Ml lHHlHlJlH‘IWl W I 1 ,IJIMIW IIWIH A STUDY OF THE SCLUBILITY AND irituzA-"L‘ION or PHOSFHORLTS m“ SOME ALKALINE CALCAREOUS ONTARIO SOILS Thesis for the Degree of M. S. MICHIGAN STATE COLLEGE Lyman J. Chapman 1938 “if! " "3 it,“ 2a '1 .,. ’3 .n ‘1’ . . < h- a \. aba- .V' .i K!- £131, <73- .. in.“ Hal-IaPVFVry A STUDY OF THE SOLUBILITY AND FIXATION OF PHOSPHORUS IN SOME ALKALINE CALCAREOUS ONTARIO SOILS BY Lyman J. Qgggfian A THESIS PRESENTED TO THE FACULTY OF MICHIGAN STATE COLLEGE OF AGRICULTURE AND APPLIED SCIENCE IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE. East Lansing 1938 THESlS ACKNOELEDGEEHT The writer takes this opportunity to err press his appreciation to Dr. C. H. Spurway and Dr. C. E. miller for helpful suggestions during the progress of this study and the preparation of the manuscript. 115902 (1) INTRODUCTION In Southern Ontario there are extensive areas of soils which.are alkaline in reaction and contain free carbonates in the plowed layer (5). Some of these soils are quite deficient in available phosphorus; plants growing on them quite commonly exhibit phosphorus deficiency symptoms and the application of phosphatic fertilizer markedly stimulates plant growth. However, other soils of the group appear to be relatively well supplied with this element. Practically all of the previous work on soil phosphorus in Ontario and the adjoining areas has dealt with the conditions found in acid soils and the methods employed are not well adapted to use with calcareous soils. Most of the studies on available phosphorus in alkaline, calcareous soil, particularly those of the Arizona school of workers, have been carried out in the more arid reagons, and their findings cannot be applied locally with- out first obtaining some basic information about the local cal- careous soils. The purpose of the present study is, therefore, two-fold: first, to develop a properly adapted laboratory method for determining available phosphorus in calcareous soil, and second, to obtain information about the solubility and fixation of phosphorus in local calcareous soils. REVIEW OF PREVIOUS KCRK The nature and composition of phosphates in soil and the manner of their solution has been the subject of numerous in- vestigations, and there is a voluminous literature on the question. It is recognized that organic phosphates may be of (2) considerable significance in plant nutrition, but they have been disregarded in this work. Furthermore, the work which relates to acid soils will not be dealt with in this dis- cussion. In connection with alkaline soils, Buehrer (3) has written a very thorough treatise on the physico-chemical nature of soil phosphates, and the publications of his co- workers, McGeorge and Breazeale, are equally important. These latter workers (19) observed a striking similarity between the reactions of rock phosphate and natural phosphates in calcareous soils. Coupled with the fact that X-ray diffraction photographs of powder samples of rock phosphate showed it to be of similar composition to bone, which, according to Hendricks and others (10), consists of a carbonate-apatite compound, this led them to conclude that the phosphates were present in calcareous soils chiefly as the carbonato-apatite, [C83(P04)3]3.03003. However, the proportion of calcium carbonate in the complex is often less than the 1:3 ratio shown by the above formula, but they think that pure tricalcium phosphate never exists in soils. Iron and aluminum phosphates are undoubtedly of minor importance. The carbonato-apatite is only slightly soluble, but it reacts readily with carbonic acid. The manner of solution of phosphorus from the above form may be illustrated by the equation, Ca3I'PO4)2 + 4112003 : CaH4(PO4)3 + 203(HCO3)3 . However, calcium carbonate has an important influence on the equilibrium since carbonic acid reacts with it to form soluble bicarbonates. In accordance with the laws of mass action and chemical equilibrium, an excess of calcium-ion will force the (3) reaction towards the left depressing the solution of phosphorus, and vice-versa.