Third instar tobacco budworm from west-central Mississippi and the Lower Rio Grande Valley of south Texas were tested for response to treatment with cypermethrin. Mortalities were observed 24, 48, 72, and 96 h after treatment. Larvae died slowly; the ultimate response of some could not be determined even after 96 h. Estimated LD(50)s based on the 24, 48, 72, and 96 h observations decreased asymptotically; asymptotic LD50 estimates may provide a means of standardizing LD(50) estimates. Because the fate of most larvae could not be determined unambiguously until at least 96 h after treatment, LD(50) estimates based on the 96 h mortality observation were used for all comparisons. The Texas and Mississippi strains of tobacco budworm were 9 and 57 times more tolerant than our susceptible colony, respectively. We examined the genetic basis of increased tolerance by crossing each field population with the susceptible colony to produce Fls, F2s, and all possible backcrosses. Our ability to draw conclusions about the genetic basis of increased tolerance in these populations was limited severely by the modest increases in tolerance in these populations and the highly variable intrapopulation response to cypermethrin that is characteristic of tobacco budworm.