L SOILS

EBB

f ff t Kff lit

LIBRARY

UNIVERSITY OP
CALIFORNIA

f y A£

/

DOCUMENTS DtFA.-:Ti«EI\
JUn £8 bo5

LIBRARY

OF THE

AGRICULTURAL SOILS

OF THE

CAPE COLONY

BT

CHARLES P. t.TURITZ, M.A., D.Sc., F.I.C.,

Senior Government Analyst of Cape Colony; President of the Chemical and

Agricultural Section of the S.A. Association for the Advancement of Science ;

Past President of the Cape Chemical Society, <&c., &c.

CAPE TOWN:

CAPE TIMES LIMITED, GOVERNMENT PRINTERS.
1909.

PREFATORY NOTE.

The investigations into the nature and composition, from an agricul-
tural point of view, of the soils of the Cape Colony, suggested and initiated
by the writer and carried on in the Government Analytical Laboratory
under his direction, constitute the only work of the character ever under-
taken in this country. Previously to their commencement a few isolated
analyses of Colonial soils had been made, some by Professor Halm — amongst
which are those quoted on page 124 — and some by the writer himself, but
these were scanty and disconnected. Never before had any systematic investi-
gational work of the kind here recorded been attempted. Even now, having
regard to the Colony's vast extent, the number of soils examined is all too
small, but the work has now reached a stage from which it is well to pause
and look back before proceeding further. The results hitherto obtained
are therefore collated here in the hope that the utility of the work may
thus be established and a desire created for its continuance and extension.

C. F. J.

Government Analytical Laboratory,

Cape Town, 12th November, 1907.

061

CONTENTS.

PAGE.

PART I. — INTRODUCTORY . . . . . . . . . . . . . . . . 1

Former theories of plant nutrition . . . . . . . . . . . . 1

Available plant-food constituents . . . . . . . . . . . . 2

Phosphate starvation and bone disease . . . . . . . . . . 3

The origin of the investigations . . . . . . . . . . . . 4

PART II. — AGRICULTURAL CHEMICAL METHODS . . . . . . . . . . 5

Divergent views on soil analysis . . . . . . . . . . . . 5

Uniformity of method desirable .... 5

Selection of samples ........ <>

Preparation in the laboratory

Chemical analysis, standard methods .... 7

Extraction of inorganic plant-food

Soil treatment with solvents . . . . . . . . . . . . . . 9

Differentiation of plant-food constituents . . . . . . . . . . 9

Choice of a solvent . . . . . . . . . . . . . . . . 12

Extraction with standard Hydrochloric acid . . . . . . . . 13

German Experiment Station method ...... 15

Maeroker's method . . . . . . . . . . 15

Hilgard's method . . . . . . . . . . . . . . . . . . 10

Extraction with Citric acid . . . . . . . . . . . . . . 10

Interpretation of results . . . . . . . . . . 17

' Field operations . . . . . . . . . . . . . . 9

Personnel and materiel . . . . . . . . . . . . . . . . 19

Means of procuring samples . . . . . . . . . . . . 19

PART III. — RESULTS OF CHEMICAL ANALYSES . . . . . . . . 21

Albany .. 23

Albert 24

Aliwal North . . 24

Barkly West 25

Beaufort West 28

Bredasdorp 28

Butterworth 32

Caledon 33

Cape

Carnarvon . . . . . . . . .... . . . . . . . . 45

Cathcart 46

Ceres . . . . . . . . . . . . . . . . . . . . 50

Clanwilliam . . . . . . . . . . . . . . . . . . 51

Colesberg 52

Elliot-Slang River . . . . . . . . . . . . . . . 53

Fort Beaufort 54

George . . . . . . . . . . . . . . . . . . . . 55

Gordonia . . . . 60

Graaff-Reinet 60

vi

PART III. — RESULTS or CHEMICAL ANALYSES — (contd.)

PAGE.

Hanover . . . . . . . . . . . . . . . . . . . . 61

Herbert 62

Hopetown . . . . . . . . . . . . . . . . . . . . 63

Humansdorp . . . . . . . . . . : . . . . . . . 63

Idutywa . . . . . . . . . . . . . . . . . . . . 64

Kenhardt 65

Kiinberley . . . . . . . . . . . . . . . . . . . . 65

King William's Town . . . . . . . . . . . . . . . . 67

Knysna 67

Komgha

Ladismith . . . . . . . . . . . . . . . . . . . . 74

Maclear 77

Mafeking . . . . . . . . . . . . . . . . . . . . 77

Malmesbury . . . . . . . . . . . . . . . . . . 78

Middelburg 89

Mossel Bay 90

Mount Currie 94

Mount Frere . . . . . . . . . . . . 95

Namaqualand . . . . . . . . . . . . . . . . . . 96

Oudtshoorn . . . . 96

Paarl 99

Piquetberg 107

Port St. John's 107

Prieska 108

Queenstown . . . . . . . . . . . . . . . . . . 109

Richmond .. .. .. .. .. .. .. .. .. .. 110

Riversdale Ill

Robertson .. .. .. .. .. .. .. .. .. .. 115

St. Mark's 121

Somerset East 121

Stellenbosch 122

Steynsburg . . . . . . . . . . . . . . . . . . . . 126

Stockenstrom 128

Swellendam 129

Tulbagh 134

Uitenhage . . . . . . . . . . . . . . . . . . . . 136

Umtata 137

Umzimkulu . . . . . . . . . . . . . . . . . . 137

Uniondale 138

Victoria East 141

Vryburg 141

Willowmore .. .. .. .. .. .. .. .. .. 145

Willowvale 145

Wodehouse 146

Worcester ..147

PART IV.— COMPARISON OF SOIL EXTRACTION METHODS 153

PART V. — GEOLOGICAL RELATIONS AND PLANT FOOD . . . J . . . 157

Aid derived from geological survey . . . . . . . . . . . . 157

Selection of geologically-typical samples . . . . . . . . . . 157

Malmesbury series soils . . . . . . . . . . . . . . . . 157

Effect of sandstone on Malmesbury soils . . . . . . . . . . 159

Granitic soils . . . . . . . . . , . . . . . . . . 159

Cango beds soils . . . . . . . . . . . . . . . . . . 161

Campbell Rand soils . . . . . . . . . . . . . . . . 161

Soils derived from Stow's " olive shales " . . . . . . . . . . 162

Soils of the Pniel volcanic series . . . . . . . . . . . . 163

Table Mountain sandstone soils . . . . . . . . . . . . 163

Bokkeveld soils 164

Soils of the Table Mountain-Bokkeveld junction 165

Witteberg soils 167

Dwyka Soils 168

Soils of the Burghersdorp beds and Stormberg series 168

Uitenhage series soils . . . . . . . . . . . . . . . . 169

Soils of the sand downs . . . . . . . . . . . . . . 170

Alluvial Silts and River deposits 170

Vll

PAGE.

PART VI. — ALKALINITY OF SOILS . . . . . . . . . . . . . . 172

Effect of rain on soil texture and composition . . . . . . . . 172

Irrigation and soil salinity . . . . . . . . . . . . . . 172

Method of taking samples for alkalinity tests . . . . . . . . 173

Distribution of salts in soil mass . . . . . . . . . . . . 173

Hilgard's and Leather's experiments .. .. .. .. .. .. 173

Methods of analysis . . . . . . . . . . . . . . . . 174

Analytical results : 175

Division of Herbert : Douglas . . . . . . . . . . . . 176

Colesberg : Oorlogspoort .. .. .. .. .. 177

Britstown : Houwwater .. .. .. .. .. 177

Steynsburg : Thebus .. .. .. .. 180

Robertson : Experiment Station . . . . . . . ; 184

Carnarvon : Van Wyk's Vlei 185

PART VII. — PHYSICAL COMPOSITION OF SOILS . . . . . . . . . . 188

Fertility and productiveness . . . . . . . . . . . . . . 188

Connection of chemical composition and texture in soils . . . . . . 188

Varieties of soil classification . . . . . . . . . . . . . . 189

The grading of soil particles . . . . . . . . . . . . . . 190

Importance of silt and clay . . . . . . . . . . . . . . 191

Methods of mechanical analysis . . . . . . . . . . . . 193

Analytical results . . . . . . . . . . . . . . . . . . 195

Geological relationship and soil texture . . . . . . . . . . 195

Mechanical analyses of various soils . . . . . . . . . . . . 197

Stellenbosch Division : Elsenberg . . . . . . . . . . . . 198

Paarl Division 198

Steynsburg Division : Thebus 199

Robertson Division : Experiment Station . . . . . . . . . . 199

Prieska Division : Buchuberg. . . . . . . . . . . . . 201

Cape Division : Koeberg . . . . . . . . . . . . . . 203

CONCLUDING OBSERVATIONS . 205

A STUDY OF THE AGRICULTURAL SOILS OF
CAPE COLONY.

PART I.— INTRODUCTORY.

Van Helmont, three centuries ago, demonstrated, to his own satis-
faction, that water was the sole source of plant food. Digby, fifty yeara
later, ascribed the nutrition of plants to a mysterious principle in the air,
and refused to consider water as anything more than a mere vehicle for
the conveyance of this principle. Yet another half-century, and both
these theories were rejected by Tull, who put forth the view that the soil
particles constituted all the nutriment that the plant needed. He held,
however, that, in order to* be available as nutriment, these particles had
to be extremely minute. Scarce a century has passed since Thaer promul-
gated the doctrine that the source of plant food was nothing other than
humus, — a doctrine that commenced to be undermined by Saussure while
still in its infancy, and was completely pulverised by Liebig when pro-
pounding his mineral theory half a century later. According to Liebig
not the organic, but the inorganic, or mineral constituents of the soil served
to build up the vegetable structure.

Fifty years have passed since the day of the great German chemist',
and at present the tendency is to lay emphasis on no single one of the
various views alluded to above, but to regard them all as partially true.
The worth of the mineral constituents of plant food is almost universally
accepted ; but not to the entire rejection of organic matter as a valuable
agent in determining a soil's fertility. The mechanical subdivision of the
soil by sifting, sedimentation, or elutriation, is the modern counterpart
of the views put forward by Tull. In the fixation of atmospheric nitro-
gen by bacteria we may trace a reflection of Digby 's theory; and the re-
cent investigations of Whitney and others in the United States of America
have led them to conclusions which, in an embryonic state, lay hidden
within the water theory of Van Helmont.

It is, nevertheless, erroneous to affirm — however much it may seem
to be the case — that modern investigators have reverted to the opinions
held by their predecessors two or three centuries earlier. It is not only in
connection with agriculture that chemical science may appear to have
moved along a circular pathway, and yet the striking theories which have
resulted from, the investigations into the nature of radio-activity, for in-
•tance, do not by any means imply that the scientists of our day have re-
treated to the alchemists' notions respecting the philosopher's stone and
the transmutation of metals. We may be trending back to the same ver-

tical Hue of thought, so to speak, without proceeding along the same dead
level: it is not a circle but a helix that marks out the path whereby
science has led its students. It is of importance to recollect this, for, un-
less we do so, the lay mind may be misled into thinking that, because in
some respects Liebig was wrong, Van Helmont was necessarily in all
respects right. The direct consequence of Liebig's mineral theory was the
view that the fertility, if not the productiveness, of any soil could be in-
ferred from a complete chemical analysis of that soil. This view, it is on ail
hands conceded, overstates the facts, but to conclude, from this concession,
that chemical analysis of soils, no matter how made and applied, is value-
less, would be a betrayal of ignorance in regard to the general tendency
of all subsequent research.

Assuming, without argument for the present, that chemical analyses
of soils have some value, the way along which such analyses should pro-
ceed may very pertinently be discussed. But before this can even be con-
sidered it becomes necessary to enquire more closely into the method by
which nutriment is conveyed to plants. Van Helmont, as has been seen,
attributed this function wholly to water, a view which was vigorously con-
tested by Liebig, who ascribed the preparation of the food of plants from
the mineral constituents of the soil to the acid excretions of the roots
themselves. As hinted above, there is a strong inclination in some quar-
ters to-day to accept the substantial correctness of Van Helmont's theory,
at all events in an expanded form.

A considerable recession has taken place from some of the views
which found currency when first Liebig's mineral theory became widely
accepted. Thus the complete chemical analysis of a soil has, for more
than a score of years, ceased to be reckoned an index of fertility. Chemists
of a later day, alive to the fact that the roots of plants are not able to ab-
sorb from the soil all the plant food constituents there present, began to-
modify the character of the solvents used in the laboratory for extracting
this nutriment: they accordingly directed their endeavours towards ex-
tracting from the soil only such constituents, and those only in such quan-
tities, as the plant rootlets are actually capable of withdrawing. The
proportions of plant food constituents thus removed were said to be pre-
sent in the soil in a form " available " to plants. Whatever the quantity
of plant food there may be in any soil, unless it was present in a form
available to the plant, it could as well, for all practical purposes, be non-
existent. As soon as this view began to be held, it became necessary, if
the analysis of a soil was to have any value for agriculturists, to employ
weaker solvents than those at first adopted, — solvents, obviously, that
would simulate the action of the plants themselves. That was the proce-
dure which Professor Cossa urged in 1866.

Now arose the problem of finding suitable solvents: various proposals
were made, and, for the purpose of putting their adaptability to the test,
constant and prolonged comparisons between laboratory and field experi-
ments were rendered necessary. A full discussion of these would be need-
less here; suffice it to say that the German experiment stations adopted
Hydrochloric acid diluted to a certain degree, and left in contact with a
specified weight of soil, for a definite period of time, at a fixed tempera-
ture. At Halle a method of determining a-vailable phosphoric oxide in
soils, by extracting it with citric acid solution, was introduced, and has
since become generally recognised for that purpose, the whole scientific
world over. An extension of this method was proposed by Dr. Bernard
Dyer in 1894, and is now usually associated with his name. Of late years
American investigators have begun to employ very small proportions of
pure water, basing their practice on the view that the soil water is the

only true medium for the conveyance of mineral salts to the plant.* The
variable root-action of different plants rendered all the more complex the
problem of discovering a solvent that would suit all cases.

Ere long one of the fundamental principles of Toll's theory was ad-
mitted to due recognition. Obviously no plant food constituents in a soil
•could be considered as "available" when present in compact impermeable
masses : hence became clear the advisability of agreement with regard to
ihe size of soil particles to be included in any determination. The result
was a consensus of opinion that 110 plant food substances in any soil could
be held to be available unless they were contained in soil particles not ex-
ceeding a definite maximum diameter. It therefore became customary,
when analysing a soil for agricultural purposes, to pass it through a sieve
of standard mesh before proceeding with the chemical analysis, disre-
garding for that purpose all that failed to pass through. f

Scientific progress with regard to the chemistry of agricultural soila
had about reached this stage- when the circumstances occurred which led
to the writer's undertaking the systematic investigation of the soils of the
€ape Colony. How this came about may be very briefly explained.

In 1890 the writer was engaged, in connection with the Fellowship
granted him by the University of the Cape of Good Hope, in the analysis
of specimens of oathay from various parts of the Colony, and, in comment-
ing upon the results of his investigations, made use of the following words :

" Another noteworthy fact, one which our farmers should bear in rnind, i.s this, that
the oathay from the Western Province contains an exceptionally small amount of lime,
this being due entirely to a deficiency of lime in the soil, a deficiency which it becomes
absolutely necessary to meet by the artificial application of lime manures to the soil. . . .
The sample of oathay from Port Alfred contained an exceedingly small quantity of
phosphoric oxide, and to this, doubtless, is due the poor appearance of the sample, its
weight being about one-fourth, or less, of what it should have been under normal
conditions. I have been informed that the Veterinary Surgeon holds the opinion that the
cattle diseases in this district are probably due to a deficiency of phosphates in the food,
and the results of these investigations appear to confirm that view ; in fact, judging from
the analyses of the plants only, I should say that the soil of the Colony generally appears
"to be rather poor in phosphates."

In the case just quoted the oats had been found to contain so little
phosphoric oxide as to cause wonder that such dwarfed and sickly-looking
plants had reached maturity at all, and the natural conclusion at the time
was that the soil upon which the cereal crop was grown was likewise de-
ficient in phosphates, and that, consequently, the opinion, said to have
been expressed by the Government Veterinary Surgeon, was, at any rate,
likely to be correct. In addressing a public meeting shortly afterwards,
the writer took occasion again to refer to the subject, and then observed :
" I do not regard the matter as settled satisfactorily, and commend it to
the attention of the Grahamstown Agricultural School, hoping that ere
long proper investigations will be made and the mystery cleared up."

* " It seems entirely unnecessary, in studying the question of the nutrition and yield
of crops, to introduce artificial digestion media known to attack minerals very slightly
soluble in water, while it seems perfectly logical to accept the nutrient solution as it
•exists in the soil as the basis for the support of plant life, and to investigate the question
along this line. In other words, it has seemed best to consider the soil as a culture medium
containing a nutrient solution — that is, to regard the soil moisture as a proper and sufficient
medium for the feeding of plants, and the soil as a reservoir and distributing agent for this
solution." (M. Whitney & F. K. Cameron : " The chemistry of the soil as related to crop
production." 1903, p. 15).

"The delivery of water from the soil to the plant must receive more consideration in
iuture chemical studies of soils for the reason that it is the mechanism by which the
mineral nutrients are supplied to the plant, and it is evident that if the delivery be email
ihe plant will suffer not only for water itself, but also for the mineral food which the
water supplies, even though ample quantities may be present in the soil solution, and in
Tvhat would usually be regarded as readily available form " (7fo'<?., p. 6P .)

t But see also page 13.

About a year later the Government Analytic*/ Laboratory was
placed under my charge, and the opportunity was thus afforded me of put-
ting the views above expressed into practice. Almost immediately some
samples of soil, from the part of the Colony particularly referred to, were
received for analysis, and then was confirmed the opinion previously put
forward that the soil in the neighbourhood was very poor in phosphates,
and that, in consequence of this, the fodder there grown would probably
be found similarly deficient, and would influence animals feeding upon it
in a very serious manner, inasmuch as phosphates were absolutely neces-
sary for the formation of bone material. Further investigations were
forthwith put in hand : other soils from the Albany Division, and from
the neighbouring Division of Humansdorp, were subjected to analysis, and
the results served to lend additional confirmation to my previous views.
The use of phosphatic fertilisers for the districts named was therefore
urged, and it was also suggested that farmers round about should learn
to utilise bone manure as generally as possible.

The time seemed to have come for more decided action; representa-
tion was accordingly made to the Government that the primary cause of
the deficiency of phosphates in the crops lay in an almost entire absence
of the former from the soil, and that this deficiency, it was at least likely,
if not absolutely certain, was at the root of all the " lamziekte " that those
districts of the Colony were belag plagued with. It was, moreover,
pointed out by me how beneficial fuller information respecting the various
soils of the Colony would be, and investigations, with the object of elicit-
ing some such information were recommended to be undertaken without
delay. Assurances of support on the part of the Government were readily
given, and in 1893 operations were commenced. The north-eastern por-
tion of the Cape Division was first visited, then the Malmesbury Division :
after that the Caledon and Bredasdorp Divisions were taken in hand;
later on those of Robertson, Swellendam, Riversdale, and Mossel Bay, and
ultimately George, Knysna, Uniondale, and Oudtshoorn ; while in the
meanwhile, in compliance with reiterated requests, a visit was paid to some
of the Eastern Province Divisions before the work was finally stopped by
the war, since when it has not yet been resumed. Along its entire
course, however, the systematic investigation of the soils of certain districts
has been supplemented by numerous casual analyses of specimens taken, as
occasion and opportunity offered, from several localities not included up to
the present in the regular soil survey.

With regard to the organisation of a comprehensive scheme of soil in-
vestigation, snch as that of which we felt the desirability, it is but sel-
dom that an opportunity offers for the satisfactory carrying out of one's
ideals; the limitations of funds, of qualified assistants, and of equipment
in this case precluded all hope of following up the preconceived plans
otherwise thai in a slow and plodding way. The work was commenced
by the writer entirely single-handed, but as, with the building up of a
then newly-established laboratory, calls on his time increased, a great por-
tion of it had, of necessity, to be performed by deputy; never at any
time was it possible to spare more than one person for the investigation,
so that all that is here recorded may be taken as representing, from first
to last, including the minutest details, one man's work during the period
comprehended therein.

PART II.— AGRICULTURAL CHEMICAL METHODS.

It was assumed, without argument, on a previous page, that the chemi-
cal analysis of soils had some value. Just what that value is, and on what
conditions it depends*, may to a limited extent be discussed later; but it
may here be convenient to allude to the fact that there are those who
would deny that it possesses any practical utility whatever. Thus Wal-
lace* remarks : —

" No analyst, using the ordinary processes for soil analysis, can determine whether or
not such infinitesimal amounts as are required by the crops are presenti or are not present
in an available form in the soil."

Professor Whitney f again observes : —

" It appears, further, that practically all soils contain sufficient plant-food for good
crop yield ; that this supply will be indefinitely maintained, and that this actual yield of
plants adapted to the soil depends mainly, under favourable climatic conditions, upon the
cultural methods and suitable crop rotation ...... and that a chemical analysis

of a soil, even by these extremely delicate and sensitive methods, will in itself give no
indication of the fertility of this soil or of the probable yield of a crop, and it seems
probable that this can only be determined, if at all, by physical methods, as it lies in the
domain of soil physics."

On the other hand, Professor Hilgard§ has the following : —

" Whitney (Bull. 22, U.S. Bureau of Soils) claims, on the basis of a large number of
(three minute) extractions of soils made with distilled water, that these solutions are
essentially of the same composition in all soils ; that all soils contain enough plant-food to
produce crops indefinitely, and that the differences in production are due wholly to
differences in the moisture supply which he claims is, aside from climate, the only govern-
ing factor in plant growth. The tables of analytical results given in Bull. 22 fail to
sustain the first contention ; the second is pointedly contradicted both by practical
experience and by thousands of cumulative culture experiments made by scientific observers;
the third fails with the second, except, of course, in so far as an adequate supply of moisture
is known to be an absolute condition both of plant growth and the utilization of plant-
food. It is, moreover, well known that it is not water alone, but water impregnated more
or less with humic and carbonic acids, that is the active solvent surrounding the plant
-root."

Granting that the chemico-agricultural analysis of soils is not an ab-
solute but a relative method of estimating their fertility — a point that
will be reverted to later — it will at o-nce be seen that such a survey of
this Colony's soils as was proposed necessitated the adoption of uniform,
or, at least, of comparable methods of procedure throughout. In order,
furthermore, to' enable comparisons to be drawn between our results and
those obtained in other parts of the world, or with those arrived at by
other analysts in this country, it becomes requisite fully to describe the
actual mode of procedure adopted. From the first it was realised that the
investigations would extend over many years, even without the unfoire-
seen interruptions which were subsequently caused by war and financial
depression. It was practically certain also that, with the advance of
scientific knowledge, there would be improvements in manipulation during
the course of the survey, and it was therefore recognised as most desir-
able to adopt stated methods at the o<utset, and to adhere to them in all
essentials throughout the whole series, a course without which results
could not be strictly comparable amongst themselves. The great obstacle
in the way of this was the fundamental differences which exist between
agricultural methods of operation in this country and those practised in
the northern hemisphere, where alone opportunity has hitherto offered of
comparing laboratory results with field experiments. To do this as it
really needed to be done would have involved years of preliminary work;
not to do it might waste years in the employment of unsuitable analyti-

* "Rural Economy and Agriculture of Australia and New Zealand." p. 169.

"The Chemistry of the Soil as related to crop production," 1903, p. 64.
$ " Soils : Their formation, properties, composition, and relations to climate and
plant growth," 1906, pp. 321, 322.

cai methods : some sort of compromise accordingly became inevitable, and &
standard method was adopted and compared from time to time with one or
two other methods, which will also be fully delineated in the sequel. How
well the method taken as a standard has fitted in with the climatic condi-
tions of the country is shown by the repeated agreement between labora-
tory results and the recorded experiences of the practical farmer. The
alternative course which was open was to adopt improvements in analytical
methods as they arose, and thus ensure in all respects up-to-date modes of
manipulation, but coupled with an inevitable unconformity to the earlier
results of the series.

In describing the methods employed the collection of soil samples
claims first attention. Dr. Wiley rightly says: — "The sole object in tak-
ing a sample of soil should be to have it representative of the type of soils
to which it belongs. Every precaution should be observed to have each
sample measure up to that standard. The physical and chemical analyses
of soils are long and tedious processes, and are entirely too costly to be ap-
plied to samples which are not representative." It will be understood that
analyses of soils from cultivated fields, whatever value they may have for
the owner of that particular plot, cannot, as a rule, be comprised in an
investigation relating to the country at large, or even to the surrounding
district ; for a soil that is modified by repeated cropping and manuring,
and altered by human agency, cannot be looked upon as typical or repre-
sentative of any extended area. The scheme of investigation which the
writer had in view aimed, for the greater part, at ascertaining the compara-
tive agricultural values of soils over wide tracts of country, and it was ac-
cordingly sought to collect, as far as possible, only virgin soils which had
not been subjected to modifying influences; in a country such as this, a
most difficult and well-nigh hopeless task.

In the actual taking of the samples for ordinary agricultural chemical
analysis the practice has been as follows : — The surface of the selected spot
;s lightly scraped with a trowel or other sharp tool, so as to clear away the
top growth and surface accumulations of semi-decayed vegetable matter not
forming a part of the true soil. A hole with vertical sides is then dug to
a depth of twelve inches, and from the sides of this hole thin slices of soil,
aggregating from five to ten pounds in weight, are removed, in as uniform
a manner as possible, and placed in a suitable bag of canvas, cloth, or other
impervious material through which the fine soil will not sift out on trans-
port during conveyance to the laboratory.

On arrival in the laboratory the sample is allowed to become air-dry
by being spread out in a thin layer, protected from dust, for some days.
Any lumps of soil that there may be are then gently broken down by means
of a wooden pestle or rolling-pin, care being specially taken to avoid
crushing any mineral fragments.

Subsequent treatment of the soil thus prepared varies according as a
complete mechanical or a more purely chemical analysis is desired. It had
been intended to combine the former with the latter in systematically in-
vestigating the country's soils, but the exigencies of circumstances rendered
such a course, though desirable, quite impracticable. In . connection with
the soil survey, therefore, only a restricted form of mechanical analysis waa
adopted, and the few instances in which the more complete mechanical dif-
ferentiation was made will be dealt with separately at a later stage, where
the method employed for that purpose will be outlined. At this point we
shall confine ourselves to the operations more closely connected with the
purely chemical analysis.

As previously remarked, the importance of finely divided particles in,
a soil, at first vaguely grasped by Tull, has been definitely recognised dur-
ing the last quarter-century. It is now understood that, for agronomic

purposes, the finer the particles which compose a soil, within limits, castcri*
paribus, the better the soil will be for general agricultural purposes. Hence
although, as above remarked, a- complete mechanical analysis was not, as a
rule, included in these investigations, a partial sifting out of the soil has
invariably preceded the chemical analysis proper, so that any plant food
material that may have been locked up within the coarser soil particles has
been left out of account in reckoning up the proportion of "available"
plant food present in the soil.

In preparing a soil for the chemical analysis, the " field sample " is
first of all freed from pebbles by dry-sifting a sufficient quantity through
a brass-bottomed sieve with circular perforations 3 mm. in diameter. All
that passes through this sieve is denominated " true soil/' 250 grammes of
this "true soil " are placed in a porcelain dish; about half a litre of dis-
tilled water is poured on, and the dish is allowed to digest on a water bath
for at least two hours, the contents being stirred at frequent intervals.
When the soil has thus been sufficiently softened, it is sifted in the follow-
ing manner : — A special sieve (Kahl's " Messingdrahtsieb " No. 50), having
meshes of diagonal measurement *45 to '50 mm. and '35 to '39 laterally, is
held over a dish containing distilled water; the moistened soil is placed in
the sieve, and the latter is then immersed with the bottom about an inch
or more below the water level in the dish. By the aid of a small brush the
soil is now stirred until particles no longer pass through, after which the
residue in the sieve is thoroughly washed with distilled water. The ma-
terial which passes through the sieve is dried, together with the wash-water,,
on a water bath, and the weight is then recorded in percentage of the*
"field sample." The residue thus obtained by evaporation of the sifted
soil and water is called " fine earth," and is subsequently utilised, accord-
ing to the standard method of chemical analysis adopted, for the determina-
tion of lime, potash, and phosphoric oxide.

Another sufficient portion of the " field sample " is dry-sifted through
a brass-bottomed sieve with circular perforations 1 mm. in diameter. That
which passes through is termed "earth," and is employed in determining
the amounts of moisture, organic matter, chlorine, and nitrogen.

The chemical analysis proper has now to be considered. As already
indicated, the proportions, of soil-constituents mentioned in all the subse-
quent tables of analysis by the standard method are calculated and stated
in terms as below : —

" Moisture " ]

" Organic matter " ( .

" Chlorine " ( m PercentaSe of earth."

" Nitrogen " )

44 Lime " )

" Potash " V in percentage of " fine earth."

44 Phosphoric oxide" )

" Fine earth " in percentage of " field sample."

Details are appended of the methods followed in determining these
constituents. Two separate portions of soil are used, prepared as already
described (the " earth " by sifting through a 1 mm. sieve, the " fine earth "
by sifting through what is practically a \ mm. sieve) ; these are, after
treatment, allowed to become air-dry and bottled.

Moisture. — Ten grammes of " earth " are weighed in a tared platinum
crucible, heated in an air bath from three to four hours, at a temperature
from 105° C. to 110° C., allowed to cool in a desiccator, and rapidly
weighed again.

Organic and volatile matter. — The crucible containing the perfectly dry
soil used in the determination of moisture is heated to redness until all
organic matter has been burnt away. It is then cooled in a desiccator and

weighed, the difference in weight indicating organic matter, water of com-
bination, ammonium salts, etc. If the soil contain appreciable quantities
of carbonates, the contents of the crucible, before weighing, are moistened
with a few drops of a saturated solution of ammonium carbonate, dried,
and heated to dull redness, after which the crucible is placed in the desic-
cator to cool previous to weighing.

Chlorine. — Fifty grammes of " earth " are boiled in a marked 500 c.c.
flask, with about 300 c.c. of distilled water, for half an hour. Cold distilled
water is added until the level nearly reaches the mark, and after cooling
it is filled up, shaken thoroughly, and filtered, by means of a Berkefeld
candle filter, into a flask holding about 500 c.c., whence 50 c.c. are removed
for titration with T^ Silver nitrate, using Chlorine-free Potassium chromate
as indicator.

Nitrogen. — Five grammes of " earth " are placed in a 150 c.c. oxidation
flask; a globule of mercury weighing '8 or '9 of a gramme is dropped in,
and a small fragment of solid paraffin to prevent frothing. 20 c.c. of con-
centrated sulphuric acid, free from nitrogen and nitrogen compounds, are
then added. The flask is closed with a loosely fitting glass stopper, and the
mixture heated over a Bunsen flame, gently at first, but more strongly after-
wards, until colourless. When cool, the stopper is removed, and the flask is
cautiously three-fourths filled with distilled water. It is then gently
shaken, and the contents are washed into a 600 c.c. Erlenmeyer flask, wash-
ing the oxidation flask three times, each time with about 30 c.c. of distilled
water : a small quantity of granulated zinc is added, and 75 c.c. of soda-lye,
prepared by dissolving 35 grammes of potassium sulphide in 200 c.c. of
water, and adding the solution to 1150 c.c. of caustic soda solution of 1*375
specific gravity. The distillation flask is closed with a rubber stopper
carrying a bulb tube which is bent above the bulb at a sharp angle and
terminates in a condensing tube 70 cm. long. This is connected, by means
of rubber tubing, with another glass tube, dipping into 20 c.c. £ Sulphuric
acid mixed with 50 c.c. jrf water in an Erlenmeyer flask as receiver. After
the mixture in the distillation flask has been heated to boiling, the flame
is so regulated that in 15 minutes' time the liquid in the receiver is at
boiling point. The boiling is then continued for an additional five minutes.
When the distillation is complete the contents of the receiver are washed
over into a 250 c.c. flask, filled up, and 50 c.c. are titrated with ^ Barium
hydrate, using Rosolic acid as indicator.

EXTRACTION OF THE INORGANIC PLANT FOOD CONSTITUENTS FROM THE BOIL.

The chemical analysis of a soil, if the object be to gain information
with respect to its agricultural qualifications, differs widely from what it
would be if the aim were to settle mineralogical or geological questions.
In the latter event the endeavour would naturally be to ascertain the
total quantities of mineral constituents present in the soil, quite irrespec-
tive of their adaptability as plant nutrients; but in the former case the
determination of these aggregates may be of very trifling assistance, inas-
much as it is highly probable that, of the total quantities of plant food
constituents present, only small proportions may be available to plants.

The chemist, desiring to estimate the amount of available plant food
in a soil, generally tries to imitate, as closely as possible, the action of the
plant itself upon the soil particles. A close copy of this natural action, is
difficult almost to the point of impracticability, and the difficulty is not
rendered any the less by the fact that certain plants have a greater facility
of withdrawing the nutrient compounds from the soil than others have.
It is essential to note the distinction between " plant food constituents "
and " plant food." A plant food constituent remains such no matter where
or how it may exist; it only becomes plant food when it is present in the
soil, and then only when it is present there in the very condition in which

the plant growing upon that soil can withdraw it and turn it to ite own
account. The- expression " plant food constituent " is accordingly used
throughout these pages with th© specific meaning above implied, and to the
term " plant food " likewise a very definite signification is assigned. Bear-
ing this in mind, it will directly follow that a chemical analysis of the soil,
in order to be of value to the farming community, should tell, not the quan-
tities of plant food constituents present, but the proportion of plant food.
It becomes obvious, then, that we are to distinguish between at least two
kinds of chemical analyses of soil, one of which supplies the farmer with
information of value, while the other does not; it will, however, be more
convenient to look upon soil analyses as capable of sub-division into three
classes or grades. First of all the plant food constituents may be present
in the soil in such a condition as to be quite incapable of being absorbed
by the plant; remembering that we are considering the subject from the
agriculturist's standpoint, we may be justified in calling these the plant
food constituents of the third or lowest grade. The chemist who wishes
to include these in his determination of the total quantity of plant food
constituents in the soil, needs to employ the strongest chemical reagents,
or the energetic action of fluxes at a high temperature in order to attain
his object, for the plant food constituents of the third grade are usually
silicates or aluminates, and do not respond to any less radical treatment :
in any case they arc not plant food. Needless to say, that the acids gener-
ally employed by the agricultural chemist fail to extract these compounds
from the soil, and hence do not give the utterly misleading results occa-
sionally attributed to them. The first and second grades of plant food con-
stituents differ from the third in being available for plants ; that is to say,
they are actually plant food. These are extractible from the soil by
mineral acids, such as hydrochloric acid. The plant food constituent® of
the first grade are readily, or immediately, available to the plants, and the
chemist can extract them from the soil by means of water or weak organic
acids, such as a dilute solution of citric acid ; those of the second grade are
less soluble, less readily available for plants, and may be extracted in the
laboratory by strong mineral acids, but not by water or weak organic acids.
They are not immediately removed by the crops, but continue in the soil
as a " reserve stock" — a term that we shall have occasion to use' again;
let it be remembered, therefore, that whenever it is employed in the course
of these remarks it signifies plant food of the second grade.

Thus we have these three grades of plant food constituents in the
soil : —

I. Soluble in water and in weak organic acids :

Immediately or readily available for plants.
II. Soluble in strong mineral acids :

Available for plants only as a reserve stock.

III. Insoluble in ordinary acids, and extracted only by fusion or by

specially powerful reagents like hydrofluoric acid.
Not available for plants.*

Obviously the agriculturist has little, if any, interest in non-available
plant food constituents, and chemical analyses — be they of a single soil
sample, or of a whole series of representative soils from various parts of
the country — which give only figures showing the amounts of plant food

* The following figures, adapted from Bulletin No. 41 of the Minnesota Agricultural
Experiment Station, illustrate the fact that soils contain considerable amounts of
plant-food constituents which are not available for plants : —

"Wheat Soil. Heavy Clay Soil. Grass and Grain Soil.
Grade II. Grade III. Grade II. Grade III. Grade II. Grade III.

Lime 2'44 "36 '48 '16 '51 '35

Potash -54 2-18 '21 3'46 '30 1-45

Phosphoric Oxide -38 -12 -08 -23 -05

la the case of the clay soil, it will be noticed, as much as 96% of the potash was in a form
unavailable for plants.

10

constituents present, and afford no indication as to the quantity of available
plant food — whatever interest they may possess for the geologist — have
none for the farmer. It is true that, by mechanical disintegration and
chemical decomposition, plant food constituents of the third grade may ul-
timately become available, as they very slowly change into those of the
second, but the process is so gradual as practically never to have any value
for the generation in occupancy. The constituents of grades I and II, on
the other hand, possess great interest for the farmer ; they comprise the
actual plant food : the former affect the land's immediate productiveness,
the latter its permanent value. In an investigation such as that under re-
view, then, we may dismiss grade III from further consideration. The
first grade, or immediately available plant food, is of less account than in a
country where agricultural lands are usually held upon short tenancy;
and, in any case, where the object is to gain information respecting the
agricultural potentialities of extended areas, insufficient date would be>
afforded by determinations of the readily available plant food in the soil :
individual farmers may profit by such investigations, but even then, to be
of lasting value, they would have to be constantly repeated, and upon very
much more comprehensive lines than the utmost range of practicability^
if they are to meet all needs.

So we are brought to this conclusion, that, if a chemical investigation
of a country's soils is to be made, it is the reserve stock of plant food in
the soil (that is to say, the constituents of the second grade) that calls for
first attention. They are continually changing into plant food of the first
grade, and are being removed from the soil by plants very much more
rapidly than they can be produced from the third grade constituents. It
may be of some value, either incidentally, in certain cases, or subsequently
to a thorough investigation of the reserve stock of plant food in the soils
of the whole country, to inquire into the conditions and quantities of the
other two grades as well, but the immediate and pressing necessity is to1
proceed with all speed along the line just indicated as demanding prime
attention.

Different plants have, as already remarked, different absorbing powers,
and hence we may almost say that the immediately available plant food in
a soil varies in amount with the crop to be cultivated. If, therefore, we
ascertain the maximum amount of plant food that can become available to
any plant from a given soil we shall gauge the utmost limits of that soil'a
natural fertility, as far as it is affected by chemical considerations. If that
be his purpose, the chemist should employ as a solvent not one which will
simulate a single crop of weak extractive power, that is to say, a solvent
that will take out little more than the constituents of grade I, leaving the
bulk of those of grade II untouched. Of course, he is well aware that the
soil which contains a large stock of plant food of the first class will in any
case be chemically rich ; it will therefore be fertile, provided that other
factors — the physical character of the soil, the supply of water, and so on —
are not defective. But he knofws equally well that if the maximum of
available constituents fall below a certain limit, the soil must of necessity
be poor, whatever other conditions may be favourable. Hence the first ob-
ject of a systematic soil survey along chemical lines should be to ascertain
the maximum proportions of plant food likely to be available to average
crops within a reasonable period of time in the area under examination.
For this reason, especially in a country like this, consisting of immense
tracts of virgin soil, amongst which there are many poor soils, the writer
considers that a great deal more information will be gained by the extrac-
tion of soils by means of Hydrochloric acid than by applying Dyer's citrio
acid solution. The latter may be employed subsequently, as supplemen-
tary, but the former method should certainly take precedence.

11

Professor Hilgard makes some incisive observations with regard to the*
threefold differentiation of plant food constituents in soils, which bear
directly on what we have said above. In order to appreciate his remark*
i:i this connection, some extracts, leading up to them, from his work al-
ready quoted are essential. He says:— ,

" While the obvious importance of the physical soil-conditions has long ago rendered
{ em subjects of close study by Schiibler, Boussingault, and others, the chemistry of soils-
vra^ very generally neglected for a considerable period, after the hopes at first entertained
by Liebig that chemical analysis would furnish a direct indication and measure of soil
fertility had been sorely disappointed in respect to the only soils then investigated, viz.
the long-cultivated ones of Europe. The results of chemical analysis sometimes agreed,
but as often pointedly disagreed, with cultural experiences ; so that after the middle of
the nineteenth century but few thought it worth while to occupy their time in chemical

soil analysis Among the few who, during the middle of the past century,

maintained their belief in the possibility of practically useful results from direct soil
investigation were Drs. David Dale Owen and Robert Peter, who prosecuted such work
extensively in connection with the geological and agricultural surveys of Kentucky and
Arkansas ; and the writer, who carried out similar work in the States of Mississippi and
Louisiana, with results in many respects so definite that he has ever since regarded this aa
a most fruitful study, and has later continued it in California and the Pacific North-west.
This was done in the face of almost uniform discouragement from agricultural chemists
until within the last two decades, with occasional severe criticisms of this work as a waste
of labour and of public funds. All this opposition was largely due to the prejudices
engendered by the futile attempts to deduce practically useful results from the chemical
analysis of .toils long cultivated, without first studying the less complex phenomena of
virgin soils ; and these prejudices persisted longest in the United States, even though in
Europe the reaction against the hasty rejection of chemical soil work had begun some time
before, as is evidenced by the methods employed at the Rothamsted Experimental Farm in
England, the Agricultural College of France, the Russian agronomic surveys, and at

several points in Germany In the United States as well, the ancient

prejudices have now gradually given way before the urgent call for more definite
information than could otherwise possibly be given to farmers by the experiment stations,
most of whose cultural experiments, made without any definite knowledge of the nature
of the soil under trial, were found to be of little value outside of their own

experimental fields In many existing treatises so much emphasis is given to

the alleged proofs of the inutility of chemical soil examination in particular that a special

controversion of these arguments seems necessary In all these discussions the

difference between the ascertainment of the permanent productive value of soils, as against
that of their immediate producing capacity, must be strictly kept in view. The former
interests vitally the permanent settler or farmer ; the latter concerns the immediate
outlook for crop-production, the ' Diingerzustand ' of the Germans. The methods for the
ascertainment of these two factors are wholly distinct, even though the results and their
causes are in most cases intimately correlated. The failure to observe this distinction
accounts for a great deal of the obloquy and reproach that has in the past BO often been

heaped upon chemical soil-analysis and its advocates The abundant presence

of the plant- food ingredients, as shown by analysis, will not avail, unless at least an
adequate portion of the same exists in a form or forms accessible to plants. Of course this
condition would seem to be best fulfilled by the ingredients in question being in the water-
soluble condition. But , . . substances in that form would be very liable to be washed

or leached out of the soil by heavy rains or irrigation It is therefore clearly

desirable that only a relatively small proportion of the useful soil-ingredients should be in
the water-soluble or physically-absorbed condition, but that a larger supply should be
present in forms not so easily soluble, yet accessible to the solvent action which the acids
of the soil and the roots of plants are capable of exercising. This virtually available
supply we may designate as the reserve food-store"

Hilgards' remarks! have been quoted thus lengthily because his boolc
on the subject has all the qualities of a standard work of reference, em-
bodying, as the work of no other writer in recent years has done, in the
clearest language, the most reasonable present-day theories on the subject
dealt with. Following out his line of reasoning, it will be seen at once
tiiat, in a series of investigations of the nature, and earned on under the
circumstances of those that have been performed in the Government Ana-
lytical Laboratories here, quantitative determinations of the ingredients of"

* Hilgard : Op. cit., pp. 313, 316-320.
t See also Snyder : " The Chemistry of Soils and Fertilisers," pp. 69, 70.

12

III possess little more than academic interest, while the readily-
available or water-soluble constituents of grade I do not urgently need in-
vestigation so long as vast fields of exploiation have to be entered upon
in respect of the reserve stock, or, as Hilgard terms it, the reserve food
store, in the various soils of the Colony. It is this programme that we have
K>een endeavouring to fulfil, as far as opportunities and circumstances per-
mitted, during the earlier portion of the past dozen years.

In the investigations conducted in the Government Analytical Labora-
tories one method of soil extraction has been adopted as a standard, but it
has been supplemented by sundry others for comparison. In deciding upon
a standard method, a solvent had to be chosen — bearing in mind what has
already been said — that would represent, not the slight action of a single
crop, nor a dissolving capacity far in excess of any action that successive
crops could exert during many years to> come. The aim was rather to make use
of a solvent that would extract from the soil just as much of the materials
composing the food of plants, as lies ready to hand for conversion into con-
stituents of the first grade as fast as the latter are withdrawn by cultiva-
tion. Obviously an adoption of such a via media necessitates the rejection,
on the one hand, of comparatively weak solvents, such as pure water, which
would extract only the directly or readily available plant food, and, on the
other, such powerful reagents as boiling aqua regia or sulphuric acid, which
would, in addition, dissolve out substances totally unavailable as plant food.

Investigations have, it is true, been made in our laboratory with dilute
citric acid, but simply to compare its action as a solvent with the standard
method.

We have yet to face the question of the actual solvent to be employed
in extracting the soil; having decided what not to use; e.g., any that would
stop short with the extraction of readily available constituent®, on the one
hand, or that would include in their action constituents that axe not avail-
able at all, it still remains to fix on what may be termed a reserve stock
solvent; and here again discrimination is necessary. "Of all the mineral
acids available," says Dr. Wiley,* " no one possesses solvent powers for soil?
in a higher degree than hydrochloric." That acid has accordingly long
been used by analysts for the purpose, and it was also resorted to in these
investigations for a similar reason. In this connection, however, three
points awaited settlement, namely, the strength of hydrochloric acid to be
employed, the length of time the acid should remain in contact with the
§oil, and the temperature at which it should be allowed to act. Lo<ughridge
foundt that hydrochloric acid of I1 115 sp. gr. exerted a greater dissolving
effect upon the soil than either a stronger or a weaker acid ; this strength
of acid had also been adopted in our investigations. He also found that
the solvent action continued to increase for five days and then ceased.
"Upon the above basis the standard method generally employed in the Gov-
ernment laboratories has been formulated; it is the first of the methods
outlined below. One respect in which this method differs- from that of
Hilgard and Loughridge is in respect of temperature. In the warm sunny
climate of this Colony it would appear that root action is more energetic
•than in colder lands. It has been found, for instance, that proportions of
phosphoric oxide which would be considered inadequate in Europe, in this
country often suffice to yield satisfactory returns. But this greater root
action results in a more rapid depletion of the reserve stock of plant food
in the soil : that reserve stock, therefore, is rendered relatively smaller than
under circumstances of lesser root activity. To represent this fact in the
laboratory when a soil is to be analysed, it accordingly becomes necessary
to employ a weaker solvent so as to indicate a relatively lower maximum

* Principles and Practice of Agricultural Analysis," Vol. I., p. 345.
| American Journal of Science," Vol. VII., p. 20.

limit 01 plant food in reserve; no direct experiments to test the validity of
this theory have been made, but the substitution of the ordinary tempera-
ture for that of the steam bath in the hydrochloric acid extraction method
seems to fit in exactly with the conditions of this Colony, and the result*
thereby obtained tally closely with the practical experiences of farmers.

It must again be observed that although every soil which may yield
large quantities of soluble plant food to this method of extraction is not
thereby proved to be fertile, yet, as the method represents a maximum,
we may definitely consider all soils that show small proportions of soluble
constituents under its treatment, to be unmi&takeably poor, and in need of
replenishment. To a certain extent, also, we may assume the fertility of
a soil which yields good results not only by that method, but also when
treated according to the citric acid method of extraction outlined below.

Two other extraction methods by Hydrochloric acid have been used in.
several of the analyses; they are the second and third of the methods de-
scribed below, the former being that of the German Experiment Stations,
and the latter one stated to have been devised by Professor Maercker, of the
Halle Experiment Station.

METHOD I. — STANDARD METHOD OF SOIL EXTRACTION WITH HYDROCHLORIC

ACID.

Two hundred grammes of the air-dried " fine earth "* are placed in
a large rubber-stoppered flask, and treated with 400 c.c. of hydrochloric
acid of specific gravity I1 115 (plus any needed for neutralising carbonates
in the soil) ; allowed to remain for five days at the ordinary temperature,
shaking thoroughly from time to time. After the prescribed period of
digestion has expired, the extract is filtered, through a. dry pleated filter,
into a, dry flask. 250 c.c. of the clear filtrate, representing 125 grammes
of soil, are evaporated to dryness in a shallow porcelain dish, at first over
a small open flame, then on the water bath, and finally on a sand bath,
or in an air oven, at 120°C, until perfectly dry. During the evaporation
a few cubic centimetres of strong nitric acid are added to the extract.
The dry residue is moistened with strong nitric acid, and again evaporated
to dryness; to expel the nitric acid, the residue is moistened with hydro-
chloric acid, and evaporated on the water bath to as near dryness as pos-
sible, taking care to stir towards the end of the evaporation, so as to
prevent the formation of crusts. This final residue, after warming in
the air bath for an hour at 105°C to 110°C, is treated with warm water
and a 20 per cent, solution of hydrochloric acid, and is then washed over
into a 250 c.c. flask, boiled for 15 minutes, allowed to cool, filled up to
the mark with distilled water, and filtered into a suitable bottle. This
filtered soil extract is then employed for the actual estimations of lime
and potash.

Determination of Lime. — 50 c.c. of the extract (equivalent to 25
grammes of " fine earth "), obtained as described in the preceding para-
graph, are removed, by means of a pipette, into a 250 c.c. boiling flask:1

* As more than once indicated, only soil grains below a certain size are included in the
material submitted to extraction. Loughridge (Proceedings of the American Association
for the Advancement of Science, Vol. 22, p. 81) has found that, of all the grades into which
soils are usually separated by mechanical analysis, the " clay," i.e. the finest grade, is by
far the richest in mineral constituents, and that the quantity of acid-extractible matter
in the grades of soil particles of over '04 mm. diameter was practically negligible. Hence
it follows that soil sifted through a \ mm. sieve will contain all the mineral constituents
available for plants. The first requisite in selecting a standard method is to utilise for
extraction only the soil passing the £ mm. sieve. Hence, too, the method t.g. of sifting
the soil through a 3 mm. sieve and pounding the sifted portion for extraction must be
rejected as furnishing misleading results.

14

after adding two or three drops of rosolic acid solution (made by dissolving
one gramme of rosolic acid crystals in 100 c.c. of 96 per cent, alcohol),
.ammonia is added, very cautiously, by means of a dropping tube, until a
permanent pinkish colour is produced in the supernatant liquid. The
mixture is then boiled until the pink colour almost disappears again, the
.alumina and oxide of iron being thus precipitated. After cooling, the
flask is filled up to the mark, thoroughly shaken, and the contents filtered
into a 300 c.c. bottle. 100 c.c. of this clear filtrate (equal to 10 grammes
-of fine earth) are removed by a pipette into a 300 c.c. Erlenmeyer flask ;
three to five drops of acetic acid are added, which should render the mix-
ture feebly acid; the liquid is then heated to near boiling, and treated
•with 20 c.c. of a four per cent, ammonium oxalate solution. The mix-
ture is placed on a water oven for six hours, and the precipitate is then
collected, ignited, weighed, and the lime calculated as CaO.

Determination of Potash. — Other 50 c.c. of the filtered soil extract
-(equivalent to 25 grammes of fine earth) are placed in a 300 c.c. Erlen-
meyer flask and boiled. 5 c.c. of a ten per cent. Barium chloride solution
.are added, and the mixture is boiled for another five minutes for the
precipitation of sulphates as Barium sulphate. After filtering into a
marked 250 c.c. flask, a few drops of rosolic acid are added, and boiling
with ammonia is proceeded with as in the case of the lime determination.
When partly cooled down, two or three grammes of crystalline ammonium
carbonate are thrown in, and the temperature is once more raised to the
boiling point, in order to separate calcium and barium. After complete
precipitation of these, the liquid is cooled, the flask filled up to the mark
with distilled water, and the contents filtered. Of this filtrate 100 c.c.
•equivalent to 10 grammes* of fine earth) are placed in a platinum basin, and
evaporated to dryness on a water bath. The dish containing the residue is
heated, at first on asbestos sheet, and then carefully over a small open flame,
until all ammonium salts have been expelled. By means of boiling water
the residue is then washed through a filter into a glass dish : 2 c.c. of a
10 per cent, solution of platinic chloride are added, and the mixture is
^evaporated to dryness on the water bath. After cooling, some dilute
alcohol (81 to 82 per cent.) is added to the residue, and it is allowed to
stand for at least half an hour. It is now filtered through a Gooch
-crucible by the aid of a filter pump, washed, at first, with 96 per cent.,
and then with absolute alcohol, and dried for two hours in a water oven.
The weight of the crucible containing the Potassium platino-chloride hav-
ing been taken, the precipitate is washed through the filter with boiling
water, and the crucible, after again washing with alcohol, is dried and
-weighed. The difference between the two weighings, multiplied by *193,
gives the quantity of potash (K20) in the ten grammes of fine earth taken.

Determination of Phosphoric oxide. — 25 grammes of fine earth are
placed in a marked 500 c.c. flask, 25 c.c. of concentrated nitric acid are
•added, and the mixture is thoroughly shaken. 50 c.c. of concentrated
sulphuric acid are next added, and the mixture is again carefully shaken
up. It is then gently heated, shaking at frequent intervals. If this does
r.ot lead to complete oxidation, more nitric acid is added, and the heating
continued. Finally the mixture is cooled, and diluted to the mark with
distilled water; it is then well shaken and filtered. 200 c.c. of the
filtrate (equivalent to ten grammes of fine earth) are placed in an Erlen-
meyer flask of suitable size, and very nearly neutralised with strong
ammonia solution, a few drops of nitric acid being used to acidify the
mixture in case the limit is overstepped. 200 c.c. of Molybdio solution —
prepared by dissolving 150 grammes of ammonium molybdate in a litre

* In the citric acid extraction process (see post) this quantity will be equivalent to 20
grammes of the soil taken.

of water, and, when completely cool, adding this to a litre of nitric acid
of specific gravity 1*20 — axe added, and the mixture is heated to a
temperature of 50°C for three hours in a water oven, and allowed to cool
completely. The liquid is decanted through a small filter, and the pre-
cipitate remaining in the flask repeatedly washed with diluted (1:1)
molybdic solution. It is then dissolved, oil the filter, with warm 5 per
cent, ammonia, and to the resulting solution, while still warm, hydroch-
loric acid is added at once, but gradually, until the precipitate at first
formed dissolves only after long agitation. After cooling 20 c.c. of Mag-
nesia mixture* are added from a burette, drop by drop, at the rate of one
drop every five seconds, and then 25 c.c. of 5 per cent, ammonia. The
mixture is allowed to stand for twelve hours at least, after which the pre-
cipitate is collected in a weighed Gooch crucible, and washed with 5 per
cent, ammonia. It is then dried on a heated iron plate, and ignited in
a furnace for fifteen minutes. The crucible is then cooled and weighed
-with the magnesium pyrophosphate.

METHOD II. — HYDROCHLORIC ACID EXTRACTION AS PRACTISED BY THE
GERMAN EXPERIMENT STATIONS.!

To 200 grammes of dry-sifted " true soil " are added 400 c.c. of a
25 per cent, hydrochloric acid solution, the quantity of acid being in-
creased, if necessary, when the soil contains an excess of carbonates. The
mixture is allowed to stand a,t the ordinary temperature for 48 hours,
with frequent shaking, and is then filtered.

Determinations of lime and potash are made in definite quantities of
the filtered extract (prepared as in Method I), exactly as in the first
method.

Determination of Phosphoric oxide. — 20 c.c. of the filtered extract
(which has been treated in a similar manner to the extract in. Method I),
equivalent to 10 grammes of " true soil," are placed in a dish and evapor-
ated almost to dryness. The residue is taken up with a little nitric acid,
washed into a suitable Erlenmeyer flask, and nearly neutralised by means
of strong ammonia solution, a few drops of nitric acid being used to
acidify the mixture should the neutral point be passed. Molybdic solu-
tion is then added, and the rest of the process conducted aa in the first
method.

METHOD III. — EXTRACTION OP IGNITED SOIL BY MEANS OF HYDROCHLORIC

ACID, AIDED BY NlTRIC ACID. (MAERCKER's METHOD.)

Twenty grammes of "fine earth" are weighed in a platinum dish,
and heated on asbestos wire gauze until all organic matter has been ex-
pelled. The dish is then cooled and the contents carefully transferred
by the aid of a brush into a porcelain dish. 100 c.c. of concentrated
hydrochloric acid and 10 -c.c. of concentrated nitric acid are now added,
and the mixture is evaporated on a water bath, and dried for three hours
at 120° C. Warm water is next added, together with some dilute hydroch-
loric acid, and the whole is washed into a marked 500 c.c. flask, in which
it is boiled for 15 minutes. The mixture is then cooled, filled up to the
mark with distilled water, and filtered.

Determination of Lime. — 100 c.c. of the filtered extract (equivalent
-to four grammes of fine earth) axe placed in a 200 c.c flask, two or three
drops of rosolic acid are added, and the determination is proceeded with
similarly to that in Method I.

* This Magnesia mixture is prepared by dissolving 110 grammes of magnesium
•chloride, and 210 grammes of ammonium chloride, in 700 c.c. of 24% ammonia solution
filtering the mixture, if turbid, and diluting with distilled water to two litres.

t Land. Versuchsstat, Vol. 38, p. 311.

16

Determination of Potash. — Other 100 c.c. of the filtered extract
(equivalent to four grammes of fine earth) are boiled in a 300 c.c. Erlen-
ineyer flask, and 5 c.c. of a ten per cent, solution of barium chloride are
added, after which the determination is continued as in the first method.

Phosphoric oxide is determined exactly as in Method I.

METHOD IV. — EXTRACTION BY HYDROCHLORIC ACID AT STEAM TEMPERATURE.
(HILGARD'S AND LOUGHRIDGE'S METHOD.)

Fifty grammes of air-dried " fine earth " are placed in a porcelain,
beaker of sufficient size, 500 c.c. of hydrochloric acid of 1'115 specific
gravity are added, and 2 c.c. of nitric acid, and, after covering the beaker
with a clock glass, the mixture is digested on the steam bath for five days.
The solution is then filtered, the residual earth is thoroughly washed with
distilled water, the filtrate and washings are evaporated to dry ness,
as in Method I, with nitric acid in a shallow porcelain dish, and finally
made up to 250 c.c.

Determination of Lime. — 100 c.c. of the extract (equivalent to 20
grammes of fine earth) are placed in a 200 c.c. boiling flask, and pre-
cipitated by ammonia, as in Method I ; the flask is then filled up to the
mark with distilled water, and the contents filtered. 100 c.c. of the
filtrate (equivalent to 10 grammes of fine earth) are precipitated with
ammonium oxalate, as in Method I.

Determination of Potash. — 100 c.c. of the extract (equivalent
to 20 grammes of fine earth) are treated with barium chloride solution in
the same way as in Method I, and filtered into a 200 c.c. flask; after pass-
ing through the usual processes 100 c.c. of this (equivalent to 10 grammes
of fine earth) are taken, as in Method I, for the actual potash determina-
tion.

METHOD V. — EXTRACTION OF THE SOIL BY MEANS OF CITRIC ACID.
(DYER'S METHOD.)

The solution here used is that recommended by Dr. Bernard Dyer, in
the " Journal of the Chemical Society," March, 1894, p. 141, which is an
adaptation of that previously used at the Halle Experiment Station, and
published by Maercker and Gerlach in 1892. In a rubber-stoppered three
litre flask are placed 200 grammes of dry-sifted " true soil " (i.e., as sifted
through a 3 mm. sieve), together with two litres of distilled water con-
taining in solution 20 grammes of pure citric acid. This solution is left
in contact with the soil, at the ordinary temperature, for seven days,
shaking the mixture thoroughly about fifty to sixty times each day. At
the end of the seven days the solution is filtered, by the aid of a filter
pump, through a porcelain funnel with a flat perforated base, or through
a Berkefeld candle filter; 500 c.c. of the filtrate would be required for
each of the determinations described below.

Determination of Lime. — 500 c.c. of the filtered soil extract (equivalent
to 50 grammes of soil) are placed in. a 1,000 c.c. flask, and a few drops of
rosolic acid are added, followed by ammonia, as in the previously described
lime determinations. The mixture is boiled, and, after filling to the mark,
it is filtered, and 500 c.c. of the filtrate, equivalent to 25 grammes of soil,
are warmed with a few drops of acetic acid in an Erlenmeyer flask, and
then treated with 50 c.c. of 4 per cent, ammonium oxalate solution, the
precipitate being collected, as before, after warming for six hours.

Determination of Potash. — 500 c.c. of the filtered extract (equivalent
to 50 grammes of soil) are evaporated to dryness in a platinum dish, and
cautiously ignited. The residue left in the dish is dissolved in hydroch-
loric acid, filtered into a 300 c.c. Erlenmeyer flask, and boiled. 5 c.c. of
10 per cent, barium chloride solution are added, and the determination
is proceeded with as in the method already described in connection with
the extraction by hydrochloric acid.

17

Determination of Phosphoric Oxide. — 500 c.c. of the filtered soil extract
(corresponding to 50 grammes of true soil) are evaporated to dryness in
a platinum dish, carefully ignited, and taken up with hydrochloric acid,
again evaporated, ignited, and extracted with hydrochloric acid; the ex-
tract is filtered and washed, and the filtrate and washings concentrated.
The concentrated solution is allowed to cool, and then 200 c.c. of a solu-
tion of ammonium molybdate in nitric acid — prepared as already de-
scribed— are added, and the rest of the procedure outlined under the
method of determining phosphoric oxide in the " fine earth " (see method
I.) is followed.

It now becomes a matter of interest and importance to agree upon
the interpretations that are to be placed upon the figures obtained by
analysis from a soil that has been extracted according to one of the fore-
going methods. In this Colony absolutely no investigations have yet been
made to show what quantities of plant food are necessary in order to ren-
der fertile a soil that is in other respects well circumstanced ; for it must
be remembered that a brack or arid soil, or one physically unfit for culti-
vation, cannot be otherwise than unproductive even when it is amply
supplied with the needful chemical constituents.

Maercker, of the Halle Experiment Station, graded soils, on the plant
food basis, as follows, the extraction being made with strong acid: —

*>« PoSh;f i» clay soil^sandy soils. »«"«-*

Poor ...... < '05 < -05 <-10 < '05 < '05

Medium ... -05— '15 -05— '10 -10— '25 -05— -15 -05 — '10

Normal ... -15— -25 -10— '15 '25 — '50 '15— '20 '10 — '15

Good ...... -25 — -40 -15— -25 '50— I'OO '20— '30 -15— '25

> -40 >'25 >1'00 > '30 > '25

This classification Professor Hilgard declares to be in remarkable
agreement with his own, with the proviso that, in the presence of high
lime percentages, relatively low proportions of phosphoric oxide and
potash may, nevertheless, prove adequate. Many of the results tabulated
in the following pages may incidentally throw light upon the subject, but
until it has been directly investigated, and sufficiently so to enable limits
to be laid down for different parts of this Colony, it will probably prove
most convenient and satisfactory provisionally to judge of soils by
Maercker's limits, and such, in fact, has been my practice hitherto, nor
has it, as already observed, been found to lead to conclusions inconsistent
with practical experience.

According to Dyer's investigations, a soil extracted with his citric
acid solution is sufficiently supplied with potash and phosphoric o-xide for
immediate purposes when the former shows over '005 and the latter over
'010 per cent. The investigations upon which these conclusions were
based, it should, however, be said, had been conducted upon well-studied
and productive soils at Rothamsted.

Just here one point calls for special emphasis : Liebig enunciated the
law that the growth and development of plants is regulated by the amount
of that particular plant food constituent which is present in the soil in
smallest proportion: if one element of plant food is deficient in the soil,
no excess, however great, of any of the others will atone for the defect.
If a soil contain abundance of lime, potash, and nitrogen, but lacks phos-
phoric oxide, no crop can reach perfection. Liebig's law of the minimum,
thus briefly set forth, must now be extended so as to fill our widened

18

vision. We know that, chemically, a soil may be all that is desired, and
yet prove unproductive; a mechanical analysis may elucidate the cause of
this; or, if that too result satisfactorily, the soil may be otherwise physi-
cally defective, for instance, in porosity; or perhaps it may lack moisture.
The various chemical constituents of plant food do not comprise the whole
chain of contingencies upon which a soil's fertility, and still less its pro-
ductiveness, depends, but they do form links in that chain, and, as the
strength of a chain is measured by that of its weakest link, so the poten-
tial fertility of a soil well provided chemically may be dependent upon,
and therefore limited by, defects along other lines. Hence chemical
analysis does not test the strength of the chain, but only that of certain of
its links. The fact that the links thus tested prove to be sound does not
render the entire chain sound. A soil which yields to acid extraction a
large reserve stock of all plant food constituents may be deficient in other
respects : even if physical unfitness or alkalinity do not affect its fertility,
defective water supply, or unfavourable situation, or many other agencies,
may cause it to be unproductive,* no matter how fertile it may be. On
the other hand, if by chemical analysis certain links in this chain are
found to be defective, then, no matter how excellent the condition of the
remaining links, the defects must be remedied ere the chain can be fit for
use.

The results of the analyses detailed in the sequel show that there are
many soils in the Colony which are thus weak in certain respects, and if
the investigations here described do no more than point out the districts
and areas where such weakness exists, they will have served a useful pur-
pose. A chemically-poor soil invariably needs working up, and such in-
vestigations as these can show where soils answering to that description
are to be found. The fact that soils rich in plant foodf are not invariably
fertile, or, if fertile, do not produce an adequate return, is no argument
against the worth of analysis. In brief, chemical analysis can designate
bad soils even if it does not aim at pointing out good ones; and yet even
this it is at times capable of doing, as the following analyses more than once
witness. Hilgard may here again be quoted. He says : — J

" It seems to be generally true that virgin soils showing high percentages of plant
food, as ascertained by extraction with strong acids (such as hydrochloric, nitric, etc.),
invariably prove highly productive : provided only that extreme physical characters do
not interfere with normal plant growth, as is sometimes the case with heavy crops or
very coarse sandy lands. To this rule no exception has thus far been found"

* " Fertility and crop production are different terms. Fertility is a property inherent
in the soil : it is what the soil is capable of doing if it is under the best possible conditions.
The yield of crops, on the other hand, is not dependent upon the fertility alone. ... If
your seed is not properly selected, if your planting season is too early or too late, if the
soil is not properly cultivated, if the climatic conditions are not favourable, your crop
yield may be affected, but the fertility of the soil — that inherent power of the soil, under
the best conditions, to produce a crop — will not necessarily be impaired." (Whitney :
"Soil Fertility," 1906, pp. 5, 6.

" The productivity of a given piece of land depends upon a large number of agencies,
any one of which may be the limiting factor in the crop yield. We may enumerate, for
example, temperature and water supply, both determined by the climate, by the natural
physical structure of the soil, and by modifications in its texture induced by cultivation :
there are, further, the aeration and the actual texture of the soil, the initial supply of
plant food of various kinds, and, again, the rate at which this last item is rendered
available to the plant by bacterial action or by purely physical agencies. All these factors
interact upon one another, to all of them, and not merely to the nutrient constituents does
Liebig's law of the minimum apply, so that any one may become the limiting factor and
alone determine the yield. It is of no use, for example, to increase the phosphoric acid
content of a soil, however deficient it may be, if the maximum crop is being grown that is
consistent with the water supply, or if the growth of the plant is being limited by
insufficient root range caused by bad texture and the lack of aeration in the soil." (A. D
Hall : "Recent Developments in Agricultural Science." Brit. Assn. Rep., 1906, p. 275.)

t i.«., yielding large proportions to strong hydrochloric acid.

j Op. cit., p. 343.

19

FIELD OPERATIONS.

Professor Whitney, Chief of the Division of Soils in the United States
Department of Agriculture, opened his report on the field operations of
his Division during 1899 with these words : —

" During the season of 1899 three well-organised parties were in the field for from six
to eight months, each equipped according to the most modern methods for surveying
investigations and mapping the soils of several important agricultural districts."

That division was established in 1894, with a personnel of 10 persons;
by the end of 1904 the number had increased to 127. Up to the middle
of 1900 about 3,500 square miles of country had been mapped out in con-
nection with the soil survey; within the succeeding four and a half years
85,500 square miles had been mapped, — a fact which shows the rapid in-
crease in the work of an institution that commenced operations a year
after our own soil investigations had been started.

At no time have we, in this Colony, been able to spare more than
one solitary unassisted individual for both field operations and laboratory
work in connection with our soil investigations, so that, while the collec-
tion of samples was going on the operations in the laboratory had to
cease, and vice versa. This is a circumstance that must be continuously
borne in mind to account for the comparative paucity of result hitherto
obtained,

The equipment of the soil collector has, as a rule, been little else
than a stock of canvas bags, a spade, a supply of census maps of each dis-
trict traversed, a small pocket compass, and a trocheameter. By the aid
of these implements samples have been taken from some 27,000 square
miles of country.

The soils whose analyses are tabulated in the following pages fall into
three classes according to the circumstances of their collection: —

(1) First of all must be mentioned the specimens taken in pursuance
of the survey scheme to which reference has already been made. In each
case these samples were collected by one of the analysts attached to the
Government Analytical Laboratory, who had been specially detailed for
the work. The following is a list of the samples so taken from each, of
•the Divisions of the Colony visited for the purpose : —

niTTi-o;rtn No. of samples Area of Division,

collected. in square miles.

Cape 25 663

Malmesbury 54 2,329

Caledon 30 1,772

Bredasdorp 21 1,577

Swellendam 37 2,362

Robertson 27 1,526

Riversdale 24 1,712

Mossel Bay 17 707

George 19 979

Knysna 13 810

Uniondale 12 1,690

Oudtshoorn 19 1,653

Worcester 30 2,623

Ladismith 16 1,256

Paarl 23 610

Komgha 27 546

Cathcart 28 995

Queenstown 4 1,749

Butterworth 7 311

Idutywa 1

Willowvale 4

St. Marks' 4 471

20

The total area from which these 442 samples of soil were collected em-
braces about 27,000 square miles, i.e., about one-tenth of the entire Colony,
so that, on an average, one sample was taken from every sixty square miles
of country.

(2) Apart from this scheme of soil investigation, samples were taken,
from time to time, at various spots of special interest, as occasion offered,
by one or other of the analytical staff, while advantage was taken of sun-
dry journeys into different parts of the country by Mr. A. C. MacDonald,
subsequently Assistant Director of Agriculture in the Transvaal, Dr. E.
A. Nobbs, and other officials of the Agricultural Department, to procure
additional specimens. Means were thus afforded for the analysis of many
soils collected within the following Divisions of the Colony : — Albany,
Aliwal North, Barkly West, Cape, Gordonia, Hopetown, Humansdorp,
Kenhardt, Kimberley, Prieska, Robertson, Somerset East, Stellenbosch,
Steynsburg, Taungs, and Tulbagh.

(3) We are also indebted for the collection of many of the samples
whose analyses appear in the following tables to unofficial persons, far-
mers and others, who, in many cases, kindly responded to requests con-
veyed to them to send specimens of particular soils which, for some reason
or another, it was thought desirable to analyse. For several reasons, most
of which are obvious, it is not possible to give as much detail regarding
these soils as in the case of those collected in pursuance of the systematic
survey scheme. Amongst the soils which, were thus procured were sam-
ples from the following Divisions : — Albert, Barkly West, Beaufort West,
Bredasdorp, Caledon, Cape, Ceres, Clanwilliam, Colesberg, Elliot-Slang
River, Fort Beaufort, George, Graaff-Reinet, Hanover, Kenhardt, Kim-
berley, King William's Town, Knysna, Kuruman, Maclear, Mafeking,
Malmesbury, Mossel Bay, Mount Cuirrie, Mount Frere, Namaqualand,
Paarl, Piquetberg, Port St. John's, Queenstown, Richmond, Robertson,
Stellenbosch, Steynsburg, Stockenstrom, Tulbagh, Uitenhage, Umtata,
Umzimkulu, Victoria East, Vryburg, Willowmore, and Worcester.

The soils included under this category have been separately
tabulated and are distinguished from the others by being marked
" privately collected," inasmuch as it has not always been abso-
lutely certain that in every case the samples so taken, however specific
previous instruction on the point may have been were thoroughly repre-
sentative, or that they were collected in the manner officially prescribed.
For these reasons the soils known to have been taken according to the
specific directions have been distinguished as " officially collected."

21

PART III.— RESULTS OF CHEMICAL ANALYSES.

It may be convenient at tins stage, before passing on to enumerate in
detail the actual analytical results arrived at, to tabulate the soils exam-
ined, grouping them according to the various divisions and districts of
the Colony whence they were collected, and in regard to the particular
methods by which they were analysed ; this is done in the list below : —

Method Method Method Method Method

Division or District, I., II., III., IV., V., Total.

Standard. German, ilaercker's. Hilgard's. Dyer's.

Albany 3f 3f* 3

Albert 3 3

Aliwal North 4 4

Barkly West 13 13

Beaufort West 1 1

Bredasdorp 1 21 22

Butterworth 7 7

Caledon o 34 40

Cape 23 4G 69

Carnarvon 2 2

Cathcart 28 — 28

Ceres 7 — 7

Clanwilliam 7 7

Colesberg 6 6

Elliot-Slang River 4 4

Fort Beaufort 4 4

George 20 6 3 29

Gordonia... 2 2

Graaff-Reinet 3 3

Hanover 6 6

Herbert 3f 3f 3

Hopetown 2 2

Humansdorp 4 4

Idutywa 1 1

Kenhardt 4 1 . 5

Kimberley 6 2 8

King William's Town ... 1 1

Knysna 16 3 19

Komgha 27 27

Ladismith 16 16

Maclear 1 1

Mafeking 2 2

Malmesbury 7 64 71

Middelburg 1 1

Mossel Bay 3 17 20

Mount Currie 2 2 4

Mount Frere 2 2

Namaqualand 3 3

Oudtshoorn 13 19

Paarl 51 5 66

Piquetberg 1 1 2

Port St. John's ] 1

Prieska 4 4

Queenstown 6 — — 6

* Phosphoric acid determination-; only.

f Same samples examined by different method?.

22

Method Method Method Method Method

Division or District. I., II., III., IV., V., Total

Standard. German. Maerckers. Hilgard's. Dyer's.

