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

Don't know how to prove that sum of all divisors of a square number is always odd.

ex: $27 \vdots 1,3,9,27$; $27^2 = 729 \vdots 1,3,9,27,81,243,729$; $\sigma_1 \text{(divisor function)} = 1 + 3 + 9 + 27 + 81 + 243 + 729 = 1093$ is odd;

I think it somehow connected to a thing that every odd divisor gets some kind of a pair when a number is squared and 1 doesn't get it, but i can't formalize it. Need help.

share|cite|improve this question

Yes, you can assign a pair to every odd divisor, except one of them.

So, we have a square number $n^2$. Extract the largest power of $2$ from $n$ to get $n = 2^k m$ where $m$ is odd. Odd divisors of $n^2$ are the same as all divisors of $m^2$. If $d$ is a divisor of $m^2$, then $m^2/d$ is also its divisor. So they all come in pairs, except for $m$ which is paired with itself.

share|cite|improve this answer

The divisors $1=d_1<d_2<\cdots <d_k=n^2$ can be partitioned into pairs $(d, \frac {n^2}d)$, except that there is no partner for $n$ itself. Therefore, the number of divisors is odd. Thus if $n$ itself is odd (and so are all its divisors), we have the sum of an odd number of odd numebrs, hence the result is odd. But if $n$ is even, we can write it as $n=m2^k$ wit $m$ odd. The odd divisors of $n^2$ are precisely the divisors of $m^2$. Their sum is odd and all other (even) divisors do not affect the parity of the sum.

share|cite|improve this answer

Sum of divisors =$\prod_{p_i\mid n}(1+p_i+p_i^2+\cdots+p_i^{2P_i})$ where $p_i$s are distinct primes and $p_i^{P_i}\mid\mid n$

Now, $p_i^r+p_i^{2P_i+1-r}$ is even for all the prime divisor $p_i$ of $n$ for $0< r\le 2P_i+1-r\implies 0<r\le P_i$ leaving $p_i^0=1$ unpaired.

Alternatively, $p_i^k\equiv p_i\pmod 2$ for $k\ge 1$

So, $1+p_i+p_i^2+\cdots+p_i^{2P_i}\equiv 1+\sum_{1\le s\le 2P_i}p_i\pmod 2\equiv 1+2P_is\pmod 2\equiv 1$

share|cite|improve this answer

The formula for $\sigma_1(n)$ for $\displaystyle{n=\prod_{i=1}^kp_i^{\alpha_i}} $ is $\displaystyle{\sigma_1(n)=\prod_{i=1}^k\frac{p_i^{\alpha_i+1}-1}{p_i-1}}$ (see this).

Therefore $\displaystyle{\sigma_1(n^2)=\prod_{i=1}^k\frac{p_i^{2\alpha_i+1}-1}{p_i-1}}$.
But $\frac{p_i^{2\alpha_i+1}-1}{p_i-1}=1+p_i+p_i^2+\ldots+p_i^{2\alpha_i}$ is always odd. Thus $\sigma_1(n^2)$ is odd as product of odd numbers.

share|cite|improve this answer

If number is 144 then 144 = 12 × 12 = 2^4 × 3^2

so every square number is of the form a^2p × a^2q

total number of divisors = (2p + 1)(2q + 1) (2p + 1) and (2q + 1) both are odd numbers, so square number is having odd number of divisors.

share|cite|improve this answer
why does the first line prove "every square number is of the form"? In your example I could choose $a=6, p=2, q=1$ but the number of divisors is not $35$ but $15$ or I could choose $a=1,p=16,q=9$ and the number of divisors is not $170$ – Ross Millikan Jan 21 '14 at 4:27

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.