1=1, 21  1  2  3  4  5  6

3  1 2 , 28  1  2  3  4  5  6  7
6 1 2  3, 36  1  2  3  4  5  6  7  8
10  1  2  3  4 , 45  1  2  3  4  5  6  7  8  9
15  1  2  3  4  5 55  1  2  3  4  5  6  7  8  9  10
66  1  2  3  4  5  6  7  8  9  10  11
78  1  2  3  4  5  6  7  8  9  10  11  12

91  1  2  3  4  5  6  7  8  9  10  11  12  13
105  1  2  3  4  5  6  7  8  9  10  11  12  13  14
231  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  ...  21

etc .similarly each of 21,28,and 35 is a triangular number.

 [such numbers can be arranged evenly along the sides of an
equilateral triangle]

1 is an odd and this odd multiple of 3 is 1.3=3 and its square

32 = 9=10-1 =difference of two triangular numbers.
3 is odd and this odd multiple of 3 is 3.3 and its square
3.32  9 2  81  91  10 =difference of two triangular numbers
Similarly, 5 is odd and this odd multiple of 3 is 5.3 and its square

5.32  225  231  6 ==difference of two triangular numbers

We observe:
 square of any odd multiple of 3 is difference of two triangular
numbers.

1.1.15 Well Ordering Property of (WOP) of positive integers.

Theorem1.4: Every non-empty subset of ℕ contains a least element.

Proof: Let S be a non-empty subset of ℕ.
Suppose S does not contain least number. Then 1 ∉ S
Let A={m∣m∈ ℕ, m < k(every k∈S)}

Observation :

16
(i) 1 ∈ A
(ii)If a ∈ A then a′ ∈A, because a ∈ A gives k › a
or k  a′(=a+1).
But k=a′ gives that a′ is the least element of S. So,k ≠ a
And
∴ k › a′ or a′< k. Thus, a′ ∈ A and by axiom of induction it follows that A=
ℕℕ and
∴ S= Φ which is not true. And hence the theorem is proved.∎

The fact that every non-empty subset of positive integers has a least
element is known as :

Well ordering principle of (WOP) of positive integers

Example 6 : Assuming the WOP of positive integers, prove the axiom of

induction.
Solution: Suppose S ≠ ℕ
∴ there exists an s ∈ ℕ, such that s ∉ S.
Let T ={ m∣m ∈ ℕ ,m ∉ S}
Observation: T≠φ and T∩S= φ, T⊆ℕ and so by WOP, T contains a least
element (say) t.
By hypothesis,
S ⊆ ℕ such that
i) 1 ∈ S
ii) if k ∈ S
then k' ∈ S.
So, 1 ∈ S gives 1 ∉ T
∴t≠1
Or t > 1 or t > t-1 >0.
∵ t is the least element of T, t-1 ∉ T. So, t-1∈ S. And
∴ by (ii) (t-1)+1=t ∈ S and also, t ∈ T which is a contradiction. Hence S=
ℕ.∎

17
Theorem 1.5: The well ordering principle and the axiom of induction are
equivalent (from above it follows).

Example7: Prove that there is no natural number in between 0 and 1.

Solution;
Let S={x ∣ x ∈ ℕ, 0 < x <1}

We show that S= φ. Suppose S ≠ φ.

∴ S is a nonempty set of positive integers. So, by WOP of positive
integer, S contains a least number a (say).
∴ 0 < a <1 or, 0 < a2 < a < 1 or, 0 < a2 < 1 or, a2 ∈ S and a2 < a
And it contradicts the assumption that a is the least element of S.
Hence S ≠ φ is wrong.
∴ S= φ .∎

Property1.6: If b is an integer >1 then for any integer r ≥ 0, br> r

Proof: Proof is by induction. For r=0 this is trivially true. We therefore
assume r ≥ 1
Let P(r) be the statement:”br >r for all integers r ≥1”
P (1) is true.
Assume that P(r) is true, so that br > r
We show that P(r+1) is true.
Now, br+1 =br.b
∵ b ≥ 2, br+1 ≥ br.2
∴ br+1 ≥ br(1+1)
≥br+br
≥ br+1(1 is the least positive integer)
≥ r+1
It follows that P(r) is true for all r ≥ 0. Otherwise we may proceed as
follows:
∵ b-1> 0, we have b-1≥1
Multiplying both sides by the positive integer (1+b+…..+br),
We get, (b-1)(1+b+b2+…..+br) ≥ 1+b+…..+br
ie.,br+1≥ 1+b+……+br ≥ r+1,
∵ br >1 for every r.

18
∴ br > r for every r.∎
Example8: Let b be an integer greater than 1 and let c0,c1,c2,…………..cr
be integers between 0 and b-1 inclusive , with cr > 0. Put
2 r r r+1
n=c0+c1b+c2b +…..+crb . Then show that b ≤ n ≤ b
Solution: We have c0+c1b+c2b2+………….+cr-1br-1 ≥ 0,
∵ every number in this expression is non-negative, it follows that
n ≥ crbr.
By assumption, cr ≥ 1.
∴ n ≥ br.
Also 0 ≤ ci≤ b-1 for each , 0 ≤i≤r
∴ c0≤ b-1,
c1b ≤(b-1)b
c2b2 ≤ (b-1)b2
…… …. ……
….. …. …..
crb ≤ (b-1)br
r

∴ c0+c1+c2b2+……+crbr
≤ (b-1) (1+b+b2+………+br)
= br+1 -1
≤ br+1
∴ br ≤ n ≤ br+1
∎
Example9: 2 is irrational
Solution: Suppose that 2 is rational. Then there would exist positive
a
integers a and b with 2 = .
b
Consequently the set S={k 2 |k and k 2 are positive integers }(≠)ℤ+
( S≠, because a=b 2 ∈ S}).
∴ by WOP S has a smallest element, say s=t 2 .
We have s 2 -s=s 2 -t 2 =(s-t) 2 .
Since s 2 =2t and s are both integers, s 2 --s=(s-t) 2 - must also be an
integer.
Furthermore, it is positive since s 2 --s=s( 2 -1) and 2 >1.
It is less than s since s=t 2 ,s 2 =2t, and 2 <2.
This contradicts the choice of s as the smallest positive integer in S.

