COMBINED ANALYSIS OF EXTANT RHYNCHONELLIDA (BRACHIOPODA) USING MORPHOLOGICAL AND MOLECULAR DATA

Our understanding of evolutionary relationships within extinct groups depends upon careful study of morphological features, but this implicitly assumes that the features studied predominantly reflect shared evolutionary history. Groups with extant lineages are an opportunity to compare phylogenetic signal between independent morphological and molecular datasets. Rhynchonellide brachiopods provide such a test case, with more than 500 extinct genera represented today by only 19 extant members. Presently, phylogenetic relationships within the Rhynchonellida are contentious, with previous studies finding that phylogenetic hypotheses generated from molecular sequence data and those constructed using morphological characters have considerable conflict. We performed a series of combined phylogenetic analyses to identify conflicting phylogenetic signals present in molecular and morphological characters among the 19 extant rhynchonellide genera. Molecular data were taken from a published dataset of 3435 nucleotide sites of 18S rDNA sequences from 12 rhynchonellide species, two terebratulide species, and four outgroup taxa (Cohen & Bitner 2013). Sixty-six morphological characters of the shell interior and exterior and soft tissues were coded for all taxa, resulting in a matrix of 26 taxa, 18 of which have both molecular and morphological data. We completed a series of parsimony-based and Bayesian analyses of molecular and morphological data each, separately and combined. Phylogenetic hypotheses generated using only morphological characters differed somewhat from previously published morphology-based hypotheses, but support the monophyly of crown clade Rhynchonellida. Phylogenetic hypotheses generated using both molecular and morphological data are in congruence with published molecular phylogenetic hypotheses. Within extant rhynchonellides, incongruence between the morphological and molecular preferred topologies is due to noise within the morphological data, and a simulation test suggests this is not driven by differences in partition size. Overall, if preserved morphology has little phylogenetic signal, systematic paleontology of that group may reflect functional and developmental constraints rather than evolutionary relationships.