Kichmond 2 2

Riversdale — 24 24

Robertson 22f 12 4f 4f 34

St. Mark's 4 4

Somerset East 2 1

Stellenbosch 2Gf 13 If 39

Steynsburg 9 11 20

Stockenstrom 5 5

Swellendaui 9 29 38

Tulbagh 94 13

TJitenhage 10 1 11

Umtata 2

Umzimkulu 2 1 3

Uniondale 12 12

Victoria East 1 1

Vryburg 16 16

Willowmore — 1 1

Willowvale 4 4

Wodehouse 3 3

Worcester . 34 — — — — 34

Totals 473 290 44 4 11 807

In the following pages are tabulated the results of each individual
chemical analysis; as a matter of convenience, the soils of each, division
or district of the Colony are dealt with separately. In respect of each,
area some preliminary remarks are offered regarding the collection of the
samples therefrom, and, as previously indicated, it has in all cases been
noted whether this sampling had been officially conducted, or through
private media. In every case, too, the method of analysis has been
specified, and in most instances a few comments are added on the general
characteristics of the soils of the district, or of such localities as may seem
to call for special observation.

f Same samples examined by different methods.

MAP OF CAPE COLONY.

Showing Divisions (underlined) from which samples of soils were collected for Agricultural
Chemical Analysis, and number of samples analysed in each case.

23

I

ALBANY.

(Officially collected.)

No. Field Coruetcy. Farm or Place. Collector.

1. Lower Riebeek. Langley Park. A. C. Macdonald.

» » »

3. ,, Highlands. „

The three soils examined were collected in the Highlands district,
Sheep running on the two farms whence the samples were taken had been
found to be greatly benefitted by giving them a regular supply of bone-
nieal mixed with salt. Lamziekte had been prevalent on the same farms,
indicating apparently an insufficient supply of phosphates in the food.
The theory prevailed, due, in part, to the foregoing facts, that the soil of
the Zuurveld generally lacked phosphatic material. The lands on the
farm Langley Park, whence the first sample was taken, were being em-
ployed as grazing ground for some of the sheep above referred to; the
second sample, which is distinctly poor in lime, was taken from the same
farm, but from a portion on which sheep which were not habitually and
specially provided with salt in their food were stocked; such a provision
did not appear to be so much needed there as in the former place. The
third sample was taken from the veld near Highlands Railway Station.
No. 1 contained a fair reserve of phosphates, but both the other soils
were lacking in this respect. The analytical results were as follows:

(Method II.)
No Percentage of Soil sifted through 1 mm. Sieve. P^centage KuteA through

Water. Organic matter. Chlorine. Nitrogen. Lime. Potash. Phosphoric oxide

1. -88 5-62 -0064 -12 -084

2. *62 2-53 -0167 — -065 — -036
•59 5-97 -0106 -11 trace

Determinations of the readily available phosphoric oxide resulted as
follows :

(Method V.)
No. Percentage.

1 ..................... -026

2 ............ . ........ trace

3 ..................... trace

Magnesia and sulphuric oxide were also determined in these soils,
with the following results per cent. : —

No. Magnesia. Sulphuric oxide.

1 ............... -023 -062

2 ............... -029 -075

3 ............... -024 -052

Research has shown that nitrifying bacteria need phosphates for their
development; hence lack of phosphates is apt to go hand in hand with
retarded nitrification. This latter process is also stopped by soil acidity,
so that the neutralising effect of carbonate of lime not only improves sour
soils, in that it removes the cause of their acidity, but it also promotes
the production of the soil nitrogen by affording the bacteria facilities for
nitrogen-fixation. The supply of nitrogen to the plant is thus dependent
upon a, sufficiency of lime and phosphates in the soil.

24

The soils of a great part of the Albany Division, and of the adjacent
divisions of Bathurst and Willowmore, are apparently derived from the
quartzites of the Zuurberg Range in the Witteberg geological series, a
formation lying over the Bokkeveld series, which forms a large portion
of the Ladismith Division. Like the soils derived from the somewhat
similar Table Mountain series in the George Divison, these sands, or sandy
loams, as they become in certain localities, are agriculturally poor, and
would be greatly improved by the admixture of clay, especially if vege-
table mould were added simultaneously, and the further addition of lime,
either as burnt lime, or, less expensively, as crushed limestone would
vastly augment the soil's adaptability for agriculture.

ALBERT.

No.
1.
2.
3.

(Privately collected.)

Field Cornetcy. Farm or place. Collector.

Lower Groot River. Odendaalstroom. .A. Struben.

These samples were collected by an officer of the Public Works De-
partment from irrigable lands on the bank of the Orange River below
Odendaalstroom. No. 1 was a brown silt, rich in lime, but of medium
quality as regards nitrogen; No. 2 a stiff red, sandy clay; and No. 3
a red sandy clay. These red soils are apparently derived from the red
shales and clays of what have been termed the Burghersdorp beds.* These
beds consist of fine-grained sandstones, but their fertility seems to be due
to the still finer clays with which they are associated.

The chemical analysesf resulted as follows : —

(Method I.)

Percentage
of Field

Percentage of Soil sifted through

Percenta

No,

, Sample.

1 mm. Sieve.

Fine earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

1.

100

3-46

4-42

•0071

•070

1-372

V.

94-6

2-44

3-71

•0071

•056

•540

3.

95-2

2-10

2-68

•0089

•014

•424

£ mm. Sieve.

Potash.

•042

•048
•051

Phosphoric
oxide.

•149
•138
•089

Of these three soils, the silt, as frequently proves to be the case,
is chemically the best, and No. 3, the most sandy, is the worst, being poor
in nitrogen and potash, with only a moderate amount of phosphates. No.
2 is well supplied with lime, although inferior to No. 1 in this and other
respects. All these soils are lacking in potash, but it must be remembered
that the good all round supply of lime compensates for other chemical
defects.

ALIWAL NORTH.

(Officially collected.)

No.

1.
2.

3!
4.

Field Cornetcy.
Aliwal North.

Farm or place.
Municipal area.

Collector.
E. A. Nobbs.

* Ann. Kept., Geological Commission, 1904, pp. 75 and 77.

t For mechanical analyses of ttese soils see under the head of " Physical Composition
of Soils," Part VII.

In texture the above soils were all fine grained, and in many respect*
resembled those from the adjoining Division of Albert. The geological
formation here too almost entirely consists of Burghersdorp beds. All
the samples were taken from the municipal area; the first from a rich,
wide level flat flanking the Orange River, the last at a spot lying some
distance back from the river. Nos. 2 and 3 were taken along the banks
of the tributary Kraai River, the latter at some distance from the water
side.

The results of the analyses are as follows : —

Percentage

of Field
No. Sample.

Fine earth.

(Method I.)

Percentage of Soil sifted through
1 mm. Sieve.

Percentage of Soil sifted through,
i mm. Sieve.

1.

2.
3.
4.

99-5
99-4
98-0
97-7

Water.

3-23
4-07

3-61
;VO«S

matter.

3 • 66
4-32

4-17

4 • 73

Chlorine. Xitrogen. Lime.

• ( )! if)
•Oftl
•001
•003

•028
113

098

•376
•276

• 192
•036

•079
•123
•12.S
123

•088
•087
•Of>7
056

It had been anticipated that the first of these four soils would prove
to be chemically of good quality, and the analysis confirmed this view,
the proportion of lime being satisfactory, although potash and phosphates
are present in only moderate amount. Taken all round, these soils possess
the chemical and physical requirements of good fertility.

BARKLY WEST.

(Officially collected.)

No.
I.
2.
3.

4.
5.
6.
7.
8.
9.

Field Cornetcy.
Hebron.

Farm or place.
Brady's.

55

Patrys Kraal.

H.V." 75.

»

Ganzepan.
Zwartputs.

Collector.
E. A. Nobbs.

These nine samples were collected by Dr. E. A. Nobbs from different
farms intended to be irrigated from the Harts River by means of a canal
and furrows. The idea was to bring some twenty to thirty thousand acres
under irrigation in this way. The specimens were selected with a view
to securing types of considerable area below the line of furrow. The first
sample, a red loamy alluvial soil, was taken on Brady's farm, at a point,
just above high flood level, 2J miles above the railway bridge, and 200
yards north of the banks of the Vaal River. The sub-soil is of similar
character to that at the surface. The area appeared to be traversed by
dolerite dykes, which would probably cause great variation in the depth
of the soil, and hinder free drainage. On land adjacent to No. 1 mealies
have been grown without manure, at frequent intervals, although not con-
tinuously, since 1874. No 2 was a red porous, friable, gravelly loam soil,
of somewhat similar type; it was collected on the same farm, about half
a mile north of the homestead, and If miles from the border: this sample
is representative of the higher-lying undulating veld in the neighbourhood.

26

The soil depth is about 30 inches, and the sub-stratum consists of water-
worn gravel. The only crops grown here are mealies and Kafir corn:
these crops are raised continuously for three or four years, after which
the land is allowed to lie fallow for a similar period of time. On the
lands represented by both samples 1 and 2 the Vaalbos, i.e., Atriplex, or
Salt bush, is to be seen.

No. 4 was taken from rising ground to the south of Patryskraalvlei,.
and is typical of a light fine-grained brown loam of an average depth of
fifteen inches : the subsoil consists largely of dolerite, and boulders fre-
quently show on the surface. No. 3 was also collected on the farm Patrys-
kraal, to the south-east of the proposed reservoir. The sample represents a
stiff red clay -loam, very level and undrained, and generally shallow, al-
though varying in depth from place to place, and resting upon a stiff blue
clay, which in turn lies upon limestone. Here too the Vaalbos grows
Further up the slopes towards Grecian Kopje, the clay subsoil disappears,
and a red sandy loam surface soil rests directly on the limestone.

The western slopes of the low hills to the east of Patryskraal posses*
a soil, apparently very much the same all along, rising from the flats
where sample No. 3 was collected, and running up to rocky summits. This
soil, represented by No. 9, is a very characteristic fine-grained rich-red
loamy sand, free from stones, and uniform to a depth of over thirty inches.
Due west of the place where that sample was collected, but on the wide
flat below the site of the suggested dam walls sample No. 5 was taken.
This represents a shallow, fine-grained, red sand, on which Mimosas grow,
It rests upon limestone, which every now and again appears on the sur-
face. The sample is typical of a wide stretch of land. No.
6 resembles No. 1 from Salisbury (see Vryburg list), and proves
the uniformity of this wide tract of country, which extends from the low
ridges lying some 2,000 yards to the east of the railway up to Ganzepan,
and from the boundary of the farms Zwartputs and H.V. 75 to Iddesleigh,
These sandy flats appear to continue down to quite near the Harts River :
the soil is a very even-grained brown sand, free from stones, and of con-
siderable depth. The red colour is characteristic of weathered surfaces,
for, when the ground is turned over with a spade, the brown always shows.

At Zwartputs a limited area of different type is met with, represented
by sample No .8. The soil is a deep humus, brown in colour, and sandy at
the surface, but it becomes very dense at a depth of about twelve inches.

Sample No. 7 shows a difference in physical character. As distinct
from the poor red sand found elsewhere, the soil of the upper portion of
the lower levels between Ganzepan and Putsfontein or Blauwboschputs,
is a brown loam about one foot deep, with a yellow clay-loam subsoil. This
type of soil may possibly extend over some five hundred acres, according
to Dr. Nobbs, from whose report on the tour the greater part of the above
description has been compiled.

(Privately collected.)

No. Field Cometcy. Farm or place. Collector.

10. Hebron. J. E. Fitt.

11. „ H.V. 67.

H.V. 63. H. C. Litchfield.

13. Daniel's Kuil. Koopmansfontein. J. Spreull.

Nos. 10, 11, and 12 were also collected from the area proposed to be
irrigated by the Harts River; Nos. 10 and 12 represent surface soils.
No. 10 was collected from Government land in the Harts River valley, and
NOB. 11 and 12 from two farms in the Burg Pits valley.

No. 13 was collected on the Government farm Koopmansfontein from:
flat country abounding in Vaalbos, sour karree, a sort of ganna bush,
reeds, and long grass, and formed a fair type of the grazing ground in the
vicinity. The ground is of a rocky nature and the soil shallow, with a
gravelly subsoil. Fragments of limestone lie strewn about the land, which,
in the spot whence the sample was taken had never been under cultivation

The following are the results of the chemical analyses* of the soils
comprised in the foregoing lists : —

(Method I.)

No.

1

2

3

4

5

6

7

8

9

10
11
12
13

No . 1 , from a chemical point of view, is poor all round : No-. 2 is fairly
well supplied with potash, but poor in other respects. Of the two Patry»-
kraal soils, No. 4 is below normal in all plant food constituents, but No. 3
has a better supply ; both soils, however, are lacking in phosphates. Nos.
5 and 9 are two soils which are chemically very poor. No. 6 has a moder-
ate proportion of nitrogen, but its mineral plant food is deficient. The
Zwartputs soil, No. 8, which was recognised by external appearance as a
soil of different type from the preceding, also proved to be superior upon.
analysis; it possesses a fair supply of all plant food, phosphoric oxide ex-
cepted. The brown loam No. 7 is altogether the best of the entire series :
chemical analysis shows the amounts of nitrogen, potash, and phosphoric
oxide to be normal, lime being present in abundance. It does not seem
impossible that the doleritic rocks in the vicinity may have contributed
greatly to the chemical constituents of this soil.

The farm represented by No. 13 is about sixty thousand acres in ex-
tent, and, in view of the fact that the cattle whose glazing ground it con-
stitutes are prone to attacks of " lamziekte/' the scantiness of phosphatic
material in the soil deserves attention. At the same time one soil sample
can scarcely be taken as1 typical of so extended an area, and if it were?
the difficulty presented of remedying a lack of phosphates in so wide a
tract is undoubtedly enormous.

Percent of Percentage of Soil sifted through Percentage of soil sifted
Field Sample. 1 mm. Sieve. through \ mm. Sieve.
n Phos-

Fine earth. Water. c Chlorine. Nitrogen. Lime Potash, phoric
matter. oxide

92-2

•94

T78

•004

•028

•044

•050

•031

84-4

1-16

2*80

•004

•042

•020

•112

•045

93-4

3-33

4-19

•003

•098

•240

•123

•031

83-4

2-02

2-98

•005

•070

•082

•056

•038

97-3

•66

1-47

•003

•028

•020

•046

•013

94-6

•63

1-31

•003

•063

•036

•035

•014

94-6

4-74

10-90

•004

•134

3-768

•187

•101

98'3

2-57

3-03

•003

•084

•084

•141

•023

96-8

•79

1-69

•002

•056

•020

•037

•022

S8-2

1-44

2-87

•0107

•098

•112

•023

•104

96-6

6'82

5-64

•0089

•070

1-037

•054

•195

97-6

6-03

4'98

•0089

•098

•461

•027

•154

91-3

2-51

8-20

•0028

•094

3-338

•125

•066

* For mechanical analyses of Nos. 10. 11 and 12. see under "Physical Composition of

Soils.'1

BEAUFORT WEST.

(Privately collected.)

Xo. Field Cornetcy. Farm or place. Collector.

1. Nieuwveld. Roos Plaats. W. C. Rose.

On the farm Roos Plaats, formerly known as Bronkhorst Ylei, and ad-
joining Doordrift and Spitskop, in the Nieuwveld district, a sample of soil
was taken from a valley in the neighbourhood of a perennial stream be-
tween two ridges. The soil had a light, sandy appearance, but the subsoil
was of a clayey nature. Upon analysis the following results were ob-
tained : —

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted
Field Sample. 1 mm. Sieve. through i mm. Sieve.

Fine earth. Water Or£anic Chlorine. Nitro" Lime. Potash. phoric
-€r< gen< oxide.

1. 94-0 2-57 5-15 "008 -i96 '356 '300 '114

The chemical analysis bears out the opinion of the occupant of the
farm, that the soil left nothing to be desired. During its present tenancy,
which had lasted ten years, the soil had received no manure of any descrip-
tion, while other lands, lower down the river on the same farm, produced
very indifferent crops until manured with guano.

It is probable that this fertility is largely due to the calcareous tufa
which forms a characteristic feature of this part of the country, and is evi-
dently the result of the disintegration and decomposition of the dolerite
which abounds in the Karroo. As a rule, soils thence derived appear to
l)e well supplied, not only with lime, but also with potash, while even the
proportion of phosphoric oxide is fairly high.

BREDASDORP.

(Officially collected.)
No. Field Cornetcy. Farm or Place. Collector.

1. Bloemfontein. Vogelstruis Kraal. J. Mulleor.

2. Koude Rivier.

5. „ Avoca „

6. Zoetendals vallei. Miere Kraal. „

7. „ Eland's Drift. „

8. Bredasdorp. Nachtwacht. ,,

Q

*• »» » 3>

10. „ Matjesfontein. „

11. Napier. Klippe Drift. „

12. „ Leeuwen Rivier. „

13. ,, Halfaampjes Kraal.

14. ;, Quarrie. „

15. „ Klippe Drift.

17. The Ruggens. Rem Hoogte.

„ Koeranna.

19. „ Haasjes Drift.

20. „ Nooitgedacht.
-21. „ Patrys Kraal.

The soils of the Bredasdorp Division are for the most part of two
broad classes; the larger portion of the Division is covered by the Rug-
gens or shaly soils of the Bokkeveld geological series, while to the south lie
the soils of the Downs. In the adjoining Division of Swellendam a third

f

belt — the sandstone soils, already noticeable in the Caledon Division on
the journey eastwards from Cape Town — again begin to acquire promi-
nence, and further east, in the Riversdale Division, these three parallel
belts, running east and west, are distinctly recognisable.

30

Sample No. 1, a stiff white clay, was taken from the slope of the hill
east of the village of Elim. About six miles W.N.W. of that village, Nog.
2, 3 and 4 were collected on the farm Koude Kivier : they represent re-
spectively a sandy clay, a sandy loam, and an alluvial sandy humus soil,
and were taken from lands S.W., south, and east of the homestead. Al-
though the subjacent rocks are those of the Malmesbury series, these four
soils had evidently been largely influenced by the Table Mountain sand-
stone and quartzites from the ranges about Elim. Basic slag had, to some
extent, been used on this farm. Proceeding some nine miles to the west
of Koude Rivier, a sandy soil, No. 5, was taken from a small granite out-
crop on the farm Avoca, which is the chief tobacco growing farm in this
district : here the surrounding hills are composed mainly of Table Moun-
tain sandstone. On the farm Miere Kraal, about 13 miles east of Elim, a
loose sandy clay, No. 6, was collected. A chain of recent limestone hills
crosses the south-eastern part of this farm. Eland's Drift, a fairly large
grain farm, was next visited : it is situated on the right bank of the
Nieuwjaars River, about 6J miles S.W. of Miere Kraal. On this farm.
No. 7, a stiff red clay soil, was taken.

From Nachtwacht, the farm of Mr. D. Albertyn, about 5J miles
south-east of the village of Bredasdorp, No. 8, a san.dy loam, and No. 9, a
loose clay soil, were taken from lands situated respectively south-east and
south of the homestead. Mr. Albertyn considers that in the dry season
the former soil is the better of the two. Between this farm and the vil-
lage, the compacted limestone which, intermingled with blown sea-sand,
forms about six hundred square miles of coast belt, appears on the sur-
face. No. 10, a loose gravelly clay, was taken from the fa<rm Matjesfon-
tein, situated on the boundary of the Ruggens and the Downs, about 8J
miles E.N.E. of Bredasdorp.

Sample No. 11, a loose clay soil, was taken from lands about half a
mile north-east of the village of Napier. Proceeding thence to the farm
Leeuwen Rivieor, another loose clay soil, No. 12, was obtained from lands
on a hill N.N.E. of the homestead. This, and all the remaining1 samples
of the series, were taken from the Bokkeveld area, and are consequently
types of Ruggeus soils. The next sample, No. 13, from the farm HaJf-
aampjes Kraal, 6J miles north-east of Leeuwen Rivier, was a gravelly
-clay. On a portion of this farm the earth of the soil is very considerably
interspersed with large fragments of clay slate. This is also the case at
Qua.rrie, where a loose clay, No. 14, was taken from, lands lying north-
-east of the dwelling-house. From Klippe Drift, about four miles N.N.E.
of Quarrie, two samples of soil were collected: No. 15 represents a loose,
gravelly clay from elevated lands, and No. 16 a soil of more Karroo type,
from low lands devoid of fragments of clay slate such as those noticed ab
Halfaampjes Kraal and Quarrie. Gravelly deposits, of the nature of
that just alluded to, are frequently met with scattered about the high lands
in this and the adjoining divisions, for example, at Zwarbklip, in the
Swellendam Division, to which reference will be made later. Regarding
the two soils last referred to, Mr. Wessels, one of the occupants of the
farm, was of opinion that during wet seasons No. 16 proved the better.

At Rem Hoogte, just about eleven miles east of Klippe Drift, No. 17.
a loose clay soil, was obtained : another such soil, No. 18, was collected
6J miles south-west of this, on the farm Koeranna, while at Haasjesdrift,
seven miles east of the last, No. 19, a loose gravelly clay soil, was taken.
The next farm visited was Nooitgedacht, or Zout Rivier, about four miles
S.S.W. of Koeranna, where No. 20, a light loose clay, was collected. No.
21, a loose red clay, was taken from the farm Patrysj Kraal, about four
miles south of Haasjesdrift.

31

On account of long continued drought the crop returns of the three
consecutive years preceding this visit to the division had been very
scanty; some farms, in the neighbourhood of those mentioned, had, in
fact, been totally deserted owiug to scarcity of water.

No.
22.

Field Cornetcy.
The Hope.

(Privately collected.)

Farm or Place.
Brakkefontein.

Collector.
C. Maxais.

On the farm Brakkefontein samples of surface soil were collected
from five different spots, at the foot of the Potteberg1 Range, and just be-
low the Potteberg Trigonometrical station. The five lots of soil were
thoroughly mixed and an average sample was then taken for analysis.
This surface soil is rather loose and sandy, and, as may be expected under
the dominating influence of the sandstone range, proves to be poor in
plant food. About five feet below the surface there is a deposit of water-
worn pebbles mixed with yellowish clay, from twelve to eighteen inches
thick, overlying a very dark loamy water-bearing stratum from five to> ten
feet in thickness. Under that lies another layer of very fine yellowish
clay. Lucerne and vegetables of all kinds, especially root crops, have
been found to flourish luxuriantly on this soil.

The analyses of all the foregoing soils of this Division are given
below : —

(Method II.)

i.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

Percentage of Soil sifted through 1 mm. Sieve.
Water. S? Chlorine. Nitrogen.

•15

•16

•13

•12

•17

•15

•15

•15

•16

•16

•18

•18

•15

•17

•16

•17

•15

•077

•15

•19

•16

0-67

3-27

•0099

1-80

8-08

•045

0-74

3-45

•012

0'71

4-06

•0092

0'38

2-36

•015

0-95

5-14

•014

1-18

3-58

•0085

0'54

2-83

•Oil

1-10

5-65

•018

0-92

5-30

•on

1-25

7-95

•0096

1-63

9-65

•019

1-39

6-94

•on

1-27

7-01

•029

0-95

5*86

•021

0-91

4-23

•028

1-32

6-91

•017

1-43

7-19

•043

1-65

6-34

•0071

0-71

3-96

•0064

1-02

2'89

•017

Percentage of Soil sifted through
3 mm. sieve.

Phos-

Lirne. Potash. phoric

oxide.

•11

•26

•12

•14

•32

•28

•11

•25

•40

•31

•20

•15

•18

•16

•37

•094

•15

•15

•16

•094

•20

•063

•0092

•045

•016

•036

•0082

•016

•022

•071

•013

•13

•026

•10

•013

•076

-.028

•12

•015

•23

•0038

•062

•on

•18

•026

•19

•032

•11

•026

•13

•028

•089

•019

•19

•038

•15

•030

•12

•024

•15

•022

•098

•010

32

(Method I.)

( Percentage of Soil sifted through Percentage of Soil sifted

pAm jle 1 mm. Sieve. through 4 mm. Sieve.

ft Phos-

No. Fine earth. Water. ; Chlorine. Nitrogen. Lime. Potash, phoric

oxide.

22. 90-6 1'75 5-35 -022 '147 '064 '017 '045

Nearly all these soils contain satisfactory quantities of Nitrogen, but they
are almost uniformly lacking in phosphoric oxide. Those of the Bloem-
fontein Field-cornetcy are badly supplied with potash, bub the rest of the
Division is apparently better off in this respect. Most of the soils contain
a fair amount of lime, although less than those of Robertson and Swellen-
dam.

BUTTERWORTH.

(Officially collected,)

No. Field Cornetcy. Farm or Place. Collector.

1. 'Nhlambe. 'Nhlambe. St>. C. O. Sinclair.

3. Butterworth. Butterworbh. „

4.

5.

6.

7.

Samples 1 and 2 of this Division were collected on the farm 'Nhlambe,
the former Residency, now the property of Mr. Burgersheim. They were
taken from lands lying along the Gcua River, and may be considered to
represent the soil derived from the sandstones of the Kentani beds belong-
ing to the Karroo system, which lie S.S.E. of Butterworth. Fruit trees
thrive here, and oats also do well. No. 3, a black soil, was collected about
six miles from Butterworth, in the same direction as the previous samples.
It represents a doleritic soil, but, contrary to what has been found else-
where, the chemical analysis (vide table below) shows it to be an all-round
poor soil, as far as mineral plant food goes. No. 4 is a sample of black
soil overlying sandstone, collected at a point about two miles E.S.E. of
Butterworth village. No. 5 was taken from a. spot two miles N.N.W. of
Butterworth, and is typical of the soil found on the hillocks — studded with
doleritic boulders — which are common in this part. The subsoil is a red,
somewhat coarse, gravel, which rises to within eighteen inches of the sur-
face. Samples 6 and 7 represent the Government plantation at Butter-
worth village. The former, a black, micaceous, somewhat clayey soil, typi-
fies the surface soil to a depth of about nine inches. No. 7 represents the
second nine inches of the soil. These soils appear to be the result of the
decomposition of the micaceous sandstone and black shales thab constitute
a portion of the sedimentary rocks to which the name of Idutywa beds haa
been assigned by the Geological Commission. On the plantation repre-
sented by the last named soils grains of all sorts thrive well; so also do
wattles, blackwoods, and sneezewoods, and varieties of cypress. Pines, on.
the other hand, do not seem to thrive.

33

The following table shows the results of the chemical analyses : —

(Method I.)

Percent, of

P*i«M

Percentage of Soil sifted through

Percentage of Soil sifted

F ItM-Q.

Sample.

1 mm.

Sieve.

through 2 rnm. Sieve.

No.

Fine earth.

Water,

Organic

mutter.

Chlorine.

Nitrogen.

Limr.

Potash.

Phos-
phoric
oxide.

1.

94'5

3-89

6-61

•0071

•175

•202

•029

•026

0

97-0

3-24

6-89

•0340

•189

•094

•071

•023

3.

94-4

2-82

5-91

•0283

•168

•080

•023

•015

4.

87-1

3-77

8-67

•0233

•203

•146

•130

•033

5.

96-4

4-77

10-04

•0290

•217

•106

•085

•052

6.

94-4

2-60

6-31

•0276

•189

•090

•134

•036

7.

94/6

1-74

3-69

•0248

•056

•066

•205

•033

All these soils exhibit a deficiency of phosphoric oxide, nor can they
be said to be well supplied with lime and potash ; indeed, the former plant
food constituent is lacking in many of them. On the other hand, moot
of the samples show a fair proportion of nitrogen.

CALEDON.

(Officially collected.)

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14,
15.
16.
17.
18.
19.
20.
21.
22.
23
24.
25.
26.
27.
28.
29.
30.

Field Cornetcy.

Upper River Zonder End.
Zwart River.

Bot and Palmiet Rivers.

Caledon.

TJilen Kraal.

jj
>>

5»

Goudini.

ii
Lower River Zonder End.

Farm or Place.

Middelplaats.
Bok Kraal.
Zwart River.

j>

Rietfontein.
The Vlei.
Langhoogte.

»

Avontuur.
Dasjesfontein.
Klipheuvel.
Muurton.

jj

Klein Steenboks River.
Weltevreden.
Dunghye Park.

))

Good Hope.
Paarde Berg.

»

Weltevreden.
Klein Wolvegat.
Goudini.
Roode Vlei.
Jongens Klip.
Alexanders Kloof.
Ganze Kraal.

»

Tygerhoek.
The Oaks.

Collector.
J. Muller.

34

The collection of soil samples in this Division commenced with A
visit to the farm Klipheuvel, 2j miles W.N.W. of the village of Caiedon;
here sample No. 11, a sandy clay soil, was taken. About 6J miles north-
west of this farm, sample No. .5, a loose clay, was collected on the farm/
Rietfontein; the soil represented by this sample had not been cultivated
for ten years. Proceeding thence about three miles to the south, No. 6,
a sandy clay, was taken from Mr. J. le Roux's farm The Vlei. Neigh-
bouring farmers stated that there is a noticeable difference between the
crops raised at that place, and those at Langhoogte, a farm lying nearly
a mile south of The Vlei. Samples Nos. 7 and 8 were accordingly col-
lected at Langhoogte, both of them sandy clays; No. 8, however, is more
of a Karroo type, and was considered to show greater fertility than No.
7. As will be seen from the table below, the only chemical superiority

•y*\»4

*• / . 'x re*

+ /" UPPKH *X - -< R ° e <r

S RIMER ZONBER ZND ] " '^^.S-^.^ ^ >

viLuEMDoyp KWWWfc ^X

^ x K /^A ZWART RH

^/"\ *

^^'

that the former of these two soils shows over the latter is in respect of the
phosphoric oxide contained ; in the same respect it is also better than the
soil taken at The Vlei.

The next place visited was the farm Avontuur, about 5i miles south-
west of Langhoogte; and here a stiff clay, sample No. 9, was collected.
Proceeding about two miles in an easterly direction along the Zwart
River, another stiff clay soil, sample No. 10, was obtained on the farm
Dasjesfontein — a soil that had not been cultivated during the last four
years, and quite different in character from any of the former samples.
It will be noticed that analysis shows it to contain a very much larger
proportion of phosphoric oxide. From the farm Muurton. or Klein
Zwart River, about 4J miles south-east of Dasjesfontein, two samples.—
No. 12, a stiff clay soil, and No. 13, a loose clay from an alkaline or brack
land — were taken. The soil represented by the latter of these two sam-
ples had not been cultivated for about five years; as will be observed

35

from the analyses below, it proved to be exceedingly poor in lime. The
farmers in this neighbourhood used, as a fertiliser, lime obtained from the
sea-shells of the Bot River Strand, scattering it along with the wheat
when sowing. They had also commenced employing artificial fertilisers
in small quantity, principally Basic slag.

On the farm Klein Steenboks River, about six miles east of Muur-
ton, a ferruginous clay soil, No. 14, was procured, and about three miles
N.N.E. of this farm, No. 15, a sandy clay, was obtained on the farm Wel-
tevreden : it had not been cultivated for the last seven years. The soil
here is very shallow, extending to a depth of about six inches only, below
Tvhich a layer of coarse gravel is met with. Four miles south-east of this,
samples of two varieties of clay soils were taken from. Dunghye Park, the
farm of Messrs. De Villiers Bros., perhaps the largest and richest farm
in this part of the Division. No. 16 represents a stiff clay soil adjoining
the vineyard, while No. 17, a loose clay, was taken from brack land on
the other side of the river. The former was considered preferable to the
latter for sowing purposes during dry seasons. As only imported seed
•was sown, rust did not appear in the crops. At Klein Wolvegat, about
six miles south-east of Dunghye Park, No. 22, a loose clay soil, was ob-
tained from the mountain side. Proceeding south from here along the
Klein River, the farm Good Hope was touched at, and a stiff clay soil,
sample No. 18, was taken from brack land. The next place visited was
the farm Paardeberg, about three miles south-east of Good Hope. Here
two samples were collected ; No. 19, a sandy, clay soil, which had not
been under cultivation for nineteen years, and which the owner of the farm
considered to be very poor, and No. 20, a very stiff clay, which practical
experience had proved to be very fertile. This sample represents the only
stiff red clay in the vicinity. The physical and chemical differences be-
tween samples 19 and 20 are mainly these: No. 20 is finer in texture and
retains moisture better; it contains more organic matter, and has fifty
per cent, more nitrogen and about six times as much potash ; these facts
apparently more than compensate for its low proportions of lime and
phosphates : it would no doubt be improved by Basic slag. No. 19, on
the other hand, which is a purely sandstone derived soil, notwithstanding
its possessing more lime and phosphoric oxide than No. 20, exhibits an
all-round poverty, without any redeeming feature whatever; hence it does
not surprise one to be told that when it is put under cultivation it faila
to bring its crops to perfection, the ears of corn generally shrivelling up
before attaining maturity. Sample No. 21, a sandy clay, was taken from
the farm Weltevreden, "about a mile north-east of the village of Stan-
ford. Further south blown sands, similar to those which cover the Cape
Flats, are met with : these sand dunes consist of finely divided particles
of sea-shells and quartz sand, with occasional compacted limestones.

Eastwards from Caledon a sandy clay, No. 23, was obtained from
Goudini, the farm of Mr. D. H. Kleyn, about 4£ miles north of Klein
Wolvegat. As a fertiliser Mr. Kleyn was in the habit of using principally
bone manure, from which he obtained very good results. This is by no>
means surprising, for the analysis shows the soil to be very deficient in
lime, while its phosphoric oxide is not much better. Proceeding about
seven miles in a north-easterly direction, the farm Roode Vlei was visited,
and a stiff clay, No. 24, was sampled. On the farm Jongens Klip, 3£
miles further W.N.W., sample No. 25 was taken, representing a sandy
clay, of somewhat stony character. Sample No. 26, also a sandy clay soil,
\vaa procured from the farm Alexanders Kloof, about seven miles E.N.E.
of Jongens Klip, and said to be more fertile than the soil represented by
the sample taken at the latter spot. From a chemical point of view, how-
ever, there is no practical difference between these two soils, nor are they
very different physically. Crossing the River Zonder End, about 8^

miles N.N.E. of Alexanders Kloof, the samples numbered 27 and 28 were
taken from the farm Ganze Kraal, on the northern bank of the River
Zonder End, and south of the Zonder End Range. The latter of these
two samples is a stiff clay, while the former is a loose alluvial clay, darker
in appearance than the' other, on account of the larger proportion of
organic matter contained. The distance between the two points is not 160
yards, the boundary line being very distinct. Both soils are planted with
vines and fruit trees, while the grain lands are all on the south side of the
river. Proceeding about 5J miles eastwards, all along the river, Tyger-
hoek, the farm of the Misses Vynes, was reached, and here sample No. 29
was obtained on the south side of the river ; it represents a stiff clay, light
yellow in appearance. The next farm to be visited was The Oaks, or
Hartebeest Kraal, nearly seven miles from Ganze Kraal, higher up the
river in a west-north-westerly direction. Here sample No. 30, a stiff clay,
was collected.

North of Caledon the first place visited was Bok Kraal, about four
miles south of Genadendal. Here sample No. 2 was collected, also a stiff
clay soil. From the farm Zwart River, about 6^ miles soutli-west of the
last one, Nos. 3 and 4 were obtained : the former, a sandy clay, rather
stony, was taken from a hill slope, whereas the latter, more sandy, was
collected from the low lying lands in the valley : as the table of analytical
results will show, the sandier soil is in all respects tlie poorer of the two.
The last sample collected on this tour, No. 1, was taken from the farm
Middelplaats, about five miles north-west of Zwart River.

With the exception of samples 5, 10, 13, 15, and 19, none of which
had been under recent cultivation, all the above represent virgin soils.

(Privately collected,)

No. Field Cornetcy. Farm or Place. Collector.

31. Bot and Palmiet Rivers Houw Hoek. C. E. Pillans.

32. „ „ „
33

34. Zwart River. Zwart River.. J. W. Hartford,

35. Upper River Zonder End. Kale Ruggens. C. Lange.

37^ " ',', !!

38. Bot and Palmiet Rivers. Geelbeks Vlei. C. Lindholm.

39. „ The Request.

40. ,, Isaaks River. „

Three samples, Nos. 35, 36, and 37, were taken from a valley near the
>n Bridge, on Mr. Charles Leonard's stud farm, Kale Ruggens,
now known as Gloria, which is situated about five miles south-west of Vil-
liersdorp, on the River Zonder End. The estate is almost encircled by
sandstone hills, and Mr. Lange, the manager, declares that his best efforts
have followed the use of guano from the Government islands, with a top-
dressing of nitrate in the case of oats, and lime and sheep manure in the
case of lucerne, which, under such conditions, thrives very well indeed.
The surface soil at the locality where these samples were taken is a sandy
loam, with a clay subsoil ; the slopes adjacent to the valley consist of the
iisual alluvial soil, locally known as " klippers grond."

Nos. 38, 39, and 40 were collected from farms situated on the Table
Mountain sandstone, of which the Pa-ardeberg Mountains are composed.
As may be anticipated, therefore, they are, generally speaking, poor. In
the case of the soil from The Request it appears likely that the sample is

37

not sufficiently typical — in other words, that the soil it represents had un-
dergone some artificial improvement whereby it had been rendered chemi-
cally richer than the average soil in the vicinity. Nos. 31, 32, and 33
were collected at Houw Hoek, from the orchard of the late Mr. Aspinall :
the plot of land whence they were taken lies across the road from the house,
on the north-west side of the main road to Caledon. Here also, the soil
is, as a rule, poor, being derived largely from the sandstone of the Table
Mountain series, whereof the Houw Hoek Mountains are built up.

Sample No. 34 was collected on the farm Zwart Eiver, adjoining the
farm Sergeant's River, on the Genadendal road, about ten miles from
Genadendal. The soil is poor all round, like the sandy soil No. 4, officially
collected on the same farm.

The analyses of the foregoing forty soils are tabulated below : —

(Method II.)
Perc€ntaseof Soil sifted through 1 mm Sieve. Percenta^

no.

Water.

Organic
matter.

Chlorine.

Nitrog'en.

Lime.

Potash.

jrnoH.
phoric
oxide.

1.

•73

2-67

, '55

•084

•27

•13

•033

o

£4 .

1-06

3-85

•0061

•13

•034

•076

•038

3.

1-24

6-69

•0086

•15

•034

•13

•059

4.

•51

2-15

•093

•091

•018

•043

•038

5.

1-60

6*57

•0038

•17

•093

•050

•058

6.

T44

6'33

•0040

•15

•028

•056

•036

7.

2-04

11-71

•0038

•25

•083

•073

•032*

8.

1-42

6-60

•0056

•15

•030

•038

•058

9.

T48

7'57

•017

•15

•026

•098

•049

10.

1-27

6-43

•0034

•13

•098

•060

•12

11.

2-05

8-30

•0031

•19

•080

•078

•077

12.

•87

4-36

•0042

•11

•039

•036

•056

13.

1-37

5'25

•018

•10

•018

•055

•038

14.

1-27

4-94

•0042

•15

•15

•076

•056

15.

1-56

6-14

•0050

•15

•024

•073

•056

16.

1-49

5-81

•0021

•14

•038

•048

•072

17.

1-67

7-42

•0055

•20

•045

•087

•059

18.

1-28

5-21

•0070

•15

•040

•044

•051

19.

•88

3'37

•0049

•091

•054

•041

•046

20.

1-57

5'21

•0055

•13

•029

•24

•032

21.

•84

3-22

•0027

•11

•038

•024

•056

22.

T38

6'57

•0097

•17

•025

•045

•036

23.

T58

7-88

•0058

•20

•016

•072

•036

24.

1-93

7-93

•0034

17

•058

•068

"13

25.

T93

7-04

•0064

•18

•032

•071

•051

26.

1-87

7-19

•0034

•20

•045

•078

•061

27.

2-10

8'20

•014

•22

•058

•049

•061

28.

1-37

4-39

•0049

•098

•030

•061

•041

29.

•95

4-60

•0042

•13

•026

•042

•038

30.

1-19

6-04

•0037

•16

•041

•045

•056

31.

5-22

11-45

•005

•45

•066

•022

32.

1-25

2-4G

•066

•021

•033

•067

33.

•89

3-64

•022

—

•022

•012

•014

34.

—

—

—

—

•091

•036

•013

(Method I.)

Per cent, of

Field
Sample.

Percentage of Soil sifted through
1 mm. Sieve.

No. Fine earth. Water. Orgamc Chlorine. Nitrogen. Lime. Potash.

35.
36.
37.
38.
39.
40.

73'5
64-3
52-5
81-6

•47
•86
•34
•58
•81
1-01

matter.

1-77
2-92
1-26
3'36

•010

•008

•008

•0099

•009

•007

•101

•087

•080

•098

•15

•16

Percentage of Soil sifted
through £ mm. Sieve.

Phos-
phoric
oxide.

017 '037

008

008

014

24

09

•010

•006

•008

•044

•18

•014

•028

•015

•0089

•06

•05

No. 1 of the foregoing soils is undoubtedly brack or alkaline, and it is
significant that the alternative name of the farm Middelplaats, whence it
was taken,, is Brakfontein : the high proportion of chlorine in this sample
will be noticed.

It is not at all surprising to find how poor in all inorganic plant food
the primary sandstone soils of this Division prove to be. In the tables of
analytical results illustration hereof is afforded by the soils from Dasje&-
fontein (No. 10), Klipheuvel (No .11), Dunghye Park (No. 16), Paardeberg
(Nos. 19 and 20), and Klein Wolvegat (No. 22).

It will be seen that very many of the soils of this part of the Colony
are distinctly deficient in lime : the area in which they lie is almost en-
tirely hemmed in by sandstone. Further east, in the Bredasdorp Division,
and in the western portion of the Division of Swellendam the influence of
the sandstone ceases, and accordingly the soils in those parts contain more
lime.

CAPE.

(Officially collected.)

Field Cornetcy.
Diep River.

No.

1.
2.
3.
4.
5.
6. Cape Town.

Farm or Place.
Tokai.

Collector.
J. G. Rose.

Parliament House

grounds. A. Simons.

7. Tygerberg and Kuils River.

Bloemhof.

J.

Muller.

8.

„

,,

9.

,,

,,

10.

)y

J?

11.

Maastricht.

C.

F. Jurite.

12.

J?

?)

13.

„

F.

Blersch.

14.

J3

}J

15.

Eversdal.

C.

F. Juritz.

16. . „

,,

n

17. Durban.

Moerendal.

F.

Blersch.

18.

Johannesfontein.

5'

Part

For mechanical analysis of No. 38, see under "Physical composition of soils,"
VII.

CAPE DIVISION

No.

19.
20.

21.
22.
23.
24.
25.
26.

27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.

Field Cornetcy.
Durban

Palen and Rietvlei.

Koeberg No. 1.

Farm or Place.
Diemersdal

»

»
Phesante Kraal.

Vissers Hok.
Government land

north of Vissers

Hok.
Vrymansfontei.il.

»

Rondeboschjesberg.
Ongegund.
Altona.
Adderley.
Hooge Kraal.

Collector.

F. Blersch.
C. F. Juritz.

J. G. Rose.

Koeberg No. 2.

Klein Olifants Kop. C. F. Juritz.
Kalkfontein. „

Uitkyk.

Blauwberg.

Dassen Vallei.
Klein Dassen Berg.
Lange Rug.

J. C. Water meyer.

The first five samples were collected at the T'okai Convict Station:
No. 1 was a very poor sandy soil from the vicinity of the camp which had
been used for interning prisoners of war; it had just produced an oat
crop, and a similar crop was grown some five years previously, a small
quantity of guano having been applied for each crop. No. 2 represents a
dark soil which it was intended to use as a vineyard, and had been culti-
vated for the last twelve or fifteen years with garden crops. Farmyard
manure and guano had been applied at various times, and basic slag three
years before the collection of the sample : the old estate vineyard used to
be located here. The third sample was a moist black vlei ground, virgin
soil, but interspersed with palmiet roots. The next was a brown soil from
the rifle range garden cultivated for five years, and manured with farm-
yard manure and guano. The last of these five samples represents a
brown soil from the neighbourhood of the Porter Reformatory ; it had
also been under cultivation, but no details regarding the manures em-
ployed could be obtained. The last two samples proved exceedingly poor
in their reserve of plant food.

Within the Municipality of Cape Town, No. 6 affords an idea of a
cultivated and frequently manured garden soil : the sample analysed re-
presents the average of the soil within the grounds of the Parliament
Houses. Samples were taken from different points in the grounds and
well mixed together, and an average sample was then subjected to che-
mical analysis.

Four samples of soil from the Hon. R. P. Botha's farm Bloemhof, or
The Hope, were analysed. No. 7 was a yellowish sandy vineyard soil,
taken from north-east of the homestead ; it had not been manured during
recent years, and was found in practice to be very poor. No. 8 was taken
from another part of the vineyard, north of the homestead ; it was more
clayey than the foregoing, but mixed with coarse gravel : it is found to be
more productive than No. 7. No. 9 represents a poor and very sandy

40

soil; the sample was taken from the vineyard iiorth-west of the house.
No. 10 was a very stiff clay, also from the vineyard west of the house, and
proved to be fairly good in practice. In the collection of these four sam-
ples the soil for six inches nearest the surface was discarded, and the next
twelve inches collected for examination. A glance at the analytical
tables will show that the first three of these soils exhibit an all-round lack
of plant food, but that No. 10 is distinctly better than those in both lime
and potash : it may therefore be said that to this extent chemical analysis
here again confirms practical experience.

Samples 11, 12, 13 and 14 were taken from the farm Maastricht,
lying S.S.W. of the village of Durbanville. Rust had worked great havoc
amongst the crops for three consecutive years, but the season previous to
the collection of the samples had witnessed a remarkable immunity from
attack. Samples 11 and 12 were virgin soils, taken at a distance of about
3J miles from Durbanville. Nos. 13 and 14 were taken from cultivated
areas which had been infested with Erysiphe graminis in the wheat crops ;
these two samples were collected some time subsequent to Nos. 11 and 127
and the disease was then in full swing. The next farm visited was Die-
mersdal, and here too samples were taken on two different occasions.
Here also a visitation of rust, similar to that experienced at Maastricht,.
had taken place : two years before the farm was first visited for the col-
lection of soils the entire crop had been destroyed, but during the suc-
ceeding season the crop had not been affected at all. Nos. 20 and 21 were
virgin soils taken at a mean distance of about 2J miles north of Durbaii-
ville. No. 19 was taken on a later occasion, and under similar conditions
to Nos. 13 and 14. Nos. 17 and 18 were also collected from lands where
the grain had been badly affected at the time with the above-mentioned
parasitic disease. At Vrymansfontein, about 3J miles N.N.W. of Dur-
banville, samples 27 and 28 were taken, the former from a hill-side, the
latter from low lying ground. No. 29 was collected from a hill-side at
Eondeboschjesberg, 4-i miles N.N.E. of Durbanville, and Nos. 22 and 23,
from the farm Phesante Kraal, somewhat over three miles N.N.E. of the
village : of these two soils, the first was taken from a hill slope, the second
from a valley, No. 24 being taken from the same farm further east. Two
soils were collected on the farm Eversdal, about 1| miles from Durban-
ville.

On most, if not on all, of these farms, the principle of applying
manure in accordance with the needs of the soil and crop seemed to be
entirely ignored : to the land adjacent to the homestead farmyard manure
was, as a rule, applied, while guano was carted to lands at greater dis-
tances, or in less accessible localities, hillsides, for instance. It was all a
question of convenience. Only in one case, that of the farm Phesante
Kraal, was the use of artificial fertilisers practised. Thus sample No. 20,
although a virgin soil, represents a type usually worked with farmyard
manure, and No. 21 one on which guano is employed; and it is a remark-
able fact that the lands represented by this latter saanple were overrun
with Eumex acetosefla, or Steenbok zuuring, as the farmers call it, from
which noxious weed No. 20 was entirely free. There were general com-
plaints amongst agriculturists with regard to the trouble caused by thia
weed, and its marvellous persistence on certain tracts, and absence from
certain others seemed to be worth investigation.

In the more northerly portion of the Division, commonly known as
Koeberg, samples were taken from valleys on each of the farms Ongegund
and Vissers Hok, and one from the top of a small hill on the Crown land
(outspan) north of Vissers Hok. Two virgin soils were collected from hill
sides on the farm Hooge Kraal, No. 33 being found by experience to be
the more productive of the two. Kraalbosch — considered locally to be a
sign of richness of soil — grows on the soil represented by No. 33, which ia

4.1.

thought to be the best soil on the farm. It is a looser soil than No, 34,
the latter being more stony and apt to cake when the ground becomes
hard and dry : No. 34 is, however, more typical of the average Koeberg
soil. No. 35 was taken from a valley on the farm Klein Olifants Kop. Of
the three soils collected on the farm Uitkyk, it must be noted that the
first two, in common with most of the other samples collected from this
Division (with the exception, that is to say, of Nos. 1 to 10, 13, 14, 17, 18,
19 and 39) represent absolutely virgin soils, never yet cultivated, dug
over, or touched by manure, No. 37 being in the occupant's opinion, the
worst, and No. 38 the best ground on that particular farm. No. 39 waa
a sample of cultivated ground adjoining No. 38, and exactly similar there-
to in nature. The analyses afford some clue as to the basis of the fanner's
differentiation between these two soils : that which he described as the
best soil on the farm is finer in grain than the poorer soil ; it is more tena-
cious of moisture, contains a larger proportion of organic matter, and
more potash ; it has more than double the reserve of lime, and nearly six
times the amount of phosphoric oxide. The cultivated soil alongside the
richer of these two had apparently been exhausted of half its reserve
stock of phosphoric oxide.

Klein Dassen Berg, from which the virgin soil Not. 41 was collected,
is not a grain farm, cereals being grown in quantity sufficient only for
home consumption. The arable land of the farm is a sandy loam. The
farm is subject to the inroads of blown sands from the west coast. The
farm Lange Rug lies on the boundary of the Koeberg clays and the sandy
soils of the coast : sample No. 42 is a sandy clay from that farm.

Of the 42 soils described above, the last 36 may be said to summarise
the grain lands of the Cape Division; Nos. 7 to 24 affording an excellent
representation of the grain farms surrounding Durbanville, while the re-
maining eighteen are typical of the Koeberg grain district.

No.

Field Cornetc.

43.

Cape Town.

44.

J)

45.

) J

46.

9t

47.

Sea Point.

48.

,%

49.

Robben Island.

50.

Newlands.

51.

n

52.

53.

tt

54.

Wynberg.

"

55.

56.

}f

57.

58.

Diep River.

59.

60.

M

61.

62.

63.

64.

Down* No. 1.

63.

66.

j?

67.

68.

(Privately collected.)

Farm or place.
Molteno Reservoir.
Government Avenue.
Table Mountain.

Fernwood.

Ke nil worth.

Cape Flats.

Tokai Forest Plantation.

Collector.

W. Gilmore.
J. M. Stephen,

H. Meyers.

B. H. Holland,
D. E. Hutchins,

Princess Vlei.
Uitvlugt.

C. E. Pillans.

D. E. Hutchina.

4:2

Sample 43 was taken from the escarpment on the upper side of the
Molteno Reservoir. No. 44 was taken from a locality in the Government
Avenue, Cape Town, where the oak trees were dying off. The cause of
this was obvious: there was a fair amount of lime and potash in the soil,
which was also well stocked with nitrogen, and although it was very poor
in phosphates, this did not afford an 'adequate solution of the difficulty,
which appeared to be physical rather than chemical, and exactly illus-
trates the closing sentence of Mr. A. D. Hall's remarks quoted on page
18. The soil was a stiff bluish clay, and exhaled a distinct sour odour
when fresh and moist: the acidity was found to be '06 per cent., calcu-
lated as oxalic acid. In situ it was a stiff, damp clay, and this fact,
coupled with the presence of much organic matter, and the lack of aera-
tion, in the soil, undoubtedly accounted for the ill effects observed.

From the top of Table Mountain samples 45 and 46 were collected :
they were, of course, typical Table Mountain sandstone soils. Practical
experience had declared the soil there to be extremely poor; in fact,
according to the Conservator of Forests, it has been found that only
Pines and Acacias flourish on it, and that on a larger scale only the
cluster pine has been successfully grown. Even this tree failed when
planted on the soil represented by No. 46, although on No. 45 it grew
well : this was attributed to the fact that the latter had the better sub-
soil of the two.

Two soils were collected from private gardens at Sea, Point. One of
these, No. 47, was taken, from the surface of a piece of ground that had
been trenched to a depth of three feet for the cultivation of vines: here,
as in the case of the Parliament House garden soil, No. 6, several lots
.were taken from various parts of the area intended to be cultivated, and
a thorough mixture made.

Many of the soils in the last list were collected with the idea of ob-
taining some information relative to the chemical composition of forest
soils, and. of areas proposed to be afforested. Thus, one sample was taken
from the site of a projected tree plantation at Robben Island. This soil,
No. 49, yielded a rather small amount of potash, but the phosphate of
lime present in an available condition was unnaturally large in quantity;
as this was hardly to be expected in so sandy a soil it appeared to point
to the presence of a great deal of bone material : the sample was there-
upon examined microscopically, and quantities of minute fragments of
bones were found therein.

Four samples were collected at Fernwood, Newlands, and one, No.
54, was taken from a garden at Kenilworth : of these, Nos. 50, 51, and 53
had been under cultivation in former years, but not latterly, while No. 52
was meadow land which had been quite recently broken up. These five
soils practically all lack potash and phosphates.

Two samples of Cape Flats clays, Nos. 55 and 56, were collected on
the Wynberg Flats, near the Kenilworth Racecourse: of these. No. 5q
was a white kaolin or China clay, and the other a plastic, dark coloured
clay, which occurs, in a thin layer, above the kaolin. These beds of pure
white kaolin underlie an extensive area of the Cape Flats sands ; appar-
ently, however, these Cape Flats clays do not extend very far east of the
railway line. Sample No. 57 represents a similar white clay, which forms
the subsoil of the forest plantation on the Wynberg Flats.

Nos. 58 to 63 represent red Coustantia loams, and were collected from
the Eucalypt arboretum at Tokai. The arboretum comprises several
plots, containing not less than 40 six-year-old trees each, and, although,
never previously cultivated, the area was in farmer times probably over-
grown with Leucodendron (Silver tree) and Mountain bushes.

43

One sandy soil was collected from the vine plantation lying south-
east of Princess Vlei. The peculiar characteristic of this soil, and of a
large area whereof it is typical, is that stone fruits are not found to do
well, whereas apples and pears thrive upon it.

The clay underlying the Government Forest Plantation at Uitvlugt
is represented by samples 68 and 69 : these typify the portions of the
estate where the tree growth is finest, No. 68 being a yellow and No. 69 a
white clay. Nos. 65, 66, and 67 were sands from the Eppiiig Forest at
Uitvlugt, a Government plantation on one of the poorest parts of the
Cape Flats, where an area of about 4,000 acres has been planted with
Cluster Pine and Wattles.

The results of the analyses of the soils from the Cape Division are
tabulated below : —

(Method I.)

Percent, of
Field Sample

Percentage of soil sifted
1 mm. sieve.

through

Percentage of soil sifted
through \ mm. sieve.

No.

Fine earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

Potash.

Phos-
phoric

oxide.

1.

31-8

•42

1-87

•024

•042

•022

•016

•015

2.

71-7

•93

3'99

•034

•111

•046

•046

•078

3.

91-1

4'36

16-77

•068

•319

•030

•059

•066

4.

61'8

1-36

4-10

•034

•014

•012

•033

•023

5.

57-3

•60

2-59

•033

•042

•014

•021

•017

6.

62-2

2'93

5-44

•0113

•120

•532

•129

•095

7.

98-3

—

—

—

—

•080

•028

•024

8.

82-7

—

—

—

—

•088

•017

•027

9.

96-8

—

—

—

• —

•086

•015

•024

10.

94-1

—

—

—

—

•160

•062

•028

33.

82-1

1-78

4-23

•0084

•448

•608

•887

•046

34.

82-4

1-33

4-09

•0142

•532

•107

*551

•035

43.

96-0

ri4

8-47

•0304

•014

•008

•on

•065

47.

66-8

2-36

7'46

•012

•103

•194

•359

•139

54.

82-4

5-76

14-97

•001

•189

•042

•010

•063

55.

57-1

1-02

8'47

•0106

•043

•086

•015

•176

56.

84-5

2-11

6-82

•0035

•057

•122

•008

•033

58.

63-0

2-28

6-77

•0057

•071

•018

•050

•046

59.

67-6

3-01

6-80

•0152

•057

•020

•047

•049

60.

63-5

2-07

7'40

•0304

•050

•024

•037

•040

61.

50-7

2-19

8'43

•0142

•078

•012

•047

•036

62.

45'2

1-96

8-31

•0230

•057

•030

•041

•063

63.

50-2

2-25

9-29

•0113

•071

•010

•035

•042

44

(Method II.)
Percentage of soil sifted through 1 mm. sieve.

Percentage of soil sifted through
3 mm. sieve.

No.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

Potash.

Phos-
phoric
oxide.

11.

1-33

15-50

•054

•128

•48

•045

•028

12.

2-97

10-52

•057

•201

•64

•27

•028

13.

2-09

6-44

•0091

•21

•157

•on

14.

•65

•26

•0090

•12

•125

•012

15.

1-37

6-94

•0053

•134

•39

•12

•044

16.

T75

7-64

•0028

•134

•35

•026

•062

17.

1-05

4-15

•0024

•20

•152

•027

18.

1-42

7-49

•on

—

•26

•036

brace.

19.

•69

5'37

•0019

—

•18

•167

•027

20.

1-03

4-60

•021

•106

•23

•14

•028

21.

1-22

6'67

•0074

•134

•25

•27

•019

22.

1-38

5-31

•0021

•089

•23

•043

•044

23.

•63

2-84

•0021

•123

•12

•025

•032

24.

1-12

5-79

•0060

•084

•32

•023

•017

25.

1-61

4-16

•0024

—

•24

•43

•035

26.

•27

•51

•0005

•056

•046

•039

•0038

27.

•70

2-85

•0095

•056

•22

•35

•020

28.

•44

2-11

•0053

•061

•11

•12

•020

29.

•94

4-00

•0021

•044

•15

•16

•026

30.

1-19

3-92

•0064

—

•23

•24

•026

31.

1-40

4'35

•0006

—

•13

•071

•062

32.

1-60

2-05

•0026

•061

•061

•070

•019

35.

•81

1-73

•0004

•061

•095

•038

•023

36.

•62

1-64

•0018

—

•061

•036

•017

37.

•96

2-36

•021

—

•067

•065

•013

38.

T65

4-20

•0016

— .

•16

•093

•076

39.

2-04

2-57

•0028

•16

•098

•040

40.

•96

2-59

•0013

•070

•094

•026

41.

•24

•81

•0015

•035

•061

•021

•029

42.

•34

2-17

•037

•028

•057

•030

•017

44.

4-84

6-81

•053

•154

•186

•111

•008

45.

•92

5'72

•0033

•126

•033

•014

•043

46.

•35

1-89

•0034

•070

•053

•010

•035

48.

1-46

3-20

•021

•084

•154

•215

•013

49.

•78

3-19

•010

—

3-320

•085

3-370

50.

3-78

6-69

•0008

•184

•188

•044

•054

51.

2-86

5-10

•0003

•089

•065

•069

•025

52.

2-66

5-34

•0007

•117

•131

•040

•027

53.

1-82

3-26

•0005

•044

•071

•044

•045

57.

•92

6-41

•0049

•042

•076

•035

•0015

64.

•05

•17

•0010

•049

•024

•on

•012

65.

—

—

—

—

•087

•049

•012

66.

—

—

—

—

•040

•014

•007

67.

—

—

—

•046

•016

•009

68.

2-81

5-70

•0113

•07

•226

•028

trace.

69.

1-56

4-11

•0624

—

•130

•0079

trace.

To refer once again to the five samples of soil — Nos. 13, 14, 17, 18,
and 19 — collected in the Durban and Tygerberg Field-cornetcies, from
areas where Erysiphe graminis had infested the wheat crops; the analyses
indicate that the average composition of the soils from the cultivated lands

where the disease had. made its appearance was poorer all round than that
oi the virgin soils from the same locality: vide the following: —

Average composition jn-r ccut. of
virgin soils. cultivated soils.

Lime '291 '194

Potash -133 -127

Phosphoric oxide '031 '015

This shows to some extent the soil exhaustion that had taken place :
the crops were apparently badly in need of phosphatic material, and it
would not be surprising if this had a great influence upon their capacity to
resist the attacks of parasitic diseases.

In this connection I may aptly quote Professor P. MaoOw&n, D.Sc.,
F.L.S., formerly Government Botanist, who was professionally investigating
the fungus disease at the time when the analyses just mentioned were being
made. Dr. MacOwan observed : —

"Phosphate of lime is the one thing needful as mineral food for all cereals —
wheat, barley, oats, mealies, and rye. Yet, beginning with a poor supply of it,*
there are hundreds of farms where cereals have been taking the phosphate out of the
soil every year for a quarter of a century. The inevitable result has come about.
Stinted in phosphates, the corn grows year by year more weakly in constitution, stools
less, gives lighter ears, gives shrivelled grain. All this is the naturai result of
phosphate starvation. On such debilitated plants the parasitic fungi and insects make
their usual attacks. The plants have so little vitality that they cannot bear the
injury and live. Naturally they succumb. The cure is restoring the original per-
centage of phosphates, to make the land what it was before five and twenty crops
had each carried away a share of this element of its fertility."

On comparing the Koeberg soils with those around Durbanville, it will
be observed that many of the latter contain more available lime than the
former do. The reason hereof has not yet been investigated : the underly-
ing rock being in both cases of the same nature, it appeared improbable
that this could have anything to do with the variation. Posaibly the
larger proportion of lime in the more southerly soils was due to the finely
divided pao-ticles of blown searsand, which extend across the Cape Flats
from the southern coast.

CARNARVON.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector,

1. No. 5. Jakhals Kolk. C. McMillan.

o

•*•• >j » »

These two soils were collected respectively from Lots 111 and 94, near

Tan Wyks Vied. The former had been under cultivation for 17 years,

but the latter was virgin soil. As they will be dealt with more fully in

* The italics are my own ; notice the small amount of phosphoric oxide in the
virgin soils.

connection with the subject of alkali in soils (see pages 185-7), the follow-
ing analyses will suffice at this stage : —

Ko.

1.

2.

Percent, of
Field Sample.

Percentage of soil sifted through
1 mm. sieve.

Fine earth. Water.

97-9
94-3

6-25
2-43

6-58
3-48

Percentage of soil sifted
through | mm. sieve.

Phos-

Chlorine. Nitrogen. Lime. Potash. phoric

oxide.

1-542 -084 -876 -399 -228

-0856 '070

116

29

193

These soils — as is usually the case with " brack soils — are well sup-
plied with plant food.

CATHCART.

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.

Field Cornetcy.
No. 3.

No. 4.

V

No. 5.

No. 4.

No. 6.

(Officially collected.)

Farm or place.
The Dales.

Exwe'll Park.

Spanover.

>j

Side Spur.
Anta.

Braemar.

j>
Stoneridge.

Inverthorn.
Blackpool.

»

»
Sledmere.

»>
»
Clapton.

Collector.
St. C. O. Sinclair.

As stated on page 36, the investigation of the soils of the Western
Province had nob progressed very far ere requests began to come in from
the east for similar investigations in that portion of the Colony. It was
in response to such requests that Cathcart and the neighbouring districts
were visited. It had, therefore to be borne in mind, in taking samples
in the Cathcart Division, that, although it was the intention that the
analyses of these samples should form part of the systematic investigation
of the Colony's soils, yet this area was being surveyed at the special re-
quest of the Eastern Province Fruit-growers' Association. It was found.

47

expedient, therefore, to concede, to a greater extent, at any rate, than
would otherwise have been the case, to the wishes of the individual far-
mers of the district in visiting special localities, and in. taking samples
from places selected by them. At the same time, samples representative
of the different types of soil occurring in the Division were also collected.

The soil of the Cat-heart Division, compared with that of Komgha,
which will be dealt with subsequently, is not so micaceous, but is, on the
other hand, more calcareous. Dolerite, although largely present, does
not appear as plentiful as in the Kom°'lia Division. The Cathcart soils
seem to be naturally derived from the upper rocks of the Karroo system.

No. 1, a sample of black loamy valley soil, was collected from the
farm The Dales. It appears to extend to a. depth of about eighteen
inches, and rests on a subsoil of yellow clay. The veld on both sides of
this valley is reported to be " sweet." No. 2 represents the hill-side sail
of the same locality. These two soils, together with Nos. 17, 18, 19 and
20, are representative of the area known as the Bontebok Vlakte, the
chief grain-producing portion of the Cathcart Division, and stretching
over practically the whole of its south-eastern part.

Nos. 3 to 8 were collected on the farm Exwell Park. Above the
homestead is a valley whence samples 3 and 4 were taken : as it was in-
tended to use this land for an orchard, those samples were taken from
the mixed contents of holes about 22 inches deep in each case. By thus
departing from the usual method of soil-collecting already detailed, a
better idea of the suitability of the soil for orchard purposes would be
obtained. Physically No. 4 differs from No. 3 in being slightly more
clayey. No. 5, a sample of clayey soil rich in organic matter, was taken
from ploughed land just above a lately constructed dam. The subsoil
here was of a more sandy nature. The soil represented by this sample is
capable of retaining a large amount of waterj due to the proportion of
clay and organic matter present therein. No. 6 represents a mixture of
surface and subsoil from the patch just referred to. No. 7 is a black soil
taken from lands below the homestead and adjacent to the railway line.
No. 8, which is also a black soil, represents the soil of the Victoria Or-
chard. This land had been in constant use for some years, and was re-
ported to be very fertile, mealies, oats, and wheat being amongst the
crops successfully borne during the last three years. These six soils
represent the eastern portion of the Waku valley, tlie main fruit-produ-
cing area of the Cathcart Division. The soils of this part of the country
contain a large amount of calcium carbonate, which is visible in the form
of " drip lime."

No. 19, taken from the farm Stoneridge, represents the valley soil
lying between the homestead on that farm and the Thomas River. No.
18 is representative of the hill-side soil near the homestead. No. 17 was
taken from cultivated land am the same farm for the purpose of compari-
son with Nos. 18 and 19 : the land from which it was taken had borne
oats and mealies and, to judge from the analytical figures, cultivation
had left its mark in a diminution of the stock of available phosphoric
oxide. It was intended to utilise this soil for grain, but in that event
it would obviously stand in great need of a phosphatic fertiliser.* No.
20, was taken from a sandy soil, situated behind the homestead and pro-
posed to be used as a vineyard, a purpose for which it did not appear un-
suitecl : it proved to be rather poorly supplied with phosphates.

No. 21 was collected, on the farm Inverthoni, from one of the older
river terraces of the Thomas River. No. 22, a sample of black soil, was
taken on the farm Blackpool, just below a ridge of doleritic boulders on

*Cf. Prof. MacOwan's remarks relative to lack of phosphates in grain soils at
Durbanville, quoted on page 45.

48

the slope towards the Thomas River. The subsoil was reported to be a
clay. No. 23 was taken from the same slope, but beyond the influence of
the doleritic ridge : it is a more sandy soil, lighter in colour than No. 22.
It was stated that barley and rye did not thrive on the lands represented
by it. No. 24 is a sample of doleritic soil lying further to the east.

On the farm Spanover two samples were taken, one, No. 9, represent-
ing the garden soil lying south-west of the homestead, and another, No.
10, the grain soil north-east of the latter. No. 11 was a stiff black soil,
representing a very productive tract of land whereon potatoes especially
were reported to do well. The soil No. 12, collected on Mr. W. H. Grout's
farm Anta, was of a sandy character, and intended to be devoted to the
cultivation of potatoes. No. 13, from the same farm, represents a rather
stiff alluvial soil of good depth. No. 14 was taken from the same slope,
but nearer to the river; the soil here is dark in colour and of a more
clayey nature than where No. 13 was collected. No. 25 is a sample of

DIVISIONAL MAP

or

CATHCART

/v

< \

(
?

soil from cultivated land just below the homestead at Sledmere. The suib-
soil at this place was found to be a yellowish clay, while the surrounding
surface soil is inclined to be sandy. No. 26 was taken from the same
land, but nearer to the bottom of the slope. No. 27, a soil rich, in organic
matter, was taken at the bottom of the slope whence the previous two
samples were collected. The portions of this slope represented by these
three soils differ considerably in their productiveness when used for the
same purpose. On the area represented by the last-mentioned sample,
wheat, for instance, does not seem to thrive. No. 28 was collected on the
farm Clapton. It is a stiff black soil, lying near to the homestead, in
the vicinity of which doleritic boulders were noticed. This soil, together
with No. 19, typifies the N.N.E. portion of the Division. Cattle are said
to be in the habit of licking the watercourses in this portion of tone dis-
trict. This fact was thought to be due to the presence of a considerable
amount of common salt in the soil; attention was not drawn to this when

41)

passing through any other part of the Division, it may, nevertheless, be
the ca«e that such a habit is prevalent throughout a much wider area,
and not peculiar to the stock in the parts where it was specially noticed.

Samples 15 and 16 are dark valley soils from the farm Braemar,
which lies north-east of Cat-heart. The latter is the more clayey of the
two. On the lands represented by these samples mealies were reported
to do well.

The following analytical figures were obtained : —

:N7o.

1.

2.

3.

4.

5.

€.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.

Percent, of

Field

Perce:

Sample.

Fine earth.

Water.

99-0

1-50

98'8

•62

96-5

1-36

97-1

1-52

99-4

2-78

99'4

2-68

99-0

2-30

99-3

2-70

98-0

T48

93-6

1-53

96-0

1-29

99-0

•55

99-3

•70

98-8

•85

99-0

1-12

98'9

1-19

97-6

1-60

94-9

1-39

98-1

1-06

97-2

1-18

97'7

1-13

98-7

2-73

97-0

•65

94-8

T87

99-0

1-10

97-4

1-50

98-6

2-65

97-3

2-57

(Method I.)

Percentage of Soil sifted through
1 mm. Sieve.

Organic
matter.

3-77
1-84
2-30
2'99
5-82
5-24
5-12
5-88
3-29
3'97
3-20
1'90
2-40
2'56
3-06
3'59
3-35
3-07
2'85
3-30
3-05
5-55
1-76
4-72
3'34
3'99
4-98
7-60

Chlorine. Nitrogen.

Percentage of Soil sifted
through £ mm. Sieve.

Phos-

Lime. Potash. phoric
oxide.

•0244
•0145
•0376
•0334
•0731
•0341
•0341
•0721
•0060
•0287
•0028
•0025
•0129
•0042
•0240
•0042
•0325
•0290
•0268
•0275
•0197
•0057
•0050
•0176
•0046
•0042
•0042
•0127

•098
•053
•070
•084
•126
•112
•112
•171
•081
•084
•133
•070
•084
•105
•140
•133
•076
•090
•076
•056
•056
•182
•028
•140
•112
•119
•091
•154

•200
•014
•059
•111
•446
•295
•151
•358
•330
•162
•128
•124
•088
•098
•054
•070
•274
•142
•152
•142
•126
•290
•072
•086
•124
•116
•148
•252

•033
•014
•066
•099
•172
•188
•179
•208
•121
•122
•095
•162
•070
•078
•067
•042
•057
•070
•066
•080
•090
•175
•137
•102
•160
•124
•126
•139

•040
•015
•025
•038
•088
•060
•082
•128
•031
•026
•036
•028
•017
•023
•031
•026
•018
•031
•033
•028
•027
•075
•029
•033
•022
•024
•035
•040

On the whole, the plant food content in these Cathcart Division
soils is more satisfactory than in the neighbouring Division of Komgha :
taken all round, they possess a fair amount of available lime and potash,
l>ut the soils, which are all more or less of a clayey nature, exhibit de-
cided poverty in respect of phosphates, if judged by European and Ameri-
can standards, although in this connection a remark already made must
be borne in mind,* namely, that proportions of phosphoric oxide which
would be deemed inadequate in Europe, in this Colony frequently suffice to
yield satisfactory returns. There appears to be generally a larger amount
of chlorides present in these soils than that found in the Western Pro-
vince soils.

See page 12.

50

In the soils of Field-cornetcy No. 3 phosphoric oxide averages better
than in the rest of the Division. Of this entire series Nos. 2 and 15 are
the worst; they show an all-round deficiency. The poorest soils in the
Division, from a chemical point of view, are those which represent the
Bontebok Vlakte : Nos. 2 and 15 have just been referred to; Nos. 1 and
16 lack potash and phosphoric oxide, and Nos. 17 and 18 are also badly
supplied with phosphates.

The fertile soil of the Victoria orchard had not received any manure
for three years ; it is one of the best all-round soils met with in the course
of these investigations. In every respect its available plant food is satis-
factory : its nitrogen is good, the lime is normal in amount, and so is the
potash and phosphoric oxide.

One of the sandier soils of the Division, No. 23, which represented
part of the incline towards the Thomas River on the farm Blackpool,
proved to be poor in nitrogen, lime, and phosphoric oxide, with but a
moderate amount of potash ; the failure of the barley and rye crops 19
thus explained, and here once more the analytical results bear out prac-
tical experience. The extent of doleritic soil to the east of this contains a
larger proportion of organic matter and more nitrogen, but less potash,
and is practically just as badly off for lime and phosphates. The produc-
tive potato lands represented by sample 11 are evidently fairly well sup-
plied with lime and potash, and also show a satisfactory nitrogen content,
but phosphoric oxide is low in amount. If potatoes be constantly grown
on this land, its reserve of potash, unless renewed artificially, would suffer
speedy exhaustion.

CERES.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Ceres. Riet Vallei. Dr. R, J. Reinecke.

2- „ » j,

o

°- )> » »

^' ?> V »»

5. Bokkeveld. The Oaks. 0. A. Ohlssou.

6- ,, „ »,

7

' • »> V »

Four samples of virgin soil were collected on the farm Riet Vallei, in
the outskirts of Ceres village, about half a mile east of the Dwars River.
At the particular places where the samples were taken, the soil had never
been cultivated or manured, and is overgrown with sour grass. The sur-
face soil is sandy and the subsoil clayey. In the locality in question soil of
this type, if well manured, is stated to have been found capable of the
most liberal cultivation : vines and tobacco do well ; without manure
lucerne does not thrive, nor can one be surprised at this, considering the
poor supply of lime in the soil.

Three analyses were made of soil from Mr. A. Ohlsson's farm The
Oaks: No. 5 represents ground that had been trenched, and No. 6 was a
mixture of soil collected on two different parts of the farm.

51

The table below gives the results of the chemical analyses of the above
soils : —

(Method I.)

P*Fie?d °f Percentage of Soil sifted through Percentage of Soil sifted
t^m je 1 mm. Sieve. through £ mm. Sieve.

