ABSTRACT EVALUATION OF CERTAIN CHEMICAL TRANSFORMATIONS OF SOLUBLE FERTILIZER PHOSPHORUS APPLIED TO THREE MICHIGAN SOILS By Carlostadio Sanchez Field, greenhouse and laboratory studies were initiated in l96h to evaluate certain chemical trans- formations that soluble ph05phate may undergo when applied to three Michigan soils. The products of the trans- formations were related to the ”available P“, extracted by the Bray P-I test,’ and to ”P fixed.” Corn was grown in the field and in the greenhouse in order to evaluate the yield response of this cr0p to fertilizer P. The use of fertilizer P in the field and greenhouse, caused significant increases in yield. However, the response in the field was not nearly as large as that obtained in the greenhouse. Corn grown on the Conover loam from Shiawassee county responded more to P fertilizer than that grown on a similar soil type from Ingham county. 0n the Pewamo silty clay loam an intermediate response to P was obtained in the greenhouse. The P content of the leaves of corn grown in the field was positively correlated with the use of increasing amounts of P fertilizer. However, the correlation was highest early in the season. The P uptake by corn plants and the P left in the soil after the application of fertilizer, as measured with the Bray P-l test accounted for only a small fraction of the total fertilizer P used. This effect was most evident on the Pewamo silty clay loam. An incubationstudy showed that the percent P fixed decreased significantly as the amount of fertilizer P increased. This was most evident on the soil from Ingham county, which contained the lowest quantity of colloidal material. Although most of the applied P was fixed during the first week after application, there was a tendency for the percent P fixed by each of the soils to increase with time, especially where high rates of P were used. The fractionation of the inorganic P at the end of twenty-two weeks, showed that from 70 to 99 percent of the applied P was recovered as water soluble, aluminum, iron, and calcium phosphate. The aluminum phosphate fraction contributed the most to these values, followed in order by the iron, the calcium, and finally, by the twater soluble fraction. A multiple regression study showed that the P extracted from the soil with the Bray P-l method, was predominately aluminum phOSphate followed by iron and then by calcium phOSphate. The relative importance of the calcium phOSphate fraction with reSpect to other P forms was discussed. It is thought that some dicalcium phosphate may have been included in the fraction generally labeled as ”aluminum phosphate.” EVALUATION OF CERTAIN CHEMICAL TRANSFORMATIONS OF SOLUBLE FERTILIZER PHOSPHORUS APPLIED TO THREE MICHIGAN SOILS By Carlostadio Sanchez A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science 1965 ACKNOWLEDGMENTS The author wishes to express his sincere gratitude to his major professor Dr. Lynn S. Robertson without his great interest, constant advice and patient counseling many aspects of the study would have been immeasurably more difficult; Dr. John C. Shickluna for his cooperation and assistance in conducting the laboratory phase of this study; Dr. Boyd G. Ellis for helpful criticism and suggestions; Professor Hubert M. Brown for his help with the statistical analysis; Dr. Ray L. Cook, Dr. Eugene C. Doll, and Dr. Norman E. Good for serving as members of his guidance committee. The Rockefeller Foundation provided financial assistance, which enabled him to pursue and complete this study. The author is indebted to his wife, Augusta, whose encouragement and assistance aided him in the fulfillment of all the requirements for his graduate study. CHAPTER VI. VII. TABLE OF CONTENTS INTRODUCTION. REVIEW OF LITERATURE. A. Reactions of phosphorus in the soil . . . . . . . l. Mechanisms of phosphorus fixation. . . . . (a) Sorption reactions. (b) Chemical precipitation. . Behavior of phosphorus fertilizer in-soils . B. Availability of primary phos- ~phorus sources . . . . . . C. Prediction of phosphorus avail- -ability to plants. EXPERIMENTAL PROCEDURE. A. Field experiments. 8. Greenhouse Trials. C. Laboratory studies I. 2. 3. RESULTS Incubation'study. Soil analysis Plant analysis AND DISCUSSION. A. Field experiments. B. Greenhouse experiments C. Laboratory experiments GENERAL DISCUSSION. . . . . . . . SUMMARY AND CONCLUSIONS . . . . . . LITERATURE CITED. iii PAGE IO Ih I6 20 20 25 26 26 27 28 29 29 3h #2 57 6O 64 TABLE LIST OF TABLES Some chemical and physical charac- teristics of the three soils where field experiments were located . Changes in P content (Bray P-l) of three soils as affected by the addition of P fertilizer and the production of one corn crop, (Field experiments) The P content of corn leaves at three ~stages of growth as affected by rates of P applied to two soils (Field experiments) . . . . . . . Linear correlation coefficients of the P content of corn leaves at three stages of growth as related to grain yields and fertilizer rates on two soils (Field experiments) The dry weight, percent P, and P uptake of corn grown on three soils as related to rates of P fertilizer (Greenhouse experiments) Simple correlation coefficients of P uptake, percent P, dry weight of corn plants, and P applied to three soils (Greenhouse experiments) Change in P content (Bray P-l) of three soils as affected by the addition of P fertilizer and the production of one corn crop (Greenhouse experiments) Effect of time of incubation and rates of P application to three soils on available P as determined by Bray P-l soil test. . . . . . . . . . . . . . Simple correlation coefficients between P a plied and soil tests for P (Bray P-l), after various incubation times iv 3l 33 35 36 39 Al #3 A5 LIST OF TABLES, Continued TABLE IO ll l2 l3 lh l5 Amount of P fixed on three soils as related to time of incubation and rates of P applied . Percent ngixed by three-soils as affected by time of incubation and rates of P application . Water soluble, aluminum, iron, and calcium phosphate in three soils after 22 weeks of incubation as affected by rates of P fertilizer. . . Percent P recovered as water soluble, aluminum, iron, and calcium phosphate “in three soils after 22 weeks of incubation . Percent of P fixed and the total P converted to aluminum, iron, and calcium phosphate as affected by rates of P fertilizer . . . . . . . . . Simple and partial correlation coeffi* cients which show the relationships between soil tests (Bray P- l) and the aluminum, iron, and calcium phosphate forms in three soils that were incubated for 22 weeks AB 50 52 5h 55 FIGURE LIST OF FIGURES Page Effect of Applied P to Two Locations on the Yield of Corn Grain (Field Experiments). . ... . 30 P uptake as Affected by the Addition of P to Three Soils (Greenhouse Experiments) . . . . . 