Seven hundred and seventeen patients were included in a prospective randomized double-blind trial comparing the efficacy of five doses of cefamandole (group I, 335 patients) with that of a single preoperative dose (group II, 382 patients) for prophylaxis against sepsis in patients who had an operation using either a Moore prosthesis, Ender or Kuntscher nails, a bone plate, or another internal-fixation device. Patients who had an open fracture or total joint replacement were not included in the study. The two groups were similar in terms of mean age, sex ratio, duration of preoperative hospital stay, underlying risk factors, and type of surgical procedure. A wound infection developed in none of the seventy-four patients in group I and in five (6.6 per cent) of the seventy-six in group II (p = 0.03) who required a Moore prosthesis. A wound infection developed in three patients (1 per cent) in group I and in fifteen (5 per cent) in group II (p = 0.006) who required an internal fixation device other than a Moore prosthesis. Staphylococcus aureus, Staphylococcus epidermidis, and gram-negative bacilli were the most common infecting organisms. The rate of mortality was similar in both groups, while the percentage of postoperative urinary-tract infections was lower in group I (p = 0.04). No adverse side effects of the drug were encountered.