Interspecific Competition and Species Co-Occurrence Patterns on Islands: Null Models and the Evaluation of Evidence

Abstract
Several recent studies have adduced non-overlapping geographic ranges as evidence that 2 [animal] species competitively exclude one another. To show that a particular pattern of species'' co-occurrence is unusually exclusive (for whatever reason), one must contrast the observed pattern with that expected were species placed independently of one another and consider all species, not only those that appear a priori to have unusually exclusive ranges. Even when one is able to show that an unusually large number of species'' distributions are improperly exclusive, in the absence of independent non-distributional evidence capable of eliminating other reasonable explanations for these odd distributions, one cannot infer that competition or any other specific cause is responsible for the observed pattern.