A remark on a theorem by F. Forstnerič (Q1813507)

Let \(\gamma\) be a simple closed curve of class \(C^ 2\) in the unit sphere in \(\mathbb{C}^ n\). If \(\gamma\) is not polynomially convex its hull must be an analytic variety \textit{A. F. Forstnerič} [Publ. Math., Barc. 32, No. 1, 145-150 (1988; Zbl 0737.32008)] has proved that this variety is nonsingular near the sphere, that the pair \((A,\gamma)\) is a \(C^ 2\)-manifold with boundary near the sphere and that moreover this manifold intersects the sphere transversally. An interesting consequence of this is that the curve must be complex transversal at every point if it is not polynomially convex. In the paper under review the author gives a direct and elementary proof of this latter fact. Assuming the existence of at least one complex tangent, he shows that polynomials are dense in \(C(\gamma)\), using a geometric construction and some facts from the theory of uniform algebras. He also shows that this result is not necessarily true if \(\gamma\) is only of class \(C^ 1\).