Objective. To test the hypothesis that elevated lead in soil is positi
vely correlated with blood lead (BPb) levels in children in an urban p
opulation surrounding a closed lead smelter, a US Environmental Protec
tion Agency Superfund clean-up site was surveyed. Method. A total of 8
27 volunteers including 490 children under 6 years of age participated
. A questionnaire was administered. Blood lead was determined as was l
ead content of samples of house dust, soil, paint, and water of the pa
rticipants' homes. Results. The arithmetic mean venous BPb in 490 chil
dren between 6 and 72 months of age was 6.9 mu g/dL (0.33 mu mol/L) ra
nge 0.7 to 40.2 mu g/dL (0.03 to 1.94 mu mol/L). The BPb of 78 (16%) c
hildren in this group was greater than or equal to 10 mu g/dL (0.48 mu
mol/L). Based on multiple regression modeling, lead in house dust acc
ounted for 18% of the variance in BPb. Lead in paint together with the
condition of the house were the main contributors to the dust lead va
riance (26%) with soil lead accounting for an additional 6%. Lead in p
aint alone accounted for 3% of the BPb variance. Lead in paint togethe
r with the condition of the house accounted for 12% of BPb variance, a
nd lead in soil accounted for an additional 3%. Factors other than env
ironmental lead such as education of parents, household income, and be
havior were associated with BPb levels. Conclusions. The mean BPb in c
hildren was below the present level of concern of the Centers for Dise
ase Control and Prevention. Children with BPb of greater than or equal
to 10 mu g/L (0.48 mu mol/L tended to live in poorly maintained older
houses. Based on these findings lead in soil and paint in well-mainta
ined homes contributed little to the lead exposure of children.