CAPILLAUY TUBES. 253 



of the surface of the interior fluid, which are extremely near Laplace's theo- 

 the sides of the tube, upon Nvhich inchnation the concavity or tk)^ a^i/de-^* 

 convexity of that surface and the length of its radius depend, pressionof 

 If by the eifect of rubbing the interior fluid against the sides of effect oLttrae- 

 the tube the curvature be increased or diminished, the capil- tion which is 

 lary effect will increase or diminish in the same proportion. ^^^.^ ^^P' " 



It is interesting to ascertain the radius of curvature ' of the 

 surface of water included in capillary tubes of glass. This 

 may be known by a curious experiment, v.'hich shows at the 

 same time the effects of the concavity and convexity of sur- 

 faces. It consists in plunging in water to a known depth a 

 capillary tube, of which the diameter is likewise known. The 

 lower extremity of the tube is then to be closed with the finger, 

 and the tube being taken out of the water, its external surface 

 must be gently wiped. Upon withdrawing the finger in this 

 last situation, the water is seen to subside in the tube and form 

 a drop at its lower base ; but the height of the column is al- 

 wa3's greater than the elevation of the water in the tube above 

 the level in the common experiment of plunging it in water. 

 This excess in the height is owing to the action of the drop 

 upon the column, on account of its convexity ; and it is obser- 

 vable that the increase in the elevation of the water is more 

 considerable the smaller the diameter of the drop beneath. 

 The length of the fluid column which came out by subsidence 

 to form the drop, determines its mass ; and as its surface is 

 spherical, as well as that of the interior fluid, if we know the 

 height of the fluid above the summit of the drop, and the dis- 

 tance of this summit from the plane of the interior base of the 

 tube, it will be easy to deduce the radii of these two surfaces* 

 Some experiments lead me to conclude that the surface of the 

 interior fluid approaches very nearly to the figure of an hemis- 

 phere. 



Clairaut has made this singular remark, namely, that if the 

 law of attraction of the matter of the tube upon the fluid differs 

 only by its intensify from the law of the attraction of the fluid 

 upon itself, the fluid will be elevated above the level, while the 

 intensity of the former of these attractions exceeds the half of 

 the intensity of the latter. If it be exactly the half, it is easy 

 to show that the surface of the fluid in the tube will be horizon- 

 tal, and that it will not be elevated above the level. If these 



two 



