There has been recent debate regarding the labile nature of stonefish venoms and the pharmacology of their breakdown products. The present study examined the cardiovascular and neuromuscular effects of lyophilised venom, and conducted a preliminary investigation of freshly milked venom. Lyophilised venom (20 microg/ml) caused endothelium-dependent relaxation in rat aortae that was abolished by atropine (0.1 microM). In contrast, an endothelium-independent contractile response occurred in porcine coronary arteries. However, in the presence of atropine (10 nM), this became a relaxation response which was attenuated by the B2 antagonist FR-173657 (0.1 microM) or by a combination of idazoxan (1 microM) and propranolol (1 microM). In rat isolated atria, lyophilised venom (4 microg/ml) caused a biphasic inotropic response consisting of an initial decrease, and then increase, in force which were attenuated by atropine (0.5 microM) and propranolol (5 microM), respectively. The increase in force produced by venom was unaffected by reserpine pre-treatment suggesting a direct action at adrenoceptors. In the anaesthetised rat, lyophilised venom (1-300 microg/kg, i.v.), caused a dose-dependent depressor response, with a subsequent pressor response at higher concentrations (30-300 microg/kg, i.v.). In the presence of atropine (1 mg/kg, i.v.), the depressor response to venom was abolished, a transient pressor response unmasked and the secondary pressor response augmented. In the additional presence of prazosin (50 microg/kg, i.v.), the transient pressor response was abolished and the secondary pressor response attenuated. Lyophilised venom had no significant effect on nerve-evoked (10 microg/ml) or directly-evoked (100 microg/ml) twitches of the chick biventer cervicis muscle preparation. Milked venom (1 microl/ml) caused a biphasic response (i.e., an initial relaxation followed by contraction) in rat aortae, a contraction in porcine coronary arteries, complete cessation of rat isolated atrial activity and markedly inhibited both nerve-evoked and directly-evoked twitches of the chick biventer cervicis muscle preparation. In the anaesthetised rat, milked venom (15 microl/kg, i.v.) caused immediate cardiovascular collapse. It appears that the cardiovascular effects of stonefish venom are mediated by a dose-dependent action at muscarinic receptors and adrenoceptors.