A case-control interview study was used to evaluate the relationship between breast cancer risk and residential proximity to industrial facilities and traffic for pre- and postmenopausal women in Nassau and Suffolk Counties on Long Island, New York. A geographic information system was used to assign industry and traffic counts to 1-km2 grid cells (5-kM2 grid cells for traffic) and to assign potential exposure values to study subjects, based on 20-y residential histories. A significantly elevated risk of breast cancer was observed among postmenopausal subjects who were ever potentially exposed to chemical facilities (Nassau County adjusted odds ratio [OR] = 1.61, 95% confidence interval [95% CI] = 1.06-2.43; Suffolk County adjusted OR = 1.58, 95% CI = 0.71-3.51. This elevated risk, however, was not observed among premenopausal subjects. Risk increased for post-menopausal subjects as number of chemical facilities increased from one (adjusted OR = 1.45, 95% CI = 0.93-2.25) to two or more (adjusted OR = 3.47, 95% CI = 1.06-11.38). Crude and adjusted ORs for high traffic density were elevated among Nassau--but not Suffolk--county subjects and were not significant statistically.