I'm trying to prove the following statement:

Let $X$ be a topological space with a countable open cover $C= \{U_\alpha : \alpha \in A\}$. If each $U_\alpha$ is second-countable then $X$ is second-countable.

However I'm stuck on how to proceed, so any hints on how to tackle it would be highly appreciated.


Let $\mathcal{B}_\alpha$ be a countable base for $U_\alpha$. Its members are by definition open in $U_\alpha$, and as all $U_\alpha$ are open in $X$, these sets are also open in $X$. So, $\mathcal{B}=\bigcup_{\alpha} \mathcal{B}_\alpha$ is a countable family of open sets in $X$. Suppose that $x \in X$ and $O$ is open in $X$ with $x \in O$. Then $x \in U_{\beta}$ for some index $\beta$. Now apply the definition of a base to see that for some $B \in \mathcal{B}_\beta$ we have $x \in B \subseteq O \cap U_\beta$. This $B \in \mathcal{B}$ and $ x \in B \subseteq O$. This shows that $\mathcal{B}$ is a countable base for $X$.


Hint: Fix a countable basis $\mathcal{B}_\alpha$ for each $U_\alpha$, viewed as a subspace of $X$. There are countably many $U_\alpha$ and each of them has a countable basis. Will $\bigcup_\alpha \mathcal{B}_\alpha$ get the job done?

Is $X\subset \bigcup_{\alpha}\bigcup_{V\in \mathcal{B}_\alpha}V$? Given two elements $A,B \in\bigcup_\alpha \mathcal{B}_\alpha$ with $x\in A\cap B$ can we find $C\in \bigcup_\alpha \mathcal{B}_\alpha$ such that $x\in C\subset A\cap B$?

  • $\begingroup$ The last argument only shows it is a base for some topology. Something else is required. It should be a base for the given topology. $\endgroup$ – Henno Brandsma Jun 22 '17 at 20:21
  • $\begingroup$ @HennoBrandsma So this means after show that $\bigcup_{\alpha} \mathcal{B}_{\alpha}$ generated a topology for $X$, we have to show that this topology is equal to initial top on $X$ ?. $\endgroup$ – Sou Jan 14 '18 at 18:21
  • $\begingroup$ @Sou yes, you should also do that. $\endgroup$ – Henno Brandsma Jan 14 '18 at 18:27
  • $\begingroup$ @HennoBrandsma Thank you. $\endgroup$ – Sou Jan 14 '18 at 19:00
  • $\begingroup$ @Sou if you follow my proof, it's not needed. $\endgroup$ – Henno Brandsma Jan 14 '18 at 19:01

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.