + bc + ca , abc ).
7. For any natural number n , (n ≥ 3), let f (n )
denote the number of non-congruent Find all triples (a , b , c ) in X such that
integer-sided triangles with perimeter n (e.g., f (f (a , b , c )) = (a , b , c ) (INMO 2006)
f (3) = 1, f (4) = 0, f (7) = 2). Show that
11. Prove that for every positive integer n there
(a) f (1999) > f (1996) ; exists a unique ordered pair (a , b ) of positive
(b) f (2000) = f (1997) . (INMO 2000) integers such that
1
n = (a + b − 1)(a + b − 2) + a
8. Let R denote the set of all real numbers. Find 2
all functions f : R → R satisfying the condition (INMO 2006)
www.pdfworld.in
www.pdfworld.in
Solutions
Level 1
1+x 5. Q f (x + y ) = f (x )f (y )
1. Q f (x ) =
1−x Putting x = y = 1, f (2) = f (1) ⋅ f (1) = 22
1 + x2
∴ f (x 2 ) = Putting x = 2, y = 1, f (3) = f (2)f (1) = 23
1 −x2
…… …… ……
1 + x  1 + x 2 
    f (k ) = 2k
f (x ) ⋅ f (x 2 ) 1 − x  1 − x2
Now, LHS = = n
1 + [f (x )]2 2 Now, from given Σ f (a + k ) = 16(2n − 1)
1 + x  k =1
1+  
1 − x  ⇒
n
Σ f (a )f (k ) = 16(2n − 1)
k =1
1+x 2
n
(1 − x )2 ⇒ f (a ) Σ 2k = 16(2n − 1)
= k =1
(1 − x )2 + (1 + x )2
2(2 − 1)
n
(1 − x )2 ⇒ f (a ) ⋅ = 16(2n − 1)
2−1
1 + x2 1
= = ⇒ f (a ) = 8 = 2a ⇒ a = 3
2(1 + x 2 ) 2
6. Given, f (x − y ) = f (x )f ( y ) − f (a − x )f (a + y )
2. Let X = x + y ,Y = x − y
X +Y X −Y Putting x = y = 0
⇒ x = ,y =
2 2 ⇒ f (0) = ( f (0))2 − ( f (a ))2
Given functional relation becomes ⇒ 1 = 1 − ( f (a ))2
X + Y   X −Y  X −Y 2 2
f (X , Y ) =    =
⇒ f (a ) = 0
 2   2  4 ∴ f (2a − x ) = f (a − (x − a ))
x 2 − y2 y2 − x 2 = f (a )f (x − a ) − f (a − a )f (x )
∴ f (x , y ) = and f ( y , x ) =
4 4 = 0 − f (x ) ⋅ 1 = − f (x )
f (x , y ) + f ( y , x )
∴ Arithmetic mean = αx
2 α⋅
αf (x ) x +1
x 2 − y2 y2 − x 2 7. f ( f (x )) = =
+ f (x ) + 1 αx
4 4 +1
= =0 x +1
2
α 2x
3 ⇒ x =
3. 64x 3 + 3 =  4x +  − 3 ⋅ 4x ⋅  4x + 
1 1 1 1 (α + 1) x + 1
x  x  x  x 
which holds only when α = − 1
= 27 − 12 × 3 = 27 − 36 = − 9
∴ f (x ) = − 9 ⇒ f (a ) = f (b ) = − 9 8. f (1 + x ) = 3 + 2 x + x = (1 + x )2 + 2
⇒ f (x ) = x + 2
2
4. Since, f (x + y ) = f (x ) + f ( y ) − xy − 1 ∀ x , y ∈ R
Putting y = 1  x + 1
9. f (x ) = f  
f (x + 1) = f (x ) + f (1) − x − 1  x + 2
⇒ f (x + 1) = f (x ) − x x +1
⇒ x =
∴ f (n + 1) = f (n ) − n < f (n ) x +2
So, f (n ) < f (n − 1) < f (n − 2) < . . . ⇒ x (x + 2) − (x + 1) = 0
< f (3) < f (2) < f (1) = 1
⇒ x2 + x −1 = 0
∴f (n ) = n holds only for n = 1.
www.pdfworld.in
www.pdfworld.in
−1 ± 1 − 4 ⋅ 1 ⋅ (−1) −1 ± 5 1
∴ x = = …(i) Replacing x by − in Eq. (i), we get
2 2 x −1
Replacing x by −x  − 1 − 1
  
 −x + 1  1  x −1  = 1 + −1
f (−x ) = f   f −  +f 
 x − 1   − 1  x −1
 −x + 2   x −1 
 
 −x + 1 
⇒ f (x ) = f   [Qf (x ) is even]  1  1
 −x + 2  ⇒ f −  + f (x ) = 1 − …(iv)
 x − 1 x −1
−x + 1
⇒ x = Adding Eqs. (iii) and (iv), we get
−x + 2
x −1 1
2f (x ) = x − +1−
⇒ −x 2 + 2x + x − 1 = 0 x x −1
⇒ x 2 − 3x + 1 = 0 x 3 − x 2 − x 2 + 2x − 1 + x 2 − x − x
=
3± 9−4 3± 5 x (x − 1)
∴ x = = …(ii)
2 2 x3 − x 2 − 1
=
∴All real values of x satisfying given relation x (x − 1)
is given by Eqs. (i) and (ii). x3 − x 2 − 1
⇒ f (x ) =
2x (x − 1)
10. Let f be a given real valued function satisfying
which is the required function.
the condition,
11. We observe that,
x − 1
f (x ) + f   = 1 + x , ∀ x ∈ R − {0, 1} …(i) f (9) = f (4 + 5) = f (4 ⋅ 5) = f (20)
 x 
= f (16 + 4) = f (16 ⋅ 4) = f (64)
x −1
Let us put in place of x so that = f (8 ⋅ 8) = f (8 + 8) = f (16)
x
= f (4 ⋅ 4) = f (4 + 4) = f (8)
 x − 1 − 1
x − 1   x −1 Hence, if f (8) = 9, then f (9) = 9. (This is one

f   +f  x  =1+ string. There may be other different ways of
 x   x −1  x
approaching f (8) from f (9). The important
 x 
thing to be observed is the fact that the rule
x − 1  −1  x −1 f (x + y ) = f (xy ) applies only when x and y are at
⇒ f   +f   =1+ …(ii)
 x   x − 1 x least 4. One may get strings using numbers
x and y which are smaller then 4 but that is
Subtracting Eq. (ii) from Eq. (i), we get not valid. For example f (9) = f (3 ⋅ 3) = f (3 + 3) =
 1  x −1
f (x ) − f  −  =x − …(iii) f (6) = f (4 + 2) = f (4 ⋅ 2) = f (8), is not a valid
 x − 1 x string.)
Level 2
ax Putting the terms equidistant from ends
1. We have f (x ) =
a + a
x
together, we get
a1 − x a a  1    2  2 
= 2   f   + f  1 − 
⇒ f (1 − x ) = 1 − x = = 1
  + f   + f 1 − 
a + a a + ax a ax + a    2n   2n     2n   2n
ax a
∴ f (x ) + f (1 − x ) = x +  3 
+  f   + f  1 −
3
a + a ax + a  + . . . +
  2n   2n  
ax + a
= x =1
a + a   n − 1  n − 1   n 
 f   + f 1 −  + f  
2n − 1
 k    2n   2n    2n  
Now, Σ 2f   = 2f   + f   + f  
1 2 3
k =1  2n    2n   2n   2n 
 1 
2n − 2   2n − 1   = 2  (n − 1) + f   
+ f   + . . . + f 
4  2 
 +f   
 2n   2n   2n  
www.pdfworld.in
www.pdfworld.in
Functions 613
 1  ⇒ f (a − c ) = f (c ) + ac + f (a ) − 1
 a2  1
= 2n − 2 + 2  1  = 2n − 2 + 2 ⋅ = 2n − 1 f (0) + 1 c 2
2 ⇒ f (a − c ) = − + ac
 2  2 2
a + a 
f (0) + 1 a 2
2n  k  + − −1
Also, Σ 2f   2 2
k =1  2n + 1 
a 2 + c 2 − 2ac (a − c )2
  1  ⇒ f (a − c ) = f (0) − = f (0) −
 2   3  2 2
= 2 f   +f   +f  
  2n + 1   2n + 1   2n + 1  As f is defined from R to R and a, c ∈ R so
 4   2n − 1   2n   ∃ x ∈ R such that x = a − c and hence
+f   + ... + f   +f  
 2n + 1   2n + 1   2n + 1   x2
f (x ) = f (0) −
2
Putting terms equidistant from the ends
a 2 f (0) + 1 a 2
together, we get ⇒ f (a ) = f (0) − = −
  1  2 2 2
 1 
= 2 f   + f 1 −  [from Eq. (i)]
   2n + 1   2n + 1 
⇒ 2f (0) = f (0) + 1
  2   2  ⇒ f (0) = 1
+ f   + f 1 − 
  2n + 1   2n + 1  x2
Hence, f (x ) = 1 −
  3  2
 3 
+ f   + f 1 −  It can be easily verified that the function
  2n + 1   2n + 1  x2
f (x ) = 1 − satisfies the given functional
  n   n  2
+ K + f   + f 1 − 
  2n + 1   2n + 1    relation.
Second Method
  1   1 
= 2  f   + f 1 −  According to the given condition
   2n + 1   2 n + 1 
f (x − f ( y )) = f ( f ( y )) + xf ( y ) + f (x ) − 1, ∀x , y ∈ R
  2   2  Let us put x = f ( y ) = 0 so that
+ f   + f 1 − 
  2n + 1   2n + 1  f (0) = f (0) + 0f (0) + f (0) − 1 ⇒ f (0) = 1
  3   3  Again putting x = f ( y ) = k so that
+ f   + f 1 − 
 2n + 1   2n + 1  f (0) = f (k ) + k 2 + f (k ) − 1 = 1

  n   n   k2
+ K + f   + f 1 −   ⇒ 2f (k ) = 2 − k 2 ⇒ f (k ) = 1 −
2
  2n + 1   2n + 1  
x2
= 2[1 + 1 + 1 + . . . upto n term] = 2n ⇒ f (x ) = 1 − is the required function
2
2. First Method satisfying the given condition.
Let f : R → R such that f (x − f ( y ))
f (x )f   = f (x ) + f  
1 1
3. We have
= f ( f ( y )) + xf ( y ) + f (x ) − 1, ∀x , y ∈ R x  x 
Let a , b ∈ R such that f (b ) = a.
⇒ 1 − f (x ) − f   + f (x ) + f   = 1
1 1
Let us put x = a , y = b in the given equation, we x  x 
get
(1 − f (x )) − f   (1 − f (x )) = 1
1
f (a − f (b )) = f ( f (b )) + af (b ) + f (a ) − 1 ⇒
x 
⇒ f (a − a ) = f (a ) + a 2 + f (a ) − 1  1 
⇒ (1 − f (x ))  1 − f    = 1
⇒ f (0) = 2f (a ) + a 2 − 1  x 
f (0) + 1 a 2 Since, f (x ) is a polynomial function of degree n
⇒ f (a ) = − …(i)
 1 
and (1 − f (x ))  1 − f    = 1, constant, such
2 2
Again let y1 ∈ R such that f ( y1 ) = c  x 
Let us put x = a , y = y1 in the given equation,
that 1 − f (x ) and 1 − f   are reciprocal in
1
we get x 
f (a − f ( y1 )) = f ( f ( y1 )) + af ( y1 ) + f (a ) − 1 magnitude which will be possible when
www.pdfworld.in
www.pdfworld.in
1 − f (x ) = x n or 1 − f (x ) = − x n ⇒ 3n = 81 = 34 (neglecting –ve sign)

