Let $T^n = (S^1)^n$ be the $n$-torus and $G = (\mathbb{Z}_2)^n$ be product of $n$ copies of the group of order $2$. Consider the action of $G$ on $T^n$ induced by the antipodal action componentwise.

Computing the cohomology $H^*(T^n/G;\mathbb{Z}_2)$, I am tempted to say that $T^n/G \cong (S^1/\mathbb{Z}_2)^n$ since the action happens componentwise, but I am not sure if that is correct. Also, I am trying to avoid a Spectral sequences argument, but if there is a solution in that way will be useful for future computations.


Your temptation is correct. If two elements of $T^n$ are related by an element of $G$, that just means that on each coordinate they are related by an element of $\mathbb{Z}_2$. So an orbit of $G$ consists of just a product of orbits of $\mathbb{Z}_2$ on each component. This gives a bijection $f:T^n/G\to (S^1/\mathbb{Z}_2)^n$ which is easily seen to be continuous using the universal property of $T^n/G$. Since $T^n/G$ is compact and $(S^1/\mathbb{Z}_2)^n$ is Hausdorff, you can conclude that $f$ is a homeomorphism.

Since $S^1/\mathbb{Z}_2\cong S^1$, this means that $T^n/G\cong T^n$, so you are just taking the cohomology of $T^n$.

  • $\begingroup$ Thanks @Eric Wofsey for your answer. $\endgroup$ – C. Zhihao Oct 12 '17 at 13:56

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.