A possible alternative host for the expression of mosquitocidal cry genes is Asticcacaulis excentricus, a gram-negative aerobic bacterium that is found in aqueous environments near the surface of the water. In a series of experiments, A. excentricus was transformed with a broad-host-range plasmid vector that carried the genes for mosquitocidal toxin proteins produced by a strain of Bacillus sphaericus (a bacterium similar to B. thuringiensis) under the control of the tac1 promoter, which is a variant of the tac promoter. This transformant produced insecticidal toxin proteins of 51 and 42 kDa and was almost as toxic to Anopheles and Culex mosquito larvae as the naturally occurring high-toxicity strains of B. sphaericus. However, unlike B. sphaericus, A. excentricus does not sink when it is sprayed onto ponds infested with mosquito larvae. Moreover, A. excentricus is inexpensive to produce, as it can be grown on much simpler media than either B. sphaericus or B. thuringiensis. It does not have a high level of proteaseactivity, so the insecticidal toxin is not readily degraded. It is well adapted to environments such as those near the surface of standing water that are exposed to relatively high levels of UV light. Thus, A. excentricus cells should not be as sensitive to inactivation by UV light as those of either B. sphaericus or B. thuringiensis. However, to use a genetically engineeredstrain of A. excentricus to control mosquito populations in the environment, it will be necessary to integrate the insecticidal toxin genes into the chromosomal DNA without any antibiotic resistance genes.