Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. THE INFLUENCE OF PHOSPHORUS FERTILISER FORMS AND RHIZOSPHERE PROCESSES ON THE PHOSPHORUS NUTRITION OF TEA (Camellia sinensis L.) A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Soil Science at Massey University New Zealand A. Kapila N. Zoysa 1997 To the Memory of My late Parents For their Enormous love, care and guidance ii ABSTRACT The understanding of the phosphorus (P) nutrition of tea, has mainly d erived from trials evaluating yield response to applied P fertilisers. The literature indicates that the fertiliser P requirement of tea is generally low (below 1 5 kg ha- 1 y(l), inspite of the generally high P fixing capacity of the U Itisols used for growing tea. Very little published information is available on the reactions of P fertilisers in tea soils and on the chemistry of P in the tea rhizosphere to explain this low P requirement of tea. Because the tea soils are highly acidic (4 . 5 - 5 . 5) a locally mined, low cost , sparingly soluble phosphate rock (Eppawala phosphate rock, EPR) has been recommended as a P fertiliser for tea in Sri Lanka. But there is no experimental information available on its suitability for tea when compared to soluble P fertilisers. The main obj ective of this thesis is to study the mechanisms involved in P utilisation from the tea rhizosphere, when both soluble (triple superphosphate) and sparingly soluble EPR fertilisers are used . An existing technique used to study rhizosphere processes of annual crops was modified to study the chemical processes involved in P utilisation from the rhizosphere of camellia plants, which are of the same family as tea. The depletion of soil and fertiliser P in slices of soil away from the rhizoplane were measured using a sequential chemical P fractionation procedure. The technique aII owed isolation of soil slices at increasing distances from the rhizoplane and characterisation of the depletion pattern of soil P forms in the camellia rhizosphere. Subsequently this technique was used to study the rhizosphere processes in tea and other crops normally grown in association with tea. A glasshouse study conducted to compare the mechanisms of P utilisation of tea (clone TRI 2025) with calliandra, Guinea grass and beans showed that all species depleted resin-P and NaOH-Pj in their rhizospheres. In contrast to other species, tea accumulated NaOH-P 0 (organic-P) in the rhizosphere. All plant species acidified their rhizospheres and the magnitude of acidification is in the order of Guinea grass > bean and tea > calliandra. The higher acidification in the rhizosphere compared t o the bulk soil caused more EPR dissolution near the roots. .iii Another glasshouse trial which examined the P utilisation efficiencies of tea clones showed that TRl 2023 and TRl 2025 had a higher external P efficiency than S 106 due to g reater root surface area and P uptake per unit root surface area. But the internal efficiencies were not significantly different between the clones. All tea clones acidified the rhizosphere and the magnitude of acidification is of the order : TRl 2023 > TRl 2025 > S 106. The dissolution of EPR in the rhizosphere also followed the same order. All three clones accumulated NaOH-Po in the rhizosphere. Rhizosphere pH of tea (clone TRl 2025) decreased compared to the bulk soil, when N was supplied as the NII/ [(1\%)2S04] or the NH/ + N03- �N03] form and it increased when N was supplied as the N03" [Ca(N03)2] form. The �)2S04 treatment caused the highest dissolution of EPR in the rhizosphere, whereas the Ca(N03)2 treatment showed the lowest in accordance with the magnitude of pH decline. C ation-anion balance in the plants showed that whatever form of N was applied, plants utilised more N 03- than N& +. High nitrification rates in the rhizosphere were probably responsible for this inspite of the addition of a nitrification inhibitor. A glasshouse trial with young tea plants (TRl 3072) showed that the agronomic effectiveness of the sparingly soluble EPR was equal to or better than the readily soluble TSP (triple superphosphate) fertiliser. This was due to the high rate of EPR dissolution in the acid soil . About 75% of the applied EPR was dissolved in the soil during the 10 month period of the study. The results also showed that the borax soil P test used to predict the P requirement of tea, as currently used in Sri Lanka, was the best of the six soil P tests investigated. This test has the advantage of requiring only one calibration curve relating yield and soil P values in estates fertilised with both soluble and sparingly soluble PR fertilisers . This thesis contributed new knowledge regarding P uptake processes In the rhizosphere of tea plants . i v ACKNOWLEDGEMENTS I am very pleased to appreciate the following people and the institutions for their contributions towards completing this thesis and my student life at Massey U niversity. My chief academic supervisor Dr P Loganathan for his worthy guidance, helpfulness, generosity and friendship throughout my life at Massey U niversity. More importantly his constructive criticism, interest and patience at all stages of my thesis research and preparation. I admire his contribution. Associate Professor M J Hedley, my second supervisor for his warm friendship and expert advise at all stages of my thesis research. My third supervisor Dr S L Amarasiri (Director General - Department of Agriculture, Sri Lanka) for his encouragement and support during my study period. I am deeply indebted to Dr P Sivapalan (Ex. Director - Tea Research Institute of Sri Lanka) for the assistance he rendered for me to pursue my studies at Massey U niversity. Professor R W Tillman (Head - Soil Science Department, Massey U niversity) and Associate Professor Paul Gregg for their help and encouragement at difficult times. All fellow postgraduate students at the Soil Science Department of Massey for providing such a friendly environment to do my studies. Especially Messors Shivaraj Gurung, Steven Trolove, Andrew Mitchell, Richard Lovell, Brett Robinson and Dr Manoharan for their support . The members o f the Soils and Plant N utrition Division o f the Tea Research Institute of Sri Lanka for their valuable help in conducting glasshouse trials in Sri Lanka. Mr Lance Currie and the laboratory technical staff for help in the laboratory and Mrs Denise Brunskill and Miss Marian Trembath for secretarial help. Mr Malcolm Boag for his help in proof reading, and Miss Ashwini Loganathan for her assistance in cross-checking the list of references. Dr S hiva Ganesh for his friendship and especially for his lectures on statistics. Drs N S Bolan and David Scotter for their friendship, support and useful discussions. Dr Nihal de Silva and the family for their love, affection and caring nature and especially for his ready assistance in solving statistical problems. Mrs Dianne Reilly and all members of the International Students office for their enormous support to make my life easy in N ew Zealand. Mr Palitha de Silva for his worthy advise and moral support at many difficult times and to his wife Kamini and children for their warm hospitality. Mrs Antoiniette Koolaard for her help in preparing a fine drawing for my thesis . Dr S Ganesalingam and his family for their support and friendship. My gratitude is also due to The Tea Research Institute of Sri Lanka for granting a scholarship to cover my living expenses in New Zealand. The Ministry of External Relations and Trade of New Zealand for granting a tuition fee scholarship to study at Massey university. The Vice Chancellor of Massey University for granting Graduate Research Award in 1994. The Academic Council of Agriculture-Horticulture Faculty of Massey University for awarding the J A Anderson memorial postgraduate Scholarship. To all the Sri Lankan friends at Palmerston N orth for their affection and care. E specially M essors Parakrama Aluvihare and Nandana Hewage for their support. Drs M artin Mainnee and John Koolaard, and Mr Van Wichern and their families for their love and friendship. In one way or another they made my life in New Zealand pleasant and memorable. Mrs Alexandra Cook and Margeret Humphrey with whom I was flatting for the last two years, for their pleasing qualities and constant support for my studies. Finally and most importantly, to my loving wife Shyama and daughter Kaumadhi for their enormous love, moral support, encouragement and sharing the j oy and sufferings during this period. v vi TABLE OF CONTENTS ABSTRACT 11 ACKNOWLEDGEMENTS IV TABLE OF CONTENTS VI LIST OF FIGURES XlI LIST OF TABLES XVlll CHAPTER 1 INTRODUCTION 1 . 1 B ACKGROUND 1 1 .2 OBJECTIVE S OF THE THE SIS 3 CHAPTER 2 LITERA TURE REVIEW 2. 1 TE A CULTIVATION IN SRI LANKA 5 2 . 1 . 1 E conomics and social impact of tea industry 7 2.2 CLIMATIC AND SOIL REQU IREMENTS OF TEA 1 0 2.2.1 Climate 1 0 2.2.2 Soils 12 2.2.2.1 Soil mineralogy and classification 1 2 2.2.2.2 Physico-chemical properties 1 2 2 . . 3 SOIL PHOSPHORU S 1 5 2 . 3 . 1 Plant available P 1 5 2.3.2 Characterisation of soil-P forms 17 2.3.3 The concept of plant-available P 20 2.3.4 Soil tests for plant-available P 21 2 . 4 PHOSPHORU S N U TRlTION OF TEA WITH SPECIAL REFERENCE vii TO SRl LAN KA 22 2 .4 . 1 Chronology of P fertiliser use on tea in Sri Lanka 2 . 5 E PPA W ALA PHOSPHATE ROCK (EPR) 2 . 6 FACTORS AFFECTING P AVAILA B ILITY FROM PR 2 .6 . 1 Characteristics ofPRs 2 .6 .2 Soil properties 2 .6 .3 C rop species 2 .6 .4 Moisture 2 . 6 . 5 Management practices 2 . 7 RE SPON SE O F TEA PLANTS T O P FERTILISERS 2 . 8 ROOT-SOIL INTERFACE (RHIZOSPHERE ) 23 33 34 34 37 3 8 43 44 44 49 2 . 8 . 1 Rhizosphere acidification 5 1 2 . 8 . 2 Relationship between cation-anion uptake and rhizosphere pH 54 2 . 8 . 3 Factors affecting P availability i n the rhizosphere 56 2 . 8 . 3. 1 pH 56 2 . 8 . 3 . 2 Release of organic acids and their chelating action 5 7 2 . 8 .3 .3 Pho sphatase enzyme activity and mineralisation of organic-P 5 8 2 . 8 . 3 .4 Mycorrhizal association 59 2 . 9 SUMMARY AND RE SEARCH NEEDS CHAPTER 3 A TECHNIQUE FOR STUDYING RHIZOSPHERE PROCESSES IN TREE CROPS: SOIL PHOSPHORUS DEPLETION AROUND CAMELLIA (Camelliajaponica L.) ROOTS 3 . 1 INTRODUCTION 3.2 OBJECTIVES 3 .3 MATERlALS AND METHODS 3.3.1 Soils 3.3 . 2 Glasshouse trial 60 62 63 64 64 64 3.4 3 . 5 4. 1 4 .2 4.3 4 .4 4.5 3 . 3 . 3 Field trial 3 .3.4 Plant and soil analysis 3 .3. 5 Soil P fractionation RE SULTS AND DISCUSSION viii 67 72 73 73 3 .4. 1 E ffect of P fertilisers on soil pH 73 3 .4 .2 E ffect of plant roots on soil pH 75 3 .4 .3 E ffect of P fertilisers on soil P fractions 76 3 .4 .4 E ffect of plant roots on soil P fractions 8 5 3 .4 . 5 Comparison of rhizosphere P depletion with P uptake processes 87 3 .4 . 6 Limitations o f the RSC technique 89 CONCLUSION S CHAPTER 4 EPPAWALA PHOSPHATE ROCK DISSOLUTION AND TRANSFORMA TION IN THE RHIZOSPHERE OF TEA (Camellia sinensis L.) COMPARED TO OTHER SELECTED PLANT SPECIES INTRODUCTION OBJE C TIVE S MATERIALS AND ME THODS 4 .3 . 1 Soil and plant sampling 4.3 . 2 Plant and soil analysis RE SUL TS AND DISC U S SION 4.4. 1 E ffect of plant species and E PR on growth characteristics 4 .4.2 E ffect of E PR fertiliser on soil pH 4.4 .3 E ffect of plant roots on soil pH and E PR dissolution 4 .4 .4 E ffect ofEPR addition and plant roots on soil P fractions 4 .4 .5 C omparison of rhizosphere P depletion with plant P uptake 4.4 . 6 E xternal and internal efficiency of P utilisation 4. 4 . 6. 1 E xternal efficiency ofP utilisation 4 .4. 6.2 Internal efficiency ofP utilisation CONCLU SION S 9 1 92 93 94 94 97 1 02 1 02 1 02 1 06 1 09 1 1 6 1 1 8 1 1 8 1 20 1 2 1 5 . 1 5 . 2 5.3 5 . 4 5 . 5 6 . 1 6 .2 6 .3 CHAPTERS PHOSPHORUS UTILISATION EFFICIENCY AND DEPLETION OF PHOSPHA TE FRACTIONS IN THE RHIZOSPHERE OF THREE TEA (Camellia sinensis L.) CLONES INTRODUCTION OBJE CTIVE S MATE RIALS AND ME THODS 5 . 3 . 1 Soil, plant and root sampling 5.3 . 2 Plant and soil analysis RE SULTS AND DISCUSSION 5 .4. 1 Dry matter yield, P and N concentration of tea clones 5.4.2 E xternal and internal P utilisation efficiency of tea clones 5 . 4 . 3 E ffect of P fertilisers and tea clones on soil pH 5 .4 .4 E ffect ofP fertilisers on soil P fractions 5 . 4 . 5 E ffect o f tea clones on soil P fractions 5 .4 . 6 Comparison of rhizosphere P depletion with plant P uptake C ONCLUSIONS CHAPTER 6 EFFECT OF FORMS OF NITROGEN SUPPLY ON MOBILISATION OF PHOSPHORUS FROM EPPA W ALA PHOSPHATE ROCK AND ACIDIFICATION IN THE RHIZOSPHERE OF TEA (Camellia sinensis L.) INTRODUCTION OBJECTIVES MATE RIAL S AND ME THODS 6 .3 . 1 Soil, plant and root sampling 6 .3 . 2 Soil and plant analysis i x 1 22 1 23 1 24 1 25 1 2 5 1 2 5 1 2 5 1 27 1 30 1 32 1 3 5 1 3 9 1 42 1 44 1 4 5 1 45 1 46 148 6 .4 6 . 5 7 . 1 7 .2 7 .3 7 .4 7 . 5 RE SUL T S AND DISCU S SION 6.4. 1 E ffect ofN forms on N and P uptake by tea 6 .4 .2 E ffect ofN forms on pH and P fractions in bulk soil 6 .4 .3 E ffect of N forms on rhizosphere pH 6.4 .4 E ffect of plant roots on P fractions in the soil 6 .4 .5 N utrient uptake and electro neutrality in plant tissues CON CLU S ION S AND IMP LICATION S CHAPTER 7 THE FATE AND EFFECTS OF PHOSPHATE FERTILISERS ON PHOSPHORUS AVAILABILITY TO TEA (Camellia sinensis L.) IN A HIGHLY ACIDIC ULTISOL IN SRI LANKA INTRODU CTION OBJE CTIVE S MATE RIALS AND ME THODS 7 .3 . 1 Plant and soil analyses 7 . 3 . 2 Relative agronomic effectiveness (RAE ) RE SULTS AND DISCU S SION 7.4. 1 E PR dissolution in soil 7 .4 .2 E ffect of P fertilisers on soil pH 7 .4 .3 E ffect of forms and rates of P fertilisers on soil P fractions 7.4.4 E ffect of P fertilisers on soil test extractable P 7 .4 .5 E ffect ofP fertilisers on growth and P uptake of tea plants 7 .4 .6 Relative agronomic effectiveness (RAE) of E PR 7 .4 .7 Relationship between dry matter yield and P uptake by plants 7 .4 . 8 Relationship between dry matter yield and leaf P concentration 7 .4 .9 Relationship between soil extractable-P and dry matter yield CONCLU SION S x 148 1 48 1 50 1 54 1 55 1 63 1 66 1 67 1 69 1 69 1 7 1 1 73 1 75 1 75 1 79 1 79 1 89 1 97 20 1 205 205 208 2 1 9 CHAPTERS SUMMARY AND CONCLUSIONS 8 . 1 AN OVERVIEW 8.2 A TECHNIQUE FOR STUDYING RHIZOSPHERE PROCESSES IN TREE CROPS 8. 3 PHOSPHORUS CHEMISTRY IN THE RHIZOSPHERE OF TEA AND ASSOCIATED CROPS 8 . 4 CLONAL VARIABILITY IN P UTILI SA TION 8 . 5 EFFECT OF FORMS OF NITROGEN SUPPLY ON MOBILISATION OF P FROM EPR 8 . 6 AGRONOMIC EFFECTIVENESS OF EPR ON TEA 8 . 7 FUTURE RESEARCH AND RECOMMENDATIONS REFERENCES xi 22 1 223 223 224 225 226 228 230 Figure 2.1 Figure 2.2 Figure 2.3 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 LIST OF FIGURES CHAPTER 2 Tea growing areas of Sri Lanka The area of cultivated tea land and tea production in Sri Lanka Proton (H+) and hydroxyl (OIT) generation during uptake and assimilation of different forms ofN into amino acids and subsequent dissociation of amino acids (Bolan et aI . , 1 99 1 ) CHAPTER 3 Schematic representation of Root Study Container (RSC) technique used in the (a) glasshouse and in the (b) field trials Bottom view of the root mats formed above the polyester mesh in root study containers (RSCs) Front view of the piston microtome Root Study Container (RSC) modified for field situation A root study container buried near a mature camellia tree in the field E ffect of P fertiliser forms on soil pH (0 . 0 1 M CaCh) in camellia rhizosphere in the glasshouse and the field trials (a) without plants - glasshouse trial (b) with plants - glasshouse trial and (c) with plants - field trial. Vertical bars correspond to Lsd at p <0.05 E ffect of soil pH (0 . 0 1 M CaCb), distance from rhizoplane and method of P extraction (with and without resin) on N CPR-P dissolution in the (a) glasshouse and (b) field trials. Vertical lines correspond to standard errors of means E ffect ofP fertiliser forms on resin-P in camellia rhizosphere in the glasshouse and in the field trials (a) with plants - glasshouse trial and (b) with plants - field trial. Vertical bars correspond t o Lsd at p <0. 05 and N represents statistically nonsignificant at p <0.05 xii 6 8 5 3 66 68 69 70 7 1 74 77 80 Figure 3.9 E ffect of P fertiliser forms on NaOH-Pj in camellia rhizosphere in the glasshouse and in the field trials (a) with plants - glasshouse trial (b) with plants - field trial. Vertical bars correspond to xiii Lsd at p <0. 05 8 1 Figure 3.1 0 E ffect o f P fertiliser forms o n H2S04-Pi i n camellia rhizosphere in the glasshouse and in the field trials (a) with plants - glasshouse trial and (b) with plants - field trial. Vertical bars correspond to Lsd at p <0.05 82 Figure 3.1 1 E ffect ofP fertiliser forms on NaOH-Po in camellia rhizosphere in the glasshouse and in the field trials (a) with plants - glasshouse trial and (b) with plants - field trial. Vertical bars corresponds to Lsd at p <0.05 and N represents statistically not significant at p <0.05 83 CHAPTER 4 Figreu 4.1 The arrangement of plant species in the glasshouse experiment (A - Guinea grass, B - Calliandra, C - Bean and D - Tea) 96 Figure 4.2 Root mats of formed on the polyester mesh for Guinea grass in E PR and control treatments 98 Figure 4.3 Root mats of formed on the polyester mesh for bean in E PR and control treatments 99 Figure 4.4 Root mats of formed on the polyester mesh for calliandra in E PR and control treatments 1 00 Figure 4.5 Root mats of formed on the polyester mesh for tea in EPR and control treatments 1 0 1 Figure 4.6 E ffect of plant species on rhizosphere pH ( l : 2 . 5 w/w H2O) of (a) control and (b) E PR fertilised soils. Vertical bars and N represent Lsd at p <0.05 and treatments that are not statistically significant at p <0 .05 respectively 1 04 Figure 4.7 E ffect of plant species on E PR dissolution in the rhizosphere. Vertical bars represent standard errors of the means 1 05 Figure 4.8 E ffect of plant species on resin-P in the rhizosphere of (a) control (no E PR added) and (b) E PR fertilised soils. Vertical bars and N represent Lsd at p <0.05 and treatments that are not statistically significant at p <0.05 respectively 1 1 1 Figure 4.9 E ffect of plant species on N aOH-Pdn the rhizosphere of (a) control (no E PR added) and (b) E PR fertilised soils. Vertical bars and N represent Lsd at p <0.05 and treatments that are xiv not statistically significant at p <0.05 respectively 1 1 2 Figure 4.10 E ffect of plant species on H2S04-Pdn the rhizosphere of (a) control (no E PR added) and (b) E PR fertilised soils. Vertical bars and N represent Lsd at p <0.05 and treatments that are not statistically significant at p <0.05 respectively 1 1 3 Figure 4.1 1 E ffect of p lant species on N aOH-Po in the rhizosphere of (a) control (no E PR added) and (b) E PR fertilised soils. Vertical bars and N represent Lsd at p <0.05 and treatments that are not statistically significant at p <0. 05 respectively 1 1 4 CHAPTERS Figure 5.1 E ffect of (a) P uptake per unit surface area and (b) total root surface area on P uptake by tea clones 1 28 Figure 5.2 E ffect of tea clones on rhizosphere soil pH in (a) Control (b) E PR and (c) TSP treatments . Vertical bars correspond to Lsd at p <0. 05 and N represents treatments are not statistically significant at p <0. 05 1 29 Figure 5.3 E ffect of clonal differences on E PR dissolution in the rhizosphere. Vertical bars represent standard errors of the means 1 3 1 Figure 5.4 E ffect of tea clones on resin-P in soil with (a) control (b) E PR and (c) TSP treatments. Vertical bars correspond to Lsd at p <0.05 and N represents treatments not statisticaJIy significant at p <0.05 1 3 6 Figure 5.5 E ffect of tea clones on N aOH-Pi in soil with (a) control (b) E PR and (c) TSP treatments. Vertical bars correspond to L sd at p <0.05 and N represents treatments not statistically significant at p <0.05 1 3 7 Figure 5.6 E ffect of tea clones on H2S04-Pi in soil with (a) control (b) E PR and (c) T SP treatments . Vertical bars correspond to Lsd at p <0.05 and N represents treatments not statistically significant at p <0.05 . 1 3 8 Figure 5.7 E ffect of tea clones on N aOH-Po in soil with (a) control (b) E PR and (c) TSP treatments . Vertical bars correspond to Lsd at p <0.05 and N represents treatments are not statistically significant at p <0.05 1 40 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Figure 6.7 Figure 7.1 Figure 7.2 Figure 7.3 Figure 7.4 CHAPTER 6 Plant growth system in the glasshouse Effect of nitrogen forms on P dissolution from EPR. (Vertical bars represent standard errors of the means) E ffect of nitrogen forms on soil pH (a) with plants and (b) without plants. (Vertical bars represent Lsd for treatment means at p <0.05) Effect of nitrogen forms on resin-P (a) with and (b) without plants. Vertical bars represent Lsd for treatment means at p <0.05 Effect of nitrogen forms on N aOH-Pdn soil (a) with and (b) without plants. Vertical bars represent Lsd for treatment means at p <0.05 E ffect of nitrogen forms on H2S04-Pi in soil (a) with and (b) without plants Vertical bars represent Lsd for treatment means at p <0.05 Effect of nitrogen forms on NaOH-Po in soil (a) with and (b) without plants. The N shows that treatments are not statistically significantly different at p <0.05 CHAPTER 7 Tea plants growing in pots in the glasshouse study Effect of time and EPR fertiliser rate of addition on P dissolution in soil with and without plants. Vertical bars correspond to Lsd at p <0. 05 E ffect ofE PR and T SP on soil pH (H20) after 5 and 1 0 months o f application (a) with and (b) without tea plants. Vertical bars correspond to Lsd at p <0.05 and N S represents treatments not statistically significant at p <0.05 Effect ofEPR and TSP fertiliser rates on resin-P in soil (a) with and (b) without tea plants. Vertical bars correspond to Lsd at p <0.05 for main effects ofP sources and N S represents treatments not statistically significant at p <0.05 xv 1 47 1 5 1 1 52 1 57 1 58 1 59 1 60 1 72 1 77 1 80 1 83 xvi Figure 7.5 E ffect ofE PR and TSP fertiliser rates on NaOH-Pj fraction in soil (a) with and (b) without tea plants. Vertical bars correspond to Lsd at p <0.05 for main effects of P sources in (a) and interaction effects of P sources. rates in (b) respectively 1 84 Figure 7.6 E ffect ofE PR and TSP fertiliser rates on NaOH-Po fraction in soil (a) with and (b) without tea plants. Vertical bars correspond to Lsd at p <0.05 for main effects ofP sources and NS represents treatments not statistically significant at p <0. 05 1 85 Figure 7.7 E ffect ofE PR and TSP fertiliser rates on H2S04-Pi fraction in soil (a) with and (b) without tea plants. Vertical bars correspond to interaction (P sources. rates) Lsd at p <0. 05 1 86 Figure 7.8 E ffect ofE PR and TSP fertiliser rates on Resin-P in soil (a) with and (b) without tea plants 1 90 Figure 7.9 E ffect ofE PR and TSP fertiliser rates on OIsen-P in soil (a) with and (b) without tea plants 1 9 1 Figure 7.10 E ffect ofE PR and TSP fertiliser rates on Borax extractable-P in soil (a) with and (b) without tea plants 1 92 Figure 7.1 1 E ffect ofE PR and TSP fertiliser rates on Bray- l P in soil (a) with and (b) without tea plants 1 93 Figure 7.1 2 E ffect ofE PR and TSP fertiliser rates o n c itric acid extractable -P in soil (a) with and (b) without tea plants 1 94 Figure 7.13 E ffect of E PR and TSP fertiliser rates on m alic acid extractable -P in soil (a) with and (b) without tea plants 1 95 Figure 7.14 Relationship between shoot dry matter yield and the rates of E PR and TSP at 5 and 1 0 months after application to the soil 1 99 Figure 7.15 Relationship between shoot P uptake and the rates ofE PR and TSP at 5 and 10 months after application to the soil 200 Figure 7.16 Relationship between shoot dry matter yield and P uptake by tea plants at 5 and 1 0 months after P fertiliser application 206 Figure 7.17 Relationship between shoot dry matter yield and P concentration in the first mature leaf for tea plants treated with E PR and T SP fertiliser 207 Figure 7. 18 Relationship between shoot dry matter yield and resin-P for tea plants treated with E PR and TSP fertiliser 209 xvii Figure 7.19 Relationship between shoot dry matter yield and Olsen-P for tea p lants treated with E PR and TSP fertiliser 2 1 0 Figure 7.20 Relationship between shoot dry matter yield and borax-P for tea plants treated with E PR and TSP fertiliser 2 1 1 Figure 7.21 Relationship between shoot dry matter yield and Bray- l P for tea plants treated with E PR and TSP fertiliser 2 1 2 Figure 7.22 Relationship between shoot dry matter yield and citric acid extractable P for tea plants treated with E PR and TSP fertiliser 2 1 3 Figure 7.23 Relationship between shoot dry matter yield and malic acid extractable P for tea plants treated with E PR and TSP fertiliser 2 1 4 xviii LIST OF TABLES CHAPTER 2 Table 2.1 Average productivity of tea in major tea growing countries (Kelegama et ai . , 1 995) 9 Table 2.2 The profitability of tea cultivation in some tea growing countries (Kelegama et al . , 1 995) 1 1 Table 2 .3 Mineralogy and classification of tea soils of the world l3 Table 2.4 Chemical properties of some major tea-growing soils of the world 1 4 Table 2.5 The chemical nature of the various soil P fractions (Hedley et aI ., 1 982b) 1 8 Table 2.6 Phosphorus fractions ( Ilg P g-l soil) in tea soils of different countries determined according to the method of Chang and Jackson ( 1 957) 1 9 Table 2.7 Mean dry matter distribution and nutrient removal by 1 00 kg processed tea and Iper ha y(1 ofland in Sri Lanka (after E den, 1 949) 24 Table 2.8 Comparison of P removal by different agricultural and tree crops 2 5 Table 2.9 Chronology of P fertiliser recommendations for tea in S ri Lanka 26 Table 2.10 The fertiliser mixtures used for tea nurseries in Sri Lanka 3 1 Table 2.11 Chemical analysis of the primary apatite and aluminous- ferruginous-siliceous phosphate matrix of E ppawala phosphate rock deposit (Dahanayake et a!., 1 995) 3 5 Table 2.12 E xamples of previous research conducted on P utilisation from phosphate rocks by different plant species 3 9 Table 2.13 Circulation of phosphorus in tea plants yielding 5050 kg processed tea ha-I in a three year pruning cycle (calculated from the data of Willson, 1 969) 46 Table 2.14 Effect of P sources on processed tea yield (kg ha-! yr-!) in South India (Ranganathan, 1 97 1- 1 980) CHAPTER 3 Table 3.1 The physico-chemical characteristics of the experimental soil Table 3.2 Estimated and predicted Ii consumption for dissolution of NCPR fertiliser in the rhizosphere and bulk soil in the glasshouse and the field trials Table 3.3 P-fractions in u nfertilised bulk soil (mean of soil slices 3 - 8 mm from rhizoplane) and % recover/ of added P fertiliser in the various P fractions in the glasshouse and the field trials Table 3.4 Comparison of phosphorus depletion from soil in the lower compartment ofRSC with plant P uptake in the glasshouse trial (values represent the mean of 5 replicates) Table 3.5 Effect of P fertiliser forms on root growth within 1 mm of the mesh and P depletion in the field trial (values represent the mean of 5 replicates) CHAPTER 4 Table 4.1 Selected physico-chemical properties of the soil used in the study Table 4.2 Comparison of plant dry matter yield and shoot : root weight ratio and P concentration of tea with those of other plant species Table 4.3 The total acid production by roots of different plant species treated with EPR into the rhizosphere of the lower compartment Table 4.4 The P fractions in the control treatment and % recovery! of added P from EPR in the bulk soil Table 4.5 Comparison ofP depletion by different plant species from soil in the lower compartment ofRSC with plant P uptake xi x 48 65 78 79 88 90 95 1 03 1 07 1 1 0 I I 7 Table 4.6 Table 5. 1 Table 5.2 Table 5.3 Table 5.4 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 -- --� �--- Comparison of external and internal efficiencies of P in plants with and without added P fertiliser CHAPTER 5 Dry matter yield, N and P concentration in shoots and roots, and 1 1 9 P utilisation efficiencies of three tea clones 1 26 The comparison of observed acidity release in the rhizosphere of tea clones in E PR treated soil and the predicted acid relea se based on E PR dissolution l 3 3 P fractions i n the control (no P fertiliser added) and % recovery' of added P from E PR and TSP treated bulk soil (3 - 5 mm) after 56 days of plant growth 1 34 Comparison of observed P depletion in the rhizosphere with predicted P depletion estimated from plant P uptake 1 4 1 CHAPTER 6 E ffect ofN forms on plant shoot and root dry matter! , shoot : root ratio, t issue N, P concentrations and uptake by tea p lants 1 49 Phosphorus fractions in control soil (without E PR) and % recovery of added E PR-P in bulk soils (3 - 5 mm) for various N treatments 1 53 The observed acid relea se by tea roots based on pH cha nges and the predicted acid production based on E PR dissolution in the lower compartment of RSC 1 56 Comparison of observed P depletion in the soil in the lower compartment of RSC with predicted P depletion calculated from plant P uptake 1 62 E ffect of different N forms and a ssumed ratios of� T : N03- uptake on the net release of H' t o the rhizosphere (0 - 3 mm) by tea roots 1 64 xx xxi CHAPTER 7 Table 7 . 1 Selected properties of the soil (Rhodustult) u sed 1 70 Table 7.2 Summary of soil P tests used 1 74 Table 7.3 The rr consumption for EPR dissolution in soil with and without tea plants 1 78 Table 7.4 The % recovery I of added P in soil P fractions in the unfertilised (control) and EPR and TSP treated soils at the end of the trial 1 8 1 Table 7.5 Effect of EPR and TSP fertilisers and their rates on shoot dry matter yield and P uptake 1 98 Table 7.6 Calculated P requirement for 95% of maximum dry matter yield 202 Table 7.7 The agronomic effectiveness ofEPR relative to TSP calculated from empirical relationships 203 Table 7.8 Correlation matrix for P extraction methods for all rates of (a) EPR and (b) TSP fertilisers 2 1 5 Table 7.9 Correlation matrix for P extraction methods for rates 0, 1 0 and 20 kg P ha-1 of (a) EPR and (b) T SP fertilisers 2 1 6 Table 7.1 0 The critical levels o f plant available-P required t o be i n soil to obtain 95% of the maximum dry matter yield in tea plants 2 1 7 1 . 1 BACKGROUND CHAPTER 1 INTRODUCTION 1 Phosphorus (P) is one of the most important macro�nutrients for tea (Camellia sinensis L . ), influencing growth, yield and quality (Gogoi et aI., 1 993; Ranganathan et aI . , 1 98 2; Yongming et ai . , 1989) . Phosphorus i s essential for a wide range of plant metabolic functions (Mengel and Kirkby, 1 98 7), most importantly carbon fixation during photosynthesis. As P is indispensable for the biochemical and physiological reactions in plant t issues, fertiliser or manure P application is essential to obtain sustainable yields. The Vitic soils on which tea is commonly grown contain very low concentrations of 1 - 4 !lg P g-l soil (borax-extractable P, Wickremasinghe, 1 986) in their native status. Young tea shoots (two leaves and a bud) that are removed regularly by plucking for processing into drinking tea have a higher P concentration (0 .36 - 0 . 29% P) than the older leaves (0. 2 1 - 0.20% P) (Hasselo, 1 965), suggesting a higher physiological demand for P in growing tissue. This P removal (6 . 8 kg P ha-1 y(l) in young tea shoots must be replaced by frequent application of P fertilisers to the soil . Organised research on the response of tea to applied P fertilisers in S ri Lanka was started in the 1 930s' (E den, 1 934). Though symptoms that are typical of P deficiency in tea are unlikely to be observed under field conditions (Biswas et aI. , 1 984; Sengupta et a! . , 1 98 6) , sub-optimum levels of soil P have been shown to reduce tea yields (Eden, 1 976) . Experimental evidence on Sri Lankan UItisols showed that there is no yield response to repeated annual application of P fertiliser beyond 1 5 kg P ha-1 y(l . Results of experiments in other tea growing countries also showed that maximum response to P was around 1 5-20 kg P ha-l y(l (Willson et a! . , 1 975) . Whilst P remains essential for growth, the reason for this low P requirement in tea is not fully understood The high P fixation capacities of Sri Lankan tea soils, due to low pH and 2 the presence of Fe and A1 oxides, cause a higher fraction of applied P fertiliser to remain unavailable to plants, resulting in high amounts of residual-P (Golden et aI., 1 98 1 ) in frequently fertilised soils . As P -diffusivity in soils is very low (Barber, 1 995) the availability of P for plant uptake is dependent on the inorganic P concentration gradient and the diffusion conditions in soil around the fine roots of tea plants. D espite the high P fixation in many tea soils, tea is found to grow well in many parts of the world with low rates of P fertiliser application. One reason for this may be its ability to utilise pools of soil P that are not traditionally considered as plant available, especially in the rhizosphere. A variety of mechanisms have been proposed to account for the increased mobilisation of soil P in the rhizosphere of various crop species. These mechanisms include the exudation of reducing substances (Gardner et aI . , 1 982), hydrolysis of organic P by microbial and plant phosphatases (Eivazi and Tabatabai, 1 977; Helal and Sauerbeck, 1 984; Tarafdar and Jungk, 1 987), action of V A-mycorrhiza (Young et al ., 1 98 6) and excretion of organic acids (Moghimi et at, 1 978). Several studies have witnessed soil pH changes in the rhizosphere as a major factor in the dissolution/desorption of soil P (Gardner et al . , 1 982; Hedley et aJ . , 1 982ac; Riley and Barber, 1 97 1 ) . Previous studies showed that tea plants secrete various organic acids (Jayman and Sivasubramaniam, 1 975; Xiaoping, 1 994), which could dissolve native soil P . Most developing countries are interested in utilising low-cost locally available phosphate rock (PR) resources for sustainable crop production. In Sri Lanka, an estimated 40 million metric tons of an apatite bearing rock deposit was discovered in Eppawala (Jayawardena, 1 976). This material is now ground and u sed as a direct P fertiliser for many perennial crops in Sri Lanka (Dahanayake et aI . , 1 995) . E ppawala phosphate rock (EPR) is recommended by the Tea Research Institute of Sri Lanka for tea plantations, but its agronomic value has been tested little. It was felt that the low pH and high rainfall in the tea-growing areas a ssist EPR to dissolve, thus releasing sufficient P to the plants. The regular application of EPR in previously well fertilised soils will not improve dry matter yield. Therefore a soil testing procedure is needed to identifY these conditions to help minimise the wasteful use of P fertiliser, in particular EPR. 3 Various soil tests are being used to determine the soil P status in different tea-growing countries of the world, but these tests have not been developed based on sound research correlating soil test values to dry matter yield, or the P uptake of tea plants. Therefore there is a need to find an easy and effective soil P test to determine the P needs for tea. In the recent past new tea clones were developed and selected by many countries (Alam, 1 994; Anandappa, 1 986; Astika, 1 994; Othieno, 1 994), for higher yields, increased resistance to pests and diseases and drought tolerance etc. , but hardly any information is available on the efficiency of P utilisation by these clones from different P sources. Perennials like calliandra (Calliandra coloth,ysus L.), which i s grown as a shade tree and Guinea grass (Panicum maximum L.), which is largely found in abondanded tea fields are known to grow well in Sri Lankan tea soils. Beans (Phaseolus vulgaris L .) which are a fast growing leguminous vegetable crop are also grown in these soils. Very little is known about the P acquisition characteristics of these plants as compared with new tea clones grown in the same soil and with different P sources. Such information on P uptake by tea and companion plants may be useful in the efficient management of soil and fertiliser P in strongly weathered acidic Ultisols. 1 .2 OBJECTIVES OF THE THESIS The research reported in this thesis attempts to fill some of the gaps in the knowledge on P availability to tea from native and fertiliser P sources. The specific objectives of the research are: 1 . To develop a suitable technique to study root induced changes on soil pH and P fractions in the rhizosphere and P uptake processes in tea plants. 2. To compare the rate of EPR dissolution and P mobilisation in the rhizosphere of different tea clones and other plants cultivated in tea soils. 3 . To determine the effect of different forms ofN supply on EPR dissolution and P mobilisation in the tea plant rhizosphere, . 4 . To compare different soil tests in predicting P uptake and yield of tea plants fertilised with EPR and TSP (triple superphosphate). 4 CHAPTER 2 LITERATURE REVIEW 2.1 TEA CULTIVATION IN SRI LANKA 5 Tea (Camellia sinensis (L.) O . Kuntze) is a perennial crop cultivated commercially as a mono culture on large scale plantations in many humid tropical countries of the world (Willson and Clifford, 1 992) . Tea belongs to the genus Camellia which comprises as many as 82 species distributed mainly in South-East Asia. It is consumed as a beverage after brewing. The popularity of tea is increasing throughout the world because of its positive effects on human health (Ganguly, 1993). Tea cultivation originated in S outhern China (The People's Republic of China) around 300 Be. The habit of tea drinking as a ritual and as a common habit was started in China and spread to Japan around 1 000 AD (Blofeld, 1 98 5). In 1834, the B ritish started cultivating tea in India by importing tea plants and experts from China. The first commercial tea plantation in Sri Lanka was started by James Taylor, a Scot in 1867, on a 1 9 acres of land in Loolecondera estate, Hewaheta in Central province. Later, tea became a timely venture as the coffee leaf rust disease destroyed the entire coffee cultivation in Sri Lanka within 25 years of its appearance. Tea proved to be a more durable and profitable resource and rapidly became the major plantation crop in Sri Lanka (Humble, 1 99 1 ) . In 187 1 , 1 00 ha of tea were planted and it increased to 6000 ha in 188 1 . The area of cultivation and the corresponding tea production increased remarkably with time The major tea-growing areas of Sri Lanka are presented in Figure 2. 1. The first consignment of Ceylon (now Sri Lanka) tea was exported in 1872. Today Sri Lanka is one of the biggest tea producing countries and the largest tea exporter in the world. Tea is cultivated on approximately 0.2 million ha planted with varying proportions of seedlings and improved, vegetatively propagated (VP) tea clones. t c:::. Pal' lV ( � � ) ". o j. o 80mr I annar Tea areas "J-t t" .Anuradhap6a q, roo Puttalam \., ) \ ic.aloa ", . � • Kandy ., Hambanto a 82° 00' Figure 2. 1 Tea growing areas of Sri Lanka 6 7 Tea production increased steadily until the mid 1 960s', and thereafter consistently declined, until it reached its lowest figure of 1 79 million kg processed tea (779 kg ha-1 yr-l) in 1 983 . Tea yields again increased and recorded the highest production of 240.7 million kg processed tea ( 1 08 5 kg ha-1 y{l) in 1 99 1 (Figure 2.2) . The increase in total production at the early stages of the tea industry ( 1 940 - 1 960) was mainly due to the u se of high yielding new tea clones coupled with a rapid expansion of the land area under cultivation. The subsequent decline in production during the 1 980s was due to mismanagement, as a result of the then government policy of fragmentation of tea estates and changes in land ownership . Neglect of cultural practices, poor land care and senility of the tea bushes also played a part. The average productivity (processed tea yield ha-l) of tea land has increased over time, although it is still lower compared to many other tea producing countries (Table 2. 1 ). If only privately owned lands are considered the productivity per ha-l in Sri Lanka is the highest in the world. The privately owned tea lands are about 42% of the total cultivated area of tea (Jayakodi, 1 996) . The key reasons for the relatively low productivity (yield ha- 1) in the state­ owned tea plantations in Sri Lanka are predominance of seedling tea, high vacancies of plants, senility, low labour productivity, low and untimely inputs of fertiliser and poor agronomic care (Humble, 1 99 1 ) . 2 . 1 . 1 Economics and social impact of tea industry Tea has been of predominant importance to the Sri Lankan economy for more than 1 3 0 years and still maintains a very influential position. The area under tea cultivation covers about 1 0% of the total cultivated land area in Sri Lanka, and the share of tea exports in total export earnings of Sri Lanka was about one-third in the 1 980s (Bodhidasa, 1 989; Bogahawatta and Herath, 1 984) . A substantial amount of revenue comes to the government from taxes levied on the tea industry (Ron, 1 986). Tea is probably the most labour-intensive industry in Sri Lanka, employing about 3.2 1 persons ha-1 This results in direct employment of approximately 0. 5 million workers, who are complimented by at least an additional 0 .2 million people dependent on the tea industry by way of product development, packing, transport, marketing, export, research, extension etc. 8 250 250 200 200 � (!) � -"0 ..s:::: (!) r<"l CIl 0 CIl (!) ,....... u * 8 '---' 1 50 1 50 0.. ro (!) OJ) ..... ..:.:: <..;... \C; 0 0 ro (!) * '- � ro :: '"0 1 00 1 00 .2 (!) ....... () ro ::l ;>- '"0 E 0 ::l ..... 0... U -e- ro 50 Cultivated area 50 (!) E- ... Tea production 1 880 1 900 1 920 1 940 1 960 1 980 2000 Year Figure 2 .2 The area of cultiv ated tea l and and tea producti on i n Sri Lanka Table 2.1 Average p roductivity of tea in the maj or tea growing countries (Kelegama et ai., 1 995) Country Yield (kg processed tea ha-I yr-1) Sri Lanka state owned estates 1 268 privately owned estates 2442 India North 2 1 27 South 2300 Kenya 223 7 Indonesia 1 64 5 Malawi 1 929 9 10 A comparison of the profitability of tea cultivation in some tea growing countries of the world is given in Table 2 .2 . Compared to the other two major tea producing countries, India and Kenya, Sri Lankan labour output from tea plantations is deplorably low, especially in the state owned estates leading to high costs of production. The gross profit ha-] is also very low from Sri Lankan tea lands, when the mean values of state and private production are considered. 2.2 CLIMATIC AND SOIL REQUIREMENTS OF TEA Tea has wide soil and climatic adaptability. It grows in a range of climates and soils in many parts of the world . At present tea is grown on a commercial scale in countries all around the world from far north as Georgia (US SR) 43° N Lat., to far south as Corrientes (Argentina) 27° S Lat . . Tea is also grown at various altitudes ranging from sea level to about 23 00 m above sea level in Kenya, Malawi, Japan and Indonesia. Tea cultivation is now beginning to occur in Australia and New Zealand as well . 2.2. 1 Climate Tea is grown in conditions ranging from Mediterranean type climates to the hot humid tropics (Carr and Stephens, 1992) . Tea plants require a warm humid climate. An annual rainfall of 2500 - 3000 mm with even distribution throughout the year, without marked seasonality is the ideal condition. The minimum annual rainfall requirement is 1200 mm (Watson, 1986). The ideal ambient temperature (mean monthly) is considered to be 18 - 25° C . Mean monthly temperatures lower than 1 3° C (average for the coldest month) or higher than 3 0° C (average for the hottest month) will affect tea production. It certainly cannot withstand frost. It varies in its ability to withstand wind (Watson, 1986). Tea growing areas in Sri Lanka are classified into three main groups based on altitude (Watson, 1986) . The areas under 600 m altitude are referred to as low-country, and those above 1200 m are categorised as up-country. The areas between 600 m and 1200 m are called mid-country. These areas are further subdivided into wet, intermediate and dry zones based on the amount of annual rainfall. Table 2.2 The profitability of tea cultivation in some tea growing countries (Kelegama et . al . , 1 995) Index Unit Sri Lanka India Kenya State estates Private estates North South Plucker intake kg dai1 1 3 . 52 24.59 26.22 25.24 4 8 . 00 Labour ha-1 No. 3 .2 1 2 .70 2 . 67 2 . 5 0 2.20 Cost of production US $ kg-I 1 . 87 1 . 54 1 . 5 2 1 . 3 9 0 .94 Revenue US $ ha-I 2574 4957 43 1 8 4669 4 3 3 8 Gross profit US $ ha-1 203 1 1 96 1085 1472 2438 1 2 2.2.2 Soils 2.2.2.1 Soil mineralogy and classification Tea grows in a wide range of soil types in tropical, sub-tropical and temperate climates, and the soils are derived from diverse parent materials (E den, 1 976; Mann, 1 93 5) . The mineralogy and the type of soils differ greatly not only between countries, but also within a country (Table 2 .3) . Most tea soils are highly weathered and strongly acidic and the major clay minerals present in these soils are kaolinite, gibbsite and geothite. These soils are broadly categorised as U ltisols or Oxisols according to the U S soil taxonomy because they are in the advanced stage of weathering. 2.2.2.2 Physico-chemical properties Soils of diverse origin and different morphological characteristics support viable tea cultivation in different countries (Child, 1 953 ; E den, 1 976; Harier, 1 97 1 ; Mann, 1 93 5) . Generally tea prefers a deep, permeable and well drained soil. In a given locality, soil characteristics such as soil depth <50 em, gravelliness >50% and rockiness >20% impose severe limitations for successful growth of tea (Watson, 1 986). S ome important chemical properties of the tea growing soils in different countries are given in Table 2 .4 . The most important soil chemical property for the good growth of tea is optimum pH. Generally the soil pH (H20), where tea is cultivated, varies from 3 . 3 - 6 .0 . The optimum range of soil pH (H20) for tea plants is 5 . 0 - 5.6 (Othieno, 1 992). The maintenance of soil pH between 4 . 5 - 5 . 5 is preferred in Sri Lanka (Anon, 1 989). Tea is considered as a calcifuge and does not grow well in soils of high base saturation. However it still needs a certain amount of Ca for satisfactory growth and for maintenance of high yield levels (E den, 1 976; Ranganathan and Natesen, 1 98 5). Tea is also known to take up large quantities of AI (Chenery, 1 95 5 ; Foy et aI . , 1 978), which is relatively easily available in most acid soils where tea is cultivated . 1 3 Table 2.3 Mineralogy and classification of tea soils of the world Country Main soil types or soil Clay minerals Reference classification India Drujeeling district Sedimentary Kaolinite, some Ramanathan and Cachar district Peaty illite and Krishnamoorthi ( 1 979), South India Latosols montmorinolite. Subramanian and Mani ( 1 98 1 ) Sri Lanka Red-yellow podzolic, kaolinite, De Alwis and Panabokke Reddish-brown gibbsite, ( 1 972) latosolic, geothite, small Immature brown amount of i l l ite. loam. Bangladesh Alluvial kaolinite, small Karim et al . ( 1 98 1 ) amounts of mica and gibbsite. East Africa Latosolic Scott ( 1 962) Malawi Alluvial Ranganathan and Natesen ( 1 985) Kenya Volcanic ash Ranganathan and Natesen ( 1 985) Tanzania Volcanic ash Ranganathan and N atesen ( 1 985) Russia Red soils, Jourbitzky and Strausberg Podzols . ( 1 966), Dey ( 1 972) Taiwan Reddish-brown, Chu ( 1 975) lateritic, Red-yellow podzolic, Yellow brown earth. Japan Volcanic ash, Ranganathan and N atesan Red yellow soils. ( 1 985) China Red soils Ranganathan and Natesan ( 1 985) Table 2.4 Chemical properties of some major tea-growing soils of the world Country Soil depth pH Organic Total CfN Total P CEC Ex. bases Available-P I Reference (area) (cm) (water) C% N% (llg g-l soil) (cmolc kg-I ) (cmoic kg-l ) (llg g-l soil) ( 1 :2.5) K Mg Ca India Assam 0-30 4.7 1 . 