THE RELATION BETWEEN FORMS OF SOIL PHOSPHORUS AND RESPONSE OF ALFALFA AND SMALL GRAIN TO ADDED PHOSPHATE t>7 ALBERT H. BOWERS A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Soil Science 1947 ProQuest Number: 10008737 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest, ProQuest 10008737 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 ACKNOWLEDGEMENT The writer expresses his sincere grat­ itude to Dr* Kirk Lawton for his guidance in the research herein reported, in preparation of the manuscript, and for furnishing p H measurements* Acknowledgements are gratefully rendered to Mr* L. S. Robertson for the field plot yield data and to Dr* J* P* Davis for assistance in the sta­ tistical analyses. The writer Is also indebted to other members of the Soil Science Department for helpful suggestions throughout the course of this study* contents INTRODUCTION REVIEW OF LITERATURE PLAN OF STUDY EXPERIMENTAL PROCEDURE DISCUSSION OF RESULTS SUMMARY AND CONCLUSIONS PLATES BIBLIOGRAPHY 49 INTRODUCTION In the earlier studies of soil fertility and plant nutrition workers sought to determine the amounts of nu­ trients in the soil by a total chemical analysis* found, however, It was that no practical correlations could be drawn between the total amount of nutrients and the require­ ments of plants. From this early concept there was a de­ cided reaction to the other extreme— as exemplified by D y e r ’s citric acid soluble phosphorus (15). This was the concept of measuring quantitatively that portion of the total amounts of nutrients which plants could actually absorb. D y e r ’s method allov/ed rather broad correlations with productivity, but its success was temporary since on many soils it failed in the determination of specific cor­ relations and separated only a minute quantity of what is now termed "acid-soluble” phosphorus. The use of carbonic acid as an extractant simulating the soil solution in the region of the root hairs is another attempt to measure the amounts of phosphorus available to plants. It has proven successful on the highly alkaline soils of the western states, but on acid soils this method has generally been unsatisfactory* Numerous dilute solutions of strong acids have been presented as separating the "available” fractions of phos­ phorus. Among these are 0.1 N HNO3 , 0,1 N HC1 and the widely used .002 N H2SO4 of Truog (27)* However, it Is believed by Bray (4) and others that this concept of avail- -2» ability is erroneous. Usually, a plant nutrient occurs in more than one form in the soil, with each form contribu­ ting to the nutrition of plants. The acid-soluble fraction is but one of the forms of the soil phosphorus. The most recent concept calls for the use of extractants designed to remove the various forms selectively or in combination* Bray and Kurtz (7) have recently presented analytical m eth­ ods for the determination of the acid-soluble, adsorbed and organic fractions of the soil phosphorus developed from stud­ ies on Illinois soils. The mere determination of the fractions is useless u n ­ til their amounts are calibrated by observation on the specific responses of crops to added phosphate on various soil types. The objective of this investigation was, therefore, to ana­ lyze, according to these methods, a number of Michigan soils showing high and low response to added phosphate, to attempt correlation of the amounts of the various fractions with crop responses and to study the relationships of the frac­ tions with various soil properties. It was hoped that the results might assist in evaluating further use of the m e ­ thods for determining soil management practices under Mich­ igan conditions. REVIEW OF LITERATURE Acid-Soluble Phosphorus In the literature, the term "forms" of soil phosphorus is used interchangeably between actual phosphorus compounds and the fractions soluble In a given extractant. The most commonly accepted fraction is the "easily acid-soluble" phos­ phorus of Truog (27), which is soluble in .002 U sulfuric ac­ id buffered to pH 3. Bray (5) states that this fraction usu­ ally represents a minor part of the total phosphorus and in Illinois soils Is generally present to the extent of 8 to 100 pounds per acre. It is more readily available to plants than the difficultly soluble fraction--that which is not sol­ uble to any Important extent in the acid solvents employed in the commonly used soilt ests--but it may vary In degree of availability from soil to soil because of differences in the chemical nature of forms included in this fraction. On Minne­ sota soils Rost and Pinckney (25) tested 112 check plot samples by Truog*s method. They found that plots carrying 25 lb. or less of this fraction responded to superphosphate in sixty-nine per cent of the cases and in seventy-three per cent of the cas­ es of those carrying between 26 and 50 lb. phosphorus. Using acid extractants of pH 2 and pH 5 on Maryland soils, Fisher and Thomas (18) divided the inorganic soil phosphorus Into (a) amorphous and finely divided crystalline phosphates of calcium, magnesium and manganese; of aluminum and Iron; oxides and that (b) amorphous phosphates (c) phosphorus adsorbed present in the form of upon hydrous *4 — apatite. By placing proper values upon the phosphorus con­ tained in each group it was found that analyses by this m e ­ thod placed 22 soils in practically the identical order of phosphorus requirements as that disclosed by pot tests. A number of Kentucky soils were extracted with 25 dif­ ferent acid, base and salt solutions by Weeks and Karraker (30) in an effort to compare the usefulness of the solutions In measuring the availability of phosphorus