210 THE ROYAL SOCIETY OF CANADA 



The Dicyanin, taken from a stock solution previously made, was 

 poured into a well-shaken solution of distilled water and alcohol which 

 had been allowed to stand for a few minutes. After being vigorously 

 stirred and again allowed to stand for some time the ammonia was 

 added. After a short time the solution was poured over the plate in 

 a shallow tray. The plate was allowed to remain in the dye for 

 43^2 minutes and was then placed in an ethyl alcohol bath for 40 

 seconds, after which it was dried in a current of cold air. It was 

 necessary to perform all the operations in total darkness and through- 

 out to maintain the solutions at a low temperature. With an exposure 

 of thirty hours the wave-length X = 10140A came out clearly along 

 with a number of others recorded by McLennan and Shaver. ^^ As 

 already stated, a reproduction is shown in Fig. 24. 



(3) Infra-red Spectra of Tin, Lead, Bismuth, Zinc and Antimony 



With arrangements identical to those used in the preliminary 

 experiments above, investigations were undertaken on the spectra of 

 tin, lead, bismuth, zinc and antimony. A carbon arc lamp was 

 used, the lower electrode having been drilled out and filled with the 

 metal to be examined. On account of the relatively low melting 

 points of the metals used in this work it was impossible to employ 

 electrodes of the metals themselves. The upper negative carbon was 

 1 cm. in diameter and the lower positive one 1.5 cms. In one case 

 an arc was tried in which the upper electrode was a cylindrical copper 

 rod^- 1 cm. in diameter, and the lower electrode a copper plate about 

 7 cms. long, 3 cms. wide and 2 cms. thick. On this a small bead of 

 the metal under test was placed, constant feeding being necessary. 

 It was thought that this should bring out the enhanced spectrum 

 more clearly but the final results were not as satisfactory as when 

 carbon electrodes were used. The arc was fed from the 220 D.C. 

 mains and carried a current of from 15 to 20 amperes. The lamp 

 was placed so that the light fell on a concave mirror, and from it was 

 reflected to the slit, no light pa.ssing from the arc directly to the slit. 

 The average length of the arc was about 8 mms. and its image was 

 kept upon the slit by adjusting the mirror in such a manner that the 

 images of the carbons fell above and below the slit, through which, 

 therefore, only radiations from the arc itself passed. An attempt was 

 made to minimize the spectrum of carbon together with its impurities 

 by keeping the positive electrode loaded with the metal under in- 



"McLennan and Shaver, Proc. Roy. Soc. A., Vol. 100, 1921. 

 12G. A. Hemsalech and A. de Gramont, Phil. Mag., Feb., 1922. 



