[ 1041 ] 
XXI. Flame Spectra at Fligh Temperatures .—Part III. The Spectroscopic 
Phenomena, and Thermo-Chemistry of the Bessemer Process. 
By W. N. Hartley, F.R.S., Professor of Chemistry, Royal College of Science, 
Didjlin. 
Received May 4, 1894—Read June 14, 1894. 
[Plate 14.] 
Contents. 
Page 
1. Introduction and Historical Notes.1041 
' 1 . Description of tlie instrument u,«ed and the method of observing and photographing 
spectra of the Bessemer dame.1047 
3. Observations made at Crewe and at Dowlais in 1893 1049 
4. Spectra of the Bessemer flame with wave-length measurements and descl■ipti^■e notes . 1058 
5. Lines of the alkali metals, and of hydi'ogen observed in the Bessemer flame .... 1067 
6. Lines and-bands of carbon and of carbon monoxide.1068 
7. Bands and lines of manganese observed in the Bessemei' flame. 1069 
8. Lines in the spectrum of the Bessemer flame, identified with lines in the solar spectrum 
and with iron lines.1071 
9. The constitution of the Bessemer spectrum.lO/d 
10. Cause of the non-appearance of lines at the commencement and termination of the 
“ blow ”. lO/.l 
11. The temperature of the Bessemer metal and flamo.1081 
12. The cause of the apj^earance of the manganese spectrum in all cases until the close of 
the “ fining period ”.1086 
13. The technical aspects of this investigation. 1088 
14. Summary and conclusions. 1089 
Introduction and Historiced Notes. 
It is well known that the llaine which issues from the mouth of the converting 
vessel used in the manufacture of steel by the Bessemer process has a very peculiar 
character ; in the first place it is intensely luminous, and of a singularly greenish- 
yellow hue at one phase of its existence, but subsequently the tint of the flame is 
amethyst coloured. 
Ill February, 1863, Sir H. E. PiOSCOE (‘Literary and Philosophical Society of 
mecccxctv.— A. 6 R 
