Previous work on the self-selection of alcohol by rats, where the fluid choice has been limited to an alcohol solution and water, indicated that on diets lacking certain essential elements, e.g., vitamins, the quantity of alcohol voluntarily chosen was greater than on diets not exhibiting such lacks. The analogy was immediately drawn that human alcoholism might stem from such dietary lacks brought about through genetic factors which induced a relatively greater need for particular dietary components. The present paper embodies the results of work on 10 rats over a 7.5 month period. On a diet known to yield good growth (46% carbohydrate, 22% protein, 25% fat) these rats voluntarily drank amounts of alcohol (on a choice between alcohol and water) approaching their capacity to metabolize alcohol. The consumption of alcohol was significantly decreased on this diet if a third fluid choice was introduced; as a third fluid choice an emulsion of fat, a solution of saccharin, or a solution of sucrose were increasingly effective (in the order given) in reducing the consumption of alcohol. Practically, sucrose solution reduced the consumption of alcohol to zero. The use of sucrose in the solid form as a second solid choice or the incorporation of additional sucrose in the diet (63% carbohydrate, 18% protein, 12% fat) also decreased the consumption of alcohol. It is concluded that the preferences expressed by these rats may be due to caloric content, palatability, form, or to other factors. These findings demonstrate that there is no specific preference for alcohol. Together with the fact that human beings are not bound by the restrictions imposed upon rats, and because human alcoholics become intoxicated while rats on a self-selection regime do not, it is impossible to assume that the etiology of human alcoholism is connected with dietary lacks.