December 29, 1922] 



SCIENCE 



735 



tant-aric, malic, oitrie, eltc, wliieli below pg 4.7 

 dissociate as monobasic a«ids, give osmotic 

 pressure curves identical with those for HCl 

 and H3PO4. We may, therefore, di-aw the con- 

 elusion that only the valency but not the nature 

 of the acid influences the osmotic pressiu-e of 

 protein solutions, that all acids which are 

 monobasic on the acid side of the isoelectric 

 point of a protein influence its osmotic pres- 

 sure in the same way, and that this influence as 

 considerably greater than the influence of 

 strong dibasic acids like H2SO4. 



If alkali is added to a solution of isoelectric 

 protein it can ibe shown thait the addition of 

 little 'a;lkali increases the osmotic pressure until 

 a maximum is reached when the addition of 

 more alkali depresses the osmotic pressure 

 again. All alkalies with monobasic caition like 

 Li, Na, K, NH4) have ithe same effect at the 

 same pjj, while alkalies and afl dibasic cations 

 like Ca or Ba act alike, the curve for ithe 

 effect of the alkalies with divalenit cation being 

 only aibout half as high as that of the alkalies 

 with donovalent cation. 



A third fact (idiscovei-ed by R. S. Lillie) 

 is thait the addition of salts to a soluition of a 

 protein salt always depresses the osmotic 

 pressure. 



The curves representing the influence of 

 aoids and salts on the osmotic pressure are 

 almost identical or very similar to those repre- 

 senting the influence of the same acids and 

 salts on swelling and viscosity. These results 

 are specific for colloidal behavior 'and any 

 theory of icolloidai behavior will have to give 

 not only a qualitative but a quantitative theory 

 of these curves. 



It was suggested by Zsigmondy thait the 

 influence of aoid on osmotic pressure was due 

 to an influence on the degree of dispersion of 

 the protein in solution, but since the degree of 

 disperaion can not be accuraitely measured7 this 

 suggestion is only a quaJitaitive speculation. 

 But it is of no use even as a qualitative specu- 

 lation since it fails to aceounit for the fact that 

 viscosity and swelling are affected in a simdar 

 way as osmotic pressure. The correct explana- 

 tion is as follo'vvs: When acid (or alkali) is 

 added to a solution of an isoelectric protein, 



part or all of this is ti-ansfoi-med into an 

 ionizable protein salt according to the amount 

 of acid added. This ionization of the protein 

 causes the colloidal behavior on account of the 

 inability of pratein ions to diffuse through 

 membranes which are easily permeable to 

 crystalloidal ions, such as icollodion or parch- 

 ment membranes or the walls of capillaries or 

 probaJbly of all cells. Now it was shown by 

 Donnan that whenever the diffusion of one 

 type of ions such as colloidal ions is prevented 

 by a membrane which is readily permeable to 

 crygtalloidal ions, an unequal distribution of 

 the diffusible crystalloidal ions results on the 

 opposite sides of the memibrane. This unequal 

 disitribution of diffusible crystalloidal ions is 

 the cause of the colloidal behavior of proteins. 

 IV 



When a collodion bag is filled with a solution 

 of gelatin chloride of Pn 3.0 and the bag is 

 immersed in an aqueous solution of HCl also 

 of Ph 3.0 but free from protein, acid is driven 

 from the protein solution into the outside 

 aqueous solution free from protein. Donnan 

 has shown thermodynamicaUy that when os- 

 motic equilibrium is established the products of 

 the concentrations of each pair of oppositely 

 charged diffusible ions (e. g., H and CI in the 

 case of gelatin ehlroide) are equal on the oppo- 

 site sides of the membrane. Let x be the molar 

 coneenh-ation of the H and 01 ions on 'the out- 

 side, y the molar concentration of the free H 

 and CI ions inside the protein solution, and « 

 the concentration of the CI ions in combination 

 with the protein; then equilibrium is defined by 

 the following equation, first used by Procter 

 and Wilson to explain the influence of acid on 

 swelling, 



x2 = y{y -f z) (1) 



The first step in an attempit to explain the 

 influence of acids, alkalies and salts on the 

 osmotic pressure of protein solutions is to find 

 out whether the variations in osmotic pressure 

 under the infiuence of acids as shown in Fig. 2 

 are accompanied by corresponding differences 

 in the concentration of diffusible ions inside 

 and outside the protein solution and whether 

 these differences can be calculated from Don- 

 nan's equilibrium equation (1). 



