[ 291 ] 



XLI. The Heats of Vaporization of Liquids. By S. Rosling- 

 ton Milnek, B.Sc, 1851 Exhibition Scholar, University 

 College, Bristol* . 



IT is well known that at the surface of a liquid in contact 

 with its vapour there exists, in consequence of the 

 excess of the downward attraction of the molecules in the 

 liquid over the upward attraction of those in the vapour part, 

 a resultant downward pull on the thin surface-film, which is, 

 of course, balanced by the ordinary hydrostatical variations 

 of pressure and density. If we assume that no more degrees 

 of freedom are opened up to the molecules in their passage 

 from liquid to vapour, so that their average kinetic energies 

 in the two conditions are the same, the difference in the 1 

 potential energies of a gram of molecules, or the space 

 integral of this surface-force per gram from the liquid to the 

 vapour, may be taken as the same thing as the " internal " 

 heat of vaporization (t. e. external work done excluded) 

 expressed in mechanical units. 



That with this assumption there must exist a definite 

 relation between the heats of vaporization and the vapour- 

 densities and temperatures of liquids may be seen at once by 

 supposing the surface-film to be indefinitely thin. In this 

 case it would follow that of the molecules striking against 

 the surface of separation, only those whose kinetic energies' 

 resolved perpendicular to it were greater than the value of; 

 the above integral for a molecule would be able to pass 

 through into the vapour. If the law of molecular motion in 

 the liquid be known, the number passing per second into the 

 vapour is thus a known function of the heat of vaporization ; 

 and evaporation goes on at this rate until the vapour-density 

 reaches a stationary value such that the molecules pass back 

 into the liquid at the same rate as they come out. Hence; 

 the value of the vapour-density at any temperature would be 

 determined by its temperature and its latent heat. 



The determinateness of this connexion it will be seen 

 depends on the assumption that there is a difference only iii_ 

 degree, and not in kind, between a liquid and its vapour — 

 that the only difference between the two, in fact, is that in 

 the liquid the mean free path of the molecules is very small. 

 However, the work of van der Waals has shown that this is 

 approximately the case, and a relation obtained in this way 

 may be expected to hold true to the same degree of approxi- 

 mation. 



* Coniniimicated by the Author. 

 Z 2 



