a. . a -. o - - o 0- a -4 - o . - u ’.‘~ 0 . ' >. - . h ‘ . O ‘ ‘ I ~>o- 0—. 0 ~ K" o 0 '~ . .. \\ . .\ . . \ .\ 1 u - v ' . M. 00- c 0- . . . . ‘ . - . . . . l‘ \ D ". l . \‘nJ... . .‘ s \ o \f. -\ 0 . ‘ot- -\ '..-“~ . . . u. \2 ?r‘. t. 1 ' Pfi ' ' 3' : .- -. ' ; n: “. -‘. ‘.. ' \'. \ , ~. . ,. ... . o\." s' I o": .\ s, .1 .'\\- -. I5 \ . ..'.\.. .‘J - ~ . . . . .1. C C— O a. ‘ t I - C - -O I . ‘. \ . ' a N .... - .uu ‘\' A o \0 ~ . o u... “\ . . L\: 9.. . -- V O. .- - ~ . /'~ . - *- 2: - ; ':~ ' . ' " ~o _ Q .\ . . ‘ ~ w o: n o o\~o"-\'-—.\ o - \' - .— \.\.‘\\'i u» :9. : I' 0-. ' f — . -"-.--. saw : a." " ~° aura . :. --:. - i . H 13 ~ . ...'§.§‘§J .‘c 0.0% WV¢o\-w ti. -ool ' ' ' "u':'_’:- f. . ' '3‘; ' :-- .4/3. ‘ ' " ’5'; . 0 ~ 0' -- , '- ‘ \. ‘. '. ». 6“...“7- .‘ug'xo ‘ OO‘I — I; \ I. - o ,- ‘~ .‘ - . .0- . — .—..- -. - . ._ - o D. u o ‘ u. ' W o. '\ ‘ i I 5 I V h ‘5 x; )1 \ This is to certify that the thesis entitled The Effect of Granulation and Placement on the Relative Uptake of Phosphorus from Superphosphate by Beans and Wheat. as Measured by Tracer Techniques presented by James A. Vomocil has been accepted towards fulfillment of the requirements for Master of Science degree in Soil Science .6)”. FM Major professor Date Novefmber 21, 1951 ..¥<:‘ - kt“ '1 ‘ .' - j ‘V-. f}. ._‘ ft ‘ _ . I“ I I ‘ ' . s 1,)!» ‘ 5" “’ ' _‘_‘l- .‘ t {_- I- . ‘n‘J, ‘ .. a ‘ "f ‘ . .' . r '. ’ i ‘ I I ‘ - 1' ' ‘; " .1 l " V . 31:16.6» 3 .. .s-fl W ‘9“ ‘..,..;~.t..-;-_‘.u. w. . u . § ‘ .- ,.‘ ~ a . I .fk‘lg A"? f V‘?“- :fi an u? m T , - EthnoT C: SHALJLAiION ANJ PLACELEET ON THE RELATI . f“ 1' q- .1 q- ‘ III. ‘ 1| 0 " UPiAnE OF PHOSPnORUS FR n SUPERPAOSPHATE E BEANS , . Q I O . AJD U37 T AS LEASURED BY TRACSR TECHHIQUES by JAZ-iES A. 1939011. 1|], A THESIS Submitted to the School of Graduate Studies of Hichigan State College of.Agrioulture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Soil Science 1951 THEE! q AC ICE-IO? TLEDC-EICE 311‘ The author wishes to express his sincere appreciation to Dr. K. Lawton for his helpful suggestions and guidance through- out the progress of the work. He is also indebted to Dr. B. Kawin for advice and assist- ance in the radioisotope measurement procedure. Grateful achnovledgement is also due to Dr. R.L. Cook for taking photographs and to Mr. P. Coleman for taking and prepar- ing micrOphotographs. The writer extends gracious acknowledgement to Dr. L.K. Turk for his criticism of the manuscript. He is indebted to the graduate council of Michigan State College for the granting of the assistantship which made the completion of this work possible. 268388 I. II. III. IV. TABLE OF CONTENTS INTRODUCTION REVIEV OF LI"EEATURE A. E Granulation Placement EX?;RIHEHTAL KATERIALS AHD KETHODS A. B. H V. RESUL Soils Superphosphate Procedure TS AND DISCUSSION Soil phosphorus Yield Phosphorus content of plant material Observations of fertilizer - root relationship 50 61 63 INTRODUCTION The consumers of fertilizers are concerned with the nutrient supplying power of a fertilizer material. The nutri- ent supplying power of a fertilizer is determined to a large extent by its composition but is also dependent upon the availability of the nutrient elements to the crOp grown under field conditions. This availability may be considered the "efficiency" of the nutrient source. Hence, in fertility studies the problem of evaluating the relative efficiency of fertilizer materials is an important one. Methods of improv- ing the efficiency of the materials Which supply the nutrients necessary for the production of creps are continually being preposed. The application of new measuring techniques using radio- active tracer elements to the problems of plant nutrition has recently renewed interest in the problem of improving the efficiency of fertilizer materials. The technique of adding material as a radioactive isotope provides a means of separat- ing quantitatively the nutrients utilized into two fractions; the native or soil contained fraction and the fertilizer or added material. On the basis of this separation,‘the amount of nutrient material obtained from the fertilizer can be readily determined. Two methods of regulating the efficiency of a fertilizer are granulation and placement. Theoretically, the efficiency of a material like super- phOSphate should be affected by granulation if it is assumed that only dissolved nutrients are utilized by the plants. Hill (ll) has shown that granulation reduces the dissolution rate of superphosphate in water. The reduction of the disso- lution rate should cause an increase in uptake if the rate of dissolution is made to more nearly correspond with the rate of phosphorus uptake by the plants. The closer correSpondence of the two rates would result in lowered phosphorus concentration in the soil solution at any one time, and this in turn would decrease the fixation of phoSphorus by the soil. Hence, it would appear desirable to decrease the dissolution rate of the soluble materials and, as previously stated, this can be accomplished by granulation. Upon consideration of the effect of placement upon the uptake of phosphorus from a given source, it is noted that by concentrating the material in a band, the material in the center of the band may not come in contact with the soil and thus escapes fixation. Also, the fertilizer may be prevented from dissolution because of the presence of a saturated so- lution surrounding it. These considerations become important only when the movement of an ion in the soil is very slow, as is the case with.ph08phorus. Secondly, there is the possi- bility that with band placement more of the phosphorus is in the root zone when the plants are growing most rapidly. Band placement is especially effective when the plants are young and have small root systems. It has been shown by many investigators that the advantage of localized place— ment diminishes as the plants age, and may disappear by the time the plants reach maturity. Collings (5) points out that it is generally accepted that localized placement may promote a heavier growth of roots within the fertilizer zone, but this does not greatly restrict the extension of the root system in all directions outside the fertilizer zone. On the other hand some workers have argued that band or localized placement may reduce the uptake of water or nutrients not contained in the zone of the placement by the concentration of the roots near the fertili- zer band. A greenhouse experiment was designed to test the effect of granulation and placement by measuring the uptake of phos- phorus by two important Michigan crOps, wheat and field beans. Radiotracer techniques were employed as a more direct approach to the problem with the hope that data of a different sort which would be of some value in the solution of the problem of fertilizer efficiency might be obtained. -4- REVIEW OF LITERATURE A. Granulation Granulation is defined by Hardesty and Clark (10) as the process by which a material is formed into small grains or masses. As applied to fertilizers, the term "granular" is commonly used to describe products which have mean par- ticle sizes greater than are customarily associated with such products and which contain relatively small portions of fine particles. Thus a fertilizer is classed as granu- lar whether the increase in mean size results from cement- ing together of individual particles, or from mechanical separation of smaller particles from the body of the material. The practice of granulation has been used for many years for improving the physical condition of fertilizers. Granulation reduces caking and dusting losses, and makes it easier to obtain a more uniform distribution in the field. Large scale commercial production of granulated superphos- phate has been carried on for about fifteen years in the United States. t was greatly stimulated by the work of Ross (17) and his associates in the United States Department Of Agriculture about thirty years ago. Agronomic experiments testing the effect of granulation on availability of super- Phosphate in the United States started about 1950. Sherman and Hardesty (20) have recently prepared a review of experiments conducted with granulated sources of phosphorus during the years 1951-1950. This work reviews 52 experiments conducted in 15 states, Canada and 7 European countries. Their paper was in turn sm1ma.rizedby'Stamstka(22). The pertinent parts of the latter sumneryarereported here. In the 47 experiments in which granulated material was compared with non-granulated under conditions of localized treatments (band placements), about 50 percent of the exper- iments showed no significant differences due to particle size. About 25 percent of the experiments showed an increase in crop yield due to the use of large granules (10 