PHOSPHORUS ADSORPTION MAXIMA OF SELECTED GHANAIAN SOILS AND THEIR RELATIONSHIP TO PHOSPHORUS AVAILABILITY A Thesis Presented to the Faculty of Agriculture University of Ghana In Partial Fulfilment of the Requirements for the Degree Master of Science by Joseph Cobbina Crop Science Department 1976 University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh ii ACKNOWLEDGMENT It is with pleasure that I express my sincere appreciation to my supervisor, Dr. E.J. Thompson, for his suggestions and constructive criticism during the investigation, and for his helpful advice during the preparation of the manuscript. Special thanks are due to Mr. K.B. Laryea, Research Fellow of the Volta Basin Research Project, who showed much interest in this work and spent some time reading and correcting the manuscript during the absence of my supervisor from the country- My thanks are due also to the staff of the Soils Division, Department of Crop Science for assisting me in diverse ways. The efficient typing of Mr. A. A. Ansah of VBRP., Legon, is very much appreciated. Joseph Cobbina. DR. E-r5T THOMPSON (SUPERVISOR). University of Ghana http://ugspace.ug.edu.gh iii TABLE OF CONTENTS CHAPTER PAGE 1 INTRODUCTION .. LITERATURE REVIEW Vii 1 Definition of the terms Adsorption and Availability .. .. 1 The Phosphorus Adsorption Phenomenon .. .. .. 4 Factors Affecting Phosphorus Adsorption in Soils.. .. 11 Methods of Determination of Available Phosphorus. .. 17 Assessment ofPRequirements of Soil from P Sorption. .. 34 Determination of the Adsorption Maximum .. .. .. 41 Determination of "Free" Fe and A1 Oxides .. .. .. 43 Determination of Silica Content of Soils .. .. .. 46 Estimation of Organic-C,pH and Clay Content of Soils .. 47 Determination of Available Phosphorus in the Soils .. 50 Greenhouse Experiment .. .. 51 Laboratory Analysis of Plant Material for P ,Ca,Mg,Fe,Mn.. 63 2 MATERIALS AND METHODS . Description of Soils Preparation and Storage of Soil Samples ....... 36 36 41 University of Ghana http://ugspace.ug.edu.gh iv CHAPTER PAGE 3 RESULTS .. .. 68 Phosphorus Adsorption Maximum of Soils U s e d ........... 68 Relation of Adsorption Maximum and Bonding Energy to Some Soil Properties .... 70 Dry Matter Yield as Related to the Adsorption Maximum .... 78 Per cent P Uptake as Related to the Adsorption Maximum. .. 90 Comparison of Dry Matter Yields and P Uptake Values for K^PO^ and K^HPO^ Treatments. 103 Plant Tissue Composition of Ca, Mg, Fe, Mn .................105 Estimated P Application Rates for the Soils Used ......... 112 4 DISCUSSION....................... 115 5 SUMMARY AND CONCLUSION ............. 136 REFERENCES....................... 141 University of Ghana http://ugspace.ug.edu.gh 12 3 4 5 6 7 8 9 10 11 12 13 14 15 v LIST OF TABLES Description of Soils Used .. .. .. 36 Some Properties of Soils Used .. .. 37 Amounts of P Applied to the Soils .. 53 Amounts of KH^PO^ Salt Added to Soils. 55 Amounts of K^HPO^ Salt Added to Soils. 57 Amounts of K?C0, Salt Added to KH?P0, Treated Soils .. .. .. .. 59 Amounts of K?C0, Salt Added to K?HP0» Treated Soils. .. .. .. 60 Adsorption Maximum Data .. .. .. 71 Phosphorus Adsorption Maximum and Bonding Energy in Relation to Some Soil Properties .. .. .. 76 Effect of P Application on Dry Matter Yield .. .. .. .. .. 80 Effect of P Application on P Concen­ tration of Millet .. .. .. 92 Relation Between Adsorption Maximum, Bonding Energy, and Uptake of Applied Phosphorus .. .. .. 102 Effect of P Application Rates on Concentration of Ca,Mg,Fe,Mn in Millet Tissues .. .. .. .. 107 Correlation Between Quantity of Phosphorus Added and Plant Tissue Concentration of Iron and Manganese 111 Estimated P Application Rates for Soils Used .. .. .. .. .. 114 PAGE University of Ghana http://ugspace.ug.edu.gh vi LIST OF FIGURES FIGURE PAGE 1 Phosphorus Adsorption Data Plotted According to the Langmuir Isotherm .... 74 2(a) The Relative Yield of Millet as Related to the P Saturation of the Adsorption Maximum for Koforidua,Akuse and Wacri Soils .. . . . . . . 85 2(b) The Relative Yield of Millet as Related to the P Saturation of the Adsorption Maximum for Ankasa and Mamfe Soils .... 86 2(c) The Relative Yield of Millet as Related to the P Saturation of the Adsorption Maximum for Agawtaw,Boi,Oyarifa and Toj e Soils . . . . . . ........ 8 7 2(d) The Relative Yield of Millet as Related to the P Saturation of the Adsorption Maximum for Tikobo Soil .. 88 3(a) The Per cent P Uptake of Millet as Related to the P Saturation of the P Adsorption Maximum for the Ankasa,Boi, and Mamfe Soils .. .. 95 3(b) The Per cent P Uptake of Millet as Related to the P Saturation of the P Adsorption Maximum for the Tikobo Soil . . . . . . . . 96 3(c) The Per cent P Uptake of Millet as Related to the P Saturation of the P Adsorption Maximum for the Agawtaw, Akuse and Toje Soils . .. 97 3(d) The Per cent P Uptake of Millet as Related to the P Saturation of the P Adsorption Maximum for the Oyarifa Soil .. .. . . .. 98 3(e) The Per cent P Uptake of Millet as Related to the P Saturation of the P Adsorption Maximum for the Oyarifa Soil .. . .. .. 99 University of Ghana http://ugspace.ug.edu.gh INTRODUCTION An adequate knowledge of the chemical reactions which occur when fertilizer is placed in soils is a pre-requisite to the development of sound fertilizer practices. Without such a knowledge, there is the possibility of applying either too much or too little fertilizer than is necessary with its consequent crop failure. Phosphorus is one of the major nutrient elements required by plants for growth. The role of phosphorus in plant nutrition includes its effect on cell division, flowering, fruiting and seed formation, crop maturation, root development, especially of lateral and fibrous rootlets, strength of straw in cereals, and crop quality of vegetables. The problem of phosphorus deficiency in Ghanaian soils and its attendant low crop -yields has long been recognized by many research workers. Between 5,0 ppm and 14.5 ppm phosphorus has been found in the topsoil CO" " 6") of many forest profiles in Ghana (Hardy and Amoroso-Centeno 1938; DeEndredy and Montgomery 1954; Nye 1952). Using Bray's rapid extraction procedure with 0,TN. HC1 and 0.03N NH^F Nye (1952) obtained 4 ppm phosphorus in the topsoil (0" - 6") of sixty-three soils from savanna sites in Ghana, In spite of these very low value o-f available phosphorus in Ghanaian University of Ghana http://ugspace.ug.edu.gh soils, Nye (1952) reports that response to phosphorus application has not been always conclusive. This lack of response has often been attributed to the high fixation capacity of tropical soils, especially the strong acid ones with high content of iron and aluminium rendering small dressings of phosphate ineffective. Several observations made by agronomists the world- over on crop response to applied phosphate give credence to those made in Ghana, These indicate that fertilizer phosphate after it has been applied is not recovered wholly in the crop that is immediately planted. Hemwall (1957) in a review, reports that crops recover only 10 to 301 of applied phosphorus, Sauchelli (1965) also reports that plants on phosphorus-fertilized soils generally recover just 20 to 30% of the added phosphate. The general consensus among soil chemists is that chemical precipitation and colloidal adsorption are chiefly responsible, for the loss. As indicated earlier on, most Ghanaian soils are deficient in phosphorus and yet would not give any response to applied phosphorus fertilizer. Investigation by many research workers have also proved the methods of assessing the availability of phosphorus quite inadequate. We therefore propose to tackle the problem of phosphorus availability studies from another angle, University of Ghana http://ugspace.ug.edu.gh ix In our view before any: economic yield can be realized from any phosphorus fertilizer application it is necessary to understand the phenomenon of phosphorus adsorption in soils and how the phosphorus is made available to plants. The purpose of the present study, therefore, is to inves­ tigate the phosphorus adsorption phenomenon in some selected Ghanaian soils with a view to greater understanding of the ways of avoiding problems associated with phosphate ferti­ lizer application to these soils and also with a view to maximizing the efficient use of such fertilizer applications. The research programme was- designed to:- (i) determine the phosphorus adsorption maxima of .selected Ghanaian soils using the Langmuir isotherm as modified by Olsen and Watanable (1957), (ii) relate these maxima to phosphorus availability, Ciii) relate the adsorption maxima to some soil properties and, (iv) estimate the rates of P application necessary to obtain optimum yield on the soil series used. University of Ghana http://ugspace.ug.edu.gh CHAPTER I LITERATURE- REVIEW 1.1. Definition.' of the' terms' Adsorption, arid Availability. Adsorption may be defined broadly as the attraction of ions or compounds to the surface of a solid. Silicate clay minerals have on their broken or exposed edges hydrous oxides of aluminium and iron. These complex hydrous oxides of aluminium and iron when in contact with soil solution become hydrolyzed as follows R-OH + HOII = R-OHJOH- ? ? * ,, . . (1 ) The hydroxyl ions of the complex aluminium or iron hydrous oxides thus formed can exchange positions with the phosphate ions in solution supplied from a native or fertilizer phosphorus source as shown in equation (’2). R-OHJOH" + U 2P0‘ = R - O H ^ P O " + OH' , . ,, (2) R-Oh JoH" + H+C1“ = R-OH^Cl-' + H+QH" ., ,. (3) R-OH^Cl" + H2P0^ . = R-OH^H^PO^ + Cl" ,, . . (4) Alternatively the hydroxyl ions can exchange places with Cl ,S04 ~ or N03 ions in solution which in turn can be exchanged with phosphate ions in solution (refer to equations (3) and (4) above). Also calcium attached to clay minerals may attract to themselves hydroxyl ions or other anions. University of Ghana http://ugspace.ug.edu.gh 2These attached ions may, in the presence of phosphate ions, exchange places. Furthermore, phosphate ions in solution may be attracted directly to the calcium attached to clay minerals. In either of these latter reactions involving calcium attached to clay minerals a clay-Ca- phosphate linkage is formed. The phosphate ions so attracted are said to be adsorbed phosphate ions and the mechanism involved is correspondingly termed adsorption or, more specifically, phosphate adsorption. Evidence in support of the above theory of phosphorus adsorption is provided by Fried and Dean (1955). They determined the phosphate-fixing characteristics of iron- and aluminium-saturated cation exchange resins. They found that these materials are capable of fixing phosphorus and concluded that a similar phenomenon could occur in the soil via the clay minerals. They also presented experimental data which show that 75% of the phosphorus retained by the ferrated exchange resins- was readily exchangeable with radioactive phosphorus. They stated that this is an "improbably high exchangeability for a precipitate'' and, hence, must be an adsorption. The term "available", as applied to nutrients, refers to their existence in the soil in a chemical condition University of Ghana http://ugspace.ug.edu.gh in which they may be absorbed by plant roots or may be readily converted into such a condition. Nutrients adsorbed on colloidal fraction in an easily replaceable state would be considered available in this context. Phosphate which is soluble and capable of entering the soil solution and of being absorbed by plants would, therefore, be considered as available phosphorus. The availability of inorganic phosphorus is largely determined by the following factors: soil pH; soluble iron, aluminium, and manganese; presence of iron-, aluminium-, and manganese - containing minerals; available calcium and calcium minerals; amount and extent of decomposition of organic matter and the activities- of microorganisms. Phosphorus availability can be assessed in diverse ways. Different research workers adopt different methods which best suit their particular objective to establish phosphorus availability. In this particular research work the straight chemical extraction procedure whereby the available inorganic phosphorus is extracted with dilute acids and bases was adopted for assessement of the availability index of the native phosphorus in the soils used for the study. Availability of applied phosphorus was also assessed by plant uptake and subsequent chemical analysis of the plant University of Ghana http://ugspace.ug.edu.gh for the inorganic phosphorus concentration. 1.2. The Phosphorus Adsorption Phenomenon. Phosphorus fixation was first recognized in Europe around 1850. At that time, reports Hemwall (1957), various workers in the field just reported that soil had the ability to "retain" phosphorus. Similar reports were known to have appeared in the United States shortly after 1900. Yet the greatest strides which helped to throw light on the basic chemistry of this phenomenon and how to control it began only in the 19 30's. G. Barbier and associates in France (quoted by Sauchelli (1965) pp. 163) concluded from an experiment conducted over a period of ten years that almost all of the soluble phosphate added to the soil remained after ten years in forms which are either extractable by dilute acids or are capable of returning spontaneously to the first form under natural conditions. The first form comprises a linkage of phosphoric ions with exchangeable cations notably calcium located on the surfaces of the clay. The resultant combination represents a sort of adsorption; phenomenon. This exchangeable phosphate, they believed, can pass back and forth in solution in the presence of dilute acids. The second form is believed to comprise a linkage of phosphate ions with exchangeable cations, 4 University of Ghana http://ugspace.ug.edu.gh notably iron, whose hydroxides strongly hold on to phosphate ions in an acid medium but are easily released in an alkaline medium. Report from an Illinois Bulletin (also quoted by Sauchelli (1965) pp. 166) stated that soluble phosphate fertilizers are either adsorbed on surfaces of clay minerals as "adsorbed phosphorus" or are changed to forms of calcium phosphate which are designated as "easily acid- soluble" phosphorus. The added phosphorus fertilizers are not reverted to unavailable forms on contact with the soil as had been reported earlier by some workers. Infact the adsorbed and easily acid-soluble forms of phosphorus are. believed to be of primary importance in plant feeding. E.G. Williams of the Macauley Institute for Soil Research, Aberdeen, is quoted by Sauchelli (1965, pp. 165) to have likened the adsorption of phosphate to the taking up of water by sponge. When only a small amount of water is added to a dry sponge it is held tightly, but as more water is added it becomes more easily squeezed out. In an analogous manner according to Williams, the reserve of phosphate in the soil must be built up to a certain extent before the phosphat&becomes available to plants. ; 5 University of Ghana http://ugspace.ug.edu.gh A research work carried out by Sell and Olsen (1946) gives credence to Williams' postulate referred to above. Sell and Olsen (1946) found that the higher the amount of phosphorus applied the greater the amount of phosphorus available to the plant that grows on the soil. It is indicative from the results that a soil has a certain capacity to fix phosphorus, and after that capacity has been filled, phosphorus may then become available. A pertinent question to pose at this juncture, however, is what are the fixation capacities of various soils of agricultural importance? This question had plagued the minds of many of the. earlier researchers. In their bid to find solutions they conducted experiments designed to study phosphorus adsorption. Davis (1935), Kurtz et al. (1946), and Russell and Low (1954) pointed out from their respective studies that the reaction of added phosphorus with soils at low concentrations can be described by means of the classical Freundlich equation. This equation expresses the relation between adsorption and concentration and may be stated as X/m-KC , where X is. the amount of solute adsorbed by m grams- of soil, C is the equilibrium concentration of the solution and K and £ are constants. 6 University of Ghana http://ugspace.ug.edu.gh Burd and Murphy (1939) proposed that a knowledge of the degree of the adsorption constituents of a soil should help provide a useful index of phosphate availa­ bility. This degree of saturation of the adsorption constituents of a soil can be calculated if the amount of adsorbed phosphate actually present in the soil is known, together with the amount of phosphate which the adsorbing minerals could hold at saturation. This latter quantity is the adsorption capacity or adsorption maximum of the soil for phosphorus. Olsen and Watanabe (1957), Thompson (1958), Rennie and McKercher (1959), Weir and Soper (1962), Gunary (1970), Udo and Uzu (1972) and Syers et al. (1973) have made use of Langmuir (1918) isotherm to calculate phosphorus adsorption maxima in soils. The adsorption equation of Langmuir (1918) is based on a theoretical consideration of the process of adsorption, It may be expressed as:- x/m =' Kbc ' 1 ’ + Kc in which x is the amount of solute adsorbed by the mass m of soil, c is the equilibrium P concentration, b is the adsorption maximum and K is a constant related to the bonding energy of the absorbent for the absorbate. In linear form: c/(x/m) 1 + c ICE b 7 University of Ghana http://ugspace.ug.edu.gh For cases in which the. equation represents the data c/(x/m) may be plotted as a linear function of c with slope 1/b and intercept 1/Kb. The Langmuir euqation is preferred by most workers to the Freundlich equation in that the former has a sound theoretical derivation, is specific for smaller amount of adsorbed phosphorus and more dilute equili­ brium phosphorus concentration (more likely to be encountered in normal phosphorus fertilizer application). Moreover with the Langmuir equation an adsorption maximum can be calculated whereas in the more empirical equation of Freundlich calculation of adsorption maximum is impossible. The Langmuir adsorption equation,like most other good things, has its drawbacks too. One disadvantage of the equation is that at higher equilibrium concentrations the plot of c/(x/m) against : c fails to give a linear relationship. Olsen and Watanable (1957) and Thompson (1958) for instance, found that in more concentrated solutions the Langmuir plots were no longer linear. Weir and Soper (1962) in adsorption and exchange studies of phosphorus observed that the phosphorus adsorption followed the Langmuir isotherm only when the phosphorus solution concentrations were less than 2 5 to 30 micrograms 8 University of Ghana http://ugspace.ug.edu.gh phosphorus per millilitre of solution. Hsu and Rennie (1962) in a study which included, among other things, an investigation of the Langmuir adsorption isotherm as an indicator of adsorption found that a straight line relationship of the plot of c/(x/m) against c suggests that the main reaction removing phosphate from solution is possibly adsorption. They found also that where precipitation of phosphate occurs the plot fails to give the straight line relationship. However, they believed that when phosphate precipitated is much less than that adsorbed on the soil surface a straight line relationship is evident. Larsen (1967) also argued that "it is doubtful that their (Olsen and Watanabe’s 1957) observation even at lower concentrations are generally applicable." Larsen based his argument on the fact that at Levington Research Station, phosphorus adsorption isotherms were determined for 120 soils and the relationship between c/(x/m) and c was curvilinear in the majority of the soils even when they were equilibrated with very dilute solutions giving equili­ brium concentration values of phosphorus less than 6 X 10-4M, Gunary (1970) in a study designed to find out adsorption isotherm for phosphate in soils observed that on a range of soils he worked with phosphate sorption does not University of Ghana http://ugspace.ug.edu.gh obey the Langmuir equation. He further questioned the reliability of phosphate adsorption maximum values calculated from the Langmuir equation. Gunary based his criticism on the fact that for 24 soils used in the above study the plot of c/(x/m) against c appeared to be slightly but consistently convex to the c axis. He therefore suggested alternative equations of which the one given below was observed to give the best fit: 1/y = B + A/C + D/C where y is the adsorbed phosphate, c is the equilibrium concentration of phosphate in solution and B, A, and D are all constants. Syers et al. (1973) also found out. from their study that two linear relationships- are obtained when phosphate sorption data are plotted according to the conventional Langmuir equation. , They proceeded further to rearrange the Langmuir equation as follows:- x/m = K2 - (x/m)/K1C in which x/m is the amount of P adsorbed per unit weight of soil, is a constant related to the bonding energy, K2 is the adsorption maximum and C is the equilibrium P concentration. The rearranged form of the Langmuir equation according to Syers £t al. (1973) should be preferred 10 University of Ghana http://ugspace.ug.edu.gh to the conventional Langmuir equation, since it is more useful for evaluating phosphate sorption at low equilibrium phosphorus concentrations. In view of the fact that the percentage saturation of adsorption maximum may serve as a measure of the capacity of the soil to supply phosphorus to the soil solution and hence to plant roots in contact with this solution, Woodruff and Kamprath (1965) studied the relationship between the growth of millet and the degree of saturation of phosphorus adsorption maximum. They found that soils with large P adsorption maximum did not require as high a P saturation as those with a low P adsorption maximum. They therefore concluded that P adsorption are important parameters in the study of soil phosphorus levels needed for optimum growth. 