The risk of second primary cancer was evaluated in more than 25,000 women with cancer of the genital organs diagnosed between 1935 and 1982 in Connecticut. Significant excesses of subsequent cancers were observed following cancers of the cervix (35%, n = 656), uterine corpus (16%, n = 1,060), and ovary (58%, n = 366). When observed and expected second cancers of the female genital tract were excluded, these excesses became 40%, 30%, and 59% after cervix, uterine corpus, and ovary, respectively. Among women with either cancer of the cervix or uterine corpus, the risk of developing a second cancer rose with increasing duration of follow-up, reaching an excess of 61 and 34%, respectively, after 20 years. In contrast, among patients with ovarian cancer, the second cancer risk decreased over time to 41% after 10 years. Cancers related to smoking, i.e., oral cavity and pharynx, esophagus, and respiratory system, were notably increased among cervical cancer patients. The twofold to threefold risks observed for these second cancers are consistent with recent evidence linking cervical cancer to cigarette smoking and seem too large to be artifacts of confounding by low socioeconomic status. An increased incidence of second cancer of the abdominal organs (colon, rectum, kidney, bladder, ovaries) was generally observed for each gynecologic site. However, only rectal cancer was consistently linked with radiation treatment for the first primary cancer. Leukemia occurred in excess after cancers of the uterine corpus and ovary, but not after cervical cancer. The predominant cell type was acute nonlymphocytic leukemia, and the excess was associated with radiotherapy for uterine corpus cancer and with chemotherapy for ovarian cancer. Cancers of the breast and colon were increased following uterine corpus and ovarian cancer and vice versa, which supports the notion that these sites share a common etiology, perhaps related to dietary or hormonal factors. Cervical cancer patients experienced a deficit of subsequent breast cancer, possibly due to ovarian removal or ablation by radiation. Investigators need to explore further the association between the smoking-related cancer sites and cervical cancer, to clarify the role of radiotherapy and chemotherapy in relation to excess cancers, and to define more fully the etiologic factors that link cancers of the breast, colon, uterine corpus, and ovary.