194 ALKALOIDS. 



rapid decomposition when precipitated in combination with 

 chloride of gold and platinum (e.g. curarine). 



184. Quantitative Separation of Several Alkaloids from one 

 another. Such cases occur in the examination of cinchona barks. 

 Although several of the many alkaloids in these barks are present 

 in such minute quantities that they may generally be neglected, 

 there are at least five the detection and estimation of which are 

 of importance in valuing samples. These are quinine, cinchonidine, 

 quinidine, cinchonine, and the so-called amorphous alkaloid. The 

 mixed alkaloids are extracted and estimated as directed in 67. 

 For their quantitative separation from one another, I propose 

 using Moens' method, which has been recognised by Hielbig, after 

 numerous experiments, as suitable for the purpose. 1 



I. The mixed alkaloids just referred to are dissolved in acetic 

 acid, 2 without the application of heat, and the solution evaporated to 

 dryness, care being taken that the residue does not turn brown. 

 This is then dissolved in the smallest possible quantity of water 

 and filtered. From the solution, which should not be evaporated, 

 quinine and cinchonidine are precipitated together by (about -5 

 of a gram of) tartrate of ammonia and soda, which is preferable 

 to the Rochelle salt usually used. After standing twenty-four 

 hours the precipitate is filtered off, washed, dried at 110, and 

 weighed. 1-6 gram of mixed alkaloids would yield about 30 cc. 

 of filtrate, and require about the same quantity of wash-water. 

 A correction must be made of '000746 gram of quinine and 

 '000441 gram of cinchonidine for each cc. of filtrate and wash- 

 ings, provided that both alkaloids are present together. If the 

 bark contains quinine alone, 0*00102 gram must be added for each 

 cc. ; or if cinchonidine alone, '000543 gram. The apparent dis- 

 crepancy in these figures is caused by the influence exercised by 

 the presence of either tartrate on the solubility of the other. 100 

 parts of precipitate indicate 79*41 anhydrous quinine or 76 '8 

 cinchonidine. 



II. To separate quinine from cinchonidine the mixed tartrates 

 are dissolved in 90 to 92 per cent, spirit containing 1 *6 per cent. 



1 Loc. cit. 



2 Hielbig has also experimented with hydrochloric and tartaric acids, but 

 obtained the best results with acetic. The chlorides formed by the hydro- 

 chloric acid appear specially liable to cause errors when subsequently precipi- 

 tating with tartrate. Whichever acid, however, be chosen, the excess must 

 in some way be removed. 



