Population cage experiments were employed to detect variability in fitness among Cry2Ab resistant and Cry2Ab susceptible genotypes of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae). In two experiments, reciprocal crosses between a Cry2Ab resistant colony (SP15) and a susceptible colony (GR) established populations where the frequency of the allele that conferred resistance was 0.5. Experimental populations were then maintained without exposure to Cry toxins. At the F₂ generation and on later occasions, the pooled egg output from each population was sampled, and emerging neonate larvae were screened to monitor the frequency of the resistant allele. Resistance is recessive so homozygous resistant insects could be readily identified as they are the only genotype to survive and grow when exposed to a discriminating concentration of Cry2Ab toxin. Assuming Hardy—Weinberg equilibrium after the F₁ generation, and the persistence of a 1:1 ratio of resistant and susceptible alleles, one quarter of the populations should be resistant. The populations in the first and second experiment were monitored for five and nine generations, respectively. The cumulative impact of any fitness costs associated with resistant genotypes was expected to result in a decline in the frequency of resistant homozygotes. In both experiments, there was no significant decline in resistance frequencies, and thus the Cry2Ab form of resistance does not seem to exhibit marked fitness costs under laboratory conditions.