JACQUES LOEB 111 



The solutions of crystalline egg albumin seem to consist (at ordinary 

 temperature and at not too high a concentration of albumin and of 

 the hydrogen ions) exclusively or almost exclusively of isolated mole- 

 cules or ions. Since the latter cannot diffuse through a collodion 

 membrane they give rise to a Donnan equilibrium across the membrane 

 and hence only the osmotic pressure of solutions of salts of crystalline 

 egg albumin is influenced by electrolytes in the way demanded, while 

 the viscosity shows such an influence only to a negligible degree. 



3. Since in connection with viscosity we have assumed the existence 

 of submicroscopic solid particles of jelly in gelatin solutions we must 

 point out where our ideas agree and where they disagree with the 

 speculations on the role of the degree of dispersion current in colloid 

 chemistry. On the basis of these latter speculations it would be 

 assumed that the osmotic pressure of a protein solution is determined 

 directly by the concentration of the protein in solution according to 

 van't Hoff's law. We assume the correctness of van't Hoff's law but 

 the osmotic pressure is only in part determined by the gas pressure 

 of the protein particles, the main source of the osmotic pressure being 

 the unequal distribution of the crystalloidal ions on the opposite 

 sides of the membrane due to the Donnan equilibrium. This has 

 been overlooked by the believers in the dispersion theory. We agree 

 with the believers in the dispersion theory that the melting down of 

 larger aggregates of gelatin ions into isolated gelatin ions must raise 

 the osmotic pressure but disagree in the explanation of this effect; 

 since in our opinion the increase in the number of isolated protein 

 ions must lead to an increase in the distribution of crystalloidal 

 ions on the opposite sides of the membrane according to the Donnan 

 equilibrium and this is the main reason why the osmotic pressure of 

 a gelatin solution is increased when the solid submicroscopic particles 

 of jelly are transformed into isolated ions.^ 



' If we wish to measure the true osmotic pressure of a protein solution, free from 

 the complication of the Donnan equilibrium, we must do so at the isoelectric 

 point of the protein. This condition was approximately fulfilled in the experi- 

 ments of Sorensen (Sorensen, S. P. L., Compt. rend. trav. Lab. Carlsberg, 1915- 

 17, .xii, 262) on solutions of crystalline egg albumin, and of Hiifncr and Gansscr 

 (Hiifner, G., and Gansser, E., Arch. f. Physiol., 1907, 209) on solutions of hemo- 

 globin. 



