Micronutrients and the Nutrient Status of Soils: a Global Study

Home , Micronutrient, Micronutrient deficiency

FAO SOILS BULLETIN 48

micron utrients and the nutrient status of SOUS:

a global study

by mikko sillanpAl

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS FAO SOILS BULLETIN 48

micronutrients and the nutrient status of soils:

a global study

by mikko sillanpäd

sponsored by the government of finland executed at the institute of soil science agricultural research centre jokioinen, finland and soil resources, management and conservation service land and water development division FAO

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIO-NS Rome 1982 The designations employed and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization oftheUnitedNations concerningthelegal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.

M-52 ISBN 92-5-101193-1

Allrights reserved. No part ofthispublication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic,mechanical, photocopyingor otherwise, without theprior permission of the copyright owner. Applications for such permission, with a statement of the purpose and extent of the reproduction, should be addressed to the Director, Publications Division, Food and Agriculture Organization of the United Nations, Via delle Terme diCaracalla, 00100 Rome, Italy.

C) FAO 1982

Printed in Finland by Werner Söderström Osakeyhtiö. Foreword

During the last two decades, the increasing use of mineral fertilizers and organic manures of different types has led to impressive yield incrcases in developing countries. Major emphasis was given to the supply of the main macronutrients, nitrogen, phosphate and potash. For a long time it was felt that under the existing farming systems and fertilizing practices the level of micronutrients was adequate and that the problem of micronutrient deficiencies was not a serious one. However, indications from developing countries show that micronutrient problems are becoming more and more frequent. In the early 1970s the Government of Finland, through its Institute of Soil Science and FAO embarked on a system of investigation of microelement deficiencies in developing countries. In 1974, the project -Trace Element Study, TF/129/FIN" started under the FAO/ Finland Cooperative Programme, with the financial support of the Government of Finland, and with the cooperation of the staff of the Institute of Soil Science, under the direction of Prof. M. Sillanpää. The facilitiesof theInstitute were made available to the project. This programme involved a worldwide study on micronutrients, in cooperation with 30 countries. The study, which is presented in this document, shows that the undertaking was very timely. It appears that micronutrients are becoming deficient in developing countries and that remedial measuring will need to be taken to ensure full soil and plant productivity. The study also gives guidance for the investigation of micronutrient deficiencies, which should be dealt with at the country level. FAO expresses its appreciation to the Government of Finland for its generous support of the study and for its excellent cooperation throughout.

F.W. Hauck Chief, Soil Resources, Management and Conservation Service V CONTENTS

page PREFACE

PART I 1

PART I SOIL AND PLANT DATA, IvIETHODOLOGY AND INTERPRETATION 3

1 MATERIALS AND METHODS 3

1.1 Collection of original plant and soil samples 3 1.2 Indicator plants 3 1.2.1 General 3 1.2.2 Sampling time 4 1.2.3 Contamination 4 1.2.4 Plant varieties 6 1.2.5 Need for better grounds for comparing plant data 6 1.2.6 Growing new indicator plants in pots 7 1.3 Analytical methods 8 1.3.1 Methods of soil analyses 9 1.3.2 Methods of plant analyses 11 1.3.3 Statistical methods 12 1.4 Expression of analytical data 13 1.4.1 Soils 13 1.4.2 Plants 14 1.4.3 Regression graphs 14

2 RESULTS AND DISCUSSION 17

2.1 General properties of soils and their mutual relations 17 2.2 Macronutrients 21 2.2.1 General aspects 21 2.2.2 Comparison of the two original indicator plants 21 2.2.3 Comparison of macronutrient status in different countries 21 2.2.4 Macronutrient contents of plants and soils in relation to four soil characteristics 28 2.3 Micronutrients 33 2.3.1 Plant analysis versus soil analysis 33 2.3.2 Molybdenum 35 2.3.2.1 General aspects 35 2.3.2.2 Soil factors affecting the molybdenum contents of plants and soils 36 2.3.3 Boron 46 2.13.1 General aspects 46 2.3.3.2 Soil factors affecting the boron contents of plants and soils 48 2.3.4 Copper 54 2.3.4.1 General aspects 54 2.3.4.2 Soils factors affecting the copper contents of plants and soils 55 2.3.5 Iron 61 2.3.5.1 General aspects 61 2.3.5.2 Soil factors affecting the iron contents of plants and soils 61 2.3.6 Manganese 64 2.3.6.1 General aspects 64 2.3.6.2 Soil factors affecting the manganese contents of plants and soils 66 2.3.7 Zinc 75 2.3.7.1 General aspects 75 2.3.7.2 Soil factors affecting the zinc contents of plants and soils 75 VI

page 2.4 Mutual relations between micronutrients 83 2.5 Micronutrient contents of plants and soils in relation to yields with special reference to the "concentratiim-dilution" phenomenon 86 2.6 Suitability of plant and soil analyses for large scale operations 90 2.7 Estimation of critical micronutrient Hants 90 2.8 Plant and soil micronutrieents and other soil data in relation to FAO/Unesco soil units 96

3 SUMMARY FOR PART I 97

PART II 101

PART II NUTRIENT STATUS BY COUNTRIES 103

4 INTRODUCTION 103

5 EUROPE AND OCEANIA 105 5.1 Belgium 105 5.2 Finland 117 5.3 Hungary 128 5.4 Italy 139 5.5 Malta 150 5.6 New Zealand 158

6 LATIN AMERICA 169 6.1 Argentina 169 6.2 Brazil 180 6.3 Ecuador 190 6.4 Mexico 194 6.5 Peru 207

7 FAR EAST 219 7.1 India 219 7.2 Korea, Republic of 232 7.3 Nepal 242 7.4 Pakistan 251 7.5 Philippines 262 7.6 Sri Lanka 271 7.7 Thailand 280

8 NEAR EAST 289 8.1 Egypt 289 8.2 Iraq 300 8.3 Lebanon 313 8.4 Syria 322 8.5 Turkey 332

9 AFRICA 343 9.1 Ethiopia 343 9.2 Ghana 355 9.3 Malawi 363 9.4 Nigeria 372 9,5 Sierra Leone 383 9.6 Tanzania 392 9.7 Zambia 401 VII page 10 SUMMARY FOR PART II 411

REFERENCES 417

APPENDIXES

1Instructions for taking plant and soil samples 421 2 National mean values of basic analytica/ data on origina/ wheat plants and respective soils 422 3 National mean values of basic analytical data on original maize plants and respective soils 428 4 National mean v-alues of basic analytical data on pot-grown wheat plants and respective soils 434 5 Indicative data on fertilizer application to the original indicator crops 440 6 Average soil properties and macronutrient contents of soils classified by FAO/Unesco soil units 441 7 Average micronutrient contents of soils and respective pot-grown wheats for FAO/Unesco soil units 442

LIST OF TABLES

Page Table 1. Comparison of nutrient content of two soils expressed on weight and volume bases. 13

Table 2. Numerical products of plant content x soil content along the lines dividing the regression graphs into five zones. 15

Table 3. Mean macronutrient contents of the two original indicator plants, of the respective soils and the mean amounts of N, P, and K applied in fertilizers to each crop. 21

Table 4. Regressions of pooled plant N, P, K, Ca, and Mg contents on respective soil macronutrients. 28

Table 5. Plant and soil macronutrients as functions of four soil characteristics in the whole international material. 28

Table 6. Directions (+ or symbols) of regressions of plant and soil macronutrient contents on four soil characteristics. 32

Table 7. Equations and correlation coefficients for regressions of A0-0A extractable soil Mo (uncorrected and pH-corrected) and plant Mo on six soil characteristics. 39

Table 8. Correlations between plant Mo and A0-0A extractable soil Mo in various countries. Soil Mo analyses are both uncorrected and corrected for pH and texture. 45

Table 9. Correlations between plant B and hot water extractable soil B in various countries. Soil B analyses are both uncorrected and corrected for cation exchange capacity. 51

Table 10.Equations and correlation coefficients for regressions of hot water soluble soil B (uncorrected and CEC-corrected) and plant B on various soil factors. 53

Table 11 Correlations between plant Cu and AAAc-EDTA extractable soil Cu in various countries. Soil Cu analyses are both uncorrected and corrected for soil organic carbon contents. 58 VIII

page Table 12.. Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Cu (uncorrected and org. C-corrected) and plant Cu on various soil factors. 60

Table 13.Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Fe and plant Fe on various soil factors. 64

Table 14.Correlations between plant Mn and DTPA extractable soil Mn in various countries. Soil Mn analyses are both uncorrected and corrected for pH(CaC12). 69 Table 15.Equations and correlation coefficients for regressions of DTPA extractable soil Mn (uncorrected and pH-corrected) and plant Mn on various soil factors. 71 Table 16.Correlations between plant Mn content and AAAc-EDTA extractable soil Mn in various countries. Soil Mn analyses are both uncorrected and corrected for pH. 73 Table 17.Mutual correlations between extractable Mn contents of soils in the whole material as determined by four methods. 74 Table 18.Correlations in various countries between plant Zn and soil Zn determined by three different methods. 79 Table 19.Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Zn (uncorrected and pH-corrected), plant Zn, and DTPA extractable soil Zn on various soil factors. 81

Table 20.Mutual correlations between extractable soil micronutrients in the whole material. 86

»Table 21.Two-year averages of micronutrient contents of 17 crops grown side by side at nine sites. 91

Table 22.Tentative critical levels of micronutrients determined by various methods. 96

Table 23.Correlations between micronutrient contents of the three types of indicator plants and respective soils. 97

Table 24.Comparison of micronutrient contents between high yielding maize varieties and local variety 'Criollo', Mexico. 206

Table 25.Comparison of macronutrient contents of high yielding and local varieties of original Indian wheats. 231

Table 26.Comparison of micronutrient contents of high yielding and local varieties of original Indian wheats. 231

Table 27.Comparison of analytical data from non-irrigated and irrigated soils of Iraq and respective plants. 311

Table 28.Comparison of macronutrient contents of high yielding and local varieties of original Nigerian wheats. 376

Table 29.Relative frequencies of low, medium and high values of soil texture (TI), pH, organic carbon content and cation exchange capacity in 30 countries. 412

Table 30.Relative frequencies of low, medium and high values of nitrogen, phosphorus potassium, calcium and magnesium in soils of 30 countries. 413

Table 31.Relative distributions of six micronutrients in five plant x soil content zones in various countries. 414 IX LIST OF FIGURES

page Fig. 1. Variations in B, Fe, Cu, Mn, Mo, and Zn contents of wheat (ppin in DM) during the growing season. 5

Fig. 2. Growing new indicator plants in pots (photo). 8

Figs 3 and 4. Mutual correlations between six soil characteristics. 18, 19

Fig. 5. Relationship between pH(H20) and pH(CaC12). 20

Fig. 6. Regressions of nitrogen content of original wheat and maize (pooled) on total nitrogen content of soil a) for the whole international material, b) for crops fertilized with > 100 kg N/ha, and c) for crops fertilized with < 50 kg N/ha. The national mean vaities are given. 23

Fig. 7. Regressions of phosphorus content of original wheat and maize (pooled) on NaHCO3 extractable phosphorus content of soil a) for the whole international material, b) for crops fertilized with > 30 kg P/ha, and c) for crops fertilized with < 10 kg P/ha. The national mean values are given. 24

Fig. 8. Regressions of potassium content of original wheat and maize (pooled) on CH3COONH4 exchangeable potassium content of soil a) for the whole international material, b) for crops fertilized with > 30 kg K/ha, and c) for crops fertilized with < 10 kg K/ha. The national mean values are given. 25

Fig. 9. Regression of calcium content of original wheat and maize (pooled) on CH3COONH4 exchangeable calcium content of soil for the whole international material. The national mean values are given. 26

Fig. 10.Regression of tnagnesium content of original wheat and maize (pooled) on CH3COONH4 exchangeable magnesium content of soil for the whole international material. The national mean values are given. 27

Fig. 11.Plant and soil macronutrients as functions of soil pH in the whole international material. 29

Fig. 12.Plant and soil macronutrients as functions of soil texture in the whole international material. 30

Fig. 13.Plant and soil macronutrients as functions of soil organic carbon content in the whole international material. 31

Fig. 14.Plant and soil macronutrients as functions of cation exchange capacity in the whole international material. 32

Fig. 15.Regression of Mo content of pot-grown wheat on ammonium oxalate-oxalic acid extractable soil Mo for the whole international material. The national mean values are given. 37

Fig. 16.Relationships of A0-0A extractable soil Mo (uncorrected and pH-corrected) and plant Mo to various soil characteristics in the whole international material. 38

Fig. 17.Relationships of A0-0A extractable soil Mo (uncorrected and pH-corrected) and plant Mo to soil pH. 40

Fig. 18.Relationship of plant Mo to A0-0A extractable soil Mo in the whole material classified for pH. 42 X

page Fig. 19.Relationship of plant Mo to pH-corrected soil Mo (AO-0A) in the whole material classified for pH. 42 Fig. 20.Regression of Mo content of the pot-grown wheat on the pH-corrected AO-0A extractable soil Mo for the whole international material. National mean values are given. 43 Fig. 21.Regression of Mo content of the pot-grown wheat on the pH- and texture-corrected soil Mo for the whole international material. National mean values are given. 45 Fig. 22.Regression of B content of pot-grown wheat on hot water extractable soil B for all the international material. National mean values are given. 47 Fig. 23.Regression of B uptake by pot-grown wheat on hot water extractable soil B for all the international material. The national mean values are given. 48 Fig. 24.Relationships of plant B and soil B (uncorrected and CEC-corrected) to cation exchange capacity. 49 Fig. 25.Regression of B content of the pot-grown wheat on CEC-corrected hot water soluble B for the whole international material. National mean values of plant and soil B are plotted in the graph. 50

Fig. 26.Relationships of soil B and plant B to various soil factors. 52 Fig. 27.Regression of Cu content of pot-grown wheat on acid ammonium acetate-EDTA extractable soil Cu for the whole international material. National mean values are given. 55 Fig. 28.Relationships of soil Cu (uncorrected and org.C-corrected) and plant Cu to organic carbon content of soil. 56 Fig. 29.Regression of Cu content of pot-grown wheat on k(org. C)-corrected AAAc-EDTA extractable soil Cu for the whole international material. National mean values are plotted in the graph. 57 Fig. 30.Relationships of AAAc-EDTA extractable soil Cu (uncorrected and org- C-corrected) and plant Cu to various soil factors. 59 Fig. 31.Regression of Fe content of pot-grown wheat on acid ammonium acetate-EDTA extractable soil Fe for all the international material. National mean values are given. 62

Fig. 32.AAAc-EDTA extractable soil Fe and plant Fe as a function of six soil characteristics. 63

Fig. 33.Regression of Mn content of pot-grown wheat on DTPA extractable soil Mn for the whole international material. National mean values are given. 65

Fig. 34.Relationships of DTPA extractable soil Mn (uncorrected and pH-corrected) and plant Mn to soil pI-1. 67 Fig. 35.Relationships of Mn content of pot-grown wheat on DTPA extractable soil Mn in the whole material classified according to pH. 68 Fig. 36.Relationships of plant Mn to pH-corrected soil Mn (DTPA) on the whole material classified according to pH. 68

Fig. 37.Regression of Mn content of pot-grown wheat on pH-corrected DTPA extractable soil Mn for the whole international material. National mean values of plant and soil Mn are plotted in the graph. 69

Fig. 38.Relationships of DTPA extractable soil Mn (uncorrected and p1-corrected) and plant Mn to various soil factors. 70 XI

page Fig. 39.Relationships of AAAc-EDTA extractable soil Mn (uncorrected and pH-corrected) and plant Mn to soil pH. 72

Fig. 40.Regression of Zn content of pot-grown wheat on acid ammonium acetate-EDTA extractable soil Zn for the whole international material. National mean values of plant and soil Zn are also given. 76

Fig. 41.Regression of Zn content of the pot-grown wheat on DTPA extractable soil Zn for the whole international material. The national mean values are given. 77

Fig. 42.Relationships of extractable soil Zn (AAAc-EDTA, AAAc-EDTA + pH-correction, DTPA) and plant Zn to soil pH. 78

Eig, 43.Regression of Zn content of the pot-grown wheat on pH-corrected AAAc-EDTA extractable soil Zn for the whole material. The national mean values are given. 79

Fig. 44.Relationships of plant Zn and soil Zn (determined by three methods) to various soil factors. 80

Figs 45Mutual correlations between B, Cu, Fe, Mn, Mo, and Zn contents of pot-grown wheat in and 46.the whole international material. 84-85

Fig. 47.B, Cu, Fe, Mn, Mo, and Zn contents of pot-grown wheat and respective extractable soil micronutrients as a function of yield. 87

Fig. 48.Regressions of B content of pot-grown wheat on hot water soluble soil B corrected for CEC, (a) all samples, (b) low yielding samples, (c) high yielding samples. 89

Fig. 49.Coefficients for correcting AAAc-EDTA extractable soil Cu for soil organic carbon content, k(org. C), and hot water soluble B for cation exchange capacity, k(CEC). 99

Fig. 50.Coefficients for correcting DTPA extractable soil Mn, AO-OA extractable Mo and AAAc-EDTA extractable Zn for soil pH. 99

Figure National Figure National Nos data for Page Nos data for page 51-68 Belgium 105-116 69-86 Finland 117-126 87-104 Hungary 128-138 105-122 Italy 139-149 123-139 Malta 150-157 140-157 New Zealand 158-168 158-175 Argentina 169-179 176-193 Brazil 180-188 194-199 Ecuador 190-191 200-217 Mexico 194-205 218-235 Peru 207-217 236-253 India 219-229 254-271 Korea, Rep. 232-241 272-288 Nepal 242-250 289-306 Pakistan 251-261 307-324 Philippines 262-270 325-342 Sri Lanka 271-279 343-360 Thailand 280-288 361-378 Egypt 289-299 379-396 Iraq 300-310 397-414 Lebanon 313-321 415-432 Syria 322-331 433-450 Turkey 332-342 451-468 Ethiopia 343-353 469-485 Ghana 355-362 486-503 Malawi 363-371 504-521 Nigeria 372-382 522-539 Sierra Leone 383-391 540-557 Tanzania 392-400 558-575 Zambia 401-409 XIII Preface

In spite of increased use of fertilizers, substantially more macronutrients are still being removed annually from the soil than are applied to it as mineral fertilizers. Some of the nutrients removed are replaced by those in straw, farmyard manure, etc., but on the average the nutrient balance is likely to remain negative. This applies especially to less developed parts of the world. Micronutrients have not generally been applied regularly to soil in conjunction with common fertilizers, and fertilizing soils with macronutrients only is likely to promote imbalances between these nutrient groups as well as between individual nutrients. Further- more, increased yields, losses of micronutrients through leaching, liming, a decreasing proportion of farmyard manure compared with chemical fertilizers and the increasing purity of chemical fertilizers are among the factors contributing toward accelerated exhaustion of the supply of available micronutrients in soils. Hidden micronutrientdeficienciesarefar more widespread thanisgenerally suspected. Micronutrient problems, which are only local today, may well become more serious and widespread in the relatively near future. They must be diagnosed and studied promptly to avoid production problems related to the quantity and quality of foods and feedingstuffs, keeping in mind also the geomedical aspects concerning both humans and animals. Even though much is known of the nature of micronutrient functions, extensive research and experimentation is needed to avoid mistaken use of this knowledge. One of the greatest difficulties encountered in the study of microelements is that different methodologies have been ,adopted for analyses in different countries and laboratories. Sometimes even slight variations in procedure may cause quite substantial differences in analytical results. Since these vary from case to case, it is difficult to compare the results obtained by different scientists. It could be said that scientists using different analytical methods are speaking "different languages". Consequently, knowledge of microelement status is rather fragmentary and there is very little basis for comparison. Discussions on this subject between FAO and Finland in 1973 led to the establishment of an international project called the "Trace Element Study, TF 129 FIN". The purpose of the study, as initially stated, "is to produce fresh information on a world-wide basis on the problems of a number of micronutrients under different soil, climatic and cultural conditions. The results to be obtained will be comparable because all analytical work will be done in one laboratory and because there will be uniformity of the analytical methods to be used. These results should then be used to make interested countries better acquainted with the problems of micronutrients when developing their agriculture and to provide them with guidelines for future research and practical work on a national basis. In other words, the aim is 1) to obtain a general picture of micronutrient status on a world-wide basis, 2) to locate and limit the problematic areas, soils and conditions where one or more of the micronutrients are likely to be deficient and where more detailed future research and field experimentation is needed and 3) to give guidelines to solve the problems in practice." The project was started at the end of 1974 in cooperation with FAO and was financed by the government of Finland. Originally, it was hoped that 25 countries would participate in the project and invitations were sent to 40 countries. Interest in the study was, however, greater than anticipated, and acceptances were received from 35 countries, XIV 30 of which took an active part in the project by collecting and sending soil and plant samples to Finland for analysis in one laboratory (the Institute of Soil Science, Finland) using the same procedures for all samples. Most of the participantswere developing countries, but some highly developed countries were invited for purposes of comparison. The following persons acted as Head Cooperators for the 30 countries participating in the study and were responsible for organizing the collection of representative plant and soil samples for the study: Argentina: Dr. P.H. Etchevehere, Prof. A. Berardo; Belgium: Prof. Dr. A. Cottenie; Brazil: Mr. A. F. Castro Bahia Filho, Mr. S. Wiethölter; Ecuador: Ing. W. Bejarano, Ing. J. Villavicencio; Egypt: Dr. Abdel Hamit Fathi; Ethiopia: Ato Desta Beyene; Finland: Prof. Dr. P. Elonen; Ghana: Dr. H.B. Obeng; Hungary: Prof. Dr. I. Szabolcs, Dr. Eva Elek; India: Dr. N.S. Randhawa; Iraq: Dr. Khalil Mosleh, Dr. Nouri A.K. Hassan; Italy: Prof. Dr. F. Mancini; Korea, Rep.: Dr. Tai-Soon Kim; Lebanon: Dr. A. Sayegh; Malawi: Mr. D.R.B. Manda; Malta: Mr. A. Scicluna-Spiteri; Mexico: Ing. G. Flores Mata, Dr. L. Lopez Martinez de Alva; Nepal: Mr. Manik L. Pradhan; New Zealand: Dr. R.B. Miller; Nigeria: Dr. B.O.E. Amon, Dr. S.A. Adetunji; Pakistan: Dr. M.B. Choudhri; Peru: Mr. J.H. Christensen; Philippines: Dr. G.N. Aleasid; Dr. J. Mariano; Sierra Leone: Dr. I. Hague; Sri Lanka: Dr. S. Nagarajah; Syria: Mr. S. Kourdi; Tanzania: Mr. J.K. Samki; Thailand: Mr. Samrit Chaiwanakupt; Turkey: Dr. M. Ozuygur; Zambia: Mr. Kalaluka Munyinda. I wish to express my gratitude to the above Head Cooperators, their colleagues and subordinates for their excellent cooperation in furnishing me with the sample material and other relevant data requested. A total of 7 488 samples (half of them soil and the other half plant) were received, amounting to 94 per cent of the original target of 8 000. I am greatly indebted to the professional and technical staff of the Institute of Soil Science, Finland, for carrying out the demanding analytical task consisting of about 170 000 chemical and physical analyses and I wish to acknowledge the stimulating discussions with Dr. J. Sippola and Mr. T. Yläranta and their practical help during various stages of the study. I would also like to thank Mr. S. Hyvärinen, who sparing no trouble has performed the statistical analyses and computer work involved. My special thanks are due to Mr. H. Jansson, my assistant, who with unfailing enthusiasm and interest in this work has been of excellent help to me during the longcourse of this study and to Dr. F. W. Hauck, Chief, AGLS, FAO, for his keen interest in this investigation and for his support and cooperation from the very beginning of the project. I wish to extend my thanks to Dr. P. Arens, Prof. Dr. A. Cottenie and Dr. K. Harmsen who read parts of the first manuscript and made several valuable comments and suggestions for clarification, and to Dr. N.E. Borlaug and Dr. R.L. Paliwal who were kind enough to check the lists of wheat and maize cultivars included in this study and furnished me with useful information on this subject. My thanks are also due to Mr. H.K. Ashby for his assistance with the final editing of the text.

Mikko PART I 3 PART I SOIL AND PLANT DATA, METHODOLOGY AND INTERPRETATION

1. Materials and methods

1.1 Collection of original plant and soil samples

Detailed instructions for taking the plant and soil samples were sent to all the field cooperators in various countries in order to standardize the sampling techniques and procedures and to obtain material of good quality for analytical comparisons. The instructions (condensed) are given in Appendix 1. The collectors of samples were also asked to provide additional background information on the sampled fields including: plant cultivars, estimated yields, use of chemical NPK fertilizers during the past three years, applications of farmyard manure (FYM) and lime, pest control measures, soil units (national and FAO/Unesco classification), known or suspected micronutrient deficiencies or toxicities, etc. These data were collected on -Field information forms" provided for the purpose.

1.2 Indicator plants

1.2.1 General

Wheat and maize were selected as original indicator plants for this study, not because they would be the optimum indicators of soil rnicronutrients, but because they are the most widely grown crops all over the world. They also complement each other territorially because maize is often grown in more humid areas while wheat grows also in morearid soils. However, the original plant samples received from different countries proved to be too heterogeneous to reflect reliably the micronutrient status of the soils where the plants had grown. It was already known that the different micronutrient contents of wheat andmaize would alone cause heterogeneity among the plant samples as a whole and make comparison of the findings difficult. In addition, among other factors causinguncontrolled variation in the results of plant analyses were: variation in the physiological age of plants at sampling. The age of maize and 4 spring wheat samples at sampling varied from less than 20 to over 60 days and similarly for winter wheat; contamination of plants by soil in the field or during sampling, sample pre-treatment or transportation; differences in main nutrient fertilization, yield levels, irrigation, use of herbicides and pesticides possibly containing micronutrients. Some on these are referred to in Sec- tions 2.2.1 to 2.2.3 and 8.2.4; variation arising from different plant cultivars. The magnitude of the effects of the above factors is variable from one sample to another and cannot be reliably estimated, but the following attempts were made to quantify some of them.

1.2.2 Sampling time

The effect of sampling time on micronutrient contents of plants was studied by planting two fields each of spring and winter wheat. Sampling, sample pre-treatment and analytical methods were the same as applied to the international study (Appendix 1) except that sampling was carried out twice a week during the whole growing season. There were no clear differences in the concentration or behaviour of micronutrients between the spring and winter wheats. The data from all fields and both cultivars were therefore combined and are illustrated in Fig. 1. They were reported in greater detail by Yläranta et al. (1979). The highest concentrations of B, Cu, Fe, Mn, and Mo were recorded at an early stage of plant growth and followed by substantial decreases toward the end of the growingseason. The decreases in B were from about 7 to less than 2 ppm, in Cu from about 7 to 3-4 ppm, and in Fe from 120 to 40-50 ppm. The changes in the Mn and Mo concentrations were less regular, and in the case of Mn a tendency to increase could be observed toward harvesting time. The behaviour of Zn differed from that of other micronutrients. The changes in Zn concentrations were relatively small, consisting mainly of a slight tendency to increase toward the later stages of growth. Clearly decreasing trends in the macronutrient (P, K, Ca, and Mg) contents of wheat were also reported by Sippola et al. (1978). Similar results have been published elsewhere with other plants. Since the concentration of micronutrients in a plant is dependent on the sampling time (physiological age of the plant), it is evident that concentrations in samples collected at different stages of growth are not strictly comparable and may give a misleading estimate of the micronutrient status of soils. See also Sections 2.5 and 7.1.4.

1.2.3 Contamination

Soil is likely to be the most contaminant of plant samples. It may arise from soil dust raised by wind, the spattering of rain or some other plant-soil contacts. The effect of soil contamination onplantmicronutrientconcentrationdepends on thedegreeof contamination, the chemical element, the total micronutrient content of the contaminating soil, etc. For example, if a plant sample of 2 g dry matter containing 6 ppm B, 7 ppm Cu, 61 ppm Fe, 112 ppm Mn, 0.32 ppm Mo, and 18 ppm Zn (average contents of pot-grown wheat in this study) is contaminated with 10 mg of soil containing 10 ppm B, 50 ppm Cu, 40 000 ppm Fe, 1 000 ppm Mn, 2 ppm Mo and 80 ppm Zn (total contents commonly 5

PPm PPm 120- 10 Fe 100 O 8 0 .0000 80 e 6 °00000 60 0 000 4 o 0. 40 0.00...... 2 20

8 Cu 80 Mn 00000 o° e .00%. 6 60 0,00 0000 000e 0.0. 0. 0.. es° 4 40 e°00 0000000 009 2 20

.000000 Mo 60

.2 o 40 .00.00000.00 0.000 .1 20

MAY JUNE JULY AUGUST SEPT. MAY JUNE JULY AUGUST SEPT. Fig. I. Variations in B, Fe, Cu, Mn, Mo, and Zn contents of wheat (ppm in DM) during the growing season. found in mineral soils) the result of the micronutrient analysis of the plant sample would theoretically be affected by the contamination as follows: The B content would be increased from 6.00 to about 6.02 ppm, i.e. an error of + 0.3 per cent. Similarly, the error for Cu would be 3.1 per cent, for Fe 327 per cent, for Mn 4.0 per cent, for Mo 2.8 per cent and for Zn 1.7 per cent. According to this theoretical calculation, the risk of soil contamination is relatively small in the case of B, Cu, Mn, Mo and Zn if compared to that of Fe. Comparison of the Fe concentration in individual original plants led to the assumption that the samples were liabletobeseverelycontaminated withsoil.Not uncommonly theplant Fe concentrations were increased by obvious contamination to a much greater extent than in the above example. Therefore, all results of Fe analyses from original plants were discarded. 6

1.2.4 Plant varieties

The ability of different plant varieties (cultivars) to absorb micronutrients from soil as well as their micronutrient requirements may vary considerably. For example, Randhawa and Takkar (1975) have drawn special attention to differences in the susceptibility of various species and varieties of crops to Zn deficiency. Because of the large number of plant varieties, data on variety effects are fragmentary and may only be used in special cases for interpreting the results of micronutrient studies. The original plant material received from various countries for this study included over 200 wheat and over 200 maize cultivars, so preventing the study of differences between individual cultivars. An attempt was therefore made to classify both species in two categories each, high yielding varieties (HYV) and local varieties, and to compare the nutrient contents of the two groups. These comparisons covering all plant samples failed for a number of reasons. For example: Neither of the two variety groups was homogeneous. Different high yielding varieties were grown in different countries and the group of local varieties especially was very heterogeneous, consistingasitdid of nativevarietiesindifferent developing countries as well as highly improved wheat and maize cultivars, the products of plant breeding work in developed countries. Great differences between the variety groups in the rates of application of N, P and K fertilizers in different countries affected the nutrient uptake. The estimated yields of the HYV plants were generally higher than those of other varieties; a factor which should be taken into account when nutrient concentrations of plants are to be compared. The nutrient contents of soils where the plants of the two variety groups had grown varied so much that often the grounds of comparison were not justified. In some countries plants of one variety group were mainly grown in irrigated and those of the other mainly in rainfed conditions. However, in a few countries, in spite of some differences in soils, yields or fertilization, the differences in the plant nutrient concentrations between the two variety groups were so clear that some indicative conclusions on their genetic diversities may be drawn. Some of these are pointed out in Part II, see e.g. Mexico, B, Mn and Zn; India, N, P, K, Ca, Mg, B, Cu, Mn and Mo; Ethiopia, Mg, Mn and Zn; and Nigeria, N and P. It must be kept in mind, however, that these comparisons are local in nature and cannot be generalized or applied to other conditions. They serve to draw attention to one of the factors partly responsible for the heterogeneity of the original plant sample material and consequently, for the discrepancy between the results of plant and soil analyses.

1.2.5 Need for better grounds for co paring plant data

In order to reduce the causes of uncontrolled variation in the micronutrient contents as found in the original indicator plants, and to establish sounder grounds for comparison of the results of plant analyses, fresh samples of indicator plants (wheat, cv. 'Apu') were grown in pots on the soils received from the participating countries. It must be understood, however, that environmental factors affecting the availability of rnicronutrients to plants in the field, such as temperature, soil moisture and aeration 7 (redox potential), were also standardized when growing the indicator plants in pots. These effects must therefore be taken into account when interpreting the results obtained from widely varying environmental conditions. Relatively few quantitative data on the effects of temperature on the availability of rnicronutrients are available, but itis generally known that reduction-oxidation processes affect the availability of Fe and Mn more than that of the other four micronutrients included in the study. The reduced forms, Fe2+ and Mn2+, are more soluble and available to plants than are the oxidiced Fe3+ and Mn3+ or Mn4+. For example, in this study the data given in Table 27 in Part II (comparison of original and pot-grown plants) indicate that the plant uptake of Mn is substantially more affected by soil moisture conditions than that of B, Cu, Mo or Zn. Another shortcoming of pot-grown plants is the effect of subsoils. Differences in the micronutrient content and/or availability in the subsoil and top soil may affect the micronutrient uptake by plants, especially during later stages of growth when the root system is more highly developed.Itisbelieved, however, that the foregoing disadvantages are overshadowed by the benefits gained by growing fresh indicator plants in pots. Although the results of micronutrient analyses of the original plants are given in Appendixes 2 and 3, all further discussion (if not specifically stated otherwise) concerning the behaviour of the six micronutrients B, Cu, Fe, Mn, Mo and Zn, is based on results obtained from analyses of the whcat samples grown in pots under controlled and uniform conditions.

1.2.6 Growing new indicator plants in pots

The small quantity of soil available from samples received from participating countries necessitated the use of small pots (200 ml) partly (75 ml) filled with quartz sand (Fig. 2). The quartz sand was washed with 6 N HC1 and thereafter repeatedly with deionized water until no reaction to AgNO3 was observed. Two-thirds (50 ml) of the quartz sand was placed below the soil (100 ml) and one-third (25 ml) above the soil to prevent destruction of the soil surface structure and prevent crust formation due to repeated irrigations as well as to minimize contamination of plants by soil. Holes in the bottoms of the pots were covered with filter paper to prevent leakage of quartz sand but to allow excess irrigation water to drain off. Any excess water was collected in a dish beneath the pot and re-used for irrigating the plants in that pot. Irrigation was carried out once a day using deionized water in quantities sufficient to keep the soil moisture content close to field capacity. To prevent deficiencies of macronutrients, fertilizers corresponding to about 60 kg N, 30 kg P, 30 kg K and 15 kg Mg per hectare were applied in liquid form in irrigation water in two equal dressings, one at the end of the second and the other in the middle of the fourth week from planting. Spring wheat (cv. 'Apu') was used as an indicator plant. Ten seeds were placed in each pot on the surface of the soil and covered with the 25 ml of quartz sand. After sprouting, the number of plants was reduced to eight. The plant tissues from 1 cm above the surface of quartz sand were harvested on the 36th day from planting, dried, weighed and analysed as explained in Section 1.3.2. About five percent of the soil samples were too small to enable new indicator plants to be grown in pots. 8

Fig. 2. Growing new indicator plants (spring wheat cv. 'Apu') in pots.

1.3 Analytical methods In order to obtain background information on factors affecting the micronutrient contents of soils and plants, a number of additional analyses were made both on soils and plants. These data includesoiltexture,cation exchange capacity, pH, electrical conductivity, CaCO3 equivalent, organic carbon content, volume weight of soils and macronutrient contents of both soils and plants. At an early stage of the study when selecting the methods for soil micronutrient analyses, preliminary studies with different extraction methods were carried out on a limited number of samples. For four micronutrients (Cu, Fe, Mn,, Zn) each of the international samples was analysed by two different methods as described below. 9 1.3.1 Methods of soil analyses

Particle size distribution was determined by dry and wet sieving and for the finer fractions by a pipette method (Elonen, 1971). Four fractions (< 0.002, 0.002-0.06, 0.06-2.0 and > 2.0 mm) were determined. For the expression of texture as a single figure a texture index (TI) was calculated (TI = 1.0 x % of fraction < 0.002 + 0.3 x % of fraction 0.002-0.06 + 0.1 x % of fraction 0.06-2.0 mm). Cation exchange capacity (CEC) of soil was determined by the Bascomb (1964) method except that the Mg concentration was measured with an atomic absorption spectrophotometer (Varian Techtron 1200; air-acetylene flame) after diluting the final filtrate by a factor of ten with a 0.24 M HC1 + 0.28 % La solution. Electrical conductivity was measured from a soil:water (1:2.5) suspension after letting the suspension settle overnight. Soil pH(H20) was determined on the above suspension after stirring. For measuring pH(CaC12) the suspension was made 0.01 M with respect to CaCl2 and stirred. The pH (CaCl2) was measured after two hours from a restirred suspension. The relationship between the two pH values is given in Section 2.1 and Fig. 5. CaCO3 equivalent of soil was determined by a method described by Trierweiler and Lindsay (1969). Organic carbon was determined by a modification of Alten's (1935) method (Tares and Sippola, 1978). An amount of soil containing 30-100 mg of humus, 25 ml 0.25 M K2Cr207 and 40 ml conc. H2SO4 were put into a 400 ml flask. The mixture was kept for 1.5 h on a hot water bath, allowed to cool for 30 min. and 175 ml of water were added. After standing overnight, the solution was measured colorimetrically and the reading compared to standards (red filter, 620-645 nm). Volume weight of soil was determined by weighing a 25 ml sample of air dried soil passed through a 2 mm sieve (Tares and Sippola, 1978). Soil total nitrogen was determined by the Kjeldahl method using Tecator equipment. A 1-3 g sample of air dry soil was weighed into a digestion tube containing 8 g of K2SO4 and 1 g of CuSO4. Conc. H2504 (20 ml) was added. The mixture was maintained at about 420 °C (for 1-2 h) until clear. To the cooled digest 50 ml of water were added, made alkaline with NaOH and nitrogen distilled into 4 % boric acid and titrated with 0.01 M HC1 using brómocresol greenmethyl red as an indicator. Phosphorus was extracted with 0.5 M NaHCO3, pH 8.5 (Olsen ad., 1954). The volumetric soil:extractant ratio was 1:20 and shaking time 1 h (27 r.p.m.). Phosphorus was measuredcolorimetricallywith ammonium molybdate-ascorbicacidreagent (Watanabe and Olsen, 1965). Potassium, calcium, magnesium and sodium were extracted from soil with1 M CH3COONH4, pH 7.0 (Cahoon, 1974). The volumetric soil:extractant ratio was 1:10 and shaking (end over end) time 1 h (27 r.p.m.). Conc. HC1 and La were added to the extract to make the extract 0.2 M with respect to HC1 and contain 0.25 % La. The cations were determined with an atomic absorption spectrophotometer (Techtron AA-4 or Varian Techtron 1200) using an air-acetylene flame for Ca and Mg and air-propane for K and Na. Boron. Hot water soluble soil B was determined by a modified Berger and Truog (1944) method (Sippola and Erviö, 1977). A 25 ml soil sample, 50 ml of water and about 0.5 ml of activated charcoal were boiled for 5 min. in a quartz flask and filtered immediately. Two ml of the extract and 4 ml of buffer masking agent (250 g CH3COONH4 and 15 g Na2EDTA dissolved in 400 ml water, and 125 ml 100 % CH3COOH added) were mixed lo and 4 ml of azomethine reagent (0.9 g azomethine-H and 2 g ascorbic acid dissolved in 200 ml water, prepared daily or weekly if refrigerated) added. The colour was allowed to develop for 1 h, intensity measured spectrophotometrically at 420 nm and compared to standards varying from 0 to 2 mg boron per litre. To correct soil B values for CEC, see Section 2.3.3. Copper, iron, manganese and zinc were extracted from all soils by two extraction methods, AAAc-EDTA and DTPA as extractants. The acid ammonium acetate-EDTA extractionsolution,0.5 M CH3COONH4, 0.5 M CH3COOH, 0.02 M Na2EDTA, was made by diluting 571 ml 100 % CH3COOH, 373 ml 25 % NH4OH and 74.4 g Na2EDTA (EDTA = ethylenediaminetetracetic acid) to 10 litres with water. The pH was adjusted to 4.65 with acetic acid or ammonium hydroxide (Lakanen and Erviò, 1971). Soil (25 ml) and extracting solution (250 ml) were shaken for 1 h (end over end, 27 r.p.m.). The suspension was filtered using Whatman No. 42 filter paper. The concentrations of Cu, Fe, Mn and Zn were determined with an atomic absorption spectrophotometer (Varian Techtron 1200, air-acetylene flame) and appropriate standards. The DTPA extracting solution was prepared to contain 0.005 M DTPA (diethylene- triamine pentacetic acid), 0.01 M CaC12, 0.1 M TEA (triethanolamine) and was adjusted to pH 7.3 with 11C1 (Lindsay and Norvell, 1969, 1978). Soil (25 ml) and extracting solution (50 ml) were shaken for 2 h (end over end, 27 r.p.m.). The suspension was filtered using Schleicher & Schtill selecta 589/3 filter paper. The concentrations of Cu, Fe, Mn and Zn were determined with an atomic absorption spectrophotometer (Varian Techtron 1200, air-acetylene flame) and appropriate standards. The results presented in this paper are based an AAAc-EDTA extraction of Cu and Fe and on DTPA extraction of Mn. For Zn the results obtained by both extraction methods are given. For details, see Sections 2.3.4 to 2.3.7. Molybdenum was extracted with A0-0A solution. The ammonium oxalate-oxalic acid extracting solution, a modified Tamm's (1922) solution, was prepared by dissolving 249 g ammonium oxalate and 126 g oxalic acid in 10 litres of water. The pH was adjusted to 3.3 with ammonium hydroxide or oxalic acid. Soil (10 ml) and extracting solution (100 ml) were shaken for 16 h (end over end, 27 r.p.m.) and the suspension was filtered. Mo was determined by the zinc dithiol method (Stanton and Hardwick, 1967). A 50 ml aliquot of the filtrate was evaporated to dryness in a quartz dish on a water bath, oven dried at 105 °C overnight and ashed by raising the temperature slowly to 600 °C. The ash was moistened with 1 ml of water and dissolved in 15 ml 4M 11C1 on a water bath for 1 h. If the residue had dissolved completely the solution was washed (4 times with 4 ml of warm 4 M HC1) into a separating funnel. If there was an insoluble residue in the quartz dish, the solution was filtered into a separating funnel and the dried (105 °C) filter paper with the residue ashed at 600 °C for 1 h. The ash was treated with 0.5 ml 40 % HF and 0.1 ml 70 % HC104 in a teflon dish, dissolved in 1 ml 4 M 11C1 and combined with the main solution in the separating funnel. The following reagents were added to the main solution: 2 ml of Fe reagent (10 g FeNH4(SO4)2 in 1 litre of 6 M HC1), 2 ml of reducing solution (15 g ascorbic acid + 7.5 g citric acid/100 ml 1170) shaken and left to stand for 2 min., 2 ml of 50 % potassium iodide, shaken, 2 ml of zinc dithiol solution (0.3 g Zn dithiol, 2 ml ethanol, 4 ml water, 2 g NaOH and 1 ml of thioglycolic acid were mixed and when clear, diluted with water to 50 ml, and 50 11 ml of fresh 50 per cent potassium iodide solution added) mixed and left to stand for 2 min., 5 ml of chloroform. The separating funnel was shaken vigorously for 2 minutes. The chloroform phase was separated. The intensity of colour in the chloroform was compared to molybdenum standards using a spectrophotometer at 680 nm wavelength. The standards (0.2-10.0 mg Mo/30 ml 4 M HC1) also contained the same reagents as the sample solution in the separating funnel and received a similar treatment. For correcting the results of A0-0A extractable soil Mo values for pH, see Section 2.3.2.

1.3.2 Methods of plant analyses

Preparation of samples. Air dried plant material was ground with a hammer mill of pure carbon steel to pass a 2 mm sieve (pure carbon steel). To avoid contamination, the samples were stored in plastic (pure polyethylene) bags. Since the mineral element contents were expressed on a dry matter basis, the DM contents of air dry samples (5.0 g) were determined by drying the samples at 105 °C for 4 h. The samples were cooled in a desiccator for 2 h and weighed before ashing. The nitrogen content of plant material was determined by the Kjeldahl method using Tecator equipment. The procedure differed from that of soil total nitrogen determination only as regards the size of sample which was 0.50 g of air dry plant material. Results were expressed on a DM (105 °C) basis. The boron content of plant material was determined by a modified azomethine-H method (Basson et al., 1969, John et al., 1975, Sippola and Erviii, 1977). A 100 mg sample of air dry plant material was weighed into a quartz dish, ashed at 450 °C overnight and cooled. A 10 ml aliquot of 0.1 M HC1 was added and covered with a watch glass. The solution was allowed to stand for 4 h and filtered into a test tube. Two ml azomethine reagent were added and the spectrophotometric measurement was made according to the procedure described for soil boron analysis. Results were expressed on a DM (105 °C) basis. Ashing and determination of Ca, K, Mg, P, Cu, Fe, Mn, Mo, and Zn. The oven dry samples described above were ashed by raising the temperature slowly (about 2.5 h) to 450 °C and maintaining this temperature overnight. The ash was moistened with a few drops of water, covered with a watch glass and 10 ml of 6 M HC1 slowly added. When the reaction had ceased, the watch glass was removed and washed with 10 ml of 6 M HC1. The solution was evaporated to dryness on a water bath, 40 ml of 0.2 M HC1 were added, covered with a watch glass and kept on the water bath for 30 minutes. The hot solution was filtered (SchleicherSchtill selecta 589/2) into a 100 ml volumetric flask. The dish, filter, and undissolved residue were washed with about 40 ml of hot 0.2 M HC1. The filter paper and residue were ashed in the original quartz dish by raising the oven temperature (over a period of about 2 h) to 600 °C, thereafter cooled and transferred to a teflon dish to which 0.2 ml HC104 and 5 ml HF were added. The dish was kept on a hot plate (200 °C) until all the HF had been removed, cooled, and 0.5 ml 6 M HC1 was added. The dish was kept on the hot plate until nearly dry, the residue dissolved in 5 ml of 0.2 M HC1, filtered, and the filtrate combined with the main sample solution in the 100 ml 12 volumetric flask. The teflon dish and filter paper were rinsed and the flask was filled to volume with 0.2 M HC1. For storage, polyethylene bottles were used. For Ca, K, Mg, and P determinations, 8 ml of the main sample solution were diluted to 40 ml with lanthanum-HC1 solution (0.2 M HC1 containing 0.31 % La). Ca, K, and Mg were determined with an atomic absorption spectrophotometer (Techtron AA-4 or Varian Techtron 1200) using an air-acetylene flame for Ca and Mg and air-propane for K. For determining P colorimetrically with a modified ammonium vanadate-molybdate method (Gericke and Kurmies, 1952), 5 ml of the dilute sample solution and 10 ml of reagent (7.5 g (NH4)6Mo7024. 4 H20 and 0.38 g NH4V03 in 1 litre of 0.13 M H2SO4) were mixed. The colour intensity was measured and compared with appropriate standards (0-70 mg P/1) one hour after mixing. Cu, Fe, Mn, and Zn were measured directly on the main sample solution with an atomic absorption spectrophotometer (Varian Techtron 1200) as on soil extracts. Mo determination was made by the zinc dithiol method (Stanton and Hardwick, 1967). To make the solution 4 M with respect to HC1, 20 ml of the main sample solution and 10 ml of conc. HC1 were pipetted into a separating funnel and shaken. Fe reagent, reducing solution, potassium iodide, zinc dithiol and chloroform were added and Mo was determined according to the procedure described for soil Mo analysis. The same Mo standards were used for plant and soil analyses. Micronutrient contents of pot-grown wheat. Owing to small pot yields, the contents of the six micronutrients (B, Cu, Fe, Mn, Mo and Zn) only were determined. The analytical procedures and methods deviated from those described above only in that smaller quantities of sample material were analysed.

1.3.3 Statistical methods

The evaluation of the micronutrient status of each soil was based on two analyses, i.e. one of the soil and one of the plant grown on that soil. Regression graphs are used to present the results of both analyses simultaneously. Furthermore, the correlation between the results of the two analyses can be considered a good measure for judging the reliability of the results. In general, for statistical evaluation of the results of chemical micronutrient analyses by countries (Part II) four simple regression equations were computed: the linear, logarithmic and two semi-logarithmic models. The best fitting equations are graphically presented. In addition to the above regression models, parabolas, cubic parabolas and their respective logarithmic modifications were adopted when studying the relations between micronutrients and various soil factors (PartI).In special cases multiple regression and linear stepwise multiple regression analyses were used. In respect of macronutrients and general soil properties the data in Part II are presented only as arithmetic means and their standard deviations and frequency distribution graphs of classified materials. 13 1.4 Expression of analytical data

1.4.1 Soils

The results of nutrient analyses of soils have traditionally been expressed on a weight basis, i.e. in ppm or mg/kg soil, but expressions on a volume basis, e.g. in mg/dm3 or mg/ litre of soil, are becoming increasingly frequent. The weights of a certain volume of soil may vary considerably depending on soil texture, mineral composition etc., but especially on the organic matter content of the soils. In the soil samples studied, which included only a few organic soils, the volume weights varied from 0.51 g/cm3 (a New Zealand peat soil) to 1.77 g/cm3 (a coarse sand from Nigeria). However, much lower volume weights have been recorded, e.g. Sippola and Tares (1978) reported an average volume weight of 0.15 for cultivated Sphagnum peat soils and a minimum value of 0.06 g/cm3. Although the volume of soil where the roots of a plant are distributed may vary from one plant to another, it can be assumed that the density of roots and nutrient absorption by plants are much better related to soil volume than to the weight of the soil, since in the latter ten- or twenty-fold variations often exist. Often, in order to understand or to interpret conflicting results, the importance of the dimension or unit in which the results are given must be realized. For example, if two soils, one a mineral soil with a volume weight of 1.5 and the other a peat soil with a volume weight of 0.1 g/cm3 are analysed and both show an equal content of 100 ppm of nutrient X when expressed on a weight basis, the result is completely different if expressed on a volume basis, as shown below:

Table 1. Comparison of nutrient content of two soils expressed on weight and volume bases.

Soil Volume Content of nutrient X as expressed weight on weight basis on volume basis ppm mg/litre of soil Mineral soil 1.5 100 150 Peat soil 0.1 100 10

Thus, in this example, the 100 ppm of nutrient X in the mineral soil is 1400 per cent higher than the 100 ppm in the peat soil when the same results are expressed on a volume basis. The above is directly applicable when the total nutrient contents of soils are determined. In the case of a determination of extractable contents, the difference is not as pronounced because of the effect of changing extraction ratio. Even so, misleading conclusions have too often been drawn because insufficient thought has been given to the dimension of the results presented. In areas where peat soils do not exist and volume weights of soils vary less, the difference between the methods of expressing the results is, however, relatively small. On the other hand, even within the same soil profile the volume weight of the topsoil may be markedly different to that of the lower horizons. The relative error due to inaccuracies in measuring a certain volume of soil for analysis, or due to variation in extraction ratio, is negligible compared to the effects of variation of volume weight among samples including both mineral and organic soils. The advantages of volume-based expressions have been pointed out, for example, by Mehlich (1972) and Sillanpää (1962a, b, 1972a, b). 14

In accordance with the above prihciples, the contents of all soil nutrients in this study are expressed on a volume basis, i.e. in mg/I. However, the total soil N content is also given as a percentage (Appendixes 2, 3 and 4) but for some of the soil characteristics such as organic carbon content and CEC the traditional dimensions are used. The quantities of various soil nutrients are expressed in terms of the element rather than oxide. The same applies also to the nutrient contents of plants and fertilizers.

1.4.2 Plants

Plant analytical data are expressed on a concentration (To or ppm) basis rather than as uptake per hectare. Both methods have certain advantages as well as disadvantages, as discussed in Section 2.6. The plant/soil micronutrient correlations calculated for B, Cu, Fe, Mn, Mo and Zn in all samples (Section 2.3) showed no indication of the superiority of one method over the other. The decision to express as concentration was due to the practical fact that in order to be able to calculate uptake, yield must also be known and, generally, at the time of plant sampling only rough estimates of yield can be made.

1.4.3 Regression graphs

To facilitatethecomparisonofresults(e.g.betweendifferentcountries),the corresponding values of the soil and plant analyses are presented in regression graphs in order to combine the results of the two techniques. The higher the correlation coefficient, the better the link between the soil and plant analyses. The regression linc and equation as well as the mean values and standard deviations for all the international material are usually given as background information and also in regression graphs concerning only individual countries or other groups of data. Because of widely varying micronutrient contents and their abnormal (linear) distributions, logarithmic scales were adopted for regression graphs. Each regression graph has been divided into five Zones (IV) in such a way that the numerical product of plant micronutrient content multiplied by soil micronutrient content is constant at every point on the line separating two zones. The numerical values along the lines demarcating the zones are given in Table 2. The locations of the zone limits have been determined so that five per cent of sample pairs (plant micronutrient content x soil micronutrient content) of the whole international material falls below the line separating Zones I and II, i.e. into Zone I. Accordingly, ten per cent of the material falls below the line between Zones II and III, i.e. five per cent into Zone II. The next 80 per cent of the material falls in Zone III and the highest ten per cent of the values in Zones IV and V, i.e. five per cent in each. Even though the most likely cases of deficiency are to be found among samples falling in Zone I and the most probable cases of toxicity in Zone V, there is no proof that the above zone limits would also be critical limits for deficiency or toxicity. Furthermore, critical deficiency and toxicity limits vary from one plant species to another. 15

Table 2. Numerical products of plant content x soil content along the lines dividing the regression graphs into five zones.

Micronutrient and Lower Lower Higher Higher soil extraction 5 % limit 10 % limit 10 % limit 5 % limit methodl) between Zones between Zones between Zones between Zones I and II II and III III and IV IV and V B (hot water sol.) 0.67 0.87 9.4 18.7 B (CEC-corr. hot w.s.) 0.72 0.98 9.2 17.5 Cu(AAAc-EDTA) 3.7 6.4 107 160 Cu(Org.C-corr. AAA c-EDTA) 3.9 6.7 102 144 Fe (AAAc-EDTA) 1680 2300 22500 31900 Mn (DTPA) 206 295 14100 25100 Mn(pH-corr. DTPA) 166 201 14100 37100 Mc> (A0-0A) 0.00260 0.0040 0.166 0.280 Mo (pH-corr. A0-0A) 0.00070 0.0017 0.275 0.441 Zn (DTPA) 2.0 2.8 86 183 Zn (AAAc-EDTA) 7.2 9.5 169 336 Zn (pH-corr. AAAc-EDTA) 6.4 8.3 180 370 I) For CEC-, org. C-, and pH-corrections, see Sections 2.3.2-2.3.7. 17 2. Results and discussion

2.1 General propertiesof soils and their mutual relations

The amounts of micronutrients removed yearly with normal crop yields represent only a very small proportion, generally less than one per cent of the total amount present in soils. The total amounts, even in serious deficiency cases, therefore far exceed the crop requirements. Cases of primary deficiency that are mainly caused by a low total content of micronutrients, are therefore very rare in normal agricultural soils, but may occur in severely leached sands or in certain peat soils. Secondary micronutrient deficiencies are the most common and are caused by soil factors reducing the availability to plants of otherwise ample supplies of micronutrients. For the effective correction in the field of a micronutrient deficiency, it is necessary to know which element is deficient and the reason why it is deficient. While primary deficiency can usually be remedied by applying salts containing the deficient micronutrient to the soils, the correction of secondary deficiencies is more complicated. Direct salt applications to the soil may not lead to a cure because the original cause of deficiency persists and renders the element added unavailable. Should it be impossible to lessen the effect of the causative factor, other measures must be taken, such as foliar applications or the use of chelates. It is most important to know the main reasons for a deficiency. Where micronutrient disorders occur, or are suspected, a general background knowledge of soil characteristics is essential. Therefore, in this study emphasis has been laid on investigations concerning the relationships of plant and soil micronutrients to various soil characteristics. For such work, the present abundant material with exceptionally wide variations in respect of all soil characteristics was well suited. Relationships of plant and soil micronutrients to various soil characteristics are presented in detail in Section 2.3, separately for each of the six micronutrients under study. General information on soil characteristics by countries is given in Appendixes 2, 3 and 4. The frequency distributions of texture, pH, organic matter content and cation exchange capacity of soils in each country against the background of corresponding distributions in the whole international material are presented in Part II. The effects of two or more different soil factors on the availability of certain micronutrients are often very similar (e.g. effects of texture and CEC on Cu, Fig. 30). In other cases the same soil factor affects the availability of two or more micronutrients very differently (e.g. effects of pH on Mo and Mn, Figs 16 and 34). Furthermore, it is often difficult to define which soil factor affects directly the availability of a certain micronutrient and which factor is only in "pseudocorrelation" with the micronutrient, owing to the mutual correlation between the two soil factors concerned. From the viewpoint of soil chemistry these questions are of importance, but in practical micronutrient studies a "pseudocorrelating" soil factor may often be as informative as a factor of direct effect. To understand better the relations between various micronutrients and other soil factors to be given in Sections 2.3 and 2.4 the mutual relations between the six soil characteristics are presented in Figs 3 and 4. 18

a f k 90F e log y =4.14-6.839 log x y =5.11+0.0628 x-0.000561 x2 y=7.45-0.695 x+0.0356x2 X +4.547flogx)2 R=0,208 -*" 0.) -F., R =0,401"" '0 70 R=0.177 - . 7 C C.5 7 ------....N o C...) a)50 17 g._ U ci, T o_ x301- 5 5 G) I.- 10 4

O log y=-9.065+24.62 log x log y=-1.729+1.635 log x log y=1.629+0.271 log x 6.4 -16.361log x)2 -0.3356(log x)` X -0 2642(log x)2 R=0 421*** a2R=0.382 -0(1) 70 R=0.413

50 ci cr) .8 x 30 e 5 4 0 .4 a) 10 .2 .2

h rIl 64,c ,ogY=0.252-,0.330x-0.0235x2 54 - log y=0.589+0.0243 x 64log y=1.377+0.5142 log x 0CD R,.0.169 '-'' C3')-0.000132 x2 cr) -0.2126110g x ) 2 a,32 032 - R=0.832*** 0 32 R=0.535 -" ,_. 0 9.) 43 (-3.3 E 16 ------15 E 16 E Li c.,5 CS 0al 8 0Lii 8 8 4

log y =1.774-0.706 x 0694 x2 -T _ log y=-0.382+0.0251x log y=02543-O.0819 log x E 25.6 E25.6 0)25.6 u R=0.571"-' o -6.000224 x2 o -0.2478(log xl2 (I) R=0.255 o R=0 099 S 54 -a-)6.4 E -c; C 1.6 C 16 a 16 O O o o oO

e o log y=-7195+1.017 x y=-2.795+0.0938 x log y =0.0445-3.374 x 6.4 r =0.887 "' 64 -0.000808 x` 6,4r=-0.278"." R=0.291'*" 7 cr1.5 1.6 a) Cr

co 4 .4 0 o 1 C-)

4 5 6 7 8 10 30 50 70 90 .2 .4 .8 1.6 32 66 Soil pH (CaC(2) Texture index Organic C, % Fig. 3.

Figs 3 and 4. Mutual correlations between soil pH(CaC12), texture (TI), cation exchange capacity (me/100 g), organic carbon (%), electric conductivity (le s cm-1), and CaCO3 equivalent (%), in the whole international material (n 3536). The best fitting regressions are given. 19

8 - log y = 0.752 0.0452 log x 8y=628+2.416 log x-1.037(log x12 8 y=7.21+ 0,626 log x-0,157(lo x12 r= = 0.578". R = 0.903"* 7 7 7

o 6 06

o o_

1. 4

b 90-iog y . 0.802 0.633 log x 90log y .1.597+0.143 log x 90log y =1.6300.0352 log x x -0.C282(log x -0.0642110g x) o o X 0.00232 (log x12 -079 -R . 0.798 ' ''' -0 70R 0.221" ." -0 7 R = 0.248 " C C

a)50 E, 50 w 50 I_ D o 4),- 30 .; 30 -5(-. 30 0 G.) a) H x)2 I- 10 10 10

rrl log y = -1.356 +1.521 log x iog y.0.0412 +0.004186 log x y.1,08- 0.1305(09 x 6.4 -0.36261l0g x12 6.4 -0.07069110g x)2 6,4 C 0804(log x12 R.0,538 "" R. 0.093 - 0.296 3.2 32

.01.6 _1.6 (,)1.6 E 0 .8 o .8 8 o.4 0 4 o 4 2 .2 2

7 log y -0.852 +1.435 log x log y =1.354 0.2348 log x 54log y=1.408 + 0.0333 tog x oE 25'6 -0 4493(log x)2 -0.1706 (log x12 -0.00897(log x12 R R = 0.223 0 32 R.0.223- U) O (7)6.4 -- 4(16 E .6 o L5 o 4

e o log y-351, +3.53 log x log y -0.8645 .og y = 0.276 +0.1743 log x 2.614 log x ,TE 25.6 - -0.8979(log x12 6,4 -0.97050og x/2 ,-0.019201 log x12 R = 0,225 = . 0.545" 6.4

6 o u Eli

4 8 15 32 64 .4 1.6 6.4 25.6 .1 1.6 6.425.6 CEC, me/100g El. cond10-4S cm-1 CaCO3 equiv,% Fig. 4.

Although all six soil factors are statistically highly significantly (0,1 per cent level) correlated with each other, the correlation coefficients vary from 0.093*** to 0.903***. (Regression curves between the best correlating soil factors, R>0.5, are drawn with thick lines). 20 The closest correlations were found_ for pHCaCO3 equivalent, CECtexture, and pHelectrical conductivity, but those for CaCO3 equivalentelectrical conductivity and CECorganic C were also high. The relations between electrical conductivity and organic C, pH and CEC, and pH and texture were the weakest but still highly significant. Three of these soil factors (pH, CEC and organic carbon content) are of special importance in connection with micronutrients because (a) they are highly responsible for the degree of availability of micronutrients to plants, and (b) they influence the extractability of micronutrients by the extractants used, and (c) because their effects during the above processes (a and b) are dissimilar. With regard to soil pH the results given in this study are based on determination of pH(CaC12). The advantages of this method have been pointed out,e.g. by Peech (1965). In addition pH(H20) was determined from all soil samples. The relationship between the two pH values is given in Fig. 5. In general, the pH(H20) is higher than the pH(CaC12) and the difference was somewhat greater in acid than in alkaline soils. To convert pH(CaC12) to pH(H20) or vice versa the regression lines in Fig. 5 can be used.

o cs,

6 X n = 3538 CI, pH (Ca Ci2) 0.808 +1.044 pH (H20) pH (H20) 0.937 + 0.934pH (CaCl2) r = 0.987 ***

4 5 6 9 pH(Ca C12) Fig. 5. Relationship between pH(H20) and pH(CaC12). 21 2.2 Macronutrients

2.2.1 General aspects The data on macronutrients are based on analyses of soils and of the two original indicator plants, wheat and maize, grown in those countries participating in the study. This plant sample material was quite heterogeneous because, in addition to two indicator plants, it consisted of numerous plant varieties, sampled at considerably varying stages of growth and fertilized with widely different dressings of N, P and K fertilizers. Unlike the micronutrient data of this study, macronutrients were not analysed from the more homogenf ous plant sample material obtained from the pot experiment because these samples were too small in size for determining macronutrients in addition to micronutrients. Furthermore, the pot-grown wheats were fertilized with NPK.

2.2.2 Comparison of the two original indicator plants

In general the sampled wheats contained more N, P and K, but less Ca and Mg than maize (Table 3). Despite the wide variations in the soil macronutrient contents, the mean values for wheat and maize soils differed relatively little. More N, P and K had been applied to the wheat than to the maize crops, so partly accounting for their higher N, P, and K contents.

Table 3. Mean macronutrient contents of the two original indicator plants (%), of the respective soils (total N %, others mg/1), and the mean amounts (kg/ha) of N, P and K applied in fertilizers to each crop.

Wheat fields Maize fields (n = 1768) (n = 1976) N P K Ca Mg N P K Ca Mg Plants '), mean 4.28 0.375 4.03 0.428 0.172 3.14 0.330 3.13 0.470 0.251 +s 1.15 0.125 0.97 0.166 0.060 0.87 0.104 0.96 0.205 0.119 Soils'), mean 0.133 20.2 365 4671 489 0.135 22.5 330 3450 446 +s 0.084 24.7 283 3076 437 0.088 33.0 356 2815 462 Fert. appl.,mean 66 21 21 27 6 7 + s 61 27 44 37 10 18

I) For analytical methods, see Section 1.3.

2.2.3 Comparison of macronutrient status in different countries

Since plant samples received from some of the participating countries consisted of wheat only and others maize only, their plant macronutrient data were not directly comparable. An attempt was therefore made to eliminate the difference in the uptake of macronutrients between the two plant species by raising (pooling) the average macronutrient contents of maize to the same level as those of wheat. For example, the N contents of individual maize samples were multiplied by a factor of 1.36 to raise the average N content of maize (3.14 %) to the same average level as that of wheat (4.28 %). The national mean data for soil macronutrients and the respective adjusted plant data are expressed in Figs 6, 7, 8, 9 and 10. 22 It must be understood that even-though the main differences due to two indicator plants were eliminated by statistical pooling (adjustment of the average macronutrient content), other causes of uncontrolled variation, such as those due to different plant varieties, varying ages of plants at sampling, fertilization, etc., remained, and good correlations between the plant and soil macronutrient contents can hardly be expected. In fact, the macronutrient correlations based on the original plants are generally poorer than those of most micronutrients where many of the factors causing uncontrolled variation in micronutrient contents of plants were eliminated by replacing the two original plants and numerous varieties by pot-grown wheat of one variety, sampled at the same time (see Section 1.2). Furthermore, compared to fertilization with macronutrients, the effects of micronutrient fertilization were negligible because fertilizers containing micronutrients were very seldom used. The general effects of recent N, P and K fertilization are illustrated by dividing the data into three fertilization classes (Figs 6, 7 and 8), but the data concerning farmyard manure application were too fragmentary to be taken into account. The influence of different wheat and maize varieties is discussed in Section 1.2.4. The data in Figs 6 to 10 give only a general picture of macronutrient status in various countries against the international background and will be discussed in more detail at national levels in Part II. Although the total nitrogen content of topsoil cannot always be considered a good indicator of the nitrogen status of a soil, the correlation between the N content of the plant and total soil N in the whole material is relatively good (Fig. 6, regression a). Nitrogen fertilization is a factor with a strong effect on plant N content but affects little the total soil N (Table 4) and consequently is likely to impair the correlation. Therefore, to obtain a general outline of the effect of applied nitrogen on the N content of the original plant samples, additional regressions were computed for two subgroups: one consisting of plants that had received more than 100 kg N/ha (Fig 6, regression b) and the other of plants that had received less than 50 kg N/ha (regression c). There is a considerable difference between the elevations of the two sub-regression lines (b and c). Most of the countries where high rates of nitrogen had been applied to the sampled crops (Hungary, Belgium, Korea, Italy, Zambia) stand clearly above the general regression line(a),while most countries of low nitrogen fertilizer use lie below that line. In case of phosphorus the plant-soil correlation is better than that for other macronutrients. Although the effects of recent phosphorus fertilizer applications can be seen from Fig. 7, it is evident that the applied P not only increased the P content of the plant but also the P extractable from the soil by 0.5 M NaHCO3. Since applied P is not leached as easily from soil as N, and plants remove much less P from soils than N or K, many countries with a traditionally high use of phosphates have built up the P reserves of their soils during the past decades. According to FAO statistics (FAO, 1980a) phosphate fertilizer consumption per hectare in 1961-78 has been considerably higher in New Zealand, Belgium, Korea, Finland, Hungary and Italy than in the other countries included in this study. In spite of relatively strong fixation of applied P in soils, the effects of recent phosphate applications on plant P contents are quite substantial as shown by the differences in the elevation of regression lines b and c representing high (> 30 kg P/ha) and low (< 10 kg P/ha) phosphate application levels, respectively. Since a large proportion of analysed samples which had received high (>30 kg P/ha) phosphate dressings came from the countries with a traditionally high phosphate consumption, the average NaHCO3 extractable P content of these soil samples was double 23

N _ b 6 6Hu - a Be0 .Ko

5 a Fi 0 It oAr oNZ ir C 4 RPo Th .Me oEt Za e ..Eg pe Q. 4 Tu eS ph Br ....,,..----- Ni c 0° SrooGh ..,ic o Mw fro Ec -I- o c: In. a) Mt Ta o 3 o Aver.fert.application

o <50 kgN /ha

la 50-100kgN/ha 2 e >100 kgN/ha

4 I I I I .04 .Ö6 .08 .1 .2 .3 .6 .8 (total) N insoil , % Fig. 6. Regressions of nitrogen content of original wheat and maize (pooled) on total nitrogen content of soil a) for the whole international material, b) for crops fertilized with >100 kg N/ha, and c) for crops fertilized with <50 kg N/ha. Regression equations are given in Table 4. The national mean values for plant and soil N contents of the 30 countries are plotted in the graph indicating also the general level of nitrogen fertilizer application to the sampled crops in each country:

Ar = Argentina, Be = Belgium, Br = Brazil, Ec = Ecuador, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le = Lebanon, Mt = Malta, Mw= Malawi, Me = Mexico, Ne Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si= Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta = Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia. that of samples receiving only small (< 10 kg P/ha) phosphate dressings (Table 4). This difference accounts for the divergence of regression line a from lines b and c. According to the present data P deficiency is likely to occur more frequently in Pakistan, Iraq, Ghana and India than elsewhere, although in most countries soils with low P status are common. 24

fertapplication

50 kg P /ha 50 -100 kg P/ha >100 kg P/ ha

10 20 30 50 70 100 200 (Na HCO3) Pin soil,mg /I Fig.7. Regressions of phosphorus content of original wheat and maize (pooled) on NaHCO3 extractable phosphorus content of soil a) for the whole international material, b) for crops fertilized with >30 kg P/ha, and c) for crops fertilized with <10 kg P/ha. For regression equations see Table 4 and for abbreviations the text for Fig. 6.

The potassium contents of plants were almost tenfold those of phosphorus. Even an average crop therefore takes more K from the soil than is usually applied in potassium fertilizers. Consequently, soil potassium reserves even in countries of traditionally high potassium fertilizer consumption have not been increased appreciably, if at all. According to FAO statistics (FAO, 1980a), among the 30 countries represented in this study, those with the highest potassium consumption per hectare in 1961-78 were Belgium, Hungary, Finland, Korea and Italy. The crops sampled from these countries for this study were also fertilized with high rates of potassium. The locations of these countries in the "international K field" (Fig. 8), however, do not show such high soil K contents as could be expected from their traditionally high potassium fertilizer consumption. On the other hand, the mean K contents of plants from countries where K had been generously applied to the sampled crops are usually above the main regression line (a). Without doubt, the greatly varying rates of recent K fertilizer application are one of the main reasons for the relatively poor plant Ksoil K correlation in the whole material (Table 4). 25

Aver. fert.application

o < 50 kg K /ha 50 - 100 kg K /ha o >100 kg K/ha

I I III 1 30 50 70 100 200 300 500700 1000 2000 (CH3 COONH4) Kinsoil, mg/I

Fig.8.Regressions of potassium content of original wheat and maize (pooled) on CH3COONH4 exchangeable potassium content of soil a) for the whole international material, b) for crops fertilized with >30 kg K/ha, and c) for crops fertilized with <10 kg K/ha. For regression equations see Table 4 and for abbreviations the text for Fig. 6.

Since the majority of crops sampled had received no or only a minimal (< 10 kg K/ha) dressing of K, the regression (line c) for low rates of applied K differs little from that for all samples (line a). Potassium fertilization at high rates considerably increased the K contents of plants as indicated by the regression line b. In general, soils in countries of high mean exchangeable K contents (e.g. Argentina, Egypt, Ethiopia, Tanzania, Mexico and Syria) are fine to medium textured (average texture index 40-62), while soils in countries of low exchangeable K (Nepal, Sierra Leone, Ghana, Sri Lanka and Zambia) are relatively coarse (mean TI 30-39). See also Fig. 12. In the last mentioned countries potassium fertilization is likely to be one of the primary requirements for obtaining reasonably high yields. Calcium is essential, not only to correct soil acidity but also as a nutrient element necessary for normal plant growth. Soil Ca plays an essential role in regulating soil pH. All countries shown in Fig. 9 where the average exchangeable Ca content of soils was less than 1000 mg/1 had a mean pH(CaC12) of 5.1 or lower, countries with a soil Ca 26

Ca .7

.6 e u Tu It .5 0_ o .4 _IIIflIjj .4- IIIIIII M.e r'jiI (t) NZe Et Eg o o.3 III e TD oo .2 I

600 1000 2000 3000 6000 10000 20000 (CH3COONH4) Ca in soil , mg/I Fig. 9. Regressionof calcium content of originalwheat and maize (pooled) on CH3COONH4 exhangeable calcium content of soil for the whole international material. For regression equationsee Table 4 and for abbreviations the text for Fig. 6.

content of 1000-1600 mg/1 had an average pH of 5.2-5.8, and those with soil Ca over 6500 mg/1 had a pH(CaC12) of 7.4 or higher. The relationship between Ca content of plants and soil pH is similar although not as clear (Fig. 11). Liming was most frequently practised in Korea and Finland where 9 out of every 10 sampled soils received lime in recent years preceding sampling. In Belgium, Brazil, New Zealand and Hungary liming frequencies (7, 6, 5 and 3 out of 10 sampled soils, respectively) were also high. Liming was uncommon in other countries. In countries of very low soil Ca status the low pH may also be accompanied by toxic or excess contents of elements such as Mn and Al. Owing to the general abundance of exchangeable Ca in soils in relation to Ca requirements of plants, the absorption of Ca by plants may rarely be directly limited by low soil Ca contents, and further, the Ca contents of soils are poorly reflected in the Ca contents of plants. It is apparent that true Ca deficiency is relatively rare and of less importance than the indirect malfunctions due to shortage of soil Ca. 27

.26-Mg

.24

6Ph .22 °Hu

20 -I eNe In C Eg oKo Q. .18 ,Tu Ni oIr o A .16 Sr e Pe oTa Le e oMw Ec Fi Za Gh° Et he Si

Bee Mt

.08 Lt2 50 70 100 200 300 500700 1000 20003000 ( CH3 COONI-14) Mg insoil ,mg/I

Fig.10.Regression of magnesium content of original wheat and maize (pooled) on CH3COONH4 exchangeable magnesium content of soil for the whole international material. For regression equation see Table 4 and for abbreviations the text for Fig. 6.

The correlation between plant magnesium and exchangeable soil Mg was only slightly better than that of Ca. Even though in almost every country there are a few soils with low contents of exchangeable Mg, their relative frequency is very low in countries such as Egypt, Syria, Philippines, Iraq, Mexico, Lebanon, Turkey and Hungary, where fine textured soils (national mean texture indexes 46-60) with medium to high pH(CaC12) (national mean 6.0-7.7) predominate. In typical low Mg countries such as Belgium, Sierra Leone and Zambia, soils with a low pH and coarse texture (national mean pH(CaC12) 4.9-5.8, TI 27-33) are in a majority. The internal variations within each country of plant and soil macronutrient contents are given in Part II. 28

Table 4. Regressions of pooled plant N, P, K, Ca and Mg contents (y) on respective soil macronutrients (x). For regression curves, see Figures 6-10.

S-4± s Tc s Regression r Fig. N, whole material 3732 4.27 ± 1.16 0.134 ± 0.086 y = 5.61 + 1.43log x 0.280***6 a N, fert. > 100 kg N/ha 1102 4.80 ± 1.11 0.131 ± 0.086 y = 6.47 + 1.78log x 0.334***6 b N, fert. 50 kg N/ha 1955 3.99 ± 1.05 0.138 ± 0.081 y = 5.11 + 1.22log x 0.274*** 6e

P, whole material 3732 0.375 ± 0.121 21.4 ± 29.4 y = 0.205 + 0.155 log x 0.556***7 a P, fert. > 30 kg P/ha 763 0.457 ±0.123 35.1 ± 32.7 y = 0.275 + 0.133 log x 0450***7 b P, fert. < 10 kg P/ha 2089 0.347 ± 0.106 17.2 ± 28.4 y = 0.212 + 0.134 log x 0.523***7e

K, whole material 3732 4.03 ± 1.12 346 ± 324 log y = 0.329 + 0.108 log x 0.304***8 a K, fert. > 30 kg K/ha 709 4.31 ± 1.01 204 ± 139 log y = 0.229 + 0.175 log x 0.398***8 b K, fert. < 10 kg K/ha 2810 3.97 ± 1.12 396 ± 352 log y = 0.286 + 0.120 log x 0.338***8 c

Ca, whole material 3732 0.428 ± 0.177 4027 ± 3004 log y =-0.663 + 0.075 log x 0.183***9

Mg, whole material 3732 0.173 ± 0.072 466 ± 451 log y = -1.028 + 0.093 log x 0.219***10

2.2.4 Macronutrient contents of plants and soils in relation to four soil characteristics

In order to obtain a broad assessment of factors influencing the macronutrient contents in this material, regressions of N, P, K, Ca and Mg contents of plants (pooled) and soils (total N, CH3COONH4 exchangeable Ca, Mg, and K, and NaHCO3 extractable P) were computed on four important soil characteristics. The regression lines are given in Figures 11-14 and the respective equations in Table 5.

Table 5. Plant and soil macronutrients as functions of four soil characteristics in the whole international material (n = 3731). For regression curves, see Figs 11-14.

Variables Regression Variables Regression Fig. y x y x

Plant N pH log y = 0.718 - 0.130 log x _0.074***Soil N pH y = 0.281 - 0.0221 x - 0.289*** 11 " TI log y = 0.485 + 0.0788 log x 0.099*** TI log y =- 1.76+ 0.510 log x 0.390*** 12 " Org.0 y = 4.24 + 0.932 log x 0.213*** Org.0log y =- 0.959 + 0.765 log x 0.890*** 13 CEC log y = 0.491 + 0.0881 log x 0.161*** CEC log y =- 1.60+ 0.484 log x 0.528*** 14

Plant P pH log y = - 0.371 - 0.0116 x - 0.090***Soil P pH log y = 1.59 - 0.0734 x - 0.190*** 11 " TI y = 0.412 - 0.000835 x - 0.113*** TI Y = 28.4 - 0.158 x - 0.089*** 12 " Org.0log y = -0.451 + 0.0824 log x 0.152*** Org.0log y = 1.08 + 0.631 log x 0.384***13 ., CEC y = 0.392 - 0.000617 x - 0.074*** CEC log y = 0.835 + 0.193 log x 0.110***14

Plant K pH y = 4.58 - 0.677 log x - 0.047** Soil K pH log y = 1.41 + 1.19 log x 0.256*** 11 TI log y = 0.509 + 0.0485 log x 0.064*** TI log y = 0.906 + 0.916 log x 0.432***12 Org.0log y = 0.585 + 0.0797 log x 0.163*** Org.0log y = 2.38 + 0.305 log x 0.220*** 13 CEC log y = 0.548 + 0.0287 log x 0.055*** CEC log y = 1.37 + 0.739 log x 0.501*** 14

Plant Ca pH y= 0.157 + 0.332 log x 0.146***Soil Ca pH log y = 0.065 + 4.14 log x 0.773***11 " TI y= 0.455 - 0.000604 x - 0.056*** TI log y = 1.43 + 1.25 log x 0.513***12 " Org.0 y= 0.439 - 0.00841 x - 0.059*** Org.0 Y = 4240 - 178 x - 0.073*** 13 CEC Y 0.438 - 0.000362 x - 0.030 n.s. CEC Y = 609 + 124 x 0.603***14

Plant Mg pH log y = -0.942 + 0.0221 x 0.149***Soil Mg pH log y = 0.471 + 2.49 log x 0.495*** 11 " TI y -= 0.176 - 0.000066 x - 0.015 n.s. TI log y = 0.364 + 1.32 log x 0.581***12 " Org.0log y = -0.792 - 0.0899 log x - 0.144*** Org.0 Y = 483 -15.1 x - 0.041* 13 CEC log y = -0.771 - 0.000638 x - 0.056*** CEC log y = 1.15 + 0.984 log x 0.620*** 14 29

10 .2 N (r.-c89., xxx)

7 ,

5 .0 _ N ( r =-.074 x") 4_ - K (r.-.047 m") 3 100

600 11'5 2 400 o Ca E200

a 100 U 600 ,04 1 o .7 -95 o 400 Mg l 4 XXX r. 146 cr, Ca / 200 (r- Z56xx 2 K p(r -.090= XXX) 100 60 o- 40 Mg (:-..149 xxxl - 20 (/)

190 .cxx) 10

4 5 6 7 4 5 6 8 pH (CaCl2) pH (CaCl2 ) Fig. 11. Plant and soil macronutrients as functions of soil pH in the whole international material (n = 3731). Regression equations are given in Table 5.

In respect to soil pH (Fig. 11) the firmest relationship was obtained for exchangeable soil Ca, which increased from about 400 to 6000 mg/1 when the pH increased from 4 to 81). The respective increase in plant Ca content was from about 0.35 to 0.45 %. If these increases (400 to 6000 and 0.35 to 0.45) aie compared to the regression given in Fig. 9, it can be seen that between soil Ca levels of 400 and 6000 mg/1 the respeçtive increase in the Ca content of plants is about from 0.34 to 0.42 %. The increase in plant Ca is therefore almost entirely due to an increase in exchangeable soil Ca, and pH seems to have practically no effect on absorption of Ca by the plants. The increasing soil Mg and decreasing soil N and P contents as functions of pH are accompanied by similar relationships between the Mg, N, and P contents of plants and the pH. Comparing these to the respective plant-soil regressions given in Figs 6, 7 and 10 (as in the case of Ca) it can be seen that the changes in the Mg, N, and P contents of plants are almost entirely due to the respective changes in the soil Mg, N and P contents, and the effect of pH on the absorption of these elements by plants is minimal. Potassium is the only macronutrient the availability of which to plants seems to be affected by soil pH. In spite of increasing exchangeable K contents of soils with increasing pH, the K contents of the plant decrease. While these relationships are statistically not as firm as those for the other macronutrients, they still point to pH as being one cause for the relatively poor correlation between the results of the plant and soil K analyses.

') Although soil Ca is given as a function of pH the cause and effect are likely to be the opposite. 30

.. 10 .2 r390' e7 5 N0-.099x") .0 4 o r = .064xxx) 3 K( 1000 600 2 400 Ca o o) I515.,----- E 200 oo 100...... II' , 60-r....-,_ Mg 40MIL 111.117 1 I= -/117--32 .5 - --- C r,._.056xxx) 2 20 D( r = -.773 )(xx) M .3 10 o 6 (AS o .2 =IIIII o Mg( r.-.015 n.s.) 4 2 02 r -089

1 I 1 10 30 50 70 90 10 30 50 70 90 Texture index Texture index Fig. 12. Plant and soil macronutrients as functions of soil texture (TI) in the whole international material (n 3731). Regression equations are given in Table 5.

The contents of soil total N and exchangeable Ca, Mg, and K increased significantly from coarse to fine textured soils but the NaHCO3 extractable P contents decreased (Fig. 12). Similar changes in the plant contents of N, P and K took place, indicating that texture has a relatively small effect on the availability of these elements to plants. In spite of the strong increases in the exchangeable soil Ca and Mg toward fine textured soils, there were no corresponding increases in the Ca and Mg contents of plants. Instead these showed a slight decrease, and indicate that plants can absorb these elements more easily from coarse than from fine textured soils. Clearly the varying relative availability of Ca and Mg in texturally different soils is one of the factors impairing the plant-soil correlations for these elements. The plant and soil macronutrient data given as functions of the organic carbon content of soil in Fig. 13 are relatively consistent with each other. The N, K and P contents of both the soil and the plant increased and those of Ca and Mg decreased with increasing organic carbon content of soils. The magnitudes of the changes are also of the same order as those given in Figs 6 to 10. For example, the total N content of soil increases from about 0.03 to 0.45 % and the N content of plants from about 3.6 to 5.0 % when the organic carbon content increases from 0.2 to 6.4 % (Fig. 13). Likewise, with the total soil N content increasing from 0.03 to 0.45 % (Fig. 6, line a) the plant N content increases from about 3.4 to 5.1 %. Thus, the increase in the plant N content is due to the respective increase in the soil total N content, and the varying soil organic carbon contents do not interfere with the plant Nsoil N correlation to any marked extent. The same is true for the relationships between the other macronutrients and soil organic carbon content. 31

10 .2 .....`'-a° NEW 7Il z- 1 LW al 3 x") 5 ( r- .0 NM moil/ )_ 4 ( y163x% Eii 1000 0_ r . 600 2 400 ..... o '2O0 =I E ..

1 100 Q) =MIMI o MI= (-) 60MIIIIIIMIIIIII MINI= o .7 EN =NE (r = - 041 x ) o 40Ell MIME C MIEN1111111111111=1111111/ .5 o 20I 11111111 - =.152xxx 1 10 7MN o o.2 CL 4ll o ("" - 144 xxx) P 2 Mail 2o .1 1 IMil illirIMEM111111111" IMg MIN MIIMINIIIIIIIIMEN minomr Ill MNIIMMIIMIIIIIIMIN 4=MN= =NM .4 .6 1.6 3.2 64 .2 .4 .8 16 3.2 64 Organic C , % Organic C ,

Fig. 13. Plant and soil macronutrients as functions of soil organic carbon content in the whole international material (n = 3731). Regression equations are given in Table 5.

The soil contents of all five macronutrients increased with increasing cation exchange capacity (Fig. 14). For N, Ca, Mg and K these relations are more firm than for P. Of the plant macronutrients, only the contents of N and K increased with increasing CEC but Ca, Mg and P decreased. The plant macronutrientCEC regressions are generally less firm than the respective relations between soil macronutrients and CEC. Because of the pronounced effects of texture and organic matter content on CEC (Figs 3 and 4), the effects of these factors on soil and plant macronutrients are indirectly reflected also in macronutrientCEC relations. The dissimilarity between the regressions of plant and soil Ca, Mg and P contents on CEC (Fig. 14) can be considered as one of the factors interfering with the plant-soil correlations for these elements. The above regression data of the plant and soil macronutrient contents on four soil characteristics are summarized in Table 6. With rising pH the N and P contents decreased but those of Ca and Mg increased. With increasing texture index the N and K contents increased but P contents decreased. With increasing soil organic matter content the N, P and K contents increased but those of Ca and Mg decreased. With increasing CEC the N and K contents increased. 32

)00,1 10 .20 .5228 N ' 7 .84-) .10

5 .05 =.161 xxx 4 N(r r =.055 xxx ) K( 1 3 MINN MNIE 600 0 ,Agilall 2 400MI= E 200 MillIPPINE 111.1 o 100mom C.) =I ..11 60 ..4111M -o MIll /NM o .7 , -. PP.'". 40MI= ..APP."- .5 Ca -.4 ____(r=-.030n.s. 220 )MilpPW pc r = -.074xxx 1 ))

6 cL [ .2 4) 0C-, Mg("-056x", 0 2) 1.0 ( r . 110 X" /

.1 1)

4 a 16 n 64 4 8 16 32 64 CEC , me/100g C EC , me /100g Fig. 14. Plant and soil macronutrients as functions of cation exchange capacity in the whole international material (n 3731). Regression equations are given in Table 5.

These relationships are supported by both the plant and the soil macronutrient analyses, and the soil characteristics specified are unlikely to have any marked adverse effect on the plant-soil correlations for the particular elements. In the cases of pHK, TICa, TIMg, CECP, CECCa and CECMg the soil characteristics have opposite effects on the plant and soil nutrient contents, and are likely to interfere with the plant-soil correlation for those nutrients. Taking these effects into consideration may help with the interpretation of the data from plant and (or) soil analyses.

Table 6. Directions (+ or symbols) of regressions of plant and soil macronutrient contents on four soil characteristics.

Soil Macronutrient characteristic N Ca Mg Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil pH(CaCl2) Texture index Organic C CEC 33 2.3 Micronutrients

2.3.1 Plant analysis versus soil analysis

Opinions of scientists differ as to the relative value of plant and soil analysis. In the present study both techniques were used on all soils and the evaluation of results is based on both (see Section 1.4.3). The analysis of over 3500 original pland from the field and the respective soils, plus over 3500 pot plants grown on the same soils, gave an opportunity to compare the data and to draw certain conclusions at theoretical and practical levels. To understand better the characteristics of these two widely used techniques, plant analysis and soil analysis, for diagnosing the micronutrient status of soils, one must realize that these analyses are based on fundamentally different principles. The most essential difference between these analyses is that the micronutrient fractions to be analysed are obtained by different ways. Micronutrient absorption by a plant is a process taking place under laws of biochemistry and plant physiology while chemical soil extraction obeys the laws of chemistry. Consequently, the same micronutrient fractions do not appear in the analysis, nor is it always the aim that they should. Therefore, it is understandable that many soil factors react differently during these two processes. Furthermore, the amounts (total) of micronutrients found in plants represent micronutrient fractions in the soil which have been available to the plant during its period of growth. Depending on the analytical method and on the micronutrient analysed, the results of soil analyses also include varying amounts of soil micronutrient reserves. The total contents of soil micronutrients, even though having an influence on the soluble or on plant available amounts, are in general, poor estimates of the available fractions. A variety of extraction methods has therefore been developed to obtain more reliable estimates of the fractions available to plants. Even the extractable fractions, although representing only a small portion of the total, exceed the requirements of plants by a considerable margin. From a theoretical point of view, the amounts of micronutrients absorbed by plants from soils are to be considered as the most reliable measure of the fraction available to plants since the process of absorption is one taking place under the laws of nature. Soil analysis again is to be considered as an attempt to imitate plants. If there is contradiction between the results there should be no doubt who is wrong, the plant or the soil chemist. If both analyses give similar estimates of the micronutrient status of soils, i.e. the mutual correlation between the results of plant and soil analyses is good, the quantitative differences between the micronutrient fractions analysed from plants and extracted from soils are of little importance. When there is contradiction between the results obtained by the two techniques, it is important to find the main reason(s) for it, i.e. which soil factor impairs the correlation between the results of soil and plant analyses by affecting the results of plant analyses in one way and those of soil analyses in another way. Such a contradiction indicates that the soil extraction method used fails to take into account the effect of a soil factor regulating the availability of a micronutrient to plant. In order to improve the method of soil analysis, its results must be corrected so that they are in accord with the results of plant analyses with regard to the effects of the soil factor in question. In other words, a 34 correction coefficient is required for eliniinating or minimizing the deviating effects of the particular soil factor(s). The prerequisite for introducing a correction coefficient is that both the plant material and the soil material meet certain requirements: large numbers (preferably thousands) of plant-soil sample pairs widely varying soil characteristics') comparable plant material consisting of one indicator plant species, the same variety, same sampling time (physiological age), same parts of plant analysed, minimum contamination, and comparable growing conditions2). The present material (soils and pot-grown plants) can be considered to meet these requirements rather well. Differences between the absorption of most microelements by a plant and the extraction of that microelement from the soil by chemical treatment are caused by more than one soil factor, but in practice all these cannot be taken into account. Therefore, in this study only the factors with the greatest effects (the key factors) are used as a basis for correcting the results of soil analyses. Examples of the effects of two correction factors (pH and TI), however, are given in case of Mo (Section 2.3.2) and an example of correcting the results of two extraction methods (DTPA and AAAc-EDTA) for one soil factor (pH) is given in the case of Mn (Section 2.3.6). Identification of the soil factor(s) responsible for impaired correlation between the results of plant and soil analyses was carried out by computing regressions for both the results of plant analyses and the results of soil analyses as functions of all six soil characteristics determined for all the international material. Comparing the regressions of plant and soil micronutrient contents as functions of various soil factors, it was usually possible to identify visually the factors with the most impairing effects on the plant-soil correlation. In verification, the actual impairing effects of all soil factors involved were computed, quantified and compared.3) According to these data the soil factors impairing the plant- soil correlation most severely appeared to be as follows:

Micronutrient Soil factor* Soil extraction method CEC Hot water Cu Organic carbon AAAc-EDTA content Mn pH AAAc-EDTA and DTPA Mo ' pH A0-0A Zn pH AAAc-EDTA * In the cases of AAAc-EDTA exractable soil Fe and DTPA extractable soil Zn, none of the correction coefficients calculated for the six soil factors improved the respective plant-soil correlation appreciably.

Identification of soil factors responsible for impaired correlation between the results of plant and soil analyses is the first step toward a more systematic interpretation of analytical results. These factors must be taken into account in one way or another when interpreting the results of soil analyses.

For example, it would not be justifiable to make general conclusions about the effect of pH, if the range of variation in pH of the soil samples were one or two pH units only. For details, see Section 1.2. In the next Section (2.3.2) the data concerning Mo are presented in detail to give a fuller account of the principles and procedures for quantifying the impairing effects of various soil factors, for identifying the soil factor with the greatest effects and for eliminating these effects by determining and applying a correction coefficient. 35 For example, in case of Mo the soil pH is one such factor. The results of soil Mo analyses must be either: interpreted differently at different soil pH levels, or the interfering effects of pH must be eliminated by making the effect of pH on soil chemical extraction equal to its effect on Mo absorption by plants. Should course (a) be adopted, a number of parallel stepwise interpretations would be needed. Thus, for example, extractable soil Mo contents obtained from a soil of pH 4 would have to be interpreted quite differently to the contents measured from soils with a pH of say 5,6 or 7. In order to simplify the interpretation of soil analytical data, the second course of action (b) was adopted and applied to the results of this study. Accordingly, the interference of soil pH was eliminated by introducing a "correction coefficient" for pH to the results of the soil Mo analyses so that the extractable soil Mo values as a function of pH became similar to plant Mo values as a function of pH. The introduction of this coefficient is not aimed to eliminate the influence of a given soil factor (in the case of Mo, the pH) but to liberate soil Mo values from an interfering factor (pH) and to minimize its impairing effects on the plant Mosoil Mo correlation. With the application of a pH correction to soil Mo values, the need for different parallel interpretations at different soil pH levels vanishes. As pointed out above, soil factors exercising the greatest interference with the plant-soil correlation vary from one micronutrient to another depending also on the method of extraction. In the next Sections (2.3.2 to 2.3.7) the effects of six soil factors on the six micronutrients are discussed. Departing from the otherwise alphabetic order of the elements, the data concerning Mo is given first and in more detail than those of other micronutrients. The underlying reasons for so doing were that the soil Mo analysis needs to be corrected more than those of other micronutrients and also because in the case of Mo two correction coefficients (for pH and for texture) are introduced.

2.3.2 Molybdenum

2.3.2.1General aspects The average Mo contents of the three types of indicator plant, the A0-0A extractable Mo contents of the respective soils, and the best fitting regressions of plant Mo and soil Mo for all the international material are shown below. The corresponding national averages are given in Appendixes 2, 3 and 4.

Molybdenumcontent in plant DM in resp. soils Indicator mean ± s mean + s Regression of plant Correlation plant n (PPrn) (mg/1) Mo (y) on soil Mo (x) (r) Original maize 1966 0.86 ± 1.35 0.212 ± 0.273 y = 0.715 + 0.661 x 0.134*** Original wheat 1766 0.94 ± 1.03 0.204 ± 0.229 y -= 0.836 + 0.525 x 0.117*** Pot-grown wheat 3537 0.32 ± 0.36 0.210 ± 0.257 y = 0.529 + 0.246 x 0.245***

The differences in the average Mo contents of soils between the three groups are small. Mo differed from the other micronutrients in that the mean Mo content of maize was 36 lower than that of the original wheat. This may not so much be due to the lesser ability of maize to absorb Mo from soils as to the generally lower pH of maize soils (mean pH 6.40) than of soils where the wheats were originally grown (mean pH 6.91). For much the same reason, the Mo contents of the original wheats exceeded those of the pot-grown wheats (mean soil pH 6.64) by a considerable margin. The effect of soil pHon extractable Mo and Mo content of plants will be discussed later in this chapter. The small amounts of soil in pots (see Section 1.2.6) may have further lowered the Mo contents of the pot- grown wheat as compared to the contents of the original wheat plants. The regression of Mo contents of pot-grown wheat on soil Mo is shown in Fig. 15, where the national mean values of plant and soil Mo are also plotted. Therewas a substantial improvement in the correlation when the original plantswere replaced by the pot-grown wheat plants. In spite of this, the correlation (r = 0.245***) was weaker than those for the other five micronutrients.0 Nor was the respective correlation between the Mo uptake (pot-grown wheat) and soil Mo (y = 0.711 + 0.341 log x;r = 0.243***) any better than that in Fig. 15. It is clear that soil Mo analysis (ammonium oxalate-oxalic acid, pH 3.3) as such does not give a sufficiently reliable index of Mo availability to p1ants.2) For instance, the average extractable Mo contents of Pakistani and Brazilian soils were approximately equal, but the Mo contents of plants from Pakistan were about 20 times higher. Studies concerning the reasons for the contradictory results of soil and plant analyses are presented in Section 2.3.2.2.

2.3.2.2 Soil factors affecting the molybdenum contents of plants and soils To obtain first a general picture of how various soil factors are related to the amounts of Mo absorbed by plants (Mo contents of pot-grown wheat) and to A0-0A extractable soil Mo, both the plant Mo and soil Mo contents were computed as functions of the six soil characteristics (pH, texture, organic carbon content, CEC, electrical conductivity and CaCO3 equivalent) determined from all the soil samples. These regressionsare presented graphically in Fig. 16 (graphs on the left and in the centre), and the respective regression equations and correlation coefficients are given in Table 7. To give a better visual picture of these relations the soil data were classified into 11 to 18 classes (columns) with respect to each soil characteristic.3) The number of samples falling into each class is given. Visual comparisons of the regressions show that the effect of a soil characteristicon plant Mo usually differs considerably from its effect on A0-0A extractable soil Mo. Only the effects of CEC on plant Mo and soil Mo seem to be in conformity; the other soil factors affect the absorption of Mo by plants to an extent considerably different to their effect on the extractability of soil Mo by A0-0A extractant. These differences account for the conflicting results between the soil and plant analyses. A0-0A extraction is thus insensitive to the effects of those soil factors on the availability of Mo to plants and, therefore, gives a poor estimate of Mo availability to plants. In fact, three of the soil factors studied (electrical conductivity, pH and CaCO3 equivalent) are better correlated

Because of the very large number of samples (n = 3535 3538) used in this study, correlation coefficients of 0.033, 0.043 and 0,055 would have been significant at the 5 %, 1 % and 0.1 % level, respectively.

At an early stage of this study other extraction methods for soil Mo were examined on limited materials, but none of these proved promising enough for acceptance. The reason for the low location of some regression curves (e.g. curve a) in relation to the columns is that the columns indicate arithmetic mean values and the regression is in a logarithmic form. 37

y -±s02 ± 3 = 0 529 +2 2 r=0 45" E 0. o_

_J . 'a .6 *I r It Pe

S F. L TuHuPh

"C"o

.06

.03 .02

.011

.01 .02 .03 .06 .1 .2 .3 .6 1 Mo in soil, mg/I.

Fig. 15. Regression of Mo content of pot-grown wheat (y) on ammonium oxalate-oxalic acid extractable soil Mo (x) for the whole international material. Abbreviations for national mean values:

Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le = Lebanon, Mt = Malta, Malawi, Me = Mexico, Ne = Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta = Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia. to Mo content of plants (r values 0.498***, 0459*** and 0.410***, respectively, Table 7) than is the A0-0A extractable soil Mo (r = 0.245***, Fig. 15). Also the other three soil factors studied (texture, organic carbon content and CEC) seem to have significant effects on the Mo content of plants (Fig. 16, regressions f,i and 1). Furthermore, all six soil factors studied affect significantly the A0-0A extractable contents of soil Mo (Fig. 16, regressions a, e, h, k, n and q). 38

2.0

1.0

5 6 7 pH (Ca C12) pH (0x012)

1.71 E

E Q. 10 30 o 30 50 70 90 Texture index (Ti) Texture 2 Texturei idex (Ti)

o .4 .8 t6 az 64 al 4 .8 16 32 6.4 .4 .8 16 32 64 OrganicC, % Organic C, % Organic C, % -8

+E. a.) o o 8 16 32 64 o 8 16 32 64 8 16 32 64 CECme/ 00g CEO me/ 009 o CEC ,me/100g

16 6.4 256 .4 16 64 zas .4 16 64 256 El.conductivity, , 10-4 Scm El.conductivi y , 10-4 Scm El.conductivi y , 10-4 Scm S

,4 16 64 256 . .4 16 64256 Ca CO equiv., % Ca CO equiv., % Co CO, equiv., % Fig. 16. Relationships of ammonium oxalate-oxalic acid extractable soil Mo and plant Mo to various soil characteristics in the whole international material. Graphs on the left: A0-0A extractable soil Mo as a function of six soil factors. Graphs in the middle: Plant Mo as a function of six soil factors. Graphs on the right: pH-corrected A0-0A extractable soil Mo as a function of six soil factors. In regard to the six soil characteristics the material is classified into 11-18 -classes. The columns indicate arithmetic mean Mo values of each class. The number of samples (n) within each class is given. The equations and correlation coefficients for regressions are given in Table 7. See also Footnote 3 on p. 36. 39

Table 7. Equations and correlation coefficients for regressions of A0-0A extractable soil Mo (uncorrected and pH-corrected) and plant Mo on six soil characteristics. Regression curves are given in Figs 16 and 17.

Variables Regression Correl. Regr. Y coeff. curve Soil Mo (uncorr.) pH 3537 log y = -6.741 + 16.40 log x - 11.15 (log x)2 .306*** a Soil Mo (uncorr.) pH 18 y = -8.558 + 4.429 x - 0.722 x2 + 0.0383 x3 .796*** al Plant Mo pH 3537 y =0.574 - 0.251 x + 0.0312 x2 .459*** Plant Mo/soil Mo pH 3537 log y = -2.300 + 0.360 x .770*** pH-corr. soil Mo pH 3537 y =0.909 - 0.336 x + 0.0338 x2 397***

Soil Mo (uncorr.) TI 3535 y = -0.274 + 0.299 log x .199*** e Plant Mo TI 3535 y = -1.559 + 2.344 log x - 0.722 (log x)2 075*** pH-corr. soil Mo TI 3535 y = -1.328 + 1.642 log x - 0.423 (log x)2 .195***

Soil Mo (uncorr.) Org. C 3537 y =0.102 + 0.0914 x - 0.00353 x2 .255*** Plant Mo Org. C 3537 y =0.380 - 0.0483 x + 0.00157 x2 .098*** pH-corr. soil Mo Org. C 3537 y =0.244 - 0.0280 x + 0.000753 x2 .078***

Soil Mo (uncorr.) CEC 3537 y = -0.390 + 0.738 log x - 0.213 (log x)2 .177*** Plant Mo CEC 3537 y = -0.202 + 0.809 log x - 0.302 (log x)2 .066*** pH-corr. soil Mo CEC 3537 y = -0.251 + 0.584 log x - 0.175 (log x)2 .115***

Soil Mo (uncorr.) El. cond. 3537 y =0.219 - 0.135 log x + 0.121 (log x)2 .122*** Plant Mo El. cond. 3537 log y = -0.856 + 0.784 log x - 0.213 (log x)2 .498*** o pH-corr. soil Mo El. cond. 3537 log y = -1.043 + 0.776 log x - 0.217 (log x)2 .498***

Soil Mo (uncorr.) CaCO3 eq. 3537 log y = -0.844 - 0.107 log x - 0.0254 (log x)2 307*** Plant Mo CaCO3 eq. 3537 y =0.406 + 0.0910 log x - 0.0243 (log x)2 .410*** pH-corr. soil Mo CaCO3 eq. 3537 log y = -0.700 + 0.118 log x - 0.0820 (log x)2 .580***

In this connection it should be remembered that although all six soil characteristics studied seem to affect significantly both the plant Mo and A0-0A extractable soil Mo, some of these correlations are only "secondary" or "pseudocorrelations" due to the significant (r values from 0.093*** to 0.903***) mutual correlations between the six soil characteristics (see Section 2.1). Some of these are referred to later in connection with Mo as well as other micronutrients. One of the principal differences between the effects of the above soil characteristics and extractable soil Mo on plant Mo seems to be that these soil characteristics act as regulators for the availability of Mo to plants, while extractable soil Mo values are more influenced by the Mo reserves of soil but fail to take into account the effects of the factors regulating the availability of Mo to plants. Therefore, a combination of one or more of the above soil factors and A0-0A extractable Mo would give a better estimate of the *amounts of Mo available to plants than any of them alone. So far the twelve regressions of A0-0A extractable soil Mo (Fig. 16 a, e, h, k, n, q) and plant Mo (b, f, i,. 1, o, r) on the six soil characteristics discussed above, indicate how soil Mo and plant Mo are correlated to these factors, but only broadly quantify the interference of these factors with the plant Mo-soil Mo correlation. To quantify the interfering effect of a soil factor its effect on A0-0A extractable soil Mo must first be equalized to its effect on plant Mo, i.e. a correction coefficient for soil Mo values is required. Applying this coefficient to all the international material gives a measure of the interfering effect of the soil factor in question in the form of a new correlation coefficient. The more the plant-soil Mo correlation is improved the greater the interference of the factor in question has been. Furthermore, by comparing the effects 40

tO E

.6 o

.4 "E. o o .2 o

Soil pH

5 6 7 8 4 5 6 Soil pH Soil pH Fig. 17. Relationships of A0-0A extractable soil Mo and plant Mo to soil pH(CaC12). Graph a: Soil Mo as a function of pH. Regression curve a is calculated for the whole material (n = 3537) and curve al for mean values of pH classes (see text). Graph b: Plant Mo as a function of pH. Graph c: Ratio of plant Mo to soil Mo as a function of pH. The regression curve also indicates the coefficient, k(pH), for correcting soil Mo for pH (right-hand ordinate). Graph d: pH-corrected soil Mo as a function of pH. Columns and points indicate mean values of each pH class, n values of pH classes are given in Graph a. Equations and correlation coefficients for the regression curves are given in Table 7. of correction coefficients calculated for different soil factors, the factor which has most impaired the correlation can be identified. A concrete example of the procedures involved is given in the following: The effects of soil pH on A0-0A extractable soil Mo are also given in Fig. 17 a. A large increase in soil Mo up to about pH 5-6 was first followed by a decrease, reaching a minimum at about pH 7.5-7.7, and then by another increase.r) The Mo contents of plants increased strongly toward alkaline soils (Fig. 17 b). This general tendency, however, was interrupted at about pH 7.5-7.7, i.e. at the very pH level where soil Mo values reached a minimum. In view of the large number of samples at this pH level, the connection between these minima is scarcely a coincidence but possibly due to soil chemical reactions affecting both the extractability of soil Mo and the availability of soil Mo to plants.

I)Because of the abnormal distribution of samples into pH classes, the regression curve (a) calculated for the whole material ignores the latter increase. Therefore, another regression curve (a1), based on mean values of pH classes, is also given. 41 The ratio of plant Mo/soil Mo plotted as a function of pH in Fig. 17 c, shows that there is a very close relationship between the plant Mo/soil Mo ratio and soil pH (r = 0.770***). It should be noted that this regression (curve c) was calculated from all the sample material, 3537 sample pairs. The regression line (c) can be used to read off the correction of soil Mo values needed at different pH levels to eliminate the difference between pH effects on soil Mo and plant Mo analyses. However, if the regression equation of curve c (modified to y = 10-2.30 + 0.36 pH) as such is used to correct A0-0A extractable soil Mo values, it would raise the soil Mo values from their original average level of 0.210 mg/1 to an average of 0.297 mg/l. There- fore, to avoid this and also to restore the original average level for the corrected values, a reversion coefficient 0.210(0.297 = 0.707) has to be introduced into the equation. The pH correction coefficient, k(pH), is now given by: k(pH) = 0.707 X 10-2.30 + 0.36 pH which in a simplified form is: k(pH) = 10-2.45 + 0.36 pH

The numerical values for k(pH) can also be read off directly from the right hand ordinate of the regression curve c in Fig. 17. The k(pH) has been calculated for pH (CaC12). The relation between pH(H20) and pH(CaC12) in this material is given in Fig. 5 (Section 2.1).

When the soil Mo values in the whole international material are corrected with k(pH), the soil MopH relation given in graph d is obtained. The latter is very similar to the plant MopH relation given in Fig. 17, graph b, i.e. the effect of pH on soil Mo has not been eliminated but it has been made equal to its effect on plant Mo. The plant Mosoil MopH relations are further illustrated in Fig. 18, where the data given in graphs a and b of Fig. 17 are combined. The respective relations after pH correction (combining data of graphs d and b) are given in Fig. 19. The S-shaped line depicting plant Mo and soil Mo in pH classes (Fig. 18) which transversely crosses the regression line for the whole material, illustrates how severely the effect of pH impairs the plant Mosoil Mo correlation. After pH-correction (Fig. 19) the new line is in good agreement with the calculated regression line. Application of k(pH) to the whole international material raised the correlation between plant Mo and soil Mo from r = 0.245*** (uncorrected, Fig. 15) to r = 0.696***. The latter regression is given in Fig. 20 with national mean values. After correction for pH, correlations between plant Mo and soil Mo within different countries were considerably higher than without correction (Table 8). The only exception was Malta, where the variation of soil pH was only from 7.48 to 7.64. The largest improvements in correlations were obtained in countries where the variation of soil pH was relatively wide. In five countries, non-significant correlations became significant: in Ghana r = 0.134 n.s. to r = 0.643***, Malawi 0.062 n.s. to 0.487***, Sierra Leone 0.040 n.s. to 0.586***, Sri Lanka 0.161 n.s. to 0.650** and Zambia 0.004 n.s. to 0.656***. The effects of the other five soil factors on soil Mo and plant Mo were investigated in a way analogous to that of pH. Correction coefficients for each of the five soil factors were 42

Fig.18. Relationship of plant Mo to M o n=c.: A0-0A extractable soil Mo in the 0.322± 136 whole material classified for pH. The 0.210 ±25 3 points indicate mean plant Mo and 110.2,61,1 soil Mo contents of various pH classes 2 inFig. 17banda.The lower E cL limit of each pH class is given. 0.1 37. MAI

MWMEMMINMEINNEtWar,64111.1-40Iy NWIMMEMENIMi."=11=11o111 .06111114111/fMIONIIIMMZIMM .03 .02

IIE .01 1116111, .01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/l (AO -0A)

12UNEMONIMMOSEMII11111111=L.==LIIMIEN11111111MOIMIONOMIIIIMIEMEMIIIMI Fig. 19. Relationships of plant Mo M o IMMOIMIIIRIMMTIMIIMMIME131111 1111 6 to pH-corrected soil Mo (A0-0A) MENNINIMEMAIIMMI111110111111111111111111MIN in the whole material classified for pH. See Fig. 17 b and d. 3 .111111=ZUMMIIIMMI11111

E o_ 1 IllaragliP1111 ifï w .6

'6..3 w, .2 UIIIIIIIPiIIhiiiH o o NorININEli o '1 1Milrr=1-16[4W.011M1210=MMENNI1MI .06 MMMENNOINMM=MMENII

.03 .02

.01 11111.11111111111.1

.006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH - corrected MOin soil, mg/l (A0-0A) calculated and their effects were tested and compared by applying them to the whole international sample material. These comparisons show that application of the various correction coefficients improved the plant Mo-soil Mo correlation from an r value of 0.245*** (uncorrected) to the following r values: 43

MN 13MNSIONSUMENNIMMIMMIIMMIIMIMIESIMMIENIMMINEMINEMIII NM ninnInnaninininnu=1Mlin M oMIMMIIIIIMIIIMENIMEMEMMOI11101MMINNIn Namminommirarmin SIIIMIIMMINIEMENI 6MEMMIII=1511ZEREB=IMMINIIIIIMMIIIIIIMIll =ME=MEMELs oBEENE3111 MUM 3IIIIII MIIIIIIMOZEILISIIIII NMI 21111111.1111111C5.9711Mill ME 11111 11111111.11111111P111-- IN-1111 11, 11 Mill11111111M1 WEI Ellini111011 1111101 o '1...... flIMMEIBMOIMIEUMMEmm...... En === .06

.03 .021111111111111111 iii 11

.01MililliNINiiiiiILI .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MO in soil,mg/t (A0-0.4)

Fig. 20. Regression of Mo content of the_pot-grown wheat (y) on the pH-corrected A0-0A extractable soil Mo (x) for the whole international material. National mean values are plotted in the graph. For abbreviations see Fig. 15.

Soil Mo corrected for Value of r pH 0.696*** Texture (TI) 0.277*** Organic carbon content 0.308*** CEC 0.260*** Electrical conductivity 0.460*** CaCO3 equivalent 0.620***

Correction for pH proved to be more effective than correction for any other soil factor involved, although those for CaCO3 and electrical conductivity also led to considerable improvements in the plant Mosoil Mo correlation. These, however, are clearly due to high mutual correlations between pH and the two soil characteristics in question (see Section 2.1). This applies partly to the other soil factors as well. Therefore, the effects of pH correction on the relations between soil Mo and the five other soil factors were further investigated. These results, given in Fig. 16 (the right hand graphs), show the following: 44 Soil Texture. A0-0A extractable soil Mo contents increased quite substantially from coarse toward fine textured soils (Fig.16e). The ability of plants to absorb Mo reached a maximum in medium textured soils and a further increase in texture index (TI) resulted in lowered Mo contents of plants (Fig.16f). Evidently, Mo is fixed in heavy textured clay soils so firmly that its availability to plants is limited. This fixation, however, does not seem to affect the extractability of Mo by a chemical extractant. These correlations (Table 7, e and f) are less close than those between Mo and pH (a and b). Correction of soil Mo values for pH brought the soil Motexture relation slightly closer in form to the respective relation between plant Mo and texture (curves g and 0 but their essential differences still remain. As the organic carbon content of soils increased, the Mo contents of plants decreased and there was a tendency for A0-0A extractable soil Mo to increase (Fig.16,h and i). When soil Mo values were corrected for pH, they also became corrected for organic carbon, i.e. the effect of organic carbon on pH-corrected soil Mo (curve j) is very similar to its effect on plant Mo (curve i). Cation exchange capacity. Both the A0-0A extractable soil Mo and Mo content of plants increased initially with increasing CEC (Fig.16, kand 1). With a further increment in CEC, there was a tendency for a decrease which in the case of plant Mo was very pronounce.d. Correction for pH also corrected slightly the soil Mo values for CEC. Electrical conductivity and CaCO3 equivaient of soils affected A0-0A extractable soil Mo values and Mo contents of plants quite differently (Fig. 16, no and qr). After pH correction, however, these discrepancies were almost eliminated (op and rs) because of strong mutual correlations between pH and electrical conductivity and pH and CaCO3 equivalent (see Figs 3 and 4). As shown above, due to mutual correlations between pH and the other five soil factors, the correction of A0-0A extractable soil Mo values for pH also reduced to varying degrees the discrepancies among the relationships of soil Mo and plant Mo to these soil factors. To quantify the remaining interfering effects of various soil factors all the procedures explained above were repeated, this time with soil Mo values already corrected for pH. They showed that texture was the factor now causing most interference with the plant Mosoil Mo correlation. The calculation of the second correction coefficient, that for texture k(TI), has been presented earlier in detail (Sillanpää,1981).The formula for the texture correction coefficient is given by:

k(TI) = 101.472 0.899 log TI

Application of the texture correction (in addition to the pH correction) to all the international material raised the correlation between plant Mo and soil Mo from an r value of0.696*** (Fig. 20)to 0.739*** (Fig. 21). A visual assessment of the efficacy of the pH and texture corrections can be made by comparing Figs15, 20and21.Application of the texture correction at the national level gave the highest plant Mosoil Mo correlations in20out of29countries (Table 8). Quantitatively, the improvement is not so marked as that of the pH correction, but substantial enough to be taken into account, especially in cases where soil materials with wide textural variation are to be analysed and accurate information on the Mo status of the soil is required. 45

NimmilmowiliklummaMN .....rommil wwwsimmenemelmollommom=1.0. =Imo i 0mini MOM MUM mmEINEHNIII111111111111111111111111111= 11111111111111111111111111111111MBH 11111111111111111111111111MINIMMINI 6IIMMOMMI=MMUNIMMEMENIMMENNII=.1. MEMEMEI 11111111111111111 R..uiii MIME MN NEE 3HEIM= Miini1 gy71g005+ 0.79 I o x 111E-' ' -"' 2 11111öt "It Ill mmo.NE= IN 111411111 MS ,P4tr 1.11 -I'm g SIEMEMiii , Be Il HuFj *Et Le. ; u h. Ko_, IL VI Ar Sr Lk. .z. Val MMMI , in .06 ... Elmin .03 Br. .02Mr

li .01 Nomm I

.006 . .02 .03 .06 .1 .2.3 .6 1 2 pH-texture -corrected Mo in soil, mg/l (A0-0A) Fig. 21. Regression of Mo content of the pot-grown wheat (y) on the pH- and texture-corrected soil Mo (x) for the whole international material. National mean values are also given. For abbreviations see Fig. 15.

