Cycle-closed permutation groups (Q1815592)

If \(G\) is a finite permutation group, then \(C(G)\) denotes the group generated by all the cycles of \(G\), and \(G\) is cycle-closed if \(G=C(G)\). By an elementary argument, the author shows that a finite permutation group is cycle-closed if and only if it is a direct product of symmetric groups and cyclic groups of prime order. Furthermore, if \(C_0(G)=G\) and \(C_{n+1}(G)=C(C_n(G))\), then for any finite \(G\), \(C_n(G)\) is cycle-closed, but there is finite \(G\) such that \(C_2(G)\) is not cycle-closed. In the infinite case, there are several notions of cycle closure. The author denotes by \(C^-(G)\) the group generated by the cycles of \(G\), by \(C(G)\) the group \(\langle G,C^-(G)\rangle\), by \(R(G)\) the group generated by all restrictions of elements of \(G\), and by \(C^+(G)\) the group generated by all permutations \(g\) such that every cycle of \(g\) is a cycle of an element of \(G\). Then \(G\leq C(G)\leq R(G)\leq C^+(G)\), for any \(G\). Again, there are inductive iterates of these definitions. The author proves many partial results, such as that if \(G\) is an infinite transitive permutation group then \(C^-_3(G)\) contains all finitary permutations and \(C^+_3(G)\) is the full symmetric group; also, that if \(G\) is the infinite cyclic group acting regularly, then \(C(G)=C^-(G)\) is \(k\)-transitive for all finite \(k\). He conjectures that for any permutation group \(G\), \(C_3(G)\) is cycle closed, proves this in special cases, and states some open problems.