1
$\begingroup$

Find fundamental group of space $X=(S^1\times S^1)/(S^1\times \{s_0\})$

I am not sure how to do it. It seems obvious that $\pi_1(X)=\mathbb{Z}$ because $X$ is torus where one class of nontrivial loops is contracted to the point.

There is also fact I found which may be used here to show that $X$ is homotopically equivalent to $S^2\vee S^1$ which would surely make above result obvious. However I am not sure how to use this fact which says:

For $i=1,2$ let $Z_i=X \sqcup_{h_i}X$ be topological space resulting from gluing $\overline{B^n}$ to $X$ alongside map $h_i:S^{n-1}\to X$. If $h_1$ and $h_0$ are homotopic, then $Z_0$ and $Z_1$ are homotopically equivalent.

$\endgroup$
  • 1
    $\begingroup$ You may either use the fact that $X$ is homotopic to $S^2 \vee S^1$ as you say, which is best proved pictorially (yes most would consider this a perfectly rigorous proof), or one may prove this by using the Seifert-van Kampen theorem using a covering of the space by a small neighbourhood around the contracted circle and its complement. $\endgroup$ – Dan Rust Jul 2 '18 at 12:02
  • $\begingroup$ Here's one point where the groupoid-version of Van Kampen's theorem is very useful, because there's a very interesting contractible subset in this space $\endgroup$ – Max Jul 2 '18 at 14:55
1
$\begingroup$

Let $X$ be your space, and note that it can be realized as a quotient of $I^2$. In particular, the usual structure of the torus is a quotient map $q:I^2 \to I^2/\sim$ by taking $(x,0) \sim (x,1)$ and $(0,y) \sim (1,y)$.

There is another quotient now $p:I^2 \to I^2/\sim$ by taking $(x_1,0) \sim (x_2,0)$ and $(x_1,1) \sim (x_2,1)$. Your space is done by doing $q$ first and then $p$ (or technically the map induced by $p$.)

However, the situation is clearer if you do $p$ first and then $q$.

Pinching two sides of the square down to a point gives a structure of $2$ zero cells, and two $1$-cells joining them, and filling this with a $2$-cell. The induced bap by $q$ glues these two $1$ cells together giving a sphere, but we still have to identify the $0$-cells, so this is a sphere with two points identified.

There is an excellent illustration that a sphere with two points identified is homotopically just $S^2 \vee S^1$ here.

Now one can apply siefert-Van kampen to obtain the result.

$\endgroup$

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.