Information which has emerged thus far relates to the overall transmitter mechanisms of sleep. The data, while conflicting, point to the involvement of many neuroregulators at numerous integrative levels of the process. However the long term question still remain: what triggers and maintain sleep, what stops sleep, what occurs to the body and brain during sleep--in essence, why sleep? These questions are now problems for behavioral neurochemists, whereas in a previous era, they were problems for philosophers. Unfortunately, our answers to date, while in another idiom, have hardly been more complete or satisfying. To answer these questions, it will be necessary to understand, in detail, the manner in which neurobiochemical processes relate to the functional physiology of sleep. Although existing studies have given invaluable insight into the neurochemical anatomy of sleep, we have only recently acquired the technical and biochemical expertise necessary to investigate sleep as it occurs normally. Future research must focus on the dynamic changes associated with the regulatory mechanisms of neurotransmitters. Many questions can be asked. With sleep transitions, what changes occur in transmitter content, synthesis, or release? Are there changes in metabolic pathways, reflecting a shift from intra- to interneuronal metabolism? What changes occur in pre- and postsynaptic neurotransmitter receptors to affect sensitivity? What constraints do genetic (245) and environmental (246) factors impose upon these mechanisms? Knowledge of such parameters will allow us to construct more complete models of the neuroregulatory basis of sleep and waking. However, as we acquire this knowledge, we must avoid the temptation of assuming causation when the evidence merely shows correlation. Neuroregulation are involved in the control of number different behaviors; and, at present, we have few, if any, methods of establishing causative links between a specific neuroregulator and a specific behavioral state. Yet, even without an understanding of what "causes" sleep, we may be able to develop pharmacological agents which permits discrete alteration of sleep mechanisms in a more physiological and specific manner. This potential for manipulation of sleep is of obvious importance in illnesses such as insomnia, narcolepsy, and sleep apnea (247, 248). In addition, it may be valuable in the treatment of such conditions as psychosis and depression, where sleep disturbances are an important component of the illness. For example, delirium tremens might be best understood as a psychotic episode which is the result of an aspect of sleep emerging into wakefulness. The range and breadth of both the basic questions and the potential application of sleep research portend an exciting future for this field.