⇒ n=4
1 − f   = n or 1 − f   = − n
1 1 1 1
x  x x  x Hence, f (x ) = x 4 + 1
Whence f (x ) = ± x n + 1  2x + 29 
6. We have, 2f (x ) + 3f   = 100x + 80
4. Let f : N → R be such that f (1) = 1 and  x −2 
f (1) + 2f (2) + 3f (3) + . . . + nf (n ) 2x + 29
Let us put in place of x , we get
= n (n + 1)f (n ), ∀n ≥ 2 …(i) x −2
⇒ f ( f ) + 2f (2) + 3f (3) + . . . + nf (n )  2 ⋅ 2x + 29 + 29 
 2x + 29   
x −2
+ (n + 1)f (n + 1) = (n + 1)(n + 2)f (n + 1) …(ii) 2f   + 3f  
 x −2   2x + 29 − 2 
On subtraction Eq. (i) from Eq. (ii)  x −2 
 
(n + 1)f (n + 1) = (n + 1)(n + 2)f (n + 1)
2x + 29
−n (n + 1)f (n ) = 100 ⋅ + 80
x −2
⇒ (n + 1)(n + 2 − 1)f (n + 1) = n (n + 1)f (n )
 2x + 29   4x + 58 + 29x − 58
⇒ nf (n ) = (n + 1)f (n + 1) ⇒ 2f   + 3f  
 x −2   2x + 29 − 2x + 4 
⇒ 2f (2) = 3f (3) = 4f (4) = 5f (5) = . . . = nf (n )
 2x + 29 
Thus, the given result reduces to = 100   + 80
 x −2 
f (1) + (n − 1)nf (n ) = n (n + 1)(n )
⇒ f (1) = n (n + 1 − n + 1)f (n ) = 2nf (n )  2x + 29   33x 
⇒ 2f   + 3f  
f (1) 1  x − 2   33 
⇒ f (n ) = = [Qf (1) = 1]
2n 2n  2x + 29 
1 1 = 100   + 80
⇒ f (2008) = =  x −2 
2 × 2008 4016
 2x + 29  −3  2x + 29 
5. We have, f (x )f ( y ) + 2 = f (x ) + f ( y ) + f (xy ) ⇒ f   = f (x ) + 50   + 40
 x − 2  2  x −2 
Putting x = y = 0, we get
With this result, the given functional relation
f (0)f (0) + 2 = f (0) + f (0) + f (0)
reduces to
⇒ ( f (0))2 − 3f (0) + 2 = 0 2x + 29 
2f (x ) − f (x ) + 150 
9
 + 120 = 100x + 80
⇒ ( f (0) − 2)f ( f (0) − 1) = 0 2  x −2 
⇒ f (0) = 2 or f (0) = 1  2x + 29 
⇒ f (x ) = 60   + 16 − 40x
But f (0) ≠ 2 (given) hence f (0) = 1  x −2 
Again putting x = y = 1, we get 120x + 1740 + 16x − 40x 2 − 32 + 80x
( f (1))2 + 2 = f (1) + f (1) + f (1) =
x −2
⇒ ( f (1) − 2)( f (1) − 1) = 0 1708 + 216x − 40x 2
⇒ f (1) = 1 or f (1) = 2 =
x −2
But f is given to be injective (one-one) and
f (0) = 1, hence f (1) ≠ 1 7. (a) Let a , b , c be the sides of a triangle with
∴ f (1) = 2 a + b + c = 1996, and each being a positive
1 integer. Then, a + 1, b + 1, c + 1 are also
Now, putting y = in the given relation, we get sides of a triangle with perimeter 1999
x because

f (x )f   + 2 = f (x ) + f   + f (1)
1 1 a < b + c ⇒ a + 1 < (b + 1) + (c + 1)
x  x 
and so on. Moreover (999, 999, 1) form the
f (x )f   = f (x ) + f   ;
1 1
⇒ [Qf (1) = 2] sides of a triangle with perimeter 1999,
x  x  which is not obtainable in the form
⇒ f (x ) = ± x n + 1 (a + 1, b + 1, c + 1) where a , b , c are the
integers and the sides of a triangle with
From the given condition we have
a + b + c = 1996. We conclude that
f (3) = 82 = ± 3n + 1 f (1999) > f (1996).
www.pdfworld.in
www.pdfworld.in
Functions 615
(b) As in the case (a) we conclude that f ( f (z )) = zf (1), ∀z ∈ R …(iv)

f (2000) ≥ f (1997). On the other hand, if Now, using Eqs. (ii) and (iv), we obtain
x , y , z are the integer sides of a triangle
with x + y + z = 2000, and say x ≥ y ≥ z ≥ 1, f ( xf (x )) = f ( f (x 2 )) = x 2f (1) …(v)
then we cannot have z = 1; for otherwise Put y = z = x in Eq. (iii) also given
we would get x + y = 1999 forcing x , y to f ( xf (x )) = xf (x ) …(vi)
have opposite parity so that x − y ≥ 1 = z
violating triangle inequality for x , y , z. Comparing Eqs. (v) and (vi), it follows that
Hence, x ≥ y ≥ z ≥ 1. This implies that x 2f (1) = xf (x ). If x ≠ 0, then f (x ) = cx , for some
x − 1 ≥ y − 1 ≥ z − 1 > 0. We already have constant c. Since, f (0) = 0, we have f (x ) = cx for
x < y + z. If x ≥ y + z − 1, then we see that x = 0 as well. Substituting this in Eq. (i), we see
y + z − 1 ≤ x < y + z, showing that that
y + z − 1 = x. Hence, we obtain c (x 2 + cyz ) = cx 2 + cyz
2000 = x + y + z = 2x + 1 which is
or c 2yz = cyz , ∀y , z ∈ R
impossible. We conclude that x < y + z − 1.
This shows that x − 1 < ( y − 1) + (z − 1) and This implies that c 2 = c . Hence, c = 0 or 1. We
hence x − 1, y − 1, z − 1 are the sides of a obtain f (x ) = 0 for all x or f (x ) = x for all x. It is
triangle with perimeter 1997. This gives easy to verify that these two are solutions of
f (2000) ≤ f (1997). Thus, we obtain the the given equation.
desired result.
10. We show that the solution set consists of
8. Putting x = 0, y = 0, we get f (0) = f (0)3 so that {(t , 0, 0); t ∈ Z } ∪ {(−1, − 1, 1)}. Let us put
f (0) = 0, 1 or −1. a + b + c = d , ab + bc + ca = e and abc = f . The
If f (0) = 0, then taking y = 0 in the given given condition f ( f (a , b , c )) = (a , b , c ) implies
equation, we obtain f (x ) = f (x )f (0)2 = 0 for all x . that
d + e + f = a, de + ef + fd = b , def = c
Suppose f (0) = 1. Taking y = − x , we obtain
Thus, abcdef = fc and hence either cf = 0 or
1 = f (0) = f (x − x ) = f (x )f (−x )f (−x 2 ) abde = 1.
This shows that f (x ) ≠ 0 for any x ∈ R. Taking Case I Suppose cf = 0. Then, either c = 0 or
x = 1, y = x − 1, we obtain f = 0. However c = 0 implies f = 0 and
f (x ) = f (1)f (x − 1)2 = f (1)[f (x )f (−x )f (−x )]2 vice-versa. Thus, we obtain
a + b = d , d + e = a, ab = e and de = b. The first
Using f (x ) ≠ 0, we conclude that
two relations give b = − e. Thus, e = ab = − ae
1 = kf (x )( f (−x ))2 where k = f (1)( f (−1))2.
and de = b = − e. We get either e = 0 or
Changing x to −x here, we also infer that a = d = − 1.
1 = kf (−x )( f (x ))2. Comparing these expression If e = 0, then b = 0 and a = d = t , say. We get the
we see that f (−x ) = f (x ). It follows that 1 = kf (x )3. triple (a , b , c ) = (t , 0, 0), where t ∈ Z . If e ≠ 0,
Thus, f (x ) is constant for all x . Since, f (0) = 1. then a = d = − 1. But then d + e + f = a implies
We conclude that f (x ) = 1 for all real x. that −1 + e + 0 = − 1 forcing e = 0. Thus, we get
If f (0) = − 1, a similar analysis shows that the solution family (a , b , c ) = (t , 0, 0) where
f (x ) = − 1 for all x ∈ R. We can verify that each t ∈Z.
of these functions satisfies the given Case II Suppose cf ≠ 0. In this case abde = 1.
functional equation. Thus, there are three Hence, either all are equal to 1; or two equal to
solutions, all of them being constant 1 and the other two equal to −1; or all equal to
functions. −1.
9. f (x 2 + yf (z )) = xf (x ) + zf ( y ) …(i) Suppose a = b = d = e = 1. Then, a + b + c = d
shows that c = − 1. Similarly, f = − 1. Hence,
Taking x = y = 0 in Eq. (i), we get zf (0) = f (0) for
e = ab + bc + ca = 1 − 1 − 1 = − 1 contradicting
all z ∈ R. Hence, we obtain f (0) = 0. Taking y = 0
e = 1.
in Eq. (i), we get
Suppose a = b = 1 and d = e = − 1. Then,
f (x 2 ) = xf (x ) …(ii)
a + b + c = d gives c = − 3 and d + e + f = a
Similarly, x = 0 in Eq. (i) gives gives f = 3. But, then f = abc = 1 ⋅ 1 ⋅ (−3) = − 3, a
f ( yf (z )) = zf ( y ) …(iii) contradiction. Similarly a = b = − 1 and
Putting y = 1 in Eq. (iii), we get d = e = 1 is not possible.
www.pdfworld.in
www.pdfworld.in
If a = 1, b = − 1,d = 1, e = − 1, then a + b + c = d For example the first entry in 5th diagonal is

