Species-Area Relationships of Insects on Host Plants: Herbivores on Rosebay Willowherb

Abstract

(1) The species-area relationship for phytophagous insect species feeding on rosebay willowherb (Fireweed), Chamerion angustifolium (L.) J. Holub was studied during 1979 and 1980. Three aspects were examined. (a) The number of species and individual insects present on patches ranging in size from 0-25 m2 to 7050 m2, at sites within Great Britain. (b) The detailed mapping of the distribution of three species within a 28 x 20 m patch of rosebay. (c) The colonization of new, transplanted patches of rosebay over a 2-year period. (2) There was a shallow, but significant regression of species numbers on patch size, the slope of which was virtually identical for both years. The height of plants (a measure of habitat heterogeneity) did not significantly increase the variation explained by the regression. (3) The number of species, and the total number of individuals, per plant also increased with patch size. (4) The abundance of individual species per plant was either positively related to patch size, had no correlation or, for one species, was negatively related. (5) There was no clear relationship between species diversity and patch size. (6) For the majority of species, the variance to mean ratio of individuals per plant within patches increased with increasing population size. For those species whose abundance was related to patch size, this implies that the distributions became more aggregated with increasing patch size. (7) The three species whose distributions were mapped within a patch all had aggregated distributions of numbers per plant, and between different regions of the patch. (8) The colonization of new patches was rapid. Mompha raschkiella had attained densities comparable to those observed naturally within one season, as did two other species sampled during the second year. (9) Habitat heterogeneity is probably not an important aspect of the species-area relationship nor can passive sampling be evoked in isolation from other factors. Instead it is thought that passive sampling, 'active sampling' (Root's resource concentration hypothesis) and area-dependent extinction all play a part in generating the species-area relationship, the contribution of each varying from species to species.