SUMMARY1. In dogs anaesthetized with sodium pentobarbitone and artificially ventilated, the gracilis muscles were vascularly isolated and perfused at a constant flow of 28-4 + 46 ml min-1 (100 g muscle tissue)-' (99-8 + 45 % of maximum free flow, means+ standard error of the mean (S.E.M.), n = 9).2. Three to five minutes of electrical stimulation of the cut peripheral end of the obturator nerve (4 Hz, 6 V, 0-2 ms) resulted in muscle contraction (0-61 +0-14 kg (100 g)-f during solvent infusion and 0-56+0 10 kg (100 g)-1 during intra-arterial adenosine deaminase infusion (50 U min-') and an immediate decrease in arterial perfusion pressure from 184-5 + 8-1 mmHg to 148-2 + 57 mmHg (187 + 3-4% decrease) during solvent infusion, and from 193-5 + 716 to 1420 + 10-2 mmHg (25-4 + 6-1 % decrease) during adenosine deaminase infusion 10 s after the commencement of muscle stimulation. After about 5 min of muscle contractions, the arterial perfusion pressure decreased to 120-8 + 7-8 mmHg (32-9 + 5-8 % decrease) during solvent infusion, and to 152-8 +11-2 mmHg (20-9 + 5-3 % decrease) during adenosine deaminase infusion (i.e. 37.9 + 6-2 % attenuation of the fall in arterial perfusion pressure). The time taken for 90% recovery of the arterial perfusion pressure was 72-1 + 10 9 s during solvent infusion, and 51.5 + 9-3 s during adenosine deaminase infusion (P < 0-05).3. Adenosine (2 x 10-3 mol I1) infusion in the resting muscle during solvent infusion (final concentration in arterial blood 1-3 x 10-4 + 6-0 x 10-5 mol l-l) resulted in a 34-8 + 7-2 % fall in arterial perfusion pressure but a fall of only 7-2 + 1-8 % during adenosine deaminase infusion (50 U min-'; P < 0 05; n = 5) indicating that adenosine deaminase infused at 50 U min-' was more than adequate to metabolize endogenous adenosine produced during muscle contractions. -4. These data suggest that adenosine contributes about 40% to the sustainedexercise vasodilatation under constant high-flow conditions and also in post-exercise vasodilatation, but does not contribute to the initiation of exercise vasodilatation.

1. The aim of this investigation was to determine the contribution of adenosine to coronary active hyperaemia in the dog denervated heart by using adenosine deaminase. 2. Beagles were anaesthetized with thiopentone sodium (500 mg, i.v.) and chloralose (100 mg kg-', i.v.) and artificially ventilated. The hearts were denervated by bilateral cervical vagotomy and cardiac sympathectomy. Blood samples were collected from the coronary sinus via a cannula passed through the right external jugular vein. The anterior descending or circumflex branch of the left coronary artery was cannulated and perfused with blood from the left subclavian artery under systemic blood pressure through an electromagnetic flow probe and a perfusion circuit. The heart was paced (3 V, 0 2 ms and a suitable frequency) via two electrodes attached to the right atrium from 109 + 7-3 to 170 + 9 8 beats min-' (means + S.E.M.) for 3-4 min, first during an infusion of the solvent, and then during an infusion of a solution of adenosine deaminase (5 U kg-' min-') into the circuit. 3. In seventeen tests in eight dogs, infusion of adenosine deaminase did not cause a significant change in the basal coronary blood flow nor in the immediate increase (within 10 s) in blood flow induced by pacing the heart from its basal rate to 170 beats min-'.However, adenosine deaminase did cause a significant attenuation by 58 + 5 2 % (P < 0 05) of the increase in coronary blood flow induced at 3-4 min of pacing from 31 + 4-6 to 43 + 5 8 ml min-' (100 g cardiac tissue)-'. Concomitantly, the pacing-induced increase in coronary vascular conductance (from 0-41 + 0-08 to 0 54 + 0-12 ml min-' (100 g)' mmHg-') was reduced by 75 + 6-6% (P < 0 02) and the increase in myocardial 02 consumption (from 13 + 3 5 to 21 + 4 2 ml min-' (100 g)-) was reduced by 50 + 12% (P < 0 05) but without significant changes in oxygen extraction or myocardial contractility. 4. The results show that although adenosine is unlikely to play a significant role in the regulation of the basal coronary blood flow, it can play a major role in the coronary active (functional) hyperaemia induced by atrial pacing to a high rate in the denervated heart of anaesthetized dogs.