The quotient of a graph by its maximal tree.

42 Views Asked by At

I was just reading the proof that a graph is a tree iff it is simply connected in Hatcher, on pg.85, which is given below:

The quotient map $X \to X/T$ is a homotopy equivalence by Proposition $0.17.$ The quotient $X/T$ is a graph with only one vertex, hence is a wedge sum of circles, whose fundamental group we showed in Example 1.21 to be free with basis the loops given by the edges of $X/T$ , which are the images of the loops $f_α$ in $X$.

The statement of Prop.0.17 is as follows:

If the pair $(X,A)$ satisfies the homotopy extension property and $A$ is contractible, then the quotient map $q: X \to X/A$ is a homotopy equivalence.

My questions about the proof are as follows:

1-Why the quotient map is a homotopy equivalence by proposition 0.17? Why is the pair $(X,T)$ satisfies the homotopy extension property?

2- I want to see a lot of examples of quotienting a graph by its maximal spanning tree please and see how the above proof is correct.

Can anyone clarify these points to me please? Thanks in advance!

1

There are 1 best solutions below

12
On BEST ANSWER

As for 1, note that $T$ is a subcomplex of $X$ by definition, i.e. $(X, T)$ is a CW-pair and thus has the homotopy extension property by proposition 0.16 (see also example 0.14).

As for 2, here's one class of examples you could consider: Take $X = \mathbb{R}$ with the usual CW-structure, i.e. we identify $X_0$ with $\mathbb{Z}$ and then glue in one 1-cell from with end points $n$ and $n + 1$ for each $n$. This is a graph and a maximal spanning tree is just $\mathbb{R}$ itself, so this shows that $\mathbb{R}$ is contractible (it might be helpful to go through the proof an construct an explicit contracting homotopy with the given recipe). Now expand $X$ by taking a larger set of 1-cells: For instance, glue in an edge from 2 to 5, or from $2n$ to $2n + 2$ (for all $n$), or glue in infinitely many edges from 0 to 1, etc. The subcomplex $\mathbb{R} \subset X$ is still a maximal spanning tree, and all the additional edges you glued in will form loops on the unique 0-cell after contraction.

For another flavor of example, take a piece of paper, draw a few nodes (your 0-cells) and then draw edges between those nodes (your 1-cells), however many you want, but making sure the result is connected. Find a maximal spanning tree and then work through the contraction by hand.

There really isn't any more to it :)