Table 8. Correlations (r values) between plant Mo and A0-0A extractable soil Mo in various countries. Soil Mo analyses are both uncorrected and corrected for pH and texture. Correction coefficients are: k(pH) = 10-2.45 + 0.36 pH and k(TI) =- 101.472 - 0.899 log TI. Correlations of the best fitting regressions are given. Regression models: (a) y = a + bx; (b) y = a + b log x; (c) log y = a + bx, and (d)log y = a + b log x.

No correctionpH correction pH + TI No correctionpH correction pH + TI Country n only correction Country n only correction regr. r regr. r regr. r regr. r regr. r regr.r Belgium 36 c -0.207 n.s. c 0.036 n.s. d 0.217 n.s. Egypt 198 a 0.544*** a 0.614*** b 0.523*** Finland 90 d 0.341*** a 0.638*** a 0.743*** Iraq 150 b 0.778*** b0.817*** d 0.800*** Hungary 201 b 0.172* d 0.642*** d 0.664*** Lebanon 16 a 0.637** a 0.779*** a 0.819*** Italy 170 a 0.418*** d 0.565*** d 0.539*** Syria 38 d 0.706*** d0.742*** d 0.749*** Malta 25 c 0.914*** c 0.906*** c 0.827*** Turkey 298 a 0.664*** a 0.751*** a 0.727*** N. Zealand 35 a -0.121 n.s. b 0.171 n.s. b 0.240 n.s. Ethiopia 125 a 0.480*** a 0.908*** a 0949*** Argentina 208 a 0.583*** a 0.790*** a 0.709*** Ghana 93 a 0.134 n.s. a 0.643*** a 0.601*** Brazil 58 c -0.078 n.s. d 0.117 n.s. c 0.287* Malawi 97 c 0.062 n.s. a 0.487*** a 0.597*** Mexico 242 b 0.252*** d 0.583*** d 0.632*** Nigeria 153 b 0.088 n.s. d0.645*** d 0.711*** Peru 68 a 0.484*** a 0.692*** d 0.651*** Sierra L. 48 b 0.040 n.s. a 0.586*** a 0.668*** Tanzania163 d 0.545*** d 0.765*** d 0.846*** India 258 b 0.514*** b0.613*** d0.685*** Zambia 44 b -0.004 n.s. a 0.656*** a 0.717*** Korea, Rep. 90 b 0.280** d0.623*** d 0.694*** Nepal 35 a 0.335* d0.749*** b 0.794*** Pakistan 237 d 0.598*** d0.618*** d 0.587*** Philippines194 a 0.265*** d0.662*** a 0.697*** Sri Lanka 18 c -0.161 n.s. a0.650** a 0.694** Whole Thailand 149 b 0.388*** d 0.671*** d 0.677*** material 3537 b 0.245*** d 0.696*** d0.739*** 46 The mutual relations between plant Mo and A0-0A extractable soil Mo, and the effects of various soil factors on plant and soil Mo are summarized as follows: Owing to the heterogeneity of plant material, the original plants (maize and wheat) were unreliable indicators of the Mo status of soils. The heterogeneity was minimized by growing fresh plant samples (wheat) in pots on soils received from participating countries. Replacing the original plants by pot-grown wheat improved the plant MoA0-0A extractable soil Mo correlations from r values of 0.117-0.134*** to 0.245***. Soil Mo analysis (ammonium oxalate-oxalic acid, pH 3.3) as such does not give a reliable index of the Mo availability to plants as the data are much affected by soil pH. Both plant Mo and A0-0A extractable soil Mo were significantly affected by all six soil characteristics studied. Soil pH proved to be the factor which most disturbed the plant Mo and A0-0A extractable soil Mo correlation. Equalizing the effects of pH on soil Mo and plant Mo by correcting soil Mo values for pH raised the overall plant Mosoil Mo correlation (r) from 0.245*** to 0.696*** and individually in 28 out of 29 countries. the pH correction also moderated and sometimes almost eliminated the differences between the effects of the other five soil factors on plant Mo and soil Mo. After correcting for pH, differences in soil texture had the greatest effect on plant Mo and soil Mo correlation. Texture correction (in addition to pH correction) improved the plant Mosoil Mo correlation from an r value of 0.696*** to 0.739***, and in 20 out of 29 countries. When the results of this study are expressed by countries (Part II), the uncorrected as well as the pH-corrected A0-0A extractable soil Mo contents are presented.

2.3.3 Boron

2.3.3.1 General aspects The national average B contents of original indicator plants, those of pot-grown wheat, and the average hot water extractable B contents of the respective soPls are given in Appendixes 2, 3 and 4. Corresponding data for the whole of the international material as well as plant/soil regressions are tabulated below:

Boron content Indicator in plant DM in resp. soils Regression plant mean + s mean ± s of plant B (y) Correlation n (PPm) (mg/1) on soil B (x) (r)

Original maize 1966 9.24 ± 8.00 .65 ± .71 y = 5.25 + 6.15x 0.548*** Original wheat 1768 6.56 ± 7.80 .81 ± .75 y = 0.68 + 7.23x 0.694*** Pot-grown wheat 3538 6.09 ± 4.80 .73 ± .75 y = 2.63 + 4.75x 0.741***

On average, the B contents of maize were almost 50 per cent higher than those of wheat. The difference between the original and pot-grown wheat was small. This may be partly due to the small difference in soil B contents, and also to the small amount of soil in pots partly filled with quartz sand (see Section 1.2.6). The variation in B values was wide. 47 The standard deviations (s) of both plant and soil B are of the same magnitude as the respective mean values. Because of the factors causing uncontrolled variation in analytical results from the original plants (see Section 1.2), the correlation between plant B and soil B was lower for the original plants than for pot-grown plants. The difference, however, was small compared with differences for other micronutrients. As in the case of other micronutrients, better quality information on the behaviour of B was obtained from pot-grown wheat. The regression line of B content of pot-grown wheat on soil B for all the international material is given in Fig. 22. The correlation is highly significant (r = 0.741***). The

100 1111111MINIIIIIMIIIIIMMIMIIMMIIIIIIIIIMINNIIIIIill 80 IMMINEIN'INNIIIMMIIIIIM11111111111=1111111111M111111MINIMIlli 60 IM=11110111111 50oil am 401111n = 3538MMIIIIIIIIIIIIIM1111111 E30ERN: :°'11111K11111 cL IIIIy = 2.63 4.75ENE i 9-. 20 .... r = 1.741* a) Ill AIM NNWINIFAM=MIMIUMEWA17ANIIIMMINION ilm Om MINI 11101111111111= 111111111111111111=1111111111 111111111. =11100irint221IMMIIMIIIIIII a) 5Miliah. MIIIILIEMItil allImill o4 IIIIIMMINSISMillillMIIIIIIIIIII o 3 iiiiMaiiiiii ME111111 co 1111 E,IIIIIII 111E1 2 lie I,

1 kig 19I .3 .4.5.6 .810 3 4 56 810 B in soil, mg/l (Hot W. so/ Fig. 22. Regression of B content of pot-grown wheat (y) on hot water extractable soil B (x) for all the international material. National mean values of plant and soil B are also given. Countries:

Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le Lebanon, Mt -= Malta, Mw = Malawi, Me = Mexico, Ne = Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta -= Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia. The boron content zones IV are explained in Section 1.4.3. 48 national mean values of plant and soil B contents are also given in the graph. Because of a wide internal variation within most countries, the national mean values only loosely define the B status in different countries. However, distinct differences between some of the countries can be seen. For example, B deficiency is likely to be more common in Nepâl, Zambia, Nigeria and Philippines than in Iraq or Mexico. Detailed examination of the data by countries is given in Part II. The regressions of micronutrient uptake as opposed to micronutrient content of pot-grown wheat, on micronutrients in soil were calculated for all six micronutrients. In general, the uptake correlations differed little from the content correlations. The greatest difference in favour of uptake was found in the case of boron, for which element the correlation (r) between the plant content and soil content was 0.741*** and the respective uptakesoil correlation 0.758***. The latter regression for the whole material is given in Fig. 23. Comparison of the regressions in Figures 22 and 23 reveals no essential dissimilarities between the two methods of expressing plant B.

100 =immiumimm i mmmoommini 80 70.01 60 50 40 6.00 30 II 0.75E111111 MINNIE 1111111..11111111 1UIIIliiI1NhIIII 8 mmospporAmmoimmEn 1111NREIPMENWINI 4 NIMPrjahEIMIII 3 ill RENEW

liiiDI .3 .4.5.6 .8 10 3 456 8 10 B in soil, mg/l (Hot w sol.) Fig. 23. Regression of B uptake by pot-grown wheat (y) on hot water extractable soil B (x) for all the international material. For abbreviations see Fig. 22.

2.3.3.2 Soil factors affecting the boron contents of plants and soils The regressions of hot water extractable soil B and plant B on six soil factors are given in Figs 24 and 26. The regression equations and correlation coefficients are given in Table 10. 49

8 16 32 64 8 16 32 64 CEO, me/100g CEO, me/100g

f I I 1P 16 32 64 CEC,me/100g CEO. me/100g Fig. 24. Relationships of plant B and soil B to cation exchange capacity. Graph a: Hot water soluble soil B as a function of CEC. Columns indicate arithmetic mean soil B values of each CEC class)) Graph b: Plant B as a function of CEC; n values within each CEC class are given in this graph. Graph c: Ratio of plant B to soil B as a function of CEC. Points indicate the ratios within each CEC class. The regression curve calculated for the whole material (n = 3538) also indicates the coefficient, k(CEC), for correcting soil B for CEC (right-hand ordinate). Graph d: Soil values as corrected with CEC correction coefficient. Equations and correlation coefficients for the regression curves are given in Table 10. i) Because the columns indicate arithmetic mean values but the regression curves are logarithmic, the latter are often located at a lower level.

Both soil B and plant B were significantly correlated with all six soil factors studied. The highest correlations were found in the case of electrical conductivity and pH. The plant B and soil B contents, however, were similarly correlated to pH as well as to electrical conductivity. A closer statistical examination showed that cation exchange capacity had the most disturbing effects on plant B and soil B. Therefore, the CEC, obviously indirectly, was the most likely soil factor impairing the correlation between hot water soluble soil B and plant B. Effectsof CEC onsoilB andplantB.Based onanalyticaldata from 3538 sample pairs of soils and plants, the contents of hot water soluble soil B increased with increasing CEC reaching a maximum at CEC values of about 35-45 me/100 g and decreased with further increase of CEC (Fig. 24 a). The B content of plants was quantitatively somewhat less dependent on CEC and the maximum was already reached at CEC values around 20 me/100 g (Fig. 24 b). The ratio of B content of plants to hot water soluble soil B is given as a function of CEC in Fig. 24 c. The regression (curve c) indicates the correction of soil B values needed at 50 10080 g n= 38 60 T±s= ±48 50 =07 072 40 E = 2 08+ 50x a_ 30 r=0 826 "

4,5;20 a) 10 8 6 u. a) L Si. cij 5 Ni *a 0 r 4C- o 4 e Th 3

.081 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 C EC -corrected B in soil, mg /l (Hot w. sol.) Fig. 25. Regression of B content of the pot-grown wheat (y) on CEC-corrected hot water soluble B (x) for the whole international material. National mean values of plant and soil B plotted in the graph. For abbreviations see Fig. 22. various CEC levels to eliminate the difference in CEC effects between soil B and plant B analyses. The CEC correction coefficient for hot water soluble soil B is given by: k(CEC) = 0.1164 X101.4 0.026 CEC + 0.000273 CEC2 where the exponent is obtained from the regression equation in Table 10 c and the constant (0.1164) is a reversion coefficient for restoring the soil B values to their original level. The formula above can be simplified to: k(CEC) = 100.466 0.026 CEC + 0.000273 CEC2 Numerical values for k(CEC) can also be read off the right-hand ordinate of the regression curve in Fig. 24 c. When soil B values in the whole material are corrected with k(CEC), the relationship between soil B and CEC given in Fig. 24 d is obtained. This relationship is very similar to the one between plant B and CEC given in Fig. 24 b. In other words, the effect of CEC on soil B has not been eliminated but it has been made equal to its effect on plant B. Application of k(CEC) to all soil B values of the whole international material improved the correlation between plant B and soil B from an r value of 0.741*** (uncorrected) to 0.826*** (Figures 22 and 25). This improvement takes place in spite of the fact that B which normally is present in anionic form is corrected for cation exchange capacity. Obviously, the total contribution of all soil factors related to CEC are indirectly reflected by CEC. 51 Application of the CEC correction at the national level improved the plant B-soil B correlation in 24 out of 29 countries (Table 9). The improvement was most warranted in those countries (e.g. Thailand, Tanzania and Zambia) where the variation of soil CEC was wide.

Table 9. Correlations (r values) between plant B and hot water extractable soil B in various countries. Soil B analyses are both uncorrected and corrected for cation exchange capacity [k(CEC) = 10 0.466 - 0.026 CEC - 0.000273 CEC2]. Correlations of the best fitting regressions are given. Regression models: a) y = a + bx; b) y = a + b log x; c) log y = a + bx, and d) log y = a + b log x.

Uncorrected CEC -corrected Uncorrected CEC -corrected Country n Regr. Regr. Country n Regr. Regr. modelr model r modelr modelr Belgium 36 a 0.521** a 0.469** Egypt 198 a 0.597*** a 0.761*** Finland 90 a 0.668*** a 0.605*** Iraq 150 a 0.945*** a 0.947*** Hungary 201 a 0.506*** a 0.579*** Lebanon 16 a 0.132 n.s. b 0.255 n.s. Italy 170 a 0.700*** a 0.740*** Syria 0.696*** d 0.715*** Malta 25 a 0.771*** c 0.784*** Turkey 29388 a 0.851*** a 0.905*** New Zealand 35 a 0.179 n.s. a 0.079 n.s. Ethiopia 125 a 0.338*** a 0.418*** Argentina 208 a 0.524*** a 0.537*** Ghana 93 a 0.162 n.s. c 0.208* Brazil 58 a 0.501*** c 0.458*** Malawi 97 a 0.336*** a 0.363*** Mexico 242 a 0.793*** a 0.814*** Nigeria 153 a 0.381*** d 0.409*** Peru 68 a 0.560*** a 0.677*** Sierra Leone 48 a 0.357* b 0.470*** Tanzania 163 a 0.195* d 0.346*** India 258 a 0.809*** a 0.828*** Zambia 44 a 0.236 n.s. a 0.500*** Korea, Rep. 90 a 0.738*** a 0.771*** Nepal 35 a 0.544*** a 0.577*** Pakistan 237 a 0.910*** a 0.886*** Philippines 194 a 0.371*** a 0.393*** Sri Lanka 18 a 0.206 n.s. a 0.236 n.s. Thailand 150 a 0.013 n.s. d 0.223** Whole material 3538 a 0.741*** a 0.826***

Effects of other soil factors on soil B and plant B. Soil pH seems to have a relatively small effect on hot water soluble soil B as also on the B content of the plant within the pH range of 4 to about 7.7. Nevertheless, a slight increase with rising pH can be observed, especially in soil B (Fig. 26, curves e and g). With a further rise in pH, however, there was a very marked increase in soluble soil B as well as in the contents of plant B. Correction of soil B values for CEC did not appreciably affect the soil B-pH relationship, apparently because of the relatively low correlation between pH and CEC (Fig. 24 c and Fig. 4 a, Section 2.1). Soil texture is a factor closely related to the cation exchange capacity of soils (R = 0.802***, Fig. 3 h). Consequently, the relations of soil B and plant B to soil texture (Fig. 26) are very similar to the respective B-CEC relations (Fig. 24). For the same reason, the correction of soil B values for CEC renders the relationship betweensoil B and texture index more like the relationship between plant B and texture (Fig. 26, curves i and j). Organic carbon content of soils is another factor accounting for the cation exchange capacity of soils (R 0.535***, Fig. 3 m). Since this relation is not so close as that between CEC and texture, the resemblance of B-organic carbon regressions (Fig. 26, curves k, 1 and m) to B-CEC regressions is not so good as in the case of texture. Itis obvious, however, that the CEC correction decreases the difference between the regressions of soil B and of plant B on the organic carbon content of soil. 52

Fig.26.Relationships of hot-water soluble soil B and plant B to various soil factors. Graphs on the left: Un- corrected soil B (wide columns and regression curves e, h, k, n, and q) and CEC-corrected soil 6 Soil pH B (narrow columns and regression curves f,i,1, o, and r) as functions of various soil factors. Graphs on theright: Plant B as a function of various soil factors. Number of samples (n) withineachclass is given.Theequations andcorrelationcoef-

10 30 50 70 90 0 30 50 70 90 ficientsforregression Texture index Texture index (al curves (e to s) are given in Table10.See also footnote to Fig. 24.

.8 16 32 64 4 .8 16 32 64 Organic C,% Organic C.%

20 o

16 64 25.6 .4 1.6 6.4 25.6 EL conductivity,10'-Scro-1 El. conductivity,10-4Scml

1.5-

.4 1.6 64 256 .4 16 6.4 256 CaCO3 eguiv ,% CaCO3 eguiv.,% 53

Table la Equations and correlation coefficients for regressions of hot water soluble soil B (uncorrected and CEC-corrected) and plant B on various soil factors. Regression curves are given in Figures 24 and 26.

Variables Regression Correl. Regr. coeff. curve Soil B (uncorr.) CEC 3538 log y = -0.711 + 0.0274 x - 0.000320 x2 0.389*** a Plant B CEC 3538 log y =0.375 + 0.637 log x - 0.263 (log x)2 0.153*** b Plant B/soil B CEC 3538 log y =1.400 - 0.0260 x + 0.000273 x2 0.524*** CEC-corr. soil BCEC 3538 log y = -- 0.760 + 0.868 log x 0.349 (log x)2 0.105*** d

Soil B (uncorr.) Pi I 3538 log y =0.260 - 0.300 x + 0.0324 x2 0.405*** e CEC-corr. soil BpH 3538 log y ----- 0.972 - 0.508 x + 0.0475 x2 0.396*** f Plant B pH 3538 log y =2.094 - 0.500 x + 0.0433 x2 0.427*** g

Soil B (uncorr.)TI 3536 log y = -0.937 + 0.0246 x 0.000187 X2 0.391*** h CEC-corr. soil BTI 3536 log y = -0.433 + 0.00794 x - 0.000075 x2 0.091*** i Plant B TI 3536 log y =0.664 + 0.00403 x - 0.000047 x2 0.097*** j

Soil B (uncorr.)Org. C 3538 log y = -0.249 + 0.284 log x - 0.316 (log x)2 0.234*** k

CEC-corr, soil BOrg. C 3538 log y = -0.238 + 0.00459 log x - 0.146 (log x)2 0.052** 1 Plant B Org. C 3538 log y =0.736 - 0.122 log x + 0.084 (log x)2 0.185***

Soil B (uncorr.)El. cond. 3538 y = 0.528 + 0.0729 x 0.436*** n CEC-corr. soil BEl. cond. 3538 y = 0.511 + 0.0789 x 0.489*** o Plant B El. cond. 3538 y =4.44 + 0.586 x + 0.00144 x2 0.594*** p

Soil B (uncorr.)CaCO3 eq. 3538 log y =- -0.237 + 0,115 log x + 0.0133 (log x)2 0.413*** q CEC-corr. soil BCaCO3 eq. 3538 log y = -0.234 + 0.0930 log x + 0.0146 (log )02 0.348*** r Plant B CaCO3- eq. 3538 log y =0.735 + 0.0563 log x + 0.0154 (log x)2 0.338*** s

Electrical conductivity and CaCO3 equivalent of soils have a high mutual correlation (Fig. 4, j and o). Furthermore, both of these characteristics are highly correlated with soil pH (Fig. 3, d and e; Fig. 4, f and k). Therefore, the relations of soil B and plant B to these factors resemble the respective B-pH relationships. A common feature of these relationshipsisthe increase of both soil B and plant B with increasing electrical conductivity and with increasing CaCO3 equivalent (Fig. 26). In short, both of these soil characteristics affect equally the soil B and plant B. Because of the relatively low correlation between CEC and the above soil factors (Fig. 4, d, e, i, n) the correction of soil B values for CEC does not materially affect the soil B-electrical conductivityor the soil B-CaCO3 equivalent relationships. The effects of the above six soil characteristics on soil and plant Bcan be summarized as follows: Determination of soil B by the hot water extraction method is clearlya useful way of estimating the B status of soils. The correlations between B content of plants and hot water soluble soil B (uncorrected and CEC-corrected) are higher than those for any other of the six micronutrients under study. Soil pH, etectrical conductivity and CaCO3 equivalent have similar effectson hot water soluble soil B as on the B content of plants. Therefore, these soil factors do not cause appreciable differences between the analytical results of soil B and plant B, and no corrections due to these factors were necessary. The correction of soil B for CEC does not materially affect the relationship of soil B to the above three soil characteristics. CEC, texture, and organic carbon have quite similar effectson hot water soluble B 54 values but very different effects on the ability of plants to absorb B. This dissimilarity is most pronounced in the case of CEC, apparently because the nature of the other two characteristics (texture and or_ganic carbon content) is partly reflected in CEC. Therefore, the correction of soil B values for CEC substantially corrects the soil B values for texture and organic carbon as well. Correction coefficients were also calculated (analogously to the procedure for CEC) for the other five soil factors. These are not presented because none of them, as applied to the whole material, improved the soil Bplant correlation as fundamentally as does the CEC correction. When the results of this study are expressed by countries (Part II), both uncorrected and CEC-corrected soil B values are given.

2.3.4 Copper

2.3.4.1 General aspects The average Cu contents of the original and pot-grown plants the AAAc-EDTA extractable Cu contents of the respective soils in the whole material, and the best fitting plant/soil regressions are given below. The corresponding mean values by countries are given in Appendixes 2, 3 and 4. Copper content in plant DM in resp. soils Indicator mean + s mean + s Regression of plant Cu (y)Correlation') plant n (PPm) (mg/1) on soil Cu (x) (r) Original maize 1966 11.6 ± 4.2 6.0 ± 7.9 log y -= 0.97 + 0.114 log x 0.344*** Original wheat 1768 9.4 ± 5.7 6.1 ± 6.4 log y = 0.90 + 0.0062 x 0.254*** Pot-grown wheat 3537 7.0 ± 2.6 6.0 ± 7.0 log y = 0.62 + 0.309 log x 0.664*** 1) Respective correlations between plant Cu and DTPA extractable soil Cu were: 0.114***, 0.125*** and 0.518***. The mutual correlation between AAAc-EDTA extractable and DTPA extractable soil Cu was 0.829***. Because of their better correlation with plant Cu, the results given in this study are based on AAAc-EDTA extraction.

The average Cu content of maize exceeds that of the original wheat by 2.2 ppm and that of the pot-grown wheat by 4.6 ppm. There is no difference, though, in the mean values of extractable soil Cu between the three groups. The higher Cu contents of the original wheat as compared to the pot-grown wheat may partly be due to the small amount of soil in the pots (the pots were partly filled with quartz sand, see Section 1.2.6). The plants were not able to absorb from this small amount as much Cu as could plants grown on the same soils under field conditions. The coefficients of variation of the Cu contents of plants ranged from 36 to 61 percent in the three groups, and those of soils from 105 to 132 percent. For reasons explained previously (Section 1.2), the correlation between the Cu contents of the original plants and the extractable Cu in soils was weak compared with that between the pot-grown wheat and soil. The regression for the latter is given in Fig. 27, where also the national mean values of plant and soil Cu contents have been inserted for a general comparison of the Cu status in the different countries. The correlation between Cu uptake and soil Cu (r = 0.674***) is about equal to that between the Cu content of plants and soil Cu. Also within individual countries, the differences between these two methods of expression are very small. 55

20 n = 3 38 y ±s 0 18 0 E a_ cal 611111g0 g 111.111111

111111111 1111111111 11021111111

(-)6 1111111EMIENIEN111111

4 11,M1111111EN1111 2IINIENIME1111ismompli MI 11111 0.1 .2 .4 .6 .81.0 2 46 810 20 40 60 80 Cu in soil, mg/l (A AA c - EDTA)

Fig. 27. Regression of Cu content of pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x) for the whole international material. National mean values of plant and soil Cu are also given. Countries: Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It = Italy, Ko =--- Korea, Rep., Le=Lebanon, Mt =Malta, Mw=Malawi, Me =Mexico, Ne = Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta = Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia.

2.3.4.2 Soil factors affecting the copper content of plants and soils The Cu content of pot-grown wheat and AAAc-EDTA extractable soil Cu as functions of six soil characteristics are given in Figures 28 and 30. The regression equations and correlation coefficients are given in Table 12. The AAAc-EDTA extractable soil Cu as well as the Cu content of wheat were significantly correlated with all six soil characteristics studied. In general, the effects of various soil factors on extractable soil Cu were in relatively good agreement with their respective effects on the Cu contents of plants. The most marked exception was the organic carbon content of the soil whose effect on the extractable soil Cu differed clearly from its effect on the Cu content of plants. Consequently, the different roles of soil organic carbon during the chemical extraction of soil Cu and during the Cu absorption by plants are likely to impair the plant Cusoil Cu correlation. 56

.8 16 3.2 6.4 .8 16 32 64 Organic C,% Organic C,%

3 3 , 7 5' 6, o _:5 6 .. . '5 2 ...2 5 Le)

o o 3 4 c C.) -a ..2 3 o 11 1 a)

8 6 (.) ¿11 til ch ci c5- .4 .8 1.6 3.2 6,4 .8 16 32 64 Organic C,% Organ c C,%

Fig. 28. Relationships of AAAc-EDTA extractable soil Cu and plant Cu to organic carbon content of soil. Graph a: Soil Cu as a function of soil organic carbon content. Graph b: Plant Cu as a function of soil organic carbon content. Graph c: Ratio of plant Cu to soil Cu as a function of soil organic carbon content. The regression curve c also indicates the coefficient, k(org. C), for correcting soil Cu for organic carbon (right-hand ordinate). Graph d: Soil Cu as corrected with k(org. C), as a function of organic carbon content. Columns and points indicate mean values of each organic carbon class; n values of classes are given in Graph b; equations and correlation coefficients for the regression curves are given in Table 12.

Effects of soil organic carbon content on soil Cu and plant Cu. AAAc-EDTA extractable soil Cu increased with increasing organic matter content of soils up to organic carbon contents of about 1-2 percent (Fig. 28 a). With further increases of organic carbon, the Cu contents of soil begin to decrease. On the basis of the results of this study, it is not clear whether this decrease is due to firmer Cu fixation by organic matter in soils of high organic matter content, or due rather to a progressive parallel decrease in the total Cu content toward organic soils. The latter possibility seems more likely. For instance, Sillanpää (1962 a, b) found a similar correlation between total Cu content of soils and soil organic matter, while the relative solubility of Cu increased slightly with increasing organic matter content of the soil. The Cu content of plants increases only slightly with increasing soil organic carbon in soils of very low organic matter content, while further increases in organic carbon cause a decrease in the Cu content of plants. The ratio plant Cu/soil Cu (Fig. 28 c) indicates the difference between the effects of soil organic carbon on plant and soil Cu contents. The regression (curve c) also indicates the correction required to eliminate this difference. The correction coefficient for correcting soil Cu values for organic carbon is given by: 57

22 Intern 20 111111111111111111

18 1111111111111111111

E16 1111111111111.111111 o_ o_ 4-7 111111111111111111111 o 14 _c 12 1111111111111111111 o 10 111111111111111111111 a) 8 11111016111111111111 o 61111111111111MEN011111111111 C.) 411111111111111MIT°111111141111111

21111111111M111111111 MI11116

o:1 .2 .4.6 .810 2 4 6 810 20 40 60100 Org.C-corrected soil CU,mg/l (AAAc-EDTA)

Fig.29.Regression of Cu content of pot-grown wheat (y) on k(org. C)-corrected AAAc-EDTA extractable soil Cu (x) for the whole international material. National mean values of plant and soil Cu are plotted in the graph. For abbreviations see Fig. 27.

k(org. C) =0.668 X100.175 0.491 log x + 0.470 (log x)2 where the exponent is obtained from the regression equation in Table12(curve c where x = organic carbon content of soil) and the constant(0.668)is a reversion coefficient for restoring the soil Cu values to their original level. The formula above can be simplified to: k(org. C) = 10-0.491 log x+ 0.470 (log 0 Numerical values for k(org. C) can be read off the right-hand ordinate of the regression curve in Fig.28 c. When soil Cu values in the whole material are corrected with k(org. C), the relationship between soil Cu and organic carbon given in Fig.28 dis obtained. This relationship resembles basically the respective plant Cuorganic carbon relationship (Fig.28b), i.e. the effects of soil organic carbon on soil Cu and plant Cu have been equalized. Application of k(org. C) to AAAc-EDTA extractable soil Cu values in the whole material improved the correlation between plant Cu and soil Cu from an r value of 0.664*** to0.731***(Figures27and29).At the national level it improved the correlation in21out of29countries (Table 11). The improvement is not so pronounced as, for example, in the case of pH corrections applied to Mn and Mo. This may be partly due to the relatively narrow range of variation in organic carbon content in this material which included only a few peat soils. 58

Table 11. Correlations (r values) betweerf plant Cu and AAAc-EDTA extractable soil Cu in various countries. Soil Cu analyses are both uncorrected and corrected for soil organic carbon contents [k(Org. C) = 10-0.491 log x + 0.470 (log x)2, where x = Org. C, Vo]. Correlations of the best fitting regressions are given. Regression models: a) y = a + bx; b) y = a + b log x; c) log y = a + bx, and d)log y = a + b log x.

Uncorrected Org. C-corrected Uncorrected Org. C-corrected Country n Regr. Regr. Country n Regr. Regr. modelr modelr modelr modelr Belgium 36 d 0.583*** d0.686*** Egypt 198 d 0573*** d 0579*** Finland 90 d 0.633*** d0.622*** Iraq 150 d 0.317*** d 0.400*** Hungary 201 b 0.289*** a 0.368*** Lebanon 16 b 0.149 n.s. b 0.097 n.s. Italy 170 b 0.804*** b 0.814*** Syria 38 d 0.644*** d 0.652*** Malta 25 d 0.760*** d 0.834*** Turkey 298 d 0.538*** b 0.523*** New Zealand 35 b 0.842*** b 0.817*** Ethiopia 125 d 0.830*** d 0.840*** Argentina 208 d 0.606*** d 0.693*** Ghana 93 b 0.726*** b 0743*** Brazil 58 b 0.625*** d 0.649*** Malawi 97 d 0503*** d 0.558*** Mexico 242 d 0.520*** d 0.587*** Nigeria 153 d 0.598*** d 0.651*** Peru 68 d 0.616*** d 0.568*** Sierra Leone 48 d 0.706*** a 0.751*** Tanzania 163 d 0.727*** da735*** India 258 b 0.396*** a 0.281*** Zambia 44 b 0848*** b0.870*** Korea, Rep. 90 b 0.456*** a 0.548*** Nepal 35 d 0.410* d 0.519** Pakistan 237 d 0.371*** d 0.282*** Philippines 194 d 0.632*** d 0.687*** Sri Lanka 18 d 0.723*** d 0.665*** Thailand 150 d 0.742*** d 0.759*** Whole material 3538 d 0.664*** d 0.731***

Effects of other soil factors on soil and plant Cu. Soil pH affects the AAAc-EDTA extractablesoil Cu and Cu content of plants considerably less than the respective contents of Mn and Mo. However, a clear increasing trend of soil Cu and plant Cu toward alkaline soils can be noticed (Fig. 30 and Table 12, curves e, f, g). Correction with k(org. C) alters the soil Cu-pH relation slightly in the direction of the respective relationship between plant Cu and pH. As soil texture becomes finer (texture index rises), the Cu contents of plants and especially the Cu contents of soils increase quite substantially (Fig. 30 and Table 12, curves h, i, j). The correction for organic carbon slightly moderates the effect of texture on soil Cu. Cation exchange capacity, owing to its close correlation with texture (Figs 3 h and 4 b; see also Section 2.1), has very similar relations to soil and plant Cu as those of texture. Electrical conductivity and CaCO3 equivalent of soil are highly correlated with each other as well as with pH (Figures 3 and 4). Consequently, the relations between soil and plant Cu to these soil factors (Fig. 30, curves n-s) do not differ substantially from the Cu-pH relations. The copper contents of plants increased steadily with increasing electrical conductivity as well as with increasing CaCO3 equivalent. Increases in the AAAc-EDTA extractable soil Cu at the highest levels of electrical conductivity and CaCO3 equivalent levels tended to result in decreases in Cu contents of plants. The correction of soil Cu values for organic carbon had relatively small effects on the above relationships. The results given in this Chapter can be summarized as follows: - Soil Cu values determined from AAAc-EDTA extracts werebetter correlated with Cu contents of plants (r = 0.664***) than those determined from DTPA extracts (r = 0.518***). Therefore, the results presented in this study are based on the former extraction. 59

Fig. 30. Relationships

10 of AAAc-EPTA extractable soil Cu and plant 8 Cuto varioussoil 6 factors. Graphs on the left: Un- 4 corrected soil Cu (wide columns and regression curves e, h, k, n, and 4 6 7 8 q) and org. C-corrected Soil pH soil Cu (narrow columns and regression curves f, i, 1, o,and r)as 10-- functions of various soil factors. 8 Graphs ontheright: E 6 Plant Cu as a function

4 of various soil factors. Number of samples (n) o 2 withineachclass is given. The equations and 30 50 70 Textu e index (To Texture index Cm correlationcoefficients for regression curves (e to s) are given in Table 12. See also footnote to Fig. 24.

8 16 32 64 6 32 64 CEO, me 1009 CEO, me/100g

8 10

16 6.4 256 16 64 256 El. conductivi y,10-4 S cm El. conductivity, 0- S cm-i

E 8 0.

'ae; 6

4 4

CO 04 03 03 N c n n 02 2 N N o7

6 64 25.6 .4 16 64 256 CaCO, equiv., / CaCO3equiv. % 60

Table 12. Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Cu (uncorrected and Org. C-corrected) and plant Cu' on various soil factors. Regression curves are given in Figs 28 and 30.

Variables Regression Correl. Regr. Y coeff. curve

Soil Cu (uncorr.) Org. C 3538log y =0.655 + 0.406 log x - 0.675 (log x)2 0.315*** a Plant Cu Org. C 3538 log y =0.830 - 0.0857 log x - 0.205 (log x)2 0.204*** b Plant Cu/soil Cu Org. C 3538 log y =0.175 - 0.491 log x -F 0.470 (log x)2 0.447*** Org. C-corr. soil Cu Org. C 3538log y =0.657 - 0.0868 log x - 0.239 (log x)2 0.104*** d

Soil Cu. (uncorr.) pH 3538log y = -1.598 + 0.600 x - 0.0390 x2 0.330*** e Org. C-corr. soil Cu pH 3538log y = -1.107 -F 0.405 x - 0.0208 x2 0.447*** f Plant Cu pH 3538log y =0.840 - 0.0712 x + 0.00984 x2 0.357*** g

Soil Cu (uncorr.) TI 3536log y = -3.682 -I- 4200 log x - 0.942 (log x)2 0.574*** Org. C-corr. soil Cu TI 3536log y = -2.334 -F 2.822 log x - 0.602 (log x)2 0.450*** Plant Cu TI 3536log y = -0.138 -F 0.904 log x - 0.193 (log x)2 0.296*** j

Soil Cu (uncorr.) CEC 3538 log y = -1.417 -F 2.223 log x - 0.524 (log x)2 0.561*** k Org. C-corr. soil Cu CEC 3538 log y = -0.439 + 1.037 log x - 0.179 (log x)2 0.391*** I Plant Cu CEC 3538 log y =0.378 -F 0.536 log x - 0.155 (log x)2 0.191***

Soil Cu (uncorr.) El. cond. 3538 log y =0.548 -F 0.543 log x - 0.287 (log x)2 0.368*** Org. C-corr. soil Cu El. cond. 3538 log y =0.562 + 0.532 log x - 0.254 (log x)2 0.389*** o Plant Cu El. cond. 3538 log y =0.776 -F 0.184 log x - 0.0300 (log x)2 0.349***

Soil Cu (uncorr.) CaCO3 eq. 3538 log y =0.707 + 0.0666 log x - 0.0311 (log x)2 0.347*** Org. C-corr. soil CuCaCO3 eq. 3538log y =0.749 + 0.0747 log x - 0.0428 (log x)2 0.441*** Plant Cu CaCO3 eq. 3538log y =0836 00513 log x 0.383***

Both the AA Ac-EDTA extractable soil Cu and the Cu content ofpot-grown wheat were highly correlated with all six soil characteristics (pH, texture, organic carbon content, CEC, electrical conductivity, and CaCO3 equivalent) under study. In general, these soil factors affected the extractable soil Cu in much thesame way as they affected the Cu content of plants. An exception, however,was the organic carbon content of soils. Correction of soil Cu values for organic carbon improved the plant Cu-soil Cu correlation for the whole material from an r value of 0.664*** to 0.73 l***. Similarly calculated correction coefficients for the other five soil characteristics did not improve the plant Cu-soil Cu correlations. -- When the results of this study are expressed by countries, both the uncorrected and the k(org. C)-corrected soil Cu values are given. 61 2.3.5 Iron

23.5.1 General aspects Various factors caused uncontrolled variation in the analytical results of the original plants, as explained in Section 1.2. There was thus no correlation between the Fe contents of original plants and extractable soil Fe contents in the respective soils. The correlations (r values) between plant Fe and AAAc-EDTA extractable Fe for the original maize and wheat were 0.007 and 0.032, respectively. For DTPA extractable Fe the respective correlation coefficients were 0.028 and 0.030. The original plants were obviously contaminated, which affected the Fe analyses to a much greater extent than those of other micronutrients. Accordingly, the results of Fe analyses of the original plants were discarded. The Fe contents of pot-grown wheat were more consistent with the contents of extractable soil Fe. The plant FeAAAc-EDTA extractable soil Fe and plant Fe- DTPA extractable soil Fe correlations (r values) in the whole material were 0.325*** and 0.263***, respectively)) The former relation is given in Fig. 31. The national mean values are shown in the graph. The correlation between the Fe contents of pot-grown plants and AAAc-EDTA extractable soil Fe (r = 0.325***) is weak compared to correlations for other micronutrients in this study, but stronger than the correlation between Fe uptake and soil Fe (log y = 1.65 X 0.106 log x; r = 0.266***).

2.3.5.2 Soil factors affecting the iron contents of plants and soils The AAAc-EDTA extractable soil Fe and the Fe content of pot-grown wheat as functions of the six soil characteristics are given in Fig. 32. The respective regression equations and correlation coefficients are given in Table 13. Contrary to the cases of the other five micronutrients, none of the six soil factors studied greatly affected the Fe contents of plants. The regression curves depart only slightly from the horizontal and consequently, the correlations are relatively weak. Soil Fe seems to have been more affected by the various soil factors than plant Fe. Increases in soil pH and CaCO3 equivalent values were accompanied by decreases in the extractable soil Fe contents. A similar, though milder relation between these soil characteristics and plant Fe can be noticed (Fig. 32 a, b, k, 1). Increasing texture index, organic carbon content and CEC were accompanied by increases in the contents of extractable soil Fe. The effects of these soil factors on Fe contents of plants are similar in direction to, but of lesser magnitude (Fig. 32 c, d, e, f, g, h). The effects of electrical conductivity on Fe (Fig. 32 i, j) are not marked. None of the above six soil factors affected the plant Fesoil Fe relations to such a degree that correction for these soil characteristics would have appreciably improved the correlation between plant Fe and soil Fe. This may be partly due to the insensitivity of plant Fe to the soil factors studied. Wheat is apparently a poorer indicator of the Fe status of soils than of the status of the other micronutrients studied. The oxidation-reduction conditions of soils are important in determining the behaviour of Fe in soils and its availability to plants. In this study the evaluation of these aspects was not possible.

I) The mutual correlation (r) between AAAc-EDTA extractable and DTPA extractable soil Fe was 0.741***. Because of their better correlation with plant Fe, the AAAc-EDTA extractable soil Fe contents only will be presented in this study. 62 300 275Fe 2 50 n . 3 38 225 y o 1. 200 s = 15 1 Lo =1 4 91 1 75 r0 32 E 1 50 125 o

100 \ 1 1 11111 45 90

4E' 80 Ph

4E. 70 o o 60 Et a) e 50

30 BI 10 20 40 60 100 200400 600 1111 2000 Fe in soil, mg/l (AAAc- EDTA)

Fig. 31. Regression of Fe content of pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Fe (x) for all the international material. National mean values of plant and soil Fe are also given. Countries: Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le = Lebanon, Mt = Malta, Malawi, Me -= Mexico, Ne = Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta = Tanzania, Th ---- Thailand, Tu = Turkey, Za = Zambia. The Fe content zones IV are explained in Section 1.4.3. 63

300 Fig.32. AAAc-EDTA extractable soil Fe (graphs on the left) and plant

Tri Fe (graphs on theright)as a function of six soil characteristics. Number of samples (n) given within eachclass.Theequationsand a, correlation coefficients for regression u_ curves (a-1) are given in Table 13. Soil pH

30 50 70 90 Texture index(,) Texture index 1,1

1500

1200

,S 900 2 .9 600 u_ 300

.8 16 az 6 Organic C,%

250

E 200 a

E 150

.5 100 a, L'- 50 a, u_

8 16 32 64 8 16 32 64 CEC,me/100g CECme/100g

16 64 256 16 64 256 El conductivity,10" S cm' El. conductivi y,10- S cm'

.4 18 6.4 256 .4 18 6.4 256 CoCO,egiuv. CaCO, equiv, % 64

Table 13. Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Fe and plant Fe on various soil factors. Regressions are given in Fig. 32.

Variables n Regression Correl. Regr. y x coeff. curve Soil Fe pH 3538 log y = -0.965 + 9.127 log x - 6.538 (log x)2 0.300*** a Plant Fe pH 3538 log y = 1.580 + 0.0757 x - 0.00688 x2 0.136*** b

Soil Fe TI 3536 log y = 1.374 + 0,442 log x 0.228*** Plant Fe TI 3536 y =59.14+ 0.040x 0.038*** d

Soil Fe Org. C 3538 y = 62.2 + 80.2 x 0.547*** e Plant Fe Org. C 3538 log y = 1.769 + 0.0629 log x - 0.0176 (log x)2 0.145*** f

Soil Fe CEC 3538 y = -37.62 + 147.7 log x 0.230*** g Plant Fe CEC 3538 y =52.74 + 5.949 log x 0.084*** h

Soil Fe El. cond. 3538 y =162.4 + 33.05 log x - 23.18 (log x)2 0.049** Plant Fe EL cond. 3538 log y = 1.769 - 0.0274 log x + 0.0368 (log x)2 0.074*** j

Soil Fe CaCO3 eq. 3538 log y =2.057 - 0.0586 log x 0.00395 (log x)2 0.241*** k Plant Fe CaCO3 eq. 3538 log y = 1.767 - 0.00868 log x 0.000533 (log x)2 0.095*** 1

2.3.6 Manganese 2.3.6.1 General aspects The average Mn contents of the three types of indicator plant, the DTPA extractable Mn contents of the respective soils, and the best fitting plant Mn-soil Mn regressions for the whole material are given below. For the respective national averages, sec Appendixes 2, 3 and 4. Manganese content in plant DM in I-esp. soils Indicator mean ± s mean ± s Regression of plant Mn (y)CorrelationO plant n (PP91) (mg/1) on soil Mn (x) (r) Original maize 1966 78 ± 48 43.2 ± 38.4 y = 68.9 + 0.200 x 0.161*** Original wheat 1768 74 ± 40 25.2 ± 33.1 y = 70.0 + 0.143 x 0.119*** Pot-grown wheat 3538 112 ± 155 34.7 ± 36.9 log y = 1.24+ 0.474 log x 0.552***

0 Respective correlations between plant Mn and AAAc-EDTA extractable soil Mn were: -0.108-***, 0.046* and 0.039*. The mutual correlation between DTPA extractable and AAAc-EDTA extractable soil Mn was 0.416***. Because of their better correlation with plant Mn, the final results given in this study are based on DTPA extraction.

The difference between the average Mn contents of the two original indicator plants is small and may be attributable more to differences in the respective soils than to the plant species. Contrary to the case with the other micronutrients, the average Mn content of the pot-grown wheat was about 50 per cent higher than that of the original wheat. Obviously, the difference is partly due to higher average extractable contents of Mn in soils in the pot experiment than in soils where the original wheats had grown. The effect of a higher Mn content of soils, however, may be nullified or even reversed by the small quantity of soil in the pots (Section 1.2.6). Another, more likely reason for the high Mn contents of the pot-grown wheats may be that the soils in the pots were in a less aerobic condition than they were in the field. The pot soils were kept near field moisture capacity during plant growth while in the wheat fields the soils must have been much drier. The availability of 65 Mn is strongly affected by oxidation-reduction reactions which largely depend on the soil moisture content. For example, Graven et. al. (1965) found a more than tenfold increase in Mn contents of plants due to flooding. See also Part II on the effect of irrigation on Mn in Iraq. The soil pH, which in the case of the pot soils was about 0.3 unit lower than in the original wheat soils (means 6.64 and 6.91, respectively), may also have been partly responsible for the increased Mn contents of the pot-grown wheats. The effect of pH will be discussed later in this Chapter. The variation in Mn contents of the pot-grown wheat is much wider than that for the other two plant types. In soils, the standard deviations (s) are of the same magnitude as the respective mean values. The correlations between the Mn content of the original indicator plants and DTPA extractable soil Mn are highly significant but poor as compared to that between the Mn content of pot-grown wheat and soil Mn. The latter regression (with national mean plant Mn and soil Mn values) is given in Fig. 33.

2000 NM NM MI INN INC= II II NOININUIPM110

1000 IMPII111111111 800 MTV. MIEN111 600 111111111111101111111111 E 400 t 300 mown g 200 II IMI& _c 100 1111111 IfJW=MMEMM=1 4E.a) 80 111=11--MEN Pi4C/INIIMMEMEN =11kWEEMEIMIMIFAMIMMEMNII 11mi g60 MMIMIONEMMEll min 40 MNIIIIIMM=11111 30 111 20

10 8

1 2 3 46 8 10 20 30 40 60 80100200 300 Mn in soil, mg/l (DTPA) Fig. 33. Regression of Mn content of pot-grown wheat (y) on DTPA extractable soil Mn (x) for the whole international material. National mean values of plant and soil Mn are also given.

Countries: Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh = Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le = Lebanon, Mt = Malta, Mw= Malawi, Me = Mexico, Ne = Nepal, NZ = New Zealand, Ni = Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy = Syria, Ta = Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia. 66 The Mn uptakesoil Mn correlation in the whole international material (log y = 1.33 + 0.481 log x; r = 0.565***) differs only slightly from that given in Fig. 33. In national materials, with a few exceptions, the differences between the two methods of expression are also small.

2.3.6.2 Soil factors affecting the manganese content of plants and soils DTPA extractable soil Mn and Mn content of pot-grown wheat as functions of six soil factors are given in Figs 34 and 38. The respective regression equations and correlation coefficients are given in Table 15. The DTPA extractable soil Mn as well as the Mn content of the plant correlated significantly with all six soil factors studied. The effects of some of these soil factors on soil Mn agreed fairly well with their respective effects on the Mn content of plants. The most striking exception was the pH of soil which affected the soil Mn and the plant Mn very differently. This difference impaired the plant Mnsoil Mn correlation. Effects of soil pH on soil Mn and plant Mn. The Mn contents of plants decreased steadily with rising pH (Fig. 34 b). This relationship is very close (R = 0.644***). The DTPA extractable soil Mn contents first increased with rising pH, reaching a maximum between pH 5 and 6, and thereafter decreased strongly toward alkaline soils (Fig. 34 a). This correlation is likewise close (R = O.695***). The ratio plant Mn/soil Mn (Fig. 34 c) illustrates the difference in pH effects at different pH levels. The equation for the regression curve (Fig. 34 c and Table 15 c) multiplied by a reversion coefficient (0.457) gives the correction coefficient, k(pH), required to eliminate the difference between the pH effects on plant Mn and soil Mn: k(pH) = 107.06 - 2.20 pH + 0.164 pH2 k(pH) can also be obtained directly from the right-hand ordinate of the curve in Fig. 34 c. When k(pH) is applied to DTPA extractable soil Mn values in the whole material (n = 3538) we obtain the relation between soil Mn and pH given in Graph d. Comparison of the relations given in Graphs b and d (Fig. 34) shows that the effect of pH on soil Mn has been equalized with its effect on Mn content of plants. In other words, the pH- corrected soil Mn values and plant Mn values are similarly correlated to soil pH. It should be noted that the pH correction coefficient is calculated for pH(CaC12). The relationship between pH(H20) and pH(CaC12) in this material is given in Fig. 5 and Section 2.1. The plant MnDTPA extractable soil MnpH relationships are also expressed in Fig. 35 where the data given in Graphs a and b (Fig. 34) are combined. The respective relationships after pH correction (combining data of Graphs d and b) are given in Fig. 36. As can be seen from Fig. 35, the DTPA extractable soil Mn values e.g. at pH levels 4 to 5 and 6 to 7 are very similar in magnitude. The Mn contents of plants grown in acid (pH 4 to 5) soils, however, are several times higher than those grown in soils having pH 6 to 7. This clear conflict between the results of soil and plant analyses was eliminated by correcting the soil Mn values for pH (Fig. 36). As numerous investigators have reported, the possibility of Mn deficiency or a response to Mn fertilization occurring in acid soils is very unlikely. Hence, the correction of soil Mn for pH will doubtless improve the reliability of soil analyses. Application of a pH correction to DTPA extractable soil Mn values in the whole of the international material improved the correlation between plant Mn and soil Mn from an r value of 0.552*** to 0.713*** (Figures 33 and 37). At the national level the correlations 67

500 E -a`i: 60 a. ct400 o

E 40 3 300 o o 2 200

c 20 o 100

Soil pH Soil pH

200

E 160 o .E 120 2 -o 80

6 6 Soil pH Soil pH Fig. 34. Relationships of DTPA extractable soil Mn and plant Mn to soil pH(CaC12). Graph a: Soil Mn as a function of pH. Graph b: Plant Mn as a function of pH. Graph c: Ratio of plant Mn to soil Mn as a function of pH. The regression curve also indicates the coefficient, k(pH), for correcting soil Mn for pH (right-hand ordinate). Graph d: pH-corrected soil Mn as a function of pH. Columns and points indicate mean values within each pH class; n values of pH classes are given in Graph a; equations and correlation coefficients for the regression curves are given in Table 15.

were improved in 25 out of 29 countries (Table 14). These were most substantial in countries having generally acid soils with wide pH variation, e.g. Belgium, Finland, Brazil, Peru, Korea, Philippines and all the African countries. In countries where the pH correction was ineffective(India, Pakistan, Egypt and Iraq) alkalinesoils with a relatively narrow pH range were predominant. Effects of other soil factors on soil Mn and plant Mn. The total Mn contents of soils increased toward fine textured -soils while the relative solubility of Mn (extractable Mn as a percentage of total Mn) increased toward coarse textured soils (Sillanpää, 1962 a, b). Clearly, the plant Mntexture relations given in Fig. 38 (curves e, f, g) result from these two opposing functions. The pH correction of soil Mn values (curves e and 0 brought the soil Mntexture relation closer to that between plant Mn and texture (curve g). Generally, the Mntexture correlations are much weaker than those between Mn and pH (Table 15). As organic carbon content of soil increased, the Mn content of plants and the DTPA extractable soil Mn content tended to increase (Fig. 38, curves h, i, j). The columns indicate a tendency for soil Mn content to decrease at the highest organic carbon levels, 68

2000 191 Fig. 35. Relationship of Mn content of pot-grown wheat to DTPA extractable soil Mn 1000 MNUlailitiliMiiiii01111111111111M1111 in the whole material classified 800MI MMEEINNLINKELEIM11111111111 600III MmnEMMIIMPNWn according to pH. The points IIIMII=EINTRE121121173MMISIEMIIIII indicate mean plant Mn and 11 E400 MEISMEINII IMMISMIII1 soil Mn contents of various a. 11 cL300 IIIIIII IIIIMEMI. pH classes. The lower limit of o 111O1111111111 IIIII1Pik each pH class is given. G) 200 _c o 1011111111 100...... 11111 80IMMOMMii=iNENNEMIZMNPAS:MANiiiIMNIMUWMCIMMENNO1=1.0rIMEN 11111111.111111 WIMMIMMIMENNIIIMMI o 60 o inlilltiiMUIDMIIIPNIMCIIii.M111 40UNIMMININIEKVIMILIMEIMIIII 300111111.1.11Mill wommul 201110Pr111111111111111imi

loIr litim IINIIIMIIIIIIIMomminlemlail 11WIMIIMIIMMEN IIIOIIIIMIMIIIIIÌIIIIMUMIIIIIIIIIIIMEMIWIIWMLMMNMMM 8IIIIIIIINIIIIIIIMINIMINIFINRIMIMMIRMILMTCMINEB 1111111.11111111MMUIIIIIIRMIMIIIIIIIIMMUNIIIMMINMEN 2 3 46 8 10 20 30 40 6080100 200 300 Mn in soil mg/l (DTPA)

2000 N1041111111MINV IMMEN101. Fig. 36. Relationship of plant Mn to pH-correct- Mn lnt n e ed soil Mn (DTPA) in 1000 - minTrarinmmummmorisimuswaiminammaitioÌÏiNÌiIlhII - the whole material 800 * 600 MINEMIll11E111111111111=11111111111=11111 classifiedaccordingto E 10.ITHEINECINEETIMMEIMIIIPE11111111111111111101211 pH. 0- 400 MEIE111 T:.; 300 111111 200 1.11111111111011111 o 100ROMMILIIM MME=1PFillal.1=MMEM11 MINIM ai 80Nommmu.-'--Nimommomariimmimmosskiwmmosnum=MMEMENPMaiRMIMMMEMIN111W7M11 o 6011111MIL. INNIMIIIIIIPPNINIMMEM11=MINIMI o IIMIIIII 1MMONEWOMMIIMINMENIII=MMENNI 4011 ElliMMINEIMMI111 =MIME 2 30ii '.1111161111111111 MINNIEN 201111.12F' 11M1111111111MIER

10MI 111111111111111 1111111 ...... IIMMINIMI1111MM=NIMINEN...... ___ .. 8MOI11 =.1111111MIRMIMMMEMEN =111MWININ=MMEN IIIIMMIll M11111111111!=11111.111MMINMENI 1=NILL1I 1 23 46 810 20 30 40 60 100200 300600 pH - corrected Mn in soil, migil (DTPA) 69 2000

1000 tiaiiiiiiiiiiiiingri=11111111M111111111111111111MMMEITINIMIWTZTPPE1N_M='MMIllillIMMEN111111MMIIIIMIMMI 800I. IMMIFLWAEM.17/ =1MMEINIIMMIIMINIMMIll 600u.IN MMINEEMIIIIMMILM11111 EMU E II MMEMPAINEEELE33 0 XWMEMINUIPAMEMIIMMEMPai Q. 400II EiguI 15; 300II Mann mmuramalmo _C 200II. I 1111111101M11111111111111 SI 100MIIIIlh,0:701MWirilIMMEN111MIMIWZMU=IMINTIOMMINIMMEN=11111MINII 80111111111MMINIMIMIIIMIMMIIIUiIENONIMINMENNIn rmiiiiatm,MMENOMW=nimitzign 60iiIMUMMIIMEN1===IMMI 1NMEMISNI 40 =MERIN =MUNN MMEIL 2 30 1111.2dlill 111111Mall 20

10IIÏIIIIIIÌIUUIIIIIII_fib IMMIliiiiMEMEMMIMMINEEM 8ImmiminmlinsaamiTamm.....EMMEME lommomreas=NEEZ:11 II NIMINIMMIMAN 3 46 810 20 3040 60 100 200 300600 pH-corrected Mn in soil, mg/l (DTPA) Fig. 37. Regression of Mn content of pot-grown wheat (y) on pH-corrected DTPA extractable soil Mn (x) for the whole international material. National mean values of plant and soil Mn are plotted in the graph. For abbreviations see Fig. 33.

Table 14. Correlations (r values) between plant Mn and DTPA extractable soil Mn in various countries. Soil Mn analyses are both uncorrected and corrected for pH(CaC12) [k(pH) = 107.06 - 2.20 pH -I- 0.164 pH2]. Correlations of the best fitting regressions are given. Regression models: a) y = a + bx; b) y = a + b log x; c) log y = a+ bx; d) log y a + b log x.

Uncorrected pH -corrected Uncorrected pH -corrected Country n Regr. Regr. Country n Regr. Regr. modelr modelr modelr modelr Belgium 36 c 0.347* c 0.737*** Egypt 198 d 0.171* d 0.139* Finland 90 a 0.104 n.s. a 0.639*** Iraq 150 d 0.057 n.s. d 0.046 n.s. Hungary 201 c 0.573*** c 0.742*** Lebanon 16 d 0.537* d 0.6425* Italy 170 d 0.462*** d 0.531*** Syria 38 d 0.384* d 0.455** Malta 25 b -0.128 n.s. b -0.090 n.s. Turkey 298 a 0.195*** a 0.232*** New Zealand 35 c 0.403* a 0.593*** Ethiopia 125 c 0.469*** d 0.665*** Argentina 208 a 0447*** a 0.664*** Ghana 93 a 0.430*** a 0739*** Brazil 58 c 0.433*** d0.704*** Malawi 97 d 0.485*** d 0.769*** Mexico 242 d 0.568*** a 0.807*** Nigeria 153 c 0395*** d 0.729*** Peru 68 c 0.508*** a a945*** Sierra Leone 48 b 0.269 n.s. b 0.635*** Tanzania 163 d 0.126 n.s. d 0.387*** India 258 a 0.438*** a 0.356*** Zambia 44 d 0.266 n.s. d 0.607*** Korea, Rep. 90 c 0.663*** a a935*** Nepal 35 d 0.368* b0.606*** Pakistan 237 a 0.287*** a 0.198** Philippines 194 d 0.517*** a 0.815*** Sri Lanka 18 c 0.674*** d 0.871*** Thailand 150 c 0.671*** a 0.827*** Whole material 3538 d 0.552*** d 0.713*** 70

30.0 Fig. 38. Relationships of DTPA extractablesoil 60 Mn and plant Mn to various soil factors. 40 Graphs on the left: Un- E corrected soil Mn (wide .! 20 columns and regression curves e, h, k, n, and q) and pH-correctedsoil 10 30 50 70 90 10 30 50 70 90 Texture index (Ti Texture index (Ti) Mn (narrowcolumns and regression curves f, i, 1, o,andr)as a

150 functionoffivesoil 60 E factors. o Graphs ontheright: 100 40 Plant Mn as a function E offivesoilfactors. 7S, Number of samples (n) .E 20 50 withineachclass is given.Theequations andcorrelationcoef- .8 16 az 64 .8 6 az 64 Organic C, % Organic C,./ ficientsforregressions (etor)are given in 60 Table 15. See also footnote to Fig. 24. E a300 40 o

E 200 E

LI 20 3100

8 16 32 64 CEC,me/ 00g CEC, me 100g

E 0a o

200

100

16 6.4 256 16 64 256 El. conduct'vi y,10-4Scrn-1 El. conductivity,10-4Scrn-1

200 oE

150 c

00 -E

.4 16 64 256 .4 16 64 256 CaCO3 equi, % CaCO3 equiv., % 71

Table 15. Equations and correlation coefficients for regressions of DTPA extractable soil Mn (uncorrected and pH-corrected) and plant Mn on various soil factors. Regression curves are given in Figs 34 and 38.

Variables Regression Correl. Regr.

Y coeff. curve Soil Mn (uncorr.) pH 3538log y = -3.075 + 1.684 x - 0.149 x2 0.695*** a Plant Mn pH 3538log y =4.900 - 0.716 x + 0.0382 x2 0.644*** b Plant Mn/soil Mn pH 3538log y =7.600 - 2.200 x + 0.1640 x2 0.513*** pH-corr. soil Mn pH 3538log y =3.892 - 0.486 x + 0.0126 x2 0.781*** d

Soil Mn (uncorr.) TI 3536 y= 35.2 - 0.291 x + 0.0055 x2 0.114*** e pH-corr. soil Mn TI 3536 y= 59.1 - 1.384 x + 0.0165 x2 0.098*** f Plant Mn TI 3536 y= 339 - 9.44 x + 0.0860 x2 0.218***

Soil Mn (uncorr.) Org. C 3538 log y = 1.358 + 0.492 log x - 0.372 (log x)2 0.312*** h pH-corr. soil Mn Org. C 3538 log y = 1.218 + 0.595 log x - 0.0256 (log x)2 0.329*** Plant Mn Org. C 3538 log y = 1.865 + 0.187 log x - 0.0848 (log x)2 0.148*** j

Soil Mn (uncorr.) CEC 3538 y = 27.0 + 0.490 x - 0.00599 x2 0.056*** k pH-corr. soil Mn CEC 3538 y= 39.9 + 4.712 log x - 5.973 (log x)2 0.044** Plant Mn CEC 3538 Y 745 - 817.8 log x + 251.7 (log x)2 0.267***

Soil Mn (uncorr.) El. cond. 3538 log y = 1.441 - 0.549 log x + 0.182 (log x)2 0.389*** pH-corr. soil Mn El. cond. 3538 log y = 1.348 - 0.807 log x + 0.395 (log x)2 0.446*** o Plant Mn El. cond. 3538 log y = 1.965 - 0.695 log x + 0.392 (log x)2 0.491*** Soil Mn (uncorr.) CaCO3 eq. 3538 log y = 1.313 - 0.227 log x + 0.0356 (log x)2 0.623*** pH-corr. soil Mn CaCO3 eq. 3538 log y = 1.088 - 0.240 log x + 0.0225 (log x)2 0.720*** Plant Mn CaCO3 eq. 3538 log y = 1.724 - 0.123 log x + 0.054 (log x)2 0.666*** but owing to the small number of samples at the highest organic carbon levels the regression curves do not reflect this tendency. Cation exchange capacity has a relatively weak effect on DTPA extractable soil Mn (Fig. 38, curves k, 1). Correction for pH slightly altered the direction of the regression curve, bringing it somewhat closer to the plant Mn-CEC regression curve (m). As electrical conductivity and CaCO3 equivalent of soils increased the Mn contents of plants as well as the extractable soil Mn decreased (Fig. 38, curves n, o, p, q, r, s). Correlations between Mn and these soil characteristics are strong, but somewhat lower than those between Mn and pH. Correction of DTPA extractable soil Mn for pH changed the relationship between soil Mn and both these soil factors slightly but distinctly toward the respective relations between plant Mn and these soil factors. The above relations of plant Mn and DTPA extractable soil Mn to the six soil characteristics may be summarized as follows: As in the case of other micronutrients, the original plants (maize and wheat) due to their heterogeneity were poor indicators of the Mn status of soils. Determination of soil Mn by DTPA extraction gave a much higher correlation between plant Mn and soil Mn than did the AAAc-EDTA extractable Mn. Both plant Mn and DTPA extractable soil Mn were significantly correlated with all six soil characteristics studied. Correction of DTPA extractable soil Mn for any of the six soil factors would improve the correlation between plant Mn and soil Mn. (Correction coefficients for the other fivesoilfactors were also calculated analogously to the procedure for correction for pH). The effects of these corrections, however, were small compared to that of pH correction. Therefore, these are not presented. 72 Correction of DTPA extractable soil Mn for pH improved the plant Mnsoil Mn correlation substantially. It also moderated the differences between the effects of the other five soil factors on plant Mn and soil Mn. When the results of this study are classified by countries (Part II), both uncorrected and pH-corrected soil Mn values are presented. The extractable Mn contents of soils in the whole material of this study were analysed by two extraction methods (DTPA and AAAc-EDTA). The main differences between these methods as well as their usefulness are discussed in the next Section.

AAAc-EDTA extractable soil Mn as an indicator of Mn status of soils. The very poor correlation between the Mn contents of plants and AAAc-EDTA extractable soil Mn (r = 0.039*) is mainly due to the predominant role of pH in plant and soil Mn analyses. The Mn contents of plants decreased strongly with rising pH (Fig. 39 b). The AAAc-EDTA extractable soil Mn increased with rising pH to about pH 7 and decreased thereafter toward alkaline soils (Fig. 39 a). Both of the above relations are very firm (R values 0.644*** and 0.404***, respectively). Thus, in acid soils the plant Mn seems to be negatively correlated, and in alkaline soils positively correlated with the AAAc-EDTA extractable soil Mn.

Q log y =-3.142+1.643 x -0.120 x2 b log y = 4.900-0.716 x +0.0382 x2 2 400 n=3538 n=3538 R=0.404" aE 500 R=0.644"

400 300 -5

"E,. 300 E'200 o CS: 210000 0 100 o Od111111111111 6 6 Soi pH Sol pH

log y =7.670-2.275 x .0.152 x2 log y=4.834-0.594x .0.0289 x2 4C 1.- 1200 n=3538 10 o) n=3538 R=0.729" R=0.547" c 1000 8 .92 Y .. - 800 á, c 6 S X Ill c -0 600 a.,

a, ! 400 u 2 Y 200 I T200a 111111iiiiiiii1Illiiiii Soil pH Soil pH Fig. 39. Relationships of AAAc-EDTA extractable soil Mn and plant Mn to soil pH(CaC12). Graph a: Soil Mn as a function of pH. Graph b: Plant Mn as a function of pH. Graph c: Ratio of plant Mn to soil Mn as a function of pH. The regression curve also indicates the coefficient, k(pH), for correcting soil Mn for pH (right-hand ordinate). Graph d: pH-corrected soil Mn as a function of pH. Columns and points indicate mean values of each pH class; n values of pH classes are given in Graph a. 73 This AAAc-EDTA extractable Mn-pH relationship is basically of the same form as that between DTPA extractable Mn and pH. There is, however, an essential difference. The DTPA extractable Mn contents reached a maximum at pH 5-6 while the maximum for AAAc-EDTA Mn was reached first at about pH 7. This means that DTPA extractable soil Mn is positively correlated with plant Mn over a wider pH range than the AAAc-EDTA Mn. This also explains the poor correlation between plant Mn and AAAc-EDTA extractable soil Mn in this material consisting predominantly of soils approaching neutrality or alkalinity. As in the case of DTPA extractable soil Mn, the correction coefficient for pH was calculated for AAAc-EDTA Mn. The ratio of plant Mn to AAAc-EDTA extractable soil Mn is given in Fig. 39 c. The regression equation in Graph c, multiplied by the reversion coefficient (2.642), gives the pH correction coefficient for AAAc-EDTA extractable soil Mn:

k(pH) = 108.092 - 2.275 pH + 0.152 pH2

The pH correction coefficient is also given in Graph c (right-hand ordinate). The pH-corrected soil Mn as a function of pH is given in Graph d. This relation closely resembles the one between plant Mn and pH (Graph b) as well as that between pH-corrected DTPA Mn and pH (Fig. 34 d). Application of the pH correction to AAAc-EDTA extractable soil Mn values in the whole material (n = 3538) improved the plant Mn-soil Mn correlation from an r value of 0.039* (uncorrected) to 0.588***. At the national level the respective correlations were (for uncorrected AAAc-EDTA Mn) non-significant or even negative in 22 out of 29 countries (Table 16). After pH correction, the correlations remained non-significant in two countries only (Malta and Iraq) and became highly significant in 20 countries.

Table 16. Correlations (r values) between plant Mn content and AAAc-EDTA extractable soil Mn in various countries. Soil Mn analyses are both uncorrected and corrected for pH [k(pH) = 108.092 - 2.275 pH + 0.152 pH2 ]. Correlations of the best fitting regressions are given. Regression models: a) y = a + bx; b) y = a + b log x; c) log y= a+ bx, and d) log y = a + b log X.

Uncorrected pH -corrected Uncorrected pH -corrected Country n Regr. Regr. Country n Regr. Regr. model r model r model r model r Belgium 36 b -0.099 n.s. 0.645*** Egypt 198 d 0.668*** d 0.666*** Finland 90 b -0.117 n.s. a 0.603*** Iraq 150 a 0.024 n.s. b 0.022 n.s. Hungary 201 d 0.028 n.s. d 0.720*** Lebanon 16 b 0.500* b 0.502* Italy 170 d 0.185* d 0.462*** Syria 38 d 0.351* d 0.352* Malta 25 b -0.376 n.s. b -0.375 n.s. Turkey 298 d 0.200*** d 0.261*** New Zealand 35 d 0.245 n.s. d 0.439*** Ethiopia 125 b-0.123 n.s. a 0.572*** Argentina 208 c 0.150* d 0.565*** Ghana 93 a -0.027 n.s. a 0.722*** Brazil 58 d 0.211 n.s. d 0.553*** Malawi 97 c 0.094 n.s. d 0.718***

Mexico 242 d 0.066 n.s. a 0.659*** Nigeria - 153 a -0.024 n.s. d 0.648*** Peru 68 c 0.124 n.s. a 0.867*** Sierra Leone 48 b 0.018 n.s. b 0575*** Tanzania 163 a -0.317*** c 0.204** India 258 c 0.075 n.s. 0.151* Zambia 44 d -0.075 n.s. d 0.556*** Korea, Rep. 90 d 0.150 n.s. 0.895*** Nepal 35 c -0.184 n .s. ab 0.496** Pakistan 237 b 0.161* a 0.186** Philippines 194 a 0.076 n.s. a 0.754*** Sri Lanka 18 c 0.312 n.s. 0.773*** Thailand 150 c -0.194* a 0.632*** Whole material 3538 c 0.039* d 0.588*** 74 Comparison of correlations between plant Mn and AAAc-EDTA Mn corrected for pH, given in Table 16, with respective correlations between plant Mn and DTPA Mn (uncorrected and pH-corrected, Table 14) shows that plant Mn correlated best with pH-corrected DTPA Mn. The r value for the whole material is 0.713*** and in28out of 29countries the r values exceed those obtained by any other method. The second highest correlations were obtained with pH-corrected AAAc-EDTA and the third highest by uncorrected DTPA, r values0.588***and0.552***for the whole material, respectively. At the national level the former is better in21and the latter in 7 countries. Comparison of relations between extractable Mn contents of soils obtained by different methods (Table 17) shows that the two methods that correlated best with the Mn contents of plants (pH-corrected DTPA and pH-corrected AAAc-EDTA) also have the highest mutual correlation.

Table 17. Mutual correlations (r values) between extractable Mn contents of soils in the whole material (n 3538) as determined by four methods.

Extraction AAAc-EDTA pH-corrected pH-corrected method (uncorrected) AAAc-EDTA DTPA DTPA (uncorrected) 0.416*** 0.672*** 0.751***

pH-corrected DTPA 0.159*** 0.840***

pH-corrected AAAc-EDTA 0.415***

It is clear that the essential difference between the two basic extraction methods, DTPA and AAAc-EDTA, liesintheir divergent responsestosoil pH. Therefore, the relationship of each of the two methods to pH mainly determines their usefulness for estimating the Mn status of soils. The difference between the methods, however, can be largely eliminated by correcting the analytical data for soil pH. Soil analysis giving results that are in harmony with the results of plant analysis is naturally a more reliable index of the availability of a nutrient to a plant than a method giving contradictory results. Often a method which is well suited for one nutrient fails in the case of another. Nevertheless, the above results with Mn show that even a method which in itself gives poor results may be successfully used if its special features and its behaviour with respect to various soil characteristics are known and properly taken into account. 75

2.3.7 Zinc

2.3.7.1 General aspects The average Zn contents of the three types of indicator plant in the whole international material, those of the respective soils determined by two extraction methods (AAAc- EDTA and DTPA) and the plant Zn-soil Zn regressions are given below. Zinc content in plant DM in resp. soils Indicator mean + s mean ± s Regression of plant Zn (y) Correlation') plant (PPIn) (mg/I) on soil Zn (x) (r)

AAAc-EDTA Original maize 1966 35.7 ± 47.2 4.13 ± 8.62 log y = 1.39 + 0.211 log x 0.391*** Original wheat 1768 27.4 ± 11.3 3.48 ± 7.72 y = 23.3 + 12.62 log x 0.405*** Pot-grown wheat 3537 18.3 ± 10.8 3.81 ± 8.26 y = 15.0 + 0.87 x 0.665***

DTPA Original maize 2.14 ± 6.49 log y = 1.46 + 0.195 log x 0.381*** Original wheat (as above) 1.77 ± 7.30 log y = 1.42 + 0.181 log x 0.473*** Pot-grown wheat 1.97 ± 7.05 log y = 1.23 + 0.321 log x 0.732***

I) The mutual correlation between AAAc-EDTA extractable and DTPA extractable soil Zn was high (r = 0.819***), and the differences between their relations to plant Zn were small. In order to get more information on their usefulness, the relations of both methods to various soil factors were investigated.

The difference between the Zn contents of the original maize and original wheat may be partly due to maize being better able than wheat to absorb Zn from soils and partly to the higher contents of available Zn in maize soils. The average Zn content of the pot-grown wheat is one-third lower than that of the original wheat plants. Evidently the plants were unable to absorb from the small amounts of soil in pots (see Section 1.2.6) as much Zn as plants grown in field conditions. The standard deviation of Zn contents of maize (± 47 ppm) was much wider than for wheat (about 11 ppm). The variation of extractable soil Zn was considerably wider than in the cases of other micronutrients. As for the other micronutrients, the correlations between the Zn content of pot-grown wheat and extractable soil Zn were much closer than those for the original maize and wheat. The former regressions (with national mean values) are shown in Figs 40 and 41. The correlations of Zn uptake with AAAc-EDTA and with DTPA extractable soil Zn were very similar to those given in Figs 40 and 41, with r values of 0.669*** and 0.707***, respectively. At the national level, too, the differences were small.

2.3.7.2 Soil factors affecting the zinc contents of plants and soils The regressions of plant and soil Zn on six soil characteristics are given in Figs 42 and 44 and in Table 19. The correlations were always highly significant with r varying from 0.055*** to 0.475***. In general, the organic carbon content of soils pH were better correlated with Zn than were the other soil factors studied. As in the case of the other micronutrients, correction coefficients were calculated for each of the six soil factors studied, as well as for the two extraction methods. The coefficients were tested over the whole soil and plant material. Possibly due to the relatively good agreement between the plant Zn and uncorrected (AAAc-EDTA and DTPA) soil Zn values, the corrections were less effective than in the cases of some other micronutrients, e.g.Mn and Mo. 76

200

100 90 80 E 70 60

4-750 -cCD 40

o° 20

10 9 8 7 6

.2.3 .4.6 .8 1 23 46 810 20 3040 6080 Zn in soil, mg/1 (.4 A Ac- EDTA)

Fig. 40. Regression of Zn content of pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Zn (x) for the whole international material. National mean values of plant and soil Zn are also given. Countries: Ar = Argentina, Be = Belgium, Br = Brazil, Eg = Egypt, Et = Ethiopia, Fi = Finland, Gh =- Ghana, Hu = Hungary, In = India, Ir = Iraq, It= Italy, Ko = Korea, Rep., Le = Lebanon, Mt = Malta, Mw= Malawi, Me = Mexico, Ne = Nepal, NZ = New Zealand, 'Ni,---- Nigeria, Pa = Pakistan, Pe = Peru, Ph = Philippines, Si = Sierra Leone, Sr = Sri Lanka, Sy 7------Syria, Ta = Tanzania, Th = Thailand, Tu = Turkey, Za = Zambia. The Zn content zones IV are explained in Section 1.4.3.

Compared to the other soil factors studied, soil pH caused the greatest difference between AAAc-EDTA extractable soil Zn and plant Zn. In the following paragraphs, the effects of pH and those of the other five soil characteristics are discussed. Effects of soil pH on soil Zn and plant Zn. Comparison of the regression lines and columns in Graphs a, b and e (Fig. 42) shows that the correspondence between the effects of pH on plant Zn and DTPA Zn is greater than for the effects of pH on plant Zn and AAAc-EDTA Zn. The contents of plant Zn decreased with rising pH over almost the whole pH range. DTPA Zn decreased from about pH 5 upwards, while in the case of AAAc-EDTA, Zn values increased up to about pH 6 and only began to fall in the alkaline pH range. pH thus had a similar influence on plant Zn and DTPA Zn over a relatively wide pH range, whereas the effects of pH on plant Zn and AAAc-EDTA Zn resembled each other within only a rather narrow pH range on the alkaline side. There- 77

IMM ... __ 200 1

100111111111111 1111 1111111 90 80 70millimil mom 60 50 40lionnolimmi 30

2011111111111111111111111

10111111111111111111111111III 9IIIIMPAMMENIIIII=MMINI MM=MEMBII I 8 7iirMilirilli MMEM"" 6 1..1.11"1.. E EL IL .2.3 .4.6 .81 23 46 810 20 3040 60 100 Zn in soil, mg /I (DTPA)

Fig. 41. Regression of Zn content of the pot-grown wheat (y) on DTPA extractable soil Zn (x) for the whole international material. For abbreviations see Fig. 40. fore, the pH correction coefficient calculated for DTPA Zn was ineffective, but that calculated for AAAc-EDTA Zn improved the plant ZnAAAc-EDTA Zn correlation. The ratio of plant Zn to AAAc-EDTA Zn as a function of pH (Fig. 42c) indicates quantitatively the systematic difference between the pH effects on the two forms of Zn, and the regression curve (c) shows the correction required to eliminate this difference. The correction coefficient for pH is given by:

k(pH) = 0.1903 X 104-1 0.976 pH + 0.068 pH2 where the exponent is the regression equation for curve c (Fig. 42 and Table 19) and the constant (0.1903) is a reversion coefficient for restoring the AAAc-EDTA Zn values to their original level. The simplified formula is given by:

k(pH) = 103.379 0.976 pH + 0.068 pH2 The numerical values for k(pH) can also be read from the right-hand ordinate of Fig. 42c and the regression line. The k(pH) applies to pH(CaC12). For the relation between pH(H20) and pH(CaC12) see Fig. 5 and Section 2.1. When the AAAc-EDTA Zn values are corrected for pH in the whole material, we obtain the soil ZnpH relation given in Fig. 42d. This curve shows that the effects of pH on plant Zn and on AAAc-EDTA Zn have been made almost equal. Application of k(pH) to AAAc-EDTA Zn values improved the plant Znsoil Zn correlation from an r value of 0.665*** (Fig. 40) to 0.706*** (Fig. 43) in the whole material and in 23 out of 29 countries (Table 18). The highest correlations, however, were 78

'le 10 g5 N

Soil pH Soil pH

4 5 8 Soil pH Soil pH Fig. 42. Relationships of extractable soil Zn and plant Zn to soil pH(CaC12). Graph a: AAAc-EDTA extractable soil Zn as a function of pH. Graph b: Plant Zn as a function of pH. Graph e: Ratio of plant Zn to AAAc-EDTA Zn as a function of pH. The regression curve also indicates the coefficient, k(pH), for correcting AAAc-EDTA Zn for pH (right-hand ordinate). Graph d: pH-corrected AAAc-EDTA Zn as a function of pH. Graph e: DTPA extractable soil Zn as a function of pH. Columns and points indicate mean values of each pH class, n values of pH classes are given in Graph b. Equations and correlation coefficients for regression curves are given in Table 19.

obtained between plant Zn and DTPA extraction (Fig. 41 and Table 18). The r value for the whole material is 0.732*** and in national materials the best correlation was obtained in 15 out of 29 countries. In general, the pH-corrected AAAc-EDTA extractable Zn was better correlated with plant Zn in countries where acid soils predominate and DTPA Zn in countries with alkaline soils. In the case of Zn, the main difference between the DTPA and AAAc-EDTA extraction methods in favour of DTPA lies in the different reaction to pH. This difference can largely be eliminated by correcting AAAc-EDTA Zn for pH. In fact, correcting AAAc-EDTA Zn for pH raised the coefficient of correlation (r) between these two extraction methods from 0.819*** to 0.911***.

Effects of other soil factors in soil Zn and plant Zn. Figure 44 shows the relationships between uncorrected and pH-corrected AAAc-EDTA Zn (left-hand graphs), plant Zn (middle graphs), and DTPA Zn (right-hand graphs) and five soil characteristics. Soil texture. The patterns of behaviour of Zn as a function of texture are quite similar irrespective of the method by which the Zn values have been determined (curves f, g, h, i, Fig. 44 and Table 19). With increasing texture index, the Zn values first decrease, reaching a minimum among medium textured soils and showing thereafter a tendency to increase towards fine textured clay soils. The effect of texture seems to be quantitatively more pronounced on extractable soil Zn values (f, g, i) than on plant Zn (h). Apparently 79

Table 18. Correlations (r values) in various countries between plant Zn and soil Zn determined by three different methods. Correlations of the best fitting regressions are given. Regression models: a) y = a + bx; b) y = a + b log x; c) log y = a + bx; d) log y = a + b log x.

AAAc-EDTA-extractable DTPA-extr. AAAc-EDTA-extractable DTPA-extr. Uncorrected pH-correctedl)Uncorrected Uncorrected pH-correctedi)Uncorrected Country n Regr. Regr. Regr. Country n Regr. Regr. Regr. model r model r model model model r model r Belgium 36 d 0.803*** a 0.895*** d 0.829*** Egypt 198 d 0.739*** b 0730*** d 0.784*** Finland 90 d 0.595*** a 0.768*** d 0.645*** Iraq 150 a 0.672*** a 0.660*** a 0705*** Hungary 201a a394*** b 0.540*** b 0.581*** Lebanon 16 d 0.715** d 0.709** d 0.792*** Italy 170d 0.751*** b 0.795*** d 0.814*** Syria 38 a 0.843*** a 0.840*** b 0.881*** Malta 25d 0.916*** b 0.941*** b 0.896*** Turkey 298 a 0.507*** a 0.536*** b 0.712*** New Zealand 35d 0.678*** b 0.684*** b 0.712*** Ethiopia 125 d 0.773*** a 0.846*** d 0.788*** Argentina 208a 0.641*** a 0.676*** a 0.635*** Ghana 93 a 0795*** b 0.827*** b 0.805*** Brazil 58a 0.778*** a 0.825*** a 0.698*** Malawi 97 d 0.542*** d 0.607*** d 0.625*** Mexico 242d 0.615*** b 0.651*** b 0.658*** Nigeria 153 d 0.758*** a 0.873*** a 0.869*** Peru 68a 0.699*** a 0.760*** b 0.752*** Sierra Leone 48 a 0.900*** a 0.893*** a 0.936*** Tanzania 163 d 0.604*** b 0.630*** b 0.669*** India 258a 0.556*** a 0.560*** a 0.584*** Zambia 43 d 0.693*** b 0.787*** b 0.752*** Korea, Rep. 90a 0.758*** d 0.815*** d 0.761*** Nepal 35a 0.626*** b 0.591*** b 0.666*** Pakistan 237d 0.682*** b 0.706*** b 0.734*** Philippines194d 0.702*** a 0.827*** a 0.812*** Sri Lanka 18a 0.943*** a 0.948*** a 0.944*** Whole Thailand 150d 0.660*** d 0.761*** d 0.842*** material 3537 a 0.665*** a 0.706*** d 0.732***

I) k(pH) = 103.379 - 0.976 pH + 0.068 pH2

200 Fig.43.Regression of Zn content of the pot- grown wheat (y) on pH- corrected AAAc-EDTA extractable soil Zn (x) 100 for the whole material. 90 Forabbreviationssee 80 Fig. 40. 70 60 E 50 o.

4-6a, 20

N 10 9 8 7 6IMO NEM

.2.3 .4.6 .8 1 2 3 46 8 10 20 3040 60 100 pH-corrected Zn in soil, mg/1 (AAAc-EDTA) 80

10 30 50 70 go 10 30 50 70 90 0 30 50 70 90 Texture index Ti) Texture hdex (TO Texture index Ti)(

.8 16 32 64 Organic C,%

8 16 32 64 8 16 32 64 CEC.me 100g CEC,me/100g

TD"

16 6.4 25.6 16 6.4 256 El. conductivi y,10- S cm-1 El. conductivity, 10" S cm"' El. conduc ivi y,10- S cm-1

.4 16 6A zas 1.6 6A 256 .4 16 6.4 25 6 COCO equiv.,/o CaCO equiv. % CoC0 equiv.,% Fig. 44. Relationships of plant Zn and soil Zn to various soil factors. Graphs on the left: Uncorrected AAAc-EDTA extractable Zn (wide columns and curves f,j,n,r, v) and pH-corrected AAAc-EDTA Zn (narrow columns and curves g, k, o, s, w) as a function of five soil characteristics. Graphs in the centre: Plant Zn as a function of five soil characteristics. Graphs on the right: DTPA extractable soil Zn as a function of five soil factors. The number of samples (n) within each class is given in the centre graphs. The equations and correlation coefficients for regressions are given in Table 19. See also footnote to Fig. 24. 81

Table 19. Equations and correlation coefficients for regressions of AAAc-EDTA extractable soil Zn (uncorrected and pH-corrected), plant Zn, and DTPA extractable soil Zn on various soil factors. Regression curves are given in Figs 42 and 44.

Variables Regression Correl. Regr.

Y coeff. curve Soil Zn (AAAc-EDTA, uncorr.) pH 3537 log y = -5.238 + 14.00 log x - 8.713 (log x)2 0.128*** a Plant Zn pH 3537 log y =0.960 + 0.177 x - 0.0204 x2 0A75*** b Plant Zn/soil Zn (AAAc-EDTA) pH 3537 log y =4.100 - 0.976 x + 0.0680 x2 0.402*** Soil Zn (AAAc-EDTA, pH-corr.) pH 3537 log y =1.660 - 0.327 x + 0.0181 x2 0.267*** d Soil Zn (DTPA) pH 3537 log y = -2.971 + 1.101 x - 0.0969 x2 0.378*** e

Soil Zn (AAAc-EDTA, uncorr.) TI 3535 y 46.50 - 51.63 log x + 15.43 (log x)2 0.089*** f Soil Zn (AAAc-EDTA, pH-corr.) TI 3535 y=66.02 - 76.14 log x + 23.04 (log x)2 0.118*** Plant Zn TI 3535 log y = 1.439 - 0.00883 x + 0.000075 x2 0.184*** h Soil Zn (DTPA) TI 3535 y=31.07 - 34.33 log x + 9.990 (log x)2 0.082*** i

Soil Zn (AAAc-EDTA, uncorr.) Org. C 3537 log y =0.324 + 0.529 log x - 0.141 (log x)2 0.343*** j Soil Zn (AAAc-EDTA, pH-corr.) Org. C 3537 log y =0.288 + 0.673 log x + 0.0449 (log x)2 0.434*** k Plant Zn Org. C 3537 log y = 1.202 + 0.210 log x + 0.0725 (log x)2 0.286*** 1 Soil Zn (DTPA) Org. C 3537 log y = -0.0731 +0.735 log x 0.423***

Soil Zn (AAAc-EDTA, uncorr.) CEC 3537 log y = -1.153 + 1.970 log x - 0.629 (log x)2 0.224*** Soil Zn (AAAc-EDTA, pH-corr.) CEC 3537 log y = -1.069 + 1.916 log x - 0.644 (log x)2 0.176*** o Plant Zn CEC 3537 log y =1.128 + 0.193 log x - 0.0925 (log x)20.075*** Soil Zn (DTPA) CEC 3537 log y =0.929 + 1.229 log x - 0.417 (log x)2 0.098*** q

Soil Zn (AAAc-EDTA, uncorr.) El. cond.3537 log y =0.280 + 0.429 log x - 0.263 (log x)2 0.261*** Soil Zn (AAAc-EDTA, pH-corr.) El. cond.3537 log y =0.292 + 0.170 log x - 0.110 (log x)2 0.099*** Plant Zn El. cond.3537 log y =1.230 - 0.0723 log x 0.135*** t Soil Zn (DTPA) El. cond.3537 log y = -0.0407+0.0924 log x - 0.170 (log x)20.089*** u

Soil Zn (AAAc-EDTA, uncorr.) CaCO3eq. 3537 log y =0.338 + 0.0169 (log x)2 0.055*** Soil Zn (AAAc-EDTA, pH-corr.) CaCO3eq. 3537 log y =0.206 - 0.00801 logx +0.0555 (log x)2 0.252*** w Plant Zn CaCO3eq. 3537 log y = 1.174 - 0.0599 log x +0.00717 (log x)20.433*** Y Soil Zn (DTPA) CaCO3eq. 3537 log y = -0.0896+ 0.129 log x -0.00995 (log x)20.340*** this is due to the wider range of variation in soil Zn than in plant Zn contents. The shape of the curves may be the result of two opposing factors, the increasing total contents of Zn and decreasing relative solubility of Zn towards finer textured soils (Sillanpää 1962 a, b). Correction of AAAc-EDTA Zn for pH had a relatively small effect on the Zn- texture relationships (f, g). As the organic carbon content of the soil increased, there was a simultaneous increase in all four regression curves (Fig. 44, j, k, 1, m). All four correlations are relatively close (Table 19). There seems to be a tendency for extractable soil Zn to decrease at an organic C content of about 0.4 percent (see columns). This, however, is not confirmed by the regressions, probably because of the small number of samples in the highest organic C classes and the general lack of organic soils in this material. Correction of AAAc-EDTA Zn for pH steepened the slope of the regression curve (j, k) but did not otherwise change the character of the relationship. Cation exchange capacity had only moderate effects on plant and soil Zn (Fig. 44 n, o, p, q and Table 19). With increasing CEC the Zn values firstincreased but the curves turn slightly downwards at high CEC levels. The maximum for plant Zn was reached at a somewhat lower CEC level than in the case of extractable soil Zn. Correction of AAAc-EDTA Zn for pH affected only slightly the Zn-CEC relationship. 82 As the electrical conductivity of soil increased, the Zn contents of plants decreased (curve t), while extractable soil Zn values show at first a tendency to increase and begin to decrease only at higher levels of electrical conductivity. This tendency is more pronounced in the case of AAAc-EDTA Zn (curve r), but was moderated by pH correction (curve s). With the increasing CaCO3 equivalent of the soil, there was a general tendency for a slight decrease in the values of plant Zn and DTPA Zn (curves y and z) while the other two curves (v and w) are more independent of CaCO3. The highest CaCO3 equivalent class with its extremely high Zn values was an exception to this tendency. This class, however, was very small (n = 38) and two-thirds of its samples came from a very restricted area (Malta). The mutual relations between plant Zn and extractable soil Zn, as well as the influence of various soil factors on Zn, are summarized as follows: Replacing the original plants (maize and wheat) by pot-grown wheat improved the plant Znsoil Zn correlations. Both the AAAc-EDTA and DTPA extraction methods for determining soil Zn gave good indexes of Zn availability to plants. The highest correlations between plant Zn and extractable soil Zn were obtained when DTPA extraction was used. Both plant Zn and extractable soil Zn contents were significantly affected by all six soil characteristics studied. Most of the difference between the AAAc-EDTA and DTPA extraction methods in favour of DTPA is attributable to their different reactions to soil pH. Correction of the AAAc-EDTA Zn values for pH largely eliminated the above difference and improved the plant ZnAAAc-EDTA Zn correlations. The plant ZnDTPA Zn and the plant ZnpH-corrected AAAc-EDTA Zn correlations (r) for the whole international material were 0.732*** and 0.706***, respectively. The former were higher in 15 countries where alkaline soils predominate, and the latter in 14 countries with predominantly acidic soils. When the results of this study are classified by countries (Part II), both pH-corrected AAAc-EDTA and DTPA extractable soil Zn values are presented. 83 2.4 Mutual relations between micronutrients

The contents of various micronutrients in plants vary with plant species and varieties and depend on the total micronutrient supply in soils and on factors controlling their availability to plants. Although there might be some differences between plant species in their reactions to soil factors controlling micronutrient availability, it is reasonable to assume that soil factors limiting the availability of micronutrients to one plant species react similarly in cases of other species. The mutual relations between the plant contents of six micronutrients presented (Figs 45 and 46) were obtained from a pot trial in which wheat, cv. 'Apu', was used as an indicator plant. Respective relations between the six micronutrients in soils (r values) are given in Table 20. As shown earlier, the micronutrient contents of plants depend both on the micronutrient contents of soils and on soil factors regulating their availability. The effects of these factors vary considerably from one micronutrient to another as well as in their relative degree of efficacy. In general, when there is good correlation between the plant contents of two micronutrients, the availability of these micronutrientsislargely controlled by the same soil factor or factors. Because of the large number of soil factors involved, these relationships are complicated but in some cases the reasons for correlations can be explained with a relatively high certainty. The best example of these is the highly significant negative correlation (r = 0.515***) between Mo and Mn. The regressions are given in Fig. 46. The availability of both these micronutrients is so strongly affected by soil pH that the other factors are overshadowed. While the Mn contents of plants decrease greatly with rising pH (R = 0.644***, Fig. 34 b), the Mo contents increase (R = 0459***, Fig. 16 b) and deficiencies of both Mn and Mo can therefore hardly exist in same soil. A deficiency of Mn is often combined with an excess of Mo and vice versa, as will be seen in Part II. Extractable soil Mn and Mo are highly significantly negatively correlated (r = 0.390***) if both are corrected for pH (Table 20). Without pH corrections this relationship would be equally significant (r = 0.391***) but positive,i.e.in strong contradiction to the plant MnMo correlation. The second highest correlation found between various plant micronutrients is that of Mn and Zn (r = 0.420***, Fig. 46). In this case also the soil pH exercises the leading effect on the availability of these micronutrients to plants. The plant contents of both these micronutrients decrease strongly with rising pH (Fig. 34 b and 42 b). Comparing the results given in Fig. 38 (graphs on the right) and Fig. 44 (graphs in the centre) it can be seen that there are also other soil factors e.g. texture and organic carbon content, which correlate with plant Mn much as they do with plant Zn. Therefore, in addition to good correlations (r values 0.346*** and 0.299***; Table 20) between extractable soil Mn and Zn, these factors may contribute to the good plant MnZn correlation. For the three next highest correlations between the plant contents of Fe and Zn (r = 0.350***), Fe and Cu (r = 0.344***), and Fe and Mn (r = 0.329***) the common soil factors are not as clearly defined, since the Fe contents of plants were less strongly affected by any of the soil factors studied than were the other micronutrients (see Fig. 32 and related text). In all three cases, FeZn, FeCu, and FeMn, the extractable soil contents are correlated to a degree similar to the plant contents (r values 0.380***, 0.270*** and 0.222***, respectively; Table 20). Plant B was best correlated to plant Mo and Mn, in the former case positively (r = 0.329***) and in the latter negatively (r = 0.244***). Respective correlations 84

Cu =5.45+210 log B 16 log B=0.675+0.00898 Cu 6 log 8=0.723+0.000244 Fe 15 r =a137." r =0.137. r =0.025 n.s.

16 g 10 15 Cu 40 80 160 Fe Fe Fe Cu Fe .59.0+2.59 log B tog Fe=1.67+0.0148 Cu Cu =-711+797 tog Fe 160 r .0.025n.s. 160 r =0.344** 5 r =0.344."

10 80 80

40 40

16 g 10 5 Cu 40 80 160 F Mn Mn Mn 512log Mn = 2.25-0506 log B 512log Mn =2.00-0.154 log Cu 512 log Mn =0.0484+1.033 log Fe r =-0.244". r=-0.076" r =0.329**

. 128 128 128

32 32 32

6B 10 15 40 80 160 Fe Moj Mol Mo Mo =-0.204+0.713 log B log Mo.-1.26+0.668 log Cu Mo =0.217+0.00172 Fe r =0.329." . r =0.243*. .8 r =0.082." . ,

.05 .05 .05

8 16 B 10 15 40 80 160 Fe Zn Zn Zn log Zn=1.35-0.186 log B log Zn=1.09+0.0170 Cu log Zn =0.107+0.625 log Fe 40 40 40 r=-0.157** r =0.221". r =0.350."

20 O 20

10 10 10

16B 5 10 15 Cu 40 80 160 Fe Fig. 45.

Figs 45 and 46. Mutual correlations between B, Cu, Fe, Mn, Mo, and Zn contents of pot-grown wheat (ppm) in the whole international material (n = 3536).Points indicate arithmetic mean contents of the micronutrients given on the y axis in content classes of the micronutrient given in the x axis. 85

16log 8=0.959-0.118 log Mn 16 log B=0.689+0.152 Mo 16 log B =0.900-0.133 log Zn r =-0. 244 r =0.329*** r =-0.157***

32 128 512Mn .05 0.2 0.8 Mo 10 20 40 Zn Cu Cu log Cu =0.884-0.0376 log Mn log Cu =0.877+0.0884 tog Mo Cu =3.53+2.86 log Zn r =-0.076*** 15 r =0.243*** 15 r =0.2 21"*"

10 10

32 28 512 mn 0.8 mo 10 20 40 zn Fe Fe Fe log Fe =1.57+0.105 log Mn Fe =59.7+3.90 Mo log Fe =1.53+0.196 log Zn 160 r =0.329'*. 160 r =0.082 '** 160 r =0.350

80 80 80

40 40 40

32 128 512Mn .05 0.2 0.8 mo 10 20 40 zn Mo Mn Mn tog Mo =0.598 -0.698 log Mn 512log Mn=1.61-0.379 log Mo 512log Mn =0986+0.734 log Zn

.8 r =-0.515 . r=-0.515**" r =0.420* . . 128 128

32 32

32 128 512 Mn .05 0.2 0.8 Mo 20 Zn Zn Zn Mo log Zn =0.767+0.238 log Mn log Zn =1.17-0.0589 log Mo log Mo =-0.306 -0.335 log Zn 40 40 r =0.420*** r =-0.140*** 80 r=-0.140***

20 20

10 10

32 128 512Mn .05 0.2 0.8 Mo 10 20 40 zn Fig. 46. 86

Table 20. Mutual correlations (r values) between extractablel) soil micronutrients in the whole material (n = 3536). Correlations of the best fitting regressions are given. Regression models are indicated by letters: a) y = a -1- bx; b) y = a -F b log x; c) log y = a -I- bx; and d) log y= a + b log x.

x B Cu Fe Mn Mo Y Cu c 0.069*** Fe d 0.178*** d 0.270*** Mn d 0.201*** d 0.251*** d 0.222*** Mo d 0.365*** d 0.469*** d 0.037* d 0.390*** Zni c 0.089*** c 0.150*** d 0347*** d 0.346*** d 0.015 n.s. Zn2 d 0.038* c 0.186*** d 0.380*** d 0.299*** d 0.064*** I) Extraction methods used: B: hot water extr. + CEC correction Mo: A0-0A + pH correction Cu: AAAc-EDTA + Org. C correction Zni: DTPA Fe: AAAc-EDTA Zn2: AAAc-EDTA + pH correction2) Mn: DTPA + pH correction 2) The mutual correlation (r) between the two extraction methods for Zn is 0.911***.

(r values 0.365*** and 0.201***) existed between the extractable soil contents of these micronutrients (Table 20). In addition to the CuFe correlation mentioned above, plant Cu correlated well with plant Mo (r = 0.243***) and with plant Zn (r = 0.221***). Especially good correlation was found between soil Cu and Mo (r = 0.469***); those between soil Cu and Zn (r values 0.150*** and 0.186***) were also relatively high. The other correlations between the plant contents of the six micronutrients were highly significant with only one exception (BFe), but compared to those specifically cited above these relations are less firm. In general, taking into account the very small concentrations of micronutrients in soils, the possibility that one micronutrient directly affects the availability of another seems less likely than that their mutual relationships are indirectly determined by other soil factors affecting their behaviour. However, on the basis of the present data the possibility of direct chemical effects between two or more micronutrients cannot be entirely ruled out.

2.5 Micronutrient contents of plants and soils in relation to yields with special reference to the "concentration-dilution" phenomenon Factors determining the yield of a plant are so numerous that the effects of a single micronutrient are almost inseparable from those of other factors simultaneously affecting the growth of the plants used in this study. However, there are certain matters which must be taken into consideration when dealing with yields of pot-grown wheat and their relationships to micronutrients. Samples of the plants acting as indicators had to be harvested at an early stage of growth when they had only developed vegetatively (see Section 1.2). At this time, the early dry matter (DM) yield may not always be a reliable measure of the final grain yield since the shortage of some micronutrients, if not severe, may appear first at relatively late stages of growth, thus limiting the grain more than the vegetative yield. Taking into account the above comment on micronutrient-yield relations, in this section the yields are not presented as functions of micronutrients but vice versa. The micro- 87 nutrient contents of plants and soils as functions of yield are of special interest in connection with the interpretation of the results of plant analyses. For example Cottenie (1980) pointed out that enrichment of a nutrient in a plant may be due to a high nutrient level in the soil or to reduced growth, and Tölgyesi and Mik6 (1977) found that with increasing maize yields the uptake of nutrients increased but their concentration in plant DM decreased. The contents of the six micronutrients in plants and soils as functions of yield are presented in Fig. 47. To obtain a better quantitative illustration of the effects of yields on the micronutrient contents of plants and soils in relation to their

12 10- 9 a) log y=186-0360 log x r=-0194rn 1.2 b) log y=-0801+0178 log x r=0 054". 8 8 8

E .8 7 0. 6 o) o E o 4 a, 6

4 ".6 5 2 o E Ei 2- o 4 log y=0.831-0.000013 x r=-0.020n.s. o 8 o log y=-0.78040.457 log xr=0.117*** o ra o o 811.10 1200 1600 20'00 860 1200 1600 2000 DM yield, mg/pot DM yield, mg/pot

80 300 y=1313-387 log x 3001 r=-0.225' 70 y=216-58.4 log x r=-0.084` 100 E 200 200 ra 60 b,- E .c 50 75 100 o, E 45 100 50 a)y=153 -297 log x r S-0155"a b)y=385-70 7 log x r =-0041' re; o o o a 40 2 800 1200 1600 2000 800 1200 1600 2000 DM yield,mg/pot DM yieldimg/pot

301- .6 a) logy =-928+0 000167xr=0 090'1' b) log y =-1372,0 000364x r=0200"

o o log y=1 26-0 000039 x r=-0 049.* o logy=-11134 342 log x r=0068" -3.5 N

800 1200 1600 2000 eio 1200 1600 2d00 DM yield, mg/pot DM yield, mg /pot Fig. 47. B, Cu, Fe, Mn, Mo, and Zn contents of pot-grown wheat (a-curves and left-hand ordinate) and respective extractable soil micronutrients (b-curves and right-hand ordinate) as a function of yield (n = 3537). The hatched areas indicate the ranges of the standard deviations (y ± s) for both the plant and the soil micronutrient contents. 88 variations, the scales for the graphs have been adjusted so that the ranges of the standard deviations (y- ± s) of the plant and the soil contents are equally wide and are indicated by the hatched areas. (For example, plant B content 6.09 ± 4.80 ppm and soil B 0.73 ± 0.72 mg/1, as in Fig. 25). The behaviour of B, Cu and Zn is similar in that their soil concentrations increase with increasing yield but the concentrations in plants decrease. In the cases of Fe and Mn both the plant and soil contents decrease and the contents of Mo increase with increasing yield. The general direction of the regression curves, however, is of little importance because apart from the single micronutrient in question other factors are involved. The most important information to be drawn from Fig. 47 concerns the mutual relationship between curves a and b. For all six micronutrients, the slopes of the a curves (regressions of plant contents on yield) show a relatively greater rate of decrease or a smaller rate of increase than the respective b curves (regressions of soil contents on yield). In other words, at low yield levels plants absorb more micronutrients per unit of produced plant mass in relation to available soil micronutrients than they do at high levels of yield. The principles of these relations, called here as concentration-dilution phenomenon, can be explained as follows: Restricted plant growth (low yield) in most cases is often due to some factOr other than the micronutrient in question. In such cases micronutrients which are available in soil in high or reasonable quantities tend to concentrate in the slowly produced plant mass, so raising the contents of these micronutrients to a higher level than when plant growth is normal or rapid. Therefore, as in this material, most of the very high plant micronutrient concentrations are to be found in those of restricted growth or in plants grown on soils where the micronutrient in question was abundantly available or, most often, where the effects of both these factors were combined. When the plant growth is rapid (high yields), the plants are apparently unable to absorb micronutrients from soil in quantities related to the mass of DM produced. Consequently, portions of the micronutrients already absorbed are diluted in the increased DM mass and the micronutrient contents of the plants in relation to available micronutrients in soil decrease with increasing yields. It is clear that the concentration-dilution phenomenon is in main principle the same as that prevailing during the growth of any plant in that variations in the micronutrient concentrations of a plant are related to the speed at which that plant produces new DM (see Section 1.2.2). Because the extractable micronutrient contents of soils are independent of plant yield, the differences between the results of plant and soil analyses due to the concentration- dilution phenomenon are to be considered as a source of error affecting the results of plant analyses only. To compare the relative importance of the effects of yield (concentration- dilution) and the extractable soil micronutrient contents on the contents of micronutrients in plants, stepwise multiple regressions including these two variableswere computed. The percentages of variation explained by the two characteristics were:

Percentage of variation explained Characteristic B Cu Fe Mn Mo Zn Soil micronutrient content alone 68.2 53.4 10.5 50.8 48.4 53.6 Soil micronutrient content + yield 69.4 54.6 10.6 51.0 48.5 54.6 89 The above comparisons show the relatively unimportant role of yield in explaining the total variations in plant micronutrient contents in this material. For further, estimationof thequantitativeroleof theconcentration-dilution phenomenon, regressions of plant micronutrient contents on respective soil contents were computed separately for low yields (<1100 mg/pot) and high yields (>1500 mg/pot), each group consisting of about one-fifth of the whole material. Since in case of B the effect of yield was more pronounced than for other micronutrients, these regressions for B are given in Fig. 48 as example. The regression line calculated for high yields (curve e) is located at a level about 25 percent lower than that calculated for low yields (curve b) . T'he locations of curves b and c on both sides of curve a (whole material) give, however, a rough visual picture of one source of variation in the plant Bsoil B correlation, i.e. that of the concentration-diltition phenomenon. On the average, this difference in location of the curves is quantitatively relatively unimportant in comparison with the total variation of plant B contents. However, in cases of some single samples it may be quite substantial as mentioned before.

100 so B rn id 60 -+ 09±4 50 R±s.3±072 40 + x E o_ 30 r= ca. -8: 20 n=672 .c St-±s=5 8±4 7 10 = 5±01 8 4i 60

4,,E, 6 - 5 b 4 3

BI r

.081 .2 .3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 C EC -corrected B in soil, mg/l (I-iot w sot)

Fig. 48. Regressions of B content of pot-grown wheat on hot water soluble soil B corrected for CEC, (a) all samples, (b) low yielding samples, (c) high yielding samples. 90 2.6 Suitability of plant and soil analyses for large scale operations

A combination of plant and soil analyses offers a better means of estimating the micronutrient status of soils than either alone, but the improvement in quality is gained at the cost of quantity and vice versa. A decision to choose one or other of the two techniques must therefore often be made. Plant analysis holds a theoretical advantage over soil analysis in that the micronutrient fractions found in plants have indeed been in the soil in forms available to plants. This advantage is offset by practical difficulties arising with work on a large scale. One such is the need for standardization of the plant material which must meet all requirements (see Sections1.2 and 2.3.1) if comparable analytical data are to be obtained. Uniform standards are almost impossible to meet in practice. Even in small countries more than a dozen important crops are grown; comprehensive interpretation systems have been developed for only relatively few plant species, very little is known about differences between plant varieties, and no general agreement has been reached as to which parts of different plants should be analysed. One of the greatest drawbacks to using plant analysis to diagnose the micronutrient status of soils may be that plant samples, in order to obtain comparable results, must be taken when the plants are at the same physiological age. The sampling time is thus limited to a very short period. Plants grown in the field are more vulnerable to contamination than soil samples. The limitations to the applicability of plant analysis on a large scale do not apply to soil analysis. Soil samples can be taken irrespective of the crop or variety growing on the land and the sampling time is not limited to any special period. Soil analysis is therefore well suited for use on a large scale but its value depends on how closely the data indicate the availability of the nutrients to plants. The basic differences between the two techniques for diagnosing the micronutrient status of soils are explained in Section 2.3 and the principles and quantitative measures needed to harmonize their results are presented in Sections 2.3.2 to 2.3.7. The aim has been to develop soil analyses in such a way that the findings would be consistent with those from the analysis of plants that meet all requirements and so combine in one method the advantages of both techniques. What- ever technique(s) may be adopted, the findings can only be considered indicative. There is a need for calibration by field trials with different crops to improve interpretation and establish critical deficiency and toxicity limits.

2.7 Estimation of critical micronutrient limits

As explained in Section 1.4.3 the zone limits presented in plant-soil micronutrient graphs are statistically defined, dividing the present international micronutrient data into five zones which do not however reveal critical deficiency or toxicity limits. Depending on micronutrient and crop, these zone limits may be too high or too low. Furthermore, the zone limits in this study are based on both the plant analyses and the soil analyses. In practical micronutrient studies both these data are seldom available and estimates of critical limits are usually based on one or other diagnosis alone. In order to obtain a general understanding of the relative differences between the micronutrient contents of various plant species, an experiment was established at nine 91 sites in different parts of Finland where 17 crops were grown side by side at each site. The micronutrient contents of the plants were analysed with the same methods as used in the present study, except that most of the plants were analysed at maturity. The results of the experiment will be published later (Yläranta and Sillanpää) but some preliminary data are given in Table 21.

Table 21. Two-year averages of micronutrient contents (ppm in DM) of 17 crops grown side by side at nine sites (some of the crops were successfully grown at only 7 or 8 sites).

