Abstract Many health disorders have been attributed to the changes brought about by “modern” civilization. Understanding how the changes in the modern environment are affecting health requires comparison with humans living in the pre-modern environment. The number of groups living in the traditional way is rapidly diminishing. Electrification and the economic development of Native American reservations eliminated such groups in the US decades ago. A similar process is ongoing in other regions of the world. Only a few groups of hunter-gatherers remain. We began a study of sleep in non-industrial humans to test the widespread belief that “modern” humans sleep much less than the evolved “natural” amount. We found that, surprisingly, at the time points that we sampled, sleep durations in all three groups we studied were lower than those seen in most studies of industrial societies. Furthermore, we found a nearly 1 hour difference between summer and winter sleep durations in the two groups we sampled for 1 week periods in both seasons, settling the longstanding controversy over whether human sleep durations respond to seasonal changes. The prior consensus was that there was no difference or only a minor change in duration. We also found that, unlike other primates, humans in the natural environment do not typically start sleep at sunset nor do they awaken at sunrise. Rather, sleep onset is several hours after sunset and awakening is 30-60 minutes before sunrise. At the time points examined, our preliminary studies suggest that hunter-gatherers do not show peripheral vasodilation at sleep onset. But, they peripherally vasoconstrict upon awakening. In contrast, laboratory studies show that “modern” humans have the opposite pattern, peripherally vasodilating at sleep onset but not peripherally vasoconstricting upon awakening. Most strikingly, hunter-gatherers do not report or exhibit substantial levels of insomnia, with rates under 2%, in comparison to the 10-15% rates reported in all industrial societies. Our data suggest that environmental temperature and the thermoregulatory responses to the daily temperature cycle may be as important, or more important, than light level in the physiological regulation, function and pathology of human sleep. Nearly all studies of the timing of human and animal sleep have been conducted at constant, thermoneutral temperatures and have focused on the role of electric light. We propose to record sleep and core, peripheral and ambient temperature as well as ambient light levels under natural, evolutionarily relevant conditions by monitoring subjects continuously across an entire year. Our integrated approach will allow us to compare the effect of light and temperature on sleep time, EEG defined sleep states and alertness. These will be the first such studies under natural conditions and will provide insights into the temperature and light parameters differentially linked to REM vs nonREM sleep, EEG spectra, sleep duration...