Systematic Parasitology 44: 49–57, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands. 49 Actinosporeans (Myxozoa) from marine oligochaetes of the Great Barrier Reef Sascha L. Hallett 1 , Christer Ers´ eus 2 & Robert J. G. Lester 1 1 Department of Parasitology, The University of Queensland, Brisbane, Queensland 4072, Australia 2 Department of Invertebrate Zoology, Swedish Museum of Natural History, Box 50007, SE-104 05 Stockholm, Sweden Accepted for publication 21st July, 1998 Abstract Four species of actinosporeans are described from marine oligochaetes (all Tubificidae) from the Great Barrier Reef, Queensland, Australia. They developed in the coelom of the oligochaete and produced spores in groups of eight in the pansporocysts. The new genus Endocapsa is proposed within the family Sphaeractinomyxidae Janiszewska, 1957 on the basis that mature spores have small valve cell processes and non-protruding polar cap- sules. The type-species, Endocapsa rosulata n. sp., has three valve cell processes, which resemble a rosette, and submerged polar capsules. It infected Heterodrilus cf. keenani from Heron Island and morphologically similar parasites occurred in Thalassodrilides cf. gurwitschi and Heronidrilus sp. from Lizard Island. E. stepheni n. sp. has asymmetrical valve cell processes and submerged polar capsules. It was found in H. cf. keenani and H. queenslandicus from Heron Island. Sphaeractinomyxon leptocapsula n. sp. has thin widely spaced polar capsules and is described from Heronidrilus sp. from Lizard Island. S. ersei Hallett, O’Donoghue & Lester, 1998 infected Tubificidae gen. sp. from Heron Island and S. cf. ersei occurred in Bathydrilus sp., Thalassodrilides cf. gurwitschi and Limnodriloides lateroporus from Lizard Island. Introduction The taxonomic affinities of the Myxozoa Grassé, 1970 have been revised frequently (Kudo, 1933; Tripathi, 1948; Schulman, 1966; Levine et al., 1980; Lom & Noble, 1984; Corliss, 1985; Kent et al., 1994; Siddall et al., 1995). Current focus is on the taxon as a whole as well as lower divisions because of recent mole- cular and life-cycle studies. Smothers et al. (1994) and Schlegel et al. (1996) proposed that the Myxo- zoa are related to the Bilateria, whereas Siddall et al. (1995) believed them to be a clade of parasitic cnidar- ians. Cavalier-Smith et al. (1996) has proposed that the Myxozoa are a subkingdom of the Animalia. Kent et al. (1994) suggested that the Class Actinosporea Noble, 1980 and five of its six families be suppressed, leaving the Phylum Myxozoa with a single class, the Myxosporea Bütschli, 1881. Life-cycle studies have shown that some fresh- water actinosporeans are alternate stages of some freshwater myxosporeans. Kent et al. (1994) conclude that alternate myxosporean development probably oc- curs in all actinosporean families and genera (except possibly Tetractinomyxon Ikeda, 1912) and as such the nominal actinosporean generic names cannot be considered distinct from myxosporean genera. They recommend that actinosporean genera be declared in- valid (except Tetractinomyxon) and that their nominal generic names be treated as collective group names. We think this is premature for two reasons. The alter- nation of generations has been demonstrated for about 1% of Myxosporea, all freshwater. Whether all species undergo this alternation is yet to be determined; there is evidence for direct fish–fish transmission of at least one marine myxosporean (Diamant, 1997). Secondly, it will be many years before the life-cycles of most myxosporeans are elucidated and in the meantime we need a precise way to categorise types of acti- nosporeans. Only 13 genera of actinosporeans exist in the literature and are available for group names. In this study we describe actinosporeans that do not