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

Show that there are infinitely many solutions of distinct natural numbers $m,n$ such that $n^3+m^2\mid m^3+n^2$.

This question appeared in Round $2$ of British Math Olympiad $2007-08$. I have been trying this problem since two days. And got no sequence of numbers satisfying it.

share|cite|improve this question
This solution ( is similar (if not identical) to the current one. I am still expecting something more intuitive – lab bhattacharjee Jul 1 '13 at 16:03
@labbhattacharjee: I believe parametric form $n^5-n^2-1$ and $k=n^3-1$ is the smallest of a kind. I'm guessing there are more polynomials $p(n) \text{ and } q(n)$ for which it is true. Thanks for the link, btw. – Inceptio Jul 2 '13 at 15:28
@labbhattacharjee I've added an intuitive explanation, and another family of solutions. – Calvin Lin Jul 5 '13 at 16:51
up vote 9 down vote accepted

Choose $a\in\mathbb{N}$, and set $m=an$. Then the condition becomes $n^3+a^2n^2|a^3n^3+n^2$, which is equivalent to $$n+a^2|a^3n+1$$

We now set $n=a^5-a^2-1$ and see that $$(a^3-1)(n+a^2)=(a^3-1)(a^5-1)=a^8-a^5-a^3+1=a^3n+1$$

Update, pulling back the veil: I looked for $k$ such that $k(n+a^2)=a^3n+1$ and I could solve for $n$. After one or two false attempts I tried $k=a^3-1$.

share|cite|improve this answer
Wow! Was there anything in particular that led you to set $n$ to $a^5-a^2-1$? – Milind Jul 1 '13 at 15:07
I recognized that $(m,n)=n$. I couldn't move anywhere after $n+k^2|k^3n+1$. Awesome observation(+1) – Inceptio Jul 1 '13 at 15:13
@MilindHegde Here's an easier way to motivate the work (as a reasonable attempt during the competition). We want $n + a^2 | a^3n + 1$, which gives us $n+a^2 | a^3n + a^5 - a^5 + 1$, hence $n+a^2 | a^5 - 1$. An obvious try is $n = a^5 - a^2 - 1$. Another is $a^4 + a^3 + a + 1$. The rest do not work well. – Calvin Lin Jul 5 '13 at 16:39

As Vadim showed, with $m=an$ yields

$$ n + a^2 | a^3 n + 1 $$

This implies that

$$ a^5 - 1 \equiv a^5 + a^3 n - a^3n - 1 \equiv 0 \pmod{a^2 + n}$$

This suggests immediately to use $n = a^5 - a^2 - 1$, as in Vadim's solution. The advantage is that I don't need to do any guesswork, or expand the product.

In fact, since $ a^5 - 1 = (a-1)(a^4+a^3+a^2+a+1)$, we could also use

$$n = a^4 + a^3 + a + 1 $$

This also shows you how to characterize all solutions. Start with any positive integer $a$, calculate the factors of $a^5-1$ that are greater than $a^2$, and that determines the value of $n$.

Note: Because we want $n$ to be positive, none of the other factors (with integer coefficients) will work to generate a family of solutions.

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.