NOTE UPON COFFEE. 
231 
a notable quantity of cafFein. During the first days of mace- 
ration, the odour of horse-radish is perceptible, but little by 
little this odour becomes more disagreeable and finally stercora- 
ceously fetid. 
The only striking facts resulting from this resumption are, 
that fine green Martinique coffee, the only kind upon which 
we experimented, contains about one-eighth of its weight of 
fixed oil, and that after torrefaction this oil is charged with an 
aromatic principle, and even with the taste of burnt coffee; 
that, moreover, caffein is found in all the products of coffee, 
whatever mode of treatment it is made to undergo. This, 
then, is a principle that must be regarded as little alterable, — 
since it resists the destructive action of fermentation and the 
elevated temperature of torrefaction. We have suspended it 
in water, and it has thus remained for many months exposed 
to the atmosphere, without appearing to have undergone the 
least alteration. 
We would be induced, in the meantime, to attribute to this 
product the essential properties of coffee, for it alone, of all, 
presents some characters a little striking; and the circumstance 
is remarkable, that it not only contains azote like the greater 
number of active vegetable principles, but like them, it forms 
an insoluble compound with tannin. We have even taken ad- 
vantage of this property to extract caffein. as has been done by 
M. Henry Jr., with respect to some alkaloids. Thus we took 
500 grammes of coffee slightly torrefied, only sufficiently so to 
render it readily powdered. We next boiled this pow^der 
twice in two litres of water at a time, and during twenty 
minutes. These decoctions, after being strained, reddened 
litmus; they were saturated by some drops of the solution of 
caustic soda, and a strong decoction of galls was poured into 
them. An abundant precipitate resulted, which sometimes 
was curdled and easily subsided to the bottom of the vessel, 
and sometimes remained suspended in the liquid. In this 
last case, it is necessary to take care to consult the state of 
saturation of the liquid; and to add, as required, either a little 
caustic soda, or diluted sulphuric acid, so as to produce 