gives c = 1. Similarly f = 1. But then 11 which is one more than the last entry of 4th
f = abc = 1 ⋅ 1 ⋅ (−1) = − 1 a contradiction. If diagonal which is 10. Observe that 5th
a = 1, b = − 1, d = − 1, e = 1, then c = − 1 and diagonal starts from 11 and ends with 15
e = ab + bc + ca = − 1 + 1 − 1 = − 1 and a which accounts for 5 consecutive natural
contradiction to e = 1. The symmetry between numbers. Thus, we see that
(a , b , c ) and (d , e , f ) shows that f (n − 1, 1) + 1 = f (1, n ). We also observe that the
a = − 1, b = 1, d = 1, e = − 1 is not possible. first n diagonals exhaust all the natural
Finally, if a = − 1, b = 1, d = − 1 and e = 1, then numbers from 1 to T (n ). (This a kind of visual
c = − 1 and f = − 1. But then f = abc is not bijection is already there. We formally prove
satisfied. the property.)
The only case left is that of a , b , d , e being all We first observe that
equal to −1. Then, c = 1 and f = 1. It is easy to f (a , b ) − T (a + b − 2) = a > 0,
check that (−1, − 1, 1) is indeed a solution. (a + b − 1)(a + b )
Aliter cf ≠ 0 implies that | c | ≥ 1 and | f | ≥ 1. and T (a + b − 1) − f (a , b ) =
2
Observe that
(a + b − 1)(a + b − 2)
d 2 − 2e = a 2 + b 2 + c 2 , a 2 − 2b = d 2 + e 2 + f 2 − − a = b − 1≥ 0
2
Adding these two, we get Thus, we have
−2(b + e ) = b 2 + c 2 + e 2 + f 2. This may be (a + b − 1)(a + b − 2)
T (a + b − 2) < f (a , b ) =
written in the form 2
(b + 1)2 + (e + 1)2 + c 2 + f 2 − 2 = 0. + a ≤ T (a + b − 1)
Suppose f (a1 , b1 ) = f (a 2 , b2 ). Then, the previous
We conclude that c 2 + f 2 ≤ 2. Using | c | ≥ 1 and observation shows that
| f | ≥ 1, we obtain | c | = 1 and | f | = 1, b + 1 = 0 T (a1 + b1 − 2) < f (a1 , b1 ) ≤ T (a1 + b1 − 1),
and e + 1 = 0. Thus, b = e = − 1. Now
T (a 2 + b2 − 2) < f (a 2 , b2 ) ≤ T (a 2 + b2 − 1).
a + d = d + e + f + a + b + c and this gives
b + c + e + f = 0. It follows that c = f = 1 and Since, the sequence 〈T (n )〉n∞ = 0 is strictly
finally a = d = − 1. increasing, it follows that a1 + b1 = a 2 + b2. But
11. We have to prove that f : N × N → N defined then the relation f (a1 , b1 ) = f (a 2 , b2 ) implies
by that a1 = a 2 and b1 = b2. Hence, f is one-one.
1 Let n be any natural number. Since, the
f (a , b ) = (a + b − 1)(a + b − 2) + a , ∀a , b ∈ N ,
2 sequence 〈T (n )〉n∞ = 0 is strictly increasing we
is a bijection. (Note that the right side is a can find a natural number k such that
natural number). To this end define T (k − 1) < n ≤ T (k )
n (n + 1)
T (n ) = , n ∈ N ∪ {0}. Equivalently,
2 (k − 1)k k (k + 1)
<n≤ …(i)
An idea of the proof can be obtained by 2 2
looking at the following table of values of k (k − 1)
f (a , b ) for some small values of a , b. Now, set a = n − and b = k − a + 1.
2
b Observe that a > 0. Now Eq. (i) shows that
a 1 2 3 4 5 6
k (k − 1) k (k + 1) k (k − 1)
1 1 2 4 7 11 16 a =n − ≤ − =k
2 3 5 8 12 17
2 2 2
3 6 9 13 18 Hence, b = k − a + 1 ≥ 1. Thus, a and b are both
4 10 14 19 positive integers and
1
5 15 20 f (a , b ) = (a + b − 1)(a + b − 2) + a
6 21 2
k (k − 1)
We observe that the nth diagonal runs from = + a =n
(1, n ) th position to (n , 1) th position and the 2
entries are n consecutive integers; the first This shows that every natural number is in the
entry in the nth diagonal is one more than range of f . Thus, f is also onto. We conclude
the last entry of the (n − 1) th diagonal. that f is a bijection.
www.pdfworld.in
www.pdfworld.in
Solved Paper 2016
RMO
Regional Mathematics Olympiad
Conducted by: Homi Bhabha Centre for Science Education, India
Exam Held on 09-10-2016
Regional Mathematics Olympiad is the first stage of 5 Stages of Mathematics Olympiad Program.
On the basis of the performance in RMO, a certain number of students from each region is
selected for Stage 2 (Indian National Mathematics Olympiad). Atmost 6 Class XII students from
each region will be selected to appear for Stage 2 (INMO).
1. Let ABC be a right-angled triangle with ∠B = 90°. Let I be the incentre of ∆ABC. Draw a line
perpendicular to AI at I. Let it intersect the line CB at D. Prove that CI is perpendicular to AD and
prove that ID = b (b − a ), where BC = a and CA = b.
2. Let a , b and c be positive real numbers such that
a b c
+ + =1
1+ a 1+ b 1+ c
1
Prove that abc ≤ .
8
3. For any natural number n, expressed in base 10, let S (n ) denotes the sum of all digits of n. Find all
natural numbers n such that n 3 = 8(S (n ))3 + 6n S (n ) + 1.
4. How many 6-digit natural numbers containing only the digits 1, 2, 3 are there in which 3 occurs
exactly twice and the number is divisible by 9?
5. Let ABC be a right-angled triangle with ∠B = 90°. Let AD be the bisector of ∠ A with D on BC. Let the
circumcircle of ∆ACD intersect AB again in E and let the circumcircle of ∆ABD intersect AC again in
F. Let K be reflection of E in the line BC. Prove that FK = BC.
6. Show that the infinite arithmetic progression 1, 4, 7, 10,…… has infinitely many 3-term
subsequences in harmonic progression such that for any two such triples a1 , a 2 , a 3 and b1 , b2 , b3 in
harmonic progression, one has
a1 a 2 a 3
≠ ≠
b1 b2 b3
www.pdfworld.in
www.pdfworld.in
Detailed Solutions
1. Let us draw the following diagram according to ⇒ a + ab + ca + abc + b + ab + bc + abc
the question. + c + ca + bc + abc
A = 1 + b + c + bc + a + ab + ac + abc
⇒ ab + bc + ca + 2 abc = 1
Now, we know that
AM ≥ GM
1
ab + bc + ca + 2 abc
I ⇒ ≥ (ab ⋅ bc ⋅ ca ⋅ 2 abc )4
4
1
⇒ 1 ≥ 4(2 a3 b 3c 3 )4
D B C
4
⇒  1  ≥ 2 a3 b 3c 3
Since, ∠AID = ∠ABD = 90°  
 4
∴ADBI is a cyclic quadrilateral.
1
∴ ∠ADI = ∠ABI = 45° ⇒ a3 b 3c 3 ≤
512
⇒ ∠DAI = 45° 1
⇒ abc ≤ Hence proved.
We also have, 8
∠ADB = ∠ADI + ∠IDB = 45° + ∠IAB 3. We have,
= ∠DAI + ∠IAC = ∠DAC n3 = 8 (S(n))3 + 6n S(n) + 1
∴∆CDA is an isosceles triangle. ⇒ n − 8 (S(n)) − 1 = 6n S(n)
3 3
∴ CD = CA ⇒(n)3 + (− 2 S(n))3 + (−1)3

Since, CI bisects ∠C. = 3 × n × (− 2 S(n))(−1)
⇒ CI ⊥ AD Hence proved. ⇒ n + (− 2 S(n)) + (−1) = 0
This shows that DB = CA − CB = b − a [If x 3 + y3 + z3 = 3xyz ⇒ x + y + z = 0]
∴ AD2 = c 2 + (b − a)2 ⇒ n − 2 S(n) − 1 = 0
= c + b + a − 2 ba
2 2 2
⇒ n − S(n) = S(n) + 1 …(i)
= c 2 + a2 + b 2 − 2 ba Again, we know that for every number n − S(n) is
= b 2 + b 2 − 2 ba always divisible by 9.
= 2 b 2 − 2 ba = 2b (b − a) ∴S(n) + 1is also divisible by 9. [from Eq. (i)]
Q ID = cos 45°  Now, for a three digit number maximum value of
Again, 2 ID2 = AD2
 AD  S(n) can be 27.
∴ 2 S(n) + 1 = 2 × 27 + 1 = 55
⇒ 2 ID2 = 2 b(b − a)
∴ 2 S(n) + 1 ≤ 55
⇒ ID = b (b − a)
2
Hence, n is either a 1-digit number or a two digit

⇒ ID = b (b − a) Hence proved.
number.
a b c
2. We have, + + =1 Hence, S(n) ≤ 18. Since, S(n) + 1must be divisible
1+ a 1+ b 1+ c
by 9.
⇒ a(1 + b ) (1 + c ) + b (1 + a) (1 + c ) ∴ S(n) = 8 or S(n) = 17
+ c(1 + a) (1 + b ) ∴ n = 17 or 35 [from Eq. (i)]
= (1 + a) (1 + b )(1 + c ) Among these n = 17 works but not 35,
⇒ a(1 + b + c + bc ) + b (1 + a + c + ac ) as S(35) = 8
+ c (1+ a + b + ab ) and 2 S(n) + 1 = 17 ≠ 35
= (1 + a) (1 + b + c + bc ) Hence, the only solution is n = 17.
www.pdfworld.in
www.pdfworld.in
RMO Solved Paper 2016

3
4. Let the number be n and sum of its digit be S(n). Also, AF = AB as ∆ABD ~
= ∆AFD
⇒ FB|| CK
Since, n is divisible by 9.
Since, FC = BK, we can say that CKDF is an
∴ n ≡ S(n)(mod 9)
isosceles trapezium.
We have, n contains 1, 2, 3 and 3 can occurs
∴ FK = BC Hence proved.
exactly twice.
∴ 10 ≤ S(n) ≤ 14 6. Consider the triplet < 4, 7, 28 >
1 1 7+1
Since, there is no value of S(n) that is a multiple Then, + =
of 9. 4 28 28
8
Thus, no such n exists. =
28
5. Consider the ∆EBD and ∆CFD.
2
∠CFD = 90° =
7
[Q∠AFD = 90°,∠CFD + ∠AFD = 180°]
∴4, 7, 28 are in HP.
∴ ∠CFD = ∠ EBD
Thus, we can say that a, b, ab are in HP and
Since, ACDE is a cyclic quadrilateral.
required condition is
∴ ∠CDE = 180° − ∠A 1 1 2
+ =
a ab b
b+1 2
A ⇒ =
ab b
⇒ b + 1 = 2a
or 2a = b + 1
E Given, AP is 1, 4, 7, 10, ……
F
∴ ak = 3k + 1
B If we take a = 3k + 1, then
C D
K b = 2 a − 1 = 2 (3k + 1) − 1
= 6k + 1
Similarly, AFDB is a cyclic quadrilateral and We observe that b is also a term of the given AP.
therefore, ∠FDB = 180° − ∠A
Again, ab = (3k + 1) (6k + 1)
∴ ∠CDE = ∠FDB
= 18k 2 + 9k + 1
⇒ ∠FDC = ∠BDE
= 3 (6k 2 + 3k ) + 1
∴ ∆EBD ~ ∆CFD
which is again a term of the given AP.
Now, AD is bisector of ∠A.
Thus, 3k + 1, 6k + 1, (3k + 1) (6k + 1) are in HP.
∴ DB = DF
Also, we have,
∴ ∆EBD ~
= ∆CFD 3m + 1 6m + 1 (3m + 1) (6m + 1)
≠ ≠
Hence, FC = EB = BK 3n + 1 6n + 1 (3n + 1) (6n + 1)
www.pdfworld.in
www.pdfworld.in
Solved Paper 2017
INMO
Indian National Mathematics Olympiad
Exam Held on 15-01-2017
Indian National Mathematics Olympiad is organised by HBCSE (Homi Bhabha Centre for Science
Education). On the basis of performance in INMO, top 35 students are selected from which final
5 students are selected to participate in International Mathematics Olympiad.
1. In the given figure, ABCD is a square paper. It is folded along EF such that A goes to a point A′ ≠ C, B
on the side BC and D goes to D′. The line A′ D′ cuts CD in G. Show that the inradius of the ∆GCA′ is the
sum of the inradii of the ∆GD′ F and ∆A′ BE .
D′
D F G C
A′
A E B
2. Suppose n ≥ 0 is an integer and all the roots of x3 + αx + 4 − (2 × 2016n ) = 0 are integers. Find all
possible values of α.
3. Find the number of triples (x, a , b) where x is a real number and a , b belong to the set {1, 2, 3, 4, 5, 6, 7,
8, 9} such that
x2 − a { x} + b = 0,
where { x} denotes the fractional part of the real number x. (For example {1.1} = 0.1 = { − 0.9}. )
4. Let ABCDE be a convex pentagon in which ∠A = ∠B = ∠C = ∠D = 120° and whose side lengths are 5
consecutive integers in some order. Find all possible values of AB + BC + CD.
www.pdfworld.in
www.pdfworld.in
INMO Solved Paper 2017

