Front Cover
 Title Page
 Table of Contents
 Land classification and land-use...
 Types of land classification
 The physical classification of...
 Land inventories
 The U.S. soil survey
 Land types and soil types
 Land-use classes
 Agricultural regions
 The use of natural vegetation as...
 Systems of land classification
 The estimation of productivity
 Appendix: Outline of a U.S. soil-survey...
 Back Cover

Group Title: Imperial Bureau of Soil Science. Technical communication
Title: Land classification for land-use planning
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00095710/00001
 Material Information
Title: Land classification for land-use planning
Physical Description: 90 p. : illus. ; 25 cm.
Language: English
Creator: Jacks, G. V. ( Graham Vernon )
Donor: unknown ( endowment )
Publisher: Imperial Bureau of Soil Science, United Kingom
Place of Publication: Harpenden, England
Publication Date: 1946
Copyright Date: 1946
Subject: Land use -- Classification   ( lcsh )
Solo (Genese Morfologia E Classificacao)   ( larpcal )
Genre: non-fiction   ( marcgt )
Bibliography: Bibliography: p. 82-85.
General Note: Imperial Bureau of Soil Science. Technical communication no. 43
 Record Information
Bibliographic ID: UF00095710
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 04166054
lccn - a 49002841

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Title Page 1
        Title Page 2
    Table of Contents
        Table of Contents 1
        Table of Contents 2
    Land classification and land-use planning
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Types of land classification
        Page 10
        Page 11
    The physical classification of land
        Page 12
        Page 13
    Land inventories
        Page 14
        Page 15
    The U.S. soil survey
        Page 16
        Page 17
        Page 18
    Land types and soil types
        Page 19
        Page 20
        Page 21
    Land-use classes
        Page 22
        Page 23
        Page 24
    Agricultural regions
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
    The use of natural vegetation as an indicator of land quality
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
    Systems of land classification
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
    The estimation of productivity
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
    Appendix: Outline of a U.S. soil-survey report
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
    Back Cover
        Page 91
        Page 92
Full Text






'rice, 4s. od.

(Published by the Imperial Bureau of Soil Science, Harpenden, England)



2 Queen Anne's Gate Buildings, London, S.W.1.

Rothamsted 'Experimental Station, Harpenden, Herts.

The Reid Library, Rowett Institute, Bucksburn, Aberdeen.

Veterinary Laboratory, New Haw, Weybridge, Surrey.

School of Agriculture, Cambridge.

Penglais, Aberystwyth.

King's Buildings, University of Edinburgh, Scotland.

East Mailing Research Station, East Malling, Kent.

Winches Farm, Hatfield Road, St. Albans, Herts.

Shinfield, near Reading, Berks.

New Bodleian Building, Parks Road, Oxford.






Price, 4s. od.

(Published by the Imperial Bureau of Soil Science, Harpenden, England)


Made and Printed in
Great Britain by the


Land types and genetic soil types -
The estimation of population capacity -
Population and land quality -
Plant indicators of forest-site quality -
Grassland indicators -
I. Michigan -
2. U.S. Soil Conservation Service -
3. Tennessee Valley Authority -
4. New York State -
5. Western Canada -
Saskatchewan -
Alberta -
6. Ontario- -
7. New Zealand -
8. Great Britain -
9. Prussia -
10. Land classification for irrigation purposes
Productivity ratings -
The Storie index -
Canadian rating systems-
Saskatchewan -
Alberta -
Ranking coefficients -
German soil ratings-Bodenbonitierung -





- 20

- 29
- 30

- 34
- 37
- 40
- 42
- 47
- 50
- 52
- 52
- 53
- 55
- 56
- 56
- 61
- 64

- 69
- 7'

- 71
_ *729

- 73




Land classification in the sense here used relates to the grouping of lands
according to their suitability for producing plants of economic importance.
Such land classification must form the starting point of land-use planning,
which is very much in vogue at present, although few land-use plans based
on systematic land classification have been put into operation. As we shall
see, there is no sharp distinction between land classification and land-use
planning, and sometimes it is necessary or convenient to initiate a plan before
a classification is made. Land classifications group lands according to their
native characteristics, present use, yield capacity and so on [6], and these
classes are then assessed with reference to their suitability for the different
uses contemplated in the plan. Classification for land-use planning may be
purely qualitative-lands being classed according to their suitability for
different kinds of use rather than according to their inherent or man-made
productivity-or quantitative, when estimates are made of the relative
productivities of different classes of land under definite uses.
A land classification for planning purposes must be related to the
,objective of the plan. "The foremost essential in a possible list of Essential
Elements in Land Classification wrote Schoenmann [75] "should be a clear
and complete definition of why the proposed classification is needed and
how it will be used when completed." The first requirement for the
successful development and application of any technical system of land
classification wrote Kellogg [45] "is a clear understanding of the problems
for which the classification is needed." According to such statements it
might be concluded that no general principles of land classification can be
propounded, and that every land-use plan requires an ad hoc classification.
It can, however, be stated that every land-use plan-except, perhaps,
in countries fully developed economically and socially-for exploiting the
agricultural and forest resources of a region for the lasting benefit of the
-occupiers must be based on the conservation of soil fertility. Plans can,
of course, be made for the rapid exploitation of the capital resources of the
land, but even under the stress of war none has been proposed with this
objective. It may be assumed that nearly all the kinds of land-use plan
with which we are here concerned have a common primary objective-soil
conservation. They may have very different secondary objectives relating
to the ways in which the land can be exploited for the benefit of the occupiers,
but these secondary objectives are dependent on the attainment of the
-common primary objective.

The United Nations Conference on Food and Agriculture that sat at
Hot Springs, Virginia, in 1943 made certain recommendations for the
long-term organization of agriculture that may be taken to reflect the
common aspirations of land-use planners everywhere [92]. The Conference
recommended, inter alia:
(a) that the inherent natural and economic advantages of any area
should determine the farming systems adopted and the commodities
produced in that area ;
(b) that farming systems should be so designed as:
(i) to maintain soil fertility at levels which will sustain yields and
ensure adequate return for labour ;
(ii) to protect crops and livestock from major pests and diseases ;
(iii) to favour steady employment throughout the year ;
(these three ends, in general and save in exceptional circumstances, can best
be assured by balanced mixed rotational farming and by avoidance of
single-crop production, or monoculture;)
(c) that production of nutritionally desirable foods which can be
obtained from elsewhere only with difficulty or not at all is a special obli-
gation of the agriculture of every country ;
(d) that in every region subject to drought (occasional or in the form of
sharply marked periodic dry seasons) suitable measures should be under-
taken, partly by storage and partly by diversification of production and
development of water resources and cultural practices ;
(e) that land used or likely to be required for agriculture should be
protected from erosion ;
(f) that the spread of existing knowledge by education and the develop-
ment of new knowledge by research should be constantly promoted, and
that in these matters nations can co-operate to great advantage.
Nearly every objective in this list requires, directly or indirectly,
the maintenance of soil fertility. In one region the emphasis will be on,
say, the control of erosion, in another on the maintenance of continuous
employment, but the different aspects are interdependent ; steady employ-
ment helps to sustain the public interest without which soil-conservation
measures are likely to fail in their purpose. The planner and land classifier,
however, will usually concentrate on the predominant factor in land
utilization, but will not forget the others.
The purpose of land classification for land-use planning has been
described as to determine the types of production, use and service that
can be obtained from the land that will yield the highest social and economic
benefits to the people dependent thereon [12], or the uses of land that are
the best for all of us together, that is, for the general public [6]. Its
function is to present the best known facts about land use that in turn
would be of assistance to the people on the land in making decisions for
themselves [12]. It matters little what social, economic or political goal
the planner has in view. In the United States land-use plans are made on
the assumption that they will operate in an individualistic, competitive

society, in Russia on the assumption that they will operate in a state-
controlled, communistic society. In each case a different path must be
trodden to reach the desired goal, but whatever the form of society aimed
at, the land must be used in such a way that its fertility is at least main-
tained, for no form of stable society can persist long in a deteriorating
environment. Land classes for land-use planning purposes, therefore,
must be distinguished primarily by the different uses to which they can be
put without losing, or by the different measures which must be taken to
conserve, their fertility. Soil conservation is the physical objective of all
agricultural land-use planning, and provides a common basis for the formu-
lation of certain general principles of land classification. Every land-use
plan is essentially a measure of soil conservation. When it is nothing else
the corresponding land classification is comparatively simple, for the
criteria for distinguishing land classes are merely the characteristics signifi-
cant for soil fertility ; but when the land-use plan has, as is usual, some
other specific object such as the production of a certain crop, regional
self-sufficiency, the liquidation of debt, the development and use of water
power, the establishment of commercial farming, or what you will, then
a special ad hoc system of land classification has to be devised for each
objective, and no general principles can be laid down.
Land classification for land-use planning is therefore based primarily
on soil-fertility characteristics, and secondarily on whatever characteristics
of the land are significant to the specific objective of the plan. For example,
the specific objective might be to develop a perishable crop which requires
immediate transport to the consumer. Here roads and railways would be
important characteristics in land classification, but the basal classification
would still be a fertility classification indicating which areas would safely
and successfully produce the crop in question, and which should be devoted
to alternative uses. There is no relationship between the two sets of
characteristics-the physical, chemical and biological properties of the
land on the one hand, and the available transport facilities on the other.
Two (in many actual cases, several) distinct classifications are involved
which have to be combined into one. This is one of the main reasons for
the indefiniteness of many land classifications ; they are compounds of
several which are not necessarily interrelated.
It is usual to distinguish between the physical, and the economic and
social, classification of land. Kellogg and Ableiter [46] and many others
insist that the physical must be considered quite separately from the
economic and social characteristics of land. The physical or inherent "
characteristics include the geology, climate and topography and are
permanent in the sense that they cannot be profoundly altered by man.
Geology, climate and topography are combined in the factor soil which,
though to some extent modifiable by man or other living organisms, is the
main factor in a physical classification. A physical land classification is
permanent, and once made can serve as the basis on which to superimpose
an economic and social classification which will need to be modified with

changes in economic and social conditions (e.g., market conditions and
communications) as well as with the changes in the objective of the plan
that may occur as the plan unfolds. Soil fertility, however, as distinct
from the separate physical factors that comprise soil, is not immutable.
Man can easily change the level of fertility up or down, but whether or not
he does so is determined mainly by economic factors. In theory, every
soil can be brought by appropriate measures to the optimal state of fertility,
but in practice the application of the required measures is often limited by
economic factors. The Sahara could be made fertile by irrigation and
manuring, but it would not pay. Chemically and physically poor sandy
soils might be classed as suitable for intensive market gardening if favourably
situated with regard to markets and supplies of manure, but quite unsuitable
if market facilities were poor or costs of manure were high. The significance
of these factors will vary continuously with the social and economic
development of the country or region.
Soil fertility thus has as much claim to be considered an economic as a
physical characteristic of land.
Much of the work on land classification has been carried out by
economists. As we are writing primarily for the agriculturist we shall
approach the subject more from the agricultural than from the economic
standpoint, but it is important to recognize that the determination in
practice of optimal land use is essentially an economic problem. In the
last analysis the success of any land-use plan depends on its economic
soundness. Gray and Regan [27], writing as economists, state that :
" Land classification is aimed at determination of economic best use from
both the entrepreneurial and the social standpoints; although the best
use of land does not imply ignoring the best use of the other factors of
"An important task in land classification is determining the advisability
of the advance, the continuance, or the retreat of agriculture-or in con-
ventional terms, the location of the extensive margin of cultivation ;-for
the one or the other of these trends has important implications with respect
to institutional patterns, capital and debt structure, public finance, and
numerous other considerations."
Astor and Rowntree [4] distinguish between the optimum and
maximum fertility level of an agricultural soil. Optimum fertility levels
are determined by the ratio between input (i.e., total costs of production)
and output (i.e., total sales). They vary not only between one crop and
another, and one soil type and another, but also between the same crops
and soils in different localities. Maximum fertility is determined simply
by yield, either yield per plant or yield per acre, and the weight of the crop
is not always as important as the quality.
The factors to be taken into account in an economic and social classifi-
cation will vary both as to their nature and the weight to be attached to
each, with the social objective of the plan. In some instances markets, in
others communications, in others tradition and custom, in others political

questions may constitute the dominant factor, and there will always be many
variable and fluctuating subsidiary factors. Veatch [97] includes among
the essential economic data uses of land, kind of ownership, land values,
transportation facilities and markets for present and potential products;
and among social data, distribution and kind of population, both rural and
urban, and economic status (standard of living) of rural classes. Allin [3]
includes among the land-use factors which must be taken into consideration
in America such things as the passing of the physical frontier, the change
of this nation from a debtor to a creditor status, the loss of foreign markets,
the approach to a stable population, and the evident conflicts that have
arisen in our economy between capital and labor." By its very nature
social-economic land classification must be labile and indefinite.
In undeveloped country where the social factor has not operated
appreciably, a physical land classification is applicable without great
modification to land-use planning. Indeed, the purpose of land-use
planning in such countries is to avoid the costly trial-and-error process of
crop adaptation that has hitherto prevailed. The idea is, instead of planting
a crop on inferior land because it is near a railway, to plant it on the most
suitable land and then bring the railway to it. In long-settled countries
the social-economic factor tends to predominate in land-use planning.
Land use has largely adapted itself in the course of time to the physical
environment, and any planning that may be thought necessary is concerned
mainly with adjustments to changed social and economic conditions.
A land-use or use-capability classification, if it is to have permanent
or long-term validity, should also allow for probable developments in
technique and science, e.g., the production of frost-resistant crop varieties,
new kinds and sources of fertilizer, new methods of cultivation [25]. Land
in England that until recently was commonly regarded on account of its
heavy texture as permanent-pasture land proved itself during the war to
be first-class arable land easily cultivable by modem machinery. But
although a classification should allow for such developments they cannot,
of course, usually be foreseen.
It is, in fact, impossible to forecast with any accuracy either economic
or scientific developments, and a practical land classification which takes
them into account must necessarily be vague. There is ample evidence "
writes Hudson [39] that the end product of land classification consists
largely of ascertaining trends in land use within definite areas and that the
function of land-use planning agencies is to facilitate and direct these
trends towards a definite goal rather than to impose some preconceived
land-use plan. The goal towards which land-use trends is guided is . .
' the utilization of lands for their greatest value'."
The nature of land-use planning is described by Lilienthal, writing
of the T.V.A. plan to develop the land and resources of the Tennessee
Basin [49].
In the unified development of resources there is such a Great Plan : the Unity of
Nature and Mankind. Under such a Plan in our valley we move forward. True, it is
but a step at a time. But we assume responsibility not simply for the little advance we

make each day, but for that vast and all-pervasive end and purpose of all our labors, the
material well-being of all men and the opportunity for them to build for themselves
spiritual strength . .
Not one goal, but a direction. Not one plan, once and for all, but the conscious
selection by the people of successive plans. It was Whitman the democrat who warned
that the goal that was named cannot be countermanded."
If this conception of planning is sound, as I believe, then it is plain that in a democracy
we always must rest our plans upon "here and now," upon "things as they are." How
many are the bloody casualties of liberal efforts to improve the lot of man, how bitter
the lost ground and disillusionment because of failure to understand so simple and yet so
vital an issue of human strategy. So frequently have men sought an escape from the
long task of education, the often prosaic day-by-day steps to "do something about it,"
by pressing for a plan-usually in the form of a law-without considering whether the
people understand the reason for the law's plan, or how they are to benefit by it.
An unwillingness to start from where you are ranks as a fallacy of historic pro-
portions; present-day planning, anywhere in the world for that matter, will fall into the
same pit if it makes the same gigantic error. It is because the lesson of the past seems
to me so clear on this score, because the nature of man so definitely confirms it, that there
has been this perhaps tiresome repetition throughout this record : the people must be
in on the planning ; their existing institutions must be made part of it ; self-education
of the citizenry is more important than specific projects or physical changes.
The points made by Lilienthal are worth emphasizing. Firstly, that
a land-use plan is dynamic; it changes as it unfolds. We can see the
goal-the highest attainable level of fertility-but we are lucky if we can
see more than one step forward at a time. Secondly, that the people must
play as intimate a part in the execution of the plan as the crops and animals
they tend. This is a political matter ; a people controlled by an autocrat
is a more pliant agent in executing a plan than a nation of individualists.
Thirdly, that the present state of society should determine the next step
forward. Here again, an autocracy can take greater liberties than a
democracy, but whatever the polity an appreciation of the facts of human
ecology is essential.
There are not enough examples of successful land-use plans to enable
us to dogmatize upon whether the people must be in on the planning ",
but the evidence available-mainly from the United States-indicates
that the active, as distinct from the passive, participation of the people
in all the operations can be a decisive factor in assuring success. If the
people are to participate actively they must understand what is being
planned, and they must understand the land classification on which the
plan is based. Several writers have stressed the importance of having a
system of land classification that is crystal-clear to the ordinary layman.
Such a system is not likely always to satisfy the taxonomic pundits; but
from the practical standpoint more is usually gained than lost by using a
classification which does not need, and sometimes will not bear, thinking
about. A unique feature of land-use planning is that the people not only
make and execute the plan, but themselves constitute one of the chief
things to be planned.
Gross [28], in an analysis of the failure of the nation-wide county
land-planning programme launched in the United States in 1938 by the

Mount Weather Agreement, attributed failure in the first place to the fact
that no desire to solve community and county problems was created in the
population of the areas in which the planning programmes were to function.
" Experts can work out beautiful expositions of problems, they can present V
exclusively alternate solutions, but unless a self-interest to achieve these
results is established, the expensive work of the expert has little or no
value." In Gross's opinion the prime task of the land planner is not to
solve certain immediate problems, but to develop a community determi-
nation to solve them. It is a task which the experts in geography, agri-
culture, soils, economics and so on are not always qualified to perform.
If county land planning recovers from its present demise the specialist in
a future planning organization should be made cognizant of his purely
advisory capacity."
That land-use planning is essentially an exercise in human geography
is obvious. It is an exercise to the performance of which many non-
geographical sciences have much to contribute, but the approach to it must
remain geographical. Colby [16] distinguishes four phases of the land-
planning process (i) survey and classification, (2) appraisal, (3) design, and
(4) effectuation, and gives the following description, taken verbatim from
his paper, of a geographer's approach to it.
The survey and classification phase of land planning calls for a critical examination
of the occupance pattern as it is to-day. For this work, the topographic and other
customary maps and the census and other statistical materials are useful, but they do
not deal with all of the features and conditions which in association make up the occupance
pattern. Such maps and materials, moreover, are not sufficiently detailed, for as McMurry
has pointed out, "as soon as land planning work leaves the theoretical plane, and the
problems assume practical application minute detail is essential." Land planning
requires an analysis in map and other forms which distinguishes the component elements
in any part of the occupance pattern separately or in their relation to the whole. Fortu-
nately, two recent advances in technique make this practical. Aerial mosaics, the great
contribution of air photography to mapping, provide the base. Site analyse., recorded
in the fractional code, an analytical technique developed co-operatively by geographers
in the last decade, furnish the primary materials. When once these materials are in hand
classification of the associated elements or of any individual elements becomes possible.
Detailed site analysis has the disadvantage of being too slow and costly for many
types of land planning. In attempting to solve this problem, experiments have been
tried with the sampling method, the traverse method, and the unit-area classification.
In the sampling method, sample surveys are made at either planned or random intervals.
In the traverse method, the surveys follow lines or strips across the area under exami-
nation. Both of these methods involve perplexing statistical questions. The unit-area
type of work contains an element of classification as well as analysis. For the writer,
the idea of unit-area classification carries back to a field conference in 1925. At that
time I became hopeful that we might be able to perfect a technique by which one could
recognize the entity of any point or area and then map that entity as far as it extended.
The idea was tried out in rapid traverse work over long distances with some success and
was reported at a meeting of the Illinois Academy of Science in 1933. In the- past year
further experiments with the method were set up but the results were unsatisfactory
until the problem was allocated to the geography section of the Tennessee Valley Autho-
rity. After some experimentation, the men of this section learned how to combine
the idea with the aerial mosaic-site analysis technique. The site was widened to an

appropriate area, classification was added to the analysis and the unit-area type of
land classification became a reality.*
The second phase of land planning calls for an appraisal of the occupance pattern
in order to discover its points of strength and its points of weakness. This involves a
study of the spacing of individual elements in the pattern and the relation of each
individual pattern to the composite pattern. It also demands that proposals for changes
in any individual pattern be weighed as to the probable effect on other individual patterns
and on the composite pattern. The ramifications of this type of work are many. It
leads to investigations of the assets and liabilities of any area for particular uses. It asks
if particular sites are deteriorating under use, and if so how rapidly. It implies analyses
of slope and soils of a new degree of intensity. It calls for much greater knowledge of
climates of slopes and thermal belts than we have at present. In this connection
practical experience is placing in action some of the investigations advocated by the
Mississippi Valley Committee, or by the Land Use Committee of the Science Advisory
Board. This phase of land planning also studies the adjustment or cultural features to
natural features, seeking areas or points of satisfactory or of unsatisfactory adjustment.
Finally, land planning raises the all-important question as to whether or not the
occupance pattern in its present status serves and promotes the physical and social
welfare of society. Many of these questions call for painstaking research on the part
of geographers or other scientists. Some call for evaluation by men of long experience
in business. Land planning does not care who does the work, but it does care greatly
for the results.
The third phase of the planning process is that of design or active planning for the
attainment of stated objectives. In land planning this means the formulation of a
design for the development of a progressively better occupance pattern. Design begins
with a preliminary intellectual conception of something to be done or produced. It
implies a penetrating knowledge of the existing occupance pattern and a full understanding
of the objectives to be attained. In the acquisition of this knowledge, geography, as we
have seen, plays a highly important part. It may be assumed, and we have some evidence
to support the assumption, that the momentum of understanding gained by geographers
during the survey, classification, and appraisal phases will enable them to make valuable
contributions in the formulation of the design. Purposive planning, after all, may be
thought of as the formulation and prediction stage of a sequence of scientific exploration.
In land planning the sequence includes observation, analysis, classification, appraisal,
formulation and prediction. Geography, with its contributions to the early stages of the
process, may be expected to contribute significantly in the later stages.
Land planning by its very nature may be expected to introduce a rational element
into social planning. Land planning, like geography, insists on the areal concept. It
demands that the social structure be held in its physical setting. It strikes at reality
and tends to hold reasoning within a rational frame of reference. It is the intelligent
application of the principles of geographical science and other sciences to the problems
and interests of society.
In the final phase of the planning process, that of effectuation, geographical science
has, as far as I can see, little or nothing to offer. The question of putting plans into
effect is a matter of administration and rests on the interest and desire of the general
public. The men who administer planning programs need not be scientists ; in fact, it
is better if they are not interested in promoting a particular science. They do need to be,
however, administrative artists in applying the findings of science to the solution of
problems of public welfare.
The best known and probably the most successful land-use plans in
operation are those of the American soil-conservation districts. In July,
1945, 3,404,000 farms covering 734 million acres were included in the 1,328
soil-conservation districts which had then been set up [78]. The prime
objective of every soil-conservation-district plan is the same-to save the
See p. 47.

soil from exhaustion and erosion. Quite a simple physical classification,
defining land in terms of its erodibility, suffices as the basis of the plans
(see p. 42). Social and economic factors do not enter into the land classifi-
cation, but they do enter into the effectuation of the plan, and may
necessitate modifications in the ideal plan based on strict adherence to
the physical land classification. The plans are made and carried out by
the occupiers of the land themselves who can call upon outside experts to
give advice and direction, if required.
Bennett [8] states that the guiding principle of the Soil Conservation
Service in planning land use has always been that the effective prevention
and control of soil erosion and the adequate conservation of rainfall require
the use and treatment of the various kinds of land, each according to its
special needs and capabilities.
In applying this general rule, however, it must be borne in mind, firstly,
that the use and treatment of any land are determined not only by physical
characteristics, but also by such factors as available facilities-implements,
power, labour, finance-and the preferences, skill and intelligence of the
occupier. The occupier, as well as his land, may require treatment".
Secondly, each parcel of land has to be considered in its physical
relation to adjoining land ; optimal use of high-lying land may be deter-
mined as much by the use of the lower-lying land as by its own physical
Other examples of land-use planning according to a land classification
are afforded by some Russian collective farms (kolkhozes). These plans
have to fit in with the larger production plans of wide regions that in turn
are co-ordinated into a national plan by the central planning authority
(Gosplan). The kolkhoz is required to produce certain amounts of certain
crops, and is expected to adopt certain rotations and farming practices
which have been shown to conserve soil fertility in the region concerned.
A fair amount of freedom in land use is permitted, but the general framework
of farm management is predetermined. Under such circumstances a
physical land classification to suit the preconceived plan, which may be
modified within limits to fit the land classes, can readily be made. Social
and economic factors enter to a variable, but usually only to a minor, extent.
Both the soil-conservation districts and the kolkhozes represent parti-
cularly straightforward examples of land-use planning because the objectives
are clearly defined and are definable directly in terms of land use. Where
the objective is less clear, or less clearly related to definite uses of land, the
land classification tends to be correspondingly vague. An objective such
as the common welfare, the public good, the improvement of the standard
of living, or merely the optimal utilization of the land forms an insufficient
basis for a land classification ; it is necessary to state how the objective
is to be attained-for example, by an economy based on cocoa, sugar, mixed
farming, etc. The Scott Committee on Land Utilization in Rural Areas [571
was instructed to consider the planning of post-war England and Wales.

