Natural resources research

XI

Titles in this series:

I.

A review of the natural resources of the African continent.

Enqute sur les ressources naturelles du continent africain.

II.

Bibliography of African hydrology

I.

III.

Carte gologique de l'Afrique

/ Bibliographie hydrologique africaine, by J. Rodier.

(1/5 O00 000). Notice explicative / Geological map of Africa (1/5 O00 000). Explanatory note, by

R. Furon and J. Lombard.

IV.
IV.
Y.

Review of research on laterites, by R. Maignien.,

Compte rendu de recherches sur les latrites,par R. Maignien.
Functioning of terrestrial ecosystems at the primary production level. Proceedings of the Copenhagen symposium/Fonctionnement

des cosystmes terrestres au niveau de la production primaire. Actes du colloque de Copenhague. Edited by F. E. Eckardt.
VI.

Aerial surveys and integrated studies. Proceedings of the Toulouse conference

de la confrence de Toulouse.

/ Exploration arienne et tudes

intgres. Actes

VII. Agroclimatological methods. Proceedings of the Reading symposium /Mthodes agroclimatologiques.Actes du colloque de Reading.
VIII. Proceedings of

the symposium on the granites of West Africa

Compte rendu du colloque sur les granites de l'ouest africain.

IX. Soil biology. Reviews of research.

IX. Biologie des sols. Comptes rendus de recherches.

x.. Use and conservation of the biosphere. Proceedings of the intergovernmental conference of experts on the scientijc basisfor rational
X.

use and conservation of the resources of the biosphere, Paris, 4-13 September 1968.
Utilisation et conservation de la biosphre. Actes de la Confrence intergouvernementale d'experts sur les bases scienti$ques de
l'utilisation rationnelle et de la conservation des ressources de la biosphre, Paris 4-13 septembre 1968.

XI.

Soils and tropical weathering. Procedings of the Bandung symposium, 16-23 November 1969.

soils and

tropical weathering
Proceedings of the Bandung Symposium
16 to 23 November 1969

..

Published by the
United Nationr Educational, Scientific and Cultural Organization
Place de Fontenoy, 75 Paris-7e
Printed by Vaillant-Carmanne,S.A., Lige (Belgium)

Q Unesco 1971
Printed an Belgium

SC.'IO/XII.ll/A

Foreword

The study of the various components of the natural

environment that can provide m a n with resources and
of the interplay of these components is at the base of
Unesco7s Natural Resources Research Programme.
The programme is designed to advance scientific
knowledge concerning natural resources; to m a k e a
synthesis of such knowledge; and to help M e m b e r
States in establishing and developing research and
teaching facilities.
The scientific study of tropical soils including their
formation under tropical conditions is of vital importance when one considers that between one-third to
one-half of the population of the world is living in
tropical regions. Like all environmental studies, soil
science must be based on sound understanding of
fundamental scientific concepts. Only with this understanding can the concepts be effectively applied to the
solving of problems which m a y be of economic and
scientific interest to man. It is on this level of basic
science that Unescos programme activities are concentrated.
Since 1955, Unesco has maintained a special prog r a m m e of scientific studies concerned with natural
resources in humid tropical regions. Within this prog r a m m e Unesco organized four symposia on H u m i d
Tropics Vegetation in South-East Asia-at
Kandy,
Ceylon, in 1956; at Tjiawi, Indonesia, in 1958; at
Goroka, Australian N e w Guinea, in 1960; and at
Kuching, Sarawak, in 1963-a seminar on the Ecology
of Tropical Highlands at Katmandu, Nepal, in 1968; a
symposium on Tropical Soils and Vegetation at
Abidjan, in 1959; and a first symposium specially
devoted to tropical soils, the Symposium on Laterites,
held in Madagascar in 1966.
In November 1969 Unesco organized a symposium
on the Different Aspects of Tropical Weathering at the
Indonesian Institute of Technology in Bandung. T h e
purpose of this meeting was to review the existing

knowledge of tropical weathering and certain connected

problems, to promote the exchange of ideas between
specialists studying these phenomena from different
points of view and to stimulate co-operation in this
field. T h e meeting was organized in consultation with
the International Society of Soil Science (ISSS), and
in co-operation with the Indonesian Institute of
Sciences (LIPI).The Government of the Republic of
Indonesia generously agreed to act as host for this
activity.
Twenty-six specialists from Belgium, Cambodia,
France, India, Indonesia, Malaysia, the Netherlands,
Pakistan, the Philippines, Thailand, the U.S.S.R. and
the Republic of Viet-Nam, invited by Unesco, and
observers from FAO, SEAMEC DIOTROP, and Caltex
Pacific, attended the meeting. T h e symposium was
directed by Dr. F. Fournier, consultant to Unesco on
natural resources research.
The symposium was mainly devoted to soil and fertility questions. T h e scientific programme included the
following topics :
1. T h e mechanisms of tropical weathering : decomposition of rocks and pedogenesis; formation and eyolution of clays, silica, metallic oxides and hydroxides.
2. Tropical weathering and soil composition :particular
characteristics of soils peculiar to the tropical zone.
3. Geomorphology and tropical weathering.
4. Tropical weathering and soil conservation :characteristics of soils in the tropical zone which influence
their rational utilization and conservation.
The symposium was inaugurated by Professor Sarwono
Prawirohardjo, Chairman of the Indonesian Institute
of Sciences. In his inaugural address he emphasized the
paramount importance of the study of rock-weathering
conditions in humid tropical zones from the economic
as well as the scientific point of view. Knowledge of
these conditions is basic to an understanding of fertility
questions which again is fundamental to scientific

efforts to increase agricultural production. H e also

referred to the role of rock weathering in the formation
of Indonesias important secondary mineral deposits,
and emphasized the serious problems of deforestation
in his country and the difficulty of geological mapping
in humid tropical areas. IIe recommended that more
interdisciplinary research should be carried out b y
geologists, geochemists and soil scientists. Addresses
were also given at the inaugural session by Professor
Dr. D.A. Tisna Amidjaja, Director of the Institute of
Technology, Bandung, and Mr. F.J. C. Pala, Deputy
Director of the Unesco Field Science Office for SouthEast Asia.
Discussions at the symposium were based on two
types of documents: (a) four keynote papers from
world-known specialists, written at the request of
Unesco and presented by them; and (b) supporting
papers on optional subjects pertaining to the symposium written and introduced by participants.
Authors of keynote papers were:
Professor J. J. Fripiat, Laboratoire de Physico-Chimie
des Minraux, Facult dAgronomie, Universit de
Louvain (Belgium).
Dr. J. v a n Schuylenborg, Laboratory of Regional
Soil Science, Agricultural University, Wageningen
(Netherlands).
Dr. P. Segalen, Services Scientifiques Centraux d e
lORSTOM,B o n d y (France).
Dr. G.D. Sherman, College of Tropical Agriculture,
University of Hawaii (United States of America).

T h e keynote papers and supporting papers are

published in full in this volume together with the
s u m m a r y report of the symposium and the texts of
recommendations unanimously adopted.
Unesco takes this occasion to express ita gratitude to
the Government of the Republic of Indonesia for having
generously agreed to act as host to the symposium.
It also thanks Professor Sarwono Prawirohardjo,
Chairman of the Indonesian Institute of Sciences and
all members and officials of his institute for their cooperation and support; Professor Tisna Amidjaja,
Director of the Institute of Technology, Bandung;
and the Local Organizing Committee, headed by
Professor J. Katili.
It further thanks all the distinguished scientists w h o
attended and assured the success of the symposium,
and those w h o contributed scientific papers.
T h e selection of material and the opinions expressed
in this publication reflect the opinions of the respective
authors and do not necessarily represent the views of
Unesco.
T h e designations employed and the presentation of
the material in this publication do not imply the
expression of any opinion whatsoever on the part of
Unesco concerning the legal status of any country or
territory, or of its authorities, or concerning the delimitations of the frontiers of any country or territory.

Contents
. . .

R E V I E W S OF R E S E A R C H

1.Formation and transformation of clay minerals in tropical soils

. .

15

Introduction . . . . . . . . . . . . . . . .
The acid character of adsorbed water and the stability of adsorbed hydronium
cations . . . . . . . . . . . . . . . . .
The ambiguous behaviour of aluminium . . . . . . . . . .
The silica phase . . . . . . . . . . . . . . .
Transformation sequences . . . . . . . . . . . . .
The genesis of clay minerals from primary silicates . . . . . . .
Transformationof micas into clay minerals . . . . . . . . .
The transformation of amorphous into crystalline secondary materials . . .
Transformation of clay minerals into other clay minerals . . . . . .
. . . . . . . .
Transformations into tridimensional silicates
Conclusions . . . . . . . . . . . . . . . .

15

18
18
19
19
20
21
21

. . . . . . . . . . . . . . . .

22

2.Metallic oxides and hydroxides in soils of the w a r m and humid areas

of the world: formation. identification. evolution . . . . . . .

25

J .J . Fripiat and
A .J .Herbillon

Bibliography

P.Segalen

15
17

17

. . . . . . . . . . . . . . . . 25
. . . . . . . . . . . . . . . . . 27
Origin of metallic oxides and hydroxides. rocks and minerals . . . . .
27
2 8
Weathering . . . . . . . . . . . . . . . .
The products of the soil . . . . . . . . . . . . . . 29'
Determination of oxides and hydroxides
. . . . . . . . . 30
.
Inuence of oxides and hydroxides on some soil properties
. . . . . 32
Part two . . . . . . . . . . . . . . . . .
32
Conditions of stability and mobility of metallic compounds
. . . . . . 32
Immobilization of oxides and hydroxides in soils . . . . . . . . . 34
Summaryof theevolution of soils . . . . . . . . . . .
35
Conclusion
. . . . . . . . . . . . . . . . . 36 ..
Bibliography . . . . . . . . . . . . . . . . .
37
Introduction

Part one

. '
.

...

I'

.
"

J .van

Schuylenborgh

3.Weathering and soil-formingprocesses in the tropics

Oxisols with plinthite
Theprofle . .
Theprocess

. . . . .

. . . . . . . . . . . . . .
. . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

'39
39
40
41

Oxisols without plinthite . .

A n oxisolprofile without plinthite
The process . . . . .
Bibliography

G . Donald

.
.
.
. . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

4.Mineral weathering in relation to utilization of soils

Sherman

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

II

49

. . . . . 51

Introduction . . . . . . . . . . . . . . . .
Application to agronomic use
. . . . . . . . . . . .
Relationship of lime requirement to mineral composition . . . . . .
The relationship of fertilizer requirement to mineral composition and weathering
stage of soils . . . . . . . . . . . . . . . .
. . .
Prediction of fertilizer needs according to stage of mineral weathering
The utilization of stages of mineral weathering in the location of mineral ores . .
The utilization of mineral weathering in identification of engineering properties of
soils and earth foundations . . . . . . . . . . . . .
The utilization of mineral weathering stage in land-use classification . . . .
Bibliography

47
47
49

51
52
52
54
56
56
56

56

. . . . . . . . . . . . . . . . .56

T R O P I C A L W E A T H E R I N G IN ASIA

I.U. Ahmed. M .S. Hussain

and M .R . K h a n

Distribution of mica in the soils of the Madhupur Tract. East Pakistan

(specially presented as supporting paper to review No .1 by J .J .Fripiat
and A .J .Herbillon) . . . . . . . . . . . . . .

61

Introduction . .
Experimental . .
Methods of analysis
Results . . .
Soil properties .
Discussion
. .

61
61
61
62
62
62

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

. . . . . . . . . . . . . .

. . . .

. . . . . . . . . . .

64

Trends in rock weathering in the southern part @f peninsular India-its

expression in morphogenesis of soils (specially presented as supporting paper
. . . . :. . . . . .
to review No .2 by P.Segalen)

65

Introduction . . . . . .
Geology . . . . . . .
Climate . . . . . . .
Natural vegetation . . . . .
Weathering processes . . . .
Soils. their properties and composition
Blacksoils . . . . . .
Laterite
. . . . . .
Redsoils . . . . . .
Conclusion
. . . . . .

65
65
68
68
69
69
70
70
71
71

Bibliography

S.V .Govipda Rajan

and R.S .Murthy

.
.
.
.
.

Bibliography

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

. . . . . . . . . . . . . . . .

72

K .V .S. Satyanarayana

Weathering of basic igneous rocks and genesis of clay minerals (specially

presented as supporting paper to review No .2 by P Segalen) . . . .

.
''

Introduction . . . . . . .
Laterites of Malabar and South Kanara .
Laterite and black soils of M a l w a Plateau

. . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .

S.V. Govinda

Rajan and
Biswas

75

Development of certain soils in the sub-tropicalhumid zone in south-eastern

parts of India. genesis and classification of soils of the Machkund Basin
(specially presented as supporting paper to review N o .2 by P.Segalen) .

77

Introduction . . . . .
Physiography. relief and drainage
Geology
. . . . .
Climate . . . . . .
Experimental . . . . .
Results and discussion . . .
Potential for land use
. .

77
77
77
78
78
80
80

Bibliography

T. Seshagiri Ilao

Ong

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

. . . . . . . . . . .
. . . . . . . . . . . . . . . .

84

85

Introduction . . .
Black cotton soils . .
Redsoils . . . .
Lateritic soils (Latosols)

. . . . . . . . . . . . .

85
85
86
86

. . . . . . . . . . . . . . . .

87

Correlation between chemical bond and chemical weatherability of the

c o m m o n rock-forming minerals (specially presented as supporting paper
. . . . . . . . . . .
to review N o .2 by P.Segalen)

89

. . . . . . . . . . . . . .
. . . . . . . . . . . . . .
. . . . . . . . . . . . . .

Introduction . . . . . . . . . .
Theory of chemical weathering . . . . . .
Chemical weathering of the c o m m o n rock-forming minerals
S u m m a r y and conclusions . . . . . . .
Bibliography

J .P.Andtiesse

.
.
.
.
.
.

Pedogenesis of some major soil groups in Mysore State. India (specially

presented as supporting paper to review No .2 by P.Segalen) . . . .

Bibliography

H. Ling

73
73
74

. . . . . . . . . . . . . . . . .

Bibliography

N .11.Datta

73

. . . . . .
. . . . . .

. . . . . .
. . . . . .
. . . . . . . . . . . . . . . .

89
90
91
.93
93

T h e influence of the nature of parent rock o n soil formation under similar

atmospheric climates (specially presented as supporting paper to review
No .2 by J .v a n Schuylenborgh) . . . . . . . . . . .

95
Introduction .
Climate . .
Altitude . .
Parent material
Soils . . .
Discussion
.
Conclusions .
Appendix

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

. . . . . . . . . . 95
. . . . . . . . . . 96
. . . . . . . . . . . 96
. . . . . . . . . . 96
. . . . . . . . . . 96
. . . . . . . . . . 98
. . . . . . . . . . 101

. . . . . . . . . . . . . . . . .

102

. . . . . . . . . . . . . . . .

110

Bibliography

V.M. Fridland

The

differences between crusts of weathering and soils developing on acid

and basic rocks in the humid tropics (specially presented as supporting
i
paper to review No .3 by J van Schuylenborgh)

. . .

111

. . . . . . . . . . . . . . . .

116

Genesis and evolution of red and black basaltic soils in Cambodia (specially
presented as supporting paper to review N o .3 by J van Schuylenborgh) .

117

Bibliography

T a n B o u n Suy

. . .

Introduction . . . . .
Parent materials . . . .
Weathering of the parent materials
Physical agents . . . .
Chemical agents . . . .
Characters of red basaltic soih~ .
Physical characters . . .
Physico-chemicalcharacters .
Cover plants and mulching
.
Conclusion
. . . . .
Acknowledgement . . . .

.
E.V. Tamesis
D.C.Salita

and

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

. . . . . . . . 117
. . . . . . . . 117
. . . . . . . . 118
. . . . . . . . 118
. . . . . . . . . 118
. . . . . . . . 121
. . . . . . . . 121
. . . . . . . . 121
. . . . . . . . 121
. . . . . . . . 122
. . . . . . . . 122

Appendix .

. . . . . . . . . . . . . . . .

122

Bibliography

. . . . . . . . . . . . . . . .

123

S o m e aspects of lateritic soil formation in the Dahican-Alayao area.

Camarines Norte Province. Philippines (specially presented as supporting
paper to review No .3 by J van Schuylenborgh) . . . . . . . 125

Introduction . . . . .
Geographic location
. . .
Climate . . . . . .
Relief and vegetation . . .
Soil-formingrocks . . . .
Morphological and chemical features
Discussion
. . . . .
Bibliography

Ng Siew Kee and

L a w Wei M i n

.
.
.
.
.
.
.
.
.
.
.

. . . . . . . . . . .
. . . . . . . . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

. . . . . .
. . . . . .
. . . . . .
. . . . . . .
. . . . . .

.
.
.
.
.

. . . . . . . . . . . . . . . .

125
125
125
125
126
126
127
127

Pedogenesis and soil fertility in West Malaysia (specially presented as

. . . . 129
supporting paper to review N o .4 by G Donald Sherman)

.
.
.
.
.
Relief . . . . .
Great Soil Groups . . .
Latosols
. . . .
Reddish-brownlateritics .
Red-yellowpodzolics . .
.
Yellow-grey podzolics
Laterites . . . .
Gleys . . . . .
Alluvial soils
. . .
Podzols . . . . .
Organic soils
. . .
Introduction . . .
Factors of soil formation
Climate . . . .
Parent materials . .
Vegetation . . .

Lithosols

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

. . . . . .
. . . . . .
. . . . . .
. . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

. . . . . . . . . . . . . .

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

129
129
129
129
130
130
130
130
. . 130
. . 130
. . 131
. . 131
. . 131
. . 131
. . 131
. . 131
e . 1 3 2

Nutrient itatus
. . . . . .
Differentiation at great roil group level.
Availability indices
Reserveindice8
. . . . .
Efficiency of reparation . . . . .
Differentiation below Great Soil Croup level
Crop response to manuring . . . .
Future development and physical factors .
Acknowledgement . . . . . .

. . . . . . . . . .

. . . . . . . . . .
. . . . . . . . . . . . . .

Bibliography

in

.
.
.
.
.
.
. . . . . . . .

.
.
.
.
.
.
.

132
133.
133:
134
135
135.
136
138
138

. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . . .
. . . . . . . . .139

FINAL R E P O R T O F T H E S Y M P O S I U M

Final report of the rympoiium

LISTOFPARTICIPANTS

. . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

..

. . ..,. .
b

143

147

I Reviews
of research

1 Formation and transformations

of clay minerals in tropical soils
J. J. Fripiat' and A. J. Herbillon'
Laboratoire de Physico-Chimie Minrale, Institut dea Sciences
de la Terre, 42 de Croylaan, Heverlee-Louvain (Belgium)

INTRODUCTION
It is well k n o w n that the main

change occurring in the

chemical composition when primary minerals are
weathered into secondary species consists in an increase
in the water content. T h e transformation of oxygen
atoms or ions into hydroxyls (constitution water or
lattice water) and the introduction of water molecules
hydrating inner 'porous surfaces appears to be the
dominant mechanism in the clay genesis or synthesis
processes. W h e n rocks of different mineralogical compositions are reduced mechanically into smaller-size
particles, the surface area per unit weight increases
progressively, allowing a higher amount of adsorbed
water to be in contact with the rock constituents.
Simultaneously the rock constituents are exposed to
more or less marked temperature effects and oxidation
processes. However, the over-all transformation process
into secondary minerals would not be understandable
without taking into consideration :(a) the remarkable
properties of water and especially of water molecules
in the adsorbed state; (b) the ambiguous behaviour of
alumina; and (c) the dynamic of the silica phase.
The first part of the contribution will be devoted to
these aspects of the weathering processes whilst, in
the second part, the generally accepted transformation
sequences will be reviewed in the light of the modifications provoked b y the hydration processes.
Elementary chemistry tells us that water is an
associated liquid, the molecules having a strong tendency to form hydrogen bonds with each other. This
accounts for the high dielectric constant and the high
. boiling point. T h e lone pairs of electrons on the oxygen
atom allows water to form strong bonds with the
orbitals of metal cations and this is at the origin of
the high hydration energy of cations.
Moreover, water is an amphoteric electrolyte undergoing the following autodissociation process :

2H,Oz H,O+

+ OH-

(K = 10-14 at 250 C).

It is thus a potential source of protons although under

'normal' conditions, it behaves as a w e a k acid.
It is appropriate to report here also the recent discovery (Bellamy et al., 1969; Wyllis et al., 1969) of
anomalous water, prepared b y condensing water vapour
into fine quartz capillaries at a temperature slightly
higher than that of the liquid water source. Anomalous
water does not display a m a x i m u m density at 4oC
as does normal water. It is m u c h more viscous and has
a specific gravity of about 1.4. It m a y be regarded as
a water polymer, hydrogen bonds of an unusual strength
being responsible for the high degree of association.
T h e presence in clays and primary materials of fine
pores, the surface of which might be more or less
comparable to that of quartz capillaries, suggests,
according to L o w and White (unpublished), that more
attention should be devoted to the study of anomalous
water in soil systems.
Nevertheless, the properties of water in the adsorbed
state have already been sufficiently explored to permit
the definition of some of its peculiarities.

THE ACID C H A R A C T E R OF ADSORBED

WATER A N D THE STABILITY. OF
A D S O R B E D HYDRONIUM CATIONS
Molecules in contact with a solid surface are submitted
to very strong surface electrical fields. Consider ,for
instance a surface cation balancing an .inner negative
lattice charge. S o m e of the water molecules belonging .
to the first layer hydrating the surface are in close
contact with this positive species: T h e y are therefore

1. The Univeriity of Louvain and YRAC. Tervuren.

J. J. Fripiat and A. J. Herbillon

strongly polarized, wiLh the consequence that the proton is rejected as far as possible from the positive core,
thus enhancing the dissociation. Such a process m a y
be schematically represented as follows :

Such a process is k n o w n for acid cations in aqueous

solutions :

I A13+6H20
I e I A12+OH,5H,0 I + H+(pKE 5 at 250 C)
but it is surely enhanced for any cation on a solid
surface for two reasons: (a) the absence of buffering
anions in the adsorbed layer, the negative charge being
inherent to the solid phase; and (b) the combined
action of the cation and of the electrical field arising
from the surface atoms or defects. Chemical evidence
(Mortland et al., 1963; Fripiat and Helsen, 1966) as
well as physical measurements (Fripiat et al., 1965;
Touillaux and Fripiat, 1966) have shown that water
molecules in the monolayer o n silica gels and m o n t m o rillonites have a degree of dissociation m u c h higher
than usual, i.e. of the order of 0.01. This means that the
ratio H+/H,O is of the order of I per cent, i.e. the
acid strength of a N/100 solution. That the degree of
dissociation on a silica gel surface is of the order of that
found on montmorillonite tends to show that the
absence of charge-balancing cations on the former
m a y be compensated by the field arising from surface
atoms or defects. The enhanced degree of dissociation
m a y thus be generalized to any ionic solid surface.
In the natural systems, however, the water contents
are usually high enough to form more than one adsorbed
layer :the hydroniums originating from the dissociation
in the first layer diffuse of course out from this first
envelope and are therefore diluted in the subsequent
layers, resulting in a rapid decrease of the over-all
acid character. Nevertheless, alternating drying and
wetting processes m a y regularly 'provide quite concentrated acid solutions around rock constituents. In

addition, in the narrow pores of a porous medium, the

monolayer situation m a y also exist for some time.
In order to understand the primary step in the
eventual hydrolyzing action of this acid layer, the
existence of stable hydronium species on silicate surfaces has to be questioned. It might'be suggested that
hydroniums exchange other metallic cations. In some
cases, the presence of H30+-balancing negative lattice
charges has been proposed, for instance on micas
(White and Burns, 1963), but protonic montmorillonite
has a short life since it weathers rapidly into an aluminium clay (Eeckman and Laudelout, 1961).
T h e behaviour of surface H,O+ is in fact very
complex. Consider for instance the following schematic
atomic arrangement which reproduces a part of the
feldspar lattices, i.e. a chain of silicon tetrahedra with
aluminium in isomorphic substitution (Formula i).
In the assumed situation,the charge-balancingcation
is supposed to be II,O+. T h e (a) configuration is not
bond is broken and the lone pair
stable :the Al-O-Si
of electrons of H,O combines with the free Al orbital
leading to the (b) configuration and, on drying, to
the (c) state. T h e resulting silanol group has been
observed b y infra-red spectroscopy (Basila, 1961) in
silico-alumina. T h e decationation process of zeolites
(Uytterhoeven et al., 1968), also clearly evidenced by
infra-red spectroscopy, leads to a similar mechanism
(Formula 2).
In this case, the (a') + (c') transition occurs
almost immediately since the decationization process
is initiated above 100-2000C.T h e (b) configuration in the
first example might also undergo deeper transformations,
allowing for the rupture of all the four O-Si bonds
and the real dissolution of Al. This might occur step
b y step or at once: no direct information is available,
but this is clearly the mechanism b y which the hydrolyzing action of water might occur. ,
In summary, both the acid character of the adsorbed
water molecules and the instability of H,O+ on silicate
surfaces could be responsible for the first steps of the
weathering process.

Formula 1.

H,O+
HO

/Al(-)\ / o \/
\
/sio\
O0

-f

(4

\ /OHZ
AI
/\

Si

O / \o

\/

HO

\ / \/O
AI

-f

./

Si

\oo/

(4

(b)

Formula 2.

N H1+
O

si

O/
"'(-)\ O 0/ \O
(al)

16

HO

\/

/\/

\/O

si

O0

(4

+NHw

Formation and transformations of clay minerals in tropical soils

T II E A M B I G U O US B E H A V IO U R
OF ALUMINIUM
According to the pH, aluminium in solution behaves
as AF+ or as a hydroxide, Al(OH),, or as a w e a k acid,
the dissociation of which produces an aluminate anion.
It is n o w well established that this crude picture
must be completed by the existence of polynuclear
complexes (Brosset, 1952 ; Brosset, Biedermann and
Sillen,1954;Fripiat,Van Cauwelaert and Bosmans,1965)

'8

such as Al,(OII) 'it, or Al,(OH)

and Al,(OII)
.;1'
In these complexes, aluminium is mostly sixfold coordinated, whilst in aluminates, aluminium is reputed
to be mainly fourfold co-ordinated. H s u (1968) has
shown that polynuclear complexes m a y be formed on
the surface of clay minerals such as montmorillonite
by treating the aluminium-saturatedclay by an alkaline
solution.
According to the structure of the solid phase, aluminium m a y be either fourfold co-ordinated to oxygen
atoms (such as in feldspars and zeolites, etc.) or sixfold
co-ordinated (such as in kaolinite, etc.), in tetrahedral
or octahedral configurations respectively. In micas and
in some aluminium spinels, both co-ordinations m a y
exist in the same lattice.
Therefore, in a solution as well as in a solid phase,
aluminium exhibits dual characteristics which are
responsible for the large variety of natural aluminosilicates.
The hydrolyzing action of water reported above
m a y therefore produce aluminium under various ionic
forms according to the local p H of the environment.
Most probably, in the usual p H range (i.e. from 3.5 to
7.5), the polynuclear complexes are well represented and
they are the usual ionic forms which will be involved
in the neogenesis processes in the presence of silica.
De Kimpe, Gastuche and Brindley (1964), Gastuche,
Fripiat and De K i m p e (1961)have shown that in the
amorphous silico-aluminas obtained b y co-precipitating
alumina and silica, the co-ordinationnumber of aluminium depends on the pH of the suspension. In acid
conditions, sixfold co-ordinatedaluminium is dominant,
whilst in an alcaline m e d i u m the fourfold co-ordinated
species is more important. It is noteworthy that this
observation is in agreement with the fact that kaolin
minerals are k n o w n to appear in desaturated media
whilst minerals in which aluminium is fourfold coordinated are formed under more alkaline conditions,
as shown below.
Recently, the unstable character of fourfold coordinated aluminium structures under acid conditions
has been clearly shown b y D e Kimpe and Fripiat (1968),
in a study of the hydrothermal transformation of fibrous
zeolites, saturated by II,O+. The transformation of
H30+-mordenite (which contains AlIV only) into
kaolinite was obtained at a rather low temperature
(170" C) and the first step involves most probably the

complete M I V + AlVI transformation.

In contrast, amorphous silico-aluminas (which contain AlIV and AlvI) have been transformed into analcime (whichcontains AlIV only) in the same temperature
range in presence of sodium carbonate b y D e Kimpe,
Herbillon and Fripiat (1966). More recently Frankart (1969)has observed the transformation of montmorillonite into the same zeolitic species in alkaline
soils.
In summary, aluminium hydrolyzed from primary
materials is present in the weathering solutions, very
probably in the state of polynuclear complexes, but
the co-ordination n u m b e r of aluminium in the n e w
solid phase obtained from the co-precipitation procees
with soluble silica depends on the pH of the environment. Under acid conditions the octahedralco-ordination
is the most usual, whilst in a more alkaline situation
the tetrahedral co-ordination is favoured.
The reason for this distinction is related to the parent
structure of either the polynuclear complexes or of
the aluminate anions. In the solid final structure, the
aluminium in octahedral co-ordination is bonded to a
variable number of hydroxyls and its charge is neutralized, whilst in tetrahedral co-ordination it is connected
to oxygen and carries a negative charge.
.

T H E SILICA PHASE
T h e hydrolyzing action of water on primary materials
liberates silicon as well as aluminium: silicon forms
polymeric hydrated SiO,, the molecular weight of which
is depolydepends firstly on the p H (6.0, 5.3)-silica
merized more at low pH than at high p H , but the
m a x i m u m degree of polymerization seems to occur
around PII 4-4.5.T h e nature of the initial silica network
has of course a profound influence on the rate of release
of silica.
In contrast to alumina, the tetrahedral co-ordination
of silicon remains unchanged under the usual conditions
in soils but the degree of polymerization allows the
number of hydroxyls attached to each silicon to vary
appreciably :the higher the degree of polymerization,
the lower the OH/Siratio.
T h e solubility of silica is very m u c h affected by the
cations in solution. W e y and Siffert (1961)have s h o w n
the influence of Al3+,Mg2+,resulting in the co-precipitation of silico-aluminas and silico-magnesia phases.
These results m a y be summarized by saying that the
stability of silica suspnsions depends on the ionic
environment and this will affect of course the m o v e m e n t
of silica in the weathering solutions. According to
D e K i m p e (1964),monovalent and divalent cations in
small proportions (MO/SiO,or M,O/SiO, E 1 per cent)
decrease the degree of polymerization. Reifenberg (1938,
1947), Demolon and Bastisse (1938, 1944) have attributed the mobility of silica in soil water to the protective action of iron oxides, forming with soluble silica

17

J. J. Fripiat and A. J. IIerLillon

ferrisilicic complexes. Tran-Vinh-An and IIerbillon (1966) have objected to these conclusions that the
above authors have worked under very high Si02/Fe203
ratios and that the more realistic situation in which
SiO,/Fe,O, ratios arc smaller-and even m u c h smallerthan 1 should be investigated. Under these conditions
silica and iron oxides do not increase mutually their
mobilities but they form amorphous ferrisilicates in
some w a y comparable to the alumino-silicates.
These n e w phases m a y be initially formed on the
clays or oxide surfaces, in this situation the isoelectric
points and thus the properties of the primary particle
m a y be affected (Herbillon and Tran-Vinh-An, 1969).
In summary, the local PII conditions and the eventual
presence of cations have a profound influence on the
degree of polymerization of silica and therefore on its
reactivity. T h e greater the depolymerization, the
higher is the OII/SiO, ratio. The silanol group is the
most obvious site for a further reaction with aluminium
polynuclear complexes, magnesium hydroxides, ferric
hydroxides etc., leading to the formation of various
amorphous or crystalline silicates. Thus, the transport
of silica from the weathering zones involves probably
a local environment depressed in polyvalent cations.

TRANSFORMATION SEQUENCES
Several transformation sequences responsible for clay
minerals genesis have been proposed: a specified clay
mineral m a y derive from primary silicates, from other
clay minerals, from amorphous materials such as
volcanic ashes, by the re-silicification of oxides or
hydroxides, by hydrothermal synthesis, etc. It is not
the aim of this paper to provide an extensivc rcvicw
of researches m a d e in this area. A few examples will
be selected in order to illustrate the importance of the
basic notions which have been mentioned above.
T H E GENESIS OF C L A Y M I N E R A L S
F R O M P R I M A R Y SILICATES

Such a transformation has been schematically represented by Keller (1964) as follows:

IMeAl silicates1 + HOH-t M e n + OH- + Aln+

+ Al(OH);++ Al(OII), + Al05 + Fe2+.,+
+ Fe(OII), + II,SiO, + Si0 + MeSi+ +

a-

j HMeA1 silicates (clay minerals)

In the absence of water, the transformation should

be impossible and it is for this reason that the peculiar

properties of water in porous media have to be considered.
The first step is of course the water adsorption
process on the external surface of the primary particles.
18

It mufit hc emphasized that hydration of minerals

of high surface areas m a y lead to some irreversible
changes. For synthetic fluorphlogopiic (which does not
contain constitution water), Ilouxhet and Drindley (1966) were able to show that ground material
was more deeply affected by hydration than bigger-sized
material. With ground mica (clay minerals size), for
instance it was impossible to remove the whole content
in hydration water even after long treatments under
high vacuum at 2000 C. This irreversible hydration
is obviously the first step of the hydrolysis process and
thus the first stage of the transformation of primary
minerals into clay minerals.
The second step m a y be assigned to the acid properties of adsorbed water. For instance, hydrolysis of
feldspars is thought by Frederickson (1951) to be
firstly a cationic exchange between II+ or H30+,
and
alkaline or alhali-earth cations located in the feldspar
lattice. Thus it m a y be assumed that the beginning of
the hydrolysis process leads to the formation of layers
of decationated feldspar lattices, very similar probably
to the decationated zeolite structure reported above.
The abrasion experiments reported by Stevens and
Carron (1948)m a y also be interpreted as the evidence
that the solid phase (the feldspar for instance) gains
protons while the solution is enriched in cations.
T h e famous Goldich (1938) stability sequences are
also a good illustration that the reactivity of the
Si-O or Ai-O bonds with respect to the proton is the
main factor which governs the primary silicatesstability.
Hydration of minerals in which silica tetrahedra are
isolated, such as in olivine, is more destructive than
hydration of minerals where silica tetrahedra share
all their four corners. Once again, it m a y be assumed
that it is easier to produce silanol groups starting from
isolated silica tetrahedra than from structures (like
quartz) containing a highly polymerized silica network.
Numerous investigations confirmed that the stability
of tectosilicates in water or in acid solutions is mainly
dependent on their Si/Al atomic ratio. Similar results
were obtained for zeolites and even for synthetic
silico-aluminas. De Kimpe and Fripiat (1968) were
unable to get stable H-zeolite from analcite (Si/AlN 2)
whilst the hydrogen forms of erionite and mordenite
(Si/AlN 5) were easily produced. Recently,Tardy (1969)
took advantage of the relative stability of tectosilicates
in order to explain h o w different clay minerals appear
selectively and afterwards coexist in the clay fraction.
As a result of the hydrolysis process, silica and
aluminium cations (monomeric as well as polymeric) are
dissolved and clay minerals or hydroxides are formed
from this mother solution. Hydrothermal alteration of
crystalline feldspars (Lagache, 1966) as well as of
glassy materials (Trichet, 1969) shows that, by controlling some very simple parameters of the solution, it
is possible to influence strongly the nature of the new
phases, in agreement with the thermodynamic data
given by Garrels and Christ (1963). Starting from

Formation and transformations of clay minerals iii tropical soils

albite for instance, it is possible to get boehmite, or

kaolinite or illite by controlling the [Na+]/[H+] or the
[K+]/[H+]
ratio and the silica concentration. Thus the
physico-chemical properties of the solution and not of
the initial solid phase appear to be the main factors
governing the neosynthcsis processes. These observations do not rule out the epitaxial influence of the
starting phase; Tchoubar (1965) for instance, has
never observed a transient amorphous phase on the
altered feldspar or in solution lut on the contrary, h e
emphasizes the importance of the epitaxial growth of
the neosynthetized mineral. In some way, these obscrvations appear to contradict those reported by Delvigne (1965)and Bates (1960),w h o have pointed out the
presence of amorphous materials around altered feldspars. T h e neoformation of kaolinite on the surface of
mica flakes has been observed by Millot (1964),
Tardy (1969),de Bthune et OZ. (personal communication), without an intermediate amorphous phase,
In summary, it s e e m s well established that the
chemical composition of parent materials does not
influence directly the nature of the secondary products,
but the nature of the secondary minerals depends on
the physico-chemical parameters of the surrounding
solutions. As weathering operates in an open system,
there is no direct relationship between the chemical
composition of the solid and of the solutes. The state
of organization of the parent material plays an important role by ruling the solubility of the mineral constituents and therefore the transient concentrations in
cations, aluminium and silica. Finally, the epitaxial
growth m a y favour .and activate to some extent the
rate of crystallization of the secondary minerals.
T R A N S F O R M A T I O N O F MICAS I N T O C L A Y
M IN E R A L S

T h e structural relationships between micas and some

clay minerals are so close tbat often their transformation
into 2 :1 clay-layer lattices does not require complete
hydrolysis. Hydration of micas and their simultaneous
transformation into vermiculite m a y occur w h e n mica
flakes are m u c h larger than 2 p, the upper size limit for
clay fractions. This fact will give us the opportunity to
study a large-scale model of what is actually a swelling
clay mineral and also the relationships between constitution hydroxyls and hydration water.
In a dioctahedral mica such as muscovite K(A1,SiJ
Al,O,,H,, the octahedral hydroxyl groups are oriented
approximately in the ab plane whilst in pure phlogopite
in which all the octahedral sites are occupied by the
Mg2+ cations, the Oare oriented perpendicular to
this plane. This m a y be easily demonstrated by infra-red
spectroscopy. Basset (1960) has suggested that the
stability of mica towards weathering could be related
to the orientation of the 0
11 bond in the mica lattice.
The perpendicular orientation allows the hydroxylic
proton to be close to the interlayer potassium, hence

the coulomlic repulsive action could be responsible

for the lability of trioctahedral micas, as compared
to the dioctahedral spccies.
Rouxhet et al. (1966) have shown that in biotites
where the two kinds of 011 populations m a y exist
simultaneously, there is a relationship between the
numhcr of O H in the perpendicular direction and the
weathering rate into vermiculite but the fluorine content
and other factors are also important.
Under natural conditions, the potassium depletion
is always followed by a decrease of the net negative
charge of the lattice; the order of magnitude of this
reduction is approximately one-third. Walker (1949)
recognized this fact very early and he suggested that
it m a y be caused by an introduction of protons into the
oxygen network. This suggestion has been checked
later on by R a m a n and Jackson (1966), N e w m a n and
B r o w n (1966) and by Robert and Pedro (1969). Once
again, the relationship between the weathering mechanism of trioctahedral micas and the hydration water
and the O H contents seems quite evident. It m a y .be
suggested that the removal of potassium allows the
acidic hydration water to penetrate between the layers
and the protons associated with these water molecules
as hydroniums to penetrate into the lattice structure.
However, the role of the oxidation state of iron in trioctahedral micas has never been completely elucidated,
nor has the real weathering processes of dioctahedral
micas. Potassium complexing agents, such as triphenylboron (Scott et al., 1960) allow replacement by sodium
but no direct proof has ever been presented that such
a process m a y occur in nature and from this respect,
muscovite and muscovite-like material appears to be
so stable that their direct transformation in illite seems
to be very questionable. As shown recently by Shutov,
Drits and Sakkarov (1969), the transformation of
montmorillonite into hydromica through an intermediate rectorite phase could be a very general process.
T H E T R A N S F O B M A T I O N OF ABIORPUOUS I N T O
C R Y 9 T A L LIN E S E C O N D A R Y M A T E RI A L S

T h e clay fraction of soils and especially of tropical

soils always contains some amorphous materials but,
with the attention focused o n crystalline clay minerals,
these is a tendency to forget that, in each particular
situation, amorphous and crystalline materials do in
fact coexist. W e do not like to use the word equilibrium because of the dynamic character of the soilforming processes. T h e mineralogical composition is a
result of a steady state that m a y be far removed from
a true thermodynamic equilibrium. W e have assumed
that the genesis of clay minerals involves a hydrolysis
followed by a co-precipitationprocess, and this requires,
as a second step, either the evolution of the amorphous
into a crystalline secondary material or the immediate
formation of both ordered and disordered structures.

19

J. J. Fripiat and A. J. IIerbillon

As compared with a crystalline phase of the same

composition, the amorphous material is always characterized by a higher hydration.and constitution water
content, and, of course by a higher specific surface
area. As far as the lattice organization is concerned,
the disorder in the amorphous materials is mainly
due to slight misfits in the arrangement of structural
units. Inside each unit, however, the atomic distances
and angles are practically the same as in the corresponding crystalline phase.
Moreover the composition appears to have a deep
influence on the yield in well-ordered structure. Coprecipitation of magnesia and silica for instance leads
very easily to the formation of montmorillonite
(Caillre and Henin, 1961) or to talc (De Kimpe, 1964),
whilst co-precipitation of alumina and silica at room
temperature produces only a very slight amount of
kaolinite ( D e Kimpe, Gastuche and Brindley, 1964)
and a large proportion of amorphous phase that does
not apparently evolve further.
As observed by numerous experimenters, increasing
the temperature and the water vapour pressure accelerates to a very great extent the rate of crystallization.
T h e reasons for this acceleration are very complex.
For instance, according to D e Kimpe, Gastuche and
Brindley (1964)the complete hydrothermal transformation of a mixture of gibbsite and silica into kaolinite
at 175oC under acid conditions could be due to the
peculiar activity of nascent boehmite obtained from
gibbsite in this temperature range and to the fixation
of silica on this nascent lattice. The generalization of
this observationmeans that any time a crystalline pbase
is thermodynamically unstable, its transformation into
a new structure through the combintion with another
constituent might be expected.
Poncelet and Brindley (1967) have observed that
montmorillonite saturated b y polynuclear aluminium
cations m a y be quantitatively transformed into kaolinite around 1750 C. T h e peculiar configuration of this
parent material would favour this transformation.
Tchoubar and Oberlin (1963) have reported that the
hydrothermal transformation of feldspars into kaolinite
m a y be preceded by the formation of boehmite followed
by a resilicification stage of this material in a w a y
similar to that described b y D e Kimpe, Gastuche and
Brindley (1964). Although these examples seem to
suggest solid-phase transformations, the acid environment does not exclude the dissolution-coprecipitation
sequence discussed previously.
T h e fundamental question thus remains : m a y a
disordered structure transform into the corresponding
ordered lattice by keeping the same approximate composition in a temperature domain and in a period of time
acceptable in pedology?
Au indicated above, chemical compositions are k n o w n
for which the answer is yes, but for the most frequently
occurring system (silico-alumina),the experimentalwork
performed so far, in m a n y laboratories fails to provide
20

an answer.

The

question might be solved morc easily

if we k n e w the growing mechanism of the clay microcrystal.

It has h e n suggested that the octahedral layer
patterns the fixation of the silica tetrahedral layer
(Caillre and Henin, 1961; Gastuche and D e Kimpe,
1961) or on the contrary, that the silica network is
formed first (Fripiat, 1961). Both hypotheses suggest
that the growing mechanism operates along the c axis.
In the study of the transformation of acid zeolites into
kaolinite, De K i m p e and Fripiat (1968)suggested that
kaolinite could grow in the ab plane, e.g., by a progreschains.
sive arrangement of rows of O-Si-O-Al-OH
The failure to synthetize kaolinite at low temperature
would then result from the lack of a sufficient number
of these chains in the mother solution.
The reason for the failure in the synthesis of kaolinite
m a y be either that adequate conditions have not been
found, or that the process is too slow, probably because
of a very high activation energy. This would explain
why kaolinite is easily produced as soon as the temperature is increased sufficiently. Determinations of
activation energies in the synthesis process of clay
minerals are rather scarce. For kaolinite, data obtained
by Rayner (1961) s e e m to indicate values of the order
of 200 kcal. This is of course high enough to require
an energetic thermal activation in order to obtain an
appreciable yield in crystalline material but very little
is k n o w n o n the influence of the nature of the starting
materials on the order of magnitude of this activation
energy. W e are inclined to believe that the time factor
m a y explain the failure of the synthesis attempts as
far as kaolinite is concerned. The situation is completely
different for the magnesium silicates. It must be
emphasized that the considerations reported here for
the evolution of amorphous materials into clay minerals
are also pertinent for the evolution of allophanes. It
has been recognized (Cloos, IIerbillon and Echeverria,
1968; Cloos et al., 1969) on m a n y occasions that allophanes for instance are very similar in composition and
structure to synthetic amorphous silico-aluminas. On
the other hand, the evolution of allophane into halloysite and kaolinite has been observed under various
climatic conditions (Sieffermann and Millot, 1969).
TRANSFORMATION OF CLAY MINERALS
I N T O OTIIER C L A Y M I N E R A L S

Very often, the replacement of a clay mineral by another

is thought to be a transformation such that the new
species inherits to some extent the lattice of its parental
material. The inherited part would undergo some
further modifications in order to achieve complete
transformation.
Obviously, transformation in which the basic oxygen
framework remains unaltered is that by which a mica
evolves into vermiculite and perhaps montmorillonite.
Illites and hydromicas are usually considered as origi-

Formation and transformationsof clay minerals in tropical soils

nating from the same kind of mechanisms as they would

derive either from degraded detrital micas or from
vermiculites or montmorillonites. However, illites are
always richer in water than micas (Wentworth, 1967)
and this simple fact shows that the hydration process
is largely irreversible.
Illites also are k n o w n to be richer in silica and this
point is easy to understand if illites come from the
aggradation of montmorillonites but more difficult to
admit if they are derived from primary micas. The
latter transformation would imply that the tetrahedral
layers of the initial mica lattice, which formerly contained aluminium in fourfold co-ordination, must be
partially renewed and re-silicificated.
This example shows to what extent the problems
arising in clay transformations are often neglected in
the very rough picture commonly used in soil clay
mineralogy. Except in a few very obvious cases, such
as those mentioned above, w e do not believe that the
transformation sequences involving real solid state
transformations are really founded. On the contrary,
it is clear that the complete dissolution of the primary
mineral followed by the neosynthesis of the n e w mineral
is a more convenient w a y to get the expected result
than the diffusion of a strongly charged cation through
the very energetic potential barrier of the oxygen
lattice. W e k n o w that some workers will object that
the random or regular interstratification of' two clay
mineral lattices which is often observed m a y evidence
some mutual transformations. T o this w e reply that
the same result would be obtained assuming epitaxial
growths.
T R A N S F O R M A T 1 O N S I N T O TRI D I fi1 E N S I O N A L
SILICATES

Pedogenetic processes m a y lead to the formation of

tridimensional silicates. Quartz neoformation during
soil development, which has long been suspected,
was clearly demonstrated to occur by Bonifas (1959).
In the course of the complete dissolution of ferromagnesian primary minerals giving rise to the absolute
accumulation of iron oxides under tropical humid
conditions (ferrallitization,laterization), silica, which is
less mobile than alkaline and alkaline-earth cations,
remains in the profile and crystallizes into secondary
quartz veins.
The relationship between palygorskite, a porous
silicate which presents some analogy with the amphibole
structure, and the montmorillonite were recognized a
long time ago. However palygorskite was reputed to
be of hydrothermal or lacustrine origin and its presence
in some soils was considered as an inheritance. Recently,
V a n den Heuvel (1966), Al Rawi and Sijs (1967) and
blillot et al. (1969)have shown that palygorskite is a
petlogcnetic mineral occurring simultaneously with the
accumulation of calcium carbonate in some aridisols.
This silicate as well as the quartz in the above example

appears thus as end members produced under extreme

conditions : the palygorskite is formed w h e n there is
an excess of alkaline-earth cations and quartz w h e n
the m e d i u m is highly desaturated. These processes
indicate that silica is poorly soluble and prefers to
polymerize on itself or with other oxides even when
conditions are no more favourable for clay minerals
synthesis.
Analcite, a sodic zeolite, was recently found to be
of pedogenetic origin. Baldar and Whittig (1968),
Frankart and Herbillon (Franckart, 1969; Franckart
and Herbillon, in press) described the occurrence of
this zeolite in alkaline soils rich in sodium carbonates.
Frankart and Herbillon emphasize the fact that in
such alkaline conditions, sodium carbonate reacts
with the clay lattice, giving rise to a change in the
number of co-ordination of aluminium. This example
illustrates the ambiguous behaviour of aluminium
reported previously. X-ray diagrams presented by
Baldar and Whittig, and Frankart and Herbillon show
clearly that in alkaline soils, montmorillonite disappears
when analcite is synthesized and that this transformation is more active in the upper horizons richer in
sodium carbonates. This illustrates the fact that
surface accessibility to the reagent is a more important
factor than the nature of the parent minerals. As pointed
'out by De Kimpe, Herbillon and Fripiat (1966),
montmorillonite, because of its high surface area
(800 mz), m a y react to give a zeolhe: it behaves in
the same w a y as pyroclastic products containing
fourfold co-ordinated aluminium.

CONCLUSIONS
T h e title and the content of this review paper m a y
appear as more or less in contradiction since no reference has been m a d e so far to the tropical soils, or more
especially to the influence of tropical weathering conditions on the formation or transformation processes of
clay minerals. In fact, w e do not believe that the
transformation sequences or the synthesis mechanisms
observed in the tropical regions are essentially different
from those occurring under other climatic conditions.
T h e reported characteristics of tropical weathering are
in fact due to the extreme conditions prevailing in
countries where the amounts of water circulating on
the earth surface m a y vary to a very great extent and
where the temperature is always quite high. Moreover,
in m a n y parts of the continents located between the
tropics, these very energetic actions have been working
for a very long time, even very long in the geological
time scale. These three points are actually those that
differentiate tropical from temperate countries as far
as the clay mineral-forming factors are concerned.
The large variationu in the soil water contents,
together with the high temperature, m a y activate to
a great extent the hydrolyzing actions described in the

21

J. J. Fripiat and A. J. IIerbillon

first part of this paper. Still, the long history of most

of the tropical soils leads very often to the end members
of the transformation sequences. In the tropical countries characterized by a more recent geological history
and especially, in those with volcanic activity, the field
observations show very fast transformation processes
leading to less organized structures. But these processes
are not basically different from those that have occurred
under other climatic conditions.
T h e desaturation degree of soils in m a n y tropical
areas favours, by lowering the pH, the formation of
kaolinitic minerals. Conversely, each time the drainage
conditions are poor, as for instance in the calcareous
and alkaline soils of the aridic areas, the pH conditions,
and perhaps the concentrations in CO:- and SO:anions, favour the transformation into montmorillonite
and even into zeolitic or zeolitic-like materials. These
t w o latter observations are clearly related to the
ambiguous behaviour of aluminium. On the other hand,
magnesia reacts with silica in forming crystalline
materials whilst alumina and silica m a y combine into
quite stable amorphous phases.

T h e weathering of micas occurs in a very dii'erent

w a y from that of fcltlspars. In the former case, transformations in which the oxygen lattice is preserved
m a y be observed whilst in the second case, clay minerals
are formed probably through the intermediate stagcs
of disorganized phases, involving a complete redissolution.
Again these processes are very dependent o n the
amount of water and o n the fluctuations in the amount
of water in the weathering zones.
Finally, it must be pointed out that, even in the
cases where structural similarities exist between the
initial and final lattices, the genetic link must be
founded on the physical nature of the transformation
processes and not on these similarities. In m a n y cases,
this procedure has been neglected with the consequence
that solid-state transformations without any real
.
significance have been proposed.
In this respect, it must also be recalled that soils
are open systems in which steady but not true equilibrium states are reached.

Bibliography
AL RAWI, G.J.; SIJS,C. 1967. A comparative study between

CLOOS,P.; HERBILLON,A.; ECIIEVERRU,J. 1968. Allophane

Euphrates and Tigris sediments in the mesopotanian flood

plain. Pdologie (Gand), vol. 17, no. 2, p. 187-209.
BALDAR, N. A.; WHITTIG,
L. D. 1968. Occurence and synthesis of soil zeolites. Proc. Soil Sci. Soc. Amer., vol. 32,
p. 235-8.
BASILA,M. R. 1961. Hydrogen bonding interaction between
adsorbate molecules and surface hydroxyl groups on silica.
Chemical physics, vol. 35, p. 1151-8.
BASSET,
W.A. 1960. Role of hydroxyl orientation in mica
alteration. Bull. Geol. Soc. Amer., vol. 71, p. 449-56.
BATES, T. F. 1960. Rock weathering and clay formation in
IIawaii. Mineral industries, Pennsylvania State Univ.,
vol. 29, p. 1.
BELLAMY,
L. J.; OSBORN,
A. R.; LIPPINCOTT,
E.R.; BANDY,
A. R. 1969. Studies of molecular structure and spectra of
anomalous water. Chem. &Ind., vol. 8, no. 21, p. 686.
DE BETHUNE, P.; NEYBERG, H.;
MARTIN,H.; HERBILLON,A.
(Personal communication to be published in J. Sedimen-

like synthetic silieo-aluminas. Phosphate adsorption and

availability. Trans. 9th Int. Congr. Soil Sci., Adelaide,
vol. II, p. 733-43.Amsterdam, Intern. Soc. of Soil Science.
-. , LEONARD, A. J.; MOREAU, J. P.; HERBILLON, A;
FRIPIAT,
J. J. 1969. Structural organization in amorphous
silico-aluminas.Clays and clay minerals, vol. 17, p. 279-87.
DELVIGNE,J. 1965. Pidogense en zone tropicale. I. La forma-

tology.)

BONIFAS,
M. 1959. Contribution l'tude gochimique de
l'altration latritique. Alm. Seru. Carte Gol. Als. Lor.,
vol. 17, 159 p.
BROSSET, C. 1952. Acta Chem. Scand., vol. 6, p. 910.
-. , BIEDERMANN, G.; SILLEN, L.G. 1954. Acta Chem.
Scand., vol. 8, p. 1917.
CAILLERE,S.; TIENIN, S. 1961. Vue d'ensemble sui le problme
de Ia Synthese des minraux phylliteux basse temprature.
Gense et synthse des argiles, p. 31-41. Paris, Coll. Int.
CNRS. (No.105.)

22

tion des minraux secondaires en milieu ferrallilique.

Ph.D.

Thesis, I,oiivain, Fac. Sciences, 177 p.