in the soil. was found that there was no best extractant. It They believed that for the practical purpose of estimating the phosphate needs of soils, experience gained with the use of any one of the extractants on a given soil is of more importance than the extractant itself. Fraps and Fudge (19) determined the solubility of phos­ phorus in 34 Texas soils of low basicity in 0.2 N nitric ac­ id, 0.75 N hydrochloric acid, .002 N sulfuric acid and 0.52 N acetic acid in 10% sodium acetate. They state that none of the extractants give quantitative estimates of the quantity of phosphoric acid which Is, or may become available to plants, though there may be a significant relation between the quan­ tity dissolved by the extractants and the quantity taken up by plants. The correlation coefficients for the relation b e ­ tween phosphoric acid removed by crops and that dissolved by solvents were much greater for the mineral acids, with the .002 N sulfuric acid giving the highest value. Comparison of the quantities of phosphoric acid dissolved by the weak sol­ vents with those of total phosphoric acid showed that the major part of the phosphorus in those soils is In the form of iron, aluminum, organic and adsorbed phosphates. Using a quick-test technique (1 gram of soil shaken with 0.7 N HC1 for 10 minutes), Olson (21) found good correla­ tions between amounts of phosphorus removed and crop response to added phosphate with corn, lespedeza and pimientos. ton did not give a comparable correlation. Cot­ It was conjec­ tured that the ability of the cotton to feed upon the adsorbed phosphorus caused this difference* Adsorbed Phosphorus Recently concepts of the available forms of phosphonus have been amplified and clarified by the general division of certain soil phosphates into the adsorbed fraction as well as the easily acid-soluble forms. As early as 1936 Truog (28) stated that below pH 6.5 and especially below pH 6.0 there Is very little calcium phosphate (acid-soluble) present unless recently supplied. Usually less than five per cent of the supply was In this form and he believed that phosphorus that may be measured as being readily available comes largely from basic iron phosphate and Is small in amount# Davis (10) found that much of the phosphorus retained at reactions below pH 6.5 is held differently when it is added as H3PO4 than it is when added as Ca(HgP04)2. bon dioxide solution was used as an extractant. Car­ Lower amounts were recovered vfcen the H3PO4 was used. It was believed by Dean (11) that phosphorus added to acid soils tends to accumulate in the alkali-soluble (ad­ sorbed) forms, while phosphorus added to neutral or cal­ careous soils tends to accumulate in acid-soluble but al­ kali-insoluble forms. Ammonium fluoride in neutral and acid solutions pro­ vided a means of fractionating the soil phosphorus as out­ lined by Bray and Dickman (6 ). Various fluoride extraction methods for phosphate were applied to a large number of soils in an attempt to measure the amount of adsorbed phos­ phates separately from the acid-soluble forms. They found that the addition of soluble phosphates to the soil increased only the adsorbed fractions. Rock phosphate increased the acid-soluble phosphate, but any conversion to adsorbed forms was found to occur only in acid soils, pH 4.8 to 5.0. Soils above pH 5.7 gave no Increase In the adsorbed forms with rock phosphate additions. Increasing the amount of adsorbed forms by addition of soluble phosphates rapidly Increased the usage of these forms by plants* Increasing the magni­ tude of the acid-soluble fraction produced no significant in­ creases In crop growth. Kurtz, DeTurk and Bray (20) found that nearly all of the phosphate added to samples of Illinois soils which was not recoverable in a water extraction was found in the ad­ sorbed fraction. Conversion of the added phosphate into acid- soluble (.002 N sulfuric acid) proceeded slowly. In parts per million this conversion was not great, but it made up fr o m 50 per cent of the total amount where additions were small to 5 per cent where additions were large. Believing that the phosphate extracted by the acid and 1 N ammonium fluoride could be utilized by plants they point out that the mm high recovery of the added phosphorus shows that little of the phosphate could be considered flfixed" in a sense that it would not he recoverable by plants* Results with both oats and cotton indicated to Coleman (9) that these plants can utilize large amounts of adsorbed phosphate that could not be removed by T r u o g fs dilute acid* This method did not remove all of the phosphate that was avail­ able to those crops* Bean and Rubins (13), and Bray and Kurtz (7) arrived at much the same conclusions concerning the effect of pH on the acid-soluble and adsorbed phosphorus* The former state that in acid soils most of the applied phosphorus occurs as ex­ changeable phosphorus whereas in sligfctly alkaline soils con­ taining a small amount of calcium carbonate the phosphorus occurs mostly as salts of divalent bases* The transition zone was found to be in the neighborhood of p H 6 * The latter workers found that in untreated soils below a p H of 6*0 the adsorbed forms are relatively more abundant than at higher pH values* Added soluble phosphates change into these forms, whereas acid-soluble fractions are finally gradually dissolved and also increase the adsorbed forms* Above pH 6 the trend was the opposite. Organic Phosphorus Pierre and Parker (23), experimenting on the availability of organic phosphorus to plants, found that the organic phos­ phorus concentration was five times that of the inorganic phos­ phorus in