54 On the Spreading of Fluids on Glass. 



the latter the edge state moves bodily over the surface, and 

 in the former the substance of the drop has to be, as it were, 

 dragged through the instable thickness. It is this latter fact 

 which I think accounts for the remarkable phenomena 

 exhibited by a drop of ethyl hydrocinnamate on water*. 



The importance of the vapour in jumping the barrier at 

 the edge of a drop must not be overlooked. Experience of 

 the formation of primary composite surfaces on water by 

 many pure chemical substances leads me to conclude that 

 the surface is formed by actual spreading of the substance 

 of the drop itself only when the sum of the tensions of the 

 upper and lower surfaces of the lens-shaped drop is consi- 

 derably less than that of water, or, using Gibbs's notation, in 

 which the vapour phase is indicated by the numeral 1, the 

 water by 3, and the substance which is to form the composite 

 surface by 2, when T 12 + T 23 is considerably greater thanT ]3 . 



Taking oleic acid and ethyl hydrocinnamate as examples: — 



J- 12 + J 23 J- 13 



Oleic acid , 31 4- 15 = 46 < 74 



Ethyl hydrocinnamate . . . 33 + 24 = 57 < 74 



When the sum of the tensions is only slightly less than 

 that of water, the drop does not flash over the surface, it 

 remains apparently unaltered, but it is ihe vapour phase which 

 condenses to form the primary composite surface, e. g.\ 



Octane 20'0 + 53 = 73'6 < 74 



These relations are clearly indicated by Gibbsf, but for the 

 purposes of some part of his argument the surface 13 is the 

 primary composite surface, and not the water-air surface. 



No fluid amongst those which I examined was found to 

 flash over glass. Judged by the fall in static friction, a 

 film takes some seconds to spread a few centimetres from 

 a drop of acetic acid. 



Whether primary or secondary spreading does or does not 

 occur on a fluid face depends mainly upon the relative value 

 of the surface tensions, but on a clean solid face it must 

 depend wholly upon the vapour tension. For consider a drop 

 o£ any fluid placed upon a clean solid surface. The tension of 

 the latter (T 13 ) may be much greater than T ]2 +• T 23 : but, since 

 the 13 surface is not contractile owing to its solidity, the 

 disparity of force acting at the edge of the drop is inoperative.. 



* Proc. Roy. Soc. A. lxxxviii. p. 324 ^191:i). 

 t Trans. Conn. Acad. vol. iii. p. 422. 



