1 In order to address mechanistic differences between arterial vessel types, we have compared the vasorelaxant actions of anandamide in resistance (G3) and conduit (G0) mesenteric arteries. 2 Anandamide produced concentration-dependent relaxations of pre-constricted G3 arteries with a maximal response that was significantly greater than seen in G0. 3 The CB1 receptor selective antagonists SR141716A (100 nm) and AM251 (100 nm) caused reductions in the vasorelaxant responses to anandamide in both arteries. Maximal vasorelaxant responses to anandamide were reduced in both arteries after treatment with capsaicin to deplete sensory neurotransmitters (10 microm for 1 h). 4 Vasorelaxation to anandamide was not affected by the nitric oxide synthase inhibitor NG-nitro-L-arginine methyl ester (L-NAME, 300 microm) in either artery. Only responses in G3 arteries were sensitive to removal of the endothelium. In G3 vessels only, vasorelaxation to anandamide was reduced by inhibition of EDHF activity with a combination of charybdotoxin (100 nm) and apamin (500 nm) in the presence of L-NAME (300 microm) and indomethacin (10 microm). 5 Antagonism of the novel endothelial cannabinoid receptor (O-1918, 1 microm) caused a reduction in the sensitivity to anandamide in G3 but not G0. 6 G3, but not G0, vessels showed a small reduction in vasorelaxant responses to anandamide after inhibition of gap junctional communication with 18alpha-GA (100 microm). 7 These results demonstrate that there are differences in the mechanisms of vasorelaxation to anandamide between conduit and resistance mesenteric arteries. In small resistance vessels, vasorelaxation occurs through stimulation of vanilloid receptors, CB1 receptors, and an endothelial receptor coupled to EDHF release. By contrast, in the larger mesenteric artery, vasorelaxation is almost entirely due to stimulation of vanilloid receptors and CB1 receptors, and is endothelium-independent.