INTERSPECIFIC ANTAGONISM AND SPECIES COEXISTENCE IN A DIVERSE GUILD OF LARVAL TREMATODE PARASITES

The salt marsh snail Cerithidea californica is first intermediate host to a diverse guild of larval trematode parasites. In Bolinas Lagoon, in central California, the site of this study, at least 15 species of trematodes infect snail populations. This study investigated patterns of interspecific association and interaction among members of this par asite guild. Seven to 19 host subpopulations were sampled annually at each of two sites in the lagoon from 1981 to 1988. Mixed-species infec tions constituted only 2.5% of the 5025 infections examined in the stu dy. A Monte Carlo simulation procedure demonstrated that the numbers o f such infections were often less than would be expected by chance, es pecially when the overall prevalence of infection was high. Patterns o f association between particular pairs of species depended on whether the species' life histories include redial or only sporocyst larvae. S pecies that develop as rediae were predominantly negatively associated with other redial species and with most species that develop only as sporocysts. There was weak evidence of positive interspecific associat ion between a few redial and sporocyst-only species, while members of other such pairs were distributed independently. Associations between sporocyst-only species were either weakly positive or neutral. Snails carrying known infections were marked, released, and recaptured at bot h study sites. During their exposure in the field, some initial infect ions were invaded by another parasite species that often excluded the first parasite. The vulnerability of a parasite species to invasion an d replacement by another differed among the tested species. Infections of the largest redial species, Parorchis acanthus, were especially re sistant to replacement, while those of the smallest redial species, Eu haplorchis californiensis, were the most frequently excluded. Four oth er species were invaded or replaced at intermediate rates. The two lar gest redial species, P. acanthus and Himasthla rhigedana, were respons ible for >90% of the invasions or exclusions. Direct observations show ed that the rediae of these species prey on the larval stages of other species, as do the rediae of Echinoparyphium sp. This direct form of interspecific antagonism is probably the primary mechanism by which su ch species exclude others from host snails, as has been widely demonst rated in similar freshwater snail-trematode systems. While hierarchica l, negative interactions prevent the coexistence of species at the lev el of the individual host, the mark-recapture study showed that rates of exclusion are low for most subordinate species, with the exception of Euhaplorchis californiensis. At the level of the host subpopulation , the assemblage of larval trematodes is diverse, and its composition is temporally and spatially variable. There is no trend toward dominan ce of the assemblage by large redial species as the level of infection rises within aging cohorts or subpopulations of hosts. These patterns of guild structure within host cohorts and subpopulations are consist ent with the hypothesis that recruitment processes rather than intersp ecific interactions primarily determine the composition and relative a bundance of species at this regional level. Several characteristics of snail-trematode systems that may promote regional coexistence of such a large number of potentially interacting parasite species are the is olated and subdivided nature of the host resource, the aggregated dist ributions of larval stages, and the differential exploitation of diffe rent-sized hosts. Many features of this system are consistent with Pri ce's (1980) non-equilibrial view of parasite communities.