141
Natural resources research XI
Titles in this series :
I. I.
II.
A review of the natural resources of the African continent. Enquête sur les ressources naturelles du continent africain.
Bibliography of African hydrology / Bibliographie hydrologique africaine, by J. Rodier. III. Carte géologique de l'Afrique (1/5 O00 000). Notice explicative / Geological map of Africa (1/5 O00 000). Explanatory note, by
R. Furon and J. Lombard. Review of research on laterites, by R. Maignien., Compte rendu de recherches sur les latérites, par R. Maignien.
IV. IV.
Y. Functioning of terrestrial ecosystems at the primary production level. Proceedings of the Copenhagen symposium/ Fonctionnement des écosystèmes terrestres au niveau de la production primaire. Actes du colloque de Copenhague. Edited by F. E. Eckardt.
Aerial surveys and integrated studies. Proceedings of the Toulouse conference / Exploration aérienne et études intégrées. Actes de la conférence de Toulouse.
VII. Agroclimatological methods. Proceedings of the Reading symposium / Méthodes agroclimatologiques. Actes du colloque de Reading. VIII. Proceedings of the symposium on the granites of West Africa
IX. IX.
x..
X.
VI.
Compte rendu du colloque sur les granites de l'ouest africain.
Soil biology. Reviews of research. Biologie des sols. Comptes rendus de recherches.
Use and conservation of the biosphere. Proceedings of the intergovernmental conference of experts on the scientijc basis for rational use and conservation of the resources of the biosphere, Paris, 4-13 September 1968. Utilisation et conservation de la biosphère. Actes de la Conférence intergouvernementale d'experts sur les bases scienti$ques de l'utilisation rationnelle et de la conservation des ressources de la biosphère, Paris 4-13 septembre 1968.
Soils and tropical weathering. Procedings of the Bandung symposium, 16-23 November 1969. XI.
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., Liège (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 make a synthesis of such knowledge; and to help Member States in establishing and developing research and teaching facilities. The scientific study of tropical soils including their
formation under tropical conditions is of vital impor- tance 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 under- standing 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 Unesco’s programme activities are concen- trated.
Since 1955, Unesco has maintained a special pro- gramme of scientific studies concerned with natural resources in humid tropical regions. Within this pro- gramme Unesco organized four symposia on Humid 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. The 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. The 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. The symposium was directed by Dr. F. Fournier, consultant to Unesco on natural resources research. The symposium was mainly devoted to soil and ferti-
lity questions. The scientific programme included the following topics : 1. The mechanisms of tropical weathering : decompo-
sition of rocks and pedogenesis; formation and eyolu- tion 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 : charac-
teristics 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 Indonesia’s 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 by 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 South- East 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 sympo- sium written and introduced by participants.
Authors of keynote papers were: Professor J. J. Fripiat, Laboratoire de Physico-Chimie
des Minéraux, Faculté d’Agronomie, Université de Louvain (Belgium).
Dr. J. van Schuylenborg, Laboratory of Regional Soil Science, Agricultural University, Wageningen (Netherlands).
Dr. P. Segalen, Services Scientifiques Centraux de l’ORSTOM, Bondy (France).
Dr. G. D. Sherman, College of Tropical Agriculture, University of Hawaii (United States of America).
The keynote papers and supporting papers are published in full in this volume together with the summary 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 co- operation 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 who contributed scientific papers.
The 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.
The 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 delimi- tations of the frontiers of any country or territory.
.
Contents . . . . .
I REVIEWS OF RESEARCH
J . J . Fripiat and A . J . Herbillon
P . Segalen
J . van Schuylenborgh
1 . Formation and transformation of clay minerals in tropical soils . . Introduction . . . . . . . . . . . . . The acid character of adsorbed water and the stability of adsorbed cations . . . . . . . . . . . . . . The ambiguous behaviour of aluminium The silica phase . . . . . . . . . . . . Transformation sequences . . . . . . . . . .
. . . . . . .
The genesis of clay minerals from primary silicates Transformation of 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 . . . . . . . . . . . . .
. . . .
. . .
. . . hydronium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15
15
15 17 17 18 18 19 19 20 21 21
Bibliography . . . . . . . . . . . . . . . . 22
2 . Metallic oxides and hydroxides in soils of the w a r m and humid areas of the world: formation. identification. evolution . . . . . . . 25 Introduction . . . . . . . . . . .
. Part one . . . . . . . . . . . . Origin of metallic oxides and hydroxides. rocks and minerals Weathering . . . . . . . . . . . The products of the soil . . . . . . . . Determination of oxides and hydroxides . . . . Inñuence of oxides and hydroxides on some soil properties
Part two . . . . . . . . . . . . Conditions of stability and mobility of metallic compounds Immobilization of oxides and hydroxides in soils . . . Summaryof theevolution of soils . . . . . .
Conclusion . . . . . . . . . . .
. . . . . 25
. . . . . 27
. . . . . 27
. . . . . 2 8 . '
. . . . . . 29' .
. . . . . . 30
. . . . . 32
. . . . . 32 ...
. . . . . . 32
. . . . . . 34
. . . . . 35
. . . . . . 36 .. I'
Bibliography . . . . . . . . . . . . . . . . . 37
. "
3 . Weathering and soil-forming processes in the tropics '39
Oxisols with plinthite . . . . . . . . . . . . . . 39
. . . . .
Theprofle . . . . . . . . . . . . . . . . 40 Theprocess . . . . . . . . . . . . . . . . 41
G . Donald Sherman
Oxisols without plinthite . . . . . . . . . . . . . 47 47
The process . . . . . . . . . . . . . . . . 49 Bibliography . . . . . . . . . . . . . . . . 49
An oxisolprofile without plinthite . . . . . . . . . . .
4 . Mineral weathering in relation to utilization of soils
Application to agronomic use . . . . . . . . . . . . 52 Relationship of lime requirement to mineral composition . . . . . . 52
stage of soils . . . . . . . . . . . . . . . . 54 Prediction of fertilizer needs according to stage of mineral weathering 56 The utilization of stages of mineral weathering in the location of mineral ores . . 56 The utilization of mineral weathering in identification of engineering properties of soils and earth foundations . . . . . . . . . . . . . 56 The utilization of mineral weathering stage in land-use classification . . . . 56
. . . . . 51
Introduction . . . . . . . . . . . . . . . . 51
The relationship of fertilizer requirement to mineral composition and weathering
. . .
Bibliography . . . . . . . . . . . . . . . . . 56
II TROPICAL W EATHERING IN ASIA
I.U. Ahmed. M . and M . R . K h a n
S . Hussain 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
S . V . Govipda Rajan and R.S . Murthy .
Introduction . . . . . . . . . . . . . . . . 61 Experimental . . . . . . . . . . . . . . . . 61 Methods of analysis . . . . . . . . . . . . . . 61
Results . . . . . . . . . . . . . . . . . 62 Soil properties . . . . . . . . . . . . . . . 62
Discussion . . . . . . . . . . . . . . . . 62
Bibliography . . . . . . . . . . . . . . . 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) . . . . : . . . . . .
Geology . . . . . . . . . . . . . . . . . Natural vegetation . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . Climate . . . . . . . . . . . . . . . . . Weathering processes . . . . . . . . . . . . . . Soils. their properties and composition . . . . . . . . . . Blacksoils . . . . . . . . . . . . . . . . Laterite . . . . . . . . . . . . . . . . Redsoils . . . . . . . . . . . . . . . .
Conclusion . . . . . . . . . . . . . . . .
65
65 65 68 68 69 69 70 70 71 71
Bibliography . . . . . . . . . . . . . . . . 72
K . V . S . Satyanarayana
S.V. Govinda Rajan and N . 11 . Datta Biswas
T . Seshagiri Ilao
H . Ling O n g
J . P . Andtiesse
Weathering of basic igneous rocks and genesis of clay minerals (specially '' presented as supporting paper to review No . 2 by P . Segalen) . . . .
.
73
Introduction . . . . . . . . . . . . . . . . Laterites of Malabar and South Kanara . . . . . . . . . . . Laterite and black soils of Malwa Plateau . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . . .
. .
Development of certain soils in the sub-tropical humid 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) . Introduction . . . . . . . . . . . . . . . . Physiography. relief and drainage . . . . . . . . . . . Geology . . . . . . . . . . . . . . . . Climate . . . . . . . . . . . . . . . . .
Experimental . . . . . . . . . . . . . . . . Results and discussion . . . . . . . . . . . . . .
Potential for land use . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . .
Pedogenesis of some major soil groups in Mysore State. India (specially presented as supporting paper to review No . 2 by P . Segalen) . . . .
Introduction . . . . . . . . . . . . . . . . . Black cotton soils . . . . . . . . . . . . . . . . Redsoils . . . . . . . . . . . . . . . . . . Lateritic soils (Latosols) . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . .
Correlation between chemical bond and chemical weatherability of the common rock-forming minerals (specially presented as supporting paper to review N o . 2 by P . Segalen) . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . Theory of chemical weathering . . . . . . . . . . . . Chemical weathering of the c o m m o n rock-forming minerals . . . . . . Summary and conclusions . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . .
The influence of the nature of parent rock on soil formation under similar atmospheric climates (specially presented as supporting paper to review No . 2 by J . van Schuylenborgh) . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . Climate . . . . . . . . . . . . . . . . . Altitude . . . . . . . . . . . . . . . . . . Parent material . . . . . . . . . . . . . . . Soils . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . Appendix . . . . . . . . . . . . . . . . . Bibliography . . . . . . . . . . . . . . . .
73 73 74
75
77
77 77 77 78 78 80 80
84
85
85 85 86 86
87
89
89 90 91 .93
93
95 95 96 96 96 96 98 101
102
110
V.M. Fridland
T a n Boun Suy
.
E.V. Tamesis and D . C . Salita
Ng Siew Kee and L a w Wei Min
The differences between crusts of weathering and soils developing on acid and basic rocks in the humid tropics (specially presented as supporting paper to review No . 3 by J . van Schuylenborgh) . . . i . . . 111 Bibliography . . . . . . . . . . . . . . . . 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 Introduction . . . . . . . . . . . . . . . . 117 Parent materials . . . . . . . . . . . . . . . 117 Weathering of the parent materials . . . . . . . . . . . 118 Physical agents . . . . . . . . . . . . . . . 118 Chemical agents . . . . . . . . . . . . . . . . 118
Characters of red basaltic soih~ . . . . . . . . . . . . 121 Physical characters . . . . . . . . . . . . . . 121 Physico-chemical characters . . . . . . . . . . . . 121 Cover plants and mulching . . . . . . . . . . . . 121
Conclusion . . . . . . . . . . . . . . . . 122 Acknowledgement . . . . . . . . . . . . . . . 122 Appendix . . . . . . . . . . . . . . . . . 122 Bibliography . . . . . . . . . . . . . . . . 123
Some 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 . . . . . . . . . . . . . . . . 125 Geographic location . . . . . . . . . . . . . . 125 Climate . . . . . . . . . . . . . . . . . 125 Relief and vegetation . . . . . . . . . . . . . . 125 Soil-forming rocks . . . . . . . . . . . . . . . 126 Morphological and chemical features . . . . . . . . . . . . 126 Discussion . . . . . . . . . . . . . . . . 127
Bibliography . . . . . . . . . . . . . . . . 127
Pedogenesis and soil fertility in West Malaysia (specially presented as supporting paper to review N o . 4 by G . Donald Sherman) . . . . 129 Introduction . . . Factors of soil formation Climate . . . . Parent materials . . Vegetation . . . Relief . . . .
Great Soil Groups . . Latosols . . . Reddish-brown lateritics Red-yellow podzolics . Yellow-grey podzolics Laterites . . . Gleys . . . . Alluvial soils . . Podzols . . . . Organic soils . . 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 . . ‘Reserve’ indice8 . . . . .
Differentiation below Great Soil Croup level
Future development and physical factors .
Availability indices . . . . . Efficiency of reparation . . . . .
Crop response to manuring . . . . Acknowledgement . . . . . . Bibliography . . . . . . .
. . . . . . . . . . 132
. . . . . . . . . 133 .
. . . . . . . . . 133:
. . . . . . . . . 134
. . . . . . . . . 135
. . . . . . . . . 135 .
. . . . . . . . . 136
. . . . . . . . . 138
. . . . . . . . . . 138
. . . . . . . . . . . 139
in FINAL REPORT OF THE SYMPOSIUM
Final report of the rympoiium . . . . . . . . . . . . 143
LISTOFPARTICIPANTS . . . . . . . . . . . . . . . . . . . . . . . 147
. . b ... . . .,
I Reviews of research
1 Formation and transformations of clay minerals in tropical soils
INTRODUCTION
It is well known 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. The 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. When rocks of different mineralogical compo- sitions 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 modi- fications provoked by the hydration processes.
Elementary chemistry tells us that water is an associated liquid, the molecules having a strong ten- dency to form hydrogen bonds with each other. This accounts for the high dielectric constant and the high
. boiling point. The 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 under-
going the following autodissociation process :
J. J. Fripiat' and A. J. Herbillon' Laboratoire de Physico-Chimie Minérale, Institut dea Sciences de la Terre, 42 de Croylaan, Heverlee-Louvain (Belgium)
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 weak acid.
It is appropriate to report here also the recent dis- covery (Bellamy et al., 1969; Wyllis et al., 1969) of anomalous water, prepared by 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 maximum density at 4oC as does normal water. It is much 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. The 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 CHARACTER OF ADSORBED WATER AND THE STABILITY. OF ADSORBED 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. Some of the water molecules belonging to the first layer hydrating the surface are in close contact with this positive species: They are therefore
.
1. The Univeriity of Louvain and YRAC. Tervuren.
J. J. Fripiat and A. J. Herbillon
strongly polarized, wiLh the consequence that the pro- ton 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 known for acid cations in aqueous solutions : I A13+6H20 I e I A12+OH,5H,0 I + H+ (pK E 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 on silica gels and montmo- 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 I H+/H,O I 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 concen- trated acid solutions around rock constituents. In
Formula 1. H,O+
\ /Al(-)\ / o \ / o / si \ -f
o
O O 0 O
(4 Formula 2.
N H1+ O O O
3 \ /\/ si
O /"'(-)\ O 0 / \ O (al)
\ /OHZ O
AI
O /\
O
(b)
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 sur- faces 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 alumi- nium clay (Eeckman and Laudelout, 1961).
The behaviour of surface I H,O+ 1 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-balancing cation
is supposed to be II,O+. The (a) configuration is not stable : the Al-O-Si bond is broken and the lone pair of electrons of H,O combines with the free Al orbital leading to the (b) configuration and, on drying, to the (c) state. The resulting silanol group has been observed by infra-red spectroscopy (Basila, 1961) in silico-alumina. The 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 100-2000 C. The (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 by step or at once: no direct information is available, but this is clearly the mechanism by which the hydro- lyzing 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.
,
H O O
-f \/ Si
O / \o
O H O \/ \ / O /. \oo/ \ AI Si
O O
(4
O HO
+NHw \/ \ / O si
O O 0
(4 16
Formation and transformations of clay minerals in tropical soils
T II E A M B I G U O US B EH A V I O UR OF ALUMINIUM
According to the pH, aluminium in solution behaves as AF+ or as a hydroxide, Al(OH),, or as a weak 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) such as Al,(OII) 'it, or Al,(OH) 8' and Al,(OII) '1;. In these complexes, aluminium is mostly sixfold co-
ordinated, whilst in aluminates, aluminium is reputed to be mainly fourfold co-ordinated. Hsu (1968) has shown that polynuclear complexes m a y be formed on the surface of clay minerals such as montmorillonite by treating the aluminium-saturated clay by an alkaline solution. According to the structure of the solid phase, alumi-
nium 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 alumino- silicates. 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 Kimpe (1961) have shown that in the amorphous silico-aluminas obtained by co-precipitating alumina and silica, the co-ordination number of alumi- nium depends on the pH of the suspension. In acid conditions, sixfold co-ordinated aluminium is dominant, whilst in an alcaline medium the fourfold co-ordinated species is more important. It is noteworthy that this observation is in agreement with the fact that kaolin minerals are known to appear in desaturated media whilst minerals in which aluminium is fourfold co- ordinated are formed under more alkaline conditions, as shown below.
Recently, the unstable character of fourfold co- ordinated aluminium structures under acid conditions has been clearly shown by 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 con-
tain AlIV and AlvI) have been transformed into anal- cime (which contains AlIV only) in the same temperature range in presence of sodium carbonate by D e Kimpe, Herbillon and Fripiat (1966). More recently Fran- kart (1969) has observed the transformation of mont- morillonite 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 number of aluminium in the new solid phase obtained from the co-precipitation procees with soluble silica depends on the pH of the environ- ment. 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 neutra- lized, whilst in tetrahedral co-ordination it is connected to oxygen and carries a negative charge. .
THE SILICA PHASE
The hydrolyzing action of water on primary materials liberates silicon as well as aluminium: silicon forms polymeric hydrated SiO,, the molecular weight of which depends firstly on the p H (6.0, 5.3)-silica is depoly- merized more at low pH than at high pH, but the m a x i m u m degree of polymerization seems to occur around PII 4-4.5. The 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/Si ratio.
The solubility of silica is very m u c h affected by the cations in solution. W e y and Siffert (1961) have shown the influence of Al3+, Mg2+, resulting in the co-precipita- tion of silico-aluminas and silico-magnesia phases. These results m a y be summarized by saying that the stability of silica suspénsions depends on the ionic environment and this will affect of course the movement of silica in the weathering solutions. According to D e Kimpe (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 attri- buted the mobility of silica in soil water to the protec- tive action of iron oxides, forming with soluble silica
17
J. J. Fripiat and A. J. IIerLillon
ferrisilicic complexes. Tran-Vinh-An and IIerbil- lon (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 much smaller- than 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 new 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. The 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 made 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 CLAY MINERALS F R O M PRIMARY SILICATES
Such a transformation has been schematically repre- sented by Keller (1964) as follows:
IMeAl silicates1 + HOH-t Men + OH- + Aln+ + Al(OH);++ Al(OII), + Al05 + Fe2+.,+ + Fe(OII), + II,SiO, + Si0 a- + MeSi+ + ’
j HMeA1 silicates (clay minerals) I 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 consi- dered. The first step is of course the water adsorption
process on the external surface of the primary particles.
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 Drind- ley (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 pro-
perties 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.
The 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 silicates’stability. 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/Al N 2) whilst the hydrogen forms of erionite and mordenite (Si/Al N 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 control- ling 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
18
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 observa- tions 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, he emphasizes the importance of the epitaxial growth of the neosynthetized mineral. In some way, these obscr- vations appear to contradict those reported by Del- vigne (1965) and Bates (1960), who have pointed out the presence of amorphous materials around altered feld- spars. The neoformation of kaolinite on the surface of mica flakes has been observed by Millot (1964), Tardy (1969), de Béthune et OZ. (personal communica- tion), without an intermediate amorphous phase, In summary, it seems 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 impor- tant role by ruling the solubility of the mineral consti- tuents 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.
TRANSFORMATION OF MICAS INTO CLAY M I N E RA L S
The 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 when mica flakes are much 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 consti- tution hydroxyls and hydration water.
In a dioctahedral mica such as muscovite K(A1,SiJ Al,O,,ûH,, 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 O€€ are 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 011 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 OH 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 Brown (1966) and by Robert and Pedro (1969). Once again, the relationship between the weathering mecha- nism 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 tri- octahedral micas has never been completely elucidated, nor has the real weathering processes of dioctahedral micas. Potassium complexing agents, such as triphenyl- boron (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 inter- mediate rectorite phase could be a very general pro- cess.
T H E TRANSFOBMATION OF ABIORPUOUS INTO CRY 9 TAL LIN E SE CON D A R Y M A T E RI ALS
The clay fraction of soils and especially of tropical soils always contains some amorphous materials but, with the attention focused on 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 equili- brium because of the dynamic character of the soil- forming processes. The 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-precipitation process, 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 charac- terized 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 corres- ponding crystalline phase.
Moreover the composition appears to have a deep influence on the yield in well-ordered structure. Co- precipitation of magnesia and silica for instance leads very easily to the formation of montmorillonite (Caillère 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 (De 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 acce- lerates to a very great extent the rate of crystallization. The reasons for this acceleration are very complex. For instance, according to D e Kimpe, Gastuche and Brindley (1964) the complete hydrothermal transforma- tion 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 observation means that any time a crystalline pbase is thermodynamically unstable, its transformation into a new structure through the combinátion with another constituent might be expected.
Poncelet and Brindley (1967) have observed that montmorillonite saturated by polynuclear aluminium cations m a y be quantitatively transformed into kaoli- nite around 1750 C. The 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 way similar to that described by D e Kimpe, Gastuche and Brindley (1964). Although these examples seem to suggest solid-phase transformations, the acid environ- ment does not exclude the dissolution-coprecipitation sequence discussed previously.
The fundamental question thus remains : m a y a disordered structure transform into the corresponding ordered lattice by keeping the same approximate compo- sition in a temperature domain and in a period of time acceptable in pedology? Au indicated above, chemical compositions are known
for which the answer is yes, but for the most frequently occurring system (silico-alumina), the experimental work performed so far, in many laboratories fails to provide
an answer. The question might be solved morc easily if we knew the growing mechanism of the clay micro- crystal.
It has h e n suggested that the octahedral layer patterns the fixation of the silica tetrahedral layer (Caillère 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 Kimpe and Fripiat (1968) suggested that kaolinite could grow in the ab plane, e.g., by a progres- sive arrangement of ‘rows’ of O-Si-O-Al-OH chains. 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 temp- erature is increased sufficiently. Determinations of activation energies in the synthesis process of clay minerals are rather scarce. For kaolinite, data obtained by Rayner (1961) seem 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 known on 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 many occasions that allo- phanes for instance are very similar in composition and structure to synthetic amorphous silico-aluminas. On the other hand, the evolution of allophane into halloy- site and kaolinite has been observed under various climatic conditions (Sieffermann and Millot, 1969).
TRANSFORMATION OF CLAY MINERALS INTO OTIIER CLAY MINERALS
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-
20
Formation and transformations of 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 known 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 con- tained 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 new mineral is a more convenient way 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 know 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.
TRANSFORM AT 1 ON S SILICATES
INTO TRI D I fi1 E N S I O N AL
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 ferro- magnesian 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, Van 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 when there is an excess of alkaline-earth cations and quartz when the medium 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 transfor- mation 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 way as pyroclastic products containing fourfold co-ordinated aluminium.
CONCLUSIONS
The title and the content of this review paper m a y appear as more or less in contradiction since no refer- ence has been made so far to the tropical soils, or more especially to the influence of tropical weathering condi- tions 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. The 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 many 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 coun- tries 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.
The desaturation degree of soils in many 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 two 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.
The weathering of micas occurs in a very diíi'erent way from that of fcltlspars. In the former case, trans- formations 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 redisso- lution.
Again these processes are very dependent on the amount of water and on 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 many 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 equili- brium states are reached.
Bibliography
AL RAWI, G. J.; SIJS, C. 1967. A comparative study between Euphrates and Tigris sediments in the mesopotanian flood plain. Pédologie (Gand), vol. 17, no. 2, p. 187-209.
BALDAR, N. A.; WHITTIG, L. D. 1968. Occurence and syn- thesis of soil zeolites. Proc. Soil Sci. Soc. Amer., vol. 32,
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-
BONIFAS, M. 1959. Contribution à l'étude géochimique de l'altération latéritique. Além. Seru. Carte Géol. Als. Lor., vol. 17, 159 p.
p. 235-8.
tology.)
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 problème de Ia Synthese des minéraux phylliteux à basse température. Genèse et synthèse des argiles, p. 31-41. Paris, Coll. Int. CNRS. (No. 105.)
CLOOS, P.; HERBILLON, A.; ECIIEVERRU, J. 1968. Allophane 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. Pidogenèse en zone tropicale. I. La forma- tion des minéraux secondaires en milieu ferrallilique. Ph.D. Thesis, I,oiivain, Fac. Sciences, 177 p.
DEMOLON, A.; BASTISSE, E. M. 1938. Sur quelques complexes colloïdaux du fer et de l'aluminium. Ann. agron., Paris, vol. 1, p. 6-33.
-. , -. 1944. Rôle vecteur de la silice dans les phénomènes géochimiques et physiologiques. Ann. agron., Paris, vol. 1,
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 Musée Royal de l'Afrique Centrale, Tervuren, Belgium.)
dans les sois sodiques de la Basse Ruzizi (Burundi). BUZZ. Groupe Français des Argiles.
FREDERICKSON, A. F. 1951. Mechanism of weathering. Bull. Geol. Soc. Amer., vol. 62, p. 221-32.
FHIPIAT, J. 1961. Jn discussion of the Caillère et Ilénin paper. Genèse et synthèse des argiles, p. 41. Paris, Coll. Jnt. CNRS. (No. 105.)
p. 265-96.
-. , HERBILLON, A. (In press.) Présence et genèse d'analcime
22
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.
dynamic 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.,
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. La genèse des minéraux argileux de la famille du kaolin. I. Aspect colloïdal. Synthèse et genèse des argiles, p. 57-65. Paris, Coll. Int. CNRS. (No. 105.)
de la famille du kaolin. II. Aspect cristallin. Genèse et synlhèse des argiles, p. 67-74. Paris, Coll. Intern. CNRS. (No. 105.)
GOLDICH, S. S. 1938. A study in rock weathering. J. Geol.,
HERBILLON, A. J.; TRAN VINII AN, J. 1969. Heterogeneity in silicon iron mixed hydroxides. J. Soil Sci., vol. 20,
IIsu, Pa 110. 1968. Heterogeneity of montmorillonite surface and its effect on the nature of hydroxy aluminum inter- layers. 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 tétraèdrique et la synthèse 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.
-0 , 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 coordi- nation 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 Trans- lation.
LAGACHE, M. 1966. Synthèse de la Boehmite, de la kaolinite et de la muscovite par altération de l'albite par l'eau à ZOOo C en présence d'acide carbonique. Bull. Groupe Fran- çais des Argiles, vol. 17, p. 71-9.
