That predators are important in structuring communities is well supported by theoretical models (Cramer and May 1972, Roughgarden and Feldman 1975, Caswell 1978), experimental studies (Connell 1970, Paine 1974, Morin 1983,1984, Carpenter et al. 1987), and surveys of natural communities (Brooks and Dodson 1965, Wells 1970). Predator-prey studies in a variety of marine, brackish, and freshwater systems demonstrate that predation can be an important force in influencing prey behavior (Peckarsky 1980, Gilliam 1989), distribution (Peckarsky and Dodson 1980, Zaret and SufFem 1976), and abundance (Crowder and Cooper 1982). The best studies demonstrating that predators control prey diversity and abundance come from the rocky intertidal zone and freshwater lakes (Sih et al. 1985). In freshwater, the best evidence supporting the role of predators as a major structuring force comes from pelagic zooplankton communities (Brooks and Dodson 1965, Carpenter et al. 1987). Work on freshwater amphibian communities similarly demonstrates that predators regulate prey communities (Morin 1983, Alford 1989). Relatively few studies have been conducted in freshwater benthic systems, and results are conflicting regarding the role of predators. Of the 26 freshwater lentic studies reviewed by Sih et al. (1985), seven were done on benthic invertebrates. The negative results of Thorp and Bergey (1981a, 1981b) are at the center of the debate on the importance of predation in benthic freshwater systems. Some argue, that because Thorp and Bergey did not document fish in their controls, their study may not be a good refutation of the importance of predation (Crowder and Cooper 1982). Results from more recent work (Cooper et al. 1990, Hall et al. 1990) suggest that the large mesh size used by Thorp and Bergey may have contributed to their negative findings as well. In another study, Gilinsky (1984) found that excluding predators from areas in a North Carolina pond resulted in changes in the invertebrate community; however, her conclusions are not strong because she failed to provide evidence that the affected species were actually preyed upon by the predator. Also, because she first removed predators from cages in August, after placing the cages in the pond in March, her results may be due to predator densities in cages from March to August rather than the result of predator manipulation; because she did not use a repeated-measures design, the results would be expected to be strongly influenced by initial starting densities of invertebrates. Based on my review of the literature, before the start of my study, I concluded that the equivocal results in freshwater systems were due to inadequacies in experimental design rather than lack of a predator effect. As a result, I examined the importance of predation by pumpkinseed sunfish (Lenomis gibbosus) as a structuring force on the benthic snail assemblages in northern Wisconsin lakes. I chose pumpkinseeds and snails because 1) there exists a variety of snail species with differing shell characteristics (that presumably determined vulnerability to pumpkinseed); 2) snails are eaten by a diverse group of predators, including fishes and invertebrates, yet no consumers within these groups are specialist molluscivores like pumpkinseeds; 3) snails are differentially susceptible to pumpkinseeds owing to variable and easily quantifiable shell characteristics (e.g., shell strength, Stein et al. 1984); 4) predator avoidance behaviors by snails are simple or nonexistent; 5) unlike aquatic insects, snails are relatively easy to quantify in the field and identify to species; 6) competition and abiotic factors appear less important in determining dominance by one snail species than predation (Lodge et al. 1987); and 7) snails are easy to collect, hold, and manipulate in laboratory and field experiments. I took a multi-faceted approach to determining if pumpkinseeds influence snail assemblage structure: 1) I quantified snail shell characteristics that influenced pumpkinseed choice of snails, and I performed laboratory experiments to determine size- and species-selection by pumpkinseed under a variety of experimental conditions (Chapter I); 2) 1 developed a foraging model to further explore how pumpkinseeds might choose snails in the field (Chapter II); and 3) I conducted field-cage experiments to determine if predator induced changes in the field were supported by predictions from laboratory snail choice experiments and model predictions (Chapter III).