ee —— Eee lee 
ON COLLOID CHEMISTRY AND ITS INDUSTRIAL APPLICATIONS. 139 
equation to one of the higher orders of parabolas. This is the reason 
why it is so difficult to reméve, by repeated changes of water or 
dialysis, the last traces of electrolytes attached to colloids or coarsely 
heterogeneous systems. 
This difficulty of removing all electrolytes and other crystalloids 
from a colloidal solution has led some workers in the past to the 
belief that the osmotic pressure shown by some of these solutions 
was due to the electrolytes present. There are several reasons, theo- 
retical and experimental, that make this view inadmissible. In many 
cases, such as gelatin and congo-red, the more effectively these 
impurities are removed, the higher is the osmotic pressure. If the 
crystalloid is in any form of combination with the colloidal particle, 
chemical or by adsorption, it forms one indivisible system with it, so 
that if a colloidal particle is too large to possess the kinetic energy 
requisite to give an osmotic pressure, it would be even less able to do 
so if attached to another molecule. In fact, the association of a 
colloidal particle with an electrolyte would decrease the osmotic 
pressure given by it. If, on the other hand, the electrolytes are free, 
they must be removed by repeated dialysis. ‘the osmotic pressure of 
hemoglobin (Hufner and Gansser, 1907) is that which it should have 
in accordance with its molecular weight as determined by chemical 
analysis. It is true that certain colloids, such as ferric hydroxide, 
become unstable if the last traces of ferric chloride are removed, but 
the particle consists of a complex of a variable number of ferric 
hydroxide molecules in adsorption with one or more ferric chloride 
molecules. The latter are not free. 
The greater number of the colloids of biochemical interest belong 
to the class called emulsoids. Their characteristic is that the two 
phases are not solid and liquid, but both contain solvent in different 
amounts. The two phases in these biochemical colloids may be 
described as being, on the one hand, a solution of a small amount of 
the solvent in the solid, and, on the other hand, a dilute solution of 
the solid in the solvent. They may also be regarded as being both 
liquid, one of them possessed of a very high degree of viscosity. 
Such systems would not show a high surface tension at the interface, 
nor a high electrical charge. They are, accordingly, relatively to the 
suspensoid class, somewhat insensitive to the precipitating action of 
ions. At the same time, as Mines has shown (1912, p. 211), the 
difference is not a fundamental one and is merely one of degree. 
Egg white is at once thrown down by a simple trivalent ion, such as 
lanthanum, even in a concentration of only 0:0016 molar. The 
complex trivalent ion of the luteo-cobalt salts (Mines, 1912) does not 
precipitate emulsoids, although it is nearly as effective on suspensoids 
as the simple lanthanum ion. The fact appears to be related to the 
low density of the charge on the large ion. 
On the other hand, these particular emulsoid colloids react in an 
important way to another property of salts, that property called by 
Freundlich (1909, pp. 54 and 412) “dyotropic.” It is manifested by 
changes in the distribution of the solvent between the two phases, 
dependent on changes in the solvent itself. In the case of water, we 
speak of the hydration of the ions and changes in the equilibrium 
between the various states of water itself. ‘These give rise io altera- 
