Little is known about the effectiveness of delta9-tetrahydrocannabinol (THC) and anandamide in blocking mechanical nociception. Even less is known about their antinociceptive efficacy in chronic inflammatory arthritis induced by Freund's complete adjuvant. The hypothesis was tested that THC and anandamide elicit antinociception in the paw pressure test, and that arthritic rats would exhibit a different response. In nonarthritic rats, THC- and anandamide-induced antinociception lasted 90 min and 15 min, respectively, while antinociception lasted 90 min and 30 min, respectively, in arthritic rats. Area under the curve calculations revealed no effect of arthritis on THC- and anandamide-induced antinociception. Another hypothesis was that paw pressure thresholds in arthritic rats reflect chronic cannabinoid receptor stimulation due to elevations in free anandamide levels. Yet, the CB1 receptor antagonist SR141716A failed to alter paw pressure thresholds in either nonarthritic or arthritic rats. Further investigation revealed that SR141716A significantly blocked THC antinociception, with no effect on anandamide. Thus, anandamide's effects did not result from CB1 receptor stimulation, and any potential contribution of endogenous anandamide in arthritis was not revealed. Finally, THC and anandamide appear to release an as yet unknown endogenous opioid, because naloxone significantly blocked their effects. This study indicates that anandamide and THC may act at different receptor sites to modulate endogenous opioid levels in mechanical nociception.