It is often assumed that species generally reach their highest densities in the centre of their range and decline in abundance towards the range edges. A number of mechanisms have been proposed that could theoretically support this pattern, and several ecological theories have been developed based on the assumption that this pattern occurs in nature. However, few studies have quantified geographical patterns of species abundance throughout species ranges. This is largely because of the logistical challenges of sampling throughout the large spatial areas of most species ranges. We use intertidal invertebrates, which have relatively well defined linear ranges, to test the hypothesis that species are most abundant in the centres of their ranges.
Our sampling programme covered all or most of the ranges of twelve intertidal invertebrate species along the Pacific coast of North America, from Cabo San Lucas (Baja California, Mexico) to Shelikof Island (AK, USA).
We sampled invertebrate density at forty-two field sites using quadrat and transect methods. We used a shape fitting procedure to find idealized range shapes that best fit the sampled distributions of abundance. The idealized range shapes represented both a distribution where abundance was highest at the range centre and distributions where abundance was highest at one or both of the range edges.
Overall, this suite of species did not show the expected pattern of high abundance near the range centre. Six of the species showed patterns indicative of high densities near one of their range edges, whereas only two showed patterns with high densities near their range centres. Furthermore, nine of the twelve species had sites near the range edges in which density ranked in the top 20% of all sites.
The hypothesis that species are most abundant in the centre of their ranges cannot be generalized to this diverse suite of intertidal organisms. The diversity of distribution shapes that we found suggest that evolutionary and ecological theories that assume high abundance at range centres should be re-examined with consideration of alternative abundance distributions. We suggest that sampled geographical distributions of abundance can be combined with demographic and physical factor data taken at the same scale to test hypotheses related to the causes of range boundaries and the responses of species ranges to climatic change.