The present study examined the efficacy of an 8-wk, cognitive-behavioral group treatment for panic disorder. Patients meeting DSM-III-R criteria for panic disorder with or without agoraphobia were randomly assigned to treatment (N = 34) or delayed treatment control (N = 33). The treatment consisted of: (a) education and corrective information; (b) cognitive therapy; (c) training in diaphragmatic breathing; and (d) interoceptive exposure. At posttreatment, 85% of treated Ss were panic free, compared to 30% of controls. Treated Ss also showed clinically significant improvement on indices of anxiety, agoraphobia, depression and fear of fear. Recovery, as estimated conservatively by the attainment of normal levels of functioning on each of the major clinical dimensions of the disorder (i.e. panic, anxiety and avoidance), was achieved in 64% of the treated Ss and 9% of the controls. At the 6 month follow-up, 63% of the treated patients met criteria for recovery. These findings mirror those from recently-completed trials of individually-administered cognitive-behavioral treatment, and suggest that CBT is a viable alternative to pharmacotherapy in the treatment of panic disorder.