I recently came across this one observation in my research on right triangles:

In any Pythagorean triangle, the sum of the hypotenuse and the base is a multiple of, at least one factor of the base, other than 1.

eg: Let $a = 28$, $b = 45$, $c = 53$.

Then $45 + 53 = 98$, with who $28$ shares a common factor: $7$.

I have been able to prove this when $a = 2m$, $b = m^2 - 1$, $c = m^2 + 1$ AND when $a = x^{1/2}$, $b = (x-1)/2$, and $c = (x + 1)/2$

I have tried and tested the same for many triangles which do not follow the above mentioned criterion, and they seem to work fine, like the example given above. I tried to use Dickson's method of taking $a = r + s$, $b = r + t$, and $c = r + s + t$, as it would exhaust all triples. It did not work out fine. Does anybody have any other method on how to go about it?


S Sandeep.


You don't need a production scheme for all pythagorean triples to solves this problem.

From $a^2=c^2-b^2=(c+b)(c-b)$ it follows that any prime factor of $c+b>1$ must occur in $a$.


The following will generate all Pythagorean triples uniquely $$a=k(m^2-n^2)=k(m-n)(m+n),\ \ \ b=k\cdot(2mn),\ \ \ c=k(m^2+n^2)$$ where $m, n$, and $k$ are positive integers with $m \gt n, m − n$ odd, and with $m$ and $n$ coprime such that $a^2+b^2=c^2$. (For example, $(k,m,n,a,b,c)=(1,7,2,45,28,53)$ is the case you wrote.)

Case 1 : $c+b=k(m^2+n^2)+k\cdot(2mn)=k\cdot (m+n)^2$ and $a=k(m-n)(m+n)$ have a common factor $k(m+n)\gt 1$ (even if $k=1$).

Case 2 : $c+a=k(m^2+n^2)+k(m^2-n^2)=2m^2k$ and $b=2mnk$ have a common factor $2mk\gt 1$ (even if $k=1$).


Since $a, b, c$ share a common factor, then the result becomes trivial. Therefore, let's consider the case where $a, b, c$ do not have a common factor.

$a^2 + b^2 = c^2 \implies a^2 = c^2 - b^2 = (c+b)(c-b)$

Now, by fundamental theorem of arithmetic, let $a =\prod\limits_{p_i \in \text{Primes}} p_i^{a_i} \implies a^2 = \prod\limits_{p_i \in \text{Primes}} p_i^{2 a_i}$.

Think of above statement as $a$ being written as product of its prime factors, like $a = p_1 \cdot p_2^3 \cdot p_3 \cdots$.
$a^2$ will be the same expression, squared.

Therefore, $(c+b)(c-b) = a^2 = \prod\limits_{p_i \in \text{Primes}} p_i^{2 a_i}$.

Therefore, each prime $p_i$ in the prime factorization of $a$ has to appear at least twice in the product $(c+b)(c-b)$.
Now, $(c+b) > (c-b) \ge 1$. This is because least possible value of $c-b$ is $1$. So, $c+b$ contributes at least one prime $p_i$ to the overall product (possibly more, or even all when $c-b = 1$).

But $p_i$ also present in $a$. Therefore, $c+b$ shares at least one factor with $a$.

  • $\begingroup$ Could you please elucidate the answer? My mathematics is quite rusty. $\endgroup$ – Sandeep Nov 27 '14 at 11:14
  • $\begingroup$ @Sandeep Which part exactly do you need more explanation on? $\endgroup$ – taninamdar Nov 27 '14 at 11:42
  • $\begingroup$ The number theory (the fundamental theory of arithmetic) part. $\endgroup$ – Sandeep Nov 27 '14 at 12:19
  • $\begingroup$ Fundamental theorem of arithmetic is just fancy way of saying that every integer can be uniquely represented as product of primes. $\endgroup$ – taninamdar Nov 27 '14 at 15:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.