1997 04 15: Opisthosomal fusion and phylogeny of Palaeozoic Xiphosura. Lethaia, Vol. 30, pp. 19-31. Oslo. ISSN 0024-1 164. Fusion of opisthosomal tergites to form a thoracetron has previously been considered a char-acteristic of the xiphosuran superfamilies Euproopoidea Eller, 1938, and Limuloidea Zittel, 1885. Evidence is presented here that fusion also occurs in Bellinuroidea Zittel & Eastman, 1913. Results of a cladistic analysis of Palaeozoic xiphosuran genera indicate that Synziphosu-rina Packard, 1886, is a paraphyletic assemblage of stem-group Xiphosura. Superfamily Pale-olimulidae superfam. nov. is erected for families Paleolimulidae Raymond, 1944, and Moravu-ridae Pfibyl, 1967. OChelicerata, cladistics, evolution. Horseshoe crabs (Xiphosura) have been held up as the prime example of a 'living fossil' group -one that has remained conservative in morphology and at low diver-sity for much of its geological history (Fisher 1984). The distinctive features of a large carapace concealing the pro-soma1 appendages, fused opisthosomal tergites (a thora-cetron), and a styliform tail spine, can be traced back to Carboniferous times, when the group was apparently much more diverse than today. Three distinct groups of Carboniferous Xiphosura are generally recognized, the superfamilies Bellinuroidea Zittel & Eastman, 19 13, Euproopoidea Eller, 1938, and Limuloidea Zittel, 1885. The latter two groups share the possession of a thoracet-ron, whilst bellinuroids have hitherto been distinguished by their free opisthosomal tergites. During a restudy of upper Palaeozoic Xiphosura by one of us (LIA), it was dis-covered that all specimens that could be referred to Bell-inurus Pictet, 1846 (Fig. 2J), and also Bellinuroopsis Chernyshev, 1933 (Fig. 2M), have all post-opercular tergites fused into a thoracetron (Anderson 1995). This observation has important consequences for xiphosuran taxonomy and phylogeny. We present here the evidence for fusion in the opisthosoma of bellinuroids, followed by a phylogenetic analysis of late Palaeozoic Xiphosura to generic level. The character matrix relies to a large extent on new information resulting from restudy of large num-bers of late Palaeozoic xiphosurans, including all available holotypes. Further details are given by Anderson (1996), and taxonomic revisions will be published elsewhere. A major conclusion of this study is that Synziphosurina Packard, 1886, is a paraphyletic assemblage and the name must be abandoned; xiphosurans with a thoracetron (and other synapomorphies) are united herein in the order Xiphosurida Latreille, 1802.