300 REPORT—1899, 
the absorption of the ultra-violet increased as the complexity of the 
molecules increases in any homologous series. It was found, however, by 
Soret and Rilliet that when the process of drying was rapidly executed, 
ethyl alcohol is not appreciably less diactinic than methyl alcohol. 
Ketones, it was found, are strongly adiactinic, but any differences which 
were perceived were not greater than might be attributed to small 
quantities of impurities which it is difficult to exclude. Pure ether is 
almost as transparent as pure water, to the ultra-violet spectrum. In 
dealing with haloid derivatives it was found that the substitution of one 
alkyl for another has very little effect upon their transparency, and this 
is particularly well marked in iodides. 
It is concluded that the action on the ultra-violet rays constitutes a 
very delicate test of the purity of an organic substance. 
Note on Soret and Rilliet’s Observations.—The difficulty in obtaining 
compounds derived from the paraffins of sufficient purity was found by 
Hartley and Huntington to be so great, in the case of hydrocarbons, that 
after repeated trials all hope of obtaining a sufficient number of pure 
homologues was abandoned. Asa rule, substances were examined imme- 
diately after they had been submitted to a final distillation and their 
boiling points taken. In the case of ethyl alcohol, it was difficult to 
obtain it quite free from traces of other substances, which affected its 
spectrum. Samples were dehydrated finally by standing over freshly burnt 
lime for twelve hours, and then submitted to distillation. This process 
was recommended by Dupré. MHaloid derivatives are remarkable for their 
transparency, particularly those of chlorine. 
With the organic acids of the fatty series great difficulty was experi- 
enced, and this is mentioned in the original memoir. In place of the 
acids themselves salts were taken, but such anomalous results were 
obtained with the same substance that a careful examination led to the 
discovery of such substances as formates being oxidised to oxalates by 
the process of evaporation and crystallisation, To obviate the difficulties 
which were thus encountered barium or calcium salts were crystallised, by 
which means any oxalate formed remained insoluble. 
Measured in terms of wave-lengths, the actual difference between 
ethyl alcohol, methylic alcohol, and pure water is very small. In fact, 
methylic alcohol is very slightly different in this respect from water, 
This is shown on Plate 1 in the ‘ Phil. Trans.’ 1879. 
Curves oF MOLECULAR VIBRATIONS. 
Researches on the Relation between the Molecular Structure of Carbon 
Compounds and their Absorption Spectra, Part VII. (Harriey),* 
In this paper the results of the examination of the absorption spectra 
of the following substances are given : 
(1) Aromatic hydrocarbons : benzene and naphthalene. 
(2) Aromatic tertiary bases and their salts: pyridine, dipyridine, 
picoline, quinoline, and their hydrochlorides, 
(3) Addition products of tertiary bases and salts: piperidine, tetra- 
hydro-quinoline, and its hydrochloride. 
? Trans. Chem, Soc. vol, lvii, p, 685, 1885, 
