224 THEORY OF COLLOIDAL BEHAVIOR 



sion of the variation of the degree of swelling of the casein 

 particles, not as much having gone into solution as after 22 

 hours. We notice that the volume occupied by the solid particles 

 in the 1-hour curve is a minimum at the isoelectric point, that it 

 rises steeply after pH 3.1, that it drops at 2.2, and that a second 

 drop commences at pH 1.8. The two drops have a different 

 cause. The drop at pH 1.8 is due to a diminution of the degree 

 of swelling of the sediment, while the drop at 2.2 in the 1-hour 

 curve is due to the fact that at pH 2.2, where the solubility of 

 casein chloride is a maximum, some of the casein chloride has gone 

 into solution. This conclusion is supported by the fact that the 

 drop at 2.2 increases in time and is very considerable after 22 

 hours (see Fig. 63), while otherwise the 1-hour and the 22-hour 

 curves show only minor differences. 



The proof that this interpretation in the volume curves of Fig. 

 63 is correct is furnished by Fig. 64, where the ordinates are the 

 dry weights of the sediments, the volumes of which are given in 

 Fig. 63. One gram of powdered casein had when dried for 24 

 hours at between 90 and 100C. a dry weight of 0.87 gm. 



That part of the casein chloride which goes into the supernatant 

 liquid (i.e., which is not contained in the sediment) consists of 

 two constituents, namely, first, solid submicroscopic particles in 

 suspension which in due time would have settled, and second, 

 isolated casein ions and molecules. The solid particles in the 

 supernatant liquid (unless they are below the limit required to 

 occlude water) undergo the same swelling under the influence of 

 the Donnan equilibrium as the particles of the sediment. In 

 addition, however, we have individual casein ions in solution (the 

 molecules being probably insoluble since isoelectric casein is 

 practically insoluble) but these ions cannot undergo any swelling 

 and hence do not add materially to the volume and the viscosity. 

 As a consequence, the more solid particles of casein chloride are 

 dissolved into isolated casein ions or particles too small to occlude 

 water, the more the relative volume occupied by the casein in 

 the solution should be diminished, and this should be accompanied 

 by a diminution in viscosity. If our theory of the origin of the 

 viscosity of the gelatin solutions is correct, it should be possible 

 to prove that where the solubility of the casein chloride solution 

 is a maximum the viscosity curve shows a drop. 



