The Fossil Record of Shell-Breaking Predation on Marine Bivalves and Gastropods

Any treatment of durophagous (shell-breaking) predation on bivalves and gastropods through geologic time must address the molluscivore's signature preserved in the victim's skeleton. Pre-ingestive breakage or crushing is only one of four methods of molluscivory (Vermeij, 1987; Harper and Skelton, 1993), the others being whole organism ingestion, insertion and extraction, and boring. Other authors in this volume treat the last behavior, whereas whole-organism ingestion, and insertion and extraction, however common, are unlikely to leave preservable evidence. Bivalve and gastropod ecologists and paleoecologists reconstruct predator-prey relationships based primarily on two, although not equally useful, categories of pre-ingestive breakage, namely lethal and sublethal (repaired) damage. Peeling crabs may leave incriminating serrated, helical fractures in whorls of high-spired gastropods (Bishop, 1975), but unfortunately most lethal fractures are far less diagnostic ofthe causal agent and often indistinguishable from abiotically induced, taphonomic agents of shell degradation. Published research has focused on gastropod and bivalve shell repairs in the fossil record (Vermeij, 1987), although reliance on sublethal damage to reconstruct predator prey interactions is not without complications. Sublethal, repaired shell damage is commonly inversely related to lethal, unrepaired shell damage (Vermeij, 1982a), a relationship that may lead to ambiguous reconstruction of predation intensity (Kowalewski et al., 1997; Leighton, in press). As strength of the predator's crushing elements increase, or, concurrently, strength of the prey shell decreases, lethal damage increases, whereas sublethal damage decreases. Furthermore, a clam victim may sustain sublethal tissue damage, such as siphon nipping (Peterson and Quammen, 1982) by fish or crabs that may not break the shell. No evidence, other than slowed growth rate (Coen and Heck, 1991) records the predatory attempt. Furthermore, time-averaging (Cadee et al., 1997) and high within-habitat variability (Schmidt, 1989) may complicate inferences on prey evolution derived from the record of shell repairs, particularly if sample sizes are small.