WHERE DOES THALATTOSUCHIA BELONG IN CROCODYLOMORPHA? AN UPDATE ON THE CROCSUPERMATRIX PROJECT

YOUNG, M.; HASTINGS, A.K.; ANDRADE, M.B.; HERRERA, Y.

Congreso; 9th International Meeting on the Secondary Adaptation of Tetrapods to Life in Water; 2021

Perhaps the most peculiar clade of crocodylomorphs is Thalattosuchia, a predominately marine group that lived during the Jurassic and Cretaceous. Within Thalattosuchia there was a major evolutionary transition, encompassing the shift from semi-aquatic to pelagic forms (the Metriorhynchidae). However, where in Crocodylomorpha they belong is contentious. There are four major hypotheses on the position of Thalattosuchia: 1) sister taxon to Crocodyliformes, 2) sister taxon to Metasuchia, 3) basal neosuchian clade, and 4) a member of a ?longirostrine aquatic? neosuchian clade with pholidosaurids and dyrosaurids. At first glance, this wide range of possible positions for Thalattosuchia, a major crocodylomorph clade, is surprising. However, thalattosuchians evolved a suite of over 40 osteological, soft-tissue and neuroanatomical characters that underpinned their aquatic and/or marine adaptations. Over the past five years the authors here have been merging their phylogenetic datasets to investigate areas of uncertainty in the crocodylomorph tree, in particular Thalattosuchia. This led to the CrocSuperMatrix project. Here we report progress on this project. Thus far the datasets of Alexander Hastings and Mark Young are merged, and the merging with the Andrade et al. (2011) dataset is over half complete. Moreover, CT-based endocranial data from the University of Edinburgh CrocTransition project has also been incorporated. At present our dataset recovers Thalattosuchia in hypothesis two ? as the sister taxon to Metasuchia. This position is recovered regardless of weighting regime used. Four key problems are revealed through our merging of these matrices: (1) Thalattosuchia is a group where a substantial number of characters reflect clade specialisations, few characters provide links to other branches, and it becomes difficult to differentiate secondary homology from convergence; (2) a Norian?Pliensbachian gap in the crocodylomorph fossil record means we lack information on the origin of most of the crocodyliform lineages (Gobiosuchidae, Hsisosuchidae, Shartegosuchoidea, Notosuchia, Neosuchia and Thalattosuchia); (3) the multiple lineages of ?sphenosuchians? present in the Jurassic have no known Triassic forebearers (including Hallopodidae, the sister taxon to Crocodylifomes); (4) these issues all imply a poor distribution of characters on the consensus topologies. These three problems will affect any matrix and sample of characters. As a result, any crocodylomorph matrix or supermatrix will suffer from too much noise (problem 1), or missing data (problems 1?4). These problems seem to explain such disparate hypotheses (and poor support) for the position of Thalattosuchia. A substantial input from the fossil record is needed to overcome this gap.