- It is evident from these re- actions that carbon dioxide is assigned a major role in the process. Thomas (26) has reviewed the literature on carbon diox- ide in relation to plant feeding very fully and his paper appends a splendid list of references. There is a consider- able controversy in the literature over the influence of acids other than carbonic in the solution of minerals in soil. It is fairly certain that nitric acid sometimes functions in this regard, but the part played by various organic acids is of uncertain magnitude. The concensus of Opinion is that carbon dioxide, or, in effect, carbonic acid is the most important factor in the solution of phosphates in soil. Various authors (24) (19) (23) provide evidence of the benefit of carbon di- oxide in liberating this element from calcareous soil, and McGeorge and Breazeale (19) stress the value of a good supply of organic matter, good drainage and aeration in relation to the production of carbon dioxide- With regard to the influence of pH, McGeorge and Brea- zeale (17) conclude from their experiments that pH 8.0-8.5 is the reaction zone where minimum solubility of phosphates occurs, and they point to pH 8.8 as the optimum for absorption of phos- phorus by plants. A high concentration of salts in solution depresses the solubility of phOSphates. This antagonistic effect is due to the calcium-ion; additions of potassium or sodium increase the calcium-ion concentration through base exchange. (4) Truog's theory (28) that plants which are heavy feeders on calcium are also efficient in absorption of phosphates is of interest here. He reasons that weak feeders on calcium allow the accumulation of calcium in the spil solution so that the solubility of phosphate is depressed in the manner previ- ously outlined. With plants which are heavy calcium feeders, the reverse is true. Truog finds ample eXperimental evidence to support his theory. Having thus briefly dealt with the nature of mineral phos- phates in alkaline calcareous soil and the manner in which they become available to plants, there remains for discussion the suitability of the various chemical methods of testing these soils for available phosphorus. It is well known that extract- ions with the more strongly acid solutions are apt to show relatively high results with alkaline soil even when deficient in this element. For this reason some less vigorous extracting solution must be employed. McGeorge and Breazeale (17) con- cluded from their studies that available phosphorus in cal- careous soil is best determined in water extracts. Electro- dialysis has been used (18) (9) but apparently the results are only satisfactory with some soils. The use of an alkaline solution of potassium carbonate as proposed by Das (7) is preferred by Hockensmith et a1 (15) while, on the other hand, various weakly ionized acids have been employed. In a few instances (20) (30) (19) carbon dioxide saturated water has been used as an extractant. In view of the action of carbonated water in the soil it is generally conceded that such a solution should be the ideal (5) extractant. But carbonic acid solutions of a definite com- position are difficult to maintain. ?uri (20) describes an extraction tube designed to allow carbon dioxide gas to bubble through a soil suspension, in this way maintaining a saturated solution and at the same time keeping it agitated. Wilcox (30) used a similar method to extract