A double-blind study comparing zopiclone and triazolam, in 48 healthy, chronic insomniacs, was undertaken in two centers to compare the hypnotic and withdrawal effects of both compounds. Comparable doses of 7.5 mg zopiclone and 0.25 mg triazolam were given at bedtime for 21 nights after a 3 day wash-out period, followed by 4 placebo nights of withdrawal monitoring. During the investigation, a 17 item post-sleep questionnaire was completed daily, the Clinical Global Impression (CGI) scale weekly, and the Hamilton Anxiety Scale (HAM-A) at baseline and at the end of the study. Withdrawal effects were evaluated with a withdrawal symptom checklist and the CGI (withdrawal). Results indicated that both compounds improved sleep and were equally effective. However, a larger number of triazolam subjects withdrew from the study because of ineffectiveness or adverse side-effects. A greater number of zopiclone subjects experienced a transient modification of taste which disappeared with discontinuation of therapy. Significant deterioration in nearly all sleep parameters were noted after the first withdrawal night of triazolam; much fewer modifications of sleep parameters were observed following the discontinuation of zopiclone.