Abstract

Although the integrity of Rosen's Neoteleostei, Eurypterygii, and Ctenosquamata has been maintained in most subsequent classifications, the evidence cited for monophyly of each clade is conspicuously disparate. Previously proposed synapomorphies are reviewed and evaluated based on my own assessment and/or that of other authors. Those that appear least ambiguous with respect to a hypothesis of monophyly for each clade are identified. Three previously unrecognized structural innovations, each congruent with a different clade, afford new evidence of monophyly. At the level of the Neoteleostei (stomiiforms and above), concomitant with the advent of the retractor dorsalis muscle, there is a shift in the insertion of the third internal levator from the dorsal surface of the fourth pharyngobranchial cartilage to the fifth upper pharyngeal toothplate. At the level of the Eurypterygii (aulopiforms and above), the medial pelvic radial fuses with the ventral half of the medial pelvic ray in early ontogeny. At the level of the Ctenosquamata (myctophids, neoscopelids, and above), the fifth upper pharyngeal toothplate is absent and with it the associated third internal levator. Recognition of the unique ctenosquamate synapomorphy follows from the resolution, based partially on ontogenetic evidence, of long-standing confusion about the identity of the fourth and fifth upper pharyngeal toothplates in higher euteleosts.