-. , JELLI, A.; PONCELET, G.; ANDRE, J. 1965. Thermo-
VOI. 69, p. 2458-61.
--. , DE KIMPE, C. 1961. La genèse des minéraux argileux
VOI. 46, p. 17-23.
p. 223-35.
,
LOW, P.; WHITE, J. L. To be published in Science. MILLOT, G. 1964. Géologie des argiles. Paris, Masson. 499 p.
l'attapulgite dans les sols à carapaces calcaires de la Basse Moulouya (Maroc Oriental). C.K. Acad. Sci. Paris, vol. 268,
MORTLAND, RI. M.; FRIPIAT, J. J.; CIIAUSSIDON, J.; UYTTER- IIOEVEN, J. 1963. Interaction between ammonia and the expanding lattices of montmorillonite and vermiculite. J. Phys. Chem., vol. 67, p. 248-58.
-- , PAQUET, H.; RUELLAN, A: 1969. Néoformation de
p. 2771-4.
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 tem- perature. 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. Genèse et synthèse 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 phénomènes d'oxydation et l'aptitude à l'ouverture des micas trioctaédriques. 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 Français des Argiles, vol. 18,
SCOTT, A. D.; IIUNZIKER, R. R.; IIANWAY, J. J. 1960. Che- mical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenyl boron. Proc. Soil Science Soc. Amer., vol. 24, p. 191-4.
SHUTOV, V. D.; DRITS, V. A.; SAKIIAROV, ß. A. 1969. O n the mechanism of a postsedimentary transformation of mont- morillonite 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 Uni- versities Press.
STEVENS, R. E.; CARRON, XI. K. 1948. A simple field test for distinguishing minerals by alrasion PII. Amer. Min., vol. 33,
TARDY, Y. 1969. Géochimie des altérations. Etude des arènes et des eaux de quelques massifs cristallins d'Europe et d'Afri- que. Ph.D. Thesis, Univ. Strasbourg, 272 p.
TCHOUBAR, C. 1965. Formation de kaolinite à partir d'albite altérée à 200° C. Etude en microscopie et diffraction électro- niques. Bull. Soc. Franç. Min. Crist., vol. LXXVIII, p. 483- 518.
-; OBERLIN, A. 1963. Altération de l'albite par action de l'eau. Etude en microscopie et microdiffraction électroniques de la précipitation et de l'évolution des fibres de boehmite formée. J. de Microscopie, vol. 2, p. 415-32.
TOUILLAUX, R.; FRIPIAT, J. J. 1966. Caractérisation des micas par résonance magnétique protonique. Magnetic resonance and relaxation. Actes XIVíh Coll. Ampère, p. 425-29. Ljubjana, Blinc.
TRAN VINII AN, J.; HERBILLON, A. 1966. Etude expérimen- tale du système SiO,-Fe,Ol. Applications à la pédogenèse. Trans. Conf. on Mediterranean Soils, Madrid, p. 255-68. Sociedad Española de Ciencia del Suelo.
TRICHET, J. 1969. Contribution à l'étude de l'altération expéri- mentale des verres volcaniques. Ph.D. Thesis, Paris, 232 p.
p. 3-9.
p. 31-49.
23
J. J. Fripiat and'A. J. IIerbillon
UYTTERIIOEVEN, J. D.: JACOBS, P.; MAKAY, K.; SCIIOON- PEYDT. R. 1968. The thermal stability of hydroxyl groups in decationated zeolites X and Y. J. Phys. Chem., vol. 72,
VAN DEN HEUVEL. R. C. 1966. The occurence of sepiolite and attapulgite in the calcareous zone of a soil near Las Cruces (New Mexico). Clayi and clay minerals, Proc, 14th Conf.,
WALKER, G. F. 1949. The decomposition of biotite in soila.
p. 1768-75.
p. 193-207.
Min. Mag., vol. 28, p. 693-793.
WENTWORTH. S. A. 1967. A n inuestigaiion offine grained micas with emphasis on their hydrous charucter. Ph.D. Thesis, Pennsylvania State University, 89 p.
WEY, R.; SIFFERT, B. 1961. Réactions de la silice mono- moléculaire avec les ions Al3+ et Mg*+. Genèse et synthèse des argiles, p. 1-23. Paris, Coll. Int. CNRS. (No. 105.)
WHITE, J. L.; BURNS, A. F. 1963. Infrared spectra of hydro- nium ion in micaceous minerals. Science, vol. 141, p. 800-1.
WYLLIS, E.: RENNIE, G. K.; SMART, C.; PETIIICA, B. A. 1969. Anomalous water. Nature (Lond.), vol. 222, p. 159-61.
o .
24
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 IIädrich, 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 régimes 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 warm; 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 up in the northern part of Africa by the Sahara, presents no interruption in south- ern Africa or in South-East Asia, for instance, where climatic conditions change gradually from one type to another. The distribution of these zones is given in Figure 1.
One 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 warm currents. The 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. Con- sequently, these differences have important effects on 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 make up 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 known 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 Amazon 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 Dekkan and the Ganges plain. All these plateaux are often bevelled through eruptive or metamorphic rocks and
25
P. Segalen
*I ,
.>
I I I I I I I I
*i
26
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 uncommon 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. On 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 Quater- nary glaciations, nothing similar can be seen in this area. No 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 on for ages on materials submitted to drastic climatic conditions.
The 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 down 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 percep- tible tinges due to gradual climatic changes up to the deserts where all vegetation comes to an end. The primary woody formations are very often associated with secondary ones such as savannahs, steppes, maquis and so on, which are due to human 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 hydrox- ides 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 accu- mulated. The presence of ferruginous products is a feature c o m m o n to the whole zone. But impor- tant 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 encoun- tered but no other three-layered clay mineral. All the ferruginous materials 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). Ferru- ginous 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 some of the above-mentioned soils.
PART ONE
After an outline of the general distribution of hydroxides and oxides in soils and the climate, parent-rock, geo- morphology, 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 them and their weathering, and next describe the main crystallized and amorphous minerals and the w a y to determine them.
ORIGIN OF METALLIC OXIDES A N D IIYDROXIDES, ROCKS A N D MINERALS
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 an actual pedogenetic interest and are met all over the world will be taken into account here.
Igneous rocks are made up essentially of iron, alu- minium, magnesium, potassium, sodium and calcium silicates in various proportions and in different struc- tures. Some other minerals containing the above- mentioned 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. They can form new 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 main constituents of rocks and ‘secondary’ minerals those which derive from the former under weathering pro- cesses.
Amongst primary miiierals, let us mention : peridots, garnets, pyroxenes, amphiboles, micas and I feldspars. Their structure is quite elaborate. The 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.
27
I’. Segalen
Peridots are magnesium and ferrous iron ortlio- silicates. Silicon tetrahedra do not join by a summit but are connected through hexacoordinated mag- 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 made of silicon tetra- hedra joining through one or two apexes, to constitute single or double lines. All the tetrahedra are set out the same 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 alka- line metals, magnesium and silica when drainage conditions are good. Residual iron accumulates as oxides or hydroxides. When 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. One 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 mate- rial. 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 he 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
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 much 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 ferru- ginous products can originate only from non-quartzose grains and these exist in small amounts (few ferro- magnesians, 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 away 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 (sur- roundings 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 forma- tion of these bauxites. Some authors state that alu- minium 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 expla- nation has been refused by many people. Some 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 away (through the action of running waters or wind) and trapped it on limestone. Each case must be examined carefully, as there is probably no 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.
The principal process is hydrolysis. Water contains not only IIOII molecules but 011- and II+ ions,’ following the dissociation of a small number of mole-
1. Or, better, IIsO+.
28
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 away of alkaline or earth-alkaline ions. In even more minerals, these metais are replaced by hydrogens. Where rather large-sized ions are replaced by much smaller ones, structures are weakened and tend to collapse. The Si-O-Al linking is affected, leading to the sepa-
ration of aluminium and silicon. If the latter is carried away, aluminium hydroxide precipitates. This happens for feldspars, but white micas are much less concerned as they are resistant to Weathering. Clay minerals, and especially three-layered ones, can also break down 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 move 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 crystal- lizes directly or through an amorphous phase. In medium drainage conditions, aluminium m a y combine anew with silica to form clay minerals ; two-layered minerals if there is not too much silica and bases; three-layered minerals if silica and bases are abundant. In many 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 medium is a reducing one, by absence of oxygen or especially by the introduction of organic material, the ferrous ion m a y subsist up 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 known 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 alu- minium. Under poor drainage conditions, these organic substances m a y be considered as responsible for the movement of iron in many soils. (Alexandrova, 1954; Uetremieux, 1951 ; Uloomfield, 1956; Schatz, 1954.)
TIIE PRODUCTS OF TIIE SOIL
Soil sesquioxides m a y be divided into two main cate- gories : amorphous and crystallized products. Both exist in the soils under study.
Amorphous products
A definition of amorphous products is necessary. They do not have a definite formula. Various names have been proposed such as cliachite or alumogel for alumi- nium, stilpnosiderite for iron, doelterite for titaninium, vernadite for manganese, etc. They are variously hydrated and this water is easily heated away at low temperatures. Their specific area is usually high (300-400 m2/g for amorphous, ten times less for crys- tallized material). As there is no crystalline, system, X-ray diffraction provides only very weak peaks, and no species can be identified. Amorphous products are very seldom alone in soils. Some soils derived from basic rocks contain up to 25 or 40 per cent amorphous material.
.
Crystallized material
Several types are known such as hydroxides, M(OH), or MOOH 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 DO 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 known 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. The 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 diíkrent iron hydroxides with the same formula of FeOOH : goethite and lepido- croci te. Goethite is a reddish-yellow mineral, 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.
The M203 oxides, or sesquioxides, are represented by corundum A120, and hematite and maghemite Fe20,. Corundum' exists in soils only when inherited from metamorphic rocks. Hematite is commonplace in trop- ical soils. It is red, non-magnetic and can be found in soils derived from volcanic rocks, in'.areas where 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 Fe,O,, which has a spinel structure, must be considered as inherited from basic parent material, but not as a soil-formed oxide.
Signijîcance of hydroxides in soils
Gibbsite is fairly widespread in equatorial and humid * tropical areas; it is sometimes encountered in the
Mediterranean area. It is uncommon 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 on broken land, even when 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 labo- ratory 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 much 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.
Amorphous materials are present in many soils. Very oîien 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 sesqui- oxides 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. Some of them behave the same way when 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. The use of this mixture was gene- ralized, at least for the soils of tropical areas. Un- weathered primary minerals, like quartz, m a y be separated from most soil-formed minerals. This proce- dure 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 know 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 deter- mined 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 much as far as identification of constituents is concerned. That is why methods leading to the selective identification of pro- ducts 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 made to determine it quickly. Dissolution of alumina in sodium hydroxide, since the oxide is ampho- teric, has been recommended. In addition alumina
30
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 quan- titative determination is not possible. The determination of free iron oxides has given ribe
to much 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. The 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 known that it does dissolve all free iron in soils like podzols, but not in tropical soils. Drosdoff and Truog (1935) recommend hydrogen sulphide, much 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). The 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 under- gone 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 tem- perature, water held between clay layers is driven off. The second goes off at a definite temperature and is related with the constitution of molecules. The 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 thermo- couples is put in the unknown product. During the
heating, endothermic reactions (absorption of heat necessary to hive away 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 tlev- eloped during the reactions, is recorded on a graph which permits the identification of the constituents. Water losses of goethite and gibbsite give rise to endo- thermic 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-ray diflraction is the safest method to idcntify soil oxides or hydroxides. The diffraction of a beam of X-rays by a powder containing the materials pro- duces a diagram in which a number of distinctive peaks can be recognized. However, where several minerals arc present, peaks m a p overlap and identi- fication 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 diíììculty arises. But hematite and maghemite are not so easy to recog- nize 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.
The identification of constituents by means of microscopic methods is unequally successful. The 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 develop- ment 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 crystal- lized products which had been obliterated can appear (Tardy and Gac, 1968). A suitable process can be fol- lowed to dissolve and determine amorphous material. Tamm’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 proce- dure.
INFLUENCE OF OXIDES A N D IIYDROXIDES O N S O M E SOIL PROPERTIES
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. The 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 when clay minerals alone are present. Yellow soils do not change in colour when one tries to dissolve away amorphous material. This leads to the idea that they do not contain any; and in such a case, goethite alone prevails. Dark colours m a y be due to abundant magne- tite. 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 oxida- tion. Aluminous and titanous material only induce light colours. Iron-poor bauxites are pink or greyish.
Structures. Many 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 known 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 much changed; but that coarse polyhedric forms are somewhat affected. However, the formation of pseudo-sands, made 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 by 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
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.
CONDITIONS OF STABILITY A N D MODILITY OF METALLIC COMPOUNDS
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
The 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 investiga- tion 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 much larger : 0.9 to 1.3 A; and metalloids such as C, Si, N, are much smaller : 0.2 to 0.4 A. The oxides of these metals depend on the ratio:
cation radius = oxygen radius
This ratio enables us to predict the co-ordination number of the metal, and the sort of structure it will fit into. The elements we are concerned with fit in octahedra Mn2+, Fez+, Fe3+, Ti4+; Al3+ fits in octahedra and in tetrahedra (Table 1).
TABLE 1. Size of ions of principal elements and co-ordination number'
Co-ordination numtirr 'Rc
Ion R (in A) = - Predicted Found
K+ 1.33 0.95 12 j 8-12 Caa+ 0.99 0.71 8 6-8 Na+ 0.97 0.69 8 6-8
MnZ+ 0.80 0.57 6 6 Fez+ 0.74 0.53 6 6 Ti4+ 0.68 0.49 6 6 Fe3+ 0.64 0.046 6 6 Ala+ 0.51 0.36 6 4-6
Si4+ 0.42 0.30 4 4
1. 1Iexagonal prism is siable when p 6~ 1.0 Cube priim is stable when Octahedral prism is stable when Tetrahedral prism is stable when
P 6~ 0.78 p w 0.41 p w 0.22.
32
Metallic oxides and hydroxides in soils of the warm and humid areas
Concentration
-8 10
M 1Õ'
M IÖ'
2
AI^+
I I
4 6 PH
Concentration
-2 M 10
I ö3 1 ó5
Fe 2+
6 8
M lo2
1 ö5 3 4 5 6 7 8 91011 pH
-2 M10.
lea
MI?+
Concentration I .
Ti4
FIG. 2. Stability areas of ions and hydroxides (after Charlot).
-0.6 -
I h I 8 9 10 PH
7 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 among the elements of low solubility (2 to 7); whereas soluble cations have a very low potential (< 2.0) and complex anions have a much higher one (9-40). All the hydroxides 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 by Herbillon and Gastu- che (1962), shows that when iron or aluminum hydrox- ides precipitate, interesting developments take place, if at the same time, one proceeds to an effective dialysis. An amorphous product precipitates first, followed soon by a depolymerization which looks very much like a solution; little by little a crystalline phase (trihydroxide) appears. This can be related with what happens in some 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 some 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 per- manent 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 man- ganese behave in very much 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+ + 1/2 II, 4 Fe2+ + II+.
(1)
(2)
Complexation and chelation
Some elements can belong to bulky molecules where they no longer have the same properties as they had when 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). The newly formed anion is liable to move in pH condi- tions when hydroxides usually precipitate. The syn- thesis 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 warm and humid arca, where mineraliza- tion of organic matter is rather fast, chelating agents are not liable to exist very long. So their influence must be taken into account only when one can show they exist.
Lastly, one must consider the possibility of displace- ment of very minute particles, by a simple mechanical washing away, The size of particles of oxides and hydroxides is very small (less than one micron), much smaller than the smallest pore space of soils. But they are very often bound to the clay minerals and move along with them. Migrations can take place inside or outside the profile.
IMMOBILIZATION OF OXIDES A N D HYDROXIDES IN SOILS
I shall try in this section to explain how, after having been solubilized as ions, metals assume one of the various aspects known in soils.
First of all, what happens to aluminum, iron, tita- nium, manganese, when the primary minerals which contain them are broken down by 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. The metals take their part in the synthesis of new
products : clay minerals such as kaolinite or halloysite (aluminium), montmorillonite (aluminum and iron), depending on 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 tita- nium 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 intro- duction in the profile of compounds coming from elsewhere. This concerns iron and manganese which are very sensitive to variations of redox potential (D’IIoore, 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 on electron micro- scope photographs.
Local concentrations of ferruginous or aluminous material occur with or without concentric structures. One 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 embody 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 much discussion and probably varies from one sample to another. They exist as nodules, concretions, accumu- lations and so on. Small concentrations of sand (quartz or magnetite), of well-crystallized clay minerals (kaoli- nite) exist, bound together with amorphous products; they are named ‘pseudo-sands ’. The usual mechanical analysis techniques are unable to disperse these small aggregates, so the sand fraction is unusually high. The use of ultrasonic vibrations is necessary to break them down. These accumulations can have a much larger extent.
The 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. They m a y be platy, vesicular, mas- sive, etc., and result in binding of coarse sand grains, quartz pebbles, rock or cuirasse residues, by oxides and hydroxides. Sometimes their aspect is conglo- meratic. Their composition is either dominantly alumi- nous (bauxitic) or ferruginous. Sometimes, the content in manganese is high, as well as in titanium. The mineralogical composition is the same as in the non- indurated 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 on flat surfaces, but not in moun- tainous areas or even on 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 some countries of west Africa for instance, as well as the subtropical and mediterranean climates do 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 forma- tion 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 OF THE EVOLUTION OF 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 on clay minerals, and binding of particles in pseudo-sands is effective; what then about concre- tions and laterite ? The 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 ? The answer to this latter question is undoubtedly positive. But the introduction of the oxides in a new pedological cycle can happen in two different ways. The 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 pos- sible 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 ‘cuirasse’ is 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 down to the surrounding gentle slopes wherc it contributes to the formation of new laterite. But it is necessary to locate an ill-drained area where local hydromorphy can allow the aceumu- lation of organic matter liable to reduce an complex the iron. But 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 down by gravity and not by solution.
Breaking down of laterite
Iron crusts are often limited by ledges which overhang the surrounding slope by 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 by the fine soil under the crust. In tropical areas, vegetation seems able to bring to a standstill all the fine earth, which thickens down 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. Diffe- rences 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 new sheet of laterite, made of pieces of the former and bound together by iron oxides, is obtained. The first, due to erosional processes, m a y disappear completely. The second, when a new water table settles may, in turn, appear above the surrounding plains. One can be surprised to find, at some height, laterite formed from debris of a sheet which has been completely destroyed by erosion. The ferruginous material m a y serve for several laterite caps and seems to step down 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 Guyana (Bleackeley, 1964), probably have a similar origin. Flat bauxitic plateaux located inland have been broken down 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 impor- tant. 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 preci- pitates. The influence of micro-organisms during the precipitation of hydroxides of iron and manganese has often been pointed out. Various bacteria and fungi are
considered to be responsible for concretions and hard- pans 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
The 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 made 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 now possible in many fields of research as more accurate methods develop and more samples are analysed.
W e are now in a position to review briefly our know- ledge 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 tita- nium as well as manganese are locally abundant. Amorphous products are seldom found in yellow soils. Iron is always outside the clay minerals lattice essen- tially kaolinitic. Concentrations of any of the above- mentioned 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 common. Gibbsite is seldom found. Amorphous mate- erials 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 concre- tions, nodules, dots, etc. In neutral or alkaline soils, iron is often trapped by 211 clay minerals such as mont- morillonites (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 down the profile; however, mottled zones rich in concretions are known, as in Australia or the United
36
Metallic oxides and hydroxides in soils of the warm 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 uncommon. 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 montmoril- lonite). Concretions are small and scarce. No laterite is known.
W h at can be said about amounts? First of all therc is no definite relation between the colour and the amount. An 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 much 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). Else- where, 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 sub- tropical 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.
The contents of titanium are generally low as far as acid rocks are concerned. O n the other hand, soils deriving from volcanic rocks contain much more (5 to 10 per cent) with exceptionally high values in Hawaii (Tamura et al., 1953). The 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 much to learn about oxides and hydroxides in many a field. Even 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 when they are formed. The knowledge of these condi- tions in the past must be studied thoroughly, as we are sometimes liable to assign to the present what is actually inherited from the past. Much progress must be made 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, much 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 come 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 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 modalités de fixation de l’hydrate de fer colloïdal par l’argile du sol. Ann. Agron.,
BETREMIEUX. R. 1951. Etude expérimentale de l’évolution du fer et du manganèse 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-reduction of ferric oxide hy aqueous leaf extracts. The role of certain constituents of the extracts. C.R. VZih Znt. Congr. !%il Sci., B, p. 427- 33. Paris, AFES.
p. 8-34.
. .
CHARLOT, G. 1946. Théorie et méthode nouvelle d’analyse quali- 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.
D’IIooRE, J. 1954. L’accumulation 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-exchange properties 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 ‘Ferric hydroxide’ by kaolinites. J. Soil Sci., vol. 16, no. 2,
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. Geolo- gical Survey, Water Supply Paper 1459 A.)
HERBILLON, A.; GASTUCHE, M. C. 1962. Synthèse et genèse de l’hydrargillite. 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 minéralogie de Madagascar. Paris, Chdanel, vol. 3.
LAMOUROUX, M.; SEGALEN, P. 1969. Etude comparée des produits ferrugineux dans les sols rouges et bruns méditer- raneens du Liban. Science du Sol, vol. 1, p. 63-76.
LAPPARENT, J. de. 1930. Lea bauxites de la France méridio- nale. Paris, Impr. Nat. 187 p. (Mém. Carte géol. 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.
p. 334-41.
MAICNIEN, R. 1958. Le cuirassement des sols en Guinée, 239 p. (Além. Serv. Carte géol. Als.-Lorr., vol. 16.) - . 1966. Compte rendu de recherches sur les latérites. 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. Stabilité structurale et composition des sols. C.R. Acad. Agric. Fr., vol. 44, p. 799-803.
MILLOT, G. 1964. L a géologie dea argiles. Paris, Masson, 499 p. PAQUET, II. 1969. Evolution géochimique des minéraux argileux dans les altérations et les sols des climats méditerranéens et tropicaux à saisons contrastées. Thcse, Fac. Sc., Strasbourg, 348 p.
PEDRO, G. 1969. Principaux types d’altération chimique à la surface du globe. Présentation d‘une esquisse géographique. Rev. Géogr. Phys. et Géol. 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. had. 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 sér. Pédol., vol. III, no. 2, p. 149-76; no. 3, p. 179-205.
-. 1968. Note sur une méthode de détermination des pro- duits minéraux amorphes dans certains sols à hydroxydes tropicaux. Cah. ORSTOM sér. Pédol., vol. VI, no. 1, p. 105- 26.
TAMM, O. 1922. Eine methode zur Bestimmung der anorga- nischen komponenten der Gelkomplexen in Boden. Medd. für Statens Skogs., vol. 19, p. 385-404.
TARDY, Y.; GAC, Y. 1968. Minéraux argileux et vermiculite-Al dans quelques sols et arènes des Vosges. Hypothèses sur la néo-formation des minéraux à 14 A. Bull. Serv. Carte géol. 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 no 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 condi- tions, whereas in the latter, brownish and greyish soils come 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, on 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. The 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-forming processes 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. Never- theless, an attempt will be made 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. From 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
The processes involved are those leading to formation of the oxic horizon and those responsible for the genesis of plinthite. The former were called laterization or
ferrallitization. The 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. The latter are practically completely leached, whereas the released silica combines with alumina to form clay minerals, of which kaolinite is the final product. The remainder of the released silica is leached. This process will be called desilication, and described as above, it is an absolute impoverishment in silica and a residual enrich- ment 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. The 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 two types of desilication will be responsible for the formation of oxic material.
The 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 on the possibility of redistribution of products of desili- cation, 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 alter- nate 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
TIIE PROFILE 5. The rotten rock (zone V), where thc green-, yellow- and brown-coloured half-weathered minerals become visible. Its appearance varies of course with the type of parent rock.
The conclusion can be drawn that such a profile can only be formed under conditions of a fluctuating ground- water 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 com- 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. The reason for this absence could be that weathering has not yet proceeded so deeply as to form a water table, due to worsening drainage condi- tions 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 plin- thite (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 by Satyanarayana and Thomas (1962). The 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.
From 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 Van der Plas and V a n Schuylenborgh (1970) for the soil (Table 1). ..
It is hardly possible to describe a representative profile of Oxisol with plinthite because of the extremely wide variations. Nevertheless, an attempt is made here to give an idealized picture of a ‘most complete’ profile, developed on hard rock. After having studied the lite- rature 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 plinthíte 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, yellowish- red, 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. The plinthite zone (indicated as zone II). This horizon m a y be partly hardened. The 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 filled’with whitish, yeliowish, or greyish material. If moist, it can be dug out and hardens upon exposure to the au. The non-hardened part is extremely firm, the firmness decreasing with depth. The horizon merges diffusely into :
3. The mottled clay zone (zone III). It is a very pale yellow to white clay matrix with many 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 grad- ually more visible when merging into : ,
TABLE 1. Normative mineralogical composition of profile 1 (weight percentages)‘
Zone Q Or Pk Bi Ho Mt - Ms Ka01 Go + Hm Mise
I 28.9 - - - - - 10.2 28.3 27.7 5.0 11.1 23.6 -- - - - - - 2.4 56.4 14.1 3.6
V 15.9 12.5 1.5 - 0.5 - 46.0 - 12.5 11.0 1.5 VI’ 2.8 41.7 41.2 3.2 7.1 2.4
11.2/111 30.8 - - - - - 6.1 45.8 12.7 4.4
- - - E quartz; Or = K-feldspar; Plg - plagioclases; Bi - biotite; 110 P hornhlende; Mt = magnetite; ,Ils = muscovite (illite); Kaul = knoliiiite; 8 o i goethite; EIm E hematite; Misc - miscellaneous.
