Recent work has shown that chronic exercise is associated with a reduction in the pain-relieving actions of opioid drugs in experimental animals. To determine whether this reduction represents an interaction between exogenously administered opioids and the endogenous opioid system, or is the result of altered drug pharmacokinetics, the antinociceptive actions of morphine and its metabolite, morphine-6-glucuronide (M6G), were compared in active and inactive female Long-Evans rats. Active animals were housed in running wheels and inactive animals in standard laboratory cages for 3 weeks preceding determinations of antinociception using the tail-flick test. At the end of the 3-week period, active rats were running the equivalent of 9-11 km a day. Antinociceptive responses, determined following subcutaneous injections of either morphine (0.625-20 mg/kg) or M6G (0.3-10.0 mg/kg), were significantly reduced in active rats relative to inactive rats. This reduction was manifested by both a lower magnitude of antinociception, and a shorter duration of antinociception after drug administration in active compared to inactive rats. This reduction was not associated with alterations in the estrous cycle or with differences in body weight between the active and inactive animals. The present results support the hypothesis that cross-tolerance develops between endogenous opioid peptides released in response to exercise and exogenously administered opioid drugs.