Summary. Potassium currents of various durations were obtained from squid giant axons voltage-clamped in artificial seawater solutions containing sufficient tetrodotoxin to block the sodium conductance completely. From instantaneous potassium currentvoltage relations, the reversal potentials immediately at the end of these currents were determined. On the basis of these reversal potential measurements, the potassium ion concentration gradient across the membrane was shown to decrease as the potassium current duration increased. The kinetics of this change was shown to vary monotonically with the potassium ion efflux across the membrane estimated from the integral over time of the potassium current divided by the Faraday, and to be independent of both the external sodium ion concentration and the presence or absence of membrane series resistance compensation. It was assumed that during outward potassium current flow, potassium ions accumulated in a periaxonal space bounded by the membrane and an external diffusion barrier. A model system was used to describe this accumulation as a continuous function of the membrane currents. On this basis, the mean peria'~onal space thickness and the permeability of the external barrier to K + were found to be 357 A and 3.21 • 10 -4 cm/sec, respectively. In hyperosmotic seawater, the value of the space thickness increased significantly even though the potassium currents were not changed significantly. Values of the resistance in series with the membrane were calculated from the values of the permeability of the external barrier and these values were shown to be roughly equivalent to series resistance values determined by current clamp measurements. Membrane potassium ion conductances were determined as a function of time and voltage. When these were determined from data corrected for the potassium current reversal potential changes, larger maximal potassium conductances were obtained than were obtained using a constant reversal potential. In addition, the potassium conductance turn-on with time at a variety of membrane potentials was shown to be slower when potassium conductance values were obtained using a variable reversal potential than when using a constant reversal potential.

Isolated and cleaned giant axons of Loligo pealii were internally perfused with solutions containing cesium sulfate and potassium fluoride. Membrane currents obtained as a function of clamped membrane potentials indicated a severe depression of the delayed outward current component normally attributed to potassium ion movement. Steady-state currents showed a negative slope in the potential range from -45 to -5 mv which corresponded to the negative slope for the peak sodium current relation vs. membrane potential which suggested long duration sodium currents. Using sodium-free sea water externally, sodium currents were separated from total currents and these persisted for longer times than normal. This result suggested that internal cesium ion delays the sodium conductance turnoff. The separated nonsodium currents showed an abnormal rectification as compared with those predicted by the independence principle, such that while potassium permeability appeared normal at the resting potential, its value decreased progressively with increasing depolarization. Pickard et al. (1964) have recently shown that when sea water in which the sodium was replaced with cesium was applied externally to the isolated squid giant axon the delayed or potassium component of outward membrane current in the voltage clamp was not decreased as might be expected if Cs+ were acting chemically in a way similar to K+ ion. Previous studies on the internally perfused squid axon (Baker et al., 1962) showed that replacing an isotonic K 2 SO 4 perfusion solution with isotonic Cs 2 SO 4 solution caused a rapid decrease in resting potential and an increase in the duration of the action potential elicited during the brief period before inexcitability occurred as a result of the depolarization. These authors were uncertain as to whether the increase in action potential duration was a direct result of the internal Cs+ ion or was an indirect effect resulting from internal K+ ion reduction. More recently, Chandler and Meves (1965), upon voltage-clamping squid axons internally perfused with half Cs and half K isotonic solution, have