38 I. INTRODUCTION When soluble phosphates are-applied to soils, most of the P is rapidly converted to relatively insoluble forms. The specific reactions and the resulting discrete compounds are poorly defined since many factors play a part in determining the chain of events that take place between the time fertilizer is added to the soil and the formation of the ultimate products in the reaction sequence. A true equilibrium may never be established since the soil system is a complex and heterogennus mixture of many dynamic components. Chang: and Jackson (l7) suggested a fractionation procedure based on the solubility in various solutions of discrete phosphate compounds. This method offers an approximation to study the amounts of specific compounds formed after phosphates are applied to soils. The available soil P, as determined by the Bray P-I test (II), is used as the basis for fertilizer recommendations in Michigan. High correlations with plant growth have been obtained under variable conditions. Therefore, by relating the Bray P-l test and the yield response of applied P to the several phosphate fractions, it should be possible to charac- terize the available forms of P in the soil. This study was initiated in l96h with the following objectives: (a) To determine the effect of several rates of fertilizer P upon the water soluble, aluminum, iron, and calcium phosphate levels in three soils. (b) To relate the Bray P-l soil test after various time intervals of P addition to the several P fractions and to the use of.applied P by corn. II. REVIEW OF LITERATURE A. Reactions of Phosphorus in the Soil l. Mechanisms of phosphorus fixation. The removal of phosphates from the soil solution, after soluble-phosphate fertilizers are applied, is a common phenomenon. The immediate recovery of fertilizer P by a cr0p is equal to only ID to 30 percent of the quantity added to the soil (#5). It is assumed that the remaining 70 to 90 percent is converted to forms less soluble and not immediately available to plants. The process by which the availability of P is decreased and its mobility restricted is designated as "phosphorus fixation” (l0, SI, 63). Several mechanisms have been proposed to eXpdain P fixation. These include: (I) adsoption of phosphates by the soil complex; (2) precipitation of P from the soil solution; and, (3) biological immobilization, which includes the use of P by soil micro-organisms. This discussion is concerned with the first two mechanisms. Very little is known about the role of soil microflora in P immobilization. (a) Sorption reactions. It is known that soil colloids adsorb anions. The practical significance of such reaction has not been determined. Several workers (SI, 66, 72) suggested a reversible exchange between phosphate and hydroxyl on the edges of the crystal lattice of clay minerals. Murphy (66), however, found that kaolinite fixes large amounts of phosphate but only when the mineral was very finely ground. This condition does\ not generally exist in the soil. Ellis and Troug (37), Coleman (29), Coleman, Thorup and Jackson (30) and Black (6) concIuded that any fixation of phosphate by clay minerals is due to an alteration of the clay crystal by the phosphate solution which liberates soluble aluminum. Aluminum and free iron precipitates the phOSphate. This type of reaction was described by Coleman t l. (30) as follows: Al-Clay + MHZPOL, + 2 MOH:M3-c1ay I AI(0H)2 HZPOL, + 2H20 ' ~ . variscite type Where M is a metal cation which encourages the hydro- lysis of aluminum. Ravikovitch (67) suggested that P is fixed by clay in a HZPOA-Ca-clay linkage. This theory was later supported by Scarseth (68) and Allison (1). Wild (8h), however, found an increase in P fixation by clay with monovalent as well as with divalent cations. Whatever the mechanism of P fixation by clay is, its relative importance, in regard to other soil components, is not well understood. Thus, Franklin and Reisenauer (#0), while correlating several soil pr0perties with P fixation, found that there was not any relation to the type or content of clay. Kanwar (50), workingwith lateritic podzols in Australia, also concluded that the clay fraction contributed to only 20% of the total P retention. Most of the retention capacity resided in the coarser fractions as related to the presence of reactive sesquioxides. In addition to the above mechanisms, Cole _£,_l.(28) hypothesized that phosphates in calcareous soils may form a monomolecular layer deposit upon CaC03 surfaces. (b) Chemical precipitation. It is possible that the three ionic forms of P into which orthophosphoric acid may dissociate, HZPOA-I HPOA'Z, and POh'B, are involved in the chemical reactions of P in the soil. Considerable evidence is available indicating that phosphate may be chemically fixed by precipitation with soil bases. In acid soils aluminum and iron play important roles in removing phosphate from the soil solution (AI, #5, 47, 63). Aluminum and iron phOSphate may be present as members of the isomorphous series variscite (aluminum phosphate), redondite (ferrian variscite), barrandite (aluminum strengite) or strengite (iron phosphate) (AI). Cole and Jackson (26, 27) showed that iron and aluminum phosphate freshly precipitated from solutions, produce minute crystals, which correspond to members of the varis- cite-barrandite-strengite isomorphous series. In alkaline soils the activity of calcium in solution seems to control the formation of insoluble P compounds (hl, SI, 63). However, the iron and aluminum compounds may be responsible for some fixation in soils with high pH's (l8, 45). The exact nature of the calcium phoSphate compounds in alkaline soils, has not been established. According to Hemwall (#5), who cited Eisenberg g£_§l. there are between CaHPOu and CaO a series of solid compounds having an apatite (CaF) Ca“ (POh)3 structure. The structure of apatite is remarkably stable, permitting a number of unusual types of substitutions which involve several ions. The identification of the phosphate compounds in soil is difficult. X-ray and Optical methods cannot be used without first concentrating the phosphate fraction. The process of concentration generally brings about changes in phosphate forms. In recent years this problem has been approached by use of solubility criteria. This relates the solubility of the soil phosphates to the solubility of known pure phosphate compounds (25, Al, 53, 60). Thus, the existence of a crystalline solid phase in soils will result, at equilibrium, in a fixed relation between its component ions in solution, and therefore, it should be possible to prove the existence of specific phosphate compounds in soils by the use of the solubility criteria. Lindsay and Moreno (60) deveIOped a solubility diagram for variscite (AlPOu.2H20), strengite (FePOh.2H20), fluoroapatite (Calo (POA)6F2), hydroxyapatite (Ca]0(POh)6(0H)2), octocalcium phosphate (CauH(P0)3.3H20) and dicalcium phosphate dihydrate (CaHP04.2H20). The activity isotherms for these compounds were represented on a single solubility diagram in which a function of the phosphate activity in solution was plotted against pH, assuming that AI3+ activity was limited by the solubility of gibbsite, Fe3+ activity by that of goethite, F'activity 2+ activity was that of a by that of fluorite, and Ca 0.005M solution. This diagram emphasizes the effect of pH on the relative stabilities of phosphate compounds that may be present in the soil. Clark and Peech (25) represented the solubilities of calcium phosphate on a single solubility diagram in which the functions of the chemical potentials of Ca(OH)2 and Ca(H2P04)2 are used as coordinates. They found that in thapresence of solid phase CaC03, the chemical potential of lime (the value of pH - l/2 Ca) depends upon the partial pressure of 002 in the