The relationship between dentine lead levels and measures of cognitive ability was examined for a birth cohort of New Zealand children studied until the age of 9. There were small, consistent and stable correlations between dentine lead measures and all measures of cognitive ability including intelligence, word recognition and teacher ratings of school performance. After adjustment for the effects of confounding covariates, sample selection factors and possible reverse causal effects, the correlations between intelligence and dentine lead levels became non-significant. However, small but statistically significant correlations persisted between dentine lead values and all measures of school performance after adjustment for sources of confounding. It is concluded that the weight of the evidence from this analysis favours the hypothesis that low level lead exposure may have deleterious effects on levels of achievement in children.