Title: Sleep and brain thermoregulation: Abstract We have found that human hunter-gatherers sleep nearly an hour more in winter than in summer and that less than 2% of this population exhibits signs of insomnia or sleep fragmentation, compared to 10-30% of “industrial” populations. In contrast to hunter-gatherers, most individuals in industrial societies (and animals in laboratories) experience a relatively constant temperature from birth to death, due to modern insulation, heating and air conditioning. We will test the hypothesis that the attenuation or elimination of natural environmental temperature rhythms disrupts sleep by comparing sleep at 3 months of age in rats experiencing a 24 hour temperature rhythm from birth to 2 months, to rats experiencing a constant temperature from birth. Seventy years after the discovery of REM sleep, the function of this state remains unclear. Despite claims that all mammals have REM sleep, we have found that cetaceans (whales and dolphins) do not have REM sleep and have only unihemispheric nonREM sleep. More recently we have found that fur seals, when they are on land, show bihemispheric nonREM sleep and REM sleep. However, when in water, where they spend about 7 months a year, they show “unihemispheric nonREM sleep” and largely cease having REM sleep, like cetaceans. They have no significant “rebound” of REM sleep upon return to land. Brain temperature has been found to decrease in bihemispheric nonREM and increase in REM sleep in all studied mammals. We have proposed that REM sleep has a key role in brainstem thermoregulation and that REM sleep facilitates alert awakening by reversing the reduced brain temperature and metabolism of bilateral nonREM sleep. REM sleep normally occurs after nonREM sleep. The reduced brain metabolism and associated temperature reduction of nonREM sleep are adaptive, because the brain consumes a large percentage of whole body energy at rest, and brain energy consumption is greatly reduced in nonREM sleep. But animals, including humans, awakened from nonREM sleep are substantially less alert than animals awakened from REM sleep. The reduced alertness upon awakening from nonREM sleep can last more than 30 minutes in humans. Thus, it is highly adaptive to have the brain warming to waking levels that occurs in REM sleep, which occupies its highest percentage of sleep time prior to awakening. We propose to determine the effect of environmental temperature on cortical and brainstem temperature. We propose to identify the brainstem sites at which temperature manipulation affects REM sleep latency and duration. Understanding the role of REM sleep in brain thermoregulation may be relevant to our understanding of narcolepsy, hypersomnia and insomnia.