1Presynaptic modulation of noradrenaline release in human atrial tissue specimens was investigated under normoxic and anoxic conditions. 2 Noradrenaline release was induced by electrical stimulation and release during experimental intervention (S2) was compared with release during a preceding control stimulation (S5). The results were expressed as the geometric means and 95% confidence intervals of the S2/Sj ratio. 3 The a2-adrenoceptor agonist, UK 14304 (0.1 ymol -') significantly inhibited noradrenaline release, resulting in a S2/S, ratio of 0.49 (0.40-0.59), and the a2-adrenoceptor antagonist, yohimbine(1 ymol I-') increased noradrenaline release (S2/S, 1.83 [1.43-2.35]) during normoxia. Both compounds were ineffective during anoxia.4 Adenosine (30 ymol -') inhibited noradrenaline release with a S2/S, ratio of 0.54 (0.42-0.66). The adenosine antagonist, 8-phenyltheophylline, alone had no effect during normoxia. During anoxia, neither adenosine nor 8-phenyltheophylline altered noradrenaline release.5 The /32-adrenoceptor agonist, terbutaline (1 ymol 1-1) increased (1.53 [1.14-2.01]) and the ,Badrenoceptor antagonist, pindolol (1 umol 1-') suppressed noradrenaline release (0.62 [0.49-0.79]) under normoxic conditions. During anoxia, pindolol significantly inhibited noradrenaline release with a S2/S, ratio of 0.66 (0.51-0.85), whereas terbutaline did not influence noradrenaline release. 6 Angiotensin 11 (0.1 Mmol 11) enhanced noradrenaline release resulting in a S2/S, ratio of 1.44 (1.34-1.54), while the angiotensin II antagonist, losartan (1 ymol 1-') had no effect on noradrenaline release during normoxia. Conversely, angiotensin II did not increase noradrenaline release and losartan significantly inhibited noradrenaline release to a S2/S, ratio of 0.60 (0.46-0.77) during anoxia. 7 In conclusion, human cardiac tissue possesses presynaptic inhibitory a2-adrenoceptors and adenosine receptors, as well as facilitatory fi2-adrenoceptors and angiotensin II receptors regulating noradrenaline release under normoxic conditions. During anoxia the responses to a2-adrenoceptors and adenosine receptor stimulation are lost, whereas facilitatory responses to f2-adrenoceptors and angiotensin II receptor stimulation are maintained and these receptors appear to be maximally stimulated. This differential presynaptic modulation in anoxia may contribute to enhanced sympathetic activity in ischaemia.