Sixty-eight patients suffering from generalized anxiety disorder were allocated to one of four groups to study how hypnotizability (high versus low) and type of suggestion (direct versus indirect) were related to treatment outcome. Patients were assessed on three occasions (Pre-treatment; Post-treatment; and three month Follow-up) using seven anxiety measures and one depression inventory. The Creative Imagination Scale (CIS) was used to assess susceptibility to hypnotic responding. Twenty Ss dropped out prematurely due to both practical and treatment related issues. Interestingly, dropouts were also found to have been significantly more depressed than completers. Patients were given six treatment sessions of hypnotherapy at weekly intervals while the author remained blind to CIS scores. Therapy focused on accessing S's unconscious abilities and resources for symptom resolution, using either direct or indirect suggestion. Patients were provided with self-hypnosis tapes for anxiety reduction in between appointments. Results indicated that completers significantly improved on all dependent measures, but few differences emerged between the four groups. Interestingly, highly hypnotizable patients who received indirect suggestion relapsed slightly on all measures by the Follow-up period, having improved from Pre- to Post-treatment. Discussion is offered on the possible implications these results have for the integrity of the independent variables, concluding that both hypnotizability and type of suggestion have a clinically significant interactive effect. Direct suggestion is recommended for highly hypnotizable generalized anxiety disorder clients, while both direct and indirect suggestion can be used with low hypnotizables, as long as all are couched within permissive language. Limitations of the present research are discussed, along with recommendations for clinical practice and future research