S90 Pressure and Efflux of noU'clastic Fluids. 



a small portion of the water will issue, but will cease to run 

 as soon as the air in the upper part of the vessel is so rare- 

 fied that its spring, together with the weight of the column 

 of water above r or g, are exactly equivalent to the re-action 

 of the atmosphere. Raise the tube ef, and the water will 

 issue with a continued and uniform stream, discharging 

 equal quantities in equal times, without regard to the height 

 of the column q r, provided it be within the limits of atmo- 

 spherical pressure. 



The reason is obvious ; for it is evident that the pressure 

 of the water at r would be proportionate to the whole height 

 of the column rq, were the atmosphere to act equally at g 

 and q; but since the air is made to enter through the tube 

 ef and thereby to counteract the column fq, it follows 

 that the pressure at q will be less than that at g by the 

 weight of the column qf. Now the pressure at o is in the 

 inverse ratio of fq, therefore the pressure at r will be uni- 

 formly the same. For while the vessel is closed at the top, 

 and the air made to enter through the tube ef the pressure 

 on any plane, htj, will be equal to the column qf together 

 with the pressure on the surface q ; for as the weight qf 

 decreases by the reduction of the water, so does the pressure 

 of the air on the surface increase by its easier access (in 

 having to counteract a less column of water), the increment 

 of the one being always equal to the decrement of the other. 



I shall now proceed to explain that part of the instrument 

 which is employed for keeping up a constant supply of 

 water when attached to a gasometer or the like, where an 

 experiment is required to be continued a considerable length 

 of time without intermission, V is a vessel provided with 

 a cock h and funnel k, having also a cock 1 communicating 

 with the top of the machine near dj and a tube mn, and 

 cock p, communicating with the bottom of the machine AB. 

 Shut the cocks 1 and p, and open k ; fiU the vessel V and 

 tube mn with water until it rises above k, which must now 

 be shut, and the cocks 1 and p opened ; by this means the 

 whole is formed into one vessel, and the operation is no 

 more than transferring the tiuid from one part of the ma- 

 chine to another, the pressure and efflux remaining the 

 fame. It will easily be perceived that the pressure may 

 be increased or decreased at pleasure by slipping the tube ef 

 upwards or downwards as circumstances may require. The 

 instrument in the Mathematical Society's repository is usc4 

 (besides the above purposes) for the illustration of all thos^ 

 experiments termed hydrostatic paradoxes. 



I-in. Qr^ 



