230 Selected Articles. 
occurs when pure tannin is placed in the same circumstances. 
This experiment was made with the bitannates of quinine and 
cinchonine. 
By means of the metallic oxides, as those of the peroxide 
of iron, tin, lead, antimony, barium, calcium, and magnesium, 
these bitannites are decomposed with more or less ease. The 
alkaloid is set at liberty, and it can be taken up by alcohol, 
which does not act on the metallic tannate. 
Potash, soda and ammonia decompose them in the same 
way, but the organic bitannate is always soluble in an ex- 
cess of these alkalies. I obtained by means of heat, a needle 
like precipitate from the bitannate of morphine, by cautiously 
adding ammonia; this on examination proved to be pure 
morphine. 
The mode of preparation of these compounds, and their 
decided action on litmus, show that they are acid salts — true 
bitannates formed by the union of one atom of the base to two 
atoms of the tannic acid. 
The quantity of pure tannin which unites with a known 
weight of the organic alkali, is exactly that pointed out by 
calculation based on their theoretic composition ; thus I ob- 
tained as follows, with one gramme each of: 
Quinine, 
Cinchonine, 
Strychnine, 
Brucine, 
Afterwards taking those organic bases, whose elementary 
analysis is best settled, as quinine, cinchonine, and morphine, 
and subjecting them in the form of perfectly pure and dry 
tannates, to decomposition with deutoxide of copper, &c, I 
found the same relations between the carbon and nitrogen, 
thus obtained. 
Bitannate of Quinine, ^ Pure and dry, af- \ 23,6 carbon. 
Cinchonine, > forded to one of > 23,7 
Morphine, ) nitrogen. ; 30,1. 
