1894.] Flame Spectra at High Temperatures. 193 



It will be readily understood that the bands can be measured -with 

 far less accuracy than lines, and that they are subject to some degree 

 of variation in width, according to variation in the time of exposure 

 and the temperature. 



A striking group of lines, the most persistent in the whole of these 

 spectra, is situated in the violet. The following measurements were 

 made : 



4036-5 4034-9 Angstrom, also Cornu. 



xmoon / 4032-91 A , .. 



4032 ' { 4031-8 )^ ngStr0m - 

 4029-5 4029-4 Angstrom. 



Another line is just visible about 4031'8, but it is so close to 4032'0 

 that it could be discerned only when the extreme points of three very 

 strong lines were examined, and the plate was in perfect focus for 

 that region. The whole group of lines appears as two bands very 

 closely adjacent, or in the manganese oxide spectrum as one band 

 with the centre appearing as if reversed, the less refrangible edge of 

 the band being very strong and sharp, the more refrangible being 

 degraded and diffuse. These lines remain after the bands in the 

 yellow and green have disappeared from the photographs, but the 

 result may be quite otherwise with eye observations, owing to the 

 greater visibility of the yellow over the violet rays. 



Photographs of the spectra obtained with a dispersion of four 

 quartz prisms of 60, and lenses of 15 inches in focal length, are pre- 

 sented with the paper. 



V. "Flame Spectra at High Temperatures. Part III. The 

 Spectroscopic Phenomena and Thermo-Chemistry of the 

 Bessemer Process." By W. N. HARTLEY, F.R.S., Royal 

 College of Science, Dublin. Received May 4, 1894. 



(Abstract.) 



The flame issuing from the mouth of a Bessemer converter was 

 first investigated by Sir Henry Roscoe* in 1863 ; by Lielegg,f and by 

 Marshall Watts in 1867 ;t by Tunner, J. M. Silliman, Rowan, || Von 



* 'Literary and Phil. Soc., Manchester, Proc.,' vol. 3, p. 57, and 'Phil. Mag.,' 

 Tol. 34, p. 437. 



f ' Sitzungsberichte Kaiserl. AkademiederWissenschaften,' Wien, vol. 56, Part II. 

 J ' Phil. Mag.,' vol. 34, p. 437. 

 ' Dingler's Polytech. J.,' vol. 178, p. 465. 

 f| ' Phil. Mag.,' vol. 41, p. 1. 

 VOL. LVI. O 