Crop and part No. B Cu Fe Mn Mo Zn of sites Tc ±s Fc ±s Tc ±s R ±s ± s' R ± s Spring wheat, grain 9 1.7 0.5 6.7 2.2 60 15 86 37 0.20 0.11 56 21 straw 9 2.2 0.8 3.4 1.2 6921 87 43 0.45 0.43 31 26

Winter wheat, grain 9 1.6 0.2 6.2 2.4 66 40 76 27 0.26 0.12 48 11 straw 9 2.1 0.8 3.81.8 86 47 106 56 0.43 0.27 33 20

Oats, grain 9 1.4 0.4 3.8 0.9 60 13 70 23 0.40 0.19 42 11 straw 9 2.6 0.5 3.51.2 79 37 98 52 0.45 0.27 21 7

Barley, grain 9 1.4 0.2 7.1 2.2 52 10 24 6 0.33 0.16 37 14 straw 9 3.4 0.6 5.0 2.3 73 15 61 37 0.37 0.31 28 20

Rye, grain 9 2.2 0.2 7.41.1 53 10 39 8 0.40 0.17 50 9 straw 9 3.1 0.8 4.31.4 7023 55 22 0.61 0.17 35 22

Timothy, silage, 1. cut 9 8.7 2.2 6.0 1.6 74 18 69 28 0.47 0.39 29 7 2. cut 8 8.0 2.2 5.9 1.4 8122 113 31 0.91 0.57 30 8 dry hay 9 6.6 1.3 4.91.3 64 18 71 35 0.41 0.36 27 10 fresh growth 9 6.7 2.1 5.4 1.2 10939 128 41 1.13 0.65 32 12

Rye grass, silage, 1. cut. 8 8.5 2.4 7.1 2.4 16450 92 31 0.81 0.56 22 5 2. cut7 6.71.7 6.81.1 19092 149 59 1.34 0.96 22 2

Red clover, dry hay 7 26.4 6.4 10.0 3.0 75 11 87 53 0.61 0.42 42 14 fresh growth7 24.1 8.0 13.13.9 105 19 119 86 0.69 0.62 48 17

Pea, seed 8 7.51.0 9.31.9 86 22 21 8 1.841.43 46 6 stalk 9 29.4 9.2 8.7 2.4 15334 133 100 1.241.20 81 47

Turnip rape, seed 7 13.2 2.2 4.9 0.6 73 8 36 13 0.27 0.10 42 7 stalk 7 17.6 6.2 3.3 0.5 47 12 45 42 0.59 0.45 24 12

Turnip, root 9 20.8 4.9 4.5 0.8 44 12 19 13 0.19 0.12 29 14

Swede, root 9 18.6 4.1 3.1 0.5 45 18 19 10 0.14 0.06 19 8 tops 9 25.1 6.4 4.9 0.8 15081 94 62 0.74 0.39 34 11

Sugar beet, root 7 13.5 0.8 4.4 1.2 37 15 92 54 0.07 0.03 29 18 tops 7 26.3 7.8 8.6 1.6 16243 277178 0.39 0.22 109 69

Red beet, root 8 17.9 2.0 7.71.2 7126 113 48 0.09 0.05 52 37

Carrot, root 9 19.3 2.8 4.6 0.7 54 29 28 11 0.04 0.01 20 5

Potato, tuber 9 6.2 0.6 5.6 1.6 41 10 8 2 0.15 0.08 16 4

Onion, bulb 9 12.1 2.0 4.7 0.7 29 7 24 13 0.16 0.25 25 11 92 These data show that different plant -species can absorb widely different amounts of micronutrients from the same soils. Critical deficiency or toxicity limits established for one plant species cannot therefore be generally applied to others, and it would be necessary to develop different analytical interpretation systems for almost every plant species. These should include clearly specified stages of growth at which the plant is to be sampled and which part is to be analysed as well as details of analytical methods. For some crops and micronutrients systems of this kind have been fairly well established but are lacking or vague for the great majority of crops. An extensive review of literature on the interpretation of results of plant analyses, including the six micronutrients dealt with in this study, was presented by Bergmann and Neubert (1976). A similar review concerning boron was recently published by Gupta (1979) who also stated that in practice there can be no single value or even a very narrow range of values to describe "critical" levels in crops. A value considered critical by workers in certain areas may not be critical under conditions in other areas, and for certain elements the margin between deficiency and sufficiency is so narrow as to cause an over- lapping of values. Most of the differences between the values considered critical for a certain crop by various investigators are clearly due to differences between the analytical procedures and methods adopted and not only to different environmental conditions. Reference to the above two reviews often shows differences of more than tenfold between the values considered critical (deficient or toxic). For example, were the lowest critical (deficiency) B level given for wheat (0.3 ppm) to be comparable to plant B values obtained in the present study, there would be no B deficiency in the 30 countries studied. Again, if the highest deficiency limit referred to (5 ppm) were to be correct, about half of these soils studied would be deficient in B. The toxicity B levels in wheat referred to by the above reviewers vary from 16 to 100 ppm. Typical B contents of healthy and B-deficient plants were also reviewed for a number of crops by Brandenburg and Koronowski (1969). For example, the B contents of healthy sugarbeets on a DM basis were 25-40 ppm in leaves and 15 ppm in roots; while those for B-deficient plants were 13-20 and 13 ppm, respectively. Healthy mangel-wurzel contained 20-46 ppm in leaves and 17 in roots (B-deficient 7-18 and 16 ppm, respectively); turnips 30-40 ppm in leaves and 18-22 in roots (B-deficient 9-20 and 8-15 ppm, respectively); potato leaves 14-30 ppm (B-deficient 4 ppm); lucerne 20-29 ppm (B- deficient 7-19 ppm); apple 20-25 ppm in leaves and 16-26 in fruit (B-deficient 12-16 and 2-6 ppm, respectively); tobacco leaves 16-50 ppm (B-deficient 4 ppm); celery leaves 26-38 and tubers 19-29 ppm (B-deficient 15 and 13 ppm, respectively). Since the situation with regard to other micronutrients studied is very similar to that of B, the range of plant micronutrient values obtained from the literature is undoubtedly too wide to help in establishing precise critical ranges for values obtained in this study. Comparison of the results of soil analysis in the present study with those of others is also limited because of the wide variety of extraction methods used by different investigators. Boron is an exception in this respect; the hot water extraction method has been widely used and estimates of critical soil B values have been presented. For example, Reisenauer et.al. (1973), referring to several sources, and Kurki (1979) considered hot water extractable soil B values < 1 ppm not high enough for optimum plant growth, and Smilde (1976) has presented a value of 0.3 ppm. Severe B deficiency (more than 100 percent yield response to B) has been reported in soil with a hot water extractable soil B content of 0.15 ppm (Hong 1972). According to several authors (e.g. Bould and Hewitt 93 1963, Jackson 1964, Mitchell 1974, and Park and Park 1966) the deficiency limit may be in the range of 0.5 ppm depending on soil factors and plant species)) Many investigators consider B deficiency to be more widespread than that of any other micronutrient. For example, in the United States the most commonly found micronutrient deficiency was that of B. It occurred in one or more crops in 41 sta tes, and in Wisconsin alone nearly two million acres (about 800 000 ha) of alfalfa showed B deficiency at one time or another during the growing season (Berger 1962). Against this background, the lower five percent zone limit and even the lower 10 percent zone limit seem for many plants to be too low to be considered as deficiency limits or values below which B deficiency can be suspected. Tentatively in this study hot water extractable B values <0.3-0.5 mg/I are considered suspect for B deficiency. The margins between the amounts of B required for normal growth and those producing symptoms of excess or toxicity vary with plant species and are considered rather narrow for B in comparison to most other micronutrients. Critical, excessive or toxic, hot water extractable soil B values varying often from 2.5 to 5 ppm have been recorded (Anderson and Boswell 1968, Gupta and Munro 1969, Kurki 1979, Mortvedt and Osborn 1965). In general, data on B toxicity seem to be more rare than those on deficiency. Many of the plantsoil B values falling in Zone IV or the lower part of Zone V may therefore still not be considered high enough to indicate conditions of B excess, especially for crops with high tolerance to B. Very little appears to be known about deficiency or toxicity limits for Mn based on DTPA soil extraction since the method has been in use for a relatively short period of time. Lindsay and Norvell (1978) when characterizing the possible usefulness of the DTPA soil test for Mn, suggested that the critical level of DTPA extractable Mn could be tentatively set at 1 ppm until further information is available. Compared to the DTPA data of the present study, this value seems very low since the minimum value recorded was 0.9 mg/1 and in only a very few sois out of more than 3500 sois from 30 countries was Mn as low as 1 ppm. Nevertheless, Mn deficiency has been reported from numerous countries and symptoms of Mn deficiency have been described for more than 20 crops including oats, rye, wheat, rice, maize, peas, soya beans, potatoes, tomatoes, cotton, tobacco, sugarbeet, tea, sugar-cane, pineapples, pecan, peaches, spinach, citrus and a number of forest trees (e.g. Bergmann and Neubert 1976, Koronowski 1969, and Sprague 1964). Against the above background, instead of DTPA values of < 1 mg/1, values of 2 or 3 seem a more appropriate deficiency limit for Mn and values of up to 4-5 mg/1 may still be considered to indicate susceptibility. For pH-corrected DTPA the respective figures would be somewhat lower. The above figures are still tentative and subject to the soil material in this study being representative enough and the soil variation range wide enough for such estimates. The widely varying Mn requirements of different plant species should also be kept in mind. DTPA values indicating excess or toxicity of Mn are not well established. Therefore, it cannot be stated whether the upper 10 or 5 percent zone lines of this study are too high or too low for predicting Mn excess. Tentatively, until further experience is gained, DTPA values exceeding 140-200 mg/1 may be considered suspiciously high. DTPA has been quite widely used to extract Zn from soils and, therefore, information on the interpretation of DTPA Zn values is more abundant than for other micronutrients.

o Taking into account that only about 0.2 percent of the soil material of this study were peat soils with a very low volume weight and also the effect of extraction ratio, the numerical ppm values can be considered about equal to those of mg/l. 94 Since Zn toxicity is generally not considered as serious a problem as Zn deficiency, the interpretation data given in literature concern mainly the estimation of critical deficiency level. Depending on the crops and other factors, various authors have proposed somewhat different DTPA values as critical deficiency limits, but in general, this limit seems to be relatively well established compared to most other micronutrients. Singh et al. (1977) considered 1.4 ppm DTPA extractable Zn as the critical limit for maize in pot trials. Randhava and Takkar (1975) proposed different deficiency limits for different Indian soils and crops (wheat, maize and rice) varying from 0.5 ppm (Deep and shallow black) to 1.0 ppm (Sierozem). Sedberry et al. (1980) found a response of rice to Zn on soils containing less than 1 ppm DTPA extractable Zn. Lindsay and Norvell (1978) considered 0.6 ppm sufficient for sorghum but a somewhat higher value (0.8 ppm) for maize. Viets and Lindsay (1973) estimated that less than 0.5 ppm was critical and 0.5-1.0 marginal for sensitive crops. Several other authors (Brown et al.,1971, Rathore et al.1978, Sakai et aL, 1979, Whitney et al., 1973, Whitney, 1980) have come to the conclusion that DTPA extractable Zn values of about 0.5 ppm represent the critical deficiency limit for several crops. A Zn value of 0.5 ppm (-,0.5 mg/1) of DTPA extractable Zn in this study would correspond to a Zn content of about 14 ppm in the pot-grown wheat and somewhat higher contents in original wheat and maize (see Zn graphs and Section 2.3.7). These figures are in relatively good agreement also with many recent data concerning deficiency limits based on plant Zn content. For example, Franck and Finck (1980), Rathore et al. (1978), Takkar et al. (1974) and Weir and Milham (1978) gave plant Zn contents varying from 12 to 20 ppm in wheat and maize as deficiency limits for Zn. If a DTPA value of 0.5 mg/lis considered to be the deficiency limit for Zn and applied to the DTPA data of this study, it would mean that even the lower 10 percent zone line (corresponding to about 0.25 mg/1 DTPA extractable Zn) given in the Zn graphs is located far too low to indicate the critical deficiency limit, and instead of 10 percent about one- third of the soils in the 30 countries could be suspected to be deficient in Zn. This would not be surprising since during recent years an increasing amount of information has emerged on Zn deficiency, especially from developing countries. Zn toxicity may be relatively rare and little information on toxic DTPA Zn levels is available. According to Takkar and Mann (1978), DTPA values of 7 and 11 ppm and plant Zn contents of 60 and 81 ppm may be considered to be levels at which wheat and maize, respectively, are susceptible. Since in this study the results given for soil Cu and Fe are based on AAAc-EDTA extraction which is a relatively new method, there are very few reference points in the literature concerning the deficiency or toxicity limits for this extractant. Therefore, direct comparisons to earlier published data are not possible. However, these elements were also extracted with DTPA from the whole of the international soil material of this study (see Sections 2.3.4 and 2.3.5) which makes some indirect approximations of deficiency limits possible. For DTPA extractable soil Cu, 0.2 ppm has been considered as a limit below which plants are likely to suffer from Cu deficiency (Viets and Lindsay, 1973, Lindsay and Nor- veil, 1978). Of the DTPA Cu data of this study, varying from 0.03 to 100 mg/1 (mean 1.8 mg/1), about 5 percent of soils show values lower than 0.2 mg/I indicating that in this material approximately 5 percent of soils are suspect for Cu deficiency. Thus, the lower 5 percent line given in the Cu graphs would give a rough estimate of the deficiency limit 95 for Cu. When converted to AAAc-EDTA and organic C-corrected AAAc-EDTA Cu values, the deficiency limits would be in the range of 0.8-1.0 mg/1 which according to the regression lines in the Cu graphs would correspond to wheat Cu contents of 4ppm or slightly less. The latter figure is somewhat higher than is considered to bea critical deficiency level for wheat e.g. by Mitchell (1974) and Panin et al. (1971) who estimated Cu contents of 2.5-3.0 ppm as critical limits. The latter authors considered 11.3ppm in wheat as a critical level of Cu toxicity but both lower and higher figures have been presented. The DTPA Fe values measured from the soil material of this study varied from 1 to 586 mg/1, averaging 37 mg/l. Lindsay and Norvell (1978) expected soils with >4.5ppm Fe not to show Fe deficiency. The present material includes about 5 percent of soils with DTPA Fe values lower than that.Since the correlation between DTPA Fe and AAAc-EDTA Fe is relatively good (r = 0.741***) the lower 5 percent line in Fe graphs could be tentatively considered as a line to separate cases of possible Fe deficiency and sufficiency. Converted to AAAc-EDTA extractable Fe values this would correspond to some 30-35 mg/l. Estimation of available Mo status of soils based on A0-0A extraction is rather difficult because the results obtained with this method as such do not correlate well with the results of plant analyses of Mo (see Section 2.3.2). By correcting the A0-0A extractable soil Mo values for pH the plant-soil correlation is considerably improved, but there are no previous data for comparison or for assessment of any deficiency or toxicity limits for the pH-corrected soil Mo values. Furthermore, individual crops differ markedly in their Mo requirement, the Cruciferae and legumes usually considered to be the most demanding and monocots the least. Critical Mo deficiency or toxicity levels in different plant species are not well defined. For example, in the literature review by Bergmann and Neubert (1976) Mo contents of wheat considered as low varied from 0.09 to 0.7 and those considered as high from 0.18 to 1.5 ppm. On the basis of the above it seems impossible to identify any soil or plant Mo values as critical limits. Therefore, until further information is available, plantsoil Mo values falling in the lowest and the highest Mo zones (I and V) may be considered indicative of possible Mo deficiency or excess, after taking into account the plant species concerned. The respective limits as applied to pH-corrected A0-0A extractable Mo data would be in the range of 0.01-0.02 mg/1 for deficiency and 0.5-1.0 mg/1 for excess. The above discussion concerning the estimation of deficiency and toxicity limits for micronutrient data of this study can be briefly summarized as follows: The statistically defined lower 5 and even the lower 10 percent limits given in B and Zn graphs seem to be located at too low a level to be considered as critical deficiency limits for B and Zn, nor may B and Zn values falling in Zone IV and lower parts of Zone V necessarily indicate an excess of these micronutrients. For the other four micronutrients (Cu, Fe, Mn and Mo) the data obtained from literature either indicate that the lower 5 percent lines locate fairly well within the ranges below which deficiencies of these micronutrients are likely to exist, or the data are too heterogeneous to give clear indications of these lines being located at too low or too high levels. The same concerns the upper statistical 5 percent lines for Cu, Mn, and Mo and possible excess of these elements. 96 Converted into soil micronutrient data obtained by various extraction methods used in this study, the critical levels (mg/1) are estimated as shown in Table 22.

Table 22. Tentative critical levels of micronutrients determined by various methods.

Micronutrientandmethod Range of deficiency excess B, hot water extraction <0.3-0.5 > 3-5 B, hot w. extr. CEC correction <0.3-0.5 > 3-5 Cu, AAAc-EDTA <0.8-1.0 > 17-25 Cu,AAAc-EDTA org. C correction <0.8-1.0 >17-25 Fe,AAAc-EDTA <30-35 Mn,DTPA <2-5 >140-200 Mn, DTPA + pH correction <2-4 > 150-220 Mo, A0-0A + pH correction <0.01-0.02 > 0.5-1.0 Zn,DTPA <0.4-0.6 > 10-20 Zn,AAAc-EDTA pH correction < 1.0-1.5 > 20-30

The values are tentative only and are subject to revision when more information is available. It is important when interpreting the results to take into account the varying micronutrient requirements of different crop species.

2.8 Plant ani soil micronutrients and other soil data in relation to FAO/Unesco soil units

Classification of soils of the present study by FAO/Unesco soil "Orders" (Dudal, 1968) was carried out in the field by the sample collectors in the various participating countries. The data are incomplete because all sample collectors did not consider themselves com- petent enough to classify the soils. Out of 3744 soils 2265 were classified (Appendix 6). Further, there is bound to be a certain degree of heterogeneity in the classification of the soil material because more than 100 sample collectors participated in classifying the soils, all of them not equally highly qualified in this particular field. In some cases when the soils were classified by sample collectors according to the national classification system only, experts at FAO Headquarters were able to convert these into FAO/Unesco soil units. Since full profile descriptions were not available, coordination of the classified s- oil data and correction of possible errors were limited only to a few orders (e.g. Chernozems, Phaeozems and Rendzinas) having diagnostic features within the surface soil layers. Because relatively little information on relations of micronutrients to FAO/Unesco soil units is available, it was considered justifiable to publish these results. However, taking into account the foregoing reservations, the data can only be considered as indicative, and therefore only the average micronutrient contents of soils and respective plants for various soil units are presented in Appendix 7. It should be noted that the background data in the graphs (regression curves, mean, standard deviations (s), zones) are calculated from the whole of the pot experiment data (n = 3538) but the soil unit data plotted in the graphs represent the number of soils classified (n = 2265) only. The average data concerning general soil characteristics and macronutrients are given in Appendix 6. 97 3. Summary for Part I

The original sample material collected from 30 countries totalled 7488 samples, half of which were soils and half plant samples growing on those soils. The plant (wheat and maize) samples, however, proved too heterogeneous to reflect reliably the micronutrient (B, Cu, Fe, Mn, Mg, Zn) status of soils where the plants had grown. Among the factors causing uncontrolled variation in analytical results were: differences due to the two plant species and cultivars (over 200 maize and over 200 wheat cultivars); variation due to differences in the physiological age of plants when sampled; differences in main nutrient fertilization, yield levels, irrigation, use of herbicides possibly containing micronutrients; contamination during sampling, sample pre-treatment and transportation. Therefore, to minimize the uncontrolled variation, fresh indicator plants (wheat, cv. 'Apu') were grown in pots on the original soils under uniform and controlled conditions. Owing to the superiority of this plant material, the discussion concerning micronutrients is based on the analytical results of the new indicator plants. After preliminary methodological studies from limited materials, soil micronutrients were extracted from all the soil samples as follows: B with the hot water extraction method, Mo with ammonium oxalate-oxalic acid (A0-0A), and Cu, Fe, Mn and Zn with both the acid ammonium acetate-EDTA (AAAc-EDTA) and the DTPA extraction methods. The correlations between the results of plant and soil analyses for all six micronutrients were considerably improved when data from the new pot-grown plants were used instead of data from the original indicator plants (Table 23).

Table 23. Correlations between micronutrient contents of the three types of indicator plants (original maize and wheat and pot-grown wheat) and respective soils.

Micronutrient and Original Original Pot-grown extr. method maize wheat wheat (n = 1966) (1768) (3537-3538) B hot water extr. 0.548*** 0.694*** 0.741*** Cu AAAc-EDTA 0.344*** 0.254*** 0.664*** Cu DTPA 0.114*** 0.125*** 0.518*** Fe AAAc-EDTA -0.007 n.s. -0.032 n.s. 0.325*** Fe DTPA -0.028 n.s. -0.030 n.s. 0.263*** Mn AAAc-EDTA -0.108*** 0.046* 0.039* Mn DTPA 0.161*** 0.119*** 0.552*** Mo A0-0A 0.134*** 0.117*** 0.245*** Zn AAAc-EDTA 0.391*** 0.405*** 0.665*** Zn DTPA 0.381*** 0.473*** 0.732***

In spite of the improvements in correlations, satisfactory conformity between the results of plant and soil analyses was not achieved. To understand the contradictions still prevailing between the results of plant analyses and soil analyses, it must be realized that these two techniques are based on fundamentally different principles. Plant analyses give measures of micronutrients which have been avail- 98 able to and have indeed been absorbed by plants. Absorption of micronutrients by plants is a process that takes place under the laws of biochemistry and plant physiology while chemical soil extraction mainly follows the laws the chemistry. Accordingly the results of plant analyses (if based on reliable and comparable plant material) are to be considered as the most reliable measure of the soil micronutrient fractions available to plants. Soil analysis in more of a "shortcut" method and an attempt to imitate plants. Therefore, if the results of plant analyses obtained from reliable and comparable plant material contra- dict those of soil analyses, and if the responsible factor (which affects plant and soil analyses in different ways) can be identified and its effects quantified, it is the soil analysis which must be corrected accordingly. In large scale micronutrient studies, results of both plant and soil analyses are seldom available and one or other technique must be chosen. The theoretical advantages of plant analyses are diminished by the practical difficulties of obtaining comparable plant material or of interpreting results from the analysis of heterogeneous materials. For example, comprehensive interpretation systems have only been developed for relatively few plant species, very little is known about differences between plant varieties and no general agreements have been reached as to which parts of different plants are to be analysed. Furthermore, to obtain comparable results plants must be sampled at the same physiological age which severely limits the sampling time. In contrast, soil samples can be taken irrespective of the crop or variety growing on it and the sampling time is not limited to any special period. Soil analyses are therefore well suited for large scale use, but their applicability depends on how closely their results agree with those from plants. The major aim in this study has therefore been to devise means of calibrating soil analyses so as to eliminate discrepancies between plant and soil analyses. To identify the factors responsible for impairing the plant-soil micronutrient correlations, and to quantify and eliminate their contradictory effects, both the plant micronutrient data and the soil (extractable) micronutrient data were studied in relation to six soil characteristics (pH, texture, organic carbon content, CEC, electrical conductivity and CaCO3 equivalent). These studies covering all the international soil samples, showed that in the case of hot water soluble B such a soil factor was the cation exchange capacity and for AAAc-EDTA extractable Cu the organic carbon content of the soil. The DTPA extractable Mn, A0-0A extractable Mo and AAAc-EDTA extractable Zn required correction for soil pH. The principles and procedures for such corrections are presented in detail. These adjustments should be considered as an essential part of soil analysis. The mathematical equations of correction coefficients for the above micronutrient extraction methods are:

Boron: k(CEC) =100.466-0.026 x 0.000273x2 where x cation exchange capacity (me/100 g) Copper: C) =10-0.491 log x + 0.470 (log x)2 where x =soil organic carbon content (%), Manganese: k(pH) =107.06 - 2.20 pH + 0.164 pH2 Molybdenum: k(pH) =10-2.45 + 0.36 pH Zinc: k(pH) =103.38 - 0.976 pH + 0.068 pH2

These five correction coefficients can also be obtained directly from graphs in Figs 49 and 50. 99 3

o o 1

O '03ta0

22 .25 .5 1 2 4 8 16 32 64 Organic C (%)or CEC ( me/100g ) Fig. 49. Coefficients for correcting AAAc-EDTA extractable soil Cu for soil organic carbon content, k(org. C); and hot water soluble B for cation exchange capacity, k(CEC).

o Q 2 o

4 5 6 7 8 pH (Ca C12) Fig. 50. Coefficients for correcting DTPA extractable soil Mn, A0-0A extractable Mo and AAAc-EDTA extractable Zn for soil pH. 100 Application of these correction coefficients to the whole of the international material (n = 3537 - 3538) improved the correlations (r values) between the results of plant analyses (pot-grown plant material) and soil analyses as follows:

Micronutrient and r value Correction r value extr. method without coefficient with correction correction B hot water extr. 0.741*** k(CEC) 0.826*** CuAAAc-EDTA 0.664*** k(Org. C) 0.731*** Mn DTPA 0.552*** k(pH) 0.713*** Mo A0-0A 0.245*** k(pH) 0.696*** ZnAAAc-EDTA 0.665*** k(pH) 0.706***

In addition to the uncorrected data the above correction coefficients are applied when presenting the micronutrient data by countries in Part II. Correction coefficients calculated for AAAc-EDTA extractable Fe and DTPA extractable Zn did not appreciably improve the respective plant-soil correlations and were not adopted. The most effective correction coefficient was that calculated for AAAc-EDTA extractable Mn for pH which improved the plant Mnsoil Mn correlation (r) from 0.039* to 0.588***. This also shows that even an extraction method which in itself gives poor results may be successfully used if its special features and its behaviour with respect to various soil factors (in this case the pH) are known and properly taken into account. As the DTPA extractable Mn generally correlated better with plant Mn the results of AAAc-EDTA Mn are not presented in Part II. The mutual relations between the contents of the six micronutrients in plants as well as those extractable from soils are presented. In general, if there is a good correlation between the contents of two micronutrients in plants or those extracted from soil, their availability to plants or their extractability from soils is largely controlled by the same soil factor or factors. The micronutrient contents of plants and soils in relation to yield are presented with a special reference to the effect of yield on the micronutrient contents of plants. In general, plants with restricted growth (low yields) contain more micronutrients per unit plant mass produced in relation to available soil micronutrients than do the plants growing more rapidly (high yields). Rough quantitative estimates of this phenomenon called here the "concentration-dilution phenomenon" are given with special reference to its effects on the results of plant analyses. Estimations of critical levels for various micronutrients are presented. These are to be considered tentative and are subject to revision when more information is available. Although the main emphasis of this study is put on micronutrients, the macronutrients (N, P, K, Ca, Mg) of soils and plants (original wheat and maize) were also analysed and data on their plant-soil relations as well as on their relations to various soil characteristics are presented, though in less detail. PART II 103 PART II

NUTRIENT STATUS BY COUNTRIES

4. Introduction

Although the main emphasis of this study is on micronutrients, a number of additional analyses on soils and plants were carried out in order to obtain further information on factors affecting the behaviour of macronutrients in soils and plants. When expressing the results of the whole study by countries, these additional data are included, though in a more general way than data concerning micronutrients. To facilitate comparison of the results from one country with those from all combined, the national results from each country are presented graphically and jointly with the respective data for all international samples. Accordingly, when presenting data on general soil properties and macronutrients, the frequency distributions of national results are expressed by columns and the respective international data superimposed on the same Figure as a slightly generalized frequency curve. Numerical data (number of samples, mean, standard deviation, minimum and maximum) for both the national and international material are also given in each graph. The data on micronutrients are presented in greater detail. The micronutrient contents of each plant-soil sample pair are shown as points in each Figure; also given are the best fitting national regression line, regression formula, correlation coefficient, mean and standard deviation. Summarized data for all the international samples are given in each graph in order to make it easier to evaluate tbe national results against the background of the international data. Since it is not possible to plot the values of single sample pairs of the whole international data in one graph, five zones (I---V) are given to show, in broader terms, the distribution of the micronutrient in question in all samples. Of the whole international material, the lowest 5 percent of plant x soil content values fall in Zone I, the next lowest 5 percent in Zone II, the "normal" values (80 percent) in Zone III, and the highest 10 percent in Zones IV and V (5 percent in each). For more details concerning the five zones, see Section 1.4,3. It must be noted that when data concerning the macronutrient contents of plants are presented, only the contents of the two original indicator plants are given, i.e. wheat and maize grown in various participating countries. Since no distinct differences in the macronutrient contents between spring wheat and winter wheat were found, the results for these crops were combined. In case of micronutrients, the study is based on the contents of 104 pot-grown wheat only. The reasons for this departure from uniformity of method are simple: because of the heterogeneity of the original plant samples (differences due to plant varieties, age at sampling, fertilization, contamination etc., see Sections 1.2.1 to 1.2.4) too much uncontrolled variation in the rnicronutrient contents of plants was involved to give a reliable picture of the actual micronutrient status in various countries. It is clear that this concerns the macronutrients as well, even if to a lesser extent. The analytical data on micronutrients presented in Part II therefore only refer to micronutrient contents of pot- grown wheat samples, although the original plants were also analysed. Unfortunately, the plant samples obtained from the pot experiment were too small to allow analyses of macronutrients in addition to micronutrient analyses. To locate approximately the origin of the samples studied, national maps indicating the sampling sites in each participating country are usually included. The data for the maps were received from the cooperators in each country who also have more detailed geographical information on the sampling sites (town, village, farmer, field) in their possession for possible future use. The national results presented in Part H are expressed in graphical forms which are self-explanatory. Therefore, these are furnished with rather short comments only. 105 5. Europe and Ozea-aia

5.1 Belgium

5.1.1 General

The sites for sampling plants and soils in Belgium are shown in Fig. 51 and the frequency distributions of four important soil characteristics, texture (expressed as texture index), pH (CaCl2), organic carbon content, and cation exchange capacity, are given in Fig. 52. The mean, standard deviation, minimum and maximum values for other soil properties (particlesize distribution, pH(H20), electrical conductivity, CaCO3 equivalent and volume weight) are given in Appendixes 2, 3 and 4, separately for soils from wheat fields, maize fields and soils used in the pot experiment.

,----

/ L., ,.....L.

43321-322 'S.., I r -..... ANTWERPEN BRUGGE\.--- si. , 0 4332 327

43328-334 GENT 5 'I SI. \ 4 (4 ., / o 43303 HAS8ELT i 04330 43105 BRUXELLES ... 43301 e 43332-340 04330-3 , 43306 \ 43315-320 \ 43305 10 LIEGE k....., o 43307 0 43300 ..; I. ( NAMUR MONS 0 CHARLEROI N \

i-':-', /

,....\

I 50° -... /I

)

...._ BELGIUM o ) ARLON N .1.- . f

10 20 60 60 7?

4 5° 6°

Fig. 51. Sampling sites in Belgium (points = wheat fields, trianglesmaize fields). Identification numbers of sample pairs are given. 106

35 BelgiumInternat Belgium iriternot Fig. 52.Frequencydistributionof o. 41 3764 n= 41 3783 texture, pH, organic carbon content = 29 44 0= 5.80 6.64 30 ts=10 16 s=.76 1.12 awl= 15 9 35 min=4.45 3.62 andcationexchangecapacityin max= 43 92 max.47.40 8.56 25 30 Belgian soils (columns). Curves show the international frequency of the same 20 25 o characters. The number of samples (n), 20 mean (TO, standard deviation (s) and o Ui minimum and maximum values are given for both national and international data.

10 20 30 0 50 60 70 80 90 TEXTURE INDEX pH (Co Cl2

Belgium Internat. Belgium Internat 40 n= 41 3779 n= 41 3777 =1.2 1.3 0 0. 18.0 27.3 14.6 -E35 os. .4 1.2 ts=3.2 Iran.= .6 rmn=12.8 max=2 1 39.1 25 max=24.5 99.9 ,a 30 o25 uj20 o

16 32 6) 12.8 25.6 16 32 ORGANIC C,% CEC, me/100g

Compared with all the international soil samples studied, the Belgian soils are relatively coarse textured and acid. The average texture index is 29 and pH 5.80 (international averages 44 and 6.67, respectively). The soil samples from the northeastern part of the country (samples 43321-34), classified as Arenosols and Podzols, are especially sandy and acid (clay content 5 %, TI 15-19 and pH 4.45-5.90), so differing from soils in other parts of the country (mainly Luvisols and Regosols, TI 29-43 and pH 5.00-7.05). These differences are often reflected in other properties of soils and plants as will be discussed later. The organic carbon contents of studied Belgian soils are at the average international level with a relatively narrow range of variation. The highest organic C contents were found in samples from the northeastern region. The average cation exchange capacity (18.0 ± 3.2 me/100 g) is low compared to the international mean (27.3 ± 14.6) and varies only slightly from one sample to another. Electrical conductivity of Belgian soils is somewhat lower and CaCO3 equivalent substantially lower than those in most other countries (Appendixes 2, 3 and 4). Low sodium contents are typical of most Belgian soils.

5.1.2 Macronutrients

The macronutrient contents of plants were analysed from the original wheat and maize samples only. The results of nitrogen, phosphorus, potassium, calcium and magnesium analyses of Belgian soils and plants are given as frequency distribution graphs in Figures 53-57, separately for wheat and maize and the respective soils. It should be noted that the differences in plant varieties, fertilization, climate and other local factors (see Section 1.2) 107

Figs 53-57. Frequency distributions of BelgiumInternet Belgium Internat. n= 21 1765 n= 20 1958 5 3.14 nitrogen, phosphorus, potassium, =3.95 4.27 =5.01 = .61 1.15 = .28 .87 min=2.68 .80 minn4.41 .88 calcium and magnesium inoriginal 30 max= 5.04 745 max5.54 6.51 wheat and maize samples andre- 25 0- spective soils (columns) of Belgium. (>320 Curves show the international frequency 1-,1 15 of the same characters. o

5 6 2 0 5 6 4 N content of wheat% N content of maize,%

70 Belgium I nternat Belgium Intermit n= 60 n= 21 1765 60 20 1958 X= .103 .133 7. .134 .135 ,s= .013 .084 *s= .023 .088 min=.081 .009 min=.088 .008 non. 129 1.023 50 max . .171 1.657

005 01 02 04 n 0 32 128 00 02 04 uo 16 32 64 1.28 Fig. 53. Nitrogen, Belgium. N in wheat soils, % N in maize soils,% may have caused variation in the macronutrient contents of plants in addition to those due to soils. The mean nitrogen contents of the original Belgian wheat and soils are somewhat below the international averages (Fig. 53). The nitrogen contents of Belgian maize (mean 5.01 %) are the highest in this study (Appendix 3) and those of the maize soils (0.134 %) about average internationally. Nitrogen was applied to Belgian crops (wheat 98 ± 21 and maize 118 ± 66 kg N/ha) at rates that were high compared to most other countries represented in this study. See also Fig. 6. The average phosphorus content of Belgian maize samples (0.535 %) is the highest among the countries studied, exceeding the international mean content by a factor of 1.6. In the case of wheat, the Belgian average (0.501 %) is exceeded only by that of Malta (Appendixes 2 and 3). The average 0.5 N NaHCO3 extractable P contents of Belgian maize soils (119.8 mg/1) and wheat soils (83.7 mg/1) are also higher than those of any other country. They are more than ten times the national mean P contents of soils in many developing countries and exceed the international mean contents by factors of 5.3 and 4.0, respectively. According to information on phosphorus fertilization received from sample collectors, the Belgian wheat and maize crops had received 21-39 kg more P per hectare than the average for the crops studied in other countries (Appendix 5). To what extent the high P contents of Belgian maize and wheat samples are attributable to the exceptionally high P contents of Belgian soils and how much to the most recent phosphorus application cannot be stated. Clearly, the exceptionally high P contents of Belgian soils are at least partly due to the liberal use of phosphates for many decades. See also Fig. 7 and related text. The higher extractable P contents of the coarse textured northeastern soils may be due to lower fixation of fertilizer P applied in past years. 108

00 00 Belgium Internat. Belgium Internet Fig. 54. Phosphorus, Belgium. 35 n o 21 1765 n = 20 1967 .30 .38 o.53 .33 *5 =.10 .12 ss o .10 .10 8 30 min o.33 .05 30 min..37 .05 0000.. .67 502 max...78 1.04 ert 0- 25 25 )- r2") 20 Ui 8 s Ui 10 EL,

2 3 P content of wheat, /o P content at maize,°/0

0. 40 Belgium Internet. Belgium Internet. n= 21 1765 o 83.7 20.2 5 n = 20 1967 ts o 213 24.7 8. 119.8 22.5 min.= 45.6 1.0 ss = 313 33.0 max.= 121.6 271.4 min.. 43.2 0.1 0 max, 187.0 656.0

D 15

u_ 10

2.5 5 10 20 40 BO 160 320 640 2.5 5 10 20 40 80 160 320640 P in wheat soils, mg/I P in maize soils, mg/I

BelgiumInternet Belgium Internat. Fig. 55. Potassium, Belgium. no21 1765 n= 20 1967 30 A= 3.9 4.0 A. 4.4 3.1 no- 0.6 1.0 - s0.5 1.0 non = 2.9 0.9 non.- 3.2 0.6 8 25 max.= 4.8 6.8 max.= 5.0

5 6 5 6 7 K content of wheat,°/0 K conterit of maize,%

Belgium niernol. Belgium Internet no 21 1765 n=20 1967 0= 213 365 0232 330 30 ss = 58 283 its oi94 356 min.= 122 20 min.= 108 18 max.= 369 2097 max = 406 5598

20 o

25 50 100 2 0 0 0800 1600 3000 6400 25 50 00 200 4 0 8001600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

The mean potassium content of the original Belgian wheat samples is at the international average in spite of the relatively low 1 N CH3COONH4 exchangeable K content of Belgian wheat soils (Fig. 55). The K contents of Belgian maize samples are considerably higher than the international average of this material, notwithstanding that the exchangeable K 109

Fig. 56. Calcium, Belgium. Belgium Internal Belgium Interne n =21 n= 1765 n. 20 1967 = .36 R. .43 8.62 .47 is = .08 is =.17 is = .11 .20 .24 min.= .42 .09 max.= .49 1.68 max.= .86 1.88

Ca content at wheat, % Ca content of maize %

Belgium internat. Belgium Internat. n = 21 n 1765 n.20 1967 35 R = 1868 = 4671 35 8.1740 3450 is =983 ts = 3076 is =866 2865 805 min.= 110 min =490 10 max.= 5045 mm.. 21930 max =4050 17996

23 25

100 2004008001603 3e00 4400 12800 25600 50 000 200 400 1300 7-00 012. 6400 12800 Ca in wheat sois, mg/I Ca in maize soi s, mg/I content of Belgian maize soils is below the international average. The high rates of K applied to Belgian wheat and maize crops (Belgian means 130 ± 29 and 122 ± 90 kg K/ha; inteimational means 21 ± 44 and 7 ± 18 kg K/ha, respectively) may explain the discrepancy between the high plant and low soil K contents, but the relatively low pH of Belgian soils may also have played a part (see Section 2.2.4). For a general assessment of K status in Belgium against the international background, see Fig. 8 and related text (Section 2.2.3). The calcium status of Belgian plants and soils is similar to that of K. The average Ca content of wheat is somewhat below, and the Ca contents of maize above the international averages, but the exchangeable Ca contents of soils are well below the international mean (Fig. 56). It is evident that there is always enough Ca in soils to meet the requirements of plants and that factors other than the Ca content of soil largely determine the uptake of Ca from soils. As mentioned in Section 2.2.4 (Fig. 12 and related text), Ca is relatively more available to plants in coarse than in fine textured soils. The Belgian maize crops sampled were grown on much coarser textured soils than wheat (mean TI values 23 and 36, respectively, Appendixes 2 and 3). This may at least partly explain the relative/y high Ca contents of maize compared to wheat. Most of the sampled Belgian soils had recently been limed. See also Fig. 9. The exchangeable magnesium contents of Belgian soils, on average, are lower than those of any other country included in this study (Appendixes 2, 3 and 4). The lowest soil Mg values come from the northeastern part of the country (samples 43321-34) but even the highest soil Mg value from Belgium (sample 43315, 146 mg Mg/1) is only about one-third of the international average. The low Mg contents of soils are reflected in the Mg contents of the plants. The mean Mg content from Belgian wheats (0.088 %) is the lowest national average in this study, and in the case of maize only three countries (New Zealand, Thailand and Sierra Leone) have slightly lower mean Mg contents than Belgium. Although 110

Belgium Internat Belgium 991 Internal Fig. 57. Magnesium, Belgium. . 21 1764 5 n.20 n= 1967 R = .088 .172 R..184 R. .251 ts. .018 .060 *5= .040 9,-. .119 .057 .044 0 mm..083 min.= .036 'Fs; max = .132 .948 max...250 mox... 1.125 k25 25

(-) 20 20

D 15 5

5 Mg content of wheat,% Mg content of maize,./0

40 40 Belgium /nter not Belgium Internat. n=20 1967 n= 21 1764 5 R. 92 489 4.76 446 -±s= 29 437 31 462 46 min,35 0 max.. 146 3214 30 max = 135 5490

Ú 25 o

111 o Lii

25 50 1 0 200 400 800 16003200 25 50 100 200 400 800 1600 3200 6400 Mg in wheat soils. mg/I Mg tn maize soils. mg/I the amount of Mg removed annually from soils by crops is relatively small (e.g. when compared to K), it is apparent that in Belgium where decade after decade the national crop yields have been among the highest in the world, considerably more Mg has been removed from soils than in most other countries. This may be one of the reasons contributing toward the present low Mg status of Belgian soils. No information is available on any possible Mg dressing or the Mg content of lime applied to the soils sampled for this study. To obtain a broad view of the Belgian soil Mg status against the international background, see Fig. 10 (Section 2.2.3).

5.1.3 Micronutrients

Boron. The B contents of individual pot-grown wheat samples and respective soils are given in Figures 58 and 59. The Belgian B points lie somewhat above the international regression line (Fig. 58) but come closer to it when the soil B values are corrected for CEC (Fig. 59). Because of the narrow range of variation of CEC in Belgian soils, the CEC correction only slightly affects the mutual relations between the points that indicate plant and soil B values of individual sample pairs. In view of the limited number of Belgian samples, the B situation in Belgium seems to be about average, internationally. Almost all the B values fall in the normal range (Zone III) with no extremely low or high B values. See also Figs 22 and 25 (Section 2.3.3). 111

Fig. loo 58. Regression of B content of ilziEsILIIIIIIMPAinii Wien ECM= pot-grown wheat (y) on hot water 80-,B lifilMIIIIIIIMMIMIE IN= MEN 60 = soluble soil B (x), Belgium. The points 1n .= 6 98 50 31 0.52IIMIIIII omIZIEE111 11.IIIIII on the graph and the short regression 40 =MINIM line indicate the position of Belgian 30 91111111MIMIN IMMI11111 plant-soilB datarelativetothe cp_ respectiveinternationaldata(long 9720 MEW= =11111 regression line and Zones IV; for 0.) details see Chapter 4). 10 1101=11MMMIILMMI11I 1111115111111111 "68 1,%1111AILMNPlia.1=iMIMIIIMPIMMENI 6 111.111 =ii1211101111111 5 IIMaiMNEFIENN=MMIMINil 4 11=ENNIII MINEMIIII o 3 1.111mOMMIii Minii co 2 I Wili1111 MEER

11011 DI11111 .2 .3 .4.5.6 .810 2 3456 8 10 B in soil, mg/I (Hot w. sol.)

Fig. 59. Regression of B content ofpot- 100 grown wheat (y) on CEC-corrected soil 80 60 n= 35 8 B (x), Belgium. tl 10 694 50 7ts=0 027 o 7 40 E .30 =578+19x Y 1. X a. r= 0 469* 08 20 08

4E.G,

o 4

.08:1 .2.3 .4 .5.6 .81.0 2 3 4 568 10 CEC -corrected Bin soil, mg/t (Hot w soL)

Copper. The Cu contents of Belgian plant and soil samples and their mutual regressions are given in Figures 60 and 61 against the background of the international data. The great majority of the Cu values fall within the "normal" international range (Zone III). The correction of soil Cu values for soil organic carbon improves the plant Cusoil Cu correlation but does not otherwise change the position of Belgium in the "international Cu field" (Figs 27 and 29, Section 2.3.4). Only one sample pair falls in the highest Cu Zone (V) and one in Zone IV. The extremely high plant Cu content (20.6 ppm) of the latter (sample pair 43301) is obviously due to Cu contamination, because both the Cu content of that soil as well as that of the original plant (6.6 ppm) are "normal". 112

22 Fig.60. Regression of Cu content of pot-grown wheat (y) on acid 20 ammonium acetate-EDTA extractable 18111 ìiinffill111111111111111 soil Cu (x), Belgium. E S.161111MEITEI MINIM o w14111111111111111111111111111111 _c .5 12111111111111111111111111811 V,10 111111111=11111E11211111 8 8 1111111111111111111M111111 06

2MiellifilltE11111111=1111; 111111111111111111M1111111 .4.6 .810 2 46810 20 40 6080 Cu in soil, mg/I (AA Ac- EDTA)

Fig.61. Regression of Cu content of 22 pot-grown wheat (y) on soil Cu correct- Internas 20i!IIIII 11111111111111111 ed for organic carbon (x), Belgium. 18 riuoi 1111111111111111 1111111111 11111111111111 1111111111M inummoinm 111111P AIN11111111 EL 1111111 IIIME1111111111 o 61111111111111119111101LIMIIIIIIII 11111111M1IIIIIMMOI11111

2 liniii1611111111INOILI

.4.6 .810 2 4 6 810 20 40 60 100 Org.C-corrected soi CU,mg/l (AAAc-EDTA)

Iron. The average Fe content of Belgian plants is slightly below the international mean, but the mean extractable Fe content of the soils is twice that of the international mean (Figs 31 and 62). Two-thirds of the Belgian Fe values fall in the normal range (Zone III) but one-third lies within the two highest Zones (IV and V). Manganese. In spite of the limited material, the variations in Mn contents of Belgian plants and soils are relatively wide (Figs 63 and 64). Most of the values, however, fall within the normal range (Zone III), and the Belgian national means of plant and soil Mn (Figs 33 and 37) deviate less from the international means than those of most other countries. Because of the quite wide pH variation in Belgian soils (from 4.45 to 7.40), pH correction of DTPA extractable soils Mn values improved the correlation between plant Mn and soil Mn substantially (from r = 0.347* to 0.737***). The Mn contents of only one 113

Fig.62. Regression of Fe content of 300 pot-grown wheat (y) on acid ammo- 275 2501/701211111E1111119 nium acetate-EDTA extractablesoil 225MHEINIVIOIMMUMI111111111111mi Fe (x), Belgium. 2 ObMIESIEUNEWINAINI1111111111111 1 75LW r°9011111/1111515-i.iiiiini3111111 k1501111111111M1111111111111111 . 125111111111111101111111111111

coD 100 15 90 80 70 11U11111 iII!!!i 60 501111.1111=1111111111

40Ni11111111111111111111 11111111111111111111 30--tontinmmmitaisin 10 20 40 60 100200400 6C0 1000 2000 Fe in soil. mg/l MAAc- EDTAl

sample pair (43327) fall in the lowest Mn Zone (I) and those of four pairs (43321-22, 43328-29) in Zone IV. All these Belgian extreme values originate from the northeastern part of the country. Molybdenum. The Mo contents of individual pot-grown wheat samples and respective soils are given in Figs 65 and 66. The ranges of variation in Mo contents of soils and plants are rather small in comparison to the whole international material. The highest Mo content found in Belgian wheat samples exceeds the lowest by a factor of 32 and in the whole international material by a factor of almost 1 000. Corresponding factors for the A0-0A extractable Mo contents of Belgian soils are 6 (uncorrected) and 17 (pH-corrected), and for the international material over 500 and 1 500, respectively. The correlation between plant and soil Mo contents in Belgian material is not significant. Correction of soil_ Mo values for pH changed the direction of the regression from negative to positive, but Belgium still remains one of the three countries where pH-correction is unable to improve the correlation to a statistically significant level (Table 8, Section 2.3.2.2). The mean Mo content of plants is slightly above the international mean and that of soils (Fig. 66) is lower than the international average. All Belgian Mo values fall within the normal range (Zone III), and there are no extreme Mo contents that might indicate possible Mo deficiency or toxicity. See also Figures 15 and 20. Zinc. The relationships between Zn contents of Belgian plants and the respective soils determined by two different extraction methods are presented in Figures 67 and 68. In both cases the correlations are very high and the mutual locations of individual points are very similar', irrespective of the extraction method used. Almost all the Zn contents of plants and soils are above the international means. The Belgian national mean for plant Zn content is twice as high as the international mean and those for soil Zn (AAAc-EDTA and DTPA extractable) are over sixfold the respective 114

2000 Fig. 63. Regression of Mn content of pot-grown wheat (y) on DTPA extract- ± 7531111751/1111111.11 able soil Mn (x), Belgium. 1000 ZION.IMEMItYMEWIAIMMIONNIiN11=011EilMMIMPII111MIIMIMMIIM 800 .kTYWOMMINIK4M147,1173V611111111 600 Ili===iMKINOUMMMENO 400 1111IMMERIIII=MINIE g: 300 nummmollimm -tai 200 111=1111111111MICI

100 11111111111111 80 =1M1==1.00.21111Mli t 60 EMBI!1MIT2=MMEPOlsuramimiamsIIImm 40 lareimmximmaimmoi 30 20 rill.1111111111 10 111111 8 MONraMILIMMIIII1111inIIMMIMMIIMINE 23 46 810 20 30 40 60 80100 200 300 Mn in soil, mg/l DTPA)

2000 Mmirigramt Fig. 64. Regression of Mn content of pot-grown wheat (y) on pH-corrected 1000 mm...6,P.IN;epv.=mmAwromussmiwisoummaEill11111111111111111111111 DTPA extractable soil Mn (x), 800 NUMMENNEME,MUMIMINNUMMWIIIIINMAL12:12EkEIESAMULAMMENNIIIMMIIIMMI Belgium. MMTIORINCHIELE=MNITITIEFEIIMMIIIMIMIMMINIIIIN 600IIM MINBEFZEIREMIEMMWMIGNMEIMUMMEMEMMEIM 400 300 200

10011.111111111.11111111111111 80sommoonsmmom-emoniimmonmmmolosriViaE.:MHIMMImAa...=monss 60II 40IIMINEMPCIUMMIEMMEIMIMMIN 30 201Iil!iilIIIIIIIUlÌiN

10iriiiiiiiiìuiiiiiiiuiiiiiiiM=IMINEN 8 =MMMIN ==MINIMMIN ieraammuna =1IMMILAI

1 2 3 46 810 20 301.0 60 100 200 300600 pH - corrected Mn in soil, mg/i (DTPA) international averages. Malta is the only other country where the plant and soil Zn contents are at the same high level as those of Belgium (Figs 41 and 43). Half of the points in the Figures fall in the highest Zn range (Zone V) and the other half are in Zone IV or in the higher part of Zone III. The highest Zn values were recorded in plants and soils from the northeastern part of Belgium. It appears therefore that the possible Zn problems in Belgium are not those of deficiency but of excess. Industrial pollution is likely to be at least partly responsible for the high Zn values of Belgian soils and crops. For example, data from the UK indicate that as much as 370 to 4340 grammes of Zn per hectare/year may be deposited in non-urban sites (Cawse 1980, Wadsworth and Webber 1980). 115

Fig. 65.Relationshipbetween Mo content of pot-grown wheat (y) and 6MO =6 42 03200 36 ammonium oxalate-oxalic acid extract- ± 0 1 85 021±025 able soil Mo (x), Belgium. 3 log 3 843x y.0 2 r2 t r 0 7 0 2 2 E 0_ a. 1 ci ,0.6 3 o

<7, .2 o o '1 .06

.03 .02

I II .01

.01 .02 .03 .06 .1 .2.3 .6 1 2 3 Mo in soil, mg/l M 0 - 0A I

MN Fig.66.Relationships between Mo M=1011M111 Mmorimas11110111111M1111111111110 M oMI 114Q11. IIIMOMMIMUlairl7Mira=LMNIMMONINI111=1.111 content of pot-grown wheat (y) and 6 uILUIIUfl pH-corrected A0-0A extractable soil EIMMIIIIMMERINEMEMESEEKEEMIIIMEINIMI Mo (x), Belgium. 3NUMB-- °11238°11111111112EILIIIIIIIMMEll 21111111°3! 5^3:

E -,c)- 1 2.6MIMIII=ESEINUIM=MIIMICRIEIIMIMMI _c ENNIIIIMEIMIIIIMM=PME1111111111MMENIIIMIMIMENIIM=IMMS112101===11M .3 r c iÍiHIIUIIIIflhIiIIhHI-*411111 a> EMMIN=1M111RINIMrnelorammiewmkme.11111.11111111111111MMI =MMUMNIM...7TIMMENNENMIMMENNEU=0O==l1M.M. LIMIMINIMIM MUMENIII =MMIIMMIIMMI=Mal I N .03 .02

.01 .006 .01 .02.03 06 . .2.3 .6 1 23 pH- corrected MO in soil,mg/l (AO-0A)

5.1.4 Summary

The Belgian soils are generally acid, mostly coarse in texture, have a medium organic matter content and a low cation exchange capacity. The P contents of plants and soils are the highest recorded in this study. K, Ca and N are usually at the normal level but the N contents of maize are very high. The Mg contents of plants and soils are among the lowest of the 30 countries in this study. The B, Cu, Mn and Mo contents correspond more or less to the normal international level, Fe is somewhat high and the Zn content of most sampled Belgian soils and plants is exceptionally high. 116

200 11111 Fig.67. Regression of Zn content of '.77 pot-grown wheat (y) on DTPA extract- 5 able soil Zn (x), Belgium. 321 log 100 90 80 70 E 60

20 o N

10 9 8 7 6 DI

.1 .2.3 .4.6 .81 2 3 46 810 20 3040 60 100 Zn in soil, mg/1 (DTPA)

200 Fig. 68. Regression of Zn content of Zn 1 pot-grown wheat (y) on pH-corrected 8 9 I AAAc-EDTA extractable soil Zn (x), . 11 100 Belgium. 90 80 1.11111.6xIIIIiiiiIMMANIN 70 60 E 50 "Iniglimillf leliIII 11 mom

30 01 1 Hopmein o I 1; 20 9 1-1111lb I1 ill o 0 1

NJ 10 11,I IL 9 MERL MMIIII MM1111 8 =NMI 11101111MAMMIIIIII 7 MMINIILME111111111MSO1111 NIMIk 111111 al 111111 6 EMMIrka WWII mnElliri .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soimg/1 AAAc-EDTA) 117 5.2 Finland

5.2.1 General

The wheat and soil sampling sites in Finland are shown in Fig. 69. The sampled area consists only of the southern part of the country since wheat is not successfully grown in the northern parts. Maize is not normally grown in Finland. The frequency distributions of texture (TI), pH(CaC12), organic carbon content, and cation exchange capacity of Finnish soils are given in Fig. 70 and data on other soil properties in Appendixes 2 and 4.

Fig.69.SamplingsitesinFinland 26° 28° 54° 36° (wheat fields). The last three numerals of each sample pair number (48151 48257) are given.

66°

64°

62°

60°

26° 118

5 Finland Internal Finland Internet Fig. 70.Frequency distributionsof n= 94 3764 n= 94 3783 40 texture, pH, organic carbon content, =45 44 R= 5.21 6.64 o g-18 16 5s= .52 1.12 min= 16 9 35 min.= 4.10 3.62 and cation exchange capacity in soils of max.= 86 92 mox = 6.90 8.56 30 Finland (columns). Curves show the international frequency of the same 20 25 characters. 20

0 10

0 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca C17)

Finland Internat. 5 Finland Internat. n= 94 3779 n=94 3777 R.3.9 1.3 31.8 27.3 0 5s= 2.8 1.2 1-s= 13.9 14.6 35 rmn, 1.4 .1 min.= 12.5 .2 max . 17.5 39.1 max = 89.7 99.9 a. 25 o 20 o ui15

2 4 6 16 32 68 12.8 25.6 16 32 ORGANICC,% CEO, me/ 009

The average texture of the Finnish soils is very similar to that of all the soils studied and the range of variation is exceptionally wide. In general, the finest textured soils occur in the coastal areas of southern and western Finland and the coarsest in the central parts of the country. Most of the fine textured soils were classified as Cambisols and the coarse textured as Podsols. The soils are generally acid. The pH(CaC12) varies from 4.1 to 6.9 and the national average (5.2) is among the lowest found in this study. The average organic matter content (org. C 3.9 %) of Finnish soils (including five peat soils) is three times as high as the international mean (1.3 %). Only the soils of New Zealand are higher in organic matter. Compared to all soils studied, the cation exchange capacity of Finnish soils is slightly on the high side and the range of variation is relatively wide. Electrical conductivity, CaCO3 equivalent and sodium contents of soils are low compared to most other countries (Appendixes 2 and 4).

5.2.2 Macronutrients

The frequency distributions of nitrogen, phosphorus, potassium, calsium and magnesium contents in original Finnish wheat samples and in respective soils are shown in Figures 71-75. To compare the average macronutrient situation in Finland to that in other countries, see Figs 6-10 (Section 2.2.3). The mean nitrogen content of the soils (0.288 %) is more than twice as high as the international mean (0.133 %) and it is exceeded only by that of New Zealand (0.340 %; Appendix 2). Only two countries, Hungary and the Republic of Korea, have higher national averages for the N content of wheat (5.46 and 5.27 %, respectively) than Finland (5.03 %). The higher rates of nitrogen applied to the sampled wheat crops in Hungary 119

4. 35 Finland Internat. FinlandInternat. n= 94 1765 ris 93 1765 Finland Interno t. 35 " 30 0: 4.3 4.0 Rs 5.03 4.27 n:94 1765 ±S. .76 1.15 35 6=.43 .38 ss= 0.6 1.0 1.5=.09 .12 4.6 min = 3.3 0.9 30 min.= 2.64 .60 ..e. max.= 7.17 7.45 min...23 .05 25 max.= 6.7 6.8 o 2330 max.=.75 102 25 03 0-25 M a' 20 >- - 20 O us 15 1015 15 o o Lu 10 u_ u-

mvz.,

3 4 5 6 2 3 4 5 6 N content of wheat, % P content of wheat, % K content of wheat% 40 Finland Internat. 70 intemcit n= 94 1765 Finland 0: 61.0 202 5 Finland Internat. ne 94 1765 55. 32.5 24.7 60 8..288 .133 mia= 142 1.0 n: 1765 +s= .160 .084 max+208.0 271.4 si 205 365 so min.= .106 .009 30 ±, 100 283 maa=1.023 1.023 nnn, 40 20 50 o o max.= 632 2097 ,cri, 25 a 40

D15 o cc 10 u_ o /19' ;,' ...AA 2 ...../..W. 2.5 20 80 160 320 25 50 100 200 400 800 1600 3200 6400 .005 0 .02 .04 .08 .6 .32 64 1.28 5 10 40 640 N in wheat soils, % P in wheat soils, mg/l K in wheat soils, mg/I Fig. 71. Nitrogen, Finland. Fig. 72 Phosphorus, Finland. Fig. 73. Potassium, Finland.

4 40 Finland Internat. Finland Internat. Figs 71-75. Frequency distributions 35 n: 94 n =1765 n.94 1764 =.40 R...43 4:142 .172 of nitrogen, phosphorus, potassium, es =.09 ±ssi.17 35..038 .060 min...22 min.=.11 30 min.=.076 .044 calcium and magnesium in original max+.65 max.= 1.66 max...285 .948 wheat samples and respective soils P.25 25 >- 3- (columns) of Finland. Curves show 0 20 0 20 the international frequency of the D15 o oD15 same characters. Ui ix 10

1.6 5 2 Ca content of wheat.% Mg content of wheat%

80 Finiand Interno). Internat. n= 94 1765 Finland 35 R= 1521 4671 35 n= 94 1764 ts =969 3076 R=207 489 min.=190 110 ±s= 213 437 max = 8800 21930 min.= 10 10 30 max.= 1075 3298

025 25

20 20 >- >- O O

Ill 15 L.LI15 o o 1.1.1 Lu ct 10

100 200 900 884 16003 00 6400 12800 25600 25 50 100 200 400 800 1600 3200 Ca in wheat soils, mg/I Mg in wheat soils, mg/I Fig. 74. Calcium, Finland. Fig. 75. Magnesium, Finland. 120 (140 ± 74 kg N/ha) and Korea (100 ± 25 kg) compared to Finland (76 ± 24 kg) and New Zealand (9 ± 16 kg) are likely to explain these differences (Appendix 5). Peat soils and other soils of high organic matter content were, in general, higher in N than other Finnish soils. See also Fig. 6. The mean phosphorus (NaHCO3 sol.) content of the soils is the third highest (after Belgium and Malta) in this study, and the average P content of wheat is exceeded only by those of Belgium, Malta and Hungary (Appendix 2). In all these countries applications of phosphorus fertilizer to the sampled wheat crops were above the international average (Belgium 42 ± 9, Finland 43 ± 26, Hungary 52 ± 55, Malta 26 ± 0, international mean 21 ± 27 kg P/ha, Appendix 5). In general, the highest soluble soil P and plant P contents in Finnish material were found in coarse and medium textured soils and in plants grown in those soils, apparently because both the applied and the native soil P were less firmly fixed in these soils than in finer textured soils. See also Figs 7 and 12 in Sections 2.2.3-2.2.4. The mean potassium content of the original Finnish wheats is at the international average in spite of the relatively low exchangeable K of the soils (Figs 73 and 8). Evidently, the high potassium application (mean 63 ± 19, international mean 21 ± 44 kg K/ha) compensates for the low contents of native soil K. The generally low pH of the soils may also contribute to the relatively high K content of the plants (See Fig. 11, Section 2.2.4). The calcium contents of the wheats are only slightly, but the exchangeable Ca of soils are considerably below the international average (Fig. 74, Appendix 2). See also Fig. 9 and related text. The magnesium content of the soils varies widely from one soil to another, but the national average is one of the lowest among the countries in this study. The low soil Mg contents are also reflected in plants. Only five countries have lower national mean Mg contents of wheat than Finland (Appendix 2), but when maize growing countries are' included in the comparison (pooled plant Mg contents, Fig. 10) Finland's position is the ninth lowest. Response to Mg is likely in several of the sites sampled.

5.2.3 Micronutrients

Boron. The B contents of Finnish soils and pot-grown wheats are given in Figures 76 and 77 and show that both soils and plants contain slightly less than the international average (see also Figures 22 and 25). A great majority of the sample pairs are within the normal B range (Zone III) and only a few in the two lowest or the two highest B Zones. No clear geographical differences between the high B and low B soils can be distinguished. Low and high B values can often be found in fields located relatively close to each other (e.g. sampling sites 48159 and -160, 48177 and -178, 48199 an -200, 48249 and -250). The limited number of samples in this study may give too high an estimate of the general B status of Finnish soils since the average hot water soluble soil B content (0.55 mg/1) is somewhat higher than those (based on more numerous samples) published earlier by Sippola and Tares (1978) and Kurki (1979), 0.38 and 0.48 mg/1, respectively. The latter data, based on the results of general soil testing, showed also that theaverage B content of Finnish soils has been improving from the years 1966-1970 (0.41 mg/1) to 1976-77 (0.48 mg/1). The improvement was attributed to increased B fertilization following soil testing and to the addition of some B to most of the fertilizers generally used in Finland. 121

100 Fig. 76. Regression of B content of pot- B imm onanium=MI grown wheat (y) on hot water soluble 80 0MINLEMO 60 n= Mr. soil B (x),Finland. For details of 50 IMMEEEMMT Mil 111M summarized international background 40 data, see Chapter 4. E30 ENEIMMEIMMI PflI 20 iiiuui "es _c 111111 10...111 111111..10=11 "68 1=1=111 MINNIIN MNIM a.) 5 11= iuimui l NMI o 4nomPosmimil inn co 3 2 N111

111 1111111 BI 1 .1 .2.3 .4 .5.6 .81.0 2 3456 8 10 B in soil, mg/l (Hot w. sol.)

Fig.77.Regression of B content of 100 80 pot-grown wheat (y) on CEC-corrected n= 3 (hot water soluble) soil B (x), Finland. 60 ±s=5 4 132 60 t 50 7±s =05 027 40 E y= 412+ 94x o o_ 30 r 0 605 ** 20

_C 0810 -ta'0)56 o4 C.) 3 CO 2

.08.1 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 C EC -corrected B in soil, mg /l (Hot w. sol.)

Copper. The national averages for Cu contents in wheat and soils are slightly on the low side compared to the whole international range (Figs 78, 79, 27 and 29). Because of the relatively high organic matter contents of Finnish soils, the correction of AAAc-EDTA extractable soil Cu values for organic carbon reduces the soil Cu values somewhat in relation to other countries and so lowers their position in the "international Cu field". Although most of the Cu values remain within the normal range, the relative number of Cu values falling in the low Cu Zones (I and II) becomes more than doubled and indicates that the possibility of response to copper fertilization is somewhat more likely than in most other countries. 122

22 Fig.78. Regression of Cu content of C d Fi len, Inte nat I\ pot-grown wheat (y) on acid ammonium 20 n= 3538 5,-± 5 99 ±11 7002.57 acetate-EDTA extractable soil Cu (x), Tc± 431 # 268 6.00-7.05 18 Finland. E Q. Lc'?6=3 54 + 3 292 . g x 11374 9'2

c .6 12 10 hill 88 06 I\ ' ..,40.1.41 \ 4 111WW_ 111 MM. _...... wI 2 --.1wommilla, IN -...... 7. - 1 ,n a n , 1 G. LI A" ,rt 1n crtnn . . Cu in soil, mg/t /AAA c- EDTA)

Fig.79. Regression of Cu content of 22-Cu pot-grown wheat (y) on organic carbon- 20 corrected (AAAc-EDTA extr.) soil Cu (x), Finland. 18

r=0622 '' 1 16l'It'211.111 11111.1 111,11A 14 Ill I 12 ;Ill

8 . .. 6 11 AIN 4 111111111tiii11ldilli1111111 1111 2 Pliillittal 1019111

O illig1131.1111MIIIIN .2 .4 .6 .810 2 4 6 810 20 40 60 100 Org.C-corrected soi CU,mg/l (AAAc-EDTA)

Most of the samples within the low Cu Zones I and II represent relatively coarse textured soils in the central parts of the country. Previous investigations (e.g. Tainio, 1960; Tähtinen, 1971) indicate that Cu deficiency occurs quite frequently in Finland, especially in the peat soils and in the coarse textured mineral soils of the northern parts of the country (for reasons explained earlier, the present material does not include samples from the northern parts). The average AAAc-EDTA extractable Cu contents of Finnish soils, based on a substantially larger number of soil samples (n = 2015) including also northern Fin- land, is 2.83 ± 2.90 (Sippola and Tares, 1978). This suggests that Cu deficiency may be more common in Finland than is indicated by the limited data of this study. According to soil testing data for the period 1955-1977 (Kurki, 1979) the Cu status of Finnish soils was declining until about 1970, but since then a slight improvement has been observed. This was attributed to the increased use of fertilizers containing Cu. 123

Regression of Fe contents 300 Fig.80. 275pgrammulooromminnme ofpot-grownwheat(y)onacid 250sammosonomummonumis ammonium acetate-EDTA extractable 225 soil Fe (x), Finland. 200=1611111116i11611111111111111111 1 75 E 1 50M1111111111111111.111 a. - 125IMM111111111111111A11111111 o 100 90 80 70 L11111111 R!M!!I o o60

40 iNIIIIIIiiÌIISI 11111111111111111111 30I=IMIIIINtillPRIEN11111M811 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (A AA c- EDTA)

Iron. The Fe contents of plant and soil samples are plotted in Fig. 80. The plantFesoil Fe correlation does not reach the 5 percent but only the 10 percentsignificance level. Both the average Fe contents of plants and soils exceed the respectiveinternational means, and Finland is the only country in which the national plant-soil averagefalls in the highest Fe Zone (V) (Fig. 31). Almost half of the samples fall within ZoneV and less than one-third within the normal range (Zone III) (Fig. 80). The soils havingthe highest extractable Fe contents (> 1000 mg/1) include the five peat soilsand soils with high clay and organic matter contents, high cation exchange capacityand low pH. The average AAAc-EDTA extractable Fe content of 2015 Finnish soils (677 ± 656 mg/1) presentedby Sippola and Tares (1978) was still slightly higher than the national averagerecorded in this study (569 ± 367 mg/1). Manganese. The regressions of Mn contents of pot-grown wheat onthe DTPA extractable soil Mn without and with pH corrections are given in Figures81 and 82. Correcting DTPA Mn for pH changes the distribution of points in the Figure,improves the correlation from being non-significant (r =0.104 n.s.) to a highly significantlevel (r = 0.639***), and moves the Finnish material to the right in the "international Mnfield" (see Figures 33 and 37). Notwithstanding that the Finnish soils are generallyacid with high availability of Mn to plants (Fig. 34), the national mean values for plant Mnand extractable soil Mn are lower than the respective mean values for the whole internationalmaterial. This is an indication of relatively low total Mn contents of Finnishsoils. However, cases of "primary" deficiency (due to low total Mn content) are very rarecompared to "secondary" deficiency (low availability mainly due to high soil pH). Most of the Finnish samples fall in the normal Mn Zone (III),only a couple of sample pairs are in the two lowest and three in the two highest zones (Fig.82). The highest Mn contents of plants and soils were found in samplescollected from sites where acid soils (pH <5) prevailed and the lowest from sites having soils of onlymoderate acidity. 124

2000 17111:-7111-1.1 MIMMIONMF Fig. 81. Regression of Mn content of pot-grown wheat (y) on DTPA extract- 1111 1000 .VXM.41MMONNEN IMAM VAINEZ01/Matb.11.1MINIMMI able soil Mn (x), Finland. 800NM VMMIllEra-MMEM:ITMIFPritervrimamil MOMIMM". 600il mmummimmmommuli MEMI 400II D300 20011=11111111111.11111 .c

"CI 10011011111111111111.41 80 mumm.. 111 1111111111MINIMINIMIMMENINI &WIER 860BINIONEMMORIMNISIMINIMMMIIII 40uiiìiuuiuiIM 30iiiiiMPAINUMMEMMIN 20 1111

1010111.1111rErird 8111111111111111111111111MINIIIIIMIlia=0.SMIMINM

2 3 46 8 10 20 30 40 60 80100 200 300 Mn in soil, mg/l (DTPA)

2000 65-1-10....1111.011 Fig. 82. Regression of Mn content of M n pot-grown wheat (y) on DTPA extract- 1000.. arrinimiimm....r...... mon=w.ra-un. able soil Mn corrected for pH (x), 800.. =1MMISMU1=1101MINMINMIMMINE. Finland. 600NI MIERMISMEMIImirrnmanimaimmou E 11111=1- 711i2E111111 PP" 0- 400 4011111 300 1111111MEMPAIIA ...41 L.1.01111111.1.1114M1111111 I 200 o 11.1111111111111111-1101111111 "`c- 100umvimmo...... 1...i.....7.01....wwX01M11111^^711 BO11._ .;MI=MMEMIUMFIIMPFainiMMMEH MIN161111 60inNOWAMEraiiiilliMMIN=IMMEMENE K1M=MITHE o 11.11=111EMEMEOPMMMMEIMENI MEMMEIEffig 40 r . 30 201111.1dillaMli11111111111.6ii OP' 111

10 ma ummatailmwmons MMMINEN 8011 INEMMITil IIM=MIM421 2 3 46 810 20 30 40 60 100200 300600 pH-corrected Mn in soil, mg/l (DTPA)

Molybdenum. The average Mo content of wheat grown on Finnish soils is somewhat lower than the international average, while the mean of ammonium oxalate-oxalic acid extractable soil Mo exceeds the international mean by a factor of 2.6 (Fig. 83). This inconsistency is eliminated by correcting the soil Mo values for pH (Fig. 84). In consequence, points in the Figure took up new positions; the plant Mosoil Mo correlation improved substantially and the whole Finnish material is moved to the left in the "international Mo field" (see also Figures 15 and 20). After pH correction only four points remain in the two highest Mo Zones (IV and V), one only is in Zone II and most of the remainder are within the normal range. 125

Fig.83.Regression of Mo content MZEPINLW=PSM of pot-grown wheat (y) on ammonium 6 .4 ...... _ =limETIMIBEIMIMICEIBLEINumuniiimamongs oxalate-oxalicacidextractablesoil mansammo 2Pill Mo (x), Finland. 3 logy 2 iiiiiils'ill E 0. 11111E1 IN MMINEMIIMIIMIIMINNIE_.....__a_NINO ninillME=CREMENMIIMENULIMINICIMEMINI ....11 1=11.1011Mill- mmunimmememinB11111MiMallaill it,\MIIIIIHME15210"- N1111111.11111111\L o '1...... amommemmErmloam AMMIIIIMMMENNEMIIIMIMPEME111.1kWIMEMENIIIMI=MEMM101MLTA. .06WimmouNi=MENENI.wIIIUMNIIMMENNII.

.03 .02

.0111111161111111...... 01 .02 .03 .06 . .2 .3 .6 1 23 Mo in soil, mg/ (A0-0A)

Fig. 84. Regression of Mo content of Moulsromm-cun.,,;,,i-maisto pot-grown wheat (y) on pH-corrected 6 extractablesoilMo(x), munizerr-9°Erafflaigiiiiirommanni A0-0A 111111114ORMIWIELICEIL1101111 Finland. 3 2

E o. 1 15; --...... a,.6 m _c immmil=5.uuiiiuiuuuniiijii ó.3 .2iniii.iiijiii.iiiiiii \ 5 oo .1121221111==0:11:111,====11:1 2 IMINEIMMINOI 06MMI111111115111MMEMBIIMMI=MMIllMMIMENNEMI11

.03 .02

.01

.006 .01 .02 .03 .06 .1 .2.3 .6 1 2 3 p14- corrected MO in soil. mg/I (A0-0A)

The highest Mo values were found in samples from sites where the soils have pH above the average and medium texture, while the lowest values were recorded from sites of low soil pH and either coarse (TI 16-30) or fine (TI 50-80) soil texture. In coarse textured soils the low Mo values are apparently due to low total contents and in fine textured soils due to low plant availability (see also Fig. 16). No clear geographical distinction can be made between the high and low Mo areas. The present material does not indicate any serious Mo problems in Finland. However, heavy nitrogen applications (300-450 kg N/ha/yr) commonly practised by fodder producers may induce Mo deficiency on acid soils if the pH is not kept at a reasonable level by liming (Rinne et al., 1974 b). 126

200 Fig.85.Regression of Zn content Zn t-8 11,7 of pot-grown wheat (y) on DTPA i' 1 ,34 1 extractable soil Zn (x), Finland. ,y c Ili 321 logK 100 90 80 70il VII E60 "i111 50 illii "t5 40

"a 30\ \ .cl

.411;11111PV -s 20 I N

10 111111111111 9 8 / wir 7 6 1 N m r'\ .3 .4 . .81 23 46 810 20 3040 60 100 ZD in soil, mg /I (DTPA)

200 Ite n Fig. 86.Regression of Zn content of Znlc°' 3 3 pot-grown wheat (y) on AAAc-EDTA 2 :4 3 I: Y extractable soil Zn corrected for pH 6 11. x isi 100 .kris (x), Finland. 90 'aIIMII. 80 MEEVill 70 60 I i MI E 50 1=1 2: 40 ill P7 o cu 30 I IlkIC o 20 PPPAn o 0.10.0L

10 9 8 : ....., 7 I MMMENIIIIIMEL 6 1 W IIi ri I Ili E \ .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soi ,mg/i AAAc-EDTA)

Zinc. The national averages for both plant Zn and extractable soil Zn are distinctly above the respective international averages (Figures 85 and 86). Irrespective of the extraction method used (DTPA or pH-corrected AAAc-EDTA), the correlations between plant and soil Zn are high. Comparison of Figures 85 and 86 shows that the locations of individual points have changed little. With only a few exceptions, the same points falling in Zones V and IV when DTPA was used as an extractant are within the same zones when the soil Zn contents were determined by AAAc-EDTA method and corrected for pH. 127 No Zn values fall in the two lowest Zn zones, and although almost one-third of the Zn points are within the two highest zones no extremely high Zn contents were found (see e.g. Figs 67 and 68). Geographically, the lowest and highest Zn values are quite evenly distributed all over the southern part of the country.

5.2.4 Summary

The Finnish soils are generally acid with widely varying texture, high organic matter content and relatively high cation exchange capacity. The P and N contents of soils and original plants are among the highest in this study and those of Mg among the lowest. The K and Ca contents of wheat are average but those of soi/s are low. The Mn and Mo contents of Finnish soils and pot-grown wheat are, with few exceptions, within the normal international range. The Cu and B contents are somewhat below but those of Zn and Fe distinctly above those of most other countries. It appears that no extensive, but some local (Cu and B), micronutrient problems are to be expected in Finland. 128 5.3 Hungary

5.3.1 General

The sarnpling sites in Hungary were well distributed across the country (Fig. 87) and soil properties therefore varied widely. About half of the soils sampied were classified as Phaeozems (125). Other soils commonly found were Chernozems (46), Luvisols (25), Vertisols (15) and Cambisols (10). The frequency distribution graphs for soil texture, pH and cation exchange capacity given in Fig. 88 show that on average, the soil properties in the Hungarian sample material correspond closely with the whole international material and the ranges of variation are almost as wide as those internationally. Organic matter contents in Hungarian soils are relatively high and quite uniform. The average electrical conductivity and CaCO3 equivalent values of Hungarian wheat soils are lower than those internationally (Appendix 2), but in the case of maize soils the Hungarian averages exceed the respective international means (Appendix 3). With a few exceptions (mainly Halos°1s) the sodium contents of the soils are relatively low.

%...1 915B-9150 05-4.0 .--,0- 121. 0252-4255

I 7.7'''':------Ì----- J....H____2-.f$ 2,2_ 4125.-4125 '1""2`±_l_7 , 1

I :,.... 5,,, :,::::)z..4,.....4,,,,i4.: 41.,3:7:14:01:5:2-1

,1 5145-2140

, 1'4129-.131 1 a,6-11aa ,, ,, , 9 5116-9J25 1 9132-4154 : .60 , 1 BUDAPEST 4252 I 914517;:, l'14..::43' 4'197-0156 1.- te427221.2 4235 9252-4232 \\''" \I 412- .1255-,4260 .,52 ,, 1 -.4 9258 e i215-,4; ::, 24 0.2 1 ' 4i82-4,52 j ---- '''..-:15''' -%15:51T4:'28'2'-"4f6"\- ' t ' :40'0: 725_46:46:7 ,11;11.0,-.4i41 4 '69 1 / 01212,,,,,_ ':?4RV94253 9169-4160 7- 4' 2 9 i 1 4252-.42Li ,L,_,...... ,i- --j., ->.1 1( 2136;9142 4167-4169 I 2.\o 4112-6113415144.273._:. 1 9146;9131

0/ 2 1 \ ., 4 0214-.215 1 .4160-4163.115-4. 4292-92,6 ';211. 6 4206-4217

4161-911 1

4170 $2,"12.

YecFL. 10 1 /.4/,ii 4,4-4,00 .;e4

1 42, tji25-4225 02 - 4300 if. I

2011

Fig. 87. Sampling sites in Hungary (points wheat fields, triangles = maize fields). The last four numerals of each sample pair number (44101-44300 and 49101-49150) are given. 129

Fig.88.Frequency distributionsof 5 Hungary Internat, Hungary Interno! nz 250 3764 40 n= 250 3283 texture, pH, organic carbon content, 6. 47 44 =6.89 6.64 30 os= 12 16 os= .88 1,12 and cation exchange capacity in soils t min.=15 9 35 min 4.20 3.62 mox a84 92 max=8.00 8.56 of Hungary (columns). Curves show 30 theinternationalfrequencyofthe 'CD 20 2 same characters. >- W 15 20 o

10 20 30 0 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

5 Hungary Internet Hungary Internet 40 n.250 3779 n= 250 3777 k = 1.6 1.3 30 0= 29.7 27.3 Z 35 os = 6 1.2 08: 10.6 14.6 min.= 4 .0 min.=4.6 2 0'./1.6 max.= 3 9 39.1 max.= 64.2 999

2 .8 16 32 6.4 12,8 25.6 2 4 8 16 ORGANICC,% CEC, me/100g

5.3.2 Macronutrients

The frequency distributions of macronutrient (N, P, K, Ca and Mg) contents of the original wheat and maize samples and respective soils are expressed in Figures 89 to 93. The Hungarian averages for nitrogen contents of soils are high compared to most other countries in this study (Appendixes 2 and 3, Figs 6 and 89). Hungarian wheats contain on average more N than those of any other country, and only Belgian maize samples have higher mean N contents than Hungarian maize. It is apparent that apart from the inherent high N contents of these soils, the liberal applications of nitrogen fertilizers to the sampled wheat and maize crops (140 ± 74 and 136 ± 58 kg N/ha, respectively) are further responsible for the high N contents of Hungarian plants. The average phosphorus contents of the original wheat and maize (0.49 and 0.50 %, respectively) are among the three highest national mean values recorded in this study (Figs 90 and 7). The contents of soluble P in Hungarian soils are also well above the average 130

Hungary Internat. Hungary Internat. Figs 89-93. Frequency distributions of 35 n= 144 1765 n= 106 1958 R= 5.46 4.27 0= 4.72 3.14 nitrogen, phosphorus, potassium, cal- =.78 1.15 ts= .50 .87 min.= 2.72 .60 min.= 2.93 .88 max.= 7.45 7.45 mox.= 5.87 6.51 cium and magnesium in original wheat 25 and maize samples and respective soils

20 (columns) of Hungary. Curves show

15 the international frequency of the same o characters. 10

2 3 4 5 6 N content of wheat, % N content of maize,%

70 HungaryInternat. Hungaryinterno). n= 144 1765 n= 106 1958 60 0..103 .133 7= .169 .135 SOSOS) .084 tS= .061 .088 min...052 .009 min.= .036 .008 max.= .331 1.023 max.= .404 1.657 (1550

o Z 30

IL 20

.005 .01 .02 .04 .06 .16 .32 .64 128 .005 0 .02 .04 08 .16 .32 N in wheat soils, % N in maize soils,% Fig. 89. Nitrogen, Hungary.

40 Hungaryinternot. Hungaryintm-i.t Fig. 90. Phosphorus, Hungary. 1765 35 n =144 n .=106 1967 = .49 .30 k= .50 .33 68= .13 .12 to= .10 .10 i.13 30 Min.= .26 .05 mm= .26 .05 max.= .83 1.02 max =1.04 1.04 0-25 >- C.) 20

815 L1.1 10 10

iririfeA7 .8 9 P content of wheat% P content of maize,%

40 40 Hungary Internat Hungary Internat. n= 144 35 1765 35 n= 106 1967 32.7 20.2 =38.9 22.5 AS= 20.3 24.7 =34.0 33.0 min.= 55 1.0 min.= 7.6 0.1 max=130.0 271.4 max.=322.0 656.0

2.5 5 10 20 40 BO 160 320 640 5 10 20 o. 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I

international level. The very high and largely varying rates of phosphorus fertilizers applied to the sampled crops (52 ± 55 and 52 ± 33 mg P/ha, respectively) also appear to have affected the P contents of Hungarian crops both by increasing the average plant P level and by widening the variation ranges. 131

Fig. 91. Potassium, Hungary. 35 HungaryInternat, HungaryInternet n= 144 1765 n= 106 1967 30 0=4.1 4.0 30 0. 4.1 3.1 es=0,9 1,0 es= 0.9 1.0 min.=1.5 0.9 mm =1.5 0.6 25 max=6.6 6.8 25 MaX =6.0 6.7 a o2- P

U..1 15 000k JAA4mil 44/.4 Ali&3 4 5 6 2 3 4 6 7 2 K content of wheat, % K content of maize,%

38 75 HungarytInternet ' Hungary Internat. n.144 7 1765 n= 106 1967 0.217 365 9,241 330 30 05=122 283 ±s= 136 356 mm.= 46 20 msn.=50 18 ..C.. mae.=843 2097 max.= 956 5598 t5-t' 25 P

20 o

U, 15

25 50 1 0 200 4 0 800 1600 3200 6400 25 50 100 200 4 0 800 1600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

Fig. 92. Calcium, Hungary. Hungary Internat. HungaryInternet. 1957 35 n=144 1765 5 n= 106 9= .53 .43 5= .77 .47 os= .11 .17 *s =.26 .20 .30 .11 0 min.=.33 .09 max.= .77 1.68 max.= 1.88 1.88

1°

.8 Ca content of wheat, % Co content of maize.%

Hungary internot Hungary Internet n= 144 1755 n= 106 1967 5 9.4910 4671 35 6.4858 3450 1893 3076 5 .2094 2815 min= 520 110 min e165 10 max, 8850 21930 max.= 8270 17995

100 200 400 80014003 00 5400 17900 25500 50 100 200 400 800 1600 3200 6400 12000 Ca in wheat soils mg/I Ca in maize soils. mg/I

The exchangeable potassium contents of most Hungarian soils are well below the international average. However, the average K content of maize is considerably above the international mean and that of wheat slightly so (Figs 91 and 8). Both crops had received generally high but widely varying rates of potassium fertilizer (98 ± 96 and 104 ± 56 kg 132

0 Hungary inter-not Hungaryintemt Fig. 93. Magnesium, Hungary. 5 nn124 1764 5 n. 105 1967 9..177 172 01,i R. .305 .251 7.9P.348 060r 01=.13E1 .159 3 1..3 min =.105 044 sin..144 .036 mox...399 9.8E1, 01.5,1,1.073 1.125 25

.05 Mg content of whea.t,% Mg content of rosa ze.%

40 Hungary Hungary Internot n-140 1764 n.106 1567 7.437 409 7- 40 445 55.310 437 ±65 262 462 mir 30 10 emn . 15 riom 1926 3290 P,X, /352 6490

25 50 170 2 0 400 003 3201 12.5 25 SO 1002 u 40 000 1600 00 5707 Mg in wheat soils, mg/I Mg in maize soils, mg/.

Klha, respectively) which is likely to explain the dissimilarity between the K contents of soils and plants as well as the wide variation of K contents in plants. The average level of potassium fertilization in Hungary was exceeded only by that of Belgium. Calcium contents of both indicator crops as well as of the respective soils are high (Figs 92 and 9). The Hungarian mean for Ca contents of maize is the highest and that of wheat is exceeded only by Turkey (Appendixes 2 and 3). Exchangeable magnesium contents of Hungarian soils as well as the Mg contents of wheat are very close to the international average but high Mg contents are typical of Hungarian maize (Fig. 93, Appendixes 2 and 3). According to these data a response to Mg fertilization seems unlikely. See also Fig. 10. 133

Fig. 94. Regression of B content of pot- 100 moiimmuss 80 igi1111111.11IIIhilL11101=1111ilfaiTi4IIMM MIMINMENIIM===121:: grown wheat (y) on hot water soluble 1:3 1=rjaKMEnall M=MMIIMI1 soil B (x), Hungary. For details of 60 50 .1.02 summarized international background 40min clUZLIE1111 .70 +1.62 data, see Chapter 4. 301.11 WHIM MEMO Q ca: 20Mil' INIM O/1111

Ell01.1MINNIV.W.MCWIMIMMENNNE111111111111111

61=11MOMMOSeriaM MMINMENI MIIIIKIEMEMIE113111MMEMIlli IMMIMINI10111111 E1111111 211 19p

.2.3 .4 .5.6 .8 10 2 3 4 5 6 8 10 B in soil, mg/I (Hot w. sol.)

Fig. 95. Regression of B content of pot- 100 grown wheat(y) on CEC-corrected 80i n. 1 3 (hot water soluble) soil B (x), Hungary. 60 --±s- 4154 6 50 7..ts. 0 8-.0 3 40 E -412+29x y 0 x - 30 r. 0 579 8

10 "5 8 C6 5 o 4 o co 3 2

I _ .081 .2.3 .4 .5.6 .8 1.0 23 4 56 8 10 C EC - corrected B in soil. mg /I ( Hot w. sol.)

5.3.3Micronutrients

Boron. The B status of the soils and the B content of pot-grown wheat are, in general, at a satisfactory level (Figures 94 and 95). Only four out of 201 sample pairs fall in the two lowest Zones (I and II). Although about 12 percent of the samples are within the two highest B Zones (Fig. 95), there were no extremely high plant or soil B values. No geographical areas of low or high B soils can be clearly defined although most of the high B values were measured in samples from sites east of Donau, either near the southern border or in the middle eastern part of the country on Phaeozem and Chernozemsoils. 134

22 Fig.96. Regression of Cu content of 20MR °Ill ritern° 111111111111111111 pot-grown wheat (y) on acid ammonium 11. 4,111111rip', 111111 acetate-EDTA extractable soil Cu (x), 18 ERIN Hungary. E 8:161111111111111:11nY,-.0- 1112111111111111 11111 o w1411111 11111 SIIUIIIISUIIIHII _c .6 12 11 li I NIIIIIIHISUPJIIIII

E,101111111111 1111111111 111111 111111 II .1111=1111111111 88111 111111 61111111111111 1111PM111111

4III IIIIIMENISINIMMI111111 211111.111111111E01111111 liblill 11111111111M111111PIMIN11111 oi .2 .4.6 .810 2 46 810 20 40 6080 Cu in soil, mg/l IA A Ac- EDTA)

Fig.97. Regression of Cu content of 22 pot-grownwheat(y)onorganic 20iii`Jin rill ,ter111111111011 carbon-corrected (AAAc-EDTA extr.) ±s6-, MIll ;_'''''7:. soil Cu (x), Hungary. 18 MINN MU 16Y173°6 1111 =_587111311 NO 14 NM 11111kMOM 12wino I MERIN

10111111111111=11111 mom

6 liora, in 4 N111li116.ardif 11155111101 2 PE MEE oMI 111111 rdilin311115111 .2 .4.6 .810 2 46 810 20 40 60100 Org.acorrected soi CU,mg/l ( A AAc -EDTA

Copper. Hungarian mean values for plants and soil Cu correspond closely to the respective international values (Figures 27, 29, 96 and 97). Since the variation of Cu contents is rather narrow and no extremely low or high Cu values were recorded, the Hungarian sample material falls almost completely in to the normal copper Zone (III). According to these analytical data, Cu problems seem unlikely. 135

Fig. 98. Regression of Fe contents of 300 pot-grown wheat (y) on acid ammo- 275 IMMINHOLIMMIIIIIIII III 250 11111111111MmEn11111 O nium acetate-EDTA extractablesoil FERWRIENallnillffill1111 II Fe (x), Hungary. 225 200 1 75ILIV,ESIMINEMIIiiiiii11111 E 1 50RIM111111111111111 1 111111 97 1 25MIIIIIIIIMI I

100MI111111111111111 I -b- 90mommg iOmni a NNW 1111111 P. -~c 80 IIIIIIg 7() MNIEMMERingrail o 60 MNIIIHNINEad2111 I u_ 50eingsiirdiiiiiiilli

40ElliiihN111EMI I 1111111 30IIADHNmum El monvonvigi 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc- EDTA)

Iron. The Hungarian material includes some plant samples with fairly low Fe contents (Fig. 98), but, in general, the Fe status of Hungarian soils and plants seems to be quite "normal". Fe data are distributed within the normal range (Zone III) with only six exceptions, four in Zone IV and two just slightly inside Zone V. In the "international Fe field" (Fig. 31), Hungary stands at the centre. Manganese. The correlation between the plant Mn and DTPA extractable soil Mn is high and because of the widely varying pH of Hungarian soils the correlation is further improved by correcting the soil Mn values for pH (Figs 99 and 100). The averages for plant and soil Mn contents deviate less from the respective international means than those of the most other countries (Figs 33 and 37). The Mn variation in Hungary, however, is very wide, and some very high as well as quite low plant and soil Mn values are included. Most of the high Mn values (falling in Zones IV and V) originated from sites in the northeastern part of the country or northeast of Budapest on soils classified as Cambisols, Arenosols or Luvisols. With several exceptions the low Mn values originate from southern areas east of Donau or from the middle eastern part of the country. These were often recorded on Phaeozem and Vertisol soils. Molybdenum. The initially low correlation between plant Mo and A0-0A extractable soil Mo (Fig. 101) improves and becomes highly significant when correction is made for pH (Fig. 102) owing to the wide variation in pH of Hungarian soils. Internationally, the Hungarian plant Mo and soil Mo contents are slightly on the low side but about 93 percent of Mo values are within the normal Mo range (Zone III, Fig. 102). See also Fig. 20. Indications of quite severe Mo deficiency are shown by samples from two sites (44133 and 44134) located near Budapest. Zinc. The Hungarian mean plant Zn content is the seventh lowest among 29 countries, and the mean soil Zn is the ninth lowest regardless of the extraction method used (Figures 136

2000 __woe Fig.99.Regression of Mn content 1511111:1111111111101 of pot-grown wheat(y) on DTPA 1000 MEMMIIMINNII1111M12M1:1.:N:WIIIMMINIMMEWc MIMMINNIIMININIEN extractable soil Mn (x), Hungary. 800 rarimilion11011mamilliusuirimmlIZZWILWITIriTVIMiTZWINWRIEMATIFITIIIII1MM 60011111MMIMIMIIIMMINEUISIMINI 400 ca- 300 200 iiiIIIIIIHhIHOiIii!!

o100 11E1111111111111111P41k Y 80mi====MMimmENIIMMIP:111MinainTIOSIMMO IIMMINIMIMMENIIIMPtiarKIISIMPIIIMI111114 o 60 4011111M1MORIIIMNIMFIVIS1111MM. 30 2011111211111111=11111111111111

10111MII/IMMEIMIN11 8siiimINMEElliaMOSTMIIMENINI WTI=1== 234 6810 20 30 40 60 80t0 200 300 Mn in soil. mg/I (DTPA)

2000 Fig.100. Regression of Mn content Mn of pot-grown wheat(y) on DTPA 1000 extractable soil Mn corrected for pH MN MUKTZ13Mi=111IMPIIIIIIMINIMMMIIIM111//1_10.011 Sat:S211.1.1=1111II:11411.6WMWMIN1111=1Mf =Mail 800 MEMEZEMMEII1 1212NMENNEMIMMIEMIMIIIIIIM (x), Hungary. 600 E IIIMMECEEMIEUTIMEM=1117111:altkVill ra- 40011111112113111111111GIBEIMILIWIMION If300IIIMMNIIIIIMMIM11111="41711111111111 200111111111111M11111111M111111111 o "E"100N11111111= a) 80laiMMEMEMMOMeat.PW=MMISIEM.Maiiii=M=11=rariMalrigVMMINIESIl=nteAlii o 60111111111=ldfir"-AIIIIMIEHME1.2111C=1M111111 o InimMiNammmunauniimirmilmuimmniii 40 30IIMIIIMPAN11111121110111111111MIESNIEll 20111.11110011.111111111111.111111 111111111111:11111111111111 10 MMMINON 8 =101111=MINNIM1131=MMENIU MO 1=MMINISI 11:11MMIIIIMIN =.11 2 3 46 810 20 30 40 60 100200 300600 pH-corrected Mn n soil, mg/l (DTPA)

41 and 43). The correlations between plant Zn and soil Zn, irrespective of the extraction method, are good and both methods give a similar general picture of the Zn status of Hungarian soils (Figures 103 and 104). It should be noted that when the DTPA extraction method was used more samples fall in the two highest Zn Zones (IV and V) than when the soils were extracted with AAAc- EDTA and the results corrected for pH, the latter method giving relatively more low Zn values (Zones I and II). The differences, however, are not marked. At some locations response to Zn fertilization can be expected. No extreme Zn values were recorded. 137

Fig. 101. Regression of Mo content of pot-grown wheat (y) on ammonium =0 353' = 62 032 ±0 6 oxalate-oxalic acid extractable soil Mo Mt 4 122 021 ±025 3 (x), Hungary. y.0 3 logxr204 r=017 2

.06

.03 .02 Ill III .01 .01 .02 .03 .06 .1 .2.3 .6 1 23 Mo in soil, mg/t (A0-0A)

Fig.102.Regression of Mo content MN 1IM =11MWM,M of pot-grown wheat (y) on pH-correct- 6 ed A0-0A extractablesoil Mo (x), MEMIIIIMIELIEM11111FEEECIMEIEEMIll MN Hungary. 31111111105 Y O06 2 ilblrEilleM1111=11nommmiilonini I4IINII11111

.111=1111D1111111111MI E:1211=75'...°1==rin:1111 mcmi MENIMINIMIMAIMIONIENreirsiliii=rAihrid&MMMEION 11MIIMMENIII ma.

MBINOMINI NEE =1111/(1 .03 .02Pill1011111111111111

.01=...... 20._=no..1111111.101111.11111111 .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MO in soil,mg/I (A0-0A)

5.3.4 Summary

The Hungarian soils included in this study vary widely in regard to texture, pH, and cation exchange capacity and somewhat less in regard to organic matter content. The P, Ca and N contents of soils and original plants are generally very high. Most of the K and Mg contents of soils are below the international average but those of plants are at an average level or above. The soil and plant contents of most macronutrients vary 138

20 H logc In et Fig. 103. Regression of Zn content of n 33 Zil '¡;:,.:.i.6..td2 7 s1 ;V..1177 pot-grown wheat (y) on DTPA extract- 7± =II . rc sl745 able soil Zn (x), Hungary.

y.1..5 I . Ig x loy1 2t + 3211og r...5:1` r 03,'.

I I .7". ) 7.' 1

) 1 \

) \ . \ '...'' I l_hiligli/If

1 ...... 111. ! IILMWM2151111111MENIMILIMM.

' 1 r ailfilltElIMIIMENIIIIk / m i I EC \ 3E 1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Zn in soil,mg/ (DTPA)

200 I te . Fig. 104. Regression of Zn content of Zn 3 3±i pot-grown wheat (y) on AAAc-EDTA 8 1 ' extractable soil Zn corrected for pH 5 .98 100 . 10 1 -31'1111111 (x), Hungary. 90 ..... o ..., 1111MMININI 80 11111I IIMMINNINII 70 1 IIIII MENIgn 60 1111 NIIII1 E 5011.1111E..,,...... o. 1 30 'id1 0111 o . 20 LI p 1 il E. o 11 ,..,:*?' r ..' .4% .: 10 1111111 9 ,...... 8 ...Hi 7 .I. .... ,... 6 I. M EP a IIi Hi-- "ppL V% .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Znin soi,mg/t (AAAc-EDTA)

widely, partly because of generally high though varying N, P, K fertilizer applicatfon. The micronutrient contents of Hungarian soils and plants are commonly at the normal international level; B is slightly on the high side, but Cu, Fe, Mn, Mo and Zn are on the low side. Compared to other micronutrients the variation ranges for B and Mn are wide: both low and high B and Mn values were recorded. 139 5.4 Italy

5.4.1 General The most important agricultural areas in Italy are relatively well represented by the soil and plant samples collected from thc country (Fig. 105). About half of the sampled soils

------i, r I T A L Y f ' - ,'..___.-'1 _...

1 k r,. "1 - \_5 4078,, .s. as. /RE,:m, ,4067A I -'i 140'79 1 ...... 40704. ' 4077-4065------' A 40406941 4("95941061 4081 Al.i...4066 if(\003 4068 9I 2 4054059 4055 4064 3943 40 002 4076 I 06204074 40 ),,,,,,A395O 4060 1:7" j-1-- -63942 :075140:21-'11:11%-4.1 ENEZIA 4083°5° 82145041 .-.../ TORINO '952 402nt A 4042 ! 1ZN "03953 9 8 .0 6 041,052 .s.--k--- 48 4°43.4080 At ",.... 3954 3848j 3949 4044 . 395501 3944 3959 3960 898 4°84049376i (394Olger 3884390907: A388.7.4 13905-06 4053 i.) 3941 1 BOLOGNblio3p97311,111902 , i GENOVA 3 09 9. \ 8,3894

3919 \ - 1 389 2 38'8--$.1..NO ".1 9,5 3916 ',.- 399, 399 3904 3912A1.3------18 NA \ ,gas:8408ED3914 308A5L54 39t3°r03882 .384906903911:1 3917 38900 392 30 ..+4--_,,___., 34 I AeRLIG L43983 0-3857 --:-."'-, 399303860 ' \ °"2 e 0 ',, r .1 3858 3859 -'1 660116 a

42. 0/ 1 - ONK --r--- 7- - 0

I 6141150948SO 2.75. \ 3870 3861 : 3871553864 ' 3868 ' 38 ''386 --- NAPO 931-3933 41,14021 POT6NZA 1,72 4022 b° 3863 3921394.0 4 I 4087 °3923 3922 3873 ,18! 3867

1 .20. 4.3/02 402; 14030 f402-'43 4027 4032:4!)2574026A4GOL2A. 0 4033 ,Iit CATeZARO

383 3880

LER0

3389h-63-68.3 03986 1 0 3qq, , 87 3984 ---03 I

3g9t441838989 5 14013 , I0...... i.-4400,14 3999 401154015. 3997.. ,.4017 39 6_ w 4016 3998 020 , 4019

, ,I.

Fig. 105. Sampling sites in Italy (points = wheat fields, triangles = maize fields). The last four numerals of each sample pair number (43854-44090) are given. 140

Italy Intermit !Oily Intermit Fig.106. Frequency distributions of n.189 3764 40 n= 189 3783 R.47 44 R.7.14 6.64 texture, pH, organic carbon content, 30 ns =16 16 nn .65 1.12 min.= 19 9 min . 4.65 3.62 and cation exchange capacity in soils of max.= 89 92 max = 7.82 8.56 3 25 30 Italy (columns). Curves show the inter-

25 nationalfrequencyofthesame characters. 20

0 _aAkNOW 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Cel C12)

Italy 47 Internat Italy Internal n= 189 3779 n= 189 3777 R=1.3 1.3 R.26.9 27.3 1.2 no.10.8 14.6 .1 min.. 9.4 max.= 5.7 39.1 max.. 74.9 99.9

1 2 16 32 60 2.8 25.6 2 8 8 16 32 ORGANICC,% CEC, me/ 00g

were classified according to the FAO/Unesco classification system. The frequency distribution of these was: Fluvisols (33 %), Luvisols (20 %), Cambisols (18 %), Regosols (13 %), Vertisols (13 %), Andosols (3 %) and Arenosols (1 %). The mean and frequency distribution of texture (TI) of the soils studied correspond closely to those of all the international samples (Fig. 106). The majority of soils have an alkaline pH(CaC12) and their organic matter contents and cation exchange capacity values are at or close to the international average. Electrical conductivity and CaCO3 equivalent values in Italian soils vary greatly. and are generally higher than in soils of most other countries (Appendixes 2-4).

5.4.2 Macronutrients

The average nitrogen contents of Italian wheat and maize soils and crops are clearly above the respective international means (Figs 6 and 107), the ranges of variation are wide. In part, the high contents and wide variation of plant N are obviously due to the generally high, but varying, rates of nitrogen fertilizer applications (Appendix 5). The phosphorus contents of the original Italian wheat and maize plants vary greatly but on average are above the international levels (Figs 7 and 108). Wide variations in the P contents of soils are also noticeable. For wheat soils the national mean is somewhat lower than for maize soils. Phosphorus fertilizer applications to the sampled crops (42 ± 29 and 51 ± 44 kg P/ha, respectively) are among the highest recorded in this study (Appendix 5). 141

Figs 107-111. Frequency distributions Italy Internet Italy Internat. ofnitrogen, n= 118 1765 n= 60 1958 phosphorus, potassium, 2=4.55 4.27 5=4.08 3.14 ±5= 1.03 1.15 os= .70 .87 calcium and magnesiumin original min =1.99 .60 30min.=2.11 .88 max =7.22 7.45 m00.=5.73 6.51 wheat and maize samples and respective 25 soils (columns) of Italy. Curves show CL >- 20 the international frequency of the same 1-1-1 15 characters. o

(1-: 10

1 2 3 4 5 6 N contenof Wheat5% N content of maize Ai

ItC11)/ Internat. ItalyInternat.

6 n= 118 1765 n= 60 1958 2..142 .133 0= .178 .135 os=.057 .084 os= .141 .088 min...065 .009 .008 50 min.= 096 m08.=.422 1.023 ern:m..1.169 1.657

eV, 0.40 o>- Z 30 LL1 O LLI 20 CC U-

.005 .01 .02 0 .08 .16 .32 .64 1.28 .005 .01 .02 .04 .08 .16 .32 .64 1.28 Fig. 107. Nitrogen, Italy. N in wheat soils, % N 'n maize soils,%

Fig. 108. Phosphorus, Italy. Italy internat. Italy 41 internat. n= 11E1 1765 n 61 n.1967 7, .42 .38 ..36 0. 33 is = .11 .12 ts ..08 ts. .10 min= .18 .05 30 min. 15 .05 max = .78 1.02 mox =.5a max . 1.04

5 6 7 .9 P contentof wheat,% P con ent of maize,%

40 RC/ iyInternat It° lyInternat. 35 n 118 1765 n= 61 1967 5. 19.6 20.2 5= 32.8 22.5 *5= 170 24.7 *5= 23.1 33.0 min=3.2 1.0 min. 2.8 0.1 max =116, 271.4 30 max =105.0 656.0

2.5 5 10 20 40 60 160 320 640 25 5 /0 CO 80 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I

There are wide variations in the potassium contents of Italian plants and soils (Fig. 109) and in the rates of application of potassium fertilizer to the sampled wheat and maize crops (29 ± 31 and 80 ± 98 kg K/ha, respectively). In general, the K contents of Italian plants are at the international average or slightly lower. The low K content of soils, 142

35 35 Italy Thternat Italy Internet Fig. 109. Potassium, Italy. n' 118 1765 n. 61 1967 30 = 3.5 4.0 30 = 3.1 3.1 = 0.7 10 es = 1.0 1.0 mu, 2.1 09 rnm.= 0.8 0.6 non' 5.3 68 25 max.= 5.2 67

5 6 2 3 5 6 K content of wheat% K content of maize,%

Italy Internal Italy Internat. n= 118 1765 n' 61 1967 30 6' 334 365 30 R. 187 330 es= 279 283 es e167 356 min..40 20 min.= 24 18 S' 25 max.= 1720 2097 25 5185 .1121 5598

25 50 1 0 2 0 400 8 0 16003200 6400 25 0 100 200 400 800 16003200 6400 K in wheat.soils, mg/I K in maize soils, mg/1

Italy Internot Italy Interne Fig. 110. Calcium, Italy. n= 118 1765 n= 61 1967 5 35 I. .50 .43 I= .69 .47 es= .21 .17 *s= .17 .20 23 30 rnm.= .18 .11 30 mm.. .33 .09 max=1.30 1.68 max.= 1.20 1.88 0,25 25 >- L) 20 20

D 15 15

Ca content of wheat % Ca conten *of maize %

Italy Interne ta lyInternot n. 118 1765 n= 61 1967 5 5= 6598 4671 35 ' I= 4847 3450 05= 4229 3076 01' 2698 2815 min.= 820 110 min.= 730 10 max.= 21930 21930 5101 .11130 17995 30 3 lE S" 25

a. 20

100 200 400 86616003 00 6400 12800 25600 50 100 200 400 800 1600 3200 64eu 12600 Ca in wheat soils mg/I Ca in maize soils, mg/I 143

40 40 Fig. 111. Magnesium, Italy. ItalyInternat. Italy interne. 35 n. 118 1764 35 n= 61 1967 9..140 .172 8,292 .251 05..081 .060 ee=.103 .119 23 30 .065 .044 30 min.= .145 .036 max.=.876 .948 max.=.711 1.125

CY./ 25

5 .< Mg content of wheot.% g content of maize,%

Italy Internet. Italyint,rflot. 61 5 n= 118 1764 n. 1967 8.341 489 R. 418 446 es= 306 437 es= 436 462 min.=45 10 min.. 59 1 30 m00e1889 3298 max =2100 6490

23 25

25 50 100 200 400 800 1600 3200 12.5 25 50 1 0 200 400 800 1600 9200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

especially maize soils, has been compensated for by heavy potassium fertilizer applications (Appendix 5). The average calcium contents of both the plants and the soils of Italy are among the highest national averages recorded in this study (Figs 9 and 110), while those of magnesium are at an average international level (Fig. 10). However, there is a difference between the two crops (Fig. 111). The Mg contents of maize and maize soils, representing the northern part of Italy (Fig. 105), are relatively much higher than the Mg contents of wheat and associated soils. About half the samples of wheat and wheat soils come from the southern part of the country including Sardinia and Sicilia, and usually have a lower Mg content than those from northern Italy.

5.4.3 Micronutrients

Boron. The B situation of Italian soils and plants is normal. In both the "international B fields" (Figs 22 and 25) Italy stands at the centre and has national mean B values near the respective international averages. However, owing to the relatively wide variation of Italian B values about eight percent of those ones show quite high B contents (Zones IV and V, Figs 112 and 113). These are not restricted to any specific geographical area but are scattered in various locations from Sicilia to Trieste. According to these data, deficiency of B is not as common in Italy as in many other countries but response to B can be expected at several locations (see also Section 2.7). Copper. Italian soils are exceptionally rich in Cu. With the Philippines and Brazil, Italy occupies one of the highest positions in the "international Cu fields" (Figs 27 and 29). The 144

100 /111111111111M 011111111111111111=1MINIMMIIIII1011111112 Fig. 112. Regression of B content of 80 IV12.11111111111i1Ex51t441111MINEIIMMIIIIMIIIMIIIIIIIIIIIIIIIMIS IEDWI=EMEMMORMINNEMIIII/111liaMMEMMIMMENIMEMMEMIIMINIMI.11111I pot-grown wheat (y) on hot water 60 "soluble soil B (x), Italy. For details of 50monimimmaTrummuummsnammuni - 0.82± o ZINIMES11111 =MEMO summarized international background 40Mg 80+2 54Ik 11111111211 30 data, see Chapter 4. a. 97 20111111111111111111 11111111

_c 10esiimk..=1=iliprairmi 'a8smisumnmmwsmaswimerniomm....solni=enrsarimminmosmmine "E. 6Emis.wsmigmossitmarmimmummu -9.) 5 o4irnith o m 3 OWER 1111 111111111 11111111i1 2 Iwo -Igr,

m .23 .4 .5 .6 .810 2 3 4 56 8 10 B in soil, mg/I (Ho t w. sol.)

100 Fig.113. Regression of B content of 80 pot-grown wheat (y) on CEC-corrected rt. 1 33 60 7±-s. ±201 (hot water soluble) soil B (x), Italy. 50 2±s=0 *04 Z ± 40 =325* 34 05 Ox .3o p 0 740 a. 15; 20 -c 1 0 8 4E a) 56 o 4 3 co 2

.08.1 .2.3 .4.5.6 .8 1.0 2 3 4 56 8 10 C EC -correctedB in soil, mg/t (Hot w. sol.) correlations between plant Cu and soil Cu are very good (Figs 114 and 115). About one- third of Italian Cu values fall in the two highest Cu Zones (IV and V), and of these two-thirds occur in Zone V. The majority of the highest (Zone V) Cu values were found in Bologna, Modena, Firenze, Arezzo, Perugia and Ascoli Piceno Provinces and bordering areas. Many of the high Cu soils were classified as Luvisols. Deficiency of Cu seems unlikely in Italy, since no low Cu values were recorded among the Italian sample material consisting of 170 sample pairs. Iron. Against the international background the Fe content of Italian plants and soils is normal (Figs 31 and 116). On average, the Fe contents of soils are slightly on the high side, but those of plants remain close to the international average. Although over ten percent of the Fe values fall within the two highest Fe zones none of these occupy any extreme 145

22 Fig. 114. Regression of Cu content of 9It± lyo Internot. pot-grown wheat (y) on acid 20 3538 ammonium acetate-EDTA extractable 04j2 700±257 R. 1 6 152 000 *7.05 soil Cu (x), Italy. 18 E iy= 9 4,1-3 g< Log y=3.620.909 R-16 80 0.664." 15; .cai14 o12a 88 Nib ()51111111MIEM1111111111

211111:1111=11111111INEi 11111111111MMEL111111111 .2 .4 .6.810 2 4 6810 20 40 6080 Cu in soil, mg/I (A AA c- EDT Al

Fig. 115. Regression of Cu content of 22 pot-grownwheat(y)onorganic `cii, Iintern., carbon-corrected (AAAc-EDTA extr.) 20 i iiuiiiiiui soil Cu (x), Italy. :°- 18 111111=1111111 IN

16 (ui,6 6 ° x I 1'117 eilligil MIR a_ o14 a) _c 12 o 10 1,\I IPIPPirLLAI.. a) 8 o o 6 limpolimill C.) 4 2 Iralm1111.!!1.9 o .2 .4 .6.810 2 46 810 20 4060100 Org.C-corrected soi CU,mg/t AAAc-EDTA) positions. The analyses of Italian samples do not indicate any severe Fe deficiency. Manganese. Because most Italian soils are alkaline, mean pH(CaC12) 7.1, the availability of Mn to plants is low. The Italian mean values for plant and soil Mn contents are far below the respective international averages and Italy occupies a low position in the international scale (Figs 33, 37, 117 and 118). Correction of the DTPA extractable soil Mn values for pH improves the plant Mn soil Mn correlation, depresses the soil Mn values and brings the Italian regression line close to the international line. About 22 percent of the Italian Mn values lie within the two lowest Mn Zones (I and II), more than half of them in Zone I, and some are very low (Fig. 118). The majority of low (Zone I) Mn values occur in the Po Valley, but soils of apparent Mn shortage are also found in the Salerno area, in Sicilia and in Sardinia. No high Mn values were found in the Italian samples. 146

3 00 Fig.116. Regression of Fe contents 2 75 1 of pot-grownwheat(y)onacid 2 50Feirligni"Miii BRIMIIIIMINVIIIIMIIIIII NI ammonium acetate-EDTA extractable 225 soil Fe (x), Italy. 200=fa21111111111MMI11111 I 1 75Illi°' MilIIMIE °AM I E 1 501iiiiiiunuiiiinhii ci .125ORRIMIIME MillI o I 100swimIIIIIIIIIIIIIIIIIIIEmu . Lumni o 90 1 -E. 80MEW IIIIIIMELEIIIIII a) 'E70 MiliniMiliteniaili o 60 Ingligi_pillealging 1 5011110-0810111111111 I

40Miiiaii..ii. IMO i

30 1MI 111111 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc EDTA)

Molybdenum. The high soil pH has the opposite effect on Mo to that on Mn and renders Mo readily available to plants. In the "international Mo fields" Italy stands fairly high (Figs 15 and 20). The plant and soil Mo contents in Italy vary widely and the sample material includes several high as well as a few relatively low Mo values (Figs 119 and 120). Although pH correction substantially improves the Italian plantsoil Mo correlation, the improvement is not as great as in many other countries and internationally where the variations in soil pH values are wider than in Italy. The highest (Zone V) Mo values are geographically widely distributed. Many of them are found in the same areas and sites as the low Mn values (see also MoMn correlation in Fig. 46). In all cases the soils are alkaline and their pH is above the Italian average. Mo deficiency is unlikely to occur in Italy except in a few locations where the soils are exceptionally acid. Zinc. The mean Zn contents of Italian wheats correspond closely to the mean for the whole international material, as do the Italian mean values for soil Zn irrespective of the extraction method used (Figs 41, 43, 121 and 122). Ranges of variation in both the plant Zn and the soil Zn are wide, though most of the Zn values are "normal". The frequency distribution of high Zn values in the Italian material corresponds to that of the whole international material: about 5 percent of Italian Zn values fall in each of the high Zn Zones, IV and V. Of the 17 high Zn sample pairs falling in Zones V and IV in Fig. 122 (pH-corr. AAAc-EDTA), 16 occur in the same zones when the DTPA method was used (Fig. 121). Many of the high Zn values originate in the same areas where high Cu values are typical (see Copper, above) but a number of samples from the upper Po Valley are also high in Zn. The low Zn soils were found mainly in the southern part of the country, especially in Sicilia where almost all soils and plants are low in Zn. Although the Zn content of most Italian soils seems to be medium to high, shortage of Zn may often be found locally. 147

Fig. 117. Regression of Mn content of 2000 Rat pot-grown wheat (y) on DTPA extractable soil Mn (x), Italy. 1 000 KVXMIGIEW.NIMXLIWESIMIUMI=UMs11,1111031111111 800 600

340000 Q.

1:1 20011111111111 11

15 10011\311111111111111111 80 60 c40IHNIsmomonsiimilemmismo 30IIMINPM115111111111111111 20

10 8 2 3 46 8 10 20 30 40 60 80100 200 300 Mn in soil, mg/l (DTPA)

Fig.118.Regression of Mn content 2000 of pot-grown wheat(y) on DTPA Mn extractable soil Mn corrected for pH 1000 800NM MitiMEILEEIA011ELIMIXMIE=MMED (x), Italy. MN 11=1=MMIIIMI NI WEVENKINKUIELT=EIMPISIEWORtlilf===EMMIIIIII 600 MINILTEMINIIIMIIMEIZINCIIIIMMIMIMEN1121 0_1§, 400 Ipp-hI -r:300 a.) _c200 o 100 411101LLLII* lhosiL,di 80 6011IIDSIMEEPP=MEMIN1 40IiIMEMBIESaitiMENEENIMIpporvi IMMUNE 30 20 JaiAll:. . 111 10 sommonnoin---min 8 i mninnonsnmai.=nnmoss =MEM.I1=WrIE 2 3 46 810 20 301.0 60 100200 300600 pH-corrected Mn in soit, mg/1 (DTPA)

5.4.4 Summary The Italian soils studied vary greatly in texture and they usually have an alkaline pH, medium content of organic matter and medium CEC. Compared with the whole international range Italian soils and plants are high in N, moderately high in P while K is at or slightly below the international average. For all the above macronutrients the ranges of variation are wide. Generally high, but greatly varying rates of N P K fertilizers were applied. On average, the Ca status of Italian plants and soils is high but that of Mg is only at a medium level. 148 ma.M1 .MOUIWINNE JIMMIE Fig.119.Regression of Mo content II0MOWN II lalini=MUM = 6MMELAMNIM IMEMMIIIIII of pot-grown wheat (y) on ammonium millekrennamo3 ZERE11111111111M oxalate-oxalic acid extractable soil Mo 3IMENIORMI°2INIIIIIIMI (x), Italy. 2milliiMitilinimM

.... 1111MI M=MIIIV MinnumizMommi+EIUM=AEZOMMErl MEMINIZIMPIHM=M211(11E=MMINEUI III 13 .3M IIMPRIMOMIIIIIii .2r---asmintNonni=1 o o -=RIBIIIMilli1111111M'EMIIIIE. rtA ammini=ME:::::.""Pi .06 linnallIIMMEMMOI1 m. .03 .02

BI .01

.01 .02 .03 .06 .1 .2 .3 .6 1 3 Mo in soil, mg/ (Ao-CA)

II mmamm::===-4---- =EliIM Fig. 120. Regression of Mo content of NI° suni. =mmuelirvirrii '..... mi. 6ournsucrocommilmancruammumniImmknoiiENIIIII am pot-grown wheat (y) on pH-corrected MENIIIIMEMMEMINIIIIMMIMIIIIIII MEi A0-0A extractable soil Mo (x), Italy. 3 inill NE° 3 AllnigillEginignill 2II var 1111111511411111111 111111B111111111011 Ill'Elbi III .EN§Iisi m.applr*--J=.--r. 3111M11EMILIIMIIIII=1111M111d4=M:11IUMNI1111ETIMENIMMEILIIMMIIIMM EMIEMMIIMIIINIMIIIII E EMOMEMPfiliklimmIIIMIESME4S11111 I 111111 MilielliIIIllM

.111111=WI a.1 -11-41114111PPillit1111111M11 IIMMIMMINaral=MMIMMME11===n21111 hmm mi._..mmmomoommm=mommom.. mom ritinimiumgmMOMMOMUMIMMMIMMOMOMMI MI INMBIIIIICIMIIMMEMIII MMENNIII MENIAMIMEMOINIMEHIE ME .03 .02 Ial .010111111111 .006 .01 .02 .03 06 . .2.3 .6 1 23 pH - corrected MO in soil, mg/I (Ao-CA)

Among the six micronutrients studied the B and Fe contents of soils and plants are usually within the "normal" international range, but especially in the case of B both low and relatively high values were recorded. Italian soils are exceptionally rich in Cu and Cu deficiency seems very unlikely. However, a great number of samples show alarmingly high contents of Cu. Owing to the high pH of most soils, high Mo and low Mn values are typical in Italy. In many areas shortage of Mn is apparent. Zn contents vary greatly, medium and high values being most common but response to Zn may be obtained locally. 149

Fig.121.Regression of Zn content 200 of pot-grown wheat(y)on DTPA extractable soil Zn (x), Italy.

100 90 80 70 E60 c` 50 o 40

t- 20

10 9 8 7 6

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zr1 in soil, mg/1 (DTPA)

Fig.122.Regression of Zn content 20 1 . of pot-grown wheat (y) on AAAc- Znlp 3 y ;1 .9 1 EDTA extractable soil Zn corrected _44 5±5 for pH (x), Italy. =1O5r 1 07 log x ; 10) II III r=u. Vardlil 9I = l.1.... 8) =I I 111111MMIIIIII 7 \ 'MI 111111111 6 in linilli E55 Ill MOM

...(?). 4 Il MOM o 3 1 .. 11"11111111 o t 2LIU ...*. in 1111111 "E'o

1 1110161 Ill ,01111 ...... , ...... MIL MMINIIIILIMINIIIIIII...... Minal 6 '""ME aElliiiII 1 111171111w NM .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soil, mg/I (AAAc-EDTA) 150 5.5 Malta

5.5.1 General

The Maltese material consists of 25 paired wheat and soil samples originating from five different areas: Rabat, Mosta, Zurrieq, Qormi and Zejtun. The wheat samples were all of one variety ('Capelli'). The soils included in this study are medium textured with high pH, having a medium organic matter content and a cation exchange capacity usually somewhat lower than the international average (Fig. 123). The national mean electrical conductivity is among the highest and that of CaCO3 equivalent is the highest in this study (Appendixes 2 and 4). The ranges of variation of all the above soil properties are narrow.

