OBJECTIVE To describe the determinants of blood lead concentration in children with long term environmental exposure to lead. METHODS Prospective cohort study. METHODS The lead smelting town of Port Pirie, South Australia, and surrounding townships. METHODS 326 children born in and around Port Pirie, 1979-1982, followed up until age 11-13 years in 1993-1994. METHODS Blood lead concentrations assessed at birth and at multiple ages up to 11-13 years; average lifetime blood lead concentration. RESULTS Mean blood lead concentration rose sharply over the ages 6 to 15 months, reached a maximum around 2 years of age, and declined steadily as the children grew older. There was no difference in blood lead concentration between boys and girls until they reached the age of 11-13 years, when mean blood lead concentration in boys (8.4 micrograms/dL [0.41 mumol/L]) was slightly higher than in girls (7.5 micrograms/dL [0.36 mumol/L]). Residential area and father's employment site were the two variables most strongly predictive of a child's blood lead concentration at the end of primary school. Poorer-quality home environment was also found to be an independent contributor to blood lead concentrations. CONCLUSIONS Age-related factors, and possibly recent concerted efforts to decrease entry or re-entrainment of lead into the environment at Port Pirie, have resulted in most children in our study having blood lead concentrations below 10 micrograms/dL (0.48 mumol/L) at the end of their primary school years. Lead exposure during a child's early years remains an important contributor to average lifetime exposure.