Objectives: Elimination of endogenous melatonin by surgical removal of the pineal gland has been widely used in animal studies to examine the roles played by the hormone in physiology and behavior. In humans, pineal resection occurs during the removal of pineal neoplasms. It results in very low (<0.5 pg/ml plasma) or undetectable melatonin levels, generally without a discernible circadian rhythm. Pineal tumor resection thus provides a unique clinical model to assess the putative role of melatonin in sleep regulation, circadian rhythmicity and seasonality. We examined self-reported sleep characteristics, circadian chronotype, mood disturbance and seasonal variation in subjects who had undergone pineal tumor resection. Methods: Subjects (n=13; 62% male) ranged in age from 23.5 to 64.0 y (mean ± SD: 40.4 ± 12.2 y) and had undergone surgery 0.5 to 6.5 years earlier. Four patients had presented with pineal cysts and 9 with various tumors. Gross resection was performed in 11 cases, while 2 patients received an estimated 40% and 70-90% pineal excision, respectively. Seven did not require postoperative radiation or chemotherapy, while 4 received both, 1 received radiation treatment only, and 1 received chemotherapy without radiation. They received a battery of standardized self-assessment instruments including the Sleep Disorders Questionnaire (SDQ), the Morningness-Eveningness Questionnaire (MEQ), and the Personal Inventory for Depression and SAD (PIDS), which provides a global seasonality score and itemizes criteria for major depressive disorder in the past year. Overnight urine samples were collected to measure the concentration of 6-sulphatoxymelatonin (aMT6s) as an indicator of residual pineal function. Results: Three subjects were at high risk for sleep apnea (SA, n=2) or periodic leg movements (PLM, n=1), with scores above the 90 th %ile in the corresponding SDQ indices. Additionally, 10 SDQ items were extracted to obtain an estimate of insomnia (i.e., difficulty falling asleep, poor/disturbed sleep, and wakefulness during sleep) independent of other sleep disorders. Nine subjects (69%) reported at least one of these symptoms at least "sometimes", while 6 (46%) reported them at least "usually", and 5 (38%) "always". Four subjects (31%) reported sleeping less than 6 h (in one case, <5 h), while the longest wake episode during sleep was estimated at >20 min by 10 subjects (80%), >60 min by 7 (54%), and >120 min by 2 (15%). Two subjects used sleep medications, while 3 used melatonin supplements PRN (discontinued at least two days prior to urine collection). Seven subjects (58%) reported clinically significant symptoms of depression, although only 3 used antidepressants. The MEQ indicated that no one was an extreme circadian type: 6 subjects (46%) were rated as moderate morning types, 1 (8%) as a moderate evening type, and 6 (46%) as neither type. The global seasonality score averaged 5.0 ± 3.1, which indicates negligible seasonal mood variation similar to that of the nonseasonal subgroup of a random sample of New York City residents, and well below the overall population mean 1. In all subjects, nocturnal urinary concentration of aMT6s, assessed using the ELISA method (Bühlmann Laboratories, Switzerland), showed abnormally low nighttime levels of the metabolite, below the detection limit of the assay (0.3 ng/ml). Conclusions: These results suggest a moderate incidence of sleep and mood disturbance in pineal surgery patients, with little seasonal variation in mood and behavior and no extreme chronotypes. A larger sample, using direct measurement (polysomnography, psychiatric examination) of potential pathology or other abnormalities, and extra-pineal surgical controls, are needed to ascertain the reliability of these findings, the validity of the self-reports and their specificity to patients lacking endogenous melatonin.