Malta Internat, M it0 Internet Fig.123. Frequency distributions of n x 25 3783 n x 25 3764 40 7= 43 44 9= 7.54 6.64 texture, pH, organic carbon content, nsx 9 16 nsx..05 1.12 min.= 32 35 min.= 7.48 3.62 max x 61 92 max = 7.64 8.56 and cation exchange capacity in soils 30 of Malta (columns). Curves show the

25 international frequency of the same

20 characters.

10 30 0 54 5u 70 00 90 TEXTURE INDEX pH (Co C12)

45 Malta 5,Internet Malta Interno) 6 40 n =25 3779 n= 25 3777 = 1.3 1.3 =22.2 27.3 x .3 1.2 ns = 4.4 14.6 "E35 .1 min=17.9 .2 max.= 2.1 39.1 max.= 29.8 99.9 0)30 25 o Ly, 20 uj 15 CC

.2 . 16 3.2 6,4 12.8 25.6 32 ORGANIC C,% CEC, me/100g

5.5.2 Macronutrients

The total nitrogen contents of Maltese soils are at an average international level but those of the original wheat are low (Fig. 124, Appendix 2, see also Fig. 6). All the wheat had been fertilized with 60 kg N/ha which is slightly less than the average amount applied to the wheat sampled elsewhere in this study (Appendix 5). The average phosphorus content of the original Maltese wheat is the highest among the national mean values and the P content of soils the second highest after Belgium (Fig. 125 and Appendix 2). All soils were fertilized with 26 kg P/ha. See also Fig. 7. Potassium contents of both wheat and soils are at an average international level (Fig. 151

40 Malta Internat. Malta Internat. n 25 n= 25 1765 35 n=1765 Internat. R=3.07 4.27 Malta = 4.2 4.0 o= 25 ts= .75 1.15 n.1765 to= 0.8 1.0 8= .53 8= .38 mm.. 2.9 0.9 0 min.= 1.63 .60 -QE') =5..15 ts =.12 max.= 5.9 6.8 o max.= 4.70 7.45 at 30 min.= .42 min.= .05 max.= .00 max= 1.02 o_ V 0-25 3- 3- o C±) 20 15 LLI 2, 15 (2/ 1.11

5 5

3 5 2 .8 .9 2 3 4 5 6 7 N content of wheat,°/0 P content of wheat% K content of wheat%

70 Malta Internat. Malta Internat. n= 25 1765 Malta Internat. n= 25 1765 5 8. 80.4 20.2 60 R. .145 .133 es= 51.3 24.7 n= 25 1765 min.= 20.7 1.0 = 365 365 ts= .034 .084 max=171.8 271.4 ±6. 226 283 min.= .115 .009 min.= 186 20 max.= .212 1.023 o max.= 789 2097 25 o- 7'- 20 O

oD 15 15

10 10

P4 A .005 01 .02 .04 .08 .16 .32 64 1.28 iii.2.5 5 10 20 00 80 160 320 640 25 50 100 200 400 BOO 1600 3200 6400 N in wheat soils, % P in wheat soils, mg/I i< in wheat soils, mg/I Fig. 124. Nitrogen, Malta. Fig. 125. Phosphorus, Malta. Fig. 126. Potassium, Malta.

40 Malta Internat. MCIII0 Internat. Figs 124-128. Frequency distribu- 35 n= 25 1765 35 n= 25 1764 8..49 .43 7..102 .172 tionsofnitrogen,phosphorus, =s= .13 .17 is= .013 .060 ú 30 min.= .31 .11 2330 min.= .075 .044 max...84 1.68 mox.= .126 .948 potassium, calcium and magnesium o_25 in original wheat samples and re- >- 0 20 spective soils (columns) of Malta. Curves show the international fre- D 15 o quency of the same characters.

5 .2 Ca content of wheat,% Mg content of wheat %

40 Malta Internat. MaltaInternat. n = 25 1765 35 8=6518 4671 5 n= 25 1764 is =744 3076 .271 489 min.= 5640 110 Ls= 47 437 mox.= 8090 21930 min.= 207 10 30max.= 335 3298

Ú 25 ú) 25

c" 20 3- O o W 15 o

100 200 400 BOO 1600 3200 6400 12800 25600 5 50 100 200 400 800 1600 3200 Ca in wheat soils, mg/I Mg in wheat soils, mg/I Fig. 127. Calcium, Malta. Fig. 128. Magnesium, Malta. 152 126). High rates of potassium fertilizer (83 kg K/ha) were applied to all soils. See also Figs 8 and 11 and related text. Calcium contents of Maltese wheat and soils are among the highest while magnesium contents, especially those of wheat are among the lowest (Figs 127 and 128, Appendix 2). See also Figs 9 and 10.

5.5.3 Micronutrients

Boron. The national mean values of B are among the highest (Figs 22 and 25) and a considerable percentage of these samples fall in the two highest B Zones (IV and V, Figs 129 and 130). The variation of B, however, is narrow and no low or extremely high B

100 __--.s Fig.129. Regression of B content of aL smismmonassi 80 pot-grown wheat (y) on hot water 60 gmh.woommuli MEMEMEn 50 33-=638NIEMOOLEMIll soluble soil B .(x), Malta. For details 40 14±aMEEinin of summarizedinternationalback- 30 IME11111 =Min ground data, see Chapter 4. 20 111111 O .0 10 .6.,,umil__1111111111111 11-amim=iumaiiME===111=1.1O=IMMININIO 15 8 IINI IMMMEMPIIPMMIPEMIGi1111111 E' 6 mh.=mimunimiammlimmEsul 5 1111 =MIN.P.ENIIMMII=MMENIII 4 NIMMINSMIiiiEMEMIni o 3 11111111MMIIIII MEMO.' E1111111 2 1111 'EMMA

.2.3 .4.5.6 .810 2 3 456 8 10 B in soil, mg/I (Hot w. sol.)

100 80 n= 35 60 ±s=6 8 145 6 I 50 7-ts=121 01 0, 0 40 E log y=06 01 0.30 r-0784 02 1E; 20 a) _c 10 8 a)56 4 o 3 2

Fig. 130. Regression of B content of .08:1 .2.3 .4.5.6 .8 1.0 23 4 5 6 8 10 pot-grown wheat (y) on CEC-corrected CEC-corrected Bin soil,mg/I (Hot w. sol) (hot water soluble) soil B (x), Malta. 153 values were measured from Maltese soils or pot-grown wheat. Copper. All Cu contents determined from Maltese plants and most soil Cu values are above the international average (Figs 131 and 132), although no sample had an extremely high Cu value. Iron. Among the 29 countries studied, Malta has the lowest national mean values for plant and soil Fe (Fig. 31). More than half of the Maltese samples are within the lowest Fe Zone (I) and only one-third lie within the "normal" range (Fig. 133). On the basis of this finding, although from a limited number of samples, it seems that the application of iron in Malta would be beneficial for crops sensitive to iron deficiency.

Fig.131. Regression of Cu content of 22 pot-grown wheat (y) on acid 20 ammonium acetate-EDTA extractable soil Cu (x), Malta.

.612

88 a66

.4.6 .810 2 46 810 20 40 6080 Cu in soil, mg/l (AAAc- EDTA)

22 Cu Nut i nb,nterno 20n=2b n.3 5 7±s.7 0± 57 18 A 5 ;±1;;4 Ifilligilli mommEsdal 11111111 IIIIIIIIMEMIENMPUI IIIMINENIUMENIII Fig. 132. Regression of Cu content of 2 11111111111111111111111171111111111111111111! pot-grown wheat (y) on organic carbon- 1111111111111MMIE corrected(AAAc-EDTA extr.)soil .2 .4 .6 .8 1.0 2 4 6 810 20 40 60 100 Cu (x), Malta. Org.C-corrected soi CU, mg/1 (AAAc-ED7A) 154

3 00 Fig.133. Regression of Fe contents 2 75 2 50Fe "1:11119111111H9 of pot-grown wheat (y) on AAAc- 225 Eiwommastitimiumi I ammonium acetate-EDTA extractable 200mimagoloostamiono o soil Fe (x), Malta. 175LINY iv, 1111111117Mliiiiii I E 1 50 cL .125 o 11111111 I 1 00MOIR o 90 mow_ L Elnii 80 mmonin IliniI 70 M1111111 spipism o 1 o60 wonmoon u_ IIIPIN IIIIIII 50 I 40 iIIflUUIillhIUI i 30 tent1ink.. msuoilimaniiil 10 20 40 60 100200400 600 1 Fe in soil, mg/l (AAAc-EDTA)

Manganese. Eighty percent of the Maltese soil and plant samples is so low in Mn that the samples fall in the lowest Mn Zone (Figs 33, 37, 134 and 135). The national mean values for soil and plant Mn are far below those of any other country included in this study. It seems that a response to manganese fertilization would frequently be obtained in Malta, providing that the Mn status of other Maltese soils is not much better than that of those studied here. Molybdenum. In Malta where all the soils included in this study were alkaline, pH(CaC12) 7.48-7.64, the availability of Mo to plants was high (Fig. 16) although the extractability of Mo by the A0-0A extractant was low. The plant Mo values given in Fig. 136 are therefore far above and soil Mo values far below the irrespective international means. Correcting soil Mo values for pH raises these values to the international average (Fig. 137). See also Figures 15 and 20. In spite of the high average Mo content of plants a great majority of the Maltese Mo values are within the "normal" range, and since the variations of both plant and soil Mo are narrow no extreme single Mo values were recorded in the Maltese material. Zinc. Contrary to the results obtained for Fe and Mn, the Zn contents of Maltese plants and soils were very high. Such high national mean values for Zn were otherwise only recorded for Belgium (Figs 41 and 43). Irrespective of the extraction method used, only five out of the 25 Maltese Zn sample pairs are within the normal Zn range (Figs 138 and 139), 15 lie within Zone IV or on the low side of Zone V and five sample pairs lie extremely high in Zone V. 155

Fig. 134. Regression of Mn content of 2000 rjrliriiiin. ta pot-grown wheat (y) on DTPA extract- L'3531 21 :I 1 6 1. able soil Mn (x), Malta. 1000NE MmliAlE5rionEMB104VMmealilIIIIIIIIIII malislf111 800 11 NERTIINIEENFINNEINEMNWIWElaretiNT:V1IMI 600IIIIMPIR:TAMIMMMEMEIKIMIMENIMMI.1... E 400 8: 300 h`iiiiiiiimilimiim V211111111111MIIIIIIIIIIMINI

"610001111111111.11111511I .d. gn11111IMEIM=MMIIMMUMIKIIIMI117.7.== %,..,ME=11MIRMIMMIIIIIMIMINI=MMEMENE110. -E. 6011111MMEMENIIPMEMMNIUM=M oliNorammoonmew=immimossimmim x!4° dillikIIIIIPPIPM1111111111111111 A. 32°0 imaltuintmlimummil liE111111110.1111111111 108...... N.....,2211111.-mayoun..., IIIM1MMINEMEIMMII011111MM111ENZi= 1 2 3 46 8 10 20 30 40 60 80100200 300 Mn in soil. mg/I (DTPA)

Fig. 135. Regression of Mn content of 2000 Immiamos. =mimom. 11=mals. pot-grown wheat (y) on DTPA extract- Mn able soil Mn corrected for pH (x), 1000MN IHNILMiNPIà1;INEM 11111111111111TIMamwsIiMNIMIlmmmil MI IMAIFIN.N/FAIBM W1.1:11MEN11111MMIMIMMENII 800ME =MINMENE 1=1=11=1=MMINII Malta. n WEIMIMINEWII manIrmwriligemmemimimMiI 600IINIMMEMITIONII 400 lIIh!UPiIII 300 h. in-Agi _c 200 101 ' o .E" 100 11=1=MMEMmalmil 80 ...111IMMMEN=1W1111=111IMMEMNIM0=1-MEME 60InmasEmmumir- MEMEMOIIMUIMIZINIM 0 iiiMLWElpiar IMMEIBUMMIMENI 40 .11 30 20 ka

10 11111111111111minmmo ON1MINIIMINZIMi11121=IMMENNE1111111111LMIUmmEms 1=1=MMEN NALMIMINI MIMMUJI 1 2 3 46 810 20 30 40 60 100200 300600 pH- corrected Mn in soil, mg/1 lDTPA)

5.5.4 Summary The soils of Malta included in this study usually have medium texture, organic matter content and CEC. Their pH, electrical conductivity and CaCO3 equivalent values are high. Many extremes of nutrient content were found. The P and Ca contents are at the highest international level and Mg contents are at a low level. The mean Fe and Mn contents of Maltese soils and plants are the lowest and those of Zn the highest recorded in this study. 156

MMINCIa manalEanaril Fig. 136. Regression of Mo content of M o g101110811 Emu. mmumagis=MIMI 61EMMMES1111111111tarMMIHII pot-grown wheat (y) on ammonium MEMPAITISIAMMICENCE0111111 oxalate-oxalic acid extractable soil Mo 057 ga 11111111 3MEMNON (x), Malta. 2MIESEfiliC Y-21111iiii E o_ CL 1 UUIIII1Ii!IIIIIII o ommommemamemunMINIMMETAMISIMIMIMMOURI 2 .6=1MINNIIIMMI=W151111 o

d) .2 0. r/r1111171.11111111111111hL o .1 2 WM1111.1:1======.... .06MENIMM112111111111=MMI11111,Wil=manummany

.03 .02

01111111111Ewen= ilesessipmNrmmwmommunr .01 .02 .03 .06 .1 .2 .3 .6 1 23 Mo in soil, mg/1 (A0-0A)

MO ,.011MMO11 IMM.MMIN==...a, WOMIIIIMCM=IMMIN11MiDi101MMI`g MirMMIMMIUM nzarammall 6

3 2IMMINIMPFFIREIRINIE=41lillsOY9:2211MANAMPI

E 0. 1 o EMINIMMEMP .6OrlIENWOMMIIIMANIMIONM-11

unansw.---mr,rmann====911111.1.11r11111ilkitni,r4k .06

.03 .02 Fig. 137. Regression of Mo content of .0111111111111.1'11 pot-grown wheat (y) on pH-corrected .006 .01 .02 .03 .06 .1 .2.3 .6 1 23 A0-0Aextractablesoil Mo(x), pH - corrected MO in soil, mg/l (,40-0A ) Malta. 157

138. Fig. Regression of Zn content 200 M ti Ine t 33' of pot-grown wheat (y) on DTPA 1;.3 77 extractable soil Zn (x), Malta. o 19 5 321 log

100 90 80 70 g_ 60 111-0 50 o 40

"a 30

-g; 20

C.) N

10 9 8 7 6 3L

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100. Zn in soil, mg /I (DTPA)

200 ID II Irte Zn 353 .... 3:5 3 133 6 on 26 5 2.1 38 .

y Oe 9510g - , ) s $ 100 90 80 IIIIIMMM 70 MillII ild 60 50 IIDA 40lopi'llilipial 1 30 dooll 1o Idol 20 I i h, li 10 Loom 9 EmilIli, MM. 1kMIEN Fig.139.Regression of Zn content 8 7 MINI M Ill LIMIIIII of pot-grown wheat (y) on AAAc- 6 EDTA extractable soil Zn corrected IIIINI 11111IRIIN for pH (x), Malta. .2.3 .4.6 .8 1 23 4.6 810 20 3040 60 100 p11-corrected Zn in soi , mg/I (AAAc-EDTA) 158 5.6 New Zealand

5.6.1 General

In addition to samples collected from the North and South Islands of New Zealand (Fig. 140), eight maize-soil sample pairs (45401-08, not shown in the map) came from Tonga- tabu, Tonga Islands, and two (45432, 45439) from Rarotonga, Cook Islands, and are included in the sample material received from New Zealand. Out of 29 soils classified, Fluvisols were the most common occurring at 8 sampling sites. Other soils were classified

Fig. 140. Sampling sites in New Zealand (points = wheat fields, triangles = maize fields). 159 as Gleysols (5), Cambisols (4), Andosols (4), Regosols (3), Histosols (2), Luvisols (2) and Phaeozem (1). The textural distribution of the sampled soils is shown in Fig. 141. The soils from North and South Island are usually medium to heavy textured whilst all Tonga soils are heavy, the texture index ranging from 70 to 82. The soils received from New Zealand are all somewhat acid. The national average pH is one of the lowest and the material includes a Histosol sample (45428) with pH(CaC12) 3.62 which is the lowest pH of any examined in this study. The latter soil failed to grow a plant sample in the pot experiment. The organic matter contents of all New Zealand soils are above the international mean (Fig. 141). Due in part to the two peat soils, the New Zealand mean (5.1 %) is higher than any other national mean content of organic carbon, and because of the generally heavy texture and high organic matter content of New Zealand soils their average cation exchange capacity is higher than that of any other country. The heavy textured Tonga soils especially show a high CEC (62-74 me/100 g). The electrical conductivity values are at an average international level or below and, with an exception of one sample from Rarotonga, the CaCO3 equivalent values were at zero level.

5.6.2 Macronutrients

The national average for nitrogen contents in New Zealand soils and original maize were the highest recorded in this study, and N contents of wheat among the highest (Appendixes 2 and 3). Furthermore, the highest single value for soil total N of this material was measured from a New Zealand peat soil. The sampled wheat crops were fertilized with relatively small amounts of nitrogen fertilizer (9 16 kg N/ha). The higher and more varying rates applied to maize (55 ± 67 kg N/ha) are an apparent reason for the wider variation of maize N than of wheat N contents (Fig. 142). For international comparison see also Fig. 6. Phosphorus contents of the original New Zealand wheat samples and associated soils (Fig. 143) vary little around the international means. The P contents of maize correspond to the international average but those of maize soils are higher. Both the P contents of maize and of maize soils vary more than those of wheat and wheat soils. The lowest P values in maize and maize soils were obtained from Tonga samples. Phosphorus fertilizer applications to sampled wheat crops were at rates much lower (16 ± 7 kg P/ha) than those applied to maize (30 ± 29 kg P/ha). In spite of this the uptake of P by maize in relation to P in the soil was-rather low. A partial reason for this may be the high CEC of maize soils (average CEC of maize soils is 60 and that of wheat soils 35 me/100 g, Appendixes 2 and 3) making P less available to plants. See Fig. 14. The P situation of New Zealand in an international context is seen in Fig. 7. Potassium content of wheat and the exchangeable K in wheat soils are clearly on the low side, but K contents of maize and maize soils are substantially on the high side compared with respective distributions for the whole international spectrum (Fig. 144, Appendixes 2 and 3). There was also a considerable difference in rates of applied potassium fertilizer to wheat and to maize crops (7 ± 12 and 44 ± 25 kg K/ha, respectively). All the highest soil K values (>800 mg/1) were obtained from Tonga samples. As a whole (in terms of general mean plant Ksoil K) New Zealand stands close to the centre of the "international K field" (Fig. 8). 160

Fig. 5 New Zealand Internat New Zealand interimt 141. Frequency distributions of n n 36 3764 0 n. 38 3783 9=54 44 9=5.39 6.64 texture, pH, organic carbon content, ts. 15 16 05=.58 1.12 min = 32 9 35 min = 3.62 3.62 and cation exchange capacity in soils max . 82 92 max it 6.77 8.56 30 of New Zealand (columns)Curves show the international frequency of the 25 same characters. 20

10 i

0 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca Cl?)

New ZealandInternat. 5 New Zealand internot n =38 3779 n=38 3777 0. 5.1 1.3 9=50.9 27.3 os. 7.3 1.2 20.2 14.6 35 min.= 2.0 min.. 20.5 .2 max =39.1 39 max.= 99.9 99.9 30 Go o25 in20

ui 15

.2 C .8 16 32 64 126 25,6 16 2 ORGANICC,% GEC, me/100g

Figs 142-146. Frequency distributions New Zealand Internat New Zealand internot 14 1765 n. 24 1958 35 n. of nitrogen, phosphorus, potassium, 9=4.68 4.27 0=3.68 3.14 05=.59 1.15 xs..89 .87 calcium and magnesium inoriginal I30 min.= 3.78 .60 30 min. 1.71 .88 max.= 5.51 7.45 ma.= 5.40 6.51 wheat and maize samples and re- 25 25 spective soils (columns) of New Zea- >3- 20 20 land. Curves show the international 5 frequency of the same characters.

2 3 4 5 6 2 3 N content of wheat, % N content of maize,./0

70 70 New Zealand Internat. New Zealand Internat. 24 1958 60 n= 14 1765 60 n= 9.340 .133 9= .481 .135 xs=.139 .084 tu- .340 .088 min.=.211 min .221 .009 50 .008 Go max =301 1.023 max.= 1.657 1.657

.005 01 .02 .04 .08 .6 32 64 128 .005 .01 .02 04 .08 o .32 .66 128 N in wheat soils, % N in maize soils,% Fig. 142. Nitrogen, New Zealand.

Calcium contents of both the original plant species are relatively low as are the exchangeable Ca contents of wheat soils (Fig. 145). The relatively high Ca average for maize soils (3215 mg Ca/1) is mainly due to the soils of Tonga and Rarotonga (3370-7120 mg/1). Since the latter soils are heavy textured and have a high CEC, their high Ca contents are not reflected as high Ca contents of maize. See Figs 12 and 14 and related text (Section 161

Fig. 143. Phosphorus, New Zealand. New Zealand50 Internat. New ZealandInternet 35 n. 14 n.1765 n= 24 1967 =.39 X= .38 R=.35 .33 *0..05 ±s. .12 *6..09 .10 min.=.32 min.=.05 min. .23 .05 max...46 max.= 1.02 mas= .51 1.04

2 3 4 s S P content of wheat, % P content of maize.%

40 4 rNew Zealand 5,0,1 Internet New Zealandint.rn.t n= 14 4.1365 n. 24 1967 35 R = 243 = 20.2 5 4= 40.0 22.5 ±S=7.8 ±s= 24.7 = 33.1 33.0 min. 15.2 min= 1.0 min.=4.8 0.1 max.= 41.6 max.=271.4 rnax.=.157.4 656.0

2.5 5 10 20 40 80 160320640 2.5 9 iO 20 40 80 160320 6)0 P in wheat soils, mg/I P in maize soils, mg/I

35 Fig. 144. Potassium, New Zealand. New Zealand /nternat New Zealand Internet n= 14 1765 n= 24 1967 9=3.5 4.0 30 7= 3.5 3.1 ±s=0.8 1.0 ,s= 0.8 1.0 min.= 2.5 0.9 min.= 2.1 0.6 max.=5.1 6.8 max.= 4.9 6.7

3 5 6 7 K content of wheat% K content of maize,%

Internet 3 New Zealand intemot New Zealand n= 14 1765 n= 24 1967 4.210 365 8= 500 330 es. 165 283 ±s= 317 356 min.= 68 20 min. 76 18 max.= 699 2097 max =1063 5598

o>- Ui o u-

25 50 1 0 200 4 4 6301600 3202 6400 50 100 200 400 000 1600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

2.2.4).In general, New Zealand stands on the low side in the international comparison (Fig. 9). As in the case of calcium the magnesium contents of original wheat and maize are low (Fig. 146). The same is true for exchangeable Mg contents of wheat soils and of maize soils obtained from the two main islands of New Zealand. 162

40 New Zealand Internat. New Zealand Internet Fig. 145. Calcium, New Zealand. n = 14 n.1765 35 n. 24 n =1967 8.30 5. .43 R =.34 _ R .47 is ..13 is ..17 50.06 is =.20 o min.=.26 min, .11 30 min.=.24 min.=.09 max...67 51.5.1.68 max.= .47 max.. 1.88

N.25 25

U 20 20

/.6 Ca content of wheat, % Co content of maize,%

40 New Zealand Internet. New Zealand Internet n= 14 n. 1765 n. 24 1967 = 1916 8= 4671 35 R =3215 3450 88. 753 85= 3076 85 =2025 2815 min.= 1170 min.. 110 min .1070 10 max.= 3610 mox..21930 , max =7120 17995

00 200 400 800 1600 3200 6400 12800 25600 50 100 200 400 BOO 1600 3200 640012800 Ca in wheat soils mg/I Co in maize soils, mg/I

40 4 [5,,3,1 New ZealandInternat. New Zealand Internat. Fig. 146. Magnesium, New Zealand. 35 n =14 1764 35 n. 24 1967 5..153 .172 R. .179 .251 08. 038 .060 55= .053 A19 min.5.109 .044 30 min.= .106 .036 o max...249 .948 max.= .291 1.125

5 5 Mg content of wheat,% Mg content of maize,% 0 40 New Zealand Internet New Zealand internot n. 14 1764 35 35 n= 24 1967 5.169 485 = 518 446 ts = 176 437 is. 395 462 min.= 48 10 min.= 41 30 max = 638 3298 max.= 1264 6490

1325

20 o3-

ILI15 o Ui ix 10 LL.

25 50 100 200 400 800 1600 3200 2.5 25 SO 100 200 403 888 1600 3200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

The high mean Mg value of maize soils (518 mg/1) is due to the high Mg contents of the ten Tonga and Rarotonga samples which all exceed 650 mg/1 and are among the 13 samples making up the three right-hand side columns in Fig. 146. As in the case of Ca, the uptake of Mg from these soils (heavy texture + high CEC) is relatively low, so explaining the low Mg contents of maize. See Figs 12 and 14 and related text (Section 2.2.4). Interna- tionally New Zealand's Mg status is similar to that of Ca (Fig. 10). 163

100 '. 117111M111=1M111O1111 1111111111111 Fig.147. Regression of B content of VIWZWAL:Is 11012111111 rai4Eril Mes pot-grown wheat (y) on hot water 80 MI1111111111111111111111111MB NMI 60 Nu=mriumn soluble soil B (x), New Zealand. For 50 MIZEMIEMEIIIITIESMIIMEMMINuni detailsof summarized international 40 -0'76±iMEIESEll 111111 background data, see Chapter 4. 30 ri 840 AILIMMENO CM

ID: 20 I111111 ti .c eismmmun '5810 b.1111.10N21=1MMIIMIIIII=MM1111111Misemsv=musib 6 EMkIMPIIMMIP.21NNIMill o 45 HOMIPMEMMilii OMMUI oo 3 II uIIuII NE1111 2 1111

1 .2 .3 .4 .5.6 .8 10 2 3 4 56 8 10 B in soil, mg/l (Hot w. sol.)

100 I Fig.148. Regression of B content of 801 In pot-grown wheat (y) on CEC-corrected n.3 33 (hot water soluble) soil B (x), New 60 V±s.5 126x 6 9 50 as.06 031 0 3 2 Zealand. 40 E .509+ 32x -2 + Ox o-30 r-007935 O 20

1 0 8 6 5 4 3 CD 2 la .081 .2.3 .4 .5.6 .8 1.0 23 4 5 6 8 10 CEC -corrected B in soil, mg/1 (Hot w. sol.)

5.6.3 Micronutrients

Boron. The B contents of soils and plants are at an averageinternational level, the variations are narrow and almost all B values (Figs 147 and 148) are within the normal range (Zone III). On the basis of these limited analytical data, response to B maybe obtained only occasionally with crops of high B requirement. 164

22 Fig. 149.Regression of Cu content of 20 pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), 18 E New Zealand. Q.

ti.14

o12 77;10 88 C.) 6

.4 .6 .810 2 4 6 810 20 40 6080 Cu in soil, mg/l (AAAc- EDTA)

Fig. 150.Regression of Cu content of 22Cu pot-grownwheat(y)onorganic NeweaL d Internt. 20 carbon-corrected (AAAc-EDTA extr.) 7±s=6 .. soil Cu (x), New Zealand. 18 :innim1I1 y.,4 23 min E 16r =081tamirT0Y.;1111111111111MIWAI Q. Q.

8 1111 o Ill" o 6 IIIIIMIIIIPMEINIIII 4 1111111MIIIMMIIII

2 I. 111811111MIllerill:

.2 .4 .6 .810 2 46 810 20 40 60 100 Org.C-corrected soi CU,mg/l AAAc-EDTA)

Copper. In spite of the limited number of samples from New Zealand, the variations of both the plant and soil Cu values are very wide (Figs 149 and 150). Only half of the samples is within the normal Cu range, one-fourth is in the two lowest Zones (I and II), indicating possible Cu shortage, and one-fourth is in the highest Zone pointing to a surplus. All low Cu values come from the South and North Islands and the high values from Tonga. 165

Fig. 151. Regression of Fe contents of 300 275 pot-grown wheat (y) on acid 250 ammonium acetate-EDTA extractable 225M164111111Vmwttirrunlimmum soil Fe (x), New Zealand. 200 1 75illi°1111111111111111 E 1 50 O- 11111111111111111 90 80 111111110111111111

-f4). 7011.11IIIMILI11111 o 60 501114111COMINIIIII 40sainfilieloom

301111111111111111111111 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc-EDTA)

Iron. The mean Fe contents of plants is only slightly above the international mean, but the mean for soil Fe is the second highest (after Finland) in this study (Fig. 31). More than half of the Fe determinations are within the two highest Fe Zones (Fig. 151). The highest extractable soil Fe value was obtained from a peat (Histosol) soil, the three highest plant Fe contents (> 100 ppm) from plants grown on Fluvisol soils and the three lowest (<34 ppm) from plants grown on Andosols (see also Fig. Fe in Appendix 7). Manganese. Because of the generally low soil pH no very low plant or soil Mn values were recorded in the New Zealand material (Figs 152 and 153). The mean values for both the plant and soil Mn are above their respective international means, nevertheless the bulk of the Mn values are within the normal Mn range (Zone III). Although some of the Mn values are within Zones IV and V, there are none in the New Zealand material with alarmingly high Mn contents. Molybdenum. Owing to the low availability of Mo to plants in acid soils the Mo contents of wheat grown in pots on New Zealand soils are low (Figs 154 and 155). The national mean is the second lowest (after Brazil) among the 29 countries (Figs 15 and 20). The New Zealand mean for A0-0A extractable soi/ Mo is only slightly below the international mean (Fig. 154) but decreases to less than one-third of it when the soil Mo values were corrected for pH (Fig. 155). The variations of plant and soil Mo values in the New Zealand material are narrow and correlations between plant and soil Mo are not significant, even though pH correction changes the direction of the regression from negative to positive. More than one-third of the Mo determinations (Fig. 155) are within the two lowest Mo Zones. Although this limited material does not include any extremely low Mo values, it is possible that Mo deficiency problems might exist in New Zealand. 166 ....,.... 2000 Ne...... 911,,9101 Fig. 152. Regression of Mn content of n= 3 pot-grown wheat (y) on DTPA extract- y ± s12 ET 1h,, 351 6 me 1000MN ErMal %NMI:,MIIMM11=MMEINI NENlcUM WANIWINIMIMIIMENI able soil Mn (x), New Zealand. 800il InTIEWMilliiiIPPEW:VrilFAMWAVIMOMI 600 minrimE=MiSliimmw=1 E 400 8: 30011.111111111111111 -5 200 111111111k .c o 100Nem.mm..T.E=----.....m.....2_rn..111E711 8011MlIMMEIMMIII211MIMMENII IIMIMMINEIRS11=M1111IIMM=IMIMENIIPMMENNIIII kv IIMMIUNIIIIMIIIMENIIIII 20IMMIIIIHM1111111

101111111111011111111 8 HimmIIIIImmq1111111 !! 1 2 3 46 8 10 20 30 40 60 80100 200 300 Mn in soil, mg/I DTPA)

2000 =mmnwIn Fig. 153. Regression of Mn content of Mn pot-grown wheat (y) on DTPA extract- 1000MN iMIMMPIMINIIIMMONIM21.MMININEiiiMilidi Mil1=INNIMIll able soil Mn corrected for pH (x), New 800u. MOMCNIEEZZELla.b.LIMMEN 1=IMMINIII u. INEMMEMEM= MONNINin Zealand. 600III INVEMEIBIMEMIECTIMILWANMOt oX WM=1E1 E IIMITWIliONIIIIIMIENNIWI 11111111M1 400 111PIP" ta:300 200 o

+Ca, 10011.11111111111111rW1111k. mom 80MI.orMeasos IsnMENI=mm-erammammomm==mw-au =IMEMEEN1=iMin uo 60IiiMMIMMEMPIMMEEN.MIMMIIII111111MMMEIMMIIP=1MMIN111111111 40 30 20irrolommnnii

1010111111E1111111 111111 8 =Emma.mimos CM:1111111 171M11111 IMMEMIN 2 3 46 810 20 30 40 60 100 200 300600 pH- corrected Mn in soil, mg/l iDTPA

Zinc. The mean values of both the plant and extractable soil Zn contents of New Zealand are somewhat above the respective international values, irrespective of the extraction method used (Figs 156 and 157). The variations of both the plant Zn and soil Zn are relatively wide and their mutual correlations are highly significant. Forty percent of the Zn values fall within the two highest Zn Zones (IV and V), but in only one sample pair (45427, Luvisol, Morringsville) are the Zn contents exceptionally high. Most of the other high Zn values occur in Tonga and Rarotonga. The six Zn values within or close to Zones I and II come from the South Island where deficiency of Zn should be suspected. 167

,M=OMMIN 11MIINION Fig. 154. Regression of Mo content of KLIVAI4L-4 NI21.5FOMEN M o CIMINO IMM=MIIMIN pot-grown wheat (y) on ammonium 6mosummoilionmE MMMIlECHEIKEINICEEDMI oxalate-oxalic acid extractable soil Mo F (x), New Zealand. 3 2 11111241211 =MINNE=MMI11111E111111 IIIIflUiI:31=177 C .2 o o o .1=MMIIMENM=1=MIWVIIMIVAMMINEN1=MIMININMill1111111111111k ///MMEMENIMIMEMIMMENNNINW=1MMENEOnMMEN- .06 11411 .03 .02 ifi .01

.01 .02 .03 .06 1 .2.3 .6 1 23 Mo in soil, mg/ (A0-0A)

EMPOO!.IMNUOIMMENMS Fig. 155. Regression of Mo content of NIVO17/4.1.11.02..1 AMVIMMITOMMI =EMI =MEN= pot-grown wheat (y) on pH-correct- 6 MINEIIIIMEMBEEEINIIIMITEMEIIMEMIWIMME ed A0-0A extractable soil Mo (x), 0 New Zealand. 3 - o 2iiiiiiiuiiiiiiiuiiiii Wommlommmenismow

2iII!IIlIIIIIIIUIIIIi o .1ii!NIÌIIIP!iiiIIIIII nonte=myeaslal 06

.03 ir WM. .02

.01IIIIIII .006 .01 .02 .03 .06 .1 .2.3 .6 1 23 pH-corrected MO in soil,mg/I (A0-0A)

5.6.4 Summary

The soils of New Zealand included in this study are usually medium to heavy textured, severely to moderately acid, have a high organic matter content and high cation exchange capacity. The P and K contents of wheat and wheat soils are generally low while those of maize and maize soils are higher. The lowest P and K values occur in Tonga. The Ca and Mg contents of plants and soils are low, those from Tonga and Rarotonga being exceptions. 168

200 Fig. 156. Regression of Zn content of Zn pot-grown wheat (y) on DTPA extractable soil Zn (x), New Zealand.

100 90 BO 70 g. 60 as 50

.c40 "6 30 o20 o N

10 9 8 7 6 E BZ

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Zr1 in soil, mg/1 (DTPA)

200 N-r - landte n Fig.157.Regression of Zn content Zr of pot-grown wheat (y) on AAAc- -_,i58 ?53 : EDTA extractable soil Zn corrected ,e7810: 1,8: for pH (x), New Zealand. 100 90 80 70 60 E 50MI I11 a_ Ill! 1111I 11111

N 10 illiirr 9 JOIN 8 ill.91NA MIME" 1.1.1 7 6 MIL Ill1,. .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soi ,mg/l (AAAc-EDTA)

Among the micronutrients, high Fe, moderately high Mn, normal B and low Mo values were typical. In the cases of Cu and Zn the contents varied widely between low and high values. For both micronutrients the lowest values were obtained from samples originating in both North and South Island and the highest came from Tonga and Rarotonga. 169 6. Latin America

6.1 Argentina

6.1.1 General The Argentinian samples were collected from two relatively small, but agriculturally important, areas (Fig. 158). Almost all the soils were classified as Phaeozems.

62. 60 6430333 651 0799 07936544652 796 0 10 20 90 60 805,9 BELLVILLE 791e, '7971798 SAN LORENZO 680,65:7 - CANADAofGOMEZ 7820 767,768 0765,, 7666" I 44800 e66E662 7 663 07"78407r786 62' I 0 ROSARIO _____74734 ea. 0 -476 077i/ 677 CAS I LOA 33. 87'/ 674 679 00763 6-,, ,7720 .075A 676. GUALEGUAV 7750 778 ' 773. 7610 o 66978,1 0 i 7770 .»7790 744.1. 529 780 / 66 SAN NICOLAS/ *527 S40 670 526A 0 776 SOSA 5747:12C 5405,41545455A503 r/ 5i4 Jr:. 6715/ 575 I. 527s, 672 A/ 52504'0509, --,:&25-_' ..,) 51, AA 5A4-- SAN PEDRO 674.0:35,39570 ,,, 554 A546 / 20 572 ATZ/' 5514, ,',85,7 BARADoER 519' StrseA;gr512 '51% .--47;50''' 5,8' 514 /.559 . 592 A7 1.0593 6 5f.:65'66 5e9 594 o 595 5e0565 £5s0 2. 4AA5. 5 5 ':3Z»05:5407 5600 562 T,, 0 561 4450IA 0548 ::::575079 549 se: 552 .88. 86.2.7.70 CHACABUCO MERCEDES 555 o

sr 6

{So ro 44691- 44725 (wheat-soil) 44726-44738( maize-soil)

ARGENTINA

Fig. 158. Sampling sites in Argentina (points = wheat fields, triangles = maize fields). The last three numerals of each sample pair number are given. 170

5 Argentina 1.,M,,a1Internet Argentina Internet Fig.159.Frequency distributions of n= 210 " 3764 40 5= 210 3783 R =43 44 0= 5.47 6.64 texture, pH, organic carbon content, 0 -es= 5 16 es..31 1.12 min.=22 9 35 min.= 4.79 3.62 and cation exchange capacity in soils 1n05:54 92 max 6. 6.96 8.56 0 of Argentina (columns). Curves show

25 the international frequency of the same characters. 20

10 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

Argentina 7- Internet Argentina Internet n= 210 3779 5: 210 3777 = 1.9 1.3 =27.7 27.3 es = .7 1.2 es = 5.4 14.6 35 mm.= 17.0 .2 max.= 4.3 39.1 max = 45.3 99.9 o25 w 20 wCI 15

.2 4 .8 16 32 64 12.8 286 16 32 ORGANIC C,% CEC, me/100g

The soils sampled are medium textured and relatively acid (Fig. 159). Over 90 percent have a texture index between 30 and 50 and pH(CaC12) between 5.0 and 6.0. Onaverage, the organic matter contents of the Argentinian soils are appreciably above the international mean but their cation exchange capacities are at a "normal" level. For these soil properties the ranges of variation are narrow. Low electrical conductivity and CaCO3 equivalent values as well as low sodium contents are typical for Argentinian soils (Appendixes 2-4).

6.1.2 Macronutrients

Owing to the high proportion of organic matter in Argentinian soils, their average total nitrogen content is one of the highest in this study (Figs 6, 160, Appendixes 2 and 3). In spite of the low rates of nitrogen fertilizer applied to the sampled wheat and maize crops (3 ± 14 and 6 ± 16 kg N/ha, respectively), the national averages for N contents of these crops are exceeded by a few countries only. The phosphorus contents of Argentinian soils and original indicator plants vary widely. In general, they are at the average international level or slightly above (Figs 7 and 161). No phosphorus fertilizer was applied to the sampled wheat crops and the rates applied to maize were relatively low (6 ± 18 kg P/ha). The contents of exchangeable potassium of all sampled Argentinian soils are higher than the international average, the variation range is relatively narrow and the national mean values for soils under wheat and maize (784 and 790 mg KA, Figs 162 and 8) are the highest recorded in this study. Although no potassium fertilizer was applied. the average K 171

4 Figs 160-164. Frequency distributions Argentina Internat. ArgentinaInternat 1765 n=90 1958 35 n= 118 35 of nitrogen, phosphorus, potassium, 0 = 4.71 4.27 9= 3.81 3.14 os=.88 1.15 es=.46 .87 calcium and magnesium inoriginal min.= 2.78 .60 ,I, min.= 2.52 .8800 li 3 max.= 6.42 7.45 max.= 4.95 6.51 wheat and maize samples and re- 25 25 spective soils (columns) of Argentina. a >- 20 Curves show theinternationalfre- c.) 5

quency of the same characters. CO

5 3 4 5 6 4 6 N content of wheat% N content of maize,%

7 70 Argentina Internat. Argentina Internat. n= 119 1765 n= 90 1958 60 0= .199 .133 60 9= .207 .135 ei .066 .084 os= .074 .088 min.. .111 .009 min= .124 .008 max.= .425 1.023 nao.. .433 1.657

o Z 30

co W 20 Li-

16 1.28 .005 . .04 08 .16 32 64 1.28 005 .01 .02 .04 .08 .32 .64 Fig. 160. Nitrogen, Argentina. N in wheat soils, % N in maize soils,%

40 Argentina intemot. Argentina Internat. 1765 1967 35 n 118 35 n= 90 0= .42 .38 0 . .33 .33 os. .13 .12 os = .07 .10 .05 30 min.= .17 .05 30 min.= .19 non.. .80 1.02 moo.= .54 1.04 0-25 5- (2) 20

815 E

8 .9 P content of wheat, % P content of maize,%

40 Argentinaint.rnot. Argentina 1nternot. n=1967 35 n 118 1765 5 n= 90 5= 24.4 20.2 10= 23.7 0. 22.5 os = 22.8 24.7 os= 16/ ts= 33.0 min.= 5.4 1.0 min.= .7 min.= 0.1 mox.= 2000 271.4 max.--105.4 mox,656.0

moI

rl ' 'L 4.4Air. hcineieoini_ 2.5 5 10 20 40 BO 160 320 640 2.5 5 10 20 40 BO 160 320 640 Fig. 161. Phosphorus, Argentina. P in wheat soils, mg/l P in maize soils, mg/I

contents of wheat and maize are among the highest (Appendixes 2 and 3, Fig. 8). See also footnote on page 173. The contents of exchangeable calcium in Argentinian soils are below the international average, as are the Ca contents of the original wheats(Fig. 163). The Ca contents of maize clearly show a bimodal frequency distribution. Since there is no similar distributioncon- 172

Argentina Internat. ArgentinaInternet Fig. 162. Potassium, Argentina. n= 118 1765 n = 90 1967 R. 4.7 4.0 0 = 4.1 3.1 = 0.7 1.0 no. 0.5 1.0 min = 3.2 0.9 min. 2.9 06 max = 6.4 68 mas. 5.6 67

2 3 5 6 7 K content of maize,%

ArgentinaInternet ArgentinaInternet n. 118 1765 n. 90 1967 30 R. 784 365 30 R. 790 330 = 198 283 os= 196 356 min =420 20 / min.= 479 18 25 max.= 1649 2097 max =1649 5598

25 50 100 200 4 0 BOO 1600 32006400 25 50 100 200 400 8 01600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

40 Argentina Internat. ArgentinaInternat. Fig. 163. Calcium, Argentina. n=118 35 n 90 1967 = .39 0= .60 .47 *s..08 ts =.29 .20 min.= .21 30 min.. .18 .09 max.= .59 mox.= 1.22 1.88

.8 Ca content of wheaf, Ca content of maize,%

166) Argentina Inter not. Argentinainterne. n= 118 n= 1765 n. 90 1967 5 8= 1993 R= 4671 6=2090 3450 ta = 632 ts= 3076 85= 309 2815 min.= 1020 min= 110 min= 1100 10 max= 5480 max. 21930 3 m00=2990 17995

vC.

2325

Ca. 20 o>- LLI 15

1.11 Q. 10 LL

00 200 400 800 16003 00 6400 12900 25600 50 100 200 400 BOO 1600 3200 6400 12800 Ca in wheat soils, mg/I Ca in maize soils, mg/I

cerning exchangeable Ca in maize soils and no lime applications to maize were reported, this bimodality cannot be attributed to soils or liming. The only marked difference between the maize samples falling into the lower Ca group (Ca <0.6 %) and the higher group (Ca > 0.6 %) that could be detected from the available data was that the low Ca maize samples were collected at later stages of growth. The age difference (41 and 54 days 173

4 Fig. 164. Magnesium, Argentina. Argentina ArgentinoInternat. 5 n= 117 1764 35 n= 90 1967 4=158 .172 R=. .242 .251 *5..032 .060 =.094 .119 mnin..096 044 30 min.= .117 .036 O max.=.244 .948 max.= .474 1.125

:5325 25

(..) 20 20

1.1.1 o15 15 Lu cc 10 10 u-

I 5 Mg content of wheat, % Mg content of maize, % 40 40 1531 Argentina Internat. Argentinaint,rne..or 3 ni= /17 n=1764 35 n=90 19671 =296 0. 489 5.297 446 ts = 72 ts= 437 = 46 min.= 175 min.=10 min.= 191 30 max.= 504 mox.=3298 max.= 537 6490

>Cne20

25 50 100 200 400 800 1600 3200 2.5 25 50 100 200 400 800 1600326 Mg in wheat soils, mg/I Mg in maize soils. mg/I

from planting) was 13 days on average. Although the Ca content of a plant may decrease considerably in two weeks when growing vigorously, it is doubtful that it can explain the whole difference between the two Ca content groups; for example, the data of Sippola et al. (1978) show a decrease of about 30-35 percent in the Ca content of wheat during much the same period of growth. See also Section 1.2.2. For the international standing of Ca in Argentinian soils and plants see Fig. 9. The magnesium contents of original Argentinian wheat and maize, as well as the exchangeable Mg of soils, are somewhat below the respective international averages (Fig. 164). Again there is very slight variation in the exchangeable Mg contents of maize soils but the Mg contents of maize vary widely. The lower and the higher maize Mg groups consist of almost the same maize samples as the respective low and high Ca groups, but the difference is not as pronouncedo. For the international rating of Mg in Argentina see Fig. 10.

I) Comparisons of N, P and K contents of maize between the two sample groups (with Ca contents <0.6 and > 0.6 %) showed that there was a statistically significant difference in the K contents of maize (means of 3.85 and 4.45 %, respectively) with no significant difference in the echangeable soil K. In the cases of N and P the comparisons were obscured by differences in soil N and P contents and N and P fertilizers applied. 174

100 'flIIIIIMIIM1111111111MEM MIIIIMMIIIWIIM11 1711111MINIMMIIIMMIIIMI=1MMEINI Fig. 165. Regression of B content of 80 ialMEMMINIEMEN11M_MINIn pot-grown wheat (y) on hot water 60 MEWI.EMMEEMIIMMINI MInnon 50MNIUM7611=MrE11111MINIM111111 solublesoil B(x),Argentina.For 40MINI-0 74 tOMEIMEIIII =minim detailsof summarizedinternational E 30 background data, see Chapter 4. 20111E111111111 1111111 o a) oiii1111111111111 '58mos.=mmuurammausinumn "E. 6noulim=mmintreVianimmoommil 11111 =1Zaifealn 11111111wW111111 M11111111

211111 iIIIIII 1111111

11111 1 EMI .111111 .2 .3 .4 .5.6.810 2 3 4 56810 B in soil, mg/I (Hot w. sol.)

100 I I " Fig.166. Regression of B content of 801 n= 8 558 pot-grown wheat (y) on CEC-corrected 60 -±-s=5 095 6 (hot water soluble)soil B (x), Ar- 50 T±s= 068 018 0 ± 2 40 gentina. E y= 3 32.8Ix = 85 30 Q. r= 0 537* 02* * *la:. 20

.08.1 .2.3 .4 .5.6 .81.0 2 3 4 56 810 CEC -corrected Bin soil, mg/l (Hot w.

6.1.3 Micronutrients

Boron. Because of the homogeneity of Argentinian soils theranges of variation in B contents of soils and plants are unusually narrow (Figs 165 and166). The Argentinian national mean values for both the plant and the soil B are close totheir international means and no extremely low or high B contents were recorded. See alsoSection 2.7. 175 22 Fig. 167. Regression of Cu content of Ar!e t ra Inte na pot-grown wheat (y) on acid ammonium 20 ne t 53: y* 309±1 7.00 2. acetate-EDTA extractable soil Cu (x), 21 3'5 .00 Argentina. 1111121E1coulliENZIIIII

.6 12 1111IPPF I10 8 06 MEMEMI

2___....1111121111111=1k:

O. 111111EMPLEM111111 .2 .4.6 .810 2 4 6 810 20 40 6080 Cu in soil, mg/l (AAAc- EDTA)

Fig. 168. Regression of Cu content of 2 pot-grownwheat(y)onorganic carbon-corrected (AAAc-EDTA extr.) 2'clan"'!IIIN li soil Cu (x), Argentina. :\;-rs1:11111111I;In's74.1111111111 1 Ilr

13m.g°3-Iniill'r-N,1111111111 pi ' 111111 1111 EMI 1111 I IIMPAIII ill! I 1111=11 ) 11E111 I PM II

3 11111111111111=11 Ii!i i1IÒII 1.111,5211 IMb-19 IN 1111551111" li .,6810 2 1. 6810 20 40 60100

Org.C-corrected soi CU,mg/l (A A Ac-EDTA)

Copper. The Argentinian national mean values for plant Cu and soil Cu are the second and the fifth lowest, respectively, recorded in this study (Figs 27 and 29). Of the 208 sample pairs, only one plant Cu and one soil Cu value exceed the respective international mean (Fig. 167). After correction for soil organic carbon content (Fig. 168) no soil Cu value exceeds the international mean and the number of Cu values falling in Zones I and II is more than doubled. Although no extremely low soil Cu values are recordedin this material, the Cu status of Argentinian soils and especially that of plants is generally low. It is possible that crops sensitive to Cu deficiency would respond to applied Cu. 176

3 00 Fig.169. Regression of Fe contents 2 75 2 50maz.11M11111111=m1111111111111M11111 of pot-grownwheat(y)onacid 225 urBwriIulftR ammonium acetate-EDTA extractable 200=111111131EUNIMEN111111111111 soil Fe (x), Argentina. 1 75 E 1 50 o_ 9:125 o 100 90 80 70 o 60 50

30 10 20 40 60 100200 400 600 1000 2000 Fe in soil, mg/I MAAc- EDTA)

Iron. As is the case for most other nutrients in the Argentinian material, the variation of soil Fe is narrow. In general, the soil Fe values tend to be high, relative to international levels while plant Fe contents are average (Figs 31 and 169). Despite a few plant samples having quite low Fe contents, none of the Fe sample pairs is within the two lowest Fe Zones. No extremely high Fe values were recorded, although 13 sample pairs fall within the two highest Zones. Manganese. The Mn situation in Argentina is very similar to that of Fe (Figs 169-171) as are the locations of Argentina in the "international Mn and Fe fields" (Figs 31, 33 and 37). Molybdenum. The discordance between generally low plant Mo and high A0-0A extractable soil Mo contents (Fig. 172) can be largely eliminated by correcting the soil Mo values for pH (Fig. 173). This also improves the plant Mo soil Mo correlation. Argentina stands close to the centre of the "international Mo field" (Fig. 20). Despite the relatively wide variation in plant Mo contents, no Mo values lie within the two lowest Mo Zones and only a few fall in Zones IV and V (Fig. 173). A single sample pair (44653) from Cordoba province yielded exceptionally high Mo contents. Zinc. Generally, the Zn contents of Argentinian soils and plants are at an average international level (Figs 41 and 43). The mean values for plant Zn content are slightly below and those for soil Zn a little above the respective international means (Figs 174 and 175). Irrespective of the method used to extract Zn from soils, the correlations are high and both methods give a very similar picture of the Zn levels in the country. No sample had a low value (Zones I and II) but about ten percent of the samples fall in the high Zones (IV and V). The Zn contents of some of these, mainly from the Gral Pueyrredon area, were relatively high in both plant and soil. Zn deficiency in Argentina is possible but less likely than in most other countries. See also Section 2.7. 177

Fig. 170. Regression of Mn content of 2000 pot-grown wheat (y) on DTPA extractable soil Mn (x), Argentina. 1000 MMMEME011MAIMIIMM11MWA IVENIIMMCVMMCYAIIM=IMNIN 800UI rravrairk-vrmsRIMIERaninraming 600iimmmumwmammommissuNI 400111MINEIIIIIIIMMENE111111 300 200liall01111111111=1111111=

100=mmolmosompomme..,-elsebtimmam11E111111111111MIN, 80rirawarlmillearastuawai 6011111111101MINIMIMMEMUNICINGINIMM 4011111111111111.1111111=111111 30IIIIIMIIMS0111=111111111111 20ima-nommiloll

101111111111111111111111 8IIIIMMINNMEIIIMIliMMIESTOMMEE.VMMMMIK MEMEMININIMM IMMUILIMMEN 23 46 8 10 20 301.0 60 80100200 300 Mn in soil, mg/l DTPA)

Fig.171. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA Mn extractable soil Mn for pH (x), Ar- 1000NM maimoimmon=1.6mgimmwoo11111111-11111 800 INEIECIEhl!ffidillEELIM1.E.qM gentina. ---....i--'..IMagatrillk0/21110.1WELEttlan 60011=119MIMINIIIMEMEMEIMISBNI a 400IMMIMMIN0111111111111 ti 300IIUUIIIIIIuuIL ce _c200110.11111111111Olil "6 100I.11.111111111111 P941111111 garmunnum.....7-.3dgit=011.IMM111M1r,MI 80aomwfammwsziZiggemainn 'IMMEW111.1111 o unalmmmosiiimiramminiwIldniSin =011111111 o 6011111MM=.MMEIMPIMMINNISIIIIMINI 40I MMEINIIMENNIENIMININGIMIll Mulit. 30 PN" 20

10111111111iIIIII 8'annul.= m.. mmsmosmom smosaswiewnenmmermievammom 11====.11Im=MMIrrn 2 3 46 810 20 30 40 60100 200 300600 pH - corrected Mn in soil, mg/1 (DTPA)

6.1.4. Summary The Argentinian soil samples (mainly Phaeozems) were rather homogeneous with medium texture, medium cation exchange capacity, low pH and high organic matter content. The K contents of soils and plants are the highest and N contents among the highest in this study. P contents are of an average international level. The Ca and Mg contents of soils are on the low side but those of maize vary widely. Among the micronutrients only some Cu and Zn contents are relatively low. No low values were recorded for B, Fe, Mn, or Mo. 1 7 8

111 mmlmImINII 1:17411511MMINI Fig. 172. Regression of Mo content of MI141111 MN IM=UNI 611101.11B11IMM1101nn pot-grown wheat (y) on ammonium 1MTIEREEIE=CESIECIII oxalate-oxalic acid extractable soil Mo 21111120 112 Ill 3 Eillitito2 1 (x), Argentina. 2 illiii immosrmgmnowour MMEINIEMMEMIlrio"-MENINOMMInsmOr UIIIIIShiIMMIEWLINY

.2M11111111111111.811,

O o o 1 1171111111111111111\ INMENNI=MMIIIIMIIMENw-3-=11:111=-Falral0==r- .06W1=IMMUIIMIIMMMINAIMEN111 .03 .02

.01

.01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/ (A 0- OA)

1nonoAIMIM=eml=1= Fig. 173. Regression of Mo content of MoIIIIIITALLasimommaminsrart"..;iIIMMOMMIN 6 pot-grown wheat (y) on pH-corrected A0-0A extractable soil Mo (x), Ar- 0± 0 C" 6 11111=111111211111 3 gentina. 21111111Yro°1511111111BEilliiiiiil

O. 11111111111111111111111 O. 1 o =MME1111g .6IENNENmmosolwimaisill,iiiiiiiiiiiiiii ó .3ENIIIIIMME11111111EMENIIIII .2unimmwoupsommum. o .11011"11111111rA11111111 ME101112=1111 APIIMIIMIMMIMMEAMININIMEMIIMIMMMENN 06=MMENIIINMINEMIMMINE 11.111 IMIIIMAMEMONIIilillaPMERIUMM=EMi=MUM .03IIIHi!LlI!AIII .02

.01 .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MO in soil,mg/I (AO- OA)

The great majority of values for these elements are within the normal international range but, depending on the element and extraction method used, one to ten percent of the values fall within Zones IV and V. This study suggests that, in the Argentinian areas sampled, micronutrient deficiencies are less likely than in most other countries. 179

Fig. 174. Regression of Zn content of 200 A i IL' Inett 3 3 pot-grown wheat (y) on DTPA extract- Z n 72., ? n ys 1 3 : 377 able soil Zn (x), Argentina. iii.= t.- as1,971 5 y=1 . 8 1 (n.y 1,23(+C321 log t r= 0.6s ' r0 32'" 100 90 I, 80 70 IIMM I g_ 60 "II 50 \ .2 40\ 30 . \ . MI 20 o .1. -4°1 .' N

10 A :.:141111U 9 8."- 7 6 1 I la \ D\ se 1 . .3 .4.6 .8 1 23 46 810 20 3040 60 100 Zn in soil, mg /I (DTPA)

Fig. 175. Regression of Zn content of 200 pot-grown wheat (y) on AAAc-EDTA Zn' e ' (3.t:" -1 . .6 131 extractable soil Zn corrected for pH 8_s- .8 8±; (x), Argentina. 100 90 80 iiiiiilniiialMr.1.6 70 60 50 1111111110111 di 40MINIMEllillWIN 30 11111111M11110/11

20 1111111PIPMI I

".X4 k 10 " P111111 9 gig 8 Ilklik IIIIIIMMII 7 6 1 1 Mt ill .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soi ,mg/1 (AAAc-EDTA) 180 6.2 Brazil

6.2.1 General

The Brazilian soil and original wheat samples for this study come from the southern part of the country (Fig. 176). No maize fields were sampled. The majority (69 %) of the sampled soils were classified as Ferralsols. Other soils were Nitosols (14 %), Luvisols (9 %) and Acrisols (8 %).

60° 50° 40°

10°

BRAZIL

56* 48°

SAO PAULO 22° 22* _

CORI-FIB 0,, 07 559 ° 5 0579e. )54.."'525 )" "" n'

F ORIANOPOL IS 26* 26°

30° 30*

100 100 300 400 500 Km

56° 52° 48°

Fig. 176. Sampling sites in Brazil. The last three numerals of each sample pair number are given. 181

Fig.177.Frequency distributions of BrazilInternet Brazil Internet. n. 71 3764 40 n. 71 3783 texture, pH, organic carbon content, 8. 67 44 8. 5.10 6.64 30 4s. 17 16 4s. .57 1.12 and cation exchange capacity in soils min.= 17 9 35 min, 3.97 3.62 6, max.. 88 92 max... 6.30 8.56 of Brazil (columns). Curves show the 0 international frequency of the same 20 25 characters. 20

0 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

Brazil Internet Brazil Internet n 671 71 40 3779 n. 3777 2.0 =27.1 27.3 .8 its=8.5 14.6 1; 35 min= .6 nmmn=6.4 .2 max 7 4.7 39.1 max 746.8 99.9

2 .4 .8 16 3.2 6.4 178 25.6 ORGANICC. GEC,IL me/100g

Although some light textured soils are represented the majority of soils are very heavy textured (Fig. 177). The national mean texture index is the highest recorded in this study (Appendix 4). All Brazilian soils are severely or moderately acid, pH(CaC12) 3.97-6.30, and a lower average soil pH occurs only in two African countries. The contents of soil organic matter are high compared to most other countries. In spite of this and the heavy texture of most soils, the cation exchange capacity values are not very high but usually at an average international level.

6.2.2 Macronutrients The total nitrogen contents of Brazilian soils are high, the applied rates of nitrogen fertilizer low (19 ± 11 kg N/ha) and the N contents of wheat at an average international level (Figs 178 and 6; Appendix 2). In spite of a fairly low phosphorus content in Brazilian soils the P contents of original wheats are at an average international level or even slightly above (Figs 179 and 7). The sampled wheat crops were heavily fertilized with phosphorus (49 ± 30 kg P/ha) which apparently increased the P contents of plants. The variations in the plant P, soil P and applied fertilizer P are wide. The potassium contents of Brazilian soils and wheats are similar to those of phosphorus: low exchangeable soil K, fairly high plant K and relatively high K fertilization (33 ± 18 kg K/ha), all varying widely (Fig. 180). The calcium contents of wheats and the contents of exchangeable Ca in Brazilian soils are low; the national mean values for both are among thelowest recorded in this study (Figs 181 and 9; Appendix 2). Many of the sampled fields had been limed during recent 182

40 35 BrazilInternat. Brazil Internat. 40 n. 70 1765 35 n= 70 1765 BrOZ i I Internat. 6= 3.96 4.27 30 0=4.4 4.0 =1.2 1.0 1.15 3 n. 70 1765 ±s= .91 8..42 .38 min.. 1.5 0.9 tID min.. 2.10 .60 Z *5..12 .12 St25 max.= 6.8 6.8 rnax.= 6.04 7.45 ?,330 min...23 .05 max. .75 1.02 5 66 a. 0.25 a >- >- (_) 220 /11 5

C3 Lit LLI CC

5 5

3 4 5 .1 .2 .3 .4 .5 .6 .7 9 2 3 4 5 6 N content of wheat, % P content of wheat, % K content of wheat, % 40 70 BrazilInternat. BrazilInternat. 35 n. 70 1765 n 71 /765 9=10.9 20.2 Brazil Internat. 60 0=.184 .133 is= 9.6 24.7 n .70 1765 min..2.4 1.0 ±s= .057 .084 30 max.= 74.2 1.4 R = 170 365 Min.= .051 .009 ts= 111 283 o -E. a, 50 max.= .325 1.023 a, min.=37 20 I.) 9 max, 632 2097

'(r) Q. 40 20 o>- Z 30 U.1 o U.1 20 CC L'-

5 10 20 40 BO 160 320640 25 50 100 200 400 BOO 1600 3200 6400 .005 .0 .02 04 08 6 .32 .64 1.28 2.5 N in wheat soils, % P in wheat soils, mg/I K in wheat soils, mg/I Fig. 178. Nitrogen, Brazil. Fig. 179. Phosphorus, Brazil. Fig. 180. Potassium, Brazil.

40 Brazil Internat. Brazil Internat. Figs 178-182. Frequency distribu- 35 n. 70 1765 35 n.70 1764 0..34 .43 /. .223 .172 tionsofnitrogen,phosphorus, 2S. .11 .17 ts=. .056 .060 ¡.13 30 non. .16 .11 2330 mm.. .140 .044 potassium, calcium and magnesium mar. 66 1.68 max...449 .948 :7) k. 25 in original wheat samples and re- >- >- spective soils (columns) of Brazil. U 20 0 20 Curves show the international fre- 111 oD 15 oD 15 quency of the same characters. (X 10 CL 10 U..

.8 5 .< Co content of wheat, % Mg content of wheat, %

Brazil iniernot. Brazil Internat. n 70 1765 5 6.1004 4671 n. 70 1764 es. 782 3076 227 489 min. 180 110 i.e. 137 437 most 5730 21930 min.. 40 10 3 30 max.. 1031 3298

ú25

'CB 20 o7-

LiJ15 o cc 10 U_

100 200 400 800 1600 3200 6400 12800 25600 25 50 100 200 400 800 1600 3200 Ca in wheat soils, mg/I Mg in wheat soils, mg/I Fig. 181. Calcium, Brazil. Fig. 182. Magnesium, Brazil. 183 years with small to moderate rates of application. In spite of the low exchangeable magnesium contents of soils the Mg contents of wheat were high (Figs 182 and 10). No information was obtained on possible magnesium fertilization or on the quality (Mg content) of applied lime.

6.2.3 Micronutrients

Boron. On average the B contents of soils and plants are slightly low in the "international B fields" (Figs 22 and 25). The ranges of variation for both the soil B and the plant B are exceptionally narrow (Figs 183 and 184). The whole of the Brazilian sample material is

Fig. 183. Regression of B content of 100 MINerwimmiseliammisomms msa 80 allIST41MMIMI6Tir-5671111milin pot-grown wheat (y) on acid ammonium 1111111111111111111111111111111111 OM 60 37WillainnilIMM soluble soil B (x), Brazil. For details of 50 summarized international background 40 -= 0 63*OMIEL1E1111 MI 1111 data, see Chapter 4. 30 a.- 20 11115' 1111111 11=1111 o

1 "58 INISNM=MMEINNIP 11111M111 a)56 INHENI=MEIPANCYMM=MEMNII11111111=Ulanill'iNIMIll o4 111MEITM:1111 o .... IIi1ÏlIIMOWER MI 3 III 111111111 2 191,

Ill 1 .2 .3.4.5.6 .810 2 3456 810 B in soil, mg/l (Hot w. sol.)

100 _11I _1111 L _L II I LA_L A

801B n= 60 -ts.4 8 089 6 9 4 50 71-s= 0 6 .0 2 0 3 7 40 E. 30 log y.0 5 .01 y= 8 5 o. r=0458 " 0 2 15; 20 .c 10 8

w 6 4 co 2

Fig.184. Regression of B content of pot-grown wheat (y) on CEC-corrected .08:1 .2.3.4.5.6.8 1.0 2 3 4 5 68 10 (hot water soluble) soil B (x), Brazil. CEC -corrected B in soil, mg/i (Hot w.sou 184

22 Br z I Inte na Fig. 185.Regression of Cu content of 20 pot-grown wheat (y) on acid ammonium ECW11=11111111 acetate-EDTA extractable soil Cu (x), 18 Brazil. 8:161111111NIZIP °6111111111111111! o Q ) 1411111111111111111111111111111

,46 12111111111111111111111111111/111

71101111111111111111115/11111 38

61111111111111M111111111111

41111111111111=11111111111111

.4 .6 .810 2 4 6 810 20 606080 Cu in soil, mg/I (AAAC- EDTA)

Fig. 186.Regression of Cu content of 2Cu i pot-grownwheat (y)on organic )Bra I i Intern.t carbon-corrected (AAAc-EDTA extr.) yr58=99±311,;;..:30± 57 1 soil Cu (x), Brazil. 3 R+s=1i+ ' 94 logy.'080 .4og logy=5870 5 g r4.0 64'' i r=0 71*" 5 A1

4 imp.

2 k II I Al

3 1111 .. Plitib 5 IRA I

5 11111111/1111.11M I t. iiniiiuuiiòmi 2 Milliiialli.....__ I Wit 1 -Iiiimil fi

3. .2 .4 .6.810 2 L. 6 810 20 40En1n0 Org.C-corrected soi CU,mg/l (AAAc-EDTA) within the "normal" B range (Zone III). There is not indication of B excess but at some sites response to B, depending on the crop, could be obtained (see also Section 2.7). Copper. In spite of the generally high organic matter contents and low pH of the soils (see Figs 28 and 30) Brazil stands in the "international Cu fields" at the very highest level on a par with the Philippines and Italy (Figs 27 and 29). This indicates high total Cu contents in Brazilian soils. The Cu contents of plants as well as the contents of extractable soil Cu vary largely. One half of the Cu values are within the normal Cu range and the other half in the two highest Cu Zones (IV and V); some would therefore indicate at least excess of Cu if not Cu toxicity (Figs 185 and 186). The distribution of the highest Brazilian Cu values is not limited to any specific geographical area but they are relatively more common in Luvisols than in other soils (see also Fig. Cu in Appendix 7). 185

Fig. 187. Regression of Fe contents of 300 275rar:7111111111111WRIMISUIPIIIIII6 pot-grown wheat (y) on acid ammonium 250maz111111111111M=MINI1111111111111 acetate-EDTA extractable soil Fe (x), 225 Brazil. 200 log y r 1 75 11111131211iiiiii11111. E 1 50 cL ilumgrinum 90mammon i kunumn 80...... ,.. Q) 70milibmwmagurail o 60mmollempirgnews 50penniiiiiiiiiimi 40ri" IIII ..1111111111111

30 -iiriuti, 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc-EDTA)

Iron. The Brazilian averages for the Fe contents of plants and soils are almost the same as those for the whole international material (Fig. 187). Since the variations arerelatively small most of the Brazilian material is within the normal Fe range and there are no extreme Fe values. Manganese. The mean Mn contents of Brazilian soils and plants are about three times as high as the respective general international means in this study (Fig. 188).Owing to the low pH of Brazilian soils, correction of soil Mn values for pH not only improves the plant Mn soil Mn correlation but also increases the difference between the Brazilian and international mean values for soil Mn (Fig. 189). The Brazilian mean for extractable soil Mn is the highest recorded in this study and the national mean for plant Mn is exceeded only by three African countries (Figs 33 and 37). Although widely varying plant and soil Mn values are included in the Brazilian material only one Mn sample pair falls in the lowest Mn Zone. More than half the material is within Zones IV and V, including many samples with extremely high Mn contents. The highest Mn values were relatively more typical for Nitosols and plants grown on these than for other soils, although many extremely high Mn contents were found in Ferralsols and associated plants. Mn toxicity to soybean in Brazil was reported e.g. by Almeida and S'fredo (1979). The risk of Mn toxicity could be reduced or even eliminated by raising the soil pH through liming (see also Fig. 34). Molybdenum. Unlike Mn, the Mo contents of Brazilian soils and plants are low (Figs 190 and 191). On average, the pot-grown wheat contains ten times as niuch MO as the wheat grown on Brazilian soils (0.322 and 0.032 ppm, respectively) and the Brazilian mean is the lowest among the 29 countries investigated (Figs 15 and 20). The apparently medium high A0-0A extractable soil Mo values become much lower when corrected for pH, and would be even lower if corrected for texture (Fig. 21). The effect of texture must therefore 186

2000 Fig. 188. Regression of Mn content of

n5E 3538 pot-grown wheat (y) on DTPA extract- 9±5 28 3 1000MNMn Bri.1%M zil /UMW MECLIVILM.101 =MME able soil Mn (x), Brazil. EN M.IMIMMINSMI 800EN IMIWZIEMITIMIENTZWIFLWAMMITICIE 600MI 400 a.D 300 V 200III 11111 -c III 3 o 100 80:1111111WrildaniIMENIMIIIINW1 g60NIflINIIMENNIMMI. o 1 . VIMME=MIManiiiill..... c40 30 20Ilarall1111

10 8MI IIIIMENIMBilOIMILIMTOMMEM ilifil IMMINNIMMUMKUIRMEN !! 2 3 46 810 20 30 40 60 80100200 300 Mn insoil, mg/l (DTPA)

2000 Fig. 189. Regression of Mn content of Mn pot-grown wheat (y) on DTPA extract- 1000 MINPiMMIMIPPO16EMENEMEMNE1111.191/111 IMote able soil Mn corrected tor pH (x), MN M. 6.1311ffirE11,1.7kIMMINEEIYAMINIEN MIIM1=--1511 800NM 21M1=WIENIM MOMMINIM Brazil. 600 MMWEININCIERME=EIMITIEEIMMII. x MOINIMMI21 E 1111WEINEMilinlIMNiNELEIRMIIII =MEWS" a_ o.400 ,so300 200 iiUIIIII o 100 11111 a) 80 Nimouni-rimmonni Nomazemr,a. 1111111MPLIii=MENNI =maunia=MICINNI 60IiMMIMMEMPril=111MUI M. -MENU 40 NriP" 30 61, 20Ii

10 8 mmiImmossoNamimmEs momms IMMEIMM-MiWanENNIMIIMMEMEN =MIF,111 2 3 46 810 20 30 40 60 100200 300600 pH - corrected Mn in soil, mg/l (DTPA)

be kept in mind when interpreting the Brazilian Mo data. Probably because of analytical inaccuracy in determining very low and only slightly varying Mo contents, the correlation between plant Mo and soil Mo for Brazilian material is not significant except when corrected for both pH and texture (Table 8). Less than one-third of the Brazilian Mo values are within the lower half of the normal range, and almost half lie in the lowest Mo Zones (Fig. 191). Low Mo contents are perhaps more typical of the Ferralsols and associated plants than of other Brazilian soils. To recapitulate, it is evident that the basic reason for the low Mo contents of Brazilian wheat is the low plant availability of Mo due to two typical characteristics of the Brazilian soils, namely the acidity and the heavy texture (see Figs 16 and 19). The great majority of Brazilian wheats with the lowest Mo contents (0.02 187

Fig. 190. Regression of Mo content of M o pot-grown wheat (y) on ammonium 6 = 8 53 037 32 ±36 oxalate-oxalic acid extractable soil Mo ± 9 072 021± 27 (x), Brazil. 3 log y- 4 3 x y = 2o 4 r= 008 0 24 2 E o_ a 1

11:1 2 .6

o .3

.2 -do o .1 .06

.03 .02 n .01

.01 .02 .03 .06 .1 .2 .3 .6 1 2 3 Mo in soil, mg/t (,40-0A)

Fig.191. Regression of Mo content M oi===111:i rrInzm =12 11 of pot-grown wheat (y) on pH-correct- 6 n=3537M11 ed A0-0A extractable soil Mo (x), ± 5- 901163 o EMU Brazil. 3IEN EZIERIMIZZatnalli 2 E 11111111111in iirldr111111Wr 0.o-1ME=MENIMMP1IMI=MMEM NNW' o TM 1MI .6

"6.311111 1111- 1111111161L .2 *C.o o .1

.06

.03 .02 IIiI!iiÌiiiAhIIIIIIII

.01

.006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MO in soil,mg/t (A0-0A)

ppm or less) had grown on very heavy textured soils (TI > 70) usually accompanied bylow pH (< 5.0). Obviously, in the long run, raising the soil pH by liming would be the best means of increasing the availability of Mo. Zinc. The average Zn content of Brazil is high in the "international Zn fields" (Figs 41 and 43). The standard deviations of both plant Zn and soil Zn, however, are larger than in 188

200 In e t Fig. 192. Regression of Zn content of nt '7 n 3 3 IZn 7-1 -' ys 1. 1 77 pot-grown wheat (y) on DTPA extractable soil Zn (x), Brazil. 7.1yr a =:::6±16I 9 "lo y i1215321Iog r. C.6 ' rO. 3 ' 100 90 80 70 g. 60 °,-; 50 40 30 6LiiAi e" 4g, 20 k iimFiRlir : 1 pp- I,

10 9 8 LA.1111..111111mm. 7 6 1111111111111111=1 1E .1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Znin soil,mg/l (DTPA)

200 Fig. 193. Regression of Zn content of pot-grown wheat (y) on AAAc-EDTA extractable soil Zn corrected for pH 100 111 (x), Brazil. 90 80.r.611:1Mitillird: 70 60 E 50hoimmigir

.40cl 3011MINK -6 20 nomplet INI 4cq; o lompop -

NJ 10 1111 1I ak 9 MEL =MEMO 1= 8IN MIIIIIILL IMMENIIIIIMMIM ni. 7111MMUliik moollnimm UM. 6111111111fil MIIIIII=E IMMINEFAMAIMINNW 11,4 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soil, mg/l (AAAc-EDTA)

most other countries and irrespective of the method of extracting soil Zn the correlations are good (Figs 192 and 193). A relatively high percentage (16-21 % depending on extraction method) of the Zn sample pairs are within the two highest Zn Zones and two of these represent very high values. Although no sample was extremely low in Zn, a response to Zn at several sites is possible (see also Section 2.7). 189

6.2.4 Summary

Most of the Brazilian soils studied, mainly Ferralsols, have heavy texture, low pH, high organic matter content and medium cation exchange capacity. The macronutrient (P, K, Ca and Mg) content of soils is usually low but that of N high. The P, K and N contents of original Brazilian wheat are at an average international level, but whereas the Ca contents tend to be low, the Mg contents tend to be high. The sampled crops were fertilized with P and K at high rates but with low rates of N. The standing of micronutrients in Brazilian soils and plants varies considerably from element to element. Only the contents of Fe are at the "normal" international level. The content of B is somewhat low, that of Mo is very low while those of Cu and Mn are very high. The contents of Zn vary widely between high and low. 190 6.3 Ecuador

6.3.1 General

The wheat and soil samples were collected from Cañar,Carchi, Chimborazo, Imbabura and Pichincha Provinces. Maize soils were not sampled. Because of the small quantity of soil in the samples received from Ecuador only a part of the analytical programme could be carried out, and therefore only incomplete data can be presented. The pH of soils vary greatly, from 4.7 to 7.7, the majority being on the acid side (Fig. 194). The organic matter contents are at an average international level and cation exchange capacity somewhat below that. Very high electrical conductivity is typical for soils sampled from Ecuador (Appendix 2).

Ecuador 6lol Fig.194. cuador 16185841 Frequency distributions of n r 15 3783 40 11r 6.22 6.64 6 r 13 3779 pH, organic carbon content, and cation r .62 1.12 Er 1.4 17 1.3 :sr 6 1.2 run r 4.66 3.62 1 exchange capacity in soils of Ecuador max.. 7.65 6.56 ma, 6 uncx r 1 3 39., (columns), Curves show the international frequency of the same characters.

.8 16 31 41 20 256 pH (Ca C12 ORGANIC C,%

Ecuador 16183661 n r 11 3777 r 21.3 27.3 :sr9.5 ' 4.6 6,6..9.7 .2 WI,. :2.1 999

15 CEC.6016/100g

6.3.2 Macronutrients

In spite of the relatively high total nitrogen contents of soils and some nitrogen fertilization (44 ± 13 kg N/ha) the N contents of the original wheats are low by international standards (Figs 195 and 6; Appendix 2). In spite of the small amount of material submitted, the ranges of plant and soil N variation are quite wide. One wheat sample especially (No. 47356) differs from the others due to its high N content. This sample had grown on a soil with the highest total N content recorded in any Ecuador soil, and had also been fertilized 191

40 3 Ecuador Internat. EcuadorInternat no 15 1765 610 16 1765 35 EcuadorInternat. 30 0= 4.3 4.0 0=3.47 4.27 n= 15 1765 is= 0.8 1.0 35 ±-s. 1.03 1.15 04.43 .36 min.= 3.4 0.9 30 min.= 2.21 .60 xs =.06 .12 St 5 max = 6.0 6.6 max, 6.43 7.45 830 min.=.33 .05 max.=.55 1.02 jj 25 a 25 >- c>..-.1 20 o

LLI (- -1 15 015 0 u_CC 10

.2 .3 4 .5 .6 .7 .6 3 4 5 2 3 4 5 6 N content of wheat, % P content of wheat, % K content of wheat%

70 Ecuador Internat. n= 15 1765 47 Ecuador Internat. R. 17.7 20.2 3 Is. 16.3 24.7 Ecuador Internat. 60 n= 16 1765 .133 min.= 2.4 1.0 no15 1765 0..209 max.= 67.6 271.4 -45..162 .084 30 0= 280 365 -E min.=.049 .009 ±60 93 283 a, 50 min.= 134 20 0 max...747 1.023 25 5 max.= 472 2097 to CD' O_ 40 3- o3- 20

Z 30 111 D 15 5 o C:2 111 L4.1 20 CC IX u. 10 U_

5 10 20 40 BO 160 320 640 25 50 100 200 400 800 1600 32006400 .005 .01 .02 04 .08 .16 .32 .64 1.28 2.5 N in wheat soils, % P in wheat soils, mg/I K in wheat soils, mg/1 Fig. 195. Nitrogen, Ecuador. Fig. 196. Phosphorus, Ecuador. Fig. 197. Potassium, Ecuador.

4 40 Ecuador Internet EcuadorInternat. Figs 195-199. Frequency distribu- 15 n=1765 35 n 15 1764 5 8.25 0. .43 R..145 .172 tionsofnitrogen,phosphorus, =.05 ±5 = .17 050.000 .060 min.. 17 minx .11 16, 30 min...096 .044 potassium, calcium and magnesium moe=.32 max.= 1.613 max = .267 .948 in original wheat samples and re- 25 spective soils (columns) of Ecuador. o 20 V Curves show the international fre- ILI oD 15 quency of the same characters.

E10

/. .8 1.5 5 2 Ca content of wheat, % Mg content of wheat, %

40 Internat. Ecuador Ecuador Internat. 15 n= 1765 15 1764 35 0.1884 R. 4671 35 56=1082 or.. 3076 421 469 min, 590 min..110 04= 287 437 min.= 78 10 max.= 4110 max=21930 3 30 mox..1038 3298

1S' 25 825

20 CL 20 o>- o>- LU 15 Lit15 o o LU c,10

100 200 400 boa ion 3200 6400 12800 25600 25 SO 100 200 400 BOO 1600 3200 Ca in wheat soils, mg/I Mg in wheat soils, mg/I Fig. 198. Calcium, Ecuador. Fig. 199. Magnesium, Ecuador. 192 with the highest amount of N (60 kg N/ha) applied to the plants sampled from that country. The phosphorus and potassium contents of soils in Ecuador are slightly below, and the P and K contents of wheats somewhat above the respective international means (Figs 7, 8, 196 and 197). The rates of phosphorus fertilizers (23 ± 8 kg P/ha) applied to the sampled crops were higher than those in most other countries in this study but somewhat less potassium fertilizers were used (13 ± 6 kg K/ha, Appendix 5). The frequency distributions presented in Fig. 198 show that the occurrence of calcium in the soils and wheat of Ecuador is low in an international context. In fact, the mean Ca content of wheat in Ecuador is the lowest national mean obtained in this study (Appendix 2). The magnesium contents are slightly lower than average (Figs 199 and 10; Appendix 2).

6.3.3 Micronutrients

The small size of soil samples received from Ecuador precluded pot experiments, and the value of having an indicator plant of one species and one variety in the international context was lost (see Section 1.2). It is therefore attempted to give a general picture of the micronutrient content of plants in Ecuador by comparing analytical data of the original wheat samples with the respective data from other countries given in Appendix 2. Boron. The B contents of the sampled soils and plants of Ecuador vary very little. All plant B values are below and only one hot water extractable soil B value exceeds the respective international means (6.56 ppm and 0.81 mg/1). The national mean for soil B (0.42 mg/1) is the fifth lowest and that of plant B (2.33 ppm) the second lowest in the entire study (Appendix 2). From these limited data, B deficiency seems likely in many locations in Ecuador. Its deficiency for several crops and response to B fertilization have been reported by Tollenar (1966) and Mestanza and Lainez (1970). Copper. The Cu content of plants in Equador, based on the results of plant analyses, seems to be generally low. However, although the national mean (7.7 ± 1.7 ppm) is the seventh lowest (Appendix 2), the Ecuador material does not include any extremely low values. Manganese. Mn contents determined from the Ecuador wheats are, with one exception, below the international average and the national mean (34.3 ± 18.8 ppm) is the very lowest recorded for original wheats (Appendix 2). The uncorrected DTPA extractable soil Mn in Ecuador soils varies from 1.2 to 76.7 mg/1 and the mean (24.1 mg/1) is only slightly below the international mean. If these soil values were corrected (see Fig. 34 and related text) it would lower the mean by about one-third and greatly reduce many of the single values. It is possible that a response to Mn fertilizer would be pbtained especially in soils where the soil pH is relatively high (e.g. sites 47353-54 in Imbabura and 47362-65 in Chimborazo). Molybdenum. Although the average Mo content of original Ecuador wheats (0.40 ± 0.40 ppm) is far below the international mean (0.94 ± 1.03 ppm), the mean for the uncorrected A0-0A extractable soil Mo (0.249 ± 0.232 mg/1) exceeds the respective international mean (0.204 ± 0.209 mg/1, Appendix 2). Correcting the soil Mo values for pH would largely eliminate this contradiction (see Fig. 16 and related text) by lowering the soil Mo values. This limited material from Ecuador does not include any exceptionally high Mo values, and although many I\4o values are low there is no indication of any severe Mo deficiency. 193 Zinc. The plant and soil Zn values determined from the Ecuador sample material are usually at or somewhat below the average international level (Appendix 2). No extremes were recorded.

6.3.4 Summary

Because of the limited material and incomplete analytical data only a vague picture of the soil nutrient content in Ecuador can be deduced. Most of the sampled soils have pH (CaCl2) below 7, a medium content of organic matter and low to medium cation exchange capacity. The macronutrient contents of soils are internationally low to medium, N being an exception. For micronutrients some very low values, especially of Mn and B, were recorded but, in general, low to medium micronutrient contents are typical for soils sampled from Ecuador. 194 6.4 Mexico

6.4.1 General

The geographical distribution of sampling sites in Mexico is given in Fig. 200, separately for wheat and maize fields. The most important wheat and maize production areas were quite well covered by sampling. Out of the 214 soils classified, Xerosols (28 %) were the most common, followed by Kastanozems (21 %), Yerrnosols (16 %) and Fluvisols (14 %). Both wheat and maize were grown on these soils. The rest of the soils were used mainly for growing maize: Regosols (5 %), Vertisols (4 %), Phaeozems (3 %), Andosols (3 %), Arenosols (2 %), Chernozems

10. 109. - 4.. 4 64,3 5 \

46:26 ..,

.,...r ...,...... :- 984030

4840 423405>Wk 8 25,HERMOSILLO 4840 48906 \

\ . 46\ 984001 .16.-7;S9.22 j 4 "''..,:g16 t 48407 4 6 96'09 983980 ,.. __, _76012 CHIHUAHUA° e 48397 0 i .. \ 4 0 9------flgT06 ',,,,..45139504 4- 16:03 ,104 --.-11:, ....,\4eael 9640 46401 40383.1.4838 483940 46441.. ',/ 098396 463135: e t .464 :11IZP".:99 46445 48394 *0412" ,s0,46446 05' ,,,,,,_.... 4644849450 48= .V 48396 Q - .415 , 4841 4e9717.9% 43'V aa7, 98916 4' ...._.4.4113 49377 983¡3 983 4 \ 49499 , 0 CteLlgAN TORREON t 8313?4" --,,j, 413390 1.10,6t

------.._010,IRANGO

05APNOW

COWIA .,_

MEr0 O. 16° CRUZ

CHILPANCINGO MEXICO a. 100 Fig. 200 a. Sampling sites in Mexico (wheat fields). 195 (1 %), Cambisols (1 %) and Ferralsols (1 %). As illustrated in Fig. 201 the Mexican soils vary greatly in texture, from very coarse to very fine. The great majority of soils are alkaline but some with quite low pH are included. Most of the soils are low in organic matter and although soils with medium or high cation exchange capacity dominate there are also some with very low CEC. Electrical conductivity and CaCO3 equivalent values vary greatly (Appendixes 2-4). Every sixth soil contains more than 500 mg/1 of CH3COONH4 extractable sodium. These have pH(CaC12) usually around 8, high electrical conductivity and high B content. In general, a wide variety of soils, with extensively varying characteristics are represented in the Mexican material. Consequently, the nutrient contents of these soils differed greatly.

46290"'' 46295

46272

ORREON

LIRANG

TEPIC o46246' o'gT51'

,,Its2A 46242 ,.;T 96185

46 0 _t2,1°79:225 4610 .-/-94e4

cj 61

Fig. 200 b. Sampling sites in Mexico (maize fields). 196

35 exicotnternat MexicoInternet Fig.201.Frequency distributions of 3764 n 247 3783 n=247 0 7= 49 44 6.7.19 6.64 texture, pH, organic carbon content, 0 rs.17 16 .83 1.12 mtn,= 10 9 35 min = 4.10 3.62 and cation exchange capacity in soils of max=83 92 max = 8.25 8.56 30 Mexico (columns). Curves show the

25 international frequency of the same o>- characters. 20 ui 15 o Lk' 10 10

0 20 30 0 50 60 70 BO 90 6 TEXTURE INDEX pH (Ca C12)

5 Mexico internal n. 247 3777 = 32.0 27.3 30 xs. 14.6 14.6 min. .2 .2 max =80.3 99.9

16 3.2 6.4 12.8 256 16 ORGANICC,% CEO, me/100g

4 Mexico Internat. Mexico Internet Figs. 202-206. Frequency distributions 1765 n= 147 1958 35 n= 100 35 7= 4.88 4.27 7= 2.86 114 of nitrogen, phosphorus, potassium, xs..87 1.15 ,s= .70 .87 30 min.. 2.28 .60 30 min.=1.40 .88 calcium and magnesium in original max.= 6.46 7.45 max.= 4.67 6.51 25 wheat and maize samples andre-

>- 20 spectivesoils (columns) of Mexico. Curves show theinternationalfre- 1,115 o quency of the same characters. LLI 10

. A A 2 3 4 5 6 N content of wheat% N content of maize,%

Mexico Internet Mexico Internat. n= 100 1765 n. 147 1958 60 0..100 .133 .135 ±5..029 .084 xs= .045 .088 .009 min.= .011 .008 max = .209 1.023 0 max...342 1.657

4,- 0. 40

.005 .01 .02 .04 .08 36 .32 64 1,213 .005 .01 .02 .04 .06 .16 .32 64 1.28 N in wheat soils, % N in maize soils % Fig. 202. Nitrogen, Mexico.

6.4.2 Macronutrients

The total nitrogen contents of Mexican soils are relatively low from the international viewpoint (Fig. 202, Appendixes 2 and 3). The N contents of both wheat and maize vary widely. This is apparently due to greatly varying rates of nitrogen fertilizer application 197

.0 40 Fig. 203. Phosphorus, Mexico. Mexico Interftat. Mexicoiqernat 8= 100 1765 35 1.147 1560 R..38 .38 6,31 .33 ts..17 .12 92.-OR r31113. .18 .05 30 7117 .7.13 .05 R,85.1.02 1.02 mol.53 104

05 F7

6 .7 It 9 P content o wheat,./0 reoize,%

MaXiC.0 I 618=731. 131521775 SS ' 00 1765 Mexico 5. 147 20.2 n.147 1867 35 -295 04.7 7=114 22.8 = 2.4 65.17.0 33.0 7185, 271.4 27.4 =8.8. 2-1 01 30 30 rnur .90.4 556.2 C

0>- 20 DiS

Oli

ve.8 2.5 5 10 23 40 80 160 320 640 os 0 70 42 00 500 020 540 P in wheat soils, mg/t P in'maize soils, mg/I

(wheat: 122 ± 54 and maize: 40 ± 49 kg N/ha). The average rate of N applied to wheat is one of the highest recorded in this study and seems to be the main reason for the high N contents of wheat. For example, all wheat samples with a N content less than 3.6 percent had received no N fertilizer at all while those with N6.0 % were fertilized with 100 to 153 kg N/ha. While the N contents of wheat were high those of maize are low, and when the plant N contents are pooled (Fig. 6) the Mexican average equalled the international mean. The phosphorus_content of Mexican wheat soils is relatively low while the P content of wheat is at an 'average international level (Fig. 203). In both cases the range of variation is wide. The P contents of maize and maize soils are at an average international level but also vary widely. The sampled wheat crops were fertilized with phosphates at somewhat higher ra tes than the maize (9 ± 13 and 4 ± 7 kg P/ha, respectively). For the international rating of P in Mexico, see Fig. 7. The national mean contents of potassium in wheat and wheat soils are among the highest in this study (Fig. 204, Appendix 2). The mean K content of maize soils is also high but that of maize itself is below the international average. Relatively wide variations were typical of all K values. Practically no potassium fertilizer was applied to either of the sampled crops. A good response to potassium fertilizer could only be expected at the lowest soil and maize K levels. See also Fig. 8. The content of exchangeable calcium in Mexican soils, especially in wheat soils, is high (Fig. 205). In wheat the Ca content is, on average, at the normal international level but the maize san-ipleg contain relatively little Ca. In all cases there are wide variations. The magnesium content of Mexican soils is generally high but again with rather wide 198

35 MexicoInternat Mexico Intermit Fig. 204. Potassium, Mexico. n= 100 1765 n= 147 - 1967 30 0= 4.7 4.0 0 = 2.7 3.1 In. 0.7 1.0 no 0.7 1.0' min. 2.9 0.9 min.= 1.0 0.6 Lo, 25"°° 6.3 6.8 25 max =4.3 6.7

it'rt

2 3 5 6 3 4 K content of wheat% K content of matze,%

Mexico Internat MexicoInternat n= 100 1765 n. 147 1967 9= 693 365 = 548 330 0 Os= 318 283 os= 338 356 rran.= 168 20 min. 51 18 max =2019 2097 max=1918 5598 o

' 25 50 100 200 400 800 1600 3200 6400 25 50 100 200 400 8001600 3200 5400 K in wheat soils, mg/I K in maize soils, mg/I

4 MexicoInternat. Mex ico Internat. Fig. 205. Calcium, Mexico. n=100 /765 n= 147 1967 0..40 .43 0= .38 .47 09= .13 .17 as= .14 .20 fl min.= .23 :11 30 min.=.16 .09 max.= .91 1.68 max.= 1.02 1.88

0_25 >- <_) 20 ai o15 E 1 o

z'T o Ca content of wheat,% Ca content of

Mexico Internat. Mexico Internat. n= 100 1765 n= 147 1967 35 2= 7023 4671 35 0= 4363 3450 as= 3020 3076 os= 2519 2815 min= 1970 110 min.= 290 10 max.=19080 21930 max.= 14880 17995

100 200 400 800 1600 3200 6400 12800 25600 50 100 200 400 BOO 1600 3200 6400 12807 Ca in wheat soils, mg/I Ca in maize soils, mg/I

variations (Fig. 206). The two indicator crops have almost equal mean Mg contents while the wheat Mg content is above the respective international average and maize below that. Some quite low Mg contents of maize are included. See also the position of Mexico in the "international Mg field" (Fig. 10). 199

40 40 Fig. 206. Magnesium, Mexico. Mexico Internat. Mexico internat. 1764 1967 35 n. 100 35 n= 147 9.185 .172 0=.186 .251 ts..037 .060 15=.051 .119 mine.118 .044 30 min.= .087 .036 max...307 .948 max.=.336 1.125

051- 20

1,1 D 15

CC 10

Mg content of wheat,% Mg content of maize, %

40 40 Mexico internat. Mexico Internet 1764 147 35 ns 100 n= 1967 R. 720 489 9= 651 446 ±s= 376 437 ±s= 381 462

min =151 10 min.= 25 1 3 11104, 1587 3298 5134.=1829 6490

A _ ! Ardiviri 25 50 100 200 400 800 1600 3200 12.5 26 50 100 200 40 80 100 3200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

6.4.3 Micronutrients

Boron. Owing to the heterogeneity of the Mexican soil material the B contents of soils and plants vary greatly from one sample to another (Figs 207 and 208). High B values dominate and in the "international B fields"- (Figs 22 and 25) Mexico stands high. The frequency of high B values (points in Zones IV and V) is more than twice as high as in the whole international material. There seem to be no distinct geographical areas of high B but high values were recorded from several states including Hidalgo, Mexico and Tlaxcala in the central part of the country, as well as Sonora, Sinaloa and Coahuila in the west and north. High B contents were relatively more common for Regosols, Yermosols, Fluvisols and Xerosols and plants growing in such soils than for other Mexican soils (see also Fig. B in Appendix 7). Some of the high B values may be due to B contained in irrigation water. Although only a few B sample pairs are within the two lowest B Zones and no extremely low B values were found, response to B fertilization may be obtained at several locations, especially if plants with a high B requirement are grown. Copper. Mexico stands close to the centre of the "international Cu fields" (Figs 27 and 29). In only one sample pair (46225) are the Cu contents exceptionally high but low values are more frequent, two of them (46249-50) being very low (Figs 209 and 210). Many of the low Cu values came from the western states of Jalisco, Nayarit and Sinaloa, but low values are also found in the central states. All four Arenosols included in the Mexican soil material were low in Cu as were many Andosols, Regosols and Fluvisols. Iron. Excluding Malta, where the number of samples was small, Mexico has the lowest 200 100B111,4"MienTaill=111=M11=11: Fig. 207. Regression of B content of 80 mommusson pot-grown wheat (y) on hot water 60onsiikargrznriummarain 50 soluble soil B (x), Mexico. For details - -12°iMESEVE111111111111 111111111111111 of summarized 40 OD +2.6 MENEM internationalback- 30ME MEBill ground data, see Chapter 4. a 20IEEE MEIN inecomi o ..,..iiiiiliiiiiiiiii...,=...... ,...... ,...... smEMMILemsomM-20ffilminommiii...... 11111111111MWfillEAMMIIMIMINSIMIIII IIIIHMIgetiliiiiii MEMUn

2111111.1101111 Emma

11111 1 IIÌii .2 .3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 B in soil, mg/l (i-iot w sol.)

100 Fig. 208. Regression of B content of 80 m I1 r4rm i. . 1E= .....170 B . pot-grown wheat (y) on CEC-corrected 60 bwra Bra 50IIMMINEEDICIEI INN (hot water soluble) soil B (x), Mexico. 40MIIMMEMBEZ 111191 E . a30IIIIIINEES MEN 111611IP 0_ 0111r ° 814 iiiiiiiiii Juilal NE Illiltraill II . ..lipMEEMIRMEMITHIPIMEIMEIllenttaRRIIMMEMIIIII Ill IN o 3 lip ,..., 0 ' VA'1EL., co r" i I : 2nid Elk101111 lbill INN hilil la HIM 081 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 CEC -corrected B in soil, mg/1 (Hot ..sol.)

national mean values for soil and plant Fe (Fig. 31). More than 25 % of the Fe values fall in the lowest Fe Zone and almost half of the Mexican material falls in the two lowest Zones (I and II, Fig. 211). Even the Fe values within Zone III are predominantly clustered toward the lower boundary. It seems likely that in many places crops sensitive to Fe deficiency would respond to iron fertilization. Soils low in Fe seem to be more common in the northern states, Nuevo León, Coahuila, Chihuahua and Sonora than in the southern parts of the country. Low Fe values were relatively more common in Yermosols and Kastanozems, but also occurred in the Regosols, Xerosols and Fluvisols. Manganese. Mexico is close to the centre of the "international Mn field" (Figs 33 and 201

Fig. 209. Regression of Cu content of 22 pot-grown wheat (y) on acid ammonium 20 m x'' 'flten° i111111111 acetate-EDTA extractable soil Cu (x), illtilli_11 37)EMI Ill Mexico. E18 neY SL-16 11111111 NM III 114 11111 11111111 38 111111 Igor 1Hi En11111111111111111111111 4 1111111E21111111111111111

2111C11111=1111111 NWT

OIII 111111111NMAMI III , I C. 04A n i aSZ l( )n / rl amain Cu in soil, mg/l (AAAc- EDTA)

Fig.210.Regression of Cu content of pot-grown wheat (y) on organic 22 carbon-corrected (AAAc-EDTA extr.) 20MLeiXIMIIin nu II soil Cu (x), Mexico. 18 4gMINTaal IIII c9°Y5:3111111111°g°Y71.2'°NIII

11111 1 l il 10 Ik 8 ENE___11NAII 6 Insworrommo L. timmillowrmi mslioniiimunoANA!wp .4.6 .8 1.0 2 46 810 20 40 60100 Org.C-corrected soi CU,mg/1 (AAAc-EDTA)

37). Although the bulk of Mexican material lies within the "normal" range for Mn (Figs 212 and 213) the variation is wide and with both high and more frequently, low values being recorded. Over ten percent of the Mn values fall in the two lowest Mn Zones. The lowest Mn contents were from alkaline soils and plants grown on alkaline soils except for sample pairs No. 46157 and 46246 (co-ordinates in Fig. 212: plant Mn 45, soil Mn 3.6; and plant Mn 33, soil Mn 0.9) where the soil pH(CaC12) values were 6.27 and 5.40, respectively. These soils were exceptionally coarse textured, low in organic carbon and apparently low in total Mn. The highest Mn Zones (IV and V) include samples only from sites with low soil pH, usually below 5.0. The geographical distribution of low and high Mn sites is 202

3 00 Fig.211. Regression of Fe contents 2 75 ofpot-grownwheat(y)onacid 2 50Fe1"1111116.11"9111M161i1 225 NEMIIHIMIRMINIIIIII III ammonium acetate-EDTA extractable 200=1113011111atialiviiiiiiEll soil Fe (x), Mexico. 1 75LWElliiiiiiP5.111111iiiIMI E 1 501t1 IIIIIIMII 111111 I o_ RIMINI IllIIIIIII 1111 111111 I 90Ma Minn E 80 I IIIIIII I 70 1 Ulla ,31111M1 60=!IInill_.01.11111 50 .Vgaraiiiiiiiii

40MIVIIIIIIIINM - 1

1 30 MIMIIffili El MIUUM1111AllI 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/I (AAAc- EDTA)

obscure. The low Mn values may be more typical of the soils of Baja California, Sonora and Sinaloa and high Mn of the Mexico and Jalisco States. Molybdenum. As in case of other nutrients the contents of Mo in Mexican soils and plants vary considerably. As a whole, Mexico stands high in the "international Mo fields" (Figs 15, 20 and 21). Because of widely varying pH and texture in Mexican soils the relatively weak correlation between plant Mo and A0-0A extractable soil Mo is substantially improved by correction for pH and texture (Figs 214 and 215; Table 8). Almost without exception plants with low Mo content (<0.1 ppm) have been grown on soils with low pH combined with either very fine or very coarse texture. Correspondingly, plants with high Mo content (> 1.0 ppm) have grown on alkaline, medium textured soils (see also Figs 16 and 19 and related text). The highest Mo values were recorded from Coahuila and Sonora, often in samples from irrigated sites. Zinc. Regarding average Zn contents of both soils and plants, Mexico stands close to the centres of the "international Zn fields" (Figs 41 and 43) but samples vary widely in Zn content between low and high (Figs 216 and 217). About six percent of the Zn values lie in the two highest Zn Zones irrespective of the extraction method used, some of them relatively far into Zone V. However, low Zn values are more common. Depending on extraction method, 8 (DTPA) to 17 (AAAc-EDTA, pH-corr.) percent of the Zn values fall within the two lowest Zn Zones with a number of samples close to these. Although the most extreme values are lacking, shortage of Zn is likely in many Mexican soils. Most of the high Zn values were measured from the Hidalgo and Jalisco samples, while shortage of Zn may be more common of Queretaro, Sonora and Sinaloa than elsewhere. 203

Fig. 212. Regression of Mn content of 2000 =MMIOMMIL.n Ilmoiman pot-grown wheat (y) on DTPA extract-

able soil Mn (x), Mexico. 1000EN WL"V2/4.% WSWRIMMEWRY Eft% IIIMMIIMINI 800 MEMINIMMEN=11UMMIll 600111noNoimmosinmommossimmimwsNIIIMIIIHMINUMMIIMMIN1113111111= E4001111MINEIIIIIIIMINIMINI111== gi 300 .5) 20011=111111111111111111111ME

100..11111111110111111T41 ""GE, 80IIMMISSEMOLEimm5PmEalgarilmm& g 60nmmommorammormanumaisom C) c 40 30IIIMMIPM111112111111111111111111=11 20

1011111111MNIMMENNIIiiIIIIIIIIUIIflllIRI111111111111MINIMIMME11MIMMENII 8IIIM1111111111111NliaM1110117MIMM11 MWM N1" 2 3 46 810 20 30 40 60 80100200 300 Mn in soil, mg/l (DTPA)

2000 M n 1000 liallatill1E111111 800iiim:Nalmiairmaratzazur=rearemi 600II==IMMLI:1:12E21311EIT122MS.LUMEJLILINM=IMIMMIN E IIMINNEMETIZIMIIINWEIVAM111111=1=151111Ind 400 o3001 flh1iîiIII 200 11111F1' . o ol

10016 1.11. Q) 801111MNIM=.111111.11KillFZUMENIMMkMr"7.MIMMIIMMIWIEel'AMMEMMENEllMMMINEEN--- o 60 11.1 11UNIMIECZEBEEIMMEMINI IIW=IMICUal C.) IImoreniseraiirmimmminuimmemum 40ImmEscrairmorwamommulnmmmli 3011111MMEENOVIIIIIIIHNI1111=11111110 20ralushrommiammoili

Fig. 213. Regression of Mn content of 10 8 szimmmtmamsmmos iMM=ImaraN pot-grown wheat (y) on DTPA extract- NIMIMH=1=MINIMMENsalMI=MEMMENnillIMMIMINMEN able soil Mn corrected for pH (x), 2 3 46 810 20 30 40 60 100200 300600 Mexico. pH - correctedMnin soil, mg/t (DTPAI 204

Fig. 214. Regression of Mo content of M osism UI 611M11.153511111111M=MIIINeaf pot-grown wheat (y) on ammonium FATEILIMIEMI=MEN1191- mmion 1E1649 Emu 1- oxalate-oxalic acid extractable soil Mo 3 (x), Mexico. 2 Ian ilitaM1111 E ca 11111141111111 Q-1 ."'"EN"I MME:111 IMMarom .6 rEsolimenrolimaim 11111, 1101.1en ó 3

4,Ú .2 immonim

.1W.:=1:11:111.===1111irommininn .06WEMENIIIIMEIBINIIMMMIMMENI

.03 .02 IIIII 01 ll .01 .02 .03 .06 1 .2 .3 .6 1 23 Mo in soil, mg/ (.40-DA)

KA ==2:121:; 1-10g1=MESiimiMMI.elreirlromalnlan Wig.6=X 6MEMIIIIMMI=EIMMINIT IMIIMMERII MM. 1101111415CEMENIKTIMmLEIMIMINEMMOIN 0 3IIII I "9 Z11111111131312211111 111 2IllI n'58 Nii iiillatitiPillil . III

E D. II 1 11111M1111111,Pill Q. .E:1111====1111.E.V1.12111 o .6 MISIM911=786111111=Eall ILUMOPl MMENIIMI N121111111Mil BUM MIAMI -11119 el I 111111 4ill ''I i IIIMINI 14 III WO1111,E11 1111101 I= IMIIIMM 1,111.-.1===09:16=rass ni imommar= .06mrat.....11 m M1 .1.11111111101=111011111=m1111111.imima.ialoram.....ENom:1= .03 IIIIIIMMIMMIIIM .1111 .02 podMIINIIMERIIIIIIIU 1 Fig. 215. Regression of Mo content of .01101111-11,111.11111SEI- pot-grown wheat (y) on pH-corrected .006 .01 .02.03 .06 .1 .2.3 .6 23 A0-0Aextractable soilMo(x), pH-corrected MO in soil, mg/l (A0-0A) Mexico. 205

Fig. 216. Regression of Zn content of 200 NA:illin Iri:e t. r1=33 pot-grown wheat (y) on DTPA extract- Zn ' '; l'..5- n :119.r 11677 able soil Zn (x), Mexico. i pi .-4c 5 IcHy12: 132110g rO.3° ° 100 90 80 if rimm 11 ii 70 iralogiirdi E60

40\I MINIM./111 ..c ó 30 HillEFM% a) 11111111M111 20 ilmi N

10 I .4111VAIMENCIIIIIMMI=INArai..- IIMINi'M 9 IIPIWAI1INIIMINI0111111k IM 8 1AMINEMIIIIIIM EIIIIILWIL 7 1 6 I 1.1 IE Di 3r .2.3 .4.6 .81 23 46 810 20 3040 60 100 Znin soil, mg/ (DTPA)

n1 1 11 1 II 1 MOW1 luirr4iII g Ewe El I NMI IIIIIIMIPAI RIMPROMMI 111116111II 1 Naito,111.I 1rmmulmmnommum_ V111111 EIIIIIIVaingra Fig. 217. Regression of Zn content of MHO MAMIENM pot-grown wheat (y) on AAAc-EDTA Ind k a MOMadill extractable soil Zn corrected for pH . . . . (x), Mexico. pH-corrected Zn in soi , mg/I (AAAc-EDTA) 206 6.4.4 Micronutrient contents of original plants with special reference to varieties

Since the original Mexican wheat material consists only of high yielding varieties, comparisons with local varieties are not possible. Variety data were obtained for 71 of the 147 maize samples supplied, of which 24 were classified as HYV. This group consisted of varieties H 407, H 412, H 505 and H 517. The other variety group was more heterogeneous comprising seven different varieties. Since the tnajority (29 samples) of these were of a local variety, 'Criollo', comparison in Table 24 is made between the HYV and 'Criollo'. On average, the maize plants of the two variety groups were sampled at the same physiological age. Differences between these variety groups were found in their uptake of several micronutrients, but these were most marked for B, Mn and Zn. Although the HYV maize plants were grown on soils lower in extractable B, Mn and Zn than plants of cv. 'Criollo', their average B, Mn and Zn contents were considerably higher.

Table 24. Comparison of micronutrient contents between high yielding maize varieties and local variety 'Criollo', respective soils and wheat (cv. 'Apu') grown on the same soils in pots. Differences between the mean contents of the two groups followed by the same index letters (aa) are not statistically significant. Letters ab indicate significant differences at 5, ac at 1 and ad at 0.1 percent level.

Micro- Average Micronutrient Contents of: nutrient original maize (ppm) resp. soils (mg/1)1) wheats grown on HYV Criollo HYV Criollo same soils in pots (n = 24) (n -=- 29) (n = 24) (n = 29) (n = 24) (n = 29) Boron 17.9a 7.8d 0.87a 1.01a 7.2a 7.0e Manganese 76a 62' 9a 66d 53a 150d Zinc 41a 28a 0.6e 4.4d 14a 20b

I) CEC-corrected hot water sol. B; pH-corrected DTPA extr. Mn; DTPA extr. Zn.

When the different maize varieties were replaced with one variety of wheat, the pla-nt and soil data for the three micronutrients were in relatively good accordance. Although factors other than the genetic differences between the varieties, such as irrigation practices, may have affected the above results obtained for original maize, it would seem that much of the apparent contradiction between the results of plant and soil analyses is due to plant variety and that the HYV maize plants are more efficient at absorbing the micronutrients in question than the local variety, 'Criollo'. See also Section 1.2.4.

6.4.5 Summary

Great variations in general soil characteristics are typical of the Mexican soil material in this study. These variations are also reflected in the macronutrient contents of soils and plants. With regard to micronutrients the concentration of Fe is exceptionally low from the international point of view. Although Cu, Mn and Zn values are generally somewhat low and those of B and Mo tend to be high, high Mn and Zn and low Mo and B contents were frequently recorded. At many locations responses to Fe and Zn fertilization would seem most likely. 207 6.5 Peru

6.5.1 General

Most of the sampling sites in Peru (Fig. 218) are located in the agriculturally important Sierra and Ceja Selva regions, where the annual rainfall varies from 200 to over 1000 mm. In many cases the sampled fields were irrigated.

46966 -46H:9453 96456,_)

96537;46540 4S'

46545 -96550 96525-asszs. bosze

4'...)L-."-'41:rs'524 46527, 46530

P E U

Fig. 218. Sampling sites in Peru (points = wheat fields, triangles -= maize fields). 208

Peru Internet Peru Internat. Fig.219.Frequency distributions of n= 70 3764 on. 70 3783 0o36 44 R. 6.86 6.64 texture, pH, organic carbon content, 30 ts = 14 16 es .85 1.12 min = 9 9' 35 non = 4.00 3.62 and cation exchange capacity in soils of oti'rm. 25 = 69 92 max = 7.70 8.55 30 Peru (columns). Curves show the inter- if)

20 25 nationalfrequencyofthe same >- L.) 20 characters. 15

o 15 LU CC 0

10 20 30 0 50 60 70 80 90 TEXTURE INDEX pH (Ca Cl2

45 Peru Internet Peru Internet 79 40 n= 70 3779 n= 3777 1.3 R=25.9 27.) to=.7 1.2 os. 10.6 14.5 C 35 min= .3 min =13.5 .2 max=3.4 39.1 max .66.1 99.9 30

.2 .< . 16 32 6.0 12.8 25.6 16 32 ORGANICC,% CEC, me/100g

The textural variation of Peruvian soils is wide (Fig. 219), with coarse soils dominating though many soils with a high texture index are included. The majority of soils are alkaline or moderately acid but the material encompasses a few samples with low pH as well. The organic matter content, cation exchange capacity, electrical conductivity and CaCO3 equivalent values are at an average international level, with fairly wide variations (Fig. 219, Appendixes 2-4).

6.5.2 Macronutrients

The mean contents of total nitrogen in Peruvian soils as well as N contents of maize are somewhat above the respective international averages but N contents of original wheats are low (Fig. 220). On average the N status is close to the international mean (Fig. 6). The sampled wheats were fertilized with relatively low rates of nitrogen fertilizers (average 38 ± 35 kg/ha; international average 66 ± 61 kg N/ha, Appendix 5). Wheats with a N content of less than 2.7 % N had not received any nitrogen fertilizers. The comparatively higher N contents of maize may be due to more abundant applications of nitrogen fertilizers (average 46 ± 58 kg N/ha; international average 27 ± 37 kg N/ha). The phosphorus contents of maize and maize soils are at an average international level but those of wheat and wheat soils are somewhat lower (Fig. 221). In general, tbe average P content of Peruvian soils and crops corresponds closely to the average international level' (Fig. 7). Only low to moderate rates of phosphates were applied to the sampled crops 209

40 Figs 220-224. Frequency distributions Peru p Internat. Peru internat n. 13 ri,1765 n.59 1958 of nitrogen, phosphorus, potassium, 5.2.54 R. 4.27 9. 3.24 914 es= .62 Is. 1.15 Bs. .55 87 calcium and magnesium inoriginal 20 min.. 1.84 rnin.s 66 1.90 88 57 651 wheat and maize samples and respect- max, 4.24 r max.. 745 max, 5.33 ive soils (columns) of Peru. Curves show the international frequency of the same characters.

J 4 5 2 3 4 5 N content of whect,% N content of moize,%

70 Peru Internot PerU Internat

Ea ns 13 1765 n.57 1358 Es .144 133 0.174 .135 8.5..056 .684 77..076 .088 .009 053 .008 50 min.064 nO max.-.285 1.023 mm4-.420 1.657

'ir]. 40 40 o>- Z 30 40 171 o w 20 ce u.