19
 It follows that 2 is irrational. ∎
Example 10. If an integer is simultaneously a square and a cube, then it is
of the type 7 k or 7 k  1 .
We verify this and the proof is left to the reader.
64  8 2  4 3 = 7.9  1 , and 7 6  7 2   7 3  i.e.,
3 2

117649  7  16807  343 2  49 3 . Students may try for other such numbers.
Note. The square of an odd integer is of the form 8k  1 and hence easily we
get the following:
For any odd integer m , clearly 3 | m  1mm  1 , and 8 | m 2  1 , 3,8  1 and so
a) 24 | mm 2  1 ,. And therefore for any odd integers m and n 8 | m 2  n 2
b) For any m  Z , m  6k  r , r  0,1,2,...,5 and choose m so that 2, 3 ∤m ,
and we get m  6k  1,6k  5 . Now m 2  23  36k 2  12k  24  12k 3k  1  24 or,
m 2  23  36k 2  60k  25  23  12k 3k  5  48 . Now in both the cases
24 | m 2  23 if 24 | 12k 3k  1 and 24 | 12k 3k  5 ,i.e. if 2 | k 3k  1 and 2 | k 3k  5 .
Whether k is odd or even , in both the cases, 2 | k 3k  1 and 2 | k 3k  5 .
Hence it is true that 24 | m 2  23 .

c) Next we consider any integer m  Z and

  
m 2 m 2  1 m 2  4  m  2 m  1mm  1m  2m and therefore, , 2,3,4,5,6 | LHS
and therefore 60, .

Example 11. Sum of squares of two odd numbers cannot be a perfect

square.
If it is, then it is an even number square. i.e. we then have
2k  12  2l  12  2t 2 or, (4k 2  4k  1)  (4l 2  4l  1) 2  4t 2 or,
4k 2  l 2   4k  l   2  4t 2 or, 2  4t 2  k 2  l 2  k  l  or,
1  2t 2  k 2  l 2  k  l  ,and is not possible.

1.2 Divisibility Theory

1.2.1 Division Algorithm.:

20
Theorem1.7: Given a,b∈ ℤ, b>0, then there exists q,r ∈ ℤ, such that
a=bq+r,o≤r<b
Proof: Let S={bx ∣ x ∈ ℤ,bx ≤ a}
Obviously S ⊆ ℤ. Again, for given a,b ∈ ℤ, there exists an n ∈ ℤ such
that bn> -a
or, -bn < a
or, b(-n) < a
or, b(-n) ∈S
∴ S ≠ φ . Thus, S is a non-empty subset of ℤ, which is bounded above (bx
≤ a).
∴ S contains a largest element, (say) bq (q ∈ ℤ). So, bq ≤ a. Hence, a=bq
+r(r ≥ 0).
Now we prove : r < b. If not, suppose r ≥ b (and ∴ r-b ≥0)
Then, a=bq+r= bq+b+r-b= b(q+1)+s , [s=(r-b)≥0]
=bq'+s, s ≥ 0
∴ a=bq+r=bq'+s, (s≥ 0)
Or, bq' ≤a
or bq' ∈S and
∴ bq' ≤ bq
or b≤ 0 and is a contradiction with b > o.
Hence r < b.
∴ a=bq+r where o ≤ r < b.∎

Now we prove:
Theorem1.8
Given a,b ∈ ℤ, b≠0, then there exists q,r ∈ ℤ, o≤ r <∣ b∣ such that
a=bq+r
Proof: ∵ 0≠ b ∈ ℤ, ∣ b∣ > 0 ,
∴ by Theorem 1.7,
a=∣ b∣q1+r , for q1,r ∈ ℤ, 0≤ r <∣ b∣
= (b)q1+r
= b( q1)+r
= bq+r

21
 Thus, a=bq+r, 0 ≤ r <∣ b∣, q1=q ∈ ℤ
Remark: q and r are unique
If possible let a=bq1+r1, 0 ≤ r1 < b
and a=bq2 +r2, 0 ≤ r2 < b
or, 0=b(q1-q2)+(r1-r2)
or, ∣b ∣∣ q1-q2∣= ∣ r1-r2∣ ……… (*)
Again, 0 ≤ r1 <| b| ……. (i)
0≤r2 < | b|
or, - | b| ≤ -r2 < 0 …….. (ii)
Adding (i) and (ii),
0–∣b∣ < r1-r2 < ∣b∣
or, ∣r1-r2∣ < ∣b∣
or,|b|∣q1-q2∣ < ∣b∣ [by (*)]
or, ∣q1-q2∣ < 1
or, ∣q1-q2∣ ≤ 0
And the only possibility is ∣q1-q2∣=0.
∴ q1=q2 and from (*) r1=r2 ∎

Definition:
(1) a=bq+r is called the principle of division identity of integer .
(2) q is called the quotient of division
(3) r is called the remainder of division.
We now look the above mentioned division algorithm from another
angle.

Consider a=17, b=59

Clearly a does not divide b (in symbol a ∤ b). We understood that division is
repeated subtraction. This way, each time the subtraction is being repeated
if the remainder b-a > a. After a finite number of steps, at a certain stage
remainder will be less than a. In this process we get the following multiple
of a viz. a, 2a, 3a, 4a,….qa, (q+1)a
And we get either b=one of these positive integers for some q,
Or, qa < b < (q+1) a and this gives, qa ≤ b < (q+1)a which gives b-aq < a . If
b-aq =r then, b=aq+r where, 0 ≤ r < a

22
This is what is known as Division Algorithm [observe that this is not a
proof.]
We now give another proof of division algorithm

Theorem1.9: To an integer a ≥ o and an integer b ≥ 1 there exist two unique

integers q and r such that a=qb+r with 0 ≤ r < b
Proof: Now, If b=1,then a=a.1+0 and the result is true.
So we now assume: b>1
If a=0, we take q=0,r=0
If a<b we have q=0, r=a
If a=b, take q=1 and r=0
We have a-b<a if a>b
Consider the case, a > b. Then 0 < a-b < a.
Suppose P(a) a=bq+r, 0 ≤ r< b, for all integers a > 0, b ≥ 1
P(1) is true, for 1=0.b+1
Assume that P(k) is true for all k ≤ a .
∵ 0 < a-b<a, there are integers q1,r
0 ≤ r < b, such that a-b=bq1+r 0≤r<b
So, a=b(q1+1)+r =bq+r 0 ≤ r < b ( q=q1+1)
Hence P(a) is true. Thus P(a) is true ∀ a > 0.∎
Uniqueness part is same as in the previous method.∎
The theorem can be extended to include the case of all integer a where a is
negative.

