fT!

THE LIBRARY

OF

THE UNIVERSITY
OF CALIFORNIA

PRESENTED BY

PROF. CHARLES A. KOFOID AND
MRS. PRUDENCE W. KOFOID

THE

ROTHAMSTED EXPERIMENTS

ON THE

GROWTH OF WHEAT, BARLEY,

AND THE

MIXED HERBAGE OF GRASS LAND.

BY

WILLIAM FEEAM, B.Sc. Lond., P.L.S., P.G.S., F.S.S.,

ASSOCIATE OF THE SURVEYORS' INSTITUTION ; CONSULTING BOTANIST
TO THE BRITISH DAIRY FARMERS' ASSOCIATION AND THE
EOYAL COUNTIES AGRICULTURAL SOCIETY ; PROFESSOR
OF NATURAL HISTORY IN THE COLLEGE OF AGRI-
CULTURE, DOWNTON, SALISBURY, AND FOR-
MERLY PROFESSOR OF NATURAL HISTORY
IN THE EOYAL AGRICULTURAL
COLLEGE, CIRENCESTER.

LONDON :

HORACE COX, " THE FIELD " OFFICE, 346, STKAND, W.C.

1888.

P '

LONDON :
PRINTED BY HORACE COX, 10, WELLINGTON STREET, STRAND, W.C.

TO

SIE JOHN BENNET LA WES, BART., LL.D, F.R.S.,

AND

PEOF. JOSEPH HENEY (HLBEET, M.A., LL.D., F.E.S.,

WHOSE LONG-CONTINUED AND SUCCESSFUL

INVESTIGATIONS IN EVERY DOMAIN OF

AGRICULTURAL INQUIRY HAVE MADE

THE NAME OF THE ROTHAMSTED

EXPERIMENTAL STATION

FAMOUS THROUGHOUT

THE WORLD,

THIS VOLUME, RECORDING THE METHODS AND

RESULTS OF SOME OF THEIR RESEARCHES,

IS BY PERMISSION RESPECTFULLY

M368238

PREFACE.

HAVING had occasion to study somewhat closely the work
of the Rothamsted investigators, it occurred to me that
their valuable memoirs, dealing as they so largely do
with actually ascertained results, might be advan-
tageously condensed into the form of a text-book.
Hence arose the present volume which, though it is
concerned with a portion only of the many questions
that have been brought within the range of experimental
inquiry at Rothamsted, yet deals with subjects of first
class importance, equally in their scientific bearing as in
their economic aspect. To the student the discussion of
concrete results should prove at least as useful as the
consideration of abstract assertions, whilst it is possible
that it may be even more suggestive.

In the endeavour to make each subject as far as
possible complete in itself, a certain amount of repetition
has been unavoidable. Many questions, again, which
are but lightly touched upon in these pages, are more
fully dealt with in other of the Rothainsted memoirs
than those which the writer has laid under contribution.
Even as this work is passing through the press a fresh
memoir issuing from Rothamsted throws further light
upon the classical inquiry as to the sources of the
nitrogen of vegetation.

vi Preface.

In several instances the information here given is
brought very nearly down to date. This has been
rendered possible only through the characteristically
kind and ready manner in which Sir John Lawes and
Dr. Gilbert have responded to my requests, and for
which my grateful acknowledgments are tendered.

At the present time, when there has been preferred a
somewhat vague demand that the Government should
undertake the support of agricultural experiment stations,
it seems a fitting moment in which to make more widely
known the nature and results of the splendid work
which private enterprise has with such conspicuous
success maintained at Rothamsted. Whatever merits
this book may possess are due to the illustrious investi-
gators with whose name the volume is associated. Its
faults, whatever they be, are mine alone.

W. FREAM.

COLLEGE OP AGRICULTURE,

DOWNTON, SALISBURY,
January, 1888.

CONTENTS.

PAGE

I. INTRODUCTORY 1

WHEAT.

II. EARLY EXPERIMENTS ON WHEAT 6

1. THE HOLKHAM EXPERIMENTS 6

2. THE RODMERSHAM EXPERIMENTS 12

III. EXPERIMENTS ON THE CONTINUOUS GROWTH OF

WHEAT UPON THE SAME LAND FOR FORTY

YEARS 18

PLAN OF THE EXPERIMENTS 19

INFLUENCE OF SEASON ON THE WHEAT CROP 23
EFFECTS OF THE UNEXHAUSTED RESIDUE OF

MANURES 30

CONTINUOUS GROWTH OF WHEAT UPON THE

SAME LAND FOR FORTY YEARS WITHOUT

MANURE 35

WITH MINERAL MANURES ALONE 38

WITH AMMONIA-SALTS ALONE 41

WITH AMMONIA-SALTS AND MINERAL

MANURES IN ALTERNATING YEARS ... 43

COMPARISON BETWEEN AMMONIA-SALTS AND

NITRATE OF SODA AS SOURCES OF NITROGEN 45

Contents.

PAGE

CONTINUOUS GROWTH OF WHEAT UPON THE
SAME LAND FOB FORTY YEARS WITH FARM-
YARD MANURE SUPPLIED EACH YEAR ... 49
QUANTITY OF AMMONIA REQUIRED TO PRO-
DUCE AN INCREASE OF ONE BUSHEL PER

ACRE IN THE WHEAT CROP 54

THE B.OTHAMSTED EXPERIMENTS ON THE
CONTINUOUS GROWTH OF WHEAT COMPARED
WITH THOSE AT WOBURN, HOLKHAM, AND
EODMERSHAM 56

SUMMARY OF THE RESULTS OF FORTY YEARS'
EXPERIMENTS ON THE CONTINUOUS GROWTH

OF WHEAT UPON THE SAME LAND 58

IV. INFLUENCE OF CLIMATE ON THE CULTIVATION

OF WHEAT 32

V. THE HOME PRODUCE, IMPORTS, AND CONSUMP-
TION OF WHEAT 74

BAELEY.

VI. EXPERIMENTS ON THE CONTINUOUS GROWTH OF
BARLEY UPON THE SAME LAND FOR TWENTY

YEARS 87

CHARACTERISTICS OF THE BEST AND WORST

SEASONS OF THE TWF.NTY YEARS 92

AVERAGE YIELD OF BARLEY PER ACRE PER

ANNUM FOR EACH DESCRIPTION OF MANURE 96
QUANTITY OF AMMONIA REQUIRED TO PRO-
DUCE AN INCREASE OF ONE BUSHEL PER
ACRE IN THE BARLEY CROP 103

EFFECTS OF THE UNEXHAUSTED RESIDUE OF

MANURES 106

THE EXPERIMENTS ON THE CONTINUOUS
GROWTH OF BARLEY COMPARED WITH
OTHER SIMILAR EXPERIMENTS . Ill

Contents. ix

SUMMARY OF THE RESULTS OP TWENTY
YEARS' EXPERIMENTS ON THE CONTINUOUS

GROWTH OF BARLEY UPON THE SAME LAND 117

VII. EXPERIMENTS ON THE CONTINUOUS G-ROWTH OF
BARLEY UPOM THE SAME LAND FOR THIRTY-
TWO YEARS 122

CHARACTERISTICS OF THE BEST AND WORST

SEASONS OF THE THIRTY-TWO YEARS ... 126

VARIATIONS IN THE CHEMICAL COMPOSITION

OF THE BARLEY CROP 128

THE HOME PRODUCE AND THE IMPORTS OF

BARLEY 130

CULTIVATION OF WHEAT AND OF BARLEY

COMPARED 132

GRASS.

VIII. AGRICULTURAL RESULTS OF EXPERIMENTS ON
THE MIXED HERBAGE OF PERMANENT

MEADOW 133

PLAN OF THE CONTINUOUS EXPERIMENTS ON

MEADOW HERBAGE 138

RESULTS OBTAINED UPON THE VARIOUSLY

MANURED PLOTS OF MEADOW LAND ... 143

1. Without manure : Plots 3 and 12 ... 144

2. Ammonia-salts alone : Plot 5 145

3. Nitrate of soda alone : Plots 15 and 17 146

4. Mixed mineral manure alone : Plot 7 149

5. Superphosphate of lime alone: Plot

4-1 153

6. Mixed mineral manure, with and with-

out potash : Plots 7 and 8 155

Effect of mixtures of nitrogenous and
mineral manures ... l'r>8

6

Content*.

PAGE

7. 400Z6. ammonia-salts, with mixed

mineral manure containing potash:
Plot 9 159

8. 400Z6. ammonia-salts, with mixed

mineral manure containing potash,

and 2000Z6. cut wheat straw : Plot 13 161

9. SQOlb. ammonia-salts, with mived

mineral manure containing potash :
Plots lla and lib 164

10. 550Z6. nitrate of soda, with mixed

mineral manure containing potash :
PlotU ... 167

11. 275Z6. nitrate of soda, with mixed

mineral manure containing potash :
PlotU ... 170

12. 400Z6. ammonia -salts, and super-

phosphate of lime : Plot 46 172

13. 400Z6. ammonia -salts, and mixed

mineral manure, with and without
potash: Plots 9 and 10 ... 175

14. Mixed mineral manure alone 7 years ;

succeeding ammonia - salts alone

13 years: Plot 6 ... 178

15. Equal nitrogen and equal potash, in

nitrate of soda and sulphate ofpot<i«Ji,
and in nitrate of potash ; in each case
with superphosphate of lime : Plots
19 and 20 181

16. Mixture supplying the ash-constituents,

and the nitrogen, of 1 ton of hay:
Plot 18 ... 184

17. Farmyard manure alone, and with

ammonite-sofa /// addition : Plots 2

and 1 189

FATE OF THE NITROGEN IN THE SOIL... 194

Contents. xi

PAGE

FATE OP THE FERTILISING MINERALS IN

THE SOIL 202

THE SECOND CROPS 206

INFLUENCE OF SEASON ON THE PRODUCE OF

HAY 209

IX. BOTANICAL RESULTS OF EXPERIMENTS ON THK

MIXED HERBAGE OF PERMANENT MEADOW . 218

INDEX 227

ERRATA.

Page 75, last line. — For " estimates" read " estimate.'
Last line but one.— For " 1787 " read " 1887.'

THE

ROTHAMSTED EXPERIMENTS.

I.-INTRODUCTORY.

THE history of the oldest agricultural experimental station in
the world is soon told. Sir (then Mr.) John Bennet Lawes,
Bart., entered into possession of his hereditary property at
Eothamsted, Hertfordshire, in 1834, and soon afterwards
commenced that unique series of experimental investigations
which are still in progress, and for which, through the muni-
ficence of their originator, provision has been made for the
continuation in perpetuity. In June, 1843, Dr. J. H. Gilbert
became associated with Mr. Lawes, and undertook the direc-
tion of the chemical laboratory ; at this time, also, the more
systematic field experiments were commenced. The general
scope and plan of these field experiments is stated in the fol-
lowing words : To grow some of the more important crops
of rotation, each separately, year after year, for many years
in succession on the same land, without manure, with farm-
yard manure, and with a great variety of chemical manures ;
the same description of manure being, as a rule, applied year
after year on the same plot. Experiments on an actual course
of rotation, without manure, and with different manures, have
also been made. In this way field experiments have been
conducted with wheat, barley, oats, beans, clover, various
leguminous plants, turnips, sugar beet, mangel wurzel,
potatoes, and permanent grass. Comparative experiments

The Eothamsted Experiments.

with different manures have also been made, on other descrip-
tions of soil, at Holkham and Eodmersham. Samples of the
soils, of the crops, and of different portions of the crops are
taken to the laboratory and subjected to exhaustive chemical
analysis, and it is on the results of such examinations that
the conclusions are based. Other inquiries have embraced
the investigation of soils, especially with reference to the per-
centages of nitrogen they contain at different depths, the
chemical constituents in rain and in drainage waters, the
amount of water transpired by plants, and experiments on
the assimilation of free nitrogen by vegetation.

Equally important experiments with farm animals were
commenced in 1847. Among the points which have been
investigated are : (1) The amount of food, and of its several
constituents, consumed in relation to a given live-weight
of animal within a given time. (2) The amount of food,
and of its several constituents, consumed to produce a given
amount of increase in live-weight. (3) The proportion, and
relative development, of the different organs or parts of
different animals. (4) The proximate and ultimate com-
position of the animals in different conditions as to age and
fatness, and the probable composition of their increase in
live- weight during the fattening process. (5) The com-
position of the solid and liquid excreta (the manure) in
relation to that of the food consumed. (6) The loss or ex-
penditure of constituents by respiration and the cutaneous
exhalations — that is, in the mere sustenance of the living meat-
and-manure-making machine. Oxen, sheep, and pigs have
been the subjects of the experiments, and the results obtained
have further been of use in furnishing data for the con-
sideration of the following questions : (1) The characteristic
demands of the animal body (for nitrogenous or non-
nitrogenous constituents of food) in the exercise of muscular
power. (2) The sources in the food of the fat produced in
the animal body. (3) The comparative characters of animal
and vegetable food in human dietaries.

Introductory* 3

Supplementary investigations have included a somewhat
exhaustive inquiry into the application of town sewage to
different crops, and especially to grass ; the chemistry of the
malting process, the loss of food constituents during its
progress, and the comparative feeding value of barley and
malt ; the changes and losses which food crops undergo in
the process of conversion into silage ; and the feeding value
of different kinds of silage when supplied to fattening oxen
and to milking cows.

The first original paper that emanated from Eothamsted
appeared in the year 1847, and from then to the middle of
1887 — that is, during a period of forty years — there were
published as many as one hundred and four separate papers
or memoirs. A rough classification of these papers leads to
some such arrangement of subjects as the following: (1)
Experiments on the continuous growth of wheat and of
barley on the same land. (2) Agricultural, botanical, and
chemical results of experiments on the mixed herbage of per-
manent meadow conducted for more than twenty years in
succession on the same land. (3) The amount and composi-
tion of rain and drainage water. (4) The application, distri-
bution, and influence of manures, and the valuation of unex-
hausted manures. (5) The fertility and exhaustion of soils.
(6) The combined nitrogen in soils and subsoils. (7) The
sources of the nitrogen of vegetation. (8) Nitrification. (9)
Clover sickness and the growth of clover by different manures.
(10) The evaporation of water from plants. (11) The botany
and chemistry of fairy rings. (12) The home produce,
imports, and consumption of wheat, and the composition of
wheat grain, mill products, and bread. (13) Agricultural
chemistry, dealing with Liebig's theories, with turnip culture,
and with the feeding of oxen, sheep, and pigs. (14) The
utilisation of town sewage. (15) The relation of temperature
to plant-growth.

It is when the inquirer, who wishes to obtain an adequate
idea and a comprehensive view of the Eothamsted experi-

B 2

4 The Eothamsted Experiments.

merits, seeks to study the published results, that the initial
difficulty is met with. For where shall they be sought ? The
answer is, in the journals of a dozen different societies, in
magazines, in parliamentary papers, and some of them in the
form of separate memoirs. About three dozen papers have
been communicated to the Journal of the Royal Agricultural
Society. Numerous other papers have appeared in the
Journal of the Eoyal Horticultural Society, the Journal-1 of
the Chemical Society, the Journal of the Linnean Society, the
Journal of the Statistical Society, the Journal of the Society
of Arts, the Eeport of the British Association, the Proceed-
ings of the Institution of Civil Engineers, and the Proceed-
ings of the Eoyal Society. Others, again, must be sought in
the Archives des Sciences physiques et naturelles, in the
Philosophical Magazine, in the Journal of Anatomy and
Physiology, and in the Edinburgh Veterinary Eeview. About
ten papers were published as independent pamphlets, whilst
between 700 and 800 of the large quarto pages of the Philo-
sophical Transactions, issued by the Eoyal Society, are occu-
pied with reports of the classical researches on the sources of
the nitrogen of vegetation, the composition of some of the
animals fed and slaughtered as human food, and the mixed
herbage of permanent meadow.

The difficulty experienced by the Eothamsted investigators
in keeping their current publications at all abreast of their
researches effectually precludes the hope of our getting from
Eothamsted itself any connected history of the work of the
last half century Such a history would be gladly welcomed
by progressive agriculturists in all parts of the world, for it
is now impossible to obtain some of the original memoirs.
Mr. Warington, of the Eothamsted Laboratory, has tersely
observed that " the best English work on agricultural
chemistry exists as yet only in detached papers." Nearly
all these " detached papers " have emanated from Eotham-
sted, and the object of the following pages is to pre-
sent to the reader in a succinct and collected form a

Introductory.

description of the Eothamsted experiments on wheat, barley,
and grass.

The first part of the book is concerned with experiments
upon the growth of wheat, and concludes with a discussion
of some important economical questions. The second part is
occupied with experiments upon the growth of barley. The
third and last part deals with experiments upon the mixed
herbage of permanent meadow. In each part there occur
many incidental references to the natural resources of the
soil, and to the nature of the loss occasioned by the drainage
waters, whilst much light is thrown upon the celebrated
problem as to the sources of the nitrogen of vegetation. The
numerical tables, most of which are quoted in extenso from
the original memoirs, are invaluable, and are specially com-
mended to the attention of students.

These experiments are unique, and their fame has spread
wherever modern agriculture is practised. To the cultivator
of the soil they afford the best examples of the method of
conducting such investigations, whilst to the agricultural
student on the one hand, and to the chemist and physiologist,
the statistician and economist, on the other, a study of them
is indispensable.

The Rothamsted Experiments.

II.-EARLY EXPERIMENTS ON WHEAT.

1. THE HOLKHAM EXPERIMENTS.

OF the experiments on the growth of wheat, by far the most
noteworthy are those concerned with the growth of this
cereal for forty years in succession on the same land. But
there are two earlier reports which possess a special value,
inasmuch as they very well serve to indicate the type of
experiment which might be advantageously repeated either
by agricultural societies or by private enterprise. The first
of these, published in the R.A.S. Journal, 1855, bears the
title " Report to the Right Hon. the Earl of Leicester, on the
experiments conducted by Mr. Keary, on the growth of
wheat upon the same land for four successive years, at
Holkham Park Farm, Norfolk."

With regard to the terms " light " and " heavy," as
applied to soils, it is pointed out in this report that, aa
applied to a surface soil, they afford a very imperfect indica-
tion of the probable native resources, and consequently of
the capabilities of growth without deterioration, of the
respective soils. Here is suggested a vein of inquiry which
has since been profitably worked at Rothamsted. The soil on
which the experiments at Holkham were made is described as
a light, thin, and rather shallow brown sand loam, resting
upon an excellent marl containing a large quantity of
calcareous matter. In such a surface soil there would be
combined the easily working qualities and the power of
rapidly yielding up manurial matter of the so-called " light "
soils, whilst in its subsoil there would be much of the native

Early Experiments on Wheat at Holkham. 7

resource of constituents, and probably the power of absorp-
tion or retention of mammal matter also, of the so-called
heavy soils. But it was of the greatest interest to ascertain
by actual experiment how far those chemical substances,
which are employed with success for the increased growth of
wheat upon heavy soils, could be used with advantage upon
those of different descriptions.

Previous to the introduction of the four-course system by
the late Earl of Leicester, the soil upon which the experiments
were made had been considered too light to grow wheat.
The land had for some years previous to the experiments been
farmed under that system ; it was clayed about twelve years
before the experiments were begun, and the crop immediately
preceding them was white turnips, manured with farmyard
dung and guano, both tops and roots being drawn off the
land. The experimental plots measured half an acre each ;
the manures were as follows, and were all sown in the
autumn, except plot 4, which was sown in spring :

Plot 1. Always unmanured.

Plot 2. Mineral manures alone.

Plot 3. Ammonia-salts alone, sown in the autumn.

Plot 4. Ammonia-salts alone, sown in the spring.

Plot 5. Both the mineral manure and ammonia-salts.

Plot 6. Rape cake.

Plot 7. Farmyard manure.

1. The unmanured plot, when once exhausted of the
accumulations derived from the more recent previous
manuring, would, of course, show the productive capability of
the soil in a comparatively normal state, in conjunction with
that of the annual climatic yield of the atmospheric elements
of growth ; and the results would provide a standard with
which to compare the produce of the different manures.

2. The mineral manure furnished a liberal supply of the
alkalies (potash and soda), alkaline earths (lime and
magnesia), and phosphoric acid ; and its produce, compared
with that of the other plots, shows whether the result of the

8 The Rothamsted Experiments.

cropping is to reduce the available supplies of such mineral
constituents in the soil below that which is requisite to
obtain the full benefit of the annual atmospheric supply of
carbon and nitrogen, or whether it is the supply in the soil
of the carbon or the nitrogen which is most exhausted.

3. The use of ammonia-salts alone, which provide nitrogen
for the growth of the crop, shows whether or not the latent
mineral wealth of the soil is more than sufficient for the
annual atmospheric supply of available nitrogen.

4. And the object of sowing one plot with ammonia-salts
in the autumn and another in the spring was to determine
whether it was practically advantageous to sow such soluble
manures in the autumn on so light a soil.

5. The mixture of mineral manures and ammonia-salts
shows, when the results are compared with those of each of
these manures used separately — first, whether the annually
available native mineral supply of the soil, taken together
with that in the manure, was not capable of producing a much
greater amount of growth than was the annual atmospheric
supply of nitrogen ; and, secondly, whether the amount of
nitrogen supplied to the soil, when such a quantity of
ammonia- salts was used alone, was not in excess in propor-
tion to the annually available supply of minerals from the
soil itself.

6. Bape-cake contains a large proportion of carbonaceous
and nitrogenous organic substances, and some mineral
matter ; and the nitrogen supplied in the quantity of it used
was nearly identically the same as, or, perhaps, rather
greater in amount than, that in the ammonia-salts of the
other experiments.

7. The farmyard manure was the product of yards in
which bullocks were fed on turnips with a moderate quantity
of oil cake. In this manure there would be added to the soil
every year a larger supply of every constituent than was
contained in the increased wheat crop grown.

The results are summarised in a series of tables, in

Early Experiments on Wheat at HolWiam. 9

which are stated the dressings of manure per acre on each
plot, and the yields of dressed corn, offal corn, and straw in
each of the four seasons, 1851-2-3-4.

In every case a larger produce, by 14 to 20 bushels, or
even more, was obtained by the same manure in the first
year than in the average of the following years. This result
speaks well for the previous " condition " of the land, and it
is also very instructive as showing how useless, for the
purposes of any general conclusions, are experiments with
manures conducted over a single season only. It is in fact
not until some of the elements of fertility, the due proportion
of which to the others is indicated in the term " condition,"
have been removed from the soil by the crop, that any safe
deductions can be formed from the results of experiments
with manures.

Plot 1, unman ured, gave 39| bushels per acre of dressed
corn the first year, 15J the second, 21^ the third, and 16|
the fourth, the average of the lour years being 23J bushels,
and that of the last three years nearly 18 bushels, which
latter amount is nearly 22 bushels less than was obtained
on the same plot in the first year.

Plot 2, manured with salts of potash, soda, and magnesia,
and superphosphate of lime, gave in the four years respec-
tively 34J, 19, 19|, and 18^ bushels of dressed corn per
acre, the average of the last three years being about 15
bushels less than the yield of the first year. A comparison
of the results on plots 1 and 2 shows that there was actually
rather more corn obtained without manure than with the
minerals, the tendency of the latter being to increase the
growth of straw, of which, taking the last three years
together, there was about half a ton more obtained by
means of the minerals.

Plot 3 received in the autumn, and plot 4 in the spring,
2001b. sulphate of ammonia, and 2001b. muriate of ammonia
per acre. Taking the four years together, there is a
difference of less than 2 bushels between the produce of

10 The Rothamsted Experiments.

the two plots, it being, however, rather in favour of the
autumn-sown manure. In straw there is a difference of only
81b. per acre in favour of the spring-sown manure. Upon
the whole, then, the results are in favour of sowing these
soluble manures in the autumn, even in so light a soil. On
the autumn-sown plot there was a fall in the produce of 18
bushels, and on the spring-sown plot of 14J bushels per acre,
from the first year to the average of the last three years.

Plot 5 received the same manure as plot 2, and as plot
3 or 4. The result was to produce over the four years from
53 to 54 bushels of corn more than was yielded by the
minerals alone, besides a larger quantity of straw. This,
therefore, was an annual average of 13 to 14 bushels of corn,
and an equivalent of straw, due to the use of the ammonia
salts. And since there was in the four years about twenty
bushels more increase by the mixture of both minerals and
ammonia-salts than by the latter alone, it is obvious that
the minerals of this last 20 bushels of the total 53 of increase
were derived from the mineral manures employed.

On plot 7, 14 tons per acre of farmyard manure were
annually applied, and gave an average annual increase of
about 10 \ bushels of corn and 13001b. of straw — less by nearly
3 bushels of corn and about 1501b. of straw than the produce
obtained by the rape-cake (plot 6). As the other plots demon-
strated, in their results, that neither minerals alone, nor car-
bonaceous organic matter (in the rape cake), had any influence
in the increase of the crop, but that wherever there was a,
supply of nitrogen in the manure there was always a very con-
siderable increase, it is concluded that it was the amount of
nitrogen liberated from the dung, in a form assimilable by
the plants, which determined the limit to the increase of
produce obtained by its use. Though the amount of nitrogen
contained in the increased produce of wheat was never equal
to that supplied in the manure, it is to be remembered that,
besides any liability to loss by drainage, to which all manurial
ingredients might be subject, the nitrogen, in several of its-

Early Experiments on Wheat at HolJcham. 11

forms of combination, is also volatile, and might be exhaled
into the atmosphere and so lost. This is not the case with
the mineral constituents of manure.

Upon the whole a careful study of these experiments
proves :

1. That the soil, even with the most unusual and very
exhausting process of carrying off the land the total grain
and straw of several successive corn crops, after a root crop
which had also been drawn from the land, still contained a
larger annual available supply of minerals than the annual
natural supplies of other constituents, nitrogen or carbon,
were adequate to turn to account.

2. That the excess of the annual supply of minerals in the
soil over that required to appropriate the natural resources
of nitrogen is proved, by the effects of ammonia-salts alone,
to have been equal to the further growth, during four years,
of about 32 bushels of wheat, or an average of about
8 bushels per annum.

3. That beyond the increased annual produce which the
supply of minerals in the soil was adequate to provide
when nitrogen was not wanting, the average capabi-
lities of the climate were competent for the maturing of a
still greater produce, if additional minerals as well as the
amrnonia-salts were provided; and, in that case, from once
and a half to twice as much corn was grown as the natural
supplies of nitrogen, even with a most liberal supply of
minerals, were sufficient to produce.

4. That carbonaceous organic matters (such as are contained
in rape cake and farmyard manure) are of themselves of little
or no effect in increasing the growth of wheat.

The concluding words of this paper, written more than
thirty years ago, are still of interest : There cannot be a
doubt of the legitimacy of the inference from these and
other experiments that, provided the land receive in a course
of years a due share of the home manures derived from
feeding of horses and other stock on the farm, the mineral

12 The Rothamsted Experiments.

supplies of the soil will be amply sufficient to sustain an
increased and even repeated growth of corn, by means of
nitrogenous artificial manures, considerably beyond that
which is recognised by the leases or the current practices
of the day ; and a further assurance that the necessary
minerals are not likely to become deficient, under the
judicious adoption of such an increased growth of corn,
is to be found in the fact that there are few really large
sources of nitrogenous manures which do not, at the same
time, bring upon the land a considerable amount of some of
the more important minerals also.

2.— THE RODMERSHAM EXPERIMENTS.

One other series of experiments, besides those at Holkham,
was carried out away from Rothamsted, and is of equal
interest with the Holkham series in suggesting the nature
and character of the field investigations which might be
conveniently prosecuted in various localities, either by private
enterprise or under the auspices of an agricultural society.
The " Report of experiments made at Rodmersham, Kent,
on the growth of wheat by different descriptions of manure
for several years in succession on the same land," appeared
in the R.A.S. Journal, 1862. The scheme of these experi-
ments was similar in essential details to that of the Holkham
Park Farm experiments, and both were modelled upon those
which then were, and still are, in progress at Rothamsted on
the continuous growth of wheat upon the same land. The
Rodmersham experiments arose in this wise : Sir John M.
Tylden, the president of an agricultural club in the neighbour-
hood of Sittingbourne, Kent, induced the members to visit
Rothamsted ; and the consequence was that they undertook
to conduct, at their own expense, a series of experiments on
the growth of wheat, the results of which might be compared
with those of the experiments at Holkham and Rothamsted.

Early Experiments on Wheat at Bodmersham. 13

Accordingly, a field of 3J acres at Rodmersham, about
three and a half miles from Sittingboume, was divided into
seven plots of half an acre each, and the experiments were
placed in charge of Mr. George Eley, of Tong, the secretary of
the club. The soil of the experimental plots is described as a
mixed clay, upon a chalk subsoil lying from 4ft. to 6ft. below
the surface. The previous cropping had been as follows :
In 1853, turnips, dressed with 2cwt. guano and 3cwt. super-
phosphate of lime per acre, and the whole of the crop fed on
the land ; in 1854, barley, and a good dressing of London
dung for beans in 1855, this being the usual preparation for
wheat in that locality. The land was, therefore, in a well-
cultivated and fertile state, and, as the results showed,
in higher condition than was desirable when the object was
to determine the character of the exhaustion, and therefore
the character of the manures required for the crop in that
particular soil, under the ordinary system of cropping and
management adopted. The action of the different manures
was, nevertheless, sufficiently characteristic after the first
crop of wheat had been taken.

The manures were obtained from the bulks used in the
Rothamsted experiments. The arrangement and quantities
per acre were as follows :

Plot 1. Unmanured.

Plot 2. Mixed mineral manure, viz., 3001b. sulphate of
potash, 2001b. sulphate of soda, lOOlb. sulphate of magnesia.
3501b. superphosphate of lime (consisting of 2001b. bone
ash and 1501b. sulphuric acid of sp. gr. 1*7).

Plot 3. Ammonia- salts, comprising 2001b. sulphate of
ammonia and 2001b. muriate (chloride) of ammonia.

Plot 4. Ammonia- salts (as Plot 3) and mixed mineral
manure (as Plot 2).

Plot 5. 5401b. Peruvian guano.

Plot 6. 20001b. rape cake.

Plot 7. 14 tons farmyard manure.

The dressings were applied annually for the first three

14 The Rothamsted Experiments.

years, and the arrangement was the same for the fourth
year, save that on Plots 2 and 4 the quantities of sulphate
of potash were reduced from 3001b. to 2001b., and of
sulphate of soda from 2001b. to lOOlb. per acre. In the
fifth and sixth seasons the crop was grown without any
fresh application of manure.

The object of experiment 1 was to ascertain the state of
productiveness of the land without any manure, and so to
provide a standard by which to compare the effects of the
different manures. Experiments 2, 3, 4, 6, and 7 were to
determine whether a specially mineral, nitrogenous, or
carbonaceous manure, or some combination of these, is the
most effective, and experiment 5 would indicate whether
increase of crop can be profitably obtained by a cheap
" artificial" manure containing a large proportion both of
nitrogen and phosphates.

The results of the six years' experiments (1856-61) as
collected in a series of tables may now be discussed. It is
noteworthy that after the land had been well dunged, and
had grown a crop of beans, the greatest increase, especially
of grain, obtained in the first year was where the manure
was the most nitrogenous. Thus, the ammonia-salts alone,
the guano, and the rape cake each gave 4 to 5 bushels
increase of dressed corn ; whilst the mineral manure, and
the mineral manure and ammonia-salts together, gave only
about one bushel. The ammonia-salts alone also gave
rather more increase of straw than any of the other
manures, more even than the mixed mineral manure and
ammonia-salts together. The produce of the unmanured
plot in the second and succeeding years showed, however,
that the condition of the land had then become reduced ;
and it is therefore from the average results of each of the
different manures taken over a series of years that it is
most convenient to judge of the character of the exhaustion
induced by the growth of the wheat crop in that particular
soil.

Early Experiments on Wheat at Rodmersham. 15

Generally, the results obtained during the four years that
the manures were applied showed that mineral manures
increased the wheat crop but little, ammonia- salts much
more, mineral manures and ammonia-salts used together
more than either, or both, used separately; that Peruvian
guano, containing both mineral and nitrogenous constituents,
gave a considerable amount of increase, but that carbona-
ceous manures had no perceptible effect. They further
showed that the condition of the land was higher than was
desirable for the purposes of the experiments, the result of
which was, not only that the seasons set a limit to the
amount of crop, and therefore to that of the increase
produced, below that which the manures might otherwise
have yielded, but that the increase consisted of a very
undue proportion of straw.

Passing now to the fifth and sixth years, in which all
applications of manures were withheld, the results corrobo-
rated the conclusion already arrived at, that the condition
of the land was too high for the full action of the manures
in the years of their application. They likewise showed
that their influence was not even then exhausted, and this
is further proved by the fact that calculation shows there
was a less proportion of the nitrogen supplied in the
manures in the four years recovered in the increase of the
six years, and in some cases much less, than is sometimes
recovered in the crop immediately succeeding the application
of a nitrogenous manure. Under favourable circumstances,
from 40 to 50 per cent, of the nitrogen supplied in an
artificial manure for wheat may be recovered in the increase
of a first crop. But it is estimated that, in the cases of
the rape cake and of the ammonia-salts alone, there was
only about one-fourth, and in those of the mineral manure
and ammonia-salts, and of the guano, under 40 per cent, of
the nitrogen supplied in the manure of the four years
recovered in the increase of the six years.

A comparison of the value of the increase of crop with

16

The Rothamsted Experiments.

the cost of the manures applied shows that the mixed
mineral manures of Plot 2 were far too expensive in pro-
portion to the amount of increase they yielded for it to
be at all worth while to reckon the cost against the increase
in their case. Ammonia-salts are generally neither so
cheap a source of nitrogen, nor are they, when used alone,
so good a manure for corn crops, as Peruvian guano, which
contains a large proportion of phosphates as well as
nitrogen. Rape cake, though a recognised manure in the
market for wheat, acts somewhat more slowly for the
amount of nitrogen it contains than does guano. It will
be well, for the sake of comparison, to show the cost of
the manures at the period of the experiments, and the
value of the increase due to the three manures — rape
cake, ammonia-salts, and Peruvian guano :

Manures applied per acre in

Increase obtained

four years.

per acre in six years.

t

§

c

Corn.

Straw.

\

g

1

I .

1

a
•5

£

5-j

Description.

«

4

— —

« &i

0

§

g

<§

J

Si

0 JH

I

5

W

*J

a<o

Ib.

£ s.

s.d.

s.d.

^ s. d.

<£ s.d.

£ s.d.

Rape cake

8000

5 10

464

7 0

75^1 3

19 12 10

21 1 4

186

Sulphate of ammonia
Muriate of ammonia

800
800

15 0
20 0

[so

7 0

69fl 3

12 10 0

16 192

492

Peruvian guano

2160

12 10

45i

7 0

91f 1 3

12 1 1

21 132

9 12 1

Eeckoning the value of the increase against the cost of the
manures, there is a considerable margin in favour both of the
ammonia-salts and the guano, particularly the latter. The
evidence further indicates that these active nitrogenous
manures are by no means fully exhausted in the first year of
their application. The quantity of guano employed — nearly
5cwt. to the acre — was, however, more than it is desirable to
apply in ordinary practice. And it should not be inferred
that the practice of growing a series of corn crops by means

Early Experiments on Wheat at Rodmersham. 17

of artificial manures is to be recommended. Nevertheless,
the practical conclusion undoubtedly is that highly nitro-
genous manures much increase the produce of grain crops
under the circumstances in which these are generally grown
n our rotations.

18 The Rofhamsted Wheat Experiments.

III.-EXPERIMENTS ON THE CONTINUOUS

GROWTH OF WHEAT UPON THE SAME LAND

FOR FORTY YEARS.

WHEAT has now been grown for more than forty years in
succession on the same land at Eothamsted, and since the
eighth year of the experiments each plot has, year after year,
received uniform treatment as regards manuring, the varia-
tions in this latter respect comprising (1) without manure,
(2) with farmyard manure, (3) with a great variety of
chemical manures. As an example of an experiment in
which the same kind of crop is grown on the same land,
subject to uniform manurial treatment, year after year, this
elaborate investigation is unique, both in its long continuation
and in the maintenance of uniformity of conditions. To the
R. A. S. Journal, Messrs. Lawes and Gilbert contributed, in
1864, the " Report of experiments on the growth of wheat
for twenty years in succession on the same land," and, twenty
years later, there appeared the " Report of experiments on
the growth of wheat for the second period of twenty years
in succession on the same land." These papers occupy
altogether about 240 pages, some eighty of which are devoted
to figures in the form of appendix tables. It will be conve-
nient to discuss the report in its entirety, as that of forty
years' continuous experiment.

In commencing their first report, the authors remark that
the records of a field of 14 acres, in which wheat has been
grown without manure, and by different descriptions of
manure, year after year, for twenty successive seasons,
without either fallow or a fallow crop, and in which the
lowest produce was in the first year 15, and in the last 17|

Continuous Growth of Wheat for Forty Tears. 19

bushels, and the highest in the first year 24 J, and in the last
56| bushels, cannot fail to be of much interest at once to the
farmer, to the economist, and to the man of science. Much
more, then, is this true of records which now extend over a
period twice as long. The experiments have been made on
what may be called fair average wheat land. The rent of
similar land in the locality ranged from 25s. to 30s. per acre,,
tithe free, and its wheat crop under the usual management of
the district, did not average more than 25 to 27 bushels per
acre every fifth year, so that the land could lay no claim to
extraordinary fertility, or to be ranked on a higher level than
a large proportion of the soils on which wheat is grown with
a moderate degree of success, under a system of rotation and
home manuring. The soil is a somewhat heavy loam, with a
subsoil of raw yellowish red clay, but resting in its turn upon
chalk, which provides good natural drainage.

PLAN OF THE EXPERIMENTS.

The reports include accounts of the amount and character
of the produce obtained in every one of the forty seasons
from 1843 to 1883, a general description of the weather in
each season being accompanied by a table showing the
manures applied and a summary of the results obtained.
The appendix tables show the effects of one manure compared
with those of another in each season separately, and the great
difference of effect of the same manure in one season compared
with another, and its increasing or diminishing effect when
used year after year on the same plot. But as it is essential
to know not only the quantity, but likewise the chemical
composition of the produce, determinations have each year
been made of the proportions of dry substance and of mineral
matter in both the grain and the straw of each plot. The
proportion and quantity per acre of the nitrogen, and the
composition of the ash, in both grain and straw, have in many

c 2

20 The Eothamsted Wheat Experiments.

selected cases been determined. The percentage of nitrogen
in the soil of some of the plots, at different stages of the
progress of the experiments, has also been estimated.

One useful result of a study of the Rothamsted experi-
ments is that intending investigators may learn, not only
what methods are best to follow, but those which it is
desirable to avoid. As already mentioned, these wheat-
growing experiments were not during the first eight years
carried out with uniformity of manuring on each plot.
Referring to this point, the experimenters observe, in the
introduction to their second report, that an attempt to inves-
tigate the growth of wheat, under circumstances so totally
different from any which had ever previously arisen, could
hardly be carried out without the commission of some errors
which experience would have enabled them to avoid. They
have now ascertained that, if it is desired to advance beyond
the question of the best manure to grow one crop of wheat,
continuity of the same manuring is of supreme importance.
With the exception of the plot receiving farmyard dung, and
the unmanured plot — neither of which has been altered from
the commencement — there is no plot in the field in which some
change did not take place in the manure applied during the
early years of the experiments. But in the autumn of 1851,
eight years after the experiments were commenced, the
manuring of the field was arranged upon a fixed plan which,
with very slight changes in one or two instances, has been
followed ever since.

As, therefore, the manurial treatment of each plot has,
since 1852, not been varied, it will be convenient to state
here, once for all, the description and quantities, estimated
per acre per annum, of the several manures :

Plot 2. — 14 tons farmyard manure (also for the eight preceding years).
Plot 3. — Unmanured (also for the eight preceding years).
Plot 20. — Unmanured (also for five preceding years).
Plot 4. — Unmanured (sulphate of ammonia and bone ash acted upon by
hydrochloric acid, for seven preceding years).

Manures Employed. 21

p^oi; o. — Superphosphate of lime alone, composed of GOOlb. bone ash
and 4501b. sulphuric acid, sp. gr. 1'7 (also for three preceding years).

P.ot 1.— Mixed alkalies, composed of GOOlb. sulphate of potash, 4001b.
sulphate of soda, and 2001b. sulphate of magnesia (also for three pre-
ceding years) ; reduced to 4001b., 2001b., and 2001b., respectively, in the
sixteenth and succeeding seasons.

Plots 5 (a and &). — Mixed mineral manure, composed of 3001b. sulphate
of potash, 2001b. sulphate of soda, lOOlb. sulphate of magnesia, and 3501b
superphosphate of lime (consisting of 2001b. bone ash and 1501b. sulphuric
.acid, sp. gr. 1*7). In the sixteenth and succeeding seasons, the sulphate
of potash was reduced to 2001b., and the sulphate of soda to lOOlb.

Plot 21. — Mixed mineral manure, as plots 5, and lOOlb. muriate of
ammonia.

Plot 22. — Mixed mineral manure, as plots 5, and lOOlb. sulphate of
ammonia.

Plots 6 (a and b). — Mixed mineral manure, as plots 5, and lOOlb. each
sulphate and muriate of ammonia.

Plots 7 (a and 6). — Mixed mineral manure, as plots 5, and 2001b. each
sulphate and muriate of ammonia.

Plots 8 (a and b). — Mixed mineral manure, as plots 5, and SOOlb. each
sulphate and muriate of ammonia.

Plots 16 (a and b). — Mixed mineral manure, as plots 5, and 4001b. each
sulphate and muriate of ammonia.

Plots 1 7 (a and b). — 2001b. each sulphate and muriate of ammonia in
the ninth and every alternate season; and mixed mineral manure, as
plot 5, in every intermediate season.

Plots 18 (a and b). — Mixed mineral manure, as plots 5, in the ninth and
every alternate season ; and 2001b. each sulphate and muriate of ammonia
in every intermediate season.

Plots 10 (a and b). — 2001b. each sulphate and muriate of ammonia.

Plots 11 (a and b). — 2001b. each sulphate and muriate of ammonia, and
superphosphate of lime as plots 5.

Plots 12 (a and b). — 2001b. each sulphate and muriate of ammonia,
superphosphate of lime as plots 5, and 5501b. sulphate of soda (reduced to
3661b. in the sixteenth and subsequent seasons).

Plots 13 (a, and b). — 2001b. each sulphate and muriate of ammonia,
superphosphate of lime as plots 5, and 3001b. sulphate of potash (reduced
to 2001b. in the sixteenth and subsequent seasons).

Plots 14 (a and b). — 2001b. each sulphate and muriate of ammonia,
superphosphate of lime as plots 5, and 4201b. sulphate of magnesia
-(reduced to 2801b. in the sixteenth and subsequent seasons).

Plot 9a. — 5501b. nitrate of soda and mixed mineral manure as plots 5
(only 4751b. nitrate in the ninth, and 2751b. in the tenth and eleventh
seasons, and no mineral manure in the ninth, tenth, and eleventh seasons,
commencing only in the twelfth).

22 The Eothamsted Wheat Experiments.

Plot 96. — 5501b. nitrate of soda alone (only 4751b. in the ninth season).

Plot 15a. — Mixed mineral manure as plots 5 (but with 2001b. hydro-
chloric instead of 1501b. sulphuric acid), and 4001b. sulphate of ammonia.

Plot 156. — Mixed mineral manure as plots 5 (but with 2001b. hydro-
chloric instead of 1501b. sulphuric acid), 3001b. sulphate of ammonia, and
5001b. rape cake.

Plot 19.— 2001b. bone ash, 2001b. hydrochloric acid, 3001b. sulphate of
ammonia, and 5001b. rape cake.

The sulphates of potash, soda, and ammonia, the muriate
(chloride) of ammonia, and the nitrate of soda, were the
ordinary articles of commerce passing under those names ; the
sulphate of magnesia was Epsom salts. The term " ammonia-
salts " may for brevity be employed to denote the mixture of
equal quantities of sulphate and muriate of ammonia.

The 14-acre field set apart for the experiments had, at the
commencement of the investigations, grown turnips, barley,.
peas, wheat, and oats since the application of manure, and
would therefore, according to the ordinary rules of practice,
be considered so far exhausted as to need re-manuring before
growing another crop. It was considered that a field in this
condition would be peculiarly fitted to show in which of the
constituents of the crop to be grown the soil had become
practically the most deficient by the removal in rotation of
the five preceding crops just enumerated. It was felt by the
experimenters at the outset that far more had yet to be done
in determining the chemical and physical qualities of soils in
relation to the atmosphere, and to manurial substances
exposed to their action, as well as in perfecting methods of
analysis, before comparative analyses could aid much in
deciding upon the relative productiveness of different soilsr
to say nothing of the still more difficult problem of estimat-
ing by such means the condition of fertility or exhaustion of
one and the same soil at different times. The field was at
first divided into plots, of which most consisted of two lands
(each about 12ft. 5in. wide), running the whole length of
the field, and comprising altogether nearly two-thirds of an
acre. After the second season, however, the double land plots

Continuous Growth of Wheat for Forty Years. 23

were each divided into two, though in most cases the two
(a and b) were similarly manured, thus providing duplicate
experiments with the same manure.

INFLUENCE OF SEASON ON THE WHEAT CROP.

In a series of forty tables are summarised the results of
each season, every table showing for each plot the total
weight of grain, and of straw and chaff, per acre, the number
of bushels of dressed grain per acre, and the weight per
bushel. In the first report the authors observe that no idea
is more fixed and prevalent in the farmer's mind than that,
after all his labour and money have been expended, he is still
at the mercy of the seasons for his reward. The tables to
which reference has just been made supply interesting
evidence on this point, but the extent of the dependence upon
season will be made more strikingly manifest by placing side
by side the results obtained by one and the same description
and amount of manure, in the least favourable and in the
most favourable of the seasons during which the same manure
has been supplied year after year on the same land. Of the
twelve seasons during which, in the first twenty years,
uniformity of manuring was maintained, the harvest of 1863
gave the best results, and that of 1853 the worst, and the
authors compare these results in a table. Taking, however,
the results also of the subsequent twenty years, whilst 1863
still occupies the best position, 1879 achieves the unpleasant
notoriety of securing the worst, and it must suffice, therefore,
to compare in Table I. the yields in these two years. The
following are the numbers of the plots compared, with the
description of manure per acre each received :
Plot.

3. Unmanured.

2.. Farmyard manure.

5. Mixed mineral manure alone.

6. Mixed mineral manure and 2001b. ammonia-salts (= 431b. nitrogen).

24 The Rothamsted Wheat Experiments.

7. Mixed mineral manure and 4001b. ammonia-salts (= 861b. nitrogen).
9. Mixed mineral manure and 5501b. nitrate of soda (== 861b. nitrogen).

8. Mixed mineral manure and 6001b. ammonia-salts (= 1291b. nitrogen).

It is only by growing the same crop continuously for many
years upon the same land, without any change in the con-
ditions of manuring, that the influence of different seasons
as so strikingly shown in this table, can be accurately
ascertained. It is seen that, in the year 1863, the produce of
the unmanured plots was 17| bushels ; whereas, in the worst
year, 1879, it was only 4£ bushels. Nineteen unmanured
crops had been grown previous to 1863, and the yield of that
year was equal to the average of the first eight years. The
unmanured yield of 1879, only 4£ bushels per acre, was about
one-third of the average yield of the plot during the whole
period of forty years. Contrary to what might be expected,
the produce on the land receiving farmyard manure, whilst it
falls greatly in yield in a bad season, does not rise as rapidly
in yield in a very favourable season. This is well illustrated
by the following comparison : Plot 7, receiving mineral
manures and ammonia-salts, gave in 1879 a crop practically
equivalent to that of the farmyard manure plot (2), 16
bushels per acre ; the average produce of the thirty-two
years is almost identical in the two cases, one being 32f , and
the other 33J bushels per acre. But in the favourable season
of 1863, whilst the farmyard manure (plot 2) gave only
44 bushels per acre, the artificial manure (plot 7) gave 53f »
being an excess of nearly 10 bushels per acre. It is therefore
evident that, under the most favourable climatic circum-
stances, the artificial manure is capable of giving a much
larger crop, both of grain and straw, than the farmyard
manure. With the heaviest artificial manuring (plot 8),
35 1 more bushels per acre of wheat, and upwards of two
tons per acre more gross produce (grain and straw), were
grown in the most favourable season, as compared with the
produce grown by the same manures in the worst season.

The character of the weather during the best and worst

Produce of Best and Worst Seasons compared. 25

TABLE I. — SHOWING THE PRODUCE OP THE BEST SEASON, 1863, THE

WORST SEASON, 1879, AND THE AVERAGE OF 32 YEARS, 1852-1883.

Dressed Grain, per Acre — Bushels.

PlotNos.

Best Season,
1863.

Worst Season,
1879.

Difference

Average of 32 years,

1852-1883.

3

m

4f

12*

13*

2

44

16

28

33*

5

19|

51

14

6

39f

10*

29|-

241

7

58|

16i

87|

32f

9

55|

22

33|

36i

8

55J

20|

35i

36i

Weight per Bushel of Dressed Grain — Ib.

PlotNos.

Best Season,
1863.

Worst Season,

1879.

Difference.

A™TS£,sri*

3

62-7

52-5

10-2

58-8

2

63-1

56-8

6-3

60-0

5

63-0

53-5

9-5

58-7

6

62-3

56-5

5-8

59-5

7

62-5

567

5-8

59-5

9

62-1

56-5

5-6

58-7

8

62-3

56-5

5-8

59-2

Straw (and Chaff) per Acre — Ib.

Plot Nos.

Best Season,
1863.

Worst Season,
1879.

Difference.

Average of 32 years,
1852-1883.

3

1600

763

837

1272

2

4279

2239

2040

3570

5

1728

855

873

1464

6

3715

1592

2123

2512

7

5866

3012

2854

3771

9

6312

4347

1965

4688

8

6602

4176

2426

4532

Total Produce (Grain and Straw) per Acre — Ib.

PlotNos.

Best Season,
1863.

Worst Season,
1879.

Difference.

Average of 32 years,
1852-1883.

3

2727 1093

1634

2090

2

7165 3303

3862

5689

5

3017

1238

1779

2421

6

6243

2283

3960 4029

7

9358

4063

5295 5845

9

9888 5809

4079 6982

8

10,216 5527

4689 6832

26 The Rothamsted Wheat Experiments.

seasons respectively is of considerable interest. The record
for the season 1862-3 is as follows : October, unusually warm,
but with a good deal of wind and rain ; November, cold,
comparatively little rain ; December to February, unusually
mild, with a fair amount of rain in December and January,
and but little in February; March, upon the whole, mild,
with but little rain, and wheat showing unusually forward
growth ; April, very dry and warm ; May brought some
refreshing rains, but temperature was occasionally extremely
low, and pretty nearly throughout rather below the average,
with frequent storms of wind ; June, with temperature rather
below the average, and with a good deal of rain, whi'-h,
though needed, was so heavy as to lay the most forward and
bulky crops ; July brought much less rain than usual, with
moderately high day but low night temperatures, and some
sharp night frosts ; August, with only moderate tempera-
tures, but less rain than usual, was, upon the whole, favour-
able ripening anu harvesting weather ; September was marked
by a good deal of rain and rather low temperatures. In June
the condition of the atmosphere as to moisture was about the
average for that month, but in July, August, and September
both the actual amount and the degree of humidity were
below the average. The extraordinary yield of the harvest of
1863 would appear to have been due to the almost unchecked
growth, from the first appearance of the plant above ground
up to the time of harvest, rather than to any extraordinary
characteristics of season at any one or more particular periods.
With the extremely mild winter and early spring, the plant
came early forward, and the rains, though sparing upon the
whole, came when needed ; whilst, though the temperature of
the summer was seldom high, it was (excepting the night
frosts of July) generally sufficient, and the condition of
atmosphere otherwise favourable, so that the whole season
contributed to a lengthened and almost unbroken course of
gradual accumulation. The harvests of 1854 and 1857, in
order, came nearest that of 1863 in total produce.

Characteristics of a Bad Season. 27

Turning now to the fateful season of 1878-79, the weather
up to Oct. 21, 1878, was fine and warm ; but it became cold
and wet from Oct. 22 to Dec. 6, when severe frosts set in, and
lasted till Dec. 26. There was, then, a good seed time, followed
by intensely cold, wet weather, till nearly the end of the year,
the temperature of the whole month of December being
6 degrees below the average. In 1879, January was one of
the coldest months ever recorded, the thermometer standing
below freezing point throughout ; snow covered the ground,
the days were nearly sunless, and the wind N. and N.E.
February very cold, with a great excess of rain, and much
snow. March first warm, then cold, with snow on the 21st ;
but the last days of the month warm. The five months
ending March 31 were described as exceedingly cold, with
much rain and snow. The next three months were cold, wet,
and sunless ; whilst, for lowness of temperature, the eight
months ending with June, 1879, have only been once exceeded
during the one hundred years and upwards which have
elapsed since the first records were kept at Greenwich. July
was dull, cold, and sunless ; rain fell every day during the
first half of the month, and frequently afterwards, sometimes
mixed with snow. August was a very cold, wet month ; there
were a few warm days, during which no rain fell ; and there
were a few fine, dry days in the early part of September, after
which rain fell almost daily until the end of the month.
The wheat crop of 1879 was the worst grown in Great
Britain since the year 1816. It was estimated that the
crop of the country was not much more than half an
average one.

If, from an examination of the results of the first twenty
years' experiments on the continuous, growth of wheat, an
attempt be made to determine whether the latter or the
earlier seasons were probably on the average the more favour-
able, the annual produce without manure would appear at first
sight to afford the safest means of judging. A comparison
of the tables showing the produce of each season serves to

The Rothamsted Wheat Experiments.

demonstrate, however, that those seasons which were the
most favourable for the unmanured, or for the merely
mineral-manured plots, were not at all the most favourable
for those manured highly with nitrogenous fertilisers — that
is, for those conditions under which alone large crops could
be obtained. Hence, the best season for land in low condi-
tion is not the best for land in high condition.

But, by comparing the increasing or diminishing amounts
of produce from year to year, under very different conditions
of manuring, a very fair judgment of the relative character of

TABLE IL— AVERAGE ANNUAL PRODUCE (GRAIN AND STRAW)
PER ACRE.

Plot.

Dressing every year.

First
half of
period.

Second Total

Duration
of total
period.

3
10a
2

3

Unmanured
Ammonia-salts alone
14 tons farmyard manure

Unmanured

! ib.

' 2711
4474

; 4828

2641

Ib. Ib.
2728 2719
4166 4312
6355 , 5591

2610 2626

Years.
20
19
20

5

lOa

Mixed mineral manure alone . .
Ammonia-salts alone

3183
4156

2927 3055
3921 4038

I "

7
2

( Ammonia-salts and mixed ^
( mineral manure )
14 tons farmyard manure

6428
5896

6546 6487
6306 6101

1*
! (1852 to
1863).

the earlier and later seasons can be formed. From a table
indicating the average produce during the first half of the
period, during the second half, and during the entire period,
it appears that without manure a slightly, though very
slightly, increased annual produce of grain and total pro-
duce (though not of straw) is marked over the last half
as compared with the first half of the twenty years ; with
ammonia- salts alone there was a decreased, and with farm-
yard manure a very much increased, rate of produce in the
later years. Table II. is a portion of the table referred to.
Thus, where the crop (plot 3) was simply dependent on

Yields of Earlier and Later Years Compared. 29

the soil and season, the produce was somewhat higher in the
later years ; where the resources of the soil were overtaxed
by the use of a large amount of ammonia salts every year
(plot 10a), the produce diminished ; but where an excess of
every constituent was annually applied (plot 2) the crop
enormously increased as the experiment proceeded. These
generalisations are based upon the first three lines of figures
in the foregoing table.

Taking the final twelve years, the table further shows that
the average annual yield was, without manure (plot 3) much
the same over the whole period ; that, notwithstanding the
exhausting effects of applying ammonia-salts every year
(plot Wa), the annual diminution of produce under their
influence was proportionally less during the latter half of
the last twelve, than of the whole nineteen years of their
use ; that, where ammonia-salts and all mineral constituents,
except silica, were liberally supplied every year (plot 7), the
produce of grain increased and that of straw somewhat
diminished ; lastly, that where an excess of every constituent
required by the crop was annually applied, as in the farm-
yard manure (plot 2), the rate of increase from year to year
was not so great during the later as during some of the
earlier years.

Upon the whole, it must be concluded that the later years
were, on the average, slightly more favourable to the crop
than the earlier ones. The fact of the unmanured plot
maintaining its produce throughout the whole twenty years
is, then, probably in some degree due to the better average
of the seasons themselves in the later years. Had it been
otherwise, the unmanured produce would have shown some
slight decline in the later years, or rather, some slight-
excess in the earlier ones, due to the accumulation of many
previous courses of manuring1 and cropping. In the report
of the second twenty years' experiments, and with the results
of forty years' continuous growth before them, the investi-
gators do not introduce a comparison between the first

30 The Rothamsted Wheat Experiments.

twenty years and the second twenty years so far as to
determine which of the two periods was on the whole the
more favourable for the growth of wheat. Nevertheless,
the foregoing illustration will serve to convey an idea of the
lines on which such an inquiry would be conducted.

EFFECTS OF THE UNEXHAUSTED RESIDUE OF
MANURES.

Having, in the report of the first twenty years, discussed
at length the amount and character of the produce obtained
in different seasons, the authors proceed in a second section
to inquire into the effects of the unexhausted residue from
previous manuring (both nitrogenous and mineral) upon
succeeding crops. The questions of the permanency of
effect of different manures, and of the tendency to exhaus-
tion which partial manuring may induce, are of great
practical importance, and are frequently discussed by prac-
tical men. By means of a series of tables and coloured
diagrams, a number of valuable conclusions are established,
of which it is only possible to give here a summary.

It appears, then, that a somewhat heavy loam, of fair
average wheat-producing quality, taken at the end of a five-
course rotation since manuring, gave scarcely any increased
produce of wheat in the year of the application, when
manured with a mixture of silicate of potash and super-
phosphate of lime ; but it gave a very considerable, though
progressively diminishing, amount of increase, when after-
wards manured for nineteen consecutive years with ammonia
salts alone. Though, at the outset, the soil still contained an
excess of annually available mineral constituents, as compared
with the annually available nitrogen supplied by soil and
season without manure, yet the annual application of large
quantities of ammonia-salts made apparent the relative
deficiency of mineral constituents, even as early as the fourth

Unexhausted Residue of Manures. 31

year. When ammonia- salts were applied, the greater portion
of the nitrogen remained unrecovered as increased yield in
the crop for which it was employed. The unexhausted
residue of nitrogen supplied as manure was but very partially
and very slowly recovered as increased yield in succeeding
years, even when followed by the liberal application of such
mineral manure as was very effective when used in conjunc-
tion with newly applied ammonia-salts. Mineral constituents
supplied in the soluble condition in the fifth and seventh
years of the experiments, though giving very little increase
when in the latter year they were used alone, continued to
increase the effect of ammonia- salts afterwards annually
applied for thirteen consecutive years. A given amount of
ammonia- salts gave very different amounts of increase,
according to the supply of available mineral constituents
within the soil, giving very much more when mineral manures
were supplied in the same, than in the preceding year, not-
withstanding that, in the latter case, there could be no
deficiency, though, doubtless, less favourable condition and
distribution of the mineral constituents. The same mineral
manures which were very effective when supplied in conjunc-
tion with ammonia-salts, gave very little increase of produce
when used alone year after year for twelve years, although
following an excess of ammonia-salts applied in preceding
years ; and they gave very little more when they were applied
every year succeeding an excess of ammonia- salts applied in
the immediately preceding year. The unexhausted residue
from previous mineral manuring, though it served as an
effective reserve against exhaustion, had little or no effect in
increasing the growth of wheat without the aid of available
nitrogen provided within the soil. An unexhausted residue
from previous nitrogenous manuring had also but little
influence upon the immediately succeeding crops, even when
aided by the application of mineral manures.

