Endocast and bony labyrinth of a stem gnathostome shed light on the earliest diversification of jawed vertebrates

Our understanding of the earliest evolution of jawed vertebrates depends on a credible phylogenetic assessment of the jawed stem gnathostomes collectively known as placoderms. However, their relationships, and even whether placoderms represent a single radiation or a paraphyletic array, remain contentious. Here we describe the endocranial cavity and inner ear of Brindabellaspis stensioi, commonly recovered as a taxon of uncertain affinity branching near the base of placoderms. While some features of its braincase and endocast resemble those of jawless vertebrates, its inner ear displays a repertoire of crown gnathostome characters. Both parsimony and Bayesian analyses suggest that established hypotheses of placoderm relationships are unstable, with newly-revealed anatomy pointing to a potentially radical revision of early gnathostome evolution. Our results call into question the appropriateness of fusiform placoderms as models of primitive gnathostome anatomy and raise questions of homology relating to key cranial features. One Sentence SummaryThe skull of a 400-million-year old fossil fish suggests that hypotheses of early jawed vertebrate relationships might have to be turned on their head.