From 8 to 18 commercial cabbage fields in upstate New York were sampled weekly from 1978 through 1980. Peak populations of lepidopterous larvae fluctuated markedly during the 3-year period, and these fluctuations could not be attributed solely to the use of insecticides. In all 3 years, larvae of imported cabbageworm and diamondback moth were detected earlier than cabbage looper, and initial infestation was not a direct function of planting date. At harvest, cabbage looper was always the most abundant species. This indicates the need for sampling individual fields for treatment decisions.