1. We have characterized the in vitro pharmacological profile of putative A2 adenosine antagonists, two non-xanthine compounds, 5-amino-8-(4-fluorobenzyl)-2-(2-furyl)-pyrazolo [4,3-e]-1,2,4-triazolo[1,5-c] pyrimidine (8FB-PTP) and 5-amino-9-chloro-2-(2-furyl 1,2,4-triazolo [1,5-c] quinazoline (CGS 15943), and the xanthine derivative (E)7-methyl-8-(3,4-dimethoxystyryl)-1,3-dipropyl- xanthine (KF 17837). 2. In binding studies on bovine brain, 8FB-PTP was the most potent (Ki = 0.074 nM) and selective (28 fold) drug on A2 receptors, whereas CGS 15943 and KF 17837 exhibited affinity in the low and high nanomolar range, respectively, and showed little selectivity. 3. In functional studies, 8FB-PTP antagonized 5'-N-ethyl-carboxamidoadenosine (NECA)-induced vasorelaxation of bovine coronary artery (pA2 = 7.98) and NECA-induced inhibition of rabbit platelet aggregation (pA2 = 8.20). CGS 15943 showed weak activity in the platelet aggregation model (pA2 = 7.43) and failed to antagonize NECA-induced vasodilatation. KF 17837 was ineffective in both models up to micromolar concentrations. 4. Antagonism of A1-mediated responses was tested versus 2-chloro-N6-cyclopentyladenosine (CCPA) in rat atria. 8FB-PTP and CGS 15943 also antagonized competitively the negative chronotropic response induced by CCPA. Conversely, KF 17837 was unable to reverse A1-mediated responses. 5. 8FB-PTP is a potent and competitive antagonist of responses mediated by A2 adenosine receptors. The data provided a basis to reduce, by further chemical modifications, the affinity at A1 receptor and therefore enhance A2 receptor selectivity.