This report covers part of the assessment of a site in northern France polluted by industrial metal emissions. Our objectives were to assess the cadmium burden in children aged 8 to 11 years and look for factors that affected it. A cross-sectional population-based study took place in 1996-1997 and included 400 children (200 in the metal-polluted area). The geometric mean of the blood cadmium (CdB) levels of children living on the polluted site was 0.51 microg/L (95% CI=[0.49-0.53]); that of the nonexposed children was 0.47 microg/L (95% CI=[0.45-0.49]) (P<0.01). The geometric mean of the urinary cadmium (CdU) levels of children living on the polluted site was 1.16 microg/g creatinine (95% CI=[1.01-1.34]); that of the nonexposed children was 0.99 microg/g creatinine (95% CI=[0.86-1.14]) (P=0.10). Mean CdB was higher among children within 4 km of a zinc smelter and consuming > or =500 ml of tap water daily (x1.34; 95% CI=[1.14-1.51]) compared with children living more than 4 km away and consuming <500 ml of tap water daily. Independently of this association, consumption of fish and shellfish (> or =once a week vs. <once a week) and homegrown vegetables (> or =once daily vs. <once daily) was associated with higher mean CdB levels, respectively: x 1.14; 95% CI=[1.07-1.21] and x 1.25; 95% CI=[1.04-1.50]. Neither sex nor age was associated with CdB. Mean CdU was associated with paternal occupational cadmium exposure.