Coincidence of harmonic and finely harmonic functions (Q618787)

In his book [Finely Harmonic Functions. Lecture Notes in Mathematics. 289. Berlin-Heidelberg-New York: Springer-Verlag., (1972; Zbl 0248.31010)], \textit{B. Fuglede} posed the following question: Given a finely open set \(U\) without interior points, does there exist a finely harmonic function \(u\) on \(U\) such that no restriction of \(u\) to a nonempty finely open subset of \(U\) admits a harmonic extension? In response, the present paper contains the following result. Theorem 1. Let \(U\) be a nonempty finely open set in \(\mathbb{R}^n\). The following statements are equivalent: (a) for any finely harmonic function \(u\) on \(U\), there is a nonempty finely open set \(V\subseteq U\) and a harmonic function \(w\) on an open set \(W\supseteq V\) such that \(w=u\) on \(V\); (b) for any finely harmonic function \(u\) on \(U\), there is a harmonic function \(w\) on an open set \(W\) such that \(W\backslash U\) is polar and \(w=u\) on \(U\cap W\); (c) \(U\) is of the second Baire category (with respect to the Euclidean topology). The author remarks that, if \(U=B\backslash Z\) for some open ball \(B\) and polar subset \(Z\) of \(B\), then \(U\) is a fine domain without interior points. If \(Z\) is a countable dense subset of \(B\), then \(U\) is of the second Baire category and so does not admit functions of the type referred to by Fuglede. On the other hand, if \(Z\) is a \(G_{\delta}\) dense subset of \(B\), then \(U\) does admit such functions.