76 



following reasons : — 1. Ordinary flames cast a shadow in sun- 

 light, and it is difficult to believe that this can happen except 

 by the intervention of solid particles ; and carbon is the only 

 solid substance which can be present. 2. Elames artificially 

 made to contain solid particles have been shown to be as trans- 

 parent as an ordinary gas flame. 3. If there be only heated 

 vapour in such a flame, soot ought not to be deposited on a 

 strongly heated rod held in it, whereas, as a matter of fact, 

 soot is deposited under such conditions. 4. By causing one 

 -flame to meet another the particles of carbon may be made as 

 it were to agglomerate, and a much coarser-grained soot is 

 obtained. 5. The quantity of hydrogen in soot is extremely 

 small, very much less than in the densest hydrocarbons. The 

 small luminosity of a Bunsen flame has been usually ascribed 

 to the complete combustion of the solid particles of carbon, 

 but it has been shown that the previous admixture of nitrogen 

 and other inert gases with the coal gas renders the flame almost 

 non-luminous. This is probably due to the lowering of 

 temperature produced by the admixture of a large volume of 

 inactive gas, for the luminosity can be restored by heating the 

 mixture before burning. In a Bunsen flame the added heat 

 due to increased combustion probably overbalances the cooling 

 effect of the admixture of air. That the heating of gas before 

 it reaches the burner greatly increases the illuminating power 

 is well known, and this principle has been applied practically 

 in recent improvements in gas-lighting in England and 

 Europe. 



