ON COFFEE. 117 



which is of so mu ch importance to be known, I divide the Three stag-* of 

 process of roasting into three distinct periods: 1, that in pr 

 which the berry loses its natural colour, and assumes that 

 of bread raspings, or dried almonds : 2, that in which it 

 acquires the brown red of a dry cltesnut: 3, that in which, 

 become almost black, it is still not charred. 



I took six ounces of Mart'mico coffee, divided them into Experiment, 

 three parts, and roasted them separately in these three de- 

 grees. 



The two ounces slisjhtlv roasted, and of .the colour of T.ost of weight 

 dried almonds, lost on the fire two drachms. These I shall '•' 

 call No. 1. 



The two ounces roasted to a chesnut colour lost three T J 7J 

 drachms. These I call No. 2. 



The two ounces roasted to blackness lost three drachms r near J. 

 forty-eight grains. These I shall distinguish as No. 3. 



No. 1. passed through the mill with difficulty. Infused Slightly roast 

 cold, the infusion contained tannin, and precipitated solu- e 

 tion of gelatine; it was very aromatic, * and had the fla- 

 vour of almonds; there was not the least bitterness, but a 

 -sufficiently decided harshness. Infused hot its aromatic fla- 

 vour was the same; and its taste reminded me of that, of 

 the almond cake called nougat. It was not at all bitter, 

 and the harshness was less perceptible. 



No. 2. was more easy to grind. To cold water it gave Higher roasted. 

 out less tannin; its aromatic flavour was weaker, and ithad 



* The object of retaining the aroma, which is dissipated by a Two methods 

 strong heat, has given rise to two processes, which are not alto- USC( * l <> retain 

 gether ineffectual. The first, adopted in India. and by some per- the aroma : 

 •sons in France, consists in putting into the cylindrical roaster a lit- butter in the 

 tie fresh butter, when the coffee begins to be coloured. No more roasting j 

 must be used than will slightly varnish the surface of the berries. — 

 The butter retains a part of the essential oil, that would have eva- 

 porated. It is not a bad method, but sometimes it imparts to the 

 coffee a peculiar flavour, which everybody does not like. 



The second consists in spreading the roasted coffee, while yet hot powdering with 

 and sweating, on writing paper, and powdering it lightly with su- su 8 ar a " tr **• 

 gar. The sugar absorbs the oil of the coffee, and retains its aroma ; 

 but it does not appear to me to increase the pleasantness of the cof- 

 fee, and renders us uncertain how much sugar to put into a cup. 



mote 



