ARTICLES 241 



the volume of solution, i.e. with the concentration of the solute, 

 then the osmotic pressure alters with the surface area of the 

 interface. This can only be so if the concentration of the 

 solute in the bulk of the solution depends upon the surface 

 area — i.e., if the concentration in the surface area is greater 

 or less than in the bulk of the solution. It can also be shown 

 that, if the difference at the surface layer of the solute per unit 

 area be T, then, 



^ ~ RT 8c 



where c is the concentration of the solution. That is, if ^ is 



negative (the substance lowers the interfacial tension), then 

 r is positive, and positive adsorption occurs. This equation 

 has been verified experimentally by Lewis ^ with fairly good 

 results. We have already seen the expression given by Wilson 

 for the potential difference which results from adsorption in 

 the case of ions. 



Coagulation. — The greater part of our knowledge of dis- 

 persoids has been derived from a study of their coagulation, 

 though the exact method, and also the rate of coagulation, seem 

 to have been neglected. Linder and Picton have observed 

 microscopically the coagulation of suspensions to be an aggre- 

 gation of the particles, and the gradual growth in size of the 

 masses so formed. This would result in the increase of in- 

 ternal energy at the expense of surface energy, and at the 

 same time the Brownian movement would cease. Attractive 

 forces between two colloid particles seem to fall off more 

 rapidly than repulsive forces, since close contact of the two 

 particles appears necessary for coagulation. We have seen 

 that the repulsive force should vary directly with some func- 

 tion of the charge on the particles. Hence neutralisation of 

 this charge will permit of closer contact. At the same time we 

 have also seen that a decrease of potential difference, due to 

 the charge on a particle, leads to an increase in the surface 

 tension. This would favour coagulation, by which the abso- 

 lute surface would be reduced. We may, therefore, expect 

 that the neutralisation of the charge upon the dispersoid phase 

 would be one effective cause of coagulation. Neutralisation 

 may most readily be brought about by adsorption of ions 

 carrying a charge of opposite sign. That this causes coagula- 

 tion is a well-known experimental observation. An interesting 

 fact, due to the work of Powis ^ on hydrocarbon oil emulsions, 

 has recently been brought to light. Powis showed that 



* Phil, Mag., vol. xv, p. 499, igo8 ; vol. xvii, p. 466, 1909. 

 2 Zeit. Phys. Chem., vol. Ixxxix, pp. 91, 179, 186, 1914. 