atmosphere. This approach permitted them to conclude that monocalcium phOSphate cannot persist in soils and dicalcium phosphate can exist only in soils in which the phosphate concentration in solution is unusually high;hydroxyapatite is probably the predominant solid phase in neutral and alkaline soils. The use of methods based on the selective extraction of chemical solutions permit a more or less complete characterization in the soil of the soil phosphate compounds. Chang and Jackson (l7) developed a method based on this principle. They subdivideflthe different forms of soil phosphate into aluminum phosphate, iron phosphate, calcium phosphate, reductant soluble iron phosphate, occluded aluminum phosphate, and occluded aluminum-iron phOSphate, with successive extraction with neutral 0.5 N NHAF, 0.l N NaOH, 0.5 N NZSOAI NaZSOA, NaZSZOA- citrate, neutral 0.5 N NHAF, and O.l N NaOH, respectively. This method was used to follow the weathering process of phosphates in soils (IS, 20, 46). Thus, Chang and Jackson (l8) concluded that in the initial stage of weathering, calcium phosphateand aluminum phOSphates were more likely to be formed because of the higher activities of the ions of calcium and aluminum than iron. Since some calcium phOSphates are more soluble than other forms, this phosphate changes rapidly to other less soluble forms. As time elapses, the calcium and aluminum phosphate gradually changes into iron phosphate which is the least soluble. This was supported by Chang and Chu (20) whot' found that superphosphate applied to the soil over 3l years, was present primarily as iron phosphate. However, three days after P was added to the soil in an incubation study aluminum phosphate was the predominant form. Since the publication of the paper on fractionation of soil phosphorus by Chang and Jackson (l7), criticisms and modifications have beensuggested by Fife (38, 39). Glen _£._l- (#2), and Khin and Leeper (52) as to the delineation of the four first main forms of inorganic phosphate. Thus, dicalcium phosphate and some iron phosphate may be dissolved in the neutral 0.5 fl NHhF, which increases the amount of extracted aluminum ph05phate. The use of an alkaline (pH 8.0 to 8.5) NHAF solution prevents this dissolution. However, Chang and Liaw (22) found that the pH of the NHAF solution in the range 7 to 8.5 does not affect the amount of P extracted. 0n the other hand, IO Chang (2]) accepted the possibility of some dissolution of dicalcium phosphate in the neutral NH4F when the phosphate ion in solution is unusually high. He added that NHAF with a pH of above 8.0 may hydrolize more iron phosphate than the neutral solution in some soils (2I). The possibility exists that the extraction of calcium phosphate with 0.5 N H2504 following the extraction with NHhF and NaOH, produces some error because sulfuric acid may dissolve some of the occluded phosphate. Khin.and Leeper (52) demonstrated that the quantity of occluded phosphate dissolved may be great, if the soll contains a large amount of it. Glenn _£,§l. (#2), therefore, suggested the possibility of extracting the reductant soluble and the occluded iron and aluminum phosphate after the removal of the iron phosphate with O.l N NaOH and before extracting the calcium phosphate with 0.5N H2804. 2. Behavior of phosphorus fertilizer in soils When a water-soluble salt such as monocalcium, monoammonium or diammonium phosphate is placed in moist .soil, water moves from the surrounding soil into the granule to form an almost saturated solution of phosphate. The water movement is due to the osmotic potential gradient between the saturated solution and the soil water (56). II The concentrated solution is drawn off by the surrounding soil so that the solution touches and reacts with the soil minerals (#7). The process of inward movement of moisture and outward movement of solution continues until the concentration of the solution is decreased by dilution or precipitation of the phosphate to the level at which no osmotic gradient remains (56, S7). The reactions that occur in the shell of soil surrounding the granule depend upon the compositions of both, the soil and the solution. The dissolution of monocalcium phosphate has been studied by Lindsay and Stephenson (57), Brown and Lehr (l6) and Lehr §£_§L (56), who showed that the initial saturated solution approximates that of a metastable triple-point solution in the system CaO-PZOS-HZO in equilibrium with Ca(H2PDh)2.H20 and Ca HP04.2H20, and in which phosphorus is 3.98 M, Ca I.## M and pH I.#8. Anhydrous dicalcium phosphate slowly replaces the metastable calcium dihydrate and the solution composition approaches that of the stable triple point (8, 57) in which Ca (HZPOQ)2.H20 and Ca HP04 are in equilibrium. The stable triple point solution is #.50 M_with respect to P, l.3# M_with respect to calcium and has pH of l.0l (58). Brown and Lehr (l6) emphasized that the composition of the solution leaving the granule of monocalcium phosphate is independent of the composition 12 of the surrounding soil. The validity of the approach was shown by their success in calculating, from phase rule data, the quantity of dicalcium phosphate that should remain as residual material at the site of therriginal pellet. According to this calculation, the fraction f of the original phosphate remaining is given by the equation: I-R f = (2-R) Where R is the mole ratio of .CaO/PZO5 in the composite solution removed from the granule. Bouldin gt a1. (8), however, showed that when superphosphate is mixed with other salts, the phase rule relationships of the three component system Ca-PZOS-HZO are inadequate. Mixtures of monocalcium phosphate and potassium chloride, for example, constitute a five component system in which six salts - Ca(H2POh).H20, CaHPOh, CaHPOh.2H20, KCl, KHZPOI4 and CaZKH7(PO#)#‘2H20 are the possible solid phases. In such systems, the composition of the released solution is controlled bythe rate at which fertilizer is dissolved and the new phases precipitate. It is expected that the highly acid ph05phate solution that moves out from the dissolving granule Iattacks clays and oxides of iron and alumffium and dissolves calcium compounds with the formation of various phosphates of iron, l3 aluminum and calcium. These will be precipitated differently depending upon the compositions of the soil and the fertilizer (##, #7, 57, 58, 59). The stable triple point solution in contact with the soil dissolves manganese more readily than aluminum, and aluminum more readily than iron (58). With time, the precipitation of these cations is in the order Fe>'Al:>Mn. Potassium tarankite, H6K3Al5 (Poh)8.l8 H20 has been identified as a product of the reaction of the metastaBle triple point solution with the soil (59). With ammonium phosphates, the ammonium analogue, ”6(NHA)3 AI5(POA)8 . l8H20, was identified by Lindsay, according to Taylor _£‘_l. (7#). In other laboratory studies of the reactions of soils with concentrated phosphate solutions more than 30 compounds were identified as possible reaction products (6l). Chang and Chu (20) in fractionation of soil P studies, found that the added soluble phosphate in six soils, with pH varying between 5.3 to 7.5, after three days fixed P mainly as aluminum phosphate, and followed by iron phosphate and calcium phosphate. After IOD days, the amount of iron phosphate increased while that of the aluminum and calcium phosphate decreased. They suggested that the initial stage of the reaction of soluble phosphate with various cations is governed by the surface area of l# the solid phases with which the phOSphates come into contact. 