Phoa-

No. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide

97-0
80-7
98-3
98-0
80-5
80-2
94-2

•36
•37
•27
•56
•33
•98
•64

1-93
1-86
1-41
2-55
5-48
4-28
2-81

•0049

•0130

•0067

•0123

•039

•019

•014

•077
•091
•084
•112
•102
•074
•074

•072
•010
•010
•008
•052
•034
•042

•015
•015
•012
•017

•024
•025
•020

•022
•028
•010
•032
•040
•041
•052

The areas represented by these soils are largely influenced by the
presence of rocks of the Table Mountain series, and the first four, in par-
ticular, lying in a tract of country practically surrounded by sandstone
mountains, can hardly be expected to be otherwise than poor in the chemi-
cal constituents of plant food.

CLANWILLIAM.

(Privately collected.)

No.

1.
2.
3.

4.
5.
6.

7.

Field Cornetcy.
Bidouw.

Farm or place.

Matjes Rivier.

»

Wupperthal.
Beukes Kraal.

>j
Kromjne Rivier.

Collector.

P. Bomemisza.
Dr. Simon.
P. Bornemisza,

«
Dr. Simon.

»
P. Bornemisza,

In this Division soils were collected from farms in/ tha Bidouw Field
Cornetcy, where tobacco is largely grown. The local practice has been,
to manure heavily, generally with goat manure, those areas which were
used for tobacco cultivation.

The results of the analyses are given below : — •

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

No.

1.
2.
3.
4.
5.
6.
7.

Water.

•91

•67
2-10

Organic,
matter.

4-18

6-42
8-60

Chlorine. Nitrogen.

•024
•023
•003
•134
•075
•170
•020 —

Percentage of Soil sifted

3 mm. Sieve.

Phos-

Lime.

Potash.

phoric

oxide.

•32

•13

•oio

•73

•11

•067

•11

•04

•09

•80

•25

•01

•34

•21

•63

•88

•78

•006

•20

•12

•03

Remembering that these represent- soils under actual cultivation, and
this in connection with a very exhausting crop, it may be obs&rved that
they all contain at least a fair amount of lime, and that this plant food
is present in quite satisfactory proportion in most of the samples taken.
The soil from Wupperthal was deficient in potash, but at Beukes Kraal
the quantity of this constituent was normal, while of the Kromme Rivier
sample No. 6 was very well stocked in this particular. Phosphoric oxide
is either wholly inadequate, or present only in moderate proportion right
through, sample No. 5 excepted : in the latter case the amount is, rela-
tively speaking, so extraordinarily high as to suggest the inference that it
is due to the method of fertilising already alluded to ; in fact, seeing that
in the soils of this Division the plant food constituents would not be ex-
pected to be present in any larger proportion than in the soils of the Ceres
Division, where similar geological influences prevail, it is probable that
nearly all these Clanwilliam soils had been altered in chemical composi-
tion by extensive fertilising agencies.

No.
1.
2.
3.
4.
5.
6.

COLESBERG.

(Privately collected.)

Field Cornetcy. Farm or place.

Upper Hantam. Oorlogspoort.

Collector,
W. Webb.

Six samples of virgin soil were collected at various localities on the
farm Oorlogspoort, formerly known as Zeekoegat. The surface soil of the
farm is ordinary Karroo veld, the underlying formation being shale, alter-
nating with lime. The chemical composition of the soil is apparently iiu-
fluenced by the rocks of the Stormberg series.

The analyses of these soils resulted as shown in tihe following table : —

Percent, of

Field

Sample.

(Method I.)

Percentage of Soil sifted through
1 mm. Sieve.

No. Fine earth. Water. Organic Chlorine. Nitrogen.

matter.

1. 91-1 2-49 3-73 '006 '064

2. 92-7 2'67 3'69 '007 '064

3. 93-8 2-94 3'84 '007 '071

4. 97-6 4'40 6-51 '003 '165

5. 98-1 2-33 3-84 '007 '101

6. 94'7 3-51 3'81 '006 '064

Percentage of Soil sifted
\ mm. Sieve.

Phos-

Lime. Potash, phoric
oxide.

•046
•094
•114
•506
•066
•216

•156
•165
•183
•203
•184
•136

•068
•072
•070
•115
•073
•069

These soils, somewhat similar in their geological origin to those of
the Albert and Aliwal North Divisions, are, like the soils of the Divisions
named, all rather fine in texture, but better, in respect of potash, than the

53

Albert soils analysed. They all contain a satisfactory proportion of pot-ash,
and a fair amount of phosphoric oxide. No. 4 is well supplied with
nitrogen and lime, and all the others have fair percentages of nitrogen.
Lime is the only element of plant food that can be described as actually-
lacking iri any of these soils; in this respect the soils represented by
samples 1, 2, and 5 were defective.

ELLIOT-SLANG RIVER.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Lutha. S. L. Hart.

o

*•' » »

>j »

4. Ettrick. W. F. Beadon,

Three samples of soil taken at Lutha, in the Elliot District, midway
between Elliot and Indwe, were analysed. In the valleys the surface soil
is usually black, with a pot-clay subsoil commencing about twelve inches
below the surface : this black surface soil is represented by No. 2. On the
hill-sides the soil is either light and sandy, or deep red, according as the>
ledges above consist of sandstone or ironstone. Samples of both these
types of soil were collected. No. 1 on the list was a fine sandy soil col-
lected from a hill ; practical experience had found it fairly good, when-
fertilisers were applied* but the deep red soil — of which No. 3 was a
specimen — is said to be the best. The subsoil, in both cases, is a clay, com-
mencing from one to ten feet below the surface. Most of the cultivated
farm lands are on the hills ; the soil there is cooler, more retentive of
moisture, and stands drought better. The valleys provide heavier crops
during good seasons, but the valley soils, being very shallow, need frequent
rains if a fair crop is to be ensured. Where available, the alluvial de-
posits of old river beds are best of all. As for the valley soils, it has been
found necessary to abandon them for cereal crops, and devote them to
lucerne, as the continual ploughing and heavy rain storms denude them
very rapidly and form large " sloots," a condition which does not result in.
the higher and sandier localities.

It may be remarked that the geological formations in the neighbour-
hood of Ida ('Mbokotwa Commonage), near to which the farm whence
these samples were taken is situated, belong principally to- the Stormberg
series ; that is to say, they are chiefly composed of Molteno sandstones,
and of red beds resembling the Burghersdorp beds. TTie former are natur-
ally poor in plant food, and the first sample on the list is a type hereof;
the Red Beds, on the other hand, have a better reputation, while the
dykes of dolerite which intersect in the sandstones, would also tend to im-
prove the soil by the addition of lime.

Another sample of soil was taken from dry agricultural lands on the
farm Ettrick, and represented fairly the soil in the neighbourhood, on
which, it is said, nothing will grow. Chemical analysis shows it to be
practically as poor as the sample of sandy soil from Lutha, and thus fully
confirms the agriculturists' views.

*Cbeuncal analysis shows the natural condition of the soil to be poor all round

The following are the analyses of the above mentioned soils: —

(Method I.)

i.

2.

3.*

4.

Percent of
Field
Sample.

Percentage of Soil sifted through Percer
1 mm. Sieve. thro

Fine earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

98-1

1-25

2-72

•0035

•073

•036

97-7

8-20

19-67

•0187

•263

•482

90-8

3-31

10-60

•0057

•170

•154

98-0

•99

4-38

•006

•074

•056

of Soil sifted
through i mm. Sieve.

Pho,-

Potash. phone
oxide.

•043
•153
•111

•049

•020
•110
•088
•038

No.
1.
2.
3.
4.

FORT BEAUFORT.

(Privately collected.)

Field Cornetcy. Farm or place. Collector.

Fort Beaufort. Fort Beaufort Asylum. Medical Superintendent.

The only samples of soil from this Division that have been analysed
are four from the grounds around the Fort Beaufort Asylum. Nos. 1 aaad
2 were taken from the garden at the male asylum, and Nos. 3 and 4 from
the female asylum garden. The site is in a valley which receives the drain-
age of surrounding hills about a quarter of a mile distant : the subsoil is
red, with fragments of limestone, and overlies a blue shale. All the sub-
soil water in the valley is reputed to possess a tulphurous odour.

The analyses of these soils resulted as below : —

(Method I.)

No.

1.
2.
3.

4.

As these soils have evidently been affected by cultivation and fertilis-
ing, very little in the way of general conclusions regarding them can be
drawn. Apparently in their original state they were not well supplied
with phosphates, for in Nos. 1 and 2 the amount of phosphoric oxide is
still low. The lime, organic matter, and nitrogen in No. 4 are unusually
high, no doubt as a result of treatment with fertilisers; for the rest it
is scarcely possible to tell whether the plant food constituents found are
due to natural causes or to manipulation, so that the practical inutility of
basing general conclusions upon analyses of cultivated soils is once again

Percent, of
Field
Sample.

Percentage of Soil sifted through
1 mm. Sieve.

Percentage of Soil sift«d
£ mm. Sieve.

Phos-

Fine earth.

Water.

Organic.

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

87-4

2-20

4-05

•009

•102

•154

•188

•070

87-8

3-24

5-45

•008

•147

•190

•189

•077

89-7

2-40

4-66

•005

•132

•444

•263

•191

77-6

3'49

10-50

•007

•442

1-058

•197

•536

*For mechanical analysis of this sample see under heading " Physical composi-
tion of soils" (Part VII.).

55

exemplified. Taking the soils, however, as they stand, none of them can
be said to be really deficiemt in any form of plant food, and No. 4, as al-
ready indicated, is rich in nitrogen, lime, and phosphates, and is more-
over satisfactorily stocked with potash. No. 3 is likewise rich in lime and
well supplied with potash and phosphates, the proportion of nitrogen
being normal. The remaining samples, with the exceptions noticed, are
satisfactorily supplied with all the principal plant food constituents.

GEORGE.

(Officially collected.)

No. Field Cornetey. Farm or place. Collector.

1. Outeniqualand. Uitkyk. J. Muller.

2. ,, Greak Brak River. ,,
3- » jj ?>

Diep Kloof. C. F. Juritz.

5. ., Palmiet. „

6. „ Modder River. „

7. George. Hans Moes Kraal. „

8. The Lakes. Outspan Reserve „

9. „ Wooclville.

10. „ Diep River.

11. Klip and Doom Rivers. Groot Doom River „
12

5J » »

13. „ Klip Drift. „

14

jj » y>

*0« jj » ,,

16. Before Long Kloof. Grootfontein. „

17. „ Schoonberg. „

18. „ Ganze Kraal. „

19. „ Kykoe.

The hamlet of Great Brak River is distant about 2| hours' cart jour-
ney from the town of George, in a south-westerly direction : a typical
sample of the intervening area is No. 1 in the above list, a " sour " soil
taken from Uitkyk, a farm in the George Division, but near to the Great
Brak River, which forms the natural boundary between the division
named and that of Mossel Bay. No. 2, a red and more sandy soil, typifies
another portion of the area referred to; it was selected from new lauds
on Mr. H. Barnard's farm, quite close to the main road, on the heights
above Great Brak River. From the lands of Mr. C. Searle, M.L.A., on
the left bank of the river, No. 3, in appearance a very similar soil, was
secured.

After leaving Great Brak River for George a short extent of granitic
soil is passed over, speedily, however, succeeded by a red pot-clay. Later
on this in turn gives place to a blacker and looser clay, of which a sam-
ple, No. 4, was collected on the farm Diep Kloof, about four miles by road
from Great Brak River. This sample is typical of the soil of the sur-
rounding farms; by practical farmers it is considered to be rather poor,
and chemical analysis quite bears out this testimony : when manured it
answers well, a fact which tends to show that its defects are wholly
chemical. Another sample of the same class, although somewhat lighter

and sandier, No. 5, was collected four miles further on, at Palmiet, part
of the farm Moeras River. Between these two points there are a few
patches of granite soil, but the predominating clay is soon reached again.
Sample No. 6 was procured at Modder River, near the entrance to the
village of Blanco. This soil, which is chemically poor except as regards
nitrogen, appears to be derived from the band of slate which stretches
eastwards, north of Blanco : it is typical of the stiffer clay soils which
skirt the southern slopes of the Outeniqua Mountains, between the Great
Brak River and George, the coast soil being of a sandy nature and sweet,
whereas the intermediate belt, from which samples 3 and 4 were taken, is
a mixture of the two classes of soil. Of the sandy belt a sample, No. 7,
was taken from the granite formation near Pacaltsdorp, on the farm
Hans Moes Kraal, at a distance of about five miles by road from George:
this also is an all-round poor soil. A sample of sandy clay, No. 8, resem-
bling the soil taken at Blanco, and lying on an extension of the band of
slate already referred to, was collected from the Outspan Reserve about
12 miles by road east of George. Another similar sample, No. 9, was

* I

BEFORE LONG* KLOOF

DIVISIONAL MAP

GEORGE.

taken about three miles nearer Knysna, from Woodville, the faxm of Mr.
J. Stevens. Tobacco has been cultivated here to some extent, and good
results are said to have been achieved as regards potatoes by the use of
basic slag. At Difp River, three miles from Woodville, sample No. 10
was collected. This soil, together with Nos. 1 and 2 of the Knysna. Divi-
sion soils, are representative of what has been called the intermediate
b^lt of soil.* According to the analysis the soil of which No. 10 was a
sample is apparently well supplied with nitrogen, but poor in other
respects.

Returning to George, and crossing the Montagu Pass, where Table
Mountain sandstone is entered upon, a sandy tract of country is passed
over on the farms Doom Rivier and Soldaats Kraal overgrown with rhe-
noster bush, alternating further on with proteas. Here we touch the
western end of the long valley known as the Long Kloof, which stretches
between the sandstone mountains for a distance of over one hundred
miles, through this division and the adjacent division of Uniondale into

See also page 29.

the H/umansdorp Division beyond. The soils in this valley are mostly
derived from Table Mountain sandstone, and are only saved from extreme
poverty by the presence of the Bokkeveld beds which here and there in-
crease the amount of lime otherwise in the soil. On the farm Klip Drift
two samples of soil were collected ; No. 15 an alluvial (vlei) soil, and No.
14, a soil from the hill behind the homestead and about 400 yards from
No. 15. This is succeeded by a Karroo soil, extending westwards to the
farm Groot Doom Rivier, where another sample, No. 11, was collected,
the ground being here of a more stony nature: this accounts for the
small proportion of " fine earth " found. The farm, it may be mentioned,.
takes its name from the mimosas which abound in the neighbourhood.
Of the Karroo soil, a sample, No. 12, was taken on the return journey to
Klip Drift, about two miles back, and another, No. 13, half-a-milei from
the homestead on the last-named farm.

On the road to Uniondale a series of samples representative of the
Long Kloof soils was taken, covering a stretch of about seventy milea.
The first of these was on the farm Grootfontein, 13j miles from Klip
Drift, No. 16, a loose though somewhat stony soil, where rhenoster bushes
prevail. The latter remark applies equally to the farm Schoonberg,
seven miles further E.N.E., where No. 17 was collected. Covering another
five miles, Ganze Kraal was reached; here there is less rhenoster bush
and more mesembryacese ; at this place sample No. 18 was taken. About
13 J miles lower down the Long Kloof, sample No. 19 was collected on the
farm Kykoe. These four samples, Nos. 16, 17, 18, and 19 represent the
soils of the Upper Long Kloof in the George Division. Together witli the
samples 4, 5, 6, 7 and 8, of the Uniondale Division soils, taken, subse-
quently in the Middle and Lower Long Kloof, they make up our series of
samples representative of the Long Kloof.

(Privately collected.)
No. Field Cornetcy. Farm or place. Collejtor.

20. Outeniqualand. Uitkyk. T. Searle.

21. „ Great Brak River. „

90

^-- D ?> »

George. Govt. Forest Plantation. C.R.Ross.

'J „ „ Distr. Forest Officer.

0 >

» •>•> »

26.

„ The Island. P. Bornemisza.

28. Kamnassie. Vogelstruisfontein. „

29. The Lakes. Lance wood. „

Noe, 20, 21, and 22 were collected from as nearly as possible the
identical spots where Nos. 1, 2, and 3 respectively had been taken on *.-
former occasion. The analyses of the latter samples had been conducted,
according to Method III., and as it seemed probable that the results so
obtained were too high, it was deemed desirable to procure fresh samples
and to analyse them by Method I. for comparison, in order to ascertain,
as far as that could be done, what differences of results could be detected
between the two methods : it may be said that about three years had
elapsed between the taking of the first and of the second set of samples.

Of the samples taken from the Forest Plantation at George, No. 2$
bore close resemblance to the other alluvial soils from that Division a»
regards texture and general physical condition, its content of moisture
and organic matter, and its almost all-round poverty in plant food. Nos.
24, 25, and 26 were taken from different parts of a site on which it was
proposed to establish a forest nursery.

58

Soils intended to be employed for the cultivation of tobacco were also
procured for analysis. One of these was obtained from Vogelstruisfon-
tein, another from the farm Lancewood or Hooge Kraal and a third from
the Ven. Archdeacon Fogg's farm The Island, situated to the north of
George town and adjoining the town commonage.

The results arrived at by chemical analysis are given in the table*
below : —

(Method I.)

No.

4.

-5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.

No.

24.
25.

26.
27.
28.
29.

Percent, of
Field
Sample.

Fine earth.

Percentage of Soil sifted through Percentage of Soil sifted
1 mm. Sieve. I mm. Sieve.

Phos-
Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric
matter. oxide.

94-7

•97 3-51 '0138

•140

•028

•034

•020

90-5

•56 2-52 -0123

•098

•020

•031

•013

98-1

•87 4-03 '0212

•161

•042

•015

•031

97-8

•47 2-24 '0216

•091

•028

•017

•019

97-8

1-62 6'57 -0283

•203

•034

•016

•022

98-2

1-41 6-02 -0127

•189

•024

•010

•015

97-0

1-12 5-94 '0481

•168

•026

•016

•031

35*0

•86 7-41 -0835

•168

•028

•102

•083

75-8

•86 5-44 -0913

•168

•052

•122

•090

90-9

1-35 6-09 -5008

•168

•096

•095

•067

73-5

•80 5-59 -0679

•189

•040

•136

•079

79-6

•60 2-17 -0626

•112

•068

•066

•028

66-8

2-78 8-06 '0580

•245

•040

•041

•095

81-9

2-84 7-37 '0569

•168

•030

•066

•068

42-6

1-22 4-35 -0948

•154

•062

•043

•038

76-7

•44 2-87 '0212

•098

•012

•068

•058

94-3

— — —

—

•050

•023

•012

93-6

— _ —

—

•094

•061

•015

91-0

— — —

—

•044

•017

•008

96-6

2-27 8-96

•02

•072

•027

•097

(Method II.)

Percentage of Soil sifted through 1 min.
Sieve.

Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
ithrough 3 mm. Sieve.
Phos-
Lime, Potash. phoric
oxide.

V43

6-73

•28

•21

trace.

T27

5-04

•318

•019

•038

1-06

6-51

•24

•24

•014

•56

14-89 '019

•11

•44

•10

•74

8-61 '008

•92

•23

•028

1-05

5-51 '034

•96

•77

•12

Percent, of

Field

Sample.

(Method III.)

Percentage of Soil sifted through 1 mm.
Sieve.

INo.

1.
2.
3.

Fine earth. Water.

92-4
89-2
85-0

1-67

2-37

•68

Organic
matter.

3-89

5'48
1-74

Chlorine. Nitrogen.

•on

•0097

•058

•026
•044
•030

Percentage of Soil sifted
through | mm Sieve.

Phos-

Licue. Potash, phoric
oxide.

•25 '14 '11
•19 '25 '029
•25 -27 -055

It will be noticed that Nos. 1, 2, and 3 show much larger proportions
of lime and potash than Nos. 20, 21, and 22. Making full allowance for the
fact that the one set of samples was collected a long time after the other,
and taking due account of their having been collected by different persons,
and possibly not from the identical spots in each case, and of their having
been analysed by different analysts, it still remains clear that the one
method yields much higher results than the other, as indeed one would
expect from the inherent differences between the respective solvents used.
From the nature of the soils- examined one could not otherwise than conclude
that by Method III. a large amount of plant food constituents was ex-
tracted from the soil which it would not be proper to consider as plant food in
the sense to which the term has been restricted in the course of the present
discussion. That such a conclusion would be just seemed clear from a com-
parison of the above results, and further proof of this will be forthcoming
when the Divisions of Mossel Bay and Riversdale have to be dealt with.

The southern part of the George Division, that is to say, the portion
lying between the Outemiqua Range and the sea, cannot, from a chemical
point of view, be considered as very promising : the samples examined,
which come from that area, are, almost without exception, poor in all the im-
portant elements of mineral plant food. The soils are very fine grained,
and there is a good quantity of nitrogenous material in the soil, but with
the lack of lime, and the acid nature of the soil, it is exceedingly doubtful
whether much of it is capable of conversion into a form in which it could
be absorbed by plants.

On a later page, in connection with the soils of the Malmesbury Divi-
sion, allusion is made to the clay slate beds which constitute the oldest sedi-
mentary rocks of the south-western part of the Colony ; above these lies the
sandstone formation with which not only the Outeniqua Mountains are
capped, but also all the mountain ranges stretching from Table Mountain
in the Cape Division to Cape St. Francis, including the Hex River Moun-
tains and the Langeberg Range. This sandstone, by disintegration, forms a
loose sand, very poor for agricultural purposes, producing a vegetation of
little more than sour grass. Firs, and similar forest trees may, neverthe-
less, find sufficient nourishment in such soils, though unsuited for ordinary
crops.

North of the Outeniqua Mountains the potash in the soil shows some
improvement : the five soils collected on the farms Klip Drift and Groot
Doom Rivier all contain a fair amount. This is also the case in regard to
phosphoric oxide, except as far the alluvial soil No. 15 is concerned. This
soil is poor in phosphates, and is also poorer than any others in the neigh-
bourhood in respect of potash. In all these soils lime is still very deficient,
although not to the same extreme as south of the Outeniquas.

The Long Kloof soils, extending from Grootfontein in the George
Division to Krakeel River in the Division of Uniondale, and represented
by samples 16, 17, 18, and 19, in the above list, together with Nos. 4, 5, ^,
7, and 8 of the Uniondale Division soils, are uniformly poor in lime, and

60

just escape a poor potash average. The soils from the western part of the
Long Kloof showed a fair amount of phosphates, Ganze Kraal exceptod,
but further east there appeared a decided deficiency.

To summarise in broad terms, the soils of the George Division are,
taken as a whole, poor in lime and phosphates, but contain a fair propor<
tion of potash, and are rich in nitrogen.

GORDONIA.

(Officially collected.)

No. Field Cornetcy. Farm or place. Collector.

1. No. 1. Upington. Dr. E. A. Nobbs,

«• » » »

The above were two very finely-grained soils collected from lands under
irrigation at Upington. The samples were very typical of the alluvial
lands along the banks of the Orange River at that point. The soil had been
under cultivation for twenty years, and had been periodically enriched by
deposits of silt from flooding. Soils of a similar nature will be referred to
in connection with the Prieska Division.

The analyses of the above soils resulted as follows : —

(Method I.)
°f Percentage of Soil sifted through I mm. Percentage of Soil sifted

Sample. * mm> Sieve>

Phos-

Ko. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide

1. 100 3-23 3-49 '0209 '052 '582 *093 '061

2. 100 1-68 2-05 '0184 '022 '400 '065 '052.

It will be noticed that the nitrogen content of these soils is low, but
they are rich in lime, and fairly well provided with potash and phosphatio
material.

GRAAFF-REINET.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector-

1. Graaff-Reinet. Graaff-Reinet Town. C. Mayer,

These three samples were taken from different depths at a locality in
the township of Graaff-Reiuet where vines were dying off from some un-
known cause. Local vignerons had been greatly perplexed in consequence.
Diligent inspection had revealed neither insects nor fungoid growth, and
in all cases decaying roots and a dead root stem indicated that dying had

commenced from below, and had worked its way upwards to the surface of
the soil. It was thought that brackishness of the irrigation water had con-
tributed to the affection, but analysis of the water did not confirm this
view : it appeared, however, that excessive irrigation was the chief cause,
and imperviousness of the soil to air owing to its compact nature and defi-
cient drainage. In any case, the affected vines were invariably associated
with the free practice of irrigation : the soils were usually irrigated once
©very month.

The chemical analyses* resulted as follows : —

(Method I.)

Percent of percentage Of Soji sifted through
£*L 1-. Sieve.

Fine earth. Water. Organic Chlorine. Nitrogen
matter.

Percentage of Soil sifted
through £ mm. Sieve.

Phos-
. Lime. Potash. phoric
oxide.

94-8

2-56

3-80

•0120 .

•154

1-148

*00

•079

95'5

2-72

4-10

•0150

•231

1-288

•318

•072

95-4

2-93

3-32

•0067

•175

1-364

•318

•073

No.

1.

2.
3.

The first of these samples was taken at the surface, No. 2 at a depth
of 10 inches, and No*. 3 at from 18 to 24 inches from the surface. The soil
is of far finer texture than those generally associated with the cultivation
of the vine in the Western Province : the three samples are all rich in lime,
well supplied with nitrogen and potash, and have a fair reserve of phos-
phoric oxide: apparently, therefore, it was not deficient storage of plant
food that was the cause of vine-failure. The amount of chlorine in the soil
was no higher than has been found to be the case in several soils of the
Paarl and Worcester Divisions. Hence it seemed to be a just inference
^that the causes were physical rather than chemical. This was evidently an
instance where the weak link of the chain was not lack of plant food.f

;No.
1.
2.
3.
4.
5.
6.

Field Cornetcy.
Hanover.

HANOVER.

(Privately collected.)

Farm or place.
Hanover.

Collector.
Forest Officer.

These soils were taken from the Government Forest Plantation at
Hanover. The plantation forms part of Hanover Commonage, and is about
ten miles distant from Hanover Road Railway Station. Nos. 1, 2, and 3
represent surface soils, and Nos. 5 and 6 clay subsoils. No. 4 was % speci-
men of the limestone which underlies the soil at this locality.

*For mechanical analysis see under "Physical composition of soils" (Part VII.).
t Vide remarks in this connection on pp. 17 and 18.

The results of the analyses of the above samples are stated in the fol-
lowing table : —

(Method II.)

No.

1.

2.
3.
4.
5.
6.

Percentage of Soil sifted through 1 mm.
Sieve.

Nitrogen.

Water.

Organic

Chlorine.

matter.t

4-25

23-29

3-94

25-31

—

1-94

3-21

•0049

5-09

29-27

—

5-83

27-03

•0191

7-35

24-22

•0104

Percentage of Soil sifted
through J mm. Sieve.

Phos-
Potasch

Lime,

22-75
15-41
1-62
32-46
26-60
19-00

•26
•13
•21
•12
•15
•21

phoric
oxide.

•010

•089

•0036

•124

•096

•052

All of these samples contained an abundance of lime, largely in the
form of calcium carbonate : they are all well supplied with potash, but tihe
proportion of phosphoric oxide varies considerably, ranging from the ex-
treme of poverty, in one case, to a fairly satisfactory amount in1 others.

No.
1.

2.
3.

Field Cornetcy.
Lower Albania.

HERBERT.

(Privately Collected.)

Farm or place.
Backhouse.
Erf No. 5.
Erf No. 36.

Collector.
A C. Martin.

The three samples enumerated in the above list were taken from the
area proposed to be brought under irrigation in connection with the
Douglas irrigation works. The first sample was taken from the farm Back-
house, near the main canal, at chainage 4 miles 30 chains. No. 2 was col-
lected on Douglas Commonage from the middle of Agricultural Erf No. 5,
and is fairly representative of all the erven lying along the river bank.
No. 3 was taken from Agricultural Erf No. 36, also on the commonage,
near the main canal, at chainage 6 miles 10 chains, and may be considered
as a type of the erven along the route of the canal.

In every case the samples represent virgin soils, and were carefully col-
lected so as to typify the surface soil to a depth of twelve inches. They were
each analysed according to Method I. and also by Method V., and the re-
sults of these analyses are tabulated below. *

(Method I.)

Percent, of
Field
Sample.

Percentage of Soil sifted through
1 mm. Sieve.

Percentage of Soil sifted
through i mm. Sieve.

Phos-

No.

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

1.

96-1

1-79

3-24

•0566

•049

•174

•061

•058

2.

99-0

1-80

3-76

•0707

•042

•555

•103

•105

3.

87-9

3-00

6-01

•0283

•098

2-128

•053

•070

also includes carbon dioxide combined as calcium carbonate.

*For the results of determinations of alkaline salts in the above soils see under
the bead of "Alkalinity of soils" (Part VI.).

S3

(Method V.)

Percentage of Soil sifted through 3 mm. Sieve.
No. Potash. Phosphoric oxide.

1. -010 -016

2. -018 -037

3. -009 -008

It will be seen that in each case about six times as much potash was
extracted from these soils by Method I. as by Method V., and in two cases
the quantity of phosphoric oxide extracted by Method I. was about 3 to 3£
times that obtained by Method V. In the third soil the directly available
phosphoric oxide was only one-ninth the available reserve.

On the basis of the minimum limits suggested by Dr. Dyer, viz., '005
per cent, of potash and '010 per cent, of phosphoric oxide, the first two
soils, at all events, are sufficiently well supplied with these elements of
plant food in a state ready for immediate use.

HOPE TOWN.

(Officially collected.)

Field Cornetcy. Farm or place. Collector.

1. North Middenveld. Fluitjes Kraal. Dr. E. A. Nobbs.

9

^' )) » »

Two fine-grained soils were collected from the farm Fluitjes Kraal:
No. 1 had the characters of a good Karroo soil, and was typical of a con-
siderable area. No. 2 was taken from what is known as vlei land.

The analyses of these two samples resulted as below: —

(Method I.)

PeF?eld °f Percenta?e of Soil ?ifted through Percentage of Soil sifted
Sample * mm* ^eve< through £ mm. Sieve.

Phos-

No. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide.

1. 96-2 3-96 3-02 '0035 '050 '478 "085 '019

95'7 7'20 4-24 '0014 '057 '332 '275 '033

The proportion of phosphoric oxide in these soils is low, nor are they
particularly well supplied with nitrogen,; on the other hand, they are both;
rich in lime, in this respect resembling many of the soils of the neighbour-
ing Divisions of Albert, Aliwal North, and Colesberg.

HUMANSDORP.

(Officially collected.)

No. Field Cornetcy. Farm or place. Collector.

1. A. C. Macdouald.

2-

Zitaikama. Lot 7,899 H. G. Fourcade.

4. Lot 7,897.

Two soils, Nos. 1 and 2, were taken in this Division from localities
badly affected with " lamziekte," and where great mortality had resulted
amongst the sheep: the soils, as will be seen from the table below, were
found to be almost devoid of phosphoric oxide, although one was well
stocked with lime and pot-ash, and the other fairly so. Taken in conjunc-
tion with the analyses of the soils from the Albany Division,* it would
appear that lack of phosphates rather than lack of lime is one of the fun-
damental causes of larnziekte. Almost throughout this Division the soil
is entirely the result of the disintegration of the great ranges of Table
Mountain sandstone, which extend along the south coast from George,
and eventually die out in the Zitzikama Range. This sandstone contains
practically nothing capable of affording nutriment to plants.

Samples 3 and 4 represent the average virgin soil of Storms River to
a depth of twelve inches, and were taken from upland plateaux. The soil
consists chiefly of fine quartz silt. The opinion of scientific agriculturists
with regard to acid soils of this type is that plant food constituents are
rendered available to such an extent by liming or cultivation that ex-
cellent crops are raised with moderate application of fertilisers.

The following are the analytical results obtained : —

1.
^2.
3.
4.

Percentage

Water.

4-12
3-74
T36
2'43

of Soil sifted
Sieve.

(Method II.)
through 1 mm.

Organic
matter.

7'46
4-48
4-07
6-72

Percentage of Soil sifted through
3 mm. Sieve.

Phos-

Chlorine. Nitrogen. Liine.

•024
•005
•053
•060

•140
•168

•09
•54
•101

•089

Potash.

•07
•13
•018
•013

phoric
oxide.

trace.

trace.
•005
•OOS

IDUTYWA.

No.
1.

Field Cornetcy.
Ibeka.

(Officially collected.)

Farm or place.
Ibeka.

Collector.
St. C. O. Sinclair.

One sample of soil was taken from, and may be considered as repre-
sentative of, the sandy tract of country near Ibeka. It is, in all proba-
ISility, similar in origin to the two soils, Nos. 1 and 2 on the Butterworth
Jist,t taken in the vicinity of the old Residency at 'Nhlambe, :~ Al*~k
district.

that

The results of the analysis of the soil from Ibeka are as follow^:

(Method I.)

Percent, of Percentage of Soil sifted through

Field 1 mm. Sieve.'

Sample.
Fine earth. Water. Organic Chlorine. Nitrogen.

matter.
97-6 2-44 5-18 '0474 '133

Percentage of Soil sifted
through i mui. sieve.

Pho»

Lime. Potash. phoric
oxide.

•074 '027 -006

* Vide page 23. t See page 32.

Nc.

1.

2.

Field Cornetcy.
Kenhardt.

65
KENHARDT.

(Officially collected.)

Farm or place.
Rooiberg.

Collector.
Dr. E. A. Nobba.

These two sa-mples were collected from irrigable lands between the
dam and the village of Kenhardt: the land consists, for the most part, of
a fine red sand, of which No. 1 is an example, varying in depth, and pasa-
ing into a red silt of even finer grain, No. 2.

No.
3.

4.

R.

(Privately collected.)

Field Cornetcy. Farm or plac«. Collector.

Kenhardt. Rooiberg. Engineer, Pub. Wks. Dept.

>» j» »

A. G. Strong.

Nos. 3 and 4 were collected respectively from the front third and
from the centre of the Rooiberg Dam. No. 5 had previously been taken
from the site of the dam, as representative of the lands proposed to be
irrigated.

The analyses of these samples are recorded in the following tables : —

(Method I.)

No.

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 mm. Siere. through £ mm. Sieve.

Sample. Phos-

Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. . oxide.

1.

55-9

1-90

2-88

•0039

•071

•454

•296

•051

2.

100.

2-24

2-85

•0025

•050

•320

•207

•069

3.

94-7

4-31

4-32

•1387

•057

•290

•348

•134

4.

95'0

4-60

4-54

•1146

•043

•420

•341

•145

No.
5.

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Water.
5-80

Organic
matter.

5-22

Chlorine. Nitrogen.
•0038

Percentage of Soil sifted
through 3 mm. Sieve.

Phos-

Lime. Potash. phoric
oxide,

•999

154

156

No. Field Cornetcy.

1. No. 2.

o

3.

4.

5.

i

6.

KIMBERLEY.

(Officially collected.)

Farm or place.
Schoolplaats.

Waterfall.
Warrenton.

Collector.
Dr. E. A. Nobbs.

E

As in the case of the Barkly West soils, the majority of the sample*
from the Kimberley Division were collected by Dr. E. A. Nobbs from a
large extent of country proposed to be irrigated by means of water con-
veyed from the Harts River. Three samples were taken on the farm
Schoolplaats. No. 1, a tough black alluvial clay, lying on a subsoil of
•ilt, represents a heavy piece of land situated close to the Vaal, obviously
very rich, but unfortunately limited in extent to a narrow strip along
the banks of the river. No. 2 represents the eastern side of a wide low-
lying bight of land on the same farm : it is a yellow fine-grained sandy
soil. The central and western portions of this bight consist of a stiff
loamy marl, grey in colour, typified by No. 3. On the farm Waterfall
occurs a stretch of some 800 acres of a red gravelly loam, about 145 feet
above river-level. The owner of the farm proposes to irrigate this land
—whereof No. 4 is a sample — by pumping up water from the river below.
Two samples of fine-grained brown alluvial soil were collected at Warren-
ton, No. 5 a garden, and No. 6 a virgin soil : both of these are from time
to time enriched by flooding and deposits of silt.

No.

7.

8.

Field Cornetcy.
No. 1.

(Privately collected.)

Farm or place.
Kimberley Borough.

Collector.
Supt. Sanitary Dept.

Nos. 7 and 8 represent land within the area controlled by the
Borough Council of Kimberley.

The following tables comprise the results of the chemical a-aalyses of
the soils collected within the Kimberley Division : —

(Method I.)

No.

1.
2.
3.
4.
5.
6.

Percent of

Percentage of Soil Bifted through

Percentage of Soil sifted

Field

1 mm.

Sieve.

through

i rim.

Sieve.

Sample.

Phos-

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

98-2

4'77

5-32

•006

•134

•168

•161

•119

95-0

•59

1-17

•002

•035

•018

•039

•031

94-5

4-83

9-01

•Q03

•171

2-860

•125

•068

79-4

2-23

4-07

•002

•013

•142

•161

•064

91-5

2-50

3-34

•003

•120

•278

•108

•074

88-2

6-29

5-07

•006

•106

•346

•094

•070

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of Soil sifted

through 3

No.

7.
8.

Water.

Organic
matter.

Chlorine Nitrogen. Lime. Potask.

•032
•036

•98
3'24

•18
•63

Sieve.
Phos-
phoric
oxide.

•07

No. 1. proves to be well supplied with every class of plant food, the
analytical results thus confirming the opinion previously formed of the soil.
The sandy soil No. 2 turns out to be far below No. 1 in chemical composi-
tion. The area whereof No. 3 is a type contains lime — largely carbonate —
in abundance, it is well provided with nitrogen, and has a moderate sup-
ply of potash and phosphates. The Waterfall gravelly loam, represented
by No. 4, lacks nitrogenous constituents, but has a satisfactory amount of
potash, and a fair provision of lime and phosphoric oxide. The Warrenton
alluvial soils may be described as chemically of medium quality.

67
KING WILLIAM'S TOWN.

(Privately collected.)

JXo. Field Cornetcy. Farm or place. Collector.

1. No. 8. Evelyn valley. J. S. Anderson.

The only sample collected for analysis within this Division was a black
loam, taken from some lands at Evelyn Valley, a forest station in the
Eastern Forests Conservancy. The soil was stated to be so poor that
nothing except trees will grow on it. The forest station is situated near
the top of the Pirie Mountain, 4,200 feet above sea-level, and 40 miles
from the coast almost due north of King William's Town. The surround-
ing farms all lie at much lower altitudes, ranging from 1,700 to 2,300 feet
below Evelyn. Valley. The soil at these lower levels, where the annual
rainfall ranges from 18 to 32 inches, has the reputation of being much more
fertile than that at the forest station, which has a mean annual rainfall
of 60 inches. The ground where the sample was taken had been broken
up from the veld two or three years previously ; the site of the sample was
a gentle slope about 200 yards from the foot of a steep kopje rising to a
height of 250 or 300 feet : the surface soil goes to a depth of from 12 to
14 inches, below which lies a red porous clay. Nearer the foot of the
kopje the surface soil increases to two feet or more in depth.

The analysis of this soil resulted as follows : —

Method I.

Percent, of Percentage of Soil sifted through 1 mm. Percentage of Soil sifted
Field Sieve. through £ inm. Sieve.

Phos-

Xo. Fine earth. Water. Organic Chlorine Nitrogen. Lime. Potash, phoric

matter. oxide.

I. 97-5 6'26 16-60 '0205 '315 '050 '082 '105

The soil was fine in grain and contained a large quantity of nitro-
genous organic matter; the proportion of phosphoric oxide was satisfac^
tory, but lime was deficient, and the reserve of potash was not much better.

KNYSNA.

(Officially collected.)

Field Cornetcy. Farm or place. Collector.

1. Zwart River. Geelhoutboom. C. F. Juritz.

Roode Kraal.
3. Gouwkamma. Millwood.

Portland Heights.
„ Balmoral.

6. „ Portland.

7. Plettenberg Bay. Ashby.

S. „ Ganze Vallei.

9. „ Holt Hill.

10. „ Witte Drift.

I 1 . Pa^rde Kop. Witte Drift.
12. Keurbooms River. Matjesfontein.
13.

On. the road from George to Knysna the first sample collected in the
latter Division, No, 1, wa3 taken on the farm Geelhoutboom, or Olyven-
hout Kraal, about six miles E.N.E. of Diep River, in the George Division ;
No. 2 being taken from Roode Kraal, some 5J miles further eastward*.
At Sour Flats, called also Balmoral, No. 5 was collected, from a spot N.E.
of the locality where the last mentioned sample was taken, and distant
therefrom about ten miles by road. The soil is a light clay, fairly loose
in texture; analysis shows it to be poor in all mineral plant food. As a
matter of fact, these three soils are all derived from rocks of the Table
Mountain series, and the exceedingly poor analytical results are only what
one can expect under the circumstances : however, on a small farm, a few
hundred yards nearer Knysna, good crops are stated to have been culti-
vated for ten years without manure. At Portland, between Balmoral and
Knysna, lying about four miles south of the former place, and 19| N.W.
of the latter, a similar light clay, sample No. 6, was collected. Here again
the analysis yielded poor results, for Portland too lies on Table Mountain
sandstooie. No samples were taken between this point and Plettenberg
Bay. At the latter place a rather loose clay predominates, but the under-
lying rock is Table Mountain sandstone. There is a quantity of rhenoster
bush in the vicinity of the local accommodation house, and here a sample

( %, r-
J r-*^< &^.

DIVISIONAL WAP OT
KNYSNA

INDIAN OCCAM

of soil, No. 7, very light in colour, was collected from a hill top. About a
mile and a half lower down in the valley a sample of " vlei ;; ground, No. 8,
was collected on the farm Gauze Vallei. Here and there large patches of
conglomerate show through the soil and along the hillsides. Another
sample of vlei ground, No. 12, looser and much darker than the last, and
an all-round better soil, was taken on the farm Matjesfontein, just after
crossing the Keurbooms River, four miles from Ganze Vallei. The organic
matter — which, moreover, contains a good deal of nitrogen — in this sam-
ple is higher than in any previously collected in this Division. The soil
is well supplied with lime, and has a satisfactory store of potash and phos-
phoric oxide. Further east on this farm Matjesfontein, the soil becomes
lighter in colour and sandier, and the vlei ground is to a large extent in-
termingled with blown sand from the sand dunes along the coast. Of one
of these mixed soils a sample, No. 13, was taken, a mile intervening be-
tween this and the previous sample : another sample — to outward seeming
of the same class — No. 11, was collected on the farm Witte Drift, about five
miles west of No. 13. On all these vlei grounds general cultivation is car-
ried on, with fair success, over a large area. At Witte Drift the rhenoster
bush is again met with, growing on a rather stiff brown clay soil : of this

a sample. No. 10, was obtained. All these latter soils yielded noticeably
lower results by analysis than No. 12. After traversing a distance of four
miles and rejoining the main road, a sample, No. 9, was taken from a
plateau covered with reeds and extending westwards for miles, up to the
borders of the forest. The soil here is almost all-through deficient.

Below Millwood a sample of black loamy soil was collected, No. 3 in the
list. This class of soil extends southwards right down to the forest, where
it gives place to more clayey soil of a lighter nature. Of this latter, a
sample, No. 4, was taken in the forest on the way from Millwood to Bal-
moral, at a distance of 2| miles from the point where No. 3 was taken.
Neither of these two soils is chemically well provided.

No.
14.
15.
16.
17.
18.
19.

Field Cornetcy.
Z wart River.
?>

5J

Gouwkamma.
Knysna.
Keurbooms River.

(Privately collected.)

Farm or place.
Sidgefield.
Lot B (2).
Salisbury.
Belvidere.
Bracken Hill.
Matjesfontein.

Collector.

J. Grant,
A. Bibbey.
P. Bornemisza.
A. V. Duthie.
P. Bornemisza.

Sample No. 14 was collected on the farm Sidgefield, at Groen Vallei.
Nos. 16, 18, and 19 were taken on the farms Salisbury, Bracken Hill (Lob
R.R.), and Matjesfontein, respectively, from lands intended to be utilised
for tobacco cultivation. The farm Salisbury adjoins Olyvenhout, or Holly-
wood, on the boundary of the George Division. North-west of Hollywood,
and nearer the George boundary, a sample of soil, No. 15, was collected
from Lot B (2).

The analytical results obtained from the soils of this Division are com-
prised in the following tables : —

(Method I.)

No.

Percent, of Percentage of Soil sifted through 1 mm. Percentage of Soil sifted
Field Sieve. through £ mm. Sieve.
Sample. Phos-
Fine earth. Water. Organic Chlorine. Xitrogen. Lime. Potash, phoric
matter. oxide.

1.

96-0

1-17

6'55

•0237

•203

•032

•012

•020

2.

98-5

•86

4-19

•1068

•161

•044

•on

•013

3.

96-6

1-59

8-86

•0672

•287

•074

•017

•031

4.

95-6

3-84

12-21

•0813

•413

•032

•028

•056

5.

94-7

•97

5-01

•0163

•154

•050

•014

•018

6.

98-2

1-27

5-81

•0347

•196

•028

•014

•046

7.

97-0

•66

2-80

•0262

•105

•028

•023

•010

8.

99-5

2-19

6-31

•0707

•231

•116

•054

•045

9.

98-0

1-20

5-39

•0651

•126

•082

•030

•018

10.

90-8

1-60

6-42

•0198

•175

•112

•069

•044

11.

82-9

•98

4-19

•0113

•154

•052

•040

•044

70

32.
13.
14.
15.
17.

Percent, of

Field

Sample.

Fine earth.

97-1

84-4
96'4
95-8
96-8

Percentage of Soil sifted through 1 mm.
Sieve.

Water.

3-05
1-33
•56
3-28
1-25

Organic
matter.

11-92

5-16

1-71

11-25

4'41

Chlorine. Nitrogen. Lime.

•0453
•0130
•0297
•0117
•214

•399
•161
•067

•462
•140

Percentage of Soil sifted
through i mm. Sieve.

Phos-
Potash.

•544
•112
•090
•040
•098

•105
•039
•050
•029
•036

phoric
oxide.

•110

•078
•051
•012
•0051

(Method II.)

No.

16.
18.
19.

Percentag-e of Scil sifted through 1 mm.
Sieve.

Water.

1-54

I'OO

•76

Organic
matter.

4-79

4-10

14-14

Chlorine. Nitrogen. Lime.

•023
•001
•010

Percentage of Soil sifted
through 3 mm. Sieve.

Phos-
Potash.

•65
•36
•75

•13

•09

trace.

phoric
oxide.

•13

trace.

•06

Of the soils, Nos. 7 to 13, collected around the mouth of the Kern-booms
River, it will be noticed that there is a considerable difference, in the
amount of available plant food, between Nos. 8, 10, 12, and 13, on the
one hand, and Nos. 7, 9, and 11, en the other. The former axe all either
wholly or very largely made up of the alluvial deposits which surround the
river mouth ; of the latter, Nos. 7 and 9 were taken from the all-prevailing
sandstone formation, and No. 11 from the Enon deposits which flank the
alluvium, and with which No. 13 is also probably to a certain extent diluted.
The following are the averages of each of these two sets of soils : —

Phos-

Nitrogen. Lime. Potash. phoric

oxide

Nos. 8, 10, 12, and 13.
Nos. 7, 9, and 11

•242
128

•221
•054

•067
•031

•024

The poverty in available plant food constituents, noticed with respect
to the southern portion of the George Division, extends, with some reserva-
tions, however, into the adjoining Division of Knysna. Here we find on an
average even less potash and phosphoric oxide, but the lime shows a slight
increase. There is a sufficiency of nitrogen all through, but the inorganic
requirements of the soil are lacking in the Zwart River and Gouwkamma
Field-Cornetcies. Potash, in fact, may be said to be lacking throughout
the Division, with the exception of a limited area round the mouth of the
Keurbooms River. Near Plettenberg Bay there is an improvement visible
on the farms Ganze Vallei and Witte Drift. Of the two samples collected
on the latter farm it will be observed that No. 10 is noticeably the better.
The richness of the forest soils near Millwood in nitrogenous material, and
organic matter generally, is worth noting, and special attention has al-
ready been directed to the quality of the alluvial soil No. 12 from the
farm Matjesfontein near the Keurbooms River mouth. There is no doubb
that these alluvial, or, as they are locally termed, " vlei " soils, are better
supplied with plant nutriment than the average soils of the Division.

71
KOMGHA.

(Officially^ collected.)

No. Field Cornetcr. Farm or place. Collector.

1. No. 3. Prospect. St. C. O. Sinclair.

t-> r> » H

3. ,, Ferndale. „

4. ,, ,, »

5. „ Reedfontein. „
tt » »

< • >J V »

S. „ Hopewell. „

9. ,, Stainland. „

10.

11- » » >» .

In
j» » >»

13. „ Zuurvlakte. „

14. ,, „ ,.

15. No. 1. A. D. Campbell. „

16. „ Residency. „

17. No. 2. Ben Hall.
IS. No. 6. Lilyvale.
19.

20. „ Westbury.

21. „ E. Sanson. „

22. „ Fort Warden. ,.

23. „ Beaconsfield. ,.

24. Badnoch. „
25.

26. „ Jessie. „

27. No. 5. Sea View. „

In taking samples in tho Komgha Division, it became necessary, for
the reasons mentioned in connection with the Cathcart Division, to place
one's self largely in the hands of the local agriculturists in respect to tha
selection of localities from which to take samples. In a few cases samples
were taken for special reasons : for instance, where experiments in fruit
growing were being carried on, and where it was the intention of the
occupier to use his land for specific crops, and it was consequently de-
sirable to have the most reliable information as to the exact physical and
chemical nature of the soil in use.

The soil of the Komgha Division is essentially derived from the pecu-
liar " Lower Karroo " sandstone (Ecca beds, or, as the Geological Com-
mission has now termed them, in that part of the Colony,* Umsikaba beds).
Dolerite, occurring either in sheets, dykes, or loose boulders, is exceedingly
plentiful, and one cannot go anywhere in the district without finding it in
evidence in one or other of these forms.

The surface soil, on the whole, appears distinctly loamy, consisting of
clay with a very fair proportion of fine sand. In some cases, however,
the soil seemed more sandy. The subsoil is generally either a red gravel,
or a rather stiff yellow pot clay, although, of course, variations between
the two are not rare.

* Geological Commission Annual Report 1902, pp. 9 and 14.

72

In judging of the latent fertility of the soil of any portion, the pre-
vailing physical conditions, such as the nature and divisibility of the sur-
face a<nd subsoil, the situation and slope of the land, the exposure to pre-
vailing winds, thunderstorms, and rains, always important factors, should
be even more largely taken into account in the case of this division than
is usually done. From a chemical point of view, external evidence leads
one to expect a soil rich, on the whole, in the essential mineral plant
foods, lime, potash, and phosphoric oxide, the last named, perhaps, not
quite so plentiful in proportion as the other two.

In some cases the clayey nature of the soil makes draining desirable.
Where this has been done, "excellent results have been reported. Very
little manuring is earned on, due to the scarcity of manure. A start has,
however, been made with artificial fertilisers. Fruit trees, such as pears
and plums, especially Japanese plum and orange, seem to thrive, and near
the coast bananas do well. Wheat does not grow satisfactorily, although
at one time its cultivation was profitable. On the other hand, the lands

/"":."•

J 'OlVloHAL MAP or

KOMCHA

stand maize well. A very possible explanation of this may be the fact
that wheat is sown on lands which have borne crops of this cereal for a
considerable time in the past, and consequently have undergone great loss
of phosphatic material.* But maize, taking out of the soil a larger
amount of lime in proportion than wheat, is able to flourish where wheat
does but poorly.

Sample No. 1 was collected on the farm Prospect ; it represents a
dark hillside soil, with a red gravelly subsoil. No. 2 was also collected
on this farm. Fruit growing is very successfully carried on here. Nos.
3 and 4 are types of black soils "from the adjoining farm Ferndale.
Around the locality whence the samples were taken, doleritic boulders
were noticed. Nos. 5, 6, and 7 were collected on the farm Reedfontein.
Of these, No. 5, a soil of good depth, was stated to yield poor results when
sown with oats. No. 6, on the other hand, which contains nearly double
the proportion of phosphoric oxide, and three times the quantity of
potash, gave good results with the same crop — another instance of prac-
tical experience bearing out the results of chemical analysis. No. 7, lying
on a subsoil of yellow clay, was reported to be very poor. The subsoil,
•which comes to within a short distance of the surface, was noticed to be

See remarks on this subject on pp. 45 and 47.

•very damp at the time of sampling. This is, no doubt, one factor which
may account for the poor productiveness of the surface soil. No. 8 repre-
sents the red soil of the farm Hopewell. The land represented was in-
tended to be used as an orchard. Nos. 24 and 25 were taken at Badnoch,
and No. 27 is a sample of black and somewhat sandy soil found on the
faa*m Sea View. No. 26 was taken a short distance from the roadside at
Jessie. No. 20 is from Westbury, taken from the top of one of the hil-
locks near the homestead. No. 21 was collected OIL the farm of Mr. E.
Sanson.

Samples 9, 10, 11, and 12 were taken at Stainland, the farm of Mr.
Coulter. No. 9 represents a valley soil from a patch which was at one
time a, SAvamp, but had, by draining, been turned into an area where oats
do well. It is very similar to No. 10, which differs from it in being more
sandy. No. 11 is a red soil taken in the vicinity of some doleritic bould-
ers. No. 12 is a valley soil which should prove valuable.

Nos. 13 and 14 were collected from Mr. Van Rensburg's farm Zuur-
ylakte. No. 15 was taken from a plot owned by Mr. A. D. Campbell,
adjoining the commonage, and would be typical of the soil of a large por-
tion of the latter. No. 18, from the farm Lily vale, was taken from a
patch studded with mimosa trees. No. 19 is from the same farm, and was
taken underneath a mimosa tree. No. 17 was collected on the farm of
Mr. Ben Hall. It represents the soil of the hill side sloping towards the
Kei River. The surrounding soil was very shallow.

The results obtained by analysis are tabulated below : —

(Method I.)

Percent, of
Field

Percentage of Soil sifted through 1 nirn.
Sieve.

Percentage of Soil sifted
through ^ mm. Sieve.

Sample.

Phos-

No.

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Linie.

Potash.

phoric

matter.

oxide.

1.

97'5

3-95

12-53

•0230

•252

•222

•042

•056

2.

92-9

2-47

6-80

•0329

•196

•134

•016

•014

3.

98-4

5-05

14-79

•0463

•252

•310

•124

•028

4.

95-8

4-47

12-01

•0275

•280

•490

•026

•036

5.

94-4

3*24

8-62

•0240

•168

•122

•025

•019

6.

97-2

1-84

5-84

•0110

•154

•106

•076

•032

7.

98-9

1-87

5-72

•0190

•154

•068

•116

•020

8.

98-1

2-84

9-45

•0162

•189

•122

•146

•020

9.

98-8

2-94

6-56

•0311

•182

•322

•017

•0089

10.

98-5

1-44

3-43

•0113

•126

•092

•014

•0064

11.

97-7

3-83

11-08

•0445

•217

•042

•045

•051

12.

96-8

2-17

7-50

•0339

•224

•098

•235

•032

13.

98-2

T60

4-78

•0204

•147

•070

•060

•032

14.

94-6

1-83

6-09

•0544

•175

•078

•154

•031

15.

87-6

1-33

4-36

•0148

•147

•100

•042

•036

16.

92-9

2-36

6-89

•0050

•175

•064

•050

•037

17.

99-3

1-05

3-63

•0106

•133

•044

•063

•023

18.

98-1

1-24

4-41

•0035

•147

•068

•035

•030

19.

95-2

2-13

7-43

•0099

•273

•162

•046

•050

20.

97'1

T25

4-17

•0311

•091

•032

•030

•015

21.

99-2

1-27

4-11

•0057

•133

•050

•033

•019

22.

98'8

•69

2-34

•0042

•098

"012

•023

•012

23.

98-4

4-19

13-34

•0057

•273

•062

•095

•067

24.

98-9

•97

3-62

•0183

•112

•056

•030

•0064

25.

97-4

4-50

14-63

•0057

•231

•120

•048

•013

26.

99-8

6-30

15-18

•0120

•245

•294

•019

•029

27.

98-5

1-49

6-10

•0290

•147

•072

•044

•0064

74

Th« proportions of lime and potash in the soils of the Komgha Divi-
sion are, taken all round, fairly satisfactory, but phosphates are decidedly
lacking : in this respect there is a strong resemblance to the adjoining divi-
sion of Cathcart.

Of the soils in the above list, the three taken; from Field-cornetciea
Nos. 1 and 2 are all deiicient in lime and phosphoric oxide: sample*
15 and 16 are likewise deficient in potash, while No. 17 is not much better
in this respect. The samples from Field-cornetcies Nos. 5 and 6 average
poor results all round. Speaking generally, Field-cornetcy No. 3 is the
best off with regard to lime and potash, but here too phosphatic material
is lacking. To come to details; Nos. 10, 24, and 27 are poor in all of the
three inorganic fertilising constituents, and phosphoric oxide, in parti-
cular, is almost entirely absent. Nos. 13, 18, 20, 21 and 22, although con-
taining a little more phosphoric oxide, are nevertheless poor all round.
No. 9 is also characterised by an almost entire absence of phosphoric
oxide.

To summarise, eight, out of the 27 Komgha soils, are deficient in all
of the three principal classes of inorganic plant food ; seven others lack
both phosphoric oxide and potash, namely : Nos. 2, 4, 5, 9, 15, 25, and 26 :
Nos. 7, 12, 14, 16, and 17 are poor in phosphoric oxide and lime. Three
samples, i.e.t Nos. 3, 6, and 8, are poor in phosphoric oxide only, No. 11
is greatly in need of both lime and potash, Nos. 1 and 19 require potash,
and No. 23 lime. Thus it happens that there is not a single soil, of all
those taken, regarding which it can be said that it is well — or even fairly
— supplied with all plant food constituents in an available condition.

It will be noticed that these soils are all of a clayey nature, and they
appear to contain, generally, a larger amount of chlorides than has been,
found by analysis in the Western Province soils.

LADISMITH.

(Officially collected,)

No. Field Cornetcy. Farm or place. Collector.

1. Anysberg. Papenkuilsfontein. J. Muller.

o

>» j> >»

3. Groot River. Buffelsfontein. ,,

4

*• >J V })

5. „ Zeekoegatsdrift. „

6. „ Buffels Kloof.

7. „ Buffels Drift.

8. Ladiamith. Elands Vallei. „

9. Huia River. Hoeko. ,r

10. „ Weltevreden. „

11. „ Zoar.

12. ,, Amalienstein. ,,

13. ,? „ „

**• »' » »>

15. „ Opzoek.

16. Buff ebf on tern. Buffelsfontein.

75

The nrst sample taken iii this Division was No. 1, a rather poor, sandy
soil, from lands 011 the slope of the hill south of De Erf, or Papenkuilsfon-
tein. On the same farm No. 2 was taken, a dark alluvial soil, of greater
productiveness than the previous one, and collected from the garden near
the house. Both samples proved rather poor in phosphates, and No. 1 in
lime as well : they also show signs of being brack, especially No. 2. A very
barren belt of country stretches between the Prins and Buffels Bivers, and
resembles the Gouph in every respect : flanked on the north by the Klein
Zwartberg Range, and on the south by the Touwsberg Mountains, both of
which ranges consist of Table Mountain sandstone, its barrenness is not
to be wondered at. At Buffelsfontein the bed of the river widens out con-
siderably, on entering a broad valley which is literally dotted with farms.
Two samples were selected here: — No. 3, a sandy soil taken from a vine-
yard on the farm of Mr. J. J. van Zyl, and said to be very fertile, and
richer than No. 4. The latter was a loose clay, taken from Mr. J. Wolf-
aardt's farm, and supposed to be as rich as the soil collected along the
Touws River* ; analysis, however, shows that there is not a superabun-
dance of mineral plant food.

0.V,5,ONAL MA* OF

LADISMITH

I- E

Passing through the farm Voorbaat, the river course was followed ta
the farm Zeekoegats Drift, and here sample No. 5 was selected from soil
adjoining the vineyard. Good results have been obtained from sweet pota-
toes and other root corps on the particular soil represented by this sample :
it is adequately furnished with potash and lime, a fact which will account
for this. At BufFels Drift a loose, sandy clay, No. 7, was sampled, repre-
senting the bulk of the lands in the valley on either side of the Zwartberga
River. Then following the main road to Ladismith, these samples were
collected : No. 6, taken from lands on the right bank of the Buffela Kloof
River, on the farm Buffels Kloof. The soil here very closely resembles a»
red Karroo soil : although not equal to such a Karroo soil in fertility, it
is said to yield much better results than No. 8. There are several large
patches of this soil under cultivation. No*. 8 represents a stiff clay, rather
shallow, from the farm Elands Vallei. It is considered to be poorer than
No. 6, and this supposition is borne out by chemical analysis. The amount
of lime is extremely low.

* See remarks later on, page 132, regarding the collection of sample from
Zevenfontein, No. 1 from the Swellendam Division.

In the Field-Conietcy of Huis River, No. 9, a rather loose, gravelly
clay was taken from uncultivated parts in lucerne camps on the farm
Hoeko, belonging to Mr. W. van der Merwe. Mr. Van der Merwe's cattle
graze on the heights above the Touws River; manure, consequently, is
rather difficult to procure on the farm, and costs too much to transport.
The soil, however, is fairly rich in plant food. No. 10, a. fine, sandy,
alluvial soil, was taken, on the farm Weltevreden, from the gardens in the
valley, and :i locally considered poorer than No. 9, which was collected
from camps along the hill slopes : the analysis certainly does bear out that
opinion. No. 11, a red sandy soil, was taken from the garden plots of the
Rev. Mr. Kretzen, at Zoar, the Mission Station.

At Amalienstein three samples were collected: No. 12, a rather sandy
soi], was taken from lands near the main road, on the way from Zoar to
Amalienstein. No. 13, a fairly stiff red clayey soil, was taken from the
garden of the Rev. C. Prozesky. No. 14 represented a loose, somewhat
sandy clay, from lands to the left of the road from Amalienstein to Calitz-
<lorp. The crops, which were just being harvested, were exceptionally
poor, the result principally of very servere drought and scarcity of kraal
manure. At the farm Opzoek No. 15 was sampled — a rather stiff, clay
soil, said to be very good from the farmer's standpoint. It contains a good
quantity of lime and potash, and a fair supply of phosphates.

Buffelsfontein was the next place visited — more of a village than a
farm — situated on the left bank of the Groot River, south-east of Ladi-
•smith. The soil along the river-bed is similar to that at Buffels Vallei,
higher up on the Buffels River, and consequently no sample was taken
here. On the heights? above are several gardens, in a loose, sandy clay, of
which No. 16 is a typical sample, collected from the garden of Mr. J. T.
van Wyk. This soil is not considered to be as rich as that in the valley
Ijelow.

The following table shows the results of the chemical analyses of the
soils above described : —

(Method I.)

No.

1.

2

3.

4.

5.

•6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.

Percent, of

Field

Sample.

Fine earth.

60-5
90-8
65-0
71*4
97-5
56-3
79-6
85-3
45-5
81'0
67'1
56'7
74-5
77-3
86-0
86-8

Percentage of Soil sifted through
1 mm. Sieve.

Water.

•64

1-84

•93

•73

•92

1-54

•95

•93

1-36

•85

1-04

•61

1-42

1-07

4*04

1-66

Organic
matter.

2-36
5'55
4-04
3-12
4-39
7-08
3-72
4-08
7'65
3-86
4-49
4-06
4-32
3-52
6-94
4'97

Chlorine. Nitrogen.

•0297
•1825
•0134
•0156
•0775
•0134
•0255
•0095
•0071
•0332
•0318
•0042
•0138
•0966
•0081
•0219

•043
•112
•126
•070
•098
•203
•098
•105
•203
•126
•133
•057
•084
•057
•161
•126

Percentage of Soil sifted
through £ mm. Sieve.

Phos-
Lirne. Potash. phoric

•056
•138
•332
•128
•344
•102
•102
•036
•138
•094
•256
•088
•470
•148
•522
•150

oxide.

•136

•042

•110

•056

•166

•106

•052

•059

•187

•097

•166

•104

•166

•077

•071

•068

•196

•160

•133

•091

•307

•147

•266

•099

•260

•095

•183

•052

•314

•069

•274

•073

Allusion has been, made above to certain of the soils of this Division
whichf resemble those of the vast stretch further inland, known as the
Karroo. So great, in fact, is this resemblance, that a portion of this
Division, and that of Oudtshoorn adjoining thereto, is often called the
" Ladismith and Oudtshoorn Karroo.'' The soil of a great part of thia
area consists of the disintegrated Bokkeveld beds which overlie the sand-
stone formation referred to in connection with the soils of the George Divi-
sion. These Bokkeveld beds consist not merely of sandstone, but also of
siliceous clay slates, and thus, in decomposing, they give rise to a sandy
loam, superior, for agricultural purposes, to the sands, for instance, of the
George Division. This " Oudtshoorn and Ladismith Karroo " covers the
stretch of country which extends eastwards between the Zwartberg and
Langeberg Ranges and their continuations. This belt of country has a
great reputation for fertility, especially in the Oudtshoorn Division, east
of Ladismith ; the cause of this is easily explained : there stretches through-
out this belt a. surface formation of calcareous marl, derived from a thin
stratum of limestone called by local farmers " Kalkbank." Its fertilising
effect is noticeable solely in the valleys, and further east, in the Oudts-
hoorn Division, it is one of the principal factors in the almost unexampled
fertility of the Grobbelaars and Wynands River Valleys, the Cango dis-
trict, and the region about Vlakte Plaats.

MACLEAR.

(Privately collected.;

No. Field Cornetcy. Farm or place. Collector.

1. Wizard's Vale. Thomson. P. W. Leach.

A sample of soil was collected on the farm Thomson, at the junction
of the Tsitsa and Mooi Rivers, from a spot where tobacco was being cul-
tivated. The quality of the tobacco produced had been very highly com-
mended by the Government Tobacco Expert. The soil is, taken all round,
of good quality. Good tobacco had also been grown on the farms Gordon
and Wizard's Vale, in the same district.

The following is the analysis of the above soil : —

(Method II.)

Percentage of Soil sifted through 1 mm. Percentage of Soil sifted

Sieve. through 3 mm. Sieve.

Phos-

No. Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide.

1- 1-75 9-18 -Oil — -87 '38 '19

MAFEKING.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

No. 1. Hartebeestlaagte. S. Minchin.

2 No. 2. West Hill. M. W. B. Webb,

Xo. 1 was a sample of sandy soil, collected from the farm Hartebee6t-
laagte, which adjoins the railway line. It was intended to plant several
hundreds of fruit and ornamental trees on this soil. The sample was
taken about two feet below the surface, and was stated to overlie a granite
formation. This formation is, in fact, exposed along the river beds of the
Molopo and Maratsani.

No. 2 was taken, at a point 25 miles north-east of Maribogo Railway
Station, from lands which it had been proposed to bring under irrigation
and utilise for the growing of lucerne.

According to the following analytical figures, both the abov$ soils are
very poor in all plant food: —

Percent of
Field

(Method I.)

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of Soil sifted
through k mm. Sieve.

Phoe-

No.
1

Fine earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

Potash.

phoric
oxide.

75-2
78-5

1-78
•93

4*53
2-99

•0007
•0025

•042
•056

•036

•064

•048
•056

•056

•041

These poor soils may be considered as representative of the large area
of surface sands and quartzites that stretches westwards from Maf eking for
very many miles.

MALMESBURY.

Groene Kloof East.

No. Field Cornetcy.

1. Mossel Banks Rivier.

2. Middle Zwartland.
3.

4.

5.

6.

7.

5

9.
10.
11.
12.
13.
14.

15. Honing Berg.

16. Zwartland.
17.

18.
19.
20.
21.
22.
23.

(Officially collected.)

Farm or place.
Kalaba-s Kraal Station.
Twee Kuilen.

»

Vaderlandsche Rietkuil.
Bloemendala Fontein.
11 eeboksf ontein .
Michiel Heyns Kraal.

»

Klipfontein.
Karnemelksf ontein .

n

Alexanderfontein.
R e eb oksf o n tein .
Platte Klip.
Geel Dam.
Witkei.

Olifants Kuil.
Geel Kuil.
New Ru&h.

Collector.

J. C. W&termeyer.
C. F. Jurite.

J. C. Watermeyer.

C. F. JuiiU.

Zwart Water.

Zout Rivier.

No. Field Cornetcy.

24. Zwartland.

25. „
26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43. Schryvers Hoek.

44.

45. Saldanha Bay.

46.

47.

48.

49.

50.

51.

52. St. Helena. Bay.

53.

54.

55.

56.

57.

58.

59.

79

Farm or place.

Schildpad Vallei.
Hooi Kraal.
Zwartfoutein.
Vogelstruisfontein.
Klein Zoutfontein.

Zoutfontein.
Droge Viei.

Zwart Water.

Hazen Kraal.

P ortugeeschf o ntei n .

Bosjesmans Kloof.

;>

Breek Muur.
Leliefontein.

»

Geelbeksfontein.
Oostenwal.
Springfontein.
Spanjaardsbosch.
Cloetes Kraal.
Lang Riet Vlei.

Muishondfontein.
Eenzaamlieid.
Nooit^gedacht.
Schuitjes Klip.
Uitkomst.
Noodhulp.
Holle Vallei.
Klip Rug.

Collector.
C. F. Juritz.

J. C. Wateonever.

F. Blersck.

J. C. Watermeyer.

"Hie Malmesbui^y Division gives its name to the geological formation,
which extends over practically the whole of its area, and conadate of a hard
clay slate that ultimately decomposes to a loamy clay soil. The strata
wherein these clay slates occur constitute what is commonly tormexi the
Malmesbury series, and, as a rule, do not yield a rich soil. A more or less
similar condition obtains all along the south-western corner of the Colony,
the Bokkeveld series replacing the older Malmesbury formation in ti«
south, and these two formations stretch practically in succession from the
Division of Piquetberg to that of Riversdale, both inclusive.

Intrusive in these Malmesbury clay slates are out-crops of granite, at
such places as Saldanha Bay, Darling, Groene Kloof, Riebeeks Kasteel,
Paardeberg, Groeneberg, Wellington, Paarl, Simonsberg, Stetieabosch,

Somerset West, etc. The soils derived from this granite are very fertile
when the rock whence they were produced has been thoroughly disin-
tegrated and decomposed but not otherwise.

The two classes of soils above mentioned form the principal broad
types of this division ; they bear a general resemblance, therefore, to those
of the Cape Division.

Proceeding from the village of Malmesbury, about ten miles in a
north-easterly direction, two soils were collected from the farm Twee Kuilen.
The occupant of the farm Vaderlandsche Rietkuil, where the next sample

waiw/j*^ BAY / t

DIVISIONAL MAP OF ^^ ^.tfjSFS^

MALMESBURY \^°^m

,' «

was taken, appeared to have progressed somewhat beyond the system of
manuring by convenience noticed in some of the Cape Division farms:*
he generally employed horse or sheep manure for fertilising his land, but
declared that, according to his experience, the portion of the farm whence
sample No. 4 was taken was sufficiently productive to need no other than
horse manure. Two and three-quarter miles further to the north, from
the comparatively small farm Witkei, situated 15| miles N.N.W. of Mal-
mesbury and occupied by Mr. Gert Rust, three samples were collected.

See page 40.

81

Mr. Rust — who has spent his entire life on the farm, and has kept a con-
tinuous farm diary for over a quarter of a century — states that on one
part of his farm the crops had never yet been seriously attacked by rust,
that rust had, in fact, very seldom so much as appeared on this part of
his lands, and that, during seasons when most of his crops had been com-
pletely ravaged, the cereal crops in this particular locality had enjoyed
entire immunity from the pest. The existence of such patches of land-
where rust seldom or never appears, and where the crops, if attacked,
suffer no serious damage — has not yet been satisfactorily explained. In
many of these cases, such as that under immediate discussion, the chlorides
have been found to be rather high ; in others, the soil has appeared to be
of an all-round better quality than the average. Of the samples of soil
collected on this farm, one — No. 16 on the above list — was taken from
lands that had been attacked by rust season after season, and the crops
destroyed; another sample — No. 17 — was taken from the rust-proof
locality above mentioned. In this ca.se it was noticeable that the under-
lying formation was largely calcareous. Both these samples were pro-
cured from lands which had been treated with horse and sheep manure.
The third sample taken from this farm, No. 18, represents a virgin soil
corresponding in nature to No. 16.

Journeying about four miles in a westerly direction from Witked, the
farm Olifants Kuil was touched at, and sample No. 19 taken. On this
farm, it was said, rust had for the previous six or seven years effected a
considerable amount of damage. From a spot three miles north-west of
this, on the farm Geel Kuil — the southern part of the farm formerly bear-
ing that name — another sample was taken, and then the northern portion
of the old farm, now called New Rush, was visited : here three samples
were collected. No, 21 was a sample of virgin soil, but Nos. 22 and 23
were both taken from ground that had been under cultivation for twelve
years : of these two, the former represents the top of a hill where rust has
occasionally made its appearance, but has never done much hairm to the
crops, while the latter was taken from a low-lying valley, a situation evi-
dently much more subject to damp than that of the previous sample. As
a matter of fact the field whence the last soil was taken is frequently
visited by rust, which at that spot does considerable harm to the crop;
nor can this excite much wonder when the situation is borne in mind.

Turning now the north-east, a sample of soil was obtained at a dia-
tance of four miles from the last— nearly five miles W.S.W. of Moorreea-
burg village, and 21 miles N.N.W. of Malmesbury — on the farm Schildpad
Vallei. Here, as in a few former instances, a visitation of rust rarely
occurs, and, when it does take place, effects comparatively little harm.
The next soil, No. 25, was collected on the farm Hooi Kraal, from a point
rather less than two miles to the north of Moorreesburg.

On the way back to Malmesbury, a sample, No. 26, was collected from
the farm Zwartfontein, about 2J miles south-east of Moorreesburg, another
from the farm Vogelstruisfontein, otherwise known as Drie Heuvel, nearly
5| miles south of Moorreesburg, and 16 miles north of Malmesbury. Two
samples were then collected on the farm Klein Zoutfontein, at distances
respectively 3J and 4§ miles south-east of the spot where the previous
sample waa procured. The last sample of this tour was taken on the farm
Zoutfontein, nearly ten miles due north of Malmesbury.