1.3. Factors 'Affecting 'Phosphorus Adsorption in Soils. Even though the work of Olsen and Watanabe (1957) provided the basis for determination of phosphorus adsorption maximum of soils, it. did not provide information concerning the mechanism by which phosphorus is retained. Several workers however, have attempted to relate phosphorus adsorption to soil properties. Compound of iron and aluminium, soluble calcium and clay minerals have 11 University of Ghana http://ugspace.ug.edu.gh been investigated to determine their role in phosphorus; fixation. The iron and aluminium oxides and hydroxides have, been recognized by many workers as playing significant role in phosphorus fixation. Several arguments in support of this view are presented in reviews, by Dean (1949) , Wild (1950) , and Hemwall (1957). The most direct argument is based on the observation that phosphate sorption is reduced markedly when oxides of aluminium and iron are removed by chemical extraction. Toth (.1937) and (1939), and Kelley and Midgley (1943) used the hydrogen sulphide method of Drosdoff and Truog (1935) to show that the removal of iron and aluminium-oxides from soil colloids reduced phosphate sorption. A similar result has been reported by a number of workers (Black, 1942; Chandler, 1941; Coleman, 1942 and 1944 (a); Metzger, 1934) after iron and aluminium had been removed by slightly modified hydrogen sulphide method after Truog et a l . (1936). Many other early workers (Davis, 1935; Kelley and Midgley, 1943; Perkins and King, 1944; Kurtz', et al., 1946 Ensminger, 1948; Swenson et al.,1949; Struthers and Sieling, 1950$ Bradley and Sieling, 1953) also have postulated and demonstrated that oxides of iron and aluminium play an important role in phosphorus fixation. University of Ghana http://ugspace.ug.edu.gh Several recent reports presented by Williams et al., (1958), Coleman et al., (1960) and. Bromfield (1965), however, indicated that aluminium plays a more dominant role in phosphorus retention than iron. The work of Weir and Soper (1963) also indicated and confirmed previous observa­ tion that organometallic complexes may also be important in phosphate sorption. Saini and MacLean (1965) in their investigation with some New Brunswick soils also found that aluminium exerted more influence on phosphorus reten­ tion capacity than iron. They also found a significant correlation between organic matter and retention capacity but clay per se was adjudged to be a less important criterion of phosphorus retention. Correlations have been established between phosphate sorption and the amounts of iron and aluminium in soils and also the ratio of Si02 to F®2®3 plus A1203- Several investigators, namely Mattson (1931), Scarseth and Tidmore (1934), and Toth (1937) have shown that phosphate sorption varies inversely as the Si0 2/(Fe20 3 + A^ 90 3) ratio of soil colloids decreased. Metzger (1941) also using forty-two soil samples found a significant correlation between total A^O^, (A1 20 3 + Fe-^Og) , and phosphate sorption in soils. Thompson (1957) reported that the amount of free iron and aluminium (that which is not combined in the clay crystal) greatly affects the solubility of phosphorus. 13 University of Ghana http://ugspace.ug.edu.gh He claimed that as the ratio of iron plus aluminium to silica increases consequent upon the weathering of silica there is a corresponding increase in active iron and aluminium combined with phosphate to form insoluble compounds. Thompson (1957) therefore subscribed to the opinion of many research workers that a high efficiency from the use of soluble phosphate can best be associated with soils low in free iron and aluminium. Soils high in free iron and aluminium have a high fixing capacity for phosphate. Sauchelli (1965) also observed that the solubility of phosphate in the soil is influenced greatly by the kind of chemical compounds present. Sauchelli therefore listed, among other things, a high silica to sesquioxide ratio as a factor favouring the solubility of phosphate in soil solutions. Coleman (1944(a) and (b) was, infact, the first soil scientist clearly to postulate that phosphate fixation by clay minerals is due to the aluminium content of the clays and has nothing to do with the intact clay minerals. He showed that the amount of phosphorus fixed by clays is proportional to the amount of free aluminium oxides on the clays. 14 University of Ghana http://ugspace.ug.edu.gh Williams et al. (1958) found that aluminium extracted by the Tamm acid-oxalate method gives highly significant correlations with phosphate sorption in all groups of soils they worked with. It is, in their opinion, the best single criterion of phosphate sorption. Bromfield (1965) from studies on relative importance of iron and aluminium in phosphate sorption concluded tentatively that for most soils phosphate sorption was due to acid soluble aluminium. The contribution of reducible iron in the original soils, according to Brom­ field, remains in doubt, but he claimed it could be but a minor one as well. Pissarides et al. (1968) however, found that the values obtained for the phosphorus adsorption maxima of soils they worked with were a function of both the type of clay and the saturating cations. Rajagopal and Idnani (1963) working with some of the laterite and acid soils of South India found that free oxides of iron and aluminium were more reactive than the combined oxides in respect of P fixation. Ahenkorah (1968) in Ghana also concluded that organic carbon, iron and their interactions with pH are the dominant factors active in phosphorus retention by Ghana cocoa growing soils. 15 University of Ghana http://ugspace.ug.edu.gh Syers et al. (1971) in their investigation on phosphate sorption parameters found that for most soils components such as crystalline iron oxides and associated alumina which are not extracted by oxalate are apparently involved in the sorption of added phosphate. These findings however, contradict those obtained from a range of Scottish soils by Williams et al.(1958) referred to in a previous paragraph. In this latter study, treatment of soils with oxalate lowered the phosphorus adsorption capacity by between 80 and 95 percent. For a large number of Australian soils (Bromfield, 1965) the phosphorus adsorption capacity after oxalate treatment was reduced by between 6 and 67 percent. Shukla et al. (1971) also found in their study that oxalate-extractable iron is the most important contributor to the sorption of added phosphorus by non-calcareous and calcareous lake sediments. Biddappa and Venkat Rao (1973) working on coffee soils of South India reported that free iron oxides were seen to show high correlation with phosphorus fixing capacity indicating that free iron oxides are more active m trapping the phosphorus in the soils thus rendering it unavailable. Phosphorus fixation in alkaline and calcareous soils is usually attributed to the formation of phosphate 16 University of Ghana http://ugspace.ug.edu.gh compounds of calcium. In addition, however, iron and aluminium are also responsible for some fixation in soils of higher pH. Burd (1948) working with calcareous soils pointed out that the very general occurrence of potentially soluble calcium-compounds in soils and the relatively low solubility of the calcium-phosphates would lead to the formation of some form of calcium-phosphate upon the addition of phosphatic fertilizers. He showed that the concentration of calcium in the soil solution is the dominant factor in determining phosphate concentration in the liquid phase of the soil, thus confirming the role of calcium in phosphate fixation. 1.4. Methods of Determination of Available Phosphorus. Generally; the determination of soil nutrient supply can be made in the laboratory in two ways: a.) by chemical methods, and b.) by biological methods. Chemical methods involve techniques of extracting the nutrient in question with different chemical extracting reagents. Biological methods include those techniques in which the nutrient is extracted from the soil by agents such as bacteria, fungi, algae, seedlings, and even entire plants. 17 University of Ghana http://ugspace.ug.edu.gh 18 •1.4.1. Chemical Methods of Determination of Available Phosphorus. Of the procedures for assessing available phosphorus in soils chemical extraction methods have been found most convenient. This is because chemical determinations can be made rapidly. Present-day instrumentation is such that chemical methods, when used in modern laboratories, can be accurate and rapid. The main problem, however, is to select a chemical method which will give as good a correlation as possible with crop response for the soils in a given region. The. chemical extraction methods so far developed for testing phosphorus in soils may be divided into: i. methods employing water and carbon dioxide saturated water, ii. acids, bases, salts and buffered solutions, . 32 iii. isotopic dilution with p tagged orthophosphate, iv. electrodialysis and ion exchangers. A critical review of the literature reveals that though acids, bases, salts, and buffered solutions etc. have been frequently used, water and carbon dioxide saturated water have been regarded with some reservations, even though this method is one of the oldest used methods and has proved reliable in many cases. University of Ghana http://ugspace.ug.edu.gh Hibbard (1931), Blenkinsop (1938), Burd and Murphy (1939), McGeorge (1939), Bray and Dickman (1942), Forsee (1945), and Bingham (1949) were among the early research workers who employed this technique with some success. The criticism levelled against the use of water as an extractant has been that water dissolves far too little phosphorus as compared to the amounts taken up by plants. Consequently the quantities of phosphorus found in water extracts do not provide a very good index of phosphorus availability to plants. Another major objection to the use of water as an extractant of soil available phosphorus determination is the frequent failure to obtain a clear extract. However- present-day instrumentation is such that the problems outlined above cease to be a serious impedi­ ment to the determination of soil available phosphorus using the water extraction method. Infact several of the contemporary workers in the field (for example Arnon, 195.3; Fried and Shapiro, 1956; Larsen et aJL. , 1959; Thompson et 'al. 1960; and Daughtrey et al., 1973) have evaluated this method and found it to be still useful. Fried and Shapiro (1956) and later Daughtrey et al., (1973) employed the water extraction method in studies on the phosphate supply pattern of some soils. 19 University of Ghana http://ugspace.ug.edu.gh A successive extraction with distilled water was used to study the intensity and the capacity of the soil to supply the soil solution with phosphorus. The capacity factor is regarded as a measure of the phosphate reserve in the soil and is related to the amount and forms of solid-phase phosphorus. The intensity factor, on the other hand, is the amount of phosphate in the soil solution at any given time. The reasoning behind the adoption of the successive water extraction was that the phosphate taken up by the plant roots from solution is continually replenis­ hed by the release of phosphate from the solid-phase. From a plot of P extracted against the extraction number they observed two different patterns of phosphorus release. In one case there was an increase in P removed with each successive extraction. In the other case there was a reduction in P removal with successive extractions. They suggested that both the intensity of soil phosphate supply and the capacity of the soil to replenish this supply must be carefully evaluated to be able to describe precisely plant available phosphorus in the soil. Oteng (1969) also concluded from studies on availability of phosphorus in Ghana soils that of the ten conventional methods employed only the water, Bray No.2, Morgan and University of Ghana http://ugspace.ug.edu.gh Olsen's method are significantly correlated with phosphorus uptake by millet. The water extractable phosphorus however, provided the highest correlation coefficient with phosphorus uptake. The acid extraction methods commonly employed in studies on soil available phosphorus can be divided into organic acid and inorganic acid methods. The organic acid extractions which have been proposed and still widely used are: i. the citric acid methods proposed by Dyer (1894), ii. the lactic acid method of Egner (1941), and iii. the acetic acid method of Hibbard (1931). The mineral acids used include the 0.7N HC1 by Olsen (1946), and 0.05N HC1 plus 0.025N H2S04 by Mehlich (unpublished). The mixture of 0.05N HC1 and 0.025N H2SO^, according to Nelson et al. (1953) is very effective in extracting a larger proportion of the difficultly available phosphorus. Other research workers have used buffered solutions of acids. Truog (19 30) used 0.002N H2S04 buffered at pH 3.0 with (NH^)2S0^. Morgan (1937) proposed acetic acid buffered near pH 4.8 with sodium acetate. Peech and English (1944) used acetic acid buffered at pH 4.8 with 21 University of Ghana http://ugspace.ug.edu.gh 22 ■ammonium acetate. Bray (1948) used hydrochloric acid buffered with NH^F. Ghani (1943) also used acetic acid in the presence of 8 hydroxyquinoline and Cooke (1951) employed 0.5N acetic acid at pH 2.5 mixed with various complexing agents. Alkali extractions have also been used primary on red soils. These include 0.-5M NaHCO^ used by Olsen et al., (1954), 0.5N NaOH recommended by Jones (1949). Others are the II potassium carbonate proposed by Das .(1930) . The numerous acid extraction methods that have been reported for assessing the available-phosphate status of soils have been adopted in the tropics. But Birch (1952) and Birch and Friend (1960) reported that crop responses were not significantly related to the amounts of acid ammonium fluoride soluble (O.OTN HC1, Bray and Kurtz, 1945) phosphate. Nye (1952) concludes that the conventional methods such as dilute acid extraction method using 0.03N NH^F solution and 0.025N HC1 for soil chemical analysis for available phosphorus have been generally disappointing -on tropical soils. University of Ghana http://ugspace.ug.edu.gh Piggot (1953) found that Truog (1945) and Purdue quick-test methods yielded no correlation with responses to superphosphate on a large number of trials on swamp rice in Sierra-Leone. In his recent studies, Stephens (1968) has also stated that the determination of available phosphorus by the Truog (1945) method was of no use in assessing the effects of superphosphate on crop yield in Uganda. Birch (1953) working in East Africa, has found that for certain soils and crops the lower the pH of the soil or the lower the percentage saturation of the exchange complex with bases, the greater the probability of a response to phosphate. On these soils he found there were no correlation between response and the amount of phosphate extracted by the usual solvents, which was usually quite high. Even in the temperate countries where the acid extraction methods have been known to give satisfactory results there have still been some reports of failure. Larsen et al., (1959) for instance, found that most of the acid extractants remove more phosphorus from mineral soil and sod muck than from virgin or deeply ploughed muck. It is evident from the work of Moser et al, (1959) „2 3 University of Ghana http://ugspace.ug.edu.gh and Thompson et al. (1960) that the phosphorus concentration in 0.01M CaCl2 gives a more reliable indication of the phosphorus uptake by crops, especially in the greenhouse, than the amount of phosphorus removed by the conventional extractants, eg., NH^F - HC1, citric, and lactic acid. ' 1.4.2. Biological Methods of Determination of Available Phosphorus. Biological methods for evaluation of soil phosphorus availability comprise the use of both higher plants and microplants. Biological methods involving growing plants in small quantities of soil in the greenhouse help to bridge the gap between soil analysis and field experiments. In this method, the comparative yields and/or uptake of plant nutrients from treated and untreated portions of soil are usually taken as a measure of plant nutrient status. The quantities of fertilizers or other materials to be added are usually calculated on the basis of pounds of soil in the jar. It is customary also to use a moisture content near the field capacity of the soil unless soil-moisture content is one of the problems involved in the study. In conse­ quence then, in extreme cases the jars are weighed 2'4 University of Ghana http://ugspace.ug.edu.gh and made up to weight everyday. A more common practice, wrote Millar (1965), is to weigh the cultures about once a week and calculate the daily loss of water. This amount of water is added each day. The extraction of mineral nutrients from soil by growing crops is a unique type of soil chemical analysis. Plant tissue analysis aids in the characterization of soil chemical properties in terms of soil fertility and mineral nutrition of plants. In many respects, the plant should be a good indicator of the soil environment as it tends to integrate all factors. It is an indisputable fact, however, that time and method of sampling, plant species, and weather will all affect plant, composition, yet from all indications the method should be preferred or help to correlate data for effective fertilizer recommendations or prediction of fertility of soils. The analysis of plants as a means of ascertaining the nutrient content of crops was undertaken early in the history of agricultural chemistry. According to Millar (1965), the method has been in use for almost 200 years now. Plant analysis was employed by early workers to establish many of the principles of plant nutrition. 25 University of Ghana http://ugspace.ug.edu.gh 2,6 It also received considerable attention as a method of approach to the practical problem of determining the availability of soil nutrients. De Saussure was reported by Ulrich (1943) to have adopted the biological method, in as far back as 1804, to analyze the ash of plants and observed that its composition varied with the soil, with the part of the plant, and with the age of the plant. Hall (1905) also analyzed the soil by means of the plant that can grow on it and concluded that " the proportion of phosphoric acid and of potash in the ash of any given plant varies with the amount of these substances available in the soil as measured by the response of the crops to phosphoric and potassium manures respectively Salter and Ames (1928) however, after due consideration of the problem of plant analysis as a diagnostic procedure concluded that so many factors influence the nutrient composition of the plant that use of plant analysis as a guide for evaluating fertilizer requirements of crops is precluded. In greenhouse experiments aimed at determining nutrient element availability by biological methods excess soluble salts may become a serious problem when large quantities of fertilizer are applied. This is because greenhouse soils are not exposed to natural leaching by rainwater. University of Ghana http://ugspace.ug.edu.gh Chlorides, sulphates, and nitrates are the major anions that are likely to contribute to excess soluble salt accumulation. Minor element shortages, as well as, excesses are also possible in greenhouse experiments. Phosphorus interaction resulting from heavy dressings of phosphatic fertilizers have been investigated to varying extents in many parts of the world, and particularly at Riverside, California, since 1953. Evidences in support of the above observations have been given by Bingham et al. (1956 and 1958) who reported that Ca(H2PO S0 4 ' 7 H 20, CaS04, NaCl and Fe SO^. Knop's solution, on the other hand, was made up of Ca(N03)2 .4H20, KN03, KH2P04 , Mg S04 .7H20, and FeP04 . University of Ghana http://ugspace.ug.edu.gh Following Sachs and Knop's initiative other workers in the field attempted to develop simpler nutrient solutions which would produce optimal growth. It is reported that American plant physiologists became particularly interested in plant nutrition in the years following 1900. Bonner and Galston (1952) , and Salisbury and Ross (1969) have reported that much work was performed by John Shive, W.R. Robbins, D.R. Hoagland, Daniel Arnon, A.L. Somner, Perry Stout, and Tottingham. These American workers and others, such as E.J. Hewitt, in England, improved the nutrient formulae given by Sachs and Knop. At present there are nutrient solutions of varied composition such as those of Pfeffer, Crones, Shive, Knudson, and Hoagland. Recent introductions, such as those of Arnon and Hoagland (1940), Hoagland and Arnon (1950), and Hewitt (1963) are very popular and are used for various greenhouse experiment. In addition to the foregoing Salisbury and Ross (1969) have also reported that the concentration in ppm of micronutrients in nutrient solutions which could be regarded suitable for many aqueous solution culture lie, depending upon the plant species, within the following ranges: Fe (0.5 to 5.0) Mn (0.1 to 0.5), B(0.1 to 1.0), Zn(0.02 to 0.2), 29 University of Ghana http://ugspace.ug.edu.gh Cu(0.01 to 0.05) and Mo (0.01 to 0.05). Higher concen­ trations than these they claim, are often toxic to many plants. The aqueous solution culture technique initiated by Sachs and Knop was later modified to involve soil samples. This pot culture technique, using plant samples to extract nutrient elements from soils and also to study the availabi­ lity of nutrient elements to plants, was proposed as far back as 1909 by Mitscherlich and, later by Neubauer in 1923. In Mitscherlich's (1909) procedure reported by Vandecaveye (1948) nutrient solution, containing N as NH^NO^j K^O as K^SO^, P2O5 as super-phosphate, and NaCl and MgSO^ was added to soil samples in pots in a case where a " complete fertilizer" treatment was required. In later work by other investigators such as Schuster and Stephenson (1940), Stephenson and Schuster (1941), Colwell (1943), and Jenny et al. (1950), just to mention only a few, nutrient solutions to supply Ca, Mg, S, Cu, Zn, B, Mn, Mo, in addition to the major nutrient elements N,P,K, were added to the soil samples in varied composition to suit the specific work envisaged. For instance, in a work by Stephenson and Schuster (1941) a nutrient solution to supply the following K, 350 ppm; P, 217 ppm; Mg, 168 ppm; S, 224 ppm; N, 200 ppm 30 University of Ghana http://ugspace.ug.edu.gh •was added to the soil samples. In greenhouse experiments using sand, gravel, soil, or sand-soil mixture culture, nutrient solution is applied in one of three ways: i. slop culture technique, ii. drip culture technique, and iii. subirrigation technique. In slpj> culture nutrient solution is periodically supplied to the surface of the culture and allowed to seep through, where as in drip culture nutrient solution is continuously dripped onto the culture. In subirrigation the nutrient solution is pumped from a bottom reservoir up through the culture or is added through a tube inserted in the culture until the solution reaches the surface. The slop culture technique, writes Bonner and Galston (19 52) is the simplest way to grow plants under conditions of controlled nutrition. Salisbury and Ross (1969) report that most greenhouse experiments can be safely conducted in borosilicate glass containers, polyethylene beakers or buckets, and in pyrex galss containers. These authors, however, suggest that borosilicate glass containers provide small amounts of boron whilst polyethylene containers supply sufficient zinc for plant growth. 31 University of Ghana http://ugspace.ug.edu.gh Pyrex glass containers, Salisbury and Ross (1969) profess, are usually very suitable. Salisbury and Ross (1969) write that certain seeds provide sufficient amounts of nutrient elements, especially the micronutrients Cu, Mo, and Zn, for the entire growth and reproductive period of greenhouse plants. They are therefore of the opinion that plants having large seeds are able to supply elements in sufficient quantities to developing seedlings than those having small seeds and hence recommended small seeds for greenhouse experiments. After sampling, plant material is: usually subjected to four different preparative steps before the actual chemical analysis is carried out: i. cleaning the material to remove surface contamination, ii. drying to stop enzymatic reactions and prepare the material for grinding, iii. mechanical grinding to reduce the material to a fineness suitable for analysis, and iv. final drying to constant weight to obtain a standardized value on which to base the analytical figures. Plant parts selected for sampling are always covered with a thin film of dust which is very difficult to remove ii University of Ghana http://ugspace.ug.edu.gh by mechanical wiping or brushing. Anyway failure to remove dust normally affects only Fe unless the dust cover is thick or of specific composition (Jones and Steyn, 1973). Steyn (1959) showed that a satisfactory way of removing contamination is by washing the tissue in 0.1 to 0.3! detergents solution followed by rinsing in pure water. After washing, plant tissue samples should be dried as rapidly as possible so as to minimize chemical and biological changes. If drying is unduly delayed, considera­ ble loss in dry weight may occur due to respiration (Locknxan, 1-970) , while proteins are broken down to simpler nitrogenous compounds. Also too highdrying temperature can affect the dry weight (Grant and MacNaughlain, 1968) . According to Tauber (1949) enzymaction is reduced or stopped if plant material is heated to above 60°C. Jones and Steyn (1973) however, recommended heating in a forced - draft oven set at 65°C. Customarily, dried plant material are ground before analysis, partly for greater ease in manipulation, partly to ensure greater uniformity in composition. Because of the laborious nature of hand grinding, particularly when samples are large, mechanical grinding in mills is 33 University of Ghana http://ugspace.ug.edu.gh favoured by most workers. Jones and Steyn (1973) have recommended storage of the tissue powder in a clean, dry bottle. Further drying for an additional 24 hours at 65°C in order to remove the moisture picked up during grinding is also recommended before sub- samples may be weighed out for analysis. 