From 1953 to 1972, 149 patients at Centre Leon Bérard, were treated by pelvic exenteration for carcinoma of the cervix. A review of the literature and of our cases showed that the mortality rate varied between 12 and 38 per cent according to the authors. In our experience, the final cause of death has been essentially pelvic reccurences within 18 months of the operation. For Ketcham, the patients died principally from metastases. To illustrate these results, criteria for patient selection for pelvic exenteration are outlined, with some suggestions for operative and post-operative management. The pre-operative medical status, the roentgenographic studies and finally exploratory laparotomy eliminated all but a very small number of patients, 15 to 20 per cent. Pelvic exenteration is appreciably beneficial only for this small group. Indeed, the exenteration is acceptable only when it does not result in excessive mutilation disproportinate with the chances of survival. It is conceivable only as curative treatment requiring a radical loco-regional excision, not only for the involved viscerae but also for the pelvic lymph nodes.