Home
  By Author [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Title [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Language
all Classics books content using ISYS

Download this book: [ ASCII | HTML | PDF ]

Look for this book on Amazon


We have new books nearly every day.
If you would like a news letter once a week or once a month
fill out this form and we will give you a summary of the books for that week or month by email.

Title: The Cultivation and Manufacture of Tea
Author: Money, Edward
Language: English
As this book started as an ASCII text book there are no pictures available.
Copyright Status: Not copyrighted in the United States. If you live elsewhere check the laws of your country before downloading this ebook. See comments about copyright issues at end of book.

*** Start of this Doctrine Publishing Corporation Digital Book "The Cultivation and Manufacture of Tea" ***

This book is indexed by ISYS Web Indexing system to allow the reader find any word or number within the document.



produced from images generously made available by The
Internet Archive/American Libraries.)



Transcriber's Note

Italic text is indicated by _underscores_, and bold text by
~swung dashes~.



  THE
  CULTIVATION AND MANUFACTURE
  OF
  TEA.



  THE
  CULTIVATION & MANUFACTURE
  OF
  TEA,


  BY
  LIEUT.-COL. EDWARD MONEY,


  THE ORIGINAL MANUSCRIPT OF WHICH SECURED THE PRIZE OF THE
  GRANT GOLD MEDAL AND Rs. 300, AWARDED BY THE AGRICULTURAL
  AND HORTICULTURAL SOCIETY OF INDIA
  IN THE YEAR 1871.


  _FOURTH EDITION_,
  REVISED AND SUPPLEMENTED BY ADDITIONAL CHAPTERS.


  LONDON:
  W. B. WHITTINGHAM & CO., 91, GRACECHURCH STREET.
  CALCUTTA: THACKER & CO.
  1883.



PREFACE

TO

THE FOURTH EDITION.


Six new Chapters are added. So much has been done in Tea since I last
wrote, I found it impossible to embody all in the former book, and so
preferred to give it separately. The new Chapters treat of--

  COUNTRIES OUTSIDE CHINA AND INDIA THAT PRODUCE TEA.

  TEA STATISTICS.

  MARKETS FOR TEA OUTSIDE GREAT BRITAIN.

  MAKING INDIAN TEA KNOWN IN THE UNITED KINGDOM.

  TEA MACHINERY.

  WEIGHING AND BULKING OF INDIAN TEAS AT CUSTOM HOUSE.

A separate and full Index of the subjects treated of in the additions
to this Fourth Edition will be found at the end of the Book.

          EDWARD MONEY.

  EAST INDIA CLUB, ST. JAMES’S SQUARE,
  _July, 1883_.



PREFACE

TO

THE THIRD EDITION.


The experience of four more years, which includes six months’ residence
in the Neilgherries, is embodied in the following, while the whole of
the letter-press of the Second Edition has been corrected and revised.

          EDWARD MONEY.

  LONDON,
  _April, 1878_.



PREFACE

TO

THE SECOND EDITION.


Three years’ further experience, and visiting two Tea districts I had
not seen before, have enabled me to amend whatever was faulty in the
First Edition. The whole has been revised, and much new matter is added
throughout. A new Chapter at the end on the Past, the Present, and the
Future of Indian Tea will, it is hoped, be found interesting. An Index
(a great want to the First Edition) is added, so that all information
on any point can be at once found. The manufacture of Green Tea, of
which I was ignorant when I last wrote, is given, and the advisability
of that manufacture is discussed.

In its present form I hope and believe this little work will be found
useful and interesting to all connected with Tea.

          EDWARD MONEY.

  DARJEELING,
  _May, 1874_.



PREFACE

TO

THE FIRST EDITION.


The following Essay was written with, _firstly_, the object of
competing for the Gold Medal and the Money Prize offered by the
Agricultural and Horticultural Society of India for the best treatise
on the cultivation and manufacture of Tea; and, _secondly_, with the
view of arranging the hundreds of notes on these subjects, which, in
the course of eleven years, I had collected.

During all these years I have been a Tea planter, making first for
myself and others a garden in the Himalayas, and for the last six years
doing the same thing for myself in the Chittagong district.

Whenever I have visited other plantations (and I have seen a great
number in many districts), I have brought away notes of all I saw. Up
to the last, at every such visit, I have learnt _something_--if rarely
nothing to follow, something at least to avoid. I have now tested all
and everything connected with the cultivation and manufacture of Tea
by my own experience, and I can only hope that what I have written
will be found useful to an industry destined yet, I believe, in spite
of the late panic--the natural result of wild speculation--to play an
important part in India.

I have endeavoured to adapt this Essay to the wants of a beginner, as
there are many of that class now, and may yet be more in days to come,
who must feel, as I often have, the want of a really practical work on
Tea.

To those who have Tea properties in unlikely climates and unlikely
sites, I would say two words. No view I have taken of the advantages of
different localities _can_ in any way affect the results of enterprises
already entered upon. But if the note of warning, sounded in the
following pages, checks further losses in Tea, already so vast, while
it fosters the cultivation on remunerative sites, I shall not have
written in vain.

          EDWARD MONEY.

  SUNGOO RIVER PLANTATION,
  CHITTAGONG,
  _November, 1870_.



CONTENTS

  CHAPTER                                                           PAGE

       I. PAST AND PRESENT FINANCIAL PROSPECT OF TEA                   1
      II. LABOUR, LOCAL AND IMPORTED                                  10
     III. TEA DISTRICTS AND THEIR COMPARATIVE ADVANTAGES.
                CLIMATE, SOIL, &C., IN EACH                           13
      IV. SOIL                                                        31
       V. NATURE OF JUNGLE                                            34
      VI. WATER AND SANITATION                                        35
     VII. LAY OF LAND                                                 37
    VIII. LAYING OUT A GARDEN                                         42
      IX. VARIETIES OF THE TEA PLANT                                  47
       X. TEA SEED                                                    54
      XI. COMPARISON BETWEEN SOWING IN NURSERIES AND IN SITU          57
     XII. SOWING SEED IN SITU, ID EST, AT STAKE                       59
    XIII. NURSERIES                                                   62
     XIV. MANURE                                                      67
      XV. DISTANCES APART TO PLANT TEA-BUSHES                         71
     XVI. MAKING A GARDEN                                             73
    XVII. TRANSPLANTING                                               76
   XVIII. CULTIVATION OF MADE GARDENS                                 81
     XIX. PRUNING                                                     86
      XX. WHITE ANTS, CRICKETS, AND BLIGHT                            89
     XXI. FILLING UP VACANCIES                                        92
    XXII. FLUSHING AND NUMBER OF FLUSHES                              97
   XXIII. LEAF-PICKING                                               102
    XXIV. MANUFACTURE. MECHANICAL CONTRIVANCES                       109
     XXV. SIFTING AND SORTING                                        134
    XXVI. BOXES. PACKING                                             147
   XXVII. MANAGEMENT, ACCOUNTS, FORMS                                152
  XXVIII. COST OF MANUFACTURE, PACKING, TRANSPORT, &C.               160
    XXIX. COST OF MAKING A 300-ACRE TEA GARDEN                       163
     XXX. HOW MUCH PROFIT TEA CAN GIVE                               168
    XXXI. THE PAST, PRESENT, AND FUTURE OF INDIAN TEA                174
   XXXII. COUNTRIES OUTSIDE CHINA AND INDIA THAT PRODUCE TEA         183
  XXXIII. STATISTICS REGARDING INDIAN TEA                            194
   XXXIV. MARKETS OUTSIDE GREAT BRITAIN                              207
    XXXV. MAKING INDIAN TEA KNOWN IN THE UNITED KINGDOM              218
   XXXVI. TEA MACHINERY                                              222
  XXXVII. WEIGHING AND BULKING OF INDIAN TEAS AT CUSTOM HOUSE        272

  ADDENDA TO THIRD EDITION                                           293

  INDEX                                                              299



PRIZE ESSAY

ON THE

CULTIVATION AND MANUFACTURE

OF TEA IN INDIA.

_PREMIUM, THREE HUNDRED RUPEES AND THE GRANT GOLD MEDAL._



CHAPTER I.

PAST AND PRESENT FINANCIAL PROSPECTS OF TEA.


  _Will Tea pay?_ Certainly, on a suitable site, and in a good Tea
  climate; equally certainly _not_ in a bad locality with other
  drawbacks.

  _Why, then, has Tea only paid during the last few years_ (?) Simply
  because nothing will pay, which is embarked on without the requisite
  knowledge; and this was pre-eminently the case with Tea.

Nothing was known of Tea formerly, when everybody rushed into it; not
much is known even now. Still, with those drawbacks and many others,
the enterprise has survived, and it is very certain the day will never
come that Tea cultivation will cease in India.

I believe there is nothing will pay better than Tea, if embarked on
with the necessary knowledge in suitable places, but failing either of
these success must not be hoped for.

It was madness to expect aught but ruin, under the conditions which the
cultivation was entered on in the Tea-fever days. People who had failed
in everything else were thought quite competent to make plantations.
’Tis true Tea was so entirely a new thing at that time, but few could
be found who had any knowledge of it. Still, had managers with some
practice in agriculture been chosen, the end would not have been so
disastrous. But any one--literally any one--was taken, and tea planters
in those days were a strange medley of retired or cashiered army and
navy officers, medical men, engineers, veterinary surgeons, steamer
captains, chemists, shop-keepers of all kinds, stable-keepers, used-up
policemen, clerks, and goodness knows who besides!

Is it strange the enterprise failed in their hands? Would it not have
been much stranger if it had not?

This was only one of the many necessities for failure. I call them
“necessities” as they appear to have been so industriously sought after
in some cases. I must detail them shortly, for to expatiate on them
would fill a book.

No garden should exceed 500 acres under Tea. If highly cultivated one
of even half that size will pay enormously, far better than a larger
area with low cultivation. Add, say, 400 acres for charcoal, &c.,
making 900 or say 1,000 acres the outside area that can be required,
and the outside that should ever have been purchased for any one
estate. Instead of this, individuals and Companies rushing into Tea
bought tracts of five, ten, fifteen, and twenty thousand acres. The
idea was that, though it might not be all cultivated, by taking up
so large an area all the local labour where there was any would be
secured. Often, however, these large tracts were purchased where local
labour there was none, and what the object there was is a mystery.
I conceive, however, there was a hazy idea that _if_ 500 acres paid
well, 1,000 would pay double, and that eventually even two or three
thousand acres would be put under Tea and make the fortunate possessor
a _millionaire_. In short, there were no bounds, in fancy, to the size
a garden might be made, and thus loss No. 2 took place when absurdly
large areas were bought of the Government and large areas cultivated.

The only fair rules for the sale of waste lands were those of Lord
Canning, which the Secretary of State at home, who could know nothing
of the subject, chose to modify and upset. Instead of Rs. 2-8 per acre
for all waste lands (by no means a low price, when the cost of land
in the Colonies is considered) and that the applicant for the land
(who had, perhaps, spent months seeking for it) should have it, the
illiberal and unjust method of putting the land up to auction with an
upset price of Rs. 2-8 was adopted, the unfortunate seeker, finder,
and applicant, through whose labour the land had been found, having no
advantage over any other bidder. The best, at least the most successful
plan in those days, though as unfair and illiberal as the Government
action, was to wait till some one, who was supposed to know what good
Tea land was, applied for a piece, and then bid half an anna more than
he did, and thus secure it. It _paid_ much better than hunting about
for oneself, and it was kind and considerate on the part of Government
to devise such a plan!

In those fever days, with the auction system, lands almost always sold
far above their value. The most absurd prices, Rs. 10 and upwards
per acre, were sometimes paid for wild jungle lands. Tracts, which
natives could have, and in some cases _did_ lease from Government for
inconceivably small sums, representing, say, at thirty years’ purchase,
4 annas per acre, were put up for auction with a limit of Rs. 2-8, and
sold perhaps at Rs. 8 or 10 per acre. Had the Government given land
_gratis_ to Tea cultivators the policy would have been a wise one. To
do what they did was scarcely acting up to their professed wish “to
develope the resources of the country.”

Since the above was written, new rules have been published for the sale
of waste lands. The objectionable auction system is continued, and the
upset price is much enhanced, as follows:--


_Schedule of Rates of Upset Prices._

                                           Upset price per acre.
  Districts of the Assam Division                Rs. 8
  Districts of Cachar and Sylhet                     8
  Districts of the Chittagong Division               6
  Districts of the Chota Nagpore Division            5
  The Soonderbuns                                    5
  All other Districts                               10

It is not likely that Government will sell much land at such exorbitant
rates.[1]

Security of title, it is generally thought, is one of the advantages
of buying land from the State; but I grieve to state my experience is
that the reverse is the case, and will so remain until the following is
done:--

_First._ The Government should learn _what is and what is not theirs to
sell_. Such an absurdity, then, as Government ascertaining, years after
the auction, that they had sold lands they had no right to sell, could
not be.

_Secondly._ That before land is sold it be properly surveyed and
demarcated; and what might so easily have been done, and which alone
would have compensated for much of bad procedure in other respects,
that the simple and obvious plan before the sale, of sending a European
official to show the neighbouring villagers and intending purchasers
the boundaries of the land to be sold, be resorted to.

This last simple expedient would have saved some grantees years of
litigation, and many a hard thought of the said grantees against the
Government. It would naturally occur to any one at all conversant with
the subject; but, _alas!_ in India this is often not the condition
under which laws are made.

But there is another difficulty at the back of all this.

Though the Waste Land Rules enact that the Government, and not the
grantee, shall be the defendant in any claim for land within a lot
sold, practically the said enactment in no way saves grantees from
litigation. Claimants for land always plead that it is _not_ within
the boundaries of the land sold, and _ergo_ the grantee is made the
defendant to prove that it is. The villagers never having been shown
the boundaries by any Government official (for it is not enacted in the
Waste Land Rules), the question whether the land claimed is within or
without the boundaries is an open one, not always easily decided, and
the suit runs its course.

I even know of cases where, though survey has been charged for at
the exorbitant rate of four annas an acre, the outer boundaries of
the lot have never been surveyed at all, but merely copied from old
Collectorate maps, which showed the boundaries between the zemindaree
and waste lands.[2] Is it strange, then, if buying lands from
Government is often buying litigation, worry, loss of time and money.

In many countries, for example Prussia (there I know it is so, for I
have tested it again and again), there are official records which can
and do show to whom any land in question belongs. This may scarcely be
practicable in India, but surely the question of title being, as it
is, in a far worse state in India than in most countries, any change
would be for the better. Anyhow, the present mode the Government adopts
in selling lands is a grievous wrong to the purchasers. Words cannot
describe the worry and loss some have suffered thereby, and it might
all be so easily avoided.

I have above detailed two of the drawbacks Tea had to contend with in
its infancy; the absurdly high price paid for land was the third.

Again, companies and proprietors of gardens wishing to have large
areas under cultivation gave their managers simple orders to extend,
not judiciously, but in any case. What was the result? Gardens might
be seen in those days with 200 acres of so-called cultivation, but
with 60 or even 70 per cent. vacancies, in which the greater part of
the labour available was employed in clearing jungle for 100 acres
further extension in the following spring. I have seen no garden in
Assam or Cachar with less than 20 per cent. vacancies, many with far
more; and yet most of them were extending. I do not believe _now_ any
garden in all India exists with less than 12 per cent. vacancies, but
a plantation as full as this did _not_ exist formerly.

As the expenditure on a garden is in direct proportion to the area
cultivated, and the yield of Tea likewise in direct proportion to the
number of plants, it follows the course adopted was the one exactly
calculated to entail the greatest expenditure for the smallest yield.
This unnecessary, this wilful extension, was the fourth and a very
serious drawback.

Under this head the fourth drawback may also be included--the fact
that the weeds in all plantations were _ahead_ of the labour; that is
to say, that gardens were not kept clean. This is more or less even
the case to-day; it was the invariable rule then. The consequence was
two-fold--_first_, a small yield of Tea; _secondly_, an increased
expenditure; for it is a fact that the land fifty men can keep always
clean, if the weeds are never allowed to grow to maturity and seed,
will take nearer one hundred if the weeds once get ahead. The results,
too, differ widely: in the first case the soil is always clear; in
the second clear only at intervals. The first, as observed, can be
accomplished with fifty, the latter will take nearly double the men.

The fifth drawback I shall advert to again later, _viz._, the selection
of sloping land, often the steepest that could be found, on which to
plant Tea. The great mischief thus entailed will be fully described
elsewhere. It was the fifth, and not the least, antagonistic point to
success.

Number six was the difficulty in the transport of seed to any new
locality, for nine times out of ten a large proportion failed; and
again the enormous cost of Tea seed in those days, Rs. 200 a maund (Rs.
500 at least, deducting what failed, was its real price). This item of
seed alone entailed an enormous outlay, and was the sixth difficulty
Tea cultivation had to contend with. It was, however, a source of great
profit to the old plantations, and principally accounts for the large
dividends paid for years by the Assam Company.

Again, many managers at that time had no experience to guide them in
the manufacture of Tea; each made it his own way, and often turned out
most worthless stuff. There is great ignorance on the subject at the
present time, but those who know _least_ to-day, know _more_ than the
best informed in the Tea-fever period. Indian Tea was a new thing then;
the supply was small, and it fetched comparatively much higher prices
than it does now. Still much of it was so bad that the average price
all round was low.

Tea manufacture, moreover, as generally practised then, was a much
more elaborate and expensive process than it is now.

This will be explained further on, under the head of “Tea Manufacture;”
I merely now state the fact in support of the assertion that the bad
Tea made in those days, and the expensive way it was done, was the
seventh hindrance to successful Tea cultivation.

Often in those days was a small garden made of 30 or 40 acres, and sold
to a Company as 150 or 200 acres! I am not joking. It was done over
and over again. The price paid, moreover, was quite out of proportion
to even the supposed area. Two or three lakhs of rupees (20,000_l._ or
30,000_l._) have been often paid for such gardens, when not more than
two years old, and 40 per cent. of the existing area, vacancies. The
original cultivators “retired” and the Company carried on. With such
drags upon them (apart from all the other drawbacks enumerated) could
success be even hoped for? Certainly not.

I could tell of more difficulties the cultivation had to contend with
at the outset, but I have said enough to show, as I remarked, “that it
was not strange Tea enterprise failed, inasmuch as it would have been
much stranger if it had not.”

Do any of the difficulties enumerated exist now? And may a person
embarking in Tea to-day hope, with reasonable hope, for success? Yes,
certainly, I think as regard the latter--the former let us look into.[3]

People who understand more or less of Tea are plentiful, and a good
manager, who knows Tea cultivation and Tea manufacture well, may be
found. It will scarcely pay to buy land of the Government at the
present high rates, but many people hold large tracts in good Tea
localities, and would readily sell.

There is plenty of flat land to be got, so no evil from slopes need be
incurred.

Tea seed is plentiful and cheap.

The manufacture of Tea (though still progressing) is simple,
economical, and more or less known. Anyhow a beginner now will commence
where others have left off.

Of course to buy a made garden cheap is better than to make one; but
the result in this case is of course no criterion of what profit may be
expected from Tea cultivation.

As many of the items to be calculated under the heads of cultivation,
manufacture, and receipts will be better understood after details on
these subjects are gone into, I shall reserve the consideration of “how
much profit Tea can give” to the end of this treatise.

FOOTNOTES:

[1] Since the Second Edition went to press, further Rules for Waste
Lands have been enacted. Generally speaking, they are only now leased
to applicants.

[2] I need scarcely observe it is impossible to define lands from maps
alone without the field-book.

[3] Note to Third Edition.--Since the above was written, Teas, both
Indian and Chinese, have had a heavy fall, due to the simple fact that
the supplies have exceeded the demands. But with increased knowledge
and experience, producers can afford to sell cheaper, and the present
absurdly low prices ruling will, I think, work their own cure.



CHAPTER II.

LABOUR, LOCAL AND IMPORTED.


When the very large amount of labour required to carry on a plantation
is considered, it is evident that facilities for it are a _sine quâ
non_ to success. Assam and Cachar, the two largest Tea districts,
are very thinly populated, and almost entirely dependent on imported
labour.[4] The expense of this is great, and it is the one, and
consequently a great drawback to those provinces. The only district
I know of with a good Tea climate and abundance of local labour is
Chittagong.[5] Several other places have a good supply of local labour,
but then their climates are not very suitable.

Each coolie imported costs Rs. 30 and upwards (it used to be much more)
ere he arrives on the garden and does any work. After arrival he has
to be housed; to be cared for and _physicked_ when sick; to be paid
when ill as when working; to have work found for him, or paid to sit
idle when there is no work; and in addition to all this every death,
every desertion, is a loss to the garden of the whole sum expended in
bringing the man or woman. Contrast this with the advantages of local
labour. In many cases no expense for buildings is necessary, as the
labourers come daily to work from adjacent villages, and in such cases
no expense is entailed by sick men, for these simply remain at home.
There is no loss by death or desertions. When no work is required on
the garden, labour is simply not employed. All this makes local labour,
even where the rate of wages is high, very much cheaper than imported.

The action of Government in the matter of imported labour has much
increased the difficulties and expense necessarily attendant on it.
It is a vexed and a very long question which I care not to enter into
minutely, for it has been discussed already _ad nauseam_; still I must
put on record my opinion, after looking very closely into it, that the
Government has not acted wisely, inasmuch as any State interference in
the relations of employer and employed (outside the protection which
the existing laws give) is a radical mistake. As for the law passed
on the subject to the effect that a coolie who has worked out his
agreement and voluntarily enters into a new one shall be, as before,
under Government protection, and his employer answerable as before to
Government, for the way he is housed, treated when sick, &c., it is not
easy to see why such enactments are more necessary in his case than in
that of any other hired servant or labourer throughout all India.

All evidence collected, all enquiries made, tend to show that
coolies are well treated on Tea estates. It is the interest of the
proprietors and managers to do so, and self-interest is a far more
powerful inducement than any the Government can devise. The meddling
caused by the visits of the “Protector of Coolies”[6] to a garden
conduces to destroy the kind feelings which should (and in spite of
these hindrances often do) exist between the proprietor or manager
and his men. I do not hesitate in my belief that imported coolies on
Tea plantations would be better off in many ways were all Government
interference abolished.

I do not decry Government action to the extent of seeing the coolies
understand their terms of engagements, and are cared for on their
journey to the Tea districts; but once landed on the garden, all
Government interference should cease.

The idea of the State laying down how many square yards of jungle each
coolie shall clear in a day, how many square feet he shall dig, &c.,
&c.! Can _any_ certain rates be laid down for such work? Is all jungle
the same, all soil the same; and even if such rates _could_ be laid
down, how can the rules be followed? Bah! they are _not_, never will
be, and the whole thing is too childish for serious discussion.

It is not difficult to sit at a desk and frame laws and rules that
look feasible on paper. It is quite another thing to carry them out.
Over-legislation is a crying evil in India, but there is still a worse,
namely, legislation and official action on subjects of which the said
officials are utterly ignorant.

I have said enough to show imported labour cannot vie with local, nor
would it do so were all the evils of Government interference removed.
I therefore believe Tea property in India will eventually pay best
where local labour exists. This will naturally be the case when other
conditions are equal, but so great are the advantages of local labour,
I believe it will also be the case in spite of _moderate_ drawbacks.

FOOTNOTES:

[4] Not so much so now as when this was first written.

[5] Note to Third Edition.--A portion of the Western Dooars may perhaps
be added, but the labour, though adjacent, is not strictly local. Up
to the present, however, I have had but little expense in importing
coolies to the gardens there in which I am interested.

[6] What a designation! Who invented it, I wonder? A clever man,
doubtless, for Government interference was probably his hobby, and he
quickly perceived the very title would, more or less, render the office
necessary.



CHAPTER III.

TEA DISTRICTS AND THEIR COMPARATIVE ADVANTAGES, CLIMATE, SOIL, ETC., IN
EACH.


The Tea districts in India, that is, where Tea is grown in India
to-day, are--[7]

   1. Assam.
   2. The Dehra Dhoon.
   3. Kumaon (Himalayas).
   4. Darjeeling (Himalayas).
   5. Cachar and Sylhet.[8]
   6. Kangra (Himalayas).
   7. Hazareebaugh.
   8. Chittagong.
   9. Terai below Darjeeling.
  10. Neilgherries (Madras Hills).
  11. Western Dooars.

In fixing on any district to plant Tea in, four things have to be
considered--viz., soil, climate, labour, and means of transport.
When--the district being selected--a site has to be chosen, all but
the second of these have to be considered again, and the lay of land,
nature of jungle, water, and sanitation are also of great importance in
choosing a site.

I will first, then, discuss generally the Tea districts given above
as regards the advantages of each for Tea cultivation. I have seen and
studied Tea gardens in all the districts named, except No. 2. What I
know of the Dehra Dhoon is from what I have read, and what is generally
known of the climate.

Before, however, comparing each district, we should know what are
the necessities of the Tea plant as regards climate and soil. Tea,
especially the China variety, will grow in very varying climates and
soils, but it will not flourish in all of them, and if it does not
flourish, and flourish well, it will certainly not pay.

The climate required for Tea is a hot damp one. As a rule, a good Tea
climate is not a healthy one. The rainfall should not be less than 80
to 100 inches per annum, and the more of this that falls in the early
part of the year the better. Any climate which, though possessing
an abundant rainfall, suffers from drought in the early part of the
year is not, _cæteris paribus_, so good as one where the rain is more
equally diffused. _All_ the Tea districts would yield better with more
rain in February, March, and April; and therefore some, where fogs
prevail in the mornings at the early part of the year, are so far
benefited.

As any drought is prejudicial to Tea, it stands to reason hot winds
must be very bad. These winds argue great aridity, and the Tea plant
luxuriates in continual moisture.

The less cold weather experienced where Tea is, the better for the
plant. It can stand, and will grow in, great cold (freezing point,
and lower in winter, is found in some places where Tea is), but I do
not think it will ever be grown to a profit on such sites. That Tea
requires a temperate climate was long believed and acted upon by many
to their loss. The climate _cannot_ be too hot for Tea if the heat is
accompanied with moisture.

Tea grown in temperate climes, such as moderate elevations in the
Himalayas, is quite different to the Tea of hot moist climates, such as
Eastern Bengal. Some people like it better, and certainly the flavour
is more delicate; but it is very much weaker, and the value of Indian
Tea (in the present state of the home market, where it is principally
used for giving “body” to the washy stuff from China) consists in its
strength. Another all-important point in fixing on a climate for Tea
is the fact, that apart from the strength the yield is double in hot,
moist climes, what it is in comparatively dry and temperate ones. A
really pleasant climate to live in _cannot_ be a good one for Tea. I
may now discuss the comparative merits of the different Tea districts.


ASSAM.

This is the principal home of the indigenous plant. The climate in
the northern portions is perfect, superior to the southern, as more
rain falls in the spring. The climate of the whole of Assam, however,
is very good for Tea. The Tea plant yields most abundantly when hot
sunshine and showers intervene. For climate, then, I accord the first
place to Northern Assam. Southern Assam is, as observed, a little
inferior.

The soil of this province is decidedly rich. In many places there is a
considerable coating of decayed vegetation on the surface, and inasmuch
as in all places where Tea has been or is likely to be planted it is
strictly virgin soil, considerable nourishment exists. The prevailing
soil also is light and friable, and thus, with the exception of the
rich oak soil in parts of the Himalayas, Assam in this respect is
second to none.

As regards labour we must certainly put it the last on the list. The
Assamese, and they are scanty, won’t work, so the planters, with few
exceptions, are dependent on imported coolies; and inasmuch as the
distance to bring them is enormous, the outlay on this head is large,
and a sad drawback to successful Tea cultivation.

The Burhampootra--that vast river which runs from one end of Assam to
the other--gives an easy mode of export for the Tea, but still, owing
to the distance from the sea-board, it cannot rank in this respect as
high as some others.


CACHAR.

The indigenous Tea is also found in a part of this province. The
climate differs but little from Assam. In one respect it is better;
more rain falls in the spring.

The soil is not equal to Assamese soil; it is more sandy, and lacks the
power. Again, there is much more flat land fit for Tea cultivation in
Assam, and there can be no doubt as to the advantage of level surfaces.

As regards transport Cachar has the advantage, for it has equally a
water-way, and is not so distant from Calcutta.

The labour aspect is much the same in the two provinces, both being
almost entirely dependent on imported coolies; but Cachar is nearer the
labour fields than Assam.

However, after discussing separately the advantages of each province,
I propose to draw up a tabular statement, which will show at a glance
the comparative merits of each on each point discussed.


CHITTAGONG.

This is a comparatively new locality for Tea. The climate is better
than Cachar in the one respect that there is less cold weather,
but inferior in the more important fact that much less rain falls
in the spring. In this latter respect it is also inferior to Assam,
particularly to Northern Assam. There is one part of Chittagong, the
Hill Tracts (Tea has scarcely been much tried there yet), which, in
the fact of spring rains, is superior to other parts of the province,
as also in soil, for it is much richer there. On the whole, however,
Chittagong must yield the palm to both Assam and Cachar on the score
of climate, and also, I think, of soil. For though good rich tracts
are occasionally met with, they are not so plentiful as in the two
last-named districts. Always, however, excepting the Hill Tracts of
Chittagong; there the soil is, I think, quite equal to either Assam or
Cachar.

As regards labour (a very essential point to successful Tea
cultivation), Chittagong is most fortunate. With few exceptions (and
those only partial) all the plantations are carried on with local
labour, which--excepting for about two months, the rice-time--is
abundant.

For transport (being on the coast with a convenient harbour, a
continually increasing trade, ships also running direct to and from
England), it is very advantageously situated.

Chittagong possesses another advantage over all other Tea districts
in its large supply of manure. The country is thickly populated, and
necessarily large herds of cattle exist. The natives do not use manure
for rice (almost the sole cultivation), and, consequently, planters can
have it almost for the asking. The enormous advantages of manure in
Tea cultivation are not yet generally appreciated: it will certainly
double the ordinary yield of a Tea garden. A chapter is devoted to this
subject.


TERAI BELOW DARJEELING.

I have seen this, and the Tea in it, since I wrote the first edition of
this Essay.

The soil is _very_ good for Tea. The climate is also a good one, but
there is not as much rain in the early part of the year as planters
could wish. Much difficulty exists about labour, owing to the very
unhealthy climate. As the jungle is cleared, however, this last
objection will be in a measure got over. As it stands now, it is
perhaps the most unhealthy Tea locality in India.

Communication will be very easy when the Northern Bengal Railway is
finished, which it will be immediately.

Except in the point of salubrity (which is, however, an important one),
I think this locality a favourable one for Tea.


THE DEHRA DHOON.

I have heard the first Tea in India was planted here. The lucky men,
two officers, who commenced the plantation, sold it, I believe, in its
infancy, to a company for five lakhs of rupees. What visions did Tea
hold forth in those days!

In climate the Dehra Dhoon is far from good. The hot dry weather of the
North-west is not at all suited to the Tea plant. Hot winds shrivel it
up, and though it recovers when the rains come down, it cannot thrive
in such a climate. One fact will, I think, prove this. In favourable
climates, with good soil and moderate cultivation, 18 flushes or crops
may be taken from a plantation in a season. With like advantages, and
_heavy_ manuring, 22 or even more may be had. In the “Selections from
the Records of the Government of India” on Tea, published in 1857 (a
book to which many owe their ruin), the following appears, showing how
small are the number of flushes in the North-west:--

  [Sidenote: Three general gatherings.]

  _Method of gathering Tea Leaves._--The season for gathering leaves
  generally commences about the beginning of April, and continues until
  October; the number of gatherings varies, depending on the moistness
  and dryness of the season. If the season be good, that is to say,
  if rain falls in the cold weather and spring, and the general rains
  be favourable, as many as five gatherings may be obtained. These,
  however, may be reduced to three general periods for gathering--viz.,
  from April to June, from July to August 15, and from September to
  October 15. If the season be a dry one, no leaves ought to be taken
  off the bushes after October 1, as by doing so they are apt to be
  injured. If, however, there are good rains in September, leaves can
  be pulled until October 15, but no later, as by this time they have
  got hard and leathery and not fitted for making good Teas, and it
  is necessary to give the plants good rest in order to recruit. Some
  plants continue to throw out new leaves until the end of November;
  but those formed during this month are generally small and tough.

When this was written, the experience detailed related to Dehra Dhoon,
the Kumaon, and Kangra gardens, and we see that five flushes or
gatherings are thought good. It however makes matters in this respect
(far from a general fault in the said “Records”) worse than they are.
Ten and twelve flushes, with _high_ cultivation, can be got in the
North-west. But what is this as against twenty and twenty-five?

Labour is plentiful and cheap. The great distance from the coast makes
transport very expensive.


KANGRA.

This is a charming valley, with a delightful climate more favourable
to Tea than the Dehra Dhoon, still it is not a perfect Tea climate. It
is too dry and too cold. The soil is good for Tea, better than that
of the Dhoon, but inferior to some rich soils in the Himalayan oak
forests. Local labour is obtainable at cheap rates. Distance makes
transport for export very difficult; but a good local market now exists
in the Punjab, and a good deal of Tea is bought at the fairs, and taken
away by the wild tribes over the border. With the limited cultivation
there, I should hope planters will find a market for all their produce.
Manure must be obtainable (manure had not been thought of for Tea when
I visited Kangra), and if liberally applied, it will increase the yield
greatly.

Kangra is strictly a Himalayan district, but the elevation is moderate,
if I remember right, about 3,000 feet, and the land is so slightly
sloping it may almost be called level. A great advantage this over the
steep lands, on which most of the Himalayan gardens, many in Cachar,
and some in Assam and Chittagong, are planted.

Kangra is _not_ the best place for a man who wants to make money by
Tea; but for one who would be content to settle there, and content to
make a livelihood by it, a more desirable spot with a more charming
climate could not be found. Land, however, is not easily procured.

The Teas produced in Kangra are of a peculiarly delicate flavour, and
are consequently highly esteemed in the London market.


DARJEELING.

This, too, I have seen since I published the first edition of this
Essay. The elevation of the station, 6,900 feet, is far too great;
but plantations lower down do tolerably well (that is, well for hill
gardens). The climate, like all hill climates, is too cold. As regards
transport the Darjeeling plantations will be well situated when the
railroad now constructing is finished. Like elevations in Darjeeling
and Kumaon are in favour of the former, _first_, because the latitude
is less; _secondly_, because Darjeeling has much more rain in the
spring. I believe, therefore, that the hill plantations of Darjeeling
have a better chance of paying than the gardens in Kumaon, but, as
stated before, no elevated gardens, that is, none in the Himalayas,
have any chance in the race against plantations in the plains, always
providing the latter are in a good Tea climate.

In two respects, however, Darjeeling is behind Kumaon. The soil is not
so good, and the land is much steeper. It is more than absurd, some of
the steeps on which Tea is planted in the former; and such precipices
can, I am sure, never pay. Gardens, barely removed above the Terai
(and there are such in Darjeeling), can scarcely be called “elevated,”
and for them the remarks applied to the Terai are more fitting. As a
broad rule it should be recognised that the lower Tea is planted in the
Himalayas the better chance it has.

All the plants in the Darjeeling gardens, with but few exceptions, are
China.

The China plant makes by far the best Green Tea, and I believe the
Darjeeling gardens would pay much better than they do if they altered
their manufacture from black to green. (See further on, under the head
of Hazareebaugh, what has been done in this way.) All Himalayan gardens
should, in my opinion, make Green Tea (Kumaon has awoke to the fact),
for all have China plants, and can therefore make far better Green Tea
than can be produced from the Hybrid which is so general in plain
gardens.[9]


KUMAON.

It was in this district (a charming climate to live in, with
magnificent scenery to gaze at) I first planted Tea in India, and I
much wish for my own sake, and that of others, I had not done so. I
knew nothing of Tea at the time, and I thought a district selected by
Government for inaugurating the cultivation must necessarily be a good
one. No hill climate _can_ be a good one for Tea; but the inner part of
Kumaon, very cold, owing to its elevation, high latitude, and distance
from the plains, is a peculiarly bad one. Yet there it was Government
made nurseries, distributed seed gratis, recommended the site for
Tea (see the “Records” alluded to), and led many on to their ruin by
doing so. The intention of the Government was good, but the officers
in charge of the enterprise were much to blame, perhaps not for making
the mistake at first (no one _at the first_ knew what climate was
suitable), but for perpetuating the mistake, when later very little
enquiry would have revealed the truth. I believe it was guessed at by
Government officials long ago, but it was easier to sing the old tune;
and a very expensive song it has proved to many.[10]

I need scarcely, after this, add that I do not approve of Kumaon for
Tea. An exhilarating and bracing climate for man is not suited to
the Tea plant. The district has one solitary advantage--rich soil.
I have never seen richer, more productive land than exists in some
of the Kumaon oak forests, but even this cannot in the case of Tea
counter-balance the climate. Any crop which does not require much heat
and moisture will grow to perfection in that soil. Such potatoes as
it produces! Were the difficulties of transport not so great, a small
fortune might be made by growing them.

Could any part of Kumaon answer for Tea it would be the lower
elevations in the outer ranges of the hills, but these are precisely
the sites that have _not_ been chosen. Led, as in my own case, partly
by the Government example, partly by the wish to be _out_ of sight of
the “horrid plains,” and _in_ sight of that glorious panorama the snowy
range, planters have chosen the interior of Kumaon. Some wisely (I was
not one of them) selected low sites, valleys sheltered from the cold
winds; but even their choice has not availed much. The frost in winter
lingers longest in the valleys, and though doubtless the yield there is
larger, owing to the increased heat in summer, the young plants suffer
much in the winter. The outer ranges, owing to the heat radiating from
the plains, are comparatively free from frost, but there again the soil
is not so rich. Still they would unquestionably be preferable to the
interior.

Labour is plentiful in Kumaon and very cheap--Rs. 4 per mensem.
Transport is very expensive. It costs not a little to send Tea from the
interior over divers ranges of hills to the plains. It has then some
days’ journey by cart ere it meets the rail, to which 1,000 miles of
carriage on the railroad has to be added.

Since the above was written, Kumaon has secured a good local market,
and I believe sells most of its Tea unpacked to merchants who come from
over the border to buy it.

It has also improved its position greatly by making Green Teas, for
which, as observed before, the China plant is so well fitted. With
those two advantages, though the climate is inferior, I suspect that
Tea there now pays better than in Darjeeling.

Gurhwall is next to Kumaon, and so similar that I have not thought it
necessary to discuss it separately. The climate is the same, the soil
as a rule not so good. There is one exception though, a plantation near
“Lohba,” the Teas of which (owing, I conceive, to its peculiar soil)
command high prices in the London market. The gardens, both in Kumaon
and Gurhwall, have been generally much better cared for than those in
Eastern Bengal. As a rule they are private properties managed by the
owners.


HAZAREEBAUGH.

This district I have resided in since I wrote the first edition of this
Essay. The climate is too dry, and hot winds are felt there. A great
compensation, though, is labour; it is more abundant and cheaper in
this district than in any other. The carriage is all by land, and it
is some distance to the rail. But the Tea gardens at Hazareebaugh can
never vie with those in Eastern Bengal, inasmuch as the climate is very
inferior. The soil is very poor.


NEILGHERRIES.

The climate is superior to the Himalayan, for the frost is very slight.
Were, however, more heat there in summer, it would be better.

Some of the Teas have sold very well in the London market, for as
regards delicacy of flavour they take a high place.

The soil is good, but the temperate climate which holds on these “blue
mountains” is not favourable to a large produce.


WESTERN DOOARS.

When the second edition of this work was printed, this district was
unknown as Tea locality.

My attention was directed to it in 1874; I was the second who planted
Tea in it, and I have now completed a garden there.

As regards climate, soil, and lay of land, it is perfect, and I believe
it will eventually prove the most paying district in India for Tea.

The Northern Bengal Railway, just opened, gives it great advantages for
transport.

Having now discussed each district, all of which, except the Dehra
Dhoon, I have seen, I give, in further elucidation, Meteorological
Tables. For those not mentioned in the tables I have failed to acquire
the necessary information.

My thanks are due to Dr. Coates, at Hazareebaugh, for his kindness in
supplying me with much of the data from which the following tables are
framed:--


TABLE OF ELEVATION AND TEMPERATURE OF TEA LOCALITIES.

_N.B.--The exact temperature of other Tea Districts not being known, I
have confined myself to these; but general remarks on the elevation and
temperature of other Tea localities will be found elsewhere._

  +----------+-------------+---------+--------+---------+---------+-------
  |Districts |  Place      |Elevation| Details|January  |February |  March
  |          |             | in feet |        |         |         |
  +----------+-------------+---------+--------+---------+---------+-------
  |          |             |        {|Monthly}|   61·7  |   63·0  |   72·6
  |         {|             |        {| Temp. }|         |         |
  |         {|Goalparah    |   386  {|Do. Max.|   77·2  |   87·9  |   94·0
  |         {|             |        {|        |         |         |
  |         {|             |        {|Do. Min.|   49·0  |   48·0  |   57·2
  |         {|-------------+---------+--------+---------+---------+-------
  |ASSAM    {|Gowhatty     |   134   |Monthly}|   63·6  |   67·6  |   74·5
  |         {|             |         | Temp. }|         |         |
  |         {|-------------+---------+--------+---------+---------+-------
  |         {|Seebsaugor   |   370   |Monthly}|   60·0  |   64·1  |   69·3
  |         {|             |         | Temp. }|         |         |
  |         {|-------------+---------+--------+---------+---------+-------
  |         {|Debrooghur   |   396   |Monthly}|   62·2  |   63·4  |   71·3
  |          |             |         | Temp. }|         |         |
  +----------+-------------+---------+--------+---------+---------+-------
  |CACHAR    |Cachar       |    76   |Monthly}|   62·9  |   66·6  |   73·4
  |          |             |         | Temp. }|         |         |
  +----------+-------------+---------+--------+---------+---------+-------
  |CHITTAGONG|Chittagong   |   191   |Monthly}|   68·5  |   72·3  |   80·5
  |          |             |         | Temp. }|         |         |
  +----------+-------------+---------+--------+---------+---------+-------
  |          |             |         |Monthly}|         |         |
  |DAR-     }|             |         |  Temp.}|   42.2  |   43.8  |   52.0
  | JEELING }| Darjeeling  |  6952   |Do. Max.|   62·0  |   66·0  |   72·0
  |          |             |         |Do. Min.|   32·0  |   28·0  |   39·0
  +----------+-------------+---------+--------+---------+---------+-------
  |          |             |         |Monthly}|         |         |
  |CHOTA    }|             |         |  Temp.}|   62.7  |   67.1  |   73.7
  | NAGPORE }| Hazareebaugh|  2010   |Do. Max.|   82·0  |   91·0  |   94·0
  |          |             |         |Do. Min.|   44·0  |   46·4  |   55·0
  +----------+-------------+---------+--------+---------+---------+-------
  |NEIL-    }|             |         |Monthly}|         |         |
  | GHERRIES}| Ootacamund  |  7490   |  Temp.}|   51·5  |   52·8  |   57·3
  +----------+-------------+---------+--------+---------+---------+-------

  -------+---------+---------+---------+---------+---------+--------
  April  |   May   |  June   |  July   | August  |September| October
         |         |         |         |         |         |
  -------+---------+---------+---------+---------+---------+--------
   77·6  |   76·0  |   80·3  |   82·1  |   81·6  |   80·5  |   77·5
         |         |         |         |         |         |
   97·0  |   91·0  |   91·0  |   92·0  |   91·5  |   92·0  |   89·0
         |         |         |         |         |         |
   62·6  |   67·0  |   70·0  |   73·7  |   73·0  |   70·1  |   62·3
  -------+---------+---------+---------+---------+---------+--------
   77·4  |   80·4  |   81·8  |   83·0  |   82·9  |   82·2  |   79·2
         |         |         |         |         |         |
  -------+---------+---------+---------+---------+---------+--------
   73·8  |   78·5  |   82·4  |   83·6  |   83·5  |   83·1  |   78·3
         |         |         |         |         |         |
  -------+---------+---------+---------+---------+---------+--------
   72·7  |   77·1  |   80·7  |   83·7  |   81·8  |   81·0  |   75·6
         |         |         |         |         |         |
  -------+---------+---------+---------+---------+---------+--------
   76·8  |   80·9  |   82·2  |   83·3  |   81·7  |   81·2  |   79·6
         |         |         |         |         |         |
  -------+---------+---------+---------+-------------------+--------
   83·5  |   84·5  |   84·0  |   82·2  |   82·3  |   83·0  |   81·6
         |         |         |         |         |         |
  -------+---------+---------+---------+---------+---------+--------
         |         |         |         |         |         |
   58.7  |   62.1  |   63.7  |   64.9  |   64.4  |   63.0  |   57.3
   78·0  |   79·0  |   79·0  |   79·0  |   75·0  |   80·0  |   78·0
   48·0  |   48·0  |   57·0  |   58·0  |   59·0  |   57·0  |   44·0
  -------+---------+---------+---------+---------+---------+--------
         |         |         |         |         |         |
   85.6  |   88.6  |   83.8  |   77.8  |   79.3  |   77.5  |   72.6
  107·0  |  100·0  |  103·0  |   89·0  |   88·0  |   87·0  |   84·0
   67·0  |   72·0  |   71·0  |   71·0  |   73·0  |   70·0  |   59·0
  -------+---------+---------+---------+---------+---------+--------
         |         |         |         |         |         |
   60·1  |   60·8  |   57·9  |   55·8  |   56·1  |   56·4  |   55·9
  -------+---------+---------+---------+---------+---------+--------

  ---------+---------+------+------+------+------+-----+
  November |December |D.J.F.|M.A.M.|J.J.A.|S.O.N.|Year |
           |         |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
     69·0  |   64·6  | 63·1 | 75·4 | 81·3 | 75·6 | 73·8|
           |         |      |      |      |      |     |
     84·3  |   78·3  |      |      |      |      |     |
           |         |      |      |      |      |     |
     50·8  |   50·0  |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
     71·1  |   65·5  | 65·6 | 77·4 | 82·6 | 77·5 | 75·8|
           |         |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
     69·4  |   62·4  | 62·2 | 73·7 | 83·2 | 76·9 | 74·0|
           |         |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
     67·4  |   61·0  | 62·2 | 73·7 | 82·1 | 74·7 | 73·2|
           |         |      |      |      |      |     |
  ---------+----------------+------+------+------+-----+
     70·6  |   65·4  | 64·9 | 77·0 | 82·4 | 77·1 | 75·3|
           |         |      |      |      |      |     |
  ---------+---------+------+-------------+------+-----+
     73·7  |   68·9  | 69·9 | 82·8 | 82·8 | 79·4 | 78·7|
           |         |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
           |         |      |      |      |      |     |
     49.4  |   44.7  | 43.5 | 57.6 | 64.3 | 58.4 |55.9 |
     69·0  |   60·0  |      |      |      |      |     |
     38·0  |   33·0  |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
           |         |      |      |      |      |     |
     64.8  |   61.4  | 63.7 | 82.6 | 80.3 | 71.6 |74.5 |
     78·0  |   76·0  |      |      |      |      |     |
     52·0  |   44·0  |      |      |      |      |     |
  ---------+---------+------+------+------+------+-----+
           |         |      |      |      |      |     |
     53·9  |   51·9  | 52·1 | 59·4 | 56·6 | 55·4 |55·9 |
  ---------+---------+------+------+------+------+-----+

N.B.--The letters in the columns, between December and the year, refer
to months; thus, D. J. F. is December, January, February. The figures
show the average temperature during those months.


TABLE OF LATITUDE, LONGITUDE, AND RAINFALL OF TEA LOCALITIES.

_N.B.--The exact rainfalls of other Tea Districts not being known, I
have confined myself to these; but general remarks on the rainfall in
other Tea localities will be found elsewhere._

  +---------+-----------+--------+---------+------------------------
  |Districts|  Place    |Latitude|Longitude|       Details
  |---------+-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |         |Goalparah  | 26°11′ |  90°36′ |{  years               }
  |         |           |        |         |{Days rain fell in 1869}
  |         |-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |  ASSAM  | Gowhatty  | 26° 5′ |  91°43′ |{  years               }
  |         |           |        |         |{Days rain fell in 1869
  |         |-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |         |Seebsaugor | 27° 2′ |  94°39′ |{  years               }
  |         |           |        |         |{Days rain fell in 1869
  |---------+-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |  CACHAR |  Cachar   | 24°48′ |  92°43′ |{  years               }
  |         |           |        |         |{Days rain fell in 1869}
  |---------+-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |         |Chittagong | 22°20′ |  91°44′ |{  YEARS               }
  |         |           |        |         |{Days rain fell in 1869}
  | CHITTA- |-----------+--------+---------+------------------------
  |  GONG   |           |        |         | Rain in 1869
  |         |Hill Tracts|    ?   |     ?   |
  |         |           |        |         | Days rain fell in 1869
  +---------+-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several }
  |         | Darjeeling| 27° 3′ |  88°18′ |{  years               }
  |         |           |        |         | Days rain fell in 1869
  |DAR-     |-----------+--------+---------+------------------------
  | JEELING |           |        |         | Rain in 1869
  |         |Western  } |    ?   |     ?   |
  |         |Dooars[11]}|        |         | Days rain fell in 1869
  +---------+-----------+--------+---------+------------------------
  |         |           |        |         |{Average rain, several}
  | CHOTA   | Hazaree- }| 24° 0′ |  85°20′ |{  YEARS              }
  | NAGPORE |     baugh}|        |         | Days rain fell in 1869
  +---------+-----------+--------+---------+------------------------

  --------+---------+---------+---------+---------+---------+-------
  January |February |  March  |  April  |   May   |  June   |  July
  --------+---------+---------+---------+---------+---------+-------
    0·42  |   0·76  |   1·84  |   4·85  |  11·72  |  23·72  |  21·33
          |         |         |         |         |         |
      2   |     2   |     4   |     8   |    19   |    24   |    22
  --------+---------+---------+---------+---------+---------+-------
    0·70  |   1·43  |   1·48  |   7·27  |  10·92  |  13·29  |  13·08
          |         |         |         |         |         |
      2   |     2   |     4   |     8   |    16   |    16   |     9
  --------+---------+---------+---------+---------+---------+-------
    1·18  |   2·43  |   3·77  |  10·15  |  11·04  |  15·56  |  14·87
          |         |         |         |         |         |
     11   |     9   |    10   |    13   |    22   |    13   |    19
  --------+---------+---------+---------+---------+---------+-------
    0·50  |   3·53  |   6·09  |  12·69  |  16·12  |  19·55  |  24·58
          |         |         |         |         |         |
      2   |     9   |    10   |    16   |    18   |    20   |    18
  --------+---------+---------+---------+---------+---------+-------
    0·37  |   1·62  |   1·31  |   5·46  |   9·42  |  22·92  |  22·54
          |         |         |         |         |         |
      1   |     7   |     3   |     4   |    14   |    15   |    21
  --------+---------+---------+---------+---------+---------+-------
    Nil   |   1·90  |   1·50  |  12·55  |   9·00  |  12·50  |  18·20
          |         |         |         |         |         |
    Nil   |     4   |     4   |     7   |    13   |    16   |    22
  --------+---------+---------+---------+---------+---------+-------
   0·76   |   1·60  |   1·65  |   3·62  |   7·01  |  27·50  |  29·40
          |         |         |         |         |         |
     2    |     3   |     5   |     9   |    17   |    23   |    26
  --------+---------+---------+---------+---------+---------+-------
   0·80   |   2·00  |   1·50  |   6·60  |  25·30  |  27·30  |  46·50
          |         |         |         |         |         |
     3    |     3   |     5   |     7   |    15   |    19   |    25
  --------+---------+---------+---------+---------+---------+-------
   0·42   |   0·52  |   0·75  |   0·42  |   1·37  |  10·99  |  14·63
          |         |         |         |         |         |
     4    |    Nil  |     7   |   Nil   |     5   |    11   |    24
  --------+---------+---------+---------+---------+---------+-------

  --------+---------+---------+---------+---------+------+
  August  |September| October |November |December |Total |
  --------+---------+---------+---------+---------+------|
   12·69  |  10·93  |   5·61  |   0·39  |   0·20  |94·44 |
          |         |         |         |         |      |
     18   |    15   |     5   |   Nil   |   Nil   | 119  |
  --------+---------+---------+---------+---------+------|
   11·98  |   6·82  |   3·20  |   0·47  |   0·12  |70·76 |
          |         |         |         |         |      |
     10   |    14   |     2   |   Nil   |     1   |  84  |
  --------+---------+---------+---------+---------+------|
   13·88  |  11·13  |   4·46  |   1·29  |   0·69  |90·45 |
          |         |         |         |         |      |
     23   |    17   |     8   |   Nil   |     2   | 147  |
  --------+---------+---------+---------+---------+------|
   16·84  |  13·90  |   7·77  |   7·03  |   0·79  |123·3 |
          |         |         |         |         |      |
     25   |    19   |     8   |   Nil   |    Nil  | 145  |
  --------+---------+---------+---------+---------+------|
   23·04  |  13·01  |   5·93  |   2·30  |   0·55  |108·47|
          |         |         |         |         |      |
     25   |    17   |     5   |   Nil   |     1   | 113  |
  --------+---------+---------+---------+---------+------+
   14·30  |  12·70  |   5·70  |    Nil  |   0·50  | 88·85|
          |         |         |         |         |      |
     19   |    19   |    4    |    Nil  |    1    |  109 |
  --------+---------+---------+---------+---------+------+
   29·09  |  18·06  |   6·56  |   0·20  |   0·14  |129·50|
          |         |         |         |         |      |
     22   |    24   |    7    |     1   |    2    |  148 |
  --------+---------+---------+---------+---------+------+
   83·50  |  46·50  |   9·60  |     ?   |   2·40  |252·00|
          |         |         |         |         |      |
     28   |    22   |    5    |     ?   |    ?    |   ?  |
  --------+---------+---------+---------+---------+------+
   11·44  |   6·26  |   3·51  |   0·19  |   0·02  | 50·52|
          |         |         |         |         |      |
     16   |    21   |    9    |    Nil  |    1    |   98 |
  --------+---------+---------+---------+---------+------+

I will now endeavour to draw up a tabular statement of the respective
advantages of the various Tea districts as regards climate, labour, lay
of land, soil, facilities of procuring manure and transport.

In importance I regard them in the order given. I place labour before
soil, because the fact is, in all the provinces suitable and good soil
for Tea can be found _somewhere_; and therefore, while soil is all
important in selecting a site, it is secondary to labour in deciding
on a district. Lay of land comes after labour. When my information on
any point is not sure I place a note of interrogation. Where advantages
are equal, or nearly so, I give the same number, and the greater the
advantage of a district on the point treated in the column the smaller
the number. Thus, under the head of Climate, Assam is marked 1.

As the following table gives no information as to which of all the
districts possesses the greatest advantages, _all things considered_,
but only gives my opinion of each under each head, and the subject
closed in this way would be unsatisfactory, I may state that, in my
opinion, the choice should lie between the three first and the last on
the list; and my choice would be the last.


_Comparative advantages of the Tea Districts in India as regards
climate, labour, lay of land, soil, manure, and transport._

  +----------------------+-------+------+-----------+----+------+------------+
  |    Tea Districts     |Climate|Labour|Lay of Land|Soil|Manure|  Transport |
  +----------------------+-------+------+-----------+----+------+------------+
  |Assam                 |  1    |  4   |     1     |  1 |   4  | 3}         |
  |Cachar                |  2    |  4   |     2     |  2 |   4  | 2} Water   |
  |Chittagong            |  3    |  2   |     2     |  2 |   1  | 1} carriage|
  |Chittagong Hill Tracts|  3    |  3   |     3     |  1 |   2  | 1}         |
  |Terai below Darjeeling|  2    |  4   |     1     |  1 |   3  | 5}         |
  |Darjeeling            |  4    |  3   |     5     |  3 |   3  | 6}         |
  |Hazareebaugh          |  6    |  1   |     1     |  4 |   2  | 4}         |
  |Kangra                |  4    |  3   |     1     |  3 |   3  | 9} Land    |
  |Dehra Dhoon           |  5    |  3   |     1     |  3 |   3  | 7} carriage|
  |Kumaon                |  5    |  3   |     4     |  2 |   3  | 8}         |
  |Neilgherries          |  4    |  3   |     2     |  2 |   3  | 4}         |
  |Western Dooars        |  1    |  3   |     1     |  1 |   4  | 1}         |
  +----------------------+-------+------+-----------+----+------+------------+

FOOTNOTES:

[7] Note to Third Edition.--I give them, as far as I know, in the
order they became Tea districts. Though in the said order there is,
I believe, no great error, I may be open to correction in one or two
instances.

[8] These are virtually one, and I shall allude to both as Cachar.

[9] When this was written the demand for Green Teas in Europe was
greater than it is now. Still Kumaon has found a local market for Green
Tea over the border, that is, among the Asiatic tribes, and Darjeeling
might do the same.

[10] Is it possible that the continued deception (it was nothing less)
was owing to the fact that Government had gardens to sell there? They
were advertised for sale a long time at absurd prices.

[11] The rainfall given for the Western Dooars cannot be relied on.
Perhaps more than the average fell in 1869, but anyhow, I should think
83 inches registered for August is an error. I know the yearly fall
there is _not_ 252 inches.



CHAPTER IV.

SOIL.


To pronounce as precisely on soil as to climate is not easy. The Tea
plant will grow on almost any soil, and will flourish on many. Still
there are broad general rules to be laid down in the selection of soils
for Tea, which no one can ignore with impunity.

When first I turned my attention to Tea, I collected soils from many
gardens, noting in each case how the plants flourished. I then sat
down to examine them, never doubting to arrive at some broad practical
conclusions. I was sadly disappointed. I found the most opposing soils
nourished, apparently, equally good plants. I knew not then much about
Tea, and judged of the Tea bushes mostly by the size (a very fallacious
test); still, after-experience has convinced me I was more or less
right in the conclusion I then came to, that several soils are good for
Tea.

Nothing, then, but broad general rules can be laid down on this point,
for I defy anyone to select any one soil as the best for Tea, to the
exclusion of others.

A light sandy loam is perhaps as good a soil as any _out_ of the
Himalayas. It ought to be deep, and the more decayed vegetable matter
there is lying on its surface the better. If deep enough for the
descent of the tap-root, say 3 feet, it matters not much what the
subsoil is, otherwise a yellowish red subsoil is an advantage. This
subsoil is generally a mixture of clay and sand. Much of Assam, Cachar,
and Chittagong is as the above, but, as a rule, it is richest in Assam,
poorest in Chittagong.

Where the loam is of a greasy nature (very different to clay), with a
mixture of sand in it, it is superior to the above, for it has more
body. All good Tea soils must have a fair proportion of sand, and if
not otherwise apparent, it may be detected by mixing a little of the
soil with spittle and rubbing it on the hand. If the hand be then held
up towards the sun, the particles of sand will be seen to glisten.

The soil so common in Kumaon, that is, light rich loam with any amount
of decayed vegetable matter on it, and with a ferruginous reddish
yellowish subsoil, is, I consider, the finest soil in the world for
Tea. The rich decayed vegetable matter is the produce for centuries of
oak leaves in the Himalayan forests, and, as all the world knows, oak
only grows in temperate climes.

It was long believed that Tea would thrive best on poor soil. The idea
was due to the description of Tea soils in China to be found in the
first books that treated of Tea. But the fact that Tea, as a rule, is
only grown in China on soil which is useless for anything else quite
alters the case. If a soil is light and friable enough, it cannot be
too rich for Tea.

Ball’s book “On the Cultivation and Manufacture of Tea in China” has
much on Tea soils, but the opinions the author collected are sadly at
variance, and on the whole teach nothing.

In conclusion, I will attempt to point out the qualities in soils in
which the Tea plant delights, as also the qualities it abhors.

It loves soils friable, that is, easily divided into all their atoms.
This argues a fair proportion of sand, but this should not be in
excess, or the soil will be poor. The soil should be porous--imbibing
and parting with water freely. The more decayed vegetable matter on its
surface the better.

To be avoided are stiff soils of every kind, as also those which
when they dry, after rain, cake together and split. Avoid also
black-coloured, or even dark-coloured earths. All soils good for the
Tea plant are light coloured. If, however, the dark colour arises
from decayed vegetation, that is not the colour of the soil, and, as
observed, vegetable matter is a great advantage. Judge of colour when
soil is dry--for even light-coloured soil looks dark when wet. Soil
which will make bricks will not grow Tea; and though I have sometimes
seen young plants thrive on stiff soil, I do not believe in any stiff
soil as a permanence.

Stones, if not in excess, are advantageous in all soils inclined to
be stiff, for they help to keep them open. But then they must not be
large, as if so they act as badly as a rocky substratum preventing the
descent of the tap-root.

The reason, I take it, why Tea thrives best in light soils is that the
spongioles or ends of the feeding roots are very tender, and do not
easily penetrate any other.

There is more nourishment in stiffer soils, but for this reason the Tea
plant cannot take advantage of it.

If a chosen soil be too stiff, it may be much improved for Tea by
mixing sand with it. However, even where sand is procurable near, the
expense of this is great. When done, the sand should be mixed with the
soil taken out of the holes in which the plants are to be placed (see
Transplanting), and it may be done again later by placing sand round
the plants and digging it in. All this though is extra labour and very
expensive, so none but a good Tea soil should ever be selected, and
it is very easily found, for it exists in parts of all the districts
discussed.



CHAPTER V.

NATURE OF JUNGLE.


I have not much to say under this head. I have heard many opinions
as to the kind of trees and jungle that should exist in contemplated
clearances, but I attach little or no weight to them, at all events in
Bengal.

In the Himalayas it is somewhat different. There oak trees should be
sought for; their existence invariably makes rich soil.[12] Fir, on the
contrary, indicates poor soil. At elevations, however, the desideratum
of a warm aspect interferes, for the best oak forests are on the colder
side. I speak of course of elevations practicable, say three or four
thousand feet; above this it is a waste of money to try and cultivate
Tea.

In Bengal I do not think the nature of the jungle on land contemplated
signifies much. As a rule, the thicker the jungle the richer the
soil; but in seeking for a site large trees should not be a _sine quâ
non_. Much of the coarse grass land is very good, and large trees add
enormously to the expense of clearings.[13] It is not cutting them down
which is so expensive, it is cutting them up and getting rid of them by
burning, or otherwise, after the former is done.

I have discussed soil fully already, and need only add here that if the
knowledge to do so exists, it is better to judge of soil from the soil
itself than from the vegetation on it, though doubtless a fact that
luxuriant vegetation indicates rich soil.

FOOTNOTES:

[12] The oak tree leaves cause a rich deposit of vegetable matter.

[13] The Western Dooars are in many parts covered with this coarse
grass, and nowhere is there better soil.



CHAPTER VI.

WATER AND SANITATION.


These may be discussed together and shortly.

Of course adjacent water-carriage is a great advantage for a garden,
and it should be obtained, if possible, in selecting a site. The
expense of land-carriage, where there is no rail, is great, and Tea
cultivation requires all advantages to make it pay well.

But it is water for a garden that particularly concerns us now. It
is not easy to find land that can be irrigated (this is discussed
elsewhere), but no labour or expense in getting such land would be
thrown away. Irrigation, combined with high cultivation in other
respects, will give a yield per acre undreamt of.

In no case should a plantation be made except where a running stream
is handy. Water is a necessity for seedlings, and a plentiful adjacent
supply of it is a great desideratum for the comfort and health of every
soul on the garden. We all know how dependent the natives are on water,
and it is evident facilities in this respect will conduce much (whether
the labour be local or imported) both to get and keep coolies. Norton’s
tube wells--a cheap and most efficient mode of procuring water--will,
I doubt not, be eventually much used on Tea plantations.

It has been observed that, as a rule, a good Tea climate is not a
healthy one. There is no getting over the fact, and we can only make
the best of it. The house, the factories, and all the buildings should
be placed as high as possible, and not very close to each other, both
for the sake of health and in the event of fire. The locality should
be well drained, and cleanliness be attained in every possible way.
Give the coolies good houses, with raised mechans to sleep on, and
sprinkle occasionally carbolic acid powder in your own house and those
of others.

Sanitation is, however, a large subject. It can be studied elsewhere.
General ideas on it, and on the properties of the commonest medicines,
are a great advantage to any intending Tea planter.



CHAPTER VII.

LAY OF LAND.


The first idea prevailing about Tea was that it should be planted on
slopes. It was thought, and truly, that the plant was impatient of
stagnant water, and so it is, but it is not necessary to plant it
on slopes in consequence. Pictures of Chinese, suspended by chains
(inasmuch as the locality could not be otherwise reached), picking
Tea off bushes growing in the crevices of rocks, somewhat helped this
notion; and when stated, as it was, that the Tea produced in such
places was the finest, and commanded the highest price--which was
not true--intending planters in India went crazy in their search for
impracticable steeps! Much of the failure in Tea has arisen from this
fact, for a great part of many, the whole of some, gardens have been
planted on land so steep that the Tea can never last or thrive on it.
This is especially the case in parts of the Darjeeling district.

Sloping land is objectionable in the following respects. It cannot
be highly cultivated in any way (I hold Tea will only pay with high
cultivation), for high cultivation consists in frequent digging, to
keep the soil open and get rid of weeds, and liberal manuring. If
such soil is dug in the rainy season, it is washed down to the foot
of the hill, and if manure is applied at any time of the year, it
experiences the same fate when the rain comes. As it cannot be dug,
weeds necessarily thrive and diminish the yield by choking the plants.

The choice is therefore of two evils: “low cultivation and weeds,”
or “high cultivation which bares the roots of the plants in a
twelvemonth.” Of the two, the first _must_ be chosen, for if the latter
were pursued, the plants, getting gradually more and more denuded of
soil, would simply topple over in two or three years. But choosing the
lesser evil, the mischief is not confined to the bad effects of low
cultivation. Dig the land as little as you will, the great force of the
rains washes down a good deal of soil. The plants do not sink as the
soil lowers, and the consequence is that all Tea plants on slopes have
the lower side bare of earth and the roots exposed. This is more and
more the case the steeper the slope. These exposed roots shrivel up as
the sun acts on them, the plant languishes and yields very little leaf.

Attempts are made to remedy the mischief by carrying earth up from
below yearly, and placing it under the plant; but the expense of doing
this is great, and the palliation is only temporary, for the same thing
occurs again and again as each rainy season returns.