Note 1: If r=0 we say that b divides a (or a is a divisor of b, or b is a factor

of a) and is written as a ∣b; otherwise a ∤ b.
Note 2: b ∣ a gives a=bq, q ∈ ℤ and

b ∤ a gives a=bq +r, 0 < r < b

a  bq  r
To illustrate division algorithm when the divisor is a negative integer,
say b  7, a  1,2,61,59 . Then

1  0 7   1 ,  2  1 7   5 , 61   8 7   5 ,  59  9 7   4 etc.

23
Example 12. Using division algorithms prove that the square of any odd
integer is of the form 8k  1 , for some integer k.
Solution: by division algorithm any integer is of the forms
4k , 4k  1 , 4k  2 , 4k  3 . Since the integer chosen is an odd one, the
number should be of the form 4k  1 and 4k  3 .
Now 4k  i 2  16k 2  8ki  i 2  8(8k 2  ki)  i 2 = 8t  i 2 [ i 2  1,9  1,8  1 ]
= 8t  1,8(t  1)  1  8m  1 . Thus it is of the form 8k  1 , for some integer
k.∎

Example 13: 1∣a a ∈ ℤ

If 1 ∤ a, then a=1.q+r when 0 < r < 1. But there is no integer in between 0
and 1 . Thus , 1∣ a , ∀ a ∈ ℤ.∎
Example 14: (Similarly) -1 ∣ a, ∀ a ∈ ℤ.∎
Example15 : if a ∣ b then ±a ∣ ±b ; ∀ a,b ∈ ℤ
Solution: ∵a ∣ b , we have b=aq, (q ∈ ℤ ) =(-a)(- q) =(-a)q1,( q1 ∈ ℤ ) ∴ -
a ∣ b.
Example16: if a ∣ b, b ∣ c then a ∣ c
Solution; ∵a∣b, we have b=aq1 ,(q1 ∈ ℤ) ,b| c
∴ c=bq2,(q2 ∈ ℤ) or, c=bq2=(aq1)q2 =a(q1q2) =aq3 where q3 = q1(q2 ∈ ℤ) or,
a|c ∎
Example17 a ∣ b gives a ∣k b , k ∈ ℤ.
Example18: if a ∣ b, a ∣ c then a ∣ kb+ lc, k,l ∈ ℤ
Example19: a∣b, b > 0 gives b > a
Proof: (i)if a < 0 then (obviously) b > a
(ii) and a > 0, a ∣ b gives b=aq, q ∈ ℤ
Now , q > 1⇒ aq > a
Or, b=aq > a or, b> a.∎
Example20: ac ∣ bc, c ≠ 0 gives a ∣b
Proof: if ac ∣ bc then bc= (ac) q, (q ∈ ℤ)
Or, c(b-aq)=0
Or, b-aq=0(∵c ≠ 0)

24
Or, b=aq
Or, a | b.∎
Example21: If a ∣ b, b ∣ a, a,b ∈ ℤ then a=±b
Example22: If a∣ b then ∣a∣∣ ∣ b∣

Some results 1.10:

1. if a|b then ± a|±b
2. if a|b, b|c then a|c
3. if ac|bc then a|b , if c ≠ 0
4. if a|b then a|bx for every x
5. a|b, a|c give a|(b+c) and a|(b-c)
6. a|b, a|c give a|(bx ± cy)

1. 2.2: Greatest Common divisor(gcd): Let a,b ∈ ℤ, b ≠ 0

Then, obviously ∣a∣ +∣b∣ > 0. Let S={ x ∣ x is a common positive divisor of a
and b}
We have , 1 ∣ n, ∀ n ∈ ℤ
∴ 1 ∣ a, 1 ∣ b and 1 ∈S
∴ S≠φ
Now if x ∣ a then x ∣∣ a∣ and if x ∣ b then x ∣∣ b ∣
Then, x ∣∣ a∣ + ∣ b∣ gives x ≤ ∣a ∣ +∣b∣
∴ S is a non empty set of positive integers which is bounded above
∴ S has a largest element, (say) g.

Definition: This g is called the greatest common divisor (GCD) of a and b.

Note: (i) g exists and g ≥1
(ii) we write (a,b)=g.

Theorem 1.11: Let a,b ∈ ℤ , b ≠ 0[or, a∈ℤ, b ∈ ℤ* (=ℤ -{0}] (a,b)=g, then,

(i)there exists x0,y0 ∈ ℤ such that ax0+by0=g

(ii)for d ∈ ℤ, if d ∣ a, d ∣ b then d ∣g

25
Proof; Consider S={ax+by| x,y ∈ ℤ, ax+by > 0}
Now, ∣ a∣ +∣b∣ > 0 ⇒ ±a ±b > 0 or, a.( ± 1)+b.( ±1) > 0 or,∣a∣+∣b∣ ∈ S.
S (≠φ ) ⊆ N and so by WOP , S contains a least element (say)m which is of
the form ax+by (say), m=ax0+by0 , x0,y0 ∈ ℤ
Assert, m=g
To prove our assertion we are to show that
(A) m is a com mon divisor of a and b (m ∣ a and m ∣ b)
(B) If d ∣ a and d ∣b then d∣ m.
Proof: Suppose, m ∤ a
∴ a=mq+r , 0 < r < m…(i)
or, r =a-mq=a-q(ax0+by0)=a(1-qx0)+b(-qy0)=ax' +by' ; x' ,y' ∈ ℤ
gives r ∈ S ,
∴ r ≥ m…….(ii)
which is a contradiction to(i).
Hence m ∣ a and similarly m∣ b
(B) Again ,d ∣a , d∣ b
gives d ∣ax0, d∣ by0
or, d∣ ax0 +by0 =m
or,d≤ m
or, m =g=ax0+by0.∎

Note: Incidentally we have proved here that d∣ m=g,

i.e.If d ∣a, d∣ b
then d ∣ (a,b)
And g is the least positive integer of the form ax+by
[The greatest common divisor can be characterized in the following two
ways:
1. it is the least positive value of ax+by where x and y range over all
integers
2. it is the positive common divisor of a and b which is divisible by any
common divisor.

