* ISg New Hydraulic Machines, [March, 



of the water be stopped, the air which issues alone occasions only 

 a slight hissing noise. 



OSCILLATING COLUMN. 



The third method thought of by tlie author, for raising the 

 water of a reservoir above its natural level, is what he calls the 

 osciliaiing column. Of his three principal methods, it is the 

 one which appears to us the newest. We know nothing that 

 could have suggested the fundamental idea of it. It is exceed- 

 ingly simple, since it consists of nothing but a bent tube, 

 adapted to a reservoir, the continuity of which is interrupted 

 towards its lowest part. It is this interruption of continuity in 

 the tube which occasions, what we see with surprise, the water 

 to mount above the reservoir, without the addition of any other 

 piece to the machine. 



To explain this phenomenon, let us suppose a syphon reversed, 

 that is to say, with the open ends of the branches turned upper- 

 most. If we drop a ball into one of these branches without 

 communicating to it any momentum, it is evident that in conse- 

 quence of the velocity which it will acquire by the fall, it will 

 ascend in the other brancii to the same height from which it fell 

 in the first, and that, abstracting the efiect of friction, this ball 

 will continue to oscillate in the syphon rising in both branches to 

 the same height. 



But if we make a second ball succeed the first, and in conti- 

 guity with it, then the first ball v/ill rise in the second branch of 

 the syphon to a greater height than that from which it descended, 

 and the second to a less height j for it is the centre of gravity of 

 their system, that is to say, their point of contiguity, which ought 

 to ascend to the point from which they set out. The same thing 

 would hold if there were a greater number of balls. 



This shows us that water poured into one of the branches of 

 the syphon ought to rise higher in the other when we continue 

 to pour it into the first : however, as it cannot rise indefinitely in 

 the second branch, a time arrives when it begins to press upon 

 the column below, and to push it back into the first branch. 

 This effect may be easily estimated by the principle of the co?i- 

 servation of liu'mg forces', for it results from that principle that 

 the moment the column becomes stationary, in order to flow 

 back, the centre of gravity of the mass ought to be precisely at 

 the height of the open extremity of the first branch of the 

 syphon ; since it is by it that all the water has been introduced, 

 and that it is supposed to have no other velocity but what it 

 acquired by flowing from that extremity. 



But according to the same principle, if at the moment when 

 the water becomes stationary, one were to subtract or annihilate 

 the small portion of fluid wliich fills the lowest part of the 



