Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If $a,b,c$ are primitive Pythagorean triples such that $c^2=a^2+b^2$, prove that this is unique primitive Pythagorean triple using $c$ as hypotenuse.

I'm not sure if I phrased this correctly. I want to ask, if $c^2=a^2+b^2$ where $a,b,c$ are pairwise coprime, prove that there are no $a_1$ and $b_1$ such that $a_1,b_1,c$ that satisfy $c^2=a_1^2+b_1^2$, and they are also Pythagorean triple.

This is obviously true based on the short list I found on Wiki but I can't seem to prove it... Help?

EDIT Since this obviously seems false, my question now is, if the process described above false only for $c=5k$? Since it seems to be false for $c=65, 145, 185, 205, 265...$ Can one prove this?

share|cite|improve this question
Not sure how to comment- but are how asking how to prove that given $c$ and $a$, $b$ is unique? – Ali Mar 1 '12 at 18:33
up vote 5 down vote accepted

$63^2 + 16^2 = 33^2 + 56^2 = 65^2$

share|cite|improve this answer
Does this happen only for $c=5k$? – Lazar Ljubenović Mar 1 '12 at 18:36
$21^2+220^2=140^2+171^2=221^2$ – Robert Israel Mar 1 '12 at 19:21
@LazarLjubenović: No. See my answer and Bill's answer. – Aryabhata Mar 1 '12 at 19:36

Hint $\rm\ \ \ c^2 = N(a+b {\it i}) = a^2 + b^2,\ \ \ C^2 = N(A+B{\it i}) = A^2 + B^2\ $ so by multiplicativity of norms

$\rm\quad\Rightarrow\ (cC)^2 = N(a\pm b{\it i})\:N(A + B{\it i}) = N(aA\mp bB + (aB\pm bA){\it i}) = (aA\mp bB)^2 + (aB\pm bA)^2 $

e.g. $\ \ \ 5^2 = 4^2 + 3^2,\ \ 13^2 = 12^2 + 5^2\ \Rightarrow\ 65^2 =\: (48\mp 15)^2 + (20\pm 36)^2 =\: \{63^2+16^2,\: 33^2+56^2\}$

share|cite|improve this answer

If the $c$ of the primitive pythagorean triple satisfies,

$c = m^2 + n^2$, then $b = 2mn$ and $a = m^2 - n^2$.

Looking in the guassian integers we see that $c = (m+in)(m-in)$. There are infinitely many $c$ which have multiple factorizations of the form $(a+ib)(a-ib)$.

So what you say is not true.

As to your edit. Any $c$ which has at least two primes factors (in $\mathbb{Z}$) of the form $4k+1$ will do and I believe any $c$ with that propertly must have at least two such prime factors. Note, if you want $m,n$ to be co-prime $c$ should not have any factor of the form $4k+3$.

Here is how you can use the above to generate Robert's example:

$65 = 5*13 = (2+i)(2-i)(3+2i)(3-2i)$

Writing it as

$[(3+2i)(2-i)] [(3-2i)(2+i)]$ gives us $(8-i)(8+i)$

which gives $m=8$, $n=1$ and so $65^2 = 16^2 + 63^2$.

Writing it as

$[(3+2i)(2+i)][ (3-2i)(2-i)]$ gives us $(4+5i)(4-5i)$

which gives $m=7$, $n=4$ and so $65^2 = 56^2 + 33^2$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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