PRELIMINARY EXAMINATION OF MOLLUSCAN BIODIVERSITY AND LIFE MODE CHANGES ACROSS THE PLIOCENE-PLEISTOCENE EXTINCTION, TAR HEEL AND WALKER'S BLUFF, SOUTHEASTERN NORTH CAROLINA

The Pliocene-Pleistocene transition exhibits patterns of extinction in nearshore marine communities of the western Atlantic, thought to be linked with the closing of the Isthmus of Panama. Studies conducted in Florida and the Caribbean suggest there were a series of extinctions during the last 4 m.y., most numerous from 2-1Ma. However, events that occurred further north in the Carolinas are not as well documented. As part of the 2010 Research Experiences for Undergraduates in Biodiversity Conservation at the University of North Carolina Wilmington, we studied the change in community structure, i.e. changes in dominance, genus richness, and life modes represented (i.e. feeding mode, substrate relation, mobility and attachment, from the NMITA molluscan life modes database) in order to gauge the severity of the series of events and to ascertain whether it resulted from decreased productivity. We took molluscan macrofossils from seven wet-sieved (1/4” mesh) bulk samples from two localities, one near Tar Heel, Bladen Co., NC, representing pre-extinction Duplin Formation material (~3.34 Ma; McGregor et al. 2011), and the other from Walker's Bluff, Bladen Co., NC, representing post-extinction upper Waccamaw Formation (1.6 Ma; Graybill et al. 2009). Specimens were identified to genus and counted; for bivalve individuals we used half the total valve count. Of the bivalve taxa that went extinct (i.e. those unique to the Tar Heel site), the most common life mode was actively mobile infaunal siphonate suspension feeding. The fact that so many extinct genera were suspension feeders supports the hypothesis of productivity decline. However, suspension feeders dominated both samples, and the most common life modes of the surviving taxa (i.e. those shared between the two localities) were the same as for extinct taxa. We generated rarefaction curves for bivalves in each sample and evaluated them at 450 specimens, which yielded 39.4 genera for Tar Heel and 40.9 genera for Walker's Bluff. The 95% confidence intervals on the curves indicate no significant difference in richness between the bivalve samples studied to date from the two localities (95% CI were 33.97- 44.81 for Tar Heel and 40.27- 41.52 for Walker’s Bluff). The study is ongoing and results may change with the inclusion of additional samples and analysis of the gastropod fauna.