562 
THE TROPICAL AGRICULTURIST. [February i, 1890. 
elusions cannot be disputed that tannin is the 
source of strength in tea ; that it is a natural 
constituent of the leaf and cannot be suppressed 
by modes of cultivation or manufacture. It does 
not follow, however, that cultivation and especially 
manufacture may not develope this constituent in 
varying degrees, and so with theine. We should 
like to see analyses of teas from natural soil of an 
Ordinary character and the produce of the same land 
when treated with nitrogenous and phosphatic 
manures. We know that, as the result of liberal 
manuring on an est ta in Chittagong, not only was 
the yield of leaf increased, but appreciably better 
prices obtained for the produce sent to market. The 
remarks with which Mr. Cochran has favoured us will 
receive respeotful attention from our readers. He 
notices that, so far as his experience shows, there 
is small difference in the proportion of tannin in tea 
grown at medium and high elevations in Oeylon. But 
certainly the results of Mr. Hooper's analyses do 
not disprove the popular idea that the proportion 
of tannin in tea is in general ratio dependent on 
altitude,— being more abundant in low-grown tea and 
diminishing with height above sea-level. Eliminating 
the ouriously exceptional case of the Travancore tea, 
which can probably be explained by local circum- 
stances, the order in which the teas are classed 
for average tannin has marked reference to ele- 
vation, thus : — 
Assam (generally under 1,000 feet) .. 18' 6 
Oeylon (generally sea level to 6,000 feet) . . 17-9 
Coorg (about 3,500 to 4,500 feet) . . 16-13 
Nilgiris (5,000 to 7,000 feet) . . 15-3 
Darjeeling (same as Nilgiris) .. 15 - 0 
If Ooorg did not intervene, the comparison would 
be as between 17-9 for Ceylon teas, many of 
which are low grown, to 15 for the Nilgiris and 
Darjeeling teas, which are grown at high altitudes. 
Granted that modes of manufacture may not affect 
the proportion of tannin inherent in tea (10 to 25 
per cent : average about 16),* Mr. Cochran's experi- 
ments show that the next most important consti- 
tuent, theine, is certainly affected. An analysis of 
tea leaves in the green state made by this 
gentleman showed a very considerably larger 
proportion of theine than was yielded by roasted 
tea. We append Mr. Cochran's remarks : — 
" According to your extract from Mr. David Hooper's 
paper the average amount of tannin in Ceylun teas 
is a high one. However, these figures appear to be 
the result of only 13 analyses, and I am inclined to 
think that a more extended series of analyses would 
show a considerably lower average. My own experi- 
ence is still very limited, I can only give data from 
12 samples of Ceylon teas, six of these from an 
estate at medium elevation, and six from an estate 
at a high elevation. At medium elevatiun the lowest 
result was 9 8 per cent of tannin (gallo-tannic acid), 
and the highest 12 , 36, the average 11-36 per cent. 
At the high elevation the lowest was 9'98 per cent, 
and the highest 1123, average 10'41. Judging, how- 
ever, from some experiments I made with green tea 
leaves, 1 do not doubt that there may be lowcountry 
teas in Ceylon which would yield up to 19 or 20 
per cent of tannin. The lowest amount of tannin I 
have found was in a sample of tea from the Nilgiris, 
which only yielded 8'94 per cent. The moisture, 
however, in this sample was high, viz., 8'80. If cal- 
culated to tea with 5 per cent of moisture the amount 
would be raised to 9'30. These figures represent 
what could be obtained from the tea leaves, after 
boiling them with water for one hour. The preoi 
pitant used was not acetate of lead, but acetate of 
copper, which I found by repeated experiments on the 
fame tea to yield very closely concordant results. 
I might mention that in a series of analyses of 
* Mr. John Hughes' analysis ol tea from a Nuwara 
Lliya gtrden ga\e "soluble tuiniu" sj low aa 6*37 
per ccnU— ECi 
Indian green tea leaves undertaken by O. Kellner, to 
ascertain the effect of season on the composition of 
tea, he found that between May 15th a id Nov. 80th 
the average amount of tannin in the particular tea 
experimented with was 106 per cent, calculated into 
the dry matter of the leaves. The lowest amount 
was on the 15th May, viz. 8-53 per cent ; and the 
highest on Nov. 30th, viz. 12-16 per cent. Mr. 
