Polychlorinated biphenyls (PCBs), persistent environmental contaminants, may affect neurodevelopment of infants following prenatal exposure. A negative impact of prenatal PCB exposure on neurodevelopment was found in the Dusseldorf (Germany) cohort study (1993-2000). PCB levels of the sum of the three indicator congeners in breast milk were negatively associated with mental/motor development as assessed by the Bayley Scales of Infant Development (BSID) in infants. Since general exposure to PCB has decreased, a new birth cohort study was initiated in 2000 in the industrial city of Duisburg, which is located 30 km downstream from Dusseldorf on the River Rhine. A subgroup of the Duisburg birth cohort study was used to compare PCB exposure and developmental effects with results from the Dusseldorf cohort. The recruitment phase of the Duisburg cohort study occurred from 2000 to 2002. Mental and motor development was assessed by means of the BSID at the ages of 12 and 24 mo. Prenatal PCB exposure of newborns from Duisburg cohort was about two- to threefold lower than the Dusseldorf cohort. Although in the Dusseldorf birth cohort mental and motor development at ages 18 and 30 mo were negatively associated with PCB exposure, there was no association observed in the Duisburg study. Evidence indicates that exposure to PCB at current exposure levels no longer apparently impair neurodevelopment of infants.