7 0 . 1 0 10 5 .8 1 .0 0.4 0 . 8 1 5 Dey ( 1 969) Anamallais 0- 1 5 4.9 7 .5 0.28 1 1 . 2 0.5 1 .7 8 . 8 2 2 Ranganathan ( 1 976) Kenya Kericho 0-5 5. 1 8 .5 0.60 8 2 . 1 2 .5 0 .9 1 4 Othieno ( 1 973) 5 - 1 0 4.5 8.0 0 .58 8 1 .9 1 .4 0.6 8 Othieno ( 1 973) 1 0- 1 5 4.6 7 . 1 0 . 50 8 1 .7 0.7 0.2 7 Othieno ( 1 973) 0-23 4 .7 3 . 7 0. 1 9 20.8 3 . 1 2 .0 0 . 5 4 Dey ( 1 969) Malawi 0-23 5 .3 3 .0 0 . 1 7 1 1 2 .7 1 . 2 1 . 3 2 . 5 1 4 Dey ( 1 969) Sri Lanka St. Coombs 0- 1 5 4.5 2 .6 0 .22 12 500 1 0. 1 - 1 3 . 3 0 .2 0. 1 0.6 73 Jayman and Hantana 0- 1 5 4.4 1 . 4 0 . 1 3 1 1 500 5 .0 - 7 .2 0. 1 0 .2 1 .0 1 70 Sivasubramaniam Passara 0- 1 5 3 .9 3 . 1 0 .20 15 400 1 . 1 - 1 .9 0. 1 0. 1 0 .3 1 9 ( 1 98 1), Kottawa 0 - 1 5 5.2 1 .6 0 . 1 3 1 3 80 3 . 4 - 3 .9 0 .2 0. 1 0.2 17 Wickremasinghe Ratnapura 0- 1 5 4.7 1 . 7 0 . 1 3 14 700 1 . 3 - 2.6 0 . 1 0. 1 0 . 1 25 ( 1986). Taiwan 1 st horizon 4.4 1 . 5 0.09 10 0 .2 0 . 1 1 .7 2 Chu (1975) 1 The available P in Sri Lankan soils refers to Borax-extractable P (Beater, 1 949); methods used were not reported for other countries. 1 5 Most tea soils are highly weathered and contain an abundance of variable charge colloidal minerals, such as oxides and hydrous oxides of Fe and AI and 1 : 1 type clay minerals. The soils are generally low in CEC (usually < 1 0 cmolc kg-I) . The soils have low base saturation and high AI saturation (often >40%). The excess weathering in these soils reduce plant-available P levels in the soil solution, because of high P sorption capacities associated with high contents of Fe and AI oxides ( Sanchez, 1 976). The overall effect is that it leads to an infertile medium for plant growth. Therefore the soils need to be managed properly to maintain a high fertility status, which wil l lead to high tea yields. 2.3 SOIL PHOSPHORUS 2.3. 1 Plant available P Tea meets its' P requirements from native soil P sources and from the added P of fertilisers. Among the macro-nutrients, P is the most susceptible to fixation by highly weathered acidic soils. Crops usually remove proportionately less P from added P fertil iser or native soil P in the short-term compared to other nutrients. In acid soils phosphorus fixation occurs by the removal of P from the soil solution either by sorption on Fe and AI oxides, or precipitation reactions with soluble Fe and AI, which are abundant in acid soils (Golden et al ., 1 98 1 ) . Generally P availability declines rapidly as soil pH falls below 5.0 (Apthorp et ai . , 1 987; Fox, 1 979; Parfitt, 1 977; Sanchez, 1 976) . Plant-available P concentrations in tea soils vary widely with location, reflecting changes in soil and management practices (Table 2 .4). It is difficult to compare the available soil P values reported in the literature as the methods of extraction are different from one country to the other. The plant-available soil P concentrations in Sri Lankan tea soils (0 - 1 5 em) as determined by borax extraction (Beater, 1 949) varies from 1 7 - 1 70 Ilg g-l soil (Wickremasinghe, 1 986). These soil P levels are found to be significantly higher than the P in adjoining forest soils (0.9 - 8 . 7 Ilg g-l soil) and 1 6 result from continuous additions of P fertilisers to tea plantations over a long period of time (Wickremasinghe, 1 986) . The P i n soils i s generally considered t o b e made u p o f the following three fractions, which are in dynamic equilibrium as shown in Equation 2 . 1 (Mengel and Kirkby, 1 987) : Soil solution P <=> Labile P <=> Non-labile P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Equation 2 . 1 ] The soil solution P fraction i s the phosphate dissolved i n soil solution and it i s the immediate source of P for plant uptake by roots. The labile P fraction is the solid phosphate, which is held on soil surfaces, in rapid equilibrium with soil solution phosphate. It consists of freshly precipitated Fe and AI phosphates and P adsorbed to the surface of soil minerals. It can be determined by means of isotopic exchange (Lar sen, 1 967). The non-labile-P is mostly composed of organic P and various mineral P compounds that are rather resistant and may release P only very slowly into the labile P pool . These definitions are very broad and it is extremely difficult to distinguish clearly between these pools from the P fractionation procedure. Phosphorus may enter into the soil solution by desorption of Pi (inorganic-P) associated with the solid phase, by mineralisation of Po (organic-P) or dissolution of lattice P (apatite). The phosphate concentration in the soil solution itself is very low and in fertile arable soils is about 1 0-5 - 1 0-4 M which is equivalent to about 0 . 3 - 3 Ilg mr! (Hossner et al . , 1 973). Fox et al . ( 1 974) showed that most annual crops reqUIre 0.2 Ilg P mr! in soil solution for optimum growth . To achieve this concentration in Sri Lankan Ultisols the P adsorbed to the soil colloids was determined to be 3 00 to 600 !-lg P g- ! soil (Loganathan and Fernando, 1 980). Thus the amount of P in soil solution in these soils is about 3 000 to 6000 times less than that in the labile P fraction at field capacity moisture content. In acid soils, soil solution P exists as HPO/- and H2P04-. The ratio of these two ionic species in the soil solution is strongly pH dependent . The increase of H+ concentration shifts the equilibrium to the more protonated form according to Equation 2 .2 . HPO/- + W ¢:> H2P04- . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . . [Equation 2 . 2] 2.3.2 Characterisation of soil-P forms 1 7 The concentration of the vanous soil P fractions can be determined by u smg sequential extraction methods, which removes labile P first followed by the more stable P forms. A widely used sequential extraction method was developed by Chang and Jackson ( 1 95 7) . This sequential procedure is based on consecutive extractions with different chemical reagents, each designed to remove a particular form of P . It employs �CI to extract labile-Pi followed by NI-LF to dissolve specifically P associated with Al. This is followed by a NaOH extraction to dissolve Fe-bound Pi, and finally dithionite-citrate extraction to dissolve occluded Pi forms. HCI is used to dissolve Ca-bound Pi and the final residue is analysed by Na2C03 fusion for the determination of total P. This procedure presented many interpretational problems such as the precipitation of Pi during fluoride extraction, the separation of Al and Fe associated Pi, and the ill-defined nature of the reductant-soluble or occluded P (Williams and Walker, 1 969). A modified P fractionation was developed by Hedley et al. ( 1 982b) and a general description of the various soil P fractions separated by this method is shown in Table 2 . 5 . This sequential extraction aimed at quantifying labile-P (plant available-P), Fe + Al-associated Pi, Ca-associated Pi, as well as labile and more stable forms ofP o. No information is available in the literature on the concentration of inorganic P fractions in tea soils determined according to the procedure of Hedley et al. ( 1 994) but information on P fractionation according to the Chang and Jackson ( 1 957) method is available and this is presented in Table 2 .6. The results reveal that the majority of P in tea soils is found as organic-P and Fe and Al bound P (Bhattacharyya and D ey, 1 978; Golden et ai . , 1 98 1 ; Yongming et aI . , 1 989). The concentrations of the Fe-P fraction in all soils were higher than the Al-P concentrations. These fractions are traditionally considered as plant unavailable and very little is known with respect to the relative importance of these fractions in supplying P to tea. The magnitude of H2S04-Pi (apatite-P) in tea soils depends on the history of PR fertiliser application Table 2.5 The chemical nature of the various soil P fractions (Hedley et aI. , 1 982b) P fraction Chemical nature of P Resin-P Inorganic P in solution and weakly sorbed P NaOH-Pj Inorganic P associated with Fe and AI hydrous oxides NaOH-Po Organic P associated with Fe and AI hydrous oxides H2SO4-Pi Predominantly calcium phosphates or "apatite like P" Residual-P Recalcitrant inorganic and organic P 1 8 Table 2.6 Phosphorus fractions (Jlg P g' ! soil) in tea soils of different countries determined according to the method of Chang and Jackson ( 1 957) Country AI-P Fe-P Ca-P Organic Reductant Solution Total Reference P soluble P P P Sri Lanka Golden et al . ( 1 98 1 ) St . Coombs (up country) 59 1 64 48 344 6 1 5 Kottawa (low country) 30 3 6 1 8 148 232 India 1 Dey and Bhattacharyya Upper Assam 45 98 54 1 86 225 608 ( 1 980) Mid and Lower Assam 39 7 7 52 1 29 1 69 466 Cachar 40 6 8 47 2 1 1 1 74 540 Dooars 1 28 1 82 202 3 5 5 292 1 1 59 Darjeeling 63 1 3 2 89 392 3 14 990 China 2 Y ongming et al. ( 1 989) Yinde 68 1 99 3 1 1 83 3 .0 484 Nanchang 1 1 1 09 1 9 258 0 .8 398 Guiyang 5 8 3 8 208 0 .4 3 04 Langxi 5 40 6 1 00 0.8 1 52 Changsha 24 3 0 5 1 9 283 3 . 8 635 ! The P fractions for Indian soils were reported in units of kg ha·'. These units were converted to Jlg P g' ! soil assuming that the bulk density of soil = 1 . 1 Mg m·3 . 2 P fractions reported are for medium yielding tea plantations. 2 0 (rates and duration of application and on the length of time taken following the application of phosphate rock prior to the collection of soil samples for analysis) because it mainly extracts the undissolved phosphate rock in these highly acid soils. The residual P fraction is less important from the plant nutrition point of view, because it is very resistant to further decomposition and P release to plants. 2.3.3 The concept of plant-available P S ince plant-available P is more a fundamental concept than a measurable quantity, its determination in soil is not a straight forward task. Only a measure of P that relates to the pool of P that is plant available can be measured. Such relationships have been developed by regressing measured soil P concentrations against plant yields or plant P concentration values for various crops (Chien et a! . , 1 990a; Loganathan and Nalliah, 1 977; Naidu et aI . , 1 99 1 ) including tea (layman and S ivasubramaniam, 1 980). The plant available P is a very small percentage of the total soil P pool and it is related to the quantity of P that can be potentially utilised by crops growing in the soil. Generally "extractable" may be a term preferred to "plant available" because the latter term implies that a certain quantity of P is present in a soil and can be absorbed by plants at a particular time. Practically this is not the case, because the quantity of P utilised by plants growing in a soil may be greater or less than that of the concentration of extractable P on a sample of soil collected at a particular time. However soil testing procedures can provide an accurate "relative index" of the quantity of P that plants may utilise from a soil, but really, can not provide an absolute measure of it. Therefore the term "extractable" is rather more appropriate than "available" soil P when plant P uptake is considered. Tea p lants may differ from other crops in their ability to extract P from soils due to differences in their root system caused by clonal variations, V AM (vesicular arbuscular mycorrhizae) associations, growth rates and root secretion of organic acids etc : . As tea is adapted to acidic conditions in highly weathered soils containing P­ fixing soil minerals it is believed to have specific mechanisms by which it is able to utilise P that is normally considered non-available to other crops. 2.3.4 Soi l tests for plant-available P 2 1 An agronomic test for plant-available P should be designed to be simple enough for routine application. It should extract sufficient P from the plant-available pool and at the same time it should not extract significant amounts of P that are not plant­ available. The principle used to achieve this is by using moderately acidic or alkaline extractants, which are able to release P that is not strongly sorbed to the soil mineral phase. Specific anions are introduced to compete with P for sorption sites, or alternatively to decrease the solubility of cations (e.g. AI, Fe) that bind P to the soil. The most common methods used to determine plant-available P are the alkaline bicarbonate extraction method of Olsen et al . ( 1 954) developed for alkaline soils and the acid ammonium fluoride extraction method of Bray and Kurtz ( 1 945) developed for acid soils. The bicarbonate extraction method, though developed specifically for alkaline soils, has been used successfully on a wide range of both acidic and alkaline soils for several crops having widely different growth characteristics (Loganathan et aI. , 1 982; Smyth and Sanchez, 1 982a) . A rationale for the use of the bicarbonate anion as an extractant is that plant roots produce CO2 which forms bicarbonate in the soil solution of calcareous soils . Therefore bicarbonate simulates the action of plant roots, thus giving a more appropriate measure of plant-available P . The acid ammonium fluoride method o f Bray and Kurtz ( 1 945) has been widely used on acid and neutral soils (Fixen and Grove, 1 990; Tiessen and Moir, 1 993). The relatively low acid strength and the extraction mechanism make this method unsuitable for calcareous or strongly alkaline soils, which would partially neutralise the acidity and eliminate the standard test conditions. There are two Bray and Kurtz extraction methods ; Bray- l uses 0 .03 M �F + 0 .025 M HCl ( 1 : 7 soil : solution) with a 5 minute shaking period and Bray-2 uses 0 . 03 M NlitF + 0. 1 M HCl (1 : 7 soil : solution) with 30 minutes shaking (Bray and Kurtz, 1 945). Bray- l test extracts mostly AI bound P whereas the Olsen test removes in addition to AI bound P, Fe bound P (Le Mare, 1 99 1 ) and some organic-Po Therefore, the Olsen method may provide higher P values in acid soils compared to the Bray- l test, especially in tea soils which have higher organic-P and Fe-P than AI-P concentrations (Table 2 .6) . 2 2 In another approach to determining plant-available P, a sink for solution Pi in the form of an anion-exchange resin is used to extract P from soil : water suspensions. The P sorbed by the resin is subsequently desorbed using NaCl and measured. The amount of P measured is assumed to simulate root extraction of P from soil. S everal different methods of resin extraction have been developed and tested, using different anionic forms, soil : water : resin ratios and times and methods of shaking, enclosure in bags or mixing with the soil suspension (Barrow and Shaw, 1 977) . A simplest method uses Teflon based anion and cation exchange resin membranes, which can be cut into strips and used repeatedly (Saggar et aI. , 1 990). This method has shown to be superior to the Olsen method in predicting P availability from pasture fertilised with phosphate rocks (Saggar et al . , 1 992a) Soil P tests used in tea soils are mostly the tests that are suitable for acidic soil conditions. Bray- l extraction method was reported to be used in India (Dey and B hattacharryya, 1 980) and China (Yongming et aI . , 1 989) for routine analysis of plant-available P and borax extraction at pH 1 . 5 is currently being used in Sri Lanka (Jayman and S ivasubramaniam, 1 980). Jayman and Sivasubramaniam ( 1 9 80) reported that Borax-P was highly correlated with leaf P concentration in tea but P extracted by other extractants failed to correlate with leaf-P concentration. They however did not report what the other extractants were (see later discussion on the borax test in Chapter 7) . 2.4 PHOSPHORUS NUTRITION OF TEA WITH SPECIAL REFERENCE TO SRI LANKA When compared with nitrogen (N), P nutrition of tea plants received scant attention in the past. Although the quantity of N required by tea is more than any other soil nutrients, sub-optimal levels of P in soil can cause reductions in tea yields (Eden, 1 976). As P is very mobile within the plant and translocated to the sites where photosynthetic activity is high, it is found in high concentrations in young shoots in tea plants. These young shoots are regularly harvested at 4 - 10 day intervals in commercial plantations and consequently tea removes large amounts of nutrients (40 2 3 kg N, 1 3 kg K and 4 kg P per 1 000 kg-l processed tea) from the soil. The quantity of nutrients assimilated by different parts of the tea plant in Sri Lanka is shown in Table 2 . 7. A comparison of P removal by tea with other perennials and a range of annual crops is presented in Table 2 . 8 . Tea removes less P per ha compared to most short­ term agricultural crops (annuals) because the short-term crops need large amounts of P in a short period for their rapid growth. The other perennial crops reported in Table 2 . 8 however removed similar or less amounts ofP compared to tea with the exception of sugarcane and coconut. Sugarcane removes a large amount ofP from soil due to its higher rate of bio-mass production compared to the other perennials . 2.4.1 Chronology of P fertiliser use on tea in Sri Lanka The importance of P fertilisers in tea cultivation was realised quite early in Sri Lanka (Eden, 1 934) . As early as 1 896 some tea growers realised that fertiliser application increases yield . At early stages the main fertiliser sources were of organic origin i .e . oi l seed cakes and bones (Eden, 1 934). Later, fertiliser use became very popular, but it did not appear to be properly organised, and fertilisers were applied at most biennially (Eden, 1 934). The first statistically designed field experiment on tea in Sri Lanka was laid down in 1 93 1 by Eden (Eden, 1 934). The objective of the experiment was to determine the yield responses of mature tea plants to varying levels of N, P and K using two 3 3 factorial experiments. The findings of these experiments led him to recommend NPK fertiliser mixtures for Sri Lankan tea plantations. Subsequently these fertiliser mixtures were changed from time to time by other workers based on other field and glasshouse trial results. The chronological order of the fertiliser recommendations for tea in Sri Lanka with special reference to P is presented in Table 2 .9 . The fertiliser mixtures used for tea in Ceylon (now Sri Lanka) before 1 93 0s' were a pruning mixture high in N and P and a general mixture high in N. The latter was applied for subsequent manuring operations in the pruning cycle. At that time the u se of large quantities of P and K in the pruning mixture was considered important for Table 2.7 Mean dry matter distribution and nutrient removal by 1 00 kg processed tea and l per ha y(1 of /and in Sri Lanka (after Eden, 1 949) Organ Mean dry matter N P K distribution (%) in the plant kg per 100 kg kg ha-1 yr-l kg per 100 kg kg ha-! yr-1 kg per 1 00 kg processed tea processed tea processed tea Tips2 3 Flush3 2 1 4 . 0 7 1 0 . 4 6 1 . 3 Foliage 23 2 . 7 48 0 . 2 3 1 . 1 Wood 53 2 .4 42 0 . 3 5 1 . 5 Total 1 00 9 . 1 1 6 1 0 . 9 1 4 3 . 9 Permanent removal 6 . 4 1 1 3 0 . 7 1 1 2 . 8 1 Processed tea yield o f 1 76 1 kg ha· 1 y(l (national average) was used in the calculation 2 young shoots that are removed at the beginning of t he pruning cycle to maintain an easy and even plucking table for harvesting 3 bud and two leaves removed at harvest kg h a-l yr-! 23 1 9 27 69 50 2 5 Table 2.8 Comparison of P removal by different agricultural and tree crops grain/dry P removal Species Age matter yield by plants (kg ha-l) or from soil per Reference (yr) dry matter year or accumulation cropping season (kg ha-l yr-l) (kg ha-l ) Annual crops Corn < 1 5 ,000 1 5 Hanway and Olson ( 1980) Sorghum < 1 4,000 1 0 Hanway and Olson ( 1 980) Soybean < 1 1 ,800 1 3 Hanway and Olson ( 1980) Wheat < 1 2,400 9 Hanway and Olson ( 1980) Oats < 1 1 ,600 7 Hanway and Olson ( 1 980) Barley < 1 1 ,900 7 Hanway and Olson ( 1980) Peanut < 1 2,763 9 Nelson ( 1 980) Rice < 1 4,980 1 3 Nelson ( 1 980) Tobacco < 1 1 , 1 20 1 5 Panabokke ( 1 967) Forest Trees Radiata pine 0- 1 0 1 7,000 3 . 5 Will ( 1 968) Radiata pine 1 0-35 1 5,600 0 .5 Will ( 1 968) Loblolly pine 1 0- 1 5 7,000 1 . 3 Switzer and Nelson ( 1 972) Jack pine 20-30 2 ,900 0 . 2 Foster and Morrison ( 1 976) Scots pine 20-30 2 ,400 1 . 3 Malkonen ( 1 974) Plantation crops Mature tea 1 ,000 6 .8 Eden ( 1949) Rubber 1 , 1 20 3 .4 Panabokke ( 1 967) Coconutl 3,000 1 0.3 Panabokke ( 1 967) Sugarcane 54,300 27 Nelson ( 1 980) I yield refers to nuts with husk Table 2.9 Chronology ofP fertiliser recommendations for tea in Sri Lanka Year Composition of composition N P K eq uivalent kg ha-I yr-I Reference fertiliser mixture (parts by weight) N P K Pre- 1 930 Pruning mixture Fish guano 220 Blood meal 70 Nitrate of potash 80 3 5 2 9 24 Eden ( 1 934) Phosphate rock 80 Concentrated superphosphate 50 Pre- 1 930 General mixture Ground nut cake 200 Fish guano 200 Blood meal 1 00 68 1 7 23 Eden ( 1 934) Ammonium sulphate 1 00 Muriate of potash 50 Superphosphate 1 00 1 946 Tsoo basic m ixtu re Ground nut cake 43 0 34 9 6 Norris ( 1 946) Saphos phosphate l 60 Muriate of potash 1 0 1 952 T 500 fertiliser m ixture Ammonium sulphate 320 74 1 5 3 5 Lamb ( 1 952) Saphos phosphate 1 05 Muriate of potash 75 1 a blend of phosphate rocks from Egypt ( 1 2 _ 1 % total P) Table 2.9 continued Year Com position of composition N P K equivalent kg h a-1 -1 yr Reference fertiliser mixture (parts by weight) N P K Mg 1 96 1 T700 fertiliser mixture Ammonium sulphate 500 1 03 1 2 5 0 Tolhurst ( 1 96 1 a) Saphos phosphate 1 00 Muriate of potash 1 00 1 96 1 Tns fertiliser mixture Ammonium sulphate 500 1 03 1 2 62 Tolhurst ( 1 96 I a) Saphos phosphate 1 00 Muriate of potash 1 25 1 96 1 T750 fertil iser mixture Ammonium sulphate 500 1 03 1 2 7 5 Tolhurst ( 1 96 1 a) Saphos phosphate 1 00 Muriate of potash 1 50 1 96 1 T200 fertiliser mixture 1 st yr Ammonium sulphate 1 00 1 24 36 74 22 Tolhurst ( l 96 1 b) Saphos phosphate 50 2nd yr Muriate of potash 25 1 5 5 45 93 27 Kieserite 25 1 983 T750 fertiliser m ixture 3 -4th yr Ammonium sulphate 500 Wickremasinghe Saphos phosphate 1 00 240 27 1 1 5 1 7 and Krishnapillai Muriate of potash 1 00 ( 1 986) Kieserite 50 Table 2.9 continued Year Composition of composition P rocessed fertiliser m i xture (parts by tea weight) Y ield slab kg ha-I yr-1 1 98 3 U346 Urea 1 74 < 800 Eppawala phosphate rock ( 1 4 . 5 % total P) 72 Muriate of potash 1 00 1 983 U709 Urea 438 800- 1 000 Muriate of potash 1 03 1 800-2000 Eppawala phosphate rock 1 68 2500-3000 N P K equivalent kg ha-1 yr-1 N P K 80 9 50 1 20 8 5 0 220 1 6 90 3 00 22 1 24 Reference Wickremasinghe and Krishnapillai ( 1 986) Wickremasinghe and KrishnapiUai ( 1 986) N CX! 2 9 recovery of the tea bush after pruning. Application of superphosphate at the rate of 49 kg P ha-1 yr-l was on record at that time (Eden, 1 934). During the period of the 2nd world war, inorganic fertiliser imports were affected and the u se of NPK fertiliser mixtures got drastically reduced. A basal fertiliser mixture T 500 containing low levels of K and P compared to N was introduced during that time (Norris, 1 946). This mixture was used till 1 950s and it resulted tea crops in some estates becoming K deficient (portsmouth, 1 953) . After the war, regular supply of NPK fertilisers for tea was restored. A balanced NPK fertiliser mixture to replace the amounts of nutrients removed by the harvested crop was given priority at that time. Eden ( 1 949), by considering the relative distribution of plant nutrient concentrations in tissues and the dry matter weights of the tissues estimated the amounts of NPK removed in 1 00 kg of tea crop (Table 2 . 7) . He assumed that the flush and wood were permanently removed at the end of the cycle. On that basis a well managed privately owned tea land in Sri Lanka with an average yield of 2442 kg ha-1 y(l of processed tea (Table 2 . 1 ) was estimated to remove 1 56 kg N, 1 7 kg P and 70 kg K ha-1 y( l at harvest. Thus the amount of P that is removed from the soil is rather low compared to the other major nutrients. In 1 952, Lamb ( 1 952) proposed a new fertiliser mixture T500 on the basis of nutrient removal by the tea crop. At that time it was felt that fixation of phosphate into insoluble forms could be a serious problem. As a remedy, saphosphosphate (a blend of Egyptian phosphate rocks) in the mixture was provided in double the quantity of P that was likely to be removed in the crop and prunings. Afterwards, Tolhurst ( 1 96 1 a) found that phosphate fixation cannot be a serious problem in tea soils and it did not limit P supply to tea plants. He introduced a fertiliser mixture T 700, which was still above the replacement level for P in crop and prunings, but lower than that provided by T 500 . He recommended T 700 for up-country tea soils which supplied 1 2 kg P and 50 kg K for every 1 00 kg of N applied. He suggested modified fertiliser mixtures with higher proportions of K for mid and low country tea soils to arrest K deficiency caused by leaching loses which are common in these soils . These fertiliser mixtures were T 725 and T 750 having 25 and 50% more K than T 700. 3 0 The need for a young tea fertiliser mixture came to light with the popularity of VP clonal tea plants, which are markedly fertiliser responsive. To obtain high growth rates and best yields Tolhurst ( 1 96 1 b) recommended T 200 mixture containing higher levels of P and K in relation to N for young tea because VP plants were reported to have a higher yield potential associated with higher fertiliser responsiveness compared to seedling tea (Anandappa, 1 986). Subsequently, based on the results of factorial NPK fertiliser trials and economics, Tolhurst ( 1 965) suggested that P could be further reduced to 1 0 kg P ha-1 y(l or even omitted for one or two years. This may be due to annual application of P fertiliser over long period, which led to a build-up of P reserves in these soils. Fernando et al. ( 1 969) also showed that for high yielding tea clones, P and K need not be increased in the same proportion as N. They recommended application rates of 9 and 1 3 kg P ha-1 y(l for seedling and clonal tea respectively. In 1 960' s sterameal (an organic fertiliser) was used to supply P for tea nurseries. Visser and Khel ( 1 96 1 ) compared the efficiency of inorganic P mixtures (superphosphate) with that of organic mixtures containing sterameal having the same composition of N, P and K on nursery tea plants. They found that there was no significant difference between the organic and inorganic fertiliser treatments in tea growth and reported that the inorganic mixture was preferable due to its cost­ effectiveness and easiness of application in nurseries. Subsequently, Tolhurst and Visser ( 1 96 1 ) found that having superphosphate in the mixture was a disadvantage due to its reactivity with soluble ammonium sulphate and potassium sulphate to form an insoluble residue of calcium sulphate. This caused several practical difficulties, including fertiliser storage and deposition of fertiliser residues on the foliage. They remedied this by substituting monoammonium phosphate for superphosphate and recommended a completely soluble nursery mixture T65 (Table 2 . 1 0) . This mixture was found to be very effective for tea nursery plants and is still widely accepted among tea growers in Sri Lanka. A few years later due to a shortage of monoammonium phosphate in the country, Tolhurst and Richards ( 1 965) revised T65 nursery mixture and recommended T55 instead (Table 2 . 1 0) . They suggested mixing of superphosphate with soil at the rate of 222 - 446 g m-3 of nursery soil to supply the Table 2. 10 The fertiliser mixtures u sed for tea nurseries in Sri Lanka Year Composition of Composition Nutrient composition Reference fertiliser m ixture ( parts by (%) of the mixture weight) ( N : P : K: Mg) 1 96 1 T6s Ammonium sulphate 1 5 1 0 .9 : 4 . 7 1 : 9 . 1 7 : 2 . 2 1 Tolhurst and Visser ( 1 96 1 ) Monoammonium phosphate 20 Potassium sulphate 1 5 Epsom salt 1 5 1 96 1 Tss Ammonium sulphate 3 5 1 3 . 1 1 : 0 : 7 . 2 1 : 1 . 74 Tolhurst and Richard ( 1 965) Potassium sulphate 1 0 Epsom salt 1 0 1 97 7 TS7 Ammonium sulphate 3 0 1 0 . 84 : 0 : 9 .04 : 2. 3 5 Ayadurai and Sivasubramaniam ( 1 97 7) Potassium sulphate 1 3 Epsom salt 1 4 3 2 basal requirement of P for tea at the nursery, followed by spraying a solution of T 55 containing N, K and Mg. In 1 977, Ayadurai and Sivasubramaniam ( 1977) formulated a nursery mixture T57 with the same components as in T 55 but with higher proportions of K and P than in T 55. Though Tolhurst and Richards ( 1 965) recommended mixing of triple superphosphate along with TS5 as an alternative to T65, Ayadurai and Sivasubramaniam ( 1 977) showed that mixing Eppawala phosphate rock (EPR, a locally mined apatite deposit; see section 2. 5) with T 57 is a better substitute than the former because EPR was cheaper. Furthermore they did not notice any significant difference between the soluble P fertiliser and EPR on dry matter and P uptake in tea plants. Until the end of 1 970s, the main source of N fertiliser used for tea in Sri Lanka had been ("NH4)2S04. Field experiments comparing the effect of different forms of N fertilisers on tea yield had shown that there was no significant yield difference between urea and ("NH4)2S04 treatments (Bhavanandan and Manipura, 1 969; Sandanam et aI. , 1 980; Watson and Wettasinghe, 1 972) . Based on these results, urea based fertiliser mixtures (U346 and U709) were recommended for mature tea (Wickremasinghe and Krishnapillai, 1 986). During the period of the 1 970s a urea plant was built in Sri Lanka, so urea was available locally. The use of locally produced urea was found to be much cheaper than the imported ("NH4)2S04. Studies were also undertaken to test the local EPR as a source of P for tea. Sivasubramaniam et al. ( 1 98 1 ) examined the chemical characteristics of this phosphate rock and found that it contained a higher total P content ( 14 . 5%) than the imported phosphate rock (saphosphosphate) ( 1 2 . 1%) (See section 2. 5) . Subsequently EPR was recommended as the source of P in urea based fertiliser mixtures for tea (Sivasubramaniam et aI. , 1981 ) . 2.5 EPPA W ALA PHOSPHATE ROCK (EPR) 3 3 EPR i s now an important source of P for tea plantations. For this reason the nature of EPR and factors affecting the plant availability of P from PRs are reviewed. In 1 97 1 the Geological Survey Department of Sri Lanka discovered a phosphate rock deposit, estimated to be about 40 million tonnes at Eppawala in the North Central Province of Sri Lanka (Jayawardena, 1976) . The weathering of the apatite rich parent carbonate rocks and associated gneisses and migmatites (rocks consisting of thin alternating layers or lenses of granite type and schist) at this site has given rise to a thick non­ carbonate and phosphate rich weathering profile (Dahanayake et al . , 1 995). This profile is being mined to exploit its phosphate ore. It has two dominant minable components that can be recognised with the naked eye. (i) greenish primary apatite crystals and (ii) brownish aluminous-ferruginous-siliceous secondary phosphate matrix. The X-ray diffraction analysis of the primary apatite crystals reveals the presence of hydroxyl chlorapatite with unit cell a-axis dimensions ranging from 9.46 A to 9.80 A. Fluorapatite or Francolite (carbonate apatite with unit cell a-axis values from 9 .35 A to 9 . 37 A), a more soluble variety resulting from weathering, is also found within and around the primary apatite crystals. The primary crystals have P contents varying from 1 5 .3 - 1 8 . 3%. Their solubilities in 2% citric acid varies from 2 .3 to 2 . 5% P. Neutral ammonium citrate (NAC) solubility varies from l .0 to l .2 % P (Dahanayake et aI. , 1 995) . The aluminous-ferruginous-siliceous secondary phosphate matrix (dominantly formed of fluorapatite with accessory hydroxyl apatite, crandallite, geothite, hematite, ilmenite, laterite and quartz) constitutes about 30 to 60% of the total deposit depending on the location of the deposit. The P content in the secondary phosphate matrix can vary from 0 t014. 5o/o(Dahanayake et aI. , 1 995). The solubility of this matrix in 2% citric acid varies from 1 . 5 to 2 . 1 . 3 4 The product available i n the market i s a mixture of the primary apatite crystals and the secondary phosphate matrix. The chemical composition of these two fractions is given in Table 2. 1 1 . 2.6 FACTORS AFFECTING P AVAILABILITY FROM PR 2.6. 1 Characteristics of PRs McClellan and Gremillion ( 1 980) classified PRs into three broad classes according to their mineralogical composition; Fe-AI phosphates, Ca-Fe-AI phosphates and apatite group of minerals . In terms of the weathering sequence, Fe and AI phosphates are often the most weathered stable end products in a sequence, wherein apatites are the least weathered (Khasawneh and Doll, 1 978). The apatites in igneous and metamorphic PR deposits are relatively inert, being coarse-grained with few internal surfaces. They contain fewer impurities as accessory minerals . Thus, the P content of these deposits is relatively high. This factor is important in the production of soluble and partially soluble P fertilisers from PR, but is of no concern when determining the reactivity of the PR when applied directly to soil as fertiliser. Sedimentary PRs, on the other hand, contain apatite minerals that are microcrystalline. They consist of fairly open, and loosely consolidated aggregates of microcrystals with relatively large specific surface areas. Usually they are carbonate apatites and also contain varying amounts of accessory minerals (Khasawneh and Doll, 1 978). The PRs of sedimentary origin are chemically more active than those of igneous and metamorphic origin (Sanyal and De Datta, 1 99 1 ) and so can be used directly as a fertiliser on acid soils receiving high rainfall (White et ai . , 1 989) . The effectiveness of PRs differs widely in chemical reactivity or solubility in acid soils due to differences in mineralogy and chemistry. Isomorphous substitution of a divalent planar C03 group for a trivalent tetrahedral P04 group creates a charge imbalance and a packing void, which weakens the crystal structure of apatite. It decreases the a value from 9 . 37 A, which is a characteristic of pure fluorapatite. This results in a reduction of crystal size and an increase in the specific surface area of the apatite. As the number of moles of Table 2. 1 1 Chemical analysis of the primary apatite and aluminous-ferruginous­ siliceous phosphate matrix ofEppawala phosphate rock deposit (Dahanayake et aI . , 1 995) Composition (%) Primary apatite Matrix p 1 7. 69 1 4. 5 1 Ca 38 .68 3 1 . 1 6 F 2 .03 2 .07 Si 0 .08 0 . 07 AI 0 .02 0 . 77 Fe 0 .22 8 . 36 Na 0 .09 0 . 1 5 K <0.0 1 <0. 0 1 Mg 0 .06 0 . 1 0 CI 1 . 86 1 . 1 2 Total S 0.27 0 .20 Mn 0 .0 1 0 . 1 8 Ti <0. 0 1 0 . 62 (AI + Fe) = R 0 .24 9. 1 3 3 5 3 6 OH or C03 per mole of P04 increases, the standard free energy of reaction of the apatite with acidic solutions becomes more negative, and its dissolution increases (Bolan et al. , 1 993; Chien, 1 977; Khasawneh and Doll, 1 978; McClellan and Gremillion, 1 980). Therefore the degree of C03 substitution is considered to be one of the important factors which determines the solubility of apatite and the dissolution rate ofPR in acid soils (Caro and Hill, 1 956). The most reactive PRs are those having a molar POJC03 ratio less than 5 (Hedley et at, 1 990). As the extent of C03 substitution in an apatite molecule is difficult to measure, the chemical reactivity of PRs are empirically assessed according to their solubility in selected chemical extractants. The solubility of PR in 2% citric acid, 2% formic acid (Chien and Hammond, 1 978ab) and neutral ammonium citrate (McClellan and Gremillion, 1 980) has been used in New Zealand, Europe and USA and Australia, respectively as a measure of the reactivity of PRs. In New Zealand PRs dissolving 2::3 0% of the total P in 2% citric-acid are classified as reactive PR and the others as less reactive PR (White et aI . , 1 989). The EPR in S ri Lanka has 2% citric acid solubility of 1 . 5% P (Dahanayake et al . , 1 995) and NAC solubility of <2% P (Riggs and Syers, 1 99 1 ) . The % of total P in EPR soluble in 2% citric-acid is about 1 4%. Therefore EPR is classified as a less reactive PR according to the New Zealand classification. Many of the PR sources in tropical countries have been shown to be relatively unreactive according to numerous agronomic trials (Hammond et al . , 1 986b). The dissolution of PR occurs at the surface of the PR particle (Barrow, 1 990). Decreasing the particle size by fine grinding will increase the surface area of PR particles exposed to the soil and therefore it increases the rate of P release from the PR, but fine grinding is not a substitute for reactivity. Reducing PR particles to a size less than 1 00 mesh ( 1 50 )lm) by grinding is generally not warranted as finer particles do not increase agronomic effectiveness greatly (Khasawneh and Doll, 1 978). The use of finely ground PRs may result in handling problems for mechanised agriculture, but would pose fewer problems for the more labour intensive agricultural systems like tea cultivation in developing countries. 3 7 2.6.2 Soil properties The major component of most PRs is apatite, the dissolution of which in an acid soil can be described by the following equation (Khasawneh and Doll, 1 978) . CalO(P04)&2 + 12 I1 <=> 10 Ca2+ + 6H2P04- + 2F . . . . . . . . . . . . . . . . . . . . . [Equation 2 . 3 ] According to this equation, the driving force for the dissolution of PR in soils is the supply ofI1 and the removal of reaction products Ca2+, H2P04- and Ffrom the site of dissolution. The ability of the soil to provide I1 to drive the dissolution process is therefore essential for the agronomic effectiveness of PR. Many field and laboratory studies show that low soil pH (Bolan and Hedley, 1 989; Kanabo and Gilkes, 1 987; Rajan et aI . , 1 996; Tambunan, 1 992; Utomo, 1 995; Zaharah and Sharifuddin, 1 995) with high P buffer capacity (Rajan et aI . , 1 996; Smyth and Sanchez, 1 982b) contribute to enhanced PR dissolution. In high P-fixing allophanic soils in New Zealand, Bolan and Hedley ( 1 990) reported an increase in P dissolution, from 29.3% to 8 3 . 5%, 1 8 .2% to 78 .9% and 1 2. 5% to 60. 3% in North Carolina phosphate rock, Jordan phosphate rock and Nauru phosphate rock respectively, by decreasing pH from 6 .5 to 3 .9. Smyth and Sanchez ( 1 982b) observed that dissolution rates of North Carolina PR and Patas de Minas PR increased in a group of acidic Oxisols in Cerrado, Brazil with increase in P-sorption capacities. They showed that the maintenance of low concentrations of P in soil solution is considered as an important driving force for PR dissolution. Tea soils certainly have high acidity (pH 4 . 5 - 5 . 5 in water, Table 2 .4) as well as high P fixing capacities (Gohain, 1 988; Golden et aI . , 1 98 1 ) to promote PR dissolution. Although PR dissolution increases with increases in P fixation the agronomic effectiveness of the PR, measured relative to triple superphosphate (TSP) or single superphosphate (SSP), may not be higher, but may be less, in a high P fixing soil than in a low P fixing soil because the fixed P is not readily available to plants (Hammond et aI. , 1 986a). The mass action of ions in equation 2 .3 also indicates that PR dissolution would be favoured by soil conditions that maintain low concentrations of Ca already dissolved in the soil solution. Mackay and Syers ( 1986) showed that low levels of exchangeable 3 8 C a associated with reduced Ca saturation of the soil' s CEC, which in turn would lower Ca concentration in the soil solution, promote an increase in PR dissolution. Acid tea soils have low levels of exchangeable Ca and low Ca saturation which i s expected to increase PR dissolution. 2.6.3 Crop species Phosphate rocks (PR) are considered to be more effective in supplying P to perennial plantation crops like tea, coffee, rubber, sugarcane and pastures compared to short­ term annual crops such as wheat, maize or millet (Chien et al. , 1 990a; Sale and Mokwunye, 1 993). Examples of previous studies carried out on P utilisation from different PRs by various crop species and the major conclusions made in these studies are presented in Table 2 . 1 2 . Numerous studies indicate that annual crops require a higher rate of P supply during their rapid vegetative growth phase compared to perennials, and this need is readily met by the P supplying characteristics of water soluble P fertilisers rather than the sparingly soluble PRs (Bolan and Hedley, 1 995; Palmer and Jessop, 1 982). Unlike annual crops, perennial crops like tea would obtain their P requirement over a considerably longer period at a slower rate and this requirement can be met by application of selected PR fertilisers. Furthermore, the slow rate of P release from PR may be an advantage in high P fixing soils, because a significant % of P released from the PR fertiliser is taken-up by the crop before it gets fixed to any significant degree compared to soluble P fertilisers where a higher proportion ofP may get fixed (Chien and Menon, 1 995) . Actively growmg plant roots are reported to have a stimulating effect on PR dissolution. The root induced acidification, where If" build up around the root surface, will result in an increase in the rate of dissolution ofPR particles (Aguilar and Van Diest, 1 98 1 ; Bolan et aI . , 1 997; Hinsinger and Gilkes, 1 995 ; Paauw, 1 965 ; Trolove et aI. , 1 996b). The plant induced processes of acidification will be discussed in a separate section in this review. Plants with higher root densities in the surface layers of the soil where the PR is concentrated would be expected to acquire Table 2.12 Examples of previous research conducted on P utilisation from phosphate rocks by different plant species Plant species Phosphate rock Observations Reference Oats Florida Soil pH of 6 .0 or below is Ellis et a1. ( 1 95 5) necessary for satisfactory utilisation ofPRs Wheat and red Florida P uptake from PR by Murdoch and clover clover is greater than Seay ( 1 955) wheat Buckwheat and Tunis, Morocco, Availability of P from PRs Armiger and Fried alfalfa Curacao, South was related to the ( 1 957) Carolina, Florida, carbonate content of the Idaho, Tennessee, PRs Montana, Virginia Potato Gafsa and Florida P availability from PR Paauw ( 1 965) increased with decrease of soil pH Wheat, rye, Reno and Langfos P availability from PR is Diest et al . ( 1 97 1 ) lettuce, barley, greater for dicotyledons cauliflower, than monocotyledons. The cabbage, maize, uptake of Ca by the pearl millet, former is an important clover, lupin and factor determining their tomato ability to utilise PR Squash, soybean, Virginia Dissolution of PR is Johnston and barley and wheat related to the rate of Ca Olsen ( 1 972) removal by plant roots Wheat, paspalum, Maranhao, Buckwheat is more Van Ray and Van maize, soybean Alvorada, Mali and efficient than others in the Diest ( 1 979) and buckwheat Patos de minas utilisation of P. Plant P uptake is related to the cation : anion uptake ratio Maize Togo Ground PR increased dry Mokwunye matter yield over ( 1 979) unground PR. Acidulation ofPR with elemental S increased PR solubility and its effectiveness 3 9 4 0 Table 2.1 2 continued Plant species Phosphate rock Observations Refe rence Ryegrass Chatham Rise Dry matter yield depends Mackay et a1. on the physical form ( 1 980) (powder or pellet), rate and time of application of the PR. Powdered form was more effective in increasing dry matter yield Soybean and Morocco and Mali P uptake from PRs is Aguilar and Van alfalfa related to the Diest ( 1 98 1 ) acidification caused by the uptake of excess cation over anions Buckwheat, Mali, Mexico and Plants enhance Bekele et aI. ryegrass, Florida solubilisation ofPR ( 1 983) rhodegrass, bean through higher cation and maize over anion uptake pattern Wheat Christmas Island, P availability from PRs Kucey and Bole North Carolina, related to their solubility ( 1 984) Khouribga, Gafsa, in chemical extracts such Sechura, Central as acidic (pH 3) or Florida, Tennessee, neutral ammonium citrate Pesca, Kadjari, or 2% citric acid. Tracer Huila, Idaho, technique provided more Jhamar Kotra, reliable index of P T apira, Cargill and availability than these two Missouri extract ants Brown top, white Chatham Rise Residual effects of the PR Mackay et a1. clover, in the field depends on ( 1 984) subterranean the soil type, P sorption clover, perennial capacity. The presence of rye grass, lotus CaC03 in close proximity decrease the rate of PR dissolution due to increase pH and Ca concentration in the solution film immediately surrounding the PR particles. 4 1 Table 2 . 12 continued Plan t species Phosphate rock Observations Reference Ryegrass Huila, Pesca, Dissolution and plant P Anderson et al. Sechura, Gafsa, availability from PR was ( 1 985) North Carolina, related to the carbonate Central Florida and content of the PRs Tennessee Pueraria Morocco and Mali P uptake from PR is De Swart and Van related to the Diest ( 1 987) acidification caused by excess cation over anion uptake Rape Mali Solubilisation ofPR in Hoffiand et al. the presence of plant ( 1 989) roots i s achieved by release of organic acids by roots Maize Huila, Capinota Agronomic effectiveness Menon and Chien of partially acidulated ( 1 990) Huila PR and Huila PR compacted with TSP were similar to TSP. Partial acidulation of Capinota PR was only half as effective as TSP. Capinota PR compacted with TSP was as effective as TSP. Partial acidulation of slightly to moderately reactive PR with high F e203 content is less effective. Only way to make profitable u se of the PR is to mix with water-soluble P sources. Compaction is one such way Ryegrass, wheat, Sechura Ryegrass acquired more Chien et al . maIze P than wheat or maize ( 1 990b) from PR fertilised soil due to its higher root density 4 2 Table 2.12 continued Plant s pecies Phosphate rock Observations Reference Ryegrass Sechura, North P extracted form PRs by Rajan et al. Carolina, Gafsa, 2% citric acid, 2% formic ( 1 992) Y oussoufia, Arad, acid and neutral Khouribga, Jordan, ammonium citrate Zin, Mexico, correlated well between Nauru, Florida themselves (R2 0 .82 - 0 .99). Formic-P was the best predictor of the agronomic effectiveness ofPRs. Citric-P was a poorer indicator of the reactivity ofPRs. White lupin, North Carolina Rhizosphere acidification Hinsinger and narrow leaf lupin and removal of P and Ca Gilkes ( 1 995) by plant uptake enhanced PR dissolution. White lupin dissolved twice as much PR as narrow leaf lupin Lotus, white North Carolina Lotus had higher internal Trolove et al . clover and external P efficiency ( 1 996a) than white clover in PR fertilised soil due to greater root length Corn Sukulu Hills Dry matter yield Butegwa et al . decreased with increase ( 1 996) in P-fixation capacity of the soil. Lotus North Carolina, Release ofH+ and the Bolan et al . Jordan removal of dissolved ( 1 997) products ofPRs by plant uptake enhanced PR dissolution 4 3 dissolved P from the PR at a higher rate, because there would be a greater likelihood that the root encountering localised ' high concentration pockets' of soluble P adjacent to the PR. This proposal is supported by evidence from glasshouse pot trials using reactive Sechura PR, where ryegrass acquired more P than wheat or maize, because of its higher root density (Chien et aI. , 1 990b). Plants promote P dissolution by increasing the root density in the vicinity ofPR particles (Kirk and Nye, 1 986). This is attributed to the reductions of the concentration of Ca and P by plant uptake. In pot trials using a loess soil ( 1 .4 mg P 1 00 g-l soil), Flach et al. ( 1 987) showed that the Ca uptake p attern (finger millet > pearl millet > maize) followed the order of crop response to applied PRs (Zimpan PR from Mexico 1 5% total P, 1 5% of total P soluble in 2% citric acid; Khouribga PR from Morocco 14% total P, 25% of total P soluble in 2% citric acid) . 