5
5. Let ABC be a triangle with ∠A = 90° and AB < AC. Let AD be the altitude from A on BC. Let P , Q and
I denote respectively the incentres of ∆ ’ s ABD , ACD and ABC. Prove that AI is perpendicular to PQ
and AI = PQ.
6. Let n ≥ 1 be an integer and consider the sum
 n  n  n  n
x= ∑ 2k 2
k ≥0
n − 2k
3k =   2n +   2n − 2 ⋅ 3 +   2n − 4 ⋅ 32 + ....
 0  2  4
Show that 2x − 1, 2x, 2x + 1 form the sides of a triangle whose area and inradius are also integers.
Detailed Solutions
p−q u − v
1. From the figure drawn below, ⇒ = …(i)
q w
∠1 + ∠2 = 90°
w + u + pv = v + pu
∠2 + ∠3 = 90°
Q
and
⇒ w = v − u + pu − pv
⇒ ∠1 = ∠3
⇒ w = p(u − v ) − (u − v )
Also, ∠3 + ∠4 = 90°
= (u − v ) ( p − 1)
and ∠2 + ∠3 = 90°
u−v 1
⇒ ∠2 = ∠4 ⇒ = …(ii)
w p−1
Again, ∠4 = ∠6 and ∠3 = ∠6
∴ From Eqs. (i) and (ii), we get
D′
p−q u − v 1
= =
D F 6 5 G C q w p−1
4 p−q 1
Now, =
q p−1
⇒ ( p − q ) ( p − 1) = q
⇒ p ( p − q − 1) = 0
3
But p≠ 0
A′ ∴ p− q − 1= 0
2 ⇒ p=q + 1
1 ⇒ pr = qr + r
A E B
2. Let u, v, w be the roots of
∴∆GCA′ and ∆ A′ BE are similar to the ∆GD′ F.
If GF = u, GD′ = v and D′ F = w, x 3 + αx + 4 − (2 × 2016n ) = 0
then we have ∴Sum of the zero’s = u + v + w = 0
A′G = pu, CG = pv, A′C = pw Sum of the product of zero’s taken two at a line
and A′ E = qu, BE = qw, A′ B = qv = uv + vw + wu = α
and product of the zero’s
Let r be the inradius of ∆GD′ F, then pr and qr will
be the inradius of ∆GCA′ and ∆A′ BE respectively. = uvw = − (4 − 2 × 2016n )
Now, AE = EA′ and DF = FD′ = 2 × 2016n − 4
Again, AB = BC = CD = AD Now, u+ v+ w=0
∴ pw + qv = qw + qu = w + u + pv = v + pu ⇒ w = − (u + v )
∴ pw + qv = qw + qu ∴ uvw = − uv (u + v ) = 2 × 2016n − 4
⇒ ( p − q ) w = q (u − v ) ⇒ uv (u + v ) = 4 − 2 × 2016n
www.pdfworld.in
www.pdfworld.in
Let n≥ 1 4. Let AB = a, BC = b and CD = c. By symmetry, we

Then, uv (u + v ) ≡ 4(mod 2016n ) may assume that c < a. We have, DE = a + b and
∴ uv (u + v ) ≡ 1 (mod 3) or 1(mod 9) EA = b + c
⇒ u = 2 (mod 3) and v = 2 (mod 3) E
∴ (u, v ) can have values
(2, 2), (2, 5), (2, 8), (5, 2), (5, 5), (5, 8), (8, 2), b+c
a+b
(8, 5), (8, 8)
A
But in each case uv (u + v ) ≡/ 4( mod 9)
∴ n must be equal to 0. G
∴ uv (u + v ) = 4 − (2 × 20160 ) = 4 − 2 × 1 = 2 F H D
Hence, (u, v ) = (1, 1), (1, − 2 ), (− 2, 1) a
c
∴ α = uv + vw + wu = uv + w (u + v )
= uv − (u + v )2
= − 3 for (1, 1), (1, − 2), (− 2, 1) B b C
3. Let us write x = n + f, where n = [x ]and f = {x}. Draw a line parallel to BC through D. Extend EA to
Then, f 2 + (2 n − a)f + n2 + b = 0 …(i) meet this line at F. Draw a line parallel to CD
The product of the roots of above equation is through B and let it intersect DF at G. Let AB
n2 + b which is greater than or equal to 1. intersect DF at H. Now, we have ∠FDE = 60° and
∠E = 60°. Hence, EFD is an equilateral triangle.
∴ n2 + b ≥ 1
Similarly, AFH and BGH are also equilateral
For solution of Eq. (i), 0 ≤ f < 1, the larger root triangles. Hence, HG = GB = c. Moreover, DG = b,
must be greater than 1. The Eq. (i) has a real root therefore, HD = b + c. But HD = AE, since
less than 1 only if FH = FA and FD = FE.
1 + 2 n − a + n2 + 2 b < 0
Also, AH = a − BH = a − c
⇒ (n + 1)2 + b < a
Hence, ED = EF = EA + AF = b + c + AH
● If n ≥ 2 , then (n + 1)2 + b ≥ 10 > a. Hence, n ≤ 1.
= (b + c ) + (a − c ) = b + a
If n ≤ − 4 , then again (n + 1)2 + b ≥ 10 > a. Thus,
we have the range for n : − 3, − 2, − 1, 0, 1. Now, we have five possibilities
● If n = − 3 or n = 1, we have (n + 1)2 = 4. Thus, we (1) b < c < a < b + c < a + b;
must have 4 + b < a. If a = 9 , we must have (2) c < b < a < b + c < a + b;
b = 4, 3, 2, 1 giving 4 values. For a = 8, we must (3) c < a < b < b + c < a + b;
have b = 3, 2, 1 giving 3 values.
(4) b < c < b + c < a < a + b;
Similarly, for a = 7 we get 2 values of b and a = 6
(5) c < b < b + c < a < a + b;
leads to 1 value of b. In each case we get a real
value of f < 1and this leads to a solution for x. In case (1) If b = 2 , then c < a < b + c are three
Thus, we get totally 2 (4 + 3 + 2 + 1) = 20 values of consecutive integers.
the triple (x, a, b ). ∴ c = 3 and a = 4
For n = −2, n = 0, we have (n + 1)2 = 1. Hence, we ∴ b + c = 5, a + b = 6
require 1 + b < a. We again count pairs (a, b ) such Hence, we have five consecutive integers 2, 3, 4,
that a − b > 1. For a = 9, we get 7 values of b; for 5, 6 as side lengths.
a = 8 we get 6 values of b and so on.
In case (2) If c = 2, then b < a < b + c form three
Thus, we get consecutive integers.
2 (7 + 6 + 5 + 4 + 3 + 2 + 1) = 56 values for the ∴ b = 3, a = 4
triple (x, a, b ).
But b + c = 5 and a + b = 7
Suppose n = − 1so that (n + 1)2 = 0. In this case
we require b < a. Thus, sides will be 2, 3, 4, 6, 7 which are not
consecutive integers.
We get 8 + 7 + 6 + 5 + 4 + 3 + 2 + 1 = 36
values for the triple (x, a, b ). In case (3) We have b < b + c are two
consecutive integers so that c = 1.
Thus, total triples = 20 + 56 + 36 = 112
www.pdfworld.in
www.pdfworld.in

