This double-blind study was performed on 10 aged insomniac patients who received, during 15 days, either triazolam (0.25 mg) or zopiclone (7.5 mg) at bed time. This in-patient period was bounded by two ambulatory periods of 5 days each, and two in-patient periods of 3 days each, during which the patients received a placebo tablet at bed time. Thirteen sleep recordings per patient were performed before (3 nights), upon initiation of active treatment (3 nights), at the end of the active treatment period (3 nights) and during the following 3 "withdrawal" nights. A clear improvement of sleep patterns was observed with both drugs, although opposite effects on delta sleep were observed, namely a decrease with triazolam and an increase with zopiclone. A carry-over effect of the enhancement of delta sleep by zopiclone took place during the first 3 withdrawal nights. As is well known, sleep problems become increasingly common with age. Epidemiological studies show that although they only represent 11% of the population. Americans over 60 years old are prescribed about 40% of sleeping pills (Mendelson, 1980). Disturbed sleep in this population is often associated with medical disorders or induced by drugs like beta-adrenergic blockers. It must be admitted, however, that, possibly due to the association with medical or situational problems there have been only few attempts to assess the efficiency of hypnotic drugs in elderly insomniac patients. These are the reasons why in this study the effectiveness of zopiclone was compared to that of triazolam in insomniac patients aged over 60 years.(ABSTRACT TRUNCATED AT 250 WORDS)