available boron from calcareous orchard soils. McGeorge and Breazeale (19) found, with phosphorus deficient soils, that extraction with a saturated solution of carbon dioxide gave lower phosphorus levels than when pure water was used, and they attributed this to the increased calcium which is brought into solution by the carbonic acid. From soils which were better supplied with available phosphorus, the carbonated water extracted more than did pure water. Schloesing (23) found carbonated water simi- larly effective with calcareous soil. Hibbard (11) recommends the use of ordinary distilled water containing some carbon dioxide in spite of minor fluctuations in composition. He prefers it to CO2-free water because it is easier to obtain and is more like the soil solution than Pure water; further- more, it has the effect of flocculating the soil so that a clear filtrate is more easily obtained. Partially saturated solutions of carbon dioxide in water have seldom been employed, due, no doubt, to the difficulties of standardizing and main- taining a reasonably constant composition. In the majority of chemical methods equilibrium extracts are made, but Hibbard (11) and Russel (22) stress the advan- tage of continuous leaching or successive extractions over equilibrium methods; by continuous leaching, or successive (6) extractions of the same sample, not only the original amount present in the soil but also the rate at which phosphorus comes into solution is determined. This latter characteristic is particularly important in calcareous soil since here the specific concentration in the soil solution is usually very low and inability to maintain that level would surely result in acute deficiency. Hibbard (12) (11) has described a per- colation apparatus for the continuous leaching of soil at a constant slow rate. The chief limitation of leaching methods is discussed by McGeorge (17). There are frequently variat- ions in duplicate results, which he attributes to "channelling", the solution having a tendency to follow Open spaces rather than to move uniformly down through the soil mass in the ex- traction tube. After due consideration of the foregoing experiences, an effort was made to devise a method for the extraction of phos- phorus from soil by continuous slow leaching with carbonated water. To attain this end Hibbard's percolation apparatus (12) was slightly modified with a view to keeping the carbon dioxide solution at a definite composition during the period of extraction. EXPERIMENTAL. The laboratory work carried out in this study may be divided into two parts; (1) the fundamental work on the meth- od of extraction, and (2) the actual study of the soils with respect to solubility and fixation of phosphorus. In the first part, ability to maintain the concentration of carbonic acid ('7) within certain limits during a 24-hour period, and the effect of various concentrations of the above solution on the amount of phosphorus extracted were primarily considered. The re- lationship between calcium and phosphorus in the extract, and the pH of the same was also studied. Having adapted a method, 22 samples from the alkaline soil areas of South-Central Ontario were tested, and the relative capacity of these soils to fix phosphates was determined. In this connection, a field test on the residual effect of superphosphate is described. PART I. The Estimation of Available Phosphorus in Calcareous Soil by Continuous Extraction with Carbonated Water. The Appargtus. The apparatus employed in this study was essentially the same as that described by Hibbard (12). With this equipment small samples of soil mixed with appropriate amounts of washed sand are leached at a constant rate, the siphons being adjusted to deliver about 1 