A classification of the phylum Sipuncula Peter E. Gibbs Marine Biological Association of the U.K., Plymouth, Devon PL1 2PB, U.K. Edward B. Cutler Division of Science and Mathematics, Utica College of Syracuse University, Utica, New York 13502, U.S.A. Synopsis A classification of the phylum Sipuncula is adopted following the analysis of Cutler & Gibbs (1985) and comprises two classes, four orders and six families. This replaces the earlier classification of Stephen & Edmonds (1972) which was based on four families only. The diagnostic characters are reviewed. Seventeen genera are redefined, one new subgenus is described and twelve other subgenera are recognised. Introduction The classification of the phylum Sipuncula has had a confused history. Early attempts to define higher taxa by grouping genera were, to a large extent, thwarted by incomplete, imprecise or erroneous descriptions of many species. Stephen & Edmonds (1972) classified the phylum into four families in providing the first compilation of species described prior to about 1970. How-ever, this monograph is essentially literature-based and consequently many errors are repeated; nevertheless, it provides a useful base-line to the present revision. The need for greater precision in defining genera has led the authors to re-examine most of the available type specimens. The definitions of genera presented below incorporate both novel observations and corrections to earlier descriptions. Where possible, nine basic characters have been checked for each species before assigning it to a genus. These characters are summarised for each genus in Table 1 . A phylogenetic interpretation of the classification used here will be found in Cutler & Gibbs (1985). Diagnostic features of higher taxa In reviewing the diagnostic characters of the phylum, particular attention has been paid to the structure of the oral disk since the arrangement of the tentacles provides a useful basis for dividing the phylum into two classes -Sipunculidea and Phascolosomatidea. Certain descriptions of tentacle arrangements are misleading or in error (see for example Stephen & Edmonds (1972) p. 16 and Table 3). No doubt these errors result from the fact that some species are not amenable to fixation in the extended state; species with long introverts are notoriously difficult to preserve with their introvert fully extended. Although dissection of the introvert is possible the details of the tentacular arrangement on a withdrawn disk are often difficult to interpret and have yet to be satisfactorily determined in some small-sized species (e.g. Apionsoma trichocephala Sluiter). The following summarises the distinctions of the two classes. The sipunculan tentacular crown exhibits many diverse forms but, basically, two tentacular patterns can be recognised. In one, that of the proposed class Sipunculidea, the tentacles are arranged peripherally on the oral disk so as to encircle the centrally-placed mouth; dorsally this circle is inflected to form an arc enclosing the nuchal organ, a feature well developed in Thysano-cardia spp for example (see Gibbs, Cutler & Cutler, 1983, Fig. 2). In the other, that of the proposed class Phascolosomatidea, the tentacles are restricted to a dorsal arc enclosing the nuchal organ and Bull. Br. Mus. not. Hist. (Zool.) 52(1): 43-58 Issued 29 January 1987
BioStor
