The clinical efficacy of moxalactam versus clindamycin/tobramycin was evaluated in a comparative, randomized, prospective study. Sixty patients were treated: 30 with moxalactam and 30 with clindamycin/tobramycin. There were 15 cases of tuboovarian abscess, 36 cases of severe pelvic inflammatory disease with peritonitis, eight cases of endomyometritis, and one wound abscess. Aerobic and anaerobic cultures from the sites of infection yielded 441 microorganisms from 53 patients; an average of 8.3 bacteria per infection (4.5 anaerobes and 3.8 aerobes). The infections tended to be mixed aerobic-anaerobic with anaerobes isolated in 90% of cases. The most frequently isolated possible pathogens were Bacteroides sp. (37), Bacteroides bivius (23), Bacteroides asaccharolyticus (12), Peptococcus asaccharolyticus (29), Peptostreptococcus anaerobius (19), unidentified anaerobic gram-positive cocci (18), Escherichia coli (17), nonhemolytic streptococci (16), Neisseria gonorrhoeae (13), and Gardnerella vaginalis (38). Clinical cure was noted in 29 of 30 moxalactam-treated and 29 of 30 clindamycin/tobramycin-treated patients. Moxalactam was effective in five of six cases of tuboovarian abscess, all 22 cases of pelvic inflammatory disease with peritonitis, the one case of endomyometritis and the one wound abscess. Clindamycin/tobramycin was effective in eight of nine cases of tuboovarian abscess, all 14 cases of pelvic inflammatory disease with peritonitis, and all seven cases of endomyometritis. No adverse hematologic, renal, or hepatic effects were noted with either regimen.