cc. per minute. However, certain modifi- cations were made in order to prevent the loss of carbon diox- ide from the system. First, the flasks which form part of the Mariette bottle arrangement are open to the air only through a length of small glass tubing. Second, the tips of the cali- brated siphons were placed within the upper ends of the levelling tubes (See Fig. 1). In this set-up diffusion of carbon dioxide can take place only through the small tubes in the constant level flasks and from the upper end of the levelling tubes. In practice, cotton was packed about the (8) Figure 1. ILLUSTRATION OF THE PERCOLATION APPARATUS. (11) TABLE 2 P04 EXTRACTED BY SOLUTIONS OF APPROX. 3H 4.6 AND 4.9 RESPECTIVELY. Approx. pH of Extract- Extract Numbers Soil _ipggSoln. 1 2 3 4 5 6 7 p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. 11 4.6 .18 .12 .11 .09 .12 .10 .13 4.9 .12 .10 .09 .09 .10 .10 .12 13 4.6 .08 .05 .06 .13 .14 .13 .14 4.9 .11 .06 .05 .09 .11 .27 .11 8 4.6 .04 .04 .07 .13 .25 .32 .29 4.9 .02 .03 .05 .11 .13 .19 .26 Figure 2. P05 LEACHED FROM SAMPLE 5 BY CARBONATED WATER 0F GRADED CONCENTRATION. 1°00[' .90 #- '80b fitfii” h ' .70 _ I} ............... ('3 to» ’/I <§60> ”/1 Q ,/ 2' 4o T .x” ‘Leo~ ,“ I‘x' .20 ’ PH 435 ______ —~ -10 11:--- __ _,_ _____ .”’__,.~————— /f.\\_k pH 63 ' 2 3 4 é ; -\: EXTRACT NUMBERS (12) solutions of approximately pH 4.6 and 4.9 respectively. Slightly higher results were obtained with the solution of pH 4.6 but the difference is not large. It was found expedient to use "dry ice" as a source of carbon dioxide in preparing the carbonated water. With this material, water in a 20-litre flask was saturated with the gas, then this stock solution was employed in adjusting the extracting solution to pH 4.50. The Beckmann pH-meter, which is equipped with a glass electrode, was found to be very satisfactory for standardizing and testing the carbonic acid solutions. The glass electrode is an almost indispensible tool in determining the pH value of this fickle solution. The Effect of Vagying¥Concentrations of Carbonic Agig; In further preliminary work the effect of various con- centrations of carbonic acid on the amount of phosphorus leached from the soil was determined. Table 3 shows the amount of phosphorus appearing in the successive portions of the continuous extracts when solutions of pH 6.0-6.3, 4.7-4.9 and 4.4-4.5 were resoectively employed. The phosphorus was determined according to the method of Truog and Meyer (30) heading the precautions suggested by Chapman (4). Fig. 2 shows the same data graphically for Sample 5. It is to be seen that the amount of phosphorus extracted from these soils by a solution of pH 4.7-4.9 was only slightly different from that extracted by one of pH 6.0-6.3, the No. 6 and 7 portions of extract showing higher levels with the more concentrated solution. But a solution of pH 4.4-4.5 possessed (13) TABLE 3 P04 EXTRACTED FROM FIVE SOILS BY CARBONATED WATER 0F pH 600-603, 407-409 AND 404-405 W I:——-—— m PH Of Ex- Extract Numbers Soil tracting Solution __; 2 3 4 5 6 7 p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. p.p.m. 6.0-6.3 0.02 0.04 0.04 0.03 0.03 0.04 0.02 1 407-409 004 006 007 005 .07 006 011 4.4—4.5 .05 .06 .09 .16 .44 .51 .60 600-603 002 005 006 003 004 .04 .02 2 407-409 001 004 005 005 005 011 018 4.4-4.5 .01 .07 .12 .28 .42 .61 .66 600-603 002 005 008 008 005 005 0U3 3 407-409 001 A .06 007 .05 006 007 009 4.4-4.5 .03 .07 .07 .22 .35 .55 .62 600‘603 .02 .03 .05 006 005 003 002 4 407-409 001 004 003 004 004 004 005 4.4-4.5 .02 .05 .15 .17 .26 .39 .54 600-603 T 009 007 006 006 004 003 5 407‘409 .04 .06 .07 .08 009 011 014 4.4-4.5 .07 .28 .45 .76 .85 .91 .90 (14) greatly increased power to dissolve phosphates in soils such as these, the