Species richness in Atlantic deep-sea fishes assessed in terms of the mid-domain effect and Rapoport's rule

A decrease in species richness with increasing latitude has been documented for a broad range of taxonomic groups. A number of hypotheses relating to biological, environmental, and historical factors have been proposed to explain this phenomenon, and the mid-domain effect (MDE) has been proposed in the form of a null model. This model considers only the geometry of spatial gradients and species' range extents, excluding any assumptions of environmental, biological or historical causes, and predicts that species richness will peak in the centre of a domain in which species occur when their ranges are randomly distributed. This model has been applied to observed latitudinal, elevational and depth gradients as a test to quantify the extent to which non-random processes influence species richness patterns in comparison to those based on geographical boundary constraints alone. We apply the M DE model to empirical datasets for the ranges of the bottom-living fish species occurring in the Faroe-Iceland Ridge, Denmark Strait, Southern New England and Northern Gulf of Mexico regions of the North Atlantic Ocean. The observed patterns show a decline in richness with depth, and do not match the richness patterns produced by the null model. Therefore it can be said that non-random processes have resulted in the observed patterns. Applied to bathymetric ranges, Rapoport's rule predicts that richness decreases and range size increases with depth and latitude. The rule explained decreasing fish species richness with depth and between latitudes, but did not appear to explain increasing range size with depth. (c) 2006 Elsevier Ltd. All rights reserved.