mesh or over) or briquets. Most of these experiments were on pod— zolic, lateritic, or low P205 soils. About 10 percent of the experiments favored a medium sized granule (10-40 mesh). Most of these were on loamy soils. About 15 percent of the exper- iments, most of which were on sandy loam soils or in pot trials, indicated either a powder or very fine granules (40- 150 mesh) to be a superior source. Not all of the results were statistically analyzed. During the years covered by the summary, 22 experiments were reported which studied the effect of particle size in mixed placement. Of these, 65 percent reported no significant differences in crop yield due to particle size. The remain— ing 35 percent favored a large particle size. All of these were either in pot experiments or podzolic soils. Not all of _ the results were statistically analyzed. - Experimental work done in Maryland in 1949 by Borland, et al (2) is similar to the work reported here. In this eXperiment hairy vetch was grown on Evcsboro sandy loam and Cecil clay loam under greenhouse conditions. According to the conclusions from the Maryland experiment there was no significant difference in yield due to granulation. With band placement, there were no significant differences in the percent of plant phosphorus derived from the fertilizer, but with the mixed placement, the percent of phOSphorus derived from the fertilizer increased with particle size from 69 percent for regular superphosphate to 92 percent for the gran- ular material of 4-6 mesh. Several experiments conducted in the Western states in 1950 using a range of particle sizes of calcium meta-phOSphate tagged with P52 have been reported. Fuller and.HcGeorge (8) reported little difference in the uptake from calcium meta- phOSphate as influenced by particle size. The only signifi- cant difference at the 5 percent level in the percent phos- phorus derived from fertilizer was found in the third cutting of alfalfa and was in favor of —lOO mesh over -lO mesh ma- terial. This experiment was carried out on a calcareous Laveen clay loam soil. Schmehl, Olsen and Gardner (19), working on a Las Animas clay soil in Colorado with sugar beets, showed that the ab- sorption of phOSphorus increased as the particle size decreased. However, differences were not statistically sig- nificant at the 5 percent level for all dates of sampling. B. Placement Around 1920 widespread interest develOped in the possi- bility of increasing the efficiency of fertilizer materials by placement which resulted in the organization of the National Joint Committee on Fertilizer Application. This committee has been responsible for the leadership in experimentation on placement and the compilation of experimental results. The ultimate purpose of these summaries is to enable the investi- gators to make fertilizer application recommendations to the farmers. The suggestions made by this organization are based on experiments conducted by investigators in virtually every section of the United States during the twenty-four;year period between 1924 and 1948. In November 1948, the National Joint Committee on Ferti- lizer Application made the following recommendations for the two crOps involved in this experiment. For field beans, the most effective placement is in a single band approximately one inch to one side and one and one-half inches below seed level. For wheat, it is recommended that the seed and ferti- lizer be drilled simultaneously with the grain drill. This places the fertilizer close to and in partial dontact with the seed. In the trials on wheat, no side-band placement had been investigated. These recommendations are for fertilizers in general and for representative soil and climate conditions. The above recommendations imply that the net result of these experiments was to favor band placement on these crops. In 1934 Cook (6) conducted an experiment comparing band application of mixed fertilizers with drill application ahead of planting. Field beans were grown on several important Michigan soils, including Miami loam. The yield results in- dicated that band application was superior. Millar (l4) and coworkers found that on ankston clay loam fertilizers placed one-half inch from the seed produced a significantly larger increase in yield of cannery peas than did the same fertilizers broadcast or placed in bands two inches from the seed. The fertilizers were 4-16-8 and 0-16—8 at 500 lbs per acre, and 0-20—0 at 240 lbs per acre. Frazier (7) working in Maryland with lima beans reported that band placement of 448 lbs of 5-10-8 caused 21 percent more yield than a broadcast application of 800 lbs of 5-10—8. These data were obtained in 1956 on a silt loam soil. Whittaker (25) has stated, as a conclusion from a ferti- lizer placement experiment, that the effect of band placement appears to be proportional to the solubility of the salt used. The greater the solubility of the fertilizer source, the more the efficiency is improved by band placement. Not all of the comparisons of band versus mixed appli- cation have resulted in favor of band placement. The summary by Starostka of the Sherman and Hardesty review included the reports from the twenty-three experiments conducted in the period 1951-1950 in which localized and mixed placements were compared. Of these twenty—three experiments, about 50 percent reported no significant difference due to placement. Almost 50 percent of the experiments showed that localized placement gave an increase in yield over mixed placement. Not all of the results were statistically analyzed. -1o— EXPERIMENTAL MATERIALS AND KETHODS A. Soils The two soils chosen for this experiment were Conover loam and Hiami sandy clay loam collected from the R.L. Cook Farm, Ingham County, Michigan by taking the tOp six inches from several areas selected at random. Although these two soils are very similar in many characteristics they differ in their capacity to retain phosphates against dilute acid extraction. The Conover series is described by Veatch (24) as dark colored loans and silt loams underlain by yellowish and mottled gray massive gritty clay to depths of several feet. Generally, these soils are non-acid or only slightly acid and are of high fertility. They are found on smooth plains and swales intermediate in drainage between Miami and Brookston types. The original vegetation was hardwood forests; large individual tree growth, mainly of elm, hickory, ash, and bass- wood. he chief agricultural value of these soils is for the production of hay, grain, beets, alfalfa, and pasture. The larger bodies of these soils constitute first class agricul- tural lands. The Miami series is described by Veatch as light brown- ish loam and silt loam over brownish compact and retentive but granular gritty clay. The clay substratum extends to depths of several feet. These soils are moist but not excessively -11- wet. The surface is usually somewhat acid but the profile is limy at shallow depths and, in general, these soils are consid— ered of relatively high fertility. The land characters where these soils are found are gently rolling upland clay plains, associated swales of wet darker colored clay land, and lakes and much swamps. Locally, the SIOpes are steep. The original vegetation was dense forest of sugar maple and beech with variable prOportion of oaks and hickory. Agriculturally these soils rank high and are used extensively for general farming purposes. Several physical and chemical prOperites of the two soils are tabulated in Table I. B. SuperphOSphate The two superphOSphates used in this experiment were radioactive materials supplied by the Unites States Department of Agriculture, Beltsville, Maryland. The specific activity of each material was reported as 0.2 millicuries per gram of P205. The pile date for each material was February 28, 1951. The granulated superphOSphate was 21.0 percent P205 by analysis made at the United States Department of Agriculture, Bureau of Plant Industry,SoilsJand Agricultural Engineering laboratory while an analysis performed in the Soil Science laboratory by the author indicated the material to be 20.7 percent P205. The material was reported by the Beltsville laboratory to be 14—20 mesh. TABLE I SOHE PHYSICAL AND CHEMICAL PROPERTIES OF THE SOILS USED IN THIS EXPERIMENT .-a-‘ .4 as- PrOperty Conover Miami Mechanical analysis (hydrometer) , greater than 50 microns. 