A number of studies have assessed the effect of REM sleep deprivation in animals on the acquisition and retention of various learning tasks. Fishbein (1969, 1970) has found deficits in retention on a one-trial passive avoidance task in mice when REM sleep deprivation either preceded or followed acquisition. If it preceded acquisition, retention deficits were found 24 hr. later bur not from 5 min. to 1 hr. later. If, however, REM sleep deprivation was inrerpolated between training and retest, impairment in retention occurred only if Ss were tested while REM deprived, but not if Ss were allowed sufficienr time to recover. The purpose of the present study was to assess the effect of REM sleep deprivation given before or after the acquisition of a one-trial passive avoidance cask on retention after a brief or extended interval of time. In Exp. I, 1 8 male albino rats approximately 300 days old received a 0.4-ma. shock of 2 sec. duration 5 sec. after passing from a lighted to a darkened comparunent of a rwocomparunent shuttle box. Ten Ss were then subjected to REM sleep deprivation via the inverted flower pot technique (Dement, 1965). Inverted flower pots having a 7 cm. base were used. An animal is placed o n top of rhe pot which is immersed in water (room temperature). A loss of muscle tonus which is concommitant with a loss of REM sleep results in S coming into contact with the water causing it to awaken. Eight Ss served as conrrols and were placed on pots 14.5 cm. in diameter. Such pots allow a certain amount of REM sleep to occur. Food was always available and within easy reach from small hanging wire containers. A record of S's weight was kept each day. As a further control, the 8 control Ss were yoked to 8 of the deprived Ss and were pushed into the water each time a deprived partner fell in. T h ~ s added control proved to be rather unnecessary since most control Ss failed to take advantage ol' the greater surface area of the larger pot while sleeping, and hence fell into the n2ntcr almost as many times as deprived Ss. After 4 8 hr. on the pots, each group was dirrded ~n half and tested for retention 30 min. or 24 hr. after being returned to their home cage. O n the retention trial, all Ss were allowed a maximum of 180 sec. to enter the dark comparunent. Results indicated that only one S in the deprived group ar the 30-min. interval entered the dark compartment within the 3-min. period. N o S in any other condition did so. N o significant weight loss occurred in either group. In Exp. 11, deprived Ss were treated exactly as those in Exp. I, except that REM sleep deprivation was initiated 72 hr. before acquisition training. The control group was housed on sawdust placed on the bottoms of the dry water tanks for the 72 hr. 10 rats served in the experimental group and 6 as controls. All Ss were 100 to 120 days of age. Groups were again divided in half and tested for retention 30 min. o r 24 hr. later. N o S in any condition attempted to enter the dark compartment during the 3-min. interval. One difference between the first experiment and this one was the significant weight loss of deprived Ss of approximately 50 grn. per animal over the 3-day deprivation period. This may be due in part to the use of younger Ss and the extra day of REM sleep deprivation. Results o f these cwo studies indicate that at least in the context of our experimental conditions, REM sleep deprivation either before or after the acquisition of a one-trial passive avoidance task, had no effect on retention in the REM deprived state or after recovery.