Ethanol (0.5 g/kg) was administered to 12 healthy, normal-sleeping men, aged 21 to 45, at two different times of the day (0900 and 1700 hr). The Multiple Sleep Latency Test (MSLT) was conducted at 1000, 1200, 1400, and 1600 hr in the day drinking condition and at 1800, 2000, 2200, and 2400 hr in the evening drinking condition. On placebo, sleepiness was greater in the daytime testing hours than in the evening, replicating findings on the circadian rhythm of sleepiness/alertness. There was a time of drinking (day versus evening) by ethanol interaction. An ethanol effect on sleep latency was found in the daytime hours, when alertness was relatively low. Ethanol failed to have a significant effect on sleep latency during the evening hours when alertness levels were increasing. Performance on a divided attention task, administered 1 hr postconsumption, was impaired by ethanol consumption, but did not vary as a function of time of drinking (day versus evening). However, at 5 hr postconsumption, mean reaction time on the first 20 min of a 40-min auditory vigilance task was slowed by ethanol to a greater extent after day drinking then after evening drinking.