1.5. Assessment of P Requirements; 'of Soils, from P Sorption. Many researchers the world-over have adopted numerous techniques designed to estimate the P requirements of soils for optimum crop yields. These invariably include field trials and laboratory determinations. The field trials are slow, laborious and the results are useful only for local applicability. Amongst the laboratory estimations the methods of Bray and Kurtz (1945), Olsen et al. (1954), and Bingham (1962) are noteworthy. The above laboratory methods give some measure of both the amount of phosphate already present in the soil and its degree of availability. However, they fail to satisfactorily indicate the amount of phosphate that will be required to give optimum crop yields in different soils Since the amount of phosphate needed to be applied to soils to give an optimum production is of more concern University of Ghana http://ugspace.ug.edu.gh 35- than the existing supply, a more direct method of assessing plant P needs was desired. In response to this world-wide desire for a method of assessing plant P needs in soils, Ozanne and Shaw (1967) tried a method for the direct assessment of plant P needs through phosphate sorption by soils. In the study of Ozanne and Shaw (1967) the amount of phosphate sorbed at equilibrium concentration of 0.3 ppm P was measured in a preliminary investigation. They subsequently examined the extent to which the above measurement would allow the estimation of the optimum amount of phosphate that need be applied to obtain near-maximum yields. From the data accumulated from their investigation they concluded that the measurement of phosphate sorption may be adequate to predict the phosphate requirements of plants. Following Ozanne and Shaw's (1967) initiative Singh et al., (1971) adopted a similar technique to determine phosphate requirement of soil for cereal crops. Singh et al.,(1971) also observed that in any programme of P application for getting optimum crop responses and yields, due consideration should be given to the capacity of soil for sorption of P rather than the initial P status. University of Ghana http://ugspace.ug.edu.gh CHAPTER 2 MATERIALS AND METHODS 2.1. Description of Soils. Soil samples used in the present study were taken from the Ocm to 23 an. layer of twelve soil series of Ghana. Table 1 provides a description of the soils as follows: series name, great soil group classification, type of parent material, textural classification and vegetation associated with the soils. Table 1. Description of Soils Used. 36 Identifi­ cation No. of Soil Soil. Series Name Great Soil Group Parent Material Textural Classifi­ cation Vegetation 1 Abenia Forest Qxysol Biotite Granite Schist Sandy Clay Forest 2 Ankasa Forest Qxysol Biotite Granite Sandy Clay Loam Forest 3 Boi Forest. Qxysol Phyllite Sandy Clay Loam Forest 4 Tikobo Forest Oxysol Tertiary Sand Loamy Sand Forest Re­ growth 5 K'dua Forest Oxysol Rubrisol In­ tergrade Biotite Granodi- orite Sandy Clay Loam Forest 6 Wacri Forest Ochro- sol Rubrisol Intergrade Hornblende Grano- diorite Sandy Clay Loam Forest 7 Mamfe Forest Ochro- sol Quartzite Gra­ nite Sandy Loam Forest 8 Oyarifa Savanna Och- rosol Sandstone Sandy Loam Thicket 9 Toje Savanna Och- rosol Tertiary Sand Sand Tall Grass Savanna 10 Akuse Tropical Black Clay Hornblende Gneiss Clay Loam Savanna 11 Prampram Tropical Black Earth Basic Gneiss Sandy Clay Loam Tall Grass Savanna 12 Agawtaw Tropical Grey Clay Acid Gneiss and . Schist Loamy Sand Savanna University of Ghana http://ugspace.ug.edu.gh Table 2 presents further information on the soil as determined in the laboratory during the present study. The information includes values for pH, organic carbon, clay content, silica content, "free" aluminium oxide (A1203) and iron oxide (Fe^^) of the soil. 37 Table 2. Some Properties of ’Soil's' Used. Identi­ fication No. of Soil Soil Series Name pH (in O.OM CaCl2) : % Org. Carbon: Clay • Content % Silica Content % ' "Free" M 2°3 % "Free" Fe2°3ft0 1 Abenia 4.00 1.51 35.92: 20.30 0.74 0.35 2 Alikas a 4.00 1.20 23.20 24.44 0.48 0.30 3 Boi 4.15 0.99 32.87’ 33.86 1.95 0.27 4 Tikobo 4.00 0.77 12.44 39.72; 0.74 0.25 5 Koforidua 6.95 1.41 22.53 23.4Z 7.04 0.20 6 Wacri 6.35 1.18 25.39 30.08 6.58 0.20 7 Manife 4.60 1.85 14.30 41.54 0.34 0.22 8 Oyarifa 6.40 0.85 19.86 ; 42.78 0.29 0.15 9 Toje 5.35 0.20 6.39 54.26 0.25 0.11 10 Akuse 6.60 0.79 32.53 ' 33.46 11.26 0.18 11 Prampram 7.12 0.79 : 34.48 32.78 13.20 0.31 12 Agawtaw 5.90 0.34 8.00 55.50 3.44 0.09 University of Ghana http://ugspace.ug.edu.gh The selected soils have pH values ranging from 4.00 to 7.12. The pH values for the soil series Abenia, Ankasa, Boi and Tikobo are in the range 4.00 to 4.15. These are' soils from high rainfall (70" or more/1780 mm or more) areas where normally there is pronounced through leaching of cations Ca2+,Mg2+,K+, and Na+ from the topsoil deep down into the profile thus resulting in acid reaction. Consequently the low pH values (well in the acid range) obtained for these soils were as expected. Mamfe series, a soil from a forest area with moderate rainfall 50" to 60" or 1270 mm to 1520 mm also has pH of 4.60 which is very acid. This particular soil series is believed to have developed over Quartzite Granite (Table I) and therefore not much soluble bases are released to the soil consequent upon weathering. Hence the pH value cannot be expected to fall in the slightly acid or near neutral range as might be the case of many soils from regions with moderate rainfall. Toje and Agawtaw series which are soils from savanna region have moderately acid reaction. These soils are underlain by inert rock materials mainly. Infact Toje is believed to have developed over Tertiary Sand whilst Agawtaw is formed over Acid Gneiss and Schist, all of which release only minute quantities of bases to the soil when weathered. 38 University of Ghana http://ugspace.ug.edu.gh t The soil series Koforidua and Wacri, which are from forest areas with moderate rainfall (50" to 60" or 1270 mm to 1520 mm per annum) have pH value which are slightly acid (pH 6.35 to 6.95). These soils are subjected to only sporadic leaching of their soluble bases and are therefore expected to show only slightly acidic reaction. Akuse and Prampram soil series have also pH of 6.60 and 7.12 res­ pectively. These soils are formed over basic parent material, therefore, the pH of near neutral and neutral is as expected since more basic materials are released when the underlying rocks do weather. As expected, Abenia, Ankasa, Boi, Koforidua, Wacri, and Mamfe, which are soils from forest region, have relatively high organic carbon content. Tikobo, however, has a comparatively low organic carbon content of only 0.771. This particular soil series has a sandy loam texture and there is the possibility of eluviation of organic matter from the topsoil into the subsoil. In contrast, the savanna soils, namely, Oyarifa, Toje, Akuse, Prampram, and Agawtaw have relatively low organic carbon content ranging from 0.2% in Toje to 0.851 in Oyarifa. This reflects on the vegetative cover and periodic burning of grasses in the savanna areas in Ghana. 39- University of Ghana http://ugspace.ug.edu.gh The clay content appears to be fairly high in the soil series Abenia (35.92% clay) and Prampram (34.48% clay). These soils have been derived from parent materials rich in basic minerals and which are less resis­ tant to weathering and as such weather easily to give clay particles to these soils. On the other hand, those soils derived mainly from sand, quartzite granite and acid gneiss such as Tikobo, Mamfe, Oyarifa, Toje, and Agawtaw have comparatively low clay content. The silica content ranges from 20.30! in Abenia to 55.50! in Agawtaw. Abenia, Ankasa, Boi, Koforidua, Wacri, Akuse, and Prampram soils with high clay content also have comparatively low silica. On the contrary, Tikobo, Mamfe, Oyarifa, Toje and Agawtaw which show low clay content have, in turn, high silica content. The "free" aluminium oxide content of the soils used is in the range 0.25! to 13.20!. Infact the "free" aluminium oxide content follows no particular trend. Akuse and Prampram soils have high "free" aluminium oxide of 11.26! and 13.20! respectively. The normal range of A1203 content of soils generally is 2 - 15!. Koforidua and Wacri also have moderately high "free" aluminium oxide content of 7.04! and 6.58! respectively. Abenia, Ankasa, Boi, Tikobo, Mamfe, Oyarifa, Toje, and Agawtaw soils, on 40 University of Ghana http://ugspace.ug.edu.gh .the other hand, have low "free" aluminium oxide content ranging from 0.25% to 3.44%. The "free" iron oxide content as Fe20 3 of the soils used in the investigation is low ranging from 0.09% in Agawtaw to 0.35% in Abenia as compared to the normal range of F©2^3 content of soils which generally is 0.1 to 8 .0 %. 2.2. Preparation and Storage of Soil Samples. The soil samples which were taken from uncultivated sites were air-dried, ground with a wooden pestle and mortar and sieved through a 2 mm. sieve with square holes to get only the "fine earth" samples. The sieved soil samples were stored in polythene bags. 2.3. Determination of the Adsorption Maximum. The method of Olsen and ffatanabe (1957) was used to obtain data for plotting the Langmuir isotherm. Subsamples of the fine earth fractions were sieved with a 72 mesh (0.211 mm) screen. The rationale behind the resieving was to obtain subsamples which contain largely clay and silt fractions. This was deemed necessary since adsorbed phosphorus can be found on the clay and silt fractions and hardly on the sand fraction. The subsamples thus obtained were then stored in wax-coated paper containers tfor the analytical tests. 41- University of Ghana http://ugspace.ug.edu.gh ■Five grams samples of the 72-mesh soil samples were weighed into 125-ml. polypropylene extraction bottles and shaken in 100 ml. of KH2P04 solutions for 24 hours. Seven different equilibrations were made with KfiLPO. solutions of the following concentrations: lXiQ4M, 2X10 4M, 3X10 4M, 7X10"4M, 9X10~4M and 12X10-4M. All KH2P04 solutions were adjusted to pH 7.0 initially. Forty millilitres aliquots of the soil suspension were centrifuged in "Sorval" superspeed angle centrifuge at 7,000 rpm. for twenty minutes, and the supernatant solutions were decanted into separate tubes. Ten millilitres aliquots of the supernatant solutions were pipetted into 100 ml. volumetric flasks and then made up to volume with distilled water. The phosphorus concentration was determined by a modified Truog and Meyer (1929) method. Suitable aliquots of the diluted clear extracts were pipetted into matched test tubes calibrated at 45-ml. volume and 2 ml. acid ammonium molybdate solution were added. The mixtures were then made up to mark with distilled water and were thoroughly shaken. Three drops of stannous chloride reductant solution were added, the mixtures were shaken again and the phosphorus content was measured with a Bausch and Lomb "Spectronic-20" spectrophotometer at 660 millimicron wavelength in exactly ten minutes. 42' University of Ghana http://ugspace.ug.edu.gh .43 The difference between the amount of phosphorus in solution after shaking and the amount initially present was taken as the amount of phosphorus adsorbed by the soil from the KF^PO^ solution. The adsorption maxima of the soils were calculated from the reciprocal of the slope of the straight-line obtained from a plot of the equilibrium phosphate concentration expressed as 4 -1C x 10 moles litre against the equilibrium phosphate concentration C x 10^ in moles litre divided by the milligrams of phosphate adsorbed per 100 g of soilT -1(C/(x/m) ) in moles litre 2,4. Determination of "Free" Pe and Al Oxides. An estimate of the "free" iron and aluminium oxides in the soils was obtained using a modified ammonium oxalate extraction technique based on the method of Tamm (1922). One gram of soil sample, previously screened through a 2 mm. sieve, was weighed and transferred quantitatively into an extraction bottle. One gram of sodium dithionite was added followed by 40 ml. Tamm solution ( a mixture of oxalic acid and ammonium oxalate). The extraction bottle was heated to a temperature of 8 5°C in an oven for twenty minutes, swirling the bottle after five minutes. University of Ghana http://ugspace.ug.edu.gh The suspension was centrifuged and the supernatant solution was decanted through a filter paper into a 200-ml. volumetric flask. The above extraction was repeated twice. The residue was then washed twice with 25 ml saturated sodium chloride solution. The extract was diluted to the 2 0 0-ml. mark and mixed thoroughly. Five millilitres aliquots of the extracts were pipetted into 50-ml. Kjeldahl flasks and 2.5 ml. aqua regia were added. The mixture was heated on a Kjeldahl microdigestion rack until nitrous oxide vapour ceased to escape. The flask and its contents were cooled. Five millilitres aqua regia were added to the contents of the flask and heated again to dryness. This latter treatment was repeated. Twenty millilitres of water and 1.0 ml. of 4N hydrochloric acid were then added to the contents of the flask and heated gently for ten minutes. The flask and its contents were Cooled and transferred quantitatively into a 50-ml. volumetric flask and the volume adjusted with water. Suitable aliquots of the extract were pipetted for the colorimetric determination of iron and aluminium. The iron content of the extract was determined colorimetrically by Olsen, R.V. method outlined in Black et. al. ed. "Methods of Soil Analysis", Agronomy No.9, Part 2, Chemical and Microbiological Properties, 1965. ,44 University of Ghana http://ugspace.ug.edu.gh A suitable aliquot, generally 1.0 ml., of the test solution was pipetted into a 50-ml. volumetric flask and then 2ml. of 5N ammonium acetate and 1.0 ml. of 101 hydroxylamine hydrochloride were added. The solution was mixed and 1.0 ml of orthophenanthroline reagent and 0.5 ml of 6N hydrochloric acid were added. The solution was diluted to volume and mixed thoroughly. The coloured test solutions were transferred to photometer tubes and placed in a Bausch and Lomb "Spectronic-20" spectrophotometer using a wave-length setting of 510 millimicron. The galva­ nometer was set to 1001 light transmission with the blank solution. The aluminium concentration of the above extract was also measured colorimetrically using pyrocatechol violet indicator. A suitable aliquot, generally 10 ml., of the extract was pipetted into a 50-ml. volumetric flask. One millilitre of hydroxylamine hydrochloride and 1.0 ml. of O-phenanthroline solution were added one after the other to the contents of the flask. Next 1.0 ml of pyrocatechol solution was also added. Twenty-five millilitres of a buffer solution, made up of aqueous ammonium acetate solution adjusted to pH 6.2 with acetic acid, were added followed by a drop of ammonia solution. The mixture was then diluted to the 50-ml. mark and set aside for two hours. 45 University of Ghana http://ugspace.ug.edu.gh -The percentage of light transmittance was measured on a Bausch and Lomb "Spectronic-20" spectrophotometer at 580 millimicron wavelength using blank solution to adjust the galvanometer to 1001 transmission. 2.5. Determination of Silica Content of Soils. The silica content of the soils- was determined by a modified method of Corey and Jackson (1953), and Shapiro and Brannock (19 56) outlined in Black et al. ed. "Methods of Soil Analysis", Agronomy No.9, Part 2, Chemical and Microbiological Properties, 1965. Ten millilitres of 15% sodium hydroxide solution measured with a plastic graduated cylinder were transferred into a platinum crucible, and evaporated the solution to dryness on a hot plate. A sample of 0.05 g of 100-mesh soil sample was placed into the crucible. The crucible was covered and heated to dull redness for about five minutes in a muffle furnace. The melt was allowed to cool, and approximately 15 ml. of water were added and then allowed to stand overnight. The content of the crucible was transferred to a 600-ml. beaker containing about 400 ml. of water and 20 ml. of 6N hydrochloric acid. The crucible was scrubbed well with a rubber policeman and the remaining residue washed into the beaker. The solution was finally transferred to a one-litre volumetric flask and adjusted with distilled water to volume. 46 University of Ghana http://ugspace.ug.edu.gh fen millilitres of the test solution were pipetted and transferred to a 100-ml. volumetric flask. One millilitre of ammonium molybdate reagent was added, swirling the contents of the flask to mix the solution well. One millilitre of reducing solution made up of sodium sulphite, 1-amino-2-naphthol-sulphonic acid and sodium bisulphite, was added and the solution was diluted to the 1 0 0-ml. mark, mixed well and allowed to stand for about thirty minutes. The coloured test solutions were transferred to photometer tubes and placed in a Bausch and Lomb "Spectronic-20" spectrophotometer using a wavelength of 650 millimicron. The galvanometer was earlier on set to 1 0 0% light transmission with the blank solution. 2.6. Estimation of Organic-C, pH and Clay content of Soils. The percent organic carbon content of the soils was estimated by the wet oxidation method of Walkley and Black (1934). A suitable amount of soil samples previously passed through a 0.5 mm sieve, generally 2 g samples, was weighed into a 500-ml. Erlenmeyer flask and to which was added 10 ml. of potassium dichromate solution from a burette. Twenty millilitres of concentrated sulphuric acid were added. The flask was swirled and allowed to stand for thirty minutes. 200 ml. of distilled water were then added followed by 10 ml. of orthophosphoric acid. Three drops of diphenylamine indicator were added and University of Ghana http://ugspace.ug.edu.gh titrated against ferrous ammonium sulphate solution to a green end-point. A blank was run in the same way as described above. The percent organic carbon was calculated from the titration readings. The pH of the soils was measured in 0.01M calcium chloride solution with a WG-Pye glass electode pH-meter on a 1:2 soil to solution ratio. Twenty grams of the 2 mm. soil samples were weighed into a 50-ml beaker and added 40 ml. of 0.01M calcium chloride solution. The suspension was stirred several times during a period of thirty minutes and then allowed to stand for another thirty minutes. The electrode was immersed and the pH was measured. Mechanical analysis to estimate the per cent clay of the soil samples was done by the pipette method. Approximately 10 g of the 2 mm soil samples were weighed into tared 250-ml. beaker and reweighed to the nearest 0.01 g . The samples were dried overnight at 105°C, cooled in a desiccator, and reweighed. About 30 ml. of water were added followed by a few millilitres of 30% hydrogen peroxide, then covered the beaker with a watch glass and stirred the contents by swirling the beaker. When the reaction had subsided additional amounts of hydrogen peroxide were added and then completed the digestion by heating the beaker for one hour on a hot-plate. 48 University of Ghana http://ugspace.ug.edu.gh Other beakers containing the soil suspension of each soil sample was placed in an oven at 105°C for about 24 hours, cooled in a desiccator, and weighed the beaker and its contents to get the weight of the organic matter-free soil samples. Twenty-five millilitres of calgon were added to the organic matter-free soil suspension and trans­ ferred into a 250-ml shaker bottle. Some water was added to the suspension to bring it to about 150-ml. volume, the bottle was stoppered and shaken for 4 hours in a reciprocating shaker. A wide-mouth funnel was put in a 500-ml. graduated cylinder and a 60 micron sieve was placed on the funnel. The partly settled suspension was poured into the sieve. The remaining soil practicles in the cylinder were washed into the funnel with a jet of water and finally washed the sand particles on the sieve with a jet of water. The sand particles were transferred•into a beaker, oven-dried, and weighed. The suspension was made up to the 500-ml. mark with water and transferred the cylinder to the sedimentation cabinet. A plunger was inserted in the suspension and moved it up and down to mix the contents thoroughly. The cylinder was moved into position in the pipette stand, then clamped the clean, dry; 1 0-ml. pipette in its holder 4.9 University of Ghana http://ugspace.ug.edu.gh and attached the tubing. Three samplings of the suspension were made at 4 minutes, 45 minutes and 6 hours 45 minutes time at the 10 cm. depth. The samples of the clay and silt suspension were poured into tared aluminium boxes, oven-dried, and weighed. The per cent clay in the soil samples were calculated from the readings recorded. 2.7. Determination of Available Phosphorus in the Soils. The initial phosphorus status of the soils used in the investigation was estimated by the Bray and Kurtz (1945) method. A 2.85 grams sample of crushed, sieved soil was weighed out into a 125-millilitres polypropylene extraction bottle. Twenty millilitres of extraction solution (0.03N NH^F in 0.02 5N HC1) were added from a pipette and the bottle was shaken for five minutes. The suspension was centrifuged and filtered through Whatman No. 42 filter paper. A two millilitres aliquot of the clear filtrate was pipetted into colorimeter tubes. Then five millilitres of distilled water and two millilitres of ammonium molybdate reagent were added in succession. The solution was thoroughly mixed on a rotary mixer. One millilitre of freshly diluted stannous chloride reagent was added and the intensity of the blue colour 50 University of Ghana http://ugspace.ug.edu.gh was measured after five to six minutes and before fifteen to twenty minutes on a Bausch and Lomb "Spectro- nic-20" colorimeter at 660 millimicron wavelength. The P standards were made in the range of 0.1 to 1.0 ppm of P through the same steps as in the above procedure. Two millilitres of extraction solution were added to each aliquot of diluted standard P solution and the final solution was of ten millilitres volume. A reagent blank was prepared and was employed for the 100 per cent transmission setting. 2.8. Greenhouse Experiment. In a greenhouse experiment phosphorus was added to ten of the twelve soil series used in the adsorption maximum studies as KF^PO^ and I^HPO^ at various rates of zero, g, 1, I, 1 and 2 times the P adsorption maximum. The soil series used were 2, Ankasa; 3, Boi; 4, Tikobo; 5, Koforidua; 6 , Wacri; 7. Mamfe; 8 , Oyarifa; 9, Toje; 10, Akuse; 12, Agawtaw. 