Although Mexico substantially reduced use of leaded gasoline during the 1990s, lead-glazed pottery remains a significant source of population exposure. Most previous studies of lead in nonoccupationally exposed groups in Mexico have been conducted in the Mexico City metropolitan area. Oaxaca, a poor southern state of Mexico, has a centuries-old tradition of use of low temperature lead-glazed ceramic ware manufactured mainly by small family businesses. We measured blood lead levels in 220 8-10-y-old children (i.e., not from pottery-making families) who were students in the innercity of Oaxaca and in the mothers of all children. The geometric mean blood lead level of the children was 10.5 microg/dl (+7.0/-4.3 microg/dl standard deviation; range = 1.3-35.5 microg/dl). The corresponding mean value for the mothers was 13.4 (+9.0/-5.4 microg/dl standard deviation; range = 2.8-45.3 microg/dl). We used cutoffs that were greater than or equal to 10 microg/dl, 20 microg/dl, and 30 microg/dl, and we determined that 54.9%, 10.3%, and 3.0% of the children were at or above the respective criteria. We accounted for 25.2% of the variance in blood lead levels of the children, using maternal responses to a questionnaire that assessed possible lead sources in a linear multiple-regression model. The most important factors related to lead levels were family use of lead-glazed pottery, use of animal fat in cooking, and family income. The addition of maternal blood lead level to the model increased accounted variance in blood lead to 48.0%. In logistic-regression modeling of children's blood lead levels, we used a cutoff of greater than or equal to 10 microg/dl, and we found that use of lead-glazed pottery was the most important of all questionnaire items that were predictive of blood lead levels (odds ratio = 2.98). In Oaxaca, as is the case elsewhere in Mexico, lead-glazed ceramic ware remains a significant risk factor for elevated blood lead levels in children.