the displaced soil solutions of 20 soils* Plants would not absorb organic phosphorus from soil extracts of the *»8»» displaced solution while in the same experiment plants ab­ sorbed all the inorganic phosphorus* They point out, how­ ever, that organic phosphates may be made available to plants by biological agencies within the soil* The rapid­ ity with which this takes place is believed to be due largely to the nature of the organic matter* Rogers et al. (24) grew corn and tomato plants in the water extracts and displaced soil solution of Webster silt loam* The water soluble organic phosphorus was neither d e ­ composed by the root enzymes nor absorbed by the plants to a measurable extent. These results are in agreement with those of Pierre and Parker* In a survey of organic phosphorus content of western Oregon soils Bertramson and Stephenson (1) found that in old­ er soils the phosphorus tends to accumulate in the form of rather stable organic compounds* They believed that these old soils have become biologically inactive and the old r e ­ sistant residue of humus no longer decomposes readily or liberates appreciable quantities of phosphorus even though some of them were high in organic matter and organic phos­ phorus • This concept is in agreement with the findings of Dyer and Wrenshall (16,17)* They state that many soils having a high concentration of organic phosphorus are very deficient In available phosphorus, and that infertile, acid soils u s ­ ually contain a high proportion of this fraction. In England, Dean (12) observed large amounts of organic phosphorus with a close relation to the carbon content of s oils• Among dark-colored Iowa soils it was found by Pearson and Simonson (22) that the Edina series (pianosol) and the Carrington series (Iowan drift) contained a greater per centage of organic phosphorus than did the Marshall, which is formed on loess of a much more recent age* The two old­ er soils are acid in reaction, while the Marshall is approx­ imately neutral* Prom results with soil fungi Chang (8 ) concluded that as long as the phosphorus requirements of organisms exceed the amount derived by the organisms from organic compounds synthesis exceeds mineralization. When these organic phos­ phorus compounds provide more phosphorus than Is required for synthesis the excess is liberated as inorganic phosphor­ us • Bray and Kurtz (7) sum up the importance of organic phosphorus thus: "The organic forms of phosphorus are of importance In fertility because they are, in general, an indirect source of the soluble forms* as nitrates, posed* Phosphates, as well are produced when soil organic matter is decom­ After liberation, soil reactions sooner or later make the phosphates a part of the adsorbed and acid-soluble forms. Thus, they help counterbalance the effect of crop removal, and in highly organic soils a good level of the available forms is often maintained over a period of years despite crop removals* But It is the level of the available forms already present, not the amount liberated from the organic matter during the growing season, which appears to determine the fertility of the soil for thsfc season as far as phosphorus is concerned.” -11- PLAN OP STUDY In order to evaluate the correlations between crop response to added phosphate and the quantities of acidsoluble, adsorbed and organic phosphorus in Michigan soils a number of soils were analyzed for these fractions accord­ ing to the procedures developed by Bray and Kurtz (7). Sam­ ples for this study were obtained as follows: In the spring of 1946 the Soil Science Department laid out six 12 * x 3 6 T field plots at 105 locations in the lower peninsula of Michigan for the purpose of studying correla­ tions between crop response and rapid soil tests for potas­ sium, phosphorus and magnesium* Plots were established as shown below, in fields where legume hay or small grain was grown* E P2 O5 -K2 0 - B Check D p2 o5-k2 o K20 F P2 O5-K2 O - Mg B P 2P5 A G ; t For the laboratory determinations, 38 locations, repre­ senting 18 soil series, which showed high or low crop r e ­ sponse to added phosphorus were selected as follows: Alfalfa: Wheat and oats: 13 soils 12 soils showing high response showing low response 7 soils 6 soils showing high response showing low response Before fertilizer was broadcast, 18 auger borings to a six-inch depth were taken from each plot at each location* Soils from the C plots (potash) and D plots ( phosphorus and potash) were used in the analytical determinations* (See Fig* 1 for locations*) For the purpose of establishing possible relationships between the fractions of the soil phosphorus and other soil properties, pH, particle size distribution and organic mat­ ter content were also determined for each location* In order to test the reliability of the field plot yield data and to observe the effects of phosphorus appli­ cations under controlled conditions soils from ten of the locations where alfalfa was grown were studied in a green­ house experiment* These included five low-responding, and five high-responding soils • Samples were composited from sites adjacent to the perimeter of the plots near, or after the close of the growing season* It was believed that sam­ ples so collected would have a nutrient content and other properties quite similar to the soils In the plots before fertilizer was applied. - 13 - «nrf£W»Q))\ luce SCHOOLCR T DICKINSON c*tPP£WA MACKINAC DELTA CNARLE OTSEGO MONTMOR ALPENA ANTRIM KALKASKA CRAW FD Pig, l--Locations of plots studied M A N IS WEXFORD ALCONA OSCODA 9i GRD TRAY MISSAUKEE ROSCOM IOSCO OGEMAW 8f S3 MASON OSCEOLA LAK E OCEANA NEWAYGO MECOSTA GLADWIN ARENAC T9 CLARE ISABELLA MIDLAND (,X S3 40 MONTCALM MUSKE GRATIOT TUSCOLA SANILAC SAGINAW 33 KENT OTTAWA LAPEER 4? CLINTON IONIA 98 20 XX OAKLAND 4Z VANBUREN KALAMA CALHOUN ST.JOSEPH BRANCH 7 JACKSON WASHTENAW HILLSDALE LENAWEE MACOMB 73 LIVINGSTN EATON ALLEGAN CASS ST CLAIR // WAYNE MONROE 14EXPERIMENTAL PROCEDURE Laboratory Determinations Preparation of Soil Samples: At time of sampling, soils had been dried and passed through a 2 mm. sieve. Pre­ liminary trials of the analytical procedures showed that for accurate results finer samples were required. All samples were therefore passed through a 55-mesh sieve. 