'As time elapses, the stability of the newly formed phases will be in accordance with the principle of the solubility product. 8. AVailability of Primary Phosphorus Sources Terman §§,§L.(75) sthdied the relative availabilities of several phosphorus containing compounds. He grew rye- grass and sudan grass in the greenhouse on three acid and two alkaline soils. Monocalcium phosphate was slightly Inore available than dicalcium phosphate on alkaline soils but less available on acid soils. Dicalcium phosphate dihydrate was considerably more available than dicalcium phosphate anhydrous on all soils. Hydroxyapatite had very low availability on all soils. Octocalcium phosphate produced yields intermediate between those of anhydrous dicalcium phosphate and hydroxyapatite. After this greenhouse study had been completed, Lehr and Brown (55) made a petrographic study of the reaction products of the primary P sources and the alterations that had been ‘made on the material that had been placed in the soil. They concluded that the availability of P from different sources may be increased by: (a) simple dissolution of crystalline products with the resultant build-up in P l5 concentration in the soil solution; (b) an increase in the solubility of some of the products as a result of biological activity in the zone where the calcium phosphates are found; (c) the root hairs interact with crystalline materials of low solubility, and, therefore, increase the solubility of the materials at the inter- face between the crystals and the living root; and (d) hydrolysis in which a product with a low solubility reacts with water to yield an even more insoluble product and a compound with higher solubility than the original material. Thus, under certain conditions dicalcium phosphate dihydrate may be converted to octacalcium phosphate by hydrolsis, as follows: 4 EaHP04.2H2(fl—9CaI+H(P04)3. 3H20 I H POL, I 5H20 3 The phosphorus released in this reaction may be readily available to the plant. Bouldin and Sample (7) reported the relative availability of monocalcium phOSphate, monoammonium phosphate, and diammonium phOSphate when applied to an acid and a calcareous soil. They found good correlations between the uptake of P by plants in the greenhouse and the integrated solubility of the reaction products of soil P. The order of availability of the various phosphates l6 on the acid soil was diammonium phOSphateJ’monoammonium phosphate:>monocalcium phosphate. 0n the calcareous soil, the availability was reversed: monocalcium phosphatej>monoammonium phosphatej>diammonium phosphate. The presence of the ammonium ion in the acid soil seemed to increase the availability of the reaction products. Taylor 2£.§l- (7#), comparing the effectiveness of various iron and aluminum complexes believed to be formed by the action of acid fertilizer solutions upon roots, found that these materials are relatively good sources of P and cannot be regarded as responsible for the "fixation” of phOSphate from water soluble fertilizers. Chang and Chu (l9) found that iron phosphate in a hydrated form as it occurs in lowland rice soils under flooding conditions, seems to be the most important source of available P to rice in Taiwan. Laverty and MacLean (5#) found that the iron and aluminum phosphate fractions are used by the oat seedlings. C. Prediction of Phosphorus Availability to Plants The Bray P-I method (II) in which 0.3 N NHhF plus 0.025 N HCl is used as extractant is now widely utilized (3, 3#, no, 65, 7I, 76, 80). In Michigan, this method provides an acceptable basis for P fertilizer recommendations (5, 31, #3, 70, 73). l7 Quantitative studies of plant growth as related to the applied P have resulted in several proposals of mathematical expressions. To this category belongs the Mitscherlich equation (83), which is based on the assumption that the yields brought about by successive increments of a plant nutrient followed a diminishing returns curve. This equation was stated as follows: Si : (A-Y).C x in which dx represents the increment of fertilizer that produces a given increment of yield dy; A is the maximum yield that could be produced by indefinitely increasing the amount of fertilizer. When integrated, assuming y :‘0 when x : 0 becomes: log (a-y) : log A -~CX One of the important points in this equation is the constancy of the factor C. Mitscherlich and his colleagues (83) considered that C for each nutrient is constant regardless of the kind of plant, type of soil, or other factors. HoweVer, other investigations (l#) indicated that C should vary with: (a) kind of plant; (b) form of nutrient; (c) fertility pattern; and, (d) planting rate and pattern. l8 Bray (l2) modified the original Mitscherlich equation to: log (a-y) : log A - Clb - CX In this equation, A = maximum yield; y : yield obtained when X units of a nutrient are added to the soil; b = original nutrient content present in the soil expressed in units of the added nutrient X; C and C] are the proportionality constants (efficiency factors) of the ’original nutrient content in the soil and of the added nutrient, respectively. Working with this equation, Bray (l2) adOpted the percentage yield concept originally described by Baule. Empirical equations based in statistics methods have been used to explain the relationship between different soil testing procedures and plant growth under particular conditions. Plant analyses also have been used as a diagnostic tool to indicate the availability of mineral elements. It is based on the assumption that there is a relationship between the amount of a nutrient present in the soil and its uptake by plants. The choice of the kind of tissue and time of sampling for a particular plant are some of the factors affecting the interpretation of plant analysis I9 as a guide of the nutritional status. Tyner and Weed (78) and Tyner (79) selected the corn leaf for analysis because it represents a seat of very active synthesis. The sixth leaf from the base of the plant was selected for sampling, because this position is easily recognized, and because it is the leaf immediately below the leaf in whose axil the uppermost ear is born. Generally, leaf samples have been collected at silking time (#, 32, 79, BI), although Ellis 2£.él- (36) and Webb and Pesek (82) observed that the effect of P applications was reflected primarily in the early growth stages. III. EXPERIMENTAL PROCEDURE A. Field experiments Three field experiments were established at the following locations: I. Ingham County - in the S.W. corner of section l5, Wheatfield township. 2. Shiawassee county - in the N.W. corner of section 26, Rush township. 