Agmatine-cannabinoid interactions are supported by the close association between cannabinoid CB(1) receptors and agmatine immunoreactive neurons and evidence that shared brain mechanisms underlie the pharmacological effects of agmatine and cannabinoid agonists. In the present study, we used the hot-plate assay of thermal nociception to determine if agmatine alters cannabinoid action through activation of imidazoline sites and/or alpha(2)-adrenoceptors. WIN 55212-2 (1, 2 or 3 mg/kg, i.p.) or CP55,940 (1, 2 or 3 mg/kg, i.p.) administration increased hot-plate response latency. Agmatine (50 or 100 mg/kg, i.p.) was ineffective. Administration of agmatine (50 mg/kg, i.p.) with WIN 55212-2 (1, 2 or 3 mg/kg, i.p.) or CP55,940 (1, 2 or 3 mg/kg, i.p.) produced response-latency enhancement. Regression analysis indicated that agmatine increased the potency of WIN 55212-2 and CP55,940 by 3- and 4.4-fold, respectively, indicating synergy for both drug interactions. Idazoxan, a mixed imidazoline site/alpha(2)-adrenoceptor antagonist, but not yohimbine (5 mg/kg, i.p.), a selective alphia(2)-adrenoceptor antagonist, blocked response-latency enhancement produced by a combination of WIN 55212-2 (2 mg/kg) and agmatine. Response-latency enhancement produced by WIN 55212-2 (2 mg/kg) was blocked by SR 141716A (5 mg/kg, i.p.), a cannabinoid CB(1) receptor antagonist; attenuated by idazoxan (2 and 5 mg/kg); and not affected by yohimbine (5 mg/kg). These results demonstrate a synergistic interaction between agmatine and cannabinoid agonists and suggest that agmatine administration enhances cannabinoid action in vivo.