2.6.4 Moisture The soil water regime is an important factor controlling PR dissolution, because PRs will not dissolve in dry soils (Sale, 1 990). PR particles need to be surrounded by moisture films to enable the dissolved products such as Ca2+ to diffuse away from the dissolving surface and to permit the inward diffusion of W towards the PR surface. Incubation studies confirmed that PR dissolution declines as soil moisture content is decreased (Gregg et aI. , 1 987). Reactive PRs (RPR) are recommended for pasture in Australia and New Zealand in areas having rainfall > 800 mm with even distribution throughout the year and soil pH <6 because under these conditions RPR is expected to dissolve at a satisfactory rate to supply adequate P to plants (White et aI . , 1 989). Tambunan ( 1 992) observed that North Carolina phosphate rock (NCPR) was more effective than TSP in increasing maize yields in an acidic Ultisol of Indonesia having higher moisture content compared to soils in a dry area because NCPR dissolution was greater in soils with higher moisture content . Most tea-growing areas in the major tea producing countries receive an annual rainfall of approximately 2000 mm or more and pH <6 and therefore, RPRs or even less reactive PRs should dissolve to supply adequate amount of P to maintain soil P levels satisfactory for uptake by tea plants. 4 4 2.6.5 Management practices Phosphate rock dissolution in soils is influenced by management practices in a number of ways. Studies have shown that broadcasting and incorporating the PR in the surface soil will result in greater PR effectiveness than placing the PR in a concentrated band in the soil (Hammond et al., 1 986b; Khasawneh and Doll, 1 978). Reduction of PR dissolution when PR is placed as bands is partly attributed to the overlapping diffusion zones around the closely spaced PR particles, which hinders the movement of Ii to the surface of PR particle and movement of Ca and phosphate ions away from the particle (Barrow, 1 990). The build-up of dissolved products at the surfaces of these particles would limit PR dissolution. This was clearly demonstrated by Hughes and Gilkes ( 1 986) in their incubation studies. Liming is normally practised to reduce A1 toxicity and increase Ca2+ in acid soils. Liming will both reduce the supply of H+ and increase the supply of Ca2+ in the soil solution (Hammond et al . , 1 986b). The outcome would be a lower rate of dissolution of PRs applied to these lime amended soils. To remedy the negative effects of lime on PR dissolution, PR has to be applied long before liming. This would allow PR to dissolve before the pH and Ca status of the soil are raised by lime application. 2.7 RESPONSE OF TEA PLANTS TO P FERTILISERS The response in tea growth to P fertilisers is more common in nurseries or on young tea plants in the field than on mature tea. The higher growth rate and the smaller volume of soil explored by the roots in the early stages of growth compared to a mature stage may be the reason for the higher P demand by nursery and young plants. Green ( 1 965) showed that mixing of superphosphate with soil in the planting hole had increased the pruning weights (dry weight of first 3 young leaves and soft wood) of young plants by an average of 37% within a year compared to untreated plants. The mixing of P fertiliser benefited the plant by greater root growth, which eventually led to an increased uptake of water and nutrients by the tea bush to synthesise more dry matter. 4 5 The agronomic effectiveness of EPR was compared with saphosphosphate (a blend of Egyptian phosphate rocks) in a glasshouse trial in Sri Lanka at an application rate of 1 5 kg P ha-l y(l by measuring the yield and P uptake of 1 2 month old tea plants (clone TRl 2025) (Sivasubramaniam et aI. , 1 98 1 ) . The dry matter yield and P uptake at the end of one year of the trial did not differ significantly between any of the two P treatments or the control (no P addition) treatment. The lack of response to P fertiliser may have been due to high residual P levels in the soil as a result of lavish quantities of P applied to the estate in the past from which this soil was collected. Sivasubramaniam et al. ( 1 98 1 ) did not report the P status of the soil they used in the trial, but Wickremasinghe et al. ( 1 986) reported that these soils had 25 .6 I1g P g-l soil of borax extractable-P which is considered to be high according to the classification of Beater ( 1 949). Sivasubramaniam et al . ( 1 98 1 ) found that malic acid is a more appropriate solvent to be used to asses the solubility of PR in soil because tea roots excrete significant quantities of malic acid, which can dissolve native P sources in the soil (Jayman and Sivasubramaniam, 1 975 ; Xiaoping, 1 994). They also found that malic acid extracted more P from soils treated with EPR compared to soils treated with imported PR and this led them to conclude that EPR is equally or more effective than the imported PRo Dey and Bhattacharyya ( 1 980) showed that continuous application of P fertiliser (40 kg P ha- 1) to a sandy loam soil in Borbhetta, India over a long period (22 yrs) had resulted in an increase in leaf P concentration from 0 .28% (no P fertiliser treatment) to 0.39%. This response could be due to the low levels of plant-available P in these soils (Bray and Kurtz extractable P of 4 - 8 Ilg g-l soil). Similar results were observed by Tolhurst ( 1963) on mature tea in a long term ( 1 7 yrs) field trial in an acid Ultisol in Sri Lanka. Tolhurst ' s trial showed that both Saphosphosphate and TSP increased leaf P concentration from 0 .25 to 0 . 30% when P rates increased from 1 5 to 29 kg ha-1 y(l but there was no difference between the two P sources on leaf P concentration and shoot yield. Willson ( 1 969) estimated the amount of P required by mature tea plants yielding 5050 kg processed tea ha- 1 in Kenya on a 3 year pruning cycle (Table 2 . l 3 ) . He quantified the removal of P by crop and the amount of P circulated as leaf fall and prunings. He 4 6 Table 2.13 Circulation of phosphorus in tea plants yielding 5050 kg processed tea ha-1 in a three year pruning cycle (calculated from the data of Willson, 1 969) Plant process Organ or tissue involved Amount of P absorbed in the process or removed by plants ( kg ha-1 prun ing cycle-I) P absorption by Roots 80 plants from soil P removal from plants Prunings 59 and return to soil Leaf fall 4 P removal from estate Young shoot harvest 1 8 4 7 reported that tea plants absorbed nearly 80 kg P ha-1 during a 3 year cycle and 1 8 kg P ha-1 i s removed as flush (bud and two leaves) at harvest. From the total removal, about 75% ofP was removed with prunings at the end of the cycle. In Sri Lanka, the prunings are normally buried into tea fields as a standard practise to recycle the nutrients and therefore the P in prunings are returned to the soil. Traditionally soluble P fertilisers are used to supply P to most agricultural crops. Odhiambo ( 1 987) studied the effect of single superphosphate (SSP) and diammonium phosphate (DAP) mixed at different rates (0, 9, 1 8, 3 5 and 70 kg P ha-1) with the soil in p lanting holes on the growth of clonal tea in Kenya. He found that replanted clonal tea treated with 1 5 - 30 g SSP or 1 0 - 20 g DAP per planting hole giving an equivalent of 1 8 - 3 5 kg P ha-1 showed the highest plant vigour and survival rate. The application ofP above this range showed no benefit to the plants. Ranganathan ( 1 97 1 - 1 980) reported the results ofa field trial on mature tea in an acid soil in South India comparing soluble P fertilisers (ammonium phosphate, diammonium phosphate, superphosphate), Suphala (a NPK fertiliser mixture, 1 5 : 6 . S : 1 2 . 5) and an unnamed sparingly soluble phosphate (Table 2 . 1 4) . During the trial period some of the P treatments were changed [i .e . inclusion of dicalcium phosphate and/or Myssoorie PR (total P% 1 l . 3 w/w and citric-acid soluble P 26% of total P) instead of Suphala] . The results of this trial showed that the form of P had no significant effect on processed tea yield during three pruning cycles. The soluble P fertilisers are much expensive because their manufacture requires high capital investment and more energy. Therefore direct application of finely ground PRs to tea soils could be considered as a cheap option than the use of soluble P fertilisers. Many tea growing countries in the humid tropics are therefore attracted by the possibility of using PRs particularly those that are indigenous. In general, PR materials are found to be most effective in P-deficient acid soils (Chien and Hammond, 1 989; Rajan et aI . , 1 996). Nevertheless, PR sources vary widely in their agronomic effectiveness depending on their chemical characteristics, soil and climatic conditions and crop growth (Bolan et aI . , 1 990; Chien and Menon, 1 995 ; Khasawneh Table 2. 1 4 Effect ofP sources on processed tea yield (kg ha-! y(l ) in South India (Ranganathan, 1 97 1 - 1 980) P Source Pruning cycle-l P Source Pruning cycle-2 P Source Pruning cycle-3 1971- 1972- 1973- 1974- 1975- 1976- 1977- 1978- 1 979- 72 73 74 75 76) 77 78 79 80 Control 2070 2083 1485 Control 34 1 6 2803 3662 Control 2346 4479 4596 Suphala 22 14 2368 1 752 Dicalcium 40 1 5 3062 3882 Myssoorie 23 1 3 4736 5 1 04 phosphate PR Diammonium 2260 23 14 1 7 1 6 Diammonium 3893 2807 3740 Diammonium 2 1 87 4608 48 1 3 phosphate phosphate phosphate Ammonium 2358 24 1 8 1 734 Ammonium 3955 3033 3902 Ammonium 2384 4797 5057 phosphate phosphate phosphate Rock phosphate 25 1 3 2457 1 688 Rock phosphate 3943 2933 3770 Rock phosphate 2522 4970 53 1 4 Superphosphate 2427 23 16 1 743 Superphosphate 3876 2899 3799 Superphosphate 2493 4820 4997 CD (p=0.05) 1 73 244 2 1 4 388 1 20 147 1 48 246 3 02 Plucking 52 44 28 45 49 5 1 32 5 1 5 1 rounds y(l 4 9 and Doll, 1 978). The factors that affect P availability from PR was adequately discussed in Section 2.6 . Several key issues still remain unresolved with respect to the P nutrition of tea. Further intensive research is imperative to fill these knowledge gaps and to obtain a more comprehensive understanding of the different processes involved in P uptake by tea from soils treated with PR especially the locally available EPR. Most research by previous workers has been on the growth response of tea to applied P fertilisers, but these studies did not examine adequately the reactions of fertilisers in tea soils, mechanisms of P uptake by tea roots and the relationship between soil P forms and P uptake by tea. In order to study the mechanisms of P uptake one needs to understand the P chemistry in the zone where the roots are active i .e . the rhizosphere. No such rhizosphere studies are known to have been conducted with tea, though studies on other crops have been documented and they are reviewed in the next section. 2.8 ROOT-SOIL INTERFACE (RHIZOSPHERE) The zone immediately surrounding the root has been termed the rhizosphere and this zone has properties significantly different from that of the bulk soil (Armstrong and Helyar, 1 992; Darrah, 1 993 ; Gahoonia et aI . , 1 992; Hedley et aI . , 1 994; Hinsinger and Gilkes, 1 996; Jungk, 1 996; McKenzie et aI. , 1 995; Tarafdar and Jungk, 1 987). The term "rhizosphere" has evolved from the very narrow definition first used by Hiltner in 1 904 to describe the narrow zone of intense bacterial activity around legume roots (Darrah, 1 993). Currently it is extended to a broader definition of "the zone of soil surrounding the root which is affected by it". Depending on plant species, the width of the rhizosphere zone has shown to extend a few mm to several cm from the root surface (Bolan et al . , 1 997; Darrah, 1 993). The rhizoplane is an elaborate word for the root surface. Plant root processes influence chemical, biochemical and physical conditions within the rhizosphere. Several complex mechanisms are involved in the uptake of nutrients by the root, particularly P, from the rhizosphere. Plants preferentially absorb P, 5 0 depleting rhizosphere labile-P levels (Dorrnaar, 1 988; Hedley et aI . , 1 994; Trolldenier, 1 992; Trolove et aI . , 1 996b) . The mobility of soil P to the roots is primarily by diffusion (Barber, 1 995). The rate of plant P uptake depends on the steepness of the concentration gradient of P between the rhizoplane and the bulk soil (Barber, 1 995; Jungk, 1 996). Rhizosphere pH which is altered by the uneven uptake of cations and anions (Gahoonia et aI. , 1 992; Gijsman, 1 990abc; Haynes, 1 990; Nye, 1 98 1 ) can influence phosphate solubility and uptake (Bolan et ai., 1 997; De Swart and Van Diest, 1 987; Trolove et al . , 1 996b) . Plant roots release enzymes, such as phosphatase, which can mineralise organic-P compounds through hydrolysis (Eivazi and Tabatabai, 1 977; Gahoonia and Nielsen, 1 992; Tarafdar and Jungk, 1 987). Bacteria and fungi which are actively engaged in P transformation in soils are about 20 - 50 times more abundant in the rhizosphere compared to the bulk soil because of abundant quantities of carbon compounds produced by roots in this zone (Newman, 1 978; Rovira, 1 979). Due to the high microbial activity in this zone labile inorganic P can get immobilised onto the organic P fraction (Armstrong and Helyar, 1 992; Darrah, 1 993 ; Trolove et al . , 1 996b) . The abundant quantity of organic compounds in the rhizosphere can also influence other forms of P (e. g. P fixed to inorganic minerals) in the soil. It is now well known that soil fertility information obtained from routine soil chemical analysis do not always correlate well with data on plant growth and nutrient uptake (Binkley, 1 986; Mahendrappa et aI . , 1 986). This may be partly due to the fact that soil samples collected for routine soil analysis are missing a small, but distinctly different and important soil fraction, the soil in the rhizosphere, which greatly influences the plant nutrient availability to plants (Curl and Truelove, 1 986) . Therefore a better understanding of the P status of the rhizosphere and its relationship to the P status of bulk soils is required to interpret routine soil analytical values. Although many workers have attempted to describe some of the P transformation processes in the rhizosphere, the understanding of the intricacies of this unique zone is still at its infancy. A significant drawback limiting research on the rhizosphere is the technical difficulty of drawing soil samples from this zone. The problems are associated with sampling small amounts of rhizosphere soils for analysis and determining the line of demarcation between the rhizosphere and bulk soil . Attempts 5 1 were made by several workers to develop suitable techniques to study the rhizosphere processes in annual crops. Some examples include, separation of soil-root zone by a nylon mesh (Dormaar 1 988; Gahoonia et al. , 1 992; Helal and Sauerbeck 1 984; Kuchenbuch and Jungk, 1 982) and porous plastic envelopes where roots have no physical contact with the soil (Brown and U1-Haq, 1 984). Other methods are more practically suited to field studies. These include gentle shaking (Haussling and Marschner, 1 989; Hendriks and Jungk, 1 98 1 ; Kirlew and Bouldin, 1 987; Majidi and Persson, 1 993) or brushing (Clemensson-Lindell and Persson, 1 992; Haussling and Marschner, 1 989) of roots collected from the field to free adhering rhizosphere soil. However literature reports on rhizosphere research on perennial tree crops are very scanty, more so under field conditions. 2.8.1 Rhizosphere acidification Plant roots acquire most of their essential mineral nutrients in cationic � +, Ca2+, Mg2+, K\ Na+ and most micro nutrients) or anionic (N03- cr, SOl", H2P04") forms. Ions with no apparent physiological role, such as various forms of Al, may also be taken-up in large quantities by the roots in acid soils. It has been reported that tea plants take-up large quantities of Al (Sivasubramaniam and Talibudeen, 1 97 1 ). However neither the function nor the charge of Al species taken-up in tea plants is clearly understood. The simultaneous uptake of several charged species in d ifferent proportions cause imbalance with respect to charge within the plant and therefore, in order to maintain electro neutrality within the plant cells, roots excrete charged ions back into the soil . If excess anions over cations are taken-up by plants, roots generally excrete orr or HC03 - causing a rise in rhizosphere pH. If excess cations over anions are taken-up then roots generally excrete W causing a decrease in rhizosphere pH (Gijsman, 1 990b,c; Haynes, 1 990) . This w/orr excretion is stoichimetrically equivalent to the charge imbalance (Breteler, 1 973; Gijsman, 1 990b; Hedley et al. , 1 982a; Troelstra, 1 983 ; Troelstra et aI. , 1 985) . Considerable differences (0 . 5 - 2 .0 pH units) between rhizosphere and bulk soil pHs have been reported for rape (Hedley et aI. , 1 982a), red spruce (Smith and Pooley, 5 2 1 989), rice (Hedley et al . , 1 994) and rye grass and white clover (Trolove et aI . , 1 996b) . Many researchers have linked pH changes primarily to N nutrition, arguing that N is the nutrient taken up in greatest quantities. Plants take up N in three main forms - as a cation � +), as an anion (N03-) or as a neutral N2 molecule (N2 fixation). Depending upon the form ofN taken up and the mechanisms of assimilation in the plant, excess of cation or anion uptake may occur in plants (Haynes, 1 983 ; Haynes and Goh, 1 978; Kirkby and Knight, 1 977). Therefore pH changes in the rhizosphere are largely as a consequence of uptake of predominantly � + (rhizosphere acidification) or N03 - (rhizosphere alkalisation) (Bekele, 1 983 ; Darrah, 1 993) . Gijsman ( 1 990bc) grew Douglas fir (Pseudotsuga menziesii) in strongly acid soil fertilised with ��, N03- or combination of both and found that the pattern of pH changes in the rhoizosphere correlated well with the form of N taken up. When �+ was used, rhizosphere pH decreased and when N03- form was used the rhizosphere pH increased and for the combination of �+ plus N03- forms the rhizosphere pH increased moderately compared to the bulk soil . A schematic representation of Ii and Off generation during the uptake and assimilation of different forms of N into amino acids and subsequent dissociation of these amino acids are shown in Figure 2 . 3 . It has been reported that tea preferentially absorbs � + compared to N03 - owing to the lower activity of nitrate reductase in roots (Ishigaki, 1 978 ; Xan and Jianyun, 1 994). According to Ishigaki ( 1 978) and Xan and Jianyun ( 1 994), in soils containing both NH/ and NO}-, tea plants should preferentially absorb �+ leaving much N03- in the soils unassimilated, which is then prone to leaching. However they did not report any experimental evidence to show that tea took-up more � + in preference to N03-. In the case ofN2 fixation by legumes, the neutral N2 is assimilated into protein and no charge imbalance is generated across the soil/root interface. Many legumes, however, commonly export H+ into their rhizospheres when actively fixing N2 (Nyatsanga and Pierre, 1 973) . This acidity is generated by the assimilation of uncharged CO2 into amino acids, which contain carboxylic acid groups with low pKa values. The H+ generated within legume roots comes from the dissociation of H+ from these carboxyl Soil Root surface Root Root/Shoot Rhizosphere Assimilation Synthesis - - - - - - - - - - - T - - - - - - - I I I NH4+ -f---I----+. ,� H 4: �+ N2 --+-------+-�:: NH3 ) .... NH3 NO -3 � _ -I------I-. ';03: �H- - - - ...- --- � - - - - - - - - - -f- ,- - / / / ...... \ O- H+ I C = O I RNI-I2 t OH I C = O I RNH2 Figure 2.3 Proton (H+) and hydroxyl (OR) generat ion during uptake and assimi lation of different forms ofN into amino acids and subsequent dissociation of amino acids (Bolan et at . , 1 99 1 ) , 5 4 groups . The acidity generated by N2 fixing legumes has been found to be equivalent to the excess uptake of cations over anions by the plant and it varies from 0 .2 - 0 .7 mol W per mol of fixed N (Jarvis and Robson, 1 983; Nyatsanga and Pierre, 1 973) . Some tropical legumes, however do not acidifY their rhizosphere as much as temperate legumes do when actively fixing N2 (Israel and Jackson, 1 978). This is partly due to the fact that their NH3 assimilation products appear to be ureides (allantoin and allantoic acid), which have high pK values (e.g. allantoin pKa 8 .96) and therefore are unlikely to dissociate releasing protons at the cytoplasmic and xylem pHs. Permanent soil acidity can also be generated under temperate legumes because the N cycle is uncoupled (Bolan et al . , 1 990). In the litter or dung mineralisation subcycIe H+ produced during nitrification is not neutralised fully by OK release through N03- uptake by plants because some N03- may be leached. 2.8.2 Relationship between cation-anion uptake and rhizosphere pH The difference (C-A) between accumulated amounts of inorganic cations � +, Na +, K+, Ca2+, Mg2+ : total = C) and inorganic anions (H2P04-, N03-, sol, cr : total = A) in plant tissues is a measure of the organic-anion or carboxylate content in the plant (Wit et aI . , 1 963). The form of N (Nli4 + or N03) taken up however must be estimated. Since micro nutrients are only present in very small quantities, these ions are generally not considered in the calculation of (C-A). A higher amount of cation uptake over anion uptake (higher C-A) is accompanied by the excretion of He and a higher uptake of anions over cations is accompanied by excretion of HC03- or OK (Haynes, 1 990). These excretions are reflected in the rhizosphere soil pH. In the case of 1 00% N03- nutrition the OK efflux can be expressed as: OH- efflux Norg + Sorg - (C-A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Equation 2 .4] � - where Norg is organic N uptake in the plant, Sorg is organic S uptake in the plant and all parameters are in units of meq per plant (Troelstra, 1 983) . Under certain circumstances this OH- efflux can be negative ( i .e . , H+ efflux). Since Sorg can be 5 5 estimated on an equivalent basis as 5 .4% of the Norg (Dijkshoorn and Van Wijk, 1 967) the equation 2 .4 can be approximated as : OH- efflux = l .054 * Norg - (C-A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Equation 2.5] In the case of 1 00% N1L + nutrition, Ii efflux can be expressed as; H+ efflux = Norg - Sorg + (C-A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Equation 2.6] or H+ efflux = 0 .946 * Norg + (C-A) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Equation 2 .7] When both N1L+ and N03- are present in soil, organic N will originate from N03- as well as N1L+. If the total organic-N is a meq planrl , the following fractions may be defined for the utilised N03- and N1L+, respectively: X and a-X meq planrl . Based on the fact that electro neutrality is maintained both inside and outside the plant, it can now be stated as: Norg (originating from N03-) + Sorg + N03- + cr + H2P04- + SO/- - Off efflux Norg (originating from N1L +) + K+ + Ca2+ + Mg2+ + Na + - H+ efflux or OH- efflux - H+ efflux = X + 0.054 a + A - (a-X) - C or net OH- efflux = 2 X - 0.946 a - (C-A) or net H� efflux (C-A) + 0. 946 Norg - 2 x. . . . . . . . . . . . . . . . . . . . [Equation 2 .8] 2.8.3 Factors affecting P availability in the rhizosphere Ample evidence has accumulated over the last several years to show that plant roots can significantly alter the chemical conditions in the rhizosphere and thereby influence the acquisition of mineral nutrients (Barber, 1 995; Darrah, 1 993 ; Hinsinger, 1 994). This review considers the factors that are controlling the plant acquisition of P from the rhizosphere. Plants draw P from the soil solution around the roots and this is buffered by re­ equilibration of solution P with P held in other forms. The soil solution P concentration is governed by P in Fe and AI precipitates, P absorbed to soil minerals and organic P. The equilibrium between the dissolution of precipitated P and solution P is governed by the solubility product of the precipitated phase (Barber, 1 995) . In soils fertilised with PR, the depletion of solution P by plant uptake will cause more PR dissolution in the rhizosphere to replenish P that is removed from the rhizosphere (Hinsinger and Gilkes, 1 995). Similarly depletion of solution P will cause desorption of adsorbed P (Barber, 1 995 ; Mengel and Kirkby, 1 987) and mineralisation of organic-P to maintain the equilibrium (Bolan et aI. , 1 997; Jungk, 1 996; Tarafdar and Jungk, 1 987). 2.8.3.1 pH Plant induced rhizosphere acidification results in an increase in the rate of dissolution of the PR particles in the rhizosphere (Hinsinger and Gilkes, 1 995; Kirk and Nye, 1 986; Trolove et al. , 1 996b) and thereby increases the availability of phosphate to the plants in soils treated with PR Trolove et aL ( 1 996b) observed more PR dissolution in the rhizosphere of lotus plants treated with NCPR fertiliser and explained this as due to lower pH in that zone compared to that in the bulk soiL Gahoonia et aI. ( 1 992) showed that the dissolution of calcium bound-P in a luvisol increased due to acidification resulting from NH4 + nutrition whereas ryegrass utilisation of P sorbed to Fe and AI increased by N03- induced alkalisation in an oxisoL 5 6 5 7 Plant roots through their effect on pH can influence the adsorption and desorption of P by soils (Hinsinger and Gilkes, 1 996), dissolution of Fe and AI phosphate (Armstrong and Helyar, 1 992) and thereby control P availability to plants. 2.8.3.2 Release of organic acids and their chelating action The roots of many plant species release various organic compounds into the soil. They can be subdivided into three groups (lungk, 1 996). 1 . Mucilage, consisting of high molecular polysaccharides and polygalacturonic acid 2. Sloughed off cell from root cap 3 . Low molecular organic acids, and complexing agents. The mucilage and sloughed off cells from the root cap do not seem to help nutrient uptake in any specific manner. Among these three groups, the third one which compnses organic acids is more important in influencing the solubility of soil phosphate particularly Fe and AI bound P (Gerke and Jungk, 1 99 1 ) . The mechanisms reported include the effect of protons, but apparently more important is the complexation of metals and ligand exchange of adsorbed phosphate by carboxyl groups (Jones and Darrah, 1 994) . The release of dicarboxylic and tricarboxylic acids (citric, malic and other acids) from the apical parts of the roots into the rhizosphere helps P mobilisation. The release of these acids was found in several dicotyledon plant species, particularly in leguminosae members (Grierson, 1 993; Hoffiand et aI. , 1 989). Grasses do not seem to secrete appreciable amounts of these acids. It was reported that tea roots secrete significant amounts of citric and malic acids which can dissolve native phosphate compounds in the rhizosphere soil (layman and Sivasubramaniam, 1 975 ; Xiaoping, 1 994) . The secretion of these acids are greatly enhanced in some plants when they are subjected to phosphate starvation. Hedley et al . ( 1 982a) observed rape plants secrete organic acids from roots when they are starved of P . Hoffiand et aI . ( 1 992) showed that rape plants are efficient users of phosphate rock, due to their release of citric and malic acids as a response to P stress. Fox et al. 5 8 ( 1990) and Jones and Darrah ( 1 994) showed that the citrate and malate secretion by roots could significantly increase the concentration of phosphate in soil solution and thus provide a higher flux from soil to the roots. 2 .8.3.3 Phosphatase enzyme activity and mineralisation of organic-P About 50% of the total soil phosphate occurs in organic forms, most of which derived from plant residues and, in part, synthesised by soil micro-organisms from inorganic sources (Sanyal and De Datta, 1 99 1 ) . The contribution of organic P (Po) i s considered as an important source for plant P nutrition in tropical countries ( Sanchez, 1 976), but it is commonly believed that Po has no direct effect on the P nutrition of plants. Organic P has to be mineralised before being absorbed by plants and this is done through hydrolysis by a group of phosphatase enzymes produced by micro-organisms. It has been found that the phosphatase enzyme activity immediately outside the plant roots is significantly higher than that in the bulk soil (Dinkelaker and Marschner, 1 992; Tarafdar and Jungk, 1 987). The higher phosphatase enzyme activity causes hydrolysis of Po to produce plant -available inorganic P. The activity of phosphatase enzymes in tea roots was studied by Xiaoping et al . ( 1 989) in pot trials on Red Earth soils in China and the results showed that the activity of these enzymes was greater in soils near the roots compared to those that are distant from the root surface. They also showed that the enzyme activity was positively correlated with organic C and Po status of the soil. The Po component of the 0 . 1 M NaOH extract is believed to contain some labile organic compounds such as RNA, nucleotides and gIycerophosphates. These organic compounds have been found to be readily mineralisable and subsequently available for plant uptake (Bowman and Cole, 1 978 ; Tarafdar and Claassen, 1 988) . Adams and Pate ( 1992) observed that inositol phosphate was as efficient as the inorganic-P source KH2P04 in supplying P to lupins when grown in a sand culture. In acid forest soils, Haussling and Marschner ( 1 989) found that readily hydrolysable Po was depleted in the rhizosphere of 60 to 1 00 year old Norway spruce (Picea abies L. Karst) grown in a Cambisol while Pi concentrations were unaffected or even 5 9 increased. This could b e due t o root-mediated production of acid phosphatases in these soils, which may have hydrolysed Po into Pi (Dinkelaker and Marschner, 1 992). Though readily soluble Po comprised a large proportion of the 0 . 1 MNaOH extract of soils, Armstrong and Helyar ( 1 992) did not observe any utilisation of this fraction by semi-arid pasture grasses from South-western Queensland. This was explained as due to accumulation of Po at a similar rate to that of plant uptake of mineralised Pi . 2 .8.3.4 Mycorrhizal association The association of mycorrhizal fungi (V AM, vesicular arbuscular mycorrhizae) is known to improve the phosphate supply to plants if available phosphate in the soil is low (Tinker, 1 984) . It was found that more phosphates per volume of soil was extracted by the fungus at the root surface. The efficiency of V AM in transferring P from soil into plants was mainly attributed to the structure of the mycelium (Barea, 1 99 1 ; Bolan, 1 99 1 ) . It is likely that plants that are heavily infected with V AM are better able to acquire dissolved P from PR because of greater volume of soil into which the mycorrhizal root system can extend, compared to non-mycorrhizal roots. Mycorrhizal hyphae could extend to a distance greater than the usual P depletion zone in the rhizosphere soil so that V AM hyphae could extract P from a soil zone beyond the reach of the roots . The increase in plant growth by mycorrhizal association is largely due to increased absorption of nutrients from the soil solution. It has been reported that the rate of nutrient uptake in mycorrhizal associated plants is faster than non-mycorrhizal associated plants (Smith et aI . , 1 985 ; Son and Smith, 1 988) . For example Sanders and Tinker ( 1 973) observed that the rate of inflow of P into mycorrhizal roots was much higher ( 1 7 * 1 0- 14 moles cm-I S-I ) than that of non- mycorrhizal associated plants (3 .6 * 1 0- 1 4 moles cm- I s-\ Sainz and Arines ( 1 988) measured different fractions ofP in an acid soil after growing red clover with and without mycorrhiza. They found that both mycorrhizal and non­ mycorrhizal associated plants decreased the concentration of inorganic-P in the soil but did not affect the concentration of organic-P and suggested that both mycorrhizal and non-mycorrhizal associated plants obtained their P requirements from the 6 0 inorganic source of P in the soil. However change of P in one fraction due to plant uptake can alter another fraction, which makes it very difficult to identify the sources ofP utilised by the plant. Benefits have been reported from the association of V AM (endo-mycorrhiza) with tea plants in India (Barthakur et aI . , 1 987), China (Zhi, 1 993) and in Japan (Morita and Konishi, 1 989), but no information is available on the extent of this association in different tea clones and its influence on the efficiency of p lant P uptake from soils. 2.9 SUMMARY AND RESEARCH NEEDS Previous studies on the P nutrition of tea were largely carried out on tea yield responses to P fertilisers in glasshouse and field trials. Most trials showed that tea yield responses to applied P fertilisers were inconsistent or irregular and in many cases there was no significant response at all . In these experiments, the soil P status or the fate of applied P was not reported adequately and therefore it was not possible to determine whether the soils had adequate plant-available P before application of P fertilisers and to what degree the p lant available P pool increased with the application of P fertiliser. Tea soils are highly acidic (pH 4 .0 - 5 . 5) and are rich in oxides and hydroxyoxides of Fe and A1 which are known to fix P. This led to the belief that high rates of P fertilisers needed to be applied to obtain yield responses in tea. But in practice rates as low as 5 - 1 5 kg P ha- I y{1 seem to be sufficient to reach maximum yield though the yield increase was small. Very little work has been carried out to understand the reasons behind the complexity of the P chemistry of tea soils and the mechanisms of P uptake by tea plants. It has been suggested that tea roots secrete organic acids and these acids may be dissolving some of the P fixed to soil so as to make them available to the roots. In order to test this hypothesis, studies need to be conducted on the chemistry ofP in soils close to the roots (i .e . in the rhizosphere). 6 1 Currently a locally available PR mined at Eppawala (EPR) in the North Central province of Sri Lanka is being used as the sole source of P fertiliser for mature tea in Sri Lanka, but without adequately testing its agronomic suitability for tea. EPR is considered to be a PR of low reactivity according to its citric acid solubility. Whether it will dissolve adequately in the highly acidic tea soils especially in the rhizosphere is a question that needs to be answered with proper research studies. More research is required in determining its suitability as P fertiliser for tea. The literature shows that different soil P tests have been used in various tea growing countries to predict availability of soil P to tea, without adequately testing their suitability in comparison to other soil tests. Research is needed to compare the various soil P tests and to choose the most suitable tests by correlating soil test values with shoot dry matter yield and plant P uptake. 6 2 CHAPTER 3 A TECHNIQUE FOR STUDYING RHIZOSPHERE PROCESSES IN TREE CROPS : SOIL PHOSPHORUS DEPLETION AROUND CAMELLIA (Camellia japonica L) ROOTSl 3.1 INTRODUCTION The review of literature presented in Chapter 2 shows that root-soil interactions in the rhizosphere markedly affect P availability to plants (Marschner et al . , 1 987) . The conditions at the root-soil interface are considerably different from, and influence plant growth more, than those at a distance from the root. For this reason, many researchers have been interested in studying the characteristics of this zone, the rhizosphere, relative to those of the bulk soil (Armstrong and Helyar, 1 992; Gahoonia et al. , 1 992; Hedley et ai. , 1 994; Hinsinger and Gilkes, 1 995 ; Marschner et al . , 1 987 ; Wang et al. , 1 995) . The rhizosphere is a narrow soil cylinder (about 0 - 2 mm radius) surrounding the root and therefore, it is technically difficult to study the root induced chemical changes in this zone. One problem is the small amount of rhizosphere soil available for chemical analysis and another is the determination of the line of demarcation between the rhizosphere and the bulk soil . Different techniques for studying chemical changes in the rhizosphere had been developed in the past for annual crops, grasses and legumes (Gahoonia et al . , 1 992; Hedley et aI . , 1 994; Jungk and Claassen, 1 989; McLaughlin and James, 1 99 1 ; Youssef and Chino, 1 988) . Some of these studies assumed that soil particles adhering to the roots are representative of rhizosphere soil and the soil distant from the roots was bulk soil and not influenced by roots (Ohno, 1 989; Riley and Barber, 1 97 1 ) . lZoysa, A K N, Loganathan P and Hedley M J 1997 A technique for studying rhizosphere processes in tree crol)S : soil phosphorus depletion around camellia (Camellia japonica L.) roots. Plant and Soil, 190, 253-265. 6 3 The practical difficulty of getting samples at known distances from the rhizoplane (root surface) is a significant obstacle in this approach. In other studies this problem was overcome by growing plants in soil in a cropping device based on the early work of Kuchenbuch and Jungk ( 1 982), where a planar mat of roots was physically separated from the soil by a polyester mesh. Thin sections of soils at various distances from the mesh (rhizoplane) were sliced and chemically analysed to determine root induced chemical changes (Gahoonia and Nielsen, 1 99 1 ; Hedley et aI . , 1 994; Wang et aI. , 1 995; Youssef and Chino, 1 989). The above studies on annual crops, grasses and legumes showed that there were marked differences in soil pH and the concentration of the different P forms between soil layers within a few mm from the root surface and the bulk soil. Nevertheless much less is known about the rhizosphere processes in tree crops especially in the field due to the absence of a dependable method for sampling the rhizosphere soil. 3.2 OBJECTIVES The objectives of the investigation reported in this chapter are: 1 . To modify the rhizosphere study container (RSC) technique ofKuchenbuch and Jungk ( 1982) to investigate rhizosphere processes in tree crops under glasshouse and field conditions. 2. To study rhizosphere acidification and soil and fertiliser P depletion patterns around the fine roots of camellia (CameWa japonica L.) under glasshouse and field conditions. 3 . To investigate the fate of applied P fertilisers in the rhizosphere and bulk soil . 3.3 MATERIALS AND METHODS 3.3 . 1 Soils 6 4 An alluvial soil (0 - 1 0 em depth) lying beyond the reach of flood water in the North Island of New Zealand was used in this study. The soil carries the soil type name Karapoti silt loam. It belongs to the Recent Order in the New Zealand classification system and is classified as a Dystric Eutrochrept in the US soil Taxonomy. Some important physico-chemical characteristics of the soil are presented in Table 3 . 1 . 3.3.2 Glasshouse trial The soil was air-dried, passed through a 2 mm sieve and amended with four P fertilisers: North Carolina phosphate rock (NCPR particle size 34. 1 % > 25011m; 52 . 5% 1 50-250 f.1m; 1 3 .4% < 1 50 f.1m, total P 1 3%, all water insoluble), single superphosphate (SSP, total P 9%, 80% total P water soluble), monocalcium phosphate (MCP, 24% total P, all water soluble) and diammonium phosphate (DAP, 20% total P, all water soluble) at the rate of200 llg P g-l soil. To ensure that N and K deficiencies did not restrict plant growth, urea and KCI were applied at the rate of 1 00 Jlg N and K g-l soil respectively. Rhizosphere study containers described by Kuchenbuch and Jungk ( 1 982) and Hedley et al . ( 1 994) were used to study the rhizosphere processes. An RSC is a two compartment device, made-up of two (PVC) cylinders, the upper compartment having an internal diameter of 82 mm and 25 mm effective depth and the lower compartment having an internal diameter of 74 mm and 50 mm depth. The two compartments were separated by a 24 !lm pore-diameter polyester mesh. The upper compartment was packed with 1 3 0 g soil (bulk density; 1 . 0 Mg m-3) and the lower compartment with 242 g of soil (bulk density; 1 . 1 Mg m-3). Eight month old seedlings of Tom Thumb, a variety of camellia (Camellia japonica L.) propagated from cuttings, were transplanted into the upper compartment of the RSCs (Figure 3 . 1 a) . Plant roots in the upper compartment striking the polyester mesh were unable to penetrate the mesh and 6 5 Table 3. 1 The physico-chemical characteristics of the experimental soil Character Unit Value Soil pH 1 : 2 . 5 w/w (0 .0 1 CaCh) 5 .0 pH buffer capacity (at pH 4-5) mmol H+ kg-! pIT! 2 1 Organic C % 2 Olsen P Ilg g-l soil 30 CEC I emole kg- l 1 5 Ex. Ca cmole kg-! 7 Ex. Mg emole kg- ! 2 Ex. K cmole kg-! 2 1 1 M�OAe pH 7 extraction (Blackmore et a1.; 1 987) (a) Glasshouse trial 24 11m polyester mesh 1 00 11m polyester mesh sand (b) Field trial Al � camellia plant soil water � e A2 I RSC L--.._ 1 m m polyester m esh roots Figure 3 . 1 Schematic representation of Root Study Container (RSC) technique used in the (a) glasshouse and in the (b) field trials . 6 6 6 7 therefore grew horizontally along the mesh forming a root mat (Figure 3 .2). The soil below the polyester mesh therefore represents the rhizosphere and the zone of transition demarcating the bulk soil. The RSCs were placed on a sand bed, which was kept moist by a water reservoir (Figure 3 . 1 a) . The watertable was fixed at 1 60 mm below the base of the RSCs. This enabled the RSCs to be kept at a constant water potential of approximately - 1 . 6 kPa. Ten ml of 1 % urea solution was added to all RSCs at the end of the seventh week. The trial examined the effects of four forms of P fertilisers on soil pH and soil P fractions in the rhizosphere of camellia plants. These treatments plus a control (with no P fertiliser added) were replicated five times and arranged in a randomized complete block design in a glasshouse maintained at 28° C maximum and 1 3° C minimum temperatures. At the end of 56 days, plant shoots were cut 5 mm above the soil surface. The soil in the lower compartment was sliced into thin sections with a piston microtome (Figure 3 .3 ) starting at the inter-cell boundary. The first four sections were sliced at a 0 . 5 mm thickness and a second set of 6 slices were taken each at 1 mm thickness in order to study the root induced changes in soil slices with increasing distance from the rhizoplane. The effect of the P fertilisers on soil pH and P fractions in the absence of plants was also measured in replicated RSCs treated with the four P fertilisers and a control (with no P fertiliser) . 3.3.3 Field trial In the field trial the RSCs were modified by mounting a 1 mm pore-diameter polyester mesh at the opening on one side of one of the compartments (Figure 3 . 1 b, 3 . 4 and 3 . 5) to allow plant roots to enter the soil inside this compartment. Phosphate fertilisers tested were NCPR and SSP. These, urea and KCl fertilisers were mixed with the soil at the same rates as in the glasshouse trial and the RSC containers were filled with this soil . In the control treatment the soil was mixed only with N and K fertilisers (No P fertiliser). A deep vertical hole was made into the soil 3 0 cm away from the base trunks of > 1 0 yr old camellia trees by cutting the roots with a sharp blade and carefully removing the soil. The RSCs were buried horizontally in the hole NCPR SSP Figure 3.2 Bottom view of the root mats formed above the polyester mesh in root study containers (RSCs) 6 8 6 9 Figure 3.3 Front view of the piston microtome 7 0 Figure 3.4 Root Study Container (RSC) modified for field situation 7 1 Figure 3.5 A root study container buried near a mature camellia tree in the field 7 2 to allow new roots to grow and penetrate through the 1 .0 mm polyester mesh into the RSC compartment facing the tree roots (Figures 3 . 1b and 3 .5) . The treatments were replicated five times. At the end of six months, the soil around the RSCs was dug out and RSCs were removed by cutting the roots entering the RSCs a few mm outside the 1 mm mesh. The soil in the compartment on the tree side was sliced 1 mm above the inter-cell boundary (24 11m polyester mesh) to obtain measurements of root length and weight at the boundary. Root radius was determined using the formula "-'(MhtpL) (M : root weight, 11: : 2217, p : density of roots, L : length of root) assuming it is a cylindrical tube of constant radius. These roots were responsible for the changes in soil characteristics on the other side of the mesh. The roots were removed from the soil, gently washed and root lengths measured using a Comair root length scanner. Rhizosphere soil in the RSC compartment away from the roots was sampled using a piston microtome as described in the glasshouse trial . 3.3.4 Plant and soil analysis Plant samples were dried at 60° C and ground to < 1 mm. Both shoot and root samples were analysed for total P (Jackson, 1 95 8) . The soils were air-dried and analysed for available soil-P (Olsen et al . , 1 954), cation exchange capacity (CEC) and exchangeable cations ( 1 M NfLOAc buffered at pH 7 .0, Blackmore et ai., 1 987), soil organic C (Walkley and Black, 1 934) and exchangeable NH/ and N03- ( 1 M KCI extraction, Markus et aI . , 1 985) . Soil pH was measured in 0. 0 1 M CaCh (0 .5 g soil : 1 .2 5 ml) . pH buffer capacity was determined by Ca(OH)2 titration method (Bolan et al . , 1 986) . The amount of NCPR dissolution in the soils was determined by the method ofTambunan et al . ( 1 993). The % ofP dissolved from EPR was calculated as follows. % dissolution of P ::::; 100 [1 - 0 .5 M HfS04 extractable P [(soil + PR fertiliser) - (soil alone)1] fertiliser P added 7 3 3.3.5 Soil P fractionation Beginning with 0. 5 g of < 1 mm air dry soil the following soil P fractions were determined sequentially by the procedure of Hedley et al. ( 1994). ( 1 ) Resin-P, by shaking end-over-end for 16 h at 25° C in 30 ml of deionised water containing a strip each of anion (AER) and cation (CER) exchange resin membrane (approximately 0 . 5 meq of exchange capacity per strip), then removing the strips and recovering P from them by eluting with 0 . 5 MNaCl. (2) NaOH-Pi, by adding 3 .3 ml of 1 MNaOH to the suspension from step ( 1 ) (i . e. final concentration 0 . 1 M NaOH) and reshaking as above. (3) NaOH-Po, by digesting 5 ml of the NaOH extract in 4 ml of conc. H2S04 and 0 . 5 ml of H202 and subtracting NaOH-Pi from the digested P . (4) H2S04-Pi, by adding 30 ml of 0 . 5 M H2S04 to the soil residue from step (2) and re-shaking as above. (5) Residual-P, by refluxing the soil residue from step (4) in 8 ml conc. H2S04 at 3 500 C for 3 h, cooling, adding 0.5 ml H2S04 and reheating, and repeating this step until the residue remained white on further reheating. The digests were finally diluted to 5 0 ml with deionised water. Phosphate concentrations in all the extracts and digests were determined by colorimetry (Murphy and Riley, 1 962). 3.4 RESULTS AND DISCUSSION 3.4.1 Effect of P fertilisers on soil pH The soils fertilised with readily soluble P fertilisers (DAP, MCP and SSP) had lower pHs than the control treatment in both rhizosphere and bulk soil in the glasshouse as well as in the field trials, while NCPR had a higher soil pH than the control treatment (Figure 3 . 6a, b, c). The decrease in pH observed for the MCP and the SSP treatments is probably due to the acidity produced when the dihydrogen phosphate in MCP and SSP dissociates to produce monohydrogen phosphate in the soil (Giroux et a! . , 1 984) . Hedley et al . ( 1 994) also observed pH reduction in soils treated with MCP compared to soils with u c<:l 5.0 4 . 8 u 4 6 � . o o '-" ::r: 4.4 0.. .- o V) 4.2 (a) 5 .0 1 1 1 1 I 1 I I I I 4 . 8 4 . 6 -+- Control 4 .4 -II- NCPR ___ DAP ----.