8 THE ROYAL SOCIETY OF CANADA 



II. Absorption Spectrum of Liquid Oxygen 



In these experiments the light from the spark between aluminium 

 terminals under water, after the manner devised by Henri/ was used 

 as the source of radiation. The electrical arrangement for the pro- 

 duction of the Henri spark is shown in Fig. 1. (See page 14). 



The secondary terminals of an induction coil A B were joined to 

 the spark gap C D and then to two condensers E, F, each of which 

 consisted of two one-gallon Leyden jars joined in parallel, the wires 

 from the secondary terminals being cbnnected to the outside coatings 

 of the two sets of jars as shown in the diagram. The inside coatings 

 of the jars were then joined to a second spark gap consisting of 

 aluminium rods, G, H, about 1 cm. in diameter and conically pointed. 

 These rods were mounted in a vertical plane inclined to each other 

 at an angle of about 45° and held in position by clamps which were 

 provided with threads so that the distance between the sparking 

 points could be readily adjusted. These aluminium terminals were 

 immersed to a depth of about 5 cm. in distilled water in a metal vessel 

 provided with a quartz plate window about 2 cm. in diameter 

 and placed at the proper height so that the spark occurred 

 immediately in front of the window. The light from the spark was 

 focussed by a cylindrical quartz lens on the slit of the spectrograph 

 which was of the large quartz type made by the Adam Hilger Co. 

 Several photographs of the spectrum of this light were taken and it 

 was found to be almost perfectly continuous from the wave length 

 X = 7000 Â.U. to the wave-length X = 2150 Â.U., as there were only 

 two very slight aluminium reversals throughout this entire range. 



The absorption spectrum of liquid oxygen in the visible region 

 was obtained by passing the light from the spark between the alumin- 

 ium terminals under water through a column of liquid oxygen con- 

 tained in a cylindrical glass Dewar flask about 5 cm. in internal 

 diameter. The flask was specially prepared for this purpose by 

 being only partly silvered so that there was a clear slit about 1 cm. 

 in width through which the light could pass. A reproduction of the 

 photograph obtained is given in Plate I, Fig. 2 (è), in which there 

 can be seen eight well-marked absorption bands. The mean wave- 

 lengths of these bands were measured and the results are given in 

 Table I, together with the wave lengths of the absorptions obtained 

 by Liveing and Dewar, and Olszewski. 



The absorption spectrum in the ultraviolet region was readily 

 obtained by the use of a small cylindrical Dewar flask, about 1.8 cm. 



1 Henri, Phys. Zeit., No. 12, p. 516, June 15th, 1913. 



