334 PHILOSOPHICAL TRANSACTIONS. [aNNO I7I84 



8th experiments, except that the narrower part is of a greater length ; and let 

 AF and BG be the heights, to which the water would spontaneously rise in the 

 two tubes ED and dc. 



Exper. 9. — If this tube have its wider orifice c immersed into the water ab, 

 and be filled to any height less than the length of the wider part, the water will 

 immediately subside to a level with the point g ; but if the surface of the con- 

 tained water enter never so little within the smaller tube ed, the whole column 

 DC will be suspended, provided the length of that column do not exceed the 

 height AP. 



In this experiment it is plain, that there is nothing to sustain the water at so 

 great a height, except the contact of the periphery of the lesser tube, to which 

 the upper surface of the water is contiguous. For the tube dc, by the sup- 

 position, is not able to support the water at a greater height than bg. 



Exper. 10, fig. 14. — When the same tube is inverted, and the water is raised 

 into the lower extremity of the wider tube cd, it immediately sinks, if the 

 length of the suspended column dh be greater than gb ; whereas in the tube 

 DE, it would be suspended to the height ap. From which it manifestly appears, 

 that the suspension of the column dh does not depend on the attraction of the 

 tube DE, but on the periphery of the wider tube, with which its upper surface 

 is in contact. 



For the sake of those, who are pleased with seeing the same thing succeed in 

 different manners, we subjoin the two following experiments, which are in sub- 

 stance the same with the 9th and 10th. abc, fig. 15, is a siphon, in whose 

 narrower and shorter leg ab, if it were of a sufficient length, might be sus- 

 spended a column of water of the height ep ; but the longer and wider leg bc 

 will suspend no more than a column of the length gh. 



Exper. 1 1 . — This siphon being filled with water, and held in the same posi- 

 tion as in the figure, the water will not run out at c the orifice of the longer 

 leg, unless dc, the difference of the legs ab and bc, exceed the length ef. 



Exper. 12, fig. 16. — U the narrower leg bc be longer than ab, the water 

 will run out at c, if do the difference of the legs exceed ep ; otherwise it will 

 remain suspended. 



In these two experiments it is plain, that the columns dc are suspended by 

 the attraction of the peripheries at a, since their lengths are equal to ef, or to 

 the length of the column, which by the supposition those peripheries are able 

 to support ; whereas the tubes bc will sustain columns, whose lengths are equal 

 to gh. 



Though these experiments seem to be conclusive, yet it may not be impro- 

 per to prevent an objection, which naturally presents itself, and which at first 