The mischief is greater on stiff than on sandy soils, for on the former
the earth is detached in great pieces and carried down the hill. I know
one garden in Chittagong, a large one, where the evil is so great, that
the sooner the cultivation is abandoned the better for the owners.

A great many gardens in India, indeed the majority, are on slopes; a
few in Assam, the greater number in Cachar, some in Chittagong, and
almost all the Himalayan plantations. Such of these as are on _steep_
slopes will, I believe, never pay, and instead of improving yearly (as
good gardens, highly cultivated, should do even after they have arrived
at full bearing) such, I fear, will deteriorate year by year.

Plantations on moderate slopes need not fail because of the slopes. The
evils slight slopes entail are not great, but the sooner the fact is
accepted that sloping cannot vie against flat land for the cultivation
of Tea the better.

Where only the lower parts of slopes are planted, the plants do very
well. The upper part being jungle the wash is not great, and the plants
benefit much by the rich vegetable matter the rain brings down from
above. I have often seen very fine plants on the lower part of slopes,
where the upper has been left in jungle, and I should not hesitate to
plant such portions _if_ the slope was moderate.

Where teelah land, in Eastern Bengal, or sloping land in the Himalayas,
Chittagong, or elsewhere, has to be adopted, aspect is all-important.
A good aspect in one climate is bad in another. In Assam, Cachar,
Chittagong, and all warm places, choose the coolest; at high elevations
(temperate climes), the warmest.

In the Himalayas, moreover, the warmer aspects are, as a rule, the
most fertile; _vice versâ_ in warm localities. Many a garden, which
would have done very well on the moderate slopes chosen had _only_ the
proper aspects been planted, has been ruined by planting all sides of
teelahs or hills indiscriminately. The southern and western slopes
of plantations in warm sites are generally very bare of plants. Not
strange they should be so, when the power of the reflected rays of the
afternoon sun is considered. Again, in cold climates plants cannot
thrive on northern aspects, for their great want in such climes is heat
and sunshine. Let the above fault, then, be avoided in both cases, for
though doubtless a garden is more handy, and looks better in one piece
planted all over without any intervening jungle, even patches of jungle
look better, and are decidedly cheaper, than bare cultivated hills.

Of flat land, after what I have written, I need not add much. It is
of two kinds, table and valley land; the former is very rare in Tea
districts, at least of any extent, which makes it worth while to plant
it. There are two gardens in Chittagong on such flat table land, and
they are both doing very well. Table land cannot be too flat, for the
natural drainage is so great no stagnant water can lie.[14]

Valley land is not good if it is _perfectly_ flat. It will then be
subject to inundation and stagnant water. There is nothing that kills
the plant so surely and quickly as the latter. Even quite flat valleys
can be made sweet by artificial drainage, but to do this a lower level,
not too far distant, must exist, and the danger is not quite removed
then. Valleys in which no water-course exists, and which slope towards
the mouth _alone_, are to be avoided, for the plants near the mouth
always get choked with sand. The best valleys are those with a gentle
slope both ways, one towards the lowest line of the valley, be it a
running water-course, or a dry nullah which carries off rain, the other
towards the mouth of the valley. Such valleys drain themselves, or at
least very little artificial drainage is necessary. A valley of this
kind, with a running stream through it, is _most_ valuable for Tea, and
if the other advantages of soil and climate are present it is simply a
perfect site. Such however are not frequent. If in such valleys, as is
generally the case, the slope from the head to the mouth is enough, the
running stream can be “bunded” (shut up) at a high level, and brought
along one side at a sufficient elevation to irrigate the whole.

I have never seen but one garden in a valley that fulfils all these
conditions exactly. It is in Chittagong; the soil is good, labour
plentiful, and manure abundant. It ought to do great things, for the
possibility of irrigating plants in the dry season (which, as observed,
is very trying in Chittagong) will give several extra flushes in the
year.

Of course in the wet season on such land the water must be allowed to
resume its natural course.

Narrow valleys are not worth planting. No narrow tracts of land, with
jungle on both sides, are worth the expense of cultivation, for the
continual encroachment of the jungle gives much extra work. The plants,
moreover, in very narrow valleys get half-buried with soil washed down
from the adjacent slopes. Narrow valleys are therefore, in any case,
better avoided.

To conclude shortly, flat lands can be highly cultivated, steep
slopes cannot. Tea pays best (perhaps not at all otherwise) with high
cultivation--_ergo_, flat lands are preferable.

FOOTNOTE:

[14] I am now commencing a second garden in the Western Dooars on flat
table land, and the site is an exceptionally favourable one.



CHAPTER VIII.

LAYING OUT A GARDEN.


By this I mean, so dividing it when first made into parts, that
later the said parts shall be easily recognised, and separately or
differently treated, as they may require it.

The usual custom is to begin at one end of a plantation, and dig it
right through to the other. In the same way with the pruning and
plucking, and I believe the system is a very bad one. Different
portions of gardens require different treatment, inasmuch as they
differ in soil, and otherwise. One part of a plantation is much more
prolific of weeds than another--how absurd that it should be cleaned no
oftener! This is only one exemplification of difference of treatment,
but in many ways it is necessary, most of all in plucking leaf.

All parts of a plantation, owing in some places to the different ages
of the plants, in others to the variety in the soil and its productive
powers, in others to slopes or to aspect, do not yield leaf equally,
that is, flush does not follow flush with equal rapidity. In some
places (supposing each part to be picked when the flush is ready) seven
days’ interval will exist between the flushes, in others nine, ten, or
twelve; but no attention, as a rule, is paid to this. The pickers have
finished the garden at the west end, the east end is again ready, and
when done, the middle part will be taken in hand, be it ready or be it
not! It may be that the middle part flushes quicker than any other; in
this case the flush will be more than mature when it is taken, in fact
it will have begun to harden; or it may be the middle part does _not_
flush as quickly as the others; in this case it will be picked before
it is ready, that is, when the flush is too young, and the yield will
consequently be smaller.

I believe the yield of a plantation may be largely increased by
attending to this. Every Tea estate should be divided into gardens of,
say, about five to ten acres each.[15] If no natural division exists,
small roads to act as such should be made. More than this cannot be
done when the plantation is first laid out, but when later the plants
yield, any difference between the productive powers of different
parts of the same garden should be noted, and these divided off into
sections. To do this latter with roads would take up too much space,
and small masonry pillars, whitewashed, are the best. Four of these,
one at each corner of a section, are enough, and they need not be
more than 3 feet high and 1 foot square. Thus each garden may, where
necessary, be divided into two sections, which, in a 300-acre estate,
partitioned off into thirty gardens, would give about forty to sixty
sections. No matter where a section may be, directly the flush on it
is ready it should be picked. Where the soil on any one garden is much
the same, and observation shows the plants all over it flush equally,
it may be left all in one. I only lay down the principle, and I am
very certain it works well, the proof of which is that, where I have
practised it, some sections during the season give three, four, and
five flushes more than others. Had the usual plan of picking from one
end to the other been adopted, they would have been all _forced_ to
give the same number; in other words, the said extra flushes would have
been lost, and further loss occasioned by some flushes being taken
before they were ready, others after a portion of the tender leaf had
hardened.

The best plan is simply to number the gardens from 1 upwards, and the
sections in each garden the same way. Thus supposing No. 5 garden is
divided into three sections, they will be known respectively as 5-1,
5-2, and 5-3. This is the best way for the natives, and I find they
soon learn to designate each section. I have a man whose special duty
(though he has other work also) it is to see each day which sections
are ready to pick the following, and those, and those alone, are
picked. Practice soon teaches the number of pickers required for any
given number of sections, and that number only are put to the work.
If a portion is not completed that day, it is the first taken in hand
the next, and if any day on no sections is the flush ready, no leaf is
picked the following.

Apart from leaf-picking, the garden and section plan detailed is useful
in many ways. Each garden, if not each section which most requires
it, is dug, pruned, or manured at the best time, and any spot on the
plantation is easily designated. The plan facilitates the measurement
of work, and enables correct lists of the flushes gathered to be kept.
It is thus seen which gardens yield best, and the worst can, by extra
manuring, be brought to equal those.

In short, the advantages are many, too numerous to detail.

Of course all this can be better done on a flat garden than on one
planted on slopes, and though it may not be possible to work it out
as much in detail on the latter, still a good deal in that way can be
done, and I strongly recommend it.

In laying out a plantation keep it all as much together as possible,
the more it is in one block the easier it is supervised, the cheaper
it is worked. Still do not, with a view to this, take in any bad land,
for bad land will never pay.

Let your lines of Tea plants, as far as practicable, run with
geometrical regularity. You will later find, both in measuring work
and picking leaf, great advantages therefrom. In gardens where the
lines are not regular portions are continually being passed over in
leaf-picking, and thereby not only is the present flush from such parts
lost, but the following is also retarded.

If your different gardens are so situated that the roads through them,
that is from one garden to the other, can be along _the side_ of any
garden without increasing the length of the road, by all means adopt
that route. There is no such good boundary for a garden as a road that
is being continually traversed. It will save many rupees by preventing
the encroachment of jungle into a garden, and more space is thus also
given for plants. It is, however, of no use to do it if a road through
the middle of a garden is shorter, as coolies _will_ always take the
shortest route.

The lines of plants on sloping ground should neither run up and down,
nor directly across the slope. If they run up and down, gutters or
water-courses will form between the lines, and much additional earth
will be washed away thereby. If they run right across the hill the same
thing will occur _between the trees in each line_, and the lower side
of each plant will have its roots laid very bare. It is on all slopes
a choice of evils, but if the lines are laid diagonally across the
hill, so that the slope _along the lines_ shall be a moderate one, the
evil is reduced as far as it can be by any arrangement of the plants.
No, I forgot; there is one other thing. The closer the lines to each
other, and the closer the plants in the lines to each other, in short,
the more thickly the ground on slopes is planted the less will be the
wash, for stems and roots retain the soil in its place, and the more
there are the greater the advantage.

Where slopes are steep (though, remember, steep slopes are to be
avoided) terracing may be resorted to with advantage, as the washing
down of the soil is much checked by it.

On flat land, of course, it does not really signify in which directions
the lines run, but such a garden looks best if, when the roads are
straight, the lines run at right angles to them.

In laying out a garden, choose a central spot with water handy for
your factory, bungalow, and all your buildings; let your Tea-houses
be as close to your dwelling-house as possible, so that during the
manufacturing time you can be in and out at all hours of the day
and night. Much of your success will depend upon this. Let all your
buildings be as near to each other as they can, but still far enough
apart, that any one building may burn without endangering others. You
need not construct any Tea-buildings until the third year.

FOOTNOTE:

[15] A garden I have just finished in the Western Dooars is 300 acres
in extent, all on flat land without any breaks in the cultivation, and
all divided into sections of 5 acres each. Being in one large block it
is not divided into gardens at all, only sections.



CHAPTER IX.

VARIETIES OF THE TEA PLANT.


These are many, but they all arise from two species: the China
plant, the common Tea-bush in China; and the indigenous plant, first
discovered some forty years ago in Assam.

These are quite different species of the same plant. Whether the
difference was produced by climate, by soil, or in what way, no one
knows, and here we have only to do with the facts that they _do_ differ
in every respect. A purely indigenous plant or tree (for in its wild
state it may more properly be called the latter) grows with one stem
or trunk, and runs up to 15 and 18 feet high. It is always found in
thick jungle, and would thus appear to like shade. I believe it does
when young; but I am quite sure, if the jungle were cleared round an
indigenous Tea-tree found in the forest, it would thrive better from
that day. The China bush (for it is never more) after the second year
has numerous stems, and 6 or 7 feet would seem to be its limit in
height. The lowest branches of a China plant are close to the ground,
but in a pure cultivated indigenous, from 9 inches to 1 foot above the
soil the single stem is clean.

The indigenous grows quicker after the second or third year than the
China, if it has not been over-pruned or over-plucked when young. In
other words, it flushes quicker, for flushing is growing.

The indigenous does not run so much to wood as the China. Indigenous
seedlings require to be watered oftener than China, for the latter do
not suffer as quickly from drought. The indigenous tree has a leaf
of 9 inches long and more. The leaf of the China bush never exceeds 4
inches. The indigenous leaf is a bright pale green, the China leaf a
dull dark green colour. The indigenous “flushes,” that is, produces new
tender leaf, much more copiously than the China, and this in two ways:
_first_, the leaves are larger, and thus if only even in number exceed
in bulk what the China has given; and _secondly_, it flushes oftener.
The infusion of Tea made from the indigenous species is far more
“rasping” and “pungent” than what the China plant can give, and the
Tea commands a much higher price. The young leaves, from which alone
Tea is made, are of a much finer and softer texture in the indigenous
than in the China; the former may be compared to satin, the latter to
leather. The young leaves of the indigenous, moreover, do not harden so
quickly as those of the China; thus, if there is any unavoidable delay
in picking a flush, the loss is less with the former. In the fact that
unpruned or unpicked plants (for picking is a miniature pruning) give
fewer and less succulent young leaves which harden quicker than pruned
ones, the two varieties would seem to be alike. The China variety is
much more prolific of seed than the indigenous; the former also gives
it when younger, and as seed checks leaf, the China is inferior in this
as in other respects. The China is by far the hardier plant; it is much
easier to rear, and it will grow in widely differing climates, which
the indigenous will not.

A patch of indigenous with a mature flush on it is a pretty sight. The
plants all appear as if crowned with gold (they are truly so if other
advantages exist), and are a great contrast to the China variety if it
can also be seen near.

I have now, I think, pointed out the leading characteristics of the
two original varieties of the Tea plant, and it stands to reason no
one would grow the China who could get indigenous. But the truth is,
a pure specimen of either is rare. The plants between indigenous and
China are called “hybrids.” They were in the first instance produced by
the inoculation, when close together, of the pollen of one kind into
the flower of the other, and the result was a true hybrid, partaking
equally of the indigenous and China characteristics; but the process
was repeated again and again between the said hybrid and an indigenous
or China, and again later between hybrids of different degrees, so
that now there are very many varieties of the Tea plant--100 or even
more--and no garden is wholly indigenous or wholly China. So close do
the varieties run, no one can draw the line and say where the China
becomes a hybrid, the hybrid an indigenous. Though as a rule the young
leaves are light green or dark green, as the plant approaches the
indigenous or China in its character, there are a certain class of
bushes all hybrid, whose young leaves have strong shades of crimson and
purple. Some even are quite red, others quite purple. These colours do
not last as the leaf hardens, and the matured leaves of these plants do
not differ from others. Plants with these coloured leaves are prolific.

The nearer each plant approaches the indigenous, the higher its class
and excellence, _ergo_ one plantation is composed of a much better
class of plants than another. Had China seed never been introduced into
India, a very different state of things would have existed now. The
cultivation would not have been so large, but far more valuable. The
propagation and rearing of the indigenous, as observed, is difficult;
the China is much hardier while young. So difficult is it to rear
successively the _pure_ indigenous, perhaps the best plan, were it
all to come over again, would be to propagate a high-class hybrid and
distribute it, never allowing any China seed or plants to leave the
nursery, which should have been a Government one. But we must take
things as they are. The Government nurseries in the Himalayas and the
Dehra Dhoon (there have never been any elsewhere, and worse sites could
not have been chosen) were planted entirely with China seeds, the
seedlings distributed all over the country, and thus the mischief was
done. The Indian Tea is vastly superior to the Chinese, and commands
a much higher price at home, but it is still very inferior to what it
would have been, had not Chinese seed been so recklessly imported and
distributed over the country.

The home of the indigenous Tea tree is in the deep luxurious jungles
of Assam and Cachar.[16] There it grows into a good-sized tree. I have
seen it 20 feet high. These are of no use, except for seed, until
they are cut down. When this is done, they throw out many new shoots,
covered with young tender leaves, fit for Tea. They are, of course,
far too big to transplant, but on some sites where they were numerous,
that spot was chosen for the plantation, and some of these are the best
gardens in Assam and Cachar.

The indigenous plant and high-class hybrid require a hot moist climate,
and will not therefore flourish in any parts of India outside Eastern
Bengal. I have tried them in the Himalayas, there the cold kills them.
In Dehra Dhoon and Kangra the climate is far too dry; besides, the hot
winds in the former, and the cold in the latter, are prejudicial. The
Terai under Darjeeling suits them. In Assam, Cachar, and Chittagong,
the indigenous and the highest class hybrids will thrive, for the
climate of all three is suitable, but perhaps Northern Assam possesses
the best climate of all for such plants.

The Himalayan gardens consist entirely of Chinese plants mixed
occasionally with a low class of hybrid. They were all formed from the
Government Nurseries where nothing but Chinese was reared. Occasional
importations of Assam and Cachar seed will account for the sprinkling
of low class hybrids which may be found. The same may be said of Dehra
Dhoon and Kangra. In some gardens in the Terai below Darjeeling a high
class of plant exists. In Assam, Cachar, and Chittagong the plantations
vary much, but all have some indigenous and high class hybrids, while
many gardens are composed of nothing else.

It is evident, then, that the value of a garden depends much on the
class of its plants, and that a wise man will only propagate the best.
Only the seed from good varieties should be selected, and gradually
all inferior bushes should be rooted out and a good kind substituted.
When this shall have been systematically done for a few years on a good
garden, which has other advantages, the yield per acre will far exceed
anything yet realised or even thought of.

Government action in the matter of Tea has been prejudicial in many
ways, but in none more so than when they were doing their best to
foster the cultivation by distributing Chinese seed and seedlings
gratis. No one can blame here (would the Government were equally free
from blame in all Tea matters!), but the mischief is none the less. It
will never be possible to undo the harm then done.

The seed of indigenous, hybrid, and Chinese is like in appearance,
and cannot be distinguished. Thus, when seed formerly was got from a
distance, the purchaser was at the mercy of the vendor.

High cultivation improves the class of a Tea plant. Thus, a purely
China bush, if highly cultivated and well manured, will in two or
three years assume a hybrid character. High cultivation will therefore
improve the class of _all_ the plants in a garden; but the cheapest
and best plan with low class Chinese plants is to root them out and
replace them with others, as will be explained hereafter. Low class
seedlings should also be rooted out of nurseries.

I cannot conclude this chapter better than by giving an extract from
the “Government Records” alluded to in a previous chapter, and I add a
few remarks at foot, as otherwise the reader might be puzzled with some
opinions expressed which are so much at variance with the generally
received opinions on Tea to-day.

[Sidenote: Several varieties.]

  _Kinds of Tea Plants cultivated._--“When Government resolved on
  trying the experiment of cultivating Tea in India, they deputed Dr.
  Gordon to China to acquire information respecting the cultivation
  and manufacture of Teas, and to procure Tea seeds. Aided by Dr.
  Gutzlaff he procured a quantity of seeds from the mountains in the
  Amoy districts. These seeds were sent to the Calcutta Botanical
  Garden, where they were sown in boxes. On germinating they were sent
  up the country in boats, some to Assam and some to Gurhmuktesur,
  and from thence to Kumaon and Gurhwal. From these plants date the
  commencement of the Tea plantations in the Himalayas.[17] Tea was
  first made in Kumaon in 1841, and the samples sent to England, and
  were pronounced to be of good quality, fitted for the home markets,
  and similar to the Oolong Souchong varieties. Thus Messrs. Thompson,
  of Mincing Lane, report on a sample sent by us to Dr. Royle in 1842:
  ‘The samples of Tea received belong to the Oolong Souchong kind,
  fine-flavoured and strong. This is equal to the superior black Tea
  generally sent as presents, and better for the most part than the
  Chinese Tea imported for mercantile purposes.’[18] By many it was
  supposed that there were different species of the Tea plant, and that
  the species cultivated in the south districts of China was different
  from that met with in the north. To solve this mystery, and at the
  same time procure the best varieties of the Tea plant, Mr. Fortune
  was deputed to China. By him large numbers of Tea plants were sent
  from different districts of China celebrated for their Teas, and
  are now thriving luxuriantly in all the plantations throughout the
  Kohistan of the North-west Provinces and Punjab. Both green and
  black Tea plants were sent, the former from Whey Chow, Mooyeen,
  Chusan, Silver Island, and Tein Tang, near Ningpo, and the latter
  from Woo-e San, Tein San, and Tsin Gan, in the Woo-e district. But
  so similar are the green and black Tea plants to each other, and the
  plants from the Amoy districts, that the most practised eye, when
  they are mixed together, cannot separate them, showing that they
  are nothing more than mere varieties of one and the same plant, the
  changes in the form of the leaf being brought about by cultivation.
  Moreover, throughout the plantation fifty varieties might easily
  be pointed out; but they run so into each other as to render it
  impossible to assign them any trivial character, and the produce of
  the seed of different varieties does not produce the same varieties
  only, but several varieties, proving that the changes are entirely
  owing to cultivation; nor do the plants, cultivated at 6,000 feet
  in the Himalayas, differ in the least in their varieties from those
  cultivated at 2,500 feet of altitude in the Dehra Dhoon.

  [Sidenote: Assam species.]

  “That the Assam plant is a marked species is true, it being
  distinguished by its large membranous and lanceolate leaf, small
  flower, and upright growth.

  “It is a very inferior plant for making Tea, and its leaves are
  therefore not used.[19] Though the plants received from the different
  districts of China do not differ from those first sent to the
  plantations, it is highly important to know that the Tea plants
  from well-known green and black Tea districts of China now exist in
  the plantations, as it is stated that local causes exert a great
  influence in the quality of the Teas as much as the manufacture
  does. The expense, therefore, incurred in stocking the Government
  plantations with the finest kinds and varieties of Tea plants
  procurable in China, though great, will be amply repaid. From them
  superior kinds of Tea are produced.”

The above extract is a sample of the said “Records.” They abound in
errors and highly coloured statements, which induced many to embark
in Tea on unfavourable sites, and “the red book” (it is bound in a
red cover) is not exactly blessed by the majority of the Himalayan
planters!

FOOTNOTES:

[16] It is a singular fact that none exists in Northern Cachar, that
is, on the northern side of the river.

[17] And also the introduction of a bad class of plants.--E. M.

[18] A single small sample of Tea very carefully made, and with an
amount of labour which could never be bestowed on the mass, is little
or no criterion. Tea is better made in Kumaon in 1878 than it was in
1842, but Kumaon Tea does not vie in price with Eastern Bengal produce.
All the Himalayan Tea is weak, though of a delicate flavour; all Tea
grown at high elevation _must_ be so.--E. M.

[19] A little enquiry would have shown this was not true, even when it
was written. All Tea planters, brokers, and all interested in Tea, know
now (many knew it then) that the “Assam species,” viz., the indigenous,
makes the most valuable Tea produced.--E. M.



CHAPTER X.

TEA SEED.


Though there is a great difference in Tea plants (see last chapter) the
seed of all is the same, and it is therefore impossible to say from
what class of plants it has been gathered.

When Tea seed was very valuable (it has sold in the Tea-fever days as
high as Rs. 200 and Rs. 300 per maund) it was the object of planters to
grow as much as possible.

High class plants do not give much seed, a plantation therefore with
much on it should be avoided in purchasing seed.

The Tea flower (the germ of next year’s seed) appears in the autumn,
and the seed is ripe at the end of the following October or early
November.

It takes thus one year to form.

Seed is ripe when the capsule becomes brown, and when breaking the
latter the inner brown covering of the seed adheres to the seed and
_not_ to the capsule.

One capsule contains 1, 2, 3, and sometimes even 4 seeds.

Though the mass ripens at the end of October, some ripen earlier; the
capsule splits and the seed falls on the ground. If, therefore, all
the seed from a garden is required, it is well to send round boys all
October to pick up such seeds.

When the seed is picked at the end of October or early November the
mass is still in capsules. It should be laid in the sun for half an
hour daily for two or three days until most of the capsules have
split. It is then shelled, and the clean seed laid on the floor of
any building where it will remain dry. Sunning it _after_ shelling is
objectionable.

The sooner it is sown after it is shelled the better.

If for any reason it is necessary to keep it, say a fortnight or three
weeks before sowing, it is best kept _towards_ germinating in layers
covered with dry mould. But if to be kept longer leave it on the dry
floor as above, taking care it is thinly spread (not more than one seed
thick if you have space) and collected together, and re-spread every
day to turn it.

For transport to a distance it should be placed in coarse gunny bags
only one-third filled. If these are shaken and turned daily during
transit a journey of a week will not very materially injure the seed.

For any long journey it is best placed in layers in boxes with
thoroughly dry and fine charcoal between the layers, and sheets of
paper here and there to prevent the charcoal running to the bottom.

It is scarcely necessary to consider how Tea seed can be utilized when
not saleable, for seed prevents leaf, and therefore it should not be
grown if there is no market for it. It will, however, make oil, but
the price it would fetch for this purpose would not compensate for
the diminished yield of leaf it had caused. It is also valuable as
manure mixed with cattle-dung, but it would not pay to grow it for this
purpose either.

My advice therefore is to allow no more seed on the garden than you
require for your own use (even the fullest gardens require some yearly)
or than you can sell at a remunerative price.

If the object is to produce a considerable quantity of seed, set apart
a piece of the plantation for it, and do not prune it at all. A large
number will then be produced on that piece.

If the object is to grow as little seed as possible after the pruning
in the cold weather, which destroys the greater part, send round boys
to pick off such of the germs as remain.

If this is done ever so carefully, some will escape, enough say to give
one maund seed from 10 acres of garden, and this as a rule is enough to
fill up vacancies in a good garden.

The following figures regarding seed will be found useful, but remember
the higher the class of plant the less durable the seed:--

Seven maunds seed, with capsules, give 4 mds. clean seed.

  One maund clean seed (fresh)           =26,000 seeds.
        „      „         (ten days old)  =32,000   „
        „      „         (one month old) =35,000   „

Say therefore, in round numbers, that one maund Tea seed=30,000 seeds.

With good Tea seed, sown shortly after it is picked, about 20,000 will
germinate.

If you get 8,000 to germinate with seed that has come a long distance,
you are lucky.

After a two months’ journey 3,000 is probably the outside which will be
realised.

My experience, with seed imported into another District from Assam or
Cachar, is that more than 4,500 Seedlings cannot be expected from each
maund.



CHAPTER XI.

COMPARISON BETWEEN SOWING IN NURSERIES AND IN SITU.


In the one case the seed is placed in nurseries at the close of the
year, and the young plants transplanted into the garden at beginning of
the following rains.

In the other the seed is (at the same time, viz., close of the year, if
you can get it so soon) sown at once in the plantation where the plants
are intended to grow.

Each of these plans has its advocates, who don’t believe in the other
plan at all! The question is which is the better?

Their respective advantages may be shortly summed up as follows:--


NURSERIES.

_Advantages._--The seed may be made to germinate early by watering.
After it germinates the plants can be watered from time to time as they
require it. Artificial shade (a great help to the germination of Tea
seed) can be given. The soil can be frequently opened, and the plants
in every way better tended in nurseries.

_Disadvantages._--The plants lose at least three months’ growth when
transplanted, and may die. The transplanting necessitates labour at
the time of the year it is much wanted for other work. The expense is
greater than the other plan, for there are the nurseries to make and
the labour of transplanting.


IN SITU.

_Advantages._--The plants gain some three months in growth by not being
moved. It saves labour at the busy time, viz., early in the rains.
It saves all the labour of transplanting, that is, it _saves_ labour
absolutely, and _gives_ labour when, as stated, it is much required.

_Disadvantages._--If the early rains (that is, rain in December,
January, and February) fail, but few seeds germinate. In the case of
a new garden, the soil must be kept clean six or seven months before
it would be necessary by the nursery plan. No artificial shade can be
given.

It will thus be seen that the advocates of both plans have much to urge
in their respective favours. Which is better?

The advocates of each plan are guided by the climate they have planted
Tea in, and the truth is simply that the better plan for one place is
not adapted to another. Planting _in situ_ where it will succeed is
by far the cheaper and better, and it will do so wherever there are
certainly cold weather and spring rains. Thus (see rain table) it will
often succeed in Assam, Cachar, Darjeeling, the Western Dooars, and
perhaps the Terai below Darjeeling. It will fail in Chittagong, Dehra
Dhoon, Kumaon, Kangra, and Hazareebaugh. In Chittagong, for instance,
a garden could never be made by planting _in situ_, or, as it is
generally called, at stake.[20]

In this and other matters adapt your operations to the existing climate.

I will now describe the above two methods of sowing seed.

FOOTNOTE:

[20] In no climate is the success of it certain, for early rains often
fail, and then it is all loss. I would, therefore, in all cases advise
nurseries in reserve.



CHAPTER XII.

SOWING SEED IN SITU, ID EST, AT STAKE.


It is named “at stake” because stakes are put along in lines to show
where the Tea trees are to be, and the seed is sown at those spots.

The _modus operandi_ is very simple. A month before the sowing time
(which should be as soon as you can get the seed), at each stake dig a
hole at least 9 inches diameter and 12 inches deep, put the soil taken
out on the sides, taking care, however, if it be on a slope, to put
none _above_ the hole. Do not put the soil near enough to the pit to
make it likely it will be washed back. Such soil as should be washed in
ought to be the new rich surface soil. For this reason the upper side
of the hole should be left free on slopes. The pits are made a month
beforehand to admit of this, and to allow the action of the air on the
open sides to improve the mould.

If lucky enough to have one or two falls of rain during the month, the
holes will be more or less filled up with soil eminently calculated
to instigate rapid growth. Just before sowing fill up the pit with
surrounding _surface_ soil. Whether to mix a little manure with it
or not is a question. If it is virgin soil, and rich in decayed
vegetation, I say no; if not virgin soil, and rather poor, yes; but it
must be strictly in moderation--not more, say, than a man can hold in
both hands to each hole. In filling up the hole, press the soil down
lightly two or three times, or it will all sink later, and your seeds
be far too deep.

When the above is all done, there is a perfect spot for the reception
of the seed. The tap-root can readily descend in search of moisture,
and the lateral rootlets can spread likewise. They (the latter) will
not reach the outer walls of the pit for six months, and will then be
strong enough to force their way through.

Now sow the seed. Put in, say, two or three, as the seed is good or
bad, six inches apart; push them into the soft soil one inch, and put
up the stake in the centre to mark the spot.

Keep the place clean till following rains, but allow only hand-weeding
near the young seedlings, and occasionally open the soil with some
light hand-instrument, as a “koorpee,” to the depth of half-an-inch.

If all the seeds germinate, and the seedlings escape crickets, and all
live, at commencement of the rains leave the best and transplant the
others to any vacant spot. You will succeed with some, not with others;
but do not be too anxious to take up the spare ones with earth round
the roots, and thus endanger the one plant left. That the seedling left
be not injured is the _great_ point, the others must take their chance.

Some people believe in two, or even three seedlings together, and would
thus advise them to be all, or perhaps two, left. I do not approve of
the plan, except, perhaps, with Chinese plants. Plant as close as you
will in the lines, but give each plant its own home.

There is another mode of planting at stake, which is, I think, better
than the above.

Lay the seed in alternate layers of seed and mould in beds. The seeds
may be laid _close_ to each other, but not _above_ each other, with
mould, say, two inches thick, above, and then seed again. When they
begin to burst, ready to shoot out their roots, examine the seeds, by
taking off the soil from each layer, every three or four days. Take
out those that _have_ burst, and plant with the eye or root side of
the seed downwards. Put all that have _not_ burst back again. Repeat
the operation again and again every second or third day. Be careful and
take them up before the root projects--that is, directly the coating
has cracked.

By this means only one seed need be put at each stake, for it is
certain to germinate, and seed may thus be made to go much further.
Great care is, however, necessary in this operation.



CHAPTER XIII.

NURSERIES.


Choose a level site, with, if possible, the command of water
at a higher level--anyhow with water handy. Either irrigating
or hand-watering for seed beds is a necessity if vigorous and
well-developed plants are to be looked for.

The soil should be of the light, friable kind recommended for the
Tea-plant (see “Soil”) and of the same _nature_ as the soil of the
garden, the ultimate home of the plants. This latter is all-important,
for seedlings will never thrive (probably not live) transplanted into
a new kind of mould, particularly a poorer kind.

If possible, the soil of the seed beds should be poorer than the soil
of the garden--on no account richer. Taking care it is of the same
_nature_ as the garden soil, choose the poorest you can find. The
principle is well known in England, and it applies equally to India.
From poor to rich soil plants thrive, but never the other way.

For the above reason, if you manure seed beds, do it very sparingly.

Artificial shade for seed beds is a necessity; at least very many more
seeds will germinate when it is given.

Natural shade over seed beds is _very_ bad; for, _firstly_, “the
drippings” are highly injurious; and, _secondly_, shade is only
required till the plants are two or three inches high; after that _any_
shade is bad, for plants brought up to the time of transplanting in
shade are never very hardy.

Seed beds, where water is handy, should not be dug deep. If so dug,
and the soil is consequently loose a long way down, the tap-root will
descend quickly, and will be too long when transplanted. As water can
be given when it is necessary, there is no need for the tap-root to go
down low in search of moisture.[21] A long tap-root is generally broken
in “lifting” the seedling from the bed.

Seed beds raised, as is the usual custom, above the paths that run
between them, are objectionable. They part with moisture too freely.
They should, on the contrary, be below the level of the paths, and
there is another advantage in this, for the said paths can then be used
partly as supports for the artificial shade, and thus do away with the
expense of long wooden stakes.

As the seed beds are only required until the beginning of the following
rains, there is no possibility of their suffering from excessive
moisture. When they are required to remain later, of course this plan
of making the beds lower than the paths will not do.

Seed is best sown in drills, six inches apart, and each seed two,
or if space can be got, even three inches from its neighbour. This
facilitates each seedling being taken up later, with more or less
of a ball of earth round the roots--an all-important point (see
Transplanting, page 76).

The length of the beds does not signify, but the breadth must not be
more than five feet, so that a man on the path on either side can reach
to the middle while hand-weeding or opening the soil.

After what has been said no lengthy directions for making the beds are
necessary.

Cut down, burn, or carry off all jungle, and then take out all roots,
whether grass or other. Now make the surface level. After this mark
off the beds and paths, the latter one foot broad only, with string
and pegs. Then raise the path six inches above the spots marked off
for the beds. This latter must not be done by earth from the beds, but
by earth from outside the intended nursery. Next dig and pulverise the
soil of the beds to a depth of six or seven inches, no more, and level
the surface.

All is now ready for the seed. A string, five feet long, with a small
peg at either end, is given to two men who stand on the path at either
side of the bed. Each man has a six-inch measure. The string is laid
across the bed, beginning at one end and pegged down on either side.
A drill is then made along the string about one inch deep, and this
done the string is, by means of the six-inch measure on either side,
removed, and pegged down again in the place for the next drill. Seeds
are then sown or placed along the first drill made, two or three inches
apart, and the earth filled in. This is repeated again and again till
the whole bed is sown.

If the character of the seed is doubtful it must be laid in thicker,
but with good seed two-and-a-half to three inches is the best distance.

The sowing finished the artificial shade has to be given. Along the
paths, at five feet apart, put in forked stakes, two feet long--viz.,
six inches into the path, and eighteen inches above it. Connect these
with one another by poles laid in the forks; now lay other (but
thinner) poles attached to the first poles at either end _across_
and above the bed; and again across these latter, that is along _the
length_ of the beds, split bamboos, and then bind the whole framework
here and there. The said framework made will then be two feet above
the beds--viz., eighteen inches of stake support, and the six-inch
raised paths. The eighteen inches of opening all round, under the
frame, that is, between the frame and the path, allows the necessary
air to circulate; while the expense, danger from high winds, and the
objectionable entrance of the sun at the sides, all of which high
artificial shade is subject to, are avoided by this low frame-work.

Mats are the best to cover the frame-work. In case of accidental or
incendiary fire they are not so objectionable as grass, for they burn
less and slower, but mats are expensive. Any coarse grass (freed from
seed) will answer, and it should be laid on as thin as will suffice to
give shade.

The beds may be watered, if there is no rain, a fortnight after the
seed is sown, and from time to time during the dry season, whenever the
soil at a depth of three or four inches shows no moisture.

The soil should also be kept free of weeds, and after the plants are
three or four inches high, the spaces between the drills should be
slightly stirred every now and then.

After the seed has germinated, and the seedlings have, say, four leaves
on them, the artificial shade should be taken away. But it must be done
gradually, taking off portions of the grass first, so that the young
seedlings may by degrees be inured to the hot sun.

Though cultivation, as described, by watering and opening the soil at
times is well, these should not be done much, or the seedlings will
be too large when the time comes to transplant them. Large seedlings
do not, as a rule, thrive as well as moderate-sized ones, after being
transplanted.

Among the many very absurd mistakes made in the cultivation of the
Tea plant, none exceeds the ridiculous way Tea seed used to be sown
in the Government plantations in the North-western Himalayas. The
seed was sown in drills, as I have advised, but in six linear inches
of the drills, where it is right to put two, or at most three, seeds,
perhaps thirty were placed! I do not exaggerate; the drill, six inches
deep, was filled with them. Many and many lacs of seed, in those days
worth many thousand rupees, were thus sacrificed. Private planters in
the Himalayas, taught by the Government method, once did the same. I
believe the absurd practice is exploded now.

Seed cannot be sown too soon after being picked. It is ripe early in
November, so the beds should be all ready by November, and if the seed
has not far to come it can thus be sown early that month.

To each maund there are in round numbers 30,000 seeds, (see page 56).
The number of plants it will take to fill an acre depends, of course,
on the distances they are set apart (see page 72), but having decided
this point, also the area to be planted, and consequently the number of
maunds of seeds to be sown (see page 56), the following table will be
found useful in calculating the size of nursery required:--


  _Table showing the size of nursery required for one maund and ten
  maunds seed, the drills being 6 inches apart, and each seed 3 inches
  or 2 inches from its neighbour._

  -------------+-----------+------------------+---------------+----------------
  Distance each|Area in sq.|Area, in sq. feet,|Area, including|Size of nursery,
  seed is set  |inches each|of beds without   |paths required |including the
  apart in the |seed   will|paths required for|for each md.   |paths to take in
  drill.       |occupy.    |each md.          |               |for 10 mds.
  -------------+-----------+------------------+---------------+----------------
   3 inches    |       18  |        3,763     |4,513 sq. feet |  100 yards
               |           |                  |       or      |      by
               |           |                  | 501 sq. yards.|   50 yards.
               |           |                  |               |
   2 inches    |       12  |        2,500     |2,995 sq. feet |  100 yards
               |           |                  |       or      |      by
               |           |                  | 332 sq. yards.|   33 yards.
  -----------------------------------------------------------------------------

If nurseries for more than ten maunds are required, then allow 100
yards to be the breadth, and for each extra ten maunds add respectively
for 3 or 2 inches (see 1st column) 50 or 33 yards to the length. Thus
fifty maunds will require nurseries 100 yards by 250 yards, or 100
yards by 165 yards, according as it is decided to plant the seed 3
inches or 2 inches apart in the lines.

FOOTNOTE:

[21] In planting “at stake” (see last Chapter) the conditions are
different. There the plant is in its permanent home, and the more
quickly and deeper the tap-root descends the better, as the plant will
then draw moisture from low down when the soil is dry.



CHAPTER XIV.

MANURE.


An idea existed formerly--got, I believe, from stray Chinamen, who I
don’t think knew much about Tea in any way--that manure, though it
increased the yield, spoilt the flavour of Tea. The idea is opposed to
all agricultural knowledge, for high cultivation, which in no case can
be carried out to perfection without manure, much improves the strength
and flavour of all edibles, the product of mother-earth.

My first experience of manure to the Tea plant was obtained in the
Chittagong district from a small garden close to the station, which has
been for some years highly manured. I was struck with the frequency and
abundance of the flushes and the strength and flavour of the Tea. My
high opinion of the Tea was later borne out by the Calcutta brokers. I
allude to the “Pioneer” garden, close to the Chittagong station. During
the best Tea months flush succeeded flush at intervals of less than a
week, while eight to ten maunds (640 to 800 lbs.) was the yearly yield
per acre! The soil was very sandy and poor.

After-experience showed me that manuring nearly doubles the yield of
plants, and that so far from injuring the flavour of Tea it improves
it, while it adds greatly to the strength.

I shall therefore beg the question that manure _is_ an advantage. If
any planter doubts, let him try it, and his doubts will soon be solved.

Any manure is better than none, but I believe one of the best manures
for the Tea plant (always excepting night-soil and the excrements of
birds, which cannot be procured) is cattle manure. It is not heating,
like horse-dung, and may be applied in large quantities without any
risk. The fresher it is applied, in my opinion, the better, for it has
then far more power. If mixed with any vegetable refuse, the bulk being
increased, it will go further, but I do not think it is intrinsically
any the better for it.

There are several chemical manures advertised for Tea plants. “Money
and Ponder’s Chemical Manure,” lately patented by Mr. Ponder and
myself, is said to have been very successful on several gardens. It is
manufactured by Mr. J. Thompson, Kooshtea, Bengal, who will supply all
details.

All garden refuse should be regarded as manure and buried between the
plants. I allude to the prunings of the bushes and the weeds at all
times from the land. To carry these off the ground, as I have sometimes
seen done, is simply taking off so much strength from the soil. The
greener, too, all this is buried the better.

When it is considered how much is taken from the Tea plant, it is
evident the soil will be exhausted, sooner or later, if no means are
adopted to repair the waste. Where manure cannot be got the waste must
be made up, as far as possible, by returning all other growth to the
soil. But manure _should_ be got if possible, for it will double the
yield of a garden; and highly concentrated chemical manures will, I am
sure, be eventually much used on Tea gardens.

The best way to apply it, if enough manure is procurable, is round each
plant; not close to the stem (the rootlets by which the plant feeds
are not there) but about 1 foot from it. Dig a round trench with a
_kodalee_, about 9 inches wide and 6 inches deep, at the above distance
from the stem, lay in the manure, and replace the soil at top. If
the plants are young the trench should be narrower, shallower, and 6
inches, instead of 1 foot, from the stems.

If enough manure is not procurable for this (the best) plan, the most
must be done with what can be got, as follows:--If the plants are full
grown, and there is say 4 feet between the lines, dig a trench down the
centre and lay in the manure. The plants will then be manured on two
sides. If the plants are young lay the manure _near_ them on two sides,
if possible, but failing that even on one side. The principle is to lay
the manure at the distance the feeding rootlets are, and the older the
plant the greater distance these are from its stem.

As to the quantity of cattle manure. Say for plants four years old and
upwards (if younger, less will be an equivalent) one maund to 20 trees
is a moderate dose, one maund to 15 trees a good dose, and one maund to
10 trees highly liberal manuring, and as much as the plants can take up.

Say in round numbers each acre contains 2,500 plants (4 by 4--a usual
distance--gives 2,722 plants, as shown at page 72), and say the manure
is procurable at three annas a maund.[22]

The following table shows the expense of each degree of manuring, viz.,
10, 15, and 20 trees per maund:--


_Table showing the possible cost and result of manuring with cattle
manure._

  Key:

  A: Maunds of manure per acre at 2,500 plants per acre
  B: Cost of manure at 3 annas per maund _N.B._--Ans. omitted
  C: Probable extra yield of Tea per acre
  D: Value of extra yield of Tea at Rs. 50 per maund
  E: Profit by manuring per acre
  F: Deducting the probable cost of putting in the manure, the following
        profit is shown per acre

  --------------------+------+-----+-------+------+-----+----
                      |      |     |       |      |     |
  Rate of Manuring    |      |     |       |      |     |
                      | A    |  B  |  C    | D    |  E  |  F
                      |      |     |       |      |     |
                      |      |     |       |      |     |
  --------------------+------+-----+-------+------+-----+----
                      | Mds. | Rs. |  Mds. |  Rs. | Rs. | Rs.
  One md. to 10 plants| 250  | 47  | 2-1/2 | 125  | 78  | 70
  One md. to 15 plants| 166  | 31  |   2   | 100  | 69  | 62
  One md. to 20 plants| 125  | 23  | 1-1/2 |  75  | 52  | 46
  --------------------+------+-----+-------+------+-----+----

It is not too much to calculate that this will add respectively 1-1/2,
2, and 2-1/2 maunds of Tea per acre to the yield, and I have carried
this out in the table and shown the results.

I quite believe the results shown will be obtained by manuring, and I
base my opinion on practice not theory.

_N.B._--I have deducted Rs. 8 for the first, Rs. 7 for the second, and
Rs. 6 for the third, as the probable cost of putting in the manure, as
it may have to be carried from the factory to the garden. If purchased
after being placed between the lines (and if manure is bought of
adjacent villagers they will so place it), the cost would be less.

The above table, of course, only applies to localities where cattle
manure can be purchased at 3 annas per maund, including carriage to the
factory.

The value of the extra yield of Tea is estimated at only Rs. 50 per
maund in the above table, because the leaf which will give one maund of
Tea is worth no more, as follows:--

                                                  Rs.   A.    P.
  Probable price obtainable for one maund or
  80 lbs. Tea in Calcutta, at 14 annas a lb.
  all round (a fair calculation, one year with
  the other, if it is well manufactured).         70     0     0

  Deduct cost, manufacture, packing, transport,
  and broker’s charges as set out in the
  chapter on “cost manufacture,” page 162         16     9     0
                                                  --------------
  Value of leaf which will make one maund Tea     53     7     0
                                                  ==============

But I prefer estimating it at Rs. 50 only, to be on the safe side.

FOOTNOTE:

[22] It is brought and placed between the lines, in one garden in the
Chittagong district, for one to two annas a maund!



CHAPTER XV.

DISTANCES APART TO PLANT TEA BUSHES.


When the idea existed, which it did once, that ploughs could be used to
cultivate a garden between the lines, these latter, with this object,
were placed unnecessarily wide apart.

All distances may be seen in different gardens, viz., 6 × 6, 6 × 3, 6
× 4, 5 × 4, 5 × 5, 4 × 3, &c., &c.

The plough idea has nowhere been found to answer, and is exploded.[23]
Still, even for hand labour to cultivate, and for facilities in picking
leaf, it is necessary there should be room enough one way to pass
along. Cultivation here means digging, and space enough for this must
be left between the lines. Giving so much, what is then the principle
that should guide us? Clearly, with a view to the largest yield
obtainable, to place as many plants on the land as it will bear.

Four or 4-1/2 feet are, I think, the best distances between the lines.

They give space enough for air to cultivate, and to pass along, even
when the trees are full grown.

Where manure is obtainable and the soil can be kept up to a rich state
by yearly applications, a garden can scarcely be planted too close.

I see no objection to trees touching each other in the lines.

On considerable slopes, to prevent the wash of soil, the plants should
be placed as close as possible, say 3-1/2 feet between and 2 feet in
the lines.

A closely planted garden will grow less weeds than a widely planted
one, and will consequently be cheaper to work.

As the expenditure on a garden is in direct proportion to the area,
and the yield in direct proportion to the number of plants (always
supposing there is power enough in the soil to support them), it
follows that a closely planted garden _must_ be very much more
profitable than the reverse.

Hybrid plants grow to a larger size than Chinese, and should therefore
have more room.

The following is a useful table:--


_Table showing the Plants to an Acre, and the Area one lakh of
seedlings will cover, at the distances named._

  -------------+---------+----------+---------------+------------------
               | Square  |          |The area in    |
  Distances in | ft.     |Plants in |acres one lakh |   REMARKS
      feet     |to each  | one acre |of seedlings   |
               | plant   |          |will cover     |
  -------------+---------+----------+---------------+------------------
  6    by 6    |  36     |  1,210   |   82-1/2}     |
  6     „ 5    |  30     |  1,452   |   69    }     |
  6-1/2 „ 4    |  26     |  1,675   |   59-3/4}     | Too wide for any
  5     „ 5    |  25     |  1,742   |   57-1/2}     |      plants.
  6     „ 4    |  24     |  1,815   |   55    }     |
               |         |          |               |
  6     „ 3-1/2|  21     |  2,074   |   48    }     | For Hybrids, but
  5     „ 4    |  20     |  2,178   |   45-1/2}     | still I think too
               |         |          |               |      wide.
               |         |          |               |
  6     „  3   |  18     |  2,420   |   41-1/4}     |  Good distances
  4     „  4   |  16     |  2,722   |   36-3/4}     |    for Hybrids.
  5     „  3   |  15     |  2,904   |   34-1/2}     |
               |         |          |               |
  4     „  3   |  12     |  3,630   |   27-1/2}     | Chinese for early
               |         |          |               |      return.
  3-1/2 „ 3-1/2|  12-1/4 |  3,555   |   28    }     |
  3-1/2 „ 3    |  10-1/2 |  4,148   |   24    }     |      Chinese.
               |         |          |               |
  6     „ 3-1/4|  19-1/2 |  2,233   |   44-3/4}     |      Hybrid.
               |         |          |               |
  5     „ 3-1/4|  16-1/4 |  2,726   |   36-3/4}     |     Chinese.
  5     „ 3-1/2|  17-1/2 |  2,489   |   40    }     |
               |         |          |               |
  3-1/2 „ 2    |   7     |  6,223   |   16          | Best distance for
               |         |          |               | Chinese on steep
               |         |          |               |       slopes.
  -------------+---------+----------+---------------+------------------

    On flat lands I advise--

                  Hybrid, if high-class  4 × 3-1/2 or 4-1/2 × 4
                  Chinese                            3 × 3

    All the following equal one acre:--

                       4 roods. 4,840 square yards.
                     160 poles. 43,560  „    feet.

FOOTNOTE:

[23] Land _before_ it is planted can be cultivated with ploughs. My
manager is using them largely in the Western Dooars, the land being
there all flat. He uses English ploughs, bought of Ransomes and Sims,
Ipswich, with bullocks, and often an elephant. He finds the latter the
best. After ploughing he uses English harrows.



CHAPTER XVI.

MAKING A GARDEN.


I have not very much to say on this head, as most of the operations
entailed are treated separately. Still a few directions on primary
matters are required.

Having selected a site and made arrangements for the Tea seed required
for the first year’s planting, you should commence operations early in
October, either by constructing the nursery, or clearing land on the
proposed site of the garden, as you may decide which mode of planting,
viz., “nurseries,” or “sowing at stake,” to adopt.[24]

If the latter, you should begin to cut the jungle somewhat earlier, but
it is no use beginning to do this before the middle of September in any
case, for before that the jungle would spring up again so soon that it
would be labour lost.

Before you do _anything_ decide how much you will cultivate the first
year, and make your arrangements for seed accordingly. Here let me
advise you in no case to attempt more than 100 acres. If you do 100
really well the first year you will have done _very_ well. Remember you
have also buildings (though few) to construct, and trying to do too
much you may simply fail in all.

Previous to October you should have made yourself thoroughly acquainted
with all your land, so that you can then fix with knowledge on the best
sites for your buildings, nursery, and Tea plantation.

You will find much on these matters in other chapters which should be
read carefully.

These respective sites having been fixed upon, and supposing you are
going to plant in both ways, from nurseries and _in situ_, construct
the nurseries as advised under that head, page 62, and also cut the
jungle on the intended garden site.

There is not much to say about cutting jungle. Cut all the brushwood
first near the ground, and the big trees later, so that when they fall
they may lie on the underwood. In the portion you intend to plant at
stake you will not have time to cut down the big trees, and had better
simply “ring” them. If this is properly done, that is, if the ring
is broad enough and deep enough (less than one foot broad and five
inches deep for large trees is not safe), they will certainly die in
a twelvemonth, and will not give objectionable shade more than half
that time. In the part to be planted “at stake” you must burn all the
cut jungle by the end of October, and it will be well, if you have
labour enough, to send men up the big trees to cut off the branches
beforehand, so that they will more or less burn with the rest. Doing
this, and piling up the underwood to be burnt round the base of the big
trees, will cause earlier death, and diminish the objectionable shade.

Having burnt the jungle, that is, as much as will burn, and carried off
the rest from the parts to be planted at stake, dig out all the small
roots, and that done, dig the whole some 4 or 5 inches deep. Then stake
it off with small bamboo stakes 18 inches long, showing where the Tea
trees are to be (see page 72 as to the best distances), and then make
your holes and plant your seed at each stake as directed at page 59.

See the way it is recommended to stake land as regards its lay at pages
45 and 46.

You will probably not have the ground ready before the end of November
(do not attempt more than you can do to that date), and then take care
and keep the seed, as directed at page 55, until it is sown.

For the part to be planted from nurseries the following June you have
plenty of time. Nowhere have I, or anyone, seen large vigorous Tea
plants under trees. It is therefore evident trees are hurtful, and no
more should be left in a garden than are required for the labourers to
sit under occasionally, and to collect leaf under before it is taken
to the Tea-house. The trees that are left should be those on the sides
of roads. One to every two or three acres is ample. After therefore
cutting down all the low jungle, cut down all but the said few trees
(it is cheaper in the end than ringing them), and then cut off and cut
up all the branches into sizes that will burn readily. Cut up the large
trunks also into lengths, for all that will not burn must be carried
off later. Leave all so lying until February, then choose a day with a
high wind and fire it from the windward side. It may burn some days.
Then collect all unburnt into heaps, and fire again and again until
nothing more will burn. Now take out all roots, big and small, and when
well dry, stack all these, and what was left before, and fire again and
again. The land should now be tolerably clear, and can be dug at once.
The roads should be marked off before this, for they are better not dug.

Now stake the land at the distances determined on, and a month before
the rains, or even more, if you are so far advanced, make holes for
the young seedlings at each stake, precisely like those recommended
for “planting at stake,” page 59. Only, if possible, these should be a
little larger each way than there advised, say 10 inches diameter and
15 inches deep.

Read carefully the direction as to those pits, and follow them out
here. Much of the success of your planting depends on these holes.

At the first commencement of the rains transplant, as directed under
that head in the next chapter.

Any large heavy trunks, which cannot be easily carried off the land,
may be placed longways between the lines, but the less of dead timber
you leave lying about the gardens the better.

FOOTNOTE:

[24] In no case trust to the latter alone.



CHAPTER XVII.

TRANSPLANTING.


If the pits for the plants have been all prepared, as directed at pages
59 and 75, this operation is simple enough.

A fortnight or so before it commences tip all the seedlings in the
nursery. Take off only the closed leaf at the head of each young plant
(see a leaf diagram, page 104), so that the bud at the base of the next
leaf be not injured. Doing this will make the seedlings hardier and
enable them earlier to recover the transplanting.

On the day you intend to take up the seedlings from any bed, if you
have water enough at command, flood the bed. This, as you take up each
seedling, will cause the soil, being moist, to adhere better to the
roots.

The difference between young plants transplanted with a ball of earth
round the roots, and those moved with their roots bare, is no less than
three months’ growth, if even it does not make the difference between
life and death.

Proceed thus to ensure the former. At one short end of the bed, the
lowest if it is on a slope, dig close to the first row of seedlings a
trench so deep that its base shall be lower than the lowest end of the
tap-roots. Then with a five or six-pronged steel fork (this is better
than a spade, for it does not cut the rootlets) put in between the
first and second row, and pressed down with the foot to its head, force
carefully so much of the row down into the trench. Then with the hand
take up each seedling separately, helping the soil with a very light
pressure (so light that it shall not change the lateral direction of
any of the rootlets) to adhere, and place it in a low basket sloping.
Do this again and again, till two baskets are full, when they will be
carried, banghy fashion, to the garden.

When the first row is finished clear away the loose soil, so that a
similar trench to the first shall be formed, and then proceed as above
with the second row, and so on.

No further directions for lifting the seedlings out of the nurseries
are required.

All is ready for their reception in the garden if the directions at
pages 59 and 75 have been followed out. The work now to be detailed
must be done by careful men well superintended.

In the soft soil of the lately filled up pit, described at page 59, a
hole is made either with the hand or a narrow kodalee (the former, if
the soil has not settled much, will suffice), large enough and deep
enough to take in the seedling with all the earth attached to it. The
seedling is then put in and the soil filled in and round it, which
completes the operation.

The manner, though, in which this is done is of great consequence. Four
things are all important:--(1) That the tap-root shall not be turned
up at the end because the hole is too shallow. (2) That any rootlets
projecting outside the attached earth shall be laid in the hole, and
shall preserve, when the soil is filled in, their lateral direction.
(3) That the collar of the plant (the spot where the stem entered the
earth in the nursery) shall be, when the pit is filled up, about 1-1/2
inch higher than the surface of the surrounding earth. (4) That in
filling in the hole the soil is pressed down enough to make it unlikely
to sink later, but not enough to “cake” the mould.

The following is the consequence of failure in these four points:--

1. Probably death, in any case very much retarded growth. I have
planted some seedlings so purposely, the majority died; those that
lived recovered very slowly, and digging them up later the tap-root was
found to have gone down after all by assuming the shape of the letter
S, the growth downwards being from the head of the letter.

2. Rootlets, turned away from their lateral direction, interfere with
other rootlets, and though they eventually grow right if the plant
lives, they retard it.

3. Fill in as you may (unless you “cake” the soil, which induces worse
evils) the plant sinks a little; thus, if not placed a little high, it
will eventually be too deep. If on the other hand placed too high, the
rootlets and collar will be exposed, which is an evil.

4. Unless this is attended to the plant will sink too much and the
collar be buried; likewise an evil, which it takes the young seedling
some time to recover.

Only first teaching and then practice will enable either European or
Native to plant well. This is how it should be done.

Take the seedling in the left hand, holding it by the stem just above
the collar; then take the very end of the tap-root between the second
and third fingers of the right hand, and thus put it down into the hole
(you thus insure the tap-root being straight). Now judge exactly the
height of the collar that it be as directed. Rest the left arm then on
the ground to keep the plant steady, release the tap-root, and fill up
the hole about one-third, pressing the soil lightly. The plant will
then be fixed, and you can employ both hands to fill up the remainder,
and keep the rootlets in a lateral position. Press the soil lightly
as you do so, and when all is filled up press it down a little harder
round the stem of the plant.

All the transplanting should be finished as early in the rains as
possible. A seedling, planted in the first fifteen days of June, is
worth two planted in July, and after the latter month it is generally
a case of seedlings and labour lost.

Days with heavy rain are not good to plant in. Those with showers or
light drizzling rain are best. When there is very heavy rain the soil
“cakes” much. Fine days, if the ground is wet, and if more rain may
soon be looked for, are good, better though if cloudy than sunny.

Where much planting has to be done, of necessity planting must be
carried on daily, for, as observed, it _must_ all be finished by end of
July at latest.

In case of a sunny break in the weather, stop planting after the second
day, for early rain to young transplants is a necessity.

In making a garden too much care cannot be given to the way seedlings
are placed in their homes.

Just before sending the third edition to press, I saw in the _Indian
Tea Gazette_ some details of “new transplanting and transporting
tools,” patented by Mr. Jeben. I hope these will prove a success, for
such are much wanted, and if they will do all it is said they can do, a
great boon will have been conferred by Mr. Jeben on the Tea industry.

Mr. J. W. Mountjoy, of Pandawbrang, Arracan, writes as follows
regarding these tools:--

  “The Transplanter has, in working, proved to be a complete success.
  Almost all the remaining seedlings have been transplanted by the
  aid of your instrument, without the slightest injury to their roots
  or check to their growth. The fact is, the young plants do not know
  that they have been transplanted, and now that sunshine has succeeded
  the late very heavy rains, new and vigorous growth is ‘bursting
  out’ from all the seedlings that were transplanted by means of your
  Transplanter. No manager of a Tea or Coffee plantation, who had once
  seen this instrument at work, would ever again be likely to recur
  to transplanting by hand, and not a single seedling should die when
  removed from the nursery and carried to its place of ultimate growth
  by means of your Transplanter. Your transplanting apparatus is better
  than baskets, and has moreover the great recommendation of being very
  economical. Your Transplanter will, with moderate care, last for many
  years, and combines thorough economy with thorough efficiency.”

I am glad to give the above extract, for I look on the invention, if
successful, as a most important one.



CHAPTER XVIII.

CULTIVATION OF MADE GARDENS.


As manuring, which is part of this, is treated separately, we have here
only to consider the best means of stirring the soil to give air to the
roots of the plants, and to keep down weeds, which, if allowed, injure
the yield vastly.

Unless when plants are full blown and in full bearing (and not even
then unless they are planted close) it is not only not necessary, but
a waste of labour and money, to open the soil all over the garden with
a view of stimulating or cultivating the plants. Much money has been
wasted in this way: for instance, in a garden planted 6 by 6 or 6 by 5,
and the plants but two years old, I have seen the whole dug many times
in the year. The roots of the said plants did not protrude at that age
more than 1 foot or so, what good could they possibly derive from the
extra space dug?

The soil _over_ the rootlets of Tea plants cannot be stirred too often.
The oftener it is done the oftener the trees flush, and when young the
more vigorously will they grow. What is the best way to do it?

I believe simply by digging _round_ each plant. I go to show why this
is, I believe, the best.

Putting aside the waste incurred in digging a whole garden when not
necessary, the way the soil is then dug near the plants is, I think,
objectionable. The ground is dug in a straight line _up to the plant_,
and in doing so, if the digging is deep, roots are very apt to be cut.
Again, when the work is task-work, the men shirk as much as possible
digging close up to the stems under the branches, and thus the soil
over much of the roots, is not stirred at all. This is not easy to
detect, for you must look under the branches of each tree to see how
the work has been done.

In “digging round plants” the men should _follow_ the kodalee round the
tree, and _the position of the blade in the same line as the roots_
makes any injury very unlikely. Even if tasked, as when the work is
examined, it is _only_ round the plants, it is more readily perceived
if the ground has not been stirred close up to the stems.

I therefore prefer digging round plants, with the view of cultivating
them, to digging the whole garden. I believe the object is better
attained. That it is much cheaper is evident.

The annulus, or space to be dug round, beginning 9 inches from the
stem, varies with the age of the plant. Up to two years one kodalee in
width will do, and after that say 2 feet.

The draw-hoe of 8 inches wide is a better tool for the above than
the kodalee, especially as it is work well suited to boys, and the
“draw-hoe” is a lighter tool.

Till plants from seed at stake are a year old, and till seedlings from
nurseries are the same age, calculating in the latter case from the
transplanting, no kodalee or even draw-hoe should come near them. The
soil round for 6 inches should be slightly opened once a month or so,
but it should be done with the “koorpee.”

We have now discussed the cultivation of the plants. The above often
done, say once a month, if possible, during the season, with judicious
pruning and liberal manuring, constitutes high cultivation. Did weeds
not grow there would be no need to do more, but weeds _do_ grow, and
must not be allowed. The richer the soil the more weeds, the more
manure you apply the more weeds also.

Weeds choke the plant and diminish the yield. Weeds take from the
soil, and from manure, when given, the strength you want for your
constantly recurring flushes. If, therefore, you have a large crop of
weeds you have a small yield of Tea.

How to stop this? There is one golden rule, “never let them get ahead
of you.” This, it is true, argues ample labour; but unless you _have_
ample labour for the area you cultivate, better let your money lie in
the Bank and not grow Tea. Reduce your area until you _can_ keep ahead
of your weeds, for keep ahead you must if you wish for success.

The secret of keeping ahead of weeds is to destroy them when young, to
do this again and again, as often as they come up, never allowing them
to bear seed. The kodalee, an excellent digging tool, is not good for
this: you want a lighter instrument, which can go over more ground and
will not open the soil in the dry season to any depth. The Dutch hoe,
the widest procurable in the blade, with a long lithe handle of 6 feet,
is perfect for this.

With weeds at the height fit for a Dutch hoe, viz., 3 or 4 inches, and
not numerous (which they will not be if you have “kept ahead”), a man
will easily do 45 square nulls, _id est_, 720 square yards. He would
not do more than 30 nulls with a kodalee.

The Dutch hoe must be well known. It is used for weeding drives and
walks in England.

To conclude shortly, for “hoeing and weeding” I recommend as follows:--

Dig the whole garden thrice in the year, viz., spring, rains, and
autumn. Bury all weeds as you dig in trenches between the lines.

In the intervals use the Dutch hoe as often as weeds appear.

Cultivate the plants by digging round them once a month if possible.

Do all this and you will find your garden is kept clean and well
cultivated, at far less cost than you incurred for cultivation when it
was choked with weeds for months together, while your yield will be at
the same time much increased.

If you keep your garden thus clean, and do not allow the weeds to
get ahead of you, the following table shows about the cost of each
cultivation operation each time you do it:--

  Key:

  A: Digging the whole surface
  B: Digging round plants
  C: Dutch hoeing or weeding
  ----------------+----------+--------+---------+---------+------+------
                  | Headman  | Men at |  Women  | Boys at |Total |Say in
   Detail of work | at 4-1/2 | 3-1/2  |   at    | 2-1/2   |cost  | Rs.
                  |  annas   |  annas | 3 annas |  annas  |      |
  ----------------+  --------+--------+---------+---------+------+------
          A       |   3/4    |    12  |    5    |   ..    |2|13|6|  3
          B       |   1/2    |    ..  |    4    |    5    |1|13|9|  2
          C       |   1/2    |     4  |   ..    |   ..    |1|14|3|  2
  ----------------+----------+--------+---------+---------+------+------

If weeds get ahead the cost in each case will be nearly double the
above.

The following table, which is as near the mark as any such estimate can
be, will be found useful. It will also be made use of when calculating
the cost of making a garden in Chapter XXIX., pages 164, 165, and 166.


  _Table showing the cost per annum of keeping up at its best 100 acres
  of Tea from the year it is planted until the sixth year inclusive._

  ------+-------------+-------------+----------------------------------
   Year |Rate per acre|Per 100 acres|    Remarks
        |  per annum  |             |
  ------+-------------+-------------+----------------------------------
        |     Rs.     |     Rs.     |
  First |      50     |    5,000    |The year the seed is sown at stake
  Second|      60     |    6,000    |
  Third |      70     |    7,000    |
  Fourth|      80     |    8,000    |
  Fifth |      90     |    9,000    |
  Sixth |     100     |   10,000    |The plants should be large plants
        |             |             | now, but they will not be at full
        |             |             | bearing until the eighth year.
  ------+-------------+-------------+----------------------------------

The above rates in the case of a 300-acre garden making will include
_everything_ but buildings.

The rates are progressive, because the expenditure on the following
increases, or should increase, yearly.

  1. Manager’s pay (say every second year).
  2. Assistant (first entertained, say third year).
  3. Cost and wear of tools.
  4. Cost of pruning.
  5. Cost of cultivation.
  6. Cost of manure.
  7. General expenses.

No cost for Tea manufacture is included in the above, as this is
estimated for separately. See table at pages 160, 161, and 162.