1-gram samples from the C and D plots of each location were used for the determinations. Duplicate Possible differences between phosphorus levels of the plots receiving no phosphorus and those on which applications were made could thereby be al­ lowed for in correlating responses. The following reagents were used in making the deter­ minations : Ammonium molybdate-hydrochloric a c i d : Dissolve 15 gm* of reagent grade ammonium molybdate in about 350 ml. of dis­ tilled water. ring. Add 350 ml. 10 N hydrochloric acid with Cool to room temperature and dilute to 1000 ml. stir­ with water. Stannous chloride: Stock solution is made by dissolving 10 gm. of reagent grade stannous chloride dihydrate in 25 ml. concentrated hydrochloric acid. The solution is kept in a dark bottle and should be prepared fresh every two months. Dilute reagent is made up by adding 1 ml. of the stock sol­ ution to l/3 liter of water. Approximately 0.5 N ammonium fluoride; 18.5 gm. solid ammon­ ium fluoride is dissolved in 1 liter of water and adjusted to pH 7. Approximately 0.8 M boric acid solution: 50 gm. reagent grade boric acid is dissolved in 1 liter of warm distilled water. 0.1 N hydrochloric a c i d ; 8.1 ml. concentrated hydrochloric acid is diluted to 1000 ml. with water. Solid ammonium flikoride 0.5 N hydrochloric acid Phosphorus-free hydrogen peroxide: 50% strength technical grade hydrogen peroxide is distilled under reduced pressure at a temperature not exceeding 60°C. Total Adsorbed Phosphorus: One gram of NH4 -saturated soil was placed in a 125 ml. Erlenmeyer flask together with 50 ml. 0.5 N ammonium fluoride and shaken for one hour in an endover-end shaker. After filtering the suspension in a glass or Buchner funnel a 10 ml. aliquot of the clear filtrate was transferred with a pipette into a 50 ml. graduate cylinder. 15 ml. of 0.8 M boric acid was added and the volume was brought to exactly 35 ml. with water. Ten ml. of ammonium molybaate- hydrochloric acid reagent was pipetted into the solution and mixed, followed immediately by 5 ml. of dilute stannous chlor­ ide solution* After mixing again and allowing 5 minutes for color to develop phosphorus concentration was determined in an Evelyn photoelectric colorimeter equipped with a 620 mu filter. Total Acid-soluble and Adsorbed Phosphorus: One gram of air-dried soil was shaken with 50 ml. 0.1 N hydrochloric acid for 30 minutes. One gram of solid ammonium fluoride was then added, making the solution approximately 0.5 N in -16 fluoride and the shaking was continued for one hour* After filtering the suspension and taking a 10 ml* aliquot the procedure outlined before for Adsorbed Phosphorus was fol­ lowed* After determining phosphorus concentration in the colorimeter the Total Acid-Soluble Phosphorus was calcu­ lated by difference* Organic Phosphorus: One gram of air-dried soil was weighed into a large test tube graduated at 50 ml. Phos­ phorus-free hydrogen peroxide, equivalent to approximately 15 ml# of 30 per cent strength and 10 ml* water were added and mixed thoroughly# The tube was then placed in a hot water bath for one hour# 15 ml* water, 10 ml* 0.5 N hydro­ chloric acid were added and the mixture was finally made up to 50 ml* with water. After shaking 30 minutes, one gram of ammonium fluoride was added and the suspension shaken for an additional hour# Following filtration, a 10 ml* ali­ quot was placed in a 250 ml# beaker, 15 ml. 0#8 M boric acid was added, and the mixture was evaporated to dryness on a hot plate. Ten ml. 0.1 N hydrochloric acid was placed in the beaker and the mixture was again evaporated. Residue was taken up with small portions of 0.1 N hydrochloric acid, transferred to a 50 ml. graduate cylinder, and phosphorus was determined according to the procedure described above for acid-soluble and adsorbed phosphates# The organic phosphorus is taken as the difference b e ­ tween the phosphorus removed by this procedure and that r e ­ moved from duplicate samples in the acid-soluble and adsorbed phosphate determination# -17 Organic matter content was determined by the wet com­ bustion method of Walkley and Black (29). The Hydrometer method of Bouyoucos (3) was used In m a k ­ ing the mechanical analyses. Hydrogen Ion concentration was determined with the glass electrode. Data on the above soil properties Is given In Tables 5 and 6 * Greenhouse E x p e r i m e n t : Soils f r o m the 10 locations were sieved through a iinch screen© The experiment was set up to provide fertil­ izer treatments similar to those received by the f i e l d plots. Since the feeding area of the roots was limited, fertilizer applications were double those applied to the field plots. Fertilizer salts were thoroughly mixed with the soils in 2 -gallon glazed pots with each treatment set up in trip­ licate as follows: A. Check. 