2. Hornblende-biotite variant of the katanorm.
40
Weathering and soil-forming processes in the tropics
TABLE 2. Normative mineralogical composition of profile 2 (weight percentages)'
Zone Q Kaol M s Go+ltm Iìu
1.1 1.2 11.1 11.2 11.3 11.4 III IV
90.1 3.4 0.4 6.1 0.3 88.3 4.2 0.4 6.8 0.4 64.5 10.2 - 25.1 0.2 45.0 10.8 - 43.8 0.3 11.3 10.0 0.3 77.7 0.7 20.8 25.9 1.1 51.4 0.8 38.2 51.3 2.2 6.4 1.9 48.9 44.3 3.8 1.2 1.8
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). The norma- tive 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 ground- water 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. The 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. The profile descriptions and the analytical details seem to indicate that the whole pro- cess can be divided into three stages. The 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 amor- phous sesquioxide colloids are raised and the base level of erosion approached, leading at a certain moment to the stage where the permeation of the soil with rain- waters 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. The second stage is therefore characterized by the occur- rence of weathering and soil-forming reactions under alternating reductive and oxidative conditions. The transformations during this stage are more or less comparable with those involved in pseudo-gley forma- tion 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.
Upon 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 moment on, under reducing conditions. There is still a fluctuating groundwater table, although thc highest water-level is 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 charac- teristics 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 liomoge- nized by biological activity, this too depending on the
41
J. van Schuylenborgh
type of parent material. h o t s penetrate into the sub- soil in the dry season and part of them will die in thc wet season because of excess of water and, eonse- quently, lack of oxygen; animals bury their tracks into the subsoil in the dry season and come to the surface in the wet season. Termites are particularly active, as they make passageways down to the level of the water table (Yakushev, 1968). In this way drainage paths for the rain-water are formed, leading to an inhomo- geneous wetting of the subsoil which contributes to the special characteristics of the plinthite material such as clay coatings along tracks and fissures.
From these considerations it 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. The 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 condi- lions. 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 com- 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 thépresent-day climate.
where brackets denote activities. The charges on 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
[KI [HI
log K, = 2 log - + 6 log [H,SiO,] (la
Similarly : 20r+2H++4HzO=Pyr+2K++2H,Si0,(2)
log K, = 2 log - [KI + 2 log [H,SiO,] (2a) [HI 2 ills + 2 H+ + 6 H,SiO, =
= 3 Pyr + 2 K+ + 12 H,O (3)
[KI [HI
log K, = 2 log - - 6 log [H,SiO,] (3a)
2 Ms + 2 H t + 3 H,O = 3 Ka01 + 2 K+ (4)
[KI [III
log K, = 2 log - Pyr + 5 H,O = Kaol + 2 H,SiO,
log K, = 2 log [H4Si0,] (5)
(54 2 ills + 2 Hf + 18 H,O =
= 3 Gibb + 2 K+ + 6 H,SiO, (6) Dedication
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 eo- workers (1961) and of Pedro (1964) are most famous. However, w e shall not discuss their experiments, but approach the problem from a thermodynamic view- point.
W e have seen that during the weathering of the rock of the first profile, micas (illite) were formed as an 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. The reactions are : 3 O r + 2 II+ + 12 II,O = MS + 2 K+ + 6 H,SiO, (i)
log K, = 2 log - [KI + 6 log [H4Si04] (6a) [Hl Kaol + 5 H,O = Gibb + 2 II,SiO,
log K, = 2 log [H,SiO,] (7)
(74
The symbols, used in the equations, have the following meaning : Or = K-feldspar, KAlSi,O,; M s =muscovite, KAl3Si,Ol~(0II),; Pyr = 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 (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 (298.150 K). 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 (lìeesman and Keller, 1968); AF; Kaol = -904.0
and AF; = AFOfprotlucts - AFJreactants
42
Weathering and fioil-forming processes 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-dimen- sional diagram with log[K]/[H] as one axis and log[H4Si0,] as the other. The 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*8 mol 1-l. At the right side of this line w e have therefore the field of K-feldspar + amorphous silica.
\
I I I I
amorphous silica saturation
Or
I I
-6 -.5 -4 -3 -2
log [ H, Si O,]
FIG. 1. Stability relations of some phases in the system K20-Al,0,-Si0,-II,0 at 25oC and 1 atm. total pressure (numbers at the boundary lines refer to equations, page 42). Gibb. = gibbsite; Kaol. = kaolinite; Pyr. = pyrophyllite; Ms. = muscovite.
The figure shows that if the H,SiO,-activity is kept low (and consequently also the activities of the dis- solved ions), e.g. when permeability is rapid and drain- nage perfect, the stable end-product of weathering is gibbsite. This is in agreement with the findings of Herbillon and Gastuche (1962), who stated that elimi- nation of extraneous ions promotes the formation of crystalline gibbsite. At higher H,SiO,-activities and in neutral or alkaline
media (high [K]/[H] ratios) as m a y occur under slow
permeability and poor drainage conditions, muscovite (illite) will be formed. A t lower [K]/[H] pyrophyllite (montmorillonite) can be formed. A t later stages, when the cations of the decomposing feldspars have been removed and when 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-I€20
at 25oC and 1 atm. was made 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 was 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
Gi.
-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-feld- spar; Kaol. = kaolinite; Ab. = albite; Gi. = gibbsite.
.
43
J. van Schuylenhorgh
t 1.0
t 0.8
-I 0.6
t 0.4
t 0.2
O
- 0.2
- 0.4
-0.6
- 0.8
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 pre- dominantly forms under conditions of poor drainage, a fact noticed by Schellmann (1964), w h o found mont- morillonite at a depth of approximately 8 metres below the surface in the weathering zone of a serpen- tinite and in many other situations which can be found in the literature.
The 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, made by Lovering (1959) with respect to
O 2 4 6 8 10 pH 12
FIG. 3. Stability fields of the stalle phases in the system Fe-O,-II,O at 250 C and 1 atm. pressure. Boundary between solids and dissolved species at C[Fe] =
the dedication process (see also Davis, 1964). H e suggested that desilication could be effected to a consi- derable 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, approxi- mately 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 by erosion, he was not able to calculate the amount of silica removed in this way relative to the amount of silica returned to the soil by decaying vegetation.
Plinthization
When 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 charac- teristics 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 move 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 moment that the mottled clay is transferred into the typical plinthite. Only weak reduction and, conse- quently, 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 tram- formation 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 trans- €ormation of thc rock into plinthite material might hc summarized from a chcmical viewpoint as follows :
44
Weathering and soil-forming processes 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 amor- phous 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 way the mineral aggregate changes into an intimate mixture of the mentioned amorphous anci cryptocrystalline products. All these products are metastable from a thermo- dynamic point of view and will be transformed (aged)
FIG. 4. Stability fields of the stable phases in the system Mn-û,-II,û 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) + H,O I
(goethite) I
(amorphous ferric- hydroxide)
Fe(OII), (a, s) + A100H(s) -+ I
(pseudo-boehmi te)
-+ a-FeOOH(s) + Al(OII), (s) I
(gibbsi te)
2Fe(OH), (a, s) + A100II (s) -+
--t a-Fe,O, (s) + Al(OII), (s) + II,O
(hematite) I
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. The silicic acid .may 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 upon 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. The 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 func- tions of Eh and pH, it should be concluded that there must always be a separation between iron- and man-
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 way to explain the mentioned simultaneous deposilion. One possibility could be the formation of intermediary, unstable iron hydroxides during the reduction-oxidation processes. 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. The results (Fig. 5) show that the field of predomi-
nance 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
Eh (VI
t 0.8
t 0.6
t 0.4
t 0.2
O
- 0.2
- 0.4
- 0.6
- 0.8
Fe3+ I '. + I '. I '. \ '.
CiO,lá,si I I _ _ - - \ \ \. \ \ \
\ \ t' - FeOOH \ t \ \ si02 (a;d \
'. \
'.
\
, \ \ \
'. , 1 Fe304 '\ +
SiO, (a.4 \
'. \ \
\. \ \
'.
'. \. '.,
Y-FeOOH
-. . .
Fe304
. FeC03 .\. '. '. '.
i 4 6 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.
frequently observed in paddy soils and pseudo-gley and gley soils (Douma 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., poly- phenols) was suggested by Bloomfield (1957, 1959) but little is known 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. The 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-all hydrolysis constant of the complexes, the higher the stability and the stronger the oxidation resistance. The relation
where KhML is the hydrolysis constant of the metal complex, ML; KML 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 KhML is smaller than 1, the complex is stable in neutral or weakly alkaline media; if larger than 1 it is stable only in acid media. The solubility products (calculated from free-energy values) of Fe(OH), and Mn(OH), are resp&tively, and lO-l3no. Hence, if KF~L is larger than the Fe(I1) complex will be stable and consequently resist oxida- tion. In the case of the complex MnL, KM^^ has to be larger than 1013.0. To explain the simultaneous preci- pitation 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 reduction- oxidation 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 poly- valent ions. Some of the Mn(1V) m a y he 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 counter- parts, 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.
46
Weathering and soil-forming processes in the tropics
A fourth possibility might le 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 inter- feres seriously in stability studies. For example, it is known 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 soil- forming processes is very useful, as it predicts the result of these processes, if the circumstances remain more or less constant.
with a thick layer of ferruginous material. Samples were taken every half metre up to and including the rock. A profile description was not made. From the analytical details and the too brief description of the geological samples, the following succession of horizons m a y be concluded:
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 movement and the structure is granular or subangular blocky and very stable. The soils are porous and have a rapid permea- bility. The 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. Man- 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 topo- graphical conditions.
A N OXISOL PROFILE 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; ow n investiga- tions on serpentinites in West 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 Kaliman- tan, Indonesia. During a geological investigation in 1956 by Gaertner and Wirtz (see Schellmann, 1964) it appeared that the Kukusan mountains were covered
Al 0-1 m Brownish friable material with some concretions.
BI 1-2 m Brown to yellowish brown friable material with very few concretions.
I), 2-5 m Yellowish Lrown to yellow friable material.
B, 5-6.5 m 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 serpenti-
nized periotite.
The profile is situated at an altitude of approximately 500 m above sea level. The rainfall varies monthly from 300 mm in January to 120 mm in September and is therefore very high. The mean annual temperature is approximately 240 C. The vegetation is a tropical rain forest. There is no dry season in the year : water move- ment 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 investiga- tions. The PII (II,O) varies from 5.0 in the A,- to 5.3 in the B3-horizon and increases from there on to 8.0 in the C,-horizon. The 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. From thereon it cannot be traced any more. The high value of exchange- able BZg in the C, and C1-horizon points to the pre- sence of a smectitic clay mineral, which was confirmed by mineralogical X-ray and DT-analyses.
The 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
TABLE 3. Normative mineralogical composition of prolile 6 (percentages given are by weight)'
Horizon S $111 Q M m Kaul Gi Go msC ?l /o % % % % % % Ser A w Mt % % %
- - - - - 13.0
76.8 43.6-
- - - 0.9 15.5 7.7 73.9 2.0 - - 1.3 6.3 17.5 71.4 3.5 - - - 2.8 1.6 17.4 74.3 4.0 - 18.3 - 3.9 1.7 3.8 68.0 4.3 - 15.5 5.8 17.1 - - 57.2 4.3 - 15.7 21.5 7.2 - - 39.6 3.0 3.9- 0- 0- 4.1 - - 9.6 0.7 O 7.2 5.8 -
-
1. Ser = serpentine; Aug augite; Mt = magnetitite; Sp E spinell; Chl= chlorite; Q = silica; Mn - montmorillonite; Kaul- kaulinite; Gi = gibhaite: 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. Some data were published by V a n Schuy- lenborgh (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. The results with respect to Si, Al, and Fe 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 much 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 assump- tion that the process acting in the surface soil differs from that in the subsoil, a suggestion already made earlier. Upon the impoverishment of the surface soil from nutrients because of prolonged leaching, the bacterial decomposition of organic matter (mineraliza- tion) slows down gradually; the consequence is that
TABLE 4. Mean composition of litter of rain forests on Java (Indonesia)
Forest on Si02 Ala02 Fen& SiOr/AlxOs Si02IFezO3 soil from : % Yo %
Andesite 2.51 1.15 0.20 3.8 42 Dacite 8.75 0.30 0.15 49 146
intermediate organic decomposition products come into play in soil formation. Some 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. Conse- quently 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 Pre- Eocene upheaval of the Kukusan mountains and it is therefore likely that the drainage conditions were poorer than at present; consequently, transformation of kaoli- nite 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 destruc- tion of the original minerals. Silica was leached and gibbsite and goethite accumulated. Upon 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 mont- morillonite 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 mont-
morillonite 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.
48
36
[MgZ' log--
CH'?
32
28
24
20
16
8 gibbrite
I -6
a - parent material b = profile 1 e i profile2 d = profile3 e = profile4
Tsukikiri f = unirrigated soil.
-
t a
\ tf
I I 1 -5 -4 -5 -2
lag:H., SiO, 1
FIG. 6. Stability fields of gibbsite, kaolinite, and montmo- rillonite in the system SiO,-Al,O,-Mg,-CaO-HZO at 250 C and 1 atm. total pressure. [Ca*+] = 5 x lo-,.
Weathering and soil-forming processes in the tropics
THE PROCESS
It is evident that plinthization is absent and that 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 zone is supersaturated with respect to silicic acid because of the high solubility (weatherability) of the serpentinite minerals and rather poor drainage conditions in the weathering rock. Upon better drainage in the layer above this zone, silicic acid is leached and montmoril- lonite becomes stable. At still better drainage conditions montmorillonite again becomes unstable and is trans- formed into kaolinite and gibbsite. These mineral trans- formations can be deduced easily from Figure 2.
Figure 6 gives a more exact stability diagram of montmorillonite, kaolinite and gibbsite. It is based on an investigation of Reesman and Keller (1968). Also this figure shows that upon stronger leaching, mont- morillonite is transformed into kaolinite and gibbsite,
Bib1 iography
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 musco- vite. US. Bur. Mines, R.I. 6356, U.S. Dept. of the Interior.
tion of gibbsite, kaolinite, halloysite, and dickite. US. Bur. Mines, R.I. 5825, U.S. Dept. of the Interior.
BENNETT, II. II.; ALLISON, R. J. 1928. The soils of Cuba. 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.)
-. , KELLEY, K. K. 1961. Heats and free-energies of forma-
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 pedo- genesis in New South Wales. IV. The ironstone soils. J i Soil Sci., vol. 4, p. 24-47.
HERBILLON, A.; GASTUCIIE, M. C. 1962. Synthèse et génèse 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 atmo- sphere. 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-energies of 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. 781- 800.
PEDRO, G. 1964. Contribution à l’étude expérimentale de l’alté- ration géochimique des roches crystallines. Thesis, Paris.
REESMAN, A. L.; KELLER, W. D. 1968. Aqueous solubility studies of high-alumina and clay minerals. Amsr. Min.,
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 thermo- dynamic properties. Nat. Bur. Standards (Circ. 500). US. Dept. of Commerce.
vol. 53, p. 929-42.
SCIIELLMAN, W. 1964. Zur lateritischen Verwitterung von Serpentinit. Geol. Jahrb., vol. 81, p. 645-79.
SCIIWERTMANN, U. 1959. über die Synthese definierter Eisen- oxide 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. Petro- chemical 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.,
-. 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.
vol. 5, p. 195-210.
50
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 decompo- sition.
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 condi- tions 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 pro- gressive 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 allophanic- halloysitic 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 same general mineral composition and similar physical and chemical properties. Their management for agricultural crop production could be based on 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 Humic Latosol (Kaisol) and the kaolinitic Terra Roxa 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.
The greatest expression of mineral decomposition and the formation of new secondary minerals occurs in the soils of the tropical regions. A considerable amount 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. Mohr (1944) has described the relationship between the
1. Emeritus Senior Professor of Soil Science of College of Tropical Agri- euiture. University of IIawaii. and Chairman of Division of Soils and Irrigation and Professor of Soils, American University of Beirut, Lebanon.
5 1.
G. Donald Sherman
internal movement 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. His 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 compo- sition which indicates the stage of weathering. It is in this field of soil mineral weathering that the soil scien- tists of the Hawaiian Islands have directed much 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, mont- morillonite 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 con-
ditions favouring rapid and almost complete decompo- sition of all primary minerals. The wide range of rainfall and climatic conditions has made it possible to study the entire range of soil mineral weathering and relate mineral composition to the physical and chemical pro- perties of the soil. Because of the intensive development of the soil management system in the Hawaiian Islands it has been possible to make 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. The 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 decom- pose. 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
:
montmorillonite formation, kaolinizalion and free- oxide formation. The free-oxide formation was refined to show peaks for ferruginous soils, titaniferous ferru- ginous 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 dev-
elopment of the oxides was established by 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 concen- tration of Fe,O, and Tio, increased and the concen- tration 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 mark- edly 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 The agriculturist has recognized that the physical properties of the soils of the tropical block earths ' (montmorillonite) and the L o w Humic 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 Humic Latosol (amor- phous 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 The early applications of lime to very acid tropical soils did not produce the anticipated beneficial results
52
Mineral weathering in relation to utilization of soils
based on experience in the temperate regions. Agri- culturists 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 on soils of the temperate regions where titration curves were based on an 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. Some 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 secon- dary clay or oxide mineral dominating in concentration and exerting the major influence on the physical and chemical properties of the soil. The 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 by 1.2.
The data given in Figure 1 indicate a dramatic
3 I 1 I I I
Tons Ca COj per acre
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. The soils of the indurated iron oxide surface horizon of the Humic 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 Humic Latosol the amount would be 8 to 10 times this amount or as m u c h as 45 metric tons per hectare. The soils of the IIydrol Humic 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 Humic 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 same 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 ammonium 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
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 Latosol- more than 50 per cent halloysite clay; C = Humic Latosol- less than 50 per cent halloysite clay + Al and Fe amorphous and crystalline oxides; D = Tropical Black Earth-Montmo- rillonite clay; E = Titaniferous Ferruginous Latosol-hydrous amorphous Te-Fe oxides; and F = IIydrol Humic Latosol- 90 per cent hydrous amorphous Al and Fe hydroxides and oxides.
53
C. Donald Sherman '
PH
12.0
11.0
10.0
9 .o
8.0
7.0
6.0
5.0
4.0
I l 1 I t 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 Low 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 = Waia- lua. The uniformity of curves make a pH-lime requirement prediction scale possible for the group.
133 pounds of ammonium sulphate per acre per month or 150 kilogrammes per hectare per month. These data indicate that with soils which owe their origin to intense chemical weathering, their lime requirement is deter- mined by the dominant secondary mineral. The kaolin rich soils will have their o w n PII lime requirement relationship as will the tropical black earths-montmoril-
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 when 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 when 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 Humic Latosol group.
THE HELATIONSHIP OF FERTILIZER REQUIREMENT TO MINERAL COMPOSITION AND WEATHERING STAGE OF SOILS
In both the tropical and temperate regions the efforts of farmers and agronomists have to a considerable extent been devoted to the development of soil manage- ment systems which would exploit with the m a x i m u m efficiency the nutrient elements released by soil minerals by chemical weathering. The 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. The application of fertilizer is a mcans to augment the mineral deficiencies produced by weather- ing or by the lack of weathering.
In the early stages of weathering mineral decompo- sition often releases adequate amounts of nutrients for the support of the growth of economic crops. In the early stages of weathering the decomposition of feld- spars 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. The alluvial soils, soils developed on relatively recent volcanic ash depositions or the loessal soils represent some of the most produc- tive 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
54
Mineral weathering in relation to utilization of fioil6
decomposition such as the feldspars, biotite and plagio- clase minerals. Hawkins and Graham (1950) and Gra- h 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 weather- ing 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) wh 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 maximum 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. The 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 cultiva- tion, 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 by the application of commercial fertilizer with greatest economic return. It is also the point of soil develop- ment 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. The high production yields reported by Mohr (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. Tamimi et al. (1963, 1964, 1967) and Liu et al. (1966) have shown that when phosphorus fertilizers are applied to these
soils along with ammonium and potassium fertilizer, metallic aluminium (ferric) phosphate complexes like tarakanite will form in the soil. The native vege- tation by 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. The former pro- duces the advanced or ultimate product of weathering according to Mohr (1944). Sherman (1955, 1958) con- siders the formation of bauxite to approach the same 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 agri- cultural 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 same fashion as iron-oxide soils but not have the same firmness. They have the same infertility. The 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 and 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 amendment applica- tions. Sherman et al. (1964), and Monteith and Sher- m 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, and an extremely high applicatiòn of phosphorus is needed to overcome the high fixation capacity and to provide an adequate level of available phosphorus. Younge and Plucknett (1966) applied as much as 1,000 pounds per acre of elemental phosphorus to these soils. T h e 1,000- pound 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 made 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 FElìTILIZER NEEDS ACCORDING TO STAGE OF MINERAL W EAT II E RING
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. The 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. The 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. The first factor in the deter- mination of the lime requirement is the mineral compo- sition of the soil. The increased fixation of phosphorus and the loss of potassium as weathering progresses is a reflection of the creation of new minerals and the loss of others. The author, while not including the hydrous mica in this paper, is aware of their role in the main- tenance of soil fertility in soils in some of the stages of advanced weathering.
T H E UTILIZATION OF STAGES OF MINERAL WEATHERING IN THE LOCATION OF MINERAL ORES
T h e development of the majority of the bauxite deposits of the world owe their origin to the mineral weathering on the earth’s surface. The conditions for the development of bauxite, iron ores and metal ores have been described by Sherman (1955, 1958) and Sherman 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.
THE UTILIZATION OF MINERAL WEATHERING IN IDENTIFICATION OF ENGINEERING PROPERTIES OF SOILS AND EARTH FOUNDATIONS Soil engineers are just beginning to recognize the impor- tance of the mineral of the weathered zone of the earth’s surface, But on the whole they are using para- meters of soil mechanics measurement which are limited to relatively unweathered earth surfaces of the tempe- rate 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 under- standing of the poperties of a montmorilloniie clay on a steep slope when its moisture and pressure relation- ships are changed. Movements of 4 inches per month 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 known for its extreme stability.
THE UTILIZATION OF MINERAL WEATHERING STAGE IN LAND USE CLASSIFICATION The population explosion and the increasing demands placed on land for industry and urbanization has made it imperative that a plan of land use be developed in order to protect our best productive lands for agricul- tural production, water conservation and our forest resources. This important function is too often left to social scientists and planning engineers who 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 who 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 soil’s chemical and physical properties which are fundamen- tally 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
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
56
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 of application of calcium carbonate and of calcium silicate on the yield of Sudan grass grown in Ferrugmaus Laiosol and O Hydra1 Humic Latosol. (IIaw. Agr. Exp. Sta. Tech. Buli. 53.)
PANDEY, S. 1969. Comparison of properties of low Ifumic Latos01 of Hawaii with Tropical Red Eearih of India. Ph.D. Thesis, Grad. School., Univ. of Naw., Honolulu. 442 p.
SIIERMAN, G. D. 1949. Factors influencing the development of lateritic and laterite soils in the Hawaiian Islands. PQC~. Sci. (Honolulu), vol. 3, p. 307-14. - . 1952. The genesis and morphology of alumina-rich laieriiic clays in problems of clay and laterite genesis. Sym- posium. St. Louis. Amer. Inst. Mining and Metal. Eng. New York, N.Y., p. 154-61. - . 1955:Some of the mineral resources of the Hawaiian
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.)
of the bauritic Halii soils. 46 p. (IIaw. Agr. Exp. Sta. Tech. Bull. no. 56.)
a new hing material. Hawaiifarm Sci. (Honolulu), vol. 13, no. 3, p. 8-9.
PQC$ Sci. (Honolulu), vol. 22, p. 458-64.
resources of the Hawaiian Islands. (A revision.) 34 p. (Haw. Agr. Exp. Sta. Bull. no. 138.)
TAMINI, Y.; KANEWIRO, Y.; SHERMAN, G. D. 1963. Ammonium fixation by amorphous Hawaiian soils. Soil Sci., vol. 95, p. 426-30. - . 1964. The reactions of ammonium phosphate with gibbsite, and with montmorillonite and kaolinitic soils. Soil Sci., vol. 98, p. 249-55. - . 1968. The effect of time and concentration on the reactions of ammonium phosphate with a humic latosol. Soil. Sci., vol. 105, p. 434-9.
TANADA, T. 1951. Certain properties of the inorganic colloidal fraction of IIawaiian soils. J. Soil Sci., vol. 2, p. 83-96.
YOUNGE, O. R.; PLYCKNETT, D. L. 1966. Quenching the high phosphorus fixation in Hawaiian soils. Proc. Soil Sci. Soc. Amer., vol. 31, p. 653-5.
-. , CADY, J. G.; IKAWA, I.; BLOMBERG, N. E. 1967. Genesis
-. , DIAS, I. I'. S.; MONTEITH, N. H. 1964. Calcium silicate
-. , IKAWA, II. 1968. Soil sequences in the Hawaii Islands.
-. , WALKER, J. L.; IKAWA, H. 1968. Some of the mineral
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 weather- ing, this mineral releases potassium slowly and that is why it has been regarded as a potential supplier of K in soils. The 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 ‘illite’ and ‘mica’ in soil science. Since mica is a broader term which includes both primary and secondary mica or illite in soils, the term mica has been used in this paper. The Madhupur 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 movement 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. The 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 sub- merged during the rainy season. Karim and K h a n (1956) classified these soils as Gray Brown Podzolic.
Previously Brammer (1964) reported that the soils of
the Madhupur Tract contain a small amount of illite along with a high content of Kaolinite. The aim of this paper is to identify and estimate the amount of mica present in the clay fractions of the soils of the Madhupur Tract and to see h o w this mica is related to the pedogenic processes.
E XP E RIM EN TAL
Three typical soil profiles were selected for this study from locations at different elevations in the Madhupur Tract. The 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 Humaquepts in subgroup level.