The bearing of the foregoing facts upon the question of
the probable influence on the mineral wealth of our soils of

32 The Rothamsted Wheat Experiments.

the use of artificial nitrogenous manures, under the circum-
stances and in the degree in which they are generally
employed in the ordinary course of agriculture in this
country, remains to be considered. But it seems desirable
to notice here what is said on the subject of the unexhausted
residue of manures in the report of the second twenty years'
experiments, for which the results obtained on plots 16a and
166 afford the necessary data. From 1852 to 1864 inclusive,
these plots received mixed mineral manure and 8001b. of
ammonia-salts per acre per annum, containing 1721b. of
nitrogen. It is very rarely that a sufficiently large crop of
wheat is grown to remove one-half of this quantity of
nitrogen. In 1863 and 1864 the seasons were highly favour-
able for the growth of wheat ; and, as the size of the crop is
regulated very much by the amount of nitrogen at its
disposal, the two seasons were well adapted for ascertaining
how much of this very large application would be employed
in the growth of the crop. In 1863, plot 16 gave 56 bushels
per acre ; in 1864, 51 bushels per acre. But plots Sa and 86
produced 56 bushels and 50 bushels in the same two years, by
means of an application of 6001b. of ammonia-salts ; and, as
the additional 200lb. supplied to 16a and 166 only added one
bushel to the crop, it was quite evident that the possible
limits of growth on 8a and 86, even in seasons so favourable,
had been reached. All further manuring was, therefore,
stopped after 1864, with the view of obtaining information,
which might prove to be very valuable, in regard to the
unexhausted residue of the manure employed.

Table III. shows the produce of plot 16 for the last two
years during which it received manure, and for the nineteen
succeeding years during which it was unmanured ; also for
comparison there is given the produce of plot 5, which
received an annual dressing of minerals alone over the whole
period.

It will be noticed that in the first year after the manures
were stopped, the produce of plot 16 was 18 bushels more

Unexhausted Residue of Manures.

33

than that of plot 5. In the next year the difference was
only 4 bushels in favour of plot 16. In each of the two
next years plot 16 yielded 5 bushels in excess of plot 5.
In the fifth year (1869) the total produce on plot 16, grain
and straw, was only 1041b. more, and in the sixth year it
was even 71b. per acre less than on plot 5. During the next

TABLE III.

Dressed grain in bushels per

Total produce (grain and straw)

acre.

per acre, — Ibs.

Year.

Plot 5.

Plot 16.

Plot 5.

Plot 16.

Minerals

Minerals and

Minerals

Minerals and

alone.

1721b. nitrogen.

alone.

1721b. nitrogen.

1863

19|

55|

3017

10,525

1864

16| ,

w*

2462

9348

Unmanured

Unmanured

1865

14i

32f

2091

5007

1866

13i .

171

2303

3081

1867

9i

14f

1613

2512

1868

171

22f

2481

3503

1869

15f .!

10*

2543

2647

1870

181 ;

18*

2564

2557

1871

HI

13£

2207

2380

1872

12f . •:

184

2166

2387

1873

12f :

12|

1806

1921

1874

13

11*

1674

1892

1875

9i

m

1714

1829

1876

10*

11

1429

1538

1877

HI

91

1570

1340

1878

14f 1

13f

2222

2181

1879

5* : -. '

41

1238

1154

1880

171

14f

2818

2383

1881

12* - '•

13£

1709

1736

1882

m

lOf

2057

1925

1883

15f

15|

2147

2131

six years, however, the total produce of plot 16 exceeded
that of plot 5 in amounts varying from 1091b. to 2211b.
After 1876, the table shows that, with one exception, the
yield on plot 16 was less than that on plot 5. It was, there-
fore, not until twelve years after the last application that all
influence due to the previous manuring ceased.

34 The Rothamsted Wheat Experiments.

The large crop grown in 1865, the first year after the
manures were stopped, is attributed to some of the nitrogen
of the salts of ammonia not having been washed out of the
soil. The year of the last application, 1864, was one of
exceptional drought — the driest experienced at Rothamsted
during the whole forty years ; it was probably owing to this
that, at all events, a portion of the produce of the first year
without manure was due. The excess of produce obtained
on plot 16, as compared with plot 5, during the next eleven
years, probably arose from the slow decay and nitrification
of the stubble and roots of the very large crops which had
been grown on this plot for so many years. During the
thirteen years of the annual application of 8001b. of
ammonia-salts, considerably more than lOOOlb. of the
nitrogen applied to the soil was not recovered in the crops.
It is therefore hardly possible to suppose that the nitrogen
(about 601b.) contained in the 66631b. of total produce
obtained in excess of that grown on plot 5, during the next
twelve years, could have been derived directly from that
supplied as manure, excepting so far as a portion of that of
the first crop was concerned. All the evidence, indeed,
points to the residue of the crop itself as the source of the
unexhausted manure ; and it is quite certain that the very
large amount of roots and other residue, possessing consider-
able fertilising influence, which some of our rotation crops
leave in the ground, has much to do with their value as
restorative crops. In 1865 samples of soil from various
plots in the experimental wheat field were analysed, and the
percentage of nitrogen in the first 9in. of soil was higher on
plot 16 than on any other plot receiving artificial manures ;
whilst, in 1881, after seventeen unmanured crops had been
taken, when the soil was again analysed, it was found that
the percentage of nitrogen was considerably reduced, and
it was in fact then not much higher than that on the
continuously unmanured plot.

Continuous Growth of Wheat Without Manure. 35

CONTINUOUS GROWTH OF WHEAT UPON THE
SAME LAND FOE FOETY YEAES WITHOUT
MANUEE.

The third section of the report of the first twenty years'
experiments is concerned with the average annual result over
the last twelve years by each description of manure applied
year after year on the same plot. As, however, the report
of the second twenty years continues the same line of
inquiry over a much longer period, it will suffice to glance
at only one or two of the leading features given under this
head in the first report. It is observed that an average
annual produce of wheat, amounting to from 15 to 16
bushels of grain, and from 15 to 16 cwt. of straw, without
manure of any kind, is looked upon by many as an extra-
ordinary yield, and as indicating a somewhat unusual
quality of land. There is no doubt it bears a higher
proportion than might have been expected to the produce
obtained, even under rotation with periodical manuring, in
a large majority of cases where land is badly farmed ;
whilst deficient range and aeration of soil, luxuriant weed
growth, and defective manuring, all contribute to the
miserable result. The experimental land, though kept
extremely clean, was not, however, ploughed more deeply
than in the ordinary practice of the farm ; and there can
be little doubt that a large proportion of those soils of the
country which are recognised as possessing average wheat-
producing qualities would yield very similar results, if kept
equally clean and otherwise as well cultivated ; whilst some
would,, under like conditions, produce much more, though
many light soils probably much less.

Eegarding the history of the permanently unmauured
plot over the two periods of twenty years each, it must be
remembered that the last time this land received any manure
was in 1839. The turnip crop of that year was succeeded
by barley, peas, wheat, and oats, all unmanured, so that by

T> 2

36

The Rothamsted Wheat Experiments.

1844 the land might fairly be considered as, agriculturally
speaking, exhausted. Since then the cultivation has been
of the simplest description, and no attempt has been made
to increase the crop by deep or subsoil ploughing ; the land
has, however, been kept free from weeds. A summary of
the results, given in four periods of ten years each, is
presented in Table IV.

TABLE IV. — PERMANENTLY UNMANURED PLOT.

Dressed grain
per acre.

Weight
per bushel.

Total produce
(grain and
straw).

Mean of 10 years, 1844-1853
1854-1863
1864-1873

1874-1883

Bushel s.
15*
16*

m

io±

Ib.
58-25
57-57
58-97
58-25

Ib.
2711
2728
1924
1614

Mean of 40 years

14

58-26

2244

The increased yield of grain during the second ten years
as compared with the first is attributed to the favourable
seasons of the second decade ; the total produce (grain and
straw), a much more accurate measure of the available
fertility of a soil than is the grain alone, is also slightly
higher over the second period than over the first. The
decline during the third decade as compared with the second
is seen to be very marked, much more so than that of the
fourth as compared with the third. But the more recent
seasons had been so unfavourable for the growth of wheat
that the produce of the fourth ten years cannot be accepted
as correctly representing the reduction due to soil exhaustion
alone. This is proved by the fact that the produce of the
last year, 1883, in a rather better season, was 13| bushels
per acre, which closely approximates to the average yield of
the forty crops.

The annual decline in the produce, due to exhaustion—

Continuous Growth of Wheat Without Manure. 37

irrespective of variations due to good or bad seasons — is,
probably, up to a certain period, equivalent to about a
quarter of a bushel per acre per annum, representing a gross
produce in grain and straw of 401b. per acre. But, with
each decline, the reduction must obviously become less and
less ; atmospheric influences, and even the small amount of
ammonia brought down in the rain, will produce a great
effect upon a declining crop. Hence, the actual process of
the exhaustion of the soil differs considerably from all the
preconceived ideas on the subject. The soil, in fact, not
only contains more fertility, but also holds it with a much
firmer grasp, and parts with it less readily, than was
previously thought to be the case.

The average annual amount of total produce (grain and
straw) removed has been 1 ton per acre, containing nearly
19001b. of absolutely dry matter; and there is strong
evidence, derived from other experiments in the field, to
prove that the carbon, indeed a large proportion of the
organic matter, is derived from the atmosphere, whilst the
nitrogen and the mineral matter are taken from the soil.
This would divide the products into from 94 to 95 per cent,
atmospheric constituents (including water), and from 5 to 6
per cent, soil constituents. The average amount of soil
constituents (minerals and nitrogen) annually removed by
this unmanured wheat crop is from lOOlb. to 1201b. per acre ;
and of the three most important constituents of plant
growth there have been removed about 171b. of potash, lOlb.
of phosphoric acid, and 201b. of nitrogen.

It appears, then, that upon a field which has been under
arable cultivation certainly for two or three centuries — and
possibly for a much longer period — and which has consequently
lost a very considerable amount of its original fertility, there
is — after the removal of forty unmanured crops — a yield
which differs very little from the average of some of the
great wheat-growing countries of the world ; the yield of the
TJnited States, India, and China being, it is stated, from

38 The Rothamsted Wheat Experiments.

12 to 13 bushels per acre. As the Rothamsted soil certainly
contains a very much less stock of fertility than the soils
upon which wheat is extensively grown in other countries,
it is impossible to attribute the comparatively large yields
there to any other cause than to the clean state of the land.
The amount of food at the disposal of the crop is small,
but it is not shared to any great extent with other plants.
The large produce of both wheat and barley on the
unmanured land in the Woburn experiments also shows how
much the crops grown upon the ordiuary cultivated land
of this country are reduced by weeds. It is true that weeds
do not exhaust a soil, as, in their decay, the fertility which
they have taken up becomes again available ; but they take
up nitrates, which, during their growth, revert to the form of
organic nitrogen (that is, nitrogen, combined with carbon),
and this must undergo nitrification in the soil before becoming
again available as plant food. Considering the high price
paid for active nitrogen,' as in salts of ammonia or nitrate
of soda, a serious loss is incurred if this nitrogen goes to
promote the growth of weeds, as so much time must elapse
before it is again available as food for a future crop.

CONTINUOUS GROWTH OF WHEAT UPON THE
SAME LAND WITH MINERAL MANURES
ALONE.

The plot which received mineral manures alone next claims
consideration. Bearing in mind the fact that uniformity of
manuring was not resorted to on all the plots till after the
expiration of eight years, the remaining thirty -two years
of the total forty may be conveniently divided into four
periods of eight years each. In Table V. the corresponding
results on the unmanured plot are given for the sake of
comparison.

During the first eight years (1844-1851) plot 5 received

Continuous Growth of Wheat with Mineral Manures Alone. 39

ammonia-salts as well as mineral manures, and the average
yield during those eight seasons was 29 bushels per acre,
or nearly 12 bushels annually in excess of the produce
on the unmanured plot ; whilst during the first eight
years of the mixed minerals, without the ammonia-salts
(1852-1859), the average produce was 19 bushels per acre,
or about 3 bushels more than the permanently unmanured
produce. Comparisons respecting subsequent periods are
afforded by the table, and it will be seen that, whilst the

TABLE V. — SHOWING THE AVERAGE PRODUCE OF DRESSED GRAIN, AND
TOTAL PRODUCE (GRAIN AND STRAW) PER ACRE ON PLOT 3 (PER-
MANENTLY UNMANURED), AND PLOT 5 (RECEIVING MIXED MINERALS

FOB A PERIOD OF THIRTY-TWO YEARS), OVER FOUR PERIODS OF EIGHT
YEARS EACH, AND OVER THE TOTAL PERIOD.

Dressed grain.

Total produce (grain
and straw).

Without
manure.
Plot 3.

Mixed
minerals.
Plot 5.

Without
manure.
Plot 3.

Mixed
minerals.
Plot 5.

8 years,

5>
»)

32 years
40 years

1852-1859
1860-1867
1868-1875
1876-1883

,1852-1883
,1844-1883

Bushels.

18*
Uft

12*
10*

Bushels.
19
15i
14
12f

Ib.
2736
2183
1833
1610

Ib.
3191
2450
2144
1899

13*
14

15i

2090
2244

2421

average produce of the unmanured plot for forty years was
14 bushels, the average of thirty-two years for the plot
receiving mixed minerals was 15J ; hence the application
of a very liberal supply of minerals has only been competent
to increase the yield by 1 £ bushels per acre per annum !
Again, the average total produce for the thirty-two years
shows a difference of only 3 3 lib. in favour of the minerals.
The amount of nitrogen in this 3311b. would not exceed 31b.,
and this represents the whole of the nitrogen which the

40 The Rothamsted Wheat Experiments.

wheat upon an acre of land, though furnished with an
abundance of minerals, and certainly at the commencement
containing more crop residue than the unmanured land,
has been able to obtain from the soil and the atmosphere
in excess of that in the wheat grown without manure !

The history of these plots 3 and 5 is very instructive.
Their yields year by year rarely differ from each other by
more than from 3 to 4 bushels per acre. The yield on both
is slowly declining, and for the last twenty-four years neither
has given as much as 20 bushels per acre ; and, without
some change in the manures, it is hardly likely this amount
can ever be grown again. The soil contains a large amount
of the mineral food of plants ; it also contains organic
nitrogen — that is, nitrogen combined with carbon, the residue
of previous vegetation. This organic nitrogen is not directly
available as food for the wheat plant, but every year a certain
portion of it is converted into nitric acid, which combines
with the lime in the soil. In this state it is very soluble in
water, is readily washed out of the soil by heavy rain, and is
a most important and essential food of the wheat plant.
The amount of nitric acid formed each year will vary, the
process of nitrification being most rapid in the hottest weather,
provided the soil is sufficiently moist. The amount of nitric
acid which the wheat crop can take up will also vary, and in
a cold and wet winter much will be washed beyond the reach
of the roots of the plant. These facts are of universal
application, and by them it is possible to explain some of the
causes which tend to the production of good or bad crops of
wheat. Analysis proves both plots 3 and 5 to have lost a
large amount of organic nitrogen, and that, in the first nine
inches1 depth, the mineral- manured soil has lost rather the
more. The total loss of nitrogen over a given area is larger
than the amount of that substance removed in the crops,
and the reason is that, except when the crop is in full vigour
of growth, the drainage waters contain nitric acid. Of the
281b. to 321b. of nitrogen per acre available each year at

The Nitrogen of the Soil. 41

Rothamsted, from soil, seed, rain, &c., only about two-thirds
are removed in the crop, whilst one-third goes into the drains
and is lost. The experiments under discussion provide an
explanation of the fact that on some soils — more especially
the newly cultivated soils of the United States and of Canada
— a large increase in the wheat crop frequently follows the
application of mineral manures. Soils rich in organic matter
may yield an increased amount of nitric acid by the applica-
tion of phosphates and potash, but in all cases the source of
the nitrogen is the soil ; and the loss by the unmanured soil
of perhaps from SOOlb. to lOOOlb. of nitrogen per acre during
the forty years of these experiments is a fact of the greatest
importance.

CONTINUOUS GROWTH OF WHEAT UPON THE
SAME LAND WITH AMMONIA-SALTS ALONE.

Leaving the mineral-manured plot, the next subject to study
is the produce of plots 10a and 106 (ammonia-salts without
minerals), and plots 17 and 18 (ammonia-salts in alternation
with minerals). Recent legislative enactments, giving the
cultivator of the soil a claim for the manure ingredients
possessing a pecuniary value, which he has applied and left
in the land, add greatly to the interest of this and allied
investigations. The difference between the plots lOa and 106
is that lOo. received one dressing of minerals followed by
thirty-nine dressings of salts of ammonia, whereas 106
received three dressings of minerals in the course of the
first seven years, since when both plots have been treated
exactly alike. Table VI. indicates (1) the produce of each
plot in each of the first eight years ; (2) the average produce
over succeeding periods of eight years each ; (3) the average
produce over thirty-two years. To be precise, it should be
stated that 106 received minerals alone in 1844 and 1850,
was unmanured in 1846, received minerals with ammonia-

The Rothamsted Wheat Experiments.

salts in 1848, and ammonia-salts alone (like 10a) in 1845,
1847, 1849, 1851, and in each of the last thirty -two years ;
10a received mineral manure alone in 1844, and ammonia-
salts alone in each of the last thirty-nine years.

TABLE VI. — AMMONIA-SALTS ALONE.

Dressed grain,

Total produce (grain
and straw).

Plot 10a.

Plot 106.

Plot 10a.

Plot 106.

1844 . .

Bushels.
15*
81|
271

25*
19i
32§
27

28|

Bashels.
15*

81|

17|
25*
25£
32f
18
28|

Ib.
2120
6246
4094
4593
3701
4992
4810
5036

Ib.
2120
6246
2671
4579
4530
5117
3120
4985

1845

1846 .

1847 .. ..

1848

1849

1850 .. .

1851 ..

8 years, 1852—1859 ...
„ 1860—1867 ...
„ 1868-1875 ...
„ 1876—1883 ...

32 years, 1852—1883 ...

22?
24
19
16£

27*
27±
20*
18*

4055
4076
3060
2618

4885
4563
3264
2935

20*

23i

3452

3912

Of the period preceding 1852, in the season of 1848, lOb
received mineral manures, as well as ammonia-salts, and
yielded 25J bushels per acre, as against 19| bushels on 100.,
which received the same ammonia- salts without minerals.
On the other hand, when 106, in 1850, received minerals
alone, the produce was only 18 bushels per acre, as
against 27 bushels obtained on 10a, manured with ammonia-
salts only. Though, in the periods of eight years, the
produce of 10& is always in excess of that of 10a, the
difference is nevertheless seen to be a declining one. As
the drainage water from cultivated fields is known to contain
but a very small amount of potash, and frequently no

Ammonia-salts and Minerals in Alternation. 43

phosphoric acid, there is no difficulty in tracing the increased
produce obtained on 106 over 10a to the minerals applied to
the former in 1848 and 1850. These large applications of
potash and phosphoric acid, although in the form of soluble
compounds, appear to enter into very fixed combinations,
somewhat similar to those already existing in the soil, and in
this respect they differ altogether from compounds of nitric
acid and ammonia, as the latter appear to be either washed
away or destroyed, unless they are fixed by vegetation, whilst
the former are fixed by the soil itself, and are only taken out
of it by means of vegetation.

CONTINUOUS GEOWTH OF WHEAT UPON THE
SAME LAND WITH AMMONIA-SALTS AND
MINERAL MANUEES IN ALTERNATING YEAES.

The ' experiments on plots lOo- and 106 show that potash
and phosphoric acid were still producing an influence upon
the wheat crop thirty-three years after their application.
Turning now to plots 17 and 18, it will be possible to trace
the unexhausted residue of another substance perfectly
soluble in water namely, salts of ammonia. On these plots
the mineral manures and the ammonia-salts are never used
together. When plot 17 receives minerals, plot 18 receives
ammonia- salts, and when plot 18 receives minerals, plot 17
receives ammonia-salts. Therefore, during the thirty-two
years, each plot received sixteen applications of mineral
manures and sixteen applications of ammonia- salts. For the
crops of the first eight years (1844-1851) the two plots received
different artificial manures, yielding a very similar produce.
Table VII. records the average produce of the mineral
manures, and also of the ammonia- salts over four periods of
eight years each, and, for comparison, the average produce
on plot 5, receiving minerals alone, is given.

The bottom line, giving the average of the entire thirty-

44

The Rothamsted Wheat Experiments.

two years, is wonderfully instructive. It shows that during
the sixteen alternate seasons in which plot 17 received
ammonia-salts, and the sixteen alternate seasons in which
plot 18 received that substance, the average produce was
30 bushels per acre, whilst during the alternate years, in
which the plots received minerals only, they yielded but 15f
bushels; or, in other words, only a fraction more than
plot 5, which received no ammonia-salts during the whole
period. It is estimated that the 4001b. of ammonia-salts
applied per acre contain 861b. of nitrogen, but the resources

TABLE "VTL — AMMONIA-SALTS COMPARED WITH MINERALS.

Dressed grain per acre.

Total produce (grain and
straw) per acre.

life

Mineral
manure*.
Plots 17 or 18.

Ammonia-
salts.
Plots 17 or 18.

liSs

-.-•r

K

ll

8 years, 1852-1859...
„ 1860-1867...
„ 1868-1875...
„ 1876-1883...

32 years, 1852-1883...

Bushels.
19

144
12f

Bushels.
15

Bushels.
32$

3H

28*

lb.
3191
2450
2144
1899

3235
2696
2404
1869

lb.
5938
5297

4781
4930

is*

m

30

2421

2551

5237

of the soil were evidently competent to furnish the nitrogen
contained in 15 bushels of wheat and its straw, inasmuch as
plot 5, receiving no ammonia, gave that produce. In the
additional 15 bushels and its straw obtained by the application
of the ammonia-salts, certainly less than 261b. of nitrogen
are carried off, thus leaving 601b. of nitrogen per acre to be
accounted for. Although analysis proved that, in 1881,
the soil of plots 17 and 18 contained rather more total
nitrogen and nitrates than plot 5, still, to a depth of as much
as 27in. there was no evidence of the existence in the soil of

Ammonia-salts versus Nitrate of Soda. 45

the large amount of nitrogen supplied in the manure, and
not accounted for in the crop. Hence, when ammonia-salts
are applied to grow wheat, it is not safe to calculate upon
any of the unexhausted residue being available for the
purpose of growing a second corn crop. The evidence also
indicates that the exhausting character which farmers attri-
bute to corn crops is quite as much due to the nitrogen
which they do not assimilate being washed out of the soil, as
it is to the amount of that substance which is removed in the
produce.

COMPARISON BETWEEN AMMONIA-SALTS AND
NITRATE OF SODA AS SOURCES OF NITROGEN.

One other series of experiments with artificial manures
remains to be discussed. The results already considered are
those of experiments in which mixed minerals alone, or salts
of ammonia alone, are employed. It is necessary now
to notice those in which one uniform quantity of mixed
minerals is used in each case, but with different amounts of
nitrogen in the form of ammonia- salts, and also as nitrate of
soda. The nitrogenous applications were as follows : Plots
6a and 6fc, 2001b. of sulphate and muriate of ammonia, con-
taining 431b. of nitrogen ; plots 7a and 76, 4001b. of the same
salts, containing 861b. of nitrogen ; and plots 8a and 86, 6001b.
of the same salts, containing 1291b. of nitrogen; whilst
plot 9a received 861b. of nitrogen as nitrate of soda, instead
of as ammonia-salts. Table VIII. is drawn up on the same
principle as the preceding tables, and similarly records the
result in four periods of eight years each.

It is apparent that in the separate periods of eight years
each, and also in the whole period of thirty-two years, the
increase of wheat obtained by adding 431b. of nitrogen
(plot 6) to the minerals varies from 8 to 11 bushels per acre,
the total increase over the whole period being not quite

46

The Rothamsted Wheat Experiments.

9 bushels per acre. The addition of another 431b. of nitrogen
(plot 7) again increases the amount of wheat by between
8 and 9 bushels over the whole period, as compared with
plot 6, receiving only half as much nitrogen. The addition

TABLE VIII. — AMMONIA-SALTS COMPARED WITH NITRATE OF SODA.
Dressed Grain, per Acre — bushels.

Mix, miu.

Mixed mineral manures and

manures

Mixed
minerals

ammonia-salts.

<fc nitrate
of soda.

alone.

Plot 5.

= 43lb.

= 861b.

= 1291b.

= 861b.

nitrogen.
Plot 6.

nitrogen.
Plot 7.

nitrogen.
Plot 8.

nitrogen.
Plot 9.

8 years, 1852—1859 ...

19

27$

35*

36J

31*

„ 1860—1867 ...

15*

26i

36±

39f

40*

„ 1868—1875 ...

14

22

31

36

39

„ 1876—1883 ...

12f

20|

28

32*

34-}

32 years, 1852—1883 ...

15±

24J

32*

36*

36i

Total Produce (Grain and Straiu) per Acre — Ibs.

Mixed
minerals
alone.
Plot 5.

Mixed mineral manures and
ammonia-salts.

Mix. inin.
manures
it nitrate
of soda.

= 86Ib.
nitrogen.
Plot 9.

= 431b.
nitrogen.
Plot 6.

= 86lb.
nitrogen.
Plot 7.

= 1291b.
nitrogen.
Plot 8.

8 years, 1852—1859 ...
1860—1867 ...
„ 1868—1875 ...
„ 1876—1883 ...

32 years, 1852—1883 . . .

3191
2450
2144
1899

4808
4276
3612
3422

6490
6262
5379
5248

7012
7363
6593
6361

5897
7862
7344
6824

2421

4029

5845

6832

6982

of a third 431b. of nitrogen (plot 8) proved far less effective
than the previous additions, as, instead of being from 8 to 9
bushels, the average increase was only 3| bushels per acre ;
the average of the whole period being, for the 1291b. nitrogen.

The Profitable Limit of Growth. 47

36J bushels per acre, whilst where 861b. of nitrogen was
applied it was only 3| bushels less. The total increase
obtained by this large amount of nitrogen (1291b.), in excess
of the produce obtained by minerals alone, amounted only
to 21 bushels per acre per annum ; hence it is clear that in
this experiment the nitrogen applied was in excess of the
quantity which could be utilised by the crop ; and even in
the best season of growth, 1863, when plot 7, receiving 861b.
of nitrogen, yielded 53-|- bushels per acre, the plot receiving
1291b. only yielded 2 bushels more !

Some important practical inferences are now deducible.
Sir John Lawes and Dr. Gilbert estimate the average yield
of wheat in Great Britain at 28 bushels, while others put it
at 30 bushels per acre. The crop obtained by 861b. of
nitrogen appears to have quite reached, if it has not exceeded,
the profitable limit of growth, and only a rise in the price of
wheat — and not lower prices — could justify the outlay in
manure which would be required to grow a larger crop.
Much of the nitrogen of the ammonia-salts has been con-
verted into nitric acid, and washed into the drains, chiefly
during the winter. At the time of applying the nitrate of
soda in the spring, the plots receiving ammonia-salts in the
autumn had already lost more or less nitrogen ; and until
1878 the trial of the relative crop -producing power of
nitrogen as ammonia, and as nitric acid in the form of
nitrate of soda, was not carried out on equal terms. For
the crops of 1878-83 plots 7a and 7b received the ammonia-
salts in the spring, whilst another plot similarly manured
received these salts in the autumn ; and, although the spring-
sown ammonia has given the largest produce, still, the
difference between the two crops is by no means what might
have been expected from the known loss by drainage which
took place when the manure was autumn- sown, so that the
question arises whether there has been loss by destruction
and evolution of free nitrogen, or whether any considerable
amount of the ammonia has remained unnitrified in the soil,

48 The Rothamsted Wheat Experiments.

which would be in a much drier state in the spring and
summer than during the autumn and winter.

From the table it may be seen that it was not until after
the experiment had been carried on for eight years that the
spring-sown nitrate showed a superiority over the autumn-
sown ammonia-salts ; but during the first three years there
were no minerals, and there was much less nitrate, applied to
the nitrate plot (9a). During the next three periods of eight
years each, the produce by the nitrate was considerably in
excess ; though in six of the last eight years the ammonia-
salts were applied in the spring. Over the thirty -two years
the nitrate gave an increase over the ammonia-salts, con-
taining an equal quantity of nitrogen, of 3| bushels per acre
per annum (36 J bushels as against 32|). Another note-
worthy feature is, that the spring-sown nitrate supplying
861b. of nitrogen gives exactly the same total average yield
(36^ bushels) as the chiefly autumn-sown ammonia-salts
supplying 1291b. ; and, taking the total produce (grain and
straw), which is by far the best measure of the power of a
manure to promote growth, it is seen that the 861b. of
nitrogen as nitrate of soda produced 69821b., whilst the 1291b.
of nitrogen as ammonia-salts produced 68321b. Although
there is often a strong prejudice against the use of nitrate,
this experiment shows that, when judiciously applied, its
properties as a manure are superior to those of ammonia-
salts. When used continuously for thirty-two years, its power
to produce growth appears to increase rather than to diminish ;
as during the last sixteen years the total produce per acre
per annum of plot 9a in excess of the produce of plot 7 — each
receiving the same amount of nitrogen — has been 70841b. as
against 53131b., a difference of 17711b. in favour of the
nitrate.

Continuous Growth of Wheat with Farmyard Manure. 49

CONTINUOUS GROWTH OF WHEAT UPON THE
SAME LAND FOE FOETY YEAES WITH FAEM-
YAED MANUEE SUPPLIED EACH YEAE.

When a plot of land which has grown wheat continuously
for forty years has every year received a dressing of farm-
yard manure, the lessons to be learnt from an examination of
the results cannot but commend themselves as eminently
deserving the attention of agriculturists. Plot 2 in the
experimental wheat-field at Eothamsted is such a plot, and for
the last forty years it has annually received ordinary farmyard
manure at the rate of 14 tons per acre. No attempt was
made to ascertain the exact composition of the manure
applied, for two reasons : (1) in the earlier years of the
experiments, the importance of knowing the exact amount
of the various ingredients applied to the soil was not well
understood; (2) later on, the uncertainty of the results
obtained in a very careful attempt to ascertain the exact
composition of some box-dung at Woburn made it apparent
that it would be preferable to trust to an estimate of the
composition based on numerous analyses of farmyard
manure. Table IX. gives the estimated amount of some of
the more important ingredients supplied to the soil of plot 2
annually in the 14 tons farmyard manure, as also the amount
of ingredients supplied in the artificial manure on plot 7,
which receives mixed minerals and 4001b. of ammonia- salts
per acre.

TABLE IX.

Dry organic
matter.

Nitrogen.

Phosphate
of lime.

Potash.

lb.

lb.

lb.

lb.

Farmyard manure

, plot 2

8540

200

155

168

Artificial manure,

plot 7

—

86

140

100*

1501b. 7 years, 1852-8, lOOlb. each year since.

50

The Rothamsted Wheat Experiments.

Although the ingredients furnished to the two plots differ
widely, the average produce is nevertheless almost the same
on each, as is clearly shown in Table X.

TABLE X. — FARMYARD MANURE (Plot 2) AND ARTIFICIAL MANURE
(Plot 7) COMPARED.

Dressed grain per acre—

bushels.

Total produce (grain and
straw) per acre— Ibs.

Plot 2.

Plot?.

Plot 2.

Plot 7.

8 years, 1852-1859
1860-1867 ...
1868-1875
1876-1883

34f
35*
35±

28|

35*
36i
31

28

6100
5926
5932

4798

6490
6262
5379
5248

32 years, 1852-1883 33*

32$

5689

5845

40 years, 1844 - 1883, } 32,
with farmyard dung >

—

5516

—

Taking the whole thirty-two years, the difference in the
average produce per acre per annum is seen to be less than
one bushel of grain — in fact, only three-quarters of a bushel.
Of total produce, grain and straw, the mixed minerals and
ammonia-salts on plot 7 gave an increase over the farmyard
manure on plot 2, over the same period, of 1561b. per acre
per annum. The most striking contrast between these two
manures, which furnish almost identical results, is that whilst
the farmyard manure supplies to the soil a large amount of
organic matter, the artificial manure supplies none. Each
year plot 2 received in the farmyard manure about 85401b.
of organic matter, whilst the crop grown upon it would not
contain more than one-half this quantity. Yet the artificial
manures on plot 7, supplying no organic matter whatever,
produced a crop which contained rather the larger amount of
organic matter of the two ; and it can be shown that, by
merely increasing the amount of nitrogen, a still larger
amount of organic matter can be obtained in the crop. Thus
plot 8, receiving 1291b. of nitrogen per annum (still much
less than that supplied in the farmyard manure) gave, over

Nitrogen in Different States of Combination. 51

a period of thirty-two years, a total produce, grain and
straw, of 68321b. per acre, or 11431b. more than that
obtained by the farmyard manure on plot 2. Hence, it is
concluded, the amount of non-nitrogenous (chiefly carbona-
ceous) organic matter in the crop bears no relation to that
supplied in the manure ; the atmosphere, and not the soil, is
the source of this supply.

In the two years 1863 and 1864 the farmyard manure
applied to plot 2 was estimated to furnish 4001b. of nitrogen,
and the total produce was 13,6531b. ; whilst that of plot 16,
which received only 3441b. of nitrogen, in ammonia- salts, was
19,8731b. There were, therefore, in the two years, 62201b.
excess of crop on the artificially manured plot. The nitrogen,
then, of farmyard manure and that of ammonia- salts, or of
nitrate of soda, must obviously be in different states of
chemical combination. This indeed is so, for in the dung the
nitrogen is mostly combined with carbon, in which form it is
both insoluble and inactive, and, for most crops, at any rate,
it only becomes available when, by the process of nitrification,
it ceases to be in combination with carbon, and passes into
the form of a soluble nitrate. Nitrate of soda contains
nitrogen in this form, and this is the reason it is so imme-
diately available to a growing crop, and becomes speedily lost
in the drainage water if there is not a growing crop ready to
make use of it. The time required for the nitrification of
different portions of farmyard manure will be very variable ;
the carbon may be separated from the nitrogen in urine in a
very short time, whilst it may take many years to nitrify
portions of the nitrogenous organic matter of straw, espe-
cially on heavy land. Since it requires a considerably larger
application of nitrogen in the form of farmyard manure to
grow the same amount of crop as that produced by a much
smaller application of nitrogen in the more active form of
ammonia- salts, or nitrates, it follows that in the soil where
dung has been employed there ought to be found a larger
amount both of carbon and of nitrogen than in the soil where

E 2

52 The Rothamsted Wheat Experiments.

artificial manures have been used. It is indeed an ascertained
fact that, to accumulate nitrogen in the soil, it must be in
combination with carbon, though very little is yet known in
regard to the various compounds of carbon existing in the
soil. But, respecting the subject now under discussion, the
evidence as to the underground fertility of plots 2, 3, and 7, as
brought out in the analyses of their soils to various depths,
is both interesting and instructive. It is estimated that,
within 27in. from the surface, the nitrogen on plot 2 (farm-
yard manure) will amount to more than 80001b. per acre, and
that it would exceed the amount on plot 7 (mixed minerals
and ammonia-salts) to the same depths by more than 17001b.,
and that on plot 3 (permanently unmanured) by more than
22001b. By far the largest amount of this difference is found
in the first 9in. of depth ; to that depth the dunged land con-
tained nearly double the amount of nitrogen which is found
on the unmanured plot, and more than one and a half times
as much as is found on the plot receiving minerals and
ammonia-salts ; and it is estimated that plot 3, after the
removal of forty unmanured crops of wheat in succession,
still contains about 24001b. of nitrogen per acre in the first
9iu. from the surface ; this, in fact, represents the residue of
the natural fertility, or, to use a term imported into the
Agricultural Holdings Act, the inherent capability of the soil.
The relation between the carbon and nitrogen in these
three soils, which differ so greatly in their total amounts of
nitrogen, indicates that they do not differ much in their
character. On the unmanured land there is not quite ten of
carbon to one of nitrogen ; on plot 7, receiving minerals and
ammonia-salts, it is ten and a half to one ; on the dunged
land it is not quite twelve to one. Yet the unmanured plot
has received neither carbon nor nitrogen in manure ; plot 7
has received a very large amount of nitrogen, but no carbon ;
whilst the dunged plot has received a very large amount of
both carbon and nitrogen. The relation of carbon to nitrogen
in the farmyard manure is about twenty to one — a proportion

Fertility of Soils. 53

totally different from that in the soil. The close relation
between the carbon and nitrogen, in the soils of plots 3 and 7,
indicates that the larger amount of nitrogen in plot 7 is not
due to the direct storing up of ammonia by the soil, but to the
nitrogen forming part of the vegetable growth, and being thus
stored up in the stubble and roots. If the nitrogen of the
ammonia-salts had been stored up in any form except that
of vegetable growth, the relation of carbon to nitrogen
would have been lower on plot 7 than on plot 3, instead
of which it is higher; there is also proof founded on
analyses of the soils of plot 7 and plot 3 that, of the two,
the latter contains by far the larger amount of unexhausted
fertility.

The adjoining field, where barley is grown continuously,
furnishes similar evidence. For twenty consecutive years,
14 tons per acre of farmyard manure were applied to one plot
of barley, after which period the plot was divided into two,
the dung being continued on the one half, and stopped on
the other. Up to 1883, twelve unmanured crops had been
taken, yielding an average of 34J bushels per acre, and, as
the last of these crops, in the rather favourable season of
1883, exceeded 35 bushels, there is evidence of a long future
before the fertility due to the residue of the twenty years'
application of dung will be exhausted. In the same field the
plots manured with rapecake yield on analysis a considerably
larger amount of nitrogen than any of the plots where
minerals, or minerals with ammonia-salts or nitrate, have
been used. It follows that whilst fertility may be stored up
in the soil, in the form of such mineral substances as potash
or phosphate, it does not appear that the more valuable
substance, nitrogen, can be stored up unless combined with
carbon ; in other words, whilst the soil fixes potash and phos-
phoric acid independently of vegetation, nitrogen is only
fixed by the agency of vegetation.

54 The Rothamsted Wheat Experiments.

QUANTITY OF AMMONIA REQUIRED TO PRODUCE
AN INCREASE OF ONE BUSHEL PER ACRE IN
THE WHEAT CROP.

One important point remains for discussion. It con-
stitutes the subject of the fourth section of the report of the
first twenty years' experiments, and deals with the amount of
increased produce obtained for a given amount of ammonia
supplied in manure. The data are furnished in Table XI.,
which shows the annual average amount of ammonia in
manure (or of nitrogen as nitrate, reckoned as ammonia)
that was required for the production of one bushel (of 601b.)
increase of wheat grain, with its proportion of straw, on
the most important plots, in two periods of six years each,
and in the total period of twelve years ; in two or three
cases there were some modifications made in the manures,
but, to avoid complicating the table, these are not introduced.

Long before the expiration of the first twenty years of
the wheat experiments, the investigators published their
opinion, founded on experiment, that the grower might
assume, for practical purposes, that he would, on the average
of seasons, get one bushel of wheat and its proportion of
straw, beyond the produce of the soil and season, for each
51b. of ammonia applied in the manure for the crop. The
results shown in Table XI. confirm this opinion. 501b. of
ammonia, or its equivalent of nitrogen, would be supplied in
rather under 2cwt. of commercial sulphate of ammonia, or
in Ifcwt. of commercial muriate of ammonia, or in about
2|cwt. of genuine Peruvian guano, or in rather more than
2|cwt. of nitrate of soda. These amounts are more than are
usually employed in common practice for the wheat crop, and
most growers would consider double these quantities to be
very heavy, if not excessive, dressings. Assuming that the
results obtained by the use, per acre, of 501b. or lOOlb. of
ammonia (or their equivalent of nitrogen in nitrate of soda)

Ratio of Increase in Crop to Ammonia Supplied. 55

most nearly represent those which may be expected in
ordinary practice, and, further, that the results obtained by
these amounts in the cases -where the mineral constituents
(unless silica) are not in relative defect are also such as are
most likely to be obtained in ordinary farm practice, the
authors subject the tabulated results to a thorough exami-
nation. The figures show how very striking is the difference

TABLE XI.

Manures per acre per annum, for 12 years,
1852-1863.

Average quantity of

ammonia required to pro-
duce one bushel (=601b.)

Plots.

increase in wheat crop.

First

Second

The

6 years.

6 years.

12 years

1852-57.

1858-63.

1852-63.

lb.

lb.

lb.

6

2001b.

amm.-salts, and mixed min. manure

5-59

4-36

4-86

7

4001k

» >5 »

5-86

4-95

5-37

8

6001b.

J> J> 5>

8-01

6-80

7-35

16

SOOlb.

»5 5> 55

9-98

9-00

9-47

m

or V
18)

4001b.

amm.-salts, in alternation with ")
mixed mineral manure ... )

6-92

6-47

6-69

10a

4001b.

amm.-salts alone (19 yrs. 1845-63)

20-52

22-74

21-57

106

4001b.

„ „ (13 yrs. 1851-63) 10*39

12-16

11-20

11

4001b.

„ andsuper-phosph.of lime 8*92

8-25

8-57

"{

4001b.

, , and sup er-phosph. , and ^
sulphate of soda . . . j

6-39

5-24

5-76

»{

4001b.

„ andsuper-phosph.,and~) fi -o
sulphate of potash., j

5-27

5-85

»{

4001b.

„ and super-phosph., and ")
sulphate of magnesia j

6-31

5-31

5-77

9a

5501b.

nitrate of soda and mixed min. man.

6-59

4-73

5-41

9&

5501b.

„ alone ... .

11-29

14-73

12-80

of effect upon the immediate increase of a given amount of
nitrogen in manure, whether used in the form of ammonia-
salts or of nitrate, according to the available supply of
mineral constituents in the soil. With the overwhelming
evidence before him which such a comprehensive summary of
experimental results on the point affords, the grower cannot
fail to see that he not only very injuriously further reduces

56 The Rothamsted Wheat Experiments.

his immediately available supply of mineral constituents, but
also pays very dearly for his increase, if he tries to obtain it
by means of purely nitrogenous manures when his soil is
already unduly exhausted of mineral constituents.

The above table is taken from a more comprehensive one,
which also gives the result on each plot for every one of the 12
years. As showing the difference of effect of a given amount of
ammonia in one season as compared with another, the results
on plot 6, receiving 501b. of ammonia (in 2001b. of ammonia-
salts) each year, and always with mixed mineral manure, are
instructive. Taking the average of the 12 years, it required
4*861b. of ammonia in manure to yield 601b. of increase of
grain and its proportion of straw ; whereas, in the remarkably
productive season of 1863, it required only 2*421b., but in
1853, 7'131b. ; in 1860, 8'851b. ; and in 1852, 12'451b.

The conclusion on the point under notice is that, great as
is the difference of effect of a given quantity of ammonia,
according to the amount applied per acre, to the mineral con-
dition of the soil, and to the season, still, when only moderate
quantities were used, when there was a sufficient supply of
mineral constituents, and taking the average of many seasons
— that is, under the conditions the most comparable with
those of the average of common practice, — the result was in
marked accordance with the early estimate, that almost
exactly 51b. of ammonia were required to be expended to
obtain an increase of one bushel of wheat grain and its pro-
portion of straw.

THE EOTHAMSTED EXPEEIMENTS ON THE CON-
TINUOUS GEOWTH OF WHEAT COMPAEED
WITH THOSE AT WOBUEN, HOLKHAM, AND
EODMEESHAM.

It may fairly be objected that these experiments on the
continuous growth of wheat furnish results true of one

Results in Different Localities Compared.

57

description of soil and of one locality only. Table XII.,
however, summarises the results of experiments made on
very various soils, under various conditions due to previous
treatment, and in various seasons, notwithstanding which the
general characters of the results are seen to be accordant.

TABLE XII. — RESULTS OF EXPERIMENTS ON THE GROWTH OF WHEAT
BY DIFFERENT MANURES, ON DIFFERENT SOILS, IN DIFFERENT
LOCALITIES, AND IN DIFFERENT SEASONS.

Dressed grain per acre— bushels. Average annual
results.

Manures.

Bothamsted, Herts.

Woburn,
Beds.

Hoik-
ham,
Norfolk.

Bodmer-
sham,
Kent.

8 years, 1856-63.

32 years,
. 1852-83.

7 years,
1877-83.

3 years,
1852-54.

4 years,
1856-59.

Broad-
balk
Field.

Hoos
Field.

Broad-
balk
Field.

Unmanured
Mix. min. man. alone ...
Amm.-salts alone (86lb.
nitrogen)

16
19

23i

38|

15
16|

26i

37|

13i

15i

20|
32f

15!
16!

23f*
37f

18
19i

27i
32f

25|

28!

31!
33^

Mix. min. man. and amm.
salts (861b. nitrogen).

* By amm.-salts = only 431b. of nitrogen.

Thus, not only is there general agreement in the character
of the results in different localities, when the averages of a
number of years are taken, but the non-effect of the residue
from previous applications of ammonia-salts is as marked in
the sandy soil at Woburn as in the very different loamy soil
at Eothamsted. This latter point is illustrated in Table XIII. ,
which shows the produce of wheat grown year after year on
the same land in Stackyard Field, Woburn.

This field received mineral manure and ammonia- salts
(= 861b. of nitrogen) for five successive years, in the year
following which one portion again received the same manure,

58 The Rothamsted Wheat Experiments.

and the other the mixed mineral manure alone. In the
next year, 1883, the portion which received the whole manure
in the previous year received mineral manure only ; whilst
the other portion received both minerals and ammonia-salts.
Tt is seen that in each year, 1882 and 1883, the nitrogenous-
manured portion yielded large crops (43^ and 45f bushels) ;
whereas, the portion on which mineral manures alone
followed ammonia-salts and large crops, yielded very small

TABLE XIII. — WHEAT GROWN YEAR AFTER YEAR ON THE SAME
LAND, STACKYARD FIELD, WOBURN.

Harvests.

Dressed grain, bushels.

1877

431

1878

27

1879 '.

1880

28*

1881

1882

(1) 13i (2) 43^

1883

(2) 455 (1) 17±

(1) Mixed min. man. alono. (2) Mixed min. man. and amm. -salts. (= 861b. nitrogen.)

crops — 13 £ and 17^ bushels respectively, against 14f and
17 1- bushels on a plot where the same mineral manures were
used year after year. There was, then, no available and
effective residue where the ammonia-salts had been previously
applied. Though in 1884 there was notable effect from
unexhausted residue of nitrogenous manure, this probably
arose from there being very little rain, and, consequently,
very little loss by drainage during the winter of 1883-4.

SUMMAEY OF THE EESULTS OF FORTY YEARS'
EXPERIMENTS ON THE CONTINUOUS GROWTH
OF WHEAT UPON THE SAME LAND.

The forty years' experiments on the continuous growth of
wheat have established, with regard to the soil, the following

Summary of Results. 59

facts. An unmanured soil has yielded forty successive crops,
averaging 14 bushels per acre, solely by means of its existing
fertility. At the outset the soil contained a large amount of
organic nitrogen, derived from previous vegetation ; it like-
wise contained a large amount of mineral food. Every
year a portion of the organic nitrogen has been nitrified by
the agency of organisms living in the soil ; part of the
resulting nitrates has contributed to the growth of the
wheat crop, and part has been washed out of the soil or other-
wise lost. This loss of nitric acid is greater in wet seasons,
and the amount taken up by the wheat crop is consequently
smaller ; so that comparatively dry seasons should be favour-
able to the production of large wheat crops. The stock of
soil fertility in the form of organic nitrogen has been
considerably reduced, and the stock of both potash and
phosphoric acid has likewise been largely reduced. Never-
theless, the stock of fertility that remains would appear to
be sufficient to grow crops of wheat for a very long period,
though the produce must necessarily lessen in course of
time.

With regard to manures, minerals alone have added very
slightly to the unmanured produce, whereas manures con-
taining nitric acid alone, or some easily nitrifiable compound
of nitrogen, have considerably increased the crop ; hence the
soil had a stock of minerals which the crop was unable to
utilise, on account of the insufficient supply of available
nitrogen. Manures consisting of potash, phosphoric acid, and
nitrogen (as ammonia-salts, or as nitrates), appear competent
to grow large crops of wheat continuously. In the ordinary
course of agriculture with rotation, as practised in this
country, the supply of mineral constituents immediately
available for the wheat crop is almost invariably in excess,
relatively, to the immediately available supply of nitrogen
from the atmosphere, or the accumulated stores within the
soil itself. Furthermore, with few exceptions, the worse the
so-called " condition " of the land — that is, the more it is in

60 The Rothamsted Wheat Experiments.

the agricultural sense exhausted — the more striking would
be the effect of exclusively nitrogenous as compared with
that of exclusively mineral manures. A given weight
of nitrogen, as nitric acid (or nitrate), has produced
more growth than the same weight of nitrogen as ammonia-
salts. The amount of nitrogen supplied in the manures is
very much in excess of the amount recovered in the increase
of the crops ; and, after a certain amount of growth has been
reached, each increase of crop requires a proportionately
larger application of manure. When the price of wheat is
high, larger crops can be grown more profitably than when
the price is low. In the form of farmyard manure, a con-
siderably larger amount of nitrogen is necessary to produce
a given increase of crop; but, though a given weight of
nitrogen, in the form of nitric acid, will produce more
growth than the same weight of nitrogen in dung, the
influence of the nitrate upon succeeding crops will be very
much less. There is no evidence to show whether the whole
available effect of the nitrogen in one manure is greater than
it is in the other.

Finally, with regard to unexhausted manures, it appears
that, in the absence of vegetation, or when applied to crops in
excess of their requirements, both potash and phosphoric
acid form insoluble compounds with the soil, and become
available for future crops. Under similar circumstances,
nitrates and salts of ammonia do not appear to form
permanent compounds with the soil, but are liable to be
washed out by rain, or to be otherwise lost. The applica-
tion of more nitrogen, as nitrates or salts of ammonia, than
the crop can utilise, does not appear to interfere with the
nitrification of the organic nitrogen of the soil. The stock
of nitrogen of the soil itself, therefore, may be reduced,
although the annual application of nitrogen may be much in
excess of the amount of that substance removed in the crop.
When large wheat crops have been grown by the use of
nitrates or salts of ammonia, with mineral manures, the

Accumulation of Fertility. 61

soil does not appear to have gained or lost fertility. Nitrifi-
cation may have gone on as usual, but the loss has been
made good by the amount of nitrogen stored up in the
stubble and underground portions of the large crops so
grown. When dung is applied continuously, the accumula-
tion of unexhausted fertility becomes very large, and the
removal by crops of the substance accumulated would
extend over a long series of years ; therefore, dung applied
to land in the ordinary process of agriculture will not be
entirely exhausted until a considerable number of years from
the time of its application.

62 The Eothamsted Wheat Experiments.

IV.-INFLUENCE OF CLIMATE ON THE
CULTIVATION OF WHEAT.

Climate and Our Wheat Crops" is a title so suggestive
of important agricultural and economic problems, that it is
hardly surprising it should have been made the subject of an
essay (E. A. S. Journal, 1880), in which Messrs. Lawes and
Gilbert sought to trace a connection between the general
character of the seasons and the amount of their respective
wheat crops. It has been remarked that, so far as climate is
concerned, the British Isles are outside the zone favourable
to the growth of wheat, and that its successful cultivation is
due to the skill of the farmer in contending against adverse
meteorological conditions ; that, however, the decline in the
wheat area cannot be attributed to any general change for the
worse in the characters of the British climate, is proved by
the circumstance that, by dividing the 108 years ending with
1880, into six periods of eighteen years each, there is even a
slight progressive increase of mean temperature from the
first to the last of these six periods (Mr. Glaisher). It is to
the greatly increased production of wheat in other countries
at a lower cost than in our own, and to low rates of transport,
by which it is brought into our markets in quantity and
at a price much reducing the value of home produce, that the
lessened area under the crop is chiefly to be attributed.

As only about 5 per cent, of the total wheat crop is
derived from the soil itself, the remainder coming, directly
or indirectly, from the atmosphere ; and as the amount of
matter accumulated from either source depends mainly on
the quantity, and the relations to one another, of heat and

Temperature and Rainfall. 63

moisture, the preponderating influence which the character of
the seasons exercises over the growth of our crops is no more
than might be expected. Though the connection between
meteorological phenomena and the progress of vegetation is
not yet so clearly comprehended as to enable us to estimate
with any accuracy the yield of a crop by studying the
statistics of the weather during the period of its growth, the
present attempt is, nevertheless, a valuable contribution
towards such an object.

Temperature and rainfall are the chief factors demanding
attention, but it is necessary to distinguish between the
total heat and the total rainfall of a season, and the distribu-
tion of temperature and of rainfall over the different periods
of the season. It is obvious that a given amount of rain
equally distributed through the spring and summer in each
of two seasons, will have a very widely different effect on
vegetation in the two cases, if the one season should be a hot
and the other a cold one. On the other hand, if the temper-
ature of the two seasons be the same, but the rainfall very
different, so also will the effect on vegetation be very different.
The defect of our climate for the production of wheat is
apparently more connected with an excess of moisture than
with a deficiency of heat, during the periods of active growth
and maturing. It is, in fact, when a cold season, or one of
only moderate temperature, is accompanied by an excess of
rain, that the yield of our wheat crops is the most defective.

During the present century, each of the following years
was more or less remarkable so far as the growth of the
wheat crop is concerned; 1816, 1832, 1833, 1834, 1835, 1853,
1854, 1857, 1860, 1863, 1864, 1868, 1870, and 1879. Of
these, 1816 and 1879 have the reputation of yielding the
two worst wheat crops within the period included by those
dates ; on the other hand, 1834 is generally referred to as
one of the most productive years of the century. Four
seasons of reputed very high productiveness which occurred
before the commencement of the Rothamsted experiments,

64 The Rothamsted Wheat Experiments.

were those of 1832-3-4-5. So abundant were these four
wheat crops, that the average price, even under protection,
went down from 54s. 5d. per quarter over the first harvest
year, to 49s. 9d. over the second, 41s. 5d. over the third, and
42s. 8d. over the fourth. The lowest price reached was 36s.
per quarter in the last week of 1835, and the first of 1836.
And such was the distress suffered by the agricultural
interest, as the result of abundant wheat crops, and the low
prices following, that select committees of both Houses of
Parliament were appointed to inquire into the matter; in
1880 a Royal Commission inquired into the distress caused,
not by abundant, but by deficient crops and large importa-
tions, though by no means such low prices as during the
period of great abundance. The characters of some of the
remarkable seasons may now be noticed.

The abundant wheat crop of 1832 was grown under the
influence of a mild and rather wet winter, a spring of
moderate characters, and a summer of only moderate
temperatures, and with heavy rains excepting in July. The
crop of 1833 was the result of a generally mild and
moderately wet winter and early spring, excepting January
and March, which were cold. The remainder of the spring
and the early summer were hot and mostly dry, and the rest
of the season upon the whole favourable. The result was a
high yielding, but not bulky, straw crop. The crop of wheat
of 1834 — one of the heaviest on record — was grown in a
season warmer than usual almost throughout, but especially
in the winter and in the spring, excepting April ; and, after
an excess of rain in the winter, there was a considerable
deficiency for four months to the end of May, again a
deficiency in June, but afterwards heavy rains, though with
high temperatures. The crop of 1835, which was of
extraordinary bulk and luxuriance, was grown in a season in
which the winter was as open and as much marked by an
absence of snow and frost as the three preceding winters ;
the spring was, upon the whole, favourable to the wheat

Seasons of High Produce. 65

crops, and the summer brilliantly fine till the last week in
June. At the end of June heavy rains and high winds laid
the crops ; but bright breezy weather in July stayed the
damage, though much did not ripen well. The rest of the
season was fine, and the wheat crops were got in in excellent
order ; but, though bulky, they were decidedly inferior in
both yield and quality to those of 1834. Generally, then,
these four consecutive seasons of abundant wheat crops were
characterised by mild and open winters, upon the whole mild
springs, and average, or warmer than average, summers,
especially the last two of the four. In each season there were
individual months of considerably more than average fall of
rain, sometimes earlier and sometimes later, and accordingly
influencing the bulk of the crop. But each season was
characterised by less than the average fall of rain during
several months of the growing period, and this was
particularly the case in the season of 1834, the one of the
most extraordinary productiveness.

Passing now to those seasons of high productiveness which
have occurred during the progress of the Rothamsted
experiments, the season of 1854 comes first in chronological
order. The crop, which was very abundant both in corn and
straw, was, after a severe winter period, grown under higher
than average temperatures during the earlier, but lower
during the later, periods of growth, and with much less than
the average fall of rain in every month from seed-time to
harvest, excepting in May and August. The heavy corn, but
not heavy straw, crop of 1857 was obtained under the influence
of about average winter and early spring, but high summer,
temperatures, and, as in other cases of high productiveness,
there was here again much less than the average fall of rain
from seed-time to harvest, the only months of any excess
being January, June, and August. The wheat crop of 1863
gave probably the highest average produce per acre over the
country at large since 1834 ; its season of growth was marked
by an extremely mild winter and early spring, with much less

p

66 The Rothamsted Wheat Experiments.

than the average fall of rain, so that the plant was brought
early forward. Then came, in the early summer, a consider-
able amount of rain, after which there was a deficiency up to
harvest. The temperature was only about the average in
June and July, conducing to continued luxuriance rather than
to early maturation ; whilst August, the harvest month, was
both warmer and drier than usual. The conditions were,
therefore, those of a lengthened and almost unbroken course
of gradual accumulation, with, finally, a favourable ripening
period. The crop of 1864 was produced under the influence
of warmer than average weather in early winter and in spring ;
only moderate, or even lower than average summer,
temperatures, but much less than the average fall of rain
from seed-time to harvest ; every month being considerably
deficient, excepting May, which was average, and March, in
which alone there was an excess. In the season of 1863 the
period of growth was almost throughout one of drought, with
high temperatures prevailing through spring and summer;
the result was a very early harvest, a not bulky but a high-
yielding crop on good and well-farmed soils, but a deficient
one on light and poorly-farmed land. After a by no means
favourable winter, followed by prolonged spring and summer
drought and heat, the wheat crop of 1870 was deficient in
straw, and also yielded less corn than that of 1868, but still
considerably more than the average, with a high proportion
of corn to straw, and high quality of grain. Generally
speaking, out of the six years of highest productiveness
throughout thirty-six seasons (1844-1879) of experiments,
the three which gave the highest produce of all (1863, 1864,
and 1854) were marked by generally higher than average
mean temperatures during the winter and early spring
(excepting the early winter of 1853-4, which was severe), but
generally only average, or lower than average, summer
temperatures. Indeed, June, 1854, was colder than June of
the disastrous year 1879. Each was also characterised by
very much less than the average fall of rain from seed-time to

Seasons of Low Produce. 67

harvest, there being in no case an excess in more than two
out of the nine months from November to July inclusive.
The other three seasons of high productiveness (1857, 1868,
and 1870), though they gave less corn than the foregoing,
and very much less straw, were, nevertheless, seasons of
considerably more than average produce of corn, and of high
quality of grain. These crops were grown under much more
variable winter conditions as to temperature, but under much
higher both spring and summer temperatures, especially those
of 1868 and 1870 ; whilst, with the higher temperature there
was, as in the cases with lower temperature and more
abundant crops, much less than the average fall of rain from
seed-time to harvest, one or two months only showing an
excess.