On a subsequent journey the first farm visited after leaving Malmea-
bury was Bloemendalsfontein, about 6 \ miles north-west of the village :
there a sample of uncultivated soil, No. 5, a sandy clay, was obtained. At
th« farm Klipfontein, about 4j miles further west, No. 9, also an uncul-
tivated soil, even more sandy than the previous one, was taken. Six miles

82

north-west of this, on the farm Karnemelksfontein, two samples (Noe. 10
and 11) were secured from land which had lain uncultivated for six years,
and was being prepared for the next season's sowing. At this spot the
soil was clay of somewhat sandy nature. Mr. F. Duckitt, the owner of
the farm, mentioned that here, on one and the same piece of laud, the
yield was apt to differ considerably; that of the low-lying portions being
inferior to that of the higher ground; two samples were therefore taken,
Xo. 10 from high ground, and No. 11 from a lower level, at spots within
a few yards of each other, in order to ascertain whether any difference in
chemical composition could be detected, or whether excessive moisture
in the low-lying parts was the cause of the variation in the crop. Of the
two soils, the former was found to possess a higher retentive capacity for
moisture; it is also better supplied with lime, but in both cases potash
and phosphoric oxide are very low. The hills about the village of Darling
are, to a great extent, granitic, and, although the farm of Mr. C. Duckitt,
The Towers, was visited, the granitic grain land was altogether omitted,
nor was any sample of the clay soil taken here, as a sample (No. 12) of
soil of the same type was procured four miles north-west of Karnemelks-
fontein, at Alexanderfontein, the farm of Mr. M. Duckitt, from land
which was being prepared for the next season's use, after having been
uncultivated for four years. On proceeding to the farm Reeboksfontein,
3J miles west of Alexanderfontein, and principally granitic soil, a sample,
No. 13, of clay soil was obtained; this soil had been uncultivated for three
years. On the farm Platteklip, 2| miles north of Reeboksfontein, a
sample of sandy clay soil, No. 14, was collected; this had not been cul-
tivated for one year, but was being prepared for the next season. These
last three soils are alluvial clays, No. 14 being evidently affected by the
granite which underlies them, all : this also manifests itself in the smaller
percentage of lime and the higher proportion of potash that it contains.
On the farm Drogevlei, two miles north-west of Platteklip, two samples,
Nos. 31 and 32, about one hundred yards apart, were taken frn-ni laud
said to be extremely fertile, lying to the north of a saltpan. The- first
soil was a greyish-coloured stiff clay, which proved to be rich in lime and
potash, and contained a fair quantity of phosphates; the second, a- lighter,
more or less sandy clay of a reddish colour, contained a moderate amount/
of lime and potash, but proved decidedly poor in phosphoric oxide. This
soil is reputed to be rust-resistent, and is cultivated without manuring.
On the farm Zwartwater, about a mile to the north-west, No. 33, a soil
similar in appearance to No. 31, was taken for comparison; it was found
to be chemically much the poorer. Why these soils should be superior
to others as regards immunity from rust, does not appear from the chemi-
cal analyses. The position of this soil with respect to the adjoining salt-
pan is similar to that of No. 31. The next sample, No». 34, was taken
from sandy soil, 1J mile further north-west, and not rust-resistenb. This
soil had not been cultivated for three years, and is never manured. Prac-
tical experience had found it poorer than the preceding sample, and chemi-
cal analysis confirms this, although the difference is not great. In this
neighbourhood too farmers have frequently observed that, while the sur-
rounding crops are affected by rust, certain patches of ground year after
year escape visitation, or, when affected, do not suffer greatly. It was
from such a patch that sample No. 35 was ta<ken. The particular
patch extends over about 40 or 50 acres; it had never been manured,
a.nd was used for the cultivation of wheat every alternate year, lying fallow
in the intervening years. Proceeding from this farm, a sandy tract is
entered upon, which stretches as far as the coast, about nine or tea miles.
\Underlying the stratum of sand, which is two feet or more in depth, is a

bed of limestone. This limestone bed had already been noticed on the
farm Klipfonteiu.

From the farm Geelbeksfonteiii, about 9i miles north-west of Zwart-
water, a sample, No. 43, was collected, consisting of hitherto uncultivated
humns soil, which was to be sown upon, without manuring, the next

ti. The only other grain land in this vicinity is granitic. This humus
soil as well as most of the sandy soils, and the limestone soils, which will
be mentioned presently, is simply ploughed and sown immediately, not
being fallowed previously, as has been the practice with the clay soils : in
this case, apparently, the opinion is that the productiveness of the soil
renders fallowing needless. At Oostenwal, 7J miles north-west of Geel-
beksfonteiii, a sample of virgin soil, No. 44 — a mixture of limestone with
granitic soil — was taken. This soil is also cultivated without manuring.
The soil of the adjacent faaun, Meeuwklip, resembles that of Oostenwal;
at Karnberg sandy soil is again met with, limestone being not far below :
over Zoutkuil to Springfontein, about nine miles north-east of Oostenwal,
where No. t"> was collected, the sand diminishes and the limestone in-
creases. At Springfontein the limestone is exposed to such an extent that
the surface of the otherwise hard material has become sufficiently friable
to be easily cut up by the plough, and mixed with the inch or two* of
sand that covers it. This soil is sown without manuring, and fanners
who were questioned 011 the subject, including the occupant of Springfon-
tein, state that the limestone soil to a great extent withstands the rust,
and that, at times, -when the grain sown on the sandy soil is almost en-
tirely destroyed bv rust, that on the lime soil is only slightly affected.

Leaving Springfontein, and passing over Kersbosch and Klipfouteiii
to Spanjaardsbosch, about six miles north of Springfontein, the soil
becomes more sandy, and is here of about the same appearance as that of
Zooitkuil and Karnberg further to the south-west. From Spanjaards-
bosch a sample (No. 46) of uncultivated sandy soil, having somewhat of
an admixture of lime, was collected. Thence the journey was directed to
Oloete's Kraal, about 3-J miles north of Spanjaardsbosch, where the soil
loses its very sandy nature, and here also a sample (No. 47) of unculti-
vated soil was procured. The next farm is that of the Brothers Kotze,
Lang Riet Vlei, at Berg Stiver, about five miles east of Cloete's Kraal.
Here, at the request of Messrs. Kotze, four samples were taken at spots
pointed out by them, namely : No. 48, " Vaalbos " ground, where the
Cape salt bush grows luxuriantly, a dark grey, somewhat stiff clay soil;
No. 49 a looser and lighter clay, said to be the richest grain land on the
farm; No. 50, a sandy soil, low-lying and yielding poor returns; and No.
51, a sandy soil, similar to the previous one, at a higher situation, and
Yielding better crops physically as well as chemically it proves to be the
Ibetter soil of the two.

Between Lang Riet Vlei and Hazenkraal, about eight miles south-
east of the former, alluvial vlei deposit stretches all the way along the
Berg River, skirted by sand. The way from Hazenkraal to the village
of Hopeneld lies along the bed of the Zout River, and the elevated land,
under cultivation on either side of the river bed from here to Hopeneld,

s a coarser sandstone formation than that previously passed over, no indi-
cations of the presence of limestone being noticed. A sample of this soil,
No. 36, was taken from uncultivated land at Hazenkraal, and again,
nearly four miles south of this, No. 37, at Portugeeschfontein. The former

£ these two samples contains a slight admixture of clay. These wore the
last samples taken on this tour, for on leaving Hopefield the Malmesbury
clay beds are almost immediately entered upon. These soils afford a fair

idea of the grain country from Darling to Saldanha Bay, and thence to
Hopefield. The intermediate sandy country is mainly devoted to cattle
and sheep farming.

Regarding the non-manuring of the lands, the impression conveyed is
that manuring is dispensed with wherever possible, as much owing te the
want of a sufficiency of stable and cattle manure, as on account of the re-
puted fertility of the soil ; but the fact cannot be too strongly emphasised,
that the richest, soil is bound to be impoverished in course of time, if cul-
tivated without manuring, and it behoves agriculturists to guard their
interests, and attend to the quality of their lands, and merchants to place
within reach of the farmer artificial fertilisers at as reasonable rates as
possible, so that the farmer may learn their value, and the whole agricul-
tural community, as also the country at large, may profit by the experi-
ence.

The final trip of this series was undertaken with a view to obtain
samples from the St. Helena Bay district, and, at the same time, to supply
the omissions of previous journeys, and thus render the investigation more
complete. St. Helena Bay, it need scarcely be said, is an important grain-
growing district, and the land, according to the farmers, is so fertile that
it can be cultivated continuously without manure, the chief obstacle to suc-
cess being the low rainfall.

The samples collected on this journey are detailed below. A light
clay soil, No. 6, was taken on the farm Reeboksfontein, about five miles
north of Malmesbury. No further samples were collected in this neigh-
bourhood, as, with those previously collected, the work had been practi-
cally completed as regards the Middle Zwartland farms. Proceeding
northwards, a stiff clay soil, No. 15, was collected on the farm Geeldam,
a portion of Holle Rivier, seven miles from Moorreesburg in a north-
easterly direction. About nine miles north-west of this Nos. 38 and 39
were taken on Mr. M. Karsten's farm Bosjesmaus Kloof ; the former of
these two is a light clay, the latter a very stony clay soil. In Mr.
Karsten's experience No. 38 is the poorer. On the farm Breek Muur,
which borders on the Berg River, and lies six miles north-west of Bosjes-
mans Kloof, sample No. 40 was taken. Here the soil is more sandy in
character, being, in fact, a sandy loam. Eight miles south-west of this,
on the farm Leliefontein, sample No. 41 was taken; this is a stiff white
clay, somewhat stony, but said to be very fertile. The lime and nitrogen
in this soil are satisfactory in amount, and the potash fair, but phosphates
are deficient. On the same farm, about 2J miles further south-west,
sample No. 42, a sandy loam, was collected. The farmers in this part
use stable manure, and also guano and artificial fertilisers in small quan-
tities.

Before reaching the St. Helena Bay a«rea, the farm Eeuza-amheid,
about 16 or 17 miles north-west of Hopefield, was visited. Two samples,
Nos. 52 and 53, were taken here; the former on the portion of the farm
known as Muishondfontein, and the latter about two miles further west
on Eenzaamheid proper. The soil here is sandy, resembling that of
Spanjaardsbosch. The farm Eenzaamheid, and other farms forming the
inland boundary of the Field-Cornetcy of St. Helena Bay, are situated
on a low-lying, sandy flat, whilst, with the exception of a narrow etrip
of similar sandy flat along the coastline, the other farms lie at elevations
of probably 500 to 700 feet, upon and around granitic hills, the highest
of which may be 900 feet. The soils of this part are, therefore, moetly
granitic. TH"e first sample taken here, No. 54, was from the faj-m
Nooitgedacht, portion of Patrysberg, about five miles north-east of Vreden-
burg; a»nd seven miles north-west of this, on th& farm Schuibjee Klip,

85

cample No. 55 was taken. These two soils are of a rather sandy nature.
On the farm Uitkonist, adjoining Schuitjes Klip, and about 2J milee to
the south-west, sample No. 56, a clay soil, in appearance resembling the
clay soils about Malmesbury, was collected. Judging by its agricultural
returns, this soil was expected to yield good analytical results, and it
certainly does show, in addition to a fair proportion of nitrogen, more
lime, potash, and phosphoric oxide than any other soil collected within
the Field-Cornetcy. The three soils which complete the circuit in the
neighbourhood are sandy loams, clay being more in evidence on the farms
west of Uitkonist than on those east of it. No. 57 was taken on the farm
Noodhulp, three miles south-west of Uitkomst, No. 58 on the farm Holle
Vallei, about 6£ miles south-east of Noodhulp, and No. 59 on the farm
Kliprug, about three miles south-east of Holle Vallei. The reputation, of
the soil is good ; chemically it shows a normal proportion of nitrogen,
and, although poor in phosphates, it contains fair amounts of lime and
potash.

It remains to add that samples 7 and 8 were taken on the farm
Michiel Heyns Kraal, eight miles south-west of Malmesbury : this is not
a grain, but a dairy farm. No. 7 represents a loam and No. 8 a humus
soil. No. 1, a sandy loam, was taken from Crown lands at Kalabae Kraal
Station, seven miles north-east of the farm Lange Rug in the Cape Divi-
sion ; it represented the only uncultivated ground in the vicinity. All
the- samples collected on this journey represent virgin soils.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

€0. Mosselbanks Rivier. Kalabas Kraal. D. E. Hutchins.

61.

62.

63. „ ,, „

64. ,, „ „

65. Greene Kloof East. Alexanderfontein. J. P. Cloete.
6t>. ., „ „

67. ,, ,, „

68. „ ,, „
69-

»
71. Middle Zwartland. Leliefontein. C. Nelson.

From the Kalabas Kraal Outspan five samples were taken: No. 60
represents the subsoil from a hill top, and No. 61 the surface soil at the
same pla/ce. No. 62 represents a light loam from the valley, and Nos. 63
and 64 are sands, the former being from sour veld and the latter mixed
sand from the river bed.

While travelling through the Malmesbury Division in connection
with the collection of the samples mentioned in the previous list, atten-
tion was frequently directed by the local farmers to numerous slight
elevations, from one to four feet in height, and twenty or more yards in
diameter; the soil of these hillocks — called " heuveltjes " by the farmers
— was alleged to be extremely rich, and cereals of all kinds were said to
grow on them with luxuriance; while on the lower ground, between the
•legations, the soil would be poor and produce scanty crops. It is not
customary— so it is asserted — ever to manure these hillocks, and some
lauds are alleged to have grown wheat for nearly a century without the
hillocks either receiving any manure or becoming exhausted. Mr. J. P.

86

Cloete, of Alexaudersfontein, through whose good offices specimens of
hillock and other soils were procured for comparative analysis, stated that
for the last four years he had been urging farmers to use lime largely on
the poor cold soils between the hillocks, and he quoted instances of very
poor lands having yielded heavy crops of wheat by the aid of a good
dressing of lime.

In order to ascertain by chemical analysis what difference, if any,,
existed between the hillock soils and those of the lower levels, samples of
each were specially collected and analysed. Those from hillocks are num-
bered 65, 66, and 67, and those from the level below 68, 69, and 70: all
of these were taken from lands that had been cultivated. In every case
the soils taken from the low-lying ground proved to be exceedingly poor
in lime, and herein lies the great difference between the hillock soils and
those below, curiously enough tending to verify Mr. Cloete's prognostica-
tions. Even the hillock soils Nos. 66 and 67 are rather deficient in limey
although considerably superior to the lower soils. No. 65 contains lime
in fair amount. The potash present in the hillock soils is fair in quan-
tity, but in this respect No. 69 is poor, while Nos. 68 and 70 show a
moderate proportion, although in each case poorer than the correspond-
ing samples from the hillocks. As far as phosphates go, there is a fair
proportion in the hillock soils, but the other three are rather poor. These
few analyses tend to confirm the popular idea; and yet the difference all
round is not as striking as sooiie of the statements made might possibly
have led one to expect. To this last observation there is just the excep-
tion already noted — that of the lime. Physically as well as chemically,
the hillock soils appear to be slightly superior : in water retentive capa-
city their average stands higher thaw that of the soils around, a point
also noted in connection with the soils collected on the farm Karnemelks-
fontein (vide p. 82). There is, moreover, a better proportion of organic
matter in the hillock soils than in the others, and the former are likewise
the richer in nitrogen. Without unduly pressing the points of difference,,
the inferiority of the low-lying soils also comes out in the amounts of
chlorine they contain, in as far as these amounts indicate tendency to be-
come brack : all along the line, therefore, the hillock soils have points in.
their favour.

Reverting for a moment to the soils from Karnemelkslontein, it may
be observed that there too the hillock soil was the better, not only in
water retaining capacity, but likewise in organic matter and nitrogen, and
there too the chief difference lay in the proportion of lime.

Another instance of comparative analyses of hillock and level soils
will be briefly referred to in connection with the farm Groenberg in the
Paarl Division.

Under all the circumstances it seems quite feasible that the process
of levelling down the hillocks — said to have been attended with general
improvement of the land in some instances — may have led to an all-round
increase in fertility, notably where these hillocks have been numerous.
It seems also reasonable to suppose, from what these results reveal, that
an addition of lime would lead to an improvement. Speaking of car-
bonate of lime as a dominant factor in soil productiveness, Professor Hil-
garcl observes : — *

" Its presence exerts a dominant and beneficial influence in many respect-, us i< readily
apparent from the prompt change in vegetation whenever it is introduced into soil*
deficient in it."

* Hilgard : " Soils ; their formation, properties, composition and relations to climate
and plant growth." 1006, pp. :}.r>:?, :5:-4.

S7

In the Malnieabury Division there are several outcrops of carbonate
of lime> for instance on the famis Drooge Vlei (Field-cornetcy Zwart-
v.-ater). Geelbeksfontein (Field-cornetcy Schryvers Hoek), Springfontem,
and Lang Riet Vlei (Field-cornetcy Saldanha Bay). Even if levelling-
down does not achieve the desired result, there should be abundance of
lime near at hand on which to draw for a supply.

The existence of these hillocks in various parts of this Colony has
been attributed to insects — ants, presumably. In this connection atten-
tion may be directed to the following : —

"The work of ants is in some regions on so large a scale as to attract the attention of
the most casual observer. Especially is this the case in portions of the arid region, from
Texas to Montana, where at times large areas are so thickly studded with hills from three
to twelve feet in diameter, and one to two feet high, that it is difficult to paas without
being attacked by the insect*. The ; mounds ' studding a large portion of the prairie
country of Louisiana seem also to be due to the work of ants, although not inhabited at
present.'"*

The last sentence seems to fit exactly the conditions of the Cape
" heuveltjes."

The results of the analyses of the soils from the Malmesbury Divisiort
a-re tabulated below: —

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of Soil sifted
through 8 mm. Sieve.

Pho*-

Sfo.

Water.

Organic

Chlorine.

Xitrogen.

Lime.

Potash.

phoric

matter.

oxide.

1.

•30

•85

•0006

•014

•059

•041

•016

60.

I'll

2-22

•0101

—

•064

•026

•041

61.

•22

•44

•0024

—

•037

•018

•052

62.

1-46

2-40

•0056

—

•097

•044

•038

63.

•34

•52

•0025

—

•046

•025

•034

64.

1-23

1-28

•0070

—

•060

•023

•023

0

•49

1-90

•0003

•061

•056

•107

•051

3]

•68

2-47

•0003

•078

•092

•171

•071

4,

1-16

5-24

•0008

•095

•136

•128

•064

5.

•14

ro7

•0004

•050

•059

•038

•025

6.

•91

2'91

•0014

•091

•049

•031

•030

7,

•67

2-30

•0011

•070

•168

•039

•038

7-16

15-36

•0056

•252

•369

•033

•080

9.

•17

•95

•0008

•067

•039

•042

•033

10.

1-03

4-44

•0014

•089

•147

•059

•041

11.

•29

2-16

•0006 .

•072

•062

•064

•022

12.

•49

3-12

•0010

•129

•081

•035

•045

13.

•80

4-16

•0032

•117

•095

•098

•048

14.

•56

2-24

•0012

•035

•046

•102

•050

15.

•94

2-50

•0009

•063

•064

•074

•032

16.

•93

3-60

•0022

•117

•160

•130

•056

17.

•80

1-86

•0347

•072

•056

•077

•044

IS.

1-05

2-79

•0010

•095

•108

•101

•051

19.

•62

2-68

•0004

•056

•104

•062

•038

20.

1-23

4-02

•0007

•033

•036

•119

•035

21.

•60

2'99

•0012

•078

•028

•144

•071

2-2.

•58

2-61

•0010

•084

•082

•090

•064

23.

•59

2-53

•0081

•061

•098

•092

•051

24.

•46

2-04

•0002

•067

•060

•020

•074

•>/>. <->t. p. hit).

88
(Method II) — continued.

Percentage of Soil sifted through 1 mm.
Sieve,

Percentage of Soil sifted
through 3 mm. Sieve.

Ko.

25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

•68

2-94

•0009

•084

•032

•033

•053

•35

1-72

•0005

•072

•064

•042

•040

•73

5-17

•0002

•100

•076

•090

•076

•37

1-63

•0003

•056

•052

•042

•066

1-68

3-17

•0005

•156

•068

•090

•063

•31

1-76

•0002

•077

•036

•045

•086

7-25

9'26

•0058

•140

1-991

•492

•063

•93

2-14

•0057

•028

•156

•122

•028

•78

1-81

•0025

•035

•125

•075

•033

•48

2-04

•0017

•028

•108

•054

•039

3-16

14-30

•018

—

12-07

•44

•17

•24

1-06

•0050

•042

•024

•045

•124

•17

•59

•0005

•021

•053

•018

•034

•87

1-83

•0103

•091

•187

•066

•042

1-08

7-90

•0042

•133

•010

•052

•058

•70

1*55

•0020

•077

•046

•048

•042

T94

4-09

•0108

•126

•256

•075

•027

•39

1-20

•0009

•035

•039

•026

•038

1-85

16-26

•2258

•325

1-159

•443

•180

•78

3-29

•0013

•042

•364

•124

•052

•52

2'43

•0016

•035

4-715

•058

•025

•42

•94

•0015

•049

•231

•037

•075

1-17

2-59

•0006

•049

•220

•068

•055

1-98

2'97

•0147

•070

1-826

•182

•053

•58

1-31

•0022

•047

•073

•063

•027

•20

•54

•0006

•028

•063

•046

•025

•51

1-01

•0008

•028

•114

•061

•034

•41

•68

•0006

•084

•084

•042

•048

•35

•67

•0010

•056

•034

•035

•027

•47

1-12

•0009

•049

•062

•046

•027

1-01

2'53

•0093

•035

•015

•021

•050

2^32

3-48

•0017

•091

•418

•105

•094

1-86

2-55

•0011

•077

•165

•062

•046

1-28

4-82

•0024

•084

•043

•039

•027

1-94

4-30

•0016

•112

•339

•060

•045

No.

65.
66.
67.
$8.
69.
70.
71.

(Method I.)

Percent, of Percentage of Soil sifted through

Field 1 mm. Sieve.

Sample.

Fine earth. Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through $ mm. Sieve.

Phos-

Lime. Potash, phoric
oxide.

72-0

1-57

5-83

•0106

•175

•146

•121

•061

73-8

1-15

3-18

•0078

•189

•072

•075

•072

80-0

1-47

5-01

•0042

•126

•096

•095

•073

72'2

1-01

4-33

•0261

•119

•014

•114

•049

68-0

1-09

2'64

-0159

•112

•014

•045

•028

76-0

1-28

4-30

•0078

•119

•014

•09f»

•055

31-8

•28

1-95

•0071

•091

•032

•020

•017

89

In the Malmetjbuiy grain soils, as in those of the Cape Division, the
^ lack is phosphatic material, and potash is almost as urgently

needed ; at the same time it must be said that, not only in proportion
to the lime present, but also absolutely, the aonount of phosphoric oxide
is higher in the Malmesbury than in the Cape Division, and seems to
continue increasing in amount as one travels northward from Tygerberg
to Zwartland.

It is noteworthy that the surface soils which are more or less in-
fluenced by the underlying lime deposits — such as those at Drooge Vlei,
Geelbeksfontein, Oostenwal, and Uitkomst — in addition to containing
more lime than the other soils, are also richer in potash. The ultimate
origin of the comparatively large amount of potash in the soils of this
.class is a point of some interest which it is worth while elucidating: it
does not seem improbable that it is caused by the debris of granitic
rocks being mixed with the compacted sand; fro-m the blown sand itself
the potash coiild certainly not have been derived.*

Out of 68 soils overlying the rocks of the Malmesbury series, col-
lected in the Cape and Malmesbury Divisions, no less than 16 were de-
ficient in all three inorganic plant food constituents in an available form;
as many as 45 are poor in phosphoric oxide; and of these latter five
also lack potash, and eight lime. Of the remaining 23 soils, eight were
deficient in lime only, three in potash only, and five in both lime and
potash. There were, therefore, only seven soils, of all those 63, that
were not lacking in respect of one or other of the three mineral fertilising
constituents, and even out of these seven, six were no better than fail
all round.

MIDDELBURG.

(Privately collected.)

No. Field CornetcY. Farm or place. Collector.

1. Nauwpoort. Carlton. Dist. Railway Engineer.

Amongst the characteristic features of the Beaufort geological series,
the occurrence of bands of dark purple mudstones and shales may be
noticed by any traveller by rail in those parts of the country where the
rocks of the Beaufort series appear at the surface, for instance, between
Nauwpoort Junction and Ludlow on the Midland Railways. References
to these mudstones and shales may be found, inter alia, on pages 33 and
174 of the Geological Commission's Report for 1903, on pages 78 and
81 of the Commission's Report for 1904, and on page 98 of that for 1905.

These purple shales are exposed in various parts of the railway cut-
tangs on the west of the railway line south of Carlton, and especially
so just at the 262nd milepost. They frequently appear to be of a cal-
careous nature, and to investigate this point, some of the weathered and
finely disintegrated shale was procured for analysis. As the ahale IB
found intermixed with limestone bands and layers of sandstone, no op-
portunity has been available of procuring any of this purple material in
a sufficiently decomposed condition to form an actual soil by itself, but
the above sample, taken from the cutting south of Carlton, represents
the weathered shale in as finely divided a condition as it was possible

* A somewhat similar problem arises in connection with the soils of the farm Hooge-
kraal, Xos. 33 and 34 of the Cape Division soils (fide pages 40 and 43 of this volume).
The point will be reverted to later when discussing the bearing which the geological
relation? of «ril? have upon the proportions of plant food which they contain : see Part V.

90

feo procure it free from other rock material that could in any way modify
the chemical composition of the sample.

In order to afford an idea of its state of sub-division, it may be men-
tioned that a mechanical analysis of the shale as collected gave the fol-
lowing1 results : —

Pebbles- 12-55

Coarse gravel 12'70

Fine gravel 36*55

Coarse sand 10*74

Fine earth (by difference) 26'75

Moisture '71

Total 100-00

Tir.c chemical analysis of this sample yielded the following resulta: —

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 mm. Sieve. through £ mm. Sieve.

Sample. Pho»-

No. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide

1.

1-86 —

— -476

120

It is obvious that the disintegration of these purple shales will pro-
duce a. very fertile soil, and indeed it was the known fertility of soils
into whose composition the shales had appeared to enter that led to th«
selection of this sample for analysis.

MOSSEL BAY.

No.

1.
2

3'

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.

Field Cornetcy.
South Middelveld.

(Officially collected.)

Farm or place.

Buff els Drift.
Ha«rtjesfontein.

Collector.
J. Muller.

Upper Gouritz River. Hemelrood.

„ Heuningbosch.

Before Attaquas Kloof. Hagel Kraal.

Brak River.

Mossel Bav.

Ruiter Bosch.
Hartebeeste Kraal.

j>
Great Brak River.

»
Klipheuvel.

Geelbeks Vallei.

Hartenbosch.

Patrysfontein.

Entering the Mossel Bay Division from Riversdale, at the farm
Buffels Drift, which lies on the boundary of the two divisions, a typical
sample of river soil was collected. No. 1 in the above list. The Gourit*

The signification of *a,ch of these terms will be explained later, under the head of
il Phj«i«*l composition of soils."

JM

River ;iows midway through the farm, aud the river bed widens outr
considerably at> that point; extensive and very deep deposits of silt are
washed down and deposited anew all along the river course, serving a»
an excellent soil for gardening. No. 1 forms a representative sample of
•uch a deposit of very rich and deep soil, which, on most of the farma
lying along this river bed, is brought under irrigation by air-motors.

The silt thus deposited by the Gouritz River is identical with that
tc which the whole of the Oudtshoorn Division owes its great fertility,
being brought down — ultimately from the Karroo — by the Olifants River
and other tributaries of the Gouritz. It is hardly open to doubt that, if
some scheme could be devised to intercept by means of retaining weir*-

OUDT3HO ORN / \

a /

UPPER GOUHITZ RfVER

\\ MOUTH
b i
X COURITZ RIVER

DIVISIONAL MAP OF

MOSSEL BAY

or otherwise, the rich Oudtehoorn silt which these rivers are constantly
carrying into the sea, and to spread it over the adjacent farms of the-
Mossel Bay and Riversdale Divisions, great benefit to those areas would
result.

Crossing the Gouritz River bridge, and leaving the main road, a*
course along the left bank of the river was taken, passing through Roode
Hoogte, the very fertile farm of Messrs. Muller Bros., the soil of which
U similar to that at Buffels Drift, lower down the river; then, proceed-
ing up the left bank of the tributary Langtouw River, to the farm,
Hemelrood, better known as Herbert's Dale — quite a small village—

92

•ample No. 4, a loose red sandy clay, was collected on the right of the
road before entering this village. It represents the predominating soil
all along the valley, which, like several others, is well cultivated, mostly
for garden purposes, all sowing being done on the slopes of the surround-
ing hills, capped as they are with a fairly deep layer of rich " Karroo ;>
•oil.

On leaving the Langtouw River valley, a very hilly tract is traversed
to Heuning Bosch, where No. 5, a sour soil, representing the upper and
more mountainous parts, was taken from old lands adjoining those just
under the plough. It was said to be a very poor soil, but yielded fairly
satisfactory crops when manured with guano. The same type of soil is
also to be found at Ruiters Kraal and Goedmoed. These poor soils are
all derived from the sandstone formation to the north. Keeping along
this ridge, a series of undulating plateaux is crossed over; they consist
mostly of sour veld, with some occasional tracts of rich land, alternately
" zwart turf" and natural red "Karroo" soil. Nos. 6 and 7, respectively
typical of each, were secured at the farm Hagel Kraal. The " kweek "
•oil along the river is literally covered with the grass which obtains for
it its vernacular appellation, a.nd proves a great inconvenience and draw-
back to the growth of lucerne. At Roode Krans and Kouma a more
or less similar soil is to be found, possibly somewhat more acid. At
Ruiterbosch a sample of very acid soil, No. 8, was taken. Both at this
farm and at Hagel Kraal, basic slag had been widely used as a fertiliser
for cereals, but too sparingly, for it is a well-known fact that these acid
soils, or " zuur veld " as they are locally termed, require principally lime
.and phosphates.

Between the last mentioned farm and Hartebeeste Kraal, there is
not much sowing carried on ; at the latter place, however, several
varieties of soil were under cultivation, of which only two were selected,
namely, No. 9, a " broken " red sour soil, not very fertile, and No. 10,
a black sour soil, inferior to No, 9. These two soils were taken from,
above the mass of granite which commences north-west of Mossel Bay,
and extends over a considerable portion of the George Division. The
relatively large amount of potash in No. 9 is, not improbably, due to
the felspar of the granite.*

From this point a course was laid for Great Brak River, the pic-
turesque little hamlet situated on the river bearing that name, the boun-
dary between the divisions of Mo&sel Bay and George ; practically the
whole of this hamlet is owned by Messrs. Searle Bros. In the valley
on the right bank of the liver, between the latter and the main road,
there is a very level tract of an intermediate " broken vlei soil " ; two
•amples were collected here, Nos. 11 and 12; the former of these is a
loose, rather sandy clay, about nine inches deep, lying upon a bed ol
yellow clay ; the other sample represents a loose sandy loam, which, after
sowing for three successive years, becomes so " brack " or alkaline, thai
its owner is compelled, through poverty of the crops, to let the land lie
fallow, for a year or two, until entirely covered with grass, when it is
again ploughed over, well manured, and sown; fairly good results have
thus been obtained.

At KHpheuvel there are several varieties of " broken vlei soil "
under cultivation. Only Nos. 13 and 14, the former a rich dark loaooo,
the latter a red sandy soil, were taken. Of these two soils, the latter ha*
constantly been subject to visitations of rust; it is in every respect
chemically the poorer soil of the two. Proceeding about three or four
miles south-east of KHpheuvel, sample No. 15, a loose brack soil,

* A reference to this feature has already hem made in connection with the

bury «oils. See p»pe 89.

collected from the farm Geelbeks Vallei. The proportion of chlorine, ib
will be noticed from the following table, is higher than in any other
of the Mossel Bay Division soils analysed. Sample No. 16, a light, yellow-
toil, with alkaline patches here and there, was taken from lands adjoining
the roadside, on the farm Hartenbosch. Journeying almost due west
from the last mentioned farm, over several varieties of " broken " soil,
to Hartjesfontein, more commonly known as Matjes Drift, some very
fertile varieties of alluvial clay soil were found. No. 2 is a good
"Karroo" soil, while No. 3 is a dark loose clay, said to be very fertile.
It is probable that this fertility is due to the lime present in the soil,
the proportion being the maximum for the Mossel Bay Division. Noe.
2 and 3 represent the two classes of soil mostly under cultivation here.
There is also a third variety, locally known as " vaal grond," but not
equally extensively cultivated. The last sample taken in this division
was No. 17, a loose gravelly clay soil, from the farm Patrysfontein. The
range of hills south of this is composed of sandstone, and forms the
natural boundary between the central belt of country, and the sand dune*
which stretch along the coast. The comparative poverty of the Harten-
bosch and Patrysfontein soils is evidently due to the influence of the
sandstone.

No.

18.
19.

20.

Field Cornet cy.
Mossel Bay.

(Privately collected.)

Farm or place.
Mcflsel Bay.

Collector.
C. W. Blade.

These soils were taken from the Public Park at Mossel Bay, whera
the underlying geological formation is Table Mountain sandstone; they
all represent poor sandy soils, No. 18 containing an admixture of a small
proportion of clay.

The analyses of the soils enumerated in the foregoing lists are
tabulated below : —

(Method III.)

No.
1.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.

Percent, of
Field

Percentage of Soil sifted through
1 mm. Sieve.

Sample.
Fine earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

97-0

2'52

6'29

•0062

•056

93-6

2-99

4'87

•037

•043

92'2

6-95

10-24

•035

•17

93-6

1-58

3'54

•on

•028

75-6

2-52

4'98

•016

•042

95'2

2-66

5'67

•on

•030

88'2

2'75

4-47

•040

•17

95-2

1-97

4*17

•012

•044

81'6

3'52

5'68

•032

•056

54-2

•92

2-47

•0071

•031

934

2'73

5-01

•0079

•046

95-2

1-97

6-29

•047

•044

79-0

2-12

4'41

•044

•067

94-0

1-05

2-13

•044

•029

92'6

2'21

5-71

•057

•056

794

•46

•89

•021

•045

i< -

•78

2-29

•02€

•13

Percentage of Soil sifted

through

£ mm.

Sieve.

Phos-

Lime. Potash.

phoric

oxide.

•43

•39

•13

•23

•46

•070

•59

•87

•12

•10

•39

•074

•16

•20

•054

•15

•25

•11

•13

•36

•059

•15

•080

•064

•15

•63

•074

•13

•18

•061

•39

•58

•056

•30

•34

•10

•31

•56

•092

•15

•26

•058

•40

•76

•15

•11

•14

•046

•10

•1$

•033

94

18.
19.
30.

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Water. Organic
matter.

ri5

•37
•81

3-95

1-66

•66

Chlorine. Nitrogen.

•0014
•0011
•0026

•123
•016
•084

Percentage of Soil sit'tel
through 3 mm. Sieve.

Phot-
Lime. Potash. phoric
oxide.

•125
•051
•055

•076
•026
•049

•038
•017

•018

It is to be expected that the method whereby most of the soils from
~this division have been analysed will yield higher proportions of lime and
potash than in the case of soil-extraction by Method I. ; to exemplify this
the analytical results given in the above table for the poor sour soil at
Helming Bosch (No. 5) should be compared with the analyses of similar
sandstone-derived soils extracted by Method I. The results obtained
from the Hartenbosch and Patrysfontein soils (Nos. 16 and 17) should
.Also be thus compared.

MOUNT CURRIE.

No.

1.

2.

3.

4.

Field Cornetcy.
No. 7.

No!' 5.

?j

(Privately collected.)

Farm or place.
Tiger Valley.

Poortje.

Collector.
J. H. Cordeioy.

J. G. King.

Two very poor gravelly soils were sampled at Tiger Valley; these
are represented in the above list by samples 1 and 2. Nos. 3 and 4 were
collected on the farm Poortje, near Kokstad. The soil, which is peaty,
is about a foot deep, generally overlying red clay, although in some cases
resting upon yellow shale. Some 200 acres had been ploughed up, but
the agricultural capabilities of the land had proved disappointing,

The following results were obtained by analysis of the above four
soils : —

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 mm. Sieve. through J mm. Sieve.

Sample. Phos-

l^o. Fine earth. Water, Organic Chlorine. Nitrogen. Lime. Potash, phorfo

matter. oxide.

1.

2.

56-3
63'7

1-16
1-74

3-56
4'56

•007
•007

•028
•125

•006
•008

•031
•020

•055
•037

3.
4.

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Water.

6-22
5'74

Organic
matter.

20-40
10-52

Chlorine. Nitrogen.

•032

•038 —

Percentage of Soil sifted
through 3 mm. Sieve.

Phos-

Lime. Potash. phoric
oxide.

•075
•062

'20
•20

•074

The disappointing character of the agricultural returns obtained from
Noe. 3 and 4 exemplifies at the same time the value of chemical analysis
as a means of judging of the fertdlity of some soils, and the lair of the

minimum* in respect of the need of all plant food in adequate proportion.
Although in the soils referred to, potash is present in satisfactory amount,
and phosphates in fair proportion, the percentage of lime is low, as in-
deed it is also in the other soil from this district; in such a case it is
scarcely reasonable to hope for good returns, however satisfactory the
*oil may prove in other respects.

No,
1.

2.

Field Cornetcy.
Mount Frere.

MOUNT FRERE

(Privately collected.)

Farm or place.
Amanzamyama.

Collector.
Distr. Forest Officer

The first of these two samples was a surface soil ; the second was taken
one foot below the surface : they were procured through the Conservator
of Forests, Umtata, and had been collected in the Amanzamyana Planta-
tion, about eight miles west of the village of Mount Frere, and adjoining
Zibokwana's location. They are representative of extensive tracts of
country in which the cereal crops usually grown do not thrive, whereas
arboriculture is fairly satisfactory. It is not surprising that shallow
rooted crops fail, for the soil is exceedingly deficient in both lime and
potash ; there is, however, a moderate store of phosphoric oxide available,
according to the analyses, which resulted as shown below : —

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 inm. Sieve. through £ min, Sieve.

Sample. Phos-

Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide.

No.

1.

2.

99-2
97-4

7-13
17-67

7-20
14-71

•0049
•0057

•287
•168

•012
•004

•051
•019

•091
•065

Field Cornetcy.
Port Nolloth.

NAMAQUALAND,

(Privately collected.)

Farm or place.

Port Nolloth.

Collector.
A. G. Hey^ood.

No.
1.

2-
»>

)» )> y>

These were three drift sands collected in the vicinity of Port Nolloth.
The first two were surface soils, taken, respectively, from the south and
from the north drift. The third was taken one foot below the surface.
Lime, as may be expected in this case, is plentiful, but in respect of other
mineral plant food the soils are poor.

determinations made resulted as follows: —

(Method II.)

Percentage of Soil sifted through Percentage of Soil sifted
1 mm. Sieve. through 3 mm. Sieve.

Phos-

'No.

Water. Organic

Chlorine. Nitrogen. Lime.

Potash.

phoric

matter.

oxide.

1.

— — '80

•063

•015

0

fim

2-07

•036

•100

3.

— —

— 1-445

•029

•014

* See page 17.

OUDTSHOORN.

(Officially collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Low*r Olifants River. Vlakte Plaats, C. F. Jurita.

L>. „ Riet Vallei.

3. ,, Hazenjacht. „

4. Grobbelaars River. Schiedam. „

5. „ Oudtshoorn. „

6. Wynands River. Gamtoos Berg. „

7. „ Blauwbosch Kuil.

8. „ Welbedaoht.

9. Attaquas Kloof. Nooitgedaeht.

10. „ Hottentots Laagte.

11. „ Saffraan Rivier.

12. Garnka Flats. Jan Fouries Kraal.

13. „ Welgerust.

14. „ Adjoining Buffels Vlei.

15. Caugo West. Lemrick.

16. „ Welgevonden. „

17. „ Matjes Rivier. „

As one passes from the Uniondale Division to that of Oudtshooru,
through Roode Krans Poort (an immediate change in the character of the
soil becomes noticeable. A closer resemblance to the Karroo soil than
on the eastern side of the " poort " begins to be traceable. The rheno*-
ter bush, which is much in evidence in the north-western portion of the
Uniondale Division, disappears completely, and the Ganna bush of the
Karroo is frequently to be seen. It is generally believed that where the
rhenoster bush flourishes lime is deficient in the soil, while the Ganna, on
the other hand, indicates a lime soil: as will appear from the sequel, the
analyses of the Uniondale and Oudtshoorn soils tend to give confirmation
to this popular idea. Of such a soil, where the Ganna bush grows, a sam-
ple, No. 1 in the above list, was taken on the farm Vlakte Plaats. Fur-
ther west, on the adjoining farm Kruis Rivier, a change in the character
of the flora would seem to be indicative of a corresponding change in the
nature of the soil ; the ganna thins out, in fact dies away practically alto-
gether, and the rhenoster bush reappears ; but as one approaches the town
of Oudtshoorn the soil again becomes more akin to that of the Karroo,
and sandier than is the case further east. Of this last class of soil, two>
ajpmples, Nos. 3 and 4, were taken at Hazenjacht, 13 miles from Oudta-
hoorn (this soil, by the way, proves chemically rather poor — for an Oudts-
hoorn soil), and Schiedam, two miles from Oudtshoorn, respectively,
while one collected at Riet Vallei, No. 2 (14£ miles W.S.W. of Vlakte
Plaats, and 4 miles north-east of Hazenjacht), was more of the nature of an
alluvial or " vlei " soil.

Underlying the surface soil throughout the Olifants River valley, ato
timea appearing on the surface, but generally within from four to eighteen
inches thereof, is a very stiff red clay, here and there showing bands of
limestone. This clay has the local reputation of being the cause of the
great fertility of the Division,* and a sample of it was taken (No. 5) from
Mr. E. T. L. Edmeades' property, just outside the town of Oudtehoorn.
From the level tract of country south-west of Oudtshoorn, sample No. 7,

* It is, of course, river alluvium, originally brought down from the Karoo.

ft

/

/

'..,

* r/

')••-"

/

/

/ ^

••;,

^ / <0

\_

/ ^

/

* »/: ?

/

/- ^

r' : *

/

\ ^

/ <0

*• u

/ ^

V

/ ^

c\

/ ^

u

•\

-v-

u ^r

'\ /
-\*\

i fc£

M ^:

0 «|. |I

J c \

^b"

' <5 '

V\

(J

Z "'! (

<x \

oo Jfi\

;-° \

\

i§ _^^-f

- \* !

K '

\v-

«*1

A
* \

8

°l

^

id

0. oj

S

4

2

r

'••. ..-.

1

*'••,

97

a fine loose clay soil, was taken on the farm Blauwboschkuil, at a dis-
tance of five miles from the town. About 6| miles further on in the same
direction, on the farm Nooitgedacht, a similar sample, No. 9, was col-
lected; that taken at Saffraan River, No. 11, 10 miles from Nooitgedaicht,
was sandier, — being largely derived from the sandstone of the Attaqua
Mountains, — and, as one may expect under such circumstances, a rather
poor soil. There are abundant evidences of limestone all over the extent
of country traversed between Oudtshoorn and Saffraan River : No. 10,
taken six miles E.N.E. of Saffraan River, on the way back to Oudtshoorn,
is a very shallow, fine, clay soil, similar in character to those just alluded
to; No. 6 being another similar soil of greater depth, collected at a dis-
tance of about 11 miles south of Oudtshoorn. These two soils, Nos. 6
and 10, were taken from hill sides. From the extent of country stretch-
ing northwards from the town of Oudtshoorn to Cango no samples wera
taken. The river bed is here practically the only cultivated portion, and
every available part of that is under actual cultivation, so that virgin soil
was unobtainable; as for the "veld," employing the term in a sense in-
clusive of hill slopes, — except close to the town, in the vicinity of which
samples had already been secured, it is very stony, and covered with
prickly pear, while farther north the hills are so steep and rocky as to
debar cultivation, except along the river side.

Along the Olifants River, west of Oudtshoorn, two samples were
taken, Nos. 8 and 12, light sandy clays, the former from the farm Welbe-
dacht, 11 J miles from Oudtshoorn, the latter at Jan Fouries Kraal, 21
miles from Oudtshoorn, and 11 J miles south-east of Calitzdorp. Sample
13, taken from the farm Welgerust, is representative of the area known
as the Gamka Flats, which extend south of the village of Calitzdorp ; the
predominating feature in the vegetation of this area is what is generally
known to KarrO'- farmers as the Kouwgoed (Aiigea capensis). On the
higher level, north of this region, and subjacent to the low mountain
range which extends eastwards to the north of Calitzdorp, the spekboom
(Portulacaria afra) grows luxuriantly over a space stretching about four
miles east. Just beyond this area sample No. 14 was collected. A red
sand soil, No. 15, was taken in Coetzee's Poorc, at the base of the con-
glomerate hills flanking the southern portion of that pass. Further north
the Karroo bushes disappear completely, and the rhenoster covers the
entire veld; the loose limestone too, so abundantly scattered about the
ground near Calitzdorp, is no1 longer to be seen, but instead there are
numerous fragments of quartz: naturally a corresponding alteration in
the character of the soil, both physically and chemically, is to be ex-
pected. Of the new typo of soil, a sample, No. 16, was taken on the farm
Welgevonden, near the top of a small hill north-east of the homestead.
Further east, on the farms Voorbedacht and Matjes River, the limestone
reappears in abundance, and to the change in the nature of the soil which
this involves is doubtless due the reputed suitability of the latter for
tobacco culture. A sample of this soil, No. 17, was taken from a gently
sloping hillside, south of, and overlooking, the cultivated lands on that
farm.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

18. Lower Olifants River. Vlakte Plaats. J. W. Louw.

On the farm Vlakte Plaats, already mentioned as the place where
sample No. 1 had been collected, two additional samples, Nos. 18 and 19,
were procured through the agency of Mr. J. W. Louw. No. 18 was taken

G

98

from the valleys amongst the mountains, and it is therefore probable that
disintegrated Table Mountain sandstone enters largely into its composi-
tion : as a virgin soil it is declared to be very rich, and capable of pn>
ducing magnificent tubers, but after some years of cultivation there is a
marked decrease in its productiveness. No. 19 is a virgin Karroo type of
soil, very good for most crops excepting potatoes.

The analyses of the above nineteen soils will be found tabulated
below : —

(Method I.)

No.

1.
18.
19.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.

Percent, of

Field

Sample.

Fine earth.

Percentage of Soil sifted through
1 mm. Sieve.

Water.

Percentage of Soil sifted
through J mm. Sieve.

Phos-

Organic Chlorine. Nitrogen. Lime. Potash. phoric
matter. oxide.

89-5

2-06

6-25

•0035

77-5

•97

2-27

•0071

96-1

2-26

7-82

•0170

56-0

•65

1-85

•0035

65-0

•33

•85

•0134

83-7

1-10

2-65

•0035

69-9

3-73

4'33

•0113

78-7

1-21

4-34

•0612

91-5

1-97

4-31

•0272

91-0

•87

2-50

•0191

93-2

1-88

4-53

•3919

68-6

2-39

6-41

•0413

76-3

•40

1-89

•0212

95-9

•84

2-36

•0081

91-1

1-57

2-65

•0026

81-9

•44

1-64

•0216

58'6

1-11

3-68

•0046

76-0

T90

5-10

•0035

61-6

3-15

17-14*

•0092

•168

1-170

•084

•104

•252

•828

•084

•154

•070

•056

•098

•170

•070

•100

•140

•098

•112

•530

•084

•606

•098

•958

•182

•206

•112

•074

•112

•172

•077

•170

•091

•046

•126

•068

•182

•084

•350

7-460

•299
•080
•211
•100
•071
•096
•149
•165
•211
•144
•141
•201
•035
•161
•170
•098
•194
•144
•087

•150
•037
•101
•074
•054
•049
•058
•123
•134
•124
•139
•119
•040
•063
•063
•049
•123
•067
•166

One cause of the superiority of the Oudtshoorn soils to those of the
surrounding Divisions stands revealed, immediately upon comparison of
the chemical results : in the former, lime, potash, and phosphoric oxide
are all present in much larger quantities than in the soils of other areas.
The fertility of the soil does not terminate with the boundaries of the
Oudtshoorn Division, but is also clearly traceable in the soils of the Buffels
Klip Fiedd-cornetcy, in the neighbouring division of Uniondale. The
transition, nevertheless, becomes very striking when passing from the lat-
ter division, the moment that the Oudtshoorn boundary is crossed.

It is, of course, lime that is chiefly abundant in the Oudtshoorn area,
and gives their distinctive character to the alluvial calcareous loams of
this division and the adjacent division of Ladismith, but the fertility is
not- to be ascribed to the lime alone, for both potash and phosphoric oxide
are present in satisfactory amount. This is due to the fact that the soils
of the Oudtshoorn basin are largely the result of ages of deposition of river
silts brought down from the Karroo by the Ganika and Olifants Rivers,
and are therefore made up of the finely divided products of decomposition
of the rocks of the Karroo system : thus they axe, both physically and
•chemically, excellently adapted for the agricultural purposes that they are
known to serve so well. Strange as it may appear, nitrogenous material
is rather less prominent in this fruitful district than, for instance, in the

This includes carbon dioxide combined as calcium carbonate.

neighbouring divisions of George, Knysna, and Uniondale, — a clear evi-
dence that the soil can do with less nitrogen than, say, the Knysna soils
provide, if only the deficiencies in other respects be made up. A remark
of Prof. Hilgard may here be quoted as bearing directly on this point;
he says : * (and the italics are in the original) : — " In general, we find that
lower percentages of potash, phosphoric acid, and nitrogen are adequate,
when a large proportion of lime carbonate is present."

The stiff red clay about the town of Oudtshoorn has potash as its dis-
tinguishing characteristic : possibly the virtue of this clay lies herein, that
on mixture with surface soils already containing lime, and perhaps phos-
phates, in adequate amount the addition of potash fulfils the one need.
Samples 7 and 9, which contain good proportions of all the elements of
fertility, exemplify the causes of productiveness in the soils of this
division.

The soils taken in the north-western part of the division are appa-
rently less fertile, lying, as they do, under the shadow of the great saaid-
stone range of the Zwartbergs, and in respect of both lime and phosphoric
oxide they show a diminution; but on approaching the Zwartberg Pass,
near which, it must be remembered, the limestone caves of the Cango dis-
trict are situated, the amount of lime in the soil naturally increases enor-
mously.

On the whole it may be said that the soils of the Oudtshoorn Divi-
sion show good quantities of lime and nitrogen, and are also satisfactory
as regards potash and phosphates.

PAARL.

No.

Field Cornetcy.

1.

French Hoek.

2.

n

3.

5J

4.

JJ

5.

}J

€.

a

7.

8.

Groot Drakenstein.

9.

10.

11.

Achter de Paarl.

12.

13.

14.

55

15.

16.

Paardeberg.

17.

18.

19

North Paarl.

20.

Wellington.

21.

22.

tt

23.

24.

25.

Wagenmakers Vallei.

(Officially collected.)

Farm or place.
Keerweder.

La Dauphine.

Cabriere.

La terre de Luque.

Zand Drift.

Le Plaisir Merle.

Bossen Dal.

Babylons Toren.

Matjes Kuil.

Groenfontein.

Kuilenberg.

Brandenburg.
Lange Rug.
Knolle Vallei.

Vrede Hof .
Blauwvallei.

Welbedacht.

Hexenberg.

Collector.
St. C. O. Sinclair.

Hilgard : ;i Soils : their formation, properties, composition, and relations to climate
and plant growth." p. 354.

100

No. Field Cornetcy. Farm or place. Collector.

26. Wagenmakers Vallei. Hexenberg. Sb. C. O. Sinclair.

27. „ Drie Fonteinen. „

28. „ Vruchtbaar. C. F. Juritz.
29.

30. „ Groenberg. „

31.

32.

33.

34. „ Openhorst. ,,

35. „ Groenfontein. ,,

36. Groene Berg. Burgers Drift. St. C. O. Sinclair.

37. „ Dryvers Vallei.

38. „ Lang Hoogte. „

Seven samples were collected in the French Hoek Field-cornetcy.
The soil of this region falls into two classes — the first is what is locally
known as the " grauw grond," from the hillsides, — a yellow soil, consist-
ting of clay with quartz pebbles and stones; and, secondly, a. black allu-
vial soil from the lowest parts of the valley. It had been expected that
the composition of the soils would be found to vary fairly considerably,
owing to the diverse nature of the rocks occurring in this district; gene-
rally, however, the soil may be regarded as the product of rocks of the
Malmesbury series, Table Mountain sandstone, and granite. The Field-
cornetcy is practically encircled by sandstone mountains, a condition which
one finds reflected in the poor proportions of plaint food in its soils.*

Nos. 1 and 6 are hillside soils; Nos. 2, 3, 4, 5, and 7 are alluvial soils
from the valley. No. 1 was taken from Mr. G. Kriel's farm adjoining
Keerweder: the soil, which is very poor, is gravelly in texture. No. 2
represents a black alluvial soil taken from the same farm. The land typi-
fied by the sample is planted with vines. No. 3, a similar soil, is repre-
sentative of that of the farm La. Dauphine. Sample No. 4, taken on the
farm Cabriere, is similar in character to Nos. 2 and 3, but the effect of the
decomposed granite is seen in the larger proportion of potash. f No. 5
was a black soil overlying potclay on the fairm La Terre de Luque. No.
6 represents a mixture of surface, and subsoil — such as is obtained when
the ground is put under cultivation — on La Terre de Luque. The subsoil
is a yellow gravel containing much clay and pebbles. Below the depth to
which the sample was taken potclay is found. The surface soil in this
part is a clay. The soil appears to have a good water capacity, and the
vegetation thereon is reported to withstaud the effects of drought to a
greater extent than is the case with the "grauw grond," which is the
ordinary hillside soil. Of this " grauw grond " No. 1, as already stated,
may be taken as representative. No. 7 represents a black soil lying close
to the river on the farm Zand Drift, and may be regarded as typical of
the lands along the river.

The remarks made in connection with the origin of the soil in the
French Hoek Field-cornetcy apply also largely to the Field-cornetcy of
Groot Drakenstein : sandstone, however, does not predominate to the same
extent, its place being taken by granite. No. 8 represents a black valley
soil from the farm Le Plaisir Merle. The soils along the Dwars River are
represented by No. 9 from the farm Bossen Dal. No. 10 represents! a hill
side soil, which was collected on the farm Babylons Toren.

*It will be remembered that in connection with the Caledon Division soils a similar

condition was observed to exist. (See p. 38).
t Compare also No. 9 of the Mossel Bay soils, page 93.

101

In the Achter de Paarl Field-cornetcy five samples were taken. No.
11, from Matjes Kuil, is a sandy soil containing a little clay, and over-
lying a gravel. When manured with stable and kraal manure, wheat, rye,
and oats appear to thrive well. On going westwards the soil becomes

Si 1N *

OAL JC&SAPHAT ^

T0i/e Mfn.
3*r

Mfn. \

,'es i

5° KLEIN DRAKENSTCIN \

^ v ^K, N-, ,

• ' ' " *> -iXVar^ -"•• >

DIVISIONAL MAP OF

PAARL

^ \

o

i»

\

\

'"••••-•• "/

-' #»'

\ '%* .., ,>
'^{.^.

if

7. M.S.

/

"HQEK / o
<,

more and more sandy. Nos. 13 and 14 furnish examples respectively of
the " hillock " and low-lying soils already mentioned in connection with
the Malmesbury Division. The occurrence of these little hillocks — " heu-
veltjes," as they are vernacularly termed — has been remarked, not only in

102

the Malmesbury and Paarl Divisions, but also iu other parts of the South-
Western Districts of the Colony. The soil of such hillocks has usually
been found to be richer than the soil round about. Whilst the rhenoster-
bosch, a bush moire or less typical of rather poor soils of the Cape Penin-
sula,* is found all around these hillocks, on the hillocks themselves we
find the " distel doom," which generally seeks the richer soil. No. 15,
from Brandenburg, represents the soil north of Kuilenberg. When pro-
perly treated it appears, from all accounts, to yield very satisfactory
results. " Laoiiziekte " is unknown on the farm, and on the pasturage
goats thrive well. This, it was thought, would seem to be evidence of the
presence of lime or phosphates in the soil in sufficient quantity, and,
curiously enough, the proportion of phosphates found — higher than in al-
most any of the other soils from this division — confirmed anticipations in
this respect. The geological formation of this area is that of the Malmes-
bury series.

In the area represented by the Paardeberg Field Cornetcy, a similar
geological formation prevails. Three samples were taken as typical of
the soils of this portion of the Paarl Division; of these, Nos. 17 and 18
represent the hillock and lower soil respectively at Knolle Vallei. It
has been found that barley grown on these hillocks withstands the
attacks of rust to a much greater extent than the barley on the sur-
rounding soil. No. 18 is somewhat more sandy, and appears generally
poorer than No*. 17.

In the JFiald Cornetcy North Paarl, sample No. 19, taken on the
farm Vrede Hof, is a type of the soil at the foot of the Paarl Mountain.
Here the intrusive granite, which constitutes the Paarl Mountain, will,
if sufficiently decomposed, bring about a change in the composition of the
soil.

No. 20 is a type of hillside soil from the farm of Mr. Stucki, part of Kee-
bok Kloof at Blauwvallei, in the Wellington Field Cornetcy, No-. 21, taken
from underneath the oaks on the farm, resembles in colour the black soil
known in the district as " turf." It is, however, a coarser soil, and contains
more lime than any other of the samples taken in the same district. This
soil was designated by Mr. Stucki as " vet heuvel grond." No. 22 re-
presents what is known locally as " turf " or " zwart grond " ; the sample
was taken from the side of a poplar plantation on the farm Welbedacht,
belonging to Mr. W. P. van der Merwe. It was somewhat mixed with
sand. No. 23 was a sample of what is called " doode turf " from the vine-
yard adjoining the homestead on the same farm. This " doode turf,"
which exists in patches about the vineyard, appears to be less fertile than
the soil around it. This must evidently be ascribed to physical causes,
rather than to difference of chemical composition. The soil is stiff and
compact, and more clayey than that represented by the previous sample.
It would naturally, therefore, hold more water, and so become cold
in spite of its warm black colour. It will be noticed that the sample
contained more than three times as much water as No. 22. There
would seem to be some similairity here to the cause of the trouble in the
vineyards of the Graaff-Reinet Division, already alluded to.f No. 24
was a sample of stiff black soil from the edge of a plot of ground which,
with suitable manipulation (such as an admixture of gravel from the hill-
sides, etc.) and manure, had borne sweet potatoes for upwards of forty
years. It was found to overlie a subsoil of gravel. The soil represented
by this sample is locally known as " Blauw pot clay."

* See pages 56, 68, and 96.
f See page 61.

103

No. 25 was a sample of clayey soil, taken from the farm Hexenberg,
in the Wagenmakers Vallei Field Cornetcy; a small portion of the sur-
rounding land was at the time being used as a brickfield, but it was the
intention to put it all under cultivation. The sample was stated to be
similar to soil found on the contiguous portion of the adjoining farm of
Mr. G. J. Hugo. No. 26 represents a stiff black soil a short distance in
front of the homestead on the same farm.

Going northwards from Wellington, No, 27 was taken from the farm
Drie Fontein. This soil, which is described as a cool soil, overlies a clay
subsoil. Vines do well on it, and do not appear to languish as when grown
on black " turf " (alluvial) soil. No'. 28 was a sample of hillside soil,
planted with an apricot orchard of 2,000 trees, on the farm Vruehtbaar,
owned by Mr. C. P. Gillie. This sample was reported to be typical of
the soil in the N.N.E. portion of the Wagenmakers Vallei Field Cornetcy.
No. 29 was taken from the edge of the orchard near the homestead of
Vruehtbaar. It consists of the fine clay washed out of the soil represented
by No. 31. From a chemical point of view, it is, on the whole, the best
of the soils collected in this district. It consists almost entirely of clay,
the orgai :. in alter present being due to incidental dropping of leaves,
etc. As a set off against the chemical qualities of this soil, it must be
mentioned that, owing to its clayey nature, it is apt to retain large quan-
tities of water, which tend to render the soil cold ; and the light colour
will have a similar tendency.

No. 30 was a sample of soil prevailing on the farm Groenberg, also
belonging to Mr. Cillie; the sample was taken from the side of a water-
course on the hill slope. No. 31 was taken from the bottom of the same
water course. The soil represented, when treated with stable manure,
was said to answer for vines, but not for fruit trees. This fact is pro-
bably accounted for by the deficiency of lime, shown by the chemical
analysis, not being corrected by the manure applied. Soil similar to the
sample was said to occur in fairly large quantities throughout the district.
No. 32, also a sample of the soil found on the farm Groenberg, was
taken from the vineyard some distance above the spot where the preceding
sample was collected. No. 33 was taken from one of the little hillocks
locally called " heuveltjes," occurring in the same vineyard as No. 32.
These hillocks, of which many are to be found in the Wellington district,
have, as has already been stated, the local reputation of being more fertile
than the surrounding ground. On comparing the results of the analyses
of these two samples, it will be seen that No. 33 is considerably the
richer hi lime and potash, whilst the amount of phosphoric oxide i& prac-
tically the same in both cases. The amount of nitrogen and of organic
matter in No. 33 is also greater than in No. 32. The texture of the two
samples, as far as shown by the mechanical analyses, is, for all practical
purposes, the same, No. 33 being very slightly coarser.

Proceeding from the Krornme Rivier towards Blauwvallei, the soil
is said to become more and more clayey ; this is borne out by the mechani-
cal analyses.

No. 34 was a sample from a part of the farm Openhorst that had
been last sown with oats. No, 35 represents a somewhat similar soil
on the farm Groenfontein ; in this case too the portion of the farm whence
the sample was taken had been sown with oats.

The soil of Burgers Drift in the Groene Berg Field Cornetcy was
sampled in No. 36, taken from a portion of the farm occupied by Mr.
Malan. In dry seasons the plants on this soil are parched. No. 37 was
taken from Mr. J. Gillie's farm Dryvers Vallei, and represents his corn
lands. No. 38 was a sample of what is locally known as " klippertjes
grond," taken on the farm Lang Hoogte. The sample is a type of the

104

poorest lands on the farm. The oats grown in the districts, from which
samples 20 to 38 were collected, had suffered severely from rust. Wheat
had not yet been appreciably affected, although signs were not wanting
to show that the disease had made its appearance. Barley had not been
attacked.

No. Field Cornetcy.

39. Groot Drakenstein.

40.

41.

42.

43.

44.

45.

46.

47. Klein Drakenstein.

48.

49.

(Privately collected.)

Farm or place.
Donkerhoek.

Collector.
R. D. Koch.

D. G. Cillie.
C. Mayer.

50. Wellington.

51. Wagenmakers Vallei.
OA. ,,

53. Groeneberg.

54.

55.

56.

Reeboks Kloof.
Vruchtbaar.
Twyfeling.
Kilkerran.

J. W. Mason.

Eight samples of soil were collected on the farm Donkerhoek from
the orchards of the Donkerhoek Fruit Syndicate. Of these, Nos. 39, 40
and 41 were collected previous to the orchard being manured. Nos. 39
and 40 were clayey soils, and No. 41 a sand. Nos. 42, 43, and 44 repre-
sent the same soils after manuring and five years of cultivation. Nos.
45 and 46 were respectively a surface soil and a subsoil, also collected
on the farm Donkerhoek, and taken to a depth of thirty inches; these
two are typical of the veld soil which constitutes two-thirds of the dis-
trict. They both represent cool gravelly sands, with a substratum of
ironstone. The local opinion was that they were decidedly poor; in fact,
farmers in former years used to consider this type of soil quite unfit for
cultivation; since then, however, vines have been planted there. It
may be said that agriculturists find the best soils of the district to lie
along the river, and in the mountain kloofs ; these soils are of limited
extent.

On the eastern bank of the Berg River, in the Klein Drakenstein
Field Cornetcy, opposite Simondium, a mixed sample of sandy soil, No.
47, was collected. This sample was taken from the surface to a depth
of 10 to 15 inches, at four different spots, and a mixture made. The
soil in the vicinity is composed of river drift sand to a depth of four
feet; below this lies a good black loamy subsoil. The surface soil is
naturally poor, but does not represent all the soil in the neighbourhood —
least of all Simondium, where, according to local opinion, the soil is as
good as any in the Western Province for vines and fruit trees.

105

Nos. 48 to 52 were samples of vineyard soil. Of these No. 51 had
not been manured for four years; the lands whence Nos. 50 and 52 were
taken had both received " kraal manure " a year before the collection of
the samples.

Four samples — Nos. 53 to 56 — were taken at Kilkerran farm, Her-
mon. No. 53 was a dark soil from a " heuvel " or hillock, on land ad^
jacent to the Berg River, 20 to 30 feet above the river bed, and about
three miles north of Hermon; Mr. Mason's experience was that these
hillocks, which consist of dark rich looking soil, are exceedingly fertile.
No. 54 was a shade lighter in colour, and was collected from the level
ground in the immediate vicinity of the preceding sample. No. 55 repre>-
sents a red siibsoil from the same locality, and No. 56 was taken from
the grain lands, at some distance from the river. The soil by the river-
side is a clayey loam, with a red subsoil of great depth; the rhenoster
bush grows luxuriantly on this land if left to itself; it ploughs freely,
when sufficiently moist, but the surface sets hard after rain, from the
amount of silt and clay present. Both this soil, as well as the grain
soil from which No. 56 was collected, vary considerably within short
distance in regard to their mechanical condition; the latter is very shal-
low in parts, with a subsoil ranging from shale to yellow clay. Mr.
Mason declared that these soils were all found to* require manuring in
order to ensure productiveness, and in this respect his experience quite
confirms the results of the chemical analyses.

The analytical results are shown in the following tables: —

(Method I.)

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
42.
43.
44.
45.
46.
47.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

Percent of

Percentage of £

Field

1 me

Sample.

Fine earth.

Water.

Orjranic

matter.

77-9

2-16

7'57

70-4

1'40

4-27

57-8

•81

2-55

84'5

T89

4-80

65-1

•97

2-78

56-9

1-56

4-77

86-3

1-41

3-85

68-4

•99

3-45

95-1

3-79

9-96

59-8

•31

1-17

70-3

2-19

6-32

80-3

2-22

6-01

36-9

•61

2-52

Soil sifted
mm. Sieve.

through

Chlorine. Nitrogen.

59-1
63'9
69-3
51'0
81-2
53-6
55-1
81-7
65-3
79-6
68-7
61-5
37-0

•41

•29

•67

1-13

1-72

T30

1-11

1-72

•32

•87

•41

2-06
T29
2-65
5'23
6-41
6-59
5-22
7-39
1-77
3-69
4-02

•0092
•0102
•0081
•0265
•0067
•0106
•0127
•0035
•0184
•0092
•0057
•0240
•0046

•005

•0071

•0212

•0127

•0212

•0081

•0109

•0170

•0149

•0060

•0049

•146
•116
•088
•116
•081
•088
•116
•116
•174
•029
•102
•059
•088

•070
•038
•095
•116
•160
•116
•088
•153
•052
•066
•294

Pcrcentaj

£e of Soil

sifted

throng

& i mm.

Sieve.

Phos-

Lime.

Potash.

phoric

oxide.

•010

•039

•042

•042

•087

•042

•008

•013

•038

•016

•124

•037

•008

•013

•022

•006

•015

•017

•058

•019

•038

•020

•021

•073

•008

•033

•098

•040

•014

•020

•096

•026

•082

•048

•019

•036

•184

•040

•064

•020

•023

•019

•010

•017

•047

•060

•013

•115

•024

•010

•017

•012

•029

•027

•072

•045

•061

•100

•073

•055

•014

•091

•115

•034

•041

•046

•098

•044

•056

•012

•013

•010

•034

•014

•055

•063

•072

•059

106

No.

21.
22.
23.
24.
25.
26.
50.
27.
28.
29.
51.
52.
30.
31.
32.
33.
34.
35.
36.
37.
38.
53.
54.
55.
56.

Percent, of
Field
Sample.
Fine earth.

Percentage of Soil silted
1 mm. Sieve.

Water. Organic Chlorine,
matter.

through
Nitrogen

Percentage of Soil sifted:
through J mm. Sieve.
Phos-
Lime. Potash, phoric-
oxide.

57-2

1-24

7

•33

0059

•025

•380

•056

•097

69-0

•

60

4

•23

006

•017

•053

•056

•oo&

88-6

2

•

20

12

•29

0064

•003

•024

•064

•159

99-3

1

79

12

•15

0106

•Oil

•138

•075

•072

77-8

1

16

5

•93

0141

•on

•027

•064

•102

70-2

58

3

•69

0081

•045

•099

•056

•077

77-0

1

80

5

•77

0049

•108

•016

•043

•051

50'6

•

52

1

•23

0216

•073

•022

•048

•019'

46-8

•

51

3

•52

0085

•025

•026

•032

•051

95-0

•

83

8

•46

0078

•003

•096

•194

•082

90-4

1

92

5

•22

0049

•115

•138

•152

•070

45-3

2

92

7

•49 '0049

•101

•038

•446

•107

47-6

•

46

5

•07

0085

•006

•036

•354

•036

21-0

•

36

4

•17

0103

•009

•020

•203

•095

66-2

•

76

5

•11

006

•003

•037

•084

•077

65-6

•

76

5

•57

0255

•006

•064

•171

•074

8-6

39

3

•93

0042

•Oil

•028

•042

•125

19-5

•

28

2-41

0049

•006

•103

•059

•067

60-3

1

•

20

4

•07

1620

•073

•026

•075

•054

51-8

1

48

4.

•87

0088

•139

•014

•039

•055

55-8

•

63

2

•33

0071

•059

•010

•042

•027

92-4

•

79

2

•73

0106

•059

•020

•035

•029

91-6

•

57

2

•11

0120

•043

•016

•033

•027

90-3

•

44

1

•68

0145

•036

•010

•026

•038-

41-3

1

•

12

3

•65

0106

•087

•018

•047

•038

(.Method II

J

Percentage
Water.

of Soil sifted through 1 mm.
Sieve.

Organic Chlorine. Nitrogen,
ma.ter.

Percentage of Soil sifted
through 3 mm. Sieve.
Phos-
Lime. Potash. phoric
oxide.

T61

5-03

•0031

168

•201

•032

•041

1-56

4-16

•0032

140

•108

•023

•010-

•49

1-48

•0012

028

•059

•0095

•009

•52

2-96

•014

—

•143

•182

•006

•17

1-01

•17

trace

trace

No.

39.
40.
41.

48.
49.

On the whole it may bo said that the Paarl soils turn out rather poor
in lime, potash, and phosphoric oxide, while the average amount of nitro-
gen that they contain is about normal. In respect of potash, some parts
of the Wagenmakers Vallei Field Cornetcy show up best, and, next to
that, part of the district lying to the north-west, especially the Achter
de Paarl and Paardeberg Field C'ornetcies; at French Hoek, however, the-
proportions of potash in the soil are very mea,gre. The largest per-
centages of nitrogen are contained in the soils of South-western Paarl, that
is to say, in the Field Cbrnetcies of French Hoek, Groot Drakenstein, and
Achter de Paarl ; in the rest of the division nitrogen is present in the
soil in moderate amount.

107
PIQUETBERG.

(Privately collected.)
No. Field Cornetcy. Farm or place. Collector.

1. Piquetberg. Do Ttiin. G. H. Dunn,

2. „ Moutons Vlei. Dr. Simon.

A sample of soil where vines and fruit trees nourish was collected on
the farm De Turn, once part of the farm Rietfontein, now commonly
known as De Hoek, and adjoining Wittewater. The soil in this neighbour-
hood varies considerably, but the sample under notice is typical of the
slopes immediately below the mountain; it is of considerable depth, in
fact sluits nearly twenty feet deep show no change all the way down. This
soil differs from that on the lower levels, such as the Piquetberg Karroo,
which, however, outwardly resembles it, but differs from it in being some-
what gravelly, containing ironstone pebbles.

No. 2 is a type of the soil used for the cultivation of tobacco on the-
farm Moutons Vlei, one of the tobacco farms of the north-western
districts.

The analyses of these two soils are given below : —

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field through 1 mm. Sieve. through % mm. Sieve.

Sample. Phoa-

No. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide

1. 72-4 1-47 3-80 '008 '140 -064 '103 '068

(Method II.)

Percentage of Soil sifted througn 1 mm. Percentage of Soil sifted

Sieve. through 3 mm. Sieve.

Phos-

No. Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide.

2. '72 7-49 -014 '40 '11 '061

PORT ST, JOHN'S.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Isinuka. H. H. Davison,

A sample of alluvial soil — excluding that upon the immediate sur-
face—was collected upon the banks of the St. John's River; the soil at
this spot had been used for mealies, and it was intended to apply it to
the cultivation of tobacco.

The analysis of this sample resulted as shown below : — *

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 mm. Sieve. through £ mm. Sieve.

Sample. Phos-

No. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide.

1. 97-2 2-73 16-5 '015 '24 '52 '171 '137

* Results of a partial mechanical analysis of this soil will be found under the head of
" Physical composition of soils."

108

This soil apparently contains a good quantity of lime and a satisfac-
tory proportion of potash and phosphates.

No. Field Cornetcy.
1. Stuurman.
2.
3.
4.

PRIESKA.

(Officially collected.)

Farm or place.
Zeekoebaard.
Stof kraal.
Keuken Draai.

Collector.
J. G. Hose.

No. 1 was a sample of the silt carried down by the Orange River
•while in flood, and deposited by the river when receding within normal
limits after overflowing its banks. The sample was collected from the
river bed, near the Buchuberg Irrigation works, at the western boundary
of the Government farm Zeekoebaard. The river was very low at the
time, and the silt was only slightly moist. It subsequently dried as hard
as a brick, without being exposed either to solar or artificial heat.
Wherever the flow of the river is sufficiently slow, this mud is deposited,
sometimes to the depth of many feet. Local farmers consider it to be
extremely fertile, but, owing to its hard compact condition when dry, it ia
capable of employment for agricultural purposes only when mixed with sand
or loose soil. If a water-furrow be made of stones closely packed, and the
interstices of the latter be filled with this silt, the furrow is said to become
perfectly watertight, so that the adhesive properties of the silt are
considerable.

Large tracts of country are covered by the blown sands of the Kala-
hari, and along the river banks a continual intermingling of the alter-
nating blown sands and river silts is in progress. Of these mixtures Nos.
2, 3, and 4 are samples; they were taken from the centre of each of the
large tracts of irrigable land, No. 2 from the farm Stofkraal, 11 J miles
down the river from the irrigation works, Nos. 3 and 4 from the farm
Keuken Draai, at distances respectively 14 and 16 miles from the works.

The results of the chemical analyses* of these samples are as follows:

(Method I.)

No.

