240 Mr. G. A. Hemsalech on the Line Spectrum of 



3. It has been shown that nitrogen is one of the deter raining 



factors in the emission of the characteristic cone spec- 

 trum of iron. In order to account for the origin of this 

 spectrum an hypothesis has been proposed based on the 

 assumption that a nitride of the metal is formed in 

 the explosion region. §§ 3, 4, & 5. 



4. The presence of ammonia in the air-coal gas and oxy-coal 



gas flames reduces the reaction between nitrogen and 

 iron to a considerable extent. The decomposition of 

 the ammonia in the cones of these flames seems to 

 initiate the emission of the so-called cyanogen bands. 



5. The presence of oxygen in the medium surrounding the 



air-coal gas flame is essential in bringing about the for- 

 mation of iron nitride in the explosion region of this 

 flame. §§ 8, 9, & 10. 



6. No reaction takes place if all the oxygen is withdrawn 



from the gas mixture of an air-coal gas flame burning 

 in an atmosphere of air. § 11. 



7. Heating of an air-coal gas flame by means of a graphite 



rod kept at a bright red heat only slightly increases 

 the intensities of the iron lines. Cooling of the flame 

 decreases their intensities. § 12. 



8. Experiments are described which prove that the higher 



velocities of the oxy-hydrogen and oxy-acetylene flames 

 do not account for the absence in the cones of these 

 flames of the supplementary spectrum of iron. § 13. 



9. The relative behaviour of Class I. lines, as brought out in 



the course of this investigation, points to the existence 

 in this class of two different types of lines. 



§ 15. Concluding Eemares. 



The results of the present investigation serve to emphasize 

 the importance of chemical reactions in the excitation of 

 both line and band spectra. In particular, the fact that 

 a well-developed line spectrum of a metal vapour can be 

 obtained by this means at a comparatively low temperature 

 may be worthy of serious consideration in the interpretation 

 of astrophysical phenomena. 



The subject has, so far, received relatively little attention, 

 and the whole problem rega rding the role and modus operandi 

 of chemical actions in the production of light radiations still 

 awaits solution. 



