Six quality-control laboratories in 4 countries independently bioassayed aliquots of a flowable formulation of Bacillus thuringiensis var. israelensis (B.t.i.) against the international standard powder IPS-82. All laboratories substantially followed World Health Organization or U.S. Department of Agriculture standard protocols. Significant differences were found in resulting potency values between laboratories. Factors that may have influenced results, such as age, stage, and strain of larvae used, amount and type of food provided to larvae, and processing of samples were examined. Use of different rearing temperatures, different strains of Aedes aegypti L., or late 3rd instars vs. the recommended early 4th instars did not explain the inconsistencies. The slope of the dose-response curve of the IPS-82 powder was influenced by particle size, which varied with the nature and duration of sample homogenization. Laboratories using low-intensity processing obtained a greater slope in the dose-response curve for the flowable product than for the powder standard. The type and quantity of food provided to larvae affected susceptibility. Larvae fed an excess of protein-rich food became 4th instars in 3 days and were less susceptible to B.t.i. than those fed smaller quantities of carbohydrate-rich food that became 4th instars in 5-7 days. Overall, deviations from standard protocols with regard to larval stage, holding temperature, and lighting regime may not be as important as differences in sample processing and pretest rearing conditions. The need to improve standardization in these areas, which are not clearly specified in current protocols, is discussed.