Turning to the seasons of unusually low productiveness, in
that of 1853 the early winter had been unseasonably wet and
warm, the land being generally saturated with water, and in
many cases flooded ; the spring was unseasonably cold, and also
wet, and the summer was, upon the whole, colder than the
average, and very wet. The characteristics of the season of
1860, yielding a crop both late and much below the average
both in quantity and quality, were a winter alternately very
cold and very mild, and upon the whole wet, followed by a
spring, summer, and autumn generally stormy, cold, wet,
sunless, and unseasonable. In the season of 1879 there was,
from seed-time to harvest, a considerable deficiency of tem-
perature, compared with the average, in every month except-
ing March. It is remarkable, however, that there was even
a lower mean temperature in June, 1854 — a season of very
great abundance — than in June, 1879, the season of the
worst crop known within the century. But it was by the
continuity and excessive amount of the rainfall that the
season of 1878-9 was especially characterised, the excess
during the eleven months, from November to September
inclusive, being more than eleven inches over the average,
and the total amount was more than double that over the

F 2

68 The Eothamsted Wheat Experiments.

same period of some of the seasons of high productiveness.
A comparison of 1879 with 1816 shows that the former year
was characterised by a season which was, from seed-time to
the end of the summer, worse than that of the latter.
1816 suffered more from low temperature, but less from
excess of rain during the summer. Both crops were, however,
very late, and, for getting in the crop, the season of 1816 was
much worse than that of 1879.

It is now desirable, disregarding as much as possible the
specialities of individual seasons, to consider the average
character of classes of seasons, arranged according to the
general character of their wheat crops. The classes are as
follows : Six years of high produce of both corn and straw —
1832, 1834, 1835, 1854, 1863, and 1864 ; four years of high
produce of grain, but not of straw — 1833, 1857, 1868, and
1870 ; four years of very low produce— 1816, 1853, 1860,
and 1879.

Taking the six years of high produce, and confining
attention, in the first place, to the period of six months,
from November to April inclusive, in only one of the six
seasons which go to make the average were there two
months, and in four others there was only one month, of the
six of in any material degree lower than average temperature,
and in only one season (1854) was there a really severe
winter month. With these few exceptions, every other
month of the six within each of the six seasons was either
about average or over average, and in many cases very much
over average, as to temperature. As to rainfall over the
same period, in two of the seasons there were two months,
and in two there was only one month with any considerable
excess of rain ; whilst in the other two there was a deficiency in
every month of the six. There were, therefore, in each of the
six seasons, four, five, or six of these six months considerably
drier than the average. Concerning the next three months
of May, June, and July : in two out of the six seasons, each
of the three was warmer than the average, in two each was

Seasons of High and Low Produce compared. 69

colder than the average, and in two there were warmer and
colder months, with about average mean temperatures. As
to rainfall, in one out of the six years there were two of the
three months, in four there was only one of the three, and in
the other there was no month, with an excess of rain. In
only one of the six years was the total rain of the three
months over the average; though in three of the six seasons
there was an excess in August. It may be added that the
average mean temperature of these six seasons was higher than
that of one hundred and eight years (1771-1878) in every
month of the twelve ; but the excess was very much greater in
the months prior to May, than in May and afterwards. Upon
the whole, then, these seasons of highest productiveness were
characterised by higher than average temperatures during
most of the winter and the early spring. Some were consi-
derably warmer during the summer also, but the majority
were characterised by but little higher, or even lower, than
average temperatures in the summer. There was also a
prevailing deficiency of rain in the winter and spring, but a
less marked deficiency in the summer.

In the second class of seasons, those of high produce of
grain, but of small produce of straw, and therefore of com-
paratively small total produce, the distribution of the excess
of temperature is exactly the opposite of that observed in the
case of the seasons of heaviest gross produce. The result
was associated with little more than fairly average condi-
tions as to temperature during the early stages of develop-
ment of the plant, but with a considerable excess during the
period of active above-ground growth, and of maturation.
There is, at the same time, though a considerable total
deficiency of rain, a much more marked deficiency during
the periods of more active above-ground growth, and of
ripening, than during the earlier stages.

In the third class of seasons, those of unusually low
produce, the averages show an actual deficiency of tempera-
ture in ten months out of the twelve ; and in only one from

70

The Bothamsted Wheat Experiments.

seed-time to harvest was there an average excess of any
importance, namely, in January. With this great deficiency
of average temperature, there appears, almost throughout,
an excess of rain, the excess being very much the greater in
April, and afterwards up to harvest, than previously. The
number of rainy days is also greatly in excess, especially in
the summer months. Very low productiveness was, then,
associated with both low temperatures and excess of rain,
especially during the periods of more active above-ground
growth and ripening.

In Table XIV. the column A refers to the six seasons of
much both grain and straw; the column B to the four
seasons of much grain but not much straw ; the column C to
the four seasons of low produce. The averages are made up
from different periods, as noted at the foot of the table, and
all the figures are deduced from the records at Greenwich.
The wheat season is, of course, from October to September.

TABLE XIV. — SUMMARY OF TEMPERATURE AND OF RAINFALL.

A.

B.

C.

Average.*

Average monthly mean temperature
(degrees F.)

49-7

49-8

47-2

48-6

Annual rainfall (inches)

21-44

19-53

32-89

25-4

Number of days on which O'Olin.
rain or more, fell

140

136

199

141

* In the first line the average temperature is that of 108 years— 1771-1878. In the
second line the average rainfall is that of 63 years— 1815-1877. In the third line
the average number of days is that of 55 years— 1815-1869.

From the comparisons which have now been made of the
various seasons, it would appear that mildness and com-
parative dryness, of at any rate considerable portions of the
winter and early spring, favouring root development, that is,
an extended possession of the soil by the plant, and a some-
what early start, have been the characteristics of the most
productive seasons. Indeed, with these favourable early
conditions, an abundant and high-yielding crop may be

Characteristics of Noteworthy Seasons. 71

obtained with only fairly average or even under average
summer conditions. But there can be little doubt that,
when high summer temperatures, without excess of rain, do
succeed upon the favourable conditions of early growth and of
plant just mentioned, the proportion of grain yielded by the
bulky crop will be the greater. The less bulky, and some-
what less abundant in grain, but still high-yielding crops
have, on the other hand, generally had less favourable
conditions for winter root development, and for early growth
in spring, but have been developed under the influence of
considerably higher than average summer temperature, with,
at the same time, deficiency of rain almost throughout, and a
considerable deficiency during the summer months.

The seasons of unusually deficient wheat crops have been
characterised by severe, or at any rate very changeable,
winter and spring conditions, with, at the same time,
generally an excess of rain during these periods, frequently
saturating the soil, causing much drainage, and discouraging
root development and early growth in spring. But the more
striking characteristic of the bad seasons is a great deficiency
of average temperature, and especially a great excess of rain,
from the period of active above-ground growth until harvest.
The crop of 1816, however, suffered more from low tempera-
ture than from excess of rain, and that of 1879 much more
from an excess of rain than from low temperature until mid-
autumri, after which 1816 continued wet, and 1879 became
dry. It would appear that any defect of our climate in appro-
priateness for the production of full and well-matured wheat
crops is more connected with an excess of rain, and conse-
quent wetness of soil and humidity of atmosphere, than with
any deficiency of average summer temperature.

A few more words, in conclusion, respecting the ne\er-to-
be^forgotten season of 1879. In that year the wheat plant,
which luxuriates in a comparatively dry soil and climate,
passed its whole existence under exactly opposite conditions,
and the result was only what might have been expected.

72

The Rothamsted Wheat Experiments.

Though it has long been known that an excess of rain is
injurious to the wheat crop, it is only within comparatively
recent years that one, at least, of the material causes of the
adverse influence has been clearly made out, namely, the great
loss of nitrogen carried off by drainage in the form of nitrates.
Moreover, when the manures are autumn-sown, this loss is
much greater during the winter than during the later period
of the season. Table XV. shows the bushels of grain and
the weight of straw, and of total produce, obtained in each of
the seven years, from 1873 to 1879 inclusive, by autumn sowing
and by spring sowing respectively, of the ammonia-salts.

TABLE XV. — AUTUMN SOWING AND SPRING SOWING OF
AMMONIA-SALTS.

1873
1874
1875
1876
1877
1878
1879

Dates of Bowing
ammonia-salts.

Grain, Straw, and Total Produce per acre.

Grain.

Straw.

Total Produce.

Autumn.

Oct. 18
„ 28
„ 23
„ 30
„ 17
Nov. 3
Oct. 15

Spring.

Autumn
sown.

Spring
sown,

Autumn
sown.

Spring
sown.

Autumn
sown.

Spring
sown.

Ib.
5031

4588
4915
4083
4795
7017
4063

Mar. 25
„ 19
„ 23
„ 24
Apr. 11
Mar. 14

» 10

Bushels.
22
39f
25|
23*

19*

22£

H

Bushels.
32|
29|
25*
25*
33i
3H
16±

Ib.
2021
4645
3422
2212
1835
3071
906

Ib.
3079
2776
3204
2428
2788
4952
3012

Ib.
3344

7094
5110
3793
3048
4486
1275

Averag

68

22*

27f

2587

3177

4021

4927

Only in 1874, therefore, was the result decidedly in favour
of autumn sowing. In 1873, 1877, 1878, and 1879 it was
decidedly against autumn sowing ; whilst in 1875 and 1876
the difference was immaterial. It is particularly worthy of
note that during the period between the autumn sowing and
the spring sowing the rainfall was far less in the season of
1874 than in any of the others. The season of 1875 ranks
next in this respect.

Loss of Nitrogen in Drainage Waters. 73

Hence, independently of other adverse effects arising from
low temperature and excess of rain, the Rothamsted experi-
mental wheat crops of 1879 suffered very considerably from
loss of the nitrogen of the manure by drainage. Experi-
mental evidence proves that similar loss may arise when
animal nitrogenous manures (such as hair, horn shavings,
woollen rags, &c.) and vegetable nitrogenous manures (such
as rape-cake) are used. Further, the drainage water from
the dunged plot in the experimental wheat-field is sometimes
found to contain a considerable amount of nitric acid, but
always very much less than that collected at the same time
from adjoining plots receiving much less nitrogen as manure,
but in the form of ammonia-salts or nitrate of soda. Whilst,
therefore, it is to be assumed that the loss of the nitrogen
of manure by drainage in the season of 1879 was propor-
tionally much greater in the experimental wheat-field than
in the case of land farmed in the ordinary way, there can
nevertheless be no doubt that much of the land of the
country at large also suffered loss in the same way.

74 The Rothamsted Wheat Experiments.

Y.-THE HOME PRODUCE, IMPORTS, AND
CONSUMPTION OF WHEAT.

INASMUCH as wheat furnishes the staple food of the British
people, all questions affecting this cereal are necessarily of
profound economic interest. Messrs. Lawes and Gilbert
have written two papers " On the Home Produce, Imports,
and Consumption of Wheat" (R.A.S. Journal, 1868, and
1880, and also Journal of the Statistical Society, June,
1880) ; and at the outset of the inquiry, they had to face very
serious difficulties arising out of the crude and imperfect
character of the statistical information at their disposal,
though since then some improvement has been effected in the
official collection of agricultural statistics. The second of the
two papers now under notice does not extend beyond the
season of 1879-80, nevertheless, in the tables which follow,
the statistical record is brought down to the harvest of 1887.
In a paper on " Our Daily Food ; its Price and Sources of
Supply," read before the Statistical Society, March 17, 1868,
Mr. (now Sir) James Caird estimated the cost of wheat and
wheat flour consumed in the United Kingdom at 30,000,OOOZ.
sterling more for the year, in consequence of the bad season
of 1867, than after the good harvest of 1863, and that out of
this total extra cost, 27,400,000?. more would have to be
paid for foreign corn after the bad harvest than after the
good one. He noticed the influence which such a result
must have upon the trade of the country, and insisted upon
the great advantages which would accrue from early know-
ledge as to the area and yield of our various crops. During
only the preceding two years (1866-7) had voluntary returns

Yield of Wheat per Acre per Annum. 75

been collected throughout the United Kingdom referring to
the number of acres under each crop, and to some other points ;
but, in regard to the important question of the amount of
produce obtained, either per acre or in the aggregate, there
were no returns nor any reliable information. In these
circumstances certain of the plots of the experimental wheat-
field at Eothamsted were, after careful observation and
comparison, selected as affording a trustworthy estimate of
the general yield over the country, and every year since 1862
there has appeared in the Times a letter containing an
estimate of the yield of wheat for the current season. The
results of the experimental wheat-field were regarded by Mr.
Caird, in the above-mentioned paper, as having "proved a
very satisfactory index of the general yield over the chief
wheat-producing area of the kingdom," and as affording
" the most instructive series of facts for the guidance of the
British corn-grower on record." The selected plots com-
prise:

Plot 3. — Unmanured every year, experiment commencing
1843-4.

Plot 2. — 14 tons farmyard manure every year, commencing
1843-4.

Plot 7. — Mixed mineral manure, and 4001b. ammonia-salts,
each year, 1851-2 and since.

Plot 8. — Mixed mineral manure, and 6001b. ammonia-salts,
each year, 1851-2 and since.

plot 9. — Mixed mineral manure, and 5501b. nitrate of soda,
each year, 1854-5 and since.

In forming the estimate of the average produce per acre of
the country at large, the plan adopted is to take the mean
produce of the unmanured plot, of the farmyard manure
plot, and of the three artifically manured plots reckoned as
one, and to reduce the result so obtained to bushels of the
standard weight of 611b. per bushel. Merely as an example,
Table XVI., published in October, 1787, is given. Though ex-
perience proves that this mode of estimates leave but little to

76

The Rothamsted Wheat Experiments.

be desired as a means of computation of the average yield of
the country over a number of years, it has not been found

TABLE XVI. —ESTIMATED YIELD OF WHEAT PER ACRE, 1887.
Bushels of Dressed G-rain per Acre.

Harvest.

Unmanured.
Plot 3.

Farmyard
manure.

I'lot _>.

Artificial manures.

Mean of Plots 7, 1
8, and 9 (or 16).

Means of Plots 3,
2, and 7, 8, 9
(or 16).

t~°

$

00

s

S

If

ss

Present year, 1887

14*

m

13f
13

34J
31f
34*
33f

29*
30*
33*
32*

34*
35
871

36*

39|
36f
36f
362

34*
34

35|
35$

28 (1)
25* (2)
28* (3)
27f (4)

Average 10 years, 1877-86
Average 25 years, 1852-76
Average 35 years, 1852-86

Weight per Bushel of Dressed Grain, in U)8.

3 «'

fa

Artificial mai

lures.

«'£

Si

PS

l::*

Harvest

|s
&

|ss

I>I 00

«£

So,

is

is

S3 oo*

lii

g(N

Present year, 1887

59J

881

62 63i

63

M$

61J

Average 10 years, 1877-86
Average 25 years, 1852-76
Average 35 years, 1852-86

58*
57*
58*

60*
60*
60±

60f 60|
59f 59*
59| 59*

59*
58|
58*

60i

59
59|

59*
59

59i

Total Straw, Chaff, $c., per Acre, cwt.

d

Artificial manures.

•>£

S*

1-

"S cj .
03 £ "N

r?s

S*^

Harvest

§ 0

g*C 0

'o'o'n

|K

*is

8

oo

i

II

||

I5*

Present year, 1887
Average 10 years, 1877-86

8
8|

30*

24*
32

40i

36i 30*

42i 38*

23i
25

Average 25 years, 1852-76

12 33

34

40|

41 -f 38 v

27*

Average 35 years, 1852-86

11 31f

33*

40*

41*

38|

27*

(1) Equal to 28| bushels, at 611b. per bushel. (3) Equal to 27£ bushels, at 611b. per bushel.

(2) Equal to 25 bushels, at Glib, per bushel. (4) Equalto26| bushels, at 611b.perbusb.el.

to be equally applicable for each individual year. The so-
calculated average produce per acre on the selected plots

Yield of Wheat per Acre per Annum. 77

gives somewhat too high a result for the country at large in
seasons of great abundance, and too low a result in un-
favourable seasons. Accordingly, in some seasons, instead of
the actual average indicated by the experimental plots, a
higher or a lower figure has been adopted ; and especially in
the case of some of the then (1880) recent bad seasons a
higher one was taken.

Independently of any such admitted differences between
the so directly calculated and the actually adopted estimates
for individual years, the question arises — Whether the
average result indicated by the several selected plots remains
as applicable as heretofore ; or whether the produce of some
is annually declining, or that of others annually increasing,
irrespectively of the influence of season, so as to vitiate the
continued applicability of such results for the purposes of
such an estimate ? This point has already been discussed in
connection with the forty years' experiments on the con-
tinuous growth of wheat, so that it is only necessary here to
repeat that the produce of the unmanured plot is gradually
declining from exhaustion, that the farmyard manure plot is
increasing in fertility, and that there is no evidence of
material increase or material decrease on either of the plots
receiving ammonia-salts other than that due to season.
Comparing the direct average of the experimental plots with
that actually adopted as the average for the United Kingdom
each year, the experimental plots indicate for the whole
twenty-eight years (i.e., up to 1879-80) about three-quarters
of a bushel less per acre per annum than the actually
adopted estimates founded upon them.

Taking the average of the twenty-eight years' adopted
estimate of produce per acre as 100, the first column of
Table XVII. shows the deviation from this general average
for the whole period, over the first eight, the second eight,
the third eight, and the last four years of the twenty-eight ;
and the second column shows the deviation, from the same
standard, of the average produce per acre on the selected plots.

78

The Rothamsted Wheat Experiments.

After making allowance for the seasonal variations in
each of these periods, it appears that, in fairly average
seasons, the mean produce of the experimental plots fairly
represents the average produce ; that in seasons of unusual
abundance the experimental plots indicate too high a figure,
and that in seasons of great deficiency they give too low a
figure. Still, it is, upon the whole, concluded that there is
no better basis for estimating the annual yield of the country
each year than that of the average produce of the same
selected plots as heretofore relied upon ; but that, as hereto-
fore, some judgment must be exercised each year, according
to the characters of the season, in deciding whether to
adopt the actual figure indicated by the experimental plots,
or in which direction, and in what degree, it should be
modified.

TABLE XVII. — SHOWING THE DEVIATION OVER EACH SEPARATE PERIOD

FROM THE ADOPTED AVERAGE OF THE WHOLE PERIOD TAKEN AS 100.

Actually adopted
averages.

Averages of Plots 3, 2,
and 7, 8. and 9.

First eight years, 1852-59 ...
Second eight years, 1860-67...
Third eight years, 1868-75 ...
Last four years, 1876-79

103

104
98
89

101
106
99
71

Total period, 28 years . . .

100

98

In illustration of the difficulty of this inquiry, it will be
appropriate to notice at this stage the nature of the data
upon which the investigators had to rely for information
upon the following points : 1. The area tinder wheat.
2. The average yield of wheat per acre. 3. The aggregate
home produce, and the amount of it available for consumption.
4. The imports. 5. The population. 6. The average con-
sumption of wheat per head of the population per annum.

1. From 1852 to 1865 inclusive, it was necessary to rely on
estimates alone in fixing the area under wheat in England

Nature of the Statistical Data available. 79

and Wales. For Scotland, there were the returns collected
by the Highland Society for 1854, 1855, 1856, and 1857;
but for the two years prior to 1854, and for the years subse-
quent to 1857, down to 1865 inclusive, estimates only were
available. For Ireland, returns were available throughout.
From 1866 onwards the " Agricultural Returns " have given
the much-needed official information as to the acreage of
wheat in the United Kingdom.

2. The only returns or official estimates as to the average
yield of wheat per acre were for Scotland for four years, and
for Ireland for each year within the period of the inquiry.
For England and Wales, comprising from 85 to 90 per cent,
of the total area under wheat, it was necessary, in the
absence of any official information whatever, to fall back
upon the average produce per acre each year of the selected
plots at Eothamsted, in the manDer already described.

3. The aggregate home produce is deduced from a know-
ledge of the acreage under crop, combined with a trustworthy
estimate of the average yield per acre, whilst the amount
available for consumption is arrived at by deducting from
the total produce the quantity annually returned to the land
as seed. On adequate grounds it is assumed that 2| bushels
per acre are required for this latter purpose.

4. The imports have throughout been taken from the
Eeturns for the United Kingdom, either of the net imports
of wheat and wheat flour, or of the imports and exports from
which the net imports can be calculated. Moreover, the net
imports have been calculated, not for the calendar years, but
for the harvest years, that is, from September 1 of one year to
August 31 of the next.

5. The figures as to population are taken from the esti-
mates of the Begistrar-G-eneral. The middle of the calendar
year being the end of June, and the middle of the harvest
year the end of February, the plan adopted has been to add
to the number officially recorded for the preceding mid-
summer two-thirds of the difference between that figure

80 The RotTiamsted Wheat Experiments.

and the number given for the next midsummer, thus
bringing the estimate up to the end of February. Any
irregularities are detected immediately after the Census
years.

6. The average consumption of wheat per head of the
population per annum is a difficult matter to determine. In
the first paper (1868) the estimate was founded on the calcu-
lation of eighty-six different dietaries, arranged in fifteen
divisions, according to sex, age, activity of mode of life,
and other circumstances ; and the result so obtained was
compared with that arrived at on the basis of the population,
and of the amounts of the available home produce, and of
the net imports of wheat each year. For Scotland, and for
Ireland, it was only possible to found an estimate on the
basis of population, and on the amounts of the home and
foreign supplies. On these bases the average consumption of
wheat in the United Kingdom collectively was estimated to
be 5J bushels per head of the population per annum, during
the later years to which the inquiry related. This estimate
is, in the course of the second paper (1880), submitted by the
investigators to close criticism. They comment on the fact
that the average consumption per head has increased in the
United Kingdom as a whole since the establishment of free
trade in corn, though probably more rapidly at first. The
quantity consumed will vary according to the prosperity or
otherwise of the people, to the price of wheat itself, and to
that of other articles of food also. Independently of the
influence of lower prices for wheat, and of the increased
prosperity of the masses of the population, among the cir-
cumstances tending to increase the consumption of wheat in
recent years may be mentioned the increased price of meat ;
whilst, among those tending to limit the rate of increase of
consumption, may be noted the fact that the proportion of
the total wheat consumed which is derived from foreign
sources is rapidly increasing, and the drier foreign wheats
will undoubtedly yield a larger percentage of flour, and flour

Consumption of Wheat per Head per Annum.

81

of better quality than much of the home-grown grain.
Upon the whole, the investigators were led to conclude that
their estimates of consumption (5'1 bushels per head per
annum) over the first period of eight years (1852-59) might
be somewhat too low ; also that their previously published
estimates of consumption for the years subsequent to the first
sixteen, were more probably too low than that their estimates
of average produce per acre, and of aggregate produce
founded upon them, were too high. For reasons assigned,
however, they adopt in their second paper their previous
estimates of average produce per acre each year without

TABLE XVIII. — YIELD OP WHEAT PER ACRE PER ANNUM.

Average produce per acre.

According to in-
creased con-
sumption and
imports.

According to

annually adopted

estimates.

Bushels. Bushels.

Average 8 years, 1852-3—1859-60...

28i

28

Ditto 1860-1—1867-8 ...

28f

28f

Ditto 1868-9—1875-6 ...

26|

26f

Average 3 years, 1876-7—1878-9 ...

27

27i

Average 27 years, 1852-3 — 187

271

27f

change. They also adopt their previous estimates of con-
sumption per head for the first two periods of eight years
each without change— 5'1 bushels and 5'5 bushels respectively.
But, for the third period of eight years, they assume the
consumption to be at the rate of 5'6 bushels per head, and
for the last three years (1876-78) at the rate of 5'65 bushels,
instead of 5.5 bushels over those eleven years, as previously
reckoned ; and, until further experience should indicate
further change to be necessary, they propose to adopt 5|
bushels as the average consumption per head of the popula-
tion per annum over the United Kingdom. Beckoning
consumption at the rate of 5'1 bushels per head per annum

G

The Rothamsted Wheat Experiments.

the first eight years, 5'5 the second eight, 5'6 the third eight,
and 5'65 the next three years (instead of, as previously, 5*5
bushels over the last eleven years), the two estimates,
recorded in Table XVIII., of the average produce per acre per
annum over the United Kingdom, are then seen to be in very
close agreement.

With reference to the circumstance that the average pro-
duce per acre, founded on the annual estimates, is slightly
higher over the last two periods than that founded on con-
sumption and imports, it must be borne in mind that the
quantity of wheat consumed by farm-stock is an unknown
and varying element, and either the estimate of the consump-
tion per head of the population must be fixed to include the
average consumption in other ways, or the annual estimates
of produce per acre, and of the aggregate home produce
founded upon them, should exceed those founded on con-
sumption and imports. It is remarked that an increase of one-
tenth of a bushel in the consumption per head per annum would,
if derived from home produce, represent an increase of one
bushel per acre per annum over the United Kingdom, assuming
a population of 33,000,000, and an area under wheat of
3,300,000 acres — figures which closely represented the actual
facts a very few years before 1880. With an increasing
population and a diminishing area under wheat, such an
assumed increase in consumption per head would, of course,
correspond to more than a bushel per acre.

A most comprehensive and elaborate table presents in one
view particulars of the home produce, imports, and consump-
tion of wheat in the United Kingdom during each of the
harvest years from 1852-3 to 1879-80. Space is too limited
to permit of the reproduction of this valuable table, but in
Table XIX. are given the details for each of the last eight
or nine harvest years, together with averages of the four
periods of eight years, of the two periods of sixteen years,
and of the entire period of thirty-two years.

In discussing the facts set forth in the complete table, it is

Home Produce, Imports, and Consumption of Wheat. 83

«J

2§l

ft 00

§co
tt

r>- •* oo 10

CO fr- lO -Tfl O CM

ib >b cb >b i cb ib ib

CM rH O CO

j> cs ip ip
cb cb 4? cb

iO CO 00 CM ! CO IO CM ,
05 000 OS OOSt- I
rH rH CM rH CM rH rH

CM 1>- rH OS
1> 00 CM 00

CD 1> (M O

CO I> OS CM

Tjl
CM

7^ rH^ r-^ 00^

<M~ O" C<f 10"

00 O 00 rH

•^ T-H_ 7^ os oq_

SO I CO" t>~ OS~ 10"

rH rH 1O
O CO O
OrH CO

CO CM lO
O CO IO
Wr-^CO^
l>^io"
CM 71 CM

Qu
6,4

rH CO CO OS

<N t» o r-

o co 10 cjs

O OS
,,!> t^'o"l>^

> %*& lO CO lO O
< O^ O OS rH -

co" ; CO"CM"CO"CM"

rH 73 CO "*
CO GO 00 00

O rH CM CO !

lO CO t^ 00

CO 00 CO 00

I I I I

-^ IO CO t^

OS <M CO rH

oo i>

^
os"co"os""^" co"t^

IO IO rH lO O 00
rH CM CO I> *>,rH
I-H"OO" CM~O"

s a a5 ! "53 M« «*o HX ' -i* «>
S^S ^acoooco-^ooio i>

gSO 2CMCMCMCMC\1<M CM

^ C3 w &

O rH rH O i CD rH I CO
.CO rH CO rH ! CO CM j lO

o 2

The Rothamsted Wheat Experiments.

to be noticed that during the thirty-two years 1852-3 to
1883-4 inclusive, the area under wheat in the United Kingdom
was reduced by about one-third. The average yield per acre
is estimated at 28 bushels ; but, owing to recent bad seasons,
the average for the whole period of thirty-two years was only
27 bushels, that for the first sixteen years having been 28 j,
but that for the second sixteen years only 25|. Thus there
has been a reduction not only in area under cultivation, but
in yield per acre also. This latter, however, is probably tempo-
rary, whilst the reduction in area will doubtless continue.
The great increase of population which has ta.ken place within
the period covered by the table has, of course, necessitated
greatly increased consumption, and the comparison of the
home production with the foreign importation, for succes-
sive periods, becomes of much interest. An inspection of
the table will show that over the first eight years only one-
fourth of the wheat consumed was obtained from foreign
sources, whilst over the last eight years nearly two-thirds
(15,781,483qrs.) of the entire consumption (24,536,234qrs.)
was imported. It is probable that the home produce will
still decline, consequent chiefly on reduction of area under
cultivation ; whilst, with increase of population, imports must
increase. It may be seen that the total consumption of
wheat per annum has increased from an average of about
eighteen million quarters over the first eight years to upwards
of twenty-four million quarters over the last eight of the
thirty-two years — an increase of one-third. And whereas the
average consumption per head per annum was under 5'1
bushels over the first eight years, it amounted to nearly 5'7
bushels over the last eight years.

Discussing this subject so recently as February, 1887, Sir
John Lawes and Dr. Gilbert say : "In conclusion, we would
remark that whilst 30 bushels per acre per annum is frequently
assumed to be the average yield of the United Kingdom, our
own estimate was, until recent years, 28 J bushels ; but the
series of bad seasons has reduced the average, over the

Ratio of Imported Wheat to Home-grown Wheat. 85

thirty-two years to 1883 inclusive, to 27 bushels. We never -
theless still adopt 28 bushels as the normal average yield.
If, however, our estimate of consumption is, as alleged, too
high, the average yield per acre would be lower, not higher,
than our figures show. If, on the other hand, it were
established that there is a large consumption by stock or
otherwise, over and above our estimate of total consumption,
it is obvious that our estimates of average yield per acre are,
so far, too low."

The following statements are based upon the authors'
second paper (1880), embracing three periods of eight years
each, and an odd period of three years at the end. Of course,
with regard to prices, most serious modification would be
needed in order to bring the statistics up to the present date,
the figures now to be mentioned coming down only to 1880.
It appears, then, that the price of wheat per quarter declined
from an average of 57s. Sd. over the first eight vears (includ-
ing the period of the Crimean War) to 49s. over the last
three years (1876-77-78). The annual value of the home
produce available for consumption declined from an average
of nearly 38,000,0002. over the first eight years (1852-59) to
less than 25,000,0002. over the last three years. The annual
value of the imported wheat increased from an average of
little more than 13,000,0002. over the first eight years to more
than 33,000,0002. over the last three years. The annual
value of the total wheat estimated to be consumed (1852
to 1878) ranged from under 40,000,0002. to more than
71,000,0002., and it increased from an average of about
51,500,0002. over the first eight years, to more than
58,000,0002. over the last three years. The average annual
cost of wheat per head was somewhat reduced in the later
periods ; it was 36s. 2c2. over the twenty- seven years.

Over the whole period of thirty -two years (1852—1884),
44'6 per cent, of the wheat consumed was derived from imports ;
and the amount supplied from foreign sources has increased
from an average of 26'5 per cent, of the total over the first

The Rothamsted Wheat Experiments.

eight years, to 64-3 per cent, of the total consumed over the
fourth period of eight years. The least proportional quan-
tity supplied from foreign sources in any one year was 15'4
per cent, in 1854-5 (2,983,000qrs. imports less exports against
1 9,410, 742qrs. available for consumption), and the greatest
proportional quantity supplied from foreign sources in any one
year amounted to 76'5 per cent, in 1879-80 (16,409,933qrs.
imports less exports against 21,457,773qrs. available for con-
sumption). The amounts carried over from one harvest year
to another will, of course, vary exceedingly according to cir-
cumstances, the influence of which cannot with any certainty
be estimated. There is no reliable information as to the
quantity of home-produced wheat held in farmers' hands,
the quantity consumed by farm-stock or otherwise used, or
the quantity of foreign wheat held over in the granaries.
Then, again, the actual length of the period to be provided
for, dependent on the earliness or the lateness of consecu-
tive harvests, is a point not to be overlooked ; it was estimated
that, at the harvest of 1865, there still remained over from
the extraordinary crop of 1863, and the abundant one of 1864r
wheat equal to from one-third to one-half of an average crop,
and that, even at the harvest of 1866 some of the crop of
1863 remained unthrashed. It may, indeed, be stated gene-
rally, that, as a rule, the excesses follow, as they should r
seasons of high productiveness, whilst the deficiencies follow
seasons of low productiveness.

Continuous Growth of Barley. 87

?!.— EXPERIMENTS ON THE CONTINUOUS
GROWTH OF BARLEY UPON THE SAME LAND
FOR TWENTY YEARS.

THE first twenty years' experiments 011 the continuous
growth of barley extended from 1852 to 1871, and the
results are given in the R.A.S. Journal, 1873. The land
selected was a portion of that immediately adjoining the
experimental wheat field. The wheat field has, however, as
a matter of experiment, been artificially drained, whilst the
barley field has not. The custom of the locality, in the case
of land of similar quality, is to take the barley crop after
roots fed off by sheep ; but the soil is too heavy for this to
be done with advantage in wet seasons. Nevertheless, good
crops, both as regards quantity and quality, are so grown on
such land in favourable seasons, and may, as a rule, be relied
upon when barley is taken, not after folding, but after
another corn crop. The object of the experiments was to
obtain answers to such questions as the following : What
are the grain-yielding capabilities of such land ? What
its powers of endurance ? In what constituents, or class of
constituents, does it soonest show signs of exhaustion?
And how far will the answers arrived at on these points in
reference to it, accord with, or be a guide to, those which
would apply to any large proportion of the arable land of
Great Britain when farmed in the ordinary way, with
rotation ?

The previous history of the land was : 1847, Swedish
turnips, with farmyard manure and superphosphate (the

88 The Rothamsted Barley Experiments.

roots carted off) ; 1848, barley ; 1849, clover ; 1850, wheat
1851, barley, with sulphate of ammonia. Having thus
grown two corn crops in succession, it was, agriculturally
speaking, in a somewhat exhausted condition for the after-
growth of grain, and would, in the course of ordinary
practice, be re-manured before growing another crop. It was,
therefore, in a suitable state for testing the effects of
different manures upon the crop, and for showing, by the
results, in what constituents, or class of constituents, the
soil had, by previous cropping, become practically the most
deficient.

The area of 4J acres was divided into twenty-four nearly
square plots ; most of these were exactly one-fifth of an
acre each, but the remainder somewhat less. Two plots
were left unmanured, one was dressed every year with
farmyard manure, and others with different manures which,
respectively, supplied certain constituents of farmyard
manure, separately or in combination. The investigators
maintain that comparative results obtained in this way are
far better calculated to indicate in what constituent or
constituents the soil is relatively deficient, so far as the
available supply for the crop to be grown is concerned, than
what is generally understood as an analysis of the soil.
This, the synthetic, as distinguished from the analytic
method, is capable of giving important and conclusive
evidence as to the conditions and requirements of the
growth of barley, as indeed of other crops.

The selection of manures for the experiments on barley
was, in many respects, the same as that adopted for those on
wheat. In reference to this point, Table XX. is instructive ;
it shows the probable average amounts of certain con-
stituents in what may be taken as fairly corresponding
crops of wheat and barley. The produce per acre assumed,
for this purpose, is : Wheat, 30 bushels of 601b. per bushel
=18001 b., and 30001b. straw=48001b. total produce ; Barley,
40 bushels of 521b. per bushel=20801b., and 25001b. straw

Composition of Equivalent Crops of Wheat and Barley. 89

— 45801b. total produce ; which will contain, approximately,
the following constituents :

TABLE XX. — NITROGEN AND ASH-CONSTITUENTS IN WHEAT AND
BARLEY.

In Grain.

In Straw.

In Total Produce.

Wheat.

Barley.

Wheat.

Barley.

Wheat.

Barley.

Nitrogen

Ib.
32
16
9-5
I
3-5
0-5

Ib.
33
17
11-5
1-5
4
12

Ib.
13
7
20-5
9
3
99-5

Ib.
12
5
18-5
10-5
2-5
63

Ib.
45
23
30
10
6-5
100

Ib.
45
22
30
12
6-5
75

Phosphoric acid

Potash

IJime

Magnesia

Silica ..

It will be observed that most of the above constituents
occur in nearly equal amount in either crop. The total
barley crop, however, would remove rather more lime, but
considerably less silica, than the wheat crop. But, looking
to the grain alone, the barley is seen to remove considerably
more silica, and rather more of each of the other con-
stituents than the wheat. Therefore, where the grain only
is sold, and the straw is returned to the land in due course
as manure, the eventual loss to the soil would be upon the
whole greater, especially in silica, by the growth of such a
crop of barley than of such a crop of wheat. In the
experiments under notice, however, both grain and straw
are always removed from the land.

The exact quantities of manures applied on a given plot
will be specified where necessary. The mineral or ash con-
stituents were supplied in the following commercial forms :
Potash, as sulphate of potash ; soda, as sulphate of soda ;
magnesia, as sulphate of magnesia (Epsom salts) ; lime,
as sulphate, phosphate, and superphosphate ; phosphoric acid,
as bone-ash, mixed with sulphuric acid in quantity sufficient
to convert most of the insoluble earthy phosphate of lime

90 The Rothamsted Barley Experiments.

into sulphate and soluble superphosphate of lime ; sulphuric-
acid, in the phosphatic mixture just mentioned, in sulphates
of potash, soda, and magnesia, in sulphate of ammonia, &c. ;
chlorine, in muriate of ammonia ; silica, as artificial silicate
of soda. Of other constituents, the nitrogen has been
supplied as sulphate and muriate of ammonia ; as nitrate
of soda ; in farmyard manure ; in rape-cake. The non-
nitrogenous organic matter, yielding by decomposition
carbonic acid and other products, was supplied in farmyard
manure, and in rape-cake. The artificial manure or mixture
for each plot is ground up, or otherwise mixed, with a
sufficient quantity of soil and turf-ashes to make it up to a
convenient measure for uniform distribution over the land.
The mixtures so prepared are, with proper precautions, sown
broadcast by hand, it having been found that the application
of an exact amount of manure, to a limited area of land,
can be best accomplished in that way.

Passing now to the field results, the investigators have
constructed tables showing the quantity and quality of the
produce obtained in each one of the twenty seasons. It will
suffice here, however, to reproduce only those tables showing
the average results of the entire period, the results in the
best season, and the results in the worst. The following are
the numbers of the selected plots, with an account of the
manures supplied, per acre, per annum :
Plots.

1 O. Unmanured.

7. 14 tons farmyard manure.

4 0. Mixed mineral manure alone.

1 A. 2001b. ammonia-salts alone.

4 A. Mixed mineral manure and 2001b. ammonia-salts.

4 A A. Mixed mineral manure and 4001b. ammonia-salts
first six years, 2001b. ammonia-salts next ten years,
2751b. nitrate of soda last four years.

4 C. Mixed mineral manure and 20001b. rape-cake first
six years, lOOOlb. rape-cake last fourteen years.

Average Results of Twenty Years.

91

The farmyard manure was made in the open yard, and
did not contain the dung of animals highly fed on purchased
food. The mixed mineral manure was composed per acre-
per annum of —

2001b. sulphate of potash (3001b. the first six years).

lOOlb. sulphate of soda (2001b. the first six years).

lOOlb. sulphate of magnesia

2001b. bone ash. )

1501b. sulphuric acid, sp.gr. 17. j SuPerPhosPhate of lime-

The ammonia-salts were formed of an equal mixture of the
sulphate and muriate of ammonia of commerce.

In Table XXI. are recorded the average results of the
twenty years :

TABLE XXI. — AVERAGE QUANTITY AND QUALITY OF BARLEY PER ACRE
PER ANNUM ON SELECTED PLOTS ; TWENTY YEARS, 1852-1871.

Dressec

. grain.

Total

Plots (manured as
above).

Quantity.

Weight
per
bushel.

Total
grain.

and
chaff.

produce
(grain and
straw).

to 100
straw.

1 O

Bushels.
20

Ib.
52-3

Ib.
1133

cwt.

1H

Ib.
2454

86'6

7

48i

54'3

2768

281

5933

88*5

40

27|

53-4

1550

14f

3162

96-4

1 A

32i

52-1

1840

18|

3919

89-2

4A
4 A A

46i
491

54-0
53'4

2630
2813

28^
32t

5817
6443

83-2
79-5

4 C

47f

53-6

2698

29£

6002

83'0

The significant fact may be noted that, over a period of
twenty years in succession, ammonia-salts (1 A) alone gave
an average, per acre per annum, of 5 bushels more grain,
and of 4cwt. more straw, than the mixed mineral manure
(4 0) alone. Again, the ammonia-salts and mixed mineral
manure together (4 A) gave an average annual produce of
about 19 bushels more grain, and 14cwt. more straw, than
the mineral manure (4 0) alone ; but only about 14 bushels
more grain, and lOcwt. more straw, than the ammonia-

"92 The Rothamsted Barley Experiments.

salts (1 A) alone. Hence, in this, in an agricultural sense,
already corn-exhausted soil, the available supply of nitrogen
was much more readily exhausted than the available supply
of mineral constituents, so far as the requirements for the
growth of barley are concerned.

€HAEACTEEISTICS OF THE BEST AND WOEST
SEASONS OF THE TWENTY YEAES.

If space permitted of the presentation here of the tables
showing each season's results, it would be necessary to bear
in mind that, so far as the influence of season is concerned,
the quantity of the produce depends greatly on the amount
and the distribution of rain during the growing period ;
and the quality (proportion of grain and quality of grain)
on a suitable adaptation of temperature. And, so far as the
influence of manures is concerned, the quantity (luxuriance)
depends greatly on the available supply of nitrogen within
the soil, and the quality of the crop (tendency to form seed
and to ripen) on the available supply of mineral or ash
constituents.

In no two years during the entire period did one and the
same manure yield precisely identical results both as to the
quantity and the quality of its produce. Nor have the
seasons which have been more or less favourable than the
average for one description of manure been equally favour-
able or unfavourable for other descriptions. Of the twenty
seasons, that of 1854 was upon the whole the most, and that
of 1856 was upon the whole the least, productive. The
results obtained in each season are stated in Table XXII.,
in which the same plots are selected as in Table XXI.,
though, in this case, the uninanured plot has been placed
second, and the farmyard-manured first. The description and
quantities of manures per acre have already been specified on
page 90.

The remarkable contrasts afforded by this table will

Yields of Best and Worst Seasons.

93

TABLE XXII. — QUANTITY AND QUALITY OP BABLEY ON SELECTED

PLOTS, IN THE LEAST, AND IN THE MOST, PRODUCTIVE SEASON O¥

THE TWENTY, 1852-1871.

Weight per Bushel of Dressed Grain.

Plots.

Least Productive
Season, 1856.

Most Productive
Season, 1854.

Excess of 1854
over 1856.

7
1O

Ib.
47-1
49-1

Ib.
53-9
53-6

Ib.
6-8
4'5

4 0

47-0

54-0

7'0

1 A

48-5

53-6

5'1

4 A

46-4

54-3

7-9

4AA

45'4

52-1

6-7

40

46-3

52-8

6-5

Total Grain per Acre, reckoned at 52lb. per bushel.

Plots.

Least Productive
Season, 1856.

Most Productive
Season, 1854.

Excess of 1854
over 1856.

7

Bushels.
31|

Bushels.
60£

Bushels.
28i

1 O

15f

374

22£

4 O .

19f

46|

26

1 A

27£

53J-

25f

4 A

30f

66

35i

4AA
4C

36i
35f

68
65|

81|

30i

Straw (and chaff) per acre.

Plots.

Least Productive
Season, 1856.

Most Productive
Season, 1854.

Excess of 1854
over 1856.

7

cwts.
19f

cwts.
37i

cwts.
17^

1 O

8f

21f

13

40

9f

23i

13J

1 A

17|

30i

13i

4 A

2H

40£

19i

4AA

33

49

16

4 C

30|

42£

Hf

Total produce (grain and straw) per acre.

Plots.

Least Productive
Season, 1856.

Most Productive
Season, 1854.

Excess of 1854
over 1856.

7

Ib.
3866

Ib.

7298

Ib.
3432

1 O

1797

4405

2608

4O

2075

4969

2894

1 A

3347

6155

2808

4 A

3981

7958

3977

4AA

5582

9026

3444

4C

5257

8125

2868

The Rothamsted Barley Experiments.

become apparent on inspection, but it is worth noting that,
with one and the same expenditure for manure, there was
a difference in the quantity of produce obtained in the two
seasons of from thirty to thirty-five bushels of grain (4 A,
4 A A 4 C), and in one case (4 A) of nearly a ton of straw, or
not much less than would represent the average barley crop
of many localities. It may further be remarked that, whilst
the season of 1856 was far worse than that of 1853 as regards
both the quantity and the quality of the barley crop,
1853 was, for the experimental wheat (which that year could
not be sown until the spring), in every particular worse than
1856. Again, whilst 1854 was a decidedly more productive
barley year than 1863, yielding under almost every condition
of manuring not only more grain, but considerably more
straw — in other words, a greater quantity of total produce,
indicating greater luxuriance — 1863 was, on the other hand,
a considerably more productive wheat year than 1854, and
especially so in grain. Both years were, however, remarkable
for very krge produce of both grain and straw, both of wheat
and of barley.

The years next to 1854, in order of productiveness of barley
were 1857 and 1864, which were very good wheat years also.
The years next in order to 1856 in point of badness of barley
crop were 1859, 1860, 1868, and 1870 ; the deficiency in the
two last-mentioned years being due to summer heat and
drought, but in the other two seasons to very opposite
conditions. The characters of the selected seasons of 1854
and 1856 may now be usefully compared.

The very unusually productive season of 1854 was preceded
by a very severe winter; March and April were upon the
whole warmer than usual, but May, June, July, and August
were pretty uniformly below the average temperature, whilst
in March, April, June, and July there was a very considerable
deficiency of rain, though more than the average number of
rainy days. In May, however, there was about double the
usual amount of rain, and an unusually large number of

Characteristics of Good and Bad Seasons. 95

rainy days. In August, again, there was a full amount of
rain, which, however, fell for the most part in heavy showers,
and the month was upon the whole favourable for ripening
and harvest. Thus, 1854 was characterised by prevailing low
rather than high temperature, an abundance of rain at the
period of early active growth (doubtless favouring root
development), and again before harvest; but otherwise by
dryness as well as coolness. It would seem, therefore, that
the large produce was due to a sufficiency of moisture within
the soil when most wanted, with, at other times, com-
paratively dry and temperate atmospheric conditions,
resulting in a continuity of unchecked growth, rather than in
very active luxuriance at intervals.

Contrasting with the foregoing the extremely unfavourable
season of 1856, it appears that there had been some rough
weather in the early part of the winter, but the later and
greater part was upon the whole mild. March, April, and
especially May, were colder than the average, whilst June,
July, and August, though showing average day -temperatures
fully as high as usual, were very changeable, and in June and
July the nights were cold. In each month from January to
July there was considerably more rain than in the corre-
sponding months of 1854 — in all nearly 6in. more ; whilst in
April there was an excess over the average, in May more than
double the average, and in August again an excess.
Generally, then, the season was very wet, with marked
alternations of heat and cold, whilst it was, for the period of
the year, the coldest during the time of the greatest excess of
rain. Moreover, there were heavy rains, with considerable
fluctuations of temperature, about the ripening and harvest
period. The very bad result in this season would seem to be
due, therefore, to an excess of rain, with, at the same time,
great alternations of temperature during the most active
periods of growth, entirely preventing continuity of progress ;
whilst the unhealthy plant thus produced was subjected to
unfavourable maturing conditions.

96 The Rothamsted Barley Experiments.

Bearing in mind that both wheat and barley will flourish
under very temperate conditions, it is worthy of remark that
both -1854, the year of the best barley crop, and 1863, the
year of the best wheat crop, in the Rothamsted experiments,
were characterised by circumstances conducive to steady and
unbroken accumulation, followed by favourable maturing
conditions, rather than by individual periods of more than
ordinarily active luxuriance.

AVEEAGE YIELD OF BAELEY PEE ACEE PEE
ANNUM FOE EACH DESCEIPTION OF MANUEE.
The second section of the report of experiments on the
growth of barley for twenty years in succession on the same
land deals with the average annual produce by each descrip-
tion of manure employed. On the unmanured plot the
average annual produce of barley over the twenty years was
about twenty-one bushels of dressed grain, and 12cwt. of
straw. The quality, as indicated by the weight per bushel1
of grain, was higher over the second than over the first
period of ten years; but the quantity of both grain and
straw was between 23 and 24 per cent, less over the second
ten years. Compared with wheat grown for many years in
succession without manure, barley gave an average of more
grain, less straw, and nearly the same weight of gross
produce (grain and straw together) ; but the barley fell off
more in produce of grain, and about equally in straw, over
the later years. Prior to the commencement of the experi-
ments, the previous cropping had been :

WHEAT FIELD. BARLEY FIELD.

Turnips (dunged). Turnips (dung and superphosphate)

carted off.

Barley. Barley.

Peas. Clover.

Wheat. Wheat.

Oats. Barley (sulphate of ammonia).

It is possible, therefore, that there would be rather more
nitrogenous condition to work out of the barley land than out

Continuous Application of Farmyard Manure. 97

of the wheat land, and tabulated results lead to the conclusion
that the wheat will eventually maintain a somewhat higher
total produce than the barley. This is what might be expected
with the autumn-sown crop, with its longer period for root
development, and consequent possession of a greater range of
soil for the collection of food.

Farmyard manure, applied yearly for twenty years
(Plot 7), gave an average annual produce of more than 48
bushels of grain and 28cwt. of straw. The weight per bushel,
quantity of grain, and quantity of straw were all considerably
higher over the second than over the first ten years. The
manure probably supplied from three to four times as much
nitrogen as any of the artificial fertilisers, and much more
of carbonaceous organic matter, and of every other con-
stituent of the crop, than was contained in the produce.
It would leave a large residue of nitrogenous, carbonaceous?
and other matters in the soil, which seem to be very slowly
available for future crops ; but the large accumulation of
organic matter increases the porosity of the soil, and its
capacity for the retention of moisture ; and the crops are
thereby rendered both less susceptible to injury from excess
of rain, and more independent of drought. As without
manure, so with farmyard manure, barley compared with
wheat yielded, over a series of years, more grain, less straw,
but nearly the same quantity of total produce (grain and straw
together). This is remarkable when it is considered that the
wheat is autumn sown and autumn manured, and the barley
spring sown and spring manured. It is interesting to note
that during the twenty years, 280 tons of farmyard manure,
containing from 80 to 90 tons of dry, solid matter, were
applied per acre ; but the produce only amounted to 24f tons
of grain and 28^ tons of straw, or in all to only 53 tons ; and
the increase over the produce without manure was only
about 14J tons of grain and 16J tons of straw, in all 30j tons
of total increase, which certainly would contain less than one-
third as much dry, solid matter as was supplied in the dung.

H

98 The Rothamsted Barley Experiments.

Mineral manures alone gave very poor crops, and the
quantity of both grain and straw fell off considerably during
the later years, but superphosphate of liine alone gave more
than salts of potash, soda, and magnesia, and not much less
than the mixture of all. It may be concluded that the soil
was not relatively deficient in any of the mineral constituents
which the manures supplied : and, from the falling off in
the produce both without manure and with purely mineral
manure, it is probable that the growth of the crop under such
conditions is gradually exhausting the available nitrogen
accumulated within the soil from previous cultivation,
manuring, and cropping. Mixed mineral manure, containing
salts of potash, soda, and magnesia, with superphosphate of
lime, gave, of barley, much more grain, rather less straw,
but considerably more total produce than of wheat. It is
probable that, with the autumn manuring for the wheat, the
various constituents are distributed by the rains, or enter
into less soluble combinations, or both, during the winter ;
that hence there is less active root development in the upper
and more highly nitrogenous layers of the soil in the spring,
and that hence the barley is more rapidly exhausting the
accumulated nitrogen of the surface soil than the wheat.
The distinction between " condition " and " fertility " is well
illustrated by the mineral-manured and by the unmanured
plots, for it is apparent that the condition of the soil, as
distinguished from its normal or natural fertility, is, at
any rate, so far as available nitrogen is concerned, being
gradually worked out by the growth of the crop.

Nitrogenous manures alone (ammonia- salts or nitrate of
soda) gave much more barley than mineral manures alone ;
the produce declined much less in the later years ; and
for twenty years in succession even fair, though not large,
crops were obtained. Whilst over the twenty years the
average annual produce was, by the mixed mineral manure
(4 0), only 27^ bushels of grain and 14| cwt. of straw, that
by the 2001b. of ammonia-salts (1 A) alone was 32 J bushels of

Continuous Application of Nitrogenous Manures. 99

grain and ISfcwt. of straw. In other words, whilst the
increase of produce by the mixed manure alone averaged, over
twenty years, only 6| bushels of grain and 2J cwt. of straw
per acre per annum, the increase by this comparatively small
quantity of ammonia-salts alone averaged over the same
period 11| bushels of grain and 6| cwt. of straw. Comparing
the effects of ammonia-salts with those of the same quantity
of nitrogen in nitrate of soda, the latter gave an "average
annual increase of 5J bushels of grain and 4fcwt. of straw
over the former. Owing to the greater solubility and more
rapid distribution in the soil and subsoil of the nitrate or
its products of decomposition, it is more liable to loss by
drainage when there is an excess of rain. On the other
hand, the subsoil in its case becomes more disintegrated,
therefore more porous, more retentive of moisture in a
favourable condition, and more permeable by the roots. It
is probably in part due to this action that the effects of
a given amount of nitrogen as nitrate of soda increase from
year to year compared with those of an equivalent appli-
cation as ammonia-salts. How much of the greater effect of
the nitrate may be due to this action, and how much to the
mineral manure which in the first year of the experiments
was supplied to the nitrated plot, it is impossible to deter-
mine. The fact that fair crops were obtained by nitrogenous
manures alone throughout the entire period is a striking
illustration of the mineral resources of such a soil ; and it
shows that when in what may, in an agricultural sense,
be called a corn-exhausted condition, it was deficient in
available nitrogen relatively to available mineral con-
stituents. Putting together the results obtained with
minerals alone, and with nitrogenous manures alone, it must
now be apparent that the exhaustion noticed in the former
case was due to a deficiency of available nitrogen in the soil.
It is, indeed, of no little interest to know that on a somewhat
heavy soil, consisting of a somewhat heavy loam with a
clayey subsoil, and of only moderate corn-yielding capa-

H 2

100 The Eothamsted Barley Experiments.

bilities, purely mineral manures will not yield anything like a
fair crop of wheat or barley ; but that, on the same soil, a
comparatively small quantity of purely nitrogenous manure
has yielded, for twenty years in succession, not much less
barley than the average crop of the country ; and that
a larger amount gave, over six consecutive seasons, con-
siderably more than an average crop. This is knowledge,
acquired of the available mineral resources of such a soil,
which chemical analysis would not have afforded, and which
supplies, if not examples for exact imitation, at any rate a
very sound basis for deduction in regard to actual practice.

By ammonia-salts (2001b. per acre) and superphosphate of
lime (3|cwt. per acre) together, an average produce of more
than 47 bushels of dressed grain, and more than 27Jcwt. of
straw, or considerably more than the average barley crop of
the country, was obtained over twenty years in succession ;
and the produce of grain increased, and that of straw in a
less degree diminished, giving a higher total produce during
the later than the earlier years. Notwithstanding the great
demand made upon the supplies of potash within the soil,
by the growth of the crop for so many years, by ammonia-
salts and superphosphate without potash, the addition of
salts of potash, soda, and magnesia gave no further increase
of grain, and very little of straw and total produce. The
potash-yielding capabilities of such a soil, and the beneficial
effects of the use of superphosphate with nitrogenous
manures for spring-sown corn crops, are here strikingly
illustrated.

When the same mixed mineral manure (3001b. the first six
years, and afterwards 2001b., sulphate of potash ; 2001b. the
first six years, and afterwards lOOlb., sulphate of soda;
lOOlb. sulphate of magnesia) and 2001b. of ammonia-salts
were applied per acre per annum for twenty years, in the
autumn for wheat and in the spring for barley, the barley
gave more than one-half more grain, nearly one-sixth more
straw, and nearly one-third more total produce than the

Nitrate of Soda versus Ammonia-salts. 101

wheat. When the same mineral manure was used with a
larger quantity of ammonia- salts, the result was still in
favour of the barley, but in a less degree than with the
smaller amount.

After applying 4001 b. of ammonia- salts per acre per
annum to barley for six years, and then reducing the
amount to 2001b., the plots so treated gave for ten years
in succession more produce than those which had only
received 2001b. annually from the commencement. From
this it is inferred that the excessive supply of 4001b. had left
a residue of nitrogen within the soil which was available for
succeeding crops.

The experiments on barley with nitrate of soda and
ammonia- salts respectively, do not happen to be exactly
comparable with one another ; but, so far as can be judged,
a given amount of nitrogen as nitrate of soda yielded more
produce than the same amount as ammonia-salts, and
especially so in dry seasons. This is probably due to the
greater solubility of the nitrate, or its products of decompo-
sition, and to their action on the subsoil, disintegrating it and
rendering it more porous ; thus affording more surface for
the absorption and retention of moisture and manure, and
greater permeability to the roots, rendering the plants less
dependent on the fall of rain during growth.

A table presenting the average annual produce by 275lb.
nitrate of soda (containing the same quantity of nitrogen as
2001b. of ammonia-salts) per acre per annum, both alone and
with mineral manure, during the four years 1868-71, brings
out the remarkable fact that the average produce is almost
identical by the nitrate alone and by the nitrate and mixed
alkali-salts together. Though much higher, it is again
almost identical by the nitrate and the superphosphate, and
by the nitrate, superphosphate, and mixed alkali-salts. The
little effect hitherto of the potash, soda, and magnesia-salts
is here again illustrated.

Eape-cake is estimated to contain, on the average, about

102 The Rothamsted Barley Experiments.

4*75 per cent, of nitrogen. It also contains a large amount
of carbonaceous organic matter, and about 8 per cent, of
mineral matter. By the annual application of rape-cake,
whether without or with the addition of mineral manures,
more barley than the average crop of the country has been
obtained ; but, in proportion to the nitrogen it contained, less
than by ammonia-salts or nitrate of soda. The nitrogen of
the nitrogenous organic matter of rape-cake is much less
rapidly available than that of ammonia-salts or nitrates ;
indeed, analysis of the soil has shown that the rape-cake has
left a considerable residue of nitrogen near the surface.

