The ubiquity of latitudinal gradients in species diversity has led to a search for biological mechanisms that have universal applicability (i.e., a primary cause). Although over 20 hypotheses have been proposed to account for such gradients (Rohde 1992), recent research suggests that the existence of a gradient, per se, could be the result of stochastic processes (Colwell and Hurtt 1994, Willig and Lyons 1998). Moreover, many of the deterministic hypotheses are circular or unsubstantiated, and all lead to qualitative predictions (i.e., increases in richness toward the tropics) rather than quantitative predictions about the form of the pattern or the number of species at particular latitudes. As a consequence, much of the supportive data for deterministic hypotheses are indirect and not subject to analyses via strong inference (Platt 1964, Pickett et al. 1994). I propose to conduct research for a 12 month period at NCEAS, during which I will evaluate in a quantitative fashion the predictions of a stochastic model for a variety of taxa (i.e., mammals, birds, reptiles, amphibians, fish) occupying different geographic domains(North America, South America, Africa, Australia, Indo-Pacific Ocean, Atlantic Ocean)

¿ First, I will refine the extant model (Willig and Lyons 1998) to include information about average range size of the taxa of interest, and expand it to two dimensions (latitude and longitude).

¿ Second, I will obtain and digitize data from a variety of sources on the distributions of taxa and incorporate them into a GIS (Geographic Information System).

¿ Third, I will test null models based on gathered data, and convene a working group to explore patterns and search for factors responsible for deviations from the null model.