It assumed* that English agriculture would be characterized by a con-
tinuance of the traditional type of English farming (balanced, mixed
farming with small, individually owned farms predominating), and on this
assumption recommended as a basis for land planning the adoption of the
classification of the Land Utilization Survey (p. 57) that distinguishes
classes according to former use under the traditional system. Here the
relationship between the system of land classification and the means by
which the planning objective is to be attained is obvious. If an entirely
new system of agriculture had been proposed, a classification according to
former use would not have been appropriate.
For land-use planning, therefore, there is a very close connexion
between the objective of the plan, the means proposed to attain it, and the
appropriate system of land classification. There is no such thing as a
general land classification, but there is a common physical classification
appropriate as a basis for all land-use plans aiming to promote and
perpetuate the public welfare, and therefore requiring the maintenance of
soil fertility. Superimposed on this basis is another social-economic
classification which is, in effect, part classification and part planning. It is
often difficult to know where the process of land classification ends, and
that of land planning begins.

The Land Committee of the United States National Resources Planning
Board, in its Report on Land Classification in the United States [62], dis-
tinguishes five types of land classification at present in use. These types
are not fixed or exclusive ; many of the land-classification projects
mentioned in the Report can be assigned to two or more types. But while
the types are somewhat arbitrarily defined to suit the special circumstances
of contemporary land classification in the United States recognition of them
brings out the general fact of the dependence of land classification on the
use which is to be made of the system. The five types are :-
Type I. Land classification in terms of inherent characteristics.
Type II. Land classification in terms of present use.
Type III. Land classification in terms of use capabilities.
Type IV. Land classification in terms of recommended use.
Type V. Land classification in terms of programme effectuation.
Of the 75 miscellaneous projects analysed in the Report, 68 are of, or
include, type-I classification. We have already referred to this type as a
physical classification. It is the least dependent on the use to be made
of the classification and is, moreover, almost essential to the further develop-
ment of any practical classification of land for agricultural or forest use.
The inherent characteristics to which most attention is usually paid are
soils and topography, though other natural land qualities assume equal or
greater importance in special cases (e.g., subsoil and water table in irrigation
Majority Report. A Minority Report by Prof. S. R. Dennison makes rather different recom-

planning). Land classification according to inherent characteristics must
obviously be the fundamental classification in undeveloped or partly
developed regions where man-made qualities such as present land use,
urbanization, political and economic factors, etc. have minor significance.
Type-II classification, according to present use, may also be considered
as fundamental ", at any rate in long-settled regions where sufficient
time has elapsed for both the inherent and the man-made qualities of the
land to have reached an equilibrium as regards their influences on land use.
The classification of the Land Utilization Survey of England and Wales
(p. 57) is a type-II classification, and as such is applicable to a plan which
aims to rationalize rather than to change the time-tested, traditional system
of land use. The Land Utilization Survey originally defined its classes
solely in terms of present use and made no claims for the system as a land-
planning tool. Later, the purely type-II classification was expanded for
planning purposes into type IV (recommended-use classes).
Both type-I and type-II classifications are essentially inventories of
actual conditions. They take no account of potentialities. The modern
American soil survey, which is evolving into a land classification, is a
combination of types I and II with a recent admixture of type III (use
For type III a specified use. is assumed (e.g., cereal culture, pasture,
forest), and land is assessed according to its potential productivity in that
use. First-rate forest land may be fourth-rate cereal land. For crop use,
land may be assessed according to the estimated yield of the crop under a
given system of management that may be based on the average, optimal
or any other standard. The system of management can be a determining
factor in the classification. For example, land might be classed as high-
yielding under dry farming owing to excellent water-holding characteristics,
and as low-yielding under irrigation owing to defective drainage. In
forestry, the classification of site quality according to actual production or
according to production estimated from the nature of the vegetation is an
example of type III.
An alternative form of type-III classification is according to the
measures which must be taken to bring the land to a certain standard
condition, e.g., immunity from erosion. This form is much used in the
U.S. soil-conservation surveys. The knowledge required for a type-III
classification always includes much of that which is obtained in making
a type-I classification.
Type-IV classification (in terms of recommended use) is in effect a
combination of the preceding types. It is not so .much an actual classifi-
cation of land as the formulation of a land-use plan on the basis of the
inherent properties, present use and capabilities of the land. The same
is true of type V-land classification in terms of programme effectuation-
which involves indicating on a map the different stages and methods whereby
land-use recommendations are to be carried out. Both in type IV and
type V complex political and social questions such as the form of govern-

ment, the aspirations of the community, the legal power of the planning
authority and the rights of landowners are involved. The five types
represent a series of increasing complexity and afford a good illustration
of the way in which land classification, starting in its simplest and most
fundamental form as a soil classification, may become merged into land
planning without any sharp distinction between them.

In a posthumously published and, unfortunately, unfinished paper
C. F. Marbut [53] makes a clear distinction between a natural and an
" economic classification of land:-
In the broadest sense possible the classification of lands may be considered from two
points of view. In one, lands are considered as natural bodies, are differentiated into-
units and the latter arranged into classes on the basis of inherent natural characteristics.
Such a classification has no necessary relationships to man or his institutions. The
other classification, however, has a definite relationship to man. It may be called an
economic classification. It is a classification based definitely on the relationship of
different kinds of land to man and his industries. In the first case, in the physical
classification, the units are being defined on the basis of definite character, and express
therefore essential and inherent, permanent characteristics. In the second case, or the
economic classification, being based on a relationship that varies with the status of man
and his institutions at any given time, the classification is not permanent. The definition
of land units for the same parts of the earth's surface differs from time to time, depending
on the changes regarding man and his institutions in a given spot at any particular time.
The lack of permanence is due to the changing character of man's institutions, not to.
changes in the inherent characteristics of lands.
Marbut asserts that land may be used by man in three main ways :
(1) as mere areas of solid earth on which to live ;
(2) as areas varying in desirability because of relationships to other kinds
of land, or to man and his institutions;
(3) as sources of usable material.
As we are here only concerned with land use for plant production our
land classifications are not concerned with the first two uses, and need be
related only to those characteristics of land which determine the kind and
amount of plants produced. As a basis for such a classification Marbut
proposes the following definition of land. Land is a term connoting all the-
features of man's natural geographical environment with which he deals in
crop production.
Marbut proceeds to establish the bases for a classification according to
the permanent physical properties of land that are significant in plant
production. Such a classification is the equivalent of a potential land-use
classification, lands being grouped according to their physical capacity for
producing different kinds of plants, regardless of whatever use is or has
been made of them. No attempt has been made to determine the kinds
of land here discussed by any sort of indirect approach such as the type and
apparent success of the local agriculture, the kinds of crops grown or the
yields obtained or the character of the natural vegetation. Such an
approach to the problem is consciously avoided because the essential

character of a physical classification requires a direct approach through
physical character. It is well known and generally agreed by students that
such indirect methods may give certain definite results. It is possible to
identify' good or productive land by a study of the character of the natural
vegetation growing on it or that grew on it. It is equally possible to arrive
at the same conclusion through a study of the existing agriculture. Not-
withstanding this possibility a serious objection to such methods lies first
in the fact that the result obtained is wholly empirical and neither the facts
nor the conclusion presents the possibility of drawing any other conclusion
than that the land is productive for the plants grown."
On the other hand, a classification based on inherent physical character-
istics provides a basis for drawing conclusions regarding the use of land for
more than one purpose.
The most important of the features of man's geographical environment
with which he deals in crop production are soil and climate, these con-
stituting the main part of the physical environment in which plants grow.
On the plant side the two main factors to be considered are (i) adapta-
bility to, and (2) productivity in, the environment. It is well known that
certain plants attain their optimal development in certain regions (e.g., corn
in the Corn Belt in the U.S.A.), whereas in other regions owing to the
presence of some limiting factor such as too short a growing season, too
much or too little moisture, too hot or too cold a temperature, growth is
partially or completely inhibited. The factors influencing adaptability
are mainly, but not entirely, climatic. Within any one, more or less
homogeneous adaptability or climatic region, however, great variations in
productivity for a given crop may occur, associated mainly, but not entirely,
with variations in the soil factor. Variability of productivity is greatest
in regions of optimal adaptability where the possible range of production per
unit land area is from zero to the maximum known. In the so-called
marginal zones soil or productivity factors are less significant.
A third factor unrelated to adaptability or productivity that has to
be considered is accessibility-not to markets, since this is a purely physical
classification, but for agriculture. Accessibility concerns mainly the relief
of a region. Some lands, for example, are too steep, or have too broken
a topography for crop production, even though the climatic and soil
conditions are favourable, because of difficulties of cultivation or the hazards
of erosion. Accessibility in this sense is less a limiting factor in forestry
than in agriculture.
The physical land characteristics significant in plant production may
thus be grouped as (i) those which control the kinds of plants grown (mainly
climate), (2) those which control the productivity of the land for the plants
grown (mainly soil), and (3) those which determine the accessibility of the
land for crop production (mainly topography). It will be seen that the
proposed classification, though physical, is expressed hi agronomic terms-
in terms of the significance of the physical characteristics for agricUlture.

It is not at present possible, owing to our inability to assess quantita-
tively the agricultural significance of physical factors, to make more than a
very general classification of land on this basis. Marbut's proposals appear
to involve a first differentiation of lands according to adaptability, the
adaptability regions being then subdivided according to productivity and
finally accessibility. This corresponds roughly to a classification scheme
in which the highest category is determined on a climatic, the next on a soil,
and the third on a topographic, basis. How Marbut intended to define the
lower and, in practice, more important categories is not indicated in his
This outline of a classification scheme shows that Marbut intended it
to be a true classification-that is, lands are arranged in categories and
classes showing the physical relationships between the different groups.
The taxonomic importance of this feature of a classification is stressed by
the Land Committee of the National Resources Board [62]. Tidal
marshes, sandy soils, oil fields, and golf courses might be present in an
area, but they are not related phenomena and cannot be assigned to classes
in a logical system of classes. Tidal marshes, rocky headlands, and sand
dunes, however, are related phenomena and can be assigned to classes in a
systematic division of shore features. In similar fashion, golf courses,
railway yards, and airports are classifiable in terms of the several categories
of urban land use."

Some systems of land classification make little attempt to indicate
relationships between classes. Provided they fulfil their purpose they are
none the less useful for that, but they would perhaps be more correctly
described as land inventories rather than land classifications. Those who
employ inventory-type classifications usually define a land type in terms
of a summation rather than a synthesis of the separate characteristics.
Thus Kellogg and Ableiter [46] define a natural land type as "land having
a particular set of defined natural characteristics, principally of soil, climate,
relief, stoniness and native vegetation." Veatch [96] says that the
natural land types [in Michigan] consist of various combinations of the
soil types, relief features, and topographic forms such as hills, basins, lakes,
and swamps." According to Veatch, for instance, a land type could be
defined as consisting of soil type A, relief B, topography C and so on, and
another as consisting of soil type X, relief Y, topography Z, etc. These
definitions or descriptions suffice to give valuable information about land-
use potentialities, but they do not indicate how class A B C is related to
class X Y Z.
The definition of the type unit for classification is a point at which the
physical and social-economic classifications of land tend to clash. In a
physical classification the boundaries of a unit land type are naturally
those at which the physical factors change, though it is sometimes difficult
to determine the actual physical boundary to be marked on a map, or

whether a change in one factor only while all others remain constant
justifies the delineation of a land-type boundary. For an economic classifi-
cation or a classification for taxation purposes a property is often the most
appropriate land unit. In some German systems of Bodenbonitierung, for
,example, points are awarded for physical land features, and the points are
integrated over an entire property to arrive at the valuation for the property
rather than for its constituent land types. Kellogg [44] points out that an
important factor determining the best use of an individual farm unit is the
manner in which the various soil or land types are associated with one
another. For example, in a hill-valley section frequently good cropland
and land suited only to pasture are so mixed that the latter may be used to
.good advantage in a farm unit having some good cropland, without which
a different use, such as forestry or extensive grazing, might be more
desirable." The orientation of physical land types in relation to property
boundaries may be a determining factor in planning land use.
What is usually described as the physical classification of land is, in
fact, more a physical inventory than a classification. There are no univers-
ally established physical land classes, as there are universal soil classes or
,easily identifiable plant families and genera. It is impossible to compound
the total physical characteristics into a single entity land, because one of the
most important characteristics of land-at least for planning-is not
physical, but economic.
Haggerty and Meyers [29] list the following characteristics as required
to make a physical inventory of land as a basis for land classification.
Soil: Flood Relationships:
Profile Frequency
Texture Intensity
Structure Duration
Chemical reaction Occurrence
Content of organic matter and Physical effects
essential plant nutrients Land Condition:
Stoniness Erosion
Internal drainage Deposition
Topography : Depletion
Elevation Water Resources:
Lay-of-the-land Kind and extent
Degree of slope Dependability
Native Vegetation : Climate :
Type Precipitation
Amount or volume 3 Temperature
Quality Frost-free period
Climatic hazards
Haggerty and Myers recognized, however, the desirability of not merely
-describing land types, but of grouping them into classes according to the
nature of the adjustment required to fit the land-use plan.

In Germany the main items of a land inventory were soil, topography,
water resources, climate and economic conditions. The soil was further
inventoried according to genetic type, texture, depth, stoniness, cultural
condition and reaction. Much more attention was paid in Germany than
is paid in the United States to economic items-market facilities, transport
and communications, labour supply, system of land tenure, size and shape
of farms, etc. An additional item which was to have been systematically
inventoried was the content of available potash, phosphate and lime.
A special service-Bodenuntersuchung-Sonderaktion-was at the beginning
of the war engaged in making these determinations. The aim was to
sample and analyse every quarter hectare of agricultural land. Between
1936 and 1942 over 4,450,000 determinations of lime status, 3,000,000 of
phosphate status and 625,000 of potash status were made on soils of the
German Reich [30].
Nearly all the factors listed by Haggerty and Myers [29] are now
inventoried in the standard U.S. soil survey which has, in fact, developed
into a combination of a soil survey proper with a land-type inventory. To
an ever-increasing extent the U.S. Soil Survey is becoming a land survey,
and it is possible that if ever a universal system of land classification is
developed it will emerge out of the present work of the U.S. Soil Survey.
Milne [56] quotes the Chief of the Soil Survey as stating that its results
" form the factual basis in the development of sound programmes of land
use." Soil surveyors are constantly reminded that it is the land rather than
the soil that is to be investigated, classified, inventoried and interpreted,
primarily though not exclusively by the soil profile.
As the survey becomes more practical-more of a land and less of a soil
survey-there is, for example, a tendency to give more weight to texture
in the definition of soil type. The American soil type is a combination of
a series and a class, the series defining the general morphological character-
istics of the soils comprising it, and the class the texture of each soil at the
surface. Thus the series name Hagerstown describes an indefinite number
of soils identical in all important characteristics except texture (class).
Hagerstown loam, Hagerstown sandy loam, etc. are distinct but related
soil types differing principally in texture. But it has been found that
whatever may be the scientific justification of classing a clay and a sand
(say) in the same series, in actual agricultural use there is likely to be. a more
pronounced difference between the behaviour of a clay and a sand of the
same series than between two clays or two sands of different series. In
other words, texture, which cannot be accorded a position of very great
importance in a genetic or morphological system of soil classification,
becomes a dominant factor in a use classification.
The trend of soil classification in the United States from a morphological
to a use basis is reflected in the growing recognition of monotype series-
i.e., the inclusion of texture in the definition of a sdil series, or the identifi-

cation pf soil series with soil type. Soils formerly regarded as belonging to
the same series by reason of common genetic and morphological character-
istics are now classed as different series by reason of their different use
capabilities as indicated by their respective textures. At most three, and
more commonly one or two, textural classes are now included min one series.
While the soil type is thus tending to disappear as a separate taxonomic
category-or to be merged into the series-other categories are appearing,
some lower than the series (e.g., phases), others higher (e.g., associations).
The phases of a soil series or type are distinguished by some use characteristic
such as stoniness, degree of erosion or slope-slope being judged not by its
steepness, but by its effect on land utilization. An association is a group
of soils, either similar or dissimilar, geographically associated in a regular
pattern over a landscape feature. It is similar to the catena described for
tropical landscapes by Milne [55].
To charge the soil type with the maximum load of land-use meaning "
[6o] is the prevailing trend in soil survey in America at the present time,
but Krusekopf [48] considers that most soil surveyors, while duly inven-
torying important land characteristics such as slope, stoniness, erosion, etc.,
have failed in their task of interpreting the primary morphological features
of the soil in terms of land use, The soil surveyor, in fact, has the difficult
task of maintaining an objective, scientific approach to soil classification
and at the same time making the results of his work useful and intelligible
to those who know nothing of the restraints imposed by pedological dis-
cipline. Milne [56] describes the soil surveyor's approach to land classifi-
cation in the following words :-
The soil surveyor regards himself as a student of the land who, being a soil scientist
first, distinguishes his land types primarily by soil profile. He knows that considerations
regarding the use of land may be in error if they leave the soil profile uninvestigated, and
his own first duty, as a key man in this sense in the rationalization of rural production,
is the expert recognition of the varieties of soil profile that exist in his piece of country
and their reduction to a manageable number of series and types. In this stage of his
work he is a pure pedologist, adding to soil knowledge or establishing the validity of
existing concepts on new ground; he may classify and describe soil types in whatever
terms the technicalities of the subject require.
As he goes abput in the field plotting soil-type occurrences, he must, however, bear in
mipc that he will have the further duty of making clear, for all who follow him on land
questions, what the signific4Ace of his data is. He must endeavour to work to the
condition that aftpr he has finished, no further visiting of the ground shall be necessary
for the ascertainment of omitted or obscured facts regarding land type. He must,
therefore, pay attention to a good deal besides soil profile. His land types are finally
to be so defined (having regard to the limitations of the scale on which he is working)
that the most desirable use of them can be inferred by other specialists (such as agrono-
mists, foresters, or the administrators whom these advise) simply by superimposing
their ownv particular technical knowledge on the pattern he has provided. Quality s
such as topographic character and slope (expressively summarized in the American
language as "fay of the land") ; stoniness ; other kinds of irregularity of surface such as
animal burrows (or, in Africa, termite mounds) ; depth available for root-development;
nature of plant cover, if uncultivated ; susceptibility to accelerated erosion ; degree of
erosion already suffered; exposure to wind or liability to river overflow-all these are

relevant, and all must be incorporated along with soil profile and texture of top soil into-
the "natural land-type units" which are to form the items in the completed inventory.
The natural land types having thus been determined on the basis of their physical
and, to some extent, their biological characteristics-the latter from the viewpoint of
describing each land type as a distinctive "habitat"-their description can be carried
further (in inhabited country) by recording whatever can be ascertained of man's experi-
ence in using them. There is clearly a wide field of relevant enquiry here : crop-plant
behaviour, crop yields and qualities ; thriftiness of stock, stock-carrying capacity, rates
of natural increase ; population densities, human health, and attainable standards of life
earned on the land . .
It will be seen that the "natural land type" signifies a great deal more than the
"soil type" does. Not only must the soil surveyor working to those objectives be a good
judge of a pit section or auger boring, and a painstaking plotter of boundaries, he also.
needs to have a good eye for country, and a quick appreciation of a situation in physical
geology. He must be a receptive observer of many kinds of natural phenomena, especially
those of vegetation and plant succession. His acquaintance with crops and cultural
practices, and his judgment in sifting the results of enquiry, must be enough to enable
him to estimate what have been, and what will be, the probable interactions between
man and the soil in both directions; for in long-occupied territory the "natural land
type" cannot easily be disentangled from the existing pattern of "social land units" and
customary usage ; the top soil has become the "M-horizon," and to understand it requires
more than mere expertness in soil morphology.
The results of soil survey, then, whilst being based on sound pedology, must be so
reported as to convey a maximum of information about potentialities in land use.
The importance attached by the Soil Survey to the collection and
presentation of miscellaneous information relating to land use is indicated
by the following quotation from Kellogg [44], describing the nature of the
soil-survey report.
The soil-survey report which accompanies the soil map describes the area surveyed,
the characteristics and capabilities for use of the soil types and phases shown on the map,
and the principal factors responsible for soil development. It is an integral part of the
complete publication and not simply an adjunct to the soil map. Except as a base map,
the soil map without the report is of little use to persons unfamiliar with the specific
nomenclature employed. First of all, the report contains a complete description of the
soil types shown on the map and their relationship to one another. As the objective of
the work is to provide information directly relating to the agricultural use of the land,
those characteristics of importance in the use of the soil for growth of plants are stressed.
Much of this information is obtained by observation during the course of the survey,
but by no means exclusively so. Climatic, chemical, economic, agronomic, and all other
data bearing on the nature and capabilities of the soil from other sources are used to the
extent that such information is available and can be interpreted in relationship to the
soil types occurring in the area. To the extent that the location of the soil areas in
respect to markets and similar factors have an important bearing on the use of the soils,
these relationships are identified and discussed.
More detailed instructions given by Kellogg, and reproduced in the
Appendix (p. 86), for the preparation and arrangement of soil-survey reports
illustrate the very wide scope of the survey which is, in fact, a land survey
based on a straightforward soil classification-a social-economic inventory
superimposed on a physical classification. It is rather more than a land
classification, but it can provide much of the information required in land-
use planning. In recently issued soil-survey reports a section has been
devoted specifically to land classification, the classification adopted being

usually the arranging of the soil types in the order of their productivity
grades (p. 69).

Whether the soil type-in the American sense-can serve as a unit for
land classification obviously depends on the objective of the latter. In so
far as that objective is to adapt crop production to the quality of the soil,
soil types and physical land types will tend to be coextensive. Moon [60]
maintained that for the Tennessee-Valley soil-survey project, the objective
of which was to provide data of value to problems of land use and manage-
ment, particularly for the production of useful plants," the soil type provided
a natural unit for land classification. The classification required was one
of soil productivity, and the class-distinguishing criteria were therefore all
the soil characteristics, both internal and external, significant for produc-
tivity. Among the internal characteristics Moon includes physical,
chemical and biological properties, and among the external characteristics
moisture conditions, topography, conditions of erosion, stoniness, and depth
to solid rock.
But whereas in the determination of soil types the relative significance
of the different characteristics is constant, in the determination of land
types the significance of each characteristic depends on the purpose of the
classification. It is not only necessary that all these soil characteristics-
internal and external-become a part of the basis of classification, but that
each of them be assigned a place appropriate to its relative significance to
the natural physical adaptation and use capabilities of the soil-if the soil
type is to fully express itself as a factor in land classification for plant-
production uses." A factor such as depth of soil or stoniness may become
" critical" at a certain intensity, and must then be given special emphasis if
the soil type is to express its full significance to land classification. The soil
type's fitness, states Moon, to serve as a unit for land classification is deter-
mined by the extent to which the variability in relative significance of the
different soil characteristics is adapted to the objective of the classification.
By taking external as well as internal characteristics into consideration
-i.e., by classifying land rather than soil, but on the basis of the soil type-
the number of classes is very greatly increased, as also is the cost of the
survey. The utility of the survey is increased proportionately. Moon
lists ten maps which can be plotted from the soil-survey map as prepared
in the Tennessee Valley :-(i) a general productivity map, (2) a potential
land-use map, (3) a lay-of-the-land map, (4) a general erosion map, either
of conditions resulting from past erosion or of natural susceptibility to
future erosion, (5) a map of stoniness, (6) a natural zoning map, (7) a relative
evaluation map, (8) a detailed natural-drainage map, (9) a crop-adaptation
map, (io) a land-management map.
Veatch [95] says that in most of the American soil-survey reports the
soil types described are in reality land types, but appear to be neither
because of the attempts in the text of the report to describe them as soil

instead of land types. But while a soil type may be identifiable with a land
type it does not follow that the same soil type in two different localities will
represent the same land type. This lack of identity is due to the intrusion
of external features and the varying emphasis placed on them in the deter-
mination of land types.

