The effects of shifting sleep two hours within a fixed photoperiod.

This study examined the effects of shifting the time of sleep within a constant photoperiod on the circadian rhythms of body temperature and melatonin secretion. Subjects lived under conditions of a long scotoperiod (dim light of less than 10 lux from 6 p.m. until 8 a.m.) for three weeks. In order to delineate dawn and dusk, subjects received one hour of bright light (2500 lux) before and after the scotoperiod (i.e., from 8 to 9 a.m. and from 5 to 6 p.m.). For the first week of the experiment they slept from 10 p.m. until 6 a.m. In the second week, sleep was advanced two hours; that is, subjects retired at 8 p.m. and arose at 4 a.m. The third week was a repeat of the first, resulting in a two-hour delay of sleep from week two to three. The six subjects who successfully completed this protocol had no significant changes in the timing of the body temperature minima and onset of secretion of melatonin. This indicates that the timing of allowed sleep has less of an immediate effect on circadian rhythms than the timing of the external light-dark cycle. The circadian effects of the timing of sleep may be due more to the light-dark cycle that is imposed by the sleep-wake cycle than from the timing of sleep itself.