One-hundred and ninety-two probands were selected on the basis of personality questionnaires as being cancer-prone (100) or CHD-prone (92). They were then randomly divided into a control and a treatment group, the latter receiving a special kind of behaviour therapy attempting to change the personality patterns in the direction of a healthier, more autonomous personality. Follow-up after 10 and 13 years disclosed significantly lower death rates in probands receiving prophylactic treatment than in controls. Of special interest was the degree of sclerosis in the fundus of the eye, rateds on a 3-point scale. This was significantly higher prior to therapy in the CHD-prone group than in the cancer-prone group. Treatment reduced the degree of sclerosis, particularly in the CHD group; lack of treatment was followed by an increase in sclerosis. Similar but less marked changes were found in the cancer-prone group. Clearly psychological treatment affects significantly bodily functions associated with CHD. Other risk factors considered were systolic and diastolic blood pressure, blood cholesterol and blood sugar, all of which, together with personality, influenced the degree of sclerosis observed.