Stremme [85] goes further than American workers in identifying land
and soil types, and maintains that the genetic soil type is a scientific unit
corresponding to natural land-use types. In Germany, where it may be
assumed that land use has in the course of time largely adjusted itself to
land type, the dry-steppe soils are used predominantly for wheat, roots,
spring barley and lucerne, oats are grown largely on Bunter Sandstones,
Marls and Clays, potatoes on heath soils, meadows on mountain soils, rye
on Eschbbden (artificial soils made from podzolized forest or heath soils by
treating them with turf or stable manure, and forming a distinct type in
north-west Germany), winter barley and clover on grey forest soils. Although
oats, rye and potatoes occur on all soil types, other crops have become
almost selective enough to be used as indicator plants of groups of
related soil types.
Studies of the yields obtained on the genetic types distinguished by
Stremme and his collaborators on purely morphological grounds showed
that four classes of soils could be distinguished according to average gross
yields of the commoner crpps :-
I. Most steppe-like soils are exceptionally good for grains, grasses and
cultivated crops.
2. The moist and better brown forest soils and the better moor-like
(anmoorig) soils are good for grains and cultivated crops and very good
fpr grasses.
3. The poorer, moderately podzolized, brown forest soils are medium-
quality grain (except wheat) soils, indifferent to good potato and fodder-
beet soils, sometimes good grassland soils.
4. The rust-coloured (podzolized) forest soils and the poorer moor-like and
moor soils are very bad for grains, and bad for potatoes, roots and
Sellke (cited by Stremme) independently distinguished four soil-type
groups, slightly different from the above, according to yield data:
(i) steppe-like soils; (2) brown forest soils, rendzinas; (3) cultivated
podzolized and heath soils; (4) mountain soils.
Several attempts have been made in Germany to evaluate the genetic
soil types according to their productivity. The genetic type being a much
wider conception than the American soil type, a correspondingly wide range
of productivity rating is to be expected. Stremme gives tables by Sellke,
Ostendorff, Taschenmacher and Przeperski showing that there are significant
differences between the ranges of variation of productivity, expressed on a
percentage scale, of different genetic types. Thus, according to Ostendorff,

the rating of steppe-like soils in Danzig and East Pomerania varied between
100 and 70, of brown forest soils between 75 and 40, of weakly, moderately
and strongly podzolized soils between 55 and 20, 44 and 20, and 30 and 0,
respectively, and of mountain soils between 30 and o.
These figures, giving the highest and lowest productivity of each type,
were derived from a study of over I,Ioo soils. Although the range of
variation is too wide for them to be used for deducing the productivity of
an unknown soil from its morphology, they indicate nevertheless that there
is an inherent relationship between the genetic type and the agricultural
value of a soil. Stremme could discover no such relationship between soil
class and productivity when soils were classified into geological, petrological
or textural groups.
Sellke states that the importance of the genetic soil type as compared
with the soil variety (Bodenart-textural distinction) in the evaluation
of productivity increases with the age of the soil. In very young soils
texture may be, but in mature soils the entire morphology of the profile
becomes, the determining factor in productivity. In mountain districts
.greater variation is caused in forest growth by differences in slope and
exposure than by differences in the geological substrate.
The Russian view on the relation between land types and genetic soil
types is given in general terms by Gerasimov et al. [22] who appear to
identify genetic soil types with geographical landscape types-i.e., the soil
type is a concrete expression of the totality of the landscape. This
view has more theoretical significance in the definition of natural landscape
types than practical significance in land-use planning. Magnitsky [51] has
briefly, and not very clearly, indicated how the first broad grouping of soils
on a genetic basis can be subdivided for land-use planning purposes. The
soil macro-complexes are classified not only on their natural characteristics,
but also taking into consideration past and present land use (agricultural
history), and particularly its effect on soil productivity. The soil types
(? land types) are distinguished according to productivity (yield data) and
the measures required to raise them to the level of fertility demanded in the
land-use plan the nature of which is determined, presumably, by both the
qualities of the land and political considerations.
The soil macro-complexes are subdivided into meso-complexes the
precise nature of which is not explained. The subdivision is made according
to the intensity of agriculture practised (whether present or planned is not
clear), the nature of the agro-technique required to ensure high yields, and
the agricultural system to be adopted.
Magnitsky's paper is the only reference we have found to the procedure
of land classification in the U.S.S.R. We have found no description of the
application of this or other procedure in practice. It would appear from
Magnitsky's paper that the soil macro-complex (association of genetic
types) determines the broad features of the plan-whether the region is
to be used mainly for grain, fruit, livestock farming, etc.-the plan is then

worked out in greater detail, and the soil macro-complexes then subdivided
into meso-complexes to fit the requirements of the plan.

Kellogg [43, 44] defines a land classification according to use capabilities
as one in which bodies of land are classified (on the basis of physical, or
both physical and economic considerations) according to their capabilities
for man's use, with sufficient detail of categorical definition and cartographic
expression to indicate those differences significant to man." In such a
classification the natural aspects are subordinated to the practical.
The emphasis is on the use rather than on the land. Kellogg puts all
land in use into eight use groups (i) cropping, (2) grazing, (3) forestry,
(4) recreation, (5) mining, (6) urban development, (7) wild-life preservation,
and (8) protection (flood control, anti-erosion, etc.). (7) and (8) are
usually combined with some of the other six, and combinations such as
forestry-grazing, grazing-cropping, forestry-protection are common.
The use class of any land unit is determined as much by economic and
social factors as by inherent environmental factors. These factors, by
their nature transitory and variable, must be considered separately from
the permanent, inherent factors. "Any change in utilization alters the
position of the land in the classification of the social land units ; whereas
the more fundamental classification of the natural land types is essentially
permanent. For example, should the boundaries of some proposed grazing
or forestry district include present cropping land, a reclassification would
need to be made giving this land its rating in the new use group in place
of its rating in the cropping use group. Where the fundamental physical
data are kept clearly separated from the economic or social, as the logic of
the method demands, such changes are easily made without additional
field work [46].
Tyler [91] distinguishes between the physical classification of land
types and the economic classification of land areas. A complication is
introduced by the fact that the boundaries of the natural land types and
of the economic land areas do not necessarily coincide.
Kellogg [43, 44] offers the following outline of a scheme of land classi-

i. Objective : Classification and extension of data regarding land;
planning of land utilization.
(Detailed expression also furnishes basic physical classification for
objective under 2.)
Climate Natural land type.
Solief To be classified into categories according to
Stoniness relative physical capability in possible types of
Stoniness utilization.

2. Objective : Rural zoning ; tax assessment.
Natural land types (detailed
expression) Social land unit.
Relationship to social groups
(a) Economic To be classified:
(b) Social (i) Into use groups:
(a) Cropping
(b) Grazing
(c) Forestry
(d) Recreational
(e) Mining
(f) Wildlife preservation
(g) Protection
(h) Urban
or definite combinations :
(i) Crop-grazing
(j) Crop-urban
(2) As to capabilities within
the use group.
If the classification is to be used for land-planning purposes there
must obviously be some correspondence between the natural and
"use" classifications, that is, the criteria used to distinguish natural land
classes must be related to those used to distinguish use classes-e.g., soil-
fertility characteristics.
Kellogg outlines the procedure (particularly for soil surveyors) to be
adopted in actually classifying and mapping land-use capabilities :
(1) The first essential includes the accurate mapping, in detail, of the physical
features of the land. The basic data are those regarding the soils that are mapped in
accordance with the modern system of soil classification. Cognizance is also taken of
relief, stoniness, native vegetation, and any other physical features of local importance
in land use. These may be regarded as the external characteristics of the soil type.
Combinations of these features give the natural land types which cannot be assigned
definite ratings of productivity in the various use groups. In order that these natural
land types may be interpreted in accurate terms, it is essential that the maps show
sufficient detail, both as to definition of the categories and as to cartographic expression.
Individual land types of a size sufficient to influence significantly the capabilities of separate
units of operation need to be separated.
(2) Coincident with, or subsequent to, the mappings of the natural land types, the
inherent productivity, including "responsiveness" to cultural practices, of these types
must be determined in each of the possible use groups. The ratings should not reflect
transitory economic considerations but should be based on the essentially permanent
physical factors of productivity. These ratings must represent the generalization of all
information available, including that obtainable inductively considering the nature of
the land, and that acquired deductively from analyses of practical experience and
experimentation. Conveniently these ratings of each land type may be expressed in
terms of percentage of a standard, taken as the best land, from a physical point of
view, in each of the use groups.
(3) The next step in the procedure is the determination of the use group for each
tract of land. Especially in areas that have been settled for many years this determi-

nation will be made largely on the basis of the physical qualities of the natural land types.
Other factors, such as accessibility to markets and the size of contiguous areas of similar
land types will also be important, especially in a new or rapidly changing section. For
example, an area having a natural land type ever so well-suited to cropping cannot be
placed in the cropping use group, should the area be too small for a unit of operation or
isolated from other land suitable for cropping. This and similar considerations make it
necessary sometimes to place land naturally suited to crops in the forestry or grazing use
groups, and subsequently the land must be rated in that use group.
(4) Finally, each social unit of land is rated in the use group (or use groups) in which
it belongs. The ratings of the natural land types within the use groups are basic, but
these need adjustment according to economic or geographic considerations which may
influence production on these social units. Such considerations include accessibility to
markets, nature of the existing vegetation on forest and grasslands, and similar factors.
By an analysis of production on the standard or ideal land, and the marketing costs at
various distances on the different classes of roads, a schedule may be prepared showing
the percentage reduction in the basic rating of the land for various locations. Similar
schedules may need to be calculated in respect to the other factors.
As a final result, each piece of land can be given a rating in terms of the percentage
of the standard for the area or region. If the classification is then used for purposes of
tax assessment, the proper local officials need to place an appraisal value on the standard
land, and every piece of land in the area takes its appraisal value strictly in accordance
with its productive capacity. Ratings given by such a land classification need to have
added to them an appraisal of buildings and improvements for loan purposes or for tax
assessment where such improvements are taxed. It must be emphasized again that the
physical data should be kept separated from the economic, including the basic ratings
of the natural land types ; as a result of any changes in economic conditions, the necessary
adjustments of the ultimate land classification can be made easily without additional
field work.
The productivity ratings referred to above are described on p. 69 and
Bausman [7] lists five different measures of land classes commonly
used in land-classification studies. None of them is a complete measure,
but each serves as a check on the others. Soil types are the most widely
used and generally the most effective measures of land classes. Their chief
weakness is that they do not reflect the effect of location on land use ; not
only the effect of distance from market or town, but also the effect of
location relative to other land types. A parcel of poor land would have a
higher use capability on an otherwise good farm than it would have on a
poor farm.
Crop yields afford a good measure of land classes if the land is used for
cropping. They are of little use on pasture or forest land. Bausman
points out that on many poor farms the proportion of cropped land is low,
but it receives all the manure produced and consequently gives higher
yields than would be expected from its inherent properties.
Size and condition of buildings are a fairly good measure of land classes.
Good land usually requires larger buildings for storage and shelter than does
poor land. As often as not the condition of the buildings is more a reflexion
of the personal habits of the occupier than of the quality of the land. On
farms near towns or owned by urban dwellers or supported by urban capital,
the size and condition of the buildings bear little relation to the class of

Net farm income may be used as a measure of land classes, but the
data are usually costly to obtain and become unreliable when any appreci-
able proportion of the income is derived from enterprises not closely
associated with land quality-e.g., poultry, flowers, market gardening.
Bausman favours intensity of land use as the most reliable single
measure of land classes, on the grounds that the occupiers will, over a period
of years, discover by trial and error the optimal land use under prevailing
conditions. He recognizes, however, that there is usually a lag in land-use
adjustments that needs to be taken into account. The relative validities
of the different measures vary widely from situation to situation, and there
are no general rules for assessing them.

As Marbut (p. 13) pointed out, the climatic factor in land-use classifi-
cation is related to crop adaptability. So far as adaptability is concerned,
in most areas coming under a single plan the climatic factor can be regarded
as constant, except in mountainous or topographically dissected regions.
Local climatic variations assume increasing significance as the intensity
of land utilization increases, but since intensive farming cannot be carried
on for long unless land use is adapted to the environment most intensively
farmed countries have not felt the need for land-use planning. Use is
already reasonably well adapted to the land. Climate has, indeed, found
more use as a basis for demarcating broad agricultural regions than in
detailed land classification. It is the dominant factor determining the
suitability or otherwise of a region for a particular crop.
Stamp [82] states that in Europe temperature-particularly in the
growing season-is the most significant climatic element governing the
adaptability of a region to the culture of one or other crop. Passing from
south to north, summer temperatures decline, and successive limits are
reached for the ripening of rice, maize, wheat, oats and barley. Thus a
physical or geographical limit is set to types of agriculture based on these
different food grains.
Winter cold is another factor influencing types of crop culture. The
intense winter cold of south-east Europe renders autumn sowing of grain
impossible, and spring wheat takes the place of winter wheat. The spring-
wheat region of Europe is practically coextensive with the black-earth or
chernozem zone, characterized by a semi-arid, highly continental, steppe
climate. Similarly, the northern limits for the olive, orange and lemon
are determined by the incidence of winter frosts, and the northern boundary
of the zones of culture of these fruits coincides with that of the Medi-
terranean climatic region.
Rainfall in Europe is usually adequate, and sometimes excessive, for
agriculture. Where it is insufficient-in the south and south-east-
irrigation is practised. Where it is excessive pastoral farming or no agri-
culture (forest), and where it is adequate arable farming, predominates.

In the delimitation of agricultural regions the soil factor tends to
follow the climatic factor, since the genetic soil type is primarily a climatic
phenomenon. It does not, however, follow that the boundary between
two recognized soil or climate zones will also be a boundary between two
agricultural or land-use regions. It is soil poverty, however, which brings
the rye-growing region into parts of Europe that climatically would have
been regarded as a wheat-growing region. The limit of wheat- or maize-
ripening is not necessarily related to the distribution of any natural plant
association and its corresponding soil type. Certain close correspondences
do occur-e.g., between the spring-wheat and chernozem zones, and between
the olive and Mediterranean-climate (terra-rossa) regions-but the
boundaries of both agricultural regions and of soil or climatic zones are
naturally very ill-defined. The transition belts from one agricultural or
soil zone to another are often very wide. It is convenient, therefore, to
make the agricultural and climatic (or soil-type) boundaries coincide as far
as possible on a map so as to emphasize the indubitable significance of
climate in determining crop adaptation. Thus, several of the boundaries
in Jonasson's map of the agricultural regions of Europe [42] correspond
closely to the boundaries of the soil zones in Sibirtsev's soil map which was
drawn partly from climatic data as well as from direct observations of the
soil, but was later considerably modified by Stremme [87].
While climate exercises the most obvious control over the distribution
of broad agricultural types it is by no means always the determining factor.
Economic, political and social controls may be equally or more potent,
particularly in populous regions.
Ahnost as obvious as the broad resemblance between maps of climate and agriculture,
are the marked discrepancies. Dry interior Asia-Africa has not the same agricultural
system as dry interior Americas and Australia. The types covering wide acreage in
the humid middle latitudes of the northern hemisphere are not extensively spread in the
southern. Most conspicuous of all, the agriculture of east and south Asia, notably China
and India, does not parallel counterpart climates in the other continents. Differences
spreading over such broad areas cannot be attributed to variations in soil or slope, which
are miniscule in comparison. They do correspond to what may be loosely defined as
Occidental versus Oriental society and progressive versus backward culture. Some
of them correspond also to marked contrasts in density of population [98].
The actual boundaries of agricultural regions, so far as they can be
defined at all, are determined not solely by physical conditions (temperature,
moisture, topography, soil), but are rather the result of the pressure of
population against these conditions, the pressure varying with the stage of
civilization and technical achievement attained by the people [5].
The effects of industrial development, when highly concentrated as in
western Europe, may outweigh those of climatic or physical controls. Most
of industrialized Europe-Great Britain, Northern France, Belgium,
Holland, Denmark, Germany, and parts of Czechoslovakia and Poland-
are included in what Jonasson [42] calls the Dairying, Hay and Root-crop
Region. The agriculture of this region is characterized by intensive, highly
diversified mixed farming, with livestock predominating. There is nothing

in the climate or soils particularly favourable to this kind of farming, which
has evolved in response to the demands of dense, urban populations.
Ellsworth Huntingdon (cited by [82]) and Jonasson consider that the
physical environment is peculiarly favourable for high production, though
it may be noted that Marbut [52] considered that humid forest soils
(pedalfers) like those of western Europe were in general much less adapted
by nature for sustained food production than the semi-arid grassland soils
(pedocals). Huntingdon supports his statement by average yield data for
the most common western-European crops. If the average yield for
Europe is 100, that of Belgium is 179, of Holland 170, of Denmark 168,
of Switzerland 155, of England and Wales 146, of Germany 127, and of
France 105. It may be noted, however, that this order of productivity has
less to do with the quality of the soil than with the consumption of fertilizers
per unit of arable land. The order of fertilizer consumption per arable acre
before the war was Holland, Belgium, Germany, Denmark, Great Britain,
France [17]. If the manurial value of imported animal foodstuffs were
taken into consideration, figures for fertilizer consumption in Denmark and
Britain would be considerably raised, and the order of fertilizer consumption
would approximate to the order of productivity. There can be little doubt
that both the type of utilization and the high level of productivity of
western Europe are less closely related to physical than to social and
economic controls. Intensive mixed farming has developed because it is
the only known system that could support a dense industrialized population.
One of the most valuable root crops in the European root-crop region is
sugar beet-an example of the effective influence of a political control,
since beet sugar can nowhere compete freely with imported cane sugar.
Jonasson [42] claimed to be able to distinguish seven zones of pro-
duction around large urban centres, and reflecting the influence on land
utilization of distance from market. These zones do not usually exist in
their entirety in actual fact, owing to the intrusion of other than commercial
factors in the determination of land utilization, but they are apparent in
part around cities like Moscow, Indianapolis and Buenos Aires. Stamp [82]
states that they can also be traced in Britain. The zones, extending,
ideally, concentrically from the urban centre are :
I. Horticulture
Zone I. Greenhouses and floriculture.
Zone 2. Market gardening-fruit and vegetables.
II. Intensive agriculture with intensive dairying
Zone 3. Dairy products, fat cattle and sheep, forage crops.
Zone 4. General farming-grain, hay, livestock.
III. Extensive agriculture
Zone 5. Bread cereals.
IV. Extensive pasture
Zone 6. Rearing of cattle, horses and sheep.
V. Forest culture
Zone 7. Peripheral areas, forests.

Baker [5] states that not only are the physical conditions the principal
factors in determining the utilization of the land in a region and the crops
grown ; but also they become more important as population increases, the
knowledge and practice of agriculture advance, transportation facilities are
improved, and the supply of capital and labor is increased and better
distributed; in brief, as agriculture and forestry become more highly
organized and commercialized."
This statement seems at first sight at variance with the view previously
expressed that a highly developed industrialism masks the effects of
physical conditions on land use, but Baker was referring to the United
States which still are more agricultural than industrial. They have only
quite recently, with the development of transport on a continental scale,
emerged from the stage of self-sufficient farming to that of commercial
farming, an early phase of which is the concentration of crops in the regions.
where they can be most economically grown. Formerly, for example,
fibre flax was grown for home consumption on almost every farm, but now
it is imported, and flax is grown commercially almost wholly for oil, and is
restricted to the sub-humid areas of the North-west where the physical
conditions are peculiarly favourable. Again, with the commercial develop-
ment of cotton, competition has restricted the crop to the area where it can
be most economically grown. There is a general tendency for the farming
community to concentrate in the more fertile areas and to abandon the less
fertile, owing to the greater capacity of the former class profitably to utilize
large amounts of capital and labour. Thus between 1910 and 1920 the most
fertile farm lands of Illinois increased in value by 79 per cent, while the
least fertile lands increased by only 41 per cent.
The general tendency at the present stage of American civilization is
for economic forces to operate in the same direction as physical conditions
in the determination of land use. The increasing use of farm machinery,
also, is emphasizing the significance of topography; accessibility for agri-
culture and adaptability to mechanization, and inaccessibility and non-
adaptability usually go together.
Three stages may be distinguished in the normal evolution of agri-
cultural regions ". At the beginning of settlement there is a pioneer stage
in which the land is made to produce anything which will allow the settler
to exist and hold his own against a still untamed Nature. This stage tends
to be exploitative, since the system of land use is dominated by the
individual's struggle for existence. Both physical and economic controls
operate, but not very specifically. Crops are grown because the cultivator
needs them for his day-to-day existence rather than because the land is
specially suited to them. It was during the pioneer stage that much of the
soil exhaustion and erosion of the modern world originated.
A subsequent stage may be distinguished as a social and economic
organism emerges, and a sense of permanence in the occupation of land,
extending beyond the span of a man's life, becomes apparent. Land use
becomes more closely adjusted to local variations in the environment, more

conservative and less exhaustive. It is at some time during this stage that
the relative influence of physical controls is maximal-and that the need
for land-use planning is most likely to be felt. Parts of the United States
at the present time afford good illustrations of the transition from exhaustive
to conservative land use, and of the application of land-use planning to
meet the problems raised thereby.
At a still later stage the influence of economic and technical controls
increases relatively to that of physical controls. Men have learnt not only
to adapt their crops to the environment, but also to modify the environment
to suit particular systems of agriculture-for example, by intensive manuring
as in western Europe, or by terracing and other soil-conservation measures as
in Japan and Java. Further desirable adjustments in land use that could
be assisted by planning are mainly to economic and technical developments.

Hollstein [37] has attempted a quantitative distinction of agricultural
regions according to the maximum populations the regions are capable of
supporting off their own produce when the land is utilized at an equal,
but arbitrary, level of intensity. He maintains that climate is the principal
natural factor determining the possibilities of land use. This becomes more
evident, the broader and more general the picture of land use. Nevertheless,
single factors such as distribution of rainfall throughout the year or in hot
or cold seasons, relation of temperature to rainfall, distribution and quality
of sunlight, etc. that cannot be incorporated into a climatic classification
are often so important for land-use determination that none of the accepted
systems of climatic classification is adaptable as a basis for land evaluation.
In other words, no practical relationship has been found between the
separate climatic factors (rain, temperature, sun, wind, etc.) that will
express the crop-producing potentialities of the land.
Hollstein measures productivity by the capacity of a region to produce
human foodstuffs. Since food production varies greatly in quantity
according to the type of agriculture (e.g., mixed or specialized farming)
grain-producing capacity is taken as standard. Grains are the most
important food crops, and their short growing periods and other character-
istics make them particularly suitable as measuring rods of productivity.
Grains are cultivated everywhere, and all the common grains have approxi-
mately the same calorie value per unit weight (3,300 per kg.), consequently
the grain yield is also a measure of the energy-producing capacity of the land.
If land is used entirely for the production of animal food the calorie value
of the animal products ultimately obtained for human consumption is about
one-fifth of what would be obtained by growing vegetable food for direct
human use. Under intensive horticulture the calorie value can increase
from four to eight times that under grain cultivation.
As basis for his calculations Hollstein uses the actual yields obtained
in the cultivated parts of the regions concerned, and from these data he
calculates the calorie value per unit area of land.

By dividing the calorie value by the arbitrary figure 2,500 the number
of human feeding days per unit area is obtained. Multiplying this by
the fraction of the total land area that is cultivable, a quality number "
(Wertzahl) for the region is obtained. It will be seen that the operative
values are the average grain yield and the proportion of cultivable land ;
the interpolation of calorie values and human feeding days involves merely
multiplication and division by two constants which do not affect the
relative significance of the quality numbers obtained. E.g., in the south-
China plains a yield of 35 dz./ha. of grain provides the calorie requirements
of 1,265 men per square kilometre. 80 per cent of the total area is cultivable,
thus the population capacity is reckoned as 1,012 per square kilometre.
The actual population density is 900-S5 per cent of the calculated popu-
lation capacity, and 71 per cent if the whole area and not only four-fifths
of it had been utilizable. The utilization factor (Ausnutzungsfaktor)
is 0.71. In one part of Shantung population density (821 per sq. km.) is
greater than population capacity (723) calculated on a basis of four-fifths
of the land being optimally used for food production, but taking the
province as a whole the ultilization factor works out at 0.55.
Goodson [24] points out that one result of this method of calculating
potential productivity is to discriminate in favour of regions where maize
is grown, since maize has a much heavier yield capacity than wheat or
barley. (It also requires about double the labour input of the smaller
grains, a factor which does not enter into Hollstein's calculations.) Con-
sequently, south-eastern Europe gets a higher assessment than actually
much more productive countries in western Europe. And by assuming an
intensity of use at present unattainable in the Amazon and Congo basins
these regions get the highest ratings of any in the world.
Goodson also draws attention to the difference between the estimates
of Hollstein (373,300,000) and of Griffith Taylor (20,000,000) of the numbers
of population supportable by Australia.
Population and land quality. Attempts have occasionally been made to
relate the population of rural communities, free from industrial or urban
influences, to the inherent properties of the land. Mohr [59] established
quite a close relationship between population density and soil quality in
some islands of the Netherlands Indies. The nature of the volcanic deposits
which cover much of Java and the age of the soils derived therefrom had
a predominant influence on population density, sometimes even greater
than that of industrialization or urbanization. Volcanic soils in the early
stages of formation are usually the most fertile, and Mohr showed how the
depositions from the latest eruption (1931) of Mt. Merapi mainly to the
south and south-west of the volcano were reflected in the higher populations
in these directions. In general, districts with the most juvenile volcanic
soils have the densest populations. Heavy rainfall, by leaching the soils,
has a contrary effect, hence the younger the soil and the lower the rainfall
(provided it is sufficient) the denser the population. Somewhat similar
relationships between soil and population were observed in other islands

of the archipelago, but were not so clear as in Java, owing to the less
extensive distribution of young volcanic soils. Where these did occur,
however, the populations tended to be densest.
Smits [77] considered that topography had a still more potent influence
than the origin of the soil on population density, the densest populations
being found in narrow valleys and on gentle slopes (of volcanoes) where
primitive and small irrigation works could be installed. The banks of large
rivers in the coastal plains (of the Netherlands Indies) where the land was
subject to regularly occurring floods in the rainy season were also marked
by high population density.

