Let's define infinite projective space space $\mathbb{CP}^{\infty}$ as direct limit $\lim \limits_{\rightarrow} \mathbb{CP}^n$. In a class I attended it was claimed that every continuous map $S^k \to \mathbb{CP}^{\infty}$ is actually valued in some subspace $\mathbb{CP}^N$ for some sufficiently high $N$. How to prove this?

  • $\begingroup$ A subspace of a CW-complex is compact iff it is closed and it intersects finitely many cells, see for example this question $\endgroup$ – Alessandro Codenotti Jan 31 at 19:01
  • $\begingroup$ If $f_i: X_i \to X_{i+1}$ is a sequence of inclusions of compact Hausdorff spaces, then the direct limit topology on the union is such that any compact subset lies in one of the $X_i$. This is a nice exercise with the definitions. $\endgroup$ – user98602 Jan 31 at 19:06
  • $\begingroup$ @MikeMiller this is precisely what I expected, but I had some trouble with verifying this property. Could you prove some additional hints? $\endgroup$ – Blazej Jan 31 at 20:04

Here's a very general statement which is enough for your situation:

Let $X_i \to X_{i+1}$ be a sequence of embeddings of $T_1$ spaces. Then any compact subset $K\subset \mathrm{colim}_i X_i$ is actually contained in an $X_i$.

A proof can be found here : Compact subset in colimit of spaces

It's a very general statement, and the proof is not hard to follow.

  • $\begingroup$ As pointed out in a comment, my argument was false. I have deleted it. $\endgroup$ – user98602 Feb 1 at 18:26
  • $\begingroup$ @MikeMiller : thanks for pointing it out, I had not checked the argument, only the statement $\endgroup$ – Max Feb 1 at 18:30
  • $\begingroup$ Thank you! I hoped for something exactly of this type, but I expected much stronger assumptions to be necessary, e.g. each $X_i$ compact Hausdorff. $\endgroup$ – Blazej Feb 1 at 21:42

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.