Because of the controversy surrounding prophylaxis for thromboembolism after total hip surgery, we undertook a prospective study comparing the results of the administration of aspirin with that of low doses of warfarin in 194 patients (200 hips) undergoing total hip replacement. The incidences of both clinically apparent and silent (asymptomatic) pulmonary emboli were determined using the objective criteria of preoperative and postoperative levels of arterial blood gases, chest roentgenograms, electrocardiograms, and perfusion lung scans. With this surveillance plan, the accuracy of diagnosis of clinically symptomatic pulmonary emboli was improved and the detection of otherwise silent pulmonary emboli became possible. The group of patients who received low doses of warfarin showed a 6 per cent total incidence of pulmonary emboli compared with a 19 per cent incidence in the group receiving aspirin (p less than 0.05). There was, however, no significant difference when the incidences of only the clinically suspected emboli were compared, the rates for the two groups being 5 and 8 per cent, respectively (p greater than 0.05). There was also no significant difference between men and women with regard to the prophylactic efficacy of aspirin in preventing pulmonary embolism.