Rapid eye movement sleep (REMS) deprivation is believed to alter the sensitivity of various neurotransmitter systems. In the present article, we studied 20 healthy volunteers divided into three groups. Group A attended the sleep laboratory for three nights: acclimatization, a baseline night, and one night of physostigmine infusion. Group B attended for eight nights; acclimatization, baseline, four nights of REMS deprivation, and two recovery nights. With the exception of the first recovery night, when group C volunteers were administered physostigmine, group C's schedule was identical to group B's. The infusions received by group A and C were composed of 1.0 mg of physostigmine, dissolved in 100 ml of saline solution. These were administered 5 min after sleep onset and thereafter every hour, except when the subjects were either awake or in REMS. All of the subjects receiving the cholinomimetic infusion were given a peripheral anticholinergic. Group A experienced a great number of awakenings with a decrease in REMS percentage. Group B recovery occurred over two nights, with an increase in the average length of REMS. Group C exhibited maximum REMS rebound on the first recovery night with an increased number of REMS episodes, as well as significant reductions in the first REMS latency. Our findings suggest that physostigmine alters REMS rebound following REMS deprivation.