Since, for twenty or more years in succession, ammonia-
salts, or nitrate of soda, with mineral manure (without
silica), have yielded considerably more of both wheat and
barley than the average crops of the country, and more also
than either farmyard manure or rape-cake, it follows that the
return to the soil of carbonaceous organic matter as manure
is unessential, so far as the successful growth of either of
these crops is concerned.

Thus, from a comparison of the produce obtained by the
different manures, it has been shown that carbonaceous
organic matter, supplied so largely in farmyard manure and
rape-cake, is, at any rate, not essential as manure for either
wheat or barley ; that mineral manures alone will not yield
fair crops of either; that nitrogenous manures give much
more produce than mineral manures alone ; and that the
mixture of nitrogenous and mineral manures will give full
crops for many years in succession. In other words, the
supply by manure of matter yielding, by decomposition,
carbonic acid and other carbon compounds within the soil
has little or no effect ; mineral manures alone will not enable
the growing plant to obtain sufficient nitrogen from the soil
or the atmosphere. When nitrogen in an available form was
liberally provided, the mineral constituents of the soil were
insufficient for its full effect ; but when so supplied, the
mineral manures, which alone had little effect, greatly

Ratio of Increase in Crop to Ammonia supplied. 103

increased the efficacy of the supplied nitrogen. The general
result is that, whilst it is essential that there should be &
liberal provision of mineral constituents, the amount of
produce is more dependent on the supply of available nitrogen
within the soil than of any other* constituent.

QUANTITY OF AMMONIA EEQUIEED TO PEODUCE
AN INCEEASE OF ONE BUSHEL PEE ACEE IN
THE BAELEY CEOP.

The third section of the report of experiments on the
continuous growth of barley for twenty years in succession
on the same land discusses the amount of ammonia in
manure (or its equivalent of nitrogen in other forms)
required to yield a given increase of grain (and its propor-
tion of straw) ; also how much the quantity will vary,
according to the amount applied per acre, to the supply of
mineral constituents, and to the characters of the seasons.
A comprehensive table is constructed to show the amount of
ammonia — or of nitrogen in nitrate of soda, or in rape-cake,
or in farmyard manure, reckoned as ammonia — required to
yield one bushel (521b.) of increase of barley grain, and its
proportion of straw, under a great variety of conditions of
manuring, and in each of the twenty seasons. In each case
increase is calculated over the produce on the corresponding
plot without nitrogenous manure, the increase on the nitrate
of soda and the farmyard manure plots being taken over the
mean unmanured produce.

One example in detail will suffice to indicate the kind of
information deducible from the table, which is far too
extensive to reproduce. Five plots each received 2001b.
ammonia - salts per acre per annum for twenty years,
but in other respects were variously treated as follows :
(1) Without any mineral manure ; (2) with superphos-
phate ; (3) with sulphates of potash, soda, and magnesia ;

104 The Rothamsted Barley Experiments. -

(4) with superphosphate and sulphates of potash, soda,
and magnesia ; (5) with superphosphate and sulphate of
potash. Taking the average for the twenty years in each
case, the quantity of ammonia required to produce one
bushel increase of barley, and its proportion of straw, was,
on the three plots with superphosphate (plots 2, 4, and 5),
2-131b., 2'411b., and 2'lOlb. : on Plot 3, with salts of potash,
soda, and magnesia, without superphosphate, 3 \591b. ; and
on plot 1, without any mineral manure, 3'681b. Thus, taking
the mean of the three experiments with superphosphate, the
amount of ammonia required was rather under 2£lb. ; but
with the mixed alkali-salts without superphosphate, and
without any mineral manure at all, it was between 3^1b. and
3|lb. That is to say, a given amount of ammonia-salts was
more than one and a half times as effective when there was a
liberal provision of mineral constituents, but especially of
phosphates, within the reach of the roots, than when there
was not. Assuming that, with otherwise favourable soil
conditions, and with an application of not more than 501b. of
ammonia per acre, an increase of one bushel of barley (521b.)
and its straw may, on the average of seasons, be obtained
for every 21b. to 2|lb. of ammonia applied, still the table
indicates that the amount may vary very greatly according to
the characters of the seasons. Thus, on Plot 2, with super-
phosphate, only about IJlb. was required in the favourable
seasons of 1863 and 1869 ; but in the bad seasons of 1853 and
1856, 5'361b. and 4'481b. respectively were necessary.

Before briefly summarising the conclusions established
in this section, one or two minor points demand notice.
During the last eight years of the twenty, one or two series
of plots, which otherwise were receiving the same manure,
also received 4001b. of silicate of soda per acre per annum.
Almost every year the eye detected a marked effect from the
silicate on the plots where no superphosphate was used, but
comparatively little, if any, on the plots with superphos-
phate. So striking was this result, that the silicate was

Behaviour of Silicates. 105

analysed to ascertain whether it contained any superphos-
phate. It was found not to contain any ; nor did it contain
nitrate, or nitrogen in any other form. It is suggested that
by the action of the alkaline silicate on the soil, otherwise
locked-up phosphoric acid was rendered available for the
plants. But it is possible that, when the superphosphate
was used, a secondary result of its action within the soil was
the liberation of silicates, which, without it, were not
available in sufficient quantity ; and hence the little effect of
the direct supply of silicates where the superphosphate was
used, and the marked effect where it was not employed.
Another suggested explanation is, that when the acid phos-
phate and the alkaline silicate are mixed together, they are
rendered comparatively insoluble and inactive. The result
may perhaps be due in part to more than one of these
actions. Another remarkable result was that after mineral
manures once applied, nitrate of soda alone, for nineteen
years in succession, yielded a result, in proportion to its
nitrogen, comparatively little inferior to ammonia-salts used
every year in conjunction with superphosphate, or with
superphosphate and salts of potash, soda, and magnesia.

From a review of all the data brought forward in this
section is drawn the practical conclusion that when an increase
of barley is obtained by means of artificial manures, such as
sulphate of ammonia, or nitrate of soda, or Peruvian guano,
an increase of one bushel of grain (521b.), and its proportion
of straw (say 631b.), may, taking the average of seasons, be
calculated upon for every 21b. to 2Jlb. of ammonia, or its
equivalent of nitrogen (1'65 to l'861b.), supplied in the
manure — provided the amount applied be not excessive, and
provided there be no deficiency of mineral constituents within
the soil. These conditions are fulfilled when barley, grown
after dunged roots carted off, or after another corn crop, is
manured by from l^cwt. to 2cwt. of sulphate of ammonia, or
Ifcwt. to 2Jcwt. of nitrate of soda, with 2cwt. to 3cwt. of
superphosphate per acre ; or, from 3cwt. to 4cwt. of Peruvian

106 The Rothamsted Barley Experiments.

guano, containing 12 per cent, of ammonia, without super-
phosphate. But when rape-cake is used, rather more
nitrogen in that form will be required to yield a given
increase of the crop for which it is applied ; a given amount
of nitrogen costs more in rape-cake than in either sulphate
of ammonia or nitrate of soda. When the increase is obtained
by sheep-folding, or by farmyard manure, very much less will
be obtained in the first crop, in proportion to the nitrogen
contained in the manure.

These generalisations may be compared with corresponding
ones, based on the continuous growth of wheat for twenty
years on the same land. For wheat, as for barley, the
quantity of increase obtained for a given amount of ammonia,
or its equivalent of nitrogen, in manure, varies exceedingly
according to the amount applied, to the provision of mineral
constituents within the soil, and to the seasons. But, under
the conditions most comparable with those of ordinary
practice, approximately 51b. of ammonia, or its equivalent of
nitrogen, were on the average required to yield one bushel
increase of wheat, and its proportion of straw. Now, one
bushel of wheat may be reckoned to weigh 611b., and its
average proportion of straw 1051b. Thus, whilst from 21b. to
2^1b. of ammonia in manure will yield 521b. of barley-grain,
and 631b. of straw (=1151b. total produce), it required 51b.
to yield 611b. of wheat-grain, and 1051b. of straw (=1661b.
total produce). It is clear that it required much more
nitrogen in manure to yield a given amount of increase of
produce when applied in the autumn for wheat, than when
in the spring for barley.

EFFECTS OF THE UNEXHAUSTED RESIDUE OF
MANUEES.

The fourth section of the report is on the effects of the
unexhausted residue from previous manuring upon succeeding
crops, on the loss of constituents by drainage, and 011 some

Unexhausted Residue of Manures. 107

allied points. One example of the many brought under
notice will suffice to show the kind of information it is
sought to obtain under this head. Two plots, 4 A and 4 A A
received the same amount and description of mineral manure
every year from the commencement. In addition, 4 A
received ammonia-salts at the rate of 2001b. per acre every
year, but 4 A A 4001b., or double the amount the first six
years, and only 2001b., or the same as 4 A, the next ten years.
Any increase, therefore, on 4 A A over 4 A, during the ten
years in which they both received the same amount of
ammonia salts, may presumably be attributed to the extra
amount applied to 4 A A during the first six years.
Tabulated results show that, during the ten years, there was
an excess of produce on 4 A A compared with 4 A, due to the
unexhausted residue of the previous nitrogenous manuring,
of nearly 28 bushels of grain, and just 28cwt. of straw, or
an annual average of 2| bushels of grain and 2|cwt. of straw.
Moreover, the excess in the tenth year was almost exactly the
same as the average of the ten years, showing that the
residue was not even then exhausted. There was, then, in
this case, a marked effect upon the succeeding barley crops
from the extra ammonia-salts applied in the first six years.
Similar experiments, with nitrogen supplied in the form of
nitrate of soda, showed that, with the nitrate, as with the
ammonia-salts, there was a somewhat lasting effect from the
extra amount applied during the earlier years.

In this section there is discussed at considerable length
the amount and destination of the nitrogen of the soil, and
Table XXIII., page 108, which explains itself, supplies for
comparison the results of experiments, not only on barley,
but on wheat and oats.

It may be explained that the increase in the amount of
nitrogen in the produce by the use of it in manure is, in the
case of the artificial mixtures of nitrogenous and mineral
manure, calculated over the amount determined in the
produce by the corresponding mixed mineral manure without

108

The Rothamsted Barley Experiments.

_g

a

a

i

1

« Is

<£> rH »C l^,t-^

i

£a5

PS

OS <M 1> CO

2

||

rH 0

3

P Ii

r» i> co I-H -^ 10

1

j!

i-H O CO OS

1/5 i/^ co co

|

|i

33

s

5 "o

.£

"o

a

s

0

1— 1

= ^

8

*

*

£

* PilODUCE, FOB

For 100 nitrog

Recovered in
Increase.

Tj< OS »/5 »C CO »
CO CO CO <M •<* rH

For 100 nitrog

a

r-l CO CO t-

For 100 nitroj

Becovered in
Increase.

01 -<f

1

i
i

O

•5

•

::::::

'

|

::::::

|| ft a s s

S

: : : : •

1

: : : : :

• 1-H 1— 1

I

(— i

• • •

^

d

2«»

g

CO

g)

J^

9"^

2

-3 *t

0 p eg

| s a s s

s

i

co

I-H

2

.-

.a

« <) *

a

1—t

xj ^ & *& PO ^

«

'2

'-i

> s ^

g

,0 j£ ^s ,a -a

09*

c

ri

g

I2 1

|

£H N CO ^1 W
T}( CO <N CD CO

1

ci

ii ii ii n ii T

B"

9
cS

sg

2

^«0

o 3 <"
15 & i

1
2

II II II II J^ •

^^^.^ 53 :

!

P.

2

1 a s»1 \

f

&

2

JL "

1

— — ""

03 — i"

^j

ri -S i 1

"*^

y ^

J

5 |

1

s

1

a s ,.j ^ s ^

°

H

si

0

§

|

08 'e'^ |

2

,0 ^3 ^3 ,0 ^3 ^5 ^

2

a'S

.0,

|

1

^ rj X X rj o

•^ 1— 1 £5 S ^H ^"

° o o o o d
o o o o »o a

B

O O O iO O O m

O O O l> O O JU

M <N «-" §

1

ff

^

1 s t s|

1

C s ^ S

lt

1

fl ^

>>

(3

CO

1

a 1

g

g

-

g

« -a

|

I

S

'a §

i §

a

1

a 2 ^

a

a

— ^

M

B

** rH

S iH

"

-1

H

3

1

|

«O t» CO ?O OS <N

1

41 4 4? r.

1

^«

H

Fate of the Nitrogen of the Soil 109

ammonia. The increase in the produce of nitrogen by
farmyard manure is also calculated over that by the purely
mineral manure. An inspection of the table will show that
when the more excessive amounts of ammonia-salts were
applied for wheat, notably less than one-third of the supplied
nitrogen was recovered, and the less the greater the excess.
The general result is that with neither crop is the whole of
the supplied nitrogen recovered in the increase of produce
obtained; that when a given amount of ammonia-salts
was applied, a much less proportion was recovered in wheat
than in either barley or oats ; but that, even with wheat?
more was recovered when nitrate of soda was employed
than when ammonia-salts were used. The nitrogen applied
in the manure, but not recovered in the increase of crop,
may be lost either by passing away in the drainage waters,
or by accumulating within the soil in a state of combination
or distribution unfavourable for being taken up by the
immediately succeeding crop. The facts elucidated by this
section of the investigation may now be briefly summarised.

When either ammonia-salts, or nitrate of soda, or nitro-
genous organic manure in the form of rape-cake, or farm-
yard manure, was applied for either wheat or barley, a
considerable proportion of the nitrogen so supplied remained
unrecovered in the increase of the crop for which the
manure was employed; nor was the whole recovered in
many succeeding crops. When ammonia-salts were applied
in the autumn for wheat, a much less percentage of their
nitrogen was recovered in the increase of crop, than when
they were applied in the spring for barley or for oats.
Analyses of the soils to the depth of 27in. revealed a
considerable accumulation, within that depth, of the nitrogen
of manure which had not been recovered in the increase of
the crop ; but indicated that a still larger amount remained to
be otherwise accounted for. Analyses of the drainage waters
showed that they contained a large amount of nitrogen as
nitrates ; that the quantity of nitrates in the drainage was

110 The Rothamsted Barley Experiments.

the greater, the greater the amount of ammonia-salts
applied as manure ; that (after autumn sowing) the quantity
was very much greater in the winter than subsequently in
the spring and summer; that, indeed, the winter drainage,
after sowing ammonia-salts in the autumn, may often
contain from two to three parts (and sometimes much
more) of nitrogen (in the form of nitrates and nitrites)
per 100,000 parts of water. Calculation showed that, for
every one part of nitrogen per 100,000 parts of drainage,
there would be a loss of 2Jlb. of nitrogen per acre for every
inch of rain passing beyond the reach of the roots. Inas-
much as a given surface of soil possesses much less capacity
of absorption for nitrate of soda, or its products of decom-
position, than for the ammonia of ammonia-salts, it follows
that heavy rains soon after sowing would carry away in the
drainage water more nitrogen from a dressing of nitrate of
soda than from an equivalent dressing of ammonia-salts
In one case, after a heavy dressing of nitrate of soda in the
spring, Dr. Voelcker examined the drainage waters, and
found 5*83 parts of nitrogen per 100,000 of water, corres-
ponding to a loss of 131b. of nitrogen per acre, per inch of
rain so passing. Owing to the much less loss by drainage
in the case of spring than of winter sowing, there was not
only more increase in the immediate crop from a given
amount of nitrogen applied in the spring for barley (or oats)
than in the autumn for wheat, but there was also much
more effect upon succeeding crops, from the at first
unrecovered amount, in the case of the barley than in that
of the wheat. Respecting the fate of the nitrogen supplied
as manure in ammonia-salts, or in nitrate of soda, it is
probable that the whole of this element is either recovered
in the immediate increase of crop, or retained in the soil in
a very slowly available condition, or drained away and lost.
Owing to the slow decomposition of the nitrogenous organic
matter of rape-cake and farmyard manure, their nitrogen is
less rapidly available than that of ammonia-salts or of

Results of Similar Experiments compared. Ill

nitrate of soda; but so far as could be judged from direct
experiments on the point, it would appear to be, at the
same time, less subject to loss by drainage. Certain impor-
tant mineral or ash constituents of manures — potash and
phosphoric acid, for example — are, at any rate, in the case
of the heavier soils, almost wholly retained by them within
the range of the roots ; and they are found to be very
lasting in their effects upon succeeding crops, provided
there be a sufficient available supply of nitrogen within the
soil.

THE EXPEEIMENTS ON THE CONTINUOUS
GROWTH OF BAELEY COMPAEED WITH OTHEE
SIMILAE EXPEEIMENTS.

As in the case of the wheat experiments, so with those on
barley, it may reasonably be objected that results obtained
on one field alone are scarcely susceptible of general applica-
tion until they have received some degree of confirmation
based on similar experiments on other land. The investi-
gators have provided against this objection by means of two
sets of experiments, in one of which barley was grown for
three years in succession on a series of plots which had pre-
viously been differently manured, and had grown ten crops
of turnips in succession ; whilst, in the other, barley has been
grown in four-course rotation, without manure, and with
different descriptions of manure.

In the Barn Field, barley was grown for three years, after
ten years (1843-52) of turnips. The average produce of
turnips over the last eight years (1845-52) was :

Boots.

Leaves.

Total.

1. Mineral manure alone

Tons cwt.
7 9

Tons cwt.
1 10f

Tons cwt
8 19$

2 Min. man. and ammonia- salts

10 4f

3 3

13 7|

3. Min. man. and rape-cake

10 19|

2 13£

13 12£

4. Min. man., amm.-salts, and rape-cake

12 Si-

4 71

16 Hi

112

The Rothamsted Barley Experiments.

Thus No. 1 gave the smallest, and No. 4 the largest, pro-
duce. Table XXIV. records the total produce (grain and
straw) in the three succeeding years during which barley was
grown, and for comparison there is introduced in the top line
the yield without manure in the second, third, and fourth of
the twenty years in which barley was continuously grown in
the Hoos Field.

TABLE XXIV. — YIELD OF BARLEY UNDER VARIOUS CONDITIONS.

Produc<

5 (grain anc

straw) per

acre, Ib.

1853.

1854.

1855.

Average
3 years.

Hoos Field-
Barley, without manure, after
three corn crops
Barn Field —
Barley, after 10 years turnips,
manured as under :
1. Mineral manures (last 8 years)..
2. Min. man. (8 years) ; ammonia-
salts (6 years)

3467

2618
2864

4462

2474
2691

3923

2206
2331

3951

2432
2629

3. Min. man. (8 years) ; rape cake
(6 years)

3558

3171

2712

3147

4. Min. man. (8 years) ; amm.-
salts and rape cake (6 years)..
5. Min. man. (8 years) ; amm.-
salts for barley, 1854

3546

(2618)

3136
7377

2555

2852

3079-
5114

6. Min. man. (8 years) ; nitrate
soda, for barley, 1854, 1855 ...

(2618)

8005

4727

6366

The average number of bushels of grain yielded during the
three years, 1853-55, was, on the unmanured plot in Hoos
Field, 31f ; and on Nos. 1, 2, 3, 4, 5, 6, in Barn Field, it was
20, 22, 25}, 25|, 39f , 47f bushels respectively. It appears,
then, that the mineral-manured turnips were followed by
three insignificant barley crops, much smaller than when
barley was grown after three immediately preceding corn
crops. In what constituent or constituents had the mineral-
manured turnips so exhausted the soil as to bring it into a
condition even far worse for the after- growth of barley than

Exhaustion of Soil.

when (after clover) three white straw crops had been grown
in succession — namely, wheat without manure, barley with
sulphate of ammonia, and barley without manure ? An
examination of the manuring and the results in experiments
2, 3, 4, as shown in the table, can leave no doubt that the
increased produce was due to an increased supply of avail-
able nitrogen within the soil where it had been applied in the
manures for the turnips. Still, in neither case is the pro-
duce so great as without manure in Hoos Field, where the
barley was grown after several previous corn crops. Experi-
ments 5 and 6, however, afford conclusive evidence that it
was of available nitrogen for the barley that the soil had.
become so exhausted by the growth of ten successive crops of
turnips. It may be seen in the table that, in the year 1854,
these two plots, by the simple addition of nitrogen in the
manures, increased the total produce three to three and a half
times.

Evidence of yet another kind indicates that it was of avail-
able nitrogen that the turnips had rendered the soil so
deficient for the after- growth of barley. It may be assumed
that an average of from 251b. to 301b. of nitrogen would be
annually removed from the Rothamsted soil by wheat or
barley griown year after year without nitrogenous manure.
But it is estimated that from the mineral-manured turnip
plots there were, over the ten years, more than 501b. of
nitrogen per acre per annum removed, After making every
allowance, it may fairly be asstimed that about one and a
half times as much nitrogen was removed from the land for
eight, if not for ten years, in succession, as would have been
taken in an equal number of crops of wheat or barley grown
without nitrogenous manure. No wonder, then, that con-
siderably less barley was grown in three years, after a series
of mineral-manured turnip crops, than was obtained in
another field after a less number of corn crops. The results
in Barn Field, moreover, afford a striking illustration of the
dependence of the turnip plant on a supply of available

114

The Rothamsted Barley Experiments.

nitrogen within the soil, and of its comparatively great power
of exhausting it.

The experiments on barley in four-course rotation (turnips,
barley, clover or beans, wheat) were commenced in Agdell
Field in 1848, and have been continued ever since. The area,
of about 2J acres, was divided into three equal portions.
One-third has been left entirely unmanured from the outset ;
one-third has been manured with superphosphate of lime
alone once every four years — that is, for the turnip crop
commencing each course ; and one-third, also for the turnip
crop only, with a complex manure, consisting of superphos-
phate of lime, salts of potash, soda, and magnesia, sulphate
and muriate of ammonia, and rape cake. From half of each
plot the whole turnip crop (roots and tops) was removed ; on
the other half the roots were consumed on the land by sheep,
and the uneaten leaves spread and ploughed in. In the first
course, clover was grown as the third crop ; but in the second
to the sixth courses, instead of clover, half of each plot was
sown with beans, and the other half left fallow. The average
produce of turnips per acre over the first five courses (the
crop failing in the sixth, owing to drought) was —

Boots.

Leaves.

Total

TTithout manure

Tons cwt.
1 6$

Tons cwt.
0 10*

Tons cwt

1 17i

^Tith superphosphate only

6 16*

1 8

8 4£

12 2*

2 2*

14 5

Without manure, therefore, there was scarcely any produce
of turnips ; consequently, there was no exhaustion of the land
by removal of the crop, so that it was, practically speaking,
left fallow for the barley. With superphosphate alone only
small crops of turnips were grown, especially in the later
courses ; still, much more was removed from the land than
without manure ; and as nothing was supplied besides what
the superphosphate itself contained, the land was, so far as

Four-course Rotation.

115

other constituents are concerned, left much more exhausted
for the growth of the barley than without any manure what-
ever. With the mixed manure fair crops of turnips were
removed in the earlier, but less in the later courses, and
(excepting in the first year) there would remain in the land a
considerable residue from the manures applied ; and hence it
would be left in a higher condition for the barley than after
either the unmanured or the superphosphated turnips. Table
XXV. shows the total produce (grain and straw) of barley in
each of the successive courses, and also the average of the
six courses ; for comparison, the top line shows the produce
in the same seasons of the unmanured plot in Hoos Field,
where barley is grown year after year without manure :

TABLE XXV. — YIELD OP BARLEY UNDER ROTATION.

Produce (grain and straw) per acre, Ib.

manures.

1849.

1853.

1857.

1861.

1865.

1869.

Averaga

Hoos Field-
Barley, unmanured,
after three corn
crops

3467

3295

2107

2042

2016

2585

Agdell Field-
Barley, in 4-conrse
rotation :
Unmanured, con-
tinuously
Superphosphate, for
turnips only
Mixed manure, for
turnips only

5656
3841
3794

4465
3560

4873

5337
3076
5168

4718
3775
7391

4182
3394
5148

3358
3686
5800

4619
3555
5363

The apparently anomalous results of the first year are
chiefly attributable to the fact that there had been removed
from the unmanured plot only about 3j tons of roots, and
2J tons of tops, but from the mixed manure plot nearly
11 tons of roots, and more than 7| tons of tops; in the
manured plot, further, there would be left much less, if any,
residue of manurial constituents for the barley of the first

i 2

116 The Rothamsted Barley Experiments.

course, than for any of the subsequent courses. The table
skows that the quantity of barley grown in rotation without
manure is very much greater than that grown in succession
without manure. Again (omitting the first year), the produce
after the removal of the full-manured and larger crops of
turnips was uniformly, and on the average, very much higher
than after the removed superphosphated turnips, and also
generally, and on the average, higher than after the un-
manured turnips ; this larger produce of barley, after the
removal of the larger crop of turnips grown by the mixed
manured, is doubtless due to the fact that there would still
be a considerable residue of the manure left within the
soil. The results of experiments already discussed have
shown that a liberal supply of mineral constituents alone
was insufficient to secure a fair crop of barley, but that the
further addition of nitrogenous manure raised the produce
to a maximum. It may, therefore, be concluded, in the case
now under notice, that the larger produce of barley after the
full-manured, than after the superphosphated or unmanured
crops in rotation, was not attributable to any residue of
mineral constituents alone which would remain after the
removal of the highly manured roots ; and that the larger
produce after the unmanured than after the super-
phosphated turnips was not due to a less exhaustion or
greater accumulation of available mineral constituents
where the smaller crop of turnips was removed. In fact^
as in other experiments, so also in these, in which barley was
grown in rotation, and under three very different conditions
as to manuring, the evidence is sufficiently conclusive, and
therefore corroborative of that in the other cases, that an
essential condition for the growth of a full crop of barley,
whether in rotation or under less usual conditions, is a liberal
supply of available nitrogen within the soil.

Summary of Twenty Years' Results. 117

SUMMARY OF THE RESULTS OF TWENTY YEARS'
EXPERIMENTS ON THE CONTINUOUS GROWTH
OF BARLEY UPON THE SAME LAND.

It remains now to draw such general conclusions as seem
warrantable from the results of the experiments on the con-
tinuous growth of barley on the same land, and to show the
practical bearings of the results. The barley field imme-
diately adjoins that devoted to the experiments on wheat,
and has a soil and subsoil of similar general character, and
described as a somewhat heavy loam, with a subsoil of raw
yellowish-red clay, but resting in its turn upon chalk, which
provides good natural drainage. Though, in wet seasons,
such a soil is not well suited for the growth of the crop after
roots fed on the land by sheep, as is the custom of the
locality, yet the recorded results abundantly prove that,
when grown under favourable conditions, large crops of
barley, of good quality, may be obtained from such land.
Without manure, the average produce over twenty years
was 21 bushels of dressed barley grain, of 5 2 Jib. per bushel,
and 12cwt. of straw. Though the quantity fell off con-
siderably, the quality was considerably higher over the
second than over the first ten years. Compared with wheat
without manure, barley gave more grain, less straw, but
nearly the same quantity of total produce. It, however,
fell off more in produce of grain, and about equally in straw,
over the later years. By farmyard manure, the average
annual produce was more than 48 bushels of dressed grain,
of 54|lb. per bushel, and 28cwt. of straw. The quantity
of both grain and straw, and the quality of the grain,
were considerably higher over the second than over the
first ten years. As without manure, so with farmyard
manure, barley, compared with wheat, yielded more grain,
less straw, but much about the same quantity of total
produce. Mineral manures alone gave very poor crops,

118 The Eothamsted Barley Experiments.

and the quantity of both grain and straw fell off con-
siderably during the later years. With barley there was
much more grain, rather less straw, but considerably more
total produce than with wheat. Nitrogenous manures alone
gave much more barley than mineral manures alone ; the
produce declined much less in the later years ; and, for
twenty years in succession, fair, though not full, crops were
obtained. Nitrogenous and mineral manures together gave,
for twenty years in succession on the same land, rather more
of both grain and straw than farmyard manure, considerably
more than the average barley crop of the country under
rotation, and an average weight per bushel of between
531b. and 541b. With the same amount of nitrogen and
the same mineral manure, applied for twenty years, in the
autumn for wheat and in the spring for barley, the barley
gave much more grain, more straw, and nearly one-third
more total produce than the wheat. Thus, then, with barley
as with wheat, mineral manures alone failed to enable the
plant to obtain sufficient nitrogen and carbon to yield even a
fair crop ; the greater effect of nitrogenous manures alone
showed that the soil, in its practically corn-exhausted con-
dition, was relatively richer in available mineral constituents
than in available nitrogen ; and the generally greater effect
by nitrogenous and mineral manures together than by farm-
yard manure — which contained not only very much more
nitrogen, but a large amount of decomposing carbonaceous
organic matter, and probably more of every mineral con-
stituent than the crop — showed that the nitrogen of the
farmyard manure was in a far less rapidly available con-
dition, and that its supply of carbon was at any rate
unessential.

Though it is a mere truism to assert that the growing
plant must have within its reach a sufficiency of the mineral
constituents of which it is to be built up, yet the results
obtained with barley, as well as those with wheat, have
shown that, whilst it is essential that there be a liberal pro-

Effects of Nitrate of Soda. 119

vision of mineral constituents within the soil, the amount of
produce is more dependent on the supply by manure of
available nitrogen, than of any other constituent. The
question has been discussed as to how much ammonia, or its
equivalent of nitrogen in some other form, will, on the
average, be required to yield a given amount of increase of
wheat or barley grain, and its proportion of straw. It was
concluded that, on the average, about 51b. of ammonia (or
its equivalent of nitrogen) were required to yield one bushel
of increase of wheat, and its proportion of straw, whilst
only 21b. to 2Jlb. of ammonia are required to produce one
bushel increase of barley, and its straw. In other words,
whilst it required 51b. of ammonia in manure to yield 611b.
of wheat grain, and 1051b. of straw = 1661b. of total pro-
duce, it only requires from 21b. to 2Jlb. to yield 521b. of
barley grain and 631b. of straw = 1151b. of total produce.
Or, for the production of lOOlb. increase of total produce of
wheat, it required 31b., and for the production of lOOlb.
increase of barley (containing a larger proportion of grain,
but about the same amount of nitrogen), it required only
from l|lb. to 21b. of ammonia in manure ; that is to say, it
required much more ammonia to yield a given amount of
increase when applied in the autumn for wheat, than when
in the spring for barley.

In connection with the action of nitrate of soda, it must
not be overlooked that a given surface of soil has much less
power to retain either nitrate of soda, or other nitrates,
than ammonia, and so far their nitrogen is, coder is paribus,
more liable to loss by drainage. Yet when frequently used
on the same land, such was the effect of the nitrate, or of
its products of decomposition, aided by increased develop-
ment of root, in causing the disintegration, and so increasing
the porosity and surface of the clay subsoil, that there
would appear to have been not only a greater retention of
moisture in an available form by the subsoil, rendering the
growing crop more independent of drought, but also a

120 The Roihainsted Barley Experiments.

greater retention of nitrates than would be anticipated con-
sidering their solubility, and hence a more lasting effect
from previous applications than would otherwise be expected.
On the other hand, where, as in the case of the experiments
at Rothamsted, nitrate of soda has been used in large
quantities so many years in succession, the surface soil has
retained so much moisture as to be difficult to work after
wet weather. It may be added that the greater liability to
loss by drainage of the nitrogen, than of the more important
mineral constituents of manure, is doubtless one factor in
the explanation of the circumstance of the prevailing excess
of available mineral constituents, relatively to available
nitrogen, in soils generally, under the ordinary course of
agriculture in this country.

It may be laid down as a general rule, applicable to the
country at large,, that, on the heavier soils, full crops of
barley of good quality may be grown with great certainty
after a preceding corn crop, under the following conditions :
The land should be got into good tilth. It should bo
ploughed up when dry, as soon as practicable after the
removal of the preceding crop. In the spring it should be
prepared for sowing by ploughing or scuffling as early in
March as possible, if sufficiently dry. The artificial manure
employed should contain nitrogen, as ammonia or nitrate (or
organic matter), and phosphates. From 401b. to 501b. of
ammonia (or its equivalent of nitrogen as nitrate) should be
applied per acre. These quantities would be supplied in
IJcwt. to 2cwt. of sulphate of ammonia, or Ifcwt. to 2Jcwt.
of nitrate of soda. With either of these there should be
employed 2cwt. to 3cwt. mineral superphosphate of lime.
Rape cake is also a good manure for barley ; from 6cwt. to
8cwt. would supply about as much nitrogen as would be
equal to from 401b. to 501b. of ammonia. With this manure,
as with guano, the addition of superphosphate is unneces-
sary. Whatever manure be used, it should be broken up,
finely sifted, sown broadcast, and harrowed in with the seed.

Nitrogen. 121

Economy in the cost of the nitrogen is the essential point
to be considered in the selection of the manure to be used.
According to the experiments at Rothamsted, it would
appear that, at equivalent prices, a given amount of nitrogen
as nitrate of soda may in the long run be more effective than
an equal amount as ammonia ; for, contrary to the current
opinion, the full effect of the nitrate was not obtained until
it had been used for some years on the same plot. The
liability to loss of the nitrogen of manure in drainage will
vary very much according to the characters of the soil and
subsoil, and of the season. But as it is much greater during
the late autumn and winter months than in the spring and
summer, nitrate of soda, sulphate of ammonia, or Peruvian
guano should always be sown in the spring.

122 The Rothamsted Barley Experiments.

V;iL— EXPERIMENTS ON THE CONTINUOUS
GROWTH OF BARLEY UPON THE SAME LAND
FOR THIRTY-TWO YEARS.

THE experiments on the continuous growth of barley on the
same land have now been in progress for nearly a second
period of twenty years, and it is desirable to examine the
results as nearly as possible down to date. Considering first
the yields without manure and with farmyard manure for a
period of thirty-two years (1852-1883), it appears that,
without manure, the yield in two years — the third and fourth
— was in each case over 30 bushels per acre ; in six of the
first thirteen years the produce was between 20 and 30
bushels, but it never afterwards reached 20 bushels ; whilst
in twenty -four out of the thirty- two years the yield was less
than 20 bushels — in two of these it amounted to only 10
bushels, and in one year (1879) to only 6£ bushels. The
tabulated results indicate that, independent of the fluctua-
tions due to season, there was a progressive decline due to
exhaustion, notwithstanding that the last four years gave a
higher average than any other four in the last sixteen years.
It is further discoverable that, without manure, there is a
decline in the produce of barley grain of 31*4 per cent, over
the second sixteen years, compared with the first sixteen.
This decline is considerably greater than was found in the
case of wheat, a circumstance doubtless due to the shorter
period of growth, and the greater dependence on the surface
soil, in the case of barley, exhaustion being thereby sooner
manifested.

With farmyard manure, as with no manure, there is very

Continuous Application of Farmyard Manure. 123

great fluctuation from year to year, according to season; but,
instead of a gradual decline, there is over the later years an
obvious increase in yield, due to the accumulation of the
manure. There is, in fact, instead of a decline of 2 7' 7 per
cent., as in the case of the unmanured plot, an increase of
7*3 per cent, over the last twelve years, as compared with the
first twenty, although the second period included a number
of the worst seasons of the entire series. In four of the
thirty-two years the farmyard manure gave more than 60
bushels of barley per acre, in thirteen years between 50 and
60 bushels, in ten between 40 and 50 bushels, and in no case
below 30 bushels. The average yield was, over the first
twenty years, 48J bushels, over the last twelve years, 51|
bushels ; and over the thirty-two years, 49J bushels, against
17f bushels without manure. These results were obtained
by the unusual application of 14 tons farmyard manure per
acre per annum for thirty-two years in succession. This
manure supplied over 2001b. nitrogen per acre per annum, or
a total of over 40001b. nitrogen in the twenty years, at the
end of which period it was estimated that not more than
about 15 per cent, of this large amount of nitrogen had been
removed in the increase of crop. Hence there was a great
accumulation, as was proved by analysis, of nitrogen within
the soil, so much so that it was calculated that, if there were
no loss of nitrogen by drainage, by evolution of free nitrogen,
or otherwise, and if the accumulated residue were as available
as that which had already been effective, the produce should
be maintained at the level of that of the first twenty years
for not far short of 150 years more.

After the first twenty years the application of the farmyard
manure was entirely stopped on half the plot, so that the
results obtained might be compared with those on the still
continuously manured half, and on the unmanured plot. The
results prove that there is a general tendency to increase in
yield where the application of the farmyard manure was
continued, and to decrease where it was discontinued; indeed,

124 The Rothamsted Barley Experiments.

during the last four of the twelve years the manure-residue
half plot shows an average of about 24 bushels per acre per
annum less than the half plot where the application was
continued. The average yield per acre over the first twenty
years under continuous manuring was 48J bushels, whilst
over the succeeding twelve years it was, where the manuring
was continued, 51| bushels, but where it was discontinued
only 34| bushels, showing, therefore, an average annual
deficiency, under the influence of the residue only, of 17f
bushels, or of 33*6 per cent. Over the whole period of
thirty-two years the total produce (grain and straw) was
without manure less than 1 ton per acre per annum, whilst
with the farmyard manure it was 2| tons, and in some years
it reached from 3| to 3| tons. It must now be apparent
that, whilst there were gradual exhaustion and reduction of
produce without manure, there were gradual accumulation
and increase of produce with the annual application of
farmyard manure. But when the application was stopped,
although the effect of the residue from the previous applica-
tions was very marked, it somewhat rapidly diminished,
notwithstanding that calculation showed an enormous accu-
mulation of nitrogen as well as of other constituents.
Indeed, determinations of nitrogen in the surface soil, after
the twenty years' application of farmyard manure, showed it
to be nearly twice as high as on the unmanured plot. The
very large amount that accumulates within the soil is in a
very slowly available condition.

It is now necessary to consider what is the character of the
exhaustion induced by the growth of the crop without manure ;
and to what constituent, or constituents, of farmyard manure
its effects are mainly due. These questions are solved bv
examining the results obtained on the plots dressed with
various purely mineral manures, purely nitrogenous manures,
and combinations of the two. A general view of the effects
of sixteen different conditions as to manuring is given in
Table XXVI., which is a summary table, recording the

Results with Various Manures.

125

average produce over thirty-two years, 1852-1883, on each
of the sixteen plots. The first column shows the results for
the four plots without nitrogenous manure, the second
column those for the four plots with ammonia-salts equal to
431b. nitrogen per acre per annum, the third those for the
four plots with nitrate of soda, and the fourth those for the
plots with rape- cake.

TABLE XXVI. — AVERAGE ANNUAL YIELD OP BARLEY UNDER SIXTEEN
CONDITIONS OF MANURING, — THIRTY-TWO YEARS, 1852-1883.

No Nitrogenous
Manure.

2001b. Am.-
salts=431b. N.

2751b. Nitrate of
Soda=431b. N.

&

if

si

1-1 0

Dressed Grain per Acre — Bushels.
Without mineral manure

17|

30f

34f

43i

Superphosphate

23

45

47

451

Sulphates of potash, soda, and magnesia...
Superphosphate, and sulphates of potash,
soda and magnesia

19f

24i

33!

44!

35i

47£

41f

45|

Straw per Acre — 16.
Without mineral manure

1128

1909

2246

2789

Superphosphate

1293

2827

3127

2953

Sulphates of potash* soda, and magnesia...
Superphosphate, and sulphates of potash,
soda, and magnesia ,

Total Produce (Grain and Straiv) per
Acre — Ib.
^Vithout mineral manure

1170
1380

2136

2151
3019

3649

2455
3348

4189

2792
3065

5249

Superphosphate

2584

5374

5791

5532

Sulphates of potash, soda, and magnesia . . .
Superphosphate, and sulphates of potash,
soda and magnesia

2278
2739

4063
5574

4490
6042

5175
5667

Referring first to the results without nitrogenous manure,
the first column plainly shows the very marked effect of
mineral manures containing phosphates as compared with
those consisting of alkalies alone. The second column shows
that ammonia- salts alone gave a great increase, which is
somewhat enhanced by the addition of alkalies, but markedly
so when phosphates are also applied. The plots referred to in

126 The Rothamsted Barley Experiments.

the third column actually received ammonia-salts equal to 861b.
nitrogen the first six years, equal to 431b. nitrogen the next
ten years, and nitrate of soda equal to 431b. nitrogen the
last sixteen years. The increase here, brought about by use
of the nitrate, is seen to be greater still, and is again by far
the greater where the superphosphate is used. As the rape-
cake itself contains phosphates, the construction to be placed
upon the results in the fourth column is that the phosphate
of the rape-cake was effective when none was otherwise
supplied ; but when it was so applied in addition, there was
more effect with the nitrate, with its more rapidly available
nitrogen, than with the rape-cake, with its greater actual
amount of nitrogen, but in a less rapidly available condition.

Since nitrogenous manures alone had much more effect than
mineral manures alone, it is obvious that the exhaustion
induced by the continuous growth of the crops is character-
istically that of nitrogen.

Both with and without nitrogenous supply, phosphates
were more effective than potash salts, showing that the avail-
able store of phosphoric acid in the soil became deficient
sooner than that of potash. With the, shorter period of
growth of barley than of wheat, and its greatei propor-
tion of surface-rooting, both nitrogenous and mineral
exhaustion are sooner developed ; and, so far as mineral
exhaustion is concerned, the available supply of phosphoric
acid was sooner exhausted than was that of potash. In deed 9
in ordinary agricultural practice, it is clearly established that
superphosphate is more effective with the spring-sown than
with the autumn-sown cereals.

CHAEACTEEISTICS OF THE BEST AND WOKST
SEASONS OP THE THIETY-TWO YEAES.

Eespecting the influence of season on the amounts of
produce, the best seasons of the thirty-two were 1854 and
1857 ; the best season for wheat — 1863 — was also a very good

Characteristics of Good and Sad Seasons. 127

year for barley. For almost all conditions of manuring, 1854
was the season of the highest total produce of barley, grain
and straw together — that is, it was the season of the greatest
luxuriance or vegetative activity; but 1857 was, especially
for the highest manuring, the one of the highest produce of
grain and of the highest quality or maturing of grain, as
evidenced by the weight per bushel. Thus, 1854 was the
highest for luxuriance and 1857 the highest for maturation of
the crop. As for wheat, so for barley, 1879 was decidedly
the worst season of the thirty-two. The contrast between the
produce on the barley plots in the two very different good
years and that in the worst season, 1879, is very striking, the
difference amounting in several cases to as much as the
average crop of the country.

Since seasons repeat themselves, the meteorological
characters of 1854, 1857, and 1879, however briefly
summarised, cannot fail to afford interesting comparisons.
In 1854, the season of great luxuriance and high total
produce, there was an excess of temperature in January,
February, March, and April, with a deficiency of rain from
November to April inclusive ; but during May, June, and
July, that is, the months of active above-ground growth,
there were lower than the average temperatures, with a con-
siderable excess of rain in May, and then a deficiency —
conditions obviously favouring continued vegetation and slow
maturation. In 1857 there was less excess of temperature
and less than the average amount of rain to the end of April ;
then from May to August inclusive there was both consider-
able deficiency of rain and considerable excess of temperature,
that is, there were throughout the period of active above-
ground growth conditions favouring seeding tendency and
maturation rather than luxuriance. Thus, then, the two good
seasons were very different in their climatic characteristics,
as they were in the character of their produce. The very
bad season of 1879 registered much lower than average
temperatures throughout the winter, spring, and summer,

128 The Hothamsted Barley Experiments.

and even somewhat in the autumn, with, at the same time,
great excess of rain from January to September inclusive,
whilst both the deficiency of temperature and the excess of
rain were very marked from April to August inclusive, that
is, during the whole period of the above-ground growth and
the ripening, if such it may be called, of the crop, for in many
cases the weight per bushel was less than 501b., whilst the
amounts of produce were very greatly below the average.

VAKIATIONS IN THE CHEMICAL COMPOSITION
OF THE BAELEY CEOP.

The influence of exhaustion of manures, and of variations
of season, on the amount of produce of barley having been
considered, it is pertinent next to inquire what influence
these factors have upon the chemical composition of the
produce. An examination of the results of analyses,
cleinonstrates the greater influence of variations of season
than of manures on the composition of barley grain. There
is much greater variation in the proportion of potash in the
grain, in different seasons with the same manure, than there
is with different manures, and further, the seasons showing
the highest amount of potash are those of much higher
maturing character than those with the lowest amounts.
There is still greater, indeed enormous, variation in the
amount of potash in the dry substance of the straw with the :
same manure in different seasons. There is also great
variation according to manure — comparatively little in.
presence of a full supply, but considerable without manure,
that is, with exhaustion.

As to the phosphoric acid in the grain, there is here again
much more variation in different seasons with the same
manure than with different manures. But whilst of potash
there is the higher proportion in the better seasons, of
phosphoric acid there are lower amounts in the better

Behaviour of Soda and Silica.

seasons. In fact, high amount of potash in the ash, and in
the dry substance of the grain, is, as a rule, associated with
high maturation, that is, with high proportion of starch,
whilst high proportion of phosphoric acid is generally asso-
ciated with low maturation and high proportion of nitrogen.
The proportion of phosphoric acid in the straw also varies
more with season than with manure, and it is the highest in
the worst seasons.

With regard to the behaviour of soda, very much more of
this ingredient was found in the crops grown without its
supply in manure, but where potash was deficient, than
where soda was annually supplied. This is strikingly shown
in the average amounts per acre per annum in the total
crops, grain and straw together. Thus, over a first period of
ten years, 1852-1861, the average amounts of soda in the
total crop were, without any supply in the manure of either
potash, soda, or magnesia, 8'401b., and with the supply of all
three, only 3'841b. ; over a second period of ten years,
1862-1871, without the supply 15'211b., and with the supply
only 3'691b. ; and lastly, over a third period of five years,
1872-1876, without the supply H'851b., and with the supply
only 3'271b. Thus, then, not only was there much more soda
taken up, or retained, by the plant where it was not supplied
than where it was, but there was the more soda taken up the
less the supply of potash.

The chief point of interest with reference to silica is, that
its percentage in the ash of barley grain ranges from 17 to
more than 20, whereas in the ash of wheat grain it ranges
only from about 0*5 to about 1'5 per cent. ; or, taking the pro-
portion of silica to 1000 dry substance of grain, in barley it
ranges from four to five parts, and in wheat only from about 0*1
to about 0'3 parts. This difference is obviously due to the chaff
being adherent in the case of the barley grain, and not in that
of the wheat grain, a circumstance leading to, as might be
expected, less definiteness in the mineral composition of the
grain of barley than in that of wheat.

130 The Rothamsted Barley Experiments.

There exists a popular notion that strength of straw is
dependent on a high percentage of silica, but direct analytical
results clearly show that the proportion of silica is, as a rale,
lower, not higher in the straw of the better- grown and better-
ripened crops — a result quite inconsistent with the usually
accepted view that high quality and stiffness of straw depend
on a high amount of silica. In fact, high proportion of
silica means a relatively low proportion of organic substance
produced. Nor can there be any doubt that strength of
straw depends on the favourable development of the woody
substance ; and the more this is attained the more will the
accumulated silica be, so to speak, diluted — in other words,
show a lower proportion to the organic substance. A large
proportion of brittle straw which breaks in plaiting has been
characteristic of recent seasons of bad harvest, and these
qualities are associated with low development of the woody
matter, and high proportion of silica.

THE HOME PKODUCE AND THE IMPORTS OF
BAELEY.

It is desirable now to inquire into such economic questions
as refer to the extent of the area devoted to the growth of
barley in the United Kingdom, the amount of our total
annual imports, and from what countries these latter are
chiefly derived. Table XXVII. shows the area under barley
in the United Kingdom in each of the years 1873 to 1887,
and also the total imports into the United Kingdom during
the year succeeding each of the first fourteen of the fifteen
harvests, reckoning from Sept. 1 to Aug. 31 in each case.

It is seen that since 1880, when the repeal of the malt tax
took place, there has been, instead of an increase, a dimi-
nution in the home area under barley, accompanied by an
increase in the imports. It would seem that the high duty
served as a bounty on the higher qualities of the home growth,

Acreage and Imports of Barley.

131

and that, when this was removed, the greater demand for
medium qualities has given an advantage to the foreign pro-
ducer. Nor has the removal of the duty led to an extended
use of malt for feeding purposes, which was one of the
main objects for which the repeal was strongly advocated by
farmers.

TABLE XXVII. — ACREAGE AND IMPORTS OF BARLEY (UNITED KINGDOM).

Harvest
Years.

Area.

Imports.

Harvest
Years.

Area.

Imports.

Acres.

Quarters.

Acres.

Quarters.

1873-4

2,574,529

2,391,785

1881-2

2,662,927

3,725,384

1874-5

2,507,130

3,667,174

1882-3

2,452,077

4,398,127

1875-6

2,751,362

2,272,081

1883-4

2,486,137

4,031,722

1876-7

2,762,263

3,684,725

1884-5

2,346,041

4.726,903

1877-8

2,652,300

3,976,384

1885-6

2,447,169

2,801.567

1878-9

2,722,879

2,798,494

1886-7

2,432,749

4,585,600

1879-80

2,931,809

3,467,147

1887-8

2,255,256

1880-1

2,695,000

2,974,892

Taking the sixteen years 1869-1884, Table XXVIII.
shows : (1) the least annual import, (2) the greatest annual
import, (3) the average annual import of barley into the
United Kingdom from each of the countries named.

TABLE XXVIII. — IMPORTS or BARLEY INTO THE UNITED KINGDOM.

Imported from : —

(1) Minimum annual
import of 16 years,
1869-84.

(2) Maximum annual
import of 16 years,
1869-84.

(3) Annual im-
port, mean of 16
years, 1869-84.

Russia
France
Germany

Quarters.
277,108 in 1869
155,203 in 1870
120 776 in 1876

Quarters.
1,544,774 in 1883
1,731,259 in 1872
1 008,144 in 1879

Quarters.

717,838
480,879
477 951

Turkey
Eoumania
Denmark
Sweden

23,391 in 1880
59,448 in 1877
63,525 in 1884
49 305 in 1884

1,059,773 in 1877
1,601,596 in 1882
594.371 in 1879
155 872 in 1882

453,393
420,904

358,847
100,259

United States ...
Egypt
Chili
Other countries..

34,342 in 1870

303,677 in 1878
76,361 in 1876
81,697 in 1884
288,413 in 1884

57,983
26,501
16,274
133,917

Total...

2,020.863 in 1870

4.609.174 in 1883

3,244,746

K 2

132 The Rothamsted Barley Experiments.

Of recent years Russia has been to the United Kingdom
the most important source of barley. France and Germany
show fairly equal average amounts, but Germany has sent
decidedly the more in the later years.

CULTIVATION OF WHEAT AND OF BAELEY
COMPAEED.

Lastly, comparing wheat with barley, it is apparent that
the requirements of barley within the soil, and the suscepti-
bility of this cereal to the external influence of season, are
very similar to those of its near ally, wheat ; but there are
distinctions of result, dependent on differences in the habits
of the plants, and in the conditions of their cultivation
accordingly. Wheat is, as a rule, sown in the autumn, in a
heavier and closer soil, and has four or five months in which
to distribute its roots and get possession of a wide range of
soil and subsoil, before barley is sown. Barley is sown in a
lighter surface soil, and, with its short period for root
development, relies in a much greater degree on the stores
within the surface soil. Hence it is more susceptible to
exhaustion of surface soil as to its nitrogenous, and especially
as to its mineral supplies ; and in the common practice of
agriculture it is found to be more benefited by direct
mineral manures, especially phosphatic manures, than is
wheat when sown under equivalent soil conditions. The
exhaustion induced by both crops is, however, characteristi-
cally that of available nitrogen; and when, under the
ordinary conditions of manuring and cropping, artificial
manuring is still required, nitrogenous manures are, as a
rule, requisite for both crops, and for the spring-sown one,
barley, superphosphate also.

The Rothamsted Grass Experiments. 133

VIII. — AGRICULTURAL RESULTS OF EXPERI-
MENTS ON THE MIXED HERBAGE OF PER-
MANENT MEADOW.

MEADOW herbage offers to the agricultural investigator
about as complex a subject for study as can well be imagined,
and it is one the enormous importance and widespread
application of which need no insistance. Some seven acres of
the land in E-othamsted Park were, in 1856, set apart
for " Experiments on the Mixed Herbage of Permanent
Meadow " and were divided at first into nine and ultimately
into twenty plots. Two of these have been left without
manure from the commencement; two have received ordinary
farmyard manure continuously ; whilst the remainder have
each received a different description of artificial or chemical
manure, the same being, except in special cases, applied year
after year on the same plot. The land has probably been
laid down with grass for some centuries. No fresh seed has
been artificially sown within the last fifty years certainly, nor
is there any record of seed having been sown since the grass
was first laid down. The land is a somewhat heavy loam,
with a red clay subsoil resting upon chalk, and, although not
artificially, is very well naturally, drained ; it is a perfectly
level area. For many years prior to 1851 the general mode
of treatment was to dress occasionally with farmyard manure,
road scrapings, and the like, and sometimes with guano or
other purchased manure. One crop of hay was removed
annually, weighing from 1J to 2 tons per acre; the second
crop was always eaten off by sheep. In the spring of 1851,
and of 1852, four separate acres of the allotted area were

134 The. Bothamsted Grass Experiments.

appropriated to the consumption by sheep of as many lots of
differently manured turnips, 10 tons of the roots being
eaten upon each acre. Neither these four acres nor the
remainder were manured in any other way during those two
seasons, nor were they manured at all in the three succeeding
ones prior to the commencement of the experiments in 1856,
at which time the character of the herbage appeared uniform
over all the plots.

During the first nineteen years — 1856-1874 — the first crop
only each year was mown, made into hay, removed from the
land, and weighed. As a rule, the second crop on each plot
was fed off by sheep, who received at the time no other food,
the object being not to disturb the condition of the manuring.
Frequently, however, the animals suffered considerably, and
in 1866, 1870, 1873, and 1874 the second crops (and third,
if any) were cut and spread on the respective plots. In the
twentieth season — 1875 — the second crops being unusually
heavy, and the weather favourable, they were for the first
time cut, weighed as hay, and removed. In 1876 they were
cut and spread on the plots. In 1877 and 1878 the second
crops were again made into hay, weighed and removed. In
1879 the second crops were cut, sampled, carted, and weighed
green ; the dry matter in the weighed samples was deter-
mined, and the produce reckoned into hay by adding one-
fourth to the calculated dry matter per acre. In 1880 and
the following years the second crop was again made into hay
and removed, and it is intended in future to adopt this plan,
whenever the weather will permit.

The earlier reports of these experiments were communi-
cated by the investigators to the Journal of the Royal
Agricultural Society of England. They are six in number,
and appeared, the first in 1858, and the sixth and last in
1863, and they extend altogether through 180 pages of the
Journal.

For a reason which will presently appear, it is not neces-
sary to examine each of these reports in detail; but,

The Earlier Results. 135

inasmuch as it will convey a good idea of the method and
objects of the experiments, it seems desirable to briefly
indicate the matters dealt with in each report. The first
report is concerned solely with the produce of hay, estimated
in cwts. per acre, yielded by each of the plots, there being
seventeen of these at the time. The second report deals with
the produce of constituents per acre on each plot, and
contains estimates of the dry matter, the mineral matter, and
the nitrogen carried away per acre by the crop on each plot,
as well as estimates of the proportion of nitrogen recovered
in increase for one hundred parts supplied in manures. The
third report embraces a description of the plants, variously
developed by different manures. These plants are appro-
priately arranged under the three heads of gramineous
herbage, leguminous herbage, and miscellaneous herbage.
This report further contains an estimate of the total yield of
each kind of herbage.

The fourth report is concerned with the chemical composi-
tion of the hay, and contains estimates of the dry matter, of
the constituents of the ash, of the nitrogen, of the woody
fibre, and of the fatty matter. The fifth report, after noting
that the most striking points so far brought out by the
inquiry were those which illustrated the very great variation
in the description and character of the plants developed by
the different manures, proceeds to discuss (1) the general
description and proportion per cent, of the different kinds of
herbage (gramineous, leguminous, and miscellaneous), and
the number of species in relation to the manures employed,
and to the amounts of crop yielded ; (2) the description and
proportion per cent, of the predominating species in relation
to the manures employed and to the amounts of crop yielded,
details being added as to the names and proportion per cent,
of the five predominating grasses, the two predominating
leguminous plants, and the three predominating weeds on
each plot ; and (3) the tendency to the development of leafy
or of stemmy produce, and the order of ripeness. This

136 The Rothamsted Grass Experiments.

report closes with a summary of the most prominent results
of the whole inquiry up to date (1863). The sixth and last
report that appeared in the R.A.S. Journal gives an account
of the produce of hay per acre, the chemical composition of
the hay, and the amount of certain constituents removed
from the land in the fourth, fifth, sixth, and seventh seasons
of the experiments.

Since 1863 the R.A.S. Journal has been silent as to the
results of this unique series of experiments on the herbage of
permanent meadow ; but on June 19, 1879, Dr. Lawes and
Dr. Gilbert read before the Royal Society a paper bearing the
title, " Agricultural, Botanical, and Chemical Results of
Experiments on the Mixed Herbage of Permanent Meadow,
conducted for more than Twenty Years in Succession on the
Same Land. — Part I." The following year this communica-
tion was printed in extenso in the " Philosophical Transactions
of the Royal Society," and occupies no less than 130 of the
large quarto pages of that publication. Since it embraces
the results of more than twenty years' observations, it was
obviously unnecessary, as implied above, to enter at any
length into a discussion of fhe first few years' results. The
investigators arrange and consider the results obtained under
three separate heads, dealing respectively with the agricul-
tural, the botanical, and the chemical results. Part I., now
under notice, treats principally of the agricultural results,
and these we proceed to examine.

Without manure, the produce of hay has varied from about
8cwt. to nearly 39cwt. per acre, and the average yield has
been about 23cwt. per acre per annum. But the plot most
heavily artificially manured, and yielding the best, has given
an average of about 64cwt. of hay per acre per annum, the
extremes being 40cwt. and 80cwt. The results on the
other differently manured plots vary greatly within these
extremes. At the same time, the botanical composition of
the herbage has varied most strikingly, so that, starting
with some fifty species of plants on the unmanured land, any

The Gramineous Heritage. 137

kind of manure induces a struggle which leads to a
diminution of the number of species down to twenty, or even
fewer, though it must not be overlooked that such diminution
of specific forms may be quite compatible with an increase
in the total yield of herbage. The subjoined details
concerning the botanical composition of the herbage are
derived from " Part II. — The Botanical Eesults " (" Philo-
sophical Transactions," 1882), further information respecting
which is given at the end of this volume.

Of Gramineae the following twenty species have been
identified upon the plots : —

GRAMINE.E.

1. Anth.oxanth.um odoratum, L. Sweet scented vernal grass.

2. Alopecurus pratensis, L. ... Meadow foxtail.

3. Agrostis vulgaris, With. ... Creeping-rooted bent grass.

4. Holcus lanatus, L Yorkshire fog.

5. Avena elatior, L., or Arrhena-

therum avenaceum, Beauv. False oat grass.

