Surface Forces in Fluids. 337 



to the earlier notion of action at a distance between electrified 

 bodies. For, on the statical theory, we regard the equilibrium 

 of any two parallel layers of uniformly distributed molecules, 

 sufficiently near to act upon each other, as maintained by 

 forces of attraction and repulsion acting across the empty 

 intervals which separate them from the intervening molecules; 

 while, on the dynamical theory, we account for the latter of 

 these two forces by what is equivalent to a pressure arising 

 from the momentum of these intervening molecules, which 

 thus constitute a medium the action of which replaces one of 

 the actions at a distance ; and it is well known that in the 

 case of electricity the results of the earlier theory are all trans- 

 latable into the language of the later. 



And it is pertinent also to remember that it is a common 

 artifice in dynamical problems, as for instance in Planetary 

 Astronomy, to make abstraction of the motion of a portion of 

 the system by the introduction of a repulsive force. In the 

 present instance the whole system probably consists of a 

 number of molecules in motion which, being endowed with 

 attractive forces, would, but for their motion, fall together. 

 If we are to imagine them stationary, and at the same time 

 to preserve the external characteristics of the system, we must 

 introduce the repulsive forces. 



I have thought it well to preface the argument which 

 follows by these remarks, since there is a tendency to forget 

 or disregard results obtained by the statical theory, because 

 the language in which they are expressed is thought to be out 

 of harmony with the newer conceptions. 



In the arguments which follow it will be found that, while 

 dealing with solids and liquids, I have preferred to use the 

 conception of separate molecules, and to speak of the intra- 

 molecular distances as decreasing or increasing with the 

 application of external pressure or tension, though the argu- 

 ments would equally well have permitted me to speak of 

 elementary volumes and variations of density: in fact, that 

 it is not till the subject of gases is considered that it 

 becomes necessary to speak of molecules. It appears to me, 

 however, that, apart from the desirability of adopting through- 

 out the paper the same system of describing phenomena, 

 there is an advantage in choosing a method which shows that 

 the conclusions arrived at in the case of solids and liquids are 

 consistent with the assumption of molecular discontinuity, 

 should it be necessary to make that assumption. 



2. In order to realize clearly the nature of the molecular 

 conceptions on which the Theory of Capillarity is based, let 

 us imagine a homogeneous cube of a solid or liquid substance 