DEMOLON, A.; BASTISSE,
E.M. 1938. Sur quelques complexes
collodaux du fer et de l'aluminium. Ann. agron., Paris,
vol. 1, p. 6-33.
-. ,-. 1944. Rle vecteur de la silice dans les phnomnes
gochimiques et physiologiques. Ann. agron., Paris, vol. 1,
p. 265-96.
EECKMAN, J. P.; LAUDELOUT, H. 1961. Chemical stability of
hydrogen montmorillonite. Kolloidzschr., vol. 178, no. 2,
p. 99-107.
FRANKART,R. 1969. Les sols halomorphes de la Basse Huzizi.
(To be published in Annales Muse Royal de l'Afrique
Centrale, Tervuren, Belgium.)
-. ,HERBILLON,A. (In press.) Prsence et gense d'analcime
dans les sois sodiques de la Basse Ruzizi (Burundi). BUZZ.
Groupe Franais des Argiles.
FREDERICKSON,A. F. 1951. Mechanism of weathering. Bull.
Geol. Soc. Amer., vol. 62, p. 221-32.
J. 1961. Jn discussion of the Caillre et Ilnin paper.
FHIPIAT,
Gense et synthse des argiles, p. 41. Paris, Coll. Jnt.
CNRS. (No. 105.)

Forniation and transformations of clay minerals in tropical soils

FRIPIAT,J.; HELSEN, J. 1966. Kinectics of decomposition of

cobalt coordination complexes on montmorillonite surface.
Clays and clay minerals,Proc. 14th Conf., p. 163-74.Berkeley,
Bailey.
-. , JELLI, A.; PONCELET,
G.; ANDRE, J. 1965. Thermodynamic properties of adsorbed water molecules and
electrical conduction in montmorillonites and silicas. J.
Phys. Chem., vol. 69, p. 2185-97.
-. , LEONARD, A.; I ~ A R A K ~ , N. 1963. Relations entre la
structure et la texture des gels de silice. Bull. Soc. Chim.
France, p. 122-40.
-. , VAN CAUWELAERT, F.; BOSMANS, II. 1965. Structure
of aluminum cations in aqueous solutions. J. Phys. Chem.,
VOI. 69, p. 2458-61.
GARRELS,
R. M.; CHRIST,C. L. 1963. Solutions, minerals and
equilibria. New York, Harper &Row. 450 p.
GASTUCIIE,
&I. C.; FRIPIAT,J. J.; DE KIMPE, C. 1961. L a
gense des minraux argileux de la famille du kaolin.
I. Aspect collodal. Synthse et gense des argiles, p. 57-65.
Paris, Coll. Int. CNRS. (No. 105.)
--. , DE KIMPE,C. 1961. La gense des minraux argileux
de la famille du kaolin. II. Aspect cristallin. Gense et
synlhse des argiles, p. 67-74. Paris, Coll. Intern. CNRS.
(No.105.)
GOLDICH, S. S. 1938. A study in rock weathering. J. Geol.,
VOI. 46, p. 17-23.
HERBILLON,A. J.; TRANVINIIAN, J. 1969. Heterogeneity
in silicon iron mixed hydroxides. J. Soil Sci., vol. 20,
p. 223-35.
IIsu, P a 110. 1968. Heterogeneity of montmorillonite surface
and its effect on the nature of hydroxy aluminum interlayers. Clays and clay minerals, vol. 4, p. 303-12.
ILER, R. K. 1955. The colloid chemistry of silica and silicates.
New York, Cornell University Press. 324 p. ,
KELLER, W.D. 1964. Processes of origin and alternation of
clay minerals. In :C. I. Rich and G. W.Kunze (eds.), Soil
clay mineralogy. The University of North Carolina Press.
DE KIMPE, C. 1964. La couche ttradrique et la synthse des
argiles. C.R. 8e Congr. Int. Sci. Sol, Bucarest, vol. III,
Commission 7, p. 1203-12.Publishing House, Academy of
the Socialist Republic of Romania.
, FRIPIAT, J. J. 1968. Kaolinite crystallization from
H exchanged zeolites. Amer. Alin., vol. 53, p. 216-30.
-. ,GASTUCHE, M.C.; BRINDLEY,
G.W. 1964. Ionic coordination in ahmino-silicic gels in relation to day minerais
formation. Amer. Min., vol. 46, p. 1370-81.
-. , HERBILLON,
A.; FRIPIAT,J. J. 1966. Synthesis of
analcite and clay minerals in relation to the reactivity of
the starting materials. Proc. Int. Clay Conf., vol. I,
p. 109-19. Jerusalem, Israel Program for Scientific Translation.
LAGACHE, M. 1966. Synthse de la Boehmite, de la kaolinite
et de la muscovite par altration de l'albite par l'eau
ZOOoC en prsence d'acide carbonique. Bull. Groupe Franais des Argiles, vol. 17, p. 71-9.
LOW, P.; WHITE, J. L. To be published in Science.
MILLOT, G. 1964. Gologie des argiles. Paris, Masson. 499 p.
, PAQUET, H.;RUELLAN,
A: 1969. Noformation de
l'attapulgite dans les sols carapaces calcaires de la Basse
Moulouya (Maroc Oriental). C.K.Acad. Sci. Paris, vol. 268,
p. 2771-4.
J.; UYTTERMORTLAND, RI. M.; FRIPIAT,J. J.; CIIAUSSIDON,
IIOEVEN,J. 1963. Interaction between ammonia and the
expanding lattices of montmorillonite and vermiculite.
J. Phys. Chem., vol. 67, p. 248-58.
-0

--

NEWMAN, A. D.; BROWN, G. 1966. Chemical changes during

the alteration of micas. Clay Min. Dull., vol. 6, p. 297.
PONCELET,
G.&f.; BRINDLEY,G.W. 1967. Experimental
formation of kaolinite from montmorillonite at low temperature. Amer. Min., vol. 52, p. 1161-73.
~XAMAN, K.V.; JACKSON,BI. L. 1966. Layer charge relations
in clay minerais of micaceous soils and sediments. Clays
and clay minerals, voi. 26, p. 57-59. Berkeley, Bailey.
RAYNER,J. II. 1961. A n examination of the rate of formation
of kaolinite from a coprecipitated silica gel. Gense et
synthse des argiles, p. 123-7. Paris, Coll. Int. CNRS.
(No.105.)
REIFENBERG,
A. 1938. The soils of Palestine. London, Th.
Murby.
.1947. The soils of Palestine. London, Th. Murby.
ROBERT, 111.; PEDRO,
G. 1969. Etude des relations entre les
phnomnes d'oxydation et l'aptitude l'ouverture des
micas trioctadriques. Proc. Int. Clay Conf., vol. 1,
p. 455-74.Tokyo, Israel Universities Press.
ROUXHET, P.G.; BRINDLEY,G. W. 1966. Experimental
studies of fine grained micas. II. The water content of wet
ground micas. Clay Min. Bull., vol. 6, p. 219.
-. ,HERBILLON,A.;FRIPIAT,
J. J. 1966. Note sur la relation
entre la vermiculitisation des micas et l'orientation des
dipoles 011. Bull. Groupe Franais des Argiles, vol. 18,
p. 3-9.
SCOTT, A. D.; IIUNZIKER, R. R.; IIANWAY, J. J. 1960. Chemical extraction of potassium from soils and micaceous
minerals with solutions containing sodium tetraphenyl
boron. Proc. Soil Science Soc. Amer., vol. 24, p. 191-4.
.A. 1969. O n the
SHUTOV, V. D.;DRITS,V. A.; SAKIIAROV,
mechanism of a postsedimentary transformation of montmorillonite into hydromica. Proc. Int. Clay Conf., vol. 1,
p. 523-32. Tokyo, Israel Uiiiversities Press.
SIEFFERMAN, G.; MILLOT,
G. 1969. Equatorial and tropical
weathering of recent basalts from Cameroun : allophanes,
halloysites, metahalloysites, kaolinite and gibbsite. Proc.
Int. Clay Conf., vol. 1, p. 417-30. Tokyo, Israel Universities Press.
STEVENS, R. E.; CARRON,XI. K. 1948. A simple field test for
distinguishingminerals by alrasion PII. Amer. Min., vol. 33,
p. 31-49.
TARDY,Y. 1969. Gochimie des altrations. Etude des arnes
et des eaux de quelques massifs cristallins d'Europe et d'Afrique. Ph.D. Thesis, Univ. Strasbourg, 272 p.
TCHOUBAR,
C. 1965. Formation de kaolinite partir d'albite
altre 200 C. Etude en microscopie et diffraction lectroniques. Bull. Soc. Fran. Min. Crist., vol. LXXVIII, p. 483518.
-;
OBERLIN,A. 1963. Altration de l'albite par action de
l'eau. Etude en microscopie et microdiffraction lectroniques
de la prcipitation et de l'volution des fibres de boehmite
forme. J. de Microscopie, vol. 2, p. 415-32.
TOUILLAUX, R.; FRIPIAT,J. J. 1966. Caractrisation des
micas par rsonance magntique protonique. Magnetic
resonance and relaxation. Actes XIVh Coll. Ampre,
p. 425-29.Ljubjana, Blinc.
TRANVINIIAN, J.; HERBILLON,A. 1966. Etude exprimentale du systme SiO,-Fe,Ol. Applications la pdogense.
Trans. Conf. on Mediterranean Soils, Madrid, p. 255-68.
Sociedad Espaola de Ciencia del Suelo.
TRICHET, J. 1969. Contribution l'tude de l'altration exprimentale des verres volcaniques. Ph.D. Thesis, Paris, 232 p.

23

J. J. Fripiat and'A. J. IIerbillon

UYTTERIIOEVEN,
J. D.: JACOBS, P.; MAKAY,K.; SCIIOON- WENTWORTH.
S. A. 1967. A n inuestigaiion offine grained micas
PEYDT. R. 1968. The thermal stability of hydroxyl groups
with emphasis on their hydrous charucter. Ph.D. Thesis,
in decationated zeolites X and Y. J. Phys. Chem., vol. 72,
Pennsylvania State University, 89 p.
p. 1768-75.
WEY,R.; SIFFERT,B. 1961. Ractions de la silice monomolculaire avec les ions Al3+ et Mg*+. Gense et synthse
VAN DEN HEUVEL.
R. C. 1966. The occurence of sepiolite and
attapulgite in the calcareous zone of a soil near Las Cruces
des argiles, p. 1-23. Paris, Coll. Int. CNRS. (No. 105.)
(New Mexico). Clayi and clay minerals, Proc, 14th Conf.,
WHITE,
J. L.; BURNS,A. F. 1963. Infrared spectra of hydrop. 193-207.

nium ion in micaceous minerals. Science, vol. 141, p. 800-1.

WALKER,
G.F. 1949. The decomposition of biotite in soila.
Min. Mag.,vol. 28, p. 693-793.

Anomalous water. Nature (Lond.), vol. 222, p. 159-61.

24

WYLLIS,
E.: RENNIE,
G. K.;SMART,C.; PETIIICA,
B. A. 1969.

2.Metallic oxides and hydroxides in soils

of the warm and humid areas of the world:
formation, identification, evolution

P. Segalen
Services Scientifiques Centraux de I'ORSTOM,
93 Bondy (France)

INTRODUCTION
Over large areas of the world, one can see that soils
are bright coloured, with red or yellow hues (both
colours can exist at different levels in the same profile).
They contain appreciable amounts of iron oxides and/or
hydroxides, and sometimes aluminium hydroxides,
manganese and titanium oxides. Sometimes, but always
locally, oxides and hydroxides of cobalt, chromium and
nickel.
A glance at a world m a p (Gamsen and IIdrich, 1965)
shows that these coloured soils are located mostly
between the tropics, but that some of them can be
found north and south of these limits. They all belong
to equatorial, tropical, subtropical, mediterranean
climatic rgimes and are seldom found in cool climates;
they are completely lacking in the deserts. They can
be found mostly where the climate is wet and w a r m ;
they are seldom found (and inherited from past climatic
conditions) in the deserts, but can be found isolated
in subtropical and mediterranean zones ; they become
unusual with altitude as well as in the cool temperate
zone. On account of past climatic conditions, it does not
appear possible to ascertain the limits of their present
genesis; all we can do is to give the actual geographic
limits.
Thus it is for the equatorial, tropical,subtropical and
mediterranean zones that I a m going to try to discuss
the formation of metallic oxides and hydroxides. This
great area, which is split u p in the northern part of
Africa by the Sahara, presents no interruption in southern Africa or in South-East Asia, for instance, where
climatic conditions change gradually from one type to
another. T h e distribution of these zones is given in
Figure 1.
O n e can see that this distribution lias been modified
by the forms of continents, the relative position of
shorelines and winds (trade winds and monsoons) and

'

cold or w a r m currents. T h e height and direction of the

main mountain ranges, the relief in general, all have
consequential effects on rainfall and temperature which
can be greatly modified.
After this quick and unavoidably hasty survey of
the climatic conditions, let us sec the materials which
rainfall and temperature are liable to affect. Ail parent
materials are present and, with some remarkable
exceptions, the main igneous volcanic, metamorphic
and sedimentary rocks are present but are not spread
in any sort of distribution; they are closely related
with geomorphology. There are roughly two major
types of relief: mountain ridges and islands on the
one hand, plains and plateaux on the other. Consequently, these differences have important effects o n
drainage.
Considered as a whole, Africa, Europe and Asia are
crossed by an almost continuous mountain range. In
Africa and western Asia, the climate is mediterranean ;
in eastern Asia and Indonesia, the climate is tropical
and equatorial. These m a k e u p the northern boundary
of the area under study. On the other hand, in America,
mountain ridges run from north to south on the western
rim of the continents. Generally speaking, they are
composed of marine sediments (mostly limestone and
marls) on which large masses of volcanic materials have
piled up, especially in America, Indonesia and Japan.
Ultra basic rocks are k n o w n in such places as Cuba,
the Philippines, New Caledonia.
The remainder of the continents consists of flat
stretches joining plateaux seldom more than 1,000
metres high, and low-level plains. Thus, in South
America, plateaux of Brazil and Guyana together with
the A m a z o n and the Parana plains; in Africa, different
plateaux from Guinea to Angola and the plains of the
. Niger, Chad, Congo, Sutld, etc.; in India, the D e k k a n
and the Ganges plain. All these plateaux are often
bevelled through eruptive or metamorphic rocks and

25

P. Segalen

26

*I

I
I

I
I
I
I
I

. >

* i

Metallic oxides and hydroxides in soils of the warm and humid areas

capped with powerful lava flows. Sedimentary rocks

are mostly detritic and of continental origin; marine
sediments are u n c o m m o n except on the outskirts.
These plateaux have undergone long erosion cycles
and presumably soil-forming factors have been in
action for very long periods. This is, to be sure, one of
the main features of these plateaux. Moreover, the
flexible tectonics which affect the mountain areas are
here completely lacking. O n the other hand, faults
with large-sized grabens are common, as in eastern
Africa. While the temperate zones (in the north as
well as in the south) are strongly marked with Quaternary glaciations, nothing similar can be seen in this
area. N o glacial tills or outwash, no loess is encountered.
In temperate zones, soils are rather young and develop
under mild climatic conditions.In warmer and variously
humid areas, weathering carries o n for ages on materials
submitted to drastic climatic conditions.
T h e vegetation which corresponds with such a variety
of situations is itself quite diversified. Rainfall is the
main factor of diversification, since, except for the top
of the mountains, temperature remains always fairly
high. Rain, pouring d o w n heavily or dropping sparingly,
is responsible for the main vegetation features. Forest
cover is most c o m m o n and varies with hardly perceptible tinges due to gradual climatic changes up to the
deserts where all vegetation comes to a n end. T h e
primary w o o d y formations are very often associated
with secondary ones such as savannahs, steppes, maquis
and so on, which are due to h u m a n activities (such as
clearing the land to bring it into cultivation, repeated
bush fires which steadily eat away the forests).
Except for extremely arid zones, w a r m and humid
areas all have in c o m m o n soils containing iron oxides
or hydroxides; manganese and titanium oxides are
always present, but in smaller amounts. In wetter
areas, in good drainage conditions, aluminium hydroxides are met; accumulation of other metallic oxides is
associated with ultrabasic rocks.
Along with these oxides and hydroxides, clay minerals
containing aluminium, iron, magnesium and various
alkaline metals, can be formed, transformed or accumulated. T h e presence of ferruginous products is
a feature c o m m o n to the whole zone. But important divisions are provided by the clay minerals.
Hence, three provisional divisions can be set out as
follows.
Presence of kaolinite together or not with aluminium
hydroxides. Small amounts of illite can be encountered but no other three-layered clay mineral. All
the ferruginousmaterials are outside the clay lattices.
This is true for ferraIlitic soils or oxisols, ferruginous
tropical soils and some hydromorphic soils.
Kaolinite is associated with different three-layered
clay minerals (especially montmorillonite). Ferruginous materials are L o ~ h
inside ant1 outside the clay
mineral lattice. Soils are fersiallitric (mediterranean
red and brown soils), semi-arid reddish-brown soils,

subtropical chesnut soils, red-yellow podzolic soils

and so on.
3. Three-layered clay minerals are dominant. Iron is
mostly inside the clay lattice and seldom outside.
This concerns vertisols associated with s o m e of the
above-mentioned soils.

PART O N E
After an outline of the general distribution of hydroxides
and oxides in soils and the climate, parent-rock, geomorphology, vegetation and time conditions that
prevail in their formation, I shall n o w try to go into
further details about the minerals which provide t h e m
and their weathering, and next describe the main
crystallized and amorphous minerals and the w a y to
determine them.
ORIGIN O F M E T A L L I C O X I D E S A N D
IIYDROXIDES, R O C K S A N D M I N E R A L S

Oxides and hydroxides which are abundant in m a n y

soils of the area under study originate in the weathering
of the minerals of which the rocks are composed. Tbese
minerals belong to various families, not all of which
can be studied bere. Only those which have a n actual
pedogenetic interest and are met all over the world will
be taken into account here.
Igneous rocks are m a d e u p essentially of iron, aluminium, magnesium, potassium, sodium and calcium
silicates in various proportions and in different structures. S o m e other minerals containing the abovementioned metals are present but their importance is
slight.
W h e n they undergo weathering, these minerals are
subject to important or minor transformations. T h e
newly created products can stay where they are, be
displaced by winds, waters or glaciers. T h e y can form
n e w materials on continents, in lakes, in seas and form
sedimentary rocks. All these materials, coming from
the depths as well as from the surface, are liable to
weathering and can be considered as starting points
for oxides and hydroxides.
It is customary to call primary minerals those
which form in the depths of the earth and are the m a i n
constituents of rocks and secondary minerals those
which derive from the former under weathering processes.
Amongst primary miiierals, let us mention : peridots,
garnets, pyroxenes, amphiboles, micas and feldspars.
Their structure is quite elaborate. T h e three main
components are the tetrahedra, the centre of which is
occupied mostly by silicon, sometimes by aluminium;
the octahedra, the centre of which is occupictl by a
metal, such as aluminium, iron, magnesium; the
hexagonal prisms with potassium in the centre.
I

27

I.

Segalen

Peridots are magnesium and ferrous iron ortliosilicates. Silicon tetrahedra do not join by a summit
but are connected through hexacoordinated m a g nesiums. Weathering of these minerals leads to the
elimination of silicon and magnesium, while ferrous
iron oxidizes and hydrolizes to ferric hydroxides. It is
worth pointing out that, in tropical regions, weathering
of peridot mountain masses leads to the accumulation
of iron products. This happens near Conakry in Guinea,
in N e w Caledonia, near Surigao in the Philippines.
Pyroxenes and amphiboles are m a d e of silicon tetrahedra joining through one or two apexes, to constitute
single or double lines. All the tetrahedra are set out the
s a m e way: three apexes are in the same plane and the
fourth juts out always on the same side, and the
tetrahedra line up opposite one another through this
fourth apex. Hexacoordinate metals such as ferrous
iron or magnesium or calcium link the lines together.
Weathering of these minerals leads to the loss of alkaline metals, magnesium and silica when drainage
conditions are good. Residual iron accumulates as
oxides or hydroxides. W h e n drainage conditions are
not so good, clay minerals m a y form. Aluminium,
unless drainage conditions are good, favours clay
minerals. Titaniferous augites release titanium to form
anatase.
In micas, silicon tetrahedra join through three
apexes to form indefinite planes. T w o of these planes
(where substitution of silicon by aluminium is important)
are kept apart by an octahedral layer; the metals are
magnesium and ferrous iron in biotite, magnesium in
phlogopite, aluminium in muscovite. The hexagonal
pits of the tetrahedral layers enclose potassium ions
which hold together two mica layers. The weathering
of these minerals leads to clay minerals, such as illite
or vermiculite but also to iron oxides and hydroxides.
In feldspars, silicon tetrahedra join through the four
apexes and are distributed in the three dimensions of
space. O n e in every four tetrahedra is centred on
aluminium instead of silicon. Alkaline or earth-alkaline
ions are located between the tetrahedra. Weathering
leads to clay minerals and/or to aluminium hydroxides.
Clay minerals as well as primary minerals are liable
to lead to oxides or hydroxides as far as iron and
aluminium are concerned. This is true for three-layered
minerals as well as two-layered minerals like kaolinites.
All rocks are liable, in w a r m and humid areas, to
supply soils with iron, aluminium and titanium material. Ultrabasic rocks are starting points for ferruginous
products but also for chromium, cobalt nickel and
manganese ones. Basic rocks, like basalt or gabbro,
lead to iron hydroxides from ferromagnesian minerals,
to titanium oxides from augite,to aluminium hydroxides
from feldspars and other aluminous minerals (sometimes
through clay minerals). It must h e pointed out here
that plenty of magnetite is found in these rocks anil
proceeds unchanged into the soil where it acts as sand.
Acid rocks (rhyolites and trachytes, granites and

28

syenites, gneiss and some other metamorphic rocks)

enclose few ferromagnesian minerals, and soils have
often small amounts of iron oxides. However, this
must not be considered as a rule, as some soils derived
from granites in very humid areas can reach 10 to
20 per cent iron oxide. Feldspars are starting points
for clay minerals and aluminium hydroxides. Titanium
contents are usually low. Ilocks containing m u c h quartz
(such as sandstones or quartzite) cannot propote
aluminium hydroxides very easily, since plenty of silica
is present. But iron oxides can form and coat sand
grains or grow into concretions. Aluminous and ferruginous products can originate only from non-quartzose
grains and these exist in small amounts (few ferromagnesians, feldspars or micas).
As for limestones, the matter is somewhat different.
During the pedogenesis, quartz is hardly eliminated,
leaving an abundant siliceous residue in the soil; on the
other hand, calcium carbonate can be washed a w a y
completely or partially. The soil develops, deriving.
from the impurities of the limestone, clay minerals or
primary minerals. This prevails under tropical (Central
America, Mexico) or mediterranean conditions (surroundings of the Mediterranean Sea); great amounts of
calcium carbonate must be got rid of before the soil
develops. This sets the problem of the genesis of
bauxites connected with limestones; but it cannot be
discussed here at length. Let us recall briefly that quite
different opinions have been set forth about the formation of these bauxites. S o m e authors state that aluminium and iron hydroxides originate through the
hydrolysis of all the impurities of limestone rocks after
the dissolution of carbonates. But some limestones are
very pure, devoid of titanium oxide whereas the bauxite
has a high percentage of this oxide. Hence this explanation has been refused by m a n y people. S o m e think
that bauxites derive through weathering from other
materials than those they are found upon. They think
that some material rich in iron and aluminium (as a
volcanic ash) was scattered over the limestone and that
the observed bauxite is nothing but the product of the
weathering of the ash, and not of the limestone. It can
also be considered that bauxite formed elsewhere on
another rock from which erosion tore it a w a y (through
the action of running waters or wind) and trapped it
o n limestone. Each case must be examined carefully,
as there is probably n o unique explanation.
W E AT11 E RING

Except for very few rocks and minerals, all are liable
to act as starting points for the oxides or hydroxides
ound in soils.
T h e principal process is hydrolysis. Water contains
not only IIOII molecules but 011- and II+ ions,
following the dissociation of a small number of mole1. Or, better, IIsO+.

Metallic oxides and hydroxides in soils of the warm and humid areas

cules. These ions are enough to ensure the breakdown

of primary minerals and modify or destroy the original
structures. They are responsible for the carrying a w a y
of alkaline or earth-alkalineions. In even more minerals,
these metais are replaced by hydrogens. W h e r e rather
large-sized ions are replaced by m u c h smaller ones,
structures are weakened and tend to collapse.
The Si-O-Allinking is affected, leading to the separation of aluminium and silicon. If the latter is carried
away, aluminium hydroxide precipitates. This happens
for feldspars, but white micas are m u c h less concerned
as they are resistant to Weathering. Clay minerals, and
especially three-layered ones, can also break d o w n by
weathering.
This hydrolysis leads to the separation of aluminium
and one can ponder on the fate of this metal. If pH is
lower than 4.2, the ion can subsist and is liable to m o v e
but most probably for short distances. If p H is higher
than 4.2, a hydroxide will precipitate. The newly
formed material is very often amorphous. If it remains
surrounded by anions crystallization is slow and some
forms, like pseudo-boehmite,m a y appear. If drainage is
good, and if anions are washed away, gibbsite crystallizes directly or through an amorphous phase. In
m e d i u m drainage conditions, aluminium m a y combine
anew with silica to form clay minerals ; two-layered
minerals if there is not too m u c h silica and bases;
three-layeredminerals if silica and bases are abundant.
In m a n y primary minerals, iron is in a ferrous form.
W h e n weathering begins, water, which is responsible
for the alterations, brings into the soil oxygen in
solution. This oxygen is liable to oxidize the ferrous
forms into ferric ones. At a pH lower than 2.5, ferric
ions m a y subsist, but such acid soils seldom occur;
at a higher pli, a hydroxide precipitates. As in an acid
medium, iron offers no affinity for silica, the hydroxide
remains as it is, or gradually transforms into goethite
or hematite. If the m e d i u m is a reducing one, by
absence of oxygen or especially by the introduction of
organic material, the ferrous ion m a y subsist u p to
pH 6.5 or higher in very dilute solutions. This facilitates
the movements of iron.
'As for manganese, the problem is more complex,
since, instead of two valences as for iron, three valences
are k n o w n in the soil. Divalent manganous ions are
stable up to p H 7.5; at higher values, a hydroxide
appears which will convert into a tetravalent bioxide.
Accordingly, iron and manganese, separated from
primary minerals as ions, are easily oxidated and will
develop in higher valence forms as oxides or hydroxides.
Titanium separated from primary structures cannot
exist as an ion above pH 2.2. A hydroxide precipitates
which readily converts to an oxide. The redox potentials
observed in soils are not low enough to allow reduction
of tetravalent titanium (Fig.2).
Organic materials are also responsible for the tearing
off of metals from primary structures. Litter on the
top of organic horizons is a source of soluble products

such as acids, aldehydes, phenols, etc., liable to attack

minerals and extract metals, especially iron and aluminium. Under poor drainage conditions, these organic
substances m a y be considered as responsible for the
movement of iron in m a n y soils. (Alexandrova, 1954;
Uetremieux, 1951 ;Uloomfield,1956; Schatz, 1954.)
TIIE P R O D U C T S O F TIIE SOIL

Soil sesquioxides m a y be divided into two main categories : amorphous and crystallized products. Both
exist in the soils under study.
A m o r p h o u s products

A definition of amorphous products is necessary. T h e y

do not have a definite formula. Various names have
been proposed such as cliachite or alumogel for aluminium, stilpnosiderite for iron, doelterite for titaninium,
vernadite for manganese, etc. They are variously
hydrated and this water is easily heated a w a y at low
temperatures. Their specific area is usually high
(300-400m2/gfor amorphous, ten times less for crystallized material). As there is no crystalline,system,
X-ray diffraction provides only very w e a k peaks,
and no species can be identified. Amorphous products
are very seldom alone in soils. S o m e soils derived from
basic rocks contain up to 25 or 40 per cent amorphous
material.
Crystallized material

Several types are k n o w n such as hydroxides, M(OH),

or M O O H and oxides R120,. Very close relationships
exist between some aluminium and iron materials.
Only one hydroxide with formula BI(OII), exists in
soils: gibbsite AI(OH),; there is D O iron equivalent.
Gibbsite is one of the basic constituents of the soils of
w a r m and humid regions; its colour is white and it is
seldom sufficiently crystallized to be seen without the
microscope; its solubility is very low. Other hydroxides
having the same formula are known. Bayerite is
obtained in some synthetic preparations and during
industrial processes. It develops easily to gibbsite by
ageing, and has not been noted in soils. Norstrandite
was discovered a few years ago; its presence in soils is
k n o w n but needs confirmation. MOOII hydroxides are
boehmite and diaspore. Both are constituents of bauxites
more than of soils. The former alone has been identified
in soils but not so frequently afi gibbsite. Boehmite is
a white mineral which can sometimes be recognized
under the microscope. T h e structure of these minerals
is n o w well known. Aluminium occupies the centre
of octahedra distributed in planes according to three
different models. There are two dikrent iron hydroxides
with the same formula of FeOOH :goethite and lepidocroci te. Goethite is a reddish-yellowmineral, often poorly
crystallized, a commonplace product in the soils of

29

P. $egalen

the area considered. Lepidocrocile, also reddish yellow,

is met only in some hydromorphic soils, seldom in
others. Goethite and diaspore have the same structure;
so have lepidocrocite and boehmite.
oxides, or sesquioxides, are represented by
T h e M203
corundum A120, and hematite and maghemite Fe20,.
Corundum' exists in soils only when inherited from
metamorphic rocks. Hematite is commonplace in tropical soils. It is red, non-magnetic and can be found
in soils derived from volcanic rocks, in'.areaswhere a
severe drought prevails during the major part of the
year. Maghemite is a magnetic oxide. It derives from
lepidocrocite by dehydration or from magnetite through
oxidation. Maghemite is seldom met in soils and occurs
especially in soils derived from volcanic rocks.
. Let us bear in mind the magnetite F
e,O,, which has a
spinel structure, must be considered as inherited from
basic parent material, but not as a soil-formed oxide.
Signijcance of hydroxides in soils

Gibbsite is fairly widespread in equatorial and humid

tropical areas; it is sometimes encountered in the
Mediterranean area. It is u n c o m m o n in very dry places.
Its presence appears to be related first of all to a humid
climate. Drainage also plays a prominent part in the
occurrence of the mineral. Gibbsite is c o m m o n in soils
developed o n broken land, even w h e n rainfall is not
too heavy, whereas in flat, level central Africa, even
under heavy rains (e.g., in the southern Cameroons,
in the Gabon, in the Congo) soils contain kaolinite
rather than gibbsite. Finally some rocks, like basalt,
contain easily weatherable minerals with a low silica
content and lead quickly to gibbsite. On the other hand,
flat but poorly drained land, containing silica-rich
minerals, shows low contents in gibbsite. Boehmite is
less c o m m o n in soils of the area under review. It is
sometimes identified in soils derived from limestone
and in bauxites related with limestones. Recent laboratory work has led to the conclusion that gibbsite is
stable under low pressure and ordinary temperature
conditions. Boehmite is stable above 150OC. It is
possible that lateritic crusts, which undergo a rapid
increase in temperature, are in favourable conditions
for a change of gibbsite into boehmite. Bauxites related
to limestones (in western or central Europe) have
been concerned with alpine folding and have undergone
increasing temperatures and pressures leading to the
transformation of gibbsite into boehmite. Bauxites
concerned by m u c h older folding (e.g., in the Urals,
in the Appalachian mountains, and in Turkey), with
still higher temperature and pressure, have diaspore
instead of gibbsite.
Goethite is widespread in all the soils. It exists alone
in usually wet soils (but not hydromorphic) of equatorial
regions. Elsewhere, the yellowish colour indicates,
amongst red soils, a wetter area. Hematite is usually
present in soils under strongly contrasted climates.

30

Amorphous materials are present in m a n y soils.

Very oien red soils depend on them for their bright

colour. O n the other hand, they lack completely in
most yellow soils.
DETERMINATION OF OXIDES
A N D IIYDROXIDES

A number of difficulties must be overcome when one

wants to ascertain the nature and amount of sesquioxides in soils. First of all, between amorphous and
crystallized material, the main difference, as far as
chemical composition is concerned, is due to a higher
water content. Usually not just one or two components
exist together, but m a n y more and of different nature.
S o m e of them behave the same w a y w h e n they undergo
chemical or thermal treatments. Finally, the products
are very small and special techniques are necessary.
Since the beginning of studies on soils, chemical
analysis has been relied upon as a means of providing
useful information. Alkaline fusion has been and still
is in use, but it does not differentiate between minerals.
This method was found too drastic and another, using
a mixture of three acids, was recommended some time
ago and used by Lacroix (1926), Harrison (1933) and
m a n y others since. T h e use of this mixture was generalized, at least for the soils of tropical areas. Unweathered primary minerals, like quartz, m a y be
separated from most soil-formed minerals. This procedure gives a fairly good estimation of the silica/alumina
ratio; moreover, it indicates if there is more aluminium
than is necessary for kaolinite. For a correct estimation
of alumina, it is necessary to k n o w if no clay mineral
other than kaolinite is present, but this is not easy
using chemical methods only. According to water
content, one can differentiate gibbsite and boehmite.
With regard ta ferruginous products, they can be determined correctly except if three-layered clay minerals,
very often iron-rich,are present. The difference between
hematite and goethite can be found on a water-content
basis. Minerals such as ilmenite or magnetite are more
or less attacked by the reagents and most of the iron
is liable to be attributed to the soil. Manganese and
titanium can be determined easily but it is somewhat
difficult to tell which product they belong to. Therefarc
the method is liable to provide valid information on
the centesimal composition but not m u c h as far as
identification of constituents is concerned. That is why
methods leading to the selective identification of products or groups of products w e r e looked for. Thus,
the determination of oxides and hydroxides of
alumina and iron were for a long period the aim of soil
chemists.
Free alumina has been considered for some time as
an unmistakable test of ferrallitization, and efforts
were m a d e to determine it quickly. Dissolution of
alumina in sodium hydroxide, since the oxide is amphoteric, has been recommended. In addition alumina

Metallic oxides and hydroxides in soils of the warm and humid areas

reacts with some dyestuffs like sulphonated alizarine.

But this dye reacts with other boil products and quantitative determination is not possible.
The determination of free iron oxides has given ribe
to m u c h work in the quest of reagents able to dissolve
these products readily. It is worth recalling that free
iron oxides stand only for soil oxides and hydroxides,
while primary magnetite and iron-holding clay micerals
such as nontronite are excluded. T h e first widespread
reagent was that suggested by T a m m (1922), i.e., a
mixture of oxalate and oxalic acid. As it dissolves little
iron, it was abandoned soon after but has regained
some interest in these last years. It is now k n o w n that
it does dissolve all free iron in soils like podzols, but not
in tropical soils. Drosdoff and Truog (1935)recommend
hydrogen sulphide, m u c h more efficient than the former
reagent. The acid was used in work on red lateric soils
and red-yellow podzolic soils of the south-eastern
united States and the results were good. But the
reagent was not generalized in soil studies, perhaps on
account of its nasty smell and toxicity. Dion (1944),
Jeffries (1946) have used nascent hydrogen provided
by the action of an acid on a metal coil, aluminium or
magnesium. But the procedure recommended by
Deb (1950) using sodium dithionite rapidly became
very popular. The reagent is easy to manipulate and
results are obtained quickly. By the addition of buffers
and complexing agents, a number of alterations of the
original procedure have been proposed. Finally, quite
a different technique using ultra-violet irradiation on
oxalic solutions was proposed by De Endredy (1963).
T h e value of all these techniques is variable. It is
necessary to test them and compare them with others
of a quite different nature before good results can be
expected.
Thermal analysis is based upon the variations undergone by a sample as it is heated. T w o ways are possible :
pondera1 thermal analysis and differential thermal
analysis. The former makes it possible to follow weight
losses of a sample while it is heated. These losses m a y
be due to organic matter combustion, decomposition
of carbonates, or oxidation of sulphides; in the products
we are concerned with, the loss is of water. T w o kinds
of water can be determined. The first goes off at low
temperature (less than 1500 C), and is related with the
amorphous material; unfortunately at the same temperature, water held between clay layers is driven off.
T h e second goes off at a definite temperature and is
related with the constitution of molecules. T h e amount
lost can be measured accurately. Constitution water of
goethite or gibbsite can be measured, but both lose this
water at about the same temperature. Oxides, of course,
cannot be determined this way. Differential thermal
analysis is based on the fact that a measurable electric
current is produced between thermocouples connected
together and heated to different temperatures. In order
to study oxides and hydroxides, one of the thermocouples is put in the u n k n o w n product. During the

heating, endothermic reactions (absorption of heat

necessary to hive a w a y the water) or exothermic
reactions (liberation of heat due to internal reactions)
take place. The second thermocouple is placed in an
inert product. The electric current produced between
the couples owing to the diKerent temperatures tleveloped during the reactions, is recorded on a graph
which permits the identification of the constituents.
Water losses of goethite and gibbsite give rise to endothermic reactions which produce peaks on diagrams.
Unfortunately these peaks are often superimposed.
The change of maghemite or lepidocrocite in hematite
produces an exothermic reaction which is expressed
in peaks opposite to the former. This technique gives
good qualitatkc information, but does not yield
quantitative data.
X - r a y diflraction is the safest method to idcntify
soil oxides or hydroxides. The diffraction of a b e a m
of X-rays by a powder containing the materials produces a diagram in which a number of distinctive
peaks can be recognized. However, where several
minerals arc present, peaks m a p overlap and identification is not so easy. Peaks of low intensity are used,
in this case, rather than strong ones. As for identification
of gibbsite or goethite, no particular diculty arises.
But hematite and maghemite are not so easy to recognize and need a sensitive X-ray set. If large amounts
of clay minerals are present, small amounts of oxides
or hydroxides are not easily identified.
Infra-red tibsorption pholometry. Few data on the
absorption of infra-red radiations are yet available.
But, this technique is certainly likely to provide very
interesting information.
T h e identification of constituents by means of
microscopic methods is unequally successful. T h e cutting
of thin sections, and the small size of the crystals have
long been a difficulty. However, the present setting
techniques using plastics have allowed a rapid development of micromorphology investigations. Furthermore,
electron-microscopy gives access to the study of shape
and arrangement of small particles, tenths of a micron
in size.
Determination of amorphous material. This material
has long been difficult to identify and study owing to
the lack of definite structure and formula. Thermic
diagrams exhibit very strong endothermic peaks at
low temperatures due to important losses of water.
Dut this happens also for montmorillonitic clay minerals.
X-ray ciiagrams display no characteristic peak.However,
if amorphous material is cleared away, peaks of crystallized products which had been obliterated can appear
(Tardy and Gac, 1968). A suitable process can be followed to dissolve and determine amorphous material.
T a m m s oxalic reagent m a y prove satisfactory; but a
technique using in turn an acid (8N hydrochloric acid)
and an alkaline (0.5 N sodium hydroxide) can exhaust
a soil of its alumina, ferric and silicic amorphous
products (Segalen, 1968). X-ray controls show that

31

P. Segalen

crystallized products are not disturbed by this procedure.

INFLUENCE OF OXIDES A N D
O N S O M E SOIL P R O P E R T I E S

IIYDROXIDES

This part

concerns the properties used to characterize

a soil, such as colour, structure or surface, that can be
recognized by sight or by touch.
The colours of the soils of w a r m and humid areas
that are due to oxides or hydroxides, can be reduced
to two, namely yellow and red (brown colours are
usually due to organic matter). These two colours are
due to ferruginous, amorphous or crystallized materiaI.
T h e removal of amorphous material enables the basic
colour to show up: reddish yellow when goethite is
the main crystallized mineral, red when it is hematite,
grey when important amounts of magnetite, greyish
w h e n clay minerals alone are present. Yellow soils do
not change in colour when one tries to dissolve a w a y
amorphous material. This leads to the idea that they
do not contain any; and in such a case, goethite alone
prevails. D a r k colours m a y be due to abundant magnetite. Black colours, that must not always be assigned
to organic matter, m a y be due to manganese bioxide;
greenish or bluish colours are usually due to ferrous
iron. Heating brings about red colours through oxidation. Aluminous and titanous material only induce
light colours. Iron-poor bauxites are pink or greyish.
Structures. M a n y discussions have concerned the
importance of oxides and hydroxides on structures
(MacIntyre, 1956; Meriaux, 1958). It seems that clay
minerals, organic matter, and associated metal ions
are responsible for structure. Little is k n o w n of the
influence of aluminous, titanous and manganeous
material. On the other hand, definite influence is
assigned to iron compounds in structural features. It
seems that the main volumes (prisms, cubes, columns)
are not m u c h changed; but that coarse polyhedric
forms are somewhat affected. However, the formation
of pseudo-sands,m a d e up of crystallized minerals (such
as quartz, magnetite, kaolinite) stuck together by
amorphous material is greatly influenced by iron.
These very small particles are not easily dispersed b y
the usual procedures of mechanical analysis, and it is
necessary to use ultrasonic vibrations to break them
down. These pseudo-particles are not specific to any
soil type. They can be noticed in different climatic
areas and in soils derived from various parent materials
such as sands or basalt.
Surfaces. Amorphous material is generally responsible
for great specific areas. When amorphous products are
dissolved, as well as free iron oxides, specific areas are
greatly reduced.

PART T W O
In this part, the dynamic aspects of the metals in soils
are examined :first of all, the features which enable us

32

to explain how they move; next, h o w they accumulate;

and last, what is expected of the accumulated forms.
In other words, we will consider here what happens to
thesc metals during pedogenesis.

C O N D I T I O N S O F STABILITY A N D
OF METALLIC C O M P O U N D S

MODILITY

A short rekiew of a few properties of ions and hydroxides

is necessaty to understand the conditions under which
they can exist in soils. Conditions prevailing when they
appear will also be considered.
Ions and hydroxides
T h e concentration/pH curve separates zones where ions
and corresponding hydroxides are stable. Figure 2
provides these curves for each element.

Ionic features

The ions corresponding to the metals under investigation are all medium-sized: 0.51 A for Al3+, 0.63 A for
Fe3+, 0.75 A for Fez+, 0.68 A for Ti4+,0.6 A for Mn4+.
Alkaline and earth-alkaline ions are m u c h larger :
0.9 to 1.3 A; and metalloids such as C, Si, N, are m u c h
smaller :0.2 to 0.4 A. T h e oxides of these metals depend
on the ratio:
cation radius

= oxygen radius
This ratio enables us to predict the co-ordination
n u m b e r of the metal, and the sort of structure it will
fit into. T h e elements we are concerned with fit in
octahedra Mn2+, Fez+, Fe3+,Ti4+;Al3+fits in octahedra
and in tetrahedra (Table 1).
TABLE1. Size of ions of principal elements and co-ordination
number'

Ion

(in A)

Co-ordinationnumtirr

= 'Rc
-

Predicted

K+

Found

1.33
0.99
0.97

0.95
0.71
0.69

12
8
8

8-12
6-8
6-8

Ala+

0.80
0.74
0.68
0.64
0.51

0.57
0.53
0.49
0.046
0.36

6
6
6
6
6

6
6
6
6
4-6

Si4+

0.42

0.30

Caa+

Na+
MnZ+

Fez+
Ti4+
Fe3+

1.

1Iexagonal prism is siable when

Cube priim is stable when
Octahedral prism is stable when
Tetrahedral prism is stable when

p 6~ 1.0
P 6~ 0.78

p
p

w 0.41
w 0.22.

Metallic oxides and hydroxides in soils of the warm and humid areas

Concentration

Concentration

Concentration
I.

-2

-2
M10.

M 10

Ti4

Fe 2+

I 3
-8
10
2

lea

1 5

PH

M 1'

lo2

AI^+

MI?+

1 5

M I'
3 4 5 6 7 8 91011

8 9 10 PH

pH

FIG.2. Stability areas of ions and hydroxides (after Charlot).

-0.6

PH

FIG.3. Ferrous-ferric chemical equilibria as function of p H

and redox potentials (after IIem and Cropper).
33

P. Segalen

Solubility of hydroxides

Another interesting characteristic is the ionic potential,

defined as th;? ratio of the charge Z on the ionic radius.
All the elements considered here have an ionic
potential allowing us to situate them a m o n g the
elements of low solubility (2 to 7); whereas soluble
cations have a very low potential (<2.0) and complex
All the hydroxides
anions have a m u c h higher one (9-40).
w e are dealing with have a very low solubility product
(10-15 to 10-39). Consequently, except for very acid or
very alkaline conditions, it appears highly improbable
that any mobility of the above-mentioned products can
be suggested when they are in the hydroxide form.
Once they are precipitated it seems very difficult for
water alone to be responsible for any solution and
movement. Recent work b y Herbillon and Gastuche (1962), shows that w h e n iron or aluminum hydroxides precipitate, interesting developments take place,
if at the same time,one proceeds to an effective dialysis.
A n amorphous product precipitates first, followed soon
by a depolymerization which looks very m u c h like a
solution;little b y little a crystalline phase (trihydroxide)
appears. This can be related with what happens in
s o m e soils of humid tropical areas where water is so
abundant that few ions are present in solution. So far,
the hydroxides behave roughly in the same way, but
s o m e differences appear when redox potentials are
concerned.
Redox potentials

Variations of redox potentials are duc, in soils, to the

presence of reducing organic material, and to the
absence of oxygen in soils and waters, related to permanent or temporary high water tables. The variations
of redox potentials concern some elements (like iron or
manganese but not aluminium). But their effect depends
on the reaction of the medium. At low pH, reduction
is very easy; at high pH,stronger redox potentials must
operate for the same result (Fig. 3). Iron and m a n ganese behave in very m u c h the same way. Iron
proceeds from the ferric to the ferrous form very
easily in moderately acid conditions but not so in neutral
or alkaline ones. Therefore, in tropical acid soils, under
poor drainage conditions, ferrous iron is easily obtained
and this is liable to explain the mobility of this element
every time even slightly reducing conditions occur. The
oxidation or reduction reactions are the following :

Fes+ -f Fe3+ + E
Fes+ + 1/2II,

Fe2++ II+.

(1)
(2)

Complexation and chelation

S o m e elements can belong to bulky molecules where

they no longer have the same properties as they had
w h e n they were ions (Martel1 and Calvin, 1959). These

elements are called electron acceptors and are liable

to fix on their outer layer a number of electrons provided
by other molecules called electron donors . Acceptors
are metals, such as iron or aluminum, donors arc
oxygen or nitrogen. Bulky molecules are constituted in
which the metal is surrounded by a number of oxygens
and belongs then to an anion liable to be solubilized
in conditions very different to those of the central atom.
If the elements of the donor are not bound together,
a complex is formed (such as potassium ferrocyanid or
cryolite). If the elements are bound together a chelate
is formed. Whenever there is a bulky organic molecule
with several functions where oxygens, oxhydrils,
amins, amids.. . are close to one another (e.g., oxalic,
citric acids, phenols, EDTA, etc.), chelates can form.
Metals are then considered to be held in a claw (chela).
T h e newly formed anion is liable to m o v e in pH conditions when hydroxides usually precipitate. T h e synthesis of chelating substances in the upper part of soils
is n o w generally admitted but their actual presence is
not always demonstrated, as it is in podzols. In the
laboratory, it has been shown that iron could be
displaced easily with the help of complexing agents.
In soils of the w a r m and humid arca, where mineralization of organic matter is rather fast, chelating agents
are not liable to exist very long. So their influence
must be taken into account only w h e n one can show
they exist.
Lastly, one must consider the possibility of displacement of very minute particles, b y a simple mechanical
washing away, T h e size of particles of oxides and
hydroxides is very small (less than one micron), m u c h
smaller than the smallest pore space of soils. But they
are very often bound to the clay minerals and m o v e
along with them. Migrations can take place inside or
outside the profile.
IMMOBILIZATION O F O X I D E S
A N D H Y D R O X I D E S I N SOILS

I shall try in this section to explain how, after having

been solubilized as ions, metals assume one of the
various aspects k n o w n in soils.
First of all, what happens to aluminum, iron, titanium, manganese, when the primary minerals which
contain them are broken d o w n b y weathering?
A crystallized mineral m a y be formed right away,
in the place of a mineral, the other constituents of which
have been entirely eliminated. Goethite or anatase
m a y crystallize in place of a plagioclase (Millot, 1964).
These crystals m a y be observed with the help of a
polarizing microscope or an electron microscope.
Sometimes the crystals are large enough to be seen
without a microscope (Lacroix, 1926, Madagascar). But
crystals m a y not develop largely enough to be identified
and amorphous material is present. X-rays are unable
to show any geometric pattern. These amorphous

Metallic oxides and hydroxides in soils of the warm and humid areas

materials are very largely distributed in some of the

soils considered in this study.
T h e metals take their part in the synthesis of n e w
products :clay minerals such as kaolinite or halloysite
(aluminium), montmorillonite (aluminum and iron),
depending o n the characteristics of the surroundings.
How are the oxides and hydroxides distributed in
soils? A relative accumulation, by exportation of
soluble material (alkaline, earth-alkaline bases, silica),
m a y occur. This concerns mainly aluminum and titan i u m which are not easily mobilized, and also iron in
m a n y situations. An absolute accumulation m a y also
occur, due to migrations inside the profile or by introduction in the profile of compounds coming from
elsewhere. This concerns iron and manganese which
are very sensitive to variations of redox potential
(DIIoore, 1954). Oxides and hydroxides concentrate
in different ways :
First, they deposit on clay minerals. This was shown
long ago by Barbier (1936) and confirmed by Fripiat
et al. (1952) and recently by Follett (1965). Products
are distributed in very minute particles along the flat
planes of kaolinite and can be seen o n electron microscope photographs.
Local concentrations of ferruginous or aluminous
material occur with or without concentric structures.
O n e can then have a series of forms differing by their
external or internal appearance or by their contents
in one or several components. These elements seldom
exist alone and e m b o d y grains of quartz or magnetite,
clay minerals like kaolinite, and unweathered rock
material. These concentrations are usually from one
to a few centimetres wide. Their origin has given rise
to m u c h discussion and probably varies from one sample
to another. T h e y exist as nodules, concretions, accumulations and so on. Small concentrations of sand (quartz
or magnetite), of well-crystallized clay minerals (kaolinite) exist, bound together with amorphous products;
they are n a m e d pseudo-sands. T h e usual mechanical
analysis techniques are unable to disperse these small
aggregates, so the sand fraction is unusually high. T h e
use of ultrasonic vibrations is necessary to break t h e m
down.
These accumulations can have a m u c h larger extent.
T h e oxide or hydroxide-enriched zones interest a whole
horizon and concern soils occupying large areas (Aubert,
1963). Cuirasse1 then forms with variable composition
or morphology. T h e y m a y be platy, vesicular, massive, etc., and result in binding of coarse sand grains,
quartz pebbles, rock or cuirasse residues, by oxides
and hydroxides. Sometimes their aspect is conglomeratic. Their composition is either dominantly aluminous (bauxitic) or ferruginous. Sometimes, the content
in manganese is high, as well as in titanium. T h e
mineralogical composition is the same as in the nonindurated horizons : goethite and hematite for iron,
gibbsite and somctimes boehmite for aluminum.
Hardening of these materials does not seem to be

directly related to the nature or to the amount of

oxides or hydroxides; they seem rather to be related
to alternate seasons and a topography dominated by
subhorizontal planes, where drainage is often poor.
This is the case for Africa and other continents where
laterite is abundant o n flat surfaces, but not in m o u n tainous areas or even o n very rough land. There seems
to be a close connexion with the general pattern of
relief. But most laterites formed in past geological
times-and
some of those formed nowadays-are
at
present high above the valley floors as a result of
reversing of relief. Relations with past and present
climatic conditions are not always easy to determine
but one can consider as a reasonable thesis that the
equatorial climate, as observed in central Africa and
parts of s o m e countries of west Africa for instance,
as well as the subtropical and mediterranean climates
d o not favour formation of laterite. On the other hand,
a tropical climate with alternate seasons, related to a
rather flat topography, is most propitious to the formation of laterite. Those that can be seen in other climatic
areas are to be considered rather as relicts of past
conditions.
S U M M A R Y O F T H E E V O L U T I O N O F SOILS

Having reached this stage, one could think that oxides

and hydroxides have found definite form and place of
accumulation. In some bright-coloured soils, oxides are
firmly fixed o n clay minerals, and binding of particles
in pseudo-sands is effective; what then about concretions and laterite ? T h e latter occupies large areas, is
sometimes millions of years old and does not seem
affected by time.
Are the materials thus stored in the landscape
brought to a definite standstill or are they concerned
with a n e w pedogenesis ? T h e answer to this latter
question is undoubtedly positive. But the introduction
of the oxides in a n e w pedological cycle can happen in
t w o different ways. T h e products are either solubilized,
or remain in a solid state but are broken down. Both
cases will be examined.

Solu bilization of sesquioxides

This process concerns only iron and can hardly be
retained for alumina. Reduction and chelation are possible and usually occur together. A first example is
provided by the bleaching of bauxites, which has been
considered by m a n y authors (de Lapparent, 1930;
Chouhcrt, Henin, Betremieux, 1952). Organic matter is
the agent (lignites for bauxites in southern France;
march in Guyana) responsible for the reduction of iron,
which is then easily leached away. Another example is
provided by Maignien (1958, 1966) in his studies of
1. In his p:per

the French term cuirasseis coneidered 88 equivalent to

laterite.

35

P. Segalen

laterite, in Guinea and in all Sudanese Africa. The top

of the laterite-capped plateaux is the source of iron;
it is mobilized there and moves d o w n to the surrounding
gentle slopes wherc it contributes to the formation of
n e w laterite. But it is necessary to locate an ill-drained
area where local hydromorphy can allow the aceumulation of organic matter liable to reduce an complex
the iron. B u t this does not always exist. So one is led
to the conclusion that the hardening of the new laterite
is a result of a redistribution of iron on the slope itself,
due to poor local drainage conditions. Iron proceeds
from the fragments of the upper laterite carried d o w n
by gravity and not by solution.

Breaking down of laterite

Iron crusts are often limited by ledges which overhang
the surrounding slope b y several metres. Erosion is
active on the soil immediately below the crust, which
gradually breaks to pieces and is scattered on the slope.
At the foot of the slope the fragments cannot go any
farther and are, little by little, covered b y the fine soil
under the crust. In tropical areas, vegetation seems
able to bring to a standstill all the fine earth, which
thickens d o w n the slope. The fragments of laterite are
joined by pieces of quartz deriving from veins and form
a regular sheet beneath the surface of the soil. Differences in texture and structure on very long smooth
slopes, hardly ever cut by talwegs, favours, during the
rainy season, a provisional hydromorphy, which is
responsible for the formation of ferrous iron and for
local movements of iron. A n e w sheet of laterite, m a d e
of pieces of the former and bound together by iron
oxides, is obtained. T h e first, due to erosional processes,
m a y disappear completely. T h e second, when a new
water table settles may, in turn, appear above the
surrounding plains. O n e can be surprised to find, at
some height, laterite formed from debris of a sheet
which has been completely destroyed by erosion. T h e
ferruginous material m a y serve for several laterite caps
and seems to step d o w n a true flight of stairs. As a
matter of fact, iron probably moves a comparatively
small distance, either as solid fragments or in solution.
Bauxitic deposits, observed in G u y a n a (Bleackeley,
1964), probably have a similar origin. Flat bauxitic
plateaux located inland have been broken d o w n by
erosion and the debris were redistributed near the sea,
where they are reworked now.
The course followed by ferrous iron when it reaches
an underground water-level is probably more important. A water normally loaded with carbonic acid and
dissolved organic substances can easily keep its iron in
the ferrous form. But as soon as it reaches the surface,
the redox potential rises, and a ferric hydroxide precipitates. The influence of micro-organisms during the
precipitation of hydroxides of iron and manganese has
often been pointed out. Various bacteria and fungi are

36

considered to be responsible for concretions and hardpans containing these products.

