176 COFFEE. 



manipulations of roasting are at present such that this very coi 

 ing and reshovelliug of hot beans must be done in the open a 

 and is the reason that it is not possible to catch and conder 

 the vapors so abundantly liberated. Practice has proven that 

 the very moment when the beans turn brovm and the first vapc 

 begin to be given off it is absolutely necessary to pull the dm 

 out of the roasting furnace and rapidly cool the coffee by shovi 

 ling and reshovelling in the air, or there is danger of its taking fi 

 in the furnace and burning to a coal. Nevertheless it might 

 feasible to connect the drums with an exhauster so as to conden 

 the gases in a receiver, and at the same time cool the bean enouj 

 to prevent its taking fire. Cech has no doubt that the oil obtain 

 in this manner would find use, at a profit, in making liqueurs. ^ 

 study the properties of oil of coffee, Dr. Cech pounded up fif 

 pounds of different kinds of coffee in a mortar, and then extract 

 it with alcohol and ether, obtaining about 1,200 grammes (tv 

 and a half pounds) of oil of coffee. The beans extracted by hi 

 were not of equal value as regards the yield of oil, for while son 

 contained as high as thirteen per cent., other kinds fell belo 

 eight per cent. The oil of coffee is a green, thick, transpare: 

 oil, and after some time a few long needles were deposited fro 

 it. These proved to be caffeine. Since caffeine is not extract( 

 from the exhausted beans by ether, and very little of it is taki 

 up by the alcohol employed, the coffee from which the oil h 

 been extracted could be employed for the manufacture of caffeir 

 The coffee oil became turbid in half a year, although it was ke 

 in hermetically closed bottles. Small groups of crystals we 

 formed in the middle of the liquid, and slowly settled to the b( 

 tom, and at the end of three years the bottle was two-thirds fi 

 of a dirty mass of crystals, consisting of the solid fatty acids, b 

 the upper layer of the liquid remained for years transparent, cles 

 and of a beautiful green color, proving that a portion of coffee i 

 consists of liquid oleic acid." 



Eecently there appeared in the Phyddcm and Surgeon, a j 

 per from the pen of Prof. Albert B. Prescott, of the Universi 

 of Michigan, from which it appears that the tannin in the coff( 

 berry is not over about one-third the quantity of that in t< 

 leaves ; it may be considerably less. Six samples of coffee we 



