Let $X$ be a mertic space, and let $\{V_{\alpha}\}$ be a collection open subsets of $X$ such that, for every $x \in X$ and for every open set $G \subset X$ with $x\in G$, there is some $V_\alpha$ such that $x \in V_\alpha \subset G$. Then the collection $\{V_\alpha\}$ is said to be a base for $X$.

Now suppose $X$ has a countable base $\{V_1, V_2, V_3, \ldots\}$, and let $\{O_\beta\}$ be an open cover of $X$; that is, let $\{O_\beta\}$ be some collection of open sets such that $X \subset \cup_{\beta} O_\beta$.

Then how to show that some countable subcollection of $\{O_\beta\}$ also covers $X$?

Of course, every element $x \in X$ is in some $O_\beta$, which in turn is a union of some subcollection of the countable collection $\{V_1, V_2, V_3, \ldots\}$. What next? How to obtain a countable subcollection of $\{O_\beta\}$?


well you need the axiom of choice. since $\{V_\alpha\}$ is countable, and is a basis, for every $V_i$ choose $O_i \in \{O_{\beta}\}$ satisfying $V_i \subset O_i$. we may write $O_i = \psi(V_i)$

every $x \in X$ is covered by at least one member of $\{O_{\beta}\}$. since this is open it must contain some $V_x$ with $x \in V_x$. now take the corresponding $O_x = \psi(V_x)$

the collection $\{O_x\}_{x \in X}$ is countable and covers $X$

  • 2
    $\begingroup$ You don’t actually need the full strength of the axiom of choice; the result is actually equivalent to $CC(\Bbb R)$, the axiom of countable choice for subsets of $\Bbb R$, which says that if $\{A_n:n\in\Bbb N\}$ is a family of non-empty subsets of $\Bbb R$, then $\prod_{n\in\Bbb N}S_n\ne\varnothing$. More important, it’s not true that every member of the base is contained in some member of the open cover. For each $i$ let $\mathcal{O}_i=\{O\in\mathcal{O}:V_i\subseteq O\}$, and then choose one member of each non-empty $\mathscr{O}_i$. $\endgroup$ – Brian M. Scott Oct 11 '14 at 5:15
  • $\begingroup$ thank you Brian. i was hoping someone might clarify what weaker assumption would suffice. and thank you also for picking up on the empty set loose end in my argument. i did have a feeling of unease, but being still somewhat of a tyro, i ignored it and blundered on $\endgroup$ – David Holden Oct 11 '14 at 5:22
  • $\begingroup$ You’re welcome. $\endgroup$ – Brian M. Scott Oct 11 '14 at 5:28
  • $\begingroup$ David Holden, how do we know that for every $V_i$, there does exist a \textit{single} $O_\beta$ that covers this $V_i$? $\endgroup$ – Saaqib Mahmood Oct 15 '14 at 0:03
  • 1
    $\begingroup$ David Holden, Yes, he has, but I'm unable even to comprehend his argument. What if no $V_i$ could be covered by a single $O_\beta$? Or what if some could be so covered, while others could not? $\endgroup$ – Saaqib Mahmood Oct 15 '14 at 1:18

What you are asking is to prove that every second countable space is Lindelof (in more common notation).

So, let's prove $\text{Second Countable}\implies\text{Lindelof}$.

Let $X$ be second countable with countable basis $\mathscr{B}$, and let $\Omega=\left\{U_\alpha\right\}_{\alpha\in\mathcal{A}}$ be an open cover for $X$. By assumption, for each $\alpha\in\mathcal{A}$ we can cover $U_\alpha$ with some collection $\mathscr{B}_\alpha$ of elements of $\mathscr{B}$. Note then that $\displaystyle \Sigma=\bigcup_{\alpha\in\mathcal{A}}\mathscr{B}_\alpha$ is a countable open cover for $X$. So, for each element $O$ of $\Sigma$ choose an element $U$ of $\Omega$ containing it (by the axiom of choice) and define the function $f: \Sigma\rightarrow\{U_{\alpha}\}_{\alpha\in\mathcal{A}}$ by $f(O) = U$. Then, $\Gamma = f(\Sigma)$ is an open cover of $X$ (since its union contains the union over all the elements of $\Sigma$ which is $X$) and is countable since $f$ is a surjection and $\Sigma$ is countable. Thus, $\Gamma$ is our desired countable subcover of $\Omega$.

  • $\begingroup$ Paul, thank you for your answer, but are you sure you haven't made a mistake with the symbols and notation you've used? I'm afraid I'm not being able to fully make sense of your answer. $\endgroup$ – Saaqib Mahmood Oct 15 '14 at 0:01

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.