Abstract
We propose two new methods, an “endobiont method” and a “model method,” to quantitatively evaluate shell-peeling predation against snails and hermit crabs in fossil gastropod assemblages. The endobiont method, based on the frequency of peeled and pagurized shells, provides the minimum and maximum estimates of predation against hermit crabs. The model method, based on the frequency distribution of shell repairs per shell in peeled and unpeeled shells, estimates directly the frequencies of predation against hermit crabs and snails. Our two methods were applied to a deep-water gastropod assemblage from the Pliocene Shinzato Formation, Okinawa, Japan. Results show that peeling predation, irrespective of the high frequency of repaired breaks (though to be an indication of high predatory attacks on snails in previous studies), is mostly attributable to hermit crabs rather than to snails in elongate shells, whereas it is the reverse in spherical shells. The results suggest that the frequency of repaired breaks is strongly controlled by the abundance of elongate shells in fossil gastropod assemblages, and that the increasing evidence of peeling predation, irrespective of predation pressure, is also caused by the increase of elongate shells (e.g., caenogastropods) in gastropod assemblages since the mid-Mesozoic. The new methods could be used for the real understanding of the history of this unique biological interaction through geologic times.
