To the Editor.— Bolos et al, 1 in contrast to us, 2 purport to find a lack of association between the dopamine D 2 receptor (D 2 R) gene and alcoholism. In the selection of 40 alcoholic subjects, Bolos et al used as an exclusion criterion "an acutely active medical disorder as revealed by the patient's medical history, physical examination, or clinical laboratory assessment." By systematically excluding severely ill alcoholics (or presumably even alcoholics with minor abnormalities in clinical chemistry), these investigators studied a quite different and less severely ill alcoholic population than we did. Our study consisted of alcoholics who had experienced repeated failures in their alcoholic rehabilitation and whose cause of death was primarily attributed to alcohol-induced damage to their bodily systems. We used a sample of 35 alcoholics and 35 nonalcoholic controls, and the two groups were matched on age and other important variables. Bolos et al