248 Professor Arthur Smithells on 



hanging horizontally a narrow cylinder with open ends and 

 made of the thinnest platinum- foil in the hottest part of the 

 flame given by a Bunsen-burner with fully open air-holes. 

 In order to keep the radiation of the hot metal from the eye 

 one looks through a narrow tube laid in the direction of the 

 axis of the platinum cylinder, and the experiment is per- 

 formed in a dark room. The gas within the tube appears 

 dark to the eye and different from that of the luminous flame. 

 Gases give the sams result when in tubes made white-hot by a 

 coal-fire." Hittorf refers to the observations of Wedgewood 

 (vide supra) and Mellon i (Pogg. Ann. lxxv. p. 62, 1848) on 

 the same subject. 



In 1882 Werner von Siemens, unaware of the work just 

 referred to, made experiments with a Siemens regenerative 

 gas-furnace and found that oxygen, nitrogen, steam, and 

 carbon dioxide possessed no luminosity at a temperature 

 somewhat above 1500° C. In confirmation of this conclusion 

 he points out that, when a flame is made hotter by increasing 

 the air-supply or previously heating it, the visible flames be- 

 come smaller although the products of combustion must be 

 hotter. If the luminosity of flame were due to the simple 

 glowing of the products of combustion, the visible flame 

 should increase and not diminish in size under the conditions 

 just described. The luminous region of flame is therefore 

 conterminous with the region of chemical action and (he 

 argues) is directly due to it. He adds : — " If we assume that 

 the gas molecules are surrounded by a sheath of aether an 

 alteration of these sheaths of aether must take place when two 

 or more such molecules combine. The resultant movement 

 of the aether particles must be compensated by vibrations 

 which can form the origin of the flow of heat and light waves. 

 In quite a similar way we can picture the light-effects which 

 appear when an electric current is passed through gases." 



The Internal Condition of Flame. 



The experiments of Hittorf and Siemens may, I think, be 

 held to establish two facts — (i.) That air, carbon monoxide, 

 steam, and carbon dioxide emit no sensible light when at an 

 average temperature of 1500° C. or perhaps 3000° 0. ; (ii.) That 

 the luminous part of flame is conterminous with the region of 

 chemical action. They do not establish, I think, a third and 

 most important conclusion, namely, that the luminosity is due 

 directly to chemical action and not to heat. Before such a 

 conclusion can be justified we must be sure that the products 

 of combustion in the flame do not exceed 1500° or 3000° 0. 



It is assumed by Hittorf and Siemens, and has also been 



