2008 Joint Meeting of The Geological Society of America, Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Gulf Coast Association of Geological Societies with the Gulf Coast Section of SEPM

Paper No. 5
Presentation Time: 9:00 AM

Comparing Extinction Selectivity among Terebratulide Brachiopod Clades: Using Phylogenetic Methods to Test Macroevolutionary Hypotheses


CARLSON, Sandra J., Department of Geology, University of California, Davis, One Shields Ave, Davis, CA 95616 and FITZGERALD, Paul C., Department of Biology, Northern Virginia Community College, 1000 Harry Flood Byrd Highway, Sterling, VA 20164-8699, carlson@geology.ucdavis.edu

Phylogenetic selectivity at mass extinction events can be difficult to recognize, characterize, and quantify. Yet, studying extinction in a phylogenetic framework is necessary in order to evaluate extinction as an agent of macroevolutionary change. We have completed phylogenetic analyses of all 71 named Devonian terebratulides, not merely selected representative genera. These analyses revealed two large sister clades, one dominated by genera in the Stringocephalidae, and the other dominated by genera in the Centronellidae. Neither family is monophyletic, and genera classified in Meganteridae occupy basal portions of both clades. The end-Emsian and end-Givetian extinctions, the two largest Devonian terebratulide extinction events, both appear phylogenetically biased regardless of familial affiliation. 41% of 41 genera in one clade became extinct at the end-Emsian, while only 20% became extinct at the end-Givetian. In the sister clade, only 7% of 30 genera became extinct at the end-Emsian, whereas 53% became extinct at the end-Givetian. Contrary to these results, preliminary phylogenetic analyses involving Permian and Triassic terebratulide genera suggest that the end-Permian extinction may not be phylogenetically biased. At least some genera in each subclade become extinct at the end-Permian; extinctions are not clustered in one or few subclades only. Earlier comparisons of Colless Indices and clade size distributions before and after the end-Givetian event revealed an increase in tree imbalance, suggesting that this extinction event was not selective with regard to phylogenetic relationships. Our more complete results now suggest either that a less-than-complete sampling of taxa may produce different results, or that Colless Indices are not necessarily accurate predictors of the selectivity of particular extinction events. These phylogenetic analyses also suggest that Devonian and Permian terebratulides were affected differently by mass extinction events, producing different macroevolutionary consequences — an observation undetectable using taxonomic comparisons alone.