Melanoma has been increasing in white populations. Incidence rates rise steeply in women until about age 50, suggesting oestrogen as a possible risk factor. Oestrogens can increase melanocyte count and melanin content and cause hyperpigmentation of the skin. We examined prospectively the association between oral contraceptive (OC) use and diagnoses of superficial spreading and nodular melanoma among 183,693 premenopausal white women in the Nurses' Health Study (NHS) and the Nurses' Health Study II (NHS II) cohorts. One hundred and forty six cases were confirmed in NHS during follow-up from 1976 to 1994, and 106 cases were confirmed in NHS II from 1989 to 1995. Skin reaction to sun exposure, sunburn history, mole counts, hair colour, family history of melanoma, parity, height and body mass index were also assessed and included in logistic regression models. A significant twofold increase in risk of melanoma (relative risk (RR) = 2.0, 95% confidence interval (CI) 1.2-3.4) was observed among current OC users compared to never users. Risk was further increased among current users with 10 or more years of use (RR = 3.4, 95% CI 1.7-7.0). Risk did not appear elevated among past OC users, even among those with longer durations of use, and risk did not decline linearly with time since last use. In conclusion, risk of premenopausal melanoma may be increased among women who are current OC users, particularly among those with longer durations of use. Further research is needed to determine whether low-dose oestrogen pills in particular are associated with an increase in risk and to describe possible interactions between OC use and sun exposure or other risk factors for melanoma.