Introduction: EXTINCTION AND NATICID PREDATION OF THE BIVALVE CHIONE VON MÜHLFELD IN THE LATE NEOGENE OF FLORIDA

INTRODUCTION

The theory of evolutionary escalation predicts that if biological enemies are indeed among the most important agents of natural selection, then the interactions between predators and prey have become more escalated over geological time (Vermeij 1987). This prediction is borne out by numerous lines of evidence preserved in the Phanerozoic fossil record (Vermeij 1987). The process of escalation since the Cambrian has not been monotonic, however, because of the repeated evolution of particular modes of predation (for example, exoskeletal drilling has apparently evolved multiple times [Sheehan and Lesperance 1978; Fursich and Jablonski 1984], and perturbations resulting from mass extinctions).

If the mechanism underlying a mass extinction generates ecological hardship within a community, then the disruption of energy supplied to that community will increase the probability of extinction for species with costly, highly developed enemy-related adaptations (Vermeij 1987). Extensive work on the dynamics of naticid gastropod drilling predation on other molluscs since the Cretaceous supports the idea that mass extinctions have interrupted the process of long-term escalation (Kelley and Hansen 1996). These studies have been conducted primarily from the standpoint of the ecological community, but similar processes may operate at the level of individual monophyletic clades that survive mass extinctions; selection will favour a reduction in the expression or frequency of enemy-related traits during an episode of mass extinction. We present here data regarding the effects of the Late Pliocene mass extinction in the southeast United States (Stanley 1986) on the relationship between a Neogene bivalve from Florida, Chione von Mühlfeld, 1811 (Fig. 1), and its naticid gastropod predators. We demonstrate that the precise interaction of extinction and escalation cannot be determined without detailed knowledge of the morphological identity of the prey species. We also present evidence supporting the microevolutionary increase of shell thickness possibly in response to drilling predation (Kelley 1991).

Naticid Drilling

Gastropods of the family Naticidae employ a stereotyped method of predation on other molluscs, including Chione. Acidic dissolution of outer shell layers, followed by radular scraping and penetration, results in a characteristic beveled or countersunk hole (Figure 2). Traces of successful and unsuccessful drilling have proven to be of enormous value to the study of predation and escalation in the geological past (Vermeij 1987; Kelley 1989; Anderson et al. 1991; Hansen and Kelley 1995). Earlier work established that naticid predators are selective with respect to drill hole location and prey size, and that prey choice is related to such parameters of prey value and escalation as tissue mass and shell thickness (Kitchell et al. 1981). An extension of the theory of escalation (the "Kelley-Hansen hypothesis") predicts that perturbation due to a mass extinction should result in a reduction of the degree of stereotypy of a naticid predator-prey system within a community immediately following an extinction, with subsequent recovery (Kelley and Hansen 1996). This, perhaps, results from the extinction of prey species that were partners in the predator-prey system. An unanswered question, however, is the effect(s) of mass extinctions at the level of a single clade. If mass extinctions do indeed disrupt the process of escalation, then two predictions can be made concerning the Chione-naticid system: (1) selection should have favoured the survival of energetically cheaper, less escalated variation in the genus if, as commonly believed, a reduction of productivity played a role in the extinctions (Allmon et al. 1993; Jones and Allmon 1995; Allmon et al. 1996; Roopnarine 1996), and (2) naticid predatory behaviour should have become less stereotyped and less predictable immediately following the extinctions, in the Late Pliocene to Early Pleistocene.

Neogene Extinctions in Tropical America

The late Neogene of the tropical western Atlantic was a time of ecological upheaval as evidenced by widespread molluscan extinctions, speciation and immigration (Woodring 1966; Stanley 1986; Vermeij and Petuch 1986; Allmon et al. 1993; Vermeij and Rosenberg 1994; Roopnarine 1996; Roopnarine 1997). The proximal cause(s) of the extinctions remains unknown, although there is increasing evidence to support a precipitous decline of planktonic primary productivity associated with disruptions of regional circulation. This decline has been linked to uplift of the Isthmus of Panama (approximately 3.5 mya), subsequent termination of flow between the Atlantic and Pacific oceans, with resultant changes in oceanographic circulation and regional upwelling. The search for an extinction mechanism has focused mainly on the documentation of the intensity of the extinction (Allmon et al. 1993; Jackson et al. 1993), and the analysis of geochemical proxies of ecological factors (Teranes et al. 1996). Here, we analyze the changes exhibited by apparent survivors of the extinction episode, examining both the relationship of Chione to its naticid predators, and morphological change within the genus.

Morphological analysis of prey is an important component of the examination of predator-prey dynamics in the fossil record. Morphological differences between pre- and post-extinction populations arise from either selection-driven changes of phenotype, or the immigration of morphologically similar individuals from other geographic regions as refugees or opportunists. The analyses should be of a morphometric nature because many defensive traits (e.g., shell thickness or external sculpture) often vary continuously. In fact, there is a significant morphometric discontinuity between the Pliocene and post-Pliocene species of Chione in the southeastern United States, Chione erosa Dall 1903 and C. cancellata Linnaeus 1767 respectively, despite their overall similarity (Roopnarine 1995). Morphometric analysis of distance measures indicates that C. erosa has a relatively larger lunule, longer shell and more dorsal disposition of the hinge when compared to C. cancellata (Roopnarine 1995). Dall (1903) originally separated the species on the basis of lunule length. There is no significant difference in size between the two species (Roopnarine 1996).

Chione erosa became extinct during the Late Pliocene, and was replaced by C. cancellata. The morphometric distinction between contemporaneous Late Pliocene Chione specimens from North Carolina (the Waccamaw Fm.) and Florida, suggests that the North Carolina specimens are assignable to a different species, or were representative of significant intraspecific geographic variation (Roopnarine 1995). The former hypothesis is supported by the morphometric similarity between Waccamaw specimens and later Pleistocene and Recent specimens of C. cancellata from Florida, as well as South Carolina, suggesting that C. cancellata was present in Late Pliocene waters north of Florida.

This raises the possibility that some extinction survivors are phylogenetically distinct "pseudosurvivors", distinguishable only by robust methods of morphometric analysis. We therefore examined the effects of the Plio-Pleistocene extinction on Chione-naticid interaction by quantifying various aspects of naticid predatory behaviour, bivalve anti-predatory morphology and the morphological distinction between C. erosa and C. cancellata.