METHODS OF ANALYSIS
Organic matter in soils was destroyed by €&O2 diges- tion. 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 Raymax-60 X-ray diffraction unit and a powder camera. Total elemental analysis of the clays was made according to the method given by Piper (1950).
1. Specially presented an supporting paper to review No. 1 by J. J. Fripiat and A. J. Herbillon.
61
I. U. Ahmed. Al. S. Ifussain and M. R. Khan
IIE S U LTS
SOIL PROPERTIES
The 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 two reasons for this : either (a) the clay fraction in the surface horizons of the Tejgaon series has moved down the profile, or (b) it has been washed away by 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 accord- ing to the 10 per cent K,O complement proposed by Mehra and Jackson (1959); the results are presented in Table 2. The 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 compo- sition of the parent material. The two profiles of the Tejgaon series are residual
soils and are regarded as nearly mature soils in East Pakistan. From 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. The 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.
TABLE 1. Distribution of sand, silt and clay in the soils of the Madhupur Tract
Soil aeries Horizon Depth ia inches Sand Silt Clay
Tejgaon (1) (highland)
Tejgaon (2) (medium highland) All
B*l
Karail (valley)
0-4 4-16 16-36 36-48
0-4 4-17 17-27 27-36 36-48
0-4 4-17 17-26 26-35 35.44
35.6 19.1 18.4 17.7
35.4 18.9 17.1 13.5 17.9
3.8 4.7 3.5 0.0 1.0
47.0 31.9 28.6 26.7
37.0 31.5 28.7 32.0 28.2
19.5 12.5 5.6 8.9 10.6
17.4 49.0 53.0 55.6
17.6 49.6 54.2 54.5 53.9
76.7 82.8 88.9 91.1 88.4
62
Distribution of mica in the soils of the Madhupur Tract, East Pakistan
35 --.. .. weathering action the mica present in the coarser
fractions (sand and silt) of the parent material has been -.-.-.-- - .-.-.- .-. 4.- *-e-.-,
-.- - 2. \ -e-.
.-,
. . . . <..:;,: ., r
30
yl 2
._ 3
6 5
4 25 .- .
2 D
..
TABLE 2. Elemental analysis and mica content in the clay fractions of the soils of the Madhupur Tract
originally present in the sand and silt fraction has n o w come into the clay fraction and m a x i m u m accumulation has taken place near the surface.
The second explanation for the higher concentration
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,
- \ \ of mica in the clay fraction of surface soils of the
i,, Tojgaonrerier(1) - 'N. '.
~~j~~~ (21 I
Soil seriee IIorizon Depth in inches
Si02 %
Alt03 %
Fez03 %
KzO %
Calculated mica in clay %
Tejgaon (1) All 0-4 54.8 29.2 3.1 5.0 3.37 34 (highland) B*l 4-16 46.0 26.2 3.2 4.1 3.25 32
Ba* 16-36 46.2 30.1 3.4 5.2 2.75 28 c, 36-48 49.8 33.2 4.6 6.2 2.62 26
Tejgaon (2) Al1 0-4 49.2 28.2 3.9 6.1 3.72 37 (medium highland) B*l 4-17 48.4 27.0 5.4 5.5 3.68 37
B** 17-27 54.3 33.2 6.1 5.0 3.52 35 B** 27-36 50.8 34.1 6.1 5.4 2.50 25 Cl 36-48 49.4 33.2 5.7 3.8 2.25 23
Karai1 APg 0-4 51.4 27.2 5.4 5.4 3.62 36 (valley) Allg 4-17 48.1 28.1 5.4 5.0 3.50 35
Al& 17-26 50.8 31.3 5.1 5.1 3.53 35 IIAIlbg 26-35 43.0 33.4 4.4 5.8 3.48 35 IIAIzbg 35-44 47.8 33.2 4.4 5.8 3.42 34
63
LU. Ahmed, bí. 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 Madhupur 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 summer 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. An elemental analysis of the clay fraction showed
that the total Fe20, and Algo contents range from 3.1 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 Fe 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 known that Kaolinite usually does not contain either' M g or Fe in appreciable quantity, hence a portion of Fe 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 geomorpho- logy 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 frac- tions 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.
KIMURA, H. S. 1966. Personal communication. MEHRA, O.P.; JACKSON, M.L. 1958. Iron oxide removal from soils and clays by a dithiouite-citrate system 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 vermiculife- illite 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.
64
Trends in rock weathering in the southern part of
J.
. -
peninsular India - its expression in morphogenesis of soils’
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 by 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 denu- dation and weathering over long ages. The peninsular mountains include the Western and Eastern Ghats, Vindhyas, Satpuras and Aravallis. The chief rivers of the peninsula are the Godavari, Krishna and Cauvery and the west-ward flowing Narmada and Tapti. The Vindhyas and Satpuras separate the northerly from the southerly flowing drainage. The peninsula thus encom- passes 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 now known as Tamil Nadu with the states of Mysore, Kerala and part of Andhra Pradesh. A cross-section of the plateau is illustrated in Figure 2.
GEOLOGY
One 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 compara- tively 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. The schistose group consists of the Dharwar system under which are grouped the Dharwar rocks of Madras and Mysore. The 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.
The 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. From 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 b’elts of schist, the constituent minerals of all types are clean and fresh, coarsely crystalline and granular in form. The hornblende schist of the Dhar- wars 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 sedi- mentary 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 crescent-
shaped basin along the middle of 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
Western coast (very wet)
Trends in rock weathering in southern peninsular India
Western Ghats Rain shadow area, Deccan Eastern Ghats (very wet) (dry) (less dry)
sea
FIG. 2. Peninsular India-section across plateau.
.
Bombay
FIG. 3. Peninsular India-geology.
67
S. V. Govinda Rajan and R. S. Murthy
peninsula. The outcrop covers most. of the Cuddapah 'and Kurnool districts. In contrast to the surrounding gneissic country, the hills of this large tract are made of precipitous scarps and gentle dip slopes. The Cud- dapahs are essentially a succession of quartzite sand- stones or quartzites and slates or shales.
13 50 c m to 100 c m
100 c m to ZOO c m CLIMATE
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 penin-
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
4*
sula for the year. According to Dudley Stamp (1965), 1
between October ani December. This rainfall comes mainly from storms which occur during the period when the south-west monsoon is retreating before the north- east monsoon has properly established itself. The 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 an average of 800 F to 550 F. The 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 be 390 F can be compared with coastal parts of Kerala having an annual range of 50 F only.
FIG. 5. Peninsular India-isotherm.
Narm
ratei
NATURAL VEGETATION
A s seen from Figure 6. forests cover considerable areas Y
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 way in the deltas to fertile irrigated tracts. On the plateau itself rainfall also provides the key to natural vegetation. The 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 much of the original vegetation remains. In districts where the annual rainfall does not exceed 200 c m but
Principal cultivatt
FIG. 6. Peninsular India-natural vegetation.
St
forest
forest
?d area
68
I
. Trends in rock weathering in southern peninsular India
B o m b a y
. -
FIG. 7. Peninsular India-soils.
is more than 100 crn, there are open monsoon forests with teak and other deciduous trees which shed their leaves in the dry season. But such forests are by no 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 aí. (1936) no single physical factor is predominant for the whole region hut on some occasions the climate and in others the parent material or the topography is decisive. Numerous instances have also been reported in which
I
- a u .... ..... ....
mn
Red sandy soils
Medium black soils
Red loamy soils
Deep black soils
Mixed red and black soils
Laterite soils .. %
Coastal alluvium
Shallow black soils
8
the factor of surface geology intervenes as the ruling clement.
Each of the soil-forming factors such as climate, rock topography and plant growth has played a part in soil formation in peninsular India. The scientific approach to the study of these soils should be made by carrying out a proper analysis in the field of the characteristics of the soil profile and in the laboratory of the proper- ties of the individual horizons.
SOILS, THEIR PROPERTIES AND COMPOSITION The 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, Red and Laterite coils which are discussed in this paper, m a y be closely ex- amined in relation to the geology of the region.
69
S. V. Govinda Rajan and R. S. ñlurthy
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, ferro- magnesian minerals and calcium and magnesium carbo- nates. They develop under climates with high average temperatures, medium 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, Ramand and Tinnevelly ; on the Mysore plateau there is but little. Some 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 many 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. They 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. Topo- graphy 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 conduc- tivity 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 occa-
sionally silty clay or silty clay loam, the clay content of most of the profiles ranging between 40 and 60 per cent. The 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 magne- sium 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
unavailable owing to potassium fixation. Calcium sul- phate 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 montmorillo- nite while calcium maintains a favourable level of PII for its formation. The alkaline earths are released by the weathering of basic rocks.
LATERITE
The bulk of the laterite of peninsular India is found near the Western Ghats. The country rock supporting any cap of high-level laterite consists of basalt, epidio- rite, 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 down into which it passes through a Kao- 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 down 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 many 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 condition- have 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 disap-
pearance 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 some- times 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
70
Trends in rock weathering in southern peninsular India
products of two distinct modes of decomposition. The presence of free silica especially in the form of quartz is an indication that the latente is of detrital origin.
Laterites as described by Maignien (1966) vary in colour but are usually bright. The shades commonly encountered are pink, ochre, red and brown but some 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 man- 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. The 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 alu-
minium as related to the other components is a unique feature of laterites. The laterites of Kerala contain 12 to 15 per cent Fe,O3 and.13 to 18 per cent Al,O,, while the laterites of Goa contain 15 to 20 per cent or more Fe,O, and 15 to 20 per cent Alzo,. Some laterites contain manganese, sometimes in sufficient quantity to be exploited as a mineral (Karwar and Goa). Quartz is an 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 on the basis of chemical composition but as demonstrated by Fox (1936), such classifications are inadequate.
RED SOILS
The 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. The 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 decompo- sition 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 mm are reddish brown in colour and this appears to be true of peninsular India. H e came to the conclusion that
there is specific relationship between rainfall zones and soil groups. The same type of soil can develop under vastly different climatic conditions depending solely on parent material. One of the main types is the coarse- textured sandy soil developed from granite or granite gneiss. It has a finer textured and heavier appearance in the high-rainfall area.
The weathered product from crystalline rocks in the peninsula has undergone more complete dcsilicification and is consequently more sesquioxidic. The predo- minantly red colours m a y be attributed to free ferric oxide with a comparatively low degree of hydration. The characters of tropical soils are therefore dominated by their parent materials. The 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 concre- tionary materials of varying sizes. T h e base status shows a wide range from low to medium and occasionally high. The p H is not so low as might be expected, a circumstance attributable to the sesquioxidic character of the clay complex. There is no marked chemical eluviation and the structure is generally friable and granular.
The red soils are not always confined to a mere zone on a particular contour of topography but are also to be found on level ridge tops and sloping valley sides as well. The 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 on 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 pre- valent, 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. The rock structure, texfure and hardness together determine its physical resistance to katamorphism. The 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
DUDLEY STAMP, L. 1965. The worfd. Bombay, Calcutta, Madras, New Delhi, Orient Longmans. 99 p.
Fox, C. S. 1936. Buchanan’s 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,
no. 83.)
. 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
INTRODUCTION
Basic igneous rocks on weathering gave rise to laterites in Malabar and South Kanara (10015’ to 13059’ N. and 74043‘ to 76015’ E.). The area is characterized by a tropical humid climate with a rainfall of from 130 to 150 inches, a mean annual temperature of 800 F and a pronounced wet and dry season. In Malwa 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 rem- 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 ann- ual maximum and minimum 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 KANARA
Satyanarayana and Thomas (1961) developed morpho- logical concepts. and horizon designations for the Buchaman type laterite occurring in the type area Angadipuram of Malabar and South Kanara. Satyana- rayana and Thomas (1962) studied two in situ laterite profiles from the type areas Angadipuram and Kasa- ragod 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 weather- ing of rocks, the prevailing differences of the primary minerals are levelled down 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
R. V. S. Satyanarayana Indian Agricultural Research Institute, New Delhi (India)
of kaolin, gibbsite or bauxite and iron in the profilc. The 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 Angadi- puram, 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. The mineralogical composition of the corresponding horizons and cation ex- change capacity of the two laterites is given in Table 1. The data reveal the marked differences in the minera-
logical make-up of corresponding horizons in the two profiles. The profile from Angadipuram contains pre- dominantly more kaolinite and small quantities of other crystalline compounds while in the Kasaragod profile kaolinite content is slightly less but the crystalline compounds are much higher. Amorphous compounds of iron are present in both the profiles, the basic rock contributing a large amount. In addition the Angadi- puram 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.
The laterite from basic rock is highly aluminous and in the case of medium 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
Surface soil IIard laterite Soft laterite Weathered rock
Ang.1 Kgq. Ang. K gq. An&. Kgq. Ang. Km. A A DLIL2 BLsh nL.,
Quartz Kaolinite Gibbsite Limonite Remaining Peso,
A1203 SiO,
c.e.c.
36.7 33.8 6.7 3.9 16.0 ... ... 4.6
1. Ang.-Angadipurarn; Kgq. =Kasaragod.
10.5 45.0 8.0 10.0 24.0 ... ... 6.5
23.9 54.4 ... ... 13.4 2.1 2.3 5.4
3.2 44.0 4.0 33.0 12.0 ... ... 4.5
28.5 51.8
12.5 0.8
1.4 3.3
...
...
9.2 26.0 32.0 1.0 29.0 ... ... 5.2
42.6 13.9 1.0
14.0 11.8
...
...
...
6.6 50.0 12.0 13.0 13.0 ... ... ...
' LATERITE AND BLACK SOILS OF M A L W A PLATEAU
Pendleton (1947) attributed the formalion of laterites in Mandsaur District of Malwa 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 Satyana- rayana (unpublished) observed a relict B-horizon of laterite in Malwa Plateau wilh an occasional thin cover of red soil or reddish-brown soil on 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 on montmorillonite formation from basic igneous rocks of this region. Multiple corre- lation analysis on the chemical data of the clay fraction of the soils indicated that with increased entry of iron in the octahedral layer the formation of montmorillo- nite is lea possible. The 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 environ- ment with a low oxygen tension is esscntial for mont- morillonite 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
TABLE 2. C a t k exchangecapacity' of the soil, silt and clay
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.
The data (Table 2) on 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. The minera- logy of these fractions indicated the predominance of montmorillonite (Satyanarayana et al., 1969). The 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. The silt and fine-sand frac- tions of the weathering zone were separated into five specific gravity fractions viz. > 2.86; 2.86-2.62; 2.62- 2.50; 2.50-2.35; and < 2.35 using bromoform-alcohol mixtures. The specific gravity fraction (< 2.35) showed c.e.c. value of 80-90 me/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.
Soil no. Nature of soil Depth inches - Soil Silt (2-20 p) Clay (< 2 d
33 36 38
lleddish brown (lateritic) 0.7 Black 0.6 Reddish brown (lateritic) 5-10 ,
25.9 36.5 (3.6)* 51.7 (20.2) 60.6 58.1 (9.4) 95.7 (50.4) 34.7 40.6 (5.7) 67.5 (32.1)
1. The c.e.c. WUB meaaured by the modilìcd ammonium aretnte meilod. 2. The 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. Pedo- genic 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. (Ade- laide), vol. 4, p. 287. Amsterdam, Int. Soc. of Soil Science.
KRISHNA MURTI, 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.
soils. I. Field characteristics of laterites of Malabar and South Kanara. J. Indian Soc. Soil Sci., vol. 9, no. 2,
-. - . 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.
-. , THOMAS, P. K. 1961. Studies on laterites ana associated
p. 107-18.
75
Development of certain soils in the subtropical humid zone in south-eastern parts of India, genesis and classification of soils of Machkund Basin'
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 km2 falls 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 mean 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, when a large number of soil profiles were examined in the field for their morphological featuring and a number of taxonomic units were recognized, based on their differentiating morphological features. It is found that variations in parent material and topography with which is associated the moisture regime and micro- climate 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 by placing them into different groupings and also locate them in the international system of classi- fication.
I
PHYSIOGRAPHY, RELIEF AND DRAINAGE
The 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,
S. V. Govinda Rajan and N. R. Datta Biswas All India Soil and Land Use Survey, Indian Agricultural Research Institute, New Delhi (India)
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 show 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 inter- sected by 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 degrada- tion by the agents of erosion, the major channels have been at grade-attaining profiles of equilibrium.
GEOLOGY
The 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-building forces. Different types of metamorphism have made the texture of these rocks either schistose or gneissose. Khondalites are the oldest rocks in this area and are highly meta- morphosed Pre-Cambrian sediments 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 plu- tonic 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
o
lization in the solid state on being subjected to plutonic metamorphism. They are characterized by the presence of hypersthene, blue-grey quartz and feldspar in acid and intermediate rocks and pale-green augite, blue- green and green-brown hornblende in the more basic rocks. The 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 super- imposition 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 many places along with the horn- blende schists. Laterite mass is found to cap some of the peaks of the hills built of khondalites. These late- rites 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.
CLIMATE
The climate of the area is subtropical monsoon type characterized by hot dry summer followed by a period of heavy precipitation and dry cold winter.
The following yearly seasonal sequence is indicated from climate data collected from the area of the Machkund B,asin:
1. Cold and comparatively ?ry winter-December,
2. Hot and dry summer-March, April and May. 3. Monsoonal warm and wet period-June, July,
4. Cold autumn-October ,and November, The mean monthly rainfall and humidity (Health Office, Machkund), averaged over ten years (1946-57) and the mean maximum and minimum temperatures for this region are given in Table 1. The distribution of rainfall and temperature are given in Tables 2 and 3.
January, February.
August and September.
’
EXPERIMENTAL
Soil samples of six soil series, namely, Kallupada, Pippoluputtu, Kangrapada, Devulupada, Kona and Hollangapada series identified in the area were collected for evaluation of chemical and physico-chemical pro- perties. A landscape of this area and relative disposition of the different soils are shown in Figures 1 and 2. A summary of the important morphological features is presented in Table 4.
Mechanical analysis was done by the international pipette method; organic carbon by Walkley and Black’s rapid titration method as modified by Walkley (1935), nitrogen by the Kjeldahl method (AOAC, 1950), p H by Beckman pH-meter in 1 : 2.5 soil-water suspen- sion (Piper, 1950); cation exchange capacity by leaching with ammonium acetate (Schollenberger and Dreibelbis,
TABLE 1. The mean monthly rainfall, humidity, and maximum and minimum temperatures of the Machkund Basin
Jan. Feb. Mar. Apr. M a y Jun. Jul. Aug. Sept. Oct. Nov. Dec.
Rainfall (mm) 13.7 2.8 9.9 45.2 53.3 208.5 482.8 626.0 411.0 155.6 7.6 3.5 Humidity (yo) 81 72 60 61 61 82 89 92 92 83 83 82 Max. temperature (OC) 28.3 32.2 36.4 38.8 39.9 37.5 30.3 29.8 29.2 30.8 29.4 28.0 Min. temperature (OC) 9.8 16.6 15.4 19.6 22.6 22.4 20.2 20.5 21.0 18.3 13.2 7.8
TABLE 2. The annual distribution of rainfall (in mm) in the Machkund Basin
TABLE 3. The annual distribution of temperature (in OC) in the Machkund Basin
. Winter: Summer: Monsoon: Autumn: December March June October
to to to to Mean February M a y September November
Winter: Summer: Mnnannn: Autumn: December March June Octobrr
to to to to Mean February M a y September November
, 2018 20.3 108.7 1724.9 16.3 Max. 32.2 28.31 38.28 31.61 30.17 Min. 16.6 8.25 19.17 21.11 15,67
78
Development of certain soils in south-eastern parts of India
Kallupada
Pippoluputtu
Kangrapada
Hollangapada
Kona
FIG. 1. A landscape in Machkund catchment showing the relative positions of the different soils.
I Kona
Devulupada
1 Kona FIG. 2. Disposition of the soils in a valley situation-Machkund catchment (Orissa).
79
S. V. Govinda Rajan and N. Iì. Datta Biewas
1930); total exchangeable bases by Piper method (1950) and base saturation by calculation. Ferric oxide, alu- minium oxide, sesquioxides, PO and C a 0 in hydro% chloric acid extracts were determined by standard chemical methods.
RESULTS A N D DISCUSSION
The area under study experiences an annual precipi- tation of 200 cm. Even though the rainfall is concen- trated within the four monsoon months, high humidity is maintained throughout the year. The difference between the mean annual m a x i m u m and minimum temperatures is wide, being of the extent of 16oC. It may, however, be mentioned that the area expe- riences dry spells as well during the year. The diverse components of climate in operation along with other soil-forming factors have been responsible for develop- ment of different kinds of soils in the area. One of the important factors has been topography. The 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. From 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 pre- sented 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.
From the foregoing discussion it would be apparent that there has been lateral translocation of finer mat- erials as also of soluble constituents on account of lateral movement of water, which flows down 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. The cation exchange capacity of the soils also varies,
being lower in respect of soils of the hills and hill
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 régime 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 defini- tely 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 accu- mulation of finer mechanical constituents. The cation exchange capacity of the soils is moderate to high showing that the process of kaolinization has not pro- ceeded 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 repre- sented by 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 Humic Gley Soils.
POTENTIAL FOR 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. The 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.
80
Development of certain soils in south-eastern parts of India
TABLE 4. A summary of the important morphological features of the different soils identified in the Machkund catchment area'
. Reac- Root dis- Other features includine: tion tribution mottling and/or concretion Soil Physiographic Depth . ~~l~~~ Texture , No. series position (em)
1 Kallupada Hill tops and 0-10 Reddish Loamy sand Structureless, Acid Common Ferruginous concre- top slope of bi&n to sandy loose and tions common with hill sides 2.5 YR 4/4 loam friable gravels and pebbles
or weathered rocks
10-35 Reddish Clay loam Weakly Acid Many Many ferruginous yellow hlocky concretions with rock 7.5 YR 6/8 fragments
2 Pippo- Convex luputtu sloping
hill aides
35-45 Yellowish Clay loam Blocky Acid red 5 YR 5/8
45 + 0-12.5 Dark Loamy sand Sthctureless, Acid Plenti-
reddish to sandy loose and fu1 brown loam; friable 5 YR 3/3 occasionally
loam
Many ferruginous concretions, lateritic pieces with fragments of weathered rocks
Laterite
Ferruginous concre- tions common, gra- vels, stones of various sizes
12.5-40 Yellowish Clay loam Weakly Acid . Many Ferruginous concre- red (gravelly) blocky tions and gravel 5 YR 4/8 common
40-75 Darkred 10 R 3/6
75+ .
3 Kangra- Foothillareas 0-15 Yellowish pada and sloping red
uplands 5 YR 4/6
15-42 Red 10 R 4/6
42-90 Dark red 10 R 3/6
Matrix of ferruginous clays, gravel and late- rite pieces; patches of black mottling
Consolidated hard ferrallitic mass, red- dish and hard
Sandy loam Weakly Acid Common Few ferruginous con- to sand . granular cretions and partially clay loam weathered fragment
of rocks
Clay ioam Weakly Acid Many Many unweathered blocky and partially weather-
ed rock fragments and many ferruginous concretions
Clay loam Strongly Acid Few Many unweathered gravelly blocky and partially weather-
ed rock fragments and many ferruginous concretions
90+ ' Duskyred 10 R 313
Matrix of detrital laterite. ferruginous clay and unweathered pieces of rock frag- ments
81
S. V. Govinda Rajan and N. R. Datta Biewas
TABLE 4 (continued)
, &y Root dis- Other featurei including tribution mottling and/or concretioni S?ii Phyriographic Depth Texture
nenes position (4 No.
4 Hollan- Sloping up- 0-15 Yellowish Loam to Weakly Acid Abun- gapada lands running red clay loam granular dant
with convex 5 YR 4/6 slope of gentle gradient
15-57.5 Dark reddish Clay loam Weakly Acid Many brown blocky 2.5 YR 314
57.5-150 Dark red Clay loam Blocky Slightly Common pea-sized 2.5 YR 3/6 (gravelly) acid hard ferruginous con-
cretions and quartz gravels
5 Devu- Nearly level 0-12.5 Dark yellow- Clay loam Sub-angular Acid Plentiful Few mustard-sized lupada to level and ish brown blocky ferruginous concre-
terraced 10 YR 414 tions lowland Yellowish
brown 10 YR 516
12.5-45 Yellowish Clay loam Angular Acid Many Many hard ferrugi- brown blocky nous concretions 10 YR 818
45-85 Yellowish Clay loam Strong Acid Many grey mottliiigs brown angular and few yellowish-red
Many soft black and brownish ferruginous concretions
10 YR 518 blocky spots;
6
85-150 Dark grey Clay Massive Acid Many pea-sized soft
tions of strong brown and black colour; red- dish brown and very dark grey-brown mottlings
7.5 YR 4/0 ferruginous concre-
Kona Depressions, 0-12.5 Pale brown Clay loam level terraced 10 YR 6/3 lowlands, reclaimed gullies and stream beds
12.5-325 Greyish Clay loam brown 10 YR 5/2
Sub-angular Mildly Plenti- blocky acid ful
Angular Mildly Many Dark yellowish-brown blocky acid mottlings; many
mustard-sized quartz gravel present
32.5-105 Dark grey Clay Strongly Mildly Few pea-sized, black, 10 YR 4/1 blocky acid soft ferruginous con-
cretions
82
Development of certain soils in south eastern parts of India
TABLE 5. Physical and physico-chemical properties of horizon soil samples of the profiles of the recognized soil series in the Machkund catchment
Mechanical constituenti (expressed as percentage Pbysico-chemical properties
Soil series on air-dry basin) B.E.C. n e . yo
Sand Silt Clay c.e.c. me. saturation PII on clay Total T.E.E. . Base calculated No. and depth (em)
me. 100 g 100 g (%)
Kallupada 0-10 10-35 35-45
Pippoluputtu 0-12.5
12.5-40 40-75
Kangrapada 8-15 15-42 42-90
IIollangapada 0-15 15-57.5
57.5-150
Devulupada 0-12.5
12.5-45 45-85 85-150
Kona 0-12.5
12.5-32.5 32.5-105 1 O 5 - 15 O
71.4 54.3 45.8
73.9 61.6 48.4
55.8 49.9 44.5
56.7 58.7 58.9
53.9 49.8 43.9 35.6
57.5 48.1 41.1 28.7
9.0 18.5 21.3
10.3 13.7 14.2
21.4 22.6 22.9
21.1 17.2 15.6
19.5 20.8 21.8 23.6
14.8 18.0 19.4 22.8
19.9 27.1 32.8
15.8 24.7 37.4
22.7 27.5 32.6
22.1 24.0 25.4
26.6 29.4 34.2 40.7
27.7 33.9 39.4 48.5
6.7 8.8 9.2
7.2 8.4 10.8
6.8 8.9 10.6
8.4 9.3 10.8
8.8 13.2 22.9 26.3
18.9 22.3 25.2 26.6
3.2 4.8 4.8
3.8 4.0 5.5
3.5 4.5 5.5
4.3 5.1 5.7
5.2 7.9 13.3 15.7
11.6 13.6 16.1 15.3
47.5 5.4 33.8 54.5 5.6 32.4 52.1 5.6 28.1
52.7 5.4 50.9 5.5 51.1 5.7
51.8 5.6 50.9 5.6 51.7 5.7
51.4 5.4 54.4 5.4 52.7 5.6
45.4 34.0 29.0
30.0 32.5 32.7
38.0 38.7 42.4
59.0 5.4 33.1 60.1 5.4 44.9 58.1 5.6 66.9 59.9 5.6 64.3
61.8 5.5 , 68.2 60.9 . 5.6 65.9 63.9 5.6 63.9 50.0 5.7 54.7
TABLE 6 '- overleafl
83
,
S. V. Govinda Rajan and N. H. Datta Uiswas
TABLE 6. Chemical composition of the horizon soil samples of the soil series
Soil series and No. horizon depth
Chemical composition (percentage on oven-dry basis)
(em) N pz.0 FezOa ~ Alzo3 C a 0 Organic carbon
1 Kallupada 0-10 10-35 35-45
0.07 0.06 0.06 0.05 0.03 0.04 .
