The ecology of an endemic gall midge, Rhopalomyia californica Felt, and its larval parasites was studied over a 3-year period in the city of Davis, California. Six species of parasites were associated with larvae of the midge: Torymus koebelei (Huber), T. baccharidis (Huber), Platygaster california (Ashmead), Zatropis capitis Burks, Mesopolobus sp., and Tetrastichus sp. However, the midge was from four to ten times more abundant in Davis than in the endemic environment. In Davis, neither parasitization by any given species nor by the aggregate of species was a consistent regulating factor. These results suggest that, whereas parasite guilds in urban environments may be structurally similar to their non-urban counterparts, adequate or comparable levels of natural biological control of the host will not necessarily result. Thus, the use of native plants in urban environments may not always minimize attendent pest problems. The data obtained provide some empirical support for multiple-species introductions in classical biological control and illustrate the utility of facultative secondary parasites. A method for assessing competitive interactions among parasites before introduction is described, and the relevance of the theory of r- and K-selection to biological control is assessed.