The optimal staging of tumors would reflect their biology and patterns of spread, permit accurate prognostication, and facilitate therapeutic decision-making. The last revision of the International Federation of Obstetricians and Gynecologists (FIGO) staging of uterine corpus tumors was in 1988, and it represented the transition from a clinical to a surgico-pathologic system. With 20 years of experience, we can now review the accuracy, reproducibility, and utility of this system. Pathologists are in a unique position to study each of these characteristics, comment on their ability to apply the criteria in daily practice, and offer suggestions to further improve the FIGO system. This paper selectively reviews some of the more problematic aspects of the current FIGO system, including the following: the distinction of tumors confined to the endometrium from those which are superficially myoinvasive; the method and utility of histologic grading of endometrial adenocarcinoma; the utility and reproducibility of the diagnosis of cervical epithelial and stromal invasion; the striking heterogeneity within and among stage III A, B, and C tumors and their differing prognostic significance. It concludes with recommendations for changes in a future revision of the FIGO staging of endometrial carcinoma.