Endomorphisms of the lattice of epigroup varieties (Q512199)

A semigroup \(S\) is called an epigroup if some power of each element belongs to some subgroup of \(S\), such that for each \(a \in S\) a pseudoinverse \(\bar a\) can be defined and in the semigroup a unary operation \(a \to \bar a\). Thus epigroups can be considered as algebras with two operations. It is not known whether there exist non-trivial (different from antiisomorphism) automorphisms of the lattice \(\mathrm{SEM}\) of all semigroup varieties or of the lattice \(\mathrm{EPI}\) of all epigroup varieties, but there exist infinitely many non-trivial injective endomorphisms of the lattice \(\mathrm{SEM}\). Here, an epigroup analogue is proved. The authors present a countably infinite family of injective endomorphisms of the lattice of all epigroup varieties \(\mathrm{EPI}\) and characterize epigroup varieties where all nilsemigroups are nilpotent.