2.8.1. Amounts of Kl-^ PO/j and ^HPO,] Applied to Soils. The following is a sample calculation, using the Ankasa series, adopted to estimate the quantity of phosphorus needed to be applied to the soil samples in the pot experiment. The P adsorption maximum for the Ankasa series is 35.714 milligrams P per 100 grams of soil or 0.0357 grams per .100 grams of soil. 51 k University of Ghana http://ugspace.ug.edu.gh But 454 grams of soil sample was used in the pot experiment. If 100 grams of soil sample required 0.0357g. P 454 grams of soil sample will require (0.0357)(454) g.P 100 = (0.0357)(4.54)g.P = 0.1620 grams P. Thus the amount of P needed to be applied to 454 grams soil sample of Ankasa series to attain the P adsorption maximum is 0.1620 grams P. To obtain the amount of P to be applied at |, \ and 2 times the P adsorption maximum the value 0.1620 grams was multiplied by j, \ and 2, respectively. The table below presents the amount of P applied to the soil samples at the varied rates of §, J, 1 and 2 times the P adsorption maximum of the various soil series used in the investigation. The following is a sample calculation, using the Ankasa series, adopted to estimate the amount of KH2P04 required to be added to the soil samples. To attain the adsorption maximum 0.1620 grams P was required. In the KH2PO4 salt there is an atom of P. 52 University of Ghana http://ugspace.ug.edu.gh Table 3. Amounts of P .Applied to the Soils. Soil Series Name P Adsorption Maximum mg/lOOg Soil Phosphorus Application Rates 1s I 12 1 2 18 i4 12 1 2 g.P per pot (4S4g. soil) kg. P per hectare Ankasa 35.714 0.0203 0;0405 0.0810 0.1620 0.3240 100 200 400 800 1600 Boi 31.250 0.0178 0.0355 0.0709 0.1418 0.2836 88 175 350 700 1400 Tikobo 27.027 0.0154 0.0307 0.0614 0.1227 0.2454 76 152 303 606 1212 Koforidua 27.778 0.0158 0.0315 0.0630 0.1260 0.2520 78 156 311 622 1244 Wacri 25.641 0.0146 0.0291 0.0582 0.1164 ,0.2328 72 144 287 575 1150 Mamfe 30.303 0.0175 0.0344 0.0688 0.1376 0.2752 85 170 340 680 1359 Oyarifa 23.256 0.0132 0.0264 0.0528 0.1056 0.2112 65 130 261 521 1043 Toje 25.641- 0.0146 0.0291 0.0582 0.1164 0.2328 72 144 287 575 1150 Akuse - 31.250 0.0178 0.0355 0.0709 0.1418 0.2836 88 175 350 700 1400 Agawtaw 26.316 0.0149 0.0299 0.0598 0.1195 0.2390 74 148 295 590 1180 University of Ghana http://ugspace.ug.edu.gh .54 The atomic weight of P is 30.98 grams and the molecular weight of KH2P04 is 136.09 grams. If 30.98 g. P are supplied by 136.09g. KK^PO^ 0.1620ff. P will be supplied by (136.09)(0.162(^.KHoP0^ 30.98 z 4 = 0.7116g. KH2P04 The amount of KH2P04 needed to be applied to 454 grams soil sample of Ankasa series to attain the adsorption ma­ ximum is 0.7116 grams. To obtain the amount of KH2P04 salt needed to be added to attain |, |, \ and 2 times adsorption maximum the value 0.7116 grams was multiplied by I, \ and 2 , respectively. Table 4 presents the amounts of KH2P04 salt added to the soil samples to get the respective amounts of P at I, {, 1 and 2 times the P adsorption maximum. The following is yet another sample calculation, using Ankasa series, adopted to estimate the amount of K2HP04 needed to be added to the soil samples. The amount of P required to be applied to attain P adsorption maximum on Ankasa series is 0.1620 grams P per pot of 454 grams soil. In the K2HP04 salt there is an atom of P. Atomic weight of P is 30.98 grams and the molecular weight of K2HP04 is 174.18 grams. University of Ghana http://ugspace.ug.edu.gh Table 4. Amounts of KJ^PO^ Salt Added to Soils. Mounts P-.o ^Added Equiv. Amounts of K Supplicid oOll Series Name 1 8 14 12 1 2 18 14 12 1 2 grams per pot (454g. soil) Ankasa 0.0890 0.1779 0.3558 0.7116 1.4232 0.0256 0.0512 0.1023 0.2045 0.4090 Boi 0.0779 0.1557 0.3115 0.6229 1.2458 0.0224 0.0448 0.0895 0.1790 0.3580 Tikobo 0.0674 0.1348 0.2695 0.5390 1.0780 0.0194 0.0388 0.0775 0.1549 0.3098 Koforidua 0.0692 0.1384 0.2768 0.5535 1.1070 0.0199 0.0398 0.0795 0.1590 0.3180 Wacri 0.0639 0.1278 0.2557 0.5113 1.0226 0.0184 0.0368 0.0735 0.1469 0.2938 Mamfe 0.0756 0.1511 0.3023 0.6045 1.2090 0.0218 0.0435 0.0869 0.1737 0.3474 Oyarifa 0.0580 0.1160 0.2320 0.4639 0.9278 0.0167 0.0334 0.0667 0.1333 0.2666 Toje 0.0639 0.1278 0.2557 0.5113 1.0226 0.0184 0.0368 0.0735 0.1469 0.2938 Akuse 0.0779 0.1557 0.3115 0.6229 1.2458 0.0224 0.0448 0.0895 0.1790 0.3580 Agawtaw 0.0656 0.1312 0.2625 0.5249 1.0498 0.0189 0.0377 0.0754 0.1508 0.3016 OlUl University of Ghana http://ugspace.ug.edu.gh If 30.98g. P are supplied by 174.18g. K2HP04 0.1620g. P will be supplied by C174.18) (0.1620)g.K-^HPO^ 30.98 = 0.9108 grams K^HPO^. The amount of iUHPO^ salt needed to be applied to 454 grams soil sample of Ankasa series to attain the adsorption maximum is 0.9108 grams. To obtain the amount of K^HPO^ needed to be added to attain |, \ and 2 times adsorption maximum the value 0.9108 grams was multiplied by § , \ and 2 respectively. Table 5 presents the amounts of JUHPO^ salt added to the soil samples to obtain the respective amounts of P at I, I, 2 > 1 and 2 times P adsorption maximum. Examination of the last five columns of Tables 4 and 5, headed "equivalent amounts of K supplied" reveals that the KH^PO^ and K^HPO^ salts added to attain 2 times adsorpt maximum on the Ankasa series supplied the highest amounts + + of K ion. The K ion supplied are equal to 0.4090 grams for KH^PO^ and 0.8180 grams for J^HPO^ per pot of 454 grams soil. Since P was supposed to be the only nutrient element whose amount in the soil samples was to be varied, it was deemed necessary to add another salt which would supply the additional K+ ion. Potassium carbonate (K^CO^) salt was accordingly added in various amounts to the soil samples to bring the added K to a level of 0.4090 grams for KH2P0 4- treated soils and 0.8180 University of Ghana http://ugspace.ug.edu.gh Table S. Amounts of K2HP04 Salt Added to Soils. Soil Series Name Amounts of K2HP04 Added Equiv. Mounts of K Supplied i 8 i 4 1 2 1 2 t B 1 4 1 2 1 2 grams per pot (454g. Soil) Ankasa 0.1139 0.2277 0.4554 0.9108 1.8216 0.0511 0.1022 0.2045 0.4090 0.8180 Boi 0.0997 0.1993 0.3986 0.7972 1.5944 0.0448 0.0895 0.1790 0.3580 0.7160 Tikobo 0.0862 0.1725 0.3450 0.6899 1.3798 0.0388 0.0775 0.1549 0.3098 0.6196 Koforidua 0.0886 0.1771 0.3542 0.7084 1.4168 0.0398 0.0795 0.1590 0.3181 0.6362 Wacri 0.0818 0.1636 0.3272 0.6544 1.3088 0.0368 0.0735 0.1470 0.2939 0.5878 Mamfe 0.0967 0.1934 0.3868 0.7736 1.5472 0.0435 0.0869 0.1737 0.3474 0.6948 Oyarifa 0.0742- 0.1484 0.2969 0.S937 1.1874 0.0334 0.0667 0.1333 0.2666 0.5332 Toje 0.0818 0.1636 0.3272 0.6544 1.3088 0.0368 0.0735 0.1470 0.2939 0.5878 Akuse 0.0997 0.1993 0.3986 0.7972 1.5944 0.0448 0.0895 0.1790 0.3580 0.7160 Agawtaw 0.0840 0.1680 0.3360 0.6719 1.3438 0.0379 0.0757 0.1514 0.3028 0.6056 University of Ghana http://ugspace.ug.edu.gh grams for K2HPO'4~ treated soils per pot of 454 grams soil. The only exception was the soil samples of the Ankasa series which received the highest P treatment of 2 times P adsorption maximum. The various amounts of K2C0 3 salt added to the KH2P0 4- treated soils are presented in Table 6 whilst those added to the K2HP04~ treated soils are presented in Table 7- 2.8.2. Incubation Technique. Weighed quantities (454 grams) of the soil series under investigation were mixed with KH2P:0 4 or K2HP0 4 salts, as the case may be, and K^CO? salt in amounts as given in Tables 4, 5, 6 and 7. Each treatment was replicated twice. The soil samples and the inorganic salts were thoroughly mixed using a "Kenwood" domestic mixer. The soil samples with the added salts were then watered daily with distilled water for two weeks. 2.8.3. Seeding of Soil Samples with Test-Crop. After the two weeks incubation period the dried soil samples were pulverised and thoroughly mixed with a "Kenwood" domestic mixer. Each 454-grams soil sample was mixed with 227 grams of acid- treated beach sand (ratio of soil to sand was 2:1). About 150 grams of thoroughly washed quartz gravel were weighed into empty plastic pot of 1340 millilitres capacity. A plastic tube measuring 11.5 centimetres long and 1.10 centimetres in diameter was placed on the quartz gravel at an 58 University of Ghana http://ugspace.ug.edu.gh 59 Table 6. Amounts of K2C03 Salt Added to KH2P04-Treated Soils. I Soil Series Name t ........... i "f, , i t ' ’ « P Saturation of Adsorption Maximum 0 18 1 4 12 1 . 2 grams K^CO^ Salt Added Ankasa 0.7226; 0.6774: 0.6322 : 0.5419 0.3612 Nil Boi 0.7226: 0.6831: 0.6435 0.5645: 0.4063: 0.0901 Tikobo 0.7226: 0.6884 0.6558 0.5857 0.4489 0.1753 Koforidua 0.7226 0.6875 0.6523 0.5821 0.4416 0.1606 Wacri 0.7226. 0.6901 0.6576 0.5928 0.4630 0.2034 Mamfe 0.7226: 0.6841 0.6458 0.5691 0.4157 0.1089 Oyarifa 0.7226: 0.6931 0.6636 0.6048 0.4871 0.2515 Toje 0.7226. 0.6901 0.6576 0.5928 0.4630 0.2034 Akuse 0.7226 0.6831: 0.6435 0.5645 0.4063 0.0901 Agawtaw 0.7226 0.6892 0.6560 0.5894 0.4561 0.1896 University of Ghana http://ugspace.ug.edu.gh Table 7. Mounts of K2C03 Salt Added to K2HP04~ Treated Soils. Soil Series Name t P Saturation of AcIsorption Maximum 0 18 14 12 1 2 grams K^CO^ Salt Added Ankasa 1.4455 1.3552 1.2649 1.0841 0.7226 Nil Boi 1.4455: 1.3663 1.2873 1.1291 0.8128 0.1803 Tikobo 1.4455 1.3769 1.3085 1.1717 0.8980 0.3506 Koforidua 1.4-455 1.3751 1.3050 1.1645 0.8833 0.3213 Wacri 1.4455 1.3804 1.3156 1.1857 0.9261 0.4068 Mamfe 1.4455 1.3686 1.2919 1.1385 0.8315 0.2177 Oyarifa 1.4455 1.3864 1.3276 1.2099 0.9743 0.5033 Toje 1.4455 1.3804 1.3156 1.1857 0.9261 0.4068 Akuse 1.4455 1.3663 1.2873 1.1291 0.8128 0.1803 Agawtaw 1.4455 1.3803 1.3117 1.1779 0.9103 0.3705 University of Ghana http://ugspace.ug.edu.gh inclined position to the wall of the plastic pot. Each plastic pot was then filled with the soil-sand mixture. A solution to supply the following nutrient elements was applied to each pot: 170mg. Ca as Ca(N0.j)2; 100 mg. N as NH^NO^ and Ca(NOj)2; 20 mg. Mg as MgS0^i7H20; 2 mg. Fe as FeS0^.7H20; 0.175 mg. Mh- as MrvCl2 .4H20; 0.08mg Cu as CuS0^.5H20; 0.25 mg. Zn as ZnS0^.7H20; 0.8 mg. B as HjBO^; 10 microgram Mo as (NH^)g MOyO^.4H 20. Twelve millet (Pennisetum t^phoides) seeds were sown in each pot. The resulting seedlings were eventually thinned after emergence to eight seedlings per pot. The soil samples were moistened daily throughout the growing period with distilled water. After the third week of growth the plants started showing symptoms of certain nutrient disorder. The symptoms were characterised by a breakdown and subsequent drying of the tips of the newly emerged leaves. The symptoms were initially only visible on plants growing on soil samples of the soil series Boi, Koforidua, Wacri, Mamfe and Agawtaw with P application rates of 1 and 2 times the P adsorption maximum. By the fourth week of growth the symptomos had become visible on almost all the plants but, as at the start, more pronounced on plants growing in the soil samples with the P application rates 61 University of Ghana http://ugspace.ug.edu.gh of 1 and 2 times the P adsorption maximum. For Koforidua and Wacri series, however, those soil samples with P application rates of only \ times P adsorption maximum showed the symptoms also by the fourth week of growth. Due to the nature of the symptoms, and since only plants on soil samples with high P application rates showed the symptoms, P-induced micronutrient element deficiencies were suspected. Further addition of nutrient solution containing mainly calcium, magnesium and micronutrient elements was applied on the twenty-third day of growth. The total concentrations of nutrient elements after the second application were as follows: 220 mg. Ca; 100 mg.N; 40mg Mg; 20 mg. Fe; 2.175mg. Mn; 0.'28mg. Cu; 0.45mg. Zn; 3.0 mg. B; 10 micrograms Mo. Following further application of nutrient elements the new leaves which emerged during the fifth week and thereafter did not show the characteristic breakdown of the tissues which was evident at the initial stages of growth. The plants were harvested when they were forty-two days old. The above ground portions of the millet plants were cut. The cut plants were thoroughly washed in distilled water. The washed plant parts were placed in clean brown paper bags and placed in a forced-draft oven set at 65°C. 62 University of Ghana http://ugspace.ug.edu.gh After drying for forty—eight hours at this temperature the oven-dried samples were weighed and the weights were recorded. 2.9. Laboratory Analysis of,Plant Material for P, Ca, Mg, Fe and Mn. The oven-dried plant materials were ground using a Willey mill. After grinding the ground plant material was stored in clean, dry bottles. Further drying for an additional twenty-four hours at 65°C was carried out just before subsamples were weighed out for digestion and eventual analysis. 2.9.1. Degestion of Plant Material. The ground plant material was digested by a procedure outlined by Black (1957). Generally one-gram sample of the ground plant material was weighed out into 150-millilitres conical flask. Fifteen millilitres of concentrated nitric acid were added to the ground plant material. Another fifteen millilitres portion of the concentrated nitric acid were added to another conical flask containing no plant material and carried that flask through a similar procedure to provide a blank. The conical flasks and their contents were heated gently at first and then more strongly until the contents of the flasks were almost.dry. The flasks were cooled and then 10 millilitres of 8 N nitric acid and University of Ghana http://ugspace.ug.edu.gh 10 millilitres of 70! perchloric acid were added to the contents of each conical flask. The resulting solutions were evaporated to dryness on a hot plate at a relatively low temperature. Fifteen millilitres of 2 N hydrochloric acid were added to the contents of the flasks which were heated for ten minutes to dissolve the salts. The digests were filtered through Whatman No.41 filter papers into 1 0 0-millilitres volumetric flasks and the contents of the volumetric flasks were diluted to volume with distilled water and the solutions were thoroughly mixed. 2.9.2. Colorimetric Determination of P in Plant Material. Analysis of the digest for phosphorus was done by a modified phospho-molybdate-vanadate colorimetric technique outlined by Black (1957). Ten millilitres aliquot of the diluted digest of the plant material and blank were pipetted into separate 150-millilitres conical flasks. Five millilitres of 2 N nitric acid were added and the solution was evaporated to dryness on a hot plate. Finally five millilitres of 0.1 N nitric acid and twenty-five millilitres of distilled water were added to the dry residue in the flask by means of a pipette. The solutions were swirled, then set aside for about ten minutes and then swirled again several times. A suitable aliquot, generally five millilitres of University of Ghana http://ugspace.ug.edu.gh the solution were pipetted into 50-millilitres test-tube and twenty-five millilitres molybdate-vanadate reagent were added from a pipette. The resulting solutions were mixed thoroughly and the transmittancy of the solutions were measured on a Bausch and Lomb "Spectronic-20" colorimeter at 420 millimicron wavelength within one to twenty-four hours. The galvanometer had been previously set at 1001 transmission using a solution prepared from five millilitres of distilled water and twenty-five millilitres of molybdate-vanadate reagent. 2.9.3. Colorimetric Determination of Fe in Plaiit Material. The concentration of iron in the plant digest was determined colorimetrically by a procedure outlined by Black (1957). Generally five millilitres aliquot of diluted digests of plant material and of the blank digest were pipetted into separate 50-millilitres volumetric flasks. Five millilitres of sodium acetate-acetic acid and one millilitre of hydroxylamine hydrochloride solutions were added. The resulting solutions were mixed thoroughly, allowed to stand for one minute, and then five millilitres of orthophenanthroline indicator solutions were added. The contents of the volumetric flasks were made up to the mark with distilled water and the solutions were thoroughly mixed again. The percentage transmittancy of the solutions were measured on a Bausch and Lomb "Spectronic-20" colorimeter at 520 millimicron wavelength. 65 University of Ghana http://ugspace.ug.edu.gh Earlier the galvanometer had been set at 100! transmission using reagent blank made up of distilled water, 0.75 millilitres of 2 N hydrochloric acid and all the other reagents enumerated above. 2.9.4. Colorimetric Determination of Mn in Plant Material. Colorimetric determination of manganese in the plant digests was carried out by a procedure outlined by Black (1957). Fifteen millilitres aliquot of diluted digests of plant samples and the blank were pipetted into separate 150-millilitres conical falsks. Ten millilitres of concentrated nitric acid were added and the solutions were evaporated to dryness on a hot plate. Twenty-five millilitres of distilled water, 2.5 millilitres of concentrated sulphuric acid and then 0.1 gram of solid potassium periodate were added to the contents of the flasks. The resulting solutions were then boiled for five minutes to develop the colour and the solutions were allowed to cool. The solutions were transferred to 50-millilitres volumetric flasks and were diluted to a volume of fifty millilitres with distilled water that previously had been boiled for ten minutes with 0.5 gram of potassium periodate per litre of water and cooled. The solutions were thoroughly mixed and then the percentage transmittancy was measured on a Bausch and Lomb "Spectronic-20" 66 University of Ghana http://ugspace.ug.edu.gh 67 colorimeter at 420 millimicron wavelength within one to twenty-four hours. The galvanometer had been previously set at 1 0 0! transmission with reagent blanks. 2.9.5. Determination of Ca and Mg in Plant Material. The calcium and magnesium concentrations of the plant digests were estimated by the EDTA titration method outlined in Black et. al., (1965) . The digests for the determination were obtained from the nitric, perchloric and hydrochloric acid digestion outlined in a previous paragraph. University of Ghana http://ugspace.ug.edu.gh CHAPTER 3 RESULTS 3.1. Phosphorus Adsorption Maximum of Soils Used. Table 8 represents the data obtained in the adsorption maximum studies. The equilibrium P concentration is expressed as 10^ X C moles per litre, and x/m represents the milligrams of P adsorbed per 100 grams of soil. A linear relationship was obtained for almost all the soils at least within the range of concentrations of 1 x 10~^M -4to 7 x 10 'M. However, beyond an initial concentration of K^PO^ solution of 7 x 10-^M, slight deviations of the isotherm from a straight line were observed for some of the soils. Figure 1 shows the Langmuir plot of the adsorption maximum data presented in Table 8 for three selected soils. Soil 5, Koforidua, represents the soils (6, Wacri; 8, Oyarifa; 10, Akuse) which closely follow the adsorption equation. Soil 3, Boi (representing soils 1, Abenia; 2, Ankasa; 4, Tikobo; 11, Prampram) shows a slight deviation of the isotherm from a straight line relationship. Infact with these soils there is a drop in the curve beyond the initial concentration of KH2P04 solution of 9 x 10_4M- Soil 9, Toje (also representing soils 7, Mamfe, and 12, Agawtaw) 68 University of Ghana http://ugspace.ug.edu.gh shows the poorest fit. Here also the last two points corresponding to initial concentration of K^PO^ solution of 9 x 10_4M, fell farther off above the curve. The textural classification of the soils used in the investigation falls under clay loam, sandy clay, sandy clay loam, loamy sand, sandy loam and sand (Table 1). Abenia series, a sandy clay soil with the highest clay content (35.92%) has the highest adsorption maximum of 50.000 milligrams P per 100 grams of soil. Prampram, a sandy clay loam, with the second highest clay content (34.48%) also has the second highest adsorption maximum of 40.000 milligrams P per 100 grams of soil. Oyarifa series has the lowest adsorption maximum of 23.2 56 milligrams P per 100 grams of soil. Toje series also has a comparatively low adsorption maximum of 2 5.641 milligrams P per 100 grams of soil. The adsorption maximum values obtained for all the soil samples used in the investigation are just as expected. Those soils from the high rain forest areas with high clay content and less silica have high adsorption maximum values. Also some of the savanna soils with high clay content and also contain less silica have high adsorption maximum values. 69 University of Ghana http://ugspace.ug.edu.gh In contrast, the other high rain forest and savanna soils formed over acid parent materials with less clay content but high silica content have low adsorption maximum values. A constant K, related to the bonding energy of the absorbent for the absorbate was also calculated from the slope and intercept values presented in Table 8 . From the linear form of the Langmuir equation expressed as c/(x/m) ± 1/Kb + c/b, c,/(x/m} may be plotted as a linear function of c with slope 1 /b and intercept 1/Kb. If therefore, the value for the slope of the linear curve is divided by the value for the intercept another value, representing the constant, K, is obtained. This constant, K, ranges, in this investigation, from 1.18 5 to 12.333. The bonding energy appears-to increase as adsorption maximum increases. However, there is no significant relationship between adsorption maximum and bonding energy. The correlation coefficient is only r = + 0.408. 3.2. Relation of Adsorption Maximum and Bonding Energy to Some Soil Properties. Table presents the relation between adsorption maximum, bonding energy, and some soil properties. Simple correlation coefficients obtained for the relationship between the adsorption maximum and some soil properties 70 University of Ghana http://ugspace.ug.edu.gh Table 8. Adsorption Maximum Data'. Identi­ fication No. of Soil Soil Series Name Equilibrium P Concentra­ tion. CxlO moles/ litre x/m mg.P/lOOjj Soil c/ (x/m) Slope Intercept Adsorption maximum mg.P/lOOg Soil Constant related to banding energy K 1 Abenia 0.015 6.098 0.003 0.020 0.002 50.000 10.000 0.025 12.232 0.002 , '» 0.043 18.324 - 0,002 0.328 28.948 0.011 0.313 41.432 0.008 1.282 47.824 0.027 2.620 58.118 0.045 2 Ankasa 0.021 6.066 0.003 0.028 • 0.003 35.714 9.333 - 0.054 12.056 0.005 0.110 17.908 0.006 0.550 27.570 0.020 0.901 37.790 0.024 3.397 34.716 0.098 5.008 43.322 0.116 3 Boi 0.028 6.022 0.005 0.032 0.004 31.250 8.000 0.078 11.908 0.007- 0.201 17.344 0.012 0.930 25.216 0.037 1.684 32.936 0.051 4.208 29.692 . 0.142 6.095 36.586 0.167 4 Tikobo 0.026 6.038 0.004 0.037 0.003 27.027 12.333 0.102 11.762 0.009: 0.249 17.046 . 