Strips coated with conidia of Metarhizium anisopliae (Metschinkoff; Deuteromycetes: Hyphomycetes) to control the parasitic mite, Varroa destructor (Anderson and Trueman) in colonies of honey bees, Apis mellifera (Hymenoptera: Apidae) were compared against the miticide, tau-fluvalinate (Apistan) in field trials in Texas and Florida (USA). Apistan and the fungal treatments resulted in successful control of mite populations in both locations. At the end of the 42-day period of the experiment in Texas, the number of mites per bee was reduced by 69-fold in bee hives treated with Apistan and 25-fold in hives treated with the fungus; however mite infestations increased by 1.3-fold in the control bee hives. Similarly, the number of mites in sealed brood was 13-fold and 3.6-fold higher in the control bee hives than in those treated with Apistan and with the fungus, respectively. Like the miticide Apistan, the fungal treatments provided a significant reduction of mite populations at the end of the experimental period. The data from the broodless colonies treated with the fungus indicated that optimum mite control could be achieved when no brood is being produced, or when brood production is low, such as in the early spring or late fall. In established colonies in Florida, honey bee colony development did not increase under either Apistan or fungal treatments at the end of the experimental period, suggesting that other factors (queen health, food source, food availability) play some major role in the growth of bee colonies. Overall, microbial control of Varroa mites with fungal pathogens could be a useful component of an integrated pest management program for the honey bee industry.