1. A whole-cell voltage-clamp technique has been used to study the alpha-effects of the adrenergic agonists noradrenaline, methoxamine and phenylephrine on the action potentials and membrane currents of rabbit atrial myocytes. Experiments were carried out at 22-23 degrees C. 2. In the presence of 10(-6) M-propranolol, all three agents prolonged action potential duration. This change could be ascribed principally to changes in membrane current early during the plateau phase of the action potential. In the presence of 10(-3) M-4-aminopyridine, no changes in calcium current (ICa) were observed on exposure to alpha-agonists. No significant shift in the voltage dependence or change in the amplitude of the calcium current-voltage relation was observed. 3. Exposure to 3 x 10(-4) M-CdCl2 to block ICa reduced the action potential prolongation caused by alpha-adrenergic agonists. Measurement of unloaded cell shortening revealed that action potential prolongation caused by alpha-agonists, especially at low stimulus rates, could contribute significantly to the positive inotropic effect of alpha-adrenoceptor stimulation. 4. The voltage-activated transient outward current (It) was markedly reduced during exposure to alpha-adrenergic agonists in a dose-dependent manner in the presence of CdCl2 (3 x 10(-4) M) and propranolol in sufficient concentration to prevent beta-adrenoceptor activation. Noradrenaline exhibited a higher potency for this effect than either methoxamine or phenylephrine. The noradrenaline concentration required to give 50% of the maximal effect was 6 x 10(-6) M compared with 2.3 x 10(-4) M for methoxamine. Noradrenaline reduced It by only about 60% of the maximum reduction produced by methoxamine suggesting that it could be classified as a partial agonist for this effect. 5. The reduction of It during exposure to alpha-adrenergic agonists was rate dependent in that larger current reductions were observed at very low rates of stimulation (less than 0.1 Hz). 6. The magnitudes of current-voltage relations for It were reduced over the entire voltage range studied during exposure to alpha-adrenergic agonists and reductions were dose dependent. No shift of these relations along the voltage axis was observed. 7. The steady-state inactivation relations for It were studied using two voltage clamp protocols. A two-step method resulted in a relatively steep sigmoid 'quasi-steady-state' relation. The half-inactivation potential of -27 mV was unaffected by alpha-adrenergic agonists.(ABSTRACT TRUNCATED AT 400 WORDS)