Composite holomorphic functions and normal families (Q642702)

Summary: We study the normality of families of holomorphic functions and prove the following result. Let \(\alpha(z)\), \(a_i(z)\), \(i = 1, 2, \dots, p\), be holomorphic functions, let \(\mathcal F\) be a family of holomorphic functions in a domain \(D\), and define \(P(z, w) := (w - a_1(z))(w - a_2(z)) \cdots (w - a_p(z))\), \(p \geq 2\). Assume that \(P_w \circ f(z)\) and \(P_w \circ g(z)\) share \(\alpha(z)\) IM for each pair \(f(z),g(z) \in \mathcal F\) and that one of the following conditions holds: (1) \(P(z_0, z) - \alpha(z_0)\) has at least two distinct zeros for any \(z_0 \in D\); (2) there exists \(z_0 \in D\) such that \(P(z_0, z) - \alpha(z_0)\) has only one distinct zero and \(\alpha(z)\) is nonconstant. If \(\beta_0\) denotes the zero of \(P(z_0, z) - \alpha(z_0)\) and the multiplicities \(l\) and \(k\) of the zeros of \(f(z) - \beta_0\) and \(\alpha(z) - \alpha(z_0)\) at \(z_0\), respectively, satisfy \(k \neq lp\), for all \(f(z) \in \mathcal F\), then \(\mathcal F\) is normal in \(D\). In particular, the result is a generalization of the famous Montel's criterion. At the same time we fill a gap in the proof of Theorem 1.1 in our original paper [Fixed Point Theory Appl. 2010, Article ID 417480 (2010; Zbl 1203.30034)].