6. Avena pubescens, L Downy oat grass.

7. Avena flavescens, L Yellow oat grass.

8. Poa pratensis, L Smooth-stalked meadow grass.

9. Poa trivialis, L Bough -stalked meadow grass.

10. Dactylis glomerata, L Bough cocksfoot.

11. Festuca ovina Sheep's fescue.

12. Lolium perenne, L Common rye grass.

13. Briza media, L Common quaking grass.

14. Cynosurus cristatus, L. ... Dogstail.

15. Festuca pratensis, Huds. ... Meadow fescue.

16. Bromus mollis, L Soft brome.

17. Phleum pratense, L Timothy grass.

18. Aira caespitosa, L Tufted hair grass.

19. Festuca elatior, L Tall fescue.

20. Festuca loliacea, Huds. ... Eye-leaved fescue.

Of the foregoing species the first twelve are of common
occurrence upon all the plots. Nos. 13, 14, 15, 16, though
very general, do not invariably occur upon each plot. Nos.
17 and 18 do not appear upon one-half, and in some years not
one-fourth, of the plots. The last two species are extremely rare.

Of Leguminosse, whilst ten species have been identified,

138 The Kothamsted Grass Experiments.

there are only four of common occurrence upon the plots ;
these are : —

Trifolium repens, L .......... White or Dutch clover.

Trifolium pratense, L ....... Purple or meadow clover.

Lotus corniculatus, L ....... Common birdafoot trefoil.

Lathyrus pratensis, L ....... Yellow or meadow vetchling.

Of miscellaneous species as many as fifty-nine have been
recorded. Below are mentioned those which commonly occur
on more than half the plots, and they are arranged in the
order of the frequency of their occurrence, each species
indicated being more frequently recorded than any of those
named after it. The upwards of forty remaining mis-
cellaneous species are of decidedly rare and uncertain

occurrence : —

MISCELLANEOUS SPECIES.

Rumex Acetosa, L .......... Sorrel.

Couopodium denudatum, Koch.,

or Bunium flexuosum, With. Earth nut.

Achillea Millefolium, L ....... Milfoil or yarrow.

Pimpinella Saxifraga, L ....... Burnet saxifrage.

Luzula campestris, Willd. Field woodrush.

Eanunculus acris, L .......... Upright buttercup.

Ranunculus bulbosus, L ....... Bulbous crowfoot, or buttercup.

Ranunculus repens, L ....... Creeping buttercup.

Centaurea nigra, L .......... Black knapweed.

Cerastium triviale, L ....... Mouse-ear chickweed.

Galium verum, L .......... Yellow bedstraw or cheese rennet.

Stellaria graminea, L ....... Lesser stitchwort.

Plantago lanceolata, L ....... Ribwort, ribgrass, or plantain.

Veronica chamaedrys, L ....... Germander speedwell.

Taraxacum officinale, Wigg. . . . Dandelion.

Carex praecox, Jacq .......... Vernal sedge.

Heracleum Sphondylium, L. ... Cow-parsnip.

PLAN OF THE CONTINUOUS EXPERIMENTS ON
MEADOW HERBAGE.

For the sake of comparison, it appears desirable to give a
general description of the manuring of each plot, first
premising that the different " mineral " or ash constituents

Composition of the Manures. 139

were supplied in the substances designated in commerce as
follows: Potash — as sulphate of potash (and in nitrate of
potash). Soda — as sulphate of soda (and in nitrate of soda).
Magnesia — as sulphate of magnesia (Epsom salts). Lime —
as sulphate (gypsum), phosphate, and " superphosphate."
Phosphoric acid — as bone ash, mixed with sulphuric acid in
quantity sufficient to convert most of the insoluble phosphate
of lime into sulphate and soluble " superphosphate " of lime.
Sulphuric acid — in the above phosphatic mixture, in sul-
phates of potash, soda, and magnesia, and in sulphate of
ammonia. Chlorine — in " muriate of ammonia " (chloride
of ammonium, or sal ammoniac). Silica — as silicate of soda
and silicate of lime (also in cut wheat straw). Other con-
stituents have been artifically supplied as follows : Nitrogen
— as sulphate of ammonia, as " muriate of ammonia," and as
nitrate of soda. Non-nitrogenous organic matter yielding,
by decomposition, carbonic acid and other products — in saw-
dust and in cut wheat straw. " Ammonia-salts," wherever
referred to, are equal parts of sulphate and muriate of
ammonia of commerce ; and " superphosphate of lime" was
always per acre, 2001b. bone-ash, 1501b. sulphuric acid (of
specific gravity 17), and water. The artificial manures
were mixed with ashes (burnt soil and turf) in quantity
sufficient to make up a convenient measure for equal dis-
tribution over the land, and the mixtures were sown broad-
cast by hand, experience having proved this to be the best
method on a limited area. The dressings of the plots were as
follows :

Plot 1. — Farmyard manure and ammonia-salts, 8 years,
1856-63 ; ammonia- salts only 1864, and each year since.

Plot 2. — Farmyard manure, 8 years, 1856-63 ; unmanured
1864, and each year since.

Plot 3. — Unmanured, every year, 1856, and since.

Plot 4.— I. Sawdust (without effect), 3 years, 1856-58 ;
superphosphate of lime, 1859, and each year since.

Plot 4.— II. Sawdust (without effect), 3 years, 1856-58;

140 The Rothamsted Grass Experiments.

superphosphate of lime and ammonia- salts, 1859, and each
year since.

Plot 5. — Ammonia-salts alone, every year.

Plot 6. — Ammonia-salts 13 years, 1856-68 (with sawdust
7 years, without effect) ; sulphates of potash, soda, and
magnesia, and superphosphate of lime, 1869, and each year
since.

Plot 7. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime, every year.

Plot 8. — Same as plot 7 for 6 years, 1856-61 ; sulphates
of soda and magnesia (without potash), and superphosphate
of lime, 1862, and each year since ; also sawdust the first
7 years (without effect).

Plot. 9. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime (as plot 7) and ammonia- salts, every
year.

Plot 10. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime (as plots 7 and 8), and ammonia-
salts, 6 years, 1856-61 ; sulphates of soda and magnesia
(without potash) and superphosphate of lime (as plot 8), and
ammonia- salts, 1862, and each year since; also sawdust the
first 7 years (without effect).

Plot Ha. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime (as plots 7 and 9), every year ; twice
as much ammonia-salts as plot 9, 3 years, 1856-58 ; same
ammonia-salts as plot 9, 3 years, 1859-61 ; double quantity
again, 1862, and each year since.

Plot life. — The same as plot Ha, with the addition of a
mixture of silicate of lime and silicate of soda, 9 years,
1862-70, and of silicate of soda only, 1871, and each year since.

Plot 12. — Unmanured, every year, 1856, and since (as
plot 3).

Plot 13. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime, and ammonia- salts (as plot 9), and
cut wheat straw, every year.

Plot 14. — Sulphates of potash, soda, and magnesia, and

Manures Employed. 141

superphosphate of lime (as plots 7, 9, 11, and 13), and nitrate
of soda containing nitrogen equal to that in the ammonia-
salts of plots 5, 9, and 13, 1858, and each year since.

Plot 15. — Nitrate of soda alone, same quantity as plot 14,
18 years, 1858-75 ; no nitrate, but sulphates of potash, soda,
and magnesia, and superphosphate of lime (as plots 7, 9, 11,
13, and 14), 1876, and each year since.

Plot 16. — Sulphates of potash, soda, and magnesia, and
superphosphate of lime (as plots 7, 9, 11, 13 and 14), and
half as much nitrate of soda as plot 14, 1858, and each year
since.

Plot 17. — Nitrate of soda alone, same quantity as plot 16,
1858, and each year since.

Plot 18. — Mixture containing the potash, soda, lime,
magnesia, phosphoric acid, silica, and nitrogen, of 1 ton of
hay, also sulphuric acid and chlorine, 1865, and each year
since.

Plot 19. — Nitrate of soda (same quantity as plots 16 and
17), sulphate of potash, containing the same quantity of
potash as the nitrate of potash of plot 20, and superphos-
phate of lime, 1872, and each year since.

Plot 20. — Nitrate of potash, containing the same quantity
of nitrogen, and the same quantity of potash, as the nitrate
of soda and sulphate of potash of plot 19, and super-
phosphate of lime, 1872, and each year since.

A careful perusal and comparison of the preceding details
will enable the reader to comprehend the scheme underlying
the system of manuring, and to understand how the various
dressings dovetail in with each other, so that the effects of
each manure alone, and of two or more variously combined,
are provided for.

In discussing the results on the various plots, it will be
useful to have some standard of comparison, such as may be
afforded between fair average yields of wheat, barley, and
meadow hay. For this purpose the following yields per acre
are assumed :

142

The Rothamsted Grass Experiments.

Wheat, 30 bushels=18001b., and 30001b. straw=48001b.
total produce.

Barley, 40 bushels=20801b., and 25001b. straw=45801b.
total produce.

Meadow hay, 1J ton, or 33601b.

Table XXIX. shows the average amounts of nitrogen, and
of most of the mineral constituents, in the above quantities
of wheat grain and straw, and barley grain and straw.
Meadow hay varies so greatly in its botanical and chemical
composition, according to soil, climate, and manuring, that it
is necessary to take mean results.

TABLE XXIX. — COMPOSITION OF AVERAGK CROPS or WHEAT, BARLEY,
AND MEADOW HAT.

In grain.

In straw.

In total
produce.

if'jj

14 ton = 3360lb.
meadow hay.

II

if

a

|j

|||

|jf

J?

jff

Nitrogen

Ib.
32-0

1-0
3-5
9-5
0-2
15-0
0-2

o-o

0-5

Ib.
33-0

1-5
4-0
11-0
0-5
16-5
0-75
0-2
11-0

Ib.
13-0

7'5
3-0
22-0
1-0
6-0
4-0
3-0
95-0

]b.
12-0

9-0
2-5
23-0
3-0
4-5
4-0
4-0
55-0

Ib.
45-0

8-5
6-5
31-5
1-2
21-0
4-2
3-0
95-5

Ib.
45-0

10-5
6-5
34-0
3-5
21-0
4-75
4-2
66-0

Ib.

40-8

27-1
7-4
72-8
7-9
16-4
11-3
21-3
59-9

Ib.
47-2

35-2
9-4
40-5
16-0
11-0
13-6
12-3
51-8

Ib.

44-0

31-2
8-4
56-6
11-9
13-7
12-5
16-8
55-9

Ib.
52-0

33-8
12-8
51-7
9-0
16-2
9-2
14-6
54-6

Lime
Magnesia

Potash

Soda
Phosphoric acid
Sulphuric acid
Chlorine
Silica

These figures, read from left to right, will be found most
instructive. Assuming as the basis of comparison the mean
composition of the manured and the unmanured hay, it is
seen that a fairly good crop of hay will remove about one-
third more nitrogen than the grain of a fairly good crop of
wheat or barley, and practically the same amount as the total
produce, grain and straw together, of either of the corn crops
Of phosphoric acid, the hay crop will remove somewhat less
than the grain alone, and only about two-thirds as much as

Mineral Ingredients of Wheat, Barley, and Hay. 143

the total produce of wheat or barley. Of potash, the assumed
average hay crop will remove five or six times as much as the
grain of either the wheat or the barley, and nearly twice as much
as the total produce, grain and straw together. Of lime, soda,
sulphuric acid, chlorine, and silica, the hay will remove many
times more, and of magnesia much more, than either
the wheat or the barley grain. Of lime, soda, sulphuric
acid, and chlorine, the hay will also remove much more,
and of magnesia more, than both grain and straw
together. Of phosphoric acid and silica alone will the total
produce of the corn crops remove more than the hay crops.
Summing up the salient points, it is obvious that in
Rothamsted Park, where the soil is a loam with a clay subsoil,
the effect of the application of a complex fertiliser like
farmyard manure, supplying as it doubtless does much more
of all the mineral constituents than the crop takes up, is in a
striking degree to increase the assimilation of potash —
notably also that of phosphoric acid, and to some degree^ that
of silica ; much more chlorine is also taken up. Indeed,
it will subsequently appear that the supply by manure of
potash has a more marked effect on the quantity, and on the
botanical and chemical character, of the herbage of the hay
crop, than that of any other of the mineral or ash con-
stituents.

RESULTS OBTAINED UPON THE VARIOUSLY
MANURED PLOTS OF MEADOW LAND.

By dressing the several experimental plots of permanent
grass land with different manures, and with different
mixtures of manures, variations are induced in the character
of the resultant herbage, the visible effect of which in the
middle of summer may fairly be described as kaleidoscopic.
To the practical man the nature and extent of these
variations, as revealed by the gross test of the weighing

144

The Rothamsted Grass Experiments.

machine, and by the subtler methods of the chemist and the
botanist, are all important. These we proceed to discuss.

1. Without Manure : Plots 3 and 12.— Table XXX., record-
ing the results on this plot, speaks for itself.

TABLE XXX. — AVERAGE PER ACRE PER ANNUM, WITHOUT MANURE.

Plot 3.

Plot 12.

Mean.

HAT.

1st period, 10 years, 1856-65

Ib.
2531

Ib.

2808

Ib.

2670

2nd period, 10 years, 1866-75
Total period, 20 years, 1856-75
2nd period, per cent, greater ( + ) or
less than ( — ) 1st period

2236
2383

-11-7

2564
2686

- 8-7

2400
2535

-10-1

NITROGEN.
1st period, 10 years, 1856-65

35-1

39-3

37-2

2nd period, 10 years, 1866-75
Total period, 20 years, 1856-75
2nd period, per cent. + or — 1st
period

30-9
33-0

-12-0

35-6
37-5

- 9-4

33-3
35-3

-10-5

MINERAL MATTER (Asn).
1st period, 10 years, 1856-65
2nd period, 10 years, 1866-75

148-5
126-1

161-7
143-2

155-1
134-6

Total period, 20 years, 1856-75
2nd period, per cent. + or — 1st
period

137-3
-15-1

152-4
-11-4

144-9
-13-2

Average per Acre per Annum.

The uniform superiority of plot 12 over plot 3 will not
escape notice. This was, at first, attributed to the fact that
plot 12 is earlier sheltered from the evening sun, and there-
fore the less liable to suffer in dry weather. But the
investigators were not content with this apparent reason, and
therefore sampled the soil of the plots, in each case to a
total depth of 54in., and from observations on the sections
dug up, as well as from estimations of the nitrogen in every
9in. of depth, were led to conclude that a considerable part,
at any rate, of plot 12 must have been made ground.
Accordingly, it was decided that the results obtained on the
unmanured plot 3 would provide fairer standards with which
to compare those yielded on the manured plots than would

Continuous Growth of Herbage without Manure. 145

the mean indications of the two unmanured plots. Hence,
unless otherwise stated, all comparisons are made with the
produce of the unmanured plot 3.

The reduction of produce on the continuous growth without
manure is, apart from influences of the seasons, attributable
in part to deficiency in available nitrogen, but probably in
greater degree to that of an available supply of potash, of
phosphoric acid, and of silica. In the absence of manure
favouring any particular species or families of plants, no
artificial struggle for existence was induced among the
herbage, and the result was that a greater variety of species
maintained their position than on any of the manured plots.
The proportion of grasses was comparatively low; that of
leguminous plants was fairly high ; but the most marked
characteristic was the large number, and high proportion by
weight, of the .miscellaneous species (weeds). Under this
varied crop there is a much greater annual assimilation of
nitrogen, and of some of the most important mineral
constituents over a given area, than in an unmanured
gramineous crop, such as wheat or barley, grown separately,
on arable land.

2. Ammonia-Salts Alone : Plot 5. — The results shown in
Table XXXI. were obtained by the annual application of
ammonia- salts alone, at the rate of 4001b. per acre per
annum, estimated to supply about lOOlb. of ammonia, corres-
ponding with about 821b. of nitrogen ; for comparison, the
results on plot 3, without manure, are also given.

Compared with the produce without manure, ammonia-
salts have given an average annual increase over twenty years
of 5631b. of hay, or, in all, ll,2601b., or rather more than five
tons, the extra demand for mineral constituents being met
solely by the soil. But the falling off in increase as time
went on is very marked. Actually less lime and potash were
taken up with the ammonia than without manure, which
appears due to the fact that, under the influence of the
purely nitrogenous manure, various species which com-

L

146

The Rothamsted Grass Experiments.

manded a more extended range of soil than those which
remained had disappeared. The total number of species
became greatly reduced, grasses largely predominating.
Thus, Festuca ovina (sheep's fescue) sometimes yielded more
than half, and Agrostis vulgaris (fine bent) more than a
quarter of the total produce ; there was scarcely any
leguminous herbage ; the miscellaneous species were few, and
added but little to the total weight, except Rumex Acetosa
(sorrel), which was objectionably nourishing. G-enerally,
the stunted foliage was of a very dark green colour, indicating
a very high percentage of nitrogen in its dry substance, or
rather a deficient assimilation of carbon in proportion to the
nitrogen taken up.

TABLE XXXI. — AVERAGE PER ACRE PER ANNUM, BY 400LB.
AMMONIA-SALTS ALONE ; PLOT 5.

Average per Acre per Annum.

Plot 3.
Without
manure.

Plot 5.
Ammonia-
salts alone.

Plot 5.
+ or -
Plot 3.

HAY.

1st period, 10 years, 1856-65
2nd period, 10 years, 1866-75
Total period, 20 years, 1856-75
2nd period, per cent. + or — 1st
period

Ib.
2531
2236
2383

-11-7

Ib.
3420
2471
2946

-27-7

Ib.
+ 889
+ 235
+ 563

NITROGEN.
1st period

35-1

57-9

+ 22-8

2nd period

30-9

47-3

+ 16-4

Total period

33-0

52-6

+ 19-6

2nd period, per cent. + or — 1st
period

— 12-0

-18-3

MINERAL MATTER (Asn).
1st period

148-5

181-2

+ 32-7

2nd period

126-1

108-9

-17-2

Total period

137-3

145-1

+ 7-8

2nd period, + or — 1st period

-15-1

-39-9

3. Nitrate of soda alone : Plots 15 and 17. — Here also are
brought out the effects of a continuous supply of nitrogenous

Ammonia-salts versus Nitrate of Soda. 147

without mineral manures, the nitrogen being, however, pre-
sented as nitrate instead of in the form of ammonia. Plot 15
received, for eighteen years, approximately the same amount
of nitrogen per acre per annum as plot 5, but as nitrate of
soda instead of as ammonia-salts ; plot 17 has received
annually half the quantity of nitrate. A slight difference in
the periods will be noticed in the Table XXXII., page 148, in
which the results are recorded.

Although plot 15 and plot 5 received the same amount of
nitrogen per annum, namely, 821b., yet, in the case in which
this nitrogen was applied in the form of ammonia- salts
(plot 5) there was an average annual increase of produce
over the eighteen years of only 4891b., whilst with the same
amount of nitrogen applied as nitrate of soda (plot 15) the
corresponding increase was 16181b., or 11291b. more. In
other words, whilst the ammonia-salts gave an average
annual increase of little more than one-fifth over the
unmanured produce, the nitrate gave more than one and two-
thirds as much. The decline of produce in the second, as
compared with the first period, was 10*3 per cent, without
manure, 25*1 per cent, with ammonia- salts, and only 3'1 per
cent, with the nitrate. The nitrate yielded one -fifth more
nitrogen than the ammonia salts ; whilst the decline of
nitrogen in the second period was nearly 9 per cent, without
manure, more than 17 per cent, with ammonia salts, and only
2J per cent, with the nitrate,

Striking as are these different effects of ammonia-salts and
nitrate, they are even inore marked in the amounts of
mineral matter taken up. The quantities of mineral matter
removed without manure and with ammonia-salts are almost
identical, whereas the nitrate of soda crops took up half as
much again. More lime, more magnesia, and considerably
more potash, phosphoric acid, and silica were removed by the
nitrate produce than by that of the ammonia-salts.

The explanation of these results must be sought in the
herbage. Under the nitrate a greater total number of species

L 2

148

The Rothamsted Grass Experiments.

g 5

5 I

I «

la <:

s!i
s

01

a

•

I S

s*

(W

+5

i «

0 CO 0

i> oo do

W CO O
(N CXI (M

+ + +

Pi;

»OO<1<M rH

*

r-i Tjt cq
I I I

r-lO

000000
!>!>!>

+ + +

m

s"gM

«s _-'3 a,- i O 1-1

^ £ 2 « , • o i>

Og S «8 ! ~ CO "f

5;§ gaj ' co <M

cxi co r» co
^- t- ^-i r»

m ><ji o f-i

I

^ » S

;|
i

CiOS(M OS

CXI O! CO CO

00 rH OS >rt

o <i> •* co

M

1-ieteH <N

Large and Small Dressings of Nitrate of Soda. 149

maintained a place than under the ammonia-salts. A greater
variety, a better development, and consequently a wider and
deeper distribution of the roots of the mixed herbage grown
by the nitrate, gave it the command of a correspondingly
increased range of soil and subsoil. Further, the nitrogen
of nitrate of soda distributes more rapidly in the soil, both in
the upper and the lower layers, and so root penetration into
the latter is induced.

Lastly, the results show that the smaller amounts of nitrate
acted more effectively than the larger dressing. Indeed, it
would appear that the limit of growth under the influence of
nitrate of soda alone — that is, without the aid of any artificial
supply of mineral constituents — was nearly reached with the
smaller quantity used. After eighteen years the heavier dose
of nitrate was discontinued, and a mixed mineral manure
(sulphates of potash, soda, and magnesia, and superphosphate
of lime) was applied, but does not seem to have been
instrumental in effecting the assimilation of any excess of
nitrogen unused by previous crops ; so that much of the
nitrogen of the heavier dressing was probably lost in the
drainage waters. But in the space of four years the herbage
had much changed; the grasses became lighter in colour,
and showed more tendency to form stem and to mature;
leguminous species (especially Lathyrus pratensis) were
gaining ground. The character of the miscellaneous herbage
was also changing.

4. Mixed Mineral Manure Alone : Plot 7. — The dressing
per acre per annum has been 3001b. sulphate of potash, lOOlb.
sulphate of soda (2001b. the first eight years), lOOlb.
sulphate of magnesia, and superphosphate of lime (2001b.
bone ash, 1501b, sulphuric acid, sp. gr. 1*7) ; the results are
shown in Table XXXIII., page 150.

Thus, mixed mineral manure alone gives considerably more
produce than does nitrogen in the form of ammonia- salts,
and nearly as much as nitrogen applied as nitrate of soda.
Moreover, there is an increase in hay, in nitrogen, and in

150

The Rothamsted Grass Experiments.

mineral matter during the second period as compared with
the first. Yet it is an established fact that nitrogenous
manures are specially effective in increasing the growth of
gramineous crops grown separately on arable land, such as
wheat, barley, or oats, though these only contain a com-
paratively small percentage of nitrogen, and assimilate a
comparatively small amount of it over a given area when

TABLE XXXIII. — AVERAGE PER ACRE PER ANNUM, BY MIXED
MINERAL MANURE ALONE ; PLOT 7.

Plot 3.
Without
Manure.

Plot 7.
Mixed
Mineral
Manure.

Plot 7.
+ or -
Plot 3.

HAT.

1st period, 10 years, 1856-65 . .

Ib.
2531

Ib.
3797

Ib.
+ 1266

2nd period, 10 years, 1866-75 ... .

2236

4118

+ 1882

Total period, 20 years, 1856-75
2nd period, per cent. + or — 1st
period

2383
-11-7

3958

+ 8-5

+ 1575

NITROGEN.
1st period . . .

35-1

55-2

+ 20'1

2nd period

30-9

58-0

+ 27-1

Total period

33'0

56-6

+ 23'6

2nd period, per cent. + or — 1st
period .

- 12-0

+ 5-1

MINERAL MATTER (Asn).
1st period

148-5

246-3

+ 97-8

2nd period

126-1

261-6

+ 135-5

Total period

137-3

254-0

+ 116*7

2nd period, per cent. + or — 1st
period

-15-1

+ 6-3

Average per Acre per Annum.

none is supplied in manure. The highly nitrogenous
leguminous crops, on the other hand, such as beans, peas,
clover, &c., are not characteristically benefited by the use of
direct nitrogenous manures, though nitrates do act more
favourably on them than ammonia-salts. Again, whilst,
under equal conditions of soils and seasons, mineral manures
alone increase comparatively little the gramineous crops grown

Effect of Mineral Manures.

151

separately, such manures, and especially potash manures, do
in a striking degree increase the yield of leguminous crops so
grown, and coincidently increase the amount of nitrogen they
assimilate per acre. Consistently with this, the mixed
mineral manure, containing potash, did very largely increase
the leguminous growth on plot 7 ; some of the gramineous
species, moreover, were increased to an extent hardly
anticipated.

A table not reproduced here shows that over the twenty
years more than half as much again gramineous herbage was
grown by the purely mineral manure than by the unmanured
plot 3. Though the leguminous herbage showed a less
actual increase of weight than the gramineous, the pro-
portional increase was much greater, there being four times
as much grown with mineral manure as without manure.
Of miscellaneous herbage there was very little increase, and a
falling off in actual amount over the later years.

TABLE XXXIV. — NITROGEN AND ASH OF MEADOW HERBAGE.

Per cent. Nitrogen.

Per cent.
Mineral Matter
(Ash).

Gramineous.
Plot 3 (without manure)
Plot 7 (mixed mineral manure)
Leguminous.
Plots

1-25
1-18

2-31

5-11
5-64

6-32

Plot 7

2-31

6-81

Miscellaneous.
Plot 3

1-45

7-30

Plot 7

1-32

8-74

Table XXXIV. shows the mean percentages of nitrogen
and of mineral matter in the different kinds of herbage,
dried as hay.

Basing the statement on the figures of Table XXXIV., it
appears that, whilst the average annual amount of nitrogen
in the gramineous herbage was 201b. without, and 29'31b.
with, the manure, or 9'31b. more with than without it, it was

152 The Rothamsted Grass Experiments.

in the leguminous herbage only 4'21b. without, but 18'21b
with, the manure, or 14'01b. more with it than without it.

The much- debated question as to the source of this
increased nitrogen of the leguminous herbage is here
answered in favour of the soil, and, in the case of clover and
beans at any rate, more probably the lower than the upper
layers. In the unmanured herbage the deeper-rooting Lotus
corniculatus (lesser bird's-foot trefoil) and Trifolium pratense
(meadow clover) contributed most of the leguminous produce ;
but in the manured herbage it is the comparatively surface-
rooting Lathyrus pratensis (meadow vetchling) that mainly
contributes the leguminous increase.

The agricultural importance of the mode of root-distribu-
tion is further illustrated by gramineous plants. Spring-
sown barley, which has but a short time in which to extend
its roots, and to gain command of the resources of the soil,
throws out a large quantity of fibres near the surface, and is
more benefited by the action of direct mineral manures than
is the autumn-sown wheat, which has four or five months
longer for root-distribution, and is less dependent on the
stores of the surface soil. Accordingly, the greatly increased
gramineous yield in the mixed herbage by the purely mineral
manure was almost exclusively due to the much denser
growth, and much greater tendency to form stem and seed, of
the creeping and surface-rooting Festuca ovina (sheep's
fescue), Agrostis vulgaris (fine bent), and Holcus lanatus
(Yorkshire fog).

How far is the increased growth of the grasses proper in
the mixed herbage arising from an increased accumulation of
combined nitrogen available to them in the upper layers of
the soil — the result of the increased growth of, and accumula-
tion by, the Leguminosse — induced by the mineral manure?
In other words, is there, where the plants are thus growing
in association, a parallel action to that which takes place
when they are grown in alternation (as wheat after clover,
the clover crop having left the upper layers of the soil rich in

Nitrogen of the Soil. 153

nitrogen) ? Analyses of the soils reveal a considerably lower
percentage of nitrogen in the first 9in. of depth of the mineral
manured plot 7 than in that of the unmanured plot 3, the
difference being more than sufficient to account for the
increased yield of nitrogen over the twenty years in both the
gramineous and the leguminous herbage of plot 7. It would,
in fact, appear that, under the influence of the mixed mineral
manure alone, both the powers of collection, assimilation, and
transformation of the plants themselves, have been consider-
ably augmented, and the accumulated stores of nitrogen in
the soil have been rendered more available.

With regard to mineral matter, plot 7, liberally supplied
with mineral manures, yielded in its herbage over the first
period nearly one and two-thirds as much, over the second
period more than twice as much, and over the entire period
nearly twice as much, as the unmanured plot 3. Potash and
phosphoric acid rank first ; of the former nearly three and a
half times as much, and of the latter nearly three times as
much, being taken up on the mineral-manured plot as on the
unmanured plot.

The general result is that a mixed mineral manure alone
has markedly influenced the description of plants developed,
and has greatly favoured the tendency to form stem and seed
— that is, to mature ; it has also much increased the amount
of total produce grown, and of nitrogen as well as mineral
matter taken up. The evidence, so far, points to the potash
of the manure as mainly conducive to these effects, the
phosphoric acid probably coming second in order of import-
ance.

5. Superphosphate of Lime Alone : Plot 4 — I. — This plot
has received phosphate alone every year from 1859. In each
of the three preceding years (1856-7-8) it received sawdust
(20001b. per acre per annum), with the object of testing
Leibig's assertion that it produced great effects by virtue of
the solvent action exerted by the carbonic acid, yielded in
decomposition, upon the mineral constituents of the soil.

154

The Rothamsted Grass Experiments.

Nevertheless, notwithstanding that each year's sawdust con-
tained nitrogen equal to between 41b. and 51b per acre, the
result was even less produce than without manure in each of
the three seasons. Table XXXV. shows the average results
on this plot.

TABLE XXXV. — AVERAGE PER ACRE PER ANNUM, BY SUPERPHOSPHATE
OF LIME ALONE ; PLOT 4 — I.

Average per Acre per Annum.

Plot 3.
Without
Manure.

-'•sal -*•' gjl
°^S§ jMf|

fcSgg A, «!

^ i «

°~°°

PW +£

•* |i-
3~g5
5 +E

HAT.

1st period, 7 years, 1859-65
2nd period, 10 years, 1866-75 ...
Total period, 17 years, 1859-75
2nd period, per cent. + or — 1st
poriod

Ib.
2495
2236
2343

-10-4

33-8
30-9
32-1

- 8-6

146-5
126-1
134-5

-13-9

Ib.
3828
4118
3999

+ 7-6

54-5
58-0
56-5

+ 6-4

247-8
261-6
255-9

+ 5-6

Ib.
2732
2384
2527

-12-7

36-3
31-6
33-5

-12-9

170-8
140-4
152-9

-17-8

Ib.

+ 237
+ 148
+ 184

+ 2-5

+ 0-7
+ 1-4

+ 24-3
+ 14-3

+ 18-4

Ib.
-1096
- 1734

- 1472

- 18-2
- 26-4
- 23-0

- 77-0
-121-2
-103-0

NITROGEN.
1st period

2nd period

Total period

2nd period, per cent. + or— 1st
period

MINERAL MATTER (Asn).
1st period ...

2nd period

Total period

2nd period, per cent. + or — 1st
period

Thus, the superphosphate plot gave more produce than the
unmanured plot, but the increase was only small, whilst the
actual produce per acre declined more than on the unmanured
plot. On the other hand, superphosphate alone gave, over
the total period, less than two-thirds as much hay as the
mixed mineral manure.

The percentage reduction in annual yield of hay, of
nitrogen, and of mineral matter, over the second period com-

Effect of Superphosphate of Lime alone. 155

pared with the first, was greater with the superphosphate
than without manure ; whereas, with the mixed mineral
manure there was not a reduction but an increase of all three
during the second period.

The number and relative predominance of the plant species
was much the same as without manure. The most noticeable
changes were a rather greater weight of total gramineous
herbage, due mainly to more of ripening tendency ; a reduc-
tion in weight of Leguminosae, with a prevalence of Lathyrus
pratensis rather than that of Lotus corniculatus ; and an
increase in weight of miscellaneous herbage, with a greater
prevalence especially of Ranunculus repens and E. bulbosus
(buttercups), Achillea Millefolium (yarrow), and Rumex
Acetosa (sorrel).

Of the constituents supplied in the superphosphate, a little
more of lime and magnesia, once and a half as much
sulphuric acid, and nearly twice as much phosphoric acid,
were taken up under its influence as without manure. Of
potash and silica very little more were taken up from the
soil than without manure, but the increase of soda was some-
what greater.

The superphosphate enabled the plants to draw somewhat
more largely on the resources of the soil, especially at first,
but the small increased available supply under its influence
rapidly diminished. A comparison of the produce by the
superphosphate alone, with that by the mixed mineral
manure, clearly shows that it was neither exclusively, nor
even mainly, to the phosphoric acid which the latter con-
tained that the results it yielded were due.

6. Mixed mineral manure, with and without potash : Plots 7
and 8. — During the first six years (1856-61), plots 7 and 8 each
received the " mixed mineral manure," comprising super-
phosphate of lime, and sulphates of potash, soda, and
magnesia, the only difference being that during those years,
and in 1862, plot 8 also received 20001b. of sawdust per acre
per annum. From 1862 inclusive, plot 7 has continued to

156

The Rothamsted Grass Experiments.

receive the same mixed mineral manure, and plot 8 the same,
excluding the sulphate of potash, but with an increased
amount of sulphate of soda. The results are given in Table
XXXVI.

TABLE XXXVI. — AVERAGE PER ACRE PER ANNUM, BY MIXED MINERAL
MANURE, WITH AND WITHOUT POTASH ; PLOTS 7 AND 8.

Average per Acre per Annum.

Plot 7.
Mineral Manure,
including potash,
every year.

Plot 8.
Mineral Manure,
including potash
6 years, without
potush, 14 years.

t^

0*2

Ij

HAY.
1st period, 6 yeara 1856-61

Ib.
3835

Ib.
4027

Ib.

+ 192

2nd period, 14 years, 1862-75
Total period, 20 years, 1856-75 ...
2nd period, per cent. -I- or — 1 st
period

4011
3958

-I- 4-6

3098
3377

-23-1

- 913
- 581

NITROGEN.
1st period . .

56-6

60-9

+ 4-3

2nd period

56-6

40-3

-16-3

Total period

56-6

46-5

-10-1

2nd period, per cent. + or — 1st
period

o-o

-33-8

MINERAL MATTER (ASH).
1st period .

254-4

271-8

+ 17-4

2nd period

253-7

181-9

-71-8

Total period

253-9

208-8

-45-1

2nd period, per cent. + or — 1st
period

— 0.3

-33-1

There is an average of rather more hay, and of rather more
nitrogen and mineral matter taken up, on plot 8 than on
plot 7, during the first six years. Whether this arose from
the action of the sawdust is uncertain, but the increased
amount of nitrogen taken up corresponded very closely with
that contained in the sawdust. It has already been observed
that sawdust alone gave no increase.

During the next fourteen years the relations are entirely

Mineral Manure with and without Potash. 157

reversed. There is an average produce of nearly one-fourth
less hay, and of over one-fourth less nitrogen and mineral
matter taken up, on plot 8 without than on plot 7 with potash,
so that the falling off in the produce on the exclusion of potash
was very great. The number of species remained about the
same on the two plots, but the percentage of Leguminosse has
greatly diminished since the exclusion of potash. Without
the potash the grasses became more leafy, less dense, and
showed much less tendency to mature. The great difference
in the amount of the total leguminous herbage on the two
plots is due to the marked increase of the surface-rooting
Lathyrus pratensis with the potash, and its greatly reduced
growth without it. Of the deep-rooting and hardy Lotus
comic ulat us there was the more without the potash.

Excepting only silica and soda, less of every mineral con-
stituent— potash, lime, magnesia, phosphoric acid, sulphuric
a^id — was taken up on plot 8 without potash than on plot 7
with the potash. In the dry substance the percentage of
potash was less than two-thirds as high on plot 8 as on
plot 7, though it was still much higher than on the
unmanured plot, as also was the percentage of phosphoric
acid.

The falling off in the total weight of produce, and in the
description and character of development of the herbage on
plot 8, must be attributed to the deficiency of available
potash.

An interesting line of inquiry is suggested as to whether on
plot 8 the potash taken up by the herbage was derived from
the residue of the potash applied during the first six years, or
was furnished directly by the soil. That certain soils do, as
a matter of fact, retain a residue of supplied potash, and in a
slowly available condition, for very many years, is con-
clusively proved by evidence of very different kinds. Thus,
in the Rothamsted experiments, in which wheat has been
grown continuously for many years on the same land, the
effects of the residue of potash supplied more than twenty-

158 The Rothamsted Grass Experiments.

eight years previously were still traceable in an increased
produce, and in an increased amount of potash in the crop
grown. Hermann von Liebig found the amounts of both
potash and phosphoric acid considerably greater, particularly
in the upper layers of the soil of this wheat field, the greater
had been the previous supplies by manure. And Dr. Aug.
Voelcker's analysis of drainage waters from different plots of
the same field showed very much less loss of potash in that
way than of either soda, lime, or magnesia, and also very
much less of phosphoric acid than of sulphuric acid or of
chlorine. There was, in fact, comparatively little loss by
drainage of either potash or phosphoric acid.

To sum up the last three cases respecting the effects of the
different mineral manures when used without artificial
nitrogenous supply, the comparison of the results obtained
without manure, with superphosphate alone, and with a
mixed mineral manure with, and without, potash, brings
clearly to view the much more striking influence of an
increased supply of potash, than of lime, magnesia, soda,
phosphoric acid, or sulphuric acid. It will be observed that
this is quite consistent with the facts already set forth in
Table XXIX., in regard to the mineral composition of the hay
crop. It is true that when there was either a direct supply,
or an available residue, of potash, more of the other con-
stituents just enumerated was taken up, and even when
superphosphate alone was used, considerably more phosphoric
acid was taken up, but coincidently there was but little
increase of produce, and very little increase in the amount of
nitrogen assimilated.

Effect of Mixtures of Nitrogenous and Mineral Manures. —
The results hitherto considered have been obtained either by
nitrogenous manures alone or by mineral manures alone. It
is now necessary to study the results following the applica-
tion of mixtures of the two. The order of discussion will be
— first, the effects of a given amount of nitrogen applied as
ammonia- salts, then of double the quantity, then of the

Mixtures of Nitrogenous and Mineral Manures. 159

original amount applied as nitrate of soda, and then of half the
quantity, also as nitrate, in each case used in conjunction
with the same mixed mineral manure, including potash, as
that applied to plot 7 ; and, following the same order as with
the mineral manures used alone, comparison will then be
made of the results obtained by the ammonia-salts alone, by
the same with superphosphate of lime, with the mixed
mineral manure including potash, and with the mixed mineral
manure excluding potash.

TABLE XXXVII. — AVERAGE PER ACRE PER ANNUM, BY 400LB. AMMONIA-
SALTS, WITH MIXED MINERAL MANURE CONTAINING POTASH ; PLOT 9.

Average per Acre per Annum.

Plot 5.
Ammonia-salts
alone.

Plot?.
Mineral Manure
alone.

Plot 9.
Ammonia-salts
and
Mineral Manure.

*
oi .2

|J

si
*i

HAT.

1st period, 10 years, 1856-65 ...
2nd period, 10 years, 1866-75 ...
Total period, 20 years, 1856-75
2nd period, per cent. + or — 1st
period .

Ib.
3420
2471
2946

-27-7

57-9
47-3
52-6

-18-3

181-2
108-9
145-1

-39.9

Ib.
3797
4118
3958

+ 8-5

55-2

58-0
56-6

+ 5-1

246-2
261-6
254-0

+ 6-3

Ib.
6002
5421
5711

- 9-7

75-7
70-7
73-2

- 6-6

368-3
301-0
334-6

-18-3

Ib.
+ 2582
+ 2950
+ 2765

+ 17-8
+ 23-4
+ 20-6

+ 187-1
+ 192-1
+ 189-5

Ib.
+ 2205
+ 1303
+ 1753

+ 20-5
+ 12-7
+ 16-6

+ 122-1
+ 39-4
+ 80-6

NITROGEN.
1st period

2nd period
Total period

2nd period, per cent. + or — 1st
period

MINERAL MATTER (ASH).
1st period
2nd period

Total period ...

2nd period, per cent. + or — 1st
period

7. 400Z&. Ammonia-Salts with Mixed Mineral Manure con-
taining Potash : Plot 9. — The average results on this plot are
recorded in the foregoing Table XXXVII.

160 The Rothamsted Grass Experiments.

The result is, over the twenty years, considerably more
than twice as much hay as without manure, nearly twice as
much as by the same amount of ammonia-salts used alone,
and nearly once and a half as much as by the same mineral
manure used alone. During the second period, however,
there was a falling off, as compared with the first, of nearly
10 per cent.

The average annual yield of nitrogen per acre was more
than twice as much as without manure, more than once and
a third as much as by the same amount of ammonia-salts
alone, and nearly once and one-third as much as by the same
mineral manures used alone. There was a decrease in the
yield over the second ten years as compared with the first.

Of total mineral constituents about two and one-third
times as much were taken up as either without manure,
or with the ammonia-salts alone, and about once and a third
as much as when the same mineral manure was used without
ammonia-salts. There was a falling off of 18 per cent, in
the second period as compared with the first.

[n discussing the character of the herbage, it is necessary
to bear in mind the corresponding results obtained (a) without
manure, (&) with mixed mineral manure, (c) with ammonia-
salts alone. On the plot now under notice, the mixture of
nitrogenous and mineral manures gave rise to herbage still
more gramineous than that produced by the ammonia-salts
alone; in fact, the weight per acre of gramineous herbage
averaged more than twice as much ; the proportion of the
produce due to both leguminous and miscellaneous species
was also less, though the average weight per acre of both
was rather more. But the botanical composition of the
gramineous herbage was more varied. Thus, whilst Festuca
ovina and Agrostis vulgaris contributed by far the larger
proportion of the total herbage grown by ammonia-salts
alone, and Anthoxanthum odoraturn (sweet vernal), Holcus
lanatus (Yorkshire fog), and Dactylis glomerata (cocksfoot)
were only very moderately, and every other grass sparingly

Mixtures of Nitrogenous and Mineral Manures. 161

represented, yet, with, the ammonia- salts and mineral manure
together, both Festuca ovina and Agrostis were much less
prominent, Holcus and Dactylis were nearly as much so,
and Poa pratensis (smooth- stalked meadow-grass) was the
most prominent of all. The character of development of the
grasses was, moreover, totally different, there being a great
tendency to form stem and seed and to mature. Under
these conditions a much larger proportion of the supplied
nitrogen was taken up than when the ammonia-salts were
used alone. As it is probable that, even when used in con-
junction with the mineral manure, not much more than half
the supplied nitrogen was taken up, the reduction in produce
over the second ten years would not seem to be connected
with a deficiency of available nitrogen. The reduction in
the amount of mineral matter taken up was greater than in
that of the nitrogen, and yet calculation shows that more,
and generally much more, of each constituent, excepting,
of course, silica, was applied each year than was removed in
the crop, so that there would be a constantly increasing
residue accumulated during the later years in all cases in
which there was no material loss by drainage. As will
presently appear, however, when the same mineral manures
were used with an increased amount of ammonia-salts,
a larger amount of each of the mineral constituents,
excepting lime, was taken up.

The reduced assimilation of both nitrogen and most of the
mineral constituents over the second period was probably
less connected with any real deficiency of either within the
soil, in a condition capable of being yielded up, than with
the description of plants favoured, and the character both of
their underground and their above-ground development.

8. 400$. Ammonia-salts, with Mixed Mineral Manure
containing Potash, and 2000Z6. cut Wheat Straw : Plot 13. —
The only difference between this and the preceding case is
the addition of 20001b. of cut wheat straw per acre per
annum, the object being to supply silica and carbonaceous.

M

162

The Rothamsted Grass Experiments.

organic matter somewhat in the same condition as they are
provided in farmyard manure. Table XXXVIII. shows the
results.

TABLE XXXVIII. — AVERAGE PER ACRE PER ANNUM, BY 400LB. AMMONIA-
SALTS, WITH MIXED MINERAL MANURE CONTAINING POTASH, AND
2000LB. CUT WHEAT STRAW; PLOT 13.

Average per Acre per Annum.

Plot 9.
Mineral Manure
and 4001b.
Ammonia-salts.

Plot 13.
As Plot 9, and
20001b. cut Wheat
Straw.

=i

ES

+

HAT.

1st period, 10 years, 1856-65

Ib.
6002

Ib.
6186

Ib.

+ 184

2nd period, 10 years, 1866-75
Total period, 20 years, 1856-75 ...
2nd period, per cent. -(- or — 1st
period ...

5421
5711

- 9-7

6679
6432

+ 8'0

+ 1258
+ 721

NITROGEN.
1st period

757

79-5

+ 3-8

70-7

87-3

+ 16-6

Total period

73-2

83-4

+ 10"2

2nd period, per cent, -f or — 1st
period

- 6-6

+ 9-8

MINERAL MATTER (Asn).
1 st period

368-3

389-2

+ 20-9

2nd period

301-0

382-8

+ 81-8

Total period

334-6

386-0

+ 51-4

2nd period, per cent. + or — 1st
period

— 18-3

— 1-6

There is here considerably more average produce of hay,
and more of nitrogen and mineral matter removed, over the
twenty years, than by the same manures without the straw.
And, instead of a reduction, there is a considerable increase
in the yield of hay and of nitrogen, and a nearly equal yield
of mineral matter, over the second ten years compared with
the first.

It is not very easy to account for the greater yield where
the cut straw was used, i.e., of plot 13 over plot 9. Analyses

Manurial Effect of Cut Wheat Straw. 163

of the respective soils reveal no essential differences, and,
independently of the manure, the only known difference
between plot 13 and plot 9 is that the former receives earlier
afternoon shade.

On plot 13 the herbage has become even more markedly
gramineous than on plot 9. Leguminosse have almost equally
disappeared on both ; and the miscellaneous species have
greatly decreased on both, Eumex Acetosa being the only
prominent weed on either, Conopodium denudatum (Bunium
flexuosum, earth-nut or pig-nut, an umbellifer) coming next.
The increase on the straw plot was almost entirely com-
posed of grass, which, compared with plot 9, contained very
much more Dactylis glomerata, this, on plot 13, becoming
the most prominent species ; about the same proportion of
Agrostis vulgaris, considerably less Holcus lanatus, very
much less of Festuca ovina, and especially of Poa pratensis ;
but, on the other hand, a fair proportion of Alopecurus
pratensis (meadow foxtail), which is scarcely represented on
plot 9 ; whilst Avena pubescens, Avena flavescens (yellow
oat grass), and Poa trivialis (rough stalked meadow grass)
have gone down on both plots.

If all the nitrogen of the straw became available, it would
supply about 91b. per acre per annum, presumably a smaller
proportion at first, and more and more each year. Hence
there was on plot 13, compared with plot 9, no excess of
yield of hay and of nitrogen in the earlier years, but a
considerable excess over the latter half of the period ; and
there was an increase of about lOlb. of nitrogen per acre
per annum over the twenty years. Probably the growing
plants were enabled to avail themselves of more also of the
other supplies of nitrogen within the soil.

There was some increase in the amounts taken up of all
the mineral constituents, but the most marked was — of
potash an average of about 231b., of phosphoric acid about
41b., and of chlorine about 81b., within each case more over the
second period than over the first. There was, too, an average

M 2

164 The Rothamsted Grass Experiments.

of about 81b. more silica annually taken up on the straw plot,
but less over the second period than the first.

Some of the increased luxuriance under the influence of
the straw manuring is probably attributable to an increased
supply of manurial constituents, and some to the mechanical
effects (e.g., the " mulching " effect, or protection given by
the straw to the young shoots). And something is due,
perhaps, to the greater powers of food collection within the
soil of the species favoured.

9. 800Z6. Ammonia-salts with Mixed Mineral Manure con-
taining Potash: Plots Ha and 116. — These plots show the
effects of a considerably increased supply of ammonia-salts,
used in conjunction with the same mixed mineral manure,
including potash, as on plot 9. During the first three years
double the amount, or SOOlb. per acre, of ammonia-salts,
containing about 2001b. of ammonia, equivalent to about
1641b. of nitrogen, was annually applied. This quantity,
from its effects, appearing excessive, was then reduced to
one-half, or to the same as on plot 9, for three years ; but as
it then appeared that the maximum growth possible was not
attained, the quantity was again doubled, and has continued
so. At the period of this change, too, i.e., after the first six
years, the plot (11) was divided, and to one-half, in addition
to the ammonia-salts and the usual mixed mineral manure,
artificial silicates were thenceforward applied ; for nine years
(1862-70), 2001b. of a crude silicate of lime, and 2001b. of a
crude silicate of soda, and afterwards 4001b. of the silicate
of soda, per acre per annum were employed. The results are
summarised in Table XXXIX. opposite.

Thus an increased supply of ammonia-salts with the same
mineral manures as on plot 9, has considerably increased the
annual produce of hay, but in a much greater degree the
assimilation of nitrogen, over a given area; and it has
increased, in about the same proportion as the increase of
hay, the amount of total mineral matter taken up over that
on plot 9, where the same amount was applied by manure.

Effect of Increasing the Ammonia of the Manure. 165

The addition of silicates during the last fourteen years of
the twenty has further increased the amounts of hay yielded,
and of nitrogen and mineral matter taken up.

TABLE XXXIX. — AVERAGE PER ACRE PER ANNUM, BY 800LB. (OR
400LB.) OF AMMONIA-SALTS, AND MIXED MINERAL MANURE, WITH-

OUT, AND WITH, THE ADDITION OF SILICATES ; PLOTS lla. AND 116.

Average per Acre per Annum.

Jl!

J §£ f

4

oj

J-g

^

l*Sf

•^'§S I

isi

0 ^

1?

pi

II

s^

K|

Sg

a <

<3

HAY.

ib.

lb.

lb.

lb.

lb.

1st period, 10 years, 1856-65 ...

6002

6913

7077

+ 911

+ 164

2nd period, 10 years, 1866-75 ...

5421

6006

6909

+ 585

+ 903

Total period, 20 years, 1856-75

5711

6459

6993

+ 748

+ 534

2nd period, per cent. + or — 1st

period

— 9-7

— 13-1

- 2-4

NITROGEN.

1st period

75-7

102-7

103-7

+ 27-0

+ i-o

2nd period ..

70'7

103-5

111-7

+ 32-8

+ 8'2

Total period

73-2

103-1

107-7

+ 29-9

+ 4'6

2nd period, per cent. + or— 1st

period

- 6-6

+ 0-8

+ 7-7

MINERAL MATTER (Asn).

1st period

368-3

420-7

435-9

+ 52-4

+ 15-2

2nd period

301-0

339-7

392-1

+ 38-7

+ 52-4

Total period

334-6

380-2

414-0

+ 45-6

+ 33-8

2nd period, per cent. + or — 1st

period

-18-3

-19-3

-10-1

* 4001b. only in 1859, 1860, and 1861.

It is remarkable that, with the enormous amount of about
2001b. of ammonia, corresponding to about 1641b. of nitrogen,
applied per acre per annum, for so many years, there
appeared to be as large, if not a larger, proportion of the
nitrogen supplied taken up by the growing herbage, as where
only half the quantity was employed. There was, in fact, a
proportionally much greater increase in the assimilation of

166 The Rothamsted Grass Experiments.

nitrogen, than in the amount of total growth — that is,
produce of hay — by doubling the application. In other
words, the percentage of nitrogen in the produce was very
much increased ; it was, indeed, abnormally high, and
indicates a deficient assimilation of other constituents in
proportion to the nitrogen taken up.

An examination of the herbage shows that, with the com-
bination of mineral manures and ammonia-salts, the greater
the quantity of the latter the more nearly does the produce
become exclusively gramineous, and the more does it consist
of a few of the most freely growing grasses (Dactylis gloni-
erata, Alopecurus pratensis, Agrostis vulgaris, Holcus lanatus,
Avena elatior). These, too, take possession of the ground
very much in tufts or patches, and grow coarse, strong seed-
stems, and broad, flaggy, dark-green leaves. The herbage is,
in fact, very coarse, often laid, and dead at the bottom before
it is ripe ; indeed, it generally matures irregularly and
imperfectly, yielding hay of low quality, though, if the grass
were fed off young, or cut green for feeding, it would probably
be fairly good food.

The ash analyses reveal some striking differences in the
mineral composition of the produce. There is a markedly
lower percentage of lime, a lower percentage of magnesia,
and, notwithstanding the produce is so prominently grami-
neous, a lower percentage of silica, in the dry substance of
the hay, than where the growth is less forced, and the
herbage is at the same time more mixed. Of phosphoric
acid, and especially of potash, there is, on the other hand, a
considerably higher percentage in the dry substance of these
coarse gramineous crops, as also there is of chlorine. The
percentage, in the dry substance, of lime, magnesia, soda^
sulphuric acid, and silica, has decreased over the later
compared with the earlier years, whilst that of the potash
and phosphoric acid (and also that of the chlorine) has
increased over the later years. The percentage of lime
especially has decreased the more, but that of the soda,

Behaviour of Silicates. 167

sulphuric acid, and silica the less, the greater the amount of
the ammonia- salts applied, and the coarser the herbage.

The only noticeable difference between the two plots 11 in
the percentage mineral composition of the herbage is, that
there is a slightly higher percentage of silica in that of the
plot on which it was applied, and with this there is a less
percentage of potash, and a less increased percentage of it
over the later years. At the same time, there is a somewhat
less excessive percentage of nitrogen where the silicates were
used. In the presence of the excessive supply of nitrogen,
and the increased activity of growth induced by it, the
silicates of soda and lime employed were probably effective
in other ways than merely as supplies of either silica, or
soda, or lime, to the plant. The effect was probably due in
part to reactions of the alkaline silicates within the soil ; for,
under their influence, there was more nitrogen, magnesia,
potash, phosphoric acid, sulphuric acid, and chlorine, as well
as more silica, lime, and soda, taken up ; and, with these,
more carbon assimilated.

Some useful facts may here be recalled. The dry matter
of the miscellaneous species generally contains a higher
percentage of mineral matter than that of the leguminous
herbage, and the leguminous a higher percentage than that
of the gramineous herbage; and the less matured the
produce, the higher, as a rule, will be the percentage of
mineral matter in its dry substance. Of the three, the ash of
the leguminous herbage contains the highest, and that of the
gramineous herbage the lowest percentage of lime, that of
the miscellaneous being intermediate. On the other hand,
the ash of the gramineous herbage is richest in potash, and
it will be the richer in potash, and the less rich in lime, the
greater the proportion of stem to leaf.

10. 550Z&. Nitrate of Soda, with Mixed Mineral Manure
containing Potash-, Plot 14. — The comparison in this case
(Table XL.) is with plot 9, which received mineral manures
along with nitrogen applied as ammonia-salts. The experi-

168

The Rothamsted Grass Experiments.

merits with the nitrate were commenced two years later than
those with the ammonia-salts, so that the periods are 8, 10,
and 18, instead of 10, 10, and 20 years.

TABLE XL. — AVERAGE PER ACRE PER ANNUM, BY 550LB. NITRATE OP
SODA, AND MIXED MINERAL MANURE, INCLUDING POTASH ; PLOT 14.

Average per Acre per Annum.

Plot 9.
Mineral Manure
and Ammonia-
salts.

Plot 14.
Miner nl Manure
and Nitrate of
Soda.

Plot 14.
+ or -Plot y.

HAT.

1st period, 8 years, 1858-65
2nd period, 10 years, 1866-75
Total period, 18 years, 1858-75 ...
2nd period, per cent. + or — 1st
period

Ib.
5904
5421
5636

- 8'2

75-4

70-7
72-8

- 6-2

356-1
301-0
325-5

-15-5

Ib.
5944
6777
6407

+ 14-0

67-6
70-6
69-3

+ 4-4

357-2
373-2
366-1

+ 4-4

Ib.
+ 40
+ 1356
+ 771

- 7-8

- o-i

- 3-5

+ 1-1
+ 72-2
+ 40-6

NITROGEN.
1st period

2nd period

Total period

2nd period, per cent. + or — 1st
period

MINERAL MATTER (Asn).
1st period

2nd period

Total period

2nd period, per cent. + or — 1st
period

As in the two cases of ammonia-salts and nitrate, each
without mineral manure, so here, with mineral manure, there
is a much greater produce of hay when the nitrogen is applied
as nitrate of soda. There is also more mineral matter taken
up under the influence of the nitrate, but rather less nitrogen ;
and whereas with the ammonia-salts there is a decline in
annual yield of hay, and in the quantity of nitrogen and
mineral matter taken up, over the second period compared
with the first, there is, under the influence of the nitrate

Ammonia-salts versus Nitrate of Soda. 169

of soda, an increase in each of the three items over the later
period.

In considering these results, special reference should be
made to the droughty season of 1870, in which the fact that
the nitrogen of nitrate of soda distributes much more rapidly
through the soil than does that of ammonia- salts was in a
marked degree illustrated.

As compared with plot 9, the nitrate plot (14) produces
more Alopecurus, very much more Bromus inollis and Poa
trivialis, and more Lolium perenne ; but, on the other hand,
very much less Poa pratensis and Festuca ovina, much less
Agrostis, and, upon the whole, less Holcus, with, in all, more
total G-raminese. The nitrate plot is the richer in leguminous
herbage, but this on both plots falls below 1 per cent. Of
miscellaneous herbage, the ammonia-salts have generally
yielded more than the nitrate, but latterly not so much.
Eumex Acetosa is most prominent under the influence of the
ammonia-salts, Conopodium denudatum and Achillea Mille-
folium coming next in order, and others occurring in quite
immaterial amounts. On the nitrate plot also Eumex Acetosa
comes to the front, but Anthriscus sylvestris (beaked parsley,
an umbellifer) is about equally developed, and is increasing ;
Achillea Millefolium and Conopodium denudatum follow
next ; whilst Taraxacum officinale (dandelion) is more plenti-
ful than on plot 9.

The striking differences in the amount of produce, in the
amount of total mineral matter taken up, in the botanical
composition of the herbage, in the character and distribution
of the roots, and in the influence of the vegetation and the
manures on the mechanical condition and on the chemical
composition of the subsoil, according as the nitrogen is
applied as ammonia-salts or as nitrate of soda, are associated
with important differences in the chemical composition of the
produce. As shown in Table XL., there was, with much
more vegetable matter produced, even less nitrogen taken up
and retained in the produce grown by the nitrate than in that

170 The Rothamsted Grass Experiments.

by the ammonia- salts. The result was a much lower per-
centage of nitrogen in the dry matter of the produce of the
nitrate plot, and, indeed, a much more normal percentage for
this almost purely gramineous herbage. This lower per-
centage of nitrogen, with at the same time increased crop,
implies, of course, an increased assimilation of carbon — that
is to say, more activity of growth — in proportion to the
nitrogen taken up. There was coincidently, more lime,
magnesia, phosphoric acid, and sulphuric acid, considerably
more silica, and very much more soda, taken up under the
influence of the nitrate ; but there was, on the other hand,
considerably less potash, and very much less chlorine, taken
up.

