A question from the proof of the diameter of curve complex is infinite (Luo's proof)

98 Views Asked by At

In Masur and Minsky's paper, p.14, Proposition 3.6 Luo's proof, it says "but $\gamma_n\to \mu$ and $\mu$ maximal implies $\beta_n \to \mu$", which I feel confused: when $\gamma_n\to \mu$ and $\beta_n$ and $\gamma_n$ are disjoint, I think $\gamma_n$ doesn't necessarily bring $\beta_n$ to converge to $\mu$. Even, thinking intuitively, it seems possible for $\beta_n$ to be farther from $\mu$ as $n$ gets larger. I don't see how combined with $\mu$ maximal would force $\gamma_n$ to make $\beta_n$ converge to $\mu$ as $\gamma_n$ converges to $\mu$.

Notations: $\mu$ is a maximal geodesic lamination and let $\gamma_i$ be any sequence of closed geodesics converging geometrically to $\mu$. For each $\gamma_n$ we may then find $\beta_n$ with $d(\beta_n, \gamma_n) = 1$ and $d(\gamma_0, \beta_n) = N-1$.

1

There are 1 best solutions below

4
On BEST ANSWER

Basically you should be thinking about $\mu$ as the stable lamination coming from a pseudo-Anosov. To help a bit more with the intuition, you should think about $\gamma_n$ as a very long curves on the surface "going all over", and longer and more "all over" as $n$ increases(the sequence of curves is not contained on any subsurface). This will mean $\beta_n$ will have to follow $\gamma_n$ for a long time (so will look like $\gamma_n$, which is why we get this convergence) to be disjoint. At least that is the basic intuition you should have/develop.

Say $\beta_n \to \beta$(maybe passing to subsequence). By disjointness of $\beta_n$ and $\gamma_n$ we get $\beta$ is "disjoint" from $\mu$, so $\beta \cup \mu$ is a lamination. With the assumption that $\mu$ is maximal we get $\beta$ must be a sublamination. The unstated jump in the sketch is that $\mu$ is minimal in the sense that any leaf in the lamination is dense in the lamination(this happens in the pA case). With that you get that $\beta = \mu$. From here you can continue the proof.