Percent, of Percentage of Soil sifted through
Field 1 mm. Sieve.
Sample.
Fine earth. Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through £ mm. Sieve.
Phos-
Lime. Potash, phoric
oxide.

1.

100 5-99

14-81

•004

•099

1-444

•473

•221

2.

99-8 2-83

1-89

•0057

•056

1-055

•056

•081

3.

99-9 4-62

5-05

•0042

•196

1-344

•164

•141

4.

90-3 3-94

2-85

•0071

•168

1-224

•115

•076

Mechanical analyses of these and other samples from the same localities will be dealt
with under the head of " Physical composition of soils."

109

The specific gravity of the silt represented by No. 1 is 2 '03 on the
basis of the dry specimen as above described, so that a cubic foot would
weigh about 127 pounds. One acre of land, covered with the silt to a,
depth of half an inch, and thus receiving a deposit of 1,815 cubic feet, or
115 Cape tons, of the silt, would therefore be enriched to the extent of:

3,314 pounds of lime,
1,086 pounds of potash,
507 pounds of phosphoric oxide.

In round figures, each such acre would receive about one ton and a
half of lime, half a ton of potash, and a quarter of a ton of phosphoric
oxide. The commercial value of the fertilising constituents, in an avail-
able form, added to each acre of land upon which the silt is deposited
to half an inch in depth is at least £30; the pecuniary value of these
silt deposits, therefore, bears comparison, and very favourable comparison,
with the cost of manuring in the ordinary way.

Many other instances may be quoted of lands having acquired fer-
tility by the accretion of similar river-borne silts. It is, for example,
largely due to the fact, already commented on," that its soil has been
built up of old alluvial deposits that the Oudtshoorn Division is so fertile ;
the rich silts and clays brought down from the Karroo by the Olifanta
and Gamka Rivers have, in great part, contributed to this. It is also
common knowledge that these rivers occasionally overflow their banks on
their seaward course through the Mossel Bay and Riversdale Divisions, and
thus contribute greatly to the fertility of the adja,cent farm lands. Fur-
thermore, 011 certain farms in the Division of Britstown, traversed by a
tributary of the Brak River, the practice, when the river comes down in.
flood, of constructing checks or weirs to control the passage of the water
and retain the silt, so as to ensure the deposition of the transported silb
on the lands, is said to have resulted in such a continual enrichment of
the latter as to render any other mode of fertilising needless. This ig
scarcely to be wondered at, and so valuable an accretion to the land surely
deserves the expenditure of time and trouble in retaining it.

It is, of course, well known — not to confine our examples to South
Africa — that this system of silting has been practised in the Nile Valley
from ancient times, and sandy soils, comparatively worthless before, have
become rich fields, and have remained productive for thousands of years.
Another case in point is that of the Rio Grande, the application of 24
inches of water from which adds nearly one quarter of an inch of soil
to the field, in the form of river sediment, and supplies every acre with
1,821 pounds of potassium sulphate, 116 pounds of phosphoric oxide, and
107 pounds of nitrogen. With regard to this, King observes:— f

" Four years of irrigation at this rate would add an inch of soil to the field, and 24
years would cover it six inches deep with a sediment containing three times the amount of
potash found in the average clay soil, and the same percentage of phosphoric acid, and a
high percentage of nitrogen."

QUEENSTOWN.

(Officially collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Gwatyu. J. Fronemann, junr. St. C. O. Sinclair.

o'

6' » » „

~t . . ,

* See pages 91 and 93.
f " Irrigation and Drainage," 2nd ed., 1902, p. 259.

110

This division was visited only incidentally, and samples were taken
from not mor© than a single farm, that of Mr. J. Fronemann, junr., lying
just across the Kei River, in the Gwatyu Field Cornetcy. The subsoil
below the place where No. 1 was taken is of a clayey character. No. 2
represents a. dark, loamy, alluvial soil, on a pot-clay subsoil, lying east
of the homestead. The land represented is reported to be fertile, oats
especially doing well upon it. No. 3 is a similar soil to No. 2, and was
collected in the same direction. No. 4 was taken from a dark clayey
soil, also lying east of the homestea-d, but nearer the river bank than either
Nos. 2 or 3.

(Privately collected.)

Uo. Field Cornetcy. Farm or place. Collector.

5. Whittlesea. Poplar Grove. F. Walsh.

Two samples of the soil occurring on the farm Poplar Grove were
collected, No. 5 being a red, and No. 6 a black soil.

The following are the results of the analyses of these soils : —

(Method I.)

Percent, of Percentage of Soil sifted through Percentage of Soil sifted

Field 1 mm. Sieve. through £ mm. Sieve.

Sample. Phos-

Uo. Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide

1. 85-0 2-36 4-82 '0060 '098 '346 '191 '041

2. 94-6 1-40 3-26 '0050 '091 '248 '127 '027

3. 96-5 1-80 4-65 '0064 '084 '170 '158 "013

4. 98-9 2-46 6'57 '0067 '126 '354 '209 "051

5. 99-2 1-34 2-68 '0831 "046 '150 '203 '052

6. 98'5 4-41 7'45 '0078 '119 T762 "215 '084

In this division, as in the adjoining division of Wodehouse, a distinct
tendency towards alkalinity is exhibited by certain soils, of which No. 5
in the above list is an instance — note its proportion of chlorine. Prac-
tically all the Queenstown soils examined were deficient in phosphates :
potash and lime are, generally speaking, present in ample quantities.

RICHMOND.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

1. East Uitvlugt, Toonbothasfontein. C. T. Ackermann.

2. „ ,, »

These two samples were taken from lands which were being used
for the cultivation of cereals. On the soil represented by No. 1, wheat
was grown, and although the straw produced was satisfactory, the grain
was less so, the ears being rather short. The soil from which No. 2 was

Ill

taken had been found to give much trouble, owing to its tendency to
form lumps after ploughing, and when water had been put on to the land
once or twice, it became compact and impermeable, and hence difficult to
work.

The results of the analyses* are as follows : —

(Method I.)

No.

Percent, of Percentage of Soil sifted through Percentage of Soil sifted
Field 1 mm. Sieve. through £ mm. Sieve.
Sample. Phos-
Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric
matter. oxid e

1.

86-3

1-63

5-12

•0021

•245

1-744

•242

•106

2.

91-1

1-73

4-48

•0190

•231

1-574

•269

•092

The soils represented by both these samples are rich in lime, and are
.also well supplied with nitrogen and potash ; phosphoric oxide is present
in fair quantity, so that the defects above referred to have their origin
apparently solely in the physical condition of the soil.

RIVERSDALE.

(Officially collected.)

i.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
•20.
21.
22.
23.
24.

Field Cornetcy.
Onder Duivenhoks River.

Vette River.
»
»
j)

35

Krombeks Rivier.
Riversdale.

5J

5)
»

Valsch Rivier.

Buff els Kraal.

>j

>j
Kaffir Kuils Rivier.

Farm or place.

Oude Muragie.

Jan Piena&r's Rivier.

Honigfontein.

Water Gat.

Brak Rivier.

»

Kweek Kraal.
Vette Rivier.
Oude Bosch.
Spiegel Rivier.

»
Novo.

jj

Klein Rivier.
Kruis Rivier.
Doom Kraal.
Assegaai Bosch.
Bosjesfontein.
Middelste Drift.
Zandfontem.

j>

Uroge Rug.
Hooge Kraal.
Tartouwa.

Collector.
J. Muller.

Broadly speaking, this division comprises three distinct belts of
•country, t clearly recognisable by the agricultural community; they run
from east to west, and each possesses a different type of soil. First of all

* For mechanical analyses of these soils tee under " Physical composition of soils "
(Part VII.). t See page 29.

112

there is the area generally known as the Berg or Zuurveld, lying to the
north of the village of Kiversdale, and parallel to the Langeberg Moun-
tains. Little sowing is done, but the soil, derived as it is from the sand-
stone of the Langeberg Range, needs a good deal of manure, a. usual-
feature with soils of such origin. Secondly, the stretch of country termed
the Ruggens; it is the intermediate belt which runs along south of the
Zuurveld, and is situated chiefly oil the geological formation known as
the Bokkeveld series. Within this belt most of the sowing is carried on,
and it is here that the best results are obtained. Lastly, to the south
of the Ruggens are the sand dunes all along the coast; these dunes are
hardly used, if indeed at all, for sowing purposes.

I — I.-'

<•'

0 HOE* OU/Y£ffHOK3
RIVIER

DIVISIONAL MAP Or

RIVERSDALE

Proceeding in a south-easterly direction from the village of Heidel-
berg, the farm Oude Muragie was first visited, and a loose clay soil, No.
1, was sampled, somewhat similar to the soil of the Ruggens in the Cale-
don Division. Soil of this nature is also found at Koega. and at Krans
Rivier. South of this farm are the sand dunes, extensive loose sandy
tracts, stretching for miles along the coast line, past the mouth of the
Gouritz River, on to< Mossel Bay and Brak River, and serving as a splen-
did pasturage, principally for cattle. As already observed, very little
or no sowing is carried on here.

About four miles to the north-east, sample No. 2, a loose clay soil,
was taken from the farm Jan Pienaar's Rivier. From this point the
nature of the soil varies very much until the farm Brak River is reached,
where two samples were selected, No. 5 from the side of a kloof destitute
of aloes (which seem to flourish better in the red Karroo soil than

113

in the ordinary Caledou Ruggens or clay soil, and hardly thrive
at all on sour veld). The other sample, No. 6, was a dark soil, collected
from lands on the opposite side of the homestead, and nearer the main
road. From the farm Kweek Kraal sample No. 7 was taken, typical of
the poorest lands under cultivation on that farm. Keeping along the right
bank of the Kaffir Kuils River, No. 3, a red sandy clay, was taken from
the plateau to the south-west of the farm Honigfontein, and a little fur-
ther west sample No. 4 was taken — a soil very similar in appearance to
the one just preceding — from the farm Water Gat, or Zwart Heuvel.

Before crossing the high watershed between Spiegel Rivier and Platte
Kloof, two samples were taken from the former place — No. 10, a black
loamy garden soil, and No. 11, a loose yellowish clay, which is said to be
an excellent soil for sweet potatoes. No. 9, from Oude Bosch, is a soil
similar to No. 10, more or less typical of the soil at Kliphoogte, Rem-
hoogte, Vergenoegd, on to Novo, where, however, the veld is not quite so
sour. On all four farms tobacco* grows fairly well.

From Vette Rivier a loose sandy soil, No. 8, was taken, as well as
Nos. 12 and 13 from lands in the valley on the farm Novo. Both of the
latter are rather dark sandy soils, differing in shade and fertility, the latter
being, it was said, somewhat less productive. Continuing along the right
bank of the Kaffir Kuils River, northward towards the mountains, one
passes over some very fine farms, and several varieties of soil ; typical
samples, Nos. 14, 15, and 16, taken from the farms Klein River and Kruis
River, and from lands on the hill slopes north of the village of Riversdale,
respectively, were selected. No. 17, a loose clay from the farm Assegaai
Bosch, taken north-east of the house, resembled the soil from the Caledon
Ruggens. This resemblance is also noticeable in the case of the area be-
tween the last-mentioned farm and Bosjesfontein, where No. 18, a rather
poor soil, was selected. Perhaps one of the largest grain farms in the
district is Middelste Drift; No. 19, a rich red Karroo soil, is typical of
the lands under cultivation on this farm; no manure whatever is used.
The amount of potash yielded by this soil, it may here be remarked, was,
with one exception, higher than that of any other soil in the Division.
Quite as good a soil, if not more fertile, is sample No. 21, taken from lands
on the hills above the level of the Gouritz River bed, while No. 20 is
typical of that below in the basin of the river; both these samples were
obtained on the farm Zandfontein. No. 20 is apparently largely in-
fluenced by the fertile silt carried down from Oudtshoorn and the Karroo
by the Gamka and Olifants Rivers, and subsequently by the Gouritz, as
noted in connection with the Mossel Bay soils.* At Onverwacht and all
farms similarly situated on either side of the river, soil more or less of
the nature of No. 21 is found. This almost dark red Karroo soil is ex-
tremely productive, and does very well without manure, but, being a very
warm soil, crops shrivel up very soon in seasons of drought. There being
so much of this good soil available, bright prospects may be awaiting this
district as a whole, and more particularly this section thereof, if only the
water question, which indeed is the great difficulty, could be solved, either
by dam making, or by more extensive irrigation by means of water raised
from the Gouritz River on to the fertile plains above. Before the advent
of the railway, farmers in this vicinity, having no market within easy
reach, nor any means sufficiently remunerative of disposing of their crops,
were reluctant to venture on any extensive irrigation schemes. Under
such circumstances, more extensive sowing was carried on during dry
years all along the less productive slopes of the Langeberg Range, where

* See page 91.

114

water was plentiful, even though the comparative poverty of the soil in-
volved the application of much manure in order to secure good results.

Beyond the fa.rm Droge Rug, where sample No. 22 was selected, the
country becomes very hilly and rugged; certainly not suited for sowing,
it is more adapted for pastoral purposes, especially for goats and sheep.
Keeping along the main road to Riversdale, several good farms were
passed over — the soil bearing great resemblance to« that at Droge Rug —
until Hooge Kraal was reached, where No. 23, a very good red " Karroo "
soil was collected; a portion of the land represented by this sample was
at the time well under cultivation with cereals. At Tartouwa, No. 24 was
sampled, a very poor soil from old lands lying waste and declared to be
unfit for further cultivation. The sample is quite typical of many of the
surrounding farms; physically the soil may be defective owing to its
coarseness of texture, but chemically its phosphoric oxide is certainly very
low, and it yields less nitrogen than any other soil collected within this
division.

The results of the analyses of the Riversdale soils are given in the
following table: —

(Method III.)

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.

Percent, of
Field

Percentage of Soil sifted through
1 mm. Sieve.

Percentage of Soil sifted
through £ mm. Sieve.

Sample.

Phos-

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

55-2

2-10

5-20

•014

•028

•013

•36

•090

58-8

1-79

4-85

•017

•056

•11

•21

•058

52-4

1-49

2-23

•004

•059

•44

•074

•098

59-6

•82

2-47

•003

•030

•19

•25

•082

65-2

1-73

4-20

•010

•15

•12

•13

•058

69-6

1-14

4-81

•012

•056

•12

•32

•081

59-6

2-66

6-59

•040

•11

•16

•22

•13

87-6

•91

1-66

•0088

•056

•093

•22

•13

83-6

1-66

3-78

•010

•056

•19

•14

•087

82-4

4'27

4-71

•020

•056

•075

•27

•065

77-2

4-35

5-95

•019

•084

•025

•26

•099

88-8

3-32

4-98

•012

•028

•15

•11

•078

77-6

2-05

4-01

•012

•15

•18

•24

•099

84-0

3-63

5-71

•013

•028

•13

•24

•15

77-6

1-13

2-99

•013

•028

•13

•24

•11

72-8

2-08

3-85

•054

•028

•18

•28

•061

68'0

1-88

3-57

•013

•030

•23

•082

•061

75-6

2-69

4-50

•062

•070

•14

•19

•069

85-6

4-05

4-87

•012

•11

•13

•34

•056

82-8

4-27

5-45

•014

•14

•93

•29

•22

93-0

2-58

2-85

•0062

•028

•11

•15

•044

77-2

•96

2-13

•0053

•028

•14

•26

•069

77'8

2-38

4-68

•on

•029

•10

•29

•089

48-4

1-61

4-32

•025

•027

•13

•24

•044

The soils of a large portion of this division, — especially the Field-
cornetcies of Onder Duivenhoks River, Valsch Rivier, and Kaffir Kuils
River, are for the most part, to judge from the above figures, rather low
in phosphates, particularly if used for the cultivation of wheat. In this
respect they exhibit the one great want of the clay slate and shale soils,
which constitute so extensive an area of the Western Province grain
districts.

115

No. Field Cornetcy.
1. Achter Cogmans Kloof.
2.

ROBERTSON.

(Officially collected.)

Farm or place.
Donkerkloof.

Collector.
J. Muller.

3.

Concordia.

4.

))

5.

Harm on ie I.

6.

5>

7.

Harmonic II.

8.

Het Kruis.

9.

5>

10.

Baden.

11.

55

12. Montagu.

Montagu.

13.

55 •

14.

55

15. „

Riet Vallei.

16. „

Het Kruis Pad.

17. „

»

18. Tradouw.

Brakkefontein.

19. Voor Cogmans Kloof.

Riet Vallei.

20.

55

21. Robertson.

Government Exp

22.

23.

24.

25.

26.

27.

28. Middle Bosiesveld.

29.

30.

31.

Station.

Keur Kloof.
Hex River.

5>

Vrolykheid.
Riet Vallei.
Bosjesmans River.

S. B. Morgenrood.
»

E. A. Nobbs.

S. B. Morgenrood.

The Robertson Division is more of a viticultural than a grain-pro*-
ducing district. Geologically, the nature of the country is most compre-
hensive : the chemically poor Table Mountain sandstone is much less in
evidence than in the more southerly and westerly divisions of the Colony ;
on the other hand, especially as one approaches the village of Robertson,
not only do granite and Malmesbury slates contribute to the formation of
the soil, but the latter is enriched by the cretaceous Enon formation, and
diversified by beds of the Ecca, Dwyka, Witteberg, and Bokkeveld series.
Naturally then, as one may expect, the soil is frequently very rich, and,
•with the Breede River and its tributaries not far distant, it is more avail-
able for irrigation than is the case in the Swellendam Division, and can
therefore the more easily be made to yield good crops.

To commence with the description of the soils collected within the
Field Cornetcy of Robertson, four soils, Nos. 21, 22, 23, and 24, were col-
lected by Dr. Nobbs from the Government Experiment Station, which
occupies about 61 acres of the Robertson Municipal Commonage. The
lands consist of alluvial valley soil, and comprise three somewhat different
typee. These three types are represented respectively by Nos. 21, 22, and
23. No. 22 is a red-coloured light sandy loam, resting upon a denser, and

lib

more clayey subsoil, over twenty feet in depth : this soil has a tendency
towards alkalinity in parts, and No. 24, in fact, represents an alkaline
patch in the area covered by the soil whereof No. 22 is a. type. No. 21 is
another red sandy loam of considerable depth, overlying a subsoil similar
to that of No. 22. The third type of soil at the station is that whereof
No. 23 is a sample. It is a grey calcareous loam, with a subsoil of almost
white appearance. In this soil brak water is met with a short distance
below the surface. Various parts of the district around possess a
" Karroo " type of soil, which inclines to become alkaline, whilst in the
direction of the mountains, and along the banks of the larger rivers, the
soil is naturally more " broken." No. 25 represents an intermediate, or
transition, " broken " alluvial soil from the farm Keur Kloof, off Lange
Vallei, on the mountain side of Robertson. At the time the sample was
taken the wheat in this valley was quite six feet high, with full ears.
Proceeding up the Vink River valley, No. 26, a sample of " vlei turf " soil,
was obtained at the junction of the Norree and Vink Rivers, on the Norree
side of the same, and another, No. 27, representing the area of the Upper
Vink River. The latter was taken close to the river bed, and could not
be expected to be as fertile as the other " Karroo " soil, such as No. 19,
for instance. Both No. 26 and No. 27 were procured on the farm Hex
River.

No. 19 represents a good " Karroo" soil from the farm Riet Vallei,
on the Klaasvoogds River, and No. 20 is what is termed a "doom'' soil
from the same area.

Crossing the Breede River towards Lady Grey, one comes upon two
similar valleys : the Konings and Keizers River valleys. No. 28 was
taken from the latter, on the farm Vrolykheid, while No. 29 is a sour soil
from Riet Vallei, representing the upper mountainous regions: its lime
content shows a diminution upon, those of the other soils of this vicinity.
Returning eastwards, No. 31 is a type of alluvial soil from the valley of
the Bosjesmans River, where there are a number of small farms, and No.
30 is a primary hill soil, representing the " Ruggens " on the right bank
of the Breede River, and to eastward of the Bosjesmans River.

The best cultivated section of the Robertson Division is the southern
portion of the Field Cornetcy of Achter Cogmans Kloof, and more es-
pecially those areas known as The Coo and The Keisie. In the Coo, from
the farm Donkerkloof, two samples were taken : No. 1 from lands on the
right bank of the Laats River, and No. 2 not far from the homestead.
The "veld" here, as well as at the neighbouring farms Concordia and
Keerom, — all of which are situated along the Coos River — is sour, as is
frequently the case with a sandstone derived soil. No. 3 was a virgin soil
of loose sandy clay, typical of most of the cultivated lands about Con-
cordia, and said to be very suitable for potatoes : chemically it proves to
be decidedly poor all round : for potatoes it may possibly be suited on
account of its physical, that is to say, its sandy nature, but much manur-
ing, especially with potash, would be needed to maintain potato crops.
No. 4, taken from the rise nearer the homestead, is somewhat more clayey,
but is at the same time a more gravelly, and, in local opinion, a poorer
soil than No. 3. Judging, however, from the analytical results, both are
deficient in plant food, and would require to be well manured with a
good all-round fertiliser. The last four samples are typical of the Coo,
which, notwithstanding external resemblances, differs inherently from the
Keisie, a belt of country running along the Keisies River towards Mon-
tagu, and bounded on the north by the Riethoek Mountains and on the
south by the Langeberg Range. The line of demarcation between the
Coo and the Keisie is the high watershed of the Coos Berg, the Coo being
several hundred feet higher than the Keisie, and also drier and more

117

Karroo-like in character. There is more sowing in the Coo than in the
Jveisie; but, on the other hand, the latter is very picturesque with its
many orchards and vineyards on either side of the river. From the results
of the chemical analyses it appears that the soils of the Keisie have a
richer store of plant food than those of the Coo, especially the more
eastern Keisie soils, i.e., towards Baden. The presence of abundance of
available lime seems in a great measure to be accountable for the supe-
riority of the Keisie for fruit culture, and it is not improbable that the
latter valley has been enriched by alluvium carried down from the higher
levels of the Coo. The first farm visited on entering the Keisie was Mr.
J. D. Conradie's Harmonic. Here two distinct varieties of soil were sam-
pled; No. 5, a virgin sandy clay, somewhat like a Karroo soil in appear-
ance, commonly known as " gebroken grond " (i.e., broken Karroo) repre-
senting the soil of the lands on the slopes, and No. 6, taken from lands
on the right bank of the river, rather further from the homestead, and all
round a better soil than No. 5, although rather alkaline. It represents a
fairly good alluvial soil, not very clayey, but retentive of moisture. It
is obvious from the analysis why, of these two soils, No. 6 succeeds better
than No. 5 ; it contains eight times the quantity of lime, and fifty per
cent, more phosphoric oxide.

Following the course of the river, the farm of Mr. E. J. van Zyl, Har-
monie, was visited, and No. 7, an alluvial soil, was taken close to the river
bed. : it is said to be richer than the two former, but slightly more brack-
ish : it contains nearly five times as much lime as No. 6, and forty times
more than No. 5, and herein it illustrates the adage that " a lime country
is a rich country." At the farm Het Kruis No. 8, a sample of " vied "
(i.e., alluvial) soil, was taken near to a vineyard. The owner of the farm,
Mr. B. B. Burger, stated that by continued cultivation and irrigation it
is rendered alkaline, — a characteristic of many Karroo soils, for by irri-
gation the surfaces of such clayey soils become quite hard, and, evapora-
tion proceeding very rapidly during warm weather, the water from lower
depths raises the alkaline salts — or " brak " as they are locally termed —
and deposits them at the soil surface. Hence the necessity in such cases,
for efficient drainage and for keeping the surface soil loose. Mr. Burger
gives it as his experience that guano is of practically no use to his soil.
No. 9, a very finely grained clay soil ,was collected next to vineyards
situated at the junction of the Keisies and Keizers Rivers: it was de-
clared to be richer than No. 8, and the explanation may be found in the
fact that it contains very much more lime: that it does not become brack
on irrigation is most probably due to the fact that the conditions for rapid
surface evaporation are less favourable than in the case of No. 8. In that
case, the vines being mostly young, and their foliage having been de-
stroyed by frost, protection for the surface of the soil from the rays of the
sun is lacking, so that a speedy caking of the surface ensues, and evapo-
ration goes on much more rapidly: as a result, the alkaline salts collect-
sooner than in a well-drained vineyard where the surface is constantly
loosened, and evaporation reduced to a minimum.

From the farm Baden two samples were taken : No. 10, a virgin sam-
ple of vineyard soil, and No. 11, a rich clay "turf" of very fine texture,
said to be an excellent potato soil. The chemical analysis of this soil cer-
tainly confirms the farm-owner's opinion regarding its fertility.

The village of Montagu being known for it® luxuriant vines and
healthy fruit, typical samples of the soils predominating in and about
that village were selected: No. 12, an average sample of a sandy clay soil,
was taken from the vineyard of Mr. D. S. du Toit. No. 13, a rather more
clayey soil, was taken from Mr. J. F. Burger's garden. In his opinion
basic slag gives good results for French grapes in a sandy soil, but not

118

in stiff clay. No. 14 was taken from fallow lands close to the road, on
the way to Baden : this was a loose sandy soil, rather poor and requiring
good general manuring. No. 15 represents an alluvial soil taken on the
farm Riet Vallei. The surrounding " veld " is very sour, and the soil
appears to have been thoroughly worked out. All three constituents of
mineral plant food — lime, potash, and phosphoric oxide — are very badly
needed. At Kruis Pad two> samples were taken: No. 16, a loose clay soil,
said to give good results with cereals when manured with bird guano, and
No. 17, a somewhat similar soil, but found to be more productive, and an
excellent soil for potatoes. These views, it may be noted, are not borne
out by the analyses.*

From the village of Montagu a north-easterly course was taken: the
character of the soil passed over, up to the Wagenbooms Bergen, is very
similar to previous samples of " Karroo " soil. On crossing this high

DIVISIONAL MAP OF

ROBERTSON

ridge, a distinct change is at once manifest, the " veld " becomes sour, and
the soil very much poorer in appearance. A typical sample of this soil
ia No. 18, collected from the farm Brakkefontein, where very little sowing
is done; cattle farming is principally carried on, so that there is an abund-
ance of manure. As the soil in this vicinity is primary sandstone soil,
from the Wagenbooms Mountains, it is probable that most samples taken
from this locality would show less lime and potash than No. 18, which
may possibly have become affected by manuring, a point regarding which
there is occasionally much uncertainty.

(Privately collected.)

No.

Field Cornetcy.

Farm or place.

Collector.

32.

Robertson.

Zand Rivier.

Forest Officer.

33.

n

yy

yy

34.

'»

"

"

* Except in so far that No. 17 is rather coarse-grained and therefore more suited for
potato culture than fine-grained soils usually are.

119

These three samples of soil were collected on the Government Forest
Plantation, about a mile and a half from Robertson village, and located
between the Worcester Road and the Breede River.

The analytical results are recorded in the following tables: —

(Method I.)

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
21.
22.
23.
24.

Percent, of Percentage of Soil sifted through
Field 1 mm. Sieve.
Sample.
Fine earth. Water. Organic Chloiine. Nitrogen
matter.

Percentage of Soil sifted
through j mm. Sieve.
Phos-
. Lime. Potash, phoric
oxide.

90-1

1-08

6-18

•0212

•175

•064

•126

•061

80-9

•65

4-33

•0057

•098

•032

•180

•068

66-4

•81

3-25

•0035

•084

•030

•027

•036

62-4

•59

2-95

•0057

•084

•013

•060

•045

70-2

•98

4-52

•0028

•112

•016

•147

•050

80-5

1-23

5-41

•0230

•154

•122

•153

•079

83-9

1-07

3-92

•0251

•112

•564

•095

•047

80-8

T13

4-64

•0095

•098

•114

•250

•120

95-2

1-49

5-71

•0138

•196

•742

•202

•193

49-3

T93

5-28

•0092

•098

•808

•331

•148

94-7

1-28

6-07

•0085

•168

•600

•247

•139

91-1

•96

4-17

•0170

•098

•260

•242

•096

88-1

1-21

5-47

•0156

•217

•120

•218

•132

70-9

•28

1-54

•0067

•070

•064

•070

•037

60-3

•21

1-21

•0046

•028

•010

•036

•010

71-1

1-20

6-45

•0078

•168

•038

•125

•079

64-7

•90

3-63

•0092

•084

•050

•078

•074

65-0

•38

1-91

•0057

•070

•116

•111

•055

90-4

•97

2-59

•0184

•084

•207

•095

•036

90-0

•95

2-98

•0099

•070

•276

•091

•033

89-2

1-97

4-72

•0248

•154

•461

•179

•042

75-0

2*85

3-35

•5489

•056

•078

•129

•040

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of Soil sifted through
3 mm. Sieve.

Phos-

No.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

Potash.

phoric
oxide.

19.

1-54

2-45

•0071

•077

2-65

•15

•017

20.

1-26

3-64

•014

•11

•89

•070

•0037

25.

1-11

4-51

•014

•11

•72

•14

•Oil

26.

2-09

7-60

•095

•098

1-20

•12

•083

27.

•38

1-53

•0053

•13

•34

•066

•036

32.

•50

1-15

•0118

•112

•025

•035

•041

33.

T84

3-16

•0982

•119

•143

•100

•054

34.

2-40

5-27

•1598

•140

1-311

•147

•092

28.

•34

1-94

•023

•077

1-08

•070

•060

29.

1-35

5-01

•014

•14

•13

•10

•021

30.

2-09

6-30

•0088

•16

•79

•32

•051

31.

1-48

4-96

•035

•056

•29

•22

•068

In regard to Nos. 21, 22, 23, and 24, additional extractions of the
soil were made, for the purpose of comparing the results, by methods
IV. and V., and by pure distilled water : in the soil-extracts obtained by

120

method V. determinations of lime, potash, and phosphoric oxide were
made, in those obtained by method IV. lime and potash were determined,
and in the aqueous extract determinations of lime alone were performed:
the results thus arrived at are appended : —

(Method IV.)

Percentage of Soil sifted through £ mm. Sieve.
No. Lime. Potash.

21. -200 -492

22. -312 -447

23. '510 -677

24. -072 '833

(Method V.)

Percentage of Soil sifted through 3 mm. Sieve.
No. Lime. Potash. Phosphoric oxide.

21. -213 '025 '0028

22. -310 "032 -0045

23. '396 '040 '0069

24. '072 -021 '0007

(By extraction with water.)

Percentage of Soil sifted through

3 mm. Sieve.
No. Lime.

24. -0444

The soils of the Robertson Division show an improvement upon those
of Caledon, Bredasdorp, and Swellendam in respect of the proportion of
phospHatic material contained. The three divisions just mentioned ha-ve
in their soils an exceedingly meagre reserve of available phosphates: it is,
therefore, by no means surprising that farmers in some portions of the
Robertson Division complain that they have purchased and applied super-
phosphates with no marked results; for, in the areas referred to potash
was quit© as essentially needed, especially in the cultivation of root crops,
and this need was in no way supplied by superphosphates alone — an illus-
tration of the applicability of the law of the minimum.

On the basis of the chemical analyses, the western part of the Achter
Cogmans Kloof Field-cornetcy is distinctly inferior to that further east, as
may easily be verified by comparing the analytical figures of Nos. 1 to 5 with
those of Nos. 6 to 11. Nos. 1 to 5 are all poor in lime, Nos. 3 and 4 being
also lacking in phosphates; in fact, No. 3 is deficient all round. In the
remaining soils of the Field Cornetcy a more satisfactory state of affairs
prevails: true, No. 8 is low in its proportion of lime, but this is quite a
solitary instance in that part of the district. In the Montagu Field Cor-
netcy the results of the analyses are lower; Nos. 14, 15, 16, and 17 are all
lacking in respect of lime, and No. 14 in phosphates as well, while No. 15
is poor in lime, potash, phosphates, and nitrogen. It is a matter of in-
terest worth noting that, beginning with the western part of Ladismith
Division, and travelling eastwards towards Oudtshoorn, there is a mani-
fest upward tendency in the potash content of the soil. Many of the soils
of this area would, nevertheless, be the better for a larger proportion of
available phosphates, and some, notably Nos. 8, 10, and 12 of the Ladi-
smith soils, are really poor with respect to lime.

No.

1.

o

3*

4.

ST. MARK'S.

(Officially collected.)

Field Cornetcy. Farm or place.

Cofimvaba. Government Plantation.

Collector.
St. C. O. Sinclair.

Of these samples, collected from the Government Plantation at Cofim-
vaba, the first, a black valley soil, was taken from No. III. compartment,
Block A. It rests on a reddish subsoil, inclined to be gravelly, and con-
taining a small quantity of ironstone pebbles. The second sample also
represents a black valley soil which lies on the same level as the preceding
sample, and was taken from Compartment II. of the same block. No. 3,
representing the subsoil of No. 2, consists largely of potclay with a, small
amount of ironstone pebbles. No. 4 is a red soil from Compartment VI.,
Block B, resting on red micaceous sandstone. The soils of this area are
apparently derived from the Burghersdorp beds (Beaufort series) of the
Karroo system, and may possibly be influenced as to their composition by
the presence of dolerite in some localities: the soils represented by the
above samples, however, are not as well supplied with plant food as those,
for instance, of the Albert Division.

The analyses resulted as follows : —

(Method I.)

Percent, of Percentage of Soil sifted through 1 rnm.
Field Sieve.

Sample.

No. Fine earth. Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through £ mm. Sieve.

Phos-
Lime. Potash, phoric

oxide.

1.

2.
3.
4.

90-1
87-9
90-1
57-9

1-52
1-61
5-38
1-39

3-08
2-91
4-95

2-69

•0276
•0361
•0439

•0'248

•056
•119
119
•091

•020
•060
•141
•032

•070
•051
•155
•061

•022
•022
•033
•020

The greatest need of these soils seems to be phosphatic material.
The three surface soils are likewise deficient in lime, nor are they parti-
cularly well supplied with potash. From a chemical point of view the
soil represented by sample No. 2 would be considerably improved by
bringing its subsoil (of which No. 3 is a specimen) to the surface, and no
doubt such a course may be found profitable in many other localities, but
to indicate such spots chemical investigation is needed.

No.
1.
2.
3.

Field Cornetcy.
Somerset.

>»
Vogel Rivier.

SOMERSET EAST.

(Officially collected.}
Farm or place.
Sterkwater.

j?
Cranmere.

Collector.
E. A. Nobbs.

A. C. "MacDonald.

Nos. 1 and 2 were collected on the Cape Government Railway farm
Sterkwater : one was taken from newly broken land in the cultivated area,
and the second from a very characteristic stretch of " rooi gras " veld.
The former of these two had been very frequently cropped, and the result
is seen in the chemical analysis which shows it to be somewhat inferior to

122

the other. In both cases the soil lacks lime, and, although in lesser
degree, phosphates: nitrogen and potash, however, are present in good
quantity.

Sample No. 3 was taken on Mr. G. Palmer's farm Cranmere or Gal-
genbosch.

The following figures were obtained by analysis: —

No.

(Method I.)
Percent, of Percentage of Soil sifted through 1 rn.m

Field
Sample.
No. Fine earth. Water.

Sieve.

Percentage of Soil sifted
through \ mm. Sieve.

Phos-

99-5
99-1

4-34
4-21

Organic
matter.

13-54
14-09

ilorine.

Nitrogen.

Lime.

Potash.

phoric
oxide.

•006
•006

•293
•321

•026
•034

•146
•124

•067
•093

(Method II.)

Percentage of Soil sifted through 1 mm. Sieve.
Water. Organic Chlorine. Nitrogen.

2'81

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.

Organic
matter.

5-65

Percentage of Soil sifted

through 3 mm. Sieve.
Lime. Potash. Phos-
phoric

•Oil —

STELLENBOSCH.

•68

'31

•01

(Officially collected.)

Field Cornetcy.
Klapmuts.

Farm or place.
Elsenburg.

Collector.
A. Simoni.

A. C. MacDonald.

Nooitgedacht. C. F. Juritz.

Nineteen samples of surface soils and subsoils were collected on the
Government Farm Elsenburg; these soils were taken from different parts
of the farm, the surface soils to a depth of eight inches, and the subsoils,
represented by Nos. 4, 6, 8, 10, 12, and 14, from eight inches to a depth

123

of two feet in each case. Nearly all of these soils were fairly coarse-
grained. The farm lies on rocks of the Malmesbury series, but quartzita
is very prominent, and potash, as well as phosphoric oxide, are accordingly
present in only small amount; the proportion of lime is, generally speak-
ing, more satisfactory. Results bearing close resemblance to those ob-
tained from the Elsenburg soils, were yielded by the analyses of many of
the soils collected in the Cape Division near Durban, from the farms
Bloemhof and Phesante Kraal, for instance, and also in the Malmesbury
Division.

,.* ,-\

!^4&<s\

tf5«- ^S_ A. 8s f. tfijii • *« \

•» *1** V. »T<*Ae? .» *

S^-V^ x o **f5f<-* \

DIVISIONAL MAP OF

STELLENBOSCH

Five soils were taken on the farm Nooitgedacht, in the same Field
Cornetcy ; the farm lies on rocks of the Malmesbury series. Nos. 20 and
21, which represented clay soils, were taken respectively from the side
of a hill west of the homestead, and from low-lying marshy ground north
of the homestead. No. 22 was collected near the centre of the farm,
while Nos. 23 and 24 were taken from vineyards, the former near the
bouse, and the latter from a hillside east of the homestead. All these
were taken from the portion of the farm lying between the highway and

124

the railroad. There is not much difference between these soils and those
from Elsenburg, already alluded to; the Elsenburg soils, like these, lie
on Malmesbury slates.

(Privately collected.)

Eerste River.

No. Field Cornetcy.

25. Klapmuts.

26.

27. Stellenbosch.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

•38.

39. Helderberg.

40. Hottentot's Holland.

Farm or place.
Nooitgedacht.

Zand Drift.

Thelma.

Neethlingshof.

Eerste River Station.

Govt. Forest Plantation.

Collector.
C. P. Lounsbury.

C. Mayer.

A.
D.

H. Mathew.
E. Hut chins.

Saxenburg.

))

Kuiken Vallei.
Fair View.

J. C. deWaal.

J.W. L! Hofmeyr.
J. Cairncross.

Nos. 25 and 26 were two red clayey soils from Nooitgedacht, rather
different from the others taken on the same farm ; they were taken from
a vineyard on the hillside to the left of the main road from Stellenbosch,
where probably they would be largely influenced by the disintegration of
the granite from the Bottelary Berg. These two samples did not repre-
sent the whole of the vineyard soil, but only patches of reddish soil,
termed in the vernacular " red dead ground " — where vines had never
flourished, and were showing signs of dying out. No. 25 was taken from
a depth of from six to nine inches. No. 26 was a clay from three feet
deep. Here, unlike the ca&e of the Graaff-Reinet soils and that of No.
23 of the Paarl soils, it appeared to be the chemical condition of the
soil that was at fault, and net the physical alone, if indeed at all. The
fact seemed to be that the nutrition of the vines was defective, if the
poverty of the soil is any criterion, both samples being deficient in all
plant food; they closely resembled in this respect some primary granitic
soils analysed by Dr. Hahn several years ago, in which the granite had
not been completely decomposed. The results of some of these analysis
,are here given : —

Locality. Lime.

Groot Constantia

High Constantia

-025

-008

-002

'081

Hout Bay '016

'026

Eerste River (Vlaggeberg) '018

'181

'063

Stellenbosch '014

Somerset West '034

•080

Potash.

•on

•020
•013
•043
•002
•012
•010
•017
•004
•019
•029
•025

Phosphoric
oxide.
•009
•019
•019
•075
•002
•Oil
•Oil
•Oil
•007
•002
trace,
trace.

Two samples of vineyard soils — Nos. 27 and 28 — of which No. 28 had
been manured with basic slag, were taken on the farms Zand Drift and
Thelma, in the immediate vicinity of the town of Stellenbosch, and from

125

granite soil at Neethlingshof in the Eerste River Field Cornetcy, near
Vlottenberg, a sample was taken representing a soil that had been
uiinianured for six years. At Saxenburg, in the same Field Cornetcy,
where No. 37 was collected, the effect of the lime from the blown sea-
sand begins to be perceptible ; it is also to be noticed in at least one of
the three Vlaggeberg soils from the same vicinity, analysed by Dr. Halm.
No. 30 was collected at a spot about 200 yards off the main road, near
Eerste River Station. The surface soil a.t this point is the usual white
sand of the Cape Flats and Downs, with a darker soil below. The locality
where this particular sample was taken lies about 14 miles distant, almost
due west, from the part of the Wynberg Flats where Nos. 55 and 56 of
the Cape Division soils were obtained, the former representing the wes-
tern and the latter the eastern margin of the Cape Flats.

In the direction of Somerset West two soils were taken for analysis;
one of these was taken on the hill near Somerset West, from part of the
farm De Hoop, now known as Fair View; the other was a granitic soil
from the farm Kuiken Vallei, where the subsoil consists of pot clay. These
two soils are represented by Nos. 39 and 40.

The analytical results obtained from the soils of this Division are
tabulated below : —

No.

(Method I.)

Percent, of Percentage of Soil sifted through 1 mm.
Field Sieve.

Sample.

Fine earth. Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through % mm. Sieve.

Phos-

Lirne. Potash. phoric
oxide.

1.

80-2

•95

2-07

•0085

•056

•065

•024

•024*

2.

57-7

•63

2-25

•0106

•056

•044

•024

•050

3.

85-4

—

—

—

—

•144

•024

•036

4.

88-3

—

—

—

—

•156

•022

•014

5.

85-2

—

—

—

—

•137

•018

•015

6.

85-4

—

—

—

—

•132

•017

•018

7.

81'1

—

— .

—

—

•219

•027

•044

8.

83-5

—

—

—

—

•160

•018

•046

9.

70-7

—

—

—

—

•036

•017

•034

10.

76-6

—

—

—

—

•034

•016

•050

11.

62-0

— .

—

—

—

•136

•036

•042

12.

63-2

—

—

—

—

•088

•029

•062

13.

80'3

—

—

—

—

•210

•061

•064

14.

79-8

—

—

—

—

•172

•051

•054

15.

84-3

—

—

—

—

•165

•038

•034

16.

90-2

—

—

—

•365

•070

•067

17.

74-3

—

—

—

—

•118

•037

•038

18.

82-2

—

—

—

—

•062

•033

•019

19.

85-0

—

—

—

—

•037

•030

•026

25.

92-2

•81

4-26

•005

•014

•008

•039

•046

26.

93-6

1-03

4-90

•006

•014

•006

•039

•045

27.

74-0

1-25

3-66

•0035

•144

•078

•068

•061

28.

92-2

2-49

7'25

•0152

•101

•026

•021

•073

29.

69-1

1-26

4'05

•0053

•057

•024

•023

•018

30.

98-1

—

-~

—

—

•066

•010

•003

40.

79-1

1-20

4-15

•0113

•059

•016

•015

•027

* In the case of this soil extraction? by Method V and by water were also made ;

see next page.

120

(Method II.)

No.

20.
21.
22.
23.
24.
31.
32.
33.
34.
35.
36.
37.
38.

Percentage of Soil sifted through 1 mm.
Sieve.

Water.

Percentage of Soil sifted through
3 mm. Sieve.

Phos-

Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide

T28
•44
•58
•36
•40
•38
•10
•11

T27
•19

No.
1.

•14

•014

•007

•016

•002

•11

•008

•008

•14
•33
•19
•13

•076
18-66
15-57
11-56
16-27
18-50
15-98

•60
trace.

•15

•082

•077

•16

•052

•28

•16

•20

•26

•12

•24

•04

•04

•029

•048

•068

•035

•018

•01

•01

•04

•07

•04

•09

trace.

trace.

(Method V.)

Potash. Phosphoric oxide.
•017 -0036

Sample No. 1, upon extraction by water, yielded '0013 per cent, of
phosphoric oxide, or slightly over one twentieth the amount extracted by
the standard hydrochloric acid method.

No.
39.

(Extraction by boiling with Hydrochloric acid.)
Water. Organic Chlorine. Lime. Potash.

•31

Organic
matter.

2-72

•019

14

•44

Phos-
phoric
oxide.

•018

No. Field Cornetcy.

1. Brak River.

2.

3.

4-

5.

6.

7.

8.

9.

STEYNSBURG.

(Officially collected.)
Farm or place.
Klerks Kraal.
Middel Water.
Zout Kuil.

Van Vuuren's Kraal.

Collector.
E. A. Nobbs.

C. F. Juritz.

The above soils were all taken from lands below the dam which
it was proposed to construct in connection with the Thebus Irrigation
project. No. 1 represents a typical Karroo* soil, and No. 2 a Karroo soil
of somewhat lighter character. No. 3 was taken from the side of a water
course after removal of the surface soil, and No-. 4 from lands which had.
been long under irrigation. Nos. 4 to 9 all represent surface soils, and,
like the preceding four samples, differed considerably in some respects,
especially in regard to the proportions of lime and potash contained from
Nos. 10 to 19 (see below).

127

So.
10.

11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

Field Cometcy.
Brak River.

(Privately collected.)

Farm or place.
Thebus.

Collector.

Engineer, Public Works
Department.

Nos. 10 to 19 were also collected within the irrigable area above
referred to, at Thebus. No. 20 is representative of the silt brought down
by the river at TSebus.

The results obtained by chemical analysis from the soils taken from,
this locality are set forth in the following tables: —

(Method I.)

Percent, of Percentage of Soil sifted through 1 mm.
Field Sieve.

Sample.

No. Fine earth. Water. Organic Chlorine. Nitrogen.

matter.

1. 98-4 4-33 4-09 '0106 '071

2. 98-8 4-27 3'79 -0276 '071

3. 94-8 6'19 3-34 '0630 '043

4. 92-9 2-89 3'23 '0149 '071

5. 97-8 5-65 4'81 '114 '070

6. 97-5 4-28 4'25 '049 '028

7. 91-1 4-00 4-26 '132 '084

8. 98-0 5-58 5-09 '141 '098

9. 98-0 3-99 3-99 '024 '084

Percentage of Soil sifted
through J mm. Sieve.

Phos-
Lime. Potash. phoric

oxide.

•238
•130
•042
•448
•062
•062
•044
•066
•188

•196
•192
•195
•152
•181
•213
•189
•257
•161

•013
•037
•031
•038
•089
•121
•097
•093'
•080

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Nitrogen.

No.

Water.

Organic

Chlorine.

matter.

*10.

4'70

6-16

•046

11.

2-86

2-46

•016

12.

6-52

4'33

•337

13.

6'35

5-00

•044

14.

4-17

3-07

•177

15.

3-20

2-47

•085

16.

5-95

5-00

•319

17.

5-08

3-92

•087

18.

3'25

2-29

•062

19.

4-21

4-42

•057

20.

—

12-99

—

Percentage of Soil sifted through
3 mm. Sieve.

Phos-

Lime. Potash. phoric

oxide.

•028

6-775

•764

•053

•031

2-192

1-038

•151

•025

3-617

1-206

•055

•021

8-570

1-218

•110

•028

3-430

•566

•076

•084

2-075

•452

•089

•035

6-675

1-314

•051

•038

2-535

•038

•076

•028

1-445

1-010

•091

•077

12-022

•978

•258

—

•580

•248

•139

* In connection with these results see also remarks on the soils of this Division under
the head of " Alkalinity of Soils " (Part VI).

There is apparently a general resemblance, both chemical and phy-
sical, between the soils of this division and those of the adjoining divisions
of Albert on the west, and Colesberg on the east : they are all fairly fine
grained soils, not inadequately furnished with lime in the condition o-f
plant food, and moderately supplied with phosphates. The Karroo rocks
of the Stormberg series, wherein lime strata are specially prominent, are
apparently responsible for these features. Between the Steynsburg and
the Colesberg soils particularly, the similarity in respect of plant food is
very strong, as may be seen from the following comparative figures : —

Division.

Colesberg.
Steynsburg.

No. of

Samples

Fine
. Earth.

Water.

Organic
matter.

Chlorine.

Nitrogen.

Lime.

Potash.

Phos-
phoric
oxide.

6

94-7

3-06

4-24

•006

•088

•174

•171

•078

9

96-4

4'58

4-09

•064

•069

•142

•193

•067

The samples available for comparison were these from Oorlogspoort
(numbered 1 to 6) in the Colesberg Division, and the nine from Steyns-
burg analysed according to the same method. The latter differ materially
from the former in one respect only, namely, in regard to the amounts of
chlorides they contain : this point will be dealt with again in connection
with the subject of alkaline soils.

STOCKENSTROM.

(Privately collected.)

No.

1.

2.
3.
4.
5.

Field Cornetcy.
Mancazana.

Balfour.

Farm or p^ace.
Ebenezer West.

Upsher.

Collector.
H. H. Davison.

J. Berwinger.

A large portion of this division is devoted to the cultivation of
tobacco, and four soils were taken for analysis from the Government
Tobacco Farm at Ebenezer West: they would appear to be, for general
purposes, fairly well furnished with lime, and also, except in the case of
No. 3, with potash. All fo'Ur, however, stand in need of phosphoric
oxide.

No. 5 represents a soil which is used for tobacco culture at Upsher.

The results of the analysis of these soils are as follows : —

(Method II.)

No.

1.

2.
3.
4.
5.

Percentage of Soil sifted through 1 mm. Sieve.
Water. Organic Chlorine. Nitrogen.

1-75

2-22
2-86
4'34
1-84

Organic

matter.

3-12
3-81
5-07
7-39
5-50

•0036
•0034
•0023
•0022
•014

•088
•10
•10
•20

Percentage of Soil sifted
through 3 mm. Sieve.

Phos-
Potash.

Lime.

•32

•44
•22
•60
•84

•16

•17
•097
•15
•3?

phoric
oxide.

•049
•048
•031
•045
13

129
SWELLENDAM.

No.

1.
2.
3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.

Field-cornetcy.
Tradouw.

Rivier Zonder End

Kluitjes Kraal.

??
Swellendam.

Breede River.
?>
?j
j>

Heidelberg.

j>

»

j>

Kamemelk River.

5)

Zuurbrak.
Klip River.

(Officially collected.)

Farm or place.

Zeve-nfontein.

Poortfontein.

Uitvlugt.

Het Goed Geloof.

Barry dale.

Tradouws Hoek.

Doom Rivers Vallei.

»

Lemoens Hoek.
Appelskraal.

Stormsvlei.
Verdwaial Kloof.
Klipfontein
Vryheid.
Kluitjes Kraal.
Klippe River.
Appelbosch.
Distelsfontein.
Oude Post.
Bonteboks Kloof.
Kinko.
Uitvlugt.
Zwartklip.
Kadies Vallei.
Rhenosterf ontein .
Klein Duine Rug.
Wagen Drift.
Asch Kraal.
Duivenhoks River.
Hooi Kraal.
Honig Klip.
Kamemelk River.
Melkhoutboom,

»
Bosjesmana Pad.

Collector.
J. Muller.

S. B. Morgenrood.

The Langeberg Range, which runs midway across the Swellendam
Division, divides it into tvo areas widely differing in their character-
istics. The southern portion resembles the adjacent division of Rivers-
dale, although the coastal belt of downs is less prominent; the northern
part — or Tradouw — is of similar type to the Ladismith Division, and
the portion of Robertson which lies north of the aforementioned range.
The soils of this entire area to the north of the Langeberg Range, be-
tween 20° 0' E. and 21° 45' E., appear to possess a better average of
plant food than the soils south of the mountains; and, as far as> analyses
have shown hitherto, North Swellendam seems, in this respect, to be
more advantageously situated than either Ladismith on the east, or
North Robertson on the west.

130

Commencing with the consideration of the southern portion of the
Swellendam Division, the first farm to be visited was Appelskraal, on
the right bank of the River Zonder End. Here three samples of alluvial
soils were taken from lands on the opposite side of the river; No. 10, a
light sandy soil; No. 11, a black drift sand or sandy loam; and No. 12,
a sandy clay. A sample of loose yellow clay, No. 13, was also taken,
from lands about a mile east of the farmhouse, in the direction of the
Ruggens. The general poverty, and especially the low lime-oontent of
the soils will become evident on contrasting them with the results yielded
by the other soils of this part of the division. The locality from which
they were taken lies just on the verge of the mass of sandstone that forms
the River Zonder End Range, and they are apparently influenced thereby.
The poverty in lime of the Caledon soils* is ascribable to a similar cause.

w.""8.^

•xsnezr-iZ \«

DIVISIONAL. MAP Or*

SWELLENDAM

* -*O> N^»SWrfl^ y
\^ B***9vtfti Sg*>; fa !

/^cawr BMvuuT-^' „ l^hm^
_ r— ^-J r* .?..*•""?»•''

No. 14, a sample of alluvial soil, was taken from the farm Stormsvlei,
about an hour's drive further along the river; this represents a stiff
clay soil mixed with, organic matter. Proceeding thence about three
miles south of Appelskraal, a loose clay, No. 15, was collected from the
farm Verdwaal Kloof, and af tea-wards No. 16, also a loose clay, from the
farm Klipfontein, about eight miles south-east of the point where the
preceding soil was sampled.

* See page 38.

131

At Vryheid, about 11 J miles south-west of the village of Swellendam,
a stiff clay soil, No. 17, was obtained, from lands to the right of the
main road. Directing a course thence towards Swollendam, the farm
Kluitjes Kraal, on the right of the Breede River, was touched at. From
this farm a stiff clay, No. 18, similar to that from Vryheid, was pro-
cured. The last farm visited on this tour was Klippe River, just at the
entrance to the village, along the main road. Here a sandy loam, No.
19, was taken from ground adjoining an area under tobacco culture.

On a subsequent journey, taking the tract of country between Swel-
lendam and Zuurbrak, and between the Langeberg Mountains and the
Breede River, one primary soil, No. 21, was collected at Distelsfontein ;
it was a fair, slightly warm soil, somewhat resembling the soil of C'ale-
don, and moderately retentive of moisture. One " doom " soil, No. 23,
was taken at Bonteboks Kloof, and two alluvial soils, No®. 20 and 22,
from Appelsbosch and Oude Post respectively.

Proceeding in a south-easterly direction, along the Port Beaufort
road, a diversion was made at Kinko. Here, on the small watershed,
was found the most fertile type of soil met with between this and Gale-
don. This class of soil extends over Koerannie, Kinko, Uitvlugt, Harte-
beeste River, and other farms eastward towards the Upper Slang River.
The formation is not such denuded slate as is generally to be seen, the
overlying sedimentary beds forming undulating hills capped by flat sand-
stone and iron " kopjes." On the higher ridges above Karnemelk River
there are a number of deposits of clay. The soil is pebbly in the best
lands, with substratum of lime, and very retentive of moisture, although
not clayey. The samples from Kinko (No. 24), Uitvlugt (No. 25), and
Karnemelks River (No, 35), sufficiently represent this area. Southwest
of this, along the Breede River as far as Paarden Kloof, is a stretch of
" geil-grond " — a fairly good but stony and warm soil, in which, except
during a season of abundant rainfall, the crops shrivel up. During the
drought that had continued for three years, the yield had been insignifi-
cant, and supplies had to be transported from the Caledon Division, a
long day's journey. The formation is bare vertical shale, with outcrops
of limestone; in the northern and southern portions there is some iron-
stone; the valleys are precipitous. One sample typical of this area, was
taken on the farm Zwartklip, near to Michiels Kraal. The gravels on
this farm, and at Rhenosterberg, mentioned below, bear some resem-
blance to that already noticed at Klippe Drift, Bredasdorp Division.
The fertility, both here and at Uitvlugt, seoms to be chiefly owing to
the presence of available lime in comparatively large amount.

Keeping along the ridge, a series of undulating plateaus are crossed,
with extensive tracts of fertile land, alternately " zwart turf," and red
gravel, with a subsoil cf limestone. On the west bank of the Slang
River the soil becomes entirely a red deep gravel, with little limestone,
until, on ascending the other side, one comes to gravelly flats with oc-
casional " turf." This continues throughout Zandfontem and Jakhalsfon-
tein to Duine Rug, where the soil becomes more sandy. Towards Melk-
hout Kraal, along the Duivenhoks River, the soil is very sandy. A
sample of " zwart turf," No. 27, was secured at Waterkloof, adjoining
Kadies Vallei, another of gravelly soil, No. 28, at Rhenosterfontein, and
one of " broken " soil, No. 29, at Duine Rug.

Between this area and Heidelberg the farms were very much sub-
divided, and several varieties of soil were under cultivation. No. 30
represents a " broken " gravel from Goedemans Kraal, on the border of
Wagendrift; No. 31 is what is locally known as "Karroo" soil, from
Asch Kraal. No. 33 is a hill soil from Hooi Kraal, and No. 32 a river
soil from the farm Duivenhoks River, outside Heidelberg.

132

Crossing the high watershed, No. 34, a sample of Upper Ruggens
soil was taken from Honigklip, where the land under cultivation lies
at an exceedingly steep angle, and two samples from the farm Melkhout-
boom were taken, representing the valleys between these hills and the
Langeberg Range, namely, No. 36, a sample of " vlei " (alluvial) soil, and
No. 37, a hill soil.

North-west of Swelleoidam, bounded by the Langeberg Mountains
and the Breede River, is a long stretch of decomposed shale and sand-
stone in alternate hills and well-watered valleys. Nearly the whole of
this picturesque. section, dotted with orchards, vineyards, and grain lands,
belongs to the Van Eeden family. Being favourably situated as regards
rainfall, etc., this area admits of being extensively cultivated, the low-
lying lands, with a potash dressing being particularly suited to the
growth of potatoes, and possibly tobacco. Sample No. 38 from the farm
Nooitgedacht (Bosjesman's Pad), together with a varient of the same
previously obtained at Klippe River, sufficiently typifies these valleys.

A considerable portion of tthe Swellendam Division is really pastoral ;
the scant water supply and the difficulties of transport previous to the
advent of the railway having rendered extension of the cultivated area
too risky, even although good soil may have been available. The water
question remains a great difficulty, and the ways and means of conserv-
ing more water have more than once been, seriously discussed. Owing
to a bend in the high Langeberg Range, the moisture laden clouds,
whether from the north-west or south, precipitate their showers round
and about Swellendam village, and northward thereof; a smaller por-
tion condenses about the Potteberg mountains and Port Beaufort, while
/ain clouds pass over the intermediate area without shedding any of their
moisture.

On a final visit to this division the northern portion, already re-
ferred to, was dealt with, and samples were collected from the Tradou/w,
the high plateau which extends right across that part of the Swellendam
Division, and embraces a considerable part of the adjoining division of
Robertson, stretching from the Touws River in the north and east, to
the Langeberg Range in the south, and the Wagenbooms Bergen in the
south-west. Sowing is carried on successfully only along the Touws
River, which widens out very considerably in some parts. No. 1 was
taken at Zevenfontein, and is a very typical sample of this fine red
11 Karroo " soil, known as " doom grond." From the eastern portion of
the Tradouw two samples were selected. Very little, if any, cultivation
is carried on between the Groot River, south-east of the village of Ladi-
smith, and the farm Uitvlugt, or Warmbad, situated at the eastern end
of the Warm water Bad. Mountain. No. 2 represents a loose, unmanured,
sandy soil, taken from the garden in front of the homestead on the latter
farm. No. 3, a loose clay soil, was collected as being typical of the
ground in the valley; it had a very good local reputation, and its quality
is excellent according to the chemical analysis. In the neighbourhood
of Barrydale, No. 9 was chosen as being more representative of the richer
" broken " soil in the valley than the sour and poorer variety near the
mountains. Beyond Doom River the farm Doom Rivera Vallei was
touched at, and two samples of soil collected. The first of these, No. 7,
was a " broken " Karroo soil. No. 8 was a loose clay, of the class usually
known locally as " broken " soil, of a rather acid character. The soil
becomes more sour on approaching the sandstone of the Langeberg Moun-
tains. No. 5, a very loose clay soil, taken from the garden of Mr. G. du
Toit, forms a representative sample of the nature of the soil in the
village of Barrydale.

133

On the way from Barry dale to Montagu, No. 6, a somewhat sandy
soil, said to be rather poor, though fairly well suited for potatoes when
well manured, was taken at Tradouws Hoek. Here again chemical

analysis confirms practical experience, for, on analysis, this soil was found
to be far the poorest of all the Tradouw soils examined. No. 4, a virgin
soil, was collected from the farm Het Goed Geloof, where the soil is rather
more sour than at Tradouws Hoek.

In the following tables will be found the results obtained by analysis
from the several soils enumerated above :

(Method I.)

No.

1.
2.
3.
4.
5.
6.
7.
8.
9.

Percent, of

Percentage of Soil sifted

through

Percentage

of Soil sifted

Field

1 mm

. Sieve.

through

i mm.

Sieve.

Sample.

Phos-

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Lime. Potash.

phoric

matter.

oxide.

97-2

•64

3-59

•0078

•064

•220

•146

•082

62-7

•31

1-32

•0120

•021

•046

•176

•036

81-0

2-04

6'92

•1180

•126

2-518

•416

•234

62-5

2-00

11-03

•0092

•392

•074

•210

•201

83-4

1-16

6-12

•0120

•168

•040

•250

•096

76-8

•77

4-15

•0134

•112

•024

•033

•023

72-8

1-21

5-76

•0057

•189

•042

•262

•134

80-3

ri8

5'95

•0099

•154

•106

•272

•151

89-0

1-21

5-69

•0460

•168

•170

•197

•122

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of Soil gifted
through 3 mm. Sieve.

Phos-

No.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

10.

•09

•77

•0056

•11

•084

•034

•0048

11.

1-05

5-41

•025

•18

•060

•035

•on

12.

•57

3-36

•013

•16

•044

•049

•015

13.

•75

3-85

•0085

•14

•058

•10

•014

14.

•68

3-59

•093

•15

•14

•078

•016

15.

•95

6-90

•018

•16

•25

•084

•013

16.

1-44

7-20

•028

•14

•16

•13

•012

17.

•51

3-73

•on

•16

•080

•055

•017

18.

•80

4-78

•086

•15

•084

•14

•022

19.

•86

6'26

•018

•16

•29

•12

•040

20.

1-50

4-64

•0028

•028

•096

•015

•036

21.

•65

4-30

•0019

•042

•16

•074

•052

22.

1-58

5-91

•0025

•065

•15

•053

•041

23.

1-81

6-76

•019

•098

•63

•23

•044

24.

1-08

4-14

•019

•098

•33

•062

•023

25.

1-76

5'65

•010

•098

1-18

•43

•009

26.

1-35

5-38

•0068

•077

•45

•23

•018

27.

3-16

9-30

•013

•070

7-57

•35

•038

28.

•59

1-87

•0073

•056

•16

•099

•010

29.

•84

2-98

•on

•042

•32

•018

•014

30.

•91

4-25

•019

•056

•37

•087

•019

31.

2-68

7-53

•086

•035

2-73

•29

•041

32.

•21

2-16

•0057

•063

•64

•033

•on

33.

1-64

4-97

•053

•15

•51

•11

•015

34.

1-70

9-26

•015

•12

•40

•045

•008

35.

4-82

4-07

•048

•077

•37

•11

•031

36.

2-86

9-07

•016

•084

•23

•060

•019

37.

2-07

4-67

•015

•084

•083

•099

•036

38.

1-43

s-cb

•015

•084

•46

•10

•010

134

The proportion of lime in the soil appears to diminish, as one passes
from Heidelberg and Breede River, through the Swellendam and Zuurbrak
Field Cornet cies in the direction of Robertson, but rises as soon as the
latter division is reached. Phosphoric oxide is very deficient through-
out. The soils of Northern Swellendam, it will be observed, are almost
uniformly well supplied with potash.

TULBAGH.

(Officially collected.)

Xo. Field Cornet cy. Farm or pl:,ee. Collector.

1. Koopmans River. Doornboom. J. Muller.

o

m* » » »

0

» » »

Three samples of rather gravelly soil were collected on Mr. F.
Baker's farm Doornboom, adjoining Porterville Road Railway Station.
No. 1 was taken from stubble land, which was just being ploughed, be-
tween the cottage and the railway line; No. 2 was collected from, land
already ploughed, and nearer the cottage; No. 3 was from new lands
beyond the cottage on the mountain side.

Geologically the Division of Tulbagh may be said to comprise four
parallel strips, alternately of Malmesburv clay slate beds and sandstone
ridges, running almost due north and south. * The three samples above
mentioned were taken from the slope of the western sandstone moun-
tains, just on the border of the Malmesbury formation; they conform,
in chemical composition, to the usual poverty of the sanolstone soils, and,
as the proportions of fine earth show, are very coarse-grained in texture.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

4. Koopmans River. Porterville Road.

5.

6. Breede River. Kluitjes Kraal. Forest Officer.

7

1 • a » »>

"• ?> » »

q

"• » a »

10. ,;

11. Waterfalls. Knolle Vallei.

?> » tf

13. Winterhoek. Misgund. J.F. Tlieron.

Two samples of soil were taken from the farm of the Imperial Cold
Storage Company at Porterville Road. In practically all respects they
resemble the neighbouring soils from the farm Doornboom.

All the remaining samples collected within this division were taken
from the tongue of slates between the two sandstone ranges, but the
surface soil is, generally speaking, influenced by the latter, and this —
added to the fact that the soils, even if undiluted in this way, and
formed solely by the disintegration of the Malmesbury beds, would pro-
bably show results no better than those noticed in the Malmesbury and
Cape Divisions — leads one to expect all-round poor soils; an expecta-
tion which is fully realised by the results of the chemical analysis.

* See Map on pacr; 160.

At Kluitjes Kraal, near Ceres Road Railway Station, is a Govern-
ment Forest Plantation, whence, from different spots around the old
homestead, four samples of sandy loam garden soil and adjacent old
garden soil were collected : these four soils are represented in the tables
by Nos. 7, 8, 9, and 10. No. 6 represents a black soil taken from the
river bed on the same farm. This soil is, as may be surmised, much finer
grained than the sandstone soils, and is fairly well supplied with organic
matter and nitrogen, but the mineral components of plant food, although
less scanty than in the sandier soils, are far from satisfactory for crops,
although probably sufficient for the tardier requirements of arboriculture.

At Knolle Vallei, a farm adjacent to Kluitjes Kraal, and purchased
by the Government for the purpose of a railway sleeper plantation, two
samples were procured through the Conservator of Forests. Of these,
No. 11 was a dark, and No. 12 a fine-grained light-coloured clay: they
were taken from a hill about a mile east of the homestead. The surface
soil in the vicinity had the appearance of a very rich clayey loam, about
fifteen inches deep, the subsoil being a stiff reddish clay.

In the Winterhoek Field Cornetcy a sample was collected on the
farm Misgund. The locality forms the very tip of the tongue of clay
slate which runs up north between the sandstone of the Witjienberg
Range on the east, and that of the Roode Zand Mountains on the west,
and is cut off by the Winterhoek Mountains on the north. The soil is
therefore naturally very poor in all plant food constituents^ for the
locality is a cul-de-sac of clay slate, surrounded on three sides by sandstone
mountains. The soil had been used for the cultivation of tobacco, and
had been manured. The sample was collected from, a hillside, and con-
sisted of a somewhat gravelly clay, the subsoil being a fine yellow clay.
The local agriculturists declare that, unless manured, this soil proves
almost worthless, and herein is again exemplified the confirmation by prac-
tical experience of the opinions based upon the chemical analysis of the
soil.

The chemical analyses* of the Tulbagh soils resulted in the figures
tabulated below : —

(Method I.)

Percent, of

Field
Sample.

Percentage of Soil sifted through 1 mm.
Sieve.

Percentage of 9oil sifted
through £ mm. Sieve.

Phos-

No.

Fine earth.

Water.

Organic
matter.

Chlorine. .

Nitrogen.

Lime.

Potash.

phoric
oxide.

1.

49-3

•49

2'84

•0067

•087

•014

•042

•027

2.

38-9

i-oo

5-16

•0071

•130

•036

•063

•045

3.

24-4

•27

T61

•0042

•043

•008

•020

•019

4.

34-7

1-12

3-22

•0850

•063

•032

•079

•063

5.

31-3

1-27

4-17

•0720

•105

•026

•084

•069

6.

94-8

2-92

11-78

•0375

•308

•070

•095

•070

11.

61-4

1-49

3-64

•0212

•095

•068

•038

•047

12.

98-7

5-75

10-09

•0127

•088

•008

•030

•033

13.

73-4

•95

3'51

•0064

•088

•044

•032

•023

* Partial mechanical analyses of Nos. 1, 2, 3, and 6 will be found under the heading
of " Physical composition of soils " (Part VII).

136

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

No.

7.

8.

9.

10.

Water.

Organic
matter.

Chlorine. Nitrogren.

Percentage of Soil sifted through
3 mm. Sieve.

Phos-

Lime. Potash. phoric

oxide.

•060
•083
•102
•050

•033
•031
•036
•058

•052
•075
•052
•089

All of these soils, with hardly an exception in any particular, are
lacking in the several mineral constituents of plant food.

UITENHAGE.

No.

I.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.

Field Cornetcy.
Sundays River.

(Privately collected.)

Farm or place.
Upper Landdrost. R.

Gouvernements Belooning.

Klaas Kraal.

> j

rt
Malmaison.

Collector.
E. Wright.

Uitenhage.

Small Kloof.

Stoneleigh. H. Griffiths.

The first ten samples in the above list were all collected on the
days River Estates, from lands intended to be placed under irrigation.
No. 11 was taken from the residential plot Stoneleigh at the town of
Uitenhage, where it was intended to plant some 50,000 tobacco plants : it
proved to be poor in phosphoric oxide.

The results obtained by the analyses of these eleven soils* are given
in the tables below : —

(Method I.)

No.

Percent, of Percentage of Soil sifted through
Field 1 mm. Sieve.

Sample.

Fine earth. Water. Organic. Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through £ mm. Sieve.

Phoa-

Lime. Potash. phoric
oxide

1.

99-5

2-70

4'72

•0085

•126

•182

•292

•073

2.

98'9

•79

2-56

•0057

•091

•156

•197

•068

3.

99'1

1-92

5-62

•0028

•140

•718

•333

•143

4.

98-3

1-07

3-30

•0028

•084

•078

•108

•075

5.

93-2

•94

3-64

•0021

•105

•392

•239

•116

6.

98-4

•90

2-90

•0021

•084

•066

•178

•068

7.

99-6

1-02

2-54

•0042

•070

•182

•185

•092

8.

95'8

1-22

3-48

•0021

•098

•278

•276

•087

9.

99-2

1-18

3'73

•0028

•168

•338

•278

•132

10.

100

1-35

3-63

•0085

•140

•592

•226

•086

* For the results of partial mechanical analyses of the first ten soils, see under
" Physical composition of soils " (Part VII).

(Method II.)

Percentage of Soil sifted through 1 mm.
Sieve.

No,

11.

Water.
3-13

Organic
matter.

6-46

Chlorine. Nitrogen.
•0034 -173

Percentage of Soil sifted
through 3 mm. Sieve.

Phos-

Lime. Potash. phoric

oxide.

1-776 '206 '036

The analytical results show that the Sundays River Estates soils are,
taken all rouind, satisfactorily furnished with Nitrogen, Lime, and Potash,
and have a fair proportion of phosphates. It frequently happens that
soils containing abundance of lime, as these do, are not inadequately pro-
vided with potash. The lime, in the present case, undoubtedly owes its
presence in the soil to the numerous bands of limestone which form an
integral part of the Sundays River beds in the Uitenhage geological
series. To these limestone bands must evidently be ascribed the great
difference that exists between the soils of the Sundays River Valley and
those of the adjacent Humansdorp Division.

UMTATA.

No.
1.

2.

Field Cornetcy.
Umtata-.

(Privately collected.)

Farm or place.
Kambu Plantation.

Collector.
Forest Officer.

These two samples of soil were collected from the Kambu Forest
Plantation, distant about fifteen miles west of Umtata: No. 1 represents
the surface soil and No. 2 the subsoil. They were taken from a steep
ridge which runs down about a thousand yards from the Kambu Forest
to the Umtata River, at a point about half-way between the two. The
surface soil in the vicinity is typical of the red sour veld, which abounds
in the mountainous parts of the district, while the subsoil is sometwhat
clayey, and bright red in colour.

The analyses of the samples collected are as follows, and it will be
observed that the soils are, like most sour veld soils, deficient in prac-
tically all inorganic plant food : —

No.

1.

2.

Percent, of

Field

Sample.

Fine earth .

99-0
98-6

(Method I.)

Percentage of Soil sifted through 1 mm.
Sieve.

Water.

4-12

Organic
matter.

16-06
14-18

Chlorine. Nitrogen. Lime.

•0075
•016

•238
•147

Percentage of Soil sifted
through i mm. Sieve.

Phos-
Potash.

•024
•010

•040
•018

phoric
oxide.

•093
•065

No. Field Cornetcy.

1 . Zuurberg.

„
3. Lower Ibisi.

UMZIMKULU.

(Privately collected.)

Farm or place.
Nieshoutfontein.

ft
Kippendavie.

Collector.
— Robinson.

H. C C.'kippen.

138

Two samples of soil, No. 1 a surface soil, and No. 2 the 8ul>aoil
below it, from the farm Nieshoutfontein, or Sueezewood, were collected
from a plantation of young cluster pines in the vicinity of Mr. James
Cole's private forest. Attempts had been made in the neighbourhood,
both by the Government Forest Department and by private persons, to
grow Cluster Pine, but the results had not been satisfactory — at least not
until the young trees had attained to some years of growth. The soil,
which was taken from a gentle slope, is of a clayey nature, with some
five or six inches surface deposit of decayed vegetable growth. Under-
lying the surface soil, at varying depths, is a yellow shale.

Soil No. 3 was taken from the farm Kippendavie, part of the farm
Hopewell (A.4) and lying on the Umzimkulu River, about four miles
from Umzimkulu village in a southerly direction. The soil represents
lands whereon tobacco was being cultivated; the tobacco which it pro-
duced was said to be altogether too strong for smoking purposes. The
piece of land whence the sample was taken is situated in a valley, and
the soil is deep and sweet.

Below are the results of analysis : —

(Method I.)

No.

1.

2.

Percent, of Percentage of Soil sifted through 1 nim. Percentage of Soil sifted

Field Sieve. through ^ mm. Sieve.

Sample. Phos-

Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide.

99-2
99-0

7-51
7-41

17-19
12-18

•0106
•0502

•350
•203

•010
•010

•037
•041

•087
•042

(Method IL)

Percentage of Soil sifted through 1 mm.
Sieve.

No.
3.

Water.
7-71

Organic
ma ter.

11-42

Chlorine.
•0007

Nitrogen.

•084

Percentage of Soil sifted
through 3 mm. Sieve.

