In a prospective study of 187 patients without known prior phlebitis who underwent total hip replacement, we compared the efficacy of warfarin, low-molecular-weight dextran, aspirin, and subcutaneous heparin for prevention of postoperative venous thrombi. Detection of thrombi was by phlebography. Warfarin dextran, and aspirin were superior to heparin. There was no significant difference between the effects of warfarin, dextran, and aspirin on the number of patients with fresh thrombi. In reducing the number of thrombi formed, warfarin and dextran were superior to aspirin. For reduction of the prevalence of thrombi in the thigh, the three drugs were equally effective. Significantly fewer bleeding complications occurred with aspirin than with warfarin. Even in patients who received warfarin, dextran, or aspirin, the prevalence of fresh thrombi was high. Thrombi often formed in the thigh without an associated calf thrombus. Fifty-one of fifty-six episodes of thromboembolic disease were clinically silent at the time of detection. The prophylactic use of warfarin or aspirin, followed by warfarin treatment if a thrombus was detected on phlebograms, provided effective protection against pulmonary embolism.