In a 4-week double-blind study comparing alprazolam with diazepam treatments, 48 outpatients suffering from mild to moderate generalized anxiety were evaluated after a 5-day placebo washout, and then after 1, 2, and 4 weeks of treatment. The optimal therapeutic doses without excessive sedation averaged 2 mg for alprazolam and 15.8 mg for diazepam. Results from the Hamilton Anxiety Rating Scale, the Clinical Global Impression Scale, a behavior checklist questionnaire, and a symptomatic patients' self-rating scale indicated that patients improved in both treatment groups. Results from the comparative phase suggest that diazepam is more efficient than alprazolam in the reduction of several symptoms of anxiety and depression in particular. Assuming that the first 2-week ratings depend on accuracy of dose adjustment and that week 4 ratings are an important evaluation of long-term efficacy, results from this study suggest that adequate control of anxiety is obtained more readily with diazepam and that symptoms of depression might benefit more from that drug. Few side effects were reported: mainly, drowsiness, tremor, light- headedness , and dry mouth. A toxic reaction to alprazolam, possibly allergic, was observed. Either alprazolam or diazepam appeared to be effective in the treatment of generalized anxiety disorder, and the statistically significant differences between the two drugs were not clinically striking.