INVESTIGATED MY THE MITHOD OF .IKT VIl:i;\T!nN 



351 



After a number of CX|M-I iments 1 have come to the conclusion that the following 

 arrangement for this s]XM-i;il use is perfectly satisfactory: 



From the tap the water is conducted direct through a rubher tube to the spout of 

 the funnel T (fig. 3), which is fastened with sealing-wax to the neck of the bottle F. 



To jet 



Fig. 3. 



This rests through a wooden frame R on a metal plate P, which is provided 

 with three adjustable screws S. The whole is borne on a bracket K, placed on an 

 outer wall. The bottle F is open above and is provided with an outlet from below as 

 shown in the figure. By help of the adjustable screws the upper edge of T is kept 

 horizontal. 



The water coming from the supply pipe will run over the edge of the funnel in the 

 form of a thin layer, and the height of the surface of water in the funnel will only be 

 very slightly dependent upon the speed of the supply, so that the unavoidable 

 variations in the pressure of the supply pipe will practically have no influence. 



The water in the funnel T is in connection with the water in the reservoir B 

 through a syphon made of glass tubes r t , r t , r a and rubber tubes </, and / 2 ; the 

 surface in B will keep the same height as the surface of water in T. B is provided 

 below with a tubulure that serves for the introduction of one branch of a T-tube, the 

 other two branches of which are provided with rubber tubes, the one serving the jet 

 apparatus, whilst the other is only used for filling or emptying the reservoir. 



It follows from the above that when the quantity of water supplied to T from 

 the supply pipe is greater than that used in the jet apparatus, the surface of water 