Ph os-
Lime. Potash. phoric
oxide.

•697 '052 '017

No.

FieJd Cornetey.

1.

Uniondale.

2.

)}

3.

l'

4.

Avontuur.

5.

tt

6.

Middle Long

Kloof.

7.

5J

8.

Lower Long

Kloof.

9.

Antonies Berg.

10.

^

11.

Buffels Klip.

12.

,,

Collector.
C. F. Juritz.

Of these soils, Nos. 1 and 2 show very poor results in all mineral
plant food. No. 3 is deficient in every respect except as regards lime.

UNIONDALE.

(Officially collected.)

Farm or plrce.

Kamnassie Wagen Drift.

Gold Diggings.

Uitvlugt.

Avontuur.

Vyge Kraal.

Misgund.

Klipheuvel.

Krakeel River.

Oude Post.

Dwaas.

Wilgehouts Rivier.

Roode Heuvel.

OUDT5HOO

139

The greater part of the Uniondale Division consists of valleys of
Bokkeveld shales, and hills of Table Mountain sandstone. It resembles,
therefore, many of the other divisions of the Colony in which rocks of the
Table Mountain and Bokkeveld series occupy the most prominent part.
Hence it may be anticipated that the Uniondale soils will exhibit similar
characteristics to those of, for instance, the Caledon Division, and a large
extent of the Bredasdorp and Swellendam Divisions. And, on the whole,
where no modifying influences are exerted, we find this to be the case.
Into this area of shales and sandstones there protrudes, from the Oudts-
hoorn Division, along the Olifants River banks, a wedge of the Uitenhage
series, and this gives a proportion of lime to the soils in the vicinity, that
serves as a preliminary indication to the traveller from Uniondale or Wil-
lowmore, that the Oudtshoorn boundary is being approached. It is not,
however, until thai boundary is actually crossed that the radical change
becomes fully perceptible. Reference was made to this when dealing
with the soils of Oudtshoorn.*

Entering the UniondaJe Division from George, sample No. 1 was pro-
cured on the farm Kamnassie Wagen Drift, about 5| miles W.S.W. of
Uniondale. This soil does not properly belong to the Long Kloof class, f
nor does No. 2, t&ken from the farm Gold Diggings, 5J miles in a south-
westerly direction from Uniondale. Of these two soils the former lies on
the Bokkeveld formation, and the latter on the sandstone: the difference
in chemical composition, to the disadvantage of the sandstone soil, is
clearly seen from the table of analyses.

Very broadly speaking, the Uniondale soils may be classified into
two groups, namely, the Long Kloof soils, derived from rocks of the Table
Mountain series, and the soils of the Bokkeveld series, which more resemble
the soil that characterises the Karroo. The Long Kloof, it is true, itself
presents two varieties of soil, for " sour veld " is met with along the
southern side of the kloof ; but, as this class of soil had already been repre-
sented by samples collected in the George and Knysna Divisions, only
sweet Long Kloof soils were taken for analysis, namely Nos. 4, 5, 6, 7, and
8. On ail these soils the rhenoster bush abounds. North of the Long
Kloof, and parellel to it, is another kloof, known as the Kouga, the soils
of which are practically similar to those of the Long Kloof. For this
reason samples of Kouga soils were not taken for special analysis.

Travelling in a north-easterly direction from Uniondale, and passing
from the sandstones to the Bokkeveld series, the character of the soil
changes, and approximates to that of the Karroo. Nowhere, however,
does it assimilate closely to the true Karroo soil, judging from the flora;
rhenoster bush continues to thrive, though less abundantly than in the
Long Kloof, whereas the Ganna (Salsola aphylla) so distinctive a mark
of Karroo soil, is entirely absent. These facts are easily understood, when
the small proportions of lime in these Uniondale soils are taken into
account. The absence of the rhenoster bush and the luxuriance of the
Ganna are generally — as remarked in connection with the soils of Oudts-
hoorn*— indications that lime is present in good proportion. From this
area No. 3 was collected on the farm Uitvlugt, about four miles north-
east of Uniondale. At Oude Post, six miles further on in the same direc-
tion, another sample, No. 9, was taken, and on the farm Dwaas, 17 miles
from Uniondale, No. 10 was procured. These samples — which contain
rather more potash than the purely sandstone soils — were all taken from
a band of Bokkeveld which extends — between the sandstone — over these
three farms. The soils are rather coarse in texture, and both the above

* Hee page 9G. f See page 57.

140

geological formations have evidently contributed to their composition. The
Bokkeveld formation extends westwards over the farms Wilgehouts Rivier,
Buffels Klip, and Roode Heuvel, but in this part of the division it is
less under the influence of the sandstone ; the soil is therefore finer in grain,
and better in chemical composition. At Wilgehouts Rivier No. 11 was
collected, about 16 miles north-west of Uniondale, No. 12 being taken from
the farm Roode Heuvel, six miles further west. The five soils just men-
tioned, Nos, 3, 9, 10, 11, and 12, which are all rather loose brown clays,
typify the soil of the entire northern part of the Uniondale division, and
form the second of the two classes of soil previously mentioned.

The results of the chemical examination of the Uniondale soils are
here tabulated : —

No.

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.

Percent, of
Field
Sample.
Fine earth.

57-9
57-1
57-3
83-4
66'8
35-8
94-2
92-4
57-2
54-1
93'2
81-7

(Method I.)

Percentage of Soil sifted thorough
1mm. Sieve.

Water.

1-54
1-78

•90

•71
1-58

•76

•74
1-07

•65

•76
1-09

•87

Organic
matter.

7-10
4-72
3-04
2-92
4-62
3-30
2'97
3'20
2-44
2-72
4-96
3-34

Chlorine. Nitrogen.

•0541
•0099
•0244
•0495
•0202
•0340
•0240
•0226
•0095
•0109
•0078
•0265

•175
•161
•098
•140
•161
•098
•126
•154
•140
•098
•154
•112

Percentage of Soil sifted
through ^ mm. Sieve.
Phos-
Lime. Potash, phoric
oxide.

•102

•182

•095

•054

•050

•077

•052

•074

•044

•058

•030

•036

•088

•026

•049

•050

•062

•052

•030

•038

•026

•040

•054

•044

•036

•094

•049

•028

•105

•067

•048

•212

•104

•098

•255

•122

Taking the Uniondale soils all round, there is a general lack of lime,
but a moderate supply of potash and phosphates, together with a good
percentage of nitrogen. The two samples taken within the Field Cornetcy
of Buffels Klip, representing the portion of the division bordering on the
fertile district of Oudtshoorn, show a marked chemical superiority over
the rest, for the reason already stated. No. 1, taken at Kamnassie Wagen
Drift, is also a fairly good soil, and differs distinctly, both in physical and
in chemical qualities, 'from the soils of the Long Kloof class. While not
exactly deficient in any one constituent, it would be greatly improved by
the addition of lime and phosphates.

The quasi-Karroo soils of this division have already been referred to,
and five samples were specially mentioned as typifying this area, namely :
Nos. 3, 9, 10, 11, and 12. The first three of these were collected in a
north-easterly direction from the village of Uniondale. They are less fine-
grained than the soils further south, are also loss retentive of moisture,
and contain a smaller proportion of nitrogenous material. Their supply
of l^me continues small — in fact, changes for the worse, but a slight im-
provement is noticeable in regard to phosphoric oxide, and a more dis-
tinct bettering with respect to potash. Turning further west, the area
from which Nos. 11 and 12 — already commented upon — is entered. In-
cluding with these five soils No. 1, we may perhaps venture to say that

141

the northern part of the Unioiidale Division is markedly superior to the
kloofs in the south with respect to- the inorganic plant food constituents
of the soil.

VICTORIA EAST.

So. Field Cornetcy.

1. Tyumie.

(Privately collected.)

Farm or Place.
Hogsback.

Collector.
C. L. Harvey.

A sample of fine-grained soil, on which it was proposed to plant
potatoes, was collected from the Hogsback Plateau, near Alice. The farm.
lies at an altitude of about 4,400 feet above sea-level. The surface soil
at the locality where the sample was collected is red and loamy; the sub-
soil is also red, but more clayey. The farm has a local reputation on ac-
count of its good forage and potato crops, but, in order to produce these,
a plentiful supply of manure is invariably found requisite: farmyard
manure is practically the sole fertiliser employed in the district.

The analysis of this soil resulted as follows : —

No.
1.

(Method I.)

Percent, of Percentage of Soil sifted through

Field 1 mm. Sieve.

Sample.

Fine earth. Water. Organic Chlorine. Nitrogen,
matter.

99-6

5-87

10-72

•007

182

Percentaj
througl

Lime.
•028

Bof Soil sifted
•} mm. Sieve.

Phos-

Potash. phoric
oxide.

•038

•055

This analysis clearly shows the reason why thorough manuring is so
necessary; lime and potash are both deficient, while phosphates are just
above the border line of poverty.

No.

1.

2

3.
4.
5.
6.

7.
8.
9.

VRYBURG.

(Officially collected.)

Field Cornetcy.

No. 7.

Farm or place.
Salisbury.
Luxmore.
Middle Park.

»
Mogogong.

Banksdrift.

Collector.
E. A. Nobbs.

The Taungs District, which is represented by the above samples, is
situated at the northern extremity of the Campbell Hand, and therefore
near the region of the crystalline and siliceous limestones which charac-
terise that formation. One may accordingly expect to find a good pro-
portion of lime in these soils, but such an anticipation is not borne out
by actual results. The lime formation extends only over the Great Kaap
Plateau, and practically terminates at its edge, with the Rand, below
which, to the south-east, along the Harts River banks, are comparatively
poor Dwyka shales — the grey and olive shales of Stow. The Taungs area

142

extends across the Campbell Rand to the westward, and takes in part of
the plateau above, but the samples of soil examined were collected only
from the region of the shales, and on the bank of the river opposite to
that flanked by the Campbell Band. Even here, however, the influence
of the lime formation may be noticed in certain localities (for instance in
the Ganzepan and H.V.67 soils of the Barkly West Division), but for the
most part the soil is poor.

The samples were collected by Dr. E. A. Nobbs, from grounds in-
tended to be irrigated by means of a system of furrows from a canal lead-
ing the water of the Harts River. The description of the soils collected
is compiled from the observations made by Dr. Nobbs on his tour. En-
tering the Taungs district from Barkly West, and leaving the long stretch
of poor sandy ground through which the railway passes, sample No. 2, a
fine and even dark grey loam, was collected on the farm Luxmore, on the
slopes of the Phokwani Hills. This soil, owing to the dry season, was in-
tensely hard, due to its fine compact texture, but not tough, and hence
probably friable when wet. No. 3 represents a similar soil, although of
somewhat looser texture, and was taken about 100 yards west of the 718th
railway mile post, on the farm Middle Park. It is a deep, very fine-
grained, and hence firm, sandy soil, of a whitish yellow colour on the sur-
face. The veld is Karroo-like in character, and the Ganna bush is notice-
able, but the entire area is not large, and appears to narrow down towards
the Harts River. Here and nowhere else a tendency towards brackishness
or alkalinity was observed. Three miles further north, however, the soil
changes to a fine-grained deep-yellow sand over forty feet deep (as seen at
the well at Railway Cottage 146). Of this soil — on which the Mimosa
grows — No. 4 is a type. Chemical analysis shows it to be very poor. No.
9, taken about five miles to the north-west, at Banksdrift, is quite different
in appearance, being a deep-red loamy sand of even consistency, and free
from stones, friable and easily worked, but of varying depth, and lying
upon shales and white sandstone; there is a very wide plain, apparently
of this character throughout. For irrigation purposes, the lie of the land
is ideal from this point, and over the farm Hartington; the chemical
results, however, show insufficient proportions of plant food.

The farms to the south of Hartington, along the river, are generally
unsuitable for irrigation purposes. A shallow sandy soil, limestone shales,
sandstones, and occasionally intrusive dolerite, constitute the general
nature, while along the river a hard clay, resting directly on shale, with-
out any deep alluvial deposits, is found. A shallow soil occurs in the
low levels, a hard clay overlying horizontal shale beds, and valueless for
irrigation, although well covered with excellent natural grass. Such are
the conditions on the farms Greefdale, Springbokfontein, Putsfontein,
H.V.70, and probably also Klip Kopje, until the soil merges into that of
which Nos. 1 (in the above list) and 6 (of the Barkly West Division soils)
are types. No. 1 represents an even grained brown sand, typical of a
wide tract of country; the sample was taken near the 710th nlile post.
Nos. 5 .and 6 were taken, less than 100 yards apart, in the valley to the
north-east of Mogogong native village. This land is stated to have yielded
good crops of mealies and Kafir corn for many years. It is a red loam,
verging upon the alluvial soil commonly known as " vlei-grond " ; there
may be from 3,000 to 4,000 acres of this land, almost flat, and surrounded
on three sides by low hills. No. 5 is a rich-red sandy loam of considerable
depth, and No. 6 a tough dark-red alluvial soil. Chemically the latter is
far the better of the two.

On the opposite of the railway line, and about four miles south-east
of Taunga railway station, Nos. 7 and 8 were collected, also near each
other; the former from a mealie patch, the latter from virgin or long

143

disused soil. These samples of red friable sandy loams were taken from
the centre of a well cultivated area, covered with good crops, and of about
2,500 to 3,000 acres extent. The underlying subsoil is a yellow clayey
loam of great depth. About five miles south-east of Taungs station this
good land gives place to an obviously less fertile and uncultivated greyish
soil, shallow, and resting upon schistose slates and sandstone®, which often
appear at the surface. Lime abounds there.

Just across the Harts River at this point, towards Thoming, there
lies a wide, level, and apparently fertile expanse, from which, however, no
samples were taken. This portion of the district would probably benefit
largely by moans of alluvial soil conveyed from the Campbell Rand to
the west. Thoming, in fact, lies upon the limestone rocks of the Campbell
Rand series.

(Privately collected.)

No. Field cornetcy. Farm or place. Collector.

10. No. 5. Geluk. H. Abt.

11. No. 12. Kuruman. C. D. H. Braine.

» » »

» » »

» » >j

» » »

» » »

A sample of somewhat sandy soil, No. 10, was collected at Geluk,
about thirty miles W.S.W. of Vryburg; the remaining six soils were
taken from the Crown Reserve about Kuruman township. Of the latter,
three were collected within the property of the London Mission, and the
other three around the Mission station. No. 11 was a red subsoil, taken
two feet below the surface to the north-east of the station. No. 12 was
taken from the surface near the central furrow in the middle of the Mis-
sion property; the spot was completely water-logged and overgrown with
reeds. Sample No. 13 was collected in front of the Mission House, and
represents a surface soil stated to have been under cultivation for over
sixty years and not recently manured. No. 14 is also a surface soil, taken
from the south-eastern portion of the Mission property. This ground waa
likewise under cultivation, but had not been manured of late. No. 15
was a subsoil, taken twelve inches below the surface, from the side of the
stream in the Dakwent valley, south-east of the Mission property. No.
16 was taken three feet below the surface, outside, and south of, the
Mission property, between the converging streams which join on the
Mission station. It was intended to irrigate all the lands represented by
these samples.

The large quantities of lime which these soils, in most oases, contain,
are obviously derived from the geological formations to which Stow* gave
the name of " The Campbell Rand Series." The Campbell Rand, a range
of precipitous cliffs, which terminate the Great Kaap Plateau towards
the south-east, and flank at first the Vaal and subsequently the Harts
River for a distance of over a hundred miles from the junction of the
former with the Orange River below Campbell, is composed of limestone
and dolomite, which extend westwards and northwards as far as Vryburg.

* " Notes on Griqualand West." Quarterly Journal of the Geological Society.
Dec., 1874, page 613.

The results obtained by the analyses of the above described soils are
as follows: —

No.

Percent, of

Field

Sample.

Fine earth,

(Method I.)

Percentage of Soil sifted through
1 mm. Sieve.

Percentage of Soil sifted
through £ mm. Sieve.

Phos-
Nitrogen. Lime. Potash. phoric

matter.

oxide.

1.

93-8

•54

1-17

•002

•013

•034

•034

•009

2.

85-2

2-95

3-11

•003

•056

•220

•101

•063

3.

95-4

2-56

3-38

•003

•028

•162

•103

•055

4.

88-2

•51

1-31

•003

•028

•010

•030

•014

5.

73-5

1-04

2-13

•002

•013

•022

•024

•045

6.

82-5

2-47

3-91

•004

•098

•130

•118

•059

7.

76-6

1-22

2-19

•003

•013

•008

•041

•045

8.

80-9

1-78

2'47

•004

•084

•054

•074

•040

9.

73-0

•84

1-71

•003

•042

•006

•039

•032

10.

89-6

•91

1-63

•0025

•161

•062

•020

•013

11.

76-4

1-01

1-48

•0039

•029

•026

•024

•024

12.

100

13-56

38-22

•0123

1-435

3'66

•030

•132

13.

88-7

2-89

7-54

•0074

•200

4-80

•084

•088

14.

85-2

3-54

4-99

•0170

•108

6-32

•059

•056

15.

94-7

4-16

6-75

•0042

•193

14-04

•028

•059

16.

87-1

T86

2-13

•0018

•071

•278

•068

•026

The results of the chemical analyses of the first nine soils are far
from promising : taking them as a whole, the soils just escape an all-round
poverty in plant-food by the fact that potash is present in fair amount.
Nitrogen, lime and phosphates are all alike lacking.

No. 10 is very poor in potash and phosphoric oxide, and its lime-
content leaves much to be desired. There is obviously nothing in its
composition to show that it overlies the limestones of the Campbell Rand
formation ; on the contrary, in respect of plant-food it resembles many a
Table Mountain sandstone soil, and is similarly made up of disintegrated
quartzites or sandstones, which, in this case, would probably be
derived from the quarzitic rocks of the Black Reef series.* A partial
mechanical analysis of this soil is given under the head of " Physical com-
position of soils."

No. 11 is poor in ©very respect, and No. 16 is deficient in lime. Nos.
12 to 15 indicate soils rich in lime, and No. 12 also shows a fair amount
of phosphoric oxide, but all these soils are Kelow normal as regards potash.
Of the last six soils in the above table, it will be noticed that the two
samples (Nos. 11 and 16) taken two and three feet below the surfa/ce, con-
tain very much less lime — in an available condition, at all events — than
the three surface soils, Nos. 12, 13, and 14, but even these latter do not
contain one half as much a* No. 15, which was taken twelve inches below
die surface. No. 11, it will also be observed, is the most gravelly of the
six.

See 10th Ann. Kept. Geol. Commsision, 1905, p. 245.

145
WILLOWMORE.

(Privately collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Willowmore. Van der Westhuyzen's Kraal. G. Kilgour.

This sample was collected from lands which it was proposed to irrigate,
oil the farm Van der Westhuyzen's Kraal, adjoining the farm Klipfontein,
and bordering on the Willowmore village commonage. The sample was
taken in the vicinity of the southerly beacon on the first-named farm.
The soil lies on a considerable and level extent of Bokkeveld beds, free
from extraneous modifying influence, and should therefore be expected to
yield better chemical results than even such soils as those from Wilgehouts
Rivier and Roode Heuvel in the Uniondale Division.

According to the analysis below, there is a good reserve of phosphatic
material in the soil in a condition available for plants : —

(Method II.)

Percentage of Soil sifted through 1 mm. Percentage of Soil sifted

Sieve. through 3 mm. Sieve.

Phos-

No. Water. Organic Chlorine. Nitrogen. Lime. Potash. phoric

matter. oxide.

1. 2-55 4-87 -0044 "077 '336 '193 '171

W1LLOWVALE.

(Officially collected.)

No. Field Cornetcy. Farm or place. Collector.

1. Willowvale. Government Plantation. St. C. O. Sinclair.

9

j> ?) jj

q

*•• J) )> ))

4

In the district of Willowvale three samples of soil were collected from
the area intended for the Government plantation at this place. No. 1,
a black clayey soil, was taken, to a depth of ten inches, from a hillside
facing north. No. 2 represents a patch covered by a long coarse5 grass,
which, in these parts, is regarded as an indication of great fertility. No.
3 is the subsoil underlying No 2. The soil in many parts of this planta-
tion consists of a rich black earth, one to two feet in depth, below which
is a layer of red gravel, containing a considerable quantity of ironstone
pebbles. Under this there is a yellow pot-clay, or rock decomposing into
clay.

No. 4 was a sample of black soil, taken about 12 miles west of Wil-
lowvale, and representative of a great part of the Willowvale district.

K

No.

1.
2.
3.
4.

140

The following results were obtained by chemical analysis :

(Method I.)

Percent, of
Field
Sample.
Fine earth.

Percentage of Soil sifted through 1 mm.
Sieve.

Water. Organic Chlorine. Nitrogen,
matter.

Percentage of Soil sifted
through % mm. Sieve.
Phos-
Lirne. Potash, phoric
oxide.

100

1-94

6-36

•0375

•147

•016

•071

•022

96-3

3-03

8'72

•0276

•203

•050

•193

•054

100

2-45

6-46

•0354

•105

•024

•340

•068

98-0

3-03

7-60

•0311

•175

•072

•024

•035

Here too phosphates are very low in quantity, being absolutely de-
ficient in the first and last samples. No. 4 is, in fact, lacking in all mineral
plant food. Nos. 2 and 3 contain fair proportions of phosphoric oxide, but
are poor in lime ; especially ia this true of the subsoil No. 3 ; both these
soils, however, have a satisfactory proportion of potash.

WODEHOUSE.

No. Field Cornetcy.

1. Groote Vlei.

2.

3.

(Privately collected.)

Farm or place.
De Boulogne.

Collector.
B. F Weir.

These three samples represent alluvial soils, and were collected on
the farm De Boulogne, now known as Fair Hope, in a valley at the foot
of the Stormberg Range, on the Indwe railway line, between Halseton
Station and Birds River Siding. The soil is a clayey loam, apparently
uniform to a depth of ten to fifteen feet. Under favourable conditions
the lauds have yielded heavy wheat crops, but needed frequent watering,
as on drying deep cracks were apt to make their appearance.

The analyses of these soils resulted as follows : —

(Method I.)

No.

1.
2.
3.

Percent, of Percentage of Soil rifted through 1 mm. Percentage of Soil sifted

Field Sieve. through £ mm. Sieve.

Sample. Phos-

Fine earth. Water. Organic Chlorine. Nitrogen. Lime. Potash, phoric

matter. oxide.

99-5 6-94
98-3 10-08
92-4 5-94

6-36

10-75

6-05

•0198
•1372
•0078

130
•344
•158

•232
•852
•346

•341 -074
•237 '069
•235 '145

Here again, as in the Division of Queenstown, the soil is inclined to
be brack; there are clear evidences of this in Nos. 1 and 2, if the pro-
portions of chlorine in those soils be taken as indication. In other respects
too the soila resemble those of the Queenstown Division; they are weli
furnished, for example, with lime and potash.

147

1.

2.

3.

4.

5.

6.

7.

8.

9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.

Field Cornetcy.
Wagenbooms Rivier.

Goudini.

Voorste Bosjesveld.

Over Hex River.

Worcester

Achter Hex River.

WORCESTER.

(Officially collected.)

Farm or place.

Eendracht.
Wilge Rivier.

j)

Breede Rivier.
Slanghoek.
Groat Eiland.
Klippe Drift.
De Doorns.
Stettin.
De Hoek.
Wagenboom.
Matjes Kloof.
Nonna.

Boven Kloppers Bosch.
Wilge Rivier.
Nooitgedacht.
Aan de Doom Rivier.
Haartebeest Rivier.
Tweefontein.
Zeekoegat.
Wyzers Drift.
Vendutie Kraal.

?>

De Doorns.
Keurbosch Kloof.

)>

Karbonaatjes Kraal.
Klein Straat.
Stinkfontein.
Ezeljacht.

Collector.
St. C. O. Sinclair

J. Muller

Geologically, the Division of Worcester is even more varied than that
of Robertson ; rocks of the Malme&bury, Table Mountain, Bokkeveld, Wit-
teberg, Dwyka, and Ecca series, as well as the surface deposits of limestone
and Enon conglomerate, all find a place within the area of the division.
With such a variety of rocks it is usual to find, not only soils of varied
chemical composition, but also, in general, soils of considerable fertility,
and this anticipation is borne out by the analyses tabulated below.

Proceeding from Cape Town by rail to the Hex River valley, five
samples typical of soil under cultivation were collected. From Vendutie
Kraal, the farm of Mr. W. Gie, two samples were taken — No. 22, an
orchard soil, under cultivation, which had been slightly manured; and No.
23, a virgin sandy clay soil, taken from lands further up the river. Mr.
Gie stated that he had used several fertilisers, and found that super-
phosphates gave the best results. No. 23, it will be observed, is very
deficient in phosphoric oxide.

The next two samples were taken on the farm Keurbosch Kloof : No.
26, a virgin sandy clay, resembling No. 22, and No. 25, a soil of rather
more clayey nature ; this latter soil was, in appearance, similar to larger
patches in the farms in the valley, where vines were being grown,
but on a less shallow soil. It was apparently the larger

148

amount of potash contained in this soil that led to its selec-
tion for the purpose of a vineyard. No. 24, a loose virgin
sandy clay, was collected on the farm De Dooms; it was de-
clared to be more productive than Nos. 22 and 23, and also than the
soils lower down the valley. This statement probably had primary refer-
ence to the adaptability of the soil for the cultivation of vines; and in
so far it is apparently correct, for of the three soils, this one is beet supplied
with potash, and the proportion of potash in the soil seems to increase
as one ascends the valley : but there the superiority ends ; in respect of
other plant food, it is certainly not the best soil. This Hex River valley,
so well known on account of its fine fruit, is almost covered with vineyards
and orchards.

The next section of the Worcester Division to be visited, was the
part bounded by the Hex River Range and the Touws River, where the
following soils were sampled: No. 28 was taken on the farm Klein Stra-at;

DIVISIONAL MAP OF

WORCESTER

it represents a loose sandy clay soil, with a reputation for being fairly
productive. It is very finely divided, as to texture, and has never been
manured, although, judging from the chemical results, it is rather poor
in lime and phosphates, due, evidently, to the close proximity of the sand-
stone mountains. On the way from here to the next farm, the sandstone
formation is directly entered upon, and the poverty of the soil increases.
From Karbonaatjes Kraal, No. 27, a sandy soil— found in practice to be
rather poor — was collected: the chemical analysis showed, it to be very
deficient in plant food. The sample was taken from the centre of the
cultivated land, and from the spot where the wheat, which was just then
in ear, was poorest. The soil at this place is much coarser than No. 28,
and affords a further instance of practical experience being abundantly
confirmed by analytical results.

149

Crossing the line of railway, the next farm visited was Stinkfontein,
about nine miles from Karbonaatjes Kraal, and six from Triangle Station.
Here No. 29, an unmanured sandy clay, was taken from the orchard.
There is a rather broad patch of the same soil under cultivation, and it
has the reputation of being more fertile than any surrounding ground.
From a chemical point of view it undoubtedly proves to be the best soil
in the Field-cornetcy, and one of the best in the entire division, of all the
soils analysed.

Ezeljacht, or Zout Rivier, was next touched at, and No. 30, a sample
of virgin soil, taken from the corner of some lands situated along the
road of! the farm, and about two miles from the previous farm. The nature
of the sour veld and soil is rather similar to that of No. 27, although
slightly better; it is, however, less fertile than No. 28.

From the foregoing remarks, coupled with a reference to the table of
analyses below, the general conclusion may be drawn that, within a certain
area enclosed by lines "drawn parallel to the railway on either side of it,
between Triangle and T'ouws River Stations, the fertility of the soil in-
creases as one approaches the T'ouws River. The fertile patch at Stinkfon-
tein is an exception, and the surrounding soil is very poor. The geological
formation is that of the Bokkeveld series, but the farms are flanked on
either side by high sandstone mountain ranges. There is not much sowing
done at any of the four farms mentioned in connection with this section
of the division; they are subject to very severe frosts and cold, in conse-
quence of which fruit trees are not plentiful, as the fruit very rarely
grows to perfection. Four samples were taken to represent the Worcester
Field-cornetcy ; these were Nos. 18, 19, 20, and 21. No. 18 was taken
from the farm Hartebeest Rivier, and typifies a clayey soil about two feet
deep. Here both vines and oats are reported to do well. No. 19 waa
collected at Tweefontein, from one of the river terraces east of Worcester,
The recent gravels and Enon formation which prevail render the soil rather
coarse and sandy, but with grain it is stated to give good results, although
here as elsewhere, rust has been known to damage the crop very consider-
ably. As a rule, however, crops standing on this soil do not suffer from
rust. No. 20 is a stiff red " Karroo " soil from Zeekoegat, and is regarded
as suitable for vines. No. 21, on the other hand, taken from the opposite
extreme of the Field-cornetcy, at Wyzers Drift — an area covered by the
above-named gravels — is a sandy alluvial soil. Chemically it is very poor,
although said to answer fairly well when planted with vines.

The soils of the Field-cornetcy Over Hex River are represented in
Noa. 13, 14, 15, 16, and 17. Sample No. 13 was taken on the farm Nonna
— a black alluvial soil — somewhat sandy — typical of much of the soil
found along the Nonnas River. The subsoil is yellow and rather sandy,
but where deep it proves very suitable for fruit culture; grain also
appears to thrive on it. The farm Boven Kloppers Bosch is represented
by soil No. 14. The local opinion was that it is a very rich soil — perhaps
too rich for oats, it was said. Vines and fruit trees, as well as barley, are
very successfully grown here. No. 15 is a sample of a dark fertile " vlei "
soil used as a vineyard on the farm Wilge Rivier. This soil, too, is con-
sidered rather rich for oats. At Nooitgedacht a somewhat sandy brown-
ish alluvial soil, No. 16, was taken, representing the third variety of soil
found in these parts; the first variety being the Karroo soil, and the
second, the dark " vlei " ground. At this place oats, wheat, vines, and
potatoes — especially the latter — are stated to thrive very well. The suc-
cessful culture of potatoes can be understood, as the soil is sufficiently
sandy for the purpose, and potash— the kind of plant food most needed
by these tuberous plants — predominates. No. 17 represents a brown soil,

150

less sandy than the last, from the Dwyka formation on the farm Aan de
Doorn Rivier. It is rather brackish, and becomes stiffer as the depth
increases. At about two feet depth lime is found, overlying sand. When
manured with guano, the soil represented by this sample has given good
results as a vineyard. The veld around is sweet. From this sample an
idea is obtained of the soils lying between the Nuy, Hex, and Nonnas
Rivers.

The next tract of country visited was the Field Cornetcy Voorste
Bosjesveld. Here samples 8, 9, 10, 11, and 12 were collected. In No. 8
is represented the soil of medium quality found on the farm D© Dooms.
This soil rests upon a clay subsoil, and, planted as a vineyard, is- stated
to yield good results. The farm Stettin contributed sample No. 9. The
land whence it was taken gave very discouraging results when put under
lucerne. The soil is somewhat clayey, overlying a white sandstone. This
sample was expected to compare unfavourably with many of the pre^
vious samples, and the chemical results justify the anticipation. No. 10
was collected on the farm De Hoek, and represents a " Karroo" soil with
a clay subsoil. Together with No. 17, it represents the soils of the
Dwyka series in this division; these two soils have been found
to contain more available lime than any others in the western
part of the Division. No. 11 represents a sandy soil from Wagen-
boom, said to be rather fertile; it is a primary sandstone soil, and very
coarse grained at that; both practical experience and chemical analysis
generally find such soils to be very poor : the reputed fertility of this one
is therefore somewhat enigmatical. It was intended to devote the lands
represented by it to the cultivation of vines. At Matjes Kloof, No. 12, a
sample of what is termed " Karroo " soil was taken. Soil of this type is
being used for wheat. The " veld " in the vicinity is sweet, but the soil,
which lies on the Witteberg formation, is of the poor and sandy type
usually associated with the quartzites of the Witteberg series.

In the Field Cornetcy Wagenbooms Rivier, Nos. 1, 2, 3, and 4 were
collected. This area forms, as it were, the root of the tongue of Malmes-
bury slate which protrudes between the two sandstone ranges into the
Tulbagh Division.* Here, as there,' it is flanked on either side by sand-
stone mountains, and, except for the Wilge Rivier soils, the results of the
chemical analyses are similar. No. 1 is a very productive alluvial soil of
good depth, taken on the farm Eendracht. Both vines and grain answer
well on this soil. Nos. 2 and 3 come from Wilge Rivier, where No. 2
represents the surface soil about one foot deep, and No. 3 the underlying
subsoil, a stiff yellow clay. This farm is famous for its vines; the vines
endure drought well, and are planted so as to get their roots into the clay
subsoil. On Mr. S. F. du Toit's farm Breede Rivier is found the soil re-
presented by No. 4 — a dark clayey soil, overlying a yellow clay. It is
said to give good results with cereals, but Mr. Du Toit's intention was to
Utilise it as a vineyard.

In the Field Cornetcy Goudini, three soils were sampled. No. 5 was
collected on the farm Slanghoek, and represents a dark clayey soil used for
vines, and overlying a red clay subsoil. No. 6 was taken from a dark,
somewhat sandy, alluvial soil, resting on a pot-clay subsoil, and representa-
tive of the land used ag a vineyard at Groot Eiland. No. 7 was taken
from the farm Klippe Drift, where a clay subsoil is brought to the surface
during tilling.

* See page 135.

No.

Field Cornetcy.

151

(Privately collected.)
Farm or place.

31. Achter Hex River

32.

33.

34.

Collector.

J. P. de Waal.

These additional soils were collected at various points on the pro-
perty of the Cape Orchard Company, for the purpose of determining the
proportions of available lime contained. The results of these determina-
tions, together with those of the chemical analyses of the other Worcester
soils above described, will be found in the table below : —

(Method I.)

Percent, of
Field

Percentage of Soil sifted throug-h 1 mm.
Sieve.

Percentage of Soil sifted
through £ mm. Sieve.

Sample.

Phos-

No.

Fine earth.

Water.

Organic

Chlorine.

Nitrogen.

Lime.

Potash.

phoric

matter.

oxide.

1.

100

2-09

6-62

•0120

•158

•066

•037

•069

2.

53'7

2-19

7-55

•0099

•266

•150

•147

•093

3.

99'4

2-53

7-15

•0290

•071

•156

•015

•025

4.

83-4

1-34

3'77

•0074

•116

•086

•027

•034

5.

98-5

2-92

10-47

•0163

•203

•022

•028

•136

6.

76-3

1-27

4-30

•0255

•144

•066

•022

•171

7.

79-5

•96

2-84

•0142

•071

•024

•018

•044

8.

74-1

•62

2-48

•0130

•116

•048

•085

•052

9.

75-2

•96

3-38

•0064

•116

•044

•067

•047

10.

84-1

3-80

6-85

•0049

•137

1-640

•157

•065

11.

33-2

•17

•44

•0078

•036

•038

•015

•022

12.

70-7

•93

1-53

•0049

•071

•088

•041

•033

13.

75-5

1-45

3-64

•0219

•158

•156

•068

•063

14.

79-1

T90

5-49

•0099

•172

•228

•084

•081

15.

87'6

2-55

6-45

•0810

•217

•262

•226

•096

16.

68-1

•44

1-13

•0237

•050

•060

•114

•035

17.

86-6

1-08

2-54

•0081

•100

•386

•119

•052

18.

70-9

•79

3'47

•0095

•081

•016

•091

•051

19.

70-0

•78

2-98

•0262

•094

•038

•158

•078

20.

89-6

1-59

4-04

•0198

•144

•260

•195

•089

21.

70-2

•56

2-00

•0163

•073

•024

•on

•034

22.

87-2

1-15

4-64

•0042

•112

•102

•077

•074

23.

88-6

•69

3-03

•0113

•091

•086

•048

•033

24.

81-6

•69

3-14

•0028

•091

•050

•113

•049

25.

71-7

•96

4-67

•0060

•105

•136

•164

•086

26.

79-0

•98

3-89

•0021

•112

•442

•139

•077

27.

52-2

•09

•57

•0046

•014

•016

•034

•024

28.

93-8

•57

3-01

•0046

•050

•022

•105

•054

29.

70-6

1-64

6-65

•0424

•182

•162

•324

•156

30.

77'0

•39

2-01

•0042

•057

•024

•077

•052

31.

79-4

.

—

•008

—

—

32.

87-7

—

•006

—

—

33.

82-9

—

—

•010

—

—

34.

70-3

—

—

—

—

•018

—

—

152

The results of these analyses of Worcester soils show that, as a rule,
there ia a fair amount of lime, potash, and phosphoric oxide in the soil,
and also a good proportion of nitrogen. As to the separate Field Cor-
tietcies, generalising with all due caution from these few results, we may
say that there are fair quantities of all three inorganic plant food con-
stituents in the Wagenbooms River district, which is, moreover, rich in
nitrogen. At Groudini there are good quantities of nitrogen and phos-
phates, but the soil is poor in lime and potash. At Voorste Bosjesveld,
the phosphates are deficient, but there are fair proportions of potash and
nitrogen and the lime is normal in amount. The Over Hex River area
shows a normal jamount of potash, with lime and phosphoric oxide in fair
quantity, and a good, percentage of nitrogen. The potash content is also
normal in the soils of the Worcester Field Cornetcy, and the phosphoric
oxide and nitrogen fair, but lime is, on the whole, rather poor. The Field
Cornetcy of Achter Hex River contains in its soils a fair quantity of all
four constituents. Of all the soils examined in this division, only in two
cases may it be said that nitrogen is absolutely deficient, namely, in the
soils from Wagenboom and Karbonaatjes Kraal. These two soils are
most characteristic Table Mountain sandstone soils ; they are coarser in
texture and less retentive of moisture than any other soil of the series;
they contain smaller proportions of nitrogen, and of organic material
generally; they are the poorest in phosphates, and amongst the poorest in
lime and potash.

153
PART IV.— COMPARISON OF EXTRACTION METHODS.

Although there have been, during these investigations, but very few
opportunities of comparing the relative solvent powers of different extrac-
tion media upon one and the same soil, it will still be possible to draw
some conclusions from the analyses of soils which may well be assumed to
be of similar chemical composition.

Of the different methods applied in the course of the investigations,
two, i.e., Methods III. and IV., may be expected to result in a greater sol-
vent action than that adopted as the standard method, and one, Method
v., would obviously exert a lesser dissolving power. For these reasons' it
seemed needless to seek by experiment for a confirmation of what was so
latent; at the same time the question still remained open in what ratio
the solvent action of the one exceeded that of the other, and whether, in
fact, there could be said to be a definite ratio at all. Less evident, how-
ever, did it appear whether there would be any appreciable difference be-
tween Methods I. and II., and, although here again no comparative tests
have been made on any particular soil, yet soils of such similar type have
been treated, some by the one and some by the other method, as to
afford a fairly safe means of comparing the two methods.

The only instance of comparison between Method IV. (Hilgard's) and
any other, is that afforded by the four soils 21, 22, 23, and 24 from the
experiment station at Robertson. These soils were extracted by Methods
I., IV. and V., and in one case also a determination of lime in an aqueous
extract of the soil was made. Comparing Methods I. and IV. we find
that, with regard to lime, Method IV. extracted practically 110 more than
what was extracted by Method I., in fact, it may be said that in the case
of No. 21 and also of No. 24, which was a patch in the area represented
by No. 22 — the proportions extracted were really identical. Apparently
in such cases Method I. extracted the entire reserve stock of lime, using
the word " reserve " as Prof. Hilgard uses it. The lime, it must be noted,
occura here largely as carbonate. As far as concerns potash, the quan-
tities extracted by Method IV. were, in soils 21 and 22 about five times
those removed by Method I., and in No. 23 nearly four times as much,
while from the brack soil, No. 24, Method IV. took out nearly seven
times as much potash as Method I. removed. From the results with re-
gard to potash obtained by Method IV. one might be disposed to infer
that the soil was very well supplied with potash in an available form,
but practical experiments with cereals on the farm have shown that
to supply the soil with phosphates was unavailing unless potash was also
given, indicating that although phosphatic fertilisers may increase the
stock of phosphates up to the limits of adequacy, yet the law of the
minimum still operated in respect to potash. It also illustrates the re-
mark already made, that in this Colony results arrived at by Method IV.
rather overstate the stock of available plant food in the soil.*

Turning now to Method III. we may expect to find an even wider
divergence from the results of Method I., inasmuch as the solvent action
on the soil is of necessity still more energetic. No comparative tests
between these two methods on absolutely identical samples of soil have
been made, but Nos. 1, 2, and 3 of the George Division soils, are, for
all practical purposes, the same as Nos. 20, 21 and 22 in the same list:
the former three were extracted by Method III., and, as pointed out when
dealing with the soils of that Division, yielded much higher figures than
the latter three, which were treated by Method I.f The soils of George,

See page 12. f See page E9.

154

unlike those of Robertson, are acid in character, or at least in tendency,
and the lime is probably present in a different state of combination :
at all events, the boiling and evaporation of the soil, according to Method
III., with concentrated Hydrochloric and Nitric acids, yielded from twice
to six times as much lime as agitation with diluted Hydrochloric acid
in the cold, as prescribed by Method I. In the case of potash the differ-
ence was even more marked, Method III. extracting from six to eleven
times as much as Method I. It will be noticed that the difference between
the two methods was greatest, not only with regard to lime, but also in
respect to potash, in the third members of the two sets of soils, namely
Nos. 3 and 22, and it was least in the second, i.e., Nos. 2 and 21.

If, for the reasons already given, Method I. is locally preferable to
Method IV., much more then is it to be preferred to Method III.

In addition to the three George soils, most of the samples taken in
the Mossel Bay and Riversdale Divisions were extracted according to
Method III. No soils from the latter two divisions have been examined
by Method I., but they may be expected to yield results not widely differ-
ing from and rather below those of the Swellendam soils. The means
of comparing the methods in these areas are not very full, the only Swel-
lendam soils examined by Method I. being those of the Tradouw, in
which the average proportions of lime so obtained fall greatly below those
yielded by Method III. for the Riversdale and Mossel Bay soils. In
this respect, as also in regard to potash, the latter show results* very
similar to those obtained from the George soils by the same method, and
so the general conclusion may be drawn that the method yields results
too high to be sufficiently comparable with those attained by natural
methods, and hence that considerable diminutions will have to be made
in its figures if the chemical aspects of soil fertility are to be judged
by them. The proportions of lime and potash, for instance, yielded by
No*. 5 of the Mossel Bay Division soils, poor and acid as it is reputed to
be, would, if obtained by Method I., have gained for it the declaration
of being a soil adequately provided with these plant food elements, and
practical experience would have dissented from such a view. A similar
observation may be made regarding Nos. 16 and 17 of the Mossel Bay
soils, and Nos. 18 and 24 of those from Riversdale.

The obvious reasons for discarding methods which were inherently
more energetic than that eventually adopted did not seem to apply to
Method II., and hence- it was given a somewhat extensive trial. It is
difficult to conceive why it should have yielded results frequently so
much higher than those of Method I. In both cases the extraction pro-
cess has been conducted, at the ordinary temperature, with almost similar
proportions of acid to soil, the acid used being, in Method I., approxi-
mately of 23J per cent, strength (M15 sp. gr.) and 25 per cent. (1'126
sp. gr.) in Method II. : in fact this, and the sifting of the soil through a
3 m.m. instead of a J m.m. sieve are the only apparent reasons for the
higher proportions both of potash and lime obtained from the soils by the
latter method.

In the Witteberg soils of the Albany Division, for instance, an
average percentage of '1 of lime was obtained by Method II. Only in
soils of fairly good fertility has this proportion been yielded when the
first method was made use of. In the sandstone soils of Humansdorp,
too, such a result was noticed.

The high results obtained in some cases by Method II. are noticeable
in the soils from the Bredasdorp Division, practically all of which were
examined by this method (the single exception analysed by Method I.
yielding lower results) and in those from the Caledon Division, although
it is to be observed that even by Method II. the Caledon soils show up

155

badly. In the Cape Division, too, it will be seen that the analyses by
Method II. give much higher results on the whole than those by Method I,
Clanwilliam and Knysna are Divisions where the Table Mountain sand-
stone series predominates; and yet here too the second method of extrac-
tion took considerably more lime and potash from the samples subjected
to this treatment than could have been done by crops. In the sandy
soils of Eerste River in the Stellenbosch Division this difference is also
visible.

In the Hanover soils may be noted the large quantities of lime ex-
tracted by the second method from soils belonging to the Karroo system.
No analyses of any soils from this area have been made by the first
method, but where that method was applied to soils of more or less
the same nature, for instance those from the neighbouring districts of
Aliwal North, Albert, Graaff-Reinet, Richmond, and Colesberg, more
limited proportions of lime were extracted. In the Division of Steyns-
burg, where a number of soils were analysed by both these methods, a
similar disproportion with regard to lime will be noticed.

Something of the same kind may be observed in respect to the soils
of the Robertson Division, where, however, the geological formation be-
longs to a different system. The Robertson soils analysed by Method I.
showed a distinctly lower average of lime.

The amounts of the phosphoric oxide extracted by Method II. do not
differ materially from the proportions extracted in cases where Methods
I., III. and IV. were applied : in these latter cases, it will be noted, the
phosphoric oxide was not determined in the hydrochloric acid extract,
but was extracted by a special process.

In one case (a granite soil from Kuiken Vallei, Stellenbosch Divi-
sion) the extraction process was varied, boiling Hydrochloric acid being
employed: apparently, however, no larger quantities of plant food con-
stituente are thus removed from the soil than by Methods II. and III.,
but it is difficult to avoid coming to the conclusion that the method
adlopted as a standard, fitting in as it does so well with the opinions of
practical men, is, on the whole, a much safer guide.

Of course, it is evident that, under certain circumstances, Methods
II., III. or IV. may not extract from the soil any more of a certain plant
food constituent than Method I.; but that may be because Method I.
extracts all that there is to extract. This may be expressed by saying
that in such a case probably all of that plant food constituent in the
soil falls easily within the limits of what has been termed grade II.* or
the reserve stock.

The poor Witteberg soils of Albany, which yielded so small amounts
of phosphoric oxide to Hydrochloric acid, naturally could not be expected
to show to advantage by the citric acid method, and the results thus
obtained are of course exceedingly low, but one of the three soils, which
gave a fair proportion of phosphoric oxide on extraction by Method III.,
yielded '026 by Method V., in other words the result was> well over Dyer's
proposed minimum limit of '010 per cent.f In one of the soils collected
at Elsenburg, Stellenbosch Division — both potash and phosphoric oxide
showed poorly — only '024 per cent in each case — by Method I. : in this
instance Method V. yielded '017 per cent of potash, but only '0036 per
cent, of phosphoric oxide, while extraction with water gave only "0013
per cent, of the latter. The soil was, according to both methods, clearly
lacking in both these two elements of plant food. The three soils from
the Herbert Division extracted by Methods I. and V. afford further in-
stance of mutual confirmation. On the basis of Method I., soil No. 1

* See page 9. t See page 17.

156

has a fair reserve of potash and phosphates, but not much to spare: No.
2 has about double that quantity in reserve, and is therefore satisfactorily
supplied: No. 3 is just on the border of potash-poverty, but has a
moderate supply of phosphoric oxide. Judging, by Method V., of the
immediate availability of these constituents, No. 2 again is satisfactory
both in potash and lime, No. 1 has about half the proportiona of each,
and No. 3 is particularly weak in phosphates. In each of these three
soils the proportions of potash extracted by Method V. was about one-
sixth that extracted by Method I., and in two cases out of the three the
latter method gave about three times as much phosphoric oxide as the
former.

Of the four Robertson soils, Nos. 21, 22, 23 and 24, No. 23, which
proved the best when examined by Method I., also turned out the best
when extracted by Method V., but all of them are lacking in reserve of
phosphates when judged on the basis of the former method, and in im-
mediately available phosphates on the basis of the latter method.
Method V. extracted from one-sixth to one-third as much potash, and from
one-sixth to one-twelfth as much phosphoric oxide as the standard method.
The lime extracted by Method V. was practically the same in amount
as that taken out by Method I., indeed, strange though it may seem at
first sight, in two cases, Nos. 21 and 22, the citric acid method extracted
slightly more lime than the Hydrochloric acid method. This is ex-
plained by the fact that the lime in the soil consists chiefly of carbonate
and so dissolves readily in either acid, and that the soil sifted through
the 3 m.m. sieve, for the purpose of Method V., would contain a larger
proportion of the coarser grades than the Jm.m. product used for Method
I., and hence more calcium carbonate, since the coarser particles in this
case consisted almost wholly of that material.

157
PART V.— GEOLOGICAL RELATIONS AND PLANT FOOD.

When these soil investigations were commenced the work of the Geo-
logical Commission had not advanced to the stage it has now attained,
and the results of that work were not available, as they are at present,
for the purposes of our own investigation. The Geological survey of the
Colony cannot but be of great worth to the scientific agriculturist, tihe
more so when supplemented by investigations such as these; in fact, it is
not too much to say that, in great part, it becomes thus fully valuable
only when so supplemented. The detailed and instructive maps issued
by the Geological Commission, showing, in very many cases, the
boundaries of the farms surveyed, cover largely the ground traversed by
the chemical staff, and are comparable with the maps issued by my own
office, in connection with which it has ever been the endeavour to mark,
as accurately as possible, the farm boundaries and all localities whence
samples of soil have been collected. Hence the soils analysed can often
be assigned to their proper geological formations, and deductions can be
drawn accordingly. Facilities were thus afforded whereby it became pos-
sible to arrange the figures in the subjoined tables in classified lists.

Before any further reference is made to these figures, the extreme
difficulty of obtaining samples of soils typical of definite geological forma-
tions must be mentioned. It does not need much discernment to classify as
a Table Mountain series soil one taken from the top of Table Mountain, or
as a Malmesbury series soil one from the upper slopes of Lion's Rump.
But when a valley composed of beds of the Bokkeveld series is> flanked
by sandstone mountains, it becomes less easy to predicate to what extent
each has influenced the chemical nature of the soil ; still more complicated
is the problem when dealing with such districts as Robertson, where, in
parts, quite a large number of rocks contribute to the formation of the
soil.

Had the samples of soils examined been so selected as specially to
typify certain definite geological series, it is highly probable that ere now
much more would have been learnt regarding the nature — from the
farmer's standpoint — of the soils derived from each of these series; but
circumstances have all along rendered any such system of collection im-
practicable; indeed, geographical rather than geological considerations
perforce ruled the selection of areas to be investigated.

In spite of the fact that conditions were not propitious for the selec-
tion of typical samples, it has been possible to sort out from the many
soils that have been analysed some to typify various geological forma-
tions, and the chemical composition of the soils so sorted out on the
whole bear out the reasonableness of the classification. For the purpose
of these comparisons only analyses conducted by the standard method
adopted in our laboratories have been made use of.

I propose, first of all, to consider the soils derived from the geolo-
gically oldest rocks, and to follow the upward sequence thence, as circum-
stances permit, ending with the superficial deposits.

Beginning, then, with the pre-Cape rocks, as the Geological Com-
mission has termed them, which underlie the Table Mountain and corre-
lated series, the soils derived from the Malmesbury series require first
notice. No analyses, performed by the standard method, of sufficiently
typical soils collected within the Division which gives its name to this
geological series, are available, but from the Malmesbury beds of the
Cape, Paarl, and Stellenbosch Divisions, the sixteen soils enumerated
below were taken, in such places as to be practically representative. The
series of rocks from which these are derived consists mostly of hard, close
clay slates; the chemical composition of the latter is, for the most part,

158

silicate of alumina, and, therefore, practically void, not only of actual
plant food, but even of plant food constituents.

Attention is directed chiefly to the percentage proportions of lime,
potash, and phosphoric oxide in the soils, but there are also incorporated
in the list the percentages of moisture retained in the air-dried soils : this
is expressed in terms of the soil sifted through a 1 mm. sieve, while the
determinations of plant food are calculated upon the fine earth. The
percentages of this fine earth in the sample, as collected in the field and
air-dried, are also given below.

Division.
Cape.

Paarl.

Stellenbosch.

Farm.

Hooge Kraal.

»

Matjes Kuil.
Groenfontein.
Kuilenberg.

Lange Bug.
Knolle Vallei.

»

Driefontein.
Burgers Drift.
Dryvers Vallei.
Lange Hoogte.
Elsenburg.

Nooitgedacht.

I. MALMESBURY SERIES.
Fine earth. Water.

821
82-4
51-0
81-2
53-6
551
65-3
79-6
68-7
50-6
60-3
51-8
55-8
80-2
57-7
92-2

1-78

1-33

•29

•67

1-13

1-72

Til

1-72

•32

•52

1-20

1-48

•63

•95

•63

•81

Lime.

•608
•107
•024
•012
•072
•100
•034
•098
•012
•022
•026
•014
•010
•065
•044
•008

Phosphoric
Potayh. oxide.

•887

•046

•551

•035

•010

•017

•029

•027

•045

•061

•073

•055

•041

•046

•044

•056

•013

•010

•048

•019

•075

•054

•039

•055

•042

•027

•024

•024

•024

•050

•039

•046

These Malmesbury series soils, it will be noticed, exhibit, almost
throughout, an all-round poverty in plankfood. The first on the list,
however (which is No. 33 of the Cape Division soils, whereof some details
were given on page 43), stands out markedly from the rest
on account of its higher plant-food content. The pebbles associated with
this soil point to its being derived from a rock of almost uniformly dull
greenish grey tint, and differ clearly from the various coloured fragments
found in connection with the next soil (No. 34), which includes brighter
red sandy ironstone* and white quartz. The unusual richness — for thia
part of the country — of the former soil makes it desirable to trace its geo-
logical horizon more closely, but for the present, owing to imperfect know-
legde of the sequence within the Malmesbury series, it must suffice to
record the existence therein of beds capable of producing soils of more
than ordinary quality. The percentages in the above sixteen soils
average as follows : —

Omitting the
All-inclusive, first two soils.

Lime

Potash

Phosphoric oxide

•079
•124
•039

•039
•039
•039

Commencing near the village of Robertson, a tongue of the Malmea-
bury beds runs at first westward for about thirty miles, and afterwards
northwards, as a depression between two sandstone ranges, for a distance
of nearly forty miles, forming the Tulbagh valley. (See sketch map at-
tached.) As will be pointed out later on, when dealing with the soils of
the Table Mountain series, the soils of the northern portion of this valley,
being practically encircled by mountains composed of Table Mountain

* Some of which contain flakes of white mica (muscovite) pointing to granitic origin.

159

sandstone, are bound to be poor all round. It would, in fact, be difficult
to understand whence they could derive any natural fertility : the root
of the tongale, nearer Worcester, is less dominated by the sandstone, and
the calcareous matrix of the Enon conglomerate, which prevails there,
exercises an undoubtedly advantageous effect upon the soil. The follow-
ing are the analyses of the soils collected from this strip, beginning with
the sandstone-encircled tip and working downwards : —

II. TONGUE OF MALMESBURY SERIES EXTENDING FROM THE VICINITV OF
ROBERTSON TO THE WINTERHOEK MOUNTAIN NORTH OF TULBAGH.

Phosphoric
Division. Farm. Fine earth. Water. Lime. Potash. oxide.

Tulbagh. Misgund. 73'4 "95 '044 '032 "023

Knolle Vallei. 61'4 T49 "068 '038 '047

98-7 5-75 "008 "030 '023

Kluitjes Kraal. 94'8 2'92 "070 "095 '070

Worcester. Breede River. 83'4 T34 '086 '027 '034

Wilge Eivier. 53'7 219 150 147 "093

Eendracht. 100 2'09 '066 '037 '069

„ Hartebeest Eivier. 70'9 '79 '016 '091 '051

Zeekoegat. 89'6 T59 '260 195 "089

Tweefontein. 70'0 "78 '038 158 '078

„ Nonna, 75'5 1'45 156 '068 '063

In many places the soils resulting from the rocks of the Malmesbury
series are diversified by the intrusion of granite. The consequent diver-
sity is more than becomes evident at first sight, for, although poverty in
plant food is practically inherent in the soils of the Malmesbury series,
it is otherwise with those derived from granite. The latter possess at
least the potentiality of fertility. This is owing to the fact, that, while
the Malmesbury rocks are deficient in plant food constituents, the intru-
sive granite is not, and thus the resulting granitic soils impart to the clay
slate soils, when mixed with them, not only plant food constituents, pro-
portionate to the relative quantities in which the two classes of soil are
mixed, but also actual plant food, proportionate to the amount of decom-
position that the granite has undergone. Where the granite has not been
sufficiently decomposed, the elements of fertility contained in its consti-
tuent minerals are not available for the plant. In other words, suffi-
ciency of plant food constituents may then be present in the soil while
there is lack of plant food. The diversification of the clays derived from
the Malmesbury slates, consequent upon the introduction of granite as a
soil-forming factor, is therefore dependent not only on the relative quan-
tities wherein these two classes of rock contribute to the formation of any
particular soil, but also on the stage which the mechanical disintegration
and chemical decomposition of the granite has reached.

These facts naturally add to the difficulty of laying down anything
like a typical composition even for unmixed granite soils. Of several
granite soils that had been analysed by Professor Hahn, in one — an allu-
vial soil — the constituent minerals of the granite had completely decom-
posed, and the results of the analysis of this soil were : —

Lime '281

Potash "151

Phosphoric oxide '172

160

The averages of the others, which were primary granitic soils, twenty
in number, were : —

Ume -037

Potash -025

Phosphoric oxide '014

It was not, let it be emphasised, that the latter contained less lime,
potash, and phosphoric oxide than the former, but that those substances
were largely present in the first soil in the condition already (see page 9)
described as grade II., whereas in the others the greater part
existed as plant food constituents of the third grade. With this explana-
tion there should be no cause for surprise that most of the nine primary
granite soils whose analyses follow are suown as deficient in mineral plant
food.

III. GRANITE SOILS. f

Phosphoric
Division. Farm. Fine earth. Water. Lime. Potash. oxide.

George. Diepkloof. 94'7 "97 "028 '034 '020

Malmesbury. Alexanderfontein. 720 T57 "146 '121 '061

„ 73'8 1-15 '072 '075 '072

„ 800 1-47 '096 '095 '073

72-2 I'Ol '014 -114 "049

680 1-09 '014 -045 "028

76-0 1-28 -014 -095 "055

Paarl. Vredehof. 61'5 "87 '034 '014 "055

Stellenbosch. Neethlingshof. 691 T26 '024 '023 '018

The averages of these soils work out as follows : —

Lime '049

Potash '069

Phosphoric oxide '048

The result of the partial decomposition of the felspar contained in
the granite is seen in the relatively larger percentage of potash when these
analyses are compared with those of the clay slate soils, and to such a>
cause may also be due the relatively large proportions of potash in the
two Hooge Kraal soils of the Malmesbury series.

More recent, geologically, than the slates of the Malmesbury series,
are what have been termed the Cango beds, in the Oudtshoorn Division.
They consist of quartzitic pebbles embedded in a slaty matrix, which at
some places becomes calcareous. It may be expected that the former
type will produce a comparatively poor soil, while a correspondingly
better class of soil will be produced where the calcareous matrix prevails.
Unfortunately, the sandstone range of the Zwartbergs so overshadows
the localities whence were collected the only three of our samples which
represent the area of the Cango beds, that the exact extent to which they
typify these beds caomot yet be arranged. The following are the three
soils referred to : —

IV. CANGO BEDS.

Phosphoric
Division. Farm. Fine earth. Water. Lime. Potash. oxide.

Oudtshoorn. Lemrick. 58'6 111 '068 194 123

Welgevonden. 76'0 T90 "084 144 '067

Matjes Eivier. 61' 6 315 7'460 "087 166

Nothing need be said to lay stress on the great difference between
the first two and the last of these three soils.

We have now to turn our attention to a class of soils of still more pro-
nounced calcareous type. As in the south-western corner of the Colony
the pre-Cape rocks comprise largely the Malmesbury clay-slates and shales,
so in the northern part we have the series of rocks which, over thirty
years ago, Stow associated with the name of the Campbell Rand,* and
into the composition of which limestone and dolomite enter largely. In
the soils representative of this series we would consequently expect to find
large quantities of lime. Here and there, however, the effects of the

* Seepage 143.

162

quartzites and sandstones of the superimposed Griquatown series, not to
mention the vast superficial deposits of sands and surface quartzites,
become apparent, and considerably dimmish the proportions of available
lime.

The following table shows the results of analyses of these soils : —

V. CAMPBELL RAND SOILS.

Division.
Vryburg.

Farm.
Kuruman.

Fine earth.

Water.

Lirne.

Phosphoric
Potash. oxide.

76-4

1-01

•026

•024

•024

100-

13-56

3-66

•030

•132

88-7

2-89

4-80

•084

•088

85'2

3-54

6-32

•059

•056

94-7

416

14-04

•028

•059

871

1-86

•278

•068

•026

89-6

•92

•062

•020

•013

Geluk.

Stowf distinguished between the ro'cks of this series and the olive
shales, as he termed them, deposited unconformably over the ancient
schistose rocks underlying the crystalline limestone of the Campbell
Band. The analyses given in the next table are those of soils which may
be taken ae representing this deposit of shales.

VI. QUARTZITES AND SHALES EAST OF CAMPBELL RAND (STOW'S OLIVE

SHALES).

Division.

Farm.

Hopetown. Fluitjes Kraal.
BarklyWest. H.V. 75.

Vryburg.

Ganzepan.
Salisbury.
Middle Park.
ii

Mogogong.

Banksdrift.

Fine earth. Water. Lime.

90-0
97-3
94-6
94'6
93'8
95'4
88-2
73-5
82-5
76'6
80-9
73-0

•73

•048

•66

•020

•63

•036

4.74

3-768

•54

•034

2-56

•162

•51

•010

1-04

•022

2'47

•130

1-22

•008

1-78

•054

•84

•006

Phosphoric
Potash. oxide.

•065

•020

•046

•013

•035

•014

•187

•101

•034

•009

•103

•055

•030

•014

•024

•045

•118

•059

•041

•045

•074

•040

•039

•032

• Although one or two of the soils enumerated in Table VI. are of
fairly good quality, nearly all of them show a deficiency in potash and
phosphoric oxide, while several lack lime as well. The Ganzepan soil
forms a remarkable exception, and its chemical richness is doubtless due
to the fact that it was collected at the very foot of a great doleritic out-
crop, from a valley whose brown loam is evidently largely derived from
the surrounding dolerites. For this reason it would be misleading to
include the soil referred to amongst the averages of the above shale-de-
rived soils, which are as follows : —

Lime '048

Potash '055

Phosphoric oxide '031

fGeol. Soc. Quarterly Journal, Dec. 1874.

163

The above table of shale soils has been taken out of its proper geo-
logical order because of the local association of these soils with those- of
the Campbell Hand. Strictly speaking, it should be considered along
•with the soils of the Dwyka series, of which these shales seem to form
part: chemically, however, the above soils differ greatly from the Dwyka
soils examined; to these reference will be made at a later stage.

Another geological formation which it will be convenient to allude to
here, but which belongs to a much earlier geological age, is that of the
Pniel Volcanic series, a group of rocks that make their appearance still
further east, i.e., on the opposite side of the shales to the Campbell Rand.
From this formation the soils enumerated in the following table were col-
lected :

VII. PNIEL VOLCANIC SERIES.

Division.

Farm.

Fine earth. Water.

Phosphoric
Lirne. Potash. oxide.

Barkly West. Brady's.

» ,5

Patrys Kraal.

JJ »

., Zwartputs.

Vryburg.

H.V. 67.
H.V. 63.
Luxmore.

92-2
84-4
83-4
93'4
98-3
96-8
96-6
97-6
85-2

•94
116
2-02
3-33

•79
2-02
6-82
6-03
2-95

•044
•020
•082
•240
•020
•082
1-037
•461
•220

•050
•112
•056
•123
•037
•056
•054
•027
•101

•031
•045
•038
•031
•022
•038
•195
•154
•063

The Pniel group underlies the limestones of the Campbell Rand
series, and, including as it does amygdaloids and diabase, it appears to be
capable of producing chemically rich soils, but only under the condition
already pointed out in conection with the granite soils, — mere disinte-
gration does not suffice ; it needs to be accompanied by chemical decom-
position in order to render the resulting soil fertile.

The following are average results of the analyses of the Pniel series
soils : —

Lime '245

Potash -068

Phosphoric oxide '069

Of the rocks of the Cape system, the lowest are those of the Table
Mountain series. Table Mountain sandstones and quartzites constitute
the greater part of the lofty mountain ranges which extend for hundreds
of -miles along the south-western part of the Colony : they tower above
all other rocks for very many miles around, and it is, therefore, a safe
assumption to make that the virgin soils, collected from these high alti-
tudes, are quite uninfluenced by any other geological formations. Fur-
thermore, the rocks cf the Table Mountain series consist of little else
than silica, namely, sandstones and quartzites, with occasional shales. It
may accordingly be anticipated that the soils thence derived will not be
of very complex chemical composition, but will, on the contrary, lack the
essential elements of plant food. Such we actually find to be the case.
The following is a table of analyses of 46 soils collected either directly
from areas formed of this sandstone, or from valleys so completely domi-
nated by ranges of the Table Mountain series as to render it practically
certain that no other ro^ck? could have appreciably contributed to their
composition.

164

VIII. TABLE MOUNTAIN SERIES SOILS.

Division.

Bredasdorp.
Caledon.

Ceres.

Farm .

The Hope.
Gloria.

Geelbeks Vlei.
Isaaks Rivier.
Riet Vallei.

George.

Knysna.

Ou tshoorn.
Pa rl.

Robertson.

Swellendam.
Tulbagh.
Union dale.

Worcester.

Fine earth.

Water.

Lime.

Phosphoric
Potash. oxide.

90-6

1-75

•064

•017

•045

73-5

•47

•010

•017

•037

64-3

•86

•006

•008

•028

52-5

•34

•008

•008

•015

81-6

•58

•044

•014

•009

1-01

•014

•09

•05

97-0

•36

•072

•015

•022

80-7

•37

•010

•015

•028

98-3

•27

•010

•012

•010

98-0

•56

•008

•017

•032

97-8

1-62

•034

•016

•022

98-2

1-41

•024

•010

•015

97-0

1-12

•026

•016

•031

66-8

2'78

•040

•041

•095

81-9

2-84

•030

•066

•068

96-0

1-17

•032

•012

•020

98-5

•86

•044

•on

•013

94-7

•97

•050

•014

•018

96-6

1-59

•074

•017

•031

95-6

3-84

•032

•028

•056

98-2

1-27

•028

•014

•046

97-0

•66

•028

•023

•010

98-0

1-20

•082

•030

•018

76-3

•40

•074

•035

•040

77-9

2-16

•010

•039

•042

70-4

1-40

•042

•087

•042

57-8

81

•008

•013

•038

65-1

•97

•008

•013

•022

56-9

T56

•006

•015

•017

86-3

1-41

•058

•019

•038

66-4

•81

•030

•027

•036

62-4

•59

013

•060

•045

60-3

•21

•010

•036

•010

76-8 .

•77

•024

•033

•023

24-4

•27

•008

•020

•019

57-1

1-78

•054

•050

•077

83-4

•71

•058

•030

•036

35-8

•76

•050

•062

•052

94-2

•74

•030

•038

•026

92-4

1-07

•040

•054

•044

98-5

2-92

•022

•028

•136

74-1

•62

•048

•085

•052

75-2

•96

•044

•067

•047

33-2

•17

•038

•015

•022

88-6

•69

•086

•048

•033

52-2

•09

•016

•034

•024

Outspan Reserve

Woodville.

Diep River.

Grootfontein.

Schoonberg.

Geelhoutboom.

Roode Kraal.

Balmoral.

Millwood.

Portland Heights.

Portland.

Ashby.

Holt Hill.

Saffraan River.

Keerweder.

M

La Dauphine.
La terre de Luque.

Zand Drift.
Concordia.

Rietvallei.

Tradouws Hoek.

Doornboom.

Gold Diggings.

Avontuur.

Misgund.

Klipheuvel.

Krakeel River.

Slanghoek.
,, De Dooms.

Stettin.

,, Wagenboom.

,, Vendutie Kraal.

,, Karbonaatjes Kraal.

It will be observed that in no single instance does the reserve stock
of either lime or potash rise as high as '1 per cent., and only in one case,
that of the soil from the farm Slanghoek, in the Worcester Division, does
the percentage of phosphoric oxide exceed '1. The average percentages
of the reserve plant food in these 46 sandstone soils are as follows : —

Lime '034

Potash -031

Phosphoric oxide '036

Lying comformably abo<ve the Table Mountain series are beds of what
is known as the Bokkeveld series. These consist of alternate layers of
shales and sandstones, the lower bands of the series being fossiliferous.
It would, perhaps, have been difficult to predict, without previous prac-
tical agricultural experience, what would be revealed by chemical analy-
sis of soils derived entirely from the Bokkeveld beds; nor is it easy to

tabulate — bearing in mind the circumstances tinder which the samples
were collected — a list of soils formed solely from the beds in question;
for, to the reasons which operate against this in other cases, there must
here be added this, that the Bokkeveld beds occur mostly in valleys, and the
resulting soils are, therefore, as a general rule, diluted, or otherwise in-
fluenced, by the disintegration and decomposition products of the rocks
composing the surrounding hills and mountains. Hence, there are very
special difficulties involved in sorting out many Bokkeveld soils that could
really be called typical, from the samples that have been dealt with in
our laboratories. I have, however, made a choice of eighteen, which
may fairly be considered to represent solely Bokkeveld-derived soils; the
results of tho analyses of these are given in the following table : —

Division.

Ladismith.
Robertson.

Swellendara

Uniondale.
Worcester.

IX. BOKKEVELD SOILS.

Fa? m. Fine earth. Water. Lime.

Buff els fontein. 86'8

Harmonic I. 80'5

Harmonie II. 83'9

Het Kruis. 80'8

95-2
Baden. 49'3

94-7

62-7

81-0

72-8

80-3

89-0

57-9

93-2

81-7

71-7

79-0

70-6

Poortfontein.

Uitvlugt.

Doom Rivers Vallei.

»

Lemoens Hoek
Kamnassie Wagen Drift.
Wilgehouts Rivier.
Roode Heuvel.
Keurbosch Kloof.

Stink fontein.

1-66

•150

1-23

•122

1-07

•564

1-13

•114

1-49

•742

1-93

•808

1-28

•600

•31

•046

2-04

2-518

1-21

•042

1-18

•106

1-21

•170

1-54

•102

1-09

•048

•87

•098

•96

•136

•98

•442

T64

•162

Phosphoric

Potash.

oxide.

•274

•073

•153

•079

•095

•047

•250

•120

•202

•193

•331

•148

•247

•139

•176

•036

•416

•234

•262

•134

•272

•151

•197

•122

•182

•095

•212

•104

•255

•122

•164

•086

•139

•077

•324

•156

On comparing the above Tables VIII. and IX., it immediately
becomes plain how greatly the Bokkeveld soils differ from those of the
Table Mountain series. While the proportions of lime, potash and phos-
phoric oxide practically never exceed '1 per cent, in the sandstone soils,
they rarely fall below that figure in the typical soils of the Bokkeveld
series. The average percentages of plant food in these eighteen soils work
out as follows: —

Lime '387

Potash '231

Phosphoric oxide '118

It is difficult to say, in default of a previous thorough investigation
of soils which are known to bo productive in this Colony, what propor-
tions! of plant food may be considered to render a soil either chemically
rich or of normal fertility, but, basing our opinions on European and
American standards, we may affirm that, while the sandstone soils shew
an all-round poverty, our analyses reveal the average Bokkeveld soil to
be satisfactorily provided with all three of the essential inorganic
elements of plant food.

Just here we touch upon a somewhat curious feature, As far as can
be gathered, the soils in Table VIII. represent undoubted sandstone soils,
and those in Table IX. purely Bokkeveld soils, but in various localities a
number of soils have been taken from near the line of junction between
the Table Mountain and Bokkeveld series. In some of theso instances the

166

underlying rock belongs to the former series, in others to the latter, but
in both cases a peculiarity is noticeable in the composition of these
" junction " soils, the nature of which will become evident upon glancing
through the next table.

X. BOKKEVELD NEAR JUNCTION WITH TABLE MOUNTAIN SANDSTONE; OR

Vice versa.

i'ivision.

Ladismith.
Robertson.

Swellendam.

»
Worcester.

Farm.

Papenkuilsfontein.
Donkerkloof.

M

Harmonie I.
Het Kruis Pad.

Het Goed Geloof.
Barrydale.
De Dooms.
Klein Straat.
Ezeljacht.

Fine earth.

Water.

Linie.

60-5

•64

•056

90-1

1-08

•064

80-9

•65

•032

70-2

•98

•016

71-1

1-20

•033

64-7

•90

•050

62'5

2-00

•074

83-4

T16

•040

81-6

•69

•050

93'8

•57

•022

77-0

•39

•024

Phosphoric

Potash.

oxide.

•136

•042

•126

•061

•180

•068

•147

•050

•125

•079

•078

•074

•210

•201

•250

•096

•113

•049

•105

•054

•077

•052

Unfortunately, not more than eleven soils are available for the pur-
pose of illustrating the point involved. It will, however, be seen that,
while these eleven soils resemble those of the Table Mountain series in
regard to the amounts of lime and phosphoric oxide contained, they
approach the Bokkeveld soils in respect of their potash content. This
feature is, perhaps, better distinguished on taking the averages, which,
for Table X., are as follows : —

Lime

Potash ... .

Phosphoric oxide

•042
•141
•075

If these were simply Bokkeveld soils diluted by alluvial sand from
the Table Mountain series, one would expect a lowering of all three of
the plant food constituents in the same proportion; but, as we see, the
reduction of lime and phosphoric oxide is proportionately much greater
than that of the potash. It was puzzling at first to account for this, but
one explanation seems feasible. Dr. Corstorphine, in his second report
on the Geological Survey of the Colony,* remarks on the distinctive char-
acter of the lowest beds of the Bokkeveld series, which enables one with
little or no difficulty to draw the line of separation between it and the
Table Mountain series below. The basement beds of the Bokkeveld
series, he says, are felspathic and micaceous, and quite different from the
hard compact quartzite of the older series. I am not aware whether it
is potash felspar that is here alluded to, and our analyses, as before ob-
served, do not furnish data with regard to the total amounts of the plant
food constituents in the soil, but it seems just possible that in this fact
lies the explanation of the predominance of potash in what appear to be
primary soils collected near the common horizon of the two series. It
would certainly be of interest to investigate the subject more closely, so
as to confirm or disprove the hypothesis here suggested. For this pur-
pose samples of soil would have to be collected in such a manner as to
render it absolutely certain that they were primary soils representing the
vicinity of the line of junction; in the past the collection of soils hag
been carried on largely without regard for geological considerations. The
time has now arrived, I think, for greater precision in this respect.

