An Example of a Retract that is NOT a Deformation Retract
Let $X$ be a topological space and let $A \subseteq X$ be a topological subspace. Recall the following definitions:
- On the Retract Subspaces of a Topological Space page we said that $A$ is a retract of $X$ if there exists a continuous function (called a retract) $r : X \to A$ such that:
- On the Deformation Retract Subspaces of a Topological Space page we said that $A$ is a deformation retract of $X$ if there exists a continuous function $r : X \to A$ such that:
We are now able to give an example of two topological spaces $A$ and $X$ where $A$ is a retract of $X$ but such that $A$ is not a deformation retract of $X$.
Let $X = S^1\times S^1$ be the torus and let $A \subseteq X$ be the equatorial circle of $X$ depicted below:
Showing that $A$ is a retract of $X$:
Define a function $r : X \to A$ as follows. Cut $X$ up into vertical circle sections. Then each circle intersects $A$ uniquely. Define $r(x)$ to be the unique point $a \in A$ for which the unique circle intersects both $x$ and $a$.
Alternatively, recall that the torus $X$ can be identified as a quotient topological space by taking $[0, 1] \times [0, 1]$ and identified opposite edges of the square in the same orientation. Then the circle $A$ can be viewed as a horizontal line segment in $[0, 1] \times [0, 1]$ and we map every point in $[0, 1] \times [0, 1]$ to its vertical projection onto this line segment:
It is easy to see that $r$ is a continuous map. This is because if $U$ is an open subset of $X$, then $r^{-1}(U)$ will be a union of open intervals in the equatorial circle. So $r^{-1}(U)$ is open. Furthermore, we have that $r(a) = a$ for all $a \in A$, and so:
(3)So $A$ is a retract of $X$.
Showing that $A$ is NOT a deformation retract of $X$.
Observe that:
(4)These fundamental groups are not isomorphic and so $A$ cannot be a deformation retract of $X$.