To estimate the incidence of MS and its relation to latitude in two ongoing prospective studies of US women. A higher incidence of MS has been found in northern areas compared with southern areas of the United States and other countries, but the attenuation of this gradient in Europe in the last few decades and the consideration of ethnic factors have led some authors to question the existence of a strong association between MS and latitude. The authors identified new cases of MS among participants in the Nurses' Health Study (NHS), which took place between 1976 and 1994, and in the Nurses' Health Study II (NHS II), which took place between 1989 and 1995. The NHS included women born between 1920 and 1946, and the NHS II included women born between 1947 and 1964. The incidence of MS among NHS participants (181 definite/probable patients) increased significantly with latitude (p = 0.03, trend). Adjusted rate ratios were 3.5 (95% CI, 1.1, 11.3) for the north and 2.7 (95% CI, 0.8, 8.9) for the middle tiers relative to the southern tier. Among NHS II women (131 definite/probable patients), no association between latitude and MS was found (p = 0.89, trend). Adjusted rate ratios were 0.8 (95% CI, 0.4, 1.6) for the northern areas and 0.9 (95%, 0.4, 1.8) for the middle areas, relative to the southern areas. The association between latitude and risk of MS in the United States was corroborated, but there was an attenuation of the north-south gradient over time. If confirmed, this finding could provide new clues to identifying environmental causes of the disease.