J. Tea Sci. 54 (1), 26-31; 1985. Printed- in Sri Lanka -
TEA CONSTITUENTS AND EXPORT STANDARDS OF 
SRI LANKA TEAS* 
K. Sivapalan, Jayanthi de Silva 
(Tea Research Institute of Sri Lanka, Talawakele, Sri Lanka) 
. and 
Chrisan'tha Perera 
. (Forbes & Walker Ltd., Colombo) 
Recently UNCTAD/GATT discussed the concept of minimum export standards for 
tradeable teas.' Since earlier work, .on Malawi teas (CTC manufactured, 'plain 
- teas') showed that a good .correlation was obtainable between the theaflavin (TF) fraction 
of. black tea and market valuation, there was a proposal that TF content be included 
as a minimum export standard for all teas. 
The theaflavin fraction of black tea is. believed to make important contributions 
'to the colour and' mouthfeel' properties of its infusion. But TF as a criterion of quality/ 
valuation may not be strictly correct for orthodox-manufactured Sri Lanka teas, because 
in these' superior - teas., TF is only one of the many constituents contributing to valuation. 
An investigation was therefore undertaken to evaluate TF as an index of market price. The 
relationship of other constituents was also examined. 
. In attempting to relate TF content to market price, consideration has to be given to the 
fact that because of market trends,- the value in monetary terms given to TF (or any other 
'. biochemical component) varies from sale to sale. This was overcome by taking a large -
range of teas .(24 samples) of one. grade sold;at the Colombo auctions in one day by one 
broker. The teas were assayed for their th'eaflavin content, total colour and soluble solid 
content. The results were collected over 16 sales, and the data were used to examine 
in Sri Lanka teas, the influence of the different constituents on market valuation. None of 
the tea constituents examined, namely theaflavins, total colour or solids were correlated 
to market price, on any of the 16 sales. 
INTRODUCTION 
Tea is a commodity that is tradeable on international markets, and therefore 
there have been attempts to set up minimum marketing standards for tea. 
The International Standards Organization (ISO) has defined certain physico-
chemical parameters which have to be met for the material to be accepted as tradeable 
tea (see Table 1.) But in practice however, it has been found that even sub-standard 
teas could frequently satisfy the above requirements and qualify for sale on the 
international tea market. 
* Based on an address presented at Annual sessions of Sri Lanka Association for the Advance­
ment of Science, December, 1982. 
( -16 .) 
TABLE 1 — Chemical requirements for black tea — tSO standards 
Characteristics Requirement 
Water extract, % (m/m) minimum 32 
Total Ash, % (m/m) 
maximum 8 
minimum 4 
Water soluble asb, as percentage of total ash 
minimum 45 
Alkalinity of water-soluble ash (as KOH), 
% (m/m)* 
minimum . l.Q 
maximum 3.0 
Acid-insoluble ash, % (m/m) 
maximum 1.0 
Crude fibre, % (m/m) 
maximum 16.5 
°When the alkalinity of water soluble ash is expressed in terms of milli-equivalent per 100 g 
of ground sample, the limits are— 
minimum 17.8 
maximum 53.6 
In an endeavour to correct this situation UNCTAD attempted to formulate 
a new set of parameters internationally acceptable as minimum quality standards 
for.tea. The UNCTAD consultant, prompted perhaps by the results of earlier 
studies in which Malawi teas showed a good correlation between theaflavin (TF) 
content and market valuation, suggested the introduction of minimum levels of TF 
in -the new set of parameters. This suggestion raised a controversy and a number 
of varying opinions were expressed about the proposal (Donahue, 1982; Owuor, 1982; 
Roberts, 1982; Davies, 1983; Mitra, 1983; Theobald, 1983). 
We examined this proposal in our own interests and this paper reports our 
views and the results of our investigations on this subject. 
Quality standards vs. Marketing Standards 
Tea quality is assessed by tea buyers', each of whom will be well aware of his 
own needs. International buyers will tend to make different assessments of quality, 
dependent upon the detailed needs of the market for which each is responsible. For 
e.g. the buyer for the English market would seek teas with a light, brisk liquor with 
bright colour, while the buyer for the Middle East market would bid for teas with a 
strong thick liquor with dark colour. In both instances the buyers seek teas 
desirable for their customers. 
-^-Thus, although a 'quality tea' generally refers to a 'desirable tea', on an 
international platform the term 'quality' signifies different attributes. Hence 
in' the area of international tea marketing, it would be more pertinent to refer to 
marketing standards (or export standards) rather than quality standards. 