- SSP 4 .2 ---*- MCP o 1 2 3 4 5 6 7 8 (b) r I l l I I I I r I 5.0 4.8 4.6 4.4 4 .2 o 1 2 3 4 5 6 7 8 (c) l I I I I I 1 : : o 1 2 3 4 5 Distance from rhizoplane (mm) Distance from rhizoplane (mm) Distance from rhizoplane (mm) Figure 3.6 Effect of P fertil i ser forms on soi l pH (0.0 1 M Cae! ) in camellia rhizosphere in the glasshouse and 2 the fi eld trial s (a) without plants - glasshouse tri al (b) with plants - glasshouse trial and (c) with plants - field trial . Vertical bars correspond to Lsd at p <0.05 . 7 5 no fertilisers. The release of protons from nitrification ofNH/ in DAP may have been the reason for the decrease in pH in the DAP treatment. The increase in soil pH observed for the NCPR treatment compared to the control treatment is due to the consumption of protons during the dissolution of phosphate rock and carbonate in apatite structure and in the accessory minerals (Loganathan et al . , 1 995) . In a yellow­ grey earth( Aeric Fragiaqualf)soil under pasture Manoharan et al. ( 1 995) also reported that application of NCPR increased soil pH and DAP decreased soil pH over the control (no P) treatment but SSP application had no effect on soil pH. 3.4.2 Effect of plant roots on soil pH A reduction in soil pH of 0.2 - 0.4 units was observed near the roots of camellia plants compared with the bulk soil in both the glasshouse and field trials for all treatments (Figure 3 . 6b, c) . No change in soil pH was observed with distance from the polyester mesh in the lower compartment in the soils with no plants (Figure 3 . 6a). This clearly showed that the soil pH changes were induced by plant roots. The extrusion of protons by roots to maintain electroneutrality in plant tissues as plant roots take up an excess of cations over anions is considered to be the dominant cause for rhizosphere acidification (Barber 1 995; Haynes, 1 990) . The nature of ions excreted to maintain electroneutrality in non-legumes is usually governed by the plant' s N nutrition. A higher uptake ofN1L� causes H+ release and higher N03- uptake causes Off release to the rhizosphere (Gahoonia et al . , 1 992; Gijsman, 1 990bc; Nye, 1 98 1 ). Proton extrusion may also be associated with the release of Jow molecular weight organic acid anions (Hoffiand, 1 992; Hoffiand et al . , 1 989; Liu et ai . , 1 990) . Tea plants (Camellia sinensis L.), which belong to the same family as camellias, are reported to secrete significant quantities of malic acid from their roots (Jayman and Sivasubramaniam, 1 975 ; Xiaoping, 1 994). The pH drop in the rhizosphere of NCPR treated soil was lower than those in the soluble P fertiliser treated soils in the glasshouse trial . This is because higher proton consumption from NCPR dissolution buffered the pH change in the rhizosphere. Differences in pH drop between treatments in the rhizosphere soils were not observed 7 6 in the field trial. NCPR dissolution decreased with distance from the rhizoplane (Figure 3 .7) because the rhizoplane was the source ofH", which diffused out into the bulk soil. The amount of H" consumed in dissolving NCPR in the rhizosphere and bulk soil was estimated using the amount of NCPR dissolved and from the relationship that 2 moles of H" are consumed for every mole of P dissolved (Loganathan et ai. , 1 995). Similarly W consumption for this dissolution was predicted by using the soil ' s pH buffering capacity, the observed pH increase in the rhizosphere and in the bulk soil of the NCPR treatment over the control treatment. The predicted and estimated H" consumption in NCPR dissolution was found to be in better agreement when the P­ fractionation scheme of Tambunan et al . ( 1 993) was used to measure the extent of NCPR dissolution, than when the P-fractionation procedure involving resin strips was used (Table 3 .2). [This is caused by an over estimation ofP-dissolution when the resin extraction was used (Trolove et al . , 1 996b) as the initial step in the fractionation scheme. This discrepancy in P dissolution between the two procedures is discussed in the next section] . Similar estimations carried out for NCPR dissolution under field conditions did not show a good agreement between the predicted and actual H" consumption for either of the two fractionation schemes (Table 3 .2). This may be due to removal of NCPR dissolution products such as F and Ca by percolating rainfall and/or acidity intrusion into the RSCs from soil surrounding the RSCs in the field trial . These processes are not likely to occur in the closed experimental system used in the glasshouse trial. 3.4.3 Effect of P fertilisers on soil P fractions At the end of the respective study periods, the P fractionation of bulk soil (3 - 8 mm from the rhizoplane - an area not significantly influenced by P depletion by plant roots) in the unfertilised (control) treatment in both the glasshouse and in the field trial showed that NaOH-Pj and H2S04-Pi ( 1 20- 1 58 jlg g-l soil) had larger P fractions than resin-P, NaOH-Po and residual-P ( 1 6-65 f.lg g-l soil) (Table 3 . 3 , Figures 3 . 8 - 3 . 1 1 ) . Application of P fertilisers increased all soil P fractions compared to control (a) (b) 70 70 60 60 � � 50 50 '-' "'0 V ;> (3 VJ . � 40 40 "'0 P-. 30 30 o Without resin • with resin 20 20 4 .60 4 .65 4 .70 4 .75 4 .80 4.85 4 .40 4.45 4 . 50 4 . 55 4 .60 4.65 Soil pH Soil pH I I I I I I I I I I I 0 0. 5 1 .5 2 4 5 o 0 .5 1 . 5 2 4 5 Distance from rhizoplane (mm) Distance from rhizoplane (mm) Figure 3.7 Effect of soi l pH (0 . 0 1 M CaCI2), distance from rhizoplane and method of P extraction (with and without resin) on NCPR-P dissolution in the (a) glasshouse and (b) field trial s . Vertical l ines correspond to standard errors of means. 4 .70 Table 3.2 Estimated and predicted Ii consumption for dissolution ofNCPR fertiliser in the rhizosphere and bulk soil in the glasshouse and the field trials Distance NCPR Estimated H+ Observed pH Predicted H+ 7 8 from dissolution consumption from increase in consumption due rhizoplane (%) NCPR dissolution NCPR treatment to pH rise (mm) (/lmo} H+ g"l soil) over unfertilised (/lmol H+ g-l soil) control Glasshouse trial (a) TEAlNaCI (pH 7) extraction of soil as the first step in P fractionation 0-0 . 5 36 .9 6. 1 6 ± 0 .22 0 .29 5 .79 ± 0 . 2 1 0 . 5- 1 .0 3 1 .6 5 .28 ± 0.28 0 .24 5 .08 ± 0.20 1 . 0- 1 . 5 30A 5 .07 ± 0. 1 7 0 .23 4 .78 ± 0 . 1 4 1 . 5-2 .0 29. 5 4 .92 ± 0. 1 2 0 .22 4 . 70 ± 0 . 3 3 6 . 0-7 .0 27.3 4 . 54 ± OA3 0 . 2 1 4 . 57 ± 0 . 24 7 .0-8 . 0 28 .0 4 .67 ± 0 .37 0.22 4 .54 ± 0 . 37 (b) Resin extraction of soil a s the first step in P fractionation 0-0. 5 50 .7 8A6 ± 0 . 86 0.29 5 . 79 ± 0. 2 1 0 . 5- 1 . 0 44.9 7A9 ± 0 .94 0 .24 5 .08 ± 0 . 20 1 .0- 1 . 5 4 1 .2 6 .87 ± 1 .26 0 .23 4 . 78 ± 0 . 1 4 1 . 5-2 . 0 42.0 7 .0 1 ± 1 .47 0 .22 4 . 70 ± 0 . 3 3 6 .0-7 .0 39 .3 6 .56 ± 1 . 56 0 .2 1 4 . 57 ± 0 .24 7 .0-8 .0 40.9 6 .83 ± 1 . 3 8 0 .22 4 . 54 ± 0 . 3 7 Field trial (a) TEAlNaCI (pH 7) extraction of soil as the first step in P fractionation 0-0 . 5 45 .6 7 .62 ± 0 .06 0 . 1 0 2 .00 ± 0 . 83 0 . 5- 1 .0 4 1 . 7 6.98 ± 0 .06 0 .07 l AO ± 0 .8 1 1 . 0- 1 . 5 4 1 . 5 6 .96 ± 0 .07 0 . 1 1 2 .20 ± 0 . 1 1 1 . 5-2 . 0 4 1 . 7 6 .98 ± 0 .06 0 .09 1 . 85 ± 0 . 1 2 3 . 0-4. 0 38 . 5 6AO ± 0 .06 0 . 1 1 2 .26 ± 0 .24 4 .0-5 . 0 3 8 . 8 6A2 ± 0 .03 0 . 1 1 2 . 36 ± 0 .24 (b) Resin extraction of soil as the first step in P fractionation 0-0. 5 60.0 9 . 88 ± 0 .07 0 . 1 0 2 .00 ± 0 . 83 0 . 5- 1 . 0 56 .8 9 .29 ± 0. 1 2 0 .07 l AO ± 0 . 8 1 1 .0- l . 5 57 . 3 9 .41 ± 0 . 1 7 0 . 1 1 2 .20 ± 0 . 1 1 1 . 5-2 . 0 54 .8 8 . 86 ± 0.42 0 .09 1 . 85 ± 0 . 1 2 3 . 0-4 . 0 54 .6 8 .85 ± 0 . 1 1 0 . 1 1 2 .26 ± 0 .24 4 .0-5 . 0 54 .4 8 . 84 ± 0 .28 0 . 1 1 2 . 36 ± 0 . 24 Table 3.3 P-fractions in unfertilised bulk soil (mean of soil slices 3-8 mm from rhizoplane) and % recovery l of added P fertiliser in the various P fractions in the glasshouse and the field trials Glasshouse trial P-fraction Unfertilised Fertiliser P recovery (%) bulk soil P ( Ilg got soil) NCPR DAP Resin-P 3 5 1 5 . 6 1 4 . 0 NaOH-Pi 1 5 1 30 .6 3 3 . 2 NaOH-Po 50 - 1 2 . 0 - 7 . 7 H2SO4-Pi 1 3 8 62. 3 47. 4 Residual-P 1 6 6 .7 6. 5 Total-P 390 1 03 .2 93 .4 1 (P fraction in fertilised soil - P fraction in unfertilised soil) * 1 00 Fertiliser P added to soil S SP 1 3 . 6 37 .8 - 1 2 . 2 46.4 7 .9 93 . 5 Field trial Unfertilised Fertiliser P recovery (%) bulk soil P NCPR SSP MCP (Ilg gol soil) 1 2 . 8 48 8 .7 1 3 . 9 3 2 . 3 1 58 3 7 . 9 5 3 . 5 - 1 3 . 9 6 5 -0.3 -4 . 5 4 5 . 5 1 20 43 . 1 20.4 1 0. 8 25 1 0. 4 1 5 . 2 8 7 . 5 4 1 6 99. 8 9 8 . 5 (a) 1 00 80 60 40 20 o N N N 1 l 1 1 1 1 1 1 2 3 4 5 6 7 8 Distance from rhizoplane (mm) (b) 1 00 80 60 40 20 o I I I I I I I r : : : /iC CO�"OI • -II- NCPR -e- DAP -.k- sSP -*- MCP 1 2 3 4 5 Distance from rhizoplane (mm) Figure 3.8 Effect of P ferti li ser forms on resin-P in camell ia rhizosphere in the glasshouse and in the field trials (a) with plants - glasshouse trial and (b) with plants - fi eld trial . Vertical bars correspond to Lsd at p <0 .05 and N represents stati stically nonsignificant at p <0.05 . (Xl o -'OJ) OJ) :::L (a) 300 250 200 1 50 o I I l l I I I I I I • • • • • 1 2 3 4 5 6 7 8 Distance from rhizoplane (mm) (b) 300 - - 250 - 200 - - I I I I 1 50 - � - I I ... ... • • --.- Conlrol -II- NCPR ____ DAP -lilt- sSP --If-- Mep • • I • • 1 00 -+---.,----,1----,1----,'--..1 o 1 2 3 4 5 Distance from rhizoplane (mm) Figure 3.9 Effect of P fertili ser forms on NaOH-P. in camellia rhizosphere in the glasshouse and in the field trials 1 (a) with plants - gl asshouse trial (b) with plants - field trial . Vertical bars correspond to Lsd at p <0 .05 . 'bI) bI) :::t � , (a) 350 300 250 "f 200 o if] N � 1 50 o 1 2 3 4 5 6 7 8 Distance from rhi zoplane (mm) 350 300 250 200 1 50 (b) I r I I I I I -.- Control -II- NCPR -e- DAP -..- sSP -+- MCP • • • .. .. ... . � . • • o 1 2 3 4 5 Distance from rhizoplane (mm) Figure 3.10 Effect of P ferti l i ser forms on H2S04-Pi in camellia rhizosphere in the glasshouse and in the field tria ls (a) with plants - g lasshouse trial and (b) with plants - field trial . Vertical bars c orrespond to Lsd at p <0.05 . (a) 1 75 1 50 1 25 1 00 75 50 25 o 1 I 1 I I I 1 N N N 2 3 4 5 6 7 8 Distance from rhizoplane (mm) (b) 1 75 1 50 1 25 1 00 75 50 25 o N N N N N N N --+- Control -II- NCPR ___ DAP ---Ik-- s.w -*- MCP 1 2 3 4 5 Distance from rhizoplane (mm) Figure 3. 1 1 Effect of P fertil i ser forms on NaOH-P in camellia rhizosphere in the glashouse and in the o field trial s (a) with plants - glasshouse trial and (b) with plants - field trial . Vertical bars correspond to Lsd at p <0.05 and N represents stati stical ly nonsignificant at p <0.05. 84 treatment (Figure 3 . 8 , 3 .9, 3 . 1 0) except NaOH-Po i n both trials (Figure 3 . 1 1b and c ). In the glasshouse trial (Figures 3 . 8 - 3 . 1 1 ) P fertiliser form had no significant effect on the extent to which soil P fractions increased. In the field trial however, significant differences were observed in the fractions of NaOH-Pj and H2S04-Pi between P fertiliser treatments. H2S04-Pi was highest for NCPR treatment in both glasshouse and field trials compared to the control treatment (Figure 3 . 1 0) . The recovery of fertiliser P in the bulk soil in both trials was close to 1 00% (88 - 1 03% in the glasshouse trial and 99 - 1 00% in the field trial, Table 3 .3) . The majority of added fertiliser P was recovered in the NaOH-Pi or H2S04-Pi fractions, which are not immediately available to plants (Table 3 . 3) . Rapid chemical reactions change the dissolved constituents of soluble fertilisers (MCP, DAP and SSP) into compounds similar to those of "native" soil-P i .e . NaOH-P; (Fe and Al bound P) or NaOH-P 0 (Golden et al . , 1 99 1 ; Hedley et ai . , 1 994; Perrott, 1 995) . The recovery of large amounts of H2S04-Pi in DAP, SSP and MCP treated soils was not expected since accumulation of P in these fractions as reaction products of soluble P has not been evident in low pH soils with lower Ca saturation (Hedley et al . , 1 994; Tambunan et aI . , 1 993). Perhaps the higher exchangeable Ca in the soils may have precipitated calcium phosphates over time. In NCPR treated soils undissolved NCPR residue dominated the H2S04-Pi fraction. The estimated NCPR dissolution in glasshouse soils showed that nearly 4 1 % was dissolved in the bulk soil when a resin strip was used as the initial step, while only 28% was dissolved when the P-dissolution scheme of Tambunan et al. ( 1 993) where TEAJNaCl (pH 7) was used as the initial step to remove exchangeable Ca (Figure 3 . 7) . The discrepancy in the amounts of P dissolution was due to the differences of pH in the equilibrium medium of the initial steps of the P fractionation in the two schemes. In the scheme where the soil was agitated with resin strips for 1 6 hr the soil suspension pH was that of the acidic soil (pH 4 . 8) and this would have dissolved some residual NCPR during extraction (Trolove et al . , 1 996b) . In the Tambunan et al. ( 1 993) scheme where a buffered neutral pH medium was used to equilibrate the soil, the neutral pH did not dissolve any NCPR (Loganathan et aI . , 1 995) . Differences in 8 5 NCPR dissolution by the two methods was observed in the field trial too (Figure 3 .7). 3.4.4 Effect of plant roots on soil P fractions The roots exerted a profound effect on the phosphate chemistry of the surrounding soi l . In both trials, reductions in resin-P, NaOH-Pi and H2S04-Pi concentrations were observed near the rhizoplane for all treatments (Figure 3 . 8, 3 .9, 3 . 1 0) . The zone of these reductions did not extend to more than 2 - 3 mm from the rhizoplane. As in the case of soil pH, no change in the concentration of these fractions was observed at any distance from the polyester mesh in RSC without plants suggesting that roots were the cause for the depletion of these fractions. The resin-P depletion in the rhizosphere of camellia roots created a P concentration gradient between the bulk soil and the root surface. This gradient is the driving force for a flux of P diffusing towards the rhizosphere from the bulk soil and the desorption! dissolution of solid phase P forms. The rate at which P moves to the root is reduced if the soil P buffering capacity i s high. An additional factor influencing the depletion of H2S04-Pi fraction will be the rate at which H+ ions can diffuse from their source at the rhizoplane into the surrounding soil (Nye, 1972). Resin-P and NaOH-Pi were the dominant forms of P depleted from the rhizosphere in soil fertilised with soluble P fertilisers in both trials. Plant and microbial P uptake could be considered as the major causes for the low resin-P in rhizosphere soil rather than P-fixation by Fe and AI because the NaOH-P; fraction, which is a measure of Fe-P and AI-P, also decreased. A significant decrease of H2S04-Pi near the rhizoplane in NCPR treated soil was due to the higher dissolution of NCPR at the low pH encountered in the rhizosphere. These results show that camellia roots possess the ability to utilise P from difficultly available P forms in the soil i . e. NaOH-Pi and H2S04-Pi. Hedley et aI . ( 1 994) also reported similar P depletions in the NaOH-Pi and H2S04-Pi fractions by upland rice in tropical soils . Attempts to model the uptake of P by these plants must consider the depletion of these insoluble P forms in rhizosphere soil as well as the diffusion of more labile P forms to plant roots. 8 6 Phosphorus fractionation of soils from the glass house trial with resin extraction as the initial step showed that the proportion of NCPR that dissolved near the rhizoplane and in the bulk soil (7 - 8 mm) were 5 1% and 4 1% respectively (Table 3 .2) . The corresponding figures for the field trial were 60% and 54%. As mentioned in the earlier section, P fractionation with an initial resin extraction resulted in higher NCPR dissolution than when NaCl/TEA (pH 7) was used for the initial extraction (Tambunan et ai. , 1 993) . The NCPR dissolution by the latter method was nearly 1 3 - 1 5% lower compared with the former method in both the rhizosphere and bulk soil of the glasshouse and field trials (Figure 3 . 7) . It also showed that 6 - 8% more P was dissolved from NCPR in the 0 - l . 5 mm zone of the rhizosphere compared to the bulk soil in both trials. As well as the lower pH of the rhizosphere soil, the removal of dissolved P and Ca by plant uptake also contributes to the higher P dissolution of NCPR in the rhizosphere. Unlike the observations made with clover by Tarafdar and Jungk ( 1 987) and with rape by Gahoonia and Nielsen ( 1 992) an accumulation ofNaOH-Po in the rhizosphere was observed in my study for all the treatments in both glasshouse and field trials (Figure 3 . 1 1 a and b) . Armstrong and Helyar ( 1 992) also observed an increase in NaOH-Po in the rhizosphere of several grass species in Mulga shrublands of South Western Queensland. Trolove et al . ( 1 996a) have also observed this with pastoral legumes. The deposition of Po near the rhizoplane could be due to the assimilation of Pi by the microorganisms, which use the organic anions secreted by camellia roots. Xiaoping ( 1 994) reported that Camellia sinensis L. secretes large amounts of low molecular weight organic acids such as oxalic, succinic, malic and citric. These organic anions excreted by camellia roots may be acting as a source of carbon for the growth and multiplication of the micro-organisms in the rhizosphere. The P that is organically bound in microbial tissues may be released at a later time and become available for plant uptake following mineralisation (Gahoonia and Nielsen, 1 992). Generally, phosphatase enzyme activity is higher at the root-soil interface than in the bulk soil (Tarafdar and Jungk, 1 987). This enzyme is expected to hydrolyse NaOH-Po and convert it to inorganic P, thus decreasing NaOH-Po in the rhizosphere. It appears in this study that the rate of immobilisation of Pi is faster than the conversion of Po to Pi. 8 7 This may be due to a high concentration of Pi in the rhizosphere and low rate of plant uptake of Pi. Hedley et al . ( 1994) also observed that an increase of phosphatase activity near the rhizoplane was not matched by an increased Po depletion. 3.4.5 Comparison of rhizosphere P depletion with P uptake processes Two Pi uptake processes occur, plant P uptake and conversion of Pi to Po in the rhizosphere soil . Both are a function of root activity, because the P depletion or transformation of Pi to Po did not occur to any significant extent at more than 3 mm from the roots. Net plant P uptake from the rhizosphere can be estimated from the total amount of P depleted in the RSC compartments (lJ.g RSC-1) . The P depleted from lower RSC compartment was estimated (Table 3 .4) by taking the difference between the total amount ofP in the slices near the root (0 - 3 mm) and total amount of P at 3 - 8 mm from the root (bulk soil - uninfluenced by rhizosphere processes). Root activity in the unfertilised treatment and soluble P fertiliser treatments depleted most of the P from resin-P (30 - 3 8%) and NaOH-Pj (50 - 60%) pools, while the contribution from the H2S04-Pi pool was only 1 5%. Root activity in the NCPR treatments depleted P mostly from the H2S04-Pi fraction (44%) while depletion of other forms were comparatively low (28% resin-P and 28% NaOH-Pj ) . Total plant P uptake was estimated by subtracting the initial P content from the final P content of the plants by destructive sampling of plants of similar size to those of test plants at the start of the trial. About 40 - 60% of plant P uptake in the soluble P fertiliser treatments was derived from the rhizosphere in the lower compartment, whereas it was 3 7% in the NCPR and control treatments (Table 3 .4). The plant P uptake unaccounted for by the P depletion in the lower compartment came from the P depletion in the upper compartment. Almost all the fine roots in the upper compartment were found lying on the mesh, where the lower and upper halves of the surface of these roots were assumed to have caused equal amounts of P depletion in the soil. Therefore the total P depletion in the soil is expected to be about twice that in the lower compartment, which agrees reasonably well with the amount ofP removed by the plant. Table 3.4 Comparison of phosphorus depletion from soil in the lower compartment ofRSC with plant P uptake in the glasshouse trial (values represent the mean of 5 replicates) Control NCPR DAP SSP MCP Depletion (f.1g RSC-l) Resin-P 265 3 1 1 477 503 479 NaOH-Pj 526 328 567 6 1 3 7 1 3 H2SO4-Pi 1 2 1 556 224 1 92 236 Total 9 1 2 1 1 95 1 268 1 308 1 428 Accumulation (Ilg RSC-1) NaOH-Po 530 507 43 1 470 496 Net depletion ( Ilg RSC-I) 3 82 688 837 838 932 Initial plant P content (Ilg porI ) 3 907 6240 5878 666 1 469 1 Final plant P content ( Ilg porI) 4980 8060 7560 8760 6200 Net P uptake ( Ilg porI) 1 073 1 820 1 682 2099 1 509 8 8 --- ------------- 8 9 Both NCPR and SSP significantly increased total root length and decreased root radius, which led to an increase in root surface area (Table 3 . 5) . Total root surface area is an important property for effective contact between soil nutrients and root surface (Barber, 1 995). The results show that depletion ofP by camellia trees per unit surface area of roots, increased with P fertiliser application, the increase being statistically significant for SSP application (Table 3 .5) . Thus the greater P uptake was caused by P fertiliser addition concentrating the pool of plant available P (see resin Pi Figure 3 . 6a, b; Table 3 .4) and greater root growth in the zone of P fertilisation. This effect of P fertiliser on root growth has been reported by Barber ( 1 995) and Fbhse and Jungk ( 1 983) . 3.4.6 Limitations of the RSC technique When using the technique to study the long-term effects of tree roots on chemical changes in the rhizosphere care should be taken to avoid penetration of roots of weeds and other plants into the RSC. Also the interruption of root growth and proliferation of lateral rootlets at the RSC mesh boundary causes a planar root surface which puts a high P-demand on the soil . This allows small changes in the chemistry of the rhizosphere to be detected, but causes larger chemical changes than would be expected from the radial geometry around single roots in the field. The difference in the radial and planar effects is mainly in the magnitude of the measured changes in the rhizosphere and not on the rhizosphere processes. Similarly proliferation of roots may result at lower soil moisture than might occur around single roots . In addition the angle at which the RSC is buried in the soil may affect the moisture status inside the RSC if water infiltration is intercepted. 9 0 Table 3.5 Effect ofP fertiliser forms on root growth within 1 mm of the mesh and P depletion in the field trial (values represent the mean of 5 replicates) Treatment Root length Root radius Root surface P depletion per (em) (em) area (em2) root surface areal (�g cm-2) Control 1 80 0 .039 44. 1 8 . 5 NCPR 400 0 .025 62.9 9 . 8 SSP 380 0 .026 56.4 1 2 . 3 Lsd (P<0.05) 1 92 0 . 006 1 5 . 6 2 . 3 Ip depletion i n the RSC containing no roots was multiplied by 2 to include P depletion in the RSC containing roots 3.5 CONCLUSIONS 9 1 The root study container (RSC) technique of Kuchenbuch and Jungk ( 1 982) adopted for camellias under glass house conditions and modified for the field situation gave similar results and provided useful information on the rhizosphere processes involved in P utilisation by camellia seedlings and mature trees. The chemistry of phosphorus in the camellia rhizosphere differs distinctly from that of the bulk soi l . Camellia roots induce acidification in their rhizosphere. Plant induced acidification in the rhizosphere created conditions conducive to the dissolution of the sparingly soluble NCPR fertiliser. Addition ofP fertilisers increased resin-P, NaOH-Pi (Fe and AI bound P) and H2S04- Pi (Ca bound P) in the soils. Camellia plants utilised NaOH-Pi and H2S04-Pi, which are traditionally considered as not readily available for plants. The accumulation of NaOH-Po in the rhizosphere suggests that soluble inorganic P was transformed into organic P forms as a result of increased microbial activity in the rhizosphere promoted by the secretion of carbon exudates from the roots. This can be of considerable benefit to the plant on a long-term, because immobilization of organic P by the microbial biomass would concentrate P in the rhizosphere and subsequent mineralisation of this organic P would release this immobilised P into plant available forms. Phosphorus fertilisation increased P uptake due to increased root growth and increased concentration of plant available P pool in the rhizosphere. 9 2 CHAPTER 4 EPPAWALA PHOSPHATE ROCK DISSOLUTION AND TRANSFORMATION IN THE RHIZOSPHERE OF TEA (Camellia sinensis L.) COMPARED TO OTHER SELECTED PLANT SPECIES 4.1 INTRODUCTION Plant species differ in their ability to utilise soil and fertiliser P due to differences in the chemical characteristics of their rhizospheres (Aguilar and Van Diest, 1 98 1 ; Fried, 1 953) . Several complex mechanisms are involved in plant P uptake from the rhizosphere soil depending on the plant species (Bekele et al. , 1 983 ; Diest et al. , 1 97 1 ) . Plants vary in the proportions of the uptake of cations and anions resulting in changes in rhizosphere pH, which influences phosphate solubility and P uptake by the plant (Nye, 1 98 1 ) . The release of enzymes by plant roots, which hydrolyse labile organic P compounds in the rhizosphere and increase plant-available P may also vary with plant species (Gahoonia and Nielsen, 1 992; Tarafdar and Jungk, 1 987). Tea (Camellia sinensis L.) is mostly grown in highly weathered acidic ultisols of the humid tropics (Ranganathan and Natesen, 1 985) . These soils contain significant quantities of Fe and AI oxides and hydroxyoxides (Golden et al. , 1 98 1 ; Karim and Rahman, 1 980), which are known to fix P . The low pH and high P fixing capacity of these soils are favourable for the dissolution of phosphate rocks (PR) (Bolan and Hedley, 1 989; Chien et aI . , 1 990b; Mackay et aI. , 1 986). Therefore tea is generally fertilised with sparingly soluble PR fertiliser sources (Ranganathan, 1 977a; Sivasubramaniam et aI . , 1 98 1 ). In Chapter 3 it was shown that camellia (Camellia japonica L.) plants, which are of the same family as tea, when grown in a Dystric Eutrochrept soil from New Zealand, were able to dissolve more North Carolina phosphate rock (NCPR) in the rhizosphere (37% of added P) in 56 days compared to that in the bulk soil (27% of added P). In this Chapter the rate of dissolution of a Sri 9 3 Lankan PR, Eppawala phosphate rock (EPR) i n the rhizosphere of tea grown in a highly acidic Sri Lankan Ultisol is compared with that in the bulk soil. Plants normally grown in association with tea have diverse growth habits . Calliandra (Calliandra calothyrsus L.), a leguminous tree crop is grown in tea fields to provide shade and organic material, which helps to conserve soil moisture (Anon, 1 990). Guinea grass (Panicum maximum L.) is usually found on abandoned sloping tea lands, which helps to control soil erosion. Bean (Phaseolus vulgaris L.) is a common leguminous vegetable crop cultivated in the same soil type in adjoining farm gardens. Generally the supply of P for tea is from sparingly soluble EPR and for beans it i s provided by a soluble P fertiliser. The intention of including crops of differing growth habits in this study was to provide information on the effect of slow and fast growing plants, as well as leguminous and non-leguminous species, on rhizosphere acidification and to test how best these plants could utilise P from the locally mined sparingly soluble PR, Eppawala phosphate rock (EPR) . There is no information available in the literature on the relative efficiencies of these crops in utilising nutrients from acidic Ultisols particularly in the soils of tea-growing areas. 4.2 OBJECTIVES The objectives of the study reported in this chapter are: 1 . To compare root induced changes in pH and P fractions in the rhizosphere of tea and other plants. 2. To quantify the release ofH� into the rhizosphere by these plants and assess their influence on PR dissolution in the rhizosphere. 3 . To understand the diversity of mechanisms involved in P uptake and the utilisation efficiency of these plants which have diverse growth habits. -- - ----------- �-------- � - - 4.3 MATERIALS AND METHODS 9 4 The soil used in this study was a Rhodustult (De Alwis and Panabokke, 1 972) collected from the top 75 mm depth after removing the litter layer of soil in a tea estate in the Southern part (Kottawa) of Sri Lanka. The selected physico-chemical properties of the soil are presented in Table 4 . 1 . The air-dried soil was passed through a 2 mm sieve and mixed with Eppawala phosphate rock (EPR, particle size 5 . 2% > 250 11m; 39 .2% 1 50 - 250 11m; 39 .9% 1 50 - 75 11m; 1 5 . 7% < 7511m, total P 1 4 . 5%, citric acid (2%) soluble P 1 . 97%, almost insoluble in water and locally mined in Sri Lanka) at the rate of 200 Ilg P gol soil. Urea and KCI fertilisers were mixed with the soil at the rate of 200 Ilg N or K gol soil before planting. The root study container (RSC) technique used for camellias in the glasshouse described in Chapter 3 was used in this study (Figure 4 . 1 ) . The upper and lower compartments were packed with l 3 5 and 242 g soil (bulk density of 1 . 1 Mg m-3) respectively. Vegetatively propagated 5 months old tea (TRI 2025) plants, 5 months old calliandra (Calliandra calothyrsus L.) plants, Guinea grass (Panicum maximum L.) cuttings and bean (Phaseolus vulgariS L.) seeds were planted in the upper compartment. The four plant species, treated with EPR and a control without EPR fertiliser were considered as treatments . The treatments were replicated 4 times and arranged in a randomised block design in a glasshouse maintained at 1 2° C minimum and 26° C maximum temperatures at St. Coombs, Sri Lanka. Four RSCs (fallow) having the same N, P and K treatments as the planted pots, but this time with no plants, were used to study any changes in the soil in the absence of plants. All RSCs were placed on a fine sand bed, which was kept moist by a water reservoir. The water level in the reservoir was fixed at 1 60 mm below the base of the RSCs. This enabled the RSCs to be maintained at a constant water potential of approximately - 1 . 6 kPa. 4.3.1 Soi l and plant sampling After 3 5 days of planting, plant shoots were cut 5 mm above the soil surface. The increase in plant dry matter during the trial period was estimated by taking the difference between final and initial dry weights of the seedlings. ---------------- - -- Table 4. 1 Selected physico-chemical properties of the soil used in the study Property Sand Silt Clay Soil pH Bulk density pH buffering capacity (at pH 4-5) Organic C Effective CEC 1 ,2 Ex. Ca Ex. Mg Ex. K Ex. Na Ex. AI Olsen-P P-fixing capacit/ Unit % % % Soil : water ( 1 : 2 . 5 w/w) Mg m-3 mmol H+ kg-l pIt! % emole kg-! emole kg-J emole kg- l cmole kg- ! emole kg-! emole kg-l J..lg g- l soil % lExchangeable (Ca + Mg + K + Na + AI + H) 2Blackmore et al . ( 1 987) Value 60 1 5 25 4 . 5 l . 1 3 0 l .70 4 .0 1 0 . 2 1 0 . 3 1 0 . 1 5 0 . 1 1 1 . 3 5 20 93 9 5 Figure 4.1 The arrangement of plant species in the glasshouse experiment (A - Guinea grass, B- Calliandra, C - Bean and D - Tea) 9 6 9 7 The soil in the lower compartment was sliced in thin sections as described in Chapter 3 . A 2 .0 mm thick soil slice in the upper compartment immediately above the inter-cell boundary (24 Ilm polyester mesh) was also sampled and root length was measured according to the line intercept method described by Newman ( 1 966). Root surface area was determined by using the formula 2...J(rc ML/p) (where rc: 2217, M: weight of roots, L: length of roots and p: density of roots) and assuming that a root is a cylindrical tube with a constant radius. The true density of roots was determined from the volume and weight of the roots. The root volume was determined by the amount of water displaced when the roots were immersed in water. The lower surface area of these roots was considered responsible for the observed changes in pH and soil P fractions in the lower compartment. The bottom view of RSCs showing roots of various plant species are presented in Figure 4.2, 4 .3 , 4.4 and 4 . 5 . 4.3.2 Plant and soil analysis Plant samples were separated into shoots and roots and dried at 600 C, weighed and ground to powder. The plant materials were dry-ashed at 5500 C, and thereafter the ash was taken up in 0.05 M HCI solution and analysed for P by the vanadamolybdate method (Jackson, 1 958), and Ca by atomic absorption spectrophotometry. The initial nutrient composition and dry weights of the plants were determined by destructive sampling of ten randomly selected plants of similar size to the test plants at the beginning of the study. The soil was air-dried and analysed for pH ( 1 : 2 . 5 soil : water ratio using a pH meter) . Resin-P, NaOH-Pi, NaOH-Po, H2S04-Pi and residual P were determined by the sequential P fractionation scheme of Hedley et al . ( 1 994) described in Chapter 3 . The amount of EPR dissolution in the soil was determined by the method of Tambunan et al . ( I 993) as described in Chapter 3 . The statistical analyses were performed according to the procedures of the SAS systems (SAS Institute, 1 985) . GUINEA GRASS - EPR Figure 4.2 Roots mats formed on the polyester mesh for Guinea grass in EPR and control treatments 9 8 Figure 4.3 Roots mats formed on the polyester mesh for bean in EPR and control treatments 9 9 CALLIAN DRA - CONTROL Figure 4.4 Roots mats formed o n the polyester mesh for calliandra in EPR and control treatments 1 0 0 TEA - CO NTR O L Figure 4.5 Roots mats formed on the polyester mesh for tea in EPR and control treatments 1 0 1 4.4 RESUL TS AND DISCUSSION 4.4.1 Effect of plant species and EPR on growth characteristics 1 0 2 Significantly higher (p <0.05) dry matter yields (final dry matter - initial dry matter) were obtained for bean and Guinea grass compared to tea and calliandra plants, whether EPR was applied to the soil or not (Table 4 .2), with bean showing the highest dry matter production. Bean being a short-term vegetable crop and Guinea grass being a fast growing grass species produced larger amounts of dry matter than the other two plant species within a short period. Tea and calliandra on the other hand are perennials and their dry matter production was less due to a slower rate of growth . The shoot : root ratio was in the same order as dry matter production (bean > Guinea grass > calliandra and tea). EPR application did not increase dry matter yield significantly in any of the plant species (Table 4 .2) . The P fertiliser application however increased shoot P concentration significantly (p <0.05) in bean and Guinea grass. 4.4.2 Effect of EPR ferti l iser on soil pH Application of EPR had no significant effect on soil pH (Figure 4 . 6) . In contrast to EPR, NCPR application significantly increased soil pH in a soil with lower pH buffering (Chapter 3) . The dissolution of EPR in the bulk soil is only about 1 0% P (Figure 4 . 7) compared to 27% P dissolution from NCPR in the trial reported in Chapter 3 . Therefore the consumption of soil acidity for EPR dissolution is much lower than that for NCPR dissolution and hence caused an insignificant rise in the bulk soil pH for EPR treatment. The lower reactivity of EPR ( 1 3 . 5% of total P dissolved in 2% citric acid) compared to NCPR (32% of total P dissolved in 2% citric acid, White et aI . , 1 989) and the absence of accessory carbonate minerals in the apatite structure ofEPR (Dahanayke et aI . , 1 995) (NCPR has 1 1 . 7% CaC03, Syers et aI . , 1 986) may have been the reason for the lower effect of EPR on soil pH. Ta hie 4.2 Comparison of plant dry matter yield and shoot : root weight ratio and P concentration of tea with those of other plant species Plant dry matter (g) Shoot : Root P concentration (%) Plant species Treatment Shoot Root Total « by weight) Shoot Root Tea -P 0.42 0 .26 0.68 1 . 62 0 . 1 47 0.062 Calliandra (Control) 0 .27 0. 1 6 0.43 1 . 68 0 . 1 1 5 0.078 Guinea grass 2 . 80 0 .98 3 .78 2 .85 0.072 0.072 Bean 5 . 1 7 0. 8 1 5 .98 6 . 38 0 .024 0 . 1 03 Lsd (p <0.05 ) 0.22 0 .2 1 0 .33 0 .88 0 .005 0 .004 Tea + P 0.49 0 .39 0.88 1 . 26 0 . 1 56 0.075 Calliandra (EPR) 0 .34 0 .20 0.54 1 . 70 0 . 1 25 0.093 Guinea grass 3 . 29 1 . 34 4.63 2 .45 0 . 1 02 0. 1 09 Bean 5 . 54 0 .99 6 .53 5 . 59 0 .052 0 . 1 1 4 Lsd (p <0.05) 0.28 0. 1 8 0 .44 0 .52 0 .005 0.002 o C/) (a) (b) 4.90 4 .85 4.90 1 N N N N N N 4.85 1 N N N N N N 4.80 4.80 4 .75 4.75 4 .70 4.70 4.65 4.65 4 .60 4.60 --+- Tea 4 .55 4 .55 ___ Call1ondra -.- Gul ... grass 4 .50 4 .50 ---.- Bran -e- Fallow 4.45 4.45 4.40 -+-----r---r-----r---r-------, 4.40 -+-------,,...----,-------,,...----r-------, o 1 2 3 4 5 o 1 2 3 4 5 Distance from rhizoplane (mm) Distance from rhizoplane (mm) Figure 4.6 Effect of plant species on rhizosphere pH ( 1 : 2 . 5 w/w H20) of (a) control and (b) EPR fertil i sed soi l s . Vertical bars and N represent Lsd at p <0.05 and treatments that are not stati stical ly significant at p <0.05 respectively . 25 • Tea • CaJliandra • Guinea grass 20 £ Bean ,.-.., 0... -0 Q) -0 -0 � ....... 0 � 1 5 '-'" 0... -0 Q) ..2: 0 en en Cl 1 0 o 1 2 3 4 5 Distance from rhizoplane (mm) Figure 4.7 Effect of plant species on EPR di ssolution in the rhi zosphere. Vertical bars represent standard errors of the means . 1 0 5 4.4.3 Effect of plant roots on soil pH and EPR dissolution 1 0 6 Soil pH decreased in the rhizosphere compared to that in the bulk soil for all plant species, but the magnitude of reductions varied distinctly among the plant species (Figure 4 .6) . The magnitudes of the pH reductions in the rhizoplane of the plant species treated with EPR were 0 .3 1 , 0.26, 0. 1 8 and 0.2 1 for bean, tea, Guinea grass and calliandra respectively. These results show that all four plant species released Ir in the rhizosphere. Part of the H+ released was consumed in the dissolution of PR in soils and the balance of H+ caused pH reductions in the rhizosphere. The amounts of Ir released into the rhizosphere (0 - 3 mm) soil in the lower compartment ofRSC was calculated by adding the Ir release calculated from the pH decrease in the rhizosphere compared to the bulk soil and the amounts of Ir that were consumed for the dissolution of EPR in that zone (Table 4 .3) . The amounts of Ir released into the rhizosphere corresponding to the pH decrease was estimated, by taking the difference in pH in the bulk soil and each of the rhizosphere soil slices. Then multiplying the respective pH differences by the soil pH buffering capacity and the weight of the corresponding soil slice, and summing the values for all the soil slices within the 0 - 3 mm distance zone from the rhizoplane. The amount of Ir consumed in dissolving EPR in the rhizosphere and bulk soil was estimated using the amount of EPR dissolved and from the relationship that 2 moles of W were consumed for every mole of P dissolved (%P in EPRlatomic weight of P * 1 1 100 * 2 = 0.00933 /lmol H+ /lg-1 EPR dissolved) . Mineralogical analysis of EPR using XRD showed that EPR has no detectable amounts of free carbonates (CaC03 or MgC03) (Tazaki et aI . , 1 987), therefore all acids consumed in the dissolution of EPR were assumed to be due to the reaction of acids with the apatite in EPR. This calculation was also done for each soil slice within the 0 - 3 mm distance zone from the rhizoplane and the values summed-up to obtain the W consumption for EPR dissolution in the rhizosphere. Guinea grass released the highest amount of acidity into the lower compartment of RSC compared to all the other plant species, but because of its larger root surface area (Table 4 . 3 ) - the amount of acidity produced per unit surface area of roots is lowest for Guinea grass. Guinea grass produced the lowest reduction in pH measured Table 4.3. The total acid production by roots of different plant species treated with EPR into the rhizosphere of the lower compartment pH drop Observed H+ Average ofEPR H+ consumption for Total H+ Surface area Amount of in the production within dissolution within EPR dissolution production of roots W released rhizosphere 0-3 mm of the 0-3 mm of the within 0-3 mm of in (0-3 mm) of within per unit surface Plant species (0-3 mm) rhizosphere in rhizosphere in the rhizosphere in the rhizosphere 0-2 mm area of roots6 compared to excess of bulk excess of bulk excess of bulk soil4 in excess of above the (flmol W cm-2) bulk soil ' soie soil3 (flmol H+) bulk soil5 mesh (/lmol H+) (P%) (/lmol H+) (cm2) Bean 0 .26 ± 0 . 0 1 1 5 .8 ± 0 .99 1 3 .28 ± 1 . 60 6 .69 ± 0 . 53 22.49 ± 2 . 5 526 0 .0427 Tea 0 .24 ± 0 .0 1 1 5 .3 ± 0 . 52 1 3 .45 ± 1 .3 1 6 . 67 ± 0 . 59 22.00 ± 2 .4 270 0.08 1 4 Guinea grass 0. 1 7 ± 0.0 1 1 0 . 1 ± 0 .79 1 8 .05 ± 1 . 82 1 7.83 ± 1 . 30 27.99 ± 3 . 5 1 1 96 0.0234 Call iandra 0. 1 7 ± 0 .02 9.6 ± 1 . 27 1 0 .67 ± 0.66 4 .66 ± 0.86 14 .3 1 ± 2 .3 256 0.0558 , Weighted mean pH of all slices within 3-5 rum from the rhizoplane minus weighted mean pH of al l sl ices within 0-3 rum from the rhizoplane. 2 Sum total of (change of pH in a rhizosphere soil sl ice compared to bulk soil {3-5 mm} • soil pH buffering capacity {30 Ilmol W g-' pH' soi l } . weight of that soil sl ice) for all sl ices in the rhizosphere (0-3 rum). 3 %P dissolved in the rhizosphere in excess of bulk soil . 4 Amount of EPR dissolved (Ilg EPR g " soil) . amount of W required to dissolve I Ilg EPR (0.00933 Ilmol W Ilg-' EPR - see text 4 .4 .3) • soil weight. 5 Observed W release in the 0-3 nun of the rhizosphere calculated from pH drop plus the amount of W consumption during EPR dissolution in the rhizosphere soils . 6 Total W released into the rhizosphere divided by the surface area of roots lying above the mesh within 0-2 rum * 0 .5 . [The multiplication factor 0.5 i s used because only the lower half of the surface area of the roots on the mesh is responsible for the H+ release in the lower compartment] . 1 0 8 in water in the rhizosphere. This is because more PR dissolved in the Guinea grass rhizosphere compared to that in other crops (Figure 4 .7) and this caused more consumption of Ii for PR dissolution giving rise to a higher pH in the Guinea grass rhizosphere. Bean, being a legume obtains its N requirement through atmospheric N2 fixation and very little through N03- uptake. The low N03- uptake by this plant may have caused a high cation-anion balance in the plants producing high acidity in the rhizosphere (Haynes, 1 992). The total amount of acidity produced by bean is only second to Guinea grass. Calliandra is also a legume like bean, but it did not produce as much acidity as bean. This may be due to the low growth rate and smaller root surface area of calliandra (Table 4 .3) . The acidity produced per unit root surface area of calliandra is however not much different from that of bean. Tea, inspite of being a non-legume reduced rhizosphere soil pH more than calliandra. The acidity produced per unit root surface area of tea is the highest among the four crops. This may be because tea plants either took up NRt + more than N03 - from the soil as suggested by Ishigaki ( 1 978) or tea roots excreted significant amount of organic acids (Xiaoping, 1 994) or both these reasons. Ishigaki ( 1 978) reported that tea roots have a low concentration of nitrate reductase and therefore prefer to take up NH/ to N03-. This suggestion has been tested and the results are presented in Chapter 6. The Ii released by plant roots into the rhizosphere contributed to more PR dissolution in the rhizosphere compared to that in the bulk soil (Figure 4.7) . The dissolution of PR was highest in the rhizosphere of Guinea grass (23% P dissolved near the rhizoplane) and lowest for calliandra ( 1 3% P dissolved near the rhizoplane). The higher dissolution of EPR in Guinea grass was attributed to the higher amount of acid release into the rhizosphere of these plants as a result of higher amounts of roots on the mesh. The higher amounts of Guinea grass roots may also have removed higher amounts of P and Ca, the dissolved products of EPR, causing more EPR dissolution compared to that in the rhizosphere of other plant species. A limitation in the calculation of acid release in the rhizosphere is that pH was determined in water. Changes in soil solution chemistry may have an effect on pH, without differences in 1 0 9 exchangeable acidity occurnng. In future studies it may be more appropriate to measure the rhizosphere pH and determine pH buffering capacity in 0 .0 1 M CaCho 4.4.4 Effect of EPR addition and plant roots on soil P fractions Increases in the P fractions in bulk soil treated with EPR (the zone unaffected by plant roots) compared to unfertilised bulk soil, accounted for nearly 1 00% of the added EPR (Table 4 .4). The dominant P fraction in the EPR treated bulk soil was H2S04-Pi (76 - 80%). According to the findings of Tambunan et al. ( 1 993), this indicated that the majority of the added P remained as undissolved EPR in the soils at the end of the trial . The amounts and forms of the P fractions extracted from soil slices close to the mesh boundary of planted and unplanted RSCs were compared. Planted RSCs showed distinct changes of P form and amount in soil 0 - 3 mm from the mesh (Figure 4 .8, 4 .9, 4 . 1 0 and 4. 1 1 ) . Similar plant induced effects on the rhizosphere chemistry in relation to P uptake were reported for camellia (Camellia japonica L.) in Chapter 3 . In the rhizosphere, the amount and nature of the P fractions varied with the plant species. The absence of plants (fallow RSCs) showed no difference in any of the P forms with distance from the polyester mesh. Therefore any difference observed in the P fractions between the bulk soil and the rhizosphere was due to the influence of plant roots. Phosphorus uptake by plant and microorganisms depleted resin-P (P in soil solution and P loosely sorbed to soil minerals) near the roots (Figure 4. 8) . The steep resin-P depletion profiles in the rhizosphere of bean and Guinea grass compared to calliandra were considered to be caused by the former growing at a faster rate and having a greater amount of roots at the inter-compartment boundary. Steep resin-P depletion profiles in the tea rhizosphere were due to plant P uptake as well as conversion of Pi to Po (NaOH-Po) by the active microbial population in the tea rhizosphere (see next paragraph) . The depleted P is replenished by desorption of P bound to Fe and AI oxides (NaOH-Pi) (Figure 4 .9) and dissolution of EPR (as indicated by a reduction in Table 4.4 The P fractions in the control treatment and % recoveryl of added P from EPR in the bulk soil P-fraction Control P recovery from (Jlg g- 1 soil) EPR (%) Resin-P 1 0 . 5 ± 1 . 1 3 . 3 ± 0. 5 NaOH-Pi 68 .2 ± 2 .5 8 . 5 ± 1 . 2 NaOH-Po 68.0 ± 2. 1 4 . 7 ± l .2 H2SO4-Pi 34 .2 ± 1 . 1 78 . 5 ± 2 .8 Residual-P 74. 2 ± 3 . 1 2 .2 ± 0 .8 Total-P 255 . 1 ± 5 .2 97 .2 ± 2 . 1 1 P recovery % = (amount of P in fertilised soil - amount of P in control soil) * 100 amount of fertiliser P added to the soil 1 1 0 ,-..., - (a) 25 20 ' 0 C/l 1 5 'bI) bI) :1. 0;< 1 0 . S C/l bean > tea > calliandra irrespective of P fertiliser addition. The higher external P efficiency of Guinea grass and bean (Table 4 .6) is associated with a greater root surface area and not due to greater P uptake per unit root surface area (Table 4 . 5) . Trolove et al . ( 1996a) also observed that higher external efficiency of white clover varieties was due to greater root surface area and not due to greater P uptake per unit surface area. Unlike other plant species, tea showed a significantly higher P uptake per unit surface area, and it may be because tea plants have mechanisms to extract more P from soil per unit surface area of roots compared to the others. 