1. The whole-cell patch clamp technique was employed to study the effects of the beta-agonist isoprenaline (ISO) on the Na(+)=K+ pump current, Ip, in acutely isolated ventricular myocytes from guinea-pig hearts. Propranolol, a beta-adrenergic antagonist, was used to demonstrate that all of the effects of ISO, stimulatory or inhibitory, are mediated by beta-receptors. 2. Below about 150 nM [Ca2+]i, we find that ISO reduces Ip, while above this [Ca2+]i ISO increases Ip. The stimulatory and inhibitory effects of ISO on Ip are independent of either intracellular sodium ([Na+]i) or extracellular potassium ([K+]o). These results suggest that the end-effect of ISO is directly on the maximum pump turnover rate (Vmax) rather than indirectly through changes in [Na+]i or [K+]o or modulatory effects on Na+ or K+ affinity. 3. The maximum effect of ISO increases Ip by 25% when [Ca2+] is buffered at 1.4 microM. A half-maximal effect is reached at roughly 10 nM-ISO and a near-maximal effect by 0.5 microM. 4. The permeabilized patch technique, using amphotericin B (Horn & Marty, 1988; Rae, Cooper, Gates & Watsky, 1991), was employed to minimize changes in the normal second messenger systems and calcium buffers. In these experiments, we used a high intracellular sodium solution (pipette sodium was 50 mM), thus sodium-calcium exchange was depressed and we expected [Ca2+]i to be above 150 nM. ISO increases Ip in these conditions as in the dialysed cells. 5. Our results suggest that beta-stimulation can increase Ip, but only if [Ca2+]i is above about 150 nM. In the beating heart [Ca2+]i rises well above this value during systole and the average [Ca2+]i, which depends on heart rate, is expected to normally be above this level. During beta-stimulation, the increase in Ip along with a concomitant increase in IK (Giles, Nakajima, Ono & Shibata, 1989; Duchatelle-Gourdon, Hartzell & Lagrutta, 1989) helps prevent action potential lengthening and allows an increase in heart rate even in the presence of increased calcium current. Further, beta-stimulation will compensate for the effects on Ip of either hypokalaemia or digitalis toxicity, and so reduce the expected rise in both [Na+]i and [Ca2+]i.