of the Bessemer -flame. 87 



The same figure shows also the result of a direct comparison 

 of the Bessemer-spectrum with that of manganese oxide : those 

 lines only of the manganese oxide spectrum are drawn which 

 were seen to coincide with the Bessemer (or Spiegel) lines. 



It is worthy of remark that the identification of these iron 

 lines proves that iron may exist as vapour at a temperature below 

 the melting-point of iron, since the Bessemer-flame is not hot 

 enough to melt wrought iron. The presence of a few only out 

 of the 180 lines which constitute the spectrum of the electric 

 spark between iron poles, is only the same thing that we observe 

 in the case of sodium, lithium, and thallium when the spectrum 

 is produced at a low temperature. 



The following lines in the Bessemer-spectrum remain un- 

 identified : — 



nA(\r\ f Jted lines not always seen. 

 S} Fine lines. 

 /3{X} Stron S redli 



S} Fine lines. 







6010" Faint line. 

 6012 , 



6006 J 

 7 5972 Strong red line. 

 5917 Faint line 



^^H Faint yellow lines. 



[58191 

 8< 5807 ^Bright light-green group. 

 L5790j 

 T5547 

 U 5532 

 [5529 



5462] 



5454 > Three faint lines. 



5443 J 



5405 Faint line. 



5167 Fine blue line. 



4481 



1ac\a ^Blue lines. 

 4404 



4383. 



Two other blue lines, whose position could not be determined, 

 were also seen occasionally. 