0.015 1.096 24.192 0.045 1.956 31.252 0.063 4.768 26.224 0.182: 7.129 30.180 0.236 University of Ghana http://ugspace.ug.edu.gh Table S.Contd. Identi­ fication No. of Soil Soil Series Name Equilibrium P Concentra­ tion. CxlO moles/ litre x/m mg.P/lOOg- Soil c/ (x/m) Slope Intercept Adsorption maximum mg.P/lOOg Soil Constant related to bondin energy K 5 Koforidua 0.101 5.568 0.018 ' 0.036 0.010 27.778 3.600 0.226 10.994 0.020 0.393 16.154 0.024 -• 1.308 22.878 0.057 5.284 23.232 0.230 7.503 27.862 0.269 6 . Wacri 0.065 5.794 0.011 O.G39 0.006 25.641 .6.500 0.164 11.374 0.014 0.335 16.512 0.020 1.253 23.214 0.054 2.259 29.376 0.077 5.445 22.028 0.247 - 7.808 25.972 0.300 7 Mamfe 0.120 5.454 0.022 0.033 0.018 30.303 1.833 0.278 10.668 0.026 0.503 15.470 0.033 1.695 20.480 0.083 2.754 26.310 0.105 6.586 14.960 0.440 8.769 20.020 0.438 8 Oyarifa 0.102 5.562 0.018 .0.043 0.009 23.256 4.778 0.247 10.862 0.023 0.332 15.912 0.027 1.414 22.220 0.064' 2.375 28.658 0.083 6.090 18.030 0.338 7.755 26.302 0.295 ■ University of Ghana http://ugspace.ug.edu.gh Table Contd. Identi­ fication No. of Soil Soil Series Name Equilibrium P Concentra­ tion CxlO moles/ litre x/m mg.P/lOOg Soil c/(x/m) Slope Intercept Adsorption maximum mg.P/lOOg Soil Constant related to bond] energy K 9 Toje 0.119 5.460 0.022 0,039 0.017 25.641 2.294 0.303 10.512 0.029 0.547 15.200 0.036 1.695 20.480 0.083 2.651 22.083 0.120 6.981 12.508 0.558 8.769 20.020 0.438 10 Akuse 0.052 5.874 0.089 0.032 • 0.027 31.250 1.185 0.154 11.440 0.013 0.291 16.786 0.017 1.183 23.650 0.050 2.058 30.618 0.067 5.124 24.016 0.213 ■ 7.327 28.954 0.253 11 Prampram 0.035 5.982 0.006 0.025 0.004 40.00 6.250 0.042 12.132 0.004 0.091 18.026 0.005 0.421 28.374 0.015 1.103 36.536 » 0.030 2.954 37.456 0.079 4.780 44.734 0.107 12 'Agawtaw 0.130 5.392 0.024 0.038 0.017 26.316 2.235 0.274 10.690 0.026 0.508 15.442 0.033 1.658 20.702 0.080 v “ 2.750 22.083 0.120 6.675 14.408 0.463 8.913 19.126 0.466 University of Ghana http://ugspace.ug.edu.gh Iq 4x C (= I0 4X M OL ES P PE R LI TR E ) X/ m (= MG . P AD SO RB ED PE R IO OG . SO IL ) m PHOSPHORUS ADSORPTION DATA PLOTTED ACCORDING TO THE LANGMUIR ISOTHERM ' University of Ghana http://ugspace.ug.edu.gh 75 indicate that adsorption maximum is significantly related to only clay, free iron oxide, and negatively to silica. A constant, K, related to the bonding energy of the absorbent for the absorbate is also significantly related to pH, free iron oxide, and silica. All the correlation coefficients between adsorption maximum and soil properties which are significant are in the range r = 0.617 to 0.815. The best relationship between adsorption maximum and soil properties, however, is given by the free iron oxide with coefficient of correlation r = 0.815 (positive) which is significant at the 0.1% level. Per cent clay is also highly correlated to adsorption maximum with a coefficient of correlation r * 0.667 (positive). This correlation coefficient is significant at the 2% level. The relation between adsorption maximum and silica is also quite good and significant, but negativeT The specific correlation coefficient is r = 0.617 (negative), and is significant at the 5% level. These simple correlation coefficients between adsorption maximum and soil properties indicate that free iron oxide is mainly responsible for the magnitude of phosphorus adsorption in the soils used. University of Ghana http://ugspace.ug.edu.gh Table 3.: » Phosphorus Adsorption Maximum and Bonding Energy in Relation to Sane Soil Properties Soil Series Adsorption Maximum mg.P/lOOg. Soil Bonding Energy K Free Fe2°3 Clay i Silica pH Organic Carbon % Free "2°3 Abenia 50.000 10.000 0.35 35.92 20.30 4.00 1.51 0.75 Prampram 40.000 6.250 0.31 34.48 32.78 7.12 0.79 13.20 Ankasa 35.714 9.333 0.30 • 23.20 24.44 4.00 - 1.20 0.48 Akuse 31.250 1.185 0.18 32.53 33.46 6.60 0.79 11.26 Boi 31.250 8.000 0.27 32.87 33.86 4.15 0.99 1.95 Mamfe 30.303 1.833 0.22 14.30 41.54 4.60 1.85 0.34 Koforidua 27.778 3.600 0.20 22.53 23.42 6.95 1.41 7.04 Tikobo 27.027 12.333 0.25 12.44 39.72 4.00 0.77 0.74 Agawtaw 26.316 2.235 0.09 8.00 55.50 5.90 0.34 3.44 Wacri 25.641 6.500 0.20 25.39 30.08 6.35 1.18 6.58 Toje 25.641 2.294 0.11 6.39 54.26 5.35 0.20 0.25 Oyarifa 23.256 4.778 0.15 19.86 42.78 6.40 0.85 0.29 -a05 University of Ghana http://ugspace.ug.edu.gh Clay also contributes but its effect relative to free iron oxide may not be as great. Increase in silica content of soils, however, appears to reduce considerably phosphorus adsorption maximum. No significant relationship was observed between adsorption maximum and organic carbon ( r = + 0.388 ), and also between adsorption maximum and free aluminium oxide .( r, = + 0.129 ). The pH is also poorly related to adsorption maximum. The correlation coefficient is only 0.306 (negative). Organic matter, pH, and free aluminium oxide appear to have very little or no influence on phosphorus adsorption maximum. The correlation coefficients between the constant, K, related to bonding energy and soil properties which are significant fall within the range r = 0.473 to 0.717. A very close relationship was found between the bonding energy and free iron oxide ( r = + 0.717). This very close relationship is significant at the II level. The pH is also fairly correlated to bonding energy ( r = - 0.594) and is significant at 51 level whilst the correlation coefficient between bonding energy and silica is r = - 0.473 and is also significant at 101 level. The negative correlation between bonding energy, and pH and silica is an evidence that acid soils will generally retain phosphorus with greater bonding 77 University of Ghana http://ugspace.ug.edu.gh energy than alkaline soils whilst soils high in silica content will retain phosphorus with much less bonding energy. There is no significant relationship between the bonding energy and organic carbon, clay, and free aluminium oxide. The correlation coefficient between the bonding energy and organic carbon is r = 0.177 (positive). The correlation coefficient between bonding energy and free aluminium oxide is r = 0.287 (negative), whilst that between bonding energy and clay is r - 0.289 (positive). 3.3. Dry Matter Yield as Related to the Adsorption Maximum. Data showing the effect of P application on dry matter yield are presented in Table 10. Columns five and six of that table give data on dry matter yield in grams per pot of 454 grams soil. The maximum dry matter yield on Ankasa, Koforidua, Wacri, and Akuse soils occurred at \ the P adsorption maximum. Dry matter yield on Tikobo, Mamfe and Oyarifa soils with KlH^PO^ treatment appears to increase with increasing addition of phosphorus. But on Mamfe and Oyarifa soils with ^HPO^ treatment maximum dry matter yield occurred at \ the P adsorption maximum and the P adsorption maximum, respectively. The maximum yield on Boi, Toje, and Agawtaw soils, however, was obtained at the P adsorption maximum. 78 University of Ghana http://ugspace.ug.edu.gh Examination of the absolute dry matter values reveals that for Koforidua, Wacri, and Akuse soils with high initial P status (lOppm and more) dry matter yield is high initially but the range of increase is narrow. The absolute range values for the KK^PO* treatment, for instance, are 5.10 to 6.92, 2.64 to 5.60, and 2.75 to 9.68 grams per pot, respectively. For Ankasa, Boi, Tikobo, Mamfe, Oyarifa, Toje, and Agawtaw soils with low initial P status (less than lOppm) although the initial dry matter yield is low the range is wider. The range in values for the KH^PO^ treatment are 0.22 to 5.20, 0.24 to 6.15, 0.25 to 6.53, 0.36 to 6.95, 0.40 to 7.73, 0.28 to 6.65, and 0.35 to 7.49 grams per pot, respectively. Those soils with P adsorption maximum of 27 milligrams per 100 grams soil and above (Ankasa, Akuse, Boi, Koforidua, Mamfe, Tikobo) produced maximum dry matter yield at \ the P adsorption maximum. The exceptions are Boi and Tikobo soils. On the other hand, those soils with P adsorption maximum of less than 27 milligrams per 100 grams soil (Agawtaw, Oyarifa, Toje, Wacri) produced maximum dry matter yield at the P adsorption maximum. An exception to this latter rule is the Wacri soil. University of Ghana http://ugspace.ug.edu.gh Table 10. Effect of P Application on Dry Matter Yield. Soil Series Name Initial P Status of Soil (ppm) Bray & Kurtz f^ethod P Saturation of Adsorption Maximum Doses of P Applied Dry Matter Yield in grams per pot Relative Yield Dry Matter at Zero Kg/ha KH2P04 kh2po4 K2HP04 Ankasa 4.50 zero zero 0.22 0.221g 100 0.48 0.43 2.2 2,014 200 2.25 1.01 10.2 4.6 ■ 2 400 5.20 ' 4.10 23.6 18.6 1 800 4.65 4.18 21.1 19.0 ' 2 1600 4.77 4.48 21.7 20.4 Boi 2.78 zero zero 0.24 0.24i§ 88 0.83 0.41 3.5 1.7 i4 175 2.50 0.99 10.4 4.112 350. 4.90 4.87 20.4 20.3 1 700 6.15 6.08 25.6 25.3 2 1400 6.13 6.05 25.5 25.2 Tikobo 3.70 zero zero 0.25 0.25ig 76 0.29 0.22 1.2 0.914 152 1.99 0.21 8.0 0.812 303 2.51 0.73 10.0 2.9 1 606 4.04 3.20 16.2 12.8 2 1212 6.53 5.70 26.1 22.8 Koforidua 31.60 zero zero 5.10 5.101§ 78 5.50 5.70 1 .1 1 .114 156 5.48 6.56 1 .1 1 .112 311 6.92 6.76 1.4 1.3 1 622 6.50 7.42 1.3 1.4 2 1244 6.91 7.06 1.4 1.4 University of Ghana http://ugspace.ug.edu.gh Table 10. Contd. Soil Series Name Initial P Status of Soil (ppm). Bray & Kurtz Method P Saturation of Adsorption Maximum Doses of P Applied Kg/ha Dry Matter Yield in grams per pot Relative Dry Matter at zero KH2P°4 k2h p o4 ffl2P°4 K2Hp°4 Wacri 15.35 zero1 zero72 2.64 4.30 2.64 2.95 1.6 1 .1 1 144 4.52 3.62 1.7 1.41 287 5.60 4.39 2.1 1.7 1 575 5.31 4.05 2.0 1.5 2 1150 4.90 4.89 1.9 1.8 Mamfe 4.04 zero zero 0.36 0.36i 85 2.20 3.19 6.1 8.9 i 170 2.85 ‘ 4.94 7.9 13.7 i 340 5.40 6.82 15.0 18.921 680 5.88 6.21 16.3 '17.2 2 1359 6.95 6.59 19.3 18.3 Oyarifa 4.12 zeroi zero 65 0.40 1.87 0.40 1.56 4.7 3.981 130 2.90 3.43 7.2 8.6 1 260 6.85 5.65 17.1 14.12 1 521 7.00 6.36 17.5 15.9 2 1042 7.73 6.37 19.3 15.9 Toje 1.78 zero zero 0.28 0.28 7.4i 72 2.06 0.80 2.981 144 4.69 1.60 16.8 5.71 287 5.75 3.34 20.5 11.921 575 6.50 3.85 23.2 13.8 2 1150 6.65 4.32 23.8 15.4 University of Ghana http://ugspace.ug.edu.gh Table lD.Contd. Soil Series Name Initial P Status of Soil (ppm). Bray & Kurtz Method P Saturation of Adsorption Maximum Akuse 10.40 Agawtaw 2.03 zero1 8 1 ! 2 12 zero i 8 141 2 1 2 University of Ghana http://ugspace.ug.edu.gh Doses of P Dry Matter Yield Relative Dry - Applied in grains per pot Matter at Zero Kg/ha at Zero kh2po4 K2HPO4 KH2PO4 w zero 2.73 2.73 88 3.98 3.43 1.5 1.3 175 6.43 4.30 2.4 1.6 350 7.35 6.82 2.7 2.5 700 7.58 6.89 2.8 2.5 1400 9.68 6.75 3.6 2.5 zero 0.35 0.35 74 2.10 2.05 6.0 5.9 - 148 - 4.08 4.03 11.7 11.5 295 6.04 5.30 17.3 15.1 590 6.95 6.20 19.9 17.7 1180 7.49 6.40 21.4 18.3 University of Ghana http://ugspace.ug.edu.gh Boi and Tikobo soils produced maximum yield at the P adsorption maximum and not at \ the adsorption maximum as would be expected from the above rule. Wacri soil also produced maximum yield at \ the P adsorption maximum and not at the P adsorption maximum. An explanation to these devia­ tions could be found in the relation of the initial P status to the P saturation of adsorption maximum at which maximum yield occurred. The initial P status of the soils appeared to influence greatly the P saturation of the adsorption maximum at which maximum dry matter yield occurred. Thus Boi and Tikobo soils with high adsorption maximum but very low initial P status of 2.78 and 3.70 ppm, respectively, produced maximum yield at the P adsorption maximum. Wacri soil, however, with its low adsorption maximum but high initial P status of 15.35 ppm produced maximum yield at \ the P adsorption maximum. It appears that those soils with high initial P status but low P adsorption maximum did not require as high a saturation of the adsorption maximum in order to produce maximum yield as did soils with high adsorption maximum but low initial P status. Figures 2(a) to (d) present a graphical picture of the relationship between P saturation of adsorption maximum and the relative dry matter yield. For Koforidua Wacri, and Akuse soils there were rapid increases in relative dry 83 University of Ghana http://ugspace.ug.edu.gh matter yields up to § P saturation of the adsorption maximum. Thereafter, for Koforidua there was a levelling off whilst for Wacri and Akuse soils there were gradual increases pp to \ P saturation of the adsorption maximum before becoming stationary. Ankasa and Mamfe soils both showed similar characteristics in their relationship between P saturation of the adsorption maximum and relative dry matter yield. Both soils produced rapid increases in relative dry matter yields up to \ P saturation of the adsorption maximum. After that point, for Ankasa with KI^PO^ treatment and Mamfe with ^HPO^ treatment there was a sharp drop in the curve to the P saturation of adsorption maximum and then levelled off. For Ankasa with ^HPO^ i treatment and Mamfe with KF^PO^ treatment there were tenden­ cies towards increased relative yields after \ P saturation of the adsorption maximum but the differences were not significant. Agawtaw, Boi, Oyarifa, and Toje soils also showed similar characteristics in their relationship between P saturation of adsorption maximum and relative dry matter yield. Boi and Oyarifa soils produced rapid increases up to I P saturation of the adsorption maximum then produced gradual increases up to the P adsorption maximum before levelling off. However, for Oyarifa with K^PO^ treatment there was further increase after the P adsorption maximum but again the difference was not significant. 84 University of Ghana http://ugspace.ug.edu.gh RE LA TI VE YI EL D; A T ZE RO 85 P SATURATION OF THE P ADSORPTION MAXIMUM Fig. 2(a) THE RELATIVE YIELD OF MILLET AS RELATED TO THE P SATURATION OF THE ADSORPTION MAXIMUM FOR KOFORIDUA, AKUSE AND WACRI SOILS. University of Ghana http://ugspace.ug.edu.gh RE LA TI VE YI EL D AT ZE RO 86 Fig. 2 (b) THE rELATIVE YIELD OF MILLET AS RELATED TO THE THE P SATURATION OF THE ADSORPTION MAXIMUM FOR ANKASA AND MAMFE SOILS. University of Ghana http://ugspace.ug.edu.gh RE LA TI VE YI EL D AT ZE RO 87 j______ i___________ _ j___________________________ i 1/8 1/4 |/2 I 2 P SATURATION OF THE P ADSORPTION MAXIMUM Fig. 2(c) THE RELATIVE YIELD OF MILLET AS RELATED TO THE P SATURATION OF THE ADSORPTION MAXIMUM FOR AGAWTAW, BOI, OYARIFA AND TOJE SOILS. J^3- — ’ y ' , / LEGEND I t ®------ • KHa PO4 TREATMENT ON AGAWTAW a----- o K2 HP0 4 TREATMENT ON AGAWTAW A----- A KHaP04 TREATMENT ON BOI A----- A K2 HP04 TREATMENT ON BOI 0 KH2 P0 4 TREATMENT ON OYARIFA 0----- 0 K2 HP04 TREATMENT ON OYARIFA 0----- 0 KH 2 PO4 TREATMENT ON TOJE 0----- 0 k2 HPO4 TREATMENT ON TOJE University of Ghana http://ugspace.ug.edu.gh . „---- .* KH2 PO 4 TREATMENT ON TIKOBO . a---- -* K2 HPO4 TREATMENT ON TIKOBO /' / / / / _____ !_____________ !____________ I 1/8 1/4 1/2 1 . 2 P SATURATION OF THE P ADSORPTION MAXIMUM. Fig. 2(d) THE RELATIVE YIELD OF MILLET AS RELATED TO THE P SATURATION OF THE ADSORPTION MAXIMUM FOR TIKOBO SOIL. University of Ghana http://ugspace.ug.edu.gh For Agawtaw and Toje soils there were rapid increases in relative dry matter yields up to | P saturation of the adsorption maximum. Thereafter, there were gradual increases up to the P adsorption maximum before becoming stationary. For Toje with K^HPO^, treatment there was rapid increase up to | P saturation of adsorption maximum before levelling off slightly. Tikobo soil, on the contrary, behaved differently from all the other soils. It appeared that successive addition of phosphorus produced increased yields in dry matter up to 2 P saturation of adsorption maximum. Comparison of rates of increase in dry matter with unit addition of P at the various P saturation of adsorption maximum reveals that on most of the soils used optimum dry matter yield could probably be produced at much lower P saturation of the adsorption maximum than the P adsorption maximum. The gradients of the slopes of the curves in Figures 2(a) to (d) calculated at the various P saturation of the adsorption maximum (i.e., I, |, \, 1 and 2) were taken as the rates of increase in dry matter with unit addition of P. For Koforidua, Wacri, Mamfe, Akuse, and Agawtaw soils the rates of increase in dry matter with unit addition of P were greatest at § P saturation of the adsorption maximum with both KH2P0 4 and K2HPC>4 89 University of Ghana http://ugspace.ug.edu.gh 90 treatments. On Ankasa, Boi, and Toje soils the rates of increase in dry matter with unit addition of P were greatest at | and \ P saturation of the adsorption maximum with KH2P0 4 and K^HPO^ treatments, respectively - The rate of increase with unit addition of P on Tikobo soil was greatest at I P saturation of adsorption maximum with Kh^PO^ treatment and the P adsorption maximum with K2HP04 treatment. On Oyarifa soil also the rate of increase in dry matter with unit addition of P was greatest at ^ P saturation of adsorption maximum with KH2P04 treatment and £ P saturation of the adsorption maximum with K^HPO^ treatment. It is evident from the foregoing that on all soils used in this investiga­ tion optimum dry matter yield could probably be obtained within the range of P fertilizer application rates of § the P adsorption maximum and the P adsorption maximum. 3.4. Per cent P Uptake as Related to the' Adsorption Maximum. Data on the effect of P application on the P concentration and uptake of the test-crop for the two P-carrier treatments are presented in Table 11. To obtain the per cent P concentration values the micrograms P per gram plant material values were divided by one million and then multiplied by one hundred. University of Ghana http://ugspace.ug.edu.gh 91 The per cent P uptake values, on the other hand, were obtained by multiplying the per cent P concentration values by the total dry matter yield in grams per pot. Generally the phosphorus concentration in the plant tissues and the total per cent phosphorus uptake increased as the amount of phosphorus added increased. The relationship between the P saturation of the adsorption maximum and the per cent phosphorus uptake is shown in Figures 3 (a) to (e). Ankasa, Boi, and Mamfe soils showed similar characteristics in their relationship between P saturation of the adsorption maximum and the per cent P uptake. For these soils, although there were slight increases in P uptake from zero-P-treatment, marked increases occurred at | the P adsorption maximum. Thereafter there appeared further increases in per cent P uptake. The graphical picture shown by Tikobo soil is slightly different from those for Ankasa, Boi, and Mamfe. There were again slight increases in per cent P uptake from zero-P-treatment but a noticeable increase appeared only at the P adsorption maximum. After that point there were further increases. Agawtaw, Akuse, and Toje soils also showed similar characteristics in their relationship between P saturation of the adsorption maximum and per cent P uptake. Marked increases in per cent P uptake occurred on these soils at § the P adsorption maximum. University of Ghana http://ugspace.ug.edu.gh Table It. Effect of P Application on P Concentration of Millet Soil Series Name Initial P Status of Soil (ppm) Bray & Kurtz Method P Saturation of Adsorption- Maximum Doses of P Applied Kg/ha P Content Per cent Per cent P Uptake kh2po4 K2w o 4 KH2P04 k2hpo4 .Ankasa 4.50 zero zero 0.09 0.09 0.02 0.0218 100 0.10 0.07 0.05 0.031 4 200 0.12 0 . 1 1 0.27 0.1212 400 0.25 0.14 1.30 0.62 1 800 0.32 0.18 1.49 0.75 2 1600 0.52 0.30 2.48 1.34 Boi 2.78 zero zero 0.06 0.06 0.02 0.02 18 88 0.06 0.06 0.02 0.021 4 175 0.13 0.12 0.33 0.1212 350 0.25 0.17 1.23 0.83 1 700 0.31 0.28 1.91 1.70 2 1400 0.52 0.55 3.19 3.33 Koforidua 31.60 zero zero 0.14 0.14 0.71 0.7118 78 0.20 0.17 1.10 0.97 1 4 156 0.27 0.23 1.48 1.501 2 311 0.38 0.30 2.63 2.03 1 622 0.44 0.36 2.86 2.67 2 1244 0.46 0.58 3.18 4.09 Wacri 15.35 zero zero 0.12 0.12 0.32 0.32 1§ 72 0.18 0.16 0.77 0.47 i 144 0.25 0.21 0.99 0.76 i 287 0.33 0.32 1.85 1.40 i 575 0.44 0.41 2.34 1.66 2 1150 0.60 0.52 2.94 2.54 University of Ghana http://ugspace.ug.edu.gh Table ll.Contd. Soil Series Name Initial P Status of Soil (ppm) Bray & Kurtz Method P Saturation of Adsorption Maximum Doses of P Applied Kg/ha P Cor Per c itent ;ent Per ct Uptc art P ike KH2P04 K2HP04 KH2P04 K^ HPO^ Mamfe 4.04 zero' zero 0.06 0.06 0.02 0.0218 85 0.09 0.09 0.05 0.0314 170 0.16 0.12 0.27 0.12T2 340 0.25 0.23 1.30 0.62 1 680 0.40 0.36 1.49 0.75 2 1359 0.52 0.52 2.48 1.34 Oyarifa 4.12 zero zero 0.06 0.06 0.02 0.0213 65 0.07 0-.08 0.13 0.1214 130 0.10 0.09 0.26 0.3112 260 0.20 0.19 1.30 1.13 1 520 0.40 0.36 2.80 2.29 2 1042 0.67 0.35 5.18 2.23 Toje 1.78 zero zero 0.06 0.06 0.02 0.02i 72 0.07 0.04 0.14 0.03 i 144 0.10 0.09 0.47 0.14 l 287 0.18 0.12 1.04 0.40 1 575 0.30 0.25 1.95 0.96 2 1150 0.66 0.58 4.39 2.51 Akuse 10.40 zero zero 0.09 0.09 0.25 0.251 88 0.13 0.13 0.52 0.4514 175 0.16 0.20 1.03 0.861 350 0.24 0.28 1.76 1.91 1 700 0.41 0.50 3.11 3.45 2 1400 0.57 0.59 5.52 3.99 University of Ghana http://ugspace.ug.edu.gh Table ll.Contd. Soil Series Name Initial P Status of Soil (ppn) P Saturation of Adsorption Doses of P Applied P Content Per cent Per cent P Uptake Bray & Kurtz Maximum . Kg/ha Method KH2P°4 k2hpo4 KH2P04 Agawtaw 2.03 zero zero 0.05 0.05 0.02 0.02 1 8 74 0.10 0.07 0.21 0.14 1 4 148 0.12 0.15 0.49 0.60 1 2 295 0.25 0.22 1.51 1.35 1 ' 590 0.42 0.38 3.41 2.36 2 '1180 0.63 0.67 4.72 4.29 Tikobo 3.70 zero zero 0.06 0.06 0.02 0.02 1 8 76 0.08 0.07 0.02 0.02 1 4 152 0 . 1 1 0.07 0.22 0.02 1 2 303 0.15 0.12 0.23 0.08 1 606 0.20 0.17 0.81 0.54 2 1212 0.45 0.39 2.94 2.22 University of Ghana http://ugspace.ug.edu.gh o9 8 7 6 5 4 3 2 I O 95 ®----- ® KH2 PO 4 0------------ e K2HPO4 Q----- a kh, P04 q----- □ Kz p,po4 &----- ■» KH2 P0 4 A----- A K2 HPC>4 LEGEND TREATMENT ON ANKASA TREATMENT ON ANKASA TREATMENT ON BOI TREATMENT ON BOI TREATMENT ON MAMFE TREATMENT ON MAMFE 1/8 1/4 1/2 I P SATURATION OF THE P ADSORPTION MAXIMUM ig. 3(a) THE PER CENT P/llPTAKE OF MILLET AS RELATED TO THE P SATURATION OF THE P ADSORPTION MAXIMUM FOR THE ANKASA, BOI AND MAMFE SOILS. University of Ghana http://ugspace.ug.edu.gh PE R CE NT P UP TA KE 96 Fig. 3(b) THE PER CENT P UPTAKE OF MILLET AS RELATED TO THE P SATURATION OF THE P ADSORPTION MAXIMUM FOR THE TIKOBO SOILS. University of Ghana http://ugspace.ug.edu.gh EB CE NT P UP TA KE 97 1 O T LEGEND 0-- — -0 KH;> P04 TREATMENT ON AGAWTAW a-- -- 0 k2h p o 4 TREATMENT ON AGAWTAW A------ A KH2 P04 TREATMENT ON AKUSE &— & k2h p o 4 TREATMENT ON AKUSE 0-- -- 0 k h2 p o 4 TREATMENT ON TOJE 0-- -- 0 k2h p o 4 TREATMENT ON TOJE 6 - P SATURATION OF THE P ADSORPT ION MAXIMUM Fig. 3(c) THE PER CENT P UPTAKE OF M IL L E T AS R ELATED TO THE P SATURATION OF THE P ADSORPTION MAXIMUM FOR THE AGAWTAW, AKUSE AND T O JE SO ILS . University of Ghana http://ugspace.ug.edu.gh PE R CE NT P UP TA KE 98 Fig. 3 (d) THE PER CENT P UPTAKE OF MILLET AS RELATED TO THE P SATURATION OF THE P ADSORPTION MAXIMUM FOR THE OYARIFA SOIL. University of Ghana http://ugspace.ug.edu.gh PE R CE NT P U P TA KE 99 THE PER CENT P UPTAKE OF MILLET AS RELATED TO THE P SATURATION OF THE P ADSORPTION MAXIMUM FOR THE KOFORIDUA AND WACRI SOILS. University of Ghana http://ugspace.ug.edu.gh Thereafter there appeared further increases except for the Ko,HP0, treatment on Akuse soil. With this Z 4 treatment there appeared a drop after the P adsorption maximum. On Oyarifa soil there were also slight increases in per cent P uptake from zero-P-treatment but marked increases occurred only at \ the P adsorption maximum. Thereafter, for Kfl^PO^ treatment there appeared further increases. However, for K^HPO^ treatment there were further increases up to the P adsorption maximum and then levelled off. For Koforidua and Wacri soils there occurred significant increases in per cent P uptake from zero-P-treatment up to \ the P adsorption maximum. After that point, although there appeared further increases the differences were not significant. Table 1& summarizes the relationship between P adsorption maximum, bonding energy, and uptake of applied phosphorus. The data on P uptake as per cent of P applied were obtained by dividing the per cent P uptake values by the amount of phosphorus in grams added to the soils (presented in Table 3). On Ankasa, Boi, and Tikobo soils with high adsorption maximum (27 milligrams per 100 grams soil and above) and equally high bonding energy (6.00 and more) uptake of applied P was relatively low. Koforidua soil with high P adsorption maximum gave 100 University of Ghana http://ugspace.ug.edu.gh high values for uptake of applied P. This soil, however, has a low bonding energy value (less than 6.00). The high uptake of applied P could probably be due to the low bonding energy. Further evidence of the influence of bonding energy on the uptake of applied phosphorus is given by Akuse and Mamfe soils. These soils with relatively high adsorption maximum comparable to that of Boi Soil (30 milligrams per 100 grams soil) gave fairly high uptake of applied P values than the Boi soil. Akuse and Mamfe soils have low bonding energy of 1.833 and 1.185, respectively. The high uptake of applied P could probably be a consequence of the low bonding energy of the soil Colloidal particles for added phosphorus. On the other hand, Oyarifa, Toje, and Agawtaw soils with low adsorption maximum and equally low bonding energy gave high values for uptake of applied phosphorus. Wacri soil, however, is an exception. Although it has high bonding energy it still gave high value for uptake of applied phosphorus. In any case its adsorption maximum value is low. It is obvious from the relationship represented in Table 13. that bonding energy is closely related to P uptake than the adsorption maximum. On those soils with high bonding energy uptake of applied phosphorus was University of Ghana http://ugspace.ug.edu.gh Table 11. Relation Between Adsorption Maximum, Bonding Energy and Uptake of Applied Phosphorus 102 Soil Series Name Initial P Status of Soil (ppm) Bray & Kurtz Method Phosphorus Adsorption Maximum Bonding Energy P Uptake as % of P Applied KH2po4 K2HP04 Ankasa 4.50 35.000 9.333 7 4 Boi 2.78 31.250 8.000 12 9 Tikobo 3.70 27.027 12.333 6 3 Koforidua 31.60 27.778 3.600 39 39 Wacri 15.35 25.641 6.500 30 21 Mamfe 4.04 30.303 1.833 15 17 Oyarifa 4.12 23.256 4.778 21 16 Toje 1.78 25.641 2.294 16 7 Akuse 10.40 31.250 1.185 25 23 Agawtaw 2.03 26.316 2.235 24 18 Range of % Uptake Values • • * . 