266 itEPOBT— 1880. 



Mr. Atfcfielil's paper induced Morren to take up the question. Starting 

 with the intention of disproving Attfield's conclusions he ended by being 

 convinced that he was right. Entirely confirming Attfield's experiments 

 Morren satisfied himself that the candle spectrum was really due to carbon, 

 and not to a hydrocarbon. He especially testifies to the brilliancy of the 

 spectrum in a flame of cyanogen and oxygen. 



Dibbits had already, before Morren, arrived at the same conclusion. He 

 was the first to furnish an answer to the theoretical objection which can 

 be raised against Attfield's explanation, and which at first sight appears 

 serious. The temperature of an ordinary flame is certainly not high 

 enough to volatilise carbon. How, then, can carbon be present in the state 

 of vapour and give us a discontinuous spectrum. Dibbits explains the 

 difficulty by saying that carbon exists before combustion, combined with 

 hydrogen ; after combustion it is combined with oxygen, and it must 

 therefore have existed during a certain stage of transition in the form of 

 simple carbon uncombined. During this stage of transition it gives us the 

 carbon spectrum. He sujjports the explanation by the fact that a flame 

 of carbonic oxide does not show the spectrum, because in it the carbon is 

 never in a free and uncombined state. Dibbits' view has received a good 

 deal of support by some very interesting experiments made recently by 

 Gouy (' C. R.' Ixxxiv. p. 231). In a Bunsen flame, the spectrum under 

 discussion is confined to a naii'ow cone ; Gouy charges the air before it 

 enters the burner with powdered salts in a finely divided state, and shows 

 that at the same place where the candle-spectrum appears we may obtain 

 the spectra of bodies which it would be impossible to volatilise in the flame. 

 Thus the salts of iron, cobalt, manganese, silver, give lines which we know 

 to be due to these metals, as they are found in their spectra obtained 

 by means of electric sparks. Even platinum salts give a spectrum in the 

 blue cone, but it is not certain that this spectrum is really due to platinum 

 in an uncombined state. Gouy believes that these experiments indicate a 

 very high temperature in the blue cone of a Bunsen flame, but we think 

 an explanation, identical with the one given by Dibbits for carbon, will be 

 found more plausible. V. de Willigen had already, before Attfield, made 

 some not quite satisfactory experiments tending to show that the candle- 

 spectrum is due to carbon. Pliicker and Hittorf, as well as Wiillner, arrive 

 (after Attfield) at the same conclusion. Watts has made a long series of 

 experiments, all tending to support Attfield's view. In addition to the 

 gases experimented on by Attfield he took carbonic tetrachloride and 

 obtained from it the candle-spectrum. Lockyer has quite recently 

 experimented with the same gas and shown that this much discussed 

 spectrum can be obtained, when a strong spark does not reveal the 

 presence of hydrogen. Huggins' attention was drawn to this spectrum 

 through his observations on comets, and he also obtained the candle- 

 spectrum in a current of cyanogen gas, and therefore considered the 

 spectrum to be due to carbon. 



On the whole it may be said that, from the publication of Attfield's 

 paper until the year 1875, every spectroscopist, whether he was a chemist 

 or a physicist, who had set to work to decide the question, came to the 

 conclusion that the candle-spectrum was a true spectrum of carbon, and 

 the question appeared to be settled. In the year 1875, after Angstrom's 

 death, Thalen published a paper in which he describes some experiments 

 jointly made with Angstrom. In consequence of these experiments the 



