MEMBRANE POTENTIALS 147 



these results could be confirmed with crystalline egg albumin. 

 This was found to be the case, and the experiments on the 

 membrane potentials of the solutions of the chloride of crystalline 

 egg albumin showed a perfect quantitative agreement with the 

 theory. 



Collodion bags of about 50 c.c. volume were filled with a solu- 

 tion of 1 per cent crystalline egg albumin containing varying 

 amounts of 0.1 N HC1, and the bags were put, as usual, into 

 beakers containing 350 c.c. of HC1 solutions of different concen- 

 tration but free from albumin. The first two horizontal rows 

 of Table XXII give the amount of 0.1 N HC1 in each solution. 

 The experiments were carried out at a temperature of 24C., 

 and after 22 hours the osmotic pressure, P.D., and pH of inside 

 (albumin) solution and pH of the outside solution were measured. 

 The albumin used was not isoelectric, but since it had been 

 prepared after S0rensen's method it was probably partly am- 

 monium albuminate, with a pH of near 6.0. The table shows that 

 the calculated and observed P.D. agree beautifully (especially on 

 the acid side of the isoelectric point) ; that the P.D. is a minimum 

 near pH 4.70 of the albumin (i.e., near the isoelectric point, which 

 is at pH 4.8), and that the albumin is positively charged on the 

 acid and negatively charged on the alkaline side of the isoelectric 

 point. This is again in harmony with what we should expect 

 on the basis of the Donnan equilibrium. 



The next problem was to determine the influence of the addi- 

 tion of a neutral salt to a solution of the chloride of crystalline 

 egg albumin. A 1 per cent solution of crystalline egg albumin 

 containing 7 c.c. of 0.1 N HC1 in 100 c.c. was made up in various 

 concentrations of NaCl. The collodion bags containing these 

 albumin chloride-NaCl mixtures were dipped into beakers con- 

 taining 350 c.c. of the same concentration of NaCl as that of the 

 albumin solution, and all made up in N/1,000 HC1. The experi- 

 ment was carried out at 24C. and the measurements were made 

 after 22 hours. 



Table XXIII gives the results, which show again a beautiful 

 agreement between calculated and observed P.D. 



We may, therefore, conclude that the P.D. of both gelatin 

 solutions and solutions of crystalline egg albumin separated by a 

 collodion membrane from a watery solution free from protein is 