Finally, if some of the iron escapes to continental
precipitation, it will reach the ocean waters, where pH
is usually high, as well as the salt content. Precipitation
of the remainder of the iron happens then, except for
the part which was consumed by some living organisms.

CONCLUSION
T h e studies accomplished in the laboratory, where
more and more accurate techniques are used to identify
amorphous and crystallized products, to determine the
amounts present and their location in a sample, have
m a d e substantial progress. Associated with field work,
they have led to a knowledge of the distribution of the
elements in a profile, or in a landscape. Recently, studies
have even been carried out on the distribution of some
elements throughout the world (Pedro, 1968). Such a
synthesis is n o w possible in m a n y fields of research as
more accurate methods develop and more samples
are analysed.
W e are n o w in a position to review briefly our knowledge of the oxides and hydroxides of the w a r m and
humid areas of the world.

1. Near the Equator, goethite is the dominant iron

mineral. Gibbsite is often encountered. Oxides of titanium as well as manganese are locally abundant.
Amorphous products are seldom found in yellow soils.
Iron is always outside the clay minerals lattice essentially kaolinitic. Concentrations of any of the abovementioned minerals can be encountered in all tabular
areas of the equatorial area. In Recent mountains, or
in volcanoes, hardened types of concentration are
hardly ever seen.
2. In the tropical zone, goethite and hematite are
c o m m o n . Gibbsite is seldom found. Amorphous mateerials are ferruginous or manganic. The former are very
often responsible for the red colours of the soils.
In smooth slopes, where oxidation-reduction process
can interfere, important migrations of iron occur. In
acid soils it gathers to form laterite sheets or concretions, nodules, dots, etc. In neutral or alkaline soils,
iron is often trapped by 211 clay minerals such as montmorillonites (Paquet, 1969).
In moderate or steep slopes where these processes
do not interfere, soils are red coloured (seldom yellow),
by amorphous material associated or not to goethite or
hematite.

3. In the subtropical zone, goethite, associated with

amorphous products, appears to be the dominant iron
mineral. Gibbsite seldom occurs. Soils are usually well
drained d o w n the profile; however, mottled zones rich
in concretions are known, as in Australia or the United

Metallic oxides and hydroxides in soils of the w a r m and humid areas

States. The iron content is high. Clay minerals are

mostly kaolinitic, but 2/i minerals such as vermiculite
or montmorillonite, containing iron in the lattice, or
presenting aluminum interlayering often occur (Rich,
1968).

4. In the Mediterranean zone, soils are dominantly red

or brown. The brownish colour is due to goethite; the
red colour is due to goethite and amorphous material,
seldom to hematite; gibbsite is u n c o m m o n . Soils
belonging to well-drained areas are red, those to poorly
drained ones are brown (Lamouroux and Segalen, 1968).
Clay minerals are varied (kaolinite, illite or montmorillonite). Concretions are small and scarce. No laterite is
known.
W h a t can be said about amounts? First of all therc
is no definite relation between the colour and the
amount. A n aluminous accumulation can be recognized
Ly its colour, if little iron is present, but a red sample
can be rich in alumina or not. A red or yellow sample
contains m u c h or very little iron. It is difficult to decide
witbout personal experience of the studied area and
without the help of analysis.
Soils containing the highest amounts of iron are
those deriving from ultrabasic rocks. Weathering gets
rid of silica and magnesia and iron accumulates by
difference. This happens near Conakry in Guinea, at
Surigao in the Philippines, in central New Caledonia
(the amounts of iron oxide reach 75 per cent). Elsewhere, the figures are quite variable according to the
parent rocks; soils deriving from basalts contain 20 per
cent, and from granites or gneiss 12 to 18 per cent.
In the tropical zone, total amounts are low In thc
fine earth (usually less than 5 per cent), with accumula-

tion taking place in concretions or laterite. In the subtropical zone, amounts go up to 15 per cent and depend
on the parent rocks. In the Mediterranean, soils deriving
from limestone range from 8 to 12 per cent.
T h e contents of titanium are generally low as far as
acid rocks are concerned. O n the other hand, soils
deriving from volcanic rocks contain m u c h more (5 to
10 per cent) with exceptionally high values in Hawaii
(Tamura et al., 1953). T h e manganese contents are
generally quite low; concentrations in soils are not
uncommon. Concentrations in chromium, cobalt and
nickel are closely related with ultrabasic rocks.
However, w e still have m u c h to learn about oxides
and hydroxides in m a n y a field. E v e n if w e have better
knowledge of the geographic area where these materials
appear in soils, w e must learn more about the conditions
of climate, drainage, and parent material which prevail
w h e n they are formed. The knowledge of these conditions in the past must be studied thoroughly, as we are
sometimes liable to assign to the present what is actually
inherited from the past. M u c h progress must be m a d e
in observing the structure of the materials and their
association :microscopic observation for microstructure,
electron microscopy to see the individual particles and
their shape. These two levels of observation, associated
with the two extremes-eye and magnifying glass on
one hand, X-ray diffraction on the other-will complete
the previous work.
In the field, m u c h remains to be done to spot,
describe, and understand h o w these products appeared
in soils and landscape. It seems that immobilization and
migration of oxides are still questionable. But it is felt
that progress can c o m e only from careful examination
of basic laws and relationships in physics and chemistry.

Bibliography
.

ALEXANDROVA,
L.N. 1954. O n the nature and properties of

CHARLOT,G. 1946. Thorie et mthode nouvelle danalyse quali-

interaction products of humic acids and humates with

sesquioxides. Pochvov., vol. 1, p. 14-29. (In Russian.)
AUBERT,G. 1963. Soils with ferruginous or ferrallitic crusts
of tropical regions. Soil Sci., vol. 95, no. 4, p. 235-42.
BARBIER;
G. 1938. Conditions et modalits de fixation de
lhydrate de fer collodal par largile du sol. Ann. Agron.,
p. 8-34.
BETREMIEUX.
R. 1951. Etude exprimentale de lvolution du
fer et du manganse dans les sols. Ann. Agron., p. 193-295.
BLEAKLEY,
D. 1964. Bauxites and laterites of Briitsh Guiana.
155 p. (Geol. Surv. Brit. Guiana, no. 34.)
BLOOMFIELD,
C. 1956. The solution-reductionof ferric oxide
hy aqueous leaf extracts. The role of certain constituents
of the extracts. C.R. VZih Znt. Congr. !%il Sci., B, p. 42733. Paris, AFES.

tative. Paris, Masson. 307 p.

CHOUBERT, B.; HENIN,
S.; BETREYIEUX,
R. 1952. Essai de
pufification de bauxites riches en constituants ferrugineux.
C.R. Acad. Sci., vol. 234, no. 25, p. 2463-5.
DEB, B. C. 1950. Estimation of free iron oxides in soils and
clays and their removal. J. Soil Sci., vol. 1, p. 212-30.
DIIooRE, J. 1954. Laccumulation des sesquioxydes libres
dans les sols tropicaux. Bruxelles, 132 p. (Pub. INEAC,
no. 62.)

1. It lias been found impossible to quote all the authors whose works were
used to prepare this paper. The complete list would bave been longer
than the paper itself,

37

P. Segalen

DION, II. G. 1944. Iron oxide removal from clays and its
influence on base-exchangeproperties and X-ray diffraction
patterns of the clays. Soil Sci., vol. 58, p. 411-24.
DROSDOFF,
M.;TRUOG, E. 1935. A method for removing and
determining the free iron oxides in soil colloids. J. Amer.
Soc. Agron., vol. 27, p. 312-17.
ENDREDY, A. S. de. 1963. Estimation of free iron oxides in
soils and clays by a photolytic method. Clay Miner. Bull.,
vol. 29, no. 5, p. 209-17.
FOLLETT, E.A. C. 1965. The retention of amorphous colloidal
Ferrichydroxide by kaolinites. J. Soil Sci., vol. 16, no. 2,
p. 334-41.
FRIPIAT,J. J.; GASTUCHE,M. C. 1952. Etude physico-chimique
des surfaces des argiles. Les combinaisons de la kaolinite
avec les oxydes defer trivalents.Bruxelles. 60 p. (Pub. INEAC,
no. 54.)
GANSSEN, R.; HADRICH, F. 1965. Atlas zur Bodenkunde.
Mannheim, Hochschulatlanten Bibliog. Inst., 85 p.
HARRISON,J.B. 1933. The katamorphism of igneous rocks
under humid tropical conditions. IIarpenden (U.K.), Imp.
Bur. Soil Sci., 79 p.
IIEM, J. D.; CROPPER W.H. 1959. Survey of ferrous ferric
chemical equilibria and redox potentials. 31 p. (US.Geological Survey, Water Supply Paper 1459 A.)
HERBILLON,
A.; GASTUCHE,
M.C. 1962. Synthse et gense
de lhydrargillite. C.R. Acad. Sci., vol. 254, p. 1105-7.
JEFFRIES,C. D. 1946. A rapid method for the removal of free
iron oxides in soils prior to petrographic analysis. Proc.
Soil Sci. Soc. Amer., vol. 11, p. 211-12.
LACROIX, A. 1926. L a minralogie de Madagascar. Paris,
Chdanel, vol. 3.
LAMOUROUX, M.; SEGALEN, P. 1969. Etude compare des
produits ferrugineux dans les sols rouges et bruns mditerraneens du Liban. Science du Sol, vol. 1, p. 63-76.
LAPPARENT,
J. de. 1930. Lea bauxites de la France mridionale. Paris, Impr. Nat. 187 p. (Mm.Carte gol. France.)
MACINTYRE, D.S. 1956. The effect of free ferric oxides on
the structure of some Terra rossa and Rendzine soils. J.
Soil Sci., vol. 7, no. 2, p. 302-6.

MAICNIEN,R. 1958. L e cuirassement des sols en Guine,239 p.

(Alm.Serv. Carte gol. Als.-Lorr., vol. 16.)
.1966. Compte rendu de recherches sur les latrites. Paris,
Unesco. 155 p. (Recherches sur les ressoiirces naturelles IV.)
MARTELL, A.; CALVIN,M. 1959. Chemistry of metal chelate
compounds. New York, Prentice-IIall.613 p.
MERIAUX,
S. 1958. Stabilit structurale et composition des
sols. C.R. Acad. Agric. Fr.,vol. 44, p. 799-803.
MILLOT,
G. 1964. L a gologie dea argiles. Paris, Masson, 499 p.
PAQUET, II. 1969. Evolution gochimique des minraux argileux
dans les altrations et les sols des climats mditerranens et
tropicaux saisons contrastes. Thcse, Fac. Sc., Strasbourg,
348 p.
PEDRO,
G. 1969. Principaux types daltration chimique la
surface du globe. Prsentation dune esquisse gographique.
Rev. Gogr. Phys. et Gol. Dynam., vol. X , no. 5, p. 457-70.
RICH, C. I. 1968. IIydroxy interlayers in expansible layer
silicates. Clays and clay minerals, vol. 16, p. 15-30.
SCUATZ,A. 1954. Chelation as biological weathering factor.
Proc. Penn. h a d . Sci., vol. 28, p. 44-51.
SEGALEN,P. 1964. Lefer dans les sols. Paris, ORSTOM. 150 p.
-.
1965. Les produits alumineux dans les sols de la zone
tropicale humide. Cah. ORSTOM sr. Pdol., vol. III,
no. 2, p. 149-76;no. 3, p. 179-205.
-.
1968. Note sur une mthode de dtermination des produits minraux amorphes dans certains sols hydroxydes
tropicaux. Cah. ORSTOM sr. Pdol., vol. VI, no. 1, p. 10526.
TAMM,O. 1922. Eine methode zur Bestimmung der anorganischen komponenten der Gelkomplexen in Boden. Medd.
fr Statens Skogs., vol. 19, p. 385-404.
TARDY, Y.;GAC,Y.1968. Minraux argileux et vermiculite-Al
dans quelques sols et arnes des Vosges. Hypothses sur la
no-formation des minraux 14 A. Bull. Serv. Carte gol.
Ab.-Lorr., vol. 21, no. 4, p. 285-304.
TAMURA,
T.;JAVKDON, M. L.;SHERMAN, G. D. 1953. Mineral
content of low humic, humic and hydro1 humic latosols of
IIawaii. Proc. Soil Sci. Soc. Amer., no. 17, p. 343-7.

3. Weathering and soil-forming processes

in the tropics
J.

J. van Schuylenborgh
Laboratory of Regional Soil Science, Agricultural University,
Wageningen (Netherlands)

Although there are n o real differences in weathering

and soil-forming processes in the humid tropics as
compared with those in the humid temperate climates,
yellowish and reddish soils are formed in the former
case under good or moderately good drainage conditions, whereas in the latter, brownish and greyish soils
c o m e into existence. This is probably the result of the
fact that the reddish soils are, as an average, poorer in
silica than the brownish soils, notwithstanding the
fact that the reddish soils are, o n an average, poorer in
streams is equal all over the world. This leads to the
conclusion that the soils of the temperate climates are
younger than those of the tropics. T h e conclusion is in
agreement with the fact that at northern latitiides thc
land surface has been considerably changed during the
ice times, thus exposing fresh material to atmospheric
influences. Actually, the red soils in the tropics are
bound to old surfaces. This makes it difficult to study
the soil-formingprocesses as there have certainly been
climatic changes (pluvials and interpluvials) as well as
geologic processes (e.g. land upheaval), which change
the weathering and soil-forming processes. Nevertheless, an attempt will be m a d e here to discuss these
processes and their results, assuming residual soil
formation as far as possible.
Jnteresting soils in the tropics are the strongly
weathered ones: the Oxisols. F r o m a genetic point of
view they can be grouped into Oxisols with plinthite
(the former Laterite Soils) and Oxisols without plinthite
(the former Latosols).

OXISOLS WITH PLINTHITE

T h e processes involved are those leading to formation
of the oxic horizon and those responsible for the genesis
of plinthite. T h e former were called laterization or

ferrallitization. T h e process is a weathering process in

which silica is released from the primary silicates and
from part of the quartz (if present); also the alkali
and alkaline earth metals are released. T h e latter are
practically completely leached, whereas the released
silica combines with alumina to form clay minerals, of
which kaolinite is the final product. T h e remainder of
the released silica is leached. This process will be called
desilication, and described as above, it is a n
absolute impoverishment in silica and a residual enrichment with the stable weathering products, amongst
which the iron-oxides and hydroxides are important,
and sometimes also quartz. However, the decrease in
silica content m a y also be achieved by deposition of
sesquioxides, transported laterally with groundwaters
coming from higher situated parts in the surroundings.
T h e decrease in silica in this sense is relative because
of the actual enrichment with sesquioxides. Both
processes, actual as well as relative, will be called
desilication. Most frequently a combination of these
t w o types of desilication will be responsible for the
formation of oxic material.
T h e forming process of plinthite, the characteristic
of which is the ability to harden upon exposure to the
air, will be called plinthization. F r o m descriptions of
the structure of plinthite (Alexander and Cady, 1962)
it is clear that this process is strongly dependent
o n the possibility of redistribution of products of desilication, of which iron plays a very important role. It
is evident that the redistribution of iron within the
preplinthite horizon will be strongly favoured by alternate reduction and oxidation. In its turn this is only
possible if a fluctuating groundwater level or apparent
groundwater level is present. Besides this mechanism,
certain recrystallization processes will be involved to
give the plinthite its firmness,

39

J. van Schuylenborgli
5. T h e rotten rock (zone V), where thc green-, yellow-

TIIE P R O F I L E

and brown-colouredhalf-weathered minerals become

It

visible. Its appearance varies of course with the type

is hardly possible to describe a representative

profile of Oxisol with plinthite because of the extremely
wide variations. Nevertheless, an attempt is m a d e here
to give an idealized picture of a most complete profile,
developed o n hard rock. After having studied the literature on laterite soils it is thought to be justified to
state that a modal profile consists of five zones :(1) the
top zone; (2) the plinthte zone; (3) the mottled clay;
(4) the pallid zone; (5) the rotten rock; with (6) the
rock.
1. The upper zone, referred to in the following as
Zone I, is a soil with texture varying from clay to
loam with greyish-brown,reddish-brown,yellowishred, yellowish-brown or brownish-yellow colour.
Sometimes an A2-horizon is present and a cambic
or argillic horizon. The consistency, when moist, is
generally friable and the structure subangular blocky
or granular. Small iron concretions m a y be present.
There is generally a clear transition to:
2. T h e plinthite zone (indicated as zone II). This
horizon m a y be partly hardened. T h e colour is red,
mottled with yellow or even violet. The texture can
be clayey or sandy, depending on the nature of the
parent rock. The structure is pisolitic or vermicular,
vesicular, cellular, if hardened. In the soft state it
is granular or Lubangular blocky. The pores are
frequently filledwith whitish, yeliowish, or greyish
material. If moist, it can be dug out and hardens
upon exposure to the au. T h e non-hardened part is
extremely firm, the firmness decreasing with depth.
T h e horizon merges diffusely into :
3. T h e mottled clay zone (zone III). It is a very pale
yellow to white clay matrix with m a n y coarse to
very coarse red, yellow, brown, blue or violet mottles.
These mottles do not harden and have the same
consistency as the white clay material. The zone
merges diffusely into :
4. The pallid zone (zone IV). It is a very pale yellow
to white homogeneous clay. The lower part of it
shows features, such as layering and folding, due to
the original rock structure and these become gradually m o r e visible w h e n merging into :
,

of parent rock.

The conclusion can be drawn that such a profile can

only be formed under conditions of a fluctuating groundwater table (e.g., as result of alternating dry and wet
seasons during the year); the upper boundary of the
pallid zone is or has been the lowest position of the
groundwater table.
N o indications are given on the thickness of the
horizons, because this is very variable. In some cases
the pallid zone is extremely thick (8 metres or more)
(Walther, 1915; Hallsworth and Costin, 1953); in
other cases the plinthite horizon is very thick as c o m pared with the mottled clay and pallid horizons. In
yet other cases the pallid zone is absent; the mottled
clay zone rests then immediately on the rotten rock and
shows in the lower parts features due to the original
rock structure. T h e reason for this absence could be
that weathering has not yet proceeded so deeply as to
form a water table, due to worsening drainage conditions with increasing weathering depth (see the paper
on formation of the latente soil profile). Another reason
could be a surface sheet erosion that keeps pace with
the weathering rate.
Very frequently zone I, the top soil, is not present,
or only a thin layer of softened hard laterite. The
United States Soil Survey staff does not classify such
soils as Oxisols but as Entisols, as the hardened plinthite (laterite) is considered as the parent material for
new soil formation.
A well-investigated example of an old red soil (a
high level laterite) is one investigated b y Satyanarayana
and T h o m a s (1962). T h e plinthite in this soil has already
been hardened and the top layer eroded. Nevertheless
it can be used to see what has happened during
weathering and soil formation. The pallid zone is
absent.
F r o m the chemical composition of the profile, we
calculated its normative mineralogical composition,
using the rules develope! by Niggli (see Burri, 1964) for
the rock and V a n der Plas and V a n Schuylenborgh (1970)
for the soil (Table 1). ..

TABLE1. Normative mineralogical composition of profile 1 (weight percentages)

Zone

I
11.1
11.2/111

V
VI

Q
28.9
23.6
30.8
15.9
2.8

Pk

Or

- -- 12.5
1.5
-

--

41.7

41.2

E quartz; O r = K-feldspar; P
lg plagioclases; Bi
o i goethite; EIm E hematite; Misc
miscellaneous.
2. Hornblende-biotite variant of the katanorm.

40

Bi

Ho

Mt

0.5

3.2

7.1

2.4

biotite; 110

hornhlende;

Ms

Ka01

Go + H m

Mise

10.2
2.4
6.1
46.0

28.3
56.4
45.8

27.7
14.1
12.7
12.5

5.0
3.6
4.4
11.0
1.5

Mt = magnetite; ,Ils = muscovite (illite); Kaul = knoliiiite;

Weathering and soil-formingprocesses in the tropics

TABLE2. Normative mineralogical composition of profile 2

(weight percentages)'
Zone

Kaol

Ms

Go+ltm

Iu

1.1
1.2
11.1
11.2
11.3
11.4

90.1
88.3
64.5
45.0
11.3
20.8
38.2
48.9

3.4
4.2
10.2
10.8
10.0
25.9
51.3
44.3

0.4
0.4

6.1
6.8
25.1
43.8
77.7
51.4
6.4
1.2

0.3
0.4
0.2
0.3
0.7
0.8
1.9
1.8

III

IV

0.3
1.1
2.2
3.8

1 .Q= freesilica;Kaol = kao1inite;Ms = muscovite(il1ite); Go = goethite;

IIm hematite; H u rutile (amatese).

Evidently the rock minerals have partly desilicated

to form micas and iron oxides, while upon further soil
formation illite is transformed into kaolinite. In the
top part of the profile the kaolinite content decreased
considerably,whereas the iron minerals enrich residually.
The loss of kaolinite is the result of lateral removal of
the soft part (kaolinite) of the hardened plinthite by
erosion. .
Another, example is the Newell-Pelliga profile,
described by Hallsworth and Costin (1953). T h e normative mineralogical composition was again calculated
from the chemical analyses, the latter being kindly
provided to us by Dr. Hallsworth, and the result can
be found in Table 2. Unfortunately the rock analysis
was absent, but this profile had a pallid zone (profile 2).
It can be seen that the pallid zone w-as practically
deprived of iron. Apparently, the iron was transported
in the reduced state upwards with the rising groundwater table and deposited in the plinthite zone. Apart
from that, a downward movement of iron can also be
noticed from the top layers of the soil. This is evidently
due to podzolization processes which happen to occur
in the upper part of the profile. Also kaolinite seems to
be unstable as it is decomposed, leaving behind free
silica. T h e aluminium oxide which should have been
foimed has been dissolved because of the high acidity
of the soil.
T H E PROCESS

General outline of the process

It is a reasonable suggestion that the three processcs

mentioned above are not acting simultaneously and
with the same intensity. Further, they will be strongly
dependent on time. T h e profile descriptions and the
analytical details s e e m to indicate that the whole process can be divided into three stages. T h e first stage is
connected with free or nearly free drainage. The rock
is disintegrated by the atmospheric effects of heating

and cnoling, wetting and drying, by hydration and

hydrolysis, carbonation, oxidation and dissolution
effects, all to a large extent influenced by soil fauna and
vegetation in an ever-increasing degree. Secondary
minerals, such as kaolinite, gibbsite, goethite, etc., are
formed ;bases,silica and part of the aluminium removed;
shortly, the desilication has a profound impact on the
parent material. As long as the weathered layer is not
too thick and the weathering zone high above the base
level of erosion, drainage will be free. As weathering
proceeds, the contents of both clay minerals and amorphous sesquioxide colloids are raised and the base level
of erosion approached, leading at a certain m o m e n t to
the stage where the permeation of the soil with rainwaters is slow enough to prevent the removal of all
the water in the wet season. Consequently a temporary
groundwater table is established. In the dry season
however, the water can still be removed sufficiently
and disappears. In this second stage (appearance and
disappearance of groundwater), reduction processes will
start, notably if the roots penetrate deep enough. T h e
second stage is therefore characterized by the occurrence of weathering and soil-forming reactions under
alternating reductive and oxidative conditions. T h e
transformations during this stage are more or less
comparable with those involved in pseudo-gley formation and they will be different from that of the first
stage. The material formed will show mottling, whereas
the upper part of the profile will have a uniform colour.
It is evident that in the second stage the plinthization
process starts to act.
U p o n further weathering, when approaching the base
level of erosion, drainage and evapotranspiration are
no longer able to remove all the water even in the dry
season and a permanent groundwater table will be
established. Normal gley formation will be superimposed
upon that of pseudo-gley.A zone of constant reduction
comes into existence and rock-weathering will proceed
from this m o m e n t on, under reducing conditions. There
is still a fluctuating groundwater table, although thc
highest water-levelis lowered gradually upon weathering
and the mottled horizon grows thicker. Part of the
originally formed mottled clay is n o w only wetted by
capillary rise above the water table and the upper part
will not be wetted any more. This part can dry to some
extent and this is assumed to lead to the special characteristics of the plinthite material. The drying causes a
certain dehydration and crystallization OP amorpkous
iron and iron oxyhydrates into cryptocrystalline and
crystalline hydroxides and oxides, thus cementing the
material.
During the three stages, the upper layer, which was
once formed by oxidative weathering of the bedrock,
has ncreasingly changed under the influence of the
vegetation. The result is sometimes accentuated by the
type of the bedrock, a surface soil with podzolic features.
Frequently the surface layer is more or lcsu liomogenized by biological activity, this too depending on the

41

J. van Schuylenborgh

type of parent material. h o t s penetrate into the subsoil in the dry season and part of them will die in thc
wet season because of excess of water and, eonsequently, lack of oxygen; animals bury their tracks into
the subsoil in the dry season and c o m e to the surface
in the wet season. Termites are particularly active, as
they m a k e passageways d o w n to the level of the water
table (Yakushev, 1968). In this way drainage paths
for the rain-water are formed, leading to a n inhomogeneous wetting of the subsoil which contributes to the
special characteristics of the plinthite material such as
clay coatings along tracks and fissures.
F r o m these considerationsit will be evident that the
composition of zone V (the weathered rock) of a mature
plinthite soil profile will strongly deviate from that of
the weathered material, formed at the beginning of the
formation process. T h e latter was the result of oxidative
weathering and the former a reductive process. It is
therefore in fact not correct to consider zone V of a
mature plinthite soil as the parent material of the top
soil. This point of view can be extended to all zones of
a plinthite soil: these are more or less individual
horizons formed under their particular forming condilions. Naturally there is some, and frequently even
strong, influence of underlying and overlying horizons;
but the interrelationships between the top zone and the
parent material are not so intimate as in temperate
climates where the depth of weathering is small c o m pared with that under tropical conditions. In this
connexion, Pedro (1964), states that only the upper
part of the profile is more or less in equilibrium with
thpresent-day climate.

Dedication

3 Or

42

log K,

+ 2 II+ + 12 II,O = MS + 2 K+ + 6 H,SiO, (i)

[KI + 6 log [H,SiO,]

= 2 log [HI

(la

Similarly :

20r+2H++4HzO=Pyr+2K++2H,Si0,(2)
log K, = 2 log

2 ills

[KI + 2 log [H,SiO,]

[HI

(2a)

+ 2 H++ 6 H,SiO, =
= 3 Pyr + 2 K+ + 12 H,O
log K, = 2 log [KI -6 log [H,SiO,]

[HI
2 Ms + 2 H t + 3 H,O = 3 Ka01 + 2 K+
log K,

Pyr + 5 H,O

(3)
(3a)

(4)

[KI
= 2 log -

[III

= Kaol + 2 H,SiO,

(5)

= 2 log [H4Si0,]
(54
2 ills + 2 Hf+ 18 H,O =
= 3 Gibb + 2 K+ + 6 H,SiO, (6)
log K,

log K,

M a n y experiments have been performed to imitate the

weathering process in the laboratory in order to be
able to study the mineral transformations more
thoroughly. Amongst these, those of Correm and eoworkers (1961) and of Pedro (1964) are most famous.
However, w e shall not discuss their experiments, but
approach the problem from a thermodynamic viewpoint.
W e have seen that during the weathering of the rock
of the first profile, micas (illite) were formed as a n
intermediate phase and finally kaolinite. As thc rock
was rich in IC-feldspars, w e can study the stability of
the various possible phases in the system K,O Alzo,
Cio, H,O. In this system K-feldspar, muscovite
(illite), pyrophyllite, kaolinite and gibbsite are possible
phases (Slager and V a n Schuylenborgh, 1970). A s the
standard free-energies of formation of these phases are
known, the equilibrium constants of the various
cquililria can be calculated. T h e reactions are :

where brackets denote activities. T h e charges o n the

ions are omitted for the sake of convenience. As, by
convention, the activities of solid substances are taken
to be unity, and as, in dilute solutions, the activity of
water is also unity, the expression for K, reduces to

= 2 log [KI + 6 log [H4Si04]

[Hl
Kaol + 5 H,O = Gibb + 2 II,SiO,
log K, = 2 log [H,SiO,]

(6a)

(7)

(74

T h e symbols,used in the equations,have the following

meaning :Or = K-feldspar, KAlSi,O,; M s =muscovite,
KAl3Si,Ol~(0II),;
P y r = pyrophyllite, N,Si,O,,(OH),;
Kaol = kaolinite, Al,Si,O,(OH),;
Gibb = gibhitet

Alzo,. 3 H,O.
In order to calculate the thermodynamic equilibrium
constants, the following relationships have to be used :

In K = -AF:/RT or log K = AF:/1.364

and AF; = AFOfprotlucts AFJreactants

(8)

(9)

where K = the equilibrium constant, AF;= standard

free-energy change of the reaction, AFJ=standard
free-energy of formation, R = gas constant (0.001987
keal/deg), and T = absolute temperature (298.150K).
All values used for the standard free-energies of
formalion, arc AFj.Or = -893.8 (Kelley, 1962);
AFTAls = -1330.1 (Barany,1964); AFjPyr= -1258.7
(leesman and Keller, 1968); AF; Kaol = -904.0

Weathering and fioil-formingprocesses in the tropics

(Reesmari and Keller, 1968); AFf Gibb = -547.0

(Barany and Kelley, 1961); AFJH,SiO, = -312.7
(Neesman and Keller, 1968);A Fo K+ = -67.5 (Rossini
et al., 1952); AFO ,O= -56.7 (Rossini et al., 1952);
AF' H+ = O; all values in kea1 mol-1. With (9)and (8)
the following values for log K were obtained:
log Ki = -15.4; log KZ = +3.3; log K3
+40.5;
log K4 = -12.2; log K, = -9.4; log K, = -21.2;

106 K7 = - 11.2.

With the aid of these data the stability fields of the

mentioned minerals can be delineated in a two-dimensional diagram with log[K]/[H] as one axis and
log[H4Si0,] as the other. T h e results are given in
Figure 1. In this figure the line of highest content of
dissolved silica is obtained from the data of Siever (1962),
which is 10-2*8mol 1-l. A t the right side of this line
w e have therefore the field of K-feldspar + amorphous
silica.

I
I
I

permeability and poor drainage conditions, muscovite

(illite) will be formed. A t lower [K]/[H]
pyrophyllite
(montmorillonite)can be formed. A t later stages, w h e n
the cations of the decomposing feldspars have been
removed and w h e n the reaction has become acid, both
minerals are transformed into kaolinite. Muscovite
(illite) can therefore be expected to be present in the
weathering zone of acid rocks, whereas kaolinite is the
c o m m o n product in the middle zone.
A study of the system Na20-K20-A120,-Si02-I20
at 2 5 o C and 1 atm. was m a d e by Hess (1966). T h e
various phase equilibria in this system could be described
as functions of the [Na+]/[H+]-and
[K+]/[H+]-ratios
and the activity of dissolved silica. This study w a s
especially interesting with regard to the occurrence of
montmorillonite, which can sometimes be found in the
weathering zone immediately above the rock. It
appeared that this mineral was stable at low values of
log [K+]/[H+],
whilst the value of log [Na+]/[+] can
vary considerably. It is important that the activity of
dissolved silica be very high. These statements are
illustrated in Figure 2 where the phases occurring at

amorphous
silica
saturation

Or

\
I
I

Gi.
-6

-.5

-4

-3

-2

log [H, Si O,]

FIG.1. Stability relations of some phases in the system

K20-Al,0,-Si0,-II,0at 25oC and 1 atm. total pressure
(numbers at the boundary lines refer to equations, page 42).
Gibb. = gibbsite; Kaol. = kaolinite; Pyr. = pyrophyllite;
Ms. = muscovite.
T h e figure shows that if the H,SiO,-activity is kept
low (and consequently also the activities of the dissolved ions), e.g. w h e n permeability is rapid and drainnage perfect, the stable end-product of weathering is
gibbsite. This is in agreement with the findings of
Herbillon and Gastuche (1962), w h o stated that elimination of extraneous ions promotes the formation of
crystalline gibbsite.
A t higher H,SiO,-activities and in neutral or alkaline
media (high [K]/[H]
ratios) as m a y occur under slow

-6

-5

-4

-3

-2
log [H4 Si O,]

FIG.2. Stability relations of some phases in the system

Na,O-K,O-Al,O,-Si0,-II,O
at 250 C and 1 atm.;log K+/II+
= 4.
Mont. = montmorillonite; Phil. = phillipsite; Or. = K-feldspar; Kaol. = kaolinite; Ab. = albite; Gi. = gibbsite.

43

J. van Schuylenhorgh

log [K]/[II]
= 4 are represented. It appears that the
phases albite, phillipsite, K-feldspar,and a part of the
montmorillonite occurrence are metastable under these
conditions. This confirms that montmorillonite predominantly forms under conditions of poor drainage,
a fact noticed by Schellmann (1964), w h o found montmorillonite at a depth of approximately 8 metres
below the surface in the weathering zone of a serpentinite and in m a n y other situations which can be found
in the literature.
T h e most important conclusion to be drawn from this
is that, at the beginning of the weathering process, the
products formed will be different from those at more
advanced stages, merely as a consequence of the change
in drainage conditions.
Finally, attention should be given to an unusual
suggestion, m a d e by Lovering (1959) with respect to

the dedication process (see also Davis, 1964). H e

suggested that desilication could be effected to a considerable extent by the accumulation of silica by tropical
vegetation and the removal of fallen litter by erosion
before the silica is released. Considering a tropical forest
of silica-accumulating trees averaging 2.5 per cent
Cio, and approximately 40 tons dry weight growth per
ha per year, he calculated that in 5,000 years, approximately 5,000 tons of silica would be removed from one
ha of soil by the forest. This amount of silica is,
according to Lovering, equivalent to the silica in a
volume of basalt of 1 ha and 30 c m thickness. As
there were no data available on the amount of litter
removed b y erosion, he was not able to calculate the
amount of silica removed in this w a y relative to the
amount of silica returned to the soil by decaying
vegetation.

Plinthization
t 1.0

t 0.8

-I0.6

t 0.4

t 0.2

0.2

-0.4
-0.6

- 0.8
O

10

pH

12

FIG.3. Stability fields of the stalle phases in the system

Fe-O,-II,Oat 250 C and 1 atm. pressure. Boundary between
solids and dissolved species at C[Fe] =

44

W h e n soil development reaches the second stage a quite

different process comes into play, viz. the dissolution of
iron and manganese and their deposition into mottles, or,
in other words, the formation of the mottled clay zone. It
is assumed that this zone will unitially have the characteristics of a pseudo-gley,as the rock acts in this stage
as an impervious layer. Consequently, part of the
profile is under conditions of poor aeration in the wet
season, whereas in the dry season water evaporates
and air will be plentiful. In other words processes of
reduction and oxidation will becorne active, whereas
in the first stage only oxidating processes prevail.
During the reduction state iron and manganese arc
reduced in places where the redox-potential is low
enough; they then become mobile and diffuse to
places where Eh is higher, e.g., in the neighbourhood
of pores which still contain some oxygen. At these
places oxidation and deposition take place. The iron
and manganese m a y be mobilized at the bottom of
the profile, transported with the rising groundwater
and oxidized in the higher situated layers, but m a y
also m o v e over short distance within individual layers;
After progressive weathering, the layers above the
highest groundwater level are wetted with groundwater
only by capillary forces and this is thought to be the
m o m e n t that the mottled clay is transferred into the
typical plinthite. Only weak reduction and, consequently, slight translocations of iron and manganese
occur in this fringe. Ageing and crystallization of the
already accumulated iron, manganese and aluminium
hydroxides come into play thus giving the plinthite its
very firm consistence. T o this m a y be added the tramformation of the silicic acid in the rising saturated or
even supersaturated soil water into opal, chalcedony, or
trydimitc. It is thought that the process of the transormation of thc rock into plinthite material might hc
summarized from a chcmical viewpoint as follows :

Weathering and soil-formingprocesses in the tropics

If a mineral aggregate is involved in the process of

weathering, one of the processes is the hydration of the
minerals. Vater is tightly bound to the surface and
in the crevices and fissures of the aggregates. At the
surface and in the fissures, therefore, micro-drainage
conditions are poor, although macro-drainage m a y be
good. Under these conditions high pH values and high
salt concentrations m a y occur at the surface and in the
fissures of the mineral aggregates. This implies that
conditions are favourable for the formation of amorphous Al-hydroxide gels or pseudo-boehmite (Gastuche
and Herbillon, 1962; H s u and Bates, 1964; Hsu, 1966,
1967), amorphous Fe(II1)-hydroxides (under high
redox-potentials), amorphous silica (Figs. 1 and 2),
and even clay minerals of the smectite group (such as
montmorillonite, Fig. 2). In this w a y the mineral
aggregate changes into an intimate mixture of the
mentioned amorphous anci cryptocrystalline products.
All these products are metastable from a thermodynamic point of view and will be transformed (aged)

FIG. 4. Stability fields of the stable phases in the system

Mn-,-II, at 250 C and 1 atm. pressure. Z[XZn] = 10-6.

into stable minerals, according to reactions

as (lo), (11) and (12):

Fe(OH), (a, s)

-+

a-FeOOH(s)

I
(goethite)

I
(amorphous ferric-

+ H,O

hydroxide)

Fe(OII), (a, s) + A100H(s)

-+

(pseudo-boehmite)
-+

a-FeOOH(s)

+ Al(OII), (s)

(gibbsite)
2Fe(OH), (a,s) + A100II (s) -+
--t a-Fe,O,
(s) + Al(OII), (s)

I
(hematite)

+ II,O

which all have negative standard free-energy changes

of reaction, as can be calculated from the standard
free-energies of formation, listed by Garrels and
Christ (1965), and Krauskopf (1967).Consequently, the
aggregate will be transformed into a more or less
porous fabric of minerals which are more stable than
the original ones. Micro-drainage then becomes
better and montmorillonite can be transformed into
kaolinite and finally into gibbsite (Fig. 2). In the wet
periods the pores and fissures are filled with the rising
Fe2+- and Mn2+-ions and II,Si04-bearing soil solution,
oxidation takes place and the pores are gradually filled
with iron-minerals, containing some manganese and
silicic acid, giving rise to a firm fabric. This is tbe
redistribution process, or plinthization, which acts over
very short as well as larger distances. A s gibbsite is
more soluble than the iron hydroxides it will be leached,
at least partly, and the spaces left can be refilled by
iron hydroxides. T h e silicic acid . m a y help in the
formation of a firm fabric by the formation of opal or
chalcedony.
It will be clear that such a material will harden
strongly upon exposure to the air and u p o n frequent
wetting and drying because of the continued loss of
moisture (dehydration) and crystallization. Such a
hardening invariably occurs after deforestation, e.g., by
shifting cultivation, with subsequent erosion of the
surface soil. T h e soft material of the laterite, being
predominantly kaolinite, is washed out by the rains
giving the laterite a porous, vesicular, vermicular, or
cellular appearance.
One of the most difficult questions to explain is that
of the simultaneous deposition of iron- and manganese
oxides in some oxisols with plinthite. On the basis of
Figures 3 and 4, representing the stability fields of
the stable Mn- and Fe-oxides and hydroxides as functions of Eh and pH, it should be concluded that there
must always be a separation between iron- and m a n -

45

J. van Ychuylentorgh

ganese-oxide deposition, as w e know, e.g., to occur in

paddy soils and pseudo-gleys. There is no possibility
of narrowing the Mn*+-field. Nevertheless, there has to
be some w a y to explain the mentioned simultaneous
deposilion. O n e possibility could be the formation of
intermediary, unstable iron hydroxides during the
reduction-oxidationprocesses. Frequently, lepidocrocite
is found in the bright orange mottles of pseudo-gley
or gley soils (Schwertmann, 1959). If it is thought that
this hydroxide has been formed a stability diagram can
be constructed assuming for instance the presence of
amorphous silica and CO, = lO-l.7 atm.
T h e results (Fig.5) show that the field of predominance of Fe2+ has grown considerably, but even so is
not comparable in size to that of Mn2+. The conclusion
seems justified that a separation between manganese
and ferric-deposits is unavoidable. In fact, this is

t 1.0
Fe3+

Eh (VI

+ I

.'
.'

CiO,l,si I

t 0.8

_ _ - -I

.'

.'

\
\

\.

.'

\
\

t 0.6

\.

t' FeOOH

\
\
\
\

'
.
,

si02 (a;d

t 0.4

Y-FeOOH
\

t 0.2

,
\

'
.

\
\

'
\

.
'

-..

Fe304

SiO, (a.4

.'

- 0.2

. '

\.

\
\

.'

Fe304

- 0.4

.
FeC03

- 0.6

.'
.'

-0.8

10

.\.

.'
pH

12

FIG.5. Stability fields of FeCO,, Fe,O,, and metastable

y-Fe0011 in the presence of amorphous silica, nt 25OC,
1 atm. total pressure and CO, = 10-1.7 atm. Boundary of
solids at activities of dissolved Fe-species = 10-5; activity of
dissolved JI,SiO, = 10-2.7.

46

frequently observed in paddy soils and pseudo-gley and

gley soils ( D o u m a et al., 1968).
A second possibility is the forqiation of stalle
ferrous-complexes. The presence of such complexes,
especially with reducing organic ligands (e.g., polyphenols) was suggested b y Bloomfield (1957, 1959) but
little is k n o w n about their nature and stability in soils.
If such a mechanism occurs it must be that the Fe2+-ions
form more stable complexes with the ligands than
Mn2+-ions do. This was actually found by Irving and
Williams (194,8).It has to be stipulated here, however,
that the magnitude of the stability constants of the
complexes is not the sole criterion for their stability.
The solubility products of the hydroxides of the metals
are also important. T h e higher the stability constants
of the metal complexes and the larger the solubility
products of the metal hydroxides, the more stable are
the complexes and consequently, the stronger is their
resistance to oxidation. More exactly, the smaller the
over-allhydrolysis constant of the complexes, the higher
the stability and the stronger the oxidation resistance.
T h e relation

where KhMLis the hydrolysis constant of the metal

complex, ML; K M L is the stability constant of the
complex, and K,, is the solubility product of the metal
hydroxide, predicts what will happen on raising the p H
of the environment. If KhMLis smaller than 1, the
complex is stable in neutral or weakly alkaline media;
if larger than 1 it is stable only in acid media. T h e
solubility products (calculated from free-energyvalues)
of Fe(OH), and Mn(OH), are resp&tively,
and
lO-l3no.
Hence, if K F ~ is
L larger than
the Fe(I1)
complex will be stable and consequently resist oxidation. In the case of the complex M n L , KM^^ has to be
larger than 1013.0. To explain the simultaneous precipitation of ferric- and manganese hydroxides K F ~
should be larger than
whereas KM^^ should not
be allowed to exceed 1013*0.Only in that case will the
conditions for the simultaneous oxidation of ferrous
and manganous compounds approach each other.
If the ligands also participate in the reductionoxidation processes it becomes difficult to predict what
will happen although this problem can be solved.
A third possibility might be the formation of impure
Mn(I1)-, Mn(II1)- and Mn(1V)-oxides or their hydrates;
e.g. Mn(I1) m a y be replaced by Fe(III), and other polyvalent ions. S o m e of the Mn(1V) m a y h e replaced by
Fe(II1) and so on (Ponnamperuma, 1969). If these
non-stoichiometric oxides of variable composition have
lower free energies of formation than their ideal counterparts, as was experienced by the author cited, the field
of dominance of soluble Mn(I1)-species will be narrowed
considerably, thus approaching that of soluble Fe(T1)species.

Weathering and soil-formingprocesses in the tropics

A fourth possibility might l e found in a rapidly

fluctuating groundwater table combined with different
oxidation rates of manganous- and ferrous-ions. If the
oxidation rate of ferrous-ions is slower than that of
manganous-ions, a simultaneous precipitation as the
oxides or hydroxides is very well possible, but as far
as known, little, if anything, has been published in this
respect. In fact, the problem of the reaction rates interferes seriously in stability studies. For example, it is
k n o w n that siderite is only found in old and never in
young paddy soils, although the environment in paddy
soils is favourable for its formation; the formation rate
of siderite, however, seems to be so slow that it appears
only after a considerable period of time. In spite of this
difficulty, a thermodynamic examination of the soilforming processes is very useful, as it predicts the result
of these processes, if the circumstances remain more
or less constant.
OXISOLS WITHOUT PLINTHITE
Typical Oxisols without plinthite are deep and friable
to very friable soils with an oxic horizon. Horizon
differentiation is indistinct as the horizon boundaries
are diffuse. There is generally no clay m o v e m e n t and
the structure is granular or subangular blocky and very
stable. T h e soils are porous and have a rapid permeability. T h e oxic horizon does not harden upon exposure
to the air as do Oxisols with plinthite. Tf the solum is
very deep (e.g., thicker than 4 metres), it is underlain
sometimes by a mottled clay horizon. Otherwise the
solum lies directly on the weathered rock. Concretions
are absent or near absent and are generally soft. M a n ganese concretions are frequently present. D u e to the
high porosity, depth and stable structure most Oxisols
without plinthite are only slightly susceptible to erosion
and certainly less than other soils under equal topographical conditions.
A N O X I S O L P R O F I L E W I T H O U T PLINTIIITE

In the course of time several types of Latosols have

been described, some with fundamental analytical
details, but only a few are analysed up to and including
the bedrock. However, this is necessary to obtain a
clear insight into the transformations that have taken
place during the formation of the soils. As far as the
authors are aware, the best investigated Oxisols 'are
those formed on serpentine rocks (Bennett and Allison,
1928; Roberts, 1942; Schellmann, 1964;o w n investigations on serpentinites in W e s t Irian, Indonesia). As the
soils in question are very similar, only the investigation
of Cchellmann (1964) will be discussed.
Profile 3 : Kukusan mountains, south-east Kalimantan, Indonesia. During a geological investigation in 1956
by Gaertner and Wirtz (see Schellmann, 1964) it
appeared that the Kukusan mountains were covered

with a thick layer of ferruginous material. Samples were

taken every half metre u p to and including the rock. A
profile description was not made. F r o m the analytical
details and the too brief description of the geological
samples, the following succession of horizons m a y be
concluded:

Al 0-1 m

Brownish friable material with some

concretions.
BI 1-2m
B r o w n to yellowish brown friable
material with very few concretions.
I), 2-5 m
Yellowish L r o w n to yellow friable
material.
B, 5-6.5m Yellow friable material.
Cl 6.5-7 m Yellow friable material.
C, 7-7.5 m Strongly weathered serpentine with
brownish discolourations.
R + 7.5 m Very weakly weathered serpentinized periotite.

The profile is situated at an altitude of approximately

500 m above sea level. The rainfall varies monthly
from 300 m m in January to 120 m m in September and
is therefore very high. T h e m e a n annual temperature is
approximately 240 C. The vegetation is a tropical rain
forest. There is n o dry season in the year :water movement in the profile is therefore always downwards.
Internal and external drainage are good. Classification
places this soil in the Typic Haplorthox subgroup.
The analyses of Schellmann (1964) include PII,
exchangeable bases, total elemental composition, X-ray
and DT-analyses, and electron microscope investigations. The PII (II,O) varies from 5.0 in the A,- to 5.3
in the B3-horizon and increases from there o n to 8.0
in the C,-horizon. T h e pH (0.1 MKCl) increased from
5.9 in the Al-to 6.5 in the B3-horizon and decreased
from thereon to 5.7 in the C,-horizon. ln this soil
pII(KC1) is therefore higher than pII(II,O) which is
indicative for strongly weathered soils with a nearly
complete removal of electronegative substances such
as silicic acid and an accumulation of electropositive
substances such as sesquioxides m a y be. This conclusion
is confirmed by the chemical and mineralogical analyses
as will be shown later.
Exchangeable Ca, Na, and K are constant throughout
the profile and amount to 3.5 m.e., 0.3 m.e. and
0.1 m.e./100 g of soil, respectively. Exchangeable RZg
decreases from 40.4 m.e./100 g of soil in the C,-horizon
to 14.1 m.e./100 g in the C1-horizon.F r o m thereon it
cannot be traced any more. T h e high value of exchangeable BZg in the C, and C1-horizon points to the presence of a smectitic clay mineral, which was confirmed
by mineralogical X-ray and DT-analyses.
T h e mineralogical changes during weathering and
soil formation are shown in Table 3.
The formation of chlorites, montmorillonite,and free
amorphous silica can be noticed in the lower part of
the profile. In the upper part montmorillonite and
chlorites are unstable and are transformed into kaolinite

47

J. van ScliuylenlJorgli

TABLE3. Normative mineralogical composition of prolile 6 (percentages given are by weight)'

Horizon

Ser

A w

Mt

S
?l

13.0
43.676.8

3.9-

Mm

Gi

Go

Kaul

/o

0.9
1.3
2.8
3.9
17.1
7.2

15.5
6.3
1.6
1.7

7.7
17.5
17.4
3.8

73.9
71.4
74.3
68.0
57.2
39.6
4.1
9.6

$111

18.3
15.5
15.7
07.2

1. Ser = serpentine; Aug augite; Mt = magnetitite; Sp E spinell; Chl= chlorite; Q

Go = goethite; Mise = miseciJaneuus (chromite + rutile).

and gibbsite.

The increase of kaolinite in the top soil

might be explained in various ways:

1. Actual enrichment in silica to form kaolinite from

gibbsite. Although this reaction is possible, it is
not likely to be the responsible factor here, as there are
no signs of a fluctuating, temporary, water table, which
could have enriched the surface with dissolved silica.
Vegetation could have supplied the surface soil with
silica by means of leaf shedding. Although data are
available on the amount of organic material falling
on to the soil, there is only little information on its
mineral content, especially with respect to SiO,, Al,03
and Fez03.S o m e data were published by V a n Schuylenborgh (1957, 1958), w h o analysed litter samples
of four rain forests in Java (Indonesia) on soils derived
from andesitic volcanic ash and of two rain forests
grown on ash-soils derived from dacitic material. T h e
results with respect to Si, Al, and F e are given in
Table 4. These data reveal that vegetation could have
contributed to the increase of the silica content in the
'surface soil, as the silica/sesquioxide ratios are m u c h
higher than of the solum. However, the available data
are too scarce to allow any general conclusion.

2. A second possibility m a y be found in the assumption that the process acting in the surface soil differs
from that in the subsoil, a suggestion already m a d e
earlier. U p o n the impoverishment of the surface soil
from nutrients because of prolonged leaching, the
bacterial decomposition of organic matter (mineralization) slows d o w n gradually; the consequence is that

TABLE4. Mean composition of litter of rain forests on Java

(Indonesia)
Forest on
soil from :

Andesite
Dacite

48

Si02

Ala02
Yo

Fen&

%
2.51
8.75

1.15
0.30

0.20
0.15

SiOr/AlxOs Si02IFezO3

%
3.8
49

42
146

5.8
21.5
05.8

= silica; Mn

montmorillonite; Kaul- kaulinite; Gi

msC
%
2.0
3.5
4.0
4.3
4.3
3.0
0.7

= gibhaite:

intermediate organic decomposition products come

into play in soil formation. S o m e of these products m a y
complex iron and aluminium and, if soluble complexes
are formed, cause an increased mobility of the metals
upon decomposition of the organic part of the complexes
or upon hydrolyzed and precipitated complexes. Consequently kaolinite is enriched residually.

3. Finally, it is possible that the initial stages of

desilication took place under drainage conditions poorer
than those of the present day. Schellmann (1964)
stated that weathering already started before the PreEocene upheaval of the Kukusan mountains and it is
therefore likely that the drainage conditions were poorer
than at present; consequently,transformation of kaolinite into gibbsite, which is favoured by prolonged and
intensive leaching (see Figs. 1 and 2), was not possible.
The deeper layers, especially the present B2-horizon,
have only minor amounts of silicate clay minerals:
apparently, drainage conditions improved gradually
upon upheaval with the subsequent complete destruction of the original minerals. Silica was leached and
gibbsite and goethite accumulated. U p o n increasing
weathering depths, leaching of the subsoil deteriorated
again and the formation of chlorites and even smectites
appears to be possible. Especially the presence of montmorillonite and amorphous silica 'in the C,- and C,horizon points to slow drainage (see Figs. 1 and 2).

As

a final conclusion it seems justified to assume tbat

mechanism 2 offers an acceptable explanation for the
peculiar distribution of minerals in the upper four
horizons.
The interdependence of amorphous silica and montmorillonite is interesting: in the Cz-horizon there is an
excess of silica over montmorillonite, whereas in the
C1-horizon the reverse is true ;moreover, serpentine has
disappeared in the C,. Apparently, montmorillonite has
been formed at the expense of serpentine and silica.
Chlorites seem to have wider possibilities for their
existence as they occur in the Cz-,C,- andB3-horizons
in equal amounts.

Weathering and soil-formingprocesses in the tropics

36
[MgZ'
log--

CH'?

32

8
a
b
e
d

e
f

THE PROCESS

It is evident that plinthization is absent and t h a t

desilicaiion is very similar to the profiles discussed
earlier.
As already mentioned, the desilication consists of a
considerable loss of silica and alkaline earths a n d
kaolinite has been formed via montmorillonite, chiorites
and amorphous silica. Apparently the rotten rock z o n e
is supersaturated with respect to silicic acid b e c a u s e
of the high solubility (weatherability)of the serpentinite
minerals and rather poor drainage conditions in the
weathering rock. U p o n better drainage in the layer
above this zone, silicic acid is leached and montmorillonite becomes stable. At still better drainage conditions
montmorillonite again becomes unstable and is transformed into kaolinite and gibbsite. These mineral transformations can be deduced easily from Figure 2.
Figure 6 gives a more exact stability diagram of
montmorillonite, kaolinite and gibbsite. It is based o n
an investigation of Reesman and Keller (1968). Also
this figure shows that upon stronger leaching, montmorillonite is transformed into kaolinite and gibbsite,

- parent material

= profile1
i

=
=
=

profile2
profile3
profile4

unirrigated soil.
Tsukikiri

28

24
gibbrite

ta

20

16

tf

I
-6

-5

-4

-5
lag:H.,

-2

SiO, 1

FIG.6. Stability fields of gibbsite, kaolinite, and montmorillonite in the system SiO,-Al,O,-Mg,-CaO-HZO at 250 C and
1 atm. total pressure. [Ca*+]= 5 x lo-,.

Bib1iography
ALEXANDER,L. T.; CADY, J. G. 1962. Genesis and hardening
of laterite in soils. 90 p. (Soil Cons. Serv. Techn. Bull.,
no. 1282.)
BARANY,
R. 1964. IIeat and free. energy of formation of muscovite. US.Bur. Mines, R.I. 6356,U.S.Dept. of the Interior.
-. , KELLEY,
K. K. 1961. Heats and free-energiesof formation of gibbsite, kaolinite, halloysite, and dickite. US.
Bur. Mines, R.I. 5825, U.S. Dept. of the Interior.
R. J. 1928. The soils of Cuba.
BENNETT, II. II.; ALLISON,
Washington, Trop. Plant Res. Foundation.
BOUMA,J. et al. 1968. O n soil genesis in temperate humid
climate. VI. The formation of a glossudalf in loess (silt
loam). Neth. J. agric Sci., vol. 16, p. 58-70.
BURRI,
C. 1964. Peirochemical calculations based on equivalents,
Jerusalem, Israel Program for Scientific Translation, 99 p.
(Translated from German.)

