SUMMARY1. Fifty to ninety per cent of the Na efflux from axons of Loligo forbesi is inhibited by ouabain. The properties of the ouabain-sensitive component of the Na efflux are different from those of the ouabain-insensitive component.2. In unpoisoned axons with an average Na content of 75 m-mole/kg axoplasm the bulk of the ouabain-sensitive Na efflux is dependent on external K.3. In the presence of 460 mm Na in the external medium, raising the external K concentration from 0 to 100 mm increases the ouabain-sensitive Na efflux along a sigmoid curve which shows signs of saturating at high K concentrations.4. The curve relating ouabain-sensitive K influx to external K concentration is similar in shape to that for the ouabain-sensitive Na efflux. At all K concentrations examined the ouabain-sensitive K influx was less than the ouabain-sensitive Na efflux.5. Potassium-free sea water acts rapidly in reducing the Na efflux. There is no appreciable difference between the rates of action of K-free sea water on the Na pump and Na-free sea water on the action potential.6. Caesium and Rb can replace external K in activating the ouabainsensitive Na efflux. Both the affinity and maximum rate of the Na efflux mechanism are lower when Cs replaces K as the activating cation. P. F. BAKER AND OTHERS 7. Isosmotic replacement of external Na by either choline or dextrose, but not Li, increases the affinity of the ouabain-sensitive Na efflux mechanism for external K without appreciably affecting the maximum rate of pumping. External Li behaves like external Na and exerts an inhibitory action on the Na efflux.8. There is a large ouabain-sensitive Na efflux into K-free choline or dextrose sea waters. Addition of either Na or Li to the external medium reduces this efflux along a section of a rectangular hyperbola. The properties of this efflux suggest that there is a residual K concentration of up to 2 mm immediately external to the pumping sites in the axolemma.9. Over the range of internal Na concentrations studied (16-140 mmole/kg axoplasm) the ouabain-sensitive Na efflux increased linearly with Na concentration.10. Tetrodotoxin (10-6 g/ml.) reduces the Na influx by about half, but does not affect the ouabain-sensitive Na efflux.11. Isobutanol (1 % v/v) reversibly decreases both the ouabain-sensitive and ouabain-insensitive components of the Na efflux.12. Application of 2 mm cyanide to axons immersed in K-free sea water produces a transient rise in the Na efflux. This rise is not seen if ouabain is included in the sea water. The rise in efflux occurs at a time when the axons are partially poisoned and contain adenosine triphosphate (ATP) but no arginine phosphate (ArgP). A similar, but maintained rise can be obtained after application of dinitrophenol (DNP) at pH 8-0. The increased Na efflux in these partially poisoned axons is also inhibited by ouabain.13. Under conditions of partial-poisoning by alkaline DNP, there is a ouabain-sensitive Na influx from K-free sea water. The ouabain-sensitive Na influx is of similar size to the ouab...

This paper is concerned with the effects of altering the composition of the internal fluid on the electrical properties of giant axons from which the bulk of the axoplasm had been removed by extrusion (Baker, Hodgkin & Shaw, 1962). To begin with, we shall consider a series of experiments which show that the difference in potassium concentration between the internal and external fluid provides the main electromotive force for generating the resting potential. In order to reduce the error introduced by junction potentials and to avoid making uncertain estimates of the activity coefficient of K+ in K2S04 it was simplest to work with an internal solution consisting of mixtures of KCI and NaCl. The high concentration of chloride did not have any markedly deleterious effect, for fibres filled with isotonic KCI were usually excitable and had a resting potential only about 5 mV less than in those filled with isotonic potassium sulphate. RESULTSThe effect of replacing internal K by Na on the resting potential Figure 1 shows the time course of the change in membrane potential caused by a temporary replacement of KCI with NaCl. The resting potential was reduced from about -50 mV to zero and returned to within a few millivolts of its original value when KCI was restored. The asymmetry of the curve, i.e. the slow depolarization and rapid repolarization, is explained by the non-linear relation between [K]1 and membrane potential. From curve a in Fig. 2 the potassium concentration at which the resting potential is reduced to half its normal value may be taken as 60 mm. Since this is only 10 % of the initial concentration, depolarization should be much slower than repolarization. The time for complete depolarization probably represents the time required for a flow of about 50 ptl./min to remove 99 % of the original fluid from a total volume of roughly 50 ,ul. (30 ,l. dead

In 1937 Bear, Schmitt & Young showed that substantial quantities of axoplasm could be squeezed out of the cut end of a giant nerve fibre of Loligo. This technique has been widely used for obtaining samples of axoplasm, but little attention has been paid to the electrical properties of the thin sheath which remains after the contents of the nerve fibre have been extruded. Since extrusion involves flattening the axon with a glass rod or roller it is natural to suppose that the membrane would be badly damaged by such a drastic method of removing axoplasm. However, in the autumn of 1960 impulses were recorded from extruded sheaths which had been refilled with isotonic solutions of potassium salts (Baker & Shaw, 1961) and on further investigation it turned out that such preparations gave action potentials of the usual magnitude for several hours (Baker, Hodgkin & Shaw, 1961). Tasaki and his colleagues at Woods Hole have also been successful in perfusing the giant axons of Loligo pealii and several methods, some evidently developed before ours, were described by Oikawa, Spyropoulos, Tasaki & Teorell (1961). We have made no serious attempt to compare different methods and all the experiments described here were carried out by perfusing sheaths from which the bulk of the axoplasm had been removed by extrusion. The first paper deals with technique and with some general properties of perfused fibres, including histology and electron microscopy. The second is concerned with the electrical effects of changing the internal solution and ends with a discussion of both sets of results. METHODS MaterialGiant axons from large specimens of Loligo forbe8i were used throughout the investigation. The mantle lengths of the squid were usually 30-50 cm, and the diameters of the axon 700-900 u. Living squid were used occasionally but as a rule we employed refrigerated mantles. In this technique live squid are decapitated as soon as the trawl is brought up and

In the previous paper (Caldwell, 1960) it was shown that the phosphagen and adenosine triphosphate (ATP) of squid giant axons break down in the presence of fairly high concentrations of inhibitors such as cyanide, dinitrophenol or azide and that some recovery occurs when the inhibitors are removed. Since these agents also reduce the outflow of sodium ions from giant axons to a low value (Hodgkin & Keynes, 1955) it is natural to suppose that ATP or some other energy-rich phosphate compound may provide the energy for running the ionic transport system. This would fit with the generally accepted view as to the function of ATP in linking biochemical and physiological events in cells (Lipmann, 1941). The aim of the experiments described here was to test the assumption by seeing whether injected ATP or arginine phosphate could restore normal ion transport to a fibre poisoned with cyanide or dinitrophenol. The results show that both arginine phosphate and ATP have a restorative action, but that the former is a more effective substitute for normal metabolic activity.

SUMMARY1. A small volume of artificial sea water containing 300 nm tetrodotoxin (TTX) was applied successively to seven lobster nerve trunks and the cumulative uptake of toxin investigated by bio-assay.2. Light and electron microscopy indicated that the nerve trunks had a total axonal area of 0 7 x 104 cm2/g.3. Sodium analysis gave a sodium space for the nerve trunks of 30 %. 4. The amount of toxin taken up by the cells in 1 g of nerve is less than 1*6 x 10-11 moles.5. It is argued that there are probably fewer than 13 sodium channels/ 2 axon in lobster nerve.