On the hyperspaces \({\mathcal C}_n(X)\) of a continuum \(X\) (Q5925883): Difference between revisions

Let \(X\) be a (metrizable) continuum, and let \({\mathcal C}_n(X)\) denote the subspace of the hyperspace \(2^X\) of \(X\) consisting of those compacta having at most \(n\) components. Wojdysławski proved in 1939 that a continuum \(X\) is locally connected if and only if \({\mathcal C}_n(X)\) is an absolute retract for every \(n\). Since Wojdysławski's work, not much has been done about these hyperspaces. In this paper, the author presents a study of them. He proves for example that for a continuum \(X\), \(2^X\) is contractible if and only if \({\mathcal C}_n(X)\) is contractible for every \(n\). In addition, \({\mathcal C}_n(X)\) has trivial shape, is unicoherent and colocally connected, etc. A nondegenerate continuum \(X\) is indecomposable if and only if for each \(n\), \({\mathcal C}_n(X)\smallsetminus \{ X\}\) is not arcwise connected. The author also presents results on the behaviour of Whitney maps on the hyperspaces \({\mathcal C}_n(X)\). Let \({\mathcal C}_\infty (X)\) denote the subspace of \(2^X\) consisting of those compacta having infinitely many components. The author proves that \({\mathcal C}_\infty(X)\) is not a \(G_\delta\)-subset of \(2^X\). The paper ends with some interesting unsolved questions, among them: Is \({\mathcal C}_\infty (X)\) always unicoherent?