Abstract Despite widespread agreement on the monophyly of several major taxa of bivalves, others remain uncertain and the relationships among them are debated. The present study compares new and published morphological phylogenies with new analyses based on 18S gene sequences. All but one family and all superfamilies in the Bivalvia were monophyletic in all the analyses. Several higher taxa, including most subclasses and orders, were also resolved as monophyletic. Only Myoida shows strong evidence for polyphyly, with at least two origins from Veneroida. Autobranchia was supported as monophyletic in the parsimony analyses. Within Pteriomorphia, Ostreoida is the sister taxon of Pterioida, if not derived from within it, rather than closest to Pectinoida. The numerous points of agreement with morphology based analyses suggests that both types of evidence are converging on a common phylogeny; however, differences remain to be resolved by further study.

Abstract Examination of sperm ultrastructure in six out of eight of the living species of giant clams (‘Tridacnidae’) supports reduction in status of the Tridacnidae to a subfamily of the Cardiidae (Tridacninae), as suggested by recent cladistic analyses based on shell, anatomical and molecular characters. Tridacninae spermatozoa are all of the aquasperm type, featuring, in anterior–posterior sequence: a conical acrosomal vesicle, an oblong to rod-shaped nucleus, a short midpiece region (proximal and distal centrioles surrounded by four round mitochondria); and a flagellum (9 + 2 pattern axoneme). Substantial differences occur between species with respect to the shape and length of the nucleus, and in the spatial relationship between the acrosomal complex and the nuclear apex. Although the two extant genera can be distinguished on sperm features – Tridacna , nuclear peg associated with acrosome, centriolar connective absent; Hippopus , nuclear peg absent, connective present – no defining feature of the Tridacninae can be detected. Within Tridacna , the species T. (Chametrachea) maxima and T. (C.) crocea are distinguished from other species of the genus by a much finer nuclear peg and a considerably smaller acrosome. In contrast, and against expectation, T. (C.) squamosa shows acrosomal and nuclear dimensions very close to those obtained for T. (Tridacna) gigas. Recently proposed phylogenies and classificatory changes for ‘tridacnids’ are discussed in the light of available sperm data.