The effectiveness of 30 mg temazepam (SaH 47-603) for inducing and maintaining sleep was evaluated in the sleep laboratory in six insomniac subjects under conditions of short-, intermediate-, and long-term drug administration. Administration of temazepam had no effect on sleep induction. In addition, effectiveness was not demonstrated for sleep maintenance: wake time after sleep onset was not significantly decreased on any of the three drug conditions, while the number of nightly awakenings was significantly decreased on all three drug conditions. Total wake time was decreased only slightly with short-term drug administration and was similar to baseline with intermediate- and long-term use. The percent REM sleep was essentially unchanged throughout the drug administration period. On the initial set of withdrawal nights, the per cent REM sleep was slightly but no significantly increased over baseline. The per cent slow wave sleep decreased significantly with short-, intermediate-, and long-term drug administration. After withdrawal, the percent slow wave sleep returned to baseline values.