EXPERIMENTS WITH THE AIR-PUMP 



285 



FIG. 190 (| real size). 



obtained the liquid would probably have already passed 

 from one vessel into the other ; and besides, the pres- 

 sure of aqueous vapour, 

 which also exerts some pres- 

 sure, would interfere with 

 the success of the experi- 

 ment. Mercury may, how- 

 ever, be used for it. One end 

 of a very narrow siphon, of 

 the form shown in fig. 190, 

 with an aperture in the 

 middle of the lower bend, is 

 dipped in a cylindrical vessel filled with mercury, which 

 is placed by the side of an empty vessel upon the plate 

 of the air-pump. The mercury is made to flow by 

 sucking, and the whole is then covered by the receiver. 

 The pump being worked, not too slowly, the thread of 

 mercury will first break at the top of the longer branch 

 of the siphon which is outside ; but as there is now a 

 vacuum in this branch, the liquid will continue to flow 

 from the vessel through the shorter branch of the siphon 

 which is inside the liquid, until finally, when the pump- 

 ing is continued, the pressure is insuificient to raise 

 the liquid in the shorter branch to the top of the siphon, 

 and the flow ceases. 



The bore of the siphon should not exceed O mm '5. The aperture 

 at the side is made before the siphon is bent. The tube, 40 or 50 cm 

 long, is closed at one end by the finger, and an assistant blows 

 strongly into the tube at the other end. The point of a blowpipe 

 flame is now directed upon the exact place where the aperture is to 

 be made, and that side is heated until it becomes soft, care being 

 taken not to heat the opposite side. A small bulb will be thus formed 

 and then burst. The edges of the aperture are then rounded in a 

 common flame, and the tube bent into the required form. 



