A mathematical model of diurnal variations in human plasma melatonin levels. Brown, Emery N., Yong Choe, Theresa L. Shanahan, and Charles A. Czeisler. Statistics Research Laboratory, Department of Anesthesia, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, Circadian, Neuroendocrine and Sleep Disorders Section, Division of Endocrinology, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115, USA
APStracts 3:0200E, 1996.
In animals melatonin facilitates processing of environmental photic information vital for timing of reproduction and other seasonally regulated behavior (27). While the hormone's function in humans is less well understood, its diurnal pattern is well-described (1, 3, 21, 31, 33, 36). The diurnal human plasma melatonin profile is characterized by a period of low to minimally detectable levels during the day; a 10 to 100 fold surge in concentration in the late evening; a decline in the early morning; and a return to low levels by mid morning. The normal 8 to 14 hour duration of elevated melatonin levels depends on light exposure patterns before and during the period of observation (36). Findings from studies in rats and several other mammals suggest that diurnal variation in plasma melatonin levels is governed primarily by the mammalian circadian pacemaker (14, 26, 28). In humans the diurnal rhythm of plasma melatonin levels correlates well with established circadian rhythms (31). As a consequence, there has been growing interest in the use of melatonin as a quantitative marker of the human circadian system.

Received 20 January 1995; accepted in final form 19 September
1996.
APS Manuscript Number E26-5.
Article publication pending Am. J. Physiol. (Endocrinol. Metab.).
ISSN 1080-4757 Copyright 1996 The American Physiological Society.
Published in APStracts on 5 November 1996