1882.] 



On the Spectrum of Carbon. 



4.09 



that on reversing the current, it shifted from one wire to the other, 

 and the wire about which it appeared was always the positive 

 electrode. In fact, the flame was the glow of the positive pole 

 attending a discharge in rarefied gas ; when the resistance of the 

 carbon thread became too great in proportion to the intensity of the 

 current, the discharge began to occur through the rarefied atmosphere 

 within the envelope of the lamp. The spectrum showed that this 

 atmosphere contained carbonic oxide. 



By interposing different flames between the incandescent lamp and 

 the slit of the spectroscope, we have been able to make some compari- 

 sons of the probable temperatures of the flames and filament. For 

 this purpose a lens of 3 inches focal length was placed 6 inches in 

 front of the slit, and an image of a horizontal part of the incandescent 

 carbon thread formed by it across the (vertical) slit. The appearance 

 in the field of view of the spectroscope was a narrow continuous 

 spectrum extending all across the field. When a flame was interposed 

 between the lens and the slit, the bright lines of the flame were seen 

 above and below the narrow continuous spectrum, and in some cases 

 were continued across it, or were seen reversed upon it. When the 

 flame was that of a Bunsen burner in which was a platinum wire 

 with sodium carbonate, the yellow sodium lines were seen bright 

 above and below the continuous spectrum of the carbon thread, but 

 reversed where they crossed it. When lithium was substituted for 

 sodium in the flame, the red lithium line was also seen bright above 

 and below the continuous spectrum, but reversed where it crossed it. 

 When an oxyhydrogen jet was substituted for the Bunsen burner, and 

 sodium carbonate held in it, the yellow sodium lines were not only 

 bright above and below the continuous spectrum of the carbon, but 

 showed as bright lines where they crossed it, in fact they were con- 

 spicuously brighter than the carbon. When coal-gas was substituted 

 for hydrogen in the jet, the same appearance presented itself, only the 

 sodium lines were not so much brighter than the carbon as they were 

 before. Fifty Grove's cells were used with the incandescent lamp, 

 which were as many as could be used without danger of rupturing the 

 threads. When barium chloride was held in the hydrogen flame fed 

 with only a little oxygen, the bright green line of barium (wave-length 

 5534) was well seen above and below the continuous spectrum, but 

 could not be traced either bright or dark across it. When a flame of 

 cyanogen burning in air was interposed, the bright bands of that flame 

 could be seen above and below the continuous spectrum, but could not 

 be traced either bright or dark across it. When sodium carbonate 

 was held in this flame the yellow sodium lines were seen feebly 

 reversed where they crossed the spectrum of the incandescent lamp. 

 We infer from these experiments that the emissive power of the 

 carbon thread for light of the refrangibility of the D lines is nearly 



