To compare the efficiency of male-to-female and female-to-male sexual transmission of human immunodeficiency virus (HIV), we studied 524 female partners of HIV-infected men and 206 male partners of HIV-infected women in 16 Italian clinical centers. All of the partners had had a sexual relationship with the index case lasting for at least 6 months and presented no other risk factor than sexual exposure to the HIV-infected partner. Among the 730 couples, 24% of the female partners were HIV positive, in comparison with 10% of the male partners. Using logistic regression analysis, including gender and controlling for condom use, frequency of intercourse, anal sex, partner's CD4+ cell count and clinical stage, sexually transmitted diseases, genital infections, and contraceptive use, we found that the efficiency of male-to-female transmission was 2.3 (95% confidence interval = 1.1-4.8) times greater than that of female-to-male transmission. Between-gender differences in the contact surfaces and the intensity of exposure to HIV during sexual intercourse are possible reasons for the difference in efficiency of transmission.