The gnathostome (jawed vertebrate) crown group comprises two extant clades with contrasting character complements. Notably, Chondrichthyes (cartilaginous fish) lack the large dermal bones that characterize Osteichthyes (bony fish and tetrapods). The polarities of these differences, and the morphology of the last common ancestor of crown gnathostomes, are the subject of continuing debate. Here we describe a three-dimensionally preserved 419-million-year-old placoderm fish from the Silurian of China that represents the first stem gnathostome with dermal marginal jaw bones (premaxilla, maxilla and dentary), features previously restricted to Osteichthyes. A phylogenetic analysis places the new form near the top of the gnathostome stem group but does not fully resolve its relationships to other placoderms. The analysis also assigns all acanthodians to the chondrichthyan stem group. These results suggest that the last common ancestor of Chondrichthyes and Osteichthyes had a macromeric dermal skeleton, and provide a new framework for studying crown gnathostome divergence.

The evolutionary history of osteichthyans (bony fishes plus tetrapods) extends back to the Ludlow epoch of the Silurian period. However, these Silurian forms have been documented exclusively by fragmentary fossils. Here we report the discovery of an exceptionally preserved primitive fish from the Ludlow of Yunnan, China, that represents the oldest near-complete gnathostome (jawed vertebrate). The postcranial skeleton of this fish includes a primitive pectoral girdle and median fin spine as in non-osteichthyan gnathostomes, but a derived macromeric squamation as in crown osteichthyans, and substantiates the unexpected mix of postcranial features in basal sarcopterygians, previously restored from the disarticulated remains of Psarolepis. As the oldest articulated sarcopterygian, the new taxon offers insights into the origin and early divergence of osteichthyans, and indicates that the minimum date for the actinopterygian-sarcopterygian split was no later than 419 million years ago.

Enamel, the hardest vertebrate tissue, covers the teeth of almost all sarcopterygians (lobe-finned bony fishes and tetrapods) as well as the scales and dermal bones of many fossil lobe-fins. Enamel deposition requires an organic matrix containing the unique enamel matrix proteins (EMPs) amelogenin (AMEL), enamelin (ENAM) and ameloblastin (AMBN). Chondrichthyans (cartilaginous fishes) lack both enamel and EMP genes. Many fossil and a few living non-teleost actinopterygians (ray-finned bony fishes) such as the gar, Lepisosteus, have scales and dermal bones covered with a proposed enamel homologue called ganoine. However, no gene or transcript data for EMPs have been described from actinopterygians. Here we show that Psarolepis romeri, a bony fish from the the Early Devonian period, combines enamel-covered dermal odontodes on scales and skull bones with teeth of naked dentine, and that Lepisosteus oculatus (the spotted gar) has enam and ambn genes that are expressed in the skin, probably associated with ganoine formation. The genetic evidence strengthens the hypothesis that ganoine is homologous with enamel. The fossil evidence, further supported by the Silurian bony fish Andreolepis, which has enamel-covered scales but teeth and odontodes on its dermal bones made of naked dentine, indicates that this tissue originated on the dermal skeleton, probably on the scales. It subsequently underwent heterotopic expansion across two highly conserved patterning boundaries (scales/head-shoulder and dermal/oral) within the odontode skeleton.

We used propagation phase contrast X-ray synchrotron microtomography to study the three-dimensional (3D) histology of scales of two osteostracans, Tremataspis and Oeselaspis, members of a jawless vertebrate group often cited as the sister group of jawed vertebrates. 3D-models of the canal systems and other internal structures are assembled based on the virtual thin section datasets and compared with previous models based on real thin sections. The primary homology framework of the canal systems in the two taxa is revised and new histological details are revealed based on the results of this work. There is no separation of vascular canals and lower mesh canals in the Tremataspis scale, contrary to previous results. The secondary upper mesh canals have a limited distribution to the anterior region of the Tremataspis scale. The upper and lower mesh canal systems of Tremataspis have different geometries, inferred to reflect different developmental origins: we interpret the upper system as a probable epithelial invagination, the lower system as entirely vascular. Oeselaspis has no equivalent of the upper mesh canal system. The upper mesh canal system of Tremataspis may have been sensory in function. In Oeselaspis, numerous polyp-shaped structures opening from the canal system onto the surface of the scale resemble the innervation tracts for neuromast organs. The growth of the Oeselaspis scale proceeds by addition of small odontodes containing unmineralized lacunae, which may further mineralize and become more compact. Our results highlight that 3D-histological investigation on scales and other dermal skeletons of osteostracans is necessary to fully appreciate the diversity of skeletal histologies in the group. Traditional 3D-models based on thin sections alone are not reliable and should no longer be used as the basis for homology assessments or functional hypotheses.