Several suggestions have been made to use plant indicators to assess
the suitability of land for producing particular crops, but the method does
not seem to be much used, partly because the indications given in one
region cannot always be relied on to be valid elsewhere. According to
Shantz [76] correlations between the natural vegetation and the crop-
producing capabilities of land in any area can be satisfactorily determined
only after careful study of the different vegetation types in relation to their
physical environments, and such correlations will need to be modified
before they can be applied in another region where the physical conditions
are different. With this proviso, Shantz, writing in 1911, considered that
natural vegetation afforded the best basis then available for classifying
land. At that time soils were distinguished on a textural or geological, not
on a genetic, basis.
More recently, Hollstein [37] has expressed the opinion that the
vegetation is a better indicator of crop-producing capacity than is the
genetic soil type. He points to the difference in capabilities between the
Eurasian steppes and the South American pampas, both of which, he says,
have soils of the chernozem type, yet the former have a hard winter and
short growing season that greatly limit the choice of crops, while the latter
enjoy mild winters, can produce a great variety of crops, and sometimes
two harvests in a season. It is, however, possible that a detailed comparison
of South American and Eurasian black earths would have revealed as signi-
ficant differences between them as exist between their respective climates
and vegetations. There is in general a close correspondence between
natural vegetation and genetic soil type. The advantage of using the
vegetation rather than the soil as an indicator is that small, but significant,
differences are more immediately obvious in the vegetation. Specific
plant indicators-e.g., Albizzia spp. as indicators of suitable soils for tea [41]
-in particular are easy to identify and to interpret, again with the proviso
that the indications may be unreliable outside the area where they have
been established by direct observation. Analogous soil indications-
e.g., high hydrolytic acidity for tea soils-are usually insufficient by them-
selves. The soil, on the other hand, is more durable than the vegetation,

and often retains many of its original characteristics after all trace of the
natural vegetation has been destroyed.
One of the chief characteristics required of a reliable plant indicator
is that it should be exacting in its habitat requirements [73]. Plants with
a wide range of growth conditions are obviously unspecific as regards
habitat. Clements [13] maintains that the dominant species which con-
stitute a climax are the best indicators since they bear the unmistakable
impress of the climate in the corresponding life-form, viz. tree, shrub, and
grass." Climax indicators, according to Clements, express the type and
degree of climatic control and the problems confronting man in maintaining
the climax or modifying or permanently replacing it by cultural forms of
land utilization. The whole plant community, since it integrates the
response to the habitat of several dominant species, is a better indicator
than are individual species.
Shantz [76] also deprecated the tendency to base judgments about
land quality on the presence of individual indicator species; the compo-
sition of the plant cover as a whole forms a much more reliable basis. In
considering the correlation of natural vegetation with crop production
many of the difficulties experienced in correlating vegetation with the
physical environment are eliminated. Broadly speaking, the native plants
obey the same physiological rules as do cultivated plants having the same
general requirements with respect to moisture, heat, and light. Hence,
it should be comparatively easy to infer from the differences in the native
vegetation produced by differences in the physical environment what would
be the effect of similar differences upon cultivated plants."
Working in Eastern Colorado he found good correlation between land
capability and vegetation. Land carrying a pure short-grass cover was
found to be supplied with water in the surface foot or two of soil only, and
usually only for a short period during spring and early summer. Land
with a uniform cover of tall grasses was supplied with water to much greater
depths, and offered favourable conditions for plant growth for a longer
period. A mixture of short and tall grasses indicated intermediate conditions
of water supply. More detailed study showed that the areas of greatest
agricultural value were those with a wire-grass (Aristida longiseta) vegeta-
tion, an association very rich in both shallow-rooting and deep-rooting
species. The conditions indicated are a fairly deep, pervious soil capable
of absorbing all the rainfall and heavy enough not to blow badly when
broken. Sampson [73] points out that the pioneer settlers selected short-
grass in preference to wire-grass land for farming because the soils appeared
to resemble those with which the settlers had become familiar in the east.
Of almost equal value to the wire-grass association were certain phases
of the grama-buffalo-grass vegetation in which Bouteloua oligostachya,
Buchloe dactyloides and Psoralea tenuiflora were dominant. Areas
characterized by other associations-e.g., the Gutierrezia-Artemisia associa-
tion, lichen formation, and species associated with blow-outs "-were
defined as less fit or unfit for agricultural purposes. These occupied

mainly bare, rocky or alkaline land which could usually have been adjudged
non-agricultural without reference to plant indicators.
As might be expected, within an area of uniform climate there was good
correspondence between vegetation and soil texture (referred to by Shantz
as soil type), but the correspondence broke down in areas of varying climate.
The modern conception of soil type embraces both climate and texture, and
a wider and more general correspondence between vegetation and the soil
types now distinguished would be anticipated.
Shantz concluded that differences in the vegetation could be used to
indicate changes in one factor of the environment only when all the other
factors remained unchanged, and that indications of soil moisture were more
easily obtained than of any other factor. In this connexion it should be
borne in mind that Shantz was working in a region where lack of moisture
was the prevalent limiting factor.
Aldous and Shantz [2] developed a complete system of use-capability
land classification based on the natural vegetation for the semi-arid region
of the United States west of the iooth meridian. When the classification
was started settlement was so recent over much of the region that no
agricultural history was available, and climatic data were also lacking over
many areas. 102 vegetation types (plant communities) were distinguished,
and classified into two major categories (i) dry-farm land, capable of
producing cultivated crops without irrigation, and (2) grazing land, on
which pasture or the natural vegetation was more valuable than possible
cultivated crops. (I) was further subdivided into good, bad and indifferent
(a) grain land and (b) forage land, and (2) into good, bad and indifferent
pasture. We have not come across any record as to whether or not this
twenty-year-old classification has proved its worth in practice.
Hilgard [34] lists the following plants which when growing in dense
stand in California, indicate a soil alkalinity too great for cultivated plants :-
tussock grass (Sporobolus airoides), bush samphire (Allenrolfia occidentalis),
dwarf samphire (Salicornia subterminalis), saltwort (Suaeda torreyana),
greasewood (Sarcobatus vermiculatus), alkali-heath (Frankenia grandifolia
campestris), cressa (Cressa truxillensis), salt grass (Distichlis spicata).
St. Clair-Thompson [72] suggested that plant indicators could be used
to determine the suitability of forest land for cocoa in the Gold Coast. He
distinguished four types of climatic forest climax in the Gold Coast. These
were :-
(1). Rain or evergreen forest-Cynometra-Tarrietia association. Cocoa
does not thrive, possible owing to lateritization of the soil.
(2). Wet mixed deciduous forest-Lovoa-Guarea-Pentadesma associa-
tion. Cocoa thrives, except in edaphic climaxes dominated by Lophira.
(3). Moist mixed deciduous forest-Khaya-Entandrophragma associa-
tion. Favourable for cocoa cultivation.
(4). Dry mixed deciduous forest-Triplochiton-Cistanthera association.
Cocoa cultivation highly speculative.

Lengthy lists of indicator species for each climax are given, for details
of which reference must be made to the original paper.
The assumption often made by pioneer settlers that the more luxuriant
the tree growth the greater will be the productivity of the soil under
cultivated crops needs to be qualified under certain conditions-e.g., a
tropical rain forest may indicate a lateritized soil incapable of producing
more than one or two crop harvests. Again, the policy of settling men
where the big trees grow proved disastrous in Victoria where the big trees
were required to protect the watershed of the Murray River [21].
After reviewing the evidence available Sampson [73] concludes that
" the character of the natural plant cover may serve to segregate virgin
lands into categories favourable or unfavourable for crop production.
The reliability of such guidance, however, is roughly proportional to the
knowledge of the ecological relations between the native vegetation, the
cultivated crops, and the environmental complex."
The intuition of an experienced observer who has taught himself to
interpret the indications of vegetation and soils can often provide more
reliable estimations of the performance of land under cultivation than any
rules of interpretation that can be expressed in words.

Cajander [II] states that there are two kinds of forest classification,
based either on some characteristic of the stand or else on the natural
qualities of the site. Stand classifications are based on the yields of the
stands, which may vary greatly on similar sites according to the tree species,
management of the forest, etc. These classifications are quite subjective.
Site classification, on the other hand, is objective, in that it represents an
attempt to combine into one class all areas with the same capacity for
growing timber, irrespective of what they are actually producing.
The principle of Cajander's system is that the inherent quality of a
forest site is expressed in the nature and composition of the ground flora,
and can be defined by the presence or predominance of certain indicator
plants which are, in fact, indices to soil conditions. The main soil groups
will be indicated by the tree species, but the important variations within
any group will be indicated by the ground flora, and in such a way that
if, say, any three plant associations under a given tree species represent
descending grades of site quality, they will do so in the same order wherever
they are found under any kind of forest.
A complete classification of Finnish forest types has been developed
from these principles ; it is considered sufficiently reliable to be used widely
in State assessments of forest-land values.
The classification can only be applied to mature and normally developed
forests in which the natural vegetation has not been disturbed by human
The following are the most important types distinguished in the
Finnish softwood forests. They are given in ascending order of site quality.

Cladina type. Occurs on very dry pine heaths, with a thin layer of
raw humus. The ground is usually greyish white, owing to the abundance
of lichens (especially Cladonia alpestris).
Calluna type. Calluna vulgaris is usually the dominant species in the
ground flora, accompanied by a rich moss and lichen vegetation. Pine is
the most common forest species, but spruce and birch frequently occur as
admixtures, and occasionally as dominants.
Vaccinium type. Moss vegetation is continuous, and lichens are
common ; grasses are fairly abundant. The dwarf-shrub vegetation is
dominated by Vaccinium vitis idaea, accompanied by Myrtillus nigra and
Calluna vulgaris. The forest is usually pine, but spruce and birch may be
dominant (more frequently than in the Calluna type), and occasionally
alder (Alnus incana).
Myrtillus type. Herb vegetation is abundant, with Myrtillus nigra
dominant, and nearly always accompanied by V. vitis idaea. There is a
luxuriant moss cover (Hylocomium and Dicranum spp.), but lichens and
grasses are unimportant. Spruce is the natural forest species, but pine,
birch, alder and poplar also occur.
Oxalis-Myrtillus type. Mosses are scantier, but herbs and grasses are
more abundant than in the three preceding types. The herbs and grasses
contain many hygrophilous species, in particular, wood sorrel (Oxalis
acetosella) which gives its name to the type. Exacting bushes, such as
Rubus idaeus, Daphne, etc. are often present. Besides the common species
of softwoods, a sprinkling of hardwoods may be met with, including ash,
maple and oak (Q. pedunculata).
Oxalis-Majanthemum type. Thin-leaved herbs and ferns (e.g., Oxalis
acetosella, Majanthemum bifolium) and flowering plants such as violets are
abundant. Mosses, grasses and dwarf shrubs are common, but not
abundant. Spruce would be the only dominant tree species in virgin
forests, but in Finland, owing to burning, cutting, etc. large areas are
covered by birch, alder, poplar and other hardwoods.
The recognition of the different types is an art which can only be
acquired by practice. The appearance and composition of the soil cover
change with the time of year, but with the more well-defined types an
experienced observer has no more difficulty in placing them than a botanist
has in determining a plant species at different stages of growth.
That there is a close connexion between this purely ecological classifi-
cation of forest types and both the actual yield capacity and soil properties
of a locality has been demonstrated by several Finnish workers. Cajander's
theory of forest type assumes that in all areas of the same type the combined
action of climate and soil is equivalent, and consequently within one
climatic region differences in type must be due to differences in soil
properties. Table i, compiled from figures representing averages of several
hundred samples, shows how certain soil properties are connected with the
forest type. The figures, except for soil reaction, are relative, taking those
for the Myrtillus type as ioo, and refer to the top 20 cm. of soil.

Yield figures and selected soil data relating to Finnish forest types [II]
Relative Amount
Type yield Loss on of
Type (75-year ignition electro- CaO P205 K20 N pH
old pine) lytes
Cladina ... ... 27 49 44 36 161 118 34 3.6
Calluna ... ... 52 88 84 54 118 96 64 4.2
Vaccinium ... ... 83 81 55 79 161 100 71 4.6
Myrtillus ... ... 100 100 100 100 100 100 100 4.8
Oxalis-Myrtillus ... 115 117 159 117 54 109 137 5.2
Oxalis-Majanthemum 143 157 140 28 114 223 5.0
Cajander's ideas are most readily applicable in countries like Finland,
where climatic conditions are comparatively uniform. His system has not
been found so suitable for forest assessment even in the neighboring
country of Sweden. There a closer connexion has been found between
forest type (vegetation) and stage of soil formation (podzolization) than
between forest type and forest yield. The different vegetation types
influence the leaching process caused by water percolation in the soil to
markedly different degrees, due to the quality of the humus layer they
produce. The strongest leaching is exercised by the Myrtillus type, then
by the Vaccinium type, and the weakest by the Cladina type [88]. Usually,
therefore, but not always, similar vegetation types are associated with the
same soil type.
Hilgard [34] found, as might be expected, a general relationship
between soil quality and the nature of the indigenous tree flora. Certain
tree species, e.g., Quercus minor and Q. marylandica, appeared to grow
equally readily on rich and poor soils, but in such cases the trees tended to
assume characteristic and different shapes according to the quality of the
soil. The shape was believed to reflect the lime content of the soil.
Coile [14] considers that the fundamental hypotheses behind the use
of ground vegetation for evaluating forest sites-namely, that the ground
vegetation reflects the inherent quality of the site better than does the
tree vegetation, and that forest types so distinguished are largely
independent of the composition, age and density of the forest stand-
are open to question except under special conditions of climate and topo-
graphy. Because the ground vegetation does not have so extensive and
ramifying root systems as trees, it is unlikely fully to reflect the conditions
under which the trees are growing. Moreover, the assumption that the
nature of the ground vegetation is independent of that of the tree species
has obviously only a very limited validity. But enough evidence has been
collected to show that certain ground-vegetation types are associated with
the more fertile, and others with the less fertile, forest-soil conditions.
Such differences in fertility are, however, as easily associated with readily
distinguishable topographical or soil characteristics as with the ground

The use of indicator vegetation types as an aid in the selection of tree
species to plant on open land falls into a different category, and is justified
where experience has shown that certain types are specific to certain sites
on which a given tree species is known to thrive. It may happen, however,
that the climax species indicated by the ground vegetation will not thrive
when planted directly onto open ground, owing to differences in micro-
climate between open and forested sites. The ground vegetation may
indicate the climax, but not the intermediate ecological stages which may
need to be traversed before the climax is reached.
Hesselman [32] has shown that certain herbaceous plants are indicative
of specific soil conditions-e.g., Epilobium angustifolium and Rubus idaeus
indicate high nitrifying capacity in Sweden. He suggested that the ground
vegetation indicated the kind of treatment required rather than the inherent
quality of the forest site.
Under the conditions prevailing in the United States Coile concludes
that if a classification of forest sites is desired, it should be based upon
fundamental and permanent features of site, namely soil and relative
topographic position of the soil mass. Characteristics of the soil mass,
the substratum, and topography, which are related to the availability and
total volume of water present for use by forests, should be the primary
criteria in any classification of site. Markedly different chemical character-
istics of soil may be secondary criteria of classification. In regions of
appreciable relief, and in northern regions with less relief, aspect of land
should be brought into the classification. The following characteristics of
site should be considered.
i. Aspect.
2. Relative topographic position and slope.
3. Texture and thickness of the surface soil or A horizon.
4. Texture and thickness of the B horizon.
5. Nature of the substratum or soil parent material and its depth
if relatively shallow."
The classification envisaged is apparently one in which the site factors
enumerated would be evaluated and awarded "points" after their signifi-
cance had been determined from existing forest stands.

On most pasture and range land indicators express the present condition
of the land resulting from past treatment rather than its inherent or potential
productivity, though indicators do express the latter in some degree on
virgin grassland. Range indicators says Taylor [89] are the clues
to range happenings. All that can be used as guides to recognition of the
real state of affairs-non-use, satisfactory use, abuse-either in the past
or present, may be classed as indicators." Indicators, not solely plant
indicators, are often specific to certain localities, and generalizations are
impossible, but certain general observations can be made about the indi-
cation of range deterioration [73].

Range deterioration well under way is shown by : weakened vitality
of the principal forage plants ; limited, or the absence of, reproduction of
the most palatable species ; close grazing of species of low palatability ;
a thinning ground cover of the entire vegetation; replacement of the good
forage plants by those regarded as of little value; evidence of relict forage
plants ; incipient gullying and evidence of increasing soil erosion. Evidences
of past range damage are : a relative absence of formerly abundant forage
plants; foliage and branches of the taller browse plants trimmed back as
high as the animals can reach; dead remnants of the browse species of low
stature ; abnormal abundance of those species which persist and reproduce
after more palatable species have disappeared; accelerated soil erosion
accompanied by numerous V-shaped gullies. Indicators resulting principally
from unsatisfactory soil conditions that may be used in conjunction with
other indication are : truncated soil horizons; lack of a normal amount
of organic soil between groups of herbs or shrubs; the conspicuous
presence of hummocks, indicating general erosion in the absence of
gullies. Doubtful or less reliable indicators of a deteriorating range are :
local denudation of the soil, sometimes caused by soil slipping or dis-
placement, or by congregation of herbivores on a restricted area ; increase
in poisonous plants resulting from a favourable successional reaction ;
general appearance and condition of the grazing animals, as where over-
stocking of an area for a single season is the principal contributing factor ;
condition of the timber reproduction, such as damage from defoliation or
destruction of the leader by insects.
These observations refer mainly to conditions in the western United
A good deal of work has been done in Germany, Switzerland and
Scandinavia [54] on the estimation of site quality of meadows and pastures
by the composition of the herbage association. The general inference
from much of this work is that the plant association reflects most clearly
the moisture condition of the habitat, and rather less clearly the reaction
and nutrient content of the soil. The investigations have considerable
ecological significance, but it is unlikely that they will have much application
to practical problems of land classification. The indicator plants are used
to classify the meadows or pastures as productive assets rather than the
land. Thus Petersen [69] distinguished six meadow types by the dominance
of certain grasses in the meadow. The best class contains meadow foxtail,
meadow fescue, timothy, canary grass, manna grass, meadow grass. A
middle class contains quack grass, soft brome grass, meadow grass. The
lowest class contains sedge grass, Scirpus, Deschampsia, sheep's fescue.
A definite hay yield, both quantitative and qualitative, is associated with
each class ; one centner of good hay is reckoned as equal to two centners.
of poor hay. The highest class yields 100 dz./ha.
Pentz [67] states that the success of any particular type or system of
farming in South Africa is closely related to the type of vegetation on the
area. Experience has shown that successful farmers are those who have

adapted their farming to the natural vegetation. Much of the current
maladjustment of South African agriculture to the environment has been
the result of booms and subsidies (e.g., on maize) which have encouraged
the wrong kinds of farming. In the majority of cases where there has
been a boom, the areas which were suited to the production of the com-
modities required have suffered very little, except from the high valuation
of land. It is those areas where conditions were not suitable for economic
production that most damage was done to the land and vegetation, and
where most failures occurred."
Pentz maintains that the only sound foundation for South African
agriculture is to apply farming systems on the vegetation types to which
they are suited, and this can only be done by a detailed botanical survey
of the country. In such an experimental survey of the Estcourt (Natal)
area the following points were observed :-(I) number of vegetation types ;
(2) area of, (3) climatic conditions and (4) soils and topography associated
with, each type. An investigation was conducted into present and possible
farming systems, and their effects on the land and costs (a) in relation to
each vegetation type, and (b) ih relation to a combination of types in order
to determine whether together they could form a complete farming unit.
Four main types were distinguished-thorn veld, tall-grass veld,
highland sourveld and mountain veld. A careful study of all conditions
prevailing on the thorn veld ruled out crop, sheep or dairy farming, and
pointed to cattle farming as the most suitable system. Ordinary ranching,
however, would require too much land, and stock breeding is recommended
as the most suitable on both ecological and economic grounds. The tall-
grass veld is likewise cattle land, although with its open plains and well-
distributed rainfall it appears ideally suited to crop farming. It has,
however, shallow and erodible soil, and can only be safely cropped on
intensive and highly scientific lines. The vegetation, topography and
climate indicate a seasonal type of husbandry, i.e., grazing the veld in the
summer and carrying the animals through the winter with hay and silage
produced from surplus summer growth. The highland sourveld has a
very severe climate, and stock farming depends on the quantity of food
that can be conserved from the veld during the summer to carry the stock
through the winter. Sheep farming is a possibility, but it is suggested that
the land is better suited to the intensive production of beef types. Both
here and in the thorn veld the farmer would dispose of every animal, except
breeding stock, as soon as possible. The mountain veld is unsuitable for
farming, but contains the head waters of the rivers, and must be
protected from denudation if farming in the areas below is not to suffer.
Pentz states that by thus adapting farming to the vegetation the whole
area could be organized as a complete unit for beef production, so that each
farmer could farm without exploiting his land and have a ready market
for his produce. The highland sourveld would produce breeding stock
for the breeder in the thorn country who would produce weaners for sale

to the grazier in the tall-grass veld who would supply a well grown-out
animal ready for feeding for export.
In the above example, only potential grazing land is considered, but
there is no reason why similar ecological indications should not be used to
determine the use capabilities of potential crop land.

There follow descriptions of several systems of land classification,
selected to indicate the points common to all systems and points peculiar
to each system. The most obvious common point is the acceptance of some
measure of soil productivity as the basis of the classification. In the
systems described the differences in soil productivity between classes are
mainly qualitative-i.e., class I is described as either better or worse than
class II, or they are described merely as classes without indicating any
relationship between them. The types and objectives of the classifications
in this section may be described briefly as follows.
I. Physical classification correlated with economic data.
2. Land-capability classification for soil conservation.
3. Fractional-code method. Physical and social inventory.
4. Economic classification according to attainable intensity of use.
5. Classification for extensive wheat production according to profit-
6. Physical classification for land-settlement purposes, without regard
to economic and social factors.
7. Classification by soil types according to inherent productivity.
8. Classification according to present use.
9. Classification for land-settlement purposes according to present
use, yields and soil-moisture conditions.
10. Classification for irrigation purposes.

The need for land-use planning in Michigan appeared acute after the
first world war when a large and increasing area of land in the northern part
of the State became idle, and forest fires were frequent and unchecked.
The importance of the land inventory as a prerequisite to land classification
and planning was clearly recognized in developing land-classification
procedure [50]. The Michigan Academy of Science, after much deliberation,
decided that a land classification must precede the operation of any land
planning, and that the classification must be based on a factual inventory
of all the essential items on which the intelligent utilization of land should
be based. The first emphasis was laid on the economic aspect, as confidence
was waning in the methods and points of view of technical agriculturists
who had permitted confidence in their technical skill, and undue
optimism-if not illusion-as to the economic practicabilities of agriculture,
to bolster the assumptions, allegations and subreptions of professional
land-boomsters [50].

The Land Economic Survey was set up in 1922 as a branch of the State
Department of Conservation. Its specific purpose was to inventory facts ;
it refused to make land classifications to be used as a basis for land-use
recommendations. The facts collected were those pertinent to the under-
standing of problems relating to forestry, agriculture and recreation, the
three major land uses in the area surveyed. Data were collected on (i)
population distribution and changes, (2) political organization and changes,
(3) assessed valuations and tax rates, (4) tax delinquency, (5) land ownership
and intent in ownership, (6) economic activities, trade and trade areas,
(7) miscellaneous data.
The data of the Land Economic Survey are reported to have been of
great use in connexion with matters such as the establishment of State
forests, the determination of acquisition policies and land-use planning,
but by themselves they do not form an adequate basis for a land classifi-
cation. For this purpose a more rapid, simple and less detailed determination
of land types was required. The concept of the land type as a unit of
inventory and classification, consisting of various unique combinations of
soil types, relief and drainage features and topographic forms, has been
much developed by Veatch [95, 96] at Michigan State College of Agriculture.
Obviously, the number of combinations of physical features is almost
infinite, and the art of the land classifier consists largely in defining the
prototypes and relating the innumerable variants to those. Veatch [96]
lists the following as types of land or combinations of natural conditions
that have influenced the determination of present land use and farming
systems in Michigan.
I. Uniformly smooth or level land with fertile, durable soils dominant,
but in combination with smaller bodies of less fertile but arable land. Such
land can be laid out in large rectangular fields, and large machinery units
can be used advantageously.
2. Arable soils and level land associated in small bodies with hills
and slopes. The soils vary distinctly in fertility, moisture content and
crop adaptations.
3. Ufiiformly level dry surface in combination with sandy soil of -low
4. Uniform mineral soil dotted with peat depressions and lakes, or
peat and muck swamps dotted with ridges and hills of dry mineral soils.
5. Fertile soils, excessively stony, in combination with steep slopes
or poor drainage.
6. Arable rolling land with gentle or moderately steep slopes containing
a dominant amount of productive loams, but with small bodies of inferior
7. Uniform soils, medium fertility, lying on level dry plains, dotted
or intersected with lakes and swamps.
8. Complex associations of wet and dry soils which are dominantly
low in productivity. The land is smooth or has only low relief features.
9. Hills or dunes of deep infertile sands subject to shifting by the

The land type usually embraces several soil types. A descriptive
inventory of a land type (taken from the legend of Emmet County land-type
map) reads as follows:
Brutus land type.-Vegetation (Major) maple, elm, aspen, ash, hemlock
(Minor) cedar, spruce, balsam. Surface, level and gently rolling. Soils
(Major) Selkirk, Brimley, Ogemaw (Minor) Saugetuck, Newton, Rubicon.
Drainage, slow to poor.
Land types are first determined by reconnaissance field observation,
and afterwards delineated with reference to soil types from standard soil
maps. Individual soil types are not, however, distinguished within a land
type on the land-type map. It will be seen that the type is essentially an
inventory of physical conditions significant to land use. Its primary
purpose is to provide a simple and intelligible presentation of soil data by
grouping soil associations and their physical characteristics into units
applicable to planning purposes. It is claimed that such grouping forms
a rational physical basis for correlation with economic and social data.

The conservation surveys of the U.S. Soil Conservation Service involves
the mapping of four main land characteristics-soil, slope, condition of
erosion and present use [62]. Soils are mapped by the standard methods
of the U.S. Soil Survey, each soil type being allocated a number in the map
legend. Slopes are measured in percentages and grouped into percentage
ranges which vary with the erodibility of the soil. Thus the slope classes
of two different soils (No. I being the more erodible) have the following
significance :
Soil No. I Soil No. 2
Slope class % slope Slope class % slope
A 0-3 A does not occur
B 3-8 B ,, ,, ,,
C 8-15 C 15-25
D 15-25 D 25-35
E does not occur E 35 and over
Erosion conditions are expressed by numerical symbols indicating the
kind and degree of erosion. The symbols are standard for the Soil Con-
servation Service, e.g., 3 indicates 25-50 per cent, 33 indicates 50-70 per
cent, loss of topsoil, 7 indicates occasional gullies more than 100 feet apart.
Land types are distinguished by a three-part symbol, the first part of which
refers to the erosion conditions, the second to the slope and the third to the
soil type-e.g., 337-B-16. A new land type is mapped wherever one part
of the symbol changes. Present land use is mapped separately and is
indicated by capital letters-L cropland, H farmsteads, P pastures, F wood-
land, X idle land. Present use is not here involved in the land classification,
but it shows to what extent the land is being inadvisedly used and where
readjustments are required.

