TAXON-SPECIFIC SAMPLING BY NOVICES AND EXPERIENCED COLLECTORS YIELDS ROBUST DATA ON DRILLING PREDATION ON FOSSIL BIVALVES

Bulk sampling has been considered to provide more robust information on predation in the fossil record than does taxon-specific (targeted) sampling. Our recent work has demonstrated that targeted sampling, even by a trained novice, can yield results comparable to bulk sampling. The present study compares drilling predation metrics for samples made by five novice collectors and three veteran collectors to determine whether trained novices could produce results comparable to those of experienced collectors, and whether targeted sampling produced results comparable to bulk sampling.

We used targeted sampling in the lower Waccamaw Formation (Lower Pleistocene) at Register Quarry near Old Dock, NC. Five replicate taxon-specific samples were made by each collector, retrieving every specimen visible within ~ 1 – 4 m² areas for the bivalves Astarte concentrica, Cyclocardia granulata,and Lirophora latilirata. Shell length and thickness were measured and position and size of drillholes determined. Frequency of left valves, drilling frequency (number of valves with a complete drillhole divided by half the number of valves), prey effectiveness (% of drillholes that were incomplete), and size selectivity (correlation of drillhole size and prey size) were determined for each taxon replicate.

For each variable, we compared the results for the novice collectors to results from bulk samples and to those for all veteran collectors. Of 298 comparisons between unstandardized data from bulk samples and replicates collected by the five novices and three veterans, 89% showed no significant difference between targeted and bulk samples. Less than 3% of 142 comparisons of standardized data for the same samples showed a significant difference between targeted and bulk samples. Of 245 unstandardized comparisons between the novices and veterans, 95% showed no significant difference. None of the 114 size-standardized comparisons between the novices and veterans showed a significant difference. Nearly all significant differences in the unstandardized data occurred in the size-related variables length and thickness (31 of 34 for collecting method and 9 of 12 for collector expertise). Previous results were not anomalous; use of targeted sampling in studying drilling predation is again validated.