* Ann. Kept. Geol. Commission, 1897, p. 16.

167

It may be pointed out just here that one of the soils classed as a true
Bokkeveld soil, viz., that from Poortfontein, in Table IX., bears a consider-
able resemblance in chemical composition to those enumerated in Table X.
Its potash, compared with the deficiency of lime and phosphoric oxide, is
high, but not so high as the general run of true Bokkeveld soils, and in
both lime and phosphoric oxide it is the poorest of the Bokkeveld soils.
It has been here classed among the latter because of its geographical posi-
tion, but it would appear almost certain, upon the basis of the reason
just stated, that it must have been derived from the basement beds of
the Bokkeveld series.

Above the Bokkeveld beds, and therefore geologically of more recent
date, are those of the Witteberg series. They consist of quartzites
associated with shales, and the soils thence derived would, therefore, be
expected to conform in chemical character with those resulting from the
older Table Mountain series. I regret that only four of the soils analysed
in our laboratories can be assigned with any clefmiteness to this formation.
The results of their analyses are as f ollo\vs : —

XI. WITTEBERG SOILS.

Phosphoric
Division. Farm. Fine earth. Water. Linie. Potash. oxide.

King William's Town Evelyn Valley. 97'5 6'26 '050 '082 '105

Ladismith. Elands Vallei. 85'3 '93 '036 '071 '068

Victoria East. Hogsback. 99'6 5'87 "028 '038 "055

Worcester. Matjes Kloof. 70'7 '93 '088 '041 '033

On account of their fewness, it would scarcely be permissible to com-
pare these four soils with the forty-six derived from the Table Mountain
series; but, as has already been remarked, it is one of my purposes to
point out the need of greater exertion, and, on that ground, it may not
be unfitting to emphasise paucity of information, so as to induce fuller
investigation. There are vast tracts of country dominated and influenced
by ranges of mountains built up of the Witteberg beds, and, if the investi-
gations proposed by me over a dozen years ago had been both embarked
upon and prosecuted with the alacrity fondly pictured at the start, it is
possible that the soils of extensive areas would ere now have been mapped
out in such a way as to shew, at all events in some respects, how they
answer chemically to the requirements of fertility.

These observations come in most suitably just at this stage, because
it was precisely in connection with the Witteberg soils that I was first
led to propose to the Government of the day that a systematic agricultural
survey of the Colony's soils should be taken in hand, impelled thereto by
the results of analyses I had made of soils from the Albany and Humans-
dorp Divisions, in both of which areas that bone disease in stock, vernacu-
larly termed " lamziekte " (literally " lame sickness ") prevailed.

The soils of a great part of the Albany Division, and of the adjoining
divisions of Bathurst and Willowmore, are apparently largely derived from
the quartzites of the Zuurberg range, a mountain chain entirely built up of
the Witteberg series; while almost throughout the whole of the Humans-
dorp Division the soil is the result solely of disintegration of the great ranges
of Table Mountain sandstone which extend along the south coast from
George, and eventually die out in the Zitzikama range. These, sandstones
and quartzites contain practically nothing capable of affording nutriment
to plants, and are almost devoid of the very mineral salts essential for the
production of bone material. It would be of more than interest to have
the areas that are covered by these poor soils mapped out as exactly as
possible, so as to shew their extent, and yet these all-important investiga-
tions have been absolutely dormant these seven or eight years past.

168

The average results of the above Witteberg soils may well be added
for comparison with those already given. They are as follows: —

Lime '051

Potash -058

Phosphoric oxide '065

These soils, as before remarked, may be expected to resemble those
from the similar sandstones of the Table Mountain series. There is, how-
ever, more clayey material in the Witteberg rocks, and hence — if the sub-
ject were more closely investigated — we should probably discover that the
soils derived from them do not exhibit quite the absolute poverty of the
Table Mountain sandstone soils. It is true that the above four analyses,
representing the Witteberg series, are along the line of this view, but they
are altogether too few in number to add any confirmation thereto.

The soils derived from the Dwyka series, which overlie the Witteberg
beds, lead us from the Cape into the Karroo system, where we meet with
an all-round richer type of soil than the average of the older system. Of
the soils which represent the Dwyka beds of shales and conglomerates we
have unfortunately examined only two that can be taken as in any way
typical, and both of these were collected within the Worcester Division.
Their analyses resulted as follows : —

XII. DWYKA SOILS.

Phosphoric
Division. Farm. Fine earth. Water. Linie. Potash, oxide.

Worcester. De Hoek. 841 3'80 1/640 157 '065

Aan de Doom River. 86'6 T08 '386 119 "052

We cannot, of course, generalise from the results of two analyses; all
that can here be said is that these two differ from those of the Table
Mountain and Witteberg sandstones, on the one hand, by shewing adequate
proportions of lime and potash, and from the Bokkeveld soils, on the other,
in their relatively low phosphoric oxide. The matrix of the rocks con-
stituting the conglomerates is composed of quite a variety of minerals —
amongst which felspar, augite and calcite are prominent — and ^would prob-
ably disintegrate into just such soils as are typified in the above two
analyses.*

Passing over the intervening rocks of the Karroo system, in respect
of which it would be difficult, for several reasons, to submit any analyses
of typical soils at the present stage, we have next to glance briefly at the
upper series of this system, namely, the Stormberg series, which we may
group with the Burghersdorp or uppermost beds of the Beaufort series just
below them. These consist of sandstones, shales and clays, interspersed
with calcareous rooks, and producing soils of a red or purplish colour. f I
do not think that we can assign any of these soils, that have hitherto
been analysed, as definitely to certain geological series, as in the case of
the older rocks, partly because many of the samples were not officially

* The Dwyka series consists of three members ; (1) the upper shales, (2) the boulder
beds, and (3) the lower shale3!: the resemblance to Stow's "olive shales" noted on an earlier
page (162) refers to the phale beds, which may be expected to yield soils differing markedly
from those resulting from the boulder beds with their calcareous matrix

f See page 89.

169

collected. At the same time a selection has been made of soils taken from
localities where the influence of the Stormberg and Burghersdorp forma-
tions was most likely to have been traced in the composition of the soil.
These are comprised in the following table: —

XIII. BURGHERSDORP BEDS AND STORMBERG SERIES.

Division.
Albert.

Farm.
Odendalstroom.

Aliwal North. Aliwal North.

Colesberg.

Steynsburg.

Wodehouse.

Oorlogspoort.

Zout Kuil.

Fine earth. Water.

Klerks Kraal.
Middel Water.
Van Vuurens Kraal.
De Boulogne.

94'6
95-2
99-5
99-4
98-0
9?7
91-1
92-7
93-8
97-6
981
94-7
94-8
92-9
9?8
97-5
91-1
98-0
.98-4
98-8
98-0
99-5
98-3
92-4

2-44
210
3-23
4-07
3-61
5-08
2-49
2-67
2-94
4-40
2-33
3-51
6-19
2-89
5-65
4-28
4-00
5-58
4-33
4-27
3-99
6-94
10-08
5-94

Lime.

•540
•424
•376
•276
•192
•036
•046
•094
•114
•506
•066
•216
•042
•448
•062
•062
•044
066
•238
•130
•188
•232
•852
•346

Phosphoric
Potash, oxide.

•048

•138

•051

•089

•079

•088

•123

•087

•128

•057

•123

•056

•156

•068

•165

•072

•183

•070

•203

•115

•184

•073

•136

•069

195

•031

•152

•038

•181

•089

•213

•121

•189

•097

•257

•093

•196

•013

•192 .

•037

•161

•080

•341

•074

•237

•069

•235

•145

The most remarkable feature about these soils is that they are all very
uniform in texture and fine in grain, as may be noticed on glancing down
the column headed " fine earth." The effect of the calcareous rocks is seen
in the relatively large proportions of lime in some of the soils; their
irregular occurrence in the very small percentages exhibited by some others.
There is, in fact, much less range in the potash than in the lime of these
soils.

The averages for these Burghersdorp and Stormberg soils are as
follows : —

Lime '233

Potash -172

Phosphoric oxide '078

The next class of soils that we have to deal with belongs to the Uiten-
hage series, and more particularly to the geological base of that series,
namely, the Enon conglomerate. This formation is of a more recent date
than any of those previously considered, and we here pass from the Karroo
to the Cretaceous system. The Uitenhage series consists of shales, clays
and sandstones, with limestone bands and marls. A fair proportion of lime

170

may, therefore, be expected in the soils to which these rocks give rise, and
in the following table this anticipation will be seen to prove correct : —

XIV. UlTENHAGE SERIES.

Phosphoric-
Division. Farm. Fine earth. Water. Linie. Potash, oxide.

Knysna. Ganze Vallei. 99'5 219 116 '054 '045

Witte Drift. 90'8 T60 112 '069 '044

Matjesfontein. 84'4 T33 112 "039 '078

Oudtshoorn. Vlakte Plaats. 89'5 2'06 1170 '299

Eiet Vallei. 56'0 '65 154 100 '074

Oudtshoorn. 69'9 3'73 100 149 "058

Blauwboschkuil. 91/5 1/97 "530 '211 134

Welbedacht. 91'0 '87 '606 144 124

Jan Fourie's Kraal. 95'9 "84 172 161 '063

Welgerust. 911 3/57 170 170 '063

Adjoining Buffels Vlei. 81'9 '44 "046 '098 '049-

Uitenhage. Upper Landdrost. S9'5 2'70 182 "292 '073

98-9 '79 156 197 '068

Gouvernements Be- 991 1/92 '718 '333 143
looning.

98-3 1-07 -078 108 "075

Klaas Kraal. 93'2 '94 '392 '239 116

100- 1-35 "592 -226 "086

98-4 '90 '066 178 '063

99-6 1'02 182 185 '092

Malmaison. 95'8 T22 "278 "276 "087

99-2 118 '338 '278 132

It is interesting to note how closely these Uitenhage soils resemble-
those of the preceding table in respect of their plant food content. In
each case the average is slightly higher than in the Burghersdorp and
Stormberg soils, namely : —

Lime '299

Potash 181

Phosphoric oxide '087

Of soils representative directly and solely of recent superficial deposits-
not many analyses have been made. Two samples, one of which is a type
of the blown sands that cover the Cape Flats, yielded the results given
below : —

XV. SAND DOWNS.

Phosphoric
Division Farm. Fine earth. Water. Lime Potash, oxide.

Knysna. Sidgefield. 96'4 '56 "090 '050 '051

Stellenbosch. Eerste Rivier. 981 '066 '010 '003

Both of these, it will be seen, are very poor in plant food generally.

Of quite a different type are the silts transported, some of them over
distances of hundreds of miles, by streams and rivers. The composition
of our river silts is far too vast a subject to discuss at this stage, but
it is one that calls for much close study and patient investigation. Silts
are proverbially fertile, a circumstance which is due, not only to their
chemical composition, but to their mechanical condition as well. To this
fact the Oudtshoorn Division, whose soils are so largely made up of the
silts and clays brought down from the Karroo by the Gamka and Olifants
Rivers, owes its productiveness. Amongst all the soils that have been
analysed in the Government Laboratories, none occupy a higher place than
those of the Oudtshoorn Division, and much of the rich soil of that area

171

is now being carried away into the sea by the Gouritz River, especially
at times of disastrous flood, although, as previously observed, some of it
is retained and deposited in the Mossel Bay and Riversdale Divisions as
the river passes through them o<n its seaward course.*

Eleven analyses of such river deposits have been made by means of
the standard method, with the results below given: —

XVI. ALLUVIAL SILTS AND RIVER DEPOSITS.

Division.
Gordonia.

Albert.

Prieska.

Kimberley.

Ladismith.

S\vellendam.

Knysna.

Port St. John's

River.
Orange.

Vaal.
Buff els.
Tomvs.

Farm.
Upington.

Odendalstroom.

Zeekoebaard.

Warrenton.

Buffelsvallei.

Zeekoegatsdrift.

Zevenfontein.
Keurbooms. Matjesfontein.
St. John's. Isinuka.

Fine

earth.

Water,

100

3'23

100

1-68

100

3'46

100

5-99

91-5

2'50

88-2

6-29

65-0

•93

97-5

•92

97-2

•64

971

3-05

97-2

2-73

Phosphoric
Lime. Potash, oxide.

•582
•400
1-372
1-444
•278
•346
•332
•344
•220
•544
•520

•093
•065
•042
•473
•108
•094
•166
•187
•146
•105
•171

•061
•052
•149
•221
•074
•070
•106
•097
•082
•110
•137

They show considerable variety, both, in texture and in chemical com-
position. The proportions of lime, potash, and phosphoric oxide average
as follows : —

Lime

Potash

Phosphoric oxide

•584
•153
•106

The following is a summary of the average results obtained from the
various types of soil reviewed. It may serve to bring out more clearly the
characteristic differences in respect of plant food. The low water-retaining
capacity of the soils of the Malmesbury and Table Mountain series will be
noted, as also the large amounts of moisture retained by the Upper Karroo
soils of the Burghersdorp-Stormsberg type: —

XVII. AVERAGES.

Geological formation.

No. of soils Phosphoric-

analysed. Water. Lime. Potash. oxide.

Pre-Cape rocks

Cape System

Karroo System

Cretaceous System
Recent deposits :

Malmesbury Series. 16 T02 '079 '124 '039

Granite. 9 119 '049 '069 '048

Campbell Rand Series. 7 3'99 4169 '048 '057

Pniel Series. 9 2'90 '246 '068 '069

Olive shales. 12 T48 '048 '055 '031

Table Mountain Series. 46 1'08 '034 '031 '036

Bokkeveld Series. 18 T27 '387 . '231 '118

Common horizon of Table 11 '93 '042 141 '075
Mountain and Bokke-
veld Series.

Witteberg Series. 4 3'49 '051 '058 '065

Dwyka Series. 2 2"44 T013 138 '059

Burghersdorp Beds and 24 2'29 '233 172 '078

Stormberg Series.

Uitenhage Series. 21 T44 '299 181 '087

Sand Downs. 2 — '078 '030 '027

Transported Silts and River 11 2'86 '584 153 106

deposits.

*See pages 91, 96, 108, and 109.

172
PART VI.— ALKALINITY OF SOILS.

Descending rain constantly carries clay particles from the surface
down to the subsoil, and so enriches the latter with a supply of the
available plant food which, as already observed, is generally contained in
the finer soil grades, i.e., the silt and clay. On the other hand, surface
evaporation, causing an upward movement of the soil water, brings to the
upper soil layers a constant supply of readily available plant food. Bearing
in mind, then, the references made on a former page* to the different grades
of plant food, it may be deduced, that the subsoil is constantly being sup-
plied from the surface with plant food constituents of grade II., while the
surface soil receives an equally continuous increment from below of the
plant food constituents of grade I. The former of these two processes takes
place more especially during the wet season, while the latter becomes
more marked during the dry, but in arid regions this increase of salts in
the surface soil may proceed so far, and result in an accumulation of salts
of such a nature, as to cause distinct injury to plant growth.

The extent to which these salts have accumulated in the surface soil
can be definitely estimated only by chemical analysis, f but such analysis
is from its very object of totally different nature from that which investi-
gates the proportions of plant food or of plant food constituents present,
for we are now dealing not with beneficial, but with harmful substances
in the soil.

Thus the agricultural chemical analysis of soils, although as a rule
confined to ascertaining whether they contain a sufficiency of plant food,
may at times have to proceed beyond this limit, and determine whether
there be not an excess of plant poison. This generally resolves itself into
an investigation of the brackness or alkalinity of the soil.

What is generally called brack or alkali in soil consists in the presence
of excessive quantities of certain sodium salts. These, under favourable
conditions, dissolve in the rain-water and drain out of the soil, but they
accumulate and act injuriously where circumstances are unfavourable to
their removal by natural drainage. During a rainy season the salts are
carried to varying depths, according to the penetrating capacity of the
rain-water. The surface soil is thus left free from injurious salts. When
dry weather sets in, however, the soil water rises to the surface as oil rises
in a wick by capillary attraction, carrying the noxious salts with it from
below. The water evaporates from the surface soil, leaving behind the
salts as a white incrustation. There is, hence, a constant downward
passage of these injurious alkaline salts during rain, and an equally con-
stant upward movement during dry weather.

It will be easily understood that irrigation will tend to increase the
accumulation of alkaline salts in the surface layers of a soil, seeing that
the irrigation water will probably penetrate to greater depths than the
rain, and carry to the surface larger quantities of salts. The danger is
naturally augmented when the water used for irrigating is itself alkaline.

Hence, when adaptability of any tract of country for irrigation has
to be pronounced upon, two chemical problems have to be decided:
Firstly, does the soil contain any constituent which may render it brack
or otherwise unproductive; and secondly, assuming a satisfactory answer
to the first point, is the land sufficiently provided with the necessary
components of the food of plants to make farming profitable?

* See page 9.

t For the present purpose this term may be considered to include the electro-
lytic process of the United States Chemists.

173

When the amount of plant food in a soil is the only point at issue,
one needs not to probe to a greater depth than the first foot or eighteen
inches, but, it is obvious from what has been stated above, that in order
to ascertain not only whether a soil is at present unproductive through
brackness, but also whether it is liable to become so when irrigated, it is
not enough to take merely a sample of soil from the surface and analyse
that. It is essential to take samples at regular intervals below the surface
to a depth at least equal to that which the irrigation water may possibly
reach. There should be no omission that may leave a loophole for mistake.
For this reason the method employed is to take samples representing
successive sections of the soil vertically downwards from the surface to a
depth of from four to six feet. Generally, one-foot sections are taken,
but, in certain cases, ten inch, eight inch, or even three inch sections

Collecting" Alkaline Soil at Thebus.

have been collected as occasion seemed to demand. Sometimes it has
been advisable to sample solid blocks of soil, for instance, in the Thebus
investigation, where the following method was used : —

A hole, several feet in diameter, was excavated to a depth of six
feet, a solid pillar of earth being left untouched in the centre. From this
central pillar six samples were taken in each case, every sample repre-
senting the average of one foot depth of soil. It will be readily seen that
such a method of sampling involves more labour than scraping a couple
of handfuls of soil from the surface, a method which has sometimes been
employed with somewhat misleading results.

It is to the presence of alkaline salts, comprising the chloride, sul-
phate and carbonate of sodium, that the brackness of a soil is due. Of
these salts, sodium sulphate is the least and sodium carbonate the most
injurious, on account of the corrosive effect of the latter on the bark ;
and its puddling of the soil, and so interfering with tillage; the last-
named salt is the cause of what is commonly termed black brack, the dark

174

-colour being due to the organic matter dissolved by the alkaline carbonate,
•while the chloride and sulphate, which do not dissolve organic matter,
form what is called white brack. In small quantities these alkaline salts
are contained in nearly all soils, being derived from the insoluble sodium
silicates, which slowly decompose, under the influence of air, water, and
the solar heat, into the soluble salts above mentioned. Hence there is
«ver going on a constant addition to the soil of soluble sodium salts, and,
unless means exist for a removal of these salts, as constant and as rapid
as their addition, they would in time so accumulate in the soil as to
render it incapable of cultivation.

As a rule, natural drainage is sufficient to remove the noxious salts
from the soil, and they are thus continually being carried down to the
ocean by the rain water in which they are dissolved. Exceptions to this
rule, however, exist: for, when the surface soil is porous but has an im-
permeable layer underneath, the alkaline salts accumulate as it were in
a basin from which there is no outlet; or the soil may itself be so com-
pact, or otherwise constituted, as to prevent proper drainage;* lastly,
a high water table in the subsoil may prevent the exit of the injurious
compounds f : in all these cases, unless a remedy be provided, there can
be but one result, an accumulation in the soil of the salts which render
it brack.

When, therefore, drainage of a soil intended to be irrigated, becomes
impossible, the question to be considered is whether the alkaline salts are
collected at any particular level, or whether, on the other hand, they are
mere or less evenly distributed throughout a mass of soil several feet in
thickness. In such localities where rain is frequent and the temperature
not excessive, the probability is that the salts will be fairly evenly dis-
tributed throughout the soil, so that comparatively little, if indeed any,
injurious effect on the vegetation ensues. If, however, the rainfall is
scanty and the climate warm, or if there even be a brief period of heavy
rainfall succeeded by a dry season, time and opportunity are afforded for
the harmful salts to rise to the surface, by capillary attraction, in the
way already outlined, and accordingly the surface soil is charged with
such an accumulation of salts that cultivation is greatly hindered, if not
rendered absolutely impracticable.

Under these circumstances it becomes an important matter to con-
sider what proportion of sodium salts a soil may contain and still be
capable of successful cultivation. Professor Hilgard has devoted con-
siderable time and attention to investigating this point in connection with
the alkali soils of California, and he states (Bulletin No. 128) : " Barley
failed to grow where the total salts were '203 per cent., but gave a full
crop where they were '159 per cent, (half of which was carbonate). Wheat
is rather more sensitive; maize fails on slightly alkaline land; but certain
sorghums do well on mild white alkali (i.e., not containing much car-
bonate)."

In India, Dr. Leather's experiments have led him to conclude that
germination was possible where the percentage of carbonate did not ex-
ceed '4, nor the sulphate and chloride 1 per cent. In after growth '2

* Such a possibility demonstrates the importance of properly ascertaining, by
mechanical analysis or otherwise, the physical condition of the soil all the way
from the surface, down to a depth of from four to six feet, wherever an irrigation
project is under consideration.

t This again shows that if a soil is likely to prove brack under irrigation, no
investigation can be considered complete unless a series of determinations of the
height of the water table in the subsoil has been made.

175

per cent, of carbonate was harmful, and '4 generally fatal. Maize was
least affected of the cereals. Wheat grew well in the presence of '137
per cent, of carbonate, but was destroyed by '2 per cent. Legumes are
IP ore affected than cereals, but even the latter were fatally affected if the
soil around their roots contained a proportion of '2 per cent, of sodium
carbonate.

METHODS OF ANALYSIS.

The chemical analysis of an alkaline soil is far from being as
empiric in its nature, and in the method by which its results have to be
interpreted as is the case with the investigations into the proportions of
plant food in soils. There is therefore not always the same necessity,
when tabulating results of analyses of alkaline soils to specify particularly
ihe methods adopted by the analyst. And yet it is for many reasons not
only convenient, but also advisable that the methods employed in Gov-
ernment Laboratories should be stated here. These methods, as now car-
ried out, are as follows: —

1. Preparation of solution:

100 grammes of true soil (i.e., the field sample sifted through a 3 m.m. sieve;
are placed in a flask marked to contain 1,000 c.c. and the flask is filled up to the
mark with distilled water and placed in a revolving continuous shaking apparatus
for twenty to twenty-four hours. At the expiration of that time the solution is
either filtered under pressure through a dry filter placed in a porcelain perforated
plate funnel, or, preferably, by means of a Berkefeld candle filter.

2. Determination of Total Soluble Salts:

50 c.c. of the above soil solution are evaporated to dryness in a tared platinum
dish, ignited at a temperature just below redness, so as to destroy any organie
matter present, and weighed.

5. Determination of Total Carbon Dioxide as Carbonates :

An aliquot portion of the above soil solution is very cautiously titrated with
N/50 sulphuric acid, a few drops of methyl orange or preferably of phenacetolin
being used as indicator : from the quantity of acid used the total quantity of carbon
dioxide present as carbonates is calculated.

A. Determination of Sodium Chloride:

To the aliquot portion used in the last determination a few more drops of N/50
sulphuric acid are added, then four drops of a chlorine-free Potassium chromate
solution, and titratipn is carried on by means of N/100 Silver nitrate, all the
chlorine thus determined being calculated as Sodium chloride.*

5. Determination of Sulphuric Oxide:

Another aliquot portion of the original soil solution is sightly acidulated by
means of hydrochloric acid. If this causes the flocculation of clay, the mixture
is filtered and the residue on the filter is thoroughly washed. The clear acidulated
solution is heated to boiling, and a large excess of boiling barium chloride solution
is then slowly added, the mixture being allowed to continue boiling for five minutes.
After the precipitate has completely settled, the supernatant liquid is poured on to
a filter ; to the precipitate remaining in the beaker is added about 20 c.c. of boiling
distilled water, with which it is after thus washing transferred to the filter. On
the filter the precipitate is again washed, at first with distilled water slightly
acidulated by hydrochloric acid, and ultimately with pure water. The filter is
then dried, ignited and weighed.

* As a proviso to this the footnote on page 176 should be noted.

17G

6. Determination of Sodium Sulphate and Carbonate :

To an aliquot part of the soil solution a measured volume of N/50 sodium
carbonate is added in excess of that required for complete decomposition of all
calcium and magnesius sulphates, chlorides, and nitrates that may be present. The
mixture is evaporated to dryness on a water bath, the residue being taken up with
water and filtered, and the filter thoroughly washed. The sodium carbonate in the
filtrate is then determined by titration with N/50 sulphuric acid. The quantity
of N/50 sulphuric acid used in the titration, deducted from the total quantity of
sodium carbonate originally added, gives the equivalents of Calcium and Magnesium
sulphates present, The difference between the amount of sulphuric acid thus com-
bined with Calcium and Magnesium and the total amount already found, gives the
proportion combined as Sodium sulphate. Should more N/50 sulphuric acid be
required than the quantity of Sodium carbonate originally added, the excess ^over
such amount is calculated as Sodium carbonate. By "total alkaline salts" is
understood the sum of the Sodium chloride, carbonate, and sulphate.

7. Determination of Total Lime:

To an aliquot part of the soil solution two or three drops of rosolic acid are
added; then dilute ammonia by means of a dropping tube, and a sufficiency of
ammonium chloride to redissolve any precipitated magnesia, after which the liquid
is warmed, 20 c.c. of four per cent, ammonium oxalate solution are added, and
the determination is further conducted exactly as in the extraction of the soil by
hydrochloric acid. It may, however, be found preferable, especially when the
quantity of lime is small, to dissolve the oxalate precipitate in dilute sulphuric
acid and determine its amount by titration with N/100 Potassium permanganate.

8. Calcium and Magnesium carbonates and sulphates :

Any carbon dioxide which may be present in excess of that combined with
soda, as Sodium carbonate, is calculated in the first instance as Calcium carbonate.
Should this not yet account for all the carbon dioxide, the rest of the latter ir
calculated as Magnesium carbonate : on the other hand, if there still remains an
excess of Lime unaccounted for, it is calculated as Calcium sulphate. Any excess
of Sulphuric oxide which may still remain is assigned to Magnesia.*

DIVISION OF HEKBERT.

According to the methods outlined above, determinations of aLkaJine
salts were made in the three samples of surface soil from Douglas, referred
to in connection with the Division of Herbert (see p. 62). These
determinations resulted as follows : — |

Sodium Sodium Sodium Total alka- Calcium Calcium Magne- Magne- Total soluble salt -«.

v chloride, sulphate, carbon- line salts, 'sulphate, carbon- sium sium calculated, found,
ate. ate. sulphate, carbon-
ate.

1 -006 .006 -021 -028 '065 '048

2 -005 '002 '007 '001 '038 '046 '060

3 '006 '003 -001) '035 '003 '048 '056

* Should there be more Lime or Magnesia, or both, than required for the sulphuric
oxide present, these are in such case calculated as chlorides, a corresponding deduc-
tion from the Sodium chloride being made. Under these circumstances the amount
of chlorine combined with Magnesium may be checked by determining the chlorine
in the heated " total salts " residue, and deducting the figure so obtained from
that arrived at under (4), the difference between the two determinations being due
to chlorine combined with Magnesium.

t The results are given in percentages, and, except where the contrary is indi-
cated, blanks in the tables denote absence of the particular substances specified.

177

These results are favourable towards the suitability of the soil for
irrigation, but, at the same time, had opportunity offered, it would have
been desirable to extend the investigation to a depth of three or four feet
lower down, for reasons already detailed in the manner illustrated above.

DIVISION OF COLESBERG.

The six soils from the farm Oorlogspoort, in the Colesberg Division,*
were also examined, with a view to ascertaining the proportions of soluble
salts present. These were found to range between '038 and '090 per cent.,
and to average '058 per cent. Such amounts are not sufficient to be harm-
ful to plant life, even if the salts consisted entirely of sodium carbonate
or " black alkali/' the most injurious form in which alkaline salts are
fcamd in the soil.

DIVISION OF BRITSTOWN.

One of the first investigations into the possible alkalinity of lands
proposed to be irrigated was carried out in connection with the farm
Houw Water, in the Division of Britstown. Five sets of samples were
taken in almost a straight line, nearly a mile and a half from end to end,
from the lands intended to be irrigated below the site of the proposed
dam. In each case samples were collected in the manner already de-
scribed, to a depth of six feet. The physical nature of the soil may be
gathered from the following table : —

Surface.

A.

B.

C.

D.

E.

1st foot

2nd foot
3-rd foot

4th foot
5th foot
6th foot

Fine loose loaru Fine loose
loam.

Loose loam loose loam ... Loose loam.

Stiff compact
clay.

Stiff hard
clay.

Stiff hard
clay.

Very loose loam Very loose
loam.

Rather stiffer !
loam.

Stiff hard clay

Rather stiffer loam.
Looser loam.

Loose loam ... j Moist loose loam.

The approximate distances apart of the points whence these samples
were taken were as follows : —

Between A and B 430 yards.

B „ C 250 „

C „ D 850 „

D „ E 900 „

Time and circumstances did not permit of any thorough investiga-
tions into the depth of the water table below the surface, but as the aver-
age annual rainfall is very low — only about 5 inches — and as E was
the only one of the five probings which showed a suspicion of moisture
within six feet of the surface, there did not appear much probability of
a high water level in the subsoil.

* See page 52.

178

The results of the chemical investigations into the alkalinity of these
soils are given below; A 1, it should be explained, denotes an average
sample of the soil, at the point A, from the surface to a depth of one
foot; A 2 signifies the same soil between one foot and two feet depth,
and so on.

No.

Total soluble
salts
per cent.

Sodium
chloride
per cent.

Sodiam
sulphate
per cent.

Al

—

•0094

— .

A2

—

•0069

—

A3

•0117

—

A4

—

•0279

—

A5

•244

•0980

•0488

A6

•280

•1587

—

Bl

__

B2

•172

•0069

—

B3

—

•0076

—

B4

—

•0163

—

B5

—

•0350

—

B6

•376

•1447

—

01

•264

•0058

02

•176

•0069

C3

•136

•0076

—

04

•168

•0117

—

05

•200

•0117

—

06

•0163

Dl

D2

•400

•0094

—

D3

•392

•0398

—

D4

•496

•1261

—

D5

•600

•2357

—

D6

•760

•2754

—

El

_

•0117

_

E2

•348

•1028

—

E3

•640

•3080

—

E4

•920

•4155

•4189

E5

1-284

•5438

—

EG

1-636

•6272

These analyses were somewhat tentative, and, owing to lack not only
of time, but also of the necessary filtering and other appliances, the tables
are studded with gaps. These drawbacks, however, do not prevent cer-
tain definite deductions. It is noticeable that, as a rule, the amounts of
Sodium chloride, and of soluble salts generally, steadily increase ae we
proceed downward from the surfa-ce to the six foot limit, and that this
downward increase becomes more rapid as we pass from A towards E.

ILLUSTRATING SCII- ALKALINITY AT (-JOU W WAT E R . Bf\J T S TOWN Oi visior
* * •«

ToA«/ Solutte Sa/fs
Sodium chlon'de

179

Taking the entire 6 feet depth of soil in each case the average per-
centages of salts, as far as they may be based on the foregoing determina-
tions, are as follows : —

Total soluble Sodium

ealts. chloride.

A -0521

B -0421

C -189 -0100

D -530 -1373

E -966 -3348

The question now to be pronounced on is : What would be the effect
of irrigation upon such soils? Broadly speaking, and having no regard to
modifying causes, there can be only one answer. The result of irrigation
would assuredly be to raise the bulk of the alkali salts, within the first
four feet of soil, to the surface.

It appears most probable that the low rainfall in the district around
Houw Water, accompanied by other causes, prevents the water from ever
penetrating the soil to any great extent, and therefore from bringing up
the salts during subsequent evaporation. Irrigation would very probably
alter all this. The large volume of water led on to the soil is bound to
penetrate, and, when evaporation follows, as it undoubtedly must, great
quantities of salt will be carried up to the surface, and will accordingly
render the surface soil more saline.*

In connection with some alkaline soils it has been found at Tulare
in California that, although the salts in the soil have been drawn up-
wards, they have not been drawn right to the surface, and, in spite of
the fact that a considerable amount of alkaline salts was present in the
soil, there was not sufficient at the surface to cause any interference with
the growth of shallow-rooted crops like barley, which flourished, to all ap-
pearances, satisfactorily. The reason for this is to be found in the fact
that the surface soil was kept under constant cultivation, whereby
evaporation from the soil itself was checked, and proceeded instead from
the verdure of the cultivated crops. If the land can be kept under con-
tinuous cultivation, there would be two simultaneously operating causes
at work to retard evaporation, and consequently to prevent the rise of
alkaline salts to the surface. Evaporation from foliage instead of from
soil would not be the only saving factor, but the shading of the ground
by the crops keeps the temperature low, and so further still diminishes
the tendency to evaporation from the soil surface.

It would appear that the most noxious of the Sodium salts, the car-
bonate, ia practically absent from the Houw Water soils, the two com-
pounds of Sodium which form the bulk of the alkaline soil constituents
being the sulphate and the chloride. Professor Hilgard estimates that,
for barley, sandy soils should not contain, within the first twelve inches,
over *10 per cent, of Sodium carbonate, '25 per cent, of Sodium chloride,
or '45 to '50 per cent, of Sodium sulphate; in clayey soils even smaller
percentages may prove injurious. It has been calculated that the average
proportion of total alkaline salts in the four feet nearest the surface
should not exceed "2 per cent, if barley is to be grown. According to
Professor Hilgard's experiments, as already stated, barley thrived on a
soil which contained on an average *159 per cent, of alkaline salts in the
first four feet; on the other hand, the cereal refused to grow where the
average amount in the first four feet reached '203 per cent. Now, of the
Houw Water soils it seems likely that A would satisfy this condition, and
so too possibly B; the average percentage in C is '186, but both I> and E
are considerably on the worse side of the danger limit.

These remarks are obviously made on the supposition that the drainage is<
inadequate.

180

DIVISION OF STEYNSBURG.

A similar investigation to that outlined above was made at Thebus, in
the Steynsburg Division, a distance of about 160 miles E.S.E. of HOUTC
Water. In this case, too, there was a proposal afoot to construct a dam.
whereby the farm Zout Kuil and part of Van Vuurens Kraal with ad-
jacent lands could be irrigated. As Parliament had voted £150,000 for
the purpose, much depended on the suitability of the area. The last-
named farm had been under cultivation for close on to half a century,
and the occupant prided himself on the returns which he had obtained
by means of irrigation. At a point within this cultivated area a set of
samples was taken in order to ascertain, if possible, the effect of con-
tinuous irrigation on the soil, by comparing the analytical results yielded
by the irrigated soil at this spot, which will be called K, with those of
samples from the adjacent tracts which had never yet been irrigated.
This point, K, was on the opposite side of the railway to that where the
dam construction works were situated, and is rather more than three miles
below the dam and 1 ^ mile from Thebus railway station ; it constituted,
in fact, the head of the area, which it was intended to irrigate. Four
samples, L, M, N, and P, were taken at other points in this area, on the
farm Zout Kuil, extending altogether over a space about two miles in
length.

The alkaline salts in these five samples were determined in similar
manner to that employed with regard to those from Houw Water, only
much more completely.

THEBUS (STEYNSBURG DIVISION) SOILS.
In this way the following percentage results were obtained : —

Sodium Sodium Sodium Total Calcium Calcium Magne- Magne- Total soluble >alr.>.
•£ chloride, sulpbate. carbon- alkaline sulphate, carbon- sium sium Calculated. Found,
ate. salts. ate. sulphate, carbon-

ate.

Kl

•031

—

—

•034

—

•006

•007

•024

•071

•204

K2

•033

•012

—

•045

—

•01*

•004

•060

•124

•232

K3

•055

—

—

•055

—

•013

•049

•053

•170

•258

K4

•083

•210

—

•293

•283

•028

•212

•816

•872

K5

•073

•237

—

•310

•030

•029

•121

—

•490

•496

K6

•103

•181

—

•284

•067

•031

•274

—

•656

•700

LI

•191

•048

—

•239

•019

•017

•023

•298

•372

L2

•161

•181

—

•342

•144

•035

•224

•745

•660

L3

•118

•120

—

•238

—

•026

•035

•015

•314

•230

L4

•173

•082

•002

•257

—

•019

—

•035

•311

•340

L5

•205

•108

•005

•318

—

•016

—

•043

•377

•380

L6

•102

•088

•039

•229

—

•on

—

•057

•297

•328

HI

•086

•086

•021

•013

•001

•121

•252

M2

•211

—

—

•211

—

•033

•025

•022

•291

•416

M3

•277

•101

—

•378

—

•026

•020

•028

•452

•582

Ml

•224

•293

—

•517

•051

•035

•080

—

•683

•800

Mo

•212

•037

—

•279

—

•021

•159

•031

•490

M6

•291

•304

—

•595

—

•024

•030

•031

•683

•880

Nl

•213

•071

•284

•013

•004

•034

•336

•34S

N2

•234

•208

—

•442

—

•034

•071

•012

• 559

•594

N3

•181

•131

•063

•375

—

•010

—

•079

•464

•668

N4

•221

•127

•050

•404

—

•010

—

•072

•486

•480

No

•251

•096

•058

•405

—

•008

—

•088

•501

• J:-L»

N6

•198

•070

•048

•311

—

•013

—

•072

•996

•394

PI

•225

•115

•012

•852

—

•008

—

•040

•400

•664

P2

•271

•762

—

1-033

•572

•023

•279

—

1-907

2-160

P3

•193

•146

•Ill

•450

—

•010

—

•101

•561

•608

P4

•159

•104

•111

•:<7!

—

•008

—

•116

•498

•60i

P5

• 205

•055

•129

•389

—

•009

—

•125

• 5:23

•673

P6

•216

•023

•156

•395

—

•019

—

•142

•556

•570

181

Apart from the fact that the samples taken at K* represent a soil that
had already been under irrigation for a long period, it is plain that of
the five places where samples were collected K is altogether the best,
and that the conditions grow worse as one proceeds down the valley from
K to P. Taking the entire six feet depth of so^l in each case, the average
percentages of alkaline salts at the various spots were as follows: —

Spot. Sodium Sodium Sodium Total

chloride. sulphate. carbonate. alkaline

salts.

K '062 -107 Nil. '169

L -159 '104 -008 '271

M '215 -122 Nil. '337

N '215 -117 -037 -369

P -211 -201 '087 -499

This confirms the views of local men with regard to the alkali in the
soil, the opinion on the spot being practically unanimous that the lower
part of the valley is unfit for cultivation; the only difference between
one man's view and another's is that ideas vary as to the extent of the
unproductive area.

It should be noted that irrigation, followed by a period of abstinence
from water, will tend to bring the soluble salts to the surface, and thus
cause the surface soils to approximate to the composition just given in the
above averages. In the first four feet of the soils under consideration
these averages are as follows : —

K '105 per cent.

L '270 „ „

M "288 „ „

N -376 „ „

P '552 „ „

It will be seen that the soil at K is the only one which falls within
the limits; of safety as laid down by Professor Hilgard, in fact K is the
only soil which one could expect to yield a profitable return, and the fact
that farming has been carried on there with fair success does not, there-
fore, settle the matter for the entire valley. The lower layers of the soil at
K contain a fair proportion of gypsum (calcium sulphate), which has con-
verted the noxious sodium carbonate into the less harmful sulphate.
There is a general prevalence of carbonate of lime throughout these soils,
and in nearly every case a stratum of gypsum runs through the soil at
various levels ; this stratum appears in the fourth foot at K and M, and
in the second at L and at P. It is invariably accompanied by an increase
in the proportion of other soluble salts, including those which produce
brack, and although the gypsum in these layers may minimise to some ex-
tent the evil effects of the black brack, by converting it into sodium sul-
phate, it is not probable that it will do away with these effects entirely. f

* The surface soils at K, L, M, N and P correspond respectively with Nos. 9,
5, 6, 7, and 8 of the Steynsburg soils in the section dealing with determinations
of plant food (See p. 127).

r From the diagrams which illustrate the distribution of salts in the soil mass
• will be seen very clearly that the soil layers in which calcium sulphate acquires
greater prominence are also characterised by augmented proportions of the other
sulphates, i.e., of magnesium and sodium.

182

There is a general resemblance between the soils K and M, the former
being an improvement on the latter. The worst kind of brack, sodium
carbonate or black brack, occurs in the soils L, N, and P. At L it makes
its appearance at the fourth foot, at N it rises to the third, and in larger
proportions, while at P it appears at the surface and averages in amount
from two to three times that in the corresponding levels at N.

It will be noticed that the Magnesium salts exceed on the whode the
lime salts in quantity, and here may be said, what does not appear from the
figures already given, that the soils K and M show a small proportion of
the rather undesirable Magnesium chloride in the first couple of feet.

The amounts of alkaline carbonates in the Thebus samples are lower
than in the soils of the Tulare Experiment Station, California, as reported
by Professor EQlgard, and the alkaline salt present in largest proportion
is sodium sulphate, at K, and sodium chloride at the other points where

Character of Alkaline Soil at Zoutkuil, Thebus. (The truncated mountains in the -
background are Thebus and Koffiebus.)

samples were taken, but then the Tulare Station may be taken as repre-
senting an extreme case, and there is no saying whether much worse in-
stances than the present may not be found at Thebus if a more complete
investigation be undertaken.

It will be noticed how irregularly the soluble salts are distributed
through the soil; this irregularity is caused by the varying permeability
of the soil layers, and by the bands of gypsum which traverse the valley.
To this variability is due the peculiar nature of the results obtained by
analysis from the samples collected at Thebus, and numbered 10 to 19 (see
p. 127). These samples were evidently not typical, and may even, in some
cases, have been taken out of the very bauds of gypsum just mentioned,
henoe their abnormal percentages of lime. Irregularities of this type are
of common occurrence in alkali soils : King (Irrigation and Drainage,
page 283) states that

"in examining soils for alkalies, it is a matter of the utmost importance to recog-
nise that the distribution of them is extremely liable to be capricious, and that it
is easy to overlook their presence by stopping the campling of the soil just short of

1 O'>

loo

the level at which all the alkalies had chanced to be concentrated ; or, again, by
taking a sample of the 1st, 2nd and 4th feet, or of the 1st, 3rd and 4th feet, when,
i) wing to the capricious distribution, all of the salts had been collected in the 2nd
and 3rd foot, and thus were overlooked because it may have been thought not worth
while to make a complete section of the soil in question."

An analysis was made of a sample of efflorescence from the soil in the
neighbourhood of the Thebus works, obtained by enclosing a quantity of
the soil, to a depth of several feet, with a metal cylinder, drenching the
soil with water, and allowing it to dry by exposure. As was to be ex-
pected, evaporation of the water led to the appearance of a layer of salts
on the surface, and this layer, on being scraped off, was found to have t»he
following composition : —

Sodium chloride 40'16 per cent.

Sodium sulphate 11'64 „ „

Calcium sulphate 10*86 ,, „

Calcium carbonate Trace.

Magnesium sulphate 37'21 ,, „

Here, too, it will be seen that of the alkaline salts, that present in
largest amount is sodium chloride, and that the Magnesium compounds
exceed those of Calcium (Lime) in quantity. In the absence of sufficient
lime this magnesia may pro>ve harmful.*

Two similar samples of efflorescence mixed with earth, collected by
Dr. Nobbs near the same spot, were analysed, with the following results : —

Total soluble salts 1'209 per cent. 1'495 per cent.

Sodium chloride '058 ,, „ '491 ,, „

Sodium sulphate 1'122 „ „ '895 „ „

Magnesium sulphate '048 ,, „

Magnesium carbonate '008 „ ,, —

Calcium carbonate '010 „ „ '020 ,, „

Here, it will be seen, sodium sulphate prevailed, f

In a sample of water, taken from a pit on the site of tine proposed dam,

the total dissolved salts amounted to 50'40 grains per gallon ; of these salts

tEe following are the principal ingredients : —

Sodium chloride

In grains
por gallon.

6-35

In parts
per thousand.

•091

Sodium sulphate
Sodium carbonate

6'76
6'68

•097
•095

Magnesium carbonate
Calcium carbonate

16-54
8'87

•236
•127

* See Storer : Agriculture in some of its relations with Chemistry. Vol. II.,
pp. 518 and 519.

t From the farm Culmstock, in the Division of Middelburg, and lying about
twenty miles south of Thebus, a sample of effloresence, such as frequently occurs on
the surface of brack soils, was collected. Analysis showed it to contain, inter alia,

Chlorine 7'8 per cent.

Carbon dioxide 6'2 per cent.

Sulphur trioxide 33-8 per cent.

The salts present therefore consisted mainly of sulphates. In both these Divisions
(Steynsburg and Middelburg) there are extensive plains of Karroo veld, which, but
for the liability to brack, would answer excellently for irrigation and cultivation

purposes.

184

Here again, as in the soils and also in the efflorescence taken from the
surface of the soil, magnesium compounds exceed those of Calcium.
Sodium carbonate is present in small quantity, the amount being ap-
parently less than in the water from Thebus dam.

The latter yielded the following results in grains per gallon : —

Total salts 45'2

Sodium chloride 6'3

Carbon dioxide in combination 12'8

The water from the dam contains but small quantities of lime and
magnesia, so that in all probability the carbon dioxide in it was combined
with soda, forming sodium carbonate, and the use of this water, unless
accompanied by effective drainage, would therefore tend to increase in the
soil that most injurious form of alkali, " black brack." The water from
the pit, on the other hand, had its carbon dioxide combined with lime and
magnesia principally.

On page 266 of King's " Irrigation and Drainage " will be found a
table, computed from Bulletin 29 of the Oklahoma Experiment Station,
page 4, showing the composition of alkaline waters that may be regarded
as safe and those which are considered unsafe to use for irrigation.

Bearing in mind that sodium carbonate represents what is called
" Black Alkali " in King's tables, and that sodium chloride and sulphate
together constitute " White Alkali," it will be seen that the amounts in
the pit water are as follows: —

Black Alkali '095 parts per thousand.

White Alkali -188 „ „

Now, according to the tables in King's book, a water may be con-
sidered safe if the black alkali — which is the most harmful constituent-
does not exceed, say, '100 parts per thousand, provided the white alkali be
not too high, but that if the former exceed the figure just mentioned the
water cannot be regarded as safe. White alkali is less injurious, and up
to about '500 parts per thousand may be passed, but beyond this limit,
even though the black alkali may be low (as in the case of sample 741 of
King's tables, where the black alkali is "026 and the white '818 per thou-
sand), it would be risky to make constant use of the water.

In the sample of water from the pit at Thebus the white alkali was
well within the limits of safety, but the black alkali was on the border
line, and in the sample of water from the darn the limit was apparently
exceeded, although the smallness of the sample sent prevented this being
definitely ascertained.

It must, of course, be kept in view that the figures with which the
analytical results of the Thebus water are compared are derived from ex-
perimental data obtained in Europe and America. If the facilities for
making such very necessary experiments existed here, figures obtained
under local conditions might be found to differ from those obtained in
Europe and in the United States, but this very fact makes it all the more
necessary to exercise great caution in proceeding with extensive irrigation
schemes.

DIVISION OF ROBERTSON.

A series of determinations, similar to those conducted in regard to the
Thebus soils, have also been made in connection with the soil of the
Government Experiment Station at Robertson. When dealing with the
determinations of plant fuod in the soils of that locality, it was stated

185

that four soils (Nos. 21, 22, 23 and 24) had been collected on
the Station.* No. 21 was a red sandy loam and No. 22 a
finer loam with a clayey subsoil and tending towards brackness
in parts. From one of these brack patches No. 24 was taken.
No. 23, a calcareous loam, showed brack water a short distance from the
surface. Under these circumstances fuljer investigations into the alkaline
character of the soil were made. The spots R and S, whence the samples
mentioned below were collected, are identical with those where Nos. 21
and 24, above referred to, were taken : in fact, Nos. 21 and 24 represent
the surface soil, to a depth of eight inches, at R and S, respectively.
In each case, the samples collected represented successive sections of the
soil to a total depth of four feet : at R, six samples, each extending ver-
tically through eight inches, were taken ; at S, an alkaline patch with a
superficial area of about 2,000 square yards, eleven samples were taken, of
which the first eight represented three-inch, and the remaining three,
eight^inch vertical sections.

The analyses made resulted as follows : —

Sodium Sodium Sodium Total Calicum Calicum Magne- Magne- Total soluble salts.

J^TO chloride, sulphate, carbon- alkaline sulphate, carbon- sium sium Calculated. Found,

ate. salts. ate. sulphate, carbon-
ate.

Rl -024 — — — _ ._ -060

R2 -018 ________ -044

R3 -010 _____ .Q70

R4 -Oil ______ -062

R5 -Oil - — ___._„ -056

R6 -019 -080

(The blanks in the upper part of this table signify " undetermined.")

81 T029 -162 1-191 -112 '020 '043 — T366 1'400

82 -942 -114 — 1-056 '088 '020 '071 — T235 1*272

83 -662 -037 -699 -037 -067 '003 -806 -858
S4 -516 -071 -587 -029 '041 '018 -675 '700

•082 -569 -024 -014 '027 '634 '674

86 -409 -082 -491 -018 '014 -030 '553 '572

57 -342 -062 -404 -016 '014 '005 -439 '504

58 -301 -067 -368 -014 '007 -005 -394 -450

59 -291 -066 -357 — -009 -012 'Oil '389 '428
BIO '307 -073 -380 -012 -012 — -404 -476
811 -321 -076 -397 -007 '012 '004 -420 "496

In the case of the soil R, there was no necessity to push the investi-
gation further, the total soluble salts being considerably below the danger
limit. At S a very different condition of things exists, and a diagram is
attached, showing the curves of the various soluble salts in the soil as we
descend from the surface to a depth of four feet.f

DIVISION or CARNARVON.

An investigation of the alkaline tracts near the large dam called Van
"Wyk's Vlei, 45 miles north-west of Carnarvon village, was conducted on
lines somewhat similar to those already indicated. By means of this dam,

* See page 115.

t In connection with the curves shown in the accompanying diagram*, it will be
noticed, each acid ion is indicated by a distinguishing feature in the line of curve as well as
each basic ion : Thus carbonates are shown by dots, sulphates by crosses, and chlorides by
+ran*cer*e dashes. These are repeated singly in the case of CALCIUM compounds, in pairs
for MAGNESIUM compounds, and unintermittently for SODIUM compounds.

186

which holds many thousand millions of gallons of water, fairly extensive
irrigation has been practised for several years. Under this treatment the
barren area has gradually spread, and at present the nearest land that is
being cultivated below the darn is no less than seven miles distant there-
from. It is, nevertheless, 'estimated that about 20,000 acres of irrigable
land of good quality are still free from any signs of alkalinity, and avail-
able for cultivation.

Two samples of water from artesian borings in the vicinity of the dam
were analysed with the following results, stated in grains per gallon : —

No. 1. No. 2.

Sodium chloride 48'94 358'98

Magnesium sulphate 49'20 177'57

Calcium sulphate 53'26 76*86

Calcium carbonate 10' 16 147'95

A specimen of water from the dam itself was found to contain chlorine
— in combination as chlorides — to the amount of 1,1 54 '6 grains per gallon,
and, on another occasion, a sample similarly taken yielded the following
figures : —

Sodium chloride 862'9 grains per gallon.

Sodium sulphate 226'3 „ „ „

At some places in the neighbourhood, beds of limestone, containing
gypsum, occur; specimens of these were analysed, but did not prove of
good quality.

The land most affected by alkaline salts is that in the immediate
proximity of the dam, and from there the saline area is working its way
downwards by degrees, the lands being drenched with water during the
season, and then left to the action of evaporation, which does not fail to
draw the salts to the surface, with the consequence that, in the long run,
one tract of land after another has been discarded as too salt for cultiva-
tion. Under such circumstances, the practice has been to transfer to a
plot of virgin soil, until that, too, in turn, becomes too alkaline for use.

To test chemically the results that this mode of cultivation brings
about, samples of soil, T and V, were taken, respectively, from Lot 111,
which had been under cultivation for seventeen years, and had, during that
period, grown fourteen crops of wheat, and from Lot 94, which had never
been cultivated, at a spot 315 yards from that where T was collected. The
crops at T were originally as good as could be desired, but have been
gradually deteriorating. The sample taken at V is typical of the Van
Wyks Vlei soil in its natural state.

The way in which these samples were collected was as follows: In
each case an excavation was made to a depth of seven feet, and, down one
side of the excavation, a vertical groove, of six inches square section, was
cut by removing for analysis blocks of soil of six inches cube at six-inch
intervals. The first sample of each series was taken from the surface to
a depth of six inches, and then samples of each alternate six inches below
that. In both cases, at a depth of six feet a stratum consisting for the
most part of large-sized stones was encountered; of this it was not con-

DIAGRAM ILLUSTRATING SOIL

Sodium chlor.dt
Sulptiaft
•. fart

'um cn/ondt.
. tulptiatt

Cartenat
Calcium ch/or.'dt

Sulphate
• Carbonate

187

fcidered necessary to make analyses, but the soils taken from the higher
levels were examined for alkaline salts, with the results tabulated below : —

Sodium
chloride.
No.

Sodium Total Calcium Calcium Calcium Magne-
sulphate. alka- chloride, sulphate, carbon- sium
line ate. chloride.

Hague- Magne- Total soluble
hium sium salts,
sul- carbon- Calculated. Found

salts.

phate.

ate.

Tl

•849

•849

1 -5-12

•106

•014

•046

—

— 2-557

2-672

T2

•432

•432

•114

•040

•022

•029

—

— -637

•708

T3

•355

•355

•03'.)

•081

•024

•017

—

— -516

•548

T.i

•359

•105

•464

•003

•038

•036

— '541

•594

T5

•304

•363

•667

•049

•034

—

•065

— -816

•840

T6

•214

•397

•(111

—

—

•037

—

•007

•Oil '666

•664

VI

•051

•051

•068

•019

•024

•012

— -174

•222

V2

•355

•355

•500

•032

•026

•090

— 1-003

1-056

V3

• 590

•590

•404

•071

•020

•0(50

—

1-151

1-210

V4

•736

•736

•082

•198

•026

• 055

— 1-097

1-102

V5

•767

•318

1-085

•042

•048

•053

— 1-228

1-238

V6

•508

•419

•927

—

-105

•017

—

•049

— 1-098

1-116

The effect of irrigation, as practised at Van Wyks Vlei, on this type
of soil is seen, in the case of soil T, by the accumulation of salts at the
surface. In its natural condition the vertical distribution of the saline
material is such as to leave the surface soil comparatively free, so that
shallow-rooted crops may be cultivated without difficulty. In the irrigated
soil the present failure to grow crops is easily explained by the existence
of the chlorides of Calcium and Magnesium in the upper soil layers. The
occurrence of these very soluble salts in " brack " is somewhat unusual,
and can only take place in arid regions like Carnarvon. At Thebus too,
where the rainfall is not abundant — although less scanty than in the Car-
narvon Division — the presence of Magnesium chloride in the soil was
noticed (see page 183). In the Robertson Division, on the other hand,
where rain is more frequent, and the atmosphere more humid, a surface
efflorescence containing these salts is an impossibility.

It will be noticed, especially from the appended diagram, that, in the
irrigated soil, the chlorides are, for the most part, accumulated within the
upper thirty inches, while the sulphates have remained at the lower levels.

Calculating from the figures in the last table, the percentage composi-
tion of the salt at the soil surface at T is as follows : —

Sodium chloride 33'20

Calcium chloride 60'30

Magnesium chloride 1'80

Calcium sulphate 4' 15

Calcium carbonate "55

With this may be compared the following results, quoted by Hilgard
(" Soils: their formation, properties, etc.," p. 442), of an analysis of
alkali occurring in California (Imperial) : —

Potassium chloride 1'15

Sodium nitrate 8' 21

Sodium carbonate *58

Sodium chloride 31'82

Calcium chloride 58'42

Magnesium chloride 2'81

There is considerable resemblance in the respective percentages of
Sodium, Calcium, and Magnesium chlorides.

188
PART VII.— PHYSICAL COMPOSITION OF SOILS.

The foregoing pages have been devoted all but entirely to the con-
sideration of questions pertaining to the chemical nature of the soil, and
more particularly to the proportions of plant food present therein. Im-
portant, however, as the chemistry of the soil is, there is often a danger
of its looming so large in the field of vision that other factors which go- to
make up soil fertility are lost sight of. Chemical analysis alone can
never suffice to measure a soil's fertility, and even still less its productive-
ness. The fertility of the soil depends upon other inherent properties
besides the presence of plant food; for instance, its texture and general
physical condition : its productiveness is dependent on variable environ-
ments and incidental circumstances, such as rainfall, atmospheric tem-
perature, conditions of drainage, and methods of cultivation; in other
words, on factors which are altogether extraneous to the soil itself. Hence
a soil well supplied with plant food is not always fertile, and a fertile soil
is not necessarily productive : the most fertile soil cannot be productive
when climatic conditions are unfavourable, and the methods of cultiva-
tion adopted unsuitable. These distinctions require to be well kept in
"view. Leaving out of sight, however, the wider subject of crop produc-
tion, and turning again to the more restricted one of soil fertility, it would
be well clearly to understand that even the chemical aspects thereof are
closely connected with the soil's mechanical condition. The supply of
water which circulates within the soil, and which is there tenaciously re-
tained for plant use, is directly dependent upon the state of subdivision of
the soil particles, other conditions being equal. The very availability of
the plant food constituents, moreover, as there has been occasion pre-
viously to- remark, is regulated by the fineness of division in which they
•exist in the soil. Under these circumstances, as Dr. Wiley observes: — *

"It is not, therefore, a matter of surpiise that the fertility of a soil is found,
cccteris paribus, to be commensurate, to a certain limit, with the percentage of fine
silt and clay which it contains. It is true that two soils, quite different in fertility,
may have approximately the same silt percentages, but in such a case it is demon-
strable that even in the poorer soil the measure of fertility is largely the percentage
of fine particles in connection with its actual content of plant food. Many soils may
have large quantities of plant food, but thse stores, owing to certain physical con-
ditions, are not accessible to the rootlets of plants. . . . The full value of silt
analysis will only be appreciated when many typical soils, from widely separated
areas, are carefully studied in respect of their chemical and physical constitution
and the character of the crops which they produce."

We see then how closely a study of the chemical composition of the
soil must be interlocked with that of its mechanical condition. The one
is never complete without the other, and it is much to be regretted that
the exigencies of circumstances compelled the jettison of a systematic in-
vestigation into the mechanical nature of the various agricultural soils of
the Colony at the time when the chemical survey was commenced.

In the mechanical analysis of a soil we aim at determining the relative
proportions in which the different sizes of soil particles are present. There
is, however, a sad lack of uniformity in the nomenclature by the medium
of which these various grades are wont to be designated : it may hence
prove profitable to examine into this subject somewhat closely.

In popular language the generality of soils has for ma»ny years been
divided into " sandy soils " and " clay soils," as though all soils consisted
essentially or solely of either sand or clay. For the purposes of rough-
and-ready classification a broad differentiation on these lines may pass, but

* " Principles and Practice of Agricultural Analysis," "Vol. 1, 2nd ed., 1906. p. 310.

189

when, apart from chemical composition and other modifying circum-
stances, the suitability of soils for cultivating certain kinds of crops has to
be considered, a more complete and detailed discrimination becomes in-
dispensable.

Thaer (see Dr. Walmschaffe's " Anleitung zur wissenschaftlichen
Bodenuntersuchung ") had distinguished between the following seven
classes, namely : stony, sand, loam, clay, marl, lime and humus soils, and
as long ago as 1838, Sclriibler* classified soils, according to their mechani-
cal condition, as follows : —

Percentage Percentage

of clay. of sand. ,

Argillaceous soils above 50 below 50

Loamy soils 30 -- 50 50 -- 70

Sandy loams 20 -- 30 70 -- 80

Loamy sands 10 -- 20 80 -- 90

Sandy soils below 10 above 90

In the above classes the amount of sand was diminished propor-
tionately when lime or humus was present in a soil.

Ingle f adopts the following classification : —

Percentage
of clay.

Clay 70 — 95

Loam 40 — 70

Sandy below 10

To the above he adds marly, calcareous, and humic soils, the appella-
tion being dependent upon chemical and physical conditions other than the
size of their constituent particles.

HilgardJ proposes the classification given below: —

Percentage
of clay.

Heavy clay soils 35 and over.

Clay soils 25 — 35

Clay loams 15 — 25

Sandy loams 10 — 15

Ordinary sandy lands 3 — 10

Very sandy soils \ — 3

Professor Whitney, Chief of the Bureau of Soils of the United States
Department of Agriculture, puts forward no less than 16 classes of soil,
viz. : stony loam, gravel, gravelly loam, dune sand, sand, fine sand, sandy

* " Grundsatze der Agrikulturcheinie."
f " Manual of Agricultural Chemistry.'' 1902, p. 56.
; Foils ; their formation, properties, composition, fcc.,'' 1906, p. 84.

190

loam, fine sandy loam, loam, shale loam, silt loam, clay loam, clay, adobe,
meadow, and muck and swamp.*

Snydert refers to " sandy," " clay," and " loam " as the terms " used
to designate the prevailing character of the soil." Sandy soils he classes
as those containing 90 per cent, or more of pure sand. He mentions fur-
ther, loam soils — i.e., mixture of sand and clay; if clay predominates it is
a clay loam; if sand, a sandy loam. Dealing with the preference of cer-
tain crops for particular classes of soil, he supplements his classification by
mentioning the classes of soil which in a large number of cases and under
average conditions (e.g. normal supply of plant food and an average rain-
fall) have proved to be satisfactory crop producers.

The better class of potato soils, according to him, are those which con-
tain about 60 per cent, of medium sand, 20 to 25 per cent, of silt, and
about 5 per cent, of clay. For fruit-growing purposes he recommends
soils containing from 10 to 15 per cent, of clay, and not more than 40 per
cent, of sand. He looks upon those as the strongest corn soils which con-
tain from 40 to 45 per cent, of medium and fine sand, and about 15 per
cent, of clay. Good grass and general grain soils should contain about 15
per cent, of clay and 60 per cent, of silt. For wheat production he dis-
criminates between three classes of soils : those of the first class contain
from 30 to 50 per cent, of clay; those of the second, about 20 per cent, of
sand, 50 per cent, of silt, and from 20 to 30 per cent, of clay ; to the third
class are assigned those soils which are composed mainly of silt, contain-
ing thereof usually 75 per cent., together with from 10 to 15 per cent, of
clay.

It will be noticed from the above how great is the variety in the class
names applied by different investigators to the several classes of soil. But
an even greater variety exists in connection with the grouping of the dif-
ferent grades of particles which go to make up any one soil, and which,
when they preponderate, impart to that soil its special character. Such
terms as medium sand, fine &and, silt, and clay have been used above : not
only do investigators differ among themselves in the terms which they thus
employ for any one grade of soil particles, but very frequently one and the
same term has been applied to widely distinct grades. Snyder, for
instance, uses the term " fine earth " to indicate that portion of the soil
which passes through a sieve with openings \ mm. in diameter, and then
goes on to grade this fine earth as follows : —

Diameter of
particles.

Medium sand '25 — *5 mm.

Fine sand '1 - "25 ,,

Very fine sand '05 -- '1 ,,

Silt -01 - - -05 „

Fine silt -005— '01 „

Clay below '005 „

This, it may be observed, is exactly the grading adopted in the Gov-
ernment Laboratories here, and it is also employed by the United States

* In the 1906 Soil Survey Field-book issued by the U.S. Department of Agriculture,
the following 1 1 classes are enumerated under the scheme of classification based on mecha-
nical composition of soils : — Coarse sand, medium sand, fine sand, sandy loam, fine sandy
loam, loam, silt loam, clay loam, sandy clay, silt clay, and clay (pp. 17 and 18).

f " The chemistry of Foils and fertiliser," 1899, p. 24.

191

Bureau of Soils,* but is far from being in general use. Thus Wolff and
Schone designate as " fine earth " all the soil that passes through a sieve
with holes 3 mm. in diameter : Knop applies the same term to that which
passes a J mm. mesh sieve. Whitney and some other American chemists
generally take 2 mm. Hilgard and others again take -| mm.

There is an equally great variety in the methods of classifying the
several finer soil grades : thus WTollnyf classifies the several grades as

follows : —

Diameter of particles.

Stones over 10 mm.

Coarse gravel 5 — 10 mm.

Medium gravel 2 ,,

Fine gravel 1 »

Coarse sand '5 „

Medium sand '25 '5 „

Fine sand '1 '25 „

Coarse silt '05 — '1 ,,

Medium silt '025 '05 „

Fine silt '005 '025 „

Clay '0001 — '005 „

The following is Hilgard's complete classification : —

Diameter of particles.

Coarse grits 1 — 3 mm.

Fine grits "5 — 1 „

Coarse sand '4 ,,

Medium sand '3 ,,

Fine sand '16 „

Finest sand '12 ,,

Dust sand '072

Coarsest silt '047 „

Coarse silt '036

Medium silt '025 „

Fine Silt '015 ,,

Finest silt '008

Clay t

While there has been, of late years, a gradual approach to uniformity
of nomenclature, a sufficiently wide variation still exists to render the use
of mere names valueless unless the method of analysis is also distinctly
stated.

Clay, for instance, is mostly determined by a process of sedimentation.
In the investigations recorded in these pages, the soil water has been
allowed to stand for 24 hours at a time, before drawing off the clay water,
but many a soil analyst would terminate the sedimentation within a much
shorter period; three hours, for instance, or one hour, or perhaps only
fifteen minutes. Obviously results thus obtained are not comparable
amongst themselves; and, as the finely divided clay exercises a most im-
portant influence on the agricultural value of the soil, it is very necessary
to know how its proportion has been determined, and to be assured that
what is called clay does not actually also include silt.

The importance of silt and clay as component parts of any soil is one
of the chief reasons for the systematic mechanical analyses of soils. Not
only does the clay bind together soil particles which would otherwise col-
lapse into drift sands, but its own fineness of texture renders soils which

* Except for the coalescing of the " silt " and " fine tilt " into one grade.
f Vide Experiment Station Record, vi, p. 762.

1 0.2

contain much clay all the more retentive of moisture, and of the nutritious
substances therein, when sands, although abounding in mineral plant food,
may be unfitted for cultivation simply through the lack of soil moisture.

Moreover, the finest particles of the soil are those which contain the
largest amount of the elements of plant food in an available form, so that
in every respect a soil containing a large proportion of silt and clay is there-
by the better fitted for plant sustenance, and, altogether apart from any
question of direct chemical analysis, the mechanical analysis of a soil will
frequently bo a good index to its probable fertility. Above and beyond
all this it is to be remembered that a fine-grained soil not only presents its
plant food in a form better fitted for the plant to assimilate, but it gen-
erally has more of it to present.

By " clay," it need hardly be said, is meant not only the term as used
in the strictly chemical sense, namely, silicate of alumina, but, in a wider
signification, it denotes that finely divided material which when stirred up
with water, does not settle down readily, but remains suspended in the
liquid for a long time, — a time generally to be measured by days — some-
times by weeks.*

Now the effect of rain beating upon the surface of the soil is to stir
up the soil particles, and, as the water percolates down through the soil, it
constantly carries with it in suspension a quantity of clay, naturally there-
fore in a rainy district the sub-soil becomes in time more clayey than the
surface soil, and as clays contain more plant food than coarser gained soils,
it also follows that the subsoil will, as a rule, be richer than its surface
soil. Hence, too, subsoils are more retentive of moisture, and less easily
penetrated by the rain -owing to the accumulation of clay.

The mechanical analyses of soils are thus of far-reaching importance ;
not only does such an analysis serve its direct purpose of determining the
proportion of particles of various sizes in the soil, but, indirectly, it sheds
light upon the soil's power to retain moisture, and also upon its permea-
bility to plant roots, and it even aids — as stated above — in forming a con-
jecture as to its chemical potentialities, since the soils that contain most
silt and other fine grades of soil particles, are frequently the richest from
a chemical point of view.

Dr. Wiley's words, already quoted, partially answer the question why
soils of certain class are required in order to cultivate particular crops, but
it does not solve the problem completely. Reference has been made to
Snyder's apportionment of differently graded soils to specific crops, but just
why such classes of soil should be so specially suited to some crops, and not
to others, has not been fully stated. A partial solution may be found in
the fact that it is a characteristic feature of each kind of plant to require
for its development a certain degree of soil heat, and specific amounts of
moisture : both of these are dependent upon the mechanical condition of
the soil. Now, according to Snyder, soils of the class that he prescribes as
best suited for potatoes are so suited because such soils generally contain
from 5 to 12 per cent, of water; similarly, those which he recommends for
fruit usually retain from 10 to 18 per cent, of water : those which are
recommended for corn should, according to the same authority, contain
about 15 per cent, of water; grain soils, he says, ought to hold from 18 to-
20 per cent, of moisture. Now all these conditions, naturally, are intended
to apply to certain parts of the United States of America. In the Cape
Colony we know as yet neither (1) whether soils physically composed as
above would retain the proportions of moisture stated, nor (2) whether such
proportions would be the optima for the classes of crops mentioned. What

" As used in a physical sense ' clay ' may be silica, felspar, limestone, mica, kaolin,
or any other rock or mineral which has pulverised until the particles are less than -C05.
mm. diameter." (Snyder : "The Chemistry of Soils and Fertilisers," 1899, p. 13).

193

needs to be ascertained here is, first of all, what the mechanical condition
of the soil is in localities where these crops do well, and next, how much
moisture such soils are capable of retaining under the climatic conditions
prevailing in this country. It was to give a start to research along these
lines that the investigations recorded on the subsequent pages were
initiated.

METHOD OF MECHANICAL ANALYSIS OF SOIL.

The complete sifting process used in the Government Analytical Labo-
ratory for these investigations included the employment of the following
sieves : — *

1. A brass-bottomed sieve with circular perforations 3 mm. in

diameter.

2. A similar sieve with perforations 2 mm. in diameter.

3. Another similar sieve having perforations 1 mm. in diameter.

4. A Kahl's (Hamburg) " Messingdrahtsieb No. 50."

Lateral measurement of meshes '35 mm. — '39 mm.

Diagonal measurement of meshes '45 mm. — '50 ,,

5. A Kahl's (Hamburg) "Messingdrahtsieb No. 100."

Lateral measurement of meshes '14 — "17 mm.

Diagonal measurement of meshes '22 — '24 mm.

6. An Ehrhardt & Metzger's (Darmstadt) " Florsieb No. 16."

Lateral measurement of meshes '09 mm.

Diagonal measurement of meshes '11 mm.

For the mechanical analysis of soils all the above sieves are used, and
the sifting process is further supplemented by sedimentation in water.
When a chemical analysis is the chief object, sieves Nos. 1, 3, and 4* only
are employed, according to the method already described.

In carrying out the mechanical analysis of a soil, a weighed average
quantity (generally from five to eight pounds) of the field sample is taken
and dry sifted through sieve No. 1. The pebbles left on the sieve are
washed and dried, and their weight noted for ultimate calculation in per-
centage of the water-free field sample. All that passes through sieve No.
1 is denominated " true soil."

250 grammes of this true soil are placed in a porcelain dish; about
half a litre of distilled water is poured on, and the dish is allowed to digest
on a water-bath for at least two hours, the contents being frequently
stirred. When the soil has been so softened that it can be washed through
the sieves by the aid of a small brush, the actual sifting operation is pro-
ceed with as follows1 : —

Sieve No. 4 is held over a dish containing distilled water; the moist-
ened sioil is placed in the sieve, and the latter is immersed with its per-
forated bottom about one inch or more below the water level in the dish.
By means of a small brush the soil is now stirred until particles no longer
pass through, after which the residue in the sieve is washed with distilled
water. This residue is next washed on to and through sieve No. 3. That
which passes through the latter sieve is dried and weighed as " coarse
sand," while the term " gravel " — using the word as conveying the idea
of dimension and not that of nature — is assigned to the residue upon thia
sieve.

* Sieve Xo. 4 is no\v definitely replaced by a brass bottomed sieve with circular per-
forations £ mm. in diameter.

N

194

The partial mechanical analysis outlined above thus separates every
soil sample thereby treated into the following grades: —

pebbles
> 3 mm.

gravel
3 — 1 mm.

Field sample

true soil
< 3 mm.

earth
< 1 mm.

coarse sand
1 — '5 mm.

fine earth
<• '5 rnm.

Both the gravel retained by the 1 mm. sieve, No. 3, and the fine
earth which passed through sieve No. 4, have to undergo further differ-
entiation. The former — by means of sieve No. 2 — is separated into*
" coarse gravel " and " fine gravel," and from a fresh quantity of the " true
soil " — 50 grammes in weight — a sufficient amount of fine earth is obtained
by sifting (as the 250 grammes were sifted before) through sieve No. 4.
This new quantity of fine earth is now well stirred and washed, in a man-
ner similar to that already described, through sieve No. 5. To the residue
on this sieve the term " medium sand " is applied. The sifted product
obtained in this way is brought into sieve No. 6 and again carefully
washed : the name " fine sand " is given to what remains on the sieve. All
the products remaining on each of the sieves are weighed after drying at
105°C.

The fine grained mate-rial which has passed through sieve No. 6 is
placed in a syphon cylinder of the construction indicated in the annexed
sketch. The height of the cylinder is about 40 centimetres, and the in-
ternal diameter is six centimetres. This cylinder is marked near the neck,
and another mark is made at a level 200 millimetres below the first one.
When the sifted soil has been placed in the cylinder the latter
is filled with water to the upper mark, after which it is stoppered, in-
verted, and well shaken: it is then placed at rest in its normal

195

position for twenty-four hours, after the expiration of which time the
water, together with the clay still in suspension, is syphoned off. It may
be necessairy to refill the cylinder with water and repeat several times the
process of shaking, subsidence, and drawing off, until no more clay can be
removed. All the liquid thus drawn off is evaporated, and the clay which it
retained in suspension is then weighed. The sediment in the cylinder is
again treated with water and shaken as before, but now the liquid is
syphoned off after a subsidence of 100 seconds, and this operation is re-
peated until a practically clear liquid comes over. What now remains in
the cylinder is washed out, dried, weighed, and returned as " very fine
sand," while that which passes over with the syphon-water is again re-
turned to the cylinder for a fresh treatment as before, except that a 1,000
seconds subsidence is now permitted. What remains in the cylinder at
this stage is called " silt," and that which is syphoned off is " fine silt,"
both being now dried and weighed.

