One hundred and two consecutive acute traumatic spinal injury patients admitted between 1976 and 1979 to a male ward at Stoke Mandeville Hospital were considered for prophylactic anticoagulant therapy. Sixty-six were anticoagulated. Thirty-six were not due to various contra-indications. Nineteen patients had pulmonary emboli. None of these were effectively anticoagulated. These results are compared with a similar previous series where there was a smaller incidence of emboli. The main reason for this was due to the delay in admitting patients to the Centre so that they could not be anticoagulated early enough to prevent pulmonary embolus occurring. In these patients who had a pulmonary embolus or were considered to be a high risk, anticoagulant therapy was continued for 6 months.