Chlamydia trachomatis was isolated from genital specimens from 21 (4.9%) of 431 female college students. Antibody to C. trachomatis was found in the genital secretions of 52 (11.9%) of 437 women. Multiple logistic regression analysis showed race, number of sexual partners, and use of barrier methods of contraception to be predictive of infection with C. trachomatis. Logistic regression analysis found race, number of sexual partners, use of barrier methods of contraception, and presence of cervical erythema to be predictive of local chlamydial antibody. White participants were infected less often (12 of 388 (3.1%)) than black participants (9 of 43 (20.9%)) (p less than 0.001) and were less likely to have local chlamydial antibody. None of the sexually inexperienced women were infected or had local antibody. Among the sexually experienced women, chlamydial infection and local chlamydial antibody increased with increasing number of sexual partners only for women who were not using barrier methods of contraception. Sexually experienced women who used barrier methods of contraception (condom, diaphragm) were less likely to be infected (one of 105 (1.0%)) than were sexually experienced women who used other contraceptive measures or who did not use contraception (20 of 276 (7.2%)) (p = 0.031). Women who used barrier methods of contraception also were less likely to have local chlamydial antibody. Women with cervical erythema were more likely to have local chlamydial antibody (4 of 11 (36.4%)) than women without cervical erythema (48 of 426 (11.3%)). Vaginal colonization with other sexually transmitted microorganisms (Mycoplasma hominis, Ureaplasma urealyticum, Trichomonas vaginalis) was noted more often among women with chlamydial infection than among uninfected women.