303 On Capillary A&ion. 



idea, though not logicaliv accurate, was ingenious, anrl had A 

 certain degree of fuccefs. It is evident that the force meafured 

 was not ftrittly that of mutual alfraClion, but only the weight 

 of the protuberant mafs of fluid the moment before its fepa- 

 ration. Kor does the height of this column depend on its 

 mere adhefion : for it can never exceed the cohefive force of 

 the fluid molecules to each other. When the plate, what- 

 ever may be its attraftive power, is raifed above a certain 

 limit, the attached column breaks, and part of it adheres to 

 the plate, while the reft falls back into the body of the fluid. 

 Thus, thofe experiments can afcertain the adhefion of the 

 plate only when it is comparatively very fmall ; in other cafes 

 thev meafure fimply the integrant attraftion of the particles 

 of the fluid. If capillary action be regarded as approximate 

 to chemical aftinity, this may be always determined, with 

 facility and with fufficient accuracy, by obferving the afcent 

 in two parallel plates fixed at fome minute but known di- 

 flance from each other. 



The elevation of a liquid in a capillary bore is produced by 

 its fpecific attra6lion to the matter of the tube. With dif- 

 ferent fubflances the effects are confiderably diverfificd : al- 

 cohol, for example, rifes little more than a third part of the 

 height to v/hich water reaches. But the flighteft alteration 

 in the conftitution of the fubftance will fometimes occafion a, 

 very material diflerence of cftc6l. Thus I find that, on di- 

 Kiting the alcohol, the afcent of the liquor is fcarcely at all 

 augmented ; but, if I reverfe the procefs, and add a few drops 

 of alcohol to the water, the capiilarv column will fuftor a re- 

 markable dcprcffion. On this principle might be conftruiSled 

 a very Ample hydrometer for mcafiu ing the ftrength of weak 

 liquors. And I would invite chemifts to try capillary a61;ion 

 with different faline and metallic folutions, as I am confident 

 that many curious fads would thus be difcovered. Nothing 

 could exceed the fimplicity of fuch an inftrument. It is only 

 neceflar)' to choofe a fine calibrcd tube, and to have it divided 

 /)n the outfidc into equal parts l)y the point of a diamond, or 

 marked with enamel colours. On plunging it into the liquid 

 and then removing it, the precife efi'cd: would be indicated 

 by the fufpended column. I have no doubt but that fuch 

 an inftrument, in the hands of a fkilful operator, would be 

 brought to mark, with lutficicnt exattnefs, the proportion of 

 ingredients contained in a compound fluid, and might in 

 fome cafes fuperfede the trouble of analyfis itfelf. 



The afcent of water through pounded glals, fand, and 

 other powdery fubftances, is juftly explained by capillary 



a6lion, 



I 



