The efficacy of perioperative antimicrobial prophylaxis in preventing infection after vaginal hysterectomy has been demonstrated. In this study 260 patients undergoing vaginal hysterectomy were randomly assigned at five centers to receive 1 gm of cefamandole (N = 129) or cefotaxime (N = 131) immediately before operation. The two treatment groups were comparable. Febrile morbidity occurred in seven (5.4%) cefamandole patients and in 17 (13%) of those given cefotaxime (p = 0.04). In contrast, the number of patients who developed major pelvic infection was equal (cefamandole, 6; cefotaxime, 5). There was no significant difference in the concentration of either drug in the serum or tissues surveyed, and there was no correlation between infection and concentration of antibiotic in either group. The mean hospital stay and the mean total hospitalization charge did not differ significantly between the two groups. Both of these agents given as a single dose before vaginal hysterectomy are effective in reducing postoperative infections.