The incidence of venous thrombosis (VT) detected by 125I-labelled fibrinogen was followed in 408 patients with acute myocardial infarction (AMI). Patients were randomized into three groups: a group receiving small doses of heparin (2 X 5000 u. subcutaneously/24 h), and an exercise group (dorsal and plantar flexion of the foot for one minute each hour). VT was present in 13.6% of the control group, in 9% of the group with mini-heparin (an insignificant difference), and in 5.1% of the group with moderate exercise (p less than 0.05). VT was statistically significantly more frequent in patients with acute MI who had heart failure, thrombosis in their medical history, had varices and were non-smokers. The results indicate that even moderate exercise of the lower limbs decreases significantly the incidence of VT in patients with AMI.