170 ELECTROLYTES IN BIOLOGICAL SYSTEMS 



selection of one mechanism if it is predominant in the system under study. 

 The possibilities are not mutually exclusive, hence experimental effects may 

 not necessarily be clear-cut as in the table. 



For frog nerve, detailed data such as called for in table 5 are not available. 

 Absolute changes in sodium and potassium concentrations, which represent the 

 differences in the influx and outflux, have been followed and compared with 

 experimental changes in E (45). Since the time course of decline in heat pro- 

 duction (11) reveals that a minimum in energy turnover precedes the minimum 

 in E in Ringer's, possibilities b and c have been discarded. Other observations 

 have led to a selection of d to account for the kinetics of potassium movement 

 (45). This has been found consistent with energy considerations and with the 

 time course of potassium emergence (44). However, these findings do not con- 

 stitute final conclusive proof that only d is operative in frog nerve. 



The requisite direct data for the desheathed toad sciatic nerve are now being 

 accumulated in our laboratory. Measurements of the rate of sodium outflux 

 reveal no change with metabolic inhibition, whereas potassium influx is re- 

 duced to about one third with a less marked increase in its outflux. Prelimi- 

 nary measurements of the resting potential under conditions preventing the 

 accumulation of potassium in the interstitial 'spaces' indicate little altera- 

 tion in E at least during the initial stages of inhibition. In this preparation, 

 therefore, direct transport takes place, but as previously concluded for frog 

 nerve from less direct data (44, 45, and table 2), this involves primarily the 

 potassium ion. Moreover, such transport probably contributes little directly 

 to the resting potential. The available facts appear to conform in part with a 

 system comparable to the CO2 models described by Osterhout (36), in which 

 potassium exchanges with a 'metabolically' generated cation, but contributes 

 to the membrane potential only to the extent that a potassium concentration 

 gradient develops across the cell 'membrane'. The selectivity of the cell for 

 potassium is at least partly dependent on the preference of metabolic transport 

 processes for this ion rather than for sodium. That this may also be true in 

 vertebrate muscle is suggested by the absence of an effect by metabolic inhibi- 

 tion on sodium outflux (27). 



More detailed evidence favoring the fourth mechanism in cephalopod axons 

 has been accumulated by Hodgkin and Keynes but is available only in the form 

 of preliminary notes and a personal communication (22, 23). In preparations 

 which had been previously stimulated to augment sodium uptake, inhibitors 

 such as dinitrophenol and cyanide greatly reduce sodium outflux with a rela- 

 tively small decrease in the influx; the same inhibitors greatly reduce potassium 

 influx. Low temperature also decreases potassium influx substantially, leaving 

 the outflux little changed. Under these conditions E is decreased only slightly. 

 These data therefore constitute conclusive proof for direct transport mecha- 



