Export of species from sources (epicenters) of differing ages and complexities was examined using laboratory microcosms. Polyurethane foam (PF) artificial substrates were colonized by protozoans for different time periods in a small pond. Substrates were returned to the laboratory and used as epicenters for protozoan colonization of barren PF ’islands’ in initially sterile microcosms. Islands were exposed to epicenters for either 24 h or continuously for 28 d. Islands from pairs of microcosms exposed to epicenters of identical ages were sampled on 1, 3, 7, 14, 21, 28 and 46 d after initial epicenter exposure. Colonization parameters were estimated by fitting numbers of colonizing species to the MacArthur-Wilson equilibrium model. Islands exposed continuously to epicenters were colonized by significantly more species than those exposed for only 24 h. Islands exposed to immature, species poor epicenters were colonized by a greater proportion of the source community than those exposed to more mature, species rich epicenters. All islands were depauperate compared to epicenters except those exposed to the most immature (1 d old) epicenter. Colonization continued at a reduced rate in spite of the absence of the epicenter. Results from communities with rapid species turnover and rapidly reproducing species suggest that the continuous presence of a species source is less important for colonization of a new habitat. Dispersal of potential colonists occurs rapidly in these communities. Less mature communities dominated by pioneer forms are more effective at producing colonists than more mature communities.