Our purpose in this report is to evaluate scientifically that body of literature relating the effects of prenatal and postnatal exposure to polychlorinated biphenyls (PCBs) upon neurobehavioral, health-related, and cognitive deficits in neonates, developing infants, children, and adults. The data derive from seven cohorts: six cohorts of mothers and their offspring who were exposed to PCBs, and one composed of adults who were also exposed to PCBs. We applied six fundamental sets of scientific criteria to evaluate the claims of each study deriving from each cohort: (1) The level of reliability and validity of assessment instruments; (2) The appropriateness of the reliability assessments made on study variables; (3) The adequacy of controlling for chance findings when multiple comparisons were made; (4) The necessity to distinguish between clinically meaningful and statistically significant results; (5) The level of success in controlling for fundamental confounding variables, and; (6) The extent to which studies were longitudinally designed and appropriately analyzed. We conclude that the studies of these cohorts failed to meet reasonable scientific standards. Studies deriving from the more recent cohorts used better research designs than those based on the early (about 1980) cohorts in Michigan and North Carolina, and produced occasional interesting hypotheses that need to be tested on new cohorts. However, the numerous claims and conclusions regarding the deleterious effects of maternal PCB exposure upon subsequent cognitive, neuropsychological, and behavioral development in their offspring, whether the exposure is in utero or postnatal, are not tenable; nor are the claims that PCBs affect the neuropsychological functioning of adults. (Contains 3 tables.)