Most people would agree that a relationship exists between drinking alcohol and smoking cigarettes, but there is very little empirical evidence demonstrating a direct causal association. Study of the relationship has been hampered by the lack of a simple laboratory methodology. This article describes an efficient experimental paradigm. Fourteen male narcotic addicts in methadone maintenance treatment volunteered to come to the laboratory for two smoking sessions, during which each subject was given either an alcoholic drink or orange juice, followed by three cigarettes at 20-minute intervals. Drinking alcohol significantly increased the amount and rate of smoking. However, not all subjects reacted to alcohol with increased smoking, and we were unable to account for those individual differences. Our finding supports the theory that a direct causal mechanism linking smoking and drinking exists. That link probably has clinical significance, because there is evidence that ex-smokers are at particularly high risk when they drink alcohol.