The phylum Myxozoa has been considered to comprise two classes, Myxosporea Bütschli, 1881 (primarily of fishes) and Actinosporea Noble in Levine et al., 1980 (primarily of aquatic oligochaetes). About 10 years ago it was demonstrated that the life cycle of Myxobolus cerebralis Hofer, 1903 (Myxobolidae: Platysporina) of salmonid fishes requires transformation of the myxosporean into an actinosporean stage in the oligochaete worm Tubifex tubifex (Tubificidae), and that the stage infective to fish is the actinosporean spore. This type of two-host life cycle has now been demonstrated or strongly implicated for 14 myxosporean species, belonging to 6 genera in 4 families. In light of these findings, the taxonomy of the Myxozoa is revised. We propose the following: suppression of the newer class Actinosporea and the order Actinomyxidia Štolc, 1899; and suppression of all families in the Actinosporea except Tetractinomyxidae. This family and its one genus, Tetractinomyxon Ikeda, 1912, are transferred to the order Multivalvulida Shulman, 1959 (Myxosporea). We also propose that actinosporean generic names be treated as collective-group names, thus they do not compete in priority with myxosporean generic names. Triactinomyxon dubium Granata, 1924 and Triactinomyxon gyrosalmo Wolf and Markiw, 1984 are suppressed as junior synonyms of Myxobolus cerebralis. The myxosporean stage of no other previously named actinosporean has been identified. Other actinosporean species are therefore retained as species inquirendae until their myxosporean stages are identified. A revised description of the phylum Myxozoa is provided that includes our proposed taxonomic and nomenclatural changes.
Prevalence and susceptibility of infection to Myxobolus cerebralis, and genetic differences among populations of Tubifex tubifex