Chapter 4

Properties of the Integers:

Mathematical Induction
1. Mathematical Induction
2. Harmonic, Fibonacci, Lucas
Numbers
3. Prime Numbers
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

The Well-Ordering Principle Any nonempty subset of Z+

contains a smallest element. (We often express this by saying
that Z+ is well ordered.)
This principle serves to distinguish Z+ from Q+ and R+.
Theorem 4.1 Finite Induction Principle or Principle of
Mathematical Induction
S(n): an open statement, n a positive integer
If (a) If S(1) is true; and /* basis, not necessarily from 1 */

(b) If whenever S(k) is true (for some k  Z ), then S(k+1)
is true; (inductive step)
then S(n) is true for all n  Z 
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Proof: Let F  {t  Z  | S (t ) is false}. We wish to prove

that F   , so to obtain a contradiction we assume that
F   . Then by the well - ordering principle, F has a
least element s. Since S (1) is true, s  1. So s > 1, and
consequently s -1  Z+ . With s -1  F (since s is the
smallest element in F ), S ( s  1) is true. By condition
(b), S (( s  1)  1)  S (s) is also true. A contradiction.

[S (n0 )  [k  n0 [ S (k )  S (k  1)]]] 

n  n0 , S (n).
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.1. For any n  Z  ,  ni1 i = n(n  1) / 2
Proof: S (n):  ni1 i = n(n  1) / 2
Induction basis: S (1) is true.
Induction hypothesis: S ( k ):  ki1 k = k (k  1) / 2 is true.
Establish the truth of S (k  1).

( Note the importance of induction basis. For example, in

Ex. 4.4, if S (n):  ni1 i = ( n 2  n  2) / 2, we also have
S ( k )  S (k  1). )
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

n
Ex. 4.5 Find a formula for  ( 2i  1)
i 1
Ans: Observe and conjecture.
n=1 1 1=12
n=2 1+3 4=22
n=3 1+3+5 9=32
n=4 1+3+5+7 16=42
n
conjecture:  ( 2i  1) =n2
i 1

then prove by induction

Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.7 the Harmonic Numbers
1 1 1
H1  1, H 2  1  , H3  1   ,
2 2 3
1 1 1
In general, Hn  1   
2 3 n
Prove n  N, (a) H n  1  n.
2
n
+
(b) If n  Z , then  H j  (n  1) Hn  n.
j 1
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.7 the Harmonic Numbers
For (a), induction hypothesis
S ( k ): H k  1  k
2
 1 1  1 1 
H k 1  1       
2  2 k k k k
2  2  1 2 2 
 1 1 
H k   
2 k
2  1 k k
2 2 
1 1
Since  , for all 1  j  2 k , it follows that
2k  j 2k

2 2   k  1 
H k 1  H k  2    H k  1  (1  k )  1  1  ( k  1)
 2k  2
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.7 the Harmonic Numbers
For (b), induction hypothesis
k
S ( k ):  H j  (k  1) H k  k .
j 1
k 1 k
 H j   H j  H k 1  ( k  1)H k  k  H k 1
j 1 j 1
1
 (k  1)[ H k 1  ]  k  H k 1
k 1
 (k  2)H k 1  (k  1)
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.8 For n  0 let An  R, where |An|=2n and the elements of
An are listed in ascending order. If r R, prove that in order to
determine whether r  An, we must compare r with no more than
n+1 elements in An.
Proof: When n=0, A0={a} and only one comparison is needed.
Assume the result is true for some k  0 and consider the case
for Ak+1.
Let Ak 1  Bk  Ck , where | Bk || Ck | 2 k and Bk < Ck
1 (1) Compare once to determine whether r  Bk or r  Ck
k+1 (2) Then compare in Bk or in Ck
k+2 Total comparisons
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.10 Prove S(n): n can be written as a sum of 3's and/or 8's
(with no regard to order) for any n  14.
basis: n=14=3+3+8 OK
induction hypothesis: k can be added up by 3's and/or 8's.

If k=...+8+..., then k+1=...+3+3+3+...