1 1 9 Table 4.6 Comparison of external and internal efficiencies ofP in plants with and without added P fertiliser Plant species Treatment External efficiency! (mg P) Shoot Root Tea -P 0 .62 1 0 . 1 6 1 Calliandra (Control) 0 . 3 1 9 0 . 1 24 Guinea grass 2 .024 0.7 1 1 Bean 1 .246 0 .839 Lsd (p <0.05) 0 . 3 1 4 0 . 1 76 Tea +P 0 . 773 0 .296 Calliandra (EPR) 0 .43 1 0 . 1 92 Guinea grass 3 . 3 64 1 .461 Bean 2 . 89 1 1 . 1 3 2 Lsd ( p <0 .05) 0 . 304 0 . 1 84 1 Plant P uptake by the respective tissue 2 Dry matter production of respective tissue Total P uptake Total 0 . 782 0 .443 2 . 73 5 2 .085 0 . 393 1 .070 0. 622 4 .825 4 .023 0 .456 Internal efficiencl (mg dry matter mg-! P) Shoot Root Total 537 3 3 1 868 623 3 54 977 1 025 3 59 1 384 2478 3 90 2868 88 49 50 462 364 826 5 5 1 3 30 86 1 683 277 960 1 377 246 1 623 76 52 95 1 2 0 Extrusion of organic acids by tea roots (Jayman and Sivasubramaniam, 1 975; Xiaoping, 1 994) thereby causing dissolution of EPR as well as desorption of P from soil (depletion of NaOH-Pj - see section 4.4 .4) may well be some of those mechanisms. 4.4.6.2 Internal efficiency of P util isation Significant differences between the plant species were observed in their internal P efficiencies for both control and EPR treatments. Bean plants had the highest internal efficiency and tea showed the lowest of the four plant species (Table 4 .6) . The plants with higher growth rates (bean, Guinea grass) could convert absorbed P into dry matter quickly compared to plant species having a slower growth rate (tea, calliandra) . For all plant species the internal P efficiency was higher in the absence of EPR fertiliser (control treatment) than with it. In a classical model of plant dry matter response to the increased supply of P, the highest amount of dry matter production per unit of P absorbed by the plant was at the lowest level of P supply. This implies that as the soil P concentration increases, the P uptake and dry matter production also increases, but the rate of increase in dry matter production per unit of P supplied decreases for each additional unit ofP absorbed. The above results show that genetic variability within and among plant species could be utilised to develop, and screen, new varieties that are suitable for a location and also could utilise locally available resources, such as phosphate rock, for sustainable production of the crops. However limitations of this method such as the initial weight of planting materials, differences in growth rates and the length of trial period may need to be taken into account when comparing plant species. The internal P efficiency index needs careful interpretation. Some apparently P efficient species may have severe P stress and not yield useful seed or marketable products . The practical potential of the above findings was studied in detail using 3 different tea clones and the results are presented in Chapter 5 . 4.5 CONCLUSIONS 1 2 1 All plant species studied acidified their rhizospheres. The magnitude of acidification varied with plant species and the extent of root growth. The plant induced rhizosphere acidification increased dissolution of EPR in that zone compared to the bulk soil . Guinea grass with the largest root mass caused the highest acidification in the rhizosphere and resulted in the highest EPR dissolution. But the rate of acidification per unit surface area of Guinea grass was the lowest among the four plant species, with tea producing the highest rate of acidification per unit surface area. It was common for all plant species to deplete resin-P and NaOH-Pj in the rhizosphere. Except for tea, all other species depleted NaOH-Po in the rhizosphere, with tea this fraction increased. These differences can be associated with the activities of phosphatase enzyme released to the rhizosphere from the roots of these plants or due to the relative rate of root C release. The organic P accumulation in the tea rhizosphere is probably due to P immobilisation by the increased microbial activity caused by the abundance of carbon exuded by the tea roots. Guinea grass and bean plants are externally more P efficient than tea and calliandra. This was largely caused by the differences in root surface areas. The plants which have greater root surface area could extract more P from the soil compared to those that have lower root surface area. The internal P efficiencies were in the order of bean > Guinea grass > calliandra > tea and it reflects the ability of each species to convert absorbed P into dry matter based on their genetic diversity. However the validity and use of this information on P efficiency of different plant species lies within the limits of the conditions under which the experiment was conducted. Long-term glasshouse and field trials are required to test these findings. 1 2 2 CHAPTER 5 PHOSPHORUS UTILISATION EFFICIENCY AND DEPLETION OF PHOSPHATE FRACTIONS IN THE RHIZOSPHERE OF THREE TEA (Camellia sinensis L) CLONES· 5.1 INTRODUCTION Tea is cultivated in many parts of the humid and sub-humid tropical regions of the world, mainly in acid soils having a pH (H20) of 4 . 5 - 5 . 5 (Othieno, 1992; Ranganathan and Natesan, 1 985) . At these low pHs, aluminium (Al) is highly soluble and reacts with P to form insoluble Al-P comolexes. Furthermore most tea soils are highly weathered and they contain large amounts of Fe and AI oxides and hydroxyoxides (Golden et aI. , 1 98 1 ; Karim and Rahman, 1 980), which are known to fix P. The precipitation of AI-P and the fixation of P in soils can cause a reduction in plant availability of P, from both native and fertiliser P sources. Phosphorus fractionation of Sri Lankan tea soils revealed that most of the applied P fertilisers are recovered as insoluble Fe and AI phosphates (Golden et aI. , 1 98 1 ) . Despite the high P fixation in these soils, tea plants do not generally suffer from P deficiencies. This suggests that tea plants have some mechanisms by which they are able to utilise the fixed soil P. The studies reported in Chapter 3 using camellia (Camellia japonica L.) plants, which is of the same family as tea, and also comparison of tea clone TRI 2025 with other plant species presented in Chapter 4 showed that tea plants can modify the root environment by acidifying the rhizosphere. The possibility of exploiting genotype differences for improving nutrient efficiency has received much attention in recent times (Fbhse et aI . , 1 988; Gahoonia and Nielsen, 1 996; Godwin and Blair, 1 99 1 ) . Phosphorus efficient genotypes can be useful for 1 Zoysa A K N, Loganathan P and Hedley M J (1998) Phosphorus utilisation efficiency and depletion of phosphate fractions in the rhizosphere of three tea (Camellia sinensis L.) clones. Nutrient Cycling in Agroecosystems (in print). 1 2 3 maintaining high productivity in low input agriculture. From a mineral nutrition point of view, a genotype is more efficient than others if it mobilises and absorbs more P from soils (external P efficiency) and! or makes better use of the absorbed P to produce greater biomass (internal P efficiency). Improvement of P efficiency by genetic selection of plants seems possible. Breeding of new crop genotypes with improved P efficiency may be a supplementary alternative to reducing the input of traditional amendments to soils such as the application of fertiliser (Batten, 1 992). New tea clones have been developed in many countries to increase yield, tolerance to drought and resistance to pests and diseases (Alam, 1 994; Anandappa, 1 986; Astika, 1 994; Barbora; 1 994). But there is no published information available on their relative efficiencies in utilising soil P. The external efficiency of P in plants is largely influenced by the size and distribution of the root system (O'Toole and B land, 1 987), the formation of root hairs (Fohse et al. , 1 99 1 ), root induced changes in the rhizosphere (Gahoonia and Nielsen, 1 992; Marschner et al. , 1 987), the kinetics of P uptake parameters (Nielsen and Barber, 1 978), soil moisture (Gahoonia et aI. , 1 994) and the plant's ability to associate with mycorrhiza (Morita and Konishi, 1 989; 2hi, 1 993). The slow diffusion ofP results in depletion ofP from the soil near the absorbing roots of the tea plants (Chapter 4). Therefore variation of P depletion profiles in the rhizosphere may provide information on the external efficiencies of P uptake in plants. The rhizosphere study container (RSC) technique described in Chapter 3 is used in this chapter to study the rhizosphere processes involved in P uptake by three tea clones having different genotypes. 5.2 OBJECTIVES The objectives of the study reported in this chapter are: 1 . To study the differences in the root induced changes in soil P fractions and soil pH in the rhizosphere of the three tea clones. 1 2 4 2. To detenmne the P utilisation efficiencies and screen the tea clones according to their ability to exploit soil and fertiliser phosphorus. 5.3 MA TERIALS AND METHODS The soil used in this study was a Rhodustult (Red yellow podsolic soil; De Alwis and Panabokke, 1 972) collected from Kottawa, Sri Lanka. The physico-chemical characteristics of this soil was presented in Table 4 . 1 , Chapter 4. The air-dried soil was passed through a 2 mm sieve and mixed with either triple superphosphate (TSP, total P 20% and 85 - 95% of total P soluble in water) and Eppawala phosphate rock (EPR, particle size 5 .2% > 250 11m; 39 .2% 1 50 - 250 11m; 39 .9% 1 50 - 75 11m; 1 5 . 7% < 75 11m, total P 14 . 5%, citric acid (2%) soluble P 1 .97%, locally mined in Sri Lanka) at rates of 200 I1g P g-I soil. Nitrogen (N) and potassium (K) fertilisers were mixed with the soil at the rate of 200 Ilg N and K g-l soil in the form of urea and KCI respectively before planting. The rhizosphere study container (RSC) technique described in Chapter 3 is used in this study. The upper and lower RSC compartments were packed with 1 3 5 g soil (bulk density 1 . 1 Mg m-3) and 242 g soil (bulk density 1 . 1 Mg m-3) respectively. Vegetatively propagated 5 month old tea plants of clones S 1 06, TRI 2025 and TRI 2023 were planted in the upper compartment of the RSC. These tea clones were selected for the study based on their differences in yield potential, resistance to pest and diseases or tolerance to drought (Anon, 1 994abcd) . The treatments used in this study were the three tea clones, two P fertilisers and a control (no P treatment) . These treatments were replicated 4 times and arranged in a randomised complete block design in a glasshouse maintained at 1 2° C minimum and 26° C maximum temperatures at S1. Coombs, Sri Lanka. Four replicated RSCs without plants (fallow) were also included in the experiment to study the changes in soil P fractions in the absence of tea plants. All RSCs were kept on a fine sand bed which was kept moist by a water reservoir. The water level in the reservoir was fixed 1 2 5 at 1 60 mm below the base of the RSCs and soil moisture in pots were maintained approximately at a constant potential of - 1 .6 kPa. 5.3.1 Soil, plant and root sampling After fifty six days of plant growth, the plant shoots were cut 5 mm above the soil surface. The soil and plant roots were sampled according to the methods described in Chapter 4 section 4 . 3 . 1 . 5.3.2 Plant and soil analysis Plant samples were separated into shoots and roots, dried at 60° C weighed and ground to powder. Both shoot and root samples were analysed for total N by Kjedhal digestion (Jackson, 1 958). Soil pH, soil P fractionation and plant P concentrations were determined as described in Chapter 4 section 4 . 3 .2 . 5.4 RESULTS AND DISCUSSION 5.4.1 Dry matter yield, P and N concentration of tea clones At the end of 56 days of plant growth, TRl 2023 and TRl 2025 had significantly (p <0.05) greater shoot and root dry matter yields than S 1 06 for the two P fertiliser treatments as well as control treatment (Table 5 . 1 ) . The differences in dry matter production in the tea clones could be attributed to their genetic variability. Application ofP and types ofP fertiliser showed no significant effect on dry matter yield. The shoot P concentrations were higher in TRl 2023 and TRl 2025 compared to S 1 06 in all the treatments (Table 5 . 1 ) . This may be attributed to increased root growth, which helped root exploration of a larger soil volume than the roots of S 1 06. The lower P concentration in the roots of TRl 2023 and TRl 2025 in soils treated with P fertiliser may be due to a dilution effect caused by increased root weight . The plants did not show any N deficiency symptoms during the course of the study. The N Table 5. 1 Dry matter yield, N and P concentration in shoots and roots, and P utilisation efficiencies of three tea clones Plant P Internal P Clone Treatment Shoot Root Total root uptake per External P efficiency surface area root surface efficiency (Shoot dry wt g Dry wt (g) N% P% Dry wt (g) N% P% (cm2) area (mg P planr l ) per rug total P (Ilg cm'2) uptake) S l 06 Control 0.482 a 1 . 5 1 a 0 . 1 1 7 a 0 . 372 a 1 .26 a 0 .049 a 242 3 .22 0 .779 a 620 a TRI 2025 1 .240 b } , 73 b 0 , 1 26 b 0 ,95 7 b 1 . 5 1 b 0 ,052 a 257 3 ,95 2 , 03 1 b 6 1 1 a TRI 2023 1 .525 b 1 . 77 b 0 , 1 30 b 1 . 1 77 b 1 .43 b 0 ,067 b 305 3 .4 1 2 , 7 1 8 c 562 b S 1 06 EPR 0 ,525 a } , 56 a 0 , 1 27 a 0 .405 a 1 .29 a 0,079 b 402 5 ,05 1 ,0 1 5 a 5 19 a TRI 2025 1 . 394 b 1 . 89 b 0 . 136 a 1 ,076 b 1 .45 b 0 ,056 a 484 5 ,07 2 ,455 b 569 a TRI 2023 1 .670 b 1 . 75 a 0 , 149 b " 1 ,289 b 1 .54 b 0 ,065 a 579 4 ,63 3 ,296 c 5 1 8 a . . S l 06 TSP 0 .544 a 1 .65 a 0 , 1 30 a 0 .420 a 1 . 36 a 0 ,074 a 568 4 ,78 1 .04 1 a 528 a TRI 2025 1 .493 b 1 .78 a 0 , 146 b 1 . 1 5 3 b 1 .45 a 0 .055 a 702 4 ,69 2 , 679 b 559 a TRI 2023 1 .704 b 1 .76 a 0 . 1 52 b 1 . 3 1 6 b 1 ,5 7 b 0 ,066 a 78 1 4 ,28 3 , 340 c 5 12 a Numbers within each cell followed by common letters indicate treatment means are not significantly different at p <0,05 according to DMRT 1 2 7 concentrations of mature leaves were 3 .6 - 4. 1% which is considered adequate for satisfactory growth of tea (3 . 5 - 4 .5% leafN, Bonheure and Willson, 1 992). 5.4.2 External and internal P utili sation efficiency of tea clones The external efficiency of P utilisation of the tea clones is in the order TRI 2023 > TRI 2025 > S 1 06 regardless of whether P fertilisers were applied or not (Table 5 . 1 ). In all three clones the fertiliser treatments had no effect on the external efficiency ofP . Higher external efficiency of P in TRl 2023 and TRI 2025 over S 1 06 i s due to greater root surface area and greater P uptake per unit surface area in the two TRI clones (Table 5 . 1 ; Figure 5 . 1 ), The higher external efficiency of P in TRI 2023 compared to TRI 2025 is however due to a greater root surface area in TRI 2023 and not due to a higher P uptake per unit surface area as observed in white clover varieties (Trolove et al. , 1 996a). These results suggests that TRI 2023 and TRI 2025 have a higher genetic potential to increase root growth, which can explore more soil volume to utilise P. These two clones have higher ability to absorb P per unit root surface area than S 1 06 probably because of greater acidification of the rhizosphere (Figure 5 .2), root exudation and/or mycorrhizal fungus association. It has been observed that tea roots have vesicular-arbuscular-mycorrhiza (V AM) relationships (Barthakur et al. , 1 987; Zhi, 1 993), but there is no information available on the extent of this association in different tea clones. Further research in this area would help to understand the differences in P utilisation efficiencies of tea clones. In general, internal P utilisation efficiency was similar among the tea clones (Table 5 . 1 ) . In all three clones however the internal P efficiency was significantly higher in the absence of any P fertiliser treatments (control treatment) than when P fertilisers were applied as observed for tea clone TRI 2025, bean, Guinea grass and calliandra in Chapter 4. In this study none of the clones had the combination of both high internal and external P use efficiencies. A combination of both these traits is expected to further benefit the plant . In essence, an ideal clone should be externally efficient to extract more P from -� ..... t:: ro 0... bfJ :.. '--' (\) � ro ..... 0.. ;:::J 0... ..... t:: ...::1 0... (a) 3500 3000 2500 2000 1 500 1 000 500 3 • • • • Y = 1 045 .5 X - 2389.45 R2 :::: 0.52 • 4 5 6 P uptake per unit surface area ( J..lg cnr2) (b) 3500 3000 2500 2000 1 500 1 000 500 200 • • • • S 106 • • TRI 2025 A TRI 2023 Y = 4 .98946 X - 244 .607 R2= 0.95 400 600 800 Total root surface area (cm2) Figure 5.1 Effect of (a) P uptake per unit surface area and (b) total root surface area on P uptake by tea clones (a) 4.8 4 .7 4.6 4 .5 4 .4 4 .3 4.2 4 . 1 a 1 2 N N 3 4 N 5 Distance from rhizopl ane (mm) 4 .8 4 .7 4 .6 4 .5 4 .4 4 .3 4 .2 4. 1 (b) a 1 2 3 N N 4 5 Distance from rhizoplane (mm) (c) 4.8 4 .7 4 .6 4 .5 4 .4 4 .3 4 .2 4 . 1 a I I I I 1 2 N N N -+- S I06 __ TRI 2Q25 --...- TRl l023 -a- Follow 3 4 5 Distance from rhizoplane (mm) Figure 5.2 Effect of tea clones on rhizosphere soi l pH in (a) Control (b) EPR and (c) TSP treatments . Vertical bars correspond to Lsd at p <0.05 and N represents treatments are not statistical ly significant at p <0 .05 . 1 3 0 the soil (P acquisition) and internally efficient to produce more dry matter from the absorbed P (P use efficiency). Therefore in the event of tea clonal screening, plant breeders should try to combine these two characters to obtain maximum yield with minimum P input. 5.4.3 Effect of P fertilisers and tea clones on soil pH Soil pH decreased in the rhizosphere of all tea clones (Figure 5 .2), but there was no change in soil pH more than 3 mm away from the rhizoplane, nor in the soil without tea plants (fallow). The pH changes in the rhizosphere were due to the influence of plant root activity. The decrease of soil pH in the rhizosphere varied with the tea clone. TRI 2023 had the highest pH decrease (0.25 - 0 .30 units) whereas S 1 06 showed the lowest pH decrease (0. 1 2 - 0 .20 units). TRI 2025 decreased rhizosphere pH by 0 . 1 9 - 0 .29 units. Greater shoot and root mass was associated with lower rhizosphere pH. The acidity released into the rhizosphere spread to a distance of about 3 mm from the rhizoplane for all the clones. The differences between clones in the release of acidity into the rhizosphere may be linked to the maintenance of electroneutrality in plant tissues after the excessive uptake of cations over anions (C­ A) (Barber, 1 995; Haynes, 1 990). Acidification of the rhizosphere has sometimes been attributed to exudation of low molecular weight organic acids (Roffiand, 1 992; Hoffiand et al. , 1 989; Liu et al . , 1 990). Tea plants are reported to secrete significant quantities of malic and citric acids from their roots (Jayman and Sivasubramaniam, 1 975; Xiaoping, 1 994) . The pKl values of citric and malic acids are 3 . 1 4 and 3 .4 respectively (Weast, 1 97 1 ), which are much lower than the cytoplasmic pH (6 - 7). Therefore these acids must leave the root cell cytoplasm in dissociated salt forms along with W to maintain electro neutrality within the cells. Dissolution of phosphate rocks is known to increase soil pH (Loganathan et al. , 1 995), but in our experiment the EPR treatment had no significant influence on soil pH (Figure 5 .2). The dissolution ofEPR in the rhizosphere (0-3 mm) ranged from 1 7 - 26% P and in the bulk soil (3 - 5 mm) it ranged from 1 2 - 1 4% P (Figure 5 . 3) . The extent of EPR dissolution in the bulk soil was however much lower than North 1 3 1 30 • S 106 25 • TRI 2025 ... TRI 2023 ,-.... � p... "-'" '"0 (j) 20 2: 0 {/) . !!? '"0 p... 1 5 o 1 2 3 4 5 Distance from rhizoplane (mm) Figure 5.3 Effect of clonal differences on EPR dissolution in the rhizosphere. Vertical bars represent standard errors of the means . 1 3 2 Carolina phosphate rock (NCPR) dissolution (30 - 32%) observed in an earlier trial, where a significant pH increase was observed in bulk soil for the NCPR treatment compared to control (Chapter 3) . Another reason for the absence of any noticeable pH rise in EPR treatments is that this PR has no accessory carbonate minerals associated with it (Dahanayake et al., 1 995) unlike NCPR and many other reactive phosphate rocks (Syers et aI. , 1 986). The higher dissolution of EPR in the rhizosphere for all clones is because of lower pH in the rhizosphere compared to the bulk soil. TRI 2023 and TRI 2025 produced significantly more EPR dissolution than S 1 06 (Figure 5 .3). The amount of EPR dissolution near the rhizoplane was in the range of 23 - 26% in TRI 2023 and TRI 2025 whereas in S 1 06 it was 1 8%. This is due to the secretion of more protons into the rhizosphere by TRI 2023 and TRI 2025 than S 1 06, as observed from the pH differences in the rhizosphere of these clones. The amount of tr released into the inner rhizosphere - an area very close to the rhizoplane (0 - 0 .5 mm zone) was calculated by adding both tr released due to p H decrease i n the rhizosphere compared t o the bulk soil and the amounts oftr that were consumed for the dissolution of EPR in that zone (see Table 5 .2 foot notes for details of this calculation) . The amount of Ii consumed in dissolving EPR in the rhizosphere and the bulk soil was determined as described in Chapter 4. , section 4 .4 . 3 . The calculations showed that less than 1 2% of the acid released was used up in dissolving EPR (Table 5 .2). This indicates that the presence of tr in the vicinity of EPR particles is not the sole factor causing dissolution of EPR. The removal of dissolved products of EPR (p, Ca and F) from the reaction site by plant uptake, soil adsorption and leaching are also key factors controlling EPR dissolution in the rhizosphere. 5.4.4 Effect of P fertilisers on soil P fractions The P fractionation of bulk soil with no P fertiliser additions showed that the soil u sed in the study had much higher NaOH-Pj, NaOH-Po and residual-P than resin-P and H2S04-Pi (Table 5 .3) indicating that the labile and Ca bound P are low in these soils Table 5.2 The comparison of observed acidity release in the rhizosphere of tea clones in EPR treated soil and the predicted acid release based on EPR dissolution Clone pH decrease I-r release within EPR H+ consumption EPR W consumption near 0-0 .5 nun of the dissolution for EPR dissolution i n for EPR rhizoplane rhizoplane2 within 0-0. 5 dissolution within the bulk dissolution (0-0 .5 mm) (Mmol W g-] soil) mm of the 0-0 .5 rum of the (3-5 mm) soil in bulk soie compared rhizoplane rhizoplane3 (P%) (3-5 nun) to bulk soi l l soil (�mol H+ g-] soil) (�mol Ht g-J soil) (P%) S 1 06 0 . 1 7 ± 0 .05 4.95 ± 1 . 62 1 7. 67 ± 1 . 86 2 . 28 ± 0. 1 1 1 3 .60 ± 1 .9 1 1 .75 ± 0. 1 2 TRI 2025 0 . 1 9 ± 0 .03 7 .98 ± 1 . 1 6 23 . 2 1 ± 1 .94 2 .99 ± 0. 1 2 1 4. 06 ± 1 .46 1 . 8 1 ± 0.06 TRI 2023 0 .30 ± 0 .02 1 2 .70 ± 0 .76 25 .93 ± 1 .90 3 .34 ± 0 . 1 2 1 4.27 ± 1 . 3 7 1 .84 ± 0 .05 ] Difference in pH (H20) of 0-0 . 5 mm soil slice and the weighted mean pH of all s lices within 3 -5 mm from the rhizoplane. 2 Change of pH in the rhizosphere compared to bulk soil * soi l pH buffering capacity (30 �mol H+ g- ] pRJ soil) . Difference in the amount of W used for d issolution of EPR in bulk and the rhizoplane (0-0 . 5 mm) soils4 (�mol Ht g - J soil) 0 . 5 3 ± 0 .04 1 . 1 8 ± 0. 1 7 1 .50 ± 0 .08 3Amount of EPR dissolved (�lg EPR g- J soil) * amount ofH+ required to dissolve 1 �g EPR (0 .00933 �mol Ht �g-I EPR - see text) . 4 H+ consumption for EPR dissolution in 0-0 .5 mm zone of the rhizosphere minus W consumption for EPR dissolution in the bu lk soil (3-5 mm). Total W production i n the (0-0.5 mm) rhizoplane5 (�mol Ht g- ] soil) 5 .48 ± 1 .62 9. 1 6 ± 0 .99 14 .20 ± 0 _77 5 Ht released in the 0-0 .5 mm from the rhizoplane plus the difference in the amount ofW used for dissolving EPR in the bu lk and the rhizosphere soils. Table 5.3 P fractions in the control soil (no P fertiliser added) and % recoveryl of added P from EPR and TSP treated bulk soil (3 - 5 mm) after 5 6 days of plant growth P-fraction Control S 1 06 TRI 2025 TRI 2023 bulk soil EPR TSP EPR TSP EPR (�g g O! soil) (P%) (P%) (P%) Resin-Pi 1 3 7 1 0 6 9 7 NaOH-Pi 1 1 3 8 66 8 69 1 4 NaOH-Po 6 1 6 1 0 6 1 0 6 H2SO4-Pi 28 78 9 77 9 7 1 Residual-P 72 2 4 1 2 2 Total-P 287 1 0 1 99 98 99 1 00 Ip% recovery = (P fraction in fertilised soil - P fraction in control soil) * 1 00 amount fertiliser P added to the soil TSP 9 73 7 8 3 1 00 1 3 4 1 3 5 (Golden et al . , 1 9 8 1 ). The small amounts of labile P in these soils is due to high P fixation (Table 4 . 1 , Chapter 4) and the low concentrations of Ca in the soiL The recovery of fertiliser P in the bulk soil was 98 - 1 0 1 % with 66 - 73 % of the P from TSP fertiliser converted into NaOH-P; and 7 1 - 78% of the P from EPR fertiliser remaining in the H2S04-Pi fraction. Most of the applied P fertilisers were recovered in NaOH-Pj and H2S04-Pi fractions, which were traditionally considered as pools ofP that are not easily available for plant uptake. Triple superphosphate had increased the readily available and weakly sorbed fractions of P (resin-P). These fractions were short-lived in soil, because of rapid chemical reactions with Fe and AI oxides and hydroxyoxides transformed this dissolved P into NaOH-Pj (Fe-P and AI-P) (Golden et al. , 1 99 1 ; Hedley et al. 1 994; Perrott, 1 995) . The recovery of large amounts of H2S04-Pi (Ca-P) in soils treated with sparingly soluble EPR fertiliser was as expected because the P in the apatite mineral in EPR is mainly bound to Ca. 5.4.5 Effect of tea clones on soil P fractions There was a significant reduction of resin-P, NaOH-Pj, and H2S04-Pi up to 3 mm from the rhizoplane in all the tea clones compared to the bulk soil (Figures 5 . 4, 5 . 5 and 5 . 6) as observed for tea clone TRl 2025 and other crops in Chapter 3 and 4 . The reduction of resin-P in the rhizosphere was significantly greater (p <0.05) in TRl 2023 and TRl 2025 compared with S 1 06 and the reduction ofNaOH-P; in the rhizosphere was significantly greater (p <0.05) in TRI 2023 compared to the other two clones. The depletion of these P fractions is due to plant uptake and microbial immobilisation of Pj . Plant P uptake and immobilisation of Pi into NaOH-Po are the major causes of the depletion of resin-P in the rhizosphere soil of all tea clones, rather than P fixation by Fe and AI, because NaOH-P; fraction which is a measure of Fe-P and AI-P also decreased. The significant differences in the H2S04-Pi among the tea clones near the rhizoplane in EPR treated soil was due to the differences in the amount of EPR dissolution caused by the differences in rhizosphere pH (Figure 5 .2). (a) 50 40 30 20 o I I I I 2 N N N 3 4 5 Distance from rhizoplane (mm) (b) 50 40 30 20 1 0 o I I I I 1 2 N N N 3 4 5 Distance from rhizoplane (mm) (c) 50 40 30 20 1 0 o l I N N N N N -+- S 106 -e- TRI 2025 --:*- TRl 2023 --B- Fallow 1 2 3 4 5 Distance frm rhizoplane (mm) Figure 5.4 Effect of tea clones on resin-P in soil with (a) Control (b) EPR and (c) TSP treatments . Vertical bars correspond to Lsd at p <0. 05 and N represents treaments not statistically significant at p <0.05 . 'Cf) cf) ::i. �- I ::c o Cd Z 300 240 1 80 1 20 60 (a) I I I N o 1 2 N N • 3 4 N 5 Distance from rhizoplane (mm) (b) 300 240 1 80 1 20 60 o I I I N 2 N N N i I 3 4 5 Distance from rhizoplane (mm) (c) 300 240 1 80 1 20 60 o N N N N N N • -+- S 106 N • -e- TRI 2025 -II- TID 2023 -B- Fallow 1 2 3 4 5 Distance from rhizoplane (mm) Figu re 5.5 Effect of tea clones on NaOH-P i in soil with (a) control (b) EPR and (c) TSP treatments . Vertical bars correspond to Lsd at p <0.05 and N represents treatments not statistical ly significant at p <0.05 . 200 1 50 1 00 50 (a) N N N N o 1 2 N N • • 3 4 N • 5 Distance from rhizoplane (mm) (b) (c) 200 I I N N N N N 200 N N N N � -� � � 1 50 � 1 50 1 00 1 00 50 50 o 1 2 3 4 5 o 1 2 N N N -+- S 106 -e- TRI 2025 -lIk-- TRI 2023 -B- Fallow i i 3 4 5 Distance from rhizoplane (mm) Distance from rhizoplane (mm) Figu re 5.6 Effect of tea clones on H2SO 4-Pj in soil with (a) control (b) EPR (c) TSP treatments. Vertical bars correspond to Lsd at p TRl 2025 > S 1 06. The differences in P depletions are in line with the differences in surface area of the roots in the boundary zone (0 - 2 mm above the mesh) (Table 5 .4) and growth rates (Table 5 . 1 ) . 5.5 CONCLUSIONS TRl 2023 and TRl 2025 had significantly higher dry matter production than S 1 06. The external P efficiency of TRl 2023 and TRl 2025 was higher than S 1 06, mainly due to greater root surface area and greater P uptake per unit surface area. TRl 2023 had higher external efficiency than TRl 2025 as a result of higher root surface area, but not due to higher P uptake per unit surface area. The higher P uptake per unit surface area in TRl 2023 and TRl 2025 clones may be due to their higher root acidification, root exudation of organic compounds and/or mycorrhizal association and these aspects needs further investigation. In all clones, P fertilisers had no influence on external P efficiency, but internal P efficiency was highest in soils receiving no P fertiliser. All three tea clones induced acidification of the rhizosphere. This caused increased dissolution of EPR in the rhizosphere compared to that in the bulk soil . Approximately 1 8 to 26% of EPR dissolved in 56 days of tea growth. This suggests that though EPR is considered a less reactive PR according to its citric acid solubility, in the vicinity of tea roots appreciable amounts of this PR will dissolve to supply P to the plants. The rhizosphere acidification and EPR dissolution were in the order TRl 2023 > TRl 2025 > S 1 06. All clones depleted resin-P and NaOH-Pi, but increased NaOH-P 0 in the rhizosphere. The rate of depletion of resin-P and NaOH-Pi was in the order of TRl 2023 > TRl 2025 > S 1 06 . The accumulation of NaOH-P 0 in the rhizosphere is probably due to transformation of Pi into Po by the enhanced microbial activity in this zone. The above findings could be utilised in future tea breeding programmes to develop new tea clones having both higher external and internal P efficiencies combined with 1 4 3 other desirable characteristics. With the rising cost of P fertilisers, the potential of using P efficient tea clones is an attractive alternative for sustainable tea production. The results demonstrated that short-term rhizosphere studies can be used to obtain quick information on the screening of plant varieties/clones for their P utilisation efficiencies. 1 4 4 CHAPTER 6 EFFECT OF FORMS OF NITROGEN SUPPLY ON MOBILISATION OF PHOSPHORUS FROM EPPA WALA PHOSPHATE ROCK AND ACIDIFICATION IN THE RHIZOSPHERE OF TEA (Camellia sinensis L.)l 6. 1 INTRODUCTION In Chapters 3 , 4 and 5 it has been shown that soil pH in the rhizosphere differs markedly from that of the bulk soil for tea and many other crops and this causes differences in PR dissolution in the two zones. Generally, rhizosphere pH decreases when � + forms of N fertiliser are used and increases when N03- forms are used, because of the release of H+ and Off or HCO:; respectively to the soi l to maintain electroneutrality within the plant cell (Gahoonia et al . , 1 992; Kirkby and Mengel, 1 967; Youssef and Chino, 1 988). In ryegrass seedlings, rhizosphere pH decreased by 1 . 6 units as a result of N1L+ supply and it increased by 0 .6 units with N03- supply (Gahoonia et al. , 1 992). The forms of nitrogen (N) supply also exert a strong influence on the plant availability of P in the rhizosphere soil through their influence on pH, which determines the rate of dissolution of PR and P fixation by soil colloids (Barrow, 1 984). In Chapter 5, it was shown that rhizosphere pH decreased by 0 .2 - 0 .3 units compared to that of the bulk soil for three clones of tea. which were fertilised with urea. This was believed to be due to plant uptake of� - formed by the ammonification of urea and the root exudation of organic anions. The uptake of N1L + and excretion of organic anions cause the release of H+ to maintain electro neutrality within root cells. It i s not clear however, if tea roots will continue to depress rhizosphere pH if N03 - or N03- + NlI/ forms ofN are applied instead ofNH/ forms . Tea plants may prefer to 1 Zoysa A K N, Loganathan P and Hedley M J 1998 Effect of forms of nitrogen supply on mobilisation of phosphorus from a phosphate rock and acidification in the rhizosphere of tea (Camellia sinensis L.) Australian Journal of Soil Research (in print). 1 4 5 use predominantly � + -N (Xan and Jianyun, 1 994) even in the presence ofN03--N. Trials have been conducted in many countries to investigate the use of calcium ammonium nitrate (CAN), a N fertiliser which creates less soil acidit)) because the continuous use of �hS04 for tea leads to increased soil acidity (Bonheure and Wil lson, 1 992; Sandanam et al . , 1 980; Watson and Wettasinghe, 1 972). All these studies however revealed that CAN produced lower tea yields than �hS04. The reduction in yield could have been caused by the lower availability of P from PR, and the lower availability of Mn at the higher pH of CAN fertilised soil (HarIer, 1 968). Tea is a calcifuge, preferring low pH and low Ca saturated soils and this may explain the lower tea yield with CAN (Watson and Wettasinghe, 1 972). These studies however did not give any information on the effect ofN forms on soil P fractions and pH either in the bulk sail or in the rhizosphere of tea. 6.2 OBJECTIVES The study reported m this chapter was designed to investigate the following objectives : 1 . To investigate the effect of �+-N and N03--N separately [�hS04, Ca(N03)2] and in combination �N03] on the pH in the rhizosphere of tea seedlings. 2. To study the effect of the form ofN supply on the mobilisation ofP from EPR in tea rhizosphere. 3 . To investigate which form ofN is preferred by tea plants. 6.3 MATERIALS AND METHODS The physico-chemical characteristics of the soil (Rhodustult) used in this study were presented in Chapter 4 - Table 4 . 1 . The soil was air-dried and passed through a 2 mm sieve and mixed with Eppawala phosphate rock (EPR - for specification details see section 4.3) and KCl at the rate of 200 !lg P or K g'! soil. Three sources of N 1 4 6 treatment and a control (no N fertiliser) treatment were used. The N sources were �)2S04 ( 1 00% NH/-N), Ca(N03)2 ( 1 00% N03--N) and �N03 (50% �+-N and 50% N03--N). The N fertilisers were mixed separately with the soil at the rate of 200 /lg N g-! soil . All soils were mixed with a nitrification inhibitor dicyandiamide (DCD) at the rate of 45 /lg g-l soil at planting to reduce as much as the possible transformation of� + to N03-. The modified root study container (RSC) technique described in Chapter 3 was used for this trial. The upper and lower compartments were packed with 1 35 g soil (bulk density 1 . 1 Mg m-3) and 242 g soil (bulk density 1 . 1 Mg m-3), respectively. A single vegetatively propagated four-month-old tea plant (clone TRI 2025) was planted in each of the upper compartments . The treatments (three N forms and control) were replicated four times and arranged in a randomised complete block design in a glasshouse maintained at 1 2° C minimum and 26° C maximum temperatures at St. Coombs, Sri Lanka. Four replicated pots without plants but having the same treatments (three N forms and control) were also included to study the changes in the soils due to fertilisers in the absence of tea plants. The RSCs were placed on top of a sand bed connected to a water reservoir and the water table was fixed at 1 60 mm below the base of the RSCs as in the trials in earlier chapters. This enabled the RSCs to be maintained at a constant water potential of approximately - 1 .6 kPa. The trial set-up in the glasshouse is shown in Figure 6 . 1 6.3. 1 Soil, plant and root sampling Sixty days after transplanting plants to RSCs, the shoots were cut 5 mm above the soil surface and dried at 60° C, weighed and ground to powder. The soil in the lower compartment was sliced into thin sections with a piston microtome, dried and ground as described in Chapters 4 . Root sampling and measurements of root mass, length, volume and surface area were also as described in Chapter 4. 1 4 7 Figure 6. 1 Plant growth system in the glasshouse 6.3.2 Soil and plant analysis 1 4 8 Shoot and root samples were analysed for total N by KjeldhaJ digestion (Jackson, 1 958) . The plant materials were dry-ashed at 5500 C and the ash was taken up in 0 .05 M HCI solution and analysed for K and Na by flame emission spectrophotometry, Ca and Mg by atomic absorption spectrophotometry, P by the vanadomolybdate method (Jackson, 1 958) and AI by the aluminon method (Jayman and Sivasubramaniam, 1 974). Plant materials were extracted with hot water and the concentration of CI measured with a Cl electrode (Adriano and Doner, 1 982) and S04 by turbidometry (Buttlers and Chenery, 1 959) . The initial nutrient composition and dry weights of the plants were determined on a sample of 1 0 randomly selected plants of similar size and foliar characteristics to the test plants at the beginning of the experiment. Soil pH, amount of EPR dissolution and soil P fractionation were carried out as described in Chapter 3 . 6.4 RESULTS AND DISCUSSION 6.4.1 Effect of N forms on N and P uptake by tea All forms of N significantly (p <0 .05) increased shoot dry matter yield and shoot : root dry matter ratio over the control treatment, but only �)2S04 and NIitN03 treatments significantly (p <0 .05) increased the root dry matter yield (Table 6 . 1 ) . However, root, shoot and total dry matter yield and shoot : root ratios were not significantly affected by the form of N fertiliser. The increase in dry matter yield obtained with all N forms compared to the control treatment is similar to the results obtained in field trials, where tea yields have always increased with N application because the N uptake rate of tea is higher than most other crops (Eden, 1 976; Tolhurst, 1 968) . The increase in total dry matter yield is due both to increases in root and shoot growth, although the shoot growth response to fertilisers is higher than the root growth response. Adding (NH4)2S04 significantly (p <0. 05) increased N and P concentrations and uptake in both roots and shoots compared to the control treatment. But the Nli.N03 treatment only significantly (p <0.05) increased shoot and root N and Table 6. 1 Effect of N forms on plant shoot and root dry matterl , shoot : root ratio, tissue N, P concentrations and uptake by tea plants Shoot Root Total dry Shoot : Root Treatment Dry matter N N uptake P P uptake Dry matter N N uptake P P uptake matter yield dry matter (g porI) (%) (mg planrl ) (%) (rug planel ) (g pori) (%) (mg planrl ) (%) (mg planrl ) (g porI ) ratio Control 0 .93 1 .36 1 2 . 8 1 0 . 1 1 6 1 .08 1 .06 1 .03 1 0.96 0.046 0.47 1 . 99 0 . 88 (NH4)zS04 1 . 77 1 .5 1 26 .68 0 . 1 32 2 .33 1 .56 1 .45 22 .67 0.054 0 . 84 3 . 33 U3 NH4N03 1 .92 1 .43 27 .39 0 . 1 26 2 .42 1 .55 1 . 55 24.23 0.053 0 . 82 3 .47 1 .24 Ca(N03h 1 .69 1 .49 25 .26 0 . 1 1 4 1 .92 1 .32 1 .3 1 1 7. 3 3 0.040 0 .52 3 . 0 1 1 .2 8 Lsd (p <0.05) 0 .36 0 . 1 4 5 . 8 5 0 .0 1 0 0.47 0 .39 0 . 1 8 7 .09 0.008 0 .23 0 .59 0 .20 I The difference in the dry matter yield at harvest minus initial dry matter content 1 50 P uptake and root N concentration compared to the control treatment. The Ca(N03h treatment only significantly (p <0. 05) increased shoot N and P uptake and root N concentration. A significantly lower P concentration and P uptake in the Ca(N03)2 treatment compared with the �N03 and �)2S04 treatments may be due to the lower solubility of EPR (Figure 6 .2) at the higher pH encountered in the rhizosphere of the Ca(N03)2 treatment (Figure 6 .3a) . It could also be due to the competition of anions, N03" from the Ca(N03)2 supply and increased OH concentration resulting from the increased rhizosphere pH in this treatment. The increased pH would have increased the valancy of the phosphate ions (H2P04" to HPO/) which are taken by plants at a lower rate than the lower valancy ions (Tisdale, 1 985). Additionally the Ca in Ca(N03)2 may have also reduced N assimilation as tea is a calcifuge, which does not perform well in soils high in Ca. Higher Ca concentrations in the soil solution may also have reduced the phosphate rock dissolution (see next section). 6.4.2 Effect of N forms on pH and P fractions in bulk soil A significant reduction in pH of the �)2S04 treated bulk soil (3 -5 mm from mesh - outside the influence of the roots) in both planted and unplanted pots compared to the control treatment, indicated that some nitrification occurred despite using an inhibitor (Figure 6 .3) . Soil pH (Figure 6 . 3 ) was lower in �hS04, �N03 and control and the dissolution of EPR fertiliser was higher (Figure 6 .2) in the bulk soil (3 - 5 mm away from root surface - zone not influenced by roots) compared to the Ca(N03h treatment. This increase in EPR dissolution with the �hS04 treatment resulted in increased resin-P and NaOH-Pi in the bulk soil (Table 6 .2). The higher amounts of H2S04-Pi in the bulk soil of the Ca(N03h treatment compared to other treatments resulted from a lower EPR dissolution (8% P dissolution for Ca(N03)2 treatment compared to 1 6% P dissolution for �hS04 treatment) at the higher soil pH of this treatment (Figure 6 .3) . In addition to the higher pH, the Ca(N03h treatment supplied soluble Ca to the soil which may have also reduced EPR dissolution (Mackay et aI. , 1 986). 30 • Control 25 • Ca(N03), .... (NH4�SO. * NH4NO, � 20 p... "'0 V "'0 "'0 CI:l ...... 0 � 1 5 <::> '--' p... "'0 V :>- '0 CI) r/l Q 1 0 5 o 1 2 3 4 5 Distance from rhizoplane (mm) Figure 6.2 Effect of nitrogen forms on P dissolution from EPR. (Vertical bars represent standard errors of the means). 1 5 1 (a) 4.80 4 .80 4 . 70 I I I I I I 4.70 I 4.60 4 .60 4 .50 ---o ::rt 4 .50 '-../ ::r: 4.40 0. � 4.40 o C/J 4 .30 4 .20 o 4 .30 4 .20 4. 1 0 1 2 3 4 5 Distance from rhizoplane (mm) (b) I I I I I I -+- Control -*- (NH4 NO), -e- Ca(NOJh -A:- (NHhS04 o 1 2 3 4 Distance from mesh (mm) Figure 6.3 Effect of ni trogen ferti l i ser forms on soil pH (a) with plants and (b) without plants (Vertical bars represent Lsd for treatment means at p <0. 05) . I 5 1 5 3 Table 6.2 Phosphorus fractions in control soil (without EPR) and the % recovery of added EPR-P in bulk soils (3-5 nun) for various N treatments P fraction Unfertilised EPR Fertiliser recovery (P%) I bulk soil Control Ca(N03h (NH4hS04 (Ilg g-l soil) (No N fertiliser) Resin-P 12 2 2 5 NaOH-Pi 1 4 1 9 2 1 5 NaOH-Po 57 2 2 2 H2SO4-Pi 28 79 95 78 Residual-P 60 5 - 1 Total-P 298 97 1 0 1 1 0 1 1 � fraction in fertilised soil - P fraction in control soil) * 1 00 Fertilised P added to soil �N03 4 1 0 3 83 1 1 0 1 1 5 4 In the non-rhizosphere (bulk) soil (3 - 5 mm), between 97 - 1 0 1% ofP was recovered by the P fractionation from the applied EPR fertiliser for all treatments (Table 6.2) with NaOH-Pi and H2S04-Pi fractions ( 1 46 - 224 !lg g-! soil) being greater than resin­ P, NaOH-Po and residual-P fractions ( 1 7 - 64 Jlg g-l soil). This agrees with the observations made in the trials reported in Chapters 4 and 5 . 6.4.3 Effect of N forms on rhizosphere pH Adding � + fertiliser lowered rhizosphere soil pH, whereas N03- fertiliser increased rhizosphere soil pH (up to 3 mm from the rhizoplane), compared to the control treatment (Figure 6. 3a) . In the absence of plant roots pH changed little with distance from the mesh (Figure 6 .3b) . Compared to the bulk soil, the pH of rhizosphere soil was 0.29 units lower in the �hS04 treatment, 0 . 1 1 units in the �N03 and 0 .24 units in the control treatments . In contrast, adding Ca(N03)2 increased rhizosphere pH by 0 . 1 3 units compared to that in the bulk soil . These results show that the form ofN supply to tea roots exerts a strong influence on their rhizosphere pH. The effect of the form of N supply on rhizosphere pH is due to the difference in the uptake of � + and N03 - by plants; uptake of � + results in production of � and N03- results in excretion of OK and HC03- (Troelstra, 1 983 ; Troelstra et al., 1 985) . The �)2S04 fertiliser being solely an � + type produced the highest reduction in rhizosphere pH whereas �N03, which contains both � + and N03- produced the smallest pH drop . The increase in rhizosphere pH compared with bulk soil pH in the Ca(N03)2 treatment is consistent with excretion of Oli and HC03- to counter balance the N03- uptake. The pH increase (0. 1 3 pH units) however is about 50% of the pH decrease observed for the �hS04 treatment (0 .29 pH units), even though the same rate ofN was supplied to the plants. A possible reason for this is the rhizosphere acidification caused by the exudation of organic anions and associated protons by the roots. Tea plants secrete significant amounts of malic and citric acids from their roots (layman and Sivasubramaniam, 1975 ; Xiaoping, 1 994). The organic acids (or organic anions plus protons) released may have acidified the rhizosphere regardless of the 1 5 5 form of N supply. Furthermore greater amounts of organic acids may have been excreted in the presence ofN03- as observed for rape roots by Hoffiand et ai. ( 1989). The observed acid release by tea roots based on pH changes and the predicted acid production based on EPR dissolution for the different N treatments were calculated as described in Chapter 4 and presented in Table 6.3 . The greatest acidity release in the rhizosphere was for the �)2S04 treatment ( 1 23 /lmol Ii) and the smallest was for the �N03 treatment (30 /lmol H+). In contrast, the plants treated with Ca(N03)2 released 7 !lmol Off or HC03- into the rhizosphere. Despite the pH rise in the rhizosphere compared to the bulk soil in the Ca(N03)2 treatment, EPR dissolution in the rhizosphere was still higher than that in the bulk soil . This may be due to the removal of dissolved products ofEPR (P and Ca) by plant or microbial uptake, which would have increased the dissolution ofEPR in the tea rhizosphere. 6.4.4 Effect of plant roots on P fractions in the soil In the absence of plant roots there was no change in any form of P with distance from the mesh (Figure 6 .4b, 6 . Sb, 6 . 6b and 6 . 7b). Therefore any difference observed in the P-fractions between bulk soil and rhizosphere was assumed to be due to either P uptake by plant roots or increased microbial activity in the rhizosphere resulting from organic carbon exudation by the roots (Tinker, 1 980) . Higher microbial activity in the rhizosphere can immobilise some soluble P (Helal and Sauerbeck, 1 99 1 ), whereas organic carbon exudates can reduce the P fixation by Fe and AI oxides (Nagarajah et aI . , 1 968) . The profile of soil P depletion, which could be used as a measure of P mobilisation was markedly affected by the forms of N added. Resin-P depletion was greatest for the �(S04h treatment and the lowest in the Ca(N03h and control treatments (Figure 6 .4a). The greater resin-P depletion in the �)2S04 treatment results from higher P fixation by Fe and A1 oxides at the lower soil pH of the �hS04 treatment (Barrow, 1 984). In contrast less NaOH-Pj was depleted in the rhizosphere of the �hS04 treatment compared to the Ca(N03)2, �N03 and control treatments Table 6.3 The observed acid release by tea roots based on pH changes and the predicted acid production based on EPR dissolution in the lower compartment of RSC Treatment pH drop (-) Observed W Average of EPR W consumption EPR H+ consumption Difference in the or rise (+) in the production (+) or dissolution for EPR dissolution i n for EPR amount of W used rhizosphcrc consumption (-) within dissolution the bulk dissolution in for dissolution of (0-3 mm) within 0-3 mm of the within 0-3 mm (3-5 mm) soil bulk soil3 EPR in the bulk soil compared to 0-3 mm of the rhizosphere of the (P%) (3-5 mm) and the rhizosphere bulk soi l ] rhizosphcrc 2 (P%) rhizosphere 3 ' 1 4 (j1mo! Hl (0-3 mm) SOl S (�lmol H+) ( !1mol H+) ( !1mo! W) Control - 0.24 ± O.o I +42 . 5 3 ± 8 .45 1 5 . 2 ± 1 .2 1 3 3 .64 ± 1 . 