Processes in the Useful Arts. 153 



lift float T, which shuts another sluice. When the rivulet rises till the 

 water flows out at aperture 6, it raises float U, and shuts the third or last 

 sluice ; the flood being now supposed so great, that the streams below the 

 reservoir are of themselves sufficient for tbe supply at Z. When the 

 streams begin to fall, the rivulet at IK will also fall; and when the water 

 ceases to flow into aperture 6, the water falls so far in the tubes, as to let 

 down float U, and open one sluice ; when it ceases to flow out at aperture 

 5, the float T falls, and a second sluice opens ; when it ceases to flow out 

 at aperture 4, the third sluice opens, which, with the other two, passes all 

 the water that the rivulet is then bringing into the reservoir Should the 

 rivulet continue to fall, so as not to flow out at aperture 3, then the water 

 ceases to flow along QR, and one sluice shuts ; should it fall below aper- 

 ture 2, the water also ceases to flow along OP, and a second sluice shuts, 

 should the rivulet become quite dry, then the third or last sluice shuts. 

 Any number of sluices may be used that are found necessary ; and in this 

 way the same quantity of water will always run in the rivulet at Z, as if 

 no reservoir had been placed upon the rivulet above, except during floods, 

 when all the water not required at Z would be detained in that reservoir. 

 Besides the immense quantity of water thus gained during floods, the ex- 

 pence of cutting an aqueduct round the reservoir is also saved ; nor is any 

 bye-water necessary,* as the main sluice on the reservoir, that regulates 

 the height of the water in the basin AB, acts also as a waster. When it 

 is necessary to supply any fixed quantity of water from the reservoir, we 

 have only to make an aperture in the basin of the proper size ; and, as 

 the water there stands always at the same height, the supply will always 

 be the same. 



2. Description of a Rotatory Gas-Burner. 

 Various attempts have been made to construct a gas-burner which should 

 revolve upon the principle of Barker's mill, by means of the reaction of 

 the gas issuing under the ordinary pressure at which it is burned. If the 

 place round which the motion is performed is an ordinary gas-tight joint, 

 the friction is so great, that a motion of rotation cannot be obtained, un- 

 less the gas has been greatly condensed, so as to issue under the pressure 

 of many atmospheres. A rotatory burner of this description was made 

 last year by Mr Deuchar, but it was nothing more than a philosophical 

 experiment, quite inapplicable to gas, as it is generally used. 



• It is not necessary that the tunnel of the reservoir should waste the water as 

 fa-st as it flows into the reservoir during the greatest floods, because the water be- 

 gins to waste several feet below the top of the embankment ; and, although the tun- 

 nel should not pass above one half of the water that flows into the reservoir during 

 the greatest floods, yet as such floods only last for a short time, the water will scarce- 

 ly ever rise in the reservoir more than a few inches during any one flood ; and, be- 

 fore another comes, it is again down to its proper level ; this, of course, depends up. 

 on the greater or less extent of the surface of the reservoir, as well as the size of the 

 rivulet or feeder ; but, in most cases, a bye-water will be found to be quite unneces- 

 sary, if a proper tunnel with a self-acting sluice be adopted. 