levels being successively higher in the later portions of the percolate. The similarity in concentration of phosphorus in the first portions of extract is notable and seems to offer an explanation of the failure of certain workers to obtain significantly increased levels of phos- phorus with carbonated, as compared to pure water, in equili- brium extracts. The Relation of Calcium to Phosphorus in the Extract. In view of the preceding experience it was decided to analyse some of the percolates for calcium to determine the relationship between that element and phosphorus in the ex- tracts. Calcium was determined according to the Cramer- Tisdall method (28). The pH values were also measured and data for all three appear in Table 4. The percolates from samples 10, 22 and 7 show a definite reciprocal relationship, calcium diminishing and phosphorus increasing in successive portions of the extracts. Soils 21 and 12 are highly buffered, the phosphorus remaining uniformly low and the calcium uni- formly high. The reaction tends to follow the calcium, al- though the correlation is not perfect. Some of the data published by Hibbard (15) and McGeorge and Breazeale (17) show this relationship. PART II. The Solubility and Fixation of Phosphorus in Some Calcareous Ontario Soils- (15) TABLE 4 COMPARATIVE DATA FOR pH, P04 AND CALCIUM IN THE EXTRACTS OF TEN SOILS. Extract Numbers “911 1 2 5' 4 5 5 7 DB 701 705 706 706 706 706 '- 21 P04 (p.p.m.) 0.04 0.04 0.05 0.04 0.04 0.05 0.05 Ca (p.p.m.) 45 44 44 44 45 45 51 pH 5.5 5.7 5.5 5.5 5.2 5.1 - 22 P04 (p.p.m.) 0.05 0.55 0.57 0.55 0.70 0.54 0.95 Ca (p.p.m.) 20 15 15 12 11 10 5 pH 7.0 7.5 7.5 7.4 7.5 7.5 - 2 904 (p.p.m.) 0.05 0.05 0.07 0.05 0.05 0.05 0.05 Ca (p.p.m.) 20 19 19 19 18 18 15 pH 5.5 5.5 5.2 5.0 5.9 5.5 - 7 P04 (p.p.m.) 0.05 0.05 0.10 0.12 0.14 0.25 0.50 Ga (p.p.m.) 20 14 10 9 7 5 5 pH 5.9 5.5 5.4 5.5 5.5 5.2 - 10 P04 (p.p.m.) 0.20 0.50 0.50 0.55 0.54 0.50 0.42 Ga (p.p.m.) 22 16 11 10 9 8 7 pa 7.0 7.5 7.5 7.5 7.5 7.5 - 11 P04 (p.p.m.) 0.15 0.11 0.10 0.09 0.11 0.10 0.15 Ga (p.p.m.) 52 55 55 54 52 51 52 pH 7.0 7.5 7.5 7.4 7.4 7.4 - 12 P04 (p.p.m.) 0.01 0.05 0.05 0.05 0.05 0.07 0.07 Ca (p.p.m.) 51 37 39 37 40 42 42 pH 5.9 7.1 5.7 5.4 5.2 5.0 - 15 204 (p.p.m.) 0.05 0.05 0.05 0.15 0.14 0.15 0 14 Ca (p.p.m.) 24 25 18 13 11 11 9 pH 7.1 7.5 7.4 7.4 7.5 7.1 - 14 P04 (p.p.m.) 0.05 0.10 0.20 0.05 0.07 0.05 0.05 Ga (p.p.m.) 29 51 52 52 52 50 21 pH 7.0 7.5 7.5 7.4 7.5 7.5 - 15 r04 (p.p.m.) 0.15 0.12 0.10 0.10 0.09 0.10 0.05 Ga (p.p.m.) 21 25 26 26 25 25 25 (15) The twenty-two samples used in this study are all of surface soil from cultivated fields and they are represent- ative of the various types of calcareous soils which occur in South-Central Ontario (5). Samples 1-5, inclusive, are from well-drained lacustrine clays, 6-11 and 16-18 from upland till soils, 12-15 are poorly drained lacustrine loans and 19-21 are marly soils. No. 22 is a neutral clay soil containing no free carbonates, and is included only for pur- poses of comparison. The data in Table 5 describe the soils from the standpoint of texture, pH and carbonate content. The pH values were determined with a Beckmann pH-meter equipped with a spear-type glass electrode, mechanical analyses were made by the hydrometer method, and the carbonate content mea- sured by means of a home-made Collins calcimeter (6). Percen- tage figures are eXpressed on the basis of soil in the oven- dry state. With the exception of No. 22, all samples contained car- bonates, the amount varying from 0.2% to 69.1% (expressed as C5003 equivalents). The pH values range from 7.1 to 7.75, none of them falling within pH 8.0-8.5 where maximum insolu- bility of phosphates occurs(17). In texture, they vary from sandy loams to clays. Available