On the basis of the three characteristics-soil type, slope and erosion
conditions-land is classified according to its use capability or suitability
for agricultural use. Eight classes are distinguished [36] :
Suitable for cultivation with
I. no special practices;
II. simple practices ;
III. intensive practices.
Suitable for occasional or limited cultivation with
IV. limited use and intensive practices.
Not suitable for cultivation, but suitable for permanent vegetation with
V. no special restrictions or practices;
VI. moderate restrictions in use ;
VII. severe restrictions is use.
Not suitable for cultivation, grazing or forestry
VIII. usually extremely rough, sandy, wet or arid land that may
have a value for wildlife.
These use-capability classes are defined very generally to begin with,
and sharper definitions are given in accordance with the conditions and
practices prevailing in the region being surveyed. The classes are deter-
mined solely on the basis of physical characteristics of the land, i.e., of the
soil and climate [64]. The chief characteristics are (i) susceptibility of the
soil to erosion when cultivated, (2) natural soil productivity, (3) factors
interfering with cultivation, e.g., stoniness, hardpan, (4) climate, particularly
temperature and precipitation.
As an illustration of the factors considered in determining a land
class [36] Class-I land is described as land highly suitable for cultivation,
for it does not have a permanently high water table ; neither is it stony or
spotted with rock ledges ; nor does it possess any other physical characters
which interfere with the use of tillage implements. Furthermore clean-
tilled crops like corn, cotton, or tobacco, the growing of which is often likely
to result in soil washing, can be raised on this land without danger of
appreciable erosion. Finally, it retains and supplies sufficient moisture
and plant nutrients to maintain those physical, chemical, and biological
conditions of the soil that favor continued production of moderate to high
yields of farm crops."
Special practices apparently include any conservation measures
such as rotations, strip cropping, terracing, drainage, not ordinarily used in
the locality. A simple rotation that is customary in one place is not a
special practice, but the same rotation where it is not customary is a special
Field mapping is usually done on aerial photographs in detailed conser-
vation surveys on a scale of four inches to one mile. Smaller scales are
used for reconnaissance surveys.
A trained soil surveyor determines and maps the soil, slope and erosion
conditions, but the actual classification or grading of the land types mapped
is the result of consultation between surveyors, planning technicians,
agricultural scientists and farmers.

This type of land classification is exceptionally easy to translate into
land-use recommendations. Table 2 illustrates how this is done [62].

Use recommendations based on use-capability classification
Land-use Supporting
capability Land use Cropping systems conservation Soil treatments
class practices

I. Cultivation Row crop 2 years, None Manure
small grain, hay
Row crop, small grain, Contour strips, Lime +manure
hay 75'-125', or terraces
do. Row crop, small grain, Contour cultivation do.
II. hay 3 years
Row crop, small grain, None do
hay 4 years
Pasture None None do.
F Row crop, small grain, Contour strips and do.
hay terraces
Cultivation Row crop, small grain, Contour strips, do.
III. hay 2 years 60'-100'
Row crop, small grain, Contour cultivation do.
hay 4 years
Pasture None None do.
do. None Controlled grazing. Lime+phosphorus
Contour furrowing
for adapted soil,
slope and vegetative
IV. cover
Woodland None Protection and None
In case it is necessary to cultivate Class-IV land, special intensive practices will
L be planned to meet the conservation needs.

The systems recommended in the third column are determined not only
by the land class, but also within limits by the capacity and preferences
of the occupier of the land.
The type of classification described might be defined more correctly as
of soil rather than of land. It is a classification with a definite and limited
objective-the control of soil erosion. Factors like inherent fertility, crop
adaptability and economics are only incidental, the overriding consideration
in using the classification for land planning being whether or not the
adoption of a certain practice will result in erosion or, more generally, soil
deterioration. There are, however, very large regions all over the world
where erosion control must now be the primary purpose of land planning,
and for such regions a relatively simple land classification, similar to that
used by the U.S. Soil Conservation Service, should be useful.
One might with some justice describe the purpose of a soil-conservation
survey to be planning for permanent settlement. It is a straightforward,
easily defined objective. Without adequate soil conservation permanent

settlement and long-range land planning become impossible. A land
classification based on the requirements of soil conservation should therefore
constitute at once a fundamental and a practical classification-
practical because it considers the economic and technical aspects of land
utilization, and fundamental because it relates these to the limitation
imposed by the natural environment. In Western Europe which, as
Eisenhower [19] has stated, is the only developed region of the world that
is practising total soil conservation, the pre-war pattern of land use was the
resultant of farming systems evolved primarily for soil conservation and
modified within the soil-conservation framework by economic, political
and social factors. It seems inevitable that those countries which are
attempting consciously to plan their evolution towards the settled state
must also base their land-use planning primarily on soil conservation.
The land classification here described has the great merit of simplicity of
determination and expression of land classes, and might serve as a model
for many other classifications with the same or a similar objective.
The land-capability classification is fairly permanent, but changes
either in the land (caused, e.g., by erosion) or in the methods of use or
protection (e.g., irrigation developments, drainage, erosion control) may
make reclassification necessary.
With the aid of a land-capability map, a present-use map and a recom-
mended-use table as given above it is a fairly simple matter to replan land
use on a conservation basis. Figs. I to 3 show how this was done for a
farm in South Carolina [36]. Normally, of course, a farm is not planned
independently, but in relation to the physical nature and use of surrounding

----- Stream U Buildings Rood

.-.. -- Intermittent stream --X- Fence ==- Form road
FIG. 1.-Land-capability map of a 151-acre farm containing land of Classes I, II, III, IV,
and VII. Symbols show the soil types, slopes, and erosion classes. Soils are: 1, alluvial
soils, undifferentiated; 19, Seneca sandy loam; 30, Cecil sandy loam; 33, Cecil clay loam;
and 40, Appling sandy loam. Slopes are: A, less than 2 per cent; B, 2-7; C, 7-10; D, 10-14;
E, 14-25; and F, 25 or more. Erosion classes are:+, alluvial land along streams; 2, sheet
erosion, 25 to 75 per cent of topsoil removed; 27, the same with occasional gullies;
37, sheet erosion, more than 75 per cent of topsoil removed, and occasional gullies. A circle
around a gully symbol indicates gullies too deep to be crossed with tillage implements.

Cropland Woodland

--- Stream

-. -- InlCTr,.leril ir, reorm

-x-- Fence


Idle land = Posture =-=:- Form rood

FIG. 2.-Land use on the farm shown in Fig. I before the new farm plan was made.

Str .eo lepseaeza Woodland

Homestead Cropland *'j Posture

'- Kudzu border Terrace ,-r Diversion ditch

Permanent stream --...- Intermittent stream Paved road

---- Rood ====== Form road -- -- Fence

I- New fence =z'uz Slobilized bullet for water disposal

FIG. 3.-Land use in the farm-conservation plan.

The unit-area method of land classification has been developed by
Hudson [38] and collaborators for use by the Land-Classification Section
of the Tennessee Valley Authority. It represents an attempt to bridge
the gap between methods of detailed field analysis and of reconnaissance.
The T.V.A. required a method of land survey by which the whole Tennessee
Basin could be rapidly and accurately covered. The best reconnaissance
methods were too general to yield adequate data ; other methods were too
slow and expensive [62].

The method finally evolved is an adaptation of the fractional-code
method-i.e., the qualities of the land significant in planning are represented
by digits, one set of digits relating to physical conditions composing the I
denominator, and another set relating to human conditions the
numerator, of a fraction which describes and defines the land class. In the
unit-area method fractional-code notations are applied to land units of not
less than 200 acres. It aims at giving an accurate quantitative portrayal
of the occupancy pattern ".

The complete symbol of a land unit comprises three parts, namely, a
long fraction, a short fraction and a Roman numeral, e.g., II 3 2D323
3 2331233
Detailed classification has been made mainly of agricultural land,
though the method can be extended, if required, to forest, urban, recreational
and other areas.

For agricultural land the physical conditions of slope, drainage,
erosion, stoniness, rock exposure, soil depth and soil fertility are first
determined. These are expressed in the above order and according to the
criteria given in Table 3 as the digits of the denominator of the long fraction.
The unit is then classified on the human or cultural features of major
agricultural use, agricultural emphasis, field size, amount of idle land, and
quality of farm buildings and equipment (Table 4). The corresponding
digits comprise the numerator of the long fraction.

The short fraction is a summary of the long fraction. The denominator
represents the judgment of the field man, based on all observable physical
factors, whether recorded or not, as to which of five quality classes the land
unit belongs ; the numerator likewise represents the field man's judgment
of quality class based on observable cultural data. In Tables 5 and 6 only
the criteria for assessing classes i and 5 are given for the sake of brevity.
From these the close relationship between the short and long fractions will
be apparent.

Major land uses or human items on the basis of which homogeneous
land uses are delimited. Noted by the digits in the numerator of the
long fraction

First Digit Second Digit Third Digit Fourth Digit Fifth Digit

Major Land Use Agricultural Field Size Amount of Idle Quality of Farm-
Emphasis Land steads and
1. General farming A. Corn 1. Large 1. Little 1. Excellent
2. Animal industry G. Grain (small) 2. Medium 2. Limited 2. Good
3. Cash-crop farming B. Beef cattle 3. Small 3. Considerable 3. Medium
4. Part-time farming D. Dairying 4. Very small 4. Excessive 4. Poor
5. Subsistence S. Sheep 5. Very poor
farming H. Hogs
6. Forest land M. Mules and/or
7. Recreational area P. Poultry
8. Rural-village area T. Tobacco
9. Urban area C. Cotton
0. Manufacturing and W. Truck
mining areas 0. Orchard
N. No emphasis
F. Forage

Major physical conditions of the land on the basis of which homogeneous
units are delimited. Noted by the digits in the denominator of the
long fraction

First Digit Second Digit Third Digit Fourth Digit

Slope Drainage Erosion Stoniness
1. Relatively level 1. Thorough 1. Little or no observable 1. Free from stones
2. Relatively level to 2. Adequate 2. Little denudation by 2. Moderately stony
undulating erosion
3. Undulating to moder- 3. Poor 3. Sheet erosion and 3. Stony
ately hilly ephemeral gullies
4. Hilly 4. Very poor 4. Excessive sheet erosion 4. Very stony
and gullying
5. Steep 5. Excessive 5. Excessive gully erosion

Fifth Digit Sixth Digit Seventh Digit

Rock Exposure Soil Depth Soil Fertility
1. Little or no rock exposure 1. Deep (6 feet or more) 1. Exceptionally fertile
2. Limited rock exposure 2. Moderately deep (3 to 6 feet) 2. Fertile
3. Considerable rock exposure 3. Shallow (1 to 3 feet) 3. Moderately fertile
4. Excessive rock exposure 4. Very shallow (less than 1 4. Very low in fertility
5. Rock exposure barren foot) 5. Very low in fertility

Criteria for classifying land on the basis of the agricultural quality
of present physical conditions
(denominator of the short fraction)
Class I. Units of 200 acres or more that are characterized by the following
physical indices, appearing individually or in combination : (I) level
or moderately undulating surface ; (2) adequate or thorough drainage ;
(3) little or no observable erosion; (4) deep stone-free soil of exceptional
fertility ; (5) little or no rock exposure.
Class 5. Units of 200 acres or more that are characterized by the following
physical indices, appearing individually or in combination: (I) steep
slopes ; (2) very poor or excessive drainage ; (3) denudation by sheet
or gully erosion beyond the point of cultivation and economically
feasible rehabilitation other than by reforestation ; (4) shallow or
stony soils, very low in fertility ; (5) excessive rock exposure.

Criteria for classifying land on the basis of the quality of the present
agricultural use
(numerator of the short fraction)
Class i. Units of 200 acres or more that are characterized by the following
use indices, appearing individually or in combination : (i) medium
and/or large uninterrupted fields; (2) fields free from weedy or brushy
cover ; (3) little or no idle land ; (4) excellent farmsteads and farm
equipment ; (5) any other evidence of an excellent standard of living.
Class 5. Units of 200 acres or more that are characterized by the following
use indices, appearing individually or in combination : (I) very small
and/or interrupted fields; (2) sparse, weedy or brushy field cover;
(3) excessive amounts of idle land; (4) very poor farmsteads and farm
equipment ; (5) any other evidence of a very low standard of living.
Finally, the land is put into one of five classes according to the severity
or absence of problems and needs for readjustments. These classes are
indicated by the prefixed Roman numerals I to V. In part, these Roman
numerals represent a summary of their accompanying short and long
fractions ; but again the entire complex in all its pertinent observable
respects is carefully considered and appraised before the field judgment is
made." The ten criteria for distinguishing each class are the same as the
corresponding criteria given, for Classes I and 5, in Tables 5 and 6. The
classes are briefly defined as follows :-
Class I. Units in which no significant agricultural problems are apparent.
These units are characterized by an excellent standard of living and by
land exceptionally well suited for both general and specialized types
of agriculture.

Class II. Units in which the apparent agricultural problems are not
critical, but can generally be solved by education and demonstration.
The units are characterized by a good standard of living and by land
well suited for general and specialized agriculture.
Class III. Units in which agricultural problems are moderately critical,
but can be solved by intensive education and demonstration. Charac-
terized by a medium standard of living and by land which can be used
for general or special crops under proper management.
Class IV. Units in which agricultural problems are very critical, and can
usually be solved only by a marked readjustment and/or reorientation
of economic activities.
Class V. Units usually only suitable for forest use, alternatively for
recreation, game preserves; also waste land. Characterized by a
very low standard of living and by land unsuitable for agriculture.
The long fraction can be regarded as an inventory of the physical and
agricultural conditions, the short fraction and the Roman numeral as
assessments or appraisals of the inventory.
A new body of land is delimited whenever a change occurs in one or
more of the items recorded. The fractions and numerals are recorded in
ink on aerial mosaics, and subsequently each unit area is put into one of
three general classes :-(i) land suited to intensive uses (with emphasis on
cropping) ; (2) land suited to extensive use (with emphasis on pasture) ;
and (3) land suited to non-agricultural uses. This classification is based
on data not only from the unit-area survey, but from all other available
Non-agricultural or forest land is classified and mapped simply by the
appropriate figure 7, 8, 9, or o of the first digit of the numerator of the long
fraction (Table 3). No further characterization is made. Forest land is
identified by the numeral 6 in the numerator of the long fraction, and in
the denominator the same seven physical conditions are noted as for agri-
cultural land, e.g., 4122114 Forest types are not distinguished by the
Land-Classification Section, as these are the concern of the Forestry Division
of the T.V.A.
The suitability of this method of land classification and mapping for
the Tennessee-Valley region is due to the fact that a considerable amount
of detailed field work has been carried out, that aerial photographs are
available as base maps, and that it is possible to conduct operations on a
fairly large scale. The method has been successfully used, with the necessary
modification in the significance of the digits, in New Zealand where similar
favourable conditions for its application obtain [18].

Land-classification studies have been made of many counties in New
York. Their main purpose has been to classify and map the land according

to its adaptability to different degrees of intensity of use. To plan a land-
use programme on the basis of such a map it is necessary to have knowledge
of the natural factors limiting intensity of use, the factors which have
determined present intensity of use, and those chiefly affecting the economic
status of the farming community. It has been shown that in New York
commercial, social and financial amenities such as roads, electricity supply,
schools and credit facilities have a preponderating influence on the intensity
of land use. At the same time the growth of such amenities is much
influenced by the quality of the land.
In the New-York system the lowest land class is numbered I, and is
natural forest land so poorly adapted to agriculture that it has seldom-
and then unsuccessfully-been cultivated. Class-II land is also better
suited to forest, nevertheless much of it has been brought under cultivation.
Classes III to VII are suited to continuous agriculture of increasing degree
of intensity of use.
The first operation in making a land classification is the preparation
of a property-classification map. Farms are classified, usually by
inspection from a moving car, according to the apparent amount and condition
of the farm capital. The main evidence is obtained from the size and
condition of farm buildings, but the state of roads and the distribution of
utility services and amenities are also taken into account. It is considered
that farmers are continually struggling to adjust their farming operations
to the character of the land in such a way as to earn as much as they can
and to accumulate the largest possible amount of capital. Variations in
the amount and condition of the capital which experienced farmers have
been able to accumulate and maintain on the land, therefore, measure the
combined economic effect of lay-of-the-land, climate, soil, and markets, and
are one of the most reliable indications of the intensity of use to which the
land is adapted [62].
A land-use map, indicating actual intensity of use, is then prepared.
Maps of topography and soils, both being important factors influencing
intensity of use, are also required.
The delineation of land-class boundaries is done primarily from the
property-classification map which roughly indicates the amount of capital
a land type is capable of accumulating. A first adjustment of the boundaries
is then made with the aid of the topographic map, since the actual boundaries
of land classes are more likely to correspond to topographic than to property
features. Further adjustments are made in turn from the land-use map
and from the soil map ; consideration of the distribution of soil types some-
times results in radical modifications of the land classes distinguished on the
basis of property condition and present intensity of use. Finally, the
land-class map is checked in the field against type of use, condition of
buildings, topography, etc. It is stated that differences between land
classes are usually immediately apparent to those who are accustomed to
land-class maps, though not always to others.

The maps are supplemented by statistical data which show the quanti-
tative differences between the land classes. Of particular significance are
the differences between classes as indicated by comparative farm-business
data. It has been found, for example, that the gross and net incomes of
farmers on the better-class lands have been consistently higher than those
on the lower classes since 19o08, the earliest date for which farm-management
data are available.

Land in the prairie provinces has two main uses-wheat production
and grazing. The former has been and will probably continue to be the
major use in Saskatchewan, though grazing may surpass it in Alberta [15].
Distinction between land classes is based on the quantity of wheat available
for sale per quarter section, converted to net productivity. There are four
steps in the classification :-
I. determination of gross productivity per quarter section of land;
2. conversion of gross productivity into terms of net revenue ;
3. classification of parcels on the basis of certain ranges of estimated
net revenue ;
4. modification of such classification on the basis of the characteristics
of individual parcels.
Saskatchewan [79]. Each quarter section is assessed in the light of all
relevant physical and economic information available, special weight being
given to the history and past productivity of the land. It is assigned to
one of five classes, class I being submarginal, class II marginal, and classes
III to V increasingly favourable for wheat production.
The classification hinges on the definition of marginal land-class II.
Marginal land is defined economically as land which . would, if
-operated with average managerial ability, in a unit of average size and
typical organization, be expected to pay expenses-including depreciation
and taxes-and provide a living for the operator and his family. It would
not yield sufficient to pay for the use of the land either as rent or interest."
Farm-management studies have shown that marginal wheat farming-
i.e., so that the operator gets the cash equivalent of a hired man's wages-
-corresponds to a production of wheat available for sale of 350 to 475 bushels
per quarter section. These are therefore taken as the lower and upper
limits of productivity of class II. Class-I land produces less than 350
bushels available for sale; class III from 476 to 720; class IV from 721
to 900; and class V over 900 bushels available for sale.
This initial classification shows the relative productivity not of specific
land or soil types, but of quarter sections. In the final classification this is
modified by taking into account factors such as the effect of physical land
characteristics on farm economy, variability of yield, progress towards
permanent settlement, and main causes of success or failure.

The productivity is calculated primarily from (i) the annual yields
per acre estimated from data obtained for different soil types for the years
1921-1936 multiplied by (2) the number of acres of cultivable land per
quarter section. 1.5 bushels per acre is deducted from the estimated yield
to allow for seed and farm use, and it is assumed further that on the average
45 per cent of the total cultivable area will be used annually in the pro-
duction of wheat for sale.
As might be expected, there is quite good correlation between these
mainly economically determined land classes and the financial and economic
status of the occupiers in each class. Thus the land class is quite closely
reflected in the price paid for land, tax assessment per acre, condition
of buildings, percentages of mortgages foreclosed, and percentages of
privately-owned and owner-operated land within a class. These character-
istics are clearly associated with the relative yields of wheat on the different
land classes. The wheat yield is an expression of the land's economic
performance, and the land-classification map based thereon would, with
minor adjustments, become a land-price or tax-assessment map, for
-example. This, presumably, is its purpose. It is claimed that the classi-
fication gives a general picture of productive capacity, and would reflect
with reasonable accuracy the relative productivity of the land in other
uses than wheat. It is questionable, however, whether this would be so
unless the other uses were operated on an economy similar to that of
extensive wheat production.
There is apparently no close correspondence between the distribution
of land classes and soil types. Sands and gravels predominate in classes
I and II, while most of the heavy clays occur in class V. The purely
economic criteria by which land classes are distinguished make it unlikely
that any close correspondence should exist, since many other variable
factors besides soil contribute to land's economic performance. To a limited
extent it was possible to equate different combinations of soil and topo-
graphic conditions within a land class, e.g., in classes I and II very poor,
coarse, alkaline flat and blow-out soils may be associated with a flat
topography favourable for cultivation, or somewhat better loams with
rolling or broken topography.
Alberta. A similar system of classification, and for similar purposes, has
been used in Alberta where, owing to large and frequent fluctuations in
weather and wheat prices, agricultural settlement and production have been
very unstable. It is felt that the classification of the land on an optimal-use
basis is the first step towards planning to establish greater rural stability.
The general approach to land classification is described by Stewart and
Porter [83] as follows. The use of land depends upon the decisions of
individual farmers; and the decisions of farmers depend on their expecta-
tions of the returns they can secure from land. Stable land use requires
that, over a period of years, farmers can expect to meet their expenses,
and secure for themselves returns sufficient to induce them to continue

production without expansion or contraction. Under given prices, and with
prevailing methods of production, some land may be incapable of providing
for these necessary costs (submarginal land) ; other land will be just suffi-
ciently fertile to cover costs (marginal land) ; more fertile land will yield a
net revenue over the essential costs. Land may therefore be classified on
the basis of its estimated capacity to yield net revenue."
The classification used by Stewart and Porter is fundamentally an
economic one, and depends on an estimate of the net revenue obtained from
the land under wheat. The determination of the net revenue is the crux
of the method. The classification assumes that the land is not in use, and
that farmers are considering the contracts and payments they should make
in connexion with securing the land.
The procedure of classification involves the following steps.
(a) Determination of a physical productivity rating for each
parcel, based on (i) the average long-run yield of wheat for the soil
type of the parcel, (ii) the acreage of tillable land, and (iii) the typical
proportion of tillable land in wheat.
(b) Conversion of this measure of gross physical productivity into
terms of estimated net revenue using (i) long-term average prices for
wheat and other farm products, and (ii) a budget of costs derived
from farm-management survey data.
(c) Preliminary classification on the basis of certain ranges of
estimated net revenue. Four classes are distinguished :
Class I. Estimated annual production of marketable wheat per
quarter available for sale less than 375 bushels, and
estimated revenue less than costs (submarginal).
Class II. Estimated annual production of marketable wheat per
quarter available for sale 375-517 bushels, and esti-
mated revenue equal to a range of costs (marginal).
Class III. Estimated annual production of marketable wheat per
quarter available for sale 518-795 bushels, and esti-
mated net revenue up to $237 per parcel of 480 acres,
and based on a wheat price of 92 cents a bushel.
Class IV. Estimated annual production of marketable wheat per
quarter available for sale 796-999 bushels, and esti-
mated net revenue of more than $237 and less than
$411 per parcel..
This classification provides a uniform basis on which to compare parcels
in widely scattered areas with significantly different general features.
Modification of the preliminary classification of parcels is necessary to allow
for particular features of individual parcels, for example, topography,
stoniness, erosion, climate and other physical characteristics. Adjustments
are based on descriptive notes made in the field, aerial photographs and
re-examination, in the .field.

Hills [35] has outlined a system of land classification applicable to
land-settlement problems in a thinly populated, slightly developed region
of the Great Clay Belt of northern Ontario. Owing to lack of communi-
cations and the density of the natural plant cover soil survey was impossible,
and the area was mapped in land types distinguished according to the
criteria of land form, geology, drainage, soils and natural vegetation.
Related land types were grouped into land classes.
Nine land classes were established, distinguished by the Roman
numerals I to IX. No qualitative significance attaches to these numerals.
The classes are objectively determined. Thus Land class No. I occurs
on the better drained areas dissected by stream courses or on comparatively
low ridges where heavier materials cover a core of rock-knob or glacial
debris. In this class are found soil profiles exhibiting a considerable degree
of podzolic leaching. Land classes II and III are half-bog soils with a
maximum peaty layer of approximately 18 to 36 inches, respectively. Land
class IV includes the deep peat or muskeg. Land class V is a complex
condition in which variable depths of intermediate and heavy textured
materials overlie the fluvio-glacial sands and gravels. Under average
conditions, the heavier surface layers supply fertility, and the lower open
layers provide good drainage with the result that the most productive soils
are found in local areas within this type. Outwash materials of intermediate
texture are included in Land class VI. Ridges of coarse sandy and gravelly
materials constitute Land class VII. Although these deposits are similar
to the underlying materials of Land class V, the soils are drought and low
in fertility because of the absence of the heavier surface layer. Fairly
compact, loamy, till soils constitute Land class VIII while the outcrops of
the Precambrian rocks, with their shallow mantle of light soil are mapped
as No. X." For some reason, No. IX has been omitted.
The agricultural or forestal capabilities of these classes are being
determined on the basis of physical data, present use and past experience-
chiefly, it would seem, of experience. For example Land classes I, II and
portions of V have the highest agricultural potentials. Local conditions
of soil drainage and climate in No. V are most favourable for the production
of potatoes. Classes III and VI have only fair agricultural potentials. Few
soils similar to those in Land class VII are used extensively for agricultural
production in America at the present time. Land classes IV and X have
practically no agricultural value. For forestal purposes areas of classes
I and V indicate the best saw-timber sites. In the mature stands of black
spruce on Land class II, 2-log trees (which make 32 feet of pulpwood) are
dominant while i-log trees are common on Land class III. Portions of
Land class VII are good pine lands."
Political, economic and cultural factors do not enter into the deter-
mination of land classes or of their use capabilities, but come into operation
in formulating a plan of development based on this land classification.
The plan, for example, might be to develop the best classes of land first.

