13 



dilute hydrochloric acid, and washing the precipitate by 

 Bunsen's filter pump, but also this product contains hydro- 

 chloric acid, which was only given off above 110^ 



By spontaneous evaporation of an alcoholic solution, 

 aurine is obtained in dull red crystals, with a green metallic 

 lustre. Dried at 110' this body contains no alcohol, but 

 still retains a large quantity of water, which only escapes 

 between 140° — 180°, the crystals not changing their appear- 

 ance at all, and they may be heated up to 200^ without any 

 further alteration, which fact does not agree with Fresenius' 

 observation, that aurine crystallised from alcohol or acetic 

 acid melts at 156°. The analysis of this body, dried at 200°, 

 which we believe to be pure aurine, gave numbers closely 

 agreeing with the formula C20H14O3 and the mode of its 

 formation may, if this formula is correct, be expressed by 



the equation. 



SCeHeO + CA = C20H, A + 2H2O. 



The substance dried at 110° lost at 180° 5*4% of water 

 corresponding to the formula C20H14O34-H2O.* 



Caro and Wanklyn obtained by the action of nitrous acid 

 on rosaniline a body, which they believe to be identical 

 with amine, and to which they assign, from the mode of 

 formation, the composition C2oHi603,t differing from our 

 formula only by two atoms of hydi'ogen. 



Nascent hydrogen converts aurine into colourless leucG- 

 aurine C20H1SO3. This reduction may be effected by heating- 

 it in an alkaline solution with zinc dust, but at the same 

 time a dark resinous body is formed, from which the leuco- 

 aurine cannot be easily freed. Better results are obtained 

 by acting with zinc dust on a solution of aurine in strong 

 acetic acid. Leuco-aurine crystallises from acetic acid or 

 alcohol in compact colourless prisms. 



A body resembling leuco-aurine is contained in crude 

 aurine ; we have not as yet obtained it in a pure state. It 



^Fresenius analysed aurine wliich was crystallized from alcohol and dried at 

 100°. His numbers agree exactly with the formula C^. ^11 ^^0.^-^-2^11^0. 

 f Proceed. Koy. Soc. xy., 210. 