9.94 9.58 10.56
17.99 0.16 22.55 0.18 25.18 0.19
0.74 0.68 0.27
2 Pippoluputta I
0-12.5 0.07 0.06 9.20 19.28 0.16 0.66 12.5-40 0.05 0.06 10.16 22.87 0.17 0.36 40-75 0.04 0.04 10.24 25.47 0.18 0.25
3
4
5
6
Kangrapada 8-15 15-42 42-90
IIollangapada 0-15 15-57.5
57.5-150
, Devulupada 0-12.5
12.5-45 45-85 85-150
Kona 0-12.5
' 12.5-32.5 , 32.5-105
105-150
0.05 0.07 9.52 . 0.05 0.05 14.16 0.04 0.04 15.80
0.08 0.07 0.05
0.10 0.06 0.05 0.04 ,
0.11 0.05 0.04 0.04
0.06 0.05 0.04
0.08 0.05 0.04 0.03
0.11 0.09 0.06 0.05
' 7.90 8.30 8.80
8.40 8.24
, 9.88 10.00
7.54 7.76 8.20 9.20
17.78 19.52. .
22.06
17.09 18.74 18.80
16.97 16.33
'.: 17.31 18.59
16.02 19.01 19.82 23.95
0.17 0.18 0.18
0.17 0.18 0.19
0.20 0.22 0.25 0.36
0.36 0.38 0.38 0.39
0.58 0.53 0.41
0.87 0.60 0.37
1.12 0.76 ,
0.55 0.30
1.06 0.62 0.58 0.3.5
Bibliography
AOAC 1950.' Oficial and tentative meihods of analysis, p. 30. Washington, Association of Official Agricultural Chemists.
GOVINDARAJAN, S. V.; DATTA BISWAS, N. R. 1968. Charac- terisation 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 classi- fication and nomenclature of Great Soil Groups in tropical
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. New 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 deter- mining organic carbon and nitrogen in soils. J. Agric. Sci., no. 25, p. 598.
84
Pedogenesis of soine major soil groups in Mysore State, India’,
INTRODUCTION
The 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. The physiographic and climatic features under which the above kinds of soils occur differ markecliy. The 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. The 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. They occupy an 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, Sep- tember and October the evapotranspiration exceeds precipitation, resulting in a water deficit which is severe from March to June.
The 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 known as the Deccan Trap on account of the step-like aspect of the weathered hills of the basalt formed from the erupted lava. The Deccan Trap is augite-basalt and basic in character containing feldspar, augite and magnesite as the main minerals and does not sliow much differentiation except in colour
T. Seshagiri Rao University of Agricul tura1 Sciences, Regional Research Station, Raichur, Mysore State (India)
and texture. The 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 com- posed of the oldest Archaen complex of granites and gneisses, which, however, vary in chemical and minera- logical composition, texture, structure and colour. The granites and gneisses consist mainly of types containing (a) quartz and feldspar; (b) quartz, feldspar and horn- blende; and (c) quartz, feldspar and mica. The feldspars vary in composition being of both types, i.e., soda-lime and potash feldspars. It is also a feature to find black- coloured soils occurring in the valleys and lighter- textured red soils in the terraces and uplands. A typical black cotton soil has the following features.
O to 30 cm, black (10 YR 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. I
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 many of these soils. The 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. The 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 make 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. The 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 mm per annum. The 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 :
I
Profile 1 O to 12 cm, reddish-brown (5 YR 5/4 dry), sandy loam, weak, fine, granular structure, loose and friable, non- sticky 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, fria- ble, 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 YR 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
and fertilizers. The 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 make these soils highly susceptible to surface erosion even with a favourablc internal drainage.
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 prevail- ing 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 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-angular blocky structure, firm, non- plastic and non-sticky, bigger iron concretions, diffuse boundary. 120 cm +. reddish-brown (5 YR 4/4), clay-loam, reti- culately mottled, strong, medium, sub-angular blocky structure, firm, slightly sticky, rounded red and black iron concretions, plinthite.
B
C
B has all the properties of an oxic horizon con- taining a mixture of hydrated oxides of iron and alu- minium, 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.
Due 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 many other soils with the same slope. They are also easy to work with because of their friable nature.
86
Pedogenesis of some major fioil groups in Mysore State, India
Bibliography
DREW, J. V. (ed.). 1967. Selected papers in soil formation and UNITED STATES SOIL SURVEY STAFF. 1967. Supplement to soil classification. Lincoln, Nebraska, Soil Sci. Soc. Amer. ClassiJication system, 7th approximation, p. 207. Soil Conser- Special Publication Series, p. 428. vation Service, U.S. Dept. of Agriculture.
VENKATA h o , B. V. 1968. Soil resources in Mysore. Raichur, ICAR, p. 470. Uniu. of Agric. Sci. Extension Series, no. 2, p. 24.
VENKATARAMIAH, P. 1934. Report on the soil survey of the Bull. Soc. belge Pédol., special no. 4, p. 134. Tungabhadra Project, p. 106. Madras, Government Press.
KANITKAR, N. V. 1960. Dry farming in India, N e w Delhi,
SMrnr, Guy, D. 1965. Pedology lectures on soil classification.
87
Correlation between chemical bond and chemical weatherabilitv of the common
1 n . . 1 1 rock-Iorming minerals’
H. Ling Ong Bagian Geologi, Institut Teknologi Bandung
I Bandung (Indonesia)
INTRODUCTION that polymorphism is c o m m o n among minerals. In connexion with this, two possibilities exist in choosing
Chemical weathering of minerals and rocks is one of ‘the formation temperature. First, the formation tem-
’
the most intriguing problems encountered by the geologist working in a tropical climate. Outcrops are rare and sometimes it is impossible to do conventional geology in this type of environment. The study of rock weathering, both theoretical and experimental, should therefore be directed towards an elucidation of the forces responsible for weathering processes.
The accepted theory on chemical weathering men- 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 weather- ing is the reverse of their order in the reaction series of Bowen (see Fig. 1).
This stability series conforms with field observations ; quartz is found to be the most resistant among the igneous minerals. Whereas olivine is the least resistant of the series.
Coldich’s principle of mineral stability m a y be explained in the following way. The minerals in igneous rocks were at equilibrium under the conditions of tem- perature and pressure at which the rocks are formed. - Having been formed at high temperature, and some- times at high pressure as well, they cannot be expected to remain stable under the very different conditions at the earth’s surface. The higher the temperature of for- mation, the less stable the mineral will be at surface environments. In a general way this seems reasonable enough although it is not a necessary conclusion, theo- retically (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 rock- forming minerals.
The formation temperature is defined as the tempe- rature at which the minerals are formed. Since the work
perature 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 Goldich’s series are assembled in Figure 1. The transformation tem- perature 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 Goldich’s 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 tempe- rature of the mineral, then quartz having a melting tem- perature 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 anor- thite. In both cases, whether the temperature of forma- tion is taken as the transformation temperature or is taken as the melting temperature, the Goldich’s series cannot be confirmed.
The 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
of coldich in 1938, severai investigations have found 1. Speeidy presented 8s supporting paper to review No. 2 by P. Segalen.
89
II. Ling Ong
Alkalic plagioclase
Alkali-calcic plagioclase
Calcic-alkalic plagioclase
1,7130 C Quartz ~
5730 c
940"-9800 C
? i
Muscovite
I 1,2350 C K-feldspar
I 7000-190000
1,1180 C - 700° C
1,2000 c ___ 5750 c
1,3300 C ___ .5250 C
1,5520 C
Augite
1,2050-1,8900 C
literature. The 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.
THE O R Y O F C II E M I C A L .WE ATH E RING The decomposition of the rock, while still in massive form or a loose material, is known as chemical weather- ing. 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.
The 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 an arid climate, although in both climates the amount of oxygen present in the atmosphere is the same. The processes of hydration, carbonation, solution and deposition cannot occur without the presence of waLer. The ullimate fate of tliíTerent minerals thus .depènds largely on their rela- tive stability in water. Chemical weathering can Le considered as a process of interaction between minerals
,
Stability decreases
! '
J. I l 1,0000-1,1000 c
Biotite ?
Hornblende I
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. The mineral surfaces of an 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 involv- ing water molecules can be predicted.
Ionic bonding between atoms A and B.occurs when atom A has a tendency to lose an electron whereas atom B has a tendency to accept the electron. The grea- ter the tendency for atom A to lose an electron and/or the greater the tendency for atom B to accept an elec- tron, the greater the ionic binding will occur between atoms A and By and consequently, the smaller the con- tribution of the covalent bond. The concept of losing and accepting electrons for atoms can be expressed in quantitative terms. This is known as the electronega- tivity and can be described as tlie electron-attracting powcr of an atom in a molecule. Thus an atom that has a tendency to lose an electron should have a low value
90
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 electro- negativity. The greater the difference in electronegati- vity between two atoms, the greater contributions of ionic bonding will be.
If we know 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 sur- face 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 earth’s surface.
CHEMIC AL W E ATH ER IN G OF .THE C O M b.2 O N ROC K - F O R M I N G MIN E RA L S In 1932, Pauling derived a useful scale of electronega- tivities. H e related the amount of ionic character in a molecule AB 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 bond- ing, Povarenykh draws a purely empirical curve between zero ionic bonding at infinitely large dif- ferences.
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 cal- culating complex compounds commonly found in min- erals. However, by comparing the amount of ionic bonding between two pairs at a time, the percentage of ionic bonding in minerals can be compared. The 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 com- m o n rock-forming oxide pairs are shown in Table 1. A high percentage of ionic bonding means a low per- centage of covalent bonding. Thus K-O which is 69 per cent ionic according to the method of Hannay and Smyth is 31 per cent covalent. The amount of ionic bonding calculated using either
IIannay’s and Smyth’s equation or Povarenykli empi- rical graph for the geologically important oxide pairs (Table 1) shows a general decrease from K-O to Si-O. The only difference is for iron. Whereas Haririay’s and Smyth’s values for iron (II) and (III) are similar to the value of Si-O (37), the values of Povarenykh
TABLE 1. Percentage of ionic bonding for the common rock- forming oxide pairs
Oside pair
K-O Na-O Ca-O
Fe(I1)-O . Fe( 111)-O Al-O si-o
Mg-o
Percentage ionic bonding (calculated by using Pauling’s scale (1932) Povarenykh empirical and Hannay’s and
Smyth’s equation (1946).
Percentage ionic bonding (calculated by using
graph, an seen in Smyth, 1963, Table 2-1)
69 65 62 55 37 35 ‘46 37
86.6
79.4 71.2 68.7 54.3 60.3 48.0
83.2
show that iron oxides (68.7 and 54.3) are ionic in character relative to the Si-O bond. However, the over- all agreement for the other oxides indicates the vali- dity of both calculations. ,
The common rock-forming minerals belong to the silicate system in which in every case a Si-O bonding is involved. Since for the geologically important oxides Si-O has the least ionic bonding, or at least comparable to that of iron if Hannay’s and Smyth’s equation is used, it can be said that the percentage of Si-O or SiO, in a mineral is directly proportional to the per- centage 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 resis- tant 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. Petti- john (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 common rock-forming minerals and their SiO,
content are listed in Table 2. The SiO, content here indi- cates the predicted values of weathering. Included for comparison are the data of Mohr 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 Mohr 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 sys- tems is taken into consideration, the IwedicLed values as a whole fit with the observed values. Thus, within
91
II. Ling Ong
TABLE 2. Correlation between SiO, content and the observed weatherability of the common rock-forming minerals
Si02 or Observed
Structure N a m e Chemical composition Predicted weatherability weatherability scale (high SiOp cuntent scale1
means resistant to weathering)
Tektosilicates Quartz Si02 100 Very low Ei-feldspar KA1Si,08 63 Low
Albite NaAlSi308 (Ab) 69 Low
Andesi te Ab,o-Ab,o 56-61 nigh Labrador¡ te Ab,,-Ab30 51-56 -
Plagioclabe
Oligoclase A h ~ A b m 61-66 Medium
Bytownite Abao-Abio 46-51 Very high Anorthite CaA1,Si,08 (An) 43 Very high
Leucite KAlSi,O, 53 Very high2 Nepheline NaAlSiO, 42 Very high2
Fcldspathoids
Nesosilicates Olivine group - Fayalite Fe,SiO, 34 Forsterite Mg,Si04 55 - Olivine (Fe, Må),SiO, 34-55 Very high
Inosilicates Pyroxenes '
Hypersthene Iledenbergite Diopside Augite Aegirine
Amphibole IIornblende
. Enstatite MgSiO, (hfg, Fe)SiO, '
CaFeSi,ûE CaMgSi,OE Ca(Mg, Fe)Si,06
, NaFeSi,06
NaCa,(Mg, Fe, Al), (Si, W8O2,(oW2
40 less than 40
48 55
48-55 52
38-583
- hledium - - Medium - Medium
Phyllosilica tes Mica Biotite (&, Fe), (AlSiaOlo) (OII), 33-363 Very high Muscovite' KAlz(A1Si 30,J (OH), 44-463 Low
1. Xohr and van Daren, 1960, Table 37, p. 114-5. 2. V a n Hise (1904). 3. Average content (Mohr and 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 musco- vite. Not 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 an increase of SiO, content.
Ohservcd values for wcatliering of ncsosilicatcß anil inosilicates are too scarce to make a reliable comparison
with the predicted values. For the nesosilicatcs, the predicted order of rcsistancc is : forsterite, olivine, and fayalite. For the inosilicates : diopside, aegirine, heden- bergite, hypersthene and enstatite.
T o find out why the structure of the different silicate systems has to be taken into consideration in the pre- diction of weatherability of minerals, w e have to refer to the assumption made in deriving the theory of che- mical weathering. The 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 on the surface area and the mor- pliology of Lhe crystals. A large surface area means an extensive chemical reaction. Thus, the physical pro-
92
Correlation between chemical bond and chemical weatherability
perties of minerals or the atomic arrangements of dif- ferent silicate systems are important and should be taken into consideration.
Weathering of the different polymorphs of the com- m o n rock-forming minerals is another structural pro- blem. The most common 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. The high-temperature polymorph has a more open struc- ture than a low-temperature polymorph; the high- temperature 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 NaA1 for Si in the open struc- tures; quartz, on the other hand is generally pure SiO, (Buerger, 1951). From this fact it can be concluded that the high-temperature polymorph will be more subjected to weathering because of the low Cio, content. Extra- polation of this finding to the weathering of other common 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-temperature polymorph should be more resistant to weathering than the high-temperature form. There is, however, no observed weathering data available to verify this finding.
S UM RIA IlY AND CO NC LUS I ON S
An attempt has been made 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 weather- ing processes.
The following conclusions can be drawn: 1. The temperature of formation of minerals used by
Goldich for explaining the weathering of minerals can- not be confirmed because of the formation of poly- morphism. 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 min- erals 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 bond 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 and 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 common rock-forming minerals shows a close agreement provided the comparisons are restricted to isostructural groups, wherein the dif- ferences are basically in the character of the bonding. 5. The predicted order of decreasing resistance to-
wards 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 neso- silicates : forsterite, olivine and fayalite; (c) for the inosilicates : diopside, aegirine, hedenbergite, hyper- sthene 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 low- temperature 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.,
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.
vol. 46, p. 17-58.
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.; VAN BAREN, F. A. 1960. Tropical soil. nruxel- les, Les Editions A. Manteau, 498 p.
93
if. Ling Ong
MOHEY, G. W. 1964. Phase-equilibrium relations of the com- mon 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.
llou, Ilustum. 1949. Decomposition and resynthesis of the
SMrTIr, F. G. 1963. Physical geochemistry. Reading, Massa-
TUTTLE, O. F.; BOWEN, N. L. 1950. High temperature albite
VAN HISE, Ch. R. 1904. A treatise on metamorphism. Mono-
micas, J. Amer. Cerant. Soc., vol. 32, p. 202-9.
chusetts, Addison-Wesley Publishing Co. 624 p.
and contiguoue feldspars. J. Geol., vol. 58, p. 572-83.
graphs of ihe US. Geol. Survey, vol. 47.
94
The influence of the nature of parent rock on soil formation under similar atmospheric ’climates’
INTRODUCTION
Am o n g the great number of processes operating in soils, those commonly referred to as ‘Podzolization’ and ‘Laterization’ are 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 pro- minent feature in most temperate soils while lateriza- tion is a process usually found in tropical soils. The most outstanding difference between these processes is con- sidered to be leaching of sesquioxides and relative enrich- ment of silica in surface horizons in podzolization, and the reverse when laterization is concerned, namely leaching of silica and relative accumulation of ses- quioxides in surface horizons. The concept of soil zonality on which many 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 pod- zols and podzolic soils in the tropics is n o w a well- established fact. Klinge summarizes much of the pre- sent 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 de- velop 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 ter- race 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, 1959). The process of illimerization (Fridland, 1958),
meaning leaching of clay without prior decomposition
J. P. Andriesse Soils Division, Department of Agriculture, Sarawak
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 on 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. Tan and van Schuylenborgh in a series of articles (1959, 1961) have shown that podzo- lization 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 alti- tudes than under a continuously wet climate. They explain this by an increase in humification at higher altitudes caused by the lower temperature resulting in the formation of humus acids while at lower altitude mineralization prevails over humification because of the higher temperatures, and carbonic acid becomes the dominant leaching agent. The author was struck by the fact that in Sarawak an almost identical range of soils was found to that described by Tan et al. In Sarawak, however, changes from laterization to pod- zolization are related to the kind of parent material: thus increasingly marked podzolization occurs in soils on increasingly acid parent materials. The 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 at- tempts to explain the apparent enteraction which takes place in soil formation between the atmospheric climate and the nature of the parent material.
1. Specially presenïed an supporting paper to review No. 3 by J. van Schuylenborgh.
95
J. P. Andriesse
TABLE 1. Summary of mean monthly and annual rainfall (West Sarawak stations) in inches (period of recording years given in parentheses after name of station)
Month Kuching Sungei Chins Malang Ban Tsrat I.iindu Simanggang (71) (72) (32) (17) (8) (20) (18)
January 25.80 34.73 33.76 20.57 20.67 21.62 16.47 February 19.19 23.12 21.35 11.60 14.80 18.47 13.45 March . 13.72 15.92 12.10 9.70 11.06 13.91 12.99 April 10.13 11.31 10.75 9.69 12.08 6.86 15.18
June 8.44 7.94 6.74 6.83 5.97 7.62 8.96
August 8.48 8.74 8.78 8.21 7.07 5.62 10.92 September 10.17 12.12 9.99 8.95 11.62 7.43 11.09 October 12.77 14.75 12.01 11.85 11.33 8.57 15.47 November 13.52 16.06 14.13 12.87 15.09 11.42 17.34 December 18.88 23.96 20.66 13.86 12.45 16.60 16.56
TOTAL 158.11 189.82 168.85 129.56 139.13 131.64 159.98
Absolute maximum 62.28 60.25 85.99 70.73 27.44 55.56 30.83 Absolute minimum 0.66 2.23 2.06 . 2.07 2.98 0.87 3.00
May 9.46 12.38 10.81 8.36 9.50 7.96 12.02
July * 7.55 8.79 7.77 7.07 7.49 5.56 9.53
- - - -- - -
CLIMATE
The soils and rock types studied are all located in West Sarawak (latitude 0050’ N.20 N., and longitude 109030’ E.-111050’ E.). Table 1 shows the rainfall for a selected number of rainfall stations having a sufi- ciently high number of recording years to be reliable and found scattered over the area. The highest mean annual rainfall is 4,750 mm 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 Mohr’s rainfall classification for pedological purposes (Mohr, 1944), the climate is considered to be continuously wet. Average daily tem- perature over the area is 25.50 C, the humidity û4 per cent daily average (Department of Civil Aviation, 1961). The climate is, therefore, characteristic of humid trop: icai lowlands.
ALTITUDE
All soils and rock types studied are found below an altitude of 500 ft (approximately 150 m) and there- fore the influence of altitude on soil formation can be ignored.
PARENT 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 pre- sented 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
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
The soil types found on 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. One 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 impor- tance. The 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 Dobritskaya’s method (1962) are shown. Since all profiles studied are mature soils, it is suggested that none or very little original iron- and aluminium-bearing material of the rocks has
96
Influence of the nature of parent rock on soil formation
TABLE 2. Analyses of igneous rock types in W e s t Sarawak
Olivine Dacite Micrograno- Rock name: gabbro Basalt Andesite porphyry diorite Granodiorite Adamellite
1 2 3 4 7 8 9 Specimen number: S6820 S8220 S13410 S9160 S6258 S728S S1822 .
Gunong Gebong, Sungai Baki, Simuja Quarry, South side Abok Quarry, Gunong Gading. Tinteng Bedil, J,ocaliiy : near Sematan near 32nd mile, near of near Lundu near
Serian Road Serian C. Sirenggok Silantek b g g a
sio, 49.58 50.50 52.70 65.50 70.50 71.07 72.11 Tio, 0.30 1.38 1.46 0.34 0.16 0.55 0.24
18.88 16.70 14.70 14.74 14.80 13.61 12.80 Fez03 0.86 1.22 2.10 1.20 0.38 0.62 1.51 Fe0 4.99 8.35 9.31 1.98 1.82 3.29 1.29 MnO 0.12 0.15 0.14 0.01 0.04 0.08 0.04
8.19 5.55 3.69 1.42 - 1.06 0.71 Ca0 14.20 8.20 6.77 4.15 2.43 2.32 1.77 Na,O 2.08 3.30 3.44 3.84 5.80 2.64 4.46 K,O 0.13 1.45 1.36 0.40 1.16 3.34 4.14 H,O+ 0.61 2.70 3.22 4.19 1.53 1.05 0.72 H20- 0.03 0.13 0.82 2.15 - 0.46 0.06 0.34 - 1.10 0.03 n.d. 0.44 0.06 -
- 0.22 0.22 0.10 0.15 0.13 0.02 COZ PZ06
TOTAL 99.97 100.95 99.96 100.02 99.67 99.88 100.15
AWa
MgO
- - - - References: S6820, S728S quoted from Wollenden and Haile, 1963. Table 7. S1822 quoted from Haile, 1954. Table 5. S13410 quoted from Pim. 19 65 Table 10. 58220 quoted from Wiltord. 1965. Table 8. S6258 quoted from Kirk, 1968, Table 25. S9160 quoted from Wollenden. 1965, Table 6.
TABLE 3. Analyses of sedimentary rock types in West Sarawak
Arkose Sandstone Sandstone Carbonaceous shale Marine sediment Rock name: (Triassic) (Triassic) ' (Tertiary) (Triassic) (Sub-Recent)
Specimen number: S13201 S8548 510849 S4307 S7391
Locality: Snngai Tarat Sungai Bukar Sungai Konggong. SerianISimanggang Sematan area
6 10 11 5 12
south-west of S e d a n road, 58th mile
SiO, 67.9 78.3 88.4 66.66 96.29 Tio, 0.55 10.75 0.14 0.75 n.a. -40, 14.5 0.84 4.65 18.54 1.60
1.63 1.55 0.45 1.60 ' 1.86 Fe0 2.65 0.41 0.91 - 0.02 Fe,O,
MnO 0.07 0.31 0.03 n.a. n.a. MgO 1.59 2.48 0.32 0.64 ma. Ca0 1.38 3.52 1.30 0.09 n.a. Na,O 3.05 0.17 0.87 ma. n.a.
2.65 0.91 1.05 n.a. n.a. n.a.
K,O II,O+ 2.65 0.22 0.30 n.a. &O- 0.78 0.25 0.91 n.a. n.a. CO, 0.37 0.06 0.63 n.a. n.a.
0.09 0.02 0.04 n.a. n.a.
