Mr. J. T. Riley on Capillary Phenomena. 193 



surface in the containing vessel above the pressure caused by 

 the concave surface in the capillary tube." 



This molecular pressure, due to the plane surface, is a 

 result only of the mathematical methods that have been 

 employed to give an explanation of capillary phenomena. 

 We cannot demonstrate experimentally the existence of this 

 pressure, since, if introduced into the calculations at the 

 beginning, it is afterwards eliminated and does not appear in 

 the results. 



In order to determine whether this molecular pressure, due 

 to a plane surface, has any physical existence, I have made 

 the following experiment, which I think conclusively proves 

 that it has not, or, if it has, that it must be very small. I 

 took a small funnel and drew out the stem so as to form a 

 fine capillary tube which I bent twice at right angles, so as 

 to bring the capillary limb parallel with the stem (fig. 6). 

 The funnel was clamped to a stand, and water poured in to 

 the level A; in the capillary limb it rose to B, the height of 

 which I read with a cathetometer. Adopting Laplace's formula, 

 and neglecting the atmospheric pressure, as its effect cancels 

 out, the pressure in the liquid immediately under the plane 

 surface at the level A is K, and that immediately under the 



curved surface at BisK— -^(^- + ^7), since R and Bf are 

 both negative. ZKU nj 



In this case K cancels out; but if, without altering the 

 level at A, we modify the surface tension, then K, which is, 

 according to Laplace, a function of H, is altered, and takes a 

 value K'. The excess of the pressure immediately below A 



over that immediately below B is now K' — K + — f^- + ^7), 

 H / 1 i\ 2 \H Ry 



instead of 9^1 p" "^-57/ as before, and the capillary surface at 



B ought to move up or down according as K'— -K has a 

 positive or negative value. 



To test this, I held a glass rod, with a drop of ether 

 adhering, close to the surface of water in the funnel: the 

 surface tension was immediately diminished; but an obser- 

 vation with the cathetometer showed that no fall had occurred 

 at B. Even when I dropped the ether on the surface I 

 could detect no movement; but immediately I held the drop 

 over the capillary tube, the surface at B began to sink, and 

 finally fell several millimetres. 



It is evident that, although H is very much diminished by 

 the solution of the ether vapour in the capillary surface, the 

 difference between K and K' is not appreciable, so that both 

 must be exceedingly small or zero. 



