147 
3C 6 H 6 0 + 2C0 = C, 0 H 14 O 3 + 2H 2 0. 
We further showed that aurin, when treated in an alkaline 
or acid solution with zinc-dust, combines, like most other 
organic colouring matters, with hydroden, yielding leucaurin 
C20H1A, which readily crystallises in large colourless prisms, 
whilst on heating aurin with ammonia under pressure a new 
red colouring matter is formed, which is called red aurin or 
peonin. 
The further investigation of this body showed that it 
contained nitrogen, which on heating with soda-lime was 
evolved in some cases completely as ammonia, while in others, 
where the action of the ammonia went on longer it was 
given off, at least partly, in the form of aniline. We then 
found, that on continuing the heating, the red colour of the 
liquid became paler, and at last a yellowish solution was 
obtained, from which water precipitated a white crystalline 
body, possessing all the characteristic properties of rosani- 
line.* The formation of this compound might be explained 
by the following equation : 
C 20 H 14 O 3 + 3NH 3 = C 20 H 17 N 3 4 - 3H 2 0. 
But Hofmann, who has most carefully examined rosaniline 
as well as a great number of its derivatives, found that the 
formula of this base is C 20 H 19 O 3 , and this result has been 
confirmed by Caro and Grabe,f who showed that rosolic 
acid, which is obtained from rosaniline by a reaction, which 
shortly may be stated to be the inverse of that above given, 
has the composition C 20 H 16 O3, while its properties are in 
almost all respects so similar to aurin, that some chemists, 
including ourselves, have been inclined to regard them as 
identical. 
In order to clear up these points it was necessary to pre- 
pare a considerable quantity of pure aurin. We have 
already shown in our memoir read before the Chemical 
Society, J that a pure compound is readily obtained by 
* Jour el. Ohem. Soc. (2) XI, 434. f Liebig’s Ann. CLXXIX, 148. 
J Journ. CTiem. Soc., 1877, II., 121. 