50.4% 50.9% 5 to 50 microns 29.2% 27.1% 2 to 5 microns 5.4% 4.4% less than 2 microns 16.8% 17.6% Organic matter content (C X 1.724) 5.22% 5.05% Cation exchange capacity 11.70 m.e. 10.10 m.e. per 100 gm per 100 gm Available P level Spurway, reserve method (21) 46 lb/A6in 25 lb/AGin Bray, total available (4) 87 lb/A6in 95 1b/A6in pH (by glass electrode) 6.58 5.74 Percent applied P not extractable after incubation Spurway, reserve method (21) 84.0 90.0 Bray, total available (4) 75.6 82.1 The non-granulated material was reported to be 19.1 percent P205 by the United States Department of Agriculture laboratory while an analysis of the material at East Lansing indicated the P205 content to be 18.4 percent. The particle size distribution of this material was determined with the following results: 8-20 mesh 16% 20-40 215 40-100 46% loo—zoo 113 less than 200 7% C. Procedure Greenhouse. In the greenhouse work, field beans and wheat were grown in three-gallon glazed clay pots having an inside diameter of 9.5 inches and a height of 10.75 inches. The pots were placed at random on the greenhouse benches. The treatments, replicated four times, were as follows: 1. Granulated superphosphate placed in bands one inch be- low and one inch exterior to the ring of seeds. 2. Non-granulated superphosphate placed the same way. 3. Granulated superphOSphate mixed with the tOp four inches of soil. 4. Non-granulated superphOSphate mixed in the tOp four inches of soil. Each treatment was at the rate of 1000 pounds per acre as 0-20-0 and was repeated for each of the two soils, Conover and Miami, and each of two crOps, Michelite Beans and Henry Variety of Spring wheat. Sufficient supplementary nutrients were applied to each pot to simulate per acre applications as follows: 1. 200 lbs of N as NaN05 2. 500 lbs of K20 as K01 5. 5 lbs of borax as H5805 4. 20 lbs of CuSO4 5. 5 lbs of Zn804 6. 20 lbs of Hn804 These applications, as well as those of superphOSphate, were made on an areal basis - the calculated value of 1.15 x 10'5 acre per pot. The supplementary nutrients were added in dry form and mixed with the top four inches of soil in each pot. Moisture content was adjusted regularly by bringing the weight of the pot, soil, and moisture up to field capacity. Depressions for the planting of the seed and the bairappli- cation of superphosphate were made by pressing the rim of a clay pot into the moist soil. The ring for the seed was five and one-half inches in diameter and three—quarters of an inch deep. That for the fertilizer was seven and one-half inches in diam- eter and one and three—quarter inches deep. Care was taken to use only a minimum of pressure to avoid compaction of the adjacent soil. The seeds were planted on April 11, 1951 and the wheat and beans were thinned to six plants per pot on.April 20 and 25 reSpectively. The above ground plant material from this thinning was retained for analyses for total phOSphorus content -15- and radioactive phOSphorus. Soil samples were collected from the "mixed placement" pots on April 25. Harvests were made on May 17, May 29 and June 21 and soil samples were collected from the "mixed placement" pots on approximately the same dates. In each harvest, except the last, one plant was harvested from each pot. In order to have sufficient dry matter for radioactive counting it was neces- sary to composite the material from the four replicates of each treatment. In the last harvest, the four plants remain- ing in each pot were harvested for yield data and for chem- ical analyses. Analyses were made for total and for radio— active phOSphorus as with the previous samples. Laboratory -_plant phosphorus. For the determination of phOSphorus in the plants which was takeh up from the ferti— lizer, one gram samples of material were ashed in the muffle furnace which was maintained at 500°C for twelve hours. The ash was dissolved in one and one-half to two milliliters of 2 normal HCl, transferred to a three milliliter volumetric flask and made made up to volume with distilled water. In- soluble silica was retained with the sample. A large eye drOpper was used to facilitate the washing and transferring of material. Aliquots were prepared for counting by transferring one milliliter (55 1/5 percent sample) of the shaken suspension to a weighed aluminum disk nine centimeters in diameter. A drop of deliQuescent salt solution (20 percent CaC12) was -15- added to each disk and allowed to dry in Open air. The Ca012 was added to prevent dusting of the sample. In order to facilitate calculating the area of residue, a thin coat- ing of stOp—cock grease was placed around the border of the disk to keep the liquid in a more circular pattern. Then the disks were dry, radiation was measured with a Tracerlab Auto Sealer ~ Mark II using a Tracerlab tube TGC-2/1B84 with a mica window; absorption thickness of 1.9 mg/cmg. To assure uniform geometry for different samples, two concentric depressions were turned into a lucite block. The deeper one was made to fit the aluminum disk containing the sample whereas the second was cut to fit the tube shield. With this arrangement the tube could be centered over the sample with a distance of about three millimeters between the sample and the mica window. This procedure was suitable for counting of plant ash samples up to about eight half lives with reasonable precision. At least five series of 256 counts each were made for each sample. Correction for mass absorp- tion was made on each average count. The counts from a given sample were compared to those from a known standard for the calculation of concentration. The standard was counted at the beginning and end of each period of counting which was usually about four hours. The counts were made following the directions outlined in the Auto Sealer Manual (9). For the radiation measurements on materials collected on the fourth sample, it was found that the precision of count- ing could be improved by use of a different procedure. Trials -17- revealed that by compressing the ground air-dry plant mater- ial into small pellets, larger samples could be effectively concentrated into the defined counting volume and thereby increase the counts per minute which could be obtained from a given sample. This method also proved much simpler, more rapid, and more accurate by reducing the number of steps in the procedure. The procedure used was an adaptation of the method devel- Oped by Dion and described by Mackenzie (12). However, with the use of the Gieger-Mueller tube with a very thin mica win- dow (TGC-2), it was found that sufficient area to get a satis- factory counting rate could be exposed to the tube by press- ing the material into a pellet instead of hollow cylinder (as developed by Dion). The diameter of this pellet was made equal to that of the mica window. The pellets were pressed in a specially designed and constructed form using a Carver press. Several pressure and time combinations were tried and the combination of 12,000 lbs per square inch for one minute (least rigorous tried) was found to be as effective as any tested. The pellets were one inch in diameter and five-sixteenths of an inch thick, containing six grams of air-dry plant mater- ial. The weight and thickness of the pellet is not critical if the absorption thickness is sufficient to completely absorb the beta radiation emitted by atoms furthest from the tube. Standards for this analysis were prepared by mixing an aliquot -18... of standard P52 solution with a portion of check material. The counting was carried out in the manner described above. Laboratory procedure for total phosphorus. The portions of samples remaining after the aliquot for counting was re- moved (2 milliliters) were filtered to remove the solid silicon dioxide and diluted for the determination of total phosphorus. A number of aliquots were dehydrated to remove soluble silica and it was found that this procedure has no effect on the accuracy of the determination. This is in agreement wdth the findings of Truog and Meyer (25) even though a different re- ducing agent was employed for the Deniges' blue reaction. The total phOSphate was determined colorimetrically by the Deniges' blue reaction. The procedure followed was as outlined by Bray (4) using the Fiske-Subbarow reducing agent which is a mixture of sodium sulfite, sodium bisulfite, and amino-naphthol-sulfonic acid. The color was allowed to develop for fifteen minutes and then its intensity zas measured with an ' Evelyn Colorimeter using a 620 millimicron filter. Standards for comparison were prepared from mono—potassium phosphate. ‘gaboratory methods for soil characterization. The most important soil characteristics in this eXperiment were the available phosphate levels at the different sampling dates and the phosphate fixing capacities. The available phOSphate levels were determined by two rapid test methods. These two differed only in the method of extraction used. One procedure was that of Spurway for the "reserve" soil phOSphorus in which phosphorus is extracted. with 0.135 normal H01. Two gram samples of soil were shaken with eight milliliters of the eitracting solution for one minute, filtered and diluted 