7
∴ a = 2 and b = 3 ∴ BD =
c2
Also, b + c = 4 and a + b = 5 a
Thus, sides will be 1, 2, 3, 4, 5. ca c2
Now, DE = BE − BD = −
In case (4) We have c < b + c are two consecutive b+c a
integers and b = 1. cb (b − c )
=
∴ c = 2, b + c = 3, a = 4 and a + b = 5 a (b + c )
Thus, sides will be 1, 2, 3, 4, 5 Thus, we get
In case (5) We have b < b + c are two consecutive AD b + c PS
= =
integers, so that c = 1 DE b − c SQ
∴ b = 2, b + c = 3, a = 4, a + b = 6 Since, ∠ADE = 90° = ∠PSQ, we can have
Thus, sides will be 1, 2, 3, 4, 6 which are not ∆ADE ~ ∆PSQ
consecutive integers.
Again, AD ⊥ PS
Therefore, the possible values of (a, b,c ) are (4, 2, 3),
⇒ AE ⊥ PQ
(2, 3, 1), (4, 1, 2).
We also observe that
∴ a + b + c = 9, 6 or 7
PQ 2 = PS 2 + SQ 2 = (r2 + r1 )2 + (r2 − r1 )2
Thus, possible sum AB + BC + CA = 6, 7 or 9
= 2 (r12 + r22 )
5. Draw PS||BC and QS|| AD
c + b2 2 2
2
∴The ∆PSQ is right-angled triangle with ∠PSQ = 90°. But r12 + r22 = ⋅ r = r [Q a2 = b 2 + c 2 ]
a2
A ∴ PQ = 2 r
Also, we have,
AI = r cosec ( A /2 ) = r cosec 45° = 2 r
∴ PQ = AI Hence proved.
I
Q
6. Consider the binomial expansion of (2 + 3 )n
Let (2 + 3 )n = x + y 3 …(i)
P S
and (2 − 3 )n = x − y 3 …(ii)
On multiplying Eqs. (i) and (ii), we get
B U D E V C ⇒ (4 − 3)n = x 2 − 3 y2
From figure, we have ⇒ x 2 − 3 y2 = 1
PS = r1 + r2 and SQ = r2 − r1 , where r1 and r2 are the Since, in the expansion of (2 + 3 )n , all the
inradii of ∆ABD and ∆ACD respectively. terms will be positive.
By AA similarity, ∆DAB ~ ∆ACB and ∆DCA ~ ∆ACB ∴ 2 x = (2 + 3 )n + (2 − 3 )n
Hence,
r1 c
=
r
and 2 =
b   n 
= 2  2 n +   2 n − 2 ⋅ 3 + L ≥ 4
r a r a  2  
where, r is the inradius of ∆ABC. Thus, we get
Thus, x≥2
PS r2 + r1 b + c
= = ∴ 2 x + 1 < 2 x + (2 x − 1)
SQ r2 − r1 b−c
∴(2 x − 1), 2 x, (2 x + 1) will form a triangle.
bc ca
Also, AD = h = and BE = By Heron’s formula, we have
a b+c
∆2 = 3x (x + 1) (x ) (x − 1) = 3x 2 (x 2 − 1)
b 2c 2  a2 − b 2 
Now, BD = c − h = c − 2 = c 2 
2 2 2 2
2  = 9x 2 y2 [Q x 2 − 3 y2 = 1]
a  a 
⇒ ∆ = 3x y, which is an integer.
c2 c4 area 3xy
= c2 × = Again, inradius = =
a2 a2 perimeter 3x
[Q∆ABC is right-angled triangle] = y, which is also an integer.
www.pdfworld.in
www.pdfworld.in
Solved Paper 2017
RMO
Regional Mathematical Olympiad
(Exam Held on 08-10-2017)
1. Let AOB be a given angle less than 180° and let P be an interior point of the angular region
determined by ∠ AOB. Show, with proof, has to construct, using only ruler and compasses, a line
segment CD passing through P such that C lies on the ray OA and D lies on the ray OB and
CP : PD = 1 : 2.
2. Show that the equation
a3 + (a + 1)3 + (a + 2)3 + (a + 3)3 + (a + 4)3 + (a + 5)3 + (a + 6)3 = b4 + (b + 1)4 has no solutions in
integers a , b.
1
3. Let P (x) = x2 + x + b and Q (x) = x2 + cx + d be two polynomials with real coefficients such that
2
P (x) Q (x) = Q (P (x)) for all real x. Find all the real roots of P (Q (x)) = 0.
4. Consider n 2 unit squares in the xy-plane centred at point (i , j) with integer coordinates, 1 ≤ i ≤ n,
1 ≤ j ≤ n. It is required to colour each unit square in such a way that when ever 1 ≤ i < j ≤ n and
1 ≤ k < l ≤ n, the three squares with centres at (i , k), ( j, k), ( j, l) have distinct colours. What is the least
possible number of colours needed?
5. Let Ω be a circle with chord AB which is not a diameter. Let T1 be a circle on one side of AB such that it
is tangent to AB at C internally tangent to Ω at D. Likewise. Let T2 be a circle on the other side of AB
such that it is tangent to AB at E and internally tangent to Ω at F. Suppose the lines DC intersects Ω
at X ≠ D and the line FE intersects Ω at Y ≠ F . Prove that XY is a diameter of Ω.
6. Let x, y, z be real numbers, each greater than 1. Prove that
x+1 y+1 z+1 x−1 y−1 z −1
+ + ≤ + +
y+1 z+1 x+1 y−1 z −1 x−1
www.pdfworld.in
www.pdfworld.in
Detailed Solutions
1. We have, So, LHS is always divisible by 7.
∠ AOB is less than 180° and P be an any interior Or
point. 7
a + (a + 1) + … (a + 6)3 = ∑r
3 3 3
(mod 7)
Therefore r =1
A 2
7(7 + 1)
=   (mod 7)
C
Q
 2 
P = (28)2 (mod 7)
= 0 (mod 7)
Now, RHS = b 4 + (b + 1)4 for any integral b.
O D B b = r (mod 7 ), where 0 ≤ r < 7
Join OP and extend OP to Q such that b 4 + (b + 1)4 = r 4 + (r + 1)4 ≡/ 0 (mod 7)
OP 2
= …(i) for any r = 0, 1, 2, 3, 4, 5, 6
PQ 1
∴ b 4 + (b + 1)4 is not divisible by 7
Draw a line through Q parallel to OB which meets
OA at C. Hence, no solution for any integral value of a and b.
1
Join CP and extend to meet OB at D. CD is 3. We have, P(x ) = x 2 + x + b …(i)
required line. 2
In ∆CPQ and ∆ DPO and Q(x ) = x 2 + cx + d …(ii)
∠CQP = ∠POD Let P(x ) = 0
[Qalternate interior angles as CQ||OD] ∴ P(x ) Q(x ) = Q(P(x ))
∠QCP = ∠ PDO 0 ⋅ Q(x ) = Q(0)
[Qalternate interior angles as CQ||OD] ⇒ Q(0) = 0
∠ DPO = ∠CPQ [vertically opposite angles] ⇒ d=0
∴ ∆ CPQ ~ ∆ DPO [by AAA similarity criterion] ∴ Q(x ) = x 2 + cx [from eq. (ii) and d = 0]
CP PQ 1 Now, P(x ) . Q(x ) = Q(P(x )) = (P(x ))2 + cP(x )
∴ = = [from eq. (i)]
DP OP 2
[Qd = 0]
∴ PC : PD = 1 : 2
⇒ Q(x ) = P(x ) + c [dividing both sides by P(x )]
2. We have, 1
⇒ x + cx = x + x + b + c
2 2
a3 + (a + 1)3 + (a + 2 )3 + (a + 3)3 + (a + 4)3 2
+ (a + 5)3 + (a + 6)3 = b 4 + (b + 1)4 1
⇒ cx = x + b + c
2
Since, 7 consecutive number appears on left
side, therefore apply modulo of On comparing the coefficient of x and constant
terms both sides, we get
1 +2 + 3 + 4 + 5 + 6 +7
3 3 3 3 3 3 3
1
⇒ c = , b + c = 0,
= 1 + 8 + 27 + 64 + 125 + 216 + 343 2
When divide by 7, we get 1
∴ b=−
1(mod 7) + 1(mod 7) + (− 1) (mod 7) 2
+ 1(mod 7) −1(mod 7) −1(mod 7) + 0 (mod 7) 1 1
Hence, P(x ) = x 2 + x − [from eq (ii)]
2 2
= (1 + 1 − 1 + 1 − 1 − 1 + 0) (mod 7)
1
= 0 (mod 7) and Q(x ) = x 2 + x …(iii)
2
www.pdfworld.in
www.pdfworld.in

3
Now, since P(Q(x )) = 0, Q(x ) is root of P(x ) = 0 Therefore, all square of type in Ist column
x 1 i.e. (x, 1) for x = 1, 2, 3, …, n are of distinct colour
i.e. x2 + − = 0
2 2 ⇒ total n distinct colours.
(x + 1)  x −  = 0
1
⇒ Similarly all square of type (x, y) for y = 1, 2, 3, …, n
 2 are of distinct colours, also no square of (x, 1) and
1 (x, y) with same colour otherwise take square
x = − 1,
2 (x, x ) with (x, 1) and (x, y), we will get contradiction
1 to given condition.
∴ Q(x ) has to be either −1or .
2 There will be atleast (2 n − 1) colours.
Case 1. Q (x ) = − 1 Now, let us prove 2 n − 1 colours are sufficient.
1
⇒ x2 + x + 1= 0 [from eq. (iii)]
2
(x, y)
⇒ 2 x 2 + x + 2 = 0 [multiplying both sides by 2] y
− 1± 15 i
⇒ x= y–1
(x+1, y – 1)
4
[by quadrotic formula]
1
Case 2. Q(x ) = x x+1 x+y=c=(i+k)
2
1 1
⇒ x2 + x − = 0 [from eq (iii)] We can see that (x, y) and (x + 1, y − 1) can have
2 2 same colours. So, point all squares with same
(x + 1) x −  = 0
1 colour for which x + y is same.
⇒
 2 Here, minimum x + y will be 2(for x = 1, y = 1) and
1 maximum x + y = 2 n
⇒ x= ,−1
2 As from 2 to 2n three 2 n − 1 values.
So total 4 roots, 2 real and 2 imaginary. So, 2 n − 1 colours will be sufficient.
1
Real roots are , − 1. 5. Let O1, O and O2 be the centres of circles T1, Ω and
2
T2 respectively.
4. We have n2 unit squares in the x y- plane.
Join OY it to meet AB at M. Join OO2 it will passes
n
through F (as both circles are tangent)
Y
(i , l ) D
l
90–θ
(j, k) O1
k
(i , k )
A E
B
M C
2 θ
1 2 3 i j n
θ
90–
Here, 1≤ i < j ≤ n O
–θ
1≤ k < l ≤ n 90 O2
F
According to problems
(i , k ), ( j , k ), ( j , l ) are of distinct colours in the
kth row no two square will be same colour as (i , k )
and ( j , k ) are of distinct colours similarly in
i th column no two square of same colour as ( j , l ) X
and ( j , k ) are of distinct colour.
www.pdfworld.in
www.pdfworld.in
4 Regional Mathematical Olympiad
Now let ∠ AEF = θ 6. We have x, y, z are real numbers and each

