Ovarian cancer has received national attention as a highly virulent disease. Its lack of early warning symptoms and the failure to develop highly sensitive screening tests have led some physicians to recommend prophylactic oophorectomies to women with relatives who have had ovarian cancer. Others have recommended routine screening of otherwise normal women for CA 125, a circulating tumor marker, and ultrasound examinations. Each of these techniques is associated with substantial false-positive rates that could lead to unnecessary surgery. A review of epidemiologic data suggests that familial ovarian cancer kindreds are rare, but women with first-degree relatives who have had ovarian cancer have a significant risk themselves for developing ovarian cancer. In addition, women with a great number of ovulatory cycles are at an increased risk for the disease. Circulating tumor markers are frequently elevated in women with advanced ovarian cancer, but their value in early detection of ovarian cancer has yet to be established. Advances in endovaginal ultrasound and color Doppler flow technology have significantly improved our ability to assess pelvic organs. This article presents the background, rationale, and structure of the Yale Early Detection Program for ovarian cancer, whose goals are to identify the best techniques for diagnosing ovarian cancer in an early stage, to determine the frequency with which such tests should be employed, to assess false-positive results, and to identify women who might benefit from prophylactic oophorectomies.