Keeping up high cultivation in every way and manuring liberally, a made
garden in full bearing can be kept up to its highest producing powers
(including the pay of the manager, establishment, and everything else)
for Rs. 100 per acre per annum.

An acre of Tea may, I am aware, be kept up in a manner for Rs. 50 or so
yearly, but the profit on such a plantation must be nil.

On the contrary, with the above expenditure per acre, on a good and
favourably located garden, the profit will be very large. See table at
page 172.

It is with Tea as with all other cultivation. It has been proved in
England, and in all other countries where really high cultivation is
followed out, that the higher the system followed the greater the
profit.



CHAPTER XIX.

PRUNING.


It is stated elsewhere at length (page 102) _why_ I conceive pruning to
be necessary for the Tea plant. Whether I am right or not, the fact is
certain that without pruning very little leaf is produced.

Pruning must be done in the cold weather when the plant is hybernating,
that is to say, when the sap is down. The sooner _after_ the sap goes
down it is done the better, for the sooner the tree will then flush in
the spring.

There have been many theories about pruning Tea bushes, but none, I
think, worth much _practically_, for the simple reason that it is
impossible to prune 250,000 plants (the number in a 100-acre garden,
at 2,500 to the acre)[25] with the care and system a gardener prunes
a favourite fruit tree. The operation _must_ be a coarse one, done by
ignorant men, in large numbers at one time, who can in a measure be
more or less taught, and the nearer they do right the better: still,
really careful and scientific pruning can never be carried out on a Tea
plantation.

The time to do it, too, is very limited. It cannot be begun before the
trees have done flushing, say, at the earliest, middle of November, or
continued, if early flushes and a large yield next season is looked
for, beyond end of January, at the latest. Thus at the most two months
and a half is all the time given.

I shall confine myself therefore to giving such directions as will be
practically useful.

The best instrument is the common “pruning knife.” It cuts far cleaner
than the “shears,” besides which the natives very seldom use the latter
well. What is called in England a “hedge-bill” is useful to trim the
outsides of the trees. If required it must be got from England, as I do
not think it is procurable in Calcutta. Whatever instruments are used
should be kept very sharp, and for this purpose, besides sharpening
them every morning on the grinding stone, each pruner should be
provided with a small pocket “hone.”

The theory, and it is correct, is in pruning, to cut near above a bud
or branch, but not near enough to injure them. The cut should be quite
clean and sloping upwards, so that nothing can lodge on it. This theory
can be, and must be, strictly carried out in cutting the thick stems
and branches, but it is quite impossible to do it with the slender
branches or twigs of the tree.

Prune so as to cause lateral growth. A Tea plant should never be
allowed to exceed, say, 4 feet in height, but the wider it is the
better.

Prune off all lower branches tending downwards,[26] for the plant
should, if possible, be clean underneath to a height of, say, 6 inches.
This clean stem high class plants have naturally, not so the Chinese,
or the Chinese cast of hybrid.

Plants should be more or less pruned out in the centre. In the
following spring young wood is then formed in the heart of the tree,
and it is only young wood and shoots that give leaf.

Plants, if above two years old (see foot note next page), exceeding
2-1/2 feet in height at the end of the season (and all plants of any
age will) may be pruned down to 20 inches, but the thick wood must be
pruned down to varying heights several inches lower.

Small plants must naturally be more lightly pruned.

The best plan is, I think, to have two gangs:

The first to go ahead and cut out the thick wood (here judgment is
necessary, so let them be the best men) to varying heights, from about
11 to 18 inches. The second gang to follow, each with a rod 20 inches
long, to cut down all the light wood left to that level.

All plants, how low or how young soever they may be, must be pruned
somewhat.[27] The lower their stature and the less their age the less
pruning they require.

Of the two extremes, at least with the Tea plant, it is probably
better to over than to under-prune. The treatment of the plants, with
reference to the leaf to be taken in the spring, must be a good deal
regulated by the way, or rather the extent, to which they have been
pruned. On this point see page 103.

The cost of pruning depends on whether it is high or low, and whether
the plants are large, middling, or small. The greatest cost is about
Rs. 6, the least about Rs. 3 per acre.

Let all prunings be buried between the lines of plants, if possible,
before the leaves have even withered. They make capital manure, but
much of the virtue escapes if they are allowed to lie on the ground any
time before they are buried.

FOOTNOTES:

[25] In a 500-acre garden the number is 1,250,000, which _ought_ all to
be pruned in two months!

[26] The best plan with the lowest branches is to _pull_ them off, with
a sharp downward action, as then they will not grow again.

[27] But not before the end of 18 months after transplanting, as the
object at first is to get a long tap-root to draw moisture from low
down, and this is best attained by allowing the plant to grow as it
will. I look on this as all-important. I care not how high a plant may
grow, for 18 months I would in no way interfere with its growth.



CHAPTER XX.

WHITE ANTS, CRICKETS, AND BLIGHT.


These insects (for blight, too, is said to be an insect) are very
destructive to the Tea plants. The cricket, however, only injures it
when quite young, so we will consider that little pest first.

When Tea seed germinates, and the young seedling is 2 or 3 inches high,
the cricket delights to cut the stem and carry, or try to carry, the
two or three green leaves attached to the upper part into its hole.
Even after seedlings are planted out, if the stems are slender, it cuts
them. To the young seedlings, in nurseries or planted “at stake,” they
often do great harm, killing in some places one-third or so.

It is much easier to prevent their ravages in nurseries than in this
latter case, simply because the spot in which they must be sought and
destroyed is circumscribed in the one, almost unlimited in the other.

Only one thing can be done. Employ boys (they soon get clever enough
at the work) to hunt for their holes and dig them out. The holes are
minute, but run down a long way. The only plan to follow them is to
put in a thin pliable stick and remove the soil along it. On getting
to the bottom of the stick, if it is not the bottom of the hole, you
repeat the operation till you _do_ get to the bottom, and there you
will generally find the cricket.

Early in the morning they can be often found and caught outside their
holes. The boys employed should be paid for them by the number they
catch. They can be placed alive and brought to the factory in a hollow
bamboo, and then killed in some merciful way.

When once a Tea plant has got a stem as thick as a thick pencil no
cricket can hurt it.

They are much worse in some places than others, and in my experience I
have found them worse on low lands.

The white ant is a much more formidable enemy than the cricket. They
_do_ (as all planters know) attack and destroy living bushes.[28]
Whether they first attack some small dead portion or not is a question,
but practically it does not signify the least, for if they do they
manage to find such in about one-third of the trees in a garden.
Beginning with the minute dead part they kill ahead of them as they go,
and will, eventually, in many cases, if left alone, kill the largest
trees.

They have a formidable enemy in the small black ant which exists in
myriads, and kills the white ant whenever the latter is not protected
by the earthen tunnels he constructs. In many places so great is the
pest that, did this small black ant not exist, I believe no Tea Garden
could stand.

From the close of the rains to the cold weather is the worst time for
white ants, and the time the planter should guard particularly against
their ravages. At that time if he examines his trees closely he will
very likely find white ants on a quarter of the whole.

Digging round the plant where they are disturbs their runs and does
much good. At the same time they should be brushed off any part of the
tree they have attacked, and the tree should be well shaken.

All this, however, only does temporary good, for they often are found
as thick as ever on the plant a week later.

Tobacco water is beneficial, but in wet weather it is soon washed off.


Kerosene oil is _very_ efficient. A little is put round the stem,
but it is expensive. The next best thing I know is the earth oil
(petroleum) from Burmah, and this is cheap enough. It is thick, but
used from a bottle it gets heated by the sun and is then quite limpid.

When white ants are found on a tree, a little with a small brush is put
on the part they have attacked. They are also well shaken off, and a
ring of oil is placed round the stem. My experience is that they will
not attack that tree again for a long time. I was at first fearful that
both it and the kerosene (the one, I believe, is only a manufacture
of the other) would injure the trees, but both are safe. I strongly
recommend others to try it, if they doubt, on a small spot only in the
first instance.

Whatever is used, or whatever is done, white ants must not be left to
work their will in the autumn. All the trees should then be examined
once at least, and once again, if possible, the following spring.

Blight (a serious matter, I hear, in Cachar) I know but little of. I
do not remember hearing anything about it when I was there, now some
fourteen years ago. It is rare in the Chittagong district, but I have
seen one or two trees attacked with it. Under its influence the young
leaves get covered with brown spots and shrivel. It is most destructive
to the yield of a garden.

From one or two experiments made I believe pruning off all the diseased
branches, and scraping back the soil for a space of 2 feet round the
stem, so as almost to lay the roots bare, will be found beneficial, but
I do not speak with certainty.

All the Himalayan gardens are free from these three pests detailed,
except that occasionally a few crickets have been seen.

FOOTNOTE:

[28] A long controversy on this point lately took place in the papers;
that is to say, the point discussed was, whether white ants do or do
not attack living tea trees.



CHAPTER XXI.

FILLING UP VACANCIES.


So difficult is this to do, that I have heard several planters declare
they would attempt it no further, but, on the contrary, accept the
vacancies in their gardens as an unavoidable evil.

That it is difficult I, too, can certify. Seedlings put into vacant
spots year after year die, either in the rains they are planted or in
the following spring. If, however, a few yards off a fresh piece of
land is taken in and planted, the plants live. What is the reason? It
can be nothing connected with the soil, for on adjacent spots they live
and die.

It puzzled me a long time, but I _believe_ I can now explain it.
_First_, seedlings planted in vacant spots in a garden are never
_safe_. When in the rains there are many weeds in the gardens, and it
is being dug, the young seedlings are not observed, are either dug up,
or injured so by the soil being dug close to them, that they shortly
after die. This is, I believe, the _principal_ cause of the failure,
and it may be in a great measure, if not entirely, obviated by putting,
_first_, a high stake on either side of the seedling, and taking care
it remains there all through the rains. _Secondly_, as an additional
precaution, and a very necessary one, before any such land is dug, send
round boys with “koorpies” to clean away the jungle round the young
plants, and at the same time open the soil slightly over their roots.
Doing this “cultivates” them, and the plants being apparent, with the
newly-stirred vacant spaces round them, are seen by the diggers, and
are not likely to be damaged.

The second cause of failure I attribute to the old plants on either
side of the young seedling, taking to themselves all the moisture
there may be in the soil during any drought. The young seedling, whose
tap-root at the time is not a long one (for it is in the spring of the
year following the year of planting that this occurs), is dependent
for life entirely on the small amount of moisture that exists in the
soil, at that insignificant depth (say 8 inches). But on two sides
of the said seedling’s tap-root, and in fact surrounding it, if the
neighbouring Tea bushes are full grown, are the feeding rootlets of the
big plants, sucking up all the moisture attainable (the necessities of
_all_ plants being then great), and leaving none for the poor young
seedling, which consequently dies in the unequal contest.

This last evil (in climates where there is a deficiency of spring
rains, and, in fact, more or less in all Tea localities, for in none
is there as much rain as the plants require in the spring) there is
no means of avoiding as long as seedlings, after transplanting, _lose
time_, the effect of the transplanting, and thus fail to attain a good
depth before the said dry season.

In fact, unless something is devised, I believe with many, trying to
fill up vacancies is a loss of time and money.

The pits to plant in, advised at page 59, should of course be made in
these vacant spots, for they help much towards the early descent of the
tap-root. Still they can scarcely avail sufficiently to avoid the evil,
if the plant is lying inert, as is generally the case for two or three
months after planting; this delay being, moreover, in the rains, the
best growing time.

If we can devise any means to avoid this delayed growth in the young
seedling after it is transplanted, then the tap-root, before the
drought of next spring, will have descended low enough to gather
moisture for itself; that is, from _lower_ depth than the greater
number of the rootlets of the neighbouring big plants traverse. Could
this be done, and if the means above detailed are resorted to, to
prevent the young plants being injured when the gardens are dug, I see
no reason why vacancies should not be successfully filled up. Then
might be seen, what nowhere can be seen now, a Tea garden full of
plants, that is, with _no_ vacancies.

When it is considered that many gardens in all the districts have 30 or
even 40 per cent. vacancies, none less than say 12 per cent., we may
strike a fair average and roughly compute the vacancies in Tea gardens
throughout the country at 20 per cent. In other words, the yield of
Tea from India, with the _same_ expenditure now incurred, would be
one-fifth more were plantations full!

I have shown how the first evil can be obviated. I _think_ the
following will obviate the second.

Get earthen pots made 7-1/2 inches diameter at the head and 7-1/2
inches deep, like the commonest flower pots, only these should be
nearly as wide at the bottom as at the top. A circular hole, 2 inches
diameter, must be left in the bottom. Fill these with mould of the same
_nature_ as the soil of the garden where the vacancies exist. Put two
or three seeds in each, all near the centre, and not more than half an
inch below the surface. Place these pots, so filled, near water, and
beneath artificial shade, as described in Chapter XIII.

When the seeds have germinated, and the seedlings have two or three
leaves, so that you can judge which is the best class of seedlings in
each pot,[29] root out all but one, the best one. Now remove the shade
gradually, water from time to time, and let the seedlings grow in the
pots till the rains. Having, before the rains, made the holes at the
vacancies as before described, after the first fall carry the pots to
the garden and place each one near a hole.

Then plant as follows. Stand the pot on the brink of the hole, having
previously with a hammer broken the bottom. Then crack the sides also
gently, and deposit pot and all in the hole at the proper depth. If
not enough broken, the sides of the pot may now be further detached,
nay, even partially removed. Now fill up with earth to the top. Pieces
of the pot left in the hole will do no harm; but it, the pot, must
be sufficiently broken at the bottom to allow of the free descent of
the tap-root, as also enough broken at the sides to allow of the free
spreading of the rootlets.

If all this has been carefully done, so that the mould in the pot shall
not have been shaken free of the rootlets, the seedlings will not even
_know_ it has been transplanted. Its growth will not be delayed for a
day, instead of two or three months; and by the time the dry season
comes, the tap-root will have descended far enough to imbibe moisture.

Another plan to effect the same object. Instead of pots, use coarse
bamboo open wicker-work baskets. The split bamboo forming the said
wicker-work about half an inch wide, the interstices about one quarter
of an inch square. Let the diameter of the basket be the same at top
and bottom, viz., 9 inches; the depth of the basket 10 inches.

When the seedlings in the nursery are large enough to enable you
to select a good class of plant, transplant one into each basket
previously filled with soil.[30] This being done when the plants are
very young, and there being _then_ no difficulty in taking them up
with earth attached to their short tap-roots and rootlets, they will
scarcely be thrown back at all. Being near water they can also be well
tended. Put basket and all into the vacant hole at the beginning of the
rains, and fill up as directed for the pots. The interstices will allow
the feeding rootlets to pass through, besides the basket rots quickly
under ground, so quickly it cannot impede the plant.

Seed is not sown at once in the baskets as in the pots, because the
baskets would not last so long. Even putting the seedling in it during
(say) February, the basket, with the occasional watering necessary,
will, more or less, have rotted before it is put into the hole.

I have concluded a contract for ten thousand pots and five thousand
baskets at half an anna each for both kinds. Two pice, to ensure the
filling up of a vacancy, is not a large outlay.

Since writing the above I have had experience of both the above plans.
The pot system is far the better, and answers very well.[31] I am
now trying to improve this still further by making the pots a little
larger, and placing a thin inner lining of tin inside each about half
an inch from the sides. This space is first filled with sand, then the
pot is filled with mould, and the tin pulled out. The same tin will
therefore do for any number of pots. The seed is then put in.

I think by this plan if, when about to plant, the mould in the pot is
well wetted, that it, with the seedling, can be turned out whole in one
piece, and then put in the hole _without_ the pot.

The same pots would then answer year after year, and the expense would
be quite nominal.

If well done, the seedling in this, as in the former case, would not
even _know_ it had been transplanted.[32]

FOOTNOTES:

[29] By “best class” I mean the most indigenous class.

[30] Mind again this be of the same nature as the garden soil.

[31] The baskets are too frail; being often wetted, they fall to pieces
before the planting time.

[32] It may be that the transplanting and transporting tools invented
by Mr. Jeben (see page 79) will solve the difficulty of filling up
vacancies.



CHAPTER XXII.

FLUSHING AND NUMBER OF FLUSHES.


The Tea plant is said to flush when it throws out new shoots and
leaves. The young leaves thus produced are the only ones fit to make
Tea, and the yield of a plantation depends therefore entirely on the
frequency and abundance of the flushes.

The way a flush is formed is fully explained under the head of “leaf
picking” (pages 103, 104, and 107).

The number of flushes in different plantations varies enormously,
owing, _first_, to climate; _secondly_, to soil; _thirdly_, to the
pruning adopted; _fourthly_, to the degree of cultivation given; and
_fifthly_, though not least, to the presence or absence of manure.

How to secure all these advantages to their fullest extent is shown
under those heads, and we have here only to consider what is a low, a
medium, and a high rate of flushing per season.

In doing this we must speak of elevated (as Himalayan) gardens
separately. The cool climate of heights makes it impossible for Tea to
flush there as on the plains.

Speaking generally of elevated gardens (the higher they are the
shorter the period, and _vice versâ_), seven months may be considered
as the average producing period, viz., from beginning of April to end
of October, and during that time twelve to fifteen flushes may be
obtained, which, I believe, with high cultivation and liberal manuring,
can be increased to eighteen.

In all localities, with favourable Tea climates, the plants flush
both for a longer period and oftener. Speaking generally also, in this
case, of the five best localities, viz., Assam, Cachar, Chittagong,
the Terai below Darjeeling, and the Western Dooars (for even in these
districts many advantages exist in one garden which do not in another),
the following is an approximation to the flushing periods:--

_Upper Assam._--February 25th to November 15th.

_Lower Assam._--February 20th to November 20th.

_Cachar._--February 20th to November 20th.

_Chittagong._--March 10th to December 20th.

_Terai below Darjeeling and Western Dooars._--March 1st to November
20th.

The opening period is a little late in Upper Assam on account of the
cold, and closes a little earlier for the same reason.

Lower Assam and Cachar are much alike.

The opening in Chittagong is later than in the two just mentioned from
want of early rains, but the season continues longer on account of the
low latitude and consequent deferred cold weather.

Roughly, then, rather more than nine months may be assumed as the
flushing period for these districts. The next point is how _often_ do
gardens in these localities flush in that time.

Not very many planters can say, certainly, how often their gardens have
flushed in a season, because they are picked so irregularly, and no
account of the different flushes kept. Enquiring on this point, when
I was in Cachar some thirteen years ago, 9 to 24 were the minimum and
maximum numbers given me at different gardens, showing how little was
really known about it.

Such knowledge as I have on the subject is mostly derived from
carefully kept records of my own garden in the Chittagong district. The
plantation is all worked in sections, in the way described previously,
and the dates given in the table below are the days each flush was
finished (that is, the picking was finished) during the seasons 1869
and 1870; 1869 being carried up to the end of the season, 1870 up to
the date I wrote the first edition of this Essay.

In the table it will be observed there is a great difference between
the two years. The section for which the dates are given was planted
from seed beds in the month of June, 1866. In 1869 it was therefore
only three years old. This will partly account for the first flush
occurring a month earlier in 1870, as it was then a year older; but
fortunate early rains in 1870 had also much to do with it.

  --------+-------------+----------+-------------+-------------
          |    1869     | Interval |    1870     | Interval
  Flushes +-------------+ in days  +-------------+ in days
          |    Dates    |          |    Dates    |
  --------+-------------+----------+-------------+-------------
     1    | March    22 |    ..    | February 22 |    ..
     2    | May       6 |    44    | March    30 |    35
     3    | „        29 |    23    | April    13 |    10
     4    | June     11 |    12    | „        25 |    12
     5    | „        23 |    12    | May       5 |     9
     6    | July      5 |    11    | „        14 |     9
     7    | „        17 |    12    | „        25 |    11
     8    | „        31 |    14    | June      4 |     9
     9    | August   10 |     9    | „        12 |     8
    10    | „        21 |    11    | „        22 |    10
    11    | Sept.     2 |    11    | July      1 |     8
    12    | „        12 |    10    | „         8 |     7
    13    | „        25 |    13    | „        16 |     8
    14    | October   9 |    13    | „        25 |     9
    15    | „        22 |    13    | August    2 |     7
    16    | Nov.      2 |    10    | „        11 |     9
    17    | „        11 |     9    | „        21 |    10
    18    | „        19 |     8    | „        29 |     8
    19    | Dec.      4 |    14    | Sept      7 |     8
    20    |      ..     |    ..    | „        18 |    11
    21    |      ..     |    ..    | „        27 |     9
    22    |      ..     |    ..    | October   5 |     7
  --------+-------------+----------+-------------+-------------
    Average intervals   | Nearly 14|             |Very little
    between Flushes.    |  days.   |      ..     |over 10 days.
  ----------------------+----------+-------------+-------------

In 1869 there was no flush between March 22nd and May 6th, a period of
44 days; and in 1870, none between February 22nd and March 30th, a
period of 35 days, a very long time in both cases, which is entirely
accounted for by the dry weather prevailing at Chittagong in the spring
(see under head of Climate), for in Cachar, Assam, and the Western
Dooars two or three flushes would have occurred in that time.

There were 19 flushes in all in 1869, and 22 in 1870, up to the time I
wrote, so there were probably in all 27 in the latter year.

In the table I give the intervals between each flush. It shows an
average of 14 days in 1869 to 10 days in 1870; the difference is due to
the increased age of the plants, and the liberal manuring given in the
cold weather 1869-70.

Such a result as is shown for 1870, and the probable result of 27
flushes to the end of that season, could not be obtained without high
cultivation and liberal manuring. The land in question had been manured
every year since it was planted, but an extra dose was given in the
cold weather of 1869-70. The ground was therefore very rich.

I think, therefore, 25 flushes in the season may be looked for on
gardens in good Tea climates, when high cultivation and liberal
manuring are resorted to. Where manure cannot be obtained, I think,
even if in other respects the land is highly cultivated, more than 22
flushes will not be obtained. Where neither manure nor high cultivation
is given, above 18 flushes will not be got.

It seems to be a general idea with planters (see diagram, page 104)
that when a flush is picked the succeeding flush, at an interval of say
seven to ten days, consists of shoots from the axis of the leaf down
to which the previous flush was picked. Thus in the diagram, supposing
the shoot to be picked down to the black line above 2, the idea is the
next flush will be a shoot springing from the same place, viz., the
axis of leaf _d_. But it is _not_ so. In the above case it will take a
whole month, after the said shoot has been picked, before the new shoot
from the base of the leaf _d_ is ready to take, probably six weeks in
Himalayan gardens.

’Tis true the flushes in favourable Tea climates follow at about
seven to ten days from each other, but these are _other_ shoots. The
replacement of the actual shoot taken is a whole month in developing.
I have carefully watched this, and am sure I am right.

With similar treatment, gardens in Cachar, Assam, and the Western
Dooars would probably give two or three more flushes in the season than
Chittagong, because there the spring rains are much more abundant; and
I am very certain that, if the day ever comes that manure in large
quantities is procurable in those districts and is applied, the yield
on those gardens will be very large.[33]

The difference between very small and very large profits is represented
by 18 and 25 flushes, so I strongly advise all planters to cultivate
highly, and to get all the manure they possibly can. If even procured
at a high figure, it (the manure) will pay hand over hand.

FOOTNOTE:

[33] Where new gardens are made on rich virgin soil, to manure them at
all for the first few years is, I think, unnecessary. But the richest
soils on Tea gardens get exhausted in time, and manure should be
applied _before_ this point is arrived at.



CHAPTER XXIII.

LEAF PICKING.


The first consideration is how to get the largest quantity of leaf
without injuring the trees.

To a certain extent, it is true that the more a Tea bush is pruned and
picked the more it will yield. It appears as if Nature were always
trying to repair the violence done to the tree by giving new mouths or
leaves to breathe with in place of those taken away. I may exemplify my
meaning in another way. A Tea bush which has as many leaves on it as
_it requires_ will throw out tardily new shoots, and their number will
be small. In other words, a plant which is not pruned, and from which
the young leaves are not taken, grows gradually large and bushy, and
then gives up flushing altogether. It has all the leaves it _requires_,
and it has no necessity to throw out more.

If, however, Nature is too much tried, that is, if too much violence
is done to her, she sulks and will exert herself no more. Up to this
point, therefore, it is well to urge her. How can we know when we have
reached it?

Only general rules can be laid down. Experience is the great
_desideratum_ on this and many other subjects connected with Tea.[34]

If the plant can always be kept in such a state that the foliage,
without being _very much so_, is still less than Nature requires, I
conceive the object will be attained.

The greatest violence is done to the plant when it is pruned, and
reason would seem to argue that when this violence is repairing, that
is, when the first shoots in the spring show themselves, and until
new mouths (or leaves) in sufficient quantities exist, until then but
little leaf should be picked.

Fortunately, moreover, while in the interests of the plant this is the
best plan, it also is the mode by which the largest yield of leaf will
be secured in the season. I go to show this.

The ordinary size of a good full-grown Tea plant, at the end of the
season, is, say, 3-1/2 or 4 feet high, and 5 feet diameter. It is
pruned down, say, to a height of 2 feet, with a diameter of 3 feet.
It is then little more than wooden stems and branches, and to anyone
ignorant of the _modus operandi_ in Tea gardens, it would appear as
if a plantation so pruned has been ruined. The tree remains so during
all its hybernating period, that is, during the time it is resting and
the sap is down (this period is longer or shorter, as the climate is
a warm or cold one, and it is always during the coldest season), but
on the return of spring new shoots start out from the woody stems and
branches in the following way:--At the axis or base of each leaf is a
bud, the germ of future branches, these develop little by little, until
a new shoot is formed of, say, five or six leaves, with a closed bud at
top. Then if it be not picked the said bud at top hardens. At the axis
or base of each of the said five or six leaves are other buds, and the
next step is for one, two, or three of these to develop in the same way
and form new shoots. The original shoot grows thicker and higher until
it becomes a wooden branch or stem. The same process, in their turn,
is repeated with the new shoots. A diagram (see next page) will make
my meaning clear. We here have a shoot fully developed, of six leaves,
counting the close leaf _a_ at top as one, viz., the leaves _a_, _b_,
_c_, _d_, _e_, _f_. The shoot has started and developed from what was
originally a bud at _K_, at the axis or base of the leaf _H_. In the
same way as formerly at _K_ a bud existed, which has now formed the
complete shoot or flush _K a_, so at the base of the leaves _c_, _d_,
_e_, _f_, exist buds 1, 2, 3, 4, from which later new shoots would
spring. These again would all have buds at the base of the leaves,
destined to form further shoots, which again would be the parents of
others, and so on to the end of the season, or until the tree is pruned.

[Illustration]

It will readily be seen the increase is tremendous. It is only limited
by the power of the soil to fling out new shoots, and the _necessities_
of the plant, for, as I have explained, when as much foliage exists as
the plant requires, but few new shoots are produced.

Now supposing the shoot in the diagram to be (with perhaps another not
shown at _L_) the first on the branch _I I_ in the spring (the said
branch having been cut off or pruned at the upper _I_). It is then
evident the said shoot is destined to be the parent and producer of all
the very numerous branches and innumerable shoots into which the plant
will extend in that direction. It is, in other words, the goose which
will lay all the future eggs. If, eager to begin Tea making early, the
planter nips it off, the extension on that part of the tree is thrown
back many weeks. It may be taken off at 1, 2, or 3 (the back lines
drawn show the proper way to pick leaf); the least damage will be done
if it is taken off at 1, the most at 3.

The said shoot _K a_ is the first effort of Nature to repair the
violence done to the tree by pruning. It is the germ of many other
branches and shoots, and it ought _never_ to be taken. I have, I hope,
made so much plain.

There is, however, another consideration. Any shoot, left to fully
develop and harden, does not throw out new shoots from the existing
buds 1, 2, 3, 4 so quickly as one checked in its upward growth
by nipping off its head. For instance, supposing the shoot under
consideration _not_ to be the first of the season, but on the contrary
to be a shoot when the plant has developed sufficiently to make picking
safe, if taken off at 2, then the new growth from 2, 3, 4 will be much
quicker than it would be had the whole shoot been left intact.

Our object then with _first_ shoots should be to secure this advantage
without destroying any buds, and this we can do by taking off simply
the closed leaf at the top _a_. This must be done so as not to injure
the bud at the base of the second leaf _b_ (I have not numbered it, for
there is no room in the diagram to do so), and we shall thus leave all
the buds on the shoot intact.

Again here the interests of the plant, and profit to the planter,
go hand in hand. The closed bud _a_ in this case will be found very
valuable. I go to show this.

The value of Tea is increased when it shows “Pekoe tips.” Only the
leaves _a b_ make these. They are covered with a fine silky whitish
down, and, if manufactured in a particular way, make literally white
or very pale yellow Tea,[35] which, mixed with ordinary black Tea, show
as “Pekoe tips.” In ordinary leaf-picking these two leaves are taken
with all the others, but unfortunately, when manufactured with them,
they lose this white or pale yellow colour, and come out as black as
all the other Tea.

As the season goes on, this is less and less the case, till towards
the end nearly all the _a b_ leaves show orange-coloured in the
manufactured Tea. Still they are not _white_ (the best colour) as they
can be made when treated separately. No means have yet been devised
to separate them _before_ manufacture from the other leaf, and though
sometimes picked separate, the plan has serious objections (see next
page). In the case, however, of the first two or three flushes the
welfare of the plants demands that no more should be taken, and though
the quantity obtained will be small, it will, if carefully manufactured
so as to make “white Pekoe tips,” add one or two annas a lb. to the
value, when mixed with it, of one hundred times its own weight of black
Tea!

More will be found under this head in the Tea manufacturing part. I
now beg the question that the said downy leaves taken alone are very
valuable.

In detailing the mode of picking I advocate, it would be tedious to go
minutely into the reasons for each and everything. I have said enough
to explain a good deal, but will add anything of importance. Of the
latter are the following.

Tea can be made of the young succulent leaves only. The younger and
more succulent the leaf the better Tea it makes. Thus _a_ will make
more valuable Tea than _b_, _b_ than _c_, and so on; _e_ is the lowest
leaf to make Tea from, for though a very coarse kind can be made from
_f_, it does not pay to take it. The stalk also makes good Tea, as far
as it is really succulent, that is, down to the black line just above
2.

The leaves are named as follows from the Teas it is supposed they would
make:--

  _a._--Flowery Pekoe.
  _b._--Orange Pekoe.
  _c._--Pekoe.
  _d._--Souchong, 1st.
  _e._--  „       2nd.
  _f._--Congou.

                 _a_, _b_, _c_--Pekoe.
  Mixed together
                 _a_, _b_, _c_, _d_, _e_--Pekoe Souchong.

If there be another leaf below _f_, and it be taken, it is named, and
would make Bohea.

Each of these leaves was at first a flowery Pekoe leaf (_a_), it then
became _b_, then _c_, and so on.

That is to say, as the shoot developed, and a new flowery Pekoe leaf
was born, each of the leaves below assumed the next lowest grade.

Could the leaves fit to make each kind of Tea it is proposed to make
be picked and kept separate, and each be manufactured in the way most
suitable to its age, and the Tea to be produced, the very best of every
kind could easily be manufactured. But this cannot be; the price of
Tea will not allow it, and the labour to do it would moreover fail. It
has been attempted again and again to do it, partly to the extent of
taking the Pekoe leaves _a_, _b_, _c_ separate from the others (for
the manufacture best suited to these upper leaves is not suited to the
lower), but it has been as often abandoned, and I doubt if it is now
practised anywhere. I am sure it will never pay to do it.

Picking leaf is a coarse operation. It is performed by 80 or 100 women
and children together, and it is impossible to follow each, and see
it is done the best way. They must be taught, checked, and punished
if they do wrong, and then it will be done more or less right; but
perfection is not attainable.

I advise the following plan in picking. Please refer to the diagram:--

If the garden has been severely pruned (as it ought to be) take only
the bud _a_ for _two_ flushes; then for _two_ more nip the stalk above
1, taking the upper part of leaf _c_, as shown (done with one motion of
the fingers). But from the fifth flush take off the shoot at the line
above 2, and by a separate motion of the fingers take off the part of
leaf _e_ where the black line is drawn. By this plan, when the rains
begin, the trees will show a large picking surface, for plenty of buds
will have been preserved for new growth. After the month of August
you may pick lower if you like, as you cannot hurt the trees. For
instance, you may nip the stalk and upper part of leaf _e_ together,
and separately the upper part of _f_.

The principle of picking is to leave the bud at the axis of the leaf
down to which you pick intact.

Some planters pick all through the season at the line above 1, and take
the _d_ and perhaps the _e_ leaf separately. I do not like the plan,
for though it will make strong Teas, the yield will be small. Moreover,
the plants will form so much foliage; they will not flush well; and
again, they will grow so high that boys who pick will not readily reach
the top.

Shortly, the principle I advocate is to prune severely, so that the
plant in self-defence _must_ throw out many new shoots; to be sparing
and tender with these until the violence done to the tree is in a
measure, but not quite, repaired; then, till September, to pick so much
that the wants of the plant in foliage are never quite attained; and
after September to take all you can get.

I believe this principle (for the detailed directions given may be
varied, as for instance when trees have _not_ been heavily pruned)
will give the largest yield of leaf, and will certainly not injure the
plants.

FOOTNOTES:

[34] See foot-note, page 86, which shows that for 18 months after
transplanting, young bushes should not be pruned or picked at all.

[35] I mean manufactured Tea. The infusion is called liquor.



CHAPTER XXIV.

MANUFACTURE. MECHANICAL CONTRIVANCES.


To manufacture your leaf into good Tea is certainly one of the first
conditions for success. It will avail little to have a good productive
garden if you make inferior Tea. The difference of price between well
and ill-manufactured Tea is great, say 4 as. or 6d. a lb., and this
alone will, during a season, represent a large profit or none.

Fortunately for Tea enterprise, the more manufacture is studied the
more does it appear that to make good Tea is a very simple process. The
many operations or processes formerly considered necessary are now much
reduced on all gardens. As there was then, that is formerly, so there
is now, no _one_ routine recognized by all, or even by the majority;
still simplicity in manufacture is more and more making its way
everywhere; and as the real fact is that to make the best Tea, but very
few, and very simple, processes are necessary, it is only a question of
time ere the fact shall be universally recognised and followed out.

For instance, panning the “roll”[36] was formerly universally
practised. Some panned once, some twice, some even three times! But,
to-day, pans are not used in most gardens at all!! Other processes,
or rather in most cases the repetition of them, have been also either
discarded or abridged. But a short statement of manufacture in old
days, and the simplest mode of manufacture, will best illustrate my
meaning:--

  -----+--------------------------------+----------------------------------
       |    One and a common old plan   |One plan to-day by which the best
       |                                |         Tea can be made
  -----+----------+---------------------+-----+----------+-----------------
       |  Number  |                     |     |  Number  |
  Days |    of    |    Detail           |Days |    of    |   Detail
       |operations|                     |     |operations|
  -----+----------+---------------------+-----+----------+-----------------
  1st  |     1    |Withering.           |1st  |    1     |Withering.
      {|     2    |1st Rolling.         |    {|    2     |Rolling.
      {|     3    |2nd  „               |2nd {|    3     |Fermenting.
      {|     4    |Fermenting.          |    {|    4     |Sunning (if sun).
      {|     5    |1st Panning.         |    {|    5     |Firing (Dholing).
  2nd {|     6    |3rd Rolling.         |     |          |
      {|     7    |2nd Panning.         |     |          |
      {|     8    |4th Rolling          |     |          |
      {|     9    |Sunning.             |     |          |
      {|    10    |1st Firing (Dholing).|     |          |
      {|    11    |Cooling and crisping.|     |          |
  3rd  |    12    |2nd Firing (Dholing).|     |          |
  -----+----------+---------------------+-----+----------+-----------------
    3  |    12    |  Total days and     |  2  |    5     | Total days and
       |          |    operations.      |     |          |   operations.
  -----+----------+---------------------+-----+----------+-----------------

So much for simplicity, and I affirm that no more than the five
operations detailed are necessary. I shall try to show this further on.

In studying Tea manufacture I first tried, in order to get reliable
data to go on, to ascertain the effect of each and every operation,
and not only that, but the effect on the made Tea of each operation
exaggerated and diminished. It would be tedious, and of no use, to set
out in detail all the experiments I conducted, the results only I will
try to give.

I began at the beginning. Why wither at all? I made Tea (following out
in each case all the other processes detailed in the old plan) of 1st,
totally unwithered leaves; 2nd, of leaves but little withered; 3rd, of
leaves medium-withered; and 4th, of leaves over-withered.

I arrived at the following results:--Unwithered or under-withered
leaves break in the rolling and give out large quantities of a light
green coloured juice during the same process. The Tea is much broken
and of a reddish grey colour. The liquor is very pale in colour,
cloudy, weak, soft, and tasteless.

Over-withered leaf on the other hand takes a good twist in the rolling,
gives out but little juice, which is of a thick kind, and of reddish
yellow colour. The tea is well twisted, “chubby” in appearance, and
blacker than ordinary. The liquor of an ordinary depth of colour,
clear, with a mawkish taste.

The medium-withered leaves make good Tea, but I found the withering
should be rather in excess of what is generally done to ensure
strength. I will show later to what extent I think leaf should be
withered.

The next point was rolling. I knew some planters rolled the leaf hard,
others lightly. That is, some rolled with force till much juice was
expressed, others with a light hand, allowing little or no juice to be
pressed out. Which was the better?

After many experiments I arrived at the following:--Hard rolling gives
darker coloured and stronger liquor than light rolling. Hard rolling
destroys Pekoe tips,[37] inasmuch as the juice expressed stains them
black.

Light rolled Tea has therefore many more Pekoe tips than hard rolled.

Hard rolled Tea is somewhat blacker than light rolled.

In all, therefore, but the point of Pekoe tips hard rolling is better.

The next question was, what is the advantage of repeated rolling? I
rolled twice, panning once between, _vide_ old plan, and found the Tea
as well made and as strong as that rolled three or four times. I then
decided to roll _no more_ than twice. The second time was, I _then_
thought, necessary, as I found the leaf of the roll opened in the pan,
and a second rolling was requisite to twist it again.

But what did panning do? I heard pans had been discontinued in some
gardens. In what way was panning an advantage? I made Tea, fermenting
it between the two rollings, but _not_ panning it, and it was
equally good. I tried again and again, but never could detect that
panning caused any difference to either the Tea, the liquor, or the
out-turn.[38] In short, though I never found panning did any harm, I
equally found it never did any good. Its use is, in fact, simply barren
of _all_ results.

I therefore dispensed with it. Having done so, why roll the second
time at all? I experimented, and found the second rolling as barren of
results as the panning.

I had now got rid of operations 3, 5, 6, 7, and 8 in the old plan. The
next was No. 9--“sunning.” I made Tea with and without it, and found as
follows:--

Sunning between the fermenting and firing processes has no effect
whatever on the liquor or the out-turn, but it makes the Tea rather
blacker, and as it drives off much of the moisture in the roll, the
firing process after it is shorter and does not consume so much
charcoal. What little effect therefore it has is good (for if not
continued too long, it does not make the Tea too black) and it is
economical. I therefore decided on retaining it.[39]

Next came the operations 10, 11, and 12, viz., “first firing, cooling
and crisping, and second firing.” Where these are done (and they are
done in some gardens now) the usual thing is to _half-fire_ the roll
the same afternoon and evening it is made, then allow it to “cool and
crisp” all night, and finish the firing next day. I tried this plan,
and also the plan I have now adopted, of doing the whole firing at one
time the same evening. I tried the experiment again and again, and
always found the Tea, the liquor, and the out-turn were the same in
both cases. In short, that the three operations did no more and no less
than the one. As the three entail extra labour and extra expense in
charcoal I abandoned them.

I thus reduced the twelve operations detailed to five, and naturally by
so doing much decreased the cost of manufacturing Tea. I in no way lay
claim to having devised this simplicity myself. Part had been done by
others before I even turned my attention to it, and I have done no more
than help with many to make the manufacture of Tea a simple process.

I was now convinced that (though I had still much to learn regarding
the said five processes) success was comprised therein, and that to
multiply them could not avail.

The next consideration is--What are the qualities desired in Tea to
enable it to command a good price at the public auctions either in
Calcutta or London? The brokers in these cases judge of the Tea first,
value it, and give their report and valuation to intending purchasers
and sellers. From what appearances and qualities do they judge?

They judge from three things, _first_, the Tea; _secondly_, the liquor;
_thirdly_, the out-turn.

_The Tea._--The colour should be black, but not a dead black, rather a
greyish black with a gloss on it. No red leaf should be mixed with it,
it should be all one colour. The Tea should be regular: that is, each
leaf should be about the same length, and should have a uniform close
twist, in all but “broken Teas.” (These latter are called “broken”
_because_ the leaf is more or less open and broken.) The Tea should
also be regular of _its kind_, that is, if Pekoe all Pekoe, if Congou
all Congou; for any stray leaves in a Tea of another kind, if even of
a _better_ kind or class, will reduce its value. In the higher class
of Teas, viz., Pekoes and broken Pekoes, the more Pekoe tips that are
present the higher, in consequence, will its price be.

_The Liquor._--In taste this should be strong, rasping, and pungent,
with, in the case of Pekoes, a “Pekoe flavour.” There are other words
used in the trade to particularise certain tastes, but the words
themselves would teach nothing. Tea tasting cannot be learnt from
books. _If_ the liquor is well flavoured, as a rule, the darker it is
in the cup the better. But to judge of Teas by the colour of the liquor
alone is impossible, for some high-class Teas have naturally a very
pale liquor.

_The Out-turn._--A good out-turn is generally indicative of a good Tea.
It should be all, or nearly all, one colour. No black (burnt) leaves
should appear in it. A greenish tinge in some of the leaves is not
objectionable, and is generally indicative of pungent liquor, but the
prevailing colour should be that of a bright new penny.

Every planter should be more or less of a Tea-taster, and should taste
his Teas daily. After a time (particularly if he gets other Teas to
taste against his own) he will learn to recognise, at all events, a
good as against a bad Tea, a strong as against a weak Tea, &c. No Tea
should be put away with the rest until it has been tasted. It may be
burnt or have other defects, not apparent till infused, and one day’s
bad Tea will bring down considerably the value of a whole bin of good
Tea.

The fancy, amongst brokers and dealers, for “Pekoe tips,” in all Pekoe
Teas, constitutes the _one_ great difficulty in Tea manufacture. If the
leaves which give “Pekoe tips” (see page 106) are separated from the
other leaves, and manufactured separately and differently, that is
rolled _very_ little and _very_ lightly, not allowed to ferment at all,
but sunned at once after rolling, and, if there is sun enough, finished
in the sun, otherwise by a very light and gradual heat--best placed
_above_ the drawers in the Dhole-house; if this is done, I say, these
will come out perfect “Pekoe tips” of a white colour, which is the best.

If _not_ separated from the other leaf, but manufactured with it, the
sap from the other leaves, expressed in the rolling, stains these said
leaves, which are covered with a fine white silk down, and makes them
black like all the rest of the Tea; the whole of which is then valued
lower _because_ there are no “Pekoe tips.”

Now, in the latter case the “Pekoe tips” are there all the same, only
they don’t _show_. The Tea is really just as good, in fact a shade
_better_, with black than with white or orange tips,[40] but it does
not sell so well, and as we cannot argue the brokers or dealers into
a rational view of the case, we must humour their fancy (they are
virtually our masters) and give them the Pekoe tips--_if we can_.

How are we to do it? The plan of picking these small leaves separately,
in order to manufacture them separately, does not answer; it is too
expensive; it diminishes the yield of a garden, and labour for it
fails. All this is shown at pages 107 and 130. Is there any other way?

It may be done during some periods of the season when there is not
leaf enough on the garden to employ all the leaf-pickers, by setting a
number of them to separate the said two leaves from the others _after_
the whole leaf is brought to the factory. This is expensive, but it
pays when there is labour to do it, for then the Teas can be made very
showy and rich with white Pekoe tips.

An ingenious planter, a Mr. McMeekin, in Cachar, invented a rolling
table with the object of separating the said leaves. It is constructed
of battens, and while rolling the leaf on it, many of the small leaves
fall through. The said table is now well known in Cachar, and is in use
in several gardens. I have tried it and find that it in a great measure
answers its object, but the objection to it is that the leaf _must_ be
rolled lightly, and lightly-rolled leaf, as observed, does not make
strong Tea.

The Pekoe tips may be, in a great measure, preserved by rolling _all_
the leaf lightly on a common table. But then again the Tea is weak, and
the plan will not give so many Pekoe tips as McMeekin’s table.

In short, in the present state of our knowledge, except by the hand
process (a tedious and expensive one for separating the leaf), strong
Teas and Pekoe tips are incompatible.

The difficulty is just where it was, and will so remain until dealers
give up asking for Pekoe tips (not a likely thing), or till a machine
is invented to separate quickly and cheaply the two said small leaves
from the others _after_ they have been all picked together. That such a
machine is possible I am certain, and the inventor would confer a boon
on the Tea interest far beyond the inventor of any other machine, for
all the other processes _can_ be done by hand without much expense,
this cannot.

I may here notice such machines and contrivances as exist for
cheapening the manufacture of Tea, or rather such as I know of.

Rolling-machines have for their object the doing away with hand labour
entirely for rolling the leaf. Kinmond’s rolling-machine is first on
the list, for it is the best yet invented.[41]

Kinmond’s consists of two circular wooden discs, the upper one moving
on the lower, which is stationary, with an eccentric motion. The
adjacent faces of the said discs are made rough by steps in the wood,
cut in lines diverging from the centre to the circumference, and over
these rough faces is nailed coarse canvas.

The leaf is placed between the discs and rolled by the motion
described. The lower disc is arranged by means of weights running
over pulleys, so that it shall press against the upper with any force
desired.

The motive power, as designed by the inventor, is either manual,
animal, or steam.

Mr. Kinmond showed me this machine, just after he had invented it, at
the Assam Company’s Plantations in Assam, and I have since seen it
working by manual and steam power. With the former it is quite useless,
for by no arrangement can sufficient or regular force enough be
applied. With the latter it does very well, and on a large garden which
will render the outlay for the machine and engine justifiable (the
former is, for such a simple machine, very expensive), it may probably
eventually prove an economy.

Not having seen it under animal power, I can give no positive opinion
as to how it would answer, but I see no reason why it should not do
well. I believe wind or water power might, on suitable sites, be easily
applied to it, and they would certainly be the cheapest of any.

Another rolling-machine was invented by a Mr. Gibbon, and a good deal
used in Cachar. I have never seen it.

Kinmond’s is, I believe, the best rolling-machine yet invented (though
it is fair to state I know no other except by report), but I do not
believe in any Tea rolling-machine superseding _entirely_ the necessity
of hand-rolling.[42] A rolling-machine may be, and is, very useful to
roll the leaves partly, that is, to break the cells, and bring the leaf
into that soft _mashy_ state that very little hand labour will finish
it. No rolling-machine yet invented can, I think, do more than this,
and it is, I think, doubtful if any will ever be invented that will do
more. Machines do not give the nice final twist which is obtained by
the hand. I was told lately that most of the gardens in Cachar that had
machines had dropped them and gone back to hand-rolling. I cannot help
thinking this is a mistake. They should use both, the hand-rolling for
the final part alone. Very few rolling-men would then suffice, with the
aid of the machine, to manufacture a large quantity of leaf.

I only know of one other Tea rolling-machine, which is Nelson’s. It
does not profess to do more than _prepare_ the green leaf for rolling,
which, as stated above, is, I think, all that any machine will ever do.
I have never seen it working, but it appears simple, being nothing more
than a mangle. The leaf is placed in bags, and then compressed under
rollers attached to a box, weighted with stones. The prospectus states,
it will prepare 80 lbs. green leaf in fifteen minutes, and that one
man can then finish as much of such prepared leaf in three minutes as
would occupy him twelve minutes if the same had not been prepared. I
see nothing unlikely in this. The machine, though inferior to Kinmond’s
in its arrangement, _ought_ to be cheap enough to bring it within the
reach of all.[43]

I have already spoken of one of McMeekin’s inventions. His
chest-of-drawers for firing Tea is, I think, superior to his batten
table. It is now so well known, and in such general use, that I
shall describe it very shortly. It is nothing more than a low
chest-of-drawers, or trays fitted in a frame one above the other, the
bottom of each tray being fine iron wire, so that the heat of the
charcoal, in the masonry receptacle over which it is placed, ascends
through all the drawers and thus dries or fires a large quantity of
“roll” at the same time. By the old plan, a single wicker sieve was
inserted inside a bamboo frame called a “dhole,” which was placed over
a charcoal fire made in a hole in the ground. On the sieve the roll was
placed, and all the heat, after passing through this _one_ sieve, was
wasted. Mr. McMeekin’s idea was to economise this heat by passing it
through several drawers.

Most planters use these drawers, and there is no doubt in the space
saved, and the economy of heat: it is a great step in advance over the
old barbarous method, where not only was the heat wasted after passing
through _one_ sieve, but a great deal was lost through the basket work
of the “dhole” itself.

Still I do not advocate four, still less five drawers one above the
other. I think the steam ascending from the lower drawers must, more
or less, injure the roll in the upper ones. I confine myself to two,
and even then in the top tray leave a small circular space vacant by
which the steam from the lower drawer can escape. I utilize the heat
that escapes, partially, by placing “dhallas” in tiers above, with roll
in them. These are supported by iron rods let into the wall, and are
useful not only for partly drying the roll, but also for withering leaf
when there is no sun.

Some planters have proposed to do away with charcoal altogether under
McMeekin’s drawers, supplying its place by hot air. The first point
in considering this invention is the question whether the fumes of
charcoal, as some assert, _are_ necessary to make good Tea. If they
are _not_ necessary (that is, if they produce no chemical effect on
the Tea, and therefore heat from wood devoid of smoke would do as
well) there can be no doubt such heat would be cheaper, and more under
command, by this or some other plan. Are then the fumes of charcoal
necessary?

I do not know that anyone can answer the query. I certainly cannot,
for I have never made Tea with any other agent than charcoal, and I
have never met with more than one planter who had. He said the Tea was
not good. Still it would, I think, require very careful and prolonged
experiments to establish the fact either way. Speaking theoretically,
as it _appears_, the only effect of charcoal is to drive all the
moisture out of the roll and thus make it Tea, I cannot but believe
other heat would do as well. It is, however, a question that only
experience can solve.[44]

I have now (four years since the above was written, and at the time
I am preparing the second edition of this essay) been for some time
employed on experiments with a view to settle the above question.
Whether I shall be able to devise a simple apparatus to effect the
manufacture of Tea without charcoal is doubtful, but I can, I think,
now safely affirm that the fumes of charcoal are _not_ necessary to
make Tea. On this point I am myself quite satisfied. The advantages
of making Tea with any fuel (wood, coal, or anything else) would be
numerous:--

1.--Economy.

2.--Absence of charcoal fumes.

3.--Less chance of fire in Tea Houses.

4.--Probably reduced temperature in Factories.

5.--Great saving of labour.

6.--Saving of fuel--for it takes much wood to make a given weight of
charcoal.

In addition to all the above, the wholesale destruction of forests that
now takes place in all Tea Districts, in order to supply the charcoal
for Tea, would be much lessened.[45]

I have seen a machine advertised for packing Tea, that is to say, for
so pressing it down that a large quantity shall go into a chest. I
have never seen the machine, and so cannot say how it works, but I do
not think such a machine at all necessary. By the mode of packing,
described at page 150, as much Tea as a chest will hold _with safety_
can be put into it. If more were forced in, the chest would probably
come to pieces in transit.

I see a sifting machine is now being advertised--“Jackson’s sifting
machine.” I have seen drawings of it, but not the machine itself. In
the one respect, that it is much larger than anything used hitherto, it
is more likely to succeed.

There is a machine for sifting and fanning Tea at one and the same
time. I know not who invented it. It is a simple winnowing machine
with sieves placed in front of the fan. By means of a rod and crank
attached to the axle of the revolving fan the sieves are made to shake
from side to side when the fanners are turned. The Tea is put into the
upper sieve, a coarse one, and passing successively through finer ones,
is thus sorted into different Teas. The open leaf at the same time is
blown out by the fan.

I purchased one, but I do not find it does the work well. Sifting Tea
is a nice process, and I did not find it sorted the Teas with any
nicety. I have taken out the sieves, and use it now only for fanning,
which it does very well, though no better than an apparatus which could
be constructed at one-third the cost.

I do not believe in _any_ present or future machine for sifting Tea,
inasmuch as it is an operation which, to be well done, has to be
continually varied. More will be said on this head further on.

I have now detailed shortly all the Tea machines or contrivances I
know, or have heard of, and I think there is plenty of room yet for
inventors.[46] The machine, as before observed, most to be desired is
one to separate the small Pekoe leaves from the others, ere the rolling
of the leaf is commenced. If such a machine existed, it would much
increase the value of all Indian Teas, and if the Agricultural and
Horticultural Society are inclined to offer a prize for any machine, it
should be this.

At the point where the separation should take place, the stalk is much
tenderer than elsewhere, and this led me to think a blow or concussion
on the mass of green leaf might effect the object. I attached a bow by
the centre to an immovable board, placed at right angles to the plane
of a table (like the back of a dressing table), and then, causing leaf
to drop from above, subjected it to sharp strokes from the string
of the bow. It effected the object partially, for many Pekoe ends
were detached, but it bruised and cut the other leaf too much also.
I believe a revolving barrel, with blunt but thin narrow iron plates
inside, which would strike the leaf placed within, as the barrel was
turned, would perhaps answer. I give the above idea for what it is
worth, for any inventive genius to improve on.

As it is impossible, as far as I can see, to construct any machine
which should _cut_ the stalk _only_ in the right place, _ergo_, I
believe some arrangement which would take advantage of the fact, that
the stalk is tenderer there than elsewhere, is the only one that could
answer.

Now to return to the manufacture of Tea. I will consider each of the
five operations detailed, which I believe are all that are necessary to
make good Tea, separately.

_Withering._--There are several tests to show when leaf is withered.
Fresh leaf squeezed in the hand, held near the ear, crackles, but no
sound should be heard from withered leaf. Again, fresh leaf, pressed
together in the palm of the hand, when released, springs back to nearly
its original bulk, but withered leaf, in like circumstances, retains
the shape into which it has been pressed. The stalk of withered leaf
will bend double without breaking, but fresh leaf stalks, if bent very
little, break. Practice, though, soon gives a test superior to all
these, viz., the feel of the leaf. Properly withered leaves are like
old rags to lay hold of, and no further test, after a time, than the
feel of the leaf is necessary.

The agents for withering leaf are sun, light, heat, and air. Of these
the most powerful is sun, for it combines all the others with it. Light
is a powerful agent, for if some leaf be placed in a partially dark
room, and some in a well-lighted verandah, the latter will wither in
half the time the former will take. If light and moderate ventilation
be present, heat is a rapid accessory to rapid withering.

There is often great difficulty in withering leaf in the rain. It
_can_ be withered in Tea pans, but “the out-turn” is then more or less
injured, for after infusion the out-turn comes out green instead of the
proper “new penny” colour. Withering in dholes is also objectionable
for the same reason, though if the heat is moderate the green effect is
less. It is further a long and tedious operation.

Space and light are the great wants for withering leaf in wet weather.
Bamboo mechans, tier above tier, should be constructed in every
available space. Large frames, covered with wire mesh, may also be made
(by means of weights running over pulleys) to run up to the roof of any
Tea building. The leaf withers well in such frames, for heat ascends,
and much heat is given out by dholes.

It signifies not though where leaf is spread as long as there is space
and light. Houses made of iron and glass would be far the best for
withering leaf, for, if well ventilated, all the necessary agents for
withering, detailed in the last page, would be present. I do not doubt
the day will come when these will be used, for properly withered leaf
is a necessity for good Tea.[47]

In dry weather, when leaf comes in from the garden, spread it thinly
anywhere and turn it once early in the night. It will generally be
withered and ready to roll next morning. If not quite ready, then put
it outside in the sun. Half an hour’s sunning will probably finish it.

In wet weather, if there is any sun when it comes in, or any time that
day, take advantage of the sun to wither the leaf _partly_, so much
that, with the after withering all night under cover, it will be ready
next morning. If not ready next morning, put it out in the sun, if
there is any, till it is ready.

In very wet and cloudy weather, when there is no sun and continual
rain, so that the leaf _cannot_ be put outside (for remember that
outside, when there is no sun, the light alone will wither it),
artificial withering of some kind must be resorted to. I have mentioned
the only means I know of for doing this.

As properly withered leaf is an important point in making good Tea, it
is well worth while to keep one or two men, according to the quantity
of leaf, for that work alone. They soon learn the best way to do it,
and if made answerable the leaf is properly ready for the rollers,
the object is generally attained. In this and every thing else in Tea
manufacture, give different men different departments, and make them
answerable. Much trouble to the manager, who should supervise all, and
much loss to the proprietor from bad Tea, will then be avoided.

_Rolling._--This is a simple operation enough when the men have got the
knack of it. Some planters advocate a circular motion of the hands when
rolling, under the impression it gives the leaf a better twist. Some
like rolling it forward, but bringing it back without letting it turn
during the backward motion. I believe in neither way, for it appears
to me to be rolled no better, or no worse, by these plans than by the
ordinary and quicker mode of simply rolling it _any way_. The forward
and backward motion is the simplest and quickest, and the way all
rollers adopt, who are given a certain quantity of leaf (say 30 lbs. a
fair amount) to roll for their day’s work. In this ordinary rolling the
ball in the hand, ’tis true, does not turn much in the backward motion,
for ’tis more or less _pulled_ back, but whether it turns or not does
not, I believe, signify the least.

Rolling in hot pans was formerly extensively practised. It is not much
done now. I have tried the plan, but found no advantage in it.

Rolling on coarse mats, placed on the floor, might be seen also. When
I visited the Assam Company’s gardens near Nazerah, in Assam, I saw it
done there. It is a great mistake. The coarse bamboo mat breaks the
leaf sadly, and much of the sap or juice from the leaf, which adds much
to the strength of the Tea, runs through the coarse mat, and is lost.

One and the principal reason why Indian Tea is stronger than Chinese is
that in India the sap or juice is generally retained, while in China it
is, strange to say, purposely wasted!

A strong immovable smooth table, with the planks of which it is formed
well joined together, so that no apertures exist for the juice of the
leaf to run through, is the best thing to roll on. If covered with
a fine seetul pattie mat, nailed down over the edges of the table,
a still greater security is given against the loss of any sap, and
I believe the slightly rough surface of the mat enables the leaf to
roll better. An edging of wood one inch above the surface of the table
should be screwed on to the edges over the mat, if there is one, to
prevent leaf falling off.

The leaf is rolled by a line of men on each side of such a table (4-1/2
feet is a good width for it) passing up from man to man, from the
bottom of the table to the top. The passage of each handful of roll
from man to man is regulated by the man at the end, who, when the roll
in his hand is ready (that is, rolled enough), forms it into a tight
compressed ball (a truncated shape is the most convenient) and puts it
away on an adjacent stand. When he does this, the roll each man has
passes up one step.

The roll is ready to make up into a ball, when it is in a soft _mashy_
state, and when in the act of rolling it gives out juice freely. None
of this juice must be lost, it must be mopped up into the roll, again
and again in its passage up the table, and finally into the ball, when
made up.

There will be some coarse leaves in the roll which cannot be twisted.
These, if left, would give much red leaf in the Tea. They should be
picked out by, say, the third or fourth man from the head of the table,
for it is only when the leaf has been partly rolled that they show. The
man who picks out the coarse leaf should not roll at all. He should
spread the roll, and pick out as much as he can, between the time of
receiving and passing it on. In no case allow roll to accumulate by
him, for if so kept it hardens and dries, and gives extra work to the
last rollers to bring it into the mashy state again. Besides which
I rather think, any such lengthened stoppage in the rolling helps
to destroy Pekoe ends, and is certainly injurious to the perfect
after-fermentation, inasmuch as it (the fermentation) partly takes
place then.

This finishes the rolling process. Each man as stated can do 30 lbs.,
but there is further work for him to be now described.

_Fermenting._--The balls accumulated are allowed to stand until
fermented. I look on this being done to the right extent and no more,
as perhaps the most important point in the whole manufacture.

Some planters collect the roll after rolling in a basket, and there
let it ferment, instead of making it up into balls for that purpose
as described. I much prefer the ball system for the following
reasons:--When a quantity is put into a basket together and allowed
to ferment a certain time, what was put in first is naturally more
fermented than what was put in last, the former probably over, the
latter under-done. The balls, on the contrary, can be each taken in
succession _in the order they were laid on the table_, and thus each
receive the same amount of fermentation. I think further the twist in
the leaf is better preserved by the ball plan, and also that a large
quantity in a basket is apt to ferment too much in the centre.

It is impossible to describe, so that practical use shall be made of
it, _when_ the balls are sufficiently fermented. The outside of the
ball is no good criterion. It varies much in colour, affected by the
extent the leaf was withered.[48] You must judge by the inside.

Perhaps as good a rule as any is that half the twisted leaves inside
shall be a rusty red, half of them green. Practice alone, however, will
enable you to pronounce when the balls are properly fermented. There is
no time to be fixed for it. The process is quicker in warm than cool
weather.

The fermentation should be stopped in _each_ ball just at the right
time. Great exactitude in this is all-important, and therefore, as I
say, the balls should be taken in rotation as they were laid down.

The fermentation is stopped by breaking up the ball. The roll is spread
out _very_ thin, and at the same time any remaining coarse leaves are
picked out.

This concludes the fermenting process.

_Sunning._--The roll is then without _any_ delay put out in the sun,
spread _very_ thin on dhallas or mats. When it has become blackish in
colour it is collected and re-spread, so that the whole of it shall be
affected by the sun. With bright sunshine, an hour or even less suns it
sufficiently. It is then at once placed in the dholes, which must be
all ready to receive it.

If the weather is wet, it must, _directly_ the balls are broken up, and
the coarse leaf is picked out, be sent to the dholes. This is the only
plan in wet weather, but the best Tea is made in fine weather.

_Firing or Dholing._--In the case of wet weather, unless you have very
many dholes, fresh roll will come in long before the first is finished.
The only plan in this case is to half do it. Half-fired the roll does
not injure with _any_ delay, but even half an hour’s delay, between
breaking up the balls and commencing to drive off the moisture, is
hurtful.

In any but wet weather necessitating it the roll can be fired at one
time, that is, not removed from the drawer until it has become Tea.

The roll in each drawer must be shaken up and re-spread two or three
times, in the process of firing. The drawer must be taken off the fire
to do this, or some of the roll would fall through into the fire, and
the smoke thus engendered would be hurtful. If the lowest drawer is
made to slide in and out a framework covered with zinc should be made
to run into a groove below it, and this zinc protector should be always
run in before the lower drawer is moved. This is part of Mr. McMeekin’s
invention, and is very necessary to prevent roll from the lowest drawer
falling into the fire when it (the lower drawer) is moved.

The roll remains in the drawers, subject to the heat of the charcoal
below, until it is quite dry and crisp. Any piece then taken between
the fingers should break with the slightest attempt to bend it.

The manufacture is now completed. The roll has become Tea.

All the above operations should be carefully conducted, but I believe
the secret of good Tea consists simply in, _first_, stopping the
fermentation at the right moment; and, _secondly_, in commencing to
drive off the moisture immediately after.

I do not say that the manufacture here detailed may not be improved
upon later, but I do say that in the results of economy, strong liquor,
and well twisted leaf, its results are very satisfactory, and not
surpassed by any other mode at present in vogue. I do not pretend that
it will give Teas rich in Pekoe tips. To attain this, light rolling as
shown must be resorted to, but just as far as Pekoe tips are procured
so far must strength be sacrificed. Until the small Pekoe leaves can be
detached and manufactured separately, this must always be the case.

From the Tea made as described by sifting and sorting, all the ordinary
black Teas of commerce, as detailed at page 137, can be produced,
excepting “Flowery Pekoe.”

To make Flowery Pekoe the closed bud and the one open leaf of the shoot
are alone taken, and these are manufactured alone. It does not, as a
rule, pay to make this Tea at all, though it fetches a long price. It
does not pay for the following reasons:--

1. After the head of the flush is taken the pickers that follow do not
readily recognise the remainder of the shoot, and consequently omit to
pick many of them. A heavy loss in the yield is thus entailed.

2. The after Teas, made without these small leaves, are very inferior,
as they are much weaker, and totally devoid of Pekoe tips.

3. The labour, and _ergo_ the expense of picking the flush, is double.

The manufacture of Flowery Pekoe is simple enough. When the two leaves
from each shoot of which it is made are collected they are exposed to
the sun, spread out very thin, until they have well shrivelled. They
are then placed over small and slow charcoal fires, and so roasted very
slowly. If the above is well done, the Pekoe tips (and there is little
else) come out a whitish orange colour. The whiter they are the better.
If the leaf is rolled _very lightly_ by the hand _before_ sunning, the
liquor will be darker and stronger, but the colour of the tips will not
be so good.

Flowery Pekoe is quite a fancy Tea, and for the reasons given above it
can never pay to make it.


_Green Tea._

The pans for this should be 2ft. 9in. diameter and 11in. in depth. They
should be thick pans, which will not, therefore, cool quickly. Many
are required for this manufacture, four or five for every maund of Tea
to be made daily. They should be set up in a sloping position, and the
arrangement of the fireplaces such that the wood to burn under them
can be put in through apertures leading into the verandah. One chimney
will do for every two pans, and it should be built high so as to give
a good draught, for hot fires are necessary.

Flat-bladed sticks are used to stir both the leaf and the Tea in the
pans, for the hand cannot bear the heat.

The men when working the Tea in the pans should have high stools to sit
on, for it is a nine hours’ job.

The bags in which “the roll” is placed at night should be made of No.
3 canvas, 2 feet long and 1 foot broad.

I will now detail the manufacture.

To make Green Tea the leaf must be brought in twice in the day. What
comes in at one o’clock is partly made the same day. The evening leaf
is left till the following morning, laying it thick (say 6 inches), so
that it will _not_ wither. But if the one o’clock or the evening leaf
comes in wet, they must both be dried, the former _before_ being put
into the pans, the latter _before_ being laid out for the night.

The manufacture thus begins twice daily, viz., morning and one o’clock,
but “the roll” of both these is treated together up to the time “the
roll” is ready to place in the bags.