26
Corollary 1.12: If (a,b)=1 then there exist x,y ∈ ℤ such that ax+by=1 [Put
g=1]
Conversely, if a,b ∈ ℤ+ and for x,y ∈ ℤ ,ax+by=1 then (a,b)=1
Proof; Suppose, (a,b)=d
Then, for some x,y ∈ ℤ we have ax+by=d. And, given that ax+by =1.
∵ 1 is the least positive integer, but d is the least positive integer of the
form ax+by.
∴ d=1. ∎

Definition: If (a,b)=1, then a,b are said to be relatively prime

Theorem 1.13: Any non-void set of integers closed under addition and
subtraction consists of zero alone or else consists of the least positive
elements and all the multiples of this element.
Proof: Let S be any non empty set of integers closed under addition and
subtraction.
Let an integer a ∈ S. Then a-a ∈ S.
∴ 0 ∈ S. Also 0-a ∈ S
∵ a, –a ∈S, of these two, at least one is positive.
By WOP, the set of all positive elements of S will contains a least element,
say d. we wish to show that every element of S is an integral multiple of d
and conversely.
It is clear that if n is any positive integer,
and nd=d+d+……….+d ( n times), ∴ nd ∈ S
∴ any integral multiple of d is in S. Suppose now that k ∈S
If d does not divide k , then by division algorithm
k=q.d+r, k,r∈ ℤ and 0 ≤ r < d
Now,
∵ k and qd are elements of S and S is closed under subtraction, r=k-q.d is
also a member of S.
∴ we must have r=0.
Thus, every element of S is a multiple of d.
Example23: (4,9)=1 ∴ 4 and 9 are rela vely prime
(16,15)=1∴16 and 15 are rela vely prime
Example24: If a,b,k ∈ ℤ , then (a+kb, b)=(a,b)

27
If a2=0, then d=a1

If a2 ≠ 0 then divide a1 by a2. there exist integers q2 and a3 such that

a1 =q2.a2+a3 with 0 ≤ a3 < a2.

If a3=0 then d=a2

If a3 ≠ 0 then we divide a2 by a3. There are integers q3 and a4

Such that a2=q2a3+a4, 0 ≤ a4 < a3.

Proceeding in this way , we observe that the remainders are getting

reduced successively since a2 > a3 > a4 ……..etc. After a finite number of
steps say k, the remainder ak would be such that ak-1 is a multiple of ak, so
that the remainder for the k th division is zero. We write this as follows:

a0 = q1.a1 + a2

a1 = q2.a2 + a3

a2 = q3.a3 + a4

.. ….. ..

…. …. ….

ak-2 = qk-1.ak-1 + ak

ak-1 = qk.ak

Now we show that ak is the g.c.d. of a0 and a1

From above we see that ak|ak-1. From the equation preceding the last, we
get ak|ak-2, since ak divides both the terms on the right. From the equations
that precedes it, we conclude similarly that ak|ak-3. Continuing up the list in
this way, we find that ak|a0 and ak|a1. We now show that if b|a0 and b|a1
then b|ak.

∵ b|a0 and b|a1, the first equation above shows that b|a2.

29
∵ b|a1and b|a2 the second equation above shows that b|a3. Continuing
down the list we see that b|ak

∴ ak=(a0,a1).∎
Definition: If d| a1,a2,……,an such that d is the greatest among the common
divisors of a1,a2,……,an then d is called the gcd of a1,a2,……,an and is denoted
d= (a1,a2,……,an)

Theorem1.14:If d=(a1,a2,…….,an), there are integers x1,x2,…,xn such that

d=a1x1+a2x2+…+anxn (Extension)

Proof:Let S={ a1x1+a2x2+………+anxn,| x1,x2,…,xn ∈ ℤ}

Now here |a1| ∈ S, |a1| = a1 sgna1 here, x1=sgn a1, x2=x3=…=xn=0.

[sgn stands for the word ‘sign’]

Thus S has a least positive member, d0 say. We show that every member is
divisible by d0. By division algorithm, if m is any member of S, we have
m=d0q+ r, 0 ≤ r < d0

Then r=m-d0q. Now d0=a1x1+a2x2+.+anxn and m=a1y1+a2y2+ …..+anyn, for

some integers x1,x2,..xn and y1,y2,……...yn.

Thus r =a1(y1-qx1)+ a2(y2-qx2)+…………..+ an(yn-qxn) ∈ S

∴ by the definition of d0 and by 0 ≤ r < d0 it follows that that d0 is a positive

common divisor of a1, a2,……an. Hence d0 | d, by the definition of d.

But d|ai (i=1,2,…n) and d0=a1x1+a2x2+.+anxn; thus d| d0.

∴ d ≤ d0. Hence d=d0 and the result follows∎

Illustration: Euclidean Algorithm
Two expressions of linear combination.
Two find the gcd 12378,3054
12378  4.3054  162
3054  18.162  138
162  1.138  24
138  5.24  18

30
24  1.18  6
18  3.6  0 thus 12378,3054  6

Now we look at the following:

6  24  18  24  138  5.24
 6.24  138
 6162  138  138
 6.162  7.138
 6.162  73054  18.162
 132.162  7.3054
 13212378  4.3054  7.3054
 132.12378   5353054
Thus 12378,3054  6  132.12378   5353054  12378x  3054 y where
x  132, y  535
and appears as a linear combination of 12378 and 3054 .
Also 12378  4.3054  162
3054  19.162  24
162  7.24  6
24   4  6   0
Note: it is to be worth mentioning here Lame’s remark: the number of steps
required in the Euclidean Algorithm is at most five times the number of
digits I the smaller integer. In the above example the smaller integer has
four digits. So the total number of divisions cannot be greater than twenty.
In actuality, thee only six divisions were needed.

Theorem1.15: If d=(a1,a2,……an) , then an integer d1 is a common divisor of

a1,a2,…..an if and only if d1|d

Proof: If d1|d, then,

∵ d is a common divisor of a1,a2,….an, it follows that d1|ai(i=1,2,…n).Thus d1

is a common divisor of a1,a2,…an.

Conversely, suppose that d1 is a common divisor of a1,a2,….,an. By the above

theorem, there are integers x1,x2,….,xn such that d=a1x1+a2x2+…+amxm.
Hence d1|d ∎

31
Theorem1.16: If a1,a2,…an are nonzero integers and if d1=a1, d2=(d1,a2),
d3=(d2,a3)…..,dn=(dn-1,an), then dn=(a1,a2,…..,an)

Proof: (is left as an exercise.)

Example 27: Express (726,275) in the form m.726+n.275

Solution: 1 275 726 2 ∴ 726=275.2+176

176 550 275 = 176.1 + 99

1 99 176 1 176 = 99.1 + 77

77 99 99 = 77.1 + 22

2 22 77 3 77 = 22.3 + 11

22 66 22 = 11.2

+ 11 Thus 11 is g.c.d

∴ 11 = 77 – 22.3 = 77-(99-77.1)3

= 77.4 - 99.3 = (176 – 99.1)4 – 99.3 = 176.4 – 99.7= 176.4-(275

-176.1).7

= 176.11 – 275.7 = (726 – 275.2)11- 275.7

= 726.11-275.29

Thus 11=m.726+n.275, where m=11, n=-29.∎

Example28: Prove that 4 ∤ (n2+2) for any integer n

Solution: (i) Let n be an odd number. Then n2 is also odd and

∴ n2+2 is d also odd number. Hence 4∤ (n2+2)

(iii) Let n be an even number, then 4|n2. Hence when 4 divide n2+2, a
remainder 2 is left.