50 lb. nitrogen per acre as ammonium sul­ fate to start seedlings. B. 300 lb. P2 O5 per acre as mono-calcium phosphate plus 50 lb. nitrogen as ammonium sulfate. C. 300 lb. K 2 O per acre as potassium chloride plus 50 lb. nitrogen as ammonium sulfate. P. 300 lb. P 2 O5 per acre as mono-calcium phosphate plus 500 lb. Kg O as potassium chloride with 50 lb. nitrogen per acre as ammonium sulfate. After mixing, the soil was moistened in excess of mois­ ture equivalent and maintained thus for one week previous to seeding. Alfalfa (Hardigan variety) was seeded In the pots on February 15th. The soils were maintained at close to -16- thelr moisture equivalent (as determined by the suction method of Bouyoucos (2) ) until six weeks after w h i c h time water was after seeding added as required. First cut­ ting was made May 4th and the secoi$ cutting June 10th. -19- DISCUSSION OF RESULTS Analytical D e t erminations: Soils were designated "high-responding" or "low-respond­ ing" according to whether increase in yield of* alfalfa or small grain was above or below the mean increase for the r e ­ spective crops. Mean yields, the amounts of total adsorbed, acid-soluble, total adsorbed plus acid-soluble, and organic phosphorus, together with the sum of the three fractions is given in Tables 1-4. This data is presented graphically in Figs. 2 and 3. Soils on Which Alfalfa was Grown: The total adsorbed fraction ranged f rom 21 to 90 ppm., with an average of 41 ppm. on the low-responding soils. On soils showing high response the range was f rom 11 to 35 ppm. and averaged only 22 ppm., or 19 ppm. less than the low-responding soils. Acid-soluble phosphorus ranged from 57 to 126 ppm., averaging 78 ppm., in the low-responding soils, and from 15 to 101 ppm. with an average of 49 ppm. in the high-respond­ ing soils. It is of interest to note that while the soil (#52) with 101 ppm. acid-soluble phosphorus was highest for this fraction it was extremely low (13 ppm.) in the adsorbed fraction and showed a significant response to added phos­ phorus . Contents of the total adsorbed plus acid-soluble phos­ phorus varied In the low-responding soils f r o m 81 to 182 ppm., averaging 119 ppm., ’while in soils of high response the var­ iation was between 29 and 116 ppm. with a mean of only 71 ppm. With organic phosphorus the trend was reversed. In the -£U- CM • H1 H< rH CM * CM CO CM • O » CM 05 CrH CM • CM H4 rH CM rH O •H • s © p bG p P O CM • as Hi o • IP H* rH O • CO 05 o IP • CM • CO CM • CM CO CM * © • d p 6 •H p P O rH P <4 O Eh o * lO 05 rH o « Ip CO O • O rH 00 O • IP 05 CM • CM CM O • O CO IP • o O• o CO CM • IP • CM rH £> Hi • 05 iH i— 1 CO • • o £rH P P o • o to ao o • rH • • 00 CO CM • CO CO s © o I—t •H • O O rH J2! PH CM * rH 05 ip o 00 CM rH E> CO rH CO rH CO rH 05 ao O O o o 05 05 « • Hi J> • IP IP O Hi * IP CM rH rH • CM rH rH • Ip CM rH CM • CO LO CM • CO £> O » tp O IP • CM to O • o CM • o ip 05 CO «H Hi to rH IP * o CM • I t> CO • I o 1 o o • o • ip Hi C- o * B aS O d S3 © © fi © P ^5 EH -P c* • O • O oo • CO • • rH IP CO o Hi CO « rH £> C- id O o CM CM • rH B © O rH O 00 o • IP to Hi O • CO • • « rH rH 05 rH rH ip IP • rH * • CM 00 £> IP *> CM o CM acre • S S3 •h d rH • © <4 O *H S3 t>3EH M rH •H CM • CO CM CM • O * 00 • rH CO • LQ CO * i— 1 Hi tons d © rH P © p -P o O © eh d <4 CM rH rH rH 00 rH © P P rH e rH |Q © • B rH CO CM • CO co per d H P® d_ H© OS P H X -poop O © <4 rH EH d O <4 H-CQ « CO 05 • B © •H S rH rH •H © -P S3 o «M © rH rH © PQ O CM d B © o B © o B © o rH •H Sh © > O S3 H B © •H rH CM 3 © B © o rH r>i © rH O B © O i— 1 © o o •H S © •H s •H 6 © nH CM CM 00 CM CO CO * £3 B o l> O • 1 O • 1 © bO © B © B E © U i— i o © rH o <4 d >> d © S3 •H rH £ © rH O O © w CM CM Hi * « 1 rH S3 © rH CO = S3 © .d -P 09 © © i—1 to S3 •H £ o .a © © rH •H O © Cm O © © Ph -P O S3 © Pi -P © S3 O o Xi o Pi «d 2 © S-ld O TO «s3 © Pi © x3 O 4-> .Cl •H Pn & o «H • S3 © P bC P P O CO o rH {>» d d © © 3 © © S3 *rl rH * © as w IP CO * B © o 1— f u © 4-5 © © S3 o 00 4* S3 3 © S3 © -P © rH rH © P © © © © M IP CO 00 O # © > rH ❖ i ^Fertilized in 1944 or 1945 Mean increase in alfalfa yield d rH © •H t>» © U rH aS CM . rH 00 rH CM o rH Si 0 I& P •H * a ft to ft 0 0 0 0 03 * rH 03 » CO O • id id 03 • rH to ■<** • O . . CM 03 00 00 o rH CM o CM * rH 00 o O c. id • 03 CO CM • CO CM rH Ld • CM CO CM 03 . CM « CM * 01 « CO Id cO rH CO . CO CO CM . id id . L> C- rH rH c- CM Id 00 to to rH o • rH o r . &to to • • id rH • 03 . 03 03 rH • CO rH 117.2 0 • » 188.4 -21- P 0 0 0 8 rj s •P 44 O 03 <4 iH EH *0 O ft . {> 03 o • CM 03 «G -*-0Q © rH •cffO •H St 8ft O rH ft . . CM « • £> CO CO rH 00 CO oo rH O 00 tO £> . CM . o Ld • CO Ld CO o • o £> CO • CO H rH O rH to . CM rH O • o O . CM . td rH • 00 CO O . c- CO Id CM • rH CM 00 CM . O • CM . o rH co CM o rH rH CO id CM CM Ld CO CO id o • rH 00 CO 00 rH . 03 O • rH rH CO CO id » i—1 8 8 to acre CO P rH rH 1> bC d bD P eh d CM CM 00 00 CO . CM rH •rl O 0 p •H d . rH O •H EH S3 N . orH• o 00 . o 03 t-H O . rH rH O .ft O ft at P o 0 P *5 o d Eh .ft 0 d p at 03 ft ft 0 .ft O 44 .ft »H (P & o CO O o SH O P o o o CM LO cm o 44 ♦ o o & rH > CM 8 at o rH O Eh 03 id . rH CM ■M* • I— 1 rH » CM . rH # 8 O PQ is; CO o to to to CO 03 P m 0 ft § CQ d at 0 O O {>» d P a> 0 8 0 O 0 rH o a o 44 09 . 8 0 t>3 d ft 0 0 at E> at O P ,P JO . CO rH 8 0 ft 44 0 H • rH rH <4 0 0 CM 0 S3 tO & 0 8 to o 8 0 to o> 0 CD CO O 44 0 o o p £> * O S^s © 0 £> jJ c P Pd 0 rH 03 0 P o * = .76 m d P 8 O ft 03 ft Average ^-Fertilized in 1944 or 1945 Mean increase in alfalfa yield rH 0 44 O 22.0 0 «*H £ O .ft per 8 & *H 4** O O ' 3S ft tons P 49.1 0 0 rH ^ rCt fH 71.5 Tt P •H - small m rH9 <© O in H o •H . g S 3 ft ’+■ 'o£ Pi oB d d o •H * E> • O oo o 02 02 • rH 00 rH . t—1 to rH o• o 00 rH O H* o OJ to o to OJ OJ • LO • O oo • 9 to • OJ H* rH o rH O to • o d than mean m less c• to IP • O 00 IP o • O' . ip G> IP C- Hi OJ to Ip • t- £> . 