Characterizing endpoints for a family of set-valued inverse limits (Q6539331)

In the present paper the authors investigate the endpoints of inverse limits of set-valued functions using \textit{A. Lelek}'s definition of an endpoint [Fundam. Math. 49, 301--319 (1961; Zbl 0099.17701)]: The point \(p\) is an endpoint of the continuum \(X\) if \(p\) is an endpoint of every arc in \(X\) that contains \(p\). Using Lelek's definition, it is shown in an earlier paper [\textit{L. Alvin} and \textit{J. P. Kelly}, Topology Proc. 54, 233--257 (2019; Zbl 07848739)] that there exists a set-valued function \(F:X\rightarrow2^X\) whose inverse is equal to the union of mappings and \((p_0, p_1, \dots ,p_n)\) is an endpoint of \(G_n=\{ \mathbf{x} =(x_i)_{i=0}^n\in\prod^n_{i=0} X: x_{i-1}\in F(x_i)\}\) for all \(n\in \mathbb{N}\), but \( \mathbf{p} = (p_0, p_1, \dots)\) is not an endpoint of \(\displaystyle{\lim_{\longleftarrow} F}\). The authors note that \((p_0, p_1, \dots ,p_n)\), being an endpoint of \(G_n\) for all \(n\in \mathbb{N}\), is a necessary condition for \(\mathbf{p}\) to be an endpoint of \(\displaystyle{\lim_{\longleftarrow} F}\) under Lelek's definition, but this leaves the question of what additional conditions must be placed on a point \( \mathbf{p}\in \displaystyle{\lim_{\longleftarrow} F}\) to guarantee it is an endpoint of the inverse limit. \N\NIn this paper, the authors provide two characterizations for a point to be an endpoint of the inverse limit for a family of set-valued functions. The first characterization utilizes limit points of intersections of special arcs in the inverse limit, whereas the second characterization focuses on sequences of branch points. The paper concludes with several examples demonstrating how endpoints can be identified using finite approximations of the inverse limit.