This article reviews studies of ethanol self-administration in non-human primates. Between approximately 1960 and 1980 research programs at three university medical centers (Michigan, Minnesota and Baylor) established that ethanol could function as a reinforcer when delivered either intravenously, intragastrically or orally. Variables such as session length, dose and intermittent reinforcement schedules were examined, and tolerance and physiological dependence on ethanol were also studied. Procedures used to establish abused drugs as reinforcers via the intravenous, intragastric and oral route are also effective in the particular case of ethanol. Under conditions of limited access (e.g. 3h per day), the variables that control behavior reinforced by ethanol are the same as those that govern behavior reinforced by other drugs. Moreover, the functional relations between these variables and ethanol-reinforced responding are similar to the functional relations between the same variables and other drug reinforcers. However, with continuous 24h access to ethanol, differences appear with respect to other sedative-hypnotic drugs. In contrast to these other drugs, intravenous ethanol use results in a suppression of food intake, weight loss, self-initiated abstinence, and often, death. When intravenous and oral ethanol self-administration are compared, two important differences emerge: the maximum intakes via the intravenous route exceed the maximum intakes via the oral route, and, perhaps surprisingly, the doses required to maintain responding are lower with the oral route. Studies since 1980 have increased the scope of ethanol reinforcement research and also corroborated and extended earlier findings. In summary, ethanol can function as an effective reinforcer for non-human primates.