Although recent survey findings have indicated a decrease in the percentage of teenage smokers, cigarette smoking remains the leading preventable cause of morbidity and mortality in the United States. A social learning smoking prevention intervention, consisting of four sessions which occurred at 2-week intervals, was conducted with fifth and sixth graders in two school systems. Schools were randomly assigned to intervention (N=476 students) or control (N=392 students) conditions. Students in the intervention groups learned about the physiological effects of cigarette smoking, the pressures which influence adolescents to smoke, ways to resist peer pressures, and the techniques of cigarette advertising. All students were pretested at the beginning of the first study year and posttested at the end of that school year and the beginning of the next school year. The results indicated a significant treatment x sex x occasion interaction with respect to experimentation with cigarettes; and a significant treatment x school system x occasion interaction with respect to recent cigarette smoking. The intervention was effective in reducing the prevalence of experimentation with cigarettes among boys, but not among girls, in both school systems; and was effective in reducing the onset of more frequent smoking among both sexes in one school system, but not in the other. Social learning intervention may not be effective in schools where a certain threshold level of smoking has been reached. (NRB)