5:1. Extracted phosphorus was measured colorimet- rically with the Evelyn Colorimeter. ' The second method of extraction used was that of Bray (4) for removal of "total available“ soil phosphorus. Two gram samples of soil were shaken with twenty milliliters of a so- lution of 0.1 normal with respect to H01 and 0.05 normal in h.4F,filtered,and diluted 12.5:1 for colorimetric determination. The phosphate fixing capacities of the two soils were measured by adding several different increments of phOSphate as mono—potassium phosphate to soil samples in tumblers in the laboratory and incubating for several weeks, then extracting with the methods described above. The phosphate was added in amounts calculated to simulate application of 60, 120, 300, 600 and 1200 lbs per acre of phOSphorus in duplicate for each soil. The cultures were incubated for six weeks under labor- atory conditions. During this period, the soil was moistened to 35 percent water and allowed to dry to 7 percent three times. The mechanical analysis of each soil was made using the hydrometer method as described by Bouyoucos (5). Organic matter content was determined using the carbon train method with ascarite (sodium hydroxide asbestos) as the 002 absorbent. The cation exchange capacity was measured according to the method of Peech (16). Normal neutral ammonium acetate was used to saturate the complex with ammonium ions, which were then replaced with 10 percent sodium chloride solution and de- termined by the KJeldahl method. Soil reaction values were determined with the glass electrode and the Beckman pH meter, model H—2, using 1:1 soil- water suspensions. The soil-water mixtures were allowed to stand for fifteen minutes before readings were made. All of the above mentioned soil prOperties were determined on duplicate samples with a third determination made when needed. Laboratory method for superphosphate analysis. The super- phOSphate used was analyzed for total phOSphorus in its origi- nal condition and after it had been in band placement in the soil for the period of plant growth. The material recovered from band placement was carefully removed from the soil with tweezers and sorted under a hand lens to remove soil particles. A reasonably pure sample of the residue was obtained in this manner. Phosphorus in this recovered material and in the ori- ginal fertilizer was determined using the standard A.O.A.C. method (volumetric Option) (1) for fertilizer phosphorus. Briefly, this method consists of digestion of the samples in HN05 and H01, filtration, dilution, and precipitation of phos- phorus as ammonium phospho-molybdate. The precipitate is dissolved in standard NaOH and the excess NaOH titrated.with standard HCl using phenolphthalein as an indicator. The particle size distribution of the non-granulated superphosphate was determined by shaking at one-quarter speed a 50 gram sample in an appropriate nest of sieves fitted to the Cenco-Meinzer sieve shaker (catalog number 18480) for fifteen minutes. The residue on each sieve was brushed into a weighing can and weighed. ~22- RESULTS AND DISCUSSION A. Soil phOSphorus Original_pho§phorus level. As shown in Table I, the Conover 8011 contained 46 lbs of phOSphorus extractable with 0.155 normal H01, while the Miami soil contained 25 lbs per acre six inches. Spurway and Lawton (21) have reported the critical level with this test to be about 40 lbs per acre. Soil con- taining less than this amount of acid extractable phosphorus can be expected to respond to phOSphate fertilization, while those containing more usually do not. Using this hypothesis, only small yield response of crOps to phosphate on these soils would be expected with a greater plant response on the Miami than on the Conover soil. It was found in this experiment that there was more response on the Miami soil as shown by the yield data in Tables VII, VIII, IX and X. Bray (4) has develOped a system for predicting the re- sponse to fertilizers. According to this system, the soil nutrients have a variable availability which depends on the mobility of the nutrients in the soil and on the nature of the plant. Those with little mobility such as phosphorus, tend to follow the Baule percentage yield relationship. The soils used in this experiment were found to contain 87 lbs per acre six inches of total available phosphorus for the Conover and 95 lbs for the Miami according to the method of Bray. These values are both high according to Bray's criteria and consequently very little if any response to fertilization would be expected. From the yield data it \ may be noted that there was no appreciable increase in yield.§ Fixation studies. Since the laboratory extraction pro- cedures used to determine the levels of available phosphorus in the soils have been found to have good correlation with yield and response measurements, these procedures could also be used to measure the capacity of soils to fix phOSphorus in unavailable form. A great many phosphate fixation studies have been carried out by investigators, and from their reports, it would appear that the process of phosphorus fixation is probably a combi- nation of chemical precipitation reactions, replacement re— actions, and physical and chemical sorption. It seems, that if fixation is effectively such a combination, the capacity of a given soil to "fix" phOSphorus should be definite, and that it should be possible to saturate each of the capacities individ- ually and therefore possible to saturate the total capacity. The large quantities of phosphorus (up to 2850 lbs P205 per acre) used in the fixation trials in this experiment were used in an attempt to saturate the capacities of the two soils used. The criterion used to determine saturation was that the . percentage of the applied phosphorus not extractable should decrease when the saturation level was attained. Several appli- cation rates were used. The data for these trials are given in Tables II, III, IV, and V. TABLE II PHOSPHORUS FIXATION BY CONOVER SOIL AS MEASURED BY THE SPURUAY RESERVE EXTRACTION METHOD* Phosphorus Extractable P Applied.P Applied P not applied - -1bs/acre/6" recovered recovered - lbs/acre/d" -lb§/acre/6" 0 4O 61 48 9 85.2 125 59 19 84.6 151 56 16 87.8 509 86 46 85.1 517 81 41 87.1 626 155 115 81.7 652 150 110 82.6 1228 271 251 87.6 1259 280 240 85.4 average 84.0 ‘All values expressed as elemental phOSphorus (P). TABLE III PHOSPHORUS FIXATION BY CONOVER SOIL AS MEASURED BY THE BRAY TOTAL AVAILABLE EXTRACTION METHOD‘ W Phosphorus Extractable P Applied P Applied P not applied - -lbs/acre/6” recovered recovered - _1b§/acre/6“ -lbs/acre/6' 0 88 61 105 15 75.8 125 120 52 74.0 151 118 50 77.1 509 165 75 75.7 517 165 75 76.4 626 248 160 74.4 652 255 165 75.9 1228 568 280 78.4 1259 580 292 76.2 average 75.6 __ Bill values expressed as elemental phosphorus (P). TABLE IV PHOSPHORUS FIXATION BY’MIAMI SOIL AS MEASURED BY THE SPURNAY RESERVE EXTRACTION METHOD“ PhOSphorus Extractable P Applied.P Applied P not applied - -lbs/acre/6” recovered recovered - lbslacrezp" ~1bs/acre/6" O 21 61 26 5 92.1 127 29 8 94.5 128 55 12 91.6 505 51 50 90.9 508 48 27 91.2 628 84 65 90.7 655 87 66 90.1 1250 177 156 88.5 1244 186 165 87.9 average 90.0 “All values expressed as elemental phOSphorus (P). TABLE V PHOSPHORUS FIXATION’BY'MIAMI SOIL AS MEASURED BY THE BRAY TOTAL AVAILABLE EXTRACTION METHOD* Phosphorus Extractable P Applied P Applied P not applied - -lbs/acre/6“ recovered recovered - _lbs/acre[6” __ -lbs/acre/6' O 95 61 107 12 81.7 127 117 22 85.9 128 122 27 79.5 505 150 55 82.1 508 145 50 84.5 628 200 105 85.8 655 208 115 82.4 1250 527 252 81.2 1244 544 249 80.7 average 82.1 WAll values expressed as elemental phOSphoruS (P). -25- InSpection of the data in Tables II, II, IV, and V shows that the percentage of applied phosphorus not recov- erable remained essentially constant over the range investi- gated. Variations which did occur are probably within exper— imental error. Thus, it can be assumed that either some factor not considered above enters into the fixation capacity of a soil, or the total phOSphorus retention capacity was not saturated even at the highest level of application of some seven and one-half tons of superphosphate per acre six inches. Effect of crop growth on available phosphorus level. Soil samples were collected at the different sampling dates and analyzed for available phosphorus by both methods. Data from these analyses show that the change in available phOSphorus during the growing period was small. The available phosphorus levels for each condition on each date are given in Table VI. It should be noted that in some cases a decrease did occur and that in some cases the reduction amounted to about 10 per- cent in the "Spurway extractable", and often somewhat more in the "Bray total available“. The fact that the amounts of ex- tractable phOSphorus is higher for non-granulated treatment than for granulated treatment probably is related to the dif- ferences in dissolution rates of the two materials. This dif- ference had virtually disappeared by the second sampling date and consequently, an indication of the effective life of the granules or the length of time during which particle size has an effect on dissolution rate of the material is given. -27- nHH sea sHH OHH 4‘ m .cdnwsno: n we as we sends moa oma ema was an we sn we coeds .qsnm os os as ea sa em an em scone ugom an bHH 80H III aw en Hm III chHa .cwnwnco: mad ONH boa ONH mm mm an an doxaa .cmnw as os es sm ma mm mm mm Mocha scone smog hwao apnea dead: oma osa osa mom mm so am mos sends .eeswueoe sea Hod sea msa mm om mm sm emwaa .essm as am so om we we we we scone madam msa ooa oma mam so we sm cad seeds .eesw;eoe Hos ass sea was so woa mos sea scans .eesm mm mm mm as we mm we we moose noes; smog nobocoo am am sH as am mm as mm on:& has has asked mean sex sex Haadd caneflasse asses seem moapomapxm obaommm hasnzmm eczema Mam.sAom hem msaonmwomm,manmaam>m yo mezzom soapddcoo g qum Mme 2H mbmommmomm mam¢AH¢5¢ zo maxomw memo m0 Bowman H> Handy _.a _/ "I "~'\' c ._ - q ,' L..