6-39 3-39 Average Values of % Uptake •• • • • . 20 16 University of Ghana http://ugspace.ug.edu.gh comparatively low. The converse, however, is true for those soils with low bonding energy. The correlation coefficients calculated for the relationship between adsorption maximum, bonding energy, and uptake of applied phosphorus support the above assertion. The correlation coefficients show that the bonding energy is a better index of uptake of applied phosphorus than the adsorption maximum. The correlation coefficients between bonding energy and uptake of applied phosphorus are r = -0.566 for KH2P04 treatment and r = -0.562 for K2HP04 treatment, and are significant at 10! level. The correlation coefficients between adsorption maximum and uptake of applied P are not significant. The specific correlation coefficients are r = -0.395 and r, = -0.211 for KB^PO^ and K^HPO^, respectively. 3.5. Comparison of Dry Matter Yields and P Uptake Values for KH,P04 and .K^HPO^ Treatments A critical examination of the data presented in Table 10 and 11 reveals striking differences between the relative efficiency of the KH2P04 and K2HP04 P-carriers. From the dry matter data it is clear that, except on Koforidua and Mamfe, dry matter yields for KHjPO* treatment were generally higher than K2HP0 4 treatment. 103 University of Ghana http://ugspace.ug.edu.gh The per cent P concentration and total P uptake values also support the above observation. The per cent P concentration of the test-crop was generally higher for KH2P04 treatment than K2HP04 treatment. The only exceptions were found on Akuse soil and a few isolated treatments on a few other soils (Boi, Oyarifa, Agawtaw). Similarly, the per cent total P uptake values were, in the majority of treatments, higher for KH2P0 ^ than for K2HP0^ treatment. Examination of the P uptake as per cent of applied P values represented in Table 1% further reveals that uptake of applied P was higher for the KH7P04 few for K^HPO^ treatment. Koforidua and Mamfe soils are exceptions. On Koforidua soil P uptake as per cent of applied P was equal for both treatments. On Mamfe soil, also the P uptake as per cent of applied P was higher for K2HP04 than for KH2P04 treatment. The range of per cent uptake of applied P values for all the soils used are 6 to 39% and 3 to 39! for KH2P04 and K?HP04 treatments, respectively- The differences in per cent uptake of applied P values observed in this investigation for the two P-carrier treatments are true reflection of the relative availability of the H^PO^ and HPO^ ions. Plants are known to take up their inorganic phosphorus principally as the H2P0 ~ ion. 104 University of Ghana http://ugspace.ug.edu.gh Infact plants may take up this inorganic phosphate ion more easily than the IIPO"" ion. Hence the low phosphorus uptake and dry matter yields on KH^PO^-treated soil samples were as expected. 3.6. Plant Tissue Composition of Ca, Mg, Fe, Mn. Data on the effect of phosphorus application rates on the concentration of calcium, magnesium, iron and manganese in the plant tissues are presented in Table 13. On all the soils used in this investigation and, for both P-carrier treatments increased phosphorus addition reduced the availability of calcium as judged by the tissue composition of the test-crop. This is evidenced by the very close but negative correlation coefficients obtained for the relationship between the quantity of phosphorus added and the plant tissue concentration of calcium for all the soil treatments put together- The specific correlation coefficients are r = 0.653 (negative) for KH^PO^ treatment and r = 0.592 (negative) for K^HPO^ treatment. Both correlation coefficients are significant at 0.1% level. In the light of the low calcium content of the plant tissue it is probable that the symptoms observed in the plants as early as the third week of growth on some of the soil samples could be calcium deficiency symptoms. 105 University of Ghana http://ugspace.ug.edu.gh The symptoms were characterised by a breakdown and subsequent dnjing of the tips of the newly formed leaves, a nutrient disorder symptom characteristic of calcium shortages. Availability of magnesium as evidenced by the plant tissue composition, on the other hand, was generally enhanced by increased phosphorus applications. The correlation coefficients calculated for the relationship between the quantity of phosphorus added and magnesium concentration of the plant tissues support the above observation. There are close and positive correlations between the quantity of phosphorus and the plant tissue composition of magnesium for all the soil treatments put together and for both P-carrier treatments. The specific correlation coefficients are r = 0.385 (positive) for Kf^PO^ and r = 0.313 (positive) for K^HPO^ treatment. The correlation coefficients are significant at 1% level and 5! level for KH^PO^ and K^HPO^ treatments, respectively. When the data on the quantity of phosphorus added and the plant tissue concentration of iron and manganese were subjected to statistical analysis no correlation was obtained between the quantity of phosphorus added and the plant content of iron and manganese for all soil treatments put together. 106 University of Ghana http://ugspace.ug.edu.gh Table IS. Effect of P Application Rates on Concentration of Ca, Mg.Fe, Mn in Millet Tissues. Soil Series Name P Saturation of P Ads. Maximum Doses of P Appl­ ied Kg.P/ha Concentration of Ca Mg, Fe, Mn in Tissue % Ca Content 1 Mg Content Fe Content (ppm) Mil Content (ppm) kh2 po4 • w • KH2P04 K2Hp°4 kh2 po4 j y ® ° 4 KH2 PO4 . K2 HPO4 Ankasa 0 zero 4.38 4.38 1.80 1.80— 287 287 56 56 1 1 0 0 2.61 3.53 1.04 0.71 160 150 92 60 1 2 0 0 1.92 2.06 1.06 0.96 1 S6 95 94 89f 2 400 0.64 0 . 8 8 1 . 2 2 0 . 8 6 127 119 1 1 0 1 0 0 1 800 0.44 0.48 1.70 1 . 1 0 128 1 1 S 126 94 2 1600 0.08 0.16 1.73 1.34 106 1 1 0 126 94 Boi 0 zero 2.58 2.58 1.31 1.31 219 219 49 49 1§ 8 8 2.40 2.16 0.39 0.44 162 176 44 54i4 175 1.60 1.82 0 . 8 8 0.96 139 125 52 63 - 1 350 ‘ 0.36 1 . 6 8 1.77 0.91 137 1 2 0 71 8 6 i 700 0.16 1 . 2 0 1.75 1.15 128 119 8 6 79 2 1400 0.08 0.80 1 . 6 8 . 0 . 8 6 115 110 79 118 Tikobo 0 zero 4.00 4.00 0.70 0.70 214 214 2 0 . 2 01 76 3.84 4.00 0.67 0.59 83 240 51 331 152 2.56 3.43 0.96 1 . 1 2 191 2 2 0 8 6 491 303 1.40 1.90 0 . 8 6 0.96 134 153 8 6 74 1 606 0.80 0.56 1.44 1.15 1533 142 103 8 6 2 1 2 1 2 0.64 0.24 1.56 1.25 141 146 189 103 Koforidua 0 zero 1.24 1.24 1.15 1.15 155 155 1 2 2 1 2 2 18 78 1 . 2 0 0.96 1.18 1.32 165 142 143 1261 3 156 1 . 2 0 0.84 1.25 1.25 133 133 189 126 12 311 0.72 0.32 1.54 1.34 162 103 160 168 1 622 0.80 0.48 l.'S4 1.30 205 215 192 2 0 0 2 1244 0.48 0.32 1.58 1.44 229 2 1 1 259 284 University of Ghana http://ugspace.ug.edu.gh T ab le lV C o n t d . Soil Series Name P Saturation of P Ads. Maximum Doses of ■ P Appl­ ied Kg.P/ha Concentration of Ca, Mg, Fe, Mn in Tissue °o Ca Content % Mg Content Fe Content (ppm) ffc Content (ppm) K2 P0 4 k2 hpo4 KH2 p0 4 i^HPO^ KH2 P0 4 W . ™ 2p04 K2 HPO4 Wacri 0 zero 1.31 1.31 1 . 1 0 1 .1 0 - 16S 165 126 -1261 72 1.25 0.56 1.09 1.15 187 1 2 0 69 168--- 1 144 1.32 0.72 1.34 1.39 171 164 2 2 1 192 1 287 1 . 1 2 0.76 1.44 1.49 182 133 189 319 1 575 0.92 0.24 1.49 1.70 2 2 0 2 1 1 208 319 ' 2 1150 0.63 0.24 1.63 1.63 229 165 225 259 Mamfe 0 zero 2.60 2.60 1.36 1.36 181 181 56 56i 85 1.56 0.96 1 . 1 0 0 . 8 6 146 146 58 71 170 1.04 0.72 1.40 0.81 129 1 2 0 63 84 1 i 340 0.48 0.16 1.73 1.44 146 67 47 8 6 680 0.64 0.48 1.62 1 . 0 1 132 80 79 79 2 1359 0.64 0.32 1.54 1.49 104 106 8 6 94 Oyarifa 0 zero 2.65 2.65 1.33 1.33 143 143 24 24 1 65 1.72 1.04 0.82 0.99 129 229 39 47 4 130 1.32 1.16 1.37 0.96 1 0 2 187 31 39 1 260 1 . 0 0 0.60 0.91 1.30 137 115 63 54 i 520 0.48 0.40 0.96 1.44 127 135 71 63 2 1042 0.48 0.48 1.44 1.39 1 2 0 137 103 94 Toje 0 zero 3.50 3.50 1.45 1.45 279 279 6 8 6 8 1 72 2.40 2.38 0.82 1.16 1 0 0 2 2 0 31 92t 144 1 . 2 0 1.40 1.04 0.76 93 140 no 39 1 287 1.04 0.48 1.06 1.30 1 2 0 297 71 94 1 575 0.54 0.40 1.54 1.54 124 215 1 1 0 125 2 1150 0.36 0.16 1.34 ' 1.73 154 225 1 1 0 155 108 University of Ghana http://ugspace.ug.edu.gh Table 15»Contd, Soil Series Name P Saturation of P Ads. Maximum Doses of P Appl­ ied Kg.P/ha Concentration of Ca, Mg, Fe, Mn in Tissue . ,------ % Ca Content % Mg Content Fe Content (ppi) Mn Content (ppm) KH2 p ° 4 k2 hpo4 . KH2 PO4 . k2 hpo4 . KH2 P04 K2HP04 K ^ p c y k2 hpo4 Akuse 0 zero 1.60 1.60 1.54 1.54 1 0 2 1 0 2 55 55 1 8 8 1.59 1.28 1.30 1.37 1 1 1 146 63 71 I4 175 1.52 1 . 1 2 1.42 1.42 115 142 -71 71 1 350 1.04 0 . 6 8 1.85 1.30 146 128 94 1 1 0 I 700 0.32 0.16 2.28 1.70 162 93 1 1 0 103 2 1400 0.64 0 . 2 1 1.63 1.44 142 146 1 2 0 126 Agawtaw 0 zero 3.27 3.27 1.48 1.48 164 164 37 37 x 74 1.16 1 . 2 0 1.03 0.81 119 128 31 55 i 148 1.16 1.04 1.06 0 . 8 6 155 174 39 65 i 295 1 . 2 0 0.48 1.39 1.25 151 137 45 94 1 590 0.56 0.16 0.43 1.34 1 1 1 142 63 103 2 1180 0.56 0.08 0.67 1.39 128 137 8 6 119 University of Ghana http://ugspace.ug.edu.gh However, when the soils were considered separately some correlation was obtained between the quantity of phosphorus added and the plant tissue concentration of iron and manganese for both treatments. The correlation coefficients for the relationship between the quantity of phosphorus added and the iron concentration of the plant tissues are negative. Only the correlation coefficients for Boi (KH2P04 treatment), Koforidua (both KH^PO^ and K^HPO^ treatments), Mamfe (KH9P0^ treatment), Tikobo (K2HP04 treat­ ment) , and Wacri (KH2P0,j treatment) are significant. The negative correlation suggests a reduction in iron availability whilst the positive correlation suggests an increase in iron availability with increased P addition. University of Ghana http://ugspace.ug.edu.gh Ill Table 1%. Correlation Between Quantity of Phosphorus " Added and. Plant Tissue Concentration ot Iron and Manganes~e Soil Series Name Correlation Coefficients IRON MANGANESE KH2P04 k2hpo4 kh2po4 ^HP04 Ankasa -0.6545 * -0.4798 0.7852* * 0.5242 **** Boi -0.7142 -0.6700 * 0.7910 ** ** 0.9451 Tikobo -0.1821 -0.7611 0.9582 0.9075 Koforidua *** * **** ***** 0.9060 *** 0.7164 0.9259 0.9936 Wacri 0.9103 ** 0.3772 0.6217 ** 0.5658 * Mamfe -0.8130 -0.4932 0.8320 * *** 0.7249 Oyarifa -0.2300 -0.4347 0.9609 0.9612 Toje -0.1316 Failed 0.6068 0.8658 ** Akuse 0.6659 0.1518 **** 0.9169 ***** 0.8719 *** Agawtaw -0.4767 -0.3805 0.9865 0.8945 Significant at 101 level * * " 5% level * * * " 2% level * * * * " 1 % level * * * * * " 0.11 level University of Ghana http://ugspace.ug.edu.gh The correlation coefficients obtained for tne relationship between the quantity of P applied and the manganese concentration of the plant tissues are all positive. This positive correlation suggests that for the individual soils increase in the quantity of phosphorus added increased manganese availability. Except for Ankasa (J^HPO^ treatment), Toje (IC^PO^ treatment) and Wacri « (both K^PO^ and K^HPO^ treatments) , all correlation coefficients obtained on the individual soils are significant. 3.7. Estimated P Application Rates for the Soils Used. Phosphorus application rates which will be needed to produce optimum yields are presented in Table IS. Ordina­ rily, the required P application rates would be estimated from the P saturation of the adsorption maximum w?hich produced maximum dry matter yields. However, the equivalent quantities of phosphorus in kilograms per hectare at the P saturation of adsorption maximum which produced maximum maximum dry matter are too high. Moreover, the rates of increase in dry matter with unit addition of phosphorus at the P saturation of the adsorption maximum which produced maximum dry matter are low. Infact, the greatest rates of increase with unit addition of P were obtained at much lower P saturation of the adsorption maximum. 112 University of Ghana http://ugspace.ug.edu.gh The specific P saturation of the adsorption maximum with the greatest rates of increase with unit addition of P for the KH2P0 4 treatment are as follows: J, I, \, I, I, I times for Ankasa, Boi, Tikobo, Koforidua, Wacri, Mamfe, Oyarifa, Toje, Akuse, and Agawtaw, respectively. For the K^HPO^ treatment the P saturation of the adsorption maximum with the greatest increase with unit addition of P are as follows: I, I, |, I, \, |-, | times on Ankasa, Boi, Tikobo, Koforidua, Wacri, Mamfe, Oyarifa, Toje, Akuse, and Agawtaw, respectively. The doses of phosphorus equivalent to the P saturation of adsorption maximum given above for the KH2P0 ^ P-carrier treatment are taken as the phosphorus application rates needed to produce optimum dry matter yields. This is on the grounds that the KH^PO^ P-carrier is a more soluble P-carrier than the K^HPO^j . Hence its availability would be a true reflection of the availability of most common P fertilizers. Moreover, plants are known to take up their inorganic phosphorus principally as the H2PO^ ion. The estimated phosphorus application rates range from 72 kilograms P per hectare on Wacri series to 260 kilograms P per hectare on Oyarifa series. 113 University of Ghana http://ugspace.ug.edu.gh 114 Table 15. Estimated P Application Rates for Soils Used, Soil Series Name Rate of P Application Kilograms per hectare Pounds per acre Ankasa 200 178 Boi 175 156 Tikobo 152 136 Koforidua 78 70 Wacri 72 64 Mamfe 85 76 Oyarifa 260 232 To j e 144 128 Akuse 88 79 Agawtaw 74 66 University of Ghana http://ugspace.ug.edu.gh 115 C H A P T E R 4 DISCUSSION A linear relationship of the plot c/(x/m) against c was obtained for all the soils within the range of con­ centration of 1 x 10 ^M. to 7 x 10 ^M. Slight deviations of the isotherm from a linear relationship were, however, observed beyond an initial concentration of KH^PO^ solu- -4tion of 7 x 10 M. This observation that the Langmuir plot of the adsorption maximum data does not follow a linear relationship was not unexpected. Infact Olsen and Watanabe (1957), Thompson (1958), and Weir and Soper (1962) noted some deviations. Moreover, Larsen (1967) at Levington Research Station, England, found that the phosphorus adsorption isotherms were curvilinear for the majority of 120 soils even at very dilute concentra­ tions. Evidence from phosphate adsorption studies on CaCOj by Cole et al. (1953) has also indicated that in the range of equilibrium concentrations where the Lang­ muir equation applies, essentially all of the phosphate _ 70 adsorbed is exchangeable with P” tagged orthophosphate added in solution. Where the isotherm deviates from a straight line only a portion of the adsorbed phosphate exchanges with P~2. A later study by Hsu and Rennie (1962) serves to buttress the observation by Cole et al. University of Ghana http://ugspace.ug.edu.gh 116 (1953). Hsu and Rennie found that where precipitation of phosphate occurs the Langmuir plots fail to give the straight line relationship. It is therefore probable in this study also that at higher concentrations of KH2PC>4 some of the soils merely precipitate the added phosphate instead of adsorbing it onto their surfaces.. The soil series with high adsorption maximum values (Abenia-, Ankasa, Boi, Mamfe, Akuse and Prampram) except Akuse and Prampram series, are mainly soils of the Forest Oxysol and Forest Ochrosol Great Soil Groups. This finding is in accord with those of De-Datta (1964), Saunders (1965) , Younge et al. (1966) , and Syers et al. (19 71 that most soils from tropical and subtropical regions, which fall in the Great Soil Groups of Latosols and Young Latosols and Red-Yellow Podzolic soils have marked ability to retain appliedinorganic phosphates. The soil series Abenia, Ankasa, Boi and Mamfe are soils from areas where the annual rainfall is b e t w e e n » « a n d 5U5‘?«j*n- Hence there has been pronounced through leaching of soluble cations like Ca++, Mg++, K+, Na+ from the surface horizon with the resultant accumulation of Fe and Al. This is evidenced by the analytical values for "free" iron oxide (Fe^^) and aluminium oxide (A^O^) presented in Table 2. The soil series Abenia, Ankasa, Boi, and Mamfe, besides their comparatively fairly high iron and University of Ghana http://ugspace.ug.edu.gh 117 aluminium oxides content, also abound in clay but contain comparatively less silica. Their high adsorption maximum values could, therefore, be attributed also to the high clay content and less silica. The soil series Tikobo which is also classed under Forest Oxysol Great Soil Group, in contrast, gave comparatively low adsorp­ tion maximum. However, this particular soil is believed to be derived from a Tertiary Sand parent material and has a loamy sand texture. Moreover, it has less clay but high silica content. Hence the low adsorption maximum value obtained is consistent since the soil lacks in materials which are believed to contribute positively to phosphorus adsorption in soils. The soil series Akuse and Prampram are soils which belong to the Tropical Black Clay and Tropical Black Earth Great Soil Groups respectively. These soils have comparatively high aluminium oxide and a fairly apprecia­ ble amount of iron oxide. Furthermore, they have high amounts of clay and a not-too-high amount of silica. Akuse and Prampram soil series, especially the Akuse series, are also known to contain appreciable amounts of calcium carbonate hence must have high calcium concentra­ tion. The relatively high adsorption maximum of these soils may have been brought about by the high clay con­ tent, as well as, the possibly high calcium content. University of Ghana http://ugspace.ug.edu.gh 118 Koforidua and Wacri soil series are soils of the Forest Oxysol and Forest Ochrosol Rubrisol Intergrade Great Soil Groups respectively. These two soils are very similar in their properties. They have a fair amount of clay, silica, and iron and aluminium oxides. In addition they have very large amounts of organic matter as represented by the organic carbon content. Therefore their comparatively fairly low adsorption maximum values are consistent. Though the fairly high amount of iron and aluminium oxides and clay should have favoured high adsorption maximum the comparatively high silica and organic carbon content tend to mask their effects. The Oyarifa, Toje and Agawtaw soil series with the lowest adsorption maximum values are soils from a coastal savanna region. They are derived from Sandstone, Tertiary Sand, and Acid Gneiss and Schist parent materials and are of sandy loam, sand, and loamy sand texture respectively. These soils have appreciably low iron and aluminium oxides and clay content. Moreover, they have very high amount of silica, a material which contributes nothing positive to adsorption of phosphorus. The low adsorption maximum values obtained for these latter soils are, therefore, as expected. University of Ghana http://ugspace.ug.edu.gh 119 On the basis of correlation values clay, silica, and "free” iron oxide showed the greatest association with phosphorus adsorption maximum. The relationship with clay content is in agreement with the findings of Pissarides et al. (1968), Galino _et al. (1972), Udo and Uzu (1972), and Schwertmann and Knittel (1973) who found that the values of adsorption maximum obtained in their various investigations were a function of clay. The highly significant correlation between clay and adsorp­ tion maximum may point to the fact that clay minerals provide adsorption sites for phosphate ions. The role of clay in phosphorus fixation has been studied by many investigators. There is the hypothesis that phosphorus fixation by clay minerals is due to the aluminium con­ tent of the clays (Coleman 1944, Ellis and Truog 1955). It is, however, likely that clay forms part of the comp­ lex gel as envisaged by Mattson et al^ (19 50). This com­ plex gel, consisting of hydrated iron oxide (J^O^) al°ng with smaller amounts of organic matter, aluminium oxide (Al^O^) and associated Si(0H) 4 and P, is considered a major site for phosphorus adsorption. There is also the age-long hydroxyl replacement mechanism. It is believed that structural hydroxyl (OH) ions which are less tightly bound readily exchange places with phosphate ions. This latter theory is supported by data on the large amount University of Ghana http://ugspace.ug.edu.gh 120 of phosphate that can be fixed by kaolinite as compared to montmorillonite reported by Murphy (.1939) and Stout (1939). Kelley and Midgley (1943). also found direct relation between the amount of phosphate fixed and the increase in pH obtained when isohydric suspensions of various solid phases and phosphate solutions were mixed. It was found in this study also that the silica content was highly correlated to phosphorus adsorption maximum but the correlation coefficient was negative. Though there is no published data an the relationship between silica per se and phosphorus adsorption maximum, marked relationship, however, has been found between the silica to sesquioxide ratio and phosphorus adsorption pacacity or phosphorus fixation. Wild (1950), Thompson (1957) , and Sauchelli (1965) have all reported indepen­ dently that many investigators have shown an inverse relationship between the silica to sesquioxide ratio and phosphate sorption or the efficiency of phosphate fertilizers for plant growth. Silica per se is an inert material and, therefore, cannot provide any sites for the adsorption of phosphate ions. The very close relationship between "free" iron oxide (Pe20 3) and adsorption maximum found in this investigation was similarly reported by Rajagopal and Idnani (1963) in India, by Ahenkorah (1968) in Ghana, University of Ghana http://ugspace.ug.edu.gh 121 by Udo and Uzu (1972) in Nigeria, and again by Bidappa and Venkat Rao (19 73) in India. This particular rela­ tionship is also in agreement with the findings of Shukla et al. (19 71) in which phosphorus sorption decreased markedly after the successful removal of oxalate-extractable iron. Williams' et al. (1958) also reported a similar relationship between iron and phos­ phorus retention, even though they, like other investigators Bromfield (1964 and 1965), Bromfield and Williams (1963), Saini and MacLean (1965), and Udo and Uzu (19 72) found aluminium to be the major factor res­ ponsible for phosphorus retention. In this work