11. 275$. Nitrate of Soda, with Mixed Mineral Manure con-
taining Potash : Plot 16. — Here the nitrate is only half as
much as was applied on plot 14, and contains of course only
half as much nitrogen as the ammonia-salts on plot 9. The
experiments were commenced in 1858, so that the periods are
8, 10, and 18 years. Table XLL, opposite, records the results.

The most remarkable fact here brought out is that an
average of only 6|lb. less nitrogen was annually removed
than where the double amount of nitrate, supplying annually
about 411b. more nitrogen, was applied.

On both plots, 14 and 16, Alopecurus pratensis is plentiful
and increasing. Plot 16 had a larger proportion of Hole us
lanatus, and much more Avena flavescens, Agrostis vulgaris,
and Festuca ovina ; but there was a smaller proportion of
Lolium perenne, less Dactylis glomerata, and very much less
of both Poa trivialis and Bromus mollis ; in all, a less pro-
portion of total G-ramineae than with the double amount of
nitrate. There was much more leguminous herbage,
especially in the later years, and chiefly Lathyrus pratensis.
There was also a larger, but a decreasing, proportion belong-
ing to other orders. In fact plot 16, with less nitrate, pro-
duced a greater variety of grasses, but a less proportion of
the more freely growing species ; whilst it grew a greater

Large & Small Dressings of Nitrate of8oda, with Minerals. 171

number of species, and a considerably greater percentage by
weight, of those belonging to the leguminous and miscel-
laneous orders.

TABLE XLI. — AVERAGE PER ACRE PER ANNUM, BY 275LB. NITRATE
OF SODA, WITH MIXED MINERAL MANURE CONTAINING POTASH ;
PLOT 16.

Average per Acre per Annum.

Plot 14.
Mineral Mftnure
and 5501b.
Nitrate of Soda.

Plot 16.
Mineral Manure
and 2751b.
Nitrate of Soda.

Plot 16.
+ or -Plot 14.

HAT.

1st period, 8 years, 1858-65
2nd period, 10 years, 1866-75
Total period, 18 years, 1858-75 ...
2nd period, per cent. + or — 1st
period

NITROGEN.
1st period .

Ib.

5944
6777
6407

+ 14-0
67-6

Ib.
5058
5332
5210

+ 5-4
63-0

Ib.
- 886
-1445
-1197

- 4-6

2nd period

70-6

62-4

- 8-2

Total period

69-3

62-6

- 6-7

2nd period, per cent. + or — 1st
period

+ 4-4

-1-0

MINERAL MATTER (Asn).
1st period

357-2

320-9

- 36-3

2nd period

373-2

307-9

- 65-3

Total period

366-1

313-7

- 52-4

2nd period, per cent. + or — 1st
period

-f 4-4

-4-1

With a greater number of species, possessing different
habits of growth, there is a higher percentage of lime, an
equal percentage of magnesia, a considerably higher per-
centage of potash and phosphoric acid, a higher percentage
of sulphuric acid, about an equal percentage of chlorine and
of silica, but considerably less of soda, in the dry substance
of the produce grown by the smaller amount of nitrate of
soda. And, reckoned per acre, there is very nearly as much

172 The Rothamsted Grass Experiments.

lime, potash, and phosphoric acid, and not much less
magnesia, though considerably less sulphuric acid, and
chlorine, and very much less soda, and, at the same time, con-
siderably less silica, gathered up under the influence of the
smaller amount of nitrate.

Although the herbage of some of the best pastures of some
of the best grazing districts of the country comprises but a
small total number of species, it nevertheless includes a con-
siderable proportion of other than gramineous species, and it
is especially rich in Leguminosae. Comparing the produce of
the different experimental plots, however, the more complex
the herbage, the higher, as a rule, is the quality of the hay,
and it is especially so when leguminous species are in fair
proportion ; and, doubtless, the quality of the hay of plot 16
would be higher than that of any other of the plots yielding
an equal weight of produce The result is, then, that with
the mixed mineral manure, and a not excessive amount of
nitrate of soda, there is both a large actual amount of pro-
duce, a large amount in proportion to the nitrate used, and a
comparatively high quality of the hay.

12. 400Z6. Ammonia-salts, and Superphosphate of Lime :
Plot 46. — As the experiment with ammonia-salts and super-
phosphate did not begin until the fourth year, the periods
selected for the comparisons in Table XLIL, on page 173, are
seven, ten, and seventeen years.

Over the twenty years, ammonia-salts alone gave about
one-fourth more produce than was obtained without manure,
and the figures opposite show that, over the seven years of the
comparative trial, the mixture of ammonia-salts and super-
phosphate gave over one-third more produce than the
ammonia- salts alone. But superphosphate alone (plot 4a,
Table XXXV., page 154) gave scarcely any more produce
than the unmanured plot, doubtless owing to a deficiency
of nitrogen available to such plants as were developed.

Not only was there much more total produce, but there was
much more nitrogen and total mineral matter annually removed

Ammonia-salts, with and without Superphosphate. 173

when the superphosphate as well as the ammonia- salts was
used ; but in both cases there was nearly the same percentage
reduction in yield of hay, of nitrogen, and of mineral matter
over the last ten compared with the first seven years (Table
XXXI.) More lime, magnesia, phosphoric acid, and sulphuric

TABLE XLII. — AVERAGE PER ACRE PER ANNUM, BY 400LB. AMMONIA-
SALTS, AND SUPERPHOSPHATE OF LIME ; PLOT 46.

Average per Acre per Annum.

Plot 5.
Ammonia-salts
alone.

Plot 46.
Ammonia-salts
and
Superphosphate.

Plot 46.
+ or- Plot 5.

HAT.

1st period, 7 years, 1859-65 .

3202

lb.

4440

lb.
+ 1238

2nd period, 10 years, 1866-75
Total period, 17 years, 1859-75 ...
2nd period, per cent. + or — 1st
period

2471

2772

— 22-8

3414
3837

— 23-1

+ 943
+ 1065

NITROGEN.
1st period . ...

56-0

67-6

+ 11-6

2nd period ....

47-3

57-8

+ 10-5

Total period
2nd period, per cent. + or — 1st
period

50-9
-15-5

61-8
-14-5

+ 10-9

MINERAL MATTER (ASH).
1st period

163-0

250-0

+ 87-0

2nd period

108-9

162'3

+ 53-4

Total period

131-2

198-3

+ 67-1

2nd period, per cent. + or — 1st
period

-33-2

— 35-1

acid were taken up than could be accounted for in the super-
phosphate supplied. There was also more potash, more soda,
and more silica. The excess of all these mineral ingredients
must have been supplied by the soil itself. In the second
period, as compared with the first, there was a very great
decline in the average annual amount taken up of almost every
mineral constituent, except phosphoric acid and magnesia, in
which the reduction was much less. Further, whilst in the

174 The Rothamsted Grass Experiments.

dry substance there was (excepting lime) an abnormally high
percentage of the mineral constituents supplied, there was
an abnormally low percentage of those not supplied, again
indicating that the point of exhaustion of available supply
was reached. With this evidence of exhaustion of mineral
matter, there was an abnormally high percentage of nitrogen
in the dry substance of the produce of both plots, but
especially in that by the ammonia-salts alone.

With this chemical evidence of repletion of nitrogenous,
and deficiency of mineral supply, the botanical character of
the herbage was equally significant. On both plots it became
more and more gramineous, particularly with the superphos-
phate, where the number of species also declined more. On
both plots Festuca ovina became by far the most prominent
grass, constantly increasing, and in 1877 it made up more
than half the total crop in each case. Agrostis vulgaris came
next in abundance, and in 1877 the two grasses, Festuca
ovina and Agrostis vulgaris, made up nearly 83 per cent, of
the total produce grown by ammonia-salts alone, and within
a fraction of 80 per cent, of that by ammonia-salts and super-
phosphate. On the plot with ammonia-salts alone, three other
grasses, Anthoxanthum, Holcus, and Dactylis, together con-
tributed about 10 per cent, more, and no other single grass as
much as a quarter of 1 per cent. On the other plot (46), Anthox-
anthum, Alopecurus, Holcus, Avena elatior, Poa pratensis,
and Dactylis, together contributed an additional 14 per cent.,
and no other grass more than a small fraction of 1 per cent.

Not only did the herbage, then, consist mainly of two
grasses, but the one which took the lead, Festuca ovina, is the
prevalent plant on poor common lands ; and on both these
plots both grasses showed very inferior characters of develop-
ment. They consisted chiefly of very dark green leafy
herbage, growing in patches or tufts, with very little tendency
to produce stem and seed, and, particularly in dry seasons,
dying at the bottom, without properly ripening, and always
yielding a soft, woolly, and very inferior hay.

Ammonia-salts and Minerals, with and without Potash. 175

13. 400/&. Ammonia-salts, and Mixed Mineral Manure, with
and without Potash : Plots 9 and 10. — During the first six
years of the twenty both plots annually received 4001b. of
ammonia- salts and the " mixed mineral manure," including'
potash. Plot 10 received 20001b. sawdust per acre per annum
during those six years, and again in the seventh year, but
without effect. After the first six years both plots were
manured as before, with the important exception that, from
that date the potash was omitted from the manure of plot 10,
and a somewhat increased amount of sulphate of soda was
applied instead. The results are summarised in Table
XLIII.

TABLE XLIII. — AVEEAGE PER ACRE PER ANNUM, BY 400LB. AMMONIA-
SALTS, AND MIXED MINERAL MANURE, WITH AND WITHOUT POTASH
PLOTS 9 AND 10.

Average per Acre per Annum.

&

<-* ~4 ^*

<a<-So

F"

Plot 10.
With Potash
6 years, without
14 years.

pi

o"S

£S

*l

HAY.

1st period 6 years 1856-61

lb.
6349
5438
5711

-14-3

78-7
70-9
73-2

- 9-9

406-8
303-7
334-6

-25-3

lb.
6223
4725
5174

-24-1

73-3

72-7
72-9

- 0-8

419-8
243-8
296-6

-41-9

lb.
-126
-713
-537

- 5-4
+ 1'8
- 0-3

+ 13-0
- 59-9
- 38-0

2nd period, 14 years, 1862-75
Total period, 20 years, 1856-75 ...
2nd period, per cent. + or — 1st
period

NITROGEN.
1st period

2nd period

Total period . ...

2nd period, per cent. + or — 1st
period

MINERAL MATTER (Asn).
1st period ...

2nd period . .

Total period

2nd period, per cent. + or — 1st
period

176 The Rothamsted Grass Experiments.

Of hay, the average produce per acre per annum over the
twenty years was : without manure, 23831b. ; with ammonia-
salts alone, 29461b. ; with ammonia-salts and superphosphate
(seventeen years), 38371b. ; with ammonia-salts and mixed
mineral manure, with potash for six years, and without
potash fourteen years, 51741b. ; and with ammonia-salts and
the mixed mineral manure, including potash every year,
57111b.

It is remarkable that there was, over the twenty years,
almost identically the same amount of nitrogen taken up
where the potash was discontinued as where it was continu-
ously applied, and during the period of the omission there
was even rather more. But of mineral matter, about one-fifth
less has been taken off, over the fourteen years, in the crop
grown without than with potash ; and the falling off in amount,
compared with the earlier period, is nearly 42 per cent, with-
out, against only 25 per cent, with, the potash. Still, on
plot 10 there was much less hay produced, and much less
mineral matter taken up, than on plot 9, where the applica-
tion of potash was continued.

It appears that, on plot 10, there was a considerable
residue of potash remaining after the six years' application ;
that this was yielded up less freely than that from fresh
supplies, but more freely in the earlier than in the later years ;
and that at the end of the fourteen years a residue still
remained, of which some still continued to be yielded up.
That this should be so is quite consistent with results
obtained on arable land, which, as already referred to, show,
both in the amounts of crop and in its chemical composition,
the effects of a residue of potash applied more than twenty-
five years previously. Supposing, however, that in the
presence of an artificial supply of potash less was yielded up
from the soil itself, the residue remaining would of course
be by so much less than 2001b. at the end of the fourteen
years.

About an equal number of species, and nearly the same

Influence of Potash on Herbage. 177

species, were represented on plots 9 and 10, but on neither
was the total number much more than half as many as with-
out manure. On plot 9, receiving potash, the six grasses
which have become most prominent are (somewhat in the
following order) Poa pratensis, Agrostis vulgaris, Festuca
ovina, Dactylis glomerata, Avena elatior, and Holcus lanatus,
and they together average more than 80 per cent, of the pro-
duce. The seven most prominent grasses on plot 10, with
the discontinued potash, are Festuca ovina, Agrostis vulgaris,
Alopecurus pratensis, Avena elatior, Poa pratensis, Holcus
lanatus, and Dactylis glomerata, which together make up
about as much as the six on plot 9. The chief differences are
that whilst Alopecurus has much increased on plot 10, there
is scarcely any of it on plot 9 ; Holcus and Poa pratensis are
the less prominent on plot 10 ; Dactylis glomerata has
decreased on plot 10, but increased on plot 9 ; Festuca ovina
has considerably increased on both plots, but the more on
plot 10. Avena pubescens, Avena flavescens, Poa trivialis,
Bromus mollis, and Lolium perenne have almost disappeared
on both plots. Leguminous species represent only a small
fraction of 1 per cent on both plots. Miscellaneous species
have considerably decreased, more in weight than in number,
on both plots.

On plot 9, with the continuous potash, the grasses showed
much tendency to produce stem and seed, and to mature. On
plot 10, with the potash discontinued, the proportion of leaf
to stem was very much greater, the herbage was patchy, of a
much darker green colour, and matured unevenly and imper-
fectly. There was, in fact, a relative plethora of nitrogen,
and with the deficiency of potash a deficient assimilation of
carbon.

Comparing the results on these plots with those on plots
7 and 8, where mixed mineral manure alone, with potash and
without potash, was employed, it was noticed that on both the
ammonia plots (9 and 10) the produce was chiefly grami-
neous, containing scar3ely any leguminous herbage, and but

178 The Bothamsted Grass Experiments.

few prominent miscellaneous species, whilst the flora was not
very materially affected by the discontinuance of the potash ;
though, without the potash, the gramineous herbage showed
very different, and much inferior, characters of development.
Without ammonia, on the other hand, and with the potash,
the produce contained a large proportion of leguminous
herbage ; and the most prominent effect of the discontinu-
ance of the potash was the reduction of the leguminous her-
bage to a very insignificant amount, whilst, at the same time,
the character of development of the grasses was deteriorated.

The importance of a liberal available supply of potash
within the soil in order to grow large, properly developed, and
well matured hay crops, is now made evident. With a fairly
mixed herbage, and only moderate nitrogenous manuring, a
liberal supply of potash will increase, or a deficiency of it will
greatly diminish, the growth of leguminous plants ; whilst,
at the same time, both the quantity and the character of
growth of the gramineous herbage will be affected. Or, with
a predominantly gramineous herbage, and full supply of
nitrogen, a deficient supply of potash will diminish the
amount of produce and consequently the efficiency of the
nitrogenous manure, and it will, moreover, lead to a deterio-
tion in the character of the growth.

14. Mixed Mineral Manure alone 7 years ; succeeding
Ammonia-salts alone 13 years : Plot 6. — The figures in Table
XLIV. illustrate the effects of applying the mixed mineral
manure, including potash, for seven years in succession on a
plot which had received ammonia-salts, without mineral
manure, in each of the thirteen preceding seasons (with saw-
dust in addition the first seven years). The comparisons are
with plot 5, which received the same quantity of ammonia-
salts (without sawdust) in each of the twenty years as was
applied to plot 6 during the first thirteen years ; and with
plot 7, which received the same mixed mineral manure
throughout the twenty years as plot 6 received during the
last seven years.

Unexhausted Residue of Manures.

179

The objects sought in discontinuing the application of
ammonia- salts on plot 6, and substituting therefor the
mixed mineral manure, were to determine the effects of the
latter, not only on the amount of produce, but on the character
of the herbage and of its chemical composition ; and especially
to acquire data in reference to the question whether any, or
how much, of the nitrogen of the previous applications, which
had not been already recovered in the increase of crop, would
be so under the influence of liberal manuring.

TABLE XLIV. — AVERAGE PER ACRE PER ANNUM, BY MIXED MINERAL
MANURE ALONE 7 YEARS, AFTER AMMONIA-SALTS ALONE 13 YEARS ;
PLOT 6.

Average per Acre per Annum.

Plot 5.
Ammonia-salts alone,
20 years.

Plot 6.
Ammonia-salts, 13
years, Mineral Manure
7 years.

Plot 7.
Mineral Manure alone,
20 years.

>0

3

Ib.
+ 108
+ 1245
+ 506

+ 1-3
- 1-1

+ 0-5

+ 3-8
+ 93-8
+ 35-3

t>^

»|

l\

HAY.

1st period, 13 years, 1856-68 ...
2nd period, 7 years, 1869-75 ...
Total period, 20 years, 1856-75
2nd period, per cent. -f- or— 1st
period

Ib.
3317
2257
2946

-32-0

57-8
43-0
52-6

-25-6

171-8
95-4
145-1

-44-5

Ib.
3425
3502
3452

+ 2-2

59-1
41-9
53-1

-29-1

175-6

189-2
1804

+ 7-7

Ib.
3914
4040
3958

+ 3-2

56-9
55-9
56-6

-1*8

254-5
252-8
254-0

-0-7

Ib.
- 439
- 538
- 506

+ 2-2
-14-0
- 3-5

-78-9
-63-6
-73-6

NITROGEN.
1st period

2nd period

Total period ...

2nd period, per cent. -for — 1st
period

MINERAL MATTER (Asn).
1st period

2nd period

Total period

2nd period, per cent. + or — 1st
period

A consideration of the facts shows it is probable that, after
the thirteen years' application of ammonia-salts to plot 6,

N 2

180 The Rothamsted Grass Experiments.

some, at any rate, of the supplied nitrogen remained within the
soil, and that some of this was available to the growing plants
during the succeeding seven years. Moreover, a glance at the
table will show that as much nitrogen was taken up as on the
plot where the ammonia-salts were still annually applied ;
but in reference to this point there is the significant fact that
the surface soil of plot 6 showed at the end of twenty years a
notably lower percentage of nitrogen than the corresponding
layer of plot 5, thus pointing to the soil itself as being the
source of nitrogen. The reader may here find it instructive to
refer back to the discussion of the results on plot 7, page 149.
Since the application of the mixed mineral manure, the
flora of plot 6 has become much more complex. Agrostis
vulgaris and Festuca ovina are less predominant, Holcus
lanatus has much increased, whilst Dactylis gloinerata,
A vena elatior, Avena pubescens, Lolium perenne, and Poa
pratensis are more favoured than on plot 5. Of leguminous
plants, the creeping and comparatively surface - rooting
Lathyrus pratensis has much increased, as has the legu-
minous herbage as a whole. With its increasing complexity,
the roots of the herbage would doubtless acquire possession
of a more extended range of soil and subsoil, and more
varied powers of underground food-collection would come
into play. Whilst, therefore, some part of the nitrogen of
the increased produce obtained on the substitution of the
mineral for the nitrogenous manuring would probably be
derived from the residue of the previous applications, it is
probable that the greater part would be due to increased
power of underground food collection, by virtue of which not
only the immediately preceding, but the earlier accumula-
tions, or what may be termed the normal stores of the soil
and subsoil, would be drawn upon. That such was really the
case is concluded from the fact of the reduction of the per-
centage of nitrogen in the surface soil of plot 6, as of plot 7,
where mineral manure alone had been applied for twenty
years, and where the very complex and highly leguminous

Effect of Minerals following Ammonia-salts. 181

herbage accumulated throughout the period an otherwise
unaccountably large amount of nitrogen.

There was, over the seven years, more of every mineral
constituent (except soda) taken up on plot 6 with the mineral
manure than on plot 5 with the ammonia-salts. Of potash
there was nearly four times, of phosphoric acid about two and
a half times, of sulphuric acid about one and a half times, and
of silica (of which there was none in the manure) more than
one and a half times, as much taken up as where the
ammonia-salts were still used. There was, however, less of
every mineral constituent (except soda) taken up than where
the mixed mineral manure had been applied from the com-
mencement (plot 7).

It is obvious that it was of potash chiefly, of phosphoric
acid also notably, but of most of the mineral constituents
more or less, that the available supply had become so
deficient under the continuous application of the ammonia-
salts. It has already been shown how ineffective was a
supply of phosphoric acid (superphosphate) when used alone
(plot 4 — I.), and how comparatively little was its effect when
used in conjunction with ammonia-salts, but without potash
(plot 4&) ; and here again is strikingly brought out the
influence of a liberal available supply of potash within the
soil, both upon the quantity and the quality of the produce.
Lastly, such evidence as is forthcoming does not favour the
supposition that any considerable proportion of the nitrogen
of the ammonia-salts applied during the thirteen years, and
not recovered in the crops during the period of the applica-
tion, remained in an available condition in the soil, and was
reclaimed in the succeeding seven years under the influence
of the mixed mineral manure.

15. Equal Nitrogen and equal Potash, in Nitrate of Soda
and Sulphate of Potash, and in Nitrate of Potash; in
each case with Superphosphate of Lime : Plots 19 and 20. —
The marked effects of nitrate of soda and of sulphate of
potash pointed to the desirability of determining whether

182

The Rothamsted Grass Experiments.

nitrate of potash would be more or less effective than a
mixture of nitrate of soda and sulphate of potash, the
mixture containing the same amounts of nitrogen and
potash as the nitrate of potash. In 1872, plots 19 and 20
were set apart for this test. In each of the seven years,
1872-78 (and the experiment is still in progress), plot 19
received 2751b. nitrate of soda, and plot 20 3271b. nitrate
of potash, both containing the same weight of nitrogen.
Plot 19 also received 2901b. sulphate of potash, containing
the same weight of potash as the 3271b. nitrate of potash.
Each plot has also received annually 3i cwt. superphosphate
of lime. For comparison there is given in Table XLV. the
average produce over the same seven years, and also over the
preceding fourteen years, on plot 16, manured annually,
during the whole period of twenty-one years, with 2751b.
nitrate of soda, 3001b. sulphate of potash, lOOlb. sulphate of
soda, lOOlb. sulphate of magnesia, and 3| cwt. superphos-
phate of lime. The quantities of manures on each plot are
not repeated in the table.

TABLE XLV. — AVERAGE PER ACRE PER ANNUM, BY NITRATE OF SODA
AND SULPHATE OF POTASH, AND BY NITRATE OF POTASH, CONTAINING
EQUAL NITROGEN AND EQUAL POTASH, IN EACH CASE WITH SUPER-
PHOSPHATE; PLOTS 19 AND 20.

Per Acre per Annum.

Plot 16.

Plot 19.

Plot 20.

14 years, 1858-71
7 years, 1872-78.

14 years

HAY.

One crop only each year . . .
C First crop only each year..
....< Including second crops,
( 1875-77-78

5451
4716

5639

66-1
54-0
69-0

331-4

283-8
369-9

lb.
4368
5273

49-1
62-5

261-4
336-4

lb.
4362
5191

47-9
61-1

254-4
322-6

NITROGEN.

7 Years

'1

MINERAL MATTER (Asn).
14 years

7 years

(

*•

Different Artificial Sources of Nitrogen and of Potash. 183

Plot 16 received rather more potash, and more soda,
magnesia, and sulphuric acid, than plot 19 ; and plot 20,
with the same amount of nitrogen and potash as plot 19,
received no soda, and less sulphuric acid, but the nitric acid
and the' potash were applied to the soil in combination.

Each plot has maintained a fairly mixed herbage. With
only a moderate supply of nitrogen, and this in the form of
nitrate, and with a liberal supply of potash, leguminous
herbage increased on all three, this increase being mainly due
to Lathyrus pratensis, but partly, on plots 19 and 20, to
Trifolium repens also. On all three plots the bulk of the
gramineous herbage is made up of a good many species ; and
011 plot 16, which has been the longest under treatment,
the mixture is greater — that is, there is less predominance
of individual species ; Festuca ovina, Agrostis vulgaris,
Alopecuras pratensis, Avena flavescens, and Holcus lanatus
are somewhat equally represented, whilst Poa trivialis,
Dactylis glomerata, and Lolium perenne each show moderate
growth. On plots 19 and 20 Festuca ovina is more
prominent, and Holcus lanatus is increasing; Agrostis
vulgaris, though abundant, is decreasing ; whilst Alopecurus
pratensis, Avena flavescens, Avena pubescens, and Poa
trivialis are each fairly represented and increasing. There
is, on all three plots, a pretty normal character of growth?
fair proportion of stem, and tendency to maturation of the
grasses.

The yields of dry matter, of nitrogen, and of mineral
matter indicate greater maturation or ripeness on plot 20 with
the nitrate of potash, than on plot 19 with the nitrate of soda
and sulphate of potash.

Analyses of the ash of the first crops for the seven years,
and of the second crops for the three years, show that the
produce of plot 16, with the fuller mineral manure and
longer continuance of the experiment, contained, both per
acre and per cent, in its dry substance, more phosphoric acid,
considerably more potash, and rather more magnesia, than

184 The Rothamsted Grass Experiments.

that of either plot 19 or 20. But it contained less lime
than that of either of the other plots, and less soda and
chlorine than that of plot 19 ; whilst that of plot 20, with
nitrate of potash, but without soda in the manure, contained
very much less soda than that of either plot 16 or plot 19
with it. The produce of plot 20 contained less sulphuric
acid, chlorine, and silica, than that of plot 19, but more lime,
and especially more potash, both per acre and per cent.

The conclusion is that there is no marked difference in the
amount, or in the botanical composition, of the produce,
whether the nitrogen and the potash be supplied as nitrate
of soda and sulphate of potash, or as nitrate of potash ; but
the data at command as to the chemical composition would
indicate a somewhat more matured condition of the produce
grown by the nitrate of potash.

The great practical value of the Rothamsted experiments
is very well illustrated in the two cases now to be brought
under notice. They deal, on the one hand, with the effects
upon the herbage of permanent meadow produced by the
application of a mixture containing the ash-constituents, and
the nitrogen, removed by the crop ; and, on the other hand,
with the results obtained after the application of farmyard
manure. In the one case the results serve to demonstrate
the fallacy of a weighty proposition first enunciated by
Liebig, and, in the second case, they throw much light
upon the solution of that important question — the fate of the
nitrogen of the soil.

16. Mixture supplying the Ash-constituents, and the Nitrogen,
of 1 ton of Hay : Plot 18.— This is the last of the series of
experiments with artificial and chemical manuring substances.
Commencing in 1865, this plot has annually received a
mixture containing the quantities of potash, soda, lime,
magnesia, phosphoric acid, silica, and nitrogen, contained in
1 ton of hay, and it also supplied sulphuric acid and chlorine
in abundance. The object in view was, in part, to put to the
test of direct experiment the principles of manuring set forth

Liebig's Theory put to the Test. 185

by Liebig, according to which all the constituents, neither
more nor less, removed in crops, should be returned to the
soil. Liebig wrote, in his " Principles of Agricultural
Chemistry : " " Our first object will naturally be, to restore
to the soil the mineral constituents in the same quantity and
in the same proportions as those in which they have been
removed in the crops ; and none must be omitted" Another
object of this experiment was to acquire data as to the
proportion in which the several constituents artificially
supplied would be removed in the increase of crop.

The manures actually applied, and the constituents they
contained, were as follows :

/•38-Olb. potash
761b. commercial chloride of potassium •< 7*01b. soda

C36'71b. chlorine

351b. sulphate of magnesia J * 61b" magnesia

(. 11 'lib. sulphuric acid

(261b. bone ash .. ( H'Olb. lime

(. 8'21b. phosphoric acid
(.261b. sulphuric acid (sp. gr. I' 7) 1 6' 91b. sulphuric acid

501b. silicate of soda ( ?lb- 8oda

(. 23'Olb. soluble silica

501b. silicate of lime ( Plb. lime

I 8-Olb. soluble silica
YSO'llb. nitrogen

1641b. " ammonia-salts " J 44'41b. sulphuric acid

(46'2lb. chlorine

If the quantity of constituents as shown in the right-hand
column be compared with the contents of IJtons of hay as
given previously in Table XXIX., page 142, it will be seen
that fully sufficient of each constituent has been supplied for
an increase of 1 ton of hay, whilst of soda, sulphuric acid, and
chlorine there is more than sufficient.

The experiment was carried on for eleven years, and in
Table XL VI., page 186, comparison is made with the results
on the unmanured plot during the same eleven years.

It may be seen that the average yield of hay over the
first half of the period was about 35cwt., and over the second

186

The Rothamsted Grass Experiments.

half rather less than 30cwt. ; whereas over the whole period it
was rather more than 32cwt. Hence there was a reduction in
the total produce during the later years. But, in order to
estimate the effects of the manure, it is necessary to deduct
something for the annual yield of the soil and seasons ; and,
as plot 18 had, like plot 3, been unmanured from the com-

TABLE XLVL— AVERAGE PER ACRE PER ANNUM, BY A MIXTURE
SUPPLYING THE ASH CONSTITUENTS, AND THE NlTROGEN, OP 1
TON OF HAY; PLOT 18.

Average per Acre per Annum.

Plot 3.
Without manure.

Plot 18.
Nitrogen and
ash constituents
of 1 ton of bay.

Plot 18.
+ or- Plot 3.

HAY.

1st period, 5| years

Ib.
2514

Ib.
3908

Ib.
+ 1394

2nd period, 5£ years
Total period, 11 years, 1865-75 ...
2nd period, per cent. + or — 1st
period ...

1787
2151

— 28'9

3301
3604

— 15-5

+ 1514
+ 1453

NITROGEN.
1st period .

35-1

50-1

+ 15'0

24-3

39-9

+ 15'6

Total period

29-7

45-0

+ 15-3

2nd period, per cent. + or — 1st
period

— 30-8

— 20-4

MINERAL MATTER (Asn).
1st period

146-7

222-7

+ 76-0

2nd period

96-9

175-7

+ 78'8

Total period

121-8

199-2

+ 77'4

2nd period, per cent. + or — 1st
period ,

-33-9

-21-1

mencement of the trials in 1856, it will be fair to deduct
from the total amount that obtained on the unmanured
plot. The differences, given in the right-hand column, show
an average annual increase of produce, due to the manure, of
not quite 12Jcwt. during the first period, about 13|cwt.
during the second, and not quite 13cwt. over the whole

Liebig's Principle Refuted. 187

period. The increased amount of nitrogen in the produce
is only about 45 per cent, of that supplied ; and the increase
of mineral matter removed is not much more than half as
much as would be contained in 1 ton of hay.

It is clear, then, that the annual supply of not only the
mineral constituents, but the nitrogen also, of 1 ton of hay>
yielded less than two-thirds instead of 1 ton of increase
of produce; moreover, the character of the, herbage was
materially modified. The gramineous herbage was much
increased, the leguminous much reduced.

The chief interest of this plot lies, perhaps, in the circum-
stance of its putting to the test the principle enunciated by
Liebig, and quoted above. Liebig, in his earlier writings,
did not recognise the fact that a considerable proportion of
the constituents removed from the land in crops is, in the
actual practice of agriculture, periodically returned to it,
and that, therefore, the loss to the soil is not measured by
the amount of constituents in the crops grown, but more
nearly by that in the produce sold off the farm. Further,
his recommendations for the carrying out of his principle
were confined to the application of the " mineral " or asJi-
constituents ; he maintaining that the atmosphere would
supply the necessary nitrogen. It is true that, subsequently,
in the course of controversy, he changed the meaning of his
terms, and then included ammonia- salts in the category of
mineral manures. It is seen, however, that even with a supply
of the amount of nitrogen, as well as ash-constituents, con-
tained in 1 ton of hay, not two-thirds of a ton of increase of
produce was obtained.

With regard to the applicability of Liebig's principle,
certain considerations must not be overlooked First, there
is no conceivable condition of chemical combination, and of
distribution within the soil, in which the various constituents
could be annually supplied so as to be all annually taken up
by growing vegetation ; and there is conclusive evidence that,
in some cases, the unrecovered residue is, in greater or less

188 The Rothamsted Grass Experiments.

part, lost by drainage ; and that, so far as it k not so, it
becomes so locked up, or distributed within the soil, that it
is — at any rate very slowly, and, in some cases, perhaps
never fully — recovered in subsequent crops.

Secondly, the principle ignores the difference in the
character and capabilities of different soils. Take two
opposite cases : A light, porous, almost exclusively sandy
soil, which itself yields up little or nothing to growing
plants, but which may, nevertheless, produce good crops
under high farming, will probably suffer great loss of
manurial constituents by natural drainage; so that, if no
more were to be supplied than were removed, there must
obviously be a decline of fertility. Suppose, on the other
hand, a rich and deep loam, which would, under good
mechanical cultivation and drainage, supply annually a
considerable amount of potash, for example, to say nothing
of other constituents, for hundreds and perhaps for thousands
of years ; surely, in such a case, it is not necessary to supply
as much in manure as has been removed in the crops.

Further, experience teaches that, in the actual condition
of our soils, and of agricultural practice, the exact compo-
sition of the crops we remove, or wish to grow, is no direct
guide to the description and the amount of manurial con-
stituents which will be most effective. Thus, an average
crop of wheat will remove even rather more phosphoric acid
than an average crop of barley; but experience teaches
that, in the case of land of the same description, and in the
same condition, superphosphate of lime is, as a rule, used
with very much more benefit to the spring-sown barley than
to the autumn-sown wheat. The wheat, being put in four
or five months earlier, has so much more time for root-
distribution, and acquires a greater capability for food
collection. The barley, on the other hand, depends very
much more upon the stores available within the surface
soil. Again, superphosphate is, in practice, of very special
benefit to the so-called " root crops," though the amount of

Rational Principles of Manuring. 189

phosphoric acid they take up, compared with other crops,
would not indicate this. Then, turning from the mineral or
ash-constituents to the nitrogen, an average crop of beans
will contain from two to three, and one of clover-hay from
three to four, or more, times as much nitrogen as one of
wheat or barley ; but land in such condition as to grow
a full crop of the rich-in-nitrogen beans or clover, without
nitrogenous manure, would not grow a full crop of wheat
or barley, containing so much less nitrogen, without liberal
nitrogenous manuring.

It is, then, under the existing conditions of practical agri-
culture, certainly not necessary to supply to the land all the
constituents that have been removed from it, or that would
be contained in the crops it is wished to grow, and neither
more nor less of them than would be so removed. On the
contrary, we should supply all, or only some, and more or
less, according to the circumstances.

17. Farmyard Manure alone, and with Ammonia-salts in
addition: Plots 2 and 1. — The effects of various important
individual constituents of manures, and of various mixtures
of them, having been discussed, it now remains to interpret
the results obtained on the application of that complex and
heterogeneous mixture, farmyard manure.

For eight years, 1856-63, plot 2 received annually farm-
yard manure at the rate of 14 tons per acre. Over the same
period, plot 1 received the same quantities of farmyard
manure, but with 200lb. ammonia-salts per acre per annum
in addition. At the end of the eight years the application
of farmyard manure was stopped on both plots, but the
ammonia-salts were still annually applied to plot 1. The
reason the farmyard manure was withheld was partly because
so large an annual application was obviously not thoroughly
taken up by the soil, and it was thought somewhat
obstructed the vegetation; and partly because calcula-
tion indicated how small a proportion of the constituents
applied was recovered in the increase of crop, and that there

190

The Rothamsted Grass Experiments.

was, therefore, a considerable accumulated residue, the
amount and the duration of the effects of which it would be
of interest to trace.

TABLE XLVIL — AVERAGE PER ACRE PER ANNUM, BY FARMYARD
MANURE ALONE, AND WITH AMMONIA-SALTS ; AND BY THE KESIDUE
OF THE DUNG, ALSO WITHOUT AND WITH AMMONIA-SALTS : PLOTS

2 AND 1.

Average per Acre per Annum.

Plot 3.

Unmanured
continously.

I'M •-'.
As stated
above.

Plot l.
As stated
above.

Plot 2.
+ or—
Unmannred.

Plot 1.
+ or—
Unmanured.

- i~

COO

fu+S

HAY.

1st period, 8 years, 1856-63
2nd period, 6 years, 1864-69
3rd period, 6 years, 1870-75
Total period, 20 years,
1856-75

Ib.
2665
2699
1692

2383

b.
4804
4846
2517

4130

Ib.
5538
5366
3331

4824

Ib.
+ 2139
+ 2147

+ 825

+ 1747

Ib.
+ 2873
+ 2667
+ 1639

+ 2441

Ib.
+ 734
+ 520

+ 814

+ 694

2nd period, per cent. + or —
1st period . .

+ 1-3

+ 0-9

— 3-1

3rd period, per cent. + or—
1st period

-365

-47-6

-39-9

NITROGEN.
1st period

37-2

58'2

68-3

+ 21-0

+ 31-1

+ 10*1

2nd period

37-3

53-3

63-9

+ 16-0

+ 26-6

+ 10-6

3rd period

23-0

27-5

41-0

+ 4*5

+ 18-0

+ 13*5

Total period

33-0

47-5

58-8

+ 14*5

+ 25*8

+ 11*3

2nd period, per cent. + or—
1st period . .

+ 0-3

— 8'4

— 6-4

3rd period, per cent. + or—
1st period

— 38*2

52'7

— 40-0

MINERAL MATTER (Asn).
1st period

160-5

329-1

370*5

+ 168*6

+ 210*0

+ 41*4

2nd period

152-2

275-0

291-8

+ 122-8

+ 139*6

+ 16*8

3rd period

91-5

136-4

164'V

+ 44-9

+ 73'2

+ 28*3

Total period ....

137-3

255*0

285-1

+ 117*7

+ 147'8

+ 30*1

2nd period, per cent. + or—
1st period

— 5-2

16'4

21-2

3rd period, per cent. + or—
1st period

— 43-0

— 58-6

— 55-5

In Table XL VII. the comparisons are made with the un-
manured plot 3. It has already been intimated that plot 2

Effect of Farmyard Manure. 191

received farmyard manure alone for eight years, 1856-63,
and then remained unmanured for twelve years, 1864-75 ;
whilst plot 1 received farmyard manure and ammonia-salts
for eight years, 1856-63, and ammonia-salts only for twelve
years, 1864-75.

It is remarkable that, over the first six years of the
cessation of the application of the manure, the average pro-
duce was almost exactly the same as, and even rather more
than, during the application — 48461b. against 48041b. ; and
the average annual increase was almost identical — 21471 b.
against 21391b. Though one or two of the six years were
very productive seasons, the result is still very striking.
The next six years were rather the reverse in this respect,
and this, with the fact that the then remaining residue
would doubtless be in a less readily available condition, led
to there being little more than half as much average produce,
and less than two-fifths as much average increase, as over
the first six years of the action of the inanurial residue.

The figures show that very much more nitrogen was
removed in the crops during the first than during the second
six years of the action of the residue. Moreover, the
amount of nitrogen taken up per acre per annum over the
first six years was less than during the eight years of
manuring. The reverse was the case with the yield of hay.

Of mineral matter, there was twice as much removed in
the first as in the second six years of the action of the
residue ; and there was of it a greater falling off in the first
six years, compared with the years of the application, than
of either the hay or the nitrogen.

The effect of the farmyard manure was to reduce the
number of species, to bring into greater prominence the
gramineous and miscellaneous herbage, but to reduce the
leguminous. The gramineous species became fewer. Poa
trivialis, Bromus mollis, and Avena flavescens were the
most prominent, whereas, without manure, not one of these
made much show, Festuca ovina, Agrostis vulgaris, Avena

192 The Rothamsted Grass Experiments.

pubescens, and Holcus lanatus being then the leading grasses.
After the supply of farmyard manure ceased, however, the
same grasses as without manure gradually became more
prominent on plot 2, whilst the Poa trivialis and Bromus
mollis had, in 1877, become insignificant in quantity. It
may be stated generally that, in the later years, there was
on plot 2, as compared with plot 3, a higher percentage and
a considerably larger quantity per acre of gramineous
herbage ; a considerably lower percentage and a smaller
amount per acre of leguminous herbage ; and a rather
lower percentage, but a somewhat larger actual amount, of
miscellaneous herbage.

As the grasses increased, and the leguminous plants
diminished, under the farmyard manure, the herbage con-
tained a considerably lower and decreasing percentage of
nitrogen ; but there was a higher percentage of total mineral
matter. Of lime, magnesia, soda, and sulphuric acid there
was more, and of potash, phosphoric acid, chlorine, and
silica very much more, removed per acre from the farmyard
manured than from the continuously unmanured plot. There
was a greater or less available residue of all the mineral
constituents, and more especially of potash, phosphoric
acid, and silica, many years after the cessation of the appli-
cation of the manure.

At what rate, and in what proportion, have the several
constituents of the farmyard manure, which were not
recovered during the years of the application, been recovered
since, and what prospect is there of their final total
recovery ?

To answer this question there is on the one hand, a toler-
ably correct estimate of the amounts of nitrogen, and of the
several mineral constituents, removed in the crops during
the separate periods, but it is a more difficult matter to
arrive at the amount of the same constituents contained in
the 112 tons of farmyard manure applied per acre to the
plot during the eight years. The following figures, however,

The Nitrogen of Farmyard Manure- 193

are based on the best available data. Of nitrogen it is
estimated that the farmyard manure will, on the average,
contain O64 per cent., and, therefore, that 20071b. were
applied per acre per annum ; or, in all, 16061b. in the eight
years. During the eight years the produce removed about
29 per cent, as much nitrogen as is thus estimated to be
supplied in the manure ; during the next six years about
20 per cent. ; and during the last six of the twenty years
about 10 per cent, more was removed, making a total of
about 59 per cent, as much removed in the whole produce of
the twenty years as had been supplied during the eight
years. But it must not be assumed that the whole of the
nitrogen of the crops came from the manure. It would be
more correct to deduct the amount obtained in the un-
manured produce, and to regard only the so-reckoned increase
as due to the manure. Estimated thus, only 10*4 per cent,
of the supplied nitrogen was recovered as increased yield
during the eight years of the application ; only 6 per cent,
more during the next six years; and only 1*7 per cent,
during the second six years since the application ; making,
in all, a recovery as increase during the twenty years of
about 18'1 per cent, of that supplied in the first eight years.

Thus, on the supposition that the whole of the nitrogen of
the produce was derived from the manure, there would still
remain, at the end of the twenty years, about 41 per cent.
of the 16061b., or about 6581b., not accounted for. But, on
the supposition that only the increase above that in the un-
manured produce was derived from the manure, which is,
doubtless, at any rate much nearer the truth, there would
remain unaccounted for, at the end of the twenty years, 81*9
per cent, of the nitrogen supplied — that is, about 13151b. ;
and yet, during the last six years of the twenty, less than
2 per cent, of the original amount supplied was recovered as
increase. The prospect of recovering the whole, or even a
considerable proportion would thus seem, to say the least,
extremely remote.

o

194

The Rothamsted Grass Experiments.

FATE OF THE NITKOGEN IN THE SOIL.

It has already been shown what large quantities of nitrogen
were supplied to the farmyard manure plots in the experi-
ments upon permanent meadow land, and how little became
recovered in the crop. Does this large amount of un-
recovered supplied nitrogen remain in the soil, and in such
a condition of combination and distribution as to be avail-
able to succeeding crops ? Or may not some of it be lost
by drainage, or in other ways, and the remainder become so
locked up or distributed as to be so slowly recoverable, if
ever, that it can be reckoned of scarcely appreciable practical
value ?

TABLE XLVIII. — ESTIMATED AMOUNT OF DRY SOIL PER ACRE AT EACH
DEPTH; NITROGEN PER CENT. IN THE DRY SOILS, AND ESTIMATED
NITROGEN PER ACRE, AT EACH DEPTH.

.

Nitrogen.

C3 ®

Per cent, in Dry

Per Acre.

i

Spil.

M fl

jhj

-1

llol

§ n C<N

«•!

:~ o
£<
<J

D

PH >•*> 2

O cj

S3
5

ill]

££

Ib.

Percent.

Percent.

Ib.

Ib.

Ib.

1st depth, l-9in. ...

2,183,375

0-2565

0-2800

5600

6113

+ 513

2nd depth, 10-18in.

2,835,339

0-0724

0-0849

2053

2407

+ 354

3rd depth, 19-27in.

2,964,176

0-0458

0-0473

1358

1402 + 44

4th depth, 28-36in.

3,049,436

0-0425

0-0402

1296

1226

- 70

5th depth, 37-45in.

3,104,583

0-0381

0-0337

1183

1046

-137

6th depth, 46-54in.

3,080,909

0-0376

0-0319

1158

983

-175

Total, 27in. . . .

7,982,890

9011

9922

+ 911

Total, 54in. ...

17,217,818

—

—

12,648

13,177

+ 529

Samples of the soils of all the experimental grass plots
were taken in February and March, 1876, that is, after the
experiments had been in progress twenty years, and before

Nitrogen Recovered, and not Recovered, in Hay Crop. 195

the next growing season had commenced. Table XL VIII.
shows, for plot 3 (unmanured) and plot 2 (farmyard manure
eight years, unmanured twelve years) :

1. The calculated average amounts of soil, free from stones,
and free from moisture expelled at 100° C., in pounds per
acre, for each layer of 9in. down to a depth of 54in. ;

2. The percentage of nitrogen in the dry mould of each
9in. layer, as determined by the soda-lime method ;

3. The calculated nitrogen per acre in pounds ;

4. The difference in amount between the two plots.

The foregoing results as to the accumulation of the nitrogen
of the manure within the soil, adopted on the basis of the
determinations to the total depth of 54 inches, may be
discussed in connection with the figures given below in Table
XLIX.

TABLE XLIX. — ESTIMATED NITROGEN SUPPLIED IN THE MANURE,

RECOVERED IN THE INCREASE OF THE HAY CROP, DETERMINED AS

BESIDUE IN THE SOIL, AND NOT RECOVERED IN EITHER THE INCRKASE

OR THE SOIL, TO THE DEPTH OF 54 INCHES.

Nitrogen c
Per Acre.

f Manure.
Per Cent.

Supplied in farmyard manure in 8 years
Eecovered in increase in 20 years (over Plot 3)

Not recovered in increase ...

Ib.
1606
291

18-1

1315
529

81-9
32-9

Residue, determined by soda- lime, in soil 54in. deep
Not recovered in increase or in soil

786

49-0

In interpreting these figures it is necessary to bear in mind
the uncertainty in the estimate of the amount of nitrogen
supplied in the manure ; the difficulty in determining how
much of the nitrogen of the produce was derived from that
supplied ; the possible natural difference, apart from the
influence of manure, in the soils and subsoils of the respective

o 2

196

The Eoihamsted Grass Experiments.

plots ; and also the unavoidable range of error in the
sampling of, and determinations of nitrogen in, the soils, and
the calculation of such data into quantities per acre. Taking
into account the figures relating to the first, second, and
third depths only — that is, only so far as the manured soil
shows more nitrogen than the unmanured — the result as to
accumulation and loss of nitrogen will stand as shown in
Table L.

TABLE L. — ESTIMATED NITROGEN SUPPLIED IN THE MANURE,

RECOVERED IN THE INCREASE OF THE HAY CROP, DETERMINED AS

EESIDUE IN THE SOIL, AND NOT RECOVERED IN EITHER INCREASE

OR SOIL, TO THE DEPTH OF 27 INCHES.

Nitrogen of Manure.

Per Acre.

Per Cent.

Supplied in farmyard manure in 8 years

Ib.
1606
291

18-1

Recovered in increase in 20 years (over Plot 3)

1315
911

81-9
56-7

Residue, determined by soda-lime, in soil 27in. deep

404

25-2

It is then practically beyond doubt that only a com-
paratively small proportion of the nitrogen supplied in the
farmyard manure was recovered in the increase of crop ; that
there was a considerable accumulation of it within the soil ;
and that there was also a very considerable amount so far
unaccounted for.

Such a result would seem to require some confirmation,
and this is not wanting. The great practical importance of
the subject now under discussion renders it desirable to
notice somewhat fully this confirmatory evidence. Among
the experiments at Eothamsted, wheat has been grown year
after year on the same land from 1843-4 up to the present

Nitrogen Recovered, and not Recovered, in Wheat & Barley. 197

time, and barley in like manner from 1852 up to the present
time. In the case of each crop, one plot has received 14 tons
of farmyard manure per acre per annum, and another a mixed
mineral manure, without nitrogen, every year. The results
shown in Table LI. were obtained from these plots.

TABLE LI. — NITROGEN SUPPLIED IN FARMYARD MANURE, RECOVERED,

AND NOT RECOVERED, IN THE INCREASE OF PRODUCE OF WHEAT AND

OF BARLEY.

Nitrogen per Acre per Annum.

Jfi

§g •

TS

•d

_n

Is

§ >>«5

> c3

.2 i»S

T?.§g

£ S s

!i

I1

£Sj

^

y

0 «

o ^

I'3

lb.

ib

lb.

lb.

Per cent.

Percent.

Wheat, 20 yrs., 1852-71 2007

20-1

49-3

29-2

14-6

85-4

Barley, 20 yrs., 1852-71

2007

23-9

45-3

21-4

107

89-3

For 100 Nitrogen
in Farmyard

It will be seen that the figures in the fourth column are
got by deducting those in the second from those in the third.

After making allowance for certain variations in the method
and duration of the experiments with wheat and barley,
on the one hand, and with the meadow herbage on the other*
the general result is that, according to the estimates, a higher
proportion of the supplied nitrogen was annually recovered
over the twenty years in the autumn-manured and autumn-
sown (and so longer grown) wheat, than in the spring-
manured and spring-sown barley ; and that about the same
proportion was recovered in the barley as over the eight years
in the grass. Obviously the estimate of the nitrogen in the
increase is likely to be nearer the truth in the case of the two
cereal crops than in that of the mixed herbage, the character
of which, and consequently the capability of collection from
normal sources, is so changed by manure. Any way, neither
with the wheat, the barley, nor the mixed herbage, was there

198 The Rothamsted Grass Experiments.

more than from 10 to 15 per cent, of the nitrogen, supplied
in the farmyard manure, recovered in the increase of crop
during the years of application.

In the case of the barley experiments, after the farmyard
manure had been applied for twenty years in succession, the
plot was divided. To one-half the dung was still annually
applied, but the other was now left without any further
manure. The result as to the recover} of the nitrogen of the
manure in the two cases is recorded in a table which is
not reproduced here. The figures indicate, however, that over
the twenty years of the application (on the whole plot) 10' 7 per
cent, of the supplied nitrogen is so estimated to be recovered
in i he increase of the crop ; over the next six years of the
continued application, on half the plot, 15*7 per cent ; and
over the total period of twenty- six years 11/8 per cent. On
the other half there was the 10'7 per cent, of the twenty
years' supply recovered during the twenty years, 2*9 per cent,
more of it during the next six years, and in all 13'6 per cent,
of the twenty years' supply recovered in the twenty-six years.
That is to say, of 40141 b. of nitrogen estimated to be supplied
in the twenty years, 86'4 per cent., or 34681b., remained
unaccounted for in the increase of crop at the end of the
twenty-six years.

As only 2'9 per cent, of the original supply of nitrogen
was recovered in the first six years after the application of
the manure ceased, or, say, O5 per cent, per annum, it would
obviously take from one hundred and seventy to one hundred
and eighty years to recover the whole of the 89'3 per cent.
(=100 — 10*7) which remained, if at the same rate as during
the first six years. There can be little doubt, however, that
part of the unrecovered amount has been lost by drainage or
otherwise, arid that, whatever residue remains, a gradually
decreasing proportion of it will be annually recovered.

With regard to the wheat plots, samples of the soils have
been taken, and their nitrogen estimated, with the view of
determining whether or not the whole of the unrecovered

Distribution of Nitrogen in the Soil.

199

nitrogen of the farmyard manure remains available within
the soil.

In the autumn of 1865, after twenty-two years' continuous
growth of wheat, samples of the soils were taken from many
of the plots, in each case from the first 9in., the second 9in.,
and the third 9in. in depth. Eather more than one and two-
thirds as much nitrogen was determined in the first 9in. of
the farmyard-manured as in the corresponding layer of the
unmanured plot, and about one and a half times as much as
to a corresponding depth of any of the plots receiving
artificial nitrogenous manures. Determinations made after
the farmyard manure had been applied some years longer
showed more than twice as much nitrogen in the first 9in. as
without manure. Table LII. is drawn out in the same form
as Table XL VIII., page 194.

TABLE LII.— EXPERIMENTAL WHEAT FIELD. — ESTIMATED AMOUNT OF
DRY SOIL PER ACRE AT EACH DEPTH ; NITROGEN PER CENT. IN THE
DRY SOILS, AND ESTIMATED NITROGEN PER ACRE, AT EACH DEPTH.

Average Dry Soil per
Acre, exclusive of
Stones.

Nitrogen.

Per cent, in Dry
Soil.

Per Acre.

cdl

I]

P

yi

*1

if

i

Plot 2.
Farmyard
Manure.

1*1

1st depth, l-9in. ...
2nd depth, 10-1 Sin.
3rd depth, 19-2 7in.

Ib.

2,287,155
2,712.508
2,848^973

Ib.
0-1090
0-0738
0-0561

Per cent.
0-1882
0-0810
0-0619

Ib.
2493
2002
1598

Ib.
4304
2197
1764

Ib.
+ 1811
+ 195
+ 166

Total, 27in. ...

7,848,636

—

—

6093

8265

+ 2172

Furthermore, in the case of the wheat experiments, as shown
in Table LIIL, page 200, the loss of nitrogen by drainage or
otherwise is estimated to be about 40 per cent, of that
supplied ; in that of the hay plot reckoned to the same, or
27in., of depth (Table L., page 196), 25'2 per cent., or, to
the depth of 54in. (Table XLIX., page 195), 49 per cent.

200

The Rothamsted Grass Experiments.

It is now clear that only a comparatively small proportion
of the nitrogen supplied in farmyard manure is recovered in
the increase of crop. The residue actually determined in the
soil is very large ; and it is possible that the whole of the
nitrogen existing as nitric acid, especially in the subsoil, is
not accounted for by the soda-lime determinations. It is
very remarkable, however, that, notwithstanding this great
ascertained accumulation, and the annually renewed supply
by manure, larger quantities of corn, or of straw, or of both,
and also of hay, are every year obtained by the use (in con-
junction with mineral manures) of much less than half as
much nitrogen applied as ammonia-salts or as nitrate of soda.
The wheat plots so manured, and so yielding, at the same
time show less than two-thirds as high a percentage of nitro-
gen in the first 9 inches of depth.

TABLE LIII. — EXPERIMENTS ON WHEAT. — ESTIMATED NITROGEN

SUPPLIED IN THE MANURE, RECOVERED IN THE INCREASE OF CROP,
DETERMINED AS EESIDUE IN THE SOIL, AND NOT RECOVERED IN
EITHER THE INCREASE OR THE SOIL, TO THE DEPTH OF 27 INCHES.

Nitrogen of Manure.

Per Acre.

Per Cent.

Supplied in farmyard manure in twenty-two years
Recovered as increase of crops

Ib.
4415
470

107

Not recovered in increase . ...

3945
2172

89-3
49-2

Residue determined by soda-lime in soil 27in. deep
Not recovered in increase or in soil

1773

40-1

But, according to the estimates, besides the actually-
determined large residue within the soil, there was also in
each case a very large amount of nitrogen unaccounted for,
either in the increase of crop or in the soil, to the depths
examined. Direct experiments have shown that the soil

Sources of Loss of Nitrogen. 201

in the wheat field, which is manured annually with farmyard
manure, retains near the surface, owing to its greatly
increased porosity, very much more of the rainfall than the
soil of the plots not so manured. Accordingly the drain
from the farmyard- manured plot runs much less frequently
than do the drains from the unmanured or the artificially-
manured plots. There will, obviously, be less loss of water
by drainage. But it is found that a given volume of the
drainage water from the farmyard-manured plot contains
from two to three or more times as much nitrogen, in the
form, of nitrates and nitrites, as that from the unmanured
plot, or from the plots with mineral without artificial nitro-
genous manure. Here, then, is a determined source of loss
of the supplied nitrogen. A considerable further loss is pro-
bably due to decomposition of the nitrogenous organic matter
within the soil and evolution as free nitrogen.

Consequently, there is cumulative evidence to show that
the nitrogen supplied as farmyard manure was recovered in
very small proportion during the years of its application ;
that in after years it was recovered in constantly decreasing
proportion ; that there, nevertheless, remained a considerable,
but very slowly available, residue ; that there was a con-
siderable loss by drainage ; and, finally, that there is probably
a further loss by decomposition, and evolution into the atmo-
sphere.

It is well, however, to remember that in ordinary agricul-
ture much less farmyard manure would be applied than in
these special experiments, and the losses by drainage would
from that cause alone be proportionately less. Much,
obviously, would also depend on the character of the soil and
the subsoil. Again, in an ordinary rotation of various crops,
more of the supplied nitrogen would probably be gathered up
before it finally passed beyond the reach of vegetation, than
in the case of a single cereal crop grown year after year on
the same land. For somewhat similar reasons a better result
might have been looked for with the mixed herbage, as com-

202 The Rothamsted Grass Experiments.

pared with the cereal crops, than the evidence would appear
to show. That it was not so may perhaps be taken to
indicate that, in estimating the proportion of the nitrogen of
the produce due to that supplied in the manure, it should not
be assumed that as much was derived from natural sources
as in the case of the unmanured produce ; but more should
be reckoned as derived from the manure.

FATE OF THE FERTILISING MINERALS IN THE

SOIL.

Having thus somewhat fully discussed the fate of the
nitrogen of farmyard manure, it remains to glance ever so
briefly at the behaviour of the mineral constituents.

Of the lime estimated to be supplied in the manure in the
eight years, only about 12^ per cent, was obtained in the
total produce of those years, about 9 per cent in the next six,
and little more than 4 per cent, in the last six years, making
in all only about 25 \ per cent, in the twenty years. Deduct-
ing the yield in the unmanured produce, however, there was an
increase obtained representing only 3^ per cent, of the
amount supplied during the first period, little more than
2 per cent, in the second period, and a small fraction of 1 per
cent, in the third period, or a total of less than 6 per cent, in
the twenty years.

Of the magnesia the proportion was much larger, that in
the total twenty years' produce representing 70 per cent., and
the increased yield about 21 per cent., of that supplied.

Of the potash the produce contained, in the three periods
of eight, six, and six years, 44 per cent., 22 J per cent., and
10J per cent., in all 77 per cent. But the increased yield,
over that from the unmanured plot, was only 30^-, 13, and 4J
per cent. ; in all, only about 48 per cent, of that supplied.
Pota.sh is, at any rate in moderately clayey soils, very little
subject to loss by drainage ; but it would appear that the

Minerals Recovered, and not Recovered, in Hay Crop. 203

unrecovered residue becomes so locked up (or distributed) as
to be but slowly available to succeeding crops.

Of soda, there was, in the presence of an abundant supply
of potash, even less taken up in the manured than in the
unmanured crop during the years of the application. In later
years there was some, but comparatively little, increase in the
amount compared with that in the unmanured crop.