.0 .92 .04 .08 .16 . 405 .01 .02 .04 .08 .16 .32 1 8 Fig. 220. Nitrogen, Peru. N in wheat soils, % N in maize soils,%

<01- Fig. 221. Phosphorus, Peru, I Perti Internet. Peru Inte,nat. 35 -. n. 13 1765 n 57 1567 6..30 .36 5..33 .33 As ...OS .12 As.05 .10 §30 .65 o .95 rnoxs49 ,i55: 1.02 max.- .51 1.94

2 .7 8 P con ent of wheat,°/0 P con ent of maize r Peru Internat. Per u I Iter not. 35 1765 n., 57 1567 it r 19 6 70.2 4= 20.9 27.5 As. 8.2 24.7 is. 15.8 33.0 min.2.8 to min.= 2.4 max.. 27.4 271.4 30 ma8 70.8 656.0

5 IL 40 43 BO 150 320 640 2,5 S 10 29 40 BO 150 326 540 P in wheat soils. mg/I P in maize soils, mg/l

(maize: 8 ± 13 kg P/ha and wheat: 10 ± 13 kg P/ha). A quarter of the Peruvian sampled wheat and maize crops had a P content of less than 0.25 percent. With only two exceptions, these crops had not been fertilized with phosphates. The generally low contents of potassium in Peruvian soils is reflected in the K contents 210 Fig. 222 Potassium, Peru. Peru Internet Peru Internal' 5= 13 1765 xi.57 1967 30 0=2.9 4.0 30 0= 2.9 3.1 as=0.8 1.0 its= 1.3 1.0 min.=1.4 0.9 min.= 0.9 0.6 max.= 4.7 6.7 fo., 25 max.=4.3 6.8 25

20 20

2 3 4 5 6 1 2 3 K content of wheat% K content of maize,%

Peru Internat, Peru Internet, n=13 1765 n= 57 1967 R. 178 365 Os 231 330 356 30 as. 65 283 as= 144 min.=78 20 mm. 56 18 max,. 273 2097 max.= 777 5598

o- 20

25 50 I 0 200 400 800 1500 3200 5400 25 50 100 200 00 800 1500200 5400 K in wheat soils. mg/I K in maize soils. mg/I

40 40 Peru Peru Internot. Fig. 223. Calcium, Peru. 1967 5 n. 13 1765 _n. 57 R. .33 .43 0..51 .47 .12 .17 is=.20 .20 .21 .09 30 min =.16 .11 30_ max ..49 1.66 mox.= 1.02 1.138 -0.25 25 - C, 20 20

IS ci0 15 Ff 10 P 5 5

.8 1.6 .e 1.6 Ca content of wheat, % Cu content of maize,%

Peru Internet Peru Internot. n. 13 1765 n. 57 1967 5 0= 4067 4671 35 7= 4556 3450 as. 2016 3076 as. 2491 2815 min= 1220 110 min.= 280 10 max. 6490 21930 max.= 10840 17995 3

?j 25

100 200 400 600 1600 3200 6400 12800 25600 0 100 200 400 800 1600 3200 6400 12800 Co in wheat soils, mg/I Co in ma ze soils. mg/I

of plants, especially in the case of wheat (Fig. 222). Rates of applied potassium fertilizer were relatively low (wheat: 12 ± 19 and maize: 9 ± 17 kg K/ha). All sampled crops with low K content (< 2.4 %) were grown on low-K soils (< 200 mg/1) and/or received no potassium fertilizer. See also Fig. 8. 211

Fig. 224. Magnesium, Peru. PeruInternat. Peru Internat. n=13 1764 3 n=57 1967 0..131 .172 0=241 .251 its = .045 .060 25..126 .119 30 min.= .085 .044 30 min.= .093 .036 max.= .194 .948 max.= .723 1.125 25

20 15 A r 0

5 rd-,A A 5 .2 .05 Mg content of wheat, % Mg content of maize,%

PeruInternat. PeruInternat. n= 13 1764 35 3 n= 57 1967 0=387 489 R= 324 446 15.265 437 es= 188 462 min..133 10 min.=49 1 max=1073 3298 max, 909 6490

ú25 25

25 50 100 200 400 B 01600 3200 12.5 25 50 100 206 400 800 1500 3200 5400 Mg in wheat soils, mg/I Mg in maize soils. mg/I

The calcium and magnesium contents of maize and maize soils were at an average international level or slightly aboyé it, while wheat and wheat soils were somewhat lower in these elements (Figs 223 and 224). Consequently, Peru lies close to the centres of the "international Ca and Mg fields" (Figs 9 and 10). 212

IIIIM II 10 INIMIWOMMIINIMMINIIIIIM1111111111111110111011110Illillia Fig. 225. Regression of B content of 8 :iel1=01111111111104iTildNIMlEhiMIZIONMINIMilaInVI11.=MillianiiiMII MO IIIIM=MIONINIIMMIIIMIEI EMWMt(WMMRNIIIIIIIIMII=MIIIIIIIIIMIIIIIIIIII pot-grown wheat(y) on hot water 6MINIMERMIMINE2eZ 0111111111111111M11111 NM soluble soil B (x), Peru. For details 5 "5 ±' ' 4NEI - MUNN IN EMU of summarizedinternationalback- E311119111ERWEVII Opal ground data, see Chapter 4. ci '.'° 111111 9: 2IIIII ANN101 ci LIR a, 11 .1111.1II I 1 "6 ...m==rogiBv_aamrians::: lummamommr....mihnummammimuirm....umormooMMIEMIL19 solumplopmanimi . lam unomam unn motal" 1 I 11111111 1111110 2 rim NMI INN _2 .3.4 .5.6 .810 2 3 56 81o B in soil, moil (Hot w. sol.)

100 Fig.226. Regression of B content of nod 1--.... 'm-mr7rmr1:1====ragrara" 80 D, wommlimminimummiamomonlommilm pot-grown wheat (y) on CFC-corrected Eirai.- Nirk.wInmammommEnallIIIIIIIIIIPAI 60 (hot water soluble) soil B (x), Peru. 50munizireswrannommummumusim IN -.±° 6 namainisemipmplim E a_ 3011111 .-4591pliNIMIIIMpi11ed111111 0. 20II `-°67' liiiii11=1111111111 III a) _c 44I 10 h.ha 111111.4"irIII Ill ci 8 IslikNmainiato.r-ihvairmairamplimram1 NIN 6 E IM a) 5 111.10F.IVZ411111NIM ilum. 4E' ululi o 4 morMman I Mil IO 2III lik IIIII mail

111,111hi.21b.. Eril .081 .2.3 .4.5.6 .810 2 3 456 8 10 C EC -corrected B in soli, mgil (Hot sol.)

6.5.3 Micronutrients

Peru is usually located close to the centres of the "international micronutrient fields" (Figs 15, 20, 22, 25, 27, 29, 31, 33, 37,41 and 43). Although none of the boron values measured in the sample material from Peru showed very low B contents (Figs 225 and 226), several of them are low enough to indicate the possibility of B deficiency, if crops with high B requirement are grown (see also Section 2.7). Excess of B seems unlikely. 213 22 Fig. 227. Regression of Cu content of P ru Inte no pot-grown wheat (y) on acid ammonium 20 acetate-EDTA extractable soil Cu (x), 18 111111i11111111111111 Peru. E 8:1611111117111111:1A11112111101111111 40o

15. 1211111111111111111.11111111 VI°111§101111111111111111211111111 88Iliii111111111111111rd111111111 (-) 611111111111111111E1111111111111111 411111111101MB11111111M11111111 211.11.31iiifilLI'111111111MM111111 11111111111111111MM111111111 .4.6 .810 2 4 6810 20 40 6080 Cu in soil, mg/l (AAAc- EDTA

Fig. 228. Regression of Cu content of 2 CU i pot-grownwheat(y) onorganic Peru I Intern. t carbon-corrected (AAAc-EDTA extr.) 2Dn. soil Cu (x), Peru. y±s..7 3± .01

r0= 568 k r=07 l''' i 6...,1111111 II 111111

24 III gill.g A : MIN i Mal 1111111 6 INIEntrigiliMillini 4 IIIIMEME11111E11111111 2 elliMal 1111 MR1111 0 IIIIMIERMIIIIMIEb 1 .2 .4.6 810 2 46 810 70 ¿n sn 100 Org.C-corrected soi CU,mgil AAAc-EDTA)

For the copper, iron and manganese values recorded from the sample material, the ranges of variation are relatively narrow (Figs 227-231), and only a few sample pairs fall outside the "normal" range (Zone III), and seldom at extreme locations within Zones I or V. Only at one site (No 46546 in Tarma) were very low Cu values measured and quite high plant and soil Mn contents were recorded at three sites (46549, -528 and -565). Two of the latter originated from Tarma in acid soils with pH(CaC12) 4.00 and 5.28 and one from Paucartambo, soil pH 4.27. 214

3 00 Fig. 229. Regression of Fe contents of 2 75 pot-grown wheat (y) on acid ammonium 2 50 225 acetate-EDTA extractable soil Fe (x), 200MIIIMINZI11111121E11111111M1 Peru. 175 E 1 50 a_ 97 1 25 o .e 100 90IIIIIIIIIMMIE1111111=1111 80 iiUUUIIIAiiIIIIIIP 70 o 60MM111111111111511111111111 a) u_ 50

40Pa10111111111E11.1

30111111111111111111

10 20 40 60 100 200400 600 1000 2000 Fe in soil, mg/l (AAA c- EDTA

The contents of molybdenum were more variable (Figs 232 and 233). In addition to four relatively high plant and soil Mo values, one especially low one was recorded. This originated from the same acid soil in Tarma (46549) as the highest Mn value. Also other low Mo values were measured from samples of high Mn content. Although none of the zinc values falls in the lowest Zn Zone, many of them show quite low Zn contents of both soils and plants (Figs 234 and 235) indicating the possibility of Zn shortage (see Section 2.7). The majority of Zn values are normal but in samples from a couple of sites originating in San Lorenzo, Janja, quite high Zn contents were measured.

6.5.4 Summary

Most of the Peruvian soils are coarse textured, slightly alkaline, have medium organic matter content and medium cation exchange capacity. The macronutrient situation is variable, with low to medium contents dominating. For micronutrients normal values are typical. However, the limited data give some indications of possible B, Mo and Zn shortages. In a few samples Mn, Mo and Zn contents were high. 215

Fig. 230. Regression of Mn content of 2000 ,=.1=1, MOI=1. i MINO MP s ilmmiNt Ma= almi pot-grown wheat (y) on DTPA extract- 1i91 8L311511/1111111.111 able soil Mn (x), Peru. 1000 II=ke^M'Flll"JAKWINEMIq1VXMRMI=MIMMS=n1M11=g-IMMI,===== 800 1 To Il11lTrIMPI W11=11:MIIMMISIMMIIIIMINIIIMIIIMIIII=17MWIM47.117171111111111=IK 600II =.1 enmmamoisimommi 400iiImm 111MMIRI1111=M Q.300ommoIIIMMISIIIIIMIIIIME 200MUNI111.11111111MMI 1111011111111M12 10011E11 MEN -,%1MIEKIEM am:11Nrou MaiMIPIIUCAMMENN0.6Bi=P,'NCMWf:IMlIlfS MIIIIMIlalleall.11.1111 1.WW. '60 60III1=ME nrialliNIOTIAM1.19.111 111=M c4011=111 7411119121EIMIIII MM 30IIMIMI liiiiMEMENiiiiii 20III/21 111=111111111M111

10liEll 111E11111111E1111111111011=MMEMI101=1=MMEMIE M 8. .WMEN Mi grnianagiTIMMENIEN I111 MENI NMliTaILIIMMIIM ii 2 34 6810 2030 40 60 80100200300 Mn in soil, mg/l (DTPA)

2000 Mn111111111111111111111111111 1000so magmenacEuia.wiarm:remmuremorimmaremmjapsaggiumImmos....6...... mmmisimmor 800 =1MMEM11Wl=WEIMM11 1==1101111111 IrraliFiNER411EITIrriFUVIIgNiZTENIIMIMMII E600 Q. o_40011=MIIIMIIIIMME111111111/.11MMillIll o300limmommilmossmormpialum 200 o 100=MME1135MENTPM=MMEEM1111111111111111111611111111 .UMEM 80 o 60uimlommiimimmulinmail=mmil 4011MMEMEDINIZIPMEIN11111111=k1=111 kr 2 30 20

Fig. 231. Regression of Mn content of 10 pot-grown wheat (y) on DTPA extract- 8 111=IMMIIIflrIMMENNI=1MIUMI 1=1=M71111 able soil Mn corrected for pH (x), 2 3 46 810 20 30 40 60 100 200 300 600 Peru. pH - corrected Mn in soil, mg/l (DTPA) 216

Fig. 232. Regression of Mo content of 6 pot-grown wheat (y) on ammonium IMMATINEIAGEMCIZEC 31 C 27 oxalate-oxalic acid extractable soil Mo 3 y-,wow 3 ".6 log (x), Peru. 2 III had m..1.-MMEENNE!1WM=MI MENNIIMMEEMIDr.

C .2 "C"o .1Ifl11100BRIMENN1111 trirmmuimmmi11111/MIMMENMIMCIMMIIMMENIMIMIMINIIMIT/S. .06 NO00111=1191 I .03 .02

I 11 III .01

.01 .02 .03 .06 .1 .2 .3 .6 1 23 Mo in soil, mg/t (A0-0A)

e wmIn mummommmma. M 0lildtbilEMIkarlalliGIME5111011 I MIMI 611116mminnin.353741011 ii MIIIIIIIRMENIIIIMECIEEMEMIII=ESSEEE s - Ogg 0 VIIMILLTMEnallni6 3 MI 2111111121611111111111M11111.1i1 11

E eia 1 1111111.11111111M1111=11 --MEMONommwsism ...... ---.....----.....---.ancion-mi =MENNEN ..m...... 6 =...... ialumodn i. .c MENEM =IMMENIII shismilimm IIIIIIMMIIIIIIIIII'AMMINIMMIMIIIIIMMIIIII M111111 1111111=1111111iM11111111WI

111101111111111111111101...... , 11==MIIIIM111.2MIMMEII kW= MIMMENNIMMINIMO.1MIMMENNE0MEMEN NKM 2 MMUMENUMniiiMMEMMOI MEMEN Mt. .0611111111M11WIMMENIIMMMI111 MN MIIMEWANOMEMUlli /Min .03 .02 I Fig. 233. Regression of Mo content of .01Piiir liiiiiii IIII .006 .01 .02 .03 .06 . .2.3 .6 1 23 pot-grown wheat (y) on pH-corrected pH-corrected MO in soil,mg/l (40-0A) A0-0A extractable soil Mo (x), Peru. 217

Fig. 234. Regression of Zn content of 200 t pot-grown wheat (y) on DTPA extract- Zn able soil Zn (x), Peru. y=18. 8 r..7 .2 100 90 80 70 g_ 60 cL50 o .c04

10 9 8 7 6

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zr) in soil,mg /I (DTPA)

100 11111111111 90 80 70LIENIMEMENIELIBSIN 60muonlimmusuimmorann E 50monommolomMINI111111 2- 40 o cl) 30Komminizramili o 11111111.1111111111111111 :7;20 o IIÍiO!!ijIIlÌiiIIIIllI

10 1 9"AMER 1111MMENIIIIMMEMI 8MIMENIIIII IMEN11111=11111111111 Fig. 235. Regression of Zn content of 7 6 16, 11H pot-grown wheat (y) on AAAc-EDTA usa k VA el extractable soil Zn corrected for pH .2.3 .4.6 .8 1 23 46 81020 3040 60 100 (x), Peru. pH-correctedZn insoi ,mg/l (AA Ac- EDTA)

219 7. Far East

7.1 India

7.1.1 General

Soil and original wheat and maize samples received from India for this study came from five States: Haryana (100 sample pairs), Punjab (70), Bihar (50), Delhi (45) and Madhya Pradesh (30). The approximate sampling sites are given in Figure 236. Of the 160 soils

171NALA

9905 99100 58 45650

O 460S1-46095 45851-95965 95891-45900 95066-95890

Fig. 236. Sampling sites in India (points = wheat fields, triangles = maize fields). In Madhya Pradesh area only the last three numerals of each sample pair number (45551-45580) are given. 220

45 IndiaInternet India Internet Fig.237. Frequency distributions of n=295 3764 40 n= 295 3783 5= 34 44 0=7.55 6.64 texture, pH, organic carbon content, 5s=12 16 es= .50 1.12 In. 15 9 35 non =5.14 3.62 and cation exchange capacity in soils of non. 73 92 max, 8.55 8.56 30 India (columns). Curves show the inter-

20 25 nationalfrequencyofthesame characters. 20 15

0 20 30 0 SO 60 90 TEXTURE INDEX pH (CaCl21

45 IndiaInternat. IndiaInternat. 3779 n= 294 3777 0 n= 294 7= .5 1.3 = 16.6 27.3 es= .2 1.2 ±s= 13.3 14.6 't35 non = .2 ,1 non =5.1 .2 max =1.3 39.1 max =76.5 999 30 Q. o25 uj20

C5 Lyi 15

. 1.6 3.2 6.4 12.0 256 ORGANICC.% CEC, me/100g classified according to the system adopted by FAO, 55 were Cambisols located mostly in the Punjab, 40 Fluvisols (Bihar and Delhi), 34 Luvisols (Bihar) and 28 Vertisols (Madhya Pradesh). Three Haryana soils were classified as Kastanozems. Texturally the Indian soils of this study are coarse (Fig. 237). The national mean texture index is among the lowest (Appendixes 2-4). The soils of Madhya Pradesh, however, differ distinctly from other Indian soils. All 30 of them are fine textured (TI = 49 - 73). Equally fine textures (TI 49 and 54) were measured only in two other Indian soils, both from Haryana. Almost all Indian soils are alkaline, contain little organic matter and have low cation exchange capacity. The national mean value for soil pH is one of the highest and those for organic carbon content and CEC among the lowest in this study. With regard to CEC, the Madhya Pradesh soils differ again from the others; all have a CEC of 30.0 me/100 g or more and other Indian soils from 5.1 to 26.0 me/100 g. Electrical conductivity values of Indian soils (Appendixes 2-4) vary considerably but are generally well below the international average. With the exception of 25 soils from Bihar (45851-75) the CaCO3 equivalent values of Indian soils are relatively low. The sodium contents are usually of the average international level.

7.1.2 Macronutrients The Indian national mean values for total soil nitrogen in both wheat soils and maize soils are the lowest recorded in this study (Appendixes 2-4, Fig. 6). The ranges of variation are also exceptionally narrow (Fig. 238) and only in three Indian soil samples N values exceeding the international mean (0.133 %) were recorded. Unlike soil N, the N contents of original wheat and maize samples vary widely but are usually low. In part, the wide variation is apparently due to largely varying rates of nitrogen fertilizer applications (wheat: 221

40 40 Figs 238-242. Frequency distributions India internot, India Internat.

n= 188 1765 5 n= 107 1958 of nitrogen, phosphorus, potassium, 0.331 4.27 5 = 2.62 3.14 os= 1.51 1.15 os= .62 .87 calcium and magnesium inoriginal i 30 min.= .60 .60 30 min.= 1.57 .88 max.= 6.14 7.45 max.= 4.34 6.51 wheat and maize samples and respect- ',I325 25 ive soils (columns) of India. Curves 20 20 show the international frequency of W 15 the same characters. o 'kJ 10 10 U-

2 3 4o 6 2 o4 5 6 N content of wheat,% N content of maize,%

70 IridiaInternat. Indiainternot. n= 188 1765 n= 107 1958 80 8.571 .133 60 R= .073 .135 ts=.021 .084 os= .017 .088 rojo.. .009 .009 min.= .042 .008 ?), 50max.= .142 1.023 50 max...128 1.657

o_40 40

C.) Z 30 30 LU

.02 .0 08 .16 .31 1.28 .005 .01 .02 .04 .08 .16 .32 .64 1.28 Fig. 238. Nitrogen, India. N in wheat soils, °A N in maize soils,%

India internat. India internat. n = 107 1967 35 n.100 1765 = .32 .38 = .31 .33 os = .13 .12 to = .07 .10 min.= .05 .05 mine .16 .05 max.= .61 1.02 max, .48 1.04 ai 25 >- Z

P content of wheat, % P content of maize,%

o India Internat. India internot. n. 107 1967 35 n= 188 1765 R= 93 20.2 =8.6 22.5 os= 8.1 247 05=7.1 33.0 min.= 1.2 1.0 min.= 1.9 0.1 max.= 54.0 271.4 max.= 46.2 656.0

2.5 5 10 20 40 BO 160 320 640 25 5 10 20 40 80 160 320 640 Fig. 239. Phosphorus, India. P in wheat soils, mg/I P in maize soils, mg/I 75 ± 51 and maize: 106 ± 57 kg N/ha). In the case of wheat the effects of varieties might seem to be very clear. Nine out of ten wheats with a N content less than 2.0 percent were classified as local varieties, but the majority of wheats having a N content over 2.0 percent are HYV wheats. For more details, see Section 7.1.4. In the case ofmaize the varietal effects are less clear, the average N contents of HYV and other varieties were 2.92 % and 2.48 %, respectively. Less than ten percent of Indian soils have a phosphorus content exceeding the international mean (Fig. 239) and the Indian national mean values for soil P are among the very 222

35 Ob rodio Internat irodioateraat. Fig. 240. Potassium, India. n=188 1769 n137 1957 30 9. 3.8 4.0 30 9.23 Es, 1.3 1.0 ts. (Q OQ Ima= 6.9 0.9 ram= 1.0 0 S 25 max= 6.2 6.8 25 X. 4.3 67

ctr,

C'20 0 7- u/ 15 C/ 1.61

Lr_

3 8 3 6 K content of wheat,% K content of maize,%,

India Internat. Indio iniernot rir 188 1765 n= 107 1987 9. 186 365 0. 126 333 95, 119 283 Es. 78 355 min =. 42 20 Ian. 41 13 mor 856 2097 max= 917 5593

25 50 103 23400 040 1600 3260 4400 25 50 100 210 400 NU1500 3240 6400 K '01 wheat soils, mg/I K in maize so.ls, rno lowest (Appendixes 2-4, Fig. 7). There seems to be no distinct difference in the P contents of soils from the five States. Although relatively heavy dressings of phosphateswere applied to the sampled Indian crops (wheat: 15L.:15 and maize: 18 ± 9 kg P/ha, Appendix 5) the P contents of both indicator plants (especially wheat) are low. There is no clear difference in the P contents of maize between the two variety groups but in thecase of wheat it seems to be substantial. See Section 7.1.4. The average potassium status of Indian soils is internationally low (Figs 240 and 8). Low soil K values were most frequently measured from the samples of Biliar and Punjab but were uncommon in the Madhya Pradesh soils. In general, the K contents of maize are low, but those of wheat are at the average international level. In both crops, especially in wheats, the variations of K contents are exceptionally wide. A part of thismay be due to varying applications of potagsium fertilizers to the sampled crops (wheat: 11 ± 20 and maize: 9 ± 19 kg Kl ha). Another source of variation seems to lie in plant varieties. Again the HYV wheats contain considerably more K than the local varieties (see Section 7.1.4). In India, as in many other developing countries, attention should be paid to preventing the present low potassium content of the soils falling any further. Especially when higher yields are sought by using improved varieties and heavier applications of nitrogen, it must be realized that (owing to the high K contents of crops) considerable amounts of K are removed from soils. If the removed K is not replaced by fertilizer or manure, the K status of soil is likely to become critically low in a relatively short period of time. This is more obvious in areas where the soils are coarse textured with.insignificant K reserves. In spite of the generally high soil pH, the exchangeable calcium contents of most Indian soils remain at or below the international average (Figs 241 and 9). High Ca contents, however, were recorded from almost all Madhya Pradesh soils which, different from other Indian soils, had fine texture and high CEC (see Figs 12 and 14 in Section 2.2.4). On average, the Ca contents of original Indian wheats are internatiorjally slightly on the low 223

40 Fig. 241. Calcium, India. India Internat. IndiaInternat. n=188 1765 35 n. 107 1967 A= .36 .43 A..56 .47 *6..16 .17 as= .26 .20 min...11 .11 30 min.= .19 .09 max.= .93 1.68 mox..1.16 1.88 0.2 25 >- 6.3 20 20

1.1.1 15 15 co

10 0 airzyj;

.2 .4 .8 1.6 Ca content of wheat, % Ca content of maize,%

India Internat. India Internat. n 188 1765 n. 107 1967 it=3530 4671 .2660 3450 80 =2795 3076 as 1221 2815 min.. 380 110 min.. 470 10 max.= 12040 21930 max.= 5330 17995

rïÌ

.4 4 100 200 400 BOO 1600 3200 6400 12 00 25600 50 100 200 400 800 1600 3200 6400 12800 Ca in whea soils mg/I Ca in maize soils, mg/I

0 40 Fig. 242. Magnesium, India. Indiainternoi India Internat. n= 107 1967 35 n= 188 1764 35 0..187 .172 = .313 .251 as = .072 .060 85=133 .119 830 min =052 .041. 30 min.= .057 .036 max.= .407 .948 m06....823 1.125 a 25 25

5 4 Mg content of wheat,% Mg content of maize,% 40 India Internat. IndiaInternat. n= 107 1967 3 n= 188 1764 35 R.= 448 489 8.269 446 5s= 383 437 as = 134 462 min.. 49 10 mm. 36 1 30 max, 3298 3298 max =600 6490

25 2

25 50 100 200 400 13 0 1600 3200 12.5 25 50 100 200 41,1 8001600 3200 6400 Mg in wheat soils mg/I Mg in maize soils. mg/I

side but those of maize somewhat higher. For both crops the variations in Ca contents are exceptionally wide. Part of it is obviously due to different ability of different plant varieties to absorb Ca from soil (see Section 7.1.4). In general the magnesium contents of Indian soils and plants are good but vary widely from one site to another (Figs 242 and 10). The lowest exchangeable Mg contents were 224

10 rtIMIIMMIII11011111101111.11111111Ilarimmersimemarsar.i.,,,,,7 illimmarememumirmisiimusimama Fig. 243. Regression of B content of 8B.,,,,,..miammoommiimmimiummommmmumwmga pot-grown wheat (y) on hot water 6 n= 8 IIINIMEEMNIMIMIIMIMI11==MINNEU ; WM- 606NMENIEM1111====M111111111 soluble soil B (x), India. For details of 5 =0 42 - MIZEIEIIII 4;MIN Mai summarized international background 3 iIiiii: 40+ 6 3IINTIRME11 Fill data, see Chapter 4.

9.- 2Mil8" RANI IIIIIIIIII o -c0.) .11111111 11111 1 1!II__II...... A...... "E5 ,...... ,...... MIIIIIBRIMIRINIEWSIMI11114111P.1i111111 1111111 EINOPUeeligliPANCIISMMIM1111111111111 MEMMIEMIginaillE MNEM o filiniMag. 11111 iN Minn CE1 11111 EIIIIIII I1Ih 111 lilt 14 III 1 . .2 .3 .4 .5.6 .810 2 3 4 56 81 B in soil, migil (Hot w. sol.)

100 Fig. 244. Regression of B content of 80 n. 8 3 pot-grown wheat (y) on CEC-corrected 60 ±s. 6 6 2 75 69 . (hot water soluble) soil B (x), India. 50 2±s06 060 -± 2 40 y3633 9x Y' 5 0.a30 r= 0 828 '` 0 2

0810 G)56 " *. . o 1 o 3 2

.08.1 .2.3 .4.5.6 .81.0 23 4 56 810 CEC -corrected B in soil, mg /I (Hot w. soi.) most frequently recorded from Bihar soils and the highest from Madhya Pradesh soils. For effects of plant varieties see Section 7.1.4.

7.1.3 Micronutrients

Boron. The average B content of wheat grown in pots of Indian soils corresponds closely to the international mean. Due to the generally low CEC of the soils, correction of soil B values for CEC also brings the Indian average for soil B close to that of the whole international material (Figs 243, 244, 22 and 25). Because of wide variation, both high and low B values occur. The frequency of high B values (Zones IV and V) is about 10 percent and corresponds to that in the whole material, but low values (Zones I and II) are more than 225

Fig. 245. Regression of Cu content of 22 pot-grown wheat (y) on acid ammonium 20Elliii1111111113111111111111111111111111 50 acetate-EDTA extractable soil Cu (x), 34 18IRON 11111 5 3E11111111111111111111 India. E a-16MEMO wEll313111111111111111N1 o w 11111111111111111111111111111111111 .c14 .61211111111111111111111111111111H1 1:10111111111111=111111111,111M1111111 88 AIIII11111111,111111=1111111111

6 41111111MINSIIIME11111111

01 .2 .4 .6 .81.0 2 4 6 810 20 40 6080 Cu in soil, mg/1 MAAc- EDTA)

Fig. 246. Regression of Cu content of 22Cu pot-grownwheat(y) onorganic inch i Inter.t. ' carbon-corrected (AAAc-EDTA extr.) 20 soil Cu (x), India. 189is=°.*.P INEMIIIII MIN 'Y282111E1 ogoY7MN 1111/1

111111111111 10 'I; . ..1111011111111111 111111rTiliiiiiM111111 111011111111ME111111 1111111=111111111 =IF r ftrinimi ' 14 oi .4 .6 .810 2 46 810 20 40 60 100 Org.C-corrected soi CU,rng/1 (AAAc-pTA)

twice as frequent. Although low B values occur in other States, most of these originate in Bihar, where four-fifths of the plant and soil B values fall in Zones I and II. Even the remaining Bihar B values are comparatively low. If soil B values 0.3-0.5 mg/1 are considered as a critical limit for B deficiency (see Section 2.7) almost half of Indian soils can be suspected of various degrees, from hidden to severe, of B deficiency. According to the present data, response to boron fertilization could be expected especially in Bihar and is least likely in Delhi. The highest B values were recorded from the Punjab and Delhi but only a few of them reach a critically high level. The above data are in relatively good agreement with those presented by Kanwar and Randhawa (1974). Copper. India's location in the "international Cu field" is central (Figs 27 and 29). Only a few Cu sample pairs fall outside the "normal" Cu range (Figs 245 and246) and only in one sample pair (45828) from Haryana were extremely high Cu contents measured. 226

300 Fig. 247. Regression of Fe contents of 2 75 pot-grown wheat (y) on acid ammonium 2 50Fe IIIME NMI 225IRS1 111111FER MIIIII acetate-EDTA extractable soil Fe (x), 200liel IiiiiiMNM India. 1 75EfiBIR111111111119;i P 111111 E 1 50 0011111131111111111111

c): 1 25i%ilIIIIIIIiLIIIIIIH o IIIIIMIIIIIII 100MI MI6111111 1,5 90milIIIIIIII -~c' 80WNWIIIIIONNIENIMIN 70 MEE i -M111111Er" o 60 MORIPMEMiliill u_ 50 trelignIMINIII

40Millikr.. 11111 111111 30II Mr lillnill iiiiMIIIIMAIIP 10 20 40 60100 200400 600 1000 2000 Fe in soil, mg/l MAAc- EDTAI

Iron. On average the Fe contents of Indian soils and plants correspond closely to the international mean values for Fe (Figs 31 and 247). The variation, however, is wide and ranges from low to high. About 12 percent of the Indian plant and soil sample pairs are within the two lowest Fe Zones, mainly in Zone II. Even those in Zone I are not in extreme positions. Judging by the present results, there is no severe Fe deficiency at sites where the Indian samples were taken but at some locations plants, sensitive to Fe deficiency, may respond to Fe fertilizer. Samples in the two highest Fe Zones represent about 15 percent of the Indian material. High Fe values were much more typical of Bihar than other States, while most of the low Fe contents were found in samples from the Punjab, Haryana and Delhi. Manganese. As in other countries with alkaline soils the Mn status in India is very low (Figs 33 and 37). Every seventh plant and soil sample pair falls in the lowest and every fourth in the two lowest Mn Zones (I and II, Figs 248 and 249). The ranges of variation of plant and soil Mn are exceptionally narrow and no high Mn values were found in the Indian material. Low Mn values (Zones I and II) are most frequent in Bihar and Delhi, quite common in the Punjab but rare in Haryana and Madhya Pradesh. Shortage of plant available Mn is apparent in many Indian soils. Molybdenum. Owing to the high pH of Indian soils Mo is usually readily available to plants. The average Mo content of wheat grown in pots on Indian soils is higher than the average for the whole international material in this study (Figs 250 and 251). The national mean for ammonium oxalate-oxalic acid extractable soil Mo is the sixth lowest among the 29 countries (Fig. 15). Correction of A0-0A Mo values for soil pH improves the plant Mo soil Mo correlation, raises the Indian mean above the international mean for soil 227

2000 11=MIMMINMEMIEMil=1.011 ammil Fig.248. Regression of Mn content Inda of pot-grown wheat (y) on DTPA n 2 extractable soil Mn (x), India. 1000 Y S1.10111:1111111111111 800 KTPT.M.swvrIMP-muivirirdreuEimiimm 600iismommenmmesinimsMN 1111711MiliNENIMWSIOMMICMIIIMIIIIMMI 400111MINIMMIIIIMME11111M 8:300 t; 200111111111111111111111111M1

-6 oo 801111!MIMEMINIIIVrall=111121111 8 60111=aliniaNCIMMUSEMENNON60 c 40 z 30iimmennigamilsomumm 20

10=11MMEMMI111111111111111111111111 8 Era IIIIMMINNIMMEMBON IMMINIMEMEN 2 3 46 8 10 20 30 40 60 80 00 200 300 Mn in soil, mg/l (DTPA)

Fig.249. Regression of Mn content 2000 ... .ommaim11 imMIIM.Vmve In of pot-grown wheat (y) on DTPA M n n 25 extractable soil Mn corrected for pH 1000 mmilipliwypesim.weempowm..r..=msIHIELINIIMI111111 (x), India. 800 ---MENIMMMI--INEMEM=1...111Mi-0E13Atiff.IMMENELIEEI,AVIM=MMIIN 600 WEUIL:1112111K101Etrilinall22.101EINIMMINEMI pr- 400 41 3009 200 uIIIi

100MIIIIMEGEM7.1111=1=MMEN 80111" MniiiililiiMrinIMI=MMEMEN21MMUEEIMaMMINIgaie=i9P:VINnEMENMMTE. 6011Ell11ILWIIIMMEMIOWI IIMICtUArinl=MMINIONIMilMMENNIINENNI=1111111 40Il ENSM1212151=MEMIIIIMIENIII 30II=1ffalliffilMMENIIIIIMMENIEN 20111Ni. :MTINONIMI111111=11111111

10IIPP- 11111111.1111111 8 III=1111111111111111111 MMIEN 8 M1111111/111111Wal=MMON =MMIITZE IN M1111111111111111111MMIEZIMMMII =IMMQ.III 23 46 810 20 30 40 60 100200 300600 pH - corrected Mn in soil, mg/I (DTPA)

Mo, and moves India from the sixth lowest to eleventh highest place in the "international Mo field" (Fig. 20). In other words, the apparent Mo status of India is substantially changed. The low Mo values in Zones I and II have almost disappeared and the number of high values in Zones IV and V is increased. According to these results problems of Mo deficiency in India are much less probable than those of excess. High Mo values occurred most frequently in samples from the Punjab and Haryana. In some cases where excess of Mo is combined with low Mn availability, it may be useful to apply manganese sulphate alone or mixed with ammonium sulphate or sulphur to obtain more lasting effects through lowering the soil pH. Zinc. In the "international Zn fields" (Figs 41 and 43), India occupies one of the very 228

rmmimasits 11.71M11111 =122;1111 Ihtfil INSIVEMMIMMIIIN Fig.250. Regression of Mo content M o MIONNIM 6 MMIMNISI of pot-grown wheat (y) on ammonium =1/1.11a2gEBEIZIMill10 oxalate-oxalic acid extractable soil Mo 3 (x), India. 2MEEDIRIBIa

.1111111111111Torm=smi Mtimmom =mma ximismiemmesumits11OWN ==1MMIHINIMINIZEWX1101=MMEMENOPMPIMillMESIMIN INIE1/11151.11PRIMII MIIMMIKEIMIZEMENINI MMON11111111111111.- o o AFIa=1211111====111. /MIIMIIIIMIIIIMENNrairINIEMMIIIINNINMMIMMEENNI 06WMAIIIII1111511111111MMNIMMI rilM11=1111111111111MIMIMMIMIN .03 'I .02

.01 .01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/t (A0-0A)

III MEMIII1====ill: =5,1 Fig. 251. Regression of Mo content of 011111.11q1 NWIIII11 i4=WINIIIIM NR1 NM NO M o1111IMENNZIi=11111111111 IMINI 6IMMENNIIM=ENZEICEEEtiiMninIMMIMMIN pot-grown wheat (y) on pH-corrected A0-0A extractable soil Mo (x), India. 31111111mmi'63 °L23111 11MEMIIMM1111111 India. r =0.61 " 21111111ndlik" figliiiiiilliiIP

E 1 111111MININk111111111/111 =MME=1MMIENNIIINIEZIall.IMMINIMME11=maillIMMINMEX191..=== Magi rg =110111M MMINEIN =MMENNENEMEMPUMIIPITMI=1M -5.6MEMEMII MMEMEMMINVMMEMINEMMMiii.M. .c MEMIIII MOINIIIIMMX121113111=1= ó .3 .2i1111=MI 111111151,11111111111111MMI

O .1inz.zoiMIIIIMME1111111111111eradliur..... EalliniimMillii=EMENI MMI 06I9iingltMainallrillEMIÏIM1 .03 .02 INIIMIIIIIIIIIIIIIIIIIII

.01111111111111101111111111111 .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MO in soil, mg/1 (AO- OA)

lowest positions. Depending on the extraction method 18 (DTPA) to 28 (pH-corr. AAAc-EDTA) percent of the Indian Zn values fall in the two lowest Zn Zones (Figs 252 and 253) and even most of the rest are comparatively low. The analyses frequently indicate quite severe Zn deficiency. Only in one sample pair from Haryana (45828) were very high Zn contents measured. The frequency of low Zn values is highest in Bihar, where almost 229

Fig.252.Regression of Zn content 200INEZ of pot-grown wheat (y) on DTPA .77 extractable soil Zn (x), India. 5 321 100 90

T5 30

o20

10 9111111iirirMARINCIII=MMIMENIIII 8111111MINEINEMIIIIMEMEHIIIIIMINON 7primmolimplm=01.1mwm 6imgremonnimMENIIIIIIMMOI nffillIONVE11111= 1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Znin soil, mg/I (DTPA)

Fig. 253. Regression of Zn content of 20 I , pot-grown wheat (y) on AAAc-EDTA Zn extractable soil Zn corrected for pH (x), India. 1 MM HiS IMffiUtelillIIIMIIIMMINIIII11. III 111111 1 MENIMII oulm.....5. ...um IMIIIIIM1110111rMINI 1 1111111 11011i opium noumemoempina ' 1 11111 mannwowilail

3111111111 111111"01d11111

2 1111111 1111111P.di 11111 ..11"1 rock1IIIII _..... 1 amaioc41011111111111_1 IIIII ETWOOkinammumniimmounisIME11111 =111101111111. WRINlliUMENNINIERNIIMMNIILIMI1111. MINAWilli 1 NI in In 1 11 .2.3 .4.6 .8 1 23 46 81020 3040 60 100 pH-corrected Zn in soi , mg/I (AAAc-EDTA)

every second sample pair falls in the two lowest Zn Zones. Low Zn values occur also in Haryana, the Punjab and Madhya Pradesh but in Delhi soils the Zn content seems to be somewhat better. In general, Zn deficiency, acute or hidden, seems to be much more common in India than in most other countries. Many investigators have reported the response of various crops to Zn in India. 230 7.1.4 Nutrient contents of original Indian plants with reference to varieties and other factors

Since the differences in the nutrient contents between the high yielding varieties and local Indian varieties were considerably more pronounced in the case of wheat than in maize, the varietal differences among the original Indian wheats warrant discussion. Of 188 wheats sampled from India, 82 could be classified with relatively high certainty as HYV and 59 as local varieties. Neither group was homogeneous. With a few exceptions the group of HYV wheats consisted of cultivars HD-1553, WG-357, 'Kalyansona', N 4 and 'Sonalika'. Most of local varieties were identified in the "Field information forms" plainly as "Local". The average macronutrient contents of the two variety 'groups are compared in Table 25 where the analytical data of respective soils and fertilizer application are also given. These data indicate that the N, P, K, Ca and Mg contents of HYV wheats are 134, 109, 56, 49 and 68 percent higher than those of the local varieties, respectively. For all five macronutrients these differences are statistically highly significant. In the cases of N and P some of the differences can be accounted for by the larger amounts of N and P fertilizers applied to HYV wheats and, compared with the local varieties, the HYV wheats were grown on soils somewhat richer in P. The differences in B, Cu, Mn and Mo contents between the original HYV and local wheats were significant without similar (significant) differences in the respective soil data (Table 26). When wheat (cv. 'Apu') was grown in pots on the same soils the differences became statistically non-significant. According to the data in Tables 25 and 26 there seem to be considerable differences between the two variety groups in their ability to absorb nutrients, especially K, Ca, Mg, B, Cu, Mn and Mo from the soil. However, there are other factors accounting for these differences. The estimated yields of HYV wheats were about 3000 kg/ha and those of local varieties about 2400 kg/ha. On the other hand, the local varieties were sampled at considerably later stages of growth; the difference was 26 days on average. These two factors, as explained in Sections 2.5 and 1.2.2, affect the nutrient contents in opposite directions but the effects of physiological age are likely to be much more pronounced. If we could take into account both these factors, the differences between the nutrient contents of the two variety groups would be much less. On the basis of the above data and discussion the striking differences in the nutrient contents of the two variety groups obviously are partly due to genetic varietal differences, partly to fertilization (N and P), partly to soil nutrient content (P) and partly to differences in yield level and physiological age of plants at sampling. Not enough is known of the quantitative effects of these factors, obviously variable from one nutrient to another, to draw distinct conclusions on their relative efficacy. However, these data give further evidence of the fact that estimation of the nutrient status of soils founded on analyses of heterogeneous plant material may lead to misleading conclusions. 231

Table 25. Comparison of macronutrient contents of high yielding and local varieties of original Indian wheats, respective soils and fertilizer applications. Differences between the mean contents of the two groups followed by the same index letter are not statistically significant. Letters ab indicate significant differences at 5, ac at 1, and ad at 0.1 percent level.

Nutrient Average Macronutrient Content Average N, P and K Original wheat Wheat soils application % N %, others mg/1 kg/ha HYV Local HYV Local HYV Local (n = 82) (n = 59) (n = 82) (n = 59) (n = 82) (n = 59) Nitrogen 4.32a I.85d 0.071a 0.071a 100a 63a Phosphorus 0.399a 0.191d 10.8a 7.6b 26a 6d Potassium 4.60a 2.94d 177a 180a 11° 7a Calcium 0.453a 0.305d 3820a 3883a Magnesium 0.229a 0.136d 475a 451a

Table 26. Comparison of micronutrient contents of high yielding and local varieties of original Indian wheats, respective soils and pot-grown wheats]) grown on the same soils. For index letters indicating statistical differences, see caption for Table 25.

Micro- Average Micronutrient Contents nutrient Original wheat (ppm) Wheat soils2) (mg/1) Pot grown wheat (ppm) grown on the HYV Local HYV Local same soils (n = 68) (n = 57) (n = 68) (n -= 57) (n = 68) (n = 57) Boron 5.7a 3.4c 0.67a 0.49a 6.4a 54a Copper 11.0a 6.0d 6.2a 5.7a 8.2a ma Manganese 70a 44d 8.3a 9.7b 46a 48a Molybdenum 1.38a 0.65d 0.190a 0.204a 0.37a 0.35a

0 Compared to Table 25 the number of samples is smaller; see Section 1.2.6. 2) Extraction methods used: B: hot water + CEC correction Mn: DTPA -F pH correction Cu: AAAc-EDTA org. C correction Mo: A0-0A + pH correction

7.1.5 Summary

The Indian soils, except the fine textured high CEC soils of Madhya Pradesh, are usually coarse textured, alkaline, low in organic matter and have a low cation exchange capacity. The soils are generally low in N, P and K, but have satisfactory contents of exchangeable Ca and Mg. The most probable microelement deficiency problems are those of Zn, B and Mn occurring most likely in Bihar. Many soils of Haryana, Punjab and Madhya Pradesh are also short of Zn and B and those of Delhi low in Mn. Some relatively high B values were recorded in the Punjab and Delhi soils. Many soils of the Punjab and Haryana were low in Fe and high in Mo. Low Fe values were found especially in Delhi and high values in Bihar samples. Of the five States represented in this study, Bihar seems to have more and Delhi and Madhya Pradesh fewer micronutrient problems than the other States. 232 7.2 Republic of Korea

7.2.1 General

The approximate sources of the Korean sample material taken from 50 wheat and 50 maize fields are given in Figure 254. Soils classified as Cambisols (37 %) were the most common, followed by Acrisols (21), Gleysols (17) and Luvisols (12). Most of the Korean soils are coarse textured (Figs. 255). Only 14 soils out of 100 had a texture index of 50 or higher. With only a few exceptions the soils are acidic but the pH varies very widely, from 4.1 to 7.5. A medium content of organic matter and low cation exchange capacity are typical as are a low electric conductivity, low CaCO3 equivalent and low sodium contents (Appendixes 2-4).

1 T 2. In. 1

/ ,.....4

. 5 19L.

il :" '2:3%4760 43478 4343G GANG REUNG i 43439 . q SEOUL 0 ,,,,, e ...9,,°;" 4'4.4'4:4" WEi JU 16 SA 3401-06 ,,t, 43403-00 '.(35 ''°' 43409,3 :., 43484 T 7. 43438 -3D 43485 43457,34345%

a 34510 43453 4,. 043428-22 434. A ,,4,:'

.T419495 A 43456

DA? JEON . Ili

344 . ..-.AI 3343441 433-34 Ocfr GU

43445 43449-50

SA 43446- 48 , . Atiltt .. .., 1P. V 0,401

. .,._3100 PO 1 '`b -a:

--° O T.3 % ' C,,..,' le o "-Sot 0 0'k> <, -. REPUBLIC OF KOREA

0,o o 4 ..0 3 0 0 \, .o

I 127° TY.123. 12.0.

Fig. 254. Sampling sites in the Republic of Korea (points = wheat fields, triangles = maize fields). 233

45 Fig.255.Frequency distributions of KoreaInternat. Korea Internat. n=100 3764 40 n r.100 3783 texture, pH, organic carbon content, = 36 44 0=560 6.64 30 Ss. 11 16 es=.92 1.12 and cation exchange capacity in soils of min.= 18 9 35 min.= 4.10 3.62 rnax.. 56 92 max.= 7.50 8.56 Korea (columns).Curves show the 30 international frequency of the same 25 characters. 20

0 20 30 40 50 60 70 80 90 TEXTURE INDEX

45 Korea Internat. KoreaInternat. n.100 3779 n 100 3777 4= 1.2 1.3 . 17.4 27.3 = .4 1.2 no =4.7 14.6 -a'35 .1 min.. 6.7 .2 max.= 2.6 39.1 max .36.0 99.9 30

0>,_ 25

u_ 0

2 .4 .8 1.6 3,2 6.4 12.8 256 e 16 32 ORGANICC,% CEC, me/100g

Figs 256-260. Frequency distributions Korea Internat. Korea Internat. 3= 50 3 1765 n= 50 1958 of nitrogen, phosphorus, potassium, 31= 5.27 4.27 7= 3,95 114 ss. 1.07 1.15 = .41 .87 calcium and magnesium inoriginal min.= 2.99 .60 30 min.= 2.90 .88 wheat and maize samples and respect- max= 7.09 7.45 max.= 4.69 6.51 ive soils (columns) of Korea. Curves show the international frequency of the same characters.

2 3 4 5 6 2 3 4 5 N content of wheat,% N content of maize,%

Korea Internet Korea Internat. 60 n= 50 1765 60 n= 50 1958 8=115 .133 0= .144 .135 1,5= 041 .084 ta. .043 .088 min.= .035 .009 50 min...041 .008 S' max.= .270 1.023 max.= .231 1.657

_47.48810 444 .005 .01 02 0 .08 36 .32 .54 028 .005 01 .02 .04 .08 .16 .32 Lc 1.28 Fig. 256. Nitrogen, Korea. N in wheat soils, % N in maize soits,%

7.2.2 Macronutrients

The contents of total nitrogen in Korean soils are at an average international level but in both original indicator plants the N contents are high (Figs 6 and 256). The average N content of Korean wheat (5.27 %) is the second highest (after Hungary) recorded in this study, but the variation range is wide. Both sampled crops were fertilized with high, but 234

Korea Internet. Korea Internat. Fig. 257. Phosphorus, Korea. 35 n= 50 1765 n= 50 1967 5.43 .38 = .37 .33 Z .6..13 .12 38. .08 .10 830 min.=.20 .05 0 min.= .21 .05 max ..713 1.02 max.= .51 1.04

6 7 8 9 P content of wheat% P content of maize,%

Korea Internet Korea Internot. n= 50 1765 5 n=50 1967 R= 45.4 20.2 =44.0 22.5 = 28.2 24.7 no=32.2 33.0 min.a6.7 1.0 min.=5.0 0.1 30 nao =116.3 271.4 30 mox =140.0 656.0

2.5 5 10 20 40 BO 160320 640 2.5 5 10 20 40 80 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I

Korea Internat. KoreaInternat. Fig. 258. Potassium, Korea. = SO 1765 no 50 1967 30 8= 4.0 4.0 30 R = 3.4 3.1 850.7 1.0 o 0.8 1.0 min 2.2 0.9 min.= 1.9 0.6 8 25 max.= 5.8 6.8 25 max.= 5.3 6.7 o>-

coUi

6 2 3 4 5 K content of wheat % K content of maize,%

Korea Internat. Korea Internat. /0=50 1765 n = 50 1967 6=131 365 0= 190 330 30 as= 48 283 0 f5 104 356 min.=45 20 min, 52 18 max, 256 2097 mats.= 492 5598 25

[77).. 20 o

Ui 15 o aLoi 10 u_

25 50 100 200 400 800 1600 32006400 25 50 100 200 400 BOO 1600 32006400 K in wheat soils, mg/I K in maize soils, mg/I varying doses of nitrogen (wheat: 100 ± 25 and maize: 107 ± 31 kg N/ha, Appendix 5) which is probably the reason for the high N contents of the crops. On average, the phosphorus content of Korean soils is twice as high as the average for the 30 countries of this study (Figs 7 and 257). The P contents of Korean wheat and maize are also high. The crops were fertilized relatively generously with phosphate fertilizers 235

Fig. 259. Calcium, Korea. Korea Internat. Korea interne. 35 n. 50 1765 n50 1967 8.39 .43 9.44 .47 10..09 .17 *0,11 .20 30 min.= .24 .11 min.= .21 .09 max.= .59 1.68 max.= .78 1.813

.6 .4 Ca content of wheat, % Co content of maize.%

Korea Internat. Korea Internal n= 50 n= 1765 = 50 1967 35 R=1068 x=4671 5 R=1391 3450 ts = 312 xs=3076 ts=831 2815 min. 400 min.=110 min.= 300 ID max= 2220 max=21930 max.= 4000 17995 30 3

"O' ':> 25

Ul 15

cc 10

100 200 41, BOO 1600 3200 6400 12800 25600 50 100 200 640012800 Ca in wheat soils, mg/I CO in maize soils, mg/I

40 Fig. 260. Magnesium, Korea. KoreaInternat. Korea Internat. n= 50 1764 35 n= 50 1967 8.184 .172 = .270 .251 as= .048 .060 as= .085 .119 min...108 .044 30 min.= .147 .036 max.= .314 .948 max.= .471 1.125 25

ID raj/1h. 5 .2 5 Mg content of wheat.% Mg content' at maize, %

KoreaInternat KoreaInternat. 1764 rm 50 35 n= 50 1967 R= 194 489 R= 262 446 as= 81 437 Xs= 266 462 min.= 48 18 min.= 42 1 30 max.= 357 3291 max = 1832 6490

25 50 100 200 400 000 1600 3200 12.5 25 50 100 200 400 BOO 1600 3200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

(wheat: 31 ± 7 and maize: 35 ± 14 kg P/ha). Korean soils are low in exchangeable potassium (Figs 8 and 258) but the K content of plants are at an average international level (wheat) or higher (maize). Obviously, this is due to comparatively high potassium fertilization (wheat: 58 ± 10 and maize: 77 ± 80 kg K/ha). See also Fig. 11 and related text in Section 2.2.4. Korean soils are generally low in exchangeable calcium and magnesium (Figs 9, 10, 259 236 and 260) but the Ca and Mg contents of plants are at an average international level. More than one-third of the soils included in this study were limed and on two-thirds farmyard manure was applied during the past few years before sampling. These practices have possibly affected the macronutrient contents of the sampled Korean crops. See also the effect of texture and CEC on the uptake of Ca and Mg, Section 2.2.4.

7.2.3 Micronutrients

Boron. The average B status of Korea is slightly low in an international context (Figs 22 and 25). Despite many B values falling in the low B Zones only a few of these indicate

100 11111111 MOM =WOW WS MI 111MOM NM BIM B710 tioliatz3 PAM C "iim 11111 IN IN 1111111111111ME 111111 IN NM 80 MEM=110111111111141111111 NM ill II MI NM 60 TE=ER151111111111111111MINIIIMINIENUI 50minusarommrtealisammummismill 401 437+ 0011/11E211111111111111111MMINIti 1119,27 +4.0 ah, 30 1111111V111111111111111113111 0_ a." 20 o _c 10JiI_111111.11111111 "68 6IMINIMMINCEINP7.4111 -(1) 5MeiniMilegarginin 4 ,mitaillit11111 co 3 211311111 Fig. 261. Regression of B content of pot-grown wheat(y) on hot water III soluble soil B (x), Korea. For details .3 .4 .5.6 .8 1D 2 3 4 56 810 ofsummarizedinternationalback- B in soil, mg/l (Hot w. sol.) ground data, see Chapter 4.

100 80 n.9 60 ±s.4 148 6. 4. 50 O35 40 E.30 .323+3 5x 0. r. 0 771* 0 2 20

.a., 56 oC 4 3 co 2

Fig. 262. Regression of B content of .08.1 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 pot-grown wheat (y) on CEC-corrected C EC - corrected Bin soil, mg /l (Hot w. sol.) (hot water soluble) soil B (x), Korea. 237 severe B deficiency (Figs 261 and 262). It is apparent that boron fertilization would be beneficial on many low B soils, especially on crops with a high B requirement such as root crops, legumes, fruits and vegetables. Response of rape to B has been reported, e.g. by Hong (1972) on a soil with 0.15 ppm available B. Copper. Korea's position in the "international Cu fields" is almost central (Figs 27 and 29). Normal Cu values dominate in the Korean material and only a few fall outside Zone III, none of them showing extreme Cu contents (Fig. 263 and 264). The analytical data from the present material indicate no serious deficiency or toxicity problems in Korea.

20Alliiiiii1111112111111111111111111111

18 ri,11111111111111111111 E cc1:161101; ECIESE:11211111111111N1 o Q)140/1111111111111111111111111/111

12 NIIIIIIIMM111111111111111011111 110 38'44111111111111111111111111111 6

4 iNIMS111111111113111111 Fig. 263.Regression of Cu content of 2 pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), 1111111111111111611111111111 0:1 .2 .4 .6 .8 10 2 4 6 810 20 40 6080 Korea. Cu in soil, mg/i (AAAc- EDTA)

)2Icilire mlIIIinerrsillilliniiiiii Y±s"601 Ill-2,-..=glIN1110 MIMI lIllbviiiiiiiill 111111111 , II 1Ill 111111 MIIIIIII

1111111 IIIIIIIM1111111 111111 ..11114E10111111 ) 1111111 ime um ' innummummmonium noirorisi11111111111111 Fig. 264.Regression of Cu content of OilindfflE111III MIMI pot-grownwheat (y)onorganic MIIIIIIINSIELMAIIIIIIIII carbon-corrected (AAAc-EDTA extr.) î .2 4.6 B 10 2 L. 68 in 20 ¿n An inn soil Cu (x), Korea. Org.C-corrected soiCU,mg/1 (AAAc-EDTA) 238

300 Fig.265. Regression of Fe contents 2 75ignearliriIMMINIETMII 2 50Mi2 1111111111 MI ofpot-grownwheat(y)onacid 225 iffiNENHIMMTOMINIII ammonium acetate-EDTA extractable 200= smanionimain soil Fe (x), Korea. 1 75IN 09,13811" E 1 50 ci c1-125 o 100 .690 80 11i"01111111 70 iiiUP o 60 M11111111111211111 u_ 50 1111111111111111*

40Elliliiiiiiiii' Ili

...11111 30uhIiiiOMIMI 'El MI111241P 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/I MAAcEDTA)

Iron. The situation of Fe in Korean soils and plants is very similar to that of Cu (Figs 31 and 265). Normal values dominate and in spite of a few within Zones I and V, none of them is in an extreme position. Manganese. Korea stands clearly on the high side in the "international Mn fields" (Figs 33 and 37). Owing to substantial pH variation in Korean soils (Fig. 255) the availability of Mn to plants varies, and consequently, the variation range of Mn contents of plants is exceptionally wide (Figs 266 and 267). Correction of DTPA extractable soil Mn values for pH raises the plant Mn soil Mn correlation to a very high level (r = 0.935***) and more than triples the standard deviation of soil Mn. Both extremely high and low Mn values are included in the Korean material. The highest values come from geographically various locations in the country from sites of low soil pH(CaC12), usually 4.1-5.0. Correspond- ingly, the lowest Mn contents were measured in samples from sites where the soil pH was close to 7. See also Figure 34. Evidently, correction of Mn anomalies in several locations would be beneficial, in low Mn sites by using fertilizer containing Mn and in high Mn sites by raising the soil pH through liming. Molybdenum. The Mo status of Korean soils and plants appears normal (Figs 268 and 269). The varying soil pH affects its availability to plants and the variation range of Mo contents of plants is relatively wide. Correction of A0-0A extractable soil Mo values for pH widens their variation and improves the correlation. However, the material shows neither low nor high extremes of Mo contents. In many cases low Mo contents are combined with high Mn contents (e.g. sample pairs 43436, -461, -463, -495), the common nominator being low soil pH. Zinc. Only two countries, Belgium and Malta, exceed Korea in the "international Zn 239

Fig.266. Regression of Mn content 2000 Emommiumom.,,,tii.m.m.=alwesamnammom. of pot-grown wheat (y) on DTPA 1000 iii0111111111111111 extractable soil Mn (x), Korea. STWVVICEItIrklMEN=RZWZIRWM==IIMOMI IIMENIMM 800 MMImms'Emermasawramemmm 600nNommunimmmmosinirmimiratariliammummislummeamummom E 400 o_ 30011=M11111111INIMMIIIIIIMMINE t 20011=111111111M11111111111E1

'15 10011E1111111111111111111111Y "c" 80...... n.w...... a) 11'1:12:1111 .71TIMMIZEINN .4E0 60ItsNmezza-lionsmimimmonommwmtimm=mmtumm.-.,.-ilimommozonams. 40IMMINIMIE01=5111111=111111111.1 30IIMIIMPAWN11101111111111111MMS 20manilowerillomil 10..ii111111111110.111111111111111...... 8NI IWNIMIllOialIMINITMINNEM...... M71M 111 IMIMMIMINOMIMMEMIIMIIIMMENII all= 1 2 3 46 8 10 20 301.0 60 80100 200 300 Mn in soil, mg/I (DTPA)

Fig.267. Regression of Mn content 2000 r,onem. mumps of pot-grown wheat(y) on DTPA extractable soil Mn corrected for pH 1000.. ..:,...... :...... -111.0! =MAIM 800 Maa.-S.'....MIIMIUWMIMMUMMEN=111111MINII titialifilimia&ilommweecmicamNIEN (x), Korea. I:5-NI WEEMMt0331ILVEITIMIIIMITITWiirAMMIIIIIII E600II1 ,MMINIIIIMILIMMIMMIIICMIMMIllnil 0- 400 300 200 o "2100 u...=ram=1.....momm.111111111111111h1111111 a) 80 mommi-woumonme=mnss 60 MINI/ÌIEIMMIIMINI11MICIONI 40 30 111111111MENIIIIIMIIEll 20 111111=11111111=11111111

10 1111111111111111111 8 mINEIMMERIIMMEINE IM=IIMMEN IM=MIELL.111=1Fria 1 2 3 46 810 20 30 40 60 100 200 300600 pH- corrected Mn in soil, mg/1 (orPA)

fields" scale (Figs 41 and 43). In spite of large variations in Korean soil and plant Zn contents there are only a few values which would indicate possible shortage of Zn (Figs 270 and 271). The Zn contents of about every fourth sample pair are so high that the samples fall in the two highest Zn Zones. Some of these are so extreme as to include the maximum plant Zn and soil Zn contents recorded for the entire international material submitted in this study. Most of the high Zn sites occur in the Suweon area on soils classified as Luvisols or Cambisols. According to the above results Zn deficiency is rare in Korea but at certain locations problems due to Zn excess may arise. 240

711/11.2111111 Fig. 268. Regression of Mo content of 00111:4Aleil 6 MISR pot-grown wheat (y) on ammonium mordalailMEEPRI oxalate-oxalic acid extractable soil Mo 395 1121J10111 3 (x), Korea. 2 Yo E Q. 111111111.1111111 asonmr--=M1111 Eimmnisummaiammin....uNammommi MMOIELORPIMMESOirn-uuuui itKIIIIIIA11111111116, IIMIZI=KAMMIIK11111=IMMII MWAEr.glMIMSMMIIMM==IMII'WAMMIMICWMIMMENIMINIIMME=MMENNI,MWAL. V .06MrWAIIIMMIUMBIIIMMIMMEMEINIIM NIINWAMINERMINIIMMINiMMIMIIMI OMMEMEMINEMOMMME .03ormmulmmomil .02

11 III .01 .01 .02 .03 .06 . .2 .3 .6 1 23 Mo in soil, mg/t (Ao-0A)

MMIMM111 Fig. 269. Regression of Mo content of M oMEACIMOIET:101aMEMEMBrainTril AMMINIMMINI 6 pot-grown wheat (y) on pH-corrected n=90 A0-0A extractable soil Mo (x), Korea. Rts=0.127±,.2 7 mc_ .11=7. 3 log y=-1.232 0.4:91o. h 09 Y- '9 r =0.623 '' 2

E o. 1'difinir==rimalsi=1mtaltinPr MMEEMBI=M=INIGNIMII=IIMMONSI `63 .6 UUuIIIUUIHIII "8 .3INEEMMENHOIMPAinni 2.2 o o 1NIMMIMMOINO=.%=WAINIMINEMMIIMMMINIMME111.11111111111111111111 IIMMOIMOMMIIIMUSIMEIMMENI=1M/MBENINIIIMMEMOIAINIIMPIIIIMMENE11=1=MMINEMI MEMIENIIIIMINIMMITINIUMMENIIMIMIMMENNO .06MEWIlnin 112.19NEHOMIMIMMEMEMil 11/121111MMEMIONIIIIIMEMMINI11111 .03 Pir .02

Ph. 111 .01IIUNMIN=MMI =11.Melli=1IL. mwm I= .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH - corrected MO in soil, mg/t (A0-0A)

7.2.4 Summary The sampled Korean soils are generally coarse in texture, severely to moderately acid, medium in organic niatter content and low in cation exchange capacity. From an international point of view the macronutrient status of Korean soils is relatively low, with the exception of P. The level of macronutrient fertilizer application is relatively high and in addition two-thirds of the soils had received FYM and one-third were limed. 241

Fig. 270.Regression of Zn content of 200 In'en t pot-grown wheat (y) on DTPA extract- o'33, Zn 2 . 7ts1 ! 77 able soil Zn (x), Korea. 25 91 3:s1 75 r-C 34og x Icqy .12 321 log r.03

80 70 E60 13- 50 o 40

o20 o N lo 9 8 7 6 UI

1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Znin soil,mg/1 (DTPA)

200 Fig. 271.Regression of Zn content of Ore3 Irte pot-grown wheat (y) on AAAc-EDTA Zn 9C 353 s.2D.) 183 extractable soil Zn corrected for pH s8 St (x), Korea. c y.1 100 90 1 80 70 60 E 50 40 o 30 o 20 'c7; o

10 9 8 7 6

.2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soi ,mg/1 (AAAc-EDTA)

Consequently, the macronutrient contents of plants were at an average international level or above. Of the micronutrients the levels of Cu, Fe and also Mo are quite normal. In case of B low values dominate and Mn values vary from very low to very high. Low Zn values were rare but some were extremely high. The most likely micronutrient problems in Korea are those of Mn and B deficiency and Mn excess. 242 7.3 Nepal

7.3.1 General The sample material received from Nepal represents the Narayani, Seti, Mahakali, Gandaki, Bagmati, Lumbini, Bheri, Koshi, Mechi and Sagarmatha areas. A map indicating the sampling sites is not available. The great majority of Nepalese soil samples are texturally on the coarse side (Fig. 272). Soil pH varies greatly but the extremes, very acid and very alkaline, are excluded. The organic matter contents of soils as well as their cation exchange capacities are generally low to medium with a relatively narrow range of variation. Values for electrical conductivity, CaCO3.equivalent and sodium content are low for the most part (Appendixes 2 and 4).

45 5 Nepal Internat. Nepal Internat. Fig.272.Frequency distributions of n=50 3764 n= 50 3783 R = 37 44 =6.16 6.64 texture, pH, organic carbon content, 30 ts = 9 16 as=.91 1.12 min.:. 23 9 35 min.= 4.50 3.62 and cation exchange capacity in soils max.= 57v, 92 max.= 7.59 8.56 25 3 of Nepal (columns). Curves show the international frequency of the same 20 25 2- characters. W 15 20 o 162 10 ir V

10 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca Clz)

Nepal Internat. NepalInternat. 3779 n= 50 3777 40 n. 50 =1.0 1.3 =19.7 27.3 30 es=.4 1.2 =5.7 14.6 35 min.=.4 min.= 9.5 .2 o max.= 2.0 39.1 max.= 36.5 99.9 30

S15

LL" 10

.2 4 .8 1.6 3.2 6.4 12.825.6 8 16 32 ORGANICC,% CEC, me/100g

7.3.2 Macronutrients

The total contents of nitrogen in Nepalese soils and wheats are at an average international level (Figs 6 and 273). For soils, the range of variation is narrow, but wide for the wheats. The use of nitrogen fertilizer (68 ± 70 kg N/ha) corresponds to the average for wheats sampled in this study but the rates varied frorn 0 to 370 kg N/ha, which partly explains the greatly varying N contents of wheats. The ten HYV wheats included in the Nepalese material had higher N contents (average 4.79 %) than other wheats (4.04 %) but they were 243

40 Nepal Internat. Nepal Internat. nr 50 1765 n= 50 1765 Nepal Internat. 0=4.19 4.27 30 0= 3.7 4.0 n. 50 1765 =0.9 1.0 ±-s =1.07 1.15 35 0..40 .38 min.= 2.2 0.9 an ..11 .12 t 30 min..2.24 .60 Eli 25 max.= 6.5 6.8 O max.=6.34 7.45 min...22 .05 30 max...76 1.02 25 0-25 (>3 20 2- '23 20 LU 15 o 015 1-1 10 u_ 5 X'

A A 4 /:;:' ...411 A A z, .4 .4 A 2 3 6 3 .2 .3 .4 .5 2 3 4 5 6 N content of wheat,% P content of wheat,°/0 K content of wheat%

70 Nepal Internet. Nepal Internet Nepal Internat. 5 n. 50 1765 X. 21.6 20.2 n=50 1765 n= 50 1765 60 as. 26.6 24.7 5= 80 365 0= .133 min.. 1.2 1.0 48 283 30 non.. 121.4 271.4 no. ±6= .042 .084 min.=23 20 min.= .061 .009 0./ max.= 323 2097 O max.= .222 1.023 O

0.401-/ o>- Z 30 U.1 o 11.1 20 Q.

2.5 5 10 20 640 25 50 100 200 400 800 1600 32006400 005 .02 .04 .08 .16 .32 8 N in wheat soils, % P in wheat soils, mg/I K in wheat soils, mg/I

Fig. 273. Nitrogen, Nepal. Fig. 274. Phosphorus, Nepal. Fig. 275. P,otassium, Nepal.

Figs- 273-277. Frequency distributions of nitrogen, phosphorus, potassium, calcium and magnesium in original wheat samples and respective soils (columns) of Nepal. Curves show the international frequency of the same characters. grown on soils higher in total N (average 0.160 and 1.109 %, respectively) and were fertilized with larger dressings of N (average 122 and 54 kg N/ha, respectively). With regard to phosphorus, the limited sample material from Nepal very closely matches P conditions in the international material as a whole. The soil and plant P averages are about equal to their respective international averages and the ranges of variation are almost as wide as in the whole material (Figs 7 and 274). The phosphate fertilization level is somewhat lower and the rates vary slightly less (15 ± 20 kg P/ha) than those for the whole material (21 ± 27 kg P/ha). The extremely low potassium contents of Nepalese soils (Figs 8 and 275) are also reflected in the K -contents of original wheats, for which the national average is the sixth lowest recorded in this study (Appendix 2). It seems that the moderate dressings of applied potassium fertilizer (16 ± 20 kg K/ha) are sufficient to keep the K contents of plants from dropping very low. Nevertheless, the Nepalese soils are almost without exception deficient in K and a good response to potassium fertilization is likely. The levels of exchangeable calcium and magnesium in Nepalese soils are generally low but the Ca and Mg contents of wheats are at or above an average international level with fairly wide variations (Figs 9, 10, 276 and 277). These contradictions may partly be due to relatively high Ca and Mg uptake from coarse textured soils with a low CEC. See Fig. 272 and Section 2.2.4. 244

NEPAL Internat. Fig. 276. Calcium, Nepal. n 50 1765 35 = .48 .43 Is .13 .17 S'30 .11 max ..76 1.68 0.25 3- L) 20 Ui 0 15

1.1.1

Ca.content of wheaf,%

NEPAL Interna). 50 1765 5 5.1766 4671 as. 1173 3076 min.. 220 110 max.. 4660 21930 0

'13-1) 25

'CT)

100 200 400 800 1600 3200 6400 12800 25600 Co in wheat soils, mg/I

Nepal Internet 35 n=50 1764 0=106 .172 ±s =.059 .060 2330 min.= .109 .044 max...358 .948 o o >- 0 20

4-1 o15 L1.1 X 10

5 .2 .4 Mg content of wheat,%

Nepal Internat.

5 n= 50 1764 R. 205 489 ±s=111 437 min.: 38 10 max.= 458 3298

25 50 100 200 400 800 1600 3200 Mg in wheat soils, mg/I Fig. 277. Magnesium, Nepal. 245 7.3.3 Micronutrients

Because of a laboratory accident the micronutrient data of 15 samples of wheat grown in pots of Nepalese soil were lost. Boron. Nepal occupies the very lowest positions in the "international B fields" (Figs 22 and 25). About half of the Nepalese plant and soil samples are within the lowest B Zone and only a third lie in the "normal" range (Zone III), and these are close to its lower limit (Figs 278 and 279). On the basis of the above results, widespread B deficiency, acute or hidden, is likely to exist in Nepal, limiting yields especially of those crops with high B requirement.

Fig. 278. Regression of B content of 10 l'1101101MOMINI11011111.0011111111111111111111111111111111 ) Du..i':-t - warcriwii7,0summailwinememmosorinamo pot-grown wheat (y) on hot water 8 L./ Isemasounemmenuammussimmisnwelin soluble soil B (x), Nepal. For details of 6Misigm,E1111111-miraremsimi=M W summarized international background 5Ì 4)iilin- '019 i ,EMIZEINIIIIIIINE li rim data, see Chapter 4. )IIIIIIIIIIIMINIIIII 1E111 E 3E MIWWII Now 1011 LINalMMENIICOMPT11111=1/1MMINIIIMD...,_ 'lod' rmil 1 Sraink 11111111111IMMillimMILIMEMAINIIIIMMIlna=1:2010.2'Zil=7/111:: iiuiiIuuui1111111111tHIMEMIPME21111 =Email=13111111111111 iiipmirm. ii-"Nm ou ummenin IiiiiN ll 'Iliqiiii hi Rh". oh .3 .4.5.6 .81.0 2 3 456 8 11 B in soil, mg/I (Hof w. sot.)

100 80 B mommilmwarazazirazil --mi. 1=MIIINDIOHI4.----mmon.MmINIMMIUMI 60MIIIBITIN. ±Q96M *M.,mumumainumum 50 00 .11, A 40NII )7-ts 0 2 MOM e Iiii -.2 44 1 In /OWR a.a.3011111r . 0.577 Kim 20 Nib 4111 h i 1111 I 10smal-. .mmairal,w-A,rlil. 081 '6,1e-rulsraikm- 65 111110111/211111 11= I MMIPMINF A Ill MI o - c.) 'N M III M IL I co UP 2 EMI

IIi, Fig. 279. Regression of B content of III1111MMIN dd pot-grown wheat (y) on CEC-corrected .08.1 .2.3 .4 .5.6 .8 1.0 2 3 4 56 8 10 (hot water soluble) soil B (x), Nepal. GEC-corrected B in soil, mg/t Not w. 246 22 Fig. 280. Regression of Cu content of 20E1111111111111111111111111 pot-grown wheat (y) on acid ammonium 1111111111111W 111111111111111 acetate-EDTA extractable soil Cu (x), 18 Nepal. E 8:1 6111111211111:1:10u 1111101111111111 1141111111111111111.111111

.6 121111111111111111111111111111111111 111111111111111111C1111 1111111111111M11111111(1111/111 88 7 611151111111MININIMMIIIII

4111111111=1111111MM1111

211111111111MIIIIIIIMMIPI 11111111111111111:11106111111111 o:1 2 .4.6 .810 2 46 810 20 40 6080 Cu in soil, mg/I (AAAc- EDTA)

Fig. 281. Regression of Cu content of 2 Cu I pot-grown wheat (y) on organic carbon- Nep. 1 corrected (AAAc-EDTAextr.)soil Vernt 111 Cu (x), Nepal. gx -61 2:: 8121111117115111 1111

6: ii?g°Y51°111r poPtiollil ° MINI 68 II IIIILMAC.p/0111111.1111111 IIIMMAIpPr IIimmll1..u..11.,__nriirl 4 iIll PENIMMIC1 2 1 1`'S1111 o 11111 911.111311111Ed 2 L. 6 810 2111121".'L. 6810 20 40 60100 Org.C-corrected soi CU,mg/I (AAAc-EDTA)

Copper. The Cu contents of Nepalese soils seem to be quite satisfactory and Nepal stands relatively high in the "international Cu fields" (Figs 27 and 29). One sample pair falls in Zone II and only three in Zone V, the rest lie within the normal range (Figs 280 and 281). Since no extreme Cu contents were found in the sample material the present results indicate no serious Cu problems in Nepal. 247

Fig. 282. Regression of Fe contents of 300 275 pot-grown wheat (y) on acid ammonium 250 acetate-EDTA extractable soil Fe (x), 225 157 1 Nepal. 200 4 C 1C9 (cg 1 75 g_150

-690 80

IA) 70 o 60 u_ 50

30 \rz\ 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/I (AAAc-EDTA)

Iron. The mean Fe content of wheat grown in pots of Nepalese soils corresponds to the international mean but the mean for extractable soil Fe contents is exceeded by only two countries, Finland and New Zealand (Figs 31 and 282). There is no correlation between the plant Fe and soil Fe values. Although a few relatively high soil Fe values exist, the availability of Fe to plants in these soils seems to be low. Since no low Fe values occur in the sampled material, Fe deficiency is unlikely in Nepal. Manganese. According to the analytical data presented in Figures 33, 37, 283 and 284, no Mn deficiency or toxicity problems are to be expected in Nepal, provided that the sample material is representative of the Nepalese soils. Molybdenum. On average, the Nepalese plant and soil samples contain less than half as much Mo as all samples in this study and Nepal's positions in the "international Mo fields" are clearly low (Figs. 15, 20, 285 and 286). The availability of Mo to plants is highly dependent on soil pH and this varies widely in Nepalese soils. After correction for soil pH the plant Mosoil Mo correlation coefficient (r) is more than doubled. About one-third of the Nepalese plantsoil Mo points are within the two lowest Mo Zones (I and II), some of them at quite low positions. Mo deficiency is likely to occur in Nepal, most probably in the soils of the Bagmati area but also elsewhere. Zinc. Nepal occupies one of the low positions in the "international Zn fields" (Figs 41 and 43). The ranges of variation for both the plant and soil Zn contents are comparatively narrow (Figs 287 and 288). Only one sample pair is within the two highest Zn Zonesand depending on the extraction method one (DTPA) or seven (pH-corr. AAAc-EDTA) lie in Zone I. According to the latter method Zn deficiency could be more severe than the results of the DTPA method indicate. Although the lowest plant and soil Zn contents are not as low as in some other countries (see e.g. India, Iraq and Turkey), Zn deficiency should be expected at many locations in Nepal. See also Section 2.7. 248

mminnmommme mmoolowmommu a ommossmmea.ro 2000 a i 1 1:at Fig.283. Regression of Mn content of pot-grown wheat(y) on DTPA 80 111 15 1000Box=mgdmorymmixiimewmor.m.:::=2..= extractable soil Mn (x), Nepal. 800illInTINIPErcatrrurRM5raiairmslesessimmsoill 11111111111MMEMENIM . railiMMINIIIMOONIIIIIIIIIIIIIIIIIN 6001111MMIIMMIII=MIZEII111111111111111111P111 E 40011IMMEMINROMIUMINEMIMIUM a 300umommIlmommillsomm t5; 200IIIII1111111111111111111111121

'6 100111111111111111111111 80....--i:=6.a.nalli==vain.-1E67, .....,,-...... _. saimommtwallow.....mam in ginink: -do 60 40ihman161.61111 1MILI nil 30limmilsop mull 2011,111=1111111112111111111111MINI

10111111111111MIIIIIIM.IMM=BIOIPM=IPÌÌIIIIMIMIOW=1 INIIMOIM1EN 811111=111111111111111ialIllITiMMEIn :NIMMEMMitillMOMEMMIIMMIIIIUMMI III MINIIIIIIMMONSiZOINIMICUMEN =MI 2 3 46 810 20 30 40 60 80100 200 300 Mn in soil, mg/I (DTPA)

2000 ugimmmwq 4,11.11MMV,MPmy au I.1 mul,wRimerio raw. . mamma Fig.284. Regression of Mn content gr5 t1 of pot-grown wheat(y) on DTPA 100011..., ....ra...r,e261..Hies,=.---ssiiiii1111111111111111101 extractable soil Mn corrected for pH 800 (x), Nepal. 600.i.. ini.,, N. Ill INNIIIIIIII IIIIIIIIIII/11,111 NIMMIy ilIkUMELV. .LiaiRallilliiMpReal 400III Ir ' ° 6' 6.1 ii mosul loomsfl 300II 1 1111111=MIIIIIIIMMICUMIll 200I mom! IllimpillITICTIM

100...... canvialra==610ma 80onmiNnamemmilvammir-=mmaImalrammMINE. ...mmEn minaminmcsi6§1 60IIIIIIMIIIIMIllitIMIllnd!MMIN111111111=1 WI 40IIMIIMWiNdiMMIIIIINk nle 30IIIIIIIIIIIIIIIRMIr=11111 In 20MEMEL

10llill...... ILIII al 8Li ...... ,...... m.....,..11...... 1.. 2 3 46 810 20 301.0 60 100 200 300600 pH- corrected Mn in soil, mg/l (OTPA)

7.3.4 Summary

The soils of Nepal are relatively coarse textured, widely varying in pH, with low organic matter content and low cation exchange capacity. Of the soil macronutrients the contents of N and P are at an average international level but P contents vary widely. The Ca, Mg and especially K contents of Nepalese soils are low. Micronutrient problems concerning Cu, Fe, and Mn are unlikely. Although the Mo content of many Nepalese soils and plants is rather low, the most evident micronutrient deficiencies in Nepal are those of B and Zn. 249

Fig. 285. Regression of Mo content of 1111111111111111111i IMEARMIL=Ira011111111111111 pot-grown wheat (y) on ammonium 6ÚnggniiiMi oxalate-oxalic acid extractable soil Mo CATICEMIEZMIENE ° 251111 (x), Nepal. 3 2 ETINIM22'S E 0.1 isnis=emsaim111111.1111111 MENUMIMMUDNIMP MMEllosawarmisoni

" o o o 1AlaINIMEN11 .06i/ NWIWROMIMMINIP=MMU11.1ESMOINEIMIMMENNII

.03 .02

.0111111111111 .01 .02 .03 .06 .1 .2 .3 .6 1 23 MO in soil, mg/t (A0-0A)

EN milonalmiwo .M11 MaiMMIIIIIMolteisso Fig. 286. Regression of Mo content of =011U11111115MIIMINIMillilll ./fflIM INI_IMI11111.111 Monig[4.1.111 1WIMMINIC11=7:1=1111=111111111MMEMINIMMININO pot-grown wheat (y) on pH-corrected 6MENNEIIIIIMM=MMNIIIIGEIKaYMMIMMIIIMINMIMIIIIIIIIMM=leniiiiMMMEMMiliailiiiiiiiMai A0-0AextractablesoilMo(x), MEN1111131:111111MBIENNIIIRIEMEEILEM11/111=M11111 6 Nepal. 3MNIIIIIMISIME IIIIIIIIMMIZEIIIIIIIMNMI 2111111VgA°9ffliiiiiiiIIRENiiiiiiIIMMRS

E o_ 11E II 11111111111111111.111 ca. 1=1=121MENIPNI1111111111111---.....- .... 6=m1§1111 "C; 1=IMMEMENIMMOUP.11IMOIRMIIN w .6Menummmosnil rArmsnimmoA. .31111111111=MEMIllil m/MIIIIIMM11 .2InomninrAminnomm "ca''o o 1111111111111111E11111111E111 .1nzasol 2 _--mwzdonl===0:::::==sia I IER .06MIMMUI11111211PAIMMIMIllIMME01111=1MME11111MIZSEINI=MIENNI IMENIIIIMAMI '4 .03 Ilr- .02

.01

.006 .01 .02 .03 .06 . .2.3 .6 1 23 pH - corrected MO in soil, mg/l (A0-0A1 250

PVIMIIMY11 mmmimm Nos 200 -1I o 1 Fig. 287. Regression of Zn content of

Z n t 77 pot-grown wheat (y) on DTPA extractable soil Zn (x), Nepal. 1 :5321I log 100 90 BO 70 g. 60 cL 50 o .240

10 9 8 7 6

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Znin soil, mg/I (DTPA)

200 rs I te n t" 3 13 ± 1. 82

100 90 80 70 60 50 40

89 7 Fig. 288. Regression of Zn content of 6 pot-grown wheat (y) on AAAc-EDTA .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 extractable soil Zn corrected for pH pH-corrected Zn in soi , mg/i (AAAc-eorm (x), Nepal. 251 7.4 Pakistan

7.4.1 General

The distribution of the 242 sampling sites in Pakistan (Fig. 289) covers relatively well the most important agricultural areas of the country. Of the 131 soils classified according to the FAO classification system, Yermosols (44 %) and Cambisols (36 %) are far more

e , 0 50 100 150 200 km

I 1 I 1.----1 ) re 7,61- 7179'',... z'\ 12, 1 7341-7350 ...... _2j .00951-8995., / F--. 70 )\ . 204\ e 7270 \ 0 ( 5 W.Whyir.) ..) B520 615N. e51254

6.9437 69 (. C. 893 7197 .8-4. "41:13 '6'5 j:° '5'67'1046' I 435 726.1° I 7195 _____3 . 150 52. ., 196 B. ,, 18395 7190B.15 155 - 716 0927. 9496 6 2a 9 15,1 \....,. 9.29 LYAL LI PUe70,17 a 6 8950 0.7 .0 / ,_,.., 79:-i8..-21::: 7269 .' 842 .57 ,.i 7"' 45 MI6 7{57 A " 25 ',K.. 7 7

, 266 ,..- o 40. U-.----in: ea 7ULT -'':::? 3 30. 30. , .71. 9 596 849 737

.0.64 71,i 0 i 7 e %VI P UR i 8977 .)/

6 es 9489 (. 0460 i 0890. 698 406

e 9497 /, 1 72 /- \ :816U KUR 00020 MINI ----'-- ,,, if orzA

f i - 0 . _ rxr r "?;,,,,

29e ; _,j ze WS'' 7,.. P ,ii./ e \I HYDERABAD :' 8 i OB 7300 \ 7308 I .4 ., DERADA ' 7310 \ 0 ,--.A A 4° 4 ,2° := 2 ./-

20. 70' PAKISTAN

Fig. 289. Sampling sites in Pakistan (points = wheat fields, triangles = maize fields). The last four numerals of each sample pair number are given. 252 common than others: Fluvisols (9 %), Halosols (7 %), and Luvisols and Xerosols (2 % each). Most soils are coarse to medium textured and highly alkaline (Fig. 290). The national average pH(CaC12) is higher than in any other of the countries studied and the standard deviation is exceptionally narrow. Low organic matter content and low cation exchange capacity are typical of Pakistani soils. The values for electrical conductivity and CaCO3 equivalent are relatively high (Appendixes 2-4). In many locations, especially in the Multan area, the alkalinity and high electrical conductivity are combined by high Na contents of soils.

35 Pakistan Internat. Pakistan Internat. Fig.290.Frequency distributions of n=242 3764 rue. 242 3783 7= 40 44 s=7.77 6.64 texture, pH, organic carbon content, 30 s= 8 16 s= .25 1.12 min.= 19 9 35 min.= 6.15 3.62 and cation exchange capacity in soils of max.= 65 55 92 max.= 8.56 8.56 Pakistan(columns). Curves show the international frequency of the same characters.

10 20 30 40 50 60 70 80 90 5 6 TEXTURE INDEX pH (Ca CIO

45 PakistanInternat. 35 Pakistan Internat.

40 n=242 3779 n= 242 3777 R.R .6 1.3 R= 16.5 27.3 Rs= .3 1.2 Rs=4.9 14.6 35 min.= .2 min.= 7.5 .2 o max.= 3.2 39.1 max.= 41.2 99.9 w30 25 U LTJ 20

S15 CC IL 10

.2 .4 .8 1.6 3.2 6.4 12.8 25.6 8 16 32 ORGANICC,% GEC, me/100g

7.4.2 Macronutrients

Because of the relatively high and greatly varying nitrogen fertilizer applications (wheat 58 ± 57 and maize 83 ± 90 kg N/ha), the nitrogen contents of original wheat and maize vary widely but are generally at an average international level in spite of the exceptionally low total content of soil N (Figs 6 and 291). Pakistani soils, on average, are poorer in 0.5 M NaHCO3 extractable phosphorus than soils of any other country, but the P contents of plants are only slightly below the international means (Figs 7, 292 and Appendixes 2-4). Greatly varying but mostly only moderate amounts of phosphates had been applied to the sampled crops (wheat, 9 ± 15 and maize, 12 ± 29 kg P/ha, Appendix 5). The discrepancy between low soil P and relatively high plant P contents may partly be due to the effects of CEC (Fig. 14, Section 2.2.4). 253

Figs 291-295. Frequency distributions Pakistan Internet Pakistan Internet 1765 1958 35 n= 156 n= 86 of nitrogen, phosphorus, potassium, =4.56 4.27 =3.13 3.14 es= .82 1.15 as=.76 .87 calcium and magnesium inoriginal 30 nan= 2.19 .60 30 min.= 1.37 .88 max.= 6.25 7.45 max.= 4.67 6.51 wheat and maize samples and respective 25

soils(columns) of Pakistan. Curves ti20 show the international frequency of the W 15 same characters. o

5 6 3 4 5 6 N content o heat, % N content of ma ze,%

Pakistan Internet_ PakistanInternet n = 156 1765 n. 86 1958 60 0..082 .133 60 =.077 .135 15=026 .084 as= .022 .088 min.= .033 .009 min.= .008 .008 mar.= .169 1.023 0 max.= .139 1.657

42.40 o Z 30 o W 20

.005 .01 .02 .04 .08 .16 .32 .28 .005 .0 .02 .04 .08 16 .32 .64 1.28 Fig. 291. Nitrogen, Pakistan. N in wheat soils, % N in maize soils,%

40 40 Fig. 292. Phosphorus, Pakistan. Pakistan Internat. Pakistan Internat. 1765 n=156 5 n= 86 n=1967 =.37 .38 8.31 8=.33 as= .10 .12 ts=.07 as=.10 min.= .12 .06 30 rnin.= .15 min..05 max.= .72 1.02 max.=.52 mox.= 1.04

1 2 .3 .4 .5 .6 1 .8 .9 P content of wheat,% P content of maize,%

40 40 Pak istaninterne Pakistan Internat.

35 n= 156 1765 5 n= 86 1967 = 7.1 20.2 8= 6.3 22.5 as=6.9 24.7 as= 6.0 33.0 min.= 1.4 1.0 min. .8 0.1 max, 51.6 271.4 max.= 41.0 656.0

o2- 20 oD 15

a_ 10

2.5 5 10 20 40 BO 160 320 640 2.5 5 10 20 40 80 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I

The potassium contents of Pakistani soils are slightly low in the international scale (Figs 293 and 8). In spite of this and of the small rates of potassium application (Appendix 5) the K contents of plants are generally high but vary widely from one sample to another. Obviously the high rates of nitrogen application have accelerated the uptake of K from soils (e.g. Rinne et al., 1974 a; Sillanpää, 1974). For example, maize plants containing more 254

35 Pakistan Intermit PakistanInternat. Fig. 293. Potassium, Pakistan. n. 156 1765 n= 86 1967 8= 4.8 4.0 O Os 3.6 3.1 SSS 0.9 1.0 ±s= 1.0 1.0 min.= 2.2 0.9 min.= 1.5 0.6 max, 6.8 6.8 5 max.= 5.5 6.7

4 5 6 2 3 K content of wheat°/0 K content of maize,%

Pakistan Internal Pakistan internat n = 156 1765 n =86 1967 = 245 365 es 188 330 30 SSS 127 283 0 ±s= 98 356 min.= 55 20 min =58 18 max = 945 2097 max.= 510 5598

20 >_

15 o tu

25 50 100 200 400 800 1600 32006400 25 50 10a 200 400 BOO 1600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

40 PakistanInternat. Pakistan Internat. Fig. 294. Calcium, Pakistan.

5 n .156 1765 3 n= 06 1967 = .39 .43 X= .50 .47 01. .13 .17 05= .23 .20 min. .13 .11 30 min.= .22 .09 max.= .78 1.68 max, 1.62 1.88

w5r, .8 .2 .8 Ca content of wheat, % Ca content of maize,%

494 40 Pakistan Internat. Pakistan internot. n =156 1765 n= 06 1967 35 1= 4516 4671 35 8= 4102 3450 ts =770 3076 *5=839 2815 min.. 1200 110 min.= 2230 10 max.= 7240 21930 max.= 7160 17995

2.14 25

ILI 15 o ix1./J 10 u_

100 200 400 800 1600 3200 6400 12800 25600 SO 100 200 400 8001600 3200 6400 12800 Ca in wheat soils, mg/I Ca in maize soils, mg/I

than 4.0 percent K had received 123 kg N/ha while plants with less than 3.0 percent K were only fertilized with 44 kg N/ha on the average. Similarly, wheats with 5.5 percent K or more had received 91 kg and wheats with less than 4.0 percent K only 51 kg N/ha on the average. 255

40 Fig. 295. Magnesium, Pakistan. Pakistan intemot. PakistanInternat. 35 ri= 156 1764 35 n= 86 1967 5= .189 .172 5.280 .251 55..044 .060 4s=.127 .119 u min.= .106 .044 30 min.= .124 .036 max.= .358 .948 max.= .652 1.125 25 >- C_2 20 20

1.11 D 15 15 z331.1.1 u- 10 5 _A, addal 5 Mg content of wheat, % Mg content of maize, % 40 Pakistan Internat. Pakistan Internat.

35 n= 156 1764 35 n= 86 1967 9.362 489 0.292 44E 4s=154 437 4s=107 min.=78 10 min.=57 1 max.= 924 3298 max.= 615 6490

o 5

a 20 3-

pm7,, ex4e4', .4.41M 25 50 100 480 400 BOO 1600 3200 12.5 25 50 100 200 400 800 1600 3000 6400 Mg in wheat soils, mg/t Mg in maize soils, mg/t

The average calcium contents of Pakistani soils and plants correspond closely to the international means (Fig. 9). In wheats and wheat soils the contents are slightly on the low side, while in the case of maize they are on the high side (Fig. 294). Although the exchangeable Ca contents of Pakistani soils vary only little, the variations in the plant Ca are wide. In general, the magnesium situation in Pakistan is quite normal (Figs 10 and 295) and in many respects is reminiscent of that of Ca. The variation of exchangeable Mg contents of soils is rather narrow but wide in the case of plant Mg.

7.4.3 Micronutrients

Boron. In the "international B fields" (Figs 22 and 25) Pakistan is among the countries occupying the highest positions. The correlations between plant B and soil B are very high (Figs 296 and 297). Correction of soil B values for CEC increases the relative share of high B values and moves Pakistan to the right in the "international B fields". About eleven percent of the B values fall in each of the two highest B Zones (IV and V), many of them in extreme positions (Fig. 297). Many of the highest B values are found in Multan but there are also soils rich in B in Sind and elsewhere. Low B values were often recorded in samples from unirrigated sites. If soil B values of 0.3-0.5 mg/1 are considered as critical 256

100 _ Fig. 296. Regression of B content of 80 monatmotinammar-TithommatommmismossisioassiNumiumcom=mMmas 60 gEREINIIIIENIENNEMII==M111111111N pot-grown wheat (y) on hot water = 8.81 86 E_Mminammowni soluble soil B (x), Pakistan. For details 50 se_ glica7713/1111 40mg.° of summarizedinternationalback- 30 MIIM ground data, see Chapter 4. Q: 20 0 iliN11111=11111111 O

10INE111111M11111111 "6811111101IMMINIORMIERIMSMIIIIIMINIMEMOIIIIMIR1111MINKMORILIPIMMINIMMEN 6111111111MliMMINNIMAildeMilli1111111M111111 5IIIIIMIZIMMINTAMMIMUIMM=M1111111 4IIIIIIIMINI,M111111 MEDIUM

co 3 2 uiniuinuiEEL! 111111111111 1111111 .2 .3 .4.5.6 .81Q 2 3 4 56 810 B in soil, rag/I (Hot w. sol.)

100 Fig. 297. Regression of B content of 80 pot-grown wheat (y) on CEC-corrected 60 rt.= 7 358 I 786 6 9 50 (hot water soluble) soil B (x), Pakistan. 2±-090 0,4 O. 0 40 E a_ 30 y. 0 41.36x r 8 0. r= 0 886 02

'20 *1 _c 1 0 8 6 a) 1..1.1 . 5 4 3 co 2

III .08:1 .2 .3 .4 .5.6 .810 2 3 4 5 6 8 10 C EC -corrected B in soil, mg/1 (Hot w sot.)

(Section 2.7), a number of Pakistani soils can be suspected of B deficiency. Copper. Almost all the Cu values measured from the plant and soil samples are within the normal Cu range (Zone III, Figs 298 and 299). The present analytical data indicate neither Cu deficiency nor excess. Only one sample pair (48462) from Multan falls in Zone V but even that is not in a very extreme position. Iron. Against the international background the Fe situation in Pakistan is relatively normal (Fig. 31). About ten percent of the Fe values are within the two lowest Fe Zones (Fig. 300), mainly in Zone II. Such values as are in Zone I lie towards its upper boundary and, at worst, indicate hidden Fe deficiency. Dry weather during the growing season may 257

Fig. 298. Regression of Cu content of 22 pot-grown wheat (y) on acid ammonium 20EISE11111111111111111011111 acetate-EDTA extractable soil Cu (x), Pakistan. `7' ± 11111111Wi 111111111M11111 E18 8:161111111111111:11m CI 113E11111

46:1411111111111111111MIIIN

.6 12

101111111111111111111111M21111 3811111111111111111111EINUM1111111 61111111111019111111111111111

41111111111M911111111Ii1111

o .2 .4.6 .810 2 46 810 20 40 6080 Cu in soil, mg/l (A A A c- EDTA

Fig. 299. Regression of Cu content of 22'Cu pot-grownwheat(y)onorganic carbon-corrected (AAAc-EDTA extr.) 20 soil Cu (x), Pakistan. -o Px,o. , ,- 18. :..- -; =...;-. taY21111611=11161 ri

10 INIII ...i, ...... 1ION, o . 6 ...... g. .. . 4 IMISINIMMIll 2 IIPIF 11111,., .4.6 .810 2 46 810 20 40"3-- 60100 Org.C-corrected soil CU,mg/l IAAAc-EDTA)

lead to oxidation of ferrous to ferric Fe due to excess soil aeration, and so decrease the availability of Fe to plants and induce Fe deficiency (e.g. De Kock, 1955; Granick, 1958; Mokady, 1961). Most of the lowest Fe values were measured from samples collected in the Peshawar and Lyallpur Sheikhupura Lahore areas. Manganese. With its highly alkaline soils Pakistan is among the countries of the lowest Mn status (Figs 33 and 37). The variations in both plant and soil Mn contents are exceptionally small (Figs 301 and 302), and nearly all Mn values are concentrated in the two lowest Mn Zones or close to the lower boundary of Zone III. There seems to be no clear geographically distinguishable low Mn districts and soils of Mn shortage are scattered 258

3 00 II Fig. 300. Regression of Fe contents of 2 751711FAMMIIMILMIWN MUIR 2 50 pot-grown wheat (y) on acid ammonium 225niMPOrwiliffirigni acetate-EDTA extractable soil Fe (x), 200MlpflI11111111011111111I I Pakistan, 1 75 r,357,111111151111iiiiiiIMI 111111 E 1 50Ill1111111E1111 I 111111 1 .125a: 1111111111111111 airs 1001111111111111111111 .690LimmummillumiI 80minumlemonaresuni 70 NIIIIIIIIIIMMIIMIll o C-) 60 MIMIll11111

50Mgeolr- lilMiii

4011111111 ' ' MIMI I 111 30 immintm111E111plimmitawmu 10 20 40 60 100 200400 600 1000 2000 Fe in soil, mg/l fAAAc- EDTA)

all over the sampled areas. On the basis of these analytical data it seems likely that Mn deficiency, acute or hidden, is relatively common in Pakistan. Molybdenum. Unlike Mn, no low Mo contents were found in Pakistani plant samples. The national average for pot-grown wheat Mo content was the highest among the 29 countries studied (Figs 15 and 20). Owing to ammonium oxalate-oxalic acid being an inefficient extractant of Mo from alkaline soils, the national mean soil Mo content is lower than the international average (Fig. 15). This contradiction is resolved by correcting the soil Mo values for soil pH (Fig. 20), whereupon the national mean for soil Mo becomes higher than that of any other country. Almost 40 percent of the Pakistani plantsoil Mo values fall in the two highest Mo Zones, with more than half of these in Zone V (Figs 303 and 304). Geographically there are no distinct high Mo regions. The origins of high Mo values were scattered all over the sampled area, being somewhat less frequent, however, in Rawalpindi and Peshawar than elsewhere. Zinc. With Turkey, India and Iraq, Pakistan belongs to the group of countries where low Zn values are more common than elsewhere (Figs 41, 43, 305 and 306). The national mean values of plant and soil Zn are similar to those of India but the variations in Pakistani Zn values are smaller. Relatively fewer Zn values therefore fall in the low Zn Zones and in less extreme positions than in the case of India. Nevertheless, the frequency of low Zn values is internationally high and in numerous locations a response to Zn can be expected. Only a few Pakistani samples showed an abundance of Zn. The low Zn values were perhaps more typical for Sind, Multan and Peshawar than for other sampled areas. 259

Fig.301. Regression of Mn content 2000 mmwr 1Imummot ___ of pot-grown wheat (y) on DTPA 1111111011111111.11 extractable soil Mn (x), Pakistan. 100011 11=MMEMI=11FirLEMMEMAIMENNWIMMeYll,=UMMIN1 800 II rninziiammnmuvsh.-smwmmenmmmrrnMiliEMENIMMMIKWIRVINITIMIEn 600r ...1 1 k. 340000iiiiiiimiiEpp 1 200LL 2 !o1 00..11E1111111E111PMleINERM=MLIMIII -aALlillik 11 C :IMCNE WilreMniMME=1 CD 80 :I IMIIM=MMEENICELMlatrillThilnE011 -a. NI 11159MITEMIIMMMEMENI101MH b U 111MMIIIIIMirilaMEMMIllMII =*.IIIVENIPIMMIMINIII1=M 2c304° IIIIMMIMBECHUINMEMIIIIIMME 2091,,,1211111111111..1111111=111

10hilliiiiMPIIIIIIINMI=MMIIIIII M 8 =1MilinfiliaMaaTiNNUMEN IffiTMMI H =1MEMIRON=C2111MMINMEMOI lii= 2 3 46 8 10 20 3040 60 80100200 300 Mn in soil, mg/l IDTPA I

Fig.302. Regression of Mn content 2000 of pot-grown wheat (y) on DTPA Mn extractable soil Mn corrected for pH 1000 mmromvp...=mumbi.m.m=moINUMMAIRMININ 800 mmoakeiEHLRNEwinGEtntomonHimmim[m (x), Pakistan. l---MH121nwEENI---1H 60011MONIIIEIMIEMIEZINOVIIMMMMMEMMIWrRrINUMEilWITEEd.11**i 0.NEMçNUMINEH 400 I 300 _c 200 .4 o 100 AIIII 6. cu 80=11MiN=111 II111=EITAINTrIMENNII'l'ECORI 60II mromitmernomlmmsemil 4011 EmiermissnimmEmmmmne 30 20

1011/111111111.1111111 8 nMIIIIINHESIMMTAMMEMEH ...... NI M===111111H111LINOMMINEMEMI =MMINUAN==Hff 2 3 46 810 20 30 40 60 100 200 300600 pH - corrected Mn in soil, rng/i (DTPA)

7.4.4 Summary

The Pakistani soils are usually coarse to medium textured, low in organic matter and have a low cation exchange capacity. The outstanding feature is their high alkalinity. Very low P and N, relatively low K, medium Mg and Ca and high Na contents are typical of most sampled Pakistani soils. Out of the six micronutrients included in this study, only the contents of Cu and Fe are usually within the normal international range. The contents of B vary widely, high values being the most common, but in a number of soils shortage of B is apparent. Owing to the high alkalinity of Pakistani soils, the availabilities to plants of Mn and Zn are low but that of Mo is high. Consequently, shortages of Mn and Zn and an excess of Mo are typical of many soils. 260

Fig.303. Regression of Mo content 1112; ; 6 111' 01111011111MINEMIMIZ17:::: of pot-grown wheat (y) on ammonium MMiMiii.iiigii*KaniEnntal oxalate-oxalic acid extractable soil Mo MEMIRM.98lo - . 3 lititr! (x), Pakistan. 2IMIENIIIMIIIIIIIII E o. 1 MIIIIIIISIIIIIII..norr,ammea...... 2 .6 -- ...... MMillitidEIMEIMI--... NECIVINICOMMIN II ó 3MMEPrilliMarinii I .2 o .11111111111111111111111& ...CM11:1110..=mwm..111MME11====11WI W& g .06W =IMIREMPAI = .....- ir 19 .03 .02 III .01

.01 .02 .03 .06 1 .2 .3 .6 1 23 Mo in soil, mg/ (A0- OA

1IMM MIIIII1111MMXIMMINIIMIIMO Mo ilikalran magsparm=rn 6 n.237 82 1131111liftrIMMEEEISIIIIIMMINIE 3 55' ZENUIPC ° 2 111111111C-MINIMINI111101

1111111111111111M11MAIMMOMMOEilli111111= MINEMENNIMMMEMIltaMall..aiIMM=11MIMIN=1710NMPIIITilliPWAWINIIIMMINNIIII MEMENICilifranriiiniEMMIN=MMENNI IIIIIIIIIPIPMEN411!1MMEN MEMENMEMIEMEniiiiMMIEME Ululi ..11111MMINEINIOIMMEMI

.2 1 1111111T11.1111111imrill MIZIO11=2:1.111=="ZE:1111==mommara 11111111116. IERMMENNI=n==711=MMENNI=11MLI MIENNUI/11. 4111111M111111111 11MIMMEIMIIIIMIE=ME 11111MT NONNI MINIIMENIIIIIMEN .03 .02 Fig. 304. Regression of Mo content of pot-grown wheat (y) on pH-corrected .01 1160111111MLI .006 .01 .02 .03. 06 . .2.3 .6 1 23 A0-0AextractablesoilMo(x), pH - corrected MO in soil, mg/t (A0-0A) Pakistan. 261 rsysou...... Fig.305.Regression of Zn content 200 1:, n= 3 r: 3 of pot-grown wheat (y)on DTPA Zn 3±=I .....R± =. extractable soil Zn (x), Pakistan. yr:h: 39 D17:11 100Ill 1 MIR 90 80Iiii1111MMENIIIMM....11 IMMEN1111111M1111111MIIMMIIIII 70 E 60 o_ 50 o .c°'40111111111111111,1 "6 30 iiupiiiir

G) 111 4g. 20IMIIIIIMPRiel1 N 1011111Pillal 9 8 7hdaiiiiiiii1111111Emid 6 111111111111111111111111MIli 3L .2.3 .4.6 .81 23 46 810 20 3040 60 100 Znin soil,mg/l (DTPA)

1 ,. 1 183 Emmiiiiiilligniiii' : 1. k 8:'8' =Ell111111=1 1 on mow 1 '1 111111 1111111=11111

, liiiilii !MIMEO 111111 01111II 1 OfII.,;'IMIL 'L

1 =2106.--;0114''....'..:161.IlkIL1001 Fig. 306. Regression of Zn content of pot-grown wheat (y) on AAAc-EDTA ' II 111 11111E11N extractable soil Zn corrected for pH .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 (x), Pakistan. pH-corrected Zn in soil, mg/( (AAAc-EDTA) 262 7.5 Phi'Wles

7.5.1 General

The sampling sites of the 197 maize-soil sample pairs collected from the Republic of the Philippines are well distributed over the most important agricultural areas of the country (Fig. 307). The sampled soils were not classified into FAO soil units but according to the FAO/Unesco Soil Map of the World (1976) the majority of Philippine soils consist of

kr, V

.51569-573 . ' 61t6.4.RAO PH ILIPPI NE S 567-56857'757'575

A. 63-565 , 699 , iArBONG 59456 ,A, . 7°. A535-5911599-550 e '5 At59-555"5

691 e9e A tl 1 561 "5" 55°" '591-592 695 ,t, 597-548 Ital.. 1 .. 517 533 585-586A ,,t' 5° I "45t-4:31Z;'

1 a et 97 59 soz ,59 69 Or_60.14...a.. 99 o IIP v .o o o"°'6S AC 'Attt..A.BOAC 691A

0 06

) % 0

, ,,,,6 -KAALOGA GA V .,t ..

AG1,0 Q^.. -619 t*-Ei et 612 , 0 s 522 '° t5z'It 607-60B 62e6e-'°Q'. 605 68 09

D .0 6,0 6p,ryc 11:68 686 se-es I -f69-672 , t.t 0 ABA 6 7-678\ AA 6 3-679 675-676 .DU GUE 99 652 655 56 -- 6 NIP 653-659 : 6 8 62g456:intler-" sse OROOUt A ' . 1,1 2 64 A 6 9, G. A enf". 6. .1What.,

eso AA . A626-628 il 82 "5'31631-632 633-636A

At' \ 6'5°T" A't. 1 5502.1

1 0j5°049' 510

1 51 A 51 1 JOLn:::),,,:,,,,,.0.i. tit

t t.-tt

o(0 , ,. a

Fig. 307. Sampling sites in the Philippines. The last three numerals of each sample pair number are given. 263 Nitosols, Acrisols, Cambisols and Andosols. The textural variation of the soils is exceptionally wide (Fig. 308) covering the whole texture scale from very coarse (texture index <20) to very fine (TI > 90). In the pH of soils the variation is also great and only the most alkaline soils are not represented. The contents of soil organic matter are usually at a medium level but the majority of soils have a relatively high cation exchange capacity. The electrical conductivity and CaCO3 equivalent values and sodium contents of soils are generally low (Appendixes 3 and 4).

35 Philippines intomot. Philippines Internet n. 197 3764 40 n= 197 3783 R. 48 44 0= 5.99 6.64 30 ±-5.18 16 so=.81 1.12 min.=17 9 35 min.= 4.23 3.62 max.= 91 92 max.= 7.56 8.56 S'2 0

20 25

o 15 Lk' io F7A u- 10

0 24 30 40 50 60 70 BO 90 TEXTURE INDEX pH(CC:C12)

45 PhilippinesInternat. 5 Philippines Internet 40 n.197 3779 nr.197 3777 n. 1.2 1.3 A. 35.6 27.3 ts. .7 1.2 xs= 12.8 14.6 4E 35 min.=.3 .1 min..5.7 .2 rnax 39.1 max.. 67.2 99.9 30

Q8_ 25 C.) uj20 Fig.308.Frequency distributions of O texture, pH, organic carbon content, U 15 and cation exchange capacity in soils of LL 10 Philippines(columns).Curves show theinternationalfrequencyofthe .2 .8 1.6 3.2 6.4 12.8 25.6 2 4 8 16 2 6 same characters. ORGANIC C.% CEC. me/100g

7.5.2 Macronutrients

The total nitrogen and NaHCO3 extractable phosphorus contents of Philippine soils are of the average international level and the N and P contents of maize are slightly lower (Figs 6, 7, 309 and 310). In all cases variations are wide. The sampled crops were fertilized with only small amounts of N and P fertilizers (8 ± 19 kg N/ha and 1 ± 4 kg P/ha). Wide variation is also typical for the potassium contents of Philippine soils as well as for the K content of maize but, in general, the K status is relatively low (Figs 8 and 311). The international minimum soil K value (18 mg K/l) and the minimum maize K content (0.6 %) were both found in samples collected in the Philippines. Only nominal amounts of potassium fertilizers (1 ± 4 kg K/ha) were applied to the sampled maize crops. A good response to potassium is likely at many locations in the Philippine Islands. With regard to calcium and magnesium the Philippine maize and maize soils tend to be high on an international scale, Ca contents only slightly so but the national mean Mg contents are among the very highest (Figs 9, 10, 312 and 313). As for many other properties of Philippine soils and plants, extensive variations are typical for Ca and Mg contents. 264

40 3 Philippines Internat. Philippines Internat n= 197 1958 n= 197 1967 35 Philippines Internat. 30 7= 2.5 3.1 .2.99 3.14 C n. 197 n.1967 so= 0.8 1.0 ±s =..51 .87 R. .29 6=.33 min.= 0.6 0.6 U 30 .88 "E. .10 (1.1 os= .07 is= ?-1 25 max.. 4.4 6.7 max..4.50 6.51 30 nin.=.16Ø min. .05 max, .60 f4 max.= 1.04 O_ 25 (1"./ 0- 25 >- 0 20 4v >- ° 20

D15 15 uJ CY 0 lo u_ u_

5 5

3 4 5 6 .2 .3 .4 .5 .5 .7 .8 2 3 4 5 6 7 N content of maize,% P content of maize,% K content of maize,%

70 Philippines intrnot. 35 Philippines Internat. Philippines internat. 35 n= 197 1967 5= 18.5 22.5 n. 197 1967 60 in 197 1958 ts = 18.9 33.0 km 212 330 7..139 .135 min.. 1.5 0.1 no.212 356 -.E. mox..138.6 656.0 4E.30 ts= .063 .088 min.= 18 18 CD min.- .039 .008 50 max.= 1997 5598 max.= .527 1.657 o 25

40 3- 0 6.3

W30

1,11 20 u-

./1 :47)%111.o... 2.5 5 10 20 40 80 160 320 640 25 50 100 200 4008001600 3200 6400 .005 01 02 .04 .08 .16 .32 64 1.28 N in maize soils,°/0 P in maize soils, mg/I K in maize soils, mg/I Fig. 309. Nitrogen, Philippines. Fig. 310. Phosphorus, Philippines.Fig. 311. Potassium, Philippines.

4 Philippinesint..n.t. 40 Figs 309-313. Frequency distribu- PhilippinesInternat. 35 n 197 1967 n=197 1967 tions of nitrogen, phosphorus, = .52 .47 5 en. .16 .20 R. .340 .251 0) min.. .22 .09 no=.174 .119 potassium, calcium and magne- moic=1.11 1.88 ?-3) min.= .102 .036 max.. 1.125 1.125 0.25 sium in original maize samples 151. 25 7- and respective soils (columns) 0 20 (±) 20 of Philippines. Curves show the 8 15 o 15 international frequency of the E 10 Q- 10 u_ same characters.

.4 Ca content of maize,% Mg content of maize, %

PhilippinesInternat. 0 n = 197 1967 PhilippinesInternat. 35 7 = 3839 3450 te = 2752 2815 35 n= 197 1967 min.= 100 10 7= 811 446 mos..12800 17995 no=877 462 30 36 1 30 min.= 6490 "E" max.=6490 25 O 25

20 (_)

..,(4 ANN 50 100 200 400 8001600 3200 640012800 12.5 25 50 100 200 400 8001600 3200 6400 Co in maize soils, mg/I Mg in maize soils, mg/I Fig. 312. Calcium, Philippines. Fig. 313. Magnesium, Philippines 265

7.5.3 Micronutrients

Boron. Of the six micronutrients included in this study B deficiency appears the most widespread in the Philippines. The national mean B contents of soils and plants are among the lowest recorded in this study (Figs 22 and 25). Every second plantsoil sample pair falls into the two lowest B Zones (Figs 314 and 315). This high frequency is exceeded only by Nepal. The plantsoil B points falling in the normal B Zone (III) occur most frequently toward the lower boundary of the zone. There seems to be no clear geographical distinction between the low and normal B areas, but soils of B shortage seem to be widely distributed over the sampled areas.

