[ 444 ] 



LVII. The Spectra of Chemical Coynpounds, 

 By James Moser *. 



IS spectral analysis indeed entitled to be ranked as a branch 

 of qualitative analysis ? 



Is the spectrum of a compound the sum of the spectra of 

 the component elements ? 



From an observation of the spectrum of a compound, is it 

 possible to determine the nature of the elements in the com- 

 pound, or has each compound a characteristic spectrum of 

 its own? 



The discoverers of spectral analysis, Kirchhoff and Bunsen, 

 in their first memoir, said that ^^ differences in the compounds 

 of any given metals have no influence upon the position of 

 the lines indicative of the presence of the individual metals.'* 

 This opinion has been generally accepted as true. The possi- 

 bility of each compound possessing a characteristic spectrum 

 was, however, recognized by Kirchhoff and Bunsen. Else- 

 where they say, "It is possible that the salts which we vola- 

 tilized suffered decomposition at the temperature to which 

 they were raised in the flame, and that the lines which we 

 observed were really due to the presence of free metals ; it is 

 possible also that each chemical compound may show lines 

 other than those caused by the elements composing iff. 



The object of the present paper is to show that each chemical 

 compound really possesses a characteristic spectrum of its own. 

 A chemical compound is characterized by " the difference 

 between its properties and the properties of its component 

 elements . . . ; the properties of the elements are lost in those 

 of the compound ; the individuality of the elements has dis- 

 appeared in the formation of a new body with new pro- 

 perties "J. 



Those physicists who have studied the subject now under 

 consideration have generally made observations on emission- 

 spectra only. 



This method is open to serious objection, and must be 

 superseded, if trustworthy results are to be obtained, by the 

 study of absorption-spectra. 



In studying the literature connected with the spectra of 

 chemical compounds, it is interesting to observe how the ori- 

 ginal view of Kirchhoff and Bunsen is gradually abandoned. 

 A. Mitscherlich found that the metallic oxides and chlorides, 

 when brought into the flame, afforded spectra different from 



* Translated and condensed from Pogg Ann. clx. pp. 177-199, by M. 

 M. Pattison Muir, The Owens College, Manchester, 

 t Pogg. Ann. cxiii. 381. 

 \ Hofmann, Einkitung in cl. mod. Chemie, Cap. iii. 



