174 THE ROYAL SOCIETY OF CANADA 



as possible in evacuated quartz tubes, and although these tubes were 

 heated to softening, the vapours obtained of all three metals were 

 not very dense. Moreover, when the light from the spark in air be- 

 tween gold, silver, and copper terminals was sent through their re- 

 spective vapours, no trace of absorption was in any case obtained. 



Through the kindness of Mr. C. D. Heycock our attention was 

 drawn to some alloys of gold and of silver which have low melting 

 points. One of these which contained 96-2% gold and 3-8% alumi- 

 nium had its melting point at 526 •5°C and another which has the 

 composition represented by Ag3Cu2 contained 28% of copper and 72% 

 of silver and had its melting point at 777 •3°C. The light from the 

 Henri spark was projected in turn through evacuated quartz tubes 

 containing these alloys and though the tubes were heated as highly 

 as was practicable, no absorption was detected in the region between 

 X = 6000 A°.U. and X = 1800 A°.U. 



VI. On the Structure of Fine Lines in the Mercury, Cadmium 

 AND Zinc Arc Spectra. 



When working with the particular form of metallic arc lamp 

 described above, it was found that when the 220-volt circuit was ap- 

 plied, cadmium and zinc arcs of extraordinary brilliancy could be 

 obtained. This made it possible to make a close examination of the 

 structure of some of the finest lines in the arc spectra of these metals. 

 Two of them in particular were carefully studied, viz., the cadmium 

 red line A = 6439-3 A°.U. and the red zinc line A = 6364 A°U. Light 

 of the former wave-length it will be remembered was used by Michel- 

 son^ in his determination of the length of the metre. In studying its 

 structure with his own type of interferometer, it was found by him to 

 be simple and not to possess any satellites. This was confirmed by 

 Janicki^ who, at a later time, investigated it with an echelon grating. 

 In our investigation with a Lummer plate interferometer it was also 

 found to be simple and a reproduction of one of the Lummer plate 

 patterns of it is shown in Fig. 12. 



The zinc line A = 6364 A°.U. is also given by Janicki^ as a simple 

 line without satellites. In our study of it with the Lummer plate this 

 line appeared visually at times to be accompaned by two faint satel- 

 lites. These came out only when high currents were passing in the 

 arc; but although numerous attempts were made to obtain photo- 

 graphs of them they did not appear on any of the plates. We are not 



iMichelson. Phil. Mag. 34 pp. 280—299, 1892. 

 2Janicki. Ann. der. Phys. (4) 29, p. 833, 1909. 

 'Janicki. loc. cit. 