Otherwise, there are at least 5 3's in k's summands.
k+1=...3+3+3+3+3+1+...=...8+8+...
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Theorem 4.2 Finite Induction Principle--Alternate Form
Let n0 , n1  Z  with n0  n1 .
(a) If S ( n0 ), S ( n0  1), , S (n1 ) are true and
(b) If whenever S (n0 ), S (n0  1), , S (n1 ), , S ( k ) are
true for some k  Z  , where k  n1 , then S (k  1)
is also true.
then S ( n) is true for all n  n0 .
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Ex. 4.11 Prove S(n): n can be written as a sum of 3's and/or 8's
(with no regard to order) for any n  14.
Assume the truth of the statements:
S (14), S (15), , S (k  1), S ( k ) for some k  Z  ,
where k  16. And now if n  k  1, then k  17
and k  1  (k  2)  3. The truth of S (k  2)
implies the truth of S (k  1).
Chapter 4 Properties of the Integers: Mathematical Induction

4.1 The Well-Ordering Principle: Mathematical Induction

Let p0  3, p1  7 , and pn  3 pn1  2 pn 2 ,
for n  Z  where n  2. Show that pn  2 n 2  1.
Proof : S ( n ) : pn  2 n 2  1. Assume S ( 0 ), S ( 1 ), ,
S ( k  1 ), S ( k ) are all true. Consider n  k  1.
pk 1  3 pk  2 pk 1  3( 2 k  2  1 )  2( 2 k 1  1 )
 3( 2 k  2 )  3  2 k  2  2  2( 2 k  2 )  1
 2k 3  1
(recurrence relation)
Chapter 4 Properties of the Integers: Mathematical Induction

4.2 Recursive Definitions

0,2,4,6,8,10,12,...: bn=2n explicit definition

1,2,3,6,11,20,37,68,125,...:an=? implicit recursive

a n  a n1  a n 2  a n 3 definition

Examples:
factorial: (n+1)!=(n+1)(n!)
1
Harmonic Number: H n1  H n 
n1
Chapter 4 Properties of the Integers: Mathematical Induction

4.2 Recursive Definitions

Ex. 4.16 The Fibonacci Numbers
(1) F0  0, F1  1; and
(2) Fn  Fn1  Fn 2 , for n  Z  with n  2
n
 2
Prove that (a) n  Z  Fi  Fn  Fn1
i 0
n
5
 
(b) Fn    for all n  N.
 3
Chapter 4 Properties of the Integers: Mathematical Induction

4.2 Recursive Definitions

Ex. 4.16 The Fibonacci Numbers
Please check out the induction basis for yourselves first.
k
( a) Induction hypothesis:  Fi2  Fk  Fk 1 ,
i 0
k 1 k
2 2 2 2
 Fi   Fi  Fk 1  Fk  Fk 1  Fk 1 
i 0 i 0
Fk 1 ( Fk  Fk 1 )  Fk 1  Fk  2
k k 1
 5  5
( b) Fk 1  Fk  Fk 1      
 3  3
k 1 k 1 k 1 k 1
5
  8
   5 24
    5 25
    5
             
 3  3  3  9   3  9   3
Chapter 4 Properties of the Integers: Mathematical Induction

4.2 Recursive Definitions

Ex. 4.17 The Lucas Number
(1) L0  2, L1  1; and
(2) Ln  Ln1  Ln 2 for n  Z+ with n  2.

Prove
n
(a) For n  N,  Li  Ln 2  1.
i 0
(b) n  Z + Ln  Fn1  Fn1
Chapter 4 Properties of the Integers: Mathematical Induction

4.2 Recursive Definitions

Ex. 4.17 The Lucas Number
Please check out the induction basis for yourselves first.
k
( a) Assume  Li  Lk  2  1
i 0
k 1
Then  Li = Lk  2  1  Lk 1  Lk  3  1
i 0
(b) Assume Ln  Fn1  Fn1 for the integers
n = 1,2,3, . . . , k  1, k where k  2.
Then Lk 1  Lk  Lk 1  Fk 1  Fk 1 
Fk  2  Fk  Fk  Fk  2 .
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

Def. 4.1 For integers a,b, we say b divides a, and we write b|a,
if there is an integer n, such that a=bn. We say b is a divisor
of a or a is a multiple of b.

