Pixley-Roy spaces over subsets of the reals (Q1110129)

\textit{E. K. van Douwen} [Set-theor. Topol., 111-134 (1977; Zbl 0372.54006)] and \textit{M. Wage} [see the preceding review] asked whether the Pixley-Roy space over the reals was homeomorphic to the Pixley-Roy space over the irrationals. In this paper this question is answered negatively. A space X is Hurewicz (weakly Hurewicz) provided that whenever \(<{\mathcal U}_ n:\) \(n\in \omega >\) is a sequence of open covers of X, there is a sequence \(<{\mathcal G}_ n:\) \(n\in \omega >\) with \({\mathcal G}_ n\in [{\mathfrak U}_ n]^{<\omega}\) such that \(V\{\) \(\cup {\mathfrak G}_ n:\) \(n\in \omega \}\) covers X (has a dense union in X. Theorem. If PR(X) is weakly Hurewicz, then \(X^ m\) is Hurewicz. Theorem. If X is metric and each \(X^ m\) is Hurewicz, then \(PR(X)^ k\) is weakly Hurewicz for each cardinal k. Theorem. If X is noncompact and k is infinite, then \(X^ k\) is not Hurewicz. Corollary. PR (reals) is not homeomorphic to PR (irrationals). This follows since the reals are Hurewicz and the irrationals are not Hurewicz. Other properties are defined and similar method are used to distinguish other special subsets of metric spaces.