The leaf having no moisture in it is placed first in hot pans, at
a temperature of say 160°, and stirred with sticks for about seven
minutes, until it becomes moist and sticky. It is then too hot to hold
in the hand.

It is then rolled for two or three minutes on a table until it gets a
little twisted.

Then lay it out on dhallas in the sun (say 2 inches thick) for about
three hours, and roll it thrice during that time, always in the sun. It
is ready to roll each time when “the roll” has become blackish on the
surface. It is not rolled more than three minutes each time, and then
spread out as before. If you put on a proper number of men to do this
they do each dhalla in succession, and when they have done the last,
“the roll” in the first dhalla will be blackish on the surface again,
and ready to roll again.

When three rollings are done, the roll should have a good twist on it.

It is then placed in the pans, at the same heat as before, and worked
with sticks as before for two or three minutes, until it becomes too
hot to hold.

It is then stuffed, as tight as it _can_ be stuffed, into the bags
described above, putting as much into each bag as you can possibly get
it to hold. The mouth is then tied up and the bag beaten with a flat
heavy stick to consolidate the mass inside, and so it is left for the
night.

Next morning it is taken out of the bags, and worked with the flat
sticks as before in the pans for nine hours without intermission. The
temperature 160° at first down to 120° at the last.

During and owing to this last process the green colour is produced.[49]
It is worked quicker and quicker as the hours pass.

The following are the kinds of Tea into which it is best sorted:--

  1. Ends         }
  2. Young Hyson  }
  3. Hyson        }   The relative value is in the order
  4. Gunpowder    } in which they are numbered.
  5. Dust         }
  6. Imperial     }

The sorting of Green Tea is a nicer operation, and takes twice as long
as sorting Black Tea.

If there are pans enough, and the work is well arranged, there should
be no night-work with Green Tea, for all should be over by 5 P.M.;
whereas with Black Tea night-work is generally a necessity.

The price obtained for Green Tea is more dependent on its _appearance_
than in the case of Black.

It is not easy to make Black and Green Tea in the same factory.

Green Tea, if well made, pays much better than Black Tea; and, as
before observed, I think all gardens with Chinese plants should adopt
the manufacture.[50] When once the building is fitted for it, and the
routine established, the Green Tea manufacture is always preferred by
those who have tried both.

The Hybrid plant makes the best Black, the Chinese the best Green Tea.


FOOTNOTES:

[36] In describing manufacture I shall call the leaf brought in “Leaf,”
until it enters on the rolling process; from that time until the drying
over charcoal is concluded, “Roll;” and after that, “Tea.”

[37] Pekoe tips are the whitish or orange-coloured ends that may be
seen in Pekoe Tea. See pages 105, 106, and 116.

[38] The out-turn consists of the Tea leaves after infusion.

[39] At the end of the season, however, sunning has more than the above
effect. It then makes the Tea “Chubby” in form, of a reddish colour,
and improves the strength of the liquor.

[40] It is better, because the “tips” having been hard-rolled give
stronger liquor.

[41] It _was_ the best, but is superseded by a new rolling-machine
(Jackson’s) I have seen quite lately.

Note to Third Edition.--Jackson’s rolling-machine, by a late Calcutta
legal decision, is declared to be simply Kinmond’s, with alterations.
As Jackson is now prohibited from selling his machines, I presume the
two inventors will come to some understanding as to the alterations,
which are most certainly improvements.

[42] I had not seen Jackson’s machine when I thought as above.

[43] Unfortunately it is not. It is advertised at Rs. 300, with a
yearly royalty of Rs. 50 the first year and 20 after. The royalty
should be dropped, and the machine sold for Rs. 150, which would give
the inventor a good profit.

[44] Note to 3rd edition.--It is a question no longer. Many besides
myself have now proved that charcoal fumes are in no way necessary.

[45] See this subject further discussed in the Addenda.

[46] I now believe Jackson’s rolling-machine, previously alluded to,
will finish the rolling entirely.

[47] Note to 3rd edition--I am now sending out the glass necessary for
a glass withering house to be erected on the garden just finished in
the Western Dooars.

[48] The more the leaf is withered the thicker in consistency and the
smaller in quantity the juice that exudes, as also the yellower in
colour. Further, the more the leaf is withered the darker the outside
of the balls. Bright rusty red is the colour produced with moderately
withered leaf; very dark greenish red with much withered leaf.

[49] Much Green Tea is coloured, but none from India has been so
treated.

[50] Note to Third Edition.--Since this was written Green Teas have
gone down considerably in value. They are still much used in America,
but in Great Britain there is but little demand for them.



CHAPTER XXV.

SIFTING AND SORTING.


Sifting is a very important item in the manufacture of Tea. Careful
and judicious sifting, as contrasted with the reverse, may make a
difference of two or three annas a lb. in the sale of Teas.

I was shown some Tea quite lately which, as regards “liquor,” was
valued by the brokers at Re. 1-3 per lb., but the “Tea” at only 14
annas! This was entirely owing to faulty sifting and sorting.

I don’t believe in _any_ machine for Tea sifting, simply because it is
not a regular process.[51] For example, you cannot say that, to make
Pekoe, you must first use one sieve, then another, and so on. The sizes
of sieves to be used, and the order in which they are to be used, will
vary continually, as both are decided by varying causes, viz., the
comparative fineness or coarseness of the Tea made daily, the greater
or less presence of red leaf in it, and (because Tea varies much during
the season, and gets coarse towards the end) by the time of the year.
These points all necessitate changes in the sizes, and the order of the
sieves.

’Tis true sieves might be changed in a machine as required, but the
only machine that could even pretend to save labour would be one in
which all the sieves were arranged one below the other, and thus the
Tea would fall through each alternately, the motion being common to
all. But this won’t do for Tea sifting. Judgment must be used to
decide _the length of time_ each sieve is to be shaken; further, with
_how much motion_ it shall be shaken, &c., &c. But this is simply
impossible with any machine, though all necessary to sift Tea well.

The cost of Tea sifting by hand (see page 161) is not eight annas _per
maund_, including picking out red leaf, which _must_ be hand-work. Good
and bad sifting will affect the value three annas per lb. or Rs. 15 per
maund!

With all parts of Tea manufacture it is well to employ the same men
continually in each department, but above all, perhaps, should this be
done in Tea sifting. A good sifter is a valuable man. He knows each
kind of Tea by name; he knows what sieves to use, and the order in
which to use them for each Tea; what the effect a larger or smaller
mesh will have on each kind, &c., &c. In fact, he knows much more of
the _practical_ part of sifting than his master can, though the latter
is, probably, a better judge how far the Teas are perfect when made.

Tea sieves are of two kinds, both round. One made of brass wire, with
wooden sides, 3-1/2 inches high, the other cane, with bamboo sides,
1-1/4 inches high only. The latter are called “Chinese sieves,” and
though the brass ones are used in many places, there is no possible
comparison between them, for the labour required in the use of the
brass ones is much greater, and the results, as regards well sorted
Tea, much better with the Chinese.

Both kinds are numbered according to the number of orifices in one
linear inch. Thus a No. 6 sieve has six orifices to the inch in both;
but in the brass kind, a No. 6 has six orifices _including_ the wire;
in the Chinese kind, the cane between each aperture is _not_ included
in the measure. Thus the orifice in a No. 6 Chinese sieve is exactly
1/6th of an inch square, but somewhat less in a brass sieve.

As I well know brass sieves cannot remain in favour after the others
have been only once tried, I shall confine my directions to the Chinese
kind.[52]

I practise, and I advise, Tea to be sifted daily. The Tea made one day,
sifted the day after, and in fact stored away in the bins ready sifted.
I find it is more carefully done this way, for by the other plan a
larger quantity being done at once by several men, they cannot, from
want of practice, be expert. But by the daily plan one, two, or three
men, as necessary, can always be kept on the work, and consequently
they learn and do it well.

To sift the following, Chinese sieves are required; and if daily
sifting is resorted to, they will be found ample for any ordinary-sized
garden:--

  4 of No. 4
  6 of No. 6
  6 of No. 7
  9 of No. 9
  9 of No. 10
  6 of No. 12
  4 of No. 16.

Previous to sifting all red leaf should be picked out of the Tea. This,
as stated under the head of “Manufacture,” should be done twice before
the “roll” is fired; but towards the end of the season especially, some
will still remain in the made Tea, and this must be carefully separated.

From what I have said it is evident that no rules can be laid down as
to what sieves to employ to get out certain Teas. Only practice can
teach this.

Further, practice can only enable you to judge in a Tea broker’s
point of view of different classes of Tea. This essay would,
however, be incomplete did it not contain a description of these.
Such a description has been ably given by Mr. J. H. Haworth in his
“Information and Advice for the Tea Planter from the English Market”
(_Journal, A. & H. Society of India, Vol. XIV._), and, as his knowledge
on the subject is far in advance of mine, and consequently more to the
point than any description I could give, I will close this chapter with
the following extract from his valuable pamphlet, and trust he will
excuse my doing so:--


_Of the Different Classes of Tea._

  Teas are arranged in various classes according to the size, make,
  and colour of the leaf. I treat first and principally of the Black
  descriptions, as Green Teas are manufactured in only a few of the
  Tea-growing districts of India.

  The following classes come under the name of Black Tea:--

  Flowery Pekoe.                        { Broken Pekoe.
  Orange Pekoe.                         { Pekoe Dust.
  Pekoe.             The various broken { Broken Mixed Tea.
  Pekoe Souchong.      kinds, viz.:--   { Broken Souchong.
  Souchong.                             { Broken Leaf.
  Congou.                               { Fannings.
  Bohea.                                { Dust.

  We occasionally meet with other names, but they are generally
  original, and ought not to be encouraged, as a few simple terms like
  the above are sufficiently comprehensive to describe all classes
  manufactured.

  Perhaps before entering into a detailed description of the various
  classes it will be well to explain the term “Pekoe” (pronounced
  Pek-oh), which as we see occurs in so many of the names above quoted.
  It is said to be derived from the Chinese words “Pak Ho,” which are
  said to signify white down. The raw material constituting Pekoe when
  manufactured is the young bud just shooting forth, or the young
  leaf just expanded, which on minute examination will be found to be
  covered with a whitish velvety down. On firing these young leaves,
  the down simply undergoes a slight change in colour to grey or
  greyish yellow, sometimes as far as a yellowish orange tint.

  When the prepared Tea consists entirely of greyish or greenish
  greyish Pekoe, with no or very little dark leaf mixed, it is called
  Flowery Pekoe.

  Flowery Pekoe is picked from the shrub entirely separate from the
  other descriptions of Tea, only the buds and young leaves being
  taken. In the preparation it is not subjected so severely to the
  action of heat as the other classes of Tea, and generally preserves
  a uniform greenish grey or silvery grey tint. Its strength in liquor
  is very great, flavour more approaching that of Green Teas, but
  infinitely superior, having the strength and astringency, without the
  bitterness, of the green descriptions. The liquor is pale, similar to
  that of Green Tea, and the infused leaf is of a uniform green hue. In
  many instances, where too much heat has been employed, we find dark
  leaves intermixed, and the prevailing colour, green, is sprinkled
  with leaves of a salmony brown tinge, which is the proper colour for
  the out-turn of any other ordinary black leaf Tea. A very common
  mistake is to call an ordinary Pekoe, that may contain an extra
  amount of Pekoe ends, Flowery Pekoe. When this class of Tea is strong
  and of Flowery Pekoe flavour, it is called by the trade a Pekoe of
  Flowery Pekoe kind. In England Flowery Pekoe sells, as a rule, from
  4_s._ 6_d._ to 6_s._ 6_d._ per lb. One parcel has sold as high as
  7_s._ 6_d._

  By many people the expediency of making Flowery Pekoe is much
  doubted. The true Flowery Pekoe leaf is the one undeveloped bud at
  the end of each twig. To pick this alone, without any ordinary Pekoe
  leaves, involves a great deal of trouble and expense, and I think
  though the Flowery Pekoe be very valuable, that the account would
  hardly balance when we consider the deterioration of the Pekoe by the
  abstraction of the young leaves.

  The ordinary Pekoe is a Tea of blackish or greyish blackish aspect,
  but dotted over with greyish or yellowish leaves which, on close
  inspection, will be found to possess the downy appearance which
  gives the name to Pekoe. In general we do not find the whole leaf
  covered with down, but only part of it, which in its growth has been
  developed later than the other parts. These are called by the trade
  “Pekoe ends” when very small Pekoe tips. A Pekoe is generally of good
  to fine flavour, and very strong, and its liquor dark. Its value is
  from 2_s._ 9_d._ to 3_s._ 8_d._ per lb.

  When the Pekoe ends are of yellowish or orange hue, and the leaf is
  very small and even, the Tea is called Orange Pekoe. In flavour it is
  much the same as an ordinary Pekoe, and many growers do not separate
  the two varieties, but send them away in the finished state mixed
  together. Its value is from 2_d._ to 4_d._ per lb. more than Pekoe.

  The term Pekoe Souchong is generally applied to a Pekoe that is
  deficient in Pekoe ends, or to a bold, Souchong class leaf with a
  few ends mixed. We often meet with it applied to an unassorted Tea,
  including perhaps Souchong, Congou, a few Pekoe ends, and some broken
  leaves. Prices range from 2_s._ 3_d._ to 2_s._ 10_d._

  The name of Broken Pekoe indicates at once what class of Tea it is,
  namely, Pekoe which has been broken in the manipulation or otherwise.
  It possesses the strength and fine flavour of a full leaf of Pekoe,
  being therefore only inferior to it in point of leaf. In value it is
  very little inferior to Pekoe, sometimes as valuable, or even more
  so, as owing to the frangibility of the tender Pekoe ends, they are
  sometimes broken off in very large quantity, thus adding to the value
  of the broken Tea, though at the same time deteriorating the Pekoe.
  Prices from 2_s._ 6_d._ to 3_s._ 4_d._

  Pekoe dust is still smaller broken, so small in fact as actually to
  resemble dust. It is of great strength, though often not pure in
  flavour, as frequently any dust or sweepings from other Tea is mixed
  with it to make the lot larger. The price of Pekoe dust may range
  from 1_s._ 6_d._ to 2_s._ 8_d._

  A Tea only slightly broken is often called by the planter Pekoe
  Dust; again an Orange Pekoe is often called Broken Pekoe, and the
  converse. A knowledge of the signification of these and other terms
  would teach the grower to be very careful in marking his Teas, as the
  nomenclature influences to a great extent the sale in the home market.

  Having described the finer Teas, we now come to the consideration of
  the classes of Tea which form the bulk of the manufacture of a garden.

  Souchong may be taken as the medium quality, and when experience and
  skilled labour are employed in the manufacture as the bulk of the
  produce of an estate. The qualifications for being comprehended under
  this term are just simply an even, straight, or slightly curved leaf,
  in length varying say from 1/2 inch to 1-1/2 inch. It has not the
  deep strength of Pekoe, but is generally of good flavour and of fair
  strength. The prices of Souchong are from 1_s._ 10_d._ to 2_s._ 8_d._

  Congou comes next. It may be either a leaf of Souchong kind, but too
  large to come under that class, or though of smallish-sized leaf,
  too unevenly made, or too much curled (so as to resemble little
  balls) to be so classified. The flavour is much the same as that of
  Souchong, but the Tea has not so much strength. Some of the lower and
  large leaf kinds may be only worth perhaps from 1_s._ 3_d._ to 1_s._
  6_d._, whereas the finer qualities sell as high as 2_s._ to 2_s._
  3_d._ per lb.

  Bohea is again lower than a Congou. It may be either of too large
  a leaf to be called Congou, or, as is generally the case, it may
  consist principally of old leaf, which on being fired does not attain
  the greyish blackish colour which is so desirable for all the black
  leaf kinds except Flowery Pekoe, but remains of a brownish or even
  pale yellowish hue. It has scarcely any strength, and is generally
  of coarse flavour, sometimes not, but is never of much value unless
  of _Namuna_ kind (a term which will be described hereafter). We may
  quote prices at from 3_d._ to 1_s._ 2_d._ per lb.

  We now come to the broken descriptions of these middle and lower
  classes of Tea.

  Broken Mixed Tea is, as its name imports, a mixture of the various
  kinds of Tea broken. It may have a very wide range, include some of
  the lower classes or approach Broken Pekoe in character and value,
  but the kind usually thus named is a Tea worth from 1_s._ 8_d._ to
  2_s._ 6_d._, generally of a blackish aspect, and containing a few
  Pekoe ends.

  The term Broken Souchong is commonly and appropriately applied to a
  Tea which, though broken, has some approach to a full leaf, and that
  of the even Souchong character. Its value may vary, say from 1_s._
  6_d._ to 2_s._ 2_d._

  Broken leaf is a term of great comprehensiveness, but generally is
  used to signify a Tea worth from 8_d._ to 1_s._ 1_d._ per lb. It
  may be of a brownish, brownish blackish, or blackish colour. Its
  strength is seldom great, but its flavour may be fair or good, but in
  the lower qualities it is generally poor, thin, or coarse. It would
  be better to employ this term only as a general name of Broken Tea,
  and not to use it to signify any particular class, as it is very
  indefinite.

  Fannings is similar in colour and class of leaf to broken leaf as
  described above; in value also much the same, perhaps on the average
  a little lower. I suppose, in most cases, the mode of its separation
  from the other classes of Tea is, as its name implies, by fanning.

  Dust is a very small broken Tea, so small, in fact, as to approach
  the minuteness of actual dust. It is often very coarse, or “earthy”
  in flavour, owing perhaps to sweepings and dust having become mixed
  with it. Its value is from 6_d._ to 1_s._ 6_d._ In any Tea of this
  class worth more than these quotations, a few Pekoe ends or tips
  will generally be found, which bring it under the name of Pekoe Dust.

  We will now look at Black Teas in a body, and point out what is
  desirable and what is objectionable in them.

  We have seen that all Teas which contain Pekoe fetch higher prices
  than others, consequently we infer that Pekoe is a desideratum. If
  we glance at the descriptions of the various classes of Tea which
  have been given above, we shall find that it is an element of
  strength and good flavour. I do not mean to say that any Pekoe is
  stronger or of better flavour than any Tea which does not contain
  Pekoe, as the soil, the climate, the cultivation, the manufacture,
  and various other causes, may influence the strength and flavour of
  different Teas; but, as a rule, in Teas that are produced under the
  same circumstances, the classes containing Pekoe are stronger and of
  better flavour than those without it.

  There is another class of Tea which I have not yet described that
  possesses very great strength and very fine flavour. This is the
  class known as the “Namuna” kind. All readers of these pages who
  have been connected with India any time will recognise the word,[53]
  though they may not quite see how it comes to occupy the position
  in which we consider it. It is said that its first application in
  this manner arose from a planter having sent to England some sample
  boxes of Tea with the ticket “namuna” on them. These Teas happened
  to be of the peculiar description which now goes by that name, and
  which I proceed to describe. The London brokers have always since
  then applied the name “namuna” to this class of Tea. The leaf may
  have, perhaps, the ordinary greyish blackish aspect, with generally
  a greenish tinge. In the pot it produces a very pale liquor, but
  on tasting it its quality belies the poor thin appearance of the
  infusion. It is very strong, stronger by far than ordinary Pekoe; in
  flavour, say, about half-way between a Flowery Pekoe and a Green Tea,
  quite distinct from the Flowery Pekoe flavour, possessing somewhat
  of the rasping bitterness of the Green Tea class with the flavour a
  little refined. The out-turn is generally green, sometimes has some
  brownish leaves mixed. Any of the black leaf Teas may be of this
  Class, from the Pekoe to the lowest dust, and all throughout the
  scale, if the flavour be distinct and pure, may have their value
  enhanced from 4_d._ to 10_d._ per lb.

  Similar in every respect, except one, is the Oolong kind. The one
  wanting quality is the strength, sometimes, by-the-bye, the flavour
  is a little different. It may have the greenish, greyish blackish
  leaf (though generally the green leaves are distinct from black ones,
  the Tea thus being composed of greyish blackish leaves with a few
  green ones intermixed), always has the pale liquor, generally the
  greenish infused leaf; but sometimes it is sadly intermingled with
  black leaves, as it is a Tea whose flavour is frequently burnt out,
  though its weakness and green appearance are no doubt often caused
  by deficient firing. Teas of this kind on the average sell below the
  ordinarily-flavoured Teas of the same class of leaf.

  In Teas of ordinary flavour the following rules hold good:--The
  darker the liquor the stronger the Tea, and the nearer the approach
  of the colour of the infused leaf to a uniform salmony brown, the
  purer the flavour. Whenever we see any black leaves mixed with
  it (the out-turn) the Tea has been over-fired, and we may either
  expect to find the strength burnt out of it, or else to find it
  marred by having a burnt or smoky flavour incorporated with it.
  When you come across an altogether black or dirty brown out-turn,
  you may be certain of pale liquor containing little or no strength
  and no flavour to speak of, unless sometimes it be sour. This is a
  quality which I shall now touch upon, and regret that I cannot with
  any certainty give any reliable information whereby the planter
  may guard against this greatest of faults. It may have various
  grades,--slightly sourish, sourish, and sour, depreciating the value
  of the Tea, say, from 3_d._ to 1_s._ 6_d._ per lb. The flavour of a
  sour Tea is hardly capable of description. It is not so acid as sour
  milk, in fact, not acid at all, rather a sweet flavour than otherwise
  being blended with the sourness. It is extremely unpleasant in its
  more developed grades, and cannot be easily understood except by
  actual tasting. To the uninitiated this fault is only perceptible in
  the more strongly marked instances, but to one of the trade the least
  tendency to it not only condemns the parcel at once, but also causes
  him to suspect any other lots made at the same or any other time by
  the same grower, and it is a curious but unaccountable fact that some
  two or three gardens (or growers?) almost always produce Teas having
  this fault. I will not cite all the different explanations that have
  been offered on this subject; I will simply quote the one which
  seems to have gained most ground, and leave those more competent than
  myself to express any opinion on the subject. The cause assigned to
  which I refer is that the Tea leaf after being picked is allowed to
  remain too long in the raw state before being fired, during which
  time it undergoes a process of fermentation; some then say that
  this causes sourness, while others maintain that the fermentation
  is absolutely necessary for the production of a Black Tea. The fact
  that we never meet with sourness in a Green Tea, one feature in the
  preparation of which being that it is fired almost immediately on
  being gathered, goes to corroborate this view.

  Burntness I have already referred to. As I said before, it may either
  destroy the strength and flavour altogether, or sometimes, without
  destroying the strength, add an unpleasant burnt flavour to it.
  When the Tea has the flavour of smoke about it, it is called smoky
  or smoky burnt. By being burnt, a Tea may be deteriorated in value,
  say from 2_d._ to 1_s._ per lb. The symptoms of burntness are a dead
  black leaf (as opposed to the greatly-desired greyish, blackish
  colour) having a burnt smell which often entirely neutralises the
  natural aroma of the Tea. In looking over a broker’s character of
  a parcel of Teas, you may occasionally meet with the terms “fresh
  burnt,” “brisk burnt,” or “malty burnt.” These phrases do not carry
  a condemnatory meaning with them. The meaning of the word burnt, as
  used here, would be better expressed by the term “fired.” The term
  “malty” means of full rich flavour, perhaps from the aroma of this
  class of Tea resembling somewhat that of malt. Teas of the three
  above descriptions, you may have noticed, often fetch very good
  prices. The meaning of the word “full,” applied to a liquor, is
  hardly appreciable except by tasting. It does not signify strength
  or flavour, but is opposed to thinness. A Green Tea may be strong or
  of good flavour, but its liquor is never full. Fulness is generally
  characterised by a dark liquor. The quality known as “body” in a wine
  is somewhat akin to fulness in a Tea. We speak of a “full” leaf Tea
  in contradistinction to a broken leaf. “Chaffy” is generally used in
  connection with Bohea and other brown leaf classes of Tea. A light
  (in weight) brown, open or flat leaf, in fact, one resembling chaff,
  would be called chaffy. The lower classes of Tea, especially the
  dusts, are often described as “earthy” in flavour. By this a coarse
  low flavour is understood, perhaps often caused by the admixture of
  real dust.

  When the make of a Tea is spoken of as a “well made,” “fairly made,”
  &c., leaf, the effect of the manipulation or rolling is referred
  to. We may have a “well made even,” or a “well made mixed large and
  small,” leaf. We may have a “straight” or “curled,” or, as the latter
  is generally expressed when applied to a large leaf Tea, “twisted”
  leaf. It may be “flattish made,” indicating that though the leaf is
  not open it wears a flattish aspect, or it may be open, which betrays
  a want of sufficient or skilful manipulation. A “wiry” leaf is small,
  perfectly rolled, and very thin (in diameter), generally rather
  curled, so as in fact to resemble small pieces of bent wire. It
  will be seen at once that only the finer Teas can have a wiry leaf,
  principally the Orange Pekoes and Pekoes. Sometimes we meet with a
  fine Souchong that may be thus described.


_Green Teas._

  As in the North-west Provinces Green Teas form the bulk of the
  produce, it will be well to give a short description of them, though
  the tenor of my remarks below will show the general opinion as to the
  desirability of making them.[54]

  Gunpowder is the most valuable description, its price ranging from
  2_s._ 8_d._ to 3_s._ 8_d._ per lb. Instead of possessing the long and
  thin finished leaf, which is the desideratum of Black Teas, it is
  rolled into little balls more or less round, varying from one-eighth
  to one-quarter of an inch in diameter. Sometimes it is not altogether
  composed of round leaf, but has some long leaf mixed.

  When the Tea is of the shape of Gunpowder, but is larger than the
  size above quoted, it is called Imperial. Prices of Imperial are from
  10_d._ to 2_s._ 6_d._

  Amongst Green Teas Hyson may be taken as the parallel of Souchong
  of the black leaf descriptions. Undoubtedly there is often much
  young Pekoe leaf in it, but all chance of _discriminating_ it in the
  finished leaf is done away with by the change in colour. Hysons sell
  from 1_s._ 2_d._ to 3_s._ 6_d._

  Young Hyson is smaller than Hyson, occasionally slightly broken. It
  fetches from 7_d._ to 2_s._ 6_d._

  Hyson skin consists of the bold broken leaf of Hyson and Young
  Hyson. A small broken Green Tea is seldom sent on the home market.
  The reason of this is obvious. When we consider that Hyson skin only
  fetches from 7_d._ to 1_s._, it is apparent that anything approaching
  a dust would give very little chance of a profit. I have seen one or
  two parcels, too much broken to come under the title of Hyson skin,
  sell at 3_d._ to 6_d._ per lb. in London. It would be well if some
  of the Indian planters would take a lesson from the Chinese, and
  not send home their very low Teas, black or green, as they are very
  difficult of sale in London, and in many cases cannot pay the cost
  of packing and shipping. The Chinese make a great quantity of their
  broken Teas into Brick Tea, and send it into the Central Provinces of
  Asia, where it meets with a ready sale. I do not see why this should
  not be done by the Indian growers. There is a large consumption of
  Tea on the other side of the Himalayas, not very far from Darjeeling
  and Assam. I hear also that in the neighbourhood of the growing
  districts, especially in the North-west Provinces, the natives are
  beginning to consume largely, and will pay 8 as. to 1 rupee for a
  Tea that could not possibly fetch more than 1_s._ to 1_s._ 6_d._ per
  lb. in England. Whether the natives of India, as a whole, do or do
  not take to drinking Tea will have a material effect on the future
  prospects of the article.

  Before dropping the subject of Green Teas, I will say a word or two
  as to the expediency of making Green Tea. I have questioned several
  experienced people on the subject, but none can tell me their
  especial object in manufacturing their leaf into Green Tea. One
  gentleman told me that he thought it was because their Tea-makers
  (Chinamen) knew better how to make greens than blacks. I have
  carefully examined the leaf of several of the North-west Green Teas,
  and, noticing their English sale prices, consider that they would
  have sold on the average at least 3_d._ per lb. higher had they been
  made into Black Tea. The best way to test this would be to have a
  Green and a Black Tea made from the same leaf, and then to value the
  one against the other. I regret that I have never had the opportunity
  of doing this. We notice that the largest and most experienced
  producers never make Green Tea.[55]

  I must not pass over Caper without a short description. It is a Tea
  which is made in large quantity in China, though I have only seen one
  parcel of Indian growth. It forms a link between the black and green
  descriptions. The colour of the leaf is a very dark green; in form it
  is similar to a gunpowder, Imperial, or round leaf Congou. The liquor
  is pale, and the out-turn green; flavour perhaps nearer to that of a
  green than of a Black Tea.

FOOTNOTES:

[51] We have yet to see what Jackson’s machine can do.

[52] Even to break Tea on them it is a mistake to use brass sieves. Tea
is best broken by a wooden roller, heavily weighed with lead, run in.
The glaze or gloss on Tea is thus preserved.

[53] I need hardly remark that the Hindustani word _Namuna_ (pronounced
_Nemoona_) means _sample_.

[54] I think I need hardly pause to correct the popular error that the
Green and Black Teas are made from two different species of plant. Most
of my readers will know that they are both made from the same leaf, the
difference lying only in the manufacture.

[55] Note to Third Edition.--As previously stated in foot-note page
133, Green Teas are now but little used in Great Britain.



CHAPTER XXVI.

BOXES. PACKING.


By far the best Tea boxes are the teak ones made at Rangoon. The
wood is impervious to insects of all kinds, even white ants. Sawn by
machinery the pieces sent to compose each box are very regular. The
plank is half inch, and each chest made up measures inside 23 by 18 by
18-1/2 inches, and necessarily outside 24 by 19 by 19-1/2 inches. The
inner cubical contents are 7,659 cubic inches, and this suffices for
above one maund of fine, and under a maund of coarse Tea.

Each box is composed of fourteen pieces--viz., for the two long sides
three each, for the two short sides two each, two for the bottom, and
two for the lid. By the arrangement of three pieces in the long sides,
and two only in the short sides, the centre piece of each long side is
attached to _both_ the short end pieces, and thus great strength in the
box is ensured, there being no place where it can possibly separate at
the joints.

These boxes are not made to “dovetail.” Each piece (and they are sawn
with mathematical regularity as to length, breadth, and thickness) must
be nailed to its neighbour. The best nails for this are the kind called
“French Pins,” 1-3/4 inches long.

The wood is sold at Rangoon in bundles, and could be landed in Calcutta
for about Re. 1-8 or 1-12 per box. The boxes need not be made up till
shortly before they are wanted, and in this form, of compact bundles of
short pieces, are very convenient for transport and stowage.

Of course in many districts these boxes are not procurable, and local
ones must be made. If so, use hard wood, and make your boxes _about_
the size given above, for small boxes add much to the cost of freight.

Let the planks be 5/8 inch thick, for 1/2 inch, that is, 4/8 inch
boards are not strong enough, except they are of teak or any other very
good wood.

Take care the joints of the several pieces composing the sides and ends
do _not_ coincide at the corners, for if they do the box is very apt to
come asunder.

The best way to arrange the pieces is as described above in the Rangoon
boxes.

“A form” must be made on which the inner leaden case shall be
constructed, that is, a well-made smooth box, to fit _exactly_ into the
box you pack in. It must be some 3 inches higher than the interior of
the original box, and have bars running across inside, for handles to
lift it up, and let the lead case slip off it, after it (the lead case)
is finished.

Solder your lead case, over your form, in the way to waste least lead.
In the Rangoon boxes described, two large, two small sheets,[56]
and one piece, 22 by 9 inches (let in between the two large sheets)
suffices, and there is little or no waste.

The lead case ready, hold up the form by the inner rods, and let the
case slide off. Put it at once into the packing-box, taking care no
nails protrude inside, or anything else which will hurt it, and thus
prepare all the boxes for the break of Tea you are about to pack.

One great advantage the Rangoon boxes, and in fact all machine-sawn
boxes, have is their equal, or nearly equal, weight. Purchasers of
Teas, at the public auctions, require “the tare” of boxes to be as
near the same weight as possible. If the tares differ, say more than
half-a-pound, the Tea will be depreciated in value.[57] It is well
there should be about the same weight of Tea in all the boxes that
contain any one kind, but this is not essential, which equality in
tares is.

Your boxes all ready and lined with lead, choose a fine day for
packing. Do this whether you finally dry the Tea in the sun or over the
dholes; for even in the latter case it is well to avoid a damp day.

But before you pack you must bulk. That is, you must mix all the Tea,
of any one kind, so intimately together that samples taken out of any
number of chests shall agree exactly. This can be done by turning out
all the Tea on a large cloth placed on the floor, and turning it over
and over. No two days’ Teas are exactly alike, and you have perhaps a
month’s Teas to pack; it is therefore necessary to mix them well.

Though I know many planters think the fumes of charcoal necessary and
beneficial for the last drying, I do not. I have tried both sun and
charcoal, and no difference was perceptible. The former costs nothing,
is more commodious, and I always apply it when possible. The sun cannot
burn the Teas; the charcoal, if the heat is too great, may.

Whether you use sun or charcoal, put the Tea hot into the boxes. The
_only_ object of the final drying is to drive off the moisture, which
the Tea will certainly, in a more or less degree, have imbibed since
its manufacture. Even the large zinc-lined bins which should be fitted
up in all Tea stores, and in which the Tea is placed after manufacture,
will not entirely prevent damp, so in all cases a final drying is
necessary.

Keep it in the sun, or over the charcoal, until it is hot throughout,
hot enough to ensure all the moisture having been driven off. Then put
into the box enough to about one-quarter fill it. Now let two men rock
the box, over a half-inch round iron bar, placed on the ground, until
the Tea has well settled. Then place a piece of carpet over the Tea,
the exact size of the box, and let a man stand inside and press it down
a minute or two with his feet. Now fill up nearly another quarter, and
press it again over the carpet as before. Repeat this, putting less and
less into the box each time, as you near the top, until it is quite
full, but do not rock it at all the last two or three times, only press
it with the feet as described. No patent screw press, or anything else,
will pack the Tea better or more closely than this plan, and when the
men are practised at it, you will find there will not be a difference
of more than two or three lbs. in the Teas of any one kind put into the
boxes.[58]

The box full, just even with the top, and well pressed down to the
last, lay over the Tea a piece of the silver paper, which is found
inserted between each sheet in the lead boxes. This prevents any solder
or resin getting on the Tea when soldering the top. Now fit on the lead
sheet top, solder, and nail on the wooden lid.

_Weight of Tea in each box._--The boxes ready lined, with a lead cover
loose, must be all weighed _before_ the Tea is packed, and again
_after_ they are filled and soldered down, but _before_ the wooden lid
is put on. The difference of these weights, minus the weight of the
little solder used in fastening down the top lead (for which allow say
one pound to give a margin also), will be the net weight of Tea in each
box.

Thin iron hooping, put round both ends of the boxes, much increases
their strength, and is not expensive.[59]

Stamp each box on its lid and on one end.[60] Use for this zinc plates,
with the necessary marks cut out in them. A brush run over these with
the colouring matter does the work well and quickly.

Let the stamp comprise the kind of Tea, the plantation or owner’s mark,
the number of the box, and the year; for instance--

[Illustration:

        Pekoe.

        \ A /
         \ /
       D  X  C
         / \
        / B \
  No. 80     1871
]

The invoice you send with the break must give for each box the number,
the gross weight, the tare, the net Tea, and the kind of Tea, with a
declaration at foot that the Teas of each kind have been respectively
well bulked and mixed together before packing.

Remember the larger the quantity of Tea, of any one kind, to be sold
at one auction, the higher the price it will probably fetch. Sell, if
possible, twenty or thirty chests of one kind of Tea at the same time,
for small quantities as a rule sell below large, both in Calcutta and
London.

Equality of tares is the most important point to attend to in packing
Teas. It may be difficult, but with machine-sawn boxes, nearly the
same weight, any difference must be made up with extra hooping, lead,
solder, or nails. Anyhow it _must_ be done, so that no tares shall
differ more than half-a-pound (see foot-note page 149).

FOOTNOTES:

[56] Large lead is 37 by 22 inches; small lead, 25 by 19 inches.

[57] Note to Third Edition.--This matter of equal tares is very
important. If they differ more than half-a-pound all the Tea is turned
out and re-weighed in London, which is a great loss in many ways.

[58] It is not essential that the same quantity of Tea shall be in each
box.

[59] This should, except the lid part, be put on the boxes before the
Tea is packed.

[60] The object of stamping the end, as well as the lid, is that when
the boxes are piled one above the other the mark can be read.



CHAPTER XXVII.

MANAGEMENT. ACCOUNTS. FORMS.


System and order, a good memory, a good temper, firmness, attention to
details, agricultural knowledge, industry, all these, combined with
a thorough knowledge of Tea cultivation and Tea manufacture, are the
requisites for the successful management of a Tea plantation.

To find men with _all_ these qualities is, I allow, not very easy,
still they do exist, and such a one must be had if success in Tea is
looked for.

Before the work is given out each day, the manager should decide
exactly what is most required, and apply it to that. He should write
down, when distributing the men, the works, and the number employed on
each. This paper he should carry in his pocket, and he can then verify
the men at work at each or any place when he visits it during the day.

The writer, the moonshee, and the jemadar (if there is one), should
write similar papers when the coolies are mustered in the morning, and
the manager should detail to each of these men which work they are
particularly responsible for. This should also be shown in the “Morning
Paper.”

Each of the above men then measures out the work to the coolies; visits
it once or oftener in the day, and measures all that remains undone at
night. A daily report of the work is kept, written by the writer in the
evening.

The two forms given below are those I have adopted. The latter is
suited to local labour paid daily, but it can easily be altered to
suit either local labour paid monthly or imported coolies.

This is the Morning Paper.

  ===================================================================
  _Work to be done on                                         188 ._
  -------------------------------------------------------------------
  Detail|No. of Garden|In whose charge|Headman on | Probable | Actual
        |             |               | the Work  | number   | number
  ------+-------------+---------------+-----------+----------+-------
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
        |             |               |           |          |
  ------+-------------+---------------+-----------+----------+-------
                                               Total Coolies |
  -------------------------------------------------------------------

The column of “Probable numbers” is given so that before it is known
exactly how many men will be present for work they can be divided in
the most likely way.

Each headman (called “Mate,” “Mangee,” &c., in different districts) is
best designated by a letter or number. In neither form would there be
room to put in names at length.

The form below written in the evening is made into a book for each
month. The advantages of it for after reference are great, and it can
of course be altered to suit the kind of labour employed on any garden.

  _Work on                   for                   188 ._
  ============================================================================
  Detail of Work|No. of |Mangees| Chupp |    |    |Coolies|Measurement and
                |Garden |       |       |    |    |       |Remarks
  --------------+-------+-------+-------+----+----+-------+-------------------
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
                |       |       |       |    |    |       |
  --------------+-------+-------+-------+----+----+-------+-------------------
                        |       |       |    |    |       |      As.
  Total at Work         |       |       |    |    |       |    × 3 =
  Command               |       |       |    |    |       |    × 2-1/2 =
  Sick                  |       |       |    |    |       |    × 2 =
  Absent                |       |       |    |    |       |    × 1-1/2 =
  ----------------------+-------+-------+----+----+-------+-------------------
  Total                 |       |       |    |    |       |    × 1 =
                        ------------------------------------------------------
                                                           Picking Leaf =
                                                           Making Tea =
                                                           Tea Sorting =
                                                               ---------------
                                                                   Cut =
                                                               ---------------
                                                                 Total =
                                                               ===============

The following is the plan I recommend for the leaf-picking and the Tea
accounts.

The leaf of each picker is best measured in the field, and, as loads
are collected, brought to the factory by one or two men throughout the
day. It entails a loss of time, and further a depreciation in the leaf,
if it is kept long in a close mass in one basket, which is the case
when each picker brings his or her leaf to the factory twice a day. The
pickers are paid so much per basket, holding in any case 2-1/2 lbs. I
find the most convenient plan is to give the mangee in charge of the
pickers tickets of any kind for this, which tickets are changed for
money in the evening. As each load of leaf comes in through the day it
is weighed, and this gives a check on the tickets given by the mangee
or mate. This is the meaning of the two columns in the form below,
“tickets by leaf” and “tickets paid.”

In the form the first column of “leaf results” shows the condition of
the leaf when picked, whether wet (W) or dry (D). Unless this were
noted the proper amount of Teas the leaf _ought_ to make could not be
known, and there would be no check against theft, which is carried on
to a great extent in many gardens.

As explained previously, only the sections ready in each garden are
picked. The sections are not entered in the form, only the number of
the garden. The flushes now noted are the 20th, in some the 21st, or
22nd in others.

The Tea is calculated from the leaf. It should be 25 per cent. if the
leaf is picked dry, and 22 per cent. if picked wet. As each load comes
in a memorandum is made as to whether it is dry or wet, and the figures
in the column “Tea should be” are thus found.

The Tea is weighed the morning after it is made and entered in the
column “Tea made.” The percentage it bears to the leaf is then
calculated and entered in the account column.

After sifting the whole is weighed again, and the result entered in the
column “Tea after sifting.” Doing this is very important, for it checks
theft. Directly after it is weighed this second time it is put in the
bins in the store.


_Daily Leaf and Tea Account._

  ---------------------------------------------------
                            TEA RESULTS.
  --------------+------+---------+---------+---------
                |Tea   |         |         |Tea after
        Date    |should|         |         |it is
                |be    |Tea made |Per cent.|sifted
  --------------+------+---------+---------+---------
  October       |      |         |         |
  Sunday, 1st   |„     |  „      |   „     |   „
                |      |         |         |
                |      |         |         |
                |      |         |         |
  Monday, 2nd   |198   |  200    |   24    |   199
                |      |         |         |
                |      |         |         |
                |      |         |         |
  Tuesday, 3rd  |231   |  230    |   25    |   233
                |      |         |         |
  Wednesday, 4th|      |         |         |
  Thursday, 5th |      |         |         |
  Friday, 6th   |      |         |         |
  Saturday, 7th |      |         |         |
                |      |         |         |
  Total for     |„     |Mds. lbs.|         |
    the week    |      |  1632   |         |
  --------------+------+---------+---------+---------

  ----------------------------------------------------
                 LEAF RESULTS.
  -----+--------------+-------+-----------------+-----
  State|   Tickets    | Number|     Flushes     |
  of   +-------+------+ of    +-----+-----+-----+Total
  leaf |By leaf|Paid  | Garden|  20 |  21 |  22 |leaf
  -----+-------+------+-------+-----+-----+-----+-----
  220W |       |     {|   3   | „   | 170 | „   |
  600D |  410  |360  {|   5   | 310 | „   | „   |
       |       |     {|   7   | 112 | „   | „   |
       |       |     {|   8   | „   | „   | 228 | 820
       |       |      |       |     |     |     |
   D   |  462  |440  {|   3   | „   | 515 | „   |
       |       |     {|   9   | 410 | „   | „   | 925
       |       |      |       |     |     |     |
       |       |     {|   1   | 430 | „   | „   |
   W   |  200  |180  {|   2   | „   | „   | 160 |
       |       |     {|   3   | „   | 210 | „   | 800
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
       |       |      |       |     |     |     |
  -----+-------+------+-------+-----+-----+-----+-----

If this system is carried out, no Tea (exceeding a pound or so) can be
stolen without its being at once missed, and the importance of this
cannot be exaggerated. Tea proprietors do not guess _how_ much is lost
in this way. Maunds upon maunds might be stolen in many gardens, and
unless the theft were accidentally discovered there is nothing in the
Tea accounts to show it to the manager.

I have suppositiously filled up the three first days of the form. The
820 lbs. leaf picked on Sunday is made into Tea on Monday. The 198
is written down Sunday evening. On Tuesday morning, when the Tea is
weighed and found to be 200 lbs., that is entered in the Monday line,
as also the percentage. On Tuesday evening, after it is sifted and made
into different Teas, it is weighed again and found to be 199 lbs., and
so entered.

In dry weather after sifting, owing to dust flying off, it is always a
little less. In wet weather, on the contrary, it increases in weight.
In the Tuesday line where “W” shows it was a wet day and the Tea 230
lbs. before sifting it, is 233 afterwards. This is owing to moisture
imbibed, and it is the only objection to sifting daily, whatever
the weather. The advantages of the plan, though, are so great, as
explained, that I put up with this, and practically I do not find it
detrimental. Of course, as previously explained, all moisture is driven
off before the Tea is _packed_. However, to make all quite safe, after
a very wet damp day, the Teas might be re-dried for a few minutes over
charcoal before being put into their respective bins. I do not do this
myself though, and do not think it necessary.

I hope now I have made the above form plain. It is in a book, and each
page will hold one week. The total of the Tea made in the week is added
up and shown at foot, and that amount is then transferred to the credit
side of the Tea store account. Thus (see both forms) 16 maunds 32 lbs.
is credited.

The form given on the next page is also kept in a book, and the total
of right-hand side subtracted from the left gives at any time the
quantity of Tea in store.

_Tea Store Account._

  ------------------------------------
               RECEIPTS
  ------------+-----------+-----------
              |  Tea made |
  Week ending |  in week  |   Total
  on Saturday +-----+-----+-----+-----
              | Mds.| lbs.| Mds.| lbs.
  ------------+-----+-----+-----+-----
  Brought     |     |     |     |
   over    .. |  .. |  .. | 405 |   8
  October 7th |  16 |  32 |     |
    „     14th|  15 |   0 |     |
    „     21st|  17 |  10 |     |
    „     28th|  14 |  40 |     |
              |-----|-----|  63 |   2
              |     |     |     |
              |     |     |     |
              |     |     |     |
              |     |     |     |
              |     |     |     |
  ------------+-----+-----+-----+-----
        Carried over      |     |
  ------------------------+-----+-----

  -----------------------------------------------------------
                            EXPENDITURE
  -----------+--------------+-------+-----------+------------
             |              |       |Tea in each|
     Date    |No. of Invoice|To whom|  Invoice  |   Total
             |              |       +-----+-----+-----+------
             |              |       | Mds.| lbs.| Mds.| lbs.
  -------+---+--------------+-------+-----+-----+-----+------
         |   |              |Brought|     |     |     |
         |   |              | over  |  .. |  .. | 351 |  14
  October| 3 |      15      |-------|  40 |  15 |     |
    „    |20 |      16      |-------|  33 |  10 |     |
         |   |              |       |-----|-----|  73 |  25
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
         |   |              |       |     |     |     |
  -------+---+--------------+-------+-----+-----+-----+------
                Carried over        |     |     |     |
  ----------------------------------+-----+-----+-----+------

Regarding accounts between the manager and his employers, I think they
should be of the simplest kind. If a man _can_ be trusted he _should_
be trusted; if he cannot, no system of accounts will restrain him, and
he should be kicked out. A simple account current, furnished monthly,
showing under few heads the receipts and expenditure, is all that can
be required. It is not by _any_ papers received from a manager that
an opinion can be expressed as to how he does his work, and how the
plantation progresses. A competent person visiting the garden can
easily ascertain, and in default of this, and combined with this, the
only true test is the balance sheet at the end of each season.

Shortly, it is not by the form, the nicety, the detail of accounts
between manager and employer, that success is ensured or even
forwarded. It is, as far as accounts are concerned, by the forms and
_system_ the manager adopts as between him and his subordinates, and
these he should be able to show are good to the employer, or anyone
deputed by him to visit the garden.

The profit shown yearly, whether it is large or small, all things
considered, is, however, the only true ultimate test.



CHAPTER XXVIII.

COST OF MANUFACTURE, PACKING, TRANSPORT, ETC.


These are as follow:--They will vary more or less according to the
district, rate of wages, &c., but in the form the tables are given, if
not suitable to any case, they can easily be made so.

I have added sorting, packing, freight to Calcutta, and broker’s
charges in Calcutta, to the cost, so that all is included from the
moment the green leaf is picked off the trees till the hammer falls at
the public auction.


_Table cost of Manufacture, Sorting, Packing, Transport to Calcutta,
and Broker’s Charges for each maund of Tea._

              _Manufacture._                 RS.  AS.  P.   RS.  AS.  P.

  1 head man with the pickers, say            0    4   0

  320 lbs. green leaf picked, at 1 pice
    per lb.[61]                               5    0   0

  1 man withering above leaf, at say 4
    annas                                     0    4   0

  1/4 share head man in rolling house         0    2   0

  10-2/3 men rolling above, at 30 lbs. leaf
    per man, and say 4 annas per man          2   10   8

  1/4 boy clearing out ashes of Dhole house,
    at say 2 annas                            0    0   6

  1/4 share head man in Dhole house           0    2   0

  1 man firing “Dhole work” say               0    4   0

  3/4 maund charcoal for Dhole work, at 8
    annas                                     0    6   0

  Lights for night work, viz., turning green
    leaf and dholing, say                     0    4   0

  Wear and tear of dhallas, baskets,
    picking baskets, fuel for artificial
    withering, &c.                            0    1  10
                                              ----------    9    7   0


          _Sifting and Sorting._

  1-1/2 boys to pick out red leaf, at say 2
    annas                                     0    3   0

  1 sifting man, at say 4 annas               0    4   0

  Wear and tear of sieves, say                0    0   3
                                              ----------    0    7   3


              _Packing._

  1 box                                       1   13   0

  4 sheets lead, viz., 2 large and 2 small.   1    6   6

  Labour of lining box with lead, solder,
    closing lead, closing wooden box,
    stamping, and cost of nails               0    0   9

  Labour of drying previous to packing,
    whether in sun, or over dholes,
    including charcoal, if the latter are
    used                                      0    0   9

  Labour of filling the box, shaking it well,
    and pressing down the Tea (2 men)         0    0   6
                                              ----------    3    5   6


            _Transport._

  Freight to Calcutta for one maund Tea,
    say                                       1   12   0
                                              ----------    1   12   0

     _Broker’s Charges in Calcutta._

  Landing, lotting, and advertising per
    chest                                     0   14   0

  Brokerage at 1 per cent. on the amount
    sale, say Rs. 70 for the maund            0   11   3
                                              ----------     1    9   3
                                                             -----------
        Total for one maund of Tea[62]                       16    9   0
                                                             -----------

N.B.--If more than two maunds Tea are made per day, some of the items
under head of “Manufacture” would be a little less. See page 70, where
it will be seen that each maund of Tea is worth to the manufacturer
(after deducting all costs) Rs. 50.

FOOTNOTES:

[61] In practice the basket in which the leaf is measured being made to
hold 2-1/2 lbs., for which a ticket is given, representing 2 pice, the
leaf to make a maund of Tea does not really cost so much.

[62] After experience has shown me this amount, when any quantity of
Tea is made, is too high--Rs. 12 to 13 would be nearer the mark.



CHAPTER XXIX.

COST OF MAKING A 300-ACRE TEA GARDEN.


In the following estimate 100 acres are supposed to be planted the
first year, 100 acres the second, and 100 acres the third.

To elucidate a table I shall draw up in the next chapter showing the
probable receipts and expenditure on such a garden for a series of
years, I shall suppose this plantation to be begun in 1875, and number
the years accordingly.

The expenditure would truly, in the supposed case, begin in the latter
part of 1874, but it is more convenient to regard it as commencing 1st
January, 1875.

I estimate all new cultivation as planted “at stake,” that is, the seed
sown _in situ_. Nurseries are only to fill up vacancies.

I shall not pretend in this to go into minute details, such as are
given at page 84, for it is simply impossible to do so. The cost of
making a plantation must vary greatly, being determined by climate,
available labour and its rates, lay of land, nature of jungle to clear,
&c., &c. In this estimate only round numbers can be dealt with. The
prices I assume are average ones, neither suited to very heavy jungle,
and very expensive labour, or the reverse:--

             _1st year_ (1875).                          RS.      RS.

  Purchase 700 acres land, at Rs. 8 per acre           5,600
  40 maunds seed, at Rs. 70[63]                        2,800
  Nurseries for vacancies and labour transplanting[64]   200
  First temporary buildings                            1,000
  All expenditure to plant 100 acres, at
    Rs. 80 per acre[65]                                8,000
  Cultivating the said 100 acres first year, at
    Rs. 50 per acre[66]                                5,000
                                                       -----     22,600

                _2nd year_ (1876).
  60 maunds seed, at Rs. 70[64]                        4,200
  Nurseries and labour transplanting[64]                 300
  Repairs, buildings and some new ones still
    of a temporary nature                                500
  All expenditure to plant the second 100
    acres, at Rs. 70 per acre[65]                      7,000
  Cultivating first 100 acres, at Rs. 60,
    second 100 acres, at Rs. 50 per acre[66]          11,000
                                                      ------    23,000

                _3rd year_ (1877).
  70 maunds seeds, at Rs. 70[64]                       4,900
  Nurseries and labour transplanting[64]                 400
  Buildings for Tea manufacture (temporary)
    and repairs to buildings                           3,000
                                                      ------     8,300

  All expenditure to plant the third 100 acres,
    at Rs. 60 per acre[67]                             6,000
  Cultivating first 100 acres, at Rs. 70, second
    at Rs. 60, third at Rs. 50 per acre[68]           18,000
                                                      ------    24,000
  Interest on first year’s outlay, two and a
    half years, second year’s outlay, one
    and a half years, third year’s outlay half
    year, at Rs. 5 per cent. per annum.                          5,357
                                                                ------
  Total expense to make the 300-acre garden                     83,257
                                                                ======

The garden is now made at a cost, including interest on all outlay of
Rs. 83,257, and I am very confident that a good 300-acre garden can, as
set out, be made for that sum. The rates assumed are so liberal that a
fair margin is allowed for bad seed or any other misfortune.

                _4th year_ (1878).
  20 maunds seed, at Rs. 70[69]                        1,400
  Nurseries and labour transplanting[69]                 500
  Repairs, buildings[70]                                 500
  Cultivating first 100 acres, at Rs. 80, second
    at Rs. 70, third at Rs. 60 per acre[71]           21,000
                                                      ------    23,400

                _5th year_ (1879).
  10 maunds seeds, at Rs. 70[72]                         700
  Nurseries and labour transplanting[72]                 500
  Repairs, buildings[73]                                 500
  Cultivating first 100 acres, at Rs. 90, second
    at Rs. 80, third at Rs. 70 per acre[74]           24,000
                                                      ------    25,700

                _6th year_ (1880).
  Nurseries and labour transplanting[72]                 500
  Repairs buildings[73]                                  500
  Cultivating first 100 acres, at Rs. 100,
    second at Rs. 90, third at Rs. 80 per
    acre[74]                                          27,000
                                                      ------    28,000

                _7th year_ (1881).
  Nurseries and labour transplanting[72]                 500
  Building a permanent Tea Factory and Tea
    Store and repairs to building[73]                 12,500
  Cultivating first 100 acres, at Rs. 100,
    second at Rs. 100, third at Rs. 90 per
    acre[74]                                          29,000
                                                      ------    42,000

                _8th year_ (1882).
  Nurseries and labour transplanting[72]                 500
  New permanent houses for Manager and
    Assistant, and repairs to buildings[73]            8,500
  Cultivating first, second, and third 100
    acres, at Rs. 100 per acre[74]                    30,000
                                                      ------    39,000

           _9th year (1883), and all years after._
  Nurseries, at Rs. 500[75]                          }
    Repairs to buildings, at Rs. 500[76]             } 1,000
  Cultivating the 300 acres, at Rs. 100 per
    acre[77]                                          30,000
                                                      ------    31,000

Nothing is allowed for interest after the third year, for soon after
that, viz., fifth year, the garden begins to give profits on the yearly
operations.

All the above figures are carried out in the table in the next chapter,
page 172, and how large the profits on Tea may be will there be seen.

In none of the estimates of cost, up to this, is the expense of
manufacturing the Tea included. It would have been very inconvenient to
do so. The cost is so much per maund of Tea, and I prefer estimating
the Tea at its market rate _minus_ the cost of manufacture shown at
pages 70 and 162.

FOOTNOTES:

[63] The cost for seed, nurseries, and transplanting increases each
year as the area over which vacancies may exist enlarges.

[64] See note 63 p. 163.

[65] The expenditure for planting the 100 acres each year includes
cutting and clearing jungle, removing roots, digging, staking, pitting,
and sowing the seed. In fact _all_ expenditure including part of the
pay of Manager and Establishment. The rate per acre _decreases_ each
year, because each year there is more expenditure of other kinds, which
helps to pay for the Manager and Establishment.

[66] The reason why the rate for cultivation on the 100 acres planted
each of the three first years increases each year is given in the table
and remarks at pages 84 and 85.

[67] See note 65, p. 164.

[68] See note 66, p. 164.

[69] The seed to be bought is now less each year, as it is produced
on the garden, and after the fifth year no more has to be purchased.
From the fourth, and all subsequent years, nurseries for vacancies
are calculated at Rs. 500, which is enough, as the garden has been
previously yearly replenished. This expenditure will be continual as
long as the garden lasts, for there will always be some vacancies to
replace.

[70] Rupees 500 is a fair sum to estimate for ordinary annual repairs
to buildings, and it will be required as long as the garden lasts. A
temporary Factory was made in 1877, and a permanent building is now
allowed for in 1881. Permanent Manager’s and Assistant’s houses are
also allowed for in 1882. The garden can afford this now, for the
profits are large. (See table at page 172.)

[71] For the rates assumed here see page 84.

[72] See Note 69, p. 165.

[73] See Note 70, p. 165.

[74] See Note 71, p. 165.

[75] See note 69, p. 165.

[76] See note 70, p. 165.

[77] See note 71, p. 165.



CHAPTER XXX.

HOW MUCH PROFIT TEA CAN GIVE.


We have already estimated the cost of making and cultivating a
plantation of 300 acres. We must now ascertain how much Tea that area
will give yearly.

It is a very wide question what produce an acre of Tea will give.

The following is an extract from the “Report of the Commissioners
appointed to enquire into the state and prospects of Tea cultivation
in Assam and Cachar,” addressed to the Government of Bengal, and dated
March, 1868:--

  “_Average produce per acre._”

  “The returns of actual produce of gardens in 1867 which we have
  obtained are so few in number that it is impossible to take any
  general average from them. The produce in these varies from
  three-and-a-half maunds to one-and-a-half maunds per acre, omitting
  the more recently formed gardens.

  “From information received during our tour we have reason to believe
  that some gardens produce more than the highest rate per acre here
  mentioned; but, in the absence of returns of exact acreage and
  out-turn, we cannot notice these instances.

  “Mr. Haworth, in his pamphlet already quoted, speaks of the produce
  of Cachar gardens as follows:--

  “‘I believe that three maunds per acre is fully one-third more than
  the present average yield of gardens in Cachar, after deducting the
  area of plant under yielding age.

  “‘There is no reason, that I am aware of, why the yield of Tea should
  not soon be raised to four maunds, and more gradually six maunds per
  acre, equal to twenty-four maunds of leaf per acre (less than one
  ton per acre for a green crop, which is still a very small one).
  Even now there are gardens in Cachar which give an average of from
  five to six maunds per acre this season. Some of these gardens have
  really no apparent advantage over their less fortunate neighbours,
  beyond that of a somewhat better system of cultivation and pruning;
  and these improvements even are to such a small degree ahead of the
  general practice, that I feel justified in saying I cannot place a
  limit on what the increased yield should be under a more rational
  system of cultivation, and the application of manures on a liberal
  scale, leaving out of consideration altogether what might reasonably
  be expected from a good system of drainage in addition.’

  “Mr. James Stuart, Manager of the Bengal Tea Company’s gardens in
  Cachar, has also given two maunds an acre as the general average of
  Cachar gardens for the past season, including young gardens of two,
  three, and four years old.

  “We do not think it necessary to quote in detail the opinions of all
  the gentlemen examined by us on the subject of average produce per
  acre. A garden that can give four maunds per acre is undoubtedly a
  good one, and we have no doubt there are such, or even better; but we
  do not think they are so common as to warrant our taking more than
  three maunds as a safe average.”

Mr. A. C. Campbell, Extra Assistant Commissioner at Burpettah, in his
“Notes on Tea Cultivation in Assam,” published in the Journal of the
Agricultural and Horticultural Society of India, part 3, vol. xii.,
page 309, says:--“Good Tea land can be made to yield as high as seven
maunds per poora.” I forget exactly how much a poora is, but I believe
it is nearly an acre.

In the Report to Government by the Commissioners, quoted above, at page
9, Mr. T. Burland, after estimating the cost of cultivation per acre
per mensem at Rs. 9-10-2, adds:--“With the above expenditure per acre
it is probable that much more than five maunds of Tea will be obtained
from an acre of fair plant.”[78]

All these estimates, however, are based on the cultivation of Tea as
carried on hitherto with few exceptions, that is to say, on gardens
covered with weeds for many months in the year, and to which no manure
has ever been given. With such cultivation, particularly on gardens
planted on slopes, I think myself that the yield will not exceed four
maunds _at the outside_.

High cultivation and liberal manuring will, I believe, at least double
the above, if the plants are of a high class. However, here I give a
table on the subject which I have carefully framed.

  _Estimate of probable yield per acre on flat land, good soil, in a
  good Tea climate, and with hybrid plants, if really high cultivation
  and liberal manuring is carried out._

  +----+-------------+-------------------+
  |Year|Supposed Year|Estimated yield per|
  |    |             | acre in maunds[79]|
  +----+-------------+-------------------+
  | 1st|    1875     |        --         |
  | 2nd|    1876     |        --         |
  | 3rd|    1877     |        1/2        |
  | 4th|    1878     |         2         |
  | 5th|    1879[80] |         4         |
  | 6th|    1880     |         5         |
  | 7th|    1881     |         6         |
  | 8th|    1882     |         7         |
  | 9th|    1883     |       7-1/2       |
  |10th|    1884[81] |         8         |
  +----+-------------+-------------------+


I do not think plants reach to perfect maturity under eight or ten
years.

That eight maunds per acre as estimated in the table just given _can_
be realised, under the conditions stated, I have no doubt whatever, but
I am equally certain that the size of some gardens in India must be
much reduced if even five or six maunds are looked for.[82] Not only
must they be reduced in size, but they must be highly cultivated, must
be manured, and no vacancies allowed. However, I have dwelt on all
these points before, and need not repeat here, for unless the reader is
convinced before this that a large area and low cultivation won’t pay,
it were waste to write more.

I now give a table showing the result for twelve years of a plantation
such as I have advised.


TABLE SHOWING THE ESTIMATED RESULTS FOR 12 YEARS OF A 300-ACRE
PLANTATION, IN A GOOD TEA CLIMATE, HIGHLY CULTIVATED AND LIBERALLY
MANURED.

  +----------------------------------++
  |         YEAR AND RATE OF         ||
  |            YIELDING IN           ||
  |              MAUNDS              ||
  +----+---------------+-------------++
  |  1 |       2       |       3     ||
  |    +---------------+-------------++
  |    |Yield per acre |             ||
  |    |as per page 170|Supposed year||
  |Year|               |             ||
  +----+---------------+-------------++
  |    |     Mds.      |             ||
  |  1 |      ..       |     1875    ||
  |  2 |      ..       |     1876    ||
  |  3 |     1/2       |     1877    ||
  |  4 |      2        |     1878    ||
  |  5 |      4        |     1879    ||
  |  6 |      5        |     1880    ||
  |  7 |      6        |     1881    ||
  |  8 |      7        |     1882    ||
  |  9 |    7-1/2      |     1883    ||
  | 10 |      8        |     1884    ||
  | 11 |      8        |     1885    ||
  | 12 |      8        |     1886    ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  |    |               |             ||
  +----+---------------+-------------++

  ----------------------------------------------------------------------------------++
                                   YIELD IN MAUNDS OF 300 ACRES                     ||
                                       AND ITS VALUE                                ||
                                                                                    ||
  -----------------+-----------------+-----------------+---------------+------------++
          4        |        5        |        6        |       7       |      8     ||
  -----------------+-----------------+-----------------+---------------+------------++
  Rate yield of 100|Rate yield of 100|Rate yield of 100| Total yearly  | Value per  ||
   acres, planted  | acres, planted  | acres, planted  |yield in maunds|   maund    ||
      in 1875      |    in 1876      |    in 1877      |               |            ||
  -----------------+-----------------+-----------------+---------------+------------++
         Mds.      |       Mds.      |       Mds.      |       Mds.    |            ||
          ..       |        ..       |        ..       |        ..     |   Rs. 50   ||
          ..       |        ..       |        ..       |        ..     | per maund, ||
         1/2       |        ..       |        ..       |        50     |   after    ||
          2        |       1/2       |        ..       |       250     |  cost of   ||
          4        |        2        |       1/2       |       650     |manufacture,||
          5        |        4        |        2        |     1,100     |  packing,  ||
          6        |        5        |        4        |     1,500     |    and     ||
          7        |        6        |        5        |     1,800     | transport  ||
        7-1/2      |        7        |        6        |     2,050     |    are     ||
          8        |      7-1/2      |        7        |     2,250     |  deducted, ||
          8        |        8        |      7-1/2      |     2,350     |  see pages ||
          8        |        8        |        8        |     2,400     |  70 & 162  ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
                   |                 |                 |               |            ||
  -----------------+-----------------+-----------------+---------------+------------++

  --------------------------------------------------------++
                                                          ||
                        YEARLY RESULTS                    ||
                                                          ||
  ------------+-----------------+-------------+-----------++
       9      |        10       |      11     |     12    ||
  ------------+-----------------+-------------+-----------++
  Receipt from|Total expenditure|             |           ||
  sale of Tea |detailed at pages|Yearly Profit|Yearly Loss||
              |  164, 165, 166  |             |           ||
  ------------+-----------------+-------------+-----------++
       Rs.    |        Rs.      |      Rs.    |    Rs.    ||
       ..     |      22,600     |      ..     |   22,600  ||
       ..     |      23,000     |      ..     |   23,000  ||
       2,500  |      37,657[83] |      ..     |   35,157  ||
      12,500  |      23,400     |      ..     |   10,900  ||
      32,500  |      25,700     |     6,800   |    ..     ||
      55,000  |      28,000     |    27,000   |    ..     ||
      75,000  |      42,000     |    33,000   |    ..     ||
      90,000  |      39,000     |    51,000   |    ..     ||
    1,02,500  |      31,000     |    71,500   |    ..     ||
    1,12,500  |      31,000     |    81,500   |    ..     ||
    1,17,500  |      31,000     |    86,500   |    ..     ||
    1,20,000  |      31,000     |    89,000   |    ..     ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
              |                 |             |           ||
  ------------+-----------------+-------------+-----------++

  -----------------------------------------------------------++
                                                             ||
                          FINAL RESULTS                      ||
                                                             ||
  --------------+-----------------+-------------+------------++
        13      |        14       |      15     |     16     ||
  --------------+-----------------+-------------+------------++
  Total receipts|Total expenditure| Balance to  | Balance to ||
    to end of   |    to end of    |credit end of|debit end of||
    each year   |    each year    |  each year  |  each year ||
  --------------+-----------------+-------------+------------++
       Rs.      |       Rs.       |     Rs.     |     Rs.    ||
       ..       |       22,600    |     ..      |   22,600   ||
       ..       |       45,600    |     ..      |   45,600   ||
        2,500   |       83,257    |     ..      |   80,757   ||
       15,000   |     1,06,657    |     ..      |   91,657   ||
       47,500   |     1,32,357    |     ..      |   84,857   ||
     1,02,500   |     1,60,357    |     ..      |   57,857   ||
     1,77,500   |     2,02,357    |     ..      |   24,857   ||
     2,67,500   |     2,41,357    |     26,143  |     ..     ||
     3,70,000   |     2,72,357    |     97,643  |     ..     ||
     4,82,500   |     3,03,357    |   1,79,143  |     ..     ||
     6,00,000   |     3,34,357    |   2,65,643  |     ..     ||
     7,20,000   |     3,65,357    |   3,54,643  |     ..     ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
                |                 |             |            ||
  --------------+-----------------+-------------+------------++

  ----------------------------------+
                                    |
                                    |
                                    |
                                    |
                                    |
                                    |
              REMARKS.              |
                                    |
                                    |
  ----------------------------------+
                                    |
  It will be seen from this         |
   table as follows:--              |
  1. About Rs. 90,000 of capital    |
   is necessary to make a plantation|
   as quick as this. If             |
   made more gradually, very        |
   much less would suffice.         |
  2. There is no yearly profit      |
   until the fifth year.            |
  3. By the eighth year all the     |
   outlay is recovered.             |
  This table has been prepared      |
   with great care, and the         |
   authority for the figures        |
   assumed has been arrived         |
   at in previous parts. (_See_     |
   headings of the Cols. for the    |
   pages and note at foot.)         |
  I believe this table represents   |
   truly what Tea, with             |
   all the necessary advantages     |
   detailed in the next             |
   page, can do.                    |
  ----------------------------------+


  At the following pages will be found the calculations for the figures
  assumed:--Col. 2, page 170; Col. 8, pages 70 and 162; Col. 10, pages 164,
  165, 166.