Theorem 4.3. For any a , b, c  Z

(a) 1| a  a |0 (b) [( a | b)  ( b| a )]  a   b
(c) [(a | b)  (b| c)]  a | c (d) a | b  a | bx , x  Z
(e) If x  y  z , for some x , y , z  Z, and a divides two of
the three integers, then a divides the remaining integer.
(f) [(a | b)  ( a | c)]  a |( bx  cy ), x , y  Z.
(g) For 1  i  n, let ci  Z. If a divides each ci , then
a |(c1x1  c2 x 2  cn x n ), where x i  Z for all 1  i  n.
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

proof of (f):
If a | b and a | c, then b  am, c  an for some
m, n  Z. So bx  cy  ( am) x  ( an) y  a ( mx  ny ).
If follows that a |( bx  cy ).
Ex. 4.21 Let a,b in Z so that 2a+3b is a multiple of 17. Prove that
17 divides 9a+5b.
Proof: We observe that 17| ( 2a  3b)  17|(4)(2a  3b).
Also 17|(17a  17b). Hence 17|[(17a  17b)  (8a  12b)]
 17|(9a  5b).
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

For n in Z+ where n>1, n is a prime number if n has only two
divisors, 1 and n. Otherwise n is a composite.
Lemma 4.1. If n in Z+ and n is composite, then there is a
prime p such that p|n.
Proof: If not, let S be the set of all composite integers that have no
prime divisors. If S is not empty, then by the well-ordering
principle, S has a least element m. But if m is composite,
m=m1m2 with 1<m1<m and 1<m2<m. Since m1 is not in S, m1 is
a prime or divisible by a prime, which means m is also divisible
by a prime, a contradiction.
lemma, theorem, corollary
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

Theorem 4.4 (Euclid) There are infinitely many primes.
Proof: If not, let p1 , p2 , , pk be the finite set of primes,
and let B  p1 p2  pk  1. Since B  pi for all 1  i  k ,
B cannot be a prime. Hence B is a composite. So by
Lemma 4.1 there is a prime p j and p j | B . Since p j | B
and p j | p1 p2  pk , it follows that p j |1, a contradiction.
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

Theorem 4.5 (The Division Algorithm)
If a, b  Z , with b  0, then there exist unique
q,r  Z with a  qb  r , 0  r  b.

a: dividend
b: divisor
q: quotient
r: remainder
Chapter 4 Properties of the Integers: Mathematical Induction

4.3 The Division Algorithm: Prime Numbers

Ex. 4.28 If n  Z + and n is composite, then there exists

a prime p such that p| n and p  n .
Proof: Since n is composite, we can write n  n1n2 , where
1 < n1  n and 1 < n2  n. We claim that one of the integers
n1 , n2 must be less than or equal to n . If not, then
n1  n and n2  n give the contradiction n  n1n2 
 n  n  n. Without loss of generality, assume n1  n .
If n1 is a prime, the result follows. Otherwise, by Lemma
4.1, there exists a prime p  n1 where p| n1 . So p| n and
p  n.
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Def. 4.2 For a, b in Z, a position integer c is said to be a common
divisor of a and b if c|a and c|b.
Def 4.3 Greatest Common Divisor
For any common divisor d of a and b, we have d|c. Then
c is a greatest common divisor of a and b.

Theorem 4.6 For any a,b in Z+, there exists a unique c in Z+

that is the greatest common divisor of a and b.
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Proof: Given a , b  Z + , let S = {as + bt | s, t  Z ,
as + bt > 0}. Since S   , by the Well - Ordering principle
S has a least element c. We claim that c is a greatest
common divisor of a , b. Since c  S , c = ax + by for some
x , y  Z . Consequently, if d  Z and d | a and d | b, then
d | ax + by , so d | c. If c| a , a = qc + r , 0 < r < c. Then r = a - qc
= a - q ( ax + by ) = (1 - qx )a + (-qy )b, so r  S , contradictory to
c is the least in S . Consequently, c| a and c| b. If both c1 and
c2 satisfy, then c1 | c2 and c2 | c1 , which implies c1  c2.

(existence and uniqueness)

Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

From Theorem 4.6, gcd(a,b) is the smallest positive integer we

can write as a linear combination of a and b.

Integers a and b are called relative prime when gcd(a,b)=1.

That is, when there exist x,y in Z with ax+by=1.

Ex. 4.30 Since gcd(42,70)=14, we can find x,y in Z with

42x+70y=14, or 3x+5y=1. By inspection, x=2, y=-1 is a
solution. But, general solution? x=2-5k, y=-1+3k or
x=2+5k, y=-1-3k..
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Theorem 4.7 Euclidean Algorithm
If a , b  Z + , we apply the division algorithm as follows:
a  q1b  r1 , 0  r1  b
b  q 2 r1  r2 , 0  r2  r1
r1  q 3r2  r3 , 0  r3  r2

ri  q i 2 ri1  ri 2 , 0  ri 2  ri1

rk  3  q k 1rk  2  rk  3 , 0  rk 1  rk  2
rk  2  q k rk 1  rk , 0  rk  rk 1
rk 1  q k 1rk
Then rk , the last nonzero remainder, equals gcd( a , b).
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Ex. 4.31 gcd(250,111)=?

