The aim of the present study was to investigate the effects of extended night sleep on subsequent daytime sleep propensity in narcoleptic patients. The possible differentiation between REM and NREM sleepiness was of particular interest. Ten unmedicated narcoleptic inpatients (8 men, 2 women) aged 23-61 years (mean, 47.9 years) participated in this study. Adapting the patients to the hospital schedule (nocturnal sleep period from 22:00 to 6:00 h and ad lib, nap during daytime) for at least 2 weeks, we conducted a baseline PSG from 22:00 to 6:00 h and subsequent 5-trial daytime sleep recordings with naps (lying on the bed for 20 min light-out period) at 9:30, 11:30, 13:30, 15:30 and 17:30 h (the baseline condition: BC). After a 1-5 day interval, we conducted an extended PSG from 22:00 to 10:00 h. Subsequent to the extended PSG, we carried out 5-trial daytime sleep recordings. We adjusted the start time of the first nap trial at least 90 min after waking time and start time in the 5th nap trial before 19:00 h (the extended condition: EC). Mean TST was 417.0 min in the baseline PSG, and 595.2 min in extended PSG. Mean number of daytime naps per subject exhibiting a sleep latency shorter than 10 min was decreased in EC. Mean sleep latency in EC was significantly prolonged in comparison with that in BC. This prolongation of mean sleep latency per subject was positively correlated with S4 duration and % S4 obtained in the morning of the extended PSG (from 6:00 to wake time).(ABSTRACT TRUNCATED AT 250 WORDS)