The association of seasonal changes in health and disease has been known for centuries. The prevalence of psychopathological symptoms with seasonal fluctuations and the use of melatonin as a biological marker of circadian and circannual rhythms is well documented. The aim of this work was to study the variability of melatonin secretion between summer and winter in our geographical area (28 degrees N, 16 degrees W) and relate the changes to the level of psychopathology. Ten drug-free, nonsmoker, healthy subjects were studied in summer (August) and winter (December). Blood samples for melatonin assays were collected every hour at night for 5 hr, from 22:00 to 02:00 hr, and next day at noon. Melatonin was assayed by an ELISA technique. Psychopathology was evaluated by means of the 28-item version of the General Health Questionnaire (GHQ-28). All subjects had a circadian rhythm of melatonin secretion in summer and winter. There was a seasonal rhythm with melatonin levels being significantly higher at night in winter than in summer. Melatonin levels at 22:00, 23:00, 24:00 and 01:00 hr and mean melatonin area under the curve (AUC) were significantly higher in winter than in summer. Melatonin AUC increased 80% in winter compared with summer. The GHQ-28 somatic and anxiety subscales and the total GHQ-28 score were significantly higher in winter than summer. Psychopathology scores were significantly and negatively correlated with melatonin production in summer and winter. Our data strongly suggest that melatonin production and psychopathology levels present seasonal fluctuations and these variations should be taken into account when conducting research in this field.