02 « to to • 02 • IP Hi ca s i 1 • to bu* per to • < 03 0 P O si ft CO o a •h d rH • © <4 O «H Ip £ M PQ iH 1 rH 1 • tO •H 1 •E to d 0> H« *H © d d aj B © o Si •p •H £ P •H > cq B © o rH X o ft -p • o o rH }2* ft © rH EH rH «H O d S o- d P © © P'S d © p B © o rH d P O ft rH •H PQ o H< t> H* to * sje S © 03 f Hc* © o 8 © d fi B © o i— i © O rH H< H* GO © p © > o> P <*1 •H o ft •H © © c © ft © s S=5 -P «H © -P 00 rH LP to .£3 © rH •H X B © * * 9.3 O • rH yield 02 oat o • o CP in oa Pi d o • o IP UP * d © N © P \ * increase d © P • P s O pi Mean rH © -P O 6-t acre CQ 3. Phosphorus HI rH O showing of soils contents E> rH r* 03 m rH 05 CO c- o 05 r> o 03 m GO o o co in 03 03 d 03 E> CO 05 LO o • 03 GO • CO 03 • rH • m CO LO CT> o in m 03 03 CO m 03 GO c- O* to * in o p o o d a ft O Off 00 to in • 03 CO 03 • in CO in » * o £> o rH 03 m rH 03 03 to 03 03 o m fr­ acr e <0 ee bu. per 0 H.D'dH cff d H P <4 d •H rH 0 O *H 2 rH * lO d H PQ 05 LO £> £> M a off o t —I c cff m H O Pn o> off 0 ,_Q Cff Off O CO * a cff o o Cff d 0 03 • CO rH rH a cff O TJ d cff Off r*> O Off or 1945 yield » 9.3 Phosphorus contents of soils showing grains with added phosphorus* greater than mean Increase in yield of P «H * 03 d a off o, ft u CO off Table 23 - 0 to cff d 0 a cff O £ d © p p P Cff 0 § a cff O > < Cff 0 d 0 •/—I d d 0 is W CQ p 05 CO CO .d fr- CO GO 05 a 0 * * 00 ❖ ■st Fertilized in 1944 Mean increase in oat smaH - low-responding soils there was w i t h a mean of 71 ppm. a range from 0 to 145 ppm. The high-responding soils, however, had an extreme range from 17 to 337 ppm. and a mean of 117 ppm. Three soils with a large organic phosphorus content were responsible for the high average. Where amounts of total adsorbed fius acid-soluble plus organic phosphorus were compared the range in low-responding soils was from 135 to 327 p p m . ; that in the high-responding soils from 78 to 338 ppm., while the averages were nearly equal--189 ppm. in the former and 188 ppm. in the latter group. Soils on Which Small Grain was drown; Differences in the contents of the various fractions of the soil phosphorus were not so marked between high- and low-responding soils as they were on the alfalfa plots. In low-responding soils the total adsorbed phosphorus varied from 20 to 61 ppm., averag­ ing 39 ppm., while on high-responding soils the range was from 26 to 78 ppm. and averaged 46 ppm. Acid-soluble phosphorus ranged from 53 to 123 ppm., with an average of 87 ppm., in the low-responding soils. high-responding soils In the this fraction showed a range from 38 to 112 ppm., ufoile averaging 75 ppm. Soils showing low response indicated a variation of from 76 to 184 ppm. total adsorbed phosphorus plus acidsoluble phosphorus while averaging 127 ppm. High-responding soils were but little different, with a range from 65 to 190 ppm. and an average content of 122 ppm* The trend In organic phosphorus was opposite to that - 25 - found in the soils on # iich alfalfa was grown* Those soils showing the greatest increases in yield contained the highest average amount of organic phosphorus* with small grain* Such was not the case In soils showing low response the range was f rom 8 to 181 ppm* with an average of 102 ppm., but in high-responding soils with a range from 0 to 78 ppm. the average was only 53 ppm. These differences in organic phosphorus caused the sum of the fractions in the low-responding soils to vary from 146 to 365 ppm* and to average 230 ppm. while the high-re­ sponding soils showed a range from 101 to 221 ppm. and an average of 175 ppm. Correlation with Crop Response: A l f a l f a : The relationship between the response of alfal­ fa to added phosphate and the contents of the subject fractions of soil phosphorus Pigs. 4, or combinations 5, 6 , 7, and 8 . thereof are shown in Correlation coefficients were de­ termined and significance established according to F i s h e r 1s Tables. For the number of determinations in this study a correlation coefficient of -.40 was necessary for a relation­ ship to be of significnee. The relation between the total adsorbed phosphorus and response of alfalfa to added phosphorus produced a correla­ tion coefficient of -.47, which is barely significant (Fig. 4). W i t h the acid-soluble fraction there was no significant rela­ tionship (ryx s -.41) (Fig. 5). This low correlation may -2 6 - I % 0 1 <0 rH •iH O tO Vi \5 * k % j..... . 1VJ 0 : C Q P o 6 X ft LL■•- .". m 1' <£* bo * 0 3 1 VN ; b0 i- 3 p O i ^W * r•■ i.-.. • to ftl dto ■r i ft 0 i— f r-i C O *H O P w o ; p 0 0 -p -0 0 £ 43 -p f t \■'“ ■ © 0 rQ O ...... ; 43 0 ft < — 1 0 *d \> 0 0 r& j'^7'- iH TZi 0 W ? ° U -P o 43 0 p.. 0 0 0 o o 3 £1 f t ft 0 0 ft p o . ---,V ~ 4 ["'' ' £Z * 5 § hI 4) 3 ■" O' O ff i0 O «£ <£ d Zt '■ !• fl O W •H P cd S $ *s £r H bo P *H & g g O O 43 O 0 bO •H P4 5 f ' ' XC I Vo Fig, 5— Comparison of phosphorus levels between soils shotting high response and soils showing low response to added phosphate on oats. -27- -V K n 0 v V SV. " ft 1 * S ft f 5 V> % V \J s X ft * K s ft S * .8 vr> r n - i o t f ■Hsu/e/ 'X -28- bo explained through reference to Fig* 2. There it will be seen that five of the soils with greater than mean yield in­ crease had an acid-soluble phosphorus content of more than 50 ppm#, or equal to the acid-soluble phosphorus content of eight of the soils showing less than mean increase in yield* Of the five soil phosphorus fractions of fractions, or combinations the adsorbed plus acid-soluble gave the most significant correlation (ryx B -*54) Fig* 6 )* This is in harmony with the observations of Bray and Dickman (6 ), who pointed out that crop growth and response for the soils and crops of Illinois are generally