The potential neurotoxicity of PCBs was first recognized in 1968 when a number of Japanese people became ill after ingesting rice oil that was contaminated with PCBs during the manufacturing process (Yusho). Later a similar exposure occurred in Taiwan (YuCheng). Children born to Taiwanese mothers who consumed PCB-contaminated rice oil were followed and a number of developmental abnormalities, including lower body weight and height, higher activity levels, greater incidence of behavior problems, and lower IQ scores, were observed. However, interpretation of these findings is complicated by the fact that there did not appear to be any relationship between available indices of exposure and severity of effects, and by the fact that the PCBs to which the Taiwanese were exposed contained unusually high concentrations of dibenzofurans, which are many times more toxic than PCBs, and may have been responsible for some or all of the observed effects. Since the Yusho and YuCheng episodes, several studies have been initiated to study the neurobehavioral effects of exposure to the lower levels of PCBs present in the environment. The two studies published to date have yielded conflicting results. Jacobson, Jacobson, and colleagues reported that in utero PCB exposure was associated with decreased birth weight and head circumference, shorter gestation, and several adverse outcomes on the Brazelton Neonatal Assessment Battery. Later they reported that the body weight deficits associated with prenatal PCB exposure were still present at 5 months and 4 years of age. Deficits in memory function were observed at 7 months and 4 years. Rogan, Gladen, and colleagues did not find any evidence of decreased birth weight or head circumference. Nor did they find any evidence of deficits in memory function. However, they did observe some similar effects on the Brazelton Neonatal Assessment Battery. They also observed a small delay in psychomotor development in the most highly PCB-exposed children, but the effect did not persist beyond 2 years of age. A number of methodological concerns have been raised about the Jacobson study, including issues related to exposure assessment, sample selection, and control of potential confounding variables. However, it is not clear that these shortcomings can explain the discrepancies between their findings and those of Rogan and Gladen. Other possible explanations include differences in exposure levels or PCB congener patterns between the two cohorts, differences in sociodemographic variables between the two cohorts, or other problems inherent in trying to detect subtle neuropsychological deficits at exposure levels that are near the threshold for effects. Hopefully, several new studies that are currently underway will help to resolve the uncertainties regarding the risks of perinatal PCB exposure that have been created by the conflicting results of these early studies.