1 70 Prof. Mitscherlich on the Spectra of Compounds 



those which, presupposing a deportment analogous to the alka- 

 lies, must belong to the metals, exhibit a regular shading by 

 which otherwise only the spectra of the compounds are charac- 

 terized. But the number of these supposed new metals would 

 soon have reached that of those at present known. The disco- 

 very of such metals appeared easiest in bismuth compounds. 

 Two lines occurred — one which was soon ascertained to be that 

 of the already known thallium, and a second near the division 

 77 of the scale. By precipitations, sublimations, reductions, 

 and oxidations conducted in the most careful manner, I could 

 not separate from bismuth the metal belonging to the line 77. 

 Subsequent investigations showed that, according to the manner 

 in which the spectrum was prepared, this line in the spectra of 

 bismuth compounds occurred or disappeared ; it always occurred 

 when deoxidizing gases came in contact with the metal at a high 

 temperature, and disappeared with oxidizing gases. The line 

 in question can be obtained quite pure when metallic bismuth 

 is volatilized in hydrogen and the latter is ignited. From 

 these observations it follows that this line belongs to bismuth as 

 metal. 



I have investigated most bodies and compounds in which a 

 spectrum might be supposed, and by different methods, in order 

 to avoid the chance of obtaining mixed or erroneous spectra by 

 foreign influences. As a source of light, I used flame and the 

 electrical discharge. I will briefly adduce the methods which I 

 used for obtaining the spectra, and designate them by numbers 

 for convenience of subsequent reference. 



1. A method which I have described in the memoir men- 

 tioned above. From a tube closed at the top and provided with 

 a very small aperture at its lower bent end, a solution continu- 

 ously flows, by means of a wick of very fine platinum wire, into 

 a flame of coal-gas, or of hydrogen ; the gas emerges from a 

 narrow aperture. The liquid in question brought by the wick 

 rapidly evaporates in the flame. Carbon does not separate in 

 the coal-gas ; it is prevented from doing so by the presence of 

 water, by which hydrogen and carbonic oxide are formed. 



2. The substances are brought into a coal-gas flame which 

 burns in oxygen. An oxy hydrogen burner is used, from whose 

 middle aperture coal-gas, and from whose external ring oxygen 

 issues. Either by regulating the oxygen or the position of the 

 substances, they can be introduced into a reducing or oxidizing 

 flame. The spectrum of carbon which would here be formed 

 disappears almost entirely, in consequence of the formation of 

 carbonic oxide, which again, in virtue of the feeble intensity of 

 its light, does not lessen the purity of the phenomena when a 

 substance to be volatilized is brought into the flame. 



