110 Mr Stewart on the Causes of Obstruction in 



water from entering the syphon until the water again accumu- 

 lates, raising the air-ball along with it, and opening the valve. 

 The second of the inconveniences referred to, is to be remedied 

 by keeping the end D immersed in a trough or vessel G ; which, 

 receiving the water as it flows from the syphon, remains always 

 full, and prevents the entrance of air. Thus furnished with an 

 air-extractor, and the means just described of preventing the 

 water from being drained off, all that is necessary to renew the 

 working of a syphon obstructed by the presence of disengaged 

 air in its vertex, is to close the valve B, by pressing down the 

 air-ball, if in practice that be found necessary, and then to press 

 down the end h of the lever h 7c, (fig. 1.), which, shutting the 

 valvey,' and opening the valve e, allows the water in the cylin- 

 der A to descend, and the air in the cylinder B to escape. The 

 air-ball being then allowed to rise, and the valve B opened 

 (fig. £.), the water will begin to flow ; when the cylinder A may 

 be again filled with water, to be in readiness when required. In 

 situations where, in setting the syphon to work, it may be diflS- 

 cult to employ a syringe for exhausting the air, the purpose may 

 be served by furnishing the lower limb D also with a valve; 

 then, shutting both valves B and D of the extremities, and 

 opening both valves of the air-extractor (by disengaging one 

 end of the lever from the working-rod), fill the syphon with wa- 

 ter from the top ; when this is done, all that is necessary is, to 

 shut the upper valve of the air-extractor, and open the valves B 

 and D, when the water will flow freely. 



Fig. 4. 



When this mechanism is to be applied to free conduit-pipes of 

 air, it is to be inserted into the place where the air is most dis- 



