1894.] 



Flame Spectra at High Temperatures. 



195 



The spectra studied extended from the red potassium line X 7697, 

 and on some of the plates to about the line P on Cornu's map of the 

 solar spectrum, X 3380'8 ; but the least refrangible line photographed 

 -was that of lithium X 6707. The bands and lines in various spectra 

 taken at Crewe and Dowlais have been measured, and their wave- 

 lengths determined. Descriptions of the spectra and how they were 

 obtained are given. The description of each band and line measured 

 is given, with its wave-length, its origin, and other references. Photo- 

 graphs of the spectra are presented, and a map has been drawn for the 

 identification of the lines on these photographs. About ninety- two lines 

 were identified with lines in the solar spectrum, with lines in Kayser 

 and K/unge's map of the arc spectrum of iron, and on spectra from 

 .steel and ferric oxide heated in the oxyhydrogen flame. 



The Constitution of the Bessemer Spectrum. 



The spectrum is a complex one which exhibits differences in con- 

 stitution during different periods of the " blow," and even during 

 different intervals in the same period. As originally observed by 

 Watts, the spectrum differs in different works, the difference being 

 due to temperature and to the composition of the metal blown. 



The lines of the alkali metals, sodium, potassium, 

 and lithium, are seen unreversed on a bright con- 

 tinuous spectrum caused by carbon monoxide. The 

 C line of hydrogen and apparently the F line were 

 ! seen reversed during a snowstorm. 

 r Bands of manganese are prominent, overlying the 

 J continuous spectrum of carbon monoxide. There are 

 ] lines of carbon monoxide, manganese, and iron, also 

 I those of the alkali metals. 



r The spectrum is the same as the foregoing, but the 

 J lines of iron are not so strong and not quite so well 

 ] defined. Some of the short lines disappear. The 

 v. lines of the alkali metals are visible. 



During the 

 first period. 



During the 

 scond period 

 The "boil." 



During the 



third period. 



The " fining 



stage." 



Tt is also probable that some of the bands of manganese oxide are 

 Dresent, but they are obscured by the continuous carbon monoxide 

 .spectrum. No absorption bands were seen, no nitrogen bands, nor 

 bands of calcium and magnesium oxide, neither did the lines of these 

 metals appear. There is no trace of cobalt, nickel, chromium, or 

 copper; certain carbon bands overlie those of manganese, and are 

 recognised by measurements of their edges. Some of the lines not 

 identified by Watts prove to be iron lines, others belong to man- 

 ganese. The manganese bands are all degraded towards the red, the 

 carbon bands towards the blue. 



o 2 





