228 THEORY OF COLLOIDAL BEHAVIOR 



which give us , aird we can calculate the volume from the 



'no 



measured volume of the sediment plus the calculated volume of 

 the casein in the supernatant liquid. The latter value is obtained 

 by deducting the dry weight of the sediment from the (known) 

 dry weight of the whole mass of casein put into the water (1 gm. 

 powdered casein, dry weight = 0.87 gm.); assuming that the 

 casein in the supernatant liquid consists exclusively of suspended 

 particles. This is partly correct for a 1-hour experiment at 20. 

 The ordinates in Fig. 67 represent the values for volume thus 



corrected and the values for log while the abscissae are the pH 



"no 



of the suspensions. The two curves are almost parallel. 



It should be stated that these corrected volumes of casein 

 include a certain amount of water between the granules. We 

 are, however, in this case not concerned with the absolute but 

 only the relative volume occupied by the casein. 



When NaCl is added in different concentrations to a casein 

 chloride solution it is noticed that the viscosity is diminished 

 as it is in the case of solutions of gelatin chloride. We shall see 

 in Chap. XV that this diminution of viscosity is accompanied by 

 a diminution in the degree of swelling of the individual particles 

 of casein which is parallel to the depression of the viscosity. 



One gram of powdered casein was put into 100 c.c. of H 2 O 

 containing 12.5 c.c. of 0.1 N HC1, and NaCl in concentrations 

 varying from to M/4. The mixture was shaken occasionally 

 and kept for 16 hours at 20. Then the viscosity, volume of 

 sediment (after settling for 24 hours), dry weight of sediment 

 (after deduction of the free NaCl contained in the sediment) 

 were determined. When the volume and the values for log 



are plotted as ordinates over the concentrations as abscissae, 



7 



it is found that the two curves agree fairly well (Fig. 68) except 



where no or little salt was added and where therefore some casein 

 particles had been completely dissolved. In this solution the 

 calculated volume was too high and our curves express the fact. 

 From these experiments we may conclude that the influence of 

 electrolytes on the viscosity of casein solutions or suspensions is 

 due to the swelling of particles of casein suspended in the solu- 



