Benthic macroinvertebrate colonization of twelve 0.04-ha experimental ponds was studied for the 1st yr of their existence. The ponds were filled in late January 1988, and artificial substrate samples were retrieved at 4-wk intervals from February 25, 1988 to February 25, 1989. The first organisms (<u>Chironomus</u> and ceratopogonidae) were collected 4 wk after filling. Diptera, primarily several genera of chironomids, were followed in colonizing the ponds by Coleoptera, Ephemeroptera, and odonata. The arrival of new taxa corresponded to adult flight periods. Non-insect taxa (oligochaeta, Nematoda, Amphipoda) were collected in relatively low numbers and then only late in the year. A total of 29 taxa were collected, 13 of which were present in all 12 ponds. Mean number of taxa, diversity, and evenness increased rapidly during the first 7 mo but never became very high. Density increased slowly throughout the spring and summer, increased rapidly during the autumn and eventually became quite high. Density increased slowly throughout the spring and summer, increased rapidly during the autumn and eventually became quite high. Community structure was dominated by Chironomidae (ca. 85%) and Ephemeroptera (ca. 9%). Trophic function was dominated by detritus-feeding collectors (ca. 79%). Factors that influenced the structure and function of the new experimental ponds appeared to be: lack of connection to colonized waters, small size, and simple, homogeneous habitat and feod resources. Although at the end of the study the aquatic insect fauna was typical of shallow lentic environments, the mesocosms were not considered ready for use in chemical testing because of low densities of other macroinvertebrates. The mesocosms were somewhat variable in regard to community structure, however, no mesocosm was found to be a consistent outlier. Density could be measured less precisely than other structural parameters (number of taxa, diversity, and evenness). All of the structural parameters, however, could be estimated with reasonable precision (within 40%) using groups of three mesocosms. Greater numbers of replicate ponds would be required to precisely measure the density of individual taxa. Secondary production was measured for seven dominant taxa. Although the estimation of secondary production required more time, it could be done at a similar level of precision as density and was considered more ecologically meaningful than structural measurements. / Ph. D.