1903.] The Spectra of Metallic Arcs in an Exhaiisted Globe. 255 



appear with great intensity. Flutings of nitrogen also appear in some 

 of the photographs. 



The enhanced line 4481*3 is of about the same intensity as &i, and 

 stronger than the lines 4703 and 4352. It is also notably sharp as 

 compared with its usually hazy appearance in the spark spectrum of 

 the metal. 



Two conspicuous pairs of sharply defined lines which appear at 

 approximate wave-lengths 4385, 4391 and 4428, 4434 cannot be traced 

 to any known impurities, and a comparison with the spark spectrum 

 leaves no doubt that they are true enhanced lines of magnesium. 

 They are very ill-defined in the ordinary spark, but are distinctly 

 visible as double lines in a photograph of the spark spectrum when a 

 small amount of self-induction is introduced into the secondary circuit, 

 though disappearing when the self-induction is increased. The line at 

 4391 is apparently identical with the enhanced line recorded by 

 Sir Norman Lockyer* at 4395. Porterf has observed lines at approxi- 

 mate wave-lengths 4390, 4430 and 4434 in the spectrum of the 

 magnesium arc in an atmosphere of ammonia or oxygen which prob- 

 ably correspond with three of the above four lines, but states that he 

 has not succeeded in identifying them. 



Another spark line of small intensity recorded by Thalen and others 

 at A, 4587 does not appear in the photographs of the spectrum of the 

 arc in the exhausted globe, but Porter finds it among the lines of the 

 magnesium arc in ammonia. 



It is important to note also that in some of the photographs there 

 is a distinct line coincident with the F (H/s) line of hydrogen. 



[Note added July 16. The identification of this line has been con- 

 firmed by visual observations showing the presence of both H/3 and 

 H..] 



Another point of importance, in considerations as to temperature 

 conditions, is that the flutings and enhanced lines appear to originate 

 in different parts of the arc. 



Zinc. 



The strong triplet of lines at AX 4810*71, 4722*26, 4680*38, and the 

 adjacent line 4630*06, appear as strong lines, but are generally sharper 

 than the corresponding lines in the spectrum of the arc in air at 

 atmospheric pressure. Most of the other lines in the region D to K 

 disappear altogether or are greatly reduced in intensity. 



The well-known spark lines 4912*2, 4924*8 (Thalen's wave-lengths 

 corrected to Rowland's scale) make their appearance with considerable 



* 'Roy. Soc. Proc.,' vol. 61, p. 163. 

 f ' Aetrophys. Journ.,' vol. 15, p. 277. 



T 2 



