and of Simple Substances. 189 



If the bodies are contained in the flame in such large quan- 

 tity that they cannot volatilize completely, or if they are not at 

 all volatile, they become white-hot and give the spectra of all 

 ignited bodies, that is, a perfectly continuous one, without dark 

 or bright lines. The former is the case if phosphorus is inves- 

 tigated by method 6 (volatilization in the glass tube, and this 

 somewhat strongly warmed) ; for the latter, the carbon separated 

 from hydrocarbons is the best example. 



If such a body, which is brought in excess into the flame and 

 hence does not evaporate sufficiently, has only few lines in the 

 spectrum (so that an absorption spectrum of it is otherwise easily 

 obtained, since in the middle there is a brightly luminous body, 

 and round about the vapours of this body), its absorption spec- 

 trum is obtained. This is the case if sodium is burned, or much 

 sodium volatilized in a hydrogen flame. This phenomenon can 

 best be investigated if more or less iodine is brought into the 

 hydrogen flame. If there is only little iodine, the new spectrum 

 which I have found is obtained ; but if there is much iodine, the 

 absorption spectrum is obtained. The middle of the flame is 

 white-hot; the white light must pass through iodine vapours 

 and these absorb a part, as is the case if a candle-light is viewed 

 through iodine vapours. 



It is not every volatile body which, introduced into the flame, 

 gives a spectrum. For particular bodies, a very high tempera- 

 ture is necessary to heat them so strongly that they disengage 

 light. This is best seen with mercury ; for mercury salts, ex- 

 cepting cyanide of mercury, investigated by methods 1 (wick of 

 platinum wire), 2 (H, or coal-gas, and 0), and 6 (volatilization in 

 the glass tube), hence at a lower temperature, give no spectrum ; 

 while, when heated by methods 7 (solid electrodes) and 8 (liquid 

 electrodes), they produce a very bright spectrum. This is fur- 

 ther observed in the case of oxygen, nitrogen, and other gases, 

 which only by the electrical discharge are sufficiently heated to 

 give a spectrum. 



To investigate whether solid bodies did not also produce 

 spectra, I caused white light to pass through a gold-leaf trans- 

 parent with blue light, or through gold which was precipitated 

 in extremely dilute solution, but I could not observe either dark 

 or bright lines in either experiment. 



Berlin, February 18G4. 



