In 1988 and 1989, cotton yields were greater with full season protection from cotton insects than where thresholds were used to trigger the application of insecticides. In 1990, plant mapping techniques were used to measure damage to fruiting structures under different levels of insecticidal protection. First position fruit retentions were compared at nodes 6-8, 9-11, 12-14 and 15-17, but the data were inconclusive. The highest yield was obtained where no insecticides were applied for F₂ generation bollworm/budworm and the cotton was afforded full protection from F₃ generation larvae. The reverse strategy produced a significantly lower yield that was no different from the untreated cotton or full-season protection.