∴ 4∤ (n2+2) ∎

32
 Example 29: Show that g.c.d. of a+b and a-b is either 1 or 2 if (a,b)=1

Solution:

CaseI. Let f=(a+b,a-b). Then f=m(a+b)+n(a-b)= (m+n)a+(m-n)b=ma + n

b. But (a,b)=1 , f=1

CaseII. If a=b , then (a,a)=1 gives a=1. Then (a+b,a-b)=(2,0)=2.

∴ The g.c.d. is either 1 or 2.∎

Example 30: If (a,4)=2 and (b,4)=2, prove that (a+b,4)=4

Solution: From hypothesis, a=2a1, where a1 is an odd number

b=2b1, where b1 is an odd number

But a1+ b1 is an even number, as the sum of two odd numbers is always
even.

∴ a1+ b1=2m(say)

So, (a+b,4)=2(a1+ b1,2) =2(2m,2)=4(m,1)=4 ( as (m,1)=1). ∎

Example31: If x-y is even, then show that x2-y2 is divisible by 4, x and y being
positive integers.

Solution: ∵ x-y is even x and y should either be both odd and both even.

∴ x+y is also even

Hence, x2-y2=(x+y)(x-y)=(Multiple of 2).(Multiple of 2)=Multiple of 4=an

expression divisible by 4. ∎

Example32: Show that the difference between any number and its square is
even.

Solution: n2-n=n(n-1)= Product of two consecutive numbers. Hence one of

the two must be even.

∴ this number is divisible by 2∎

33
Example33: If 4x-y is M(3), show that 4x2+7xy-2y2 is M(9),

[M(n) denotes a multiple of n.]

Solution: 4x2+7xy-2y2

=(4x-y)(x+2y)

=(4x-y){(4x-y)-3(x-y)}

=M(3){M(3)-M(3)}

=M(9).∎

Example 34: Show that the square of any integer b is of the form 4k or
8k+1.

Solution: By division algorithm, any integer b is represent able as 2q or

2q+1. If b=2q, then b2=4q2. Thus b2 is of the form 4k.

If b=2q+1, then b2=4q2+4q+1=4q(q+1)+1.

∵ q(q+1) is divisible by 2, we get that b2 is of the form 8k+1.∎

Example35: Find all integers n such that n2+1 is divisible by n+1

Solution: Let n be an integer such that n+1|n2+1.

Note that n+1|(n+1)(n-1) ie., n+1|n2-1.

Hence, n+1|(n2+1)-(n2-1) i.e., n+1|2.

∴ n+1= +1, +2. So, n= -3,-2,0,1 ∎

Some properties1.17:
a b
1.If (a,b)=d then, ( , )=1
d d

2. If a|bc and (a,b)=1 then, a|c

a b
3.If c ≠0, c|a, c|b, ( , )=1 then c=(a,b)
c c

34
And this gives , [a, b] (a ,b)=ab.∎
Example38: If (a,b)=1 then a!b!|(a+b-1)!

(a  b  1)! (a  b  1)!
Solution: Let =c and =d
(a  1)!b! a!(b  1)!
Then (a+b-1)!=(a-1)!b!c=a!(b-1)!d
And so bc=ad. ∵ (a,b)=1, then a|c, that is , c=ar.
Thus (a+b-1)!=a!b!r, which implies that a!b!|(a+b-1)!.
Hence the result follows.∎

Example 39: prove that the product of r consecutive integers is divisible by

r!

Proof: Let Pn =n(n+1) (n+2)……..(n+r-1).

Then, Pn+1=n(n+1)(n+2)…………..(n+r)

∴ Pn+1 –Pn =(n+1)(n+2)….(n+r-1){(n+r)-n}

Pn
=(n+1)(n+2)…(n+r-1)r = r
n

= r times the product of (r-1) consecutive integers.

We now use induction as follows: Assume that the product of (r-1)

consecutive integers is divisible by (r-1)!,

Therefore, Pn+1 –Pn=r. M({(r-1)!}=M(r!)

∴ P2 –P1 =M(r!); But P1 =r!; So, P==M(r!)

∴ by induction P3, P4, …..,Pn are all multiples of r!

Again the product of any two consecutive integers is divisible by 2!. Hence
conclude that the product of any three consecutive integers is divisible by
3!, and so on ∎

1.3: Theory of scales of numeration:

38
In the decimal system of numeration then digits 1,2,3,4,5,6,7,8,9 and 0 are
used. Can we think of any other system using some or all of these digits?
We have a few such system viz., The decimal system and the binary system
We observe the following :
13=1.23+1.22+0.2+1=(1101)2

31=1.24+1.23+1.22+1.2+1=(11111)2

387=1.28+0.27+1.26+1.25+1.24+0.23+1.22+1.2+1=( 101110111)2 the binary

representation of the number 387.
On the other hand 13=1.10+3; 31=3.10+1;387=3.102 +8.10+7
We now observe the following:
The ordinary decimal notation uses the representation of integers in in the
scale of 10 in the binary system the same is in the scale of 2 etc.,
We now discuss some general result in this regard

Theorem1.24: Let S be a positive integer. If n is a positive integer such that

bnSn + bn-1Sn-1+………+b1S+b0 = 0 where the integers bi are such that |bi| ≤ S-
1, then bi = 0 (i=1,2,…,n)
Proof: From bnSn+bn-1Sn-1+………+b1S+b0 = 0 we have b0=k0S for some integr
k0

Thus |b0| ≥ S unless k0=0 and so b0=0

Then we have bnSn+bn-1Sn-1+………+b1S = 0 and so for some integer k1we
have b1=k1S and in the same way we have b1=0 and so on. Hence finally we
have bi=0 for all i∎

Theorem1.25: Suppose S is a positive integer. Then each positive integer a

can be expressed uniquely in the form
a= cnSn+cn-1Sn-1+………+c1S+c 0, where 0 ≤ ci ≤ S-1 (i=0,1,2,….,n-1) and
0 < cn ≤ S-1
Proof: Let a be a given integer and B ={ k ∈ ℤ+| Sk+1 > a}
∵ S > 1, Sk+1 tend to infinity as k tends to infinity. Thus B is a non-empty
subset of positive integers and therefore it contains a least element say n.
Then Sn ≤ a < Sn+1 ………….(*)
and n is uniquely determined. By division algorithm we have,

