434 PHILOSOPHICAL TRANSACTIONS. [aNNO I7I9. 



remainder of the pressure of the atmosphere on the column of water at a is 

 sustained by the force with which that column adheres to the capillary tube, 

 which therefore exactly balances the weight of the cylinder of water pgbc, and 

 is the real, though not the immediate, cause of its suspension. 



The experiment succeeds in the same manner when a column of quicksilver 

 is raised into the funnel, instead of the column of water fgbc, the top of the 

 tube being touched with a wet finger as before. But then the height of the 

 quicksilver in the funnel must be as much less than that of the water, as its 

 specific gravity is greater. 



I proceed now to examine whether the experiments in N°355 would succeed 

 in vacuo; and whether water could be suspended in a wide tube by means of 

 a capillary at top, at a greater height, than what it can be raised to by the 

 pressure of the atmosphere. 



In order to this, I boiled some water, and afterwards drew out its air by 

 means of the air-pump; then those experiments all succeeded in the exhausted 

 receiver in the same manner as in the open air. The 13th experiment in 

 particular was made with a tube of about 33 inches in length, and a quarter of 

 an inch diameter, the top of it being drawn out into a fine capillary. Which 

 being filled with water freed of its air, as before mentioned, the whole quan- 

 tity continued suspended in the exhausted receiver. This plainly shows, that 

 the success of that experiment does not depend on the pressure of the air, 

 since the small quantity of air left in the receiver was by no means capable of 

 sustaining the water at so great a height; and consequently that the height, 

 at which water may be suspended in this manner, is not limited by that 

 pressure. 



But here I must not omit taking notice of a considerable difl[iculty, which 

 presents itself to those who attentively consider this experiment. In order to 

 make which the better appear, it will be proper to observe what happens, 

 when a simple capillary tube is filled with water freed of air, and inclosed in 

 the exhausted receiver. In this case, the whole column of water contained 

 in the tube acb, fig. 3, is suspended by the attraction of the annulus at the 

 top of the tube, a. And though that annulus does not immediately act on 

 any part of the water, except what is either contiguous to it, or so near as to 

 be within the sphere of its attraction, which extends but to a very small dis- 

 tance ; yet it is impossible that any other part of the water, as for instance that 

 at c, should part from the water above it and sink down, because its descent 

 is opposed by the attraction of the contiguous annulus at c. For this, being 

 equal to the upper annulus at a, is capable of sustaining a column of water of 

 the length ab, and consequently is more than suflicient for supporting the 



