INSECTICIDE RESISTANCE AND RESISTANCE MANAGEMENT Oviposition on and Mining in Bolls of Bt and Non-Bt Cotton by Resistant and Susceptible Pink Bollworm (Lepidoptera: Gelechiidae) YONG-BIAO LIU, 1 BRUCE E. TABASHNIK, TIMOTHY J. DENNEHY, YVES CARRIE ` RE, MARIA A. SIMS, AND SUSAN K. MEYER Department of Entomology, University of Arizona, Tucson, AZ 85721 J. Econ. Entomol. 95(1): 143Ð148 (2002) ABSTRACT Transgenic cotton that produces insecticidal crystal protein Cry1Ac of Bacillus thu- ringiensis (Bt) has been effective in controlling pink bollworm, Pectinophora gossypiella (Saunders). We compared responses to bolls of Bt cotton and non-Bt cotton by adult females and neonates from susceptible and Cry1Ac-resistant strains of pink bollworm. In choice tests on caged cotton plants in the greenhouse, neither susceptible nor resistant females laid fewer eggs on Bt cotton bolls than on non-Bt cotton bolls, indicating that the Bt toxin did not deter oviposition. Multiple regression revealed that the number of eggs laid per boll was negatively associated with boll age and positively associated with boll diameter. Females also laid more eggs per boll on plants with more bolls. The distribution of eggs among bolls of Bt cotton and non-Bt cotton was clumped, indicating that boll quality rather than avoidance of previously laid eggs was a primary factor in oviposition preference. Parallel to the results from oviposition experiments, in laboratory no-choice tests with 10 neonates per boll, the number of entrance holes per boll did not differ between Bt cotton and non-Bt cotton for susceptible and resistant neonates. Also, like females, neonates preferred younger bolls and larger bolls. Thus, acceptance of bolls by females for oviposition and by neonates for mining was affected by boll age and diameter, but not by Bt toxin in bolls. The lack of discrimination between Bt and non-Bt cotton bolls by pink bollworm from susceptible and resistant strains indicates that oviposition and mining initiation are independent of susceptibility to Cry1Ac. KEY WORDS Pectinophora gossypiella, oviposition, neonate, Bacillus thuringiensis, host preference, resistance management MILLIONS OF HECTARES of transgenic crops that produce insecticidal proteins from the bacterium Bacillus thu- ringiensis (Bt) are being grown to control insect pests (James 2000). Such Bt crops can promote increased yield and decreased insecticide use (Fernandez- Cornejo and McBride 2000). Although the potential beneÞts of Bt crops are great, evolution of resistance by pests could cut short their efÞcacy (Tabashnik 1994a, Gould 1998, Frutos et al. 1999). Diamondback moth has evolved resistance in the Þeld to sprays of Bt toxins and at least 10 other pests have evolved resis- tance in the laboratory to Bt toxins (Tabashnik 1994a, Schnepf et al. 1998, Frutos et al. 1999). The beneÞts of a Bt crop and the risk of resistance are exempliÞed by the interactions between Bt cotton and pink bollworm, Pectinophora gossypiella (Saun- ders) (Lepidoptera. Gelechidae), a worldwide pest of cotton (Ingram 1994). So far, transgenic cotton that produces Bt toxin Cry1Ac has been very effective against pink bollworm (Henneberry and Jech 2000, Tabashnik et al. 2000a). In Arizona, Bt cotton has helped to greatly reduce insecticide use (Carrie ` re et al. 2001). It has accounted for 50% of cotton grown statewide since 1997, with much higher levels in some areas (Carrie ` re et al. 2001). This widespread adoption increases the risk of resistance to Bt cotton. Labora- tory selection readily produced 100-fold pink boll- worm resistance to Cry1Ac and increased survival on Bt cotton in a strain recently derived from the Þeld and in a long-term laboratory strain (Liu et al. 1999, 2001a, 2001b; Patin et al. 1999; Tabashnik et al. 2000a, 2000b). To delay evolution of resistance by pink bollworm and other pests targeted by Bt cotton, the U.S. Envi- ronmental Protection Agency requires refuges of cot- ton that does not produce Cry1Ac (non-Bt cotton). The potential effectiveness of the refuge strategy has been illustrated mainly with mathematical models (Tabashnik 1994b, Gould 1998, Carrie ` re and Tabash- nik 2001), whereas experimental data have been scarce (Liu and Tabashnik 1997, Shelton et al. 2000, Tang et al. 2001). In principle, refuges provide many susceptible adults to mate with rare resistant adults emerging from Bt cotton Þelds. Ideally, the hybrid offspring produced by matings between resistant and susceptible adults are killed by Bt cotton. Thus, two 1 Current address: Agricultural Research Station, USDA-ARS, 1636 East Alisal Street, Salinas, CA 93905 (e-mail: yb_liu@yahoo.com). 0022-0493/02/0143Ð0148$02.00/0  2002 Entomological Society of America