BLOOMFIELD,
C. 1957. The possible significance of polyphenols
in soil formation. J. Sci. Food Agric., vol. 8, p. 389-92.
1959. Mobilization of iron in podzol soils by aqueous
leaf extracts. Chern. 6% Ind., no. 9, p. 259-60.
CORRENS,C. W. 1961. The experimental chemical weathering
of silicates. Clay Min. Bull., vol. 4, p. 249-65.
DAVIES,S.N. 1964. Silica in streams and groundwater. Am.
J. Sci., vol. 262, p. 870-91.
CARRELS,
R. M.; CHRIST. Ch. L. 1965. Solutions, minerals,
and equilibria. New York, IIarper &Row. 403 p.
HALLSWORTII,
E.G.; COSTIN,A. B. 1953. Studies in pedogenesis in New South Wales. IV. The ironstone soils. J i
Soil Sci., vol. 4, p. 24-47.
HERBILLON,
A.; GASTUCIIE,M. C. 1962. Synthse et gnse
de i'hydrargyllite. C.R.Acad. Sci., Paris, vol. 254, p. 1105-7,

-.

49

J. van Schuylenborgh

HESS, P. C. 1966. Phase equilibria of some minerais in the

K,O-Na,O-Al,O,-SiO,-~~*O
system at 250 C and 1 atmosphere. Am. J. Sci.,vol. 264, p. 289-309.
IRVING,II.; WILLIAMS R.J. P. 1948. Order of Stability of
metal complexes. Nature (Lond.), vol. 162, p. 746-7.
KELLEY,
K. K. 1962. Heats and free-energiesof formation of
anhydrous silicates. US.Bur. Mines, R.I. 5901.
KRAUSKOPF,
K.I3. 1967. introduction to geochemistry. New
York, McGraw Hill.657 p.
LOVERING,T. S. 1959, Significance of accumulator plants in
rock weathering. Bull. Geol. Soc. Amer., vol. 70, p. 781800.
PEDRO, G. 1964. Contribution ltude exprimentale de laltration gochimique des roches crystallines. Thesis, Paris.
REESMAN,
A. L.; KELLER, W.D. 1968. Aqueous solubility
studies of high-alumina and clay minerals. Amsr. Min.,
vol. 53, p. 929-42.
ROBERTS,
R.C. 1942. Soil survey of Puerto Rico. Bur. Plant
Ind. (Ser. 1936, no. 8.). U.S. Dept. of Agriculture.
ROSSINI,F.D. et al. 1952. Selected values of chemical thermodynamic properties. Nat. Bur. Standards (Circ. 500). US.
Dept. of Commerce.

50

SCIIELLMAN,W. 1964. Zur lateritischen Verwitterung von

Serpentinit. Geol. Jahrb., vol. 81, p. 645-79.
SCIIWERTMANN,U. 1959. ber die Synthese definierter Eisenoxide unter verschiedenen Bedingungen. Z. anorg. allg.
Chem., vol. 298, p. 337-48.
SLAGER, S.; VAN SCIIUYLENBORGH,J. 1970. Morphology and
geochemistry of some clay soils of the coastal plain of Surinam
(S.A.). (In press.)
VAN DER PLAS, L.; VAN SCHUYLENBORGH, J. 1970. Petrochemical calculations applied to soils (with special reference
to soil formation). Geoderma. (In press.)
VAN SCIIUYLENBORGH,J. 1957. Investigations on the classi-

fication and genesis of soils derived from andesitic tuffs

under humid tropical conditions. Neth. J. agric. Sci.,
vol. 5, p. 195-210.
1958. On the genesis and classification of soils derived
from andesitic tuffs under humid tropical conditions. Neth.
J. agric. Sci., vol. 6, p. 99-123.
WALTIIER, J. 1915. Laterit in Westaustralien. Z. dtsch.
Geol. Ges., vol. 67B, p. 113-32.
YAKUSHEV, V. M. 1968. Influence of termite activity on the
development of laterite soil. Sowj. Soil Sci., no. 1, p. 109-11.

-.

4.Mineral weathering in relation to utilization

of soils
G. Donald Sherman'
College of Tropical Agriculture,
University of Hawaii (United States)

INTRODUCTION
Mineral weathering of rocks and soil-forming materials
under comparable environmental conditions will develop
soils or geological formations which have similar mineral
composition. The secondary mineral will have similar
physical and chemical composition. Jackson and
Sherman (1953)have described a sequence of secondary
mineral formation which is the result of the degree and
type of mineral decomposition. These authors have
identified these mineral products as stages of weathering
based on the susceptibility of minerals to decomposition.
Sherman (1948)and Sherman and Ikawa (1968)have
developed the concept of a sequence of soil development
which is related to the secondary mineral products
resulting from their pedogenetic weathering. In the
latter report they emphasize that soil genesis is the
product of weathering and leaching and that the
ultimate mineral composition is determined by the
weathering environment where the system reaches a
static dynamic equilibrium. Sherman (1952) described
a sequence of soils which developed under different
degrees of leaching and which can be identified by their
mineral composition. In this sequence clay minerals of
the montmorillonite group were developed under conditions of mineral decomposition and a lack of extensive
leaching where both the concentration of bases and
soluble silica are high in the soil solution. As leaching
increased and base concentration decreased with progressive dedication, the minerals of the Kaolin group
were formed and a different group of soils was developed.
Finally the processes of desilication proceed to a point
where the Kaolin minerals decompose giving rise to
the development of oxidic soils-a sequence of soil
genesis which occurs in the Hawaiian Islands under
conditions of free and unimpeded internal drainage.
Subsequent observations have shown that where the

leaching potential is great, both clay mineral groups

will be by-passed with the development of amorphous
hydrated oxidic soils, or if not so intense, an allophanichalloysitic soil will develop.
Matsusaka and Sherman (1950) have shown that
all soil families and series belonging to the Great
Soil Groups of the Hawaiian Islands have the s a m e
general mineral composition and similar physical and
chemical properties. Their management for agricultural
crop production could be based o n the properties
resulting from their c o m m o n mineral composition.
Pandey (1969)has shown that a c o m m o n relation can
be drawn between the L o w H u m i c Latosol (Kaisol)
and the kaolinitic Terra R o x a soils of Brazil and the
kaolinitic Red Earths of India. Subsequent research to
these important findings have shown that the soils of
each mineral combination (Great Soil Group) have their
o w n characteristic properties as to lime requirement,
base saturation, availability of plant nutrients and
water, fertilizer requirements, plant nutrient fixation,
water-holding capacity and movement, crop adaptation
and even their adaptation to mechanical cultivation.
These findings have important implications to future
policies governing the use of land and soil management
practices.
T h e greatest expression of mineral decomposition
and the formation of new secondary minerals occurs in
the soils of the tropical regions. A considerable a m o u n t
of research has been conducted in the Hawaiian Islands
and in the past by Dutch scientists in Indonesia to
identify the relationships which exist between climate,
drainage and parent materials on the nature of mineral
weathering and its products on the genesis of soils.
M o h r (1944)has described the relationship between the

1. Emeritus Senior Professor of Soil Science of College of Tropical Agrieuiture. University of IIawaii. and Chairman of Division of Soils and
Irrigation and Professorof Soils,American University of Beirut, Lebanon.

51.

G. Donald Sherman

internal m o v e m e n t of water within the soil and the

genesis of the soil. IIe described soils in terms of their
age where age refers to the degree of mineral weatliering.
H i s age descriptions of tropical soil development were as
follows in sequence of weathering :fresh, juvenile, virile,
senile and laterite. While soil scientists m a y differ with
this concept there is growing evidence that the age of the
soil is an expression in the form of a mineral composition which indicates the stage of weathering. It is in
this field of soil mineral weathering that the soil scientists of the Hawaiian Islands have directed m u c h of
their research activities. Up to about 20 years ago
practically all mineralogical research was centred
on the unweathered primary and the clay minerals of
the 2 :1 lattice-type minerals-hydrous micas, montmorillonite and interstratified groups. Most of the
soils were those which were in the early stages of
mineral weathering, the soils of the temperate region.
Their properties were dominated by texture and organic
matter rather than from the secondary minerals as in
the tropical regions.
The Hawaiian soils have been developed under conditions favouring rapid and almost complete decomposition of all primary minerals. The wide range of rainfall
and climatic conditions has m a d e it possible to study
the entire range of soil mineral weathering and relate
mineral composition to the physical and chemical properties of the soil. Because of the intensive development
of the soil management system in the Hawaiian Islands
it has been possible to m a k e an evaluation of the role of
soil weathering in soil management for crop production,
their engineering properties, mineral and industrial
uses and for their land-use classification.
T h e studies in soil mineral weathering were initiated
immediately after the completion of the soil survey of
the Hawaiian Islands (Cline, 1955). Tanada (1949)
found that as the rainfall increased above 20 inches
per annum, the concentration of kaolin clay minerals
decreased. At about 180 inches of annual rainfall the
1:l layered alumino silicate clay minerals cannot be
detected.
Subsequent research (Sherman, 1952) over the same
range of rainfall and in the same soils has shown
that as the kaolin minerals decreased the amount of
free oxides of aluminum, iron and titanium increased.
Fieldes et al. (1952) from studies in the Cook Islands,
Bates (1960), and Sherman et al. (1964)have shown
that as leaching progresses in both the removal of bases
and lowering of soluble silicate concentration in the
soil, the kaolin minerals become unstable and decompose. Their decomposition produces amorphous hydrated
colloidal oxides and hydroxides which on dehydration
will eventually become crystalline oxide and hydroxide
minerals. Sherman (1952)has shown that as weathering
progresses in Hawaii either by time of exposure to
constant leaching or to increasing intensity of leaching
due to exposure to different amounts of rainfall soil
could be identified which represented the peaks of

52

montmorillonite formation, kaolinizalion and freeoxide formation. T h e free-oxide formation was refined
to show peaks for ferruginous soils, titaniferous ferruginous soils and bauxitic soils (Sherman, 1955). Later
the amorphous hydrated oxides and hydroxides were
recognized by Sherman (1958) and the ferruginous
bauxitic soils (Sherman, 1967).
The basis for the recognition of the pattern of development of the oxides was established b y an earlier
work by Sherman (1949). In this study it was shown in
the humid tropical rain-forest regions of the Hawaiian
islands having an alternating wet and dry season, that as
the amount of annual rainfall increased, the concentration of Fe,O, and Tio, increased and the concentration of SiO, and A1,0, decreased. In the areas
which received sufficient monthly rainfall (exceeding
4 inches) as the annual rainfall increased, the A1,0,
content increased markedly and Tio, moderately. At
the same time the SiO, concentration decreased markedly and the Fe,O, content showed a moderate decrease.
These findings served to establish the role of desilication
as the intensity of leaching increased. The decrease in
soluble silicon in the soil solution will lead to the
complete decomposition of the layered alumino silicate
clay minerals. This can only occur under free-drainage
conditions. Impeded or restricted drainage stops the
process of desilication and the kaolin minerals stability
is enhanced.
The soils developed as the result of these weathering
conditions have served to apply their properties to their
utilization.

APPLICATION TO AGRONOMIC USE

T h e agriculturist has recognized that the physical
properties of the soils of the tropical block earths
(montmorillonite)and the L o w H u m i c Latosol-Tropical
Red Earths (halloysitic) soils differ greatly. Their
management for crop production varies greatly as to
their fertilizer needs, their lime requirements, and to
workability by mechanical implements and tools.
Likewise the soils of the Hydro1 H u m i c Latosol (amorphous hydrated hydroxide and oxide gels and colloids
of aluminum and iron) have an extremely high lime
requirement as shown in Figure 1. Likewise, their
capacity to fix phosphate in difficult available form is
abnormally high. On the other hand these soils possess
excellent physical stability which is surprising as they
often contain from 300 to 400 per cent water on their
dry basis in the field. With this high content of moisture
they are capable of bearing fairly heavy traction
machines.

RELATIONSHIP OF LIME REQUIREMENT

TO MINERAL COMPOSITION
T h e early applications of lime to very acid tropical
soils did not produce the anticipated beneficial results

'

Mineral weathering in relation to utilization of soils

based o n experience in the temperate regions. Agriculturists at that time did not have the knowledge of
the differences in the titration curves of the different
secondary minerals occurring in tropical soils. Their
experience was based o n soils of the temperate regions
where titration curves were based on a n organic matter
fraction, a large fraction of unweathered relatively inert
primary mineral and a relatively small fraction of
secondary clay minerals. Matsusaka and Sherman (1950)
have determined the titration curves and buffering
capacities of the soils of the tropical Great Soil Groups
occurring in the Hawaiian Islands. S o m e of the data
they obtained in this study is presented in Figure 1.
Each Great Soil Group represented soils having a
c o m m o n secondary mineral composition with one secondary clay or oxide mineral dominating in concentration
and exerting the major influence on the physical and
chemical properties of the soil. T h e buffering capacity
and the characteristics of the titration curve reflected
those of the major mineral constituent. In order to
convert the characteristics of the neutralizing action
of the titration curve to lime requirement in the field
soil it was found necessary to multiply the units of
calcium carbonate equivalents required for a desired
reaction change as shown by the curve and multiply

difference in the amount of calcium carbonate units

that would be required to change the soil reaction from
p H 4.0 to p H 7.0 for the soils of each of the tropical
Great Soil Groups. T h e soils of the indurated iron oxide
surface horizon of the H u m i c Ferruginous Latosol
would require the least lime for this reaction change,
approximately 2 tons per acre or about 4.5 metric
tons per hectare. In the case of soils of the IIydrol
H u m i c Latosol the amount would be 8 to 10 times
this amount or as m u c h as 45 metric tons per hectare.
T h e soils of the IIydrol H u m i c Latosol contain over
90 per cent hydrated colloidal amorphous hydroxides
and oxides of aluminium and iron.
In Figure 2 are given the titration curves of the soils
of the soil families belonging to the L o w H u m i c Latosol
group which contain more than 50 per cent halloysite
clay minerals. All soils of this Great Soil Group have
identical titration curves and thus would have the s a m e
general low-lime requirement for a given change in soil
reaction. Likewise these soils would acidify rapidly
with the application of acidifying fertilizer materials
such as a m m o n i u m sulphate. A soil of this group has
a recorded change of soil reaction from p H 7.2 to
p H 5.4 in 18 years due to the annual application of

by 1.2.
T h e data given in Figure 1 indicate a dramatic

FIG.1. Titration curves in CaCO, equivalents per acre of

typical soils of some IIawaiian Great Soil Groups :
A = Indurated A horizon of Titaniferous Ferruginous
Latosol-Ti-Fe crystalline oxides; B = Low Humic Latosolmore than 50 per cent halloysite clay; C = Humic Latosolless than 50 per cent halloysite clay + Al and Fe amorphous
and crystalline oxides; D = Tropical Black Earth-Montmorillonite clay;E = Titaniferous Ferruginous Latosol-hydrous
amorphous Te-Fe oxides; and F = IIydrol Humic Latosol90 per cent hydrous amorphous Al and Fe hydroxides and
oxides.

Tons Ca COj per acre

53

C. Donald Sherman

'

lonite and oxide soils and their respective oxide

minerals.
In the application of calcium carbonate (using the
titration curves), to oxide soils, very divergent results
were obtained, In the early 1960s it was found that the
application of calcium silicate gave consistent beneficial
results w h e n applied to bauxitic soils of Hawaii.
(Monteith and Sherman, 1963) In a later report it was
shown that a 3 tons of sugar increase per acre would
be obtained with the application of calcium silicate over
a comparable application of calcium carbonate w h e n
applied to the bauxitic Puhi soils of Hawaii (Sherman
et al., 1964). Calcium silicate is applied in preference
to calcium carbonate o n all ferruginous and bauxitic
soils of the Hawaiian Islands including the amorphous
hydrated oxide soils of the Hydro1 H u m i c Latosol
group.

PH
12.0

11.0

10.0

9.o

8.0

T H E HELATIONSHIP OF
FERTILIZER REQUIREMENT TO
MINERAL COMPOSITION A N D
WEATHERING STAGE OF SOILS

7.0

In both the tropical and temperate regions the efforts

6.0

5.0

4.0

20

40.

60

80

1 O0

m e . NaOH/100 g soil

FIG.2. Titration curves of typical soils of the soil families

belonging to the L o w Humic Latosol Great Soil Group.
N-1 = Molokai (the typic soil family of the group;
N-2 = Lahaina; N-3 = Wahiawa; N-4 = Kahana;
N-5 = Kohala an intergrade to oxidic soils;and N-6 = W a i a lua. The uniformity of curves make a pH-lime requirement
prediction scale possible for the group.
,

133 pounds of a m m o n i u m sulphate per acre per month or

150 kilogrammes per hectare per month. These data
indicate that with soils which o w e their origin to intense
chemical weathering, their lime requirement is determined by the dominant secondary mineral. T h e kaolin
rich soils will have their o w n PII lime requirement
relationship as will the tropical black earths-montmoril-

54

of farmers and agronomists have to a considerable

extent been devoted to the development of soil management systems which would exploit with the m a x i m u m
efficiency the nutrient elements released by soil minerals
by chemical weathering. T h e genetic improvement of
the rooting vigour of economic plants or the selection of
varieties tolerant to soil acidity or low nutrient level
are just some of the methods to adapt plants to soil
weathering conditions. Soil treatments are being
devised to increase the availability of the native minerals
of the soil. T h e application of fertilizer is a mcans to
augment the mineral deficiencies produced by weathering or by the lack of weathering.
In the early stages of weathering mineral decomposition often releases adequate amounts of nutrients for
the support of the growth of economic crops. In the
early stages of weathering the decomposition of feldspars and the formation and presence of hydrous mica
are to provide an adequate source of potassium. Thus
in general the early stages of mineral weathering,
especially in the tropical regions, are associated with
fertile and productive soils. T h e alluvial soils, soils
developed on relatively recent volcanic ash depositions
or the loessal soils represent some of the most productive areas of the world. Their high potential fertility is
due to the fact that their weathering conditions
are in the early stage and under the full expression of
the influence of the factors of chemical weathering,
thus facilitating the rapid decomposition of primary
minerals and the release of essential elements.
In the youthful stage of soil weathering the fertility
of a soil can IJC rclated to the presence of some of tlie
primary minerals which are very susceptible to chemical

Mineral weathering in relation to utilization of fioil6

decomposition such as the feldspars, biotite and plagioclase minerals. Hawkins and Graham (1950) and Grah a m (1949) have shown that the concentration of
feldspars and plagioclase feldspars occurring in the silt
fraction wag in direct proportion to the fertility levels
of the soils and in inverse proportion to the degree of
weathering. In the tropical regions soil mineral weathering is extremely rapid in the early stages. This causes
a rapid release of bases. Baren, quoted by Jackson
and Sherman (1953), concluded that the fertility of the
tropical soils of Sumatra was correlated with their
content of primary minerals. Mohr (1944) w h o has
divided weathering in the tropical regions into five
stages-fresh, juvenile, virile, senile and laterite-has
pointed out that plant growth increases rapidly in the
juvenile atage and reaches its m a x i m u m in the early
part of the virile stage. This is due to the m a x i m u m
release of plant nutrients to the soil solution which
maintains the base saturation at a high level thus
providing the point of highest capability of nutrient
availability in the soil. These soils have a high content
of the 2:l alumino silicate clay minerals.
Plant growth begins to decrease at the end of the
virile stage and decreases rapidly in the senile stage of
soil weathering. T h e end of the virile stage is where
one would expect to find the peak of kaolinization, a
point in soil mineral weathering where there is declining
availability of plant nutrients but it is also the point
of soil development where the best combination of
physical properties exist as to structure, ease of cultivation, and availability of water to the plant. The soils
of the peak of kaolinization provide the best opportunity
for intensive soil and crop management for m a x i m u m
crop production. It is in these soils that the nutrient
element deficiencies produced by the advancement of the
stage of weathering can be effectiveIy supplemented
b y the application of commercial fertilizer with greatest
economic return. It is also the point of soil development which offers the greatest efficiency in the use of
water under irrigation. Kaolin soils are easy to manage
under intense cultivation and maintain their stability
to erosion. T h e high production yields reported b y
M o h r (1944) and the sugar and pineapple industry of
the Hawaiian Islands attest to the high economic
productivity of soils at the peak of kaolinization when
fertilizers are used to supplement the decreasing
beneficial effects of weathering.
The senile stage of weathering is the peak of oxide
mineral formation in soils. It also represents the
advanced stages of leaching, thus the complete depletion
of bases or plant nutrients. The process of desilication
of mineral weathering has destroyed all silicates,
primary and secondary, leaving the system devoid of
soluble silica in any form. The capacity of the soil to
fix phosphate, native or added, has increased to its
peak due to its reaction with the oxides. T a m i m i et
al. (1963, 1964, 1967) and Liu et al. (1966)have shown
that w h e n phosphorus fertilizers are applied to these

soils along with a m m o n i u m and potassium fertilizer,

metallic aluminium (ferric) phosphate complexes like
tarakanite will form in the soil. The native vegetation b y evolutionary processes has adapted itself to
this low state of fertility of this stage of mineral
weathering in soils.
In the senile stage of weathering the oxide products
can be those offerrollitic weathering or those which
produce bauxite or aluminous oxides. T h e former produces the advanced or ultimate product of weathering
according to M o h r (1944). Sherman (1955, 1958) considers the formation of bauxite to approach the s a m e
stage as laterite except that its development is restricted
to regions of high rainfall and continuous unrestricted
leaching. Mohr's (1944) concept however would have
both the laterite and bauxite in the final product which
is possible under free drainage. However, the terruginous
soils of the senile stage have little to offer in the w a y of
agricultural production. They couple infertility with
unsurmountable physical problems which m a k e agricultural operations unprofitable. Their property of
induration on exposure produces a hard layer which
resists normal cultivation practices. Induration is
caused by the crystallization of the amorphous hydrated
oxides to crystalline goethite, hematite and maghemite.
Titanium oxides will indurate in p u c h the s a m e fashion
as iron-oxide soils but not have the same firmness.
They have the same infertility.
T h e bauxitic products of the senile stage, while
capable of induration, do not produce crusts or layers.
They form indurated fragments,nodules and aggregates.
Induration produces a loss of specific surface a n d cation
exchange capacity in the surface horizon but it retains
the more active surfaces of the amorphous hydrated
oxides of the subsurface. Because bauxification occurs
in areas of relativeIy high and well-distributed rainfall,
they offer an excellent potential for crop production
with proper fertilization and soil a m e n d m e n t applications. Sherman et al. (1964), and Monteith a n d Sherm a n (1963)have shown with the application of calcium
silicate that normal yields of crops can be obtained
with adequate fertilization. Adequate fertilization
requires an application 'of nitrogen to equal the total
plant requirement; the same is true for potassium, a n d
an extremely high applicatin of phosphorus is needed
to overcome the high fixation capacity and to provide
an adequate level of available phosphorus. Y o u n g e and
Plucknett (1966)applied as m u c h as 1,000 pounds per
acre of elemental phosphorus to these soils. T h e 1,000pound application has produced a high yield with no
decrease for 10 years, whereas the application of 250
and 500 pounds are n o w producing very unsatisfactory
yields. The discovery of the beneficial effects of the
application of calcium silicate has m a d e it economical
for the Hawaiian sugar industry to expand into these
areas to offset the loss of their highly productive kaolin
soils to urbanization and resort development.

55

G. Donald Sherman

PREDICTION OF FElTILIZER NEEDS

ACCORDING TO STAGE O F M I N E R A L
W EAT II E RING

T H E UTILIZATION OF MINERAL
W E A T H E R I N G IN IDENTIFICATION O F
ENGINEERING PROPERTIES O F SOILS
A N D E A R T H FOUNDATIONS

Very little lias been done in attempting to relate

management practices to the mineral composilion of
the soil. Agriculturists are using these properties but
indirectly. It is the contention of this author that both
lime and fertilizer requirements can be predicted from
the mineral composition of the soil and the nutrient
requirement of the crop. T h e author will go on to say
that these predictions will be more reliable than the
rapid chemical tests when applied to the complexity
of soils which are the products of the intense processes
of chemical weathering which occur in the tropical
regions. Let us take for example the lime requirement
of a soil. T h e lime requirement of a soil is determined
by the following: buffering capacity of the soil; the
desired change of soil reaction; and the p H and lime
tolerance of the plant. T h e first factor in the determination of the lime requirement is the mineral composition of the soil. T h e increased fixation of phosphorus
a n d the loss of potassium as weathering progresses is a
reflection of the creation of new minerals and the loss
of others. T h e author, while not including the hydrous
mica in this paper, is aware of their role in the maintenance of soil fertility in soils in some of the stages of
advanced weathering.

Soil engineers are just beginning to recognize the importance of the mineral of the weathered zone of the
earths surface,But o n the whole they are using parameters of soil mechanics measurement which are limited
to relatively unweathered earth surfaces of the temperate region. In the n e w residential area of Aina Haina
Valley of the city of Honolulu, scores of homes are
slowly moving downhill due to the lack of understanding of the poperties of a montmorilloniie clay on
a steep slope when its moisture and pressure relationships are changed. Movements of 4 inches per m o n t h
have been measured.
In contrast, homes are built on steeper slopes in
Honolulu and they are stable even though subjected
to same change of conditions. In this case the homes
are built on a ferruginous bauxite, a secondary mineral
product k n o w n for its extreme stability.

T H E UTILIZATION OF STAGES OF
M I N E R A L WEATHERING IN T H E
LOCATION OF MINERAL ORES
T h e development of the majority of the bauxite
deposits of the world o w e their origin to the mineral
weathering on the earths surface. T h e conditions for
the development of bauxite, iron ores and metal ores
h a v e been described b y Sherman (1955, 1958) and
S h e r m a n et al. (1968). These ore bodies represent the
weathering products of the senile and laterite stage
where the stable oxide minerals become dominant
residues of mineral decomposi tion.

T H E UTILIZATION O F MINERAL
WEATHERING STAGE IN L A N D USE
CLASSIFICATION
T h e population explosion and the increasing demands
placed on land for industry and urbanization has m a d e
it imperative that a plan of land use be developed in
order to protect our best productive lands for agricultural production, water conservation and our forest
resources. This important function is too often left to
social scientists and planning engineers w h o have no
background in the use of any of these resources except
as a user. At best they use the outmoded provincial land
capability index based on texture and not the true
capability of the land to produce economically.
Land-use classification should be done by a natural
scientist w h o has an appreciation for the economic and
social problems. In order to identify those areas of land
use, the potential for fertility management for the
greatest productive return must be done on the soils
chemical and physical properties which are fundamentally related to the stage of mineral weathering. Its use
would place land-use classification on a scientifically
sound basis and a basis on which natural-resource
planning can be placed in a defendable position.

Bibliography
BATES,
T.F. 1960. Rock weathering and clay formation in
Hawaii. Min. Ind., vol. 29, no. 8, p. 1, 2, 8.
CLINE,M.G. 1955. Soil survey of territory of Hawaii. Soil
survey series 1939, no. 25, p. 644. Washington, US. Dept.
of Agriculture.
FIELDES,
M.; SWINDALE,L.D.; RICHARDSON,
J. P. 1952.
Relation of colloidal hydrous.oxides to high cation exchange

56

capacity of some tropical soils of the Cook Islands. Soil

Sci.,vol. 74, p. 197-208.
GRAIIAM,
E.Il. 1949. The phagioclase feldspar as an index to
soil weathering. Proc. Soil. Sci. Soc. Amer.,vol. 14, p. 300-2.
~IAWKINS, R. II.; GRAHAM,E.R. 1950. Mineral content of
the silt separates of some Missouri soils as these indicate

Mineral weathering in relation to utilization of soils

the fertility level and degree of weathering. Proc. Soil Sci.

Soc. Amer., vol. 15, p. 347-54.
JACKSON,M. L.; SHERMAN,G. D. 1953. Chemical weathering
of minerals in soils. Advanc. Agron., N.Y.,vol. 5, p. 219-318.
LIU,I'.;SIIERMAN,
G. D.; SWINDALE,L. D. 1966. Laboratory
formation and characterization of Taranakite. Paci6 Sci.
(IIonolulu), vol. 20, p. 496-500.
MATSUSAKA,
Y.; SHERMAN, G. D. 1950. Tiiralion curves and
bufering capacities of Hawaiian soils. 33 p. (Haw. Agr.
Exp. Sta. Tech. Bull. no. 11.)
MOIIR,E. C. J. 1944. Soils of equatorial regions. Ann Arbor,
Edward Bros. 644 p.
MONTEITH, N. II.; SIIERMAN,G. D. 1963. Comparative efecis

Islands. 28 p. (Haw. Agr. Exp. Sta. Special Publ. no. 1.)

-.1958. Gibbsiie-rich soils of the Hawaiian Islands. 26 p.

(IIaw. Agr. Exp. Sta. Bull. no. 116.)

,CADY,J. G.;IKAWA, I.; BLOMBERG,N. E. 1967. Genesis
of the bauritic Halii soils. 46 p. (IIaw. Agr. Exp. Sta. Tech.
Bull. no. 56.)
-. , DIAS,I. I'.S.; MONTEITH, N. H.1964. Calcium silicate
a new hing material. Hawaiifarm Sci. (Honolulu), vol. 13,
no. 3, p. 8-9.
-. , IKAWA, II. 1968. Soil sequences in the Hawaii Islands.
PQC$ Sci. (Honolulu), vol. 22, p. 458-64.
-. , WALKER,
J. L.; IKAWA, H. 1968. Some of the mineral
resources of the Hawaiian Islands. (A revision.) 34 p. (Haw.
of application of calcium carbonate and of calcium silicate
Agr. Exp. Sta. Bull. no. 138.)
on the yield of Sudan grass grown in Ferrugmaus Laiosol
TAMINI,
Y.; KANEWIRO,
Y.; SHERMAN, G.D. 1963. Ammonium
and O Hydra1 Humic Latosol. (IIaw. Agr. Exp. Sta. Tech.
fixation by amorphous Hawaiian soils. Soil Sci., vol. 95,
Buli. 53.)
p. 426-30.
PANDEY, S. 1969. Comparison of properties of low Ifumic
. 1964. The reactions of ammonium phosphate with
Latos01 of Hawaii with Tropical Red Eearih of India.
gibbsite, and with montmorillonite and kaolinitic soils.
Ph.D. Thesis, Grad. School., Univ. of Naw., Honolulu.
Soil Sci., vol. 98, p. 249-55.
442 p.
. 1968. The effect of time and concentration on the
SIIERMAN,
G. D. 1949. Factors influencing the development of
reactions of ammonium phosphate with a humic latosol.
lateritic and laterite soils in the Hawaiian Islands. P Q C ~ .
Soil. Sci., vol. 105, p. 434-9.
Sci. (Honolulu), vol. 3, p. 307-14.
TANADA, T. 1951. Certain properties of the inorganic colloidal
. 1952. The genesis and morphology of alumina-rich
fraction of IIawaiian soils. J. Soil Sci., vol. 2, p. 83-96.
laieriiic clays in problems of clay and laterite genesis. SymYOUNGE, O. R.; PLYCKNETT,
D.L. 1966. Quenching the high
posium. St. Louis. Amer. Inst. Mining and Metal. Eng.
phosphorus fixation in Hawaiian soils. Proc. Soil Sci. Soc.
New York, N.Y., p. 154-61.
Amer., vol. 31, p. 653-5.
.1955:Some of the mineral resources of the Hawaiian

-.

57

II Tropical weathering
in Asia

Distribution of mica in the soils of the

Madhupur Tract, East Pakistan
1. U. Ahmed, M. S. Humain and M. R. Khan
Department of Soil Science, Dacca University (Pakistan)

INTRODUCTION
The study of soil mica is important from both the
fertility and pedogenic points of view. During weathering, this mineral releases potassium slowly and that is
why it has been regarded as a potential supplier of K
in soils. T h e amount of mica in soils and its distribution
in the profile is helpful in understanding the genetic
processes in soils. This is because the primary mica
m a y transform to secondary mica or illite under suitable
conditions (Jackson, 1956). In some Australian soils
Karim (1954) determined illite and observed that this
mineral is genetically related to those soils. Juang and
Uehara (1968),working with some tropical soils formed
from the same type of parent material, reported that
the amount of rainfall is related to the amount of mica in
soils. They also stated this mica to be pedogenic in origin.
There is considerable confusion regarding the use of
the terms illiteand micain soil science. Since mica
is a broader term which includes both primary and
secondary mica or illite in soils, the t e r m mica has been
used in this paper.
The M a d h u p u r Tract is the second largest Pleistocene
terrace in East Pakistan and extends over an area of
around 1,600 square miles. This tract which was
uplifted by earth m o v e m e n t is believed to be more
than 1 million years old and the red clays were derived
from the surrounding areas lying beyond the territorial
limits of East Pakistan (Morgan and McIntire, 1959).
Gulley erosion has dissected the Madhupur Tract, and
the soils have been classified on the basis of the
breadth of this dissection. T h e soils that occur on the
ridges or upland areas are red in colour because they
have been well oxidized. On the other hand, the soils
in the nearby valley regions are grey and remain submerged during the rainy season. Karim and K h a n (1956)
classified these soils as Gray Brown Podzolic.
Previously B r a m m e r (1964)reported that the soils of

the Madhupur Tract contain a small amount of illite

along with a high content of Kaolinite. T h e aim of this
paper is to identify and estimate the amount of mica
present in the clay fractions of the soils of the M a d h u p u r
Tract and to see h o w this mica is related to the pedogenic
processes.

E XP E RIM E N TAL
Three typical soil profiles were selected for this study
from locations at different elevations in the M a d h u p u r
Tract. T h e first profile was selected from the highest
elevation while the second profile was selected from an
intermediate elevation. Both these profiles are members
of Tejgaon series and have been classified as Typic
Dystrochrepts. The third soil profile was selected from
a valley-bottom and was designated as Karai1 series.
This soil has been classified as Cumulic H u m a q u e p t s
in subgroup level.
METHODS OF ANALYSIS

Organic matter in soils was destroyed by &O2 digestion. Free iron oxides in soils were removed b y the
dithionite-bicarbonate-citrate method of Mehra and
Jackson (1958). Free alumina and free silica in soils
were removed by boiling them with 2 per cent
Nazco3 for five minutes. Dispersion of soil materials
and the fractionation of the clay fraction (<2p
were done according to the procedures outlined by
Jackson (1956). The minerals in the clay fractions were
identified by X-ray analysis using a R a y m a x - 6 0 X-ray
diffraction unit and a powder camera. Total elemental
analysis of the clays was m a d e according to the method
given by Piper (1950).
1. Specially presented an supporting paper to review No. 1 b y J. J. Fripiat
and A. J. Herbillon.

61

I. U. Ahmed. Al. S. Ifussain and M. R. Khan

IIES U LTS
SOIL P R O P E R T I E S

T h e profile distribution of sand, silt and clay fractions

in the soils of the Madhupur Tract is shown in Table 1.
Sand represents a higher percentage in the soils of the
Tejgaon series compared to that of the Karail series.
Again, the sand content in the surface horizon of the
soils of the Tejgaon series is higher than that of the
subsurface horizons. There m a y be t w o reasons for
this :either (a) the clay fraction in the surface horizons
of the Tejgaon series has m o v e d d o w n the profile, or
(b) it has been washed a w a y b y surkce run-off and has
been deposited on the valley floor, and this could explain
the uniformly high clay content in all the horizons of
the Karail series which are situated on the valley floor.
The accumulation of clay in the subsurface horizons of
the high land soils (Tejgaon series) m a y lead one to
suggest that there m a y be formation of argillic
horizons.
Since this study is centred on the minerals present
in the clay fraction, o d y this particular fraction was
analysed with the help of X-rays. X-ray results showed
that Kaolinite was the dominant mineral in the clay
fraction. The next most important mineral in the clay
fractions was illite.
A very useful indication of the amount of mica in the
clay fractions has been derived from the amount of
K,O present in them (Jackson, 1956; Karim, 1954).
Mica was estimated from elemental K,O results according to the 10 per cent K,O complement proposed b y
M e h r a and Jackson (1959); the results are presented in
Table 2. T h e percentage of mica thus obtained in the
clay fraction ranges from 22 to 37 per cent in the
Tejgaon series, while in the Karail series the mica

conlent ranges between 34 and 36 per cent. Allocation

of all K,O to mica in the clay was justified since there
was hardly any fine-grained feldspar in this fraction.
Jt m a y be noted that the amount of mica in the clay
fraction decreases from the surface downwards in the
Tejgaon series whereas in the Karail series the amount
remained more or less constant throughout the profile.

DISCUSSION
The Karail series which is a young alluvial soil shows
a constancy of mica distribution throughout the
profile (Fig. 1). This is rather to be expected since the
soil materials in this profile did not undergo chemical
weathering long enough to bring about a change in the
mineralogical composition in this soil. Kimura (1966)
in a study of some alluvial soils in the tropics reported
that the content of mica remained uniform throughout
the profile. In other words, the amount of mica in the
clay fraction depended on the mineralogical composition of the parent material.
The two profiles of the Tejgaon series are residual
soils and are regarded as nearly mature soils in East
Pakistan. F r o m Figure 1, it m a y be seen that the mica
content in the clays of the Tejgaon series decreases
gradually with depth. This higher concentration of
mica near the surface m a y be related to the pedogenic
process. T h e climatic conditions under which these
soils have developed is tropical with around 60 inches of
rainfall in a year, most of it during the south-west
monsoon season. As a result the weathering intensity
in the Tejgaon series is high. It m a y be mentioned that
this area has a natural forest cover of deciduous
forest.

TABLE1. Distribution of sand, silt and clay in the soils of the Madhupur Tract
Soil aeries

Horizon

Tejgaon (1) (highland)

Tejgaon (2) (medium highland)

All
B*l

Karail (valley)

62

Depth ia inches

Sand

Silt

Clay

0-4
4-16
16-36
36-48

35.6
19.1
18.4
17.7

47.0
31.9
28.6
26.7

17.4
49.0
53.0
55.6

0-4
4-17
17-27
27-36
36-48

35.4
18.9
17.1
13.5
17.9

37.0
31.5
28.7
32.0
28.2

17.6
49.6
54.2
54.5
53.9

0-4
4-17
17-26
26-35
35.44

3.8
4.7
3.5
0.0
1.0

19.5
12.5
5.6
8.9
10.6

76.7
82.8
88.9
91.1
88.4

Distribution of mica in the soils of the Madhupur Tract, East Pakistan

--..
-.-.-.-

-.35

30

*-e-.-,

4
.
-

..

.-.

-.-.-.--

-e-.

2.

weathering action the mica present in the coarser

fractions (sand and silt) of the parent material has been

originally present in the sand and silt fraction has n o w

c o m e into the clay fraction and m a x i m u m accumulation
has taken place near the surface.
T h e second explanation for the higher concentration
of mica in the clay fraction of surface soils of the
Tejgaon series m a y be given in the light of the findings
of Juang and Uehara (1968).Because of recycling of K+
by the forest vegetation, secondary mica m a y form in
the soils under study. Since the concentration of K+,
released after the decomposition of organic matter,

4.-.25 yl

,,i

._
3

Tojgaonrerier(1)

.
N
'

~~j~~~

'
.
I

(21

.-,

. .<.:;,:
.,

..

TABLE2. Elemental analysis and mica content in the clay fractions of the soils of the Madhupur Tract

Soil seriee

Tejgaon (1)
(highland)

IIorizon

All

B*l

Ba*

c,

Tejgaon (2)
(medium highland)

Al1

B*l

B**
B**
Cl
Karai1
(valley)

APg
Allg

Al&
IIAIlbg
IIAIzbg

KzO

Calculated
mica in clay %

5.0
4.1
5.2
6.2

3.37
3.25
2.75
2.62

34
32
28
26

3.9
5.4
6.1
6.1
5.7

6.1
5.5
5.0
5.4
3.8

3.72
3.68
3.52
2.50
2.25

37
37
35
25
23

5.4
5.4
5.1
4.4
4.4

5.4
5.0
5.1
5.8
5.8

3.62
3.50
3.53
3.48
3.42

36
35
35
35
34

Depth
in inches

Si02

Alt03

Fez03

0-4
4-16
16-36
36-48

54.8
46.0
46.2
49.8

29.2
26.2
30.1
33.2

3.1
3.2
3.4
4.6

0-4
4-17
17-27
27-36
36-48

49.2
48.4
54.3
50.8
49.4

28.2
27.0
33.2
34.1
33.2

0-4
4-17
17-26
26-35
35-44

51.4
48.1
50.8
43.0
47.8

27.2
28.1
31.3
33.4
33.2

63

LU. Ahmed, b. S. IIussain and IX. R. Khan

the formation of secondary mica in soils by fixation

of K+ from soil solution. They reported further that the
secondary mica formed in soils is fine grained and is
usually found in the clay fraction of soils. It m a y be
suggested here that in the residual soils of the M a d h u p u r
Tract the mica near the surface is probably authigenic
and the supply of K+ has been provided by falling
leaves of the deciduous forest cover. The presence of
pronounced wet and dry seasons together with a long
wet s u m m e r which causes luxuriant vegetative growth
m a y facilitate the formation of this pedogenic mica.
Further investigation in this line m a y prove useful.
A n elemental analysis of the clay fraction showed

that the total Fe20, and Algo contents range from 3.1
to 6.1 and 4.1 to 6.2 per cent respectively (Table 2).
The clay fraction was defcrrated well, and it m a y thus
be suggested that a portion of the F e and AIg contents
is present in the crystal lattice of mica, since Kaolinite
is the only other dominant mineral in the clay fraction.
It is k n o w n that Kaolinite usually does not contain
either' M g or Fe in appreciable quantity, hence a
portion of F e and Mg m a y be allocated to mica. In that
case the mica in the clay fraction m a y be trioctahedral
in nature. Reesman and Keller (1967) reported this
kind of mica in soil clays.

Bibliography
BRAMMER,H.1964. A n outline of the geology and geomorphology of East Pakistan in relation to soil development.
Pak. J. Soil Sci., vol. 1, p. 1-23.
DYAL,R.S.; HENDRICHS,
S. B. 1952. Formation of mixed
layer minerals by potassium fixation in montmorillonite.
Proc. Soil Sci. Soc. Amer., vol. 16, p. 45-51.
JACKSON, M. L. 1956. Soil chemical analysis. Advanced
course. Department of Soils, University of Wisconsin,
Madison, Wisconsin.
JUANC, T. C.; UEHARA, G. 1968. Mica genesis in Hawaiian
soils. Proc. Soil. Sci. Soc. Amer., vol. 32, p. 31-5.
KARIM, A. 1954. A mineralogical study of the colloid fractions of some great soil groups with particular reference to
illites. J. Soil Sci., vol. 5, p. 140-4.
-. , KHAN, D.H. 1956. Coil of the Nanakhi series, East
Pakistan: II. Chemical investigation and classification.
Soil Sci., vol. 81, p. 389-98.

64

KIMURA,
H.S. 1966. Personal communication.
MEHRA, O.P.; JACKSON, M.L. 1958. Iron oxide removal
from soils and clays by a dithiouite-citratesystem buffered
with sodium bicarbonate. Clays and clay minerals,p. 317-27.
New York, Pergamon Press. (Monograph no. 5.)
-. 1959. Constancy of the sum of Potassium unit cell
surface and inter layer sorption surfaces in vermiculifeillite clays. Proc. Soil Sci. Soc. Amer., vol. 23, p. 101-5.
MORGAN, J. P.; MCINTYRE,
W.C. 1959. Quaternary geology
of the Bengal Basin, East Pakistan and India. Bull. Geol.
Soc. Amer., vol. 70, p. 319-42.
PIPER,C. S. 1950. Soil and plant analysis. Adelaide University
Press, Australia.
REESMAN,A. L.; KELLER, W.D. 1967. Chemical composition
of illite. J. sedim. Petrol., vol. 37, p. 592-6.

Trends in rock weathering in the southern

part of peninsular India - its expression
in morphogenesis of soils
.

J.

S. V. Govinda Rajan and R. S. Murthy

All India Soil and Land Use Survey,
Indian Agricultural Research Institute, New Delhi (India)

INTRODUCTION
Peninsular India as described b y Krishnan (1958)and
illustrated in Figure 1, is one of the three component
physiographical units of this sub-continent. It is an
ancient land mass owing its present features to denudation and weathering over long ages. T h e peninsular
mountains include the Western and Eastern Ghats,
Vindhyas, Satpuras and Aravallis. T h e chief rivers of
the peninsula are the Godavari, Krishna and Cauvery
and the west-ward flowing N a r m a d a and Tapti. T h e
Vindhyas and Satpuras separate the northerly from the
southerly flowing drainage. The peninsula thus encompasses a very large area. But the subject matter of this
paper is confined to the portion of the peninsula lying
to the south of the main expanse of the Deccan Trap
and the Krishna delta which includes most of Madras
n o w k n o w n as Tamil N a d u with the states of Mysore,
Kerala and part of Andhra Pradesh. A cross-section of
the plateau is illustrated in Figure 2.

GEOLOGY
O n e of the most conspicuous features of the ancient
rocks of the Indian peninsula is a profound unconformity
separating a .highly compressed and metamorphosed
assemblage from an overlying set of beds which are
most appreciably folded and have undergone comparatively little mineral change (Pascoe, 1965). Exposures
of the oldest rocks forming the Archaean complex cover
a vast area in the peninsula, nearly three-fifths,which
can be seen from Figure 3. Also termed crystallines,
they comprise two types :gneissic and schistose. T h e
schistose group consists of the Dharwar system under
which are grouped the Dharwar rocks of Madras and
Mysore. T h e peninsular gneisses foliated and banded
form a heterogeneous suite, the members of which vary
greatly in age. With the exception of Dolerite dykes

and some Pre-Tertiary deposits of laterite, the state

of Mysore is composed entirely of Archaean rocks.
T h e rocks of the Dharwar system in south India
consist principally of hornblende schists, chlorite schists
and mica schists. There is a progressive increase in
metamorphism in the Dharwar rocks from the Dharwar
and Bellary districts southwards across Mysore, i.e.,
in the charnockite region of Salem and Coimbatore.
F r o m west to east also across northern Madras towards
the charnockite province of the Eastern Ghats, there is
a general increase in schistocity in the rocks of the
Dharwar belts.
In the north of Mysore and adjacent parts of Madras
where the Dharwar schists are well developed, chlorite
is the conspicuous constituent. In the centre of Mysore
where the narrow Dharwar bands are wedged in
between the masses of granite, dark hornblendes are
conspicuous in the basic.rocks. ,In the south where
there are no wide belts of schist, the constituent minerals
of all types are clean and fresh, coarsely crystalline and
granular in form. The hornblende schist of the Dharwars of south India is generally tough, dark greenish
grey in colour. The chlorite as well as the hornblendic
schists affords evidence of extensive igneous activity
though a large proportion of it appears to be of sedimentary origin.
The charnockite suite has a wide distribution in the
southern and eastern parts of peninsular India. It
covers a large prdportion of the country south of a line
connecting Mangalore, Bangalore and Madras and is
found again in the coastal parts of Nellore and Eastern
Ghats. It occupies the highest peaks as well as most of
the loftiest hill ranges and plateaux of peninsular India.
The Cuddapah beds of Madras occupy a crescentshaped basin along the middleof the eastern side of the

1. Specially presented an oupporting paper to review No. 2 by P. Segalen.

65
I

S. V. Govinda Rajan and R. S. hlurtliy

FIG.1. Peninsular India.

66

Trends in rock weathering in southern peninsular India

Western coast
(very wet)

Western Ghats
(very wet)

Rain shadow area, Deccan

(dry)

Eastern Ghats
(lessdry)

sea

FIG.2. Peninsular India-section across plateau.

Bombay

FIG.3. Peninsular India-geology.

67

S. V. Govinda Rajan and R. S. Murthy

peninsula. T h e outcrop covers most.of the Cuddapah

'and Kurnool districts. In contrast to the surrounding
gneissic country, the hills of this large tract are m a d e
of precipitous scarps and gentle dip slopes. T h e Cuddapahs are essentially a succession of quartzite sandstones or quartzites and slates or shales.

13

100 c m to ZOO c m

CLIMATE

50 c m to 100 c m

Peninsular India receives most of its rain from the

south-west monsoon and hence rain falls from June to
October. Figure 4 is a simple rainfall m a p of the peninsula for the year. According to Dudley S t a m p (1965),
it can be divided into three main divisions : (a) good
rainfall division with more than 200 c m of rain in the
year; (b) moderate rainfall division wih 100 to 200 e m
of rain in the year; (c) poor rainfall division with 50 to
100 c m of rain in the year.
In most parts, rains cease in September or October
but the Madras coast gets a considerable amount of rain
between October a n i December. This rainfall comes
mainly from storms which occur during the period w h e n
the south-west monsoon is retreating before the northeast monsoon has properly established itself. T h e centre
of peninsular India lies in the rain shadow of the
Western Ghats.
As regards the temperature in January, the sun is
vertically a long way'to the south of India and gets
cooler from south to north from a n average of 800 F
to 550 F. T h e isotherms are shown in Figure 5.
Those places which are a very long w a y from the sea,
especially in the dry regions, have hot days and cold
nights, thus leading to a wide range in temperature.
Similarly dry regions, a long w a y from the sea, have a
big annual range of temperature. T h e difference between
the hottest and coldest months in the north which will
b e 390 F can be compared with coastal parts of Kerala
having a n annual range of 50 F only.

4*
1

Narm

ratei

FIG. 5. Peninsular India-isotherm.

N A T U R A L VEGETATION
A s seen from Figure 6. forests cover considerable areas
in the peninsula. Owing to heavy rainfall (over 2CO c m
per annum) luxuriant forests with coconut palms are
found along the Malabar Coast, while the windward
slopes of the Western Ghats are also thickly wooded;
very different is the Coromandel Coast, where, owing
to the lower annual rainfall (under 100 cm), long
stretches are covered with evergreen forests which,
however, give w a y in the deltas to fertile irrigated
tracts. On the plateau itselfrainfall also provides the key
to natural vegetation. T h e north-west has less rain than
the south-east and the driest part is the belt lying in
the rain shadow of Western Ghats. Despite cultivation
m u c h of the original vegetation remains. In districts
where the annual rainfall does not exceed 200 c m but
Y

68

Principal cultivatt?d area

St

forest
forest

FIG.6. Peninsular India-natural vegetation.

Trends in rock weathering in southern peninsular India

Bombay

R e d sandy soils
M e d i u m black soils
Red loamy soils
Deep black soils

- a

Mixed red and black soils

u
mn
....
.....
....

FIG. 7. Peninsular India-soils.

Laterite soils

..

Coastal alluvium
Shallow black soils
8

is more than 100 crn, there are open monsoon forests

with teak and other deciduous trees which shed their
leaves in the dry season. B u t such forests are by n o
means continuous and are interspersed with stretches
of scrub while a further contrast is provided by the
densely wooded river valleys.

WEATHERING PROCESSES
Several investigators in the field of tropical science have
realized that neither the climate as such, nor the rock
is the single, predominant factor in soil formation
(Mohr et al., 1954). According to Milne et a. (1936)n o
single physical factor is predominant for the whole
region hut o n some occasions the climate and in others
the parent material or the topography is decisive.
Numerous instances have also been reported in which

the factor of surface geology intervenes as the ruling

clement.
Each of the soil-formingfactors such as climate, rock
topography and plant growth has played a part in soil
formation in peninsular India. T h e scientific approach
to the study of these soils should be m a d e by carrying
out a proper analysis in the field of the characteristics
of the soil profile and in the laboratory of the properties of the individual horizons.

SOILS, THEIR PROPERTIES

A N D COMPOSITION
T h e m a p showing the distribution of soils in the southern
part of the peninsula is presented in Figure 7. These
soils, particularly the Black, R e d and Laterite coils
which are discussed in this paper, m a y be closely examined in relation to the geology of the region.

