We have studied current (IStr) through the Na,K pump in amphibian oocytes under conditions designed to minimize parallel undesired currents. Specifically, IStr was measured as the strophanthidin-sensitive current in the presence of Ba2+, Cd2+ and gluconate (in place of external Cl-). In addition, IStr was studied only after the difference currents from successive applications and washouts of strophanthidin (Str) were reproducible. The dose-response relationship to Str in four oocytes displayed a mean K0.5 of 0.4 microM, with 2-5 microM producing 84-93% pump block. From baseline data with 12 Na(+)-preloaded oocytes, voltage clamped in the range [-170, +50 mV] with and without 2-5 microM Str, the average IStr depended directly on Vm up to a plateau at 0 mV with interpolated zero current at -165 mV. In three oocytes, lowering the external [Na+] markedly decreased the voltage sensitivity of Ip, while producing only a small change in the maximal outward IStr. In contrast, decreasing the external [K+] from 25 to 2.5 mM reduced IStr at 0 mV without substantially affecting its voltage dependence. At K+ concentrations of less than 1 mM, both the absolute value of IStr at 0 mV and the slope conductance were reduced. In eight oocytes, the activation of the averaged IStr by [K+]0 over the voltage interval [-30, +30 mV] was well fit by the Hill equation, with K' = 1.7 +/- 0.4 mM and nH (the minimum number of K+ binding sites) = 1.7 +/- 0.4. The results unequivocally establish that the cardiotonic-sensitive current of Rana oocytes displays only a positive slope conductance for [K+]0 greater than 1 mM. There is therefore no need to postulate more than one voltage-sensitive step in the cycling of the Na, K pump under physiologic conditions. The effects of varying external Na+ and K+ are consistent with results obtained in other tissues and may reflect an ion-well effect.