The following scheme illustrates the entire sifting and sedimentation
process : —

Wet sifting. Wet sifting and sedimentation.

Field
sample

Dry sifting.

pebbles
> 3 mm.

true soil
< 3 mm.

f gravel

3 — 1 mm.

f coarse gravel
3 — 2 mm.

fine gravel
2 — 1 mm.

coarse sand
1 — mm.

f medium sand
•5 — '25 mm.

earth J fine sand

< 1 mm. | -25 — '1 mm.

I

very fine sand
*1 — '05 mm.

silt
•05— -Olmm.

fine silt

•01 — -005mm.

clay
_< -005 mm.

EESULTS OF MECHANICAL ANALYSES.

fine earth
< mm.

•By wet sifting.

100" sedimenta-
tion.

1000"

24 h.

in suspension

In most of the samples collected, and enumerated in the foregoing
pages, where any mechanical separation at all has been made, it has been
only of a rudimentary description; that is to say, it has consisted simply
in grading the soil into two portions, the one comprising the " fine earth "
and the other the coarser soil. And yet, elementary though these opera-
tions have been, they have elicited figures that do not entirely lack
interest. For instance, if we look at the soils taken from the south-
western corner of the Colony, where the Malmesbury and Table Mountain
geological series prevail, we find, on the whole, a coarser type of soil than
that of the area within which, for the most part, the Bokkeveld series
occurs; and this again, yields a soil of coarser texture than the rest of
the country from which soils have been examined, and covered chiefly by
rocks of the Karroo system, more particularly by those of the Beau-
fort series.

196

In the very broadest sense possible, then, the soils examined may be

grouped in these three sections, and so we get the results shown below.

1. SOILS OF THE MALMESBURY AND TABLE MOUNTAIN AREAS.

No. Percentage of

Division. examined. Fine Earth.

Caledon ............ 4 68'0

Cape ............ 21 70-2

Malmesbury ......... 7 67 '7

Paarl .'. .......... 51 65'0

Piquetberg ......... 1 72'4

Stellenbosch ......... 26 80' 5

Tulbagh ............ 9 56*3

... ... 119 00*1

2. SOILS OF THE BOKKEVELD AREA.

No. Percentage of

Division. examined. Fine Earth.

Bredasdorp ......... 1 90 '6

Ceres ............ 7 89'8

George ............ 23 84'3

Ladismith ............ 16 73-8

MosselBay ......... 17 87'2

Oudtshoorn ......... 19 79'1

Riversdale ............ 24 73-4

Robertson ............ 22 77'7

Swellendam ......... 9 78'4

Uniondale ............ 12 69'3

Worcester ............ 34 77'9

Summary ...... 184 77 '0

3. SOILS FROM THE BEST OF THE COLONY.

No. Percentage of

Division. examined. Fine Earth.

Albert ............ 3 96'6

Aliwal North ......... 4 98'7

BarklyWest ......... 13 93 "7

Beaufort West ......... 1 94'0

Butter worth ......... 7 94 '1

Carnarvon ........ 2 96*1

Cathcart ........... 28 97'8

Colesberg ............ 6 94-7

Elliot— Slang River ...... 4 96 '2

Fort Beaufort ......... 4 85'6

Gordonia ............ 2 100

Graaff-Reinet ......... 3 95-2

Herbert ............ 3 94'3

Hopetown ............ 2 96'0

Idutywa ............ 1 97'6

Kenhardt ............ 4 86'4

Kimberley ............ 6

King William's Town ... ... 1 97-5

Krysna ............ 16 94'9

Komgha ............ 27 97'0

Mafeking ............ 2 76'9

Middelburg ......... 1 27'5

Mount Currie ......... 2 60'0

Mount Frere ......... 2 98*3

Port St. John's ......... 1 97-2

Prieska ............ 4

Queenstown ... ... ... 6 95*5

Richmond ............ 2 88'7

St. Mark's ............ 4 81 '5

Somerset East ......... 2 99'3

Steynsburg ......... 9 96'4

Uitenhage ............ 10 98'2

Umtata ............ 2 98'8

Umzimkulu ......... 2 99'1

Victoria East ......... 1 99'6

Vryburg ............ 16 8.V7

Willowvale ......... 4 93'6

Wodehou?e ...... 3 96 '1

197

It will be observed that the 22 soils from the divisions of Kenhardt,
Mafekiug, and Vryburg — a widely extended area where the geological for-
mation may be said to correspond in some sort to that of the south-western
corner of the Colony, inasmuch as sandstone and quartzite enter largely
into the composition of the soil — are again of a coarser texture, their pro-
portion of fine earth averaging 85 '0.

The Division of George is naturally divided into a northern and a
southern portion by the great range of the Outeniquas : the northern part
continues into the Uniondale Division, and of the southern the Knysna
Division is the natural continuation. In each case there is a corresponding
similarity in the texture of the soil, as the following shows : —

No. of samples Percentage of

Area. examined. fine earth.

George, northern part 9 69-2

Uniondale... 12 69 -3

George, southern part
Knysna

14

94-0
94-9

While fairly complete mechanical analyses have been comparatively
few, a number of soils have been graded in greater detail than those re-
ferred to above ; some of these are tabulated below : —

Division.

Albert

ucniiij rvwv
55

Caledon

•

Graaff-Reinet

.

55

Port St. John's

*

Malm

eebury

•

Richmond
Tulbagh

Uitenhage

Vryburg

Serial No.
in list of
samples.

1

2

3

10
11
12
38

1

2

3

1

65
66
67
68
69
70

1

2

1

2

3

6

1

2

3

4

5

6

7

8

9
10
10

Pebbles^
> 1 mm.

nil

1-5

2-0

0-7

3-2

1-5

0-8

1-2
•9
•9

nil
15-8
10-1
10-0
12-0
14-6
12-4

7-7

5-8
34-9
54-6
54-0

2-1

nil

nil

nil

0-4

0-4

nil

0-2

0-5

0-2

nil

2-1

Coarse

nil

3-9

2-8

1-1

0-2

0'9

17-6

4-0

3-6

3-7

2-8

12-2

16-1

10-0

15-8

17-4

1T6

6-0

3-1

15-8

6-5

21-6

3-1

0-5

1-1

0-9

1-6

5-4

1-6

0'2

3-7

0-6

nil

8-3

Fine

earth

< i mm.

100
94-6
95-2
98-2
96-6
97-6
81-6
94-8
95-5
95-4
97'2
72-0
73-8
80'0
72-2
68-0
76-0
86-3
91-1
49-3
38-9
24-4
94*8
99-5
98-9
99-1
98-3
93-2
98-4
99-6
95-8
99-2
100

In the three orchard soils, Nos. 31, 32, and 33, collected in the Worcester
Division, the proportions of silt and clay were found to be respectively
17'3, 14'2, and lO'O. Soils of this description, containing over 75 per cent,
of fine earth, whereof less than 20 per cent, is composed of silt and clay,
and hence containing at least 55 per cent, of more or less fine sand, may
very well be classified as medium sands.

Of the four soils representing the Elliot-Slang River District, a partial
mechanical analysis was made of one, No. 3, with the following results : —

Pebbles
Coarse gravel
Fine gravel
Coarse sand
Fine earth ...

> 3 mm. -38%

3 — 2 mm. -58%

2 — 1 mm. 1-39%

1 — J mm. 6-87%

< £ mm. 90-78%

The sample examined, as previously stated, typifies a red soil of fino
texture, and one of the most productive in the district : chemically, there
are other soils in the neighbourhood which are better, but in these soils
the conditions of environment more than counterbalance their chemical
superiority.

Similar mechanical analyses have been made in the case of the soils
from Elsenburg numbered 3 to 19 in the Stellenbosch Division list. The
following are the results of these analyses : —

No.

3

4

5

6

7

8

9
10
11
12
13
14
15
16
17
18
19

Pebbles
> 3 mm.

2'38

1-55

1-47

1-18

1-64

2-12

14-02

22-90

13-19

12-85

1-96

1-34

1-17

1-91

2-96

13-11

1-62

Coarse

gravel

3 — 2 mm.

•37
•49
•68
•72

•40
•22

2-82

1-57

5-55

5-44

•79

•95

•83

1-17

2-35

3-61

1-83

Fine

gravel

2 — 1 mm.

1-76

1-68

2-30

2-40

1-80

1-36

7-06

4-55
11-96
10-96

3-34

3-25

Coarse

sand

1 - £ mm.

12-11

9-31
11-59
11-35
16-37
14-53
15-30
11-89
15-48
15-64
15-20
15-70
10-83

5-37
15-49

7-78

9-32

Fine
earth
J mm.

83-38
86-97
83-96
84-35
79-79
81-77
60-80
59-09
53-82
55-11
78-71
78-76
83-27
88-49
72-09
71-43
83-63

Nos. 18 and 19 of these soils contained a great proportion of ironstone
gravel, and, on reference to their chemical analyses, it will be seen that
they are both very poor.

Fourteen soils representing the Paarl Division were examined in like
manner, with the results stated in the following table. As in the last
table, the serial reference number connects each analysis with the sample
in the list of Paarl soils already given : —

No.

20
21
22
23
24
25
26
28
30
31
32
33
34

Pebbles
> 3mm.

13-8
5-4
7-2
3'2
Nil.

13-6
2-4
3-2
9-8

12-8

•2

4-6

19-6

28-1

Coarse

Fine

gravel

gravel

3 — 2 mm.

2 — 1 mm.

15-0

23-4

9-4

15-0

3-0

5-8

1-0

2-0

•1

•2

•6

1-4

3-0

9'6

6-8

22-0

9-2

19-8

16-4

36-4

4-6

14-8

4-8

12-6

21-8

35-8

15-1

25-3

Coarse

sand

1 — i mm.

10-8
13-0
15-0

5-2
•4

6-6
14-8
21-2
13-6
13-4
14-2
12-4
14-2
14-0

Fine
earth
£ mm.

37-0
f>7'2
69-0
88-6
99-3
77-8
70-2
46-8
47-6
21-0
66-2
65-6
8-6
19-5

199

In the Division of Steynsburg ten samples of soil were collected on the
farm Groene Vlei, in the Brak River Field C'ornetcy, north of the railway
line near Thebus Station, by Mr. W. B. Gordon, late Director of Irriga-
tion. They axe not included in the list of soils collected in the above men-
tioned division,* and were examined by mechanical analysis with the fol-
lowing results : —

No. a ^. .

A 36-18 14-22 49'60

B 42-42 28-54 29'04

C 44-92 23-92 31'16

D 25-14 35-34 39'52

E 37-82 30-72 31'46

F 46-14 16-96 36'90

G- 28-72 35-84 35'44

H 23-72 32-86 43'42

I 24-96 49-06 25'68

J 23-68 39-06 37'26

The [joils in the neighbourhood may be generally described as sandy
loams, with a very distinct tendency towards alkalinity, and rather cal-
careous in parts.

More comprehensive than any other series of mechanical analyses
undertaken in the Government Analytical Laboratories was the investiga-
tion into the physical composition of the different types of soil which occur
on the Government Experiment Station at Robertson. In dealing with the
chemical analyses of the soils from the division named, mention was made
of four samples, Nos. 21, 22, 23 and 24, taken from the grounds of the Ex-
periment Station. When the surface soils, which were subsequently
chemically examined, had been removed to a depth of eight inches, further
samples were collected at each spot, representing every succeeding eight
inches in depth, up to a total depth of four feet, thus making six samples
a-s representing the soil-profile at each of the four spots, or 24 samples in
all. In the tables below, the numbers 21, 22, 23, and 24 signify the sur-
face soils to eight inches deep; Nos. 21a, 22a, 23a., and 24a mean the second
eight inches of soil, and so on. |

* See page 126.

t It may be explained that the figures in the following tables are all the results of
direct determinations upon the air-dried soil. Any loss of material would therefore be
included in the " moisture," which would represent the total weight of the air-dry soil less
the sum of the weights of the different grades of soil particles.

200

1

b- CO O 1C CO "*i
CO 00 b- b- 1-1 b-

CO O ^J b- 1C (N

§CO CO OS O b-
i— i OO OS b- OS

CO CO —i CO C^ •*
>O (N iO CO CO •*

o S

00 00 OS C<J O OS
i-H <M <N rH

CO CO O O OS b-
W CO 5N CNJ

6N TH CO t^ CO N

I-H rH rH

OS OS tO b- CN C«
r-i <M 5N C^ CO CO

V

£

£

c^i

CO 1C rH CO l-H i-H
00 1C 1C b- CO OS

CO OS rH CO CO b-
CN O O b- -* OS

b- O "* b- 1C O
•<*< •* b- 00 00 b-

CO 0 W b- b- CO
b- <N I-H »C »C 1C

U. »: |

00 CS b» O 1C CO

1— t r— 1 rH

co i-t ou co b- b-

rH

^ rH •* 1C CO CO
rH rH i-H (N (^ cq

^^^^^^

8

5 1

b- >O b» CO CO -H

CO OS rH rH CO OS

I-H >.-: o co os co

1C OS CO CS 1C 00

1C "* O i— i OS CO

•* I-H CO 1C O b-
CM b- -* O <N <N

1

OS b- «O OS CO 3S

b- 05 b- <* b- b-

O OS -* I-H to »C

10 CO CO Th CO •*

*

«, a

1«

»O b- OS OS »C b-

(M 1C 00 b- CN b^

b- C75 >C i-H b- CO
b- tO 0 •* rH <M

h» «>. -"f O b« OS

CO 1C 00 S b- <N

<s * i

<-H f-H 3J M CO b-

0000 b- »C CO CO

CS CO t> ^ CO 00

b- »C b- 00 b- QO

> j*

I-H rH l-H rH

OD

j a

3

8 a

o

O O OS 00 OS CO
•* CO 1C «C OS i-H

cigg§^^

«o »o co -* -^ oo

CO O •> «> O 1C

S^b25c^o

s

g esi
1 »

O CO OS CO rH -*
•* CO C<) <M W (N

»-H 00 <M O 00 00
»C CO CO CO <N (N

OS J(5 CO to b- r-4
CN CO W W IM (M

co co co co co co

5c

p

H

fl

tf

oil

S]
PQ

M • — C V-

OS •* CO b- 00 •*
OS b- rH i— i CS 1C

«C O »C CO OS OS

CO »C OS CO 1C C^

CO b- 00 tO rH •*
CO •*< OS CO C\) W

-* CO N -^ O CO
CO OC -h CO CO ^H

S

| S5*^

o co N co co ic

OS OS CO b- b- OO

« o co co to co

CO b- OS O O OS

rH rH

O

1 a

OQ

W

S ||J

"* CO CO 00 C<J CO

— i O N b- O CO
<M b- <N CO CO OS

OS OS '^ CO »C CO
00 TM "* CO »Q CN

££££££

02

Kj

5w i

00 CO ^H b- «0 10

CO <M CO rH rH rH

OS rH 0 00 10 -*

CO 00 b- CN !M CN

r?

ill

os co o co So co

OJ C£ to »C CO CO

OS 0 J« ^H co <M

SCO •**< 00 CO 1C
b- CO O OS i-H

fe ^ i

rH

rH IN rH rH

C) .1- _ 7; — -N

0

rH

o*

-

2

0

|f a

M M 00 CO « O

w2§§c5;2

•M — — --T _ C-l
(M 7v| rH -*« 00 OS

CS 1C I-H CO CO OS
•* 00 <M O «> •""

s

0* |

-I

^H rHrH FH

IB

1 «

iS2SS§§

SS3S22

rH OS •«** <N Q UB

rH 0 <N 00 0& CO

•* rH O b- N "**
O OO OS b- b- OS

* A

N tO U5

1C CO -f

1

1

111

1 t

I4

1

•fr * *i

*^>^ r = s

>> r =*

>»......,.-

o * *

Jci

§ .s

fl

0} r3

cc

CO P«H

rfl

4

e3 ,Q o T! a?

M*^wScN

j3,D OT? v

eS^ 0^3 v

<N CS W W W «

W ?\ M (M c4 5S

WCNWCNCNC^

Diagram illustrating mechanical Analyses of Soil No. 21 from the Government

Experiment Station. Robertson.

Petal led Analysis. Summary.

1 2345*6 78 9 10 '

• •§

I &

? =

S3 £

cc En

^
o

Perctnt.

O

too

to
O

100

r

fo

z

so

uo

o
too

1*

9o
7°

tt>

o
too

o
100

I

Z

uo

*0
JO
IO

0

Diagram illustrating mechanical Analyses of Soil No. 22 from the Government

Experiment Station, Robertson.
Detailed Analysis. Summary.

123456789 10

s S

1 2 1 3 I"

PM C5 cc 02 a

Perant:

Jo

10

o

too

1°
9o

z

fo

Diagram illustrating mechanical Analyses of Soil No. 23 from the Government

Experiment Station, Robertson.
Detailed Analysis. Summary.

123456789 10

s -S

j» fe

Diagram illustrating mechanical Analysis of Soil No. 24 from the Government
Experiment Station, Robertson.

Detailed Analysis.

Summary.

201

The results detailed in the foregoing table may be summarised as fol-
lows:—

Pebbles Gravel Sand Silt Clay

Xo. > 3 mm. 3 — 1mm. 1 — -05mm. '05 — -005mm. < -005 mm.

21
21a
21b
21c
21d
21e

•15
•12

2-26
•54
1-08
•09

1-15
I'll

1-89
1-00
•85
•45

71-28
72-64
02-47
55-82
48-94
42-90

18-75
17-50
13-68
19-89
28-97
36-82

8-67
8-83
19-70
22-75
20-16
19-74

22
22a
22b
22c
22d
22e

•68
•07
•04
•01
•18
•15

1-14
•75
•75
•58
•82
•74

78-03
69-44
62-98
55-03
54-10
55-69

13-79
21-04
15-31
13-71
15-05
15-80

6-36
8-70
20-92
30-67
29-85
27-62

23
23a
23b
23c
23d
23e

•11

•09
•24
2-82
3-80
15-35

roi

1-21

roi

3'57
2-09
2-64

61-35
63-70
55-23

48-34
44-97
40-28

24-93

20-84
29-64
37-28
40-44
38-76

12^60
14-16
13-88
7-99
8-70
2-97

24
24a
24b
24c
24d
24e

15-04
3-81
4-90
•77
•72
•94

3-49

2-58
2-85
3-11
2-74
3-34

51-97

56-47
58-13
60-12
56-15
54-48

9-97
7-91
7-61

8-62
I'll
8-80

19-53
29-23
26-51
27-38
32-62
32-44

It will be observed that the alkaline patch, which is also chemically
the poorest of the four in plant food, contains very much less silt than the
others, whereas No. 23, the soil best supplied with plant food, is likewise
the richest, in silt.

In order to illustrate the character of these soils the better, a diagram
showing the texture curves of soil No. 24, for each eight inch section,
is appended. Such diagrams reveal to the eye, at a glance, more clearly
than mere figures can, how, on descending from surface to sub-soil, in No.
21, the sand diminishes in proportion, while the silt and clay increase; in
No. 22 sand also diminishes, while clay increases ; in No. 23 both sand and
clay diminish, but silt increases ; and in No. 24 the proportion of clay in-
creases, and there is a diminution of pebbles.

On a former page reference was made to the Orange River clay col-
lected in the Prieska Division (No. 1, Prieska Division soils) : the chemical
analyses of sojls from farms situated along the river banks lower down were
also given (Nbs. 2, 3, and 4) : these three samples, in each case, represented
the surface soil to a depth of twelve inches, but opportunity was taken at
the same time to secure specimens of the subsoil to* a depth of three feet,
and mechanical analyses of all these were made, with the results stated
below : —

Pebbles Gravel Sand Silt Clay

No. > 3 mm. 3 — 1mm. 1 — '05mm. -05 — -005mm. < '005 mm.

1 nil nil -29 2S-53 71-18

2
2a
2b

nil
nil
nil

•05
•10
•03

77-81
60-46
38-26

10-58
31-37
53-29

5-50
8-07
8-42

3

3a
3b

nil
nil
•08

•04
nil

•18

36-14
55-91
59-78

4.V53
32-15
25-64

18-29
11-94
14-32

4
4a
4b

1-61
•06
•17

3-64

•44
•57

50 ' 05
59-05
53-88

30-63
28-31
31-36

13-47
12-14
14-02

202
These results are set forth in greater detail in the next table : —

g

>>s 2S

S 0 5

C5 'ti (M

c<» o co

^S 0

•9s r

OS

»s oo oo

00 P"H •*

CO C<l «i<

V

8*f ^

CO t^ CO

<N CO O
O5 C<J 00

co cs t-

O 0 f

^ 3 *

4* 00 r*H

OS r*H <N

0 (N CO

a

1

P-< -* O

C<l CO OJ

5^00

CC O 00

35 '| «

2§5i

«0^l

S2^

8

, a

aj 3

Swo

§ss

•^ »c t-»

CO *O OO

£§T

«x |

00 O "*
N CO (N

2§4^

Ol QO |>.

3 J

o cc a

* g ftr

CO •* O5
<N CO 00

C<J CO i-"
t> <M (M

O o m

t*~ ^^ CS

00 t- <N

t-OCi

I-H co co

S^w^

GO Q ^

fi ° jj

S5B

00 O 00

r- t- P-I

o 1 :1*|

si*

•:*•

r-t (N •*

i4*

H o

1-3 PH g'gj j-j

i-H 0 CO

i-H CO rH

•^ oc »~

00 «O O

| S"l

•<*! r-H pl|

i .si

CO 0

<3^ 00 »c

HH 0 ^

o o o

(N <N CO

^ S 17 '3
^

'3

H

I ni _

§§„

i— c CO

Oj-H 0

gjsa

«BJ 'S

'3

'3

1—1

sa
sa ^

S

1-1 1C t~

O O i-H

«; «

9 as

sf*

rH

•

i!:!

' a' !.

I ^

1

'O

•&

•a

— :

il

i

- a

a *

K ~ s

|

rl

II

1 "

1

e8 —
(N (N 0*

d ^a

CO CO CO

c3X»

Diagram illustrating mechanical Analyses of Orange River Clay and of Soils from the
farm, Stof Kraal, in the Prieska Division, adjacent to the Orange River.
Detailed Analysis. Summary.

128456789 10

I 2

o «

. 1 t>

Ills I

Diagram illustrating mechanical Analyses of Soils from the farm Keuken Draai,
in the Prieska Division, adjacent to the Orange River.

Detailed Analysis. Summary.

123456789 10

I I

CO

* I Is

co PH O

O
100

<JO

fa

Z

fa

203

Rich though it may be chemically, the Orange River deposit contains
too much clay — and is therefore too dense and heavy — to enable it to be
advantageously used in its original condition; but when deposited on the
surface of a sandier soil, with which it may be manipulated, it would add
considerably to the agricultural value of the latter. If it could b© directed
on to a very sandy soil, containing, say, 90 per cent, of sand, a single flood-
ing of the lands would — assuming a half-inch annual deposit — increase the
proportion of silt and clay in the soil to about 17, and diminish its sandi-
ness proportionately; after a second flooding the soil so treated would con-
sist of 76 parts of sand and 24 of silt and clay, and the third year would
see the sandy soil converted into a loam composed of 70 per cent, of sand
and 30 per cent, of clay.

At a previous stage a remark of King in connection with the Rio
Grande deposits was quoted ; he continues the passage as follows* : —

" When such sediments are laid down upon coarse sandy soils, it will be readily
appreciated that the gain to the field is far greater than that due to the mere plant
food which the sediments contain ; for such sediments, being composed of very fine
grains, their influence in improving the texture of the soil is quite as great as that
due to the fertilisers contained."

The Orange River clay, although rather heavier, appears in some
degree to resemble what has been called Vergennes clay, a type that occurs
on the terraces surrounding Lake Champlain, and consists of deep-water
sediments deposited in post-glacial times. The subsoil of this clay, which
more particularly resembles that from the Orange River, is a heavy drab
or light-brown clay, somewhat tenacious when wet, but extremely stiff,
compact, and intractable when dry. The Vergennes clay is said to be an
excellent soil for hay, while corn, oats and barley are also profitably
grown; it yielded, when mechanically examined by the United States
Department of Agriculture, the following results : —

Surface soil. Subsoil.

Fine gravel : 2 — 1 mm. ..... 2 0

Coarse sand :1 — '5mm. ... . 1

Medium sand : '5 — '25 mm. 2 2

Fine sand: -25 — -1mm. ..... 3 2

Very fine sand : *1 — '05mm. 3 2

Silt: -05— -005mm 18 22

Clay : less than -005 mm 69 71

Nos. 2, 3 and 4 illustrate the effect of the Orange River clay being
deposited on and mixed with the Kalahari sands which prevail for many
miles around ; these sands are brick red in colour, and constitute the
coarser portions of the soils on the neighbouring farms. In the process
of mechanical separation into grades, the point of transition from the
coarser red sands, contributed to the soil by the Kalahari, to the finer-
grained drab-coloured silts, derived from the river deposits, is readily dis-
tinguished. This sharp contrast of colours enables one to draw the infer-
ence that the silt deposits are often more coarsely grained than that of
which No. 1 is a sample. In each of the three soils sampled the particles
of greater diameter than 'I mm. are of local origin, whereas those below
that size are river-borne sands, silts and clays. No. 2, as the table
of results shows, becomes less sandy as one penetrates deeper, but Nos.
3 and 4 are sandier lower down than at the surface.

As representing some of the choicest lands in those districts which
have been termed the Granary of the Colony, the two Koeberg soils num-
bered 33 and 34 amongst the Cape Division samples (vide page
39), were chosen for mechanical analysis. They have already
been described, and their chemical condition referred to. t Each of
these two soils was sampled, in twelve-inch vertical sections, to a depth

* " Irrigation and Drainage," 2nd ed., 1902, p. 259.
fSee pp. 40 and 158.

204

of three feet, and from the samples so collected the following results were
obtained by mechanical analysis : —

S'S £ 2 So

So — T j

Og 10 /O CO

00 O O5

V

a
a

g^i =^ r r

Sot-

00 ft re

o

s 2 ^

£
~S J5 2 2

s^ 2 ^ ^

s s 2

g

3

u §

I

1?

0

|lj i i «
J

% r ?

r-H T-H OS

HH
>

s

S a

§ Sg'| * t- A.

R

3

co co r-

CO -H CO

o
tf

a

o a I

3«e* *o »o ro

w i§M » ^ ^

0 «g'| CO (N (M

!:!

a
o
M

g e'ga « 3 j^

Rss

o

g%^ co co co

•* * *

a

02

ft

HJ :2£

to oo co

CO CO CO

BCHANICAL

i^|

g§« ? ? r

hi tO Io

s a « t- in

2 g co «N y?

*£c*5 04 »C h-

O CO h-

I

- ' 1

>»

ir

1

s

0- w ^ jO

5n CO CO «

* -

cc co ro

Diagram illustrating mechanical Analyses of Soils from the farm Hooge Kraal
in the Cape Division.

Detailed Analysis.
45 6 7

Summary.

10

T!
fl
c3

:£

00

.

Xo. 38.— CAPE DIVISION — LOAM.

Soil Grade.

Percent. 9'

a

o

1-58

&

fe
3-21

O
3-58

I I

3-o«; 8-44

,9

I

m

n;-35

m
32-28

No. 24.— BOBERTSON DIVISION — SANDY CLAY.
Percent. 15'04 1'49 2'00 6'76 7'34 31-50 7'37 5'24 4'73 19'53

205

The figures of the table on the preceding page are summarised below,
and the results are also presented in diagrammatic form in order to facili-
tate comparison with the other Colonial soils similarly examined.

Pebbles Gravel Sand Silt Clay

No. > 3 mm. 3 — Inirn. 1— ;05 mm. -05 — -005 mm. < '005 mm.

33 9-86 4-79 21 "37 48'63 15-35

33a 5-27 5-17 18'59 45 "35 25 -62

33b 7-65 4-25 19 '22 45-36 23-52

34

10-16

5-40

38-98

36-57

8-89

34a

13-71

5-50

37-70

32-98

10-11

34b

17-65

3-90

36-93

32-31

9-21

It will be found of interest to compare these results with the figures
quoted from Snyder (see page 190) as the most satisfactory for wheat
production. As before observed, it do'es not by any means follow that
the optima of mechanical composition for wheat soils in the United States
of America are also the optima in the Cape Colony ; in fact, it would be
strange if such were the case, in view of the widely different climatic con-
ditions.

By way of supplementing the diagrams which illustrate what I have
termed the " texture curves " of some of the soils examined, a photograph
is appended showing the actual proportions of the several grades of soil
particles separated by the above described processes from the former of the
above two soils, and from No. 24 of the Robertson soils.* It will be in-
teresting to compare this photograph with the diagrams drawn from the
calculated results of the mechanical analyses of the respective soils.

CONCLUDING OBSERVATIONS.

In bringing this record to a conclusion, the writer must express hia
sense of its incompleteness. True, no investigations of this nature can
ever bear the impress of finality, they must perforce continue incomplete
as long as agricultural chemistry has room to progress. But the incom-
pleteness that attaches to this special record is the inseparable adjunct of
contingencies which have all along restricted to one pair of hands at a
time the work of investigating the chemical problems of the soil in a
country so vast and varied, and possessing such diversities of temperature,
rainfall, and geological conditions, as this Colony presents. When, above
all, this investigation had to be sandwiched in amongst multifarious other
calls on time and attention — or passed on to assistants, when time failed
entirely, as it did almost at the very outset, and the work had then to
be fitted in amongst their other duties — the difficulties were greatly in-
creased ; and moreover, as no other researches of like nature were being
carried on within a radius of many thousand miles, special appliances
had to be procured from abroad, and perhaps tested for a considerable
time ere their unfitness stood revealed. Thus much time was continually
being lost, and great periodic gaps occurred in the pursuance of the work.

There are also comparably great gaps of totally uninvestigated tracts of
country, and even the areas that were visited have been practically
skimmed over. Thus it has come about that much had to be done by
delegation, through written instructions — a course which always fails to
secure uniformity — for instance, in regard to collection of samples. Fre-
quently failure to carry out these instructions to the letter has either

* It is in each case the surface layer of soil that is represented in the photograph.

200

hindered, or neutralised the value of, the subsequent laboratory work, or
prevented its performance, more especially in sparsely populated parts of
the country, where, through lack of conveyance, months have at times
elapsed before fresh samples could be procured.

The directions in which investigation was made also show omissions
that will be patent on glancing through these pages. Some of these de-
fects will, it is hoped, be supplied in the near future; others must of
necessity remain defects until contingencies enable one to widen the scope
of investigation. Amongst the latter, unfortunately, has to be classed one
of the most important studies that pertain to the soil, that is, the soil
nitrogen, its exhaustion and replenishment.

And yet, however beset with difficulties and obstacles the work has
hitherto been, the writer thinks that — thanks to the assistance of those
who have, from time to time, been members of his staff, and whom, first
one and then another, he charged with the details of the work — these
records will show the existence of solid reason for initiating this under-
taking, and that the investigation has not been wholly void of fruit must
also be apparent. The connection between the composition of the soil
and the ravages of bone disease in stock has not been fully traced out, but
it is at least remarkable that these diseases find the greatest foothold in
just those areas where lime and phosphates are deficient in the soil. In
connection herewith the question has been put, why, if the sandstones of
the Table Mountain and allied series are so poor in the plant food con-
stituents named, " lamziekte " does not make its appearance in the south-
west, but is confined to the districts further east? One answer obviously
is that, in the more easterly districts, the rocks in which these con-
stituents are deficient stretch uninterruptedly over vast extents of country,
while in the west they a<re sufficiently diversified by granite and the rocks
of the Bokkeveld series.

The confirmatory testimony given, in many cases, during the course
of these investigations, by practical farmers, has shown that, presuming
certain necessary data, correct deductions as to the probable fertility of
a soil are capable of being made from analytical results. It would be idle
to assert that discrepancies do not occur, but in the present state of en-
quiry one must needs meet with seeming anomalies : there is a plain neces-
city for pushing investigation very much further. The best means of
determining the proportions of plant food constituents in the soil have to
a certain extent been studied, and the methods finally adopted appear,
judging from the confirmatory evidence just spoken of, to be the most
suited to the country's circumstances. To transfer hither, unmodified, and
without enquiry, European or American methods of treatment, or
standards of interpretation, would scarcely have been wise.

Regarding the irrigation schemes which must continue to bulk largely
in the country's future agricultural development, it has been shown how
needful it is previously to enquire into the condition of the soil, and here
not only the alkali deposits themselves need investigation, but equally so
the physical condition of the soil. The important problems of differen-
tiating soil types, and of classifying the Colony's soils from a laboratory
standpoint — an investigation which has proved so useful in America —
still remain to be grappled, and it is the writer's increasing view that
just at this point a thorough study, and, if possible, a mapping out, of the
mechanical condition of the soil should go conjointly with, if indeed not
actually precede, enquiry into the quantities of plant food present. In
connection with a scheme like this an adequate knowledge of the distribu-
tion of indigenous plants and of their soil relation is also
most necessary. Such a scheme, the writer trusts, will ere long
be put into practical operation. How frequently barrenness may

207

be caused by other factors than the lack of plant food in
the soil has been shown more "han once in these page®, so
that there is no need to labour the matter further; and in many cases
where there is no question of barrenness at all, the difference of texture
in soils suited, say, to the cultivation of potatoes, and soils of a more
clayey nature, has exercised a controlling influence upon the class of farm-
ing to be practised; such differences of texture in the soil profile may be
actually mapped out, and would afford a means of incalculable benefit. It-
has, for instance, been seen that to a certain extent geological influences
perform an important part in this connection (see p. 195) and, placing
this over against what has been referred to above regarding the mapping
out of soil types, it becomes evident that an interesting and most valuable
phase of the subject invites closer study. Hitherto the study of the soil's
physical condition has extended little beyond the mechanical analyses of
more or less typical soils of three broad classes, namely, the grain soils
of the south-western districts, the fruit and general farm soils of the Wor-
cester-Robertson area, and the deposited silts of the Orange River, all of
which have been dealt with under the head of " Physical composition of
soils." A far wider application of this branch of investigation should now
be embarked upon.

From a chemical point of view the richness of the soils where the
Bokkeveld series and the Karroo system prevail has been shown, nob to
mention the chemical value of the river silts, the products of the erosion
that is going on elsewhere; and the efforts to retain and utilise these
latter also deserve extended application as a matter of economic import-
ance.

In these and other respects, briefly and inadequately touched upon
hitherto, and in many more as yet untouched, which need not be specially
mentioned here, but will occur at once to anyone reading the foregoing
records and acquainted with the state of agricultural science to-day, the
practical value of a soil survey to this vast and fertile country is immense,
and all the more is it important at this juncture, when a general re-
awakening to the need of agricultural development is manifesting itself,
and when it is beginning to be more thoroughly realised how great are
the capacities of this sub-continent for such development, and for taking
full advantage of the magnificent climate wherewith it has been endowed.

I must not conclude without some words of acknowledgement of the
services of those gentlemen who, as members of my staff, have from time
to time individually undertaken many of the analyses which have enabled
me to compile this record. Although most of the work detailed under the
heads of " Alkalinity " and " Physical composition " was carried out by
myself in person, of the analyses enumerated in Part III. opportunity was
not afforded me, for the reasons already explained, of performing more
than about eighty. My especial acknowledgments are due to Mr. A.
Simons, B.A., who performed over 400 analyses. The services of the late
Mr. J. C. Watermeyer, and of Messrs. J. Muller, B.A., and St. C. O.
Sinclair, M.A., who undertook some 60 analyses each, are also entitled to
special recognition here. Without the valued aid of these and the other
members— past and present— of the laboratory staff, this compilation could
not, for lack of material, have been undertaken. My thanks are also due
to my brother, Mr. J. W. Juritz, for his ready assistance in drawing the
maps with which this record has been illustrated.

INDEX.

(N.B. — Names of farms have not, as a rule, been included in this Index).

Achter Cogmans Kloof, F.C., 115, 116, 120.

Achter Hex River, F.C., 147, 151, 152.

Acid, Carbonic, as a plant food solvent in the soil 5.

„ Citric, „ „ „ „ „ 2, 9, 10, 12, 13, 155.

Acid excretions from roots, 2.
Acid, Humic, as a plant food solvent in the soil, 5.

„ Hydrochloric, as a plant food solvent, 2, 9, 10, 12, 13.
Acidity of soil (see also "sour veld ") 23, 42, 92, 116, 118, 133, 154.
„ „ „ and liming, 64, 92.
„ ,, ,, retards nitrification, 23.
Agricultural chemical methods, 5.

,, Experiment Station, Minnesota, 9.

Robertson, 115, 153, 155, 184, 185, 199, 200.

„ methods at the Cape, different from those of northern hemisphere, 5.

Air, source of plant food, 1.

Albany Division, 4, 20, 21, 23, 64, 154, 155, 167.
Albert Division, 20, 21, 24, 25, 52, 63, 121, 128, 155, 169, 171.
Aliwal North Division, 20, 21, 24, 25, 52, 63, 155, 169.
Alkali, Injurious proportions of, Prof. Hilgard's estimations, 179, 181, 182.

,, Amount of, at soil surface kept low by cultivation, 179.
Alkaline efflorescence, 183.

soil, 34, 35, 93, 110, 116, 117, 128, 142, 146, 150, 153, 172.
„ at Thebus, Steynsburg Division, 173, 180-184.
collection of, for analysis, 173, 182, 186.
at Thebus, 173.
,,VanWyk'sVlei, 186.

„ „ need of sampling complete soil profile, 182.

distribution of salts in, 174, 182.
effect of irrigation on, 172, 179, 180, 187.
evil minimised by gypsum, 181.
irregular salt distribution in, 182.
in Britstown Division, 177-179.
., Carnarvon ,, 185-187.
„ Robertson „ 116, 184, 185, 201.
„ California, 174.
„ India, 174.

Magnesium salts in, 182.
methods of analysis described, 175.
,, need for draining, 174.

,, proportions of salts that cause barrenness, 174.
„ results of analyses of, 176, 178, 180, 185, 187.
water, analyses of, 183, 186.
Amazamyama plantation, 95.

Analyses and practical agriculture, instances of concordant results, 6, 11, 13, 17, 28
50, 53, 55, 66, 72, 82, 85, 86, 105, 117, 133, 135, 141, 148, 149, 150, 155, 181
Analysis of soil, chemical, value of, 2, 5, 11, 18.

„ ., „ methods of, 2, 5, 13-17, 175.

„ „ „ different kinds of, 8, 9, 11, 59, 172.
„ „ „ pounding may give misleading results, 13.
„ „ „ prejudice against, 11.

210

Analytical figures, interpretation of, 17.
Antonies Berg, F.C., 138.
Anysberg, 74.

Aqua Regia as a plant food solvent, 12.
Arkansas soil survey, 11.
Atriplex, 26, 83.
Attaquas Kloof, 96
Augea capensis, 97.

Availability of plant food, 2, 3, 7, 8, 9, 10, 12, 153, 159, 172, 188.
Available phosphoric oxide, 2, 156.
Avoca, Bredasdorp Division, 28, 30.
Avontuur, Caledon Division, 33, 34.
„ Uniondale Division, 138.

Bacteria, 1, 23.

Balfour, 128.

Barkly West Division, 20, 21, 25, 142, 162, 163.

Barrenness produced by causes other than lack of plant food, 18 (n), 42, 61, 73, 102,

103, 111, 188, 192.
Barren patches, 102, 124.
Bathurst Division, 24, 167.
Beaufort geological series, 89, 121, 168.
Beaufort West Division, 20, 21, 28.
Before Attaquas Kloof, 90.
Bidouw, 51.

Black brack (sodium carbonate), 173, 179, 182, 184.
Blauwberg, F. C., 39.
Blauw pot clay, 102.
Bloemfontein, F. C., 28, 32.

Blown sands, 30, 35, 45, 68, 93, 95, 108, 112, 125, 170, 171.
Bokkeveld and Table Mountain series soils compared, 165.

geological series, 24, 29, 112, 115, 139, 140, 145, 149,*157, 166.

„ ,, ,, felspathic nature of basement beds of,^166.

„ soils, chemical analyses of, 165, 166, 171.

,, ,, derived from line of junction with Table Mountain series, 165, 16tt.

„ „ proportions of fine earth in, 196.

„ „ satisfactorily provided with plant food, 165.

Bone manure, 4.
Bontebok Vlakte, 47, 50.
Bot and Palmiet Rivers, F.C., 33, 36.
Bot River, 35.
Boussingault, 11.
Brack : see " Alkaline."
Brak River, Mossel Bay Division, 90.

„ „ Steynsburg Division, 126, 127.

Bredasdorp Division, 4, 19, 20, 21, 28, 38, 120, 139, 154, 164.
Breede River, 115, 129, 131, 134.
Britstown Division, 109, 177.
"Broken" soil, 92, 93, 116, 117, 131, 132.
Buchuberg irrigation works, 108.
Buffelsfontein, 74.
Buffels Klip, 138, 140.
Buff els Kraal, 111.
Buffels River, 171.

Burghersdorp beds, 24, 25, 53, 121, 168, 171.
Butterworth District, 19, 21, 32.

Calcareous tufa, 28.

Caledon Division, 4, 19, 20, 21, 29, 33, 112, 113, 139, 154, 164,

Caledon, F.C., 33.

California, alkaline soils of, 174, 179, 182, 187.

Cameron, F. K., 3.

Campbell Rand, 141, 143, 144, 162, 163.

„ „ series soils, analyses of, 162, 171.

Cango, 77, 96, 99.

beds, 161.

fertility of, 77.

Cape Division, 4, 19, 20, 21, 38-45, 80, 123, 125, 134, 155, 157, 158.
Cape Flats, 42, 125.

211

Cape geological system, 163, 171.
Carbonic acid as a plant food solvent in the soil, 5.
Carlton, 89.

Carnarvon Division, 45, 185.
Caroxylon salsola : see " Ganna."
Cathcart Division, 19, 21, 46-50, 74.

Cattle diseases and deficiency of phosphates, 3, 23, 64, 102.
Cereals and phosphates, 45, 47, 72, 114.
Ceres Division, 20, 21, 50, 52, 164.
Chemical analysis of soil, its value, 2, 8, 11.
„ „ methods of, 2.

,, „ complete, purpose served by, 2.

„ ,, index of fertility, 2.

and mechanical composition of soil, relation between, 188.
decomposition distinguished from mechanical disintegration in the produc

tion of plant food, 10, 80, 124, 159, 163.
Chlorine, determination of, in soil, 8, 175.
Chumie : see Tyumie.
Citric acid for soil extraction, 2, 9, 10, 12, 13.

„ method of soil extraction, 16, 17, 155, 156.
Clanwilliam Division, 20, 21, 51, 155.
Classification of soils on basis of plant food percentages, 17.

„ „ „ mechanical condition, 188-190.

Clay, application of the term, 192, 195.

carried down to subsoil by rain, 172, 192.

in mechanical analysis, definition of, 192.

Orange River, 108, 203.

richest of soil grades in plant food, 13, 24, 192.

soil, American heavy, analysis of, 10.

soils, physical composition of, 188.

Vergennes, 203.

,, mechanical analysis of, 203.
Coarse sand, application of term, 193.
Cofimvaba, 121.

Colesberg and Steynsburg soils compared, 128.
Colesberg Division, 20, 21, 52, 63, 128, 155, 169, 177.
Collection of samples, 6, 173, 186.

Comparison of soil extraction methods, 59, 63, 94, 119, 120, 126, 153.
Constantia loams, 42.
Coo, The, 116.

Corn (i.e., Indian corn) soils, physical composition of, 190, 192.
Corstorphine, Dr. G. S., 166.
Cossa, Professor, 2.
Cretaceous System, 169, 171.

Cultivation without manuring, habit of, 25, 28, 50, 82, 83, 84, 85, 113, 143.
Culture experiments, 5.
Cylinder for sedimentation, 194.

Decomposition, chemical, and mechanical disintegration, 10, 80, 124, 159, 163.
Definitions of soil grades, 7, 12.

Clay, 195.

Coarse gravel, 194.
„ sand, 193.

Earth, 194.

Field sample, 7.

Fine earth, 7, 194, 195.
,, gravel, 194.
„ sand, 194.
„ silt, 195.

Gravel, 193.

Medium sand, 194.

Silt, 195.

True soil, 7, 193.

Very fine sand, 195.
Deterioration of irrigated lands, 186,
Diabase, 163.
Diep River, 38, 41.
Digby, 1.
Disintegration and decomposition, 10, 80, 124, 159, 163.

212

Distel doom, an indication of fertile soil, 102.
District of Butterworth, 19, 21, 32.

„ Elliot-Slang River, 20, 21, 53, 196.

„ Idutywa, 19, 21, 64.

„ Maclear, 20, 21, 77.

„ Mount Currie, 20, 21, 94.

„ Mount Frere, 20, 21, 95.

„ Port St. John's, 107, 171.

„ St. Mark's, 19, 21, 121.

„ Umtata, 20, 22, 137.

„ Umzimkulu, 20, 22, 137.

„ Willowvale, 19, 22, 145.

Divisions (Fiscal) of the Colony will be found indexed under their respective names.
Dolerite, intrusive, 25, 27, 28, 47, 48, 71, 72, 112, 121
Doleritic soils, rich in plant food, 27, 28, 32, 53, 162.
Dolomite, 161.
41 Doode turf," 102.
"Doom" soil, 116, 131, 132.
Downs, 29, 30, 125.
Downs, F.C., 41.

Drainage, the remedy for brack, 174.
Drift sands : see " Blown sands " and " Downs."
Drip lime, 47.

Dunes : see " Blown sands " and " Downs."
Durban, 38, 44.

Dwyka geological series, 115, 141, 147, 150, 163, 168.
„ „ „ soils, analyses of, 168, 171.

„ „ „ „ compared with those of other geological series, 168, 171.

Dyer, Dr. Bernard, 2, 10, 17.

„ „ method of soil extraction, 16, 21, 155.

Eastern Province, 4.

East Uitvlugt, 110.

Ecca (see also " Umsikaba), 115, 147.

Eerate River, 124, 155.

Efflorescence, analysis of, 183.

Elim, 30.

Elliot-Slang River District, 20, 21, 53, 196.

Elsenburg, 122, 123, 155, 198.

Elutriation methods, 1.

Enon conglomerate, 70, 147, 149, 159.

Erysiphe graminis, 40, 44.

Evaporation from soil surface retarded by cultivation, 179.

Evelyn Valley, 67.

Exhaustion of soil, 45, 118.

Extraction of soil by water, 8, 153, 175.

Fallowing, 83, 92.
Farmers, soils collected by, 20.
Felspar in Bokkeveld series, 166.
Fertilisers, Bone, 4.

„ Cultivation of crops without, 25, 28, 50, 82, 83, 84, 85, 113, 143.

Fortuity and productiveness, 2, 10, 18, 188.

„ as gauged by chemical analysis, 2, 5, 13.

„ due to lime, 93.

„ lack of, may be due to other causes than deficiency of plant food, 18, 42,

60, 73, 102, 103, 111, 188, 192.

„ mechanical soil analysis often affords a good index of. 188.
Field operations, 19.

„ sample, application of term, 7.
Fine earth, application of term, 7, 191, 194, 195.
„ „ average proportions of, in soils of the Malmesbury, Table Mountain, and

Bokkeveld series, 196.
„ sand, application of term, 191, 194.
Finer grades of soil more retentive of water than the coarser, 188, 192.

„ „ richer in plant food, 13, 24, 53, 172, 192, 201.

Fluxes, use of, for extraction of plant food constituents, 9.
Forest soils, 42, 43, 57, 61, 67, 70, 95, 119, 135, 137, 138.
Fort Beaufort Division, 20, 21, 54.

213

France, Agricultural College of, 11.
French Hoek, 99, 100, 106.
Fruit culture and lime soils, 117.

„ orchards, 117, 132, 148, 197.

„ ,, in Worcester Division, mechanical analyses of soils of, 197.

,, soils, physical composition of, 190, 192.

Gamka Flats, 96, 97.
Gamka River, 109, 113, 170.
Gcinna bush, 96, 139, 142.

,, ,, indicative of lime soil, 96, 139.

Garden soils, 39, 42, 48, 54, 76, 113, 117, 132, 135.
Gcua River, 32.
" Geil grond," 131.
Genadendal, 36, 37.
Geological Commission, 89, 157, 166.

,, formations and selection of soil samples, 157, 166.

,, relation of soils as influencing plant food content, 157.

,, ,, ,, „ ,, their mechanical condition, 196, 197.

George Division, 5, 19, 20, 21, 24, 55, 70, 77, 139, 153, 154, 164, 197.
German Experiment Stations, 2, 11, 13, 15.

,, ,, ,, method of soil extraction, 15.

Gordonia Division, 20, 21, 60, 171.
Goudini, F.C., Caledon Division, 33, 35.
Goudini, F.C., Worcester Division, 147, 150, 152.
Gouritz River, 91, 112, 113.
Gouwkamma, F.C., 67, 69.
Government Analytical Laboratory, 4, 12.

Experiment Station, Robertson, 115, 153, 156, 184, 199,
„ ,, „ „ results of mechanical analyses of soils from,

200.

Graaff-Reinet Division, 20, 21, 60, 155.
Grac; of soil : see " Mechanical analysis."

various, of plant food, 9, 10, 160.
Grahamstown Agricultural School, 3.
Grain districts, 114, 203.

,, soils, American, analyses of, 9.

„ physical composition of, 190, 192, 204.
Granite as a source of potash, 92, 100, 161.

intrusive, 79, 82, 84, 89, 92, 100, 102, 115, 124, 159.
„ undecomposed, poor in plant food, 124, 160
Granitic soils, analyses of, 124, 161, 171.

,, ,, compared with those of Malmesbury series, 159, 161.

" Grauwgrond," 100.

Gravel, application of the term, 191, 193.
Great Brak River, 55, 92.
Griquatown geological series, 162.
Grobbelaars River, 96.

„ „ valley, fertility of, 77.

Groene Kloof, 78, 85.
Groeneberg, 79, 100, 103, 104.
Groot Constantia, 124.
Groot Drakenstein, 99, 100, 104.
Groote Vlei, F.C., 146.
Groot River, 74.
Guano, 28, 117, 118, 150.
Gwatyu, F.C., 109.
Gypsum at Thebus, 181.

at Van Wyk's Vlei, 186.

Hahn, Prof. P. D., 124, 159.

Hall, A. D., 18.

Halle Experiment Station, 13, 17.

„ method of determining available phosphoric oxide, 2.
Hanover Division, 20, 21, 61, 155.
Harts River, 25, 26, 142.
Hebron, F.C., 25, 26.
Heidelberg, 112, 129, 134.
Helderberg, 124.

214

Herbert Division, 21, 62, 155, 176,.
Herbert's Dale, 91.
Heuveltjes : see " Hillock soils."
Hex River Range, 148.
„ Valley, 147.

Highlands, Albany Division, 23.
Hilgard, Prof. E. W. 5, 11, 12, 17, 18, 87, 99, 153, 174, 179, 189, 191.

„ soil classification of, 189, 191.

„ ,, soil extraction, method of, 16, 21, 153.

Hillock soils, 85, 86, 87, 101, 102, 103, 105.

„ more fertile than lower levels, 85, 86, 102, 103.
„ „ withstand rust, 102.

Hope, The, F.C., 31.
Hopetown Division, 20, 21, 63, 162.
Hottentot's Holland, 124.
Huis River, 74.

Humansdorp Division, 4, 20, 21, 57, 63, 137, 154, 167.
Humic acid as a plant food solvent in the soil, 5.
Humus theory, Thaer's, 1.

Hydrochloric acid as a plant food solvent, 2, 9, 10, 12, 13.
„ ., boiling, soil extraction by, 126, 155.

„ „ methods of soil extraction by, 13-16.

Tbeka, 64.

Idutywa beds, 32.

Idutywa District, 19, 21, 64.

Ignited soil, extraction of, by Hydrochloric acid, 15.

Immediately available plant food, 9, 10, 11, 156, 172.

Immediate productiveness and permanent value, 10, 11.

Ingle, H., 189.

Inorganic plant food, estimation of, 8.

Insects as fertilising agents, 87.

Irrigation, 108, 109, 113, 115, 117, 126, 127, 136, 142, 143, 172.

effect of, on alkaline soils, 172, 179, 180, 186.
„ works should be preceded by soil analyses of certain kinds. 172.

Kaffir Kuils River, 111, 113, 114.
Kahl's soil sieves, 7.
Kalahari, 108, 203.

„ sands, 203.
Kalkbank, 77.
Kambu Forest, 137.
Kamnassie, 57.
Kaolin, 42.

Karnemelk River, 129.
Karroo bushes : (see also Ganna), 97.

„ system, 32, 121, 154, 168, 169, 171.
soil transported from, 109, 113, 170.
''Karroo" soil, 92, 93, 96, 112, 113, 114, 116, 117, 118, 126, 132, 139, 140, 142,

149, 150.

Keisie, The, 116, 117.
Keizers River valley, 116.
Kenhardt Division, 20, 21, 65.
Kentani beds, 32.
Kentucky soil survey, 11.
Keurbooms River, 67, 69, 171.
Kimberley Division, 20, 21, 65, 171.
King William's Town Division, 20, 21, 67, 167.
Klapmuts, 122, 124.
Klein Drakenstein, 104.
Klein Steenboks River, 33, 35.
Klein Zwart River, 34.
Klip and Doom Rivers, F.C., 55.
" Klippers grond," 36, 103.
Klip River, 129.
Kluitjes Kraal, 129.

Knysna Division, 4, 19, 20, 21, 67, 139, 155, 164, 170, 171, 197.
Koeberg, 39, 40, 41, 45.
Kokstad, 94.

215

Komgha Division, 19, 21, 47, 49, 71.
Konings River valley, 116.
Koopmans River, F.C., 134.
Komvgoed : see Augea capensis.
Kraai River, 25.
Kraal bosch, 40.
Krombeks Rivier, 111.
Kuruman, 143.
' Kweek " soil, 92.

Laboratory and field experiments, Comparison between, 2, 5.
Ladismith Division, 19, 21, 24, 74-77, 98, 129, 165, 166, 167, 171.

,, Karroo, 77.

Lakes, The, 55, 57.
Lamziekte, 4, 23, 64, 102, 167.

Langeberg Range, 77, 112, 113, 116, 129, 131, 132.
Langley Park, Albany Division, 23.
Langtouw River valley, 91.

Leather, Dr. J. W., investigations on alkaline soils, 174.
Liebig, 1, 2, 11, 17.

Law of the Minimum, 11, 17, 95, 120, 153.
Lime, a dominant factor in soil productiveness, 86, 99.
as a source of fertility, 93, 96, 98, 99, 117, 131.
deficiency of, 38, 59, 67, 130, 139.
„ determination of, 7, 13, 15, 16.
„ in clay and sandy soils, 17.
„ large proportions of compensate for lack of phosphoric oxide and potash, 17,

24, 25.
„ needed for fruit culture, 117.

Shell, as a fertiliser, 35.
Limestone formations, 30, 77, 83, 89, 96, 97, 131, 137, 141, 142, 143, 144, 147, 161,

162, 169, 186.

Loam, Physical composition of, 189.
Local investigations, Need of, 184, 192, 193.
Long Kloof, The, and its soils, 55, 57, 139, 140.
Loughridge, Dr. R. H., 12, 13, 16.

„ method of soil extraction, 16.

Louisiana, Hilgard's work in, 11.
Lower Albania, F.C., 62.
Lower Groot River, F.C., 24.
Lower Riebeek, F.C., 23.
Lower River Zonder End, F.C., 33.

MacDonald, A. C., 20, 23.

Maclear District, 20, 21, 77.

MacOwan, Prof. P., 45.

Maercker, Prof. M., 13, 17, 21.

Maf eking Division, 20, 21, 77.

Malmesbury Division, 4, 19, 20, 21, 78, 123, 134.

geological series, 30, 79, 89, 100, 102, 115, 123, 134, 147, 150, 157, 158,

159, 161, 171.

„ „ „ soils, chemical analyses of, 158, 159, 171.

,, „ „ „ proportions of fine earth in, 196, 197.

„ „ „ „ water retaining capacity of, 171.

Mancazana, 128.

Manuring, Cultivation without, 25, 28, 50, 82, 83, 84, 85, 113, 143.
Mechanical analysis in the investigation of alkaline soils, 174.
of soil, 1, 3, 7, 18, 90, 188, 193.

„ good index of fertility, 192.

„ „ „ „ lack of uniform procedure, 191.

„ „ „ „ methods described, 193.

,, ,, „ ,, scheme of, 195.

„ ,, ,, ,, sieves used for, 193.

„ value of, 192.

,, ,, various soil grades defined, 7, 191-195.
results of, 195.

„ „ Koeberg soils, 204, 205.
„ „ Prieska soils, 201, 202.

„ Robertson soils, 200, 201.
„ „ Vergennes clay, 203.

216

Mechanical disintegration distinguished from chemical decomposition. 10, 80, 124,

159, 163.

„ fineness of soil particles requisite, 1, 3, 6, 13.

Medium sand, application of the term, 190, 191, 194.
Methods of collection and analysis of samples, description of, 6, 13-17, 173, 175, 176,

193-195.

„ „ plant food extraction, 2, 13-17, 153.
Middelburg Division, 89, 183.

„ efflorescence at, 183.

Middle Bosjesveld, 115.
Mimosa, 57, 142.
Mineral theory, Liebig's, 1, 2.
Minimum, Law of, 11, 17, 95, 120, 153.
Minnesota Agricultural Experiment Station, 9.
Mississippi, Hilgard's work in, 11.
Mogogong, 142.

Moisture, determination of, in soil, 7.
Molteno sandstones, 53.
Montagu, 115, 116, 118, 120.
Moorreesburg, 81.
Mosselbanks River, 78, 85.

Mossel Bay Division, 4, 19, 20, 21, 90-94, 109, 113, 154.
Mount Currie District, 20, 21, 94, 112.
Mount Frere District, 20, 21, 95.

Namaqualand Division, 20, 21, 95.

Napier, 28, 30.

Nauwpoort, 89.

Newlands, 41, 42.

'Nhlambe, 32, 64.

Nieuwjaars River, 30.

Nieuwveld, F.C., 28.

Nile valley fertilised by river deposits, 109.

Nitrification retarded by lack of lime and phosphates, 23.

„ „ „ soil acidity, 23.

Nitrogen, determination of, in soil, 7, 8.

„ fixation of, by bacteria, 1, 23.

„ proportion of, in Oudtshoorn soils, 98.
Nobbs, Dr. E. A., 20, 25, 26, 115, 141, 142, 183.
Nomenclature, variety of, in soil classification, 189-191.
North Middenveld, F.C., 63.

Oaks, The, 33, 36.

Oathay, Cape, poor in lime and phosphates, 3.

Odendalstroom, 24.

Oklahoma Experiment Station, 184.

Olifants River (Oudtshoorn), 91, 96, 97, 109, 113, 139, 170.

Olive shales, 141, 162, 171.

Onder Duivenhoks River, 111, 114.

Orange River, 24, 25, 108, 143, 171.

clay, 108, 203.

Organic matter, determination of, in soil, 7, 8.
Oudtshoorn calcareous loams, 98.

Division, 4, 19, 21, 77, 91, 96-99, 113, 139, 161, 170.

causes of fertility, 91, 96, 98, 109, 170.
„ Karroo, 77.

Outeniqualand, 55, 57.

Outeniqua Range, Difference in soil texture north and south of, 197.
Over Hex River, F. C., 147, 149, 152.
Owen, Dr. D. D., 11.

Paardeberg, 79, 99, 102.

„ Mountain, 36.

Paarde Kop, 67.
Paarl Division, 19, 20, 21, 79, 86, 99-106, 157, 158, 164, 198.

„ soils, mechanical analyses of, 198.
Palen and Rietvlei, F.C., 39.

Permanent value and immediate productiveness, 10, 11.
Peters, Dr. R,, 11.

217

Phokwani Hills, 142.

Phosphate starvation, 45.

Phosphates, Lack of, 49, 59, 60, 63, 74, 89, 110, 114, 120, 121, 134, 146.

„ „ retards nitrification, 23.

Phosphoric oxide, available, 2.

determination of, 7, 14, 15, 16, 17.

European limits of adequacy differ from Colonial, 12, 49.
needed for cereals, 45, 47, 72, 114.
poverty of certain districts in, 4, 23, 32, 33.
scanty store of, may be compensated for by lime, 17
Physical classification of soils, variety of methods, 189, 190.

,, composition of soils : (see also Mechanical analysis), 177, 188.
Physics, Soil, 5, 10, 11, 42.
Piquetberg Division, 20, 21, 107.
Plant food, 1, 2, 8.

„ abundant in doleritic soils, 27, 28, 32, 53, 162.
„ „ and plant food constituents, difference between, 8, 9, 159.

„ available, 2, 3, 7, 8, 9, 10, 12, 153, 159, 172, 188.
„ „ „ carried to the subsoil by rain, 172.

„ „ constituents, determination of ; (see Soil constituents).

,, ,, ,, third grade, Fluxes needed to extract, 9.

„ ,, „ various grades of, 9, 10.

„ ,, deficient in Malmesbury series soils, 158.

„ „ „ Sandstone soils : (see Sandstone soils, Poverty of).

,, „ derived from air, 1.

,, ,, ,, ,, humus, 1.

,, ,, ,, „ mineral constituents, 1, 2.

,, ,, ,, ,, soil particles, 1.

,, ,, ,, „ water, 1, 2.

,, ,, how conveyed, 2.

„ ,, in soil, influence of geological relations on, 157.

,, ,, mechanical fineness requisite, 1, 13, 188.

,, ,, most abundant in finest grades of soil, 13, 188.

,, ,, nitrogen as a source of, 1.

„ „ of first grade, affects land's immediate productiveness, 10.

,, ,, ,, brought to the surface by evaporation, 172.

,, ,, of second grade, affects land's permanent value, 10.

„ „ carried down to subsoil by rain, 172.

percentages, soil classification on basis of, 17.
readily available, brought to the surface by evaporation, 172.
reserve stock of, 9, 10, 11, 12, 153, 155.
solvents of, 2
Plettenberg Bay, 67.
Pniel geological series, 163.

„ „ „ soils, Analyses of, 163, 171.

Port Nolloth, 95.

Port St. John's District, 107, 171.
Portulacaria afra, 97.
Potash, determination of, 7, 8, 14, 16.

„ proportion of, in certain soils, 120.
„ scanty store of, may be compensated for by lime, 17.
Potato soils, Physical composition of, 190, 192.
Potteberg, 31, 132.

Pounding of soil furnishes misleading analytical results, 1' .
Pre-Cape rocks, 153, 161, 171.
Preparation of soil samples for analyses, 6.
Prieska Division, 20, 21, 60, 108, 171, 201.

,, soils, Mechanical analyses of, 201, 202.
Productiveness and fertility, 2, 10, 18. 188.
Purple shales and mudstones, 89.

Quarrie, Bredasdorp Division, 28, 30.
Queenstown Division, 19, 20, 21, 109, 146.

Radio-activity, 1.

Rain renders subsoils clayey, 192.

Readily available plant food, 9, 10, 12.

Reserve stock of plant food, 9, 10, 11, 12, 153, 155.

,, ,, „ „ and objects of soil survey, 10.

218

Rhenoeter Bush, 56, 57, 68, 96, 97, 102, 105, 139.

,, ,, an indication of poor soil, 56, 68, 96, 102.

Richmond Division, 20, 21, 110, 155.
Rio Grande silt, 109, 203.

River deposits, fertility of (see also Silts), 53, 60, 66, 70, 91, 98, 108, 113.
„ „ improvement of sandy soils by, 203.

,, „ retaining weirs for, 91, 109.

Riversdale Division, 4, 19, 21, 29, 90, 109, 111-114, 129, 154.
Robben Island, 41, 42.
Robertson Division, 4, 19, 20, 22, 32, 115-120, 129, 132, 133, 147, 155, 184.

Experiment Station, 115, 153, 156, 184, 199-201.

„ „ „ results of mechanical soil analyses, 200, 201.

Roode Zand Mountains, 135.
Root action, increase of, results in depletion of reserve plant food, 12.

„ variable in different plants, 3.
Roots, Acid excretions from, 2.
Rothamsted Experimental Farm, 11, 17.
Ruggens, The, 29, 112, 113, 116, 130.

types of Bokkeveld soil, 29, 30, 112.
Rumex acetosella, 40.
Russian agronomic surveys, 11.
Rust in cereal crops, 40, 81, 104, 149.
Rust-proof localities, 81, 82, 83, 102.

St. Helena Bay, 79, 84.

St. John's River, 107, 171.

St. Mark's District, 19, 21, 121.

Saldanha Bay, 79, 87.

Salisbury, Taungs District, 26.

Salsola aphylla : see " Ganna."

Salt bush, 26, 83.

Salts, Proportions of, in soil compatible with growth of crops, 174, 175.

Sand downs : see Blown sands.

Sandstone, Influence of, upon chemical composition of soil, 139, 159, 161, 163.

Sandstone soils, Poverty of, 24, 36, 38, 39, 42, 51, 53, 57, 59, 64, 75, 92, 93, 97, 99,

100, 134, 135, 139, 148, 150, 152, 159, 163, 164, 168.
Sandy soils, Physical composition of, 189.
Saussure, H. E. de, 1.
Scheme of mechanical soil analysis, 195.
Schryvers Hoek, F.C., 79, 87.
Schiibler, J. J., 11, 189.
Sea Point, 41.

Sedimentation in mechanical soil analysis, 1, 191, 194.
Sieves for mechanical soil analysis, 3, 7, 193.
Sifting of soil, 1, 3, 13, 154, 193.
Silicates and plant food, 9.
Silt, application of the term, 190, 191, 195.

„ importance of, in mechanical soil analysis, 191.
Silts, 24, 91, 108, 113, 170, 171, 191.

„ and clays richer than coarse grained soils, 13, 24, 53, 191.
,, deposited by rivers, analyses of, 171.
Simondium, 104.

Situation, unfavourable, effect on productiveness, 18.
Slates, Malmesbury : see Malmesbury series.
Snyder, Prof. H., 11, 190, 192.

Sodium sulphate and carbonate in alkaline soils, 173.
Soil analysis, Prejudice against, 11.
„ constituents, determination of *7.

chlorine, 7, 8.

lime, 7, 13, 15, 16.

moisture, 7.

nitrogen, 7, 8.

organic matter, 7.

phosphoric oxide, 7, 14, 15, 16, 17.

potash, 7, 8, 14, 16.

chemical analysis of

its value, 2, 5, 8, 11.

olutriation of, 1.

exhaustion, 45.

extraction by boiling hydrochloric acid, 126, 155.

219

Soil, extraction methods, comparison of, 59, 63, 94, 153.
fertility, due to various causes, 1.
fine state of division requisite, 1, 13.
mechanical analysis of : see Mechanical analyses,
moisture, the medium for feeding plants, 3n.
physics, 5, 10, 11, 42.
particles, source of plant food, 1.

rounding of before analysis, gives misleading results, 13.
sa ^ples, choosing of, 6, 157.

collection of, 6, 19, 20.
,, preparation of, for analysis, 6.
survey carried on single-handed, 4, 19.
commencement of, 4.

object of, to ascertain reserves of plant food, 10.
equipment for, 19.
of Arkansas, 11.
,, Kentucky, 11.
origin of, 3.

should include only virgin soils, 6, 54.
,, stopped by war, 4.
„ uniform methods needed for, 5.
surveys, Russian, 11.
So Is chemically poor cannot be productive, 18.
Solvents of plant food, Citric acid, 2, 9, 10, 12, 13.

„ Hydrochloric acid, 2, 9, 10, 12, 13.
„ „ „ natural and laboratory, 2, 8, 12.

„ water, 2, 3, 9.

Somerset East Division, 20, 22. 121,
Sour veld: see also Acid soil, 116, 118, 132, 137, 139, 149.

„ ,, soils, deficient in plant food, 137.
South Middelveld, F.C., 90.
Spekboom : see Portulacaria afra.

Standard method of chemical soil analyses, 6, 7, 12, 13.
Steenbok zuuring, 40.

Stellenbosch Division, 20, 22, 79, 122-126, 155, 157, 198.
Steynsburg and Colesberg soils compared, 128.

Division, 20, 22, 126-128, 155, 180-184, 199.
Stockenstrom Division, 20, 22, 128.
Stormberg and Uitenhage series soils compared, 170.
geological series, 52, 53, 128, 168, 169.
Range, 146.

„ series soils, analyses of, 169, 171.

,, ,, ,, water retaining capacity of, 171.

Stow, G. W., 141, 143, 161.
Stuurman, F. C., 108.

Suitability of soils for specific crops, Reasons for, 192.
Sulphuric acid as a solvent, 12.
Sulphur waters, 54.
Sundays River, 136.

„ beds, 137.
Swellendam Division, 4, 19, 22, 29, 30, 32, 38, 112, 115, 120, 129-134, 139, 154, 164,

165, 166, 171.
Systematic investigation of Cape soils : see Soil survey.

Table Mountain and Bokkeveld series soils compared, 165.

geological series, 24, 30, 36, 37, 42, 51, 56, 57, 59, 64, 68, 75, 92, 93,
98, 99, 100, 115, 134, 139, 147, 148, 149, 150, 152, 155, 157,
163, 164, 167, 168, 171.

„ „ series soils, average proportions of fine earth in, 196.

„ „ „ „ chemical analyses of, 164, 171.

„ „ „ „ low water retaining capacity of, 171.

Taungs, 20, 26, 142, 143.

Texture of soil, an indication regarding fertility, 41.
Thaer, 1.
Thebue, 127, 180-184.

„ soils compared with Calif ornian alkali soils, 182.

„ „ mechanical analyses of, 199.

„ water from, analysis of, 183, 184.
Thomas River, 47.

220

Tobacco soils and tobacco farms, 30, 77, 107, 128, 131, 135, 136, 138.

Touws River, 132, 148, 149, 171.

Tradouw, 115, 129, 132, 154.

True soil, application of term, 7, 193.

Tulare, California, 179, 182.

Tulbagh Division, 20, 22, 134, 150, 159, 164.

Tull, Jonathan, 1, 3, 6.

"Turf," 92, 102, 103, 116, 117, 131.

Tygerberg, 38, 44.

Tygerhoek, 33, 36.

Tyumie, 141.

Uilen Kraal, F.C., 33.

Uitenhage, Burghersdorp, and Stormberg series soils compared, 170.

geological series, 137, 139, 169.

Division, 20, 22, 136, 170.
,, series soils, Analyses of, 170, 171.

Umsikaba beds, 71.
Umtata District, 20, 22, 137.
Umzimkulu District, 20, 22, 137.
Uniform methods, need of, 5, 6.

„ „ obstacles to, 5.

Uniondale Division, 4, 19, 22, 56, 57, 96, 138-141, 145, 164, 165, 197.
United States Department of Agriculture, Soil work of, 19, 189, 190.
Upington, 60.
Upper Gouritz River, 90.
Upper Hantam, F.C., 52.
Upper River Zonder End, F.C., 33, 36.

Vaal bos (Atriplex), 26, 83.

" Vaal grond," 93.

Vaal River, 25, 143, 171.

Valsch Rivier, 111, 114.

Van Helmont, 1, 2.

Van Wyks Vlei, 45.

„ „ alkaline soil at, 185-187.

„ „ analyses of water from, 186.

Variety in physical analyses of soils, 190, 191.

classification of soils, 189, 190.
Vergennes clay, 203.

Veterinary Surgeon, Government, opinion of regarding phosphate deficiency, P..
Vette Rivier, 111.

Victoria East Division, 20, 141, 167.
Virgin soils, analysis of, 11, 54.

"Vlei" soil, 57, 68, 70, 92, 96, 116, 117, 132, 142, 150.
Vlei, The (Caledon Division), 33, 34.
Vogel Rivier, 121.
Voor Cogmans Kloof, F.C., 115.
Voorste Bosjesveld, F.C., 147, 150, 152.
Vryburg Division, 20, 22, 141-144, 162, 163.

Water as a solvent of plant food, 2, 5, 9, 11.

„ extraction of soil by means of, 8, 153, 175.

„ theory, Van Helmont's, 1, 2.
Wagenboom Range, 132.
Wagenbooms Rivier, F.C., 147, 150, 152.
Wagenmakers Vallei, 99, 104, 106.
Wallace, Prof. R., 5.
Water, analyses of, 183, 184, 186.
Waterfall gravelly loam, 66.
Waterfalls, F.C., 134.

Water retaining capacity of soils, 171, 188, 192, 193.
Water table, height of, in alkaline soils, 174n., 177.
Weirs for river silt deposits, 91, 109.
Western Province, 3.
Wellington, 79, 99, 103.
Wheat soils, American, analyses of, 9n..

,, ,, physical composition of, 190.

White brack, 174, 184.

221

Whitney, Prof. M., 1, 3n., 5, 18n., 19, 189, 191.

,, „ soil classification of, 189.

Whittlesea, 110.

Wiley, Dr. H. W., 6, 12, 188, 192.
Willowmore Division, 20, 22, 24, 145, 167.
Willowvale District, 19, 22, 145.
Winterhoek, F.C., 134, 135.

„ Mountains, 135, 159.

Witteberg geological series, 24, 115, 147, 150, 154, 155, 167.

„ „ soils, analyses of, 167, 168, 171.

Witzenberg Range, 135.
Wizard's Vale, 77.

Wodehouse Division, 22, 110, 146, 169.

Worcester Division, 19, 20, 22, 147-152, 159, 164, 165, 166, 167, 168.
Wynands River, 96.

Valley, fertility of, 77.
Wynberg, 41.

Zitzikama, 63, 167.
Zoetendals Vallei, F.C., 28.
Zonder End River, 36, 129.
Zout Rivier, Bredasdorp Division, 30.
„ „ Malmesbury Division, 79.

Zuurberg Range, 24, 167.
Zuurbrak, F.C., 129, 131, 134.
Zuurveld : see also Acid soil or Sour soil, 23, 92, 112.

„ lack of phosphates in, 23.
Zwartberg Range, 75, 77, 99, 161.
Zwartbergs River, 75.
Zwartland, 78, 85.

Zwartputs, Barkly West Division, 25, 26.
Zwart River, Caledon Division, 33, 34, 36.

„ ,, Knysna Division, 67, 69.

" Zwart turf " soil, 92, 131.
Zwart Water, 79, 87.

I