On Depression of Mercury in the Tales of Barovieteis. 3t 



in the hands of hundreds, it has ceased to be a wonder to Mr. 

 G. Stevenson, wlio, it is well known, when he first saw it. said 

 that it could not possibly succeed. 

 I am, sir, 



Your obedient humble servant, 

 London, Dec. 90, 1816. PhilaleTHES. 



VI. Remarks on Mr. Laplace's Talle of the Depression of 

 Mercury in the Tubes of Barometers*. 



To Mr. Tilloch. 



Sir, — JVIr. Laplace speaks in this paper of having "discovered'* 

 and '■' reduced to the fundamental principle of chemical affinities" 

 the theory of capillarv action : at least there can be no doubjt 

 that his expressions are meant to be understood as relating en- 

 tirely to his own investigations. I must confess that I have 

 never thought hi.s reasoning on the corpuscular powers of matter 

 sufficiently demonstrative: but it is only at this moment that 

 I have been able fully to satisfy myself, by a difierent train of 

 arguments, with respect to the true foundation of the laws of 

 capillary action. 



It has been showni by an anonymous ivriter in Nicholson's 

 Journal for September 1807, that tlie sujierficial strata of co- 

 hesive Itodies, of a thickness less tlian tiie distance to which the 

 cohesive force extends, must be less dense, and less coherent 

 vvitli the strata beneath them,, than other parts of tlie body. 

 Hence it mav be inferred that a small portion of a liquid, in tiie 

 middle of sucii a stratum, will be less stronglv })ressed al)ove and 

 below than a similar portion of the internal part of the li(juid, 

 while the lateral pressure is undiminished, the particles in the 

 direction of the surface on each side exerting their full cohesive 

 power. This inequality must necessarily occasion a tendency 

 in the stratum X'> acquiic a greater thickness and a less length, 

 or in otlier words a contiactile force, wiiich will be the same in 

 every part of the surface, and which will sufficiently explain all 

 the piienoi!iena of capillarv action. 



Ml. Laplace has indeed denied the existence of such a ten- 

 sion : but there is some reason to apprehend that he may have 

 b'.en since disposed to retract this opinion ; as he most as.surcdly 

 would distiaiin the preference which he has given to tlic slovenly 

 method in which his table is computed, if he should ever have 

 an opportunity of compaiiiig the accuracy of its rcf^ults with the 

 detaiN of the same calculation by means of the scries published 



* SccPI.il, Ma", for Fcbniii'-y 1810. 



