THE STABILITY OF PROTEIN SOLUTIONS 267 



Precipitation occurred in (NH^SCK when the concentration of 

 this salt in the casein solution was 2 M. Precipitation occurred 

 in low concentrations of CaCl 2 , namely M/128. In this latter 

 respect the solution of Na caseinate differed from a solution of 

 Na gelatinate in water. The facts indicate that the stability of 

 a solution of Na caseinate in water is not due to a Donnan 

 equilibrium. 



It can be shown that the solution of granules of isoelectric 

 casein in HC1 depends on forces regulated by the Donnan equi- 

 librium and that the rule of Hardy is only a consequence of this 

 fact. This can be proven by microscopic observation of the 

 mechanism of the solution of solid particles of originally isoelectric 

 casein in solutions of acids of different concentration. It was 

 found that the particles of casein swell in a solution of HC1, 

 becoming more and more transparent the more they swell, and 

 that when the swelling has reached a certain stage, the particles 

 disappear they are dissolved. When in the swollen stage, 

 slight agitation may make them fall apart. T. B. Robertson 

 had suggested such a mechanism for the solution t)f Na caseinate, 

 but we have seen that the mechanism of solution in this latter 

 case is different. There is no doubt, however, that the swelling 

 of casein particles is a necessary prerequisite for the solution of 

 casein-acid salts, since such particles are only dissolved when 

 their swelling exceeds a definite limit. 



The method of procedure was as follows: A small number of 

 granules of isoelectric casein of the same size (going through a 

 sieve with mesh 100 but not through a sieve with mesh 120) 

 were put into 50 c.c. of water containing different quantities of 

 different acids and kept at 24C. At various intervals, i.e., 

 after 15, and 60 minutes, and 6, and 24 hours, the diameter of 

 about 15 grains was measured with a micrometer under a micro- 

 scope and the average diameter calculated. The particles were 

 not stirred, and care was taken to avoid their breaking into 

 smaller fragments. The averages after 1 hour are plotted in 

 Fig. 78. The abscissae are the logarithms of the concentrations 

 of acid of the aqueous solution, the ordinates are the average 

 diameters of the particles. It is obvious that the average 

 diameter of the particles increases at first with the increase of 

 the concentration of the acid, reaching a maximum at about pH 