69

S. V. Govinda Rajan and R. S. lurthy

B L A C K SOILS

Newbold (1846) has cbtimated that at least one-third

of southern India is covered by black soil also called
Regur. The parent materials that have given rise
to black soils are generally basic in character being
relatively high in one or more of plagioclases, ferromagnesian minerals and calcium and magnesium carbonates. They develop under climates with high average
temperatures, m e d i u m to low rainfall and a marked
dry season with alternation of wet and dry periods
(Dud,al, 1965). They generally have shallow to deep
profiles. Regur is found everywhere on the plains of the
Deccan Trap country. In south India, tracts of black
soil are scattered throughout the valley of the Krishna
and occupy the lower plains and flats of Coimbatore,
Madura, Salem, Tanjore, R a m a n d and Tinnevelly; on
the Mysore plateau there is but little. S o m e occurs on
portions of the coastal plain along the eastern shore of
the peninsula and the great alluvial flat of Surat and
Broach in eastern Gujarat.
In m a n y cases the black soil is directly derived from
the basalt by surface decomposition. Schists, calcareous
shales and slates have also given rise to black soils.
Although the soils are derived from a variety of parent
materials, whether of sedimentary, metamorphic or
volcanic origin, they share a number of important
characteristics. The soils are mainly dark in colour, fine
in texture and low in organic matter. T h e y expand and
contract appreciably with changes in moisture and
commonly lack distinct horizons in their profiles. They
are confined to land forms having low relief. Topography is most often undulating to level. The slope of
the major proportion of such soils does not exceed
5 per cent. B y and large, infiltration rates arc low to
moderate. There is relative accumulation of salts in the
lower horizons and tendency to a downward increase
of sqdium in the exchange complex. Electrical conductivity of 2-6 m m h o s has been measured in the surface
horizons. The p H is generally on the alkaline range and
the high p H is associated with high percentage of
calcium and magnesium.
The texture of black soils is generally clay but occasionally silty clay or silty clay loam, the clay content
of most of the profiles ranging between 40 and 60 per
cent. T h e silt proportion amounts to 25 to 40 per cent
of the total soil. Gravel m a y occur locally in the form
of calcareous or iron-manganese concretions. The cation
exchange capacity in the surface horizons ranges from
40-60 me/100 g. Calcium saturation in the exchange
complex is of the order of 60 to 80 per cent and magnesium 10 to 30 per cent.
Although the fertility level of black soils varies widely
in relation to the kind of parent material from which
they arc derived, the nutrient status reveals several
features in common. They are relatively low or even
deficient in phosphorus and nitrogen. Potassium m a y
be present at a moderate level but is sometimes

70

unavailable owing to potassium fixation. Calcium sulphate occurs in a finely divided shape as gypsum
crystals.
The synthesis of expanding clay minerals takes placc
in the presence of alkaline earths in the weathering zone.
Magnesium plays a role in the synthesis of montmorillonite while calcium maintains a favourable level of PII
for its formation. T h e alkaline earths are released by
the weathering of basic rocks.
LATERITE

T h e bulk of the laterite of peninsular India is found

near the Western Ghats. T h e country rock supporting
any cap of high-level laterite consists of basalt, epidiorite, shist, gneiss, granite and other types c o m m o n in
the peninsular region. Most frequently laterite forms
flat or slightly undulating plateaux on the Deccan Trap.
In the Malabar Coast of Kerala, South Kanara and
Shimoga, laterite has been formed from the underlying
granite rocks d o w n into which it passes through a K a o linized zone. In the plain of South Kanara the horizontal
table-like capping of laterite occurs on peninsular
gneisses, large boulders of which can be seen altering
into massive laterite. In Vizagapatam, its occurrences
are mostly confined to the Archaean schists from which
they have been directly derived.
Primary laterite passes d o w n into the underlying
parent rock, whether this be igneous, metamorphic or
sedimentary. If the underlying rock is basalt or gneiss,
its upper part is decomposed or kaolinized to form a
clay or lithomarge passing upwards into the laterite.
This intermediate material is extremely variable in
character but always more ferruginous above than
below. In m a n y other places, this lithomargic zone
bhows no tendency to pass into the underlying rock
though usually exhibiting a marked transition into the
laterite above. In such cases, the laterite and lithomarge
together constitute a group of beds superposed confors
mably upon older rocks of various kinds. Such conditionhave been observed in parts of the central peninsula,
the laterite resting with a clear junction upon the trap
like a distinct formation.
Most of the secondary laterite is found in the low-lying
coastal regions on the eastern and western sides of the
peninsula. O n the east coast, laterite occurs almost
everywhere rising from beneath the alluvium which
fringes the coast and sloping gradually upwards towards
the interior. It is m u c h less a massive formation than
the rock of the west coast and consists of large rounded
or sub-angular fragments of gneisses and other rocks.
The latenzation of any rock involves the disappearance in solution of the silica, lime, magnesium and
alkalines of the parent rock with the concretion of the
hydrated oxides of aluminium, iron titanium and sometimes manganese. The essential feature of laterite is the
small amount of combined silica in proportion to the
alumina present. Laterite and clay are in fact the end

Trends in rock weathering in southern peninsular India

products of t w o distinct modes of decomposition. T h e

presence of free silica especially in the form of quartz
is a n indication that the latente is of detrital origin.
Laterites as described by Maignien (1966) vary in
colour but are usually bright. T h e shades commonly
encountered are pink, ochre, red and brown but s o m e
occurrences are mottled and streaked with violet. A
single sample m a y exhibit a whole range of colours.
This variegated colour pattern is due to iron oxides in
various states of hydration and sometimes also to m a n ganese. Alumina is often found mixed with iron in hard
pans. Silica which is ordinarily whitish and impregnated
with the hydroxides of iron, yields a red or rust colour.
Although estimation of colours gives only a rough idea
of composition, it makes it possible to estimate the level
of evolution.
Density varies within fairly considerable limits in
relation to chemical composition, increasing with iron
content and decreasing with alumina content. T h e
specific gravity of oxidized forms is higher than that of
hydrated forms. Old crusts are denser than recent
crusts.
High content ofethe sesquioxides of iron and/or aluminium as related to the other components is a unique
feature of laterites. T h e laterites of Kerala contain 12
to 15 per cent Fe,O3 and.13 to 18 per cent Al,O,, while
the laterites of G o a contain 15 to 20 per cent or more
Fe,O, and 15 to 20 per cent Alzo,. S o m e laterites
contain manganese, sometimes in sufficient quantity to
be exploited as a mineral (Karwar and Goa). Quartz
is a n important component and some of the laterites
in the peninsula contain more than 20 to 25 per cent.
There have been some attempts to classify laterites o n
the basis of chemical composition but as demonstrated
by F o x (1936), such classifications are inadequate.
R E D SOILS

T h e ferruginous soils c o m m o n on the surface especially

of the metamorphic formations occupying extensive
areas in the peninsula are quite a contrast to the black
soils. T h e most c o m m o n form of red soil which is
classified as rcd sandy or red loamy derives its colour
from ferric oxide resulting from the decomposition of
the rock in situ or from the same products of decomposition transported to a lower elevation by rain-water.
Hardy (1949)quotes a statement to the effect that
soils developed under a rainfall of less than 875 m m
are reddish brown in colour and this appears to be true
of peninsular India. H e c a m e to the conclusion that

there is specific relationship between rainfall zones and

soil groups. T h e same type of soil can develop under
vastly different climatic conditions depending solely on
parent material. O n e of the main types is the coarsetextured sandy soil developed from granite or granite
gneiss. It has a finer textured and heavier appearance
in the high-rainfall area.
T h e weathered product from crystalline rocks in the
peninsula has undergone more complete dcsilicification
and is consequently more sesquioxidic. T h e predominantly red colours m a y be attributed to free ferric
oxide with a comparatively low degree of hydration.
T h e characters of tropical soils are therefore dominated
by their parent materials. T h e red colours and the
presence of high proportions of sesquioxides should be
attributed rather to rock weathering than to profile.
development. Owing to high rate of evaporation the
soil becomes quite dry and hard between the rains.
This accounts for the frequent occurrence of concretionary materials of varying sizes. T h e base status
shows a wide range from low to m e d i u m and occasionally
high. T h e p H is not so low as might be expected, a
circumstance attributable to the sesquioxidiccharacter
of the clay complex. There is no marked chemical
eluviation and the structure is generally friable and
granular.
T h e red soils are not always confined to a mere zone
o n a particular contour of topography but are also to
be found o n level ridge tops and sloping valley sides as
well. T h e hard pan soils are the characteristic response
of a parent material consisting of granite or granite
gneiss decomposed by the action of a semi-arid climate
existing o n a topography which includes steep hillsides
or bottom lands.

CONCLUSION
T o conclude, in peninsular south India where high
temperatures and high to moderate rainfall are prevalent, the chemical disintegration of rocks and the
subsequent end product of the processes of alteration
rind decomposition, i.e., soil, are directly dependent o n
the properties and features of the parent material. T h e
rock structure, texfure and hardness together determine
its physical resistance to katamorphism. T h e chemical
composition, crystal structure and susceptibility to
alteration and decomposition are important from the
pedologic and agricultural points of view.

71

S. V. Govinda Rajan and R. S. Alurthy

Bibliography
DUDAL, R. 1965. Dark clay soils of tropical and subtropical
regions, Rome, Food and Agriculture Organization of the
United Nations. 24 p. (Agricultural Development Paper
no. 83.)
DUDLEYSTAMP,L. 1965. The worfd. Bombay, Calcutta,
Madras, New Delhi, Orient Longmans. 99 p.
Fox, C. S. 1936. Buchanans laterite of Malabar and Kamara.
Records Geological Survey of India, Calcutta, vol. 69,
p. 389-422.
HARDY,
F. 1949. Soil classification in the Caribbean region,
p. 64-75. Harpenden (Herts.), Commonw. Bur. Soil Sci.
(Tech. Commun. 46.)
KRISHNAN,
M.S. 1958. Introduction to the geology of India,
p. 1. Madras, IIigginbothams.

MAICNIEN,R. 1966. Reeiew of research on laterites. Paris,

Unesco. 148 p. (Natural resources research IV.)
MILNE,
G. et al. 1936. A provisional roi! map of East Africa.
Amani, East Africa. 34 p. (Amani memoirs.)
MOHR,E.C. J.; VANBAREN.F.A. 1954. Tropical soils. The
Hague and Bandung, N.V. Uitgeverij W. Van IIoeve.
125 p.
NEWBOLD.
1846, Regur soils. J. A. Asiat. Soc., London, vol. 8,
p. 252.
PASCOE,
E.H. 1965. A manual of the geology of India and
Burma, vol. I, p. 57. Delhi, Manager, Govt. of India Press.

Weathering of basic igneous rocks and genesis

of clay minerals1
R.V. S. Satyanarayana
Indian Agricultural Research Institute, New Delhi (India)

INTRODUCTION
Basic igneous rocks on weathering gave rise to laterites
in Malabar and South Kanara (10015 to 13059 N.
and 74043to 76015 E.).The area is characterized by a
tropical humid climate with a rainfall of from 130 to
150 inches, a m e a n annual temperature of 800 F and a
pronounced wet and dry season. In M a l w a Plateau
22022 to 2902 N. and 74032 to 76028 N.) with an
elevation of 1,600 feet (MSL)basaltic rocks underlie the
black, red and brown soils and isolated laterite r e m nants. The climate is arid to sub-humid. The rainfall
varies from 30 to 50 inches, a major part of which is
received during the period July to September. The annual m a x i m u m and m i n i m u m temperatures arc 1100 F
and 450 F respectively. Data are presented here on the
weathering and clay mineralogy of soils in the two areas.

LATERITES OF M A L A U A R A N D
SOUTH K A N A R A
Satyanarayana and T h o m a s (1961)developed morphological concepts. and horizon designations for the
B u c h a m a n type laterite occurring in the type area
Angadipuram of Malabar and South Kanara. Satyanarayana and T h o m a s (1962)studied two in situ laterite
profiles from the type areas Angadipuram and Kasaragod to follow the changes undergone in composition
of the rock in its development into laterite and overlying
soils. They concluded that, in the first stage of weathering of rocks, the prevailing differences of the primary
minerals are levelled d o w n by the loss of alkalis and
alkaline earths and gain in water in the two types of
basic rocks examined by them. In tlie next stage,
resynthesis of the weathered products is possiLiy tlie
same as indicated by the general trends of accumulation

of kaolin, gibbsite or bauxite and iron in the profilc.

T h e chemical data of the rock, the laterite and soil
layers is re-calculated assuming,constant Ti percentage
and the losses and gains are worked out. At Angadipuram, the hornblende granulitic rock lost 35 per cent
in the first stage of weathering to mantle 41 per cent
to the laterite layer and the surface soil presents a
further loss of 5 per cent and is equivalent to 36 per
cent of the original rock. In a laterite developed on
basalt at Kasaragod (South Kanara) the loss was of
46 per cent to give only 38.6 per cent of the original rock
in the mantle and 32 per cent in clay horizon but the
return of bases by plants to the surface soil was greater
on the basic than on the acid rock. T h e mineralogical
composition of the correspondinghorizons and cation exchange capacity of the two laterites is given in Table 1.
The data reveal the marked differences in the mineralogical make-up of corresponding horizons in the t w o
profiles. The profile from Angadipuram contains predominantly more kaolinite and small quantities of
other crystalline compounds while in the Kasaragod
profile kaolinite content is slightly less but the crystalline
compounds are m u c h higher. Amorphous compounds
of iron are present in both the profiles, the basic rock
contributing a large amount. In addition the Angadipuram profile shows the presence of some amorphous
aluminium and silica compounds. The cation exchange
capacity is in fair agreement with the nature of the clay.
T h e laterite from basic rock is highly aluminous and
in the case of m e d i u m acid rock it is predominantly
kaolinitic. The presence of high amounts of quartz in
acid rocks possibly permits the development of red
loams and red earths of great depth without laterite
structure.

1. Specially preaented as suppurling paper to review No. 2 by I. Segalen.

73

li. V. S. Satyanarayana

TABLE 1. Mineralogical cornposition and cation exchange capacity of laterite profiles

IIard laterite

Surface soil

A1203

SiO,
c.e.c.

Kgq.

nL.,

Ang.

Km.

3.2
44.0
4.0
33.0
12.0

28.5
51.8

9.2
26.0
32.0
1.0
29.0

42.6
13.9
1.0

...

...

6.6
50.0
12.0
13.0
13.0

4.5

1.4
3.3

Kgq.

DLIL2

Ang.

K gq.

36.7
33.8
6.7
3.9
16.0

10.5
45.0
8.0
10.0
24.0

23.9
54.4

...

...

...

...

4.6

6.5

...
...
13.4
2.1
2.3
5.4

Weathered rock

An&.

Ang.1

Quartz
Kaolinite
Gibbsite
Limonite
Remaining Peso,

Soft laterite

BLsh

...
12.5
0.8

...

...
...
5.2

...
14.0
11.8

...
...

...
...
...

1. Ang.-Angadipurarn; Kgq. =Kasaragod.

'

LATERITE A N D B L A C K SOILS
O F M A L W A PLATEAU
Pendleton (1947) attributed the formalion of laterites
in Mandsaur District of M a l w a Plateau to have taken
place in the Tertiary period, before the Western Ghats
cut off a great deal of monsoon rain, making the climate
semi-arid from that period onwards. Sahi and Satyanarayana (unpublished) observed a relict B-horizon of
laterite in M a l w a Plateau wilh a n occasional thin cover
of red soil or reddish-brown soil o n gentle slopes a red
lateritic soil with lime concretions below and black soils
at the lowest end of the catena.
Krishna Murti and Satyanarayana (1969)investigated
the role of environment o n montmorillonite formation
from basic igneous rocks of this region. Multiple correlation analysis o n the chemical data of the clay fraction
of the soils indicated that with increased entry of iron
in the octahedral layer the formation of montmorillonite is lea possible. T h e clay mineralogy of the soils
indicated that kaolinite is absent in the black soils and
is nearly 40 per cent in the lateritic soils. Thus the
presence of iron and magnesium in the chemical environment with a low oxygen tension is esscntial for montmorillonite formation.
Jackson (1968) observed that feldspars weather
readily to kaolinite hut have insufficient silica potential
to form montmorillonite directly. H e concluded that
poor drainage and production of reducing conditions

TABLE2. C a t k exchangecapacity'
Soil no.

33
36
38

greatly influences the course of pedochemical weathering

of minerals by affecting the activity of Fe, Al, Mn and
Si normally maintained at low levels in well-oxidized
and well-leached boils.
T h e data (Table 2) o n the cation exchange capacity
(c.e.c.) of the lateritic and black soils and their silt
and clay fractions indicated a definite contribution of
the coarser fraction to the c.e.c. of the soils. T h e mineralogy of these fractions indicated the predominance of
montmorillonite (Satyanarayana et al., 1969). T h e high
c.e.c. values of the coarse size fractions of these soils
find support from the views expressed by Flach et
al. (1968). They observed that under the influence of
pedogenic processes the transition from saprolite to soil
B horizon is accompanied by a change in the texture
although mineralogical composition and c.e.c. remain
constant.
Further, the work to characterize the component
responsible for the high c.e.c. values of these size
fractions was attempted. T h e silt and fine-sand fractions of the weathering zone were separated into five
specific gravity fractions viz. > 2.86; 2.86-2.62;2.622.50; 2.50-2.35;and < 2.35 using bromoform-alcohol
mixtures. T h e specific gravity fraction (<2.35) showed
c.e.c. value of 80-90me/100 g. This fraction is highest
in the zone of weathering and decreases the solum to
the surface, a point of interest in the weathering of
basalts.

of the soil, silt and clay

Nature of soil

lleddish brown (lateritic)

Black
Reddish brown (lateritic)

Depth inches

0.7
0.6
5-10

Soil

Silt (2-20p)

Clay (<2 d

25.9
60.6
34.7

36.5 (3.6)*
58.1 (9.4)
40.6 (5.7)

51.7 (20.2)
95.7 (50.4)
67.5 (32.1)

1. T h e c.e.c. W U B meaaured by the modilcd a m m o n i u m aretnte meilod.

2. T h e values in parenthescs indicate the contributions of the various size fractions to the c.e.r. of ihe soils.

Weathering of basic igneous rocks and genesis of clay minerals

Bibliography
FLACII,K. W.;CADY, J. C.; NETTLETON, W.D. 1968. Pedogenic alteration of highly weathered parent materials.
Trans. 9th Congr. Int. Soc. Soil Sci. (Adelaide), vol. 4,
p. 343. Amsterdam. Int. Soc. of Soil Science.
JACKSON, M. L. 1968. Weathering of primary and secondary
minerals in soils. Trans. 9th Congr. Int. Soil Sci. (Adelaide), vol. 4, p. 287. Amsterdam, Int. Soc. of Soil Science.
KRISHNAMURTI,G. S. R.; SATYANARAYANA,
K.V. S. 1969.
Significance of magnesium and iron in montmorillonite
formation from basic igneous rocks. Soil Sci., vol. 107.
PENDLETON, R. L. 1947. Soils of India: Four soil surveys in
Gwalior State. Soil Sci., vol. 63, p. 421-35.

SATYANARAYANA,
K.V. S.; SAIII,B. P.; KRISHNA
MURTI,
G. S. R. Cation exchange capacity of basaltic soils of
Malwa Plateau. Curr. Sci., vol. 38.

-.

,THOMAS, P. K. 1961. Studies on laterites ana associated

soils. I. Field characteristics of laterites of Malabar and
South Kanara. J. Indian Soc. Soil Sci., vol. 9, no. 2,
p. 107-18.
-.
. 1962. Studies on laterites and associated soils.
II. Chemical composition of laterite profiles. J. Indian Soc.
Soil Sci., vol. 10, no. 3, p. 211-22.

75

Development of certain soils in the subtropical

humid zone in south-easternparts of India, genesis
and classification of soils of Machkund Basin'
S. V. Govinda Rajan and N. R. Datta Biswas
All India Soil and Land Use Survey,
Indian Agricultural Research Institute, New Delhi (India)

INTRODUCTION

The catchment of the Machkund River above the

Jalaput d a m covers an area of 2,211 km2 falling in the
districts of Koraput and Vishakapatnam, respectively
in the States of Orissa and Andhra Pradesh. Of this
area 729 km2falls in Orissa State and the rest in Andhra
Pradesh. This area forms a part of the Eastern Ghats
where elevation ranges between 610 and 1,680 m
above m e a n sea level with highly undulating and rolling
topography, interspersed with valleys. Soil surveys
were carried out in these areas to obtain basic soils
information which could be used for soil-conservation
planning in this river valley catchment.
The surveys were carried out by the Regional Centre,
Calcutta, of the All India Soil and Land Use Survey,
w h e n a large number of soil profiles were examined in
the field for their morphological featuring and a number
of taxonomic units were recognized, based o n their
differentiating morphological features. It is found that
variations in parent material and topography with
which is associated the moisture regime and microclimate of different land features have had great impact
on soil genesis. These are reflected in the differential
physico-chemical properties significant in taxonomic
groupings of the soils. This paper aims to characterize
these soils b y placing them into different groupings and
also locate them in the international system of classification.

PHYSIOGRAPHY, RELIEF
A N D DRAINAGE
T h e catchment area of the Machkund River is covered
over by numerous rivulets and gullies starting from the
hill ranges and' which feed the river. Steep slopes,

heavy rainfall within a short duration, erosive nature

of the soil, faulty method of cultivation, cutting d o w n
of forest vegetation have all combined together to cause
severe soil erosion of the area. The hills in the area are
the remnants of the hard erosional surface. There are
considerable evidences to s h o w that the area is a
rejuvenated polycyclic peneplain that has attained
more than one base level of erosion. The 'hills are
discontinuous and numerous and this region is intersected b y the basins of several stream and water
courses. The drainage pattern is mostly dentritic in the
valley and mesas of low gradients, while it is mostly
radial and subdentritic near the monadnocks of sharp
gradient. The rivers of the terrain, as obvious, are of
antiquity but due to resurrected rejuvenation of the
topography and consequently by the ceaseless degradation by the agents of erosion, the major channels have
been at grade-attaining profiles of equilibrium.
GEOLOGY

T h e old geological formations consist of a series of

metamorphosed sediments of the Archaean system
which have been intruded by granites, charnockites and
dolerites. The metamorphism of these ancient sediments
have been accounted for by the moutain-buildingforces.
Different types of metamorphism have m a d e the texture
of these rocks either schistose or gneissose. Khondalites
are the oldest rocks in this area and are highly metamorphosed Pre-Cambriansediments consisting of quartz,
garnet and sillimanite, with lesser amounts of feldspar,
graphites, manganese and iron ores. The charnockites
are younger than khondalites and represent the plutonic igneous rock series which have undergone recrystal-

1. Specially presented as supporting paper to review No. 2 by P. Segalen.

77

S. V. Govinda Rajan and N. R. natta niswas

1. Cold and comparatively ?ry winter-December,

lization in the solid state on being subjected to plutonic

metamorphism. They are characterized b y the presence
of hypersthene, blue-grey quartz and feldspar in acid
and intermediate rocks and pale-green augite, bluegreen and green-brown hornblende in the more basic
rocks. T h e charnockites and hornblende schists generally
occur in the area as intrusives in khondalites..
In places, presence of charnockites as prominent hill
mass arc observed due to the upliftment and superimposition in Pre-Archaean times which m a y be to some
extent explained by the general freshness of their
appearance and comparatively light weathering.
Massive granite-gneiss form occasional outcrops at
places and also occur in bands in the hilly region.
Quartzites occur in m a n y places along with the hornblende schists. Laterite mass is found to cap some of
the peaks of the hills built of khondalites. These laterites are mostly ferruginous and occasionally rich in
alumina. Blown sand, newer alluvium, and water-borne
deposits of red and lateritic soils are the most recent
deposits and occur in thin layers in the valley areas.

January, February.

2. H o t and dry summer-March, April and May.

3. Monsoonal w a r m and wet period-June, July,
August and September.
4. Cold autumn-October ,and November,

The m e a n monthly rainfall and humidity (Health

Office, Machkund), averaged over ten years (1946-57)
and the m e a n m a x i m u m and m i n i m u m temperatures
for this region are given in Table 1. The distribution of
rainfall and temperature are given in Tables 2 and 3.

EXPERIMENTAL
Soil samples of six soil series, namely, Kallupada,
Pippoluputtu, Kangrapada, Devulupada, K o n a and
Hollangapada series identified in the area were collected
for evaluation of chemical and physico-chemical properties. A landscape of this area and relative disposition
of the different soils are shown in Figures 1 and 2. A
s u m m a r y of the important morphological features is
presented in Table 4.
Mechanical analysis was done b y the international
pipette method; organic carbon by Walkley and Blacks
rapid titration method as modified by Walkley (1935),
nitrogen by the Kjeldahl method (AOAC, 1950),
p H by B e c k m a n pH-meter in 1 :2.5 soil-watersuspension (Piper, 1950); cation exchange capacity b y leaching
with a m m o n i u m acetate (Schollenbergerand Dreibelbis,

CLIMATE

T h e climate of the area is subtropical monsoon type

characterized b y hot dry s u m m e r followed by a period
of heavy precipitation and dry cold winter.
T h e following yearly seasonal sequence is indicated
from climate data collected from the area of the
M a c h k u n d B,asin:

TABLE1. The mean monthly rainfall, humidity, and maximum and minimum temperatures of the Machkund Basin

Feb.

Mar.

Apr.

May

Jun.

Jul.

Aug.

Sept.

Oct.

Nov.

Dec.

2.8
72
32.2
16.6

9.9
60
36.4
15.4

45.2
61
38.8
19.6

53.3
61
39.9
22.6

208.5
82
37.5
22.4

482.8
89
30.3
20.2

626.0
92
29.8
20.5

411.0
92
29.2
21.0

155.6
83
30.8
18.3

7.6
83
29.4
13.2

3.5
82
28.0
7.8

Jan.

Rainfall (mm)
Humidity (yo)
Max. temperature (OC)
Min. temperature (OC)

13.7
81
28.3
9.8

TABLE2. The annual distribution of rainfall (in mm) in the

TABLE3. The annual distribution of temperature (in OC) in

Machkund Basin

Mean

December
to
February

Summer:
March
to
May

Monsoon:
June
to
September

Autumn:
October
to
November

2018

20.3

108.7

1724.9

16.3

78

Winter:

the Machkund Basin

Mean

Max.
Min.

32.2
16.6

Winter:
December
to
February

Summer:
March
to
May

Mnnannn:
June
to
September

Autumn:
Octobrr
to
November

28.31
8.25

38.28
19.17

31.61
21.11

30.17
15,67

Development of certain soils in south-easternparts of India

Kallupada
Pippoluputtu
Kangrapada

Hollangapada

Kona

FIG.1. A

landscape in M a c h k u n d catchment showing the relative positions of the different soils.

Kona

Devulupada

1
FIG.2.

Disposition of the soils in a valley situation-Machkund

Kona

catchment (Orissa).

79

S. V. Govinda Rajan and N. I. Datta Biewas

1930);total exchangeable bases by Piper method (1950)

and base saturation by calculation. Ferric oxide, aluminium oxide, sesquioxides, PO and C a 0 in hydro%
chloric acid extracts were determined by standard
chemical methods.

RESULTS A N D D I S C U S S I O N
The

area under study experiences an annual precipitation of 200 cm. E v e n though the rainfall is concentrated within the four monsoon months, high humidity
is maintained throughout the year. The difference
between the m e a n annual m a x i m u m and m i n i m u m
temperatures is wide, being of the extent of 16oC.
It may, however, be mentioned that the area experiences dry spells as well during the year. The diverse
components of climate in operation along with other
soil-forming factors have been responsible for development of different kinds of soils in the area. O n e of the
important factors has been topography. T h e landscape
features are characterized by steeply sloping hills,
convex hill tops and nearly level or moderately sloping
valleys. These topographical situations have created
conditions of micro-climatic variation within the area
resulting in hill tops and slopes suffering moisture
deficiency or aridity of local climate and the concave
land areas receiving excess moisture. The orogenic
influence has resulted in the development of kinds of
S'oils on convex land surfaces which are quite distinct
from those in other landscape situations. F r o m the
tabular statement it will be seen that the soils on the
hill tops and top slope OP the hill sides, namely, soils of
Kallupada series are shallow, occasionally attaining
moderate depth. As the slope eases up to the foothill
areas, the soils attain greater depth. The soils of the
valley and depressions are very deep. The data of
mechanical and chemical composition of the soils presented in Tables 5 and 6 (pages 83 and 84) show that
there has been an increase in the content of clay in the
soils from hill top to the valley. A similar increase is
also noticed in case of dibasic constituents of the soils.
F r o m the foregoing discussion it would be apparent
that there has been lateral translocation of finer materials as also of soluble constituents on account of
lateral movement of water, which flows d o w n as run
off. The above situation is suggestive of lack of deep
percolation through depth in hill tops and slopes giving
rise to shallow soil. Further accelerated erosion due to
slope conditions depletes the soil materials from convex
and steep-slope situations giving rise to shallow soils,
Deep percolation of water in other areas accentuating
weathering and receipt of soil material from the upland
and through the drainage courses have given rise to
deep soils in valleys and depressions.
T h e cation exchange capacity of the soils also varies,
being lower in respect of soils of the hills and hill

80

slopes and higher in respect of the soils in the low-lying

situation. The high content of dibasic constituents in
the low-lying soils coupled wiLh high moisture rgime
appears to have helped resilicification of the secondary
clay minerals enriching the soils in contents of 2 :1
lattice minerals of high exchange capacity. The derived
values of cation exchafige capacity of the clays definitely show a transition from hills to low-lying areas.
The heavy precipitation received in the area creates
conditions of accelerated leaching (lateral and/or
internal)thereby depleting the soils of basic constituents
and making them acidic with p H values around 5.5.
The pedologic processes have been operative in the
development of profile characters in the soils of the area,
the profiles having assumed illuviated zones with accumulation of finer mechanical constituents. The cation
exchange capacity of the soils is moderate to high
showing that the process of kaolinization has not proceeded too far and under the acidic environment there
has not been rupture of the secondary minerals releasing
sesquioxidic constituents. The base saturation is high
being more than 50 per cent in all cases except one,
namely the surface layer of the Kallupada series. The
colours as represented by Munse11 colour notations in
respect of first four soils which occur on the hills, the
hill slopes and the foothills vary the hue ranging from
7.5 YR to 2.5 YR. In most cases the hue is 5 YR or
less, the colour approaching spectral red particularly
in the textural B horizon. The above characteristics are
suggestive of the grouping of the hili soils as Entrophic
reddish-brown lateritic according to the Food and
Agriculture Organization system of classification or
more broadly as Latosol with high base saturation
(Plinthic, 1949) as recently applied for tropical soils of
Brazil. The soils of the valleys and depression represented b y last two series have features associated with
wetness (colours and/or mottlings indicating saturation
with water at some period of the year) and are classed
as L o w H u m i c Gley Soils.

P O T E N T I A L F O R L A N D USE

While considering the use of the different kinds of soils

for crop production it will be apparent that limitation
of depth of the hill soils creates a condition of very low
available water capacity keeping the soils droughty.
In spite of moderate cation exchange capacity of these .
soils, the above limitations in regard to depth and
available water capacity render these soils to be of low
to moderate productive value and are consequently of
limited suitability for intensive agricultural use. T h e
soils of the valleys and depressions are free from the
above limitations but the condition of wetness render
them suitable for paddy cultivation. With effective
drainage alternative crops can also be grown. Further,
for efficient fertilizer use, these soils would require
application of amendments to mitigate the acidity.

Development of certain soils in south-easternparts of India

TABLE4. A summary of the important morphological features of the different soils identified in the Machkund catchment area'

No.

Soil

series

Physiographic
position

Depth
(em)

Kallupada H
ill tops and 0-10
top slope of
hill sides

~~l~~~

Texture

.
,

Reac-

tion

Root dis-

Other features includine:

tribution mottling and/or concretion

Reddish
bi&n
2.5 YR 4/4

Loamy sand Structureless,Acid

loose and
to sandy
loam
friable

C o m m o n Ferruginous concretions c o m m o n with

gravels and pebbles
or weathered rocks

10-35

Reddish
yellow
7.5 YR 6/8

Clay loam

Weakly
hlocky

Acid

Many

35-45

Yellowish
red
5 YR 5/8

Clay loam

Blocky

Acid

M a n y ferruginous
concretions, lateritic
pieces with fragments
of weathered rocks

45 +
2

Pippoluputtu

Convex
sloping
hill aides

Laterite

0-12.5

Dark
reddish
brown

5 YR 3/3
12.5-40 Yellowish
red
5 YR 4/8

40-75

Loamy sand Sthctureless,Acid

to sandy
loose and
loam;
friable
occasionally
loam

Plentifu1

Ferruginous concretions common, gravels, stones of various

sizes

Clay loam
(gravelly)

Many

Ferruginous concretions and gravel

common

Weakly
blocky

Acid .

Matrix of ferruginous
clays, gravel and laterite pieces; patches of
black mottling

Darkred

10 R 3/6

75+

Kangrapada

Many ferruginous
concretions with rock
fragments

Consolidated hard
ferrallitic mass, reddish and hard

Foothillareas 0-15
and sloping
uplands

Yellowish
red

5 YR 4/6

Sandy loam Weakly

to sand .
granular
clay loam

Acid

C o m m o n F e w ferruginous concretions and partially

weathered fragment
of rocks

15-42

Red
10 R 4/6

Clay ioam

Weakly
blocky

Acid

Many

M a n y unweathered
and partiallyweathered rock fragments and
m a n y ferruginous
concretions

42-90

Dark red
10 R 3/6

Clay loam
gravelly

Strongly
blocky

Acid

Few

M a n y unweathered
and partially weathered rock fragments and
many ferruginous
concretions

90+

'

Duskyred
10 R 313

Matrix of detrital
laterite. ferruginous
clay and unweathered
pieces of rock fragments

81

S. V. Govinda Rajan and N. R. Datta Biewas

TABLE4 (continued)
No.

82

S?ii

nenes

Phyriographic
position

Depth

Hollan- Sloping up- 0-15

gapada lands running
with convex
slope of gentle
gradient

Devulupada

Kona

Texture

(4

Loam to
clay loam

Root dis-

Other featurei including

tribution mottling and/or concretioni

Weakly
granular

Acid

Abundant

15-57.5 Dark reddish Clay loam

brown
2.5 YR 314

Weakly
blocky

Acid

Many

57.5-150 Dark red

2.5 YR 3/6

Blocky

Slightly
acid

Nearly level 0-12.5

to level and
terraced
lowland

Yellowish
red
5 YR 4/6

&y

Clay loam
(gravelly)

Dark yellow- Clay loam

ish brown
10 YR 414
Yellowish
brown
10 YR 516

C o m m o n pea-sized
hard ferruginous concretions and quartz
gravels

Sub-angular Acid
blocky

Plentiful F e w mustard-sized
ferruginous concretions

Many

12.5-45 Yellowish
brown
10 YR 818

Clay loam

Angular
blocky

Acid

45-85

Yellowish
brown
10 Y R 518

Clay loam

Strong
angular
blocky

Acid

Many grey mottliiigs

and few yellowish-red
spots;
Many soft black and
brownish ferruginous
concretions

85-150

Dark grey
7.5 YR 4/0

Clay

Massive

Acid

Many pea-sized soft

ferruginous concretions of strong brown
and black colour;reddish brown and very
dark grey-brown
mottlings

Pale brown
10 YR 6/3

Clay loam

Sub-angular Mildly
acid
blocky

Plentiful

12.5-325 Greyish
brown
10 YR 5/2

Clay loam

Angular
blocky

Mildly
acid

Many

32.5-105 Dark grey

10 YR 4/1

Clay

Strongly
blocky

Mildly
acid

Depressions, 0-12.5
level terraced
lowlands,
reclaimed
gullies and
stream beds

Many hard ferruginous concretions

Dark yellowish-brown
mottlings; many
mustard-sized quartz
gravel present

Few pea-sized, black,

soft ferruginous concretions

Development of certain soils in south eastern parts of India

TABLE5. Physical and physico-chemicalproperties of horizon soil samples of the profiles of the recognized soil series in the
Machkund catchment
Mechanical constituenti
(expressed as percentage
on air-dry basin)

Soil series

No.

and depth (em)

Pbysico-chemicalproperties

B.E.C.
Total

Sand

Silt

Clay

T.E.E. .

c.e.c.
m e . 100 g

me.
100 g

Base
saturation

PII

n e . yo
calculated
on clay

(%)

Kallupada
0-10
10-35
35-45

71.4
54.3
45.8

9.0
18.5
21.3

19.9
27.1
32.8

6.7
8.8
9.2

3.2
4.8
4.8

47.5
54.5
52.1

5.4
5.6
5.6

33.8
32.4
28.1

Pippoluputtu
0-12.5
12.5-40
40-75

73.9
61.6
48.4

10.3
13.7
14.2

15.8
24.7
37.4

7.2
8.4
10.8

3.8
4.0
5.5

52.7
50.9
51.1

5.4
5.5
5.7

45.4
34.0
29.0

Kangrapada
8-15
15-42
42-90

55.8
49.9
44.5

21.4
22.6
22.9

22.7
27.5
32.6

6.8
8.9
10.6

3.5
4.5
5.5

51.8
50.9
51.7

5.6
5.6
5.7

30.0
32.5
32.7

IIollangapada
0-15
15-57.5
57.5-150

56.7
58.7
58.9

21.1
17.2
15.6

22.1
24.0
25.4

8.4
9.3
10.8

4.3
5.1
5.7

51.4
54.4
52.7

5.4
5.4
5.6

38.0
38.7
42.4

Devulupada
0-12.5
12.5-45
45-85
85-150

53.9
49.8
43.9
35.6

19.5
20.8
21.8
23.6

26.6
29.4
34.2
40.7

8.8
13.2
22.9
26.3

5.2
7.9
13.3
15.7

59.0
60.1
58.1
59.9

5.4
5.4
5.6
5.6

33.1
44.9
66.9
64.3

57.5
48.1
41.1
28.7

14.8
18.0
19.4
22.8

27.7
33.9
39.4
48.5

18.9
22.3
25.2
26.6

11.6
13.6
16.1
15.3

61.8
60.9
63.9
50.0

5.5
. 5.6
5.6
5.7

Kona

0-12.5
12.5-32.5
32.5-105
1O 5 15O

TABLE6

'
-

68.2
65.9
63.9
54.7

overleafl

83
,

S. V. Govinda Rajan and N. H.Datta Uiswas

TABLE6. Chemical composition of the horizon soil samples of the soil series
Soil series and

No.

horizon depth
(em)

pz.0

FezOa

0.07
0.06
0.03

0.06
0.05
0.04

Pippoluputta
0-12.5
12.5-40
40-75

0.07
0.05
0.04

Kangrapada
8-15
15-42
42-90

Alzo3

Ca0

Organic carbon

9.94
9.58
10.56

17.99
22.55
25.18

0.16
0.19

0.74
0.68
0.27

0.06
0.06
0.04

9.20
10.16
10.24

19.28
22.87
25.47

0.16
0.17
0.18

0.66
0.36
0.25

0.05
0.05
0.04

0.07
0.05
0.04

9.52
14.16
15.80

0.17
0.18
0.18

0.58
0.53
0.41

IIollangapada
0-15
15-57.5
57.5-150

0.08
0.07
0.05

0.06

'

0.05
0.04

7.90
8.30
8.80

17.09
18.74
18.80

0.17
0.18
0.19

0.87
0.60
0.37

Devulupada
0-12.5
12.5-45
45-85
85-150

0.10
0.06
0.05
0.04

0.08
0.05
0.04
0.03

8.40
8.24
, 9.88
10.00

16.97
16.33
'.:17.31
18.59

0.20
0.22
0.25
0.36

1.12
0.76
0.55
0.30

0.11
0.09
0.06
0.05

7.54
7.76
8.20
9.20

16.02
19.01
19.82
23.95

0.36
0.38
0.38
0.39

1.06
0.62
0.58
0.3.5

0.18

Kallupada

0-10
10-35
35-45

Chemical composition (percentageon oven-dry basis)

17.78
19.52.
22.06

Kona

0-12.5
12.5-32.5
32.5-105
105-150

'

0.11
0.05
0.04
0.04

Bibliography
AOAC 1950.'

Oficial and tentative meihods of analysis,

p. 30. Washington, Association of Official Agricultural

Chemists.
GOVINDARAJAN,S. V.; DATTABISWAS, N. R. 1968. Characterisation of certain soils in the sub-tropical humid zone
in the South-Eastern part of India-Soils of Machkund
Basin. J. Znd. Soc. Soil Sci., vol. 16, no. 2, p. 179-86.
KELLOGG, C. E. 1949. Preliminary suggestions for the classification and nomenclature of Great Soil Groups in tropical

84

and equatorial regions. In: Commonw. Bur. Soil Sci. Tech.

Commun., no. 46, p. 76. IIarpenden (Herts.).
PIPER,C. S. 1950. Soil andplant analysis,p. 9,189. N e w York,
Academic Press.
SCHOLLENBERGER, C. J.; DREIBELBIS,
F. R. 1930. Analytical
methods in base exchange investigations on soils. Soil Sci.,
no. 30, p.'161.
WALKEY, A. J. 1935. A n examination of methods for determining organic carbon and nitrogen in soils. J. Agric. Sci.,
no. 25, p. 598.

Pedogenesis of soine major soil groups

in Mysore State, India,
T. Seshagiri Rao
University of Agricultura1 Sciences,
Regional Research Station, Raichur, Mysore State (India)

INTRODUCTION
T h e three major soil groups which arc found in Mysore
State are called locally : (a) red soils; (b) black cotton
soils; and (c) lateritic soils (Latosols), These occupy
extensive areas in the State. In addition there are
a large number of intergrades and sub-groups of the
above major groups whose distribution is not extensive.
T h e physiographic and climatic features under which the
above kinds of soils occur differ markecliy. T h e object
of this paper is to present some important pedogenic
characteristics of these soils.

BLACK COTTON SOILS

These soils are also called Regur by some workers.
T h e physical and chemical properties of these soils
would classify them as belonging to the Vertisols in the
N e w Soil Classification proposed by USDA. T h e y occupy
a n extensive area in the northern part of Mysore State
which is characterized by semi-arid and arid climatic
conditions. Except in the months of August, September and October the evapotranspiration exceeds
precipitation,resulting in a water deficit which is severe
from March to June.
T h e black cotton soils are derived from diverse kinds
of parent materials but the most c o m m o n feature in
all the different situations is the similarity in climate.
Thus in the region north of the River Krishna the soils
are derived from volcanic lava materials thousands of
feet in thickness with almost horizontal layers. This
geological formation is k n o w n as the Deccan Trap o n
account of the step-like aspect of the weathered hills
of the basalt formed from the erupted lava. T h e Deccan
Trap is augite-basalt and basic in character containing
feldspar, augite and magnesite as the main minerals
and does not sliow m u c h differentiation except in colour

and texture. T h e usual colour is a greyish-green tint

but even black- and red-coloured types m a y be found.
On weathering these rocks have given rise to black
soils.
In the region south of the River Krishna and in the
Tungabhadra River basin the parent material is c o m posed of the oldest Archaen complex of granites and
gneisses, which, however, vary in chemical and mineralogical composition, texture, structure and colour. T h e
granites and gneisses consist mainly of types containing
(a) quartz and feldspar; (b) quartz, feldspar and hornblende; and (c) quartz, feldspar and mica. T h e feldspars
vary in composition being of both types, i.e., soda-lime
and potash feldspars. It is also a feature to find blackcoloured soils occurring in the valleys and lightertextured red soils in the terraces and uplands. A
typical black cotton soil has the following features.

O to 30 cm, black (10 Y R 2/1 dry), silty clay with few

lime concretions, strong, fine, sub-angular structure,
friable, sticky and plastic when wet, slight effervescence
with dilute IICl, gradual boundary.
30 to 58 cm, very dark grey (10 YR 3/1 dry), silty clay
with lime concretions, strong, fine, blocky structure,
firm, sticky and plastic when wet, slight eflervescence
with dilute ICI,clear and smooth boundary.
58 to 92 cm, very dark grey (10 YR 3/1 dry), silty clay
with abundant lime concretions of medium size,massive,
very firm, very sticky and plastic when wet, strong
effervescence with dilute IICl, gradual and smooth
boundary.
92 to 108 cm, very dark grey (10 YR 3/1 dry), clay
with abundant big-sized lime concretions, massive, hard,
sticky when wet, vigorous effervescence with dilute IICl.

In m a n y situations gypsum is also present beneath the

calcic horizon.
1. Specially presented as supporting paper to review No. 2 by P. Segalen.

85

T. Seshagiri Rao

The colour of these soils is due principally to organic

matter content but the amount of the latter itself is
low, less than 1 per cent in m a n y of these soils. T h e
black soils show considerable variations in depth and
clay content, depending on their position. They do not
ordinarily show marked soil horizons, but the subsoil
has a slightly higher percentage of clay than the surface
soil. The clay belongs to the montmorillonoid group of
minerals. The soils swell and shrink enormously with
the change in moisture content. T h e infiltration capacity
and permeability of these soils are very low and hence
they are susceptible to erosion. Even on a very small
slope of 1 per cent the losses of water and soil due to
erosion are considerable. When the soils are not irrigated
they are under dry-land farming whose main feature
is to m a k e the best of available moisture by adopting
proper soil and water conservation practices.

RED SOILS
This group

includes soils of texturally different classes

such as sands to clay loams. T h e c o m m o n feature of
this group is the red colour which is due to the presence
of ferric oxide. These soils occupy an extensive area
in the southern region of Mysore State. The rainfall in
this area is between 650 and 1,000 m m per annum.
T h e soil depth varies from 90 to 150 c m for loams to
only about 45 to 60 c m for sandy soils. They are also
derived from diverse parent materials such as granites,
gneisses and other rock materials neutral or acidic in
character. Profile descriptions are given as follows :

LATERITIC SOILS (LATOSOLS)

In the New Soil Classification proposed by USDA
(7th approximation) the Latosols belong to the order
Oxisols. These soils occur in the western part of the
state where the land form is hilly and mountainous.
The rainfall in this area is well over 200 c m and at
places m a y be as high as 750 cm. They are derived from
diverse parent materials. The high temperatures prevailing in the region have resulted in increased weathering
and the high rainfall is responsible for intense leaching
of bases. The profile of a lateritic soil has the following
features :

A
B

Profile 1

O to 12 cm, reddish-brown(5 YR 5/4 dry), sandy loam,

weak, fine, granular structure, loose and friable, nonsticky and non-plastic,clear and smooth boundary.
12 to 35 cm,yellowish-red (5 YR 4/6 dry), sandy loam,
moderate, medium, sub-angular blocky structure, friable, slightly sticky, gradual smooth boundary.
35 to 150 cm, red (2.5 YR 4/6 dry), sandy clay loam,
moderate, coarse, sub-angular hlocky structure, hard,
slightly sticky.

Profila 2
O to 40 em, reddish-brown(5 YR 4/8), loam, weak, fine,
granular structure, friable, non-sticky and non-plastic,
graduai boundary.
40 to 200 em, dark reddish-brown (5 Y R 3/6), loam,
moderate, coarse, sub-angular blocky structure, hard,
dark iron and manganese concretions, clear boundary.
200 c m +.dark-brown gravelly material mixed with
quartz pebbles.
These soils are sandy to loam in texture, containing
about 10 to,20 per cent clay. The internal drainage is
good and it is easy to cultivate these soils. They are
not potentially productive Lut respond well under
proper management with assured inputs such as water

86

and fertilizers. T h e physiographic features of the areas

occupied by these soils are characterized by gentle to
steep slopes. It is not unusual to find deep and widc
gullies with vertical soil columns standing for several
years. The soil surface becomes very hard when dry
probably due to the irreversible dehydration of colloidal
iron. The undulating nature of the terrain coupled with
hard nature of the surface m a k e these soils highly
susceptible to surface erosion even with a favourablc
internal drainage.

O to 45 cm, reddish-brown(5 YR 4/3), silty clay, strong,

fine, granular structure,friable,iron concretions,smooth
boundary.
45 to 120 cm, reddish-brown (5 YR 4/3), clay loam,
strong,medium, sub-angularblocky structure,firm, nonplastic and non-sticky, bigger iron concretions, diffuse
boundary.
120 cm +.reddish-brown (5 YR 4/4), clay-loam,reticulately mottled, strong, medium, sub-angular blocky
structure, firm, slightly sticky, rounded red and black
iron concretions, plinthite.

has all the properties of an oxic horizon containing a mixture of hydrated oxides of iron and aluminium, 1: 1 lattice clays and insoluble quartz sand.
The c.e.c. also is very low. Plinthite is presen) in the
soil profile. It io usually soft when not exposed but
changes irreversibly to ironstone hardpans or irregular
aggregates on exposure to repeated wetting and drying.
The dried bricks (L. Laterite = brick) are used as
building materials.
D u e to intense leaching of bases these soils are acidic
in reaction. The agriculture is mainly dependent on
rainfall. They are deficient in phosphorus and potash.
The amount of organic matter and nitrogen is adequate
but the latter needs to be mineralized for making it
available. The soils need liming. Lateritic soils support
luxuriant forests.
The soils have good physical conditions for plant
growth. They are less susceptible to erosion than m a n y
other soils with the same slope. They are also easy to
work with because of their friable nature.

Pedogenesis of some major fioil groups in Mysore State, India

Bibliography
DREW,J. V. (ed.). 1967. Selected papers in soil formation and

UNITED
STATES SOILSURVEYSTAFF.1967. Supplement to soil

classification. Lincoln, Nebraska, Soil Sci. Soc. Amer.

Special Publication Series, p. 428.
KANITKAR,
N.V. 1960. Dry farming in India, N e w Delhi,
ICAR,p. 470.
SMrnr, Guy, D. 1965. Pedology lectures on soil classification.
Bull. Soc. belge Pdol., special no. 4, p. 134.

ClassiJicationsystem, 7th approximation,p. 207. Soil Conser-

vation Service, U.S. Dept. of Agriculture.

VENKATAh o , B.V. 1968. Soil resources in Mysore. Raichur,
Uniu. of Agric. Sci. Extension Series, no. 2, p. 24.
VENKATARAMIAH,
P. 1934. Report on the soil survey of the
Tungabhadra Project, p. 106. Madras, Government Press.

87

Correlation between chemical bond

and chemical weatherabilitv of the c o m m o n
.
.
rock-Iorming minerals
1

H. Ling Ong
Bagian Geologi, Institut Teknologi Bandung
Bandung (Indonesia)

INTRODUCTION

Chemical weathering of minerals and rocks is one of

the most intriguing problems encountered by the
geologist working in a tropical climate. Outcrops are
rare and sometimes it is impossible to d o conventional
geology in this type of environment. T h e study of
rock weathering, both theoretical and experimental,
should therefore be directed towards a n elucidation
of the forces responsible for weathering processes.
T h e accepted theory o n chemical weathering m e n tioned in most geology textbooks, is the one offered
by Goldich (1938). H e pointed out that the order of
stability of minerals of igneous rocks toward weathering is the reverse of their order in the reaction series of
B o w e n (see Fig. 1).
This stability series conforms with field observations ;
quartz is found to be the most resistant a m o n g the
igneous minerals. Whereas olivine is the least resistant
of the series.
Coldichs principle of mineral stability m a y be
explained in the following way. T h e minerals in igneous
rocks were at equilibrium under the conditions of temperature and pressure at which the rocks are formed.
Having been formed at high temperature, and sometimes at high pressure as well, they cannot be expected
to remain stable under the very different conditions at
the earths surface. T h e higher the temperature of formation, the less stable the mineral will be at surface
environments. In a general w a y this seems reasonable
enough although it is not a necessary conclusion, theoretically (Krauskof, 1967, p. 105).
This observation led to a survey of the literature for
the temperatures of formation of the c o m m o n rockforming minerals.
T h e formation temperature is defined as the temperature at which the minerals are formed. Since the work
of coldich in 1938, severai investigations have found

that polymorphism is c o m m o n a m o n g minerals. In

connexion with this, t w o possibilities exist in choosing
the formation temperature. First, the formation temperature is the temperature at which the mineral starts
to crystallize from its melt and is called the melting
temperature. Second, the formation temperature is thc
temperature at which the mineral changes from one
polymorph to another and is called the transformation
temperature. Both temperatures for Goldichs series
are assembled in Figure 1. T h e transformation temperature at one atmosphere is shown as a denominator
(data from Buerfer, 1948; Tuttle and Brown, 1950;
Mason, 1959) and the melting temperature at one
atmosphere is shown as a numerator (Roy, 1949, and
Morey, 1964).
If the transformation temperature is taken as the
formation temperature of the mineral, and if Goldichs
principle holds, then quartz, having a transformation
temperature of 5730 C, should be at least as stable as
the calcic-plagioclase and the alkali-calcic plagioclase,
both plagioclases having transformation temperatures
of 5750 C and 5250 C respectively. On the other hand, if
the melting temperature is taken as the formation temperature of the mineral, then quartz having a melting temperature of 1,713OC, should be less stable than the
plagioclase series, which have melting temperatures
ranging from 1,1180 C for albite to 1,5530 C for anorthite. In both cases, whether the temperature of formation is taken as the transformation temperature or is
taken as the melting temperature, the Goldichs series
cannot be confirmed.
T h e aim of this article is to find a theoretical basis for
predicting the chemical weatherability of the c o m m o n
rock-forming minerals and to compare the predicted
values with those observe! in the field as reported in

1. Speeidy presented

8s

supporting paper to review

No. 2 by P. Segalen.

89

II. Ling Ong

1,7130 C
Quartz

Stability
decreases

5730 c

!'

940"-9800 C

Muscovite

1,2350 C

K-feldspar

1,1180 C
Alkalic plagioclase

J.

7000-190000

1,0000-1,1000c

Biotite

700 C
1,2000 c
Alkali-calcicplagioclase

___

5750 c

Hornblende
I

1,3300 C
Calcic-alkalicplagioclase

___

Augite

.5250 C
1,5520 C

literature. T h e present work is limited to the weathering

of c o m m o n rock-forming minerals, mainly because of
the availability of the observed weathering data for
these minerals.

T H E O R Y O F C IIE M I C A L .WEATH E RING

T h e decomposition of the rock, while still in massive

form or a loose material, is k n o w n as chemical weathering. Jt produces changes in the nature and composition
of the body, as distinguished from physical weathering
which produces a change only in the form of the body.
T h e five principal agents involved in chemical
weathering are : oxidation, hydration, carbonation,
solution, and deposition (see for discussion, Joffe, 1949,
p. 81-6).In all cases, the process of weathering involves
the use of water molecules. For example, oxidation of
ferrous to ferric iron takes place most readily in a
humid climate rather than in a n arid climate, although
in both climates the amount of oxygen present in the
atmosphere is the same. T h e processes of hydration,
carbonation, solution and deposition cannot occur
without the presence of waLer. T h e ullimate fate of
tliTerent minerals thus .depnds largely o n their relative stability in water. Chemical weathering can Le
considered as a process of interaction between minerals

90

1,2050-1,8900C

and water molecules. T h e extensive chemical weathering

in humid regions verify this statement.
Water being a polar molecule will be attracted also
by another polar molecule. Molecules having ionic or
polar bonds provide a favourable place for attracting
water. T h e mineral surfaces of a n ionic crystal are a loci
of reaction with water molecules. However, this is not
the case with molecules having covalent or non-polar
bonds :these molecules have no w a y of attracting polar
water molecules due to their equal distribution of
charges. Therefore, if the amount of ionic bonding in a
mineral is known, the magnitude of the reaction involving water molecules can be predicted.
Ionic bonding between atoms A and B.occurs w h e n
atom A has a tendency to lose a n electron whereas
atom B has a tendency to accept the electron. T h e greater the tendency for a t o m A to lose an electron and/or
the greater the tendency for atom B to accept a n electron, the greater the ionic binding will occur between
atoms A and Byand consequently, the smaller the contribution of the covalent bond. T h e concept of losing
and accepting electrons for atoms can be expressed in
quantitative terms. This is k n o w n as the electronegativity and can be described as tlie electron-attracting
powcr of an atom in a molecule. T h u s a n atom that has
a tendency to lose a n electron should have a low value

Correlation between chemical bond and chemical weatherability

of electronegativity, whereas an atom that has a

tendency to gain an electron has a high value of electronegativity. T h e greater the difference in electronegativity between two atoms, the greater contributions of
ionic bonding will be.
If we k n o w the electronegativity values for each
atom on the periodic table, then the amount of ionic
bonding between any two atoms can be calculated.
Since the amount of ionic bonding is related to the
amount of water molecules being attracted to the surface of the minerals, the amount of ionic bonding in
minerals is related to the degree of chemical weathering.
In other words, minerals arranged according to their
percentage of ionic bonding, also shows the same
arrangement when they are subjected to weathering
at the earths surface.

CHEMIC A L W E ATH E R IN G O F .THE

C O M b.2O N ROC K F O R M IN G MIN E R A L S

In 1932, Pauling derived a useful scale of electronegativities. H e related the amount of ionic character in a
molecule A B to electronegativity difference (X,-X,)
based on an empirical graph. Several investigators have
modified his formula, the most successful of these is
that of IIannay and Smyth (1946):

yo ionic bonding = 16 (XA-XB)-3.5 (XA-xB)2.

