240 SCIENCE PROGRESS 



capillary effects alone, the basis of his theory being that the 

 range of molecular attraction varies with different substances. 

 It would, however, seem impossible to formulate any theory 

 of colloid state of general application which neglected elec- 

 trical phenomena. As a result of his theory, particles of the 

 same colloid appear to be acted upon by two effective forces — 

 positive and negative surface energy — tending respectively to 

 increase and decrease the surface area. The particles would 

 then have, within limits, a critical size. 



B. Negative Surface Energy. — ^The idea of an effective 

 negative surface tension has recently been developed by Wo. 

 Ostwald.^ Reasoning from analogy to liquids and gases, he 

 would suspect a form of surface energy to exist in dispersoid 

 systems, of which the intensity factor might be called expan- 

 sive or negative surface tension. The increases in surface 

 which occur in strictly diphasic systems can only be explained 

 by assuming the existence of such a surface energy. Such in- 

 creases in surface have been observed by Traube, Mengarini, 

 and Scala,* and Amann,' when apparently no electric energy 

 is available. It can be shown that negative surface energy 

 should increase with the existence of a potential difference at 

 the surface. In the homogeneous state of matter, the increase 

 of positive surface energy is equivalent to the decrease of 

 expansive surface energy. Ostwald considers that excessive 

 development of absolute surface is accompanied by liberation 

 of positive surface energy, and this characterises the act of 

 dispersion. This view is by no means new — we have seen 

 that Donnan recognised an effective negative surface tension 

 — but Ostwald has laid upon it considerable stress. 



Osmotic Forces. — Osmotic forces are due to a difference in 

 concentration of the ions at the interface of two phases, and in 

 the bulk of solution. Little experimental work is possible on 

 this subject, for we are unable to examine directly the consti- 

 tution of the dispersoid phase. Work on the swelling of gels f 

 would seem to show that the dispersoid was permeable by 

 the solvent molecules, but generally, when we consider the 

 magnitude of the surface tension, this seems improbable to 

 any extent. The subject of osmotic forces and capillary 

 chemical effects in general may best be approached by a study 

 of Gibbs' adsorption equation * : 



So- _ 8P 



hv ~ hS 



That is, if the surface tension o- at an interface alters with 



^ Loc. cit. ^ Koll. Zeit., vol. vi, p. 65, 1910. 



" Ibid., vol. vi, p. 235, iQio- ^ Scientific Papers, vol. ii. 



