8

This is regarding the fundamental group computation of the complement of a toral knot in $S^3$ in Hatcher's algebraic topology book. See page 48. I have understood till the stage where the cross section of the torus minus the knot deformation-retracts to the radial segments as the arrows indicate. What is not clear is "Letting $x$ vary, these radial segments then trace out a copy of the mapping cylinder $X_m$ in the first solid torus."

I tried imagining this with a simple cases like the trefoil knot, but can't fathom this statement. Any help would be greatly appreciated!

Karthik C
  • 2,581
  • 7
    It helps to think of the 3 sphere as the union of two solid tori. The circles hatcher uses in the mapping cylinder construction are the cores of the solid torus plus a curve parallel to the knot in the separatingtorus. – Ryan Budney Jul 11 '12 at 16:44
  • 1
    Never thought of it this way! Will chew upon this. Thanks a ton! – Karthik C Jul 11 '12 at 16:58

2 Answers2

9

It helps to think of the 3 sphere as the union of two solid tori. The circles Hatcher uses in the mapping cylinder construction are the cores of the solid torus plus a curve parallel to the knot in the separating torus.

Ryan Budney
  • 22,958
  • If this is the case, then even if each torus deformation retracts onto each of these cylinders it doesn’t deformation retract onto X_mn, the original required space, since the two « cylinders » don’t intersect in the required circle no? – Little Narwhal Feb 13 '22 at 19:54
  • @LittleNarwhal the two cylinders should intersect on the curve parallel to the knot in the separating torus - which is a copy of $S^1$. This is also the 'domain' of either mapping cylinder, so it is precisely what deformation retracts onto the respective ends. – porridgemathematics Mar 11 '22 at 14:42
2

To expound a bit:

The center of the "Y" figure is a point of $S^1$ of the mapping cylinder. This center also wraps around the torus once as $x$ varies. However, the tips of the "Y" wrap around the torus $m$ times.