I want to know whether there is a way of proving that $S_g$ (the orientable surface of genus $g$) covers $S_1$ (the torus) if and only if $g=1$ without invoking the Euler characteristic. I know that any covering $S_g \rightarrow S_1$ induces an injection on fundamental groups, but does this also necessarily yield an injection on the abelianizations for example? In that case, the assertion would follow.

I also found the older post Surface of genus $g$ does not retract to circle (Hatcher exercise) in which the top answer asserts that a certain map does induce an injection on abelianized fundamental groups, but I do not understand why.


If $S_g \to S_1$ is a covering, then the induced map on fundamental groups $\pi_1(S_g) \to \pi_1(S_1)$ is injective. Therefore $\pi_1(S_g)$ is isomorphic to a subgroup of $\pi_1(S_1)$. As $\pi_1(S_1)$ is abelian, every subgroup of $\pi_1(S_1)$ is abelian. For $g > 1$, $\pi_1(S_g)$ is non-abelian and therefore there is no covering $S_g \to S_1$.

  • $\begingroup$ Thanks! I was being silly, thinking that the fundamental group of the torus was the free product. $\endgroup$ – Thomas Bakx Apr 10 '18 at 14:44

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.