This would require modification in practice according to costs of develop-
ment, proximity to social and economic services, pressure of population,
market facilities, etc., in each area. The political factor of whether costs
of and responsibility for development were assumed by public or private
authorities would also affect the execution of the basic plan.

Grange [26] has attempted to classify land in the North Island, New
Zealand, according to its inherent productivity, on the basis of the soil
survey. The soils are firstly classified into genetic groups which are
then subdivided into stages-young, immature, semi-mature, mature-
and finally into series and types as in the American system. The stage
subdivision is in effect an inherent-productivity classification within a
genetic group, young soils being the most, and mature soils the least, fertile.
The soil map itself has but limited practical use, as there are more than
500 different types in the North Island, but from it single-factor maps
(e.g., of lime or phosphate requirement, erosion, texture) are being prepared
which have a direct application to advisory work. The soil types have
also been grouped into six fertility or use-capability classes, with special
reference to pasture utilization, namely [63] :-
Class I.-Level or undulating land, not too elevated, with deep soils
and favourable moisture conditions, and which are, or can be,
converted into high-quality farming land.
Class II.-Ploughable land which can be converted into only fair- or
medium-quality farming land on account of some limiting factor
to productivity.
Group (a) : Soils in which moisture is a limiting factor.
Group (b) : Soils in which some other factor such as texture,
structure, drainage, elevation, or depth of soil
is a limiting factor.
Class III.-Ploughable land which has severe limitations to productivity
and requires further investigation before development is attempted.
Class IV.-Hilly or steep land which will maintain grass pasture with
little or no top-dressing. Both topsoil and subsoil are of high
fertility, and erosion is not a serious problem.
Class V.-Hilly or steep land of moderate to low fertility. Light
top-dressing is required to maintain a cover of grass, and careful
management is necessary to prevent serious erosion.
Class VI.-Hilly or steep land which has severe limitations to utilization,
such as low fertility or erodibility. This class is probably more
suited to forest than to grass.

In a country of long and politically stable settlement, present land use
(except in apparently abnormal periods such as wartime) constitutes a

natural basis for a land-planning classification. Such a statement implies
that land planning is not required, or would not involve much drastic
modification of present use, in a country which has enjoyed long political
stability, and in general this is borne out by facts. The countries of Western
Europe are politically the most mature and before 1940 seemed the most
stable in the world; they did not evince the same interest in land planning
as the politically immature countries of the New World or of Central and
Eastern Europe. A prolonged period of undisturbed national development
allows an equilibrium to be approached between the numerous interacting
forces affecting the utilization of the land. In Britain much of the land
has been settled and farmed for upwards of a thousand years, some of it for
two thousand. The present utilization does therefore reflect the combined
influence of the varied factors concerned : the physical factors of relief,
soil and climate ; the more purely economic and social factors of markets,
prices, transport, labour ; the historical factors of land ownership, local
usage, tenure, and even tradition or custom. It is only when the attempt
is made to analyse present land use by reference to only one set of factors
that dangerous fallacies of interpretation may be promulgated [81].
No nation-wide soil survey has been carried out, and the only com-
prehensive scheme of land classification is that adopted by the Land
Utilization Survey of the University of London, and based originally on
existing land use, i.e., during the period 1931-1938. This period witnessed
the lowest intensity of land utilization since the Industrial Revolution.
The land is divided into ten major types numbered I to Io. Whilst types
I to 4 inclusive are land of high agricultural value, types 8 to 10 land of low
agricultural value, and types 5 to 7 land of intermediate quality, there is
not a steady gradation in utility and value from I to 10. First-class arable
land and first-class grazing land may be equally valuable, but the two types
of utilization require different physical conditions. Certain types primarily
suitable to arable farming and crop production are distinguished by the
addition in brackets of the symbol A (arable) ; others naturally suited to
pasture or meadow are distinguished by the symbol G (grassland). Still
others can serve in either utilization (A-G), but the poorer types of land
agriculturally sub-marginal under most economic conditions are normally
under heath, moor or rough pasture, and are given the suffix H
The ten classes are grouped into three Major Categories, distinguished
by the following characteristics of site and soil:
Major Category I-good-quality land
Site (i) not too elevated,
(2) level, gently sloping or undulating,
(3) favourable aspect.
Soil (I) deep,
(2) favourable water conditions,
(3) texture, mostly loams, but including some peats, sands,
silts and clays.

Major Category II-medium-quality land
Land of limited productivity by reason of the unfavourable operation
of one or more site or soil factors, e.g.,
Site (i) high elevation,
(2) steepness,
(3) unfavourable aspect.
Soil (i) shallowness,
(2) defective water conditions.
Major Category III-poor-quality land
Land of low productivity by the extreme operation of one or more site
and soil factors. Four groups of extreme factors are recognized.
(a) extreme heaviness and/or wetness of soil, giving poor-quality
heavy land or land in need of extensive drainage works;
(b) extreme elevation and/or ruggedness and/or shallowness of soil,
giving mountain moorland conditions;
(c) extreme lightness of soil with attendant drought and poverty,
giving poor-quality light land;
(d) several factors combining to such an extent as to render the
land agriculturally useless or almost so-such as shingle
beaches or moving sand dunes.

The complete ten types and the estimated percentages
Wales occupied by each are :-
Major Category I-good
i. First-class 5.5
2. Good general-purpose farmland-
2 (A) Suitable for ploughing 20.3
2 (AG) Crops or grass 6.3
3. First-class, restricted use,
unsuitable for ploughing 3.6
4. Good but heavy land 13.1

Major Category II-medium
5. Medium light land-
5 (A) Suitable for ploughing
5 (G) Unsuitable for ploughing
6. Medium general-purpose farmland


Category III-poor
Poor heavy land
Poor mountain and moorland
Poor light land
Poorest land .



of England and



Residue-closely built over

That present land use in Britain is, broadly speaking, indicative of land
quality is shown by the remarkable constancy of land use over long periods
of time on the best and worst types, and the maximum of change on the
intermediate types. Historical records show that the best arable land has
often remained under the plough almost continuously for a century or
more, while the worst types have remained uncultivated. Neither the
long, pre-war slump in arable farming nor the war-time boom appreciably
affected the utilization of these types. The intermediate types, on the
other hand, include those which largely went out of cultivation during the
last seventy years and were the first to be ploughed up during the war.
These classes were originally distinguished according to present use,
not directly according to soil, climatic or geographical characteristics except
in so far as these have determined present use-as in many cases they have
to a very large extent. The purpose of the Land Utilization Survey was
merely to inventory land use, not to classify land. But as public interest
in land planning has grown, a need for a complementary land classification
has also become apparent, and has been filled by adapting the land-use
inventory. It will be seen from the above descriptions of the classes that
they are distinguished not solely on a use basis, but also according to the
characteristics of soil and site. There is implied in this classification, as in
all others, the outline of a land-use plan-i.e., the classes are distinguished,
and the criteria for distinguishing them are selected, with more or less
definite uses in view. The plan here postulates the reservation of the most
productive lands for agriculture, consequently, the classes are distinguished
mainly by agriculturally significant criteria. Stamp [81] suggests :-
(1) that the good agricultural lands, categories I to 4 inclusive, which
occupy about half of the surface, should be reserved for agricultural
use, and that other forms of utilization should only be permitted
in exceptional cases, each specific case to be fully justified ;
(2) that schemes requiring extensive tracts of land such as national
parks, afforestation, green belts, main highways of parkway type,
water-supply and power schemes, should be restricted as far as
possible to the poor lands;
(3) that the integration of national needs in industry, housing, com-
munications, including provision of sites for garden cities and
satellite towns should be particularly on land of intermediate
quality (5 and 6), since poorer land is only in some areas suitable
or available, it being borne in mind at the same time that some
of the poorest land is particularly adapted for heavy industry.
For a highly developed country like Britain it is probably impossible
to make more detailed recommendations for land use on the basis of a
physical and present-use classification, since it is economic rather than
physical factors which are often the determinant.

A more detailed ecological classification of grasslands has been made
by Stapledon and Davies (cited by [81] ) who distinguish the following
eleven classes :
(a) First-grade ryegrass pastures in which perennial'
ryegrass contributes 30 per cent or more to the
(b) Second-grade ryegrass pastures containing 15 to
30 per cent of perennial ryegrass.
(c) Agrostis-with-ryegrass pastures : being basally
Agrostis, but perennial ryegrass contributing up Chiefly
to 15 per cent. lowland
(d) Ordinary Agrostis pastures sometimes containing types
mere traces of ryegrass.
(e) Agrostis pastures with excess of rushes and sedges.
(f) Fescue pastures, including (i) mountain fescue,
(ii) open downland, and (iii) lowland heaths
carrying fescue swards. Chiefly
(g) Nardus pastures often with excess of heath rush rough-
(Juncus squarrosus). *and
(h) Molinia pastures. hill
(j) Cotton-grass and deer-grass moors. grazing
(k) Heather moor.
(1) Fern, gorse and the like.

Osmond [65] has proposed the use of indices in regional land classi-
fication, defined in terms of the three factors site, soil and climate that
influence the growth of crops and a fourth-the development factor-
which embraces the influences of human activities and of such natural
phenomena as are not included in the first three factors.
Each factor in a region is classed as either good (i), intermediate (2),
or poor (3). A region is defined by a four-figure index made up of the class
figures of the four factors in the order site, soil, climate, development.
Thus 1213 indicates that the region has a good site, fairly good soil, good
climate, but poor development due, for instance, to past agricultural
neglect or to poor drainage. The index shows that if the conditions
indicated by the figure 3 can be remedied the area could become of use to
agriculture, whereas if the figure 3 had been first in the index (referring to
site) improvement might have been precluded in any planning scheme.

The chief site factors considered are elevation, slope and topography ;
the chief soil factors are depth, water-holding capacity, nutrient content

and structure ; and the chief climatic factors are rainfall and temperature
during the growing season. The indices are given not to soil types or
individual farms, but to regions. No definition of a region is given, but
presumably the word is used in the modern sense used by geographers.
The indices do not represent a classification, but regions can, if desired,
be classified into three categories A, B and C on the basis of the indices-
e.g., regions with indices IIII, 1211, 2111 would go into category A, 2221
and 2121 into category B, and 3212 and 3333 into category C.

A land classification based on the soil survey and used by the Preussische
Geologische Landesanstalt is described by Miickenhauser [61] who points
out that the ordinary soil map leaves out much information essential to the
land planner-for example, the cropping capacity and water economy of
the soils. To include such details would make the map too complex for
general purposes. Special evaluation maps (Auswertungskarten), for the
making of which expert knowledge of both soils and the objectives of the
land-use plan are necessary, are used to convey the information. In general
the planner wants facts about the cropping capacity and water economy
of the soils and the nature of the subsoil foundations (Baugrund). It has
been found convenient in Prussia to express these facts on two special maps,
one showing soil utilization and another showing water and subsoil
The soils are first classified as arable, grassland, forest or wasteland
soils, with the recognition that certain grassland soils, for example, could
be transferred to the arable class by drainage, etc. The arable and grassland
soils are then divided into five and two classes respectively, according to
the average yields obtained from them, and these classes are given standard
colours on the map. In the legend of the map the commonly grown plants
of each soil class are indicated in four separate columns-for grains, inter-
tilled crops, fodder and green-manure crops, and trees-and two further
columns give a general idea of the cost (Aufwand) of maintaining soil
fertility, and suitability for settlement.
The various colours for the second map are chosen so that all yellow,
red and brown tints indicate land not requiring drainage, grey tints land
with impeded drainage, and blue tints land with a water table nearer than
I metres to the surface. The corresponding soil-moisture conditions,
drainage and irrigation requirements, and subsoil conditions from the
building standpoint are indicated in the map legend. Parts of the legend
are shown in Tables 7 and 8.

TABLE 7. Legend of soil-utilization maps of the Preussische Geologische Landesanstalt


1. Red Good arable 24-27 dz./ha. Wheat, oats, Beet, beans, Lucerne, Mixed forest Moderate Very good
wheat barley, rye oil plants clover, beans liming and

3. Light brown Rather poor 20-23 dz./ha. Rye, oats, Potatoes, Clover, blue Mixed forest Some expendi- Suitable
arable rye some barley swedes lupin, ture on im-
seradella provements
5. Yellow Bad arable 10-13 dz./ha. Rye Potatoes Yellow lupins Pine and birch Heavy outlay Unsuitable
rye on improve-


1. Dark blue Fairly good 40-50 dz./ha. Sundry grasses and legumes Ash, birch, Moderate im- Well suited for
grassland hay spruce provement grass
2. Light blue Poor grassland 25-35 dz./ha. Ash, birch, Greater outlay Suitable
hay spruce needed


TABLE 8. Legend of the water and subsoil map

Colour Moisture conditions (a) Drainage Quality of subsoil for
of soil (b) Irrigation building foundations

(a) Not needed
Yellow Dry (b) Irrigation Dry, load-bearing

Light brown Good (a) Not needed Dry, load-bearing
(b) Not needed

Light grey Periodically damp (a) Very beneficial Reinforcement of
cellar necessary.

Light blue Damp. Water table (a) Necessary Reinforcement
0.8-1.2 m. deep necessary.

Stremme and Ostendorff [87] have attempted to classify land in East
Prussia especially, and in Germany generally, according to the area of a
holding required to give a minimum standard of living to the occupier.
This is in effect a classification according to soil productivity, since the
same standard of living is procurable from a smaller area the more productive
the soil is. Soils are grouped in genetic classes subdivided into textural
classes. The minimum area for existence (Mindestbetriebsgrbsse) is calcu-
lated for each soil class from the gross yield of all crops cultivated on the
soil, in two ways. In the first the yield is multiplied by the average price
-of the produce without reference to the cultivation or cropping system.
This gives the average money yield per hectare, from which the area required
for minimal existence can be calculated. The classification is simple and
said to be applicable without serious modification to the whole of Germany.
In the second way a correction is applied to allow for the standard of
management practised and attainable.
A further adjustment is made to allow for the different tractive powers
required to work soils of different texture. Estimates are based on a
peasant family of seven, but allowance has also to be made for the number
of horses needed to till the land, which is 4 on heavy soils and 2 on sandy
or moor soils. In the Marienburg district of East Prussia, the minimum
area varied between 9 and 20 hectares, the smallest (9 ha.) being on a
steppe-like brown forest soil, sandy loam overlying loam. This was two-

horse land. The minimum area for a similar, but heavier, four-horse soil
was 13.7 ha., and the lowest minimum area for any four-horse land was
12.7 ha.
The number of agricultural holdings into which the whole of East
Prussia could be divided on a minimum-area basis would be between
234,053 and 161,690 corresponding to minimum areas of 11.2 and 14.3 ha.,
respectively. The actual number of holdings in 1925 was estimated at
234,228 with an average area of 11.2 ha. The most equable redistribution
of the land among the then occupiers would have involved the enlargement
of 162,000 and the reduction of 23,000 existing holdings.

The classification of irrigated land or land intended for irrigation
presents certain peculiarities since one of the natural factors of the
environment-moisture-is artificially transformed, and the transformation
may profoundly modify another natural factor, soil. Moreover, a
faulty determination of the suitability of land for irrigation may involve a
large and irrecoverable expenditure of money, and cause permanent damage
to the soil.
The physical factors which determine suitability for irrigation are
climate, soil, relief, drainage and water supply [99]. If any of these is
unfavourable, irrigation is likely to be a failure. Geographical, economic
and social factors, such as location of the land, extent and distribution of
land types, accessibility of water supply, markets and transport, determine
whether it is feasible to supply irrigation water to potentially productive
land, and what farming systems will succeed. In some areas the character
of the population must also be taken into account.
The ideal physical features of an irrigation area are (I) a smooth land
surface, (2) a gentle slope sufficient to facilitate irrigation and adequate
drainage, and (3) deep, fertile loam soils. The first step in the land-
classification system evolved by the U.S. Bureau of Reclamation [62] is to.
determine whether the location and area of the land which approaches these
conditions in the region surveyed justify the execution of an irrigation
project. The boundaries of the arable land which can be irrigated with the
available water supply are then tentatively determined, and a detailed
topographical survey, on a scale of I : 4,800, is made, to form the basis for
planning the lay-out of irrigation canals, etc. Finally, a soil survey is made
in which, in addition to the usual soil characteristics, special attention is
paid to salinity or alkalinity, permeability of the underlying rock, and
sub-surface drainage.
The actual classification of lands for irrigation purposes is made with
particular reference to the three factors soil, topography and drainage. Five
or six land classes, designated by numerals and letters to show limiting
factors, are distinguished. i is the highest, and 6 the lowest, class. The
reason for placing land in a lower class than i is indicated on the field maps

by placing the letter S (soil), T (topography), D (drainage), or R (loose rock
in the arable horizon) after the class number. For example, the symbol
2ST indicates that the land is put in the second class because of inferior
soil and rough or steep topography. Further distinctions are sometimes
made to indicate low water-holding capacity.
Both deductive and inductive methods are used in the classification.
Observations on the success or otherwise of existing irrigation schemes are
used in judging what is likely to happen under similar conditions of soil and
topography elsewhere. Comparable conditions where the results of irriga-
tion are known are not, however, always to be found. In these cases the
soil or land types are evaluated by awarding points for the different
significant factors. Frequently a combination of deductive and inductive
methods is used.
The following are the specifications for the six land classes distinguished
by the Bureau of Reclamation [62, 66]. They are generally applicable
throughout the western United States.
First-class lands are those constituting the highest type of arable
lands, suitable in all respects for cultivation and production of crops
by irrigation. They are smooth-lying with slopes up to 5 per cent,
soils at least 4 feet in depth, medium or fairly fine in texture, mellow,
open structure, allowing easy penetration of roots, air and water,
having free drainage yet good water-holding capacity and free from
accumulation of soluble salts.
Second-class lands are of intermediate character, suited to the
cultivation and production of quite a wide range of crops. They are
not so desirable nor of such high value as lands of class i, because they
are more costly to prepare owing to slightly unfavourable topography,
more costly to farm because more irrigation water must be used, or
because water is difficult to apply. Soils may be comparatively
shallow (12-18 inches of good plough soil, 36 inches of soil penetrable
by roots and water). Drainage may be impeded and soils may contain
moderate concentrations of soluble salts (0.5 per cent or less), and
display alkalinity up to pH 9.0 or less in the top 2 feet. Texture of
soil may be too sandy or it may be a heavy clay. Topography may be
of an uneven or rough slope of less than 5 per cent or smooth slopes
of from 5 to o10 per cent.
Third-class lands are of inferior or low quality and poor pro-
ductivity. They have a restricted crop'adaptation owing to such soil
deficiencies as shallow depth (6 to 12 inches over gravel and loose rock,
36 inches penetrable by roots and water) ; extremes of texture (sandy
or heavy clay) ; or presence of sufficient loose rock on the surface and
in the plough zone seriously to interfere with cultivation. In some
instances they are lands having impeded drainage and moderate con-
centrations of soluble salts. In some areas they are lands of uneven
topography where erosion might occur.

Fourth-class lands are those which sometimes have a specific limited
utility. They are made up of heavy compact soils with tough imper-
vious layers in the subsoil. Sometimes they can be artificially drained.
For the most part these lands are considered as non-arable.
Fifth-class lands are sub-marginal. They often include lands
which are waterlogged and have an accumulation of an injurious
amount of soluble salt, or lands which are subject to correction only
through construction of flood-control or drainage works.
Sixth-class lands are permanently non-arable, consisting of lands
which have a rough, broken or eroded surface, rendering them unfit
for cultivation. Other areas consist of rough regions of raw shale, or
shallow soils over shale and sandstone. Here also are included lands.
which are too high for delivery of irrigation water in this class.
Besides the detailed surveys, reconnaissance surveys on a scale of
I : 24,000 are made to determine the general extent and location of cultivable
areas, or for use on areas under consideration for supplemental water
supply, and semi-detailed surveys (scale I : 12,000) the results of which are
used in conjunction with engineering investigations to determine the
feasibility of projects, or the extent and character of lands utilized within
a given drainage area [62].
Page [66] states that the purpose of the land-classification work of the
Bureau of Reclamation is to determine exactly where and in what quantities.
irrigable lands can be found for subdivision into tracts of sufficient size and
productive capacity to support single families and to repay irrigation costs.
The economic aspect of land classification is therefore important. The
determination of costs is mainly an engineering problem, but the determi-
nation of potential earning power after irrigation has been installed is a
problem of land classification, as earning power depends on type and yield
of crops that can be produced, besides market requirements, transport
facilities, etc. Experience in the United States has been that the stability
and economic security of an irrigation settlement depend on charges for
water being based on the quality of the land rather than entirely on the
quantity and cost of delivery of water supplied, since the same quantity
of water will give different returns on different qualities of land, and if
charged at the same rate would make one farm unit economically feasible,
and another unfeasible [62]. Hence, paying classes are distinguished,
consisting of land types each of which can be subdivided into economic
farm units of approximately equal size, ranging from 40 to 80, and not
exceeding 16o acres.
Blanch and Stewart [9] describe a scheme of classification for a region
already extensively irrigated (the Uinta Basin, Utah) with a view to
securing the best distribution of the available water and the best utilization
of the irrigable land. The classification starts from the premise that society
profits more when the available water is used on the best lands.

Two classifications are presented : (i) a present-use classification, and
(2) a desirable-use classification. The first shows the apparent need for
adjustment of water to land, based on present (1943) conditions, the second
shows what would be the most desirable combination of land and water
over an extended period of time, as indicated by analysis of all available
information. In the present-use classification three sets of factors were
taken into consideration, namely:
Soil factors : alkali, drainage, topography, slope, erosion.
Irrigation-water factor : quantity of water received, and distribution
throughout the growing season.
Location of land: with respect to community institutions and source
of water supply.
Other factors such as climate and transport facilities were taken as
uniform over the area surveyed.
The area was divided into 130 sub-areas, usually of not less than
30o acres and uniform with respect to the three factors listed above. The
sub-areas were then classified as (A) satisfactory soil and water conditions,
(B) adjustments in soil and water use desirable, (C) adjustments imperative,
(D) land mostly not irrigated, but good enough to irrigate.
In the desirable-use classification the same factors were considered as
in the present-use classification. Since water is the only factor that lends
itself to redistribution the problem was one of determining how the available
water could be most effectively used. The available supply is not adequate
for all the irrigated land, but it is adequate to irrigate the most productive
soils. Hence the sub-areas were regrouped on the assumption that each
received a full water supply. Three desirable-use classes (out of the nine
in a system applicable to all rural land in Utah) were distinguished, namely,
class II range or pasture, class V extensive arable, and class VI more
intensive arable.
Comparison of these two classifications is said to have given clear and
precise information about the most desirable adjustments to be made in the
irrigation of the region and the economic feasibility of the adjustments.
A system of classification for irrigable lands in South Africa was
described in Technical Communication No. 15 [40]. Irrigable-value "
soil maps are made on a scale of I : 6,ooo, using detailed topographical
maps, when available, as base maps. Soil profiles on representative sites
are studied and classified into four grades of permeability profile ", as
follows :-
I. An eight-foot depth of soil free from undesirable layers such as
hardpan, dense clay, coarse sand, etc.
II. At least four feet of soil of grade I resting on soil containing
interfering layers, or on semi-permeable hardpan or clay, very
open sand, etc.
III. At least four feet of soil of grade I resting on solid rock or other
impenetrable formations, or 30 inches of soil of grade I resting
on semi-permeable or gravelly layers.

IV. Shallow soils which do not conform to the above grading, such
as less than four feet of soil on impenetrable rock, or less than
30 inches of soil on semi-impenetrable or markedly interfering
The soils are also classified according to the average salt content of the
profile (so-called brak) :-
Grade I. Soils showing an average salt content (deduced from electrical-
resistance figures and checked by actual analysis of the water
extract of the soil type) of less than 0.15 per cent.
II. Average salt content from o.16 to 0.35 per cent.
III. ,, ,, ,, ,, 0.36 to 0.55 .
IV. ,, ,, ,, over 0.55 per cent.
These two factors are combined to deduce the final irrigable value
thus :-
Profile Salt Irrigability
Grade I I I
other combinations IV
The grading may further be influenced by such factors as development
of soils and agricultural experience, site of concentration and nature of brak
salts, abnormal pH values, general topography, etc., but these are usually
only applied in doubtful cases, their main use being in connexion with the
giving of advice to farmers.

The simplest land classification to understand-and simplicity is an
important virtue in the execution of a land-use plan-is one in which
classes are distinguished by numbers expressing some significant land
property, particularly productivity. In some examples already given,
land classes are distinguished as Class I, II, III and so on, but there is
nothing to indicate how much better or worse Class I is than Class II. The
numbers are merely expressions of the classifier's judgment. Many attempts
have been made to arrive at a more objective and quantitative estimate
of productivity. The methods used may be classed generally as inductive
(e.g., by adding or otherwise integrating marks awarded to certain
properties of the land or soil that influence productivity) or deductive
(e.g., deduced from yield data), but many methods in actual use are a
combination of both types.
It is important to appreciate what is meant by productivity. It is
the capacity of a soil to produce crops in general, or one specific kind of
crop, and is the resultant of the qualities of the so-called natural factors
of the environment-especially soil, climate and topography-plus human

labour or management ". The productivity of any untouched soil for
cultivated crops is nil ; it is possible to speak of the productivity of natural
grass and forest land, but even here labour must be expended before the
produce can be used. The relative importance of management in the
total productivity may be almost negligible in an open range, and pre-
dominant over that of all other factors in irrigated agriculture. As already
stated productivity is a partly economic concept, and can only partly be
expressed in terms of the inherent characteristics of land. Every time
one says that a certain piece of land is productive . there is implied in
that statement a prediction of agricultural prices and management costs
just as much as if they had been stated in dollars and cents [45].

