Habitat influences diet overlap in aquatic snake assemblages A. M. Durso 1,3 , J. D. Willson 2,3 & C. T. Winne 3 1 Department of Biology, Utah State University, Logan, UT, USA 2 Department of Biological Sciences, University of Arkansas, Fayetteville, AR, USA 3 Savannah River Ecology Laboratory, University of Georgia, Aiken, SC, USA Keywords niche overlap; niche partitioning; competition; isolated wetlands; community dynamics; occupancy. Correspondence Andrew M. Durso, Department of Biology, Utah State University, 5305 Old Main Hill, Logan, UT 84321, USA. Tel: 919 349 7967; Fax: 435 797 1575 Email: amdurso@gmail.com Editor: Mark-Oliver Rödel Received 27 January 2013; revised 25 June 2013; accepted 25 June 2013 doi:10.1111/jzo.12061 Abstract Competition for prey is thought to be important in structuring snake assemblages. However, due in part to the secretive behavior and low detectability of many snake species, this generalization is based on a limited number of studies, most of which focus on a single study site. We examined differences in diet composition, trophic niche overlap, site occupancy and detectability of five sympatric aquatic snake species between two different habitat types in the Southeastern US, replicated at the landscape scale: permanent wetlands with fishes (n = 13) and isolated, often ephemeral wetlands without fishes (n = 10). We collected >3700 prey items from snakes and compared diet composition among snake species to examine niche breadth and overlap, correcting for relative availability of prey captured indepen- dently in the same wetlands. We evaluated evidence for competitive exclusion by estimating the probability of co-occupancy for pairs of snake species in each habitat type using occupancy modeling. In wetlands with fishes, niche overlap was low, suggesting resource partitioning. Conversely, in wetlands without fishes, niche overlap was high, with most species feeding on larval or paedomorphic ambystomatid salamanders, but competitive exclusion did not occur. We suggest that high co-occupancy of aquatic snakes in wetlands without fishes despite the apparent lack of resource partitioning is due to a combination of seasonally high abundance of high quality amphibian prey, unique aspects of predator physiology and stochastic abiotic processes that prevent these systems from reaching equilib- rium. Our results demonstrate that snake diets can be highly context (e.g. habitat)- specific. Studies should consider other factors in addition to competition for prey when attempting to understand snake population and community dynamics. Introduction When resources are limited, consumers are thought to reduce interspecific competition by partitioning (Pianka, 1974). By occupying discrete niche space (e.g. by feeding on different types or sizes of prey), individuals of each consumer species within a community minimize competitive interactions with heterospecifics. Theoretically, this permits a greater number of syntopic consumers to occupy the same trophic level (MacArthur, 1970). An increase in the number of competitors or a decrease in the availability of resources can lead to a reduction in niche overlap due to increased partitioning of resources over evolutionary time (Hardin, 1960). Although competition for prey is only one of several interacting factors contributing to resource partitioning (Toft, 1985), it is consid- ered a key factor in structuring many species assemblages (e.g. Luiselli, 2006c). Of the evolutionary radiations of tetrapods, snakes are unusual in being exclusively carnivorous. Despite gape limita- tion, many species have a wide diet breadth, by virtue of highly kinetic jaws that allow them to eat large prey relative to their body size. Snake assemblages typically consist of many species of both specialists and generalists, and often include many similar, often related, species feeding at similar trophic levels (Luiselli, 2006a). Several reviews have concluded that exploitative competition for food is the most important force in structuring snake assemblages (Arnold, 1972; Schoener, 1977; Toft, 1985). Data from either field or museum studies form the basis for generalizations about the composition of snake diets and the structure of snake assemblages (Toft, 1985). Although many excellent studies on snake diets exist, field studies often focus on a single study site (e.g. Luiselli, 2006b), whereas studies based on museum specimens generally integrate records over large spatial and temporal scales, without replication (e.g. Henderson, Dixon & Soini, 1978). Thus, little is known about spatiotemporal variation in snake diets, or the effects thereof on the structure of snake assem- blages. Both Luiselli (2006c) and Gibbons & Dorcas (2004) Journal of Zoology Journal of Zoology. Print ISSN 0952-8369 Journal of Zoology 291 (2013) 185–193 © 2013 The Zoological Society of London 185