The introduction of behavior modification in the treatment of obesity a decade ago resulted in a substantial increase in weight loss of persons treated for mild to moderate obesity. It has been hoped that this increased effectiveness of treatment would extend also to maintenance of weight loss, and the first controlled clinical trial suggested that it did. This article reports the results of (1) a five-year follow-up of this first trial and (2) all follow-up studies of behavior modification for obesity yet conducted, six published and four previously unpublished. A new method of data analysis shows that weight losses are only modestly maintained, although the question of how their maintenance compares with that of other treatments cannot be answered because comparable data on other treatments are not available. In contrast to the vast amount of work on the initiation and generalization of behavior change, study of its maintenance has been a neglected area of behavior modification. We suggest that this area constitutes a new frontier for research in behavior modification.