762 TRIVALENT AND TETRAVALENT IONS 



albumin for each other becomes greater than the attraction between 

 the molecules of albumin and water. 



If the albumin is practically non-ionized (as is the case at the iso- 

 electric point) no Donnan equilibrium between the nascent micellae 

 and the surrounding solution can be established and no p.d. 

 between the nascent micellae and the solution can prevent the 

 coalescence of the micellae. When, however, part of the albumin 

 is ionized, the molecules of albumin will also unite upon heating 

 to form micellae, but these micellae will begin to repel each other 

 as soon as they contain protein ions. For in this case the protein 

 ions in the nascent micellae will cause the establishment of a 

 Donnan equilibrium between the m.icellae and the solution, and the 

 electrical charge produced thereby on the particles will prevent the 

 further coalescence of the nascent micella. This charge will 

 increase with the relative concentration of ionized protein con- 

 tained in the micellae. It is obvious that the average size of the 

 micellae will remain the smaller the greater the relative concen- 

 tration of protein ions in solution; since the greater the relative 

 concentration of ionized protein the smaller will be the average 

 number of protein molecules which can form an aggregate without 

 including protein ions. This argument is supported by the well 

 known fact that when we add some acid or alkali to isoelectric 

 albumin, the solution will become only opalescent on heating but 

 heat precipitation of the albumin will no longer occur. A compari- 

 son of the effect of increasing concentrations of acid shows that 

 the relative size of the micellae will become the smaller the greater 

 the relative mass of ionized protein. To demonstrate this, 10 cc. 

 of an aqueous 0.2 per cent solution of almost isoelectric crystalline 

 egg albumin and containing varying amounts of 0.1 n HCl were 

 put into test-tubes, and these test-tubes were put into boiling 

 water until the temperature of the albumin solution rose to 90°C. 

 Then the test-tubes were allowed to cool at room temperature and 

 the appearance of the solution was noticed. Table I gives the 

 result. 



When the 10 cc. contained 0.01 cc. of 0.1 n HCl the protein 

 remained practically isoelectric (pH 4.8), practically no ionization 

 was produced, and hence flocculation occurred upon heating. 



