No abstract

Ostracoderms are a paraphyletic group of extinct jawless fishes comprising the gnathostome stem and are fundamental to our understanding of early vertebrate evolution. However, only a handful of these clades have robust phylogenies in place, hindering our interpretation of early vertebrate histories. A new phylogeny is proposed for the Pteraspidiformes À the largest and most-studied clade of heterostracan ostracoderms. Difficulties such as large amounts of missing data and the limited morphological variability within the group have led us to explore different coding strategies such as the inclusion of quantitative data and implied weighting. We present a new comprehensive data set including all described genera of Pteraspidiformes (47 taxa) analysed using discrete characters only, a combination of discrete and continuous characters, and gap-coding strategies (transforming the continuous into discrete characters), along with a Bayesian analysis. Two representatives of the Psammosteidae (Drepanaspis and Psammosteus) are also incorporated within the analysis to elucidate their inclusiveness within the Pteraspidiformes. Well-resolved trees are only achieved under re-weighted (implied weighting) analyses. Here, we show that many 'classic' Pteraspidiformes clades hold true under our different coding methods, with the implied weighting of discrete characters and inclusion of continuous characters giving very similar topologies. In all instances, the Psammosteidae are found to belong within the Pteraspidiformes, nested with the Spitsbergen genera Doryaspis, Xylaspis and Woodfjordaspis. Gap coding, however, results in a different tree topology to other analyses, perhaps due to the high sensitivity to missing data. Our results indicate that careful consideration and justifications should be applied to quantitative characters when reconstructing relationships of homoplastic ostracoderms. Superfamily Doryaspidae superfam. nov. is introduced.http://zoobank.org/urn:lsid:zoobank

The fossil record of early vertebrates has been influential in elucidating the evolutionary assembly of the gnathostome bodyplan. Understanding of the timing and tempo of vertebrate innovations remains, however, mired in a literal reading of the fossil record. Early jawless vertebrates (ostracoderms) exhibit restriction to shallow-water environments. The distribution of their stratigraphic occurrences therefore reflects not only flux in diversity, but also secular variation in facies representation of the rock record. Using stratigraphic, phylogenetic and palaeoenvironmental data, we assessed the veracity of the fossil records of the jawless relatives of jawed vertebrates (Osteostraci, Galeaspida, Thelodonti, Heterostraci). Non-random models of fossil recovery potential using Palaeozoic sea-level changes were used to calculate confidence intervals of clade origins. These intervals extend the timescale for possible origins into the Upper Ordovician; these estimates ameliorate the long ghost lineages inferred for Osteostraci, Galeaspida and Heterostraci, given their known stratigraphic occurrences and stem–gnathostome phylogeny. Diversity changes through the Silurian and Devonian were found to lie within the expected limits predicted from estimates of fossil record quality indicating that it is geological, rather than biological factors, that are responsible for shifts in diversity. Environmental restriction also appears to belie ostracoderm extinction and demise rather than competition with jawed vertebrates.

Reconstructing evolutionary histories requires accurate phylogenetic trees. Recent simulation studies suggest that probabilistic phylogenetic analyses of morphological data are more accurate than traditional parsimony techniques. Here, we use empirical data to compare Bayesian and parsimony phylogenies in terms of their congruence with the distribution of age ranges of the component taxa. Analysis of 167 independent morphological data matrices of fossil tetrapods finds that Bayesian trees exhibit significantly lower stratigraphic congruence than the equivalent parsimony trees. As such, taking stratigraphic data as an independent benchmark indicates that parsimony analyses are more accurate for phylogenetic reconstruction of morphological data. The discrepancy between simulated and empirical studies may result from historic data peaking practices or some complexities of empirical data as yet unaccounted for.