- r r" ._- D CrOp growth during the swigfrom April 25rd to May 17th y could scarcely account for the large decreases in available phOSphorus noted in some cases. Hence, the difference must be due to a change of state; that is, from a soluble salt at the first date to a reverted or "fixed" form on the second date. B. Yield Yield measurements were made on the material collected on the fourth sampling date when the plants were 71 days old. At this time the fruit of both the beans and the wheat was well develOped but not sufficiently ripe to allow threshing. For this reason, fruit weights include the entire fruiting body in each case. The fruiting bodies were separated from the foliage when the material was first harvested to avoid shattering losses. Yield measurements are reported as follows: [.1. Total weight per plant. 2. Number of fruits per plant. 5. Weight of fruit per plant. 4. Weight of foliage per plant. The data are recorded in Tables VII, VIII, IX, and X. The values given represent the average for sixteen plants from four pots in each case. The data reveal that the treatments had a greater effect on the yields of plant material from Miami than from Conover soil but, as was expected, in neither case was the increase very large. -29- TABLE VII YIELD OF FIELD BEANS 0N COUOVER SOIL :: wt./2 fruits/2 £55157 roffggg/ plant plant plant plant Treatment grams number grams grams check 8.70 5.50 4.15 4.55 gran. band 8.64 5.75 4.05 4.61 gran. mixed 8.79 5.56 4.05 4.74 non-gran. band 8.40 5.19 5.85 4.57 non-gran. mixed 9.06 5.50 4.05 5.01 no significant difference at 5% level TABLE VIII YIELD OF FIELD 57 US ON HIAHI SOIL 2:2 ‘ wt.]; fruits?» fruit/ foliage/ Treatment $251; 133.33; 2533: $3.33 check 8.14 5.50 4.51 5.55 gran. band 8.59 6.94 5.46 5.85 gran. mixed 9.27 6.50 4.55 4.94 non-gran. band 8.40 6.15 4.26 4.14 non—gran. mixed 9.54 6.56 4.54 4.50 no significant difference at 5% level TABLE IX YIELD OF SPRING WHEAT 0H CCNOVER SOIL ”“5 Dnt./ fruit57» fruit/5’ foliaé:7 plant plant plant plant Treatment grams humber_ grams grams__ check 5.58 5.50 2.58 5.00 gran. band 6.56 5.69 2.74 5.62 gran. mixed 6.01 5.86 2.74 5.27 non-gran. band 5.56 5.52 2.57 2.79 non-gran. mixed 6.24 5.87 5.01 5.25 no significant difference at 5% level TABLE X YIELD OF SPRING-WHEAT 0N MIAMI SOIL wt./ fruits/ fruit/ foliage/ plant plant plant plant greatment grams number grams _grams check 4.59 5.58 2.25 2.15 gran. band 6.50 4.51 2.80 5.50 gran. mixed 6.42 4.58 5.05 5.57 non-gran. band 5.80 4.02 5.00 2.80 non-gran. mixed 6.51 5.27 5.02 5.49 —__ no significant difference at 5% level Statistical analysis of the yield data failed to reveal any significant differences between treatmentsbecause of the great variability within a single treatment. Certain trends can be pointed out in the yield data. It is noted that the values cited are usually higher for mixed placement than for band placement. This is true except in the case of spring wheat on.Miami soil. In the comparison of granulated and non-granulated material, the yield differences are inconsis- tent, consequently very little can be deduced from the data. From a summary of all yield comparisons it would appear that the non-granulated material may have been slightly superior on both soils for both crOps. Attempts were made to measure the early effects of the treatments by making height measurements of the plants at the ages of thirty-six and forty-eight days. Again the great variation between individuals became a problem but by discarding the individuals with extreme deviations from the mean in each pot, average heights were obtained which could be considered reliable indicators. These averages are given in Table XI, and it is noted that the differences due to treatment were small. They were of about the same order of magnitude as the differences in final yields. During the growing season some general observations were recorded. There were no marked differences in the vigor of the plants, nor in the rate of growth. Blossoms appeared on all plants at about the same time. Most wheat plants were -52- TABLE XI INFLUENCE OF TREATMENT ON THE HEIGHT OF PLANT _A_..- Average height per plant - inchgs Age - 56 days Age — 48 days Treatment (average of 18 (average of 14 plants) plants) Conover Loam non-gran. mixed * average of 10 plants ** average of 8 plants Wheat check 14.1* 22** gran. band 14.7 26 gran. mixed 14.9 26 non-gran. band 14.9 25 non-gran. mixed 14.8 26 Beans check 7.7* 25'M gran. band 9.7 22 gran. mixed 9.7 25 non-gran. band 9.5 22 non-gran. mixed 8.7 24 Miami Sandy Clay Loam Wheat check 14.1 25** gran. band 14.5 24 gran. mixed 14.5 25 non-gran. band 15.2 2 non-gran. mixed 14.6 26 Beans check 6.8 20** gran. band 8.5 24 gran. mixed 7.7 22 non-gran. band 9.5 22 7.8 22 -55- Jointing at or about the age of 56 days, and the beans had an average of seven leaves per plant regardless of treatment. At the age of forty days, heads were appearing on the wheat plants, but the order of appearance of heads was apparently not related to treatment. Blossoms appeared on many of the bean plants at the age of forty days but again the appearance of blossoms seemingly was unrelated to treatment. C. Phosphorus content of plant material Total phosghorus content. The values for total phos— phorus content appear in Table XII for wheat and in Table XIII for beans. In spite of the high phosphorus level on which the test crops were grown, the total phosphorus contents were not much higher in the mature plants than those reported as aver- age values for these crops by Morrison (15). These data were analyzed statistically by the analysis of variance method and it was found there was no significant difference at the five percent level although certain trends are indicated. For the last sampling date, as in the total yield data, mixed placement resulted in slightly higher phosphorus content than band place- ment for both crops and both soils. There appears to be no consistent effect of granulation in the last date of sampling. In the data for the third sampling date, band placement appears superior for the wheat on both soils. However, the beans grown with mixed placement are slightly higher in phos- phorus content. Granulated phosphorus resulted in a higher total phOSphorus content in the wheat grown on Conover soil, TABLE XII CCNCEN TRATION 0F PHO SPHOPUS IN DR" WHEAT ILATERIAL 5011‘ A; Conover i557 Kiami Fertilizer Gran. Non-gran. Gran. Non-gran. Placement band mix band mix band mix band mix _ milligrams perg gram * milligrams per_gram * April 25 5.40 7.56 4.74 6.45 6.15 4.55 5.25 5.92 May 17 5.60 5.90 4.55 4.55 4.07 5.90 5.68 4.05 May 29 5.55 2.70 2.70 2.48 5.00 2.17 5.57 2.85 June 21 2.99 5.54 5.88 5.58 5.45 2.61 2.84 5.12 * to convert to percent phOSphorus, divide by 10 ‘5 no significant difference at 5 percent level by analysis of variance TABLE XIII COUo-LlRATI‘” 0F EhCSEHORUS IN DRY BEAN L'ATEHIAL 55il Conover Miami Fertilizer Gran. Non-gran. Gran. Non-gran. Placement band mix band mix band mix band mix .