Instar specific survival and feeding preference studies of the tobacco budworm, Heliothis virescens (F.) and bollworm, Helicoverpa zea (Boddie) on nontransgenic and transgenic Bt cottons, Gossypium hirsutum L. expressing an insecticidal protein from Bacillus thuringiensis Berliner subsp. kurstaki were conducted in 1992-93 at the Texas A&M University Agricultural Research and Extension Center, Corpus Christi, TX. The objectives of the study were: (1) to determine and compare the percent survival of each instar of bollworm and tobacco budworm when fed on Bt flower buds and non-Bt flower buds; and (2) to determine if bollworm and tobacco budworm third-instars show a feeding preference for non-Bt cotton or Bt cotton when feeding on mixtures of the two cotton plants.

In the instar specific survival study a cohort of larvae (150 larvae/instar) was reared on diet until the appropriate instar for the test was reached (i.e., first-, second-, third-, fourth-, or fifth-instar). The larvae were placed on Bt or non-Bt flower buds to simulate larvae feeding on that plant genotype the remainder of their lives. Bollworm and tobacco budworm larvae that fed on Bt flower buds as first-, second-, third-, and fourth-instars failed to pupate. However a large percentage of the fifth-instars of both species survived to become adults on the Bt flower buds. Significantly more bollworm fifth-instars pupated and survived to the adult stage when fed on the non-Bt cotton (76.0% and 51.3%, respectively) compared with those fed on the Bt cotton (46.6% and 27.1%, respectively). In contrast, significant differences were not found in the number of tobacco budworm fifth-instars that pupated and emerged as moths when fed on non-Bt flower buds (73.3% and 41.3%, respectively) compared with those fed on Bt flower buds (75.3% and 48.0%, respectively).

In the feeding preference study five different mixtures (i.e., ratios) of Bt:non-Bt plants were evaluated. Ratios of Bt to non-Bt plants in the mixture were 100:0, 75:25, 50:50, 25:75 and 0:100, respectively. Plants for each mixture were clustered together in a cage. Significantly fewer larvae were found on the Bt cotton plants compared with non-Bt plants after 24 hours of feeding in the mixtures. The number of tobacco budworm injured flower buds and bolls were lower in all mixtures compared with the pure stand of non-Bt, 0:100 mixture.

This study showed that Bt cotton increased the mortality of bollworm and tobacco budworm larvae but that the fifth-instar was capable of feeding, pupating and becoming an adult on Bt cotton plants. It was also demonstrated that the bollworm and tobacco budworm larvae showed a preference for non-Bt cotton plants compared with Bt cotton plants. However, this apparent preference was probably a result of larval intoxication, reduced feeding and increased plant abandonment on the Bt plants compared with the non-Bt plants, rather than a classical behavioral preference. Additional studies under field conditions are needed of bollworm and tobacco budworm injury, and cotton yields for mixtures containing less than 25% non-Bt plants.