Liquid Spheres — Molecular Diameters. 881 



It is difficult to conceive, however, that the work done by 

 either of the other two components of internal pressure 

 parallel with the surface can he equal to thnt done by the 

 one perpendicular to the surface. For in each of the former 

 components the forces which attract the molecule on 

 opposite sides are equal, and the energy is not available, 

 since no work can be done by forces in equilibrium ; whereas 

 in the case of the latter component the forces which attract 

 the molecule towards the interior are greater than those 

 which attract it towards the exterior, and the energy is 

 available, since work may be done by their resultant. Still, 

 from the best available data for d, E, and L for low 

 temperatures, the calculated values of these two expressions 

 for twenty-nine substances bear out his statement that " the 

 relation above fits these facts remarkably well." 



Another objection, which applies also to similar reasoning 

 by other writers, lies in the division of the latent heat into 

 two pa'ts. If the external layer of 'molecules contains 

 potential energy equivalent to only one-sixth of the latent 

 heat, would not the other layers within the range of mole- 

 cular attraction contain the remaining five-sixths and the 

 surface energy be 6 E ergs per sq. cm. ? Or, is the latent 

 heat composed of two distinct kinds, or does it arise from 

 two distinct sources? 



Now it so happens that Mr. Hammick's formula itself 

 suggests a means of completely meeting both these objec- 

 tions. For his expression — for the area occupied by the 



molecules of a gram molecule when they are all arranged in 

 the surface layer, assuming, as it does, that the molecules fill 

 the whole volume, is also the expression for the area occu- 

 pied by one molecule. This area is thus •-, d d /d or - . ird 2 , 



which is exactly one-sixth of the area of its free surface. 

 If, therefore, we suppose that the surface of the free mole- 

 cule possesses surface energy of the same intensity as that 

 which he supposes the surface layer to have acquired, the 

 whole surface energy would be six times as much as 

 Mr. Hammick thinks, and his relation should be : 



E. T .j-L„ 



which may evidently be written 



E.Nird 2 .- 



where N is Avogadro's constant. Also, since his molecules 



E.Nrf.-| = U 