The necessities for success in Tea are:--

  1. A good climate.

  2. A good site.

  3. Perfect knowledge in Tea cultivation and Tea manufacture on the
  proprietor’s part or that of his manager.

  4. Seed from a high class of plants.

  5. Local or cheap imported labour.

  6. Facilities for manuring.

  7. Cheap transport.

Do not dispense, though, with even _one_ of the seven points named, for
the truth is simply, that Tea will pay _very_ well with all the above
advantages, but will utterly fail without them.

Such is my advice to intending beginners. To those who have gardens, I
say, reduce your areas till of the size you can really cultivate them
highly, and procure manure at any cost.

I shall not have written in vain, and Tea enterprise in India will
flourish, if the motto of planters in future be--

  “A full area, highly cultivated.”

FOOTNOTES:

[78] See my estimate for cultivation at page 84. I there estimate Rs.
100 per acre per annum from the sixth year, so that Mr. Burland six
years ago had come to the same opinion about high cultivation that I
hold.

[79] Calculating Tea by maunds is convenient, inasmuch as pounds
necessitate such lengthy figures for all calculations. The maund here
employed is, however, quite an arbitrary measure. It is _not_ the
Indian maund, it equals and is represented exactly by 80 lbs. Any
number of maunds multiplied by 80 will naturally give the lbs. of Tea.

[80] Up to this point, viz., the fifth year inclusive, the figures
given have been much more than realised, and that on a garden with
15 per cent. vacancies. It has been, though, highly cultivated and
liberally manured from the first.

[81] From the fifth to the tenth year is assumption, except that I know
one garden which, to my certain knowledge, has given _more_ then ten
maunds an acre, and this in spite of about 15 per cent. vacancies. The
garden is an old one, planted about 18 years ago. It is also a very
small one. The soil is _very_ poor, but the plants are of the highest
class. It was much neglected till about eight years ago. From that
time it has been highly cultivated in every way except in the point of
irrigation, for it has not that advantage. It has been _most_ liberally
manured.

[82] Note to Third Edition.--With high cultivation on a favourable site
and in a really good Tea climate, I now believe 10 maunds per acre will
eventually be realised.

[83] With interest, _see_ pages 164 and 165.



CHAPTER XXXI.

PAST, PRESENT, AND FUTURE OF INDIAN TEA.


A few words on the past, the present, and the future of Indian Tea will
now conclude this Essay, and will, I hope, be acceptable to the reader.

The subject is one of growing importance, but being a new one, there
are points connected with it on which the public are very ignorant, and
should be enlightened.

To begin with, the following facts are not disputed by those who know
anything of the subject:--

1. Indian Teas have far more body, that is strength, than Chinese Teas.

2. Indian Teas consequently command a higher price at the London sales
than Chinese Teas.

3. In spite of its higher price, it is far more economical than the
Chinese produce, as, generally speaking, one-third of the quantity
suffices.

4. There are lands enough in India to grow all the Tea required for
England’s use, and, indeed, for all her colonies.

If these _are_ facts, and I confidentially affirm they are so, how is
it that the following holds in England?

1. Indian Tea is not known to the public.

2. Except in one or two shops in London and Glasgow, unknown to the
mass of the people, not an ounce of pure Indian Tea can be bought in
all England.

3. That India is even a Tea-producing country is scarcely known in
England.[84]

I think I can explain some of these anomalies.

Tea is an acquired taste: by which I mean, not only that the adult who
had never tasted Tea would not like it when first offered to him, but
also that, with those who consume it regularly, any Tea that differs in
flavour from what is habitually drunk is not relished.

It matters not whether it is intrinsically better or worse, enough that
the flavour is different, for that reason it is not liked.

Indian Tea differs widely from Chinese Tea, and for that reason is
rarely appreciated by those accustomed to the latter.

For a long time it appeared as if this difficulty would be a bar to the
general introduction of Indian Teas in England, and so indeed it would
have proved, had the short-sighted policy adopted at the commencement
by one or two Indian Companies that their Teas should be sold retail
and pure, that is, unmixed with Chinese, been followed out. It did not
avail to tell John Bull it was better Tea, that it was far stronger,
that it was in no way adulterated; for he simply shook his head, the
flavour was different to what use had made him familiar with, and he
would none of it.

But little by little, in spite of the above, it made its way. Grocers
soon found that the worst, _id est_, the weakest class of Chinese Teas
received _body_ and were made saleable by an addition of Indian Tea.
It was not long after this that the trade discovered that pretty well
_all_ Chinese Teas were improved, if proportions of Indian Teas were
mixed with them. In short, the fact was recognised by Tea vendors that
Chinese Teas were weak, and much improved if mixed with Indian.

The public were thus _educated_ to relish the superior flavour of
Indian Tea, and did so, when the quantity mixed with the Chinese was
not so great as to make the new flavour too _prononcé_. Little by
little the custom of so mixing became very general, so much so that it
may almost be said to-day that if Indian Teas cannot be purchased pure,
no more can Chinese. A mixture of Chinese and Indian Tea, the latter
small as compared with the former, is what is now generally used in
Great Britain.

This is the case to-day. What will it be in the future?

As the English palate is educated to like the flavour of Indian Tea,
more and more of it will be demanded in the mixture made up for the
public, and though the day is distant, nay, may never arrive, on
account of its greater cost, when it will be generally drunk pure, I do
not myself doubt that the demand for it will go on steadily increasing
for years to come, as it has for years past.[85]

It is an important query if, with a largely increased demand, the
supply will be equal to it. Very far from all India has a good Tea
climate, which is a peculiar one, and only exists in perfection in
Assam, Cachar, Chittagong, and lands in Bengal close to the foot of the
Himalayas.

But in these districts alone there are lands sufficient to supply
nearly the whole world with Tea, so that it is not the lands which are
wanting, though the Government prices for the lands are prohibitory and
will check cultivation. But in Assam, Cachar, and the Terai below the
Himalayas labour is very scarce, while in Chittagong the area fit for
Tea is not large, so that I do not anticipate any very sudden increase
of the cultivation, though year by year it is on the increase and will
so continue.

On the other hand, I do not--for the reasons stated, viz., that
Tea is an acquired taste and thus a new kind is not at first
palatable--anticipate any very sudden increase in the demand. If,
however, I am wrong, and from a largely increased demand the prices of
Indian Teas rise, I do not doubt that the cultivation will be greatly
extended, and that after an interval of four years (it takes that time
for the Tea plant to produce) the supply will be equal to the then
wants of the English market.

The future of Indian Tea is, I think, a bright one, and I know nothing
in which capital can be more profitably invested if the business is
conducted with knowledge and experience, but to embark in it without
these two requisites is ruin.

A few figures may be given here. The imports into Great Britain of
Indian Teas have been yearly increasing, till in 1873 they amounted to
18,367,000 lbs., and, judging from the estimate out here of the produce
this year, viz., 1874, the imports into Great Britain in 1874 will not
be far short of 20,000,000 lbs.[86]

But as the annual consumption of Tea in the United Kingdom is not less
than 130,000,000 lbs., India is still very far from supplying enough to
give a mixture of three-fourths Chinese and one-fourth India Tea.[87]

The finest Chinese Tea sells in London in bond at 2_s._ 4_d._ to 2_s._
6_d._, while the finest Indian in bond fetches 3_s._ to 3_s._ 6_d._[88]

What, then, will be the future of Indian Tea? It is an important query.
The industry is one which, if successful, might attain to wide limits,
and help not a little to relieve the Indian State Exchequer, while it
would afford occupation to many a class of Englishmen who at present
look about in vain for employment.

Tea speculation has passed through the first two preliminary phases
to which most new ventures are liable. First, we had the wild rush,
the mad fever, when every man thought that to own a few Tea bushes
was to realise wealth. In those days existing plantations were bought
at eight and ten times their value; nominal areas of 500 acres were
paid for which, on subsequent measurement, proved to be under 100; new
gardens were commenced on impossible sites, and by men as managers who
not only did not know a Tea plant from a cabbage, but who were equally
ignorant of the commonest rules of agriculture. Boards highly paid,
with secretaries still more liberally remunerated, were formed both in
Calcutta and London to carry on the enterprise; and, in short, money
was lavished in every conceivable way, while mismanagement ran rampant
in each department. It is not strange that the whole thing collapsed:
the wonder is it did not do so earlier.

The second stage was then entered upon. Numbers had been bitten, and
the idea, once formed, grew apace, that Tea could not pay at all.
Everyone wanted to sell, and down went all Tea shares to a figure
which only increased the general panic. Many companies, and not a few
individuals, unable to carry on, had to wind up and sell their estates
for whatever they would fetch. Gardens that had cost lakhs were sold
for as many hundreds, and the very word “Tea” stank in the nostrils of
the commercial public. A few of the best companies held on, as also
such individuals embarked in the speculation as could weather the
storm; but some of the companies were bowed down with heavy debts, and
it has been with many, from that cause, a losing race ever since.

This great smash occurred in 1867. I purpose, therefore, to examine
into the future prospects of the industry, now that time has been given
to test its vitality. Naturally the mistakes made at the first have not
been repeated since, so the speculation has had more or less of a fair
chance to show what it can do.

In the first place, the share list of Tea companies in the public
prints does not at all represent the true position of Tea property
to-day. It only gives the dividends declared and the value of the
shares in those few limited liability companies which were able to
weather the storm, but who, in common with all the others, were bowed
down with debt, and are suffering up to the present time, both from
that and the numberless mistakes made at the commencement of the
enterprise. There are a few notable exceptions, even among the Tea
companies. Some of these have done very well, pay large dividends, and
are quoted at a high premium, which shows that Tea can and will pay
even with the disadvantages attached to limited liability companies.
I mean that in these latter work is always expensively done, and that
much of the profits are swallowed up by secretaries, directors, &c.,
besides which, generally from interested motives, the Teas are sent
home for sale which private planters know from experience is _not_ the
best plan.

But to return to the share list. The very many gardens held by firms
or private individuals are absent, and inasmuch as many of these were
begun more lately, and consequently, the blunders made in other gardens
were avoided, it is evident that _their_ position, if it could be
ascertained, would give the true picture needed.

There is one class of plantations which it would be by no means fair to
include. I mean those gardens bought for a mere song during the panic.
On many of these necessarily enormous profits have been made, but it
proves nothing, inasmuch as the profits, to be legitimate profits for
criticism, should on the debit side include the whole cost incurred in
making the plantation. To form a fair appreciation of the profits Tea
planting can give, we must select gardens constructed after knowledge
on the subject was attained, where good management, combined with
economy in all details, has been carried out, and where the necessary
natural conditions for success exist--and such are rare.

But first let me explain what I mean by the “necessary natural
conditions for success.” Manageable areas; flat or nearly flat land
for the garden; a good class of indigenous and hybrid plants; local
labour, or anyhow a good proportion of this; facilities for manuring; a
good soil; a good Tea climate; and cheap means of transport constitute
these, and where they exist I hold Tea _must_, and _does_, pay well. I
don’t believe in plantations of 600 or 800 acres; some of these pay,
but they would pay much better if reduced in size. A garden of 300
acres, yielding even at the rate of four maunds an acre, will pay much
better than another of 500 acres, yielding but two and a-half or three
maunds.

The reason is obvious, the larger produce is against a smaller
expenditure. Were I to commence a Tea plantation to-day, it should
not exceed 300 or 400 acres in size. This passion for large areas is
the rock on which, more than any other, Tea Companies have wrecked
themselves; experience has already shown this, and will show it more,
as time goes on.

Flat land for Tea gardens is a great desideratum. Steep lands are
difficult to cultivate; the soil is continually washing away from the
roots of the plants; it is impossible to manure them successfully, and
the consequence of all this is that the Tea bushes do not thrive.

The Chinese plant gives a small and inferior produce, the indigenous
and hybrid kind a larger and very superior one; thus I think the latter
one of the “necessary conditions for success.” On the other points,
with the exception of manuring, nothing need be said, inasmuch as
their necessity is evident; but on the point of manure I must say a
few words. The Tea plant is being continually denuded of its leaves;
nothing is returned to the soil; and consequently in process of
time that soil is exhausted. It was held once that manure destroyed
the flavour of Tea. This idea, at variance with all agricultural
experience, is now completely exploded, like many others received from
the Chinamen who first came from the Flowery Land to teach the art of
Tea cultivation and Tea manufacture to the Indian public. Many of them
had never perhaps seen a Tea bush, anyhow in many respects theirs was
faulty teaching, and all experienced planters are convinced, and it is
truth, that more knowledge on Tea exists in India than China at the
present time.

But to return to the subject of manure. It is, and is now generally
allowed to be, a necessity to the lengthened and successful maintenance
of a plantation. Means for its production are now largely adopted in
Assam and Cachar, and the results will be a yield per acre the most
sanguine have never dreamt of. Chittagong, on this head, has great
advantages; manure in any quantity can there be procured for a trifle,
and the results have shown its great value.

We have scarcely yet entered on the third stage to which any new
speculation, after the two first (the wild venture, and the unreasoning
panic have passed), tends; but as knowledge of the financial results of
Tea plantations in the hands of private firms and private individuals
increases, that third stage will dawn, if it has not done so already.
It consists in a sober appreciation of the subject opposed to both the
extremely exulting and depressing views passed through, and when it
arrives, the great and successful future of Indian Tea will be only a
question of time.

FOOTNOTES:

[84] Note to Third Edition.--The above three statements, quite true
when written, are not so now. The heavy fall in the value of both
Indian and Chinese Teas in 1877, while pressing hard on the Indian
producer, has certainly had the one good effect for him of making
Indian Teas more widely known. They _are_ generally known now, in many
cases sold pure as Indian Tea, and used by all retail dealers to give
the body, or strength lacking in most Chinese kinds.

[85] Note to Third Edition.--Yes; the demand has largely increased,
but, alas! production has increased in a greater ratio. In short, the
supply exceeds the demand, and hence the low prices now ruling. As
regards the use of Indian Teas, so much have the English public been
now made familiar with their flavour, they, as a rule, reject any Teas
which have it not more or less. In fact, the English public, as I
predicted years ago, have now begun to like the new flavour, and even
pure Indian Teas are now relished by many.

[86] Note to Third Edition.--The imports have been as follows during
the last three years:--

  1875           25,615,000.
  1876           29,384,000.
  1877           31,882,000.


[87] Note to Third Edition.--The annual consumption of all Teas in
Great Britain in 1877 was:--

  Chinese       158,000,000
  India          28,000,000
                -----------
  Total         186,000,000


[88] Note to Third Edition.--In 1876 the _average_ prices of the two
kinds in bond were:--Chinese, 1_s._ 2_d._; Indian, 1_s._ 10_d._ per lb.



_ADDITIONS TO THE FOURTH EDITION._



CHAPTER XXXII.

COUNTRIES OUTSIDE CHINA AND INDIA THAT PRODUCE TEA.


So much has the industry marched since the Third Edition was published
in 1878 that I think it well to add the following pages to my book.

I will first consider the countries _outside_ India and China which
produce Tea, or wish to do so.


CEYLON.

This is likely to prove a formidable competitor. As far as I can
gather, Tea plants (of both the Assam and China kinds) were introduced
into Ceylon in 1841, but it is only during the last few years Tea
planting has been taken up in earnest. A Mr. Shand, who seems to have
studied Tea in Ceylon, estimates 500 lbs. per acre as the produce when
in full bearing. This is 6-1/4 maunds, and though less than the best
Indian gardens give, it is considerably above the average all over
India. Ceylon Tea finds a ready market in London. The parcels vary
much, as they do from India, but in the past year (1882) many very
desirable lots were sent home. I believe, take it all in all, Ceylon
Tea is no better, and no worse, than Indian Tea.

With Tea prices as they are to-day, I would not myself _commence_ Tea
cultivation in India, Ceylon, or _anywhere_. I feel sure, therefore,
if Ceylon planters rush into Tea, as they did in India in times past,
they will regret it. But I hear that made Tea gardens can be bought
there cheap, and under these circumstances Tea will probably pay the
purchasers well.


JOHORE.

H.H. the Maharajah there has started a small Tea garden, but as there
are in it only two acres of Tea, the whole thing is quite an experiment
yet. The climate is said to be favourable, and land easily acquired.
Cheap labour is the difficulty. May it long continue so! There is too
much Tea already; the low prices ruling result simply from supply
exceeding demand. Thus I hope Johore will _not_ produce Tea. The
following is from the _Tea Gazette_:--


  TEA IN JOHORE.

  We have lately published several articles on the subject of Tea in
  Johore and the prospects of Tea plantations in the Malay Peninsula.
  The soil and climate are all that can be desired for the successful
  cultivation of the Tea plant; there is abundance of land lying
  idle which can be obtained on advantageous terms; but all hopes of
  establishing the Tea industry on a prosperous footing are frustrated
  by the want of cheap Indian labour. A correspondent writes on this
  subject to the _Ceylon Observer_ as follows:--

  “I was pretty well disgusted with Johore at first. I got such fever
  as nearly finished me up twice. A new comer from Ceylon says he had
  Wellawaya fever and all other fevers in Ceylon, but he never felt
  anything to come near the severity of Johore fever. Liberian coffee
  does first-class in the low country. Cocoa is being tried with
  apparent success. Tea is also promising. You may have seen about some
  samples sold in London, at a high figure. All this is nice enough,
  but what’s the good of it when we have not a plentiful supply of
  labour over which we can have complete control? So you see, the
  burden of my letter is an indefinite supply of labour.”

  Strenuous efforts are, however, being made to arrange with the
  Government for the importation of labourers from this country, which,
  if successful, will inevitably result in the cultivation on a large
  scale of the Tea plant in Johore. In another column will be found a
  description of the Maharajah of Johore’s experimental Tea plantation
  at Tanjong Putri, Johore.


JAPAN.

Sends its Teas principally to America. The Tea is of a greenish nature,
and experiments to manufacture black Tea have not, it seems, been
successful. The following should give a hope to Indian Tea planters:--


  JAPAN TEA.

  To the Editor of the _Japan Herald_.

  Dear Sir,--I read your article on Tea contained in last Saturday’s
  paper, anent the deterioration in quality of one of the country’s
  principal articles of export, and can fully confirm the chief points
  contained therein.

  But in addition, from my personal experience, there appears to exist
  a steadily increasing disregard of care in the preparation of the
  leaf up country, and the evil, though existing for the last three
  or four years, is much more manifest this season, and is worthy of
  being brought under the especial notice of parties interested in the
  welfare of this country’s produce.

  I submit for your inspection a sample of coarse leaf sifted out of
  a parcel of good quality, and the proportion of similar stuff in
  the chop amounts to fully 3 per cent., very much affecting the good
  appearance of the fired leaf. This defect no doubt arises from the
  attempted production of too great an amount of cured leaf for each
  hand employed per diem in the process, to be attributed no doubt
  to the enhanced cost of labour in the interior. But the defect is
  of vital importance for the future of Japan Teas in America. The
  buyer for distribution amongst consumers in that country is greatly
  influenced by the “appearance” of the leaf, despite its relative
  intrinsic quality in infusion, in comparison with a Tea of worse
  appearance, hence the high facing and colouring at present so much
  in vogue. If the Japanese producers continue the present style of
  manufacturing the leaf up country, so surely will Japan Teas decline
  in favour in America, as the foreign shipper here cannot make up the
  leaf prepared up country to the standard required by the American
  buyers, and with the prospect of a possibility of Oolongs and even
  blacks becoming ere long dangerous rivals with consumers in the
  United States, it behoves the Japanese Tea growers to turn their
  attention towards an improvement in production in their own country
  before they attempt to rival foreign competitors at this side.--Yours
  faithfully,

          YAMATO.

  Yokohama, _Aug. 19, 1881_.

  [We have inspected the sample of coarse leaf referred to in the above
  letter, and though we cannot pretend to any critical knowledge of
  Tea, we can confirm the statements of our correspondent, and hold the
  specimen at our office, where it can be seen by anyone desirous of
  doing so.--_Editor, Japan Herald._]

I know not where this next extract came from, but it appears they
understand adulteration in Japan:--

  Mr. Yanagiya might, however, have gone a step farther, and have given
  particulars of the various analyses, and have mentioned that the
  “leaf” of the various samples showed a large proportion of leaves
  quite different to those of the Tea shrub, and for the presence
  of which not even the astute foreigner--that bugbear of Japanese
  commerce--can be held accountable.

  We have heard this season loud complaints of the presence of leaves
  entirely distinct from those of the Tea plant amongst purchases.
  These consisted principally of wisteria, willow and a species of ash,
  but the native growers were impartial, and several other species of
  shrubs also contributed their quota to the frauds practised by the
  Japanese.

  The probable reason of the falling off in the quantity of one of the
  leading articles of export from Japan is not difficult to guess at,
  nor can the Japanese say that they have not received full and timely
  warning of the danger threatening the popularity of Japan Tea. A
  reckless over-production, excessive and close picking of the shrubs,
  and great carelessness in pruning and manuring the tree--caused,
  no doubt, in no small degree by the high rate of wages in the
  interior--is militating against the realisation of a good crop,
  and the peasant is too intent upon immediate profits to forego the
  picking of the third crop of a season under existing circumstances.
  The result of all this has been that--at the close of last season--a
  quantity of worthless leaf was poured upon the market, finally sold
  at almost nominal figures, and shipped across to the United States,
  where it remains an incubus on the figures of stock, and a source
  of future abhorrence to any unfortunate purchaser towards anything
  bearing the name of Japan Tea.

The following is from the report of the Japanese Consul at San
Francisco. I should have thought the Americans were too sensible “to
prefer coloured Teas:”

  It has however come to my knowledge that in the Eastern States the
  Tea was analyzed, and adulteration was discovered; such as the
  admixture of other leaves and poisonous ingredients which are used
  for colouring the Tea before it is exported, and that the markets
  in the Eastern States being overstocked, no Tea, unless of the best
  quality, can find purchasers. This is a very deplorable state of
  affairs. The colouring is made by the foreign merchants residing
  in Japan, for Americans prefer coloured Tea, and a few Japanese
  merchants may have imitated them, and exported on their own account.

The exports of Japan Tea to America have declined from seventeen to
fourteen million in one year! Not strange if all the above is true.

The following from the _Tea Gazette_ bears out what I say above as to
Japan black Tea:--


  JAPAN BLACK TEA.

  Mr. Consul Euslie writes from Kanagawa (Japan) as follows concerning
  black Tea:--This has, on the whole, proved a failure, although the
  production continues on a limited scale. The climate and soil of this
  country appear unfitted to the growth of plants producing a leaf of
  the quality necessary to make good black. Teas resembling good leaf
  congous can be made with good and even handsome leaf, several samples
  being in appearance very similar to Indian Teas of pekoe class,
  but lacking strength, and not being nearly equal to good Chinese
  Foochow Teas in that respect. A small amount of these Teas has been
  shipped to Germany on native account, a German financier providing
  the necessary funds; but thus far the outcome of these shipments has
  not transpired. The results generally of 1881 have not proved as
  satisfactory as those of the preceding year; the whole crop, and more
  particularly the first picking, shows signs of hasty and careless
  preparation. The amount of Tea exported from Japan was decidedly
  in excess of the requirements of the United States and Canada, and
  a considerable portion of the shipments for the year had to be
  sacrificed at prices which did not cover laying down cost.

All this is hopeful for our Indian Teas, as we _can_ manufacture the
greenish Tea they like--that is, we can do it if they won’t take our
black, but they have begun to do so.

The above mode of manufacture in Japan is new to us in India.

  The process of steaming the Tea is as follows:--As soon as picked
  it is at once steamed, all damp or wet leaves being thrown on one
  side, excepting those that may be a little wet with dew. In order
  to obtain the proper application of heat, a few leaves are put into
  a shallow basket, spread out evenly, and the lid put on; the basket
  is then placed over a charcoal fire box or stove, a perfume is at
  once perceived. When the greeny smell has subsided, the leaves are
  removed, spread on a piece of new matting, and fanned briskly so as
  to draw out the heat. After the lapse of some little time the Tea is
  placed in a tray, and then undergoes a firing process, the length of
  which is regulated either by the minute hand of a watch or the beats
  of a pulse, and depends a good deal on the manipulator’s own ideas.


JAVA.[89]

Much of this Tea goes to Holland and Northern Germany. I know
Tea cultivation in Java is carried out very carefully and very
successfully, but this one fact is all I know as to Java or its Teas.


AMERICA.

That the proper climate for Tea can be found there (a huge Continent to
choose from!) goes without saying. But equally sure is it that Tea will
not pay except labour is cheap.

By the extract below, it appears Georgia has been selected for
experimental Tea cultivation, and I doubt not it is a good selection:--


  TEA PLANTING IN AMERICA.

  Successful experiments have been made in this branch of cultivation
  in the United States, as is shown by a report just published by
  Mr. Jackson, a Scotch gentleman now settled in America, who was at
  one time manager of the estates of the Scottish Assam Company. The
  Commissioner of Agriculture has, at Mr. Jackson’s advice, selected
  a tract of land in Georgia for an experimental farm, on which the
  raising of Tea on an extended scale will be carefully and thoroughly
  tried. Samples of the Teas already produced by Mr. Jackson have been
  sent to Messrs. Thompson, tea merchants, Mincing Lane, London, to be
  examined. The reply was that--“They represented Teas of a high type.
  The flavour, though not strong, is remarkably fragrant. In appearance
  they resemble Indian Tea, but the flavour is more like that of the
  finest Chinese black Tea, or of the hill Teas of India.”

No reason why the Teas should not be good, but the labour difficulty
will, I think, prevent Tea paying there, as elsewhere in America, for
Mr. Jackson himself, who continues the above, asks further on, “Can we
afford to pay our labourers four times as much as they pay in India and
still make Tea a success?” He, strange to say, tries to prove “yes.” I
say no, a thousand times no, in spite of all Mr. Jackson says. I like,
however, to give both sides of a question, and so will let Mr. Jackson
speak for himself:--

  The stock cry continually raised against Tea culture in this country
  is, how can you raise Tea in a country where wages are so high?
  You can cultivate Tea at a profit only in a country where labour
  is at the lowest possible minimum, and so on. And so it is taken
  for granted that the Tea culture is to be allowed to retain its
  antiquated forms and systems for all time, and that the skill and
  intelligence of a civilized nation can do nothing to raise it to a
  level with corresponding branches of agriculture, such, for instance,
  as rice growing. What would the people of South Carolina say if
  told that the only way to cultivate rice at a profit was to sow all
  their seed in nursery beds and, when sprouted, to transplant their
  entire crop, seed by seed, by hand, as is done in India? What would
  a Minnesota farmer say if told that the only implement with which
  he could profitably rear a crop of corn was the hoe, wielded by an
  attenuated skeleton of a man? If the hundreds of wealthy Tea planters
  in Assam were told that they must return to the original system of
  manufacturing their Teas by hand, they would throw up their farms in
  despair. Seventy million pounds of Tea are now annually manufactured
  by machinery in India and Java, and I have satisfied myself that
  green Teas, suitable for the American market, can be manufactured at
  one-third the cost of the black Teas prepared by machinery for the
  English market. There is but one division of Tea culture into which
  the labour question would enter at all, and that is the picking of
  the leaves. Everything else can be carried on with the mechanical
  precision of the cultivator, reaper, and floutring mill. Is not the
  real truth of the matter to be found in the fact that the American
  people know nothing--absolutely nothing--of the modern system of Tea
  culture and manufacture, and are therefore in no position to form a
  sound judgment of the possibilities of their country and countrymen
  in regard to Tea? I say again, as I have often said before, that the
  question of labour will prove no barrier to successful Tea culture
  in America. Let any who are interested enough in this subject to
  feel sceptical about it favour me with a call at No. 229 East Fourth
  Street and I will take pleasure in showing them what achievements
  modern skill and mechanical genius have already attained and what may
  very easily be accomplished in America. I believe that then there is
  a bright future in store for successful Tea culture.

In another place Mr. Jackson says he has always cultivated with
ploughs, and done it successfully. Naturally this would make his
cultivation much cheaper, and it is high time, as I say elsewhere, that
we in India should try and do the same thing.

We, all the world knows, how ingenious, how inventive the Americans
are, and thus it is _possible_ they may by the use of machinery for
all branches of manufacture, by improved steam-ploughs and other
agricultural instruments which shall dispense with hand labour for
cultivation, so cheapen the cost of Tea that its production will pay
in spite of the high rate of wages ruling. Only in this way, however,
_can_ the industry succeed in America, and if it be done (I hope it
may, for we in India shall then benefit by the ideas carried out), the
United States will add one more laurel to the many they have achieved
already in other branches of commerce.


NATAL.

Tea here too! Where, alas! is it not? The following is in a report from
Natal:--

  We have glanced at the past and seen the present condition of the
  Tea enterprise. The most important matter is still before us--the
  future of Tea in Natal. It must be remembered that an industry may
  be profitable to encourage for local consumption, and yet fail when
  it comes to be exported. This production (Tea) must be looked at as
  one which, if it progresses, must shortly be exported. Three hundred
  acres of Tea in full bearing will supply to the full the present need
  of the Colony and its surroundings, _i.e._, taking the import returns
  as our guide; and even with increasing demand, that demand can soon
  be met. Therefore the importance attaching to the question, “Can
  we in Natal grow Tea to pay, so as to compete with other countries
  in the markets of the world?” The Tea-growing districts of India,
  till lately confined to Assam, Cachar, and neighbourhood, have now
  been extended to the Neilgherries and Ceylon, and these places,
  till lately confining themselves to coffee, are leaving that most
  precarious crop and growing Tea and cinchona. Both these districts
  at present export Tea to England. It is said that Tea is to be
  introduced into Queensland and the northern territory of Australia;
  and that the Southern States of the Union and California contemplate
  Tea growing. Are we justified, then, in believing that Tea may be
  profitably exported from Natal? Before considering this, we have
  to bear in mind that Tea is placed in the London market in large
  quantities at a very low price; China Teas as low as 5-1/2_d._ per
  lb. in bond, and Indian Teas as low as 8_d._ per lb. in bond; but
  again the price extends to 3_s._ and even 4_s._ per lb. in bond. All
  these things have to be weighed and well considered by anyone before
  embarking in this enterprise to any great extent. In all matters of
  agriculture the labour question is the foremost, after it is known
  that the plant will flourish. To attempt to base the success of this
  particular industry upon any other than that of coolie labour would
  be foolish, as we all know that Kafirs are not to be relied upon;
  therefore the cost of labour has to be considered. We well know
  that all the Tea estates of India have to be supplied with imported
  coolies--the natives of the districts concerned will not work unless
  casually--but the cost of coolies in Natal must be more than the
  cost of coolies in Assam--therefore in that item the advantage must
  be in favour of the Indian planter. Countries such as Queensland
  and the United States of America must either import labour, or pay
  a much higher rate than we do; hence so far as these countries are
  concerned, Natal will hold its own.

I have not done yet. Here is another place where it seems they mean to
try Tea. I do hope that climate, soil, labour, something will be found
unsuitable.


  FIJI TEA.

  Mr. J. E. Mason, of the Alpha Tea and Coffee Estate, Taviuni, Fiji,
  has forwarded to Mr. J. O. Moody, the expert, of Melbourne, samples
  of the first Fijian Tea produced in his part of the world; at the
  same time writing that early next year he hoped to pluck off 30 acres
  planted with Tea, and that the samples sent were hastily made in a
  barrel with a frying-pan of charcoal. Mr. J. O. Moody reports:--“Fiji
  Pekoe leaf: Handsome, small, even, golden tipped, evenly and well
  fermented. Fiji Pekoe liquor: Very strong, full, rich, and pungent
  pekoe flavour, thick, with deep red infusion. An invaluable Tea for
  mixing, and worth about 2_s._ 6_d._ per lb. in bond. Fiji Pekoe
  Souchong leaf: Well made, wiry, twisted, rich, black tippy leaf,
  evenly and well fermented Fiji Pekoe Souchong flavour, with good,
  bright, red infusion. A fine Tea to drink alone, and worth about
  1_s._ 9_d._ per lb. in bond. These Teas have the character of good
  Ceylon growths, and are in every respect suitable Teas for general
  consumption, and such samples are sure to meet with ready sale in
  Australasia or Great Britain.”

Here again, I am told, the labour question is the doubtful point. Tea
cannot be made to pay without cheap labour, and the sooner all these
new Tea countries learn the lesson the better for the pockets of the
projectors. I may add, the better for the Indian planter’s pocket too,
for _any_ increase in the supply is hurtful.

In closing this chapter, I would give one word of advice to intending
Tea planters in India, or indeed anywhere. There is too much Tea
already, why plant more? If you must “go into Tea” you may do so and
probably make money, but it will not be by _planting_ it. If you look
about you can buy a plantation ready made for far less than you could
make such, and in doing so there is no reason why it should not pay,
and pay well. If you make a garden you will have five or six years to
wait for any return; you attempt what requires knowledge and experience
to succeed in, and begging your success, who can say what the market
will be then?

I have far from exhausted the subject of these new Tea fields, but my
space is limited, and several other points demand attention.

FOOTNOTE:

[89] Much about Java and its Teas can be found in a book entitled
“Java, or How to Manage a Colony,” by J. W. B. Money. Crown 8vo, 2
vols. Hurst and Blackett.



CHAPTER XXXIII.

STATISTICS REGARDING INDIAN TEA.


As early as 1780 a few Chinese plants were to be seen in Calcutta
cultivated by a Colonel Kyd.

The possibility of cultivating Tea in India was first mooted in
1835-36, and the Indian Government started an experimental garden at
Lukimpore (Assam) at that time.

Indigenous Tea was first discovered in Assam by a Mr. Bruce in 1830.

In 1845 and following years the Government imported large quantities of
China seed and established nurseries on the Himalayas.

Tea planting was commenced in India by the Assam Company about 1840,
and the cultivation was undertaken in other districts in the following
years:--Kumaon and Gurwhal, 1850; Cachar, 1855; Dehra-Dhoon, 1855;
Sylhet, 1857; Kangra, 1858; Darjeeling, 1860; Terai, 1860; Chittagong,
1860; Neilgherries, 1862; Chota Nagpore, 1872; Dooars, 1875.

Thus it may be said Tea planting in India dates from 1840 by one
Company, but 1850 by individuals.

The following figures show the imports of Indian Teas into Great
Britain since 1870 in millions of pounds:--

  +----+----------+
  |    | Millions |
  |Year|of Pounds |
  +----+----------+
  | 70 |    13    |
  | 71 |  15-1/4  |
  | 72 |    17    |
  | 73 |  18-1/4  |
  | 74 |  17-1/4  |
  | 75 |  25-1/2  |
  | 76 |  29-1/4  |
  | 77 |  31-3/4  |
  | 78 |    36    |
  | 79 |  38-1/2  |
  | 80 |    44    |
  | 81 |  45-3/4  |
  | 82 |  54-3/4  |
  +----+----------+

I may here remark that while the imports of Indian Teas have, since
1877, increased by 23 million pounds, the imports of China Teas have
increased by 4 million pounds only in the same time.

The deliveries and stocks were as follows for 1881 and 1882 in millions
of pounds:--

                                 1881.    1882.
  Deliveries                    48-3/4   50-1/2
  Stocks on 31st December       18-1/2   21-3/4

During the _last 3 months of_ 1882 the deliveries averaged 5 million
pounds per month. In January, 1883, they were 5-1/2 millions, and in
February (I write in March) 5 millions.

Deliveries at this rate mean 60 millions a year.

    I estimate Indian produce for 1883 at about 62     millions.
    Deduct the probable quantity to be sent  }
  to countries outside the United Kingdom,   }   5-1/2   „
  and local consumption in India             }
                                              ------------------
    Leaving available for the home market    }  56-1/2 millions.
                                              ==================

Thus, if deliveries continue at the present rate, demand must soon
equal, if not exceed, supply, and the consequence naturally must be
enhanced prices, which, however, will surely to some extent check the
deliveries.

There is, however, a hopeful feature regarding Indian Teas. The taste
for them is increasing greatly. A very small per centage of the public
drink them pure (a large per centage in Ireland), but the public
generally are now accustomed to the strength attained only by mixing,
say one third of Indian to two thirds China, and _will_ nothing weaker.
Thus retail dealers must continue to use them, and thus though, as
remarked, increased prices _will_ check deliveries, they will not do so
with Indian Teas to the same extent they would with China.

Another hopeful feature is (for details see further on) a growing
demand for Indian Tea is now established in Australia, and has quite
lately commenced in America. Thus, I think, the increased produce from
India (which in no case can be large for some years[90]) will probably
be met by this outside demand, leaving no greater quantity than now
available for the home market. True Ceylon (a new field) will increase
the supply, but it will not be by much for some time. Take it all in
all, I look hopefully at the prospects of Indian Tea in the future. I
never anticipate a range of prices as good even as ruled in 1881, but
a good deal better than we had in 1882, and thus enough to make the
industry a paying one.

I have lately received a valuable paper on Indian Tea statistics from
Messrs. Gow and Wilson, Indian Tea brokers. I cannot transcribe the
diagram they allude to, but otherwise I give the complete paper as sent
me:--

  19, Little Tower Street, Mincing Lane,
      London, _15th February, 1883_.

      “INDIAN TEA STATISTICS.”

  Dear Sir,--Now that the annual figures are made up, we beg to submit
  a statement showing the continued progress made by Indian Tea in
  public estimation, together with comparative figures relating to the
  consumption of China and Indian Teas,--And remain, dear Sir, yours
  faithfully,

  GOW AND WILSON, _Indian Tea Brokers_.

  The very considerable increase in the home consumption of Indian
  Tea during the last quarter of 1882, and January this year, once
  more attracts attention to the growing importance of India as a
  field of production, and the increasing appreciation of the British
  public for Indian Teas, whether used alone or mixed with China sorts.
  Notwithstanding the check to consumption in the early part of 1882,
  when Indian medium and common Teas were just 50 per cent. dearer
  than they now are, the average monthly deliveries of the first three
  months were 3,670,000 lbs., or 230,000 lbs. a month more than the
  average of the first quarter of 1880, with prices much the same at
  both periods. Quotations last year receded step by step, and, as
  prices dropped, so we found the consumption grew, till for the last
  quarter of 1882, with its very low range of prices, the average
  monthly deliveries reached the unprecedented figures of over 5-1/4
  million pounds.

  The average monthly deliveries in each quarter of the last five years
  have been as follows:--

  (In thousands of lbs., 000’s omitted.)

           Jan-Mar.   April-June.  July-Sept.   Oct-Dec.
  1878      3,216       3,129        2,869       3,041
  1879      3,444       2,688        2,461       3,155
  1880      3,441       3,418        3,522       4,228
  1881      4,197       4,172        3,824       4,094
  1882      3,670       4,125        4,116       5,254

  During the year 1878, out of every 100 lbs. of all descriptions of
  Tea consumed in this country, 23, or one in about four-and-a-third,
  was Indian Tea. Last year the proportion was 31 per cent., or nearly
  one in three.

  These figures show, in the clearest manner, how steadily Indian Tea
  is becoming popular.

  The unevenness of the quarterly deliveries of China Tea in the year
  1878 and 1879 is due to the apprehensions felt in those years that
  the duty would be increased. In consequence, clearances were hastily
  made before the Budget announcement, and the deliveries immediately
  after sank to very low figures, increasing again as stocks of
  retailers were depleted. In March, 1880, again there was a pressure
  to clear Teas, which brought up the total deliveries of the first
  quarter to a high level. A considerable check was given to deliveries
  of Indian Teas during the latter part of 1881 and the early part of
  1882 through the rise of prices during that period.

  The most noticeable feature of the last three calendar years is the
  stationariness of deliveries of China Teas at the reduction from the
  level of both 1878 and 1879. Approximately the deliveries of China
  and Indian Teas in the five years ending 31st December in each case
  may be given, in millions of pounds, as follows:--

             1878.   1879.   1880.   1881.   1882.
  China       121     125     115     111     113
  Indian       37      35      44      49      51

  The deliveries of China Tea have _receded_ from 125 million pounds
  in 1879 to 113 million pounds in 1882, while the home consumption
  of Indian Tea has _increased_ from 35 million pounds to 51 million
  pounds in the corresponding years.

  Notwithstanding the supply of Indian Tea for the season 1882-3 is
  estimated at the unprecedented figure of 55 million pounds against
  the actual imports in the previous two season years of about 50
  million and 46 million pounds respectively, it appears not unlikely
  the consumption of 1882-3 will once again, as in 1880-81, overtake
  the supply and reduce the stocks by July next to the equivalent of
  less than three months’ deliveries.

  We find that there has been received to the 31st December last,
  33,218,000 lbs., leaving to arrive 21,782,000 lbs. for the six
  months ending 30th June next, to make up the estimated supply of
  55 million pounds which, according to Indian advices, will be
  available for shipment to this country. The imports of the current
  six months, therefore, will be but slightly in excess of those of
  the corresponding half-year, when 20,948,000 lbs. were received; for
  although the shipments of 1882-3 are expected to show an excess of
  more than five million pounds over 1881-2, exporters this season have
  hurried forward their Teas early and in the six months ended December
  31st the arrivals in the United Kingdom were 32,218,000 lbs. against
  only 28,947,000 lbs. in the corresponding half of 1881.

  The significant feature of the movement, however, is the very
  agreeable surprise, month by month, caused by the publication of the
  delivery figures. These compare for the last few months as follows:--


  _Deliveries of the last Four Months compared with corresponding
  Months._

             October.    November.    December.     January.
               lbs.         lbs.         lbs.         lbs.
  1882-3    5,132,000    5,174,000    4,457,000    5,502,000
  1881-2    4,353,000    4,205,000    3,724,000    4,104,000
            ---------    ---------    ---------    ---------
  Increase    779,000      969,000      733,000    1,398,000

  A continuance of similar large increases is most probable, especially
  as we compare with the relatively small deliveries of February to
  April inclusive of last year, when only 10,489,000 lbs. were taken
  from warehouse, against 12,782,000 lbs. in the same months of 1881--a
  decrease of 2,293,000 in three months. Part of the decrease may be
  attributed to the then higher range of medium and common Indian Teas.

  The net result of the above statistics is that--with no materially
  larger arrivals visible, even with the liberal allowance of 5,000,000
  lbs. increased shipments for the crop year--the deliveries give every
  promise of showing very considerable expansion, and, as we have said,
  threaten for the complete year to more than absorb the extra supply.

  To enable anyone to check and form an independent opinion on the
  forecast we venture to give, we present below the imports and
  deliveries, half-year by half-year, for the four seasons 1879-80 to
  1882-83 inclusive, with the totals of each crop year:--


  IMPORTS OF INDIAN TEA.

                 1879-80.      1880-81.      1881-82.      1882-83.
                   lbs.          lbs.          lbs.          lbs.
  1st July to}
    31st Dec.}  23,537,000    29,142,000    28,947,000     33,218,000

  1st Jan. to}
    30th June}  15,868,000    16,819,000    20,948,000 [91]21,782,000
                ----------    ----------    ----------     ----------
  Season year   39,405,000    45,961,000    49,895,000 [91]55,000,000


  DELIVERIES OF INDIAN TEA.

                 1879-80.       1880-1.       1881-2.       1882-3.
                   lbs.           lbs.          lbs.          lbs.
  1st July to}
   31st Dec. }  16,847,000    24,352,000    23,755,000     27,109,000

  1st Jan. to}
   30th June }  19,465,000    25,106,000    23,386,000 [92]28,802,000
                ----------    ----------    ----------     ----------
  Season year   36,312,000    49,458,000    47,141,000 [92]55,911,000

  These figures show an estimated supply to the end of June next of
  21,782,000 lbs., and an estimated consumption of 28,802,000. The
  former is based on the statements that the available supply for
  the United Kingdom will be 55 million pounds, and the latter on
  the actual delivery to January inclusive, and estimated average
  deliveries of 4,500,000 lbs. a month for the remainder of the
  half-year. Having these figures before us, we can proceed to
  calculate the effect on stocks.

  At 31st December last we had in warehouse            lbs.
    a stock of                                      21,716,000
  Add six months’ imports to June 30th              21,782,000
                                                    ----------
          Total supply                              43,498,000
  Deduct estimated deliveries six months            28,802,000
                                                    ----------
  Leaving probable stock at 30th June               14,696,000
  Which will compare with (at 30th June, 1882)      15,991,000

  In stating the deliveries at an average of 4,500,000 lbs. for the
  next five months, we have taken this amount merely for the purpose
  of arriving at a conclusion. In case the deliveries of the months
  of February to June inclusive average, as is generally expected,
  5,000,000 lbs. a month, the stocks at 1st July next will be under
  12,200,000 lbs., or 3-3/4 million lbs. less than at the corresponding
  date.

  We can but consider this a very healthy outlook, especially as it is
  simultaneous with the estimated decrease in the supply of China Tea,
  and the possibility that the shipments to the United Kingdom may not
  reach the estimate. With respect to the latter contingency, we must
  recollect that new markets are being rapidly developed for Indian
  Tea. Australia, America, and other parts than the United Kingdom took
  over three million pounds from 1st May to 31st December last year,
  compared with less than a third of that quantity shipped thence from
  India in the corresponding period of 1881. A continuance of this
  rapid rate of outside demand would considerably curtail our supply,
  and develope the growing Indian industry.

      GOW AND WILSON,
  _19, Little Tower Street, London, E.C._

The diagram omitted shows as follows: it gives the results quarterly,
I only give them yearly in millions of pounds.


CONSUMPTION OF CHINA AND INDIAN TEAS IN THE UNITED KINGDOM FOR THE LAST
FIVE YEARS.

  -------+--------+--------+--------+--------+--------+-----------------
         |  1878. |  1879. |  1880. |  1881. |  1882. |
  -------+--------+--------+--------+--------+--------+-----------------
  China  | 128-1/2| 125-1/4| 114-1/2| 111-1/4| 114    |}All are Millions
  Indian |  36    |  35-1/2|  43-3/4|  48-1/2|  51-1/2|}  of Pounds.
  -------+--------+--------+--------+--------+--------+-----------------
  TOTALS.| 164-1/2| 160-3/4| 158-1/4| 159-3/4| 165-1/2|
  -------+--------+--------+--------+--------+--------+-----------------

Thus, while China Tea consumption has decreased in five years by
fourteen and a-half millions of pounds, Indian has increased by fifteen
and a-half millions!

It must be remembered that the former table given at page 194 deals
with imports, this with consumption, and thus the difference in the
figures.

The following extract from the _Tea Gazette_ (January, 1883) is
interesting in a statistical point of view:--

  The exports to Australia (which, as it is well known, have increased
  more than twenty-fold in six years) now occupy a position only second
  to that of the United Kingdom; and if the P. and O. Company would see
  its own interest, it would facilitate by every means in its power so
  important a development of a great industry.

  The Straits Settlements, in spite of their proximity to China,
  took last year ten times the quantity they took six years ago;
  and Persia, strange to say, has taken 54,712lbs. against 334lbs.
  in 1876-77--Turkey in Asia and Egypt, together, having taken also
  21,488lbs. against 886lbs. in the same period.

  Mr. Liotard is of opinion that these are not a tithe of the openings
  that might be found; and it is to be hoped that at the Amsterdam
  Exhibition and elsewhere the Tea Syndicate will in no way relax its
  efforts.

  The imports of China Tea to India have, in the six years also above
  referred to, increased about 60 per cent. Speaking on this point, we
  are of opinion that the Syndicate might well make efforts to increase
  the local consumption of Indian Tea to the replacement of Chinas, and
  we feel sure that such organised exertion would be followed by very
  satisfactory results.

  The re-export of China Tea from India shows four times the quantity
  of six years ago--by far the greater quantity going to Persia and
  Turkey in Asia. Mr. Liotard thinks that the N.-W. P. and the Punjab
  might appropriate a good deal of this trade; and from the character
  of the Teas of these districts we are disposed to agree with him.
  The great increase of export from Karachi shows that this, to some
  extent, is being done.

  The abolition of the duty on China Tea imported to India, under
  the recent free trade policy, appears to have had a prejudicial
  effect on the planters in Northern India--who can ill afford it.
  The figures given at the commencement of the pamphlet show that
  the number of plantations in Northern India has increased in six
  years from 851 to 1,422, and the area from 4,246 to 7,466 acres;
  the outturn from 1,311,113 lbs. to 2,271,773 lbs. These figures
  speak of great activity in production, and show the necessity for
  every exertion being employed to open out new markets. A combined
  Syndicate for all the districts in Northern India, on the lines of
  the one now established in Calcutta, is suggested, but how far this
  is practicable we are not at present prepared to say.

The following, too, from the _Tea Gazette_ this year is in some
respects hopeful:--

  We see from the _North China Herald_ that the exports of Tea from
  Shanghai and the Yang-tse ports to England during the current year
  have fallen off some six and a-half million pounds (8-1/2 per cent.),
  and that there is also a decrease of some six million pounds in the
  quantity sent to America this last year--making a difference of
  nearly thirty per cent.

  On the other hand, there has been an increase of nearly three million
  pounds (45 per cent.) in the direct export to Russian ports. Two
  large cargoes--one of nearly three million pounds destined for
  England, and one of almost two million pounds bound for Russia--were
  lost, so that the real increase of China Tea sent to Russia is only
  one million pounds, which would reduce the increase to about 15 per
  cent. The decrease of Tea sent to England becomes even greater,
  reducing the receipts as compared with last year by nearly ten
  million pounds. Another aspect of the question must, however, be
  considered, namely, that the real displacement, _i.e._, in the amount
  of Tea destined for the English market, would be only six and a-half
  million pounds, and it is not safe to reckon on a recurrence of loss
  of such a heavy quantity by shipwreck.

  The decrease of China Tea sent to America is almost entirely in green
  Teas, there being only a falling off of 40,000lbs. in black, as
  compared with one of over six millions in green Tea. The falling off
  as regards the English market is much more evenly distributed between
  the two varieties, the difference being greater in that of black than
  of green Tea.

  While America takes from the Shanghai ports over four times as much
  green as black Tea, England, on the other hand, takes eleven times as
  much black Tea as it does of green Tea.

  These last facts might make it worth the while of planters in the
  North-West Provinces and in the Punjab to combine to exploit the
  American markets with half-chests of green Tea, for the manufacture
  of which those districts are famous, and for which there is so little
  demand at present in the Central Asian market. We merely throw out
  the suggestion, knowing that most of the planters in these parts
  could ill afford to risk much in such an experiment. It is possible
  that the Syndicate here, which already ships largely to America,
  might arrange to ship green Tea for such of the planters in Northern
  India as cared to join the Calcutta body.

  The American demand for green Tea is so large, that a quantity
  representing the entire outturn of Northern India would form but
  a small percentage of the whole, and if Indian green Tea from the
  Himalayas were taken up in that market, a demand for the whole
  quantity produced might easily arise. Whether it would ever be able
  to compete with China green Tea in the matter of price we do not
  know, and we should think it would be up-hill work, and attended with
  some loss--in the first instance at any rate.

The more Tea each individual drinks, the better doubtless for the
producers. It is satisfactory therefore to find the consumption _per
head_ is increasing in the United Kingdom as follows:--

  1870.     1875.     1880.
   lbs.      lbs.      lbs.
  3.81      4.44      4.59

Nearly 1 lb. per head more in 1880 than 1870!

A few figures as to Indian Teas in Australia and America will finish
this Chapter of Statistics.

The consumption of Indian Tea in Australia and the Colonies stands as
follows:--

  1880-81       Little under 3/4 of a million lbs.
  1881-82       Nearly        1       million „
  1882-83       Estimated     2       million „

This is a satisfactory increase, but when we consider how vast is the
great Australasian field, it stands to reason two millions is but a
small fraction of what it eventually may be.

The outdoor rough life, led more or less in the Colonies, makes its
inhabitants the largest Tea drinkers in the world. For instance, each
white denizen in New Zealand drinks nearly three times as much Tea as
each person in Great Britain!

The following was the consumption in lbs. per head in 1878:--

   United                                               New
  Kingdom.   Victoria.   N. S. Wales.   Queensland.   Zealand.
    4.66       6.92         7.53           9.16        11.05

I now give, in millions of lbs., the consumption in 1880 in the same
Colonies:--

  Victoria.    N. S. Wales.    Queensland.    New Zealand.
    5-1/2           5              2               3

But what vast tracts exist outside these. The total consumption of all
the Colonies must be very large. We make the best Tea in the world in
India, why should we not have a large share of the market?

The population of Australia is nearly 2-1/4 millions, and of Tasmania
and New Zealand nearly 3/4 million, say three millions in all, or say
three-quarters of the population of London. What a field exists there
for Indian Tea!


AMERICA.

The yearly consumption (Canada is included) is over eighty million
pounds, nearly all supplied by China and Japan. It is quite lately
Indian Teas have been sent to America; so far, their reception has been
favourable. But the Americans are accustomed to a greener Tea than we
make in India, and this will prove a difficulty. Still we can make the
Tea they like, _if_ they will buy it. It is early to speculate much as
to America, but I think we shall succeed little by little, especially
as in the States they are awaking to the fact that both China and Japan
Teas are adulterated.

In closing this chapter I must put on record the fact, known to all in
India, that the great success achieved in Australia, and the opening
thus early attained in America, is entirely due to the labours of the
Calcutta Tea Syndicate, and that I firmly believe, much as they have
done, they would have done still more had they been properly supported
by larger supplies of Tea by the planters in India, who, as a class,
are strangely blind to the advantages of co-operation. I can only hope
in this respect they will do better in future.

FOOTNOTES:

[90] It will not be large because much in the way of extensions has not
been executed lately. A higher range of prices will doubtless cause
more land to be cultivated, but no produce from such will be available
for four or five years.

[91] Estimated on basis of 55,000,000 lbs. available for shipment to
the United Kingdom.

[92] Estimated on average monthly deliveries, February to June
inclusive, of 4,500,000 lbs.



CHAPTER XXXIV.

MARKETS OUTSIDE GREAT BRITAIN.


I have forestalled a good deal on the above in the last chapter, so
this will be short, but, I hope, cheering.


AUSTRALIA.

This, from the correspondent of the _Tea Gazette_ in Melbourne, as to
the size of chests, should be attended to:--

  If the planter wishes to get his Tea direct into consumption, the
  packages must be small, to suit buyers. In the Colonies a large trade
  is done in 38lb. half-chests. They are within the purchasing power of
  a numerous class, and are easy to handle.

A fierce fight has been going on in Melbourne between the advocates of
China and Indian Tea. The latter say China Tea is often adulterated,
but this is disputed by the former. Of course _I_ cannot say which is
right, but chemical analysis, to which China Teas have been subjected
in Melbourne, would seem to prove that in some cases they are _not_
pure. We all know China Teas in London have, in several instances, been
pronounced unfit for consumption, so it is _possible_, of course, that
similar Teas are sent to Australia.

The Tea trade in China has taken alarm at our attempts on the
Australian market. This is what the _North China Herald_ (an organ of
the China Tea trade) said lately:--

  There are no squeezing mandarins in India; there is European
  supervision in the packing and firing of the leaf, and the
  plantations are connected with civilisation by the railway and the
  telegraph. Everything is done to give India an unfair advantage over
  China. Consequently, India tea of the same quality is far cheaper in
  London than the ill-regulated produce of Hankow and Foochow, and it
  is only the conservatism of the consumer, _who is not yet entirely
  habituated to the Indian flavour_, that prevents our losses being
  much heavier than they are. Every year this preference for the leaf
  that has been longer known is wearing away, and our buyers will soon
  have to reckon with its disappearance. As yet, Indian Tea is hardly
  taken on the continent of Europe at all; but here, too, it will
  penetrate sooner or later, as it is doing into America and Australia,
  and then there will be no corner of the earth where the sway of China
  Tea will be undisputed. Until foreigners can supervise the packing
  of the leaf in China as they do in India, the produce of the latter
  country will continue to have an unfair advantage. The time no doubt
  will come when we shall be able to go up and buy the raw leaf on
  its native hills, pack it by our own methods, and bring it down by
  railway to Shanghai for shipment; but _for years yet we labour under
  the disadvantage of having to buy it just as the Chinamen choose to
  prepare it, without any real knowledge of the total crop at any time,
  or any immediate power to manipulate the Teas to suit the tastes of
  consumers_.

Mark you, this is an enemy’s opinion. May his prognostications be
accomplished to the letter!

The following is from the _Tea Gazette_ lately received:--


  THE CALCUTTA TEA SYNDICATE.

  We are glad to learn that this most useful body intends to continue
  its operations in opening up, wherever possible, new markets,
  although there will be no more soliciting supplies of Tea for
  Australia--the feeling being that the trade in this direction may now
  be left to take care of itself.

  The Tea Syndicate has done a great good, and those able to ship to
  Australia should at once arrange to take the fullest advantage of the
  opening made for them. We would have wished that the Syndicate had
  continued actively its operations there, but perhaps they are right
  in leaving, now, the further development of the trade they have so
  successfully founded to private enterprise. It will be the fault of
  owners themselves if they do not take advantage of the large market
  opened to them.

I conclude my notice of Australia as a market by the following, also
from the _Tea Gazette_. Matters there certainly look promising for the
Indian planter:--


  THE NEW AUSTRALIAN TEA ASSOCIATION.

  Our friends in Australia, now that they are convinced of the purity
  and good quality of our Indian Teas, have determined, we are glad to
  see, to follow in the wake of the Calcutta Tea Syndicate, and push
  by _united_ effort Indian Teas throughout the Australian Colonies.
  Knowing full well that no half-hearted measures would be likely to
  succeed, and that the efforts of a few individuals would not meet
  the requirements of the market, our friends in Victoria and New
  South Wales have _combined_, and formed an association under the
  title of the “Calcutta Tea Association,” for the sale of pure and
  unadulterated Indian Teas to wholesale merchants, storekeepers, and
  customers in general. Large and handsome premises have been taken in
  King Street, Melbourne, and Charlotte Place, Sydney, in which the
  operations of the Association are to be carried on on a large scale.


AMERICA.

The following is from the _Daily News_--a Calcutta paper:--

  We were glad to note that our American cousins were being induced to
  give some orders. If only Indian Tea was once taken up, and became
  popular, its future would be secured. The teeming masses of people
  in the States would consume more Tea we should imagine than all
  the English public, provided Indian Tea took the place of China.
  Australia so far has done well, but the market there would be easily
  glutted, whereas, if its use became general, it would be almost
  impossible to glut the American market. The millions of settlers
  in America and in Canada all use Tea at their meals very much as
  an Englishman takes his beer, so that the inland consumption must
  be very large. In Australia, every shepherd carries his pannikin
  of Tea, and the amount he swallows in twelve months must be pretty
  considerable. In the backwoods of America and Canada, each woodcutter
  consumes nearly half a pound of Tea weekly, so that, with its
  millions of people, America could easily dispose of millions of
  pounds of Tea, which would not only clear off all the surplus Tea in
  the London market, but would probably cause a deficit. We wonder if
  in our time this golden era will take place.

This from the TEA GAZETTE:--


  TEA IN AMERICA.

  A petition has been presented to the United States Congress asking
  for the prohibition of the importation of adulterated Teas from
  China and Japan, which are at present extensively sold. This, it is
  thought, will lead to increased attention being paid to Indian Teas,
  which are well known to be pure and unadulterated.

Again from same paper:--

  The circular lately addressed to the local Tea planting interest by
  the Committee of the Calcutta Syndicate, reporting the results of Mr.
  Sibthorp’s efforts to create a market for Indian Teas in America,
  opens up a vista of unprecedented prosperity in the future.

  That the population of America, the bulk of which consist of the same
  races among whom Indian Tea has grown in favour so rapidly in the
  United Kingdom, should persist in rejecting it after a fair trial was
  _à priori_ highly improbable. It was, therefore, reasonably to be
  presumed that whatever difficulty might beset the opening up of this
  new market would consist chiefly in the obstacles to securing such a
  trial.

  Mr. Sibthorp’s report not only bears out this view of the case, but
  justifies a confident expectation that the obstacles in question,
  so far as they have any real existence, will speedily disappear. In
  Chicago, so far from having had to encounter any of those strong
  trade prejudices which were met with at first in Australia, Mr.
  Sibthorp found the leading importers, Messrs. J. Doane and Co., ready
  to render every assistance and confident of being able to dispose
  of five thousand half chests the first season, without forcing the
  market. Similar success seems to have attended his efforts in New
  York, and a telegram has been received from him ordering a thousand
  half chests for shipment to that port.

  The importance of this new market is immensely enhanced by the
  circumstance that the American consumption of Tea is destined to
  increase, owing to mere growth of population, at a rate not to be
  looked for in any other country; at such a rate, in fact, that if
  India could only secure the annual addition to the demand from this
  cause, she would probably have to double her production in less than
  a generation to enable her to meet it.

  So far from seeing any reason why she should not secure this amount
  of custom in the New World, we see none why the proportion of India
  to other Teas consumed in America should not ultimately be as large
  as in England, where there was once a strong prejudice against Indian
  Tea.

What possible foreign markets have we besides Australia and America?
Russia and many European countries are on the cards, and if the
Calcutta Syndicate will continue its work great results may ensue.
Those who know the Continent often say, and it is true, that no good
Tea can be had in France, Germany, or Italy (it is _not_ so in Russia),
and retail dealers have offered again and again (made the offers to
me) to take large quantities of the Indian Tea of which I have shown
them samples. As this is so, why not supply them? But it cannot be done
well to any extent by individual planters. The Calcutta Syndicate could
easily do it, and I quite believe they would find the work in Europe
easier than in America.

The Amsterdam Exhibition, so soon to take place, affords a great
opening, and from all I hear it will be taken advantage of. Inhabitants
from all countries will be there, and the fame of our Teas should thus
spread throughout Europe. The _Tea Gazette_ says:--


  THE AMSTERDAM EXHIBITION.

  It is intended to have Indian Tea well represented at the forthcoming
  Exhibition at Amsterdam; and we trust that the most will be made of
  the opportunity. There is no reason why we should not succeed in
  Holland as well as we have succeeded in _America_ and _Australia_.
  The rapid strides going on in production must be met by exceptionally
  active exertions to open out new markets, and to see that those
  recently opened out are not allowed to drop for want of fostering.

  The effect of this opportunity will be by no means limited to
  Holland, as in all probability thousands will flock to the Exhibition
  from adjacent countries, and many from all parts of the world.

  We hope that every advantage will be taken of future International
  Exhibitions in any part of the world by an adequate quasi-permanent
  organization in Calcutta, and we sincerely trust that the existing
  Calcutta Tea Syndicate will not cease its most useful operations
  until all the world bows to the great god Indian Tea. The operations
  in countries other than Great Britain during the last few years show
  what important developments in the Tea trade of this country are
  now taking place, and every exertion is necessary to maintain these
  successful results--for which the industry is so much indebted to the
  Syndicate.


THIBET.

This is a large, mountainous, and table land country on the northern
side of the Himalayas. It is at a very high elevation, intensely cold,
and very thinly populated. The Thibetans drink much Tea per head, but
they use Brick Tea; this is made of the coarsest leaves compressed with
some glutinous substance.

There is no difficulty in its manufacture. At present it is supplied by
China, which is close by, but not nearer than India. Many think much
of our coarse Tea (particularly from the Himalayan gardens) might find
a market in Thibet, and I incline to the belief they are right. The
quantity would not be very large, “but every little helps.”

Formerly much Tea was sold to the native tribes over the northern
border by the gardens in Kumaon, Gurwal, and Kangra. Why I know not,
but I hear the trade has fallen off to some extent; the Teas are taken
to the Central Asian markets.

I have done with foreign markets, but there is yet another and a very
large one regarding which nothing has yet been done: I allude to the
market among the natives of India, in other words


THE LOCAL MARKET.