( 27 ) 
From this point-of. view, >we would like.to stress the following facts regarding 
(he role of TF as a marketing standard: ' We do not contest the fact that' TF is ah 
important chemical constituent that contributes to the-desirable characteristics of 
a cup of tea. But it is also a fact that there does exist a strong market for certain 
teas, well-made in their own sense, but with relatively low TF contents. This 
underlines the position that in the present discussion, we have to evaluate TF as a 
'marketing standard' and not as a 'quality index*. 
Theobald (1983) has discussed the subject of quality and chemical specifica­
tions very lucidly. He feels that the basic approach to the problem is wrong, and 
questions the wisdom of raising' levels of'chemical specifications to improve quality 
of tradeable teas. 
Cash Valuation vs. Quality Hating 
While assessing the cash valuation of a tea for any particular market, the tea 
taster takes into consideration mainly the liquor characteristics. such as colour, 
strength, flavour brokness and the quality of these. Thus, the term 'quality' is 
rather an ambiguous attribute and conveys different meanings to different persons. 
In the widest sense, quality describes the appeal to the palate as a whole, but in a 
restricted sense, it is described as a particular liquor characteristic recognizable by 
the tea taster. 
A study of these liquor characteristics show that they offer little scope of being 
translated at-their face .value into quantitative definitions. But it may be possible 
to draw some semi-quantitative correlations between individual liquor, characteris­
tics and the chemical constituents of the tea brew. . . 
Strength and colour are characters developed during the fermentation 
process and therefore it is reasonable to ascribe these characters to the enzyme 
oxidation products of the polyphenols, namely Theaflavins and Thearubigins 
(Roberts, 1962). - The 'briskness' of a tea liquor has been described as a liveliness 
on the palate and has been related jointly to theaflavins and caffeine. Flavour is 
obviously related to the volatile constituents. ' Regarding the relationship of quality, 
Roberts (1962) offers the following observations, "When it is considered what an 
important part, is played by the theaflavins in determining the colour, strength 
and briskness of a liquor it is not surprising that teas with a high content of . 
theaflavins are frequently considered to be quality teas. However, there are by 
no means.infrequent-occasions when there is no correlation particularly when 
volatile constituents are of special importance. 
These exceptions emphasize that polyphenols, and their oxidation products 
are not the only factors determining liquors characters ". 
The total evaluation of black tea is thus seen to be determined by a complex • 
combination of chemical parameters. 
Chemical evaluation 
Nevertheless attempts have been made to relate. the valuation of a tea to. 
certain chemical constituents in the tea. Probably in teas where the total character 
of the tea is dependent on just one constituent, a simple correlation may, be found 
between such a constituent and the valuation of the tea: In fact such a direct corre-. 
lation has been reported between theaflavin content and valuation for Malawi teas —.: 
which are known to be '.plain' teas (Hilton and Ellis, 1982), , 
(i 2X„ ); 
But if the favourable attributes of the tea are due to a composite of chemical 
constituents, the above contention would most-probably run into difficulties/ This 
is true of any food item where the sensations of taste and smell are involved; In 
spite of the advances in analytical technology and the Ultra-sophisticated instrumen­
tation available throughout the world* certain food items are still evaluated by the 
human senses — may be following the adage," the proof of the cake is in the eating". 
It was our intention therefore to examine the correlations between tea cons­
tituents and market evaluations of Sri Lanka teas which are generally considered 
vantage teas. 
Market evaluation 
Ultimately the best assessment of a tea's commercial value is the price offered 
on a freely competitive auction. But this, too is subject to certain, considerations. 
i.e. Market valuation = Cash valuation + Market factors. 
Therefore in attempting to relate a chemical, constituent or liquor characteristic 
to sale price, consideration has to be given to the fact that because of market trends' 
e.g. demand for a particular grade or. variety of tea at that auction.or net, weight of 
a particular tea being offered etc, the value in monetary' terms associated with a 
particular liquor characteristic o.r chemical characteristic, may vary, from sale to 
sale. This difficulty may be overcome by adopting the workplan detailed below. 
Workplan 
We examined a large range of teas (one batch) sold at. the Colombo-auctions 
on one day. Each batch consists of 24 invoices of tea of 
(i) One type of manufacture — orthodox"' : 
(ii) One variety — high grown, medium grown, etc. 
(iii) One grade — BOP, BOPF etc, 
(iv) Sold on one day: 
(v) Sold by one broker •• . * 
(vi) To cover the highest to lowest price range'fetched for that sale. 
For each batch, correlations are worked out between market price and chemical 
constituent. Sixteen such batches i<e. sold on 16 different Occasions are examined. 
Justification . . . 
