The serum melatonin concentration was measured in 11 runners before and after a long-distance race, and during one menstrual cycle of another 11 runners who were performing a strenuous training programme, and in their 12 control subjects. The concentration of melatonin was significantly increased (p less than 0.05) after the competition, whereas during one menstrual cycle of female runners it did not differ from the corresponding concentrations of non-athletes. Training did not cause any chronic changes in melatonin secretion, and the serum melatonin increases following the physical exercise remained clearly below the nocturnal level. Hence, the antigonadotropic effect of endurance running activity is very likely not a melatonin-associated phenomenon.