No label defined (Q4392550)

Let \(\Gamma\) be a compact Riemann surface of genus \(g\geq 2\). A point \(P\in\Gamma\) is said to be a 2-Weierstrass point of \(\Gamma\) if there exist a quadratic differential with a zero of order \(\geq 3g-3\) at the point \(P\). In the paper under review the following theorem is proved: Theorem. Let \(P_\pm\) be two fixed points on a compact Riemann surface \(\Gamma\) of genus \(g\geq 2\), none of them a 2-Weierstrass point of \(\Gamma\). Then, there exists a basis \(\{E_i\}\) of the space of meromorphic quadratic differentials, holomorphic away from the points \(P_\pm\), with the following behaviour near \(P_\pm\): \[ E_i= c_i^\pm z_\pm^{\pm i+g_0-2} (1+O (z_\pm))(dz_\pm)^2,\quad c_i^\pm \in \mathbb{C}, \] where \(z_\pm\) is a local coordinate around \(P_\pm\), \(g_0={3g\over 2}\) and \(i\) takes on integral values \((i=\dots, -1,0,1, \dots)\) for even \(g\) and half-integral values \((i=\dots, {-1\over 2}, {1\over 2}, \dots)\) for odd \(g\). This result is the natural generalization of the analogous one for 1-differentials or vector fields. The paper is very clearly written and uses just elementary techniques.