Another approach to the problem of electronegativity
is given by Povarenykh in 1956, and discussed by
Smith (1963). T o calculate the percentage ionic bonding, Povarenykh draws a purely empirical curve
between zero ionic bonding at infinitely large differences.
Both approaches to the problem of calculating the
amount of ionic bonding can only be applied to two
atoms at a time and, therefore, cannot be used for calculating complex compounds commonly found in minerals. However, by comparing the amount of ionic
bonding between two pairs at a time, the percentage
of ionic bonding in minerals can be compared. T h e most
appropriate pair to choose for the silicate minerals
would be the oxide pairs. The percentage of ionic
bonding calculated by the two methods for the c o m m o n rock-forming oxide pairs are shown in Table 1.
A high percentage of ionic bonding means a low percentage of covalent bonding. Thus K-Owhich is 69 per
cent ionic according to the method of H a n n a y and
Smyth is 31 per cent covalent.
The amount of ionic bonding calculated using either
IIannays and Smyths equation or Povarenykli empirical graph for the geologically important oxide pairs
(Table 1) shows a general decrease from K-O to Si-O.
T h e only difference is for iron. Whereas Haririays
and Smyths values for iron (II) and (III) are similar
to the value of Si-O (37), the values of Povarenykh

TABLE1. Percentage of ionic bonding for the common rockforming oxide pairs

Oside pair

K-O
Na-O
Ca-O

Mg-o
Fe(I1)-O .
Fe(111)-O
Al-O

si-o

Percentage ionic bonding

(calculated by using
Paulings scale (1932)
and Hannays and
Smyths equation (1946).

Percentage ionic bonding

(calculated by using
Povarenykh empirical
graph, an seen in Smyth,
1963, Table 2-1)

69
65
62
55
37
35
46
37

86.6
83.2
79.4
71.2
68.7
54.3
60.3
48.0

show that iron oxides (68.7 and 54.3) are ionic in

character relative to the Si-Obond. However, the overall agreement for the other oxides indicates the validity of both calculations.
,
T h e c o m m o n rock-forming minerals belong to the
silicate system in which in every case a Si-Obonding is
involved. Since for the geologically important oxides
Si-Ohas the least ionic bonding, or at least comparable
to that of iron if Hannays and Smyths equation is
used, it can be said that the percentage of Si-O or
SiO, in a mineral is directly proportional to the percentage of ionic bonding. Since the amount of ionic
character is also related to the weatherability of a
mineral, the amount of SiO, should indicate h o w resistant a mineral will be against weathering. A high SiO,
content means the mineral is resistant to weathering
or the weatherability is low.
Quartz, having a 100 per cent SiO, content should
be the most resistant of all the silicate minerals. Pettijohn (1941), comparing the mineral3 of sedimentary
rocks of increasing geologic age with those of recent
sediments, has found that quartz is one of the most
persistent of the silicate minerals.
The c o m m o n rock-forming minerals and their SiO,
content are listed in Table 2. The SiO, content here indicates the predicted values of weathering. Included for
comparison are the data of M o h r and van Baren (1960)
about the degree of weatherability of the most c o m m o n
rock-forming minerals as observed in soils. Although
there is no complete agreement upon the exact order
of mineral stability during weathering processes, the
data of M o h r and van Baren in general, conform with
the accepted Goldich series.
At first glance there seems to Le no correlation
between the SiO, content, or the predicted weathering
values, and the observed weatherability scale (Table 2).
However, if Lhe btructure of the different silicate systems is taken into consideration, the IwedicLed values
as a whole fit with the observed values. Thus, within

91

II. Ling Ong

TABLE2. Correlation between SiO, content and the observed weatherability of the common rock-forming minerals
Si02
Structure

Name

Chemical composition

Tektosilicates

Quartz
Ei-feldspar
Plagioclabe
Albite
Oligoclase
Andesite
Labradorte
Bytownite
Anorthite
Fcldspathoids
Leucite
Nepheline

Si02

Nesosilicates

Inosilicates

Phyllosilicates

Pyroxenes '
.Enstatite
Hypersthene
Iledenbergite
Diopside
Augite
Aegirine
Amphibole
IIornblende
Mica
Biotite
Muscovite'

100
63

KA1Si,08
NaAlSi308(Ab)
Ah~Abm
Ab,o-Ab,o
Ab,,-Ab30
Abao-Abio
CaA1,Si,08 (An)

69
61-66
56-61
51-56
46-51
43

KAlSi,O,

53
42

NaAlSiO,

Olivine group
Fayalite
Forsterite
Olivine

Fe,SiO,
Mg,Si04
(Fe, M),SiO,

34
55
34-55

Observed
weatherability
scale1

Very low
Low
Low
Medium
nigh

Very high
Very high
Very high2
Very high2

Very high

48
55
48-55
52

hledium
Medium
-

NaCa,(Mg, Fe, Al),

(Si,W8O2,(oW2

38-583

Medium

(&, Fe), (AlSiaOlo)(OII),

KAlz(A1Si 30,J (OH),

33-363
44-463

Very high
Low

MgSiO,
(hfg,Fe)SiO,
CaFeSi,E
CaMgSi,OE
Ca(Mg, Fe)Si,06
,

or
Predicted weatherability
scale (high SiOp cuntent
means resistant to weathering)

40
'

NaFeSi,06

less than 40

1. Xohr and van Daren, 1960, Table 37, p. 114-5.

2. V a n Hise (1904).
3. Average content (Mohrand van Baren, 1960).

the tektosilicates general agreement is evident. And

the predicted order of resistants gives the following
series : quartz, albite, K-feldspar, oligoclase, andesite,
labradorite, leucite,bytownite, anorthite, and nepheline.
Within the phyllosilicates, the predicted order to
resist weathering is in accordance with the observed
facts; muscovite being more resistant than biotite.
This is also confirmed by the experimental work of
Keller, Belgord and Reesman (1963) w h o found that
biotite is more readily dissolved in water than is muscovite. N o t uncommonly in weathering muscovite alters
to illitc or to montmorillonite. This is also true with
biotile which changes to glauconite. In all cases the
transition can be explained in terms of a n increase of
SiO, content.
Ohservcd values for wcatliering of ncsosilicatc anil
inosilicates are too scarce to m a k e a reliable comparison

92

with the predicted values. For the nesosilicatcs, the

predicted order of rcsistancc is :forsterite, olivine, and
fayalite. For the inosilicates:diopside, aegirine, hedenbergite, hypersthene and enstatite.
T o find out why the structure of the different silicate
systems has to be taken into consideration in the prediction of weatherability of minerals, w e have to refer
to the assumption m a d e in deriving the theory of chemical weathering. T h e assumption is that the degree
of chemical weathering is related to the extent of the
chemical reaction between water molecules and the
mineral surfaces. It is further assumed that the chemical
reaction is purely a chemical phenomenon. This is not
the whole account, since the extent of the chemical
reaction also depends o n the surface area and the morpliology of Lhe crystals. A large surface area means a n
extensive chemical reaction. Thus, the physical pro-

Correlation between chemical bond and chemical weatherability

perties of minerals or the atomic arrangements of different silicate systems are important and should be
taken into consideration.
Weathering of the different polymorphs of the c o m m o n rock-forming minerals is another structural problem. The most c o m m o n example is that of SiO,
having four types of modifications as indicated below
(at atmospheric pressure) :

5730

8670

1,4700

L o w quartz = High quartz = Tridymite = Cristobalite.

T h e high-temperaturepolymorph has a more open structure than a low-temperature polymorph; the hightemperature form is dynamically maintained by thermal
agitation (Mason, 1958).
The open structure formed at the high temperature
causes the incorporation of foreign ions other than Si.
Analyses of those minerals show the presence of N a
and Al, suggesting of N a A 1 for Si in the open structures; quartz, on the other hand is generally pure SiO,
(Buerger,1951). F r o m this fact it can be concluded that
the high-temperaturepolymorph will be more subjected
to weathering because of the low Cio, content. Extrapolation of this finding to the weathering of other
c o m m o n rock-forming minerals gives the following
results: (a) for the K-feldspar, microcline is the most
resistant to weathering followed by orthoclase and
sanidine; and (b) for the ragioclase and leucite series,
the low-temperaturepolymorph should be more resistant
to weathering than the high-temperature form. There
is, however, no observed weathering data available
to verify this finding.

S U M RIA IlY A N D CO NC LUS I O N S

A n attempt has been m a d e in this paper to describe

the chemical weathering of minerals on a quantitative

chemical basis, with the purpose of providing a better

understanding on the forces responsible in the weathering processes.

T h e following conclusions can be drawn:
1. The temperature of formation of minerals used by
Goldich for explaining the weathering of minerals cannot be confirmed because of the formation of polymorphism.
2. In chemical weathering, water is the major
cause of the disintegration of minerals. The degree of
weathering is proportional to the magnitude of the
reaction between water molecules and the surfaces of
minerals. A high percentage of ionic bonding in minerals favours an extensive reaction because more of the
polar water molecules will be attracted to the mineral
surfaces.
3. The amount of the ionic bonding, calculated by
using the electronegativity scale for the c o m m o n
rock-forming oxide pairs, shows that the Si-O b o n d is
the least ionic (most covalent) of them ail. Minerals
can, therefore, be arranged according to increasing
Si-O(or SiO,) content or decreasing ionic bonding a n d
consequently, decreasing weatherability.
4. A comparison between the predicted values, based
on the SiO, content, and the observed values for the
weathering of the most c o m m o n rock-forming minerals
shows a close agreement provided the comparisons are
restricted to isostructural groups, wherein the differences are basically in the character of the bonding.
5. T h e predicted order of decreasing resistance towards weathering for the different silicate systems is as
follows : (a) for the tektosilicates : quartz, albite,
K-feldspar, oligoclase, andesite, labradorite, leucite,
bytownite, anorthite and nepheline; (b) for the nesosilicates : forsterite, olivine and fayalite; (c) for the
inosilicates : diopside, aegirine, hedenbergite, hypersthene and enstatite; (d) for the phyllosilicates :
muscovite and biotite.
6. For the polymorph minerals, the high-temperature
form is less resistant to weathering than the lowtemperature form.

Bibliography
BUERGEB,M. J. 1948. The role of temperature in mineralogy.
Amer. Min., vol. 33, p. 101-21.
.1951. Crystallographic aspects of phase transformations.
Phase transformation in solids, p. 181-211. New York,
John Wiley &Sons.
GOLDICII,
S. S. 1948, A study in rock Weathering. J. Geol.,
vol. 46, p. 17-58.
HANNAY,
N.B.; SMYTH, C. P. 1946. The dipole moment of
hydrogen fluoride and the ionic character of bonds. Amer.
ehern. J.,vol. 68, p. 171-3.

JOFFE. J. S. 1949. Pedology.

2nd ed. New Brunswick, Pedology

Publications, 662 p.
KELLER, W.D.; BALGORD, W.D.; REESMEN, A. L. 1963.
Dissolved products of artificdy pulverised silicate rocks and
minerals. Part I. J. sediment. Petrol., vol. 33, p. 191-204.
KRAUSKOPF,K.B. 1967. Introduction to geochemistry, New
York, McGraw-Hill,721 p.
MASON,I). 1958. Principles of geochemistry. 2nd ed. New York,
Wiley, 310 p.
hfOHR, E.c. J.; V A N BAREN,F.A. 1960. Tropical soil. nruxelles, Les Editions A. Manteau, 498 p.

93

if. Ling O n g

MOHEY,G.W. 1964. Phase-equilibrium relations of the comm o n rock-forming oxides except water. In : Michael
Fleischer (ed.), Data of geochemistry, 6th ed., chapter I,.
158 p. (U.S.Ceol. Survey Prof. Paper 440-L.)
PAULING, L. 1932. The nature of the chemical bond, IV. The
energy of single bonds and the relative electronegatively
of atoms. J. Amer. Chem. Soc., vol. 54, p. 3570-82.
PETTIJOIIN,F. J. 1941. Persistency of heavy minerals and
geologic age, J. Geol., vol. 49, p. 610-25.

94

llou, Ilustum. 1949. Decomposition and resynthesis of the

micas, J. Amer. Cerant. Soc., vol. 32, p. 202-9.
SMrTIr, F. G. 1963. Physical geochemistry. Reading, Massachusetts, Addison-Wesley Publishing Co. 624 p.

TUTTLE,
O. F.; BOWEN,N. L. 1950. High temperature albite
and contiguoue feldspars. J. Geol., vol. 58, p. 572-83.
VAN HISE,
Ch. R. 1904. A treatise on metamorphism. M o n o graphs of ihe US. Geol. Survey, vol. 47.

T h e influence of the nature of parent rock on soil

formation under similar atmospheric climates
J. P. Andriesse
Soils Division, Department of Agriculture, Sarawak

INTRODUCTION
A m o n g the great number of processes operating in
soils, those commonly referred to as Podzolization
and Laterizationare probably the most controversial.
A full review of the existing literature on them would
fill volumes and therefore it m a y suffice to say that
most soil scientists agree that podzolization is a prominent feature in most temperate soils while laterization is a process usually found in tropical soils. T h e most
outstanding difference between these processes is considered to be leaching of sesquioxides and relative enrichment of silica in surface horizons in podzolization, and
the reverse when laterization is concerned, namely
leaching of silica and relative accumulation of sesquioxides in surface horizons. T h e concept of soil
zonality on which m a n y genetic soil classifications are
based, e.g. those of Vilensky (1927), Marbut (1928),
Thorp and Smith (1949), implies that these processes
are typically zonal. However, the occurrence of podzols and podzolic soils in the tropics is n o w a wellestablished fact. Klinge summarizes m u c h of the present knowledge on the occurrences of H u m u s Podzols
and concludes that Tropical Podzols occur on both
uplands and lowlands and that upland podzols develop mostly from acid hard rocks, rich in quartz or
river deposits, while lowland podzols occur mainly on
unconsolidated beach and derived dune sands, poor in
clayed material and on Pleistocene and Holocene terrace deposits of similar composition (Klinge, 1968,
p. 49-50).He, however, did not include the Podzolic
soils in his study. Podzolic soils are considered to be
soils showing either translocation and accumulation
of sesquioxides, or clay migration, or both processes
in combination (Stobbe and Wright, 1959 and RlcCaleb,

and subsequent leaching of the clay components, is

commonly regarded as typical for soils of a podzolic
nature while breakdown of clay and subsequent removal
of the components is typical for true podzols. Studies
in Sarawak o n these soils (Andriesse, 1969a and 1969b)
have shown that in podzolization as defined above both
processes occur but that illimerization is a process
typical of clay soils. T a n and van Schuylenborgh in a
series of articles (1959, 1961) have shown that podzolization in the tropics becomes increasingly dominant
with altitude (given the same parent material). This is
the case both under a monsoon climate and under a
continuously wet climate but they consider that under
a monsoon climate laterization moves to higher altitudes than under a continuously wet climate. T h e y
explain this by an increase in humification at higher
altitudes caused b y the lower temperature resulting
in the formation of h u m u s acids while at lower altitude
mineralization prevails over humification because of
the higher temperatures, and carbonic acid becomes
the dominant leaching agent. T h e author was struck
b y the fact that in Sarawak an almost identical range
of soils was found to that described by T a n et al.
In Sarawak, however, changes from laterization to podzolization are related to the kind of parent material:
thus increasingly marked podzolization occurs in soils
o n increasingly acid parent materials. T h e climate in
Sarawak is a uniform factor (continuously wet) and
altitude in the areas studied is too subdued to play a role.
The present paper presents evidence of this and attempts to explain the apparent enteraction which takes
place in soil formation between the atmospheric climate
and the nature of the parent material.

1959).

The process of illimerization (Fridland, 1958),

meaning leaching of clay without prior decomposition

1. Specially presened an supporting paper to review No. 3 by J. van

Schuylenborgh.

95

J. P. Andriesse

TABLE1. Summary of mean monthly and annual rainfall (West Sarawak stations) in inches (period of recording years given
in parentheses after name of station)
Month

January
February
March .
April
May
June
July *
August
September
October
November
December

TOTAL
Absolute maximum
Absolute minimum

Kuching

Sungei Chins

Malang

Ban

(72)

(32)

(17)

Tsrat
(8)

I.iindu

(71)

(20)

Simanggang
(18)

25.80
19.19
13.72
10.13
9.46
8.44
7.55
8.48
10.17
12.77
13.52
18.88

34.73
23.12
15.92
11.31
12.38
7.94
8.79
8.74
12.12
14.75
16.06
23.96

33.76
21.35
12.10
10.75
10.81
6.74
7.77
8.78
9.99
12.01
14.13
20.66

20.57
11.60
9.70
9.69
8.36
6.83
7.07
8.21
8.95
11.85
12.87
13.86

20.67
14.80
11.06
12.08
9.50
5.97
7.49
7.07
11.62
11.33
15.09
12.45

21.62
18.47
13.91
6.86
7.96
7.62
5.56
5.62
7.43
8.57
11.42
16.60

16.47
13.45
12.99
15.18
12.02
8.96
9.53
10.92
11.09
15.47
17.34
16.56

158.11

189.82

168.85

139.13

131.64

159.98

62.28
0.66

60.25
2.23

27.44
2.98

55.56
0.87

30.83
3.00

- - 85.99
2.06 .

CLIMATE
T h e soils and rock types studied are all located in
W e s t Sarawak (latitude 0050 N.20 N.,and longitude
109030E.-111050E.).Table 1 shows the rainfall for a
selected number of rainfall stations having a suficiently high number of recording years to be reliable
and found scattered over the area. T h e highest m e a n
annual rainfall is 4,750 m m and the lowest 3,240 mm.
The rainfall is well distributed over the year, there
being no single month in which the rainfall is less than
150 mm. According to Mohrs rainfall classification for
pedological purposes (Mohr, 1944), the climate is
considered to be continuously wet. Average daily temperature over the area is 25.50 C, the humidity 4 per
cent daily average (Department of Civil Aviation, 1961).
T h e climate is, therefore, characteristic of humid trop:
icai lowlands.

ALTITUDE

All

soils and rock types studied are found below a n

altitude of 500 ft (approximately 150 m) and therefore the influence of altitude on soil formation can be
ignored.

P A R E N T MATEI1IAL
Tables 2 and 3 show a selection of the rock types
occurring in the area. Table 2 concerns igneous rock
types; Table 3 sedimentary rocks (consolidated and
unconsolidated). The different rock types are presented in a sequence showing increasing values of SiO,,
having gabbro at one end of the sequence and an almost
pure quartz sand at the other. For ease of reference the

96

-129.56
70.73
2.07

rock types have been assigned the numbers 1 to 12,

shown below the names.
All rock types are from the same localty as the soil
types studied, although it was not possiible to collect
rocks from exactly the site at which the soil profiles
were dug. Tt is believed, however, that the rock types
are representative for the true parent rocks of the soil
profiles studied.

SOILS
T h e soil types found o n this sequence of rock typeti
are described in the Appendix in which the analytical
data available on these soils have also been collated.
Unfortunately, the analytical data on these profiles
is of a varied nature, coming from different sources.
O n e could have selected just the profiles for which
the same chemical data is available but this would not
have shown the full range of characteristics which,
particularly for the Red-Yellow Podzolics, is of importance. T h e chemical, clay mineralogical and sand
mineralogical data, however, indicate that over the
full range mature soils are concerned. The leaching of
bases is shown adequately by the chemical values of
the exchange complex and the pH range. Leaching of
sesquioxides or silica could not be shown for all profiles
by molecular ratios of the colloidal fractions which is
the most indicative method (Jenny, 1941). For other
profiles therefore total silicate analysis of the fine earth
fraction had to be used instead. In the absence of these,
values for the group III elements (total A1,0,, Fe20,
and Tio,) as analysed using Dobritskayas method (1962)
are shown. Since all profiles studied are mature soils,
it is suggested that none or very little original ironand aluminium-bearing material of the rocks has

Influence of the nature of parent rock on soil formation

TABLE2. Analyses of igneous rock types in W e s t Sarawak

Rock name:
Specimen number:
J,ocaliiy :

sio,
Tio,
AWa
Fez03

Fe0
MnO
MgO
Ca0

Na,O
K,O
H,O+
H20COZ
PZ06

TOTAL

Olivine
gabbro

Basalt

Andesite
3

S6820

S8220

Gunong Gebong,
near Sematan

Granodiorite

Microgranodiorite
7

S13410

S9160

S6258

S728S

S1822

Sungai Baki,
near 32nd mile,
Serian Road

Simuja Quarry,
near
Serian

South side

Gunong Gading.

Tinteng Bedil,
near

C. Sirenggok

A b o k Quarry,
near
Silantek

50.50
1.38
16.70
1.22
8.35
0.15
5.55
8.20
3.30
1.45
2.70
0.13
1.10
0.22

52.70
1.46
14.70
2.10
9.31
0.14
3.69
6.77
3.44
1.36
3.22
0.82
0.03
0.22

65.50
0.34
14.74
1.20
1.98
0.01
1.42
4.15
3.84
0.40
4.19
2.15
n.d.
0.10

70.50
0.16
14.80
0.38
1.82
0.04

71.07
0.55
13.61
0.62
3.29
0.08
1.06
2.32
2.64
3.34
1.05
0.06
0.06
0.13

49.58
0.30
18.88
0.86
4.99
0.12
8.19
14.20
2.08
0.13
0.61
0.03

99.97

100.95

Dacite
porphyry

of

99.96

Lundu

100.02

bgga

Adamellite

2.43
5.80
1.16
1.53
0.46
0.44
0.15

72.11
0.24
12.80
1.51
1.29
0.04
0.71
1.77
4.46
4.14
0.72
0.34

0.02

99.67

99.88

100.15

References: S6820, S728S quoted from Wollenden and Haile, 1963. Table 7. S1822 quoted from Haile, 1954. Table 5. S13410 quoted from
Table 10. 58220 quoted from Wiltord. 1965. Table 8. S6258 quoted from Kirk, 1968, Table 25. S9160 quoted from Wollenden. 1965, Table 6.

Pim. 19 65

TABLE3. Analyses of sedimentary rock types in W e s t Sarawak

Rock name:
Specimen number:
Locality:

SiO,

Tio,
-40,

Fe,O,

Fe0
MnO
MgO
Ca0
Na,O
K,O
II,O+
&OCO,
P206

TOTAL

Arkose
(Triassic)

Sandstone
(Triassic)
10

12

S13201

S8548

510849

S4307

S7391

Snngai Tarat

Sungai Bukar

Sungai Konggong.
south-westof S e d a n

SerianISimanggang
road, 58th mile

Sematan area

88.4
0.14
4.65
0.45
0.91
0.03
0.32
1.30
0.87
1.05
0.30
0.91
0.63
0.04

66.66
0.75
18.54
1.60

96.29
n.a.
1.60
1.86
0.02
n.a.
ma.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.

67.9
0.55
14.5
1.63
2.65
0.07
1.59
1.38
3.05
2.65
2.65
0.78
0.37
0.09

78.3
10.75
0.84
1.55
0.41
0.31
2.48
3.52
0.17
0.91
0.22
0.25
0.06
0.02

99.9

99.79

'

Sandstone
(Tertiary)
11

Carbonaceous shale
(Triassic)

100.0

n.a.
0.64
0.09
ma.

n.a.
n.a.
n.a.
n.a.

'

Marine sediment
(Sub-Recent)

88.28

99.77

n.a.

Referencei: S13201 quoted from Pimm,1965, Table 9. S8548 quoted from Wilford. 1965, Table 5. S10849 quoted from Wilford, 1965. Table 84. S4307 quoted
from Andriesne, 1969. Table III. S7391 analysed by S o h Laboratory, Sarawak.

97

J. i. Andriesse

remained in the solum, and that total analyses of the

fine earth for these fractions are indicative for leaching
processes. Therefore, although the various chemical
data cannot be used to compare levels of leaching processes they can be used to show trends.
. Jn the text, Sarawak series names have been used;
these series were correlated with the United States
genetic soil classification (Great Soil Groups) by Thorp
and Smith (1949). Although this classification m a y be
out of date, and is n o w replaced b y the 7th Approximation and amendments (United States Soil Survey, 1960,
1967), for genetic studies and correlation of genetic soil
types it is still a useful tool, particularly since m u c h of
the existing literature on the subject has employed
this genetic classification.
In Sarawak, on the various rock types, we distinguish, therefore, the following Great Soil Groups :
Ileddish-Brown Lateritic soils, Red-Yellow Podzolic
soils, Grey-White Podzolic soils, and Podzols. T h e
Grey-White Podzolics are a group distinguished in
Sarawak and comprises soils with all the characteristics
of the Red-Yellow Podzolic soils except for the very pale
to white colours caused by very low iron content of the
parent rock. T h e correlation is based on a study of
these soil groups b y McCaleb (1959), and Stobbe and
Wright (1959). Since some of the characteristics of the
soils, G r e y W h i t e Podzolic soils and Podzols. T h e
texture and colour, for this study the full sequence of
local beries is used to indicate the full range present.
Texture in particular appears to be an important factor within the group of Red-Yellow Podzolics and can
frequently be related to the intensity of the podzolization process.

DISCUSSION
If w e take it that in both podzolic soils and podzols
podzolization in soil. profiles can be summarized as
is at work then the morphological manifestation of
podzolization in soils profiles can be summarized as
follows :
1. Removal of clay from A horizons and enrichment of
clay in the B horizons. Clayskins and fillings around
structural units and in old root channels or cavities
(argillic horizon-7th Approximation).
2. Removal of sesquioxides from A horizons and subsequent accumulation in the I3 horizons indicated
b y mottling and iron concretions.
3. Pale colouration of the A, horizon, coupled with
correspondingly deeper colours in the B horizon
(formation of albic horizon-7th Approximation).
4. Removal of organic material d o w n cavities; organic
staining of the A, and B horizons.
5. T h e formation of iron-humus accumulation horizons
in the I3 position (formation of the spodic horizon-7th Approximation).
98

Table 4 shows these morphological features as found

in the Great Soil Groups occurring in Sarawak.

TABLE4
Morphological features
of podzolization

Soil groups

Reddish-Brown
Lateritic soils
Red Podzolic Soils
Yellow Podzolic soils
GreyWhite
Podzolic soils
IIumus Podzols

- - - - X
X

X
X

-Not present
x Preeent
I.

Depending on characteristics (texture mainly) of soil series within the

group.

Table 4 shows that the Reddish-Brown Lateritic soils

do not show morphological evidence of podzolization
whereas through the sequence d o w n to Humus Podzols
the morphological evidence increases. When following
the c o m m o n definition that in laterization silica is
selectively removed from the A horizons then in Saraw a k even on basic igneous rocks no laterization takes
place. This is shown by the,molecular ratios of the
colloidal fraction of the Tarat series (see Appendix).
T a n et al. (1959) refer to a process of laterization in
,their soil on andesitic material at low altitude under a
monsoon climate. This soil is classified as a ReddishB r o w n Latosolic soil. For comparison the description
and chemical data are given below.

9. Karangpandan. Intersected region. Altitude :

600 m. Flat and horizontal part of a hill. Vegetation:

Profile

Manihot utillissima. M e a n annual air temperature :

22.70 C.
A, 0-12 c m Reddish-brown (6YR 5/4 to 5YR 4/3)
clay with well-developedcrumbly to nutty
structure. Friable. M a n y roots.
Reddish-brown (5YR 5/4 to 5YR 4/3)
clay with well-developedfine nutty structure. Iron coatings. Friable. Few roots.
31-90 c m Reddish-yellow (5YR 6/6) to reddish
brown (5YR 4/3) clay with well-developed m e d i u m nutty structure. Iron
coatings. Friable. Termite and ant activity.
+90 c m Light-brown (7.5YR6/4)to reddish brown
(5YR4/3) clay. Greyish, strongly weathered stones are present. Friable.

BI 12-31 c m

n,

n3

Jnfluence of the nature of parent rock on soil formation

Analyses of ProJile 9
Molecular ratios of the colloidal iraciion

SiOz/R.LOa SiOz/AlzOa SiOzIFezOa AIzOalFez03

A,
13,

13,

0-12cm
12-31c m
31-90c m
> 90cm

1.58
1.63
1.67
1.73

1.89
1.93
1.97
2.03

9.70
10.26
11.12
11.82

5.14
5.31
5.66
5.82

Reference: Tan and van Schuylenborgh (1959).

Morphologically, this profile is comparable to the Tarat

series, and asbuming that latosolic is synonymous
with lateritic, the classification at Great Soil Group
level is the same. The chemical data, however, reveals
that a laterization process is operative in,the former.
This could be explained as follows:
M o h r and van Baren (1954)in their review on lateritic soils consider all boils with sesquioxide accumulations to be lateritic. These accumulation horizons m a y
be found either at the surface or at depth. T h e y come
to the conclusion that most lateritic soils are fossil
soils and even in Indonesia they appear to be extremely
rare. They, surprisingly, pay little attention to the great
majority of soils occurring under the names of red
loams, red earth, Red-Yellow Podzolic, Red-Yellow
Latosolic soil which probably form the bulk of tropical
soils. In the view of the writer, what is considered a
lateritic soil by Mohr and van Baren are senile soils,
the end stage in soil formation. T h e observation by
M o h r and van Baren that these soils are mainly fossil
m a y accord with this view. The great bulk of red loams,
red earths, etc. maybemature or immature soils showing
great similarities and on the w a y to becoming a lateritic
soil (in the sense of M o h r and van Baren), a podzolic
soil or a Podzol. The question whether they will ever
form does not arise, since this m a y be dependent on a
number of factors which m a y start to play a role in
later btages of development. F r o m the available
evidence one must conclude that T a n and van Schuylenborghs Reddish-Brown Latosolic soil and the ReddishB r o w n Lateritic boil of Sarawak are both mature stages
of the same soil type, the latter, however, m a y have
slightly further progressed towards a podzolic soil.
The Red-Yellow Podzolic soils show certain aspects
of podzolization in the profile but this is dependent on
the clay content of the soil. In the case of sandy textures clay removal is more obvious and in the case of
iron-poor soils, such as the Nyalau series, bleaching
of the A, is more apparent. T h e analysis suggests that
in all the Red-Yellow Podzolic soils sesquioxides increase with depth. Therefore, both the morphological
and chemical features show that a podzolization process is involved. Organic staining is only visible in the
very pale coloured soils, such as the GreyWhite Pod-

zolics, and even in a clay profile, such as found in the

Kerait series, this is noticeable.
Summarizing this, since there is as yet no method to
measure quantitatively the intensity of the podzolization process, one must be guided by the morphological
manifestation in the profile, which m a y show stages of
maturation. It is obvious that the intensity of the process follows the trend as indicated by Table 4, the clay
profiles following behind the sandy profiles within the
group.
Figure 1 shows the position of all these soil series and
the genetic soil groups arranged in order of total SiO,
content of the parent rock (indicated by the numbers 1
to 12). It is immediately obvious that a good correlation exists between the genetic soil types and the nature
of the rock types. Within the groups, owing to differences in texture (as explained), the position of the
series changes somewhat. There is a shift from the top
left to the bottom right in the diagram from the soil
showing the least signs of podzolization towards soils
showing signs of very intensive podzolization. This is
related to the increase of total SiO, and decreases of
weatherable minerals of the parent rocks.
T o show the influence of the chemical composition
of the rocks on soil formation more clearly, Table 5 was
set up. T h e total cation content of the rocks comprises
the s u m total of Ca, M g , K and N a oxides. It is logical
to assume that these o n their release influence the c o m position of the h u m u s and acidity of the leaching agent.
There appears to be well-expressedrelationship between
the lateritic-podzolicrange of soils and the range found
in the s u m total of rock cations.
Secondly, since Figure 1 and field evidence suggested that texture plays an important role in the intensity of the podzolization process, total Alzo3and total
quartz contents of the rocks (where the information
was available) were compared with the texture range
of the soils. It was here assumed that Al,03 and original
quartz content of the rocks influence the texture of the
soil formed after rock weathering. The table tends to
confirm this. Finally, the colour of the soil was found
to be somewhat controlled by the total iron content
of the rocks. This is more so the case with clay soils
than with sandy soils. This can be expected since in the
sandy soils podzolization appears to be more severe
and m u c h of the original iron content of the rock m a y
have been removed by this process.
Figure 1 and Table 5 show clearly that a strong
relationship is present between the chemical composition of the rock type and the intensity of podzolization.
It should be remarked that this correlation m a y not be
present if one deals with soils in various stages of development. T h e clay mineralogy and the sand mineralogy,
however, indicate that the examples quoted are all in a n
advanced state of weathering. According to Jenny (1941)
relationships between the nature of parent material
and the genetic soil types formed on them are not
always conclusive because the correlations attempted

99

J. P. Andriesse

Podzolic soils

.eddish-brownLateritic soilr

Podzols
Red Podzolic soils

Dlivine asalt
Gabbro

Andesite Dacite
porphry

Jebong

Tarat
(3)

(1)

Tarat
(2)

Yellow Podzolic soils

Grano- Microgranoliorite
Gading Ibok
(7)
(8)

4damellite
;ranite

diorite

Gumbang
(4)

lagoi

(9)
General
decrease
in
weatherable
minerals

GreyWhite
Podzolic
soils

Carbon

Arkose

Sandstone Quartzitic
sandstone

Serin
(6)

Nyalau

aeeous

shales
Kerait

(5)

(10)

Silantek
(11)

Quart-

zitic
alluvium
Pueh
(12)

General increase in SiO, content = decrease total iron content.

General decrease in A1,0, content increase in grain size.
FIG. 1. Relation between genetic soil types and kind of parent material under a continuously wet tropicalclimate.

TABLE5. Relation between the chemical composition of rocks and

Total cations
rock

soil groups

Total quartz
rock

nil

Texturesoil

Clay-clayloam
Clay-clayloam

3.18
4.91
4.28

Red-yellow
Red-yellow
Red-yellow

Clay

0.75

White

2.20
2.80
2.39

Yellow
Yellow
Yellow

2.36
1.86

White-pale yellow
White

4.4

Gumbang
Gading
Serin

9.81
9.36
8.67

14.74
13.60
14.50

30.18
34.4
23.50

S.cl.-s.cl. loam

GreyWhite
Podzolic
6Oih

Kerait

2.45

18.54

n.a.

Yellow
Podzolic
soils

Abok
Jagoi
Nyalau

9.23
11.08
6.31

14.80
12.80
10.75

26.1
35.9
n.a.

Clay-clayloam
S.cl.-s.el. loam
Sand-s.cl.loam

Podzols

Silantek
Pueh

3.49
0.25

4.65
1.60

ma.

n.a.

Sand-sandyloam
Sand

Red
Podzolic
Soil8

nil

Soil colour

5.85
9.57
11.41

18.88
16.70
14.70

Jebong
Tarat
Tarat

Total iron
content rock

Clay
Clay
Clay

24.60
18.50
15.26

Reddish-Brown
Lateritic soils

100

Total Al208
rock

soil characteristics

'

RedRed
Red

Influence of the nature of parent rock on soil formation

in the past are as a rule suffering from lack of control

of soil-forming factors, particularly climate. This is in
the present correlation not the case, the climate, as
shown, is constant and care has been taken to include
in this study only profiles from topographically identical
sites.

CONCLUSIONS
The similarity of this bequeme of genetic soil types
found in Sarawak on the various rock types, with a
sequence of genetic soil types usually linked with either
an iucreasing altitude in the tropics or, on a global
scale, with increasing latitude has interesting consequence. It m a y indicate that neither altitude nor climate can be regarded as being solely responsible for
such a sequence. This was also indicated by the studies
of T a n et al. (1959),w h o found that under a monsoon
climate but with andesitic material laterization m o v e d
to higher altitude while under a continuously wet climate podzolization was still found operating at almost
sea level. However, for the later study (1961)they used
soils developed over acid parent material. They reached
their conclusion by blaming the climatic difference
but the present study bears out the fact that the difference in parent material m a y well be responsible for
this. Also, although from their study of the chemical
composition of the h u m u s they arrived at the conclusion
that the differencesare related to an increase of humification at higher altitudes and an increase.in mineralization at lower altitude, they suggest that in the
case of their Red-Yellow Podzolic on acid parent material at sea level humification was caused by the acidity
of the parent material.
F r o m these studies, one may, therefore, conclude that
the functions of climate and parent material in soil formation are interchangeable and are interacting. If w e
accept the fact that the rate of mineralization or humification is responsible for either laterization or podzolization, another conclusion can be drawn, namely that
humification and mineralization are as m u c h dependent
on the nature of parent material as on climatic factors.
T h e still generally accepted idea that mineralization
in the humid tropical lowlands is rapid because of the
high temperature is not always borne out by facts. In
Sarawak, as shown in the profile descriptions, podzols
have a thick O horizon under Primary Forest. Also, a
Yellow Podzolic soil, such as the Nyalau series, has
generally a thick O horizon under Primary Forest.
T h e Reddish-Brown Lateritic soils usually have no O
horizons.
F r o m these facts the following theory is propounded.
Humification or mineralization are dependent on two
factors, parent rock and climate. O n basic igneous rock
types, with adequate bases the vegetation will initially
produce litter rich in bases and with a pH which m a y

he around neutral to weakly acid. Under high t e m perature and high humidity biological activity is at
its m a x i m u m and rapid mineralization will occur,
resulting, as shown by T a n and van Schuylenborgh
(1961) in the formation of mainly carbonic acid
which selectively leaches silica from the soil. However,
with increasing maturation the litter will become
increasingly more acid and the biological activity
decreases resulting in an increase of humification
and consequently the formation of fulvic and humic
acids which react with the sesquioxides in the soil
(Alexandrova et al., 1968) and m o v e them d o w n the
profile. In Sarawak, clay mineralogical studies have
shown that a young soil on basalt (an immature Tarat
beries) contains largely gibbsite and goethite. T h e
mature Tarat series contains mainly Kaolinite with subsidiary goethite and gibbsite. This m a y show that with
increasing age the aluminium m a y become more mobile.
This could be the onset of podzolization. Apart from
mobilization of sesqioxides as chelates leaching of aluminium in an ionic form can be expected at a pK of
less than 4, as shown by Pickering (1962). Therefore,
initial laterization m a y be followed by podzolization due
to rapid leaching of the bases from the soil and subsequent impoverishment of the humus. This m a y explain
why in almost similar soil types a change from laterization to podzolization can be noticed (cf. R a n and
Schuylenborghs Reddish-Brown Latosolic soil and
Sarawaks Reddish-Brown Lateritic soil).
Similarly, podzolization on increasingly more acid
rock types m a y start at an earlier stage in the maturation process because the solum becomes poor in bases
more rapidly. This m a y also explain why podzolization
is more intense in sandy textures. As shown the sandy
textures in a mature profile in the tropics are related
to quartz content and total aluminium content of the
parent material. The ultimate is reached in pure quartz
sand where complete podzolization is reached in a
comparatively short period (Andriesse, 1969b).
This theory consequently means that under a continuously wet tropical climate complete podzolization
is the ultimate stage in soil weathering regardless of the
parent material. In actual fact, this does not happen
because on basic igneous rocks mainly clay soils form
and the complete breakdown and removal of all clay
minerals m a y only be reached in a period usually
measured in a geological time scale and climate or landscape m a y have undergone such changes that this stage
is never reached. However, the lateritic sheets (which
are called fossil by M o h r and van Baren) m a y fit into
this picture and they m a y well be the result of removal
and accumulation of iron/aluminium compounds through
podzolization, rather than that their occurrence could
be explained by relative enrichment caused by laterization, an opinion also held by Thorp (1933,quoted from
M o h r and van Baren, 1954).
Following the samc trend of thought, laterization
should also exist at higher latitudes than the tropics

101

J..'I Andriesse

if the parent materials are basic, keeping the climate

constant. In other words, at a given latitude, one should
find increasingly more signs of podzolization o n increasingly more acid parent rocks. Pickering (1962)
from his leaching studies concludes that laterization
o n Lasic materials in temperate regions cannot b e ruled
out. It is suggested that the role of the parent rock is,
in large measure, taken over by the climate and at a n
increasing rate at higher latitudes where the rate of
humification (biological activity) is largely controlled
by low temperature and the acidifying effect of selected
forest types (coniferous). T h e resulting increase in the
formation of fulvic and humic acids will acidify the
topsoils rapidly. In a n early stage of soil formation,
even o n basic igneous rock types, podzolizationwill then
occur. T h e great difference between these processes
in the tropics and in the temperate regions is that in the
tropics weathering is intense to great depth; particularly clay formation is rapid and therefore if in a
later stage of soil formation, podzolization follows
laterization, the former process will be operating in
a totally different m e d i u m than in the temperate
regions.
It can, therefore, be concluded that laterization
and podzolization can both operate o n a world-wide

scale at different periods in the maturation process

in the same soil profile, depending o n the nature of the
parent material and the climate. It is contended that
laterization is a process more dominant in the tropics
than in the temperate regions mainly because the acidifying effect of the climate through the higher rate of
humification is more pronounced in the temperate .
regions and that in a n earlier stage in soil formation
than in the tropics. Laterization m a y follow podzolization but never the reverse under normal conditions.
T h e great number of intergrades between the lateritic
soils and podzols are indicative to the interaction
between climate and parent rock, particularly in the
subtropics. It should be realized that where site factors
overrule the factors of climate and parent rock, other
criteria will have to be used. Possibly the occurrence
of Lateritic Earth on limestone in the Mediterranean
m a y s h o w the influence of basic parent rock o n laterization at high latitudes at one extreme, while the h u m u s
podzols in the tropics, formed largely o n quartz sands,
m a y be the other extreme. In both cases, the parent
materials, because of their highest possible contents '
of bases and silica, overrule any influence of climate
o n the resulting soil-forming process. .

Appendix

TARAT SERIES
AZtitude:
Vegetation:
Parent material:
Topography:
Rainfall:

150 ft.
Young rubber with dense undergrowth with fern dominant.
Andesite.
Broken terrain,strongly dissected. Slope 2.50.
f 140 in. annually, 7.07 lowest mean monthly.

Description:

Very thin layer of leaf litter (1/10in.).

Al

0-4 in.

Reddish-brown (5YR 4/4), clay, fine subangular blocky to fine angular blocky structure (nutty). Dry.
Friable. Abundant rootlets. Porous. Distinct boundary to

BI

4-15 in.

Red (2.5YR 5/6), clay. crumbly. Moist. Friable. Many roots. Shiny surfaces of natural ped surfaces
possibly indicating orientated clay. Macro structure-coarse prismatic. Indistinct boundary to

B*

14-24 in.

Red (2.5YR 5/6), clay. As above horizon but the soil is firm and does not break into crumbs on pressure.
Slightly moist. Indistinct boundary to

B3

21-52 in.

Red (2.5YR 5/6), clay loam which breaks into small crumbs and fine angular blocky peds if slight
pressure is applied. Dry.Soft. Porous. Slight development of shiny ped surfaces. Many roots. Scattered
small weathered rock pieces (possibly colluvial).

51-80 in.

Mixture of red (2.5 YR 5/6), clay loam very friable to powdery,and brittle thoroughly weathered parent
material. SO per cent-50 per cent. Slightly moist. Few rootlets. Porous.

1 o2

Influence of the nature of parent rock on soil formation

GUbIBANG SERIES
Altitude :
Vegetation:
Parent material:
Topography :
Rainfall:
'.

50 ft.
Rubber garden with secondary jungle. Undergrowth mainly ferns.
Porphyritic Dacite.
Strongly dissected, mountainous terrain. Slope 100 (foot slope).
130 in. annually, 6.83 in. lowest mean monthly.

Descripiion:

O-&

A,

4-9 in.

Yellowish brown (1OYR 5/8), clay, weak subangular blocky to crumb structure. Moist. Slightly plastic.
Firm. Many rootlets. W e a k clayskins development around larger peds. Scattered small, partly
weathered rock pieces (colluvial and some hard rock fragments, mainly quartz (smaller than 3 in).
Total volume of stone less than 5 per cent. Distinct wavy boundary to

BI

9-20 in.

Reddish yellow (7.5YR 6/8), clay, with sporadic quartz grit (2-3rnm in size). Weak, fine angular blocky
structure, on pressure plastic and slightly sticky. Moist. Firm. Few roots and rootlets. Many partly
weathered rock pieces (colluvial) about one inch size and some larger. Small pockets of bluish-black

in.

Partly decomposed litter of fern origin mainly.

powdery material in weathered rock. Clayskins moderately developed on old ped faces. Illuvial clay
mainly coating larger cracks evidenced by colour dierence. Weathered rock approximately 20 per
cent of soil volume. Indistinct boundary to

Ba

20-40 in.

Similar to the horizon between 9 and 20 inches but clayskins are stronger developed. Weathered
colluvial boulders and gravel 40-50per cent in volume. Indistinct boundary to

Bdd

40-60 in.

Reddish yellow (7.5YR6/8), clay loam,veryfine angular blocky to crumb structure. Many small,black
spots (sand size). Plastic and non-sticky.Moist. Clayskins developed along large cracks. Weathered
colluvial boulders and gravel about 50 per cent volume. Distinct boundary to

60-74 in.

Red (2.5 YR5/8), soft, silty-textured highly weathered bouldery material with white spots (indicating
former phenocrists) mixed with reddish-yellow clay similar to that found in horizons above. Slightly
moist. Distinct wavy boundary to

74-90 in.

Red (2.5YR 5/8), soft, silty, very friable thoroughly weathered porphyritic Dacite, many white
phenocrists.

GADING SERIES
Altitude:
Vegetation:
Parent material:
Topography :
Rainfall:

50 ft.
Old rubber garden with weed undergrowth.
Granodiorite.
Mountainous terrain. Foot slope, 150.
f 132 in. annually, 5.56 in. lowest mean monthly.

Description:

Al
AZ

BI

'

0-5 in.

Dark brown (10YR 4/3), weak subangular blocky sandy clay loam. Friable. Organic matter present.
Weil rooted. Gradual but distinct change to

5-16 in.

Yellowish brown (IOYR, 5/8), sandy clay loam. Smeary. Individual coarse sand grains, (gritty). No
apparent structure. Friable. Very few roots. Moist. Very gradual change to

16-44 in.

Reddish yellow (7.5YR 6/6), sandy clay loam. Slightly plastic, structureless to weak crumbly. Slightly
firm.No roots. At lower depth slightly more reddish coloured.

13,

44-56 in.

Yellowish red (5YR 5/8), sandy clay loam, in places gritty. W e a k crumbly. Slightly firm. No roots.
Gradual change to

56-60 in.

Red (2.5YR 5/6), gritty .clay (sandy clay), smeary. Structureless. Much glimmers in rock debris.
Material in this horizon can be separated into partly weathered rock, disintegrated but coarse
particles not yet weathered, and red clay material. No roots. Moist.

103

J. P. Andriesse

ABOK SERIES
Altitude:
Vegetation:
Parent material:
Topography:
Rainfall:

'

150 ft.

Old rubber garden (seedlings) mixed with secondary growth.

Microgranodiorite.
Broken hilly terrain. Slope 250.
f 160 in. annually, 8.96 in. lowest mean monthly.

Description:

Al

0-1 in.
1-2 in.

A,

2-13 in.

BI.*
BI

13-46 in.
46-52 in.

Dark brown (1OYR 3/3), litter, mainly decomposed leaves and roots.
Very dark greyish, brown (IOYR 3/2), clay loam with weak grey mottles (surface gleying). Crumbly.
Friable. Well rooted. Moist. Distinct change to
Yellow (1OYR 3/8), clay loam. Crumbly. Friable. Well rooted. Moist. Diffuse boundary to
Yellow (1OYR7/8), clay loam. Crumbly to subangular blocky. Compact. Moist. Distinct change to
Yellow (IOYR 3/8), clay loam with reddish yellow (7.5YR 6/8), mottles. Compact. Hard iron concretions. Large roots. Moist.
,

Distinct iron accumulation in B,.

Remarks:

SILANTEK SERIES
Altitude:
yegetation:
Parent material:
Topography:
Rainfall:

250 ft.
Lowland Tropical Heath Forest.
Quartzitic sandstone. (Tertiary).
Moderately dissected hilly terrain. Dip slope 80.
f 132 in. annually, 5.56 in. lowest mean monthly.

Description:

1o4

0-2 in.

Partly decomposed (5YR 2/2), dark reddish brown, organic matter with few sand grains, mixed with
dense rootmat of fine roots mainly, some large roots, slightly moist. Clear over

2-5 in.

Dark reddish brown (5YR 3/2), sand. Much organic matter. Friable. Crumbly. Moist. Individual sand
grains are white in colour and clear. Abrupt but in places wavy boundary to

5-9 in.

Reddish grey (5YR 5/2), medium sand (humus stained). F e w roots. Moist. Single grain. Firm. Clear
but wavy change to

9-13 in.

Light grey (IOYR7/1), medium sand with reddish grey staining in places (75 per cent light grey25 per cent reddish grey). Single grain. Firm. Some veins of humic material run through this,horizon
without any apparent direction. No roots. Abrupt over

13-18 in.

Dark reddish brown (5YR 2/2 and 3/2), loamy medium sand. Weakly cemented. Some fine roots at
boundary with horizon above. Sand grains are bleached and glitter in cemented material. Irregular but
clear change to

18-22 in.

Light yellowish brown (1OYR 6/4), fine' sandy loam. Slightly wet. Many old decomposed roots. Small
pockets of (5 YR 2/2), colour where material is cemented. Platy structure with humus accumulation
'
between structure elements laid, distinct change to

22-33 in.

Very pale brown (1OYR 7/3), loamy sand to sandy loam. Compact. Structureless. Slightly wet. Many
old root channels with organic material (5 Y R 2/2). Some organic material accumulated along fracture
planes. Clear change to

33-52 in.

Pale yellow (2.5YR 8/4), sandy clay with (10 YR 6/6),mottling, in some places as lateralbands in others
along old root channels. Sticky and plastic. Some quartz pebbles (rounded), at 44 in. becoming more
sandy and resembling sandstone.

52-68 in.

.White medium sandstone. (Deep augering confirms occurrence of white clay bed at 68-76in.). Perched
water table at 48 in.

Influence of the nature of parent rock on soil formation

rumr

SERIES
10 ft.
Bracken type ferns.
Sub-Recent beach sand.
Flat, raised beach.
132+ in. annually, 5.56 in. lowest mean monthly.

Altitude:
Vegetution:
Parent material:
Topography :
Ruinfal:

&

Description :

O-$

in.

Light grey (10YR 7/1), medium sand. Loose. Very dry. No organic matter. Abruptly overlying

Al

8-7

in.

Light brownish grey (10YR6/2),medium sand. Loose. Single grain structure. Many roots. Many dark
grey infilled worm holes. Slightly moist. Gradual regular change to

A2

7-14 in.

White (10YR 8/2), medium sand. Loose. Structureless. Slightly moist. Very small roots stopping
at 12 in.

A,

14-15 in.

Brown mottled white sand, boundary irregular, sloping down with slightly steeper inclination than
surface slope.

Bh(ir)

15-19 in.

Dark brown (7.5YR 4/4), medium to fine sand with common hard (2.5 YR 3/6), dark red, concretions
and soft mottle-like concretions of same colour. Structureless. In places this horizon is brittle and
weakly cemented. Slightly moist. Clear regular change to

BI&)

19-24 in.

Reddish brown (5YR 4/4), medium sand. Structureless. Firm. C o m m o n dark red concretions and
mottles (as 15-19 in. but colour less brown). Slightly moist. Clear, very wavy change to

BZ.1

24-44 in.

Strong brown (7.5YR 5/8), Cine sand. Firm. Structureless. No roots. Slightly moist. Gradual wavy
change to

44-54 in.

Strong brown (7.5YR 5/6), fine sand with abundant strong dark brown (7.5 YR 4/4)mottles and some
red. Structureless. Irregular wavy change to

54-60 in.

Light olive brown (2.5 YR 5/6), medium sand. Wet. In places strongly stained black to strong brown.

Bz.2
,

B(mn)

JAG01 SERIES
Altitude:
Vegetation:
Parent material:
Topography:
Rainfall:

350 ft.
Secondary forest.
Adamellite.
Mountainous terrain,strongly dissected. Slope 15".
Estimated between 140 and 160 in. (no records).

Description:

o .
Al

0-5 in.
4-4 in.

Mainly coarse litter.

4-20 in.

Light olive brown (2.5Y 5/4), sandy clay loam. Crumbly. Firm. Many rootlets. Gradual change to

Greyish brown (2.5 Y 5/2), sandy loam. Single grain. Much decomposed organic litter. Organic staining
and infiltration of organic material along root channels. Friable. Many rootlets. Merging into

AZ
Bl
BI

20-34 in.

Yellow (10YR 8/6), sandy clay loam, angular blocky. Firm. Clear infiltration of clay. Gradual change to

34-46 in.

Yellow (10YR 7/6), sandy loam clay to sandy clay. Very firm. Greenish small mottles. F e w roots.
Moist. Gradual change to

46-55 in.

Yellow (1OYR 7/6), sandy (gritty) clay. Weathered granite embedded in sandy clay. Limonitic hard
iron present and red coloured soft iron accumulation. F e w fine root#.

105

J. I. Aiidriesse

KERAIT SERIES
f 150 ft.

Altitude:
Vegetation:
Iararent material:
Topography :
Rainfall:

Secondary growth, mainly ferns (bracken type).

Carbonaceous shales. (Triassic).
Broken hilly terrain. Slope 150.
4140 in. annually, 7.07 in. lowest mean monthly.

Description:

A,

0-3 in.

Light grey (2.5Y 7/2), sandy clay loam. W e a k humus staining from 2-3in., surface gleying also present.
W e a k platy structure, densely rooted, moist. Clear regular boundary to

3-15 in.

Light grey (2.5Y7/2), sandy clay loam with faint few light grey (1OYR7/1), and yellow (1OYR 7/8),
mottles. Massive. Compact, Large cracks give rise to formation of large prisms when soil dries out. In
the cracks dense root system, remainder of soil sparsely rooted. Clayskins along cracks. Gradual
increase in occurrence of yellow small mottles. Moist. Gradual change to

u,

15-30 in.

White clay (2.5Y 8/0), maximum concentration of (10 YR 7/8), yellow mottles, particularly where
quartz grit is present. Quartz grit occurs in pockets and as disturbed thin stonelines (from quartz
strings in parent material). Roots only present in extending cracks from surface horizons. Massive.
Compact. Moist.

Bm

30-60 in.

White (2.5Y 8/0), clay with pockets of quartz-grit, weakly mottled yellow. Light grey colour of
A horizon persists along cracks (clay illuviation). Roots mainly confined to craeks. Massive. Compact.
Moist. Gradual change to

B2

60-100in.

Light grey (1OYR 7/1), clay, massive and compact with small common strong brown (7.5 YR 5/8).
mottles. Illuvial clay noticeable in large cracks. No roots. Pockets of quartz-grit.Moist. Abrupt but
irregular boundary to

100-172in.

Light grey (IOYR 7/1), silty clay, massive and very compact. No cracks or roots, slightly moist.
Abrupt irregular boundary to

172+

Soft, easily cut black shale with quartz-strings. Inclusions of fossil roots and olive yellow coloured
pyritic material particularly along fracture planes.

A,

B/C
C

in.

SERIN SERIES
Altitude :
Vegetation:
Parent material:
Topography :
Rainfall:

150 ft.
Secondary forest (many wild rambutan and durian trees).
Arkose.
Broken hilly terrain.
f 140 in. annually, 5.97 in, lowest mean monthly.

Description:

A,

0-4 in.

Dark greyish brown (1OYR 4/2), clay loam. Abundant roots and rootlets. Friable. Weakly developed
subangular blocky. Moist. Irregular boundary

A,

.4-10 in.

