Some remarks on the geodesics in infinite dimensional Teichmüller spaces (Q4392007)

Let \(\Gamma \) be a Fuchsian group acting on the unit disk \(\Delta \). Denote by \(M(\Gamma)\) the set of Beltrami differentials of \(\Gamma\), that is, the set of measurable functions \(\mu \) on \(\Delta \) with essential supremum \(\| \mu \|_\infty <1\) satisfying \((\mu \circ \gamma)\overline{ \gamma^{'}}/\gamma^{'}=\mu \) all \(\gamma \in \Gamma \). The Teichmüller space \(T(\Gamma)\) of \(\Gamma \) is defined as the set of equivalence classes of the elements of \(M(\Gamma)\). A Beltrami differential \(\mu \) in \(M(\Gamma)\) is said to be extremal if for all \(\mu^{'}\) in \(M(\Gamma)\) equivalent to \(\mu\), there holds that \(\| \mu \|_\infty \leq \| \mu^{'}\|_\infty \). Let \(B\) denote the set of points \([ \mu ] \) in \(T(\Gamma)\) with the Teichmüller metric such that a geodesic between the points \([ 0] \) and \([ \mu ] \) is not unique, then a natural question to ask is whether or not there always exists infinitely many geodesics joining two points \([ 0] \) and \([ \mu ] \) if the points \([ \mu ] \) is in \(B\). Theorem 1: For an extremal Beltrami differential \(\mu\), we have: (1) There exists infinitely many geodesics joining \([ 0] \)and \( [ \mu ] \) if and only if there exist infinitely many extremal Beltrami differentials equivalent to \(\mu \), or there exists some extremal Beltrami differential \( \mu_0\) equivalent to \(\mu \) such that the set \(\{z\in \Delta :| \mu_0(z)| <\| \mu_0\|_\infty \}\) has positive measure. (2) There exist precisely \(n(1<n<\infty)\) geodesics joining \([ 0] \) and \([ \mu ] \) if and only if there exist precisely \(n\) extremal Beltrami differentials \(\mu_i(1\leq i\leq n)\) (including \(\mu \)) equivalent to \(\mu\), and each \(\mu_i\) satisfies \(| \mu_i| =\| \mu_\infty \| \) almost everywhere. For an extremal Beltrami differential \(\mu \) in \(M(\Gamma)\), denote by \( K_0(\mu)\) its maximal dilatation and by \(H(\mu)\) its boundary dilatation. Proposition 1: Let \(\mu \) in \(M(\Gamma)\) be extremal. Then the following conditions are all equivalent. (a) \([ \mu ] \in \overline{B};\)(b) \(H(\mu)=K_0(\mu);\) (c) \(\mu \) has a degenerate Hamilton sequence. Proposition 2: For any Fuchsian group \(\Gamma \) infinitely generated or of the second kind, there exists an integrable extremal Teichmüller differential \(k\frac{\overline{\phi }}{| \phi | }(0<k<1,\phi \in Q(\Gamma))\) in \(M(\Gamma)\) which possesses a degenerate Hamilton sequence. Corollary: \(B\) is not closed in \(T(\Gamma)\) when \(T(\Gamma)\) is of infinite dimension.