3. St. Clair county - in the N.W. corner of section 6, Columbus township. The soils in experiments l.and‘2 are Conover loams which are imperfectly drained Gray-Brown Podzolic soils devel0ped on loam or silt loam calcareous till. A description of the most important horizons occurring in these two profiles is as follows: -Conover loam (Ingham County) Depth Horizon (inches) Description Ap 0-6 Loam; dark grayish brown (l0YR#/2); weak, granular structure; very friable when moist and slightly hard when dry; pH 6.2 A2 6-l# Loam; pale brown (l0YR6/3) mottled with abundant yellowish brown (l0YR5/6) and light brownish gray (l0YR6/2); moderate, medium granular to weak, thin, platty structure; friable when moist and slightly hard when dry; pH 5.9 20 2| Depth Horizon (inches) Description Bt l#-2# Sandy cla loam; yellowish brown (l0YR 5/6 mottled with grayish brown (l0YR 5/2); moderate, medium to coarse, subangular blocky structure; firm when moist and hard when dry; pH 7.0 Conover loam (Shiawassee County) Ap 0-8 Loam; darkgrayish brown (l0YR#/2); fine, granular structure; friable -when moist and slightly hard when dry; pH 6.3 B] 8-l# Clay loam; brown (l0YR#/2), mottled with l0% of grayish brown (2.5Y5/2) with 25% yellowish brown (l0YR5/6); subangular blocky structure; firm when moist and hard when dry; pH 6.8 82 l#-2# Clay loam; brown (l0YRS/3) mottled with light brownish ray (lOYR6/2 and yellowish brown (l0YR5/6); sub- angular blocky structure; firm when moist and hard when dry; pH 7.2 Experiment 3 was located on a Pewamo silty clay loam. This soil is a poorly drained Humic Gley in the Gray Brown Podzolic Region, devel0ped in calcareoussilty clay loam or clay loam till. The description of this profile at the site of the experiment is: Depth Horizon (inches) Description Ap 0-7 'lty clay loam; very dark brown SI (l0YR2/2); granular structure; firm when moist, slightly plastic when wet, and hard when dry, pH 6.l 22 Depth Horizon (inches) Description A2 7-IO Clay loam. Very dark gray (l0YR3/l); granular structure; firm when moist; pH 6.] 8219 l0-l7 Silty clay loam; dark gray (l0YR#/I) mottled WIth olive brown (2.5Y#/#) and dark brown (l0YR#/3), coating of gray (l0YR5/l) and grayish brown (l0YR5/2) on faces of peds; fine angular blocky structure; very firm when moist; pH 6.l 8229 l7-#0 Silty clay; mottled with grayish brown (l0YR5/2-2.#Y5/2), olive brown (2.5Y#/#), dark gray (l0YR#/l), and yellowish brown (l0YR5/6-5/8), mottles are many; medium, angular blocky structure; firm to very firm when moist; pH 6.5 C] #0+ Silty clay loam; yellowish brown (l0YR5/6-5/8), mottled with grayish brown (2.5Y5/2)-(l0YR5/2) and dark grayish brown (2.5Y#/2), mottles are~common; weak, coarse, angular blocky to massive structure; very firm when moist; pH 7.l Certain chemical and physical characteristics of the surface horizon of each soil are shown in Table l. The field plots in the three experiments were arranged in-a randomized block design with four replications of each treatment. There were # rows, #2 inches apart and 50 feet long for each plot. Six levels of P, namely 0, IO, 20, #0, 80 and I60 pounds per acre of elemental P were studied. co_umo_m_umcumcouc_ .l Eoocmc .m_mo;ucocma mou____ _ “mu_c__omx x mou__:o_ELo> > “ou_co_;u u \N >u_omamo omcmcoxo co_umo \fl xom _ x0m x No. AzI>IoV m.~: o.~: N.m_ _.o m.m_ m.m~ m.m L_m_u .um wmww >u__m OEmzom x0: _ xom x &o~ > m.o~ o._m ~.~: m.o ..m m.~_ m.~ oommmmezm Emo— Lm>0cou Xem . xou x No. > Emo— xom _A_-uv m.m_ o.k~ N.mm ~.o m.m m.o_ o.~ sagas. tm>0coo . EE EENOO. EENo. mlco_uomcm Noo. Iwo. I.~ s 0 web co Axe Axe Axv sag .Lmoo_\me x comu_manoo mums >m_o .__m vcmm Ia A_Im >mcmv Louumz >ucaoQ oa>p I_xoLaam ocm oc_x m_m>_mc< m .l.u.m.u o_cmmco . __0m IpmpwcmcomZN \_ I voumuo_ mum: mucoE_Loaxu p_o_m.oco;3, m__0m mocgu any mo mu_um_couomcm;o .mo_m>:a ocm .mu_Eo:o meow, ._ o_nmh 23 2# Corn was the indicator crop. The other major nutrients were made adequate by adding l50 pounds per acre of nitrogen and 166 pounds per acre of potassium. P, N, and K were applied as diammonium phosphate, ammonium nitrate and potassium chloride, respectively. One-half of the P and all the N and K were broadcasted on the surface of the soil before plowing. The remaining P was applied after plowing and disked in. Soil samples consisting of 20 borings per plot to a depth of six inches, were taken prior to fertilizer application and after harvesting the corn. A second composite soil sample composed of 50 sub- samples from each experimental area, was taken before fertilizer was applied. This soil was used in the green- house and in the incubation studies. The experiments in Ingham, Shiawassee, and St. Clair counties were planted, respectively, on May ll, May 25 and May 2l of l96#. The crop in St. Clair was lost beqkabe of a poor stand caused mainly by lack of rainfall for some time after planting. Leaf samples were collected at different stages of growth. Each sample consisted of IO leaves selected randomly from l0 plants in one of the central rows of each plot. The fifth leaf from the base of the plant was selected. In Ingham county, samples were taken on June 25, 25 July 20 and August l8; in Shiawassee county on July 9, August #, and September I. The last sampling was made when the ears were completely developed. In Ingham County, grain harvests were made on October l3. Two areas per plot, each being l/l000 acre in size were harvested. Areas were selected that contained the equivaTent of exactly l5,000 plants per acre. In Shiawassee-county, one central row per plot with a population of l2,000 plants per acre, was harvested on October l3. The corn grain yields were calculated at a l5.5% moisture content. 8. Greenhouse Trials For greenhouse work, the soil samples from the three experimental areas were air-dried and crushed to pass a 2-mm sieve. Six levels of P equivalent to 0, 20, #0, 80, I60 and 320 pounds per two-million pounds of soil were applied as diammonium phosphate. Uniform applications of N and K were added to all treatments. Ammonium nitrate was used to bring the total N up to a rate equivalent to 300 pounds per acre of N. Potassium chloride was used to supply K at the equivalent rate of 332 pounds per acre of K. 26 The treatments in the three experiments were set up as randomized block experiments with four replications of each treatment. Alf fertilizer materials were thoroughly mixed with the soil. Corn was grown in the greenhouse in 3-gallon glazed jars, each filled with l3.0 kgr. of soil. Eight seeds were planted. After seedling emergence, the population was reduced to four plants per jar. Distilled water was used as needed. Plants were grown for 35 days. Then, the plants were cut off at the level of the soil, dried in a draft oven at 60°C and weighed. Samples for chemical analysis were ground in a Wiley Mill. After the cr0p harvest, soil samples consisting of 5 borings in each jar were taken. C. Laboratory Studies I. Incubation Study Nine P levels, 0, IO, 20, #0, 80, I60, 320, 6#0 and l280 ppm P were established with the use of KHZPOh. One hundred grams of soil for each treatment were placed in 200 ml beakers. Phosphorus was added from a standard solutiOn and well mixed with the air-dried soil. Each soil sample was adjusted to the moisture equivalent level . and tightly covered with plastic Saran wrap, which permitted the air to diffuse in and out but limited the passage of 27 water vapor. The samples were incubated at 25°C. There were three replications for each treatment. At the end of one, two, eight and twenty-two weeks, soil samples were taken and air-dried for chemical analysis. 