In this study we tested whether the efficacy of behavior therapy for obesity might be improved by lengthening the duration of treatment. Forty-eight obese clients were randomly assigned to either a standard treatment of 20 weekly sessions or to an extended treatment of 40 weekly sessions. The content of each program was identical, but the treatment procedures were introduced in a more gradual manner in the extended condition. At Week 20, the conditions showed equivalent weight losses. At 40- and 72-week evaluations, however, the extended treatment produced significantly greater mean weight losses than did the standard treatment.