It is rather a singular coincidence that the same alkaloid should 
prevail in all the principal substances employed in a similar manner 
as beverages in different parts of the world— in the tea of China and 
India, the coffee of Arabia, the Cacao of Central America, the Mate 
of South America, and the guarana of Brazil. M. Fournier 20 has 
found in the last-named substance, besides tannate of caffein, the 
following principles: — Gum, starch, an acrid green fixed oil, a 
concrete volatile oil scarcely soluble in water, a peculiar principle not 
precisely determined, and tannic acid.’ 
Tannic Acid was found in guarana by F. V. Greene, 21 and he 
describes the reactions of this as ‘ strikingly dissimilar from those of 
tannic acids generally.’ Thoms, 22 in 1894, found in guarana 8-63 per 
cent, of moisture. Later work by E. Kirmsse 23 showed the presence 
of tannin. From the crude tannin this worker obtained crystals of a 
catechin the properties and reactions of which were found to be 
identical with that of Pegu catechu. The author confirms Thoms’s 
statement that the amount of caffeine in guarana has been over 
estimated; three samples were found to contain 2 68, 297 and 3 10 
per cent, respectively. Cocoa, which has been stated to be an 
adulterant of lower grades of guarana, was not detected in any of the 
samples. 
Oesterlen 24 notes also the presence of a resin, and says that the 
tannin of guarana appears to act like that of Monesia and Rhatany. 
A large number of authorities 25 have reported the presence of 
Caffeine, which is the most important constituent of guarana, an 
identical with guaranine and theine. In 1853 Oesteilen- 6 refers to a 
special indifferent crystalline substance guaranin. Hanbury 27 spea 'S 
of caffeine as a constituent of guarana. d here is some variation 
the statements of authorities as to the percentage of caffeine presen , 
thus we have the following data : — 
In 1886, 4-5 per cent. Bochefontaine and Gusset. 28 
372 to 5 or 6 per cent. Kremel. Feemster, Squibb 
and Fluckiger. 29 
1888, 312 to 3-80 per cent. A. Kremel. 30 
1894, 2*6 per cent. Thoms. 31 
1897, 4 32 to 4*68 per cent. C. H. La \\ all. 32 
2 5 to 5 per cent. Humphrey- and British Pharma- 
ceutical Codex. 33 
