I have two problems, I´m a little complicated with the problem , I know that it´s easy but I need just a little help. Are two problems

i) Suppose that the identity map $ i:X \to X $ is homotopic to a constant path, and the space Y is path connected, then the set of homotopy classes of maps of X into Y has a single element.

ii) Let $ p:E \to B $ be a open covering , B connected. Show that if for some $ b_0 $ we have $ \left| {p^{ - 1} \left( b_0 \right)} \right| = k\,\,\, $ then for every b we have $ \left| {p^{ - 1} \left( b \right)} \right| = k $

My try:

i) Im not sure if this is ok, but i "proved" that any path $ f:X \to Y $ is homotopic to a constant map, and this implies that Y is simply connected. But i did not used the fact that Y is path connected, I think that something that I did, is wrong. Let f be a path from X to Y. Since the identity in X is nulhomotopic, then there exist a homotopy $ F:I\times X \to X $ such that: $ \eqalign{ & F\left( {0,x} \right) = i\left( x \right) = x \cr & F\left( {1,x} \right) = k_0 \in X \cr} $

Then the homotopy $ H = f \circ F:I\times X \to \,Y $ "deforms f to a point " since $ \eqalign{ & H\left( {0,x} \right) = f\left( x \right) \cr & H\left( {1,x} \right) = f\left( {k_0 } \right) \in Y \cr} $

For ii) I suppose that the set of elements such that its preimage is "k" (for every k ) elements must be open, but I´m not sure if this is true, if this is not true, I have no idea , How to do the problem.

Help please Dx


For (i) you've done a lot of the work, but some care is needed. You have indeed shown that any $f\colon X \to Y$ is homotopic to a constant map, but if $Y$ is not path connected then there may be different constant maps from $X$ to $Y$ that are not homotopic. As a silly example, consider the space of maps $\{x\} \to \{y_1, y_2\}$, where these finite sets have the discrete topology.

So, picking up from where you leave off and using that homotopy is an equivalence relation, suppose we have constant maps $f, g\colon X \to Y$. As $Y$ is path connected, there is a path $\gamma$ from $g(k_0)$ to $h(k_0)$. Use this to construct a homotopy between $f$ and $g$ (At this stage, $X$ does not play a significant role).

I would also avoid thinking of these maps $X \to Y$ as paths. A path in $Y$ is a map $S^1 \to Y$, and $S^1$ is not nullhomotopic. So you certainly can't conclude that $Y$ is simply connected, and indeed that need not be the case.

For (ii), I think you have the right idea; you just need to make an argument. Take an open neighborhood $U$ of $b$ such that $p^{-1}(U)$ is a disjoint union of $\{V_i\}_{i \in I}$, each $V_i$ mapping homeomorphically onto $U$. If $b'$ is another element of $U$, then define a map $\Phi\colon p^{-1}(b) \to p^{-1}(b')$ as follows. If $e \in p^{-1}(b)$, then $e$ is an element of $V_i$ for some (unique) $i \in I$. Show that there is a unique element $e'$ of $V_i$ such that $p(e') = b'$. Set $\Phi(e) = e'$, and show that this is a bijection.

It follows that if $p^{-1}(b)$ has $k$ elements, then so does $p^{-1}(b')$ for $b'$ in an open neighborhood around $b$. You can say something similar if $p^{-1}(b)$ does not have $k$ elements. Now use connectedness.

Let me know if I should say more!

  • $\begingroup$ And How can I use the hypothesis of being path connected in i) ? $\endgroup$ – Daniel Oct 30 '11 at 18:28
  • $\begingroup$ @Daniel Updated. $\endgroup$ – Dylan Moreland Oct 30 '11 at 18:48
  • $\begingroup$ Thanks! And in the problem ii) At least the problem, it´s from Munkres, section 53, problem 3. I don´t know If I need more hypothesis or not. $\endgroup$ – Daniel Oct 30 '11 at 19:22
  • $\begingroup$ @Daniel Woops, paths on the brain, I guess. Fixed it up. $\endgroup$ – Dylan Moreland Oct 30 '11 at 19:45
  • $\begingroup$ I´m not sure, how to construct, the new homotopy D: $\endgroup$ – Daniel Nov 1 '11 at 20:19

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.