2. Soil Analysis Available phosphorus was determined by the Bray P-l method (ll),_ using an extracting solution of 0.03 N NHAF plus 0.025 N HCI. The soil to extracting solution ratio was I to 8. Water soluble phosphates, aluminum phosphates, iron phosphates and calcium phosphates were determined by the methods of Chang and Jackson (J7). Organic matter was determined by the method of Walkley-Black (#9). The cation exchange capacity was evaluated with Na0Ac at a pH of 8.2, as described by Jackson (#9). The Bouyoucos (9) method was used for 'mechanical analysis. The identification and approximate rcomposition of the clay fractions were made-with X-rays, DTA, surface area, cation exchange capacity, and infrared .absorption spectra, with procedures described by Jackson (#8). The pH was determined in a l:l soil-water ratio with a potentiometer. 28 3. Plant Analysis Corn plants grown in the greenhouse and leaves from plants grown in the field were analyzed for P. Ground tissue was dried at 65°C. To 0.2 grams of plant tissue was added 5 ml. of alcoholic Mg(N03)2. The sample vwas dry ashed in a muffle furnace at 500°C. The ash was dissolved in 2N HCl and filtered into a 200 ml. flask. The pH was adjusted to 3.0 with the indicator 2-# dinitrophenol. A 50 ml. aliquot was taken and phosphorus was determined by the Deniges colorimetric method as modified by Truog and Meyer (77). Percent of transmission at 660)\was determined in a Bauch and Lomb spectro- photometer. IV. RESULTS AND DISCUSSION A. Field experiments Corn was planted at three locations. The weather at each location during the entire growing season was characterized by significantly less than normal pre- cipitation. This is an important consideration in inter- preting the data obtained from each field experiment. Corn yields were not obtained from the Pewamo silty clay loam because an extended dry period immediately after planting on this fine textured soil resulted in a very poor and uneven stand. The grain yields produced in Shiawassee and Ingham counties are shown in Figure I. As can be seen, the use of fertilizer increased the yields in each experiment, but more so in Shiawassee county. Maximum grain yields were produced with approximately #0 pounds of P in Ingham county and with 80 pounds in Shiawassee county, despite the fact that the soils in each experiment had similar soil test results before applying fertilizer and before planting the crOp (Table l). The residual available P levels, as measured by soil tests, after the crops were harvested are shown in Table 2. In Ingham County, soil test levels for P increased only #2.8 pounds per acre where I60 pounds had been used. 29 30a Figure I. Effect of Applied P to Two Locations on the Yield of Corn Grain (Field Experiments) 30 a no wmo< mum mQZDOm ow. 0?. ON. 00. om -I d. dl IW u bnm.0n. $2.23 a {3: + ...-=23 n: 3.6 Lad?“ o C 75:35 V Eco. Lo>ocou ¢mmd u. ~.u.__2o-& 3390-33.25 In: 36.0 3.1.9.“ 333:6va Eco. ..osocoo ow 4 oc IqI ON 0 0.0m 383V 83d S'IBHSOB-NUOD :IO 013M I 0.00— %9.0: Figure l. goon >Au> mm: cowumasmom ch00 age \M mcowvmowamoh : mo owwno>< ‘M 0.H~+ 1.9: m.hm :.mm+ 0.Hm 3.0a 0.N:+ 1.50 0.:H 00H 00H 00H 0.0H+ 0.1: «.mm 0.0H+ 0.00 0.0a 0.0H+ 0.00 «.5H 00H 00 00: «.0 + 010m 0.20 0.0 + :.0H 0.0H m.mH+ 0.~m 3.0a 00H 0: 0m m.o + ~.Hm 1.9m o.H + quad o.mH o.HH+ o.mm p.5H mad ow cad 0.m I 1.00 N.mm 0.0 I :.HH 0.0H 0.0 + . «.50 Noma 00H 0H 00H 0.0 I 0.00 m.flm 0.: I 0.0H 0.:H 1.0 + m.mu 3.0H 00H 0 om omcmno soapfiunm coflpwppm mmcmno.:owuflopm cowuflvnm, mmcmao :owuwpum acaufipn< x m z 0 909mm m whommm m hopmm m whomom m Lo+%m m whowumawnowxmoc:omy ”Lamrc-+.wmwwqaww u aaaam mmmp mMH mzww boa . \M . oammmm noommmzmwnmv smog po>ocoo Asmnmch smog am>ocoo mucmEpmmLH A090m\mocsomv m \H .Awp:05wnomxm uaowmv .mono choc oco mo cowuosuona ozp can noufiawpnom m mo cowpflvvm any an popuuwmm mm maaom ooaau mo AHIm smamv acoucoo m ca momcmno I N mam< IIII Abommmammgmv ado; po>ocoo IIIIIIIIIIIII.AawnmcHV smog ao>ocoo IIII \H -mao.o one. omo.o awe o oao.o baa.o .Ho. .p.m.q oao.o :ao. .mao.o ado.o whoo.s mao.o mod .Aum.a am.o. mm.o :m.o a~.o. a~.c mm.o mad ,oom. and muwo om.o am.o:l- m~.o a~.o Hm.o mmmi om oma mm.oe -om.o- om.o :~.o m~.o m~.o mmml o: oma mung..,-,.am.a a~.o :~.o m~.o m~.o maa Iau and Mano a~.o a~.o m~.o mano.. m~.o mad 0H .oma onto m~.o- H~.o o~.o o~.o -.o mud o oma ----oaaiiiaih-~k--- ---a,-hu --.--- InIIIIOOH--- IrioaII- h--:a----- In, a z I . , . whom\m0::om IIIIIIIIJIIIIIIIIII Amawucmao noumm mmucv mcwaoamm mo mafia IIIIIIIIIIII :- .;,i. . , : mpcoaumoha “mucoewhooxo oaowmv mawomiozv,ou omflamam m mo mupmp an oopoommm mm nuzonm mo , mommum moan» um mo>mmH cpoo mo pampcoo m «:9 I m mgm<9 33 3# The P content of the leaves waSe closely related to both the grain yield and the P applied (Table #). The highest correlation coefficient involving yields from the Ingham county experiment were obtained with the July sampling. The same in Shiawassee county was obtained for the September date. The relationship between the P content of leaves and applied P was highest at the first sampling date at each experiment. This observation agrees with those of Ellis ._£‘§l. (36) and Moham (6#). 8. Greenhouse Experiments Before any fertilizer was applied in the field experiments, soil was collected from all three sites for use in the greenhouse. Similar rates of P, as were used in the field, were added to each soil. Corn was then planted. The data in Table 5 show the yield (dry weight), the percent of P content in the corn plants, and the P uptake by the :plants. In the greenhouse, the use of each increment of fertilizer P, increased the dry weight of the corn plants grown on all three soils. The use of even the highest rate of P, equivalent to 320 pounds per acre, was not sufficient to produce maximum yields. This was not the case in the field. Ha>oa Ho.o um pamowmwcmam «a ..ama.o ¢WH~a.o «.nwa.o. «humane .aama.a ..kma.o- pmaamau a ....mmm.o «coma.o. :«mom.o «unease «aasm.a ahaaa.o pamaw campu chop. mcwpcmao :wdmucmaa..w:wpcmHo maeuamao ucwucwaoi mcwpcwam pound, . Adamo novmm aopmm noumm 909mm ammo ooa...msmp «a mama a: . mama ooa mama ck mama :: a noasouamm : «manna a sane ma pmsms<. om says :u once ommmmzmwnm-Iramoa no>oaoo . F EmnmaH I ado; po>ocoo mo>moq :« m .Amucoawnomxuivaowmv unOm can no mupwn nonwawnnom one moaowz owmnm op omyuflon mm nuaonm.mo-mommum_oon£u pm.mo>moa.cpoo mo 9:09:00 m:onu mo mucowowmmuoo cowumaonmoo.pmuaflq I : mqm<9 35 36 macapmowaaon : Hm mpasmon ommno>< \m 00.0 HH0.0 0H.H 0:.H 0H0.0 05.0 .an0 0H0.0 .50.H I H060 .a.m.# :H.H 000.0 00.0 00.H HH0.0 0040 m0.H 0H0.0 0~nmll 00.0 .n.m.q 0.:: 00.0 H.5H 0.:5 00.0 :.00 0.00 00.0 0.0a 000.0 .IM0 000 000 0.00 0H.0 0.0a 5.:0 Hm.0 0.5H 0.00 00.0 0.0a 0H0.H 000 00H 000 :.0H 0H.0 5.0 5.HH 00.0 5.5 H.:H 00.0 5.0 000 000 00 000 0.0a 5H.0 0.0 0.0 :H.0 0.: :.0H 0H.0 0.0 H:00 N00 0: 000 5.0 5H.0 0.0 0.0 00.0 0.: :.0 0H.0 0.0 500 000 00 000 :.5 I 0H.0 0.: 0.0 NH.0 0.0 0.0 :H.0 0.0 0 000 0 000 “wage a a Amamhwv amass a a Ame..wc rammmw a » Amemhwc x a z oxmwo: vnwfloz oxmums “£0003 uxmums vnwwmz pom\wa m ham 0 >99 0 zhn vmoom "whom\mocsomv m mucmaumoph Ahamao .pmo smog xoummmzmanmo xemgmch hmHo zuawm oemzom Smog ho>ocoo Emoq po>ocoo \H .Ampcoawnomxu mason Icoonwv nonwawunum 0 mo muumn op noumaon mm madam 6093p no czohm choc mo uxmum: m can .m “coupon .vnwwmz 090 0:9 I 0 mqmoa H0.0 “a unmowmwcmwm «a «*000.0 c.050.0¢¢000.0 Smumm.0.emcmmmu0.ccmmm.0 «‘000.0 ¢c~5m.0 4&050.0I. .wxmwna m «c000.p. maH00wb. II «0000.0 «40:0.0 II ae000.0 «amam.0 «I .m 0 I.mmm.o II II ..mam.o II . II, I.H:m.