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

So I'm going through Otto Forster's "Lectures on Riemann Surfaces", and I need another hint (shame). This is in the "Cohomology Groups" sections, as part of a problem to show that for $X$ a compact Riemann Surface, $H^1(X,\mathbb{C})$ is a finite dimensional $\mathbb{C}$-vector space.


Suppose $X$ is a manifold, $U \subseteq X$ is open, and $V \subseteq U$ is relatively compact in $U$. Show that $V$ meets at most finitely many connected components of $U$.

My Attempt (so far)

Let $\{U_i\}_{i \in I}$ be the connected components of $U$, with the $U_i$ disjoint, and $I$ perhaps infinite. Since all the $U_i$ are open (in $U$), and $V \subseteq U$, $\{V \cap U_i\}_{i \in I}$ is an open cover of $V$ (relative to $V$). Since, by assumption, the closure of $V$, which I'll denote $\overline{V}$, is compact in $U$, and $\{\overline{V} \cap U_i\}_{i \in I}$ is an open cover of $\overline{V}$, there exists a finite subcover $\{\overline{V} \cap U_i\}_{i=1}^n$ (sorry for the run-on sentence...).

I then (want to) claim that $V$ meets only the finite set of components $\{U_i\}_{i=1}^n$ given by the finite subcover of $\overline{V}$. This is where I'm getting stuck. Obviously, I need to work in the assumption that $X$ is a manifold someone in here, but I'm not seeing it yet.

Am I on the right track, or should I scrap it?

share|cite|improve this question
up vote 2 down vote accepted

Finitely many $\bar{V} \cap U_i$ covers $\bar{V}$, and remember that $U_i$ are disjoint to start with. What can you say about all the $\bar{V} \cap U_j$ that does not show up in your finite subcover?

P.S. You already used that $X$ is a manifold when you say that $U_i$ are open. Connected components in general may not be open.

share|cite|improve this answer
Ahhh, thank you. So, suppose there is some $j$ such that $\overline{V} \cap U_j \neq \emptyset$, with $j \neq 1,\cdots, n$. Then $x \in \overline{V}$, and since the $\{\overline{V} \cap U_i\}_{i=1}^n$ cover $\overline{V}$, there is some $i=1,\cdots, n$ such that $x \in \overline{V} \cap U_i \cap U_j$. But the $U_i$ are disjoint, so this is a contradiction. Gotcha. And since $V \subseteq \overline{V}$, this means $V$ meets only finitely many of the $U_i$. – Brian H Feb 10 '13 at 20:38

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.