of Light Radiations emitted by Iron Vapour, 3 



2. Experimental Methods. 



Hie feeding of the Flames. — For the air-coal gas flame 

 the high pressure Grouy sprayer (working at from 1\ to 

 2 atmospheres) and a new low-pressure electric sprayer were 

 used ; for the oxy-coal gas and oxy-hydrogen flames only the 

 latter. The spark method described above could not be 

 employed, for it requires a specially powerful electrical 

 equipment. In our original experiments we made use of a 

 Hemsalech-Tissot resonance transformer * in connexion with 

 a Moscicki condenser of 0'05 micro-farad, working at a 

 pressure of from 20,000-30,000 volts. 



For a description of the Grouy sprayer, the reader is 

 referred to de Watteville's original memoirs. 



r lhe Electric Sprayer. — This is an enlarged and improved 

 form of a simple device originally intended for demonstration 

 purposes only f« An inverted bell-jar (fig. 1) 10-J- in. high 

 by 5 in. diameter is fitted at its narrow end with a rubber 

 stopper, through which passes a stout wire carrying the iron 

 electrode E 1? and a bent glass tube G communicating by 

 means of rubber tubing with the reservoir R. A wooden 

 disk L, provided with a flange, forms the lid and fits air-tight 

 on the ground edge of the jar by means of the rubber washer 

 W. The disk is securely held by six clamps C pressing upon 

 a steel ring F, which rests upon the flange. Through this 

 disk pass the following glass tubes : — 



1. An inlet tube A, which passes down to within an inch 

 from the inner surface of the rubber stopper. 



2. A slightly wider outlet tube B, which only just passes 

 through the disk and is provided with a widened mouth to 

 prevent larger drops of liquid, which collect here, from 

 interfering with the free passage of the gas. 



3. A glass tube D^ 6 in. high by 2 in diameter, closed at 

 its upper end by a cork stopper, through which passes a 

 second tube D 2 , 12 in. long by 1 in diameter. Into this tube 

 is fitted the capillary D 3 passing right down the middle of 

 the bell-jar to within about 1 in. from the electrode E x . 

 Through this capillary passes a stout iron wire, the uncovered 

 end of which forms the electrode E 2 , leaving a spark gap of 

 \ in. or more between it and Ej. 



The object of these three concentric tubes is to prevent the 

 formation of a continuous layer of conducting liquid along 

 the inside walls of the apparatus with consequent short 

 circuiting of the electrodes. 



* Hemsalecli et Tissot, Comptes Rendus de I'Academie des Sciences, 

 vol. cxliv. p. 262 (1907). 

 t Hemsalech, Reports Brit. Assoc. Dublin, 1908. 



B2 



