Newborn or adult rats and mice were treated with capsaicin. The effect of systemic or intrathecal treatment on thermonociception, chemonociception, content and release of immunoreactive substance P (I-SP) was investigated. Treatment of two day old rats caused a small, but life-long elevation of the hot plate or tail withdrawal latency. Treatment of adult rats led to a large increase in the reaction time on the hot plate for 4--10 days but the tail withdrawal latency was only slightly elevated for not more than 1--2 days. Mice treated on the 2nd day of life had normal reaction times on the hot plate and a small and inconsistent prolongation of the tail withdrawal latency. In contrast, mice treated on day 7, 10 or as adults had greatly prolonged latencies in both tests for at least 3 months. The changes in latencies were not affected by naloxone or methysergide. Responses to noxious chemical stimuli were moderately inhibited in mice treated on the 2nd day of life, but almost abolished in mice treated on day 7, 10 or as adults. Neonatal capsaicin treatment of rats resulted in a depletion of I-SP in spinal cord and sciatic nerve for 20 months. Capsaicin-evoked release of I-SP from rat spinal cord was reduced by 93% after neonatal treatment, but only by 69% 2 weeks after adult treatment. Treatment of mice on day 2 caused a similar decrease of the I-SP content in spinal cord and of the capsaicin-evoked I-SP release (88%) as treatment on day 4 or 7 although behavioral changes were different. After treatment of adult mice release of I-SP was reduced by 93%. Capsaicin administered intrathecally to rats or mice depleted I-SP in the spinal cord but not in the sciatic nerve. The animals were almost insensitive to noxious heat (tail withdrawal test) and to local application of mustard oil or capsaicin to the hindpaw. Chemosensitivity of the eye, however, remained unchanged. The experiments indicate that systemic or intrathecal capsaicin treatment of rats or mice affects thermo- and chemonociception but species differences were found. It appears, furthermore, that changes in substance P alone cannot explain all the observed behavioral effects after capsaicin treatment.