A total of 104 men who were heavy smokers (minimum of a pack, and averaging 35 cigarettes a day) participated in a 10-week program to break the habit. The majority (N=83) received group therapy, and the remainder were seen individually. Each S was randomly assigned to one of the following drug conditions: no pills, placebo, lobeline, dextroamphetamine and imipramine. Pretesting established each S as a good (N=67) or poor (N=37) risk to stop smoking. Results indicated that, both at the end of treatment and at followup 3 months later, group was superior to individual therapy, treatment without drugs was more effective than taking medication (especially for high risk cases), and low-risk did better than high-risk Ss. Successful Ss stayed in treatment longer than did failures. Relapse was associated with life situations of loneliness, passivity, boredom, tension and personal tragedy. The best predictor of resistance to relapse was the abrupt and complete breaking of the habit during the first 2 weeks of the program ("cold turkey").