Cushion plants as islands.

The acarine fauna of two abundant species of cushion plant on the high, short-grass prairie of S.E. Wyoming were used to test The MacArthur-Wilson Theory of Island Biogeography. Multiple regression analysis using area, distance and percent moisture as independent variables and number of mite species and number of mite individuals were run for the two sampling dates. Results showed area alone to be consistently and highly correlated (r=0.84-0.94) with both species and individuals for one cushion species. The slopes of the species-area and individuals-area curves are among the highest recorded and were significantly higher on the second sampling date. Selective seasonal changes in the fauna were shown by increases both in numbers of species and individuals, mainly on larger cushions, for the later sampling period. It is hypothesized that seasonal changes are due to an increase in the number of predator species in response to an increase in the number of prey items. The slopes of the species-area curves are compared with those in the literature and it is argued that slope values are more dependent upon the taxonomic group being studied than on whether the island is insular or oceanic. Finally, we suggest that The MacArthur-Wilson Theory is not applicable to islands which 1) exhibit continuous growth, 2) lack a discrete species source, and 3) are relatively transitory.