Lead and the relationship between maternal and child intelligence.

Using regression analysis, we show that the IQs of children with elevated levels of dentine lead (greater than 20 parts per million) are below those expected, based on their mothers' IQs. Moreover, the amount by which a child's IQ falls short of the expected value increases with increasing levels of dentine lead in what may be a nonlinear fashion. Although lead level contributed nothing to the prediction of IQ for children with low levels of dentine lead (less than 10 parts per million), it rivaled maternal IQ in importance as a predictor in the group with elevated lead values. Thus for schoolchildren with lead burdens in the highest decile of the distribution for the urban area we sampled, the usual relationship between maternal and child IQ appears to be disrupted in a manner systematically related to lead levels in dentine.