no relationship was found between "free" aluminium oxide (A1.?0„,) and phosphorus adsorption maximum. This observation is consistent with that of Ahenkorah (1968) who, in working with some cocoa growing soils of Ghana, found no relationship between aluminium and phosphorus retention. The lack of agreement between the results of this work and those of other investigators quoted in previous paragraph may be due to several dif­ ferences, among which are the variation in sesquioxides of podzolic and latosolic soils and the dominant parent materials involved. The lack of any relationship between "free" aluminium oxide (A^O™) and phosphorus adsorption maximum may be further explained as being due University of Ghana http://ugspace.ug.edu.gh 122 to ineffective extraction of the "free" aluminium oxide by the single extraction with- the Tamm solution compri­ sing oxalic acid and ammonium oxalate from some of the extremely acidic soils used in this investigation. Even though Ahenkorah (1968) working with some cocoa growing soils of Ghana and some other investiga­ tors have found significant ^relationship between organic carbon and phosphorus retention, this particular work failed to establish such a relationship. There was some correlation ( r = + 0.388) which was not significant. The lack of significance between adsorption maximum and organic carbon is consistent with the findings of Udo and Uzu (1972) . Saunders (1965) also did not get any direct evidence to show that phosphorus is retained by soil organic matter. He, however, indicated that organic matter could be part of the complex gel postulated by Mattson et 'al., (1950). Kardos (1964) has also stated that in reactions involving the adsorption of phosphate ions organic compounds, being dominantly anionic in character, are likely to compete with the phosphate anion in polar adsorption phenomena. Mortensen and Himes (1964) reported that in most agricultural soils which are near a neutral pH, the soil organic matter has a net negative charge. The negatively charged anions and polyanions are apparently linked to clay surfaces through polyvalent University of Ghana http://ugspace.ug.edu.gh 123 inorganic cations and ionized carboxyl groups (clay-M-OOR). Mortensen and Himes (1964) again reported that positive charges due to exposed lattic-edge aluminium and aluminate surfaces have been shown by several investiga­ tors to be adsorption sites for negatively charged, car- boxylated polymers. The extent of adsorption, they believe, is governed by, among other factors, the pH, The majority of the soil samples used in this work have pH in the range 4.60 to 7.12- thus the reactions described above may be possible. Hence the lack of singnificance in the correlation between organic carbon and phosphorus adsorption maximum is consistent. Furthermore, products of organic matter decay, such as organic acids and humus, are thought to be ef­ fective in forming complexes with iron and aluminium, compounds. Such complexing of iron and aluminium compounds reduces inorganic phosphate adsorption to a remarkable degree. All the soil samples used in this investigation except Toje (organic - C = 0.201) and Agawtaw (organic - C = 0.341) gave appreciable amounts of organic carbon content when analysed. Therefore, it is possible that the complexing of iron and aluminium ■compounds by organic acids and humus mentioned above is very operative thus rendering the adsorption sites rather ineffective in attracting phosphate ions. A University of Ghana http://ugspace.ug.edu.gh 124 report by Mortensen and Himes (1964) that mutual coagu­ lation and peptization reactions occur between organic matter and aluminium and iron gives credence to the above observation. Also acids produced in organic matter transformation could decrease the pH thus resul­ ting in enhanced solubilization of iron and aluminium from clay minerals and other iron and aluminium compounds. The dissolved iron and aluminium would then form complexes with other products of organic matter decay or merely precipitate phosphate ions in solution instead of just adsorbing them. The strong correlation found between the constant K, related to bonding energy and "free" iron oxide suggests that this soil property is mainly responsible for the greater tenacity that some soils have for inorganic phosphate ions. Soils which abound in silica may possibly hold on to inorganic phosphate ions with less tenacity since the correlation found between bonding energy and silica is negative. This observation is in order since sand hardly retains any inorganic phosphate ions on the surfaces of its particles. The highly significant but negative correlation found between the constant K, related to bonding energy and pH is in agreement with the findings of Olsen and Watanabe (19 57) . As the value of this constant University of Ghana http://ugspace.ug.edu.gh 125 increases the bonding energy of the soil for phosphorus increases. Thus the acid soils retain more phosphorus per unit area and also hold the phosphorus with greater bonding energy than the alkaline soils. Those soils with high adsorption maximum generally produced maximum dry matter yield at \ the P adsorption maximum whilst those with low adsorption maximum produced maximum dry matter yield at the P adsorption maximum. This finding is in accord with that of Woodruff and Kamprath (1965). However, in this investigation the P saturation of the adsorption maximum at which maximum dry matter yield accurred appear to be markedly influeu- nced by the initial P status of the soil. Hence for the soil series Koforidua and Wacri with high initial P status maximum dry matter yields were produced at much lower P saturation of adsorption maximum than would be expected. It appears that those soils with low adsorp- maximum but high initial P did not necessarily have to saturate the adsorption maximum in order to produce maximum yield. It is clear from the relationship between the relative dry matter yield and the P saturation of adsorption maximum that optimum dry matter yields could be obtained at lower Q, §) p adsorption maximum. University of Ghana http://ugspace.ug.edu.gh 126 Infact one needs not necessarily saturate the P adsorp­ tion maximum in order to obtain optimum dry matter yield. Fertilizer phosphorus application rates must therefore be limited to P saturation of adsorption maximum below the P adsorption maximum. Of course, the initial P status of the soil must always be taken into consi­ deration. There are ample evidence in the data accumulated to show that the initial P status greatly influenced the P saturation of the adsorption maximum at which optimum dry matter yield was produced. The soils with high initial P invariably produced optimum dry matter yield just at § the P adsorption maximum. The high increases in dry matter yields from zero-P-treatment with unit addition of phosphorus up to I the P adsorption maximum show that there is moderate response from phosphorus application even at § the P adsorption maximum. These modest responses even at very low saturation of the adsorption maximum support the hypothesis of the partially reversible nature of fixed phosphorus. It was found in this investigation that for those soils with low initial P status (Ankasa, Agawtaw, Boi, Mamfe, Oyarifa, Tikobo, Toje) the range of increase in dry matter was wider. On the other hand, for those with high initial P status (Akuse, Koforidua, Wacri) the range of increase in University of Ghana http://ugspace.ug.edu.gh 127 dry matter was narrower. This finding is in accord with, results of F.A.O. fertilizer phosphorus trials reported by Ahn (1968). In one of the trials involving maize and cassava carried out for eight years in the forest zone of Ghana, phosphorus gave large and increa­ sing responses when applied to maize but had little ef­ fect on cassava. In the savanna areas it was found out that responses to phosphorus, although erratic, may be small to moderate. The Ankasa, Boi, Tikobo and Mamfe soils are from forest areas whilst Agawtaw, Oyarifa and Toje are from a savanna zone. The finding that these soils gave wider range of increase in dry matter with the addition of phosphorus was consistent. Data presented on per cent uptake of applied phos­ phorus reveal that the uptake of applied phosphorus was in the range 6 to 39 per cent for BH^PO^ and 3 to 39 per cent for K^HPO^ treatment. These ranges of recovery of applied phosphorus by the test-crop, although slightly low as regards their lower limits, are in close agreement with the range value of 10 to 30 per cent reported by Hemwall (1957) and Sauchelli (1965) . Fertilizer phospho­ rus added to soils are believed to be subjected to chemical precipitation and colloidal adsorption. Con­ sequently only a small proportion of the added fertilizer P becomes available to the plants which immediately University of Ghana http://ugspace.ug.edu.gh 128 grow on the soil. It is evident that on the acidic soils of the Forest Oxysol and Forest Ochrosol Great Soil Groups ( A n k a s a , Boi, Tikobo Mamfe) uptake of applied phosphorus was generally low. On the contrary on Koforidua and Wacri soils which belong to the Forest Oxysol and Forest Ochrosol Rubrisol Intergrade Great Soil Groups, respectively, uptake of applied phosphorus was very high. Again on the Savanna Ochrosols (Oyarifa and Toje) and the loamy sand Agawtaw soil uptake of applied phosphorus was moderately high. The uptake of applied phosphorus on the Akuse Soil of the Tropical Black Clay was also moderate. If the correlation coef­ ficients between adsorption maximum, bonding energy and uptake of applied phosphorus reflect on the relative importance of these parameters to uptake of applied phosphorus then the constant K, related to bonding energy is a much better index than adsorption maximum. There were marked increases in per cent P content and per cent P uptake by the test-crop with increased P application. These are evidences that when a certain saturation of the P adsorption maximum is reached availa­ bility of adsorbed phosphorus compounds increases con­ siderably. This observation adds support to a similar one made by E.G. Williams of Macauley Institute for Soil Research quoted by Sauchelli (1965) and also that University of Ghana http://ugspace.ug.edu.gh 129 of gell and Olson (1946). These workers observed that the phosphate reserve of every soil must be built up to a certain extent before phosphorus becomes available to plants. The observation made in this investigation is also consistent with that of G. Barbier and associates in France quoted by Sauchelli 0-965) . They found that a greater proportion of soluble phosphate added to soils remained after ten years in forms which are either extractable by dilute acids or are capable of returning spontaneously into the soil solution under natural conditions. This points to the fact that phosphate attached to phosphorus adsorption sites can, with time, be returned into the soil solution. On this note then, the above observation can be explained on the grounds that soils with high adsorption maximum are still capable of returning significant amounts of adsorbed phosphate into the soil solution for plant use, although over some lapse of time. Another report from an Illinois Bulletin, also quoted by Sauchelli (1965) which stated that added phosphorus fertilizer are not reverted to unavailable forms on contact with the soil also serves to buttress the above reasoning. The significance of the data on dry matter yield and uptake of applied P accumulated is that those soils with high adsorption maximum values but produced maxi- University of Ghana http://ugspace.ug.edu.gh 130 mum dry matter yield at lower saturation of the adsorp­ tion maximum would probably have larger residual ef­ fects. This is in view of the fact that although those soils with high adsorption maximum released enough phosphorus to produce maximum yield in dry matter at relatively low P saturation of adsorption maximum, their respective uptake of applied P were very low. This would indicate that still large amounts of phosphate remained attached to the P adsorption sites which would not be dislodged immediately. The attached phosphate would probably come into solution in the long run. These residual effects of adsorbed phosphorus may proba­ bly be very important for tree crops like oil palms, rubber and citrus which require considerable amount of phosphate supplied over a long period. The KH^PO^ phosphorus-carrier treatment resulted in comparatively high phosphorus uptake by the test-crop on almost all the soils than the K^HPO^ phosphorus- carrier. The former phosphorus-carrier has relatively high solubility in water than the latter. In this particular study the fertilizer phosphorus materials were added to the soils in the solid state. Hence their relative solubilities would greatly influence their uptake by plants. Moreover, plants are believed to take up their phosphorus almost exclusively as University of Ghana http://ugspace.ug.edu.gh 131 inorganic phosphate and principally as the YL^O 4 ion (Hagen and Hopkins, 19.55). The relatively high phos­ phorus uptake from the KH2P0 4 phosphorus-carrier is therefore consistent. The estimated P application rates are slightly high especially on the Ankasa, Boi, Tikobo, and Oyarifa soils. The NPK fertilizer application rates recommended for some food crops in Ghana are in the ranges 20 to 2 0 0 , 20 to 1 0 0 , and 10 to 100 kilograms per hectare, respectively, (CRI, 1974). For other parts of the tro­ pics and subtropics the recommended P application rates range from 20 to 200 kilograms P per hectare for a variety of crops (de Geus, 1973). The estimated P application rates for the Agawtaw, Akuse, Koforidua, Mamfe and Wacri soils fall within the ranges of P application rates recommended for some crops in Ghana and parts of the tropics and subtropics. In contrast, the estimated rates for Ankasa, Boi, Oyarifa, Tikobo and Toje are just too high in respect of recommended rates for Ghana but are still within the range for the tropics and sub-tropics. Except Oyarifa and Toje soils, the Ankasa, Boi and Tikobo soils are the extremely acidic soils of the high rainfall areas of Ghana. The estimated high application rates may be advantageous in that they will help build up the phosphorus fertility University of Ghana http://ugspace.ug.edu.gh 132 level to the point where further applications may be just beneficial. The high rates would probably leave a lot of phosphorus in the soils as residual phosphorus. Such a residual phosphorus may be beneficial to tree crops such as oil palm, rubber and citrus which are known to thrive best on those soils. The findings in this investigation that magnesium and manganese concentrations increased with increased phosphorus applications are in accord with those of Bingham et al. (1958). Although they used citrus as test-crop Bingham et al. Q.958) found that actual increases in magnesium and manganese absorption occurred with increased phosphorus applications. They found that iron availability was not very much affected. The only explanation Bingham et al. (1958) could offer for the increased manganese availability consequent upon in­ creased phosphorus additions was a possible formation of very soluble manganese phosphates in the treated soils which are readily absorbed by the plants. Nothing, however, is known of the mechanism bringing about in­ creased magnesium concentration of plant tissues with increased phosphorus additions. The decrease in cal­ cium content following increases in phosphorus appli­ cations appear unprecedented since no published work could be found in the literature. Bingham et. al. (1958) University of Ghana http://ugspace.ug.edu.gh 133 did not report of any decrease in calcium content of their indicator plant as a result of increased phos­ phorus additions. In any case Hahne (1966) found that the application of potash fertilizer markedly decreased the calcium and magnesium content in maize leaf. However, in this study equal amounts of potassium were added to all soil treatments. This therefore, rules out any pos­ sible effect of potassium treatment on some soil samples. Infact, phosphorus was the only element whose rates of application were varied. The formation of complex calcium-phosphate compounds with reduced solubility and hence availability to plants is not precluded. Further research, however, would be needed to ascertain the possible mechanism resulting in reduced availabi­ lity of calcium following increased phosphorus additions found in this investigation. Variable results were noted for the plant tissue content of iron as influenced by increased P additions. However, the finding that on some soils increased P additions caused reduced iron availability is in agreement with the findings of Bingham (1963), and of a research work on soybean by U.S. Department of Agriculture scientists reported by Tisdale and Nelson (1966). Bingham (1963) noted that high P substrate concentrations may restrict the movement of iron in some plants. Tisdale and Nelson (1966) also University of Ghana http://ugspace.ug.edu.gh 134 reported that high concentrations of phosphorus cause a deposition of iron on the surface or just inside the root of soybeans resulting in decreased iron avai­ lability, hence iron chlorosis. No published work was found in the literature to support the finding in this study in which on a few soils increased P additions increased the availability of iron. The plant tissue concentrations of calcium, magnesium, iron and manganese on soil samples given P treatments equivalent to the estimated P applica­ tion rates, generally fall within the ranges of suf­ ficiency given by some workers. No published data on nutrient sufficiency ranges for millet could be found in the literature. Nevertheless the sufficiency ranges for grain sorghum estimated by Lockman (19 72) and quoted by Jones and Eck Q.973) could be taken as standards. Grain sorghum and millet both thrive well on soil types with similar physical and chemical pro­ perties, and under the same climatic conditions. It is therefore reasonable to equate them in regard to their nutrient requirements. From Lockman1s 0-9 72) data a whole sorghum plant, twenty-three to twenty-nine days old gave phosphorus content of 0.30 to 0.60%. The calcium and magnesium concentrations were in the ranges 0.9 to 1-3% and 0.35 to 0.501, respectively. The iron University of Ghana http://ugspace.ug.edu.gh 135 and manganese concentrations were also in the ranges 160 to 250 ppm and 40 to 150 ppm respectively. In the investigation reported herein also the test-crop (millet) was harvested when the seedlings were forty- two days old. The whole seedlings were ground and analysed for the nutrient elements composition. Except for the iron concentration, the calcium, magnesium and manganese content of the plant tissues on soil samples given P treatments equivalent to the estimated P application rates fall within the ranges estimated by Locfcman. This is an evidence that with the apparantly high estimated P application rates there would still be a fair amount of secondary and micronutrient elements available to the plants for their proper growth and yield. University of Ghana http://ugspace.ug.edu.gh 136 CHAPTER 5 SUMMARY AND' CONCLUSION The present study has been concerned with: Ci) the phosphorus adsorption maximum of twelve selected Ghanaian soils using the Langmuir isotherm, (ii) the relation of adsorption maximum to availability of applied phosphorus to millet, Ciii) the relation of adsorption maximum to some soil properties, and Civ) estimation of rates of P application necessary to obtain optimum yield on the soil series used in the study. The soils of the Forest Oxysol, Forest Ochrosol, as well as, the Tropical Black Clay and the Tropical Black Earth Great Soil Groups (Ankasa, Abenia, Boi, Mamfe, Akuse, and Prampram, respectively) have comparatively high phosphorus adsorption maximum. The loamy sand Tikobo series and Koforidua soil have moderately high adsorption maximum. Agawtaw, Wacri, Toje and Oyarifa soils have comparatively low adsorption maximum. Abenia soil has the highest adsorption maximum whilst Oyarifa has the lowest adsorption maximum. University of Ghana http://ugspace.ug.edu.gh 137 Those soils with high adsorption maximum produced maximum dry matter yield at lower P saturation of the adsorption maximum. In contrast, those soils with low adsorption maximum produced maximum dry matter at higher P saturation of the adsorption maximum. The initial P status of the soils, however, greatly influenced the P saturation of the adsorption maximum at which maximum dry matter yield occurred. Those soils with high phos­ phorus adsorption maximum apparently are ahle to supply sufficient phosphorus for growth of crops at a lower saturation of the adsorption maximum than those with low phosphorus adsorption maximum and also low initial P- Evidence accumulating from this investigation indicates that presumably those soils with high adsorption maximum would have larger residual phosphorus effects. The residual phosphorus effects may be essential for tree crops such as oil palms, citrus and rubber- On those soils with extremely acidic reaction and which also have high phosphorus adsorption maximum, as well as, high bonding energy uptake of applied phosphorus by the test-crop was comparatively low. However, those soils responded quite favourably to phosphorus applica­ tion as indicated by the high increases in dry matter yields with unit addition of phosphorus. Only Akuse, University of Ghana http://ugspace.ug.edu.gh 138 Koforidua, and Wacri soils with, high initial P did not show any significant response to phosphorus applications. It is clear that the initial phosphorus status of a soil greatly influence its response to phosphorus applications. From data accumulated it can be concluded also that optimum dry matter yields could be obtained at lower P saturation of the adsorption maximum. Infact it may not be beneficial to saturate a soil beyond the P adsorption maximum. Economic fertilizer P applications should, there­ fore, always be limited to lower P saturation of the adsorption maximum than the P adsorption maximum. The soils of the Forest Oxysol Great Soil Group - Ankasa, Boi, Tikobo - will require 200, 175 and 152 kilograms P per hectare, respectively to produce optimum dry matter yield. Mamfe series which belongs to the Forest Ochrosol Great Soil Group will also require 85 kilograms P per hectare to produce optimum yield. On the Forest Oxysol, Forest Ochrosol Rubrisol Intergrade Great Soil Gaoups - Koforidua, and Wacri - 78 and 72 kilograms P per hectare, respectively, will be needed to produce optimum yield. The Savanna Ochrosol soils - Oyarifa and Toje will need 260 and 144 kilograms P per hectare, respectively, whilst the Black Clay soil, Akuse will need 88 kilograms P per hectare to produce University of Ghana http://ugspace.ug.edu.gh 139 optimum yield. Agawtaw also will require 74 kilograms P per hectare for optimum yield. This conclusion is tentative and must be confirmed by field trials to ascertain the beneficial effects in terms of crop yields and economics of these high phosphorus application rates. The KE^PO^ phosphorus - carrier resulted in higher dry matter yields and P uptake bn all the soils used than the K^HPO^ phosphorus-carrier- It is clear from this study that the H^PO 4 ion is preferred by plants to the HPO^ ~ ion. On the basis of plant tissue concentrations of calcium, magnesium, iron and manganese, it is concluded that successive additions of phosphorus fertilizer to soils will probably . Induce calcium shortages in plant tissues. Of course, further research will have to be conducted both in the greehouse and in the field to ascer­ tain the validity of this finding and also throw some light on the mechanism involved. Actual increases in magnesium and manganese availability are likely to occur with increased phosphorus additions. Availability of iron appear to be reduced with increased additions of P only on Boi and Tikobo soils. On Koforidua and Wacri soils however, iron availability appear to be enhanced by increased P additions. University of Ghana http://ugspace.ug.edu.gh 140 Clay, silica and free iron oxide C F e ^ J were all better correlated to adsorption maximum than pH, organic carbon, and free aluminium oxide (AljPg)* F°r the soils under investigation, therefore, clay, silica, and free iron oxide content appear to be factors which greatly ' influence phosphorus adsorption maximum. Bonding energy of the absorbent for the ahsorbate, was also better correlated to pH, free iron oxide, and silica. The correlation coefficients for pH,silica content and the bonding energy were negative. This points to the fact that the highly acidic soils with less silica content hold on to soil phosphorus with greater bonding energy than the alkaline or slightly acid soils with high silica content. The bonding energy of the absorbent for the absorbate was found to be a better index of uptake of applied phosphorus than the adsorption maximum. Thus soils with high bonding energy, no matter the value of the adsorption maximum,will still make less phosphorus available to plants for their growth and yield. University of Ghana http://ugspace.ug.edu.gh REFERENCES 141 1. AHENKORAH, Y. (1968). Phosphorus Retention Capacities of Some Cocoa Growing Soils of Ghana and Their Relationship with Soil Properties. Soil Sci. 105: 24-30. 