100 ;fiutiminisilitaiv;211:1 80 miammEnsammommo 60 NI -43376=MM:in 50 0 28_0 40 MENE11111 30 rinffingirnani

9.-20 -6 G) Ill 11111M11111111 10 low =simixiir 15 8 Immwm=mnimeimi MINIESi11111111111111111 o5 suanommeemrea mmummun "C"o4 11111111WFMNII1II MI1111111 o MEEM co 3 Fig. 314. Regression of B content of 2 EN11111 INN pot-grown wheat(y) on hot water soluble soil B (x), Philippines. For de- Ill MINI 1111 tailsofsummarized international .2 .3 .4.5.6 .810 2 3 4 56 810 background data, see Chapter 4. B in soil, mg/l (Hot w sol)

100 80 5.1 358 60 ±s.4 083 6 _ 50 2±s=02 012 0 ± 40 E y.3 68+2 2x r 8 55 CL 30 r=0393* 0 2 *

.ö

-sr 2

Fig. 315. Regression of B content of I In pot-grown wheat (y) on CEC-corrected (hot water soluble) soil B (x), Philip- .081 .2.3.4 .5.6 .8 1.0 2 3 4 56 8 10 pines. C EC -corrected B in soil, mgil (Hot w. 266

22 P ntea Fig. 316. Regression of Cu content of 20 pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), 18miaow rimollini E Philippines. 8:16EweocESCIIIMM111111 o w1411111111111 111111111111111/111 .c 1512111111111111

77,10M1111111 111116%22111111 880111111 1121116111111111 7 (-) 611111111=11111111E11111

4 2111111INIE9!!PRI11111 .4 .6.810 2 4 6810 20 40 6080 Cu in soil, mg/I (AAA c- EDTAI

Fig. 317. Regression of Cu content of 22Cu pot-grown wheat (y) on organic car- 20 bon-corrected(AAAc-EDTA extr.) u!1III soil Cu (x), Philippines. 18 4.:7410110 iniu irog6=80saimmeal"! To

o 10 111 1 1110111111

8 inipowomiigul o 6 IMIII1111111111111111

4 IIIIMIIIIIIIIME111111

2 IIIIIIME1111111MPIII: , ftralmann .2 .4 .6 .8 1.0 2 4 6 810 20 40"4" 60 100 Org.C-corrected soi CU,mg/I (AAAc-EDTA)

Copper. Contrary to the occurrence of B, the soils of the Philippines are exceptionally rich in Cu (Figs 316 and 317). With Brazil and Italy the Philippines occupy the highest positions in the "international Cu fields" (Figs 27 and 29) and high Cu value (Zones IV and V) were found more frequently in the Philippines than in any other country. Only a couple of relatively low Cu values were measured in samples which came from Camarines Norte (Bicol) but there are no indications of Cu shortage anywhere else in the country. The highest Cu values were recorded in Ilocos samples, but high Cu conients were also found in samples from Northern and Eastern Mindanao and occasionally elsewhere. 267

Fig.318. Regression of Fe contents 300 275Fe ' P?.3 int rna of pot-grown wheat (y) on acid ammo- 250 nium acetate-EDTA extractablesoil 225 PINFENNIMIMION1 Fe (x), Philippines. 2001='±s' 3.11111EIME 1 75 Etom2ViEliiiNgiiiiiii ° °" E 1 50 ca. .125E 111 11 o 100IN ME . EMIMilli mlRatag.iiiiiampi- 111111RESI 50 pp., 1 1 \ 40 ,\ \ . \\ 30 III \rg\V. 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/I (A AA c - EDTA)

Iron. The Philippine national average Fe content of plants is the highest and that of AAAc-EDTA extractable soil Fe the fifth highest recorded in this study (Fig. 31). Every third sample pair (Fig. 318) falls within the Fe Zones IV and V but none within, nor even close to, the low Fe Zones (I and II). It seems therefore that the possibility of any response to iron fertilization is minimal. Most of the high (Zone V) Fe values originate in Ilocos and Central Luzon. Manganese. Substantial variations in the plant and soil Mn contents were found in the Philippine material (Figs 319 and 320). In general, high Mn values were more frequent and the national averages are higher than those of most other countries (Figs 33 and 37). Only one sample pair (43688 from Bohol) indicates possible Mn shortage but the relative frequency of high Mn (Zones IV and V) values is almost double the international frequency. Almost all of the highest (Zone V) Mn values were found in samples from Northern and Eastern Mindanao. Molybdenum. In the "international Mo fields" (Figs 15 and 20) the Philippines are placed centrally, but as in the case of several other soil properties extensive variations in both plant Mo and soil Mo contents are typical (Figs 321 and 322). The correlation between plant Mo and A0-0A extractable soil Mo is substantially improved by pH- correction owing to wide pH variations. The sites for both the high and the low Mo contents are geographically scattered; high contents were associated with soils of relatively high pH and the low contents with soils of low pH. At locations where Mo deficiency might be suspected, raising the soil pH through liming may be a sufficient treatment to increase the availability of Mo to a normal level. Zinc. Generally high but widely varying Zn contents are typical of Philippine soils and plants (Figs 323 and 324); from several locations abundant Zn contents were measured. In 268

2000 Roweam=mml Fig.319. Regression of Mn content of pot-grown wheat(y) on DTPA 1000 rawT-wvcwagginanilimmeme iimimml= extractable soil Mn (x), Philippines. 800 1==l-1NMM:1011111111101-111=Mr 600IIMMMINEMInMsmoniu1714:111511NIMIIhMILMREHI111111==== E 400 E-.300 20011=1111111111111111111.11PMEN

45 100111111MMININIM:i1M2MINEMMIYMMIMENn11111.111111NEIMMIN 801111r1/11=1111:11MEMMOI .5 60insmagilimmortirrammmonm=rammg 40 30iimmPrissiumummuomma 20imailniummounimus

loMal ININ=MMINN=1111MEMEEllillillE1111112:.°34:b9 I 8 IMME11.11MMIE Il:171 MIMMENIIMINEMlialEMITOMMEMI. NYWi= 1 2 3 4 6 8 10 20 30 40 6080100200 300 Mn in soil, mg/l (DTPA)

2000 =11-== um. .mommwo. Fig. 320. Regression of Mn content of MnNisigi,,:Jim iiiimalimi pot-grownwheat(y) onDTPA 1000I: extractable soil Mn corrected for pH 800.. armarem=s.73.32:10mrimaln (x), Philippines. 6001. WYPECIINVE...... H...... n. EllFM1151EIEWIWIR:MINFAMMIIImpil E 11=MENENIM" MIIIIMIMim3IMMIIIMMPANIIIIMIIMI a 400 1-; 300 cu _c 200molmommenlegoomin o C 100 80 bIONMEIPIIIMP" uoc 60 111 MMEMIMP I INEL 40 11Pr- 1, 30 . '11* 20

10111111111E1111111 8NO1111IMMUNIIMMTA=MMEMBEIMMIMIIIMMMMIMME IllhilEMMMIMMEN ummismimmonsos IhAIIIMMININ INIINIM=1=mimas 1 2 3 46 810 20 30 40 60 100 200 300600 pH - corrected Mn in soil, mg/l (DTPA)

two sample pairs (43684 from Cebo and 43686 from Bohol) the Zn contents were particularly high. Other high Zn values were distributed in various sites from Southern Mindanao to Ilocos. Unfortunately, the soils of eastern Mindoro where serious Zn deficiency in rice has been reported (FAO, 1980 a) were not sampled for this study. On the basis of the present analytical data, Zn deficiency is likely to occur in other parts of the Philippines though not as severely nor to such an extensive degree as in many other countries (e.g. Iraq, Turkey, India, Pakistan). See also Section 2.7. 269

Fig. 321. Regression of Mo content of 71eIVIIMMIIIIIIMINIIMIMINMEINE=Mwilikumigragn pot-grown wheat (y) on ammonium 6=1MIERIIIIIIIIIIIIIIIMMI111111 oxalate-oxalic acid extractable soil Mo 1METIMENIEEEMIZEMEINI

(x), Philippines. 3 y= r . 0. 6 2 c 111111111111 iME11111E1111.MIIIIMMIIIIIII mmmulumplEMISIIIMMEMINIMIIIMIMIU MEMNIIII!"-- MIIIIIIIIumandllIBP.'

a, .2MMLIERMINENII o o 1 riliiiiMilk wAMIIIIIHMEETir=v-MIIIIIIIIIMONNIMMIIMMEM .06...WIM1611111111111111MINEN mmE111111111111111111111111 .03Monsporminuu .02 111 01

.01 .02 .03 .06 .1 .2.3 .6 1 23 Mon soil, mg/l (A0-0A)

Fig. 322. Regression of Mo content of Mo111=SEMpirilrr===11111 pot-grown wheat (y) on pH-corrected 6MEIIIMENIINGEIKEEDMIMMUIll=MI 1=1MINSI A0-0AextractablesoilMo (x), 3 Philippines. 100Y.- r=0.66 2 gi 11111111111111111111111 =MMEMSIMmwEENS1==Weii MMEMENII1111IMMIMPII ISLMIM11111P4/1 MEMMINMMIENUINEM iniMI11:11 MEINMINNEEMM=MEMMINIMP,211111111111MIIIMERIERMINMNI

.1111111Militill11.1111M"1:21:1===1:11 EVM. emernawrIallormwmmEnnoSWILIa'rirPORIMINIIIIM= 11111111MMEAIROMMU MEMIIIII .03 .02

.01rill1M1111111111 .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH - corrected MO in soil, mg/l (A0-0A)

7.5.4 Summary

Wide variations in most of the general soil properties as well as in the soil macronutrient contents are typical of the Philippines. Of the six micronutrients included in this study B deficiency seems to constitute the most widespread micronutrient problem in the country. The level of Mo is effectively at a normal international level while that of Fe, Mn, Zn, and especially of Cu, are internationally at the highest level. As locations where soils are low in Zn, response to Zn application is likely. 270

20 1-4,...... Fig. 323. Regression of Zn content of pot-grown wheat (y) on DTPA extractable soil Zn (x), Philippines.

10 111, 111 9; inummwmosni 8 7 6) 5 111111111111111 111

4) inimmusenimmin

3 111111M11111M1 II

2 Milliffeal ll

1 prn.ILIIMMIIIMIMENIIIM limII_ UMMEN11111MMENIIIIIUNNIIII 111 1

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Znin soil, mg/I (DTPA)

200 Zn '1-s ± 2: 1 3 1 gi

100 Ai 90 eciAiw_ &i: iimismoino 80 11111ME 111111111111M1111/11 70 60 111111111111111 1111111 g_ 50 Inium_mplrilmi 30 minimmitim$iIII

20 4ci; 11111014Ili

10 1111111111111111111111 9 I =mosionTipl 8 7 Fig. 324. Regression of Zn content of 6 li I VIIII pot-grown wheat (y) on AAAc-EDTA .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 extractable soil Zn corrected for pH pH-corrected Zn in soi mg/I (AAAc-EDTA) (x), Philippines. 271 7.6 Sri Lanka

7.6.1 General

Of the 21 soils collected from Sri Lanka, 13 were classified as Luvisols and 8 as Acrisols. The latter soils occur south of the Puttalam-Batticaloa line and most of the Luvisols north of the line (Fig. 325). With three exceptions the soils are coarser textured than the average of soils in this study (Fig. 326). The range of variation in pH is wide but acid soils are

el at*

10 SRI LANKA

t4F5F9NA4 tlei 0 50 75 Q.)

VAVUNI VA O 45953

TRINCOMALEE

OANURADHAPURA

45954

PUTTALAM A 4T 45952

BATTICALOA

45959

45962

AD L'A

COLOMBO 45955

AMBANTOLA

MATARA

el 111.

Fig. 325. Sampling sites in Sri Lanka. 272 more common. A medium organic matter content is typical but the cation exchange capacity, electrical conductivity, CaCO3 equivalent value and sodium content are very low (Fig. 326 and Appendixes 3 and 4).

4 Sri Lanka internat. Sri Lanka imemat Fig.326. Frequency distributions of 21 n= 21 3764 co n. 3783 0=31 44 6= 5.62 6.64 texture, pH, organic carbon content, 30 os= 10 16 Os=.85 1.12 min.= 13 35 1010 .5 4.38 3.62 and cation exchange capacity in soils 1100.5 55 92 max = 7.60 8.56 36 of Sri Lanka (columns) Curves show

25 theinternationalfrequencyof the same characters. 20

/111 rls7.4 20 30 40 50 60 70 80 TEXTURE INDEX pH (Ca Cl2)

45 Sri Lanka 48 Internat. 35 Sri Lanka Internat. 40 n= 21 3779 ni= 21 3777 SS 1.2 1.3 7= 14.2 27.3 1.2 30 ts 4.3 14.6 t35 min.=.5 1 min.= 6.8 .2 max.= 3.9 39.1 max =24.6 99.9 "0 30

o25

LU o LU

U- 10

.2 .4 .8 1.6 3.2 6.4 108 25.6 16 32 ORGANICC,°!e CEC, me/100g

7.6.2 Macronutrients

The total nitrogen and NaHCO3 extractable phosphorus contents of soils and the N and P contents of the sampled maize crops tend to be somewhat on the low side in the "international N and P fields" (Figs 6 and 7). The variation in the N 'contents of maize is relatively wide (Fig. 327) which may partly be due to varying nitrogen fertilization (24 ± 26 kg N/ha). The relatively low P contents of soils (Fig. 328) are partly compensated for by phosphate applications (10 ± 9 kg P/ha). See also CEC in Fig. 326 and Fig. 14 in Section 2.2.4 and related text. Potassium applications to sampled maize crops (14 ± 12 kg K/ha, Appendix 5) are high compared to most other developing countries and in spite of the low level of the native soil K the K contents of maize are above the average international level (Figs 329 and 8). To some extent this contradiction may also be due to the relatively higher uptake of K by plants from acid than from neutral or alkaline soils. See Fig. 11 (Section 2.2.4) and related text and Fig. 326. Calcium and magnesium contents of Sri Lanka soils and maize are internationally fairly low (Figs 330, 331, 9 and 10). In all cases the variations are wide. At some sites a response to Mg could be expected. See also texture and CEC in Sri Lanka soils (Fig. 326) and Figs 12 and 14 and related text in Section 2.2.4. 273

40 Sri LankaInternat. Internat. Sri Lanka no 21 1967 35 n= 21 1958 Sri Lanka Internat. 30 = 3.5 3.1 0 = 2.82 3.11. n. 21 1967 es= 0.8 1.0 ±s=.76 .87 9. .32 .33 o min.= 2.0 0.6 a'0 min.= 1.56 .88 as..10r .10 L)25 max.= 4.8 6.7 max.= 4.17 6.51 min.../7 .05 nao.. .54 88f 1.04 o (T) a a 20 0>- W 15

w 10 cc 10 U_ U._ 5

2 6 2 3 4 5 6 7 N content of maize, % P content of maize,% K content of maize,% 40 70 Sri Lanka internat Sri Lanka Internat. Sri LankaInternat. 35 n.21 1967 3 R. 14.0 22.5 n= 21 1967 60 6= 21 1958 ±s. 12.3 33.0 0=141. 330 = .122 .135 min.= 2.2 0.1 es= 59 356 - 49 = .061 .088 30 non.- 48.4 656.0 mm.. 21 18 U 50 min.= .075 .008 max, 258 5598 max.= .363 1.657 25 a) CL CL 40 20

w 30 D 15 5

0 LU W o LI 10 CC U_

2.5 5 10 20 40 BO 160 320 640 25 50 100 200 400 BOO 1600 3200 6400 .005 .01 .02 .04 .08 .16 .32 64 1.28 P in maize soils, mg/I K in maize soils, mg/I N in maize soils, % Fig. 327. Nitrogen, Sri Lanka. Fig. 328. Phosphorus, Sri Lanka. Fig. 329. Potassium, Sri Lanka.

4'Sri Lanka Internat. 4 Figs 327-331. Frequency distribu- = 21 1967 Sri Lanka Internat. 34 5'.41 .47 n= 21 1967 tions of nitrogen, phosphorus, po- =.13 .20 2..231 .251 min.= .22 09 es=.098 .119 tassium, calcium and magnesium 30 max.= .70 1.813 33: min.=.106 .036 max...499 1.125 in original maize samples and re- ?Li_ 25 spectivesoils (columns) of Sri 0 20 Lanka. Curves show the inter-

15 coLU national frequency of the same LU u-cc o characters.

5 Ca content of maize.,% Mg content of maize,%

40 Sri Lanka Internat. 40 = 1967 21 Internat. R.1159 3450 Sri Lanka 35 21 es. 886 2815 35 n. 1967 min.= 300 10 no 161 446 max.= 3700 17995 es= 136 462 min.=35 1 max.= 574 6490

Affric 1600 3200 6400 0 100 200 400 800 1600 3200 6400 12800 2.5 25 50 100 00 400 600 Ca in maize soils, mg/I Mg in maize soils, mg/I Fig. 330. Calcium, Sri Lanka. Fig. 331. Magnesium, Sri Lanka. 274

100 'NWIMIIIIIIIIIII IIMM NM IN Ell II011MIIMENNIIIIIIMIIIIIMII MIN III IBM 0 11A111111120111111111411111111111111111111111111111111111111111M Fig. 332. Regression of B content of 80 .1iiii=1MOIM1Einiumes pot-grown wheat (y) on hot water 60 lE RIMMINEMIIM11=IIMINIMMIIII 50MMITIMMITWATETIIIIIIIMME111111 soluble soil B (x), Sri Lanka. For de- 40MI- - 0 32+ MIZEINIIIII MIMUM tails of summarized international back- MI!: 58 + 1 0 30 ground data,. see Chapter 4. 97 20 ..c II 1111111111111111111 1 IIMISSOMM=1MMINIMM, "5 8INNIIMimimiess=immin M11 6M111111=11EMPLIIIIN=M11111111eamme=mmomem.wimmen 511111111EIMMEN11/1 =MIMI 4IIIIIIMM1111111 INME1111 co 3 E1111111 2Ell 14111 1111111i1

ifi .3 .4 .5.6 .810 2 3 456 8 10 B in soil, mg/l (Hot w.

100 N1 1 L / t - 1 - Fig. 333. Regression of B content of 80 pot-grown wheat (y) on CEC-corrected 60 n=1 38 50 -±s= 31 078 6 9 4 (hot water soluble)soil B (x),Sri 40 2±s04024 0, 0 Lanka. E .30 y=355+76 = 8 5 0. r= 0 236s 0 2 20 a) 10 8 56 4 3 co 2

.08.1 .2.3 .4 .5.6 .8 1.0 23 4 5 6 8 10 CEC -corrected B in soil, mg/l (Hot w. sol.)

7.6.3 Micronutrients

Boron. A low B level is typical of soils and plants of Sri Lanka (Figs 332, 333, 22 and 25). Almost all B points are within the low B range (Zones I and II) or in the lower position of Zone III. No extreme values, however, were recorded. The sites of the lowest B values are in the more northern parts of t,he country. 275 22 Fig. 334. Regression of Cu content of nter not pot-grown wheat (y) on acid ammonium 20 acetate-EDTA extractable soil Cu (x), 751± 18 ± N7E3 Sri Lanka. E 816Illintli=011.9.2 1131111111111 o cy 141111111111111M1111111111M

.6 121111111111MM 1111111111111111 11111111111111 1111,111111111 111111111111111 ILIM11111111 1111111111111=2,111111111111111

41111111111111111211111111111111 21111111111111\011111111011% 1111111111111MISEENNIIIIIii .4.6 .810 2 46 810 20 40 6080 Cu in soil, mg/I (A A A c - EDTA)

Fig. 335. Regression of Cu content of pot-grown wheat (y) on organic carbon- 22Cu ' corrected(AAAc-EDTAextr.)soil Sri Lnit internt 2)n.48 n=358 Cu (x), Sri Lanka. I 2!s:7 1'±'17 2t..s: 7.5 0±57 1 i Al

3rc 3 illgiax II' 5)1 E 1 iti II ailig 1 min o 1

1i 1 11111 1110111111111 o 1 ill, 1 INININIIIIII ) 110I--Ipuraimaing o o ! uuuuiiiiiiiiiiiii : 11111111811111111111111111111 . _similimmuliniimmilli ) in 1mogulimish 1 .2 .4.6 .810 2 46810 2f1 a) ArtInn Org.C-corrected soil CU,mg/l (AAAc-EDTA)

Copper and Iron. Both Cu and Fe values determined from Sri Lanka soils and plants correspond to the average international level (Figs 27, 29 and 31). No extreme (Zone I or V) values were recorded for either of these elements (Figs 334, 335 and 336). Manganese. Only Zambia, Brazil and Malawi have national mean Mn contents at levels as high as those of Sri Lanka (Figs 33 and 37). In samples from three locations (45962, -67 and -71) exceptionally high Mn contents were measured (Figs 337 and 338), all having acid 276

3 00IMMIN111111011MUSIMIMMINII Fig. 336. Regression of Fe contents of 2 75 pot-grown wheat (y) acid ammonium 2 50 acetate-EDTA extractable soil Fe (x), 225 200 Sri Lanka. 1 75IM E 1 50

.!-:125 o -;100 90111111111111 80 70mommlnimmalmr- o o60Mminimismaind 50IIIIIM11111111 40rintimpi 30wow 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/1 (AAAc-EDTA)

soils with pH(CaC12) 5 or below. The possible Mn toxicity could be corrected by liming. Molybdenum. The results of plant Mo analyses and A0-0A extractable soil Mo are contradictory and lead to a non-significant negative correlation (r =-- 0.161 n.s., Fig. 339). Furthermore, the average Mo contents of Sri Lanka plants are about half the international average, but the A0-0A extractable soil Mo contents correspond to the international mean. When the effect of soil pH on A0-0A extraction is taken into account the correlation is improved to 0.650**, and at the same time the mean of soil Mo is reduced to half the international mean (Fig. 340). In the "international Mo field" (Fig. 20) Sri Lanka is clearly on the low side but even so, only one Mo point falls in Zone II, the others being within the normal range (Fig. 340). Zinc. Irrespective of the soil extraction method used for Zn, the limited number of samples from Sri Lanka indicates a relatively high Zn level in soils and plants (Figs 41 and 43). However, with the exception of the very high Zn contents measured in one sample pair from Kalutara, the Zn contents are mainly within the "normal" Zn range (Figs 341 and 342).

7.6.4 Summary

Considering the limited soil sample material of Sri Lanka, it may be concluded that typical soils are of relatively coarse texture, varying pH (mainly rather acid), of medium organic matter content and a low cation exchange capacity. The contents of macronutrients (N, P, K, Ca, Mg) are internationally fairly low. Of the six micronutrients, normal values were measured for Cu, Fe, Mo and Zn, but low B and high Mn levels were recorded. 277

Fig.337. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA extractable soil Mn (x), Sri Lanka. 1000 .VX-rWVIIM^11:111tIMMctilfEWOIMIIIMMIIMIMB!../MMI 800NM 111UNIIMBIfdl MI IIITEIIKEIRRiPg1401PZEIIM474111711111111MIPINIM=M 600 400 30011M1111111111=IMMEIIIMMI 200

100KIIMI=IMMIMML110IN.11111111E111111111111 11 80POIIIIIMMIMMIN=MMOBP/MEMENNOMWI1111=MMININIINNIMIA'=11111 60 11.11111.11.11MMIIIIMPME- MMMEMINII MM. 40IIIMMIIIROMIMMEM11111M 30Imommtioniumwounlim 20111111111111111111111111=

10IIMPMMUMMIIIIIIIMMEIESMIOMINEM11111111111 8HEMINNII=MBEIMMIIN1111111111111,11MIIIMIAMAIMMEN El N1 2 34 6810 20 30 40 6080100200 300 Mn in soil, mg/l IDTPAI

Fig.338. Regression of Mn content 2000 of pot-grown wheat (y) on DTPA M n extractable soil Mn corrected for pH 1000 osams-roulion P. (x), Sri Lanka. 800 E600 CL 400 "6 300 a) 200 o 00I.19111161111111 .111111011111 80AMMIMIMMMIMENMIMIEr:li.=1MMENNErMil=11MMENNIMMIMMEE: ni iiim.mminm-=mmilCommINKUI o 60IIinl 40llIMII11111111=M=IMENIIIIMEIMMI 2 30 20111111111111111.1111

10II 8ms immommuma-mordanrje. ==.9- 2 3 46 810 20 30 40 60 100 200 300600 pH-correctedMn in soil, mg/l (DTPA) 278

rzsmsmrson ommimmosimi Fig. 339. Regression of Mo content of momilitis 1111011Milsmelin MMEIMMINI MINMENHIMMI 61MR1113311INIIIMMMIMINUI pot-grown wheat (y) on ammonium EMMIIMISTMEE=M171111I oxalate-oxalic acid extractable soil Mo 3IIMMICEMEIREMILIZEN1111 (x), Sri Lanka.

2IMMIENNIEW, 22111111 111111111111111111 1IMNIMMMIIM11.11111111

C .2MIN1111111111111111111 o o .1A11=NES411NOMII=1=M11411 4=1111=MEMMEI nEINNOM 06wirAMmMENI1MnENNI

.03 I .02

.01

.01 .02 .03 .06 . .2 .3 .6 1 23 Mo in soil, mg/t (A0-0A)

NI oIMMEETNOMMEIHNIINIr1771 6 Illyn;s18.0.1.110 ilitst3s5= 07322 O.Mil 3 2 r = 0.650 NIMME611.MMiniuRy.'00°1r17.1 7uuiiuii IMM=/'M =IMIMEMEN

11111 .2 11111= Milli

.1 1111.11167VM,./PANNEIrirmlMIINEM IIIMIMPUMME11/.:41.

.03 Ilialt111111111111...111 .02 Fig. 340. Regression of Mo content of

.01 pot-grown wheat (y) on pH-corrected .006 .01 .02 .03 06 .1 .2.3 .6 1 23 A0-0AextractablesoilMo(x), pH-correctedMO insoil,mg/l (A0-0A) Sri Lanka. 279

Fig. 341. Regression of Zn content of 200 pot-grown wheat (y) on DTPA extract- Zn able soil Zn (x), Sri Lanka.

100 90 80 70 Q5E60

o20 o C Ni

10 9 8 7 6 III

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zn in soil,mg/I IDTPA)

200

, ". 100 111 111111111 90 simumsaaammimmiwzaiiv e* mossoni. 80 70IMMI1111111111111111111MIMIMIll 60411111m1.110111 g_ 50 40 o 301111111111111111UMOMIll o 204111111M110111111111 o

10 9 11111111ME1011 Fig. 342. Regression of Zn content of 8 pot-grownwheat(y)AAAc-EDTA 7 extractable soil Zn corrected for pH 6 EON .11111.11111 (x), Sri Lanka. .2.3 .4.6 .8 1 23 46 81020 3040 60 100 pH-corrected Zh in soi ,mg/1 AAAc-EDTA) 280 7.7 Thailand

7.7.1 General

The sample material received from Thailand consists of maize and respective soil samples only since wheat is not an important crop in this country. According to the geographical distribution of sampling sites, illustrated in Fig. 343, the central parts of the countryare

. 1

' 20. N...... -,,...

! .) k i cl"...

', HR. \ G X.1 798-803

3808-810

.1 --1 os ii,3o os ¡- 4.1=111111 4377°9- 7 ' 63780 ' 4378,060, i63818- 8,4 CE, ,..» .70, -704 IMON PAICXADIANI

A I gA7,4,,,. ' 2, N.11 0

'., er . ' 1 4353_ `s. 45761,43764 ',..f \ 4 3784-797 - o ANGKOK l'

t 1 S.- ? I\ \ Ao.oils t

;ay :6

o ' i'

... Ili

I A 8

THAILAND ,

"w . H°Ma. lo Fig. 343. Sampling sites in Thailand. 281 those best represented. Of the 150 soils sampled, 33 were classified as Rendzinas, 25 as Acrisols and 21 as Vertisols. The majority of Thai soils are fine textured, though many medium and a few coarse textured soils are included in the sample material (Fig. 344). The range of variation in soil pH is relatively wide but very extreme pH values are not represented and alkaline or slightly acid soils are most typical. Owing to the relatively high organic matter content and heavy texture, the national average cation exchange capacity of Thai soils is among the highest recorded in this study (Appendixes 3 and 4). The average electrical conductivity and sodium content are somewhat below the respective averages for all the international material. CaCO3 equivalent values for the majority of soils are low but a few quite high values were recorded.

Fig.344.Frequency distributions of 35 Thailand Internat. ThailandInternat. n.150 3764 40 n= 150 3783 texture, pH, organic carbon content, R. 56 44 =6.80 6.64 30 ts x14 16 =.80 1.12 and cation exchange capacity in soils of min.=12 9 35 min.= 4.50 3.62 max.= 85 92 max.= 7.70 8.56 Thailand (columns). Curves show the 2)8 international frequency of the same characters.

5 r,4 10 20 30 40 50 be 70 80 90 TEXTURE INDEX pH {Ca C12)

Thailand Internat. Thailand Internal

40 n.150 3779 n. 150 3777 7. 1.6 1.3 =41.8 27.3 es=.5 1.2 0 ts =17.0 14.6 min.=.7 .1 min, 11.8 .2 max.= 3.7 39.1 max.= 82.8 99.9

IT, 3 y 25 o ill20 o

U-10

.2 .4 .8 1.6 3.2 6 12.8 256 16 32 ORGANIC CA CEC, me/100g

7.7.2 Macronutrients

According to the information received from the sample collectors, no chemical fertilizers were applied to the sampled crops. On average, the total nitrogen contents of Thai soils and maize are slightly above their respective international averages (Figs 345 and 6). On average, the phosphorus and potassium contents of both the Thai soils and maize are a little low by international standards but vary substantially from one sample to another (Figs 7, 8, 346 and 347). At many locations a good response to applications of both P and K should be expected. The average calcium and magnesium contents of maize grown in Thailand are low compared to most other countries (Appendix 3, Figs 348 and 349) despite the relatively medium to high exchangeable Mg and high Ca levels found in Thai soils. The variations in values of both these elements are relatively wide. To obtain a general picture of the Ca and 282

40 ThailandInternat. Thailand Internet 1958 40 n 150 1967 35 n= 150 Thailand interne. 0=3.27 3.14 30 S. 3.0 3.1 n=150 n.1967 ss = 0.6 1.0 ±s= .52 .87 35 C R..29 = .33 min.= 1.3 0.6 min.= 1.52 .88 no= .07 s= .10 max.= 4.8 6.7 o30 max.= 4.63 6.51 4.23 min, .14 min.=.05 g25 max...47 max, 1.04 ai CT) 25 ai a- a 25 20 3- 3- 20 al 15 os o iy10 a. LL

3 4 5 6 .2 .3 .6 .7 .8 .9 2 3 4 5 6 7 N content of maize, % P content of maize,% K content of maize,% 40 Thailand Internat.

Thailand Internat. n= 150 1957 3 Thailand Internat. 7= 11.6 n= 150 1958 22.5 n= 150 1967 60 ns= 18.8 33.0 0.216 330 0= .157 .135 min,0.1 0.1 .... is= .047 .088 0 max..165.2 656.0 xs= 131 356 c min.= 40 18 ai min...076 .008 max...380 1.657 max.= 849 r 5598 25

620

1.11 o15

,ix1 A A .01 .//1". xiie 2.5 5 10 20 .01 .02 .04 08 .16 .32 40 BO 160 320 640 25 50 100 200 4008001600 3200 6400 N in maize soils, % P in maize soils, mg/I K in maize soils, mg/I Fig. 345. Nitrogen, Thailand. Fig. 346. Phosphorus, Thailand. Fig. 347. Potassium, Thailand.

40 Thailandini,not. Figs 345-349. Frequency distribu- Thailand Internat 1967 35 ..150 1967 7= .32 .47 n= 150 tions of nitrogen, phosphorus, po- es = .10 .20 "E. 7..182 .251 830 min.= .16 .09 .119 tassium, calcium and magnesium max, .64 1.88 0 30 min.= .081 .036 max.= .424 1.125 in original maize samples and re- g. 25 g.35 >- spective soils (columns) of Thai- 0 20 ° 20 land. Curves show the interna- 15 15 o 0. tionalfrequencyof the same io Cro 10 u_ LL characters.

.2 1.6 5 Co content of maize,% Mg content of maize, %

Thailand Interne n. 150 1967 ThailandInternat. 35 7= 6444 3450 Ls =4244 2815 n. 150 1967 min, 600 10 S. 440 446 max.. 15390 17995 Ls. 305 462 min.. 39 1 max,1762 6490

50 100 200400 8001600 3200 6400 12800 12.5 25 SO 100 200 400 BOO 1600 3200 6400 CO in maize soils, mg/I Mg in maize soils, mg/I Fig. 348. Calcium, Thailand. Fig. 349. Magnesium, Thailand. 283 Mg status of Thai plants and soils, see also Figs 9 and 10. The heavy texture and high CEC of the Thai soils (Fig. 344) may have contributed to the low uptake of Ca and Mg by plants. See Figs 12 and 14 and related text in Section 2.2.4.

7.7.3 Micronutrients

Boron. In the "international B fields" Thailand features among the low-B countries (Figs 22 and 25). Unlike most other nutrients the B contents of Thai soils and of the respective pot-grown wheats vary within narrow limits. Almost all plant and soil B values are lower than their respective international means. The correction of hot water extractable soil B values for CEC not only improves the plant Bsoil B correlation but also moves many of the samples from the "normal" range (Zone III) into the low B Zones (I and II, Figs 350

loo ao B n+150 60 s=407_ 82 8 50 s=042± 07 7 40 =404+0 x = 63+ 7 E30 =0 013ns 4 9:20 o .c ó 810 t6 o 4 co53 Fig. 350. Regression of B content of 2 pot-grown wheat (y) on hot water Ill soluble soil B (x), Thailand. For details 1 of summarizedinternationalback- .1 .2.3 .4.5.6 .810 2 3 456 8 10 ground data, see Chapter 4. B in soil, mg/I (Hot w. sol.)

100 80 n-10 5 60 Vis. 08 50 5c-±s =0 3 +014 3 72 40

E I x +5 50x o_ 30 log y=0 5+ lo 0. r-0223 ++ 20

081 0 C6 41) 5 t.4. o 4 o co 3 2

Fig. 351. Regression of B content of pot-grown wheat (y) on CEC-corrected .08.1 .2.3 .4 .5.6 .8 1.0 2 3 4 56 8 10 (hot water soluble) soil B (x), Thailand. CEC - corrected B in soil, mg /l (Hot w. sol.) 284 22 Fig. 352. Regression of Cu content of 20 pot-grown wheat (y) on acid ammonium 21111111'67311111111111111 acetate-EDTA extractable soil Cu (x), amoccuansumillim Thailand. 111111111111111111111111111111111E

z- 12

E,10I11111111111111111111111116111111111111111E1111111111122111111

81111111111111MIIIIIIIMI1111111 z 611111111111111111MINE111111111

411111110111MIMUNI111111111111111

2 111111111111111111MILIMIM ol .2 .4.6 .810 2 46 810 20 40 60 80 Cu in soil, mg/l (A AA c - EDTA)

Fig. 353. Regression of Cu content of 22Cu pot-grownwheat(y)onorganic Tha nd 20 111111111111111111111111111111 carbon-corrected (AAAc-EDTA extr.) 6_1 y s soil Cu (x), Thailand. 8 11111 4111114111111E11111111 09YIjII .17158161111111111 mulmommormanpil11111111111111M1111411111 impommilmowano

81111111111111IMAIK 11111111111

61111111111111111111411M11111111 11111111111541111111MENIIIIII

21111111:111-11111111111111111141111 111i111111110111112111311kikill .4 .6810 2 4 6810 20 40 60100 Org.C-corrected sotCU,mg/t (AAAc-EDTA)

and 351). In only a few countries is the relative frequency of low B (Zones I and II) values higher than that in Thailand. On the basis of these analytical results B deficiency, acute or latent, seems to be more likely in Thailand than in most other countries. Copper. The national average Cu contents of Thai soils and plants correspond closely to the international mean values and Thailand is located at the centres of the "international Cu fields" (Figs 27 and 29). Over 90 percent of the sample pairs are within the normal Cu range (Zone III) but there are a few rather high as well as low values (Figs 352 and 353). According to the present data, Cu deficiency does not seem to be common, but shortage of Cu may be a factor limiting normal crop growth in some localities. 285

Fig. 354. Regression of Fe contents of 300 pot-grown wheat (y) on acid ammonium 275 250FEJIIIIIIIIM111111111 acetate-EDTA extractable soil Fe (x), 225 Thailand. 200 y 175

E_ 1 50ktIli1111111111111111111111111 .125iiiiniiuiiiuiniiii o t 1001111111111M1111P1111111 17, 90 80INOWN111111111111110511111111111

-'11 70 u601111111111EME11111111111111

401111111111111111111111M1

30IIIAiiIIUIIIiiuiiI

10 20 40 60 100 200400600 1000 20000 0 Fe in soil, ,mg/1 AAc- EDTA)

Iron. The analytical data on Fe given in Figs 354 and 31 show that, in general, the Thai soils and plants correspond to the normal international levels and distributions for Fe. Although a few sample pairs are within the two low Zones (I and II) their locations do not indicate any severe Fe deficiency. Manganese. Thailand stands at the centres of the "international Mn fields" and its national mean values for plant and soil Mn deviate only slightly from the mean values of all the international material (Figs 33 and 37). The ranges of variation for both the plant and soil Mn are comparatively narrow with no low (Zones I and II) values among the Thai samples. Most of the high values are relatively close to the normal range (Figs 355 and 356). Only in one sample pair (43823) were exceptionally high Mn contents measured. These originated from a site in Pakchong with a heavy textured acid soil, TI 81, pH(CaC12) 4.5. Low pH is also typical of other Thai soils where high Mn contents were recorded. Molybdenum. The Thai national means for plant and soil Mo correspond well to those of the whole international material (Figs 15 and 20), although some relatively low and high Mo values were found among the samples (Figs 357 and 358). The present analytical data give no indication of any severe Mo problems in Thailand. Zinc. The Thai national mean values for plant and soil Zn are close to the international means (Figs 41 and 43), as are those of other micronutrients included in this study (B is an exception). In spite of the wide variations in both plant and soil Zn contents, the Thai samples do not occupy very extreme positions in the international scale (Figs 359 and 360). Low Zn values seem to be more typical for Sara Buri and Nakornsawan than the other areas sampled. There, and elsewhere in the country, a good response to Zn fertilization is likely. See also Section 2.7. 286

2000 7,---r...... irloomi Fig.355. Regression of Mn content mn Tn.,ht IL of pot-grown wheat (y) on DTPA 7±s 11 11 ,.: 10000===.2zawn...... ,parawan===I extractable soil Mn (x), Thailand. 800 600gonwrioNWirmitirrguramwarligrazirmameein11=1111111111MAIMIIIIIM 4001111111=MMINE1111=111111MINCEI1111111111111111111 cS: 300IINION.MMMENNUIMINKIiiiiiMiiiialIM=M111101111111 liffilhillillall 110 200911111111111111111 INIONIEN

10011E1 IN Ellere 80 0 60 1,14 40IMMINIMEIBM INIMMIIIIMIIMMEll 30IIIMMIENIIIMIMINIPPMS=1111111111111i P.50111111111111 li 201111112 INíO UIHIIUI

10..1.1...... 11 e...... 5.6 2 3 46 8 10 20 30/.0 60 80100 200 300 Mn in soil, mg/l (DTPA)

2000 ..msum mamsames T ola Fig.356. Regression of Mn content 15 of pot-grown wheat (y) on DTPA 1000 1 ± 7 extractable soil Mn corrected for pH 800 47 N (x), Thailand, E600- 111161 ka1.4B'9FINIMIN a.400 1111111111111111=11111 6.300 .c 200 o 100 a) 80 60 40 2 30 20

2 3 46 810 20 30 40 60 100200 300 '600 pH - corrected Mn in soil, mg/I (DrFA)

7.7.4 Summary

Most of the Thai soils are heavy to medium textured, alkaline or slightly acid, relatively high in organic matter and have a high cation exchange capacity. Compared to data obtained from other countries the average Ca content of Thai soils is quite high, N and Mg are at the average international levels and P and K somewhat below. Most of the micronutrient contents (Cu, Fe, Mn, Mo) analysed from the Thai samples represent normal values. At some locations a deficiency of Cu is likely. The most likely micronutrient problem in Thailand is that of B and Zn deficiency. 287

Fig.357. Regression of Mo content 7/11WRIMINFIEN=11MMEnIPIIIMEN of pot-grown wheat (y) on ammonium 6 111111=E1111 oxalate-oxalic acid extractable soil Mo 3MMN` (x), Thailand. 11 < 2MIMESIS. AIM

MMMMUIIIMMESEFMNIEMI1111111111 monommommir C .2

O o 1101111111111111 o .1worAmoni==mgm=lvaik.Aállomosomodniymmm MMI111111111UNINIM .06WA=11104111=1=IMINEMIIIII

.03 .02

11 .01

.01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/ (A0-0A)

EN oEN M.111M=M11111111 6mEnsolacimmommulusimaimummilIIMIIMINI11111MIMI

3MMA . 0MEDIUM "1:1221311E11111 21111111°90Y67011112111hZolitilliiiiil

IIIIIII 112=11tIMMINEINII ...... IMMIMMEMP!RW =IMMIN1MILIMICIIIII=MMEMME=MMEN1.=.11115111MVMI=MMIUMERII wommmmossiummempapsfat milommwounumisordeolum imomm-m.,...... =msirimoitterionni IMIMMIMO.1Pr.,111MM11111O11111MMIMEMINIBImommramimirealww=1Humm 1.....1114r7"INIMMEMINIW.40MMEMENI1MMEMMI 1111111111====MIll

.03 .02 Fig. 358. Regression of Mo content of pot-grown wheat (y) on pH-corrected .01 A0-0A extractable soil Mo (x), Thai- .006 .01 .02 .03 06 . .2.3 .6 1 23 pH-corrected MO in soil,mg/I (.40-0A) land. 288

200 BM =IIMIMMIO IIIMMINIMMUM Fig.359. Regression of Zn content of pot-grown wheat (y)on DTPA extractable soil Zn (x), Thailand. 100 :11 90 80 INN/ 70IIIIMMIN11111=1111MIIIII E60 cl 50111111111111IMMIIIIEM 40 misenuommelm 48 30iiiamomobinti I

2011111111111111111111 I o N 10MI11111191111AAR k 8ligrArmium=11111111i.. 7lummEOLNE11111IMMIIIII=ILIIIIMA 61111111111M111111=1M111111ZEMINII IMEINIVE111111M x .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zrt in soil, mg/t (DTPA)

200 I

100 90 80 70 60 E 50 0. onomdiarta 40 o 30magoperi o 20wpm I o

10 9 8 7 Fig. 360. Regression of Zn content of 6nommingiti pot-grown wheat (y) on AAAc-EDTA .2.3 .4.6 .81 2 36 6 810 20 3040 60100 extractable soil Zn corrected for pH pH-corrected Zn in soil, mg/1 (AAAc-EDTA) (x), Thailand. 289 8. Near East

8.1 Egypt

8.1.1 Genera/

The Egyptian sample material consists of 200 plant-soil sample pairs, half of which (100) were collected from wheat fields and half from maize fields. The locations of sampling sites illustrated in Fig. 361 show that the most important agricultural areas of the country,

52. 33. 30. 33°

..267 214 --- _.,-.-" 2 _----- 2227 4 PORT 9,9,, ',¡'2% P097 sely eALEXANDR 228 ._,._,...... 1tr, AL6S3.7,,A..4 252 6 201 .202 31. 9 1732HU G m.2990R2 ,7.0411410V 4,94V94 . 226A 200x 19,6 1 7 34A 227. 22 t sale. 194A " ..32G 54*563 235*226. 1,3 .IL 19 161 321.,. .1o 28.0 za 33 2 L1A 747 7 282 L AM O ?67'2. 334- 24 4° 336 .6 236. 24725 338 .2 .-M.4 3 4. 2 i . 4, 1.1 24;7-7-A I9G1152

174* CAIRO 331 CAIRO 30 SU 2 30. !IA 24

2,54* Is

.. ..-i72. 4"..-- 25.5 . ELrm EL FA1VUM .70 04 o y ,A17 \ 1 6 156.11 BENI SURF' \ 29 29 164 \ .3 .I 26I 0

262 \ \..., I 160A

159. 264.. 265 66. 28 MA . I- 26 a 67A I 266. i i56A 269* 27 155,, AS UT 97117 27 174. I- .1149---- 1614

SOHAS'P, $17543.

464 A

25. NAG8 MAO NAG14 1911401 26

LU OR 1.1809

25. ARAB REPUB IC , 19 V 6 OF EGYPT (Nile district)

a la MI 26 a 120 va 26949

33W 94VAN 24.

30. Sr 50. 31. 32.

Fig. 361. Sampling sites in the Egypt (maize fields on the left and wheat fields on the right). The last three numerals of each sample pair number (45151-45350) are given. 290

the Nile valley and delta, are well covered. Almost all (197) the sampled soilswere classified as Fluvisols; the remaining three soils as Xerosols. The range of textural variation of the soils is very wide but the great majority is fine textured (Fig. 362). About 85 percent have a texture index exceeding the international average of soils in this study. Unlike texture, the variation of soil pH(CaC12) is very limited, from 7.4 to 8.2 only, and the national mean (7.74) isone of the highest recorded in this study. The organic matter contents correspond to theaverage international level and vary within relatively narrow limits The national mean cation exchange capacity is one of the highest among the 30 countries studied, and in only a few soils were low CEC values recor.ded. The values of CaCO3 equivalent are usually at theaverage international level but those of electrical conductivity are very high and theaverage sodium contents are higher than in any other country (Appendixes 2-4).

45 35 EgyptInternat, EgyptInternat. 7 Fig.362. 5=200 3764 Frequency distributions of 40 n= 200 3783 7= 58 44 =7.74 6.64 30 texture', pH, organic carbon content, 5,5= 14 16 =.14 1.12 min.. 15 9 3 min.= 7.38 3.62 max.=67 92 max.= 8.20 8.56 and cation exchange capacity in soils of 25 30 Egypt (columns).Curves show the 20 25 international frequency of the same

20 characters.

5 /0 5. 4 4. r fr

Air 4.55.74 .4 , /I 10 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca CIO

45 Egypt Internat. EgyptInternat. 40 n = 200 3779 n=200 3777 7 =1.2 1.3 0=43.1 27.3 58 = .3 1.2 ±5=10.5 14.6 "E35 .1 min.= 5.9 .2 o max.= 1.8 39.1 max.=58.5 99.9 30 [1. 25 w 20 o Lu 15 co U_

.2 .4 .8 1.6 3.2 6.4 12.8 25.6 8 /6 32 ORGANIC C,% CEC, me/100g

8.1.2 Macronutrients

The nitrogen contents of original Egyptian plants as well as the total N contents of soils are of average international level (Figs 6 and 363). The variations, especially of soil N, are narrow. Nitrogen fertilizers were applied at high rates to both sampled crops (wheat 129 ± 31 and maize 161 ± 39 kg N/ha). The phosphorus contents of both plants as well as the NaHCO3 extractable P contents of the respective soils tend to be slightly on the low side in the context of the whole international material in this study (Figs 7 and 364). The P valuesvary relatively more 291

Figs 363-367. Frequency distributions Egypt Internat. Egypt Internat. n= 100 1765 1958 35 n= 100 o,o, of nitrogen, phosphorus, potassium, 7. 4.31 4.27 0=3.02 3,14 es=.58 1.15 es =.45 .87 calcium and magnesium inoriginal min.= 2.81 .60 30 min.= 1.72 .88 max.= 5.76 7.45 max.= 3.83 6.51 wheat and maize samples and respect- 25

ive soils (columns) of Egypt. Curves 20 show the international frequency of the same characters. 10

4 5 6 3 4 5 6 N content of wheat.% N content of maize../0

EgyptInternat. Eg ypt Internet n= 100 1765 n= 100 1958 60 2.132 .133 60 5= .117 .135 Os =.030 .084 os= .028 .088 min.=.061 .009 min. .044 .008 max.=.221 1.023 50 max.= .208 1.657

/ 005 0 02 .0 .08 16 2 .64 1.28 .005 .01 .02 .04 .08 16 .32 LL 728 Fig. 363. Nitrogen, Egypt. N in wheat soils, °/. N in maize soils,%

Fig. 364. Phosphorus, Egypt. Egypt Internat.

35 n=100 n=1765 8= .36 4= .38 rE as.= .07 Is= .12 LS'30 min. .05 mox.= .58 max.= 1.02 0-25 >- zU 20 Lt1 015 L11 10

P content of wheat,% P content of maize,%

40 Egypt Internat. Egypt Internat. 5 5. 100 1765 35 n= 100 1967 R. 78.7 20.2 R. 75.8 22.5 as= 13.8 24.7 as = 13.4 33.0 min.=4.2 1.0 min.=3.4 0.1 30 max.= 85.8 271.4 30 max=101.1 656.0

25 a. o>-20

L11 oD 15

2.5 5 10 20 40 60 160 320 640 2.5 5 10 20 40 80 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I than those of N but no extremes were measured. The dressings of P applied in fertilizers were small (wheat: 9 ± 7 and maize: 2 ± 6 kg P/ha). The average CH3COONH4 exchangeable potassium contents of Egyptian wheat and maize soils are about twice as high as the respective international averages (Fig. 365). Only in a few other countries were such high soil K contents recorded (Fig. 8). Although no potassium fertilizers were applied to ihe sampled wheat and maize crops, their K 292

EgYPt Internat Egypt Internal Fig. 365. Potassium, Egypt. n = 103 1762 n.109 1967 / 4.0 4.0 39- 0: 3.3 2 3 05/. 0.5 1.6 to / 0.5 nIls./ 3.0 0.9 06 MO% r.5.8 6.8 25 non -4,8 67

5 7 K content of wheat,% K content of mciize,%

Egypt 1nternat, 0 Egypt Intemot. 5= 10C, 1765 n. 106 1967 7. 722 265 /630 33C tor 308 283 sc. 306 356 rnio, 228 20 82 18 1085. 2597 1297 5000.. 2325 5598

25 50 190 220 6 08 U 1600 32005400 SO 100 250 190 8 5 600 3200 6400 K 'in wheat soils, mg/I K in maize soils, mg/I

co Egypt Inte,flet Fig. 366. Calcium, Egypt. 35 35 1,106 ns 1567 n .47 .06 8s=.23 '13 30 30 = .12 min...93 ma., .53 rnox 1.08

tal. 25

20 22 n15 tsE

Co content of wheat% Ca content of maize,%

0 0 Egypt ,nternot. Egypt nth-not n 100 1765 /100 1907 9.7063 4671 6.8348 3450 60. 1319 3070 985 285 min.. 1640 110 5115..3452 10 30 mox./ 9920 21930 max .8998 17995

40:1 75

LO.

10 00 400 8901600 3200 6400 12000 25500 50 102 200 400 837 1600 3200 6402 12820 Ca in wheat soils mg/I Co in maize soils, mg/, contents correspond well to the average international level. The soils of Egypt are generally calcareous. Their average exchangeable calcium content is higher than that of most other countries but the Ca contents of both plants are slightly lower than the respective international averages (Figs 9 and 366). 293

40 Fig. 367.Magnesium, Egypt. Egypt Interne-, Egypt 151.0301 35 n= 100 17E4 35 r, too 1957 0..184 ir" .172 14,253 .251 -.. . 1:7,031 .060 te=.071 .179 4.93 36 min -.139 .044 030 max.= .280 .946 08,691 1.37$ 2G 25 >- 0-) 20 20

411 o11, CC 10 1-14

51-

05 Mg con ent of wheat.% Mg content ot maize,% M Egypt Internat. Egypt intvnai n= 190 1734 n. 103 1507 2=1580 489 2=1206 046 tar 525 437 f-5, 372 462 min.= 395 10 min.= 303 max.= 2946 3296 max =1996 1,090

443 2

0>- '5

25 50 100 200 400000 1504 9200 12.5 25 50 100 260 000 9 01400 3200 5400 Mg in wheat soils. mg/I Mg in maize soils. mg/I

In the case of magnesium the situation is similar to Ca. The average contents of plants are only slightly above the respective international mean values although the national averages for exchangeable soil Mg are higher than in any other country (Figs 10 and 367). Obviously, the generally heavy texture and high CEC of Egyptian soils (Fig. 362) have lowered the uptake of Ca and Mg by plants in relation to the exchangeable Ca and Mg of the soils. See Figs 12*and 14 and related text in Section 2.2.4. 294 8.1.3 Micronutrients

Boron. The B levels of Egyptian soils and pot-grown plants seem quite normal. In the "international B fields" (Figs 22 and 25) Egypt stands near their centres. There were no values indicating severe B deficiency but some were so low as to suggest at least the possibility of hidden 13 deficiency (Figs 368 and 369). Although many Egyptian samples showed an abundance of B, none of the B values was extraordinarily high. According to

100 _____... 80B IliiiIMINIMMINIII11111111 n =18 ' ItEREMMUMMIEMMI 60EMEMEMMETMEMIMI NIMMEIMIIII 50 O; MU"0.53 ± DEREIIII MINI Hoc + 2.44x 30IllirlindifIntillIMIN INIA a_ 20 Kum "E) -cG) illimplim 10il...,...... d...... immirmmoiromp...riiiimm.....p.r.s.-1.1..ma....-.Alkummomias "E" 6NossommNinotlitaraimmilimmumeil 5111MIEIIIIIIMPIMilliEN=IZMIIIIIIII 4MINI NiplEraramiiiiiMENIMUI m 31111111111.11.4111111 illilim1111111 101111111 2111 Pil Fig. 368. Regression of B content of MI pot-grown wheat(y) on hot water 111111 1 soluble soil B (x), Egypt. For details .2.3 .4.5.6 .810 2 3 456 8 10 of summarizedinternationalback- B in soit, mg/l /Hot w. sot/ ground data, see Chapter 4.

100 80 loritiMmairra 1mligrall 60 IMIOMMagnilill 50 4097-"triail Ila E a_ 30Ill IMMIE EME- Rohm o. -6: 2011111 ' °"' i II_ A0111111 (a) _c 08...... =ErairmPill 111111.111111111 6Illat..miromi-,YdmirquipmEm ..111.-wm .traiSUMMINE1111111 Ili MI 5NIMMIMMOYAIOSIMIGMEMITIOUN o 4111111M1M01111111 IE PM 2NM MNIIIIIII Ilk 1111E1

11111 EMIIlk DI 1118I. Fig. 369. Regression of B content of .08.1 .2.3 .4 .5.6 .8 1.0 23 4 56 8 10 pot-grown wheat (y) on CEC-corrected CEC -corrected Bin soil, mg/1 (Hot w sot) (hot water soluble) soil B (x), Egypt. 295 Elseewi and Emalky (1979) much of the water soluble B in Egyptian soils comes from Nile water used for irrigation. They also considered the B contents of soils sufficient but incidences of toxicity unlikely. Copper. Relatively high Cu contents are typical of Egyptian soils and plants (Figs 27, 29, 370 and 371), and about 30 % of the Cu values are within the two highest Cu Zones (IV and V) indicating an abundance but not likely toxicity of Cu. Deficiency of Cu seems unlikely in Egypt. Iron. According to the present analytical data (Figs 31 and 372) no severe Fe problems are to be expected in Egypt.

20Elli1/1111 '1111111 MI 15g1111111 18WE! E togx 1111111,1111:e allilli olI ELI 11111 111111111 PI 1111 11111 II 11111111/11 111E111111 II BM 8811111111111 U IIIiIIUl 86611101111M 11111111L111 411111111Walli IIIIIII Fig. 370. Regression of Cu content of 2IliellitillM IIII Min: pot-grown wheat (y) on acid ammonium III 111111111SIMPILL IN o i,4 /Co An -In IA anon acetate-EDTA extractable soil Cu (x), ir. . Egypt. Cu in soil, mg/l (A AA c - EDTA

22ICU Egy Inter t 20 11 18fir11:7V1;11 i'i'-'8,0, Fi. log y.0030 301io2 , logy.11587O. 5E 1.9 . r=0575'' r.0 7 1`" Fr 16 a_

t's' 1

1 I 10 L .1110e'l>4111 o o EINInighlom II opimmimmozoin Fig. 371. Regression of Cu content of Immigi umil Noll, , ii. N-. pot-grownwheat(y)onorganic ...... carbon-corrected (AAAc-EDTA extr.) al .4.6 810 2 46 810 20 40 60100 soil Cu (x), Egypt. Org.C-corrected soi CU,mg/i (AAAc-EDTA) 296

300 Fig.372. Regression of Fe contents 275IMINIEM11111111IIMMISIMMINIS 2 50manIIIMI11111111111111111111111111111111211 of pot-grownwheat(y)onacid 225IMINERIIMISIMM111111111111111 ammonium acetate-EDTA extractable 200 soil Fe (x), Egypt. 1 75 E 1 50 Q- .125iiiuuiiiiiióiiiiinhiii o 1 0010111111111111111111 90maILIMINUMMI61111111=1111 80 c 70 8 60INIIIIIIMMON111111111 50

3011111111111111111111 10 20 40 60 100200400600 IWO 2000 Fe in soil,rng/l AAc- EDTA)

Manganese. Unlike most other nutrients the level of available Mn in Egypt is low (Figs 33, 37, 373 and 374), which is a natural consequence of the high alkalinity of soils. For the same reason the ranges of variation in plant and soil Mn contents are exceptionally narrow. All plant Mn contents, as well as pH-corrected DTPA extractable soil Mn contents, are below the respective international means and about 20 % of the sample pairs fall in the two lowest Mn Zones (I and II). Problems due to the small amount of available Mn may arise at several locations. Molybdenum. The high level and limited variation of plant and soil Mo contents in the Egyptian sample material are due to the high pH of Egyptian soils. The apparent contradiction between high Mo contents of plants and low A0-0A extractable soil Mo contents (Fig. 375) is eliminated by pH-correction (Fig. 376). In spite of the generally high Mo level only a couple of plant-soil sample pairs fall in the highest Mo Zone (V). On the basis of these data Mo problems in Egypt are unlikely. Zinc. Whichever extraction method is used for determining soil Zn contents, the Egyptian Zn levels are very similar. The national mean values for plant and soil Zn are internationally somewhat low (Figs 41 and 43), the variation ranges are relatively narrow and, therefore, only few Zn values fall outside the normal Zn range (Zone III; Figs 377 and 378). Many of the plant Zn contents, however, are unusually low and indicate lower Zn availability than do the soil analyses. At these sites response to Zn may be obtained. See also Section 2.7 where the critical limits for Zn are discussed. 297

Fig.373. Regression of Mn content 2000 MII=SIMMEMoulmen.mo.now vlowymol ed....ra.*/, of pot-grown wheat(y) on DTPA extractable soil Mn (x), Egypt. 1000 _,111MIIIMIR:r=i1=1111111111 80012=8.41MITIMIM7ITMr 600 I.

aE 360000IIIuIIILUIIuIU N. MIMIIIIIMIMENIQUINI11.111 g 200=poll

-6 100fin4.1110160.," 80 111111111111111111117111111.11IMUMMig -?" 601 MIN ImummerrasTamimmunilusim 6011111111111111111111ROMIZMAMMMIIIIMMIIME xc 30 20111111101111111111=11111111=111

10111111111111=111111111 MNNA.MMEM11.1401101.0111110MEN1111111&MMMINEIMMI MIMEOS RIM MIMPWE CIINIPIMINIMMIIIENIMMMI111 811111000111111111.111=1111114111UNNOMMERNISTIAMMli 11111=111111101111111111111.111110 2 34 6810 20 30 40 60 80100200 300 Mn in soil. mg/l OTP.41

Fig.374. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA t 1 extractable soil Mn corrected for pH 1000Iiii17111M Iii11111111 800gi M5116*)1C121=1El*Cal.1==sizig: (x), Egypt. II.. E600 a. a- 400 k7s212-IfimmowSiZigigidi "ti3001 immminumommann. 1111 1 I 200 I EMI IlI o 100 G) 1.---16,...,-1------lw-mium.,--,-m-nik 80Il UI 60 -rdow------aummiranIIIMMIENE 40fl111 EpiApam=mwW - nsi a 3011 antiv,.. II 20 kiila-.. N

10n=0 rdsdra=kmagill FiraE=91I 8 I il.mnui isanniZZIFILemeslumemomedSal 2 3 46 810 20 30 40 60 100200 300600 pH- correctedMI1 in soil, mg/I (DTPA)

8.1.4 Summary

Typical Egyptian soils are fine textured, highly alkaline, have a medium organic matter content and high cation exchange capacity. The macronutrient content of soilsis generally good. The national average of P is slightly below the international P mean, but K, Ca, and Mg contents of Egyptian soils are higher than in most other countries. Micronutrient problems concerning Cu, Fe and Mo are unlikely but occasionally responses to B, Mn and Zn can be expected. 298

Fig. 375. Regression of Mo content of 6 pot-grown wheat (y) on ammonium 77 .32 ± am 3 oxalate-oxalic acid extractable soil Mo .. 1.. III . 1 3 (x), Egypt. :5. kilik : 2 Ill E

1 IIIIIEI oo_ ....urremovmmasmCONIIE1.116W. a) moniororg:massim IMMORINIEMMKEMINIIIPOOMIN11-_- Igloo.111111111111E o o . 1 MIMMEEMBIIIIMM .06 imaymegims.9 olo .03 .02 III El .01

.01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/l (,40-0A)

M oIIMMIIliMMAIIIIMI117111717111=1=M1111 6MENNEIIIIINIMMMS11111311131=1111somimm=mmin

3NEM= 03.noinconstom 2111111Y, °°1:11,6111111111:ffidiliiiil E 1111111.1111111111541111111 1 wiminms=mi nteamom.R. ==1111==1WAR1501111:11nimusannowmokenuiaminmr .5.6MIMMEMEIMMIMMEMENIMlittg5rd.101MI .c Immossolmmmonolmmairolitramsi MONIIIIMMEMMENNIIIIMINIFAMIEM11 .3

77, .211111111/1/0111110MINIIIIIM. "E.o o .111=11MMENLME PIMMEINE=11=MMINNIII 2o =MIMENI0.1LIIMSIMMENNOImomml=milmn=nns .06=MMINZIIIMIETIMMEIMMEIN111EN

.03 .02 Fig. 376. Regression of Mo content of .01 .006 .01 .02 .03 06 . .2.3 .6 1 23 pot-grown wheat (y) on pH-corrected pH -corrected MO in soil,mg/1 (A0-0A) A0-0A extractable soil Mo (x), Egypt. 299

Fig. 377. Regression of Zn content of 200 pot-grown wheat (y) on DTPA extract- 77 5 able soil Zn (x), Egypt. 321Iag < 100IznNEE 90 80 70 E 60I mom 50 o 40 mommor- 30

201111111111E o N 101511111 9 8 7 6 lu EL

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zn in soil, mg /I (DTPA)

20 .t Zn En:93 1313te3 1 n .7c±31t26 at

y=.C-3 '73/to L108 S . 1 u. / I ill iii 1 rI

1 k

r I 1111111 1 liii .111111rii-,. I 110,11, 1 ...111ILII , ----11PA...... ,. , Fig. 378. Regression of Zn content of 1 1 , pot-grown wheat (y) on AAAc-EDTA H. WI I extractable soil Zn corrected for pH (x), Egypt. pH-corrected Zn in soil, mg/1 (AAAc-EDTA) 300 8.2 Iraq

8.2.1 General

The sample material from Iraq consists of 150 plantsoil sample pairs, of which 119 were collected from wheat fields and 31 from maize fields. The sampling sites are illustrated in Fig. 379 and cover the agriculturally important areas of the country. The Western Desert and Jazirah regions are excluded. One-third of the sampled crops had been grown under rainfed conditions in the northern part of the country where the annual precipitation usually exceeds 400 mm, two-thirds were irrigated and came mainly from the Mesopotamian plain. The Iraqisoils were classified as follows: Halosols(all Solonchaks, 45 soils), Xerosols (35), Fluvisols (31), Vertisols (30) and Lithos°ls (8). Most soils are fine to

' 4134 ! /50a 555.'!.45;:67 90 I 4 522 i 48 571 k..., 4:: 59:\ r40..55 . ', 56 5'2 52.6.0,_____' \ 552. 5 5 526 .528 401.. 5:6 568 4.2506,1 o t 0--. SULTAIMANITA 556. KIRKt., 55.5,S) .98 592* 500 i 580 ,, I, 5 587. ,....) I 494 .91 596 (s I . 5 .. .581

.5 54* 166:518,0 ' tiA0.5A" 584\ 4 6 .4 .555 RAMAN t, 2 BAGHDAD

0 RUTBA 1.512 , 5244

606 A 455 46 . 45IB u5 'II.

46 542 I 4

i I It 470 .6, 501 5A, ±....___I

-\ ) 4 a? 42 O6 --A&RA .2476 549 540. i 23,524.42 5 6 i 545 0548 S8 i I

7. 542 i I .ADDASIRD'A 5' I-IN. .535 I 6 -. 53:

i IRAQ i .

...

NEVTRAL 20

0 AO or, 120 ma no Ft.

.21. o an

Fig. 379. Sampling sites in Iraq. The last three numerals of each sample pair number are given (dots = wheat fields, triangles = maize fields). 301 medium textured and all are alkaline; the national average pH is one of the highest (Fig. 380). The organic matter contents are low and the cation exchange capacity is at the average international level. The national mean value for electrical conductivity is the highest and those for CaCO3 equivalent and sodium content the second highest recorded for countries participating in this study (Appendix 4).

380. IraqInternat. Iraq Internat. Fig. Frequency distributions of n=150 3764 11= 150 3783 A= 51 44 A= 7.72 6.64 texture, pH, organic carbon content, 30 Asr. 10 16 As=.16 1.12 min.= 18 9 min.= 7.22 3.62 and cation exchange capacity in soils max.= 74 92 max.= 8.55 8.56 of Iraq (columns). Curves show the international frequency of the same characters.

10 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

Iraq Internat. Iraqinte,not 40 n=150 3779 n= 150 3777 k=.8 1.3 A= 26.2 27.3 As=.3 1.2 Is= 7.1 14.6 "E" 35 min.= 11.0 .2 max.= 1.8 39.1 max.= 43.8 99.9 " . 2 O6 25 w 20 o7 LU 15 CA U- 0

.2 .4 .8 1.6 3.2 6.4 12.8 25.6 16 32 ORGANICC.% CEC, me/100g

8.2.2 Macronutrients

The total nitrogen contents of Iraqi soils are quite low (Figs 6 and 381). Low N contents are typical of Iraqi wheats but for maize the national mean equals the international mean. This may be due to nitrogen fertilization; the sampled wheats had only received 5 ± 9 kg N/ha while the maize crops were fertilized with 27 ± 24 kg N/ha. The national average phosphorus contents of both plants and soils are among the lowest of this study (Figs 7 and 382). Wheat crops were fertilized with only nominal amounts of phosphates (2 ± 4 kg P/ha) but maize received somewhat more (10 ± 10 kg P/ha). A good response to phosphorus fertilizers could be expected at most locations examined, provided there are no other soil factors limiting the crop growth. The potassium contents of Iraqi soils as well as the K contents of both indicator plants seem to correspond to the average international level (Figs 8 and 383). Potassium fertilizers were applied only occasionally (wheat 1 ± 3 and maize 0 ± 1 kg K/ha). Both the exchangeable calcium and magnesium contents of Iraqi soils are high but the Ca and Mg contents of plants con-espond more or less to the average international level (Figs 9, 10, 384 and 385). 302

Iraq internat. Iraq Intermit Figs 381-385. Frequency distributions 8= 119 1765 35 n= 31 1958 4.27 9=3.14 3.14 of nitrogen, phosphorus, potassium, ns= .79 1.15 Is= .40 .87 30 min.= 1.90 .80 min.. 2.50 .88 calcium and magnesiuminoriginal max.= 5.16 7,45 max.= 4.12 6.51 wheat and maize samples and' respect-

6- 20 ivesoils (columns) of Iraq. Curves show the international frequency of the ELI15 o same characters. Ui 0 CC

2 3 4 5 2 3 N content of wheat, % N con ent of maize.%

Iraq Interno) Iraq Internat. n = 119 1765 nt 31 1958 60 7,097 .133 9=087 .135 = .035 .084 16..032 .088 .009 min...041 .008 rnax.= .239 1.023 50 max.= .197 1.657

.005 .01 02 .04 .08 .16 .32 64 1.28 .005 01 .02 .04 .08 16 .32 .64 1.28 N in wheat soils. % N in maize soils,% Fig. 381. Nitrogen, Iraq.

40 Iraq internat. Iraq Internat. Fig. 382. Phosphorus, Iraq. II=31 n =1967 35 n. 119 1765 35 = .30 .38 R..28 R..33 ts . .08 .12 ts = .06 ss. .10 3 min.. .17 .05 0 min.= .17 min.=.05 max.= .52 1.02 max = .41 max.= 1.04

a-25 25 >- (2-) 20 20

E 0

.8 .9 .2 .3 .< .5 .6 .7 .8 9 P content of wheat, % P content of maize,%

Iraq Internat. Iraq Internat. n. 119 1765 n . 31 1967 35 R. 8.3 20.2 5 = 7.5 22.5 Is= 10.2 26.7 ts =4.6 33.0 min.. 1.2 1.0 0.1 max, 69.7 271.4 max.= 20.9 656.0 30 C

?;._3 25

>- 20

11J on 15

u_ 10

25 5 10 20 40 80 160 320 640 2.5 5 10 20 40 80 160 320 640 P in wheat soils, mg/I P in maize soils. mg/I

Of the 119 wheat samples, 88 were classified as HYV ('Mexipak') and 30 as local varieties. This last group consisted of varieties 'Sabir Biek' (11 samples), unclassified, marked 'Local' (9), 'Harama' (2), 'Moseele' (2), 'Bakra Toe' (1), 'Rashboal' (1) and four samples called 'Italian'. The contents of all macronutrients in the HYV wheat were somewhat higher than of the local varieties. However, this comparison is of limited value because in most cases the HYV had either grown on soils richer in macronutrients or had 303

Fig. 383. Potassium, Iraq. IraqInternat. IraqInternat. n=119 1765 rm 31 1967 30 R= 3.5 4.0 30 R = 3.5 3.1 As= 0.6 1.0 As= 0.4 1.0 ' min = 1.8 0.9 min.= 2.3 0.6 6.7 (,13 25 max.= 4.8 6.8 25 max.= 4.5

a 20 o>- W 15

ix 10 U-

1 2 3 2 3 4 5 6 K content of wheat,°/0 K content of maize,°/0

Iraq Internat. Iraq Internat. n= 119 1765 ns31 1967 Os 341 365 R = 304 330 30 Ss= 141 283 As= 126 356 min.= 130 20 min.= 120 18 max.=1028 2097 max.= 645 5598 25

>. 20

25 50 120 200 400 800 1600 3200 6400 25 50 100 200 400 BOO 1600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

40 40 Fig. 384. Calcium, Iraq. Iraq Internat. Iraq Internat. n = 31 1967 35 n =119 1765 5 8= .37 .43 7.48 .47 .E. *s = 09 .17 35 ,08 .20 S, 2, min.= .19 .11 30 min.=.34 .09 mas.. .78 1.68 mox...65 1.138 k25 25

020 20

D 15

(X 10 u

Ca content of wheaf, % Ca content of maize,%

'71 Iraq Internat. Iraq lnternot. n. 119 1765 n 31 1967 2= 7315 4671 5 0.6456 3450 37 =1608 3076 As = 2851 2815 min. 4400 110 min = 4200 10 max.= 15431 21930 max =17995 17995

100 200 400 000 7600 3200 6400 12800 25600 50 100 200 400 600 1600 3200 6400 12800 Ca in wheat soils, mg/I Ca in maize soils, mg/I

received more macronutrient fertilizers than the local varieties. Furthermore, the majority of local varieties were grown under rainfed conditions and the majority of HYV under irrigated conditions, rendering the comparison between the variety groups stillless justified. See Section 8.2.4. 304

40 Iraqinternat Iraq Internat. Fig. 385. Magnesium, Iraq. 35 n= 119 1764 35 n= 31 1967 x= 162 172 ==.291 .251 C os= 045 060 ss=.089 .119 30 min.075 044 30 min.=.160 .036 max.= 272 948 max.=.483 1.125 25

.05 Mg content of wheat.% Mg content of maize, % 0 0 Iraq Internat. Iraq Internat 35 n= 119 1764 n= 31 1967 8= 678 489 X= 688 446 ts= 334 437 xs= 245 462 rein, 190 10 30 min.= 237 max,1555 3298 max.=1272 6490

25 25 fr a 20 o)-

U.1 15

CS Ul CC 10 4V/r,r A\_ 25 50 100 200 400800 1600 3200 12.5 25 SO 100 200 <00 800 160032006400 Mg in wheat soils, mg/I Mg in maize soils. mg/I 305 8.2.3 Micronutrients

Boron. The national mean B contents of Iraqi soils and pot-grown wheat were considerably higher than those for any other country (Figs 22, 25, 386 and 387). A third of the plant- soil sample pairs falls in the highest B Zone (V) and half in Zones IV and V. The ranges of variation of both plant and soil B contents are exceptionally wide with some values falling in the two low-B Zones. The latter do not indicate any serious B deficiency but it is apparent that a response to B could be obtained at several locations in Northern Iraq. (See also Section 2.7). B toxicity could become a problem at several locations in Iraq if crops sensitive to excess B are grown. This would be most likely in the Mesopotamian

100 80B..litelemisaGJwilsompuilmmummommenumsihsammumemlisingiummaimuattra 60 PILIMIMMTIMMIIIKIMINIIIMINIIIMMINThill 50MEITTINIMAINIMPW111111111111=111111111111Mmill 40INIZEICTIEMEE191111111111111O IMMIlld E 30nigniiMENNiii Align a. 20mew Nopumnipmgm

10iiiiinliiligliiNumnmNolowommas"Firmilhrmaslimliii 15 8 immoimwmawasawmummonmenWag 'romrWarilignramesumumwain munamapprAimmummomunniMinnffirmiirepigialMMIIMIE111111111 IlliiM.WIGOWNIIMINEIIIMIIIi Fig. 386. Regression of B content of 2IliiEon SIMI pot-grown wheat (y) on hot water soluble soil B (x),Iraq. For details 11 Minn IE 11111 of summarizedinternationalback- . .3 .4 .5.6 .8 1.0 23 4 56 8 10 ground data, see Chapter 4. B in soil, mg/I (Hot w. sot.)

100 1111111MINE MinallmIllall111MMIInnIMINM.MrAl 80 MOMIliiIKAIR 60 B"iiIIIMIllNamiumins 50MIIIII 71121:11112.0.E01 11101461M 40iiill-- L'4ZIEREENNO grin aE 30IIIIIMMEEMIVI lindrEilli riamm 111 AdilIN iddeid oripAsvamisani imumnawswracianimamommutuRarAMISAVIMM...1`.11.11.. 111111111111 mmummenummorqui111111111111M011012111111111. MINNIIIIIII CO 20111111111111

1116. Fig. 387, Regression of B content of 111111111E11 drillikih. ! pot-grown wheat (y) on CEC-corrected .081 .2.3 .4 .5.6 .8 1.0 23 4 5 6 8 10 (hot water soluble) soil B (x), Iraq. CEC -corrected B in soil, mg/I (Hof w. 306

22 Irq Inte na. Fig. 388. Regression of Cu content of pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), 2°1811111161MINEs Illi111111111 E Eno :Isom Iraq.

-,a,-'116411111111111111MIIIIHIMIllt .612111111111111=1111111111111%1

-1-1011111111111111111115111 C 8111111111IN1!d11111 0611111111111NIS1111111

4111111111011MBOIHIMMIll

.4.6 .8 10 2 4 6810 20 40 6080 Cu in soil, mg/l (AAA c - EDTA)

Fig. 389. Regression of Cu content of 22Cu 1 pot-grown wheat (y) on organic carbon- Iraq intern.t 20 I corrected (AAAc-EDTA extr.)soil Cu (x), Iraq. 18Iiss:8'2;:.'6( INE 57I I Ai

E16mY4.0Iiiiiill',°*-iWallin 11/1 O. 14 cs

12 "6 10 1111 1111111 , 8 .., i o o , 6 ....i...... 4 ,,,,,,...... , 2 ... _..,..,,..,.., o .4.6 .810 2 46 810 20 40 60 100 Org.C-corrected soi CU,mg/l (AAAc-EDTA)

plains where the soils are very rich in B and where the irrigation waters apparently contain B. See Section 8.2.4. Copper. The locations of Iraq are relatively close to the centres of the "international Cu fields" (Figs 27 and 29). The variation of soil Cu contents is exceptionally narrow and with a single exception') the samples are within the normal Cu range (Zone III, Figs 388 and 389). Response to Cu in Iraq is unlikely at the sites examined.

I) The extremely high Cu corttent of pot-grown wheat in this sample (47541) may be due to contamination. Neither the soil Cu value nor the Cu content of the respective original wheat indicated excess Cu. The Fe and Mn contents of this sample were also exceptionally high. 307

Fig. 390. Regression of Fe contents of 300 275IMIN11111111111111/111 11111.11111 pot-grown wheat (y) on acid ammonium 250EnoIN11111111111111 11 acetate-EDTA extractable soil Fe (x), 225 DINE11111111Rill 11111 1 Iraq. 200WIR-11111111111;111 11111 I 175IIII`rcgoEll E 1 50ILvuuiiiuiiuuiiiiii o. 1 25 o ., 100 1 . 90simm Imo. .-Ilk 11111 1 MEW 1111111 %IIUIIIIP E 80 MINIMUM:illIIIIMill 3)* 70 o 60 M111111101111k '11111 I 50OWEN 11111 I

40MI1011111 ..- '1111 I

30 IllI MIMIIll El 11111M1111 10 20 40 60 100200400 600 1030 2000 Fe in soil, mg/1 (AAAc-EDTA)

Iron. According to the data presented in Figs 31 and 390, problems with Fe are unlikely in Iraq unless the soil oxidation-reduction conditions are such as to limit its availability. These conditions are more likely to develop in the northern parts of the country where crops are grown under rainfed conditions, and where increased oxidation potential may lead to the oxidation of Fe from ferrous to ferric forms so decreasing the Fe availability to plants. See also Section 8.2.4. Manganese. Because of the generally high and only slightly varying pH of Iraqi soils the Mn contents of both plants and soils are low and their ranges of variation exceptionally narrow (Figs 33, 37, 391 and 392). Although a few samples fall in the low- Mn Zones (I and II) these do not indicate any severe degree of Mn deficiency. However, a response to Mn could be obtained at several locations especially at sites where soil moisture conditions are unfavourable for Mn uptake by plants. See also Section 8.2.4. Molybdenum. Unlike Mn the contents of Mo vary widely in spite of a uniformly high soil pH (Figs 393 and 394). In general, high values dominate and in the "international Mo fields" Iraq is among the high-Mo countries (Figs 15 and 20). Although the national plant Mosoil Mo correlation is not greatly affected by pH correction, the location of the whole Iraqi Mo data on the regression graphs is more consistent with those of other countries when the A0-0A extractable soil Mo values are corrected for soil pH. See also Section 8.2.4. Zinc. Irrespective of the extraction method used for determining soil Zn, the positions of Iraq are the lowest in the "international Zn fields" (Figs 41 and 43) and both soil methods give good correlations with the results of plant Zn analyses (Figs 395 and 396). Still, the DTPA extractable soil Zn values are relatively lower than those extracted with AAAc-EDTA. In other words, more than half of the plant ZnDTPA Zn points are 308

2000 Fig.391. Regression of Mn content of pot-grown wheat (y) on DTPA 1000 INN:ligni 11I extractable soil Mn (x), Iraq. 800 IITE731WWWilIPArTIPMEMIFINWMI____1=111_1111 7404Mr 600IIMMENNE11MINEMili 40011INEMIIIIMEMINI011111IN o_ 300 15, 200IlL1111111111=111111111MPI o 1 11011111111111111111M1'4.1=MIWNIC1.11:1011MMEMEN=11. -C*0080nimmargengramorni= 6011 10.1111.0.11.1.11111 40111111111111EMENCIUMMIIIIIIII111M 3011111111111PININIMMINI11111 2011=21111111.11.0111111

10011111111111E1111111 OS 11M=MMIIMOMIONWOMMINENIM=MINE 8 011=marnia=aliTimmEMI MAIMMINE *7. 2 3 46 810 20 30 40 60 80100200 300 Mn in soil, mg/l (DTPA)

2000a momemamm ammEn. 441mma Fig. 392 Regression of Mn content Mn of pot-grown wheat (y) on DTPA 1000 mum=mplioN=vem.rwmarER101111111111 iimmiummai extractable soil Mn corrected for pH NE =. EvaaLviiiimmrami.a.immissmNumoii. 800SI =INIMMMMMOMMINEIIMM=MMil (x), Iraq. 600 MINEIHMIIVIS/MWEIRIIIIMEM miiirmimommer 400 300iiiiiiiiiiuniiiini 200MM.inomminueNiin

100 111111 808=1.,mmistss eadrommINIIN=Markválmusam ino MELIA 60iiMMitaiiMilniIMEMEnnMSINimmomMalotionwrimemomwomugum 401111MOMPRICIIMM11111111110=MIMElli 30IIIIIMRMINIIM111111=111110 20owrisinummilomimmipp- lo Ilk MMEMAII MMINIMMINO OBMMNIMMIMML111111MIUMME INE=MIIMJEIIMMENIN 2 3 46 810 20 30 40 60100 200 300600 pH-corrected Mn in soilimg/i (DTPA) within the two lowest Zn Zones while in case of AAAc-EDTA there is a strong concentration of samples in the lower side of Zone III. As pointed out in Section 2.3.7.2 (Table 18 and related text), the DTPA extractable soil Zn is generally better correlated with plant Zn in areas with alkaline soils and pH-corrected AAAc-EDTA Zn in areas where acid soils predominate. Therefore, it would seem that in the case of Iraq the DTPA extraction method gives a better estimate of the soil Zn status (Fig. 395) than does the AAAc-EDTA method. See also Section 2.7. According to these analytical data Zn deficiency seems more common in Iraq than in any other country studied. The lowest values were measured from samples originating from widely scattered areas, and no distinct low-Zn areas could be distinguished. See also Section 8,2.4. 309

Fig. 393. Regression of Mo content of min --sizzumannum.....---rara 6,,, i....-0...... n. CE pot-grown wheat (y) on ammonium l.. 111111IMMIMMin oxalate-oxalic acid extractable soil Mo -?,...1210111111MEEmmagi 7-''° 2ifiallENZIONIII (x), Iraq. 3 2 : 5 7 ;-111 - -111filii E

2/". 1 wwwwwwwwwaraummwitie_l_0111111111 cl ,6 MISIEIWW.'4LiMMEMMIRIM,..iliNier - - .5411-imairi ErIENIIIIiii - Pi,;,__.i I \ hip AOS 11 Pr" " o ...... 1 iiinui o ....A-Wismia :=1:1111:: Cr /Liar mmaniinualm .06mimriaimignusmin mri IANNMII I O1:1111ff le .03MIII I1111ii . .02MMIiM 1 .01 I III u 01 .02 .03 .06 .2 .3 .6 23 Mo in soll, mg/ (.40-0A)

Fig. 394. Regression of Mo content of p=1111 pot-grown wheat (y) on pH-corrected nImmikliiiill'.....voni 6 EN n150 MEMO In . 53 AO-OAextractablesoilMo(x), y±0.s. 620. 96in±t,.0 22 0.-6 me 5 t s .0. 6/..t 0.276 0 210 T 0. I Iraq. 3 y=1.186 .89,Ians i 1-.0.817 r°00Y.6 .67 '... 2IN

E 4. a 1 =mu 431: ...... uwiarli IIIIIIII//MMIIIIII 1111111111111111111111110 MIII 1111=10114111.,_ 1.1WIWIPP.511116411Mun MENEMMINIMMI=ONHOMENONE -5.6 ii=r--4"Urgaralailralai IMO= anntlir AIMENEMEIMIll o .3 III Ell Ir AI111111714111

.2 o OP . .1 MOM mil OM . ==11111=m, 111111M1111111111 MN= .0611111111111111111MIllrating jIUU IIIINEUMWMIMINIIIIIiiii IIII 11111= .03 IPREMIER .02 smog al Ja .01I f IMIPIMIONIMINIM MUM .006 .01 .02 .03 .06 . .2.3 .6 1 23 pH - corrected MO in soil.mg/I (40-oA)

8.2.4 Nutrient status of soils and plants of Iraq with special reference to irrigation

Successive civilizations have lived in the Mesopotamian riverplains for approximately 6000 years. They founded their existence on irrigated agriculture, and consequently, irrigation practices have greatly influenced the soil formation. At present almost the whole plain is covered by a layer of irrigation deposits of usnally several metres in thickness (Delver, 1960). About two-thirds of the samples in this study originate from the Mesopotamian plains and represent irrigated agriculture while one-third were collected from Northern Iraq where the crops were grown under rainfed conditions. The analytical 310

2001 glE Fig. 395. Regression of Zn content of pot-grown wheat (y) on DTPA extractable soil Zn (x), Iraq. r2 100 90 8011111MIElliall I 111111111111VP2 70 E60urrimfrimardniqui 9750MINN NMI 111101 MAM o 40NUM MIIMMilliiir/Iill S" 30E l 111Eniliniamu 11111 20INiiliiriiIIIME o N 10111111111111111...... iammumgii=mumena II 9 8pg 7 alivinommiesi no 6du ..11111 NImin=iiiIin= nu EEL 1 EITmommin .2.3 .4.6.8 1 23 1,6810 203040 60 100 Zr1 in soil,mg/ (DTPA)

200 Irg \ I tn Fig. 396. Regression of Zn content of n50 \ 3 pot-grown wheat (y) on AAAc-EDTA yt. it\ 1 31 1. ' \ 8 .t extractable soil Zn corrected for pH

091.2.1 x r.18; o ,f 100 \ (x), Iraq. 560' ' 90 , 80MI \ 'ME 70Ill III1 60 i 50 \ iIll i I 40 E \ , 30 \ 14., Iiir,

20 \ \\ ..:;,e \

.;11. 1

10 9 8 i I 7 I\ A 1! 6 I1\ 111 V .2 3.4.6 .8 1 2 3 46 810 20 304060 100 pH-corrected Zn in soil,mg/I (AAAc-EDTA) data on the soils and wheato of these two groups are given for comparison in Table 27. When comparing the data it should be borne in mind that although the irrigation methods practised since ancient times and their consequences may be the most likely cause for the differences between the soil and plant properties given in Table 27, the samples under comparison are from different geographical areas and this may account for

1) Maize crops and respective soils are omitted from the comparison because the original maize crops, with one exception, were all irrigated. 311

Table 27. Comparison of analytical data from non-irrigated and irrigated soils of Iraq and respective plantso. Statistical differences (t-values) between the groups are given at 5*, 1** and 0.1*" percent significance levels.

Characteristics Non-irrigated Irrigated (n =44) (n 75) (t) pH(CaC12) 7.55 ± 0.08 7.77 ± 0.11 11.98*** Texture index 54 ± 11 50 ± 10 2.19* Organic C, % 0.97 ± 0.33 0.77 ± 0.27 3.43** CEC, me/100 g 30.9 ± 8.3 24.5 ± 5.6 4.54***

. 10-4 S El. conductivity, 1.7 ± 0.9 12.2 ± 10.9 8.27*** CaCO3 equivalent, % 23.8 ± 8.3 26.0 ± 4.9 1.55 n.s. Na in soil, mg/1 21 ± 64 510 ± 465 8.97*** N in orig. wheat, % 2.91 ± 0.59 3.67 ± 0.76 6.06*** N in soil, % 0.104 ± 0.034 0.093 ± 0.036 1.76 n.s.

P in orig. wheat, % 0.269 ± 0.062 0.317 ± 0.087 359*** P in soil, mg/I 6.1 ± 3.7 9.6 ± 12.4 2.25*

K in orig. wheat, % 3.20 ± 0.52 3.65 ± 0.60 4.34*** K in soil, mg/1 430 ± 117 288 ± 127 6.19***

Ca in orig. wheat, % 0.383 ± 0.092 0.370 ± 0.095 0.77 n.s. Ca in soil, mg/1 7929 ± 1447 6955 ± 1597 3.41**

Mg in orig. wheat, % 0.124 ± 0.025 0.185 ± 0.038 12.70*** Mg in soil, mg/1 375 ± 107 856 ± 260 12.95***

B in orig. wheat, ppm 3.64 ± 2.84 18.32 ± 16.00 7.74*** B in pot wheat, ppm 4.23 ± 1.34 17.07 ± 13.83 7.98*** B in soil, mg/1 0.45 ± 0.37 1.89 ± 1.68 7.10***

Cu in orig. wheat, ppm 7.31 ± 2.81 10.95 ± 15.05 2.03* Cu in pot wheat, ppm 7.02 ± 1.02 10.18 ± 2.14 10.85*** Cu in soil, mg/I 4.97 ± 0.85 5.86 ± 1.26 4.49***

Fe in pot wheat, ppm 53.3 ± 7.0 74.7 ± 18.2 9.09*** Fe in soil, mg/I 64.0 ± 17.7 137.3 ± 32.7 15.84***

Mn in orig. wheat, ppm 57.4 ± 17.8 99.7 ± 42.5 7.57*** Mn in pot wheat, ppm 69.5 ± 17.7 70.3 ± 25.8 0.21 n.s. Mn in soil, mg/1 6.8 ± 1.5 6.0 ± 1.8 2.09*

Mo in orig. wheat, ppm 0.28 ± 0.23 1.72 ± 0.85 13.85*** Mo in pot wheat, ppm 0.12 ± 0.08 0.72 ± 0.26 18.59*** Mo in soil, mg/1 0.096 ± 0.049 0.370 ± 0.280 8.28***

Zn in orig. wheat, ppm 21.2 ± 4.6 21.1 ± 6.5 0.07 n.s. Zn in pot wheat, ppm 10.8 ± 1.8 12.0 ± 3.5 2.39* Zn in soil, mg/I 0.28 ± 0.11 0.28 ± 0.40 0.00 n.s.

I) For extraction and analytical methods, see Section 1.3. B: hot water extraction + CEC correction Cu: AAAc-EDTA org. C correction Fe: AAAc-EDTA Mn: DTPA + pH correction Mo: A0-0A + pH correction Zn: DTPA 312 the differences. Since the effects of irrigation depend on the quality of the irrigation water in relation to soil properties, and as no analytical data of the water used at different sampling sites are available, too far-reaching conclusions on these effects cannot be justified. However, the very large differences in the contents of such highly soluble elements as Na and B would suggest that much of the extractable Na and B in the irrigated soils is due to accumulation of these elements brought to the soils by irrigation water. The same may be true for Mo and Fe; e.g. Fireman and Kraus (1965) found Mo in irrigation waters. Irrigation seems to have practically no effect on Zn levels, and for Cu the quantitative differences are relatively small. There are only small and statistically non-significant differences between the Mn contents of wheat grown in pots of formerly irrigated and rainfed soils, though the original irrigated wheats contained significantly more Mn than did the rainfed wheats. It would seem that in irrigated soils the moisture (redox) conditions are more favourable to Mn absorption by plants than in the drier field conditions in the rainfed areas. This difference does not appear in the Mn contents of pot-grown wheat since all pots were maintained under similar soil moisture conditions. El-Dujaili and Ismail (1971) measured 2.6 me Ca and 2.2 me Mg per litre of Tigris River water at Aziziya. If these concentrations are typical of the water used for irrigating the present crops, it it possible that the irrigation water leached Ca from the soil which is high in Ca and contributed Mg to the soil which is low in Mg. For N, P and K the comparison may be obscured to some degree by fertilizer application, even though the amounts applied were generally small. According to the field information data, the average N, P and K applications to irrigated wheat were 8, 3 and 1 kg/ha and to non-irrigated wheat 2, 1 and 0 kg/ha, respectively. On the debit side the higher yields obtained frorn irrigated fields would remove more nutrients from the irrigated than from the non-irrigated soils. This applies especially to K which crops extract from soils in considerable quantities. It is apparent that the accumulation of elements such as Na and Mg in the irrigated soils contributes to the higher electrical conductivity and pH of these soils. The above results indicate in a general way some of the differences between the irrigated and non-irrigated soils. Further studies, including detailed data on the quality of irrigation water and ground water at various locations, are needed to obtain a more comprehensive picture of the behaviour of different nutrients and other elements and of their leaching-accumulation characteristics.

8.2.5 Summary

Most of the Iraqi soils sampled for this study are medium to fine textured, highly alkaline, low in organic matter and have a medium cation exchange capacity. Very high electrical conductivity, CaCO3 equivalent, and sodium content are typical, especially for the soils of the Mesopotamian plains. In general, the P and N contents of soils are low, the K contents correspond to the average international level and the Ca and Mg contents are high. The most likely problems to arise with micronutrients would be due to deficiency of Zn but those due to both shortage and excess of B can also be expected. High Mo and low Mn are typical for Iraqi soils but most of the Cu and Fe values are within the "normal" range. There are distinct differences concerning most nutrients and other soil properties between the northern parts of the country, where the crops are grown under rainfed conditions, and the Meso- potamian plain where irrigated agriculture has been practised since ancient times. 313 8.3 Lebanon

8.3.1 General

The geographical sources of the Lebanese maize-soil sample material is shown in Fig. 397. The soils are medium to fine textured and alkaline with relatively low organic matter content and high cation exchange capacity (Fig. 398). The high CaCO3 equivalents and electrical conductivities and relatively low sodium contents are typical of most Lebanese soils sampled (Appendixes 3 and 4).

40. ao

1 -,

i .L.A .45465 .../ ° 05466

TITIPOLI

4059 61118BEK

45455 95456

BEYROUT 6 ZAV4 454 ,4 2 05461,,--- ....,....! BAt604 546f ;5'!9'. : gágP i

SOTEl6 DNE

4 .9 ',.. ) 95460 / O 45461 ...... ,'' RAC VITA ') )

ARJAYOUN .... . f : -i L E BANON

o a 12 Is 22 24_ 24 24 as 4224.

39.50' 90'50'

Fig. 397. Sampling sites in Lebanon. 314

5 Lebanon internat. ebanonInternat. Fig.398.Frequency distributions of 3764 n. 16 40 n. 16 3783 44 30 0.60 R. 7.44 6.64 texture, pH, organic carbon content, -4s. 13 16 4-sr..29 1.12 min.= 38 35 min.= 6.95 3.12 and cation exchange capacity in soils of max.= 84 92 max.. 7.75 8.56 25 Lebanon (columns). Curves show the tr, 20 international frequency of the same >- C) characters. w /5 co

0 20 30 40 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

Lebanon Internet 35 Lebanon Internet 40 n. 16 3779 ' n= 16 3777 R..9 1.3 =38.3 27.3 4-5=.3 1.2 I's=9.4 14.6 ":S. 35 min...4 .1 min.= 23.8 .2 mox.= 1.3 39.1 max.. 61.9 99.9 'I,30 25

LTJ 20

¿Du 15 cr LL 10 44A. 2 .4 .8 1.6 3.2 12.8 25.6 8 16 ORGANICC,% CEC, me/100g

8.3.2 Macronutrients

The total nitrogen contents of soils are low. However, owing to the relatively high rates of applied nitrogen fertilizers (53 ± 99 kg N/ha), the mean N contents of maize are slightly above the international average (Figs 6 and 399). The wide variation in N contents of maize may be attributable to varying fertilizer rates. The phosphorus contents of soils and maize are relatively high (Figs 7 and 400). The potassium contents also slightly exceed the respective international means (Figs 8 and 401). The sampled crops had received relatively heavy dressings of phosphates (33 ± 54 kg P/ha) but no potassium. The national mean for the exchangeable calcium content of Lebanese soils is the second highest (after Syria) in this study but the average Ca content of maize exceeds only slightly the respective international mean (Fig. 402). The mean exchangeable magnesium content of Lebanese soils exceeds the international mean but the mean Mg content of maize is lower on the international scale (Fig. 403). These results may be due to the relatively low uptake of Ca and Mg by plants from soils of heavy texture and high CEC which characterize most Lebanese soils. See Figs 12 and 14 and related text in Section 2.2.4. 315

Lebanon Internat. "Lebanon Internat. 40 n. 16 1967 n 16 35 1958 Lebanon Internat. 30 0.3.3 3.1 9=3.43 3.14 is = 0.5 1.0 35 n. 16 1967 4E. is =.73 87 5..36 .33 a, min.= 2.2 0.6 min.= 1.83 .88 ts..06 .10 25 max.= 4.2 6.7 Max.= 4.69 6.51 0 30 mm.. .29 .05 non,, .43 1.04 s_ 2 0. 2 U>-0

1- .1 15 0 X 10 Q. U_ u_

.1 .2 .3 .4 .5 .6 .7 .8 .9 3 4 5 6 2 3 4 6 N content of maize, % P content of maize,% K content of maize,%

70 Lebanon Internat. Internat. Internat. Lebanon n. 16 1967 3 Lebanon n. 16 1958 R. 42.5 22.5 n= 16 1967 60 9,097 .135 tss. 31.0 33.0 13= 354 330 min..5.3 0.1 is= 234 356 no .023 .088 max..104.6 656.0 min. =.061 .008 min . 101 18 a) max .1070 5598 (.0 50 max.=.137 1.657 25

Q. 40 >-

20 u-

2 5 10 20 40 80 160 320 640 25 50 100 200 400 8001600 32006400 .005 .0 .02 .04 .08 .16 .32 64 1.28 K in maize soils, mg/I N in maize soils,% P in maize soils, mg/I Fig. 399. Nitrogen, Lebanon. Fig. 400. Phosphorus, Lebanon. Fig. 401. Potassium, Lebanon.

40 40 Figs 399-403. Frequency distribu- Lebanon internat. Lebanon Internat. 35 n16 1967 n= 1967 .47 n. 16 tions of nitrogen, phosphorus, po- 9..50 R..251 es =15 .20 8=.219 no. .119 S.'30 min...22 .09 ts..075 tassium, calcium and magnesium 707..85 1.88 Ú 30 min.=.143 min. .036 max.=.406 max..1.125 in original maize samples and re- 'eg 25 25 >- spective soils (columns) of Leba- (2) 20 (2) 20 ILI non. Curves show the interna- o15 Ui tional frequency of the same co 10 LL IL characters.

05 Co content of maize,V0 Mg content of maize,%

50 Lebanon Internat. 40 n. 16 1967 Lebanon Internat. R. 7789 3450 35 n= 16 1967 2815 35 es. 1919 446 min.= 5000 10 6= 545 max..11950 17995 no. 412 462 min.= 177 1 max.=1553 6490

100 200 400 800 160032006400 50 100 200 400 800 1600 3200 6400 12800 12.5 25 50 Ca in maize soils, mg/I Mg in maize soils, mg/I Fig. 402. Calcium, Lebanon. Fig. 403. Magnesium, Lebanon. 316 8.3.3 Micronutrients

Judging from the limited number of samples the micronutrient levels in Lebanon seem to be quite satisfactory. In the "international micronutrient fields" the position of Lebanon is relatively close to their centres, often slightly lower (Section 2.3: Figs 15, 20, 22, 25, 27, 29, 31, 33, 37, 41 and 43). Since the ranges of variation are usually narrow the points

100 MI Minimum 80 E...:0.1111.1111111MattilriiTiAlliOMM1=1=INIMp_ne11111111=111M1111=111111insw MIN 60INIMEMIMMIPX011111MIMII1111111 50 .0 54_ O. 40um MILESELEIIIII IN111111111 30 .82 o.5 11031191 MI= Q: 20111111111111111111111 11111111 11111 111111.11111111 oIIIMMU1111=111113.11.111101 MIIMMIMM111.411110111MIMINMMIMIIM 13 8inisnINEMinh.W.MINMENEPOIMEMOMENaiilinnimENEmPionminlimmis 1111111111M1111111=IMI 1111110111111 MENHIR m3 11111 2 Ileum Noma Fig. 404. Regression of B content of pot-grown wheat (y) on hot water 11111111 111111 soluble soil B (x), Lebanon. For details .4.56.810.810 2 3 456 810 of summarizedinternationalback- B in soil, mg/I (Hot w. sol.) ground data, see Chapter 4.

100 80 n= 5 - 60 7.ts= 065 50 7±s=0 4 +0 H ,7 40 E y.5 77+1 710 x .30 r0 255s 8 * 20

10 8 6 5 4 o 3 03 V 2

Fig. 405. Regression of B content of .08:1 .2.3 .4 .5.6 .8 1.0 23 4 56 8 10 pot-grown wheat (y) on CEC-corrected C EC -corrected B in soil, mg/1 (Hot W. 601 ) (hot water soluble) soil B (x), Lebanon. 317 indicating micronutrient values for single plantsoil sample pairs are usually within the "normal" range (Zone III in Figures 404 to 414). Even the few Fe, Mn and Mo values outside Zone III are not in any extreme positions. However, at some locations response to B and Zn may be obtained (see Section 2.7). On the basis of these limited analytical data, micronutrient problems in Lebanon seem less likely than in most other countries involved in this study.

22 20ZliiiiiiIIIIES11111111111111111111111

1811111MMI1 11111111.11111111111111 E a-1611111111MITI'R6111:111S1111111111111 o 11111111111111111111111111111111 .ca) 14 .6 1211111111111M11111111111111111111111 V,10 8c 8111511111111111111111111=111111111 6 4tommuumilimminim 2 Fig. 406. Regression of Cu content of pot-grown wheat (y) on acid ammonium simmummishiab..Aimmulb 2 .4.6 .810 2 46 810 20 406060 acetate-EDTA extractable soil Cu (x), Cu in soil, mg/I (A AA c- EDTA) Lebanon.

2 '

I) ) nrei6 7.s=n .3580* 57 51±5.77±1 e . 1111111

1 i(7)9.§)*11111 inisiv , - "1.1 Nil 11111NE

' 1111 ' Ell :IN 1111111 3 IEEE1.1111111111111 11PNUiiiflillI ' Fig. 407. Regression of Cu content of UP91111 111113 pot-grownwheat(y)onorganic carbon-corrected (AAAc-EDTA extr.) 31 .4.6 .810 2 46 810 20 40 60 100 soil Cu (x), Lebanon. Org.C-corrected soi CU,mg/l (AAAc-EDTA) 318

3 00 Fig. 408. Regression of Fe contents of 2 75 pot-grown wheat (y) on acid ammonium 2 50iirffinlaw"mm 225 acetate-EDTA extractable soil Fe (x), Lebanon. 200Eingratinli0)% 1 75 E 1 50 .125 o 100 is90 t"8011111111 70 o 60

40 301111 10 20 40 60 100 200400 600 1000 2000 Fe in soil, mg/l (AAAc-EDTA)

8.3.4 Summary

With exception of relatively low total N contents the macronutrient status of sampled Lebanese soils is generally good. Fewer problems concerning micronutrients could be expected than in most other countries. 319 IMMEMM.V87 =1110O Fig.409. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA extractable soil Mn (x), Lebanon. 1000MN ECM 111111.7111111111111VA EIMIMMENtLq/lanXK fAIN=IMINII=4 800 111117FIEEKEWIRgIMIEWMWEEMILMWITrin=1 600 11II ream,.....mumiiimmimumumm 11=1 K400nimiimmillumm o_ 300 IL .4.: a, 200 ll MIIIIIIIMMIIMILVISI .c 3 "isinn 1111h. ,,=MIMMEMM.161.,:L'E01116 C 80:1...,kozninEmmr...romini==7. G) c80PIIMMMIRIMMENII=IM%IiiMMENNIMMIN o o 11MMED111%%1=MMENNI=M=M c _5"30 IIIMINiispoommuiIIIIRIIIIIIIMMIMI

2°10111111111110111111111111 8 Il;; MEMEMOBIUVIIIIIMMOIK1=IminniM111.=11MMEBialM3ISTOMMIIIIMWM= 2 34 6 8 10 20 30 40 60 80100200 300 Mn in soil, mg/i ( DTPA)

2000 M n 1000 NIMMIMPFIMUMMMMIMM.,6UMIIM11IMIIN11111M161011111111 800 11=WINENNOIIMIIMMMONNMadelglIMAZELELTlekMMIN011MINIMMIN 600 MCITEIRMIDIMIWPMERDX4i [EnTilMINEMMEMI E IIEigTHiliNUME=WSTRIMMINMEMMINEMPM o. pp- 0- 400 1; 300 200 1. o Ó 10011.11111111111111111111111111 G) ..----...... 80...... n...... ,.. Il...... IMUMBEliliiIMINMENIIIIM=IMEN1111 6011 =MMINErraMEMMMINIMMEWI4 40 1Pg''9 1 11 K 2 30 20

10 Fig.410. Regression of Mn content EN11111111M111111111111111moimmosommmmwommo immulmos 8MI 1=MINIMMIEiiiiIIMMMEMEN MMIMMIM of pot-grown wheat (y) on DTPA El 1=1MIln Maal=MINIMO =MMUIN extractable soil Mn corrected for pH 2 3 46 810 20 30 40 60 100200 300600 (x), Lebanon. pH- corrected Mn in soil, mg/I (DTPA) 320

=mommtm 10111P11111111111111=111111111111MISIIIIIMI=1.... Lq/..110.4111111111111Mbliffilik161111111111I Fig.411. RegressionofMocontent Mo MIIIM1111111=111111111111111M111111 6EMMEIMMIIIIIIIMMINIMIIN 111 of pot-grown wheat (y) on ammonium EMMFAMIIIIEEIMEEEIMEIMIII11111M oxalate-oxalic acid extractable soil Mo 3 (x), Lebanon. 2=EMIR= ° MM.= E O. 'a- 1.1111111.11111111.11MMENM=1 lgal= o nn.1WW=ME11111MMIEl 1111M1111111111111111111!= .6MMENIIIIMESINIIIII=1111ENrar=11111.ENIwilimmiimmuNMEN o .3 C.2MENIONIM1111110111 o o o .1EMI" =MMEMIMILM= .06 WAINIMINEMIMMIMIIIMEIMME91

.03 .02 I Li la 01 .01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/ (A0-0A)

.. EN:11111==2:1111 =.11 Momom mmsoutunrracrilwommman -- 6 I MIIIIIIIIMMIIIIMIMI EMI Emmun=16 MIUMEI:MIMMIIMIll MI onny: illurammaromun MI 3Mininy=0.093.0.9 owl mama= n 2III II 111111C31111iiiiil al 0.1E ...... 11111110779"...... Ipolomman ...... _._ o o.6 -c imali illiiitall=11 .3 g.2imimilmammi o o o .1imr\ Gronnow .06Emilwapiii=ina .03lipmendimmilm" .02INNI IMEI 11911MM. 111 Fig. 412. Regression of Mo content of pot-grown wheat (y) on pH-corrected .01MO blind" II .006 .01 .02 .03 .06 .1 .2.3 .6 1 23 A0-0A extractable soil Mo (x), Le- pH-corrected MO in soil, mg/l (A0-0A) banon. 321

Fig.413. Regression of Zn content 200

of pot-grown wheat (y) on DTPA 77 extractable soil Zn (x), Lebanon. 5 3211og

100 90 80 70 g_ 60 o. 50

_t)40

10 9 8 7 6 Ill FL

.1 .2.3 4 .6 .8 1 23 46 810 20 3040 60 100 Zn in soil.mg/1 (DTPA)

200

100 , 111111111 90MINIMEMIllMIIM-TaJillIMIIMIE1111111&1111 80IEENIIIIMIMINION1111=111111111111511111 70MIIIII11111111=111M111511111 60IIIIIMMWEIVIIIIII1111111112111111111 g. 50 11.111111111.1111111111110111111111 2.:-Lo 40 30111111111111111111101111111fil 11101101pRionno

1011111111111liliiiIMMIIIIIIMENNI `MINIM 9...... 1!IIM11111111111.M111111111 8EIMMIIIIIMIMEN11111=1=111111111111 Fig. 414. Regression of Zn content of 7 pot-grown wheat (y) on AAAc-EDTA 6MINIM IIIIIMEINIMENEENEE extractable soil Zn corrected for pH .2.3 .4 .6.8 1 23 46810 20 3040 60 100 (x), Lebanon. pH-corrected Zn in soi mg/1 (AAAc-EDTA) 322 8.4 Syria

8.4.1 General

The original Syrian sample material consisting of 20 wheat-soil and 18 maize-soil sample pairs was collecfed from seven out of 11 Syrian Provinces: Latakia (11), Aleppo (7), Hama (7), Deir-ez-Zor (7), Derá (2), Homs (2) and Raqqa (2). The geographical distribution of sampling sites is illustrated in Fig. 415. Of the sampled soils, 15 were classified as Fluvisols and 12 as Luvisols. The Syrian soils vary widely in texture but the majority are fine textured (Fig. 416). Very high pH, medium organic matter content and a high cation exchange capacity are typical. The national average for CaCO3 equivalent is one of the highest recorded in this study. Electrical conductivity values are relatively high but the sodium contents correspond to the average international level (Appendixes 2-4).

1 ,41)

, 1 w.,-AELs,40AuRs0/,. -., .95729

.044 a 810 .X4 °AMINE 08728

EV07714 J 45727 a

. \ -- .... 40717 ,

OIVICOVR

°Étir12110AN E AT°Li. ' 40 08735 45 19 ' '

aa . 1 - -MITINF r-- " 4573 0°° '84%704 1 4° 7 . 48725 703 1,1111.08 tr 1:172'48706 4 48 4872.1 A" I ! PALesIVRE 48720. Kft?4.,"7" I

04,rEINE a

:50.0,0,

D.4`8"- \

°AMA

l,ONVRA

ET:8702

o 508610A 1-N.tir°7. SYRIA 700.014 SAWA° ----

e 10 0,44..0_0,471 4 .o4000

00.

Fig. 415. Sampling sites in Syria (dots -= wheat fields, triangles = maize fields). 323

8.4.2 Macronutrients

The total nitrogen contents of the sampled Syrian soils as well as the N contents of the original wheat and maize are at the average international level or slightly below it (Figs 6 and 417). The sampled crops, especially maize, were fertilized with high, but widely varying dressings of nitrogen (wheat: 52 ± 59 and maize 117 ± 87 kg N/ha) which may have extended the range of variation for the N contents of plants.

5 Syria Internat. Syria Internat. sq n = 38 3764 40 n=38 3783 0.54 44 g= 7.59 6.64 30 06416 16 Rs= .20 1.12 min. = 18 35 min.= 6.90 3.62 ma, = 82 92 max.= 7.92 8.56 'S 25 30

20 25 o>- w 15 20

o 15

LeL2 0 LL 10

90 pH (Ca C12)

5 Syria Internat. Syria Internat. 40 n= 38 3779 n= 38 3777 31= 1.2 1.3 0=38.4 27.3 Os= 1.0,===, 1.2 ±s = 13.0 14.6 -E. 35 min. .3 ' .1 min.= 9.3 .2 non. 5.6 , 39.1 000. 74.3 99.9 Ú ,,b_ 30

o25 uj20 Fig.416. Frequency distributions of o texture, pH, organic carbon content, W r,94 and cation exchange capacity in soils of 0 Syria(columns).Curvesshowthe international frequency of the same .2 .4 .8 1.6 3.2 6.4 12.825.6 2 characters. ORGANICC,% CEC, me/100g

Figs 417-421. Frequency distributions Syria Internat. Syria Interne n 20 1765 35 n, 18 n=1958 of nitrogen, phosphorus, potassium, 0, 4.14 4.27 0=2.98 0. 3.14 ±s=.76 1.15 05= .83 ±s ,.87 min, 2.82 .60 30 min.=.88 calcium and magnesium inoriginal max.= 5.55 7.45 max.= 4.22 L.. max.= 6.51 wheat and maize samples and respect- 25 ive soils (columns) of Syria. Curves 20 show the international frequency of 15 the same characters. 10

5 6 2 5 6 N content of wheat,% N content ot maize,%

Syria Internal' Syria 1nternat n= 20 1765 60 n=18 1958 0= .121 .133 0.134 .135 no. .078 .084 06 .101 .088 non,. .040 .009 11110.=.023 .008 max...379 1.023 max.= .480 1.657

.005 .01 02 .8 16 32 .64 1.28 .005 .01 02 .04 .08 .16 _32 63 108 Fig. 417. Nitrogen, Syria. N in wheat soils, % N in maize soils,% 324

4 40 Syria Internat. Syria Internot. Fig. 418. Phosphorus, Syria. n. 20 1765 n= 16 1967 35 5..38 .30 35 5=28 .33

4.C.. 9s..09 .12 to= .07 .10 min.= .24 .05 minx .20 .05 re' , 30 30 t max...56 1.02 rnax.=.42 1.04 0.25 25 3- rt"' 20 20

815

EC_ 10

,e P content of wheat% P content. of maize.%

Syria Internat. Syria Mternot n= 20 1765 n. 18 1967 35 5= 19.5 20.2 n=16.6 22.5 24.6 24.7 to =19.1 33.0 min = 2.0 1.0 min.=4.4 0.1 max =111.9 271.4 rnax..82.0 656.0

>- 20 CD

D 15

LU u_CC 10

2.5 5 10 20 40 80 160 320 640 2.5 5 0 20 40 BO 160 320 540 P in wheat soils, mg/I P in maize soils, mg/I

SyriaInternat. Syria Internat. Fig. 419. Potassium, Syria. n o 20 1765 n= 18 1967 30 004.0 4.0 0 5= 3.7 3.1 neo 0.7 1.0 tO. 1.1 1.0 t min..= 3.0 0.9 min.= 2.2 0.6 825 max.= 5.6 6.8 25 8106 .0 5.3 6.7 a 20 o3- Lu 15 o ix 10 U_

5 6 2 3 5 6 K content of wheat% K content of maize:a/0

SyriaInternat. Syria Internat no 20 1765 n. 18 1967 60 465 365 5= 419 330 so 226 283 55 = 249 356 min, 566 20 min.. 136 18 max o 965 2097 max = 887 5598

25 50 100 200 400 BOO 16003200 6400 25 50 100 200 400 800 1600 3200 6400 K in wheat soils, mg/I K in maizesoils,mg/I

The NaHCO3 extractable phosphorus contents of Syrian soils as well as the P contents of plants are, on average, at the normal international level while varying substantially from one sample to another (Figs 7 and 418). Only moderate amounts of phosphates were applied to the sampled crops (wheat: 7 ± 11 and maize 7 ± 12 kg P/ha). The mean exchangeable potassium contents of soils and K contents of plants are at the average international level but the ranges of variation are quite wide (Figs 8 and 419). No 325

40 Fig. 420. Calcium, Syria. Syria Internat. Syria Internat. n. 20 1765 35 n= 18 1967 3..41 .43 =.60 .47 -.6 ts../0 .71 xs../5 .20 23 30 min...22 .11 0 min.= .28 .09 max.=.62 1.68 max...85 1.68 4-, Cs 25 25 7- 20 20 a15 15 E 10

1.6 Ca content of wheat,% Ca content of maize

SyriaInternat Syria Internat. n= 20 1765 n= 18 /967 .9159 4671 5 0. 7618 3450 ±3 3391 3076 X3= 2763 2815 min.= 4900 110 min. 3310 10 max = 17700 21930 max./6910 17995 30

100 200 400 8001600 3200 6400 12800 25600 SO 100 200 400 8001600 3200 6400 12800 Ca in wheat soils, mg/I Co in maize soils, mg/I

Fig. 421. Magnesium, Syria. 40 Syria Internat. SyriaInternat. n=20 1764 n=18 1967 4.162 .172 0..263 .251 Z 15= .051 .060 Xs=3399 .119 Ú min.= .109 .044 30 min...159 .036 max.= .315 .948 max...476 1.125 a-, 25

Mg content of maize,%

40 40 Syria Internat Syria Internat. n=20 1764 35 n=18 1967 7= 911 489 0=1006 445 Xs= 586 437 no= 510 462 non. 108 10 30 30 min.= 268 max .2620 3298 non .2359 64901

ú 25 25

20 70

LU 15

OX 10 U_

25 50 100 200 4 0 BOO 1600 3200 25 25 50 100 200 4 0800 1600 0200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

potassium fertilizers were applied. The national mean values for exchangeable calcium and magnesium in Syrian soils are among the very highest recorded in this study, but the average Ca and Mg contents of both original indicator plants correspond to their respective international mean values (Figs 9, 10, 420 and 421). See also texture and CEC of Syrian soils (Fig. 416) and the effect of these soil characteristics on the relationship between Ca and Mg in soils and plants (Figs 12 and 14, Section 2.2.4). 326 8.4.3 Micronutrients

Boron. On average the Syrian soils and plants contain somewhat more B than usually found in the soils and plants of this material (Figs 22 and 25). Although the B content of soils and plants varies considerably from one sample to another (Figs 422 and 423) and several high (Zones IV and V) plantsoil B values were measured, these are still much lower than those found in neighbouring Iraq. The differences in the B contents of soils and

10 IN:IMMINIIIMENVEIIIM MIRE 1MIIMMINIIIIIIII VIL4LeWIMIGIPUBIRMIIIM MI=MIMINI 8 IMM.IiMIENE MEMMEIMMIIII EMEIMMEMMEIIIMI MIIIMIll 6 IMELEMEMMILMIIII =MENEM 5 0..108± 4 immassu mu E 3 998+242MMEEMI 1E111 C- 20I. MIME Alin 1111111 o (a)

1 'MallI MMIwIEN11!'.EMI' 11111 liMh=MINCIMPPAIMIlhiCkIII=MEMINENffill=MailliarPMOMmiummail Ell I=IMEMPEITIMIMME1111111 111=111111-11121AMMEM=MEMII o 11=MTBNIiill MENIIIIIII o 3 Imiwoo Eno merlon co II El111111 1111143 2 1 AI li! Fig.422. Regression of B content of I. pot-grown wheat (y) on hot water NI Ill soluble soil B (x), Syria. For details 1 . .2 .3 .4.5.6 .810 2 3 1.56 1111 of summarizedinternationalback- B in soil, mg/l (Hot W. sol.) ground data, see Chapter 4.

100 111101111IMIUIIIMIE =I= MINIM KVAliiIMMEMNIMIMMITil 80- Me=1MMENIEN NM NMMIME 60 IZEEIN I=MINHEIDN MI EMI 50 IIIMICEIEEMME300.3:=MEMMEININ 40 E MIEN.iii72 Mg an E 5500 CL30 1121EINENII MEEPAVIII IIr'0715- Illiiii 1=1111

il 1111 Mraml 1111011==1:11 114. MN sio=1miNnaiiiMMe=1MMErgilelii11.111M=MMERNIVAIMPSEM iiiME 11MMENCIIIIIolmmamouracom , 11111111 m 1 IIMENNIN11111 MMEIIIII

2 HM111111111 MMINI

I11111 inMN Fig. 423. Regression of B content of 08.1 .2 .3.4.5.6 .810 2 3456 810 pot-grown wheat (y) on CEC-corrected C EC -corrected B in soil, mg/I (Hot w. sol.) (hot water soluble) soil B (x), Syria. 327 plants between the irrigated and non-irrigated sites are smaller than those found in Iraq. The present limited data from Syria do not indicate any serious problems due to shortage of B but at some locations a response to applied B may be obtained (see also Section 2.7). Copper. Syria's position in the "international Cu fields" (Figs 27 and 29) is central. Since only a couple of sample pairs fall outside the normal Cu range (Figs 424 and 425) no problems concerning this micronutrient are to be expected in the areas sampled in Syria.