__ pmilligramSper gran * milligrams per gram * April 25 5.02 4.71 5.81 5.57 7.58 5.22 .28 5.94 May 17 2.94 2.25 5.60 2.29 2.70 2.48 4.57 2.57 May 29 2.40 2.70 2.50 2.72 2.48 2.40 2.48 2.62 June 21 5.88 5.98 5.90 4.06 5.85 4.12 5.82 5.85 * to convert to percent phosphorus, divide by 10 no significant difference at 5 percent level by analysis of variance while non—granulated superphosphate resulted in a higher phosphorus content in the wheat grown on Hiami soil. The effect of granulation was inconsistent for the bean plants of the third sampling date. An inspection of the data in Table XII, for the second sampling date, reveals that for wheat grown on Conover soil the non-granulated superphosphate gave a somewhat higher phos— phorus content, while for wheat grown on Miami the granulated fertilizer gave slightly higher values. The effect of place- ment was inconsistent. According to the data in Table XIII for the second sampling date, phosnhorus content of beans, at the age of .L thirty-six days, was_appreciably higher in those grown with Eagd piacempnt on both soils. It also appears that the uptake of phosphorus from non-granulated superphOSphate was greater where the band placement was used. 0n the first sampling date, mixed placement resulted in higher phOSphprus content for the wheat grown on Conover, lower phosphorus for the wheat grown on Miami, and lgwer‘phos- phorus content for the beans grown on either soil. The effect I 0f granulation was somewhat inconsistent on the first sampling date, as shown in Table XII, but granulated material increased the uptake of phosphorus by wheat, reduced uptake by beans on Conover soil and increased the uptake by beans on Miami soil. For all combinations of soil, crop, and treatment the total phosphorus content was high in the young plants, de— creased to a minimum at the third sampling, and then increased again but never attained the high level found in the young plants. Phosphorus from fertilizer. The content of fertilizer phosphorus in the plants was determined as described in the procedure by measuring the radiation due to the radioisotope content. The values obtained by this measurement are given in Table XIV for wheat, and in Table XV for beans. It may be seen from the tables and also from Figures 1 to 4 that the content of fertilizer phosphorus in the tissue was low at the first sampling, increased to a maximum by the time of the second sampling, and then gradually and almost linearly diminished. This trend was true regardless of the soil, crOp or treatment and is entirely as would be expected. The first sampling was made when the plants were only eleven days old. At this time, probably an appreciable portion of the phos- phorus in the tissue came from the reserve in the seed from which the plants grew. As the plants grew older, more of their phosphorus came from the fertilizer until such time as the root system extended beyond the zone of influence of the fertilizer phOSphorus. This condition should be expected to occur earlier with the band than with the mixed placement, but this difference did not show up in this experiment. Probably samplings were not made with sufficient frequency to find such a relationship. It will be noted that the increase in fertilizer phos- phorus from the first to the second sampling was always greater -37- TABLE XVI CONCENTRATION OF FERTILIZER PI.% HIOPUS IN DE’N WHEAT LATERIAL fi- Soil Conover Miami Fertilizer Gran. Non-gran. Gran. Non-gran. Placement band mix band mix band mix band mix milligrams per gram * milligrams per gram * April 25 1.21 1.87 0.75 1.96 1.55 1.15 1.55 0.95 May 17 5.26 5.01 4.54 5.20 4.05 5.24 5.55 2.99 May 29 2.86 1.84 2.61 2.29 2.68 1.78 5.10 2.08 June 21 1.92 1.70 2. 22 1.62 2.24 1.56 1.81 1.40 * to convert to percent, divide by 10 TABLE XVII CONCENTRATIC 1 OF FERTILIZER PHOSPHORUS N DP.Y BEANI ATBHIAL Soil Conover Miami Fertilizer Gran. Non-gran. Gran. Non-gran. Placement band mix band mix band mix band mix milligrams per ggam * milligrams per gram * April 25 0.78 0.70 1.06 0.60 0.95 0.51 1.29 0.48 May 17 2.85 1.85 5.10 1.91 2.76 1.28 2.81 1.59 May 29 2.52 1.75 .15 1.87 2.26 1.57 2.25 1.61 June 21 1.68 1.58 1.45 1.56 1.51 1.55 1.01 1.24 —‘ * to convert to percent, divide by 10 ~58- .daOu hopoaoo no steam ”was: :« masonauozn uuuaaavncu uo doapdhucoocoo no omd panda no vacuum one H .w«& chad t uncdan no ow< douas o caudasnanwacoa asap 1 dopunssdhmacoz venue a douaaacduc camp I cooafinndao 0X00 H.o L «.0 R I gamma» amend aha ca n.o ushonnuona 4.. L06 chauOGOAUGm .Haom «swax no aboam poems :« msaonnmonn Aouaaauaou no coupmnucoocoo no owe pcmad no poomuo any m .Mab aha-0 I mac-5H0 ho omd Ob 00 on . 8 on ON OH - q q . u q . . a.o / . n w.0 % I onuudu pecan had a« 0.0 ushonnaonn obdpooodudm Gouda ... vowed-558-com - 053 o voyage-unwise," o .. n «.0 sends t mopdasaaao -. 055 .. coped-:35 . among suuaaaawoe . a aohocoo so cream mason an ushon “woGOApaapcoonoo no 005 sedan no noouho any n .mnh . as I 3-3.3 no 004 —4o- ” JNoO a o cam-a» . V psdan and ad InOO usaonnnona .5 583-3- 533 .. eouaasfieméoa a 15.5 snap : eoaaaseeswneoz . wands a ooaoasadao x dean . deceaseaeo . .Hdou «and: no macaw scoop an «manganese nonnaavnon no coaumnucoocoo no owe panda no poonno may v .wdh undo o unnean no 9w4. On. 0% CO 3 00 ON OH - - a, - - a I . O i \I /o JNOO R s manna» pecan nae an Ln.o usaoaauona downs I copwflsmdnwtcoz I obdpodOHUGm 53 .. cop-4858-52 o Jae UOHHE I dopmfiscmka n 23 .. eon-355.6 . no.0 for the band placement than for the mixed placement. This was eSpecially trhe in the case of wheat. This fact suggests that a fully effective root system had not developed to in- clude the fertilizer band (one inch below and one inch to the side) within twelve days. The concentration of fertilizer phosphorus was generally somewhat higher in the mature wheat than in the mature beans, although this trend did not follow for total phosphorus. This was probably due to the difference in the root systems of the two crops. The beans foraged more effectively in the soil below the fertilizer level (fertilizer phosphorus never applied to depth greater than four inches). “However, the decrease in radioactive phOSphorus content in both crOps after the second date, regardless of placement, would indicate that both crOps undoubtedly derived considerable phOSphorus out- side of the zone of fertilizer influence. The effect of placement on the concentration of ferti- lizer phosphorus in the plants may be noted from Tables XIV and XV. It is readily seen that the band placement was superior in all except the first sampling date in the case of wheat grown on Conover soil. Here, for some reason, mixed placement produced greater plant absorption of phosphorus. The effect of placement did, however, show some variation with time, being greatest at the second sampling. This tendency is graphically illustrated in Figures 1 to 4. Figures 1 to 4 also illustrate the effect of granulation, on the concentration of fertilizer phOSphorus in the plant material. The effect of granulation is inconsistent and although no analysis of variance was made of these data, there probably was no significant difference due to granulation. From the conclusions made in the section on soil phosphorus studies, such a result would be expected and certainly no differences due to granulation would be evident after the first sampling date. Fraction on plant phosphorus derived from fertilizer. Among the experiments in recent years using radioactive phos- phorus for evaluating a fertilizer source or practice, the most commonly used basis has been the percent of total plant phosphorus derived from the fertilizer. This criterion is also included in this experiment although it is believed that the basis discussed in the last section (page 56) is more suitable for measuring the efficiency of the treatments, sep- arately and in combination. -_,_~__o_ The percent plant phosphorus derived from the superphos- ' K phate is given in Table XVI for wheat and in Table XVII for beans. The values presented were calculated from the data in Tables XII, XIII, XIV and XV. These data are graphically de- picted in Figures 5 to 8. The data indicate that band placement resulted in a larger percentage of fertilizer phOSphorus in the plant in all cases except at the first sampling of wheat. Here the mixed place- ment produced the higher efficiency. It is interesting to note that in the first sampling the percentage fertilizer phosghorus TABLE XVI PERCENT PHOSPHCRUS IN WHEAT DER 723 FR M FERTILIZER Soil Conover Miami Fertilizer Gran. Non-gran. Gran. Non-gran. Placement band mix band mix band mix band min April 25 22.7 24.7 15.4 50.4 24.9 26.4 25.4 26.1 May 17 90.6 77.~ 99.6 75.7 98.8 85.4 96.1 75.9 May 29 80.6 68.5 96.7 84.0 89.4 82.1 52.1 75.1 June 21' 54.2 51.0 57.2 48.0 55.4 52.2 51.2 45.0 TABLE XVII PERSENT PHOSPHORUS IN BEANS DERIVED FROM FERTILIZER ___ a 2:: Soil Conover Miami Fertilizer Gran. Non-gran. ran. Non-gran. Placement band mix band mix band mix band mix April 25 15.5 14.8 18.5 11.2 12.8 9.8 24.4 8.1 May 17 96.4 82.5 86.2 85.4 100 51.7 61.5 66.6 May 29 100 62.5 86.1 69.5 91.2 65.5 90.9 61.4 June 21 45.4 54.6 57.2 58.6 42.0 52.8 26.6 52.2 on vacancy and: 05. 