Surface Forces in Fluids. 349 



will be — ve, and equilibrium can only be attained by a de- 

 crease of density in the liquid (a) as we pass inwards from 

 the surface. In this case therefore the liquid (u) will be 

 condensed upon the surface of the substance ($); and its 

 surface-layers will exhibit, instead of a tension, a pressure, 

 and the surface of contact will tend to enlarge itself, the liquid 

 in the neighbourhood of the surface possessing, per unit of 

 volume, less energy than the liquid in the interior. 



9. Now that we know how a liquid of uniform density, to 

 start with, would behave if placed in contact with a uniform 

 substance, we can tell how two different liquids, each supposed 

 uniform to start with, will behave when placed in contact with 

 each other. There will be a readjustment of each near the 

 surface in the direction either of an extension or a compression 

 according as what may, for brevity, be called the mutual 

 attraction of the liquids is less or greater than what we may 

 call the self-attraction. We shall return to this result after 

 examining the equilibrium at the surface of a gas or vapour. 



It is necessary, before we can decide how far our results 

 apply also to vapours and gases, to extend to their case the 

 same statical treatment which we have used with liquids. 



] 0. When the temperature and pressure of a gas are such 

 that it is far removed from its point of liquefaction, we find 

 no trace of any attraction between the molecules, but only of 

 repulsion ; and we may regard the phenomena of gases from 

 a statical point of view by considering the molecules to be 

 beyond the radius of each other's molecular attraction, but not 

 beyond that of molecular repulsion, which latter force now 

 varies inversely as the volume, and therefore inversely as the 

 cube of the molecular distances. 



That we know, in the case of gases, something of the 

 dynamical conditions which produce what we are calling the 

 repulsive force, need not prevent us from adopting the statical 

 point of view, which may also be regarded as statistical. 



If the temperature be below what is called the critical tem- 

 perature of the gas, we can, by sufficiently increasing the 

 pressure, liquefy it, and the liquid will only exist as such so 

 long as the pressure above it is maintained. In fact, the 

 interior of a liquid in equilibrium with its own vapour is in a 

 state of compression. There is, however, nothing in this that 

 is inconsistent with the rarefaction of the surface-layers, as 

 may be seen from the following considerations. 



For we may imagine that upon the surface of a non-volatile 

 liquid exposed to a vacuum one or more additional layers of 

 molecules are laid, which molecules will be attracted down- 

 wards by an excess of cohesive force, and will therefore exert 