Brown to strong brown (5YR 5/6), clay loam with many fine to medium brownish and greyishmottles.
Moderately developed subangularblocky. Slightly firm. Slightly porous. Moist. Distinct wavy boundary
to

BI

10-34 in.

Reddish yellow (7.5YR 6/8), clay. Strongly developed subangular blocky. Firm. F e w roots. Slightly
porous. Moist. Indistinct boundary to

BI.*

34-44 in.

Reddish yellow (5YR 6/6), clay with few faint light reddish mottles. Strongly developed medium
subangular blocky. Firm. Abundant iron concretions. Moist. Indistinct boundary to

B*.2

44-56 in.

Yellowish red (5YR 5/8), clay. Strongly developed medium blocky structure. Firm. Abundant fine
weathered arkose. Non-prous.Diffuse boundary

B/C

56-68 in.

Yellowish red (5YR 5/8), clay, with few very fine yellow mottles. Strongly developed medium blocky
structure. Firm. Abundant weathered arkose. Moist.

106

Iniuence of the nature of parent rock on soil formation

NYALAU SERIES
Altitude:
Vegeiaiion:
Parent maierial:
Topography:
Rainfall:

350 ft.
Lowland Dipterocarp forest. .
Thick bedded sandstone (Tertiary).
Near summit of hill. Slope 150. Broken hilly terrain.
f 190 in. annually, 7.94 in. lowest mean monthly.

Descripiion:
Surface scattered with litter of varying depth. (4-2in.).

0-2 in.

H u m u s containing abundant fine to medium roots, leaves in varying stages of decomposition.

A,

2-6) in.

Pale brown (10YR 6/3) sandy loam. Distinct common strong brown mottles. Subangular blocky to
crumbly. Friable. Roots common coarse to fine. Moist. Clear, undulating change to

As

64-22 in.

Pale yellow (2.5Y 7/4), sandy clay loam. C o m m o n diffuse greyish brown (10 YR 6/3), mottles. Pale
brown coloured material from Al leached through root channels and cracks. Subangularblocky. Friable.
F e w coarse and medium roots. Moist. Difluse change to

Ba

24-43 - in.

Yellow (2.5Y 7/6), sandy clay loam with few distinct white and abundant distinct light grey and
brownish yellow mottles. Subangular blocky. Firm. Moist to very moist. Diffuse change to
.

B/C

45-54 in.

As 24-43in. but with also common white and strong brown mottles. Massive. Subangular blocky. Firm.
Moist to very moist.

Remarks :

Distinct O and A, horizons.

ANALYTICAL DATA OF A RANGE OF S A R A W A K SOIL SERIES

ON INCREASINGLY MORE ACID PARENT ROCKS-overleaf

107

s:t!
w
B

4
4

2
O

108

w m m
I??
w m m
o r i m

O??

mc-m

ri

*.

d r i d r i

-ri--

00000

0 0 0 0 M

Mc.lcJM

VI

El

t-

O 0 0

&me

r m

&0000,+0d

00 0 00 o

O N

' m 0 0 0 t -

rlmwot-

23
73977
0 n ) m o o o e l

F l d

O 0 0 0

I0,OdH

I
00000
***el

el

r,

bi

II
O

II
m

II
di
r(

109

J. P. Andriesse

Bibliography
ALEXANDROVA,L. N.; ARSHAVSKAY,
T. Th.; DORFMAN,E.M.;
LYUZIN,M. F.; YURLOVA,
O.V. 1968. H u m u s acids and their
organo-mineral derivates in soil. Proc. 7th Znt. Congr.
Soil Sci., yol. 3, p. 143-51. New York, Elsevier.
ANDRIESSE,
J. P. 1969a. A study of the environment and characteristics of tropical podzols in Sarawak (East Malaysia).
Geoderma, vol. 2, p. 201-27 (to appear in 1970).
. 1969b. The development of the podzol morphology in
the tropical lowlands of Sarawak. (In press.)

L
-

DEPT.
OF CIVILAVIATION
AND METEOROLOGICAL
SERVICES.
1961. Rainfall statistics of the British Borneo Territories
(period 1896-1957). Sarawak, Government Printer.
DOBRITSKAYA, Yu. I. 1962. Quick method for a short total
chemical analysis of soils. Soviet Soil Sci., p. 965-72.
FRIDLAND,
F.M. 1958. Podzolization and illimerization. Soviet
Soil Sci., p. 24-32.
IIAILE, N. S. 1954. The geology and mineral resources of the
Strap and Sadong valleys, west Sarawak. Geological Survey
.
Department, British Territories in Borneo.
JENNY, H.1941. Factors ofsoilformation. New York, McGrawHill. 281 p.
KIRK,
H.J. C. 1968. The igneous rocks of Sarawak and Sabah.
Geological Survey, Borneo Region, Malaysia. (Bull. no. 5.)
KLINGE, H. 1968. Report on tropical podzols. (First draft.)
Rome, FAO.
MARBUT, C. F. 1928. A scheme for soil classification. Proc.
1st Znt. Congr. Soil Sci., vol. 4, p. 1-31.
MCCALEB, B. Stanley. 1959. The genesis of the red-yellow
podsolic soils. Proc. Soil Sci. Soc. Amer., vol. 25, p. 164-8.
MOHR,Jul. C. E. 1944. The soils of equatorial regions. (Trans.
, Pendleton.) A n n Arbor, Edwards, 766 p.
-;
VAN BAREN, F.A. 1954. Tropical soils. The Hague,
van Hoeve. 498 p.
PICKERING,R.J. 1962. I. Some leaching experiments on three

'110

quartz-freesilicate rocks and their contribution to an understanding of laterization. Econ. Geol. (Lancaster, Pa.),
vol. 57, p. 1185-206.
PIMM, A. c. 1965. Serian Area, west Sarawak, Malaysia.
Geological Survey, Borneo Region, Malaysia. (Report 3.)
STOBBE, P. C.; WRICIIT,
J. R. 1959. Modern concepts of the
genesis of podzols. Proc. Soil Sci. Soc. Amer.. vol. 23,
p. 161-4.
TAN, K.H.; V A N SCHUYLENBORGII,J. 1959. O n the claseification and genesis of soils, derived from andesitic volcanic
material under a monsoon climate. Neth. J. agric. Sci.,
vol. 7, p. 1.21.
-.
1961. O n the classification and genesis of soils developed
over acid volcanic material under humid tropical conditions.
Neth. J. agric. Sci., vol. 9, p. 41-57.
THORP,
J. 1935. A provisional soil m a p of China with notes on
Chinese soils. Trans. 3rd Int. Congr. Soil Sci., vol: 1,
p. 275-6 (quoted from Mohr and van Baren).
-. , SMITH,Guy, D. 1949. Higher categories of soil classifications :Order, Suborder and Great Soil Groups. Soil Sci.,
vol. 67, p. 117-26.
UNITED STATES SOIL SURVEY. 1960. soil classification. A
comprehensive system. 7th approzimation. (Revised 1967.)
Soil Conservation Service, U.S. Dept. of Agriculture.
VILENSKY,
D.G. 1927. Concerning the principles of genetic
soil classification. Contributions to the study of the soils of
Ukraine, vol. 6, p. 129-51 (quoted from Jenny).
WILFORD,
G. E.1965. Penrissen Area, west Sarawak, Mahysia.
Geological Survey, Borneo Region, Malaysia. (Repurt 2.)
WOLFENDEN, E.B. 1965. Bau Mining District,west Sarawak,
Malaysia. Geological Survey, Borneo Region, Malaysia.
Buli. no. 7.)
-. , IIAILE, N. S. 1963. Sematan and Lundu Area, West
Sarawak. Geological Survey Department,British Territories
in Borneo. (Report 1.)

The differences between crusts of weathering aiid

soils developing on acid .andbasic rocks
in the humid tropics
V. M. Fridland
Institute of Geography, Academy of Sciences of the U.S.S.R.,
Moscow (U.S.S.R.)

T h e differences between crusts of weathering and soils

developing on the acid and basic rocks in the tropics
were described in the classical works by Lacroix and
Harrison as long ago as 1913 and 1933 respectively
and by a good m a n y others authors. As has been shown
by the investigations carried out in North Viet-Nam,
the crusts of weathering and soils formed on pure,
deeply metamorphosed limestones-marbles are rather
similar to those formed on the basic igneous rocks.
However, the crusts of weathering and soils developing
on weakly crystallized limestones and especially on
those containing appreciable amounts of admixtures,
possess essentially new properties bringing them closer
to the black soils developing on marls and called
margalitic by M o h r and van Baren (1954).The crusts
of weathering and soils formed on the basic volcanic
tuffs have, in their turn, very m u c h in c o m m o n with
the margalitic soils.
It is evident, therefore, that the character of the products of weathering can be vitally influenced not only
by the chemical composition but also by the physical
properties and crystalline-chemical peculiarities of
primary rocks.
While most generally considering this influence, it is
possible to affirm that the character of products of
weathering and, consequently, the soils is stipulated
by the ratio of disintegration rate of primary rocks
and minerals to that of leaching out the soluble products
of disintegration. The ferrallitic crusts of weathering
and soils are formed on acid rocks, solid igneous basic
rocks and on marbles whose weathering under humid
tropical conditions is characterized by prevalence of
the rate of leaching the substances out over that of
their release.
The above crusts of weathering and soils are generally
k n o w n to be distinguished by a strongly acidic reaction, unsaturation with bases and formation of both clay
minerals belonging to the kaolinite group and min-

erals-aluminium and iron oxides (gibbsite, boehmite,

limonite, goethite) which condition their fine texture,
almost complete absence of the silt fraction, stability
of aggregates, low cation exchange capacity and inadequate cation saturation, considerable anion-absorbing capacity.
These properties of ferrallitic soils should be sup-.
plemented by a great predominance of fulvoacids in the
organic matter. T h e ratio of carbon of the humic acids
to that of the fulvoacids in the red-yellow ferrallitic
soils formed on acid rocks (gneisses) and dark-red ferrallitic soils developed on the basic rocks (andesites)
in hilly and low-mountain regions is 0.2. This ratio
increases to 0.4-0.6 in the mountain humus-ferrallitic
soils of the m e d i u m high mountain zone and mountain
allitic-humus boils of the high mountain zone (foggy
forests).
Apparently, a decrease in temperature with altitude
slows d o w n the decomposition of organic remains thus
contributing to formation of more polymerized organic
substances.
It should be also pointed out that the humic acids in
these soils bave a slightly expressed aromatic ring and
are very similar to fulvoacids which is proved by high
thresholds of coagulation of the humic acids of the
ferrallitic soils in North Viet-Nam (30-40 meq. of
CaC12 per one litre of humate) and the impossibility
of the complete coagulation in certain boils (mountain
allitic-humus soils).
T h e relatively low optical density of humic acids
(Fig. I) proves their closeness to the fulvoacid.
All the considered properties are characteristic of
both the crusts of weathering and soils developed on
acid as well as solid basic rocks. However, there exist
some essential differences between them.
1. Specially presented as eupporting paper to review No. 3 by

J.

van

Schoylenhorgh.

111

V. M. Fridland

the entire thickness such crusts of weathering pohselih

the structure of a well-structured (granular-cloddy)

726
length of waves

665

574

533

496

465

niir

FIG. 1. Optic density of humic acids in :

A-dark-red ferrallitic soil on basalt;
B-mountain allitic-humussoil on gneiss;
C m o u n t a i n humus-ferralliticsoil on rhyolite.
T h e crusts of weathering of the acid rocks are distinctly subdivided into several horizons. T h e lower one
is notable for a considerable amount of primary minerals
though metamorphosed but still possessing certain
significant initial properties. T h e m e d i u m horizon
contains only the most stable primary minerals such as
quartz, ilmenite, etc., because other primary minerals
were already transformed into the minerals of intermediate stages of weathering and, first of all, into
hydromicas. In this horizon there are also some appreciable quantities of kaolinite, halloysite and mineralsiron and aluminium hydroxides. T h e upper horizon is
fully composed of the most stable primary minerals,
minerals of the kaolinite group and minerals-hydroxides. T h e whole crust of weathering is several and
frequently several dozen metres thick. It preserves
the structure of the primary rock throughout the
thickness except the upper part (the h u m u s horizon
of soil) homogenized by the root systems of plants
and essentially differing from the underlying rocks
affected by weathering.
The crusts of weathering of the basic rocks anp
marbles have no subdivision into horizons. They are
also from several to several dozen metres thick. Several
millimetres above the primary rock and throughout
%

112

masn greatly differing from that of the primary rock.

T h e soil-forming process going o n in the upper part
of the weathering crust does not essentially affect its
structure, and the soil, from the standpoint of its morphology, is almost identical with the crust of weathering.
The chemical and mineralogical composition of the
weathering crust of solid basic igneous rocks as well
as marbles is the same throughout the thickness which
makes it possible to assume that weathering of basic
rocks affects only a layer of several millimetres contrary to that of the acid rocks where weathering is a
phasic (stadial)process affecting considerable thickness.
E v e n under good drainage conditions the weathering
erusts of acid rocks have a mottled colouring (reddish,
yellow, bluish) caused b y the gleization and bydration processes and responsible for their widely k n o w n
n a m e mottled clay. The weathering crusts of the
basic rocks are affected by gleization only in special
geomorphological-hydrological situations.
W e shall consider below only those crusts of weathering and soils which are characteristic of a sufficiently
dissected and adequately drained topography. Their
comparison has revealed considerable differences in
their physical properties (Table 1).
Both the gneiss products of weathering and the redyellow ferrallitic soil formed on them have a considerably greater volume weight than that of the basalt
products of weathering and the dark-red ferrallitic
soil. T h e difference in the specific weights of these
rocks and soils is not so well pronounced which stipulated rather a great variation of the total porosity
value. The values of aeration porosity, i.e. the volume
of pores unfilled with moisture under a moisture content in the soils and rocks equal to the m a x i m u m field
moisture capacity, differ even more considerably (twoto threefold).
T h e above differences reach their m a x i m a in the crust
of weathering and decrease in the h u m u s horizons of
soils which makes it possible to assume that they arise
in the process of weathering crust formation and bcc o m e less distinct with the development of soil-formation process.
The results of long-term investigations on the moisture content in the ferrallitic crusts of weathering and
soils developed in North Viet-Nam on both the basic
and acid rocks (Fridland, 1964)have demonstrated that
most of the year it is equal (in the soil thickness d o w n
to a depthof 2.5-3 m), to the m a x i m u m field moisture
capacity and slightly decreases during the rest of the
year. Moisture content in the crust of weathering
underlying the soil profile below a depth of 2.5-3 m
equals the m a x i m u m field moisture capacity the year
round. Therefore, the degree of aeration of weathering
crusts and soils is dependent upon aeration porosity
and the data supplied in Table 1 give grounds to assume
that the ferrallitic soils (the weathering crusts espe-

Differences between crusts of weathering and soils

T h e wet sifting which preserves only water-stablestructural separates has confirmed the minimal differences
between the t w o groups of ferrallitic soils and has
revele'd the high stability of structure : only 1520 per cent, and in the upper horizons less than 10 per
cent of the investigated soil mass disintegrates into
particles less than 0.25 mm, whereas the bulk of the
soil mass is preserved as larger aggregates. T h e mountain
humus-ferrallitic soil and especially its upper part,
richest in humus, is characterized by the biggest aggregates. Smaller structural separates are typical for
the red-yellow ferrallitic soil. T h e smallest structural
separates (prevailing sizes from 0.5 m m to 2 mm)
are characteristic of the dark-red ferrallitic soil.
Thus, the weakly aerated red-yellow ferrallitic soils
developing o n acid rocks have in the h u m u s horizon
similar or even more stable structural separates than
the well aerated dark-red ferrallitic soils developing
o n the basic igneous rocks and marbles. Hence, it is
possible to conclude that the differences in physical
properties are stipulated not b y various degrees of
water stability but by diverse structure and arrangement of structural separates. Weathering of basic rocks
results in the formation of loosely connected macroporous structural separates with considerable spaces
between them.

cially) developed o n acid rocks are unsatisfactorily

aerated contrary to those developed o n the hasic ones.
Consequently the reduction regime and gleization process, responsible for the specific mineralogical and chemical features, are characteristic of the former, whereas
the latter are characterized b y the oxidation regime
conditioning their specific mineralogy and chemism.
Therefore, the destinations of the primary rocks
manifest themselves not only in the chemical and mineeralogical but also in the physical properties of the
weathering crusts and soils formed on t h e m and, first
of all, in the character of their structure.
T h e following investigation o n the structure of ferrallitic soil has been carried out to elucidate the essence
of these distinctions. W h e n viewed through a binocular
the structural separates of all ferrallitic soils have
appeared to possess uneven surfaces which testifies
to high cementing properties of the connecting cements
(Figs. 2(a) and 2(b)).
Structural analysis of the soils under investigation
has been m a d e b y employing the methods of dry and
wet sifting (Figs. 4(a) and 4(b)). T h e dry sifting fails to
reveal the difference between the t w o groups of ferrallitic soils as both of t h e m are characterized by rather
big structural separates (more than half of the total
mass of soil consists of separates larger than 2 mm).

TABLE1. Physical properties of gneiss (A) and basalt (B)crusts of weathering and red-yellow (A)and dark-red (B)ferrallitic
soils formed on them

Porosity (volume percentage)

Volume weight

Specific weight
Total

Aeration

Depth
cm)

'

0-10
10-20
20-30
30-40
40-50
50-60
60-70
70-80
80-90
90-100
100-110
110-120
120-130
130-140
140-150
240-250

0.96
1.20

0.94
0.89
0.86
0.86
0.83
0.81
0.85
0.83
0.84
0.79
0.78
0.78
0.82
0.83
0.83
0.83

2.73
2.73

2.54
2.53
2.59
2.59
2.50
2.50
2.59
2.59
2.59
2.58
2.58
2.58
2.58
2.58
2.58
2.58

64.9
56.1

63.0
64.9
66.0
66.8
66.8
67.0
67.2
67.5
69.4
69.8
69.8
68.6
68.6
68.6
68.6
68.6

28.1
14.1

23.2
27.2
28.0
30.0
33.2
32.0
28.7
30.5
31.8
35.8
37.6
33.6
29.3
30.0
25.6
28.2

1.17
-

1.22
1.22
1.26

1.24
1.28

2.78

2.80

2.77

2.80

2.79
2.80

58.0
56.5
56.0
55.0
-

55.5
54.3

15.7
11.7
11.6
8.7
-

10.7
8.5

'

113

V. Til.

Fridland

e . .

**

"

.e**<

(4

(b)

FIG.2. Structural separates of the h u m u s horizons of soils in North Viet-Nain :(a) dark-red ferrallitic soil in hasalt;
(b) red-yellow ferrallitic soil on gneiss.

FIG.3. Microphotographs of thin sections of structural separates of soils of North Viet-Nam

soil on basalt; (b) red-yellow ferrallitic soil o n gneiss.
,

114

(x 100) : (a) dark-red ferrallitic

._

Differences hetween crusts of weathering and soils

100

50

------

50

A, 0-15
B, 0-22
e, 0 - 5
D, 2 - 7

_--_
-- -

A, 15:30
B. 22-35
c, 5-ta
0, 7-15

50,

LA, 30-50

------

al

S, 35-60
C, 18-29
D. 15-45

O25

0.5

(a)

5 7 1 0 1 5

E
2,
20.25

0.5

Logarithmsof fractions (diameters,m m )

3
b)

FIG. 4. Results of the structural analysis of soils of North Vietnam (percentage of fractions smaller than): (a) dry sifting;
(b) wet sifting. Letters and figures designate soils and depths in centimetres respectivelv :
A-dark-red ferraiiitic soil on basalt:
I3-dark-red ferraliitic soil on marble;

C-red-yellow ferrallitic soil on gneiss;

D-mountain humus-ferrallitic soil on rhyolite.

Weathering of acid rocks leads to the formation of

structural separates with narrow inner pores and close
packing of separates repeating that of the primary
rocks (Figs. 3(a) and 3(b)).
This difference is, probably, stipulated by more
intensiveremoval of substances in the course of weathering of basic rather than acid rocks.
T h e character of structural separates is significantly
influenced by the chemical composition of substances
cementing the structural separates. A special investigation employing the method proposed by AntipovKarataev, Kellerman and K h a n (1948) has been
carried out to elucidate this problem. Water-stable
structural separates from 3 m m to 5 m m in size were
selected from the fioils being investigated, placed o n
glass discs and subjected to successive treatments

with various soIvcnts. In view of the possible destructive influence o n the aggregates, if kept moistened for
a long time, the control structural separates from each
sample were continuously moistened with distilled
water.
Fresh portions of reagents were carefully pipetted
twice a day and the waste reagents were also removed
by a pipette. T h e tretment of samples was conducted
in consecutive order by the following reagents :
1. Alkaline buffer mixture of decinormal sodium hydroxide and sodium oxalate at a ratio of 1 :4 removing the loosely connected and free organic matter.
2. Tamm's reagent removing the slightly crystallized
sesquioxides.
3. Normal sulphuric acid extracting the stronger crystallized sesquioxides.

115

V. M. Fridland

4. The alkaline buffer mixture (second treatment)

aimed at removing the organic matter which was
fixed by the sesquioxides.

The

results of the five-month experiment have shown

that :
All the ferralitic soils bave rather stable structural
separates (more stable than those of chernozems and
krasnozems);
Structural separates of the red-yellow and mountain
h u m u s ferrallitic soils are less stable than those of
the dark-red ferrallitic ones;
Iron oxides act as cementing substances w h e n the structure of red-yellow and dark-red ferrallitic soils is
being formed;
A significant role in the structure formation of m o u n tain humus-ferrallitic soils is played by organic
substances.

It will b e very interesting to study by this method also

deep horizons of soils and crusts of weathering. Ilowever these data also prove that differences in physical
properties between the weathering crusts and soils
developed on acid and basic rocks is connected not
with thc stability of structural aggregate, but with
its construction.
It should be pointed out in conclusion that the
character of weathering rocks essentially affects not
only the chemical and mineralogical but also the
physical properties of products of weathering. The
latter, in their turn, influence the processes going on
in the crusts of weathering and soils and significantly
determine the further chemical and mineralogical
transformations.

Bibliography
ANTIPOV-KARATAEV, I, N.;KELLERMAN, V. V.; KHAN, D.V.
1948. About soil aggregate and methods of its study, p. 42-3.
Moscow, Publ. House Acad. of Sciences. (In Russian.)
FRIDLAND, V. M. 1964. Soils and crusts of weathering of humid
tropics, p. 208-27. Moscow, Publ. House Nauka. (In
Russian.)

116

HARRISON, J. . 1933. The katamorphism of igneous rocks

under humid tropical conditions. IIarpenderi (U.K.).
LACROIX,
A. 1913. Les latrites de Guine. Paris, Nouvelles
Arch. Museum histoire natur., ser. 5, t. 5.
MOIIR,E. C. J.; VAN BAREN, F.A. 1954. Tropical soils. London, Interscience Publishers, 498 p., tabl., fig., pl.

Genesis and evolution of red and black

basaltic soils in Cambodia'

Tan Boun Suy

Agronomist, Dean of the Faculty of Tropical Crops at the
Royal University of Kompong-Cham (Cambodia)

INTRODUCTION
T h e Faculty of Tropical Crops of the Royal University of K o m p o n g - C h a m was established with the aim
of forming agronomists, trained to problem solving
in the agricultural area of north-east Cambodia, with
emphasis on the area of the K o m p o n g - C h a m province.
Particular attention would be given to the studies of
crops grown on red and black soils of basaltic origin.
In this respect, the need for a more thorough investigation pertaining to the formation and weathering
of these soils could not be overemphasized.

PARENT MATERIALS
The basaltic block of the K o m p o n g - C h a m province is
subsequent to the former alluvion. This assertion
sprang from the drilling of wells at the University of
K o m p o n g - C h a m in which sandy soil has been brought
out from a depth of 21 m, under a layer of basaltic
material. Its age is assumed to be from the late Quaternary era. Thickness of the basaltic layer ranges
from 21 m, at the University in K o m p o n g - C h a m , to
125 m in Chamcar-Krauch, 30 k m away).
Structure of the materials shows great variabilities,
due to different states of cooling. The rocks could be
found in the compact, vesicular, or microlitic form,
lying as continuous or fractured tables or as shells.
Around the standing crater of Phnom-Pros, distant
by 7 k m from the city of K o m p o n g - C h a m , one can
observe scoria cones and belt of volcanic ash.
D u e to different states of decomposition, nodular
rocks and gravels of various sizes can also be found.
Current minerals of basalt of K o m p o n g - C h a m are
found as follows :
Alkaline plagioclase feldspath: This is a mixture of
albite (Na,O; Alzo,, 6 SiO,) and of anorthite (Cao,

Al,O,, 2 SiO,) with a predominance of the latter

(more than 60 per cent).
This is a metasilicate with approximate c o m position Ca M g (SiO,),.
Olivine : Magnesium silicate (2Mg0, SiO,) in mixture
with iron silicate (2 Feo, SiO,) at various molecular
proportions.
Magnetite :Fe304or (Feo,Fe,O,).
Analyses which were carried out by Raoult, cited
by Henry (1931) and those of Baleine: reported by
Deresmaux (196!J), result comparably. Sampling by
Raoult was m a d e in C h u p and that of Baleine w a s
m a d e at the university (see Table 1).
Augite:

TABLE1. Percentage composition of basait

Analysis of

Lost on ignition
SiO, (free)
Silicate (total)

Fe0
Fe,O,
ALO,
Tio,
Ca0
MgO
MnO
PZ06

Na,O

K,O
LiO,
TOTAL
:

F r o m Chup
by M. Raoult

F r o m the University
by O. Baleine

0.69

0.77
0.11
49.52
8.25
3.70
13.66
1.89
8.48
8.46
0.14
0.33
2.27
1.01
0.00
98.48

46.98
8.46
3.01
13.67
1.98
10.38
9.73
0.17
0.51
3.18
1.59
0.00
100.35

1. Specially preaented as supporting paper to review No. 3 b y J. van

Schuylenborgh.
2. O. Baleine, Facult Polytechnique de Mons (Belgium).

117

Tan Doun Suy

Transformation of basaltic mineral rocks is as follows :

WEATHERING OF THE PARENT

MATERIALS
I

iddingsites + iron hydroxide

(Phylitic mineral, &th a reflexion at
15.6
and goethite are components
of iddinesite.)
muscovite (white potassium mica)
Anorthite +
+ epidotite (hydrated silicate of Ca, Al,
Fe)
+ calcite (Caco3)
clay
+ bauxite
Augite
-+
chlorite (hydrated silicate of Al, Fe, Mg)
Olivine

Degradative agents of parent rock are of physical and

chemical nature.

-+

A,

PHYSICAL A G E N T S

Physical degradation starts during the process of cooling

of the lava which o n contraction splits into blocks.
Degradation continues by the alternating effects of
humidity and dryness,combined with the great variation
of temnerature.
W e t during the rainy season and then dried out during
the dry season during which temperature fluctuates
over a wide range, the rocks undergo continuous stress
during the process of expansion which in turn causes
the rocks to crack.
C H EM I C A L A GEN T S

Rain-water lodged in cracks is considered as the main

chemical weathering agent. Water is a very w e a k electrolyte, dissociated into H+ and OH-. T h e dissociation of
this w e a k electrolyte is however enhanced by a rising
temperature; at 340 C it is four times as active as at

I-+

Magnetite+

epidotite + calcite + limonite

+ quartz
oligiste (red hematite) Fe203
+ limonite (brown hematite) Fe, (OH),.

The mineral rock splits into Concentric shells and loses

its greyish-black coloration; its hardness decreases.
Fragments of rocks are rounded off and concentric
shells are separated by larger and larger fissures filled
with secondary oxidized materials; each independent
sheet becomes more altered than the layers underneath.

160 C (K16
= 0.63 x lO-l4; K, = 2.51 X
Demolon (1960), stated that the proton in water has

Under conditions of good drainage

a potential kinetic energy w h e n lodged in the soil, as

it causes the hydrolytic reaction to go o n indefinitely.
This process of decomposition would reach its maximum intensity in a tropical climate. On the other hand,
free carbonic acid in water contributes to increase the
dissociation constant of the m e d i u m 3 to 5 fold, which
then enhances its hydrolytic activity. Similar action is
intestigated by sulphuric acid formed by oxidation of
sulphur-containing organic compounds and more generally by all intermediate acidic compounds stemming
from oxidation of organic materials.

In the case of basalts of Kivu (Pecrot, 1960), fresh

rocks are composed of augites and of plagioclases;
olivines have been decomposed to iddingsites and iron
hydroxide (see Table 2 below).
T h e first layer of decomposition is some millimetres
thick : the plagioclases have already disappeared
whereas augite remains unaltered. The next process of
decomposition gives birth to a mixture of halloysite
with some parts of kaolinite. Further on, halloysite
turns to kaolinite and then to gibbsite, under very
intense drainage.

TABLE2. Weathering of basalts of Kivu

Firnt period clay

First layer decomposition

Primary minerals

Soil clay

Good drainage

Plagioclases
Augites
Olivines (decomposed to iddingsites and
iron hydroxyde)

Poor drainage
Plagioclases
Augites
Olivines (not decomposed)

Augite

fiel

IIalloysite
+ kaolinite
+ augite (in the process
of decomposition)

Kaolinite
+ gibbsite

Montmoriilonite
+ gibbsite
-I- halloysite

Montmorillonite
+ halloysite

Plagioclases
Augites
Montrnorillonite

Genesis and evolution of red and black basaltic soils in Camhodia

TABLE3. Chemical analysis of'red

Analyses of samples from XIondulkiri, performed ,y

Cruys (1964), have resulted in very high contents of
Alzo, and Fe,O, and a low content in SiO, (Table 4).
High values obtained for the loss o n ignition arc
assumed to he caused by the presence of trihydrate
gibbsite (laterited gibbsite) in the bauxite.

soil, by Baleine (samples

from the University)

Analysis

Loss on ignition
SiO, (free)

sio,(total)
Fe0
Fe203
A1203

.Tiol
Ca0

MgO
MnO
p20,

Na,O
K,O
Li02
CO,

Basalts

Iled soil

0.77
0.11
49.52
8.25
3.70
13.66
1.89
8.48
8.46
0.14
0.33
2.27
1.01

10.38
17.07
42.38
0.81
17.37
21.21
3.77
0.27
0.39
0.23
0.90
0.06
0.18
0.00

0.00
0.00

TOTAL

Under conditions of poor drainage

0.00
-

98.48

100.05

Mineralogical identific hasation been

performed

with an X-ray diffractometer.

In Cambodia, samples extracted at the university

by Deresmaux (1968), have been analysed in Belgium

by Baleine and R0bazynski.l (See Table 3.)
T h e moderate leaching state of alkaline and alkalinecarth metals and the presence of a large quantity of
free SiO, have proved that red soils are of relatively
recent origin.
Further proof has been provided by F. Robazynski
in his diKrfractometerstudies :the kaolinite analysed

SiO,/Al,O, = 27.30/21.20 = 1.28

is of recent formation.
This is mixed with minute traces of halloysite,
fated to disappear in a short lapse of time.
Henry (1931), stated that red soil underwent
the following evolution : silica, alkaline and alkalinecarth metals are progressively exported from the solid
mass. There remains a mixture of hydratcs of F e and
Al forming laterite completely free of soluble bases.

TABLE4. Analyses carried out

17 $1 (lateritic basalt)
17 N (lateritic red soil)
18 A (lateritic gravel)
18 II (lateritic basalt)

SiO,/Al,O,

= 28.02/12.07 = 2.32

is found to be higher than that of red soils (1.28).

A lesser extent of leaching of alkaline and alkalineearth metals gives rise to the formation of m o n t m o rillonite. (This has been found by X-ray diffraction
studies.)
T h e total content of iron is relatively high, although
the reduced form of this metallic element is scarce.
High concentration of the oxidized statc of iron is
found in pisolithic concretions.

1. F. Robazynski, Facult do Polytechnique de Il&,

Belgium.

on samples from AZondulkiri, by the Bureau de Recherches Gologiques et Minires, Paris

04,Si02

Sample

'

T h e conditions of poor drainage o n basalt of Kivu have

been studied by Precrot (1960)(see Table 2).
Relatively intact rocks are composed of unaltered
olivine, besides plagioclases and augite.
A mineral of the type montmorillonite appears with
the plagioclases and augite in the first layer of alteration. T h e former enriches in the subsequent layers,
whereas the latter t w o diminish. At last, a mixture of
montmorillonite and halloysite is formed.
Chemical analyses have shown that in the case of
poor drainage, silica, alkaline and alkaline-earth metals
are retained, whereas high silica content and high p H
contribute to the montmorillonite formation.
In K o m p o n g - C h a m , wherever there exists a nonpermeable underground, black basaltic soil (named
basaltic regurs by Crocker (1962), forms a barrier
completely isolating masses of red soils. T h e profile
shown in Figure 1 gives a clear aspect of such
morphology (Henry, 1931).
Samples of soil extracted in Mien, distant by s o m e
20 km from the University and analysed by previously
n a m e d workers in the field, have yielded the data
shown in Table 5. .
T h e ratio,

46.60
30.80
43.40
50.00

yoSioz

(total)

(quarzt)

1.60
4.30
1.70
1.10

0.11

0.20
0.03

Fe30z

Ti02

y,
Lops on
ignition

20.50
33.00
23.70
17.20

3.40
5.20
3.20
2.80

27.80
24.60
27.00
29.00

119

Tan Boun Suy

Red soil
Basalt or dominant conglomerate
Black or brown-redsoil
Black m
il with basalt or conglornasts

FIG.1. Profile of Kompong-Thm-Chamcar-Krauch.

TABLE6. Mean values of analytical characters of red soils from different regions
Exrhangeable bases in meq.
Site

%d.s.

PII

Number
of '
P
blocks
p.p.m.
studied

,
K

Na

Ca

Yg

1120

KCI. N

KCI
N/SO

253

0.23

0.04

1.75

0.32

2.35

8.4

25.2

4.9

4.4

4.4

17

1122

0.42

0.09

4.61

0.66

5.78

10.7

52.2

5.0

4.9

4.8

Ch. Loeu

1003

1.12

0.03

7.57

2.05

10.78

13.2

81.2

5.8

5.7

5.7

Tapao

2 009

0.64

0.11

6.86

1.38

9.00

12.7

70.3

5.8

5.4

5.5

Chup

17

296

0.25

0.05

2.60

0.68

3.58

7.9

44.2

4.8

4.6

4.6

330

0.16

0.06

0.32

0.36

0.89

10.4

8.6

4.2

4.0

3.9

Prek Chhlong

210

0.22

0.08

1.07

0.72

2.10

8.4

25.0

4.3

4.3

- 4.2

Chalang

456

0.22

0.04

1.61

0.67

2.54

9.1

27.6

4.7

4.5

4.5

'14

166

0.22

0.04

1.35

0.59

2.18

8.4

25.5

4.8

4.4

4.3

Prekkak

Ch. Andong

Mimot
and

Snoul

120

Genesis and evolution of red and black basaltic soils in Cambodia

TABLE5. Chemical contents of basaltic regurs (by Baleine)

Compact besalt

Loss on ignition

SiO, (free)
SO,(total)
Fe0

Fe#,
A1208

Tio2
Ca

Mg0
MnO
p*os
.

Na,O
K20
Li02
CO,
TOTAL

It is

Basaltic regura

4.34
1.28
47.66
5.15
6.10
14.24
1.86
10.47
7.13
0.09
0.33
1.77
0.74
0.00
1.72

9.79
17.09
45.11
1.03
24.81
12.07
2.84
0.96
1.68
0.33
0.19
0.53
0.36
0.00
0.00

99.88

99.70

worth mentioning here the work undertaken

by Correns (1962), on weathering of basalts in Iceland,

in which he found that in contact with hot and acidic
springs the final decomposition product is always a
definitely crystallized kaolinite. Whenever the acidity
of the environment decreases, montmorillonite is formed and then degraded into kaolinite.

C H A R A C T E R S OF H E D
BASALTIC SOILS
F r o m the pedological standpoint, red basaltic soils are
classified, after Aubert and Ducbaufour (1960)into ferrallitic soils.
Such a substratum is mainly planted with Hevea
trees in Cambodia (about 60,000 ha). It can be noted
that our country is in the lead a m o n g the worlds producers of natural rubber, as far as the yield per hectare
is concerned. This position is due partly to the favourable
natural environment and in particular to the red
basaltic soils.

PHYSICAL CHARACTERS

Depth

With the exception of the relatively recent soil at the

university, which does not exceed 1 m in thickness,
the depth is great (Manil, 1959). A depth of 20 m
or more has been found in Krabao (Henry, 1931) and
Sor T h a y Seng (1966), has even reported in certain
areas a depth of 60 m.

The depth of the red soil is crucial for a good penetration of the tap-root of Hevea which, during the dry
season, must reach the underground water.
Texture

R a m b e a u x and Danjard (1963), have reported a clay

content of red soil ranging from 62 to 85 per cent.
T h e component clay plus loam is of a high value
(82 to 92 per cent) and appears to be relatively constant.
T h e sand component remains low in value and ranges
from 2.7 to 10.6 per cent.

Structure

In spite of the high content in clay, red basaltic soils

have a grumous structure. This has been explained by
the richness in iron hydrate which is an electro-positive
colloid; in contact with the kaolinite of opposite sign,
floculation occurs (Demolon, 1960).
This particular structure favours an easy infiltration of rain-water and preserves at the same time the
soil from erosion. On the other hand, it promotes an
adequate aeration of the soil, thus providing a good
protection against fungal root diseases.

P H Y S I C O C II E n I C A L C H A R A C T E R S

Table 6 shows data obtained from soil of different areas

grown with Hevea ( R a m b e a u x and Danjard, 1963).
T h e optimum value of pH is around 4.9 for a good
growth of Hevea. This is a condition which is generally
found in Cambodia.
T h e pH of more recent soils is higher than the optim u m limit. Correction can however be achieved b y
the use of a m m o n i u m sulphate fertilizers (Ninane, 1969)
such as in the case of the estate in Peam-Cheang
where the lowering of p H in the soil has proved to be
beneficial for the growth of the rubber trees.
This is the first time that response of the red soils
to fertilizer application has been noted in Cambodia.

C O V E R - P L A N T S A N D nIULCHING

Experiments undertaken by Niv Narin (1969), at.the

university have shown that covering of soil is very
important for the growth of young Hevea trees, at the
time when the canopy has not fully developed. T h e
following cover plants have been tried for such purposes ; Tithonia diversifolia, Trixacum P a x u m , Illimosa
invisa, Eupatorium odoratum, and graminacea, M i m o s a
inoisa appears to yield the best effect.
Mulching with c u t d o w n cover plants, (luring the
dry season, has resulted in a better growth of young
trees (Ninane, 1969).

121

Tan Boun Suy .

. ... . .

. .

ing into the ground. This practice resulted in a decreabed

out-flow of various springs (e.g. the springs in T u k Chha).

CONCLUSION
Iled basaltic soils grown with Hevea trees in Cambodia
are in the main well protected against weathering.
W h e n subjected to yearly crop planting, they undergo
a rapid degradation. Silica and exchangeable baies are
leached. There would remain a c o m p o u n d of iron,and
aluminum hydrates in the process of laterization.
In the case of soils in hiondulkiri, the practice of
'rays' by the hill people has been detrimental to the
soils. E v e n the forests which formerly covered the
area cannot he reconstituted.
Deorestation would cause a lowering of the water
table. Rainfall runs offsuperficially instead of infiltrat-

A C K N O w L E D GEM'Ei3~
T h e author thanks, &Ir. Deresmaux for permission to
use his analytical results on rock samples. Thanks are
also due to Messrs. Resing, Langlois and Chai Kim C h u n
for their m a n y helpful suggestions during the preparation of this paper.

Appendix
Meteorological data from the Institut des Recherches sur le Caoutchouc au Cambodge (IRCC)is given in the follow'ingfive
IRCC station of the Royal University is located at a distance of 10 km, on the left bank of the
Mekong river.

figures (Ninane, 1967). The

100Maxima 1966

36

90

Mean 1961-66
.

.......

80

'

70

60

2401

Minima 1966

';

F ' M ' A ' M ' J ' J ' A ' S ' O ' N ' D '

1966
...........Average 1961 66

Average monthly temperatures.

122

'
.

>

Relative humidities.

Genesis and evolution of red and black basaltic soils in Cambodia

mrn

mrn

400.

zoo0

300

1500

!
100

10.0

90

9.0

8.0
1 O00

200

7.0

500

1 O0

6.0

Y
J

____

1966
Cumulative 1966
.......... Average 1959.66

I
A

Daily average insolation (in hours) and relative insolation.

Rainfall in 1966 and average rainfall during 1959-66.

k
u
.
E
c
._

Potential evapotranspiration.

Bibliography
CORRENS,W. 1962. Observations sur la formation et la transformation de minraux argileux lors de la dcomposition
des basaltes. Gense et synthse des argiles.. Paris, Coll.
Int. CNRS.
CROCKER,C. D. 1962. Soils of Cambodia.
CRUYS. 1964. Mission de bauxite. Paris, BRGM. (Private
communication.)
DEMOLON, A. 1960. Dynamique du sol, p. 29, 136.
DERECMAUX.
1969. (Private communication.)
DUCHAUFOUR, P. 1960. Prcis de pdologie. Paris, Masson.
438 p.
HENRY,
Y. 1931. Terres ro'uges el terres noires basaltiques
d'Indochine, p. 157, 159, 40-41.
MANIL, P. 1959. Cours l'Institut agronomique de l'Etat
Gembloux (Belgique).
NINANE, F. 1969. Exprience de fumure dans les terrains peu
acides. Commission technique 484, p. 9-10,41. Institut des
Recherches sur le caoutchouc au Cambodge (IRCC).

-.1967.
Climatologie de la station IRCC en 1966.
1969. Rsultats d'expriences de mulch. Commission

-.

technique 483. Institut des Recherches sur le caoutchouc au

Cambodge (IRCC).'

NIVNARIN.
1969. Les plantes de couverture et quelques donnes
sur leur relation avec la croissance de I'IIva Brasiliensis.
Mmoire de fin d'tudes, Faculty of Tropical Crops, Royal
University of Kompong-Cham.
PECROT,A.; DELVIGNE,J. ; GASTUCHE,M. C.; VIELVOYE, L.;
FRIPIAT,J. J. 1960. L'altration des roches el la formation
des sols QU Kivu. Publ. INEAC (Srie Sc. no. 86).
RAMBEAUX, J.; DANJARD,
J. C. 1963. Terre rouge basaltique
et nutrition de l'hva dans les conditions cologiques du
Cambodge, p. 9-10,41. Institut des Recherches sur le caoutchouc au Cambodge (IRCC).(Opuscule technique no. 2/63.)
SOR THAYSENG.1966. Les sols du Cambodge rt les possibilirs
de leurs utilisations, p. 18.

123

Some aspects of lateritic soil formation in the

Dahican-Alayao area, Camarines Norte Province,
Philippines
E. V.

Tamesis and D. C. Salita

Department of Geology and Geography, University of the
Philippines (Philippines)

INTRODUCTION
The occurrence of deep lateritic soils is a c o m m o n
feature of m a n y tropical forested areas in the Philippines. They are found in different climatic regions
(defined by rainfall frequency and distribution), topographic situations, and underlying rock types. While
m u c h is k n o w n about the morphological and chemical
features of agriculturally arable lands in m a n y parts
of the country, there is a dearth of observations on this
important group of soils-which support vast tracts
of commercial timber forests.
T h e following report presents the results of preliminary pedological studies conducted on a typical soil
profile developed under a tropical rain forest in a
south-eastern province in the island of Luzon,
Philippines. It is part of a continuing study on the
weathering characteristics and soil-forming features
of different rock types as influenced by relief, climate
and other edaphic factors in similarly vegetated areas.

GEOGRAPHIC LOCATION
The Dahican-Alayao area is located in the northern
part of Camarines Norte Province, about 45 km northeast of the capital town of Daet, and some 396 k m
south-east of Manila (1220 30-1220 37E.; 140,17140 20 N.). The area covered in this study is bounded
on the west by the Alayao River, on the east by the
Tigbi River, and on the south by a jeep trail that links
the sitios of Banogbog and Alayao.

CLIMATE
T h e province of Camarines Norte falls within the Etype of climatic region (Huke, 1963, p. 44-7) where
there is no dry season, but a pronounced winter rain-

fall. The annual precipitation is about 150.5 in. The

heaviest rainfall generally occurs during the months of
November, December and January, and the least in
M a y and June. Atmospheric temperatures are likewise
high. The mean annual temperature is about 27.2OC
with the m e a n m a x i m u m and m i n i m u m temperatures
obtaining during M a y and January respectively.

RELIEF A N D VEGETATION
T h e topography of the Dahican-Alayao area is generally
mountainous and rugged, with numerous gulleys and
ravines cut into slopes of the predominantly north-west
trending ridges. Because of the thick canopy of the
tropical forest, the area is deceptively flat and plateaulike when observed from the air, but the ground is
deeply dissected.in m a n y places. There are a few isolated small plains in the northern part, some of which are
swampy. The average relief is approximately 180 m.
Several big rivers drain the area to the north. A m o n g
these are Alayao River, Alasanan River and Tigbi
River. These streams flow through youthful valleys
near their source, but meander on reaching the low
coastal lands to the north where they empty into
Alayao B a y and Dahican Bay. Most have courses that
are apparently structurally controlled.
T h e vegetation of the interstream area between
Alayao River and Tigbi River, including the m o u n tainous country farther south consists of primary and
second growth forests, a f e w open grassland covered
with cogon and related species of grass and shrubs.
Mangrove as well as other swamp-thriving plants are
c o m m o n along the coastal fringes and partially submerged mouths of the larger streams. The tropical
forest which is distinctly dipterocarp in composition
1. Speciaiiy presentrd as oupporting paper to review No. 3 by J. van
Schuylenborgh.

125

E.V. Tamesis and D.C. Salita

occupy the steep and h o a d slopes.1 In places where

logging operations have ceased for some time, or where
a shifting type of agricultural activity locally k n o w n
as kaingin has been practised, a second growth forest
has usually taken root.

TABLE1. hforphological characteristics of soil profile

1Iorizon

Depth

Description

0.30

cm

SOIL-FORMING R O C K S
T h e principal soil-forming rocks of the DahicanAlayao area belong to a sequence of engeosynclinal
sediments consisting of greywacke, arkosic wackes,
limy shale, chert and interbedded spilitic basalts of
Cretaceous-Paleogene age (?), and a m u c h later group
of rocks (Eocene to early Miocene) composed of COnglOmerates, arkose, carbonaceous shale, limestone and
associated andesite flows and pyroclastics (De Guzman,
1966; Gervasio, 1966). Both groups of rocks underlie
most of the hilly terrain in the northern part of C a m a rines Norte Province along a depositional strike that
trends roughly north-west to south-west.
A s in m a n y areas of tropical weathering, relatively
fresh outcrops are difficult to find. Information however, o n specific lithologies are generally available in
deep stream banks, and in s o m e recent road cuts m a d e
during logging operations in the area. Of the different
rock types just mentioned, the +volcanics, both of
basaltic and andesitic composition, constitute the
bulk of the soil formers.
In a recent soils m a p published by the Bureau of Soils
(P. Lucas et al., 1966) the soils of the area studied
were m a p p e d as part of the Alaminos clay which is
described as a reddish-brown to brick-red compact
clay which becomes sticky w h e n wet. This soil description is applicable to practically most of the localities
surveyed regardless of the underlying rock type. It
has been observed however, that deep red, lateritic soils
are more prominent in places where the bedrock is
basaltic in composition. Similarly,very sticky, yellowish,
brown soils are c o m m o n where the underlying parent
materials are shales and the coarser clastic members of
the younger rock units.

M O R P H O L O GI C A L
A N D C H E M I C A L FEATURES
In order to obtain s o m e insight into the weathering
characteristics and soil-forming processes involved in
the development of the lateritic soils in the area, a
detailed study of the morphology and chemical c o m position of typical soil profile was undertaken. T h e
profile section selected is about 4 km south of the sawmill site of the Dahican L u m b e r C o m p a n y at a n approxi1. Lateritic-following
!.

Mohrs (1954,p. 354) definition-as terms indicating

that the weathering tends to assist the accum,ulation of iron and
alumina, and the leaching of silica nnd basen.

0126

30-100 crn

Pale, reddish brown (1OR 5/4) clay

loam,friable and granular when dry,
sticky when wet. Numerous aggregates, colour somewhat mottled.

Light to yellowish brown (5YR5/6

10YR 5/4) loam, friable and

to

granular under dry conditions, but

heavier and more compact than
overlying horizon, and very sticky
when wet. Less mottled than
horizon 1. Upper boundary gradational.
3

100-300ern

Partly weathered bedrock, moderate yellowish brown (10YR 5/4)

clayey and very sticky when wet
with white specks of weathered
ieldspar phenocrysts.

mate elevation of 450 ft., and on a spilitic basalt parent

material with the following petrographic description:
Megascopic : Dark-greenish grey (56 4/1)with porphyritic
textures, altered largely to chlorites and with
patches of epidotes.
Microscopic: Porphyritic texture, consisting of phenocrysts,
andesitic and labradorite zone to calcic interiors,
chloritized and fresh augites or pigeonite in a
pilotaxitic matrix of albite-oligoclase, labradorite microlites, some augite, epidote, magnetite
and ilmenite. A few plagioclase phenocrysts
altered to albite-oligoclase.Alteration of rock is
chiefly chloritic, with feldspars altered to clay.
Epidotization is evident in local patches.

On the basis of changes in colour and texture, three

horizons (in Table 1) designated as 1, 2 and 3 from
top to bottom are recognized.
A total of chemical analysis of representative bulk
samples from the different horizons was undertaken to
determine.the quantitative distribution of the oxides.
Because most of the samples tended to form lumpy
masies on drying, they were initially disaggregated
and quartered to 100 g of material before reducing
t h e m to a fine powder as required by the analytical
technique. T h e pH values of soil from each horizon were
determined using a B e c k m a n p H meter. Additional
measurements of the pH were undertaken of another
soil profile having the same consistency, colour and
textures .as the one studied in detail for comparison
purposes. (See Table 2.)

Some aspects of lateritic soil formation

TABLE2. Chemical analysis of

60il profile from decomposed

basalt, Alayao, Camarines Norte, Philippines

r .

Snil horizon:
Constituenti

'

"

'

'

SiOl
ni203

Ca0

MgO
Na,O
K,O
Tio,
Loss on ignition

TOTAL(yoby

. I
,

1 (0-30em)

2 (30-100cm)

48.81
13.56
22.25
0.52
1.19
1.21
0.23
0.90
11.41

53.00
7.36
22.98
0.54
1.79
0.08
0.24
0.70
12.58

wt.) 100.08

p H values

5.2

3 (100-300,cm)

55.23
8.47
22.18
0.56
1.35
1.75
0.78
0.82
9.03

99.27

100.17

4.95

DISCUSSION

Although n o chemical analyses of the unweathered

parent material was made, the percentages of the oxides
from the different horizons reflect the intense leaching
process that has taken place in the area. Depletion
of bases is s h o w n by the fairly low percentages Cao,
MgO, NA,O, K,O, compared to published chemical
analysis of fresh rocks of similar mineralogical composition. Slight but progressive increase with depth are
exhibited by Cao, K,O, and possibly Na,O, but the
latter's abnormally low percentage in the second horizon is not easily explained. T h e gradational character
of the contacts between the three horizons, in addition
to the tight, compact nature of the intermediate horizons, seems to preclude selective leaching. On the other
hand, the possibility of analytical error is not beiog
discounted. No special variations or trend are shown
by both Tio, and MgO.
T h e percentages of the more stable components of
the soils as revealed by the analysis is more consistent
with their k n o w n chemical behaviour, and explains to
a certain extent some of the physical and chemical pro-

perties of the horizons. Silica progressively increases

with depth and suggests slight desilication in the upper

horizons. However, its over-allpercentage is fairly close
to fresh rocks having comparable mineralogical c o m positions (Johannssen, 1937; Mohr, 1944). T h e high
concentration of Fe,O, in the uppermost horizon in
relation to horizons 2 and 3 is the result of its relative
enrichment at the expense of the bases. Alumina, which
is another stable component, is more or less uniformly
distributed through all horizons. Like SiO,, the amount
of A1,0, is slightly higher than what is given in published analyses of fresh basaltic rocks and is therefore
believed to represent its concentration by retention.
Silica-alumina ratios of the soil samples from all three
horizons are greater than 2, which is Joachim's (1935)
terminology could be regarded as a n immaturc lateritic
soil.
T h e soils found in Dahican and neighbouring places
are generally acidic. T h e p H values range from 4.9 to 5.5,
and in both profiles there is a slight increase in acidity
with depth. T h e acidity of the soil is as expected of most
humid soils, and m a y be attributed to the combined
effect of high precipitation and luxuriant vegetation
(Krauskopf, 1967). Abundant rainfall in low latitudes
promote the development of tropical forests whose
decomposition products contribute to high acidity of
soil water. In turn this acid water can effectively leach
the bases from the upper horizons and m a k e it give a n
acidic reaction.
T h e significance of the slight decrease in p H with
depth in both profiles is not known. Elsewhere similar
occurrences have been attributed by M o h r (1954,
p. 390-1)to the presence of bases liberated from decaying plants to the topsoil, or the result of the upward
migration of alkali ions during the dry season. However,
because of the non-seasonal character of the climate in
the Dahican-Alayao area, it does not seem likely that
the latter interpretation satisfactorily explains the
observed geochemical feature. T h e soil in the arca is
generally moist and does not become dry for any significant length of time. It is also possible that s o m e of
the leached bases are actually adsorbed by the clay
minerals formed from the breakdown of aluminium
silicates in the upper horizons and is therefore responsible for the blightly higher pH at these levels.

Bibliography
DE GUZMAN,K.A. 1966. Uranium mineralization in the Paracale mining district, Camarines Norte. Bulletin of the
Institute of Filipino Geologists,Munila, vol. 1, no. 1, p. 1-13.
GERVASIO.
F. G. 1966. A study of tectonics of the Philippine
Archipelago. The Philippine geologist, voi. 20, no. 2,
p. 51-75.
IIUKE, R. E. 1963. Shadows on the land, un economic geography
of the Philippines, p. 44-50. Manila, Bookmark.

JOACHIM,A. W.R. 1935. Studies on Ceylon soil. II. General

characteristics of Ceylon soils, some typical soil groups of
the Island, and a tentative scheme of classification. Trop.
Agriculturist, Ceylon, vol. 84,'p. 254-74.
JOHANNSEN, A. 1937. A descriptia7e petrography of the igneous
rocks, vol. III, p. 260-1. Chicago, Ill., University of Chicago
Press.

127

E.V. Tamesis

and D.C. Salita

LUCAS,
L. R. 1966. Soil survey of Camarines Norte Province,
I'hilippines. Republic of the Philippines, DANR, Manila,
Bureau of Soils. (Soil Report no. 23.)
MOHR, E. C. 1944. The soils of equatorial regions with special
reference io the Netherlands East Indies. p. 31-2. A n n Arbor,
Mich., Edwards. 766 p.