Phosphorus. This series of samples was leached in the percolation apparatus for about 23 hours using 5 grams of soil. At the outset it must be said that duplicate tests were sometimes in poor agreement, necessitating a third. The figures appear in Table 6. (17) TABLE 5 A DESCRIPTION OF THE SOILS USED IN THIS STUDY, FROM THE STANDPOINT OF REACTION, TEXTURE AND CARBONATE CONTENT. Mechanical Analysis C5003 Textural Soil pH Equivalents Sand Silt Clay Class 35 75 % 2 1 7.35 4.2 28.0 41.2 30.8 Clay 2 7.40 3.5 25.1 40.4 34.5 Clay 3 7.25 4.8 19.3 42.1 38.6 Clay 4 7.25 4.6 19.0 40.6 40.4 Clay 5 7.25 4.0 40.5 32.5 27.0 Clay 105m 6 7.55 10.4 56.4 23.1 20.5 Sandy loam 7 7.10 3.7 43.9 31.0 25.1 Clay loam 8 7.65 0.8 45.7 34.2 20.1 Clay loam 9 7.50 1.3 45.6 30.2 24.2 Clay loam 10 7.50 1.5 55.9 25.7 18.4 Sandy 105m 11 7.50 8.2 52.5 26.2 21.3 Sandy 105m 12 7.50 14.6 20.6 32.5 46.9 Clay 13 7.60 0.4 45.4 39.6 15.0 Loam 14 7.75 1.7 33.8 52.2 14.0 Silt loam 15 7.65 8.1 31.0 47.4 21.6 Loam 16 7.50 0.2 32.2 40.6 27.2 Clay loam 17 7.55 2.1 46.0 31.9 22.1 Clay loam 18 7.45 3.3 45.3 37.5 17.2 Loam 19 7.60 69.1 13.6 29.1 57.3 Clay 20 7.60 63.0 13.7 36.2 50.1 Clay 21 7.45 57.7 11.4 35.9 52.7 Clay 22 7.10 0 16.4 16.3 67.3 Clay (18) TABLE 6 P04 EXTRACTED BY PERCOLATION AND EQUILIBRIUM METHODS. Soil Extract Numbers Sums Equi- librium 1 2 3 4 5 6 7 Extract pope pop. pope pop. pop. p.p' pop. pop. pip. mil. mil. mil. mil. mil. mil. mil. mil. mil. 17 T 0.03 0.03 0.03 0.03 0.04 0.02 0.15 T 9 T .02 .03 .03 .03 .03 .04 0.18 T 19 .02 .03 .03 .04 .04 .04 .03 0.23 .03 21 .04 .04 .05 .04 .05 .05 .03 0.30 .04 20 .06 .08 .07 .06 .06 .05 .04 0.42 .04 4 T .04 .03 .04 .05 .05 .06 0.27 '04 3 T .06 .07 .05 .06 .07 .09 0.40 .03 1 .04 .06 .07 .06 .07 .06 .11 0.47 .03 6 .06 .05 .08 .05 .08 .07 .10 0.49 .03 12 .01 .08 .06 .08 .07 .07 .06 0.43 .08 2 .05 .05 .06 .06 .06 .11 .16 0.55 .03 5 .04 .06 .07 .08 .09 .11 .14 0.59 .05 18 .10 .10 .08 .08 .08 .10 .12 0.66 .08 11 .13 .11 .10 .09 .10 .09 .10 0.72 .11 16 .06 .08 .06 .06 .09 .14 - - .04 13 .10 .06 .06 .11 .13 .20 .13 0.79 .03 15 .23 .21 .10 .13 .16 .14 .14 1.11 .14 7 .04 .05 .08 .11 .15 .25 .28 0.96 .03 8 .03 .04 .06 .12 .19 .26 .28 0.98 .03 14 .07 .11 .25 .24 .29 .33 .36 1.65 .07 10 .23 .46 .77 .71 .67 .55 .45 3.84 .13 22 .03 .38 .37 .56 .70 .84 .98 3.80 .04 (19) These data are informative in three ways. They indicate (1) the initial amount of available phosphorus present in the soil, (2) the rate of solution of phosphorus, and (3) the buffering power of the soil. The first point is not very well illustrated by these samples. That most of them are in a mini- mum state of fertility is shown by the very low levels of phos- phorus in the No. 1 portions of the extracts; on the other hand samples 15 and 11 yield the most phosphorus to the first port- ions of the percolates indicating a better supply of phosphorus in the soil solution or in very readily soluble form. The rate of solution is shown by the amount of phosphorus in the extract and this is, in a general way, related to the buffering capacity. The status of this latter characteristic is deduced from the relative composition of successive portions of the extracts; the samples at the t0p of the list, such as No. 17, are highly buffered since the level of phosphorus remains uniformly low- No. 11 shows the same characteristic although the specific level is higher. On the other hand, when successive portions of the percolate contain increased amounts of phosphorus, as with No. 22 at the bottom of the list, it means that the buf- fering power is lower and the phosphates are more easily avail- able to plants. In Table 6 the samples are arranged in order of their soluble phosphorus content. In general, the lacustrine clay loams and the marly soils are found at