1 3 1 2 . 23 ± 0 .0 1 22.34 ± 0. 1 5 1 1 . 30 ± 0.22 (NH4)2So'l - 0.29 ± 0 .0 1 + 1 07.89 ± 1 .0 1 1 8.40 ± 1 . 1 9 45 .36 ± 1 . 59 14 . 16 ± 0 .0 1 30.58 ± 0.03 1 4 .78 ± 0 .22 NH4N03 - O. l l ± 0.02 + 1 7.90 ± 1 .87 1 5.26 ± 0 .78 3 7.90 ± 1 .3 0 1 2 . 3 2 ± 0 .0 1 25.47 ± 0.20 1 2.43 ± 0 . 3 5 Ca(N03h + 0 . 1 3 ± 0.03 - 1 5 .03 ± 0.62 9 . 3 5 ± 0 .43 22 .94 ± 0.84 7.42 ± 0 .05 1 5.33 ± 0 .09 7 . 6 1 ± 0. 1 1 I Difference in pH (H20) between 0-3 mm soil slice and the weighted mean pH of all slices within 3-5 mm of the rhizoplane. Total Hi" producti on (+) or consumption (-) in (0-3 mm) of the rhizosphere 5 (j1mol Hi") + 5 3 . 83 ± 0.52 + 1 22.67 ± 0 .42 + 30. 3 3 ± 0 .37 - 7 .42 ± 0 .2 1 2 Sum total of (change of pH in the rhizosphere soil slice compared to the bulk soil {3-5 nun} * soil pH buffering capacity {30 flmol H+ g. l pHI soi l } * weight of that soil sl ice) for all sl ices in the rhizosphere (0-3 mm) . 3 Amount of EPR dissolved (flg EPR g.l soil) * amount of Hi" required to dissolve 1 flg EPR (0 .00933 flmol H+ flg·l EPR - see text) * soil weight. 4 H' consumption for EPR dissolution in 0-3 nun zone of the rhizosphere minus W consumption for EPR dissolution in the bulk soil (3-5 mm). 5 H' released in the 0-3 mm of the rhizosphere plus the difference in the amount of W used for dissolving EPR in the bulk and the rhizosphere soils. (a) 30 I I I I I I 25 -- � 20 o r:/l 'bI) � 1 5 0... I . S � 1 0 � 5 o 1 2 3 4 Distance from rhizoplane (mm) (b) 30 I 25 20 1 5 1 0 5 5 o I I I I 1 2 I 3 I -+- Control -e- Ca(NOJh ---.Ik- (NH.>zso, -*- NH.NO J 4 Distance from mesh (mm) Figure 6.4 Effect of ni trogen forms on resin-P (a) with and (b) without plants. Vertical bars represent Lsd for treatment means at p <0.05 . I 5 (a) (b) 200 j 200 I [ I I I I I r I I I I I � : : =:==: 1 75 1 75 � - ' 0 r/l +---t .---+ • .. + On /)I) :::t 1 50 1 50 --0.:,-, ::c: 0 til Z -+- Control 1 25 1 25 -e- Ca(NO�2 -.tr (NH'>2S0, -*- NU.NOJ 1 00 1 00 0 1 2 3 4 5 0 1 2 3 4 5 Distance from rhizoplane (mm) Distance from mesh (mm) Figure 6.5 Effect of nitrgen forms on NaOH-Pj in soil (a) with and (b) without plants. Vertical bars represent Lsd for treatment means at p <0.05. f-' Vi OJ (a) (b) 275 275 250 I I I I I I I 250 1 I 1 1 I I � ...- 225 . -0 225 . . . -- • • VJ I 0 ; 1)1) ; 1)1) 200 ::j., "-' 0,.;-I .,. 0 C/) 1 75 N X � * * 200 � .. .. 1 75 -+- Control -e- Ca(N°3)2 1 50 1 50 -k- (NH'>lso. -*- NH4NOJ o 1 2 3 4 5 o 1 2 3 4 Distance from rhizoplane (mm) Distance from mesh (mm) Figure 6.6 Effect of ni trogen forms on H2S04-Pi (a) with and (b) without p lants . Vertical bars represent Lsd for treatment means at p <0.05 . 1 • * A 5 (a) 90 80 70 60 o N N N N N N N 1 2 3 4 5 Distance from rhizoplane (mm) (b) 90 80 70 60 o N N N N N N N --+- Control -e- Ca(NOJ)2 -.- (NH4l2SO, -*- NH4NOJ 1 2 3 4 5 Distance from mesh (mm) Figure 6.7 Effect of nitrogen forms on NaOH-P in soil (a) with and (b) without plants. The N shows that o treatments are not stati stical ly significantly different at p <0.05 . 1 6 1 (Figure 6 .5a and Table 6 .4) . This cannot be explained by differences in plant P uptake as plant P uptake in the (NRt)2S04 and N1:LtN03 treatments were similar (Table 6 . 1 ) . Excretion of Off and/or HC03- from roots due to N03- uptake in the Ca(N03h treatment may have released phosphate ions adsorbed to Fe and Al oxides in the soil by ligand exchange with the Off and HC03- (Gahoonia et al. , 1 992). Also the higher amounts of organic acids excreted in the presence of N03 - compared to the N1:Lt + source of N (Homand et aI . , 1 989) may have dissolved more fixed P in the soil because of the organic anions complexing with Fe and Al (Earl et aI . , 1 979� Nagarajah et al. , 1 968) thereby increasing resin-P and reducing NaOH-Pi. The H2S04-Pi depletion was lowest in the Ca(N03)2 treatment (23% of total-P depletion) compared with all other treatments with (NRt)2S04 having the highest depletion (34% of total depletion) (Figure 6.6; Table 6.4). The higher rhizosphere pH and Ca input in the Ca(N03)2 treatment may have reduced the dissolution of EPR in the soil thus lowering H2S04-Pi depletion. An accumulation of NaOH-P 0 was observed within 2 mm from the rhizoplane for all treatments (Figure 6. 7a) . Higher concentrations of NaOH-P 0 in the rhizosphere compared to the bulk soil may be due to the transformation of labile Pi into Po by microbial utilisation of Pi as discussed in Chapter 4. However there was no difference in NaOH-Po concentration between N sources suggesting that the source ofN or the pH changes due to these sources had no significant effect on the utilisation of Pi by microorganisms in the rhizosphere. An attempt was made to predict rhizosphere P depletion using plant P uptake. Plant P uptake, observed total P depletion and predicted total P depletion in the rhizosphere were estimated for each treatment as explained in Chapters 4 and 5 . The predicted P depletions in the rhizosphere were lower than the observed P depletion by 1 0 - 26% (Table 6.4) . In a similar experiment, but with different tea clones fertilised with urea, the predicted P depletion varied between - 1 5 and 23% of the observed P depletion (Chapter 5) . The reasons for the differences between observed and predicted P depletions were given in Chapter 4. 1 6 2 Table 6.4 Comparison of observed P depletion in the soil in the lower compartment ofRSC with predicted P depletion calculated from plant P uptake 1 Control Observed depletion in the lower RSC (Ilg RSC1) Resin-P 283 NaOH-Pi 503 H2SO4-Pi 377 Total 1 1 63 Observed accumulation in the lower RSC (Ilg RSC1) NaOH-Po 200 Observed net depletion in the lower RSC(llg RSC1) 963 Surface area of boundary (0-2 mm above mesh) roots (cm2) 236 Plant P uptake (fAg plant-I ) 1 524 Total root surface area (cm2) 25 1 Plant P uptake/root surface area (Ilg cm-2) 6. 1 Predicted Total P depletion in the lower RSCI (fAg RSC -1) 7 1 6 Deviation of predicted P depletion from observed (%) -26 Plant P uptake * boundary root surface area * 0 . 5 Total root surface area Treatments (NI-L!hS04 N"ILN03 Ca(N03)2 525 3 76 327 587 755 790 580 442 333 1 692 1 573 1450 262 303 222 1 430 1 270 1 228 290 306 266 3 1 73 3247 2450 3 86 435 3 58 8 . 2 7 . 5 6 .8 1 1 92 1 1 42 9 1 2 - 1 7 - 1 0 -26 [The factor 0 . 5 is used because only half the root surface area was assumed to cause depletion in the lower RSC] 6.4.5 Nutrient uptake and electroneutrality in plant tissues 1 6 3 Plants generally absorb different amounts of cations and anions. The excess positive charge created by greater uptake of cations (NRt +, Na +, K+, Ca2+, Mg2+ : total C) than inorganic anions (H2P04-, N03-, cr, sot : total A) in the plant tissue (C-A) is balanced by formation of organic anions (e.g. carboxylate) in the plant (Troelstra, 1 983) . Tea plants absorb large quantities of AI from acid soils (Sivasubramaniam and Talibudeen, 1 97 1 ), but because the form and the charge of the absorbed AI ions are not known it was not included in the (C-A) calculation as many others did in their work (Gijsman, 1 990b; Troelstra, 1 983) . The fractions of N taken up as N03- and NH/ by the plants were approximately estimated by matching the proton release into the rhizosphere soil with (C-A) calculated for different proportions of � + and N03 - taken up by the plants (Table 6 .5) . Such matching showed that in the (NRt)2S04 treatment, the proportions of N taken up as �+ and N03- were probably around 29% and 7 1% respectively whereas in the Ca(N03)2 treatment it was 23% and 77%. The proportion of N03-taken up by the plants was also estimated by usmg the following formula (Troelstra et al . , 1 985) : X= (C-A) + 0.946 NQrg - H+ efflux 2 where all parameters are in units of !leq per plant. X is the concentration of organic N (Norg) contributed by N03- taken-up by the plant (the balance contribution to Norg is from NH/ taken-up by the plant), C-A is the cation - anion concentration difference in plants, H+ efflux is the acidity released by the roots into the rhizosphere soil . The plant uptake ratios of � + : N03- was calculated from the X values estimated using the above formula (Table 6 . 5) These ratios agreed very well with the corresponding ratios calculated in the preceding paragraph by matching NH4 + and N03- uptake with charge balance in the plants. It should be noted however that these calculations rely on the accuracy with which the rhizosphere pH change and the soil pH buffering 1 6 4 Table 6.5 Effect of different N forms and assumed ratios ofNI-L + : N03- uptake on the net release ofIr to the rhizosphere (0-3 mm) by tea roots Assumed ratios ofNI-L + : N03- (Cations - Anions) uptake in the plantl ()leq pJanrl) taken-up by plant Control (NIL)2S04 NI-LN03 Ca(N03)2 0 : 1 00 -606 -2398 -2592 - 1 867 20 : 80 43 -569 -620 -3 53 22 : 78 1 08 -386 -423 -20 1 25 : 75 205 - 1 1 2 - 1 27 26 28 : 72 303 1 62 1 68 253 30 : 70 368 345 365 404 32 : 68 433 528 563 556 45 : 55 855 1 677 1 844 1 54 1 47 : 53 9 1 9 1 860 204 1 1 69 1 50 : 50 1 0 1 7 2 1 74 2337 1 9 1 9 52 : 48 1 082 23 1 7 2543 207 1 5 5 : 45 1 433 2591 2829 2297 1 00 : 0 264 1 6745 7265 5704 Total Ir release (+) or OIT release (-) in the rhizosphere of all 1 1 4 326 86 -20 roots2 ( )leq Ir planrl) Predicted ratio of (NIL : N03) uptake by the j:llant 23 : 77 29 : 7 1 26 : 74 23 : 77 Calculated (NIL : N03) uptake using the imperical formula3 24 : 76 3 3 : 67 29 : 7 1 27 : 73 1 Lcharge of total cation uptake ()leq) by plant - Lcharge of total anions uptake ()leq) by plant (NIL + and N03 - calculated from Norg in plant according to the ratios in column 1 ) . 2 9th column in Table 6.3 * Total root surface area Surface area of roots above mesh (0-2 mm) * 0 . 5 (The factor 0. 5 is used because the acidity produced in the lower RSC is assumed to be due to half of the surface area of the roots on of the roots on the mesh). 3 X = (C-A) + 0 .946 NQI£.. - H+ efflux 2 where X represents the contribution ofN03- to Organic-N (Troelstra et aI . , 1 98 5). See text for units and explanation of the equation. 1 6 5 capacity (pHbc) are determined. Changes in pH with ionic strength and pHbc with time may affect this calculation. The high proportion of N03 - compared to � + taken up in the �hS04 treatment suggests that a large proportion of � + in the �hS04 treatment may have been nitrified inspite of the nitrification inhibitor used in this study. Tolhurst ( 1 955) reported significant nitrification in Sri Lankan tea soils even at a pH as low as 3 .7 . Sandanam et al . ( 1 978) reported a recovery of 39% of fertiliser N as N03- from acid tea soils treated with �hS04 fertiliser within 27 days of incubation at 20 - 22° C at 3 5% (w/w) moisture content. Therefore it is quite possible that a significant proportion of the � + added in the present trial would also have been converted to N03 - within the trial period of 60 days despite addition of dicyandiamide to stop nitrification. Nitrification rates may even be higher in the rhizosphere than in the bulk soil due to the higher activity of nitrifying bacteria in the rhizosphere, as a result of abundant energy provided by carbon exudates from the roots. Rovira and McDougall ( 1 967) reported that the activity of nitrifying bacteria (nitrosomonas and nitrobacter) in the rhizospheres of wheat, maize and lucerne was much higher compared to that in the bulk soiL It is possible that the higher activity of these bacteria may have nitrified much of the � + in the fertilisers. This would have reduced pH in the rhizosphere of �)2S04 and �N03 treated soils - even though the plants took-up predominantly N03- from these fertilisers causing the roots to release OR or HC03- into the rhizosphere. Nitrification produces two moles of H+ per mole of N whereas N03- uptake produces one mole of Off per mole of N. The net result is therefore rhizosphere acidification. In contrast to the results reported in this Chapter, Xan and Jianyun ( 1 994) reported that tea plants preferentially absorb � � compared to N03-, because tea roots have less nitrate reductase activity to reduce any N03- taken-up. However, Selvendran ( 1 970) reported that there were appreciable amounts of nitrate reductase enzyme in the active white roots of tea. In the present study the soils may have had mainly the 1 6 6 N03- form ofN and low amounts ofNH/-N even when N was supplied in the NH/ form and therefore the plants had access mainly to N03--N. As the soils in this trial were not analysed at the end of the trial to determine the relative proportions of soil NII/ and N03-, it is not possible to verify the above suggestion. 6.5 CONCLUSIONS AND IMPLICATIONS The form of N supplied to tea had a marked influence on the rhizosphere pH and on the uptake and utilisation of soil P and P from EPR fertiliser. The supply ofNH/-N decreased rhizosphere pH resulting in higher P dissolution and adding N03- increased rhizosphere pH, which reduced P dissolution in the rhizosphere. Irrespective of the form of N, phosphate rock dissolution in the rhizosphere was always greater than in the bulk soil . Exchangeable (resin-P) and weak alkali (0. 1 M NaOH-P; ) extractable soil P forms were depleted in the rhizosphere due to plant and microbial uptake of P . Weak alkali extractable organic P (0 . 1 M NaOH-Po), accumulated and this may be due to the transformation of Pi into Po, by the high microbial activity in the rhizosphere compared to that in the bulk soil . The depletion of NaOH-P; in the rhizosphere was lowest when NH/-N was supplied to the plant due to the higher P fixation caused by the lower rhizosphere pH compared to when N03--N was supplied to the plant. The cation-anion balance studies in the plant indicated that tea removed more N03--N than NHt -N from the soil irrespective of the form of N added to the soil. These results together with the results of the forms of N supply on rhizosphere pH suggest that nitrification of� � was reasonably rapid relative to N uptake. The calculation of the proportion of N taken-up as �+ and N03- in this study was based on several assumptions. These results require confirmation through further investigation, perhaps using 1 5N labeled � + and N03 - fertilisers, higher doses of split application of nitrification inhibitors and measuring � + and N03- concentration in the soils at different periods during plant growth. Alternatively, the plants could be fed continuously with dilute solutions ofN and nitrification inhibitors in the experiments. 1 6 7 CHAPTER 7 THE FATE AND EFFECTS OF PHOSPHATE FERTILISERS ON PHOSPHORUS AVAILABILITY TO TEA (Camellia sinensis L.) IN A HIGHL Y ACIDIC UL TISOL IN SRI LANKA 7.1 INTRODUCTION Tea soils frequently present problems that have constrained the development of a successful sustainable tea production, due to conditions associated with high soil acidity and deficiency in plant-available P caused by fixation of P by Fe and Al oxides and hydroxyoxides (Bhattacharyya and Dey, 1 983 ; Golden et aI . , 1 98 1 ). As these soils have very low pHs (pH in water <5 . 5) and receive high rainfall (>2000 mm), phosphate rock (PR), when applied to these soils, is expected to dissolve and supply adequate amounts of P to plants (White et al . , 1 989). Therefore direct application of finely ground locally available PR may be an economically attractive alternative to the use of more expensive imported soluble P fertilisers. The rate of dissolution of PRs in soil, and hence the potential availability of P to plants, depends on the properties of the PR and on soil factors such as pH, P sorption capacity (Bolan and Hedley, 1 989; Chien et aI. , 1 980; Mackay et al . , 1 986), exchangeable Ca content and CEC (Mackay et al . , 1 986; Robinson and Syers, 1 99 1 ) . Although the supply of acidity is a prerequisite for PR dissolution it may not necessarily result in an increase in plant­ available P (Syers and Mackay, 1 986) because low soil pH also causes high P fixation resulting in a decrease in plant-available P (Apthorp et al . , 1 987; Sanchez, 1 976). In recent times many countries have been attracted by the possibility of using PRs, to increase agricultural production, particularly those having indigenous PRs (Sale and Mokwunye, 1 993). In mid 1 970s a large PR deposit estimated to be about 40 million metric tones (Jayawardene, 1 976) was discovered at Eppawala in the North-central province of Sri Lanka (Chapter 2, section 2 . 5) . This PR (Eppawala phosphate rock, EPR) is now recommended as a P fertiliser for direct application to many crops 1 6 8 including mature tea in Sri Lanka (Dahanayake et al. , 1 995) in spite of inconclusive experimental evidence available on the agronomic effectiveness ofEPR on tea. Mature tea plants seldom showed yield responses to the application of soluble and sparingly soluble P fertilisers (Willson and Clifford, 1 992). Eden ( 1 949) reported that the application of a PR fertiliser (saphosphosphate - a blend of Egyptian PRs) at a rate of 1 5 kg P ha-1 y(l gave the maximum yield response by tea on an Ultisol in Sri Lanka with no further response accruing from higher rates of application. In India, a comparison of the yield responses to a range of easily soluble and sparingly soluble P fertilisers showed that all P forms applied at a rate of 26 kg P ha-1 yr-l increased tea yield over the control (no P fertiliser) treatment but no significant difference in yield between the P forms (Ranganathan, 1 97 1 - 1 980). In Sri Lanka, EPR was compared with imported saphosphosphate as P fertiliser for young tea at an application rate of 1 5 kg P ha-1 under glasshouse conditions and found that there were no yield responses to either fertilisers (Sivasubramaniam et al., 1 98 1 ) . Therefore it was not possible to compare the relative agronomic effectiveness of the two P fertilisers. There is also no published information on the reactions and transformation of P applied in fertilisers to tea soils in any of the tea growing countries. The present experiment was therefore designed to test the effect of Eppawala phosphate rock (EPR) and Triple superphosphate (TSP) applied at different rates on the dry matter yield of tea and the transformation of fertiliser P in a Sri Lankan acid soil. Soil P tests are important tools for assessing the availability of soil P to plants and determining P fertiliser requirements of plants. This aids efficient use of P fertilisers, limits wastage of fertiliser materials and minimises pollution hazards in soils and water bodies. Many soil tests have been developed in the past to estimate the pool of plant­ available P in soils. The performance of different tests is influenced by soil properties, the climate and the crop grown. The Olsen-P test (NaHC03• pH 8 . 5) has been used for wheat in India (Gattani and Seth, 1 973) and Bolivia (Waugh and Manzano, 1 97 1 ) and for coconut in Sri Lanka (Loganathan et al . , 1 982). This method i s also successful in predicting soil P availability to pasture in soils fertilised with soluble P fertilisers in New Zealand (Saunders et aI . , 1 987) and Australia (Colwell, 1 963). In acid soils, Bray- l and Bray-2 tests have been found to be suitable in estimating plant-available P 1 6 9 for a number of crops (Fixen and Grove, 1 990), An acid borax extractant (pH 1 . 5, Beater, 1 949) has been used in Sri Lanka to predict P availability to tea (Jayman and Sivasubramaniam, 1 980), but no experimental evidence is available on which to base a comparison of the suitability of the borax test with other soil test methods. 7.2 OBJECTIVES The objectives of the study reported in this chapter are: ( 1 ) To determine the chemical changes in the P fractions in a highly acidic Sri Lankan Ultisol fertilised with two forms ofP fertilisers (EPR and TSP) over 1 0 months of tea seedling growth. (2) To determine the extent of plant-induced dissolution ofEPR in the acid soil. (3) To compare the agronomic effectiveness ofEPR with TSP in the above trial . (4) To determine the most suitable soil test that can predict dry matter yield of tea in the above trial. (5) To determine whether leafP concentration is a good index ofP supply to tea plants. 7.3 MATERIALS AND METHODS The soil used in the study was collected in St. Coombs estate( 1 382 m amsl), Talawakelle, Sri Lanka, where tea has been cultivated for over 50 years. The soil belongs to the Red Yellow Podsolic Great Soil Group (RhodustuIt according to US Soil Taxonomy, De Alwis and Panabokke, 1972). Selected properties of the soil are presented in Table 7 . 1 . After removing the surface litter, the soil was collected from a o - 1 5 em depth, air-dried at room temperature, lumps broken and passed through a 2 . 0 mm sieve. Approximately 4 . 5 kg of air-dried soil was then weighed into each of 1 07 plastic pots. Eight month old tea plants (TRI 3072) of similar size were removed from nursery bags and soil adhering to roots were carefully removed by immersing in water. The Table 7. 1 Selected properties of the soil (Rhodustult) used Soil property Unit Sand % Silt % Clay % pH (Soil : H20 1 : 2 . 5 w/w) Organic C % Effective CEC 1 emolc kg- l Total N % Ex. Na emoic kg-l EX. K emole kg-l Ex. Ca emolc kg- l Ex. Mg emolc kg"! Ex. Al emolc kg-l pH buffer capacity (at pH 4-5) mmol H+ kg-l pIT! Resin-P �g g- ! soil Olsen-P ).tg g-! soil Bray- l P �g g-l soil Borax-P ).tg g- l soil P-fixing capacitl % lExchangeable Ca + Mg + Na + K + AI + H 2Blackmore et al . ( 1 987) 1 7 0 Value 49 23 28 4.55 2. 1 2 4.07 0 . 1 6 0 . 1 2 0 . 1 7 0 . 1 6 035 1 .29 1 7 2 40 4 6 95 1 7 1 plants were weighed and then planted in pots. Each pot received either Triple superphosphate (TSP, total P 20% and 85-95% of total P dissolved in water) or Eppawala phosphate rock (particle size 5 .2% > 250 !lm; 39.2% 1 50 - 250 !lm; 39.9% 1 50 - 75 !lm; 1 5 .7% < 75 !lm, total P 1 4.5%, citric acid (2%) soluble P 1 .97%, almost insoluble in water) as the P source. Triple superphosphate and EPR were applied evenly on the surface of the soils at the rates of 0, 1 0, 20, 3 0, 40, 5 0 and 60 kg P ha-1 ( 1 0 kg P ha-1 is equivalent to 6 !lg P g"l soil assuming a bulk density of 1 . 1 Mg M3 and 0- 1 5 cm soil depth) at the beginning of the trial. The treatments were replicated six times. Nitrogen (N) and potassium (K) were applied on to the soil surface at rates of 1 20 kg N and K ha-1 as CN"H4hS04 and KCI respectively in four split applications, with the first application at the time of planting and the final application nine months later. Duplicate pots were filled with the same quantity of soils receiving the same P treatments, but without tea plants (fallow pots) to study the effect of the P treatments on soil P status in the absence of plants. Nitrogen and potassium were also applied to these pots as was done in the case of planted pots. All the pots were arranged in a completely randomised design and kept in a glasshouse maintained at 1 3° C minimum and 25° C maximum at the St. Coombs estate in Sri Lanka. The soil in the pots was kept at field capacity moisture content throughout the trial period. At the end of 5 months, three replicates of each treatment with plants were dismantled for soil and plant analyses. Soil samples were also taken from the unplanted pots (fallow pots) for analysis at the same time. After 1 0 months the remaining plants were harvested, plant and soil samples were taken for analysis as in the first harvest. The plant growth system in the glasshouse is shown in Figure 7 . 1 . 7.3.1 Plant and soil analyses Plant samples were oven-dried at 60° C and ground to < 1 . 0 mm. Both shoot and root samples were analysed for total P by the vanadomolybdate method (Jackson, 1 958) . Soil samples were air-dried and analysed for plant-available soil P according to the 1 7 2 Figure 7.1 Tea plants growing in pots in the glasshouse study 1 7 3 methods of mixed cation and anion resin strip test (Saggar et al., 1 990), Olsen test (Olsen et al. , 1 954), Bray-l test (Bray and Kurtz, 1 945), borax extraction (Beater, 1 949), 2% citric acid extraction and 2% malic acid extraction (Sivasubramaniam et al. , 1 98 1 ) (see Table 7 .2). Exchangeable cations were extracted by 1 M N'l40Ac buffered at pH 7.0 and determined by the method of Blackmore et al. ( 1987) and soil organic C by the method of Walkley and Black ( 1934). Soil pH, pH buffering capacity, P fractionation and the amount of EPR dissolution in the soil was determined as described in Chapter 3 . The statistical analyses were carried out using the Statistical Analysis System (SAS) software package (SAS, 1 985). 7.3.2 Relative agronomic effectiveness (RAE) The agronomic effectiveness (RAE) of EPR, relative to the standard TSP was calculated from the yield and P uptake response relationship according to the "vertical" comparison method (Saggar et al. , 1993) using equations 7. 1 , 7 .4 and 7.5 described below. (a) The RAE ofEPR was calculated from the cumulative dry matter yield at each harvest using the definition : RAE (%) = Yield with EPR (averaged over all rates ofP) - Control . 1 00 . . (7. 1 ) Yield with TSP (averaged over all rates ofP) - Control (b) The RAE was calculated from cumulative dry matter yield and P uptake for both P sources at five and ten months after P fertiliser application using a Mitscherlich-type equation described as: Y = a + b [ 1 - exp (-eX)] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7.2) where Y is the yield (g porl) at a P application rate of X (kg P ha-l), a and b are parameters that describe the yield on an unammended soil and the maximum yield increment when the nutrient (P) is not limiting. The parameter b is considered to be Table 7.2 Summary of soil P tests used P-test Extractant pH Soil : solution Time of ratio extraction Olsen 0.05 MNaHC03 8 .5 1 :20 30 min Bray- l 0.03 M�F + 0.025 MHCl 3 .0 1 : 7 5 min Borax extract 0.000 1 5 MNa2B407 buffered with c. H2S04 (98%) 1 . 5 1 : 10 30 min Anion and cation exchange resin strips Resin strips saturated with HC03- and Na+ - 6.8 1 :30 1 6 hrs Citric acid 2% Citric acid 2.5 1 : 1 0 1 hr Malic acid 2% Malic acid 1 . 5 1 : 1 0 1 hr 1 7 5 the responsiveness of the plant to the nutrient P added to the soil and parameter e describes the steepness of the response curve (Bennett and Ozanne, 1972). The Mitscherlich equation is often used to describe plant growth responses to nutrients (Campbell and Keay, 1970� Bennett and Ozanne, 1 972� Spencer et aI. , 1 980� Saggar et aI. , 1 993). This form of response curve implies that each successive increment in nutrient supply produces a diminishing increment in the yield. The Equation 7.2 can be rewritten as Y - a = b [ 1 - exp (-eX)] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7.3) The RAE ofEPR relative to TSP at any one of the rates ofP ( 1 0, 20, 30 ,40 ,50 and 60 kg P ha-1) was estimated using Equation (7.4), RAE (%) = bi [1 - exp (-elm * 1 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7.4) b2 [1 - exp (-ezX)] Where the subscripts 1 and 2 refers to EPR and TSP respectively. The effectiveness of EPR relative to TSP was also calculated from the ratio of the initial slopes represented as a product of maximum response (b) and the rate constant (c) of the fitted models (Barrow, 1 985� Chien et aI. , 1990a) i. e. RAE (%) = (blcl / b2C2) * 1 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7 .5) 7.4 RESUL TS AND DISCUSSION 7.4.1 EPR dissolution in soil The dissolution of EPR was higher in the soil with tea plants compared to that in the unplanted soil. Plant induced dissolution ofPR is now well documented (Bekele et aI. , 1 7 6 1 983; Bolan et al., 1 997; Hoftland et al., 1 989; Chapters 3, 4 , 5 and 6). One of the reasons for higher dissolution of EPR in the presence of plants is rhizosphere acidification (Chapters 4, 5 and 6) . Trolove et al. ( 1996b) also observed a higher dissolution of North Carolina Phosphate rock (NCPR) in the rhizospheres of white clover and lotus due a to higher acid production by roots in that zone. Other reasons are the removal of PR dissolution products, Ca and P by plant uptake and root secretion of organic acids (Jayman and Sivasubramaniam, 1975� Xiaoping, 1 994) as discussed in Chapters 3 , 4, 5 and 6. The proportion of EPR dissolved was greater at lower rates of EPR application compared to that at higher rates (Figure 7.2) though the amount of EPR dissolution was higher at higher rates (Table 7.3) . This was because at lower rates of EPR additions, there were proportionately higher amounts of acidity available for dissolution and also more sinks for the removal of the dissolved products P, Ca and F per unit of applied EPR. The amount of EPR dissolution in the soil was significantly higher at the 10 month sampling compared to that at the 5 month sampling, however significant differences (p <0.05) could be observed only among rates that were below 20 kg P ha-I (Figure 7.2). At the 1 0 months sampling, in the presence of plants, more than 50% of added EPR has dissolved at P application rates < 30 kg P ha-I . This suggests that at least 50% of EPR has the potential to dissolve in tea soils under field conditions, when applied at the rate of 20 kg P ha-I y(I currently recommended by the TRI (Tea Research Institute) of Sri Lanka. In line with the results of this experiment, Tambunan ( 1992) observed in field trials on Ultisols of Indonesia under Calopogonium the dissolution of NCPR and MPR (Moroccan phosphate rock) added at the rate of 80 kg P ha-I had increased with increased contact time of PR and soil. For example after 545, 360 and 1 80 days the dissolution of NCPR was 98, 82 and 40% respectively. As the experimental soil is highly acidic (pH 4 .55) and exchangeable Ca levels (0. 1 6 cmole kg-I) and % Ca saturation of the exchange complex (2.5%) are low, it provided favourable conditions for PR dissolution (Bolan and Hedley, 1 989; Mackay et al., 1986). Though the initial rate ofPR dissolution is high, with time it decreased because ,,--... p... ""0 Q) ""0 ""0 � <.0-< 0 "$. '--' ""0 Q) >- '0 CI) CI) . - ""0 p... 1 00 90 80 70 60 50 40 30 20 1 0 o I Lsd for main effect of plants (S months) I Lsd for interaction effects of plant . rate (10 months) ----+-- EPR 5 month "'ith plants � EPR 10 month "'ith plants -a- EPR 5 month without plants ____ EPR 10 month without plants 1 0 20 30 40 P rate (kg ha- 1 ) 50 60 Figure 7.2 Effect of time and EPR fertiliser rate of addition on P dissolution in soil with and without plants. Vertical bars correspond to Lsd at p <0.05 . 1 7 7 Table 7.3 The H+ consumption for EPR dissolution in soil with and without tea plants Rate ofEPR Experimental Amount of EPR application condition dissolved (kg P hao !) after 5 months (Ilg P gO ! soil) l O With tea 3 . 1 2 ± 0. 1 7 20 plants 4.8 1 ± 0 .24 30 6 .28 ± 0 .26 40 7. 1 6 ± 0 .2 1 50 7.49 ± 0 .43 60 7.62 ± 0 .28 l O Without tea 2.28 ± 0 .08 20 plants 3 . 59 ± 0.2 1 30 4 .54 ± 0 .25 40 4 .88 ± 0. 3 1 50 4 .61 ± 0.39 60 3 .67 ± 0 .27 H+ . 1 D consumptton or EPR dissolution in soil after 5 months (Ilmol H+ gO! soil) 0.20 ± 0.0 1 1 0 .3 1 ± 0.0 1 5 0 .4 1 ± 0.0 1 7 0 .46 ± 0.0 14 0 .48 ± 0.028 0 .49 ± 0.0 1 8 0. 1 5 ± 0.005 0 .23 ± 0.0 1 4 0 .29 ± 0.0 1 6 0 .3 1 ± 0.020 0 .30 ± 0.026 0.24 ± 0.0 1 7 Amount of EPR dissolved after 1 0 months (Ilg P gO! soil) 4.50 ± 0 .22 6.60 ± 0 .48 8 .35 ± 0 .2 1 9 .72 ± 0 .36 l O. 52 ± 0.23 1 1 . 9 1 ± 0 .44 3 .32 ± 0. 1 2 5 . 1 6 ± 0. 1 6 6.80 ± 0 .24 6.97 ± 0 .26 6 .21 ± 0 .32 5 .69 ± 0 .3 1 H+ . 1 £ consumptton or EPR dissolution in soil after 10 months (Ilmol H+ gO ! soil) 0.26 ± 0.035 0 .43 ± 0.03 1 0 .54 ± 0.0 1 4 0 .63 ± 0.024 0.68 ± 0.0 1 5 0.77 ± 0.029 0 .2 1 ± 0.008 0.33 ± O.O l O 0.44 ± 0.0 1 6 0.45 ± 0.0 1 7 0.40 ± 0.02 1 0.37 ± 0.020 Contribution of soil pH increase due to EPR dissolution2 0.0 1 5 0.025 0 .032 0.037 0.040 0.045 0.0 1 2 0.0 1 9 0.025 0.026 0.023 0.02 1 lAmount ofEPR dissolved (Ilg EPR gO! soil) * amount ofH+ required to dissolve EPR (0.00933 Ilmol W Ilgo1 EPR - see text) 2 Amount of H+ consumed for dissolution of EPR divided by the pH buffering capacity 1 7 9 the dissolution products, Ca and P accumulate and soil pH increases near the PR particles in the soil (Hammond et al., 1 986b). 7.4.2 Effect of P fertilisers on soil pH At the beginning of the trial, the soil pH was 4.55 (H20) and it significantly decreased during the trial period for both P treatments (EPR and TSP) in pots without plants (Figure 7.3). This may be due to nitrification of � +, which was derived from (NH..)2S04 added as N fertiliser at the beginning of the trial and subsequently at each 3 month interval. Another reason for the decline in pH could be due to the increase in ionic strength caused by the addition ofKCI and (NH..)2S04 and the mineralisation of organic matter; these processes would have reduced the pH measured in H20. The EPR dissolution would have consumed some of the acidity, but this seems to be of a lower order of magnitude (Table 7.3) compared to the acid generating process discussed above. The reduction in soil pH was however not statistically significant in the pots with tea plants. This may be due to N03- uptake by the plants resulting in the release of Off or HC03 - to the soil, which neutralised part of the acidity produced in the soil through nitrification. Another reason could be that the ionic strength in the presence of plants may have been lower because of plant uptake of ions and therefore the pH measured in H20 would have been influenced to a lesser extent in soils with plants compared to those without plants. The amount of acidity consumed for EPR dissolution in the soil with and without plants at the 5 and 1 0 month samplings was determined as described in Chapter 4, section 4.4.3 (Table 7.3) . At both sampling times the acid consumption for EPR dissolution in the presence of plants was greater than in their absence. 7.4.3 Effect of forms and rates of P fertilisers on soil P fractions At the end of the trial (at 1 0 months) approximately 1 00% of the added fertiliser P was recovered in the various soil P fractions from the unplanted pots for both P fertiliser treatments (Table 7.4). The lower recovery ofP from soils in planted pots is (a) 4.60 4 .50 4.40 4.30 4.20 4. 1 0 4.00 o NS (5 months) NS (10 months) 1 0 ---+- EPR 5 months -+- EPR 10 months -D- TSP 5 months _ TSP 10 month. 20 30 40 P rate (kg ha- 1 ) 50 60 (b) 4 .60 4 .50 4 .40 4 .30 4.20 4 . 1 0 4 .00 o Initial soil pH I Interaction (P sources " rate) Lsd (5 months) I Lsd main effects of P sources (10 months) 1 0 20 30 40 P rate (kg ha- 1 ) 50 60 Figure 7.3 Effect of EPR and TSP on soil pH (H 0) after 5 and 1 0 months of appl ication (a) with and (b) without tea plants. 2 Vertical bars correspond to Lsd at p <0.05 and NS represents treatments not stati stically significant at p <0.05 . f-' 00 o Table 7.4 The % recoveryl of added P in soil P fractions in the unfertilised (control) and EPR and TSP treated soils at the end of the trial Rate of P With or Resin-P NaOH-Pi NaOH-Po application without (Ilg g-I soil) (Ilg g-I soil) (Ilg g-I soil) (kg ha- I ) plants EPR 0 With 1 . 9 197.5 1 14.8 10 1,Iants 2.5 199.7 1 16.3 20 3 .3 202.3 1 1 5 .9 30 4.3 205.9 1 1 7 .5 40 5.3 206.9 1 17.6 50 6.4 208.7 1 1 8 .5 60 8.0 2 10.2 1 1 8 .5 EPR 0 No 1,Iants 2.8 209.7 122 .8 10 3 .8 2 1 3 .7 123 .9 20 4.9 2 1 5.2 1 25 .3 30 5.4 2 16. 1 1 24.7 40 6.8 2 17.3 124.8 50 8.8 2 17.0 126.3 60 10. 1 2 18.3 128 .3 TSP 0 With 1 .9 197.5 1 14.8 10 plants 2.6 200.4 1 1 4.4 20 3 .4 206.6 1 1 5 .4 30 4.3 207.9 1 16.6 40 5.5 209.7 1 17.5 50 6.2 2 1 1 .0 1 1 9.7 60 8.2 2 1 5.0 1 2 1 . 1 TSP 0 No t,tants 2.8 209.7 122.8 10 3 .7 2 16.5 124. 1 20 4.9 220.4 126.2 30 5 . 1 224.8 127. 1 40 6.0 229.6 1 27. 1 50 7.5 234.3 128. 1 60 9.4 237.0 1 3 1 .2 1 %P recovery = 0: P fractions in fertil ised soil - L P fractions in unfertilised soil) * 100 amount of fertiliser P added to the soil H2SO4-Pi Residual-P Total-P Total P (Ilg g-I soil) (Ilg g-I soil) (Ilg g-I soil) recovery from applied P (%) 34.2 1 20.5 468.9 ---- 35. 1 1 1 9.5 473 .0 70 34.2 1 2 1 .4 477. 1 69 40.2 1 1 8.2 486. 1 96 42.5 1 1 9.4 49 1 . 7 95 44. 2 1 20.7 498.5 99 44.6 1 1 9.7 500.9 89 35. 1 1 24.2 494.6 ---- 37.6 1 22. 1 50 1 .0 106 40.6 1 2 1 .0 506.9 102 45.2 1 23 .5 5 14 .9 1 1 2 48.9 1 23 .0 520.8 109 52. 2 1 23 .2 527 .5 109 52.0 1 24.2 533.0 107 34.2 1 20.5 468.9 ---- 34.0 1 2 1 .4 472.7 65 35.7 1 17.0 478.0 76 36.2 1 1 8.7 483.7 83 37.3 1 2 1 .8 49 1 .7 95 36. 5 124.2 497.5 96 37.5 120.4 502.3 93 35. 1 1 24.2 494.6 ---- 35. 1 1 2 1 .6 50 1 .0 106 36. 1 1 20. 1 507.6 108 37.5 1 19.9 5 14.5 1 10 36. 9 122.9 522.4 1 16 39.0 1 1 7.7 526.7 107 37.7 1 1 8.9 534.2 1 10 1 8 2 due to P uptake by plants. The changes caused by P fertiliser application in the individual soil P fractions, resin-P, NaOH-Pi, NaOH-Po and H2S04-Pi at the 5 and 10 month sampling times are shown in Figures 7 .4, 7.5, 7 . 6 and 7 . 7 respectively. Unlike EPR, the highly soluble TSP fertiliser had quickly dissolved and mobilised into various soil P fractions. The applied TSP was mainly recovered as NaOH-Pi whereas the majority of EPR was recovered as undissolved EPR in the H2S04-Pi fraction as observed in Chapter 5 . In the TSP treatment water initially moves into TSP granules and dissolves monocalcium phosphate forming a metastable triple point solution containing dicalcium phosphate and free phosphoric acid. The solution coming out of the granule has a pH of 1 to 1 . 5 (Sanchez, 1976). In acid soils Fe and AI in the solution and the exchange phase are abundant and they react with P to form relatively insoluble Fe and AI phosphates (Coleman et aI. , 1 960). These P forms are extracted by NaOH in the P fractionation scheme. In both P fertiliser treatments the concentration of all P fractions increased with increasing rates of P application, but the rate of increase per unit increase of P application is highest for NaOH-Pi in the TSP treatment and for H2S04-Pi in the EPR treatment (Figure 7.5, 7 .7) . Trolove et al. ( 1 996b) also found an increase in the resin­ P and NaOH-Pi fractions in MCP (the chemical components in TSP) treated bulk soil compared to the control treatment, whereas in NCPR treated soil only resin-P and H2S04-Pi increased. My results compared well with the results obtained from a trial in Sri Lanka on a sandy soil, where long-term ( 10 yrs) annual application of concentrated superphosphate to coconut, had significantly increased Fe + AI bound P (NaOH-Pi) concentration, whereas application of a PR increased the concentration of P bound to Ca (H2S04-Pi) compared to the no fertiliser treatment (Loganathan and Nalliah, 1977). In an acid (pH 4.6) Ultisol in Cavinti, Philippines, Hedley et al. ( 1 994) recovered only 5% ofP from applied MCP in the resin fraction because 75-80% ofP was transformed into NaOH-Pi fraction within six weeks in upland rice growth. The labile resin-P concentration in the TSP treated soils is generally higher than that in the EPR treated soils, the differences were statistically significant (p <0.05) and prominent in the unplanted soil for the majority of the rates of P fertiliser application especially at 5 months (Figure 7.4). This is due to the higher solubility of TSP -.. '0 '" - , on on :::1. "-'" 0... I c: '" <: v ><: CI:S ..... 0 CO (a) 40 � • 30 C • 20 1 0 o 1 0 EPR 5 months R2= 0.98 EPR 10 months R2= 0.96 TSP 5 months R2= 0.92 TSP 10 months R2= 0.97 20 30 40 P rate (kg ha- 1 ) 50 60 (b) 40 � • 30 C • 20 1 0 o 1 0 EPR 5 months R2= 0.99 EPR 10 months R2= 0.98 TSP 5 months R2= 0.97 TSP 10 months R2= 0.97 20 30 40 P rate (kg ha- 1 ) • 50 60 Figure 7. 10 Effect of EPR and TSP fertiliser rates on Borax extractable-P in soil (a) with and (b) without tea plants. (a) (b) 1 2 1 2 � EPR 5 months R2= 0.85 � EPR 5 months R2= 0.96 • EPR 10 months R2= 0.89 • EPR 10 months R2= 0.96 1 0 C TSP 5 months R2= 0.94 1 0 [] TSP 5 months R2= 0.96 • TSP 10 months R2= 0.88 • TSP 10 months R2= 0.96 ,--.. 8 8 .......... . - 0 en '7 on on 6 6 =1- '--" 0... I >, ro .... CO 4 4 2 2 o 1 0 20 30 40 50 60 o 1 0 20 30 40 50 60 P rate (kg ha- 1 ) P rate (kg ha- 1 ) Figure 7.1 1 Effect of EPR and TSP fertiliser rates on Bray- l P in soil (a) with and (b) without tea plants. 6 ,-.. 5 -'0 EPR 5 months R2= 0.97 � EPR 5 months R2= 0.94 • EPR 10 months R2= 0.99 • EPR 10 months R2= 0.99 C TSP 5 months R2= 0.99 C TSP 5 months R2= 0.99 • TSP 10 months R2= 0. 89 5 • TSP 10 months R2= 0.86 4 3 2 1 o 1 0 20 30 40 50 60 o 1 0 20 30 40 50 60 P rate (kg ha- 1 ) P rate (kg ha- 1 ) Figure 7.12 Effect ofEPR and TSP fertiliser rates on Citric acid extractable-P in soil (a) with and (b) without tea plants. 6 5 1 (a) (b) 6 � EPR 5months R2= 0.98 � EPR 5 months R2= 0.91 • EPR 10 months R2= 0.93 • EPR 10 months R2= 0.97 a TSP 5 months R2= 0.98 [J TSP 5 months R2= 0.96 • TSP 10 months R2= 0.90 5 • TSP 10 months R2= 0.97 • 4 3 Initial Malic-P o 2 1 1 0 20 30 40 50 60 o 1 0 20 30 40 50 P rate (kg ha- 1 ) P rate (kg ha- 1 ) Figure 7.13 Effect of EPR and TSP fertiliser rates on Malic acid extractable-P in soil (a) with and (b) without tea plants. 60 1 9 6 compared to those at 5 months could be due to the higher amounts of NaOH-Po at 1 0 months (Figure 7.6). Tambunan ( 1 992) observed that applications of NCPR and TSP to an acidic moist Ultisol of Indonesia at rates of 1 40 kg P ha-l upto 560 kg P ha-l increased resin, Bray- l and Olsen extractable P. In my trial the extractable P values generally increased in the order of Olsen > borax > resin and Bray- I > citric and malic acid. Olsen P values (3 5- 1 00 Ilg g-l soil) in the soils used in this study are very high compared to those in other tropical soils (El Swaify et al., 1 98 5 ; Loganathan et al., 1 982), because of the high rates of previous P applications which have raised total soil P in tea soils (mean of 6 1 6 Ilg g-l soil, Golden et al., 1 98 1 ) . In contrast to tea soils, the coconut growing soils in Sri Lanka have been reported to have low amounts of total P (37 - 3 3 8 Ilg g-l soil with a mean of 140 Ilg g-l soil), because the latter soils have not been regularly fertilised (Loganathan et al., 1 982). Olsen P concentrations in the coconut soils of Sri Lanka were reported to be an average of 2 Ilg g-l soil (Loganathan et al. , 1 982), which was much lower than the values obtained in this study (Figure 7.9). The Olsen test produced higher soil test P values than the acidic extractants, Bray- I , citric and malic acids because the Olsen method extracts both F e and AI bound P (NaOH-Pi) and organic P, whereas the acid extractants dissolve mostly AI-P (Le Mare, 1 99 1 ) or Ca-P. The other reasons for the higher P values with the Olsen extraction were due to the reduction of P fixation at the high equilibrium pH (pH 8 . 5) (Barrow, 1 984) and desorption of fixed-P (Fe and AI bound P) by ligand exchange with HC03-. The difference between the Olsen P values and those obtained by other methods was more marked, because the soil used in this study had much higher Fe-P values than AI-P values (Fe-P of 1 69 Ilg g-l soil vs AI-P of 59 Ilg g-l soil; Golden et al. , 1 98 1 ). The soils also had large amounts of organic P ( 1 1 0 - 1 3 0 Ilg g"l soil, Figure 7 . 6) and it is possible that part of this organic-P (labile pool of organic-P) may have dissolved in the alkaline NaHC03 exatractant resulting in higher Olsen-test values. The lower time (5 mins) of extraction may be another reason for the low Bray- l P values in this high P-fixing soil. Saggar et al. ( 1 995) also showed that Bray- l extractant gave very low recoveries of added P in a high P-fixing New Zealand soil 1 9 7 (Dystrandept, 4 . 2 J.1g gol soil) compared to that in a low P-fixing soil (Haplohumult, 75 .5 J.1g gol soil) when the soils were treated with monocalcium phosphate. The acid extractant borax dissolved greater amounts of P from all soils than the other acid extractants. The reason for higher borax extractable P values compared to the other acidic extractants (Bray-1 , malic acid and citric acid) may be that the tetra borate ion (B40l) is more effective in extracting fixed P because of the strong energy of adsorption ofB4olo anion to Fe and Al oxides in soils. In addition in EPR fertilised soils the pH (pH 1 . 5) of borax will cause increased dissolution and extraction of P from dissolved EPR residues. This may be the reason why the borax test extracted more P from EPR treated soils than TSP treated soils (Figure 7. 1 0). 7.4.5 Effect of P fertilisers on growth and P uptake of tea plants The application of P fertilisers increased shoot dry matter yield and total P uptake in tea plants compared to the control treatment (Table 7.5). The P rates beyond 20 kg P haol did not significantly improve shoot dry matter yield nor shoot P uptake in any of the two P fertiliser treatments at both sampling times. There was no significant difference in shoot dry matter yield between the two forms of P fertilisers. This shows that EPR is able to dissolve at a rate sufficient to provide amounts of plant-available P equivalent to that from the completely soluble TSP fertilisers. Tambunan ( 1992) also found that NCPR was more effective than TSP in increasing maize yields in an acidic Ultisol from Indonesia with higher moisture content compared to soils in a dry area because soils with high moisture increased NCPR dissolution. Similar results have been reported for acid soils in many parts of the world and now there is ample evidence to support the claim that PRs are as effective as soluble P fertilisers to maintain P requirements for many crops provided the soils have conditions favourable to PR dissolution (Alston and Chin, 1 974; Rajan and Gallingham, 1 986; Tambunan, 1 992). Increases in shoot dry weight and P uptake in response to P fertiliser application fitted very well to Mitscherlich-type equations (Equation 7.2; Figure 7. 14 and 7. 1 5). Using Table 7.5 Effect ofEPR and TSP fertilisers and their rates on shoot dry matter yield and P uptake After 5 months Rate ofP Shoot Leaf-P Shoot Shoot kg P hao l Dry matter yield 1 . 1 ,2 concentratIon Plant P uptake 1 Dry matter yield 1 (g pori) (%) (mg P pori) (g pori) EPR TSP EPR TSP EPR TSP EPR TSP 0 3 1 .8 5a 3 1 . 85a 0. 1 78a 0. 1 78a 34.28a 34.28a 40.45a 40.45a 1 0 35 .03b 33 .30ab 0. 1 83a 0 . 1 83a 39.48b 38.27a 44.74ab 43 .36a 20 35 .70b 33 .95ab 0. 1 87b 0. 1 86a 43 .26b 4 1 .22b 46.21b 44.20ab 30 35 .88b 35 .88b 0. 198c 0. 1 96b 45.2 1bc 44.89bc 48. 3 1b 46.22b 40 36. 1 1 b 35 .54b 0.200c 0.200b 45 .82bc 44.89bc 47. 38b 45 .94b 50 35 .99b 35 .62b 0. 199c 0.200b 45 . 5 1 bc 43 .99bc 47.99b 45 .63b 60 36.46b 35 .50b 0. 202c 0. 199b 47. 19bc 43 .88bc 48.09b 45.33b Lsd3 (p <0 .05) NS NS NS NS 1 Values representing the same letters in columns are not significantly different at p <0.05 by DMRT test 2First mature leafP concentration 3F or comparison of EPR and TSP After 10 months Leaf-P Shoot . 1 ,2 concentratIOn Plant P uptake 1 (%) (mg P pori) EPR TSP EPR TSP 0. 1 77a 0. 1 77a 43 .73a 43.73a 0. 1 82a 0. 1 84a 50.22b 50.03b 0. 1 86a 0. 1 87a 52.94bc 52.32bc 0. 197b 0. 195b 57.94c 55 .30bc 0.20 1b 0. 199b 56.32c 57 .82c 0. 199b 0.20 1b 56.63c 56.36c 0.20 1b 0. 1 98b 62.03d 55 .84bc NS NS "'"' - , ..... 0 0. eo '--' "'0 Q) .- ;>. .... � :t: ro E ;>. .... "'0 ..... 0 0 ..s:: en 50 45 40 35 30 o • - - . C JL - - .. - - - -• o EPR 5 months Y = 31.865 + 4.273 (1 - exp (-0.126 X) R2 = 0.99 --&- EPR 10 months Y = 40.437 + 7.707 (1 - exp (- 0.080 X) R2 = 0.98 • TSP 5 months Y = 31.710 + 4. 114 (1 - exp (- 0.069 X) R2 = 0.95 -___ - TSP 10 months Y = 40.399 + 5.455 (1 - exp (- 0.080 X» R2 = 0.94 1 0 20 30 40 P rate (kg ha- 1 ) 50 60 Figure 7.14 Relationship between shoot dry matter yield and the rates of EPR and TSP at 5 and 1 0 months after application to the soil 1 9 9 -.. ..... 0 0.. 00 E '--' � ..!Io:: S 0.. :3 ,:l.... .... 0 0 � CZl 65 60 55 50 45 40 35 30 25 o � () • --e- � -.- • EPR 5 months Y = 34. 