(^50 CA-PILLARY TUBE3. 



That the at- truth of this principle. If the interior surface of a tube of glas5 

 *^Kss'tube mi ^^ covererf with an extremely fhih coating of any greasy suEr- 

 vvater does not stance, the capillary effect wi'll be destroyed as to sense. 



act at any per- Never tl>eless the tube always acts in the same manfter upon flie 

 ceplible nis- ... . 



tance is shewn Column of fluid in the axis ; for the capillary attraction musi 

 by greasing it. y^^ transmitted through bodies in the same manner as is observ- 

 ed with regard to gravity and the attractions and repulsions of 

 magnetism, and even of electricity. Newton,. Clairaut. and all 

 geometers who have subjeefed this class of attractix^ns to com- 

 putation, have proceeded upon that hypothesis : since therefore 

 the capillary attraction is destroyed by the interposition of a. 

 coating of fat matter, however thin it may be, it must follow 

 that the action of the tube will be insensible at any sensible 

 distance. 

 Mei-cuiy at- The following phenomenon affords an additional proof of 



tracts and ad- ^^^ principle here announced. It is known that by strong 

 hores to a glass i r ^ j a 



tube; but in ebullition of mercury in a capillary tube the fluid becomes ele- 



peHesTby aU*ihi ^"^^^^ ^^ ^^^ ^^""'^^^ ^"^ ^^^" ^^^"^^ ^^^ ^^^^^ ^^ ^^® boiling be 

 coating of continued. This phenomenon appears to me to depend upon 

 wa ei. ^j^g ^j^Ij^ coating of water which in the ordinary state lines the 



inner stirface of • the tube and weakens the mutual action of 

 the glass and the mercury j an action which becomes more 

 and more manifest in proportion as the thickness of that coat- 

 ing is diminished by the heat of boiling.. In the experiments 

 which I made with M. Lavoisier upon barometers by boiling 

 the mercury for a long time in them, we caused the convexity 

 of its interior surface to disappear ; we even succeeded in ren- 

 dering it concave ; but we always restored the effect of the 

 capillarity by introducing a drop of water into the tube. If 

 we consider the extreme thinness which the aqueous coat must 

 have, particularly when the tube and the mercury has been 

 well dried, which process is not sufficient to destroy the capil- 

 larity, we may form a judgment that the action of the glass 

 on this fluid is not sensible but at insensible distances. 

 The author's Proceeding on this principle, I determine by the formulas of 



process of ana- ^ly treatise de INlccanique Celeste, the action of a mass of 

 lysis, or the n ■ i • 



theory of capil- fluid termmated by a concave or convex spherical surface upon 



lary tubes. ^^ interior column of fltiid included in an infinitely narrow 

 canal whicli passes through the axis of that surface. By this 

 action I understand the pressure which the fluid included in the 



canal 