the top of the list, showing that they are relatively poor in this respect, while the poorly drained loans and upland till soils are usually better supplied. Soil 22, with a pH value of 7.1 and no free (20) carbonates, is better supplied than any of the calcareous soils. One would expect the fixation of phosphorus by these soils to be in general agreement with these results. Phosphate fixation. The study of phosphate fixation includes further labor- atory work on the same 22 samples and some plot work in the field on a typical area of alkaline calcareous clay soil. The main function of laboratory work is to give a comparative rating to the soils in this respect, and field tests of the residual effect of superphosphate are necessary to show what happens under natural conditions. Needless to say, the problem is of practical interest in relation to the use of commercial fertilizers. In the spring of 1934, a duplicate set of small plots was. established on which graded amounts of 20% superphosphate, from 50 to 2000 pounds per acre, were applied. The arrange- ment of the plots is shown in Fig. 4. Since then the field has received no fertilizer of any sort so that the residual effect of the phosphate might be observed. The material was broadcast by hand and raked into the surface soil, oats were grown on the plots and it is notable that there was no benefit from the fertilizer the first year. In the autumn the field was plowed and in 1955 was again seeded to oats. That season the response was very marked on the fertilized plots, even the 100 pounds per acre application producing a distinct improve- ment in colour and size of the oats. No yields were taken, but instead, tests for inorganic phosphorus in the stem tissue of the plants were made in June with a Simplex soil testing (21) Figure 4. ARRANGEMENT OF SUPERPHOSPHATE PLOTS. 3 Tons 250 lbs. 400 lbs. 50 lbs. per Acre per Acre per Acre per Acre Ca(OH)3 S.‘ Phos. S - Phos. S - Phos. 300 lbs. 200 lbs. 300 lbs. per Acre per Acre per Acre S - Phos. Kcl S - Phos. 1000 lbs. 200 lbs. 500 lbs. 100 lbs. per Acre per Acre per Acre per Acre Sulphur S - Phos. S - Phos. S - Phos. per Acre per Acre per Acre S - Phos. Kcl S - Phos. 2000 lbs. 150 lbs. 750 lbs. 150 lbs. per Acre per Acre per Acre per Acre S - Phos. S - Phos. S - Phos. S - Phos. 500 lbs. 50 lbs. 200 lbs. per Acre per Acre per Acre S - Phos. Kcl S - Phos. 1500 lbs. 100 lbs- 1000 lbs. 3 Tons per Acre per Acre per Acre per Acre S - Phos. ' S - Phos. S - Phos. Ca(0H)2 750 lbs. 2000 lbs. 200 lbs. per Acre per Acre per Acre 1000 its) so lbs. 1500 lbs. 1000 lbs. per Acre per Acre per Acre per Acre S - Phos. S - Phos. S - Phos. Sulphur (22) outfit (25), the procedure being the same except that finely chopped tissue was used instead of soil. Nitrates and potass- ium were also determined, since it is necessary to have know- ledge of these other elements in interpreting the results for phosphorus. According to Thornton's directions (27) stem tissue was used for nitrates and phosphorus and leaf tissue for potash. Since 1935 the field has been in alfalfa, and in 1936 tissue tests were again made on selected plots and the visible differences in growth recorded. These data are set forth in Table 7. With regard to these results it will suffice to say that 250 pounds per acre gave a good response the year following its application, having been plowed down the first autumn, and the effect was still noticeable the following season. Plots receiv— ing heavier amounts could be distinguished in the alfalfa crop the third year after its application. In other words, the re- sidual effect of superphosphate was quite good on this cal- careous clay soil. The concentration of inorganic phosphorus in the plant tissue was generally higher