Of phosphoric acid, in the twenty years about 57 per cent.
of the quantity estimated to be supplied was contained in the
total produce, whilst the increased yield represented 33 per
cent. The residue of the phosphoric acid, like that of the
potash, is very little subject to loss by drainage.

Of sulphuric acid there is proportionately a much less
increased amount than of phosphoric acid.

Of chlorine, the increased amount found in the produce is,
during the years of application of the manure, greater in
proportion to the estimated supply than that of any other
constituent. Both chlorine and sulphuric acid are very subject
to loss by drainage.

Lastly, of silica, the produce of the twenty years contained
about 41J per cent, as much as there was estimated to be
supplied of soluble silica in the manure, and the increased
yield of it represented about 22 per cent.

The general results may now be summarised. Of the three
more important constituents of manure — nitrogen, potash,
and phosphoric acid, when these are supplied in farmyard
manure, the nitrogen is recovered in the least proportion in
the increase of the crop for which it is supplied. It leaves
a large determinable residue within the soil, which, however,
is very slowly available to succeeding crops ; and, finally, it
is subject to serious loss by drainage, and probably by evolu-
tion into the atmosphere also. The potash so supplied is
recovered in increase in much greater proportion during the
years of the application, in much greater, though still rapidly
decreasing proportion, in subsequent years, and is very little
subject to loss by drainage. The phosphoric acid again, is

204 The Rotkamsted Grass Experiments.

recovered in much greater proportion than the nitrogen, but
not in so large a proportion as the potash ; it, too, like potash,
is but little subject to loss by drainage.

The much less immediate effect of a given amount of
nitrogen when supplied in farmyard manure than when in
ammonia-salts or nitrate of soda, the consequent necessity to
supply so much more in that form to obtain a given result,
and the very slow action of the remaining residue, are
important factors in the scientific explanation of the prac-
tically recognised much lower money value of a given amount
of nitrogen so supplied.

Reverting now to plot 1, the results on which have already
been given in Table XL VII., page 190, it remains to point out
the difference in effect when, besides the farmyard manure,
2001b. of ammonia-salts were also annually applied per acre,
both over the eight years of the application of the farmyard
manure, and over the next twelve years of the action of the
residue.

The ammonia-salts caused the herbage to assume a darker
green colour ; gramineous species became more, and both
leguminous and miscellaneous species less prominent, than
either on the unmanured plot 3, or on plot 1 with farmyard
manure alone. Compared with the latter, in the early years,
Poa trivialis and Bromus inollis were even more prominent, as
also was Dactylis glomerata, and these three grasses made up a
large proportion of the total produce. During the later years,
as on the plot without ammonia, but in a greater degree,
Agrostis vulgaris and Holcus lanatus became very prominent,
as also did Anthoxanthum odoratum and Festuca ovina;
whilst, on both plots, Poa trivialis, Bromus inollis, and
Dactylis glomerata diminished very much. Of leguminous
plants, Lathy rus pratensis is the most prominent on both
plots, but much more so on plot 2 than on plot 1 ; and, whilst
without ammonia (plot 2) there was nearly 1 per cent, of
Trifolium pratense, with it there was none. A much greater

Effect of Ammonia-Salts. 205

number of species, gramineous, leguminous, and miscel-
laneous, contribute to the produce without (plot 2), than to
that with (plot 1) , the ammonia-salts.

Although, as is shown in Table XL VII. , the increase of pro-
duce due to the ammonia-salts was pretty constant, the actual
amount of produce per acre was nearly 40 per cent, less over the
last six years than over the first eight. Where the farmyard
manure was used alone, however, the decline was greater still,
being nearly 48 per cent.

The 2001b. of ammonia-salts annually applied are estimated
to supply about 411b. of nitrogen, equal to about 501b. of
ammonia. Yet the figures show that the increased yield of
crop represents an average of only about 27-| per cent, of the
nitrogen so supplied.

Of mineral matter there was also an increased amount
taken up under the influence of the ammonia-salts, though it
shows a steady falling off. Over each period there was more
lime, magnesia, phosphoric acid, and sulphuric acid, and very
much more chlorine, taken up with, than without, the
ammonia- salts.

As the increased amounts of lime and magnesia, of potash
and soda, and of phosphoric acid, must have had their source
in the previous supplies within the soil, or in the residue
from the farmyard manure, the action, so far, of the
ammonia- salts has been more rapidly to utilise, and, there-
fore, the more to exhaust, these otherwise dormant stores.
But it is probable that, of both the potash and the phos-
phoric acid supplied in the dung, part remains unliberated
from its original condition of combination in the manure, and
part becomes so locked up (or distributed) within the soil as
to be only very slowly available.

Upon the whole, the evidence goes to show that the effect
of the ammonia-salts was to reduce the complexity of the
herbage, to render it more gramineous, to increase the
amount of produce, and, with this, to draw more upon the
mineral stores within the soil. It is also clear that, although

206 The Bothamsted Grass Experiments.

the application of the ammonia-salts was the means of turning
to account some of the accumulated residue of the mineral
constituents supplied in the dung, the limit of the imme-
diately available supply was very soon reached, the remainder
becoming less and less rapidly recoverable. It was, in fact,
retained in a condition so slowly available as to be of but
little effect in increasing immediate crops, and, therefore, of
but little practical value, except as a storehouse against
exhaustion.

Two points only now remain for consideration, namely, the
character of the second crops, and the influence exercised by
the nature of the season upon the produce of hay. Both of
these are matters of high practical interest.

THE SECOND CEOPS.

So far the produce of the first crops only, of each year,
has been taken into account. The estimated amounts of
second 'crop, on each plot, in each season, are collected in
an elaborate table in an appendix to the paper published in
the Philosophical Transactions. But the general character
and bearing of the results are sufficiently brought to view
in an abstract table, too extensive to be reproduced here,
in which are recorded the figures for a selection of the
plots only, representing very characteristically different con-
ditions of manuring. The plots for which the results are so
given are :

Plot 3. Unmanured every year.

Plot 7. Mixed mineral manure (including potash), without
nitrogenous manure, every year.

Plot 9. Mixed mineral manure (including potash), and
4001b. ammonia- salts, every year.

Plot 11. Mixed mineral manure (including potash), and
SOOlb. ammonia-salts.

Plot 14. Mixed mineral manure (including potash), and
nitrate of soda containing the same amount of nitrogen as
the ammonia-salts on plot 9.

Proportion of Second Crop to First Crop. 207

The particulars given are : The actual amounts of produce
of the first crops, the estimated amounts of the second crops,
and the proportion of the second to the first reckoned as 100 ;
and at the bottom of the table are stated the averages of the
first crops, of the estimated second crops, and the percentage
of the second to the first, for the first eight years of the
twenty, in every one of which the second crops were fed off by
sheep, and their quantities estimated, and for eight subse-
quent (though not consecutive) years, in seven of which the
second crops were fed and estimated, and in the eighth (the
last of the twenty), in which the second crops were cut,
removed, and weighed. There are also given the particulars
of the produce for the same plots for the three years subse-
quent to the first twenty (1876-78). The salient features
presented in this numerical summary may now be noticed.

Although, without manure, the amounts of produce, of
both first and second crops, are small, the proportion of second
crop to first is greater than under either of the selected
manurial conditions ; that is, it is greater where the total
removed from the land is comparatively small, and where,
especially, the variety of the herbage is the greatest, and
where, consequently, the possession by the roots of the upper
layers of the soil, and the capabilities of food-collection
generally, will be the most varied.

Next in proportion of second crops to first comes the
mineral-manured plot (7). Here again, the crops, though
much larger than without manure, are not really large ; but,
as without manure, the herbage is complex, and the command
by the roots, especially of the upper layers of the soil, will be
very varied.

On plot 9, the first crops average about one-and-a-half
times as much as with the mineral manure alone, but the
estimated average of the second crops is very nearly the
same in the two cases. Thus, with the much more luxuriant
growth of first crops under the influence of the ammonia-
salts, and the much more simple and almost exclusively

208 The Rothamsted Grass Experiments.

gramineous herbage, the actual quantity of the second crop
is small, and its proportion to the first little more than half
as much as without manure, and only about two-thirds as
much as with the mineral manure alone.

On plot 14 the first crops averaged more still ; they, also,
consisted almost exclusively of free-growing (though chiefly
other) grasses, and they comprised but few species. With
these characters, the second crops averaged even rather less
than with the ammonia- salts, and bore a smaller proportion
to the first.

In the case of plot 11, with double the amount of
ammonia-salts of plot 9, there were upon the whole still
larger first crops, and almost exclusively gramineous herbage,
which contained, however, a very abnormally high percentage
of nitrogen, and, with the obvious excess available, there is
here more second crop, and a higher proportion of second
crop to first, than with the smaller amount of nitrogenous
manure. There is, moreover, a tendency to a greater amount,
and proportion, of second crop in the later years.

The general result is that, when (with mineral manure)
active nitrogenous manures are used, but not in excessive
amount, the increase of the first crop will, in favourable
seasons, be such as to leave comparatively little available
nitrogenous residue for the second crop ; whilst, the produce
under such circumstances being characteristically gramineous,
and comprising comparatively few species, the condition of
the herbage is not very favourable for subsequent growth.
The percentage of both mineral matter and nitrogen is
generally, however, much higher in the dry substance of the
second and less matured produce, than in that of the first
and more matured, so that the removal of the second crops
is a considerably greater drain upon the resources of the soil
than might be suspected from the comparatively small
amounts of the produce. It is evident, too, that the actual
and relative amounts of second crop depend not only on the
balance of available coustituents remaining within the soil,

Influence of Season on Crop. 209

and on the climatal conditions, but also on the variety, and
the unexhausted condition, of the plants themselves which
are comprised in the mixed herbage.

INFLUENCE OF SEASON ON THE PEODUCE
OF HAY.

The object of this discussion is to endeavour to trace the
connection between certain measurable characters of season
on the one hand, and the luxuriance or sluggishness of growth
of the mixed herbage on the other, with comparatively little
reference to the effects of the different manures.

Common observation recognises a general connection
between the characters of the weather as to moisture, heat,
and light, and the luxuriance or scantiness of vegetation.
When, however, the amounts of growth in different seasons
are compared with the usual meteorological records of the
period, it at once becomes apparent how complicated is the
connection, and how inadequate are such records for a fuller
explanation of the differences of result obtained in different
seasons. And, however difficult and intricate the subject
may be when the growth of a crop consisting of a single
species only is concerned, it is far more so when the problem
has to deal with the relations of the various climatal condi-
tions to the development of a great variety of species growing
together, as in the case of the mixed herbage, and numbering,
as they do, from less than twenty to more than fifty, accord-
ing to the varied manurial conditions provided.

The method adopted is to draw attention to the actual and
the comparative characters of season under which some of
the largest, and some of the smallest, amounts of produce
have been grown. In a series of tables are recorded the actual
amounts of produce of hay per acre (first crops) on the same
five selected plots as were taken (page 206) to illustrate the
second crops, the monthly rainfall in inches (at Eothamsted),

210 The Rothamsted Grass Experiments.

the number of days in each month when the rainfall exceeded
O'Olin., together with the monthly mean iriaximum tempera-
ture, the monthly mean minimum temperature, the monthly
mean temperature, and the monthly mean range of tempera-
ture, all at Greenwich. Further, in the respective tables,
there are given, for each season, the total rainfall, the total
number of days on which O'Olin. or more fell, and the means
for each item of temperature : —

1. For the total 12 months ; July to June inclusive ;

2. For 4 months — July to October ;

3. For 4 months — November to February ;

4. For 4 months — March to June :

5. For 3 months — April to June ;

6. For 2 months — April and May.

With each of the five very different conditions of manuring
(on plots 3, 7, 9, 11, 14) the year 1869 gave the highest amount
of produce. On the other hand, 1870 gave, with two out of
the five conditions of manuring, the lowest produce, with two
others the lowest but one ; but the remaining or nitrated
plot was an exception, giving, in this year of drought, 1870,
an abnormally high produce for the season — a fact which was
considered when the general results on plot 14 were under
discussion. There can be no hesitation, therefore, in taking
the season of 1869 as that of the highest productiveness of
the twenty, and that of 1870 (with the exception mentioned)
as one of the lowest productiveness ; the next in order in this
latter respect being 1874. The produce of these two most
contrasted seasons is set forth in Table LIV., page 211, where
also a comparison is made with the average of the twenty
years, and the deficiency in 1870 compared with 1869.

A very full table, not given here, shows some of the
meteorological conditions under which the two very different
crops were grown.

The lowest line in Table LIV. shows that (excluding the
nitrate plot 14) there was an average yield per acre of nearly
40001b. (about If tons) more hay in 1869 than in 1870;

Characteristics of a Good Season.

211

and without manure, and with purely mineral manure,
the excess amounted to more than the average produce of
those plots.

TABLE LIV.— PRODUCE OF HAT PER ACRE ON SELECTED PLOTS.
AVERAGE OP THE 20 YEARS ; PRODUCE OF 1869, THE YEAE OF
HIGHEST PRODUCTIVENESS; PRODUCE OF 1870, THE YEAR OF LOWEST
PRODUCTIVENESS ; DIFFERENCE OF EACH FROM THE AVERAGE ; AND
DIFFERENCE OF THE ONE FROM THE OTHER.

13 GO

•g«5

£

1*3

§1

gl

d

•I

£ 1!

gj

c^

CO g

«•§!

S s'^

"*>rt "S

CO

og
S|

fc|

||I

|||

III

d

1

II

II

*i

Average, 20yrs., 1856-75

2383

lb.
3958

lb.
5711

lb.
6726

lb.
6407

lb.
5037

1869

4256

6124

7700

8610

8526

7043

1870

644

1968

3306

5150

6300

3474

< 1869 i+ 1873

+ 2166

+ 1989

+ 1884

+ 2119

+ 2006

+ or— average ^ -.g^Q , -1730

-1990

-2405

-1576

- 107

-1563

1870 less than 1869 | - 3612

-4156

-4394

-3460 -2226

-3569

The character of the weather of 1868-9 is thus summarised :
After five months of unusually high temperature (May to
September, 1868), and unusual drought during the first three
of them, the two following months (October and November,
18o'8) were again dry, but cold. The three winter months
were very warm, and all more or less, but December especially,
very wet. The result was an unusual winter growth of grass.
The dry and cold weather of March, however, checked vege-
tation ; but, with its early start and marked progress in the
winter, it recovered rapidly under the influence of the very
warm and sufficiently wet weather of April. The two
remaining months of the grass season were, however,
unusually cold, May being at the same time very wet, but
June dry, a condition which was compensated by the previous
abundance of moisture ; whilst, although the ruling temper-
atures were low for those months, they were yet actually

p 2

212 The Rothamsted Grass Experiments.

sufficient for active vegetation. The result was luxuriant and
succulent, though not maturing, growth. Indeed, according
to notes taken on the ground, the plots generally manifested
great luxuriance ; but the most prominent plants, whether
gramineous, leguminous, or miscellaneous, were much more
backward than usual at the time of cutting.

Thus, the excessive produce of grass in 1869 was due not
so much to the climatic conditions during the limited period
of really active accumulation and above-ground growth, as to
the preceding very favourable, instead of as usual detrimental,
conditions throughout the three winter months, thus bringing
the herbage unusually forward, and rendering it more capable
of turning to the best account such climatic elements of
growth as followed. This result is somewhat analogous to
that observed in the case of the Rothamsted experiments on
the continuous growth of wheat ; the seasons of extraordinary
productiveness of that crop having been marked rather by
moderately favourable conditions throughout, than by spe-
cially favourable ones during the period of most active above-
ground growth.

The conditions which prevailed in 1870, the year of least
productiveness, must now be examined. After the enormous
first crops of 1 869 less than average second crops were grown.
Not only would there be comparative exhaustion of manurial
constituents, but, succeeding upon the dry weather of June,
and the cold weather of both May and June, there was a
considerable deficiency of rain in July and August, but little
more than the average in September, and again a deficiency
in October ; and, with the continued lack of rain in July and
August, July was warmer, but August for the most part
unseasonably colder, than usual ; whilst September, with its
fair amount of rain, was generally warmer, and October, with
its defect, at times much colder than the average.

The autumn conditions were therefore, upon the whole,
adverse to growth. Over the five months — November, 1869,
to March, 1870, inclusive — the rain gauge indicated more

Characteristics of a Sad Season. 213

than the average total fall, though there was a considerable
deficiency in January. There were heavy and continuous
falls of rain in November and December, with great fluctua-
tions of temperature, some very warm, and some very cold
weather, and numerous gales. The first three months of
1870 again were marked by frequent alternations of warm
and very cold weather, the colder periods prevailing, and
being sometimes very severe. Snow was frequent, but the
total fall was deficient in January, and but little above the
average in February and March. Vegetation generally was
very backward, and grass land was very brown and bare.
April, May, and June, the three months of most active
accumulation and growth, were largely deficient in rain, and
with the exception of about a fortnight at the end of April
and the beginning of May, when the weather was cold and
cloudy, the whole period was unusually warm and sunny, the
three months together being not only much warmer than the
average, but very unusually deficient in rain. The day
temperatures especially were high, though the night tempera-
tures were, in April and May, low ; and throughout the three
months the degree of humidity of the atmosphere was
considerably below the average.

Thus, after an autumn very deficient in rain, and fluctuating
as to temperature, a winter and early spring very stormy, very
fluctuating as to temperature, in fact very inclement, and
vegetation consequently very backward to start with, the
three months of most active above-ground growth were very
unusually dry, very unusually hot in the days, and frequently
colder than the average in the nights. It was under these
conditions that the smallest crop of the twenty years was
obtained.

Upon the whole then, the registered meteorological condi-
tions of the season of least productiveness more obviously
account for the deficient crop, than do those of the previous
season account for its excessive yield. In the case of the
crop of 1870, the conditions previous to the period of active

214

The Rothamsted Grass Experiments.

growth were strikingly unfavourable for the herbage, and
the period of active growth was itself strikingly adverse,
both in its extreme dry ness, and in its coincident high day
and low night temperatures. In the case of the excessive,
but succulent and immature, growth of 1869, the climatic
conditions previous to the period of most active vegetation
were obviously very favourable ; but those of the period of
active above-ground growth itself were such as would only
conduce to great luxuriance provided there were an already
forward condition of the herbage.

The years 1868 and 1874 are also noticed, the former
being perhaps the second in order of productiveness
and the latter the second in unproductiveness. Table LV.
summarises the results in these two years, and may be
usefully compared with Table LIV., page 211.

TABLF. LV.- -PRODUCE OF HAT PER ACRE ON SELECTED PLOTS.
AVERAGE OF THE 20 YEARS ; PRODUCE OF 1868, THE TEAR SECOND IN
ORDER OF PRODUCTIVENESS ; PRODUCE OF 1874, THE TEAR SECOND IN
ORDER OF UNPRODUCTIVENESS ; DIFFERENCE OF EACH FROM THE
AVERAGE; AND DIFFERENCE OF THE ONE FROM THE OTHER.

Plot 3.

Plot 7.

Plot 9.

Plot 11.

Plot 14.

Means.

Ib.

Ib.

Ib.

Ib.

Ib.

Ib.

Average, 20yrs., 1856-75

2383

3958

5711

6726

6407

5037

1868

1960

4264 6622

7616

7728

5638

1874

1412

3088 3290

3540

5484

3363

( 1868

- 423

+ 306 + 911

+ 890

+ 1321

+ 601

— average | jgy^

- 971

- 870 -2421

-3186

- 923

-1674

1874 less than 1868

- 548

-1176 -3332

-4076

-2244

-2275

Another table gives a very full abstract of the meteoro-
logical conditions of the two years under notice, and, as
before, the relation of the season to the yield of crop is
discussed, but the briefest summary must suffice here.

The season second in order of productiveness was charac-
terised by unusually high temperatures throughout the whole
period of growth, with a sufficiency of rain up to the end of
April, conditions which brought the herbage very early

Good and Bad Seasons. 215

forward, and rendered it comparatively independent of the
extreme heat and drought of the months of May and June,
which would otherwise have been fatal at that period, and
which were, in fact, very injurious where the conditions of
manuring had not been such as to bring the vegetation
sufficiently forward previously.

It is a curious result that the year of the lowest produce of
the twenty, 1870, was the one of the most extreme heat and
drought of the series, and that the year of the highest pro-
duce but one, 1868, was only second to 1870 in heat and
drought of the growing period. But there was this
difference : the winter and early spring of 1870 had been
very adverse, the herbage was in a very backward state, and
the heat and drought commenced a month earlier ; whereas,
from the commencement of 1868, for a period of nearly four
months, the conditions both as to heat and moisture were
favourable, and it was not until May that the heat and
drought set in, then serving to elaborate and mature, rather
than materially to check, vegetation in such condition of
luxuriance and forwardness.

With regard to the season second in order of unproduc-
tiveness, it appears that, although the winter and early
spring of 1873-4 were, upon the whole, considerably warmer
than usual, there were periods of some severity as to
temperature, and the whole period was very deficient in rain ;
so that, instead of the warmth of the usually cold months
availing to bring vegetation forward, it remained very back-
ward when it was overtaken by the unfavourably cold days,
and the very unusually severe frosty nights, of the greater
part of May, whilst the short period of warmer weather
which then set in was unaccompanied by sufficient moisture,
and the herbage was already too much damaged to recover.
According to notes taken on the ground at the time, the
foliage of the grasses became spotted, and the earlier flower-
ing stems were bleached, and in many cases killed. The
greatest damage was done on the plots highly manured with

216 The Rothamsted Grass Experiments.

ammonia-salts, where Dactylis glomerata, which was both
abundant and forward, suffered very much.

The four selected seasons most strikingly illustrate the
intricacy and difficulty of the attempt to trace the relation
between the amount of growth of the mixed herbage of grass
land, and such meteorological conditions as are sufficiently
recorded. Very dissimilar climatal conditions characterised
the two seasons of highest productiveness, and again very
dissimilar ones those of lowest productiveness. The character
of the produce was also very different in the two cases of
the largest crops, and again very different in the two of the
smallest crops.

Taking a comprehensive view of the four seasons, it is
apparent tTiat in both seasons of high productiveness the
period prior to that of most active above-ground growth had
brought the herbage into an unusual state of forwardness ;
when, in the one, abundance of rain, with, upon the whole,
low temperatures, gave great luxuriance, but comparatively
leafy, succulent, and immature produce ; whilst, in the other,
the luxuriant early growth was followed by both unusual
drought and unusual heat, yielding quantity by virtue of
high development and maturation, as distinguished from
succulence and immaturity. As in both the cases of high
productiveness the period antecedent to that of most active
growth had been favourable, so in both those of very defec-
tive growth they had been very unfavourable. The winter
and early spring of 1870 had not, upon the whole, been
deficient in rain, but the period had been extremely variable
as to temperature, frequently very inclement, and, upon the
whole, colder than usual. The herbage was, from these
causes, very backward at the commencement of the active
growing period. April, May, and June followed, with a
great deficiency of rain, very high day and low night
temperatures, yielding very stunted and prematurely ripened
produce. The winter and early spring of 1874 had, on the
contrary, been very unusually deficient in rain, whilst the

Character of Herbage dependent on Season. 217

temperatures had ruled higher than the average, both day
and night, especially the latter, and the already backward
herbage was very materially damaged, yielding not only
checked and stunted, but reallv damaged crops.

It has already been shown how greatly both the botanical
and the chemical composition of mixed herbage vary accord-
ing to the description of manure applied ; and the foregoing
typical illustrations of the effects of the varying climatic
conditions of different seasons clearly indicate how different,
both botanically and chemically, will be the character
of the produce dependent on the character of the sea-
sons. In fact, a given quantity of gross produce of the
mixed herbage may be one thing in one season and quite
another in another season, both as to the proportion of the
different species composing it and as to their condition of
development and maturity.

218 The Rothamsted Grass Experiments.

IX. - BOTANICAL RESULTS OF EXPERIMENTS
ON THE MIXED HERBAGE OF PERMANENT
MEADOW, (a)

THERE is at Rothamsted nothing which will more impress
the visitor than the seven acres of meadow land in the Park,
the many years' experiments upon which with different
manures constitute the subject of this memoir. The twenty
parallel plots into which this area is divided appeal at once
and forcibly to the eye by the obvious differences in their
herbage. A plot here with rich green grasses waving
luxuriantly upon it ; another, on which the yellow meadow
vetchling apparently constitutes the leading feature ; a third,
irregular, patchy, and much afflicted with sorrel ; and yet
another, on which, at the time of my visit (August), the
white-flowered umbels of the earth-nut put everything else
in the shade—these and the like appearances convince with
an eloquence which the pen is powerless to imitate.

The land in Rothamsted Park has probably been laid down
with grass for some centuries. No fresh seed has been arti-
ficially sown within the last fifty years certainly, nor is there
record of any having been sown since the grass was first laid
down. The experiments commenced in 1856, at which time
the herbage appeared to be of uniform character. With

(a) " Agricultural, Botanical, and Chemical Eesults of Experiments on
the Mixed Herbage of Permanent Meadow, conducted for more than
twenty years in succession on the same land." Part ii., the Botanical
Eesults. By Sir J. B. Lawes, Bart., F.E.S., Dr J. H. Gilbert, F.E.S.,
and Dr. M. T. Masters, F.E.S. Phil. Trans., part iv., 1882. Pp. about
250. — (This notice is taken from an article which I contributed to Nature,
vol. XXIX., page 81.— W. F.)

Botanical Analysis of the Herbage. 219

few exceptions the same description of manure has been
applied year after year to the same plot ; and two plots, the
third and twelfth, have been continuously unmanured. For
the first nineteen years the first crop only was cut and carried
away, and the second crop was usually fed off by sheep who
were receiving at the time no other food. Of recent years it
has been more and more the practice to make the second
crop also into hay, and it is intended to adhere to this plan
in future, weather permitting.

The produce of every plot is weighed as hay, and the
result calculated per acre. Taking the average of the first
twenty years, the unmanured plots, 3 and 12, gave the
lowest yields of all, 21J and 24cwt. respectively. Next above
these is plot 5, manured with ammonia-salts at the rate of
4001b. per acre per annum, the yield giving an annual
average of 26^ cwt. per acre. The highest average recorded,
62f cwt. per acre, resulted from a mixed manure, containing
5001b. sulphate of potash, lOOlb. sulphate of soda, lOOlb.
sulphate of magnesia, 3|cwt. superphosphate of lime, 6001b.
ammonia-salts, and 4001b. silicate of soda, — a tremendous
dressing, by the way. The average yields on the rest of the
plots, each one of which received different manurial treatment
from that of the others, range themselves between these
extremes.

But the mere quantitative estimation of the yields per
acre was a comparatively simple task to that of making a
qualitative botanical examination of each crop. The proxi-
mate analysis was into the three classes of gramineous
herbage, leguminous herbage, and miscellaneous herbage,
the last-mentioned containing all plants not referable to the
Graminese or the Leguminosae ; and even this task would not
be a very difficult one. But when it is stated that in certain
seasons a complete botanical analysis was made, whereby
each species of plant was separated from all the others, then
the irksomenesss of the work will be appreciated. For the
details of these analyses I must refer to the memoir itself,

220 The Eoihamstecl Grass Experiments.

but the following is worth reproducing. "To quote an
extreme case in illustration of the difference in the character
of the herbage, and of the difference in the degree of
difficulty of separation accordingly, it may be mentioned that
whilst a sample of 201b. from one plot in 1872 only occupied
from four to five days in botanical analysis, a sample of equal
weight from another plot n* the same year occupied thirty

The total number of different species of plants that have
been detected on the plots is 89; of these, 20 are gramineous,
10 are leguminous, and the remaining fifty-nine belong to
miscellaneous orders. The 89 species comprise 59 dicoty-
ledons, 26 monocotyledons, and 4 cryptogams, three of which
are mosses (Hypnum) ; they are arranged under 63 genera
and 22 orders. Of the miscellaneous plants there are 13
species of Compositse, 6 of Rosacese, 5 each of Ranunculaceae
and Umbelliferse, 3 each of Labiatae, Polygonacese, Liliacese,
Caryophyllese, Scrophulariacese, and Musci, 2 each of
Rubiacese and Plantaginese, and 1 each of Cruciferse, Hype-
ricinese, Dipsacese, Primulacese, Orchidacese, Juncacese,
Cyperaeese, and Filices. Six genera only were represented
by more than one species ; these were Eanunculus, 5 species,
Rumex 3, and Potentilla, Gralium, Leontodon, and Veronica,
2 each. The 20 species of grass comprise 14 genera; Festuca
is represented by 4 species, Avena by 3, Poa by 2, and
Anthoxanthum, Alopecurus, Phleum, Agrostis, Aira, Holcus,
Briza, Dactylis, Cynosurus, Bromus, and Lolium by 1 each.
The fact that the four genera whose names are here italicised
were only represented by one species each serves to indicate
somewhat the nature of the land. Had it been wet or
marshy in parts, Alopecurus geniculatus might have been
looked for as well as Alopecurus pratensis. Had not the
plots been quite away from hedgerows, several species of
Bromus might have accompanied Bromus mollis, whilst
Brachypodium sylvaticum might also have been looked for.
The total absence of Glyceria further shows the fairly dry

The Flora of Eoihamsted Park. 221

character of the soil. Lastly, the 10 species of Leguminosse
fall under 5 genera — of Trifoliuui 4 species, Lotus and
Vicia 2 each, Lathyrus and Ononis 1 each.

Ten species of grasses occur on all the plots : Anthoxanthum
odoratum, Alopecurus pratensis, Agrostis vulgaris, Holcus
lanatus, Avena flavescens, Poa pratensis, Poa trivialis,
Dactylis glomerata, Festuca ovina, and Lolium perenne.
Festuca elatior was only found in one plot, and Festuca
loliacea in two. Phleuni pratense occurred in about one-
fourth the number of plots, Aira caespitosa in about one-half,
Briza media, Cynosurus cristatus, Festuca pratensis, and
Bromus mollis in sixteen or seventeen. No leguminous plant
occurred in all the plots, but Lathyrus pratensis was found
in nineteen plots, Trifolium repens and Trifolium pratense in
seventeen, Lotus corniculatus in sixteen, and Trifolium minus,
Trifolium procumbens, Lotus major, Ononis arvensis, Vicia
sepium, and Vicia Cracca only in one each.

These details will serve to indicate the nature of the flora
of the plots. Certain miscellaneous plants common on many
old pastures in this country are conspicuous by their absence.
The dry and level character of the meadow will account for
the absence of Caltha and Juncus. No species of Geranium
is recorded. But the most noteworthy fact appears to be the
absence of certain scrophulariaceous genera, which are by no
means uncommon on old grass lands, namely, Bartsia,
Euphrasia, and Ehinanthus. The quality of the land is
probably too good for the first two, and the application of
manure would certainly be against Euphrasia, but Ehinan-
thus Crista-galli (yellow rattle) is very common on old
meadows, as, for example, in Derbyshire and Worcester-
shire.

The object which the authors kept in view in writing this
section of their report was, in their own words, "to show
both the normal botanical composition of the herbage, aiid
the changes induced by the application of the different
manuring agents, and by variation in the cliinatal conditions

222 The Rofhamsted Grass Experiments.

of the different seasons ; and, as far as may be, to ascertain
what are the special characters of growth above ground or
under ground, normal or induced, by virtue of which the
various species have dominated, or have been dominated over,
in the struggle which has ensued." At the outset it was
noticed that those manures which are most effective with
cereals grown on arable land, were also most active in
increasing the quantity of grass amongst the herbage, and
that the manures which are most beneficial to beans or clover
produced the greatest proportion of leguminous herbage.
Thus, the highest gramineous produce resulted from a highly
nitrogenous manure, such as ammonia-salts or nitrate of
soda, with alkaline-salts, particularly potash ; but side by
side with the increase in the total gramineous herbage, there
was a decrease in the actual number of species of grass.
On the other hand, the highest percentage of leguminous
produce was the result of a mixed mineral manure with
potash. The percentage (by weight) results on the following
plots illustrate these points :

Plot 7.

Plots 3 and 12.

Plot 11.

GrramineaB

61-03

67-43

Qtr.n-i

Leguminosae . . .

i 23-06

8-20

O'Ol

Other Orders

15-91

24-37

4-08

100-00

100-00

100-00

Plot 7 was the most favourably manured for leguminous
produce, it received mixed mineral manure alone, including
potash ; plots 3 and 12 were the two unmanured ones ; plot
11 was the most favourably manured for gramineous produce,
it received SOOlb. ammonia-salts per acre, with mixed mineral
manure, including potash.

Special observations and complete botanical separations
made at intervals of five years to determine the influence of
seasonal variations show that " a given quantity of the pro-

The Struggle for Existence. 223

duce grown under the same conditions as to manuring might
be composed very differently in two different seasons."

The influence due to the special medium through which a
particular plant-food, such as nitrogen, is presented to the
plant, is aptly illustrated in the following words : " Because
a particular grass, or other plant, is little benefited by
ammonia-salts for instance, it does not follow that it will not
be favoured by nitrates ; nor, because if while growing in
association with other species it may not be specially benefited
by a particular manure, does it follow that it would not
derive advantage from the same substance when growing
separately."

Nearly all the plants on the plots are perennials, very few
are annuals, Bromis mollis being the only case amongst the
grasses. The advantage possessed by deep-rooting over
surface-rooting plants was well brought out in the droughty
season of 1870, when the latter suffered considerably from
lack of moisture. The locomotive power of underground
stems is of great use to some plants ; " the stock continues to
grow at one end, year after year, the opposite end gradually
dying away. In the course of a few years the plant therefore
occupies quite a different position from that which it a,t first
had." Notwithstanding the general rule that the chief effect
of nitrogenous manures is to favour the extension of foliage
and give it depth of colour, whilst that of mineral manures
is to encourage stem formation and the production of seed,
and notwithstanding that excessive nitrogenous manuring
prolongs the development of the vegetative organs till
perhaps the resources of the plant are exhausted or the
season is over, whilst excess of mineral manures may
induce premature ripening, yet so far as the experiments
have gone no absolute change in the distinctive form of any
plant has been effected by the prolonged use of the different
manures, though changes of degree are sometimes very
marked, as in the tufts of Dactylis glomerata.

The battle for life between the various species of plants

224 The Eothamsted Grass Experiments.

growing in the meadow is dependent much less on the
chemical composition of the soil than on its physical character,
its capacity for holding water, and its permeability to roots.
The immediate source of victory lies very generally in the
powerful root-growth of the survivors, the term " root " here
covering all kinds of underground stem. The various
influences affecting the struggle for existence amongst
meadow plants are discussed by the authors in a fascinating
manner, and this part of the memoir is of special value to the
botanical student.

Every plant occurring on the plots is dealt with individu-
ally, and in the case of each grass and leguminous plant and
of the more commonly occurring weeds, a table showing the
relative predominance is given. The fact that plants closely
allied morphologically may yet differ widely in their physio-
logical endowments is strikingly illustrated by the two species
Poa trivialis and Poa pratensis. These two plants, sprung at
some former period from a common ancestor — for this, we
presume, is the morphological significance of their being
placed in the same genus — differ only in the most trivial
points : Poa pratensis (smooth-stalked meadow grass) is
smooth, stoloniferous, and has a blunt ligule ; Poa trivialis
(rough- stalked meadow grass) is rough, has no stolons, and
possesses a long pointed ligule. We read that " the stolon-
bearing Poa pratensis is especially benefited by nitrogenous
manure in the form of ammonia-salts (in combination with
mineral manure), but not at all by nitrate of soda, whereas
the more finely-rooted and non- stoloniferous Poa trivialis has
declined markedly on the ammonia plots, but has remained
very prominent on the nitrate plots, especially where the
larger amount of nitrate was used with the mixed mineral
manure." Thus in 1872, on plot 9 (mineral manure and
ammonia- salts), Poa pratensis gave 22'67 per cent of the total
produce, and Poa trivialis only 0'64 ; on plot 14 (mineral
manure and nitrate of soda) Poa trivialis gave 2476, and
Poa pratensis only 2'57 per cent. It is suggested that the

Conclusion. 225

relatively shallow-rooting Poa trivialis predominates on the
nitrate plots by reason of its fine surface-roots arresting
and taking up the nitrate before it has had time to penetrate
too deeply ; this plant invariably makes rapid growth upon
the application of the nitrate of soda in the spring.

The remaining portion of the paper is devoted to a dis-
cussion of the botany of each separate plot in each season of
complete botanical separation, and is carried out with the
same elaborate detail as the earlier portion. No one can read
this memoir without being impressed with the great power,
too frequently overlooked, possessed by the subterranean
members of the plant body in deciding the struggle for
existence ; much of the internecine warfare is carried on in
the dark.

Such a splendid series of experiments on grass land has
never before been consummated, and the results deserve the
most careful study not only of the agriculturist, but of
the botanist, the chemist, and the evolutionist. It may
perhaps be long before the great lessons learnt in
Eothamsted Park have filtered down to those to whom they
should be of most practical value, but I do not despair of a
time coming when the intelligent manuring of grass lands
for very specific objects will form a part of ordinary
agricultural practice. Those who put their hands to the
plough in the field of agricultural research must be content
to trudge along, laboriously and unnoticed, in the furrow.
Their discoveries cannot be made in a week, or a month, as
are many in electricity or in chemistry, but, like those at
Rothamsted, where the investigations were commenced more
than half a century ago, they can only be looked, for, even
after the expenditure of much thought and of unflagging
industry and perseverance, as

"the long result of Time."

INDEX.

ACCUMULATION, 61, 109, 123, 124, 196
Achillea Millefolium, 138, 155, 169
Acreage of wheat (United Kingdom), 78, 83

barley, 131

Agricultural results of the grass experiments, 133
Agricultural Holdings Act, 52
Agrostis vulgaris, 137, 146, 152, 160, 161, 163, 166, 169, 170, 174,

177, 180, 183, 191, 204, 220, 221
Aira caespitosa, 137, 220, 221
Alopecurus pratensis, 137, 163, 166, 169, 170, 174, 177, 183, 220,

221
Ammonia, quantity required for given increase in wheat crop, 56

in barley crop, 103
Ammonia-salts, 8, 15, 16, 41, 43, 45, 60, 72, 91, 98, 145, 146, 159,

161, 164, 169, 172, 175, 178, 181, 189, 200, 204, 205, 219, 224
" Ammonia-salts " defined, 22, 139
Analytic method, 88
Annual grasses, 223

Anthoxanthum odoratum, 137, 160, 174, 204, 220, 221
Anthriscus sylvestris, 169
Arrhenatherum avenaceum (see Avena elatior)
Ash constituents, 89, 111, 184
Autumn manuring, 97, 98, 188, 197
Avena, 220

Avena elatior, 137, 166, 174, 177, 180
Avena flavescens, 137, 163, 170, 177, 183, 191, 221
Avena pubescens, 137, 163, 177, 180, 183, 191

BARLEY, composition of, 142, 188
cultivation of, 120, 132
Barley crop, nitrogen recovered in, 197
Battle for life, 223
Bean crop, nitrogen in, 189

228 Index.

Bedstraw, yellow (see Gali.mii)

Bent grass (see Agrostis)

Birdsfoot trefoil (see Lotus cornicnlatus)

Black knapweed (see Centaurea)

Botanical results of grass experiments, 218

Botanical analysis of hel'bage, 219

Briza media, 137, 220, 221

Bromusmollis, 137, 169, 170, 177, 191, 192, 204, 220, 221, 223

Bunium flexuosum (see Conopodium denudatum)

Burnet saxifrage (see Pinapinella)

Buttercup (see Ranunculus)

CANADA, soils of, 41

Carbon of the soil, 52

Carbon, source of, 37

Carbonaceous manures, 11, 15, 51, 102, 161

Carex preecox, 138

Caryophyllacese, 220

Centaurea nigra, 138

Cerastium triviale, 138

Chickweed, mouse-ear (see Cerastium)

China, yield of wheat in, 37

Chlorine, 142, 163, 166, 170, 183, 192, 205

recovered, and not recovered, in crops, 203
Clover (see Trifolium)
Clover crop, nitrogen in, 189
Cocksfoot (see Dactylis)
Cornpositse, 220
Composition of barley, 128, 142

crops, no direct guide to manures required, 188

hay, 142

manures, 32, 139

second grass crops, 208

wheat, 142

" Condition" of soils, 9, 13, 15, 28, 59, 88, 96, 98, 115
Conopodium denudatum, 138, 163, 169, 218
Consumption of wheat, 74, 80, 81, 83
Cow parsnip (see Heracleum)
Crowfoot (see Ranunculus)
Cruciferss, 220
Cryptogams, 220
Cultivation of barley, 120, 132

Index. 229

Cultivation of wheat, 132
Cynosurus cri status, 137, 220, 221
Cyperacese, 220

DACTYLIS glomerata, 137, 160, 161, 163, 166, 170, 174, 177, 180, 183,
204, 216, 220, 221

Dandelion (see Taraxacum)

Dicotyledons, 220

Dipsacese, 220

Dogstail (see Cyiiosurus)

Downy oat grass (see Arena pubescens)

Drainage waters, 40, 42, 47, 51, 58, 59, 60, 73, 99, 109, 119, 158, 161

188, 198, 199, 201, 203, 204
Drought, effect of, 223
Dutch clover (see Trifolium repeus)

EARTH nut (see Couopodium denudatum)

Exhaustion of soils, 92, 98, 113, 122, 124, 126, 132, 174, 206

FALSE oat grass (see Aveua elatior)

Farmyard manures, 8, 10, 49, 97, 123, 189, 193, 200

Fate of minerals in the soil, 202

nitrogen, 184, 194

Fertility of soils, 9, 36, 37, 38, 53, 59, 98
Festuca, 220

elatior, 137, 221

loliacea, 137, 221

ovina, 137, 146, 152, 160, 161, 163, 169, 170, 174, 177, 180,
183, 191, 204, 221

pratensis, 137, 221
Field woodrush (see Luzula)
Filices, 220
Foxtail grass (see Alopecurus)

G-ALITJM, 220

verum, 138

Germander speedwell (see Veronica)
Gramineous herbage, 135, 137, 145, 146, 151, 155, 157, 160, 163,

166, 169, 170, 172, 174, 178, 183, 187, 191, 192, 204, 208, 219,

220, 222

HAIR-GRASS (see Aira)
Hay, composition of, 142

230 Index.

" Heavy " soils, 6

Heracleum Sphondyliurn, 138

History, 1

Holcus lanatus, 137, 152, 160, 161, 163, 166, 169, 170, 174, 177,

180, 183, 192, 204, 220, 221
Holkham experiments on wheat, 6
Home produce of wheat, 83

barley, 130
Hypericinae, 220
Hypnum, 220

IMPORTS of wheat, 74, 79, 83

barley, 130
India, yield of wheat in, 37

JUNCACEJE, 220

KNAPWEED, black (see Centaurea)

LABIATE, 220

Lathyrus pratensis, 138, 152, 155, 157, 170, 180, 183, 204, 218, 221

Legislation, 41

Leguminous herbage, 135, 138, 145, 146, 150, 151, 155, 157, 160,

163, 169, 170, 172, 177, 178, 180, 183, 187, 191, 192, 204, 219, 220,

222

Leontodon, 220

Lesser stitchwort (see Stellaria)
Liebig, 153

his theory, 184, 185, 187
Light soils, 6, 188
Liliaceae, 220
Lime in barley and wheat crops, 89, 142

in hay crop, 142, 145, 147, 155, 157, 161, 166, 167, 170, 171,

173, 183, 192, 205

recovered, and not recovered, in crops 202
Loam, 188

Locomotive power in plants, 223
Lolium perenne, 137, 169, 170, 177, 180 183 220 901
Lotus, 221

corniculatus, 138, 152, 155, 157, 221
major, 221

Index. 231

Luxuriance of crops, 94, 127
Luzula cainpestris, 138

MAGNESIA in wheat and barley crops, 89, 142

hay crop, 142, 147, 155, 157, 166, 170, 171, 173, 183,

192, 205

recovered, and not recovered, in crops, 202
Malt tax, 130

Manures, composition of, 32, 139
Manuring, autumn and spring, 97, 98, 188, 197, 225

rational principles of, 189
Maturation, 129, 153, 183, 208
Meadow clover (see Trifoliuin pratense)
fescue (see Festuca pratensis)
foxtail (see Alopecurus)
vetchling (see Lathyrus)
Milfoil (see Achillea)
Mineral manures, 7, 9, 15, 39, 41, 43, 53, 59, 91, 98, 111, 149, 153,

159, 161, 164, 167, 170, 175, 178, 187, 207, 211, 223
Minerals of the soil, fate of the, 202
Miscellaneous herbage, 135, 138, 145, 146, 151, 155, 160, 163, 169,

170, 177, 191, 192, 204, 219, 220, 222
Monocotyledons, 220
Morphological characters of plants, 224
Mosses, 220

Mouse-ear chickweed (see Cerastiuin)
" Mulching," effect of straw, 164

NITRATES, 38, 60, 201
Nitrate of potash, 181

soda, 45, 51, 98, 99, 101, 146, 167, 169, 170, 181, 200, 224
Nitrites, 201

Nitrification, 38, 40, 51, 59
Nitrogen, evolution of free, 201, 203

in ammonia-salts, 205

in bean crop, 189

in clover crop, 189

in different forms of combination, 51, 60, 110, 121, 147,
168, 181, 183, 200, 204, 223, 224

in farmyard manure, 193, 200, 204

in hay crop, 142, 165, 187, 191, 192

in wheat and barley crops, 89, 142, 189, 197

232 Index.

Nitrogen in wheat straw, 163
loss of, 40, 73, 201
of the soil, 34, 39, 44, 52, 101, 107, 113. 116, 124, 153, 184

194

organic, 38, 40, 59

recovered, and not recovered, in crops, 201. 203, 205
source of, 37, 41, 152, 180
Nitrogenous condition of soil, 96

manures, 14, 17, 98, 100, 150, 179, 189, 223

OAT grass (see Avena)
Oats, 108
Ononis, 221
Orchidacese, 220
Organic matter, 50, 97, 139
nitrogen, 38, 40, 59

PERENNIAL grasses, 223
Peruvian guano, 13, 16
Phleurn pratense, 137, 220, 221
Phosphoric acid in barley crop, 89, 142, 188

in drainage waters, 158

in hay crop, 142, 147, 153, 155, 157, 163, 166 170
171, 173, 181, 183, 192, 205

in soil, 43, 53, 59, 60, 126

iii wheat crop, 37, 89, 142, 188

recovered, and not recovered, in crops, 203
Physiological characters of plants, 224
Pirnpinella Saxifraga, 138
Plantaginese, 220
Plantago laiiceolata, 138
Plantain (see Plantago)
Poa, 220

Poa pratensis, 137, 161, 163, 169, 174, 177 180 *n 004
Poa tririalis, 137, 163, 169, 170, 183, 191, 192, 204 Q21 224
Polygouacege, 220

Population of United Kingdom, 79
Potash in barley crop, 89, 142
in drainage waters, 158

in soil, 43, 53, 59, 60, 126, 129

Index. 233

Potash iii wheat crop, 37, 89, 142

manures, 151, 155, 159, 161, 164, 167, 170, 171 175
recovered, aiid not recovered, in crops, 202

Potentilla, 220

Primulacese, 220

Purple clover (see Trifolium prateuse)

QUAKING grass (see Briza)
Quality of hay, 172

crop, 92, 94, 96
Quantity of crop, 92, 94, 96

RAINFALL, 63, 70, 94, 95, 210
RanunculacesB, 220
Ranunculus, 220

bulbosus, 138, 155
repeiis, 138, 155

Rape cake, 7, 8, 10, 11, 13, 16, 53, 102.. 126
Ribgrass (see Plantago)
Ribwort (see Plantago)
Rodmersham wheat experiments, 12
" Root-crops," 188
Root distribution in the soil, 149, 152, 180, 188, 207, 223, 225

growth, 224
Rosaceae, 220

Rotation, barley grown in, 114, 116
Rough cocksfoot (see Dactylis)
Rough-stalked meadow grass (see Poa trivialis)
Rubiaceae, 220
Rumex, 220

Acetosa, 138, 146, 155, 163, 169, 218
Rye grass (see Lolium)
Rye-leaved fescue (see Festuca loliacea)

SAWDUST, 154, 156, 178

Scrophulariacese, 220

Season, characters of a bad, 27, 67, 95, 127, 212, 215

characters of a good, 26, 65, 94, J27, 211, 214

favourable for wheat, 59

influence of, 23, 63, 209
Second grass crops, 206

234 Index.

Sedge, vernal (see Oarex)

Sheep's fescue (see Festuca oviua)

Silica in barley and wheat crops, 89, 129, 142

in hay "crop, 142, 147, 155, 164, 166, 170, 171, 173, 181, 184,

192

recovered, and not recovered, in crops, 203
Silicates, behaviour of, 167
Silicate of linie, 164

of soda, 104, 164

Smooth-stalked meadow grass (see Poa pratensis)
Soda, 129, 142, 155, 166, 170, 171, 173, 181, 183, 192, 205

recovered, and not recovered, in crops, 203
Soda-liine process, 200
Soft brome (see Bromus mollis)
Soil, 6? 44, 133, 188

distribution of nitrogen in, 194
inherent capabilities of, 52
Sorrel (see Rumex Acetosa)
Speedwell, germander (see Veronica)
Spring manuring, 97, 98, 188, 197, 225
Stellaria graminea, 138
Stitchwort, lesser (see Stellaria)
Straw as manure, 161, 164

strength of, 130
Struggle for existence, 145. 223
Sulphate of soda, 175

of potash, 181
Sulphuric acid in barley and wheat crops, 142

in hay crop, 142, 155, 157, 170, 171. 173, 181. Ib4

192, 205

recovered, and not recovered, in crops, 203
Superphosphate of lime, 98, 100, 139, 153, 172, 174, 181, 188
Sweet vernal (see Anthoxanthum)
Synthetic method, 88

TALL fescue (see Festuca elatior)
Taraxacum officinale, 138, 169
Temperature, 63, 70, 94, 95, 210
Timothy grass (see Phleum)
Trifoliuin, 221

minus, 221

pratense, 138, 152, 204, 221

Index. 235

Trifolium, procumbeus, 221

repens, 138, 183, 221
Tufted hair grass (see Aira)
Turnips, 114

UMBELLIFER^E, 220

Unexhausted manures, 30, 32, 45, 60, 106, 179, 190, 206, 208

source of fertility in, 34
United States, yield of wheat in, 37

soils of, 41

Unmamired produce, 9, 14, 24, 36, 96, 112, 116, 122, 136, 144, 204,
207, 211, 219

VARIATIONS in chemical composition of barley crop, 128
Vernal sedge (see Carex)
Veronica, 220

chamsedrys, 138

Vetchling, meadow (see Lathyrus)
Vicia, 221

WEATHER, factors of, 209

Weeds, 35, 36, 38

Wheat, composition of, 142. 188

crop, cause of lessened area, 62

nitrogen removed by, 197, 200

crops, remarkable, 63
White clover (see Trifolium repeus)
Woodrush, field (see Luzula)

YARROW (see Achillea)

Yellow bed straw (see Galium)

Yellow oat grass (see Avena flavescens

Yield of wheat, 75, 79

Yorkshire fog (see Holcus lanatus)

1888.
A

CATALOGUE OF BOOKS

RELATING TO

, §0&titt0, CrirM,

USEFUL FOB

COUNTRY GENTLEMEN,

TEAVELLEES, ETC.,

PUBLISHED BY

HORACE COX,

"THE FIELD" OFFICE, 346, STBAND, LONDON, W.C.

*#* Orders for any of the following works, with postage
stamps or post-office order (payable at the Money Order Office,
369, Strand) for the amount, should be sent to HOEACE Cox,
Publisher, at the above address, or they tnay be obtained by
order of any bookseller.

ItTOTIE.

T BEG to call the attention of Country Gentlemen,
Travellers, Sportsmen, Farmers, and others to
the works quoted in this Catalogue. They are
written by authors who are well known and acknow-
ledged authorities on their respective subjects.

The illustrations have been intrusted to competent
artists, and neither pains nor expense have been
spared to make the works as complete as possible.

HORACE COX,

Publisher.

CATALOGUE OF BOOKS

PUBLISHED BY

HORACE COX.

Second Edition, Greatly Enlarged.

Royal 4>to., bevelled boards, gilt edges, price 15s., by post 15s. 9d.

THEIR

NATURAL HISTORY AND PRACTICAL MANAGEMENT.

BY

W. B. TEGETMEIEK, F.Z.S.,

(Member of the British Ornithologists'' Union.)
AUTHOR OP "THE NATURAL HISTORY OF THE CRANES," Ac.

Illustrated with numerous foil-page engravings drawn from Life by T. W. WOOD.

CONTENTS.

The Natural History of the Pheasants, Habits, Food, Structure, &c. ; Manage-
ment in Preserves and in Confinement, with detailed descriptions of the Different
Species adapted for the Covert and the Aviary, and an account of their Diseases
and Transport.

NOTICES OF THE SECOND EDITION.

From Yarrell's " British Birds," fourth edition.—" For further details, as well
as for instructions as to the management of pheasants, both in the covert and the
aviary, and the disorders to which they are liable, the reader is referred to Mr.
Tegetmeier's excellent work, to which the editor is under great obligations."

From the Pall Mall Gazette. — " This new and sumptuous edition contains so
large an amount of fresh matter that it may be regarded, practically, as another
work. In its own way, it is quite exhaustive. Illustrated by admirable and life-
like full-page illustrations. On the technical details of rearing and preservation,
Mr. Tegetmeier will be found a thoroughly trustworthy and scientific guide."

From the Daily Telegraph. — " Both in its description and practical aspects, the
treatise is admirable."

" THE FIELD " OFFICE, 346^ STRAND, W.C.

4 A CATALOGUE OP BOOKS

Now ready, super-royal Svo., price £2 2s., by pout £2 3s.

YACHT ARCHITECTURE.

By DIXON KEMP,

Associate Institute Naval Architects (Member of the Council).

THIS WOKK enters into the whole subject of the laws which govern the resis-
tance of bodies moving in water, and the influence a wave form of body and
wave lines have upon such resistance. It also deals comprehensively with the
subject of STEAM PKOPULS1ON as applied to yachts. An easy SYSTEM for
DESIGNING is provided, and every necessary calculation is explained in detail.
The latter part of the work is devoted to YACHT BUILDING, and engravings are
given of every detail of construction and fitting, including laying off, taking bevels,
Ac. The List of Plates (exclusively of over thirty devoted to the elucidation of the
text, and nearly two hundred woodcuts) exceeds fifty, and comprises the LINES of
some of the most CELEBRATED YACHTS AFLOAT by the most successful
builders and designers.

SUMMARY OF CHAPTERS.

Chap.

I. — Displacement, Buoyancy, and Centre of Buoyancy explained.

II.— Proportions of Yachts and Tonnage Enles ; Eules for Freeboard, Depth, <fec.

III. — Stability as influenced by the Proportions, Form, Weight, and Ballasting

of Yachts ; their Centres of Gravity and Buoyancy. Profusely Illustrated.

IV.— The Motions of Yachts among Waves as influenced by their Forms and

Proportions.
V. — Lateral Resistance, and the adjustment of its centre in relation to the

Centre of Effort of the Sails.
VI.— Power to Carry Sail ; the Impulse of the Wind as a Propelling Force ;

Apportioning Sail for Speed ; Speed Formulae, <fec.
VII.— The Action of the Rudder and Steering Efficiency. Proportions of

Rudders, &c.

VIII. — Resistance of Vessels moving in Water. The Wave Line Theory; the
Stream Line Theory; the Wave Form Theory. Mr. Froude's Admiralty
Experiments, clearly defining the exact influence of Form on Speed
IX.— Resistance Experiments with Models.
X.— The Wave Form in Theory and Practice. The Forms of many celebrated

Yachts analysed.
XI.— Steam Yachting. The Boiler, Engine, Indicator, &c., practically explained

at great length.

XII.— Propulsion by Steam. The action of the screw propeller and all the
conditions which influence propulsion by steam explained, with
numerous practical examples.
XIII.— Rules and Formulae in use for determining the Displacement, Stability

and other qualities of a yacht fully explained.
XIV.— Working Examples for making all the necessary Calculations concerning

a Yacht, every sum being given in detail.
XV.— Yacht Designing: being a complete system for putting into effect the Art

of Designing Yachts by Scientific Methods.
XVI.— Laying Off, Making Moulds, Taking Bevels. &c.; Taking Off a Yacht's

Lines, &c.

XVII.- Yacht Building: Giving detailed examples for constructing yachts, with

numerous plates and engravings of the various parts ot the vessel,

including two large coloured lithographs on a half inch scale giving

YA7Trr 0 ^ctional views of a 40-tonner, complete with all her fittings, &c

XVIII.-Spars and Blocks. ^ Rules for fixing upon their lengths, girths, Ac. Sizes

XIX.— BaHa8t?ng! C°

AppENDix.-Comains^much information concerning small yachts, various useful

" THE FIELD " OFFICE, 346, STRAND, W.C.

PUBLISHED BY HORACE COX.

FIFTH EDITION. GREATLY ENLARGED.

Super-royal 8vo., ivith Seventy -jive full-page Diagrams, price 25s.,
by post 26s.

A MANUAL

OF

YACHT AND BOAT SAILING.

BY

DIXON KEMP, A.I.N.A.,

AUTHOR OF "YACHT DESIGNING."

(The Lords Commissioners of the Admiralty have ordered this work to be
supplied to the Libraries of the ships of the Boyal Navy.)

THIRD EDITION. Price 7s. 6rf., by post 7s. Wd.

FIGURE SKATING;

BEING

THE THEORY AND PRACTICE OP THE ART AS DEVELOPED
IN ENGLAND,

WITH

A GLANCE AT ITS ORIGIN AND HISTORY

BY H. C. VANDERVELL AND T. MAXWELL WITHAM

(Members of the London Skating Club).

There are thousands of skaters who attain a small amount of skill in Figure
Skating, and there stop, because they neither know what to do or how to do it. A
reference to this, the acknowledged Text Book on Figure Skating, will solve any
difficulty that may have stopped progress for years.

FOURTH EDITION. In post Svo., limp doth, gilt, price 2s. 6d., by post 2s. 8d.

THE ART OF SKATING;

WITH

ILLUSTRATIONS, DIAGRAMS, AND PLAIN DIRECTIONS FOR THE

ACQUIREMENT OF THE MOST DIFFICULT AND GRACEFUL

MOVEMENTS.

By GEORGE ANDERSON ("Cycles"),

Vice- President of the Crystal Palace Skating Club, and for many years President
of the Glasgow Skating Club.

" THE FIELD " OFFICE, 346, STRAND, W.C. 6

A CATALOGUE OF BOOKS

ILLUSTRATED WITH FULL-PAGE ENGRAVINGS DRAWN PRINCIPALLY FROM
LIFE BY HARRISON WEIR,

In Imperial 4-to., bevelled boards, gilt edges, price 18s., by post 18s. 9cL

THE

CATTLE OF GREAT BRITAIN:

BEING

A SERIES OF ARTICLES

ON TEE

VARIOUS BBEE OF CATTLE OF THE UNITED KINGDOM,
THEIR HISTORY, MANAGEMENT, &c.