230a .3555 no 5.9% neon: ho o ashozamonn ”swan no coaudmonscou mommmm amid : upqdan Ho owd l! 00 on Ca. on ON 0H - u - q q - _ ..om \ 2.. a \ J. 2 O . .Loc x q .aom .0058 I doudflflcdhwscoz a 0 use; . uoouascugwcaoz. . .Loon fiends I dopmflsqdho n camp 0 covuaannno . n a aauaadpp.u song ce>ahou an: noun) a paaan Have» no caduceus -46... A .32... saga no Esohw neon: .39 0983 on pomauon and: unnonnuona panda no coapamomsoo m .mah ammo s upcwfin Mo ow< as _oo on cc on ow oH . q . . . a . 4 ml. nouaadanou . om Bonn copapov . o¢ ow: nods: x m andam 0 4 on Hdep no . coupodna . om wands o oopwascwhwucoz n 00H can» a dopwnscmawtcoz ponds I nonmaacdmo scan s eopsascune .Hflom.nm>ocoocc cream mamas hos con50m 0p academy npas mayonnmcmm pecan Mo ccapfimogsoo s .w.E wand I mpcsaa mo ow¢ Ob 00 CD 0* 00 ON OH lull — _ . d d u - ‘Vk d \ a v fiQNdepHOH ow eon» fiQFdhOfl l 00 mw3 30.2.3. -47- m undag Have» no o coapomam doxda a Unp¢a3¢¢hMtzoz ”“19 I dflflflHSGdhwlfloz ponds t doucHstuw cusp a eouaascans 09H -48— .Hdom dead: so :Sonw ocean now oonzou on poonmoa and: msaosmmoan pcead ho codpamonsoo m .m«& chap I mecca“ no 0&4 .11 . . A a . 4 1\ O L om . u nonaaauhou o L+ 0”? BOHH x dobdhod 9 .low on: moan} m pecan 1 aspen no om coaaoagfi 0 onds 0 doucazcmnwanoz a [III/III 33 .. 533552-32 o . 1 co.“ ponds I deceasednw x . damn t coumHScaaw . in the total phosphorus of the wheat was lower for band place— ment than for mixed placement on both soils, while the total plant phosphorus concentration was lower for band placement for wheat only on the Conover soil. This situation probably results from the difference in the original available phosphorus levels of the two soils. It may be noted from Table XVI that on the second sampling date (plant age - 56 days) over 90 percent of the total phos— phorus in wheat was from the fertilizer when placed in bands. Then the fertilizer was mixed with the tOp four inches of soil, about 75 percent of the phosphorus in wheat came from the su- perphOSphate. Beans, on the average, were not quite as effective in the utilization of fertilizer phosphorus and in three instances the peak utilization appeared on the third sampling date in- stead of the second. The average tOp utilization percentage for beans was 87 percent on Conover soil, and 81 percent on the Miami. These peaks are appreciably lower than those for wheat and it appears that the greatest utilization for both crops is somewhat higher on Conover than on Miami soil. This difference is undoubtedly due to the fact that Hiami soil has about 10 percent higher "fixing" capacity than the Conover soil. As seen from Tables XVI and XVII and Figures 5, 6, 7 and 8, the effect of granulation on the percentage of plant phosphorus derived from the fertilizer was inconsistent, but in general -50.. granulated material produced slightly higher results on the final sampling date. D. Observations of fertilizer - root relationship Concentration of roots in bands. The use of rather heavy application in band placement made it possible to inspect the band after the plants were harvested. It was noted that there was a very heavy concentration of roots growing within the band of fertilizer material. Good distribution of roots outside the band was also noted. It was impossible to ascertain when the heavy concentration of roots in the bands develOped, but since the soil phosphorus levels indicated an early dissolution and fixation of the fertilizer phosphorus, it would seem probable that this concentration of roots did not develOp until after a major portion of the fertilizer phosphorus had dissolved and was "fixed“. It is extremely difficult to decide whether the effect was due to chemotrOpism or simply to the fact that the roots were following the path of least physical resistance. The fertilizer band probably provided the path of least physical resistance for two reasons. First, although extreme care was taken to avoid compression of the soil, some probably did occur when the depressions were formed for the fertilizer. Secondly, the dissolving of phosphate material might leave pores for root elongation and enlargement. Also, in the case of the granulated material, the average size of the particles was much larger than the average size of the soil particles and as such would cause the band to provide a more porous medium for growth than would the surrounding soil. It was -51- observed that the concentration of roots in the bands of granulated material was greater than in the bands of non— granulated material. Photographs were taken of the roots in the fertilizer bands and are included here as Figures 9 and 10. In Figure 9 there is evidence of heavy accumulation of bean roots in a band of superphosphate. It may be noted that the soil is massive in structure. The size of some of the bean roots found in the fertilizer band can be seen in Figure 10. The two large roots shown perpendicular to the stem were in the band of fertilizer. The presence of fertilizer granules clinging to branches of the main roots can also be noted. Contact of roots and fertilizer particles. MicrOphoto- graphs were made of plant roots which were recovered from the bands in an attempt to show more clearly the close contact of the fertilizer particles with the plant roots. These photo- graphs are presented as Figures ll—l9. Figure 11 shows a bean root in very intimate contact with several granules of granulated superphosphate. Some rootlets are shown growing into and through the superphosphate granules. Whether this penetration occurred before or after the dissolu- tion of the major portion of the easily soluble phosphate salt cannot be ascertained. The large granule in the center of the photo apparently caused a bending of the root branch going toward the upper right hand corner of the photo. Multiple fine rootlets are shown going into or around each of the granules. It is interesting to note that soil particles are not clinging -52.. Fig. 9 Accumulation of roots of bean plant in band Of granu- lated super- phosphate. (Miami soil) f Fi . 10 Lar e bean roots growing in band 0 g gragulated superphosphate. (Conover soil) F10 0. ll Weathered granules in intimate contact with root of bean plant. Fig. 12 Particles of non—granulated superphOSphate adhering to wheat rootlets. Fig. 15 Contact between bean root and granule of superphOSphate. Note enlargement of root at contact. Fig. 14 Root passing through eroded—out cavity of granule. Fig. 15 Granule which had not been in soil. Fig. 16 Cluster of several granules. Note erosion of large granule on right. Fig. 17 Granule removed from band. Note effect of dissolution. Fig. 18 Granule removed from band. Note effect of dissolution. Fig. 19 Two granules removed from band. Note difference. A small root is seen in cavity of granule on left. l C CO I to the root in a similar manner. In Figure 12 a wheat root is shown removed from a band of non-granulated superphOSphate. The population of rootlets found on the roots recovered from the band was extremely dense. Practically every rootlet present has attached to it one or several particles of super,LOSphate. The two large particles shown (upper left and lower center) were macroscop- ically identified as soil particles rather than superphosphate. Study of the picture indicates that some of the roots are vis- ible in outline in the interior of the particle. A large granule Just slightly below the focus of the camera is evident in Figure 15. he large root shown extend— ing from the lower right hand corner is in contact with it and possibly growing into it. The root appears to be enlarged at the point of contact. The smaller granule at the center of the picture is attached through fine rootlets to both of the large roots shown. Figure 14 shows a large root passing through an eroded cavity in a single granule. Both the white masses on opposite Sides of the root are of the same fertilizer granule and are connected under the root. It is also possible to see fine rootlets growing into the cavity in the portion of the granule shown above the root. The manner in which fertilizer particles eroded in the soil is presented in Figures 15, 16, l7, l8 and 19. In Figure 15 a granule is shown which had not been in the soil and is in -59- its original condition. The surfaces are not entirely smooth and regular, but a few small depressions and partition planes are visible. Figures 16, 17, 18 and 19 illustrate granules recovered from the soil. The particles shown in these figures illustrate several types of dissolution but it is apparent in each case that the dissolution was not limited to the external surface of the particle. As shown by the chemical analysis data, about two-thirds of the phOSphate had dissolved from these particles, hence the residue was probably essentially gypsum. Figure 16 shows a cluster of several large particles and a number of small ones about a plant root. Some soil particles are also present which may be distinguished by their darker cOlor. The large granule at the right shows appreciable inter- nal dissolution. Changes in_phosphorus content of superphosphate. Dupli- cate samples of the materials recovered from the soil were analyzed for total phosphorus to determine the effect of incu— bation in the soil on the analysis of the materials. These samples were composites of material collected from both soils. The pots from which they were collected had grown both crOps. The values are reported in Table XVIII along with the original phOSphorus contents. -50- TABLE XVIII EFFECT OF ONE EASON IK SOIL ON THE P205 C LITENT OF SUPERB-ICES RATE 3‘ —=é5!