179 + 12.903 (1 - exp (-0.057 X» R2 = 0.99 EPR 10 months Y = 44.018 + 17.471 (1 - exp (-0.038 X) R2 = 0.92 TSP 5 months Y = 33.947 + 11.030 (1 - exp (-0.062 X) R2 = 0.95 TSP 10 months Y = 43.657 + 13.503 (1 - exp (-0.062 X» R2 = 0.96 1 0 20 30 40 50 60 P rate (kg ha-1) Figure 7.15 Relationship between shoot P uptake and the rates ofEPR and TSP at 5 and 1 0 months after application to the soi l 2 00 2 0 1 these fitted equations, the P requirement to achieve 95% of the maximum yield was calculated for both forms of P fertilisers and presented in Table 7.6. The estimated P requirement values were very low (7 - 14 kg ha-1) indicating that the experimental soil is only marginally deficient in P and the response to P application was obtained only at the very lowest rate and then it diminishes to show no response at all. Sivasubramaniam et al. ( 198 1 ) compared an imported phosphate rock (saphos phosphate) with EPR at rates of 0 and 1 5 kg P ha-1 in a glasshouse trial on tea and found that there was no significant yield increase due to the application of any of the P fertilisers compared to the control treatment and therefore they were not able to compare the relative agronomic effectiveness of the two P fertilisers. But they observed that malic-acid extractable P was greater in the EPR treated soil compared with the imported phosphate rock treatment and this made them to conclude that EPR is as good as the imported phosphate rock for fertilising tea. They used malic acid as the extractant because tea roots were found to excrete large quantities of malic acid (Jayman and Sivasubramaniam, 1 975; Xiaoping, 1 994). 7.4.6 Relative agronomic effectiveness (RAE) of EPR The RAE of P fertilisers with different degrees of solubility may vary with crop species. In general, the RAE of PRs with respect to water-soluble P sources is expected to be higher for long-term or perennial crops than for short-term or annual food crops (Chien et al. , 1 990ab). Consistent with this, PRs have been used extensively for tree crops in Asia, particularly in Sri Lanka for tea (Wickremasinghe and Krishnapillai, 1 986), rubber (Peris, 1 970) and coconut (Loganathan, 1 978) and for rubber and oil palm in Malaysia (Chien and Menon, 1995). In this trial, the RAE values calculated for dry matter yield, using EPR as the test P­ source relative to TSP (standard P fertiliser ) are given in Table 7 .7 . The RAE values ofEPR were calculated using three methods (Equation 7. 1 , 7.4 and 7.5). The results obtained using 7. 1 and 7.4 show that in general the agronomic effectiveness of EPR is equal to or higher than TSP. At the 1 0 month sampling the RAE values were much higher compared to the 5 month sampling. The possible reason for the higher RAE of 2 0 2 Table 7.6 Calculated P requirement for 95% of maximum dry matter yield P fertiliser Growth Growth response curves Calculated P period (Y - yield, X - P rate) requirement for (months) 95% of the maximum yield (kg ha-1) EPR 5 Y = 3 1 . 865 + 4.273 ( 1 - exp ( - 0. 1 26 X» 7 10 Y = 40.437 + 7.707 ( 1 - exp ( - 0.080 X» 14 TSP 5 Y = 3 1 . 7 1 0 + 4. 1 14 ( 1 - exp ( - 0.069 X») 12 1 0 Y = 40.399 + 5 .455 ( 1 - exp ( - 0.080 X» 1 1 2 0 3 Table 7.7 The agronomic effectiveness ofEPR relative to TSP calculated from empirical relationships Equation Rate of P Dry matter yield Plant P uptake used for kg ha-1 5 months 10 months 5 months 10 months the calculation 7. 1 all 1 22% 143% 1 180/0 1 12% 7.4 10 149% 14 1% 1 10% 88% 20 128% 14 1% 1 12% 97% 30 1 16% 14 1% 1 13% 104% 40 1 10% 14 1% 1 14% 1 10% 50 106% 14 1% 1 15% 1 15% 60 105% 141% 1 16% 1 19% 7.5 all 189% 14 1% 107% 79% 204 EPR relative to TSP may be due to a continuous supply o f P by EPR through dissolution to meet the plant P requirement. Whereas TSP, dissolves within a short time and the dissolved P is immobilised into organic or inorganic soil P fractions with a lower P availability to tea. Tambunan ( 1 992) reported similar results, for field trials on a maize crop in Indonesia. The RAE of residues of NCPR applied seven months before planting maize was 26% more effective than TSP residues in a moist Ultisol at Sembawa, but 42% less effective on a dry Ultisol at Sarong. He further stated that at Sembawe the RAE of freshly applied TSP was higher than freshly applied PR, but its effectiveness decreased progressively with increasing time and became lower than that of PR after seven months. Utomo ( 1 995) compared a locally available phosphate rock from Lamongan, East Java (total P 14.4% and 2% citric acid extractable-P 5 .8%) with TSP on com (Zea mays L.) growth under glasshouse conditions using an Ultisoil (pH 4.6 in water) from Sumatra and reported that the RAE of PR is almost equal to that of the TSP fertiliser. Zaharah and Sharifuddin ( 1 995) compared the RAE of NCPR and an unreactive Chinese phosphate rock (CPR) with TSP consecutively for four cultivations of com on an acid soil (pH 4.99 in water) of the Tebok series (Typic Kandiudult) in Malaysia. They found that the agronomic effectiveness of NCPR was equal to TSP from the second harvest and for the CPR at the third and fouth harvests. In my experiments, the RAE values obtained using Equation 7 .5 were highly variable and higher at 5 months growth than at 1 0 months. In Equation 7.5, RAE is calculated from the ratio of the initial slopes of the response curves (Saggar et al. , 1 993). Because the yield or P uptake response to P fertilisers in this study is very low, the initial slopes are low and therefore the RAE calculated from the ratios of these slopes will lead to greater errors. This may be the reason for the highly variable results obtained using this equation. 7.4.7 Relationship between dry matter yield and P uptake by plants 2 0 5 A Logistic model (Equation 7.6) was fitted to the data relating dry matter yield to plant P uptake at each sampling time (5 and 1 0 months after P fertiliser application) (Figure 7 . 1 6). The logistic model is a sigmoid (growth) type function, which can be described by the following equation: Y = a /(l + exp (b - cX)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [7.6] where Y is the shoot dry matter yield and X is the plant P uptake, a is the maximum shoot dry matter yield (upper asymptote) when the plant P uptake is not limiting and c is related to the rate parameter, a higher c value indicates a rapid rise of the function between the lower and upper asymptote and ble defines as the value of X at the point of inflection (Causton and Venus, 1 98 1 ) . The models gave good fits t o data at both sampling times with R2 = 0 . 9 7 and 0 . 8 8 showing that the variability i n the yield i s satisfactorily explained b y the model. This relationship provides an estimate of the internal efficiency of P use within the plant. The curvilinear relationship shows that the internal efficiency of P utilisation by tea plants is lower at high amounts of P uptake. It indicates that the amount of dry matter produced per unit of absorbed P diminishes with increasing levels of plant P uptake and this could be attributed to limitation by other factors affecting dry matter yield. 7.4.8 Relationship between dry matter yield and leaf P concentration The shoot dry matter yield of tea plants harvested at 1 0 months was regressed against P concentration in the first mature leaf using a logistic type function (Figure 7. 1 7). The first mature leaf is the one with the axil from which pluckable shoots emerge. The rationale for taking the P concentration in the first mature leaf of tea as a diagnostic tool is that the mineral composition of this leaf varied least under the :::' .... 0 0... Ol) '-" '"0 � ';;" .... (1) � cd E >. .... '"0 ..... 0 0 ..c CIJ 50 45 40 35 30 25 • / / • . -,f4' . ... , ,- " . ,- 5 month Y = 37.49005/(1 + exp (2.74274 - 0.130 X» R2= 0.97 10 month Y = 51.88107/(1 + exp (1.91318 - 0.072 X» R2= 0.88 EPR .. TSP 20 �----�----r---�----�----�----.---� 30 35 40 45 50 55 60 65 Shoot P uptake (mg pori) Figure 7.16 Relationship between shoot dry matter yield and P uptake by tea plants at 5 and 1 0 months after P ferti l i ser appl ication 2 0 6 -- ...... o c. 00 '-" ""0 � >. .... CI) ..... .... c:I:$ E >. .... ""0 .... 0 0 ...s::: C/.) 50 • 48 • 46 e-- . • 44 42 40 I • EPR Y = 48.08011(1 + exp (30.3720 - 181.02035 X» Rl = 0.99 -. - TSP Y = 46.35111(1 + exp (18.4506 - 115.0669 X» Rl = 0.97 38 �-----.------.------.-----,------.-----� 0 . 1 75 0 . 1 80 0. 1 85 0. 1 90 0. 1 95 0.200 0.205 P concentration (%) in first mature leaf Figure 7.17 Relationship between shoot dry matter yield and P concentration in the first mature leaf for tea plants treated with EPR and TSP fertil iser 2 0 7 208 influence of changing environmental conditions (Hasselo, 1 965). Wilson ( 1969) found that the chemical analysis of the bud or the third leaf provides a better index of the fertiliser requirements in tea plants. The validity of this claim was tested by Sivasubramaniam and Jayman ( 1976) and they showed that the analyses of the first mature leaf, rather than the third leaf or bud would be a more reliable index of the nutrient needs of tea plants. Hasselo ( 1 965) also confinned that the first mature leaf provided a better index of the nutritional status of tea plants than the older or younger leaves grown under a wide range of environmental conditions. The shoot dry matter yield as a response to leaf-P concentration (Figure 7. 1 7) showed very good fits to Equation 7.6 (R2 = 0.99 and 0.97). The concentration of P (%) in the first mature leaf that was required to obtain 95% of the maximum dry matter yield was calculated from the regression equations and found to be 0. 1 85% for EPR treatment and 0. 1 86% for TSP treatment. These values agrees well with the 0.2% P value reported by Jayman and Sivasubramaniam ( 1980) for good tea growth. 7.4.9 Relationship between soil extractable-P and dry matter yield The shoot dry matter yield harvested at 1 0 months was regressed against the corresponding soil P test values determined by different chemical reagents (Table 7.2) using a Mitschelich-type equation (Equation 7.2). The plots for resin-P, Olsen-P, borax-P, Bray- I , citric acid extractable P and malic acid extractable P are presented in Figures 7. 1 8, 7. 1 9, 7.20, 7 .21 , 7.22 and 7.23 respectively. The R2 values for all plots were very high showing that any one of these soil tests can be used to predict P availability to tea. This is because all these extractants themselves are highly correlated whether all (Table 7 .8) or only low rates ofP application (Table 7.9, range where 95% of maximum yield was obtained) were considered. For each soil test, critical soil P levels associated with 95% of the maximum dry matter yield (g porl) were calculated using the regression equations (Table 7. 1 0). These critical values of soil P were found to vary highly between the type of "'0 � >. .... a.> � � E � "'0 ... 50 • • -- - . - - --• g 40 ..c:: C/.) , I • EPR Y = -26.3723 + 74.3761 (1 - exp (-1.2250 X) Rl = 0.98 - *- TSP Y= 2.561588 + 43.2681 (1 - exp (-1. 1160 X) Rl = 0.95 -- - --- Pool data Y = 29.396 + 17.82 (1 - exp(-O.645 X) Rl = 0.71 35 �--�--��--�--�--�--�----�--�--� 2 4 6 8 1 0 Resin-P (Ilg g-l soil) Figure 7.18 Relationship between shoot dry matter yield and resin-P for tea plants treated with EPR and TSP fertili ser. 2 0 9 50 :::' '0 45 0.. 00 '-' :::9 Q) >. .... Q) � E C -0 ..... g 40 (;3 I • - ...-..- - - - - - -e - - e - - e e / • EPR Y = -159.191 + 207.878 (1 - exp (- 0.0903 X)) R2 = 0.94 - -. - TSP Y = 4.32072 + 41.7164 (1 - exp (-0.0559 X)) Rl = 0.87 - -- - Pool data Y = - 5363.45 + 5409.65 (l-exp (-0.192 X)) R2 = 0.52 30 40 50 60 70 80 90 1 00 Olsen-P (llg g.1 soil) Figure 7.19 Relationship between shoot dry matter yield and Olsen-P for tea plants treated with EPR and TSP ferti li ser 2 1 0 ,.-... � 0 0. 00 '-" "'0 � >. .... v :t:: te E C "'0 ... 0 0 ..s:: C/.) 50 45 40 o - - /. // - 1/- --.- EPR Y = 37.8776 + 10.7794 (1- exp ( -0. 1313 X)) R1 = 0.94 -... - TSP Y = 37.731 2 + 8.87373 (I - exp (- 0. 1763 X)) R1 = 0.87 - -- - Pool data Y = 38.139 + 10.74 (1 - exp(-O.109 X)) R1 = 0.86 5 1 0 1 5 20 25 Borax-P (�g g-l soil) 30 Figure 7.20 Relationship between shoot dry matter yield and borax-P for tea plants treated with EPR and TSP fertili ser 2 1 1 50 45 .. g 40 ...s:: r/) 1 • EPR Y = 32.9527 + 17.0330 (1 - exp (- 0.2872 X) R1 = 0.97 - • - TSP Y = 33.8480 + 12.5386 (1 - exp (- 0.3712 X) R1 = 0.93 � -- - Pool data Y = 34.948 + 12.43 (1 - exp (-0.356 X» R2 = 0.68 2 3 4 5 6 7 8 9 1 0 Bray- l P (Jlg g -l soil) Figure 7.21 Relationship between shoot dry matter yield and Bray- l P for tea plants treated with EPR and TSP fertil iser 2 1 2 50 ,-... -;:.... o 45 0. 00 '-./ ::g � >. ..... � � E C "'0 ...... g 40 ...c: en ! • I .,­" • , .- . -., • - - - - ! .... - - • EPR Y = 36.9511 + 17.0341 (1 - exp (-0.2872 X) Rl = 0.97 -• - TSP Y = 46.3011/(1 + exp (-1.006 - 0.6132 X) Rl = 0.93 - - - - Pool data Y = - 61.65 + 108.43 (1 - exp (-1.682 X) Rl = 0.64 1 .5 2 .0 2 .5 3 .0 3 .5 4 .0 Citric acid extractable P ( Ilg g.lsoil) 4.5 5 .0 Figure 7.22 Relationship between shoot dry matter yield and citric acid extractable P for tea plants treated with EPR and TSP fertil iser 2 1 3 ,.-... ' ..... 0 0- 00 '-" ""0 d) � .... Q.) ..... ..... � E � .... ""0 ..... 0 0 -= en 50 • • 45 , " . "" v. �. A � /. " � I 40 • EPR Y = 17.5538 + 32.7495 (1 - exp (-0.5873 X) R2= 0.95 -___ - TSP Y = -27.8329 + 73.9101 (1 - exp (-1.2591 X» R2= 0.90 - - - Pool data Y = 13.246 + 35.592 (1 - exp (-0.697 X) R2 = 0.81 35 ��--��--��--���--��--��--�� 1 .5 2 .0 2 .5 3 .0 3 .5 4 .0 4 .5 5 .0 Malic acid extractable P (Ilg g-lsoil) Figure 7.23 Relationship between shoot dry matter yield and Malic acid extractable P for tea plants treated with EPR and TSP fertil i ser 2 1 4 2 1 5 Table 7.S Correlation matrix for P extraction methods for all rates of (a) EPR and (b) TSP fertilisers (a) Resin Bray Borax Resin 1 Bray 0.957* 1 Borax 0.984** 0.965* 1 Citric-acid 0.98 1 * * 0.973 ** 0.984** Malic-acid 0.96 1 * 0.996*** 0.973 * * Olsen 0.987** 0.968** 0 .998*** (b) Resin 1 Bray 0.9 1 8* 1 Borax 0.936* 0.979** 1 Citric- acid 0.977** 0.977** 0.973 ** Malic-acid 0.979** 0.969** 0.976** Olsen 0.975 * * 0.955* 0.985* * '" Correlation coefficient significant at p = 0.05 * * Correlation coefficient significant at p = 0.0 1 Citric-acid 1 0.972** 0.999*** 1 0.996*** 0.98 1 ** *** Correlation coefficient significant at p = 0.00 1 Malic-acid Olsen 1 0.973 ** 1 1 0.987* * 1 Table 7.9 Correlation matrix for P extraction methods for rates 0, 1 0 and 20 kg P ha-1 of (a) EPR and (b) TSP fertilisers (a) 2 1 6 Resin Bray Borax Citric -acid Malic-acid Olsen Resin 1 Bray 0.998* * * 1 Borax 0.988* * 0.995* * * 1 Citric-acid 0.957* 0 .970* * 0.989* * 1 Malic-acid 0.995* * 0 .998* * * 0.998* * * 0.98 1 ** Olsen 0.998* * * 0.999* * * 0.994* * * 0.970* * (b) Resin 1 Bray 0.980* * Borax 0.998* * * 0.968* * 1 Citric-acid 0.989* * 0.998* * * 0.980* * 1 Malic-acid 0.998* * * 0 .970* * 0.999* * * 0 .98 1 * * Olsen 0.999* * * 0 .972* * 0.999* * * 0.983 * * * Correlation coefficient significant at p = 0.05 * * Correlation coefficient significant at p = 0 .0 1 *** Correlation coefficient significant at p = 0.00 1 1 0.998* * * 1 1 0 .999* * * Table 7.10 The critical levels of plant available-P required to be in the soil to obtain 95% of the maximum dry matter yield in tea plants Soil P-test R 2 values of the model Critical levels of soil P (Ilg g-l soil) EPR TSP EPR + TSP EPR TSP Pool data Resin 0.95 0.90 0.71 3 3 3 Olsen 0 .94 0 .86 0 .52 49 52 40 Borax 0 .93 0 .86 0 .86 1 1 8 1 3 Bray- l 0 .96 0.97 0.68 7 5 5 Citric-acid 0 .98 0 .91 0.64 6 3 2 Malic-acid 0 .94 0 .88 0.8 1 4 3 4 N ...... -....J 2 1 8 extractant used, but did not differ greatly between the two fertiliser treatments for any of the extractants. Generally P tests for diagnostic purposes are designed with several aims. An ideal test should be simple enough for routine application, extract a sufficient amount of P to be easily measurable, extract sufficient P to represent a significant portion of the potential plant uptake and it should not extract significant amounts of P that are not plant available (Fixen and Grove, 1 990). Additionally a good P extractant should be able to provide values within a reasonable range, when soils are treated with different P sources. Saggar et al. ( 1 992b) showed that the Olsen extractant was good for predicting P availability to pasture in soils treated with soluble P fertiliser, but not in soils treated with sparingly soluble P fertilisers, whereas resin-P was able to predict P availability regardless of the type of P fertilisers used. The R 2 values for the relationship between tea shoot dry matter yield and soil test values were higher for the borax (R2 = 0.86) and malic acid (R2 = 0.8 1 ) extractants, when the data for EPR and TSP are pooled and analysed (Table 7. 1 0). Therefore, for these two extractants a single calibration curve could be used for predicting P availability to tea in soils treated with EPR or TSP (Figure 7 .20 and Figure 7.23). For the other extractants two calibration curves are required depending on whether EPR or TSP was used (Figure 7. 1 8, 7. 1 9, 7.2 1 and 7 .22). The possibility to use a single calibration curve for both soluble and sparingly soluble P fertiliser forms in the borax test is an advantage for the analyst, because it has no influence on the tea growers preference of the form of P fertiliser used for the plantation and one calibration curve can be used when the history ofP fertiliser source is unknown. Between the borax and malic acid tests, which showed the highest R2 values for the combined data for EPR and TSP treated soils, the borax method could be considered as the most suitable test for routine P analysis, because in addition to giving the highest R2 values of all the extractants it provided reasonably high P values with an acceptable range to distinguish between sufficiency and deficiency of P levels in the soil. The critical borax P concentration of 1 3 Ilg P g-l soil obtained in this study agrees very well with the value of 1 0 - 1 5 Ilg P g-l soil reported by Jayman and 2 1 9 Sivasubramaniam ( 1 980) as an adequate P concentration for mature tea growth in the field. Figure 7.20 also shows that for maximum tea growth, the borax P concentration is about 1 5 Ilg P g01 soil. The Olsen test though needs two calibration curves depending on the P fertilisers used, gave very high critical soil P values and therefore it can be easily measured, so soils deficient in or with sufficient P can be distinguished. As PR is the only P fertiliser used in many tea estates and likely to be used in the future too the Olsen calibration curve for PR (Figure 7. 1 9) can be used to assess the P availability in these soils. Irrespective of the type of P fertiliser used, the soil P concentration required to achieve 95% of the maximum shoot dry matter yield was found to be at the very beginning of the response curve in this study because initial soil P levels were high due to previous P fertiliser applications. Therefore it was not possible to classify soil test values according to a wide range of P availability classes (low, medium and high etc.). It is recommended that further glasshouse and field trials be conducted on soils with low soil P concentrations to test the findings reported in this Chapter. 7.5 CONCLUSIONS The results of the glasshouse trial showed that application of TSP or EPR fertiliser at the lowest rate of 1 0 kg P ha01 tested is sufficient to obtain maximum tea yield in a previously well fertilised soil. The agronomic effectiveness of EPR was equal to or slightly better than that of TSP at the 5 and 1 0 month periods of the trial. This shows that even though EPR is considered as a non-reactive phosphate rock according to its citric acid solubility, it dissolves at a rate fast enough to supply the P needs of tea plants in the highly acidic and high P fixing Ultisols of Sri Lanka. Therefore the low­ cost locally available EPR can be recommended for use as a P fertiliser for tea in Sri Lanka. Transformation of P from EPR and TSP fertilisers into various soil P pools varied due to differences in their solubilities. The resin-P was higher in TSP treated soil at 5 2 2 0 months due to its greater solubility, but at 1 0 months EPR produced higher resin-P due to higher dissolution with time. In the presence of tea plants, at 5 months 52% of the P from EPR applied at the rate of 1 0 kg ha-1 dissolved compared to 75% P dissolution at 1 0 months. The NaOH-Pi (Fe + AI bound P) concentration was significantly higher in the TSP treated soil due to the rapid conversion of easily soluble P to P forms fixed by Fe and AI oxides. In EPR treated soils, the H2S04-Pi (Ca-P) concentration was greater due to the recovery of undissolved PR in the soil by this extractant. In the unplanted soil the NaOH-Po concentration increased with time due to increased microbial activity, but during the latter part of the trial the concentration of this fraction decreased in the presence of plants due to the mineralisation of Po into Pi and the subsequent uptake by plants. Soil P values obtained by all the soil tests were very highly correlated with tea yield, when the data for each of the two P fertilisers was considered seperately. When the data were combined, only borax and malic acid extractants showed high R2 values. Of these two tests, borax test gave sufficiently high P values that can be measured easily, making it simple to distinguish between sufficiency and deficiency levels. Therefore the borax test is recommended as the most suitable test to predict P availability for tea. The Olsen test, which gave the highest soil test values, also can be used in tea plantations where only one type of P fertiliser (PR in most cases) has been used previously. For wider extrapolation of the conclusions reached in this study further research is required using glasshouse and field trials with soils of lower P status than the soil used for this study. CHAPTER 8 SUMMARY AND CONCLUSIONS 8.1 AN OVERVIEW 2 2 1 The literature review (Chapter 2) indicated that the amelioration of soil phosphorus (P) deficiency is an important part of managing tea (Camellia sinensis L.) nutrition in the highly weathered acidic soils (Ultisols) of humid and sub-humid tropics. Evidence derived from the management of tea in Sri Lankan soils indicated that the P fertiliser requirement of tea plant is low and in most instances in soils with relatively low inputs ofP at between 10 - 1 5 kg P ha"l yr"l there was no response to further applications of P. Considering the low P requirement in previously fertilised soils the use of locally available low-cost phosphate rock (PR) is an attractive source of P for tea compared to the expensive soluble P fertilisers because PRs are soluble in tea growing soils which have pHs <5.5 and rainfall >2000 mm. In Sri Lanka, a locally mined PR (Eppawala phosphate rock, EPR) is used to meet the plant's P requirements for most perennial crops including tea. The chemical characteristics of PRs and the conditions required for their dissolution were adequately discussed. Despite the favourable conditions available for the dissolution of EPR, the high acidity in these soils can cause applied fertiliser P to get converted into forms (Fe and AI bound P) that are more stable and relatively insoluble. These P forms have been traditionally considered to be not easily available to plants. Amidst such complexities, it is striking to note that tea grows well with low P fertiliser inputs of both soluble and insoluble forms without showing any visual symptoms of P deficiency under field conditions. Literature speculates that the ability to use sparingly soluble P forms results from the secretion of significant amounts of organic acids (citric and malic acids) by tea roots. These have been reported to be 2 2 2 capable of dissolving PRs in soils in the vicinity of roots. Additionally these organic anions were reported to chelate Fe and AI and release P that are bound to them. A large volume of literature discussing the differences in chemistry and biology of the rhizosphere of many plants compared to that of the bulk soil is now available. Little is known however, on the chemical changes and the mechanisms of P uptake in the rhizosphere of tree crops including tea plants. This is mainly due to a lack of dependable techniques for sampling soils in the rhizosphere zone. The literature on the techniques currently used to study rhizosphere processes was reviewed. Much information is now available on the factors influencing the dissolution of PR in agricultural soils. However the information available on tea plants on this subject is very scanty. The lack of information on the rate of PR dissolution in tea soils, especially in the rhizosphere zone, hampers accurate recommendations of PR fertiliser additions for tea plantations without polluting soil and water bodies by indiscriminate use. The first objective of this thesis was to develop a suitable technique to investigate rhizosphere processes in tree crops under glasshouse and field conditions in order to study root processes involved in the P nutrition of tea plants. This technique was subsequently used to study the rhizosphere processes in tea and other crops with diverse growth habits to test the differences in the mechanisms involved in P uptake. The genetic variability in tea clones and the effect of the form of applied nitrogen (N03- or NJL+) on P utilisation from EPR by tea plants were also investigated. Finally the agronomic effectiveness of EPR was compared with that of triple superphosphate in a glasshouse trial with tea plants. The relative suitability of soil tests in predicting P availability to tea plants in soils treated with these fertilisers were also determined. 8.2 A TECHNIQUE FOR STUDYING RHIZOSPHERE PROCESSES IN TREE CROPS 2 2 3 A modification of the root study container (RSC) technique ofKuchenbuch and Jungk ( 1982) in combination with a sequential soil P fractionation technique was used to study rhizosphere processes of P depletion around fine roots of Camellia (Camellia japonica L.) plants which are of the same family as tea (Camellia sinensis L.). The trials were carried out under glasshouse and field conditions using a top soil sample of Dystric Eutrochrept from New Zealand treated with North Carolina phosphate rock (NCPR), monocalcium phosphate (MCP), single superphosphate (SSP) and diammonium phosphate (DAP). Both glasshouse and field trials gave similar results and provided useful information on the rhizosphere processes involved in P utilisation by camellia seedlings and mature trees. The chemical properties in camellia rhizosphere differed significantly from that of the bulk soil. Camellia roots induced acidification in their rhizosphere. Plant induced acidification in the rhizosphere created conditions conducive for the dissolution of the sparingly soluble NCPR fertiliser. The chemical fractionation of soil P showed that the plants depleted the resin-P and NaOH-Pj fractions from the rhizosphere while accumulating NaOH-P o. The accumulation of Po indicated a transformation of more soluble forms ofPj into Po due to the high microbial activity in the rhizosphere. The RSC technique proved to be a viable aid to study the rhizosphere process in tree crops in the glasshouse as well as in the field. 8.3 PHOSPHORUS CHEMISTRY IN THE RHIZOSPHERE OF TEA AND ASSOCIATED CROPS A glasshouse study was carried out on a top soil sample from a Rhodustult in Sri Lanka using the modified RSC technique to understand how tea (clone TRI 2025) plants differ in their ability to utilise P from EPR compared to other plant species grown in the same locality. Calliandra (Calliandra calothyrsus L.), a leguminous tree 2 2 4 grown in tea fields to provide shade and organic material, Guinea grass (Panicum maximum L.) found abundantly in sloping tea lands which helps control soil erosion, and bean (Phaseolus vulgaris L.) a common leguminous vegetable grown in the same soil, were used in this study. All plant species acidified their rhizospheres. The magnitude of acidification differed among the plant species in the order of Guinea grass > bean and tea > calliandra. The highest acidification found in Guinea grass was due to the largest root surface area of this plant among the four plant species. Guinea grass however represented the lowest rate of acidification per unit surface area, whereas tea produced the highest rate of acidification per unit surface area. The rate of EPR dissolution in the rhizosphere followed the same order as that of the acidification. All plant species depleted resin-P, NaOH-P; in the rhizosphere. Except for tea all other species depleted NaOH-P 0 in the rhizosphere. Tea plants like camellia plants accumulated NaOH-P 0 in the rhizosphere probably due to increased microbial activity caused by a large supply of carbon exudates from tea roots. Guinea grass and bean were found to be externally more P efficient to extract P from soil compared to the other two species because they had a larger root surface area. The internal efficiency which describes the plant's ability to convert absorbed P into dry matter was in the order of bean > Guinea grass > calliandra > tea. 8.4 CLONAL VARIABILITY IN P UTILISATION Another glasshouse study was conducted on the same soil type as the one used in the first glasshouse trial in Sri Lanka to investigate the effects of clonal differences in tea on the utilisation of P from sparingly soluble EPR and soluble triple superphosphate (TSP) P fertilisers. The tea clones used were S 1 06, TRI 2023 and TRI 2025 . TRI 2023 and TRI 2025 produced significantly higher dry matter yield and P uptake than S 1 06 for both P treatments but there was no significant difference in P uptake by any of the clones between the two P fertilisers. The external P efficiency of TRI 2023 2 2 5 and TRI 2025 was higher than S 1 06 mainly due to greater root surface area and greater P uptake per unit surface area. The reason for the higher external efficiency in TRI 2023 compared to TRI 2025 was due to higher P uptake per unit surface area and not due to higher root surface area. The higher P uptake per unit surface area may be associated with higher root acidification and root exudation of organic compounds. All tea clones acidified their rhizospheres compared to the bulk soil in both P fertiliser treatments. The rhizosphere pH decrease among the clones was in the order of TRI 2023 > TRI 2025 > S 1 06. As in the previous glasshouse study more EPR dissolved in the rhizosphere compared to the bulk soil and the amount of EPR dissolution in the rhizosphere was related to pH decrease. All three tea clones depleted resin-P and NaOH-Pj but accumulated NaOH-Po in the rhizosphere as observed in the previous glasshouse study. The results of this study showed that the RSC technique could be successfully used to provide information on the preliminary screening of clones for their P utilisation efficiencies. 8.5 EFFECT OF FORMS OF NITROGEN SUPPLY ON MOBILISA TION OF P FROM EPR Nitrogen is the major nutrient input to tea plantations because it is constantly removed through regular harvesting of young tea shoots at 4 - 10 day intervals. The form of N supply could have a profound effect on the uptake of other nutrients notably P through its influence on rhizosphere pH. A glasshouse study similar to the one described in section 8 .3 was conducted to test the effect of N forms (NH/, N03- or both) on the availability of P from EPR applied to tea clone TRI 2025 . Ammonium sulphate, Ca(N03)2 and Nl4N03 were added to the soil to supply Nl4 +, N03- and Nl4+ + N03- respectively. The rhizosphere pH decreased compared to the bulk soil when N was supplied as NH/ or Nl4+ + N03- forms, and it increased when N was supplied as N03- form. The 2 2 6 estimations of cation-anion charge balance in the plants showed that the plants had taken-up more N03- than NIL. + even in the (Nl4hS04 treated soil. This was explained as due to high nitrification in the soils in spite of the presence of a nitrification inhibitor. The rhizosphere acidification in soils treated with (Nl4)2S04 was considered to be mainly due to nitrification. The dissolution of EPR in the rhizosphere increased regardless of the N forms. The (NH4)2S04 treatment caused highest dissolution of EPR in the rhizosphere whereas the Ca(N03)2 treatment showed the lowest. The degree of P dissolution was in agreement with the degree of acidification in the rhizosphere as observed in previous studies. The dissolution of EPR in the rhizosphere of tea treated with Ca(N03)2 was higher than in the bulk soil even though the rhizosphere pHs were higher. This is probably due to the plant uptake of Ca and P, the reaction products of EPR dissolution. Sequential fractionation of soil P indicated that the (Nl4hS04 treatment caused the highest depletion of resin-P but lowest depletion ofNaOH-P; probably due to the fixation of P by the soil at the low pHs in the rhizosphere. The concentrations of resin-P and NaOH-P; were lower and that of NaOH-P 0 was higher in the rhizosphere than that of the bulk soil as observed in the earlier studies. 8.6 AGRONOMIC EFFECTIVENESS OF EPR ON TEA Another glasshouse trial was conducted in Sri Lanka on the same soil type as in section 8 .3 to compare the agronomic effectiveness of EPR with TSP on tea. The results showed that TSP or EPR fertiliser at a rate as low as 1 0 kg P ha-1 was sufficient to obtain maximum tea yield. The agronomic effectiveness of EPR was equal to or slightly better than that of TSP at the 5 and 1 0 month samplings of the trial. This is due to the high rate of EPR dissolution in the highly acidic and high P fixing Ultisols which supply adequate P to meet the P needs of the tea plants. The results showed that the low-cost locally available EPR can be used profitably as a P fertiliser for tea plantations in Sri Lanka. 2 2 7 The composition of the soil P fractions varied according to the type of P fertiliser, reaction time in soil and the presence or absence of plants in the soil. The amount of P dissolved from EPR and the amount of TSP converted into various soil P fractions varied due to differences in their solubilities as observed in the other glasshouse trials. The resin-P was higher in the TSP treated soil at 5 months due to its greater solubility but at 1 0 months the EPR produced higher resin-P due to its increased dissolution over time. In the presence of tea plants, 52% of P from the EPR applied at the rate of 1 0 kg ha-l was dissolved at 5 months compared to 75% of dissolution at the 10 month sampling. The NaOH-Pi (loosely characterising Fe + AI bound P) concentration was significantly higher in the TSP treated soil due to rapid conversion of easily soluble P to P forms fixed to Fe and AI oxides. In the EPR treated soils, H2S04-Pi (Ca-P) concentration was greater due to the recovery of undissolved EPR. Tea yields were correlated against P extracted by various soil P tests. The Olsen test extracted the highest quantity of P from tea soils (3 5 to 100 �g P g-l soil) compared to all other extract ants tested (borax, Bray- I , citric acid and malic acid; extractable P values ranged from 2 to 30 �g P g-l soil). The Olsen test extracted both Fe and AI bound P, resin-P and also some labile organic P whereas the other extractants removed resin-P and mainly Al-P from these soils which are rich in Fe-P. Soil P values obtained by all soil tests were very highly correlated with tea yield when the data for each of the P fertilisers was considered separately. All tests except borax and malic acid extractions required two calibration curves depending on whether EPR or TSP fertiliser was used. Therefore they cannot be used to predict soil P availability to tea in estates where both types of fertiliser have been used. Between borax and malic acid, the former gave sufficiently high P values which could be measured easily and distinguished between the sufficiency and deficiency levels. Therefore borax test is recommended as the most suitable soil test to predict P availability to tea. Olsen test can also be used in tea plantations where only one type of P fertiliser (PR in most cases) has been used previously. 8.7 FUTURE RESEARCH AND RECOMMENDATIONS 2 2 8 The clonal evaluation made in the utilisation of P from EPR and native soils which indicated that TRI 2023 and TRI 2025 had high P utilisation efficiency. This needs further testing in the field in different agro-ecological regions and soil types. It is proposed that plant breeders include P utilisation efficiency trait in the selection of tea clones for different regions of the country to obtain high tea yields with low P inputs. In all trials, Po accumulated in tea and camellia rhizosphere but it was depleted in the rhizosphere of other crops tested. The accumulation ofP 0 was explained as due to the immobilisation of Pi by the highly active microbial population in the tea rhizosphere. The depletion of Po in other crops was explained to be due to the hydrolysis of Po by the increase concentration of phosphatase enzyme in rhizosphere compared to the bulk soil. The reasons for the differences between tea and other crops in the mobilisation and immobilisation of Pi in rhizosphere needs further investigation. The nitrification inhibitor used in the rhizosphere study testing the effect of different forms of N on rhizosphere acidification was not completely effective. This caused difficulties in the estimation of the ratios of uptake of NH/ and N03- by the tea plants. To avoid such difficulties, in future studies, it is recommended that the nitrification inhibitor be split applied to the soil several times during the period of investigation. Also it is appropriate to periodically monitor the concentrations of NIL + and N03- in the soil to determine the degree of nitrification, if any during the trial. Perhaps using 1 5N labelled � + and N03- fertilisers in the presence of a nitrification inhibitor would be more appropriate. Tea plants are known to secrete significant quantities of organic acids from their roots. The role of organic acids in rhizosphere acidification and PR dissolution is now well documented for many crops. However no literature is available on the relative contribution of organic acid exudation by roots and proton release caused by the plant' s uneven uptake of cation and anions towards rhizosphere acidification. Further studies on this aspect are required. 2 2 9 The glasshouse study reported on Chapter 7 showed that on a high P status soil EPR is as effective as TSP in supplying P to young tea plants confinning the anecdotal evidence of planters alreadey using EPR. In low P status soils the value of EPR may need further testing. The conclusion made in this study also needs further testing in the field in different agro-ecological regions of the country. These trials could also be used to further examine the suitability of borax soil test in predicting soil P availability to tea plants. REFERENCES Adams M A and Pate J S 1992 Availability of organic and inorganic fonns of phosphorus to lupins (lupinus spp). Plant and Soil 145, 1 07- 1 1 3 . Adriano D C and Doner H E 1982 Bromine, chlorine and fluorine. In Methods of Soil Analysis Part 2. Chemical and Microbiological Properties (Eds. A L Page, R H Miller and D R Keeney) 2nd edition, American Society of Agronomy, Madison, Wisconsin, USA. 459-460. Aguilar S and Van Diest A 198 1 Rock phosphate mobilisation induced by the alkaline uptake pattern of legumes utilising symbolically fixed nitrogen. Plant and Soil 61, 27-42. Alam A F M B 1994 Influence of improved genetic material towards higher productivity of tea in Bangladesh. In Proceedings of the international seminar on "Integrated Crop Management in Tea: Towards Higher Productivity", International Potash Institute, Basel, Switzerland. 33-49. 2 3 0 Alston A M and Chin K W 1974 Response of subterranean clover to rock phosphates as affected by particle size and depth of mixing in the soil. Australian Journal of Experimental Agriculture and Animal Husbandry 14, 649-655. Anandappa T I 1986 Planting material. In Handbook on Tea. (Eds. P Sivapalan, S Kulasegaram and A Kathirvetpillai) Tea Research Institute of Sri Lanka, Talawakelle. 6-9. Anderson G 1 980 Assessing organic phosphorus in soils. In The Role of Phosphorus in Agriculture. (Eds. F E Khasawneh, E C Sample and E J Kamprath) ASA­ SSSA-CSSA, Madison, USA. 4 1 1 -430. 2 3 1 Anderson D L, Kussow W R, Corey R B 1985 Phosphate rock dissolution in soils : indication from plant growth studies. Soil Science Society of America Journal 49, 9 18-925. Anon 1 989 Dolomite application for tea soils. In TRI Circular Serial No. F-12, Serial No. 3/89, Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. Anon 1 990 The planting and management of shade trees in tea. In Circular No. 0-1, Serial No. 5/90. Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. Anon 1994a Clones recommended for the up-country districts. In TRI Circular No. C-2, Serial No. 15/94. Tea Research Institute 'of Sri Lanka, Talawakelle, Sri Lanka. Anon 1 994b Clones recommended for the low-country districts. In TRI Circular No. C-3, Serial No. 16/94, Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. Anon 1 994c Clones recommended for the mid-country wet zone. In TRl Circular No. C-4, Serial No. 1 7/94, Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. Anon 1 994d Clones recommended for the mid-country semi-dry zone. In TRI Circular No. C-5, Serial No. 18/94, Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. Apthorp J N, Hedley M J and Tillman R W 1987 The effect of nitrogen fertiliser form on the plant availability of phosphate from soil, phosphate rock and monocalcium phosphate. Fertiliser Research 12, 269-284. Armiger W H and Fried M 1957 The plant availability of various sources of phosphate rocks. Soil Science Society of America Proceedings 21, 1 83- 188 . 2 3 2 Annstrong R D and Helyar K R 1 992 Changes in soil phosphate fractions in the rhizosphere of semi-arid pasture grasses. Australian Journal of Soil Research 30, 1 3 1 - 1 43 . Astika W 1 994 Plant improvement and field practices towards improved productivity of tea in Indonesia. In Proceedings of the International Seminar on "Integrated Crop Management in Tea: Towards Higher Productivity" International Potash Institute, Basel, Switzerland. 6 1 -66. Ayadurai S and Sivasubramaniam S 1977 The influence of types of phosphatic fertilisers on the growth and nutrient uptake of tea nursery plants. Tea Quarterly 47, 23-27. Barber S A 1 995 Soil Nutrient Bioavailability. 2nd edition John Wiley and Sons, New York. Barbora B C 1 994 Practices adopted in North-east India to enhance tea productivity. In Proceedings of the International Seminar on "Integrated Crop Management in Tea: Towards Higher Productivity", International Potash Institute, Basel, Switzerland. 67-68. Barea J M 1 991 Vesicular -arbuscular mycorrhizae as modifiers of soil fertility. Advances in Soil Science 15, 1 -39. Barrow N J 1984 Modelling the effects of pH on phosphate sorption by soils. Journal of Soil Science 35, 283-297. Barrow N J 1985 Comparing the effectiveness of fertilisers. Fertiliser Research 8, 85- 90. Barrow N J 1 990 Relating chemical processes to management systems. In Phosphorus Requirements for Sustainable Agriculture in Asia and Oceania. Symposium Proceedings, International Rice Research Institute, Philippines. 200-209. Barrow N J and Shaw T C 1977 Factors affecting the amount of phosphate extracted from soil by anion exchange resin. Geoderma 18, 309-323 . Barthakur B K, Dutta P and Begum R 1 987 Scope of V AM application in tea plantation. Two and a Bud 34, 1 7-20. Batten G D 1 992 A review of phosphorus efficiency in wheat. Plant and Soil 146, 1 63- 168 . Beater B E 1 949 A rapid method for obtaining readily soluble phosphate and phosphate fixation in soils. Plant and Soil 1, 2 1 5-220. 2 3 3 Bekele T, Clino B J, Elhert P A I, Van der Mass A A and Van Diest A 1983 An evaluation of plant borne factors promoting the solubilisation of alkaline rock phosphates. Plant and Soil 75, 361 -378. Bennett D and Ozanne P G 1972 Deciding how much superphosphate to use. In CSIRO Division of Plant Industry Annual Report. 45-47. Bhattacharyya N G and Dey S K 1 978 Phosphate in tea soils and changes on long term manuring. Two and a Bud 25, 32-36. Bhattacharyya N G and Dey S K 1 983 Role of pH and aluminium on phosphate availability of tea soils. Two and a Bud 30, 6 1 -64. Bhavanandan V P and Manipura W B 1969 Fertiliser responses of tea in the up­ country districts. Tea Quarterly 40, 135- 144. Binkley D 1 986 Forest Nutrition Management. John Wiley and Sons, New York. 2 34 Biswas A K, Sarkar S K and Biswas A K 1984 Manuring syndrome for mature tea : a long term study. Two and a Bud 31, 22-27. Blackmore L C, Searle P L and Daly B K 1987 Methods of Chemical Analysis of Soils. NZ Soil Bureau, Scientific Report 80, NZ Soil Bureau, Lower Hutt. Blofeld J 1 985 The Chinese Art of Tea. George Allen and Unwin, London. Bodhidasa H D T 1989 Welcome address. Journal of Tea Science (Conference Issue) 58, 3 . Bogahawatta C and Herath M A L B 1 984 Pricing and technical efficiencies of smallholder green tea production. Evidence from Kotmale area in the Nuwara Eliya district. Sri Lankan. Journal of Agrarian Studies 5, 3 7-5 1 . Bolan N S 1 99 1 A critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. Plant and Soil 134, 1 89-207. Bolan N S, Elliott J, Gregg P E H and Wei I S 1997 Enhanced dissolution of phosphate rocks in the rhizosphere. Biology and Fertility of Soils 24, 1 69- 1 74 . Bolan N S and Hedley M J 1 989 Dissolution of phosphate rocks in soil. 1 . Evaluation of extraction methods for the measurement of phosphate rock dissolution. Fertiliser Research 19, 65-75. Bolan N S and Hedley M J 1990 Dissolution of phosphate rocks in soils. 2. Effect of pH on the dissolution and plant availability of phosphate rocks in soil with pH dependent charge. Fertiliser Research 24, 125-1 34. Bolan N S and Hedley M J 1 995 Slow release phosphate fertilisers for sustainable production, Fertiliser Industry Annual Review 18, 1 79- 196. 2 3 5 Bolan N S, Hedley M J and Loganathan P 1993 Preparation, forms and properties of controlled release phosphate fertilisers. Fertiliser Research 35, 13 -24. Bolan N S, Syers J K and Tillman R W 1986 Ionic strength effects on surface charge and adsorption of phosphate and sulphate by soils. Journal of Soil Science 37, 379-388. Bolan N S, White R E and Hedley M J 1 990 A review of the use of phosphate rocks as fertilisers for direct application in Australian and New Zealand. Australian Journal of Experimental Agriculture 30, 297-3 1 3 . Bolan N S , White R E and Hedley M J 199 1 Processes of soil acidification during nitrogen cycling with emphasis on legume based pastures. Plant and Soil 134, 53-63 . Bonheure D and Willson K C 1992 Mineral nutrition and fertilisers. In Tea Cultivation to Consumption. (Eds. K C Willson and M N Clifford) Chapman and Hall, London, UK. 269-32 1 . Bowman R A and Cole C V 1978 Transformations of organic phosphorus substrates in soils as evaluated by NaHC03 extraction. Soil Science 123, 49-54. Bray R H and Kurtz L T 1945 Determination of total, organic and available forms of phosphorus in soils. Soil Science 59, 39-45 . Breteler H 1973 A comparison between ammonium and nitrate nutrition of young sugar-beet plants grown in nutrient solutions at constant acidity. I. Production of dry matter, ionic balance and chemical composition. Netherlands Journal of Agricultural Science 21, 227-244. Brown D A and Vl-Haq A 1984 A porous membrane-root culture technique for growing plants under controlled soil conditions. Soil Science Society of America Journal 48, 692-695 . Butegwa C N, Mullins G L and Chien S H 1996 Induced phosphorus fixation and the effectiveness of phosphate fertilisers derived from Sukulu Hills phosphate rock. Fertiliser Research 44, 23 1 -240. Buttlers B and Chenery E M 1 959 A rapid method for determination of total sulphur in soils and plants. The Analyst 84, 234-245. Campbell N A and Keay J 1970 Flexible techniques in describing mathematically a range of response curves of pasture species. Proceedings of the 11th International Grassland Congress, 332-334. 