Abstract

The species-area curve has been used to link the biological with the geographical. Larger areas of land would seem to contain more species as a result of both the effect of sampling (i.e. more samples are taken to represent larger areas) and ecological processes (i.e. island biogeography theory and hypotheses relating to habitat diversity, successional development, species-energy, target-area, incidence function, small island habitat and disturbance). Unfortunately, the species-area curve is usually interpreted as though it was due entirely to ecological processes when it could be due largely to sampling. Modelled and real data (for forests in Ghana) demonstrated that while the effect of both ecological processes alone and sampling alone increased species number with area, only ecological processes could be expected to increase the number of species per unit area. These results suggest that before a species-area curve could be used as an indicator of ecological processes the effect of sampling on the species-area curve must first be removed.