EDITED BY J. COLEMAN,

Editor of the Farm Department of " The Field," and formerly Professor of Agriculture
at the Royal Agricultural College, Cirencester.

ILLUSTRATED WITH FULL-PACE ENGRAVINGS DRAWN PRINCIPALLY FROM
LIFE BY HARRISON WEIR.

In Imperial 4£o., bevelled boards, gilt edges, price ISs., by post IBs. 9d.

THE

SHEEP AND PIGS OF GREAT BRITAIN:

BEING

A SERIES OF ARTICLES

ON THE VARIOUS

BREEDS OF SHEEP AND PIGS OF THE UNITED KINGDOM,
THEIR HISTORY, MANAGEMENT, &c.

EDITED BY J. COLEMAN,

Editor of the Farm Department of " The Field," and formerly Professor of Agriculture
at the Royal Agricultural College, Cirencester.

" THE FIELD " OFFICE, 346, BTEAND, W.C.

PUBLISHED BY HORACE COX.

NEW AND CHEAPER EDITION OF THE CATTLE, SHEEP, AND PICS OF GREAT
BRITAIN.

Now ready, with Illustrations from the Original Drawings by Harrison Weir,
in 1 vol., price 12s. 6d, by post 13s.

The Battle, Sheep, and Pigs of Great Britain :

Being a Series of Articles on the Various Breeds of the United Kingdom,
their History, Management, &c.

Edited by JOHN COLEMAN,

Editor of the Farm Department of "The Field." and formerly Professor of Agriculture
at the Royal Agricultural College, Cirenoester.

CONTENTS.

THE CATTLE OF GREAT BRITAIN.

L— Introductory.

II.— Breeding and General Manage-
ment.

III.— Principles of Feeding, and
Value of Different Kinds of
Food.

IV.— Buildings, and the Manage-
ment of Manure.
V.— Dairy Management, the Milk

Trade, Ac.

VI.— Shorthorns. By John Thornton.
VII.— The Hereford Breed of Cattle.

By T. Duckham.
VIII.— Devon Breed of Cattle. By

Lieut.-Col. J. T. Davy.
IX.— The Longhorns. By Gilbert

Murray.
X.— The Sussex Breed of Cattle.

By A. Heasman.

XI.— Norfolk and Stiff oik Bed-Polled

Cattle. By Thomas Fulcher.

XII.— Galloway Cattle. By Gilbert

Murray.

XIII.— The Angus-Aberdeen Cattle.
XIV.— The Ayrshire Breed of Cattle.

By Gilbert Murray.
XV.— West Highland Cattle. By

John Eobertson.
XVI.— The Glamorgan Breed of Cattle.

By Morgan Evans.
XVII.— Pembrokeshire or Castlemartin

Cattle. By Morgan Evans.
XVIII.— The Anglesea Cattle. By Mor-
gan Evans.
XIX.— The Kerry Breed of Cattle. By

the late R. O. Pringle.
XX.— The Jersey Breed of Cattle. By

John M. Hall.

XXI.— The Guernsey Breed of Cattle.
By " A Native."

THE SHEEP OF GREAT BRITAIN.

I. — Introductory.

II.— The Management of Ewes up
to Lambing.

III.— Preparations for and Attention

during Lambing.
IV.— Management from Birth to

Weaning.

V.— From Weaning to Market.
VI.— On Wool.
VII.— Leicester Sheep.
VIII.— Border Leicesters. By John

Usher.

IX.— Cotswold Sheep.
X.— Long-Woolled Lincoln Sheep.
XL— The Devon Long- Wools. By

Joseph Darby.
XII.— Bomney Marsh Sheep.
XIII.— Southdown Sheep.
XIV.— The Hampshire or West
Country Down Sheep. By
E. P. Squarey.
XV.— Shropshire Sheep.
XVI.— Oxfordshire Down Sheep. By
Messrs. A. F. M. Druce and
C. Hobbs.
XVII.— The Boscommon Sheep. By

the late B. 0. Pringle.
XVIII.— Negrette Merino Sheep.
XIX.— Exmoor Sheep.
XX.— The Black-faced or Scotch

Mountain Sheep.

XXL— Cheviot Sheep. By John Usher.
XXII.— Dorset Horned Sheep. By

Joseph Darby.
XXIII.— Welsh Mountain Sheep. By

Morgan Evans.
XXIV.— The Badnor Sheep. By Morgan

Evans.

XXV.— Herdwick Sheep. By H. A.
Spedding.

THE PIGS OF GREAT BRITAIN.

I. — Introductory.
II.— The Berkshire Pig.
III.— Black Suffolk Pigs.
IV.— Large White Breed of Pigs.
V.— Small White Pigs.
VI.— Middle Bred White Pigs.
VII.— The Black Dorset Pig.
VIII.— The Tamworth Pig.

•'THE FIELD" OFFICE. 346,: STRAND, w.€.

A CATALOGUE OF BOOKS

FIFTH EDITION.

Re-written, with additions and new full-page Engravings. In one
volume, levelled boards, gilt edges, price 15s., ly post 15s. 9<Z.

THE

DOCS OF THE BRITISH ISLANDS:

BEING

A SERIES OF ARTICLES

ON

THE POINTS OP THEIR VARIOUS BREEDS,

THE TEEATMENT OP THE DISEASES TO WHICH
THEY AEE SUBJECT.

REPRINTED FROM "THE FIELD" NEWSPAPER.

BY J. H. WALSH,

" STONBHKNQE," EDITOR OF " THE FIELD."
(WITH THE AID OF SEVERAL EXPERIENCED BREEDERS.)

CONTENTS.
General Management.

Book

I.— Management of Dogs in Health.
II. — Drugs Commonly Used for the
Diseases of Dogs, and their Modes
of Administration.

III.— The Ordinary Diseases of the Dog

and their Treatment.
IV.— Judging at Dog Shows and Field

Trials.

Book
I.— Dogs Used with the Gun.

Sporting Dogs.

| ff.— Hounds and their Allies.

Non-Sporting Dogs.

I.— Watch Dogs.
II.— Sheep and Cattle Dogs.

I Book
III.— Terriers (other than Fox and Toy).*
IV.— Toy Dogs.

THE FIELD" OFFICE, 346, STRAND, w.c.

PUBLISHED BY HORACE COX.

THIS WORK CONTAINS ABOVE 700 PAGES, AND NEARLY 400
ILLUSTRATIONS.

New and cheaper Edition, with additions, price 25s,, by post 26s.
cloth gilt.

Shifts and Expedients

OF

CAMP LIFE, TRAVEL, and EXPLORATION,

W. B. LOKD AND T. BAINES.

(Royal Artillery.) (F.R.G.S.)

CONTENTS.

INTRODUCTION.

Chap.

I._0utflt to take abroad.
II.— Boats, Eaf ts,and Makeshift Floats
III.— Working in Metal.
IV.— Huts and Houses.

XV.— Fish and Amphibious Annuals.
XVI.— Poisoned Weapons, Arrows,

Spears, <fec.

XVII.— Tracking, Hunting, and Trap-
sing.

V.— Extempore Bridges and Make- ! XVIII.— Palanquins, Stretchers, Ambu-
shifts for Crossing Rivers and lances, &c.

Ravines. XIX.— On Sketching and Painting

VI.— Timber and its Utilisation. under the Ordinary Difflcul-

VII.— Sledges and Sledge Travelling, j ties of Travel.

VIII.— Boots, Shoes, and Sandals. XX.— The Estimation of Distances,

IX.— Waggons and other Wheeled

Vehicles.

X. — Harness and Pack Animals.
XI.— Camels.
XII.— Cattle Marking.

XIII.— Water, and the Sap of Plants. XXIII.— Bush Veterinary Surgery and
XIV.— Camp Cookery. Medicine.

and Hints on Field ' >bserving.
XXI.— Hints to Explorers on Collect-
ing and Preserving Objects
of Natural History.
XXII.— Ropes and Twines.

Extract from the INTRODUCTION.

Like two voyagers returned from a long cruise in far off seas, we throw together
our joint gleanings in many lands. These do not consist of jewels, gems, gold, or
furs; no piles of costly merchandise do we lay at the reader's feet as offerings from
distant climes, but simply the experiences of two roving Englishmen who have
" roughed it." By those who have to pass through a campaign, travel wild
countries, or explore little known regions, shifts must be made, and expedients of
many kinds had recourse to, of which the inexperienced in such matters would but

little dream In our travels and adventures we have not been associated, the

paths trodden by us being widely separated. Whilst one was exploring the wilds
of North Australia, the other was dwelling in a canvas-covered hole in the earth
before Sebastopol. The scenes chansre; Southern and Tropical Africa is visited
by the late Australian traveller, whilst the Crimea, with its ragged hills and wild
ravines, is exchanged for the jungles of Central India by the other.

"THE FIELD" OFFICE, 346, STRAND, w.c. c

10 A CATALOGUE OP BOOKS

IN TWO VOLUMES,
Demy Svo., price 15s. each., by post 16s.,

THE

MODERN SPORTSMAN'S
GUN AND RIFLE,

INCLUDING

CAME AND WILDFOWL GUNS, SPORTING AND MATCH RIFLES
AND REVOLVERS,

Vol. I.— Game and Wildfowl Guns.
Vol. II.— The Rifle and Revolver.

By J. H. WALSH,

" STONEHENGE," EDITOE OF "THE FIELD,"
Author of" Dogs of the British Islands" " The Greyhound," " British Rural Sports, Ac.

PRESS OPINIONS ON VOLUME I.

" A perusal of Mr. Walsh's book has forced upon us the conclusion, one that will
be shared by nearly every reader, that it is indisputably the standard work on the
subject, and is likely long to remain so— a position it richly merits. Sportsmen
will anxiously look forward to the second volume of the work, for there is every
reason to anticipate that the same high standard will be maintained, and that the
rifle will receive as complete an exposition as the ' Game and Wild Fowl Guns.'
We are only fulfilling a duty to the public when we say that no man connected in
any way with guns or gunnery should be without a copy of Mr. Walsh's masterly
volume."— The Birmingham Daily Gazette, Nov. 21, 1882.

" It will be seen that the work contains a variety of hints which may be useful to
intending purchasers of guns, so that we can confidently recommend an intelligent
glance through it as likely to save money and prevent disappointment." — Saturday
Review.

" Taking the work as a whole, the sportsman will find in it much information on
guns, shot, and kindred topics." — Pall Mall Gazette.

" The most complete work that has yet been written on sporting guns."—
St. James's Gazette.

" For bre«dth of view and completeness this treatise could hardly be excelled.
It has, moreover, the advantage of reporting authoritatively on the very latest
improvements, both as regards weapons and powder and shot, all which objects
of a sportsman's consideration seem to be susceptible of indefinite progress "—
Daily News.

"With such a guide as this, all who appreciate sport will be able to enjoy it
fully, and, what is of importance, will be able to avoid much of the danger
attending the use of imperfect weapons. — The Era.

" THE FIELD " OFFICE, 346, STRAND, WlC.

PUBLISHED BY HORACE COX. 11

Demy 4to., with 12 full-page illustrations, some of which contain Portraits of
Sporting Celebrities, and 24 vignettes, price £1 Is., by post £1 2».

SPORTING SKETCHES

WITH

PEN AND PENCIL.

BY

FRANCIS FRANCIS AND A. W. COOPER.

CONTENTS.

The First of September.

A Day in a Punt.

Mark Cock !

Trouting.

Long Tails and Short Ones.

Paying the Pike.

Eabbit Shooting.
Eoaching.
Grouse Shooting.
Salmon Fishing.
Snipe Shooting.
Grayling Fishing.

Crown 4to., printed on toned paper, price 15s., by post 16«.
THE

ANNALS OF TENNIS

BY

JULIAN MARSHALL.

THIS work will be found very complete, and, it is thought, justly entitled to take its
place as the standard work on Tennis. It has cost its author much laborious
research ; and, independently of its great value to tennis players and all lovers of
the game, it is trusted, from the vast amount of curious lore it contains, the volume
will be found not unworthy of a place on the shelves of the scholar. The author,
himself a well-known amateur, is fully competent to speak with authority on the
game> having had the opportunity of studying the play of the best Continental, in
addition to that of the best English, masters, and, therefore, may be taken as a safe
guide by learners.

CONTENTS.

I. — Tennis Abroad.
II. — Tennis in England.
III. — The Court and Implements.

TV. — The Laws and their History.
V.— The Game.
VI. — Appendix.

THE FIELD" OFFICE, 346, STRAND, w.c.

A CATALOGUE OF BOOKS

A PRESENTATION VOLUME FOR CLUBS.

4«o., levelled boards, gilt edges, (500 pages), with appropriate
illustrations, price One Guinea, by post £l 2s. 4<f.

THE

ENGLISH GAME OF CRICKET:

COMPRISING A DIGEST OP ITS

ORIGIN, CHARACTER, HISTORY, AND PROGRESS,

TOGETHER WITH

AN EXPOSITION OF ITS LAWS AND LANGUAGE.

CHARLES BOX,

keters' Manual," " Reminiscences of Celebrated Players." Essays on
" Songs and Poems," " Theory and Practice of Cricket," 4o.

CONTENTS.

chtpv

TO > Introductory.
VI.)

Glances at the Past and Present State of

Connty Cricket.
VII.— Middlesex.
VIII.— Public School Matches.
IX.— Kent.
X.— Hampshire.
XI.— Surrey.
XII.— Sussex.
XI II.— Nottinghamshire.
XIV.— Yorkshire.

XV.— Warwickshire and Derbyshire.
XVI. — Gloucestershi re.
XVII.— Lancashire and Leicestershire.
XVIII.— The Eastern Counties.

^x'} Intercolonial Matches.
XXl'.— School and Village Matches.
XXII.— Curiosities of Cricket.
XXIII.— Cricket Grounds.
XXIV.— Laws of the Game.

XXV.— Poems, Songs, and Ballads.
XXVI.— Glossary of Words and Phrases.
POSTSCRIPT. — Shakespeare and Cricket
—An Enforced Dissertation.

OPINIONS OF THE PRESS.

" The most complete and interesting work on cricket ever published. No expense
has been spared in making that which is really useful a very handsome volume." —
Belts Life.

" We welcome with heartiness a writer like Mr. Charles Box, who has so
pleasantly united in the splendid volume before us the old order (of cricket) with
the new." — Sporting and Dramatic News.

' This work will prove interesting to all lovers of cricket." — Times.

" The volume is a very handsome one indeed, destined, doubtless, to become an
authority on the essentially ' English Game of Cricket.' "—Morning Post.

"The best work on cricket that has yet come under our notice." — Nottingham
Journal.

" A handsome and well got-up volume, the author being the man of all others
qualified and in a position for compiling such a work. — Kent Herald.

"THE FIELD" OFFICE, 346, STRAND, w.c.

PUBLISHED BY HORACE COX. 13

Price Is., by post Is. Id.

PASTURES, OLD AND NEW:

A Plea for the Improvement of Old Turf, Better

Systems of Grassing-down, and the Prolonged Tenure of

Alternate Husbandry Grass Layers.

By JOSEPH DARBY.

SECOND EDITION, large post 8vo., price 5s., by post 5s. 3d.

.A. IN" O- L I 1ST GK

By FRANCIS FRANCIS.

Author of " A Book on Angling," " By Lake and River," " Hot-Pot," &c.

CONTENTS.

PREFACE.

Chap.

I.— The Art of Angling.
II.— Mid-Water Fishing.
III.— Surface or Fly Fishing.
IV.— The Gudgeon, the Pope or Buff,
the Bleak, the Eoach, the Eudd,
the Dace, the Chub, the Barbel,
the Bream, the Carp, the Tench,
the Perch.

Chap.

V.— The Pike.

VI.— Trout Fishing with Bait
VII.— Fly Fishing for Trout.
VIII.— Trout Flies.
IX.— Grayling Fishing.
X.— Salmon Fishing.
XI. — Salmon Flies.
XII.— On Tackle Making.
ADDENDA.

Post 8vo., in cloth, price 5s., by post 5s.

MISCEJLLANEOUS PAPERS

BY FRANCIS FRANCIS,

Author of " A Book on Angling," " By Lake and Eiver," " Angling," &c

CONTENTS.

A Christmas Eeverie— The First Day of the Season— A Strange Fishing Match—
The Poacher — The Banker — Reminiscences of an Angler ; or, Justices' Justice —
Christmas in the Fisherman's Snuggery— St. May Fly— Catching Tartars—
Under the Boughs, "Now and Then" — Eeminiscences of an Angler, "•Farmer
Gumshun"— Bankers and Tinkers— The Pleasures -of Grayling Fishing— Will
Whistle — An Angler's Christmas Tarn — Eeminiscences of an Angler, " Squaring
the Keeper "—A Week on the Brattle— A Storm on the Bra wle— White Trout and
Salmon Fishing in Galway— Eeminiscences of an Angler, " Anglers' Miseries "
—Sam Coventry— Piscatory Prosings " De Omnibus Eebus," <fcc.— Chewton Pike.

" THE FIELD " OFFICE, 346, STRAND, W.C. d

A CATALOGUE OF BOOKS

Demy 8vo., with folding plates and full-page illustrations printed on
toned paper, price 21s., by post 21s. 9<Z.

MODERN WILDFOWLING.

BY

LEWIS CLEMENT,

" WlLDFOWLER."

CONTENTS.

PUNTING. — Introduction.

I.) Muzzle-loading Punt Guns
II.)- (Flint, Percussion and Copper
III.) Tube Ignition.
IV.)

V.S-BreechloadingPunt Guns.
VI.)

VII.— Loading Punt Guns.
VIII.— Aiming and Firing Punt Guns.
X.— The Setting of Punt Guns.
XL— Eecoil and After-recoil Appa-
ratuses.
XII.
XIII.

XIV.

Punts.

XV.
XVI.
XVII.

XVIII.— Launching Punts and Canoes.
XIX. — Punting Accessories.

XXIII.

XXIV.-My First Single-handed Punt-

ing Trip.

XXV.) Amateur and Professional
XXVI. f Puntsmen.

XX VII. ^)

XXVIII. Uhoulder Guns.
XXIX J

XXX.— Flapper Shooting.
XXXI.— Inland Duck Shooting.
XXXII.— Sailing to Fowl.

XXXIV;} Decors to the Gun.
XXXVl— Decoying in America.
XXXVI.— American Blinds.
XXXVII.— American Canvas - back

Shooting.

XXXVIII.— Shore Shooting.
XXXIX.— Flighting.

XL.) Curious Wildfowl and Sea-
XLI. j fowl Shooting Expedients.
XLII.— Close Time and Wildfowl

and Sea-fowl Acts.
XLIIL— "WildfowlerV1 Table of

Loads.
XLIV.— Netting Plovers and Snipe

Snaring.

XLV.— Snaring and Hooking Sea-
fowl on the Continent.
XLVI.— Decoying into the " Pipes."
XLVII.— Flight Ponds and Bock

Fowling.
XL VIII.— Concluding Remarks.

OPINIONS OF THE PEESS.

" An excellent work indeed, and full of capital illustrations, is ' Modern Wild-
fowling; to recommend it aright I should have, if I were clever enough, and it did

'™se' '* book no ^<^^ ****

urn o puns, ot snge and double handed; sails; punt guns, muzzle-loading as
wellasbreechloading; recoil apparatus; and shoulder guns of all patterns, with
the varying loads required Tor different bores. In addition to this are several
chapters devoted to a narration of the adventures of the author while in pursuit of
w-S^1' S-at h,°me a^d abr°ad-which are very pleasant reading
With the addition of a good index, sportsmen will have in this work a capital vade
mecvm on the art of wildfowling.-Vte Zoologist for November 1880

THE FIELD" OFFICE, 346, STRAND, W.C.

PUBLISHED BY HORACE COX.

15

SECOND EDITION. Large post 8vo., price 7s. 6d., by post 8s.
THE

"IDSTONE" PAPERS.

A SERIES OF ARTICLES AND DESULTORY OBSERVATIONS ON
SPORT AND THINGS IN GENERAL

BY

"IDSTONE,"

OP "THE FIELD."

Chap.

I. — Agricultural Labourers.
II.— The Bough Eider.
Ill— The First of May.
IV.— « Strictly Confidential."

V.— Shooting Dress.
VI.— Some Old Portraits.
VII.— Dens and Sanctums.
VIII.— The Bat-catcher.
IX. — Early Morning in London.
X.— The Earthstopper.
XL— The Shooting Pony.
XII.— Whistle and Whip.
XIII.— Old Traps and Spring-guns.
XIV.— Tom Frere the Hard-riding

Fanner.
XV.— Expecting Brown.

CONTENTS.

XVL— Brown in the Country.
XVII.— The Earthstoppers' Feast.
XVIIL— The White Snipe.
XIX.— Swans and Eagles.
XX.— The Philosophy of Missing.
XXL— Shooting in Aiderney.
XXIL— Shirkers.
XXIIL— Our Black Heath.
XXIV.— Traps and Calls.
XXV.— Northward.
XXVL— A Bright October.
XXVIL— Varied Shooting.
XXVIIL— The End of the Season.
XXIX.— On Beating for Game.
XXX.— Land Valuers and Stewards.
XXXI.— Snipe Shooting.

WOEK BY THE LATE FEANCIS FEANCIS.

Now ready, crown 8vo., price 3s. 6d., by post 3s. 9d.

ANGLING REMINISCENCES,

By the late FRANCIS FRANCIS,

Author of " A Book on Angling," &c.

CONTENTS.

A Christmas Betrospect.

Luck.

Loch Tay.

The Angler's Wish.

Grayling Fishing.

Spring Salmon Fishing.

Barbel Fishing.

Cover Shooting.

Fly Fishing for Ladies.

Saint Mayfly.

My First Salmon Bun.

The Mayfly Mess.

A Month in the West.

Trout Fishing.

In and Out Dales.

Up and Down.

Tom Bowers's Christmas.

THE FIELD " OFFICE, 346, STRAND, W.C.

A CATALOGUE OF BOOKS

Just published, Svo., pp. 463, with 32 illustrations, price 16s., by post
16s. 10d.

ESSAYS

ON

SPORT AND NATURAL HISTORY.

By J. E. HARTING.

CONTENTS.

Shooting — Hawking — Fishing — Training Hawks — Lark Mirrors — Plover
Catching— Fishing with Cormorants— Decoys— The Irish Wolf hound— The Badger
—Wild Turkeys— The Great Bustard— Seals— Wild Swans, Ac.

Thirty-eight Essays: concluding with Practical Hints on Bird Preserving for
the use of Travellers and Collectors.

In demy Svo., price 3s. 6d., by post 3s. 9d.

HINTS ON THE MANAGEMENT OF HAWKS,

By J. E. HARTING,

Author of" A Handbook of British Birds," " Essays on Sport and Natural History"
Large post 8vo., price 6s., by post 6s. 4<Z.

RAMBLES AFTER SPORT;

OE,

TRAVELS AND ADVENTURES IN THE AMERICAS AND AT HOME.
BY "OLIVER NORTH."

CONTENTS.

A Week's Duck Shooting at Poole— That Sheldrake— Quail Shooting in California
—Bear Hunting in Mexico— Bear Shooting in California— My First Elk— My Last
Bear— Bound Cape Horn, Valparaiso, Santiago— Andacollo, Lima, Panama,
Jamaica— Country Sports and Life in Chile— Shooting in Chile— Two Days' Fishing
in Chile—" Toling" for Docks in California— Up the Sacramento— The White Elk
of Astoria— Sport in the Coast Bange Mountains.

In large post 8vo., limp cloth, price 2s. 6<Z., by post 2s. Sd.

ITS

AGRICULTURE, STOCKFEEDING, SCENERY, AND SHOOTING.

By S. NUGENT TOWNSHEND, J.P.

("ST. KAMBS.")

" THE FIELD " OFFICE, 346, STEAND, W.C.

PUBLISHED BY HORACE COX- 17

Now ready, VOLUME I. (containing Parts L, II., and III.), in
crown 8vo., red cloth, price 6s., ~by post 6s. 6d.

THE

HUNTING COUNTRIES

OP

ENGLAND,

THEIR FACILITIES, CHARACTER, AND REQUIREMENTS,

A GUIDE TO HUNTING MEN.

"BROOKSB Y."

By

CONTENTS.

PART I.— Introduction— The Belvoir— The South Wold— The Brocklesby—
The Burton and The Blankney— The Fitzwilliam— The Quorn— The Cottesmore—
The Puckeridge— The Old Berkeley.

PART II.— The North Warwickshire— The Pytchley— The Woodland Pytchley
—The Atherstone— The Billesdon or South Quorn— The Meynell— The Bicester and
Warden Hill Hunt— The Heythrop— The Old Berkshire— The South Oxfordshire—
The South Nottinghamshire— The East Kent— The Tickham— The Vine— The
South Berkshire— Mr Garth's— The H. H.— The Tedworth— Lord Ferrers'— The
Warwickshire.

PART III.— The Dulverton— The Stars of the West— Mr. Luttrell's— Lord
Portsmouth's— The Essex and the Essex Union— The Hertfordshire— The Whaddon
Chase— The Vale of White Horse— The Cheshire and South Cheshire— The Black-
moor Vale— The Cambridgeshire— The Duke of Grafton's— The Holderness— The
Oakley— The North Herefordshire— The Duke of Buccleuch's— The Tynedale—
Lord Percy's— The Morpeth— The Eufford.

Also now ready (VOLUME II.).

PART IV.— The Badsworth— The Southdown— The East Essex— The Bram-
ham Moor— The East Sussex— The Essex and Suffolk— The York and Ainsty— Lord
Fitz William's— The Crawley and Horsham— The West Kent— Sir Watkin Wynn's
—The Hursley— The Hambledon— Lord Coventry's— The Grove— The West Norfolk
—The Bedale— Lord Zetland's— The Craven— The Surrey Union.

PART V.— The Old Surrey— Mr. Eichard Combe's— The Burstow— The Hur-
worth— The Cattistock— The Suffolk— The Shropshire— The Earl of Eadnor— Capt,
Hon. F. Johnstone's— The South Durham— The Worcestershire— The Ledbury—
The South Herefordshire— The South Staffordshire— The North Staffordshire— The
Duke of Beaufort's— The Cots wold— The Dumfriesshire— The Albrighton— The
North Cotswold.

PART VI.— Lord Middleton's— The Sinnington— The Wheatland— The United
Pack— The Chiddingfold— Lord Fitzhardinge's— Hon. Mark Eolle's— South- and-
West Wilts— Lord Portman's— The Cleveland— The North Durham— Braes of
Derwent— The Eadnorshire and West Hereford— The Monmouthshire.

Each Part is published separately, price 2s. 6d.

Now ready, price 2s. 6d., free by post 2s. 8d.

PRACTICAL DINNERS:

CONTAINING 108 MENUS AND 584 EECIPES.

By " The G. C.,"

"Author of Eound the Table.'

THE FIELD" OFFICE, 346, STRAND, w.c.

18 A CATALOGUE OP BOOKS

Now ready, royal 8vo., price 10s. 6d., by post 11s.

HORSE BREEDING RECOLLECTIONS.

BY

COUNT G. LEHNDORFF,

CONTAINING :

Notes on the Breeding of Thoroughbred a— Tn-breeding and Out-
crossing — Pedigrees of all the Principal Sires — and Genealogical
Tables of Celebrated Thoroughbreds.

Post 8vo., price 7s. 6d., by post 8s.

MOSS FROM A ROLLING STONE;

MOORISH WANDERINGS AND RAMBLING
REMINISCENCES.

" Sarcelle" of " The Field" <fcc., Author of " The Diamond Diggings of South Africa."
Price 5s. cloth, by post 5s. 4cf.

A Year of Liberty ; OP, Salmon Angling1 in Ireland,

BY W. PEARD, M.D., LL.B.

Price 2s. 6cL, by post 2s. 8d., in limp cloth.

RABBITS FOR PROFIT AND RABBITS FOR POWDER.

A Treatise upon the New Industry of

Hutch Babbit Farming in the Open, and upon Warrens specially intended for
Sporting Purposes ; with Hints as to their Construction, Cost, and Maintenance.

By R. J. LLOYD-PRICE.

Price 6d., by post Q±d.

THE "FIELD"
LAWN TENNIS UMPIRES1 SCORE-SHEET BOOK

(with Instructions for the use of Umpires).

Adapted for the use of Umpires and Players, as used at the Championship
Meetings.

"THE FIELD" OFFICE, 346, STRAND, w.c.

PUBLISHED BY HORACE COX.

19

THIRD EDITION. Large post 8vo. price 7s. 6d. cloth, by post 7s. 10<f.

FACTS AND USEFUL HINTS

RELATING TO

FISHING AND SHOOTING:

TO WHICH IS ADDED

A LIST OF RECIPES FOR THE MANAGEMENT AND CURE

OF DOGS IN DISEASE.
Edited by I. E. B. C.,

Editor of " The Gamekeeper's and Game Preserver's Account Book and Diary," &c.

FISHING.

Baits— Fish— Fish Hatching— Flies and Fly Making— Flights— Floats— Gut— Lines
—Miscellaneous— Nets— Ponds and Streams— Eods— Wading Boots— Wax.

SHOOTING.

Birds and Beasts — Breeding — Coverts — Deer— Dogs— Ferrets— Foxes— Guns-
Kennel— Miscellaneous— Nets— Preserving— Eabbits— Eifles— Traps— Vermin.

APPENDIX.— Diseases of Dogs.
In post 8vo., with Illustrations, price 3s. 6d., by post 3s. 9d.

THE PRACTICAL MANAGEMENT OF FISHERIES.

A BOOK FOE PEOPEIETOES AND KEEPEES.

BY FRANCIS FRANCIS,

Author of " Fish Culture," " A Book on Angling," " Beports on Salmon
Ladders," <fec. &c. &c.

CONTENTS.

Chap.

I.— Fish and Fish Food.
II.— How to Grow Fish Food and how

to Make Fishes' Homes.
III.— On the Management of Weeds and

the Economy of Fishing.
IV.— The Enemies of Trout and how to

Circumvent them.
V.— The Artificial Incubation of Ova.

.. — On the Hearing of Fry and
the Conduct of Ponds, Stews,
&c.

VII.— Some Hatcheries.
VIII.— Coarse Fish.
IX.— On Salmon and Trout Ladders

and Passes.
APPENDIX.— Notes, Ac.

Price 2s. 6d. by post 2s. 9d.

GAME REGISTER,

GIVING AN ACCOUNT OF EACH HEAD OF GAME KILLED, AND HOW

DISPOSED OF.

Containing also Divisions for Begistering Sporting Engagements and
General Observations.

THE FIELD" OFFICE, 346, STRAND, w.c.

A CATALOGUE OF BOOKS

Now ready, quarto, price 5s., by post 5s. Qd.
THE

GAMEKEEPER'S AND GAME PRESERVER'S

2Ut0mit Book anb Marg.

By I. E. B. C.,

Editor of " Facts and Useful Hints relating to Fishing and Shooting," " The
Angler's Diary," &c.

ITS SECTIONS COMPRISE-

THE SHOOTING AND ITS GUARDIANS.
MEMORANDUM OF AGREEMENT.
RECEIPTS AND PAYMENTS.
VERMIN DIARY—

General Summary.

POULTRY DIARY—

Receipts and Payments.
General Balance Sheet.

PHEASANT DIARY—

Receipts and Expenses.
General Balance Sheet.

DOG DIARY—

Kennel Names, Ages. Value. &c.
Kennel Occupants at the beginning of

each Quarter of the Year.
Produce Register— Bitches.
Stud Register.

ts and Expenses.
General Balance Sheet.

GAME DIARY—

Total Summary of the Season .
Produce of the Beats or Coverte.
Tenants, &c., to whom Game should be

given.
Inventory of Appliances, &c.

STOCK VALUATION.

General Balance Sheet for the Year.

In handy pocket size, price Is. 6d., by post Is. Id.

THE GAMEKEEPER'S SHOOTING MEMORANDUM BOOK

FOR THE

REGISTERING OF GAME SHOT, MEMORANDA OF SALE, Ac.

By I. E. B. C.,

Editor of "Facts and Useful Hints relating to Fishing and Shooting," " The Game-
keeper's and Game Preserver's Account Book and Diary," Ac.

Crown Svo., price 2s. 6d., by post 2s. 9d.

PUBLIC SHOOTING QUARTERS

IN ENGLAND, WALES, SCOTLAND, IRELAND, AND ON THE
CONTINENT.

By " W I L D F O W L E R ,"

Author of « Shooting and Fishing Trips," « Modern Wildfowlmg," "Table of
Loads," &c.

" THE FIELD " OFFICE, 346, STRAND, W.C.

PUBLISHED BY HORACE COX. 21

THIRD EDITION, ENLARGED AND REVISED.
Large post Svo., with Illustrations, price 5s. cloth, by post 5s. 4d.

THE COUNTRY HOUSE:

A COLLECTION OF USEFUL INFORMATION AND RECIPES,

Adapted to the Country Gentleman and his household, and of the greatest utility
to the housekeeper generally.

By I. E. B. C.,

Editor of " Facts and Useful Hints relating to Fishing and Shooting," and " The
Gamekeeper's and Game Preserver's Account Book and Diary."

PUBLISHED ANNUALLY. In post 8vo., price Is. 6d., by post Is. Sd.

THE ANGLER'S DIARY

AND

TOURIST FISHERMAN'S GAZETTEER

CONTAINS

A Record of the Rivers and Lakes of the World, to which are added a Lrat of
Rivers of Great Britain, with their nearest Railway Stations.

Also Forms for Registering the Fish taken during the year; as well as
the Time of the Close Seasons and Angling Licences.

By I. E. B. C.,

Editor of " The Gamekeeper's and Game Preserver's Account Book and Diary," &c.

THIRD EDITION. In /cap. Svo., price Is., by post Is. Id.

WILD BIRDS' PROTECTION ACT, 1880,

WITH COMMENTS ON THEIR RESPECTIVE SECTIONS

Explanatory of their bearing as regards owners and occupiers of land, sportsmen,
bird catchers, bird dealers, Ac. ; together with Notes on the Birds named in the
Schedule, their provincial names, <fec.

An accurate exposition of and commentary on the recent measure, and will
dispel many misconceptions of its scope." — Quarterly Review.

'A capital annotated edition of the Act." — Saturday Review.

" You have not the book of riddles about you, have yon?"— Merry Wives.
Price 5s-, by post 5s. 3d.

"WHETSTONES FOR WITS;"

OR

DOUBLE ACROSTICS.

IB1T -V^IEeiOTJ-S IHI^ICsriDS-
EDITED BY " CRACK."

"THE FIELD" OFFICE, 346, STRAND, w.c.

22 A. CATALOGUE OP BOOKS

SECOND EDITIQN. In demy Svo., price 10s. 6d., ly post 11s.

ESTATE MANAGEMENT:

A PRACTICAL HANDBOOK FOB LANDLORDS, STEWARDS,
AND PUPILS,

WITH A

LEGAL SUPPLEMENT BY A BARRISTER,

ALSO

Cenant SRtjgflt fr0m a HanMtarlr'* point 0f SFtcto.
By CHARLES E. CUBTIS.

CONTENTS:

Dhaf.— Letting and Leases.
II. — Farm Valuations.
III.— Forestry.
IV.— Underwood.
V.— Fences.
VI.— Grasses suitable for Woods and

Plantations.
VII.— The Home Farm.

VIX } EePairs and Materials.

X.— The Blights of Wheat and other

Cereals.
XI. — Accounts.

XII.— Useful Eules of Arithmetic and
Mensuration.

In crown 8vo., price Is., by post Is. Id.

CATECHISM OF ESTATE MANAGEMENT.

SECTION I.

LETTING AND LEASES.

BY

CHAS. E. CURTIS, F.S.I.,

Professor of Estate Management at the College of Agriculture, Principal of the
School of Estate Management, Author of " Estate Management," &c.

In crown Svo., with Thirteen full-page Plates, price 2s. 6d., by
post 2s. 9<L

THE SWIMMING INSTRUCTOR:

A TREATISE ON THE ARTS OF SWIMMING AND
DIYING.

By WILLIAM WILSON.

Author of " Swimming, Diving, and How to Save Life," " The Bather's Manual,"
' Hints on Swimming."

" THE FIELD " OFFICE, 346, STRAND, W.C.

PUBLISHED BY HORACE COX. 23

SECOND EDITION, Greatly Enlarged (with Illustrations and Plans of Silos).
Price 6s., by post 6s. 6d.

SILOS

FOB

PRESERVING BRITISH FODDER CROPS STORED
IN A GREEN STATE.

Notes on the Ensilage of Grasses, Clovers, Vetches, &c.

COMPILED AND ANNOTATED

BY THE

OIF " TIHIIE

CONTENTS.
INTRODUCTORY.— Cattle-feeding v. corn-growing; previous information on the

storage of green fodder in pits ; silos in ancient and modern times.
CHAP.

I. — Summary of Practice.
II. — CROPS FOR THE SILO. — Grasses, clovers, lucerne, vetches, maize, green rye

and oats, spurrey, buckwheat, comfrey, roots and miscellaneous crops.
III.— THE VARIOUS KINDS OF SILOS.— Earthern pits and other simple forms of
silos ; barn and other converted silos ; specially constructed silos of stone,
brick, concrete, wood, &c. ; ensilage stacks and barrels.

IV.— COST OF SILOS.— Estimation of capacity; cost of British silos, specially
constructed; converted buildings ; patent 'silos of concrete slabs, slate,
and wood; French silos; relative capacity and cost of silos and hay
barns; roofs.

V. — FILLING THE SILO. — Mixture of dry material with green fodder ; influence
of wet weather ; chopping up the fodder; slow v. quick filling ; curbs or
super-silos ; the use of salt ; trampling down the fodder.

VI.— COVERING AND CLOSING THE SILO.— Straw and other materials ; the cover-
ing boards ; closing the doorway.

VII. — WEIGHTING THE SILO. — Amount of weight to put on; consequences of
insufficient pressure; expression of juice from the fodder; mechanical
arrangements for pressure ; cost of apparatus.
VIII. — OPENING THE SILO. — Precautions to be taken in uncovering the silage, and

removing the weights.

IX. — EFFECT OF ENSILAGE ON FODDERS. — Fermentation in the pit; advantages
and losses produced by fermentation ; reduction in weight and altered pro-
portion of constituents of silage, resulting from excessive fermentation.
X. — FEEDING QUALITIES OF SILAGE. — Effect of amount of moisture in diluting
the nutrient matters ; comparative money value of green fodders and
silage ; results of feeding experiments on the condition of the animals
and the production of milk and cream ; silage for horses, sheep, Ac.
XL— EFFECT OF SILAGE ON DAIRY PRODUCE.— Complaints of bad flavour in

milk and butter; causes of conflicting reports.

XII.— COST OF HAYMAKING v. ENSILAGE.— Differences of cost of the process
under various conditions.

Price 6d., by post Id.

SHORT NOTES ON SILO EXPERIMENTS AND PRACTICE.

(Extracted from "Silos for Preserving British Fodder Crops.")

"THE FIELD" OFFICE, 346, STRAND, w.c.

24 A CATALOGUE OF BOOKS

Price 6d., by post 7d.

HARVESTING CROPS INDEPENDENTLY
OP WEATHER:

Practical Notes on the Neilson System of Harvesting.

By "AC RICO LA,"

AND OTHER CONTRIBUTORS TO " THE FIELD."

In crown Svo., price 2s. 6d., by post 2s. Sd.

MANURES:

THEIR EESPECTIVE MEEITS FROM AN ECONOMICAL
POINT OF VIEW.

BY -A.. "W. CIR.IE'WS,

Author of " Guano : its Origin, History, and Virtues," " The Potato and its Cultivation," &o.

- CONTENTS.

PAKT. I.— Definition of the Word " Manure "—Nature's Modes of Applying
Fertilisers— History— Classification.

PART II.— The Value of Ploughing Down Green Crops— Weeds— Sea-weed—
Straw— Sawdust— Tanners' Bark— Wood Ashes— Peat— Rape Cake— Hemp-
Poppy, Cotton, and Cocoa-nut Cakes — Bran — Malt Dust — Brewers' Grains — Coal —
Soot— Charcoal.

PAET III.— Dead Animals— Fish— Blood— Animalised Charcoal— Bones— Horn
—Woollen Bags, Hairs, Feathers, &c.— Night-soil— Farmyard Manure— Guano.

PAET IV.— Salts of Ammonia— Salts of Magnesia— Salts of Potash— Salts of
Soda — Common Salt — Lime and its Compounds — " Ooze."

In crown 8vo., price 2s., by post 2s. 2d.

THE POTATO AND ITS CULTIVATION.

.A.. W.

y, an
lerits, &c.

Author of " Guano : its Origin, History, and Virtues," " Manures : their Respective

CONTENTS.

Derivation — History — Constituents — Varieties — Sprouting - Soils — Planting —
Manures— Earthing up— Disease— Scab— Storing— Forcing— Producing New
Varieties— Substitutes for the " Potato "—Miscellaneous Information.

"THE FIELD" OFFICE, 346, STRAND, w.c.

PUBLISHED BY HORACE COX.

25

Demy 8w., price 3s. 6d, by post 3s. 9d., Illustrated with several Diagrams.
THE

PRACTICAL SURVEYOR :

A TEEATISE UPON SURVEYING.

SPECIALLY AEEANGED FOE THE GUIDANCE OF PUPILS, STEWAEDS,
THE SCHOLASTIC PEOFESSION, AND INTENDING EMIGEANTS.

BY THOMAS HOLLOWAY.

Cha?.

CONTENTS.

[.—The Man and his Outfit.
II.— The Chain — Cautions to Be-
ginners— Best Figure for Chain
Surveying.

III. — Boundaries.

IV.— Setting-out Lines by the Eye and

passing Obstructions.
V.— Division of the Circle and Use of
Box Sextant — Chain Angles
Condemned — Cross Staff Con-
demned—The Optical Square
— Measuring Inaccessible Dis-
tances.

VI.— The Theodolite — Setting-out

Lines with the Theodolite.
VII.— Eeduction of the Measure of Un-
dulating Ground to Horizontal
Measures and Table of Vertical
Angles.

VIII.— Measuring Lines — The Offset
Staff and taking Offsets.

IX.— To prove the Correctness of Ob-
servations taken with the Sex-
tant—Single Fields Measured
with the Chain and Optical
Square, so that the Areas can
be directly Calculated.
X.— To Set-out a Eight Angle with
the Chain— Figures of the Lines
of Measurement best adapted
to Irregular Fields

XI. — Equalising Boundaries, and
Drawing a Triangle equal to a
given Figure.

XII. — Computation of Areas of Irre-
gular Fields.

III.— Example of a Survey of several
Fields together, and the Field
Book.

XIV.— Eeference Numbers to Maps—
To put Detached Buildings
in correct Positions on a Plan
by Means of Unmeasured
Lines — Lines Measured on
the Work— Making Stations.
XV.— Plotting — Selection and Ma-
nagement of Paper — Inking
in.

XVI. — Surveys made for the purpose
of Dividing Land into Stated
Quantities.
XVII.— Setting-out Allotments and

Building Plots.

XVIII— Angles and Bearings, and Use
and Adjustment of Circular
Protractor.

XIX.— Traverse Surveys.
XX.— Trespass.

XXI— Quality Lines— Superstructures
and Works Underground —
Harvest and Coppice Work
— Seducing Plans from a
Large Plan to a Small One.
XXII— To Copy a Map— Colouring,

Penmanship, &c.

XXIII — Commencement of a Parish
Survey — Surveying to a Scale
of Feet.

XXIV.— Town Surveying.
XXV.— Testing the Accuracy of a

Survey — General Eemarks.
XXVI— In Memory of the Past.

Price 6d., by post Td. ; or 2s. 6d. the half-dozen.

"THE FIELD" DUPLICATE JUDGING BOOK

Facilitates the work of the Judges at Poultry and other Shows, by a very simple
method of entering and preserving a duplicate judging list.

•'THE FIELD" OFFICE, 346, STRAND w.c.

26 A CATALOGUE OF BOOKS

SECOND EDITION. In Three Parts, large post 8vo., price 5«., by post 5s. 4rf.., each.
THE

FAEM, GARDEN, AND STABLE

By I. E. B. C.,

Editor of " The Gamekeeper's and Game Preserver's Account Book and Diary."

Part I.— The Farm.

Cattle— Crops— Dairy— Diseases— Fencing— Food for Stock— Manures-
Miscellaneous— Pigs— Sheep— Soils— Weeds— Woods.

Part II.— The Garden.

Flowers— Fruit— Houses— Lawns— Manures— Miscellaneous— Seeds— Trees and
Shrubs — Vegetables — Vermin — Weeds.

Part III.-The Stable.

Carriages— Diseases— Feeding— Harness, &c.— Miscellaneous— Stable Management.

PUBLISHED ANNUALLY. In large post Svo.
THE

KENNEL CLUB STUD BOOK :

CONTAINING A COMPLETE

RECORD OF DOG SHOWS AND FIELD TRIALS,

WITH

Pedigrees of Sporting and Non-Sporting Dogs.

Vol. I., from 1859 to 1873, price 12s. 6d., by post 13s.

PRICE 10s. 6d., BY POST 10s. lOd. EACH—

Vol. III., 1875; Vol. IV., 1876; Vol. V., 1877; Vol. VI , 1878-

VoLVIL, 1879; VoLVIIL, 1880; Vol. EX., 1881; Vol. XL, 1883;

Vol. XII., 1884 ; Vol. XIII., 1885 ; Vol. XIV., 1886.

Demy 8vo., price Is., by post, Is. Id.

THE EAELY MATURITY OF LIVE STOCK

By HENRY EVERSHED,

Writer on AgricuHure in the « Journal of the Eoyal Agricultural Society of
.ungiand, The Field," '< Qaarterly Review," <fcc.

"THE FIELD" OFFICE, 346, STRAND. W.C.

PUBLISHED BY HORACE COX.

27

PUBLISHED ANNUALLY. Demy 4>to., price Is., by post Is. 2d.

AND SPORTSMAN'S ILLUSTRATED CALENDAR FOR 1888.

Articles on the following Subjects are included in the List of Contents :
NOTES ON THE PAST RACING SEASON.
YACHT RACING IN 1887.
LIST OP HUNTS, THEIR MASTERS, &c.

FIELD TRIALS WITH POINTERS AND SETTERS IN 1887.
DOG SHOWS AND CANINE MATTERS IN 1887.
SUMMERING HUNTERS.
FOXHUNTING AND FARMERS.
COUNTY CRICKET IN 1887.
FISHERMEN'S COMFORTS.
BAITS AND BAIT CATCHING.
TENNIS SEASON OF 1887.
FLOWERS WORTH GROWING WELL.
VELOCITY OF THE FLIGHT OF BIRDS.
POULTRY IN LIMITED RUNS.
INVISIBLE APPROACH FOR WOOD PIGEONS AND OTHER

BIRDS.

THE GOLFING GREENS OF THE WORLD.
TABLES OF AMATEUR AND PROFESSIONAL PEDESTRIAN,

BICYCLING, AND SWIMMING PERFORMANCES, the best

on Record.
STALLIONS FOR BREEDING BLOODSTOCK AND HUNTERS

(List of about 300 Stallions, with their Pedigrees, and Fees for

Thoroughbred and Half -Bred Mares).

ALSO SUMMARIES, TABLES, RECIPES, &c., viz.,

Angling close seasons

Athletic championships

Beagles, packs of

Bicycling, best times on record

Boat-races, Oxford and Cambridge

Boots, waterproofing and boning

Brown harness, dressing for

Cambridgeshire winners

Cesarewitch winners

Close seasons for game

Cricket on cocoa-nut matting

Derby winners

Dog clubs, list of

Fairs for horses, &c.

Foxhounds, packs of

Game, legal season for killing

Golfing greens of the British Isles

Harriers, packs of

Huntsmen, changes of

Jumping records

Lawn, improving the turf of

Oaks winners

Otter hounds, packs of

Public Schools athletics n 1887

Eaces of 1888, dates of

Eacquets, Schools challenge cup

Eunning, best times

Stimulants for ferns

St. Leger winners

Swimming, amateur performances

Tennis, University matches

Terms, University and Legal

Tricycling performances

University athletic sports

University boat-races

University racquet matches

University tennis matches

Walking, best times.

" THE FIELD '' OFFICE, 346. STKAND, W.C.

28 A CATALOGUE OF BOOKS

FOURTH EDITION. In demy ^to., on toned paper, and in fancy cover, price 2s.,
by post 2s. 2d.

THE BOOK OF DINNER SERVIETTES;

CONTAINING

A NEW INTEODUCTION ON THE DECORATION OF DINNER TABLES,
AND GENERAL DIRECTIONS FOR FOLDING THE SERVIETTES.

There are Twenty-one different kinds given, with Ninety-two Woodcuts Illustrative
of the various Folds required, and the Serviettes complete.

Just published, price 5s., by post 5s. 2d.

"COMBINED FIGURE SKATING;"

Being a collection of 300 combined figures, as skated by the Skating Club, London,
the Wimbledon Skating Club, &c., illustrated by 130 scaled diagrams, showing the
correct direction of every curve executed by the skater, and the recognised amount
of circling round the centre; together with a progressive series of alternate
" calls." The figures are named in accordance with the revised system of nomen-
clature and rules for combined figure skating, compiled by the Skating Club,
London, Sept. 11, 1882. Diagrams of the combined figures in the first and second
class tests of the National Skating Association are included.

BY MONTAGU S, F. MONIER-WILLIAMS AND STANLEY F. MONIER-WILLIAMS

(Members of the Wimbledon Skating Club).
Now ready, post free, 6d., cloth gilt..

RULES OP THE GAME OP HOCKEY

AND OF

THE HOCKEY ASSOCIATION.

Price Qd., by post,

TIKE

Eules and Bye-Laws as to Boating, Fishing, the Use of Steam
Launches, &c.

By C. E. GODDARD, Solicitor.

THE FIELD" OFFICE, 346, STRAND, w.c.

PUBLISHED BY HORACE COX.

29

In crown 8vo., price 5s., by post 5s. 4d.

OR,

THE ARTS OF ROWING AND TRAINING.

EDWIN DAMPIER BRICKWOOD

(EX-AMATEUR CHAMPION OP THE THAMES).

CONTENTS.

ROWING.

Chap.
I. — Introduction: Past and Present

Condition of Boat-racing.
II.— Eacing Boats : Their History and

Fittings.
III.— The Sliding Seat: Its Invention,

Adoption, and Theory.
IV.— How to Use an Oar, and Sculls.
V.— Faults and Errors : What to avoid.
VI.— Steering : Coxswain and Non-
coxswain.

VII.— Teaching Beginners.
VIII.— Coaching for Eaces, and Selec-
tion of Crews.
IX.— The Varieties and Conduct of

Boat-races.
X.— The Laws of Boat-racing.

Chap.

XL— The Qualifications of Ama-
teurs.
XII.— Boat Clubs : Their Organisation

and Administration.
XIII.— Historical Eecords, A.D. 1715 to

1838.
XIV.— Historical Eecords, A.D. 1839 to

1855.

XV.— Historical Eecords, A.D. 1856 to
1875.

TRAINING.
XVI.— Its Principles.
XVII.— Its Practice.
XVIII.— Prohibitions, Ailments, <fcc.
APPENDIX.— Eules for Betting.
INDEX.

PUBLISHED ANNUALLY. Price Is., by post Is. Id.

THE ROWING ALMANACK AND OARSMAN'S
COMPANION FOR 1887,

Edited by E. D. BRICKWOOD

(EX-AMATEUR CHAMPION OF THR THAMES),

Author of " Boat-Eacing ; or, the Arts of Bowing and Training."

CONTENTS.

A Calendar with Space for Memoranda
and High Water Table, with a Table
of Tidal Observations.

A Eeview of the Bowing Season.

Eecord of all Eegattas and Principal
Club Eaces, with a copious Index.

The Championship of the World.

Amateur Bowing Association.

The Definition of an Amateur.

The Laws of Boat-Eacing.

Henley Begatta Bules.

Metropolitan Amateur Eegatta Eules.

A Bowing Directory.

The Lengths of the different Bacing

Courses.
An Itinerary of the Biver Thames from

Oxford to Putney, showing all the

points of interest, with Hotels, &c.
The Eule of the Eoad on the Biver.
Thames Navigation Bules.
The Thames Preservation Act, 1885.
The Thames Bye-Laws, 1886.
Tables of Winners of all the principal

Baces and Begattas.

" THE FIELD " OFFICE, 346, STRAND, W.C.

30 A CATALOGUE OF BOOKS

T2EPORTS on SALMON LADDERS, with Original

•*-* Drawings, Plans, and Sections. By FRANCIS FRANCIS. In post 4to., price

2s. Gd., by post 2s. Id.

A MANUAL of the LAW of SALMON FISHERIES in
ENGLAND and WALES, with a copious Index. By SPENCER WALPOLK,
one of Her Majesty's Inspectors of Salmon Fisheries. Price 2s. 6d, by post 2s. 8d.

A TABLE of CALCULATIONS for use with the "Field"

•*"*• Force Gauge for Testing Shot Guns. Also an Illustration and Description
of the Apparatus. In demy 4to., price 2s. 6rf.

RULES of PIGEON SHOOTING. Published by

Special Permission, the Hurlingham Club and the Gun Club Eules of Pigeon
Shooting. SECOND EDITION. Bound together in cloth, gilt edges, price 6d., by
post Id.

HHHE COURSING CALENDAR, for the Sprint Season

-1- 1887, contains Eeturns of all the Public Courses run in Great Britain and
Ireland. A revised List of Addresses of Coursing Secretaries. Public Coursers,
Judges, Slippers, and Trainers, with List of Waterloo Cup Winners, Greyhound
Sales, <fcc. Edited by C. M. BROWNE (" ROBIN HOOD "). Price 105. &d.

LAWS of LAWN TENNIS, as adopted by the

Marylebone Cricket Club and the All England Croquet and Lawn Tennis
Club. Entered at Stationers' Hall. Price 6d., by post 6jd.

Price 6d., by post 7d.
ALL ENGLAND LAWN-TENNIS CLUB REGULA-

^ TIONS for the MANAGEMENT of LAWN-TENNIS PRIZE MEETINGS,

1887.

HHHE

Price Is., by post Is. 3d.
FIELD" LAWN TENNIS CALENDAR for

1887. Containing a List of Secretaries and their Addresses; Synopsis of

and a

ITALIAN SYSTEM of BEE KEEPING; being an

Exposition of Don Giotto Ulivi's Economical Frame Hives and Honey
!££**£, ? ^ J; ?ai£ell'^te CaPt- H'M- 3l8t Regiment. With lllustra-
th« ?• > ^-^ lS-f\d- This Pamphlet contains practical, directions for
ffLTh g anf u*lllsatlon of frame hives, costing less than 2s. each, and a centri-
fugal honey extractor costing 5s. or 6s.

" THE FIELD " OFFICE, 346, STRAND, W.C.

PUBLISHED BY HORACE COX.

In 4/0., printed on toned paper, with plates, price f>.<?., by post 5s. 4d.

THE QUEEN LACE BOOK:

AN

Historical and Descriptive Account of the Hand-made
Antique Laces of all Countries.

This work contains the whole of the series of articles on Antique Point Lace
which have been published in " The Queen." It will prove an invaluable guide
and book of reference to ladies interested in Antique Lace, and, with its highly
ornamental embossed cover, will form a handsome ornament for the drawing-room
table. _

ENGLISH TEANSLATIONS OF THE CLASSICS.

Post 8vo., 540 pages, price Is. 6d.

HALF-HOURS WITH GREEK AND LATIN AUTHORS.

FEOM VAEIOUS ENGLISH TEANSLATIONS, WITH
BIOGEAPHICAL NOTICES.

By G. H. JENNINGS and W. S. JOHNSTONE,

Authors of a "A Book of Parliamentary Anecdote."

In post 800., price 5s., by post 5s. id.

THE BARB AND THE BRIDLE:

HANDBOOK OF EQUITATION FOR LADIES,

AND

MANUAL OF INSTEUCTION IN THE SCIENCE OF EIDING FEOM THE

PEEPAEATOEY SUPPLING EXEECISES ON FOOT TO THE FOEM

IN WHICH A LADY SHOULD EIDE TO HOUNDS.

By "VIEILLE MOUSTACHE."

Handsomely bound in cloth, price 3s. Gd., by post 3s. 9d.

ACTING CHARADES FOR OLD AND YOUNG.

Author of "The Enchanted Toasting Fork," &c.
In pa/per cover, price 6d.

QTJiEEisr " ZR/EOXIFIES.

By " THE G. C." (Author of " Bound the Table").
" THE QUEEN " OFFICE, 346, STRAND, W.C.

32 A CATALOGUE OF BOOKS.

Now Ready, Price One Shilling ; by Post, Is. 2$d.

THE QUEEN ALMANAC,

AND

LADY'S CALENDAR for 1888.

AMONG ITS CONTENTS WILL BE FOUND

A CHROMO-LITHOGRAPH PLATE OF DESIGNS

FOR ARRASENE, CHENILLE, AUTUMN LEAF AND LINEN EMBROIDERY IN
COLOURS. DESIGN FOR PRIMROSE LEAGUE DIPLOMA.

TWO COLOURED PLATES OF NOVELTIES IN KNITTING
AND CEOCHET.

Specimens of China and Pastel Painting ; Ornamental Chests, Caskets,
and Keys ; Nail, Band, and Bepousse Work ; Suggestions for Fancy
Needlework ; Artistic Arrangement of Rooms ; Evening Toilettes,
Headdresses, and Jewellery ; Fashions in Muffs, Boas, Bags, Fans,
and Sunshades, &c.

FORTY-NINE PORTRAITS:

Her Majesty the Queen's Daughters, Daughters-in-law, Grandchildren

and Great Grandchildren, Prince Alexander and Princess Victoria of

Teck, the Emperor and Empress of Brazil.

Fashionable Mantles, Dresses, and Headdresses for Indoor and Out-
door wear ; Winter Mantles and Hats ; Children's Costumes, Hats,
and Toques ; Suggestions for Fancy Costumes ; Lingerie, Under Linen,
and Children's Costumes ; Suggestions for Ornamental and Useful
Arrangements of Flowers and Plants ; Chip Carving, &c.

Full information is given relating to— The Eoyal Family ; the Royal Household ;
the Government ; British and Foreign Ambassadors ; Lord Lieutenants of Counties
in the United Kingdom; Irish and Scotch Representative Peers; Peers who are
Minors; Peeresses in their own right ; Alphabetical List of the Surnames of the
Peers Temporal; Complete List of the House of Peers, with their Surnames and
Titles, and the Titles of their Eldest Sons; Jewish Calendar ; Bank of England;
Post Office Regulations; Eclipses in 1888; List of Charities, Associations, &c.;
Obituary of Ladies of Distinction during the Past Year.

RECIPES FOR SOUPS, FISH, ENTREES, SWEETS, BREAKFAST
DISHES, AND SAVOURIES,

" THE QUEEN " OFFICE, 346, STRAND, W.C.

YB 16470