Although several writers have emphasized the habit-like characteristics of bulimia nervosa, no study has investigated the extent to which bulimia responds to treatment specifically for habit disorders. Habit reversal is a general treatment plan which teaches individuals to regain lost awareness of all aspects of a habit. The habit is then interrupted at the earliest movement in the chain by a physically competing response. Habit reversal (Azrin & Nunn, 1973) has been effective in treating other habits (e.g., stuttering, tics, bruxism). In the present study, habit reversal was adapted for the treatment of bulimia. In response to newspaper advertisements, ten females (aged 21-26) with a minimum of two binges/week were included in the study. Bingeing and vomiting ranged from .5/day to 3.57/day during baseline; duration of the disorder ranged from one to eight years. In a multiple-baseline, across-subjects design, subjects were randomly assigned to begin treatment following 14 days of baseline data collection. Subjects self-monitored and reported daily binge eating, vomiting, fasting and laxative use. Daily social contacts and athletic activities were self-monitored and reported as secondary dependent measures. A blood chemistry analysis to assess metabolic functioning, the Beck Depression Inventory (BDI) and the Symptom Checklist 90-Revised (SCL-90R) were administered pre- and post-treatment. Two subjects reported cessation of bulimic episodes during baseline data collection apparently as a result of self-monitoring and frequent phone contacts. A rapid and substantial reduction in the target behaviors for the subjects who received treatment was reported following presentation of habit reversal (mean bingeing reduction=66.7%; mean vomiting reduction=52.0%). As reports of bingeing and vomiting decreased, concomitant increase in reported social contacts and athletic activities was observed in five of seven subjects without treatment for social behavior. Scores on the BDI and SCL-90 R did not change as a function of treatment. However, the self-reported frequency of binge eating during treatment correlated positively with psychopathology (r=.95, p