The following is from the _Calcutta Englishman_ on the subject:--

  The letter of our correspondent “A. E. T.” calls attention for the
  hundredth time to the failure of the planting interest to make the
  most of the local demand for Indian Teas. It is only necessary to
  compare the prices realised at the public auctions with those at
  which even the most liberal of our retail firms offer to supply
  their customers with such Teas to see that but a very small fraction
  of the difference between the prime cost of the Tea and what the
  consumer has to pay for it goes into the pocket of the planter. It is
  probably no exaggeration to say that while the consumer pays, on the
  average, from twelve annas to a rupee per pound more than the actual
  cost of the Tea laid down in Calcutta, the planter may think himself
  fortunate if he can appropriate from half an anna to an anna of this
  sum. By whatever course of argument the fact may be justified, it is
  certainly not justifiable by the equity of the case as it appears
  to ordinary minds. For it is the planter who has borne the heat and
  burden of the day, and the proportion which the capital invested by
  him bears to the ultimate return is immensely greater in his case
  than in that of the retail dealer.

  On whom does the blame for the continuance of this state of things,
  if blame there be in the matter, rest? Hardly on the public. They
  would only be too glad to allow the Tea planter, say, four times
  his present profit instead of allowing twelve times that profit to
  a middleman or a series of middlemen. The public, however, can give
  their custom only to those who bid for it, and who consult their
  convenience in the arrangements they make to secure it.

  It is evidently the planter, and the planter alone, who can move in
  the matter. But whether out of regard for the interests of the retail
  dealer, or from a belief that the game is not worth the candle, he
  does not move. If there were a retail Tea trade worthy of the name,
  in the proper sense of the term, in Calcutta, it would probably not
  be to the interest of planters to enter into competition with it.
  But though we have many retail establishments who deal in Tea, its
  sale is, in the great majority of cases, only one item of a very
  multifarious business, the profit on which, as a whole, is probably
  not excessive under all the circumstances of the case.

  As to the game not being worth the candle, that is possibly the case
  if only the present demand is considered. But we are persuaded that
  it is otherwise if regard is had to the expansion of which that
  demand is capable.

  If Indian Tea were procurable in the bazaars in parcels of moderate
  size at a reasonable advance on auction prices, we believe that
  a large native demand for it would rapidly grow up. As it is, an
  extensive business goes on in China Tea of the most wretched quality,
  some of it sold in packets of a few ounces, and some of it loose in
  still smaller quantities. Even in Calcutta this Tea is sold at prices
  which would pay the Indian Tea planter a handsome profit, while in
  the interior it is sold at rates which would have been high fifty
  years ago.

  Surely a Syndicate which extends its efforts for the popularisation
  of Indian Tea to such distant and widely separated markets as
  Australia and America might profitably make some systematic effort to
  promote its use among the vast population at its doors.

  The time may be far distant when the great bulk of this population
  will adopt Tea as an ordinary beverage; but the way in which the
  habit of using it has spread during the last ten or fifteen years,
  among all classes of the vast population of Calcutta, affords an
  indication of possibilities very well worth testing.

When last in India I wrote on this subject largely, but all to no
avail. The following was one of my letters which appeared in the _Tea
Gazette_:--


  THE MARKET AT OUR DOORS.--CONSUMPTION OF CHINA TEA IN INDIA.

  The _Statesman_, in a recent article, observes as follows, while
  discussing the maritime trade of British India:--

  “Perhaps the most anomalous import we have is Tea. It is hardly
  conceivable that while Indian Tea continues to advance in public
  estimation at home, we should not only use China Tea in India, but
  that in increasing quantities.”[93]

  In 1876-77 the imports of China Teas were a little _under_ two
  millions, but in 1880-81 as much _over_ three millions! The
  _Statesman_ states, and truly, that the reason of this is simply
  “that Indian Tea is sold in too large packets to be easily obtainable
  by the general public, for it seems, as regards Indian Tea, the
  smallest quantity that can be bought is one pound, whereas an ounce
  of China Tea can be purchased.”

  Further on, the _Statesman_ kindly alludes to my advocacy in the _Tea
  Gazette_ of a company to sell Tea in small packages to the natives,
  stating also that such a trade is “capable of almost unlimited
  expansion at a fair profit,” which is exactly what I have, for some
  time, been trying to hammer into the heads of those interested in the
  Tea industry of India.

  Now, Sir, is it not absurd that while the _bête noir_ of our industry
  is “supply in excess of demand,” and while, with this dread, we
  are trying (it seems with success) to open up new markets at the
  Antipodes and in America, we are neglecting a market at our very
  doors, the limits of which, I hold, no man can foresee, for is it not
  a market where the possible buyers number 200 millions?

  Is it not also more than absurd, nay a very shame to those interested
  in our industry, that while we have a better article than China Tea,
  we allow, by our supineness and lack of enterprise, more than three
  million pounds of an inferior article to be sold in the birth place
  of the better? And why? simply because we _will_ not supply it in the
  form the teeming crowd of natives willing, nay anxious, to buy can
  avail themselves of it!

  Since I advocated in your paper the formation of a company to sell
  Tea to natives in small packets, and showed, I thought conclusively:
  1--That the capital required was not large (say one and a-half
  lakhs). 2--That the shareholders might expect very fair dividends.
  3--That there was no assignable limit to the trade which might be
  developed. 4--That if such a company was started and worked well, all
  fear for the future of Indian Tea would be at an end. 5--That every
  Tea owner, who became a shareholder, would advance his own interests
  by many times more than the dividends he would receive--since then I
  have obtained from England estimates of all the machinery required to
  bulk and pack the Teas, advice from the best firms as to _the mode_
  so successful in England, and I am more than ever convinced that the
  company would be a money-making one, and that, in two words, we shall
  sadly neglect our own interests if we do not accomplish it.

  Again, since my former articles I have spoken to dozens of Tea
  planters and Tea owners on the subject, and all of them think highly
  of the scheme, while many only wait for the company to be launched
  to take shares. I could name more than one influential native also
  who is willing to join, and this is a good sign, for, in my opinion,
  a moiety of the directors should be natives. I will myself become a
  large shareholder, though I cannot offer my services on the board,
  for it must be in Calcutta, and I do not reside there.

  I am convinced, if the company is launched, the shares will be taken
  up in a week.

  But if no one in Calcutta is public-spirited enough to launch such
  a company, why should not an association of a few individuals try
  to carry out the scheme. I quite believe Tea proprietors would help
  them, at starting, by supplying, on reasonable credit, the coarse
  Teas suitable. Were this done, the thin edge of the wedge would
  be driven in, and, if the association succeeded, they might later
  transfer the business at a fair profit to a company.

  I had written so far when I saw your remarks on the same subject in
  your last issue. I cannot agree with you in thinking an association
  would be _better_ than a company, but I say, failing the last let us
  have the first--in fact, let us make a beginning.

  I give here below, to save the trouble of reference, the last part of
  my former article:--

  “I will now, in conclusion, shortly estimate for how much two and
  four ounce packets could be sold to the consumer.

  “Supposing suitable Teas could be bought at six annas per lb. (and
  all Tea planters know that a very large supply of broken Teas with
  some red leaf would be available at that price), one ounce would
  equal 4-1/2 pie or 9 pie for 2 ounces. We may then calculate thus for
  each 2 ounce packet:

                                                           R. A. P.
  Tea                                                      0  0  9
  Tin foil, company’s mark, labour of making up packet,
    wear and tear, bulking machinery                       0  0  3
  Profit to company                                        0  0  3
                                                           -------
  Price at which company could sell 2 ounce packets        0  1  3
  Profit to dealer or middleman                            0  0  3
  Profit to retailer                                       0  0  3
                                                           -------
  Cost to consumer for 2 ounce packet                      0  1  9

  “As making up a 4 ounce packet would be cheaper in proportion, and
  the profit to company, middleman, and retailer need not be double the
  2 ounce rate, we may fairly say that 4 ounce packets could be sold at
  3 annas.

  “I have sent to England for an estimate of the necessary machinery,
  so that if my project meets with favour, there will later be no delay
  on that score.”

  Surely the above figures, and I believe they are sound, have _the
  look_ of success about them.

  I hear it has been suggested that paper packets would deteriorate by
  keeping, but protected by a good wrapper of tin-foil inside, I feel
  sure this would not be the case.

  EDWARD MONEY.

Nothing has been done to this day; and thus, to our shame be it said,
we are allowing a market capable of indefinite expansion to remain
dormant.

FOOTNOTE:

[93] With few exceptions it is bought by the Natives alone and for the
reason given above.--E. M.



CHAPTER XXXV.

MAKING INDIAN TEA KNOWN IN THE UNITED KINGDOM.


Several plans have at times been proposed in India with a view to make
the merits of pure Indian Teas known in England. When I was last out
there I saw the following letter in the _Calcutta Statesman_, and it
appeared to me the plan suggested was in every way an excellent one:--


  HOW TO PUSH THE SALE OF TEA.

  _To the Editor “Statesman.”_

  Sir,--Referring to your leader of to-day on the subject of selling
  Tea at home, I agree with you that Tea-growers should combine for
  retailing, as they have, through the Syndicate, combined for opening
  up new markets, but there must be the same spirit of enterprise in
  the one case as in the other. Now, the mere opening up of shops for
  the sale of Indian Teas, involving, as it would, rents, expensive
  establishments, and bad debts, would not afford the necessary scope,
  nor would it meet the case.

  The system of auction in Mincing Lane must with all its drawbacks
  continue, but it is surely possible to extract some good from it. Let
  agencies for such a combination as you propose be established in all
  the large towns in Great Britain, and weekly auctions of packets of
  Tea from 2 ounces to 5 lbs. or so be held in different parts of each
  town, so that every day except Sunday there would be an auction going
  on somewhere. Let the sales be _bonâ fide_ to the highest bidder
  and for cash on the nail, and I will promise that before a year is
  over, as high prices will be paid at these auctions as are at present
  realised by Cooper and Cooper, whilst the demand would soon greatly
  exceed the supply.

  If something of the same kind were done in the bazaars of India, the
  taste which so decidedly exists among natives would develop rapidly.

          MATT. DREWS.
  Calcutta, _January 4th, 1882_.

I wrote the following remarks on the above to the same paper:--


  INDIAN TEA SALES AT HOME.

  _To the Editor “Statesman.”_

  Sir,--Your article of Wednesday on the above, and a letter from Mr.
  Drews in Thursday’s paper, have interested me much. As you truly say,
  it is more than absurd that the public at home should pay 150 per
  cent. for our Teas above the prices at which they are sold in Mincing
  Lane, and that this tremendous profit, minus 6_d._ duty, should all
  go into the hands of the retail dealer. Absurd as it is, it is still
  a great fact, and the absurdity can only be increased in one way, and
  that is, if we remain quiet, accept the position, and do nothing.

  That we ought to move, and move quickly, is very certain. How best to
  act requires serious consideration, and ample discussion.

  You advocate a company or association to sell our Teas retail in all
  the large towns in Great Britain, and advocate a subscription of
  Rs. 10 per month by each garden in India, until the business could
  support itself. Nothing _can_ be done unless we all subscribe a
  small sum to set it going, and the amount you mention (Rs. 120 for
  one year; the necessity would most assuredly last no longer) should
  frighten no one, while, if done generally by the Indian gardens, it
  would be ample. I would suggest, therefore, that we should begin the
  matter as set out below.

  The following none of the very many interested in Tea can deny:--

  1. The large profits made on Indian Teas at home are not realised by
  the producers, but by the retail dealers.

  2. We can easily undersell the said retail dealers, to the tune of 50
  per cent. or more, and still work at a large profit.

  3. If the retail dealers were so undersold, an enormous custom would
  ensue to us, or rather the agents we employed.

  4. If Indian Teas were procurable at a fair price all over Great
  Britain, because Indian Tea is superior to China, because those who
  have drunk Indian never revert to China, because thousands would then
  taste our Teas for the first time, and continue their use--I say,
  because of all this, little by little, the consumption would increase
  in a ratio we do not dream of now.

  5. The consumption so increased, we should necessarily, because
  demand exceeded supply, get good prices at the public marts in
  Calcutta and London, and in consequence thereof the value of all Tea
  property in India would be greatly enhanced.

  I believe all the above would _certainly_ follow on a general
  well-combined movement on our part; but let us take the worst view.
  No one can deny that they _might_ do so. Would Rs. 120 be a large
  stake from each garden for even the chance? Let us _begin_ thus:
  Open a list in your office for the names of those gardens willing to
  join. One year’s subscription, at Rs. 10 per month, should be the
  limit from each garden. When enough names are collected to warrant
  further movement, call a meeting in Calcutta, and let the next steps
  be decided on, and in the interval--agitate; I will help to the best
  of my power, and collect opinions from all sides.

  Open the list with the names of the three gardens I represent (as per
  enclosure), equivalent at once to a subscription of Rs. 360.

  Now, as to the question--_how_ to do it? I give you my views, but let
  them be criticised and discussed. We want to do it, and to do it the
  best way.

  What I have been suggesting for months in the _Tea Gazette_, as
  the best thing to do in India--_viz._, to sell Tea by auction in
  convenient forms as to quantity for native consumption--is really
  what I advise for England. I am quite at one with Mr. Drews on this
  point. (I wish you would reprint his valuable letter above, and then
  my allusions to it would be understood.) Retail shops and all they
  would entail, _viz._, intricate supervision, rents, establishments,
  and what not, necessitate details quite outside our legitimate
  sphere as producers. No organisation we could devise would carry on
  successfully two or three hundred shops at home. We (that is, the
  company or the association) _could_ not efficiently superintend such
  a complicated business, and we should be cheated right and left. But
  let others, I say, do the work for us at their own risk, as follows:--

  Sell Teas in whole, half, and quarter chests, in tins of 10, 5, and
  1 lbs., in packets of 8, 4, and 2 ounces once a week (the market
  day) in country towns; daily, in different localities in London,
  Birmingham, Edinburgh, Glasgow, Dublin, and such like cities--all by
  auction to the highest bidder for _cash_, in lots which would suit
  both retail dealers and retail purchasers. Never mind if there be a
  loss at the commencement; the quantities sold, till we felt our way,
  need not be large.

  What would be the result? Retail dealers would shortly sell as much
  Indian as China Tea, _if they could get it_. Our Teas would go into
  thousands of houses where it has never been tasted yet. The demand
  would increase on all sides; prices in Mincing Lane, and consequently
  in Calcutta, would rise, and no fear of a glutted market could then
  exist. In two words, Indian Teas would, I believe, six months after
  such operations were commenced, become the rage in England, and we,
  the owners of Tea property, would add 50 per cent. to the value of
  our estates.

  Is not even the _chance_ of all this worth an outlay of Rs. 120 for
  each garden? I am proud to head the list with my Rs. 360, and I do
  beg of all interested in Tea to follow my lead.

  In the plan I have sketched, like Mr. Drews, all the operations would
  be simple. The necessary supervision would be small: the details
  easily arranged. The Teas would of course be bought in the open
  market in London and distributed for public auction to the different
  localities. There might be some loss at first (it is for this the
  capital is wanted), but if always sold to the highest bidder, there
  would be none--nay, a handsome profit after a time; and though I do
  _not_ think with Mr. Drews, nor should I wish, that the prices would
  eventually equal Cooper and Cooper’s, I do think that the said firm
  would soon find it useless to advertise their _cheapest_ Indian Tea
  at 3 shillings a pound--Tea for which they certainly paid no more
  than 13 pence!

  I may add that I quite agree with the last paragraph of Mr. Drews’
  letter; but a sale for India and a sale for England are two different
  things, and I will not treat of both together.

  EDWARD MONEY.

  Western Dooars, _January 7, 1882_.

Alas! in this case, like the one of supply of Tea to natives, nothing
practical came of it. A _very few_ gardens agreed to subscribe, and the
matter dropped.

Of all the plans that have been mooted, this of Mr. Drews I believe to
be the best. I wish a small company in England would try to initiate
it. No greater boon, in my opinion, could be conferred on the Indian
Tea industry; and were such a Company, with good names, launched in
England, a large proportion of the shares would probably be taken in
India. A very moderate capital would suffice.



CHAPTER XXXVI.

TEA MACHINERY.


So much has it been extended and improved since the Third Edition was
published, I have much to say on this subject.

I will divide it into two headings, “Tea Cultivation” and “Tea
Manufacture.” Of course the machines for the last far outnumber the
first, which are very few, but much of great importance to the industry
will find its place under the first heading--


_Machinery and Implements for Tea Cultivation._

Formerly, with prices as they ruled, Tea paid under most circumstances.
It is _not_ so now. Unless Tea, and good Tea, can be made cheap it is
hopeless to look for profit from a Tea garden. To cultivate cheaply,
and efficiently, is therefore all important (far more important than
has hitherto been recognized), and assuredly the more machinery can
be made to take the place of hand labour, the sooner shall we attain
that end. On this point I need only observe that in most of the Tea
districts in India labour has to be imported at a great cost, varying
from Rs. 50 to Rs. 100 per coolie, and anything which would lessen this
want would materially help to success.

The following, signed “Nil Desperandum,” appeared in the _Tea Gazette_
in August, 1881. I quite agree with the writer and have myself often
expressed the same opinions:--


  PLOUGHING AND HOEING MACHINERY.

  Dear Sir,--On looking over your columns I have been surprised to
  see the small attention paid to agricultural machinery: in fact, I
  can’t find the subject mentioned, although one would imagine it was
  as important if not more so than manufacturing machinery. Various
  agricultural instruments, such as ploughs, &c., have, I know, been
  tried in old times, and not with the best results to the bushes; but
  there is no reason why, because the ordinary machines have failed,
  that planters should be sunk in the belief that that costly article
  the coolie must endure as long as Tea does.[94]

I will now consider the cultivation implements I know of.

       *       *       *       *       *

_Planting Pots._--These are made of clay, cow dung, and cut straw. They
are placed in the nurseries and the Tea seed planted in them. When the
seedlings are big enough to put out, pot and all is buried where the
Tea bush is to be. The pot being broken a little when placed in the
ground, the rain soon destroys it. The seedling _does not know_ it has
been transplanted, and the check of six weeks or more, experienced by
all transplants, is entirely avoided. I know not who invented the pots,
but the idea is an excellent one.

_Jebens Transplanter._--This is an implement for lifting seedlings
without injuring the rootlets or disturbing the soil around them. It
is noticed at page 79 favourably: since that time (1878) it has been
used more or less in all Tea districts. I have seen many opinions both
for and against it. I believe the truth is it works very well in light
soil, and with smallish seedlings, but does not answer in hard soil or
with plants above 2-1/2 feet high. Where the soil and size of seedlings
are suitable, it certainly saves much of the check experienced
otherwise by transplants.

I know of no other peculiar implements for Tea cultivation.

The greatest expense connected with cultivation is, naturally, opening
the soil or digging; the spade is never used in India and would not
answer. Coolies dig with a kodali, a thing something like a spade,
with the handle set at right angles to the blade. Could we dispense
with this, and cultivate between the lines of Tea with ploughs of any
suitable pattern, whether worked by steam or animal power, an enormous
saving would be effected. I am sure the _whole_ space between two lines
of Tea can never be so done, round each and every bush the soil _must_
be opened by hand; but the centre space, say about 2-1/2 to 3 feet,
could, I am convinced, be so worked, and I think it is only a question
of time when it will be so treated.

The planting community are gradually appreciating the fact that
something _may_ be done in this way. The following appeared in the _Tea
Gazette_, end of 1881, _re_ ploughing by steam:--


  PLOUGHING _v._ HOEING.

  Dear Sir,--I am glad to see by the letter of a “Man in the Kundah”
  that some managers have taken up the idea of ploughing instead of
  hoeing. It is an idea which I have been dinning into the ears of Tea
  planters ever since I saw a Tea garden. Mr. Lyell deserves credit,
  and so will everyone who assists to introduce ploughing instead of
  hoeing. The saving of labour would be immense. The gentlemen who are
  interested in the subject will be glad to learn that I wrote home
  last month to several leading agricultural machinery people asking
  the fullest particulars as to _steam ploughing machinery_, with a
  view to seeing how far suitable it would be for Tea cultivation. As
  soon as all my information arrives, and I have thought the matter
  out, I will give the planting community my opinion. I have, as far as
  I am personally concerned, already formed it, and am confident that
  at no very distant date the steam plough will supersede the _dhangar_
  or other hand labour. But of course I must make out a strong case for
  it, or my opinions would be supposed to arise from a professional
  predilection for machinery.

  F.

  Siligoorie, _27th November, 1882_.

Again, “Nil Desperandum,” quoted above, continues:--

  I enclose a report on Darby’s Digger from the _Times_ and _Pioneer_,
  which shows that it is an instrument possessing the principle we
  require in deep hoeing, _viz._, turning the earth completely over,
  and bringing the subsoil to the surface, although of course far too
  unwieldy, costly, and weighty to be used in Tea. It is, however, the
  first step in the right direction, as it closely copies spade action;
  and we may hope that before long a machine with that principle, and
  capable of being worked in a Tea khet, will be brought out. For light
  hoeing, last cold weather I procured from Messrs. Vipan and Headly,
  Church Gate, Leicester, England, two expanding horse hoes, which I
  worked all the hot weather, and which did their work admirably and
  at a much cheaper rate than can be done by hand labour. Two of these
  hoes hoe a 12-acre khet in six days up the lines of Tea and across
  them, but to make a thorough job it is better to go over the work
  again. The total cost of this:--

                     Planted 4′ × 4′
              { Pay of boy and man 12 days         =  3  6  0
  For one hoe { Food of bullocks @ 4 as. per}
              { diem, Barley @ 24 per Rupee }      =  2  0  0
                                                      -------
  Cost of light hoeing 12-acre khet                =  5  6  0
                                                            2
                                                      -------
                   Against                           10 12  0
                                                     ========
  Nirrikh for 136 bildars, light hoeing, 240 spaces,
    4′ × 4′, per diem @ 0-2-9 each                 = 23  6  0
                                                     ========
  Or a saving of more than 100 per cent.

  I gave one 12-acre khet four of these light hoeings during the hot
  weather, which so thoroughly destroyed the grass seeds that, although
  heavy rain has fallen here for the last month and a-half, the grass
  in this khet is thin and not more than 6″ high, a fact which, to
  those who know how the jungle springs up in cultivated ground in
  the Doon when the rains set in, will be a sufficient proof of the
  success of these instruments. The frame of the hoe is only 7″ high,
  and when the blades are buried in the ground is only 4″, and as the
  handle projects from the centre of the back of the hoe and not from
  the sides, there is no danger of the bushes being injured. The hoe
  will expand from 14″ to 20″ at back, and from 3″ to 7″ in front;
  and as the standards of the blades are curved outwards, the hoe in
  its greatest expansion cultivates a breadth of 27″ of ground. I
  found that one bullock was too weak to drag a hoe, although a good
  pony was quite equal to the work, so put in a pair of bullocks. The
  bullocks and hoe take up between them three rows of Tea at once, the
  bullocks on each outside row and the hoe in the centre one. A boy
  walking up the centre row leads the bullocks, which are harnessed to
  the hoe in the same manner as bullocks are harnessed to the country
  ploughs, but with longer julas of course. These hoes are, I find,
  useless during wet weather, as they clog dreadfully, but during hot
  dry weather they are invaluable. What we now want is a machine that,
  either by bullock, horse, or steam power, will do our deep hoeing as
  well as the light hoe does the light hoeing. This is a matter which
  I consider of vital interest to owners and shareholders, as, unless
  in these days of very low prices we can reduce the cost of production
  considerably, we cannot hope that Tea will pay a fair interest on the
  money expended, and great length of time lost in getting up a garden.

  NIL DESPERANDUM.

In the above, two bullocks to drag the plough or digger are evidently
contemplated. My experience is, that two draught cattle cannot be used,
simply because there is not room for them between the lines of Tea.[95]
If animal power is used, it must be a single bullock alone. How to
harness a single bullock to the plough is the question. A collar with a
hinge below, which allows it to open at top, may be put on from below,
and then the sides fastened together at the top. But I advise another
plan, which I have seen most successfully practised in Austria. The
traces, joining together, and thus becoming one behind the bullock,
are fastened to the horns, and tightly connected with a leather pad
across the animal’s forehead. The bullock thus pulls by his head, and
I am sure he can pull in no more efficient or easier way to himself.
Bullocks in pairs, or singly, are thus harnessed for plough work in
Austria, and I have seen single animals dragging ploughs of much
greater weight and power than we should require in our Tea gardens.

Given a proper plough, and I feel sure a large strong bullock thus
harnessed would be successful.

A really good Tea garden plough has yet to be invented. All that is
necessary is to give some agricultural machinists here at home the
conditions necessary for success, and I predict what we want would be
soon forthcoming. I will myself try to do so, let others do the same;
one of us is sure to succeed.

I give all these extracts to show that many think as I do.

Cultivation with ploughs of any kind can never be feasible except on
flat land. The hill gardens in India must in no case hope to introduce
it; but I sincerely trust the planters in India who own level gardens
will not rest till they have solved the problem, and that Messrs.
Kinmond, Jackson, and other inventors of Tea machinery will give their
valuable aid. The following two letters from the _Tea Gazette_ show the
difficulties to be encountered in steam cultivation:--


  STEAM CULTIVATION FOR TEA.

  Sir,--As promised in your last issue, I now write to say that I have
  received from England the catalogues and price lists of Messrs.
  Howard, of Bedford, Messrs. Barford and Perkins, and other makers
  of steam ploughing machinery. Messrs. Howard seem to think that
  the greatest difficulty would be in lifting the return rope over
  the bushes. This would be certainly a difficulty, but the idea of
  steam cultivation for Tea is so valuable that it is well worth
  while thinking this out. I will in your first issue for January
  give a _resumé_ of all the information gleaned from the illustrated
  catalogues and the letters from the engineers at home on the subject
  of steam ploughing, and will then be glad to co-operate with any
  gentlemen interested in Tea by giving my professional opinion and
  assistance without fee in endeavouring to solve this matter. I
  trust, should I ever have to write another series of articles on
  Tea machinery for the _Tea Gazette_, the steam plough may figure as
  one of the machines which I will have to describe as in use on Tea
  gardens.

  Meanwhile the principal difficulty in the way seems to be the
  shifting of the long wire pulling rope over the row of bushes. Let
  those interested in the subject try to devise a speedy and economical
  method of doing this.--Yours, &c.,

  C. B. FERGUS, C.E.

  Siligori, _17th December, 1882_.


  AS TO STEAM-PLOUGHING ON TEA GARDENS.

  Sir,--As promised, I give you a letter regarding the question as
  to whether steam-ploughing could be wholly or partially introduced
  as a substitute for manual labour in Tea gardens. I have been in
  communication with several of the leading makers of steam-ploughing
  machinery in England, but notably with Messrs. Howard, of Bedford,
  and Messrs. Barford and Perkins, of Peterborough. These gentlemen
  forwarded me their illustrated catalogues in duplicate, one set of
  which I sent to you.

  The first question that ensues in regard to the subject is, “Would it
  pay, even if found feasible?”

  In Assam, Cachar, Sylhet and other places, where labour is scarce,
  it is probable that the introduction of steam cultivation would be a
  great boon to the Tea planter. The first cost of a steam-ploughing
  apparatus with ropes, plough, and everything complete as in use in
  England on what is called the “single system,” that is, working
  with only one engine, is about £950. This is heavy, but as a much
  lighter cultivator would be used for Tea, I think the cost might
  be reduced to £800--say Rs. 10,000 on the garden. Under moderately
  favourable circumstances the machinery, making all allowances for
  native attendants, and the usual difficulties we have to encounter
  in India through their laziness and stupidity, should cultivate 800
  to 1,000 acres per month of twenty working days. The remaining ten
  days might be occupied in the rains by taking the engine and gear
  from place to place where it might be required; for, as the expense
  of a steam-ploughing engine and apparatus would be too much for any
  concern, except a very large one, to bear, I suggest that two, three,
  or four gardens unite and purchase one. There need be no clashing
  or quarrelling about terms at the end of the season: each should
  pay his share of the cost of fuel, up-keep, wages, &c., according
  to the number of days it was on each garden. It would thus be to
  the interest of each manager to forward it on to the man whose turn
  was next, without delay. Remember, please, that in saying that it
  would cultivate so many acres in such a time, I mean that it would
  cultivate two ways--that is up and down and across. There would
  remain a little hand-hoeing, &c., round the inner part of the roots
  of the bushes, but not much, as the cultivator I would design would
  go partly underneath the laterals and still not hurt the roots, the
  outer lines being much shorter than the inner ones.

  Now it is a simple matter to calculate, according to the rates of
  the district in which the reader may be, the comparative cost of
  cultivating 1,000 acres of Tea by hand and by the steam-plough. The
  plough would be worked for Tea by an 8-H. P. portable engine of
  any maker’s manufacture. Wages for one engineman, one cooly to cut
  wood, possibly one pair bullocks and cart-driver to bring barrels
  of water, two coolies to shift the anchors, and two more to assist
  them (possibly) in shifting the rope, added to the cost of fuel, and
  15 per cent. per annum added for repairs and deterioration, seems
  to be the cost of working. This would be lessened by the rope and
  anchor-men and the woodcutter on the days when the plough was not at
  work. Add, however, the cost of elephants or bullocks to take the
  engine, &c., from garden to garden, and I think it will be found
  that the saving in expense would be very great on the side of the
  steam-plough as regards cooly labour.

  Now, as to the feasibility of the scheme. It is difficult, without
  the aid of plates, to describe how steam-ploughing is done. The
  engine remains stationary at one corner of the field. Near it is a
  large double windlass, which, when the cultivator is at work, winds
  up the dragging rope with one barrel of the windlass, whilst from
  the other the rope is uncoiling, which will drag the plough down the
  next furrow. When the plough comes to the end of the furrow, two
  men, one at each end of the rope, shift the anchors, on which are
  the pulleys round which the rope runs: one furrow breadth forward
  the plough is double, one set of coulters and shears being at work,
  while the other set is tilted up in the air by the weight of a man
  who sits on and guides the plough. When the plough is to return it is
  not turned round, but the man simply tilts up into the air the set
  of ploughs that have done their work, and brings down the others. Of
  course ploughs like this would not do for Tea: a special cultivator
  would be needed. At the end of the furrow the motion of the windlass
  is reversed, and the drag rope becomes in its turn the following
  rope. In England there is an ingenious mechanical contrivance for
  shifting the anchors, which does away with two men, as it works
  automatically. Now the greatest difficulty in the whole matter, will
  be best explained to the reader in Messrs. Howard’s own language in
  their letter to me. They say:--

  “The obstacle to the use of steam-ploughs through rows of bushes or
  trees is the practical difficulty of bringing the slack or following
  rope into position for following the implement back on its return
  journey. The rope cannot be lifted over the intervening row of
  bushes, and to employ draught animals to take the rope up the next
  alley between the bushes would add to the expense of the work, and
  would impede it.” They continue: “If it is important that the land be
  broken up to a depth of 9 inches, and the obstacles to effecting this
  by animal power are practically insuperable, the steam plough worked
  on the single system, with animals to convey the slack from end to
  end of the land, would probably be the most effectual and economical
  method of working.”

  Now if this difficulty could be overcome (and I confess it is a
  rather formidable one), I quite believe that on fairly straight land,
  even if somewhat sloping, with straight rows of bushes, and the land
  clear of stumps, steam cultivation would be easy. On hill gardens,
  or gardens where the Tea is irregularly planted, on ground much
  traversed by nullahs or having stumps left in, the steam cultivator
  could not work. There may be some method of lifting the rope over
  the bushes. Coolies might be stationed at intervals along the row,
  and with the aid of a very light block and tackle might hoist long
  bights of the rope high enough to clear the bushes. The block and
  tackle would be fastened to the top of a light pole. One man would
  hold the pole while the other hove up, and (the pole being midway
  between the two rows) might incline it over till above the next row
  and then lower away. A strong 10 ft. bamboo, a pair of light wooden
  blocks, and an inch and a half Manilla rope, would be all that would
  be requisite. Other projects for effecting this may strike some of
  your readers, and what I want is, that those who may think the idea
  of steam-ploughing of any value should co-operate together to work it
  out in a practical form: I will give every assistance in my power.

  We can scarcely hope, in the present depressed state of the Tea
  market, that proprietors will club together to subscribe to bring
  out a set of steam cultivating apparatus in order to institute
  experiments on the subject. Should 1883 bring better times,
  something of the sort might be done, and it is as well to have the
  matter well thought out and discussed beforehand, so that should a
  series of experiments take place, people would be prepared for any
  contingencies which might arise, and perhaps be better prepared to
  overcome these difficulties through the matter having been previously
  well discussed.

  It is now the season for opening out Tea gardens, and one piece of
  advice I would give to planters--that is this. It is quite possible
  that steam ploughing for Tea cultivation is a thing of the future,
  or may be nearer than you imagine: therefore be careful to have your
  lines of Tea very straight, both along and across, so that there
  would be no obstacle to the plough or cultivator working. If you
  object to the expense of taking out stumps, they may remain in, as
  they could be taken out afterwards.

  I trust your readers, Mr. Editor, will not view this subject with
  indifference, but will co-operate in endeavouring to solve the
  problem.--I am yours faithfully,

  “TEA MACHINERY.”

Though the signatures differ, I conceive Mr. C. B. Fergus, C.E., wrote
the second as well as the first. He has evidently pondered the matter
well. Let others do so too, and I foretell that the day is not far
distant when _flat_ Tea gardens will, in a great measure, be cultivated
by steam or animal power. When this is so, even 8 annas (say 10_d._)
per lb. for our Tea all round should pay us well.


_Tea Manufacturing Machinery._

The processes in Tea manufacture, as generally practised in India
to-day, are--

  1. Plucking.
  2. Withering.
  3. Sorting green leaf in a measure, and separation of Pekoe Tips.
  4. Rolling.
  5. Fermenting.
  6. Drying or firing.
  7. Sorting.
  8. Final heating before packing.

No. 3 is not always done, the others invariably.

I will consider the machines invented for each process, in the order of
the said processes.

_Plucking._--No machine has ever been invented for this, and I do not
think any is possible.

_Withering._--In any but continued wet weather no artificial means are
necessary. The leaf, spread thinly and exposed to the action of the
air below and around (former attained by any kind of mesh), withers
perfectly.[96] In continued wet weather artificial means are sometimes
required. The various Dryers in use (see further on) are sometimes
supposed to furnish the means, but their use necessitates much labour,
nor is the result satisfactory. A good withering machine (it must be
on a large scale) might, I think, be easily invented; there is none at
present. Why do none of the inventors of other Tea machinery try to
succeed in this?

_Sorting Green Leaf._--This is sometimes attempted in a rough way by
the use of sieves of different meshes. To separate the fine from the
coarse leaf, and in some cases to eliminate the Pekoe tips, is the
object. A machine by John Greig and Co., of Edinburgh, professes to do
the latter. I have never seen it, but I doubt any machine abstracting
the Pekoe tips alone. A machine which would, however, separate the fine
from the coarse leaf previous to rolling is, I think, quite feasible,
and it would conduce much to good Tea. This, again, is an opening for
inventors.

_Rolling._--This is perhaps the most important of all processes in Tea
manufacture. The object of it is to break the cells in the Tea and
liberate the sap (fermentation could not take place otherwise), and
further to give a tight roll or twist to the leaf. Formerly this was
always done by hand (it is so done in China, I believe, to this day),
but the process was lengthy, expensive and dirty. I might perhaps add
inefficient, for doubtless machine-rolled Tea is better done (better in
appearance, better in liquor) than hand-rolled.

I will now consider--


_Tea Rolling Machines._

The inventors are Jackson, Kinmond, Haworth, Lyle, Greig and Thompson.
There may be others, but I have not heard of them.

_Jackson_ has invented five machines. The details of each, how much
each can do, the testimonials regarding them, &c., would fill many
pages. All can be seen in the illustrated catalogue he supplies, so
I will only offer a few general remarks. All planters know Jackson’s
rollers, and they are held in high estimation. His last invention (if
I mistake not) is the Rotary Tea Roller, which is on quite a different
principle to the others. It consists of an elongated revolving barrel
or cylinder, with a polygonal _internal_ surface, and a roller with a
fluted _external_ surface, mounted within the said barrel its whole
length. These revolve in opposite directions (the roller the quicker)
and the leaf is rolled in the annulus between. It is not yet known
what the success of this last invention will be. Not so with his
Cross-action and Excelsior Rollers. These are first-rate machines, and
all who have tried speak well of them.

_Kinmond_ invented the first Tea roller (see page 117), many years ago.
Many improvements resulted, eventually, in his “Improved Double Action
Tea Roller,” which is a very good machine and has given satisfaction to
the many who have used it. From all I have heard and seen, however,
I doubt if, take it all in all, it is equal to Jackson’s Cross Action
Excelsior. Kinmond, some two years ago, invented a “Centrifugal
Roller.” It was made in two sizes. The smaller seems to have done well,
not so the larger; one of the latter on the Phoolbarry garden (in which
I am interested) has proved a failure. But Mr. Kinmond has quite lately
materially altered the said Centrifugal machines, and is confident
that they _will_ do well. He is now leaving for India with one, and
anticipates good results.

Tea machinery is still so much in its infancy that the best machines
are likely to be improved upon, and perhaps superseded by others, but
as things are now, I think, though some do not agree with me, that
Jackson has carried off the palm in rollers.

The following two letters on rollers appeared in the _Tea Gazette_, and
are well worth attention:--


  KINMOND’S IMPROVED PATENT DOUBLE ACTION TEA ROLLING MACHINE.

  Sir,--You have so repeatedly asked planters to supply you with
  information regarding “Tea machinery” that it is a matter of surprise
  to me you have not been flooded with letters on the subject. I know
  very little about Tea machinery, as I am not an engineer, but I
  gladly contribute my quota of knowledge on the subject. I have been
  rolling leaf for some time past in one of Kinmond’s old machines,
  styled his “Improved Patent Double Action Tea Rolling Machine.” A
  machine for _fine_ leaf I do not believe there is in existence.
  I have seen several machines at work on different factories, and
  I should say for _fine_ leaf this machine of Kinmond’s cannot be
  beat.[97] A few improvements could no doubt be made, and I feel sure
  Mr. Kinmond himself is aware of this, and is quite competent to
  make them. I have seen Mr. Kinmond’s “Compound Action Centrifugal”
  at work. I do not consider it a success. It certainly cannot hold
  a candle to his “Patent Double Action.” I would strongly recommend
  Mr. Kinmond to improve the latter, and forego the former, unless he
  can make some very material alterations to it. The roll from the
  “Centrifugal” comes out hot and flat, whereas that from his “Patent
  Double Action” is turned out not only perfectly cool, but has a
  perfect twist.[98] For _coarse_ leaf, Jackson’s “Excelsior” is a
  splendid machine. I should say a factory could not want two better
  machines than one of Kinmond’s “Patent Double Action” and one of
  Jackson’s “Excelsior” Rollers--the former for _fine_, the latter for
  _coarse_ leaf. Will some of my brother planters kindly give their
  experience, and thus further enlighten an anxious

  ENQUIRER.


  TEA ROLLING MACHINERY.

  Dear Sir,--I will be glad if some of your numerous readers will
  kindly furnish results of trials, or of experience, of Kinmond’s
  Compound Action Centrifugal Tea Rolling Machine. I have tried it
  repeatedly, and find it not only heats the green leaf a great deal
  too much, but in addition cuts, I may say into mincemeat, about 5%
  of the leaf in the process of rolling. I am not an engineer, and
  therefore cannot state for certain where the fault lies, but I fancy
  the ribs of the two revolving plates are somewhat at fault. If they
  were broader and bolder, the machine might, perhaps, be a better
  success. The green leaf does not come out sufficiently rolled. The
  major portion of the roll is too flat. Perhaps Mr. Kinmond will
  kindly help by giving a hint or two to a perplexed Tea-house

  “ASSISTANT.”

_Haworth’s Roller._--This machine was invented long ago. The leaf is
placed in bags and so rolled. In some respects the machine resembles
a mangle. It has not been largely used, and thus is not much known.
I have no personal experience of its worth, but have heard much of
it from an old friend of mine, Mr. Carter, of the Chandpore Tea
Estate, Chittagong. He has, I believe, had one from the first on his
plantation, and thinks very well of it. Mr. Carter is a first-rate
judge on all Tea matters. He conducted some experiments to test the
value of Tea rolled by Jackson’s and Haworth’s Rollers, and did it with
great care, that the quality of leaf, the withering, the drying, all
but the two modes of rolling should be _exactly_ the same. The samples
were then sent to Calcutta and valued. Results as below:--

  Messrs. Carritt and Co.’s report on the samples is dated Calcutta,
  29th October, 1881, _viz._:--

  Chandpore leaf rolled by Haworth’s machine:--Large irregular open
  unassorted leaf, brisk, fair flavour, little strong--Re. 0-9-9.

  Chandpore leaf rolled by Jackson’s machine:--Leaf preferable, closer
  rolled, liquor inferior, not very strong--Re. 0-9-0.

  Sungoo leaf rolled by Haworth’s machine:--Rather large irregular
  loosely twisted unassorted leaf, flavoury, little brisk--Re. 0-9-3.

  Sungoo leaf rolled by Jackson’s machine:--Leaf little preferable,
  liquor inferior, wanting briskness--Re. 0-8-9.

By above it appears Haworth’s gave better liquor, and Jackson’s the
best Tea in appearance. From all I have heard I think it likely
Haworth’s roller has not received the attention it deserves.

_Lyle’s Roller._--I have never seen this. From the drawing before me
it has no resemblance to other rollers. The inventor claims for it
simplicity, cheapness, strength, durability, good rolling, and large
outturn with a minimum of labour. One testimonial I have seen speaks
very highly of its capabilities.

_Greig’s Roller._--This I have not seen or heard of. I can only give
the description sent me by the inventor:--

  The Greig Link and Lever Tea Rolling Machine, worked by one man,
  and suitable for rolling the finest nibs without breaking them, or
  to crush the coarsest leaf into broken black at will. It can roll
  a large or small quantity equally well. Price £70, delivered in
  Edinburgh. Small size suitable for cattle gear, £45. Cattle gear,
  £20, delivered in Edinburgh.

  The Calcutta Agent of the Luckea Moung Lung Tea Estate, Sonada,
  Darjeeling, in sending remittance for a large size machine which has
  been working there all the past season, says: “I am informed the
  machine does its work in a most satisfactory manner, rolling better
  than by hand: I am pleased to have to state this.”

_Thompson’s Challenge Roller._--This (quite lately invented) though
given last is likely, by all I hear, to stand well among rollers.
I have no drawing or description of it, but _why_ I think well of
it is that a Tea engineer, Mr. Ansell, of Kurseong, who thoroughly
understands Tea machinery, thinks so highly of the machine that he
has recommended its purchase by the Phoolbarry Tea Company. I have
every faith in Mr. Ansell’s judgment, and feel confident therefore the
machine must be a good one. One feature and advantage claimed for it
is, “free contact of the leaf throughout the roll with the outer air.”

I may conclude my remarks on rollers with a quaint letter (from _Tea
Gazette_) by a native. If he can judge of Tea machinery as well as he
can write English his opinion is worth preserving:--


  TEA ROLLING MACHINERY.

  Dear Sir--On the subject of Tea-leaf rolling machinery, the (to all
  appearance) strangely opposite results I have obtained from machines
  of the same make have led me to the following conclusions, viz.:--

  1. All “genuses” of machines are equally good.

  2. There are hardly two “species” of the same genus which give
  similar results.

  3. Changing the “fixings” of a machine makes all the difference in
  the world.

  _Ergo_ a good mechanic will have a good machine whether he patronize
  Jackson, Kinmond, Haworth, or any other inventor.

  I think with your correspondent “A Voice from Assam” that the machine
  that gives the roll quickly, and in a continuous supply, is the best.

  I would defy any man to prove that any _inventor_ has it “all his own
  way,” for I certainly have not found it so in my experience.

  Yours truly,
      KOL MISTRY.

Before going to press I received drawing and description of “Thompson’s
Challenge Roller.” It is impossible to judge of its merits by the
drawing, but some very strong testimonials are appended--one much in
its favour from Mr. Ansell, the Tea engineer above mentioned. By the
testimonials (more than one from men I know) the following advantages
appear to have been obtained:--

  “Balling” of the leaf is avoided.
  The tips are kept quite bright.
  Heating prevented.
  Simplicity of “feed” and “discharge.”
  One attendant, a minimum of motive power, and low priced.
  A good twist attained.
  Simplicity in the machine, and ease of transport and erection.

_If_ all the above are facts, I quite think the “Challenge” will prove
a great success.

       *       *       *       *       *

The following, written by me to the _Tea Gazette_, may be worth the
attention of Tea-rolling machine inventors:--


  SUGGESTIONS FOR IMPROVING THE DRUMS AND THE FACES OF THE ROLLERS IN
  TEA ROLLING MACHINERY.

  Sir,--The following idea, suggested to me by a planter up here,
  may be practicable or not, but in any case it is worth letting the
  patentees of Tea-rolling machines know it.

  In days gone by when iron worked in contact with iron on the faces of
  rollers the colour of the outturn (that is the infused Tea leaves)
  was quite destroyed. That is now remedied, but there is still an evil
  of less importance. The wood on the said faces of the rollers absorbs
  the sap of the leaf, and unless they are washed very clean, the said
  old sap is apt to contaminate, more or less, the new leaf. Could not
  this be rectified by making the faces of the rollers of porcelain or
  iron (like camp crockery) and the drums of opaque coarse glass? Both
  these, if they would stand, could easily be washed quite clean.

  I give the idea, given to me, for what it is worth, and would invite
  the opinion of other planters on it.

  EDWARD MONEY.

  Darjeeling, _November 10th, 1880_.

_Fermenting_ is the next process in the list. After the leaf is rolled
it is put together; some make it up in truncated balls, some put it
in baskets, but in either case it is allowed to stand until a given
amount of fermentation has set in. This is done in the warm atmosphere
of the factory. Naturally no machine is required for this process;
but shelves, at varying height from the factory floor, are useful to
regulate the fermentation, inasmuch as the higher the shelf the warmer
the air, and warmth hastens the process. This plan of shelves was
devised by Mr. J. Fleming, at the Phoolbarry Garden, and it seemed to
me to answer well.

_Drying_ or _Firing_ comes next. Up to this point the leaf is of a
brownish green colour, and soft. After the drying it is black and
crisp, in fact, made Tea. By the drying process all the moisture in
the mass is driven off. For many years charcoal only was used to fire
Tea, and it was an established belief that _the fumes_ given out by
the said charcoal had some chemical effect on the Tea--in fact, that
good Tea could not be made without it. When, twelve years ago, I
published the First Edition of this Essay, I had begun to doubt the
soundness of the above belief, and four years later I had thoroughly
satisfied myself of its fallacy. It was not, however, till 1877-78 that
I devised _a means_ of firing Teas without charcoal. The invention
was well received, and thought well of. At all events, it _proved_
what I had long urged--viz., that any fuel, if contact with the smoke
was avoided, would dry Tea. My invention was a very crude one, and
quickly superseded by far more perfect designs; still I have the
satisfaction of knowing that on this head I have done much to perfect
Tea manufacture, and that the conviction I had attained to in 1874 is
now general and practised throughout India. (Pages 119-121, 295, to end
of Addenda, bear out the above remarks.)

I will now consider the various


_Tea-Drying Machines_.

_Robertson’s Typhoon._--This is a late invention: it was noticed in the
_Tea Gazette_ in 1881. It had, however, made a great noise at end of
1880, and so well was it spoken of, many, in the early part of 1881,
purchased it. The following was the report as to its merits (_Tea
Gazette_, September, 1881):--


  ROBERTSON’S TYPHOON.

  Mr. J. M. Robertson, manager of the Arcuttipore Tea Company’s
  Gardens, has invented a new Tea-drying apparatus which he has named
  the “Typhoon.” A number of the planters of his district met at his
  garden, by invitation, to test the merits of his machine. We quote
  the verdict recorded by them in their own words, and also append the
  brokers’ report on the Teas which were manufactured in their presence
  during the trial.

  The “Typhoon” is a simple and inexpensive construction of brick
  and iron, which can be erected without skilled labour. The heating
  material used is coke, and the quantity of coke required for a maund
  of Tea is stated to be one quarter of a maund.

  The out-turn from the “Typhoon” we found to be at the rate of one
  half maund of thoroughly dried Tea per hour, and the manner in which
  the work was done was to our entire satisfaction, some of us thinking
  that the apparatus was capable of doing more.

  The inventor leads us to understand that the entire cost of
  construction and material will not be over Rs. 300, and we do not see
  that this sum need be exceeded.

  We are unanimously of opinion that unless the dryers at present in
  use are very materially reduced in price, they will be beaten off the
  field by the “Typhoon.”

  Messrs. William Moran & Co.’s report on the Teas is as follows:--

                          Typhoon Teas.           London      Cal. Equi.
                                                  value.     Ex. 1-8-1/2.
  Pekoe, very well made leaf, with ends, good
    brisk flavour                                   1 10        14
  Orange Pekoe, very well twisted leaf, good
    amount of tip, very good brisk flavour          2  0        15-1/4
  Br. Pekoe, leafy black Br. Pekoe, some
    ends strong                                     2  1         1
  Pekoe Souchong, well twisted leaf, some ends
    good flavour                                    1  3         9
  Souchong, small good even grey leaf, brisk        1  1         7-1/2

  The above are very desirable Teas as regards leaf and liquor.

  The following are some of the chief features and advantages of this
  machine:

  1. The low cost.

  2. Durability, there being nothing except the trays that can suffer
  from wear and tear.

  3. The small quantity of fuel required--about 1/4 maund of coke for
  kutcha firing 1 maund of Tea.

  4. Ease in stoking, the furnace not requiring attention oftener than
  once every one and a-half to two hours.

  5. Absolute and immediate control over the temperature, which can be
  raised or lowered instantaneously.

  6. No “getting up heat” required. In fifteen minutes after beginning
  to light the fire the apparatus is ready for work.

  7. Requires no troublesome cleaning out.

  8. Quantity. The apparatus is capable of drying at least 40 lbs. an
  hour, and has frequently dried over 50 lbs.

  9. Quality of Tea is equal to that obtained by any process hitherto
  introduced.

Of course all the above was very favourable, and its low price gained
it many purchasers. I think, as a _first_ success, it beat any machine
yet invented. But, alas! its fall was sudden as its rise, for, judging
from several letters in the _Tea Gazette_, the purchasers were not
satisfied with its capabilities, and I doubt consequently if it is now
manufactured; still I may be wrong.

_Allen’s Tea Drying Apparatus._--I have never seen this, and have not
heard much about it. Advantages claimed for it are--1. Quick drying.
2. Coke can be used as a drying agent, 10 seers to one maund of Tea.
3. Only manual labour required. 4. Not necessary to turn the Tea. 5.
Perfect control over temperature. I have three testimonials to its
merits before me, one from an engineer, and all three speak highly of
it.

The following letter from the inventor to the _Tea Gazette_ gives
further information:--


  ALLEN’S PATENT DRYING MACHINE.

  Dear Sir,--Some time back your valuable paper contained a description
  and rough drawing of my Patent Drying Machine. I now beg to say that
  the machine is in the market.

  I will simply state here that it can dry one maund of Tea per hour,
  or about equivalent to four maunds of leaf.

  It cannot burn the Tea as in other machines, yet it thoroughly dries
  it at one fill of the machine.

  It takes half a maund of Tea at each fill, and every leaf of this is
  done in exactly the same time; no turning over, changing of trays, or
  further looking after the Tea, after the roll has been placed in the
  machine on the trays.

  Temperature can be lowered from 300° to 100° in two or three seconds,
  and run up again in five to seven minutes.

  It will burn any fuel. Fireplace 2-1/2′ × 3′, when kept regularly
  three quarters full of firewood or coal about 6 to 8 inches thick,
  while machine is drying, will suffice (half a maund of fuel to a
  maund of Tea should be ample). The appearance and fine flavour of
  Tea dried in this machine by fan beats charcoal; no gloss is lost on
  the Tea from shaking up and turning over, and the Tea is black, with
  glossy appearance and good flavour.

  The following are valuation and reports on this machine’s dried Tea,
  by Messrs. William Moran and Co., to whom some of bulk or rough Tea
  was sent.--Yours, &c.,

  J. C. ALLEN.

I omitted to extract the broker’s reports, but they were favourable.
I think it likely this Dryer is well suited to small gardens, which
cannot afford steam motive power.

_Davidson’s Sirocco._--Many of these, over 200, have been set up in all
the Tea districts; it has done good work in its time: had it not done
so it would not so long (some years) have commanded attention. When it
came out it was, I think, the best machine going. I doubt much that
being the case now. It requires no motive power, and is thus, in that
respect, cheap to work. The following letter to the _Tea Gazette_ in
many respects embodies my views of the machine:--


  THE SIROCCO.

  Dear Sir,--As both sides of a question, _viz._ for and against,
  should be stated before the public for their judgment, I think I
  may say that, as far as we have seen _in print_, the “Sirocco” is a
  “first rate Tea-drying machine.” I beg to state that _all_ does not
  appear in print, though what does appear there may be quite true,
  and quite right too for the seller to get as many sales of it as he
  can, for who would be such an ass as to cry down his own invention
  or anything else he wished to profit by. The “Sirocco,” as I have
  seen (and I have seen over ten, and amongst them the latest improved
  ones), does not thoroughly fire off the Tea without burning it: the
  Tea must be taken out of the machine when three parts fired, and
  allowed to cool, when its own heat, and the fact of it being gathered
  in one place, give sufficient heat to finish the kutcha firing, but
  pucka batti is required after that. Again, the advertisement would
  lead one to suppose that the drying is effected by means of the
  draught of hot air entirely: now if this were the case, when the
  fires are first lighted in the machine, the hot air would at once
  be of sufficient heat to dry Tea; but this is not the case, for the
  whole iron work, in fact, the whole apparatus, iron work, &c., has
  to be heated up by fire, and when a little off red hot, the Tea is
  put in and fired. I do not mean to say that hot air does not ascend
  through the Tea, but I contend that the heat of the iron has more
  to do with the drying; there is no detriment to the Tea, I feel
  convinced, whether it is dried by hot iron or hot air, but there is a
  very considerable detriment to the machine. Let purchasers ask any
  engineer, or even blacksmith, how quickly iron burns away, and he can
  tell them.

  Up to date no doubt the “Sirocco” has seen its run: over 200 are
  advertised as in use, but it is now beaten by two machines which
  have come out lately, and which beat the “Sirocco” entirely as to
  quantity dried and simplicity of working, and for durability should
  last any time by careful looking after. One is Robertson’s, which is
  firebrick, and the other Allen’s; both these machines for durability
  cannot be surpassed: the difference in results between the two is,
  that one dries every tray of Tea in the same time without turning
  over, and the other requires to have the Tea turned over and the
  trays changed, &c., as in the “Sirocco.”

  The “Sirocco,” no doubt, was a good Tea-drying machine in its time,
  and the inventor deserves the greatest credit for it, but it has been
  improved upon, as is always inevitably the case in machinery.

  I trust no offence will be taken by the “Sirocco” inventor, as such
  is not intended. Any answer of his will be gladly read.

  Cachar.       Yours faithfully,       PUCKA TEA.

There may have been an answer, but I did not see it.

_Gibbs and Barry’s Tea Dryer._--This machine has been lately invented.
I saw it when not as complete as it is now. I have tried to get
details, but failed. It must have merit, however, for though a late
machine, some thirty-six are now in operation; I heard one good judge
speak very well of it. More are, I hear, being despatched to India. No
trays are used with this Dryer.

_Shand’s Dryer._--This hails from Ceylon. Steam for drying Tea is not
quite a new idea. I saw an apparatus to use steam in Cachar years ago.
The great advantage claimed for this Dryer is that Tea cannot be burnt.
It is quite a new invention. This, from the _Tea Gazette_, describes
it:--


  A NEW TEA DRYER.

  A gentleman in Colombo, Ceylon, a Mr. C. Shand, as we mentioned in
  our last, has invented a new patent Tea Dryer. The following is a
  description of his invention:--

  The barbacue-shaped steam-heated Tea Dryer is the cheapest, most
  economical and safest drying machine.

  As this machine can be made any length and width, the quantity of
  leaf which can be manufactured is only limited by the extent of
  drying surface. One, 5 feet wide, and 15 feet long, will admit of
  about forty pounds of Tea being spread as thinly as on Sirocco trays,
  and, if heated to 150° Fahrenheit, would dry a maund per hour. The
  steam for heating thin galvanized iron drying surface is generated in
  the space (3 inches) between it and the thin boiler plate bottom.

  The machine, which is made steam-tight, is partially filled with
  water, and placed on a fire stove. It is evident that a comparatively
  small quantity of fuel will generate sufficient steam to heat a large
  surface, especially if the smoke flue is placed under the whole
  length of the machine.

  As it is impossible to fire-burn the Tea, dried by the steam-heated
  Dryer, the enormous advantage of being independent of the care and
  judgment of coolies, and of the necessity of uninterrupted European
  supervision, is too evident to require comment.

  Then comes the figure of the Dryer with the following note:--

  “Barbacue-shaped Tea Dryer.--The far end should be slightly higher
  than that over the fire, to allow the space over it to be full of
  water.

  An apparatus for escape of steam and supplying water is inserted in
  the end plate covering the boiler.”

  The _Ceylon Observer_, referring to the above, asks the following
  questions:--

  Is it really impossible by means of steam to over-heat, though we may
  not, indeed cannot, “fire-burn” Tea? And when a boiler is employed
  to generate steam, do we become quite independent of the care and
  judgment of coolies, and avoid the necessity of uninterrupted
  European supervision? Will not a thermometer be necessary to indicate
  the proper degree of heat, will it not require close watching, and
  will there not be danger of the boiler exploding if neglected? The
  danger may be reduced to a minimum, but we should be glad of proof
  that it cannot exist.

  Mr. Shand in reply writes--With reference to your remarks and queries
  regarding my Tea-drying machine, will you allow me to mention that,
  as it is not intended to sustain any pressure of steam, the drying
  surface cannot easily be heated over 150 degrees.

  As a matter of course, the Tea takes a longer time to dry than when
  made by Siroccos, in which the temperature is maintained at 275
  degrees, but the extent of drying surface available makes this a
  matter of secondary importance.

  I did not mean that no care or attention is required to keep up
  fire and supply boiling water periodically from a cistern placed
  over the flue; but you can understand that the same care, judgment
  and observation is not required to dry Tea at a comparatively low
  temperature as at a very high one: for instance, it does not injure
  coffee to allow it to remain on the barbacue after it is thoroughly
  dry; but put it in a roaster, and what care and judgment is not
  required to perfect the roasting!

  No doubt, by the use of Siroccos and other modern appliances,
  the risk of fire-burning is now greatly diminished, but these
  still require great care in shifting the trays and watching the
  thermometer. This constant watching is obviated by the use of my
  machine, and all the superintendent has to do is to feel when the Tea
  becomes crisp and dry. He has the security that, if this is neglected
  to be done at the moment it is sufficiently dry, no injury takes
  place by its remaining on the heated surface.

  The machine is especially adapted for redrying Tea before packing,
  this being an operation carried on at a low temperature, and
  requiring a good deal of care.

  There are, it is well known, two difficulties connected with the
  proper manufacture of Tea, requiring at present the constant
  supervision of the superintendent: these are fermentation and
  firing. If the necessity of closely watching the latter can be
  dispensed with, it gives the superintendents more time to direct
  the fermentation, on which the colour of the infused leaf, and
  consequently the value, so greatly depends.

I have now considered all the Dryers I know of except Kinmond’s and
Jackson’s. I have purposely left these to the last. While in the case
of Rollers I thought Jackson had done best, in Dryers I most decidedly
award the first place to Kinmond.

_Jackson’s Dryer._--A long and exhaustive report upon it from Mr.
Carter, of the Chandpore Garden, Chittagong, appears in the _Tea
Gazette_, November 7, 1881. It is too long to insert here. No one can
read it and doubt that the trials were most carefully conducted, and
without bias of any kind. The results are not in favour of the machine.
Moreover, were Jackson’s Dryers a real success I should have been aware
of the fact long ago. I incline to the belief Mr. Jackson thinks he
can do better, for he has lately brought out a Self-acting Tea Dryer
regarding which the following appeared in the _Tea Gazette_:--


  JACKSON’S NEW SELF-ACTING TEA DRYER.

  Messrs. W. and J. Jackson have invented a new apparatus that will
  deal with the Tea itself throughout the drying process, and thus,
  they submit, secure a perfection in the dessication of the leaf not
  hitherto obtained. The objects arrived at by the new invention are as
  follows:--

  1.--After the leaf is fed into the machine it requires no more
  attention until it is discharged dry.

  2.--Every individual leaf is simultaneously exposed in precisely a
  similar manner to the action of the heated air, thus producing an
  unvaried and perfectly even dried leaf.

  3.--The Tea is steadily but very slowly kept in motion, thereby
  dispensing with the tedious and tiring watchfulness of attendants,
  hitherto required in Tea drying on the tray system.

  4.--There are no trays about the machine to handle, and it is,
  therefore, thoroughly durable and cannot get out of order.

  In operating with the machine, a boy or attendant has simply to
  spread the leaf on a slowly moving feeding web or band, which carries
  it forward and places it in the machine, where it is steadily but
  inactively kept in motion, and in due course is discharged dry and
  crisp from a shoot at the delivery end; so long, therefore, as
  the attendant continues to supply the machine with leaf, it will
  steadily dry and discharge it, and should he have occasion to leave
  the machine at any time, no injury can take place to the leaf in the
  apparatus, as it must pass on and be discharged.

  The leaf is continuously, but very slowly, turned over, disentangled
  and individually presented to the action of the heated air by a
  peculiar combination of concentric cylinders, thus ensuring not only
  the most uniform fermentation, but the drying of each leaf being
  simultaneously effected alike must produce an unvaried briskness and
  quality of liquor not obtainable from any of the methods of drying at
  present known.

  The machine will dry about forty maunds of green leaf per day, and
  will be approximately 9′ long, 3-1/2′ wide, and 8′ high.

  The apparatus will take very little driving, which can either be
  effected by steam or hand power. It is very simple, easily erected,
  and self-contained.

I know nothing about this new Dryer beyond what is printed above, and I
rather doubt if any have yet been set up. _If_ the advantages detailed
are truly all realised, they are doubtless of much value.

_Kinmond’s Dryer._--I shall devote extra space to this, for I _believe_
in it. I have seen it working for a long time on the Phoolbarry Garden,
and I continue since I left India to receive good reports of it. This
is what the inventor himself says of it recently:--

  This Tea-drying machine continues to give great satisfaction. The
  improvements made last year considerably increased the out-turn of
  Tea, and reduced the amount of fuel required. Further improvements
  have this year been introduced in fastening the iron plates at the
  corners of the trays with copper rivets, and otherwise strengthening
  the trays, remedying many small defects suggested by planters who
  are using the Dryers, and in improving the arrangement of the
  fire-bricks over the furnace. The latter, as well as some of the
  smaller alterations, were suggested by Mr. Ansell (inventor of the
  sifting machine which bears his name), an engineer who has had great
  experience in and around Darjeeling in erecting and working all the
  three sizes of these Dryers.

  This is the only Tea-drying machine which can keep pace with the
  largest rolling machines. It is made in three sizes. The capacity of
  the smallest or No. 1 Dryer is one maund of pucka Tea per hour. The
  capacity of No. 2 Dryer is two maunds per hour, and that of No. 3
  Dryer is three maunds per hour. The consumption of fuel is less than
  one maund of wood fuel to one maund of pucka Tea dried.

  One of the great advantages of this Tea Dryer is the facility it
  gives for _final firing_ before packing. The enhanced price of Tea
  which has been dried and final fired in this Dryer is well shewn in
  the high average of 1_s._ 6_d._ per lb., which the Scottish Assam
  Company’s Teas have fetched this season. See letters annexed from
  their superintendent in Assam, Mr. Cruickshanks, and their secretary
  in Edinburgh, Mr. Moffat.

  When _final_ firing Tea with the Dryer, it is found convenient to
  place a fine gauze cover over the top trays in each compartment, to
  prevent any of the Tea dust being carried away with the hot air which
  passes through the Tea.

  In order to get the maximum quantity of work from the Dryer, the
  trays must be spread with rolled leaf twice as thick as that used
  when Tea is dried over charcoal, where there is no forced current of
  air, and after the Tea has been _half-dried_, then the Tea on _two_
  trays should be spread on _one_ tray, and the drying finished. In
  the Dryers now in course of construction, the trays have been made
  one-half deeper, so that the half-dried Tea on _three_ trays should
  be finished in _one_ tray. The out-turn of the machine is greatly
  diminished when the foregoing method is not observed; and owing to
  its non-observance, many of the Dryers in use have never been worked
  to their greatest capacity.

  The Dryer should be lined outside with one thickness of bricks--they
  are the cheapest and best non-conductors of heat--inferior or
  badly-burned bricks may be used. Both ends of the Dryer should be
  lined, and both sides and elbows as high as the trays. The top may
  either have a lining of bricks, or four inches thick of sand or
  clay. When the Dryer is lined round with bricks, it not only greatly
  reduces the consumption of fuel, but by preventing the radiation of
  heat, it enables the men to increase the out-turn of pucka Tea.

  The Dryer is extremely simple and compact--the No. 2 size occupies
  a space of about 7 feet long and 3 feet wide. The fan of this Dryer
  requires about half a horse-power to drive it.

  The fan should be driven at a speed of 500 revolutions per minute.
  The pulley on the fan spindle is 7-1/2 inches diameter and 4 inches
  wide.

  Owing to the satisfaction given by these Dryers this season, an
  exceptionally large number of orders are on hand, and although a
  number of each size is generally kept in stock, the patentee will be
  obliged to those requiring Dryers for next season to kindly send in
  their orders early.

  No. 1 Dryer, capable of drying one maund of pucka Tea per hour, £150;
  No. 2 Dryer, capable of drying two maunds of pucka Tea per hour,
  £220; No. 3 Dryer, capable of drying three maunds of pucka Tea per
  hour, £300. These prices are f.o.b. in London.

  London Agents--Messrs. Geo. Williamson and Co., 7, East India Avenue;
  Calcutta Agents--Messrs. Williamson, Magor and Co., 4, Mangoe Lane.

The best of the three sizes is No. 3. I have quite lately sent out two
of them, one for the Phoolbarry, one for the Leesh Company’s Gardens,
both in the Western Dooars. I think the prices are much too high, and
might with advantage (to both inventor and planters) be reduced; but
as to the excellence of the machine there can, I think, be no doubt.
My opinion is shared by many. I have before me many testimonials as to
its excellence. Space forbids me inserting them here, but Mr. Kinmond
or his agents will send them on application.

