150 Profs. Liveing and Dewar. Spectra of the [Feb. 5 y 



not at present aware of any practicable method by which it can be- 

 got rid of. 



The arc was taken in different gases inside a small glass globe (act 

 in figure) about 60 millims. in diameter, blown in the middle of a tube. 

 The two ends of the tube (bb) were closed with dry corks, through 

 which were passed (1) the carbons (cc), inserted through two pieces 

 of narrow glass tubing, and a packing of platinum foil; (2), two 

 other glass tubes (del) through which currents of the gases experi- 

 mented with were passed. 



Spectrum of the Arc in Air. 



The arc taken in a globe of air gave a tolerably bright continuous 

 spectrum, superposed on which the green and blue hydrocarbon bands 

 were seen, also the seven bands in the blue (wave-lengths 4600 to 

 4502, Watts) as in the flame of cyanogen, three faint bands, or lines, 

 in the indigo (wave-lengths about 4364*5, 4371*5, 4381*5), and much 

 more brightly the six bands in the violet (wave-lengths 4220 to 4158,. 

 Watts) and five ultra-violet (wave-lengths about 3883*5, 3871, 

 3862, 3854*5, 3850.) Besides these bands, lines of iron, calcium,, 

 and sodium were visible. The arc in this case was practically taken 

 in a mixture of nitrogen and carbonic oxide, for in a short time the 

 oxygen of the air is converted into carbonic oxide, as was shown by 

 the flame which always filled the globe on stopping the arc for a 

 second or two. 



Spectrum of the Arc in Carbonic Acid Gas. 



On passing through the globe continuously a current of carbonic 

 acid gas, the seven bands in the blue, and those in the violet and ultra- 

 violet gradually died out until they ceased to be at all visible con- 

 tinuously, and when now and then momentarily seen were only just 

 discernible. On the other hand, the hydrocarbon bands, yellow, green, 

 and blue, came out stronger and were even brilliant. Lines of iron 

 and calcium were still visible. On stopping the current of carbonic 

 acid gas and allowing air to diffuse into the globe, the violet and ultra- 

 violet bands soon began to appear and presently became permanent 

 and bright, the hydrocarbon bands remaining bright. 



Spectrum of the Arc in Hydrogen. 



The globe had next a continuous current of dry hydrogen passed 

 through it. The arc, contrary to what would be expected from the 

 behaviour of the spark discharge in hydrogen, would not pass through 

 more than a very short space, very much less than in air or carbonic 

 acid gas. There was a tolerably bright continuous spectrum, no trace of 

 the seven bands in the blue, or of those in the violet or ultra-violet, and 

 no metallic lines, but a fairly bright line in the red, which we identi- 



