558 On the Spectra of Carbon and Hydrocarbon. 



narrower Fe: spectrum was superposed on that of the Meeker 

 flame. 



Fig. 13, PI. XIII., is an enlargement from a negative 

 obtained with the prismatic spectrograph on which the 

 unknown bands are recorded. 



The least refrangible edge falls about midway between 

 the two strongest iron lines and appears to consist of a 

 double line. The second and stronger head is nearly coin- 

 cident with the iron line X 4891. A third head is also 

 somewhat vaguely indicated, but the most remarkable 

 feature is the ribbed or fluted structure of the band, the 

 close and regular sequence of which can be traced nearly 

 up to the carbon fluting (A, 4737'2) in the blue. 



The f aintness of the detail, however, makes the components 

 somewhat difficult to recognize. 



In addition to these bands, a well-marked single line 

 occurs close to the blue carbon band. Owing to the long 

 exposures necessary the " Swan " spectrum is greatly over- 

 exposed; and this accounts for the appearance of the bright 

 green carbon band (A, 5165), for an ordinary plate has little 

 sensitivity to this region. 



The approximate wave-length measurements are as 

 follows : — 



Fe : Comparison Spectrum. 

 X. 



4957-6" 



4939-0 



4924-1 

 f 4920'7 

 1 4919-1 



4891-3" 



4878-4 



4871-9" 



4860-0 



Unknown Bands. 

 X. 



(1) f 49420 

 14937-5 



(2) 4890-0 



(3) 4853 (?) 



(4) line 4743 



In order to show the line X 4743 well, the exposure must 

 be suitable. It is seen more clearly in fig. 14, PI. XIII., 

 where it is recorded at the expense of the bands. 



These features appear to be of constant occurrence in 

 association with the " Swan " spectrum as given by the 

 Meeker coal-gas flame. There does not seem to be any 

 reason for attributing them to the presence of some chance 

 impurity, and the inference is that they are probably due to 

 carbon or one of its compounds. The fainter details are 

 liable to variations and changes in relative intensities, but 



