During systemic hypoxia the contribution of adenosine to skeletal muscle vasodilatation is dependent on the presence of NO; NO is required for the release of adenosine from the endothelium (Edmunds et al. 2003) and mediates dilatation via endothelial A₁-receptors (Ray & Marshall, 2005). By contrast, skeletal muscle vasodilatation accompanying muscle contraction (exercise hyperaemia) is mediated by adenosine acting at A_2A-, but not A₁-receptors (Ray & Marshall, 2008). Adenosine can release NO from endothelium by acting at A_2A-receptors (Ray et al. 2002). Thus, we investigated the role of NO in exercise hyperaemia. In three groups of rats, anaesthetized with Saffan (7-12 mgkg^-1hr^-1 I.V.), we recorded arterial blood pressure (ABP), femoral blood flow (FBF) and tension in the extensor digitorum longus. Isometric twitch contractions were evoked by stimulation of the sciatic nerve at 4Hz. Integral femoral vascular conductance (IntFVC) was calculated off-line. Group 1 (n=7) was the time control for, Group 2 (n=10), which received NOS inhibitor L-NAME before the third, and A_2A-receptor antagonist ZM241385, before the fourth contraction. Group 3 (n=12) received L-NAME before the third, the NO-donor SNAP to restore baseline FVC during the fourth and fifth contraction and ZM241385 before the fifth. Time controls showed consistent tension and hyperaemic responses. In Group 2, baseline IntFVC was reduced by L-NAME (0.555±0.04 (mean±SEM) to 0.297±0.02CU*, ANOVA for repeated measures, p<0.001) but not by ZM241385. L-NAME reduced exercise hyperaemia (13.91±1.31 to 9.52±1.09CU*), and it was further attenuated by ZM241385 (to 5.46±1.12CU*). In Group 3, SNAP after L-NAME restored baseline IntFVC to control levels (Control: 0.702±0.09, L-NAME: 0.377±0.05*, L-NAME + SNAP: 0.616±0.09CU); ZM241385 had no further effect. Exercise hyperaemia was also restored to control levels after L-NAME by SNAP (Control: 17.10±1.18, L-NAME: 10.87±1.09*, L-NAME + SNAP: 16.99±1.38CU), and this response was further attenuated by ZM241385 (12.75±0.98CU*). These results confirm that adenosine acting via A_2A-receptors contributes to exercise hyperaemia. However, they indicate that NO is not required for adenosine to be released during contraction, as adenosine acting on A_2A-receptors still contributed to exercise hyperaemia when NOS was inhibited. Further, as the contribution of adenosine acting via A_2A-receptors to exercise hyperaemia was fully restored when tested against a background of NOS inhibition and tonic NO-induced dilatation, it seems that adenosine released during contraction does not depend on new synthesis of NO to produce vasodilatation. We therefore propose that, during muscle contraction adenosine is released from the skeletal muscle fibres independently of NO and acts directly on A_2A-receptors on the vascular smooth muscle to cause vasodilatation.