correlated with the r ela­ tive amounts of all fractions present, not with the amount of any one fraction* Bray# has also stated that the rapid test which removes a portion of both fractions gives a b e t ­ ter correlation with crop response than laboratory methods specific for the total amount of each form* There was apparently no relationship whatsoever between the organic phosphorus and response of alfalfa (Fig# 7)# this case the correlation coefficient was +.33# In This valuef if taken literally, would indicate that alfalfa responded more to added phosphate on soils of high organic phosphorus content than on soils low in this fraction* This Is, of course, an impossible situation, but it does serve to point out that the organic phosphorus is totally unavailable to alfalfa. The correlation coefficient of -#02, calculated for the relation between the sum of the fractions #Private communication. and yield increase -29- > 434- of alfalfa (Pig. 8 ) emphasized the futility of attempting to link the organic phosphorus in any way with crop response In a siggle season. It will he observed here that the sig­ nificant correlation found w i t h the adsorbed plus acid-sol­ uble fractions correlation (ryx *= -.54) is completely nullified when is attempted between those fractions plus organ­ ic phosphorus and response of alfalfa (ryx =* -.0 2 ), Oats: Since analytical data was available only for nine soils on whi c h oats were grown, It was decided that with so limited a number of items statistical treatment was of no value. F rom a comparison of Tables 3 and 4, however, It will be seen that although the high- and low-responding soils con­ tain similar average amounts of adsorbed and acid-soluble phosphorus there is a sharp difference between the average amounts of organic phosphorus present. The high-responding soils contained half as much of this fraction (53 ppm.) as the low-responding soils (102 ppm.). This relationship appears to be in direct opposition with correlations in alfal fa yield increases where the organic phosphorus had no effect on response. Since none of the high-responding soils are regarded as inherently fertile, mfoile three of the four lowresponding soils have a medium to high natural productivity, an explanation may be afforded by the observations of Wynd and Noggle (32). They indicate that the growth of oats on some Kansas soils was more affected by the amount of replaceable bases In soils than by the content of adsorbed, acid-soluble or adsorbed plus acid-soluble phosphorus. Although no data on content of exchangeable bases in the subject soils was com­ piled, the fact that those showing the lowest phosphate r e ­ sponse were of good fertility indicates a similar relation­ ship. The average organic matter content was considerably above that of the high-responding soils, also. (Tables 5 and 6 ). Relationships between Soil Phosphorus and Soil Properties: Past M a n a g e m e n t : A study of fertilizer applications by farmers on all locations previous to establishment of the plots was made. There was no uniformity either as to the amounts or kinds of fertilizers used, but the over-all study revealed that soils to which phosphate had been added in 1944 or 1945 averaged 41 ppm. total adsorbed phosphorus, while unfertilized soils averaged only 27 ppm* No such r e ­ lationship was evident with the acid-soluble fraction.« Fer­ tilized soils averaged 72 ppm. to 67 ppm. in the unfertilized soils. Nine of the 22 high-responding soils and 11 of the 16 low-responding soils were fertilized in 1944 or 1945 (Tables 1-4)• Since superphosphate rather than rock phosphate Is the prevailing amendment in Michigan the observations of Bray and Dickman are born out (6 ). Under Illinois conditions they found that in acid soils added soluble phosphates rapidly change into the adsorbed forms, whereas acid-soluble portions are gradual­ ly dissolved and very slowly Increase the adsorbed forms* Influence of Soil A c i d i t y : Of the soils above pH 6.6 It was found that 11 showed little or no response to added phosphate while only 2 showed a high response. soils (Tables 5 and 6 ) were definitely acid in Twenty-nine reaction Table 5. Plot Ho. Soil properties. Per cent organic matter, p H and per cent sand, silt and clay of soils showing high response to added phosphate. Soil Type Organic Matter pH ^ Sand Silt Clay % ; % 24 Conover loam 1.48 6.7 56 34 10 25 Conover loam 1.43 6.9 58 26 16 27 Coloma sand 0.85 6.7 87 9 4 33 Oshtemo loamy sand 3.84 6.6 80 10 10 38 Brooks ton sandy clay loam 2.08 6.6 60 22 12 49 P o x sandy loam 0.83 5.6 72 22 6 50 Miami clay loam 1.47 6.2 50 26 24 52 Miami sandy clay laam 1.43 6.6 64 24 12 54 Hapanee clay loam 1.89 6.5 64 22 14 60 Isabella sandy loam 1.40 6.7 74 20 6 63 Ogemaw sandy loam 3.00 7.2 74 14 12 69 Miami clay loam 2.10 6.0 38 20 42 73 Brookston sandy clay loam OJ OJ o • 6.7 52 28 20 75 Napanee clay 1.79 5.9 50 12 38 79 Arenac sand 0.96 6.1 92 5 3 83 Hester loam 0.83' 6.0 48 52 10 88 Emmett sandy loam 1.95 6.8 58 26 16 93 Selkirk loam 1.97 6.7 52 28 20 96 Conover loam 1.87 6.5 42 40 18 98 Berrien loamy sand 1.30 5.3 88 12 7 Table 6 . Plot Ho. Soil properties (cont'd.). Per cent organic m a t ­ ter, p H and per cent sand, silt and clay of soils showing low response to added phosphate. Soil Type 7 F o x loan Organic Matter pH % Sand -o f" ' /o ' Silt -Olay —a ¥ /0 1.43 6.9 48 30 22 1.32 6.9 52 25 33 20 Beliefontaine sandy : l oam 0.83 6.8 70 18 12 21 Miami loam 1.82 5.3 48 28 24 22 Conover loam 1.56 6.6 58 26 16 28 Miami loam 1.74 7.8 58 24 18 34 Brady sandy loam 1.50 7.0 74 14 12 36 Miami clay loam 1*30 5.8 44 34 22 42 Coloma sand d.54 6.1 89 7 4 47 Gilford