While the protective nature of parity with respect to ovarian cancer has been well documented, whether a history of incomplete pregnancy affects ovarian cancer risk is uncertain. Data collected from 739 epithelial ovarian cancer cases and 1,313 community controls in the Delaware Valley from 1994 to 1998 were used to evaluate the relation between gestational length and timing of first induced or spontaneous abortion and ovarian cancer risk. Incomplete pregnancy was not associated with ovarian cancer among nulliparous women or among ever-pregnant women either before or after adjustment for relevant confounders (for nulliparous women, odds ratio (OR) = 1.12, 95% confidence interval (CI): 0.66, 1.89; for ever-pregnant women, OR = 0.95, 95% CI: 0.76, 1.18). Among unigravid women, one full-term pregnancy was more protective than an incomplete pregnancy (adjusted OR = 0.29, 95% CI: 0.15, 0.57). These results were independent of the type of pregnancy loss. Among ever-pregnant women, a spontaneous abortion before a first birth provided significant protection (adjusted OR = 0.47, 95% CI: 0.30, 0.75), while no significant effect was found for an induced abortion prior to a first birth (adjusted OR = 0.80, 95% CI: 0.44, 1.47). These data do not support an independent association between incomplete pregnancies, either spontaneous or induced, and ovarian cancer risk.