158 L. J. MULLINS 



o 



the electron micrographs of Geren and of Gasser are of the order of 300 A 

 thick or less. 



A fourth point should be made regarding the high permeability of cell mem- 

 branes to water (Lucke, Hartline and Ricca, 1939). It might be objected that 

 a cell membrane containing lipid, in which water is only slightly soluble, could 

 not have a high enough water permeability to account for the experimental 

 results. The work of Rideal (1925) has shown, however, that condensed mono- 

 layers of long-chain fatty acids lower the rate of evaporation of water (into a 

 moderately evacuated space) to a value not less than about half that for a 

 clean surface. A continuous lipid film in the cell membrane would therefore 

 present no great barrier to the passage of water. 



The fifth point has to do with the permeability of ions at oil-water interfaces, 

 for if the cell membrane has a continuous lipid phase, any ions traversing the 

 membrane will have to cross into a somewhat hydrophobic medium and out 

 again. J. T. Davies (1950) has studied the diffusion of salts across a water-nitro- 

 benzene interface, obtaining experimental values of the rate at which various 

 alkali halides cross the interface from water to oil. He found permeability con- 

 stants for the whole salts which were in the range between 1.4 X 10~ 10 and 6.2 X 

 10 -8 cm/sec. This indicates that the free-energy barrier between an aqueous 

 phase and an oil phase is itself sufficient to account for permeabilities which 

 would be low enough, in fact too low in this case, to agree with the permeability 

 values for the passage of ions across a cell surface, such as the membrane of the 

 giant axon of the squid. The values for squid (Hodgkins and Katz, 1949) range 

 from 7.2 X 10~ 8 to 1.8 X 10 -6 cm/sec. Thus it is conceivable that the resistance 

 of the squid axon to the passage of ions is due merely to the presence of a free- 

 energy barrier which must be surmounted as they pass into a lipid phase. The 

 existence of membrane potentials can also be accounted for on this basis, as due 

 to differences in the height of the barrier (before any potential difference arose) 

 for the various anions and cations involved. The cell membrane will have two 

 such interfaces, of course, one on each side of the lipid layer, and the net mem- 

 brane potential will be the difference between the two interfacial potentials, 

 which will in general be unequal. Furthermore, if the height of the barrier for a 

 given ion, such as sodium, is different at the two interfaces the system exhibits 

 the phenomenon of active ion transport. Such a difference could be due to the 

 presence of an interfacial chemical reaction involving the ion at one of the inter- 

 faces. The picture of cell membrane potentials as compounded of two inter- 

 facial potentials might also account for the existence of the two separable com- 

 ponents of membrane potential observed in frog muscle. I am referring to Dr. 

 Ling's A and B potentials (1952). Two distinguishable fractions in the resting 

 potential of frog nerve have also been observed by Shanes and Brown (1942) 

 and Lorente de No (1947). This would be possible if the free-energy barriers at 

 one of the interfaces could be changed without affecting those at the other inter- 



