Two benzodiazepine hypnotics, triazolam, 0.25 mg, with a short elimination t1/2, and quazepam, 15 mg, with a long t1/2, were evaluated in 22-night sleep laboratory studies. Quazepam improved sleep significantly during both short- and intermediate-term use. Daytime sleepiness, which decreased with continued use, was the side effect most often associated with quazepam dosing. In contrast, triazolam dosing did not significantly improve any of the major sleep efficiency parameters, and there was a rapid development of tolerance for the drug's slight initial effectiveness. In addition, there were a number of behavioral side effects including amnesia, confusion, and disinhibition. Withdrawal of triazolam was associated with sleep and mood disturbances (rebound insomnia and rebound anxiety), whereas quazepam exerted carryover effectiveness. Thus the data in this study show that the 0.25 mg dose of triazolam, which is being prescribed increasingly, has a profile of side effects that is similar to that of the 0.5 mg dose.