548 Mr. C. ~W\ Raffety on some Investigations of 



The arc between soft-cored carbon rods 8 mm. in dia- 

 meter, connected to an alternating current supply of about 

 200 volts, was struck within a name of coal-gas in air, 

 the current being regulated by an inductance and a variable 

 resistance in circuit. It was thus hoped to intensify the 

 carbon bands and also to discover whether any modifi- 

 cations occurred in the hydrocarbon bands which are 

 produced under these conditions. 



An image of a horizontal arc was projected with a lens 

 on the slit of the spectrograph in the usual way, the arc 

 having been struck in a flame of coal-gas which completely 

 enveloped it. After careful adjustment it was found that 

 the normal blue region at the centre of the arc became 

 replaced by one of a pronounced green colour, and at 

 the same time the arc emitted a characteristic sound by 

 which the change could always be recognized. 



A visual examination of the spectrum revealed the carbon 

 flutings with especial brilliancy, the series of lines forming 

 them appearing much extended so as to fill the spaces sepa- 

 rating the groups. At the same time the red-orange bands 

 of the normal flame spectrum emerged from their obscurity, 

 being exalted in brightness until their " heads " became 

 quite conspicuous. The red line of hydrogen (Ha, \ 65630) 

 also appeared. 



In the more refrangible part of the spectrum the cyanogen 

 bands still possessed great brilliancy, there being, appa- 

 rently, sufficient access for atmospheric nitrogen in the 

 flame ; but together with them, the hydrocarbon bands were 

 well developed. 



Fig. 1, PI. XIII., is an enlargement from a spectrogram 

 of the normal carbon arc and the arc-in-flame, taken with a 

 grating spectrograph, showing the yellow and green bands 

 of carbon. 



The difference in intensity between the two spectra, how- 

 ever, is actually much greater than is here indicated as, in 

 order to produce clear photographs and avoid extreme con- 

 trast, longer exposure was given to the one of the normal arc. 

 Fig. 2, PI. XIII., shows the corresponding intensification 

 of the red-orange bands as compared with the normal arc. 



From experiment it appeared that the flame most favour- 

 able for the production of the carbon bands was the ordinary 

 luminous jet rather than the air-fed Bunsen or Meeker 

 flames ; and the arc was moved about from one part of 

 the flame to another, but was finally located about the 

 centre of the luminous region. 



