CAN WE LEARN ANYTHING FROM ALL THOSE PIECES? OBTAINING DATA ON DRILLING PREDATION FROM FRAGMENTED HIGH-SPIRED GASTROPOD SHELLS

Studies on drilling predation often analyze site stereotypy and drilling frequency. However, incomplete specimens create particular problems for these analyses. Without a method of normalizing drill hole location on incomplete shells, it is difficult to study site stereotypy. Additionally, in order to calculate drilling frequency, a method is needed to address the possibility that multiple fragments may have been produced by the same individual. Shell breakage—due to transport, compaction, and/or predation—is a significant taphonomic challenge in studying high-spired gastropods. As a result, many previous studies have considered only “complete specimens”.

We applied two approaches to taking account of incomplete shells in a study of drilling predation in Turritella alumensis from the Pliocene Jackson Bluff Formation at Alum Bluff, Florida. First, we studied site stereotypy using a Theoretical Apex System to identify hypothesized apices in order to normalize drill hole location. Using this method, we found that including incomplete specimens gave a different picture of site stereotypy than some previous authors have noted, especially towards the apical end where smaller predators drill. We also observed two populations of drillers acting on different portions of the shells with an undrilled region in between. This pattern might have been missed by using only “complete” specimens. Second, we calculated drilling frequency using shells of varying completeness by creating a new calculation for the minimum number of individuals within a bulk sample. This calculation treated all incomplete shells as isosceles trapezoids such that each shell could only occupy a given width once. The most commonly occupied width was treated as the minimum number of individuals present. Drilling frequency for shells drilled by larger predators generated using this method was consistent with that calculated in previous work using only “complete” specimens. Although this was consistent with previous work, “completeness” is rarely well defined. Our method normalizes the calculated minimum number of individuals in a sample, making it possible to compare samples with variable amounts of shell breakage.