The aim of the study was to find out whether there is a daily rhythm in goat serum cortisol concentrations, whether the concentration profiles differ between normal light:dark and constant dark conditions, and whether any seasonal variations might be detected in daily cortisol secretion patterns. Seven Finnish landrace goats were kept at indoor temperature (18-23 degrees C) under artficial lighting that approximately simulated the annual changes of daylength at 60 degrees N. Blood samples were collected for cortisol measurements by radioimmunoassay at 2h intervals during six times of the year: winter (light:dark 6:18h), early spring (10:14h), late spring (14:10h), summer (18:6h), early fall (14:10h), and late fall (10:14h). Cortisol profiles were determined for two consecutive days, first in light:dark (LD) conditions and then in continuous darkness (DD). There was no significant daily rhythm in serum cortisol levels in any time of the year, nor did the profiles in LD and DD conditions show any differences. A significant seasonal variation was, however, detected among the overall cortisol levels. In winter, the concentrations were higher than in any other season, and from early spring to summer they were at their lowest. Under equal photoperiods, the cortisol levels were higher in fall than spring. The difference between winter and summer was confirmed the following year in LD conditions. There was on correlation between serum cortisol and progesterone levels. The results suggest that the possible circadian variation of cortisol secretion in goats is completely masked by external factors, and lighting conditions do not have immediate effects on the daily secretion patterns. The seasonal variation in the overall cortisol levels is most probably related to the changes in photoperiod, because other conditions were relatively constant during the experiment.