and of Simple Substances. 1 77 



comparison of the spectra of all compounds has hitherto been 

 impossible. 



The metals whose compounds are decomposed at such a low 

 temperature, and hence only give the spectrum of the metal 

 itself, are potassium, sodium, lithium, magnesium, zinc, cadmium, 

 silver, and mercury. Potassium, sodium*, lithium, as metal, 

 cyanides or chlorides, investigated by method 6 (volatilization 

 in the glass tube) or by method 2 (H and 0), give, especially 

 in the spectra produced by the first method, several lines in addi- 

 tion to those depicted by other observers. 



By method 2 (H and 0) there are further obtained the spectra 

 of magnesium, zinc, cadmium, silver, mercury — by the use of 

 chloride of magnesium and chloride of ammonium, of chloride 

 of zinc, of carbonate of cadmium, of cyanide of silver, and of 

 cyanide of mercury. By the use of other mercury compounds, 

 such as chloride and sulphate of mercury, no lines can be 

 obtained by this method. In burning magnesium, the same 

 spectrum is obtained as in the methods adduced. 



By method 8 (liquid electrodes) these spectra can also be 

 more or less well obtained with the use of various solutions. 

 In the spectra obtained by this method other lines occur, which, 

 however, are for the most part more feebly luminous than those 

 formed by the flame. If zinc is burned in iodine, a brightness 

 only is observed in the spectrum, doubtless produced by ignited 

 particles of iodide of zinc. 



That in the case of sodium compounds the metal actually 

 gives the spectrum, I have shown in the paper already mentioned. 

 I have found the same when I obtained this spectrum with 

 the use of the metal by method 7 (solid electrodes), excluding 

 oxygen. That the spectrum found with the use of the com- 

 pounds of magnesium, zinc, cadmium, silver, and mercury is in 

 each case the spectrum of the metal itself, I proved in the 

 same manner j\ 



From the fact that in sodium compounds the metal gives 

 the spectrum, I thought myself justified, in the above paper, in 

 expressing the opinion that in the oxygen compounds of ba- 

 rium, strontium, and calcium the spectrum is also produced by 

 the metal itself. This opinion, as I have already stated, has not 



* Brightness without any shading, as formed in the sodium and potassium 

 spectra, and which I could only ascribe to the ignited solid particles in 

 the flame, I have omitted in the spectra depicted. 



t The objection urged against me in an English paper ('The Photo- 

 graphic News') after the publication of my former paper, that spectra 

 produced by the volatilization of carbonate of soda, chloride of sodium, &c. 

 which are brought into heated tubes could not have been formed, because 

 the temperature is too low for the volatilization of these salts, does not 

 hold ; for I have myself observed the vapours. 



