66 



Prof, J. Dewar on the 



[Mar. 15, 



the blue colouring-matter cyanine, have a more complicated structure, all 

 attempts to reach gradationally simpler substances of known constitution 

 having been unsatisfactory. G-erhardt remarked that " En general les 

 reactions de la quinoleine sont fort peu nettes"*. It is the object of this 

 communication to render the relations of this substance a little more 

 definite. 



In the paper above referred to it was shown that pyridine might be 

 obtained by the distillation of dicarbopyridinic acid with soda-lime ; 

 and as the acid had been made from the oxidation of picoline this may 

 be regarded as a means of passing from the six- to the five-carbon base. 

 It seemed, therefore, important to ascertain if the members of the 

 chinoline series yielded on oxidation any similar acid, and if a simpler 

 base could be obtained by subjecting it to a like treatment. 



In order to carry out this investigation, a quantity of the crude high 

 boiling-point bases occurring in coal-tar was procured, and after repeated 

 fractionation on the large scale was treated in the following manner, to 

 ensure its freedom from pyrrols, phenols, and high boiling-point hydro- 

 carbons. The bases were dissolved in strong nitric acid, separated 

 by means of soda, distilled, and again treated with nitric acid. The 

 addition of small quantities of arsenious acid to the solution was con- 

 tinued for some hours. The bases now liberated from the solution of 

 the nitrate were fractionally distilled, and the specimens boiling between 

 220° C. and 240° C. were regarded as tolerably pure leucoline. It has 

 been satisfactorily proved by G-reville Williams that no amount of frac- 

 tional distillation will yield chemically pure members of this series ; so 

 that this product must be regarded as a mixture of leucoline and irido- 

 line, the names given to the bases in coal-tar isomeric with the chino- 

 line and lepidine obtained from cinchona. In various experiments 

 specimens were used which had been more carefully fractionated, the 

 boiling-point not varying more than 2° C. ; but the products obtained were 

 in all cases similar. 



The mode in which the oxidation was conducted was similar in all 

 respects to that described in my former paper ; but the high boiling-point 

 of the base and the rapidity of the action rendered the use of the con- 

 denser unnecessary. The general method of working was as follows : — 

 129 grams of leucoline were dissolved in an equivalent of sulphuric acid 

 and the solution diluted to 600 cubic centimetres. This was then 

 divided into three equal parts, to each of which was added a hot solution 

 of 100 grams of permanganate of potash in about a litre of water, with 

 constant stirring so as to avoid too vigorous an action. In a few minutes 

 the whole of the permanganate is reduced, and the solution, which ought 

 to be neutral, having been made slightly alkaline, is filtered from the 

 oxide of manganese and evaporated to a small bulk, when a large pro- 

 portion of the sulphate of potash crystallizes on cooling. The mother- 

 liquor was next carefully acidulated with dilute sulphuric acid, when a 

 * 'Traits de Chimie Organique.' par M. C. Gerhardt, torn. iy. 



