The risk taking behavior of sober subjects on the choice dilemma questionnaire was compared to that of subjects drunk on bourbon, vodka and synthetic alcohol. The results indicated that (a) contrary to previous findings, drunk subjects will accept greater subjective risk than sober subjects and (b) subjects drunk on bourbon, which is very high in congener content, will take greater risk than those drunk on the other two beverages, which are low in congener content. Alcohol produced greater elevation on some items in the questionnaire than on others. Earlier research has shown that older age produces a decrease in risk on the same items, while group discussion produces an increase in risk on a different subset of items. This may indicate, among other things, that the shift toward greater risk taking that occurs under the influence of alcohol is not produced by the processes that are responsible for the risky shift in group discussion.