1903.] The Spectra of Metallic Arcs in an flxhamted Globe. 253 



" The Spectra of Metallic Arcs in an Exhausted Globe." By 

 A. FOWLER, A.RC.Sc., F.R.A.S.,. and HOWARD PAYN, F.K.A.S. 

 Communicated by Sir NORMAN LOCKYER, K.C.B., LL.D., 

 F.R.S. Received June 10, Read June 18, 1903. 



[PLATE 14.] 



Working on the supposition that the bands which occur in the arc 

 spectrum of magnesium are due to possible combinations of the metal 

 with gases in the outer part of the arc, we have recently attempted to 

 obtain the line spectrum free from bands by photographing the spectrum 

 when the arc is passed in an exhausted globe. It is found, however, 

 that while the band beginning at A 5007*5 is sometimes greatly reduced 

 in intensity under these conditions, the bands attributed to ** magnesium 

 hydride " by Messrs. Liveing and Dewar* are very strongly developed. 



Besides this, the characteristic spark, or " enhanced " line at A. 4481 '3, 

 whicb is almost invisible in the ordinary arc in air, is one of the strongest 

 lines in the spectrum. Special interest attaches to this line on account 

 of its application to the determination of relative stellar temperatures, 

 based on the variations in the spectrum of magnesium under different 

 conditions, which were first systematically investigated in relation to 

 temperature by Sir Norman Lockyer in 1879.f 



The occurrence of the line 4481 in the arc spectrum under special 

 conditions was first recorded by Liveing and Dewar in 1888J, and has 

 since been investigated by other observers. Crewg found that the line 

 appeared in the spectrum of magnesium when the arc was surrounded 

 by an atmosphere of hydrogen, and a little later Porter) | found among 

 other results, that a similar effect was produced by oxygen. Experi- 

 ments by Sir Norman Lockyer and the writers have confirmed these 

 observations. 



More recently, Hartmann and EberhardU have observed that enhanced 

 lines appear in the arc spectra of magnesium, zinc, and cadmium when 

 the arc is made to pass in water. The spectra obtained in this 

 manner are stated to be almost identical with those obtained when 

 the arc is passed in a current of hydrogren, arid the opinion is ex- 

 pressed that the change is produced by hydrogen released by elec- 

 trolysis around the electrodes in water. Hartmann has since found 



* ' Roy. Soc. Proc.,' vol. 32. p. 196. 



t 'Roy. Soc. Proc./ vol. 30. p. 29; also ' Chemistry of the Sun/ p. 242; and 

 * Inorganic Evolution/ p. 75. 



t ' Roy. Soc. Proc./ vol. 44, p. 241. 



' Astrophys. Journ./ vol. 12, p. 167, 1900. 



|| ' Astrophys. Journ./ vol. 15, p. 281, 1902. 



IT ' Sitz. der Preuss. Akad. der Wiss./ 1903, IV, p. 40 ; ' Astrophys. Journ./ 

 vol. 17, p. 229. 



VOL. LXXII. T 



