Red herring or low illumination? The peninsula effect revisited
The peninsula effect is the prediction that the number of species declines from a peninsula's base to its tip. We evaluated evidence for and against the peninsula effect, and conducted a field study designed to test alternative hypotheses for that effect. Location
The Florida peninsula, USA. Methods
First, we critically reviewed the accumulated literature on peninsula effects; second, we sampled microcrustaceans in palustrine wetlands on the ridges of peninsular Florida. Site selection in our field study accounted for historical effects and partially controlled for habitat effects. Statistical analyses further accounted for habitat effects, leaving peninsular geometry as the remaining causative mechanism for residual variation in species richness regression analyses. Results
Our literature review found mixed evidence (49% of cases) for a peninsula effect. However, most study designs did not control for alternative hypotheses, most comparisons of alternative hypotheses were qualitative, and most studies focused on vertebrate animals. Our field study found that freshwater microcrustaceans inhabiting isolated wetlands on Florida’s peninsular ridges do not exhibit a peninsula effect. Essentially, no variation in microcrustacean species richness could be attributed to peninsular geometry, but 82.5% of variation in species richness was attributed to habitat and sampling effort. Main conclusions
Although our research results support the ‘red herring’ label for the peninsula effect, our literature review leads us to argue that more illumination (in the form of study design and quantitative analysis) is needed if mechanisms causing the peninsula effect hypothesis are to be resolved. Future studies of peninsula effects need to control for alternative causative hypotheses (geometry, habitat or history) in study design, and compare quantitatively the effects of hypothesized mechanisms on peninsular diversity patterns. Additionally, studies of taxa other than vertebrate animals need to be conducted for generality. Our study may serve as an example of such an approach.