2 36 Caro J H and Hill W L 1 956 Characteristics and fertiliser value of phosphate rock from different fields. Journal of Agricultural and Food Chemistry 4, 684-687. Carr M K V and Stephens W 1 992 Climate, weather and the yield of tea In Tea Cultivation to Consumption. (Eds. K C Willson and M N Clifford) Chapman and Hall, London, UK. 87- 135 . Causton D R and Venus J C 1981 The Biometry of Plant Growth. E Arnold Ltd. Publishers. London. Chang S C and Jackson M L 1957 Fractionation of soil phosphorus. Soil Science 84, 133- 144. Chenery E M 1955 A preliminary study of aluminium and the tea bush. Plant and Soil 6, 1 74-200. Chien S H 1977 Thermodynamic considerations on the solubility of phosphate rock. Soil Science 123, 1 1 7- 1 2 1 . Chien S H and Hammond L L 1978a A comparison of various laboratory methods for predicting the agronomic potential of phosphate rocks for direct application. Soil Science Society of America Journal 42, 935-939. 2 3 7 Chien S H and Hammond L L 1978b Long-term reactions of phosphate rocks with an oxisol in Columbia. Soil Science 144, 257-265. Chien S H and Hammond L L 1989 Agronomic effectiveness of partially acidulated phosphate rock as affected by soil phosphorus - fixing capacity. Plant and Soil 120, 1 59- 1 64. Chien S H, Leon L A and Tejeda H R 1980 Dissolution of North Carolina phosphate rock in acid Colombian soils as related to soil properties. Soil Science Society of America Journal 44, 1 267- 1270. Chien S H and Menon R G 1 995 Factor affecting the agronomic effectiveness of phosphate rock for direct application. Fertiliser Research 41, 227-234. Chien S H, Sale PW G and Friesen D K A 1 990a Discussion of the methods for comparing the relative effectiveness of phosphate fertilisers varying in solubility. Fertiliser Research 24, 149- 1 57. Chien S H, Sale P W G and Hammond L L 1990b Comparison of the effectiveness of phosphate fertiliser products. In Phosphorus requirements for sustainable agriculture in Asia and Oceania, Symposium Proceedings. International Rice Research Institute, The Philippines. 143- 1 56. Child R 1 953 The Selection of Soils Suitable for Tea. Pamphlet No. 5, Tea Research Institute of East Africa, Kericho, Kenya. Chu H M 1975 Studies on soil properties and their grading for tea land improvement. Taiwan Tea Experimental Station Bulletin 67, 35-66. Chu C R, Moschler W M and Thomas G W 1962 Rock phosphate transformations in acid soils. Soil Science Society of America Journal 26, 476-478. 2 3 8 Clemensson-Lindell A and Persson H 1 992 Effect of freezing on rhizosphere and root nutrient content using two sampling methods. Plant and Soil 139, 39-45. Coleman N J, Thorup J T and Jackson W A 1 960 Phosphate sorption reactions that involve exchangeable aluminium. Soil Science 90, 1 -7. Colwell J D 1 963 The estimation of the phosphorus fertiliser requirement of wheat in Southern New South Wales by soil analysis. Australian Journal of Experimental Agriculture and Animal Husbandry 3, 1 90-197. Curl E A and Truelove B 1986 The Rhizosphere, Springer-Verlag, Berlin, Germany Dahanayake K, Ratnayake M P K and Sunil P A 1995 Potential ofEppawala apatite as a directly applied low-cost fertiliser for rice production in Sri Lanka. Fertiliser Research 41, 145-1 50. Darrah P R 1993 The rhizosphere and plant nutrition: a quantitative approach. Plant and Soil 155/156, 1 -20. De A1wis K A and Panabokke C R 1 972 Hand Book of the Soils of Sri Lanka (Ceylon). Journal of Soil Science Society of Ceylon 2, 83-85. De Swart P J and Van Diest A 1 987 The rock phosphate solubilising capacity of Pueraria javanica as affected by soil pH, superphosphate priming effect and symbiotic N2fixation. Plant and Soil 100, 1 35-147. Dey S K 1969 A study of some soils from Assam, Central and East African tea areas. Two and a Bud 16, 65-69. Dey S K 1972 Tea in Russia. Two and a Bud 19, 73-84. Dey S K and Bhattacharyya N G 1980 Studies on soil phosphate. Two and a Bud 27, 2 1 -23 . 2 3 9 Diest J, Marais P G, Harry R B A and Heyns C F G 1 97 1 Relative availability of rock phosphate to different plant species. Agrochemophysica 3, 35-40. Dijkshoom W and Van Wijk A L 1 967 The sulphur requirements of plants as evidenced by the sulphur-nitrogen ratio in the organic matter: a review of published data. Plant and Soil 26, 129- 1 57. Dinkelaker B and Marschner H 1992 In vivo demonstration of phosphatase activity in the rhizosphere of soil-grown plants. Plant and Soil 144, 1 99-205 . Dormaar J F 1 988 Effect of plant roots on chemical and bio-chemical properties of sorrounding discrete soil zones. Canadian Journal of Soil Science 68, 233- 242. Earl K D, Syers J K and McLaughlin J R 1 979 Origin of the effects of citrate, tartrate and acetate on phosphate sorption by soils and synthetic gels. Soil Science Society of America Journal 43, 674-678. Eden T 1 934 The manuring of tea. Tea Quarterly 7, 6 1 -74. Eden T 1949 Monograph on Tea Production in Ceylon (No 1 ) . Tea Research Institute of Ceylon, Talawakelle, Sri Lanka. Eden T 1976 Tea. 3rd edition, Longman, London. Eivazi F and Tabatabai M A 1977 Phosphatases in soil. Soil Biology and Biochemistry 9, 167- 1 72. Ellis R, Quader M A and Truog E 1955 Rock phosphate availability as influenced by soil pH. Soil Science Society of America Proceedings 19, 484-487. 2 4 0 EI Swaify S A, Pathak P, Rego T J and Singh S 1 985 Soil management for optimised productivity under rain-fed conditions in the semiarid tropics. Advances in Soil Science 1 , 1 -64. Fernando L H, Bhavanandan V P, Wettasinghe D T and Manipura W B 1 969 Fertiliser recommendations for tea in Ceylon - 1969. Tea Quarterly 40, 1 29- 144. Fixen P E and Grove J H 1990 Testing soil phosphorus. In Soil Testing and Plant Analysis. (Ed. R L Westerman) SSSA book series: 3 , Wisconsin, USA. 14 1 - 1 80. Flach E N, Quak W and Van Diest A 1 987 A comparison of rock phosphate­ mobilising capacities of various crop species. Tropical Agriculture (Trinidad) 64, 347-352. Fohse D, Claassen N and Jungk A 1 988 Phosphorus efficiency of plants. I. External and internal P requirement and P uptake efficiency of different plant species. Plant and Soil 110, 10 1 - 109. Fohse D, Claassen N and Jungk A 199 1 Phosphorus efficiency of plants. II. Significance of root radius, root hairs and cation anion balance for phosphorus influx in seven plant species. Plant and Sot1 123, 26 1 -272. Fohse D and Jungk A 1 983 Influence of phosphate and nitrate supply on the root hair formation of rape, spinach and tomato plants. Plant and Soil 74, 359-368. Foster N W and Morrison I K 1 976 Distribution and cycling of nutrients in a natural Pinus banksiana ecosystem. Ecology 57, 1 10- 120. Fox R L 1979 Comparative responses offield grown crops to phosphate concentrations in soil solutions. In Stress Physiology in Crop Plants (Eds. H Massele and R Staples) John Wiley and Sons, New York. 82- 1 06. 2 4 1 Fox R L, Nishmoto R K, Thompson J R and de la Pena R S 1 974 Comparative external P requirements of plants growing in tropical soils. Transactions IOth International Congress on Soil SCience, Moscow 4, 232-239. Fox T, Comerford N and McFee W 1990 Phosphorus and aluminium release from a spodic horizon mediated by organic acids. Soil Science Society of America Journal 54, 1 763-1 767. Foy C D, Chancy R L and White M C 1978 The physiology of metal toxicity in plants. Annual Review. Plant Physiology 29, 5 1 1 -566. Fried M 1953 The feeding power of plants for phosphate. Soil Science Society of America Journal 17, 357-359. Gahoonia T S, Claassen N and Jungk A 1 992 Mobilisation of phosphate in different soils by ryegrass supplied with ammonium or nitrate. Plant and Soil 140, 24 1 - 248 . Gahoonia T S and Nielsen N E 1 99 1 A method to study rhizosphere processes in thin soil layers of different proximity to roots. Plant and Soil 135, 1 43- 146. Gahoonia T S and Nielsen N E 1 992 The effects of root-induced pH changes on the depletion of inorganic and organic phosphorus in the rhizosphere. Plant and Soil 143, 1 85- 1 9 1 . Gahoonia T S and Nielsen N E 1996 Variation in acquisition of soil phosphorus among wheat and barley genotypes. Plant and Soil 178, 223-230. Gahoonia T S, Raza S and Nielsen N E 1 994 Phosphorus depletion in the rhizosphere as influenced by soil moisture. Plant and Soil 159, 23 1-2 1 8 . Ganguly D K 1 993 Tea in health - an overview. Proceedings of International Symposium on Tea Technology, Calcutta, India. 1 07- 1 08. 2 4 2 Gardner W K, Parbery D G and Barber D A 1982 The acquisition of phosphorus by Lupinus albus L. II. The effect of varying phosphorus supply and soil type on some characteristics of the soiVroot interface. Plant and Soil 68, 33-4 1 . Gattani P D and Seth S P 1 973 Phosphorus soil test correlation studies in Rajasthan. Journal of Indian Soil Science 21 , 375-378. Gerke J and Jungk A 199 1 Separation of phosphorus bound to organic matrices from inorganic phosphorus in alkaline soil extracts by ultrafiltration. Communications in Soil Science and Plant Analysis 22, 162 1 - 1630. Gijsman A J 1 990a Rhizosphere pH along different root zones of Douglas-fir (Pseudotsuga menziesii), as affected by source of nitrogen. Plant and Soil 124, 1 6 1 - 1 67. Gijsman A J 1 990b Nitrogen nutrition of Douglas-fir (Pseudotsuga menzies;;) on strongly acid sandy soil. I. Growth, nutrient uptake and ionic balance. Plant and Soil 126, 53-6 1 . Gijsman A J 1 990c Nitrogen nutrition and rhizosphere pH of Douglas-fir (Pseudotsuga menziesii) on strongly acid sandy soil. II. Proton excretion and rhizosphere pH. Plant and Soil 126, 63-70. Godwin D C and Blair G J 1 99 1 Phosphorus efficiency in pasture species. V. A comparison of white clover accessions. Australian Journal of Agricultural Research 42, 53 1 -540. Gogoi A K, Choudhury M N D and Gogoi N 1 993 Effect of phosphorus on the quality of made teas. Two and a Bud 40, 1 5-2 1 . Gohain K K 1988 Phosphorus supplying capacity of tea soils and its relation to P­ uptake by young tea (Camellia sinensis L.). Two and a Bud 35, 62-65. Golden D C, Weed S B, Sivasubramaniam and Nalliah P 198 1 Studies on the performance of two phosphatic fertilisers on tea soils with respect to the P­ adsorption, mineralogy and plant species using 32p labelled fertiliser. Tea Quarterly 50, 50-60. Golden D C, White R E, Tillman R W and Stewart R B 199 1 Partially acidulated reactive phosphate rock (P APR) fertiliser and its reactions in soil. I. Initial movement of dissolved ions and solubility of the phosphate rock residue. Fertiliser Research 28, 28 1 -293 . Green M J 1 965 Planting techniques. In Annual Report, Tea Research Institute of East Africa, Kericho, Kenya. 37. Gregg P E H, Mackay A D, Tillman R W and Currie L D 1 987 A summary of phosphate rock research at Massey University. In ' The Use of Reactive Phosphate Rocks and their Derivatives as Fertilisers ' Occasional Report No. 1. (Eds. R E White and L D Currie) Massey University, New Zealand. 1 1 5- 124. Grierson P 1993 Organic acids in the rhizosphere ofBanksia integrifolia. Plant and Soil 144, 259-265 . 2 4 3 Griroux M, Dube A and Barnett G M 1984 Effect de la fertilisation phosphatee sur la pomme de terre en relation a vec l' analyse du sol et la source de phosphore utilisee. Canadian Journal of Soil Science 64, 369-38 1 . Hammond L L, Chien S H and Easterwood G W 1 986a Agronomic effectiveness of Boyovar phosphate rock in soil with induced phosphorus retention. Soil Science Society of America Journal 50, 1 60 1 - 1 606. Hammond L L, Chien S H and Mokwunye A U 1986b Agronomic value of unacidulated and partially acidulated phosphate rocks indigenous to the tropics. Advances in Agronomy 40, 89- 140. 2 4 4 Hanway J J and Olson R A 1 980 Phosphate nutrition of corn, sorghum, soybeans, and small grains. In The Role of Phosphorus in Agriculture. (Eds. F E Khasawneh, E C Sample and E J Kamprath) SSSA Proceedings Madison, Wisconsin, USA. 68 1 -690. Harler C R 1 968 Tea soil acidity. World Crops 20, 60-62. Harler C R 197 1 Tea soils. World crops 23, 275 . Hasselo H N 1 965 The nitrogen, potassium, phosphorus, calcium, magnesium, iron, copper, boron, zinc, molybdenum and aluminium contents of tea leaves of increasing age. Tea Quarterly 36, 1 22- 1 36. Haussling M and Marschner H 1989 Organic and inorganic soil phosphatases and acid phosphate activity in the rhizosphere of 80-year-old Norway spruce [Picea abies (L.) Karst. ] trees. Biology and Fertility of Soils 8, 1 28- 1 3 3 . Haynes R J 1 983 Soil acidification induced by leguminous crops. Grass Forage Science 38, 1 - 1 1 . Haynes R J 1 990 Active ion uptake and maintenance of cation and anion balance: A critical examination of their role in regulating rhizosphere pH. Plant and Soil 26, 247-264. Haynes R J 1 992 Relative ability of a range of crop species to use phosphate rock and monoca1cium phosphate as P sources when grown in soil. Journal of Science Food and Agriculture 60, 205-2 1 1 . 2 4 5 Haynes R J and Goh K M 1978 Ammonium and nitrate nutrition of plants. Biological Reviews of the Cambridge Philosophical Society 53, 465-5 1 0. Hedley M J, Hussin A and Bolan N S 1 990 New approaches to phosphorus fertilisation. In Phosphorus Requirements for Sustainable Agriculture in Asia and Oceania. Symposium Proceedings. International Rice Research Institute, The Philippines. 125- 142. Hedley M J, Kirk G J D and Santos M B 1 994 Phosphorus efficiency and the forms of soil phosphorus utilised by upland rice cultivars. Plant and Soil 158, 53-62. Hedley M J, Nye P H and White R E 1982a Plant-induced changes in the rhizosphere of rape (Brassica napus var. Emerald) seedling II. Origin of the pH change. New Phytologist 91, 3 1 -44. Hedley M J, Stewart J W B and Chauhan B S 1 982b Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Science Society of America Journal 46, 970-976. Hedley M J, White R E and Nye P H 1982c Plant-induced changes in the rhizosphere of rape (Brassica napus var. Emerald) seedling III. Changes in L value, soil phosphate fractions and phosphatase activity. New Phytologist 91, 45-56. Helal H M and Sauerbeck D R 1984 Influence of plant roots on C and P metabolism in the soil. Plant and Soil 76, 1 75- 1 82. Helal H M and Sauerbeck D R 199 1 Soil and root phosphatase activity and the utilisation of inositol phosphates as dependent on phosphorus supply In Plant Roots and Their Environment (Eds. B L McMichael and H Persson) Elsevier Science Publishers B. v. , The Netherlands. 93-97. --- --- - ----- - ---- Helal H M and Sauerbeck D R 1984 Influence of plant roots on C and P metabolism in the soil. Plant and Soil 76, 1 75- 1 82. Hendricks M and Jungk A 1 98 1 Frfassung der Mineral-stoffsverteilung in Wurzelnahe durch getrennte Aalyze von Rhizound restboden. z. Pjlanzenernahe Boden 144, 276-282. 2 4 6 Hinsinger P 1994 The acquisition of mineral nutrients by roots; rhizosphere processes. In Proceedings of the Third Congress of the European Society for Agronomy. Padova, Italy. (Eds. M. Borin and M Sattin) European Society for Agronomy, Colmar. 428-437. Hinsinger P and Gilkes R J 1 995 Root-induced dissolution of phosphate rock in the rhizosphere of lupins grown in alkaline soil. Australian Journal of Soil Research 33, 477-489. Hinsinger P and Gilkes R J 1 996 Mobilisation of phosphate from phosphate rock and alumina-sorbed phosphate by the roots of ryegrass and clover as related to rhizosphere pH. European Journal of Soil Science 47, 533-544. Hoffiand E 1 992 Quantitative evaluation of the role of organic acid exudation in the mobilisation ofrock phosphate by rape. Plant and Soil 140, 279-289. Hoffiand E, Findenegg G R and Nelemans J A 1 989 Solubilisation of rock phosphate by rape. II Local root exudation of organic acids as a response to P starvation. Plant and SoU l t3, 1 6 1 - 1 65 . Hoffiand E, Vandenboogaard R, Nelemans J and Findenegg G 1 992 Biosynthesis and root exudation of citric and malic-acids in phosphate-starved rape plants. New Phytologist 122, 675-680. 2 4 7 Hossner L R, Freeouf J A and Folsom B L 1973 Solution phosphorus concentration and growth of rice (Oryza sativa L.) in flooded soils. Soil Science Society of America Proceedings 37, 405-408. Hughes J C and Gilkes R J 1 986 The effect of rock phosphate properties on the extent of fertiliser dissolution in soils. Australian Journal of Soil Research 24, 209-2 17. Humble R 1 99 1 Tea Area Changes in Sri Lanka - Analysis of Regional Distribution, Processes, Mechanisms, and Correlating Factors of Changes in the Area Cultivated with Tea Since 1956. Department of Geography, University of Zurich, Zurich. Ishigaki K 1978 Studies on the nutritive characteristics of tea plant. Bulletin of National Research Institute of Tea (Japan) 14, 30-1 52. Israel 0 W and W A Jackson 1 978 The influence of nitrogen nutrition on ion uptake and translocation by leguminous plants. In Mineral Nutrition of Legumes in Tropical and Subtropical Soils. (Eds. C C Andrew and E J Kamprath) CSIRO, Melbourne. 1 1 3- 1 29. Jackson M L 1958 Soil Chemical Analysis. Englewood Cliff, Prentice Hall Inc. Englewood Cliffs, New Jersey, USA. Jarvis S C and Robson A D 1 983 The effects of nitrogen nutrition of plants on the development of acidity in Western Australian soils. I. Effects with subterranean clover grown under leaching conditions. Australian Journal of Agricultural Research 34, 341 -353 . Jayakodi A 1996 Tea Census 1994, TRi Update, Publication of Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. 1 , 4 . 248 Jayawardena D E de S 1 976 The Eppawala Carbonatite Complex. Geological Survey Department Publication, Geological Survey Department. Colombo, Sri Lanka. Jayman T C Z and Sivasubramaniam S 1 974 The use of ascorbic acid to eliminate interference from iron in the aluminon method for determining aluminium in plant and soil extracts. The Analyst 99, 296-30 1 . Jayman T C Z and Sivasubramaniam S 1 975 Release of bound iron and aluminium from soil by the root exudates of tea (Camellia sinensis L.) plants. Journal of Science Food and Agriculture 26, 1 895-1 898 . Jayman T C Z and Sivasubramaniam S 1 980 The use of fertilisers for tea in Sri Lanka 3 - Foliar and soil analysis with particular reference to P, Ca and Mn. Tea Quarterly 49, 1 37- 14 1 . Jayman T C Z and Sivasubramaniam S 1981 The mineral status and physical composition of some cultivated and uncultivated tea soils of Sri Lanka. Tea Quarterly 50, 74-83 . Johnston W B and Olsen R A 1972 Dissolution offluoraptite by plant roots. Soil Science 1 14, 29-36. Jones D L and Darrah P R 1994 Role of root derived organic acids in the mobilisation of nutrients from the rhizosphere. Plant and Soil 166, 247-257. Jourbitzky Z I and Strausberg D V 1 966 Foliar diagnosis of the tea plant. Nitrogen and potassium status. Fertilite 26, 3-6. Jungk A a 1 996 Dynamics of nutrient movement at the soil-root intrerface. In Plant Roots The Hidden Half. (Eds. Waisel Y, Eshel A, Kafkafi U) 2nd edtion. Mercel Dekker Inc. , New York. 529-556. Jungk: A and Claassen N 1 989 Availability in soil and acquisition by plants as the basis for phosphorus and potassium supply to plants. Z. Pflanzenemahr. Bodenkd 1 52, 1 5 1 - 1 57. 2 4 9 Kanabo J A and Gilkes R J 1 987 The role of pH in the dissolution of phosphate rock fertilisers. Fertiliser Research 12, 1 65- 1 74 . Kanabo I A K and Gilkes R J 1 988 The effect of soil texture on the dissolution of North Carolina phosphate rock. Journal of Soil Science 39, 1 9 1 - 1 98. Karim Z and Rahman A 1 980 Cationic environment in some selected tea soils of Bangladesh. Tea Quarterly 49, 53-57. Karim Z, Rahman A and Rahman H 1 98 1 Mineralogy of acidic tea soils and sorption of phosphate. Tea Quarterly 50, 1 75-1 82. Kelegama J B, Gunaratne N W H D, Samaraweera D S A and Abeygoonawardena S K 1 995 Report of the Presidential Commission on the Tea Industry and Trade. Office of the Presidential Commission on the Tea Industry and Trade, Colombo 7, Sri Lanka. Khanna S S, Tomar N K and Gupta A P 1983 Efficiency of incubated phosphatic fertilisers varying in water solubility with organic matter to wheat. 3rd International Congress on Phosphorus Compounds. Brussels. 567-579. Khasawneh F E and Doll E C 1 978 The use of phosphate rock for direct application to soils. Advances in Agronomy 30, 1 59-206. Kirkby E A and Knight A H 1977 Influence of the level of nitrate nutrition on ion uptake and assimilation, organic acid accumulation and cation-anion balance in whole tomato plants. Plant Physiology 60, 349-353 . 2 50 Kirkby E A and Mengel K 1 967 Ionic balance in different tissues of the tomato plant in relation to nitrate, urea and ammonium nutrition. Plant Physiology 42, 6- 1 4. Kirk G I D and Nye P H 1 986 A simple model for predicting the rate of dissolution of sparingly soluble calcium phosphate in soil 1 . The basic model. Journal of Soil Science 37, 529-540. Kirlew P W and Bouldin D R 1987 Chemical properties of the rhizosphere in an acid sub-soil. Soil Science Society of America Journal Sl, 1 28- 1 32. Kucey R M N and Bole J B 1 984 Availability of phosphorus from 1 7 rock phosphates in moderately and weakly acidic soils as determined by 32p dilution. A value, and total P uptake methods. Soil Science 138, 1 80- 1 88 . Kuchenbuch R and Jungk A 1 982 A method determining concentration profiles at the soil-root interface by thin slicing rhizospheric soil. Plant and Soil 68, 391 -394. Lamb J 1 952 Tea - a note on manurial mixtures. Tea Quarterly 23, 87. Larsen S 1 967 Soil phosphorus. Advances in Agronomy 19, 1 5 1 -2 10. Le Mare P H 1 99 1 Rock phosphates in Agriculture. Experimental Agriculture 27, 4 1 3-422. Liu Z Y, Shi W M and Fan X 1990 Rhizosphere effects of phosphorus and iron in soils. Transactions 14th International Congress on Soil SCience, Kyoto 2, 1 47- 1 53 . Loganathan P 1 978 Some recent developments in fertilising coconut. Ceylon Cocon Quarterly 29, 87-96. 2 5 1 Loganathan P and Fernando T W 1 980 Phosphorus sorption by some coconut­ growing acid soils of Sri Lanka and its relationship to selected soil properties. Journal oj Science Food and Agriculture 31, 709-71 7. Loganathan P, Dayaratne P M N and Shanmuganathan R T 1982 Evaluation of the phosphorus status of coconut-growing soils in Sri Lanka. Journal oj Agricultural Science (Cambridge) 99, 25-33 . Loganathan P , Hedley M J and Saggar S 1 995 Liming effects of reactive phosphate rock: A laboratory evaluation. In Plant-Soil interactions at Low pH: Principles and Management. (Eds. R A Date, N J Grundon, G E Rayment and M E Probert) Kluwer Academic Publishers, Dordrecht, The Netherlands. 641 -645 . Loganathan P and Nalliah V 1977 Downward movement and transfonnation of phosphorus in soils after long-continued fertiliser application to coconut (Cocos nucifera). Journal oj Agricultural Science (Cambridge) 89, 279-284. Mackay A D and Syers J K 1 986 Effect of phosphate, calcium and pH on the dissolution of a phosphate rock in soil. Fertiliser Research 10, 1 75- 1 84. Mackay A D, Gregg P E H and Syers J K 1980 A preliminary evaluation of Chatham Rice phosphorite as a direct use phosphatic fertiliser. New Zealand Journal oj Agricultural Research 23, 44 1 -449. Mackay A D, Gregg P E H and Syers J K 1 984 Field evaluation of Chatham Rice phosphorite as a phosphatic fertiliser for pasture. New Zealand Journal oj Agricultural Research 27, 65-82. Mackay A D, Syers J K, Tillman R W and Gregg P E H 1 986 A simple model to describe the dissolution of phosphate rock in soils. Soil Science Society of America Journal 50, 29 1-296. Mahendrappa M K, Foster N W, Weetman G F and Krause H H 1986 Nutrient cycling and availability in forest soils. Canadian Journal of Soil Science 66, 547-572. 2 5 2 Majidi H and Persson H 1 993 Spatial distribution offine roots, rhizosphere and bulk soil chemistry in an acidified Picea abies stand. Scandinavian Journal of Forest Research 8, 147- 1 55. Malkonen E 1974 Annual primary production and nutrient cycle in some Scots pine stands. Communicationes Instituti Forestalis Fenniae 84, 5 . Mann H H 1935 Tea Soils. Technical Communication No. 32, Imperial Bureau of Soil Science, Harpenden, England. Manoharan V, Loganathan P and Tillman R W 1995 Effects oflong-term application of phosphate fertilisers on soil acidity under pasture in New Zealand In Plant­ Soil interactions at Low pH: Principles and Management. (Eds. R A Date, N J Grundon, G E Rayment and M E Probert). Kluwer Academic Publishers, Dordrecht, The Netherlands. 85-9 l . Markus D K, McKinnon J P and Buccafuri A F 1 985 Automated analysis of nitrite, nitrate and ammonium nitrogen in soils. Soil Science Society of America Journal 49, 1208- 12 1 5 Marschner H, Haussling M and George E 199 1 Ammonium and nitrate uptake rates and rhizosphere pH in non-mycorrhizal roots of Norway spruce (Picea a/hies (L.) Karst). Trees: Structure and Functions 5, 14-2 1 . Marschner H, Rornheld V and Cakmack I 1987 Root induced changes in the nutrient availability in the rhizosphere. Journal of Plant Nutrition 10, 1 1 75- 1 1 84. 2 5 3 McClellan G H and Gremillion L R 1980 Evaluation of phosphatic raw materials. In The Role of Phosphorus in Agriculture. (Eds. F E Khasawneh, E C Sample and E J Kamprath) SSSA , Madison, Wisconsin, 43-80. McKenzie R H, Dormaar J F, Schaalje G B and Stewart J W B 1 995 Chemical and biochemical changes in the rhizosphere of wheat and canola. Canadian Journal of Soil Science 75, 439-447. McLaughlin M J and James T R 1 991 Effect of phosphorus supply to the surface roots of wheat on root extension and rhizosphere chemistry in an acidic subsoil. Plant and Soil 134, 73-82. Mengel K and Kirkby E A 1987 Principles of Plant Nutrition. 4th edition. International Potash Institute, Worblaufen - Berne, Switzerland. Menon R G and Chien S H 1 990 Phosphorus availability to maize from partially acidulated phosphate rocks and phosphate rocks compacted with triple superphosphate. Plant and Soil 127, 1 23- 128. Moghimi A, Lewis D G and Oades J M 1 978 Release of phosphate from calcium phosphates by rhizosphere products. Soil Biology and Biochemistry 10, 277- 28 1 . Mokwunye U 1979 Phosphorus fertiliser in Nigerian Savannah soils. II. Evaluation of three phosphate sources applied to maize at Samaru. Tropical Agriculture (Trinidad) 56, 65-68 . Morita A and Konishi S 1989 Relationship between vesicular-arbuscular mycorrhizal infection and soil phosphorus concentration in tea fields. Soil Science and Plant Nutrition 35, 1 39- 143 . Murdoch J T and Seay W A 1 955 The availability of greenhouse crops of rock phosphate phosphorus and calcium in superphosphate-rock phosphate mixture. Soil Science Society of America Proceedings 19, 199-203 . 2 54 Murphy J and Riley J P 1962 A modified single solution method for the detennination of phosphate in natural waters. Analytica Chimica Acta. 27, 3 1 -36. Nagarajah S, Posner A M and Quirk J P 1968 Desorption of phosphate from kaolinite by citrate and bicarbonate. Soil Science Society of America Proceedings 32, 507-5 1 0. Naidu R, Syers J K Tillman R W and Kirkman J H 1 991 Assessment of plant-available phosphate in limed, acid soils using several soil testing procedures. Fertiliser Research 30, 47-53. Nelson L E 1 980 Phosphorus nutrition of cotton, peanuts, rice, sugarcane, and tobacco. In The Role of Phosphorus in Agriculture. (Eds. F E Khasawneh, E C Sample and E J Kamprath) SSSA, Madison, Wisconsin, USA. 693-736. Newman E I 1 966 A method of estimating the total length of roots in a sample. Journal of Applied Ecology 3, 1 39- 145. Newman E J 1 978 Root micro-organisms: Their significance in the ecosystem. Biological Reviews of the Cambridge Philosophical Society 53, 5 1 1 -544. Nielsen N E and Barber S A 1978 Differences among genotypes of com in the kinetics ofP uptake. Agronomy Journal 70, 695-698. Norris R V 1 946 Fertiliser rationing for 1946. Tea Quarterly 18, 1 -2 . Nyatsanga T and Pierre W H 1973 Effect of nitrogen fixation by legumes on soil acidity. Agronomy Journal 65, 936-940. 2 5 5 Nye P H 1972 The measurement and mechanism of ion diffusion in soils VII. A theory for the propagation of changes of pH in soils. Journal of Soil Science 23, 82-92. Nye P H 1 98 1 Changes in pH across the rhizosphere induced by roots. Plant and Soil 61, 7-26. Odhiambo H C 1987 Exploratory investigation of types and rates of fertilisers in the planting hole on replanted tea, Camellia sinensis (L.) Kuntze. Tea 8, 43-48. Ohno T 1989 Rhizosphere pH and aluminium chemistry of red oak and honeylocust seedlings. Soil Biology and Biochemistry 21, 657-660. Olsen S R, Cole C V, Watanabe F S and Dean L A 1 954 Estimation of Available Phosphorus in Soils by Extraction with Sodium Carbonate. USDA Department circular 939. Government Printing Office, Washington D C. Othieno C 0 1973 Physical characterisation of soils. In Annual Report, Tea Research Institute of East Africa, Kericho, Kenya. 38-4 1 . Othieno C 0 1 992 Soils. In. Tea Cultivation to Consumption.(Eds. K C Willson and M N Clifford). 1 st edition, Chapman and Hall, London. 137- 1 72. Othieno C 0 1 994 Agronomic practices for higher tea productivity in Kenya. In Proceedings of the international seminar on "Integrated Crop Management in Tea: Towards Higher Productivity", International Potash Institute, Basel, Switzerland. 67-78. O'Toole J C and Bland W L 1987 Genotypic variations in crop plant systems. Advances in Agronomy 41, 91- 140. Paauw W 1 965 Factors controlling the efficiency of rock phosphate for potatoes and rye on humic sandy soils. Plant and Soil 22, 8 1 -89. Palmer B and Jessop R S 1982 The relative value of water soluble and ammonium­ citrate soluble phosphorus for wheat production. Journal of Agricultural Science (Cambridge) 98, 467-470. 2 5 6 Panabokke C R 1 967 Soils of Ceylon and Fertiliser Use. Ceylon Association for the Advancement of Science, Colombo, Sri Lanka. Parfitt R L 1 977 Phosphate adsorption on an oxiso1. Soil Science Society of America Proceedings 41, 1064- 1067. Peris 0 S 1 970 Handbook of Rubber. Rubber Research Institute of Sri Lanka. Agalawatta, Sri Lanka. Perrott K W 1 995 Changes in soil P fractions resulting from P depletion and addition in a glasshouse experiment. In Fertiliser ReqUirements of Grazed pasture and Field Crops -Macro and Micro Nutrients. (Eds. L D Currie and P Loganathan) Massey university, Palmerston North, New Zealand. 1 2 1 - 128 . Portsmouth G B 1953 Report of the plant physiologist for 1 952. Report of Tea Research Institute of Ceylon. 3 1 -39. Rajan S S S and Gallingham A G 1986 Phosphorus rocks and phosphate rock/sulphur granules as fertilisers for hill country pasture. New Zealand Journal of Experimental Agriculture 14, 3 1 3-3 1 8. Rajan S S S, Brown M W, Boyes M K and Upsdell M P 1992 Extractable phosphorus to predict agronomic effectiveness of ground and unground phosphate rocks. Fertiliser Research 32, 291 -302. Rajan S S S, Watkinson J H and Sinclair A G 1996 Phosphate rocks for direct application to soils. Advances in Agronomy 57, 77- 1 59. Ramanathan K M and Krishnamoorthi K K 1979 Clay mineralogy of Tamil Nadu soils. Myssoorie Journal of Agricultural Science 13, 283-290. Ranganathan V 1971 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Tea Scientific Department, Cinchona, India. 73-74. 2 5 7 Ranganathan V 1972 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 50- S ! . Ranganathan V 1973 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 72- 75. Ranganathan V 1974 Annual Report of the Chemistry Division, United Planters Association of South India. Tea, Scientific Department, Cinchona, India. 86- 87. Ranganathan V 1975 Annual Report of the Chemistry Division, United Planters Association of South India. Tea, Scientific Department, Cinchona, India. 4 1 - 42. Ranganathan V 1976 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 39- 66. Ranganathan V 1977 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 4 1 - 42. Ranganathan V 1977 a Tea soils of South India with special reference to nutritional and manuring problems. Fertiliser news - February 1997. United Planters Association of South India. Tea Scientific Department, India. 19-24 2 58 Ranganathan V 1 978 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 44- 45. Ranganathan V 1 979 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 5 1 - 52. Ranganathan V 1980 Annual Report of the Chemistry Division, United Planters Association of South India. Tea Scientific Department, Cinchona, India. 47- 48. Ranganathan V and Natesen S 1985 Potassium nutrition in tea. In Potassium in Agriculture. (Ed. R D Munson) ASA-CSSA-SSSA, 677, South Segoe Road, Madison, WI 5371 1 , USA. 98 1 - 102 l . Ranganathan V, Natesen S and Bhat S S 1982 P and K interaction in tea, a pointer to balanced nutrition for high yields. Indian Potash Journal 7, 1 0- 16. Ravan J A and Smith F A 1 976 Nitrogen assimilation and transport in vascular land plants in relation to intercellular pH regulation. New Phytologist 76, 4 1 5- 43 l . Riggs K S and Syers J K 1 99 1 Dissolution of Potash and Phosphate Rocks in Open Leaching and Close Incubation Systems. British Geological Survey Technical Report WC/91145/R. Riley D and Barber S A 1971 Effect of ammonium and nitrate fertilisation on phosphorus uptake as related to root induced pH changes at the root-soil interface. Soil Science Society of America Proceedings 35, 301 -306. _. -- --------- 2 5 9 Robinson J S and Syers J K 199 1 Effect of solution calcium concentration and calcium sink size on the dissolution of Gafsa phosphate rock in soils. Journal of Soil Science 42, 389-397. Ron R 1 986 A Taste of Bitterness. The Political Economy of Tea Plantations in Sri Lanka. Free university press, Amsterdam. Rovira A D 1 979 Biology of soil-root interface. In The Soil Root Interface. (Eds. J L Harley and R S Russell) Academic Press. New York. 145- 160. Rovira A D and McDougall B M 1 967 Microbiological and biochemical aspects of the rhizosphere In Soil Biology and Biochemistry (Eds. A D McLaren and G H Peterson). Edward Arnold Publishers Ltd. London. 4 1 8-463 . Saggar S, Hedley M J and White R E 1990 A simplified resin membrane technique for extracting phosphorus form soils. Fertiliser Research 24, 1 73 - 1 80. Saggar S, Hedley M J and White R E 1992a Development and evaluation of an improved soil test for phosphorus. I. The influence of phosphorus fertiliser solubility and soil properties on the extractability of soil P. Fertiliser Research 33, 8 1 -9 1 . Saggar S, Hedley M J, White R E, Gregg P E H, Perrott K W and Cornforth I S 1 992b Development and evaluation of an improved soil test for phosphorus 2. Comparison of the Olsen and mixed cation-anion exchange resin tests for predicting the yield of rye grass grown in pots. Fertiliser Research 33, 1 35-144. Saggar S, Hedley M J, White R E, Gregg P E H, Perrott K W and Cornforth I S 1993 Assessment of the relative agronomic effectiveness of phosphate rocks under glasshouse conditions. Fertiliser Research 34, 14 1 - 1 5 1 . 2 60 Saggar S, Hedley M J, White R E, Gregg P E H, Perrott K W and Cornforth I S 1995 Evaluation of soil tests for predicting pasture response on acid soils to phosphate supplied in water soluble and sparingly soluble forms. In. Plant-Soil Interactions at Low pH. Principles and Management. (Eds. R A Date, Grundon, G E Rayment and M E Probert) Kluwer Academic Publishers. Dordrecht, The Netherlands. 657-660. Sainz M J and Arines J 1 988 Effects of native vesicular-arbuscular-mycorrhizal fungi and phosphate fertiliser on red clover growth in acid soils. Journal of Agricultural Science (Cambridge) 1 1 1, 67-73 . Sale P W G 1 990 The agronomic efficiency, the economics and the quality of directly applied ground rock phosphate and partially acidulated phosphate rock. In Proceedings of IFAIIMPHOS Seminar on Role of Phosphates in Balanced Fertilisation 1 , IFA, Paris. 1 86-202. Sale P W G and Mokwunye A U 1 993 Use of phosphate rocks in the tropics. Fertiliser Research 35, 33-45. Sanchez P A 1 976 Properties and Management of Soils in the Tropics. John Wiley, New York. Sandanam S, Krishnapillai S and Sabaratnam J 1 978 Nitrification of ammonium sulphate and urea in an acid red yellow podzolic tea soil in Sri Lanka in relation to soil fertility. Plant and Soil 49, 9-22. Sandanam S, Sivasubramaniam S and Rajasingham C C 1 980 Response of seedling tea to forms and levels of nitrogenous fertilisers, levels of potassium and liming in the up-country tea growing districts of Sri Lanka. Tea Quarterly 49, 20-29. 2 6 1 Sanders F E and Tinker B P 1 973 Phosphate flow into mycorrhizal roots. Pesticide Science 4, 385-395. Sanyal S K and De Datta S K 1991 Chemistry of phosphorus transformations in soil. Advances in Soil Science 16, 1 - 1 19. SAS Institute 1 985 SAS Stat User's Guide, Version 5th edition, Cary, NC: SAS Institute Inc., USA. Saunders W M H, Sherrell C G and Gravell I M 1 987 A new approach to the interpretation of soil tests for phosphate response by grazed pasture. New Zealand Journal of Agricultural Research 30, 67-77. Scott R M 1962 Summary of the soil survey observations on the Sambert Valley. East African Agriculture and Forestry 10urna1 27, 22. Selvendran R R 1970 Changes in the composition of the xylem exudate of tea (Camellia sinensis L.) during recovery from pruning. Annals of Botany 34, 825-833 . Sengupta A K, Bhowmik S and Sen A 1 986 Phosphate fixation and its fertility status in acid soils of Do oars. Two and a Bud 33, 49-52. Sivasubramaniam S and Jayman T C Z 1 976 The use of fertiliser for tea in Sri Lanka 2. Foliar and soil analysis with particular reference to potassium. Tea Quarterly 46, 4- 1 1 . Sivasubramaniam S and Talibudeen 0 1 971 Effect of aluminium on growth of tea (Camellia sinensis) and its uptake on potassium and phosphorus. Journal of Science Food and Agriculture 22, 325-329. Sivasubramaniam S, Wickramasinghe K N and Ayadurai S 198 1 The use ofEppawala apatite for tea in Sri Lanka. Tea Quarterly 50, 4- 10. 2 6 2 Smith S E, St John B J, Smith F A and Nicholas D J D 1985 Activity of glutamine synthetase and glutamate dehydrogenase in Trifolium subterraneum L. and Allium cepa L. : Effect of mycorrhizal infection and phosphate nutrition. New Phytologist 99, 2 1 1 -2 17 . Smith W H and Pooley A S 1989 Red spruce rhizosphere dynamics : spacial distribution of aluminium and Zn in near-rock soil zone. Forest Science 35, 1 1 1 4- 1 124. Smyth T J and Sanchez P A 1982a Phosphate rock and superphosphate combinations for soybeans in a Cerrado Oxiso1. Agronomy Journal 74, 730-735 . Smyth T J and Sanchez P A 1982b Phosphate rock dissolution and availability in Cerrado soils as affected by phosphorus sorption capacity. Soil Science Society oj America Journal 46, 339-345 . Son C L and Smith S E 1988 Mycorrhizal growth responses: Interaction between photon irradiance and phosphorus nutrition. New Phytologist 108, 305-3 14. Spencer K, Govaars A G and Hely F W 1980 Early phosphorus nutrition of eight forms of two clover species Trifolium ambigum and T. repens. New Zealand Journal oj Agricultural Research 23, 457-475. Subramanian K S and Mani G 198 1 Genetic and geomorphic aspects of laterites on high and low landforms in parts of Tamil Nadu, India. In Proceedings of International Symposium on lateritisation processes, Trivandrum, Kerala, India . . Oxford and ffiH Publishing Co. New Delhi. 237-245. Switzer G L and Nelson L E 1972 Nutrient accumulation and cycling in loblolly pine (Pinus radiata L.) plantation ecosystems: the first twenty years. Soil Science Society oj America Proceedings 36, 143-147. 2 6 3 Syers J K and Mackay A D 1986 Reactions of Sachura phosphate rock and single superphosphate in soil. Soil Science Society of America Journal 50, 480-485. Syers J K, Mackay A D, Brown M W and L D Currie 1986 Chemical and physical characteristics of phosphate rock materials of varying reactivity. Journal of Science Food and Agriculture 37, 1 057-1064. Tambunan D 1 992 Dissolution and Plant-availability of Phosphate Rocks in Selected New Zealand and Indonesian soil. PhD Thesis. Massey University, New Zealand. Tambunan D, Hedley M J, Bolan N S and Turner M A 1993 A comparison of sequential extraction procedures for measuring phosphate rock residues in soils. Fertiliser Research 35, 1 83- 1 9 1 . Tarafdar J C and Claassen N 1 988 Organic phosphorus compounds as a phosphorus source for higher plants through the activity of phosphatases produced by plant roots and micro-organisms. Biology and Fertility of Soils 5, 308-3 1 2. Tarafdar J C and Jungk A 1 987 Phosphatase activity in the rhizosphere and its relation to the depletion of soil organic phosphorus. Biology and Fertility of Soils 3, 1 99-204. Tazaki K, Fyfe W S and Dissanayake C B 1 987 Weathering of apatite under extreme conditions of leaching. Chemical Geology 60, 1 5 1 - 162. Tiessen H and Moir J 0 1 993 Characterization of available P by sequential extraction. In Soil Sampling and Methods of Analysis (Ed. M R Carter). Canadian Society of Soil Science, Lewis Publishers. 75-86. Tinker P B 1980 Role of rhizosphere micro-organisms in phosphorus uptake by plants. In. The role of Phosphorus in Agriculture (Eds. F E Khasawnch, E C Sample and E J Kamprath) ASA- CSSA-SSSA, Wisconsin, USA. 6 1 7- 654. 2 6 4 Tinker P B 1984 The role of micro-organisms in mediating and facilitating the uptake of plant nutrients from soil. Plant and Soil 76, 77-9 1 . Tisdale S L, Nelson W L and Beaton J D 1985 Soil Fertility and Fertilisers. 4th edition, Macmillan Publishing Co. New York. Tolhurst J A H 1 955 Report ofthe Agricultural Chemistry Division. Tea Research Institute of Ceylon Bulletin 37, 3 1 -32. Tolhurst J A H 196 1a Maturing mature tea: The new T 700 series. Tea Quarterly 32, 148- 1 5 1 . Tolhurst J A H 1961 b Manuring young tea. Tea Quarterly 32, 1 52- 1 54. Tolhurst J A H 1963 Concluding report on a phosphate manurial trial, St. Coombs. Tea Quarterly 34, 144- 147. Tolhurst J A H 1965 Economising on phosphate and potash for mature tea. Tea Quarterly 36, 45-47. Tolhurst J A H 1968 Report of the Agricultural Chemistry Division. Tea Research Institute of Ceylon. 25-4 1 . Tolhurst J A H and Richards A V 1965 A revised fertiliser mixture for nursery ­ Tss. Tea Quarterly 36, 43-44. Tolhurst J A H and Visser T 196 1 Manuring of nurseries: 2 TRI nursery manure: A completely soluble inorganic mixture. Tea Quarterly 32, 220-22 1 . Troelstra S R 1983 Growth of Plantago major on a NOYNH. medium and the estimation of the utilisation of nitrate and ammonium form ionic-balance aspects. Plant and Soil 70, 1 83-197. Troelstra S R Van Dijk K and Blacquiere T 1 985 Effects ofN source on proton excretion, ionic balance and growth of Alnus glutinosa (L.) Gaertner: comparison ofN2 fixation with single and mixed sources ofN03 and �. Plant and Soil 84, 361-385 . Trolldenier G 1992 Techniques for observing phosphorus mobilisation in the rhizosphere. Biology and Fertility of Soils 14, 1 2 1 - 125 . 2 6 5 Trolove S N, Hedley M J, Cardus J R and Mackay AD 1 996a Uptake of phosphorus from different sources by Lotus pendunculatus and three genotypes of Trifolium repens. I. Plant yield and phosphate efficiency. Australian Journal of Soil Research 34, 1 0 1 5- 1026. Trolove S N, Hedley M J, Cardus J R and Mackay A D 1 996b Uptake of phosphorus from different sources by Lotus pendunculatus and three genotypes of Trifolium repens. II. Forms of phosphate utilised and acidification of the rhizosphere. Australian Journal of Soil Research 34, 1027- 1040. Utomo M 1 995 Effect of rock phosphate on soil properties and apparent phosphorus recovery in acid soil of Sumatra In. Plant-Soil Interactions at Low pH: Principles and Management, (Eds. R A Date, N J Grundon, G E Rayment and M E Probert) Kluwer Academic Publishers, Dordrecht, The Netherlands. 653-656, Van Ray B and Van Diest A 1979 Utilisation of phosphate from different sources by six plant species. Plant and Soil 51, 577-589. 2 6 6 Visser T and Khel F H 1961 Manuring of nurseries. Experiments on the frequency and methods of application of inorganic and organic mixtures. Tea Quarterly 32, 2 1 6-2 19. Walldey A and Black I A 1934 An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil Science 37, 29-38 . Wang J , Luo Z and Loeppert R H 1 995 Nutrient status ofrhizosphere and phosphorus response ofraddish. Journal of Plant Nutrition 18, 385-399. Watson M 1 986 Soil and climatic requirements In Handbook on Tea. (Eds. P Sivapalan, S Kulasegaram and A Kathirvetpillai). Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka, 3-5 . Watson M and Wettasinghe D T 1972 Some effects of treatments designed to lower soil acidity on the growth of the tea plant. Journal of the National Agricultural Society of Ceylon 9/10, 32-42. Waugh D B and Manzano A 1 97 1 The correlation of phosphorus response with soil analysis in tall and drawf wheat varieties in Bolivia. Proceedings of International Symposium of Soil Fertility Evaluation (New Delhi) 1, 377-382. Weast R C 1 97 1 Handbook of Chemistry and Physics 52nd edition. ( 197 1 - 1972). The Chemical Rubber Co. Ohio, USA. White R E, Hedley M J, Bolan N S and Gregg P E H 1 989 Recent developments in the use of phosphate fertiliser on New Zealand pastures. Journal of Australian Institute of Agricultural Science 2, 25-32. Wickramasinghe K N 1986 Soil survey in tea plantations and adjoining uncultivated jungle areas in different agro-ecological regions In Annual Report of TRI - 2 6 7 1986. (Eds. P Sivapalan, S Kulasegaram and A Kathiravetpillai) Tea Research Institute of Sri Lanka, Talawakelle, Sri Lanka. 40-41 . Wickramasinghe K N and Krishnipillai S 1986 Fertiliser use in tea In Handbook on tea. (Eds. P Sivapalan, S Kulasegaram and A Kathiravetpillai). Tea Research Institute of Sri Lank, T alawakelle, Sri Lanka. 63 -77. Will G M 1 968 The uptake, cycling and removal of mineral nutrients by crops of Pinus radiata. Proceedings of the New Zealand Ecological Society 15, 20-24. Williams J D H and Walker T W 1 969 Fractionation of phosphate in a maturity sequence of New Zealand basaltic soil profiles. Soil Science 107, 22-30. Willson K C 1969 Tea Nutrition. Tea, Journal of Tea Board of East Africa 10, 1 6- 23 . Willson K C and Clifford M N 1992 Tea Cultivation to Consumption. 1 st edition, Chapman and Hall, London. Willson K C, Hainsworth E, Green M J and O'Shea P B T 1975 Studies on the mineral nutrition of tea. Plant and Soil 43, 259-278. Wit C T de, Dijkshoorn W and Noggle J C 1 963 Ionic Balance and Growth of Plants. Versl. Landbouwk, Onderz, Pudoc Wageningen, 68. Xan W and Jianyun R 1 994 Effect of potassium and magnesium interaction in tea production and quality of green, Oolong and black tea in China. In: Proceedings of the International Seminar on "Integrated crop management in tea: Towards higher productivity", International Potash Institute, Basel, Switzerland, 203-2 14. Xiaoping W 1 994 Study of the organic acids extracted by tea roots. Journal of Tea Science (China) 14, 17-22. Xiaoping W, Xun W and Y ongming Y 1 989 Study on activities of phosphatases in red earth in tea field. Journal of Tea Science (China) 9, 99-1 08. 2 6 8 Yongming Y, Xun W, Xiaoping W and 2hi L 1989 Phosphorus nutrition and transformation in the red tea soil in low hilly regions. Journal of Tea Science (China) 9, 23-34. Young C C, Juang T C and Guo H Y 1986 The effect of inoculation with vesicular­ arbuscular mycorrhizal fungi on soybean yield and mineral phosphorus utilisation in sub-tropical and tropical soils. Plant and Soil 95, 245-253 . Youssef R A and Chino M 1988 Development of new rhizobox system to study the nutrient status in the rhizosphere. Soil Science and Plant Nutrition 34, 461 - 467. YoussefR A and Chino M 1 989 Root-induced changes in the rhizosphere of plants 1 . pH changes in relation to the bulk soil. Soil Science and Plant Nutrition 35, 46 1 -468. 2aharah A R and Sharifuddin H A H 1995 Use of rocks for sweet com production on a Malaysian Ultisol. In Plant-Soil Interactions at Low pH: Principles and Management, (Eds. R A Date, N J Grundon, G E Rayment and M E Probert). Kluwer Academic Publishers. Dordrecht, The Netherlands. 647-65 1 . 2hi L 1 993 Effect of VA Mycorrhiza on the growth and uptake of mineral nutrients of tea plant. Journal of Tea Science (China) 13, 1 5-20.