° II II #00003 006 -W.wa¢oaoakmmoo.cowvaonnoo oanfiwm omaamnm .ps sun 6 a omaanmm .pz spa 0,0 ouaaagm .u: app 0 a , m m . m _ . moanmwam> Aaamdo.umo -.- II Iamoq zmao muawmIInoomwmzmwnmv smog ao>ocooIIIIIAamcwch ado; po>ocooIII - oamzom , .: .Amucoafihooxo.ousoncuonwVImH«Om mouse on nowaamm m 0cm..mucmHn choc mo #30063 age .0 pcoonmo .oxuums.m mo mucuwowmmooo coaumaonnoo maaawm I 0 mqm<9 39 #0 are statistically significant at the one percent level. The correlation coefficients obtained from the Shiawassee county soil tend to be slightly higher than those from the other two soils. Soil tests were made before fertilizer was added to the soil in the greenhouse and immediately after the cr0ps were harvested. The results are shown in Table 7. The level of P, before fertilizer was used, in the two Conover soils was. similar. Each value is considered to represent a "low” level of P availability. The level in the Pewamo silty clay loam is considered as ''medium”. After the crops were harvested, soil test levels increased with the use of fertilizer P. In Ingham county soil, up to one-half of the fertilizer P remained in an available form; whereas, only up to one-fourth was available in the Shiawassee and St. Clair county soils. The P removed by the crops and P left in the soil as measured by standard soil test procedure do not account for all of the applied fertilizer P. Therefore, some must have been converted into non-extractable forms or ”unavailable forms" not measured by the Bray P-l test. This effect was more evident in the soils from Shiawassee and St. Clair counties. The effect was greatest at low levels of applied P. mcowvmowaooh : mo uwmno>< \w m.om+ mome m.Hm :»mm+.0.mww0H NowH mommH+ moflbfl 0.5H mmm omm oo o.o:+ mean ,mon m.m:+ 0.00 «.0H I0.hh + Nomm mohH mmm QmH com o.mH+ 0.0: meHm m.:H+ oodm NowH 0.5m + :Imm mobfl «mm om com 0.0 + 0.0m mon Now + - :.HN NowH m.mH + :.mm mafia «mm 0: com H.H + momm moan o.H + N.hH NImH o.mH + 0.0m... mbhH . Nmm ON oom mom I o.mN mem NAN 4 . 0.:H Nomd N.H + mama 0.5H «mm 0 com MT wwwmco cowpwoom cowuwvom mflmmso :Oflpflvom coavwnom ummwno GOfiwflonm cowuwvnm x m z m pwumm whommm m cupmm whomom m nmumm whomom mama mm mzum mm 65mm mm . I pnwmao .pvv smog AuommmzmaAmvemo; no>ocou namzwch smog no>ocoo unflp >pamm oamzmm Accom\mocsoav mwcofiumona Ixmhom\mucsoav a \H “mucoaflaooxo omsoncounwu mono caoo moo mo cowvosvopm unp,uam,nonwafiuhow 0 mo :owuwoum.ony an oupoummm-mm maaou conga mo naIm ampmv paoucou 0 ca «mango I a mqm<. \m .MI5MIIbIbeIhIh5. 5.55 .5.Hm :.:m E a... .....R ...: E . . mum 5.. 0.0H o.ma 0.0H c.0H o.a N. a «.0 :.oa 5.0a c.0H «.ma _o.ma . I .....I 5m.5HImwmpmv a emu . . an m 0.. H. «a a N A an. a a . H cocoa IIIIIIIIIIIIIIIIIIIIIIIIIII50x0030 0Sap coapmm:ocH IIIII IIIIIIIIIIIIIIIIIIIII 0 San -- -...w.fi.w..%w--................ .... .... .PmOu. Hwom HIm 50am 0n 0060590900 00 m 0HAMHHM>0 co mawom 009:» ow cowvmowammm 0 mo 00509 new cowvmnnocw no 0809 no voumwm I 0 mqmoa 86.6 um unmo000am0m .I . «0000.0 «1050.0 «0500.0 00 I «Imam.o . ..0omm.a «.5ma.o a I ..amm.o , «.000.6 I., ..5mm.o 0 «0000.0 «0000.0 «0000.0 0 .08000000000 80000009900 000800 Ap0mwo.pmv . A00wwmzm0nmv Aaunmch IImxoosIHI 8000 0000.50000 080300. 8000 p0>ocoo 8000 n0>ocou. 080v :00uunuocH .0080p 00000naoc0 mso0nm> m0umm..AHIm 00000 0 now 0000» 0000 van 000aaqm.m c00300n 008000000000 80000009000 000800 I 0 mam00 00.0 00 00000000000 .0 0o>00 00.0 00 08000008000 0 08000000000 0 00 0m000>< \m [.irlP mom.o wmmwo mmm.o tnm.0. :mm.o th.o OQQPQ.Mmm.o «mm.o 433:0. mam.o hum.o wvcmflowmmooo a0 «0 «a «a a. 04 a0 «a a. «0 cc 0 80000000000 000800 0.000200000 0.000.0.000 0.00: 0.000 0.000 0.000..0.00: 0.00:..0.000 0.000 . . 0000 :omwm conflm m.mHm «.mom. :«mmm N.::m «.mmm 3.:Hm -:.mmu 0.0mm m.o:H m.hh 01m :.de w.mmH moozfl $.mmH- h.HmH mommH «.mmH m.ooN. hom3H H.NmH .Hufiwd condfl - can bomHH oom0H :.NOH m.mm «.mHH m.mHH 9-mm «.mm moam m.Hm m.h: mom: OWH 0.0m 0.:m m.~m m.bm :oum oon m.mm 3.0m :omm 0.0m m.mN m.mN { om moon 0.0m 3.0m m.mm m.mm ::.:m .Nomm 0.:N w.mu m.bN m.mn m.HN o: 0.00 0.00 0.00 0.0m: 0.00 0.00 0.00 0.00. 0.00 0.00 0.00 0.00 00 m.m mom coca O0OH o.m hom :.m N00 .mwm 0.6 Hum. from OH .I 00:0 a00mv 00x00 0 800 \H 0 r 00 0 0 0 00 0 0 ‘0 00 0 mi 0 UGUU< V . 0 8mm unnuanuwnuunuuuunnnnun 0000030 0800 8000008080 Inunuunuunnnnuuuuunnuunuunu1nn 1:11:00000o.0mv 8000 II|A00000300nmv8000 00>0800uuuuu080nw800 8000 00>08ooun| >000 >MH00 D8030n .0000000 0 00 00000 080 8000000080 00 0800 00 0000000 00 00000 000:0 80 00x00 0 00 08:08< u 00 mqm<0 47 The data in Table II were calculated from those presented in Table 10. The values in this table illustrate the percent of fertilizer P that was fixed by each soil. Generally the percent P fixed decreased with each increment of added fertilizer P. This was the situation for each soil and for each incubation period.. The fixation rate was slowest in the Conover loam from Ingham county, ranging between l0 and 84 percent at the end of 7 days and most rapklh1de Pewamo silty clay loam which had a range in percent fixed P between 34 and lOO percent at the end of I week. The values at the end of 22 weeks of incubation were 34 to 89 for the Conover soil and 6l to 98 for the Pewamo. Such differences suggest that different kinds of reactions were involved in the P fixation process for each of the soils. At high rates of applied P, the percent P fixed by all three soils increased with time. The greatest range occurred in the Pewamo-soil, which received 1,280 ppm P. During the twenty-two week incubation period the percent ' P fixed increased from 35 t0 6] percent. The data in Table ll suggest that the two samples of Conover were not as similar perhaps as they should be considering that each soil represented the same soil type. The percent P fixed in the Shiawassee county soil was much greater and, on the average, the fixation of fertilizer P occurred at a more rapid rate than in the Ingham county sample. III: A0000oq0mww0 islri 8000 >0 0m 0803 will; 80000000000 0 00 00000 080 8000009080 00 0000 0n 00000000 00 00000 000:0 00 00000 0 0800000 . 00 000<0 wm111000000300nmv8004 00>08ooun|10800w80v 8004 00>08oolnun .08000000000 00020 00 0m000>< \0 00000;;00000::00000 00000 00000. 00000 00.00 700.00 00.00 00.00 00.00 00.00l 000.0 00.00 .00.00. 00.00 00.00 00.00 00.00 00.00 00.0: 00.00 00.0: 00.00 00.00 000 ...00000 .00000--00000g.00.00 ,00.00 00400 00.00 .00.00 -0000: 00.00 00.00 00.00 000 00.00 00.001.00000 00.00 00.00 00.00 00.00 .00.00 00.00 00.00 000.00 00.0w; 000 00.00 00.00 00.00 00.00 00.00. 