TOTAL 99.9 99.79 100.0 88.28 99.77 - - - P206 -
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 Soh 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 pro- cesses 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 by the 7th Approxima- tion 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 much of the existing literature on the subject has employed this genetic classification. In Sarawak, on the various rock types, we distin-
guish, therefore, the following Great Soil Groups : Ileddish-Brown Lateritic soils, Red-Yellow Podzolic soils, Grey-White Podzolic soils, and Podzols. The 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. The correlation is based on a study of these soil groups by McCaleb (1959), and Stobbe and Wright (1959). Since some of the characteristics of the soils, GreyWhite Podzolic soils and Podzols. The 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 fac- tor within the group of Red-Yellow Podzolics and can frequently be related to the intensity of the podzoliza- tion 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.
2.
3.
4.
5.
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). Removal of sesquioxides from A horizons and sub- sequent accumulation in the I3 horizons indicated by mottling and iron concretions. Pale colouration of the A, horizon, coupled with correspondingly deeper colours in the B horizon (formation of albic horizon-7th Approximation). Removal of organic material down cavities; organic staining of the A, and B horizons. The formation of iron-humus accumulation horizons in the I3 position (formation of the spodic hori- zon-7th Approximation).
Table 4 shows these morphological features as found in the Great Soil Groups occurring in Sarawak.
TABLE 4
Morphological features
1 2 3 4 5
Soil groups of podzolization
Reddish-Brown Lateritic soils - - - - -
Red Podzolic Soils X X X x - Yellow Podzolic soils X X X x - GreyWhite Podzolic soils X X X x -
IIumus Podzols X 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 down to Humus Podzols the morphological evidence increases. When following the common definition that in laterization silica is selectively removed from the A horizons then in Sara- wak 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). Tan 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 Reddish- Brown Latosolic soil. For comparison the description and chemical data are given below.
Profile 9. Karangpandan. Intersected region. Altitude : 600 m. Flat and horizontal part of a hill. Vegetation: Manihot utillissima. Mean annual air temperature : 22.70 C. A, 0-12 c m Reddish-brown (6YR 5/4 to 5YR 4/3)
clay with well-developed crumbly to nutty structure. Friable. M a n y roots.
BI 12-31 c m Reddish-brown (5YR 5/4 to 5YR 4/3) clay with well-developed fine nutty struc- ture. Iron coatings. Friable. Few roots.
n, 31-90 c m Reddish-yellow (5YR 6/6) to reddish brown (5YR 4/3) clay with well-dev- eloped medium nutty structure. Iron coatings. Friable. Termite and ant acti- vity.
n3 +90 c m Light-brown (7.5YR 6/4) to reddish brown (5YR 4/3) clay. Greyish, strongly weath- ered stones are present. Friable.
98
Analyses of ProJile 9
Molecular ratios of the colloidal iraciion
SiOz/R.LOa SiOz/AlzOa SiOzIFezOa AIzOalFez03
A, 0-12 cm 1.58 1.89 9.70 5.14 13, 12-31 c m 1.63 1.93 10.26 5.31
31-90 c m 1.67 1.97 11.12 5.66 13, > 90cm 1.73 2.03 11.82 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:
Mohr and van Baren (1954) in their review on late- ritic soils consider all boils with sesquioxide accumula- tions to be lateritic. These accumulation horizons m a y be found either at the surface or at depth. They 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. The observation by Mohr 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 way to becoming a lateritic soil (in the sense of Mohr 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. From the available evidence one must conclude that Tan and van Schuylen- borgh’s Reddish-Brown Latosolic soil and the Reddish- Brown 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 tex- tures 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. The analysis suggests that in all the Red-Yellow Podzolic soils sesquioxides in- crease with depth. Therefore, both the morphological and chemical features show that a podzolization pro- cess is involved. Organic staining is only visible in the very pale coloured soils, such as the GreyWhite Pod-
Jnfluence of the nature of parent rock on soil formation
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 podzoliza- tion 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 pro- cess 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 correla- tion exists between the genetic soil types and the nature of the rock types. Within the groups, owing to dif- ferences 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. The total cation content of the rocks comprises the s u m total of Ca, Mg, K and N a oxides. It is logical to assume that these on their release influence the com- position of the humus and acidity of the leaching agent. There appears to be well-expressed relationship between the lateritic-podzolic range of soils and the range found in the s u m total of rock cations.
Secondly, since Figure 1 and field evidence sug- gested that texture plays an important role in the inten- sity of the podzolization process, total Alzo3 and 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 much 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 composi- tion 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 deve- lopment. The clay mineralogy and the sand mineralogy, however, indicate that the examples quoted are all in an 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-brown Lateritic soilr Podzols
Red Podzolic soils Yellow Podzolic soils - Micro- grano- liorite Ibok (7)
-
Dlivine ßasalt Andesite Gabbro
Dacite Grano- porphry diorite
4damellite ;ranite
lagoi (9)
Jebong Tarat Tarat (1) (2) (3)
Gumbang Gading (8)
Quartzitic sandstone
Silantek (11)
Quart- zitic alluvium
Pueh (12)
(4)
Grey- White Podzolic soils
Carbon - aeeous shales
Kerait (5)
General decrease in weatherable minerals
Arkose
Serin (6)
Sandstone
Nyalau (10)
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.
TABLE 5. Relation between the chemical composition of rocks and soil characteristics
Soil colour Total cations Total Al208 Total quartz Texture soil Total iron content rock rock rock rock soil groups
Reddish-Brown Jebong 24.60 18.88 nil Clay 5.85 Red- Lateritic soils Tarat 18.50 16.70 nil Clay 9.57 Red
Tarat 15.26 14.70 4.4 Clay 11.41 Red
Red Podzolic Soil8
Gumbang Gading Serin
9.81 9.36 8.67
14.74 30.18 Clay-clay loam 13.60 34.4 S.cl.-s.cl. loam 14.50 23.50 Clay-clay loam
3.18 Red-yellow 4.91 Red-yellow 4.28 Red-yellow
GreyWhite Podzolic 6Oih
Kerait 2.45 18.54 n.a. Clay 0.75 White ,
Yellow Podzolic soils
Abok Jagoi Nyalau
9.23 11.08 6.31
14.80 26.1 Clay-clay loam 12.80 35.9 S.cl.-s.el. loam '
10.75 n.a. Sand-s.cl. loam
2.20 Yellow 2.80 Yellow 2.39 Yellow
Podzols Silantek 3.49 4.65 ma. Sand-sandy loam 2.36 White-pale yellow Pueh 0.25 1.60 n.a. Sand 1.86 White
100
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 conse- quence. It m a y indicate that neither altitude nor cli- mate can be regarded as being solely responsible for such a sequence. This was also indicated by the studies of Tan et al. (1959), who found that under a monsoon climate but with andesitic material laterization moved to higher altitude while under a continuously wet cli- mate 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 dif- ference in parent material m a y well be responsible for this. Also, although from their study of the chemical composition of the humus they arrived at the conclusion that the differences are related to an increase of humi- fication at higher altitudes and an increase.in min- eralization at lower altitude, they suggest that in the case of their Red-Yellow Podzolic on acid parent mat- erial at sea level humification was caused by the acidity of the parent material.
From these studies, one may, therefore, conclude that the functions of climate and parent material in soil for- mation are interchangeable and are interacting. If w e accept the fact that the rate of mineralization or humi- fication is responsible for either laterization or podzo- lization, another conclusion can be drawn, namely that humification and mineralization are as much dependent on the nature of parent material as on climatic factors. The 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. The Reddish-Brown Lateritic soils usually have no O horizons.
From these facts the following theory is propoun- ded.
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 tem- 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 move them down the profile. In Sarawak, clay mineralogical studies have shown that a young soil on basalt (an immature Tarat beries) contains largely gibbsite and goethite. The mature Tarat series contains mainly Kaolinite with sub- sidiary 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 alu- minium 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 subse- quent impoverishment of the humus. This m a y explain why in almost similar soil types a change from lateriza- tion to podzolization can be noticed (cf. R a n and Schuylenborgh’s Reddish-Brown Latosolic soil and Sarawak’s Reddish-Brown Lateritic soil).
Similarly, podzolization on increasingly more acid rock types m a y start at an earlier stage in the matura- tion 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 con-
tinuously 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 land- scape m a y have undergone such changes that this stage is never reached. However, the lateritic sheets (which are called fossil by Mohr 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 lateriza- tion, an opinion also held by Thorp (1933, quoted from Mohr 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 in- creasingly more acid parent rocks. Pickering (1962) from his leaching studies concludes that laterization on 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 an 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). The resulting increase in the formation of fulvic and humic acids will acidify the topsoils rapidly. In an early stage of soil formation, even on basic igneous rock types, podzolization will then occur. The great difference between these processes in the tropics and in the temperate regions is that in the tropics weathering is intense to great depth; parti- cularly 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 medium 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 on 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 aci- difying effect of the climate through the higher rate of humification is more pronounced in the temperate . regions and that in an earlier stage in soil formation than in the tropics. Laterization m a y follow podzoliza- tion but never the reverse under normal conditions. The 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 show the influence of basic parent rock on lateriza- tion at high latitudes at one extreme, while the humus podzols in the tropics, formed largely on 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: 150 ft. Vegetation : Parent material: Andesite. Topography: Rainfall:
Young rubber with dense undergrowth with fern dominant.
Broken terrain, strongly dissected. Slope 2.50. f 140 in. annually, 7.07 lowest mean monthly.
Description:
Al
Very thin layer of leaf litter (1/10 in.). Reddish-brown (5YR 4/4), clay, fine subangular blocky to fine angular blocky structure (nutty). Dry. Friable. Abundant rootlets. Porous. Distinct boundary to
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
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
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).
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.
0-4 in.
BI 4-15 in.
B* 14-24 in.
B3 21-52 in.
C 51-80 in.
1 o2
Influence of the nature of parent rock on soil formation
GUbIBANG SERIES
Altitude : Vegetation: Parent material: Topography : Rainfall:
'. Descripiion:
O O-& in. A, 4-9 in.
BI 9-20 in.
Ba 20-40 in.
Bdd 40-60 in.
C 60-74 in.
C 74-90 in.
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.
Partly decomposed litter of fern origin mainly. Yellowish brown (1OYR 5/8), clay, weak subangular blocky to crumb structure. Moist. Slightly plastic. Firm. Many rootlets. Weak 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 Reddish yellow (7.5YR 6/8), clay, with sporadic quartz grit (2-3 rnm 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 powdery material in weathered rock. Clayskins moderately developed on old ped faces. Illuvial clay mainly coating larger cracks evidenced by colour diíïerence. Weathered rock approximately 20 per cent of soil volume. Indistinct boundary to Similar to the horizon between 9 and 20 inches but clayskins are stronger developed. Weathered colluvial boulders and gravel 40-50 per cent in volume. Indistinct boundary to Reddish yellow (7.5YR 6/8), 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 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 Red (2.5YR 5/8), soft, silty, very friable thoroughly weathered porphyritic Dacite, many white phenocrists.
GADING SERIES
Altitude: Vegetation: Parent material: Topography : Rainfall:
Description:
Al 0-5 in.
AZ ' 5-16 in.
BI 16-44 in.
13, 44-56 in.
C 56-60 in.
50 ft. Old rubber garden with weed undergrowth. Granodiorite. Mountainous terrain. Foot slope, 150. f 132 in. annually, 5.56 in. lowest mean monthly.
Dark brown (10YR 4/3), weak subangular blocky sandy clay loam. Friable. Organic matter present. Weil rooted. Gradual but distinct change to 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 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. Yellowish red (5YR 5/8), sandy clay loam, in places gritty. Weak crumbly. Slightly firm. No roots. Gradual change to 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:
Description:
O 0-1 in. Al 1-2 in.
A, 2-13 in. BI.* 13-46 in. BI 46-52 in.
Remarks:
150 ft. Old rubber garden (seedlings) mixed with secondary growth. Microgranodiori te. Broken hilly terrain. Slope 250. f 160 in. annually, 8.96 in. lowest mean monthly.
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 (1OYR 7/8), 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 concre- tions. Large roots. Moist. ,
Distinct iron accumulation in B,.
SILANTEK SERIES Altitude: yegetation: Parent material: Topography: Rainfall:
Description:
0-2 in.
2-5 in.
5-9 in.
9-13 in.
13-18 in.
18-22 in.
22-33 in.
33-52 in.
52-68 in.
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.
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 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 Reddish grey (5YR 5/2), medium sand (humus stained). Few roots. Moist. Single grain. Firm. Clear but wavy change to Light grey (IOYR 7/1), medium sand with reddish grey staining in places (75 per cent light grey- 25 per cent reddish grey). Single grain. Firm. Some veins of humic material run through this ,horizon without any apparent direction. No roots. Abrupt over 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 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
Very pale brown (1OYR 7/3), loamy sand to sandy loam. Compact. Structureless. Slightly wet. Many old root channels with organic material (5 YR 2/2). Some organic material accumulated along fracture planes. Clear change to
'
Pale yellow (2.5YR 8/4), sandy clay with (10 YR 6/6), mottling, in some places as lateral bands in others along old root channels. Sticky and plastic. Some quartz pebbles (rounded), at 44 in. becoming more sandy and resembling sandstone.
. White medium sandstone. (Deep augering confirms occurrence of white clay bed at 68-76 in.). Perched water table at 48 in.
1 o4
Influence of the nature of parent rock on soil formation
rumr SERIES Altitude: Vegetution: Parent material: Topography : Ruinfal:
Description : O-$ in.
Al 8-7 in.
A2 7-14 in.
A, 14-15 in.
Bh(ir) 15-19 in.
BI&) 19-24 in.
BZ.1 24-44 in.
Bz.2 - 44-54 in.
, B(mn) 54-60 in.
10 ft. Bracken type ferns. Sub-Recent beach sand. Flat, raised beach. & 132+ in. annually, 5.56 in. lowest mean monthly.
Light grey (10YR 7/1), medium sand. Loose. Very dry. No organic matter. Abruptly overlying Light brownish grey (10YR 6/2), medium sand. Loose. Single grain structure. Many roots. Many dark grey infilled worm holes. Slightly moist. Gradual regular change to
White (10YR 8/2), medium sand. Loose. Structureless. Slightly moist. Very small roots stopping at 12 in. Brown mottled white sand, boundary irregular, sloping down with slightly steeper inclination than surface slope.
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 Reddish brown (5YR 4/4), medium sand. Structureless. Firm. Common dark red concretions and mottles (as 15-19 in. but colour less brown). Slightly moist. Clear, very wavy change to Strong brown (7.5YR 5/8), Cine sand. Firm. Structureless. No roots. Slightly moist. Gradual wavy change to 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 Light olive brown (2.5 YR 5/6), medium sand. Wet. In places strongly stained black to strong brown.
JAG01 SERIES
Altitude: Vegetation: Parent material: Topography: Rainfall:
Description : o . 0-5 in.
Al 4-4 in.
AZ 4-20 in. Bl 20-34 in. BI 34-46 in.
C 46-55 in.
350 ft. Secondary forest. Adamellite. Mountainous terrain, strongly dissected. Slope 15". Estimated between 140 and 160 in. (no records).
Mainly coarse litter. 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 Light olive brown (2.5Y 5/4), sandy clay loam. Crumbly. Firm. Many rootlets. Gradual change to Yellow (10YR 8/6), sandy clay loam, angular blocky. Firm. Clear infiltration of clay. Gradual change to Yellow (10YR 7/6), sandy loam clay to sandy clay. Very firm. Greenish small mottles. Few roots. Moist. Gradual change to
Yellow (1OYR 7/6), sandy (gritty) clay. Weathered granite embedded in sandy clay. Limonitic hard iron present and red coloured soft iron accumulation. Few fine root#.
105
J. I’. Aiidriesse
KERAIT SERIES
Altitude: Vegetation: I’ararent material: Topography : Rainfall:
Description:
A, 0-3 in.
A, . 3-15 in.
u, 15-30 in.
B m 30-60 in.
B2 60-100 in.
B/C . 100-172 in.
C 172+ in.
f 150 ft. Secondary growth, mainly ferns (bracken type). Carbonaceous shales. (Triassic). Broken hilly terrain. Slope 150. 4 140 in. annually, 7.07 in. lowest mean monthly.
Light grey (2.5Y 7/2), sandy clay loam. Weak humus staining from 2-3 in., surface gleying also present. Weak platy structure, densely rooted, moist. Clear regular boundary to
Light grey (2.5Y 7/2), sandy clay loam with faint few light grey (1OYR 7/1), 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
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.
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 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 Light grey (IOYR 7/1), silty clay, massive and very compact. No cracks or roots, slightly moist. Abrupt irregular boundary to Soft, easily cut black shale with quartz-strings. Inclusions of fossil roots and olive yellow coloured pyritic material particularly along fracture planes.
SERIN SERIES Altitude : 150 ft. Vegetation: Parent material: Arkose. Topography : Broken hilly terrain. Rainfall:
Secondary forest (many wild rambutan and durian trees).
f 140 in. annually, 5.97 in, lowest mean monthly.
Description:
A,
A,
0-4 in. Dark greyish brown (1OYR 4/2), clay loam. Abundant roots and rootlets. Friable. Weakly developed subangular blocky. Moist. Irregular boundary Brown to strong brown (5YR 5/6), clay loam with many fine to medium brownish and greyishmottles. Moderately developed subangular blocky. Slightly firm. Slightly porous. Moist. Distinct wavy boundary
Reddish yellow (7.5YR 6/8), clay. Strongly developed subangular blocky. Firm. Few roots. Slightly porous. Moist. Indistinct boundary to 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 Yellowish red (5YR 5/8), clay. Strongly developed medium blocky structure. Firm. Abundant fine weathered arkose. Non-pÖrous. Diffuse boundary Yellowish red (5YR 5/8), clay, with few very fine yellow mottles. Strongly developed medium blocky structure. Firm. Abundant weathered arkose. Moist.
.4-10 in.
to
BI
BI.*
B*.2
10-34 in.
34-44 in.
44-56 in.
56-68 in. I B/C
106
Iníiuence of the nature of parent rock on soil formation
NYALAU SERIES Altitude: Vegeiaiion: Parent maierial: Topography : Rain fall:
Descripiion:
O 0-2 in.
A, 2-6) in.
As 64-22 in.
Ba 24-43 - in.
B/C 45-54 in.
Remarks :
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.
Surface scattered with litter of varying depth. (4-2 in.). Humus containing abundant fine to medium roots, leaves in varying stages of decomposition. 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 Pale yellow (2.5Y 7/4), sandy clay loam. Common diffuse greyish brown (10 YR 6/3), mottles. Pale brown coloured material from Al leached through root channels and cracks. Subangular blocky. Friable. Few coarse and medium roots. Moist. Difluse change to 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 As 24-43 in. but with also common white and strong brown mottles. Massive. Subangular blocky. Firm. Moist to very moist. Distinct O and A, horizons.
.
ANALYTICAL DATA OF A RANGE OF SARAWAK SOIL SERIES ON INCREASINGLY MORE ACID PARENT ROCKS-overleaf
107
-ri--
0000M
dridri
00000
Mc.lcJM
wmm
I??
wmm
or
im
mc-m *.
O??
ri
t! s: w 4 B 4 a 2 k O
108
I0,OdH
I O000
O00
00000
el
***el
II O r, U
bi
- o d
c - VI
M
&me t- a
d r
m
&0000,+0d
00 0 00 o
II m
ON
'm000t-
2373977
0n
)m
oo
oe
l
rlmwot-
Fld
r - El
II di
r( 109
J. P. Andriesse
Bibliography
ALEXANDROVA, L. N.; ARSHAVSKAY, T. Th.; DORFMAN, E. M.; LYUZIN, M. F.; YURLOVA, O. V. 1968. Humus 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 cha- racteristics of tropical podzols in Sarawak (East Malaysia). Geoderma, vol. 2, p. 201-27 (to appear in 1970).
L- . 1969b. The development of the podzol morphology in the tropical lowlands of Sarawak. (In press.)
DEPT. OF CIVIL AVIATION 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, McGraw- Hill. 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.) Ann 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
.
quartz-free silicate rocks and their contribution to an under- standing of laterization. Econ. Geol. (Lancaster, Pa.),
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,
TAN, K. H.; VAN SCHUYLENBORGII, J. 1959. O n the clasei- fication and genesis of soils, derived from andesitic volcanic material under a monsoon climate. Neth. J. agric. Sci.,
-. 1961. On 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 map of China with notes on Chinese soils. Trans. 3rd Int. Congr. Soil Sci., vol: 1, p. 275-6 (quoted from Mohr and van Baren).
fications : Order, Suborder and Great Soil Groups. Soil Sci.,
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.)
Sarawak. Geological Survey Department, British Territories in Borneo. (Report 1.)
vol. 57, p. 1185-206.
p. 161-4.
vol. 7, p. 1.21.
-. , SMITH, Guy, D. 1949. Higher categories of soil classi-
vol. 67, p. 117-26.
-. , IIAILE, N. S. 1963. Sematan and Lundu Area, West
'110 ,
The differences between crusts of weathering aiid soils developing on acid .and basic 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.)
The 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 many 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 Mohr and van Baren (1954). The crusts of weathering and soils formed on the basic volcanic tuffs have, in their turn, very much in common with the margalitic soils.
It is evident, therefore, that the character of the pro- ducts 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
known to be distinguished by a strongly acidic reac- tion, 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 in- adequate cation saturation, considerable anion-absorb- ing capacity. These properties of ferrallitic soils should be sup-.
plemented by a great predominance of fulvoacids in the organic matter. The 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 fer- rallitic 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 medium high mountain zone and mountain allitic-humus boils of the high mountain zone (‘foggy forests’).
Apparently, a decrease in temperature with altitude slows down 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). The 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
726 ‘length of waves
665 niir
t 574 533 496 465
FIG. 1. Optic density of humic acids in : A-dark-red ferrallitic soil on basalt; B-mountain allitic-humus soil on gneiss; Cmountain humus-ferrallitic soil on rhyolite.
% The crusts of weathering of the acid rocks are dis- tinctly subdivided into several horizons. The lower one is notable for a considerable amount of primary minerals though metamorphosed but still possessing certain significant initial properties. The medium horizon contains only the most stable primary minerals such as quartz, ilmenite, etc., because other primary minerals were already transformed into the minerals of inter- mediate stages of weathering and, first of all, into hydromicas. In this horizon there are also some appre- ciable quantities of kaolinite, halloysite and minerals- iron and aluminium hydroxides. The upper horizon is fully composed of the most stable primary minerals, minerals of the kaolinite group and minerals-hydro- xides. The 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 humus 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
-
the entire thickness such crusts of weathering pohselih the structure of a well-structured (granular-cloddy) masn greatly differing from that of the primary rock. The soil-forming process going on in the upper part of the weathering crust does not essentially affect its structure, and the soil, from the standpoint of its mor- phology, 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 con- trary to that of the acid rocks where weathering is a phasic (stadial) process affecting considerable thickness.
Even under good drainage conditions the weathering erusts of acid rocks have a mottled colouring (reddish, yellow, bluish) caused by the gleization and bydra- tion processes and responsible for their widely known name ‘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 weather- ing 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 red-
yellow ferrallitic soil formed on them have a con- siderably greater volume weight than that of the basalt products of weathering and the dark-red ferrallitic soil. The difference in the specific weights of these rocks and soils is not so well pronounced which stipu- lated 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 con- tent in the soils and rocks equal to the maximum field moisture capacity, differ even more considerably (two- to threefold).
The above differences reach their maxima in the crust of weathering and decrease in the humus horizons of soils which makes it possible to assume that they arise in the process of weathering crust formation and bc- come less distinct with the development of soil-for- mation process. The results of long-term investigations on the mois-
ture 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 down to ’a depth’of 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 maximum 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-
- .
112
Differences between crusts of weathering and soils
cially) developed on acid rocks are unsatisfactorily aerated contrary to those developed on the hasic ones. Consequently the reduction regime and gleization pro- cess, responsible for the specific mineralogical and chem- ical features, are characteristic of the former, whereas the latter are characterized by the oxidation regime conditioning their specific mineralogy and chemism.
Therefore, the destinations of the primary rocks manifest themselves not only in the chemical and mine- eralogical but also in the physical properties of the weathering crusts and soils formed on them and, first of all, in the character of their structure. The following investigation on the structure of fer-
rallitic 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 made by employing the methods of dry and wet sifting (Figs. 4(a) and 4(b)). The dry sifting fails to reveal the difference between the two groups of fer- rallitic soils as both of them are characterized by rather big structural separates (more than half of the total mass of soil consists of separates larger than 2 mm).
The wet sifting which preserves only water-stable struc- tural separates has confirmed the minimal differences between the two groups of ferrallitic soils and has reveäle'd the high stability of structure : only 15- 20 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. The mountain humus-ferrallitic soil and especially its upper part, richest in humus, is characterized by the biggest aggre- gates. Smaller structural separates are typical for the red-yellow ferrallitic soil. The smallest structural separates (prevailing sizes from 0.5 mm to 2 mm) are characteristic of the dark-red ferrallitic soil. Thus, the weakly aerated red-yellow ferrallitic soils
developing on acid rocks have in the humus horizon similar or even more stable structural separates than the well aerated dark-red ferrallitic soils developing on the basic igneous rocks and marbles. Hence, it is possible to conclude that the differences in physical properties are stipulated not by various degrees of water stability but by diverse structure and arrange- ment of structural separates. Weathering of basic rocks results in the formation of loosely connected macro- porous structural separates with considerable spaces between them.
TABLE 1. 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)
Total Aeration Volume weight ~ Specific weight
Depth ícm)
A B A B A B A B
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
1.17
1.22
1.22
1.26
- - - - - - - - 1.24 1.28
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
2.78
2.80
2.77
2.80
- - - - - - - - 2.79 2.80
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
58.0
56.5
56.0
55.0
- - - - - - - - 55.5 54.3
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
15.7
11.7
11.6
8.7
- - - - - - - - 10.7 8.5
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
113
V. Til. Fridland
x
w
e . . ** 4 " S .e**< i (4 (b)
FIG. 2. Structural separates of the humus 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 ( x 100) : (a) dark-red ferrallitic soil on basalt; (b) red-yellow ferrallitic soil on gneiss.
._ , ..
