Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I'm now learning multiple integrals. I reached this formula through computation of multiple integrals: $$ V_{2n}(R)=\frac{\pi^n}{n!~}\cdot R^{2n}, \quad V_{2n+1}(R)=\frac{2^{n+1}\pi^n}{(2n+1)!!~}\cdot R^{2n+1}. $$ Here $V_m(R)$ denotes the volume of a $m$-dimensional ball of radius $R$.

It looks quite strange that $V_{m+2}(R)$ has one more $\pi$ than $V_m(R)$, while the pattern of degree of $\pi$ are not the same between odd dimensions and even dimensions.

Question: How to interpret the power of $\pi$ occur in the formula above?

Geometric interpretations could be nice, since higher dimensional spaces is difficult for me to imagine.

Thank you !

share|improve this question
    
Take a look at this. I am not sure if the power of $\pi$ is important here since there are different multipliers in the front. –  Patrick Li Jan 6 '13 at 6:48
2  
I'd say it has to do with the rank of the groups $SO(2n)$ and $SO(2n+1)$ both of which is $n$. This roughly says that there are $n$ "independent rotations" in both $2n$ and $2n + 1$ dimensional space since for each one you need to pick a plane where the rotation occurs. The rotation in this one plane then corresponds to a circle, which is where the $\pi$ comes from. But take this with a lot more than a grain of salt! :) –  Marek Jan 6 '13 at 8:30
    
@Marek Thank you very very much! I'm really interested in your comment, also looking forward to a more detailed answer:) –  rhenskyyy Jan 9 '13 at 9:30

1 Answer 1

The formula is actually quite symmetric for even and odd dimensions. The Jacobian in $n$-dimensions is: $$J_n = r^{n-1}\prod_{k=1}^{n-2}\sin^k \phi_{n-k-1}$$ And the resultant volume is also symmetric in odd and even powers of $n$: $$V_n \sim \frac{\pi^{n/2}}{\Gamma(n/2+1)}$$ So in essence the difference in odd and even volumes comes from the insistence to convert a perfectly friendly gamma function into factorials, thus leading to an "extra" factor of $\sqrt{\pi}$. Of course you may intuitively expect there to be whole powers of $\pi$ since that is what we are used to in low dimensions, but I think we should just get used to the fact that the familiar $\pi$ is just a reduction of the "real $\pi$" which involves the Gamma function.

share|improve this answer
1  
I don't agree that we should treat the odd and even cases uniformly. There is no sense in which the factor jumps by $\sqrt\pi$ in every dimension. Rather, it only jumps by $\pi$ every two dimensions. There is just a whole world of difference between the odd and even case. E.g. you can't comb the hair on $S^2$ (Hairy ball theorem) but there's no problem doing it on $S^3$. This essentially boils down to Euler characteristic which is $\chi(S^n) = 1 + (-1)^n$. Similarly, observe that antipodal map ($x \mapsto -x$) is not a proper rotation only for odd $n$. –  Marek Jan 9 '13 at 11:27

Your Answer

 
discard

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.