This case-control study investigated the potential association between ambient levels of carbon monoxide in a pregnant woman's neighborhood of residence and her chance of delivering a low birth weight infant. Low birth weight infants and normal birth weight infants were contrasted with respect to ambient levels of CO during the 3 months prior to delivery in the neighborhoods where their mothers lived at birth. After adjustment for the confounding effects of maternal race and education, there was no association between higher CO exposure and higher odds of low birth weight. These data do not support a strong association between maternal exposure to neighborhood CO during pregnancy and odds of delivering a low birth weight infant. Further investigation of the effects of CO exposure on birth weight, with direct measurement of total CO exposure, is needed.