114
100
50
O
50
O
50,
al c m E
~ 2,
Differences hetween crusts of weathering and soils
- A, 0-15 - B, 0-22 ---- e, 0-5 --- D, 2-7
- A, 15:30 - B. 22-35 _--_ c, 5-ta -- - 0, 7-15
L A, 30-50 - S, 35-60 ---- C, 18-29 --- D. 15-45 ¿O25 0.5 1 2 3 5 7 1 0 1 5 20.25 0.5 I 1 2 3
íb) Logarithms of fractions (diameters, mm) (a)
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 intensive removal of substances in the course of weather- ing of basic rather than acid rocks.
The character of structural separates is significantly influenced by the chemical composition of substances cementing the structural separates. A special investiga- tion employing the method proposed by Antipov- Karataev, Kellerman and K h a n (1948) has been carried out to elucidate this problem. Water-stable structural separates from 3 mm to 5 mm in size were selected from the fioils being investigated, placed on glass discs and subjected to successive treatments
with various soIvcnts. In view of the possible destruc- tive influence on 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. The treátment of samples was conducted in consecutive order by the following reagents : 1. Alkaline buffer mixture of decinormal sodium hy-
droxide and sodium oxalate at a ratio of 1 : 4 remov- ing the loosely connected and free organic matter.
2. Tamm's reagent removing the slightly crystallized sesquioxides.
3. Normal sulphuric acid extracting the stronger crys- tallized 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 humus ferrallitic soils are less stable than those of the dark-red ferrallitic ones;
Iron oxides act as cementing substances when the struc- ture of red-yellow and dark-red ferrallitic soils is being formed;
A significant role in the structure formation of moun- tain humus-ferrallitic soils is played by organic substances.
It will be very interesting to study by this method also deep horizons of soils and crusts of weathering. Ilow- ever 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.
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.
It should be pointed out in conclusion that the 1
Bibliography
ANTIPOV-KARATAEV, I, N.; KELLERMAN, V. V.; KHAN, D. V. HARRISON, J. ß. 1933. The katamorphism of igneous rocks 1948. About soil aggregate and methods of its study, p. 42-3. Moscow, Publ. House Acad. of Sciences. (In Russian.) LACROIX, A. 1913. Les latérites de Guinée. Paris, Nouvelles
FRIDLAND, V. M. 1964. Soils and crusts of weathering of humid tropics, p. 208-27. Moscow, Publ. House ‘Nauka’. (In MOIIR, E. C. J.; VAN BAREN, F. A. 1954. Tropical soils. Lon- Russian.)
under humid tropical conditions. IIarpenderi (U.K.).
Arch. Museum histoire natur., ser. 5, t. 5.
don, Interscience Publishers, 498 p., tabl., fig., pl.
116
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
The Faculty of Tropical Crops of the Royal Univer- sity of Kompong-Cham 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 Kompong-Cham 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 inves-
tigation pertaining to the formation and weathering of these soils could not be overemphasized.
PARENT MATERIALS
The basaltic block of the Kompong-Cham province is subsequent to the former alluvion. This assertion sprang from the drilling of wells at the University of Kompong-Cham 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 Qua- ternary era. Thickness of the basaltic layer ranges from 21 m, at the University in Kompong-Cham, to 125 m in Chamcar-Krauch, 30 km 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 km from the city of Kompong-Cham, one can observe scoria cones and belt of volcanic ash.
Due to different states of decomposition, nodular rocks and gravels of various sizes can also be found.
Current minerals of basalt of Kompong-Cham are found as follows : Alkaline plagioclase feldspath: This is a mixture of albite (Na,O; Alzo,, 6 SiO,) and of anorthite (Cao,
i I
Al,O,, 2 SiO,) with a predominance of the latter (more than 60 per cent).
Augite: This is a metasilicate with approximate com- position Ca M g (SiO,),.
Olivine : Magnesium silicate (2Mg0, SiO,) in mixture with iron silicate (2 Feo, SiO,) at various molecular proportions.
Magnetite : Fe304 or (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 made in Chup and that of Baleine was made at the university (see Table 1).
TABLE 1. Percentage composition of basait
From Chup From the University by M. Raoult by O. Baleine Analysis of
Lost on ignition SiO, (free) Silicate (total) Fe0 Fe,O, ALO, Tio, Ca0 MgO M n O PZ 0 6 Na,O K,O LiO, TOTAL :
0.69
46.98 8.46 3.01 13.67 1.98 10.38 ,
9.73 0.17 0.51 3.18 1.59
-
0.00 I - 100.35
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 -
1. Specially preaented as supporting paper to review No. 3 by J. van
2. O. Baleine, Faculté Polytechnique de Mons (Belgium). Schu ylenborgh.
117
Tan Doun Suy
WEATHERING OF THE PARENT MATERIALS
Transformation of basaltic mineral rocks is as follows : Olivine -+ iddingsites + iron hydroxide
I Degradative agents of parent rock are of physical and chemical nature.
(PhylÏitic mineral, &th a reflexion at 15.6 A, and goethite are components of iddinesite.)
PHYSICAL AGENTS
Y ,
Anorthite + muscovite (white potassium mica) + epidotite (hydrated silicate of Ca, Al,
Fe)
+ clay + calcite (Caco3)
+ bauxite
Physical degradation starts during the process of cooling of the lava which on contraction splits into blocks. Degradation continues by the alternating effects of humidity and dryness, combined with the great variation of temnerature. I
Augite -+ chlorite (hydrated silicate of Al, Fe, Mg) 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.
I-+ epidotite + calcite + limonite + quartz
Magnetite+ oligiste (red hematite) Fe203 + limonite (brown hematite) Fe, (OH),.
CHE M I CAL A GE N T S
Rain-water lodged in cracks is considered as the main chemical weathering agent. Water is a very weak electro- lyte, dissociated into H+ and OH-. The dissociation of this weak electrolyte is however enhanced by a rising temperature; at 340 C it is four times as active as at 160 C (K16 = 0.63 x lO-l4; K, = 2.51 X
Demolon (1960), stated that the proton in water has a potential kinetic energy when lodged in the soil, as it causes the hydrolytic reaction to go on indefinitely. This process of decomposition would reach its maxi- mum intensity in a tropical climate. On the other hand, free carbonic acid in water contributes to increase the dissociation constant of the medium 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 gen- erally by all intermediate acidic compounds stemming from oxidation of organic materials.
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.
Under conditions of good drainage
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).
The 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.
TABLE 2. Weathering of basalts of Kivu
Primary minerals First layer decomposition Firnt period clay Soil clay
Good drainage Plagioclases Augite IIalloysite Kaolini te Augites + + kaolinite + gibbsite Olivines (decomposed to iddingsites and iron hydroxyde) of decomposition)
fiel + augite (in the process
Poor drainage Plagioclases Augites Olivines (not decomposed) 4
I
Plagioclases Montmoriilonite Augites + gibbsite Montrnorillonite -I- halloysi te
Montmorillonite + halloysite
Genesis and evolution of red and black basaltic soils in Camhodia
TABLE 3. Chemical analysis of'red soil, by Baleine (samples from the University)
Analysis Basalts Iled soil
Loss on ignition SiO, (free) sio, (total) Fe0 Fe203
.Tiol Ca0 MgO M n O p20, Na,O K,O Li02 CO,
TOTAL
A1203
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 0.00
98.48 -
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
100.05 -
Mineralogical identific hasation been performed
In Cambodia, samples extracted at the university , with an X-ray diffractometer.
by Deresmaux (1968), have been analysed in Belgium by Baleine and R0bazynski.l (See Table 3.)
The moderate leaching state of alkaline and alkaline- carth 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 diKrfractometer studies : 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 alkaline- carth metals are progressively exported from the solid mass. There remains a mixture of hydratcs of Fe and Al forming laterite completely free of soluble bases.
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 on ignition arc assumed to he caused by the presence of trihydrate gibbsite (laterited gibbsite) in the bauxite.
Under conditions of poor drainage
The conditions of poor drainage on 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 altera- tion. The former enriches in the subsequent layers, whereas the latter two 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 Kompong-Cham, wherever there exists a non-
permeable underground, black basaltic soil (named basaltic regurs by Crocker (1962), forms a barrier completely isolating masses of red soils. The profile shown in Figure 1 gives a clear aspect of such morphology (Henry, 1931).
Samples of soil extracted in Mien, distant by some 20 km from the University and analysed by previously named workers in the field, have yielded the data shown in Table 5. .
The ratio, 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 alkaline-
earth metals gives rise to the formation of montmo- rillonite. (This has been found by X-ray diffraction studies.) The 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, Faculté do Polytechnique de Il&, Belgium.
TABLE 4. Analyses carried out on samples from AZondulkiri, by the Bureau de Recherches Géologiques et Minières, Paris
Sample 04, Si02 (total)
yo Sioz (quarzt)
% Fe30z
% Ti02
y, Lops on ignition
17 $1 (lateritic basalt) ' 17 N (lateritic red soil) 18 A (lateritic gravel) 18 II (lateritic basalt)
46.60 1.60 0.11 20.50 30.80 4.30 - 33.00 43.40 1.70 0.20 23.70 50.00 1.10 0.03 17.20
3.40 27.80
3.20 27.00 2.80 29.00
5.20 24.60
119
Tan Boun Suy
Red soil
Basalt or dominant conglomerate
Black or brown-red soil
Black mil with basalt or conglornasts
FIG. 1. Profile of Kompong-Thmâ-Chamcar-Krauch.
TABLE 6. Mean values of analytical characters of red soils from different regions
Exrhangeable bases in meq. % d.s. PII Site Number
of ' P blocks p.p.m. , studied K Na Ca Y g S T V 1120 KCI. N KCI
N/SO
Prekkak
Ch. Andong
Ch. Loeu
Tapao
Chup
Mimot and Prek Chhlong
Chalang
Snoul
6
17
9
5
17
9 + 2
5
'14
253
1122
1003
2 009
296
330
210
456
166
0.23 0.04 1.75 0.32
0.42 0.09 4.61 0.66
1.12 0.03 7.57 2.05
0.64 0.11 6.86 1.38
0.25 0.05 2.60 0.68
0.16 0.06 0.32 0.36
0.22 0.08 1.07 0.72
0.22 0.04 1.61 0.67
0.22 0.04 1.35 0.59
2.35
5.78
10.78
9.00
3.58
0.89
2.10
2.54
2.18
8.4 25.2 4.9 4.4 4.4
10.7 52.2 5.0 4.9 4.8
13.2 81.2 5.8 5.7 5.7
12.7 70.3 5.8 5.4 5.5
7.9 44.2 4.8 4.6 4.6
10.4 8.6 4.2 4.0 3.9
8.4 25.0 4.3 4.3 - 4.2
9.1 27.6 4.7 4.5 4.5
8.4 25.5 4.8 4.4 4.3
120
Genesis and evolution of red and black basaltic soils in Cambodia
TABLE 5. Chemical contents of basaltic regurs (by Baleine)
Compact besalt Basaltic regura
Loss on ignition SiO, (free) SO, (total) Fe0 Fe#, A1208 Tio2 Caü Mg0 M n O p*os Na,O
. K20 Li02 CO,
TOTAL
4.34
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
1.28 9.79 17.09 45.11 1.03
12.07
0.96 1.68 0.33 0.19 0.53 0.36 0.00 0.00
24.81
2.84
99.88 99.70
- It is 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 form- ed and then degraded into kaolinite.
CHARACTERS OF HED BASALTIC SOILS
From the pedological standpoint, red basaltic soils are classified, after Aubert and Ducbaufour (1960) into fer- rallitic 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 among the world’s pro- ducers 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 Thay Seng (1966), has even reported in certain areas a depth of 60 m.
The depth of the red soil is crucial for a good pene- tration of the tap-root of Hevea which, during the dry season, must reach the underground water.
Texture
Rambeaux and Danjard (1963), have reported a clay content of red soil ranging from 62 to 85 per cent.
The component clay plus loam is of a high value (82 to 92 per cent) and appears to be relatively constant.
The 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 infiltra- tion 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 ní I CAL C H A RA C T E RS Table 6 shows data obtained from soil of different areas grown with Hevea (Rambeaux and Danjard, 1963).
The optimum value of pH is around 4.9 for a good growth of Hevea. This is a condition which is generally found in Cambodia.
The pH of more recent soils is higher than the opti- m u m limit. Correction can however be achieved by the use of ammonium 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.
COVER-PLANTS 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. The following cover plants have been tried for such pur- poses ; Tithonia diversifolia, Trixacum Paxum, Illimosa invisa, Eupatorium odoratum, and graminacea, Mimosa inoisa appears to yield the best effect.
Mulching with cutdown cover plants, (luring the dry season, has resulted in a better growth of young trees (Ninane, 1969).
121
. . .. . . Tan Boun Suy . . . .
100-
90 -
80 -
70 -
60 -
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 compound 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. Even the forests which formerly covered the area cannot he reconstituted.
Deîorestation would cause a lowering of the water table. Rainfall runs off superficially instead of infiltrat-
ing into the ground. This practice resulted in a decreabed out-flow of various springs (e.g. the springs in Tuk- Chha).
A C KN O w L E D GEM'E i3~ The 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 Chun for their m a n y helpful suggestions during the prepara- tion of this paper.
Appendix
Meteorological data from the Institut des Recherches sur le Caoutchouc au Cambodge (IRCC) is given in the follow'ing five figures (Ninane, 1967). The IRCC station of the Royal University is located at a distance of 10 km, on the left bank of the Mekong river.
36
Maxima 1966
Mean 1961-66 . I .......
. >
Minima 1966
I 2 4 0 1 , , ' ; , , , , , , , J ' F ' M ' A ' M ' J ' J ' A ' S ' O ' N ' D ' , J F M A M J J A S O N D - 1966
........... Average 1961 66
Average monthly temperatures. '. Relative humidities.
122
Genesis and evolution of red and black basaltic soils in Cambodia
mrn
400.
300
200
1 O0
mrn
zoo0 10.0
9.0 1500
8.0
1 O00
7.0
500 6.0
- 1966 _ _ _ _ Cumulative 1966 .......... Average 1959.66
Rainfall in 1966 and average rainfall during 1959-66.
k E u . c ._ n
Potential evapotranspiration.
100 ! 90
Y I J F M A M J J A S O N D
Daily average insolation (in hours) and relative insolation.
Bibliography CORRENS, W. 1962. Observations sur la formation et la trans- formation de minéraux argileux lors de la décomposition des basaltes. Genèse et synthèse des argiles.. Paris, Coll. Int. CNRS.
CROCKER, C. D. 1962. Soils of Cambodia. CRUYS. 1964. Mission de bauxite. Paris, BRGM. (Private
DEMOLON, A. 1960. Dynamique du sol, p. 29, 136. DERECMAUX. 1969. (Private communication.) DUCHAUFOUR, P. 1960. Précis de pédologie. 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 à l'Institut agronomique de l'Etat à Gembloux (Belgique).
NINANE, F. 1969. Expérience de fumure dans les terrains peu acides. Commission technique 484, p. 9-10, 41. Institut des Recherches sur le caoutchouc au Cambodge (IRCC).
communication.)
- . 1967. Climatologie de la station IRCC en 1966. -. 1969. Résultats d'expériences de mulch. Commission
technique 483. Institut des Recherches sur le caoutchouc au Cambodge (IRCC).'
NIV NARIN. 1969. Les plantes de couverture et quelques données sur leur relation avec la croissance de I'IIévéa Brasiliensis. Mémoire de fin d'études, Faculty of Tropical Crops, Royal University of Kompong-Cham.
PECROT, A. ; DELVIGNE, J. ; GASTUCHE, M. C. ; VIELVOYE, L. ; FRIPIAT, J. J. 1960. L'altération des roches el la formation des sols QU Kivu. Publ. INEAC (Série Sc. no. 86).
RAMBEAUX, J.; DANJARD, J. C. 1963. Terre rouge basaltique et nutrition de l'hévéa dans les conditions écologiques du Cambodge, p. 9-10, 41. Institut des Recherches sur le caout- chouc au Cambodge (IRCC). (Opuscule technique no. 2/63.)
SOR THAY SENG. 1966. Les sols du Cambodge rt les possibilirés 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 many tropical forested areas in the Philip- pines. They are found in different climatic regions (defined by rainfall frequency and distribution), topo- graphic situations, and underlying rock types. While much is known about the morphological and chemical features of agriculturally arable lands in many parts of the country, there is a dearth of observations on this important group of soils-which support vast tracts of commercial timber forests. The following report presents the results of prelimi-
nary 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 north- east of the capital town of Daet, and some 396 km south-east of Manila (1220 30’-1220 37”E.; 140, 17’- 140 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
The province of Camarines Norte falls within the E- type 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 mean m a x imum and minimum temperatures obtaining during M a y and January respectively.
RELIEF A N D VEGETATION The 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 plateau- like when observed from the air, but the ground is deeply dissected.in many places. There are a few isola- ted 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 Bay and Dahican Bay. Most have courses that are apparently structurally controlled. The vegetation of the interstream area between
Alayao River and Tigbi River, including the moun- tainous country farther south consists of primary and second growth forests, a few open grassland covered with cogon and related species of grass and shrubs. Mangrove as well as other swamp-thriving plants are common along the coastal fringes and partially sub- merged mouths of the larger streams. The tropical forest which is distinctly dipterocarp in composition
I
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 known as ‘kaingin’ has been practised, a second growth forest .
has usually taken root.
SOIL-FORMING ROCKS
T h e principal soil-forming rocks of the Dahican- Alayao 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 much later group of rocks (Eocene to early Miocene) composed of COnglO- merates, 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 Cama- 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 how- ever, on specific lithologies are generally available in deep stream banks, and in some recent road cuts made 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 mapped as part of the Alaminos clay which is described as a reddish-brown to brick-red compact clay which becomes sticky when wet. This soil descrip- tion 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 CAL AND CHEMICAL FEATURES
In order to obtain some 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 com- position of typical soil profile was undertaken. The profile section selected is about 4 km south of the saw- mill site of the Dahican Lumber Company at an approxi-
1. Lateritic-following Mohr’s (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’.
TABLE 1. hforphological characteristics of soil profile
1Iorizon Depth Description
. I
1 0.30 cm Pale, reddish brown (1OR 5/4) clay loam, friable and granular when dry, sticky when wet. Numerous aggre- gates, colour somewhat mottled.
2 30-100 crn Light to yellowish brown (5YR 5/6 to 10YR 5/4) loam, friable and granular under dry conditions, but heavier and more compact than overlying horizon, and very sticky when wet. Less mottled than horizon 1. Upper boundary grada- tional.
3 100-300 ern Partly weathered bedrock, mod- erate 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, labrado- rite 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 them to a fine powder as required by the analytical technique. The pH values of soil from each horizon were determined using a Beckman 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.)
0’126
Some aspects of lateritic soil formation
TABLE 2. Chemical analysis of 60il profile from decomposed basalt, Alayao, Camarines Norte, Philippines
r . . I . . Snil horizon: , ' , " '
Constituenti ' . 1 (0-30 em) 2 (30-100 cm) 3 (100-300,cm)
SiOl
ni203 Ca0 MgO Na,O K,O Tio, Loss on ignition
48.81 13.56 22.25 0.52 1.19 1.21 0.23 0.90 11.41
53.00 55.23 7.36 8.47 22.98 22.18 0.54 0.56 1.79 1.35 0.08 1.75 0.24 0.78 0.70 0.82 12.58 9.03 - - -
TOTAL (yo by wt.) 100.08 99.27 100.17
p H values 5.2 5 4.95
DISCUSSION
Although no 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 shown by the fairly low percentages Cao, MgO, NA,O, K,O, compared to published chemical analysis of fresh rocks of similar mineralogical compo- sition. 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 hori- zon is not easily explained. The gradational character of the contacts between the three horizons, in addition to the tight, compact nature of the intermediate hori- zons, 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.
The percentages of the more stable components of the soils as revealed by the analysis is more consistent with their known 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-all percentage is fairly close to fresh rocks having comparable mineralogical com- positions (Johannssen, 1937; Mohr, 1944). The 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 pub- lished 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 an immaturc lateritic soil.
The soils found in Dahican and neighbouring places are generally acidic. The p H values range from 4.9 to 5.5, and in both profiles there is a slight increase in acidity with depth. The 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 make it give a n acidic reaction. The significance of the slight decrease in p H with
depth in both profiles is not known. Elsewhere similar occurrences have been attributed by Mohr (1954, p. 390-1) to the presence of bases liberated from decay- ing 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. The soil in the arca is generally moist and does not become dry for any signi- ficant length of time. It is also possible that some 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 res- ponsible for the blightly higher pH at these levels.
Bibliography DE GUZMAN, K. A. 1966. Uranium mineralization in the Para- JOACHIM, A. W. R. 1935. Studies on Ceylon soil. II. General cale mining district, Camarines Norte. Bulletin of the characteristics of Ceylon soils, some typical soil groups of Institute of Filipino Geologists, Munila, vol. 1, no. 1, p. 1-13. the Island, and a tentative scheme of classification. Trop.
GERVASIO. F. G. 1966. A study of tectonics of the Philippine Agriculturist, Ceylon, vol. 84, 'p. 254-74. Archipelago. The Philippine geologist, voi. 20, no. 2, JOHANNSEN, A. 1937. A descriptia7e petrography of the igneous p. 51-75. rocks, vol. III, p. 260-1. Chicago, Ill., University of Chicago
Press. of the Philippines, p. 44-50. Manila, Bookmark.
IIUKE, R. E. 1963. Shadows on the land, un economic geography
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. Ann 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’
INTRODUCTION
Soil classification is a special objective of pedologists who carry out soil surveys to define taxonòmic units in the field, but it also holds interest for soil fertility specialists and crop scientists w h o deal with soil man- agement and nutritional requirements of crops. The term ‘fertility’ is used here in a more restricted sense to mean the nutrient status of soils. If meaningful corre- lations 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 fer- tility 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 pur- poses has limited scope. On the other hand, in many 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; Thomas and Allen, 1966; Soil Survey Staff, 1966; and Moorman and Rajanasoothon, 1968), but their relationships with fertility status have not been closely examined. As an understanding of the pedogenetic- fertility 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 made here in the light of data available at pre- sent. This evaluation is only tentative because ana-
Ng sew Kce Rubber Research Institute, Malaya and Law Wei Min Department of Agriculture, West Malaysia (Malaysia)
lytical data are not sufficiently comprehensive for all soils.
FACTORS OF SOIL FORMATION
Factors of pedogenesis in West Malaysia are outlined in brief as follows.
CLIMATE
The Malayan Peninsula lies between the Equator and 70 N and has a hot, humid climate. D a y temperatures usually lie between 26-320 C with night temperatures about 4-50 cóoler. Annual precipitation generally ranges from 1,800 mm to over 3,500 mm (Oh, 1965), the heaviest rainfall areas being the main mountain ranges and the lowest areas being in the extreme north- west and part of the central region of the country. In the extensive lowlands and alluvial flats, rainfall averages mostly 2,000-3,000 mm. 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 mm per month in the wet season can occur (Wycherley, 1967). Relative humidity during the day is usually in the nineties.
P A R E N T MATERIALS . -
As far soil formation is concerned, the geology of West 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 rela- tively minor extent basic rocks of andesite, l~asalt and tuffs. The 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 in- significant. Finally, the alluvial deposits mainly occupy the major coastal and river-flood plains of the coun- try, although some deposits occur in inland situations. Seven main types of deposits have been distinguished, mainly according to characteristic land forms (Gopi- nathan, 1968). These are : (a) high-level terrace deposits which have been dissected and occur at elevations of about 80 m in southern and central states; (b) inter- mediate-level deposits at elevations of 16-50 m which are weakly dissected and gently undulating in topo- graphy occurring mainly in the northern states; (c) low- level 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 depo- sits mainly along the east coast and, lastly; (6) peat deposits found in depressions within the major coastal or flood plains.
VEGETATION
From the classification of Malayan forest types by Wyatt-Smith (1964), three major groups of forests can be distinguished, namely : hill and mountain vegetation; lowland (dryland) vegetation; and swamp 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 vegeta- tion mainly comprises primary Dipterocarp forests with a multitude of species but the principal being Dipterocarp and Shorea species. Under swamp and low- lying vegetation, species of Rhizophora and Avicennia predominate in mangrove swamp forests in coast lines, particularly in the west coast. In the big fresh-water swamps, typical peat vegetation with limited species is found.
RELIEF
The 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
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. From these mountain systems, flow the major rivers through hilly, rolling and undulating lowlands towards the flood plains and coastal flats. The intermediate low- lands 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.
GREAT SOIL GROUPS
Owen (1951) first classified provisionally soils in West 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 classi- fication, based on more recent field and laboratory data, is adopted for the purpose of this study. The 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. The humus horizon is not very distinct. Macrostructure is weak. Texture is commonly clay although appears more to be a loam by feel of hand. Internal drainage is good and the surface layer tends to dry out readily during dry spells. The iron content of these soils is usually 20-30 per cent Fe,O,.
RED DIS II-B R O W N L A T E RITI CS
Soils in this group are characterized by a B horizon with moderately developed subangular blocky struc- tures 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 8-12 per 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,.
130
Pedogenesis and soil fertility in West Malaysia
TABLE 1. Great soil groups of West Malaysia
Soil group Major parent materids Horizonation Estimated area
Terrain (in thousand ha)
1. Latosol
2. Reddish-Brown Lateritic
3. Red-Yellow Podzolic
4. Yellow-Grey Podzolic
5. Caterite
6. Gley
7. Alluvial Soil
8. Podzol
9. Organic Soil
10. Lithosol
Uasalt, Andesite, Schist
Diorite, Tuffs, Shale
Dominant U horizon
Weak Ae/B t
Granite, rhyolite, sandstone, Distinct Ae/Bt sandy shale, older alluvium
Shales, shales interbedded Distinct Ae/Bt with siltstones coarse structures
Shales, Phyllites B position occupied by iron rich concretions
A-BG or A-G Marine sediments, subrecent alluvium
Recent riverine alluvium AC
Beach sands, granites Bleached A,, humus and/or iron B
Dead swamp vegetation Organic horizon
Granite, quartzite Incipient U /
Roiling to hilly
Undulating to hilly
Undulating to hilly
Undulating to rolling
Undulating to rolling
Flat
Flat
Undulating
Flat
Very steep TOTAL :
80
480
2 340
1160
320
1300
490 '
290
720
5 400 12 580
Y E L L O W - G R E Y PODZOLICS
Textures are usually silty loam to silty clay. The A horizon is thin with moderate subangular blocky aggregates which are ñrm. The subsoil has coarse angular blocky or primatic structures with firm or very firm consistence. An eluvial/alluvial sequence is fairly distinct. Mettling, indicative of impeded drain- age, 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
The presence of a dominating, often compacted hori- zon 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.