In March, 1881, so satisfied was I even then with the Dryer (both the
manager, Mr. Pillans, at Phoolbarry, and I am still more so _now_), I
wrote the following to the _Tea Gazette_, and I give it here as details
are embodied:--


  KINMOND’S TEA DRYER.

  To all interested in Tea in India, and their name is legion, Tea
  manufacturing machinery and its capabilities must be a subject of
  great interest.

  Though Tea prices may, and I think to a certain extent will, revive,
  the old scale which existed previous to the late serious fall
  will never probably return. How serious the fall has been will be
  appreciated when I state that gardens which previously realised 14
  annas to 1 rupee for their produce think now they do well if they
  obtain an average of 10 annas. Thus, an average of 12 annas (even
  if the partial rise I hope for takes place) will probably be more
  than most Indian plantations will get in the future. In two words,
  the Tea industry of India is passing through a period of depression
  and a crisis which argues “the survival of the fittest.” Not only
  must plantations, destined to last, produce largely, they must
  also make good Teas at a small cost. This latter, I hold, both as
  regards quality and economy, _can_ only be attained by the use of
  machinery; and thus, what is the best kind of rolling machine, the
  best description of dryer, equaliser, and sifting apparatus, is an
  all-important point.

  Tea machinery is still quite in its infancy. Various as are the
  machines in use, and superior as some are to others, perhaps none of
  them are yet quite perfect. Still, planters cannot afford to wait
  for ultimate perfection, for though any machines bought to-day will
  probably be more or less out of date in a few years’ time, he who
  waits _must_ go to the wall in the meanwhile. Realising this fact, as
  those who know the subject do, they (and they are many) ask eagerly:--

  “Which of the several machines for the different processes in Tea
  manufacture shall we buy?”

  I have not now, perhaps, the knowledge to discuss fairly the several
  merits of the various machines for each different process, but as Tea
  Dryers hold an important place in the list, and I have, perhaps, an
  exceptional experience of one kind, I purpose to give your readers
  the benefit thereof.

  Years ago, when I first mooted the idea that Tea could be fired
  without charcoal, it was scouted. It was said, “The fumes of charcoal
  had some chemical and necessary effect.” “The Chinese would not
  have used it from time immemorial had a substitute, and a cheaper
  one, been practicable.” Such were the objections. It is now no
  longer a question. A great part, perhaps the greater part, of the
  Indian produce is to-day worked with other fuel, and it is only
  a question of time when _all_ of it will be so. It is generally
  admitted that Tea prepared in Dryers is more valuable than that fired
  over charcoal; and begging the question that the fumes of charcoal
  are _not_ necessary (the old idea is very nearly exploded), it is
  reasonable that it should be so; for, if there is one thing certain
  in Tea manufacture, it is that speed is necessary. Charcoal drying
  took on an average 45 minutes; Tea is fired in the best Dryers in
  eight minutes. In respect of speed, Kinmond’s Dryer (which is the one
  I advocate) is certainly unequalled.

  When, as in large factories, 30 or 40 maunds of Tea have to be made
  daily, it is evident that, _cæteris paribus_, the machine which will
  do most in a given time and given space must be the best. In these
  respects also Kinmond’s Dryer stands well, for the small size (No.
  1) will do one maund, and the larger size (No. 2) will turn out two
  maunds per hour. In other words, in a working day of 12 hours (and I
  allow no more, for I do not believe in night work) 12 and 24 maunds
  daily are the capacities of the two sizes. Considering that the said
  two sizes, with necessary stokehole, tables, &c., occupy respectively
  not more than 200 and 260 square feet of space in a factory, the
  satisfactory results, in both the above respects, are unquestionable.

  Tea made at night, both because the colour of it in its different
  stages cannot be well seen (let the light be what it will), and also
  because superintendence cannot then be so close, is never so good as
  day-made Tea. This is _why_ I do not believe in night work; and it is
  also a very important extra reason why machinery (which by its speed
  enables all the necessary Tea to be made by daylight) will prove such
  a great and lasting advantage.

  When Kinmond’s Dryer was first constructed, it was proposed to work
  it at 300 degrees. Later experience has proved 260 degrees is better
  and sufficient; but of course more time is thus taken, and with the
  old sizes one and two maunds per hour could not be turned out at the
  lower temperature. The machines are now made one-fifth larger to
  obviate this.

  The fan is worked at 600 revolutions per minute, and this is found to
  be the best speed.

  Several alterations, and important ones, have been made since the
  first machines were constructed, but I will mention them shortly,
  for they will only be understood by those who know the Dryer--1. The
  trays now take out alternately both sides. 2. The fine Tea or dhole
  trays take out independently. 3. Outside bearings are supplied to
  the fan shaft or spindle: thus the lubricating oil cannot now run
  down into the fan casing. 4. The chimney is moved forward, and thus
  heats a larger amount of air and reduces fuel. After the necessary
  temperature has once been obtained, one maund of wood will fire one
  maund of Tea. This is an outside estimate.

  The great feature in Kinmond’s Dryer is the fact that a _separate_
  blast of hot air is forced through the Tea on each tray. In all
  other Dryers I have heard of, the _same_ hot air passes through each
  tray successively, and moisture is consequently more or less carried
  upwards through each. It is principally in this respect, and in the
  large quantity of work it executes, that I consider the excellence of
  Kinmond’s Dryer to consist.

  It remains only to give shortly the results of a long series of
  experiments with Kinmond’s Dryer. The valuations were made by more
  than one Calcutta broker:--

         _Class._   _Charcoal dried._  _Machine dried._

  Pekoe               Rs. 0 11  0        Rs. 0 14  0
  Broken Pekoe        Rs. 0 10  0        Rs. 1  1  6
  Pekoe                £  0  1  6         £  0  1 10
  Broken Pekoe         £  0  1  5         £  0  2  7

  These were made from the same leaf, at the same time, with every
  care. In one of my gardens, after Kinmond’s Dryer was obtained, the
  Teas averaged upwards of 2 annas per lb. more all round.

  The Dryer can also be used for withering leaf, but in my opinion no
  Tea Dryer is fit for that work, inasmuch as to do a large quantity
  takes far too much time.

  Artificial withering is only necessary when the weather is wet and
  cold, and the machine to do it should do a large quantity _at a
  time_. No Tea Dryer _can_ do this. A machine fitted for that work has
  yet to be invented, unless Baker’s Wet Leaf Dryer, of which I have
  heard good accounts, but have not seen, would answer.

  EDWARD MONEY.

Since the above was written, further improvements and alterations
(suggested by Mr. Ansell, the Tea engineer, and Mr. Pillans, manager at
Phoolbarry) have been carried out. The machine is now very perfect, and
I consider it the best Dryer at present in the market.

Mr. Kinmond has invented quite lately a coke-burning Dryer. He is now
taking this with him to India to try it, and has sent me the following
prospectus of it:--

  The Coke Burning Tea Dryer has been made to meet the want of Tea
  districts where wood fuel is scarce, and coke can be obtained at a
  reasonable price. The upper part of the Coke Burning Tea Dryer is
  exactly the same as the No. 2 Wood Burning Dryer, which is adapted to
  burn any kind of fuel, but its capacity is a little more, being from
  2-1/4 to 2-1/2 maunds pucka Tea per hour. One maund of pucka Tea can
  be dried with the consumption of about 1/4 maund of coke. Besides its
  large capacity for doing work, and its small consumption of coke, the
  Coke Burning Dryer has other advantages. It is nearly one-half less
  in weight than the Wood Burning Dryer, which means one-half saving
  in freight. It requires no foundation or brickwork of any kind; and
  taking into consideration the quantity of work it does, it is the
  cheapest Dryer in the market--costing only £180, f.o.b. in England.

I know nothing of this Coke Dryer. Its price compares favourably with
his other Dryers.

In April, 1881, the following leader, written by me, appeared in the
Calcutta _Statesman_. Though other Tea matters are included (all
of interest), I give it here as further testimony to the merits of
Kinmond’s Dryer:--

  The days are passed when Tea planters hoped to make a fortune in
  a few years. There are mainly two reasons for this. Firstly, the
  prices of Tea have fallen greatly, in many cases 30 and 40 per cent.
  This is due to the fact that supply, in the case of Indian Tea, has
  overtaken demand. Still, there is some comfort to all interested
  in the industry to be derived from the low prices which have ruled
  during the last two years. So cheap have Indian Teas been that
  the attention of the trade has thereby been directed to them, and
  consequently the deliveries of the last few months have exceeded any
  known previously.[99] It is calculated by those best able to judge,
  that if the present rate of deliveries in London continues, the stock
  in June next will not exceed twelve million pounds, and the truth is,
  strange as it may appear, that below this point it is not well that
  the stock in hand should fall, because, if it does, dealers will not
  be able to meet their requirements, and will then perforce buy more
  China. Low as prices are, we therefore, nevertheless, consider the
  statistical position of Tea to-day as good. There is another point
  which should give comfort and hope to the Indian planter, in spite
  of the fact that we are heavily handicapped in our race with China,
  inasmuch as owing to more expensive labour our cost of production
  _must_ exceed theirs. This source of hope is the great point now
  generally admitted, that Indian Tea is better and goes further than
  China Tea. The experience of each of us can quote instances of
  individuals dropping China Tea, and taking to India; who knows of
  anyone doing the reverse? We admit the taste for Indian Tea is more
  or less an acquired one. Still, the public at home have already been
  educated to the taste by the yearly increasing proportion of Indian
  mixed with China Tea. Speaking generally (though the exceptions
  are many and increase yearly), it is true that Indian Tea is not
  obtainable pure, but no more is China. The bulk of the Tea now sold
  to the public in the United Kingdom is a mixture, three parts China
  and one Indian, and all points to the fact that in a few more years
  the general mixture will be half-and-half.

  We are thus surely paving the way, in other words, teaching the
  English public to like Indian Tea, and the broad fact that, once
  used, it is never abandoned for its rival is surely a very hopeful
  feature. The truth is that were it possible to _make_ the population
  of England, Australia, and America drink Indian Tea for one week
  only, the demand after that week would be enormous, and we should
  hear no more of “supply exceeding demand;” nay, more, many thousands
  of, acres would at once be added to the present cultivation in India.

  But we have somewhat wandered from the question we set out with,
  _viz._, _why_ Tea does not pay now as it once did. The first reason
  we have given; the second is that there is now no market for Tea
  seed. This last reason is little dwelt on, but it is a very important
  factor. The days were when Rs. 300 per maund, and even more, were
  paid for Tea seed, and though this did not last long, the price for
  many years up to 1878 was about Rs. 100. Now it is simply unsaleable.
  The receipts for Tea seed, during all these years, formed a large
  part of mature garden earnings, and, to quote one instance, thereto
  in a great measure were due the big dividends paid by the Assam
  Company.

  But though Tea prices may, and we think will, improve, it is not
  likely we shall ever again see the rates obtainable formerly. This
  being so, it is probable that only those plantations in the future
  will pay that produce Tea cheaply. How is this to be done? Those
  gardens that are heavily weighted by unsuitable climates, by a bad
  class of plant, by slopes which are too steep, by inordinately
  expensive labour, or other causes, will have a hard time of it, but
  plantations with natural advantages need in no way despair. Though,
  as we said above, we cannot, in the matter of cheap labour, vie with
  China, we have a great advantage over the Flowery Land as regards
  economy of production in another respect. We allude to the use of
  machinery, which does much now, and will do more and more as each
  year passes, to reduce the cost of production. Machinery in the
  manufacture of Tea is, we believe, almost unknown in China. There
  each and every operation is performed by hand; here in India many now
  do, and eventually all will, wither, roll, fire, and sort by the
  help of machines. It says not a little for the enterprise and the
  inventive genius of the Anglo-Saxon race that, while in China the
  manufacture of Tea dates back many centuries, and yet all the Tea is
  still made by hand, we in India, who have only planted Tea some forty
  years, have invented machines and use them to-day for each and every
  operation in manufacture. It is but as yesterday that we imported
  Chinamen to teach us the _modus operandi_. We now know far more than
  they do on the subject, and verily the pupil has beaten his master.

  Though machinery reduces the cost of production, and in more than
  one case improves the quality of Tea, and planters know it, the
  difficulty before them to-day is to know which is the best machine
  for each operation. Unanimity on this point is not to be expected
  yet. One swears by Jackson, another by Kinmond, others by Ansell,
  Barry, Lyle, the inventor of the Sirocco, and so on. The machines
  and names of inventors are many, and each has its disciples. Perhaps
  the most favourite rolling machines are Jackson’s and Kinmond’s, but
  we see the latter has just produced what he calls a “Centrifugal
  Rolling Machine” which he thinks will supersede all others. We have
  not seen it, though it is at work on several gardens, and so can give
  no opinion about it; but another of Kinmond’s machines, his Dryer,
  we know well. It was long a moot point if Tea could be efficiently
  fired by any other agent than charcoal. Many affirmed that the fumes
  of charcoal were necessary; and when, years ago, Colonel Money, so
  well known by his writings in Tea matters, affirmed from experiments
  that charcoal was not necessary, but that any fuel would do the
  work, few believed him, for people said it was impossible to credit
  that the Chinese would have gone on using charcoal (so much more
  expensive than other fuel) for centuries, were it not a necessity.
  What Colonel Money then predicted has already come to pass. Much of
  the Tea now produced in India never sees charcoal at all, and it is
  very certain that in two or three years all Indian Tea will be fired
  by machinery. We say this is certain simply because, apart from the
  saving effected by using other fuel, the value of Teas fired by
  machinery is increased. It is natural it should be so because, by the
  use of the best machines invented for that purpose, the heat can be
  regulated to a nicety, an impossibility by the old mode of charcoal
  firing.

  Kinmond’s Dryer is, in our opinion, the best Tea Dryer machine
  yet invented. Space forbids our describing it minutely (besides,
  only those, and they are few, who understand Tea machinery would
  appreciate our description), but its general features we will shortly
  touch on. In the comparatively small space it occupies in a factory,
  and in the large quantity of work it does in a given time, we think
  it unrivalled. This last feature does away with the necessity of
  night-work, which, apart from other drawbacks, is prejudicial to the
  excellence of Tea, because, among other reasons, its colour cannot
  then be appreciated in its several stages. Tea made at night is
  never very good. With sufficient motive power, sufficient rolling
  machinery, and Kinmond’s Dryers, the factory (let the leaf gathered
  be what it may) can be shut up at dark. Kinmond’s Dryer may yet
  be improved upon by himself or by others, but as it now stands it
  possesses a feature peculiar to itself, and all important. The hot
  air, driven by a fan (the speed of which, under control, regulates
  the temperature), does not pass successively through the different
  trays, for the hot air, drying the Tea in each tray, has a separate
  inlet and outlet. By this means is avoided the objection of carrying
  the moisture absorbed by the hot air from one tray to the other.
  Another peculiarity in the machine is, that the same air is used
  again and again, being re-dried and re-heated each time. By this two
  advantages are obtained: (1) fuel is saved, it is easier to heat air
  which still retains caloric than fresh air; (2) the aroma of Tea is
  very volatile, and when hot air, which dries it, passes away, some of
  the essence and strength of the Tea goes with it. But here the same
  air being used again and again, the volatile essence (how much who
  can say?) is returned to the Tea. It is reasonable to suppose that
  this will increase the value of the Tea; indeed, we know it did so
  materially in one garden last season.

  We do not doubt that the unanimity wanting at present amongst
  planters as regards machinery will more or less come with time, but
  only long experience can settle the merits of rival machinery. One
  thing, however, is very certain--if the exports of Indian Tea ever
  vie in quantity with China, it will be due to the use of machinery in
  manufacture.

I may state that Kinmond and some other inventors of Dryers claim for
them that in wet weather green leaf may be withered by their means.
But, as I stated some pages back, I do not think any Dryers suitable
for withering. _That_ machine has yet to be invented.

To conclude my remarks on dryers, I give (again from the _Tea Gazette_)
an estimate of the cost of drying by the old primitive mode with
charcoal, and with machines. There was no signature to the letter. I
cannot say if the figures assumed are quite correct, but in any case
the machines have much the best of it:--


  TEA DRYING MACHINERY _v._ CHARCOAL.

  Dear Sir,--Tea drying by machinery _versus_ Tea drying by charcoal
  fires over choolahs is, I believe, still discussed as to the relative
  merits of each. I will try and give you a fair estimate of cost, and
  speak from experience as far as I know relative to the merits, ills,
  &c., &c., of both modes of firing.

  _1st. Charcoal firing and its merits._--Except for those who persist
  that the fumes of charcoal are necessary to make good Tea, I can
  see no merit whatever in charcoal drying, either in cost, quality,
  rapidity, saving of labour, or anything else, over machine-dried Tea.


  _Cost per maund Tea of Tea dried over choolahs by charcoal._

                                                    R.  A.  P.
  Charcoal at 8 annas per maund, 1-1/2 maunds    =   0  12   0
  1 Battiwallah at annas 4-6, kutcha firing      =   0   4   6
        Do.     pucka firing, say                =   0   0   6
                                                     ---------
       Cost of firing by charcoal                Rs. 1   1   0

  N.B.--Notice the labour staff required for three months in the year
  to make charcoal; the immense space (and heat) taken up by choolahs;
  cost of timber used for charcoal; the number of trays, gauze, iron,
  &c., &c., required; the masonry and carpenter’s work always more or
  less out of repair; loss of small tea falling through trays, &c., &c.

  Now let us take


  _Cost of machine-dried Tea per maund._

                                                     R.  A.  P.
  1st. Those machines which dry by coke, say cost
    of coke                                      =   0   8   0
  3 men at annas 4-6 per 5 maunds Tea    =    about  0   2   8
                                                     ---------
  Cost of drying per maund Tea for a machine,
    drying by coke 5 maunds in 10 hours              0  10   8

  I now give an estimate of cost of 1 maund Tea dried by a machine of
  similar capabilities, but drying with any sort of fuel--coal, wood,
  grass, bamboo, &c., say 2 maunds of firewood at 6 pie per maund = 1
  anna per 1 maund Tea.

  N.B.--Price of firewood at 3 pie per maund should be nearer the mark.

  3 men’s pay, annas 4-6 for 5 maunds in 10 hours = annas 2-8 per
  maund. The analysis of the above comes to this--

                             R. A. P.
  Charcoal drying         =  1  1  0
  Coke      „             =  0 10  8[100]
  Wood fire „             =  0  3  8

  We read of machines drying with any fuel, and doing double the Tea
  of what I have estimated above, and how people can still stick to
  charcoal beats me.--(No signature.)

_Sorting or Sifting_ is the next process--that is to say, dividing the
Tea (by passing it through sieves) into different kinds, as Pekoe,
Broken Pekoe, Pekoe Souchong, and Broken Tea. All do not divide it
thus, for some make other kinds also. In the body of this Essay (page
122) I say, “I do not believe in any present or future machine for
sifting Tea.” I did not then; that was in the early days of Tea; but I
was wrong. A sifting machine, on the large scale on which Tea is now
made, is essential for every garden.

_Jackson’s Sifter._--I have seen this, and heard it well spoken of, but
I have no experience of it.

_Greig’s Sifter._--This I have not seen, but from the drawing I have I
should doubt if it would sift enough per day for a large garden.

_Pridham’s Sifter._--This is quite a new thing. I know nothing of it.

The fact is, the manager at Phoolbarry and I have been so thoroughly
satisfied with the Sifter we use there (Ansell’s) I could conceive
nothing better, and I have not therefore looked into the matter of
Sifters.

In January, 1881, I sent an article to the _Tea Gazette_ describing
_Ansell’s Sifter_, and as I thought then I think now. I believe it is
by far the best Tea Sifter yet invented. Many are the testimonials,
too, in its favour. The price, £80, is too high; but the manufacturers
(Ransomes, Head and Jeffries, of Ipswich) advise me they propose
reducing it to £70. Even that, I think, is too much; but there can be
no question the use of it effects a great saving in a factory.

  This is my article:--


  ANSELL’S SIFTING, SORTING, AND FANNING MACHINE.

  _January 27, 1881._

  In the days gone by, Tea cultivation was, to those commencing a
  Tea career, _the_ thing to study. Those days are passed. None
  are embarking in new gardens, and but few are extending existing
  cultivation. Prices have fallen so wofully that all that Tea planters
  think of to-day is how to make what they have _pay_. _I_ believe in
  Tea still. I think the present low range of prices cannot last, and I
  think so simply because I know Tea will not be cultivated year after
  year at a loss. But the present crisis is very serious: it means,
  in five words, “the survival of the fittest,” and even the fittest
  will not succeed, unless every advantage is taken of all existing Tea
  knowledge.

  Tea manufacture is now the most important branch in the industry. We
  have advanced greatly in the last few years; but Tea manufacture, as
  regards economy in doing it, is yet comparatively in its infancy.
  Still we have done a great deal since the indigenous plant was
  discovered in the jungles of Assam, now nearly fifty years ago; we
  have advanced more in Tea manufacture than the Chinese, who have
  been making Tea many centuries. That is to say, I affirm that the
  Indian Tea planter of ordinary intelligence knows more of both Tea
  cultivation and Tea manufacture to-day than any of his Chinese
  contemporaries. The Chinaman grows Tea, and makes Tea, as he taught
  _us_ to do it twenty to thirty years ago. The pupil in this case
  has certainly beaten his master. We have made some improvements
  in Tea planting and Tea cultivation, but where we have left our
  teachers far behind is in manufacture. “Johnny” makes his Tea as
  his father made it before him, taught by his grandfather who made
  it the same way; and, for aught we know, no improvements, in that
  way, have taken place in the course of many centuries. All is hand
  labour; machinery to them is unknown. The most primitive ideas in Tea
  manufacture are still adhered to. In support of the latter, I will
  quote one instance: Tea, from time immemorial, has always been dried
  by charcoal in China; no other way is known there now. How is it
  here in India? A large proportion of the produce is fired with other
  fuel, aided by machinery; and it is only a question of time (and a
  very short time) when the whole of it will be thus prepared. I could
  quote other instances: let this suffice, for no comparison can be
  drawn between Tea manufacture as followed out in China and India in
  this year 1881. The former is as crude as it was two or five hundred
  years ago: the latter (though still far from perfection) in its many
  details, in its numerous machines cleverly contrived to save labour
  and better the Teas, is a striking illustration of the activity, the
  energy, the inventive genius of the Anglo-Saxon race!

  An Indian Tea factory, well set up with machinery--that is to say
  with a green-leaf drying apparatus, rolling machines, Tea dryers,
  equalisers, and sifting and sorting machines, all driven by an
  engine of 15-horse power--offers a wonderful contrast to a Chinese
  Tea factory, where all is handwork. But more strange still is the
  comparison alongside of the fact, that in the former case the
  industry dates back only some thirty years; in the latter many
  centuries.

  Tea machinery is destined to work great results in India. When
  brought to perfection (it is far on the road now), it will so cheapen
  the cost of manufacture that, though labour is dearer with us than
  in China, we shall, thanks thereto, be able to lay down our Teas at
  cheaper rates than the produce of the Flowery Land. If Indian Tea
  ever vies in quantity with China in the Tea-consuming countries of
  the world, it will be due entirely to the economy effected by our
  machinery. I do not myself anticipate that Indian Teas will ever
  beat China out of the field, but, inasmuch as our Teas are better,
  because the taste for Indian Tea is growing apace, I do believe the
  day will come (it will scarcely be in our time) that the Tea exports
  from India will equal those from China; and, as I said before, to
  machinery, far more than to anything else, will that end be due.

  There is therefore no question of more importance to the Indian
  planter to-day than Tea machinery. It is a difficult question too,
  because so many machines, for each of the different necessary
  processes, are vieing in competition for public favour. “Which is
  the best machine to buy?” is the question one hears asked daily.
  I propose, with your leave, to write a series of articles on Tea
  machinery, pointing out, as far as in me lies, the advantages and
  defects of those which commend themselves most to me, for I wish to
  give planters, through your paper, the advantage of my experience;
  and as my expressing an opinion in no way precludes others from
  doing the same, and I know your columns are open to all, I would
  invite discussion on rival merits, and thus certainly benefit the Tea
  industry.

  I will to-day describe what, I think, is the best Tea sifting and
  fanning machine extant. It is true it is the last machine used in
  manufacture, but that does not signify; I will take all the others in
  turn.

  The said machine is the invention of an able man and engineer, Mr. C.
  W. Ansell, well known in the Darjeeling district for his knowledge of
  Tea machinery. He has been for many years employed as an engineer in
  Tea factories. I heard of his machine when I was lately in England,
  and went down to Ipswich to the manufactory of Messrs. Ransomes,
  Sims, and Head to see it. Though difficult to judge of it, as there
  was no Tea wherewith to test it, I was so pleased with the principle
  that I ordered one. The cost was £80. It has now been working on one
  of my gardens some thirteen months, and in every way it has proved a
  great success. But to describe it, as far as I can, in a few words:--

  Its length is 19 feet, its breadth 5 feet. The Tea, in bulk, is
  delivered through a hopper from an upper floor, on what I will call
  the A end of the machine, to distinguish it from the other end, which
  I will name B. The principle of all other sifters (except Jackson’s),
  as far as I know, is, that the succeeding trays of differing wire
  mesh are arranged one below the other, the slope all being the
  same way, that is--from A to B. This plan is objectionable in the
  following way: if the Tea has been well rolled and clings together, a
  good deal of the fine Teas that are in the mass or bulk often passes
  some distance down, perhaps over half the tray or wire-mesh length,
  before falling through. If they do so, and the object is to sift out
  any particular class on the next succeeding tray, there is only half
  the length of mesh left to traverse to effect the object, instead of
  the whole length of the tray. This is obviated in practice by pushing
  the Teas continually back up the inclined tray; but this is done at
  the expense of extra labour and making the Teas dusty and grey.

  The above objection is obviated in Ansell’s machine. It consists
  of four slopes, but each of these incline downwards, alternately,
  different ways--_viz._, No. 1 (the upper), from A to B; No. 2, from
  B to A; No. 3, from A to B; No. 4, from B to A, and below the mesh
  of each slope is a carrying tin tray, sloping the same way, which
  carries all the Tea which falls through each mesh down to _the head_
  of the succeeding slope, while in each case the Tea which will _not_
  pass through the mesh is delivered separately. The above arrangement,
  however, does _not_ hold with the upper or No. 1 slope. This consists
  of two wire trays or meshes, with the carrying tray below the lower
  one. Such of the bulk as will not pass through the upper tray is
  delivered on the head of No. 2 slope, at the B end of the machine.
  What passes through the upper tray, but will not pass through the
  lower, is delivered by a side shoot at the B end of the machine, and
  is “No. 1 Pekoe.” What passes through both sieves on to the carrying
  tray is also delivered by an opposite side shoot from the B end of
  the machine, and is “Broken Pekoe.” Between Nos. 1 and 2 slopes is
  an air chamber, which, as the bulk left on the upper sieve of No. 1
  slope falls on the head of slope No. 2 (a blast being sent through it
  by a fan at the A end of the machine), drives out of the said falling
  bulk all red leaf, stalks, fannings, &c.

  No. 2 slope receives the bulk at the B end of the machine, after the
  red leaf and fannings are taken out as stated above, and what will
  not pass through the mesh is delivered at the back of the A end of
  the machine, and is “Congou;” while what does fall through the mesh
  into the carrying tray below it (which is still bulk, consisting of
  “Pekoe,” “Pekoe Souchong,” and “Souchong” mixed) is delivered at the
  A end of the machine on to the head of No. 3 slope.

  What will not pass through the mesh of No. 3 slope is delivered at
  the B end of the machine in front, and is “Souchong;” while what does
  pass through the mesh of No. 3 slope on to the carrying tray below
  (still bulk, consisting of “Pekoe” and “Pekoe Souchong”) is delivered
  on to the head of No. 4 slope at the B end of the machine.

  No. 4 slope has no carrying tray: it would be useless. What will not
  pass through the mesh is delivered at the A end of the machine, and
  is “Pekoe Souchong;” while what does pass through the mesh falls
  on the floor of the factory and is the remaining “Pekoe,” that is,
  “Pekoe No. 2.”

  The sorting is so far finished, and the results are the following
  Teas, placed round the machine thus:--“Pekoe No. 1,” at the left
  side of B end; “Broken Pekoe,” at right side of B end; “Red Leaf
  and Fannings,” some distance in front of B end; “Souchong,” also in
  front of B end, but nearer to the machine; “Congou,” at back of A
  end; “Pekoe Souchong,” also at back of A end, but nearer the machine;
  “Pekoe No. 2,” on the floor below the machine.

  With Teas thus minutely sorted, all possible requirements are
  provided for, and the planter can, by mixing or otherwise, make any
  number of classes he may choose.

  It will be observed that “Pekoe” is taken out twice, resulting in
  “Nos. 1 and 2 Pekoe.” These differ slightly, but are better mixed
  together. “Why take them out separately,” some exclaim, “to mix them
  together again?” But there are three very good reasons: firstly,
  the “Pekoe” is taken out at the commencement, previous to fanning,
  to prevent the small or broken Pekoe tips being blown out in that
  process; secondly, the “1st Pekoe” being taken out thus early,
  its appearance is not injured by passing over a large amount of
  sieve-mesh area; and thirdly, _all_ the “Pekoe” is thus extracted,
  which it could not be, as far as I can see, by any other process.

  From all the kinds detailed above, I make only four--viz., “Pekoe,”
  “Broken Pekoe,” “Pekoe Souchong,” and “Broken Tea;” but others can do
  as they will.[101]

  The machine is of course driven by steam.[102] The movement of all
  the trays is a backward and forward one of 3 inches longitudinal
  semi-circular motion, the latter movement being imparted by the steel
  spring hangers. Only a small amount of power is required to drive the
  machine, viz., under half horse.

  I must here conclude my description.

  Now as to the amount of work the machine will do. I speak from actual
  experience when I state what follows:--

  It will sift and fan seven maunds of Tea per hour. The only hand
  labour required to supplement it is a few (a very few) women to pick
  out any foreign substances out of the “Congou.”

  At our garden in Western Dooars, 1,260 maunds of Tea were made in
  1880, and all sifted by this machine, the hand labour besides being
  only 44 women during the whole season, or about one-fifth of a woman
  per day.

  The machine requires only two men to work it continually, and one boy
  to feed it from the upper floor.

  I can think of no possible objection to this machine, or even of any
  possible improvement. I believe, in the case of a 300-acre garden
  with a decent amount of produce, the machine, in its saving of
  hand labour, pays for itself in one year, whilst the Teas are much
  improved in appearance by its use, and fetch higher prices.

  EDWARD MONEY.

I add two more letters in favour of the machine from the same paper:--


  ANSELL’S PATENT TEA SORTING AND WINNOWING MACHINE.

  Sir,--In respond to your call for information regarding Tea
  machinery, I am happy to supply you with my experience of Ansell’s
  Patent Tea Sorting and Winnowing Machine. I have been sifting the
  whole of my Teas, through it this season, and am therefore in a
  position to state what I think of it. I consider it a most useful
  machine, and a great saver of labour. With four men, I do with it in
  one day an amount of work which without it I would have to employ
  from twenty to twenty-five men to accomplish.--Yours, &c.,

  “SIFTER.”


  ANSELL’S SIFTING MACHINE.

  A correspondent writes from London to the _Ceylon Observer_ as
  follows:--Ansell’s Patent Tea Sorter seems to be an article which
  will later be much used in Ceylon. In a memo. before me there is
  an extract from Messrs. George Williamson and Co., who say:--“The
  manager of our Majilighur Garden writes:--‘I have now had sufficient
  experience of Ansell’s Sifter to be able to report very favourably
  upon it. It does its work thoroughly and cleanly, and, owing to the
  comparatively small space it occupies, little or no loss occurs even
  of the finest dust. Sixteen maunds in nine hours is what I find to be
  about its capabilities, and four boys do all the work connected with
  it. It has effected a great saving in the Tea house this year, and
  has quite done away with hand-sieving, except equalizing the broken
  Pekoe and broken Tea--a very trivial operation.’”

_Packing._--This is the final process. Unless Teas are packed directly
they are made, they require to be heated once more to drive off any
moisture imbibed. This can be done in a way in most of the dryers
described, perhaps in Kinmond’s best of all.[103]

This concludes my remarks on Tea machinery; but I shall not have a more
appropriate place than this to mention the ornamental tin boxes devised
by Messrs. Harvey Bros. and Tyler, as a new mode of packing Teas. The
following is an article of mine on the subject to the _Tea Gazette_,
written in 1880:--

  I saw lately tin Tea boxes made to hold 20 lbs., which are
  manufactured by Messrs. Harvey Brothers and Tyler, 21, Mincing Lane.
  I was much pleased with them, for I foresaw that by their use great
  good to the Indian Tea industry would accrue. I went to Mincing Lane,
  and had a long talk with the firm, and came away convinced that the
  fact of the said boxes should be known far and wide in India.

  The boxes measure 15-3/4 by 10 15/16ths by 10 5/16ths. They are
  handsomely illustrated with Indian Tea plantation subjects.[104]
  Each piece runs into a groove in the adjoining one, so that one
  minute will put a box together, and a touch of solder here and there
  completes it; they are then perfectly air-tight. The boxes are very
  sightly. Price is now 2_s._ 5_d._ per box. Boxes sent to Calcutta
  up to this have been charged 2_s._ 7_d._ The price is dependent on
  the fluctuating price of tin, which is somewhat lower now. Of course
  they are sent out in pieces. Cases holding pieces for 100 boxes weigh
  4 cwt. The firm tell me that Messrs. Schœne, Kilburn and Co., and
  Messrs. Begg, Dunlop and Co., in Calcutta, have consignments of the
  boxes, so any of your readers can see them.

  In my opinion there are several advantages, to be derived from their
  use:--

  1. They will help to open up new markets. The ungainly, unwieldly
  packages we have used hitherto are certainly detrimental, at least
  where Indian Teas are not known. By the use of these tin boxes the
  sale of our Teas would, I am sure, be extended at home, and they
  would also give great facilities for successfully introducing Indian
  Tea into Australia, Canada, the United States, the Cape, &c. It seems
  some Indian Tea has already been sent home in these tins, and I am
  told it met with a ready sale, quite to 8_d._ per lb. over what it
  would have brought in chests. This is, of course, too good to last,
  but less than one penny a lb. increase would pay for their use.

  2. The sale of Indian Tea in India would be developed by using them.

  3. The tares of these boxes is and must be exact, viz., 3 lbs. 15-1/2
  oz., so only a few would be opened at the Custom House,[105] and the
  great loss by the deterioration of Tea being exposed (few know _how_
  great it is) would be avoided.

  4. There is no doubt Tea will keep better in transit in these boxes
  than in our old packages. How often are the latter broken and the
  lead torn! This evil would be quite avoided.

  There seems to me to be but one doubtful point. The boxes cannot be
  sent loose on board ship: how then are they to be packed? Chests
  holding four tin boxes were recommended, but they do not smile on me.
  True, they might be made very light: still they would add to the
  size, weight, and cost considerably. I think crates of strong light
  battens would answer perfectly, and six, or perhaps eight boxes might
  then be placed in each. However, this is a matter of detail, which
  experience would quickly decide. To continue the advantages:

  5. Teas packed in these boxes, and so sold, would not be used for
  bolstering up China rubbish. They would be drunk pure, and thus the
  great desideratum of teaching the public, both here and abroad, to
  use Indian Tea by itself, would be, in a measure, attained.

  I do not say that any planter should pack _all_ his Teas in this
  new way. The mass of Indian Tea, do what we may, will still be used
  to mix with China. Again, the highest class of Indian Teas are not
  the ones to commence with. As a rule they are too expensive for the
  public to use them alone. Ordinary Teas, or perhaps a mixture which
  could be sold cheaply, and would be a good household Tea, is what I
  should recommend. It is just this kind which is now such a drug in
  the market, and necessarily the diversion of some of this into other
  channels would help us greatly.

  6. A considerable saving in the loss of Tea at the Custom House would
  result by the use of these boxes, as the following figures will show.
  To begin with, the trade allowance of 1 lb. per package which is now
  allowed the buyer, and which is of course a loss to the producer,
  would be avoided; for this allowance does not apply to any package
  under a gross weight of 28 lbs., and these tins with 20 lb. 2 oz. of
  Tea in them, will weigh gross only 24 lbs. 1-1/2 oz.

  To make the figures below clear, I must state that the rule of the
  Custom House is to discard fractions of a pound both in the gross
  and the tare. But in the gross the number _below_ is written, in
  the tare the number _above_. Thus, if the gross weight of a package
  is 132-1/2 lbs., the gross is written 132. If the tare of a package
  is 37-1/4 lbs., it is written 38. Now to take one extreme case, to
  show the loss on our ordinary Indian packages: a chest weighs gross,
  say, 132 lbs. 15 oz.; it is still written 132 lbs. The tare of the
  said package weighs, say, 37 lbs. 1 oz.: it is written 38. The tare
  deducted from the gross gives the net weight of Tea. In this case 132
  minus 38 equals 94 lbs., which is all the producer is paid for. But
  the net weight of Tea in the box is 132 lbs. 15 ozs., minus 37 lbs.
  1 oz., equals 95 lbs. 14 ozs., and thus on such a package there is a
  loss of exactly 1 lb. 14 ozs. Add to this the trade allowance of one
  pound, and the whole loss is 2 lbs. 14 ozs., which is about 3 per
  cent.

  It will be observed that by this custom the advantage, as regards
  the duty of 6_d._ per lb., is on the side of the payee, but none the
  less is it to the loss of the producer. The case quoted above is,
  of course, an extreme one, but in practice I believe the loss of
  Tea on Indian packages, including the trade allowance, is not much
  under 2 lbs. In the case of our ordinary Indian packages, if we could
  regulate our tares exactly, so as to make the gross weight only one
  ounce above the whole number, and the tare one ounce below the whole
  number, the loss would necessarily be much decreased. This, however,
  is impossible, for, as a rule, the tares are one or two pounds less
  when they arrive in England than when they left the garden, owing to
  the wood drying in transit; and thus it is quite a chance what the
  real tares come out here.

  But, with the tin boxes in question, the tares, that is their weight,
  being fixed and equal, and not liable to change, we can so arrange
  the weights that the loss will be very trifling, thus:--

                                      lbs.  ozs.

  The box weighs                        3   15-1/2
  We put in Tea                        20    2
                                       --------
           Gross Weight                24    1-1/2
                                       --------

  In the Customs the gross is written  24 lbs.
  And the tare is written               4 „
                                       ------
          The Tea paid for will be     20 lbs.

  that is a loss of only 2 ounces, or not much above half per cent.,
  instead of three per cent., as shown in the old packages.

  Shortly, to conclude this point. In the case of the old packages by
  no means can we help ourselves; but, as shown, with the tin boxes,
  the loss need be very little.

  Roughly, the cost of using these tin boxes would be, all told, from
  1-1/2_d._ to 1-3/4_d._ per lb., and with our lead-lined boxes it
  averages perhaps per penny. The difference of a halfpenny, or even
  three farthings, one pound would not be much for the advantages
  detailed.

  One point I have forgotten. If 500 boxes are ordered, the plantation
  mark is put on the ends of the boxes gratis. If less than 500 are
  ordered, the additional cost for this would be about £5.

  I hope the Syndicate in Calcutta will try these boxes. I shall
  certainly do so.

  I enclose the directions for making up the tins, and hope you will
  insert them at the foot of this letter.

  Reading over the above, there is one point I find not observed on
  as regards the loss of Tea at the Custom House. By the mode of
  weighing, as explained, the producer often loses 2 or 3 per cent.,
  but still, strange to say, in practice, this loss is sometimes more
  than counterbalanced by the increased weight of the Tea due to the
  moisture imbibed while exposed (if boxes are broken in the transit)
  anyhow at the Custom House. But I need not point out that this gain
  is dearly bought by the deterioration of the Tea. The Custom House
  procedure is bad in every way. More on this subject later.

  EDWARD MONEY.

The following is also from the _Tea Gazette_, and is much in favour of
the boxes:--


  PACKING OF TEA IN TIN BOXES.

  In our issue of November 7th, 1881, we inserted a short editorial
  note questioning, on the authority of certain correspondents, the
  advisability of using tin Tea boxes for the packing of Tea, at the
  same time asking our readers to favour us with their opinions on
  the subject, in case we were misinformed. Our invitation has met
  with a response from several quarters, and the correspondence we
  have received leads us to alter the opinion we formerly held on
  the subject. A gentleman largely interested in Tea, but in no way
  connected with the manufacturers of the patent tin boxes, writes to
  us from England:--

  “I made enquiries as to the condition in which Tea packed in Messrs.
  Harvey Brothers and Tyler’s lacquered tin boxes is turned out in
  London. I found that the Tea was not at all injured by this method
  of packing, but that its condition is quite as good as that of Tea
  packed in chests. Messrs. W. J. and H. Thompson assured me that you
  were entirely mistaken in your remarks as to the contamination,
  but they thought that an objection to the packing in the lacquered
  tin boxes was the labour of putting up in these boxes. Catalogues
  were shown me in which I saw that the Teas in the lacquered tin
  boxes fetched higher rates than the same Teas packed in chests, the
  difference being in one case 3_d._ per lb.”

  This is certainly a most favourable testimony, and coming as it does
  from a disinterested party, who writes simply in defence of what he
  considers the right, we cannot but accept of his statement in its
  entirety.

  Another correspondent writes:--

  “I now give you a few of the sales of these boxes made at public
  auction during the last month, shewing the preference of the trade
  for Tea so packed, and the higher prices realised.

                _Public Sale 3rd November._  _s._  _d._

  Koliabar.    {28 chests Pekoe               1   10-1/2 per lb.
  K. Assam.    {28 cases, each 4 tin boxes    2      3/4   „

                _Public Sale 16th November._

  M.L.B.D.S.A. {30 chests Pekoe               2      1/4   „
               {30 cases, each 4 tin boxes    2    2-1/2   „

      „        {20 chests Souchong            1    3-1/2   „
               {20 cases, each 4 tin boxes    1    4-3/4   „

                _Public Sale 23rd November._

  M.L.B.L.P.   {20 chests Pekoe               1    6-3/4   „
               {19 cases, each 4 tin boxes    1    9-3/4   „

  “In every case the above Teas were packed out of the same heap
  in India, and the difference in the selling price arises chiefly
  from the _better condition_ of the Tea on arrival, and the growing
  preference of the country trade for Teas so packed.”

The following is worth notice:--


  HOOP IRON.

  The _Ceylon Observer_ says: “The planters should note the following
  (writes to us a London firm)--From quotations lying before us the
  prices of 22 gauge iron hooping are as follows: 1/2in., 165_s._
  per ton; 5/8in., 110_s._ per ton; 3/4in., 70_s._; 7/8in., 60_s._;
  1in., 50_s._ Thus by using one inch hooping, less than one-third the
  price is paid. The narrower the hooping, the more difficult is it to
  manufacture.”

It is also not so strong.

FOOTNOTES:

[94] “Nil Desperandum” evidently foresees what must be sooner or later.
All interested in Tea, owners, planters, and inventors, should aid to
achieve the result.

[95] “Nil Desperandum” takes up three spaces--one bullock in each
outside space and the hoe in the centre. I don’t like the plan. It
_could not_ be done where the tea plants are high.

[96] In wet weather especially the warm air generated in the factory by
the fires in it helps the process.

[97] This, after the previous sentence, is obscure.--E. M.

[98] I agree with “Enquirer” in this.--E. M.

[99] They are still higher now. The last three months they have
averaged five millions.--E. M.

[100] I should be glad to be set right if I have not rightly calculated
the price of coke.--(_Writer of the letter._)

[101] I advise only these four kinds. When the trader becomes more
sensible, three or even two would be better, but as it is now four are
necessary.

[102] With a driving belt from the engine shafting.

[103] Heating before packing has to be done on a large scale. None
of the Dryers notified are large enough. A special machine should be
devised.

[104] Top is “The Tea Garden;” front, “Weighing Leaf;” back, “Packing;”
ends, “Elephant with Howdah,” or, if desired, the plantation mark.

[105] The following are the numbers to be opened by the Custom House
regulations:--

  From    1  to   5              1 to be turned out.
   „      6 „    40              3   „    „
   „     41 „    80              4   „    „
   „     81 „   120              5   „    „
   „    121 „   200              6   „    „
   „    201 „   300              8   „    „
   „    301 „   500             10   „    „
   „    501 „   800             12   „    „
   „    801 and upwards         16   „    „

This applies to packages of all sizes and kinds, if the tares are equal
or nearly so. If the difference in the tares are not great, an average
is struck. If tares are various all are turned out!



CHAPTER XXXVII.

WEIGHING AND BULKING OF INDIAN TEAS AT CUSTOM HOUSE.


One misapprehension with some exists on this head. The _weighing_ is
done by the Customs to ascertain the amount for duty. The _bulking_ is
done at the request of the vendor, the broker who is to sell it, or the
purchaser, and it has to be paid for.

Two distinct injuries are inflicted on the producer by the present
Custom House system--

  1. The Tea is much damaged by exposure.
  2. The quantity found is _always_ less than the actual.

Now as to No. 1. When we consider how damp the London atmosphere is
at the best, how in foggy days it teems with moisture, is it not very
certain that Teas exposed to it, often for days, deteriorate? What
care we take in India heating before packing--carefully with lead
and solder, excluding all air--and then the Teas on arrival here are
treated as above! It is simply monstrous.

The following extract from a letter to _Home and Colonial Mail_ sets
out the case forcibly:--

  The blame ought not entirely to be laid upon the planter, however,
  for certain facts have come to our knowledge during the present week
  as regards the manner in which Indian Teas are bulked at some of
  the London warehouses, which somewhat explains how depreciation in
  quality comes about. We bought several breaks of Tea in the sales
  this week, which were stated to be bulked and ready for sampling six
  days before the sale; and yet we know for a fact that some of those
  very Teas were not put back into the chest till the day after the
  sale, if even then. More or less moisture is always to be found in
  the London atmosphere, particularly in rainy weather, and there can
  be no question that incalculable injury would be done to a fine Tea
  by seven days’ exposure on the floor of a warehouse. The damage and
  loss falls entirely on the buyer. The effects of it are not seen at
  once, but there can be little doubt that a gradual depreciation sets
  in, consequent on the absorption of moisture. No redrying process
  follows; the Tea is simply filled back into the chests when seven
  days of neglect have done what mischief is possible. Is it to be
  wondered at that samples drawn from such a break of Tea a few months
  after it has been bulked in London will have lost all their freshness
  and malty smell?

  J. C. TAYLOR AND COLMAN.

I have no reason to think the delay above is very unusual, and I must
add to the above, that when the chests are closed no attempt is made
even to cover the top with lead, much less to resolder it. Some paper
on top is all attempted. I need say no more to prove that the quality
of Indian Teas is _most_ seriously damaged at the Custom House.

Now as to No. 2. The loss in quantity to producer.

The following article, which I wrote to the _Indian Tea Gazette_ in
1881, shows how invariable the loss must be:--

  The loss of Tea by the mode adopted at the Custom House in England is
  great.

  When Teas are sold at Calcutta, though the English Custom House
  regulations do not then affect us immediately, they do so indirectly.
  If purchasers in Calcutta gain by our Teas, they will bid more; if
  they lose, they will bid less. Besides, many Teas are sold in London.

  To understand what follows, it is necessary to remember that--

  Garden Invoices _never_ go to Custom House. Custom House arrives at
  weight of Tea by weighing the package for “gross,” and then turning
  out Tea, weighing box, lead, nails, iron hooping, in fact all but
  Tea, for “tare;” gross weight, minus tare, is the weight of Tea they
  demand duty on, and the weight so found by Custom House is all the
  producer or importer gets paid for.[106] It follows, therefore, that
  the less Tea declared by Customs means a loss to producer and a gain
  to buyer. To the latter in two ways, _viz._, less Tea to pay for
  than is really there, and a saving of 6d. per lb. duty! But to show,
  now, how the loss occurs. When weighing for gross, the fractions of
  a pound are discarded; when weighing for tares, the pounds above the
  actual weight are written. The _greatest_ loss that can occur by
  this method, on one package, is 1 pound 14 ounces of Tea. It (this
  greatest loss) _must_ always occur when the gross is 1 ounce short of
  a pound, and the tare 1 ounce more than the pound.


  NO. 1 EXAMPLE.

  Gross and tare can be put at any figures as to pounds. It will always
  come out the same. Say, therefore,

                              lbs.   oz.
    Gross                     132    15} actual weights taken at
    Tare (deducted)            37     1}  Custom House.
                              ----------
  Actual Tea in chest          95    14

  By rule quoted the gross and tare weights are set down at Custom
  House--

                                lbs.
    Gross                       132
    Tare (deducted)              38
                               -----
  Actual Tea thus paid for  =    94 pounds--on which duty is also paid.

  Therefore the loss on the chest is 1 pound 14 ounces.

  The _least_ loss that can take place (when ounces occur in gross and
  tare) is 2 ounces. To insure this the gross must be 1 ounce more than
  the pound, and the tare 1 ounce below.


  NO. 2 EXAMPLE.

  Say any figures in pounds.

                                lbs.  oz.
    Gross                       133    1}  actual weights taken at
    Tare (deducted)              36   15}   Custom House.
                                --------
  Actual Tea in chest            96    2
                                ========

  But again, by rule quoted, it is written by Customs--

                                lbs.
    Gross                       133
    Tare (deducted)              37
                              -----
  Actual Tea paid for            96 pounds, on which duty is also paid.
                              =====

  Therefore the loss on chest is 2 ounces only.

  Now did weights turn out the same in London that they were on the
  garden, we could, by doing as in last example, insure only the above
  trifling 2 ounce loss. But it is _not_ so. The wood dries and thus
  makes both the gross and tare less. The loss then comes out anything
  between 2 ounces and 1 pound 14 ounces.

  I find the following simple rule will give the exact loss on each and
  every weight of both gross and tare.

  _Rule._--Add the ounces above a pound in the gross to the ounces
  short of a pound in the tare. The sum of the two, in ounces, will be
  the loss of Tea on the package.

This is only part of the article. I break off here to add a few remarks
more appropriate now than what I then wrote.

There are means by which this varying loss, of which the maximum is
1 pound 14 ounces, can be reduced to 4 ounces only on each and every
chest.

I admit the procedure is scarcely practical, but as nothing can
demonstrate better the absurdity of the system as pursued at the
Customs, I give it here.

How can we insure the _least_ loss, taking into consideration the fact
that the weights of both the gross and tare, because of the wood drying
and lightening in transit, can never come out the same at the Custom
House in London as they were on the garden.

We can do it thus: the Tea if well packed in a chest in no way
alters in weight during transit. If dry, when put up, it cannot
become lighter; if the leaden covering is air-tight, it can absorb
no moisture, which would of course make it heavier. I therefore beg
the question that it is a _fixed_ quantity, for it must be so if well
packed.

We have therefore only to consider the gross and the tare, and, as
shown, the loss in Tea, varying from 2 ounces to 1 pound 14 ounces,
depends entirely on the weights these are found to be at the Custom
House. In other words, if we can insure the gross there being but
little over any even number of pounds, and the tare there being but
little below any other even number of pounds, we attain (approximately)
the least loss we can be mulcted in.

Begging the question that we can add to, or detract from, the gross
weight of each chest in the Custom House (before it is put into the
scales by the officer there) by the addition or subtraction of a few
nails if the weight is nearly what we want, or pieces of hoop iron if
the actual varies much from the desired weight--I say, if we can do
this, we can insure approximately the minimum of loss. I go to show how
this is to be done.

Pack the Tea in the usual way, but whatever the quantity it is desired
to put into the chest (it can be varied with each class, for it matters
not what the weight is in pounds) add to it 4 ounces, and be very
careful that the whole weight of Tea is exactly the number of pounds
required, plus 4 ounces--for the whole success of the plan depends on
this weight being exact. Nothing more is required to be done at the
Factory than has been done hitherto, for it matters not one straw,
as regards the success of the plan, what the gross and tare of each
package is, nor what the weight of Tea is, as long as exactly 4 ounces
above an even number of pounds is there; neither does it signify how
much the wood lightens in transit, and thus decreases the weights which
were found at Factory for gross and tare.

The next step must be taken at the Custom House in London. Let the
importer or the producer’s agent attend and weigh each package himself
nicely, any time before the weights are to be taken by the Customs.
Then let him _make_ each package 2 ounces above the even number of
pounds. This will be easy enough, by the addition or subtraction of
a few nails or hoop iron. For instance, suppose the chest to weigh
140 pounds 6 ounces, he would take away nails or hoop iron weighing
4 ounces. If it weighed 140 pounds 13 ounces, he would, by adding 5
ounces more nails or hoop iron, make it 141 pounds 2 ounces. All would
then be finished, and each and every package so treated would give a
loss in Tea of 4 ounces only.

If my plan could be carried out (as the minimum loss otherwise is 2
ounces, and the maximum 1 pound 14 ounces the mean is one pound),
we save a loss of the said pound on _each_ chest, minus the loss we
compound for, _viz._, 4 ounces. That is to say, we gain 12 ounces on
each package which, in a break of 2 or 3 hundred chests, means a good
deal to the producer or Customs!

I will give one example in figures. Any other possible figures can be
tried: it will always come out the same, _if_ the weight of Tea is
exactly 4 ounces above any given number of pounds.


NO. 3. EXAMPLE.

                                                lbs.  oz.
  Results at Garden. Tea, any number of pounds
       with 4 ounces added (say)                 100   4
         Tare (any figure) (say)                  43   6
                                                 -------
                              Gross at Garden    143  10
                                                 -------

  The wood lightens in transit any amount (it is immaterial),
      say 15 ounces.
                                                lbs.  oz.
  The weights at the Custom House    {Gross      142  11
      then become                    {Tare        42   7
                                                 -------
                       Weight Tea as before      100   4
                                                 -------

  At Custom House (as detailed) by adding 7 ounces
      of nails or hoop iron make                       lbs.  oz.
                                            Gross       143   2
  The tare will thereby necessarily be}
      increased 7 oz. and become      }     Tare         42  14
                                                        -------
                             Weight tea as before       100   4
                                                        -------
                                             lbs.
  These weights are written at     {Gross    143
      Custom House                 {Tare      43
                                            ----
         Weight of Tea found by Customs is   100 pounds

which is a loss of 4 ounces only as stated.

Were the plan feasible, the gain to the Indian planters would be
large. Say this year (1883), fifty-seven million pounds are imported,
and ninety pounds per chest is taken as the average, this gives over
600,000 chests, and 12 ounces saved on each = 450,000 pounds, of Tea,
which at 12 annas per pound, Rs. 3,37,000.

The gain to the Customs would be 450,000 sixpences = £11,250.

This increase to the Customs would be attained by simply (though still
keeping under the actual weight of Tea in each chest) taking the
contents more correctly.

The above shows, if figures will show anything, that a great loss to
both the producer and Customs takes place by the system in vogue. As
the only object of the Customs _should_ be to arrive at the true weight
of Tea in the most expeditious and simple way, how very absurd is the
system pursued! What the _tare_ is can in no way signify to them; all
they really want is the weight of the Tea. The absurdity of the system
is proved by the fact (demonstrated) that the results to both producer
and Customs _can_ be altered by the addition or subtraction in the
Custom-house of a few nails! How easy to weigh the Tea itself! What
possible objection can exist?

The Indian Tea Districts Association having failed to move the Customs,
have quite lately addressed the following Memorial to the Secretary of
State for India:--


  TO THE RIGHT HONOURABLE THE EARL OF KIMBERLEY, HER MAJESTY’S
  SECRETARY OF STATE FOR INDIA.

  The Petition of the Indian Tea Districts Association sheweth--

  That your Petitioners are a body representing the interests connected
  with the cultivation of Tea in British India, in which enterprise
  British capital to the extent of over fifteen millions sterling has
  been invested.

  That the industry dates from the year 1838, when the first
  consignment of Indian Tea, consisting of 456 lbs., reached the London
  market.

  That the imports of Indian Tea for the year ending 30th June,
  1882, were 49,503,000 lbs., having a value of more than £3,300,000
  sterling; while the estimated importation for the current season
  is upwards of 55,000,000 lbs., or fully one-third of the entire
  consumption of the United Kingdom for the year.

  That the contribution to the Revenue accruing from Customs’ import
  duty on the above quantity of Tea will exceed a million and a quarter
  sterling.

  That the whole of this large quantity is manufactured and packed on
  between 2,700 and 2,800 separate estates, situated on various parts
  of H.M.’s Indian dominions.

  That the boxes in which the Teas are packed are in great part made
  of such wood as can be obtained on the several estates, or purchased
  from the neighbouring Forest Department, and it is very important on
  economic grounds, as also in the manifest interests of the districts,
  that this should be exclusively the case.

  That it has been found, under these conditions, practically
  impossible to meet the imperative Custom-house standard of close
  uniformity of tare weight when the chests reach the Bonded Warehouses
  here.

  That your Petitioners have reason to complain of the system of
  weighing the Teas in the said warehouses for the purpose of levying
  the duty.

  That the present system of weighing is to weigh each package in the
  gross, then to turn out the contents, weigh the empty case, and thus
  arrive at the nett weight of the contents.

  That the only exception to this rule is when the package, _i.e._, the
  empty cases, in a Break closely approximate in weight.

  That by the said system of weighing, two serious injuries are
  inflicted on the grower and importer of Indian Tea, viz.:--

  In the first place, a loss of weight is sustained by the fractions
  over the even pound in both gross and tare being given against the
  seller, and in favour of the buyer, amounting, it may be, to 1 lb.
  15 oz., or an average of about 1 lb. in every package weighing over
  28 lbs. gross, in addition to the usual trade allowance of 1 lb. per
  package.

  Secondly, and by far the more serious grievance, very great injury
  is caused to the Teas by the process of turning them out of the
  packages, in which they arrive hermetically sealed, for the purpose
  of weighing the empty packages. The Teas are thus exposed to the
  atmosphere, the humidity of which they readily absorb, and sustain
  further serious injury and depreciation by breakage from rough
  handling in the process of repacking: the lead linings also are so
  torn in the process as to be rendered comparatively useless for the
  purpose for which they were intended, eliciting loud complaints from
  the trade of the rapid loss of condition of the Teas.

That the concession of this Petition, by rendering it unnecessary to
turn out more than a small percentage of the chests to test the correct
weight of contents, would admit of the Teas being bulked in India;
and while it would free the industry from an injurious and vexatious
restriction, and admit of the Teas reaching the consumer in a purer and
sounder condition, it would also greatly simplify and reduce the work
of the Customs.

That the foregoing statistics significantly demonstrate the importance
of the Indian Tea industry to both England and India, and constitute a
claim to the favourable consideration of both Governments, especially
that of India, on the ground of the benefit accruing to the districts
in which it is conducted, and the increment of State revenue to which
it has directly and indirectly conduced.

That having regard to the existing close and hardening competition
with China, Japan, and other Tea producing countries, your Petitioners
naturally feel aggrieved that the important industry they represent
should be hampered in the contest by the restrictive and superfluous
impediment forming the subject of their petition.

That your Petitioners have unsuccessfully urged on the Commissioners
of Her Majesty’s Customs the adoption of this change of system, and
therefore venture to address your Lordship.

That your Petitioners beg to refer to the accompanying copies of
correspondence between the Association and the Commissioners of Her
Majesty’s Customs annexed to this Petition.

That the accompanying Memorial signed by the leading mercantile firms
and others in Calcutta, interested in the growth and export of Indian
Tea, is an illustration of the feeling in India on the subject of this
Petition.

Your Petitioners therefore pray--

That your Lordship will kindly take such steps as may be necessary to
secure for your Petitioners the relief sought for.

  And your Petitioners will ever pray, &c.

      T. D. FORSYTH,
          Chairman of the Association.

      ERNEST TYE,
          Secretary.

The following reply was received:--

  INDIA OFFICE, S.W.,
      28th February, 1883.

  SIR,--I am directed by the Secretary of State for India in Council
  to acknowledge the receipt of the Memorial addressed to the EARL
  OF KIMBERLEY by the Indian Tea Districts Association, respecting
  the method of weighing Indian Tea at the Custom House. In reply, I
  am to inform you that the Memorial has been forwarded to the Lords
  Commissioners of Her Majesty’s Treasury, with the expression of LORD
  KIMBERLEY’S hope that whatever is practicable may be done to remedy
  the grievance complained of by the memorialists in the interests of
  the Indian Tea trade.

      I am, Sir, your obedient servant,
          (Signed)       J. K. CROSS.

  The Secretary, Indian Tea Districts Association.

It is possible, therefore, that some improvement will now be
accomplished.[107]

But at the CRUTCHED FRIARS Warehouse (belonging to the East and West
India Docks) a great advance has already been made. The Tea there is
now bulked, and re-packed by machinery. The Directors most kindly
invited me to come and witness the process. I went, and was more than
pleased with what I saw. The machinery, and all connected with the
process, is so well described in an article in the _Home and Colonial
Mail_, I cannot do better than give it here:--


  TEA BULKING AT THE EAST AND WEST INDIA DOCK COMPANY’S WAREHOUSES, IN
  CRUTCHED FRIARS.

  It is not a little strange that the importance of effecting
  improvements in the present system of Tea bulking, which has
  exercised the minds of Tea growers and importers so much of late,
  should have hitherto been neglected or ignored by the proprietors of
  the various bonded warehouses in London wherein the Tea is bulked
  and stored. That Tea may be, and only too commonly is, bulked by an
  antiquated and unsatisfactory process is a fact which is well known
  to all who are interested in the matter. How this result is arrived
  at will be seen later on; at present we desire to show that at least
  at one warehouse the question has received the attention which it
  deserves, and to explain, so far as may be possible, the steps which
  have been taken in the matter.

  It is, then, that old and powerful body, the East and West India Dock
  Company, who have taken up the matter. At the instance of Mr. Du Plat
  Taylor, the able and energetic secretary of the company, supported
  by the equally energetic warehouse superintendent, Mr. Robert Adams,
  the arrangements for bulking Tea at the warehouse of the Company have
  been very greatly improved. More than this; there has been invented
  and set up a special and very ingenious machine for the bulking of
  Tea in a manner which avoids all the failings of the old system. What
  this machine is, and what its peculiar merits are, will best, and
  perhaps only, be clearly understood by a brief description of the
  two systems as we lately saw them in operation at the warehouses of
  the company in Crutched Friars, which we may mention are nearer than
  any others to Mincing-lane, an advantage securing to planters and
  importers the certainty that their Teas will be sampled by the trade
  generally.

  Under the old system, then, each chest of a break, after having been
  subjected to certain preliminary formalities, is opened, and the
  Tea turned out in a heap on the floor of the warehouse. When this
  is done the Tea is bulked by means of wooden spades, each spadeful
  being thrown to the top of the central heap, so that it falls over
  and on all sides. Here the Tea lies until it is placed back again in
  the chests after they are tared, there being a considerable interval
  at some of the London warehouses between the bulking and refilling.
  The refilling is thus accomplished. The Tea is first put into bags
  and weighed on a machine at the side of the bulk. The bag and chest
  are then taken off the weighing machine and the contents of the bag
  are emptied into the chest. The Tea, however, requires some pressure
  to force it into the chest, and this pressure is obtained by an
  expedient of a very primitive kind. When the chest is partly filled
  a man gets in and presses down the Tea by treading on it. So soon as
  the Tea is all in the chest the package is properly secured, and the
  operation is completed.

  Now the serious faults of this plan are at once apparent. In the
  first place the Tea, being in heaps on the floor of the warehouse
  with a large surface exposed to the atmosphere, runs the risk of
  losing a great deal of its freshness and aroma, this risk being
  largely increased by the doors of the warehouse being kept open in
  order to discharge or to receive merchandise in all weathers. No
  atmospheric influences are calculated to benefit Tea. Then, again,
  the shovelling of the Tea by means of wooden spades, and the treading
  into the chests, can hardly do otherwise than injure the Tea--the
  filling in a minor degree and the treading to a more serious extent,
  the result being, of course, that the Tea is depreciated.

  The East and West India Dock Company have made the best of this
  primitive method of Tea bulking. In the first place it is insisted
  on in their warehouses that previous to trampling the Tea into the
  chests, a cloth shall be placed over it to preserve it from the dirt
  of the man’s boots, and to some extent from injury--a precaution
  which, strange as it may seem, is not taken in every bonded
  warehouse. Then, again, Mr. Adams, the warehouse superintendent--who
  could hardly have the interests of planters and importers more at
  heart were he “in Tea” himself--uses his best endeavours to refill
  the boxes with as little delay as possible, and thus to prevent it
  from being injured by undue exposure to the atmosphere. He also keeps
  the floors of the warehouse as clean as practicable. But feeling
  that the best efforts, however well devised, and however strenuously
  carried out, must necessarily be attended with but partial success,
  the East and West India Dock Company have erected--as has already
  been mentioned--a Tea bulking machine, a device which is ingenious
  and meritorious, and which seems to be, so far as it has been tried,
  a great success.

  This machine, designed by Mr. Tydeman, of the company’s engineering
  staff, and constructed under his supervision, consists, firstly, of
  a large hollow revolving drum weighing nearly two and a-half tons,
  and of sufficient capacity to thoroughly bulk about 50 chests of Tea.
  The drum is made to hold about 100 chests of Tea, which leaves ample
  space for the bulking of the above quantity. Inside this drum are
  frames fitted at intervals with iron rods, and extending at varying
  angles from the axle of the drum to its extremity. Externally the
  drum has two openings for the reception of the Tea, and two smaller
  ones for its discharge. In a line with the axle of the drum, some
  height from the floor, is a platform to which the chests are conveyed
  by a double lift which simultaneously ascends with a full chest and
  brings down an empty one. Adjuncts to the machine are a weighing
  machine, a presser, and four beaters--of the two latter the nature
  and object will be immediately apparent. The process of bulking
  as effected by this machine is briefly as follows: The drum being
  revolved till its receiving openings are level with the platform,
  a chest of Tea is raised, as before explained, and the contents
  examined on the door of the drum, which falls back into a horizontal
  position for that purpose, then by closing the tray or door the Tea
  is passed into the drum. The lift then brings up another full chest
  and takes down the emptied one, which is at once taken to a scale for
  taring purposes, and so the process is continued till the break is
  exhausted. This filling process can be carried on at both sides of a
  drum at once, as there are two openings and two lifts. The Tea being
  in, the drum is made to revolve, when the iron frames thoroughly mix
  the Tea in a very few revolutions--three would suffice.

