462 Mr. J. Evershed's Experiments on 



other end being open to the atmosphere. This was suspended 

 horizontally, and a Bunsen flame brought under the central 

 part ; when about 5 or 6 cm. of the tube had become nearly- 

 red-hot, the flame was held for a few seconds under the part 

 containing the iodine, which immediately volatilized, filling 

 the whole tube with coloured vapour. A bright reddish 

 glow then appeared in the heated portion of the tube. 



Exp. II. — Iodine was heated in a tube as in I. but closed to 

 the atmosphere, and in which air was replaced by nitrogen. 

 No difference was seen in the emission of light. 



Exp. III. — The same as II., excepting that hydrogen was 

 substituted for nitrogen. In this case also the glowing appeared, 

 the same as before, but it could not be maintained as combina- 

 tion took place between the hydrogen and iodine, and the 

 colourless hydriodic acid formed emitted no light. 



Exp. IV. — In C0 2 the same result was obtained as in 

 nitrogen. 



Exp. V. — A thick glass tube containing iodine was exhausted 

 with an air-pump until the pressure fell to about 20 mm. It 

 was then heated so as to expel the remaining air by volatilizing 

 a portion of the iodine, and then sealed off. The whole length 

 of the exhausted tube was next warmed up until the density 

 of the iodine vapour was sufficient to give the usual deep 

 violet colour. On strongly heating a short section of the 

 tube at this stage, the usual reddish-yellow glow r appeared 

 exactly as in the other experiments. 



These preliminary experiments, therefore, as far as they go, 

 entirely favour the view that the glowing is determined by 

 heat alone, and not by chemical " luminescence. 9i For the 

 phenomenon appears to be quite independent of the gas in 

 which the iodine is heated, and in III., where chemical com- 

 bination actually occurs, the light, instead of being intensified, 

 is extinguished. 



To determine the character of the light, things were 

 arranged as in I., the glowing tube being observed with a 

 spectroscope, having a lens in front to throw an image of the 

 hot part of the tube across the slit of the instrument. Before 

 heating the iodine a faint streaky spectrum, due to glowing 

 opaque particles in the glass, was all that could be seen ; but 

 immediately the iodine was vaporized a bright spectrum 

 shone out. This appeared perfectly continuous, and similar 

 to that given by a red-hot iron wire, with which it was 

 directly compared. No sign of resolution into lines could be 

 made out even with a highly dispersive train of five prisms. 



Next a ray of white light from the flame of a paraffin lamp 

 was made to pass through the glowing tube and enter the spec- 

 troscope, after having suffered absorption by the heated vapour. 