22 S Inte na 20

18 5°3°811411111111111 E 8:16E111111111/111M11111111111

**É; w 141E11111111 111111d1111111111 .c 15 1211111111111 1111111111111111/E1 1,1011111111111 11111Elik1211111 8811111111111 1111111PAN1111111 051111111111111111/11/111101111111

41111111111111M211111111L1111111

Fig. 424. Regression of Cu content of 2 pot-grown wheat (y) on acid ammonium 111111111111611l!i111111111111111 acetate-EDTA extractable soil Cu (x), .6 .810 2 4 6 810 20 40 6080 Syria. Cu in soil, mg/l (AAA c EDTA)

2 C i Sri 1 intern

3;AMIDE -Y72-4 11111111 IIINI twaS1118111 mi 111111

) IBEM,,,,ll ...i.,,,,11111

Fig. 425. Regression of Cu content of !.;,,...... ,..... pot-grownwheat(y)onorganic _...... ,.....,,..,,, carbon-corrected (AAAc-EDTA extr.) soil Cu (x), Syria. Org.C-correctedsoiCU,mg/l (AAAc-EDTA) 328

300 Fig.426.Regression of Fe content 275 of pot-grownwheat(y) 2 50inzEMILWARIMEI onacid 225 ammonium acetate-EDTA extractable 200 soil Fe (x), Syria. 1 75=:111111111THERININIII E 150 o_ 125 o 100 o 90 80 70 o 60 50

30 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (4AAc-EDTA)

Iron. The Fe values measured from pot-grown wheat samples grown on Syrian soils vary from 46 to 95 ppm and approach the international mean closely (Figs 31 and 426). The AAAc-EDTA extractable soil Fe values are below the international mean. On the basis of these limited data Fe deficiency should not be expected to be common in Syria. However, it has been reported in the deciduous orchards of the Damascus Ghouta and in certain areas along the eastern fringe of the Anti-Lebanon mountains (Loizides, 1967). Manganese. Low Mn availability is typical for countries with alkaline soils and Syria is no exception (Figs 33 and 37). Almost all the Syrian plant and soil Mn values are lower than the international means and about 25 percent of the sample pairs fall in the two lowest Mn Zones (I and II, Figs 427 and 428). Low Mn availability may be a factor limiting the normal growth of crops in Syria although extremely low Mn contents were measured in only one sample pair. Molybdenum. Despite the high pH of Syrian soils the average Mo content of pot-grown wheat grown on these soils only slightly exceeds the international mean (Figs 15, 20, 429 and 430). The low soil Mo values (mean 0.119 mg/I) given in Fig. 429 (due to low extractability of Mo by A0-0A in alkaline soils)is doubled (mean 0.235 mg/1) by correction for pH (Fig. 430) and the apparent contradiction between the analytical results of plant and soil analyses is resolved. The corrected data indicate no Mo deficiency in Syria, rather, both the plant and soil Mo values are seen to be normal or somewhat high. Zinc. Both the methods used for extracting Zn from soils give similar estimates of Zn levels in Syria (Figs 41, 43, 431 and 432). High Zn contents were recorded for a few locations only, the majority of values are quite low, a few falling in the two lowest Zn Zones. Although none of the values indicates a very severe Zn deficiency it is possible that a favourable response to Zn fertilization would be obtained at several locations. 329

=WEIIM 11.1110.001111,111.111.017Maera1.10 Fig.427. Regression of Mn content 2000 of pot-grown wheat (y) on DTPA ill 111:1111111111111 extractable soil Mn (x), Syria. 1000MII :CS.-IMI la 11CM/111.XLII M.A.'S 00=1 WM. IMMIllIMENIA 800BS IIMIMMMIIIIMIIII01111 =ME 11=6VIAmmilanirrtwiiireirkArreimiANIMMIMTIMMAMEM111111111 MU 600UNEMINEMOIM=MMLINIIIIIMMIMMig E 40011 =1...1111.=IMMEReilliMailli 30011=11.IIIIIMMPHIIMMII 200HEM111111=mounimem .c o 1001111I...... 111111111111:411 80... IIIIIIIMMINIL IIIMEMINSEM11MPP.MIIMIIIMPN o 60IniM piniiIIMMIMMIgninMMIEM 40iiIMMIES - leeTiliallIMMEMIIIIMMIER 3011111..1 -gliummullummis 20UWE IIIMINI111111111111

10MI .11111\311111111111...... IMII11IMMINIMMIN _ =1In NMI MEN117111S7VMMENNO awm= IMIMMIILMMIRMIIME in r=im 2 34 6810 20 30 40 60 80= 200 300 Mn in soil, mg/l (m-PA)

111=1MMINtiOMMID=IMAM NMI Fig.428.Regression of Mn content 2000 of pot-grown wheat (y) on DTPA Mn extractable soil Mn corrected for pH 1000 mMIMPORWOREME11==7704PMNIMMEMMININIMMisalli RUMP 800 MEIMMEMEM0=M=WIINIMENIIMMIMMINiiErielia (x), Syria. 600BB E 111111114MIMMIIIMIS1=111411IMINIIM11111111.111n11 o_400illMEMIUMMEILEMEADMIUMpirilliii o300 200Imolommosnippownw o 11\111111111M1111111011111111 100 raa 80 III101InEMENE 11PlitidIMIMBMMMOOKIiinni o 60NIMMIIMMEMONEMMI11=kENEMMi 40 30 20upsehmimsionimmlop

10inlinmalponum 8 MIMMINIMINI 113MMEMINI IN1.111 2 3 46 810 20 301,060100 200 300600 pH-corrected Mn in soil, mg/l (DTPA)

8.4.4 Summary

Most of the Syrian soils are medium to finetextured and have medium organic matter contents. The other essential features aretheir high alkalinity, high cation exchange capacity, high CaCO3 equivalent and relatively highelectrical conductivity. High Ca and Mg and medium N, P and K contents are typicalof most sampled Syrian soils. Most of the micronutrient values measuredfrom Syrian sample material are within the "normal" international range. Due to thehigh soil alkalinity the availabilities of Mn and Zn to plants are low, and the most likelymicronutrient problems in Syria are shortages of these elements. At some locations a responseto B fertilization may be obtained 330 Nionl M o N Fig. 429. Regression of Mo content of 6 AMIN. pot-grown wheat (y) on ammonium oxalate-oxalic acid extractable soil Mo 3 toga (x), Syria. 2Ewait E 1 ImmilMEMmorns MIIMMENI71=Q11111 _c4., .611MEMMIrMEMEN

4-- o .3M=111111/%1M1111 E111111111111111.11 ve 2MMME=MIIME141ZUMMEM=1MINEMILIMNII OM MMWAIIMIONEENIIMMEMMINIAWAMMMIMIMEMIMMIMMMIk .06rMMOIMENIIIMWMINIMONNIIIIMIMI .03 .02

01 III

.01 .02 .03 .06 .1 .2 .3 .6 1 23 Mo in soil, mg/l (A0-0A)

EMP1 MSIMM I MNIMENZIMMEI MIEMBEN 6mnnunsitommumnomiumummmilimBliiWISMI1111 IMIMEEINEEKEEMIIIIMMEMBEHMEMI 3 2

°- 1 16111216 -5.6MAIREIPEI ohummuloollot =M1111111Wd=nommumm.

.03 .021111111111111111 Fig. 430. Regression of Mo content of .01momkomm .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pot-grown wheat (y) on pH-corrected pH - corrected MO in soil, mg/l (A0-0A) A0-0A extractable soil Mo (x), Syria. 331

Fig. 431. Regression of Zn content of 200 pot-grown wheat (y) on DTPA extract- 77 able soil Zn (x), Syria. 5 321 tog 100 90 80 70 E60 °- 50 o 40 "6 30

o20 N

10 9 8 7 6 lU rz

.1 .2.3 .4.6 .8 1 23 46 8 10 20 3040 60 100 Zfl in soil, mg /l (DTPA)

200 r a Irte n. =3f 3 z .6 133 s ± 4 84. 8 100 C.51 90 80 70 60 E 50 40 o 30

" a 20 o

NJ 10 9 8 7 Fig. 432. Regression of Zn content of 6 pot-grown wheat (y) on AAAc-EDTA III extractable soil Zn corrected for pH .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 (x), Syria. pH-corrected Zn in soil, mg/l (AAAc-EDTA) 332 8.5 Turkey

8.5.1 General

The sampling sites of the 250 wheatsoil and 50 maizesoil sample pairs collected from Turkey, are well distributed over the agriculturally important areas of the country (Fig. 433). Five out of the seven natural regions (Black Sea, Marmara, Aegean, Central

-----t

\ 1 --

i .,..i.

' E ..; G7 ' aE .. I r- \t ..., _...--' .... ___ __I----4;------T 0. .\ .

,; 14 __.------F----. _a__ . \ i ! `.. 2 I i

1 d t 1E 1 !

i,..: g 1E i.! IV .,

4 E ., r) I .,4. 1 ¡E** lo0''t I..,,i. z F..:-T- R .. H-744 ' ------tl't 'Ve" 0-, 1l'N-----/i ' ì L../

'...el. g! ' If ,

'i \ I -0 / --1-- *----*-1*'' . T. i

i ...; I 1 . r IV C

1 t 1 R". ,-- _,..

0 E5' 5 - :, 1 1 . .1 /-r-

--- _ -,,1 el,. Il

IL 1_ ,,, [?I ' ",A

I Fig.433.Sampling sitesin Turkey. i ------__ I_ The last three numerals of each sample 1 pair number are given (dots = wheat k fields, triangles = maize fields). 333 Anatolia and Eastern Anatolia) are represented and these constitute about four-fifths of Turkey's arable land area. Of the total Turkish plantsoil sample material, 87 percent had been collected from non-irrigated and 13 percent from irrigated sites. About two- thirds of the sampled soils were classified by FAO/Unesco soil units: Fluvisols (81 soils), Luvisols (62), Kastanozems (17), Vertisols (16), Lithos°is (12), Rhegosols (2), and Xerosols (2). The sampled soils vary extensively in texture from TI 17 to 85 (Fig. 434). Highly alkaline soils, pH(CaC12) > 7.5, predominate but some neutral and a few relatively acid soils are included. The organic matter contents of soils are usually rather low but their cation exchange capacity tends to be high compared to the international distributions of these properties in the whole material of this study. Relatively low 'electrical conductivity values and sodium contents are typical of most Turkish soils but the values for CaCO3 equivalent are usually quite high (Appendixes 2-4).

Fig.434. Frequency distributions of Turkeynternat Turkey Internat. n. 303 3764 n. 300 3783 texture, pH, organic carbon content, R.L9 44 R. 7.40 6.54 0 os.13 19 os..57 1.12 and cation exchange capacity in soils min.. 17 9 35 min.. 5.20 3.62 non. 92 'nao.. 8.55 8.55 25 of Turkey (columns).Curves show 0 theinternationalfrequencyofthe 25 same characters. 20 15

10 20 30 0 59 0 70 80 93 TEXTURE INDEX pH (CaC12)

Turkey In1e1nat 5 Turkey Internal n= 300 7779 n. 300 3777 R. 9 1 1,3 0. 31.5 27.3 os. 4 1 2 13.s 14 6 min.. 2 .1 min..7.7 .2 max.= 78 39.1 max - 54.2 999 30

25 20

w20 15

wo 15

.2 . 1.6 3.2 15 ORGANICC,% GEC, me/100g

8.5.2 Macronutrients

The total nitrogen contents of Turkish soils are low compared to most other countries (Figs 6 and 435). The N contents of the original Turkish wheat are also internationally rather low but those of maize tend to be high. In both cases the N contents vary substantially. The sampled maize crops were fertilized with nitrogen (61 ± 50 kg N/ha) at double the rate applied to the wheats (34 ± 30 kg N/ha). The relatively higher N contents of maize as well as the wide variations in the N contents of both crops may partly be due to fertilization. The NaHCO3 extractable phosphorus contents of Turkish soils, especially of wheat grown soils, are internationally somewhat low (Figs 7 and 436). Although moderate 334

40 40 Turkey Internet Turkeyinternat. Figs 435-439. Frequency distributions n n249 3765 35 5 n= 50 1959 5= 4.12 4.27 3.28 3.14 of nitrogen, phosphorus, potassium, es. .78 1.15 -re= .63 30 min.. 1.92 .50 min '.59 173 calcium and magnesium in 30r111,1..6.73 7.45 max.= 4.56 551 original k25 25 wheat and maize samples and respect- 9- 23 20 ive soils (columns) of Turkey. Curves show the international frequency of the 13-1 15 a same characters. Ic'cj U_

2 AiL N con ent cl wheat.% N contentofmaize,%

73 Turkey Internet TUr Key internat. 6 r- 249 1755 0 n.50 1950 12= .994 ,133 = .1'9 t se .034 .084 05..047 .088 min.= .021 .039 min - .047 noc io max = 788 1 923 max -,270 1.657

.91 .32 .04 6 .19 .3 22 .000 .41 02 .0/. e E .32 6 t29 N in wheat soils, % N in maize soils,% Fig. 435. Nitrogen, Turkey.

40 Turkey internat. Turkey Internat. Fig. 436. Phosphorus, Turkey. n . 249 1755 3 I 50 1907 = .3@ R ..30 33 48 .09 .12 es = .06 10 .14 830 min.= .05 , Till, . '17 05 ml,.- .576;. 1.02 1880.. .45 02 3,00 o -0.25 zL) 20 ej 15 U.I CC 10

.3 5 .6 P content of wheat.% P conlent of rnaize../o3.6 Le Turkey Internet, Turkey Internat.

35 n. 249 1765 5 n 50 1967 11-= 10.9 20.2 I= 17.5 22.5 es= 7.9 24.7 14.4 33.0 min. 1.0 1.3 min..3.0 0.1 man.=139.6 271.4 37 mae. 73.2 655.0

2.5 5 20 40 80 100320 540 2 0 20 P in wheat soils, mg/I P 71 maize soils, mg/I dressings of phosphates were applied (wheat: 18 ± 13 and maize: 6 ± 11 kg P/ha) the average P contents of both plants remain below the respective international means. The level of exchangeable potassium in Turkish wheat soils is almost double that of maize soils (Fig. 437) but, on average, the K contents of Turkish soils correspond closely to the respective international mean for soil K (Fig. 8). Practically no potassium fertilizers were applied, and the national mean K contents of both indicator crops remain rather 335

35 3 Fig. 437. Potassium, Turkey. TurkeyInternat, Turkey Internat. n=249 1765 n= 50 1967 30 8=3.3 4.0 30 0=2.7 3.1 26= 0.6 1.0 S= 0.8 1.0 min.= 1.0 0.6 .-C-' min.= 1.8 0.9 25 max, 4.2 6.7 825 max.= 5.4 6.8

5 6 2 3 4 5 6 K content of wheat.% K content of maize,%

Tu r key Internat. TurkeyInternet n= 249 1765 n,, 50 1967 7=411 365 7= 214 330 0s= 115 356 0 ts= 210 283 min.= 58 20 min.= 46 18 max.= 1338 2097 max.= 556 5598 25

...a9/21 41 / 25 50 100 200 400 800 160032006400 25 50 100 200 400 BOO 32006400 K in wheat soils, mg/I K in maize soils, mg/t

4 Fig. 438. Calcium, Turkey. Turkey 8a8880t. Turkey 'Ioteroot. n= 50 1967 3 n=249 1765 35 7= .57 .43 6, .57 .47 35= .22 .17 es = .14 .20 30 min...33 .09 S'30 min.= .21 max.=1.68 1.68 max.= .90 1.88

(71 122 25 7- t.) 20 oD 15 Lai Ffr 10

.2 .8 Ca content of wheat, °A Co content of maize,%

40 Turkeyinteroot. Turkey internat. n= 249 1765 n= 50 1967 et 5 3. 6408 4671 35 X. 5411 3450 es=2140 3076 3s. 2282 2815 min.= 660 110 minx 1290 10 max.= 13160 21930 max.= 10375 17995 30

?..1) 25

/A. 100 200 400 800 1600 3200 6400 12800 25600 so loo 200 400 800 1600 3200 6400 12800 Ca in wheat soils, mg/I Co in maize soils, mg/I

low internationally. Apart from the lack of potassium fertilization, the low K contentof wheats may be due to the effect of soil pH which was considerably higher in wheat soils than in maize soils (means 7.52 and 6.79, respectively, Appendixes 2 and3).For the effect of pH on the plant K-soil K relationships see Fig. 11 and related text inSection 2.2.4. 336 On average, the calcium contents of Turkish soils and plants are high and the average magnesium contents correspond closely to the respective international means (Figs 9, 10, 438 and 439).

40 TurkeyInternat. Turkey Internet Fig. 439. Magnesium, Turkey. n= 249 1764 35 n= 50 1967 0..174 .172 6= .290 .251 es= .069 .060 es =.131 .119 min.= .067 .044 3 min.= .134 .036 max.x .948 .948 max.= .614 1.125 25

.05 2 .05 Mg content of wheat,% Mg content of maize,%

Turkey Internat. Turkey Internat. 5 n= 249 1764 35 n= 50 1967 5= 495 489 0= 469 446 es= 305 437 es= 363 462 min.= 70 10 rem.= 95 1 max.= 1755 3298 max= 1742 6490

20 7-

LL 15 o LU 10 u.

25 50 100 200 400 800 1600 3200 12.5 25 50 /00 200 400 800 1600 32006400 Mg in wheat soils, mg/I Mg in maize soils, mg/I

8.5.3 Micronutrients

Boron. Wide variations in B contents are typical of Turkish soils and plants (Figs 440 and 441) but high B values dominate and in the "international B fields" Turkey clearly stands high (Figs 22 and 25). In spite of the variation, 84 percent of the plantsoil B values are within the "normal" B range (Zone III, Fig. 441) and only two percent of these fall in the two low B Zones, one percent in each. The lowest B values occur in the Black Sea, the Aegean and the Marmara regions. The majority of the high (Zones V and IV) B values are found in Central Anatolia. The highest B values were measured in sample pair No 46673 which came from Ankara Province in an irrigated field with highly calcareous soil, pfl(CaC12) 7.9. Although the effect of irrigation is not as marked as, for example, in Iraq, it is apparent that in many cases the practice of irrigation is partly responsible for the high B values found in Turkish plants and soils. The relative frequency of high 337

Fig.440. Regression of B content of pot-grown wheat (y) on hot water 80tbd 60 n=298 3 soluble soil B (x), Turkey. For de- 50- -46 59 50 07 tails of summarized international back- 40 -*r_s=110±1 5 ground data, see Chapter 4. .= 63 3074+440r 851". 4 20

CD 10 "68 6 5 4 o 03 2

BI 1 .1 .2.3 .4.5.6 .81.0 2 3 456 8 10 B in soil. mail (HotW. so!.)

Fig. 441. Regression of B content of 100 pot-grown wheat (y) on CEC-corrected 8Oß 5 (hotwatersoluble)soil B (x), 608±-sx6 9 -±5 50 6 9 4 50 7±s=0 9090 0 3 0 Turkey. 40 E .126-1-5 3x r 8 o_ 30 r. 0 905 * 02 is 20 a) 10 8 . . . 4 3

.08:1 .2.3 .4.5.6 .8 1.0 2 3 4 5 6 8 10 C EC - corrected B in soil, mg /I (Hot w sol

(Zone V) B values is about three times as high for irrigated as for non-irrigated sites, while none of the sample pairs from irrigated sites fall in the two lowest B Zones (I and II). See also Section 8.2.4. Judging from these analytical data, B deficiency occurs in Turkey but is not widespread. The analysed B values are usually "normal", but at several locations attention should be paid to possible disorders due to an excess of available B caused either by naturally high B contents of soils or induced by high-B irrigation waters. See also Section 2.7. 338

22 Tu k nter no Fig. 442.Regression of Cu content of pot-grown wheat (y) on acid ammonium 20 n 8 3538 48± 780r2 7 acetate-EDTA extractable soil Cu (x), Y± c 18 Turkey. Q.1611111171t111ElitilrE1111 o 1111111111111MIIIIIMMIllr

.6 12111111111IMIIIIMI

10 o8aliordno 06J 111111111111MOMME11111

4111111M1111111111E1111

21111Enifillain 111111111D1

o .2 .4.6 .810 2 4 6 810 20 40 6080 Cu in soil, mg/I (.4AAc- EDTA)

Fig. 443.Regression of Cu content of 22Cu i pot-grown wheat (y) on organic carbon- .yi m. 20, corrected (AAAc-EDTA extr.)soil Cu (x), Turkey. 18ryi::s%INN-9-53-insiii noll E 16Y=307E1111111 rogoliiiiiill 1111111 o.

10 11111111 1111111111111

8 ...., :...,r4.mum o 6 IIIIIMIKENINE1111111

4 111111kalkillIME1111111

2 111111011101111 , MIN

O .;allImilll .2 .4.6 .810 2 46 810 20 40 60 100''' Org.C-corrected soi CU,mg/l (AAAc-EDTA)

Copper. The Turkish national averages of plant and soil Cu contents correspond closely to the international respective mean values in this study (Figs 27, 29, 442 and 443). Relatively small variations in Turkish plant and soil Cu values are typical, and no extremely low or high Cu contents were recorded. The present analytical data suggest that problems due to shortage or excess of Cu are unlikely in Turkey. 339

Fig. 444. Regression of Fe contents of 300 275fmNimessim pot-grown wheat (y) on acid ammonium 250 acetate-EDTA extractable soil Fe (x), 225 riorionommulipms Turkey. 200\ "iiiiiiiiiilLINOMMIN1101111 175\`"'' ? 'illiklMiliiiiiiiIIIIII E 1 50Ilia IIKII111111111111111 Q. MLIIIIIIIMG111111 -125

c100Mil111111111111111 1-5 90 =XiMIA11111 011111=1 t 80 MINI= plINSWII MiliiiiiigNiegaililli

50 111010M11111111111 I 40ardigiiiiIIIIiII.I

30 liopromilE ..,111111 III MINI'AMIN 10 20 40 60 100200400 600 1000 2000 Fe in soil. mg/1 MAAc- EDTA)

Iron. Compared to most other countries the Fe content of Turkish soils and plants is low (Figs 31 and 444). More than 20 percent of the Fe values fall in the two low-Fe Zones (I and II), and some of these show distinctly low plant and soil Fe contents. It seems likely that in many places crops sensitive to Fe deficiency would respond to iron fertilization. Almost 90 percent of the low (Zone I) Fe values occur in Central Anatolia and half of these are in the Province of Konya. Manganese. Turkey's placing in the "international Mn fields" tends to be low (Figs 33 and 37). However, in spite of the generally low Mn content, only about two percent of the Turkish samples fall in the lowest Mn Zone (I), and none of these indicates any severe shortage of available Mn (Figs 445 and 446). Molybdenum. The two Mo graphs (Figs 447 and 448) give dissimilar pictures of the Mo status in Turkey. This is because A0-0A is unable to extract Mo from alkaline soils in quantities related to its availability to plants (see Section 2.3.2.2), so distorting the relationship between soil and plant Mo. After pH correction the soil Mo data are in better conformity with the plant data and only a few sample pairs indicate shortage of available Mo. Several relatively high, but not extreme, Mo values were recorded from the samples. No distinct geographical division between low and high Mo areas could be drawn but, for example, of the 62 pot-grown wheats with Mo content <0.1 ppm only two were grown on soils originating in irrigated sites. See also Section 8.2.4. Zinc. The standing of Zn in Turkey is one (if the very lowest recorded in this study (Figs 41 and 43). Irrespective of the method used for extracting Zn from the soils, the great majority of extractable Zn contents of soils and plants are below the respective international means. About 35 percent of the sample pairs fall in the two lowest Zn Zones (Figs 449 and 450) and only a few in the high Zn zones. Low (Zones I and II) Zn values were recorded most frequently in samples which came from Central and Eastern Anatolia where alrnost every sample pair gave such low figures. About 20 percent of samples from the Black Sea, Marmara and Aegean Regions fell in the low Zn zones. it appears, therefore, that Zn deficiency occurs cornmonly in Turkey. 340

JIMEMINLAIIIMi IMINICEMMEIN, Fig.445. Regression of Mn content of pot-grown wheat (y) on DTPA LVM,01,C31,4t1VMENIWf/ESECal.1111111MII1111=11111111.11111111N1 extractable soil Mn (x), Turkey. 80011 MilaMinTirn=r7MiWieVird4rmilogN=111111ISIIIM1 600II E 400 St300;km.1111111111111111111111111 15200 11111111=11111111MIE1 11111111111111111111:41, =MCMIIIIIIMIPUIMMISIMMEZIMIMLIEMIMII ,3, 80UNIMINIMMIMISIMISMICTOMIergEMNI.11111aTilEMI MENIMMINMENINSPECaMianglat...iiii1MMIMINK 5 6° iiiiiiiiiniMWrr.iidli;VVWSINCHZkininiii 40mimmimmalimaggrumigniulinmq0."I :Alibi11111111111 1111111111IIMIMIMMIlialMIOSTAMIMMEIISVA1111MM 1EaNIMMINInli1=1 1 2 3 46 810 20 30 40 6080100200300 Mn in soil, mg/l (DTPA)

2000 =121MONI IMMIIMIMMININE Fig.446. Regression of Mn content of pot-grown wheat(y) on DTPA 1000 1911111111111111 extractable soil Mn corrected for pH 800E .....Encraml.nangurF...... ".....mcw.m..m.....-mLnin (x), Turkey. 600III MIIMIIIIIIIIII MEIZEPAgtIgacnIc IMMINO118 E nwwinagaummmoztommumwmpri 400rEIMINEFOIliral 1 300 .c 200 1 100...... 11.1,_,...... 11111101001.11 80...... ,..,...... ,.. 6011111=1- EJEISPIITIIIIMIMIIIIMIMMIIIIIIII o 11=ga'arcitANASIMUIIMMISIMMIIMMEENII 40IIMIMIWIEMPNEMIlmsomimmui 2 30111=MatIll MEW illi ir 20

lo .... MMEIN111 ,a.. MIMIIIIII=101111Naill=11MMENZE..... ==MMIN irIII...... IIMMEM=L1111111EZALMMERNMI i MIMMIMAI 2 3 46 810 20 30 40 60 100200 300600 pH - corrected Mn in soil, mg/l (DTPA)

8.5.4 Summary

The Turkish soils vary greatly in texture. Most of the soils are alkaline and low in organic matter but have a relatively high cation exchange capacity. Compared to other countries participating in this study the N and P contents of the soils are usually low, while those of K and Mg are at a medium level and Ca contents are high. The most evident micronutrient disorders in Turkey are those due to deficiency of Zn. The level of Fe is also low as is that of Mn. Low B values were recorded occasionally. In general, the B and Mo values are "normal" but some of them are very high. Problems due to shortage or excess of Cu are unlikely. 341

Fig. 447. Regression of Mo content of o pot-grown wheat (y) on ammonium 6111111MIIIIIIIIIMI=1111111111111 oxalate-oxalic acid extractable soil Mo monswaumaima gillm (x), Turkey. 3 M- 2=MERIN= 111111116111111111111 ==WEINMMISIMMZITAIIIII:MilM=....M.M.1 1111111111,4111 .6 IMMEMMUOMMEMENNIMMUIallIMEN

.211111111011111111111111MMI AISIIICOMMIUM1=mN ONWM/11=1001.11111MEMMEMETEMENISIIMIMVPINIMMEMINIIMIMMIM=MEMENIMPILIII )6MIMMIIMMININIIIMIII=11110111 MI11111M /111111111111111111111 )3 )2 I Lit, )1 IIIIIIL .01 .02 .03 .06 . .2 .3 .6 1 23 Mo in soil, mg/l ta 0-0A)

immENNImslii== M oIIMMUNMel.....TrAT:t=IIM=MIMUM111111111111 INI111111111=11111 6MII1111111MELIMIM1LIIn . 3 37 111111111111MININ MENIIIINICEMILISMINNIIL-±s -0 3220. ammo= sum IIIIIIIIIIMIIIIIIIIIIRCEILIEIIIIII Il 3 111/1 21111111-20%6a 111111Effilliiiiil

Q. 1111111 11111E11111111/1111 Q. IMI=MOI=MMINISM1172111MMEMEI,11==..IIMIIMM.qmWMIMmuror. MEMNIIMI=MMEINZEIIMMIIMMIZO*5111;1111= MUMBIBI=MMIMMEMENIONIMMIMIWOMM=MMEMEM .6NEIME111MENNIIIMMIMIINIVAINSINIMMINNI MOINNIIMMMINIMEINEMMEINFA2M11MIIME 3 .31 I N IIVA' ' I 2 ilik Ill ,ll o 1111111. RPM'.-f 1111111101h" 11.L.N .1ord.g...... 111=EMMEMW6S...... -Ail 2:::=E-Zannia==MMIUUMEIMG: .06mmullmormommirmommulmmr..

.03 Fig. 448. Regression of Mo content of .02 pot-grown wheat (y) on pH-corrected ,i, 0\ . .01 d i A0-0Aextractablesoil Mo(x), .006 .01 .02 .03 .06 .1 .2.3 .6 1 23 Turkey. pH - corrected MO in soil, mg/I (A0-011) 342

200 IMMIMMEIN Fig. 449. Regression of Zn content of pot-grown wheat (y) on DTPA extract- 577 able soil Zn (x), Turkey. 321 log II 100 ItMI 90IIflE011111111 80 u_ 70 E 60 a 50 o 40iiauiiiiiiiiiiiiiiiiiir

20mosolommon

1011111111111111111111 9 8IIIIMM119111111111MMENIIIII=111 7 6 Ill 3E

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zn in soil, mg/ (DTPA)

-I I ten. 3 3

i±'- o 3 1 x, 1=03 o r" 'iiiM iiil iiii '0 ii1111 U12 50 MN In 50 111111 lul sum 1 Il li IIIIII

30 1 WM iii

20 k 111111 Igo...... 0iii11111 10 9 8 MIER, II UM ' 7 -minMI8111-'...: . iIL.:::: Fig. 450. Regression of Zn content of 6 111111IIIIIIIMIE pot-grown wheat (y) on AAAc-EDTA .2.3 .4.6 .8 1 23 46 8 10 20 3040 60 100 extractable soil Zn corrected for pH pH-corrected Zn in soil,mg/I (AAAc-EDTA) (x), Turkey. 343 Africa

9.1 Ethiopia

9.1.1 General Soil and original wheat and maize samples received for thisstudy from Ethiopia came from Shoa (40 sample pairs), Sidamo (32), Arussi (27), Wollega(16), Harar (5), Gamu- Gofa (4) and Kaffa (3). These provinces represent about half onthe arable land of the country. The approximate sampling sites are givenin Fig. 451.

. . .

1 ETHIOPIA i

1 1=6,...1=oremers.==1*0 25 50 100 350 200 250,on i 16 .-- i 1 i A554W,I, I POOSOAT

ASSARA

ANSUM ri----- k ii

6ONSAR /

1 I

IDErE L.. ..

0689tolAPPHOS 1 . k..... 46928 47054-9706J 13186047470,3_,02, .,....,

470421 0 97046-47053 46972-46982 HAP. 47043 - 97047,4,,2 ADDIS ABAB

695, 96916-96919 97100 '6932-9 46921- 46927 47078;271:133..-14=-46971' 68PE 97028-9702

47035-97037 r.,_,40iD 1--J figig::7T'r(7 .8889 96929 , 46931 97061-47077 9708-97099 GOPPAHEI.

47094-47097

881,0

blE,SELL1

....

, .....-....

56 3 .10 Fig. 451. Sampling sites in Ethiopia (dots = wheat fields, triangles =maife fields). 344 The range of textural variation of Ethiopian soilsis very wide, although heavy textured soils predominate (Fig. 452). The national mean texture index (62) is exceeded only by that of Brazil (Appendix 4). The pH values vary widely, from 4.1 to 7.7, but most of the soils show moderate to strong acidity. Due to the relatively high organic matter contents and heavy textures the cation exchange capacities of Ethiopian soils are very high (see also Fig. 3, h and m). The electrical conductivity and CaCO3 equivalent values and sodium contents are generally low (Appendixes 2-4).

45 5 Ethiopia intemot. Ethiopia Internat. Fig.452. Frequency distributions of n.126 3764 40 n= 127 3783 4. 62 44 7.$$ 5.74 6.64 texture, pH, organic carbon content, 30 en.16 16 ts ..87 1.12 min.= 20 9 35 min.. 4.10 3.62 and cation exchange capadity in soils of o max.= 92 92 eran.. 7.65 8.56 Ethiopia (columns). Curves show the international frequency of the same characters.

10 20 30 40 50 60 70 80 ,90 5 6 TEXTURE INDEX pH (Ca C12)

45 Ethiopiainternat 35 EthiopiaInternat. n $.127 40 n=127 3779 3777 =2.1 1.3 =48.2 27.3 In. .9 1.2 .45 .13.8 14.6 t 35 min.= .8 .1 min.= 16.2 .2 8 max.= 4.0 39.1 max.. 79.0 99.9 43 30

0 25 20 11.1 tC3

.2 .4 .8 1.6 3.2 6.4 148 25.6 2 4 8 16 32 ORGANICC,% CEC, me/100g

9.1.2 Macronutrients

In assessing the macronutrient contents of the original wheat crops it must be borne in mind that the Ethiopian wheats were sampled at rather late stages of growth (mainly between 54 and 62 days after planting). See Section 1.2.2. The analytical results of Ethiopian wheats therefore show lower macronutrient contents than they would have done if sampled about three weeks earlier as was done in the case of maize. This may also explain some of the contradictions between the results of soil and original wheat analyses. The mean total nitrogen contents of Ethiopian soils are somewhat higher than the respective averages for the whole international data of this study (Figs 453 and 6). The average N contents of maize also exceeds the international average but that of wheat is lower. Relatively light dressings of nitrogen fertilizer were applied to the sampled wheat and maize crops, 11 ± 10 and 7 ± 11 kg/N ha, respectively (Appendix 5). From an international point of view the phosphorus contents of both soils and plants are somewhat low but the potassium contents are relatively high (Figs 454, 455, 7 and 8). Perhaps arising out of previous experience of fertilization, the sampled wheat and maize crops were given moderate amounts of phosphorus (10 ± 10 and 5 ± 7 kg P/ha, re- 345

Figs 453-457. Frequency distributions Ethiopia sotomot. EthiopiaInternat. n= 54 1765 71 35 35 n= 1958 of nitrogen, phosphorus, potassium, 0=3.053.05 4.27 =3.90 3.14 -Ss= .77 1.15 06=.67 .67 calcium and magnesium inoriginal 30 min.= 1.94 .60 0 min.= 1.59 .88 42 max.= 5.43 7.45 max.= 6.51 6.51 wheat and maize samples and respect- c-c, 25 25 ive soils (columns) of Ethiopia. Curves 7.- 20 20 show the international frequency of Z4.2) the same characters. 01, u_ 5 Pori ar/ 5 6 2 4 5 6 N content of wheat% N content of maize,%

70 70 Ethiopia Internal Ethiopia Internat. 54 60 n= 1765 60 n= 71 1958 0= .176 .133 7=200 .135 =.093 .084 145=.068 .088 min.= .056 .009 min.= .064 .008 50 14105= 360 1.023 max.= .394 1.657

005 .01 02 .04 08 .16 .32 55 1.28 .005 .01 .02 .0 08 16 32 .65 1.28 Fig. 453. Nitrogen, Ethiopia. N in heat soils, % N in maize soils,%

Fig. 454. Phosphorus, Ethiopia. Ethiopia Internat. Ethiopia Internet. 5 n= 54 8765 n= 71 1967 6.31 .38 1.30 .33 *se .08 .12 Ss= .12 .10 min.= .23 .05 rain.= .13 .05 max.= .56 1.02 max...81 1.04 S. 25 >- <2) 20

13.1 S15

P con ent of maize.%

Ethiopia Internat. Ethiopia Internet 5 n= 58 1765 n= 71 1967 =10.0 20.2 R. 18.7 22.5 os' 9.6 24.7 OS= 43.0 33.0 min.=3.0 1.0 min= 1.2 0.1 mox.= 57.8 271.4 max=311.0 656.0

..4A15444 A 2.5 5 10 20 50 BO 160 320 640 75 5 10 M 40 80 160 320 550 P in wheat soils, mg/l P in maize soils, mg/I spectively), but no potassium (Appendix5).For example, Gertsch (1972), when reporting the results of several hundred fertilizer trials, stated that a response to K was found only occasionally in Ethiopia and to a lesser extent than to P and N. The calcium and magnesium contents of maize and respective soils are lightly low on the international scale (Figs 456 and 457) and in spite of the relatively high Ca and Mg 346

35 Ethiopia Interned Ethi010.10 Internat. Fig. 455. Potassium, Ethiopia. n 54 1765 n= 71 1967 30 2= 3.8 4.0 30 9=3.6 3.1 s= 0.8 1.0 ±s= 0.8 1.0 min.= 2.6 0.9 min.= 2.8 0.6 13 25 max.= 6.4 6.8 max.= 5.2 6.7

a 20 2 ILI15 co 10 IL easeel°

4 5 6 2 3 4 5 K content of wheat% K content of maize,%

Ethiopia Internat. EthiopiaInternat. n, 54 1765 n= 71 1967 = 740 365 = 549 330 ±s = 260 283 ±s= 390 356 min.= 127 20 min.= 87 max.= 1336 2097 max.= 2077 5598 o

[1.

(.2

25 50 100 200 400 800 1600 3200 6400 25 50 100 200 400 BOO 1600 3200 6400 K in wheat soils, mg/I K in maize soils, mg/I

Ethiopia Internat. Ethiopia Internat. Fig. 456. Calcium, Ethiopia. 1765 n= 54 n = 71 1967 =.31 .43 =.40 .47 *8 =08 .17 es =.11 .20 rnin.=.16 .11 30 18111...19 .09 8101= 55 168 max.=.71 1.88

2 .4 1.6 .e Co content of wheat. °A Ca content of maize,%

Ethiopia Internet Ethiopia Internat. n= 54 1765 n= 71 1967 5 = 5662 4671 35 8=2292 3450 es =3261 3076 ±s=1608 2815 min.=1250 110 min.= 510 10 mime 11550 21930 max.= 7660 17995

, 100 200 400 800 1600 3200 6400 12800 25600 50 100 200 400 8001600 3 00 6400,32800 Co in wheat soils, mg/I Co in maize soi s, mg/I contents of wheat soils the contents of these elements are low in wheats. This contradiction may be due to the advanced maturity of the wheats when sampled, as pointed out earlier. Other factors which also may be responsible for the strong contradiction between low Ca and Mg contents of Ethiopian wheats and high exchangeable Ca and Mg contents 347

40 Fig. 457. Magnesium, Ethiopia. EthiopiaInternat. Ethiopiaintemoi, 35 n=54 1764 35 n. 71 1967 0,108 .172 R=i .210 .251 7.6= .017 .060 6.5..069 .119 min.= .075 .044 3 min.= .074 036 max.= .162 .948 max.= .449 1.125 25

05 5 .2 Mg content of wheat,% Mg content of maize,%

40 40 Ethiopia intemcit Ethiopia Internat. n= 71 n= 54 1967 35 1764 5 0. 417 446 0= 718 489 ns. 295 462 06= 394 437 min.= 87 min.= 189 10 ma.= 1753 max.= 1432 3298 30 6490

/ ,AL 25 50 100 200 400 800 1600 3200 12.5 25 50 100 200 400 800 1600 3200 6400 Mg in wheat soils, mg/l Mg in maize soils, mg/I

of wheat soils may be soil texture and CEC. The relative availability of Ca and Mg to plants in heavy textured soils with high CEC is low. See Figs 12 and 14 and related text in Section 2.2.4. In the case of Ethiopian wheat soils the average texture index was 68 and CEC 56.7 me/100 g which were the highest national means for wheat soils recorded in this study (Appendix 2). These two soil factors affect less the uptake of Ca and Mg by maize since maize soils are somewhat coarser textured (aver. TI 56) and have a lower CEC (aver. 41.5 me/100 g). For a general assessment of Ca and Mg status in Ethiopia against the international background, see Figs 9 and 10. The majority of the Ethiopian plant samples consisted of local varieties (43 local and 11 HYV wheats; 53 local and 18 HYV maize samples). Differences in Ca contents between the two variety groups were obscure but the HYV of both crops show higher average Mg contents than the local varieties in spite of the change-around in the exchangeable Mg contents of the respective soils:

HYV Local HYV Local (n -= 11) (43) (18) (53) Wheat, Mg % 0.116 0.106 Maize 0.243 0.199 Soil, mg Mg/1 393 809 Soil 360 436

The above differences between the two variety groups are statistically significant at the 5 percent level in Mg contents of both plants, at the 0.1 percent level in wheat soils but non-significant in maize soils. See also Section 9.1.4. 348 9.1.3 Micronutrients

Boron. The average B contents of Ethiopian soils and pot-grown wheat are lower than the respective international averages (Figs 458 and 459) and consequently, Ethiopia stands somewhat low in the "international B fields" (Figs 22 and 25, Section 2.3.3). The ranges of variation in B contents compared to many other countries are relatively narrow, and in spite of the generally low B level no extremely low values were recorded. According to these data no severe deficiency of B exists in Ethiopia but at several locations a response to B should be expected.

100 80 n=15 3 60 s=476± 2 i 8 50 s=054 t 0 07 7 40 02 + 13 x = 63 E30 338 4 20 o

10 '58 "E. 6 5 o 4 , . co 3 2 Fig. 458. Regression of B content of pot-grown wheat (y) on hot water soluble soil B (x), Ethiopia. For details 1 .2.3 .4 .5.6.8 10 2 34 568 10 of summarizedinternationalback- B in soil, mg/I (Hot w. sol.) ground data, see Chapter 4.

100 80 n=1 60 ±s= -1092 6, 9 4 50 T±s=04 01 40 E =377+31x y= 5 .30 r =0 418 0 2 -8 20

W5C6 o4 C.) co 3 2

III Fig. 459. Regression of B content of .081 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 pot-grown wheat (y) on CEC-corrected CEC -corrected B in soil, mg /l (Hot w. sou (hot water soluble) soil B (x), Ethiopia. 349 Copper. In the "international Cu fields" (Figs 27 and 29, Section 2.3.4) Ethiopia is among the low-Cu countries. Contrary to B, the Cu contents vary greatly and very low plant and soil Cu contents (including the minima of the whole international material) were measured (Figs 460 and 461). Of the whole Ethiopian sample material, 21 percent of the sample pairs fall in the lowest Cu Zone (I) and 6 percent in Zone II (Fig. 461). Most of the low (Zone I) Cu values occur in Sidamo Province, where two-thirds of the sample pairs show very low contents, and most of the remainder occur in Zone II. Although a few low Cu values were also recorded in samples from the neighbouring Provinces of Shoa, Gamu-Gofa and Arussi, the Cu levels of soils in these and other provinces seem to be

22 20 P1111.1.11111111111111

18 hill11111111111111111111111 E R-16111111171111111:11m01111111111111 T-3w14111111111M11111111111111 -c 111111111111111111111111Will 111111111111111111121111

81111111111111111111111M1111111

Fig. 460. Regression of Cu content of 2 JEMIR11211211111111111111% pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), .2 .4.6.810 2 46 810 20 40 60 80 Ethiopia. Cu in soil, mg/I AAc EDTA)

' °AIM"=7,11111 )11111111111Y s 7,....,, 11111611 Y's ' 'gilllogAliiia )r M 111111 IN AEI '? INIi LI II-'1101r1 11111 I 11111A=1111111

Fig. 461. Regression of Cu content of 11111141111111111 pot-grownwheat(y)onorganic 211 .6 .810 carbon-corrected (AAAc-EDTA extr.) 2 26 4060 100 ;:rd'Ilialiirilbi:1:104 soil Cu (x), Ethiopia. Org.C-corrected soil Cu, mg/l (AAAc-EDTA) 350

3 00 Fig. 462. Regression of Fe contents of 2 75 2 50amMILIMEIIIIIIIMMOB pot-grown wheat (y) on acid ammonium 225 acetate-EDTA extractable soil Fe (x), 200 Ethiopia. 1 75 113111111111123111i111111111 E 1 50 o_ ct 1 25iiiuiiiuiiiiiuiuiuii o a) 1011111111111111111111 100MIIIII111111=111611111111M111 11-, 90 80MIIME111111=1.111111111MI 11111111111111111111111111M111 o07 u60 50

40111111111111111111111M1

30 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc-EDTA)

within the normal range. No values showing a possible excess of Cu were recorded. These analytical data point to the presence of Cu deficiency in Ethiopia, especially in the Sidamo area, and a response to Cu fertilization in several locations is very probable. Iron. On average, the Fe contents of wheat grown in pots on Ethiopian soils correspond closely to those of the international material as a whole, but the extractable Fe contents of soils are somewhat higher (Figs 462 and 31). Normal (Zone III) values dominate and in spite of a few samples within Zones II, IV and V, these are insufficient as an indication of Fe problems in Ethiopia. Manganese. Ethiopia stands clearly on the high side in the "international Mn fields" (Figs 33 and 37) and in spite of a relatively wide variation of Mn contents (Figs 463 and 464) no low (Zone I or II) values were recorded. Instead, a substantial percentage of Ethiopian samples are found in the high Mn Zones, IV and V. Most of the highest (Zone V) Mn values were measured from samples which came from Sidamo Province but occasionally some were obtained from the samples from other provinces. All sites where excess Mn was recorded had acid soils, the pH(CaC12) was usually below 5.0, and therefore, by raising the soil pH through liming the availability of Mn could be reduced and the possibility of its toxic effects eliminated. Molybdenum. In Ethiopia the problems associated with Mo differ from those of most other countries because typical Ethiopian soils are exceptionally heavy textured and usually acid. In such soils the availability of Mo to plants is low although soil analyses may show relatively high Mo values (Fig. 16). This explains why many Ethiopian low plant Mo contents are combined with moderately high A0-0A extractable soil Mo values (Fig. 465). This contradiction is partly overcome when the soil Mo values are corrected for pH (Fig. 466). Since the correction for texture (see Section 2.3.2.2) is not applied to the Ethiopian Mo data, the effect of texture must be taken into account when interpreting the 351

Fig.463. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA extractable soil Mn (x), Ethiopia. 1000 DCWM:114"=11plstqUol10. 1==imiNME 800MI IM:173147Mlinii1RW.VirIMNMMiNIMEN=munms //1 MgII 600iimmmImmeolinsmmimaniam rasimaimmoufsiammimmam E 400 ca" 30011=11111111111=111=1111111

2 00

1 0011111111111114111116ts. 80::=Enwmagni=rewrzancINLMMIMMCi Minedig11 60 o nwimmowesilimminimminniimmm. 40 30 20imminiammouno

10ea11111111111111111111111N1=MIIMM=1MME 8 1=MMINEENiaMIOSTIMMEINIM1.111=11=.11 2 3 1, 6 810 20 30/.0 6080100200 300 Mn in soil, mg/t (DTPA)

Fig.464. Regression of Mn content 2000 m...... 51alm=19 ...i...... of pot-grown wheat (y) on DTPA Mn E extractable soil Mn corrected for pH 1000 WINOPIRPOII,=7UM.WIURIIIMM1411111111111111111 (x), Ethiopia. 800=MENNEINWonINNIMMMIIIIMUMMPIMMINlia&WO1EENIM.2VIELIWAtenININ011MIMIWICIMEn; 6001111=-11114MBEMEMEREIMMKTErnMSIONMEMEINENIMIIINSI Minall1111NiNnilian 11INIIMMINKII 400 300IN rmw111111h111111111 200

1001111...- mipsms-ammmo.-ism;=mon111111111111111111111111111 80h.... NERMIMr..4MIMIMIEWINMENIMUME.al...... ,..s...... ,-.. 60NiMMIZIMI, MERNIMIIMIMMIMMIMM=EISIIII 1111111111M., NIMPRIMININIIIIIMIMIIIIMMILIINIM 40II MM `11PROMEMIIIIIII11111111MNIMMIIMIL 30iiMpr, IgniiMMEMIOMMINÌNER 20

10161 11111111111111 811 MEIMMIIMMILIMMINMENO. ....N6...... M=MIIMM=NNW. MN musemmasvaumn Immoloso 1 2 3 46 810 20 3040 60 100 200 300600 pH- corrected Mn in soil, mg/1 (DTPA)

Mo results sample by sample. See also Figs 15, 20, 21 and Table 8 in Section 2.3.2. Low Mo values are rather more common in Sidamo than in other provinces but are also found in Shoa, Wollega and Arussi, usually in sites with soils of low pH and of heavy texture. Many of the highest Mo values were analysed in Sidamo samples from sites where the soils were less acid or alkaline. The highest plant Mo content (9.79 ppm) and the highest A0-0A extractable soil Mo content (uncorrected 3.6 and pH-corrected 4.8 mg/1) in the whole international material were both obtained from a Sidamo sample pair (47079). In this case Mo toxicity should be suspected. On several sites low Mo values are combined with high Mn values. The application of lime would increase the availability 352

Fig.465. Regression of Mo content M o INIMMINNINESMIMOMMENEU 6MMIURNIE1111111111MMEMMIIIIIIIIIIHI1 of pot-grown wheat (y) on ammonium =k1-21.1FaIMEINIZEEEIEEMIIIIIMI oxalate-oxalic acid extractable soil Mo 3mmigammat ammo= (x), Ethiopia. 2 oliNiall-221311111.1 11101110111r IIIIIMMESIERFM MIIIIMMIIIIP mminumarimausum==101111MEMMinMlegainliMi 't.2 o 111111111111111111 o AINICWROMMENFM1111VRIMEM1AM010INMIIMMEIMMI 'MUM. IMOM11=1IMMENIMIMIIIIMIENNO L.1=M. .06WMMNIONIIIIIMINNIIIIIIIIIMM111MMININNEliMINION I' MMEMENIMBMINNEFEM .03IMMIMMINIFILIMMIIIIIIME .02

.01 .01 .02 .03 .06"i. .2.3 .6 1 23 Mo in soil, mg/ f AO- OA I

Fig. 466. Regression of Mo content of iiirdonmsammeenintrrcri==E:1111== 1=1111 MIIII=I Mo 1MMMEMAnIN!=M EMMIAN 6 .125IIMMINHINILIIIMMIIN 0111=I4111 pot-grown wheat (y) on pH-corrected =°. 73= EZEINIIIIMICIEEMELE A0-0AextractablesoilMo(x), 3 .'71111111111211GILIE11111111111111111 Ethiopia. 2 0=.2.05i=MMEIMIIIIIIii'IIi!iiI 741 ME MI 11119,21i11111IMM EMEENEWANiiiiiMIENN 2 8C 1111111111111111hal .1 Lclamon...1==2:1111=ME: 06 WMENTANMElliiii==ni

.03 .41111Newrograin .02 01,11111,4

.01 11,411111.111 .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH-corrected MOin soil,mg/I (AO- OA

of the former and decrease that of the latter, and would be the recommended treatment.. Zinc. The Ethiopian national mean values for plant and soil Zn are high compared to those of most other countries (Figs 41 and 43). The variation, however, is wide and both high and relatively low Zn contents are found. Independent of the soil extraction method the plant-soil correlations are good (Figs 467 and 468). According to both methods every third sample pair falls in the high Zn Zones (IV and V), a few of them showing very 353

Fig.467.Regression of Zn content 200 of pot-grown wheat (y) on DTPA 177 extractable soil Zn (x), Ethiopia. 5 321 og 100 1111111 90nimmimmommisrmolmolimummou 8011111MINIIIIII=MMINIIIIMINIMIN111111 70IUlIIIIUkLUIlIIIUullIIIP E 60 a 501111111111111111111111111111E1111 o _E 40

301111M11111111111111111K11111111111 111111111111111111111111111111

101111111011111111111111 9 MENU 8 7131111MINLIMIIIIIIMMENUIIIMILIM1111111 6 iiiimminumwmunawnsounuiuiiuiuuiìîuiiixuuu .2.3 .4.6 .81 23 46.81020 3040 60 100 Z11 in soil, mg/1 (DTPA)

Fig. 468. Regression of Zn content of 200 pot-grown wheat (y) on AAAc-EDTA extractable soil Zn corrected for pH (x), Ethiopia. 100 1. 1 NMI! 90 80MEN milarmenni 70 60 E 50MUMINNIIIII o40 1011111101 30limI o 20INiploN11111 o

1011110101P11 9 81...miL morm.asam 7munowninliommmum. 6momonummonommusui .2.3 .4.6 .8 1 23 46 81020 3040 60 100 pH-corrected Zn in soil, mg/l (AAAc-sorA) high Zn contents. When Zn contents are low, the DTPA method reveals a slightly better Zn content of soils than does the AAAc-EDTA extraction. Neither of them indicates severe Zn deficiency but at many locations a hidden shortage of Zn is likely. The lowest Zn contents mainly come from Harar and Wollega and the majority of high Zn values from Sidamo. 354

9.1.4 Micronutrient contents of original plants with special reference to varieties

The most obvious varietal differences in the micronutrient contents of original indicator plants were those between the Mn and Zn contents of the HYV and local maize plants. The HYV maize (varieties H 512, H 611, H 613, H 632, Composite II and KCB, n = 18) showed higher average Mn and Zn contents that the local varieties (n = 53) although they were grown on soils poorer in these elements.

Mn Zn HYV Local HYV Local Maize, ppm 90 75 52 43 Soil, mg/10 98 132 2.7 5.6

0 pH-corr. DTPA extr. Mn, DTPA extr. Zn

When one wheat cultivar ('Apu') was grown on the above soils in pots, theaverage Mn and Zn contents were as follows:

Mn Zn Wheat ('Apu'), ppm 154 161 22 29 Soil, mg/10 98 132 2.7 5.6

0 pH-corr. DTPA extr. Mn, DTPA extr. Zn

The latter Mn and Zn contents are more compatible with the results of soil analyses than those obtained by analyzing the original plants of different varieties. The above differences between the two groups did not quite reach statistical significance at the 5 percent level, except in the case of soil Zn. However, since the differences in the original maize Mn and Zn contents were diametrically opposed to these in the respective soils, it would seem possible that there are some genetic differences between the two variety groups and that the HYV maize plants are the more efficient in absorbing the above micronutrients from the soil. This above example hopes to draw attention to one of the difficulties investigators face when interpreting results of plant analyses obtained from materials consisting of several varieties.

9.1.5 Summary

Typical Ethiopian soils included in this study are fine to medium textured, and show moderate to strong acidity. They are relatively high in organic matter and have a high cation exchange capacity. Compared to the international general means, the mean N and especially K contents of Ethiopian soils are high, P contents are somewhat low, and Ca and Mg contents are at the international average. The micronutrient content of Ethiopian soils and plants varies considerably depending on the element. Of the six micronutrients studied, the one most likely to be deficient is Cu, but responses to B and Zn can also be expected at several locations. The contents of Fe are usually at the normal international level but many high Mn values were recorded. The Mo and Zn contents typically vary widely and range between relatively low and high. 355 9.2 Ghana

9.2.1 General

The Ghana sample material consists of93maize and related soil samples and was collected from geographically limited but agriculturally quite important areas, namely the Central and the Ashanti regions. All the sampled crops were grown under rainfed conditions. Of the sampled soils 46 were classified as Nitosols,39Ferralsols, 5 Gleysols,2Lithosols and 1 Vertisol. In Ghana coarse textured soils predominate (Fig.469)and only a few have a medium texture. The national average texture index (TI = 30) is one of the lowest among the 30 countries investigated (Appendixes3and 4). Strong to moderate acidity, low cation exchange capacity, low electrical conductivity, low CaCO3 equivalent and low sodium content are typical for Ghana soils.

45 Ghana Internat. Ghana rmemot, Fig.469.Frequency distributions of n= 94 3764 40 8. 94 3783 =30 44 =5.85 6.64 texture, pH, organic carbon content, ±5= 9 16 ts= .66 1.12 min.= 13 9 35 min.= 4.35 3.62 and cation exchange capacity in soils of max.= 55 92 max.= 7.05 8.56 0 Ghana (columns)Curves show the

25 international frequency of the same

20 characters.

10 20 30 cu 50 60 70 80 90 TEXTURE INDEX pH (Ca Cl2)

45 Ghana Internat. GhanaInternet n= 94 3779 n=94 3777 0 0= 1.3 /.3 5=14.7 27.3 ±s=.5 1.2 rs= 6.0 11..6 35 min =.6 .1 min.= 4.9 .2 o max.= 2.7 5 39.1 max =41.2 99.9 a.,30

25 C> w 20 w 15

rer3.4 is Zs e: //YAM 2 .8 1.6 3.2 6.4 12.13 25.6 ORGANICC,% CEC, me/100g

9.2.2 Macronutrients

The total nitrogen contents of soils are at the average international level (Figs 6 and470). In spite of relatively high rates of nitrogen fertilizers(48 ± 40kg N/ha) applied to the sampled crops, the average N content of maize remains below the international mean. The NaHCO3 extractable phosphorus and CH3COONH4 exchangeable potassium contents of Ghana soils are very low compared to most other countries (Figs7, 8, 471and 472).The applied dressings of phosphate and potassium fertilizers (11 ± 11 kg P/ha and 14 ± 12kg K/ha), although not high, were about double the average P and K applications 356

40 3 Ghana r Internat Ghana Internat. n= 93 n=1958 n= 93 1967 35 Ghana Internet 5=2.80 x. 3.14 30 5= 3.1 3.1 ±s=.41 35 n= 93 1967 no= 0.5 1.0 ts= .87 R..32 .33 30 min.= 1.58 nun.=.88 a, min.= 1.9 0.6 ±5= .07 .10 5 max.= 4.4 6.7 max.= 3.84 max.= 6.51 0 30 min.= .17 .05 max.= .51 aQt° 25 1.04 0,75 20 >- 0_2 20 3- 4.3 20 LLI LLI15 o15 o LU ce 10 cc 10

5 5

4 A .4.47.,

2 3 6 9 6 N content of maize, % P content of maize,% K content of maize,% 4 70 Ghana Internat. Ghana Internat. n. 93 1967 Ghana I n elrgn6a7t 1958 =8.2 22.5 n= 93 *5 .10.1 33.0 n= 93 7=.132 .135 min.. 18 0.1 R 1729 330 -Zs = .056 .088 30 max.= 95.8 656.0 no. 356 min. =.053 .008 min.= 34 18 max.= .333 1.657 o a, max.= 321m 5598 25 ,96 0./ o- f' 3- 20

o15 ' 7 LCf. 10

i1i1ø4 09m 2.5 5 20 .005 01 .02 .04 .08 .16 .32 64 1.28 10 40 80 160 320 640 25 50 100 200400 800 600 3200 6400 N in maize soils, % P in maize soils, mg/I K in maize soils, mg/I Fig. 470. Nitrogen, Ghana. Fig. 471. Phosphorus, Ghana. Fig. 472. Potassium, Ghana.

Ghana Internat. Figs 470-474. Frequency distribu- Ghana Internat. 35 n= 93 n. 1967 9..33 R..47 n= 93 1967 tions of nitrogen, phosphorus, po- *5..06 35..20 R=203 .251 830 min.../6 min...09 6s=.043 .119 tassium, calcium and magnesium max...50 max . 1.88 U 30 min.=.119 .036 max.=.316 1.125 CT, in original maize samples and re- 2t_25 0_ 25 >- >- spective soils (columns) of Ghana. 0 20 0 20 11.1 Curves show the international 15 frequency of the same characters. Cr 10 L.L.

,4eXee.- .4 .2 Co content of maize,% Mg content of maize, %

Ghana Internat 40 n. 93 1967 GhanaInternat. 5 R. 1145 3450 n. 520 2815 3 n= 93 1967 71 10 R. 215 446 max.. 3150 17995 ±s= 183 462 30 min.= 72 1 max.= 1455 6490

0 100 200 400 800 1600 3200 6400 12900 12.5 25 50 100 200 400 800 1600 32006400 Ca in ma ze soils, mg/I Mg in maize soils, mg/I Fig. 473. Calcium, Ghana. Fig. 474. Magnesium, Ghana. 357 to maize crops in this study. These may have contributed to raise the P and K contents of the Ghana maize to the respective average international levels. Also the uptake by plants of K from acid soils and that of P from soils of low CEC is high in relation to the contents of these nutrients in soils. See Figs14and 11 and related texts. The exchangeable calcium and magnesium contents of Ghana soils and the Ca and Mg contents of maize are lower than in most other countries (Figs 9, 10,473and474).

9.2.3 Micronutrients

Boron. The B contents of Ghana soils, and of wheat grown in pots on these soils, vary within quite narrow limits (Figs475and476).The national B averages correspond

loo 80B 60 =6020 0 50 =057± 07 40 64+0 6 x =6 E 30 162 n s 0 4 20

.c 10 "a8 6 5 o 4 3

Fig. 475. Regression of B content of 2 pot-grown wheat (y) on hot water soluble soil B (x), Ghana. For details 1 of summarizedinternationalback- .1 .2 .3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 ground data, see Chapter 4. B in soil, mg/I (Hot w. sol.)

100 80 n. _3 60 ±09 50 9±s07 02 7 + 40 E Q-30 log y.07 +0 x Q. r0208 8 a

la Fig. 476. Regression of B content of pot-grown wheat (y) on CEC-corrected .08.1 .2.3 .4 .5.6 .8 1.0 23 4 5 6 8 10 (hot water soluble) soil B (x), Ghana. CEC -corrected Bin soil, mg/I (Hot w. So() 358

22 G a a I nte not. Fig. 477.Regression of Cu content of pot-grown wheat (y) on acid ammonium 20 n 353 484±14 700 2.5 acetate-EDTA extractable soil Cu (x), 194 12 6. 7.0 18 E Ghana. 8:16 11a00smimmuni

.c161.14 .6 121111111111111111111111 I10 88111111111111111=11111M1111111 067 11111111111111111=11=11111 4

.2 .4.6 .810 2 4 6 810 20 40 60 80 Cu in soil, mg/l (AAAc- EDTA)

Fig. 478.Regression of Cu content of 22Cu i pot-grownwheat(y)onorganic , Gha s 1 Intern t 2-,n ,. carbon-corrected (AAAc-EDTA extr.) y±s.4 j3..7:7i3r1g.±, u a1 soil Cu (x), Ghana. 13P-s.11 y=39291Io x Irogoy;isx0 5.1,1gi A411 15r -°74 1

142 1110 1 11111111111 13

6 111111111111111111111

4 nontiiimmanill

2 11111141FIRMIIIIIII1 MIMI

gning. Ill 01 7 L. 6 810 2 4 6 810 20 40 60 100141 Org.C-corrected soi CU,mg/l (AAAc-EDTA) closely to respective international mean values and in the "international B fields" Ghana stands at the centre (Figs 22 and 25). Since practically all B values are "normal" and fall in B Zone III, these limited analytical data do not indicate any severe B disorders in Ghana. Copper. Most plant and soil Cu contents measured from Ghana sample material are lower than the respective mean values in the whole international data of this study, and Ghana occupies one of the lowest positions in the "international Cu fields" (Figs 477, 478, 27 and 29). Only about half of the plant-soil Cu values are "normal" so falling in Cu Zone III, and the other half is distributed evenly in the two low Cu Zones. In general, Cu deficiency, acute or hidden, seems to be more likely in Ghana than in most other countries. 359 300 Fig.479. Regression of Fe contents 2m of pot-grownwheat(y)onacid 2501122111111:1111111 ammonium acetate-EDTA extractable 225 soil Fe (x), Ghana. 200Enitraliiii 1 175augh;,,3 E 1 50 1111111111U11111 o_ o_ - 1 251111111111M111111111 1 o 100111111111ME11111 I 1590Er%111. milipI t 80 t 70iiiimilliiMij1111=1 o o60Iliniin11111A16111111 50 111118111111111111111 1

40MiiiiiiiiiiMini

30 simsnummeommui4wfxEEOi 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l (AAAc- EDTA)

Iron. The Fe data presented in Figs 31 and 479 do not indicate any disorders due to this micronutrient in Ghana. Manganese. The variation range of the plant Mn contents is relatively wide (from 45 to 601 ppm) and clearly due to the varying pH in Ghana soils, but the values for DTPA extractable Mn vary only from 14 to 138 mg/I (Figs 469 and 480). Correction of DTPA Mn values for pH more than doubles the variation range and improves the plant Mnsoil Mn correlation (Fig. 481). Ghana stands above many other countries in the "international Mn fields" (Figs 33 and 37). Nine percent of the plant-soil sample pairs fall in the highest Mn Zone (V) and 12 percent in Zone IV, all representing acid soils, pH(CaC12) < 6.0, and most of these come from Ashanti. No low Mn values were recorded from the sample material of Ghana. See also Mo in the next paragraph. Molybdenum. Owing to the generally low pH of Ghana soils, the availability of Mo to plants is low and the national average for plant Mo content is less than one-third of the international mean and is one of the lowest among the countries investigated (Figs 482, 483, 15 and 20). The correlation between plant Mo and A0-0A extractable soil Mo contents is non-significant but becomes highly significant when the effect of pH is taken into account by pH-correction. More than one-third of the Ghana plantsoil Mo points fall in the two lowest Mo Zones (I and II). The low Mo values were most frequently found in samples coming from the acid soils of the Ashanti region and were often combined with high Mn values. Since low pH seems to be the main reason for the low availability of Mo, as well as for the high availability of Mn, the first logical approach to correct both possible disorders would be 360

2000 Fig.480. Regression of Mn content of pot-grown wheat (y) on DTPA 1 000 .(3:MWYlo-ZWVA,1VIWZMC,DMINIMMIIMIOS=101M1/11311131111101111 extractable soil Mn (x), Ghana. 800I. nwifesimnsumr:mnip-mirtrmiratasffillial=mmni.....=musummisammum 600111MIIMMENIIIMMRIMM1121111=1MIIMIMIMMEMOIMI.- 11111111=1111111 E 400ii.==IMMIMMIMMERIME=MMIE 8: 300 -'06, 200IM11111 IM111111111111111 .c 100rosmorma._-___Ers.cm111111111111.11111111 Y 80inionrumml=zuwwarliwuramsea 11111111111111MIIIWEBOIMP111161111MMEIMI LIMER 'do 60iiliniailiMUMMOPMMENNOMMIIM 401110111MIERVANIIMMINIENIIIIIM 230 20

10Iill1111111111111111111117111/1111111111111ilidMMIROMMISINNIMIIMMIE111 IMO imm 8IIIIIMNINIIMME=MME111111i71111,01iTiMMEM 111111111111111111111MOM1111LIIIIIIIMMIMMEMMI =ZEN 1 2 34 6810 20 30 40 6080=200 300 Mn in soil, mg/I (DTP.4)

=MMEINI1 71.11111 .1.011111POIMMIMMIIIIMR 2000 a Fig.481. Regression of Mn content Mn 93 of pot-grown wheat (y) on DTPA s. 1000 WPOMMIWONRIO,_111ME1111111111111* SIMONS_ ININMVIMM111111.1. extractable soil Mn corrected for pH 800 (x), Ghana. 600BB mainmixtimxuarlivaziniuminumitin E IIMMOT zonnwmasonsiumuramancis 0- 400II MIIIIM=IMM111111111%111111.0111111 ts'300IMMEM cD ; 200 1111=111111111111M111111111111 o 100OIUIIIIIU!1hUiIIII1 80 6011111111. MEMMOMMIAM=MENEMENFM=Mini fj IIIMIMM=M1111111=MM111 40 IiiiiMEN=NAMMEMMMIN 30 20

10OiiiiiiiIiIIIIIIIIIIiÌÏiiiNIMMEN 8 1=1 MINIOnlalMMEINIMINIENIiMM771.1=IMMEENI =NIMMWEVI 1 2 3 46 810 20 3040 60 100200 300600 pH- co r rected Mn in soil, mg/I (DTPA) to raise the soil pH through liming Zinc. The mean Zn contents of Ghana soils and plants are somewhat on the low side in the "international Zn fields" (Figs 41 and 43). Comparison of the results obtained by the two extraction methods (Figs 484 and 485) shows, however, that DTPA extraction yields a generally somewheat higher soil Zn level than does the AAAc-EDTA extraction with pH correction. In both cases the correlations with the results of plant analyses are very good. The lowest plantsoil Zn values, almost without exception, were measured from samples which came from the Ashanti region. The present analytical data do not indicate very severe Zn deficiency but response to Zn may be obtained at several locations. See also Section 2.7. 361 Fig. 482. Regression of Mo content of : MILINC-="m122111MOMTVIi(Mari IMMINN pot-grown wheat (y) on ammonium 6 IM1W -aimsNM oxalate-oxalic acid extractable soil Mo Immaya3-97 0.32,mg. (x), Ghana. 3orrasommtimmuu 2 SEE

act 1 111111111111111MIIMVVVESVMiMsnavs a, .6 MMENOMMMEMMONEVin "-3mmonumeamili .2ilt11111101111111111L o o .1SWA:=E11111:=MES:1111'WAIMM.NMIIMMINNOIONIINIMMIMM , .06wwImummunonmmommuna 111/111111111111 =MINIM .03 111 .02 II\ 01 .01 .02 .03 .06 . .2 .3 .6 1 23 Mo in soil, mg/ (A0-0A)

Fig. 483. Regression of Mo content of M o mom ND ut=1111MRSIIIIIIIITIFIZT1MalagrinrnaZZ:::: VIIMMINIMII011 111=MI=MI-- pot-grown wheat (y) on pH-corrected 'V MMIUMMI ININ 6 .93 L. 0=3MMIlnliMMIliMnill A0-0AextractablesoilMo(x), °.13 ' .0211111111EIMIBBEEZIMINNUISII s 0.42± 02 illINFIZINONIIIIIMIMMI Ghana. 3 y=0,013 2 :.;1 111111110111111111M11111 111111111111_,1111 =Morns --...... --- =NIEMEN EHMENIIIIMMENMINEMINE411111MMINI===:iMmuMgmmiaiimEmisi MOM= MMENIIIIIIIIMMIIIMMIIIIIIIIMINNIMMIN MINENIIIMMINIIIPAIMIIIMINIMME ,MinniliMiNEEM 111111imiter MINIIIMMIN11E/AINIMINIIN=MOIRE 171MEMILIM=6=MISmipmel 1=MIWIND UPIIVIMINVNIVIM MO=NEVIIMM =IMMEMINIAIMIMMEUMMIMMINIMMIMIUMVIMMORVMMIVIMI MiMMENIUMEmmimli

.03 .02ounimminommummoul

.01MPIWOmMint111111111111111111111111111111111111111011 =MINIM MNIMB .006 .01 .02 .03 .06 .1 .2.3 .6 1 2 3 pH- corrected MO in soil, mg/I (A0-0A)

9.2.4 Summary

Coarse texture, low pH, medium organic matter content and low CEC are typical of most Ghana soils. With the exception of total N contents which are medium, the soils are generally poor in macronutrients. The most likely disorders due to micronutrients are those of Cu and Zn deficiency. Shortage of available Mo and excessively available Mn may cause problems with acid soils. Occasional responses to B are possible but disorders due to Fe are unlikely. 362

200 Fig. 484. Regression of Zn content of Zn pot-grown wheat (y) on DTPA extract- 5 able soil Zn (x), Ghana. 321Iog 100 90 80 70 E 60 cl 50

.c w 40

"6 30

o20 N

10 9 8 7 6

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Zn in soil, mg/I (DTPA)

200

100 1111111111111 90 80 70 60 E 50 III111 2: 40 o 30 MI ION o 20 o me NJ I 10 irim4 PI1 9 IIIILI a 7 Fig. 485. Regression of Zn content of 6 hillIll E pot-grown wheat (y) .on AAAc-EDTA .2 .3 .4.6 .8 1 23 46 810 20 3040 60 100 extractable soil Zn corrected for pH pH-corrected Zn in soi .mg/1 (AAAc-EDTA) (x), Ghana. 363 9.3 Malawi

9.3.1 General The sample material collected from Malawi consists of 100 maize-soil sample pairs. The sampling sites, shown in Fig. 486, are well distributed over the country. The majority of soils are coarse textured and acid, have a medium organic matter content and a low cation exchange capacity (Fig. 487). Low CaCO3 equivalent, low sodium content and medium electrical conductivity are typical of most Malawi soils (Appendixes 3 and 4).

33° 39° 35° 36°

9 11

Vs.

k 48530

48528 1148531-53 10° 10° 'NA 48533 r 48534 I%\ /.'...... , 7 48446 t 8193.

r 1.121128[1 1) A 8527 %. 498526 1 r-'`'`. 90523 98528 i iAA 48524 i 4 7 12 i / 2° i a (1 46522 i

\ ...... 913521 1 ...... 9..../ 1 1 ... i 13. NKOTA 4856 074;1. 3° I ; aoseiAt.10: i

i96592 -543 A 5A5V.:5" i * ./. 99999 ;:gig;9,554aa 485713 e 48538-591 48572 -,,----1 .1/4 4:556-560 485 6A 4'3" 78 985 557 91,,, , 19° LILONGWE 19 1344/V-9563 483164- Se 4 56T-539 48570,\ 13573-.175 \ \ 4.6.41:2949.959,0 48551 48567-598 99999 -....' 411595-566 ANT-600 1i..''.... \ 46593-594 48 2 -603 1 A 48591-5 2 98589,90 148604-6 \ A4858T-re i 42glitg, r 15° 48610-611 /1.1 ZOMBA0 4D601 1 i 98618-619i A , 5 44p822 / aLANTYRE "9-68(SIG-t17' ( 43612-613 I 16° \ 16° i \ 1 MAL AWI \ 1 . s.\ i\ .; IT° i 17.

L. 53. 36° Fig. 486. Sampling sites in Malawi. 364

35 Malawiinterne. Malawi Internat. Fig.487. Frequency distributions of n .100 3764 40 nx 109 3753 0. 35 44 6.5.36 texture, pH, organic carbon content, ts 10 16 tali.51 1.12 min, 15 9 3.62 max.... 51 92 and cation exchange capacity in soils ma4,6.80 5.56 of Malawi (columns). Curves show

5 theinternationalfrequencyofthe same characters.

10 29 40 50 GO 70 BO TEXTURE INDEX pH (Ca C121

45 Malawi Internat. 3' Malawi Internet 40 n r 100 3779 ne 103 3777 7=1.3 1.3 0417.7 27.3 .6 8448.3 14.6 min,. 5.9 .2 max,. 3 3 max.-41.1 99.9 lg. 30

25 20 ci 620 s 15

LL 10

.2 6 1.6 3.2 64 12.0 25.5 4 8 16 64 ORGANIC C,% CEC, me/100g

9.3.2 Macronutrients

The average nitrogen contents of soils is slightly lower than the mean for the whole international soil material in this study (Figs 488 and 6). The N contents of maize remain at a relatively low level in spite of the large nitrogen dressings applied to the sampled maize crops (81 ± 63 kg N/ha). The wide variation in plant N contents is obviously due to nitrogen fertilization. All 15 maize plants with a N content of 3.6 percentor higher (upper graph in Fig. 488) had been fertilized with more than 100 kg N/ha. On the other hand, the 15 plants with the lowest N contents (< 1.7 %) had receivedno nitrogen fertilizer. With a few exceptions the phosphorus status of Malawi soils is good (Figs 7 and 489). The P contents of maize are usually at the normal international level but the national mean is higher because of a few plants with exceptionally high P contents. The latter plants (P % > 0.65) were grown on soils rich in P (> 74 mg P/1) and (withone exception) received large amounts of phosphate (> 20 kg P/ha). On average, the sampled Malawi maize crops had received 17 ± 14 kg P/ha. The average potassium contents of soils and plants are slightly below the respective international means (Figs 8 and 490). Varying quantities of potassium (11 ± 26 kg K/ha) were applied to the sampled maize crops. None of the plants with a K content of less than 2.0 percent had been fertilized with potassium. The plants with the highest K contents (> 4.0 %) had either received large amounts of potassium (>45 kg K/ha)or their K uptake was accelerated by large dressings of nitrogen fertilizers (Rinne et al., 1974 a). The exchangeable calcium and magnesium contents of Malawi soils and the Mg contents of maize are somewhat /ow in an international context but the Ca contents of maize correspond to the international average (Figs 9, 10, 491 and 492). 365

35 Malawi Internat Malawi internat. n= 100 1967 1958 35 n= 100 Malawi Internat. 30 R. 2.9 3.1 9=2.72 3.14 os. 1.0 1.0 35 n 100 1967 Os= .80 .87 R. .39 .33 min.= 1.0 0.6 ca3 30 min.= 1.18 .88 ts .15 .10 O 25 max.. 5.1 6.7 max.= 4.32 6.51 if30 min.. .21 .05 max.. .93 1.04 O. k 25 20 f1-'a-2 25 t) 20 o (-3 20 W 15 o cr 10 U_

.1 .3 .4 .5 .6 .7 8 7 3 4 5 6 .2 N conent of maize,V0 P content of maize,V0

70 Malawi Internat Malawi Internat. Malawi Internat. n.100 1967 R. 39.4 22.5 n= 100 1967 1958 60 n= 100 *s. 26.3 33.0 a. 227 330 R.= .115 .135 min.. 2.6 0.1 86= 143 356 is= .041 .088 nao.. 113.4 656.0 min.=38 18 U 5 min.= .039 .008 max.= 693 5598 1.657 max.= .257 5 a, 40 o LuZ 30 o W 20 CC u- , 10

./PP,A.14 440,- 400800 600 3200 6400 2.5 5 10 20 40 BO 160 320640 25 50 100 200 .005 .0 02 .04 .06 .16 .32 .64 1.28 N in maize soils,V0 P in maize soils, mg/I K in maize soils, mg/l Fig. 488. Nitrogen, Malawi. Fig. 489. Phosphorus, Malawi. Fig. 490. Potassium, Malawi.

40 Figs 488-492. Frequency distribu- Malawiinternat. Malawi Internat. 1967 35 n100 n. 100 1967 tions of nitrogen, phosphorus, .47 0.48 .251 BBB .14 .20 6..215 ts..078 .119 potassium, calcium and magne- Zi30 min...18 .09 max, 64 1.88 .036 max...595 1.125 sium in original maize samples at25 and respective soils (columns) of 2- 0 20 20 rjr Malawi. Curves show the interna- o oD /5 tionalfrequencyof the same U.1 cc 10 characters. u_ u_ 2'4

5 P., .2 1.6 5 .2 Ca content of maize % Mg content ot maize,°/.

MalawiInternat. 40 n 100 1967 Malawi Internat. 35 6.1534 3450 n. 100 1967 tn. 8139 2815 R. 222 446 min.. 200 10 no. 156 462 max .3836 17995 non. 17 1 max.. 769 6490

25 SO 100 200 400 800 1600 3200 6400 50 100 200 400 800 1600 3200 6400 12800 12.5 Co in maize soils, mg/l Mg in maize soils, mg/I Fig. 491. Calcium, Malawi. Fig. 492. Magnesium, Malawi. 366

9.3.3 Micronutrients

Boron. Almost all B values determined from Malawi samples are lower than the respective international mean B values and the country stands very low in the "international B fields" (Figs 493, 494, 22 and 25). The relative frequency of low (Zones I and II) B levels is higher than in most other countries and consequently, deficiency of B can be suspected at several locations especially for crops with high B requirements. Low B values are not typical of any distinct geographical area but seem to be rather more frequent in the Southern Region of Malawi than elsewhere in the country.

, ...... arain )).1121.`-. Eilni111'Erliilill -129± .721111231111 IN NENIEd )IIIIMINIERINNIIII : IIIMMINIIIIMMINIIIIHill

,1111 1111011 11 ,MINEENI MII IM=Maik*Wiall EMEIMINION 1 alliali miLlallinirmi=alaniira .MEMILIMMIIII IMIEMEPT2 ,IMINIMIONNI ? INIIIIMiliMillii2111111111. M111111111111011

,1111111111111MMINIIIIMM31111111111 111111M011131111111111111111EMILIIII

ill 11111111111111 Milliil Fig.493. Regression of B content of ... pot-grown wheat (y) on hot water 111111111111111 nig soluble soil B (x), Malawi. For de- .1 .2 .3 .4.5.6 .8 10 2 3 4 5 6 8 1( tails of summarized international back- B in soil, mg/I (Hot w sol.) ground data, see Chapter 4.

100 80 n lgiMr...99111 MKSIOLUME 60 ocKwmilimag rsrummumnirA 50IMIIMIZEIZZINMI.I.MIIIIIIIMMEN1111111115111 40E 11635101Ellkvi. 2 1 NM 1111111101 E liiiiy 3 28,2 mg N , gm .3O r 0363 EVIIII Q. 11110363 iin '2O 11/11111110 RP .1,111Wilim 0810huiti,===rmospr-glow-4mmmWICASI..mg

4E- 6innimm.=Milli="aririni 5 0C 41111gSiiiiiin 'EN 03 1E 11 a) 111111WW0111111 2 "Iniimitialin moo Fig. 494. Regression of B content of IIII hat lif ireL. v pot-grown wheat (y) on CEC-corrected .081 .2.3 .4 .5.6 .8 1.0 2 31. 56 8 10 (hotwatersoluble)soilB (x), CEC -corrected B in soil, mg/l (Hot w. sot.) Malawi. 367 Copper. The level of Cu in Malawi soils and plants seems to be somewhat better than that of B (Figs 495 and 496). Nevertheless, low Cu values predominate and in the "international Cu fields" Malawi's positions are on the low side (Figs 27 and 29). The relative frequency of low (Zone I and II) Cu values slightly exceeds the international frequency and at some locations a response to Cu is likely. Most of the low Cu values were measured from samples which came from sites of coarse textured soils and were relatively more frequent in the Southern and Northern Regions than in the Central Region.

22 20 18111111411.111141111111111111111 E Q. IIIMPICIIIIMAK:13111111111111E o cu14111111111M111111111111111111111111 .6 12111111111111111111111111111111111101

77,101111111111111M11111111111111111111111 81111111111111111111111111E111111111 o 61111111111111111111111111111111111 411111111111M11111111111\1111111 Fig. 495. Regression of Cu content of 2IlliiiiiiiiiiiM111111111111111V1411 pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), 01 .2 .4.6 .810 2 46 810 20 40 6080 Malawi. Cu in soil, mg/1 (A AAc- EDTA)

22Cu 20 114cist7c1 1 i inn.t,3r5n8t ±s=56.1; 9±s-o0+ 57 18

16 tr?¡°Y5=51till 3II iroitia !i mill

12 1111 IHNINII nu 10 iim a . 8 EN I: NEI MIMI 6 11111111111MIIM11111111 4 irniiai I 11111\11111111 Fig. 496. Regression of Cu content of 2 111111,1MLII ill Iii 1911 pot-grown wheat (y) on organic carbon- o corrected(AAAc-EDTAextr.)soil .4.6 .8 10 2 46 810 20 40 60 100 Cu (x), Malawi. Org.C-corrected soi CU,rng/1 (AAAc-EDTAI 368

3 00 Fig. 497. Regression of Fe contents of 2 75 2 501/3111111151.1111 pot-grown wheat (y) on acid ammonium 225 acetate-EDTA extractable soil Fe (x), Malawi. 200 o 175IMEEM1111117151'il E 1 50

1 25 o 100111111111111111 75 90 80 70 o 60

40sionionmeno 30INNIS 1111 10 20 40 60 100200400 600 1000 2000 Fe in soil, mg/I (AAAcEDTA)

Iron. The Fe contents of Malawi plants and soils are usually within the "normal" Fe range (Zone III) or relatively close to it (Figs 497 and 31). Manganese. The Mn contents of Malawi soils and plants are on average two to three times as high as those in the other countries, and up to ten times as high as those in countries of the lowest Mn contents (Figs 33 and 37). More than half the sample pairs are within the two highest Mn Zones (Figs 498 and 499). The highest Mn valueswere typical for sites on acid soils and were recorded most frequently for samples whichcame from the Northern Region, where 10 out of 16 sample pairs fall in the Mn Zone V. Molybdenum. Unlike Mn, the average Mo content of wheat grown on Malawi soils is low being less than one-third of the international mean plant Mo content (Figs 500, 501, 15 and 20). This is due to the low availability of Mo to plants in acid soils (Fig. 16, b). Contrariwise, the extractability of Mo to A0-0A from acid soils is high (Fig. 16, a) and, therefore, the soil Mo values given in Fig. 500 and 15 are almost of the average international level. This contradiction is largely eliminated through pH correction. In consequence, the internal correlation between Malawi plant Mo and soil Mo values improves from being non-significant (r = 0.062) to highly significant (r = 0.487***) and Malawi's position in the "international Mo field" (Fig. 15) moves from its earlier remoter position to a point closer to the international regression line (Fig. 20). Almost every third Malawi plantsoil Mo value is found in the lowest Mo Zones (Fig. 501). These are often combined with high Mn values. As in the case of the high Mn values the low Mo values are more typical for samples from the Northern Region than from other Regions. Zinc. Irrespective of the extraction method used to extract Zn from Malawi soils, the plant Znsoil Zn correlation is high and the data obtained on Zn levels by the two methods are much alike (Figs 502, 503, 41 and 43). "Normal" Zn values are typical and even the few sample pairs falling outside Zn Zone III are not in any extreme positions. However, many Zn values are low enough to indicate the possibility of Zn deficiency at several of the sampled sites. 369

Fig.498. Regression of Mn content 2000 of pot-grown wheat (y) on DTPA extractable soil Mn (x), Malawi. 1000 I.CMTRI:47VECINzFiiiiiíiìiiiiiiììì ,1,11IIMMOMIIMIM12111111111 800ME WIIMMInIMEMEMIMI=111!1=MMIR MN IVIIIMIGIMMEMaMMINIMOICIIMINEMIEFTETRIMiMil411:MMMOMM.111 1/111 600IIIMIEMENEMBUMMMUMEMEMNEINKMEI 400immmumilimmummumnimritom 8: 300 o200 /11±1111111111111111111011 .c

o10011111111111111111111 a) 80ummE _hmmimsstrilimmwennumimmkile 60 o IIMMIIIIIMMEINIIIITAMMEN111111111MM 4011111111111111111MINIMEIMIIIIIIMMININI 30 2011111111111111=1111111111M111111

lolizzgimmimmE111211:=0..klEreMICE="1111111111111E1111111111 8InsmMMIIMMISINNIINialISTSMINIM110111171, IMO 2 34 6810 20 30 40 6080100 200 300 Mn in soil, mg/l (DTPA)

Fig.499. Regression of Mn content 2000 of pot-grown wheat(y) on DTPA M n --11111111111111-11111111 extractable soil Mn corrected for pH 1000 wtmanat.a.maaasgamasimmerimmamoniesIMMEMIROMMUME1110111111111MMINIIIWIIIIIIII IIMMIMIM 800d MIIIIMIMMEIMMIIIMIIMINIMIMMIIIIIIMMINIMIIIIIPAIIIIMIIIIIINI (x), Malawi. MI MITETLIMPIEIEHMEMIIIIIM=MMIIIIIITIIINIVAIONIMIIIIIIII 6001111111CHEIMMUINIMMIEZMIIMMIIIMIIIIIPAIIIMIlltli a.a 4001111111111011111=LIIIMEIIMIIIMIPERIM "Cl". 300IIIESIMIIIIIIIIIMIIIIIIIIIMIRETIMMIO _c200netwiloommiormniummu "6 "C"100 =MEMIC/ArolIMEN hilk a) 80ri===drolmmar.11111=11=1""":...... 60iummmiammatilMfaigaraina=MISELli o liMMINKNEMEHMEMMINEMMIMEMIMMMENEM 40Immormormandimmammummumn. 2 30 II 1 201P° 141 10 nimmio 8001Mwim=NO MEN111Millminr, ms 2 3 46 810 20 30 40 60 100200 300600 p1-4- corrected Mn in soil, mg/l (DTPA)

9.3.4 Summary The soils of Malawi are mostly coarse textured with low pH, low cation exchange capacity and medium organic matter content. Relatively high P, low to medium N and K, and low Ca and Mg contents are typical of most Malawi soils. Of the six micronutrients included in this study, the analytical data for Fe and Zn show usually "normal" values, but at some locations shortage of Zn likely. The contents of B and Cu are low, those of Mo lower than in most other countries, while the Mn contents are among the highest. 370

AMINIIIENIMl= MENEM Fig. 500. Regression of Mo content of 6 pot-grown wheat (y) on ammonium

0.21t. 0 2 oxalate-oxalic acid extractable soil Mo 3 (x), Malawi. 211.9111191.112.1T E cL °- 1 MENEM O .6 3 IIIIIIUfll .2 o o . I 11.111111111111111111L o IlMOINIIMMIMMII=MIMIIMI=MIMEMEINIVOIS.1IIMMMENNIMOINMIMMIIM=M NEf11MENIMINSIMIIIIIM1=MMIMIIIMMIZEMIN11.11.11MMEMO .06MIUMINMOMIMM=1. /M1111111=1111111MMIll194 .03 .02 I II ifi .01

.01 .02 .03 .06 .1 .2 .3 .6 23 Mo in soit, mg/ (A 0-0A)

MIPOW1=0,1MMINNMa 17.WAMOnella =INDEM. In M oONIC/DiDivn=nalliMMIMrTrnEM=1MIMIII IIEMNMEN 6 n =97 1.0. MENU EN. 1111111FIEDEEKEZIMI1132131:101LE11111111 zio 3 °IZMIR& s 21111111 g°4:15111111114.1111111.1111 =MI=MMEM11111111111111111111111111 MIIMMEME1P'MIHNIMENNE MEMIMMO=1MMINfirig

. 77, .2 IL o u di o ' 2 MIIMIBM13111111k.W.111111:41111111111115immlommewwwwiswinineuves .06mmosnomminzarvasimmuismummus Emninriormsmunimmommuln .030111111 11111M111111MMINIIIII .02IIIMMENE111111111111111111 Fig. 501. Regression of Mo content of .01=11111111111111111E11111111MIMMIINENM=MMINEN pot-grown wheat (y) on pH-corrected .006 .01 .1 .02 .03 .06 .2.3 .6 1 23 A0-0AextractablesoilMo(x), pH-corrected MO in soil, mg/l (A0-0A) Malawi. 371

Fig. 502. Regression of Zn content of 200 pot-grown wheat (y) on DTPA extractable soil Zn (x), Malawi.

100 90 80 70 g_ 60 50 ci 40

-630

Nio

10 9 8 7 6

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zn in soil, mg /I (DTPA)

200 Zn

100 90 80 70 60 ga. 50 .2- 40 ci 30 o 20 o

10 9 8 Fig. 503. Regression of Zn content of 7 pot-grown wheat (y) on AAAc-EDTA 6 extractable soil Zn corrected for pH .2.3 .4.6 .8 1 23 46 810 20 3040 60 00 (x), Malawi. pH-corrected Zr in soil, ing/i (AAAc-EDTA) 372 9.4 Nigeria

9.4.1 General

The Nigerian sample material consisting of 103 maize-s- oil, 42 wheat-soil sample pairs plus 31 soil samples was collected from the sites shown in Fig. 504. About half the samples, including all wheat-soil sample pairs, originate from the Northern States and half from the Western, Mid-Western and Eastern States. Of the 91 soils classified into FAO/Unesco soil units, Ferralsols (31 soils) and Acrisols (27) were the mostcommon. The other soils were classified as Luvisols (8), Fluvisols (6), Nitosols (5), Regosols (5), Cambisols (4), Vertisols (2), Arenosol (1), Gleysol (1) and Lithosol (1). The great majority of the sampled Nigerian soils are very coarse textured buta few medium and fine textured are included (Fig. 505). The soil pH varies widely covering the range from pH 4 to 8 quite evenly. Most of the soils are low in organic matter and have a low cation exchange capacity. With a few exceptions the values for electrical conductivity, CaCO3 equivalent and sodium content are low (Appendixes 2-4).

1 7.

4 14 .-. -...... -.,

...- SOKOTO . C447579-13813 ..----.-. 1 LA.,

7"3 07,41 2. "Q"Ats97849 -94.5 1 e47871-877 77807 -"7te 47866-839 i 97899,00 14410119891 i. .

47,,,,%,!sg2,Acet7B60-861

kAOUNA 97824 009785

T. i 47802-803 31.44173'6°-.971940, 97892 10 ) 4' Tis 0047862-865 470 5 . vow-- MINNA OA 97823 .

----. 820-822 4- 97807 OLA ! f --- 1 97703-709 a 97808 i 4,078 18

1 447701-70201-702 J . a.

1777771D 07016£ 1.1-.7.3- .) 1.0094 ' . '4.-.72--,:r...--Hu, IBAD k47718 978A airs-av i. 9 977 22- 72 9 I CI. 97832 97730 A i_ ...... ,... , 97732 733 AA47714-0S5 97726-727 1.7 97753 OA97769 ,.1 A i 97752 A97 5 4ffil"''' LAGOS 9,31 7760 4.397751 ....f 4'0'147ftlAfb '2457-758 AA:M,Ii.'97;73, ,.) ' 97759--- 4770,p7, 9778,1734A-AW774--- 9 761.76y. .77. 4777777O5' 47766 vIr 7769 ) 97765 97776 I NIGERIA wrip J--' 47103 97777 ) NW. aU7 ne2 ..47,;., e 147,788 e % 778 - ..0 7 ,

o co o so $40

10 12.

Fig. 504. Sampling sites in Nigeria (dots = wheat fields, triangles = maize fields). 373

Fig.505.Frequency distributions of 5 NigeriaInternat. NigeriaInternat. n=176 3764 n. 176 3783 texture, pH, organic carbon content, R. 29 44 0x. 5.74 6.64 as = 15 16 tsx 1.09 1.12 and cation exchange capacity in soils of min.= 12 9 35 min.= 3.95 3.62 92 max, 7.96 8.55 Nigeria (columns). Curves show the max.= 83 international frequency of the same characters.

10 20 30 50 60 70 80 90 TEXTURE INDEX pH (Ca C12)

NigeriaInternat. 5 Nigeria Internat. n=176 3779 = 176 3777 0=.9 1.3 0= 14.4 27.3 30 es=.6 1.2 ss =8.0 14.6 35 min.= .1 .1 min.=2.4 .2 max.= 4.9 39.1 max.= 47.1 99.9

.2 .8 1.6 3.2 6.4 148 25.6 B 16 32 ORGANICC.% CEC, me/100g

Figs. 506-510. Frequency distributions Nigeria Internet Nigeria Internat. n= 42 1765 5 n= 103 n=1958 9= 4.28 4.27 8=2.48 g= 3.14 of nitrogen, phosphorus, potassium, its=.72 1.15 4s=.52 as=.87 rnin = 2.44 .60 30 min.= 1.31 min = 88 calcium and magnesium inoriginal max = 5.95 7.45 max.= 3.80 max.= 6.51 wheat and maize samples and respect- 25 ive soils (columns) of Nigeria. Curves 20

show the international frequency of 15 the same characters.

5 6 2 0 4 5 6 N content o wheat.% N content of maize.%

70 Nigeria Internet. Nigeria Internat. n= 103 1958 60 n=42 1765 0=053 .133 /2.- .092 .135 ±s=.041 .084 asx .049 .088 min...030 .000 min.= .041 .008 max.=.227 1.023 max...392 1.657

'6-C1_ 40 o7- Z 30 o CO UJ 20

.005 01 .02 .04 .08 .16 .32 .64 1.28 .005 MI .02 .04 .08 .64 1.28 Fig. 506. Nitrogen, Nigeria. N in wheat soils, % N in maize soils,./.

9.4.2 Macronutrients

The total nitrogen contents of Nigerian soils are very low compared to those of most other countries (Figs 506 and 6). Because of the relatively high rates of nitrogen fertilizers applied to the sampled wheat crops (85 ± 27 kg N/ha), the national mean N content of 374

40 Nigeria internat. Nigeria internat. Fig. 507. Phosphorus, Nigeria. 35 n 42 1765 35 n. 103 1967 =.34 .38 =.32 .33 is=.11 .12 ts=.10 .10 0 min.= .21 .05 min, .14 .05 max...73 1.02 max.. .59 1.04

a 25 25 2- (±) 20 20 S Ui E10

.2 .3 4 .5 6 7 P content of wheat% P content of moize,%

o 40 Nigerialotrooi. Nigeria iotoroot. n = 5 42 1765 5 n= 103 1967 R =16.7 20.2 R. 21.7 22.5 OS =13.1 24.7 *5= 41.4 33.0 min.= 1.2 1.0 min.= 1.2 0.1 30 rna...56.4 271.4 max,303.2 656.0

V r 4 .4 id h. X A" .4"..;,45r0,4 2 5 10 20 40 BO 100 320 640 2.5 5 10 20 40 80 160 320 640 P in wheat soils, mg/I P in maize soils, mg/I

3 Nigeria intwimt. Nigeria Internat. Fig. 508. Potassium, Nigeria. n= 42 1765 n=103 1967 0 6=4.5 4.0 30 2= 3.4 3.1

±s= 0.7 , 1.0 ±s= 0.8 1.0 t min.= 3.0 0.9 min.= 1.8 0.6 25 2,3 29 max.. 6.0 6.8 max.= 6.0 6.7 rr jr/4 _ 44%deAM_- 2 3 4 5 6 7 2 3 K content of wheat, % K content of maize,%

Nigeria Intermit, Nigeria intern& n= 42 1765 n= 103 1967 = 207 365 =19/ 330 30 = 134 263 ±s= 175 356 min.= 50 20 min.=35 18 max.= 633 2697 mox.=1042 5598 rÓl 0 25

25 50 100 200 400 BOO 1600 3200 6400 25 50 100 200 400 8001600 3200 5400 K in wheat soils, mg/I K in maize soils, mg/I

Nigerian wheats (4.28 %) corresponds to the international mean (4.27 %). Much less nitrogen was applied to maize (13 ± 23 kg N/ha) and the N contents of maize remain at a low level. This national mean is the second lowest after Tanzania (Appendix 3). The average NaHCO3 extractable phosphorus content of maize soils and P content of 375

4 Fig. 509. Calcium, Nigeria. Nigeria interriat. Nigeria Internat. 5 n. 42 1765 35 n.553 1967 St .46 .43 4. .44 .47 is .17 .17 . .14 .20 mine.26 .11 30 min.. .17 .09 mox .1.10 1.68 mas.. .79 1.88 k25

20 20

<3 is Si

.e .2 1:6 Ca content of wheat.% Ca content of maize %

Nigeria Internat, Nigeria Internat. 5. 42 5.1765 n=103 1967 5 8.2009 6=4671 35 5.1099 3450 ±01852 25.3075 ±5.1107 2815 min.=320 NI min. 110 min.= 65 10 ma,. 7000 553 max..21930 max.= 4880 17995

025

20 o>-

1.1.1 15

/3./ a10 u-

A d . 4 100 200 400 80016,30 3200 6400 12900 5600 50 100 200 400 6001 00 3200 6400 BOO Ca in wheat soils, mg/I Ca in maize soi s, mg/I

40 Fig. 510. Magnesium, Nigeria. Nigeria Internet NigeriaInternat 1764 n. 42 35 n. 103 1967 9..231 .172 5..232 .251 Os..072 .060 7-5..075 .119 min...126 .044 min.= .080 .036 max...483 .948 868.5.530 1.125 25

5 2 .05 0 2 Mg content of wheat, % Mg content of maize, %

Nigeriainternot. NigeriaInternat. 35 n.42 1764 n= 103 1967 345 489 s. 180 446 4s. 362 437 4s= 258 462 min.= 49 o 10 min.= 15 1 max..1795 3298 max.= 2468 6490

o Lu L u_

, , 25 50 /00 -20334009001600 3200 12.5 25 50 100 200 400 8001500 3200 6400 Mg in wheat soils, mg/I Mg in maize soils, mg/F maize correspond to the international averages but those of wheat soils and wheat are somewhat low (Figs507and 7). In all cases the variations are wide. The Nigerian crops received somewhat less phosphate fertilizers than the average for crops in this study (Appendix5). The average exchangeable potassium content of Nigerian soils is slightly on the low side 376 internationally but the mean K contents of both original indicator plants are somewhat higher than the respective international averages (Figs 8 and 508), although only small to moderate applications of potassium fertilizers were used (Appendix 5). This contradiction may partly be due to the general acidity of Nigerian soils. The plants seem to be able to absorb more K from acid soils in relation to exchangeable soil K than from alkaline soils. See Fig. 11 and related text. In spite of the relatively low exchangeable calcium and magnesium contents of Nigerian soils, the Ca and Mg contents of the indicator crops are at the average international level (Figs 9 and 10), and in the case of wheat even slightly above it (Figs 509 and 510). As pointed out in Section 2.2.4 (Figs 12 and 14 and related text) these elements are relatively readily available to plants in coarse textured soils of low CEC, and these characteristics are typical of most Nigerian soils.

9.4.3 Macronutrient contents of original plants with special reference to varieties

Since the original Nigerian plant sample material included only a few HYV maize samples, comparisons between HYV and local varieties were not possible. In the case of wheat, 24 samples represented high yielding varieties and 18 were classified as local varieties. The HYV group consisted of cultivars 'Indus' (16 samples), 'Siete Cerros' (7), and 'Inia' (1). The average N, P, K, Ca and Mg contents of each variety group, their respective soils and amounts of applied fertilizers are given in Table 28.

Table 28. Comparison of macronntrient contents of high yielding and local varieties of original Nigerian wheats, respective soils and fertilizer applications. Differences between the mean contents of the two groups followed by the same index letter are not statistically significant. Letters ab indicate significant differences at 10 and ac at 5 percent level.

Nutrient Average Macronutrient Content Average N, P and K Original wheat Wheat soils application % N %, others mgll kg/ha HYV Local HYV Local (n = 24) (n = 18) (n = 24) (n = 18) (n = 24) (n .----- 18) Nitrogen 4.542 410 0,080a 0.0812 872 822 Phosphorus 0.3712 0.311b 12.42 20.3e 182 152 Potassium 4.7424,33b003b 218a 1852 4a 3' Calcium 0.499 2349 15062 Magnesium 0.2322 343a 333a

The HYV wheats were sampled 40 days and the local wheats 35 days from planting, on the average. Therefore, the effect of physiological age is more likely to increase the differences in macronutrient contents of wheat between the two variety groups than to decrease these (see Section 1.2.2). The data in Table 28 indicate that the two variety groups may differ genetically in their ability to absorb P and N. In the cases of plant K and Ca the differences are more likely to be due to similar differences in the soil K and Ca. With regard to Mg and micronutrients no clear differences were found between the two variety groups. 377

9.4.4 Micronutrients

Boron. With relatively few exceptions the B contents of Nigerian soils and wheats grown in pots of these soils are lower than the respective international averages (Figs 511 and 512), and Nigeria is one of the countries occupying the lowest positions in the "international B fields" (Figs 22 and 25). A quarter of the Nigerian samples falls in the two lowest B zones (I and II) with some of them in quite extreme positions (Fig. 512). According to these analytical data B deficiency is likely at several locations in Nigeria, and response to B, especially by crops with a high B requirement such as root crops, legumes and some fruits and vegetables could be expected. With only a few exceptions the lowest (Zones I and II)

100 INIMIIIMILINIMMIERIPPIIIMMI1111111MIMMINIMININIIIIIMMINI MUM 80 lozittelMIIMIIITirraid11111111111111111111=11M1111110 60 n =EiR111ReWIN1111111111111111MMINMENIMII 50 40 rF1150351nc127ill112746111111111=111121 Ing.80 + 2.41916, 30 NORM Winn I 1111111111 er .Illir aiiiiii 15 a) 11111 I _c 101201111 imodhaii. iiii -6 81.....01=min...... Ims.MIMMI=0EiNMEMMIMIWEIZOINOMMillan -E. 6 MIMIII=120111111111111112PIIMMIMBISIIMMI cu 5 1111111111111111111111PISTAIMMMINIIIIIIIIIIIIII "-c- 111_11, 111111110.MMOMIll MIN1411111111 4 (-) ifilliMMIIIMMIll m 3 mum Fig. 511. Regression of B content of 211111 n11111111,411 1111111-il pot-grown wheat (y) on hot water lr, soluble soil B (x), Nigeria. For details 1 la 1 of summarizedinternationalback- .2 .3 .4 .5.6 .81.0 2 3 4 56 8 10 ground data, see Chapter 4. B insoil, mg/I (Hot w. sol.)

100

80B n= I 3 60 Sas.4 5 110 50 2ts.0 3017 0 72 40 E log.07 +0 log 550x -3o R 0. r'0409 *. B 15; 20

l0 8 48 6 611 . . s a- 4 U CO a 2 I- Fig. 512. Regression of B content of pot-grown wheat (y) on CEC-corrected .081 .2 .3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 (hot water soluble) soil B (x), Nigeria. C EC - corrected B in sil. mgil (Hot w. sot.) 378

22 Fig. 513. Regression of Cu content of 20Eliiii11111 ;111111111111111111111 pot-grown wheat (y) on acid ammonium acetate-EDTA extractable soil Cu (x), :i-flu 11 111111111111 Will III Nigeria. 111111111011:Log. gl :11:11111Illini Mil 11111 I IIIIIIIII 111111

.6 12'u11111 MI111111111 11/11N11 glo11111111111 ...i11111,11 111111 38MIN1 NI111111P1411 111111 0611111 1J111111111MININII IIIIII 4ill 111PPLIM111111111111011 2MI Iiii01111,41 11111 1111141 111111111111111111111111111111111 01 .4.6 .810 2 4 6 810 20 40 6080 Cu in soil, mg/1 (AAAc- EDTA)

Fig. 514. Regression of Cu content of 22cu pot-grownwheat (y)onorganic 20Nige11111111internt 11111 11111 carbon-corrected (AAAc-EDTA extr.) soil Cu (x), Nigeria. 18Y=1_5R111111111il;-.7MIEN 11 TY-15111110 Ell .....,L111111111116.. -...IIII rii ...,. 111111111111. I 11I loci 1 6N1IINiirn!!iIÌLII 4 Ipipmram 1 21111111.1dinikliNiii11111 MI 11111111M 111:1311Munibil oi .2 .4 .6 .8 1.0 2 4 6 810 20 40 60 100 Org.C-corrected soil CU,mgil (AAAc-EDTA)

B values originate from the Northern States. A response to B by cotton has been reported e.g. in Niger State (FAO, 1979). Copper. Only about two-thirds of the Nigerian Cu values are within the "normal" Cu range (Zone III) and one-third falls in the low Cu zones (Figs 513 and 514). Nigeria's position in the "international Cu fields" is well down (Figs 27 and 29). The great majority of the low (Zones I and II) Cu values originate from the Eastern States where such low values were recorded for more than half the sample pairs. In this part of the country Cu deficiency seems much more likely than elsewhere in Nigeria where low Cu values are relatively rare. 379 Fig. 515. Regression of Fe contents of 300 275MITIM111111111111,M111111111MONCI pot-grown wheat (y) on acid ammonium 250 acetate-EDTA extractable soil Fe (x), 225=ffillIMIIINIEldill1911111111111111 Nigeria. 200 1 7511=111111111111Waiiiiiiii1111111111 E 1 5011111111111111111111111111111 o_

'6 90wommunnamalkuninuris t 80 3.) 70 o 60 u. 50IMIIIIIBM1311111111111111

30....linciummumuntAlweit11111111111111 10 20 40 60 100200 400 6(13 1000 2000 Fe in soil, mg/1 (AAAc-EDTA)

Iron. In the "international Fe field" (Fig. 31) Nigeria stands slightly low. Although several quite low Fe values were measured from the Nigerian sample material, shortage of Fe is unlikely to be among the primary micronutrient problems of the country (Fig. 515). Manganese. The national mean Mn content of Nigerian plants is almost double the international mean but the respective DTPA extractable soil Mn values about equal it (Fig. 516). Correction of soil Mn values for pH raises the national mean from 35.3 to 55.1 mg/1, triples the standard deviation and itnproves the plant Mnsoil Mn correlation (r) from 0395*** to 0.729*** (Fig. 517). See also Figs 33 and 37. About a quarter of the sample pairs are within the high Mn Zones (IV and V) but only a few fall in the low Zones (I and II) indicating that problems due to excess Mn are more likely than those due to Mn shortage. Two-thirds of the highest (Zone V) values were measured from samples from the Eastern States. Molybdenum. The non-significant correlation between p/ant Mo and A0-0A extractable soil Mo (Fig. 518) becomes highly significant when the soil Mo values are corrected for pH (Fig. 519). The ranges of variation of both plant Mo and soil Mo are wide, but the majority of values are low and Nigeria's position in the "international Mo field" is relatively low (Fig. 20). Only a few Mo points are in the high zones and the relative frequency of low (Zones I and II) Mo values is exceeded by only a few other countries. About a fifth of the samples fall in Zone I and a third in Zones I and II (Fig. 519) indicating that problems due to shortage of Mo are likely in Nigeria, especially in the case of legumes which have a specific need for Mo. A response by groundnut to Mo has been reported e.g. in Niger State (FAO, 1979). Two-thirds of the Mo values in Zone I originate from the Eastern States at sites with a soil pH(CaC12) of 5.0 or below and these are often associated with high Mn values. 380

2000 Fig.516. Regression of Mn content of pot-grown wheat (y) on DTPA 1000 =im...... =miorni.....~11111114111111111111111111 moms extractable soil Mn (x), Nigeria. 800 InraiiiimnI11maiiminnowannseirrinmremiiimwiralkmartinnwsmimm 600IIMM=INIMMINHINIMMNMZ1110111WA11111111 400IlmImumunramilammuilinnum 300 IMI11011=111111111111613 1101111111111111PPM 100 Y 80 60;11=Malailill==g3111111LMIT: 401111111111MINIEMOUNIEMMIIIIMMIIIII=IMM 30IIIMME2111111151.111111MMEIN 20

10allINIMIIIMMINImMIONII11101M7MINEMENNE1111111111111111111;4::°:MIll1"I 8 MR Millailil=11 0=61=ITAIMMININUIRMMATOMUMEN iWMINN niNIMMIIII1111111LXMILIMIIIISI 234 6810 20 30 40 60 80100 200 300 Mn in soil, mg/l DTPA)

2000 11.0.00 Fig.517. Regression of Mn content Mn of pot-grown wheat (y) on DTPA 1000 11111111, extractable soil Mn corrected for pH MfAtIMMEtiWl!1mmmirosprom. ML.1211a wastaiii: 800 (x), Nigeria. 60011ErliarlirlNalREETEUELIMIIOMM nnulwrAmeao 40011MMEN 3001111112111MMIIIIIMMI 20001711111.11111 y 113 100 11111111111111:. 801ENEEME/PNLAm.

40 IMMENII= 30 2011,1511111111

10 11111111111.11 11immlislins-as.waiiommumworrsa 1111111111111111111111111=11111111111111U112111. 2 3 46 810 20 30 40 60 100200 300600 pH- corrected Mn in soil. mg/l (DTPA)

Zinc. Irrespective of the method used for determining extractable soil Zn contents, the patterns of Zn levels in Nigeria are similar (Figs 41, 43, 520 and 521). In the "international Zn fields" Nigeria stands clearly high. However, the Zn values vary considerably from one site to another and both very high and relatively low values were determined The highest Zn values were measured in samples from the Eastern and Northern States, but low Zn contents were more typical for sites in the Northern States than elsewhere in Nigeria. The above analytical data do not indicate very severe Zn deficiency but at many locations a response to Zn fertilization is likely (see also Section 2.7). Also Osiname (1976) reported increases in yield of between 240 and 1920 kg/ha for maize after applying Zn at rates varying from 1 to 8 kg/ha at three locations in Nigeria. 381

Fig.518. Regression of Mo content 2:=SIIIMMIllgra MoEgiSrinesummulailationimmi mow of pot-grown wheat (y) on ammonium 6 =oftamiammemmommon ma = 2=11111111111MENOMIIIIIIIT MIN oxalate-oxalic acid extractabie soil Mo 9' ....,-:gignimwanalliI ENE (x), Nigeria. 3 7-6 sakbliiimMiiall 1 2 Yr:,.: !Rig, l'Aili'Hi E 0_ 11 11 ...... 1rec....,... 1.1.monniatnomal-manisomms millOPP' 111MN .2 i o libi o .1..m.:4;=.1. 11 .06nr=ari=11121janimmmiairi IgillIMMIgilliKt issinsim... .03rIMMEllibli liii 02 \.,. II I 1 m .01 _.___ ...... 01 .02 .03 .06 .1 .2 .3 .6 1 23 Mo in soil, mg/ (A0-0A)

Fig. 519. Regression of Mo content of a pot-grown wheat (y) on pH-corrected Mo IMEM3iii wilMwo 6 BIIMMIMMNIILIIIIUni1 il n - 35 7 ManiII1Mm A0-0Aextractable soilMo(x), t s0 91 MOM O s - o 22 O. EMU. i I so 79 at 0 n 02210 1). Nigeria. 3 In is 11098,60 ala i48 og,,_ 2

E Q- 1 mikplli 0. 6' 1:=..., leirtit w ,.6...... m.m A..iii le an-isre, r ,,,,,MrKiiiinmENI 'MUM MU ó .3 NEI ffl 1 2 OM 2 ul Ail a 1 o M fir 111122EN=i ANIMZEI ...... MINELL.6 rainPVIMWATAMS: N .06 NE usoluiagragrosiN =mom .03MINIM" .02upiiiuiN

.011111111 la OM s MIIIINIiihNI'1111 i = 'MU .006 .01 .02 .03 ,06 .1 .2.3 .6 23 pH-corrected MO in soil,mg/I (A0-0A)

9.4.5 Summary

Most Nigerian soils studied are coarse in texture, low in organic matter and have a low cation exchange capacity. The pH varies widely. The macronutrient status of soils is relatively low. The micronutrient status of Nigerian soils varies considerably from one element to another. Micronutrient problems due to shortage of B, Cu, Mo and Zn and excess of Mn are likely. 382

200 Fig.520.Regression of Zn content of pot-grown wheat (y) on DTPA extractable soil Zn (x), Nigeria.

100 90 80 70 E 60 cl 50 o a' 40

"6 30

o20 o N

10 9 8 7 6 nr

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Zn in soil,mg/ 4DTPA)

200

100Zn 1 1 Nil 111 90 80 70 60 ct 50 1111I I1"7.111Ill 11/11 40 11111 1NI 111 .E/III o a) 30 vmi NMI Ill o in 20 ion o o

NJ 10 IIIII 9 8 till 7 Fig. 521. Regression of Zn content of 6 lik.11111-1 111 L 1 pot-grown wheat (y) on AAAc-EDTA .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 extractable soil Zn corrected for pH pH-corrected Zn in soi ,mg/l (AAAc-EDTA) (x), Nigeria. 383 9.5 Sierra Leone

9.5.1 General

The geographical distribution of the sampling sites for the 50 maize-soil sample pairs collected from Sierra Leone is shown in Fig. 522. All the sampled crops were grown under rainfed conditions in areas where the annual precipitation varies from 2000 to 4000 mm. The majority of the soils are coarse textured and with a few exceptions very acid (Fig. 523). The national mean pH is lower than that for any other country. In spite of a generally high organic matter content the cation exchange capacity of soils remains at a

12.

5 5.... 5 a 5 (s C..

% n: 7 ( ) r j- I ..; 1 I. 1

48., 480416 8018 48014 48017.4801'9 448020 480134 ... 21021 £48525 48023 4 480244466°015 FREE7011 ; ' 11/648012 4801Z0,9,480015 48027 a. 3.08 111,7 , 48007 A '1-'48" A 40dI i 301 A48004 48028*". I ( 48003A / 4800 £48030 J )5. 4804 t:454.804497.4.04. 48050 £48031 f 48032 ...48034 49033 A r 48035 4A:04063083e i I i 4809 tzg::44Ø43 ,,-- 48042

4 / SIERRALEONE

o zo ,Go ea . 10 te,

12.

Fig. 522. Sampling sites in Sierra Leone. 384 relatively low level due to their coarse texture (see Fig. 3, h and m). Low electrical conductivity, CaCO3 equivalent, and sodium contentare typical of Sierra Leone soils (Appendixes 3 and 4).