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00; 00 00.00 00.00 00.000 00.000 00.00 00.00 00.000 00.00 00.00 00.00 00.00 00.00 00 00.00 .00000 00.000,00.o00 -myd0ll00.00 00.00 200.00 00.00 00.001‘.0.00 00.00 00- \H 00 0 0 0 00 0 0 0 00 0 0 0 4 Hi ll. 00000 uunnuunuauluuutulluununuuuuunJu00x0030 0800 800000308Huuunluuununnnnnuuuuwuunuaa¢n 0 800 “3 49 Because fertilizer P was fixed by these three soils in different amounts and at different rates, the question naturally arises as to why such differences occurred. The results of the P fractionation after 22 weeks of incubation help to explain this situation. The fractions labeled by Chang and Jackson (‘7) as water soluble, aluminum, iron, and calcium phosphates are shown in Table l2. The data in this table show that where no fertilizer P was used, there was approximately four times the calcium phosphate in the Pewamo soil as in either of the two Conover soils. The Pewamo also contained about twice as much iron phosphate ‘as the two Conover soils, as well as significantly more aluminum phosphate. All three soils,where no fertilizer P was used,contained only traces of water soluble phosphate while the greatest quantity of P was in the iron phosphate fraction. In general all of the P fractions were increased by the use of increasing rates of fertilizer P. In all three soils the fraction labeled aluminum phosphate increased the most. The iron ph05phate was also dominant where relatively low rates were used up to and including 80 ppm P. At l60 ppm P, in the Ingham county soil the values for the iron and aluminum P fraction were similar while in the other two soils the iron P continued to dominate. At the 320 ppm 08000000000 00000 00 0m000>< \a 00m m.mH How II Nam how HONH II :.N :om mom ll H000 OQOmI‘H mom 80m m.m II m.N mom m.m ll mod m0m «.1 II mo.o .Q.m.q 8 :NH c.00m manic bnmad momm m.mmm Norma momNN mom: momma m.:mm m.HmN emu—H m HNH 80mmm m0mmm m08m H.8m meNN b.8om m.on mobm Hoomfl hohww mobdd 03o w wHH mound m.wHN HoHN m.Hm .mwmmd «.08H m0m: H.0m H.MNH m.NmH comm omm m moH NomMH 80000 mww moon mowdd :.mb moad m.mu m.HOH m.000 m.0~ omH m0wm moomH w.Hh 00008 N.om Momm :.N: 000.08 mom“ comb mobm N0m om 80mm m.m~H m.om 00008 momm :.N8 moon, .00098 moflm 10mm mohm mam o: Nomw .ooHHH 8.0: 00098 H.ow puma 10mm 0030.08 m.:N 00mm moan 000R8 om mama momon.m0mm- 00008 momd boom oom0 000h8 m00N comm mama mmflh8 OH Noam memOH m.Hm 00008 :cma nowm .oowH mom..08 muHN mam: m.HN 00008 0 I 0 8mm \0 000: ., . 000:0 00008 00080 000:0 000:0 -0onm 000:0 000:0 000:0 -0onm 00080 00000 00000 -0o:0 00000 100nm imonm imonm 00080mm «003m umonm imoam 00na0om imorm umogm Imonm 000800m m 800 -00 -00 -0< o m -00 -00 -0< 000 -00 -00 -00 000 IIIIIAQHMHUO va EMOJ >MHUI. 00000 080300 1000000300£mv80oq 00>0800. .IIA80nm8HV 8000 00>08oo-II .000000000m m 00 00000 an 00000000.00;8000000080 mo 0x003 «N 00000 00000 00000 80 000000000 8800000 080..8on0..83808800..00nu0om 00003 a N0 m0m<8 50 Sl treatment, and at rates above this, the P in the soil was associated with the fraction labeled as aluminum phosphate. The aluminum phOSphate content of the two Conover soils were relatively similar at all fertilizer rates, while the Pewamo soil contained more P in this fraction. The difference between this soil and the two Conover soils became greater at the highest two levels of application of fertilizer P. The data in Table l3 were derived from those presented in Table 12. These data show the percent recovery of fertilizer P, in the form of water soluble, aluminum, iron, and calcium phosphate. These data were calculated by taking the difference between the amounts of a particular soil phosphate fraction in the treated and untreated soil and expressed as a percent of applied P. These data show that most of the P applied as fertilizer was recovered in the total of the four P fractions. In general, the total P recovered in the four fractions accounted for more than 70 percent of the applied P. These values are slightly less than those calculated by Laverty and McLean (5#). Similar results are also presented by Yuan £3 £1. (86). Most of the fertilizer P was recovered in all three soils as a part of the aluminum phosphate fraction. This was followed by the iron phosphate, and then by calcium 000000000900 000:0 mo 0m000>< \M 0:00 00.:0 00.00 00.00 00.0 00.00 00.m:,.00.:0 00.0. 00.00 0:.0: 00.00- 000.0 00.0 00.00 00.00 00.0 00.0 00.00 00.0: 00.00 0:.0 00.00 00.0: 00.000 0:0 00:00 00.0. 00.00 ,00.0 «040 -00.:0 00.0: 00.00 00.0 00.00 00000 00.00. 000 00.00 00.00 .00.m:.-00.m 0000 00.00 00.00 .00.: ::.~ 00.00 00.0: 00.0m-, 000 00.00 .00.00 .00.0: ,00000 .00.00. 00.0:. 00.00 .00000 400.:e 00.00 00.0: 00.0 00 00.00- 00.00 00.0: 00000 00.:0 00.0: 00.00 00000 00.0 00.00 00.0: 00.:mi. omi- 0.m~ 00.0 .00.0: 00000 00.0 00.00 00.00 00000 00.00 00.00 00.0: 00000 on 0.00 00.0 00.:: 00000 00.0 00.:: 00.00 00000 00.0 00.:0 00.00 00000 00 : ~fl_.h00>000mtm w 000:: 000:0 001:: 000:0 000:0 000:0 nmo:m 000:0 000:0 000:Q imo:m 000:9 000:0 000:0 imo:m -0000 -0000 -0000 0000000 -0000 -0000 -0000 00000om.-0000 -0000 -0000 0000000 -00 -00 -0< 000 -00 -00 -0< 000 -00 -00 -0< 000 00000 iuuuula0000000mvi8moq >000I1|n0000030m:mvamoq 00>ocooIIJIA80:w:HV amo: 00>Ccoollii m 800 a0awm 080300 . L coa0musoc0 mo 0x003 «N 00000 00000 000:0 :0 000:mmo:a 8500000 0:0 .co0w .8saw8aam .0H::Hom 00003 00 0000>0000 m 0:0000m - m0 mqm<9 .52 53 phosphate. The water soluble phosphate tended to increase with the use of increasing rates of fertilizer P. These trends are supported by the work of others (l8, 20, 2h, 33, 51+). The question arises as to the relationship between the -percent of fertilizer P fixed in the soil and the amount of P recovered in the aluminum, iron, and calcium phOSphate forms. As described previously, the total P fixed is considered to be that which was not measured by the Bray P-l test. The values for this, expressed on a percentage basis are shown in Table lh. As can be seen, there is no direct or well-defined relationship between the percent fixed P and that recovered with the fractionation procedure. It is possible that the P that was measured as ”fixed” with these procedures was not all fixed. There were, however, direct relationships between-soil P levels as measured by the Bray P-l test and the three fractions of P for each of the soils as can be seen in Table l5. Each of the simple correlation coefficients were very high and statistically significant to the one percent level. This seems logical because the Bray P-l extractant is composed of hydrochloric acid which can dissolve some forms of calcium phosphate and the fluorine ion in acid solution can complex aluminum and iron ions, with consequent release of P held by these trivalent ions to the solution (23). 00.00. 00.00 i 00.:0 00.00 00.:0 00.:0 000.0 00.00, 00.00 00.00 00.00 00.00 :0.0: 0:0 00.00 00.00 00000 00.00 00.00 00.0: 000 00.00 00.00. 00.00 00.00 00.:0 00.00 000 :0.00 00.00 00.00 00.00 00.00 00.00 00 00.00 00.00 00.00 00.:0 00.00 00.00 0: 00.00 00.00 00.00 00.00 00.00 00.00 00 00.00 00.00 00.00 00.00 00.00 00.00 00 a 000:mmo:m w 00m:mmo:m a 000:amo:a -00 0:0.-00.000 0 -00 0:0.00-.0< 0 -00 000 .-00-0< 0 00 U0000>coo m w0xwm n 00 U0000>coo 0 00x00 0 00 ONP00>coo m U0xwm m M0WMM 000000 .0mv Emo: A00mwmzmfi:mv 800: 00>ocoo “Em:m:Hv Emo: 00>ocoQ >0Ho >0Hmm,oemz0m .0000000000 0 mo 00000 0: 00000000 00 000:Qmo:a a=0oamo 0:0 .0000 .E::HE:00 O0 p0000>coo 0 00000 0:0 0:0 00x00 0 mo 0:00000 w :0 mqm¢0 5h 00>00 00.0 000 00 00000000000 .c 000.0 «0000.0 «00m0.0 «0000.0 «0000.0‘ «0000.0 000000 .000 E000 >000 >000m 05030n m:m.o 00mmm.o . 00:00.0 0000000 00mmmoo 00000.0 Ammmwmzmwnmv 3000 00>0000 my :omoo «0mmmoo 00000.0 «00mm.o «0300.0 «0000.0 050£w0H0 E004 00>ocow humor mumm ml0< mlmo \mn0m \nn0< AHom 000000000000 00000000000 0000000 000000000000 00000000000 00mE0m .0000: «N 000 000000000 0003 0000 00000 00000 00 05000 000000000 8000000,000 .0000 .09000000 000 000 00tm.>00mv 00000 0000 0003000 0000000000000 000 3000 000:3 000000000000 00000000000 0000000 000.000600 n 00 mam