∴ ∠O2 EF = 90° − ∠ AEF greater than 1
⇒ ∠O2 EF = 90° − θ Let x ≥ y ≥ z > 1
Since, O2 E and O2 F are radius of circle T2 ⇒ x 2 − 1 ≥ y2 − 1 ≥ z2 − 1 > 0
∴ ∠O2 EF = ∠O2 FE = 90° − θ On taking reciprocal of each term except zero, we
[Qangles opposite to equal get
1 1 1
sides are equal] ⇒ ≥ ≥ >0
Also ∠OFY = ∠OYF z2 − 1 y2 − 1 x 2 − 1
[OY = OF radii of same circle O] [Qsign of inequalities change due to reciprocal]
∠ FOX = 2 ∠OYF Also, x − y ≥ 0, y − z ≥ 0
x− y x− y
[Qangle subtended by an arc at the ∴ ≥
centre of the circle is thrice the angle y2 − 1 x 2 − 1
subtended by the same arc at y− z y− z
and ≥
any point on the remaining part z2 − 1 x2 − 1
of the circle]
x− y y− z x− y y− z
∴ ∠ FOX = 2(90° − θ) ⇒ + ≥ +
y2 − 1 z2 − 1 x2 − 1 x2 − 1
= 180° − 2θ
x− y y− z x− z
In ∆OFX ⇒ + ≥
y −1
2
z −1
2
x2 − 1
∠OFX + ∠OXF + ∠ FOX = 180°
x− y y− z z−x
[by angle sum property of triangle] ⇒ + + ≥0
y2 − 1 z2 − 1 x2 − 1
⇒ ∠OFX + ∠OFX = 180°−∠FOX
1  x − 1 x + 1 1  y − 1 y + 1
[Q∠OFX = ∠OXF as OF = OX, ⇒  − + −
2  y − 1 y + 1 2  z − 1 z + 1
radii of same circle]
⇒ 2 ∠OFX = 180°− 180° + 2θ 1  z − 1 z + 1
+  − ≥0
2  x − 1 x + 1
⇒ ∠OFX = θ
x −1 x + 1 y−1 y+ 1
∴ ∠ XFY = ∠OFX + ∠OFY ⇒ − + −
y−1 y+ 1 z−1 z+ 1
= θ + 90° − θ = 90° z−1 z+ 1
[Q∠OFY = ∠O2 FE = 90°−θ] + − ≥0
x −1 x + 1
∠ XFY = 90° x −1 y−1 z−1 x + 1 y+ 1 z+ 1
⇒ + + ≥ + +
∴XY is a diameter of circle O. y−1 z−1 x −1 y+ 1 z+ 1 x + 1
www.pdfworld.in
www.pdfworld.in
Solved Paper 2018
INMO
Indian National Mathematical Olympiad
(Exam Held on 21-01-2018)
1. Let ABC be a non-equilateral triangle with integer sides. Let D and E be respectively the mid-points
of BC and CA; let G be the centroid of triangle ABC. Suppose D , C , E , G are concyclic. Find the least
possible perimeter of triangle ABC.
2. For any natural number n, consider a 1 × n rectangular board made up of n unit squares. This is
covered by three types of tiles; 1 × 1 red tile, 1 × 1 green tile and 1 × 2 blue domino. Let tn denote the
number of ways of covering 1 × n rectangular board by these three types of tiles. Prove that tn divides
t2n + 1.
3. Let u1 and u 2 be two circles with respective centres O1 and O2 intersecting in two distinct points A and
B such that ∠ O1 AO2 in an obtuse angle. Let the circumcircle of triangle O1 AO2 intersect u1 and u 2
respectively in points C and D. Let the line CB intersect u 2 in E; let the line DB intersect u1 in F. Prove
that the points C , D , E , F are concyclic.
4. Find all polynomials with real coefficients P (x) such that P (x2 + x + 1) divides P (x3 − 1).
5. There are n ≥ 3 girls in a class sitting around a circular table, each having some apples with her. Every
time the teacher notices a girl having more apples than both of her neighbours combined, the teacher
takes away one apple from that girl and gives one apple each to her neighbours. Prove that this
process stops after a finite number of steps. (Assume that the teacher has an abundant supply of
apples.)
6. Let N denote the set of all natural numbers and let f : N → N be a function such that
(a) f (mn ) = f (m) f (n ) for all m, n in N
(b) (m + n ) divides f (m) + f (n ) ; for all m, n in N
Prove that there exists an odd natural number k such that f (n ) = n k for all n in N.
www.pdfworld.in
www.pdfworld.in
Detailed Solutions
1. We have, 2. Consider a 1 × (2 n + 1) board and imagine the board
ABC is non-equilateral triangle. to be placed horizontally.
D and E are mid-point of BC and AC Let us label the squares of the board as
respectively. C − n , C −( n − 1) , …, C −2 , C −1, C 0 , C1, C 2 , … C n − 1, C n
A from left to right. The 1 × 1 tiles will be referred to as
red and green tiles, and the blue 1 × 2 tile will be
referred to as a domino.
Let us consider the different ways in which the
centre square C 0 can be covered.
E
There are four different ways in which this can be
G done.
(i) Let blue domino covering the square C −1, C 0 . In
this case there is 1 × (n − 1) board remaining on
B D C left side which can be covered in t n − 1 ways and
1 × n board remaining on right side which can be
DCEG are concyclic covered in t n ways.
∴ AG ⋅ AD = AE × AC (ii) If blue domino covering the square C 0 , C1, then
2 1 1 × (n − 1) board remaining on right side which
⇒ AD . AD = AC. AC
3 2 can be covered in t n − 1 ways and (1 × n) board
2 remaining on left side which can be covered in
[Q G is centroid of ∆ ABC, AG = AD t n ways.
3
Also E is mid-point of AC] (iii) If red tile covering the square C 0 . In this case
3 there is 1 × n board remaining on both sides of
⇒ 2 AD = AC 2
2
this tile, each can be covered in t n ways.
2
(iv) If green tile covering the square C 0 . In this case,
In ∆ ABC by Appolonius theorem, we get
there is 1 × n board remaining on both sides of
BC 2
AB2 + AC 2 = 2 AD2 + this tile each can be covered in t n ways.
2 Now, putting all the possibilities mentioned
3 BC 2 above together, we get
⇒ AB2 + AC 2 = AC 2 +
2 2 t 2 n + 1 = t n − 1 ⋅ t n + t n − 1 ⋅ t n + t n2 + t n2
⇒ AC 2 + BC 2 = 2 AB2
= 2(t n − 1 t n + t n2 )
2 2
AC − BC   AC + BC  = AB2
⇒   +   ⇒ t 2 n + 1 = 2t n (t n − 1 + t n )
 2   2 
which implies that t n divides t 2 n + 1
AC − BC AC + BC
Thus  , , AB are 3. We will first prove that C, B, O2 , E are collinear and
 2 2 
pythagorean triplet. Consider the first triplet (3, this line is bisector of ∠ACD
4, 5) this gives AC = 7, BC = 1, AB = 5. But A
AC, BC and AB are not forming a triangle. E
F
The next triplet is (5, 12, 13), we get
AC = 17, BC = 7 and AB = 13. In this case we O1 x O2
get a triangle and its perimeter is
17 + 7 + 13 = 37. B
Since 2 AB < AB + BC + CA < 3 AB. It is
C D
sufficient to verify upto AB = 19.
www.pdfworld.in
www.pdfworld.in

7
Let ∠ ABO2 = x Thus we have

Then, ∠ AO2 B = 180° − 2 x β 3 − 1 + γ 3 − 1 ≥ β 3 − 1+ γ 3 − 1
O2O1 is angle bisector of ∠ AO2 B.
1 = α β + γ −2 = 3 α
Q ∠ AO2O1 = ∠ AO2 B = 90° − x [Qβ + γ = −1]
2
Since A, O1, C, O2 are concylic. This show that the absolute value of atleast one of
β 3 − 1 and γ 3 − 1 is not less than 3 α / 2
∴ ∠ AO2O1 = ∠ ACO1 = 90° − x
⇒ ∠ AO1C = 180° − 2(90° − x ) = 2 x It we take this as α1, we have α1 > α
From this Now, α1 is a root of P(x ) = 0 and we repeat the
1 1 argument with α1 in place of α. We get infinite
∠ AFC = ∠ AO1C = (2 x ) = x sequence of distinct roots of P (x )= 0, which is not
2 2
possible for any polynomial. This contradiction
and ∠ ABC = 180° − x proves that all the roots of non-constant
Thus ∠ ABC and ∠ ABO2 are supplementary. polynamial must be ‘O’.
Hence C, B, O2 , E are collinear. Now, O2 A = O2 D ⇒ P(x ) = ax n , a ∈ R, n ∈ N.
implies that O2 is the mid-point of arc AO2 D.
Hence, CO2 is the bisector of ∠ ACD. Similarly we or P(x ) = C, c ∈ R − {0}
obtain that D, B, O1, F are collinear. Thus, BE and 5. We have n≥ 3 girls in a class sitting around a
BF are diameters of the respective circles. circular table, each having some apples with her.
⇒ ∠ BAE =∠ BAF = 90°; Let i th girl have ai apples at any given time.
⇒ FAE are collinear. Consider two quantities with this distribution.
Finally, we get ∠ ECD = ∠ BCD
s = a1 + a2 + a3 +…+ an and
= ∠ ACB = ∠ AFB = ∠ EFD
t = a12 + a22 + a32 +…+ an2
Therefore C, D, E, F are concylic.
Using Cauchy - Schwartz inequality, we get
4. If P(x ) is constant , say C, then P(x 2 + x + 1) = C
(a1 + a2 + a3 +… an )2
and P(x 3 − 1) = C, where C ≠ O. Thus, in this case a12 + a22 + a32 +… an2 ≥
n
P(x 2 + x + 1) divides P(x 3 − 1).
s2
If P(x ) is non-constant polynomial, then ⇒ t≥
n
P(x 3 − 1) = P(x 2 + x + 1)⋅Q(x ), where Q(x ) is some
s2
polynomial in x. Therefore, t ≥ at any stage of the above
n
⇒ P((x − 1)(x 2 + x + 1)) = P(x 2 + x + 1) ⋅ Q(x ) process. Whenever teacher makes a move, s
⇒ Whenever x 2 + x + 1 is a root of P(x ), then increases by 1. Suppose the girl with am apples
(x − 1)(x 2 + x + 1) is also a root. has more than the sum of her neighbours. Then
the change in t equals to
We prove that the only root of p(x )= 0 is O
(am − 1)2 + (am −1 + 1)2 + (am + 1 + 1)2 − am
2
− am
2
−1 − am + 1
2
Suppose that there is a root ‘α’ with α > 0
= (am − 1)2 − am
2
+ (am −1 + 1)2 −
Now, take x 2 + x + 1=α
−1 + ( am + 1 + 1) − a m + 1
2 2 2
am
Let β and γ are root of x 2 + x + 1− α = 0
Then, β + γ = −1 = (am − 1+ am )(am − 1− am )
+ (am −1 + 1+ am −1 )(am −1 + 1− am −1 )
Since, P (x + x + 1) divides P (x − 1)
2 3
+ (am + 1 + 1+ am + 1 )(am + 1 + 1− am + 1 )
∴ P(β 3 − 1) = 0, P(γ 3 − 1) = 0
= − 2 am + 1+ 2 am −1 + 1+ 2 am + 1 + 1
We also see that = 2 (am + 1 + am −1 − am ) + 3
β 3 − 1+ γ 3 − 1= (β − 1)(β 2 + β + 1) + (γ − 1)(γ 2 + γ + 1) = 3 + 2 (am + 1 + am −1 − am ) ≤ 3 + 2 (−1) = 1
⇒β 3 − 1+ γ 3 − 1 = (β − 1)(α ) + (γ − 1)α = α (β + γ −2 ) [Qam > am −1 + am + 1]
[Qβ + β + 1= γ + γ + 1= α]
2 2
www.pdfworld.in
www.pdfworld.in
If s1 and t 1 denote the corresponding sums after 2 n + 1 divides f(2 n) + f(1)

one move, we observe that ⇒ (2 n + 1) divides f(2 ) f(n) + 1
s1 = s + 1 and t 1 ≤ t + 1 [Qf(2 n) = f(2 ) f(n) and f(1) = 1]
Thus after teacher performs k moves, if the This shows that
corresponding sums and sk and t k , we get
HCF of f(2 ) and (2 n + 1) is 1 for all n.
sk = s + k, t k ≤ t + k
Since, 2 n + 1 is an odd number.
sk2 (s + k )2
tk ≥ ⇒ t + k≥ ∴f(2 ) must be even number i.e. f(2 ) = 2 k for some
n n
natural number k.
⇒ nt + nk ≥ s 2 + k 2 + 2 sk
Since 3 = 1 + 2 divides f(1) + f(2 ) = 1 + 2 k , k is odd
⇒ k 2 + 2 sk − nk + s 2 − nt ≤ 0
Now take any arbitrary power of 2, say 2 m , and an
⇒ k 2 + (2 s − n)k + s 2 − nt ≤ 0 arbitrary integer n.
Since this cannot hold for large k, we see that the ∴ 2 m + n divides f(2 m ) + f(n)
process must stop at some stage. ⇒ 2 m + n divides (f(2 ))m + f (n)
6. We have, [Qf(mn) = f(m) (n) f(2 m ) = f(2 ) ⋅ f(2 ) … m times]
f(mn) = f(m) f(n), m, n ∈ N [Qf(2 ) = 2 k ]
Putting m = n = 1, we get Thus 2 m + n divides 2 km + f(n)
f(1) = f(1) f(1)
But 2 km + f(n) = 2 km + nk + f(n) − nk
⇒ f(1)2 − f(1) = 0
= M (2 m + n) + f(n) − nk , k is odd
⇒ f(1) (f(1) − 1) = 0
This means 2 m + n divides f(n) − nk .
⇒ f(1) = 1, f(1) ≠ 0
Now, put m = 2 and n = n, we get By varying m over N, we conclude that
f(2 n) = f(2 ) f(n) f(n) − nk = 0
Also given (m + n) divides (f(m) + f(n)) ∴ f(n) = nk from all n ∈ N.
∴ Putting m = 2 n and n = 1, we get
www.pdfworld.in
www.pdfworld.in
Solved Paper 2018
RMO
Regional Mathematical Olympiad
Max Marks : 102 (Exam Held on 07-10-2018) Time : 3 hours
1. Let ABC be a triangle with integer sides in which AB < AC. Let the tangent to the circumcircle of
triangle ABC at A intersect the line BC at D. Suppose, AD is also an integer. Prove that
gcd ( AB, AC ) > 1
2. Let n be a natural number. Find all real numbers x satisfying the equation
n
kxk n (n + 1)
∑ 1+ x
k =1
2k
=
4
3. For a rational number r, its period is the length of the smallest repeating block in its decimal
expansion. For example, the number r = 0.123123123.... has period 3. If S denotes the set of all
rational number r of the form r = 0 ⋅ abcdefgh having period 8, find the sum of all the elements of S.
4. Let E denote the set of 25 points (m,n) in the xy-plane, where m, n are natural numbers, 1 ≤ m ≤ 5,
1 ≤ n ≤ 5. Suppose the points of E are arbitrarily coloured using two colours, red and blue. Show that
there always exist four points in the set E of the form (a , b), (a + k, b), (a + k, b + k), (a , b + k) for some
positive integer k such that at least three of these four points have the same colour (i.e. there always
exist four points in the set E which form the vertices of a square with sides parallel to axes and having
at least three points of the same colour.)
5. Find all natural number n such that 1 + [ 2n ] divides 2n. (For any real number x, [x] denotes the
largest integer not exceeding x.)
6. Let ABC be an acute-angled triangle with AB < AC. Let I be the incentre of triangle ABC, and let
D , E and F be the points at which its incircle touches the sides BC , CA and AB respectively. Let BI , CI
meet the line EF at Y , X , respectively. Further assume that both X and Y are outside the triangle
ABC. Prove that
(i) B, C ,Y , X are concyclic, and
(ii) I is also the incentre of triangle DYX.
www.pdfworld.in
www.pdfworld.in
Detailed Solutions
1. c2
=
A  a + b −c 
2 2 2
 a2 + c 2 − b 2 
b  − c 
 2 ab   2 ac 
4
2
c 2a
=
1 O b −c22
3
6 5 Also, we know that
D B C DA 2 = DB × DC
c 2a  c 2a 
We know that, =  + a
b − c 2  b2 − c 2
2