35 Sierra LeoneInternet Sierra Leone Internat. Fig.523.Frequency distributions of n50 3764 40 n= 50 3783 0=33 44 texture, pH, organic carbon content, 30 St =4.90 6.64 tax 8 16 ts= .58 1.12 "E min.= 22 9 35 min.= 4.10 3.62 and cation exchange capacity in soils of 0<13 25 max.= 58 92 max.=7.10 8.56 30 Sierra Leone (columns). Curves show

25 theinternationalfrequencyof the same characters. 20

U- 5

AVA VA VAVA 0 20 30 40 50 60 70 80 SO 6 TEXTURE INDEX pH (Ca C12)

Sierra Leone 8 Internat. 5 Sierra Leone Internat. n = 50 3779 40 2 n= 50 3777 02.5 1.3 0=20.9 27.3 Xs. 1.0 1.2 xs x7.6 14.6 -E 35 rein.=.8 .1 min.= 10.7 .2 max.= 5.7 39.1 max.= 44.7 99.9 1530

Lu o o 5 CC u_

.2 .4 .8 19 12 6.4 118266 16 32 ORGANICC,% CEC, me/100g

9.5.2 Macronutrients

The total nitrogen contents of the soils are relatively high due to their high organic matter contents (Fig. 524). The average use of nitrogen fertilizers was minimal (5 ± 22 kg N/ha), so explaining the low N contents of maize. In fact, nitrogen had only been applied to three sampled crops (samples No 48045, -48, and 50; 20, 100 and 112 kg N/ha, respectively). Consequently, the N contents of these maize samples were high (3.32, 3.76 and 4.14 percent, respectively) compared to the other maize samples from Sierra Leone. Most of the NaHCO3 extractable phosphorus contents of Sierra Leone soils are lower than the average in this study (Figs 525 and 7). The low soil P content is reflected in the P contents of maize which are lower than in most other countries. Only three of the 50 sampled maize crops had received phosphate fertilizer The average exchangeable potassium content of soil is lower but the K content of maize is higher than those of most other countries (Figs 526 and 8). This contradiction may partly be due to the higher K uptake from acid than from neutral or alkaline soils. See Fig. 11 and related text. The national means for exchangeable calcium and magnesium contents of Sierra Leone soils and those for the Ca and Mg contents of maize are among the very lowest of this study. (Figs 527, 528, 9, 10 and Appendix 3). 385

40 3 Sierra LeoneInternat. Sierra Leone Internat. n=49 1958 40 n= 49 1967 35 Sierra Leone Internat. = 2.83 3.14 7. 3.7 3.1 as=.53 .87 n. 49 1967 Ls= 0.7 1.0 2..27 .33 min.= 1.6 0.6 30 .88 min.= 1.72 is ..06 .70 max.= 5.2 6.7 max.= 4.14 6.51 1.) 30 min...17 .05 max...47 1.04 41/ 25 a. a 2S c:, 20 3- (2) 20 a 2, is o / ce 17 u-

2 3 4 5 6 .5 3 4 5 6 7 N content of maize,% P content of maize,% K content of maize,%

70 Sierra LeoneInternat. Sierra LeoneInternat. n= 49 1967 X. 12.3 22.5 n= 49 1958 Sierra LeoneInternat. 60 0..198 .135 Ls 21.1 33.0 min.. 3.0 0.1 n. 49 1967 as = .092 .088 max..152.0 656.0 0.108 330 min. =.103 .008 55 356 0 max. = .492 1.657 mM =47 18 max = 280 5598 á; 0- 40 o3- LTJ II 30 o / LIJ 20

5 0 25 50 100 200400 8001600 3200 6400 .005 .01 .02 .04 .08 16 .32 64 1.28 2.5 20 40 80 160 320 640 N in maize soils,% P in maize soils, mg/I K in maize soils, mg/I Fig. 524. Nitrogen, Sierra Leone. Fig. 525. Phosphorus, Sierra Leone.Fig. 526. Potassium, Sierra Leone.

Sierra LeoneInternat. Figs 524-528. Frequency distribu- rt.. 49 1967 Sierra Leone Internat. 5..29 .47 35 n. 49 1967 tions of nitrogen, phosphorus, xx = .07 .20 0,168 .251 min...17 .09 .//9 potassium, calcium and magne- 1.88 min...111 .036 max.= .267 1" 1.125 sium in original maize samples a 25 3- and respective soils (columns) of o Sierra Leone. Curves show the 0 15 O international frequency of the

10 u- same characters.

.4 1.6 5 Ca content of maize,% Mg content of maize, %

Sierra Leone Internat. 40 n 49 1967 Sierra LeoneInternet. r585 3450 ts .547 2815 5 n= 49 1967 min.. 100 10 2= 92 446 max.= 2840 17995 no.67 462 30 minx 15 1 max.. 411 6490 o

50 100 200 400 800 1600 3200 6400 12800 12.5 25 50 100 200 400 BOO 1600 3200 6400 Co in maize soils, mg/I Mg in maize soils, mg/I Fig. 527. Calcium, Sierra Leone. Fig. 528. Magnesium, Sierra Leone. 386 9.5.3 Micronutrients

Boron. Almost all Sierra Leone soils and wheat grown in pots on these soils have B contents lower than the respective international averages (Figs 529 and 530) and the country occupies relatively low positions in the "international B fields" (Figs 22 and 25). However, the B contents vary very little from one sample to another and only a few B values fall in the low B Zones (I and II). According to these analytical data, deficiency of occurs at several of the sampled Sierra Leone sites and a response to B is likely, especially where crops with high B requirements are grown. See also Section 2.7.

10, 1 -,...... ,...... 8 q,,,,,mairierfailmnommagui .111111101:111111111MINEN=MMENI 6I ITINENTEMMEMIZIIM1111111=MMEMII 5I 1 .0.39kCIMINEEMEE31 11111" 4 1

1 yi36 3 + 2.0U9!!. pl 11/1111 511111 9: 2 1 NM O _c I 1 , i iiiiiiiiiiiii o =1MENCIMP51=lniMM a i miimlimmirm a) 111 =111%1M-111-111 =M111111111111111 i o , lllimilinifiallll MENU= O

, 11'1111111lift 111111 Fig. 529. Regression of B content of lr, pot-grown wheat (y) on hot water iviI 11, i soluble soil B (x), Sierra Leone. For 1 :, .2 .3 .45.6 .810 2 3 456 81 detailsof summarizedinternational B in soil, mg/l (Hot w. sol.) background data, see Chapter 4.

100 80 n= 3 60 -±s= 5 ±082 6. 50 2±s'04 010 0 ± 40 E =647,3 310 x OX -3o r= 0 470 ". " 20 a, _c 10 8 6 5 4 3 CO 2

III Fig. 530. Regression of B content of -I pot-grown wheat (y) on CEC-corrected .081 .2.3 .4 .5.6 .8 1.0 2 3 4 5 6 8 10 (hot water soluble) soil B (x), Sierra CEC -corrected B in soil, mg/l (Hot w.soi.) Leone. 387 Copper. The average Cu contents of Sierra Leone soils and plants are lower than those in any other country participating in this study (Figs 27, 29, 531 and 532). Only 10 percent of the sample material show "normal" (Zone III) Cu contents and 69 percent fall in the lowest Cu Zone (Fig. 532). It seems that the typical characteristics of Sierra Leone soils: low pH, coarse texture, high organic matter content and low cation exchange capacity, contribute to the low Cu levels of the soils and plants in this country (See Figs 523, 28 and 30). Cu deficiency has been reported in coffee and cacao groves (Hague and Godfrey-Sam-Aggrey, 1980).

22 20Elliiiiiiiiii11110211111111111111111 1111111112,1111115 11111111111111111111 11111111111111MAIIIIIII11111111 1111111111111111111111111111111 .612111111111111111111111111111/1111 I 11111111111111111111111/111111 811111111111111111111111=11111111 ()51111111111111111111111101111111111

41111111111M211111111M111111 Fig. 531. Regression of Cu content of 2IIILMINMENE11111111111111111 pot-grown wheat.(y) on acid ammonium 11111111111111111M1RMIIIIIIII acetate-EDTA extractable soil Cu (x), .2 .4.6 .810 2 46 810 20 40 6080 Sierra Leone. Cu in soil, my( (AAA c- EDTAI

22 itelr it: \ 1 L: 2

1 r!..A111Illi'±-s'iiilEll MIMI

1 iillIll m; NMI MIIIIIIII

1 i 1 111 1111101 ., ,RA.i.,. 5 ...... , NIMBI!1111111111111 Fig. 532. Regression of Cu content of ? .mgOrgiliMMI I 119111 pot-grownwheat(y)onorganic I -"411114111,Mani In 1-14:5 carbon-corrected (AAAc-EDTA extr.) )1 .2 .4.6 .810 2 46 a in 20 al 60 im soil Cu (x), Sierra Leone. Org.C-corrected soi CU,mg/t (AAAc-EDTA) 388

3 00 Fig.533. Regression of Fe contents 2 75 2 50 of pot-grownwheat(y)onacid 225 ammonium acetate-EDTA extractable 200 soil Fe (x), Sierra Leone. 1 75 E 1 50 o. c1:125 o 100 .690 .680 70 o 60

10 20 40 60 100 200400600 1000 2000 Fe in soil, mg/l (AAAc-EDTA)

Iron. The Fe situation in Sierra Leone seems quite "normal", with, almost all the Fe sample pairs falling within the Fe Zone III (Fig.533),and consequently, problems due to Fe are unlikely. Manganese. Owing to very low pH (< 5) of most Sierra Leone soils, there is high availability of Mn to plants but only moderate extractability by DTPA (Fig. 34). This is the reason for the apparent contradiction between high plant and low soil Mn values given in Fig. 534. This contradiction is partly eliminated by pH correction which doubles the average soil Mn values and improves the plant Mnsoil Mn correlation from a non- significant to a highly significant level (Fig. 535). Because the variation of Mn values in the Sierra Leone sample material is relatively narrow, and only "normal" values are included, Mn problems seem unlikely. Molybdenum. The absorption of Mo by plants is lowest from soils of low pH, but coarse texture, high organic matter content, low CEC, low electrical conductivity and low CaCO3 equivalent are factors which also contribute toward low Mo availability (Fig. 16). Many of these factors affect the extractability of Mo by A0-0A in different ways but when the A0-0A soil Mo values are corrected for pH the effects of these factors on soil Mo are very similar to their effects on plant Mo. That the above mentioned soil properties are typical of Sierra Leone soils (Fig. 523 and Appendix 4) explains why Sierra Leone occupies the lowest positions in the "international Mo fields" (Figs 15 and 20). The correlation between plant Mo and uncorrected A0-0A soil Mo is not significant but is highly significant for pH-corrected soil Mo values (Figs 536 and 537). More than80 pergent of the Sierra Leone Mo sample pairs fall in the lowest Mo Zone (I) and over 90 percent in the two lowest Zones (Fig. 537). Only four out of 48 sample pairs show "normal" (Zone III) Mo contents. The extremely low Mo content of Sierra Leone soils 389

Fig.534. Regression of Mn content 2000 nommunnamomrnmememws of pot-grown wheat (y) on DTPA 1000 11103511111111111111111 extractable soil Mn (x), Sierra Leone. NO I.erWiin.rmiti;am104/Mim.toomimlo=lou 800NI VEMIMIUM1171111:051FRINOVAIIMINEN11111==MENI-----111111 600liIII mmoisiniblumninimmmumm121171111M1111 1iIMMINIMINIIIIIMII=IMM 34000011MINEMIIIIEIMERIIII IMIIIN ELLE iiMEMMIENMEMIliiiiM g 200IIMIIIIIIINIIIIIIIIIIIIIMISI

100111111111111111111111111E,_Al 80=moc...,==...... = -E. 60IIIIIImMmmummimummuMma 8 IIIIIMINMIRIMINNIPPMMMIIIMM 40IIIIIMMIIIENSIIMIIIIINg 30INIMMIIPM11111=111111111111=1111S 20INE21111111111111111111 i11111111111111111111111 10...... 80111MIMI=INIMMEMOinMIONTONIMMINEMliolaillimMimasse..MTN= IIIMMIMM=MIIIIIIIIMiZIMMIILy:01Ma21 n MOM 234 6810 20 30 40 60 80100200 300 Mn in soil, mg/i (DTPA)

Fig.535. Regression of Mn content 2000 of pot-grown wheat (y) on DTPA Mn extractable soil Mn corrected for pH 1000E. mmmirmionmo=wirrwmamgIiíEI 11111 800 =. k13LILIAl!IIMININEL3 LIIMMEN mmomormai (x), Sierra Leone. IN WIIIMIZIENVEITIMMINTETIREglatit=mmolmst Emma4ixTrammimmosin E 60011MISMU121111 WITNEIMENII a 40011MM1111=plant12.111111 300 Pr 200 111 o 100 1.11111 80 Mn=narl:iMM=11:11 60 =MIgumi 40 =MOW 30 MINER 20

10 8

2 3 46 810 20 3040 60 100200300600 pH - cor rected Mnin soil, mg/L (DTPA)

must be taken into account especially if crops with a high Mo requirement, e.g. nitrogen fixing legumes, are grown. For example, Hague and Bundu (1980) reported a good response of soybean to Mo in Sierra Leone. Zinc. According to the analytical data on Sierra Leone soils and plants, the Zn level seems quite satisfactory irrespective of the method used for extracting Zn from the soils (Figs 538, 539, 41 and 43). With the exception of a few relatively high Zn values, the samples fall in the "normal" range (Zone III) and the analytical data obtained by plant and soil analyses are in agreement. At some sites a response to Zn could be expected. 390 Fig. 536. Regression of Mo content of WOMMENIM=MMENNI 6 MINSMUNIN=NEPEMMInEn pot-grown wheat (y) on ammonium oxalate-oxalic acid extractable soil Mo 3 (x), Sierra Leone.

2 IuuIIIIIIIIIIIIIu 111111111101

ViiIIIIIUUIIIllk411=11RilIMIMMEMMMINIMEN . 11IIIMINUMMEMENN11MMMENUM14EMNIMMIMINEMBEIN11=1=MMOINNW' 06//111111111111011111111=MIll WIIIIIIMMINIMIEIMI11MMEM

.03 .02

UI .01

.01 .02 .03 .06 . .2.3 .6 1 23 Mo in soil, mg/i (A0-0A)

Fig. 537. Regression of Mo content of itnwtomi-Mrannimrrma.m k. pot-grown wheat (y) on pH-corrected 6 IIIMEMONI MENIIIIFIZIEDEESE0111=IMEGAIIII A0-0AextractablesoilMo(x), 3 Sierra Leone. 211111111111d1M1111 ilIlIllUllIllIll

.1iiiiiiiiiir__iu_____ III pricamramonsi III .03 .021115101

.01 11111111111.1III .006 .01 .02 .03 06 .1 .2.3 .6 1 23 pH - cerrected MOin soil, mg/l (A0-0A)

9.5.4 Summary

The essential feature of the Sierra Leone soils is their low pH. The majority are coarse textured and have a relatively low CEC but are high in organic matter content. With the exception of nitrogen, the macronutrient contents of soils are very low. Of the six micronutrients studied, the analytical data for Fe and Mn are usually "normal" but there are strong indications of Cu and Mo deficiency and some of B and Zn. 391

Fig. 538. Regression of Zn content of 200 /.19M NIVIMUMMENI pot-grown wheat (y) on DTPA extract- Zn 77 able soil Zn (x), Sierra Leone.

100 111111115 90 80 70 E 60

o20 0 N

10 9 8 7 6 III

.1 .2.3 .4.6 .81 23 46 810 20 3040 60 100 Znin soil, mg/1 (DTPA)

20 Zn 7. -2°5\neIIIMI

10, s4: 1 1111 I 9 8i Miliillini"iramui 7 6¡ IllIlE111Blr i E 5 MIN MIlIO gNIME .2-4 liniwAss 1 o Hlipq

1 : 3 1Ildh EMI II o 1 11011111111111id +0E; 2 o

1 1 ) 10"-lopi Fig. 539. Regression of Zn content of pot-grown wheat (y) on AAAc-EDTA iii v z ' ..m.IIK.1La EEO extractable soil Zn corrected for pH :1.2.3 .4.6 .8 1 23 46 810 20 3040 60 100 (x), Sierra Leone. pH-corrected Zr in soi , mg/1 (AAAc-EDTA) 392 9.6 Tanzania

9.6.1 General

The sample material of Tanzania consisting of 175 maize-soil and 5 wheat-soil sample pairs was collected from the following Regions: Arusha (25 sample pairs), Coast (7), Dodoma (8), Iringa (10), Kilimanjaro (27), Mara (16), Morogoro (13), Mtwara (13), Mwanza (10), Ruvuma (12), Shinyanga (3), Singida (13), Tabora (2), and Tanga (21). The approximate distribution of sampling sites is given in Fig. 540. With two exceptions the sampled crops were grown under rainfed conditions. The soils vary widely in texture (from TI 13 to 88) but coarse and medium textures predominate (Fig. 541). Wide variations are characteristic also for pH, organic matter content and cation exchange capacity but soils with moderate acidity, medium organic matter content and low to medium CEC were most Common. As will be seen later, the

30° 36° 38°

1 SU013A

USO4,.., P 44,359- 05,1\ 1Ai44366-375.... w ... 1 g 1---J ANZA BIHAF7AMIJI.0 ' I /---) /A/ i /44347-368 +-\'--N 4439406 392 \,, A\ ' 1../. 44376-

/ , \ KAHrIA --711"57-'-9-17?\ C'.. / 0 /---\ 44418 423 S3 r49340-343 4/713;2\ -335 ....A .,,/ 1 49338-339 '-' AI ,..,...... , 44424 4 KIGO A TABSRA __ 431-439i \ N. ; \--;s-44344-346 ___44433-4, TANA.G \ ,,..._.) \ Z"'Ak °444542-45 w 49938-490/ MANYON1 \ ./ ,, __,_.9 '---- 4444 447 ' 4, I k i44337 1 \ JI ZA IBA (0'...0,Ak("ii,i114303-306 ,,--\ T9,,1S99 - 4 5 A1 49324 - 331 ._ / .' !'. Ciii_E; sr 9407 - 313..4o / (444 01-302 DAR-ES-S LAAM --../ MOR0608*

01111111S /MM. 41SUBAWANGA ww- N,

WA \ NJE

49490.LINDI 10 10. 70'. 94987 99985A 94456A 444133 AFtA 94986* 94459 . 49997 TANZANIA 4t,eAe44499

120 le0 .0.1 12°

30° 32 36° 38 90°

Fig. 540. Sampling sites in Tanzania (dots wheat fields, triangles = maize fields). 393 wide variation in the above soil properties are reflected as wide variations in the macro- and micronutrient contents of Tanzanian sample material. The electrical conductivity and CaCO3 equivalent values and sodium contents are usually low (Appendixes 2-4).

45 Fig.541. Frequency distributions of 35 Tanzania Internat. Tanzania internat. 3764 n. 180 40 n= 180 3783 =39 44 8.6.27 6.64 texture, pH, organic carbon content, 30 Os= 20 16 05= .77 1.12 and cation exchange capacity in soils min.= 13 9 35 min.= 4.39 3.62 max.. 88 92 non.=7.80 8.56 of Tanzania (columns). Curves show 30 theinternationalfrequencyofthe 25 same characters. 20

10 20 30 40 50 613 70 80 90 TEXTURE INDEX pH (Ca C12)

Tanzania inter nat. 35 Tanzania Internat. 40 n= 180 3779 n= 180 3777 =1.4 1,3 O=23.2 27,3 = .9 1.2 30 no =15.4 14,6 35 min.. .2 .1 min.= 3.1 .2 max.= 4.8 39.1 max.= 60.0 99.9 o 25 30

y.25 20 O uj20 /5

L9 15 10 U- 10

.2 .4 .8 lb 3.2 6 2.8 256 16 32 64 ORGANICC. % CEC, me/100g

9.6.2 Macronutrients

Because only five wheat-soil sample pairs are included in the sample material, the following graphs show only the frequency distributions of macronutrient contents of maize and the respective soils. The averages for wheat and wheat soils are given in Appendix 2. It is noteworthy that only very low grain yields (200-1000 kg/ha) were expected from all original sampled wheats and from half of the sampled maize crops. The total nitrogen contents of Tanzanian maize soils vary widely but are somewhat low when viewed in the international context of this study (Figs 542 and 6). The average N content of maize is lower than that of any other country. This must, at least in part, be due to the minimal amounts of nitrogen fertilizer applied (8 ± 17 kg N/ha) to the sampled crops. Very wide variations are characteristic of the NaFICO3 extractable phosphorus contents of Tanzanian soils as well as of the P contents of maize (Fig. 543). The average soi/ P content is somewhat higher than the respective international mean but the average P content of Tanzanian maize corresponds to the international mean. Only about 10 percent of the sampled maize crops had received phosphate fertilizer. The rates varied from 4 to 48 kg P/ha and for all sampled crops averaged 2 ± 6 kg P/ha which probably explains the relatively low P contents of maize. See also Fig. 7. The potassium status of Tanzanian soils is generally high but varies largely from one soil to another (Figs 8 and 544). The K contents of Tanzanian maize are close to the 394

LO Tanzania/nternat Tanzania Internot n. 175 1967 35 0. 167 1958 i.t.,n.t. 9. 2.16 3.14 Tanzania 30 5. 3.1 31 35 n.175 1967 0.9 1.0 ±5..45 .87 9..32 .33 cd3 30 min...88 .88 min.. 1.0 0.6 025 Max.= 6.7 6.7 max.= 3.50 6.51 30 min ..05 .05 max...60 1.04 a. 0.25 U 20 6 W 15 o a15 LL1 cc 1c u_

2 3 4 5 6 2 3 6 7 N content of maize.% P content of maize,% K content of maize,% 9 70 TanzaniaInternat. TanzaniaInternat. 15r.. 175 1967 5. 32.7 22.5 Tanzaniaini,mct nr. 175 1958 75. 66.6 5.,116 33.0 n. 175 1967 .135 min.- .8 0.1 5. 639 330 00,077 .088 0 max 456.0 656.0 35 Os. 734 3512 .008 1- min.. 18 le rnax,.425 1.657 max -.5598 5596 '3'25 825 a.Lc1.3 >- 20 0 L.)

lii15

10 lo u_ 10 - LL A-,-,41/,"AwA/"., A 5 10 20 8 3212 5 .005 .01 .22 .124 .08 .6 .32 .62 1.29 25 10 160 25 50 100 200 200 5051500 3200 6400 N in maize soiis.% P in, maize soils, mg/1 K in maize soils, mgil Fig. 542. Nitrogen, Tanzania. Fig. 543. Phosphorus, Tanzania. Fig. 544. Potassiurn, Tanzania.

Tanzania internat. Figs 542-546. Frequency distribu- n.175 1967 rTanzaniaintemai .43 .47 5 n. 175 1967 tionsof nitrogen, phosphorus, t5 .17 .20 7..223 .251 min.. .09 .09 z6..113 .119 mox. .99 1.88 .036 potassium, calcium and magne- r1,3%. -.772 1.125 sium in original maize samples 41_ 25 - 7- and respective soils (columns) of 20 Tanzania.Curves show the inter- o 15 L.. national frequency of the same 10 u_ characters .

Go content of maize Mg content at maize,%

Tanzania Internot <6 r 175 1967 5 7= 2104 3450 TanzaniaIntern., Is- 1727 2815 35 r, 175 1967 min..155 10 no.., 9265 17995 R.368 406 is, 293 462 min., 23 1 max.. 1355 6499

0 100 260 202 090 1630 3293 6403 12800 2.5 25 00 100 200 400 500 1605 3202 6432 Co in maize soils, mg/I Mg in maize soils, mg/. Fig. 545. Calcium, Tanzan'a. Fig. 546. Magnesium, Tanzania. 395 international mean with correspondingly wide variations. No potassium fertilizer was applied to the sampled crops. Wide variations are also typical for exchangeable calcium and magnesium contents of the Tanzanian soils and for Ca and Mg contents of maize (Figs 555 and 556). The average contents of these elements are somewhat lower than the respective international averages (Figs 9 and 10).

9.6.3 Micronutrients

Boron. With relatively few exceptions the B contents of soils and plants are within the "normal" B range (Zone III, Figs 547 and 548) and Tanzania locates near the centres of

10 311=1M1111=111111WAIMMININallIMMIMI ) Le iin(lblillIMOMIWATiMMIIMIlliiMIIMMINIMIMINMININIIIIIIMIIIIIIN 8 IMIIMMEINIIMININIIIINE 0:5111IERWM11111111111 6) MIUMMIIIIIII 51 IIIIIIIMEEM 89WrifIllIIIMININIMINIIIIIII 4' MEM-= ° 78 ± IZACEME1111 IN NW E3))1NRI 71+a33WEINIII Igil 0_ 2MEN111111111 1/1111 o _c 11,111 1 11MUMM.W.IMEIIMIMEMIIMMEEmil nu o1 alinaMirgirliniWkilikagran:111 IMIIIIIIIIIMMIIIIIEIIIIIIPIIIII dilliMUCIUM11111 a, iiiiiirMaltataiiiMaterallIMMOrni 8 111111MaarrMaSengliniNEEMMill 0 111111MAKI ME MIN1111111 ME R1111L Fig. 547. Regression of B content of Ellinlillin pot-grown wheat (y) on hot water lit soluble soil B (x), Tanzania. For de- 1 fig ni I 1 tails of summarized international back- . .2.3 .4.5.6.810 23 4 5 s 8 ii ground data, see Chapter 4. B in soil, mg/l (Hot w. sol.)

100 / I 'LLJ_LL 80 n1 3 60 Sits. 7089 e 50 i±s.09 +09 0 72 40 a.30 lo .0 7 +0 lo 2 +5.50x 0. r..0346** 8 4* 15, 20 a) _c

8 w56 4E. 4 3

2 Fig. 548. Regression of B content of pot-grown wheat (y) on CEC-corrected (hotwatersoluble)soil B (x), .081 .2.3 .4.5.6 .8 1.0 2 3 4 5 6 8 10 Tanzania. CEC -correctedB in soil, mg/1 (Hot w. sol.) 396 22 nte no Fig. 549. Regression of Cu content of Cjc±t" latit3 20 pot-grown wheat (y) on acid ammonium 3 acetate-EDTA extractable soil Cu (x), 18 7 1211i11111111111111111 E Lo Tanzania. Q. 12111:211111:11:211111111111 o 14

.5 12 111111111111111111

10 111111111111111121111111 - 8 1111111111110M11111111 o6 11111111111111111=1111111 111111111E1-21111111111111111111

2 MiiiIIILVNIp1111111111%11 1111111111111MEMMIIIIIND .4.6 .810 2 4 6 810 20 40 6080 Cu in soil, mg/I tAAAc-EDTA)

22Cu Fig. 550. Regression of Cu content of pot-grown wheat (y) on organic carbon- Tonni 20 corrected(AAAc-EDTAextr.)soil Cu (x), Tanzania. 18 FUNIIIIIIMM11111011011111MGM 'rcniii5.111111111111111111 iummomummon'Emmummoommenn 10ommoonmemmum 8ommmmortmo 0o 6 C_) 41111111M12111111111M1111111

2 111111111011MPOMIONE 1 .2 .4.6 810 2 4 6 810 20 40 60 100 Org.C-correctedsoiCU,mg/l (AAAc-EDTA)

the "international B fields" (Figs 22 and 25). No extremely low or high values were recorded and, consequently, these analytical data do not indicate any widespread B problems in the country, although at some locations shortage of B may reduce the growth of crops sensitive to B deficiency. Copper. Tanzania's national mean plant and soil Cu values are only slightly lower than the respective international averages (Figs 549 and 550) and itslocations in the "international Cu fields" are almost central (Figs 27 and 29). However, variation among the Tanzanian samples is wide and several very low and a few relatively high Cu values occur. About 14 percent of the sample pairs fall in the lowest Cu Zone (I) and 23 percent 397

Fig. 551. Regression of Fe contents of 300 275 li pot-grown wheat (y) on acid ammonium 250 siantatin acetate-EDTA extractable soil Fe (x), 225 PRUNIMMIR Tanzania. 200MININIMINEASIMUNINIU 1 75IMEEMINIKESSUMMIN E 1 50klEMMErninnfiniini a. 9:12511111111111110111MMINO D cG) 100iunuiiuiiiiiii .5 90 t 80 a) 70IIMMEMMELMIIMMTOI o 60111111111110MONIMINI

50111121141111.11111111111111

30111111111111111111 10 20 40 60 100200 MX/ 600 1000 2000 Fe in soil, mg/I (AAAc- EDTA)

in the two lowest Zones (I and II, Fig. 550). Low Cu values were relatively more frequent for samples which came from Mara than from other Regions. More than two-thirds of Mara samples show low (Zone I) Cu values indicating apparent shortage of Cu in this part of the country. Iron. The average analytical data on Fe contents of Tanzanian sample material given in Fig. 31 and those of single sample pairs presented in Fig. 551 indicate that both the averages and the variations of Tanzanian plant and soil Fe values are of the same order of magnitude as those for the total international material in this study. Although several Fe points fall outside the "normal" range (Zone III) these are not in such extreme positions that severe disorders due to Fe should be expected. Manganese. The non-significant plant Mnsoil Mn correlation (Fig. 552) becomes highly significant when he DTPA soil Mn values are corrected for pH (Fig. 553). No low (Zones I and II) Mn values were recorded from Tanzanian samples but some relatively high Mn values were measured from samples which came from sites with acid soils. These were somewhat more frequent for samples from Tanga and Singida than from other Regions. Molybdenum. In the "international Mo fields" Tanzania stands c/early on the high side (Figs 15 and 20). However, the ranges of variation of Tanzanian plant and soil Mo values are exceptionally wide and both very low and very high Mo contents were recorded (Figs 554 and 555). Low (Zone I and II) Mo values occurred in samples from several Regions but were relatively more frequent in samples from sites with acid soils in Ruvuma, Mtwara, Morogoro and Dodoma. Almost all high (Zone IV and V) Mo values were obtained from Arusha and Kilimanjaro samples from sites where the soils are alkaline or only slightly acid. Zinc. The Tanzanian national averages for plant and soil Zn correspond to the respective international mean values placing Tanzania at the centres of the "international 398

2000 Fig.552. Regression of Mn content of pot-grown wheat (y)on DTPA 1000 OMM11mmoimiNZNION1111111111111111111111 extractable soil Mn (x), Tanzania. 800iirsawiwnii=wimmanii.. 600111=1111MMANIE011111111=satrawrirmamEmn=enunnimm E 400umannamminmimomm Q.300111111111MMIIIIMMENINIIIIMINI 200INII).1111111=1111111511itillIS .c o 100rri4wwzz4frimignimpiammli. 80 11 3111 47A161:111111r` "do 60111==MwEINEINIIMPPIMINNIMPIIMMEM 4011=1/01.1111MVIIIIIIII 30 20

lo'11===1111=011111111111E111111111111 111IMMIIII111 MWM 811.11111i41=2.1117aMMEMEN maz 111 =IIIMMIMINNOINi.1MICZMMINEN 234 6810 20 30 40 60 80100200300 Mn in soil, mg/l (DTPA)

2000 Fig.553. Regression of Mn content Mn of pot-grown wheat (y) on DTPA 1000MN MIdElgWEEM3/1.1=LIWEEIIMMIIMOININ!PE71.7.111=1"ru.MM=11111111Nr11119111111111111 extractable soil Mn corrected for pH 800 OMMMEMENImIMMINMEMM11=1MIMII11 (x), Tanzania. E600111WMINT4ThlinliniMENFORECIMMEMMErriii 0- 400 9 "E; 300 200!UlIIIIlI!IIIIUUIII

't100 egg 80NIIbamPCIVINIF121MmommaniMIMMnallIMI o 60 40111111MMEIPEMIMINIMIIIIMM=M1111 30 PP" 20 10 Itimmnlmm===II IM=IMMEMIIMIEMMEEMINIMIlri%TaMUMENIleaMINNI =MMICLEINIM=MMETU 2 3 46 810 20 3040 60 100200300 600 pi-l-correctedMnin soil,mg/t fDTPA)

Zn fields" (Figs 41 and 43). As in the cases of most other nutrients, the variations in Zn contents are very wide (Figs556and557).The results obtained by either of the extraction methods (DTPA and pH-corrected AAAc-EDTA) show a good correlation with the plant Zn analyses, but the former method gives relatively somewhat higher Zn values than the latter, especially at low Zn levels, i.e. more than twice as many plant Zn AAAc-EDTA Zn points fall in Zone I than in the case of DTPA extraction. Nevertheless, both methods give quite clear indications of a shortage of Zn at many locations in Tanzania. The lowest Zn values were measured from samples of Mwanza, Shinyanga, Singida, and Mara while most of the high Zn samples came from Kilimanjaro, Tanga and Iringa Regions. 399 Fig. 554. Regression of Mo content of glinlign=a1MMIMISTM pot-grown wheat (y) on ammonium 6 oxalate-oxalic acid extractable soil Mo (x), Tanzania. 3 2 E D. 111111111111111111111111 111111WPISCIUMMUSW:M=1INIMEMMUM=Mm71=11 MEMINIMMIMMEMIMISSIIMIN111111Md MMMMUNiniiiZSIVSMMMUMM 11111EMPRIMENIMININIII

MIWM/MMPdtfillAIMMIMIMMINOMMIMINIENNWIIMA. NO/MIinliMEMINIIIMiMMIN=M11MENEMMEni. .06WEINEngain MilIMMIMMEM

.03 .02

0111111111.11111111 .01 .02 .03 .06 .1 2 .3 S 1 23 Mon soil, mg/ (Ao-oA)

NIN igiMiNLINimmi Fig. 555. Regression of Mo content of EMPJ11111MEINIMOIIIMIIIII =mama MINE M o WIIIMEMNIII1171117114 MIIIIMMINNMEN pot-grown wheat (y) on pH-corrected 6 Lo A0-0AextractablesoilMo(x), misunitinnumbikummon'"37msturnmul Tanzania. 3 121321111113141;151212111.111 2MEINr°g

o_ 111111111111111111011111111 111=1MIENMMIOINMNIIIMMINW1001141MMIRIIIMI, INEMIIMMIIIIMME=IMMININIMMEMIERMEMIMAIIIIKIN V,.6NIMINEHIIIINMENINNIIMMIIMIWAROMPI 1111111111111=11111111111/111111.1.21131111, o.31111111111111111111111111111111Ezionow

w .2 o o .111111110111191111111110 LIEILIOIWM=E61211121 1=1=16=111111 .06 UUIIIIId1UMMIIINUNIMMENIIMIN II 11111111111111 .03 .02

.01111111111111.111111 .006 .01 .02.03 06 .1 .2.3 .6 1 3 pH- corrected MO in soil, mg/I (A0-0A)

9.6.4 Summary None of the soils sampled from Tanzania for this study could be called a "typical Tanzanian" soil since soils varied greatly in all properties examined. However, soils with coarse to medium texture, moderate acidity, medium organic matter content, and low to medium CEC predominated. 400

200 Fig. 556.Regression of Zn content of pot-grown wheat (y) on DTPA extractable soil Zn (x), Tanzania.

100 90 80 70 E60 °- 50 o .240

Nio

10 9 8 7 6

.1 .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 Zri in soil, mg /I (DTPA)

20 1,1, Fig. 557.Regression of Zn content of pot-grown wheat (y) on AAAc-EDTA extractable soil Zn corrected for pH I IlI (x), Tanzania. ¡ Emoceak. .4ilMi1111=11111111114 )...... ffil -mgoolummmouzal )111111MEN.E"1=1 Iii MI"IlmNII :i1 II=NAN we 1 V EllmilMifiNMN 11111111Millillik.iliiI 111111PitIII

1 11411111111011111I vismmilumi wrrias.low 6 EtaIiiii i u ,,, . . . .8 1 23 46 810 20 3040 60 100 pH-corrected'Ll"IiIIIII Zn in soil, mg/l (AAA.-EarA)

Wide variations are characteristic also for the contents of all macronutrients, buton average the N, Mg, and Ca contents of soils are slightly on the low side while P and K contents are on the high side compared to other countries in this study. The analytical data of the six micronutrients studied indicate that the most likely micronutrient problems in the country are those caused by shortage of Cu, Zn, B and Mo. 401 9.7 Zambia

9.7.1 General

The Zambian sample material consisting of 46 maize and respective soil samples was collected from the Central and Southern Provinces and effectively covers the maize growing areas of the country (Fig. 558). A great majority (41) of the sampled soils were classified as Nitosols. Two soils were classified as Arenosols and the remaining three as a Ferralsol, a Fluvisol and a Gleysol.

26 27 28° 29°

SAM F TA

REPU LIC OF I .----- GHILIL4[70-1-90WE) ti 0 \...-9.-^e-9,4. ZAMBIA CH3GOL is., (CENTRAL ANOSOUTHERA;PART) L"'N !COLA() ',.,,, 13° 0 5 30 60 0

...- (..' 48969 4:58

A 48950

98951 o. o 98964 o A "t 'AE:876956,24 8 e e I- 48953 '? ,,,,,,,,e9/9 469,7i 48957 41"9tA 48966 48970 4 956 15° - 48954 48962 1 48965, A 48973 98951

LUSAKA O

48967

4899 48975 A

fEJ . 482.74619774A1189 48978A A 48979-4841 "V4:995

18982 48983 489135-98987 48988 7 17° K°4849/ril9 48989 r

4TA°48991

i je48993 A 48992 / Li G TONE

° 18 25 2S 27 . e

Fig. 558. Sampling sites in Zambia. 402 About two-thirds of the soils are coarse textured (TI < 30) and one-third have a medium texture (Fig. 559). The soil pH varies from 4.2 to 7.1 but acid soils predominate and the national mean pH is the second lowest (after Sierra Leone) among the 30 countries in this study. The national mean organic matter content is also among the lowest and that of CEC the absolute lowest recorded in this study. The values of electrical conductivity, CaCO3 equivalent, and sodium content are also low (Appendixes 3 and 4).

Zambia internat. Zambia Intern' n.45 3764 = 46 3783 13.27 LL 5.02 6.54 20 00.11 16 004 .73 1.12 m,n 11 min - 4.10 3.62 1300 55 92 300.7.14 8,56

20 ai o u_

d'd 10 3 0 40 50 63 /5 50 90 TEXTURE INDEX pH (Ca II.12)

Zambia InternCt Zambia InternOt r 45 3779 n= 45 3777 5. 8 1.2 -;211.4 27.3 t, .4 1.2 -432 6.2 min.= .2 min.. 3.4 6498.=0-0 39.1 802.428.2 99 S

Fig.559.Frequency distributions of texture, pH, organic carbon content, and cation exchange capacity in soils of Zambia (columns). Curves show the internationalfrequencyof thesame . 15 3.2 5.4 129 25 8 16 32 ORGANIC C,% CEC, me/100g characters.

9.7.2 Macronutrients

The average total nitrogen content of Zambian soils is the second lowest (after India) in the international scale (Fig. 6), but the N contents of maize are of the average international level (Fig. 560). This is because almost all sampled crops were heavily fertilized with nitrogen (Appendix 5). The contents of NaHCO3 extractable soil phosphorus are of the average international level but since the sampled crops were rather heavily fertilized with phosphates the P contents of maize are relatively high (Figs 7, 561 and Appendix 5). The exchangeable potassium contents of Zambian soils vary considerably but are generally low and several very low values are included (Figs 8 and 562). Although moderate dressings of potassium fertilizers were applied to the sampled crops (Appendix 5) the K contents of maize remain at a very low level. The Zambian national averages for exchangeable calcium and magnesium contents are among the lowest in this study (Figs 9, 10, 563 and 564). The Mg contents of maize are also relatively low but in relation to soil Ca the Ca contents of maize are high. As pointed out in Section 2.2.4 (Figs 12 and 14 and related text) plants seem to absorb these elements easily from coarse textured soils with low CEC and these properties are typical of Zambian soils. 403

40 Zambia Internat. Zambia Internat. 5= 45 5=1958 40 n. 45 1967 35 ZambiaInternat. 30- 0.2.1 3.1 7= 3.16 7,3.14 rim 45 1967 is = 0.5 1.0 .41 OS. .87 ts, C 8..40 .33 Min, 0.9 0.6 S30 min.= 2.29 min., .88 o s4. .09 .10 O 25_ max.= 3.2 6.7 m08., 3.94 max.= 6.51 30 min...19 .05 max.= .59 1.04 25 o) o- 20 o2-

1.1.1 15 15 o o ccLi] /0 u- u_

3 4 5 6 2 .3 .4 .5 .6 4 5 6 N content of maize, % P content of inaize,% K content of maize,% 0 70 Zambia Internot 1967 Zambia Internat. 5 n= 45 35 Zambia Internet 7= 20.8 22.5 60 n =45. 1958 65.20.4 33.0 n. 45 1967 ii- g=.073 .135 min.=2.8 0.1 = 145 330 C in =.035 .088 ....a. 30 mox.=114.5 656.0 rs = 202 356 6., 0 50 sin. =027 .008 mino 18 18 max.,..180 1.657 max. .1272 5598 L. 25 a. CL 40 >- <>3 20 z 0 o U-I20 II-

2.5 5 10 20 40 80 160 320 640 60 100 00 .005 01 .0 .04 .08 .16 .32 64 1.28 N 'n maize soils,% P in maize soils, mg/I K in maize soils, mg/I Fig. 560. Nitrogen, Zambia. Fig. 561. Phosphorus, Zambia. Fig. 562. Potassium, Zambia.

Zambia Internat. Figs 560-564. Frequency distribu- ZambiaInternat n 45 1967 n= 45 1967 =.48 .47 tions of nitrogen, phosphorus, S5 =.14 .20 =.108 .251 min.= .27 .09 =.054 .119 max.= .82 188 4) 30 min =.092 .036 potassium, calcium and magne- RICIX = 369 1.125 O. 25 sium in original maize samples and respective soils (columns) of Zambia. Curves show the international frequency of the same characters.

5 Ca content of maize % Mg content of maize,%

Zambia Internat. n. 45 1967 Zambia Intermit 55= 665 3450 55.570 2815 35 n= 45 1967 min. 10 10 0.120 446 mox .2550 17995 05= 132 462

= 1 1 max.= 466 5490

0 WO 200, 400 800 1500 3200 5400 12800 25 25 50 100 200 400 8001850 3200 6400 Ca in maize soils, mg/I Mg in maize soils, mg/I Fig. 563. Calcium, Zambia. Fig. 564. Magnesium, Zambia. 404 9.7.3 Micronutrients

Boron. The B contents of Zambian soils and pot-grown wheats vary within quite narrow limits (Figs 565 and 566). Almost all plant and soil B values are lower than the respective mean values for the total international material in this study, and the location of Zambia in the international scale (Figs 22 and 25) is low. Although no extremely low B contents were recorded in the Zambian samples, the analytical data indicate B shortage at several places in the country. B deficiency on cotton in Zambia was discovered in 1966 by Rothwell and co-workers (Anon. 1981). During recent years B fertilizers have commonly

10 I 1=1111MIZVIIMIIIIIIIIIIIIIIMIIIIIMIIIII111.111111111111111111=110111/11ff ALOICAPROMUMNIfirCrffinVMMEIMMONINIM NM Iiiiilill 8 M11111.6.111111.1111111.1111==.1 II.IN 6 5 1 .11LIE7'11,01VAIRhi1nillarillill MINOil wan _0.25 + ciI 073 I 4 .1 ug Egg87+2.0III", 2 631.I E3 I MI Aill 0- MEWIN 111111

h. 1 1616141111111111111 1:wilm mulrillt:269:=1111 inIRIIMS1111 MmilPr' 0101111MEMMINII , Wliumm Al 11=1.111111MMi 11111111111*Wil Mill EIMMONii IMMIMMORIMMLIMMIMMilliiiii iiiiiigiliEMMIlit Illillibl Fig. 565. Regression of B content of i pot-grown wheat (y) on hot water IIIIIIIIIII1iiii [11111 11, soluble soil B (x), Zambia. For details 1II III .1 .2.3 .4.5.6 .81.0 2 3456 811 of summarizedinternationalback- B in soil, mg/I (Hot w sol.) ground data, see Chapter 4.

100 % ei MI i rare=IN "..111.11. MIN 80 r'tPIMIIIIIKIIIMI 60 n- '''alllaaall V ....1 :is-071t ie 50 lf 40 5 4 11....KIL E.3O 111l0il :.,58.=. 1, 201 1 . mi gli,, dl a) _c gill. liii soIiimpo.to 6 Imoort_now-lit ...4 MAIMIP1111111111111M1== MIR MP NIP ITRIIIMMIMPUI gap IIIIII 11111111M111111111:r.10"1 111111111111 MS limaimmaal 1 NIIMMIIIVII INEMIrgill':INMaiErn uum11111111111I Mill

Ill NMI 1111113 Fig. 566. Regression of B content of .08:1 .2.3 .4.5.6 810 23 456 8 10 pot-grown wheat (y) on CEC-corrected C EC -corrected Bin soil. mg/1 (Hot w sot) (hot water soluble) soil B (x), Zambia. 405 been applied, often together with insecticides. Copper. Zambia falls among the low-Cu countries in the "international Cu fields" (Figs 27 and 29). More than half of the Zambian sample pairs fall in the two lowest Cu Zones (I and II) and most of these are in Zone I (Figs 567 and 568) indicating apparent Cu shortages. Although no certain low-Cu areas could be geographically defined many of the samples lowest in Cu came from the Kalomo, Mkushi and Kabwe areas. Iron. On average, the Fe contents of Zambian plants are of the average international level but the AAAc-EDTA extractable soil Fe values are low (Figs 569 and 31). Nevertheless, although several sample pairs fall in the low Fe Zones (I and II), indicating

20 zail1111111111111111111111111111111

181111MNPI350,1111119111111111 E Q. (0.06 0611:1311111111111 o w14 1111111111111111111111111111/1

12111111111111110111111111111111/111 111111111111111.mmardmi 38111111111111IIMMILVIIII1I (-) 61111111111111INMENIIIM111111

Fig. 567. Regression of Cu content of 2 pot-grown wheat (y) on acid ammonium 111111111111ENIMMIllii acetate-EDTA extractable soil Cu (x), .2 .4 .6 .810 2 4 6 810 20 40 6080 Zambia. Cu in soil, mg/I (AAAc- EDTA)

20 NM interning! NM

18 DAMN 411111111111111N3

12 11111111111111 IMINSIMINI 111011111111pOilM11111110 Imosindommtummi Fig. 568. Regression of Cu content of 2 gintiostemmoun pot-grownwheat (y)onorganic Ominnummompammoua carbon-corrected (AAAc-EDTA extr.) .4.6 .810 2 46 810 20 40 60 100 soil Cu (x), Zambia. Org.C-corrected soil CU,mg/l (AAAc-EDTA) 406 3 00 Fig. 569. Regression of Fe contents of 2 75reramonpernmom 2 50 pot-grown wheat (y) on acid ammonium 225 acetate-FDTA extractable soil Fe (x), 200!-Igralitrair Zambia. \n411111111131151 iiiii 175IIUUIIIIIIUIIflIN E1 50 MI .1259.- I"IIlIuIlIIIuuIIuII o 111111 -5 1 00MIMI 111151 z-90=LI 11111111=11IIII 80INIKE111111111 111111111 .1) 701111111111111111=alpflall o o60 ME111111111.MMI 5011111111111=1111 40Milo II ,

30 II 1111111E1111111MORS I Mil 10 . 20 40 60 100200400 600 1000 2000 Fe in soil, mg/l fAAAc-EDTA)

possible response to Fe of crops sensitive to Fe deficiency, a shortage of Fe is not likely to be among the primary micronutrient problems of the country. Manganese. The Zambian mean for plant Mn content is higher than that of any other country in this study (Figs 33, 37, 570 and 571). It is five times that of the international mean and 25 times that of the lowest national mean (Malta). However, the Zambian national mean for DTPA extractable soil Mn is only the eighth highest, but moves up to second place when corrected for soil pH. Simultaneously, the plant Mnsoil Mn correlation is improved from non-significant (r = 0.266 n.s.) to a highly significant (r 0.607***) level. More than half (55 %) of the Zambian plant-soil sample pairs fall in the highest Mn Zone (V) and only 30 percent of those show "normal" (Zone III) Mn contents (Fig. 571). According to these analytical data, excess of Mn is a factor which may cause disorder of plant growth in many parts of Zambia. The samples containing most Mn came from geographically scattered areas where very acid soils, pH(CaC12) < 5, existed. Molybdenum. Unlike Mn, the Mo contents of Zambian plants are low. The national average is only one-third of the international mean and Zambia occupies one of the lowest positions in the "international Mo fields" (Figs 15 and 20). When pH correction of the A0-0A extractable soil Mo values takes place, the negative non-significant plant Mo- soil Mo correlation (r = 0.004 n.s.) becomes positive and highly significant (r = 0.656***, Figs 572 and 573). At the same time the Zambian national average for soil Mo decreases from 0.105 to 0.042 mg/1 and Zambia's position in the "international Mo fields" moves closer to the international regression line. Half (50 %) of the plant-soil sample pairs are within the lowest Mo Zone (I), 9 percent in Zone II and 41 percent in the "normal" (Zone III) Mo range (Fig. 573). Problems due to the low Mo status of Zambian soils are likely to arise, especially if crops are grown which are sensitive to Mo deficiency, such as nitrogen fixing legumes. 407

Fig.570.Regression of Mn content 2000 Ilualromemominiiimeli of pot-grown wheat (y)on DTPA extractable soil Mn (x), Zambia. 1000 s t o IIMPSIIIIMPUMEI 800riLIWIlailtWmilMilia== E111111111111111=11111111111111111 600liMMEMMANOMI NERIVEMENCESINIM 111104419 E 400u r - u -torFpSCIEEIIIIIMMIIIIIIIIIEUIEIIIEIMIIIIIIIBIIII a 300 iuuin Ill Intnot imiii 111PNIMIRIN 14G) 200 L II;ii111111111111111111 .15 100ropmairaliazzormriermIL.& 80liniarmmin=16.ralraN=Ma 7, 601111111111111=1111EMIIIIIMMIIIMMINIiii 40iiiiiiiMINIMIEMMEMOMMEEMEMME 2 30II MOW 111111M1 I 20IIiIIhIiIiIIIHIi

10IiiIIIflIHIIIIIIUII11111111MIIMIIIMIIIIM9111MMIIIM MMIIIMI aEMSBEINALMErineurnantriMMItal 2 3 46 810 20 30 40 6080100200 300 Mn in soil, mg/t fLTPAI

Fig.571.Regression of Mn content 2000 of pot-grown wheat(y) on DTPA extractable soil Mn corrected for pH 1000 800 PIi1i.Ìii!T (x), Zambia. E 600

2-1- 400 .1.1Lampeam 14 300 a) 3 200 Arlo o Ó 100 Ind 80 o 60 40 30 20

10 8

2 3 46 810 2C 3060 60 100200 300600 pH-corrected Mn in soil, mg/I (DTPA)

Samples with low Mo contents occuring in geographically scattered sites with acid soils, pH(CaC12) < 5, are usually combined with high Mn contents. Obviously, in the long run, raising the soil pH by liming would be the best means of increasing the availability of Mo as well as reducing the risk of Mn toxicity. Zinc. Irrespective of the method used for determining extractable soil Zn, the distributions of Zambian soil and plant Zn contents are much alike (Figs 574 and 575) and in both cases the plant Znsoil Zn correlations are good. Almost all plant-soil Zn values are within the "normal" (Zone III) range, but many of them show such a low Zn content that disorders due to Zn shortage are likely (see also Section 2.7). 408

CJIMMITMENNIMIMMWENNImmwm Fig.572. Regression of Mo content 6 of pot-grown wheat (y) on ammonium TrEFORMEZEIni oxalate-oxalic acid extractable soil Mo 3 cç (x), Zambia. 2

uIIIIIIIiiIIIIIIm,nmaimm i.. 1 =MEE MMENIZIMMIELM=1111=1 monomat.Prommun INI1110111101111 11A111111111111111111116. ll=MIMINI 2 =MMEMIN1=rn .06M1=1Ifflk.UINUMMI EMMEMOMMINEW

.03 .02 ifi .01

.01 .02 .03 .06 . .2 .3 .6 1 23 Mon soil, mg/ (A0-0A)

Fig. 573. RegressiOn of Mo content of 6 MENIII6111111 pot-grown wheat (y) on pH-corrected 111111IMMIIIIIMENIBENIIIIMEEREM131111111 EN A0-0AextractablesoilMo(x), 3 31111211111101210gE11111 Zambia. 2 Yr 111

E O.a. 1 1111111111111111111kh1111111PAII o G) .6111=111111=i1MIIIMMIMMINIIIMil ó .3wounimimmunipmpirommon c2.INIMIMEMENIMMEOMMI.19 "E".o oo .111111111111111111INIMP- .06mikera-aa,pasNomm=1.....11

.03 .02

.01111111111111.1 Ill II .006 .01 .02 .03 06 . .2.3 .6 1 23 pH - corrected MO in soil, mg/I (A0-0A)

9.7.4 Summary

Most of the Zambian soils in this study are acid, coarse to medium textured, low in organic matter, and have a low cation exchange capacity. The levels of macronutrients (N, K, Ca and Mg) are usually low but that of P corresponds to the average international level. The micronutrient content of Zambian soils 409

mom!MI 110111=111110 Fig.574. Regression of Zn content 200 1. of pot-grown wheat(y)on DTPA extractable soil Zn (x), Zambia. ii 10011.Il i 90 80 1111' =EN 70 E 60 50 o 4011.111111111111111111(

30 ' ¡Mil ini Ca; iI 20IMPLOgilt..1111111111111111111 o N 1011111111111P 9 8fm."1:111111167MILIuMommul 715dWIMIIIIIIMI 111111MINK1111111011 61111 111111111111=IIIIIIIIMMIllion IIIIIIMINEIIIII ni 111111111=111111111111 . .3 .4.6 .81 23 46 810 20 3040 60 100 Zr1 in soil, mg /I (DTPA)

200 Fig. 575. Regression of Zn content of u I pot-grown wheat (y) on AAAc-EDTA 3 3 extractable soil Zn corrected for pH 4 8_ (x), Zambia. 100 2 lo Ill 90 SiltighIMIMMMIIIIIINIMI 80eldill IIIIIIMIIME1111511111 70 60 II I IliFRO E 50M I I IFI Ili 1 40, o 30 ,IiiilillViiiiilli o 20 k Ai 1.111111 4<-7; ,,ogis o LII

10 11111 9E WEENIE 8Emo INIPIININ 7EM III IMMIIIIIIIM 6 MEE II WEENIE 111 II Lk II II IMMEINEE .2.3 .4.6 .8 1 23 46 810 20 3040 60 100 pH-corrected Zn in soi mg/I (AAAc-EDTA)

and plants varies considerably depending on the element. Only the contents of Zn are usually of the "normal" international level, but at several locations shortage of Zn is likely. Those of B and Fe are somewhat on the low side. Very low Mo and Cu and very high Mn values occur more frequently in Zambia than in most other countries. 411 10. Summary for Part II

The national data for each of the 30 participating countries on the most important soil characteristics, macronutrient contents of the original indicator plants (wheat and maize) and respective soils, as well as the micronutrient contents of pot-grown wheats and respective soils are presented in graphical forms in such a way that in the background of each graph the respective data for the whole international material are shown. Analytical data for soil micronutrients are based on hot water extraction (B), acid ammonium acetate-EDTA extraction(Cu,Fe), DTPA extraction (Mn, Zn) and ammonium oxalate-oxalic acid extraction (Mo). Soil B values corrected for CEC, soil Cu values corrected for soil organic carbon content, Mo and Mn values corrected for pH, and AAAc-EDTA Zn values corrected for pH are also presented. The nutrient analytical data for each country are briefly discussed and compared with the relevant international data as a whole and in the light of the soil characteristics typical of the country or area in question. Data on four important soil characteristics are summarized by countries in Table 29, where the limits for the three categories for each characteristic are such that of the whole international soil sample material approximately 25 percent falls into the "low", 50 percent into the "medium" and 25 percent into the "high" class. Coarse texture is typical of soils sampled from most African countries (except Ethiopia) and the majority of soils of Sri Lanka, India and Belgium fall into the same class. The highest percentage of fine textured soils was recorded from Brazil but soils of this type are also common in most Near East countries, Ethiopia and Thailand. Low soil pH is most characteristic for Sierra Leone where 92 percent of the sampled soils have pH lower than 5.6. Relatively high percentages of acid soils were recorded from many other African countries as also in Finland, Brazil, Argentina, New Zealand and Korea. High soil alkalinity is most common in Near East countries, Pakistan and India. The content of organic matter of soils varies considerably from one country to another. Soils of low organic matter content are especially common in India, Pakistan, Zambia and Nigeria while the soils sampled from New Zealand, Finland, Sierra Leone, Brazil, Argen- tina, Ethiopia and Thailand are generally rich in organic matter. Low cation exchange capacity of soils is most typical for countries where the soils are coarse textured and/or low in organic matter such as Zambia, India, Nigeria, Sri Lanka, Ghana and Pakistan, while fine texture and/or high organic matter content generally contribute to high soil CEC, as in Egypt, New Zealand, Ethiopia, Thailand, Lebanon and Syria. The macronutrient contents of soils of the 30 countries are summarized in Table 30 using the same low (25 %) -medium (50 %) -high (25 %) criteria as in Table 29. The relative frequency of low total nitrogen content of soils is highest in countries with soils of low organic matter content (India, Pakistan, Zambia and Nigeria) and conversely, the high N soils are typically found in countries where soils have a high organic matter content (New Zealand, Finland, Argentina, Ethiopia, Hungary and Brazil). 412

Table 29. Relative frequencies (percentages) of low, medium and high values of soil texture (TI), pH, organic carbon content and cation exchange capacity in 30 countries.

Country TI pH(CaC12) Org. C., % CEC, me/100 g low med. high low med. high low med. high low med. high <33 33-55>55<5.65.6-7.6>7.6<0.80.8-1.5>1.5<1616-36>36 Belgium 56 44 44 56 17 61 22 27 73 Finland 31 29 40 76 24 2 98 3 71 26 Hungary 10 68 22 12 73 15 6 46 48 5 71 24 Italy 19 50 31 5 83 12 6 78 16 14 69 16 Malta 4 84 12 88 12 80 20 100 New Zealand 6 61 32 63 37 100 18 82

Argentina 3 97 68 32 34 66 90 10 Brazil 4 20 76 75 25 4 24 72 7 76 17 Ecuador no data 20 73 7 15 46 38 36 55 9 Mexico 18 46 36 7 48 45 41 56 3 13 47 39 Peru 46 43 11 10 83 7 13 51 36 6 83 11

India 58 34 8 1 39 60 82 18 76 15 10 Korea, Rep. 47 51 2 53 47 11 72 17 36 64 Nepal 28 68 4 32 68 32 56 12 24 74 2 Pakistan 13 85 2 13 87 72 28 0 51 48 0 Philippines 18 55 27 36 64 19 55 25 4 53 43 Sri Lanka 62 38 48 52 14 76 10 67 33 Thailand 6 38 56 12 84 4 1 45 54 2 41 57

Egypt 6 26 68 14 86 5 86 10 5 12 83 Iraq 3 63 35 27 73 47 49 4 5 85 11 Lebanon 38 62 62 38 25 75 44 56 Syria 13 34 53 63 37 32 47 21 5 42 53 Turkey 10 57 33 3 52 46 31 64 5 7 59 34

Ethiopia 4 33 63 51 48 1 34 66 20 80

Ghana 67 33 35 65 10 67 24 66 33 1 Malawi 47 50 3 69 31 12 58 30 49 48 3 Nigeria 73 20 7 46 52 2 54 37 10 74 22 4 Sierra Leone 62 36 2 92 8 6 94 24 74 2 Tanzania 46 33 21 18 78 4 27 32 41 41 36 23 Zambia 70 30 78 22 59 35 7 80 20

High NaHCO3 extractable phosphorus contents of soils occur most commonly in countries with traditionally high use of phosphates where the soil P reserves have been built up during past decades. These include most European countries, New Zealand, Korea and Lebanon. Soils of low P content are most typical for Pakistan, Ghana, Iraq and India. High exchangeable (CH3COONH4) potassium contents were recorded most frequently from soils of Argentina, Egypt, Mexico and Ethiopia. Low K values again were relatively most common in Nepal, Ghana, Sierra Leone, Zambia, Nigeria, Sri Lanka and India. Unlike in the case of phosphorus, in countries with the traditionally highest K fertilizer consumption (New Zealand, Belgium, Hungary, Finland, Korea, Italy, Sri Lanka, Lebanon and Brazil) the soils are not generally higher in exchangeable K than soils in other countries. This is apparently due to the high uptake of K by plants preventing the build-up of soil K reserves. Exchangeable (CH3COONH4) calcium and magnesium contents of soils are generally 413

Table 30.Relative frequencies (percentages) of low, medium and high values of nitrogen, phosphorus, potassium, calcium and magnesium in soils of 30 countries,

Country N, % P, mg/1 K, trtg/1 Ca, mg/1 Mg, mg/1 lowmed. high.lowmed.highlowmcd.highlowmed. highlowmed. high <.083 .083 >.16<6 6-25 >25<140 140 >450 1500 190 190 >550 ,16 450 <1500 6000 >6000 550 Belgium 2 93 5 100 14 86 59 41 100 Finland 10 90 7 93 22 76 1 55 44 1 68 22 10 Hungary 4 37 59 38 62 17 77 6 6 56 38 18 60 23 Italy 3 65 33 10 51 39 31 50 19 8 50 42 38 46 17 Malta 80 20 4 96 64 36 28 72 100 New Zealand 100 5 37 58 21 42 37 26 66 8 53 11 37

Argentina 29 71 1 64 35 2 98 7 93 0 99 1

Brazil 3 39 58 27 70 3 46 51 3 79 21 49 49 1 Ecuador 13 38 50 13 67 20 7 87 7 40 60 27 40 33 Mexico 34 62 4 19 64 17 4 30 66 9 53 38 5 41 54 Peru 9 43 49 16 54 29 30 63 7 7 66 27 19 69 13

India 73 27 46 49 5 51 47 2 25 67 8 24 60 16 Korea, Rep. 11 62 27 32 67 45 55 76 24 44 56 Nepal 22 62 16 26 48 26 92 8 52 48 48 52 Pakistan 57 43 0 61 36 2 25 69 5 0 98 2 11 81 8 Philippines 14 57 29 18 60 2245 45 10 21 62 17 15 32 53 Sri Lanka 10 86 5 38 48 14 52 48 76 24 62 33 5 Thailand 1 60 39 38 55 7 33 61 6 11 43 45 18 56 26

Egypt 6 83 11 10 72 18 1 22 78 22 78 5 95 Iraq 35 60 5 50 48 2 1 83 16 25 75 42 58 Lebanon 31 69 6 31 63 13 63 25 12 88 6 63 31 Syria 26 55 18 29 50 21 3 55 42 ")1 79 3 24 74

Turkey 39 55 627 65 8 7 67 27 1 44 56 11 55 34

Ethiopia 13 26 61 45 44 11 3 31 65 25 54 21 7 57 36 Ghana 19 54 27 53 45 2 74 26 81 19 56 40 Malawi 20 69 11 9 43 48 34 58 8 56 44 54 42 4 Nigeria 57 37 7 33 50 18 55 38 7 84 14 2 70 24 6 Sierra Leone 50 50 26 68 6 70 30 -- 94 6 94 6 Tanzania 35 47 18 41 28 30 3 52 44 57 39 4 32 48 19 Zambia 62 36 2 9 65 26 63 33 4 91 9 76 24 low in African countries (except Ethiopia), Belgium, Finland, Brazil, Korea, Nepal and Sri Lanka and high in Near East countries and Mexico. Several very low Mg values indicating apparent deficiency were recorded particularly from Belgium, Zambia, Sierra Leone, Sri Lanka, New Zealand and Finland. To facilitate the comparison of the relative abundance of the six micronutrients in different countries, the frequency distributions of analytical micronutrient data in five zones are summarized in Table 31. The zone limits are based on both the plant and the soil micronutrient contents (plant micronutrient content x soil micronutrient content) and adjusted so that each of the two lowest Zones (I and II) contain 5 percent, the "normal" Zone (III) 80 percent and the two highest Zones (IV and y) 5 percent each of the whole international material (see the uppermost line in the Table). The zone limits are statistically defined and therefore do not necessarily refer to critical deficiency of toxicity limits. Especially for B and Zn, a considerable number of plant-soil analytical data falling in the lower part of Zone IH would still be suspect of deficiency. 414

Table 31. Relative distributions (percentages) of the six micronutrients in five plant x soil content zones in various countries.1)

Zone Boron Copper Iron Country ti----. I II III IV V I II III IV V I II III IV V

Whole material 3538 5 5 80 5 5 5 5 80 5 5 5 5 80 5 5

Belgium 36 92 8 94 3 3 72 17 11 Finland 90 3 1 93 1 1 6 13 80 1 ' 26 44 Hungary 201 0 1 87 9 2 99 1 97 2 1 Italy 170 92 5 3 68 10 22 1 I 86 7 5 Malta 25 72 20 8 92 8 60 8 32 New Zealand 35 97 3 17 11 49 23 43 26 31

Argentina 208 99 0 0 4 19 77 95 3 3 Brazil 58 100 48 12 40 2 91 5 2 Mexico 242 2 74 12 13 2 3 93 1 0 27 15 57 0

Peru 68 1 3 87 7 1 1 1 93 3 1 3 90 6 1

India 258 12 8 68 6 6 97 2 1 3 9 74 4 11

Korea, Rep. 90 3 13 79 2 2 1 7 89 2 1 2 92 2 3 Nepal 35 46 23 31 3 89 9 71 9 20

Pakistan 237 2 1 75 11 11 97 3 0 1 8 88 1 1 Philippines 194 30 25 45 1 35 27 38 64 16 19 Sri Lanka 18 11 89 94 6 11 83 6 Thailand 150 10 13 77 3 1 91 2 2 4 3 86 3 4

Egypt 198 94 4 2 67 26 7 1 86 11

Iraq 150 3 3 45 13 36 99 1 2 7 91 1 Lebanon 16 100 100 13 88 Syria 38 3 76 16 5 3 95 3 8 13 79

Turkey 298 I 1 84 6 8 94 5 1 14 9 74 1 1

Ethiopia 125 4 6 90 21 6 74 1 90 6 4 Ghana 93 98 2 23 24 54 1 99

Malawi 97 9 14 76 7 6 86 1 4 93 2 1

Nigeria 153 11 14 75 12 20 67 1 5 12 81 2 1 Sierra Leone 48 2 98 69 21 10 4 92 4 Tanzania 163 1 88 9 2 14 9 73 2 2 6 7 81 3 2 Zambia 44 2 11 86 36 16 48 16 11 70 2

1) Extraction methods for soil micronutrients:

B: hot water extraction + CEC correction Cu: AAAc-EDTA org. C correction Fe: AAAc-EDTA Mn: DTPA + pH correction Mo: A0-0A + pH correction Zn: DTPA

Although boron deficiency can be suspected at some locations in almost every country it seems to be relatively most common in Far East countries, Nepal, Philippines, India and Thailand as well as in Nigeria and Malawi. Problems due to excess B are most likely to arise in Iraq, Mexico, Pakistan and Turkey, often at irrigated sites. The analytical data on copper indicate that deficiency of this micronutrient is relatively common in all African countries studied but also many soils of New Zealand and some of Finland are low in available Cu. Elsewhere, shortage of Cu seems to be rare. 415

Table 31 (cont.)

Zone Manganese Molybdenum Zinc Country I II III IV V I II III IV V I II III IV V Whole material 3538 5 5 80 5 5 5 5 80 5 5 5 5 80 5 5

Belgium 36 3 83 11 3 100 17 31 53

1 11 Finland 90 1 I 94 2 1 1 94 3 71 18 Hungary 201 4 4 85 3 3 1 3 93 1 1 96 2 1 Italy 170 12 9 79 2 85 7 6 3 4 82 5 6 Malta 25 80 16 4 80 16 4 20 36 44 New Zealand 35 91 3 6 9 29 63 60 34 6

5 Argentina 208 92 5 3 98 0 1 90 4 Brazil 58 2 47 19 33 45 26 29 84 7 9 Mexico 242 6 4 87 1 2 1 1 84 6 8 2 5 86 2 4 Peru 68 1 4 89 1 4 1 3 84 3 9 87 4 9

India 258 14 11 75 2 88 6 5 11 7 81 0 I Korea, Rep. 90 8 6 66 7 14 4 10 81 2 2 78 9 13 Nepal 35 9 91 20 14 66 3 3 91 3 Pakistan 237 11 15 74 62 17 22 8 12 79 1 8 Philippines 194 1 I 80 12 7 3 9 85 1 3 1 80 11 Sri Lanka 18 72 11 17 6 94 83 11 6 Thailand 150 91 7 3 1 2 88 6 3 1 2 87 5 4

Egypt 198 8 13 79 86 13 1 2 95 1 2 Iraq 150 2 5 93 1 70 14 15 34 23 43 1 Lebanon 16 13 88 94 6 6 6 88 Syria 38 13 11 76 92 3 5 5 11 76 3 5 Turkey 298 / 5 92 0 2 92 2 4 17 18 64 1 1

Ethiopia 125 62 19 18 4 89 2 6 1 65 15 19 Ghana 93 80 12 9 16 22 62 2 95 2 1 Malawi 97 45 26 29 7 24 69 1 1 89 9 Nigeria 153 1 3 71 14 12 20 12 65 3 I 1 87 5 8 Sierra Leone 48 2 98 81 10 8 88 4 8 Tanzania 163 91 7 2 6 7 69 7 12 3 2 85 5 5 Zambia 44 30 16 55 50 9 41 2 95 2

Exceptionally high Cu values were frequently found in samples from the Philippines, Brazil, Italy and Tonga') but only occasionally elsewhere. Although the highest percentage of low (Zone I) iron values were recorded from Malta, most of the lowest single plant x soil Fe values originated in Mexico and Turkey. Indications of low Fe availability were also obtained occasionally from several other countries. Low availability of manganese is usually associated with alkaline soils and excess of Mn with acid soils. The most probable cases of Mn deficiency are therefore to be found in countries with soils of generally high pH. The highest percentage of low (Zone I) Mn values were recorded from Malta where all sampled soils had a pH of 7.48 or higher. Among other countries with a relatively high frequency of alkaline soils and high probability of Mn deficiency are India, Pakistan, Syria, Italy, Egypt and Lebanon. Indications of Mn deficiency were very rarely obtained from African countries, where

0 Tonga samples are included in New Zealand material. 416 acid soils predominate, and indeed in many of them an excess of Mn is a more likely problem as it could be in other countries with acidic soils such as Brazil, Sri Lanka and Korea. Unlike Mn, a deficiency of molybdenum seems to be most widespread in countries with acid soils such as most of the African countries, especially Sierra Leone, Zambia, Nigeria and Ghana. Low Mo values were most frequently recorded also from samples from Brazil, Nepal and New Zealand and high Mo values from samples obtained at irrigated sites in Pakistan and Iraq. According to the present analytical data deficiency of zinc can be suspected somewhere in almost every country, Belgium and Malta being the most likely exceptions. Itseems to be most widespread in Iraq, Turkey, India and Pakistan but in several other countries such as Syria, Lebanon, Mexico, Italy, Nepal, Tanzania and Thailand the data indicate some shortages of Zn. High plant Zn x soil Zn values were recorded chiefly in samples from Belgium, Malta, Korea and Ethiopia. When interpreting the nutrient data by countries, attention has been drawn to certain soil characteristics and other factors (such as texture, pH, organic matter content, CEC, fertilization, varietal differences in original indicator plants and irrigation practices) affecting the nutrient contents of soils and/or plants. It should be stressed that the limited number of soil and plant samples collected from the countries participating inthisinvestigation impose a restriction on the application of the findings. These findings cannot be regarded as being applicable to the whole agricultural areas of a country, but rather should be considered as general guidelines to draw attention to the nature of problems likely to arise and to the future work neéessary to resolve them. 417

References

ALMEIDA, A.M.A. & S'FREDO, G.J. Leaf curl and 1979 in soils and waters of Egypt. Soil Sci.Soc. Amer. 1979 dwarfing of soybean associated with manganese J. 43: 297-300. toxicity. Fitopatol., Brasil. 4,2: 333-335. FAO. Fertilizer demonstration and distribution programme, ALTEN, F., WANDROWSKY, B. & KNIPPENBERG, E. 1979 Nigeria. Project findings and recommendations, 1935 Beitrag zur Humusbestimmung. Erg. Agric. Chem. Vol, 2, Niger State: 104 p. FAO, Rome. 4: 61-69. FAO. 1979 FAO fertilizer yearbook 29: 143 p. FAO, ANDERSON, O.E. & BOSWELL, F.C. Boron and 1980aRome. 1968 manganese effects on cotton yield,lint quality, FAO. Fertilizer demonstration and pilot scheme distribu- and earliness of harvest, Agron. J. 60: 488-493, 1980btion 1978-79. The Philippines. AG: DP/PHI/ ANON. Small packs of boron for Zambian cotton. Micro- 77/009, Terminal Report: 129 p. FAO, Rome, 1981 nutrient News 1, 2: 2. FAO/UNESCO. Soil map of the world. Southeast Asia. BASCOMB, C.L. Rapid method for the determination 1976 Sheet IX, Edition 1/1976. 1964 of cation exchange capacity of calcareous and non- FIREMAN, M. & KRAUS, Y. Salinity control in irrigated calcareous soils, J. Sci.Fd.Agric. 15: 821-823. 1965 agriculture, Tahal, Israel. 46 p. BASSON, W.D., BQHMER, R.G. & STANTON, D.A. FRANCK, E.v. & FINCK. A. Evaluation of critical zinc 1969 An automated procedure for the determination of 1980 contents of oats and wheat. Z.PfLernähr. Bodenk. boron in plant tissue. Analyst 94: 1135-1141. 143(1): 38-46. BERGER,, K.C. Micronutrient shortages, Micronutrient GERICKE, S. & KURM/ES, B. Die kolorimetrische 1962 deficiencies in the United States. Agrie. Food, 1952 Phosphorsäurebestimmung mit Ammonium Vana- Chem. 10: 178-181. dat-MolybdatundihreAnwendung inder BERGER, K.C. & TRUOG, E.Borontestsand Pflanzenanalyse.Z.Pfl.ernähr.Diing.Bodenk, 1944 determination for soils and plants. Soil Sci. 57: 59: 235-247. 25-36. GERTSCH, M.E. Fertilizer demand in Ethiopia. FAO BERGMANN, W. & NEUBERT, P. Pflanzendiagnose und 1972 Joint. Proj, ETH-054-A(SIS) Subrep. Ethiopian 1976 Pflanzenanalyse. 711 p., Jena. Gov.Tech. Agency, Addis Ababa. 111 p. BOULD, C. & HEWITT, E.J. Mineral nutrition of plants GRANICK, S. Iron metabolism in animals and plants. 1963 in soils and in culture media. Plant Physiology 1958 Trace Elements (ed. C.A. Lamb et al). Academic (ed. F.S. Steward) Vol. III: 15-133. Press, New York, BRANDENBURG, E. & KORONOWSKI, P. Die nicht- GRAVEN, E.H., ATTOE, O.J. & SMITH, D. Effect of 1969 parasitarenKrankheiten,Vili.Bor.Handbuch 1965 liming and flooding on manganese toxicity in der Pflanzenkrankheiten. Verlag Paul Parey, Ber- alfalfa. Soil Sci. Soc.Amer.Proc. 29: 702-706. lin. 2,Teil: 132-170. GUPTA, U.C. Boron nutrition of crops. Adv.Agron. BROWN, A.L., QUICK, J. & EDDINGS, J.L. A 1979 31: 273-307. 1971 comparison of analytical methods for soil zinc. GUPTA, U.C. & MUNRO, D.C. The boron content of Soil.Sci.Soc. Arner.Proc. 35: 105-107. 1969 tissuesand roots of rutabagas and ofsoil CAHOON, G.A. Handbook of reference methods for soil associated with brown heart condition. Soil Sci. 1974 testing. The council of soiltesting and plant Soc.Amer.Proc. 33: 424-426. analysis. 101 p. Athens, Georgia. HAQUE, I. & BUNDU, H.S. Effects of inoculation, N, CAWSE, P.A. Deposition of trace elements from the 1980 Mo and mulch on soyabean in Sierra Leone. 1980 atmosphere in the U K, Inorganic pollution and Commun. inSoilSci. & Plant Anal.11(5): agriculture. Ministry Agric. Fish.Food (U K), 477-483. Reference Book 326; 22-46. HAQUE, I. & GODFREY-SAM-AGGREY, W. Nutri- COTTENIE, A. Soil and plant testing as a basis of 1980 tional survey of coffee and cacoa growes in Sierra 1980 fertilizer recommendation. FAO Soils Bull. 38/2: Leone. Commun. in Soil Sci. & Plant Anal. 11(5): 100 p, FAO, Rome. 485-505. DE KOCK, P.C. Iron nutrition of plants at high pH. HONG, C.W. Thefertilitystatus of Koreansoils. 1955 Soil Sci. 79: 167-175. 1972 ASPAC Food & Fert.Technol.Center,Techn. DELVER, P. Saline soils in the Lower Mesopotamian Bull. 10: 56-87. 1960 Main. Rep,Iraq, Ministry Agrie.Tech.Bull.7: JACKSON, M.L. Chemical composition of soils. Chemis- 54 p. 1964 try of the Soil (Ed. F.E. Bear). Reinhold, New DUDAL, R. Definitions of soil units for the soil map of York. 71:141. 1968 the world. World Soil Resources Rep. 33: 72 p. JOHN, M.K., CHUAH, H.H. & NEUFELD, J.H. EL-DUJAILL A. & ISMAIL, H.N. Reclamation, improve- 1975 Application of improved azomethine-H method 1971 ment and management of samt-affected and water- to determination of boron in soils and plants. logged soils in Iraq. Salinity seminar, Baghdad Anal. Lett. 8: 559-568. Dec. 5-14, 1970: 157-193. FAO, Rome. KANWAR, J.S. & RANDHAWA, N.S. Micronutrient ELONEN, P.Particle size analysis of soil. Acta Agr. 1974 Research in Soils and Plants in India. A review. 1971 Fenn. 122: 1-122, Indian Council of Agrie. Res. Tech. Bull. (Agrie.) ELSEEWI, A.A. & EMALKY, A.E. Boron distribution No. 50. 418

KORONOVVSKI, P. Die Nichtparasitären Krankheiten, RINNE, S,-L., SILLANPÄÄ, M., HUOKUNA E. & 1111- 1969 Handbuch der Pfianzenkrankheiten, 2.Tell; X. 1974a VOLA, S.-L, Effects of heavy nitrogen fertilization Mangan; 211-.246. on potassium, calcium, magnesium and phosphorus KURKI, M. Suomen peltojen viljavuuden kehityksestä. contents in ley grasses. Ann. Agric. Fenn. 13: 1979 41 p. Viljavuuspalvelu Oy, Helsinki. 96-108. LAKANEN, E. & ERVIÒ, R. A comparison of eight RINNE, S.-L., SILLANPÄÄ, M., HUOKUNA, E. & HII- 1971 extractants for the determination of plant available 1974bVOLA, S.-L. Effects of heavy nitrogen fertilization rnicronutrientsinsoils.Acta Agr.Ferm.123: oniron,manganese,sodium,zinc,copper, 223-232. strontium, molybdenum and cogalt contents in ley LINDSAY, W.L. & NORVELL, W.A. Development of a grasses. Ann. Agrie. Fenn. 13: 109-118. 1969 DTPA micronutrient soiltest. Agron.Abstr. p. SAKAL, R., SINHA, H., SINGH, A.P. & THAKUR, K.N. 84. 1979 A critical limit for the response of rice and wheat LINDSAY, W.L. & NORVELL, W.A. Development of a to applied zinc in Tarai soils. J.Agric.Sci. 93: 1978 DTPA soil test for zinc, iron, manganese, and 419-422. copper. Soil Sci Soc.Amer,..T. 42: 421-428, SEDBERRY, J.E. Jr., PETERSON, F.J., WILSON, RE., LOIZIDES, P. Report of UNSF FAO Damascus Agri- 1980 MENGEL, D.B., SCHILLING, P.E. & BRUP- 1967 cultural Research Station Project, Iron deficiency BACHER, R.H. Influence of soil reaction and and its control in the Syrian Arab Republic, 7. p. applications of zinc on yields and zinc contents of FAO, Rome. rice plants. Commun. Soil. Sci. Pl. Anal. 11 (3): MEHLICH, A. Uniformity of expressing soil test results. 283-295. 1972 A case for calculating results on a volume basis. SILLANPÄ,Ä, M. Trace elements Comm.Soil Sci.Plant Analysis. 3; 417-424, inFinnishsoilsas 1962arelated to soil texture and organic matter content, MESTANZA, S.S. & LAINEZ, C. The correction of boron J.Sci.Agric. Soc. Finland. 34: 34-40. 1970 deficiency in cacao in Ecuador. Trop. Aric. 47: 57-61. SILLANPÄÄ, M. On the effect of some soil factors on the MITCHELL, R.L. Trace element problems on Scottish 19626solubility of trace elements. Agrogeol, publ. 81: 1974 soils. Neth.J.agric.Sci. 22: 295-304. 1-24. MOKADY, R. The effect of partial pressure of oxygen in SILLANPÄÄ, M. Trace elements in soils and agriculture, 1961 field soils on lime induced chlorosis. Plant and I972aSoils Bull. 17: 67 p., FAO. Rome. Soil 15: 377-386. SILLANPÄÄ, M. Distribution of trace elements in peat MORTVEDT, J.J. & OSBORN, G. Boron concentration ,1972bprofiles. Proc. 4th Intern. Peat Congr. Otaniemi, 1965 adjacent to fertilizer granules in soil and its effect Vol 5, p. 185-191. onrootgrowth.SoilSci.Soc.Amer.Proc.29: SILLANPÄÄ, M. Unbalanced use of fertilizersas a 187-191. 1974 possible cause of soil degradation. Rep. Expert OLSEN, S.R., COLE, C.V., WATANABE, F.S. & DEAN, Consultation on Soil Degradation. FAO, Rome 1954 L.A. Estimation of available phosphorus in soils 10-14 June 1974: 12-13. by extraction with sodium bicarbonate. U.S.Dept. SILLANPÄÄ, M. Preliminary results of the World wide Agrie., Circ. 939. 1981 microelementstudy.Mineral elements'80. A OSINAME, O.A. Research aspects in soil fertility and nordic symposium on soil-plant-animal-man inter- 1976 fertilizeruseinWesternState.Rep. FAO/ relationship. Proc.Part. I: 63-73. Kemira, Hel- NORAD and Fed. Dept. Agrie. Seminar on sinki. fertilizer development in Nigeria: 24-29. FAO, SINGH, C.P., PRASAD, R.N., SINHA, H. & PRASAD, B. Rome. 1977 Evaluation of the criticallimit and extractants PANIN, M.S. & PREDTE6ENSKIJ, S.A. Zur Frage der for the determination of available zinc in cal- 1971 Cu-Dynamik im System Boden-Sommerweizen. careous soils. Beiträge zur Tropischen Landwirt- Chimija y sel'skom Choz. 9,8: 20-23. schaft u. Veterinärmedizin. 15: 131-136. PARK, C.S. & PARK, N.J. Studies on the available boron SIPPOLA, J. & ERVIÒ, R. Determination of boron in 1966 content in Korean upland soils. The Research 1977 soils and plants by the azomethine-H method. Rept. Office of Rural Development. 9: 163-174. Finn.Chem.Lett. 1977: 138-140. PEECH, M. Hydrogen-ion activity in Methods of Soil SIPPOLA, J. & TARES, T. The soluble content of mineral 1965 Analysis Part 2; C.A. Black, ed. pp. 914-926. 1978 elements in cultivated Finnish soils. Acta Agric, RANDHAWA, N.S. 8c TAKKAR, P.N. Micronutrient re- Scand., Suppl. 20: 11-25. 1975 search in India. The present status and future SIPPOLA, J., YLÄRANTA, T. & JANSSON, H. Macro- projections. Repr.Fert. News May 1975: 9 p. 1978 nutrient contents of wheat during the growing RATHORE, G.S., GUPTA, G.P., KHAMPARIA, R.S. & season. Ann.Agric.Fenn. 17: 158-162. 1978 SINHA, S.B. Response of wheat to zinc applica- SMILDE, K.W. Minor elements in the nutrition of cereals. tion in alluvial soils of Morena District, Madhya 1976 Semaine d'étude céréaliculture, Gembloux: 303- Pradesh. J.Ind.Soc.Soil Sci. 26: 58-62. 312. REISENAUER, HM., WALSH, L.M. & HOEFT, R.G. SPRAQUE, H.B. Why do plants starve? Hunger signs in 1973 Testing soils for sulphur, boron, molybdenum and 1964 crops. (ed. H.B. Sprague) p. 1-24. chlorine,p.173-200. L.M. Walsh and J.D. STANTON, RE. 8c HARDWICK, A.J. The colorimetric Beaton: Soil Testing and Plant Analysis. Soil 1967 determination of molybdenum in soils and sedi- Sci.Soc.Amer. Inc., Madison. ments by zinc dithiol. Analyst 92: 387-390. 419

TAHTINEN, H. Copper content of the soil and the VIETS, F.G.Jr. & LINDSAY, W.L. Testing soils for zinc, 1971 effectof copper fertilization.Acta Agr.Fenn. 1973 copper, manganese, and iron. p. 153-172. L.M. 123: 136-142. VVaish and J.D. Beaton: Soil testing and plant TAINT°, A. Maan kuparipitoisuus ja kuparilannoituksen analysis. Soil Sci,Soc.Amcr.Inc., Madison. 1960 vaikutus. Koetoirn. ja käyt. 17: 3. WADSWORTH, G,A. & WEBBER, J. Deposition of TAKKAR, P.N., MANN, M.S. & RANDHAWA, N.S. 1980 minerals and trace elements in rainfall, Inorganic 1974 A note on relationship of zinc contents in wheat pollution and agriculture. Ministry Agric.Fish. plants exhibiting different degree of zinc deficiency Food (U K), Reference Book 326: 47-55. symptoms with wheat yield. /ndian J.Agric,Sci. WATANABE, F.S. & OLSEN, S.R. Test of an ascorbic (ref. Randhawa, N.S., Takkar, P.N. 1975 Fert. 1965 acidmethodfordeterminingphosphorusin News, May 1975). water and NaHCOA extracts from soil. Soil Sci. TAKKAR, P.N. & MANN, M.S. Toxic levels of soil and SocAmer.Proc. 29: 677-678. 1978 p/ant zinc for maize and wheat. Plant and Soil WEJR, R.G, & MIEHAM, P.J. Use of plant analysis to 49: 667-669. 1978 asses zinc status of rnaize seedling. N.Z. Dept,Sci. TAMM, O. Eine Methode zur Bestimmung der an- Ind,Res. 134: 547-552. 1922 organischen Kornponenten des Gelkomplexes im WHITNEY, D.A. Micronutrient soil tests zinc, iron, Boden. Medd.Stat. Skogsförsöksanst.19: 387 1980 manganese, and copper. Recommended chemical 404. soil test procedures for the North Central Region. TARES, T. & SIPPOLA, J. Changes in pH, in electrical Agricultural Exp.Sta, & U.S.Dept.Agric.Bull. 499: 1978 conductivity and in the extractable amounts of 18-21, mineral elements in soil, and the utilization and WHITNEY, DA., ELLIS, R. Jr., MURPHY, L.S. & losses of the elements in some field experiments. 1973 HERRON, G. Identifying and correcting zinc and Acta Agric.Scand.,SuPpl.20: 90-113. iron deficiency in field crops. Kansas Coop. Ext. TC1LGYESI, G. & MIKO, Z. T. (Correlations between Ser. Leaflet L-360, Manhattan, KS. 1977 yield and mineral uptake in maize). Eng. summary. YLARANTA, T., JANSSON, H. & SIPPOLA, J. Seasonal Növénytermelés 26, 2/3: 169-175. 1979 variation in rnicronutrient contents of wheat. Ann. TOLLENAR, D. Boron deficiency in cacao, bananas, and Agric.Fenn. 18: 218-224, 1966 other crops on volcanic soils of Ecuador. Neth. YLÄRANTA, T. & SILLANPÄÄ, M. (Unpublished data J.Agric.Sci, 14: 138-151. on micronutrient contents of various plant species). TRIERWEILER,J.F.& LINDSAY, W.L.EDTA- Institute of Soil Science, Jokioinen, Finland. 1969 ammonium carbonate soil test for zinc. Soil Sci. Soc.Amer.Proc. 33: 49-54, 421

APPENDIX 1. Instructions (condensed) for taking plant and soil samples.

Selection of sampling sites: The main wheat and maize growing areas of the country s'hould be included in such a way that samples from the major soil types will be represented. Areas where deficiencies of one or more trace elements are suspected or known to exist should be included in proportion to the total area to be sampled. To minimize the risk of contamination, sampling sites close to roads or other dusty places must not be chosen. The locations of sampling sites must be indicated on a suitabie large scale working map in such a way that the exact sites can later be relocated easily. From these large scale working maps the sampling sites and numbers should be transferred to srnall scale maps covering the whole country.

Plant sampling: To avoid contamination by soil, plant samples must always be taken with clean hands before taking soil samples. Size of plant sample: not less than 350 g fresh material collected from various parts of the sampling plot of about 10 X 10m. Timing: Wheat samples should be taken at mid-tillering stage of the plant growth (i.e. for spring wheat about 35-40 days after planting depending on weather conditions, for winter wheat the corresponding stage of growth). Maize samples at 5-6 leaf stage (i.e. approximately 35-40 days after planting). Plant part to be sampled: Wheat the upper half of the plant (use scissors). Maize two uppermost fully expanded leaf blades of each plant. Equipment: Clean hands, clean scissors, clean knife (rusted equipment should not be used). Bags: Paper bags, which will be made available, shou/d be used when collecting the plant samples in the field and during transportation from field to a laboratory (or to any clean dust-free room) for drying. Every effort should be made to avoid contamination. Drying: All plant samples must be air-dried in a dust-free room, or even oven-dried at 60-65 °C. Milling: No milling should be done.

Soil sampling: Equipment: Stainless steel soil bores (which will be made available) only should be used to avoid trace element contamination due to equipment. Sampling depth: Full plough layer (appr. 0-20 cm). Size of the soil sample: 0.6 dm3, i.e. full pre-numbered carton box composed of a minimum of 6-10 subsamples collected from various parts of the plant-soil sampling plot of about 10 X 10 m. Drying: Soil samples must be air-dried in a dust-free area. Contamination: Every effort should be made to avoid contamination at all stages.

440

APPENDIX 5.Indicative data on fertilizer application to the original indicator crops in different countries1).

Number of Nitrogen Phosphorus Potassium sites kg N/ha kg P/ha kg K/ha

Country Wheat Maize Wheat Maize Wheat Maize Wheat Maize fields fields fields fields fields fields fields fields Belgium 21 20 98 ± 21 118 ± 66 42 ± 9 45 ± 26 130 ± 29 122 ± 90 Finland 94 76 ± 24 43 ± 26 63 ± 19 Hungary 144 106 140 ± 74 136 ± 58 52 ± 55 52 ± 33 98 ± 96 104 ± 56 Italy 118 70 105 ± 66 186 ± 127 42 ± 29 51 ± 44 29 ± 31 80 ± 98 Malta 25 60 ± 0 26 ± 0 83 ± 0 New Zealand 14 24 9 ± 16 55 ± 67 16 ± 7 30 ± 29 7 ± 12 44 ± 25

Argentina 119 90 3 ± 14 6 ± 16 0 ± 0 6 ± 18 0± 0 0± 0 Brazil 71 19 ± 11 49 ± 30 33 ± 18 Ecuador 16 44 ± 13 23 ± 8 13 ± 6 Mexico 100 147 122 ± 54 40 ± 49 9 ± 13 4± 7 1 + 5 0± 1 Peru 13 57 38 ± 35 46 ± 58 10 ± 13 8 ± 13 12 ± 19 9 ± 17

India 188 107 75 ± 51 106 ± 57 15 ± 15 18 ± 12 11 ± 20 9 ± 19 Korea, Rep. 50 50 100 ± 25 107 ± 31 31 ± 7 35 ± 14 58 ± 10 77 ± 80 Nepal 50 68 ± 70 15 ± 8 16 ± 20 Pakistan 156 86 58 ± 57 83 ± 90 9 ± 15 12 ± 29 0 + 3 2 ± 8 Philippines 197 8 ± 19 1 ± 4 I + 4 Sri Lanka 21 24 ± 26 10 ± 9 14 ± 12 Thailand 150 0 ± 0 0 ± 0 0 ± 0

Egypt 100 100 129 ± 31 161 ± 39 9 ± 7 2 + 6 0 + 0 0 ± 0 Iraq 119 31 5 + 9 27 ± 24 2 ± 4 10 ± 10 1 ± 3 0 ± I Lebanon 16 53 ± 99 33 ± 54 0 + 0 Syria 20 18 52 ± 59 117 ± 87 7 ± 11 7 ± 12 0 ±0 0 ± 0 Turkey 249 50 34 ± 30 61 ± 50 18 ± 13 6 + II 0 ± I 0 ± 0

Ethiopia 54 71 11 + 10 7 ± 11 10 ± 10 5 ± 7 0 + 0 0 ± 0 Ghana 93 48 ± 10 11 ± 11 14 ± 12 Malawi 100 81 ± 63 17 + 14 II ± 26 Nigeria 42 103 85 ± 27 13 ± 23 17 ± 8 4 ± 6 4 ± 11 6 ± 10 Sierra Leone 49 5 ± 22 2 + 7 3 ± 15 Tanzania 5 175 16 ± 36 8 ± 17 2 ± 4 2 + 6 0 +'0 0 ± 0 Zambia 45 190 ± 84 35 + 35 25 ± 29

Internat. average 1768 1976 66 ± 61 27 ± 37 21 ± 27 6 ± 10 21 ± 44 7 ± 18

I) The data are based on information given by co-operators and sample collectors in "Field Information Forms". Because of several missing figures the data are not complete and, therefore, must be considered only as indicative. It should be pointed out that in many cases the samples come from fertilizer trial fields and therefore, the indicated fertilization level may be considerable higher than elsewhere in the country. 441

APPENDIX 6. Average general soil properties and macronutrient contents of soils classified by FAO/Unesco soil units. For analytical methods, see Section 1.3.

Texture El. CaCO3CEC Org. N P K Ca Mg Na indexpH(CaCl2)cond. equiv. me C total (extractable) 10-4 s 100 g Tr. mg/1 mg/I mg/I mg/1 mg/I

Acrisols 42 5.19 1.2 0.0 19.4 1.1 0.122 15.0 156 1208 238 14 (n = 73) s 16 0.49 0.9 0.0 11.1 0.7 0.059 14.7 97 653 174 15

Andosols 33 5.26 1.9 0,4 35.5 2.6 0.297 16.7 388 1581 111 14 (n =-- 8) s 14 0.63 0.9 0.7 13.0 1.9 0.157 6.8 226 803 94 13

Arenosols 2 19 5.13 1.6 0.2 13.7 1.1 0 107 82.2 163 793 83 13 (n 22) s 8 0.83 1.0 0.2 8.1 0.6 0.053 58.8 89 682 86 11

Cambisols 2 40 6.41 1.8 2.1 22.2 1.8 0.159 29.5 179 2549 271 46 (n =- 246) s 13 1.27 1.0 5.7 12.3 2.2 0.136 29.4 104 1688 186 77

Chernozems 2 47 7.20 2.3 5,6 28.1 1.6 0.183 30.4 174 5350 289 16 (n .= 48) s 7 0.70 0.7 5,1 4.3 0.4 0.044 15.4 49 1609 171 18

Ferraiso Is x 32 5.26 0.9 0.0 14.2 1.3 0.116 12.6 112 813 136 9 (n 127) s 13 0.74 0.5 0.1 5.3 0.5 0.049 18.2 72 632 89 7

Fluvisols Tc 51 7.44 5.2 7.6 33.4 1.1 0.123 8.1 475 5859 904 397 (n 470) s 16 0,68 6.8 9.6 13.9 0.5 0.073 21.2 327 2539 613 543

Gleysols 2 36 5.62 2.1 0.8 21.3 1.7 0.179 47.2 206 1600 269 17 (n = 31) s 14 1.00 1.1 2.3 12.7 1.2 0.116 40.8 131 1504 276 17

Halosols x 50 7.78 12.3 21.0 23.7 0.8 0.097 10.7 265 6017 780 713 (n 60) s 9 0.29 11.8 11.2 6.0 0.3 0.036 13.1 142 1964 336 767

Histosols 4.54 2.2 0.0 82.0 20.3 0.968 44.2 151 2079 320 22 (1)= - 7) s - 0.51 0.7 0.0 15.3 9,6 0.412 13.7 54 1068 243 16

Kastanonms x 52 7.43 3.2 11.2 34.7 0.9 0.105 12.3 530 7053 674 135 ( n 64) s 15 0.50 3.6 16.5 15.9 0.3 0.032 12.4 267 3535 393 214

Lithosols x 40 7.25 1.3 10.4 26.6 0.8 0,088 8.2 356 5786 305 8 (n = 23) s 18 0.64 0.4 11.6 12.7 0.4 0.046 4.4 158 3231 155 3

Luvisols x 43 6.79 1.9 7.7 24.1 1.0 0.105 24,0 244 4127 404 53 (n = 217) s 14 0.98 1.3 12.6 12.4 0.5 0.047 29.4 186 2814 382 115

Nitosois x 28 5.60 0,9 0.1 13.9 1.1 0.104 14.9 152 1008 187 5 (n =-- 106) s 9 0.85 0.6 0.1 6.8 0.5 0.054 17.6 151 687 168 6

Phaeozems 2 44 6.08 1.8 1.3 29.1 1.9 0.198 26.4 581 3280 355 34 (n ----- 307) s 9 0.88 1.6 3.4 8.1 0.7 0.067 21.4 324 1893 178 112

Planosols x 52 6.32 3,3 0.6 37.3 2.2 0.210 24.6 545 4220 625 74 ( n = 117) s 14 1.07 2.2 1.0 8.6 0.8 0.087 12.2 277 2107 343 36

Podzols x 23 5.23 1.9 0.0 24.0 3.1 0.221 80.3 182 1460 87 14 (n 1!) s 8 0.46 0.8 0.0 7.1 1.3 0.079 40.2 91 1619 34 16

Regosols x 42 6.62 2.6 5.0 21.8 1.1 0.114 33.8 348 4037 361 151 (n = 42) s 18 0.97 2.9 10.5 11.9 0.5 0.047 31,2 283 3524 363 484

Rendzinas x 60 7.40 1.6 12,4 57.8 1.8 0.168 11.9 264 10730 573 22 (n = 48) s 14 0.29 0.4 14.3 11.9 0.5 0.053 10-3 169 2869 315 25

Verusols X' 58 ' 7.16 2.8 8.9 41.1 1.1 0.112 12.0 352 7256 681 104 (n = 135) s 12 0.72 5.3 12.1 14.4 0.6 0_059 14.2 237 2854 451 278

Yermosols 2 44 7.83 6.2 6.1 20.3 0.6 0.082 7.3 438 5268 480 447 ( n 92) s 10 0.24 9.7 4.3 10,6 0.2 0.020 7.8 307 1770 270 591

Xerosols x 48 7,62 4.4 12.4 29.0 0.8 0.096 13.5 601 6207 659 227 (n = 101) s 13 0.27 5.9 13.7 10.5 0.3 0,032 15.1 344 2300 362 301 442

APPENDIX 7. Average mieronutrient (B, Cu, Fe, Mn, Mo and Zn) contents of soils and respective pot- grown wheats for FAO/Unesco soil units: Acr Acrisol (n 73), And = Andosol (8), Are = Arenoso! (22), Cam = Cambisol (246), Che = Chernozem (48), Fer Ferralsol (127), Flu = Fluvisol (470), Gle = Gleysol (31), Ha! -= Halosol (60), His = Histosol (7), Kas = Kastanozem (64), Lit -= Lithosol (23), Luv .--- Luvisol (217), Nit = Nitosol (106), Pha = Phaeozem (307), Pla = Planosol (11), Pod = Podzol (27), Reg = Regosol (42), Ren = Rendzina (48), Ver = Vertisol (135), Yer = Yermosol (92), Xer Xerosol (101). See also Section 2.8.

100 80- 60 MkIM111110111111111 50 n=3538 40 E ca. 30 Rts,_ 0.7 0_ I 11)Y=8:.5°P°111111

MMOMME=MMENNIMPT,I Ilihin II II *8 8 `11,11=MMENIIIIMleik=MMErN11-04.4KE 111111111M,=RTEEM11511iMMUmEME -C66IIMEWW=MIRPagaMIN= NUM 54- o 4 =MIIÏIIIUIIIlI 3 iIIIIH ining 111111! 081 .2.3 .4.5.6 .810 2 3 4 5 6 8 10 CEC-corrected B in soil, mg /l (Hot w.sol.)

20 18 MIMII I1iIIIilIiIUINEI III MI A I le 111111 I 1 ii MBE IIIINFIRMIN 4 11.11641111im

2 PirFIMIII Ell . o uu h I' IIIIIII_ - .4.6 .8 1.0 2 46 810 20 40 60100 Org.0 corrected soi CU,mg/I (AAAc-EDTA; 443 300 275 ranMiliiiiiiM11111 0 250 Noll i nllEINII1111111111111 225mimingrIaapielIin II 200=11111E111 1-A20111111 I 1 75MINIM !i gYVEIMITaiiII I E 1 50IIII IN IllIIIll 1 cL INI IIIII 11liIMO

1 00iiiii iiiiII I 90INIII1111111111111111 NMI 11 80MEN E11111111111111111111

-4E 70 MIIIIIIII MIIIIILINTII o °60 1 .-F-4,F) )liiiIiii u_ .lerche';'. LlYrarn ow 1 50 11...4 Ad 40mils ill

30 IN iiiill I1111111111.111 In MIRA111 10 20 40 60 100200400500 1000 2000 Fe in soil. mg/I (AAAc-EDTA)

2000 motutimamsmotrimmosis Nimairsamis 1 5n . NEI 1000 8001mommiummwsocnrimirminwamimomma1: ImAigretr'm111111MONms. 01111111111111111 600milIMMONIMEASINEMI.rikNI NENAW 400 woomml Lam= 300 .1...11 minnimurlip 200I9 Ill MEN ' Ill 01in4'd 100 illi hiIN 80 .1=====i 1- Lt:J°pirmon anElbas421 50 ....2111flp --W-44, Mailai ririni11 Immirst l'r IRRIMIIIIIIIIIMIIIIIIM111111111=11,1 40 M. ElPtillIMIIII III Mil1 30 ofiEll II Mil v 20 OMNI IIII 1111

10 Ill 111 1111 8 IIMITIar=111171115P1 1 M itaukummi ainuAN 1 2 3 46 810 20 3040 60inn200 300 Rno pH- corrected Mn in soil, mg/l (DTPA) M o MN. =MIME aMNIMMIIIINIMMMEIntII IIIIIIMIMMINNIE11111111111111 6 MNIOSIGEERENIMINIIMIIIIMINI=111= MINIM s - 0ES31312111111 MIN 3 Emu 2 111111k 1111111111 11111

iiiiiiiiiiiiiiirniIMENIMMEME 1Mio.MIMIIMIIPIMEN MNIMIEINMI=MMEM IMMENIMITTriumrimmmmmr rmemil 1.4111 11111111 IIIIIMMAINIK411111 EMI 111111IL111511"1"11cheA111111111 airaraammir2MWMINIONNIViIMMEMMIZIMMIN11IMIM .0651111117IFAIMMIIIMMI:21IMENUMMMEMMEMMENMM. NONNIMINIMINI IMME .03uiiuiivui MENEM MI .02111 "1111111111111111 NMI

.01 111111111611MININIK 11111.111111=MI=IM I .006 .01 .02 .03 06 .1 .2.3 .6 1 2 3 pH - corrected MO in soil, mg/t (A0-0A)

200

100liii !III,11111111111111111 90 , iimiiNiiuui 80 70 IMMIIIMIIIIIIMM111111112M1 E60 fl" 50 1111l11111111M1111111111111= o 40 IM1111111111111/11111111M1

30 11=1111111311111111111111

"g 20

C 'Flu R N Ren 1011111114=1MO1== 111=113 101111111Milli 9 r NEE 811111)1MM1=MMIIMMEIMMIUMMINI 7III 6MI MIMIi___uiii.isuiuiuiii 1111111M1111 III111111= 111111111 .1 .2.3 .4.6 .8 1 2 3 46 810 20 3040 60 100 Zlel in soit,mg/t (DTPA) FAO SALES AGENTS AND BOOKSELLERS

Algeria Société nationale d'édition et de diffusion, 92, rue Didouche Mourad, Algiers. Argentina Editorial Hemisferio Sur S.A., Librería Agropecuaria, Pasteur 743, 1028 Buenos Aires. Australia Hunter Publications, 58A Gipps Street, Collingwood, Vic. 3066; Australian Government Publishing Service, P.O. Box 84, Canberra, A.C.T. 2600; and Australian Government Service Bookshops at 12 Pire Street, Adelaiden S.A.; 70 Alinga Street, Canberra, A.C.T.; 162 Macquarie Street, Hobart, Tas.; 347 Swanson Street, Melbourne, Vic.; 200 St. Georges Terrace, Perth, W.A.; 309 Pitt Street, Sydney, N.S.W.; 294 Adelaide Street, Brisbane, Old. Austria Gerold & Co.. Buchhandlung und Verlag, Graben 31, 1011 Vienna. Bangladesh ADAB, 79 Road 11A, P.O. Box 5045, Dhanmondi, Dacca. Belgium Service des publications de la FAO, M.J. de Lannoy, 202, avenue du Roi, 1060 Brussels. CCP 000-0808993-13. Bolivia Los Amigos del Libro, Perú 3712, Casilla 450, Cochabamba; Mercado 1315, La Paz; René Moreno 26, Santa Cruz; Junín esq. 6 de Octubre, Oruro. Brazil Livraria Mestre Jou, Rua Guaipà 518, Sào Paulo 05089; Rua Senador Dantas 19-52051206, 20.031 Rio de Janeiro, PRODIL, Promoçáo e Dist. de Livros Ltda., Av. Venáncio Aires 196, Caixa Postal 4005, 90.000 Porto Alegre; A NOSSA LIVRARIA, CLS 104, Bloco C, Lojas 18/19, 70.000 Brasilia, D.F. Brunei SST Trading Sdn. Bhd., Bangunan Tekno No. 385, JIn 5159, P.O. Box 227, Petaling Jaya, Selangor. Canada Renouf Publishing Co. Ltd, 2182 St Catherine West, Montreal, Que. H3H 1M7. Chile Tecnolibro S.A., Merced 753, entrepiso 15, Santiago. China China National Publications Import Corporation, P.O. Box 88, Beijing. Colombia Editorial Blume de Colombia Ltda., Calle 65 N° 16-65, Apartado Aéreo 51340, Bogotá D.E. Costa Rica Librería, Imprenta y Litografía Lehmann S.A., Apartado 10011, San José. Cuba Empresá de Comercio Exterior de Publicaciones. O'Reilly 407 Bajos entre Aguacate y Compostela, Havana. Cyprus MAM, P.O. Box 1722, Nicosia. Czechoslovakia ARTIA, Ve Smeckach 30, P.O. Box 790, 111 27 Praha 1. Denmark Munksgaard Export and Subscription Service, 35 Norre Sogade, DK 1370 Copenhagen K. Dominican Rep. Fundación Dominicana de Desarrollo, Casa de las Gárgolas, Mercedes 4, Apartado 857, Zona Postal 1, Santo Domingo. Ecuador Su Librería Cía. Ltda., García Moreno 1172 y Mejía, Apartado 2556, Quito; Chimborazo 416, Apartado 3565, Guayaquil. El Salvador Librería Cultural Salvadoreña S.A. de CV., Calle Arce 423, Apartado Postal 2296, San Salvador. Finland Akateeminen Kirjakauppa, 1 Keskuskatu, P.O. Box 128, 00101 Helsinki 10. France Editions A. Pedone, 13, rue Soufflot, 75005 Paris. Germany, F.R. Alexander Horn Internationale Buchhandlung, Spiegelgasse 9, Postfach 3340, 6200 Wiesbaden. Ghana Fides Enterprises, P.O. Box 14129, Accra; Ghana Publishing Corporation, P.O. Box 3632, Accra. Greece G.C. Eleftheroudakis S.A., International Bookstore, 4 Nikis Street, Athens (T-126); John Mihalopoulos & Son S.A.. International Booksellers, 75 Hermou Street, P.O. Box 73, Thessaloniki. Guatemala Distribuciones Culturales y Técnicas "Artemis", 5a. Avenida 12-11, Zona 1, Apartado Postal 2923, Guatemala. Guinea-Bissau Conselho Nacional da Cultura, Avenida da Unidade Africana, C.P. 294, Bissau. Guyana Guyana National Trading Corporation Ltd, 45-47 Water Street, P.O. Box 308, Georgetown. Haiti Librairie "A la Caravelle", 26, rue Bonne Foi, B.P. 111, Port-au-Prince. Hong Kong Swindon Book Co., 13-15 Lock Road, Kowloon. Hungary Kultura, P.O. Box 149, 1389 Budapest 62. Iceland Snaebjörn Jónsson and Co. h.f., Hafnarstraeti 9, P.O. Box 1131, 101 Reykjavik. India Oxford Book and Stationery Co., Scindia House, New Delhi 110001, 17 Park Street, Calcutta 700016. Indonesia P.T. Sari Agung, 94 Kebon Sirih, P.O. Box 411, Djakarta. Iraq National House for Publishing. Distributing and Advertising, Jamhuria Street, Baghdad. Ireland The Controller, Stationery Office, Dublin 4. Italy Distribution and Sales Section, Food and Agriculture Organization of the United Nations, Via delle Terme di Caracalla, 00100 Rome; Libreria Scientifica Dott. Lucio de Biasio "Aeiou", Via Meravigli 16, 20123 Milan; Librería Commissionaria Sansoni S.p.A. "Licosa", Via Lamarmora 45, C.P. 552, 50121 Florence. Japan Maruzen Company Ltd, P.O. Box 5050, Tokyo International 100-31. Kenya Text Book Centre Ltd, Kijabe Street, P.O. Box 47540, Nairobi. Kuwait Saeed & Samir Bookstore Co. Ltd, P.O. Box 5445, Kuwait. Luxembourg Service des publications de la FAO, M.J. de Lannoy, 202. avenue du Roi, 1060 Brussels (Belgium). Malaysia SST Trading Sdn. Bhd., Bangunan Tekno No. 385, JIn 5159, P.O. Box 227, Petaling Jaya, Selangor. Mauritius Nalanda Company Limited, 30 Bourbon Street, Port Louis. Mexico Dilitsa S.A.: Puebla 182-D. Apartado 24-448, Mexico 7, D.F. Morocco Librairie "Aux Belles Images", 281, avenue Mohammed V, Rabat. Netherlands Keesing Boeken V.B.. Joan Muyskenweg 22, 1096 CJ Amsterdam. New Zealand Government Printing Office. Government Printing Office Bookshops: Retail Bookshop, 25 Rutland Street, Mail Orders, 85 Beach Road, Private Bag C.P.O., Auckland; Retail, Ward Street, Mail Orders, P.O. Box 857, Hamilton; Retail, Mu!grave Street (Head Office), Cubacade World Trade Centre, Mail Orders, Private Bag, Wellington; Retail, 159 Hereford Street, Mail Orders, Private Bag, Christchurch; Retail, Princes Street, Mail Orders, P.O. Box 1104, Dunedin. Nigeria University Bookshop (Nigeria) Limited, University of lbadan, lbadan. Norway Johan Grundt Tanum Bokhandel, Karl Johansgate 41-43, P.O. Box 1177 Sentrum, Oslo 1. Pakistan Mirza Book Agency, 65 Shahrah-e-Quaid-e-Azam, P.O. Box 729, Lahore 3. Panama Distribuidora Lewis S.A., Edificio Dorasol, Calle 25 y Avenida Balboa. Apartado 1634, Panama 1. Paraguay Agencia de Librerías Nizza S.A., Tacuarí 144, Asunción. Peru Librería Distribuidora "Santa Rosa", Jirón Apurimac 375, Casilla 4937, Lima 1. Philippines The Modern Book Company Inc., 922 Rizal Avenue, P.O. Box 632, Manila. Poland Ars PoIona, Krakowskie Przedmiescie 7, 00-068 Warsaw. Portugal Livraria Bertrand, S.A.R.L., Rua Joño de Deus, Venda Nova, Apartado 37, 2701 Amadora Codex; Livraria Portugal, Dias y Andrade Ltda., Rua do Carmo 70-74, Apartado 2681, 1117 Lisbon Codex; Edioóes !TAU, Avda. da República 46/A-r/c Esqdo., Lisbon 1. Korea, Rep. of Eul-Yoo Publishing Co. Ltd, 46-1 Susong-Dong, Jongro-Gu, P.O. Box Kwang-Wha-Moon 362, Seoul. Romania Ilexim, Calea Grivitei N° 64-66, B.P. 2001, Bucharest. Saudi Arabia The Modern Commercial University, P.O. Box 394, Riyadh. Sierra Leone Provincial Enterprises, 26 Garrison Street, P.O. Box 1228, Freetown. Singapore MPH Distributors (S) Pte. Ltd, 71177 Stamford Road, Singapore 6; Select Books Píe. Ltd, 215 Tanglin Shopping Centre, 19 Tanglin Road, Singa- pore 1024; SST Trading Sdn. Bhd., Bangunan Tekno No. 385, Jln 5/59, P.O. Box 227, Petaling Jaya, Selangor. Somalia "Samater's", P.O. Box 936, Mogadishu. Spain Mundi Prensa Libros S.A., Castello 37, Madrid 1; Librería Agricola, Fernando VI 2, Madrid 4. Sri Lanka M.D. Gunasena & Co. Ltd, 217 Olcott Mawatha, P.O. Box 246, Colombo 11. Sudan University Bookshop, University of Khartoum, P.O. Box 321, Khartoum. Suriname VACO n.v. in Suriname, Dominee Straat 26, P.O. Box 1841, Paramaribo. Sweden C.E. Fritzes Kungl. Hovbokhandel, Regeringsgatan 12, P.O. Box 16356, 103 27 Stockholm. Switzerland Librairie Payot S.A., Lausanne et Genève; Buchhandlung und Antiquariat Heinimann & Co., Kirchgasse 17, 8001 Zurich. Thailand Suksapan Panit, Mansion 9, Rajadamnern Avenue, Bangkok. Togo Librairie du Bon Pasteur, B.P. 1164, Lomé. Tunisia Société tunisienne de diffusion, 5, avenue de Carthage, Tunis. United Kingdom Her Majesty's Stationery Office, 49 High Holborn, London WC1V 61-IB (callers only); P.O. Box 569, London SE1 9NH (trade and London area mail o-ders); 13e Castle Street, Edinburgh EH2 3AR; 41 The Hayes, Cardiff CFI 1JW; 80 Chichester Street, Belfast BT1 4JY; Brazennose Street, Manchester M60 8AS; 258 Broad Street, Birmingham B1 2HE; Southey House, Wine Street, Bristol BSI 2B0. Tanzania, United Rep. of Dar es-Salaam Bookshop, P.O. Box 9030, Dar es-Salaam; Bookshop, University of Dar es-Salaam, P.O. Box 893, Morogoro. United States of America UNIPUB, 345 Park Avenue South, New York, N.Y. 10010. Uruguay Libreria Agropecuaria S.R.L., Alza/bar 1328, Casilla de Correos 1755, Montevideo. Venezuela Blume Distribuidora S.A., Gran Avenida de Sabana Grande, Residencias Caroni, Local 5. Apartado 50.339, 1050-A Caracas. Yugoslavia Jugoslovenska Knjiga, Trg. Republike 5/8, P.O. Box 36, 11001 Belgrade; Cankarjeva Zalozba, P.O. Box 201-IV, 61001 Ljubljana; Prosveta, Terazije 16, P.O. Box 555, 11001 Belgrade. Zambia Kingstons (Zambia) Ltd, Kingstons Building, President Avenue, P.O. Box 139, Ndola. Other countries Requests from countries where sales agents have not yet been appointed may be sent to: Distribution and Sales Section, Food and Agriculture Organization of the United Nations, Via delle Terme di Caracalla, 00100 Rome, Italy.

October 1981 M-52 ISBN 02+101103-1