gcd(250,111)=1, what is x, and y such
250 111
that 250x+111y=1?
*2 222
28
1=28-1(27)=28-1[111-3(28)]=
84 *3 (-1)111+4(28)=(-1)111+4[250-2(111)]
27
=4(250)+(-9)(111)
*1 27
1 x=4+111k, y=-9-250k
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Ex. 4.32 For any n in Z+, prove that 8n+3 and 5n+2 are relative
prime.
gcd(8n+3,5n+2)=1. But we could also
8n+3 5n+2 arrived at this conclusion if we had
5n+2 noticed that
3n+1 (8n+3)(-5)+(5n+2)(8)=1
3n+1
2n+1
2n+1
n
2n
1
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Ex. 4.34. Griffin has two unmarked containers. One container
holds 17 ounces and the other holds 55 ounces. Explain how
Griffin can use his two containers to have exactly one ounce.

gcd(17,55)=1, 1=13(17)-4(55), Consequently, Griffin must

fill his smaller container 13 times and empty the contents
into the larger container.
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Ex. 4.35. Diophantine equation
Find nonnegative integer solutions for 6x+10y=104.
Ans: 3x+5y=52, gcd(3,5)=1, 1=3(2)+5(-1), 52=3(104)+5(-52)
x=104-5k, y=-52+3k, 104-5k  0 and -52+3k  0. So
k=18,19,20.
Theorem 4.8. If a,b,c in Z+, the Diophantine equcation ax+by=c
has an integer solution if and only if gcd(a,b) divides c.
Chapter 4 Properties of the Integers: Mathematical Induction

4.4 The Greatest Common Divisor: the Euclidean Algorithm

Def 4.4. For a,b,c in Z+, c is called a common multiple of a,b if
c is a multiple of both a and b. Furthermore, c is the least common
multiple if it is the smallest of all positive integers that are common
multiple of a,b. We denote c by lcm(a,b).
Theorem 4.10 For a,b in Z+, ab=lcm(a,b)gcd(a,b)
Chapter 4 Properties of the Integers: Mathematical Induction

4.5 The Fundamental Theorem of Arithmetic

Lemma 4.2 If a , b  Z+ and p is a prime, then
p| ab  p| a or p| b.

Lemma 4.3 Let a i  Z for all 1  i  n. If p is a prime

and p| a1a 2  a n , then p| a i for some 1  i  n.
Ex 4.38 Show that 2 is irrational.
Proof: If not, let 2  a / b, where a , b  Z+ and
gcd( a , b) = 1. Then 2 = a 2 / b 2  2b 2  a 2  2| a 2
 2| a  a  2c  a 2  4 c 2  2b 2  4c 2  b 2  2 c 2
 2| b 2  2| b  gcd(a , b)  2, which contradicts the
earlier claim that gcd( a , b) = 1.
Chapter 4 Properties of the Integers: Mathematical Induction

4.5 The Fundamental Theorem of Arithmetic

Theorem 4.11 (The fundamental theorem of arithmetic) Every
integer n>1 can be written as a product of primes uniquely, up
to the order of the primes.
Ex. 4.41 For n in Z+, we want to count the number of positive
divisors of n.
e e e
By Theorem 4.11, n = p11 p22  pkk , where for
each 1  i  k , pi is a prime and ei  0. If m| n, then
f f f
m = p1 1 p2 2  pk k where 0  f i  ei for all
1  i  k . So by the rule of product, the number of
positive divisors of n is ( e1  1)( e2  1) ( ek  1).
Chapter 4 Properties of the Integers: Mathematical Induction

4.5 The Fundamental Theorem of Arithmetic

Ex. 4.43 Can we find three consecutive positive integers whose
product is a perfect square, that is, do there exist m,n in Z+ with
(m)(m+1)(m+2)=n2?
Sol: Suppose m,n do exist. Since gcd(m,m+1)=gcd(m+1,m+2)=1,
so for any prime p, if p|(m+1), then p|m and p|m+2. Furthermore,
if p|(m+1), then p|n2. Since n2 is a perfect square, the exponents
on p in the prime factorizations of both m+1 and n2 must be the
same even integer. So m+1 is a perfect square. And m(m+2) must
be a perfect square too. But m2<m(m+2)<m2+2m+1<(m+1)2. So
m(m+2) cannot be a perfect square. m,n do not exist.