∠ 1 = 2 ∠C
[Angle subtended at the centre by a chord is double [QDC = DB + BC]
the angle subtended by it at remaining part] abc abc
∴ DA = 2 =
Now, OA = OB b − c 2 (b + c )(b − c )
⇒ ∠2 = ∠3 Since, DA is integer and let us assume b and c are
π abc
∴ ∠2 = ∠3 = − ∠C [QOA ⊥ AD] coprime, then ∈ Z, if (b − c )
2 (b + c )(b − c )
Again , ∠4 = 90 ° − ∠2 and (b + c ) both are divisible by a, which is not
⇒ ∠4 = 90 ° − (90 ° − ∠C ) = ∠C possible as b + c > a (triangle inequality).
Also, ∠5 = 180 ° − ∠B ∴b and c can’t be coprime.
In ∆ABD, Hence, GCD( AB, AC ) > 1 [by contradiction]
∠4 + ∠5 + ∠6 = 180 ° 2. We have
⇒ ∠6 = 180 ° − [∠C + 180 ° − ∠B]
n
kx k n(n + 1)
⇒ ∠ 6 = ∠B − ∠C
∑1+ x 2k
=
4
k =1
In ∆ABD, x 2x 2 3x 3 nx n
⇒ + + + ......... +
Using sine formula, 1+ x 2
1+ x 4
1+ x 6
1 + x 2n
DB AB
= n(n + 1)
sinC sin D =
4
DB c
⇒ = [QIn ∆ABC, AB = c] ⇒ −1
1
+ −2
2
+
3
+ ...... + − n
n
sinC sin(B − C ) x + x x + x 2 x −3 + x 3 x + xn
c sinC n(n + 1)
⇒ DB = = …(i)
sin Bcos C − cos BsinC 4
c ⋅ Kc Case I When x > 0
⇒ DB =
Kb ⋅ cos C − cos B ⋅ Kc We know that, AM ≥ GM
[In ∆ABC, by sine formula ∴ x − k + x k ≥ 2 x − kx k ≥ 2 …(ii)
= k (let )
sin A sin B sinC
= = From Eqs. (i) and (ii), we get
a b c 
1 2 3 n n k n(n + 1)
c2 + + + .... + ≤ Σ ≤
= 2 2 2 2 k =1
2 4
b cos C − c cos B n
kx k n(n + 1)
By cosine formula, in ∆ABC ∴∑ = is only possible when, x k = 1,
k = 11 + x
2k
4
a2 + b 2 − c 2
cos C = ∀ k∈ N
2 ab
∴ x =1
www.pdfworld.in
www.pdfworld.in

3
Case II When x = 0 condition ensures at least one square with at least

n
kx k 3 red vertices, then obviourly, the other two
LHS = ∑ = 0, but starting conditions for first row (i.e. 4 vertices
k =1 1 + x
2k
coloured in blue or all 5 vertices coloured in blue)

n(n + 1)
RHS = ≠ 0 for any natural number n. would automatically ensure at least one square
4 with at least 3 red vertices.
So, x = 0 is not a solution. 5 x x
Case III When x < 0
4
from the case I, it is clear that LHS is less than
n(n + 1) x x x
, if x < 0. Since, the odd indexed terms 3 x x
4
would becomes negative. 2x x x
∴No negative solution x exists.
1
Hence, x = 1is only solution. 1 2 3 4 5
3. We have, B B B
B B B
r = 0. abcdefgh (1) (2)
⇒ 108 r = abcdefgh. abcdefgh x x
⇒ (108 − 1)r = abcdefgh x x x
P x
abcdefgh
⇒ r=
108 − 1 S x x
Q
1 2 999.....(8 times)
Total sum = 8 + 8 + .... +
10 − 1 10 − 1 108 − 1 x x x x
1 + 2 + 3 + K + 99K99
= ( 8 times)
108 − 1
B B B B B B
(108 − 110
) 8
= = 5 × 107 (3) (4)
2 (10 − 1)
8
This sum contains all the numbers having period

1, 2, 4 and 8.
Now, period = 4 x x x
x x
If unit digit is 1, then the next three places can be
filled in 103 ways.
∴ Sum of all the digits at unit places B B B
(1 + 2 + 3..... + 9)103 45 × 103 (5)
= =
108 − 1 108 − 1
Then the vertices marked with (x) can not be
Hence, sum of all such numbers coloured by blue, so they has to be coloured in
(1 + 10 + ... + 107 )(45 × 103 ) red, marking at least one square with at least
= = 5 × 103
108 − 1 3 red vertices (as shown in marked). In figure 3,
position p can either be red or blue. If P is red,
It include numbers with period 1 and period 2.
then we have upper left|x| square with 3 red and if
∴Required sum = 5 × 107 − 5 × 103 P is blue, then either Q or S is red, making another
= 49995000. |x| square with 3 red.
2n
4. 5 vertices of each row has to be coloured using 5. Let =x
exactly two colours. We shall start with first row in 1
which 3 vertices have one colour say blue and ⇒ 2n = x …(I)
rest 2 vertices have other colour, red. If this initial
www.pdfworld.in
www.pdfworld.in
4 Regional Mathematical Olympiad
We know that, We have, ∆AIF ~

= ∆AIE
a − 1 < [ a] ≤ a ∴ ∠AIF = ∠AIE = 90 ° −
A
∴ 2n − 1≤ 2n 2
⇒ x [ 2 n − 1] < x ≤ x 2 n ∴ ∠EIF = 180 ° − A
⇒ x [ 2 n − 1] < 2 n − x ≤ x 2 n [from eq (i)] ∴ ∠IEF = ∠IFE =
A
Let 2n = t 2
∴xt − x < t 2 − x and t 2 − x ≤ xt ∴ ∠FEC = ∠FEI + ∠IEC
⇒ xt < t and t − xt ≤ x
2 2
A
= + 90 °
2
x2 2
x < t and  t −  ≤ x +
x
⇒ [Qt = 2 n > 0]
 2 4 Now, in ∆XEC,
2 2
∠EXC + ∠XEC + ∠ECX = 180 °
x < t and  t −`  <  + 1
x x
⇒
 2 2  A C
⇒ ∠EXC + + 90 ° + = 180 °
x x 2 2
⇒ x < t and t − < + 1
2 2 A C
⇒ ∠EXC = 90 ° − −
⇒ x < t and t < x + 1 2 2
⇒ t − 1< x < t [Qt = 2 n] 180 ° − ( A + C )
=
⇒ =x …(II) 2
180 ° − (180 ° − B) B
from Eqs. (i) and (ii), we get = =
2 2
2 n = x + x2
x(x + 1) B
⇒ n= , x ∈ N. ∴ ∠EXC = ∠EBC =
2 2
6. ∴BCYX are concyclic .
A
IDCE is a cyclic quadrilateral.
∴ICYE is also a cyclic quadrilateral.
A/2 A/2
Hence I, D, C, E, Y lie on same circle.
∴ ∠ICD = ∠IYD
C
X F E Y ⇒ ∠IYD =
B/2 C/2 2
B
Similarly, ∠I XO =
2
I
∴IX, IY , ID are angle bisector.
B/2 C/2 ∴ I is also incentre of ∆DYX.
B/2 C/2
B D C
www.pdfworld.in
www.pdfworld.in
Solved Paper 2019
INMO
Indian National Mathematical Olympiad
Max Marks : 102 (Exam Held on 20-1-2019) Time : 4 hours
1. Let ABC be a triangle with ∠BAC > 90°, let D be a point on the segment BC and E be a point on the
line AD such that AB is tangent to the circumcircle of triangle ACD at A and BE is perpendicular to
AD. Given that CA = CD and AE = CE, determine ∠BCA in degrees.
2. Let A1 B1C1 D1 E1 be a regular pentagon. For 2 ≤ n ≤ 11, let An BnC n Dn E n be the pentagon whose vertices
are the midpoints of the sides of An −1 Bn −1C n −1 Dn −1 E n −1 . All the 5 vertices of each of the 11 pentagons
are arbitrarily coloured red or blue. Prove that four points among these 55 points have the same
colour and form the vertices of a cyclic quadrilateral.
3. Let m, n be distinct positive integers. Prove that,
gcd (m, n ) + gcd (m + 1, n + 1) + gcd (m + 2, n + 2) ≤ 2|m − n| + 1.
Further, determine when equality holds.
4. Let n and M be positive integers such that M > n n − 1 . Prove that there are n distinct primes
p1 , p2 , p3 ,..... pn such that pj divides M + j for 1 ≤ j ≤ n.
5. Let AB be a diameter of a circle Γ and let C be a point on Γ different from A and B. Let D be the foot of
perpendicular from C on to AB. Let K be a point of the segment CD such that AC is equal to the
semiperimeter of the triangle ADK.Show that the excircle of triangle ADK opposite A is tangent to Γ.
6. Let f be a function defined from the set {(x, y) : x, y real, xy ≠ 0} to the set of all positive real numbers
such that
(i) f (xy, z ) = f (x, z ) f ( y, z ), for all x, y ≠ 0
(ii) f (x, 1 − x) = 1, for all x ≠ 0, 1.
Prove that,
(a) f (x, x) = f (x, − x) = 1, for all x ≠ 0;
(b) f (x, y) f ( y, x) = 1, for all x, y ≠ 0.
www.pdfworld.in
www.pdfworld.in
Detailed Solutions
A If we consider any two sides which are parallel,
1.
they are the parallel sides of an isosceles
2α 90–α
trapezium, which is cyclic.
If we consider any pentagon, its two adjacent
90–α 2α vertices have the same colour. Consider all such
B C 11 sides whose end points are of the same
D
colour. These are in 5 fixed directions. By
Pigeon-hole principle, there are 3 sides which are
in the same directions and therefore parallel to
90º 2α
each other. Among these three sides, two must
E have end points having one colour (again by P-H
principle). Thus, there are two parallel sides
Let ∠C = 2α. Then, ∠CAD = ∠CDA = 90° − α .
among the 55 and the end points of these have
Moreover ∠BAD = 2α as BA is tangent to the
one fixed colour. But these two sides are parallel
circumcircle of ∆CAD. Since AE = CE. It gives
sides of an isosceles trapezium. Hence the four
∠AEC = 2α . Thus ∆AEC is similar to ∆ACD.
end points are concyclic.
Hence,
AE AC 3. We know that, for a > b
= .
AC AD gcd (a, b ) = gcd (a, a − b ).
But the condition that BE ⊥ AD gives AE = ABcos2α ∴ let gcd (m, n) = gcd (m, m − n) = x
= ccos2α . It is easy to see that ∠B = 90 ° − 3α . ⇒ m − n = αx …(i)
Using sine rule in triangle ADC, we get
AD AC gcd (m + 1, n + 1) = gcd (m + 1, m − n) = y
= .
sin2α sin(90 −α ) ⇒ m − n = βy …(ii)
and gcd (m + 2, n + 2 ) = gcd (m + 2, m − n) = z
This gives AD = 2 b sinα . Thus we get,
⇒ m − n = γz …(iii)
b 2 = AC 2 = AE ⋅ AD = (c cos 2α ) ⋅ 2 b sinα .
Using, b = 2 R sin B and c =2 R sinC, this leads to ∴ gcd (x, y) = 1, gcd ( y, z) =1and gcd (x, z) = 1or 2.
Eq. (i) Now, let gcd of (x, y) =1, gcd ( y, z) = 1
cos 3α = 2 sin2α cos 2α sinα = sin 4α sinα
Writing cos 3α = cos(4α − α ) and expanding, we
and gcd (x, z) = 1
get, cos 4α cos α = 0. Therefore, α = 90 ° or ⇒ m − n is divisible by xyz
4α = 90 °. But α = 90° is not possible as ∠C = 2α . ⇒ m − n ≥ xyz
Therefore 4α = 90°, which gives ∠C = 2α = 45° . ⇒ 2(m − n) + 1 ≥ 2 xyz + 1 …(iv)
D1
2. If x = y = z = 1, then
D2 C3 2 xyz + 1 = 3 = x + y + z
C2
Again, m and n are two consecutive integers.
C4 B5 B4
E1
∴ 2(m − n) + 1 = 3
C1
D3
C5 A5
B3 ∴ x + y + z = 2(m − n) + 1
O If atleast one of x, y, z ≠ 1, let z > 1
D4 A4
B2 ⇒ z≥ 2
E2
D5 E5 ∴ 2 xyz + 1 = 4xyz + 1 > x + y + z
E3 E4 A3 ⇒ x + y + z < 2(m − n) + 1
A1 B1 Eq. (ii) Now, gcd (x, y) = 1, gcd ( y, z) = 1
A2
and gcd ( z, x ) = 2
We first observe that all the eleven pentagons are
⇒ z = 2 p, x = 2q for relative prime p and q
regular. Moreover, there are 5 fixed directions and
all the 55 sides are in one of these directions. ∴ m − n ≥ 2 pqy
www.pdfworld.in
www.pdfworld.in

