Rapoport's bathymetric rule and the latitudinal species diversity gradient for Northeast Pacific fishes and Northwest Atlantic gastropods: Evidence against a causal link

Abstract

Aim: Few studies have explicitly considered the recurrent pattern of declining species diversity and increasing geographical range size that exists for numerous taxa across a variety of physical gradients. We extend Stevens' [Stevens, G.C. (1996) Journal of Biogeography, 23, 149] work on Rapoport's bathymetric rule, using a more complete latitudinal assemblage of Northeast Pacific fishes and new data from Northwest Atlantic gastropods, to show that bathymetrical range size and species diversity are not causally linked. Location: Fishes from the Northeast Pacific (0°-60° N) and gastropods from the Northwest Atlantic (0°-74° N) distributed from the surface to depths greater than 200 m. Methods: Species pools were divided into three bathymetrical subgroups: (1) species restricted to shallow waters, between the surface and 200 m, (2) species that occurred in waters, both shallow and deep of 200 m, and (3) species restricted to waters deeper than 200 m. Median bathymetrical range size and total number of species were plotted against latitude (2° bins) using Stevens' method, for the entire species pool and individual bathymetrical groups. Results: For both fishes and gastropods, the apparent link between extratropical diversity and bathymetrical range size is an artefact resulting from the disproportionately high number of shallow restricted species in tropical latitudes, and the loss of these species in temperate latitudes. Furthermore, the extratropical gradient in gastropod diversity and bathymetrical range size are decoupled by approximately 15°, and while the latitudinal pattern for diversity is consistent across bathymetrical groups, median bathymetrical range size is highly irregular. Main conclusions: These results suggests that functional groups can contribute disproportionately to patterns apparent at larger scales and that analysis of ecographical patterns by subregion is a novel approach that can help resolve debates over causality when patterns are seemingly coincident.