80 DONNAN EQUILIBRIUM AND VISCOSITY 



This assumption of a second type of viscosity seems unnecessary 

 since it is possible to account for the viscosities of protein solutions 

 on the basis of Einstein's law when the relative volume occupied by 

 the protein in solution is small, and on the basis of Arrhenius's formula 

 when the volume exceeds the limits within which Einstein's formula 

 holds. According to our view the former is true when the protein 

 exists in the solution exclusively or almost exclusively in the form of 

 isolated molecules or ions or particles too small to occlude water 

 and this seems to be the case for solutions of crystalline egg albumin, 

 at ordinary temperature, at a pH above 1 .0 and when the concentration 

 is not excessive. 



The viscosity of such protein solutions is not only of a low order 

 of magnitude but is not influenced by electrolytes in the way as is, 

 e.g., their osmotic pressure. Where we have viscosities of a higher 

 order of magnitude, as in the case of gelatin solutions, we notice 

 that the viscosity is influenced by electrolytes in the same way as 

 is the osmotic pressure of these solutions. In the case of such proteins 

 we assume that both the high order of viscosity as well as the character- 

 istic influence of electrolytes on the viscosity are due to the same cause ; 

 namely, the existence in the gelatin solution of a certain number of 

 submicroscopic particles of solid jelly occluding large masses of 

 water, the exact amount of which is regulated by the Donnan 

 equilibrium. The occlusion of large masses of water increases the 

 relative volume occupied by the gelatin in solution so that Einstein's 

 formula is no longer applicable. We are dealing, however, in both 

 cases with the same type of viscosity which is primarily a function 

 of the relative volume occupied by the protein particles in the solution. 



Measurements of the influence of concentration on the viscosity 

 of solutions of crystalline egg albumin of pH 5.1, i.e., quite near the 

 isoelectric point, were carried out at 15°C. and showed that the viscosity 

 is under these conditions practically a linear function of the concen- 



tration (Fig. 4). If Einstein's formula— = 1 + 2.5 v? can account 



for the measurements, it should be possible to calculate the volume 

 (p of the albumin in 100 cc. of solution from 



^=£-)^ 



nl2? 