7
⇒ 2(m − n) + 1 ≥ 4 pqy + 1 a+ b+c

r= − a =x −a
Again, equality satisfy only when p = q = y = 1. 2
If x = 2 = z and y = 1only when m and n are two Again, AC 2 = AD × AB
consecutive even integers. ⇒ x 2 = a × 2R …(i)
If y > 1 ⇒ y ≥ 2, then Also, OE = | AD + DE − AO| = |a + r − R|
4 pqy + 1 > 2 p + 2q + y = x + y + z = |x − R|
∴ 2(m − n) + 1 > x + y + z Now, in ∆OIE
⇒ x + y + z < 2(m − n) + 1. OI 2 = OE 2 + EI 2
4. If some number M + k, 1≤ k ≤ n, has at least n ⇒ OI 2 = (x − R )2 + r 2
distinct prime factors, then we can associate a ⇒ OI 2 = x 2 + R 2 − 2 xR + r 2
prime factor of M + k with the number M + k ⇒ OI 2 = R 2 + r 2 + 2 aR − 2 xR
which is not associated with any of the remaining ⇒ OI 2 = R 2 + r 2 − 2 R(x − a) [From Eq. (i)]
n − 1numbers.
⇒ OI 2 = R 2 + r 2 − 2 rR
Suppose, M + j has less than n distinct prime
factors. Write ⇒ OI 2 = (R − r )2
α α
M + j = p1 1 p2 2 .... prα r , r < n. ⇒ OI = R − r.
n −1
But M + j > n . Hence, there exist t,1≤ t ≤ r ∴The two circles touch each other internally.
such that ptαt > n . Associate pt with this M + j . 6. We have,
Suppose pt is associated with some M + l. Let ptβ t f(xy, z) = f(x, z) ⋅ f( y, z)
be the largest power of pt dividing M + l. Then, Put x = y = 1, we get
ptβ t > n. Let T = gcd ( ptα t , ptβ t ). Then T > n. Since f(1, z) = f(1, z) ⋅ f(1, z)
T | (M + j ) and T | (M + l ), it follows that T|( j − i )|. ⇒ f(1, z) = (f(1, z))2
But| j − i|< n and T > n, and we get a contradiction.
This shows that pt cannot be associated with any ⇒ f(1, z) = 1 ∀ z ≠ 0. ...(i)
other M + l. Thus each M + j is associated with Again, put x = y = − 1, we get
different primes. f(1, z) = f(−1, z) ⋅ f(−1, z)
5. A ⇒ 1 = (f(−1, z))2
⇒ f(−1, z) = 1 ∀ z ≠ 0 ...(ii)
c a
from Eq. (i)
K b D
C f(1, x ) = 1 ∀ x ≠ 0
O ⇒ lim f(x 1 / 2 x , x ) = 1
n→ ∞
n− 1
I E ⇒ lim f(x 1 / 2 , x) = 1
n→ ∞
⇒ f( x , x ) = 1
1/ 2
⇒ f(x, x ) = 1 ∀ x ∈ (0, ∞ )
B Similarly from Eq. (ii)
f(1, − x ) = 1 ∀ x ≠ 0
Let AD = a, KD = b, AK = c, AC = x , EI = r, n
AB = 2 R ⇒ lim f(x 1 / 2 ,− x ) = 1
n→ ∞
Now, we have ⇒ f(x, − x ) = 1 ∀ x ∈ (0, ∞ )

a + b +c
AC = x = ∴ f(x, x ) = f(x,− x ) = 1 ∀ x ≠ 0
2
Again,
Again, Ex-radius of
1 = f(xy, xy) = f(x, xy) f( y, xy)
KD + AK − AD
∆ADK = r = = f(x, x ) f(x, y) f( y, x ) f( y, y)
2
b + c − a b + c + a − 2a ⇒ f(x, y) ⋅ f( y, x ) = 1 ∀ x, y ≠ 0.
∴ r= =
2 2
www.pdfworld.in

Rajeev Manocha Good Quality

Uploaded by

Copyright:

Available Formats

Rajeev Manocha Good Quality

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rajeev Manocha Good Quality

Uploaded by

Copyright:

Available Formats

www.pdfworld.

ARIHANT PRAKASHAN (SERIES), MEERUT

Arihant Prakashan (Series), Meerut

ADMINISTRATIVE & PRODUCTION OFFICES

SALES & SUPPORT OFFICES

Published by Arihant Publications (India) Ltd.

/arihantpub /@arihantpub Arihant Publications /arihantpub

1. Theory of Numbers 1-94

— RMO & INMO 2016-2017 1-8

ABOUT THE EXAM

Indian National Mathematical Olympiads

Awards and Recognition

Number of Students Appearing

Month and Frequency of Examination

Craze of the Examination

Skill Sets Required

Difficulty Level of Problems

How to Prepare for Different Subjects in Stipulated Time

General Mental Set up Required for the Examination

Do’s and Don’t on the Day of Examination

How this Book is Useful for You

Basic Properties of Natural Numbers

Greatest Common Divisor

Note 1. (a, b ) = (b, a ) 2. If a|b; then (a, b ) = a

3. If d |b and d |c and d > 0, then 

For example a = 6, b = 21, c = 10

Note 1 is neither prime nor composite.

Proof Let (a , b ) = d and (b , r ) = e

Example 1 If (a, b ) = d , then  ,  = 1.

There exists integers x and y such that

Example 2 If a|bc and (a, b ) = 1, then a|c.

Example 3 Prove that every two consecutive integers are coprime.

Example 4 Show that GCD of a + b and a − b is either 1 or 2, if (a, b ) = 1.

i.e., d | 2(a, b ) [Q(ma, mb ) = m (a, b )]

Example 8 If a |c, b |c and (a, b ) = 1, then ab |c.

Example 9 If a 2 − b 2 is a prime number, show that a 2 − b 2 = a + b, where a, b are natural numbers.

∴ One of the two factors = 1

Q The only divisor of a prime number are 1 and itself.

Theorem 3 Prove that the product of any r consecutive numbers is divisible by r ! .

i.e., every product of (r − 1) consecutive integers is divisible by (r − 1)! .

Example 2 Prove that square of each odd number is of the form 8j + 1 .

Solution Let n = 2m + 1 be an odd number .

Example 3(b) If n is an integer. Prove that product n(n 2 − 1) is multiple of 6.

Example 4 If r is an integer, show that r (r 2 − 1)( 3r + 2) is divisible by 24.

Again (r − 1)r (r + 1) is divisible by 3! = 6

Example 5 If m, n are positive integers, show that (m + n )! is divisible by m ! n ! .

Example 6 If ( 4x − y ) is a multiple of 3, show that 4x 2 + 7xy − 2y 2 is divisible by 9.

Example 7 If n is an integer, prove that n (n + 1) ( 2n + 1) is divisible by 6.

Example 8 Prove that 4 does not divide (m 2 + 2) for any integer m.

Note Two important formulae.

Example 9 Prove that 8n − 3n is divisible by 5.

Example 10 If p > 1 and 2p − 1 is prime, then prove that p is prime.

Theorem 4 The number of primes is infinite.

∴ The number a must have a prime say factor p i.e., p | a.

Example 1 Every natural number other than 1 admits of a prime factor.

∴ An odd prime n is either of the form

Theorem 6 The number of divisors of a composite number n : If n is a composite number of the

Note If dk (n ) denotes the sum of k th power of divisor of n, then

Example 1 Find the sum of the cubes of divisor of 12.

Example 2 Find the number of divisor of 600.