THE EVOLUTIONARY THEORY OF SLEEP AND WAKEFULNESS NATHANIEL KLEITMAN* Which came first—the hen or the egg? In the alternation ofsleep and wakefulness, which of the two states interrupts the other? Is the onset of sleep an active process or a mere cessation ofwakefulness? By an evolutionary theory I presented earlier [i], sleep is a passive condition, and the phylogenetic and ontogenetic development of the CNS (central nervous system) is reflected mainly in the characteristics ofinduced wakefulness : die subcortically controlled wakefulness of necessity gradually evolves into a cortically regulated wakefulness ofchoice, along with establishment of the 24-hour sleep-wakefulness rhythm. A certainty was expressed at that time that, as new facts were brought to light, modifications would have to be made in the theory. Discoveries since then suggest that there are evolutionary changes in sleep as well as in wakefulness, and revisions ofthe theory dictated by these discoveries are presented now. The onset and continuation ofsleep, as well as awakening and the maintenance ofwakefulness, can conceivably be explained by postulating the stimulation and inhibition, or activity and rest, ofa sleep center, wakefulness center, or two reciprocally innervated centers [2, chap. 35]. But examination offindings associated with destructive lesions in the brain stem results in the conclusion that the principal, ifnot the only, mechanism involved is a BSRF (brain stem reticular formation) center or system which, active, induces and maintains wakefulness and, inactive, leads to sleep. The possible existence of accessory or secondary hypnogenic systems— rostral, caudal, or within the BSRF—does not affect this conclusion fun- * Professor Emeritus, Department of Physiology, University of Chicago. Address: 222 Washington Avenue, Santa Monica, California. This article is a slighdy modified version ofChapter 36 in the revised and enlarged edition ofthe author's book Sleep and Wakefulness, published by The University ofChicago Press, 1063. I69 damentally. In the medulla, for instance, one can stimulate certain spots to produce vasoconstrictor effects and others to cause vasodilation; but cutting the cervical spinal cord leads to a profound drop in blood pressure, showing that the principal tonic influence is a vasoconstrictor one. In any case, there is general agreement that the location of the primitive sleepwakefulness system or systems is subcortical. Cortical processes are influenced by and, in turn, influence the mesodiencephalic centers, although they are not needed for the centers' elementary functioning in producing an alternation ofsleep and wakefulness. An innate alternation ofprimitive sleep and wakefulness can be seen in anencephalous children and in decorticated dogs. In these creatures there is no consciousness (as I defined it elsewhere [2, chap. 1])—nothing to learn individually. A similar sleep-wakefulness cycle prevails in the normal newborn infant. The temporal aspects ofthe cycle are: (a) sleep duration of2 to 4 hours, bearing little relation to the succession ofnight and day; and (b) dominance ofthe sleep phase, with a sleep-to-wakefulness ratio of two-to-one. The criteria for the repetitive passage from primitive wakefulness to primitive sleep are: (a) decrease or cessation of muscular activity and (b) raised threshold ofreflex excitability. The periodic awakening is adjusted to the organism's nutritional needs and is essentially a gastric cycle. Either hunger contractions or some humoral agents furnish the internal stimulus, in the absence ofexternal disturbances. The length ofthe cycle represents a coalescence ofseveral basic short-term rest-activity periodicities of50 to 60 minutes' duration, as first described by Denisova and Figurin [3] and later observed in our laboratory [4]. On a self-demand infant feeding routine, the interfeeding interval is usually an integer ofthese basic activity cycles. Ifthe infant is not aroused through external or internal stimuli during the shallow phase ofthe periodicity, he is not likely to awaken till the shallow phase recurs. The relatively short primitive wakefulness is maintained through the activityofthewakefulness centeror system oftheBSRF, which, independently ofits ability to arouse the cortex, has extensive feedback connections via the DRAS (descending reticular activating system) with caudal regions ofthe nervous system and peripheral receptors and effectors. After feeding is completed and other general body needs or animalistic functions fulfilled , the activity ofthe wakefulness system abates and sleep sets in. Primitive sleep, though undoubtedly dreamless, is not ofuniform depth, for it is 170 Nathaniel Kleitman · The Evolutionary...