D. C. TOSTESON 149 



PEP to that of KCl is the same as the activity coefficient ratio of Na PEP to 

 XaCl. Snell (Biochem. et. Biophys. Acta, 10: 188, 1953) has made similar 

 measurements of several organic phosphate compounds, none of which showed 

 a selective affinity for Xa or K. 



MODEL SYSTEM 



In concluding, we shall discuss some model systems of relevance to cation 

 transport in red cells. First we shall treat artificial membranes which have 

 cation diffusion characteristics comparable to that exhibited by red cells. 

 Then we shall mention theoretical and experimental work on artificial mem- 

 branes which display cation selectivity (e.g. K over Na or vice versa). 



The plasma membrane of red cells allows rapid passage of chloride, bi- 

 carbonate and other small anions. On the other hand, as noted above, the 

 diffusion of cations through the membrane, if it occurs at all, is very slow. 

 The ratio of the rate constant for chloride influx to that for K influx by diffusion 

 is of the order of 10^ in human red cells. This behavior is comparable to that 

 shown by positively charged perm- selective membranes (99). The fixed positive 

 charges in such membranes are generally quaternary ammonium groups. The 

 selectivity of such membranes for anions over cations is of the order of 15 X 

 10-, somewhat less than that exhibited by the human cell membrane (M. 

 Gottlieb, personal communication). 



The general problem of the transport of ions in artificial membranes con- 

 taining fixed charges has recently been treated theoretically, and to some 

 extent experimentally, by several workers (e.g. 67, 90, 98, 99). However, 

 most of the experiments in which tracer diffusion rates rather than electrical 

 potential differences and concentrations were measured, deal with the critical 

 ion (i.e. the ion of opposite charge to the fixed charges) rather than the non- 

 critical ion. The relevance of these observations to cation diffusion in red cells 

 is therefore questionable since the cation is analagous to the non-critical ion 

 in this case. Nevertheless, it is interesting to note that the rate of diffusion of 

 the critical ion through a charged membrane is a function of the concentration 

 in the outside solution only at very low and very high concentrations (67, 98). 

 In the former case, the rate limiting step is diffusion across the unstirred water 

 layer adhering to the surface of the membrane, and in the latter case a sig- 

 nificant amount of the non-critical ion penetrates the membrane. At inter- 

 mediate concentrations, diffusion of the critical ion across the membrane is 

 approximately a zero order process. 



More pertinent to the problem of cation diffusion in red cells are a few 

 recent measurements of non-critical ion diffusion across perm-selective mem- 

 branes (67). In these studies, the rate of diffusion of chloride in exchange for 

 nitrate across a cation permeable membrane separating equal concentrations 

 of KCl and KNO3 was measured. The rate of exchange was found to increase 



