See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304636728

Divisibility

Article · September 2016

CITATIONS READS

0 47

1 author:

Sachchidanand Prasad
National Institute of Technology Rourkela
2 PUBLICATIONS 0 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Sachchidanand Prasad
letting you access and read them immediately. Retrieved on: 18 October 2016
 AN APPROACH TO
DIVISIBILITY

Abstract: A unified and simplest test of divisibility is proposed by

using elementary facts of linear congruence,Euclids algorithm.

Key words: Key words and phrases: Linear congruence, modular

arithmetic,fundamental theorem of arithmetic

1
Contents
1 Introduction 3

2 Congruence 3
2.1 Congruence Modulo m . . . . . . . . . . . . . . . . . . . 3
2.2 Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Linear Congruence . . . . . . . . . . . . . . . . . . . . . 3

3 Divisibility by 7 4
3.1 Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

4 Divisibility test for other integers 5

4.1 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.2 Proposition . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3 Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.4 Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

5 Conclusion 8
name10

2
1 Introduction
There are a lot of works had been done in the field of number theory
[4], but specially in the field of divisibility very less number of works has
been done. For checking of any given number is divisible by any number
is still a time taken work.We have some algorithm,[3, 1] for example,
Euclid’s algorithm[2] is one of the best method till known. As we know
till the date we don’t have any generalized test for divisibility. I’ve used
some facts of congruence which is an important tool in number theory.In
this paper I’ve try to illustrate the generalized result for divisibility using
basic facts of linear congruence and modular arithmetic.

2 Congruence
2.1 Congruence Modulo m
Let m be a positive integer. Then an integer a is said to be congruent
to an integer b modulo m if m|(b − a) and we write this in symbols as,
a ≡ (mod m); m is the modulus of the congruence relation.
Example:- 13|(45 − 6), 45 ≡ 6(mod 13)

2.2 Properties
We are discussing some properties without dealing the proof. You can
get the proof from any standard number theory book[2, ?, 5].

1. ≡ is an equivalence relation.

2. a ≡ b (mod m) ⇐⇒ a and b leave the same remainder when

divided by m.

3. If a ≡ b (mod m) and c is any integer, then following holds

• a + c ≡ b + c (mod m)
• ac ≡ b (mod m)
• a2 ≡ b2 (mod m)

2.3 Linear Congruence

Now let’s see some of the congruence containing variables, such as 3x ≡
4 (mod 5), x2 ≡ (mod 8), and x2 + 2 ≡ 3x (mod 5). The simplest of
such congruence is linear congruence ax ≡ b (mod m). We will see
that the necessary and sufficient condition for a linear congruence to be

3
solvable.

By a solution of the linear congruence, we mean an integer x0 such

that ax0 ≡ b (mod m). For example, 5 · 4 ≡ 2 (mod 3), so 3 is a solution
is s solution of the congruence 5x ≡ 2 (mod 3).

Now let’s see the necessary and sufficient condition for the existence
of solution for the linear congruence equation.
Theorem 2.3.1. The linear congruence ax ≡ b (mod m) is solvable if
and if (a, m)| b.
Remark. The proof is left to the reader.
For any further information you can see[2, 6]
Before we introduce the new approach we observe the following:
Lemma 2.3.2. Let Z denote the set of integers.Let a, b, c ∈ Z be such
that
a + b ≡ 0 (mod c)
Then, we have
a ≡ 0 (mod c) ⇐⇒ b ≡ 0 (mod c)
Proof. The Proof is trivial and hence left to the reader.

3 Divisibility by 7
Let’s start with divisibility by 7.

Let N be a given positive integer. We express N in the form of 10a+b,

where b is the unit digit of N in decimal representation and a = (N10−b) ,
the rest of N . Now,
(10a + b) − 3(a − 2b) = 7(a + b) ≡ 0 (mod 7)

so,
(10a+b) ≡ 0 (mod 7) ⇐⇒ 3(a−2b) ≡ 0 (mod 7) ⇐⇒ (a−2b) ≡ 0 (mod 7)
[∵ gcd(3, 7) = 1]
Suppose, N 0 := (a − 2b).
Hence,
7|N ⇐⇒ 7|N 0 (1)
This shows that to check whether N is divisible by 7 or not, it is sufficient
to check whether 7 divides N or not.

4
3.1 Note
We observe that the integer N 0 has less digit then N . So, it will be
comparatively easier to work with N 0 . Moreover, the condition (1) is
necessary and sufficient. So, if N is still a big number, then we can
perform the same checking on N . We can continue it until we arrive
at a small number so that it can be checked by inspection whether this
number is divisible by 7.

4 Divisibility test for other integers

Let us explain the method with the help of some examples:

4.1 Example
Suppose we want to check whether 7 divides 13756853284 or not. We
perform the following computations:
1375685328 4 −→ 137568532 0 −→ 13756853 2 −→ 1375684 9 −→
137566 6 −→ 13755 4 −→ 1374 7 −→ 136 0 −→ 13 6 −→ 1.

∵ 7 - 1 so, 7 - 13756853284.
So, we can conclude the following

4.2 Proposition
Let N be any given +ve integer, b be its unit digit and a be the rest of
the number N ,then 7|N ⇐⇒ 7|(a − 2b) as a divisibility test for 7.

4.3 Note
We observe that such a rule is not unique because,
(10a + b) − 3(a − 9b) = 7(a + 4b) ≡ 0 (mod 7).
So, 7|N ⇔ 7|(a − 9b).

We can also find such divisibility test for 11,13,17. Let us discuss
those tests:

Divisibility Test for 11

(10a + b) + (a − b) = 11a ≡ 0 (mod 11)
So, 11|(10a + b) ⇐⇒ 11|(a − b).

