Tight analytic surfaces (Q803600)

A smooth immersion of a compact surface f: \(M\to {\mathbb{R}}^ n\) is said to be tight if it minimizes the total absolute curvature among all immersions of M. The main result of the paper is the following nice theorem: Let f: \(M\to {\mathbb{R}}^ 4\) be an analytic tight immersion of an orientable compact surface that is not contained in a proper affine subspace. Then M is a torus. Moreover, M is the intersection of two developable ruled hypersurfaces, possibly with singularities, whose rulings are two-dimensional. This is in strong contrast to \(C^{\infty}\) tight immersions of orientable surfaces into \({\mathbb{R}}^ 4\) which exist for any genus \(\geq 1\) according to \textit{N. H. Kuiper} [Invent. Math. 10, 209-238 (1970; Zbl 0195.511)]. Finally some open questions about tight analytic surfaces complete the paper. For instance, which nonorientable surfaces admit tight analytic immersions into \({\mathbb{R}}^ 4\) that do not lie in a hyperplane? The only known example is the projective plane, or more precisely the stereographic image in \({\mathbb{R}}^ 4\) of the Veronese surface in \({\mathbb{R}}^ 5\).