In recent years there have been included in the county soil-survey
reports of the U.S. Department of Agriculture tables showing the pro-
ductivity ratings of all the soil types of the areas surveyed. The producti-
vity rating is the crystallized expression of the experience of the people
who have used or are using the land [i]. Ratings are given to specific
soil types as defined by the U.S. Soil Survey. In the soil-survey reports
tables are given of productivity ratings or indices in relation to all the
main crops of the district, e.g.,
Soil type Corn Wheat Rye Alfalfa Sugar Productivity
beet grade
Miami loam 70 90 70 100 80 100 70 90 60 70 3 1
The figures under A and B refer respectively to the percentage of the
standard yield of the crop obtainable with (A) common practices of manage-
ment (average farming) in the area, and (B) the best current practices. The
standard yields are selected to represent the approximate average yield
obtained for that crop on the more extensive apd widely developed soils of
the regions in the United States in which the crop is a principal product [i].
The standards refer to average yields obtained without the use of amend-
ments, although it has been arbitrarily agreed that nitrogen fixed by legumes
and manure produced frpm feed grown on the land are not to be considered
as amendments. E.g., the standard for corn is 50, and for wheat 25, bushels
per acre ; for clover and timothy it is 2, for alfalfa 4, tons. Under average
management for the county in question the Miami loam will yield 35 bushels
of corn and 17: bushels of wheat, and under first-rate management 45
bushels of corn and 25 bushels of wheat. The difference between the
(A) and (B) ratings is a measure of the soil's response to good management.
The productivity grade indicates the soil's general productivity for
the common crops of the region, and is obtained by a simple percentage
weighting of the crop ratings according to each crop's importance in local
agriculture. Soils with a weighted average between o100 and 90 are graded i,
between 90 and 80, 2, and so on. Those with low A and high B productivity

grades would be described as low in fertility (plant-nutrient content), but
productive (responsive to management). Productivity grading is admitted
to be a somewhat arbitrary procedure, depending, as it does, on the validity
of the system of weighting. From the example given above it would appear
that the Miami loam is used predominantly for cereal growing since its
productivity grade corresponds to its productivity ratings with respect to,
Formerly the inherent productivity based on the qualities of the
soil and environment, and the actual productivity based on crop yields
obtained under an average level of management, were estimated, but the
inherent productivity has now been discarded on the grounds that few soils
have any agricultural productivity at all without at least some management,
and its place has been taken by the potential productivity, which is
that obtainable by the best management. The inherent productivity has
to be obtained inductively, since there are no yield data for unmanaged "
soils, and there is no true connexion between this and the deductively
obtained actual productivity. Calculation of the ratings is based entirely
on agricultural experience as indicated by crop yields. Study of the soil
is required only to identify soil or land types. The yield figures from which
the ratings are determined are estimated from data obtained by the soil
surveyor from interviews with farmers and county officials, AAA records,
elevator records, personal observations, etc. The chief difficulty in deter-
mining ratings from yield data is to establish a standard of management
and to estimate the degree of divergence from the standard on each farm.
The usual standard of management is defined as that which will maintain
soil productivity at or near the level of the nearly virgin soil. It is
important to note that productivity ratings are not convertible to crop
yields except under certain definite conditions. The ratings are meaningless
unless a physical definition is given of the system of management assumed.
At present sufficient data are not available of yields obtained under standard
or comparable systems of management to give the ratings a fixed character.
Nevertheless, provided the soil types and phases are properly defined, the
concept of the productivity rating makes available a means for synthesizing
the great background of research work and experience in one figure of
expected yield under a defined system of management [45].
Tyler [91] maintains that productivity ratings, if they are to be used
for agricultural land classification, should be based to a greater extent than
hitherto on actual yields obtained over a period of years. A difficulty here
is that many, if not most, fields, which are the units for yield measurements,
contain more than one soil type. Another difficulty is that in areas where
the use of soil amendments is common, accurate data of yields obtained
without amendments are not usually obtainable.

An inductive method of rating the agricultural value of soils has been
suggested by Storie [84]. Regardless of actual cropping experience with

the soils Storie compiles the ratings from percentage values accorded to
certain characteristics of the soil profile. Three factors are used, one
referring to the general character (excluding texture) of the profile and
particularly to the stratification and degree of weathering, one referring
to the surface texture, and a third to soil-modifying conditions such as
drainage, acidity and alkalinity, erosion, etc. The values of these three
factors expressed as percentages of the optimal conditions for plant growth
are multiplied together to obtain the rating, and the product is expressed
as a percentage of the maximum. The advantage of multiplying instead
of, as is more usual, adding the points credited to each soil characteristic is
that thereby any one abnormal factor can dominate the final rating. Thus
a soil may have excellent profile conditions rated at 100 for the first factor,
excellent texture rated at ioo for the second factor, but bad alkali accumu-
lation rated at io for the third factor. The product of the three gives a
percentage rating for the soil of io, according with the fact that despite
other favourable conditions the alkali renders the soil unproductive. On
an addition basis the rating would have worked out at 70 per cent.

Saskatchewan. Mitchell [58], who points out that the rating system must
be adapted to the type of agriculture practised, has used in Saskatchewan
a system similar to Storie's and designed to indicate a soil's ability to
produce grain. He used three factors : (A) soil profile (texture 40, structure
30, natural fertility 30 ; maximum ioo points) ; (B) topography (maximum
Ioo, minimum 10 points, according to slope conditions) ; (C) climate 25,
salinity* 25, stoniness* 25, tendency to drift* 25; maximum ioo. The
productivity index is obtained by adding the component marks of each
factor, multiplying the total marks for the three factors, and expressing
as a percentage. It is provisionally suggested that an index of 30 is the
margin for arable land.
The index is being used in an assessment of rural municipalities as the
basis of a natural land classification. Superimposed on this is a
" practical classification which takes into account factors such as freight
costs, distance from markets, etc.
Alberta. The Alberta Soil Survey prepares land-class maps to accompany
the soil maps [Io]. The land-class map interprets the physical data of the
soil map in terms of productive capacity. Economic factors are not
considered, although it is recognized that certain factors such as price have
influenced the kind of crops grown in the past, the yield records of which are
used as a measure of probable productivity. Eight land classes are dis-
tinguished-three pasture (the poorest), one marginal arable, and four
arable (the best).
The soil type on the soil map is distinguished by a 3-digit number
referring to the soil-colour zone (reflecting rainfall and fertility), the mode
Full marks for none, and deduct proportionally.

of deposition of the parent material and the nature of the soil profile, i.e.-
First (or hundreds) digit: brown soils i ; dark brown soils 2 ; shallow, or
southern black 3 ; black 4.
Second (or tens) digit: residual o ; sorted residual i ; glacial, unsorted and
variable 2 ; resorted or uniform glacial 3 ; gravelly outwash 4 ; alluvial 5;
alluvial or aeolian 6 ; aeolian or lacustrine 7.
Third (or units) digit : 0-7, distinguished according to profile characteristics,
among which salinity is specially noted.
To determine the land classes these digits are given percentage ratings,
and other ratings are given for texture, topography, erosion, stoniness,
alkali, and any other local factor affecting productivity. The ratings are
multiplied together and expressed as a percentage of the total possible,
Characteristic Description Rating
Colour brown to dark brown 50
Deposition glacial 80
Profile normal 1oo
Texture light loam 70
Alkali absent Ioo
Stones numerous 60
Topography very hilly 40
Multiplied together, these give a percentage rating of 6.7, placing the
land in Class 2 (Class 8 is best)-fair to good pasture. The method
represents an effort to meet present-day requirements by taking an
inventory of soil resources and interpreting the data in terms of a probable
productivity classification. This information should be of material aid in
any programme of land-use planning."

Kendall [47] proposed a rather different method of rating the relative
productivities of English counties. He took the county as land unit, since
he was using the official yield figures of the Ministry of Agriculture, and
these are compiled on a county basis. The county has obvious disadvantages
as a unit since its boundaries have no relationship to those of geographic-
ally, economically or agriculturally determined areas. Kendall developed
four different measures of productivity, using available statistics of the
ten major crops-wheat, barley, oats, beans, peas, potatoes, turnips and
swedes, mangolds, hay from temporary grass and hay from permanent
(i) The productivity coefficient. This coefficient was obtained by regarding
the average yields as the coordinates of a point in io-dimensional space.
From the counties of England 48 points were obtained, and the line which
fitted these points best (the sum of squares of the perpendicular distances
of the 48 points from the line being a minimum) was usedi to rank the

counties. This coefficient involves very intricate calculations, and does.
not seem to have any advantage over the other three more simply deter-
mined coefficients.
(2) The ranking coefficient. Each county was ranked for the average yield
of each of the ten crops, and the coefficient of each county was the mean
of these rankings. Counties with high yields had low ranking coefficients,
and vice versa.
(3) The money-value coefficient. Obtained from the average value produced
per acre from the ten crops.
(4) The energy coefficient. The average gross digestible energy produced
per acre.
Neither the productivity coefficient nor the ranking coefficient takes
any account of the area devoted to each crop in the counties. As there
are obviously large differences between counties in this respect, the use of
equal weights for all crops gives an unfair comparison between counties,
particularly as the correlations of the yields of grass and fodder crops with
those of cereal crops are often negative. The money-valued coefficient
has some advantages over the two previous ones-for example, account is
taken of the volume of production. One disadvantage is that certain parts
of the country are better suited for the more profitable crops-because of
proximity to markets-and the coefficient will tend to measure this factor.
Kendall's last coefficient-the energy coefficient-has more to recommend
it, but it is unfortunate that he chose to work in terms of gross digestible
energy. This has the effect of favouring the wheat-growing counties, since
about half of the energy value attributed to wheat is due to straw of which
most of the nutritive value is available for fuel purposes only. Moreover,
pasture which is hayed is only assessed with respect to the hay produced,
and as hay is of the order of half the total output for the year counties
with a large amount of hay have low coefficients. This is probably the
reason for the differences in the grading of Somersetshire, for example,
which was given a high grade by the productivity and rank methods, but
a low grade by the others.
On the basis of the four coefficients for each of four years, Kendall
classified the counties as excellent, good, moderate, indifferent or poor.
This was done for each of the four years 1925, 1930, 1935 and 1936. On the
whole, counties with good reputations for productivity came out con-
sistently high in the ranking list, and those with poor reputations consistently
low. Intermediate counties fluctuated rather widely from year to year, and
from coefficient to coefficient. Kendall was estimating actual rather than
potential productivity [24]. Actual productivity will obviously vary from
year to year according to weather, price level and other external influences.

GERMAN SOIL RATINGs-Bodenbonitierung
Much work has been carried out in Germany on the quantitative
assessment of soil quality. After the first world war a general reassessment

of taxable land values was necessary and had to be carried through rapidly.
Several schemes for rating agricultural land were proposed, a common
feature of all being that productivity deriving from skill in management
was not taken into consideration. It was thought that a man should not
be penalized by higher taxes for improving his land. Assessments were
based on the yield capacity determined by the natural conditions of soil,
climate, topography and water, and the local conditions of farm economy,
especially market and labour conditions, as well as the nature of the holding
(compact, scattered, distance of farmhouse from the fields, etc.). These
were almost the identical bases on which A. von Thaer, the father of German
Bodenbonitierung, had constructed his first land-valuation tables in the
early years of the nineteenth century [90o].
The general procedure in Germany was to classify the land of an
estate firstly as arable, pasture, meadow or forest land, and then make
further separate classifications inside each class. The classifications are
based entirely on the properties of the soils. The influence of economic and
commercial factors and of factors such as size, lay-out and nature of the
farm is assessed quite separately, and then superimposed on the soil classifi-
cation, usually by allotting a certain percentage of possible points to
internal or soil factors and a certain percentage to external factors. Assess-
ment of arable land is made mainly inductively, from the properties of the
soil, and of grass and forest land deductively, from yield data.
In order to secure a uniform standard of rating, a national standard
estate (Reichspitzenbetrieb), that of Frau Else Haberhauffe, Eickendorf,
near Magdeburg, was given a rating of Ioo. Local standard estates in
different parts of Germany were rated relatively to the national standard,
and in this way a series of suitably located standards was set up to which
all ratings could be referred [74].
Schattenfroh [74] states that the allotment of points for the
separate conditions must by its very nature be so subjective that the
points can only be regarded as a rough guide. Only experience can teach
the assessor his job. The natural and economic factors are quite
independent. Relative prices or rentals of land give good indications of the
effect on productivity of variable natural conditions under similar economic
conditions, or of variable economic under similar natural conditions. The
usefulness of a system of land assessment based on awarding points for
various factors in productivity depends on the correct judgment of the
relative points values to be allotted to the totality of natural and economic
factors, respectively.
According to Schattenfroh the points value of the economic factors
should ordinarily be about 60 per cent of that of the natural factors for
flat agricultural land. This, presumably, refers specifically to German
conditions in the inter-war period. Near large cities and industrial centres
the relative value of economic factors rises and may exceed that of natural
factors. In the open market, however, economic and particularly commer-

.cial factors exert a much greater influence on the price actually paid for
land than do the intrinsic qualities of the soil.
Several points systems are cited by Schattenfroh. In Fackler's Bavarian
system [20] points are awarded as follows :-
For soil conditions up to 30 points
,, climate ,, ,, 20 ,,
,, other natural factors ,, ,, 60o ,,
,, economic factors ,, ,, 10 ,,

Total 120

Soil condition is determined mainly by texture and humus content
,(highest is humous loess-Schwarzerde) with marks added or subtracted
according to the depth of the arable horizon, the nature of the subsoil and
the general cultural condition. Economic factors appear to be confined to
the distance of the farm from the nearest railway station or town.
B6mer [74] allots points for Westphalian farms as follows :-
Nature of arable layer 20
Depth of arable layer I1
Groundwater conditions 10
Surface topography 8
Total for soil factors 48
Climate (temperature and rainfall) 18 18
Marketing conditions 17
Type and state of culture 7
Distance of fields from farmstead 5
Labour conditions 5
Total for economic factors 34


Thus whereas Fackler estimates soil influences at 25 per cent and
economic influences at less than 8 per cent, B6mer estimates the former at
nearly 50 per cent and the latter at 34 per cent.
Rothkegel and Herzog [71] worked out a system which was accepted
in 1928 by the German Treasury for the valuation of agricultural
estates. Different scales of valuation are used for arable and grassland.
For arable land, eight soil groups are distinguished with points values
varying from 100 (loamy chernozem) to 6 (coarse sand, pure clay, high
moor). Pastures are marked directly according to the length of the grazing
season and the carrying capacity.
For meadows the scale is based on the quantity and quality of the
natural meadow plants, and hence on the quantity and quality of the
harvest. Points values vary from 100 (corresponding to an attainable
harvest of 75 double centners of good hay per hectare) to 12 (corresponding
to 20 dz./ha. of poor but eatable hay).

Besides the soil characteristics in arable land, and the yield and ecolo-
gical characteristics in grassland, an apparently rather arbitrary and
subjective estimation is made in Rothkegel and Herzog's system of the
actual and potential effects on productivity of climate, exposition, possi-
bilities of drainage and irrigation, and cultural condition of the soil. In the
assessment of economic factors special attention is paid to means of transport
and communication.
Criticizing this system Schattenfroh states that it represents an attempt
to establish a scientific basis for tax assessment, but that it is much too
complicated for ordinary practice, and could not be operated except by
highly trained assessors. The range of marking for soil quality, from ioo to 6,
is, moreover, too wide and leaves too much to the assessor's personal
judgment. In group I (mild loams) the range is 22 points, and in group VI
(heavy clays) it is 50 points. Similar ranges occur in the grouping of
pastures and meadows. Consideration of the subsoil, also, is left largely
to the assessor's discretion.
Schattenfroh himself devised a scheme (modified from an earlier one
of Weisslein) for Bavaria, in which soils were first grouped into four classes
according to texture-heavy soils (marked 0-8), loams (marked 9-12),
light soils (marked 1-8) and humus soils (marked o-io). The depth of arable
soil is ignored, because this depends more on the means of cultivation at
the disposal of the farmer than on inherent soil properties, and few farmers
in Bavaria could afford to cultivate deeply and thus to take advantage of a
potentially deep soil. The humus content is evaluated as a percentage
addition to the points mark, e.g., 10-20 per cent addition for a 5 per cent
humus content, and 30-50 per cent addition for 15-20 per cent of humus.
Acid soils get a percentage deduction, and calcareous soils a percentage
addition. The deduction for acidity is greater in regions devoted pre-
dominantly to acid-sensitive crops (wheat, barley, sugar beet, clover) than
in rye, oats and meadow regions. A percentage deduction for stoniness is
made according to the quantity of stones and the soil texture. Other
additions or deductions are made for climate (according to quantity and
distribution of rainfall and the length of the local compared with the
regional frost-free period), for compactness of the holding and distance
from a market. Extensive tests of the system in practice, however, indicated
that the relative importance and quantitative assessment of humus content,
acidity, lime status, stoniness, groundwater relationships, etc., varied
greatly from place to place.
The number of other German points systems for soil evaluation that
have been proposed is legion [33]. Van Aartsen [93] has described several
not mentioned here, and has compared the German systems as a whole with
the American productivity ratings.
G6rz [25] states that there is now no dispute that the soil must be the
starting point of any modern system of agricultural land evaluation, but
that some authorities favour the Bodentyp (genetic soil type) whilst others
favour the Bodenart (textural soil variety) as the basis of the classification.

From the important practical standpoint of getting the man-in-the-field to
understand what is being done, there is much to be said for using the soil
variety, and all the numerous German systems for the points evaluation
of land lay particular weight on soil texture. G6rz, however, points out
that a classification based solely on soil-variety distinctions is only valid so
long as a certain extensity of utilization allows differences in value between
different soil varieties to be clear. As land utilization becomes more
intensive these differences tend to lose their significance, while that of
soil-type differences increases relatively. A practical soil evaluation must
take into consideration and combine both practical experience and the
scientific premises of the valuation system in a form which is immediately
comprehensible to the practical man. For the same reason, information
about land quality obtained from practical experience and the "feel" of the
soil (Fingerspitzgefiihl) should be expressed in a simple quantitative form.
These practical requirements are best satisfied by means of a points
system. According to G6rz, points are allotted in the standard Bonitierungs-
system for texture (varying from loam io.o-8.8 to sand 4.1-2.0) and for soil
condition or type, marked from 10 (mild loess) to 4 (podzol with ortstein).
The rating is obtained by multiplying the texture marks by the type marks.
Thus at one end of the scale a loess loam gets ioo points, and at the other
end a sandy heath podzol gets 8 points.
Points values for soil types and varieties are obtained from the gross
and net yields obtained from large numbers of different soils accurately
described and classified by experienced assessors.
Herzog [31] gives more factual details about how the official system of
soil assessment had developed by 1939. In 1939 there were 2,000 soil
assessors in the Reich besides numerous other workers indirectly engaged.
The purpose of the assessment was to form a basis for taxation, but as the
work proceeded the possibilities of using it for advisory purposes and for
land-use planning came increasingly into the foreground. The intention
was that every quarter hectare should be sampled (and even smaller areas
where the soils were very variable), the soil conditions described and a map
The assessment is arrived at by giving a soil marks according to certain
characteristics, but the marks appear to be given for the combined character-
istics as a whole, not for each one separately. Different characteristics are
used for arable land and grassland, respectively.
Arable soils are first classified into 8 textural varieties (Bodenarten:
S=sand, Sl=slightly loamy sand, IS=loamy sand, SL=-very loamy sand,
sL=sandy loam, L==loam, LT=heavy loam, T==clay), and moor (Mo).
The second differentiating characteristic is the origin : L6= loess, D= dilu-
vium, Al=alluvium, V=non-transported weathered material. The suffix g
attached to V (Vg) indicates very coarse or stony material. A third differen-
tiation is made according to the condition (Zustandsstufe) or stage of develop-
ment of the soil. The most immature soils represent stage 7, fully mature
soils stage I, at which the productivity of the soil is, other things being

equal, greatest. Many soils never reach stage i, owing to peculiarities in
local conditions, such as topography, rainfall and groundwater relationships,
others revert by processes of degradation from stage i to less productive
stages. It is not clear how the stage of development is determined.
For grassland soils only four Bodenarten* (S, IS, L, T-also Mo) are
distinguished and three Zustandsstufen. The origin of the soil is not taken
into consideration, but marks are given according to mean annual tempera-
ture (three classes a, b and c) indicating the length of the growing season,
and moisture relationships (five classes).
The description and rating of a soil profile are given in symbols, e.g.-
h-h mi fs L 2.5-3 L 2 L6 83
h' mi fs L 3
ka fs L
According to the symbols on the right, this is a loam derived from loess,
Zustandsstufe 2, rating 83 (relative to the best agricultural soil in Germany,
rating ioo). The symbols on the left describe the horizons of the profile in
more detail. The top horizon is a medium to highly humous (h-h) mild
(mi) fine sandy (fs) loam (L), underlain by a weakly humous (h') mild fine
sandy loam, underlain again by a strongly calcareous (ka) fine sandy loam.
No indication is given in Herzog's paper as to the meaning of the figures
following the symbol L in the profile description ; they do not seem to refer
to the same Zustandsstufe indicated in the symbols on the right.
A profile description of a grassland soil is as follows :-
zer gt Hmo 1-2 Mo II a 3-31
r u. to Hmo
i.e., a well (gt) decomposed (zer) highmoor (Hmo) overlying a raw (r) and
(u.) peaty (to) highmoor. For rating purposes the soil is a moor (Mo),
Zustandsstufe II, climate a (= mean annual temperature 8'C. or over)
moisture condition 3, rating 31.
The profile descriptions can be very varied. There are 46 symbols
referring to soil characteristics, 9 to water relationships, 18 to types of
cultivation and other peculiarities, and about a dozen "other abbrevia-
tions". The ratings are calculated from Tables 9 and 10.
The ratings have no absolute significance, being merely relative numbers
expressing differences in gross yields which can be obtained under ordinary,
average management. In estimating the ratings, moderate climatic
conditions-8C. mean annual temperature and 600 mm. rainfall-level or
gently sloping topography, and market conditions corresponding to those
in central Saxony are assumed.
It has been proposed, however, to use the ratings to put a cash value
on soil for taxation purposes. It is argued [70] that the productivity of land
is the result of a combination of the factors soil, capital and labour, but it
is impossible to determine what part of the yield of a field is attributable
Gericke [23] showed from a statistical analysis of more than 2,000 experiments that soil texture
had practically no effect on hay yields.

to soil productivity, and what parts to machinery, buildings, stock, labour,
and so on. Although the value of the bare soil of a field or of a soil type
cannot therefore be estimated from the results of farming, that of the soil
of a complete farm can be. The soil of the farm is given a percentage
productivity rating relative to the best soil (100 per cent) in Germany.
The land is capitalized on a basis of 4 per cent of the net profit, which is
210 RM. per ha. for soil with rating 100-i.e., 5250 RM. per ha. (in 1944).*
Land rated at 20 would thus have a capital value of I05o RM. The capital
value of the buildings and the living and dead equipment of the farm is
calculated and divided by the number of hectares in the farm. This is
subtracted from the capital value of the land per hectare determined from
the net profit (Reinertrag), giving the value of the soil. E.g., land rated
at 20 might have a value of 490 RM. per ha. for buildings, 300 RM. for
stock, and 140 RM. for other equipment, total 930 RM., leaving 120 RM. for
the value of the soil. Land rated at 100 might have a value of 1450 RM.
per ha. for buildings, 500 RM. for stock, and 300 RM. for other equipment,
leaving 3000 RM. for the value of the soil (total 5250 RM.). It is claimed
that this method of soil assessment provides a sound basis for calculating
land taxation.

In 1929 the capital value of 100-rating land was 4140 RM. per ha. [74].

