Lead-contaminated water, soil, and paint have been recognized as potential sources of children's lead exposure for decades, but their contributions to lead intake among urban children remain poorly defined. This analysis was undertaken to estimate the relationship of environmental lead exposures to lead intake among a random sample of urban children, adjusted for exposure to lead-contaminated house dust. Analyses of 183 urban children enrolled in a random sample, cross sectional study were conducted. Children's blood and multiple measures of household dust, water, soil, and paint were analyzed for lead, and interviews were conducted to ascertain risk factors for childhood lead exposure. Environmental sources of lead, including house-dust, soil lead, and water lead, were independently associated with children's blood lead levels. In contrast, paint lead levels did not have a significant effect on blood lead levels after adjusting for other environmental exposures. An increase in water lead concentration from background levels to 0.015 mg/L, the current EPA water lead standard, was associated with an increase of 13.7% in the percentage of children estimated to have a blood lead concentration exceeding 10 micrograms/dL; increasing soil lead concentration from background to 400 micrograms/g was estimated to produce an increase of 11.6% in the percentage of children estimated to have a blood lead level exceeding 10 micrograms/dL, and increasing dust lead loading from background to 200 micrograms/ft2 is estimated to produce an increase of 23.3% in the percentage of children estimated to have a blood lead level exceeding 10 micrograms/dL. These data support the promulgation of health-based standards for lead-contaminated dust and soil and the progressive lowering of standards for lead-contaminated water as the definition of undue lead exposure has been modified.