The associations between early dentine lead levels measured at the age of 6-8 years and educational outcomes measured at 18 years were examined in a birth cohort of 1265 New Zealand children. Analyses showed significant (p < .005) dose/response relationships between early dentine lead levels and later outcomes: at age 18 children with early elevated lead levels had poorer reading abilities, had more often left school early, had more often left school without qualifications, and had lower levels of success in school examinations. These associations persisted after statistical control for a range of social and familial confounding factors. A number of potential threats to the validity of the findings are examined, including sample selection bias, statistical undercontrol of covariates, and errors of measurement. It is concluded that the findings are consistent with the view that early mildly elevated lead levels have modest but detectable effects on individual achievement, with these effects extending to late adolescence.