loamy sand 2.02 7.3 82 11 7 48 F o x loamy sand 2.10 5.9 82 13 5 51 Mi ami sandy 1 oam 1.43 6.0 76 18 6 53 Napanee clay loam 2.13 6.6 52 30 18 62 Kawkawlin loamy sand 1.74 7.3 82 12 6 65 Kawkawlin sandy loam 2.86 7.7 68 25 17 80 Hester loam 1.76 7.0 64 20 16 85 Nester sandy loam 1.82 6.7 78 12 10 1 0 8 Isabella sandy loam 1.27 6.9 75 20 5 11 Miami clay loam (below p H 6.8 )* Six of these were low in their response while 23 gave significant Increases in yield. This com­ parison is in accord with the recommendations of Scarseth (26) and numerous other workers, who show that maximum phosphate availability occurs just below neutrality. The relation between p H and per cent total adsorbed phosphorus In the total adsorbed plus acid—soluble com­ b i n a t i o n is shown i n Fig. 9. There is a wide scattering of points and no trend is evident. Dean and Rubins (13) stated that the adsorbed fraction Increased with acidity, especially below p H 6 , while in slightly alkaline soils the acid-soluble fraction predominated. zone, The transition they believed, began in the neighborhood of pH 6 . Since the majority of the soils studied fell into this transition zone between p H 6 and pH 7 the lack of definite relationship may be explained. Influence of Organic M a t t e r : As is shown in Fig. 30, the quantity of organic phosphorus generally increased with organic mat t e r content. This might well be expected, but, as Dickman and DeTurk (14) point out, the amounts are not proportional. of the soils The organic matter contents of the majority studied varied in the narrow range between 1 and 2 per cent, as found by the wet combustion method, and w i t h i n such limits a clean-cut trend in the relation­ ship is difficult to follow* Wynd and Noggle (31) found a positive relationship between organic matter content and adsorbed phosphorus in -39- X) © Xi U JO & © phosphorus o £ oo o T) © © *CJ as H © rH P © O P p O P o *rJ © £ • bO P • £ £ o o JC3 *H OrP TO O O © .£ £ CX 0 © © © a9q.q.em ojmSao ^u©o aa Davis, P.L. 1946, Retention of phosphates by soils. IV: Solu­ bility of phosphates retained by virgin Ham­ mond very fine sandy loam treated w i t h cal­ cium hydroxide and phosphoric acid. Soil Sci. 61:179-190. (11) Dean, L.A. 1937. Distribution of the forms of soil phosphorus. Soil Sci. Soc. Amer. Proc. 2:223-228. (12 ) ________ 1938. An attempted fractionation of the soil phos­ phorus. Jour. Agr. Sci. 28:234-246. (13) ___________ and Rubins, E.J. 1947. Anion exchange In soils. I: Exchangeable phos­ phorus and the anion exchange capacity. Soil Sci. 63:377-387. (14) DIckman, 1938. S.R. and DeTurk, E.E. A method for the determination of the organic phosphorus of soils. Soil Sci. 45:29-39. (15) Dyer, Bernard 1894. On the analytical determination of probably available "mineral11 plant food In soils. Jour. Chem. Soc., (London) Trans. 65:115-167. (16) Dyer, W.J. and Wrenshall, C.L. 1941. Organic phosphorus In soils. I: Extraction and separation of organic phosphorus compounds from soil. Soil Sci. 51:159-170. (17) __________________________________ 1941. Organic phosphorus in soils. Ill: The decom­ position of some organic phosphorus compounds in soil cultures. Soil Sci. 51:323-329. (18) Fisher, ^.A. and Thomas, R.R. 1935. The determination of the forms of inorganic phosphorus in soils. Jour. Amer. Soc. Agron. 27:863-873. (19) Fraps, G-.S. and Fudge, J.F. 1945. The nature of the phosphates dissolved b y var­ ious soil extractants. Jour Amer. Soc. Agron. 37:532-541. (20) Kurtz, L., DeTurk, E.E. and Bray, R.H. 1946. Phosphate adsorption b y Illinois soils. Sci. 61:111-124. Soil (21) Olson, L.O. 1945. Factors affecting the relation between labor­ atory tests for soil phosphorus and crop r e ­ sponse to added phosphate. Soil Sci* Soc. Amer. Proc. 10:443-445. (22) Pearson, R.W. and Simonson, R.W. 1939. Organic phosphorus in seven Iowa profiles: DIstirbution and amounts as compared to organic carbon and nitrogen. Soil Sci. Soc. Amer. Proc. 4:162-167. (23) Pierre, W.H. and Parker, F.W. 1927* Soil phosphorus studies: II. The concentration of organic and inorganic phosphorus i n the soil solution and soil extracts and the availability of the organic phosphorus to plants. Soil Sci. 24:119-128. (24) Rogers, H.T., Pearson, R.W. and Pierre, W.H. 1940, Absorption of organic phosphorus by corn and tomato plants and the mineralizing action of exo-enzyme systems of growing roots. Soil Sci. Soc. Amer. Proc. 5:135-143. (25) Rost, C.O. and Pinckney, R.M. 1932. Colorimetric methods for the determination of readily available phosphate in soils. Jour. Amer. Soc. Agron. 24:377-395. (26) Scarseth, G.D. 1941* The Hills and Valleys of Phosphate Fixation. Better Crops with Plant Food Magazine. (27) Truog, E. 1930. The determination of readily available phos ­ phorus in soils. Jour. Amer. Soc. Agron. 27: 874-882. (28) 1936. Availability of essential soil elements— a relative matter. Soil Sci. Soc. Amer. Proc. 1:135-142. (29) Walkley, A. 1947. A critical examination of a rapid method for determining g organic carbon in soils— effect of variations In digestion conditions, and of inorganic soil constituents. Soil Sci* 63: 2 5 1 —264. (30) Weeks, M.E. and Karraker, P.E. 1941, A comparison of various extracting solutions for measuring the availability of phosphorus in soils of known fertilizer treatment and crop response. Soil Sci. 51:41-54. -56 (31) Wynd, F.L* and Noggle, G.R. 1943# Associations b e t w e e n phosphorus fractions and other chemical components of the soil* Soil Sci# 56:383-392. (32) _______________________________ 1946* Relationships between fractions of phosphorus in the soil and growth of cereals* Food Res* 11:210-215*