Various aspects of the MacArthur-Wilson noninteractive model were tested using protozoan communities on artificial islands in both field and laboratory experiments. Laboratory experiments produced results comparable to those obtained in the field. In addition, the effects of epicenter (source pool) maturity and the distance of islands from the epicenter were easily tested in the laboratory systems; these would be very difficult to test in field experiments involving microbial communities. The size of an island, its distance from an epicenter, its maturity, the species composition of the epicenter community, the season during which the islands were colonized and the quality of the surrounding water (particularly its trophic status) all seemed to influence the colonization rate and equilibrium number of species on islands. These factors seem to exert their greatest effect during the noninteractive phase of colonization, which is characterized by species accumulation and a temporary equilibrium and which, in turn, is produced by a balance of immigration and extinction of colonists.