TABLE 9. Arable-land rating (German system)

Soil OStage of development
variety Origin
1 2 3 4 5 6 7

S D 41-34 33-27 26-21 20-16 15-12 11- 7
Al 44-37 36-30 29-24 23-19 18-14 13- 9

Sl D 51-43 42-35 34-28 27-22 21-17 16-11
(S/lS) Al 53-46 45-38 37-31 30-24 23-19 18-13
V 42-36 35-29 28-23 22-18 17-12

D 59-51 50-44 43-37 36-30 29-23 22-16
L6 62-54 53-46 45-39 38-32 31-25 24-18
iS Al 62-54 53-46 45-39 38-32 31-25 24-18
V 50-44 43-37 36-30 29-24 23-17

D 67-60 59-52 51-45 44-38 37-31 30-23
SL L6 72-64 63-55 54-47 46-40 39-33 32-25
(IS/sL) Al 71-63 62-55 54-47 46-40 39-33 32-25
V 67-60 59-52 51-44 43-37 36-30 29-22
Vg 47-40 39-32 31-24 23-16

D 84-76 75-68 67-60 59-53 52-46 45-39 38-30
L6 92-83 82-74 73-65 64-56 55-48 47-41 40-32
sL Al 90-81 80-72 71-64 63-56 55-48 47-41 40-32
V 76-68 67-59 58-51 50-44 43-36 35-27
Vg 54-45 44-36 35-27 26-18

D 90-82 81-74 73-66 65-58 57-50 49-43 42-34
L6 100-92 91-83 82-74 73-65 64-56 55-46 45-36
L Al 100-90 89-80 79-71 70-62 61-54 53-45 44-35
V 82-74 73-65 64-56 55-47 46-39 38-30
Vg 60-51 50-41 40-30 29-19

D 78-70 69-62 61-54 53-46 45-38 37-28
LT Al 82-74 73-65 64-57 56-49 48-40 39-29
V 78-70 69-61 60-52 51-43 42-34 33-24
Vg 57-48 47-38 37-28 27-17

D 63-56 55-48 47-40 39-30 29-18
T Al 74-66 65-58 57-50 49-41 40-31 30-18
V 62-54 53-45 44-36 35-26 25-14
Vg 50-42 41-33 32-24 23-14

Mo 45-37 36-29 28-22 21-16 15-10

Grassland rating (German system)

Soil Moisture conditions
Stage Climate 2
variety 1 2 3 4 5








_ I




















I I: ________________________________________



11- 7


12- 7

Climate a = over 8.00C. mean annual temperature
b = 7.9-7.00C. .
c = 6.9-5.70C. OP






1. ABLEITER, J. K. Productivity ratings of soil types. Missouri Agric. Expt. Sta. Bull-
421, 1940 (13-24).
2. ALDOUS, A. E. AND SHANTZ, H. L. Types of vegetation in the semi-arid portion of the
United States and their economic significance. J. Agric. Res. 28, 1924 (98-128).
3. ALLIN, B. W. Soil scientists have a planning job. Proc. Soil Sci. Soc. Amer. (1941), 6,
1942 (3-7).
4. ASTOR, VISCOUNT, AND ROWNTREE, B. SEEBOHM. Mixed farming and muddled thinking.
MacDonald, London (n.d.), pp. 143.
5. BAKER, 0. E. Agricultural regions of North America. Econ. Geog. 2, 1926 (459-493).
6. BARNES, C. P. AND VOGEL, H. A. Land classification for general land use planning.
Missouri Agric. Expt. Sta. Bull. 421, 1940 (283-288).
7. BAUSMAN, R. 0. Merits and limitations of some of the land classes. J. Farm. Econ. 23,
1941 (899-903).
8. BENNETT, H. H. There's but one way to do the job. Soil Conservation 11, 1945
9. BLANCH, G. T. AND STEWART, C. E. Utilization of irrigable land in the reservation area
of Uinta Basin, Utah. Utah Agric. Expt. Sta. Bull. 303, 1943, pp. 53.
10. BOWSER, W. E. Soil surveys in relation to land classification in Alberta. Sci. Agric. 20,
1940 (285-290).
11. CAJANDER, A. K. The theory of forest types. Acta Forest. Fenn. 29, 1926, pp. 108.
12. CAMP, J. R. The classification of lands for forest management. Missouri Agric. Expt.
Sta. Bull. 421, 1940 (203-211).
13. CLEMENTS, F. E. Experimental ecology in the public service. Ecology 16, 1935 (342-363).
14. COILE, T. S. Forest classification ; classification of forest sites with special reference to
ground vegetation. J. Forestry 36, 1938 (1062-1066).
15. COKE, J. Land classification in Western Canada. Sci. Agric. 21, 1941 (378-387). '
16. COLBY, C. C. Changing currents of geographic thought in America. Ann. Assoc. Amer.
Geogrs. 26, 1936 (1-37).
17. CROWTHER, E. M. Fertilizers during and after the war. Bath and West Soc. Pamph. 13,
1945, pp. 51.
18. CUMBERLAND, K. B. The survey and classification in New Zealand ; a basis for planning.
Trans. Roy. Soc. N.Z. 74, 1944 (185-195).
19. EISENHOWER, M. S. Federal responsibilities in total conservation. In Conservation of
Renewable Natural Resources, pp. 175-189. Univ. Pennsylvania Press, 1941.
20. FACKLER. Wochenbl. Landw. Ver. Bayern 114, No. 41, 1928. (Cited by [74].)
21. GALBRAITH, A. V. In Forests in relation to climate, water conservation and erosion.
S. Afric. Dept. Agric. Bull. 159, 1935, pp. 62.
22. GERASIMOV, I. P., ZAVALISHIN, A. A. AND IVANOVA, E. N. [A new scheme for a general
soil classification for the U.S.S.R.] Pedology No. 7, 1939 (10-43).
23. GERICKE, S. Die Wirkung verschiedener Wachstumfaktoren auf die Ertragsleistung
deutscher Wiesen. Bodenk. PflErnahr. 34, 1944 (213-239).
24. GOODSON, J. B. The appraisal of agricultural productivity. Thesis, London, 1939,
pp. 188. (Typescript.)
25. G6RZ, G. Wissenschaft und Praxis in der Bodenbonitierung. Trans. Third Internat.
Cong. Soil Sci. 1, 1935 (276-279).
26. GRANGE, L. I. A basic scheme for land classification. N.Z. J. Sci. Tech. 26A, 1944
27. GRAY, L. C. AND REGAN, M. Needed points of development and reorientation in land
economic theory. J. Farm. Econ. 22, 1940 (34-46).
28. GRoss, N. C. A post mortem of country planning. J. Farm. Econ. 25, 1943 (644-661).
29. HAGGERTY, J. J. AND MYERS, A. M. Materials and techniques of modem land classifi-
cation. Missouri Agric. Expt. Sta. Bull. 421, 1940 (100-122).
30. HERRMANN, R. Die Ergebnisse der Bodenuntersuchungen. ForschDienst. 16, 1943

31. HERZOG, F. Boden und Bodenbeschaffenheit. In Geffige und Ordnung der deutschen
Landwirtschaft, pp. 87-103. Berlin, 1939.
32. HESSELMAN, H. Studier 6ver barrskogens humustacke, dess egenskaper och beroende
av skogsvArden. Medd. Statens Skogsf6rsjksanst. 22, 1925 (169-552).
33. HEUSER, 0. Der Kulturboden, seine Charakteristik und seine Einteilung vom landwirt-
schaftlichen Gesichtspunkt. In Blanck's Handbuch der Bodenlehre 8, 1931 (1-48).
34. HILGARD, E. W. Soils. Macmillan, New York, 1914, pp. 593.
35. HILLS, G. A. An approach to land settlement problems in Northern Ontario. Sci. Agric.
23, 1942 (212-216).
36. HOCKENSMITH, R. D. AND STEELE, J. G. Classifying land for conservation farming.
U.S.D.A. Farm. Bull. 1853, 1943, pp. 45.
37. HOLLSTEIN, W. Eine Bonitierung der Erde auf landwirtschaftlicher und bodenkund-
licher Grundlage. Petermanns Mitt. Ergiinzungsh. 234, 1937, pp. 49.
38. HUDSON, G. D. The unit area method of land classification. Ann. Assoc. Amer. Geogrs. .
26, 1936 (99-112).
39. HUDSON, G. D. The theory of land classification. 2. The contribution of geography to --
land classification. Missouri Agric. Expt. Sta. Bull. 421, 1940 (174-183).
40. IMPERIAL BUREAU OF SOIL SCIENCE. Soil survey for irrigation purposes in South Africa.
Tech. Commun. 15, 1930, pp. 11. (Mimeo.)
41. IMPERIAL BUREAU OF SOIL SCIENCE. Tea Soils. By H. H. Mann. Tech. Commun. 32,
1935, pp. 66.
42. JONASSON, 0. Agricultural regions of Europe. Econ. Geog. 2, 1926 (19-48).
43. KELLOGG, C. E. A system of land classification. Trans. Third Internat. Cong. Soil Sci. 1, .
1935 (283-286).
44. KELLOGG, C. E. Soil survey manual. U.S.D.A. Misc. Pub. 274, 1937, pp. 135.
45. KELLOGG, C. E. The theory of land classification. I. The contributions of soil science -
and agronomy to rural land classification. Missouri Agric. Expt. Sta. Bull. 421, 1940
46. KELLOGG, C. E. AND ABLEITER, J. K. A method of rural land classification. U.S.D.A.
Tech. Bull. 469, 1935, pp. 30.
47. KENDALL, M. G. The geographical distribution of crop productivity in England. J. Roy.
Stat. Soc. 102, 1939 (21-62).
48. KRUSEKOPF, H. H. Land classification in relation to the soil and its development.
I. Physical aspects. Missouri Agric. Expt. Sta. Bull. 421, 1940 (39-44).
49. LILIENTHAL, D. E. T.V.A. Penguin Books, Harmondsworth, 1944, pp. 208.
50. LOVEJOY, P. S. Theory and practice of land classification. J. Land Pub. Util. Econ. 1,
1925 (160-175).
51. MAGNITSKY, N. K. [Agropedological division into regions as a method of planned study
and utilization of the soil.] Pedology No. 5, 1941 (98-100).
52. MARBUT, C. F. Russia and the United States in the world's wheat market. Geog. Rev. 21,
1931 (1-21).
53. MARBUT, C. F. Land classification. "Life and Work of C. F. Marbut," pp. 215-227.
Soil Sci. Soc. Amer. (n.d.), pp. 271.
54. MEVIUS, W. Die Bestimmung des Fruchtbarkeitszustandes des Bodens auf Grund des
natiirlichen Pflanzenbestandes. In Blanck's Handbuch der Bodenlehre 8, 1931
55. MILNE, G. Composite units for the mapping of complex soil associations. Trans.
Third Internal. Cong. Soil Sci. 1, 1935 (345-347).
56. MILNE, G. A report on a journey to parts of the West Indies and the United States for
the study of soils, February to August, 1938. E. Afric. Agric. Res. Sta. Amani 1940,
pp. 78.
57. MINISTRY OF WORKS AND PLANNING, LONDON. Report of the committee on land utili-
sation in rural areas. Cmd. 6378, 1942, pp. 138.
58. MITCHELL, J. A method for obtaining a comparative rating of Saskatchewan soils.
Sci. Agric. 20, 1940 (281-284).

59. MOHR, E. C. J. The relation between soil and population density in the Netherlands
Indies. C. R. Cong. Internal. Giog. Amsterdam 2, Sect. IIIc, 1938 (478-493).
60. MOON, J. W. The soil type as a unit for land classification in the Tennessee Valley area.
Proc. Soil Sci. Soc. Amer. (1937) 2, 1938 (489-493).
61. MfCKENHAUSER, E. Auswertung von Bodenkarten ffir Planungszwecke. C. R. Cong.
Internal. Geog. Amsterdam 2, Sect. V, 1938 (87-89).
62. NATIONAL RESOURCES PLANNING BOARD. Report on land classification in the United ,
States. Washington, 1941, pp. 151.
63. NEW ZEALAND. 18th Rept. Dept. Sci. Indust. Res. 1944, pp. 57.
64. NORTON, E. A. Soil conservation survey handbook. U.S. Dept. Agric. Misc. Pub. 352,
1939, pp. 40.
65. OSMOND, D. A. An index for use in the regional classification of land for agricultural
purposes. Long Ashton Agric. Hort. Res. Sta. Ann. Rept. (1944), 1945 (219-224).
66. PAGE, J. C. Land classification to determine the feasibility of reclamation projects.
Missouri Agric. Expt. Sta. Bull. 421, 1940 (247-255).
67. PENTZ, J. A. The value of botanical survey and the mapping of vegetation as applied to
farming systems in South Africa. Bot. Surv. S. Africa Mem. 19, 1938, pp. 15.
68. PENTZ, J. A. An agro-ecological survey of Natal. Proc. Eighteenth Ann. Cong. S. Afric.
Sug. Tech. Assoc. 1944 (65-68).
69. PETERSEN, A. Taxation von Wiesenlindereien. Berlin, 1927. (Cited by [54].)
70. ROTHKEGEL, W. Die Auswertung der Ergebnisse der Bodenschitzung. ForschDienst.
17, 1944 (71-84).
71. ROTHKEGEL, W. AND HERZOG, H. Das Verfahren der Reichsfinanzverwaltung bei der
Bewertung landwirtschaftlicher Betriebe. Ber. iiber Landw. 10, 1928, Sonderh.
72. ST. CLAIR THOMPSON. Forest conditions in the Gold Coast. Imp. For. Inst. Pap. 1,
1936, pp. 10. (Mimeo.)
73. SAMPSON, A. W. Plant indicators-concept and status. Biol. Rev. 5, 1939 (155-206).
74. SCHATTENFROH. In Praktische Bodenuntersuchung und Bodenbeurteilung. By A. von
Nostitz. Paul Parey, Berlin, 1929, pp. 156.
75. SCHOENMANN, L. R. Essential elements in land classification. Proc. Soil Sci. Soc. /
Amer. (1936) 1, 1937 (453-454).
76. SHANTZ, H. L. Natural vegetation as an indicator of the capabilities of land for crop
production in the Great Plains area. U.S.D.A. Bur. Pl. Indust. Bull. 201, 1911,
pp. 100.
77. SMITS, M. B. Population density and soil utilization in the Netherlands Indies. C. R.
Cong. Internal. Gdog. Amsterdam 2, Sect. IIIc, 1938 (500-506).
78. Soil Conservation 11, 1945, p. 126.
79. SPENCE, C. C. AND HOPE, E. C. An economic classification of land in fifty-six municipal
divisions, south central Saskatchewan. Canada Dept. Agric. Pub. 728 (Tech. Bull. 36),
1941, pp. 42.
80. STAMP, L. D. The land classification scheme of the land utilisation survey of Britain. &'
Trans. Third Internal. Cong. Soil Sci. 3, 1936 (138-140).
81. STAMP, L. D. The land of Britain and its use. In the press.
82. STAMP, L. D. A regional survey of European agriculture. Advance. Sci. 2,1942 (156-158).
83. STEWART, A. AND PORTER, W. D. Land use classification in the special areas of Alberta.
Canada Dept. Agric. Pub. 731 (Tech. Bull. 39), 1942, pp. 73.
84. STORIE, R. E. An index for rating the agricultural value of soils. Calif. Agric. Expt.
Sta. Bull. 556, 1933, pp. 44.
85. STREMME, H. Die B6den Deutschlands und der freien Stadt Danzig. Petermanns Mitt.
Ergdnzungsh. 226, 1936, pp. 74.
86. STREMME, H. International soil map of Europe. Gea Verlag, Berlin, 1937.
87. STREMME, H. AND OSTENDORFF, E. Petermanns Mitt. Ergdnzungsh. 228, 1937.
88. TAMM, 0. Markstudier i det nordsvenska barrskogsomrAdet. Medd. Statens Skogsf6rsdks-
anst. 17, 1920 (49-300).

89. TAYLOR, M. W. Indicators of southwestern range conditions. U.S.D.A. Farm. Bull.
1782, 1937, pp. 35.
90. THAER, A. VON. Versuch einer Ausmittlung des Reinertrages der produktiven
Grundstiicke. Berlin, 1813.
91. TYLER, H. S. Materials and techniques of modern land classification. Discussion. V
Missouri Agric. Expt. Sta. Bull. 421, 1940 (129-134).
1943, pp. 40.
93. VAN AARTSEN, J. P. Land classification in relation to its agricultural value : a review
of systems applied. Mo. Crop Rept. Agric. Stat. 35, 1945 (79S-166S).
94. VEATCH, J. 0. Graphic and quantitative comparisons of land types. J. Amer. Soc.
Agron. 27, 1935 (505-510).
95. VEATCH, J. 0. The idea of the natural land type. Proc. Soil Sci. Soc. Amer. (1937) 2,
1938 (499-503).
96. VEATCH, J. O. Agricultural land classification and land types of Michigan. Mich. Agric.
Expt. Sta. Spec. Bull. 231, 1941, pp. 67. (First revision.)
97. VEATCH, J. 0. Land classification for general land use planning. Discussion. Missouri v
Agric. Expt. Sta. Bull. 421, 1940 (289-292).
98. WHITTLESEY, D. Major agricultural regions of the Earth. Ann. Assoc. Amer. Geogrs. 26,
1936 (199-240).
99. YOUNGS, F. 0. Factors influencing irrigation development and classification of land
for irrigation. Proc. Soil Sci. Soc. Amer. (1937) 2, 1938 (505-507).


(Reproduced from [44] )
In order to achieve some uniformity in the treatment and arrangement
of subject matter in the report, especially for the convenience of those
using large numbers of these reports, the outline which follows has been
prepared. This outline cannot be followed blindly, as each area has its own
particular features and problems requiring emphasis, and certain sections,
given only broad headings, must be carefully organized in subdivisions.

i. Description of the area surveyed.
(A) Location and extent of the area.
(1) General location within the State or Territory.
(2) Distance of county seat or principal town from one
or more important places.
(3) Size of area in square miles.
(B) Physiography.
(1) Mention of the physiographic division of the United
States in which the area is located.
(2) General description of the physiography and geology
of the area.
(C) Relief.
(1) Discussion of any modification of the physiographic
surface by natural dissection.
(2) Sketch map of relief areas if such a sketch is of
material assistance to the discussion.
(D) Elevation.
(1) General elevation of the area and ranges in elevation.
(2) Altitude of some of the main topographic features.
(3) Altitude of towns or other known points. (Cite
authority for data.)
(E) Vegetation.
(i) General but brief discussion of the vegetation,
including especially the original and present forests,
grasses, or shrubs. (Ordinarily there should be no
detailed description of species unless this is impor-
tant where many comparatively uncommon plants
are mentioned. The common names should be
used, but a glossary of botanical names should be
given, especially where noncrop plants need to be
discussed at length.)

(F) Organization and population of the county.
(1) Date of settlement.
(2) Important historical data.
(3) Source and distribution of the population.
(4) Nationality of the settlers. (To be supported by
census data.)
(G) County seat and principal towns.
(1) Name of county seat.
(2) Principal towns and their relation to the agriculture
of the area.
(H) Transportation, markets, and other cultural features.
(1) Railroads and highways and the service rendered by
(2) Any service rendered by steamship lines.
(3) The disposition and marketing of farm products.
(4) Conditions of public roads.
(5) Schools, churches, telephones, and other features
relating to rural culture and social life.
(I) Industries.
(i) Important nonagricultural manufacturing industries,
mining, and other industries affecting the area
should be mentioned and any relationship these
have to the agriculture of the area noted. (The
plants engaged in processing agricultural products,
such as flour mills and cheese factories, are dis-
cussed under H (3).)
2. Climate.
(1) General type, that is, oceanic, continental, etc.
(2) Variations among seasons.
(3) Distribution of rainfall during the growing season.
(4) Influence of physiography and bodies of water on climate in
different parts of the area.
(5) Climate as a factor in the production of special crops.
(6) Discussion on value of the data on average dates of frost and
average frost-free periods.
(7) Discussion on value of data on average annual means for
(8) Conditions, favorable or unfavorable, for farm work.
(9) Tables from Weather Bureau of temperature, precipitation,
and frosts.
(io) Discussion of unusual weather conditions, such as winds,
storms, and hail.

3. Agricultural history and statistics.
(A) Early agricultural development of the area.
(B) Important changes that have taken place in the use of the
land since settlement.
(C) Census data on the agriculture of the area as far back as
available and up to and including latest census reports.
(D) Present condition of agriculture.
(1) Use of fertilizers, lime, and other amendments.
(a) Amount and cost.
(b) Kind of fertilizers used.
(2) Labor.
(a) Kind and availability of labor. (Do not discuss
the specific wages paid in different farm
operations as conditions change from year
to year.)
(3) Size of farms.
(a) General range in the size of farms and the size
of the average farm. (To be checked by
census data.)
(b) Any trends toward change in size of farms and
the reasons for such change.
(4) Tenure of farms.
(a) Percentage of farms operated by the owners
and by tenants.
(b) Systems of rental, such as cash rent, share of
crop, or other.
4. Soil-survey methods and definitions.
(1) Description of methods used in mapping soils.
(2) Definition of terms used in the description and classification
of soils.
5. Soils and crops.
(1) General characteristics of the soils of the area.
(2) Systems of agriculture practiced.
(3) Relationship between soils and agriculture.
(4) Grouping of soils on the basis of capability for use.
(5) Brief description of each group.
(a) Agronomic relationships.
(b) Characteristics common to soils in the group that
determine use.
(c) The names of series in the group and the character-
istics of the types of the series brought into this

(d) Description of each type and phase of the group,
including all the following features, if applicable.
I. Describe each horizon as to :-
a. Color.
b. Texture.
c. Structure.
d. Consistence.
e. Thickness.
f. Reaction.
g. Content of organic matter.
h. Stoniness.
i. Root penetration.
j. Salt or alkali.
2. Important variations within the type.
3. Location and extent of soil.
a. General location of the larger areas
and definite location of areas of
minor types.
b. Estimate in square miles of areas of
each type.
4. Relief.
5. Geologic origin of parent material.
6. Drainage (external and internal).
7. Native vegetation, if important.
8. Uses to which soil is put and crop yields
woven in with capabilities of soil.
Approximate acreage in different crops.
9. Management of the soil. Systems of
management compared with those of
area as a whole. Susceptibility to erosion
or deterioration from other causes under
wrong management.
io. Methods of drainage or irrigation.
II. Water supply if related to soil type.

6. Land uses and agricultural methods.
(A) Capabilities of soils for use.
(1) Crops, (2) native pasture, (3) forests.
(2) Successful and improved methods of management as
demonstrated within the area.

(3) Results obtained by experiment on soil types or
closely related types, including :-
(a) Use of various fertilizers, lime or other
chemical treatments.
(b) Rotations.
(c) Farm implements.
(d) Prevention of erosion.
(e) Tillage.
(f) Drainage.
(g) Varieties of plants.
(h) Plant diseases, insect pests, and noxious
(This discussion should be specific in respect to the
separate soils.)

7. Drainage, irrigation, or alkali amelioration (not always necessary).
(Indicate exact meaning of the term alkali in each particular

8. Productivity ratings.
(Instructions on this chapter will be supplied in each instance.)

9. Morphology and genesis of soils.
(A) Location of the area with reference to the great soil groups.
(B) Parent materials.
(C) Factors of environment influencing soil development.
(D) Description of normal regional profile and a few of the more
important variations.
(E) Character of intrazonal and azonal soils.

10. Summary.
(A) Brief discussion of area and its agriculture.
(B) Uses made of soils and reason for such use.
(C) General statement of the character of the soils.
(D) Brief description of the soil groups and their relation to
(E) Names of the principal soils, their characteristics, and their
influence on the agriculture.

This outline should be followed in preparing the report, unless some
special individual feature of an area makes some change imperative in
order to achieve a logical presentation of the material.

No. pp. Pri
and J. W ishart) ... ... ... ... ... ... ... 24 2s. o
September, 1930 ... ... ... ... ... ... ... 44 is. 6
(by H. G. Thornton) ... ... ... ... ... ... 39 is. 6
de 'Sigmond) ... ... ... ... ... ... ... 26 is. 6
27. LAND AMELIORATION IN GERMANY (by H. Janert) ... ... 32 2s. o
29. SOIL, VEGETATION AND CLIMATE ... ... ... ... ... 43 2s. o
F. Yates) ... ... ... ... ... ... ... ... 96 5s. o
36. EROSION AND SOIL CONSERVATION* ... ... ... ... 206 5S. 0
37. SOIL STRUCTURE (by E. W. Russell) ... ... ... ... 40 2s. o
S. D. Garrett) ... ... ... . ... ... 54 2S. 6
39. THE MINOR ELEMENTS OF THE SOIL ... ... ... ... 86 4s. o
WHEAT (by Sir E. J. Russell and D. J. Watson) ... 163 7s. 6
41. THE TAKE-ALL DISEASE OF CEREALS (by S. D. Garrett) ... 40 2s. 6
42. THE MINERALOGY OF SOIL COLLOIDS (by G. Nagelschmidt) ... 33 2S. 6,
*Also published by the Imperial Bureau of Pastures and Forage Crops as Bulletin No. 25.








A comprehensive abstract journal of soil science, issued four times yearly. Each
issue contains about 5oo abstracts and references to current literature besides short
editorials, reviews and other matter of general interest. A special edition printed
on one side only of the page, so that the abstracts can be cut out and filed, is

Subscription Rates
Ordinary edition, per annum
Special edition ,, ,, ...
Single copies (ordinary edition only

British Commonwealth
... ... ... 20S. od.t
... ... ... 25s. od.f
) ... ... 7s. 6d.

INoTE.-These reduced rates are obtainable only when the order is sent direct to the
Bureau from the subscriber. They are not applicable to foreign orders or to British
orders sent through a trade agent.

25s. od.
30s. od.
7s. 6d.

Nos. I, 2, 4, 7 to Ii (1940) ... ... ... ... ... ... ... each, 6d.

No. i (October, 1932) ... ... ... ... ... ... ... Is. 6d.
Subsequent issues ... ... ... ... ... ... ... ... each, 6d.

AGRONOMY, 1931-1934; 1934-1937; 1937-1940; 1940-1944
Complete reference books to the recent literature of soil and allied sciences.
Each bibliography contains over 6,ooo classified references, a comprehensive subject
index, and lists of authors, journals and periodicals.
Crown octavo. Bound in cloth. Vols. 1-3, price 25s. od. each; vol. 4, price
30s. od.

"The Katamorphism of Igneous Rocks under Humid Tropical Conditions." By
the late Professor Sir John Burchmore Harrison, C.M.G.

Published on behalf of the British Association and Demerara Proprietors, Ltd.,
this monograph gives a detailed account of the author's work during thirty-seven
years in British Guiana, on the processes of lateritization.
Crown quarto: 8o pp., with a foreword by Sir E. J. Russell and a preface by
Professor F. Hardy. Price 5s. od.

The following series of multigraphed periodicals were discontinued in December,
1937. Annual sets (some incomplete) are obtainable at the prices given below.
"Publications Relating to Soils and Fertilizers" (monthly), 1935 (8 issues),
6s. od.; 1936 (io issues), 7s. 6d.; 1937 (Io issues), 7s. 6d.

"Monthly Letters," 1935 (Io issues) ; 1936 (12 issues); 1937 (ni issues). 4s. od.
per annum.

"Recent Developments in Soil Analysis" (quarterly). 16 issues (Nos. 5 and 1I
missing), 7s. od.


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs