Within the last ten years, there has been a large increase in the number of commercial cotton cultivars, Gossypiun hirsutum L., available for production. Most cotton cultivars available in the past were similar morphologically and the more recently developed cultivars set bolls at a different time period than older cultivars. Breeders have inadvertently been selecting plant types with smaller bolls and with fewer and smaller seed per boll. Due to the inverse relationship existing between number of bolls per plant and boll weight, the objective of the research was to determine if the economic injury level could be varied based on boll weight for Tobacco Bud Worms (TBW) (Heliothis virescen Fabricius) and Boll Worms (BW) (Heliothis zea Boddie). The experiment was conducted at the Macon Ridge Research Station, in 1995, comparing six cotton varieties; Stoneville LA 887 (STV LA887), Stoneville 132 (STV 132), Chembred 830 (CB 830), Hartz 1215 (H1215), Delta and Pine Land 5409 (DPL 5409), and Delta and Pine Land (DPL 5415). The experimental design was a split plot with three replications: main plot = insecticide treatment, control and untreated; split plot = varieties. For the control insecticide treatment, all economically important insects were controlled all season until 40% open boll was achieved. For the untreated insecticide treatment all economicily important insects were controlled through out the season. When the crop reached mid-bloom for untreated insecticide treatment, control of TBW/BW was discontinued to inflict mature fruit damage. Results from the analysis of variance showed main plot effects, seed cotton yield, predicted seed cotton yield, lint yield, predicted lint yield, open bolls per 2 meter, bolls per acre, and total bolls open at 90% maturity were significant at alpha =.05. Results from the analysis of variance showed split plot effects, seed cotton yield, boll weight, and total bolls open at 90% maturity were significant at alpha =.05. Numerically , the medium (STV 132 and H1215) and heavy (STV LA887 and CB 830) boll weight varieties had higher seed cotton yields, open bolls per 2 meters, and predicted seed cotton yield than the light boll weight varieties (DPL 5415 and DPL 5409). The lighter boll weight varieties also had a higher number of worm damaged bolls per 2 meters than did the heavier boll weight varieties. The regression analysis indicated when boll weight was regressed against seed cotton yield, it had a positive relationship to yield; y = 16.22 + 238.2x₁, R² = .15, C.V. =26.5. The full model of seed cotton yield = open bolls per 2 meters + boll weight + worm damaged open bolls per 2 meters had a fit of R² = .75 , and C.V. = 15, (y = -756 + 154.77x₁ + 139.05x₂ + -11.28x₃).