Abstract
Two experiments compatible with that used to investigate the effect of clinical doses (200 mg.) of amylobarbitone were set up to investigate the effects of clinical doses of chloral hydrate (800 mg.), dichloral phenazone (1,300 mg.), and Mandrax (methaqualone 250 mg. and diphenhydramine 25 mg.) over a period of one to two weeks.Four healthy male subjects were used in each experiment and received placebo or drug throughout a period of six to eight weeks when control records, drug records, and drug withdrawal records were obtained.Chloral hydrate was found to depress rapid eye movement (R.E.M.) sleep appreciably though less consistently than amylobarbitone. No withdrawal R.E.M. sleep rebound was found.Neither dichloralphenazone nor Mandrax was found consistently to depress R.E.M. sleep, though occasional nights when R.E.M. sleep was low occurred more often with Mandrax.In the light of other experiments it is postulated that there exists a "threshold" in the dose of a hypnotic, and that when this is exceeded the drug will produce R.E.M. reduction. Thus it may be possible to prescribe a drug which is clinically useful while avoiding withdrawal effects.