DESCRIPTION (provided by applicant): About 70 million Americans suffer from a sleep problem; among them, nearly 60 percent have a chronic sleep disorder. Sleep disorders account for approximately $16 Billion of the annual US health care bill, yet basic questions about the functions and evolution of sleep remain unanswered and understudied. The purpose of this proposal is to investigate the distinct evolutionary histories and physiological functions of the two major mammalian sleep states REM (rapid eye movement or active sleep) and NREM (non-REM or quiet sleep) using state of the art analytic procedures for assessment of phylogenetic change. Such an investigation would clarify the role of each of the sleep states in the restorative capacities of sleep and would illuminate the ways in which breakdowns in the separate functions of each sleep state could lead to clinical disorders of sleep. Our approach is to use phylogeny-based comparative analysis to examine how, and in association with which other traits, REM and NREM evolved. Research during the past two decades has produced major advances in understanding sleep within particular species. Simultaneously, molecular advances have made it possible to generate phylogenetic trees, while new analytical methods provide the tools to examine macroevolutionary change on these trees. Thus, comparative methodology has entered a new "Golden Age" - but these methods have yet to be applied to questions concerning the evolution of distinctive sleep state functions. Our specific aims are to 1) Develop a Web-accessible comparative database on mammalian sleep patterns. Exhaustive literature searches have shown that quality data on sleep quotas is now available for approximately 150 species. We present analyses in this re-submittal that show that data on sleep quotas that were obtained over the past few decades in the laboratory are generally reliable and valid indices of sleep processes as they occur in the wild. Investigators who collected these data made strenuous efforts when recording sleep values to approximate natural conditions for the animals under study. Our other aims involve analyses of the data set constructed under Aim 1. These aims include 2) Investigate allometric trends, quantify phylogenetic signal and identify confounding variables. 3) Test a priori hypotheses for the evolution of and functions of active/REM versus quiet non-REM Sleep: 4) Develop a cost-benefit model of sleep states: Development of a cost-benefit model of sleep will allow us to further explore independent functions of REM/NREM and to derive clinical implications of perturbations in sleep durations of the two sleep states that typically occur in major human sleep disorders.