Chap. 2.] The Siphon. 417 



of its dtfcharge through the fiphon. For fuppofe G F 

 the confines of the atmofphere, all the points of the 

 furface A of the liquor will be equally prefied by the 

 column of air AF; if, therefore, at fome point of 

 this {lirface, the prefiure is fufpended, the liquor mud 

 flow at that point, becaufe it finds lefs refiftance there 

 than in any other part j this is therefore the obvious 

 reafon why the fiphon becomes full immediately after 

 the air is drawn out at the extremity C. 



If the two branches of the fiphon were of equal 

 lengths, as B A, B D, the flow through the bent tube 

 would not take place ; becaufe the column of air D G 

 which would refift in D, being of an equal height 

 with that which preiTes at A, would alfo be in equili- 

 brium with it, in the fame manner as the two columns 

 of the fluid B A, B D. But fince B C, one of the 

 legs, is longer than the other, though the column of 

 air G C, which ahfwers to it, is really longer than that 

 which preffes in A ; yet it is not capable of preventing 

 the pafiage of the fluid. To underftand this more 

 perfectly, let us confider the column of air G C to be 

 divided into two parts, one of which G D, would form 

 an equipoife wich the column of air F A, and would 

 be capable of (lopping the flow from the tube if the 

 branch B C ended in D. The portion of fluid which 

 fills the part D C of the fiphon, will find no other re- 

 fiftance in C than one column of air D C of the fame 

 length with it, which is evidently very inferior to it 

 in weight. This portion of fluid then flows out, be- 

 caufe it greatly exceeds in weight the column of air 

 which is oppofed to it. But while it continues to 

 flow, nothing fuftains that which is above it, which 

 flows neceflarily, while the prefiure of the air at A 

 furnifhes a new fupply of fluid to replace that which 

 runs out. It is by thefe means, that the water in' the 



VOL. II, E e fiphon 