  The drum has now to be emptied, and this operation is effected in
  the following manner:--The revolution of the drum is stopped when
  the openings through which the Tea is released are brought over the
  weighing machines--there are two for greater expedition--on which are
  placed the chests ready to receive it. The delivery doors (worked
  by levers) being opened, the Tea is allowed to descend till the
  chest is about half full, when the presser and beaters are brought
  into play by hydraulic pressure. The presser is a piece of flat
  iron about an inch in thickness, removable at pleasure, and varies
  in size to fit either a chest or a box. The beaters are four pieces
  of the same metal, which support the chest so soon as it is on the
  weighing machine. When the chest is partly filled, the beaters are
  released, and, by the action of a wheel, are made to strike all four
  sides of the chest, and thus shake the Tea down. The presser is
  also brought down to press the Tea in. The action of both of these
  agents can be regulated to any required degree of force. Thus by
  degrees the chest is filled, and (the supporting beaters having been
  released and the presser raised) is weighed and ultimately removed.
  Such, in brief, is the action of the new Tea bulking machine. One
  or two points, however, remain to be mentioned. The power by which
  the machine is actuated is hydraulic. The presser will not injure
  the Tea. The beaters serve the triple purpose of holding the chest
  in position on the weighing machine, of supporting it should it be
  of weak construction, and of materially assisting the repacking of
  the Tea. The beating action does not in any way injure the chests.
  Our readers will also be pleased to know that certain very marked
  improvements even upon the above described are already in hand by
  this Dock Company--improvements which will greatly increase the value
  and usefulness of their machinery for bulking Teas.

  To descant on the advantages over the old system of bulking which are
  possessed by the machine which has been described would be little
  better than a waste of time. Yet some few points may be briefly
  referred to. First, cleanliness is secured, for from first to last
  the Tea is never touched by hand or foot. Again, the Tea cannot be
  injured, nor can it lose its aroma, for it is never exposed to the
  atmosphere at all. Instead of being allowed to lie on the floor
  of the warehouse for any period, the entire process of bulking is
  completed without break or delay. The Directors of the East and West
  India Dock Company are not, of course, so sanguine as to imagine that
  the old system of bulking will be at once abandoned; indeed, they
  have, as has been mentioned, taken steps to improve that system;
  they do, however, think that it should be abandoned, and to that end
  have adopted the Tea Bulking Machinery as an alternative, and an
  immeasurably superior process. That they are justified in this view
  there can be no doubt in the minds of those who have witnessed both
  the systems in operation.

The said machinery is at the CRUTCHED FRIARS Warehouse alone, and it
is, of course, very desirable the machinery should be adopted in all
Tea warehouses. This end will be quickly brought about if those who
send their Tea home, and the importers here, insist on their Tea being
sent to this one warehouse that has the machinery.

What an advantage to owners and managers of Tea estates is the fact
that Tea bulked by machinery at CRUTCHED FRIARS is not exposed to the
changeable English atmosphere, or at least not for more than a few
minutes, and consequently is not so likely to be classed as “flat.”
How many planters are there who, after taking especial care in the
manufacture of their crop, find to their chagrin that on arrival in
London (and after exposure probably for some days), the shipment is
described as “flat,” and worth so many pence per lb. less than if the
atmospheric exposure had not occurred.

It appears to me that very little, added to the help this new machinery
gives, would now do away with _all_ the injury the producer and the Tea
has hitherto borne in the Customs. So much has now been accomplished
by this machinery, the Tea is well bulked, and _receives no injury
whatever thereby_. But two further improvements are required:--

  1. That the actual weight of Tea in each chest (discarding ounces)
  be recorded, and that thus the loss to the producer and the Customs,
  detailed above, be avoided.

  2. That the lead at top of the Tea be carefully replaced and
  resoldered, so that every chest shall leave the Custom House in as
  good condition as it entered it.

Very little addition to the machinery detailed above would accomplish
the first. The chest ready to receive the Tea, plus the lid and top
lead (which should have been carefully removed), might be weighed on
the platform at the side of the big drum (by simply making the said
platform a weighing machine) and weighed again when filled, with the
lid and lead laid on it. The difference of the two weights would, of
course, be the weight of the Tea.

The second is a question of expense; it would not be great if done
systematically. The chest should be carefully opened, and the top lead
removed in a square piece _nearly the size of the box_. When replaced,
a narrow strip of lead, soldered down on either side, would make the
covering complete.

Justice will not be done to Indian Teas till this last is accomplished.

Who should bear the expense? The chests are received into the Customs
for the benefit of the Revenue, and who can doubt, were the question
tried in a Court of Law, that they are bound to return them in as good
condition as they were received. They do _not_, and have never done
so, and I only wonder the trade has stood it so long, and has not sued
them. Were the course I advise followed out, there would remain no
cause of complaint, and the trifling cost of soldering on the lid again
should doubtless, therefore, be borne by the Customs.

But in reality the Customs would sustain no loss--in fact, the other
way. I have shown clearly at page 278 _that were the weight of Tea
correctly recorded, the Customs would receive in duty upwards of
£11,000 each year from Indian Tea more than it does now_. To re-solder
the lids on the boxes would cost nothing like that; and highly as
Indian Tea is thought of now, how much higher still would it stand were
it not injured to the frightful extent it is in passing through the
Customs.


CONCLUSION.

I lit on the following in the _Home and Colonial Mail_ just before
going to press, and it is too pertinent to much in preceding pages to
omit:--


  THE CHINA TEA TRADE.

  The influence of the expansion of the Indian Tea enterprise on the
  trade in China is being felt. We have more than once adverted to the
  fact that the growing use of the well-flavoured Teas of India would
  diminish the consumption of the better grades of China Tea, and that
  the effect of the competition between the two countries would be
  first seen in the falling off in the demand for so-called fine China
  Tea.

  The following letter, which appeared in the _Times_ Money article
  lately, confirms this view, and refers to the present unsound
  condition of the China Tea trade:--

  “Sir,--In view of the opening of the Tea season in China, a few
  remarks upon the present position and future prospects of this
  important trade may not be inopportune.

  “It is no secret that for some years past the losses of merchants
  have been serious, and that while most of the wealthy firms so long
  known as connected with China have either entirely ceased to import
  Tea, or have reduced their operations to a very small compass,
  the trade has been carried on by new houses possessing but little
  capital, who are enabled, by the competition of the banks, to do a
  large business by drawing bills on China, not only for the whole
  cost of the Teas purchased, but also for their commissions on these
  purchases--that is to say, for an unrivalled profit of 3 per cent.
  The question, Who has so far paid the losses of the past two years?
  is one that greatly exercises the minds of the trade. Many suppose
  that large balances are being carried over in the books of some of
  the banks, or by the Chinese, and that it is the hope of recouping
  a portion of this loss that induces the banks or the Chinamen to
  support those who would otherwise be obliged to relinquish the
  trade. The Chinese have also a further inducement to support such
  firms, since it is partly through them that those high prices are
  established in China at the opening of the season which entail so
  much loss afterwards. As a result of these prices, about 30 per cent.
  more fine Congou is produced than (on account of the competition of
  the Indian growth) can be consumed except at the price of medium Tea.
  How large the excess is may be gathered from the fact that, although
  5,000,000 lbs. of this class of Tea was lost last July in the
  ‘Moskwa’ and the ‘Fleurs Castle,’ yet stocks in Russia have increased
  by about 30,000 half-chests, and there is still so large a quantity
  on this market that it can only be realized at a loss of from 5_d._
  to 6_d._ per lb. on the China cost; thus some Teas, said to have cost
  in Hankow 1_s._ 8_d._ to 1_s._ 9_d._, have been recently sold as low
  as 1_s._ 3_d._, and others costing 1_s._ 7_d._ in Foochow, have been
  sold at 1_s._ 1_d._ per lb.

  “It is evident from the above that merchants as a rule do not realize
  the immense change that has been brought about in the conditions
  of the trade by the enormous increase in the use of Indian Tea,
  which now forms about one-third of the entire home consumption, and
  competes mostly with the finer qualities of China congou; nor the
  fact that all engaged in the trade are becoming year by year more
  averse to holding stock on account of the heavy charges involved,
  and the risk of deterioration in quality. Yet, as the whole twelve
  months’ supply of first crop Tea arrives within three months of
  the opening of the season, it is plain that some one must hold the
  balance, which can only be done with safety if the Tea be bought at
  a very low price.

  “The one remedy for the present condition of things is that the great
  bulk of the so-called fine Teas should be bought in China at their
  present value on this market--viz., at about 5_d._ to 6_d._ below the
  prices given for them in recent years. With the large accumulated
  stocks in Russia, and consequently reduced orders from that country,
  the yearly-increasing supply of Indian Tea, and the present prices
  here, one would think that such a course would at once be adopted.
  Unfortunately, however, so much of the Tea is bought on commission,
  and the Russian agents seem so reckless as to the prices which they
  give, that any such prudent action can hardly be hoped for. It would,
  therefore, be wise for holders of shares in Eastern banks, as well
  as all who have been in the habit of intrusting orders to buying
  agents in China, to ponder the foregoing facts, which can be easily
  verified by a reference to any of the trade circulars lest their
  money should be lost in the crash which must certainly take place if
  the past policy of Tea buyers in China be continued.--I am, &c.,

  “A. B.”

Will those warned be wise in time, and not swamp the Home Market with
China Teas certain to be sold at a loss? Who can say? But “A. B.” is
evidently master of the subject, and if his advice in not taken, the
China Tea “crash” he predicts will not be a small one.

When China Teas are _not_ sent home to realise a certain loss, our
Indian Teas will have fairer play.

I cannot conclude without acknowledging the great help I have derived
from the pages of the _Tea Gazette_ in writing these additions to my
Fourth Edition.

       *       *       *       *       *

Since my remarks on Ceylon were printed, I have acquired much further
information regarding the Tea industry in that island, and the
prospects certainly seem very favourable. Anyhow, there seems to be no
doubt that Ceylon for Tea offers quite as good a field as any part of
India, always supposing that good sites are selected and the area to
choose from is large.

The future market for Tea is really, as regards Ceylon, the only
doubtful point, and consequently (as at page 183) I advise the planters
there to act with caution.

Where it is proposed to put coffee lands under Tea, of course one
great advantage in economy will be gained, inasmuch as there will be
no jungle clearing or previous cultivation. But here again caution is
necessary. Make sure the soil is not worn out, for Tea, though it will
grow, will not yield largely on such.

       *       *       *       *       *

  _June, 1883._

P.S.--The following are the new rules lately issued by the Customs
regarding the future treatment of Indian Teas.

  The weight of Indian Tea for duty may, if desired by the importers,
  be ascertained under the following regulations:--

  1. The Tea on arrival to be weighed to ascertain the gross weight of
  each package.

  2. With each entry the importer to give an endorsement of the net
  contents of each package.

  3. To test the accuracy of this endorsement, 10 per cent. of each
  break to be turned out and weighed net.

  4. If the difference between the weight given of any package and the
  weight found exceeds or is less than 3 lbs., the whole parcel should
  be weighed net.

  5. Duty to be charged on the average weight of the packages weighed
  net, unless the importer elects to weigh the whole parcel in the
  usual way.

  6. When the average of the packages weighed net amounts to so many
  pounds and a-half, an additional pound will be charged on each of the
  whole parcel; when the fraction is less than half a pound it is to be
  rejected.

  7. The new system to come into operation on July the 1st next.

FOOTNOTES:

[106] If tares are nearly equal, and if Teas are well bulked in India,
only some packages (about 10 per cent.) are opened, and an average tare
struck. But this in no way saves the loss in _quantity_ of Tea, though,
of course, less Tea is thus _injured_.

[107] Since I wrote the above the Customs have framed new rules for
Indian Teas. The absurd tare system is done away with.



ADDENDA

_TO THE THIRD EDITION_.


The following from the _Indian Economist_, regarding Indian Teas in
general and Neilgherry Teas in particular, is not out of place here. At
the same time I do not agree with the writer, for I believe that in the
strength and pungency of Indian Teas consists their value:--


  INDIAN TEA.

  “That the Teas of India have at length come to be fully appreciated
  in England may be taken, we presume, as an admitted fact; and it
  is of importance that planters should direct their attention to
  modifying their methods of manufacture so as to suit the public
  taste, and, if possible, turn out an article free from the objections
  still advanced against the Indian leaf as a daily beverage. There
  are, we know, those who argue that enough has been done, and that
  consumers will acquire a taste for the produce of our gardens in
  time; but we have daily evidence that in the most trivial matters
  there is no greater tyrant than the public. It behoves those then who
  cater for this tyrant to consult its taste and satisfy its demands,
  however exacting and capricious they may be. The remarks we are about
  to make are based on experiments and enquiries extending over some
  years in this country and in England, and we leave those engaged in
  the enterprise to estimate their value. All Teas grown in the plains
  of India are known to the trade in London under the general name of
  Assam, and are chiefly used for mixing, seldom reaching the consumer
  in a pure state. When they do, the objections raised are that the
  leaf is too pungent and rough for most palates; and purchasers are
  in the habit of mixing it with Chinese to tone down those astringent
  qualities. In other words, it wants the delicacy of flavour which
  is the chief characteristic of the Chinese leaf, meaning of course
  that vended by respectable houses, not the abominable trash that
  formed part of the cargoes of the _Lalla Rookh_ and _Sarpedon_,
  containing, according to Dr Letheby’s Analysis, ‘40 to 45 per
  cent. of iron filings and 19 per cent. of silica.’ Nor is this lack
  of delicacy of flavour to be lightly regarded, for the efforts of
  our manufacturers have been directed unwittingly and indirectly
  to foster the peculiarity, as the test of Indian Tea has hitherto
  been its strength and pungency, to fit it for _salting_ weak, thin,
  inferior sorts of Chinese. This is what the dealers have demanded,
  and what, consequently, brokers in their turn have insisted on, with
  the result that the out-turn of our Assam and Cachar plantations is
  now, if anything, too powerful to suit public taste. Whether means
  of manipulation may be hit upon by which aroma can be retained
  without sacrificing strength, we leave those most interested to
  determine; but it is worthy of note that this objection to strength
  and roughness is almost confined to women, the sterner sex preferring
  Assam unmixed, while the working classes of both sexes are unanimous
  in favour of the unadulterated Indian article. Experiments were
  further tried by substituting Neilgherry Tea, and after a short
  interval the verdict of the majority was in its favour. We need
  now only point out the difference in the manufacture between the
  two Teas, leaving others to decide questions regarding the bearing
  of climate or altitude. Up to the time of finishing rolling, the
  manipulation of the leaf is identical, care being taken to retain the
  juice; but that made on the hills instead of being almost immediately
  placed over _choolas_ was spread out thinly on tables all night, in
  a temperature of 54 deg., sustaining consequent loss of strength by
  evaporation, but developing an aroma that established it at once in
  favour. So successful has this Neilgherry Tea been at home, that
  offers are now received by plantation proprietors for their produce
  at half-a-crown per lb. free on board, in Madras. This would seem to
  indicate that the aroma is generated by the action of cold upon the
  damp leaf while in a state of ‘suspended fermentation;’ for, previous
  to experimenting with consumers, the samples were submitted to
  Mincing Lane brokers and pronounced sound, in corroboration of which
  opinion the bulk from which they were taken sold at auction for 2_s._
  2-1/2_d._, so that fermentation (_i.e._ sourness) had been carefully
  avoided. We know that the climate of Assam and temperature of the
  Tea-houses render the keeping of rolled leaf even for an hour fatal
  to soundness; but should the development of this aroma be really due
  to ‘suspension of fermentation’ is it not worth while adopting some
  contrivance for cooling down a chamber set aside for the purpose of
  spreading out the rolled leaf to the temperature required?

  “The question whether delicacy is due to altitude alone and not
  to manufacture might be ascertained by experiment. Let a quantity
  of green leaf be sent _down_ from one of the Neilgherry gardens,
  and worked up in the plains at the foot of the hills, and an equal
  quantity sent _up_ from one of the Assam gardens, say to Shillong,
  and manufactured on the Neilgherry principles there, and the result
  then compared. This experiment would cost little and determine a not
  unimportant question: for all engaged in Tea are interested in using
  their best endeavours to fit it for public consumption, and to guard
  it against Chinese in any shape or form whatever.”

  _Note by the Author._--That “delicacy of flavour,” and “want
  of strength” with it, _is_ due to altitude has long ago been
  admitted, and any experiments on that head would, I think, be quite
  unnecessary. The experiments as to manufacture on the Neilgherries
  are interesting, and should be further looked into.

  E. M.

  I have at last completed experiments with a view to do away with the
  use of charcoal in Tea manufacture, and I think with success.

  The “Furnace Teas,” for so I purpose naming them, have in most cases
  been pronounced by the Calcutta brokers to be superior to similar
  samples of the same day’s leaf, made in the usual way over charcoal.

  Nothing but the heat generated by _any_ fuel placed in furnaces sunk
  under ground outside the Tea-house is used. No motive power of any
  kind is employed. The apparatus is very simple. It is cheap to erect
  and very durable in character.

  As the apparatus with which the Teas up to the present time have been
  made is a rude and imperfect one, having disadvantages which must
  tell more or less on the excellence of the Teas so manufactured, and
  as, even with these disadvantages, the Teas are pronounced by the
  brokers _at least_ equal to charcoal-dried Teas, it is not too much
  to hope that with a perfect apparatus (one of which will be erected
  immediately) Teas will be improved in value by this new invention.
  The following will be shortly the advantages of this new process,
  even supposing the Teas are no better:--

  1. _Economy._--This will possibly be even greater than what is set
  out in the extract of the local paper below; for the fact that the
  Tea is never placed over charcoal until the whole is ignited, and has
  become “live charcoal,” is not there recognized, much of the caloric
  thus escapes.

  2. Cleanliness and absence of charcoal dust.

  3. Absence of the objectionable fumes of charcoal.

  4. Immunity from fire in Tea-houses.

  5. Greater speed in the firing process, and the saving of all the
  labour employed to make charcoal.

  6. Reduced temperature in Tea-houses.

  If all the advantages are, as I expect they will be, attained, the
  life of a Tea planter will be more pleasant than hitherto.

  The following is the opinion of the new process expressed by the
  _Darjeeling News_ of 1st August:--

  “It has long been a question, which all planters were desirous to
  solve, if the fumes of charcoal were necessary to make Tea, that is
  to say, if any chemical action was produced on the Tea by the said
  fumes, and if not, whether it would not be possible to do the firing
  in some other and far cheaper way.

  “The question has, we believe, been solved by Colonel Edward Money,
  and if so, for the invention is quite a new one, a boon of great
  magnitude will have been conferred on the Tea interest of India. We
  congratulate this district as being the birthplace of the improvement.

  “The apparatus at present in use at Soom, and which we have seen
  working, is a rough and crude one made on the spot. This, and the
  more perfect plans from which larger and better ones are to be made,
  are readily shown by Colonel Money to anyone visiting Soom; but until
  the invention is patented, it is not well to describe it in print.
  Suffice if we say the invention is a remarkably simple one--cheap to
  erect--durable in its character, and the working thereof unattended
  with any expense whatever, beyond the cost of the fuel (which may
  be of any kind), and which of course will be many times less than
  charcoal.

  “If true, as we hear, that it takes 3-1/2 maunds of wood generally
  to make one maund of charcoal, and if also true, as Colonel Money
  suggests, that the caloric in one maund of wood equals the caloric
  in two maunds of charcoal, it then follows that each maund of wood,
  put into Colonel Money’s furnace, equals seven maunds of wood to make
  charcoal.

  “Of course the above are more or less random figures, but they
  suffice to show that the saving of fuel will be very great--a boon
  of course to planters, but a boon also to the Forest Department and
  to India.

  “We knew of the invention some time back, but we forbore to notice
  it until the brokers’ reports on the Tea so made had been received.
  We have now seen these. Samples of ‘charcoal’ and ‘furnace’ Tea were
  sent down, made from the same leaf, the same day, and manufactured
  in one up to the “firing” process. Two brokers give the higher value
  to the furnace Tea, one to the charcoal kind--but the difference is
  small.

  “We believe, as one of our most experienced planters, who has tasted
  the Teas, been to Soom, and seen the brokers’ reports, says, that
  ‘the Tea dried by the furnace apparatus will be _at least_ equal to
  that prepared over charcoal.’

  “As Colonel Money is already known as an authority in Tea, and as
  he has stated to us his belief that ‘charcoal days’ for Tea are
  now at an end, we await with confidence the ultimate success of
  his invention, which even if it makes no better Tea will certainly
  make it far cheaper, while the dirt from charcoal dust will be done
  away with, the temperature of the Tea-houses much reduced, and the
  deleterious fumes of charcoal, so very objectionable from a sanitary
  point of view in Tea manufacture, will be known no more.”

  Again, 29th August, a month later, the _Darjeeling News_ further
  remarks:--

  “We alluded recently to Colonel Money’s very ingenious plan for
  drying Tea without charcoal. Since then his apparatus has been
  in full work at Soom, and has been inspected by numbers of the
  Darjeeling planters, one and all of whom have, we understand,
  reported most favourably on its working. Samples of Tea manufacture
  have been from time to time sent to Calcutta brokers for their
  opinion, and reports have been received from fifteen, of whom seven
  are in favour of Tea made by the old charcoal process, seven are in
  favour of the new furnace process, and one reports that the Tea made
  by each process is exactly the same.

  “Colonel Money is now taking steps to erect his improved furnace,
  which will be in working order by the end of September, and the whole
  October crop of Soom Tea will be fired by the new furnace.

  “Colonel Money has applied for a patent, and as soon as this is
  granted we hope to give our readers a description of the apparatus.
  For obvious reasons it would not be advisable to do so before then.
  We may mention here that one of the most intelligent and practical
  planters in this district has ordered one of Colonel Money’s flues
  for his private garden.

  “Of the commercial success of Colonel Money’s apparatus we have no
  doubt whatever, and we trust that Colonel Money will reap a handsome
  profit from his very ingenious invention, which will be an undoubted
  boon not only to this district, but to all the Tea-producing
  districts of India.

  “One point which has struck us as good in Colonel Money’s apparatus
  is that the temperature of the Tea-house is considerably lowered
  during the firing process as compared with the open _chulas_, and
  that there is no free carbonic acid gas allowed to escape into the
  Tea-house, so that those very unpleasant symptoms of slow poisoning
  which often show themselves in planters and Tea-makers will be
  unknown in future. At our suggestion Colonel Money has decided to
  keep a register of the maximum temperature of the Tea-house, whilst
  the open _chulas_ continue in use, and to compare it with the
  temperature when the new apparatus has superseded them, also to test
  for free carbonic acid gas in the air with each process.

  “We are convinced that when the figures are available our readers
  will be rather astonished at the difference from a sanitary point of
  view.

  “On the whole, we think that Colonel Money’s invention is by far the
  most important application of _common sense_ and scientific knowledge
  to Tea manufacture that we have yet seen, and we are almost certain
  that his apparatus will before long be adopted throughout the Indian
  Tea districts.”[108]

FOOTNOTE:

[108] Note to Third Edition.--No. The furnace has been erected but
on two or three gardens. Other inventions have since been brought
forward, and the whole matter is still in an uncertain state--I mean
as to which of the several apparatuses is the best. I believe in mine
still, and intend to erect it on the Western Dooar Gardens in which I
am interested, but, of course, I am not an impartial judge! One thing,
however, I lay claim to, and that is, that I was the first to show by
practical results that the fumes of charcoal are in no way necessary to
make Tea.

Note to Fourth Edition.--Since the above note was written (now five
years ago) many Tea Drying Machines have been invented (see pages 240
to 259), and I most willingly admit they are _all_ better than my
furnace apparatus. The first inventor rarely attains perfection, and
as in my case, he generally labours for the benefit of those who come
after!--EDWARD MONEY.



INDEX.


  Area required for a garden, 2
    -- large, a mistake, 2


  Boxes, 147
    -- cost of, 161


  Climate, 14
    -- in each district, 14 to 25
    -- wanted, 14
    -- rainfall, 14, 28
    -- rain table, 28
    -- temperature table, 26
    -- elevation table, 26
    -- cold, 14
    -- hot winds, 14
    -- affects flavour of tea, 15
    -- good for tea, bad for man, 15, 35

  Cultivation, 81
    -- what is it, 81
    -- when a waste of labour, 81
    -- by digging round each plant, 82
    -- weeds not to get ahead, 83
    -- Dutch hoe for, 83
    -- cost each operation per acre, 84
    -- cost of, to 6th year, 84
    -- cost of in full bearing, 85


  Districts, 13
    -- which best, 30
    -- rainfall in, 28
    -- cold in, 26
    -- of Assam, 15
    -- of Cachar, 16
    -- of Chittagong, 16
    -- of Terai below Darjeeling, 18
    -- of Dehra Dhoon, 18
    -- of Kangra, 19
    -- of Darjeeling, 20
    -- of Kumaon, 22
    -- of Gurhwal, 24

  Districts of Hazareebaugh, 24
    -- of Neilgherries, 24
    -- of Western Dooars, 25
    -- meteorological table of, 26
    -- comparative advantages of, 30
    -- soil of, 13 to 25
    -- jungle of, 13 to 25
    -- lay of land of, 13 to 25
    -- price waste lands in, 4
    -- elevation of, 26
    -- temperature of, 26

  Distances for plants, 72
    -- table of, 72
    -- regulated by class, 72
    -- best, 72


  Flushes, 97
    -- number of, 97 to 101
    -- way formed, 104
    -- differ in districts, 98
    -- intervals between, 99


  Hills and Plains--
    -- comparison of, chap. iii.
    -- high elevations bad, do.
    -- table of elevation, 26


  Jungle, 34
    -- what best in Himalayas, 34
    -- not of much consequence in Bengal, 34

  Jungle, coarse grass, 34
    -- cutting, 75


  Labour, 10
    -- local, 10, 11, 12
    -- imported, 10, 11, 12
    -- government action, 10, 11
    -- cost of imported, 10

  Labour in tea districts, chap. iii.

  Laying out a garden, 42

  Lay of land and aspect, 37
    -- flat, sloping, steep, 7, 13, 35
    -- aspect, 39
    -- valleys, 40
    -- narrow valleys, 40

  Lay of land and selection of steep land, 7, 37
    -- disadvantages of steep land, 37
    -- lines on steep land, 46
    -- plants close on steep land, 45

  Leaf-picking, 102
    -- principles of, 102
    -- diagram of shoot, 104
    -- teas made from each leaf, 107
    -- cannot make separate teas in practice, 107
    -- pruning connected with, 102
    -- mistakes in, 42
    -- how shoots form, 104
    -- mode of, 104


  Manufacture, 109
    -- importance of good, 109
    -- old and new plan, 110
    -- withering, 110, 111, 123
    -- rolling, 111, 112
    -- panning, 109, 112
    -- sunning, 112, 128
    -- tea, how judged, 113
    -- Pekoe tips, 105, 106, 114, 115, 116, 122
    -- strong teas and Pekoe tips incompatible, 116
    -- fermenting or colouring, 127
    -- firing or dholing, 128
    -- of flowery Pekoe, 130

  Manufacture of green tea, 130, 144
    -- sifting and sorting, 134, 135, 136, 161
    -- sieves, 135
    -- Chinese sieves best, 135
    -- classes of tea, 137
    -- cost of, 160
    -- ignorance of, 7

  Manufacture, coarse leaf, 126
    -- burntness, 143

  Manure, 67
    -- advantages of, 17, 67
    -- how to apply, 68
    -- quantity, 69
    -- cost of, 69
    -- kinds of, 67
    -- results of, 69

  Management, accounts, forms, 152
    -- what qualities required for a Manager, 152
    -- forms, 153
    -- accounts, 158

  Making a garden, 73
    -- general instructions for, 73

  Mechanical contrivances, 116
    -- McMeekin’s rolling table, 116
    -- Kinmond’s rolling machine, 116
    -- Nelson’s rolling machine, 118
    -- Jackson’s rolling machine, 116
    -- McMeekin’s drawers, 119
    -- Money’s furnace, 121, 296
    -- sifting machines, 121
    -- machine required to separate the leaves, 122
    -- packing machines, 121

  Miscellaneous:--
     Transport, chap. iii.
       -- in each district, chap. iii.
     Green tea, 21, 130, 144
     Stagnant water, 40
     Inundation, 40
     Sections, 42
     Yield, 43, 170
     Lines of plants, 45, 46
     Roads, 45
     Relative price green and black teas, 133
     Yield first 10 years, 170
     Necessities for tea, 173, 180
     Past, present, and future of Indian tea, 174
     Strange facts about tea, 174
     Imports, 177
     Annual consumption, 177
     Collapse of tea speculation, 178
     Share list to-day, 179

  Money matters: will tea pay?, 1
    -- why has it not paid sometimes?, 1, 6
    -- cause of failures, 1 to 8
    -- wilful extensions, 6
    -- price paid for gardens, 8
    -- faulty area sold, 8
    -- cost of making a 300-acre garden, 163
    -- how much profit tea can give, 168
    -- table, result 300 acres for 12 years, 172


  Packing, 147
    -- lead case for, 147
    -- larger each break the better, 151
    -- cost of, 161

  Planting at stake, 59
    -- advantages of, 57
    -- disadvantages of, 58
    -- mode of, 59

  Pruning, 86
    -- time for, 86
    -- instruments for, 87
    -- height to prune, 88
    -- cost of, 88


  Sale Lands, Waste Lands--
    -- sale waste lands, 3
    -- auction system, 3
    -- price waste lands, 4
    -- title, 4

  Sanitation, 35

  Seed, 54
    -- transport of, 55
    -- price of, 7
    -- shade, natural, 62
    -- do. artificial, 64
    -- how to sow, 57
    -- when ripe, 54
    -- treatment of, 54

  Seed as manure, 55
    -- number in 1 maund, 56
    -- proportions that germinate, 56
    -- Government gave seed, 51
    -- indigenous hybrid and China alike, 51
    -- how to increase, 55
    -- nurseries or stake planting best, 57

  Soil, 31
    -- only general rules for, 31
    -- sandy, 31
    -- greasy, 32
    -- poor, 32
    -- Ball on, 32
    -- friable and porous, 32
    -- in Tea districts, 13, 25, 31
    -- clay, 33
    -- decayed vegetation, 33
    -- for seed beds, 62


  Transplanting, 76
    -- holes for, 59, 76
    -- mode of, 77
    -- results of bad, 77
    -- when to be finished, 78
    -- best days for, 79


  Vacancies, 92
    -- difficult to fill up, 92
    -- why difficult, 92
    -- best plan to fill up, 92
    -- large proportion of, 6

  Varieties of tea plants, 47


  White-ants, Crickets, Blight, 89
    -- harm done by crickets, 89
    -- harm done by white-ants, 90
    -- harm done by blight, 91
    -- remedies for crickets, 90
    -- do. white-ants, 91
    -- do. blight, 91

  Weeds, 82
   -- ahead of labour, 83



INDEX

_TO THE ADDITIONS IN FOURTH EDITION._


  Agricultural machinery, 223

  America, 185, 204, 205, 209

  Amsterdam Exhibition, 202 to 211

  Any fuel _versus_ charcoal, 239, 258

  Australia, 201, 202, 204, 205, 207 to 209


  Brick tea, 212


  Calcutta Syndicate, 202, 206, 208, 210, 211, 212, 214

  China tea trade, 288

  China, 194 to 198, 201 to 207, 210 to 212, 288

  Consumption of China and Indian Tea,     201

  Continent of Europe, 202, 211


  Damage to tea by procedure in London, 272, 273

  Darby’s digger, 225

  Date of commencement of tea cultivation in each district, 194

  Deliveries and stocks, 195, 197 to 201

  Discovery of indigenous tea, 194

  Dryers, by Robertson, the Typhoon, 240, 241

     „       Allen, 242

     „       Davidson, the Sirocco, 243, 244

     „       Gibbs and Barry, 244

     „       Shand, 244, 245, 246

     „       Jackson, 246 to 248

     „       Kinmond, 248 to 257


  Fermenting Shelves, 239, 258

  First tea in India, 194


  Greatest and least possible loss by Custom House procedure, 274, 275

  Green tea, 203, 204


  Himalayan gardens, 212

  Hoop iron, 271

  How loss by Custom House procedure could be avoided, 275 to 278


  Imports into Great Britain, 194, 195, 198 to 200, 203

  Increase of Indian Imports into Great Britain, 195

  Indian produce for 1883, 195

  Indian _versus_ China tea, 219


  Jebens’ transplanter, 223


  Local market in India, 213 to 218

  Loss of tea by procedure in London, 272, 273

  Loss on China teas, 288 to 290


  Machinery, 222 to 271

  Making Indian tea known in United Kingdom, 218 to 221

  Manufacturing machinery, 231 to 271

  Markets outside Great Britain, 207 to 217

  Money loss to producers and Customs by method of weighing in vogue, 278


  New mode bulking at warehouse in Crutched Friars, 282 to 286
    -- required further, 287, 288

  New Zealand, 205


  Ornamental tin boxes by Harvey Bros. and Tyler, 266 to 271


  Petition of Indian Tea Districts Association _re_ mode of weighing teas,
        279 to 281

  Planting pots, 223

  Plantations in Northern India, 203

  Ploughing, 223 to 231

  Processes of manufacture, 231 to 271


  Rollers by Jackson, 233, 235, 237

      „      Kinmond, 233, 234, 235, 237

      „      Haworth, 235, 237

      „      Lyle, 236

      „      Greig, 236

      „      Thomson, 237, 238

  Russia, 203, 211


  Sifters, by Jackson, 259

      „       Greig, 259

      „       Pridham, 259

      „       Ansell, 260 to 266

  Sorter for green leaf by Greig & Co., 232

  Statistics of Indian tea, 194 to 206


  Tea outside China and India, 183 to 192
    Ceylon, 183
    Johore, 184
    Japan, 185 to 188, 205, 210
    Java, 188
    America, 188 to 190, 201, 203, 205, 206, 208
    Natal, 191
    Fiji, 192

  Tea consumption per head, 204, 205

  _Tea Gazette_--This is alluded to in most pages, (see 290)

  Thibet, 212


  Weighing and bulking by Customs, 272 to 288

  Weighing teas by Customs, The new rules, 290, 291

  Withering machine, 232, 253, 257



  W. B. WHITTINGHAM AND CO.,
  PRINTERS,
  91, GRACECHURCH STREET,
  LONDON.



RANSOMES, HEAD & JEFFERIES,

IPSWICH & LONDON.

[Illustration]

_SOLE MANUFACTURERS_

OF

ANSELL’S PATENT TEA SORTER.


LATEST REPORT.

(From Mr. D. M. LUMSDEN, Manager of the Borelli Company’s Gardens,
through Messrs. J. WILLIAMSON & Co., London, February 8th, 1883.)

“I find that using ANSELL’S Sifter I have spent Rs. 780 less on my
sorting than last year, besides sorting 500 maunds more Tea; so I may
safely calculate that last season’s working paid for the machine.”

MANUFACTURERS ALSO OF

Portable and Fixed Engines, for burning Coal, Wood, and Vegetable
refuse.

Ploughs specially suited for Indian Agriculture.


Catalogues and all information on application to RANSOMES, HEAD &
JEFFERIES, 9, Gracechurch Street, London, and Orwell Works, Ipswich;
Mr. C. W. ANSELL, Woodcot, Kurseong, Bengal; Messrs. GILLANDERS,
ARBUTHNOT & Co., Calcutta.



BRITISH INDIA

STEAM NAVIGATION COMPANY,

LIMITED.


FLEET OF STEAMERS.

  ABYSSINIA
  AFRICA
  AGRA
  *ALMORA
  ARABIA
  ARCOT
  ASIA
  ASSYRIA
  AVAGYEE
  BAGHDAD
  BANCOORA
  BHUNDARA
  BOOLDANA
  *BULIMBA
  BURMAH
  BUSHEER
  BYCULLA
  CALCUTTA
  CANARA
  CHANDA
  CHILKA
  CHINDWARA
  CHINSURA
  CHUPRA
  *CHYEBASSA
  COCONADA
  COLABA
  COMMILLA
  CULNA
  *DACCA
  *DORUNDA
  *ELDORADO
  ELLORA
  ETHIOPIA
  EUPHRATES
  GOA
  GOLCONDA
  GOORKHA
  *GOALPARA
  HENZADA
  HIMALAYA
  HUZARA
  INDIA
  JAVA
  KANGRA
  KERBELA
  KHANDALLA
  KILWA
  MADRAS
  MADURA
  MAHRATTA
  MALDA
  MECCA
  MEDINA
  *MERKARA
  *NAVARINO
  NERBUDDA
  NOWSHERA
  NUDDEA
  ORIENTAL
  PACHUMBA
  PATNA
  PEMBA
  PUNJAUB
  PURULIA
  *QUETTA
  RAJPOOTANA
  RANGOON (Paddle)
  *REWA
  *ROMA
  SATARA
  SCINDIA
  SECUNDRA
  SIMLA
  SIRDHANA
  SIRSA
  *WAROONGA (Building)

  *Steamers of the British India Association, running on the Trunk
  Lines from London to India and Queensland.


_The various services of the Company are arranged to correspond with
the arrivals and departures of the Overland Mail Packets._

~London to Colombo, Madras, and Calcutta.~--Fortnightly.

~London to Java and Queensland.~--Every 28 days. Calling at Port Said,
Suez, Aden, Batavia, Thursday Island, Cooktown, Townsville, Bowen,
Mackay, Rockhampton, and Brisbane.

~London to Kurrachee, Bombay, and Malabar Coast Ports.~--Fortnightly.
Calling at Algiers, Port Said, Suez, Aden, Kurrachee, Bombay, and
Malabar Coast Ports; and every alternate Steamer calling at Lisbon, the
latter connecting at Aden with the Company’s East African Mail Steamers
for Zanzibar, Mozambique, Quillimane, Inhambane, and Delagoa Bay.

~Calcutta to Chittagong, Akyab, Kyouk Phyoo, and Rangoon.~--Weekly.

~Calcutta to Rangoon and Moulmein.~--Weekly.

~Calcutta to Rangoon, Moulmein, Penang, Malacca, and
Singapore.~--Fortnightly.

~Rangoon to Penang~ (Coasting).--Five-Weekly.

~Rangoon and Moulmein.~--Weekly.

~Rangoon, Tavoy, and Mergui.~--Fortnightly.

~Madras and Rangoon.~--Fortnightly.

~Calcutta and Bombay~ (Coasting).--Weekly.

~Bombay and Kurrachee.~--Bi-weekly.

~Bombay and Persian Gulf.~--Weekly.

The Netherlands India Steam Navigation Company’s Mail Steamers for
Padang, Samarang, Sourabaya, &c., run in connection with the British
India Company’s Steamers.

Passengers booked through to all Ports in India, Burmah, the Straits
Settlements, the Malay Archipelago, the Soudan, North and East Coast of
Africa.


For all information apply to GRAY, DAWES & CO., 13, Austin Friars, and
GELLATLY, HANKEY, SEWELL & CO., 109, Leadenhall Street, and 51, Pall
Mall, S.W., London; BINNY & CO., Madras; ALSTONS, SCOTT & CO., and
DARLEY, BUTLER & CO., Colombo; MACKINNON, MACKENZIE & CO. (Managing
Agents in India), Calcutta, Bombay, and Kurrachee.



HENRY S. KING & CO.,

AGENTS AND BANKERS,

65 CORNHILL, and 45 PALL MALL, LONDON.

  Banking and India Office Agency   45 Pall Mall.
  Shipping and Supply Department    65 Cornhill.
  Baggage Warehouse      14 Worship St., Finsbury.

Branch Firms:

  KING, KING & CO.        _Bombay._
  KING, HAMILTON & CO.    _Calcutta._
  KING, BAILLIE & CO.     _Liverpool._


  ~Banking Accounts~ kept upon the usual terms.

  ~Deposits received~ for fixed periods at rates of interest which may
  be learned on application.

  ~Circular Notes and Letters of Credit~ issued, payable in all the
  principal cities of the world.

  ~Investments and Sales~ effected in all British, Indian, American,
  and Foreign Securities free of charge.

  Remittances ~made to~, and Drafts purchased or collected on, any city
  in the world.

  ~Pay, Pensions, &c.~, collected, and Dividends and Coupons realised.


_SUPPLY AND FORWARDING DEPARTMENT._

  ~Miscellaneous Goods~ supplied on the most advantageous terms.

  ~Newspapers and other Periodicals~ despatched with the utmost
  regularity.

  ~Berths selected and Passages~ engaged by all lines of Steamers
  without charge.

  ~Baggage collected~, shipped, and insured.

  ~Passengers Homeward~ met on arrival at London, Liverpool, or
  Plymouth, and their effects cleared through the Customs and
  forwarded, or warehoused to await instructions.

  ~Goods and Parcels~ forwarded to all parts of the world, or received
  from Abroad and forwarded to destination.


  “THE OVERLAND MAIL,” a Summary of Home Intelligence for India, China,
  and the East, for despatch weekly. Annual Subscription, including
  postage, _viâ_ Brindisi, £1 12_s._ 6_d._, payable in advance.

  “THE HOMEWARD MAIL AND OFFICIAL GAZETTE,” a Summary of Intelligence
  from India, China, and the East, published on the arrival of each
  Mail _viâ_ Brindisi. Annual Subscription, including postage, £1
  6_s._, payable in advance.



W. B. WHITTINGHAM & CO.’S

PUBLICATIONS.

91, GRACECHURCH STREET,

LONDON.

⁂ Agents to Government by Appointment for the Sale of the India Office
Publications.


_Fourth Edition, with important Additional Chapters. Price ~10~s. ~6~d._

TEA--CULTIVATION AND MANUFACTURE.

  By Lieut.-Col. EDWARD MONEY.


OPINIONS OF THE PRESS ON THE THIRD EDITION.

The ~Saturday Review~, in the course of an extended notice, says:--“We
think that Col. Money has done good service by throwing into the form
of a book an essay which gained the Prize awarded by the Agricultural
and Horticultural Society of India, in 1872. The author is one of a
well-known Anglo-Indian family.... He has had plenty of practical
experience, and has tested the labours of other men.... Col. Money’s
general rules and principles, as far as we can form a judgment, seem to
have reason as well as experience on their side.... No tea planter can
afford to disregard his experience.”

The ~Indian Agriculturist~ says:--“Col. Money has advanced with the
times, and the work under review may well be considered the standard
work on the subject, and it ought to be in every tea planter’s hand in
India, Ceylon, Java, Japan, China or America; the merit and sterling
value of his essay has been universally and deservedly acknowledged....
We recommend our readers who require full information and sound advice
on the subject to procure Col. Money’s book.”

~Allen’s Indian Mail~ says:--“The particulars of this great industry,
which comprises (Tea) Cultivation and Manufacture, are given in the
work of Col. Money. The Third Edition expanded from the original prize
Essay published in 1872, by the results of the author’s practical
experience and observations up to the present time, supplies full
details of the origin and progress of an Indian Tea Garden, and that
in a very lucid and readable form.... The publication of so thorough,
clear and instructive a directorium as Col. Money’s work is in itself a
proof of the attention devoted to this important industry, which has a
great future before it. No one who desires to understand the condition
of its development; still more--no one who has a pecuniary interest in
a Tea Garden, can feel that the subject of tea is known until this work
has been studied.”

The ~China Express~ says:--“The experience gained since 1872 is added
to the work, and it now forms a most complete guide to the tea planter.
The great progress the cultivation of tea is making in India renders
a practical work of this kind very valuable; and the method in which
Colonel Money deals with the subject shows his thorough knowledge of
it.”

The ~Scotsman~ says:--“With respect to the conditions of climate and
soil necessary for successful tea cultivation, the requirements of the
plant in the way of water, &c., the varieties best suited for culture
in the various districts, the laying out of the tea garden, and all the
various details of cultivation and manufacture, Colonel Money writes
with the authority derived from many years of experience; and in the
present edition the fruits of his latest experience are embodied. To
new beginners in tea cultivation this book must be of the greatest
value, while it will be found full of interest by outsiders who may
be desirous of information about the condition and prospects of an
important department of agricultural industry.”

The ~Produce Markets Review~ says:--“Colonel Money is a practical tea
planter, and his work is the standard work on the subject, so that it
should be procured by all who are interested in the subject. The new
edition is greatly enlarged, and corrected by the experience of the
past six years.”

The ~Planters’ Gazette~ says:--“The cultivation of tea in the British
dominions is becoming a rapidly extending industry, and we are glad to
see that Colonel Money’s prize essay has reached a third edition, for
it is full of practical information and deserves to be studied by every
tea planter.”

The ~Manchester Examiner~ says:--“During the last few years the fact
that India is a tea-producing country has become more generally known
in England; but few people know that the finest Indian teas are more
expensive than the best of Chinese growth, and, that the average price
of the tea grown in India is higher than that which comes from the
Flowery Land. Another piece of information given in this book is not
less suggestive; we mean that which assures us that India is capable of
producing as much tea as would meet the wants of great Britain and all
her colonies. But the culture is yet in its infancy. Colonel Money’s
treatise is one of the most complete and exhaustive of the kind we
have ever read. He seems to anticipate all possible difficulties; his
warnings and his counsels embrace every branch of the subject, and
only a practical man could have written them. One would think that a
tea grower of common sense could scarcely make blunders with such an
admirable guide before him; and the commercial side of the enterprise
is discussed in the same careful manner as the agricultural.”

The ~Broad Arrow~ says:--“In this work we have the results of eighteen
years’ experience of a tea planter in India, and the author has so
written it that the beginner will find it invaluable, for he has had
his wants specially in view. It is, so far as we know, the best, as
it is certainly the most practical, book about tea that has been
published.”



_Third Edition. Crown 8vo. Cloth, ~3~s. ~6~d._

  THE ART OF TEA BLENDING.

  A Handbook for the Tea Trade; a Guide to Tea Merchants, Brokers,
  Dealers and Consumers, in the Secret of Successful Tea Mixing.


OPINIONS OF THE PRESS.

The ~Field~ says:--“This is a practical and authentic little text book
on the principles involved in Tea Blending.”

The ~Grocer’s Chronicle~ says:--“The book ought to be in the hands of
every grocer of the United Kingdom.”

The ~Grocer’s Journal~ says:--“We cordially recommend ‘The Art of Tea
Blending’ to our readers as giving useful instruction and guidance.”

~Allen’s Indian Mail~ says:--“The author gives full technical
instructions for the professional tea-blender and tea-taster; and in
doing so, he imparts much information that will be found both valuable
and interesting to the tea-drinking public.”

~Broad Arrow~ says:--“A trader should be able by its aid to make a name
as a teaman, and realise the result which the housewife only needs--a
gentle hint as to the opportunity of exercising real judgment and
correct taste in a matter of such important family interest as tea.”

The ~Grocer~ says:--“This is the third edition of a book which we have
previously noticed with favour, and which has met with considerable
success. Although the art of successful tea-blending is not one
which can well be gleaned from mere book-lore--practical experience
being essential to its acquisition--there are many young beginners,
and possibly also some older hands, who will derive a good deal of
information from the work now under notice. It has evidently been
prepared with much care, and in its way is a very useful handbook.”

The ~Daily Chronicle~ says:--“This capital handbook, which will prove
of great service to merchants, brokers, and all engaged in the tea
trade, has reached a third edition. For consumers we may extract the
information that water for making tea should be soft and pure; it
should be boiled quickly, and used when at the boiling point; the tea
will be at its best in rather less than ten minutes, losing part of its
flavour if allowed to stand longer.”

The ~Grocer’s Gazette~:--“This is a work which has now reached its
third edition, and which fully bears out its claim as an excellent
handbook on the subject. Not only is it a useful book to all
professionally engaged in the trade, but it is also calculated to
educate those who have not had the benefit of a practical experience,
by teaching them how to obtain a knowledge of the different classes of
teas and the proper method of mixing them. To the mature grocer the
work will be of interest, while the uninitiated may by its aid learn
how to select proper teas, please his customer’s palate, and sustain
his reputation by keeping up the character of his mixings.”



_Price ~28~s._

  THE TEA CYCLOPÆDIA.

  A Compilation, by the Editor of the _Indian Tea Gazette_, of
  Information on Tea, Tea Science and Cultivation, Soils and Manures,
  Statistics, &c., with Coloured Plates on Blights. 350 pages.


The ~Grocer~ says:--“One of the most valuable and exhaustive
contributions to tea literature which we remember to have seen.... The
cultivation of the plant in the different districts and provinces, the
selection of soils and manures, and buildings for its manufacture, &c.,
are all ably treated in this work; and as it deals thoroughly with the
scientific, statistical, and domestic branches of the subject, it is a
manual deserving the attention of the tea planter, importer, dealer,
and consumer. The experience of practical growers and cultivators is
here fully narrated, the opinions of the most competent authorities
on disputed points are clearly given and explained; and, in short,
every matter connected with the history of the tea trade, as a growing
industry and a widening channel of commerce, is gone into with a
completeness and precision which leave nothing to be desired.”

The ~American Grocer~ says:--“The Tea Cyclopædia is one we can commend
to our importers, grocers, and dealers, as being the most complete work
of its kind on Indian teas, as well as furnishing innumerable items of
interest to those engaged in the sale of China and Japan teas.”



_Fancy boards, price ~2~s. Cloth, ~3~s. ~6~d._

  WOMAN’S FORTITUDE--

  A Tale of the Indian Mutiny. By Lieut.-Col. EDWARD MONEY.


OPINIONS OF THE PRESS.

~Public Opinion~ says:--“The author has managed to convey the
characteristic tone of garrison talk in a very clear manner.... There
is much good narrative and brilliancy of dialogue.”

The ~Scotsman~ says:--“Written with much spirit ... it will be full of
interest to anybody who cares to know what European life and Society
were in India in the last days of ‘John Company.’”

The ~Daily Chronicle~ says:--“The horrors enacted at Cawnpore during
the Indian Mutiny give a tragic interest to this thrilling tale.”

~Capital and Labour~ says:--“The plot of the tale is carefully
constructed and well worked out, and while the main purpose is always
kept in view, opportunity is taken to depict some of the phases of
Anglo-Indian life, while the characters in the story are cleverly
portrayed, and the attention of the reader is never allowed to flag.”



_Crown 8vo. cloth elegant, bevelled boards, gilt edges. Price ~5~s.
Plain, ~3~s. ~6~d._

  JEMIMA.

  A Story of English Family Life. By ADELAIDE. With Frontispiece.


OPINIONS OF THE PRESS.

The ~Times~ says:--“‘Jemima,’ by Adelaide, is another tale that girls
should care to read, with sufficient proportion of story, and of a more
original type than girls’ books generally are. The humour, of which
there is an unusual proportion for such works, is not, perhaps, of a
very subtle or rich quality, but it is easy and simple, and appropriate
to the characters. Any humour, so long as it is neither vulgar nor
obscene, is surely preferable to the long-drawn melancholy which is too
apt to pervade girls’ books--for what reason we could never understand;
girls are no more naturally prone to sadness than boys.”

The ~Scotsman~ says:--“A better story of its kind than ‘Jemima’ cannot
easily be met with. The book is written with a freshness and exuberant
buoyancy of manner that suit the subject admirably.”

The ~Academy~ says:--“‘Jemima’ is a very natural and charming story of
a very natural and charming little girl. It is exactly what it pretends
to be--‘a story of English family life’--but it has a distinctness of
quality which is by no means common in stories of English family life.”

The ~Daily Chronicle~ says:--“The story of English family life told
by Adelaide, under the title of ‘Jemima,’ is of a much more realistic
character. Lively and amusing throughout, there is also an element
of good sense introduced, which keeps the juvenile escapades within
reasonable bounds, and extracts a lesson even from naughtiness.”


_Crown 8vo. 2 vols. ~10~s. each_.

  VOCABULARY OF THE ENGLISH-MALAY LANGUAGES--

  With Notes. English-Malay Vocal Dialogues. By FRANK A. SWETTENHAM.


_Demy 8vo., cloth, price ~7~s. ~6~d._

  HISTORICAL SKETCHES OF SAVAGE LIFE IN POLYNESIA--

  With Illustrative Clan Songs. By Rev. W. W. GILL, B.A.



_Second and Revised Edition. Crown 8vo. cloth elegant. Price ~2~s._

  SERMONIC FANCY WORK

  On the Figures of our First Acquaintances in Literature. By JOHN PAUL
  RITCHIE.

     I. Little Jack Horner. The Spirit of Self-Satisfaction.
    II. Peter White. How we are led by the Nose.
   III. Humpty Dumpty. The Spirit of Exclusiveness.
    IV. Little Miss Muffit. The Education of Fear.
     V. Jack Spratt and his Wife. The Perfect Law of Liberty.
    VI. Jack and Jill. The Climbing Spirit and its Incumbrances.
   VII. Little Bo-peep. The Recovery of the Lost Sheep.
  VIII. Beauty and the Beast. The Union of the Strong and Beautiful.


OPINIONS OF THE PRESS ON THE FIRST EDITION.

The ~Literary Churchman~ says:--“Pungent, amusing, and replete with
clever satire.”

The ~Christian~ says:--“In this ingenious and novel experiment, gravity
and mirth go hand in hand. The style is energetic and pointed, and the
matter pregnant and suggestive.”

The ~Sword and Trowel~ says:--“Very clever.”

The ~Nonconformist~ says:--“Under the guise of commentary on texts
from old nursery rhymes and stories, Mr. Ritchie really gives us
some admirable discourses--‘Sermonic Fancy Work’ in very deed. It is
astonishing how, by the help of a slight vein of paradox and a nimble
fancy, he can pass, almost imperceptibly, from mild fun to very sad
earnest, touching not a few of our most ingrained faults in the most
efficient way.”

The ~Scotsman~ says:--“A clever, wholesome, readable little book.”

The ~Homilist~ says:--“The sermons are really good. They have satire,
but it is satire which consumes religious rubbish and nonsense. They
have fun and humour, but you are made to laugh in order that you may
think with more vigour and seriousness.”

The ~Freeman~ says:--“The ‘Familiar Texts’ are the old nursery rhymes
treated homiletically. In the styles adopted we fancy we can trace
resemblances to those of some of the popular preachers of our day. The
wit is not without wisdom. The satire is not destitute of sense. It
is the sort of book that a reader with any humour in him will find it
difficult to lay down before he has read it right through.”

~Capital and Labour~ says:--“A droll book and yet containing much
quaint wisdom in searching out and applying principles of truth and
common sense.... As a whole, and considering its healthy tone and
practical scope, we heartily commend this handsome little volume. It is
a fine specimen of the combined arts of the typographer and bookbinder,
and its attractive exterior ought to draw many readers, who will then
be charmed with the contents and with the unconventional method of
treatment.”



_Crown 8vo. Cloth elegant, gilt edges, price ~5~s. Plain, ~4~s. ~6~d._

  WON BY WAITING.

  A Story of Home Life in France and England. By EDNA LYALL. With
  Frontispiece by FRANK MURRAY.

OPINIONS OF THE PRESS.

The ~Daily News~ says:--“The book is full of promise, the story
soon deepens into real interest and develops considerable power of
construction and character drawing.”

The ~Spectator~ says:--“The characters are drawn with considerable
skill, with force, and without exaggeration.”

The ~Academy~ says:--“The Dean’s daughters are perfectly real
characters--the learned Cornelia especially;--the little impulsive
French heroine, who endures their cold hospitality and at last wins
their affection, is thoroughly charming; while throughout the book
there runs a golden thread of pure brotherly and sisterly love, which
pleasantly reminds us that the making and marring of marriage is not,
after all, the sum total of real life.”

The ~Freeman~ says:--“A very pleasing and well-written tale: full of
graphic descriptions of French and English life, with incidents and
characters well sustained. A book with such pleasant reading, and with
such a healthy tone and influence, is a great boon to the young people
in our families.”


_Cloth elegant, ~5~s._

  POEMS, DOMESTIC AND MISCELLANEOUS.

  By JAMES GILES.

The ~Sheffield Independent~ says:--“Very pretty poems, full of a dainty
and airy melody. It is beautifully got up.”

~Public Opinion~ says:--“Mr. Giles has evidently a true poetical
instinct.”

The ~Literary World~ says:--“Full of gentle human feeling, domestic
tenderness, and patient submissive thinking.”


_8vo. cloth. Price ~7~s. ~6~d._

  OLD CEYLON.

  Sketches of Ceylon Life in the Olden Time. By JOHN CAPPER. With
  Illustrations by Ceylon Artists.

“Readable and entertaining sketches.”



_New and Enlarged Edition. Crown 8vo. cloth elegant. Price ~7~s. ~6~d._

  SEATS AND SADDLES,

  Bits and Bitting, Draught and Harness, and the Prevention and Cure of
  Restiveness in Horses. By Major FRANCIS DWYER.

OPINIONS OF THE PRESS ON THE NEW EDITION.

~Bell’s Life~ says:--“The work which Major Dwyer has so successfully
carried through the press in two former editions is, for the third
time, presented to the public in a new and enlarged form. In all the
details of horse management the author is perfectly at home, and the
practical way in which he deals with his subject cannot fail to be
appreciated by equestrians or those who keep studs. The first portion
of the volume is devoted to a lengthy dissertation on the all-important
subject of Seats and Saddles. These chapters afford much valuable
information gained by a careful study, not only of the framework of
the animal considered from a mechanical point of view, but also of the
influence of the saddle in its relation to the seat of the rider.... We
assure our readers that the whole contents of the book are well worth
perusal. It may be well, however, to mention that the question of bits
and bitting is thoroughly considered, while Part III. is taken up with
remarks on the true principles which should be observed in matters of
draught and harness. The concluding portion of the book deals with that
worst of all vices in the horse, restiveness, its prevention and cure.”

The ~Saturday Review~ says:--“It is a book which we should recommend to
the notice of young cavalry officers.”


_Limp cloth, plain, ~1~s. Cloth gilt, gilt or red edges, ~1~s. ~6~d._

  THE LIVING EPISTLE;

  Or, The Influence of Christian Character.

The ~Christian~ says:--“Unflinching in its loyalty to the highest of
all standards, simple in its delineation of what Christian character
should be, earnest in its appeals to the heart and to common sense,
this little book brings to its readers a draught of clear, pure air,
and ought to send them on their way invigorated and quickened in their
desires after holiness.”

The ~Freeman~ says:--“A really ingenious and beautiful exposition of
the inspired description of Christian life. The volume is from the pen
of the late Dr. Jenkyn, formerly of Coward College, and is worthy of a
place by the side of other works we owe him.”



_In Large Crown 8vo. cloth elegant, bevelled boards. Price ~14~s._

  PERTHSHIRE IN BYGONE DAYS--

  One Hundred Biographical Essays. By P. R. DRUMMOND, F.S.A.

OPINIONS OF THE PRESS.

The ~Nonconformist~ says:--“The volume is simply full of the raciest
material, on the whole well laid out, and cannot fail to prove of
interest to many beyond the circle of Perthshire men into whose hand it
may have the good fortune to come. Mr. Drummond had no purpose to serve
in writing the book beyond giving vent to his wide knowledge and his
love of the subject. He was a bookseller in Perth, and it is evident
that to nothing in literature or in human life was he indifferent. All
the notables he knew; and he treasured up _ana_ year by year simply
because it fell in with his tastes and enjoyments to do so.... The book
is full of delicious morsels.”

The ~Athenæum~ says:--“It contains a great deal of sound sense, and
many amusing stories.”


_Price ~1~d. Stiff Paper, ~2~d. With Gold Bead, head and foot, ~4~d.
Oak Bead, ~6~d. Handsomely Framed, ~5~s. ~6~d._

  THE PEOPLE’S POLITICAL ALMANACK,

  A Sheet Calendar, edited by JOHN NOBLE, of the London and Counties
  Liberal Union.

The ~Daily News~ says:--“It is ornamented with an excellent engraving
of the Victoria Tower. It should hang in every Liberal Club in the
Kingdom.”

100 Copies, post free, 8s.; 500 Copies, post free, £1 15s.; 1,000
Copies, post free, £3. If 1,000 copies are taken, the names of the
Officers of Local Societies and their Branches are printed at foot free
of charge.


_Price ~3~s. ~6~d._

  SACHS’S GERMAN GRAMMAR--

  A Complete Grammar of Pure Modern High-German. A New and Practical
  Method of Learning the German Language. By H. SACHS.

The ~Daily Telegraph~ says:--“A complete introduction to pure modern
High-German on true principles.”



_Price ~6~d. Cloth, ~1~s._

  STRAY HINTS ON LAWN TENNIS,

  With latest Club Rules. By NEMO.

  Including Directions for making the Best and Cheapest Poles yet
  invented--without Guy Ropes.


_Price ~7~s. ~6~d._

  WHITTINGHAM’S SKELETON TELEGRAPH CODE--

  A Secret Expansive Code for ordinary Business Purposes--with Code
  Words representing Pounds Sterling, Numbers, every Date in the
  Year, Weights, &c., in addition to Words and Spaces available for
  4,500 Special Messages--all Code Words revised under the latest
  International Regulations.

The ~Standard~ says:--“Appears to us to answer admirably the purpose
for which composed.”


_8vo., roan, price ~5~s. The “1881” Code, ~12~s. ~6~d._

  THE SOCIAL CODE:

  Compiled for Family Use in Telegraphing to Friends abroad.

“The best of its kind.”


_Price ~5~s._

  THE PREMIUM CALCULATOR--

  An Office Manual, with Cut Index, for the use of Underwriters, &c. By
  C. MCKAY SMITH.


  DR. AGER’S TELEGRAPH CODES.

  General Commercial, £2 15s.; Standard, 100,000 Words, £5 5s.;
  Shipping, 21s.; Skeleton, 16,000 Words and Spaces, 12s. 6d.; Corn
  Merchants, £1 11s. 6d.



W. B. WHITTINGHAM & CO.,

MANUFACTURING AND

EXPORT STATIONERS & PRINTERS,

Booksellers and Publishers, and Account Book Makers,

91, GRACECHURCH STREET, LONDON.

(Works: 4, White Hart Court, Bishopsgate.)


  STATIONERY DEPARTMENT.

All Articles of COMMERCIAL STATIONERY on Export terms. Merchants,
Bankers, and Companies’ LEDGERS and ACCOUNT BOOKS, Parchment, Vellum,
Loan, and all kinds of Printing and Writing Papers, Wholesale and
Retail. Letter and Note Paper and Envelopes of every description.
DESIGNS supplied for Bills of Exchange, Bills of Lading, Scrip
Certificates and Bonds, Promissory Notes, and Bankers’ Cheques. Card
Plates Engraved, Dies Cut, and Crest or Monogram Stamped.

  PRINTING DEPARTMENT.

Every Branch of LETTERPRESS PRINTING in Latest Styles from New Types.
Business Circulars, Prices Current, and Mercantile Printing of all
kinds. Books, Periodicals, and Pamphlets, Printed to Estimate.

  LITHOGRAPHIC PRINTING.

Lithographic Letters, Fac-simile or otherwise, Bankers and Merchants’
Circulars, Auctioneer’s Estate Plans, Policies and Calendars Designed
and Produced in every variety of Style.

  BOOK DEPARTMENT.

Books sent _Post Free_, on receipt of the Published Price, to any part
of England, the United States, Egypt, British India, and Continental
Countries within the Postal Union.

  NEWSPAPERS and PERIODICALS.

Despatched with regularity to any part of the World.


  _Authorised AGENTS for the Supply of the Bills of Lading of the
  “Peninsular and Oriental,” the “British India,” the “Messageries
  Maritimes,” and other steam Navigation Companies._


ACCOUNT BOOKS.

Messrs. W. B. Whittingham & Co. invite attention to the unsurpassed
Strength and Finish of the Ledgers which they Manufacture. The same
Workmanship and Materials which for more than Half-a-Century have
established the reputation of these Account Books are guaranteed.


W. B. WHITTINGHAM & CO., 91, Gracechurch-st., London.



Transcriber's Note


The following apparent errors have been corrected:

p. 14 "it well" changed to "it will"

p. 94 "wonld" changed to "would"

p. 97 "fifteeen" changed to "fifteen"

p. 107 "_a_." changed to "_a_",

p. 139 "leafs" changed to "leaves"

p. 177 (note) "was--" changed to "was:--"

p. 187 "similiar" changed to "similar"

p. 203 "considered." changed to "considered,"

p. 243 "litle" changed to "little"

p. 264 "viz," changed to "viz.,"

p. 264 "seive-mesh" changed to "sieve-mesh"

p. 273 "loose" changed to "lose"

p. 273 "tare." changed to "tare,"

p. 280 "he gross" changed to "the gross"

p. 299 "Hazareebagh" changed to "Hazareebaugh"

p. 300 "Transport, chap" changed to "Transport, chap."

p. 301 "title" changed to "-- title"

Advertisement "Chapers" changed to "Chapters"

Advertisement "PERIODICALS" changed to "PERIODICALS."


Inconsistent spelling, hyphenation, italics and punctuation have
otherwise been kept as printed.


Tables have been refactored for accessibility.





*** End of this Doctrine Publishing Corporation Digital Book "The Cultivation and Manufacture of Tea" ***

Doctrine Publishing Corporation provides digitized public domain materials.
Public domain books belong to the public and we are merely their custodians.
This effort is time consuming and expensive, so in order to keep providing
this resource, we have taken steps to prevent abuse by commercial parties,
including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Doctrine Publishing
Corporation's ISYS search for use by individuals, and we request that you
use these files for personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort
to Doctrine Publishing's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a
large amount of text is helpful, please contact us. We encourage the use of
public domain materials for these purposes and may be able to help.

+ Keep it legal -  Whatever your use, remember that you are responsible for
ensuring that what you are doing is legal. Do not assume that just because
we believe a book is in the public domain for users in the United States,
that the work is also in the public domain for users in other countries.
Whether a book is still in copyright varies from country to country, and we
can't offer guidance on whether any specific use of any specific book is
allowed. Please do not assume that a book's appearance in Doctrine Publishing
ISYS search  means it can be used in any manner anywhere in the world.
Copyright infringement liability can be quite severe.

About ISYS® Search Software
Established in 1988, ISYS Search Software is a global supplier of enterprise
search solutions for business and government.  The company's award-winning
software suite offers a broad range of search, navigation and discovery
solutions for desktop search, intranet search, SharePoint search and embedded
search applications.  ISYS has been deployed by thousands of organizations
operating in a variety of industries, including government, legal, law
enforcement, financial services, healthcare and recruitment.



Home