472 Dr. T. Carnelley on the 



also Li and Na, are exceptions to the latter of these state- 

 ments*. 



Spectra of Chemical Compounds.' — Lockyer (Proc. Roy. 

 Soc. 1873, p. 285) has investigated the spectra of metals when 

 combined with various negative elements, and finds that regular 

 relations obtain. Thus in the case of PbF 2 , PbCl 2 , PbBr 2 , 

 and Pbl 2 it appears that the length and number of the metallic 

 lines observed diminish as the atomic weight of the non- 

 metallic element increases. This was also found to hold good 

 with the haloid compounds of the metals of the alkalies and 

 alkaline earths. 



Mitscherlich (Pogg. Ann. cxxi. p. 459 ; Phil. Mag. [4] 

 xxviii. p. 169) has also pointed out a singular relation between 

 the atomic weights of the haloid compounds of Ca, Sr, and Ba 

 and the distance between the chief lines of their spectra. He 

 finds that in the case of Ba, the distances between the most 

 prominent lines given by the Ba haloid compounds respectively 

 are directly proportional to the molecular weights of these 

 compounds, whilst for the haloid compounds of Sr and Ca 

 these distances are inversely proportional to the molecular 

 weights of the compounds. 



Colour and Tinctorial Properties of Carbon Compounds, — 

 All organic colouring-matters are nearly related in composi- 

 tion. They all belong to the aromatic group of compounds, 

 and, on reduction, become colourless, owing to combination 

 with hydrogen. All quinones and azo-compounds are coloured; 

 and they all combine with nascent hydrogen and become 

 colourless. In these bodies, too, the or N is combined in 

 a manner different to what they are in other compounds. 

 Thus we find that all bodies (quinones) which contain 



united in the following manner, ( n y ), are coloured, and 

 combine with H, thus: ( rycr)* forming colourless com- 

 pounds (hydroquinones). We may also obtain similar colour- 

 less bodies by using monatomic alcohol-radicals instead of 

 H. In azo-compounds we have the N combined thus, 

 ( — N = N — ) ; and they also combine with H to form colourless 



* Since the above was in proof, I have noticed a paper by Ciamician 

 ( Wien. Akad. Ber. lxxvi. p. 429; Chem. Soc. Journ. Abstracts, Sept. 1879) 

 in which, from a study of the spectra of thirty-one elements, he concludes 

 "that the spectra of' chemically analogous elements correspond either 

 singly or in groups, and that therefore every group of elements has a spec- 

 trum which only differs from those of the various members of the group 

 in that the homologous lines or groups of lines are transposed either in one 

 direction or another " — and, further, that the properties of the elements 

 are a function, not only of their atomic weights, but also of the wave- 

 lengths of their rpsppctive lines. 