2. AHN, P.M. (1968). A Review of Fertilizer Responses in West Africa, with Special Reference to Food Crops. Legon Jour. Agric. I: 16-23. 3. ARNON, D.I. (1953). The Physiology and Biochemistry of Phosphorus in Green Plants. In Pierre, W.H. and Norman, A.G. (eds) Soils and' Fertilizer Phosphorus in Crop Nutritions. Agronomy _4: 1-39. Academic Press Inc., New York. 4. ________ and HOAGLAND, D.R. (1940). Crop Production in Artificial Culture Solutions and in Soils with Special Reference to Factors Influencing Yields and Absorption of Inorganic Nutrients. Soil Sci. 50: 463-483. 5. BIDDAPPA, C.C., and VENKAT RAO, B.V. (1973). Studies on the Relationship Between Sesquioxides, Phosphorus Contents and Phosphorus Fixing Capacity of Coffee Soils of South India. J. Indian Soc. Soil Sci. 21: 155-159. 6 . BINGHAM, F.T. (1949). Soil Test for Phosphate. California Agriculture 3: No.8 , pp.11,14.(Soils and Fertilizers XIII, 482). University of Ghana http://ugspace.ug.edu.gh 7. BINGHAM, F.T. (1962). Chemical Soil Tests for Phosphorus. Soil Sci. 9_4: 87-95. 8 . -------- — — (1963). Relation Between Phosphorus and Micronutrients in Plants. Soil Sci. Soc. Amer. Proc. 2_7: 389-391. 9. BINGHAM, F.T., and MARTIN, J.P. (1956). Effects of Soil Phosphorus on Growth and Minor-Element Nutrition of Citrus. Soil Sci. Soc. Amer. Proc. 20: 382-385. 1 0 . -------------* ’ and CHASTIAN, J .A. (1958). Effects of Phosphorus Fertilization of California Soils on Minor-Element Nutrition of Citrus. Soil Sci. 8_6 : 24-31. 11. BIRCH, H.F. (1952). The Relationship Between Phosphate Response and Base Saturation in Acid Soils. J.Agric. Sci. 42: 276-285. 12. ------------- (1953). Phosphate Investigations. East.Afr. Agric. For. Res. Org. Rept. 1952. 70-73. (Soils and Fertilizers, XVI, 1689). ______________, and FRIEND, M.T. (1960). Phosphate Responses in Relation to Soil Tests. J.Agric. Sci. 54: 341-347. 142 University of Ghana http://ugspace.ug.edu.gh 14. BLACK, C.A. (1942). Phosphate Fixation in Kaolinite and Other Clays as Affected by pH, Phosphate Concentration, and Time of Contact. Soil Sci. Soc. Amer. Proc. 7_: 123-133. 15 . ------------ (1957). Laboratory Methods of Soil Investigation. Soil Fertility. 3rd Edition. Agronomy 556B: 29-30. 16. ------------ , EVANS, D.D., WHITE, J.L., ENSMINGER, L.E., and CLARK, F.E. (1965). Methods of Soil Analysis, Part 2. Chemical and Micro­ biological Properties. Agronomy 9: 959, 963. 17. BLENKINSOP, A. (1938). Soil Studies for Advisory Purposes. Agronomy Progress, IJS: 184-193 (Soils and Fertilizers II, 2). 18. BONNER, J., and GALSTON, A.W. (1952). Principles of Plant Physiology.W.H. Freeman and Co., San Francisco, pp. 48, 53, 54. 19. BRADLEY, D.B., and SIELING, D.H. (1953). Effect of Organic Anions and Sugars on Phosphate Precipitation by Iron and Aluminium as Influenced by pH. Soil Sci. 76: 175-179. 143. University of Ghana http://ugspace.ug.edu.gh BRAY, R.H. (1948), Correlation of Soil Tests with Crop Response to Added Fertilizers and with Fertilizer Requirement. In Kitchen, H.B. (ed.) Diagnostic Techniques for Soils and Crops, Chapter II, pp. 5.3-86. American Potash Institute, Washing­ ton, D.C. ---------------- , and DICKMAN, S.R. (1942). Adsorbed Phosphates in Soils and Their Relation to Crop Responses. Soil Sci.Soc.Amer. Proc. 6: 312-319. ------------ , and KURTZ, L.T. (1945). Determination of Total, Organic, and Available Forms of Phosphorus in Soils. Soil Sci. _59: 39-45 . BROMFIELD, S.M. (1964), Relative Contribution of u Iron and Aluminium in Phosphate Sorption by Acid Surface Soils. Nature 201: 321-322. ------ ^ '6 -, . studies of the Relative Importance of Iron and Aluminium in the Sorption of Phosphate by some Australian Soils. Aust.J. Soil Res. 3: 31-44. — , and WILLIAMS, E.G. (1963). An Examination of Biological Reduction Method for Estimating Active Iron in Soils.J.Soil Sci. 14: 346-359. University of Ghana http://ugspace.ug.edu.gh 26. BURD, J.S. (1948). Chemistry of Phosphate Ion in Soil System. Soil Sci. 65: 227-247 27. _____________ and MURPHY, H.F. (1939). The Use of Chemical Data in the Prognosis of Phosphate Deficiency in Soils. Hilgardia 12^ : 323-340. 28. CHANDLER, W.V. (1941). Phosphorus Adsorption by Five Alabama Soils as Influenced by Reaction, Base Saturation, and Free Sesquioxides. J.Amer. Soc.Agron. 3_3: 1-12 29. CHAPMAN, H.D. (1951). Why so much Nitrogen. Citrus Leaves. 31: 6-7, 24-26, 42. 30. COLE, C.V., OLSEN, S.R., and SCOTT, C.O. (1953). The Nature of Phosphate Adsorption by Calcium Carbonate. Soil Sci.Soc.Amer.Proc. 17_: 352-356 . 31. COLEMAN, R. (1942). The Adsorption of Phosphate by Kaolinite and Montmorillonitic Clays. Soil Sci. Soc.Amer. Proc. 7_: 134-138. 32. --------.---*— (1944a). Phosphorus Fixation by the Coarse and Fine Clay Fraction of Kaolinitic and Montmorillonitic Clays. Soil Sci. 5_8: 71-77. 33 . ----------- — (1944b). The Mechanism of Phosphate Fixation by Montmorillonitic and Kaolinitic Clays. Soil Sci. Soc.Amer. Proc. 9: 72-78. 145. University of Ghana http://ugspace.ug.edu.gh 34. COLEMAN, N.T., THORUP, J.T., and JACKSON, W.A. (1960). Phosphate Sorption Reactions that Involve Exchangeable Aluminium. Soil Sci. 90: 1-7 35. COLWELL, W.E. (1943). A Biological Method for Determining the Relative Boron Contents of Soils. Soil Sci. 56: 71-94. 36. COOKE, G.W. (1951). Fixation of Phosphate During Acid Extraction of Soils. J.Soil Sci. 2: 254-262 37. C.R.I. (19 74). Guide to the Production of Some Crops in Ghana. Crop Research Institute of the Council for Scientific and Industrial Research, Kumasi, Ghana. 38. DAS, S. (1930). An Improved Method for the Determina­ tion of Available Phosphoric Acid in Soils. Soil Sci. 30: 33-49. 39. DAUGHTREY, Z.W., GILLIAM, J.W., and KAMPRATH, E.J. (1973). Phosphorus Supply Characteristics of Acid Organic Soils as Measured by Desorption and Mineralization. Soil Sci. 115: 18-24. 40. DAVIS, L.E. (1935). Sorption of Phosphate by Non-Calcareous Hawaiian Soils, Soil Sci. 40: 129-158 41. DEAN, L.A. (1949). Fixation of Soil Phosphorus. Advances in Agronomy 1: 391-411. 146 University of Ghana http://ugspace.ug.edu.gh 42. De-DATTA, S.K. (1964). Availability of Phosphorus and Utilization of Phosphate in Some Great Soil Groups. Hawaii Diss.Abstr. 25: 716. 43. De-ENDREDY, A.S., and MONTGOMERY, C.W. (1954). Some Aspects of the Gold Coast Forest Soils. Trans. Int. Cong, on Soil Sci. 5th Congress, Leopodville 3: 268-273. 44. De GEUS, J.G. (1973). Fertilizer Guide for the Tropics and Subtropics. 2nd Edition.Published by Centre d'Etude de 1'Azote, Zurich. 45. DROSDOFF, M. , and TRUOG, E. (1935). A Method for Removing and Determining the Free Iron Oxide in Soil Colloids. J. Amer. Soc. Agron. _27: 312-317. 46. DYER, B. (1894). On the Analytical Determination of Probably Available Mineral Plant Food in Soils. Trans. Chem. Soc. 6_5: 115-167 47. EGNER, H. (1944). The Egner Lactate Method for Phosphate Determination. Amer. Fertilizer 94-: 5-7, 22,24,26. 48. ELLIS, R. , and TRUOG,E. (1955). Phosphate Fixation by Montmorillonite. Soil Sci.Soc.Amer. Proc. 19: 451 - 454. 147 University of Ghana http://ugspace.ug.edu.gh 49. ENSMINGER, L.E. (1948). The Relationship Between Water Lost and PO^ Adsorbed on Phosphating Clay Minerals and Soil Colloids.Soil Soc. Amer. Proc. 170-174. 50* FORSEE, W.T. (1945). Soil Investigation. Soil Test Methods. Florida Agric.Expt. Station Ann.Report. 1944/45: 199-202. (Soils and Fertilizers XIII, 47). 51. FRIED, M., and DEAN, L.A. (1955). Phosphate Retention by Iron and Aluminium Cation Exchange Systems. Soil Sci.Soc. Amer. Proc. .19: 143-147. 52. ------- and SHAPIRO, R.E. (1956). Phosphate Supply Pattern of Various Soils.Soil Sci.Soc.Amer.Proc.20: 471-475. 53. GALINDO, G.G., and SCHALSCHA, E.B. (1972). Phosphate Sorption Capacity of Clay Fractions of Soils Derived from Volcanic Ash. Geoderma, 1_: 225-232 . 54. GHANI, M.Q. (1943). The Use of 8-Hydroxyquinoline as a Means of Blocking Active Iron and Aluminium in the Determination of Available Phosphoric Acid of Soils by Dilute Acid Extractions. Indian J.Agric.Sci. 13: 562-565. 55. GILE, P.L. (1933). The Effect of Different Colloidal Soil Material on the Efficiency of Superphosphate. U.S. Dept. Agric. Tech. Bull. 371. 1933. 56. GRANT, C.L., and MACNAUGHLAIN, D.J. Jr.(1968). Preparation of Biological Samples and Standards for Spectro- chemical Analysis. Abstr.Appl.Spectr. 2 2 : 365. 148 University of Ghana http://ugspace.ug.edu.gh 57. GUNARY, D. (1970). A New Adsorption Isotherm for phosphate in Soil. J. Soil Sci. 21: 12-11. 58. HAGEN, C.E., and HOPKINS, H.T. (1955). Ionic Species in Orthophosphate Absorption by Barley Roots. Plant Physiol. 30: 193-199. 59.HAHNE, H.C.N. (1966). Nutrient Balance in the Maize Leaf in Relation to Fertilizer Treatment and Grain Yield. (Research Project N-Th.13). Agric Res. Part.I. - 1-2 and 3, pp. 369-373. 60. HALL, A.D. (1905). The Analysis of the Soil by Means of the Plant.J.Agric.Sci. 1: 65-88. 61. HARDY, F., and AM0R0S0-CENTEN0,E.H. (1938). Some Gold Coast Cocoa Soils-.II. Imperial Coll.Trop.Agric. Trinidad. 6th Ann.Rept.on Cocoa Res.1937 pp. 32. 62. HEMWALL, J.B. (1957). Fixation of Phosphorus by Soils. Advances in Agronomy. 9.: 95-112. 63. HEWITH,E.J. (1963). The Essential Nutrient Elements: Requirements and Interactions in Plants. In Steward, F.C. (ed.) Plant Physiology- A Treatise, Vol.III. Academic Press, New York. pp. 137-360. 64. HIBBARD, P.L. (1931). Chemical Methods for Estimating the Availability of Soil Phosphorus. Soil Sci. 31: 437-467. 149 University of Ghana http://ugspace.ug.edu.gh 65. HOAGLAND, D.R., and ARNON, D.I. (1950). The Water Culture Method for Growing Plants without Soil. California Agric.Expt. Sta. Cir. 347. 6 6 . HSU, Pa Ho, and RENNIE, D.A. (1962). Reactions of Phosphate in Aluminium Systems: I. Adsorption of Phosphate by X-Ray Amorphous Aluminium Hydroxide. Canadian J.Soil Sci. 4_2: 197-209. 67. JENNY, H., VLAMIS, J., and MARTIN, W.E. (1950). Green­ house Assay of Fertility of California Soils. Hilgardia 20_: 1-8. 6 8 . JOHNSON, C.M. (1966). "Molybdenum". In Chapman, H.D. (ed.) Diagnostic Criteria for Plants and Soils. Univ. of California Division of Agric.Sciences. Chapter 20, pp. 286-301. 69. JONES, G.H.G. (1949). Proceedings on the First Common­ wealth Conference on Tropical and Subtropical Soils. Discussion on Soil Fertility Problems. Commonwealth Bur.Soil Sci.Tech.Comm. £ 6 : 180 70. JONES, J.B. Jr., and ECK. H.V. (1973). Plant Analysis as an Aid in Fertilizing Corn and Grain Sorghum. In Walsh, L.M., and Beaton, J.D. (eds). Soil Testing and Plant Analysis. Revised Edition. Soil Sci. Soc. Amer.Inc.,Madison, Wisconsin. Chapter 21, pp. 349. 150 University of Ghana http://ugspace.ug.edu.gh 71. JONES, J.B. Jr., STEYN, W.J.A. (1973). Sampling, Handling, and Analyzing Plant Tissue Samples. In Walsh, L.M., and Beaton, J.D. (eds). Soil Testing and Plant Analysis. Revised Edition. Soil Sci.Soc.Amer.Inc., Madison, Wisconsin. Chapter 16, pp. 249. 72. KARDOS, L.T. (1964). Soil Fixation of Plant Nutrients. In Bear, F.E. (ed). Chemistry of the Soil, 2nd Edition. Amer.Chem.Soc.Monograph Series pp.381. 73. KELLEY, J.B. , and MIDGLEY, A.R. (1943). Phosphate Fixation- An Exchange of Phosphate and Hydroxyl Ions. Soil Sci. 55: 167-175 74. KURTZ, L.T. (1953). Inorganic Phosphorus in Acid and Neutral Soils. Agronomy 4-: 59-88 . 75 __________ __, De-TURK, E.E., and BRAY, R.H. (1946) Phosphate Adsorption by Illinois Soils. Soil Sci. 61: 111-124. 76. LANGMUIR, I. (1918). The Adsorption of Gases on Plane Surfaces of Glass, Mica and Platinum J.Amer.Chem. Soc. 40: 1361-1403. 77- LARSEN, J.E. (1967). Soil Phosphorus. Advances in Agronomy 19: 151-210. 78. , WARREN, G.F., and LANGSTON,R. (1959).Studies of Phosphorus Availability in Organic Soils. Soil Sci.Soc.Amer.Proc. 2 2 : 336-339 1 University of Ghana http://ugspace.ug.edu.gh 79. LOCKMAN, R.B. (I970)• Plant Sample Analysis as Affected by Sample Decomposition Prior to Laboratory Processing. Comm. Soil Sci. Plant Anal. L^: 13-19. 80. — — (1972). Mineral Composition of Grain Sorghum Plant Samples. Part III. Suggested Nutrient Sufficiency Limits at Various Stages of Growth and Under Different Environments. Comm. Soil Sci. Plant Anal. 3: 271-282. 81. MATTSON, S. (1931). The Laws of Colloidal Behavior: VI*Amphoteric Behavior. Soil Sci. 52: 343-365. 82. ------- , ALVSAKER, E., KOUTLER-ANDERSON, E., BARKOFF, E., andVAHTRAS, K. (1950). Phosphate Relationship in Soil and Plant: VI. The Effect of Phosphate Solubility in Pedalfer Soils. Ann. Roy.Agric. Coll. Swed. 1]_: 141-160. 83. McGEORGE, W.T. (1939). Factors Influencing the Availa­ bility of Native Soil Phosphate and Phosphate Fertilizers in Arizona Soils. Agric.Expt. Sta. Tech. Bull. %2_: 293-331. (Soils and Fertilizers III, 3). 84. METZGER, W.H. (1934). Distribution of Fertilizer Residues in the Soil After Fourteen Years of a Fertilizer Experiment with Alfalfa. J.Amer.Soc. Agron. 2_6: 620-625. 85. — ---------------(1941). Phosphorus Fixation in Relation to the Iron and Aluminium of the Soil.J.Amer. Soc. Agron. 33: 1093-1099. 8 6 . MILLAR, C.E. (1965). Soil Fertility. John Wiley and Sons, Inc., New York. pp. 248. 87. MORGAN, M.F. (1937). The Universal Soil Testing System. Connecticut Agric. Expt. Sta. Bull. 392: 129-159. 152 University of Ghana http://ugspace.ug.edu.gh IS 3 8 8 . MORTENSEN, J.L., and HIMES, F.L. (1964). Soil Organic Matter. In Bear, F.E. (ed) Chemistry of the Soil. 2nd Edition,Amer. Chem.Soc. Monograph Series, pp. 209. 89. MOSER, U.S., SUTHERLAND, W.H., and BLACK, C.A. (1959) Evaluation of Laboratory Indexes of Absorption of Soil Phosphorus by Plants: I. Plant and Soil 10: 356-374. 90. MURPHY, H.F. (1939). The Role of Kaolinite in Phosphate Fixation. Hilgardia 12: 343-382. 91. NELSON. W.L., MEHLICH, A., and WINTERS, E. (1953). The Development, Evaluation, and Use of Soil Tests for Phosphorus Availability. Agronomy 4-: 153-188 . 92. NYE, P.H. (1952). Studies on the Fertility of Gold Coast Soils!III. The Phosphate Status of the Soils. Empire J.Expt. Agric. 20: 47-55. 93. OLSEN, S.R., COLE, C.V., WATANABE, F.S., and DEAN, L.A. (1954). Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate. U.S. Dept. Agric.Circ. 939. 9 4 . -------------, and WATANABE, F.S. (1957). A Method to Determine a Phosphorus Adsorption Maximum of Soils as Measured by the Langmuir Isotherm.Soil Sci.Soc.Amer.Proc. Tl: 144-149. 95. OLSON, L.S. (1946). Factors Affecting the Relationship Between Laboratory Tests for Soil Phosphorus and Crop Response to Applied Phosphate. Soil Sci. Soc.Amer.Proc. 1 0 : 443-445. 96. OTENG, J.W. (1969). Studies on the Availability of Phosphorus in Representative Ghana Soils.M.Sc.Thesis, Dept, of Crop Sci.Univ. of Ghana,Legon. 1969. University of Ghana http://ugspace.ug.edu.gh 154 97. OZANNE, P.G., and SHAW, T.C. (1967). Phosphate Sorption by Soils as a Measure of Phosphate Requirement for Pasture Growth. Aust.J .Agric. Res. 18; 601-612. 98. PEECH, M., and ENGLISH, L. (1944). Rapid Micro-Chemical Soil Tests. Soil Sci. 5_7: 167-195 . 99. PERKINS, A.T., and KING, H.H. (1944). Phosphate Fixation by Soil Minerals: II. Fixation by Iron, Silicon, and Titanium Oxides. Soil Sci. 5_8: 243-250. 100. PIGGOT, C.J. (1953). The Use of Phosphate Fertilizers in Sierra Leone. Ann.Rept. Dept.Agric. Sierra Leone, Freetown. 101. PISSARIDES, A., STEWART, J.W.B., and RENNIE, D.A. (1968). Influence of Cation Saturation on Phosphorus Adsorption by Selected Clay Minerals. Canadian J.Soil Sci. 48: 151-157. 102. RAJAGOPAL, C.K., and IDNANI, M.A. (1963). Some Aspects of Phosphorus Fertilization in the Nilgiri Soil. J. Indian Soc.Soil Sci. IT: 141-150. 103. RENNIE, D.A., and McKERCHER, R.B. (1959). Adsorption of Phosphorus by Four Saskatchewan Soils. Canadian J. Soil Sci. 39:64-75. 104. REUTHER, W., and LABANAUSKAS, C.K. (1966)."Copper." In Chapman H.D. (ed.) Diagnostic Criteria for Plants and Soils. Univ. of California Division of Agric. Sciences. Chapter II, pp. 157-179. 105. RUSSELL, G.C., and LOW, P.F. (1954). Reactions of Phosphate with Kaolinite in Dilute Solution. Soil Sci.Soc. Amer.Proc. 18: 22-25. University of Ghana http://ugspace.ug.edu.gh 155 107. 108. 109. 110. 111. 1 1 2 . 113. 114. 106. SAINI, G.R., a n d MACLEAN, A.A. (1965). Phosphorus Retention Capacities of Some New Brunswick Soils and Their Relationship with Soil Properties. Canadian J. Soil Sci. 4_5: 15-18. SALISBURY, F.B., and ROSS, C. (1969). Plant Physiology. Wadsworth Publishing Co. Inc., Belmont, California 1969. Chapter, 10. pp. 191. SALTER, R.M. , and AMES, J.W. (1928). Plant Composition as Guide to the Availability of Soil Nutrients. J.Amer.Soc.Agron. 20: 808-836. SAUCHELLI, V. (1965), Phosphate in Agriculture. Reinhold Publishing Corp., New York. 1965. SAUNDERS, W.M.H. (1965). Phosphate Retention by New Zealand Soils and Its Relationship to Free Sesquioxide, Organic Matter and Other Soil Proper­ ties. New Zealand J. Agric.Res. 8^: 30-57 SCARSETH, G.D., and TIDMORE, J.W. (1934). The Fixation of Phosphate by Soil Colloids.J.Amer.Soc.Agron. 26: 138-151. SCHUSTER, C.E., and STEPHENSON, R.E.(1940). Sunflower as an Indicator Plant of Boron Deficiency in Soils. J.Amer.Soc.Agron. 32: 607-621 SELL, O.E., and OLSON, L.C. (1946). The Effect of Surface-Applied Phosphate and Limestone on Soil Nutrients and pH of Permanent Pasture. Soil Sci. Soc.Amer. Proc. II: 238-245. SHUKLA, S.S., SYERS, J.K., WILLIAMS, J.D.H.,AMSTRONG,D.E ., and HARRIS, R.F. (1971). Sorption of Inorganic Phosphate by Lake Sediments.Soil Sci.Soc.Amer. Proc. 35: 244-249. University of Ghana http://ugspace.ug.edu.gh 156 115. 116. 117. 118. 119. 1 2 0 . 1 2 1 . 1 2 2 . SINGH, T., GANGWAR, B.R., and MEHROTRA, C.L.(1971). Phosphate Sorption Studies to Determine Phosphate Requirement of Soil for Cereal Crops. International Symposium on Soil Fertility Evaluation Proceedings. New Delhi. I: 111-118. STEPHENS, D. (1968). Fertilizer Trials in Cotton and Other Annual Crops on Small Farms in Uganda. J.Expt. Agric. _4 : 49-59. STEPHENSON,R.E., and SCHUSTER,C.E.(1941). Laboratory, Greenhouse, and Field Methods of Studying Fertilizer Needs of Orchard Soils. Soil Sci. 52: 137-153. STEYN, W.J.A. (1959). Leaf Analysis. Errors Involved in the Preparative Phase, J.Agric.Food Chem. 7: 344-348. STOUT, P.R. (1939). Alterations in the Crystal Structure of Clay Minerals as a Result of Phosphate Fixation. Soil Sci.Soc.Amer.Proc. 4: 177-182. STRUTHERS, P.H., and SIELING, D.H. (1950). Effect of Organic Anions on Phosphate Precipitation by Fe and Al as Influenced by pH. Soil Sci. 69: 205-213. SWENSON, R.M., COLE, C.V., and SIELING, D.H.(1949). Fixation of Phosphate by Fe and Al and Replacement of Organic and Inorganic Ions. Soil Sci. 67: 3-22. SYERS, J.K., BROWMAN, M.G., SMILLIE, G.W. and COREY, R.B. (1973). Phosphate Sorption by Soils Evaluated by the Langmuir Adsorption Equation. Soil Sci. Soc.Amer.Proc. 37: 358-363. University of Ghana http://ugspace.ug.edu.gh 124. 125. 126. 127. 128. 129. 130. 131. 132. 12 3. SYERS, J.K., EVANS, T.D., WILLIAMS, and MURDOCK, J.T. (1971). Phosphate Sorption Parameters of Representative Soils From Rio Grande De Sol, Brazil. Soil Sci. 112: 267-275. ------------, MURDOCK, J.T., and WILLIAMS, J.D.H (1970). Adsorption and Desorption of Phosphate by Soils. Soil Sci. Plant Analysis 1: 57 - TAMM, 0. (1922). "Medd. Skogsforsoksanst. Stockh." 19: 387-404. TAUBER, H. (1949). Chemistry and Technology of Enzymes. John Wiley and Sons, New York, pp. 4. THOMPSON, E.J. (1958). Evaluation of Different Laboratory Indexes of Soil Phosphorus Availability- M.Sc. Thesis. Iowa State College. Ames. ------------------- » OLIVERIA, A.L.F., MOSER, U.S., and BLACK, C.A. (1960). Evaluation of Laboratory Indexes of Absorption of Soil Phosphorus by Plants: II. Plant and Soil 13: 28-38. THOMPSON, L.M. (1957). Soils and Soil Fertility. 2nd Edition McGRAW-HILL Book Co., Inc. New York. pp. 251. TISDALE, S.L., and NELSON, W.L. (1966). Soil Fertility and Fertilizers. 2nd Edition. Macmillan Co., New York. pp. 326. TOTH, S.J. (1937). Anion Adsorption by Soil Colloids in Relation to Changes in Free Iron Oxides. Soil. Sci. 44: 299-314. "— (1939). The Effect of Free Iron Oxide Removal on Some Properties of Soil Colloids. Soil Sci. 48: 385-401. 157 University of Ghana http://ugspace.ug.edu.gh 133. 134. 135. 136. 137. 138. 139. 140. 141. TRUOG, E. (1930). The Determination of the Readily Available Phosphorus of Soils•J.Amer.Soc.Agron. _23: 874-882. _ 5 a n d MEYER, A.H. (1929). Improvements in the Deniges Colorimetric Method for Phosphorus and Arsenic. Ind. Eng. Chem.Anal. Ed. I: 136-139. ---------- ’TAYLOR, J.R., PEARSON, R.M., WEEKS, M.E. and SIMONSON, R.W. (1936). Procedure for Special Type of Mechanical and Mineralogical Analysis. Soil Sci.Soc.Amer.Proc. 102-112. UDO, E.J., and UZU, F.O. (1972). Characteristics of Phosphorus Adsorption by Some Nigerian Soils. Soil Sci. Soc. Amer. Proc. _36: 879-883. ULRICH, A. (1943) . Plant Analysis as a Diagnostic Procedure. Soil Sci. 5_5: 101-112. VANDECAVEYE, S.C. (1948). Biological Methods of Determining Nutrients in Soil. In Kitchen, H.B. (ed.) Diagnostic Techniques for Soils and Crops. American Potash Institute. Chapter VII pp. 199. WALKLEY, 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 Sci. 37: 29 - 38. WEIR, C.C., and SOPER, R.J. (1962). Adsorption and Exchange Studies of Phosphorus in Some Manitoba Soils. Canadian J. Soil Sci. 4_2: 31-42. 'and----------(1963). Interaction of Phosphate with Ferric Organic Complexes. Canadian J. Soil Sci. 43: 393-399. 158 University of Ghana http://ugspace.ug.edu.gh 159 142. WILD, A. (1950). The Retention of Phosphate by Soil. J. Soil Sci. I: 221-238. 143. WILLIAMS, E.G., SCOTT, H.M., and McDONALD, MARGARET, J. (1958). Soil Properties and Phosphate Sorption. J.Sci.Food Agric. 9: 551-559. 144. WOODRUFF, J.R., and KAMPRATH, E.J. (1965). Phosphorus Adsorption Maximum as Measured by Langmuir Isotherm and Its Relationships to Phosphorus Availability. Soil Sci. Soc. Amer. Proc. 29: 148-150. 145. YOUNGE, O.R., and PLUCKNETT, D.L. (1966). Quenching the High Phosphorus Fixation of Hawaiian Latosols. Soil Sci.Soc.Amer.Proc. 30:653-655. University of Ghana http://ugspace.ug.edu.gh