228 M. Arago m Artesian Wells 



Let us suppose, then, that the left branch of this tube opens 

 towards the top, with a large reservoir which can maintain it- 

 self always full ; rhat the right branch is cut across towards its 

 lower part ; that only a short portion of its vertical part is left, 

 and that this portion is fitted with a stop-cock. When this 

 stop-cock is open, the water will be projected into the air, 

 through the remaining portion of the right branch, to exactly 

 the height it would have risen if this branch had remained en- 

 tire. It will ascend as far as it has descended from the level 

 of the reservoir, which, without ceasing, supplies the opposite 

 branch *. 



The two hypotheses I have just alluded to have been realized on 

 a great scale, the former in the souterazi of the Turks, and in 

 the greater number of conduit pipes which serve to distribute 

 the water of an elevated spring to the different parts of a town, 

 and the different stories of houses ; and the latter in the sub- 

 terranean conduits which are destined to produce jets d'eau, as, 

 for example, those of Cassel, Versailles, Saint Cloud, or that in 

 the garden of the Tuilleries. When the Romans wished to con- 

 duct water from one hill to another, they, at a great expense, 

 constructed in the intermediate valley aqueduct bridges, such as 

 those of the Pont Du Gard, or such aqueducts as that of Jouy, 

 near Metz, &c. &c. The Turks manage the matter in a way that 

 is infinitely more economical. They place along the declivity 

 of the first hill a pipe of metal, or a tunnel built of bricks or 

 stones ; they carry this along the intermediate valley, making it 

 follow the different inflections it may meet, and finally cause it 

 to ascend the slope of the second hill. In virtue of the princi- 

 ple stated above, the water which runs in this pipe, rises, after 

 having crossed the valley, to very nearly the same height whence 

 it had descended. This is the origin of the name south-azi — 

 the equilibriutn of water — which the Turks give to the conduit 

 pipes which they have substituted for the aqueducts. 



Let us now suppose that the tube is carried to the middle of 

 the valley only, and that there a single opening is made for its 



• Experimentally, we find that the jet of the water is somewhat less; but 

 the difference does not interfere with the principle : it depends upon the 

 friction, the resistance of the air, and the opposing currents of the ascending 

 and descending liquid p;.rticles. 



