Flow-mediated vasodilation (FMD) in the brachial artery measured by ultrasound, and the increase in forearm blood flow (FBF) induced by local infusion of a muscarinic-receptor agonist have both frequently been used to evaluate endothelium-dependent vasodilation (EDV) in the human forearm. The present study intended to evaluate the relationship between these techniques and to investigate if vasodilation induced by the muscarinic receptor-agonist methacholine (MCh) was owing to production of nitric oxide (NO). FMD during hyperaemia was assessed by ultrasound and FBF was measured by venous occlusion plethysmography during local infusion of MCh or L-arginine in the human forearm. Both these methods were applied in 26 individuals. In another 12 individuals forearm arterial and venous plasma concentrations of nitrate/nitrite (NOx) were measured together with FBF before and during local MCh infusion. While the change in brachial artery diameter induced by sublingually given nitroglycerine and the vasodilatory response to sodium nitroprusside (SNP) given locally in the forearm were significantly correlated (r = 0.70, P < 0.01), FMD showed no relationship with the vasodilation evoked by MCh (r = -0.03) or L-arginine (r = 0.04). The five-fold increase in FBF during MCh infusion was associated with a significant increase in venous plasma NOx concentrations (P < 0.05) and a more than 11-fold increase in forearm NOx-release (P < 0.01). Thus, a significant relationship between the two methods regarding the evaluation of endothelium-independent vasodilation evoked by NO-donors was found, but no relationship was found between the two methods regarding the evaluation of endothelium-dependent vasodilation. Furthermore, vasodilation induced by MCh in the forearm seems to be induced by NO-release.