Using Island Biogeographic Distributions to Determine if Colonization is Stochastic

Abstract

A number of models for the process of island colonization have been proposed in the literature, ranging from completely stochastic and noninteractive through deterministic and heavily influenced by interspecific competion. Biogeographic distributional data are cited in support of these models, but statistical tests have often been omitted, or are so weak where performed that one is not completely inclined to accept the hypotheses in question even where rejection is not yet indicated. Broadly speaking, plant and invertebrate distributional data have been interpreted as not heavily influenced by interactions, while vertebrate data have often been cast in a context of diffuse competion. Analysis to date has not been sufficiently profound that one must discard Occam''s razor and seek different explanations for the distributions of different taxa.