=_== Material percent P20 before after difference Granulated super 20.7 6.1 14.6 Non-granulated super 18.4 5.2 15.2 The total phosphorus content changed about the same amount in both states of aggregation. Available phosphorus content of the extracted samples was not determined. The total phOSphorus data indicate that granulation had no appreciable effect on the phOSphorus remaining after one period of plant growth. This is the same conclusion made by Sayre and Clark (18) who made a test of this sort in 1957. However, these investi- gators found there was very little change in the available phosphorus level of the material even after one year. The soil used was Ontario loam with a pH of 6.9. No plants were grown on the fields where the trials were made and the precipitation during the period of the experiment was not reported. -61... swam A greenhouse experiment was conducted to test the effect — E of granulation and placement on the efficiency of uptake of phosphorus from superphOSphate. Conover loam and Miami sandy -, clay loam soils were used and field beans and spring wheat were grown as indicator crOps. Both soils were relatively high in available phOSphorus \ before the fertilizer applications were made. PhOSphate fix- ation studies showed that the two soils differed appreciably “ in their ability to fix phOSphates. The "fixation" by Miami soil was appreciably greater than by the Conover soil. The growth of the crOps was found to have very little effect on the available phosphorus levels of the soils. The only general trend noted was the reduction in the available phosphorus level between the first and second sampling dates. This fact indicates the completion of the dissolution of the fertilizer phOSphorus and its conversion to a non-extractable form. Yield and plant height data failed to reveal any statis- tically significant difference due to placement or granulation. However, it was observed that mixed placement gave slightly higher yields in more cases than band placement. The differ- ence was very small. In the comparison of granulated and non- granulated material, the difference was variable and very little can be deduced from the data. Differences due to placement did show up by measuring the radioactive phosphorus content of the plant tissue, and in the percentages of plant phosphorus derived from the ferti- lizer. Band placement resulted in higher fertilizer phos- phorus contents in all cases except that of the first sampling of wheat (plant age - 11 days). The effect of granulation on these parameters was thoroughly inconsistent, and although no analysis of variance was made of these data, it was con~ cluded there was no significant difference due to granulation. Some micrOphotographs are presented to illustrate the characteristics of the dissolution behavior of the superphos- phate granules, and to indicate the intimate contact between the granules and the plant roots. No attempt was made to de- termine Just when the high frequency of intimate contact between the roots and fertilizer particles deve10ped, but it is believed that this occurred after the major dissolution rate of the monocalcium phOSphate had passel. 9. -53- «war-w may «A PQEL-J.JIJ C1113 Association of Official Agricultural Chemists, 1945. Official and tentative methods of analysis. Washington, Doc. Exlo 6:20-22. Borland, J.J. and Ar singer, J.I ., 1950. Crop response tests on granulated superphosphate containing radio— active phosphorus. Office Memorandum, U. S. Dept. Agr., Bur. Plant Industry, Soils, and Agr. Eng., Div. Soil Management and Irrigation. January 12. Bouyoucos, G.J., 1956. Directions for making mechanical analysis of soils by the hydrometer method. Soil Sci. 42:29:25-2 500 Bray, R.H., 1949. Correlation of soil tests with crOp response to added fertilizers with fertilizer require- ments. Diagnostic Techniques for Creps and Soils. American Potash Institute, Washington, D.C. pp. 55-96. Collings, G.H., 1947. Commercial Fertilizers, ed. 4. The Bl akiston 00., Philadelphia, Toronto. p. 450. Cook, R.L., 1954. A brief report of th is fertilizer placement work on beans as conducted at the Michiga Experiment Station in 1954. Proc. Tenth Ann. Meeting Mtl Joint Comm. on Fert. Appl. The Natl Fert. Assn., Jas sliington, D.O. p. 57. Frazier, V.A., 1956. Fertilizer placement on lima beans in Maryland. Proc. Twelth Ann. Meeting of the Natl Joint 30mm. on Fert. Appl. The Natl Fert. Assn., “ash iington, D.C. p. 116. Fuller, M.H. and McGeorge, W.T., 1950. Effect of source of phosphorus on the uptake of phosphorus by alfalfa and cotton in Arizona. Mimeo. Summary Qf the 1950 Field Research in Western States xith PO:3 Tagged Fertilizers. Phosphorus Sub-comm. for Jestern Region of the Phosphorus Work Group of the Natl Soil and Fert. Res. Comm. Handbook of Operating Instructions for the AUTOSCALER. Tracerlab, Inc. 55 Oliver St., Boston, Mass. 10. ll. 12. 14. 15. 16. 17. 18. 19. Hardesty, J.O. and Clark, K.G., 1951. Granulation of fertilizers. gricultural Chemicals 6(1):54-CS, 95 and 97. Hill, W.L., 1951. Particle size - plant nutrient rela- tionships in phosphate fertilizers. Mineo paper presented on program on Fert. Tech. arranged by Fert. Conn. of Soil Sci. Soc. Amer. Penn. State 3011., August 27. MacKenzie, A.J., 1950. Measurement of P02 activity in plant material by use of hollow briquets. Special Rep. No. 42. U.S. Dept. Agr., Agr. Res. Admin., Bur. Plant Industry, Soils, and Agr. Eng., Div. of Soil Management and Irr., Beltsville, Maryland. Methods of Applying Fertilizer. Recommendations of the Natl Joint Comm. on Fert. Appl., Washington, D.C. Revise Nov. 1949. Millar, C.E., Cook, R.L., Davis, J.F., and.McAllister, A.W., 1941. The effect of placement and fertilizer analysis on the yield and stand of cannery peas at Michigan Exper- iment Station in 1941. Proc. Seventeenth Ann. Meeting Natl Joint Comm. on Fert. Appl. The Natl Fert. Assn., Washington, D.C. p. 89. Morrison, F.B., 1949. Feeds and Feeding. The Morrison Publishing Co., Ithaca, N.Y. Ed 21 pp. 1156 and 1158. Peech, 4., 1945. Determination of exchangeable bases and exchange capacity of soils. Soil Sci. 59:25-58. Ross, U.H., 1951. Recent development in the preparation and use of fertilizers. Ind. and Eng. Chem. 25:19—20. Sayre, 0.3. and Clark, A.W., 1958. Changes in composition of granular and powdered fertilizers in the soil. Jour. Amer. Soc. Agron. 50:50-57. SchmeHLIM.R., Olsen, R.S. and Gardner, R., 1950. Utili- zation of phosphorus by sugar beets as influenced by source of phosphorus and methods of application. Mimeo Sugmary of 1950 Field Research in Western States with POL Tagged Fertilizers. Phosphorus Sub—Comm. for Western Region of the PhOSphorus work Group of the Natl Soil and Fert. Comm. 20' 21. 22. (0 CI] 0 25. I (1) U1 I Sherman, M.S. and Hardesty, J.O., 1950. Review of Experimental \orlz on the agronomic effects of particle size of r‘un7r77mariate and of mi:: d fertilizers con— taining superfio osphate or otfzier water soluble phos- phates. Plant Food Kemore ndum Peportl Io. 20. U.S. Dept. A3r., A3r. Res. Admin., Bur. Plant Industry, Soils, and A3r. En3., Div. of Fert. and.A3r. Lime, Beltsville, Maryland. Spurway, 0.3. end Lav.ton, K., 1949. Soil testin3; a practical system of soil diagnosis. Mich. A3r. Exp. Sta. Tech. Bull. 152 revised. Starostka, R.U., 950. Sunnerv of 52 e>2periments on placement and granule tion of superphosp1iate (1951—1950). Mimeo leaflet, inform7tion derived from PLAHT- FOu KEIWORAN 1H REPORT NO. -A3ronomic effects of particle size of superphospllate and of m xed fertilizers contain— in3 superphOSph7te or other water-soluble phospl rates -- by 3.3. nernan and .O. Hardesty (see ref No. 20 above) Tru03, E. Mnd rieyer, A.H., 1929. Improvements in the Deni3es colorimetric meth ed for pho phorus and arsenic. Ind. Eng. Chem. Anal. Ed., 1:1-6 -159. Veatch, J.O., 1941. Agriculture 1 land classification and land ty7es of Michigan. Mich. A3r. Exp. Sta. Spec. Bull. (“5" ol. Whittaker, C.W., Coe, D.G., Bartholomew, R.P., Volk, .W. and Rader L.F. Jr., 1947. Placement of calcium phosphates. Jour. Amer. Soc. Agron. 59:859—968. muulmumgufluuulmvummlllmmwltleI