A consecutive, randomly allocated, controlled clinical trial of the prophylactic effect of intermittent pneumatic compression of the calf on the incidence of postoperative deep vein thrombosis showed that in patients without malignant disease there was a highly significant reduction in the incidence of thrombosis. In patients with malignant disease the incidence of thrombosis was higher than in those without, and there was no reduction in incidence by the application of intermittent compression. In the absence of malignant disease, severity of operation and the age of the patient were the most significant aetiological factors. We found no relation between the incidence of deep venous thrombosis and obesity, length of preoperative stay, location of hospital, or duration of anaesthesia. We suggest that intermittent pneumatic compression as used in this trial is a safe, effective, and extremely practical method of preventing postoperative deep vein thrombosis in patients not suffering from malignant disease.