Surface Forces in Fluids. 339 



It will be observed, however, that, so far as we have gone, 

 there is nothing to tell whether the attractive force increases, 

 decreases, or is constant for a diminution of distance. 



We are unable to separate with certainty the repulsive 

 action we have spoken of into two parts, one of which might 

 be due to the temperature and the other io what might 

 be called an inherent elasticity of the molecule. The whole 

 action was called by Poisson and writers before him " the 

 repulsive force of heat " — a phrase which, because it is not 

 entirely in harmony with the modern theory of molecular 

 motion, has fallen into disuse. 



Unfortunately the important relation between phenomena 

 which it served to express has also been lost sight of. What- 

 ever the repulsive and attractive forces and the intramole- 

 cular distances that we are speaking of may turn out to be, 

 we may be quite sure that what used to be called the repulsive 

 force of heat does in solids and liquids vary with alterations 

 of molecular distances more rapidly than does the force of 

 molecular attraction. It is on this fact, clearly pointed out both 

 by Young and Poisson, that the whole theory of capillarity is 

 based. I shall therefore not hesitate to use the phrase " repul- 

 sive force of heat " precisely as it was used by Poisson, to 

 signify the whole of the repulsive action of which I have 

 spoken. Those who recognize in the phenomena of molecular 

 diffusion and heat-conduction evidence that the internal pres- 

 sure not only in gases, but also in homogeneous solids and 

 liquids, is really due to a transfer of molecular momentum 

 taking place equally in opposite directions across any ele- 

 mentary area within the substance, will have no difficulty in 

 perceiving that it is of this transfer of momentum that " the 

 repulsive force of heat " is the statical equivalent, and that in 

 assigning equal efficiency as a centre of force to each of the 

 molecules we are really dealing with the matter statistically. 



3. It is evident that but for the attractive forces between 

 the molecules none of the phenomena of cohesion and adhe- 

 sion could take place; and in the case of solids we are able to 

 derive from observation of the breaking-stress some knowledge 

 of the numerical value of the difference between the two 

 forces. 



We may, consistently with -the phraseology we have used, 

 regard a vertical solid bar of uniform section and subject to no 

 atmospheric pressure as composed of horizontal layers of mole- 

 cules, the distance apart of which layers is determined partly 

 by the temperature of the bar and partly by the tension to 

 which it is subjected. Let us fix our attention on the action 

 across a plane horizontal section between the molecules at 