39
Here each hex digit is converted to a block of four binary digits (the
initial zero in the initial block (0010)2 corresponding to the digit(2)16 are
omitted).And thus the corresponding binary representation is
(10111110110011)2
And to convert from binary to hex , if we take(11110111101001)2 then
we break this into blocks of four starting from the right . The blocks are
from right to left , 1001,1110,1101,and 0011. And thus we get in hex as
(3DE9)16

When performing base r addition, subtraction and multiplication by

hand, we can use the same familiar technique as use in decimal addition
Example42: To add (1101)2 and (1001)2 we write
1 1
11 01
+ 10 01
1 01 10

Where we have indicated carries by 1’s in italics written above the

appropriate column. We found the binary digits of the sum by noting
that 1+1=1.2+0,0+0+1=0.2+1,1+0+0=0.2+1,and 1+1+0=1.2+0.

To subtract (101110)2 from (11011)2 we have

-1
1 1 0 1 1
1 0 1 1 0
1 0 1

Where –1 in italics above a column indicates a borrow. We found the

binary digits of the difference by noting that 1-0=0.2+1,1-1+0=0.2+0,
0-1+0=-1.2+1.1-0-1=0.2+0, and 1-1+0=0.2+0.
To multiply (1101)2 and (1110)2 we write

41
 1 1 0 1
x 1 1 1 0
0 0 0 0
1 1 0 1
1 1 0 1
1 1 0 1
1 0 1 1 0 1 1 0
We first multiplied (1101)2 by each of (1110)2 , shifting each time by the
appropriate number of places , and then we added the appropriate
integers to find our product.
To divide (11101)2 by (111)2 , we let q=(q2q1q0)2. We subtract
22(111)2=(11100)2 once from (11101)2 to obtain (1)2, and once more to
obtain a negative result, so that q2=1. Now R1=(111010)2-(11100)2=(1)2.
We find that q1=0,since R1-2(111)2 is less than zero, and likewise q0=0.
Hence, the quotient of the division is (100)2 and the remainder is (1)2.

1.4: Prime numbers:

If a =1 then a has only one positive divisor viz.1.If |a| ≠ 1, then a has at
least two positive divisors viz, 1 and |a|.
The numbers of positive divisors of a is 1 if a=1 and >1 if |a| ≠1.

Definition: If a>0 and a has exactly two positive divisors then a is called a
prime number.

Definition: If a >0 and a has more than two positive divisors, then a is called
a composite number.
Remark: (i) 1 is not a prime number for, it has only ‘one’ positive divisor.
(ii) If p is a prime number then, 1 and p are the only positive
divisors of p.

Lemma1.26: If p is a prime, then (p,a)=1 or p|a.

Proof: Suppose (a,p)=d

42
Then, d|p, d|a and p being a prime, it therefore follows that d=1 or d=p
If d=1 then we are done and if d=p then from above it follows that p|a.
Hence either d=1 or d|a ie., (a,p)=1 or (a,p)|a. ∎

Properties1.27: If p is a prime and p|ab then, p|a or, p|b, where a,b ∈ ℤ

Proof: Let p∤a, then (p,a)=1

∴ ∃ x,y such that px+ay=1
And this gives, pbx+aby=b…(i)
Now p|ab
gives ab=pk, k ∈ ℤ ,
∴ from (i) we get, pbx+pky=b
⇒ or, p(bx+ky)=b
or, pq=b, q ∈ℤ
or, p|b
Similarly if p ∤ b, then p|a.∎

Extension1.28:If p|a1.a2….an then, p|a1,or p|a2 ,or ..p|an

( proof is repeated application of the above result.)

Corollary1.29: If p|p1.p2... pn ( p’s are primes) then, p is one of p1,p2,..,pn

Proof: By one preceding theorem p|p1p2…pn gives either p|p1or, p|p2p3
….pn . If p∤p1,then again by the same result we have p|p2 or p|p3p4..pn .
Hence with a finite number of application of the same result we get that
p|pi , for some i. And
∵ p is a prime we must have p = pi∎

Converse of the theorem1.30: Let p > 1 and p has the property that if for
any a,b∈ℤ p|ab gives p|a or p|b, then is a prime.

[Note: Both the theorems can be combined as for a,b ∈ ℤ , p|ab gives
p|a or p|b if and only if p is a prime.]
Proof: Suppose p is not a prime. Then it is composite..

43
∴ q ∈ ℤ+ such that q | p and q ≠1,p ie. 1 < q < p.
∴ p=qr, 1 < r < p
Now p|p=qr, p|q or p|r. But both q and r are positive integers (< p) .
∴ p < q or r. So we meet a contradiction.
∴ our assumption is wrong. So, p is prime∎
Example43: If n=ab, then at least one of a and b must be less than n

Theorem1.31: If a ≠ 1, a ∈ Z, a must have a prime factor.

Proof: case i): If a is a prime then a is itself a prime factor.
Case ii): Let a be not prime. And S={ d|d > 1, d ∈ ℕ, d|a}
Now |a| |a, |a| > 1, |a| ∈ S .
∴ S ≠φ.
∴, S is a non-empty set of positive integers, and by W.O.P it has a least
integer, say, p.
Assert: p is a prime
And for p ∈ S, p >1, p|a. So 1 and p are positive divisors of p. If possible let
q ≠1,p and q|p, q ∈ ℕ.
∴ q < p….(i)
But p|a, q|p ⇒ q|a and so, q ∈ S
∴q ≥ p and this contradicts (i)
Hence p is prime∎

Properties1.32: Every integer n (>1) can be expressed as a product of finite

number of primes.
Proof: If the integer is a prime then it itself stands as a product with a single
factor. Otherwise by above theorem we have a prime factor p1 such that
n=p1n1, 1<n1<n
If n1 is not a prime then again by the same result we have another prime p2
such that n2=p2n3 , 1<n3<n2. Similarly for n3 . This process of writing each
composite number that arises as a product of factors gives us n > n2 > n2 >
n3 > ….., a descending chain of positive integers and it must therefore
terminate after a finite steps.
Thus we can write n as a product of finite number of primes.