GLEYS
The prominence of gleying and associated features of mottling in root channels and structural faces are
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.
ALLUVIAL SOILS
These are juvenile soils with AC profiles. Colours are yellowish to strong brown and textures are variable, from loam to clay. Structures are generally weak and consistence friable. Primary minerals can be com- monly observed in these profiles.
PODZOLS
A bleached A, horizon overlying a humus and/or B horizon is characteristic of these soils. They are largely found on old beach sands on the east coast but also occur at altitudes of above, 1,350 m on acid granites (Panton, 1958). ,
ORGANIC SOILS
These soils have more than 65 per cent organic matter which is greater than 30 cm in depth. Woody mor
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 made on them.
'
NUTRIENT STATUS
Nutrient status is gauged by conventional soil ana- lytical 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.
6
5
PH
4
' 3
6
5
PH
4
3
- . ,I- Topso i l
400
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 small- holdings which were unlikely to have received much fertiliser. Data are derived from published works (Dumanski, 1966; Joseph, 1965; Leamy, 1966; Ng, 1965, 1966; Panton, 1957, 1958; Wong, 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. The numbers of profiles examined for each soil group are as follows : Latosols 14, Reddish-Brown Lateritic 18,
I I I I I I I I I I Su bsoi I
I I I I I
LS R B L RYP YGP L
I 300
200
1 O0
O' I I I I I 1 I I
400
300
I 200
G A P 0
Topsoil
l o o i ! O € i * € I I € - Ls RBL RYP YGP L G A P
FIG. 1. PII (in water).
132
FIG. 2. 0.1N NaOII soluble inorganic P.
Pedogenesis and soil fertility in West Malaysia
-
--T
i I I
the majority being in the range 3.5-5.5. The Organic soils have the lowest values as well as some members 'of the 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 by strong weathering and leaching processes.
2. O.IN N a O H soluble inorganic phosphate (Fig. 2). Only the Latosols have distinctly higher levels of phosphate, their mean levels being greater than the m a x i m u m values found for the other groups. Diffe- rences between the other groups are not clearly evident although the Yellow-Grey Podzolics and Podzols incline to possess the lowest levels. Mean levels for the
Red-Yellow Podzolic 34, Yellow-Grey Podzolic 19, Laterites 10, Gley 18, Podzols 5 and Alluvials 8.
Only topsoils of 10-15 c m in depth and subsoils down to a depth of 30-45 c m are taken in consideration.
DIFFERENTIATION AT GREAT SOIL GROUP LEVEL
AVAILABILITY INDICES
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, are below 80 p.p.m., a groups except the Latosols
1.6
1.4
1.2
1 .o
0.8
0.6
0.4
0.2 cn o 7 > 2 0
T I
- Topsoil - - _ I Subsoil -
% 0.5
0.4
0.3
0.2
0.1
Topsoil
Subsoil
-
T I I I I I 4- I I I I
T I I I I I I I I I I I I -i- I I I I
T I I T T
I T I I
I I +
t 1 I r
t I I I I 1 I I
Ls R B L RYP YGP L G A P LS RBL RYP YGP L G A P
FIG. 3. Exchangeable K. FIG. 4. Nitrogen contents.
133
Ng Siew Kee and Law Wei Min
TABLE 2. Exchangeable Me values in soil groups (m.e./100 g)
Soil group
Topsoil Subsoil
Range Mean Range Mean
Latosol Reddish-Brown Lateritic Red-Yellow Podzolic Yellow-Grey Podzolic Laterite Gley Alluvial Podzol
0.10- 0.93 0.61 0.10- 0.59 0.21 0.21- 0.82 0.55 0.10- 0.30 0.17 0.16- 0.59 0.34 0.05- 0.42 0.17 0.10- 0.73 0.42 0.05- 0.58 0.21 0.26- 0.78 0.47 0.05- 0.58 0.22 0.37-12.6 3.17 0.33-10.0 3.11 0.74- 3.94 2.40 0.16- 4.26 1.61 0.05- 0.74 0.32 0.05- 0.32 0.15
value below which response to phosphate application is likely (Saunder, 1956).
Exchangeable potassium (Fig. 3)
The 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 mean values are less than 0.2 m.e./100 g soil. Thus about three classes can be differentiated well and a fourth marginally.
Exchangeable magnesium (Table 2)
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 groups are not distinct and values are generally less than 0.6 m.e./100 g soil, indicating very low availability.
TABLE 3. Exchangeable Ca values in soil groups (m.~./lOO g)
Soil group Topsoil Subsoil
Range Mean Range Mean
Latosol Reddish-Brown Lateritic Red-Yellow Podzolic Yellow-Grey Podzolic Laterite Gley Alluvial Podzol
-- 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
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
- 0.11 0.13 0.09 0.12 0.14 1.25 0.87 0.09
Exchangeable calcium (Table 3)
The 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, mean values are very low, being less than 0.5 m.e./100 g soil.
The 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.
‘RESERVE’ INDICES
Nitrogen
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 ammonium and nitrate forms. However, the organic forms differ in their susceptibility to bac- terial 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 sub- soils, Alluvials are also low. The over-all pattern ap- pears to indicate a higher nitrogen status in the Gleys and Latosols. than the rest.
6N HC1 extractable nutrients 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.
TABLE 4. 6N HCl extractable K in soil groups (m.e./lOü g) :.
Topsoil Subsoil
Soil group Range Mean Range Mean
Latosol 0.37- 1.97 0.93 0.32- 1.69 0.84 Reddish-Brown Lateritic 0.57- 9.83 2.83 0.81-10.8 3.28 Red-Yellow Podzolic 0.32- 4.73 1.40 0.30- 5.93 1.67 Yellow-Grey Podzolic 0.96-14.1 6.89 2.52-18.1 9.04
I ’Laterite 1.30- 3.74 2.52 1.09- 5.04 2.61 Gleys 0.80-12.5 3.57 0.64-13.6 5.22 Alluvial 1.02- 6.19 3.11 1.46-10.0 4.87
134
Pedogenesis and soil fertility in West Malaysia
PPm 1,500
1,250
1,000
750
500
2 50
O
- Topsoil ____ Subsoil l- I I I I I I I I I I I I I I t I I I I I I I I
T
T I I I I I I + I [i I
I I l
Ls R B L RYP Y G P L G A
FIG. 5. 6 N HCI soluble P.
TABLE 5. 6N €IC1 extractable M g in soil groups (m.e./100 g)
Topsoil Subsoil Soil group
Range Mean Range Mean
Latosol 1.13- 6.91 Reddish-Brown Lateritic 1.08- 9.48 Red Yellow Podzolic 2.48- 4.78 Yellow-Grey Podzolic 2.36- 6.51 Laterite 1.60- 6.08 Gley 2.38-12.4 Alluvial 3.84-1 1.9
3.68 4.37 3.94 3.60 3.70 5.89 7.31
1.60- 5.28 3.08 1.28-15.3 5.28 0.80- 6.10 3.30 2.84- 5.95 3.82 1.28- 8.64 4.52 2.67-16.7 6.51 3.86-10.9 7.29
Potassium (Table 4). The ranges show that the Lato- sols are poorest in potassium reserves while the Yellow- Grey Podzolics and Gleys tend to be richest. According to mean values, the Latosols and Red-Yellow Podzolics have less than 2 m.e./100 g in both horizons, the Red- dish-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.
Magnesium (Table 5). By range and mean values, only the Gleys and Alluvial can be differentiated to have higher magnesium contents.
\
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 mean values, the number of distinct classes which each soil index can achieve is shown in Table 6. It is evident that the nutrient in- dices make considerably less differentiation than pro- file morphology, the relative efficiency being about 30 per cent. This is an indication that nutrient differ- ences at the start of pedogenesis have been drastically reduced by the intense weathering and leaching pro- cesses 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 members of most soil groups except the Gleys (Ng, 1965).
DIFFERENTIATION BELOW GREAT SOIL GROUP LEVEL
A great soil group is rather heterogeneous in mor- phological terms and it is possible that generalizations can mask 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
TABLE 6. Number of nutrient classes separable by soil indices
No. of separable classes Average Number of Soil index
sOi'.~Oups Topsoil Subsoil
PII NaOH-P Exch. Ei Exch. M g Exch. Ca N 6N IICl P
' 6 N HCl K 6N HCl M g
2 2 3 2 2 2 3 3 2
2 2 3 2 ' 2 2 3 3 2
135
Ng Siew Kee and Law Wei Min
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 on 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 indices.
CROP RESPONSE TO MANURING
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 down 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 two major crops in the country, rubber and oil palm are pre- sented in Tables 9 and 10. S o m e of the data are for cultivated soils but they should be fairly pertinent to
TABLE 7.cNaOH-P, exchangeable K and M g of main soil families
Topsoil S&soil
P (p.p.m.) K (me. %) Mg (me. yo) P K Milg Family
Latosol Kuantan Segamat
Munchong Kampong Kolam
Rengam Serdang IIarimau
109-406 65-205
0.06-0.16 0.06-0.60
0.31-0.74 0.63-0.93
0.21-0.82 0.26-0.83
80-389 49-174
12-80 29-80
0.03-0.06 0.02-0.34
0.13-0.20 0.10-0.59
Reddish-Brown Lateritic
26-95 ' 36-95
0.14-0.40 0.05-0.40
0.03-0.49 0.02-0.13
0.10-0.30 0.05-0.26
Red-Yellow Podzolic
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 Anam
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 0.23-1.57 , 1.27-12.6 32-135 0.34-0.97 0.42-10.0 27-42 0.12-0.59 0.41-1.08 11-26 0.05-0.15 0.33-0.67
TABLE 8. N, 6 N IICl soluble P, K and M g in major soil families
Topsoil Subsoil Family
N(%) Pb.p.m.) K (me.%) Mg(m.e.%) N P K Mg soil group
Latosol Kuantan 0.15-0.36 400-885 0.37-0.64 1.68-3.08 0.04-0.22 400-750 0.32-0.64 1.60-2.00 Segamat 0.10-0.38 220-1320 0.57-1.97 4.02-6.91 0.10-0.17 214-1110 0.57-1.69 2.41-5.28
Reddish-Brown Munchong 0.07-0.41 90-260 1.04-9.83 1.08-9.48 0.05-0.14 26-140 0.81-10.8 1.60-8.50 Lateritic Kg. Kolam 0.10-0.41 41-224 0.57-3.13 1.61-8.39 0.04-0.14 31-231 1.09-6.39 1.28-15.37
Red-Yellow Rengam 0.06-0.18 161-190 0.64-4.73 3.54-4.60 0.02-0.11 113-142 0.32-5.93 0.80-3.64
Harimau 0.11-0.15 52-70 0.50-0.64 2.48-4.56 0.05-0.06 60-94 0.30-0.64 1.92-4.56 Podzolic Serdang 0.09-0.25 60-148 0.32-2.96 3.21-4.66 0.03-0.10 46-131 0.32-3.74 2.88-4.68
Yellow-Grey Durian 0.06-0.31 46-187 2.37-13.4 3.04-6.51 0.06-0.12 40-160 2.52-18.1 2.84-5.58 Podzolic Batu Anam 0.06-0.16 52-120 0.96-14.1 2.36-4.96 0.03-0.08 28-120 2.97-16.4 2.34-5.95
Gley Selangor 0.15-0.48 209-580 6.50:12.5 2.38-10.6 0.07-0.31 168-500 6.36-13.6 2.67-17.8 Manik 0.09-0.15 78-157 0.80-3.19 n.d. 0.03-0.05 90-142 0.97-2.90 n.d.
136
Pedogenesis and soil fertility in W e s t Malaysia
TABLE 9. Rubber response to fertilizers
Soil scries Soil group Response to Percentage response source
Serdang (ex-jungle)
Rengam (ex-jungle)
Rengam (cultivated)
Kuantan (ex-jungle)
Munchong (cultivated)
Malacca (cultivated)
Selangor (cultivated)
Sitiawan/Sogomana (cultivated)
RYP
RYP
RYP
La
RllL
L
G.
G
NPK NP PK
N
P
K
K Nil to P, M g
% NP
NK NPK
N
N
22 yo in yield 17 yo in yield 13 yo in yield 8 yo in growth 1.5 yo in yield 14% in growth 3 yo in yield 22 yo in yield
9 yo in immature growth
7 Yo in yield 9% in growth 3 yo in yield 18 yo in yield 12 yo in growth 5-10 yo in yield
6 yo in growth
Guha and Pushparajah (19660)
Guha and Pushparajah ,
(19660)
Guha and Pushparajah (1966~)
Guha and Pushparajah (1966a)
Guha and Pushparajah (19660)
Guha and Pushparajah (1966b)
Guha and Pushparajah (1966a)
Guha and Pushparajah (1966b)
TABLE 10. Oil palm response to fertilizers
Soil series Soil group Response to Percentage response Source
Rengam (ex-jungle) . RYP N 8 yo in yield Rosenquist P 5 yo in yield (1964) K 33 yo in yield w? 2 yo in yield
Durian (ex-jungle) YGP N 12 yo in yield Rosenquist P 22 yo in yield (1964) K 2 yo in yield
' Rengam (ex-jungle) RYP NPK M g 25 yo in yield Rosenquist (1964)
Batu Anam (ex-jungle)
Selangor (replant)
YGP N P K w 3
G N K
10 yo in yield 15 yo in yield (1969) 22 yo in yield
Martineau et al.
6 yo in yield 8 yo in yield
Hew (personal com- munication, 1969)
137
Ng Siew Kee and Law 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, parti- cularly 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 seems to accord satisfactorily with the soil potas- sium 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. The 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 Red- Yellow Podzolics, Reddish-Brown Lateritic and Yellow- Grey Podzolics but not in the Gleys and Latosols.
The requirements for three or four of the macro- nutrients 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 West Malaysian soils are impov- erished of nutrients and therefore their nutritional differences are fewer than their morphological varia- tions. Thus, the strong weathering processes and in- tense leaching have polarised the fertility status of West Malaysian soils, with the disproportionate majo- rity in the low end, producing a skewed type of distri- IJution.
FUT U RE DEVE L O PM ENT AND PHYSICAL FACTORS
T w o major conclusions can be drawn from the pre- ceding analysis.
First, the successful development of the largc reserves of virgin soils in West Malaysia with agriculture de- pends 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 Eden (1964), ‘as true of temperate soils, the fertility of tropical soils also must be man-made or controlled’, and this is at least per- tinent 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 pedo- genetic classification and inherent fertility in West 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 by 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 by 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 ep- hance the agronomie value of soil classification. Unfortii- nately, such data are extremely limited at present and greater emphasis should be accorded in future to this area of soil research in this region.
AC I< NO W L E D GERI E 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 when one of us (N.S.K.) was also a member 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. .
138
Pedogenesis and soil fertility in West 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 rela- tionships between the genetic group as a soil and its agri- cultural chemical properties. 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 Con., Kuching, Sarawak,
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 Co- op., 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 reconnais- sance soil map 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 map of Thailand. Proc. 3rd Malaysian Soil Con., Kuching, Sarawak, p. 225-8.
p. 45-50.
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 map of Malaya. J. trop. Geogr.,
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.
SOIL SURVEY 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 map of Sabah. Proc. 2nd Malaysian Soil Con., 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 map 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.)
VOI. 18, p. 119-24.
'
139
III Final report of the symposium
Final report of the symposium
1. During the inaugural session, addresses were deli- vered 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. At the opening session, the Symposium nominated Dr. D. Muljadi (Indonesia), Dr. N g Siew Kee (Malaysia), RIr. 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. The first working session was devoted to the pre- sentation of the key paper on ‘Formation and trans- formation of clay minerals in tropical soils’, by Pro- fessor 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 pro- ceeds 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 tem- perate climate. In general, the formation and transformation of
primary minerals to secondary minerals are through hydration and hydrolysis of the primary minerals. The monolayer water in the capillary pores in the soil is dissociated and acts as au acid forming the hydro- nium H,O+ ion. As far as the formation of polywater is concerned, it is possible that cleanliness of the capil- lary 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-phase substitution are few and only possible in the formation of illite from montmorillonite and vermiculite.
As the soil is an open system in which leaching and movement of soil solution would he continuous, an equilibrium between the ions in soil solution and the primary minerals would not be achieved. Conse- quently there is no relationship in the chemical com- positions of the primary parent material and that of the secondary minerals or that of the soil solution. Simi- larly, in well-developed soils, there is no direct corre- lation 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 common. In the transformation processes, organic matter m a y operate as a pump, accumulating or rejecting elements such as Fe2+, AP+, PO:-, etc., at the right (or the wrong) moment. It has never been proved that it has a more direct role.
The formation of polynuclear complexes of aluminium is pH dependent and the change of aluminium from fourfold to sixfold co-ordination occurs when 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 kao- linite 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. The transformation of kaolinite to gibbsite has been
observed to be enhanced by presence of organic ions.
143
Final report of the symposium
The organic anions comhine with gibbsite (AL(OH),) to form complexes and kaolinite will further be decomposed and removetl according to the following schcnie :
Kaolinite + II,O 3 Al(OII), + II,SiO, IIn acid + 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. Some remarks were made about this paper. The method of Jackson, used by 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 attri-
buted 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, identifica- tion, evolution’.
8. The discussion emphasized many points : Because ilmenite is a very resistant mineral, it
cannot be considered as a good source of iron in the weathering process. The presence of ilmenite which has been transported in beaches (for example, Madagascar), confirms the above statement.
Calcium carbonate and laterites do not occur to- gether. 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 way 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 crys- talline 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. The 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 by Taylor, of CSIRO (in press, Clay Minerals) and by 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 Fe- glycerol complex formed will decompose in boiling water to form magnetite and maghemite, which can LC separated quantitatively by using a magnet.
9. The third key paper : ‘Weathering and soil forming processes in the tropics’, was presented by its author, Dr. J. van Schuylenborgh.
10. The discussion emphasized many points. The author, in his thermodynamic calculations, con-
sidered silica as orthosilicic acid; however, in. the case of polymerized silica, the equilibrium equation m a y not
N e w cemented laterite
000 O00
144
Final report of the symposium
be the same. The 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. The real process depends on the activation free energy which is, in no way, a direct function of AF. Soils are open systems. The equilibrium condition has therefore to be changed and, instead of stating it as TAS = AQ, it needs to be stated as TAS = AQrev. + AQirrev..
The removal of Fe in the profiles in tropics is com- 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,O system, a good relationship has been found between the theory and the observed mineralogical data in a few cases.
The plinthite is part of the oxic horizon, in which Fe, as oxides and hydroxides, is deposited and silica is leached out. Dehydrated Fe 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 on the nature of parent rock on soil formation under similar atmospheric cli- mates (Rh. J. P. Andriesse) that soil formation, on parent rocks with varying contents of silica attains dif- ferent aspects. Neither climate nor the altitude is res- ponsible 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 lateri- tization m a y occur at high altitude if the parent mat- erial is basic.
The 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. Mineraliza- tion in the humid temperate low land is generally rapid but, with acid parent material, is retained.
13. Dr. V. M. Fridlaud presented a paper on the dif- ferences between crusts of weathering and soils devel- oping on 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 known that in identical topo- graphical conditions, mottles are developed in poorly drained soil.
The 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 ele- ments with big inner porosity whereas acid rocks dev-
elop into angular ones with thin inner porosity.Therefore, these structural irregularities affect the different water régimes of the weathering crusts and soils formed on them.
The 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. There- fore, they give diKerent forms of porosity.
The general pore volume only of the weathering crust does not conform to the drainage and gleization characteristics. The 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 Suy 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. The comment on the paper: ‘Some 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 incom- plete chemical weathering is given, an analysis of organic carbon should also be made.
17. In the absence of Dr. Donald Sherman, Dr. Ng Siew Kee presented the key paper on ‘Mineral weather- ing in relation to utilization of soils’ and Dr. J. J. Fri- piat commented on it.
18. This presentation was followed by a paper and a discussion on ‘Pedogenesis and soil fertility in West Malaysia’ (Dr. Ng Siew Kee and Mr. L a w W e i Afin).
19. The Red-Yellow Podzolics as recognized in Malay- sia were so classified according to the definitions by Thorpe and Smith (1949). This definition is not based on the occurrence of an 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 classifi- cation. It has been observed that, although an 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 many of the soils, even recent alluvial soils, had dev- eloped 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 made in the 7th Approximation, it was suggested that if the oxic nature of the soils is more pronounced than horizona- tion, 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
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 uncommon 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).
Paddy soils in Malaysia are presently grouped with the gley soils but separate classification m a y be neces- sary in future.
14 6
List of participants
Belgium
Cambodia
France
India
Indonesia
Professor J. J. Fripiat
&Ir. Tan Boun Suy
Dr. P. Segalen
Dr. T. Scshagiri nao
Dr. H. Ling Ong
Dr. D. Muljadi
Mr. Soepraptohardjo
Professor Ur. G o Ban Hong
Dr. Achmad II. Satari
Ir. T, Sukarna
Laboratoire de Physico-Chimie Minérale, Institut des Sciences de la Terre, 42 Decroylaan, Heverlee-Louvain.
Doyen de la Faculté des Cultures Tropicales, Université Royale de Kompong-Cham.
Services Scientifiques Centraux de l'ORSTOM, 70 Route d'Aulnay, 93 Bondy.
Chief Scientific Officer, Regional Research Station, University of Agricultural Sciences, Raichur, Mysore State.
Lecturer in Geochemistry and Geology, Department of Geology, Institut Teknologi Bandung, Djl. Ganeca 10, Bandung.
Director, Soil Research Institute, Djl. Ir. II. Djuanda 98, Bogor.
Deputy Director, Soil Research Institute, Djl. Ir. H. Djuanda 98, Bogor.
Director, Central Research Institute for Agriculture, Djl. Merdeka 99, Bogor.
Associate Professor in Soil Sciences, Soil Science Department, Bogor Agricultural University (IPB), Bogor.
Soil Scientist, Faculty of Agriculture, University of Padjadjaran, Bandung.
.
147
List of participants
hlalaysia
Netherlands
Pakistan
Philippines
. Thailand
Dr. Ng Siew Kee
Mr. J. P. Andriesse
Mr. H e w Choy Kean
Dr. J. van Schuylenhorgh
Dr. Iajuddiii Alimed
Dr. Domingo Salita
Mi. Vichai Boonyawat
United States of America Professor G. D. Sherman
U.S.S.R.
Viet-Nam
Dr. V. Bl. Fridland
Professor Le Khac Pho
INSTITUTIONS AND ORGANIZATIONS
Indonesian Instiiute of Sciences (LIPI)
Professor Sarwono Prawirohardjo
Professor J. A. Katili
Miss Sjamsiah Ahmad
Head, Analytical Chemistry Division, Rubber Research Institute of Malaysia, P.O. Box 150, Kuala Lumpur.
Senior Officer, Soil Survey Division Department of Agriculture, Kuching, Sarawak.
Research Officer, Oil Palm Research Station, Harrisons and Crosfield, P.O. Box 207, Banting, Selangor.
Laboratory of Regional Soil Science, Agricultural University, Wageningen.
Head, Department of Soil Science, Dacca University, Dacca.
Professor of Geology and Geography, University of the Philippines, Quezon City.
Soil Survey Division, Department of Land Development, Ministry of National Development, Bangkok.
Senior Professor of Soil Science, College of Tropical Agriculture, University of Ilawaii, IIonolulu, Hawaii.
Senior Scientist and Professor of Moscow University, Institute of Geography, Academy of Sciences of U.S.S.R;
Professor of Geography at the Faculty of Education and Faculty of Letters, IIue University 2, Le loi, Hue.
Chairman, Indonesian Institute of Sciences, Djl. Teuku Tjhik Ditiro 43, Djakarta (Indonesia).
Deputy Chairman for Natural Sciences, Indonesian Institute of Sciences, Djl. Teuku Tjhik Ditiro 43, Djakarta (Indonesia).
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
BIO TROl’/SEAIIlEC
Dr. W. Soegeng Reksodihardjo
Food and Agriculture Organization of the United Nations
Dr. V. S. Subramanian
Calten I’aciJic Indonesia
Dr. T. C. J. Zwartkruis
Unesco
Dr. F. Fournier
Mr. F. J. C. Pala
BIr. Pita Niramaya
Rector, Institute of Technology Bandung, Djl. Kapt. Pattimura 64, Bandung (Indonesia).
Deputy Director of BIOTROP, Bogor (Indonesia).
FAO Expert, Bogor (Indonesia).
Head, Regional Geologist,
Kebon Sirih 52, Djakarta (Indonesia). c/o CPI
Consultant, Natural Resources Research Division, Environmental Sciences and Natural Resources Research Unesco, Place de Fontenoy, 75 Paris 7e (France).
Deputy Director, Unesco Field Science Office for South-East Asia, Djl. Imam Bondjol 30, P.O. Box 273/DKT, Djakarta (Indonesia).
Senior Administrative Secretary, Unesco Field Science Office for South-East Asia, Djl. Imam Bondjol 30, Djakarta (Indonesia).
149
![Weathering of Petroleum Biomarkers: Review in Tropical ......Weathering, is a natural defense mechanism in ecosystems [5] [6], and it acts like the main remediation process for areas](https://static.documents.pub/doc/80x56/6085e428a47e3f5d3e52106b/weathering-of-petroleum-biomarkers-review-in-tropical-weathering-is-a.jpg)
