The theorem is: Let be f, g : X→Y two continuous maps between topogolical spaces and

H : X$\times$[0, 1] a homotopy such that H(x,0)=f(x) and H(x,1)=g(x).

Given $x_0\in X$ let be $y_0=f(x_0)$, $y_1=g(x_0)$ and $\tau(t)=H(x_0,t)$ a path in Y from $y_0$ and $y_1$.

Then, $g_*=\varphi_{[\tau]}\circ f_*$

i.e. $\require{AMScd}$ \begin{CD} \pi_1(X,x_0) @>f_*>> \pi_1(Y,y_0)\\ @V V V @VV \varphi_{[\tau]} V\\ @>>g_*> \pi_1(Y,y_1) \end{CD}


$f_*:\pi_1(X,x_0) \rightarrow \pi_1(Y,y_0),\space f_*([\gamma])=[f\circ \gamma]$,

$g_*:\pi_1(X,x_0) \rightarrow \pi_1(Y,y_1),\space g_*([\gamma])=[g\circ \gamma]$ and

$\varphi_{[\tau]}:\pi_1(Y,y_0) \rightarrow \pi_1(Y,y_1),\space\varphi_{[\tau]}([\gamma])=[\tau^{-1}]·[\gamma]·[\tau]$ In the diagram it's supposed that $\pi_1(X,x_0)$ goes directly to $\pi_1(Y,y_1)$ by $g_*$

$f_*$ and $g_*$ are group homomorphism and $\varphi_{[\tau]}$ is a group isomorphism.

I began supposing

$g_*([\gamma])=\varphi_{[\tau]}\circ f_*([\gamma])=[\tau^{-1}]·[f\circ\gamma]·[\tau] \iff [\tau]·g_*([\gamma])·[\tau^{-1}]·f_*([\gamma]^{-1})=e_{y_0}$

and from here I don't know how to continue, I think there's something with four components which can help me to complete the proof.



I'd rather do this explicitly on the level of paths.

A closed path on $(X,x_0)$ is pushed via $f$ to a closed path on $(Y,y_0)$. $\tau$ is a path that starts at $y_0$ and ends at $y_1$. By composing paths $\tau * (f\circ \gamma) * \tau^{-1}$ a closed path $\gamma$ on $(X,x_0)$ is instead pushed to a closed path $(Y,y_1)$.

A homotopy between this path and $g \circ \gamma$ is constructed via $H$. The "pause" function:

$P: [0,1]\times [0,1] \to [0,1]$, $\quad (s,t) \mapsto \begin{cases}s \quad t<s\\t \quad t≥s\end{cases}$

is continuous.

For any s, $(\tau \circ P)(s,t) = \begin{cases}\tau(s) \quad t<s \\ \tau(t) \quad t≥s \end{cases}$ becomes a path that starts at $\tau(s)$ and follows $\tau$ from there until $y_1$. Let $(\tau \circ P)^{-1}(s,-)$ be (for fixed s) the path that runs backwards through $\tau \circ P$.

For fixed $s$ if we put a closed path into $H$, it starts and ends at $\tau(s)$. So for any s the concatenation of $(\tau \circ P)* H(\gamma,s) *(\tau \circ P)^{-1}$ will be a closed path at $y_1$. In the case $s=0$ this path is $\tau * (f\circ \gamma)*\tau^{-1}$, in the case s=1 it is $g \circ \gamma$.

If you show that map is continuous on $I \times I$ you will have constructed a homotopy between these two paths, so they are in the same homotopy class.

If you want to show that the map is continuous explicitly write out the 3 cases in the path concatenation.

A picture of what the homotopy does is that it deforms the $f\circ \gamma$ part of $\tau * (f\circ \gamma) * \tau^{-1}$ into parts of the $g \circ \gamma$ curve via the homotopy between $f$ and $g$ while unzipping the doubled part of path from $\tau$ and $\tau^{-1}$.


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.