On total reality of meromorphic functions (Q2466308)

The authors consider the following problem, which is higher genus extension of the one-dimensional Shapiro-Shapiro conjecture: given a real curve \((C,\sigma)\) (a compact Riemann surface with an antiholomorphic involution \(\sigma\)) and a meromorphic function \(f:(C,\sigma)\to\mathbb{C}\mathbb{P}^1\) such that all its critical points and values are distinct, and all its critical points belong to \(C_\sigma\) (the real part of \(C\)), is it true that \(f\) becomes a real meromorphic function after an appropriate choice of a real structure on \(\mathbb{C}\mathbb{P}^1\)? The main result states that this is so for the functions \(f\) of degree \(4\) and curve \((C,\sigma)\) of any genus (the case of functions of degree \(\leq 3\) was settles by the second, fourth and fifth authors [Preprint, \url{arXiv:math/0605075}]). The proof follows the lines of the above cited paper and finally reduced the question the non-existence of a real curve of bi-degree \((4,4)\) on a quadric ellipsoid having \(8\) real cusps. The latter is established via consideration of real \(K3\)-surfaces, arising as double covers of a quadric surface ramified along real curves of bi-degree \((4,4)\), and their intersection lattices.