The Mid-Domain Effect: There’s a Baby in the Bathwater

In a recent article (Colwell et al. 2004; hereafter CRG), we reviewed all studies of the mid-domain effect (MDE) published up to that time and responded to earlier critiques of MDE by (among others) Hawkins and Diniz-Filho (2002) and Zapata et al. (2003). The discussion continues here with comments on our article by Hawkins, DinizFilho, and Weiss (2005; hereafter HDW) and by Zapata, Gaston, and Chown (2005; hereafter ZGC). In this article, we respond to HDW and ZGC. Null models and null hypotheses. ZGC defend their view of MDE models as null hypotheses subject to falsification. In contrast, proponents of MDE theory have always held that richness patterns have multiple causes. For example, referring to the effect of geometric (boundary) constraints on geographic patterns of species richness, Colwell and Lees (2000, p. 79) stated that “the question is not whether geometry affects such patterns, but by how much.” We view any assessment of the role of geometric constraints as a problem of estimating the magnitude of the contribution (if any) of MDE to richness patterns (“how much”). Just as with other candidate causes of richness patterns (e.g., climatic, topographic, or historical drivers), the appropriate null hypothesis (if any) is that MDE makes no contribution to richness patterns, not that MDE uniquely and fully accounts for richness patterns. It is for this reason that we distinguish between MDE as a model and the hypotheses that can be framed regarding its predictions.