On a conjecture of Agronsky and Ceder concerning orbit-enclosing \(\omega \)-limit sets (Q1978909)

Let \(f:A\rightarrow A\) be a continuous map of a topological space \(A\). Denote \(\gamma (x,f)=\{f^n(x):x\in A, \^^Mn\in \mathbb N\}\) and let \(\omega(x,f)\) be the set of all limit points of the sequences \(\{f^n(x)\}^{\infty }_{n=0}\). We say that a subset \(B\subset A\) is an orbit-enclosing \(\omega\)-limit set if there exist \(f\) and \(x\in A\) such that \(B=\gamma (x,f)\subset \omega(x,f)\). The author disproves the conjecture by \textit{S. Agronsky} and \textit{J. G. Ceder} [Real Anal. Exch. 17, 371-378 (1991/92; Zbl 0799.26013)] which claims that all compact connected orbit-enclosing \(\omega\)-limit sets \(B\subset A=E^k\) \((k\geq 2)\) necessarily are arcwise connected.