A three-generation bioassay was developed to determine the response curves of the western spruce budworm,Choristoneura occidentalisFreeman, to compounds that occur in the foliage of Douglas-fir,Pseudotsuga menziesii(Mirb.) Franco, one of its principal hosts. Insects from a nondiapausing laboratory colony ofC. occidentaliswere reared on artificial diets containing defined levels of various nutrients and allelochemicals. An experimental base diet (nutritionally similar to Douglas-fir foliage in concentrations of nitrogen, sugars, and minerals) was modified to test responses. Insects spent their entire larval period, for three consecutive generations, feeding on the treatment diets. Responses measured include survival rates for all life stages, pupal masses, and larval development times and growth rates. Data on survival and fecundity (predicted from female pupal masses) were used to estimate population growth over three generations for the levels of the compound(s) tested. The tremendous variation in the survival and reproduction ofC. occidentalisamong 29 replications of the bioassay on the experimental base diet emphasizes the importance of replicating whole experiments for this type of bioassay. The great variability that occurred in these feeding studies underscores the critical importance of using large sample sizes and proper controls on each experiment to ensure that conclusions made from feeding bioassays are reliable. Average performance on the base diet was: 68% of the first instars survived to the late larval stage, 57% of the initial cohort survived to the pupal stage, 40% of the cohort survived to the moth stage, 45% of the egg masses produced had viable eggs, mean fresh pupal mass was 97 mg for females and 66 mg for males, mean larval development time was 44 d for females and 40 d for males, and larval growth rate was 2.3 mg/d for females and 1.7 mg/d for males.