VOL. XXX.] PHILOSOPHICAL TRANSACTIONS. 333 



tube, it follows that the surface, containing the suspended water, in every tube, 

 is always a given quantity : but the column of water suspended is, as the dia- 

 meter of the tube. Therefore, if the attraction of the containing surface be 

 the cause of the water's suspension, it will follow that equal causes produce 

 unequal effects, which is absurd. 



To this it may perhaps be objected, that, in two tubes of unequal diameters, 

 the circumstances are different, and therefore the two causes, though they be 

 equal in themselves, may produce effects that are unequal. For the less tube 

 has not only a greater curvature, but those parts of the water, which lie in the 

 middle of the tube, are nearer to the attracting surface, than in the wider. But 

 if any thing follows from this, it must be, that the narrower tube will suspend 

 the greater quantity of water, which is contrary to experiment. For the 

 columns suspended are as the diameters of the tubes. But as experiments are 

 generally more satisfactory in things of this nature, than mathematical reason- 

 ings, it may not be amiss to make use of the following, which appear to me to 

 contain an experimentum crucis. The tube cd, fig. 11, is composed of two 

 parts, in the wider of which the water will rise spontaneously to the height bp ; 

 but the narrower part, if it were of a sufficient length, would raise the water to 

 a height equal to cd. 



Exper. 7. — This tube being filled with water, and the wider end c immersed 

 in the stagnant water ab, the whole continues suspended. 



Exper. 8, fig. 12. — ^The narrower end being immersed, the water im- 

 mediately subsides, and stands at last at the height dg equal to bp. 



From which it is manifest, that the suspension of the water in the former of 

 these experiments, is not owing to the attraction of the containing surface : 

 since, if that were true, this surface being the same, when the tube is inverted, 

 would suspend the water at the same height. 



Having shown the insufficiency of this hypothesis, I come now to the real 

 cause of the phenomenon ; which is the attraction of the periphery, or section 

 of the surface of the tube, to which the upper surface of the water is con- 

 tiguous and coheres. For this is the only part of the tube, from which the 

 water must recede on its subsiding, and consequently the only one, which by 

 the force of its cohesion, or attraction, opposes the descent of the water. This 

 is also a cause proportional to the effect, which it produces; since tiiat periphery, 

 and the column suspended, are both in the same proportion as the diameter of 

 the tube. Though from either of these particulars it were easy to draw a just 

 demonstration, yet to put the matter out of all doubt, it may be proper to con- 

 firm this assertion, as we have done the former, by actual experiment. 



Fig. 13. Let therefore edc be a tube, like that made use of in the 7th and 



