REASSESSMENT OF THE AFFINITIES OF THE EXTINCT GENUS CYLINDRACANTHUS (OSTEICHTHYES)

The perplexing fossil genus Cylindracanthus has long been a taxonomic problem. Virtually a form genus of ribbed spine-like marine fish fossils (Cretaceous-Eocene), it has only recently been definitively shown to be rostral, based on bilateral symmetry and the presence of teeth. Best-known from circumAtlantic sites, the first South Dakota specimen revealed the teeth for the first time and gave impetus for this review. Although the study is on-going, an emergent concept of relationships of the genus is now hypothesized, as well as potential biostratigraphic utility. Traditionally Cylindracanthus was discussed in relation to the billfishes (marlins and swordfishes), particularly by those who considered the fossils to be rostral fragments. Aside from the anatomical considerations, such a relationship is awkward for biostratigraphic and evolutionary reasons. The extant billfish are highly derived Perciformes, an order that is virtually confined to the Cenozoic. Cylindracanthus and associated genera originated during the Cretaceous, which would place them among the earliest perciform genera, if correctly referred. In fact, they are in some respects even more derived than the living billfish, presumably with a skeleton entirely of cartilage except for the rostrum. (Extant billfish are bony, not cartilaginous.) It is thus unlikely that Cylindracanthus would have appeared so much earlier than supposedly related forms which are less derived. Ancestral Acipenseriformes (sturgeons and related forms) appeared in the geologic record at about the same time in the Cretaceous as did Cylindracanthus , and a significant number of acipenseriforms survived the Cretaceous/Tertiary extinction event, as did Cylindracanthus. All of the living members of the order have tendencies toward projecting rostral development, and all are highly cartilaginous. As many taxonomic groups of fishes have produced experiments in bill-like projections of the rostrum, it is reasonable to speculate that Cylindracanthus was such an experiment among the acipenseriforms, with which all known features are consistent. Of known actinopterygians, only