Diurnal variations in tonic pain reactions have been described in mice tested in Spring, but the underlying mechanisms are still unknown. We tested the potential role of melatonin, a key hormone in the control of neuro-endocrine circadian rhythms. The experiments were performed in male CBA/J mice housed under controlled temperature, humidity, and light (12/12 dark/light cycle) conditions, during the Light (7-10a.m.) or Dark (7-10p.m.) phases of the diurnal cycle. In a first group of experiments, animals were either pretreated with i.p. saline (controls) or with the melatonin receptor antagonist, luzindole (30 mg/kg), before the s.c. injection of a dilute formalin solution into a hindpaw. In control animals, pain-related behavioral reactions (licking and flinching) were higher in the evening (Dark) than in the morning (Light), both during the first (0-10 min) and the second (11-55 min) phase of the response to s.c. formalin. In animals pre-treated with luzindole, no diurnal changes occurred, pain reactions in the Dark being similar to those of the Light Control group. In a second group of experiments, artificial pinealectomy, obtained by exposing animals to continuous light for 48 h, also reduced pain reactions in the evening to levels comparable to those in the morning. Receptor autoradiography showed lower binding availability at spinal cord level in mice sacrificed during the Dark, as expected from the circadian pattern of melatonin secretion. A further significant decrease of melatonin receptor binding was induced by noxious stimulation. These results suggest a proalgesic role of endogenous melatonin in tonic pain.