44
Remark: Every integer can be expressed as product of primes.
For n= ± [p1n1] , p1 is prime, 1 < p1 < n
= ± [p1,p2.n2] 1 < p2 < n1
=………
=± [p1,p2,p3,..,pn]; p’s need not be all different.
[ n=100=2. ×50=2×2×25=2×2×5 ×5]
Note: Disregarding the order in which the primes are written, every natural
number can be expressed as a product of primes and this is the result that
we are going to prove below. Before giving the proof of the theorem we not
the following
Does the fundamental theorem need a proof at all? As Prof. H R Gupta says
“Has any one ever come across an example where the theorem has failed?
The following example will convince the reader about the necessity of a
proof.[Selected Topics in Number Theory;H Gupta:ABACUS PRESS,1980]
Take the set
{1,4,7,10,13, 22,25,28,31,…} of natural numbers. It is easy to see that the
set is closed under multiplication. Call an element of the set a ‘prime’ if it
has exactly two divisors in the set. Thus, the primes of the set are
4,7,10,19,22,25,31,..
We observe that 100 belongs to the set and it cam be written as a product
of the primes of the set in two distinct ways:
100=4.25; also 100=10.10.

Fundamental Theorem of Arithmetic (also known as unique factorization

theorem) 1.33:

The presentation of n(> 1),as a product of primes is unique apart from the
order of factors
Proof :
Suppose that n=p1p2..pr=q1q2…qs, where p1,p2,…pr, q1,q2,…qs are primes and
suppose that the primes are ordered so that p1 ≤ p2≤…..≤ pr and q1≤q2 ≤
…..≤ qs. We now prove that r=s and pi=qi(i=1,2,….r)
The proof will be by induction.

45
 The result is true for n=2. Suppose that it is true for 2,3,..,n-1 and consider
the number n.
If n is a prime the result is true. Suppose n is not a prime . Then in the
expression n= p1p2…pr=q1q2…qs we have r >1 and s > 1. Then p1=qj and
q1=pi for some i and j (by corollary1.29) ∵ p1 ≤ pi =q1 ≤ qj=p1, it follows
that p1=q1
Then the integer n/p1 is such that 1 < n/p1 < n, and we have n/p1
=p2….pr=q2…q2. Thus from the inductive hypothesis r=s and pi=qi(i=2
,,,….r)Hence r=s and pi=qi(i=1,2,…r)
And the result follows by induction. ∎
In the application of the fundamental theorem we frequently write any
integer (> 1) in the form, sometimes called the “ fundamental form”
ak
n= p1 1 . p2 … pk
a a2

One may prove the above result if this form is used to write
ak bj
Proof: Suppose n= p1 1 p2 . … pk = q1 1 q2 . …. qj ,
2a a 2 b b
p’s and q’s are
primes…………..(*)
Assume that, p1 < p2 <p3 < . .…<pk and also q1< q2< q3<.. …<qj
ak
Now p1| p1 1 . p2 … pk (obviously).
a a2

∴ p1|q1q2…qj So, p1|qibi , for some i ∈ {1,2,..j} so, p1|qi ,

Hence, p1=qi.
∴ every p is some q.
Similarly, starting from right p’s and q’s are arranged in ascending order, so
follows that p1=q1, p2=q2. and k=j.
Now to prove that ai=bi.
If not, say bi > ai. Dividing (*) by piai ,we get
b ai b
a a a
p1 1 p2 2 . pi1 i1 pi1 … pk =
a i1 k a b
p1 1 p2b . …. pi i
2
… . pk k .
Hence pi∤ LHS.
But pi | RHS and is a contradiction.. Thus, aj=bj.
a a
Corollary1.34: If n ∈ ℤ then n= ± p1 1 p2 2 . … pk , p1 < p2< …<pk
ak

Note:
i) Pythagoreans deserve the credit for being the first to classify numbers
into odd, even prime, and composite.

46
ii) a prime p is of the form 3n  1 . Now if n is odd , then n  2k  1 and then
p  32k  1  1  6k  4  23k  2 and is an even that is not prime. Hence n
must be even i.e. of the type n  2k . And so p  3.2k  1  6k  1 . Thus the
prime is also the form 6k  1
As 3n  23m  2  3k  1 , 3n  23m  1  3k  2 , 3n  13m  1  3k  1 , it
therefre follows that
iii) Each integer of the form 3n  2 has a prime factor of this form.
As n 3  1  n  1n 2  n  1 . Now to be a prime of this form, either n  1 or
n 2  n  1 is unity. And thus the possible case is n  1  1 or, n  2 . Hence
iv) 7 is the only prime of the form n 3  1
Suppose p is a prime . Then if 3 p  1  m 2 for some integer m ,
then 3 p  m  1m  1 and by observation the possible values
are m  1  3 , m  1  p and so p  5 . Thus
v) the only prime p for which 3 p  1 is a perfect square is 5 .
If p is a prime then p  6k  1 or, p  6k  5 , for p  2 and then
6k  12  6m  1 for some m. and 6k  52  6k  1for some k and in both
cases 6m  1  2  6m  3  32m  1 .
Hence
vi) for a prime p  2 , p 2  2 is always composite. It is also easy to see that
vii) for any prime p and any integer a , p | a n  p n | a n . As for an odd
number n a n  b n  a  b a n 1  ...  b n 1  , we can easily see that
viii) any integer of the form 8 n  1 is composite.

Theorem1.35: There are infinitely many primes . `

Proof; Suppose 2,3,5,7,11,………,p be the finite set of primes up to p.Then
let q=2.3.5.7.11…..p+1
Now q is not divisible by any of the primes 2,3,5,7,11,…,p because if q is
divided y any of these primes 1 is left as remainder.
Hence q is either a prime number itself or is divisible by some prim
between p and q in either case there is a prime number greater than p.
∴ the number of primes is not finite.∎

Theorem 1.36: No rational algebraic formula can represent prime numbers

only .
Proof: If possible let the formula
a+bx+cx2+dx3+………+kxn …………(1)

47
represents prime numbers only.
When x = m, let its value be p.
Then p=a+bm+cm2+dm3+……….+kmn
When x=m+np, (1) gives
a+b(m+np)+c(m+np)2+………+k(m+np)n
i.e., a+bm+cm2+dm3+….+kmn+ a multiple of p
= p+a multiple of p
= M(p), where symbol M(p) stands for a multiple of p
= an expression divisible by p
Hence, when x=m+np, (1) does not give a prime number.
This shows that there is no simple general formula for the nth prime p, ie.,
a formula by which we can calculate the value of pn for any given n ∎