5
Divisibility Test for 13

(10a + b) + 3(a − 9b) = 13(a − 2b) ≡ 0 (mod 13)

So, 13|(10a + b) ⇐⇒ 13|(a − 9b).

Divisibility Test for 17

(10a + b) + 7(a − 5b) = 17(a − 2b) ≡ 0 (mod 17)

So, 17|(10a + b) ⇐⇒ 17|(a − 5b).

4.4 Theorem
For every n ∈ N co-prime to 10, we can get such a divisibility test.

Before we proceed for the proof of above result we see that what we
have done in the earlier examples. It is easily visible that we are trying
to find an integer of the for ax + b where x ∈ Z with the property that
n|(10a + b ⇐⇒ n|(a + xb) where n is a positive integer co-prime to 10
whose divisibility test we want to find out. So, our problem is reduced
to -“Can we find such an x for every positive integer n which is co-prime
to 10 ? ”
proof of the Theorem 4.4. Suppose that N = 10a + b be any given
integer where b is its unit digit in the decimal representation and a be
the rest of N.
For example:- Let N = 1846937487 then b = 7 and a = 184693748.
If we can find an integer x such that

(10a + b) + (n − 10)(a + xb) ≡ 0 (mod n) (2)

then clearly this x will have the desired property n|(10+b) ⇐⇒ n|(a+xb)
because gcd(n − 10, n) = 1 since gcd(n, 10) = 1.For proof see[6].
Consider the equation (2),

10a + b + na + nxb − 10a − ‘0xb ≡ 0 (mod n).

which will happen if x satisfies

10x ≡ 1 (mod n) (3)

So the x we are trying to find is basically the solution of the above linear
congruence. This linear congruence has a solution because gcd(10, n) =
1. Moreover, since gcd(10, n) = 1, (3) has a unique solution in modulo

6
n. So, {x0 + nt : t ∈ Z} is the set of all solutions of (3) where x0 is a
solution of (3).

Thus we have infinite number of choices for x. So, now the question
comes that how will we decide our required x. This is not a very difficult
work. We shall make our choice which will make the computations least
complicated. That is why we choose that solution of (3) which will be
negative and have the smallest magnitude.

Now we can find a solution of (3). For this the unit digit of n in
decimal representation is necessary. Since n is co-prime to 10, its unit
digit must belongs to {1, 3, 7, 9}.

The following table shows a particular solution of (3) for different

unit digits of n.
Unit digit of n Particular solution
1 (1 − n)/10
3 (1 − 7n)/10
7 (1 − 3n)/10
9 (1 − 9n)/10
Table-1

If we observe carefully our solutions then we see that all are negative
and have the least magnitude among all the solutions. So, if we choose
these solutions then our calculation become least complicated. Now let’s
illustrate with some examples:

1. N = 214379 and n = 43,

Let’s see in the table-1 the unit digit of 43 is 3, corresponding to
that the solution is (1 − 7n)/10 = −30. It means that
43|N ⇐⇒ 43|(a − 30b)
,where a and b are same as discussed above. Now,
21437 9 −→ 2116 7 −→ 190 6 −→ 10.
∵ 43 - 10 =⇒ 43 - 214379.
See, here the required result comes in 3 steps but if we perform the
usual division then we have to perform 4 steps to get the desired
result and the multiplications and subtractions done in the above
calculation are simpler than the usual division method.

7
 2. N = 45637827 and n = 131,

1−n
The unit place is 1, corresponding to that the solution is 10
= −13
It means that 131|N ⇐⇒ 131|(a − 13b). Now,

4563782 7 −→ 456369 1 −→ 45635 6 −→ 4555 7 −→ 446 4 −→ 39 4 −→ −13

∵ 131 -=⇒ 131 - 45637827.

Thus without performing the division we can determine whether a

given positive integer is divisible by any positive integer co-prime
to 10 by performing some simple calculations.

5 Conclusion
Now we take up the case of divisibility by n where n is a positive inte-
ger co-prime to 10. Applying ‘Fundamental Theorem of Arithmetic’ we
can write n = 2α 5β k where α and β are nonnegative integers and k is a
natural number co-prime to 10. Now, since k is co-prime to 10 so, we
have a divisibility rule for k. We also know that for p ∈ {2, 5}, r ∈ N the
divisibility rule for pr is as follows:

“For any given positive integer N , consider the integer N 0 formed by

its last r digits of decimal representation. Then pr |N ⇐⇒ pr |N 0 .”

So, by applying divisibility tests for 2α , 5β and k separately, we can

easily determine whether any given positive integer N is divisible by n
or not.

References
[1] Walter William Rouse Ball, Garnett Carper, and Florian Cajori. His-
tory of Mathematics. Macmillan, 1919.

[2] David M Burton. Elementary Number Theory. Tata McGraw-Hill

Education, 2006.

[3] David M Burton. The History of Mathematics: An Introduction,

volume 10. McGraw-Hill Science/Engineering/Math;, 2010.

8
[4] R. Health Brown J.Silverman G. H. Hardy, E.M.Wright and A. Wiles.
An Introduction to the Theory of Numbers. Clarendon press Oxford,
6th edition, 2008.

[5] Kenneth Ireland and Michael Ira Rosen. A classical Introduction to

Modern Number Theory, volume 1982. Springer, 1982.

[6] Thomas Koshy. Elementary Number Theory with Applications. Aca-

demic Press, 2002.

9
 Sachchidanand Prasad
Third year Integrated M.Sc.
Department of Mathematics
National Institute of Technology,Rourkela

10
Index
divisibility by 7, 4
divisibility test for 11, 6
divisibility test for 13, 6
divisibility test for 17, 6

Euclidian Algorithm, 3

linear congruence, 3

Table-1, 7

11