Other factors being constant, the sale price of a tea (of one variety, one grade, 
sold on one day etc.) would be more or less based on the 'quality rating'- of the 
tea. Further if we exclude seasonal/flavoury/exotic teas from this experiment, this 
sale price would generally work out to be an average valuation accepted by a number 
of tasters each of whom would have evaluated this tea. independently. Therefore 
it would be reasonable to assume that the market price or sale price would be an 
index of the relative quality rating of the tea offered for sale oh that day. 
( 29 ) 
EXPERIMENTAL 
Teas were collected over 16 sales as detailed above and they were assayed for 
theaflavin content, total colour (Hilton and Ellis, 1982) and. soluble solid content. 
Linear regression correlations were worked out between each of the above 
chemical constituents and the market price for each sale. Thus 16 correlation 
coefficients (r) were obtained for each constituent. 
RESULTS AND DISCUSSION 
The results of the linear regression analyses presented in Table 2 show that 
none of the tea constituents namely, theaflavins, total colour or soluble solids were 
correlated to the market price, on any of the 16 sales. 
This result is perhaps understandable, because in Sri Lankan black tea, the 
overall evaluation is a composite of a number of chemical parameters present in 
various proportions, and in particular the volatiles. Further it is the delicate 
balance of these factors that gives Sri Lankan tea the final * bouquet'. This appar­
ently is why attempts to quantify ' quality' has not been so far successful. 
TABLE 2 — Linear regression analysis •— sale price vs constituents 
r r r 
SP vs TF SP vs SS SP vs TC 
1. High Grown BOP 0.30 0.50 0.04 
2. High Grown BOPF 0.31 0.44 0.05 
3. High Grown Dust No . 1 0.60 0.70 0.16 
4. High Grown BP - 0 . 0 4 0.22 0.27 
5. Medium Grown BOP 0.45 0.42 0.22 
6. Medium Grown BOPF 0.26 0.15 0.06 
7. Medium Grown Dust No. 1 —0.45 0.65 - 0 . 2 3 
8. Medium Grown BP —0.13 0.01 - 0 . 4 0 
9. High Grown BOP - 0 . 0 6 0.54 —0.22 
10. High Grown BOPF 0.08 0.33 —0.35 
11. High Grown Dust No. 1 0.38 0.05 0.40 
12. High Grown BP 0.03 0.51 - 0 . 0 3 
13. High Grown BOP. -rO-27 0.69 -O.03 
14. High .Grown BOPF . -O.04 0.67 —0.15 
15. High Grown Dust No. 1 0.03 0.19 - 0 . 0 9 
16. High Grown BP 0.39 0.16 —0.47 
Note : 
SP — Sale Price. 
TF — Theaflavin Content. 
SS — Soluble Solid Content. 
TC — Total Colour. 
( 30 ) 
Sri Lankan teas are mostly orthodox-manufactured, and would be associated 
with a relatively higher content of volatiles than CTC manufactured teas. Trace 
amounts of flavour could confer special characteristics to the tea and bring about 
a disproportionate increase in its valuation. This is specially true of our high-
grown teas. In this context we are reminded of the Darjeeling teas which are 
valued highly despite the relatively lower TF .they,contain. 
The above discussion supports the contention of Roberts (1982) that it is not 
possible to fix minimum values for TE as an export standard for Sri Lankan teas. 
REFERENCES .* . 
DA VIES, A. G. (1983). Theaflavins — objective indicators of quality. Tea & Coffee Trade Journal 
Feb. 1983, 34. 
DONAHUE, J. (1982). The TF factor. World Coffee and Tea Dec. 1982, 6. 
HILTON, P. J. and ELLIS, R. T. (1982). Estimation of the market value of Central African tea 
by theaflavin analysis. J. Sci. Food Agric. 25,227-232. 
MITRA, K . K. (1983). Quality standards for tea. International Tea Journal (S) Aug. 1983,6-8. 
OWUOR, P. O . (1982). Can Theaflavins content alone be adequate parameter in black tea quality 
estimation? A review. Tea 3 (N6.2) Dec 1982,36-40. • • 
ROBERTS E. A. H. (1962). Economic importance of flavonoid substances. Tea fermentation, 
pp. 468-512. In The chemistry of flavonoid compounds, ed. T. A. Geissman, Pergamon 
Press, London. 
ROBERTS, G. R. (1982). World trade in tea and minimum standards. International Tea Journal 
No. 4, October 1982, 13-16. 
THEOBALD, D. J. (1983). How do we define quality ? Tea and Coffee Journal Oct. 1983, p. 36. 
(• 31. )