BAREN, F. A. 1954. Tropical roils. London, Interscience

Publishers.
PHILIPPINE
WEATHER
BUREAU.1961. Annual climatological
review. Republic of the Philippines, Dept. of Commerce
and Industry, Weather Bureau. (Scientific Paper, no. SO.)

-;

/--

I
.

,/'

Pedogenesis and soil fertility in West Malaysia

Ng sew K c e
Rubber Research Institute, Malaya
and
Law Wei Min
Department of Agriculture, W e s t Malaysia (Malaysia)

INTRODUCTION
Soil classification is a special objective of pedologists
w h o carry out soil surveys to define taxonmic units
in the field, but it also holds interest for soil fertility
specialists and crop scientists w h o deal with soil m a n agement and nutritional requirements of crops. The
t e r m fertilityis used here in a more restricted sense
to m e a n the nutrient status of soils. If meaningful correlations between fertility levels and genetic groupings
can be established, then soil classification can serve
an additional practical purpose in defining nutritional
priorities in each soil group. If this delineation can be
further expressed on a map, the strategy to solve fertility problems can be better planned.
In agriculturally advanced countries, however, where
land has been intensely cultivated for decades and
modern fertilizer practice widely adopted, inherent
fertility differences between soil groups have been
either considerably diminished or eliminated and the
usefulness of genetic classification for fertility purposes has limited scope. On the other hand, in m a n y
tropical regions such as South-East Asia, vast tracts of
land still remain under primary forests and in Thailand
and Malaysia at least, reconnaissance soil surveys have
been undertaken to determine the agricultural potential
of undeveloped lands. Soils in these countries have been
classified mainly along the genetic concept (Panton,
1964;T h o m a s and Allen, 1966; Soil Survey Staff, 1966;
and M o o r m a n and Rajanasoothon, 1968), but their
relationships with fertility status have not been closely
examined. As an understanding of the pedogeneticfertility relationship can have significance in assessing
nutrient inputs in future croppings of new lands and
thus enhance the agronomic value of soil classification,
an evaluation of this relationship for West Malaysian
soils is m a d e here in the light of data available at present. This evaluation is only tentative because ana-

lytical data are not sufficiently comprehensive for all

soils.

F A C T O R S OF SOIL F O R M A T I O N
Factors of pedogenesis in West Malaysia are outlined
in brief as follows.
CLIMATE

T h e Malayan Peninsula lies between the Equator and

70 N and has a hot, humid climate. D a y temperatures
usually lie between 26-320C with night temperatures
about 4-50 coler. Annual precipitation generally
ranges from 1,800 m m to over 3,500 m m (Oh, 1965),
the heaviest rainfall areas being the main mountain
ranges and the lowest areas being in the extreme northwest and part of the central region of the country. In
the extensive lowlands and alluvial flats, rainfall
averages mostly 2,000-3,000mm. In the southern and
central districts, rainfall is fairly evenly distributed but
in the north-east and north-west, a marked monsoonal
rainfall pattern prevails and rainfall in excess of 450 m m
per m o n t h in the wet season can occur (Wycherley,
1967). Relative humidity during the day is usually
in the nineties.
PARENT MATERIALS
.

As far soil formation is concerned, the geology of W e s t

Malaysia can be said to be relatively uncomplicated.
With the exception of the older Langkawi Islands in the
north-west,the majority of rocks are Triassic or younger.
The principal geological formations are : igneous;
sedimentary including metamorphics; alluvial deposits.
i. Specially presented as supporting paper to review No. 4 by G. Donald
Sherman

129

Ng Siew Kee and Law Wei Min

Igneous rocks comprise predominantly granites with

ancillary outcrops of rhyolite or dacite, and to a relatively minor extent basic rocks of andesite, l~asaltand
tuffs. T h e major mountain ranges consist of granite
rocks but in the south, granite forms an extensive part
of the lowlands. Sedimentary formations are composetl
of quartzite, sandstone, shale, mudstone, and schist.
Sandstones and shales, often interbedded, constitute
the major portion of sedimentary rocks. Limestones
occur as distinct land forms in the north-west and
central parts of the country, but in area they are insignificant. Finally, the alluvial deposits mainly occupy
the major coastal and river-flood plains of the country, although some deposits occur in inland situations.
Seven main types of deposits have been distinguished,
mainly according to characteristic land forms (Gopinathan, 1968). These are :(a) high-levelterrace deposits
which have been dissected and occur at elevations of
about 80 m in southern and central states; (b) intermediate-level deposits at elevations of 16-50 m which
are weakly dissected and gently undulating in topography occurring mainly in the northern states; (c) lowlevel terrace deposits which lie on flat topography but
above the coastal and flood plains; (d) recent coastal
marine clay deposits which are extensive all along the
west coast; (e) riverine flood deposits which occur in
the plains of major rivers and are most developed in
the river systems of the east coast; (f) beach sand deposits mainly along the east coast and, lastly; (6) peat
deposits found in depressions within the major coastal
or flood plains.
VEGETATION
F r o m the classification of Malayan forest types b y
Wyatt-Smith (1964), three major groups of forests
can be distinguished, namely : hill and mountain
vegetation; lowland (dryland) vegetation; and s w a m p
and low-lying vegetation.
Hill and mountain vegetation occurs on central
ranges above an altitude of 350 m and consists of mainly
hill and upper Dipterocarp forest in the lower altitudes
and Montane forest at higher altitudes. Lowland vegetation mainly comprises primary Dipterocarp forests
with a multitude of species but the principal being
Dipterocarp and Shorea species. Under s w a m p and lowlying vegetation, species of Rhizophora and Avicennia
predominate in mangrove s w a m p forests in coast lines,
particularly in the west coast. In the big fresh-water
swamps, typical peat vegetation with limited species
is found.
RELIEF
T h e physical relief of West Malaysia is dominated by
the central mountain range which runs nearly through
the middle of the country and up to a height of about
2,400 m. Secondary ranges fan out from this mainly
130

in thc northern half of the country. These ranges

consist of steepland with dope8 exceeding 250 and
constitute nome 40 per cent of the total land surface.
F r o m these mountain systems, flow the major rivers
through hilly, rolling and undulating lowlands towards
the flood plains and coastal flats. The intermediate lowlands lie mainly between contours of 20-160 m. Thus,
relatively free drainage and varying degrees of erosion
are prevalent features in this region whereas wetness
is more characteristic of the flood plains.

G R E A T SOIL G R O U P S
Owen (1951)first classified provisionally soils in W e s t
Malaysia into six great soil groups. Later, Panton (1964,)
established nine soil groups and expressed them in
m a p form. A modification of Panton's scheme of classification, based on more recent field and laboratory data,
is adopted for the purpose of this study. T h e various
soil groups, major parent materials, characteristic
profile horizonation and approximate extent based on
original data of L a w (1968),are given in Table 1. Brief
descriptions of t+ main morphological characteristics
are as follows.
LATOSOLS

The solum is deep, dark-brown and red coloured and

has strong granular or crumb structure and very friable
consistence. T h e h u m u s horizon is not very distinct.
Macrostructure is weak. Texture is commonly clay
although appears more to be a loam b y feel of hand.
Internal drainage is good and the surface layer tends
to dry out readily during dry spells. T h e iron content
of these soils is usually 20-30 per cent Fe,O,.
R E D D I S II-BR O W N L A T E RITI CS

Soils in this group are characterized b y a

horizon

with moderately developed subangular blocky structures with moderately firm to friable consistence.
However, clay eluviation is indistinct. Textures vary
from clay loam to silty clay. Fragmentary concretions
of laterite m a y occur in the subsoil. Iron contents in
these soils range from 8-12per cent Fe20,.
R E D - Y E L L O W PODZOLICS

This soil is the most extensive and the typical feature

is a distinct eluvial horizon overlying an argillic B
horizon, which is relatively stronger in colour. Textures
are commonly sandy loam to sandy clay and structures
subangular blocky with friable to moderately firm
consistence. Iron contents range from 2-5 per cent

Fe,O,.

Pedogenesis and soil fertility in West Malaysia

TABLE1. Great soil groups of West Malaysia

Estimated area
Soil group

(in thousand ha)

Major parent materids

Horizonation

Terrain

1. Latosol

Uasalt, Andesite, Schist

Dominant U horizon

Roiling to hilly

2. Reddish-Brown

Diorite, Tuffs,Shale

W e a k Ae/Bt

Undulating to hilly

480

Granite, rhyolite, sandstone,

sandy shale, older alluvium

Distinct Ae/Bt

Undulating to hilly

2 340

4. Yellow-Grey
Podzolic

Shales, shales interbedded

with siltstones

Distinct Ae/Bt
coarse structures

Undulating to rolling

1160

5. Caterite

Shales, Phyllites

B position occupied by

Undulating to rolling

320

80

Lateritic
3. Red-Yellow

Podzolic

iron rich concretions

6. Gley

Marine sediments, subrecent

alluvium

A-BGor A-G

Flat

1300

7. Alluvial Soil

Recent riverine alluvium

AC

Flat

490

8. Podzol

Beach sands, granites

Bleached A,,
humus and/or iron B

Undulating

290

9. Organic Soil

Dead swamp vegetation

Organic horizon

Flat

720

'

10. Lithosol

Granite, quartzite

Incipient U

Very steep

TOTAL
:

Y E L L O W - G R E Y PODZOLICS

Textures are usually silty loam to silty clay. T h e A

horizon is thin with moderate subangular blocky
aggregates which are rm. T h e subsoil has coarse
angular blocky or primatic structures with firm or
very firm consistence. A n eluvial/alluvial sequence is
fairly distinct. Mettling, indicative of impeded drainage, is c o m m o n in the I3 horizon and discontinuous
iron concretions can occur. However, iron contents in
these soils are usually below 2 per cent Fe,Os.
LATERITES

T h e presence of a dominating, often compacted horizon of iron-rich concretions within 45 c m of the surface
is characteristic of soils in this group. The topsoil is
usually strong brown to reddish brown and loam to
clay loam in texture with friable consistence. In the
laterite horizon, the concretions constitute over,50 per
cent in weight. Often, the laterite can be massive,
particularly o n higher slope?. Below the concretionary
horizon variegated clay is typically found.

5 400
12 580

charzcteristics of soils of this group. Textures are

generally clayey and structures of subsoils range from
angular blocky to' prismatic. .Topsoils vary in organic
matter content and the gley horizon is usually found
at about 1 m depth. These soils are generally poorly
to very poorly drained.
A L L U V I A L SOILS

These are juvenile soils with AC profiles. Colours are

yellowish to strong b r o w n and textures are variable,
from loam to clay. Structures are generally w e a k and
consistence friable. Primary minerals can be c o m monly observed in these profiles.
PODZOLS

A bleached A, horizon overlying a h u m u s and/or B

horizon is characteristic of these soils. T h e y are largely
found o n old beach sands o n the east coast but also
occur at altitudes of above, 1,350 m o n acid granites
(Panton, 1958).
,

GLEYS

O R G A N I C SOILS

The

These soils have more than 65 per cent organic matter

which is greater than 30 cm in depth. Woody m o r

prominence of gleying and associated features of

mottling in root channels and structural faces are

131

Ng Siew Ree

and Law Wei Min

humus, low bulk density, presence of buried timber and

very poor drainage are characteristics of such soils.
LITHOSOLS
'

These are young soils with an incipient or weak 13

horizon on very steep slopes and relatively little detailed
study has been m a d e on them.

NUTRIENT STATUS
Nutrient status is gauged by conventional soil analytical data which are somewhat arbitrarily divided
into: (a) indices which indicate more labile or easily
available forms of nutrients, and (b) indices indicating
a less readily available or reserve category of nutrients.

T o eliminate the variable of past fertiliser use, data for

profiles under virgin forests or that are uncultivated
are used except in the case of the Gleys because they
have been largely developed for agriculture. However,
in this instance, sample sites were restricted to smallholdings which were unlikely to have received m u c h
fertiliser. Data are derived from published works
(Dumanski, 1966; Joseph, 1965;Leamy, 1966;Ng,1965,
1966; Panton, 1957, 1958; W o n g , 1966) as well
as unpublished records. Data for the Organic soils and
Lithosols are very meagre and consequently they are
omitted from consideration. The Podzols and Alluvial
soils also have not been studied in sufficient detail:
therefore conclusions about them are more tentative.
T h e numbers of profiles examined for each soil group
are as follows :Latosols 14,Reddish-Brown Lateritic 18,

Topsoi l

PH
4
- .

'

,II

400

Topsoil

300

200

1 O0

O'

400

Subsoi I

300

PH

200

LS R B L RYP YGP

FIG.1. PII (in water).

132

looi! i * I I
O

Ls

RBL

RYP YGP

FIG.2. 0.1N NaOII soluble inorganic P.

Pedogenesis and soil fertility in West Malaysia

Red-Yellow Podzolic 34, Yellow-Grey Podzolic 19,

Laterites 10, Gley 18, Podzols 5 and Alluvials 8.
Only topsoils of 10-15c m in depth and subsoils d o w n
to a depth of 30-45 c m are taken in consideration.
'

DIFFERENTIATIONA T
G R E A T SOIL G R O U P LEVEL

the majority being in the range 3.5-5.5. T h e Organic

soils have the lowest values as well as some m e m b e r s
'ofthe Gleys. Otherwise, there are no marked differences
between soil groups. This is an indication that the
bases in the -exchange complex have been largely
replaced by hydrogen or aluminium ions, brought
about b y strong weathering and leaching processes.

2. O.IN N a O H soluble inorganic phosphate (Fig. 2).

AVAILABILITY I N D I C E S

1. pH (water). This is the simplest soil measurement

but as a single index, it probably has the most value
as it gives an indication of the stage of soil weathering,
availability of certain nutrients and need for liming for
certain crops. Figure 1 shows that all the soils are acid,

Only the Latosols have distinctly higher levels of

phosphate, their m e a n levels being greater than the
m a x i m u m values found for the other groups. Differences between the other groups are not clearly evident
although the Yellow-Grey Podzolics and Podzols
incline to possess the lowest levels. M e a n levels for the
groups except the Latosols are below 80 p.p.m., a

1.6

Topsoil

1.4
Subsoil

--_I

0.5

Topsoil

Subsoil

1.2
0.4

1 .o

I
I

0.8

I
I
I
I

0.6

I
I

0.4

T
I
I
0.2
cn

>
2

0.3

0.2

T
I
I

T
I

I
I

-i-I

I
I
+

T
I

I
I
I

0.1

--T

iI

4I

R B L RYP YGP

FIG. 3. Exchangeable K.

t
1

I
I
A

LS

RBL

Ls

I
I
I

RYP YGP

FIG.4. Nitrogen contents.

133

Ng Siew Kee and Law Wei Min

TABLE2. Exchangeable Me values in soil groups (m.e./100g)

Topsoil

Exchangeable calcium (Table

T h e pattern of calcium is similar to that of magnesium,

the highest status being in the Gleys, followed IJY the
Alluvials, the means exceeding 1 m.e./100 g. For the
other soil groups, m e a n values are very low, being less
than 0.5 m.e./100 g soil.
T h e preceding data indicate the strongly leached
condition of most of the soil groups although they were
derived-from different materials. In the case of less
mobile phosphate, however, parent material appears
to be the cause of the main difference found.

Subsoil

Soil group
Range

Latosol
Reddish-BrownLateritic
Red-Yellow Podzolic
Yellow-Grey Podzolic
Laterite
Gley
Alluvial
Podzol

0.10- 0.93
0.21- 0.82
0.16- 0.59
0.10- 0.73
0.26- 0.78
0.37-12.6
0.74- 3.94
0.05- 0.74

Mean

Range

Mean

0.61
0.55
0.34
0.42
0.47
3.17
2.40
0.32

0.10- 0.59
0.10- 0.30
0.05- 0.42
0.05- 0.58
0.05- 0.58
0.33-10.0
0.16- 4.26
0.05- 0.32

0.21
0.17
0.17
0.21
0.22
3.11
1.61
0.15

3)

RESERVE INDICES

Nitrogen

value below which response to phosphate application

is likely (Saunder, 1956).

Exchangeable potassium

Soil nitrogen occurs mainly in a rather inert organic

form and generally only a very small fraction of it
can be expected to be mineralized by microorganisms
into labile a m m o n i u m and nitrate forms. However,
the organic forms differ in their susceptibility to bacterial attack and therefore soils with similar nitrogen
contents m a y differ in their power to supply labile
forms of nitrogen. Figure 4 shows that topsoil means
of Red-Yellow Podzolics, Yellow-Grey Podzolics and
Podzols have lower values than the others but in the subsoils, Alluvials are also low. T h e over-all pattern appears to indicate a higher nitrogen status in the Gleys
and Latosols. than the rest.

(Fig.3)

T h e Gleys have the widest range of values for both

topsoils and subsoils and the Podzols, being sands,
have the lowest values. The Latosols, Reddish-Brown
Lateritics and Alluvial soils seem to have slightly
higher values than the Podzolics and Laterites. Except
for the Gleys, subsoil m e a n values are less than
0.2 m.e./100 g soil. Thus about three classes can be
differentiated well and a fourth marginally.

6N HC1 extractable

Exchangeable magnesium (Table 2)

nutrients

groups are not distinct and values are generally less

than 0.6 m.e./100 g soil, indicating very low availability.

Phosphorus (Fig.5). As in the case of the more mobile

fraction, the Latosols stand out clearly as the group
richest in this nutrient. The Gleys have a wider range
of values than the other groups while the Podzolics
tend to have lowest levels. Thus, about three classes
can be separated out by this criterion.

TABLE3. Exchangeable Ca values in soil groups (m.~./lOO

g)

TABLE4. 6 N HCl extractable K in soil groups (m.e./lO g) :.

The

Gleys and Alluvials are separated out by their

higher values, with the Gleys tending to have the

highest levels. Differences between the other soil

Subsoil

Topsoil

Soil group

Latosol
Reddish-BrownLateritic
Red-Yellow Podzolic
Yellow-Grey Podzolic
Laterite
Gley
Alluvial
Podzol

134

--

Range

0.10-0.78
0.10-0.45
0.05-0.32
0.10-0.42
0.15-0.58
0.10-6.90
0.57-4.68
0.05-0.26

Mean

Range
~

0.35
0.23
0.15
0.30
0.30
1.80
2.62
0.14

0.05-0.21
0.05-0.21
0.05-0.20
0.05-0.21
0.05-0.37
0.05-5.20
0.15-1.47
0.05-0.16

Topsoil
Mean

0.11
0.13
0.09
0.12
0.14
1.25
0.87
0.09

Subsoil

Soil group

Latosol
Reddish-BrownLateritic
Red-YellowPodzolic
Yellow-GreyPodzolic
Laterite

Gleys
Alluvial

Range

Mean

Range

Mean

0.37- 1.97
0.57- 9.83
0.32- 4.73
0.96-14.1
1.30- 3.74
0.80-12.5
1.02- 6.19

0.93
2.83
1.40
6.89
2.52
3.57
3.11

0.32- 1.69
0.81-10.8
0.30- 5.93
2.52-18.1
1.09- 5.04
0.64-13.6
1.46-10.0

0.84
3.28
1.67
9.04
2.61
5.22
4.87

Pedogenesis and soil fertility in W e s t Malaysia

PPm
1,500

Potassium (Table 4). The ranges show that the Latosols are poorest in potassium reserves while the YellowGrey Podzolics and Gleys tend to be richest. According
to m e a n values, the Latosols and Red-Yellow Podzolics
have less than 2 m.e./100 g in both horizons, the Reddish-Brown Lateritic, and Laterites have 2.4 m.e./100 g
and the rest greater than 4 m.e./100 g except for the
topsoils of the Gleys and Alluvials.

____

Topsoil

Subsoil

1,250

Magnesium (Table 5). By range and m e a n values,

only the Gleys and Alluvial can be differentiated to
have higher magnesium contents.

l-

I
I

1,000

E F F I C I E N C Y OF S E P A R A T I O N

On the basis of range and m e a n values, the n u m b e r of

distinct classes which each soil index can achieve is

shown in Table 6. It is evident that the nutrient indices m a k e considerably less differentiation than profile morphology, the relative efficiency being about
30 per cent. This is an indication that nutrient differences at the start of pedogenesis have been drastically
reduced by the intense weathering and leaching processes operating in a hot, humid environment where
only the more resistant constituents such as quartz
and sesquioxides remain. Further evidence of advanced
weathering comes from the dominance of kaolinite
in the clay fraction of m e m b e r s of most soil groups
except the Gleys (Ng,1965).

I
I
I
I
I

750

I
I
I
500

250

t
I
I
I
I
I
I

I
I

I
I
I
I

+ [i

I
l

O
Ls

RBL

RYP

YGP

FIG. 5. 6 N HCI soluble P.

DIFFERENTIATION B E L O W
G R E A T SOIL G R O U P L E V E L

great soil group is rather heterogeneous in morphological terms and it is possible that generalizations
can m a s k differences within groups. In order to assess
whether intra-group differences exist, data for major
. families in five great soil groups are examined. Table 7
shows that only in Gleys is further clear separation

TABLE6. Number of nutrient classes separable by soil indices

TABLE5. 6 N IC1 extractable M g in soil groups (m.e./100 g)

No. of separableclasses
Soil index

Subsoil

Topsoil

Number of
sOi'.~Oups

Average

Topsoil

Subsoil

2
2
3
2
2
2
3
3
2

2
2
3
2
2
2
3
3
2

Soil group

Latosol
Reddish-BrownLateritic
Red Yellow Podzolic
Yellow-GreyPodzolic
Laterite
Gley
Alluvial

Range

Mean

Range

Mean

1.13- 6.91
1.08- 9.48
2.48- 4.78
2.36- 6.51
1.60- 6.08
2.38-12.4
3.84-11.9

3.68
4.37
3.94
3.60
3.70
5.89
7.31

1.60- 5.28
1.28-15.3
0.80- 6.10
2.84- 5.95
1.28- 8.64
2.67-16.7
3.86-10.9

3.08
5.28
3.30
3.82
4.52
6.51
7.29

PII
NaOH-P
Exch. Ei
Exch. M g
Exch. Ca

N
'

6 N IICl P
6 N HCl K
6 N HCl M g

'

135

Ng Siew Kee and L a w Wei Min

C R O P RESPONSE TO M A N U R I N G

possible and in the Latosols, the Kuantan family tends

to have a higher phosphate but lower potassium status.
In the other soil groups, further separation is not
sustained although in the Red-Yellow Podzolics, the
Harimau family has lower potassium content.
In general therefore, only in the Gleys has there been
the masking of distinct families due to generalisation,
the Manik family being a more typical low humic
gley formed o n terrace alluvium. Although not shown,
a further family, Linau, being acid sulphate in character
can be separated but by p H and sulphur contents
(Chow, 1969) rather than by the c o m m o n nutrient

If the general lack of fertility differences depicted by

the soil indices used is valid, it should be substantially
sustained by agronomic experience. Unfortunately,
few fertiliser experiments in the past have been laid
d o w n according to well-defined soil series that had not
been cultivated, and consequently, relevant information
is rather limited. Nevertheless, in order to project an
overall impression, crop response data for t w o major
crops in the country, rubber and oil palm are presented in Tables 9 and 10. S o m e of the data are for
cultivated soils but they should be fairly pertinent to

indices.

TABLE7.cNaOH-P,exchangeable K and M g of main soil families

S&soil

Topsoil

Family

Mg (me. yo)

%)

P (p.p.m.)

K (me.

109-406
65-205

0.06-0.16
0.06-0.60

Milg

0.31-0.74
0.63-0.93

80-389
49-174

0.03-0.06
0.02-0.34

0.13-0.20
0.10-0.59

Latosol

Kuantan
Segamat

Reddish-Brown
Lateritic

Munchong
Kampong Kolam

26-95
36-95

0.14-0.40
0.05-0.40

0.21-0.82
0.26-0.83

12-80
29-80

0.03-0.49
0.02-0.13

0.10-0.30
0.05-0.26

Red-Yellow
Podzolic

Rengam
Serdang
IIarimau

20-47
21-75
41-59

0.03-0.32
0.07-0.20
0.07-0.17

0.16-0.51
0.20-0.42
0.25-0.59

30-40
15-52
35-65

0.03-0.10
0.04-0.13
0.04-0.07

0.05-0.31
0.05-0.42
0.10-0.20

Yellow-Grey
Podzolic

Durian
Batu A n a m

23-86
17-72

0.21-0.29
0.04-0.22

0.20-0.73
0.10-0.67

19-64
13-62

0.12-0.16
0.03-0.11

0.15-0.58
0.05-0.25

Gley

Selangor
Manik

65-158
27-42

0.23-1.57
0.12-0.59

1.27-12.6
0.41-1.08

32-135
11-26

0.34-0.97
0.05-0.15

0.42-10.0
0.33-0.67

'

'

TABLE8. N, 6 N IICl soluble P, K and M g in major soil families

Subsoil

Topsoil
soil group

Family

N(%)

Pb.p.m.) K (me.%)

Latosol

Kuantan
Segamat

0.15-0.36
0.10-0.38

400-885 0.37-0.64
220-1320 0.57-1.97

Reddish-Brown
Lateritic

Munchong
Kg. Kolam

0.07-0.41
0.10-0.41

90-260 1.04-9.83
41-224 0.57-3.13

Red-Yellow
Podzolic

Rengam
Serdang
Harimau

0.06-0.18
0.09-0.25
0.11-0.15

161-190 0.64-4.73
60-148 0.32-2.96
52-70 0.50-0.64

Yellow-Grey
Podzolic

Durian
Batu A n a m

0.06-0.31
0.06-0.16

46-187 2.37-13.4
52-120 0.96-14.1

Gley

Selangor
Manik

0.15-0.48 209-580 6.50:12.5

0.09-0.15
78-157 0.80-3.19

136

Mg(m.e.%)

1.68-3.08 0.04-0.22 400-750 0.32-0.64

4.02-6.91 0.10-0.17 214-1110 0.57-1.69
1.08-9.48
1.61-8.39

0.05-0.14
0.04-0.14

Mg

1.60-2.00
2.41-5.28

26-140 0.81-10.8 1.60-8.50

31-231 1.09-6.39 1.28-15.37

3.54-4.60 0.02-0.11 113-142 0.32-5.93 0.80-3.64

3.21-4.66 0.03-0.10 46-131 0.32-3.74 2.88-4.68
2.48-4.56 0.05-0.06 60-94 0.30-0.64 1.92-4.56
3.04-6.51
2.36-4.96

0.06-0.12
0.03-0.08

40-160 2.52-18.1
28-120 2.97-16.4

2.84-5.58
2.34-5.95

2.38-10.6 0.07-0.31 168-500 6.36-13.6 2.67-17.8

n.d.
0.03-0.05 90-142 0.97-2.90
n.d.

Pedogenesis and soil fertility in W e s t Malaysia

TABLE9. Rubber response to fertilizers

Soil scries

Soil group

Response to

Percentage response

source

Serdang (ex-jungle)

RYP

NPK
NP
PK

22 yo in yield
17 yo in yield
13 yo in yield

Guha and Pushparajah

(19660)

Rengam (ex-jungle)

RYP

8 yo in growth
1.5 yo in yield
14% in growth
3 yo in yield

Guha and Pushparajah

(19660)

P
Rengam (cultivated)

RYP

22 yo in yield

Guha and Pushparajah

(1966~)

Kuantan (ex-jungle)

La

K
Nil to P, M g

9 yo in immature

Guha and Pushparajah

(1966a)

%
NP

7 Yo in yield
9 % in growth
3 yo in yield

Guha and Pushparajah

(19660)

Munchong (cultivated)

RllL

growth

Malacca (cultivated)

NK
NPK

18 yo in yield
12 yo in growth

Guha and Pushparajah

(1966b)

Selangor (cultivated)

G.

5-10yo in yield

Guha and Pushparajah

(1966a)

Sitiawan/Sogomana(cultivated)

6 yo in growth

Guha and Pushparajah

(1966b)

TABLE10. Oil palm response to fertilizers

Soil series

Rengam (ex-jungle) .

'

Soil group

RYP

Response to

Percentage response

Source

N
P
K

Rosenquist
(1964)

w?

8 yo in yield
5 yo in yield
33 yo in yield
2 yo in yield

Durian (ex-jungle)

YGP

N
P
K

12 yo in yield
22 yo in yield
2 yo in yield

Rosenquist
(1964)

Rengam (ex-jungle)

RYP

NPK M g

25 yo in yield

Rosenquist
(1964)

Batu Anam (ex-jungle)

YGP

N
P
K

10 yo in yield
15 yo in yield
22 yo in yield

Martineau et al.
(1969)

6 yo in yield

Hew (personal com-

w3
Selangor (replant)

N
K

8 yo in yield

munication, 1969)

137

Ng Siew Kee and L a w W e i Min

nitrogen and potassium as residual effects of these

nutrients applied are low.
For both crops, with the exception of the Latosol,
all the soil groups show responses to nitrogen, particularly in the Red-Yellow Podzolic and Yellow-Grey
Podzolics as better indicated by the more nutrient
demanding oil palm. This is in general consonance with
the soil nitrogen data. The result for the Latosol m a y
not be the true longer term picture as it only refers
to immature growth. Potassium shows a very strong
response in the Red-Yellow Podzolics for both crops
and nil or correspondingly lower response in the Gleys.
This s e e m s to accord satisfactorily with the soil potassium pattern. The Yellow-Grey Podzolics show a very
low or high response to potassium, probably associated
with the very variable potassium reserves in these
soils. T h e Latosol, being low in potassium reserves,
has already shown a growth response for rubber. The
Reddish-Brown Lateritic alsa shows no response to
potassium, again probably associated with its moderate
potassium reserves. Response of rubber is also recorded
for the Laterite, a reflection perhaps of the small amount
of fine earth fraction in the solum, although it contains
moderate potassium reserve. Phosphate response is
registered in all the soils except the Gleys and Latosols,
a result in fair agreement with soil phosphate &tatus.
Magnesium i3 also indicated to be short in the RedYellow Podzolics, Reddish-Brown Lateritic and YellowGrey Podzolics but not in the Gleys and Latosols.
T h e requirements for three or four of the macronutrients in the Red-Yellow Podzolic, Yellow-Grey
Podzolic, Reddish-Brown Lateritic and Laterite soils
give reasonable confirmation to the earlier deduction
that the majority of W e s t Malaysian soils are impoverished of nutrients and therefore their nutritional
differences are fewer than their morphological variations. Thus, the strong weathering processes and intense leaching have polarised the fertility status of
West Malaysian soils, with the disproportionate majority in the low end, producing a skewed type of distriIJution.

F U T U R E DEVE L O PM ENT
A N D PHYSICAL FACTORS
T w o major conclusions can be drawn from the preceding analysis.

138

First,the successful development of the largc reserves

of virgin soils in West Malaysia with agriculture depends to a conbiderable extent on the availability and
relative cheapness of fertiliser inputs. Little can be
relied upon the natural soil fertility to sustain large
yields. As pointed out by E d e n (1964), as true of
temperate soils, the fertility of tropical soils also must
be m a n - m a d e or controlled, and this is at least pertinent to most parts of Malaysia and Thailand, if not
the whole of South-East Asia. The use of fertilizers
would tend to minimize inherent fertility as well as.
productivity differences (Davtyan, 1963) if nutrients
alone were the limiting factors.
Second, the lack of close relationship between pedogenetic classification and inherent fertility in W e s t
Malaysia is a natural consequence and should not
invalidate the broad scheme of genetic classification.
However, such a classification cannot be stretched to
cover the fertility aspect as well. As commented b y
Northcote (1964),a single operation cannot be expected
to classify soils as soils and also as media of differing
degrees of suitability for plant growth. In mitigation
however, it should be stressed that soil nutrients
alone do not determine crop growth. ,The effect of
nutrients can be greatly influenced b y physical soil
properties which affect rooting and water availabilit
Thus, it appears that certain physical measuremen$
such as bulk density, water-holding capacity, available
water and permeability of soil individuals might ephance the agronomie value of soil classification. Unfortiinately, such data are extremely limited at present and
greater emphasis should be accorded in future to this
area of soil research in this region.

A C I< N O W L E D GERIE NT
W e thank the Director, Rubber Research Institute of
Malaya, and the Director of Agriculture, West Malayc;ia,
for permission to present this paper. The work reported
on was carried out w h e n one of us (N.S.K.)was also a
m e m b e r of staff of the Department of Agriculture.
Grateful thanks are also due to Messrs. M. L. Leamy,
H. Smallwood, J. Dumanski, Ignatius W o n g and Chin
Kim W a h for assistance in sampling and analysis. .

Pedogenesis and soil fertility in W e s t Malaysia

Bibliography
Criow WENC
TAI.1968. A preliminary study on acid-sulphate
soils in West Malaysia. proc. 3rd hfdaysian soil conf.,
Kuching, Sarawak, p. 51-8.
DAVTYAN, G.S. 1963. General Problemes of soil science relationships between the genetic group as a soil and its agricultural chemicalproperties. Soviet Soil Sci., vol. 7, p. 619-26.
DUMANSKI,J.; 001, Cheng Hock. 1966. Reconnaissance soil
survey of the Temerloh-Gemas region. Kuala Lumpur,
Dept. Agric. 67 p. (Malayan Soil Survey Report no. 5/
1966.)
EDEN,T. 1964. Elements of tropical soil science. London,
Macmillan &Co., 164 p.
GOPINATHAN, B. 1968. Terrace and alluvial soils in W,est
Malaysia. Proc. 3rd Malaysian Soil ,.noC
Kuching, Sarawak,
p. 45-50.
GUHA,M. M.; PUSHPARAJAH,E. 1966a. Responses to fertilisers
in relation to soil type. R.R.I. Plant Bull. (Kuala Lumpur),
no. 87, p. 178-83.
. 1966b. Responses to fertilisers in Hevea Brasiliensis,
in relation to soil characteristics. Proc. 2nd Malaysian
Soil Conf., p. 194-201. Kuala Lumpur, Min. Agric. and Coop., Malaysia.
JOSEPH, K.T. 1965. The reconnaissance soil survey of Kedah.
Kuala Lumpur, Min. Agric. and Co-op., Malaysia. 39 p.
(Bull.no. 117.)
LAW Wei Min; SELVADURAI,K. 1968. The 1968 reconnaissance soil m a p of Malaya. Proc. 3rd Malaysian Soil Con.,
Kuching, Sarawak, p. 229-39.
L E ~ Y M.
, L. 1966. Proposals for a technical classijcation of
Malayan soifs. Kuala Lumpur, Dept. Agric., 62 p. (Malayan
Soil S w e y Report No. 3/1966.)
MARTINEAU, P. G.; KNECHT, J. C. X.; RAMACHANDRAN, P.
1969. An oil palm manurial experiment on an inland soil.
Progress in oil palm. Kuala Lumpur, Incorp. Soc. Planters,
Malaysia, p. 82-104.
MOORMAN,
F. R.; RAJANASOONTHON
SANTHAD.1969. The soil
m a p of Thailand. Proc. 3rd Malaysian Soil ,.noC
Kuching,
Sarawak, p. 225-8.

NG,S. K. 1965. The potassium status of some Malayan soils.

Malay. agric. J., no. 45, p. 143-61.
-. 1966. Soils. The oil palm in Malaya, p. 24-47. Kuala
Lumpur, Min. Agric. and Co-op., Malaysia.
NORTHCOTE, K.H.1964. Some thoughts concerning agronomy
and soil classification. J. Aust. Inst. agric. Sci., p. 241-6.
OH, Kong Yew. 1965. Hydrology in Malaya, rainfall and
river discharge. MQ~QY.
agric. J., no. 45, p. 182-90.
OWEN,G. 1951. A provisional classification of Malayan soils.
J. Soil Sci., vol. 2, p. 20-42.
PANTON, W.P. 1957. Soil survey report No. 5. The Federal
Experiment Station, Jerangau, Trengganu. Malay. agric.
J., no. 40, p. 19-29.
.
1958. Reconnaissance soil survey of Trengganu. Kuda
Lumpur. 59 p. (Dept. of Agric. Bull. No. 105.)
. 1964. The 1962 soil m a p of Malaya. J. trop. Geogr.,
VOI. 18, p. 119-24.
ROSENQUIST,
E.A. 1962. Fertiliser experiments on oil palm
'
in Malaya. Part I. Yield data. J. W.Afric. Inst. Oil Palm
Res., vol. 3, no. 12, p. 291-301.
SAUNDER, D.H. 1956. Determination of available phosphorus
in tropical soils by extraction with sodium hydroxide. Soil
Sci., vol. 82, p. 457-63.
SOILSURVEY STAFF. 1966. A classification of Sarawak soils.
Proc. 2nd IlfaZaysian Soil Con., p. 33-78. Kuala Lumpur,
Min. Agric. and Co-op., Malaysia.
THOMAS, P.; ALLEN, A. W. 1966. Provisional soil m a p of
Sabah. Proc. 2nd Malaysian Soil ,.noC
p. 13-31. Kuala
Lumpur, Min. Agric. and Co-op., Malaysia.
WONG,
I. F. T. 1966. Reconnaissance soil survey of Selangor.
Kuala Lumpur, Dept. Agric., 63 p. (Malayan Soil Survey
Report No. 6/1966.)
WYATT-SMITH, J. 1964. A preliminary vegetation m a p of
Malaya with descriptions of the vegetation types. J. trop.
Geogr., vol. 18, p. 200-13.
WYCHERLEY, P. R. 1967. Rainfall probability tables for
Malaysia. Rubb. Res. Inst. Malaya, 85 p. (RRI planting
manual No. 12.)

139

III

Final report
of the symposium

Final report of the symposium

1. During the inaugural session, addresses were delivered by Professor Dr. D. A. Tisna Amidjaja, Rector,
Institute of Technology Bandung; Professor Sarwono
Prawirohardjo, Chairman, Indonesian Institute of
Science; and Mr. F. J. C. Pala, Deputy Director,
Unesco Field Science Onice for South-East Asia.
2. A t the opening session, the Symposium nominated
Dr.

RIr.

D.Muljadi (Indonesia), Dr. N g Siew K e e (Malaysia),

Boonyawat (Thailand), Dr. Domingo Salita (Phi-

lippines) and Dr. Seshagiri R a o (India) as chairmen for

the different sessions, and Professor Dr. J. A. Katili
as Deputy Director of the Symposium.

3. T h e first working session was devoted to the presentation of the key paper o n Formation and transformation of clay minerals in tropical soils, by Professor J. J. Fripiat.

4. During the discussion, the following points were

emphasized.
In the tropics, because of higher temperatures and
greater evaporation and precipitation, formation and
transformation of clay minerals hy hydrolysis proceeds at a faster rate. It is assumed that the chemical
reactions involved in the hydrolysis process are not
different from the similar process operating in temperate climate.
In general, the formation and transformation of
primary minerals to secondary minerals are through
hydration and hydrolysis of the primary minerals.
T h e monolayer water in the capillary pores in the soil
is dissociated and acts as a u acid forming the hydron i u m H,O+ ion. As far as the formation of polywater
is concerned, it is possible that cleanliness of the capillary pores will be one of the factors of its synthesis.
It is believed that this polywater has a sheet structure.

Exceptions to the dissolution process in the formation

of secondary minerals by direct solid-phasesubstitution
are few and only possible in the formation of illite
from montmorillonite and vermiculite.
A s the soil is a n open system in which leaching and
movement of soil solution would he continuous, a n
equilibrium between the ions in soil solution and the
primary minerals would not be achieved. Consequently there is n o relationship in the chemical c o m positions of the primary parent material and that of the
secondary minerals or that of the soil solution. Similarly, in well-developed soils, there is n o direct correlation of soil fertility and its parent material. However
under direct solid-phase transformation, it is possible
that lattices of the secondary minerals m a y be related
to those of the primary minerals. This exception is
not c o m m o n . In the transformation processes, organic
matter m a y operate as a p u m p , accumulating or
rejecting elements such as Fe2+, AP+, P
O
:
,
etc.,
at the right (or the wrong) moment. It has never been
proved that it has a more direct role.
T h e formation of polynuclear complexes of aluminium
is pH dependent and the change of aluminium from
fourfold to sixfold co-ordination occurs w h e n the p H
is low. However, uncombined cations in the solutions
could also influence the change of co-ordination of
aluminium. For example, it is more dificult under
similar hydrothermal conditions to synthesize kaolinite
in acid conditions in the presence of sodium.
It is possible that the transformation process of kaolinite to gibbsite, under certain conditions, is reversible.
This could possibly explain the presence of kaolinite
in older soils and gibbsite in younger soils in the tropics.
It is also necessary, in this respect, to examine the
amorphous materials in these older soils derived from
sedimentary parent materials.
T h e transformation of kaolinite to gibbsite has been
observed to be enhanced by presence of organic ions.

143

Final report of the symposium

T h e organic anions comhine with gibbsite (AL(OH),) to

form complexes and kaolinite will further be decomposed
and removetl according to the following schcnie :
Kaolinite

IIn acid

+ II,O 3 Al(OII), + II,SiO,

+ AI(OII), -+ AIII
+ II,O.

5. After this discussion, Dr. Iajuddin Ahmed presented

a paper on Distribution of mica in the soils of the
Madhupur Tract, East Pakistan.

6. S o m e remarks were m a d e about this paper.

The method of Jackson, used b y the authors for the

indirect measurement of the amount of mica in the clay,
might underestimate the amount of mica because of the
presence of potassium-depleted mica.
In the distribution of mica, it was noted that biotite
mica was present in the clay fraction. In the silt and
fine sand, both mica and feldspars were present.
The higher amount of illite in the top soil was attributed to the synthesis of these minerals in situ. This
synthesis was brought about by the return of K from
the organic matter. It was noted that the previous
vegetation in this area (top hill) was decidious forest,
the K content of which was apparently high.
7. The second session of the meeting was devoted to
the presentation, by Dr. Y. Segalen, of his key paper
on Metallic oxides and hydroxides in soils of the w a r m
and humid areas of the world: formation, identification, evolution.
8. T h e discussion emphasized m a n y points :
Because ilmenite is a very resistant mineral, it
cannot be considered as a good source of iron in the
weathering process. T h e presence of ilmenite which has
been transported in beaches (for example, Madagascar),
confirms the above statement.
Calcium carbonate and laterites do not occur together. If they are found together, then it must be due
to contamination of an old weathered product by
recent calcium carbonate on an ancient surface.

In Africa, as in other parts of the world, physical

breakdown of latcritcn occurs. The new transported
laterile ru1)llcs would be cemented together by iron
oxides to form another layer of laterite. Sometimes
this new laterite layer is found at depth, having been
covered by other transported materials as shown in
the lower part of Figure 1.
In some shallow soils of the tropics, the pH decreases
with depth. This can be explained by the action of roots
penetrating all the w a y into the parent material from
which they are able to remove bases from the rocks.
As a consequence, the amount of bases in the topsoil
(A horizon) increases and the upper horizon of the soil
will be more basic than the lower horizon.
Amorphous materials in soil can be removed by
repeatedly treating the. soil with 8N HC1 and/or
0.5N N a O H . This will of course also attack the crystalline materials to some degree. By removing these
amorphous materials, the basic colour of the soil will
show up and the presence of minerals can be recognized.
T h e colour is reddish yellow when goethite is the main
mineral, red when it is hematite, dark grey when
significant amounts of magnetite are present, greyish
when clay minerals alone are present. Yellow soils, in
general, do not contain amorphous materials.
Recent studies b y Taylor, of CSIRO (in press,
Clay Minerals) and b y Fuls, Rodrigue and Fripiat
(in press, Clay and Clay hfinerals) indicate that crystal-
line and amorphous iron in soil can be completely
removed by treating the soil with glycerol. The Feglycerol complex formed will decompose in boiling
water to form magnetite and maghemite, which can LC
separated quantitatively by using a magnet.

9. T h e third key paper : Weathering and soil forming

processes in the tropics, was presented b y its author,
Dr. J. van Schuylenborgh.
10. The discussion emphasized m a n y points.
T h e author, in his thermodynamic calculations, considered silica as orthosilicic acid; however, in.the case
of polymerized silica, the equilibrium equation m a y not

N e w cemented laterite

000 O 0 0
144

Final report of the symposium

be the same. T h e AF value of the reaction m a y not be

.the real significant value to be taken into consideration
for deciding which process will occur. T h e real process
depends o n the activation free energy which is, in n o
way, a direct function of AF. Soils are open systems.
T h e equilibrium condition has therefore to be changed
and, instead of stating it as TAS = AQ, it needs to
AQirrev..
be stated as TAS = AQrev.
T h e removal of F e in the profiles in tropics is c o m paratively less than in the temperate regions because
of accumulation of organic matter in the surface and
greater leaching capacity.
Nevertheless, in the thermodynamics applicable to
the K,O-AlzO,-SiOz-H,Osystem, a good relationship
has been found between the theory and the observed
mineralogical data in a few cases.
T h e plinthite is part of the oxic horizon, in which Fe,
as oxides and hydroxides, is deposited and silica is
leached out. Dehydrated F e is crystallized. However,
it is necessary to find out whether this hardening of
the plinthite horizon m a y not be due to transformation
in microstructure.

11. Four papers were presented in relation to this third

topic.
12. It was noted in the paper o n the nature of parent
rock o n soil formation under similar atmospheric climates (Rh.J. P. Andriesse) that soil formation, o n
parent rocks with varying contents of silica attains different aspects. Neither climate nor the altitude is responsible for the sequence. It seems that both climate
and parent material are interacting in such a w a y that
podzolization can become dominant if the parent
material contains sufficient silica. Similarly, the lateritization m a y occur at high altitude if the parent material is basic.
T h e humification and mineralization of organic
matter in the humid tropics and humid temperate zones
are not only caused by climatic factors but also by the
general base content of parent material. Mineralization in the humid temperate low land is generally rapid
but, with acid parent material, is retained.

13. Dr. V. M. Fridlaud presented a paper o n the differences between crusts of weathering and soils developing o n acid and basic rocks in the tropics, which was
followed by a discussion.

14. Mottling in the weathering crust of acid rocks is

generally more easily developed than in that of basic
rocks. It is also generally k n o w n that in identical topographical conditions, mottles are developed in poorly
drained soil.
T h e weathering of acid and basic rocks is affected
by structure and porosity of the two types of weathering
crust. Basic rocks develop into round structural elements with big inner porosity whereas acid rocks dev-

elop into angular ones with thin inner porosity.Therefore,

these structural irregularities affect the different water
rgimes of the weathering crusts and soils formed o n
them.
T h e pore sizes of the weathering crust are more
subject to the chemical nature of acid and basic rocks
than to physical properties of extrusive and intrusive
rocks. For example, rhyolite and andesite are physically
the same but different in chemical composition. Therefore, they give diKerent forms of porosity.
T h e general pore volume only of the weathering
crust does not conform to the drainage and gleization
characteristics.T h e differences in the pore volume of the
crusts and of the soils with respect to the radius of
structural elements and inner pores must be taken into
consideration.

15. After the presentation by &fr. T a n D o u n S u y of his

paper on Genesis and evolution of red and black
basaltic soils in Cambodia, it was remarked that
although differences in base contents of the black and
red basaltic soils are not very large, the differences arc
considered great enough to lead to different clay
minerals.
16. T h e c o m m e n t o n the paper: S o m e aspects of
lateritic soil formation in the Dahican-Alayao areas,
Camarines Norte Province, Philippines (Messrs. E. V.
Tamesis and D. C. Salita) was that, although high
Si0,/R,03 (2:1) is shown and a statement of incomplete chemical weathering is given, a n analysis of organic
carbon should also be made.

17. In the absence of Dr. Donald Sherman, Dr. Ng

Siew K e e presented the key paper o n Mineralweathering in relation to utilization of soils and Dr. J. J. Fripiat commented o n it.
18. This presentation w a s followed by a paper and a
discussion o n Pedogenesis and soil fertility in West
Malaysia (Dr.Ng Siew K e e and Mr. L a w W e i Afin).

19. T h e Red-Yellow Podzolics as recognized in Malaysia were so classified according to the definitions by
Thorpe and Smith (1949).This definition is not based
o n the occurrence of a n argillic horizon as introduced
in the 7th Approximation. Investigations are being
carried out to determine to which great groups these
Red-Yellow Podzolics could be placed in the classification. It has been observed that, although a n increase
in clay content could be noticed in the field and by
granulometric analysis, clay movement is not evident
in thin sections. However, it was observed in Surinam
and in the United States that in some soils a textural B
horizon has been found without any evidence of clay
orientation. Apparently the clay movement took place
in the past but the orientation had been destroyed.
1115

Final report of the symposium

Indications are that such soils could be placed in the

u1 tisols.
Observalions in East Malaysia (Sarawak) show that
m a n y of the soils, even recent alluvial soils, had developed from an oxic material which could qualify as an
oxic horizon and hence these soils could be classified
as Oxisols. Nevertheless, clay movement is evident in
these profiles. As no provision has been m a d e in the
7th Approximation, it was suggested that if the oxic
nature of the soils is more pronounced than horizonation, these soils should be included in the Oxisols
rather than in the Ultisols.
According to the classification system used in Africa,
it was noted that Ferrallitic soils are those containing
high amounts of 1 :1 clay minerals, iron oxides and/or
aluminium hydroxides, residual minerals and quartz.
These soils also have low base saturation and a crumb
structure. Fersiallitic soils, on the other hand, are those
containing large amounts of 2: 1 clay minerals, having
iron oxides but no gibbsite and some Fe present in the

146

clay lattices. They are usually shallow with prismatic

ant1 blocky structure and with high base saturation.
It was noted that the occurrence of Persiallitic soils is
unlikely in Malaysia.
It was noticed that the alluvial soils have lower
exchangeable potassium than the soils derived from
sedimentary materials. This could possibly be due
to the higher content of illite in the sedimentary rocks.
It was further noticed that, although the relative
amount of potassium in soils derived from sedimentary
rocks appears high, this is equivalent to the order of
1.5 per cent of illite and will not be detected by X-ray
analysis.
Soils derived from limestone and ultra basic rocks
are u n c o m m o n in Malaysia and are localized. Where
these soils occur on limestone and on ultrabasic rocks,
they contain high to toxic amounts of ore (Ni + Cr).
P a d d y soils in Malaysia are presently grouped with
the gley soils but separate classification m a y be necessary in future.

List of participants

Belgium

Professor J. J. Fripiat

Laboratoire de Physico-Chimie Minrale,

Institut des Sciences de la Terre,
42 Decroylaan, Heverlee-Louvain.

Cambodia

&Ir. Tan Boun Suy

Doyen de la Facult des Cultures Tropicales,

Universit Royale de Kompong-Cham.

France

Dr. P. Segalen

Services Scientifiques Centraux de l'ORSTOM,

70 Route d'Aulnay, 93 Bondy.

India

Dr. T. Scshagiri nao

Chief Scientific Officer,

Regional Research Station,
University of Agricultural Sciences,
Raichur, Mysore State.

Indonesia

Dr. H. Ling Ong

Lecturer in Geochemistry and Geology,

Department of Geology,
Institut Teknologi Bandung,
Djl. Ganeca 10, Bandung.

Dr. D. Muljadi

Director,
Soil Research Institute,
Djl. Ir. II. Djuanda 98, Bogor.

Mr. Soepraptohardjo

Deputy Director,
Soil Research Institute,
Djl. Ir. H. Djuanda 98, Bogor.

Professor Ur. G o Ban Hong

Director,
Central Research Institute for Agriculture,
Djl. Merdeka 99, Bogor.

Dr. Achmad II. Satari

Associate Professor in Soil Sciences,

Soil Science Department,
Bogor Agricultural University (IPB), Bogor.

Ir. T, Sukarna

Soil Scientist,
Faculty of Agriculture,
University of Padjadjaran, Bandung.

147

List of participants

hlalaysia

Dr. N g Siew Kee

Head,
Analytical Chemistry Division,
Rubber Research Institute of Malaysia,
P.O. Box 150, Kuala Lumpur.

Mr. J. P. Andriesse

Senior Officer,
Soil Survey Division
Department of Agriculture, Kuching, Sarawak.

Mr. H e w Choy Kean

Research Officer,

Oil Palm Research Station,

Harrisons and Crosfield,

P.O. Box 207, Banting, Selangor.

Netherlands

Dr. J. van Schuylenhorgh

Laboratory of Regional Soil Science,

Agricultural University,
Wageningen.

Pakistan

Dr. Iajuddiii Alimed

Head, Department of Soil Science,

Dacca University,
Dacca.

Philippines

Dr. Domingo Salita

Professor of Geology and Geography,

University of the Philippines,
Quezon City.

. Thailand

Mi.

Vichai Boonyawat

Soil Survey Division,

Department of Land Development,
Ministry of National Development,
Bangkok.

United States of America

Professor G. D. Sherman

Senior Professor of Soil Science,

College of Tropical Agriculture,
University of Ilawaii,
IIonolulu, Hawaii.

U.S.S.R.

Dr. V. Bl. Fridland

Senior Scientist and Professor of Moscow University,

Institute of Geography,
Academy of Sciences of U.S.S.R;

Viet-Nam

Professor Le Khac Pho

Professor of Geography at the Faculty of Education and

Faculty of Letters,
IIue University
2, Le loi, Hue.

INSTITUTIONS AND ORGANIZATIONS

Indonesian Instiiute of Sciences

(LIPI)

Professor Sarwono Prawirohardjo

Chairman,
Indonesian Institute of Sciences,
Djl. Teuku Tjhik Ditiro 43, Djakarta (Indonesia).

Professor J. A. Katili

Deputy Chairman for Natural Sciences,

Indonesian Institute of Sciences,
Djl. Teuku Tjhik Ditiro 43, Djakarta (Indonesia).

Miss Sjamsiah Ahmad

Chief, nureau for Institutional Relations,

Indonesian Institute of Sciences,
Ujl. Teuku Tjhik Ditiro 43, Djakarta (Indonesia).

14'8

List of participants

Institute of Technology Bandung

(ITB)

Professor Dr. D.A. Tisna Amidjaja

Rector,
Institute of Technology Bandung,
Djl. Kapt. Pattimura 64, Bandung (Indonesia).

BIO TROl/SEAIIlEC

Dr. W. Soegeng Reksodihardjo

Deputy Director of BIOTROP,

Bogor (Indonesia).

Food and Agriculture Organization of the United Nations

Dr. V. S. Subramanian

FAO Expert, Bogor (Indonesia).

Calten IaciJic Indonesia

Dr. T. C. J. Zwartkruis

Head, Regional Geologist,

c/o CPI
Kebon Sirih 52, Djakarta (Indonesia).

Unesco

Dr. F. Fournier

Consultant,
Natural Resources Research Division,
Environmental Sciences and Natural Resources Research
Unesco, Place de Fontenoy, 75 Paris 7e (France).

Mr. F.J. C. Pala

Deputy Director,
Unesco Field Science Office for South-East Asia,
Djl. I m a m Bondjol 30,
P.O. Box 273/DKT, Djakarta (Indonesia).

BIr. Pita Niramaya

Senior Administrative Secretary,

Unesco Field Science Office for South-East Asia,
Djl. I m a m Bondjol 30, Djakarta (Indonesia).

149