Theorem1.37: There are arbitrarily large gaps in the series of primes. ( In

other wise, there exist k consecutive numbers whose length exceeds any
given number k.
(given any positive integer n , there exist n consecutive composite integers.)
Proof: Consider the integers
(k+1)!+2, (k+1)!+3……………(k+1)!+k, (k+1)!+k+1.
Each of these numbers is a composite number because the number n
divides (k+1)!+n
if 2 ≤ n ≤ k+1 and there are the k consecutive integers which are
composite. Hence the theorem.∎
Example44: There are infinitely many primes of the form 4n+3
Solution: Suppose 2,3,5,7,11…p are the primes up to p, and let
q=22.3.5…..p -1,
Now q is of the form 4n+3 and is not divisible by any of the prime 2,3,5…p.
It cannot be a product primes 4n +1 only, because the product of two
numbers of this form is of the same form. It is therefore either a prime or
divisible by a prime of the form 4n+3, greater than p. Hence there are
infinite number of primes of the form 4n+3∎
Example45: Prove that there are an infinite number of composite numbers
among the numbers represented by the polynomial f(x)≡a0xn+a1xn-1+….+an,
where n>0, a0,a1,…..,an are integers and a0 >0
Solution: Suppose m is an integer such that f(x)>1 and f(x)>0 for x ≥m.
Suppose, f(m) =M. Then all the numbers given by
f(x+Mt), t=1,2,….. are composite as they are multiples of M.

48
∴ ak+1 is not a prime, if a is odd.
(IV) a > 1 , k > 1, a is even , k is odd , a+1|ak+1 ⇒ak+1 is not prime

(V) a >1, k > 1, a is even and k is even.

Let k= 2k1.k2, k2 is odd, k1 ≥ 1.

∴ ak+1= a2
k1
.k2
  +1 = b
+1= a
2k1
k2
k2
+1 ;
k2
∴ b+1| b 2 +1 gives b +1 is not prime
k

and k2 is odd and ≠ 1 gives ak+1 is not a prime.

∴ ak+1 is may be prime if k2=1
k
(vi) a> 1,k >1, a is even , k is even and of the form 2 1 and then ak+1=
2k1 2r
a +1 is prime for some k1.In particular, 2 +1 is prime for some r.
m
2
Definition: Fm= a +1 are called Fermat number
Fermat’s conjectured that Fm is prime for all m ∈ ℤ.
25
F0=3,F1=5,F3=257, F4=65537 , that F5= 2 +1=4,294,967,297 is not a prime
was shown by Euler in 171
At the beginning Fermat conjectured that all the Fermat numbers are
primes. In 1732, Euler pointed out that F5 is a composite. This negated the
Fermat conjecture. So, there are some primes and others are composite in
Fermat numbers.
Up to now , we know only that the first five Fermat numbers are
F0=3,F1=5,F2=17,F3=257,F4=65537 are primes and other 49 numbers Fn
are composite; their respective n’s are:
5,6,7,8,9,10,11,12,13,14,15,16,18,
19,20,21,23,25,26,27,30,32,36,38,
39,42,52,55,58,63,73,77,81,117,
125,144,150,207,226,228,250,267
, 268,284,426,452,556,744,1945,

Besides these we do not know whether or not there are infinitely many
Fermat primes. No new Fermat prime has been discovered for the last
30 years(since 1995), so many people conjecture that there are no more

50
 Fermat primes. This is still one of the unsolved problems in the theory of
numbers.

The following elementary proof that 641|F5 is due to G.Bennett(which does

not explicitly involve division)
Example 49 :The Fermat number F5 is divisible by 641
Proof: We put a=27 and b=5, so that
1+ab=1+27.5=641
it is easily seen that
1+ab-b4=1+(a-b3)b=1+3b=24
But this implies that
25
F5= 2 +1 =232+1
=24a4+1
=(1+ab-b4)a4+1
=(1+ab)a4+(1-a4b4)
= (1+ab)[a4+(1-ab)(1+a2b2)] which gives 641| F5 ∎

Conjecture: Fm is prime for finite number of m: Prove or disprove !

2n
Example50: d=( a +1,
2m
a +1)|2 if m ≠ n

Solution: Given that m≠ n. suppose m < n

2mk
=X say. Then, a = a
m k m k
2m 2 .2 2 2 2
=a
2n
∴ n=m+k,say. Then, a =( a ) =x
k

2mk
2k
Now x+1| x -1, k ≥1 .Now d=( a +1,
2m
a 2k 2k
+1=(x+1, x +1)| x -1, x+1 .
k k
2 2
∴ d| x -1, x +1 ⇒d|2.∎

Corollary1.38: (Fm,Fn)=1
Proof: (left as an exercise)

Let us consider ak-1

(1) a-1|ak-1 ak-1 may be a prime if a=2

51
 (2) a=2, let k=lm(ie. We are assuming 1 < l < k; 1< m < k ,that k is
composite)
∴ 2k-1=(2l)m-1⇒2l-1| (2l)m-1. ∵ l >1, 2k-1 is not a prime if k is composite.
(3) ak-1 may be a prime if a=2, k >1and k is a prime.

Numbers of the form Mn =2n-1 n>1 are called Mersenne numbers after
a French monk Marin Mersenne who made an incorrect but
provocoative assertion concerning their primality.Those Mersenne
numbers which happen to be prime are said to be Mersenne primes.
In 1644, Marin Mersenne stated without proof that the number
2251 –1 is composite. In the 19th century, mathematicians finally proved
Mersenne correct when they discovered that this number was divisible by
both 503 and 54,217. However the issue still remained –what are the
prime factors of Mersenne number? In February 1984, two researchers
finally settled the matter using a 32 hour search on a Cray super computer
. The astonishing truth is this:

2251 –1 =
503Х54217Х178230287214063289511Х61676882198695257501367Х12
070396178249893039969681

Definition: If p is a prime then Mp =2p-1 is called a Mersenne prime.

(4) Mersenne said that Mp is a prime for p ≤257. Subsequently it is
found that Mp is a prime for p=2,3,5,7,13,17,19,31,67,257
(5) There are also mistakes as, p=61 gives Mp a prime.
Among the Mersenne numbers, there are some primes and others are
composite s . Up to now (1995) we know only the following 31 Mersenne
primes for which the respective p’s are
2,3,5,7,13,17,19,31,61,81,107,127,
521,601,1279,2203,2281,3217,4253,
4423,9689,9941,11213,19937,21701,
23209,44497,86243,110503,132049,216091

26972593-1 is the 38th known Mersenne prime discovered on June 1,1999 by

one of the 12000 participants in the Great Internet Mersenne Prime search.
It was also the first meagaprime found (prime with more than a million
digits) . Some websites predict that the first bevaprime (prime with more
than a billion digits) will be found by 2006.The complete list of the record

52