A total of 1489 patients were included in a prospective, randomized study that compared the efficacy of a single dose of cefonicid in 474 patients (Group I) with that of three doses of cefamandole in 510 patients (Group II) and five doses of cefamandole in 505 patients (Group III), for prophylaxis against infection after an operation on bone. The operations involved the insertion of a Moore prosthesis, an Ender and Kuntscher nail, a bone-plate, or another device for internal fixation. Patients who had an open fracture or a total joint replacement were not included in the study. The three groups were similar with regard to mean age, sex ratio, duration of preoperative hospitalization, underlying risk factors, and type of operation. The rates of wound infection were not significantly different in the three groups (p = 0.8) or when the rates were stratified according to the type of operation (p greater than 0.3). Staphylococcus aureus and gram-negative bacilli were the most common infecting microorganisms. The rate of mortality related to infection was similar in all three groups (p = 0.2). No adverse side-effects of drugs were encountered. A single preoperative dose of cefonicid, three doses of cefamandole, and five doses of cefamandole were equally effective prophylaxis against infection of the wound in these patients.