Kellner exhausted the leaves not with water but with 
alcohol. Dr. Battenhall gives, as the result of the 
analyses of Indian teas by American chemists, the 
following figures for tannin : 13 04 to 18 868 per cent, 
average 15'323. The samples analysed represente 
2,414 packages, and the process used for the estima- 
tion was in many, probably in most, ca6es the acetate 
of copper process. I am of opinion that the popular 
view that tannin decreases with elevation is true for 
Ceylon; but it may be found, as the few analyses I have 
made would indicate, that while the difference in 
the tannin between low-grown tea an! tea grown 
at a medium elevation is marked, the difference in 
respect of tannin between medium -and high-grown 
tea is relatively small. I do not think tbnt the differ- 
ent modes of preparation will appreciably affect the 
amount of tannin in teas. It requires a temperature 
of 482 F. to decompose gallo-tannio acid." — M. C. 
TANNIN IN INDIAN AND CEYLON TEAS. 
(From the Chemist and Druggist, Jan. 4th J 
Mr. David Hooper communicates to the Chemical 
News a paper on this subject which is based on ana- 
lyses of 65 specimens, 29 of them from the Nilgiris 
(including 10 green teas), 6 from Travancore, 3 from 
Coorg, 13 from Ceylon, 6 from Assam, and 8 from 
Darjeeling. There has hitherto existed most erroneous 
notions regarding the amounts of tannin in tea. Thus 
Mulder's figures for black tea are 12 88, and green tea 
17'80 per cent.; Dragendorff found in teas of Russian 
commerce 9 - 42 to 12 70 per cent. ; Janlc obtained a 
maximum of 914, and a minimum of 6 92 per cent, 
in eighteen samples ; Wigner, analysing some astrin- 
gent teas, reported as much as 27-7 to 42-3 per cent., 
Hassell gives as the average 15-24 in black, and 18-69 
per cent, in green teas ; Clarkfound from 5 to 19 per 
cent.; and Geisler, an American ohemist, obtained 
14 87 as the average percentage of tannin in a large 
number of packages of Indian teas imported into New 
York. 
Mr. Hooper's analyses show that the percentage in 
black teas ranges from 10 to 21 percent., his lowest, 1014, 
being that of a Nilgiris Congou, aud the highest 2122 
per cent., afforded by a Travancore broken Pekoe. The 
following figures are calculated from Mr. Hooper's 
table, which is too long to reproduce, and show the 
totals of tannin extracted : — 
Samples. Highest. Lowest. Average. 
Travancore (6) 
Assam (6) 
Oeylcn (13) 
Ooorg (3) 
Nilgiris (19) 
Darjeeling (8) 
The specimens of green tea were taken from two 
different estates on the Nilgiris, and showed an aver- 
age of 19 - 1 per cent of tannin, the highest being 
24 37 and the lowest 11*52 per cent., the latter being 
a coarse leaf. The effect of the figures is to show that 
the finest teas are those which contain the most tannin, 
and there is a gradual declension of this principle as 
we approach the Souchongs and Congous. The eleva - 
tion does not appear to affect the amount of tannin, 
as has been supposed. The Dodabetta estate, the highest 
in the list, shows a smaller percentage of tannin in 
the leaf than all the rest, but does not very much 
differ from the samples from Aratapara, some 4,400 
feet below ; and in this respect there is very little to 
learn from the altitudes of the Oeylon teas. The highest 
tannin oontent occurs in Travancore tea of the Sea- 
field estate, a tea of great fragrance, and considered 
to rank very high-class in the borne maiket, and the 
Glenorchy Broken Pekoe of Ceylon, althtujh contain- 
i ing 19 p;r qeut of tannin, ia one cf the bet t-pneed te^a 
21-22 
17-37 
19-63 
20-8 
16-18 
186 
20-87 
15-00 
17-9 
16-93 
1515 
16-13 
18-55 
10-14 
15-3 
17-74 
13-61 
150 
