MathOverflow is a question and answer site for professional mathematicians. 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

Let $X$ be a finite CW complex and $x_0$ a point in $X$.

My question is then just:

Is $X-\{x_0\}$ still homotopy equivalent to a finite CW complex?

share|cite|improve this question
The answer is yes if $x_0$ lies in the interior of a top dimensional cell. – John Klein Jun 11 '12 at 13:41
The answer is also yes if $X$ is a regular CW-complex (i.e. if all its attaching maps are homemorphisms). – André Henriques Jun 11 '12 at 14:38
up vote 42 down vote accepted

The answer to your question is no. Here is a counterexample.

Let $X$ be the CW-complex obtained by attaching a 2-cell to the space $[-1,1]$ via the attaching map $S^1\cong [-1,1]/(-1\sim 1) \longrightarrow [-1,1]$ given by (the continuous extension of) $x\mapsto x\sin(1/x)$.

Then $X\setminus \{0\}$ is homotopy equivalent to an infinite wedge of $S^1$'s.

share|cite|improve this answer
All I have to say about this example is "yuck". Good answer though. – Spice the Bird Jun 11 '12 at 15:30
There's nothing yuck about this particular example. But I'll agree that the notion of CW-complex is a bit "yuck". Depending on one's perspective, the notion of CW-complex is either too general, or not general enough. The two notions that I believe to have a "good level of generality" are 1) regular CW-complex, and 2) retract of a CW-complex. – André Henriques Jun 11 '12 at 15:42
@Matin: I guess that Andr\'e thinks of $\mathrm{S}^1$ as the quotient of the interval $[-1,1]$ by its boundary. Its attaching map takes the same value at $-1$ and at $1$. – Oblomov Jun 12 '12 at 8:11
@Andr\'e: Actually, are you sure that the resulting space is really a a infinite wedge of circles. It seems to me that it might even not be a CW complex but rather a ''sort of Hawaian earrings`` space. – Oblomov Jun 12 '12 at 9:02
The Hawaiian earing space has a sequence of loops that converge to a constant loop. In the space I constructed, there is no sequence of loops converging to a constant loop, for the simple reason that the limit point has been removed. Here's another example where a similar phenomenon shows up: $\mathbb R^2\setminus \{1/2,1/3,1/4,...\}$ is h.e. to the Hawaiian earings, but $\mathbb R^2\setminus (\{0\}\cup\{1/2,1/3,1/4,...\})$ is homotopy equivalent to an infinite wedge of $S^1$s. – André Henriques Jun 12 '12 at 9:07

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.