While latitudinal gradients in diversity are one of the clearest patterns in ecology, with diversity peaking in the tropics, we know surprisingly little about the ecological and evolutionary processes that underlie this pattern.  Many hypotheses have been proposed, and one in particular, the ‘biotic interactions hypothesis’ is receiving increasing attention. This hypothesis predicts that higher speciation rates occur at lower latitudes because of stronger and more specialized species interactions.  If true, this evolutionary mechanism has important consequences for how species coexist at ecological timescales.  For example, predation pressure is assumed to be a primary factor in shaping marine communities, but the effect of predation on species diversity per se has never been tested in a comprehensive, latitudinal-scale experiment.  To test the hypothesis that predation will have a stronger effect on diversity in the tropics, we allowed communities of epibenthic marine invertebrates to develop in the presence or absence of predators in four regions along a latitudinal gradient: Connecticut (41°N), Florida (27°N), Belize (16°N), and Panama (9°N).  Communities were either caged to exclude predators or left open and were sampled after three months.  Additional experiments were conducted in Belize and Panama to further elucidate the effect of predation on diversity. 

Results/Conclusions

Results indicate a clear latitudinal shift in the impact of predation on community diversity.  While there was no effect of predator exclusions on diversity in the temperate and subtropical zones (Connecticut and Florida, respectively), epifaunal communities that developed in the absence of predators in the tropics (both Belize and Panama) were nearly three times more diverse than communities where predators were present.  When diverse tropical communities that had developed in the absence of predators were then exposed to predation, diversity was quickly reduced in comparison to communities that remained in predator exclusions.  While overall effects were strong and consistent in the tropics, results also indicate between-habitat spatial variation in recruitment and predation pressure.  Communities in Connecticut did not vary among treatments in diversity, although relative abundances of certain taxa (i.e., tunicates) were strongly affected.  These results demonstrate that while predation is an important community driver at all latitudes, predation has the greatest impact on diversity in the tropics.  Therefore, at ecological timescales, strong and spatially variable interspecific interaction strengths coupled with spatial and temporal variability in recruitment and community composition allow a diverse assemblage of tropical species to coexist at a regional scale.