D. C. TOSTESON 1 33 



time. When ihe K concent ration in the medium is 4 mM/ 1. and the hematocrit 

 about 18, about 90% of the initial plasma radioactivity must be transferred 

 to the cells to reach isotopic equilibrium, while if [K],„ = 74 mM/1. only about 

 one fourth of the initial plasma radioactivity must be transferred. Thus, small 

 errors in the measurement of plasma radioactivity will have more efifect on the 

 derived K llux when [K],„ is high than when it is low. For that reason, the 

 single experiment of Raker et al. cannot be interpreted to rule out a slight 

 increase in K influx with increased [K]m • Our more recent experiments in 

 which K influx was derived from direct measurements of the cell radioactivity 

 indicate that K influx is probably higher at high [K]ni . These points included 

 in figure i give rise to the upward slope of the line relating K influx to [Kjm 

 when the latter is above 5 mM/I. Streeten et al. (103) have also recently ob- 

 served that K influx is increased with increasing [K]n, . Because of the results 

 of Raker et al., they chose to fit their data with an isotherm-type equation. 

 The points could just as well be fitted by a relation such as equation 8. The least 

 square analysis of their data yields the relation, 'jSIk = i-93 + (-056 ± .014) 

 [K]m • The value of D'k = -056 may be compared with the value .017 derived 

 from the flux ratio analysis of their data (K outflux was 1.95 mM, (1. RBC) X 

 (hr.)). Due to the small range of medium concentrations studied ([K],n = 

 4.50 — 16.75 itim/1.), the value of D'k derived from the concentration curve 

 analysis of their data must be considered tentative. Glynn has also recently 

 mentioned evidence for a small diffusion 'leak' of K in the human red cell 

 membrane (24). Our recent measurements of the variation of cell K content, 

 [K]c , after long exposure to a high K concentration in the medium also support 

 the idea that slow diffusion of K occurs in the human red cell membrane 

 (table 3). When [KJm was ca. 50 mM/1., cell K content, expressed as mM/ 

 (that number of cells which occupied i 1. at the start of the experiment), in- 

 creased by 20 in 23 hours at 37°C. When the K concentration in the medium was 

 increased to 130 mM/1. the cells hemolyzed in 24 hours. In all cases the K 

 concentration per volume of cells remained approximately unchanged while 

 [K]c per kg of cell water fell slightly. The value of D'k calculated from the 

 relation, 



^ = iMK + D;[K],„ - ^ [K]e , (P) 



dt 1.2 



was .018, in good agreement with the value obtained from flux ratio analysis. 



Few comparable studies of K outflux as a function of [K]c have been re- 

 ported due to the difficulty in varying the cell K concentration without altering 

 other factors. Such studies as have been done are inconclusive (109). 



Solomon (100) has studied the effect of varying [Xa],„ on sodium influx. 

 He found that the data could be described by the equation: 



»]\lNa = 0.72 + .019 [Na]m 