on the treated plots, but with alfalfa in 1936 the levels did not increase until the 400 pounds per acre application was reached. This index was not as sensitive a criterion as visible improvement in colour and size of the plants. Laboratory tests. In the laboratory the fixation capacity was determined in the following manner. Mono-calcium phosphate in solution was added to weighed samples of soil in evaporating dishes, enough water was used to saturate the soil and the suspension was (23) TABLE 7. THE RESIDUAL EFFECT OF SUPERPHOSPHATE AS INDICATED BY TISSUE TESTS AND GROWTH OBSERVATIONS. 1935 Results (Oats) Applications of N03 P K Growth Spperphosphate Hp.p.m. p.p.m. (p.p.m. Observation Check 25 1 High 50 lbs. per Acre 25 1-2% " 100 " " " 15 2%-5 " good response 200 " " " 15-05 2% " " 2000 " " " 10 5 " " Check 25 1 " '“I 1936 Results (Alfalfa) Check 5 §-1 150 50 lbs. per Acre 0 fi-l 100 100 " " " o 1 100 200 " " " 0 1 150 good response Check 0 1 125 " 300 " " " o 1 100 " 400 " " " o 1.2% 75 " 500 " " " o 2 125 " . 2000 " " " o 2A-5 125 " Check 0 é-i 100 (24) thoroughly mixed then set aside to dry at room temperature. When dry, extractions were made with carbonated water of pH 4.5 using a 1:100 soil to water ratio and shaking inter- mittently for one hour before filtering. Phosphorus was determined in the extracts as before. Similar measurements were made on untreated samples so that the amount of phos- phorus retained by the soil could be calculated. Table 8 shows the results obtained. In this table, the samples are arranged according to their fixing capacity. Like the solu- bility tests, these show a wide variation. They are of value mainly in a comparative sense and the most significant point is the position of the No. 2 sample which was taken from the field where the superphosphate plots were situated. It is to be found near the centre of the group, or in other words, about half of the samples have lesser while the remainder have greater fixing power. In view of the good residual effect of superphosphate on No. 2 soil, the indications are that much of the calcareous soil in South-Central Ontario possesses very little ability to fix phosphates in forms which are unavailable to plants. Considering these results in relation to the kind of soil, it is to be seen that there is as much variation within the different classes, as between them. Actually, the two highest (6 and 9) and the second, third and fourth lowest (10, 11, and 18) are samples of upland till soil. The corre- lation between these data and the class of soil is not so marked as it was with the solubility tests. (25) TABLE 9. RELATIVE FIXING CAPACITY AS INDICATED BY P04 RECOVERED IN 1:100 WATER EXTRACTS FROM SAMPLES. Treated Untreated T-U % Soil p.p.m. p.p.m. p.p.m. Retained 6 .07 .03 .04 97 9 .06 .01 .05 96 17 .08 .01 .07 94 21 .12 .04 .08 93 12 .16 .08 .08 92 19 .12 .03 .09 91 16 .14 .04 .10 91 7 .13 .03 .10 91 2 .13 .03 .10 91 3 .13 .03 .10 91 4 .15 .04 .11 90 1 .17 .03 .14 88 20 .19 .04 .15 87 13 .19 .03 .16 86 22 .21 .04 .17 85 15 .32 .14 .18 84 8 .26 .03 .23 80 5 .29 .04 .25 78 18 .35 .08 .27 77 11 .39 .11 .28 76 10 .49 .13 .36 69 14 .48 .07 .41 65 (26) It was thought that the results might have been in better agreement with the solubility data if available phosphorus in these fixation tests had been determined by the percolation method. To do this, untreated and phos- phate-treated samples must be leached side-by-side for an extended period until the level of phosphorus in the per- colate from the treated sample lowers to that of the un— treated. As a supplementary experiment, this method was attempted with four samples, two of them having maximum and the other two with minimum fixing capacity according to the initial method. Ideally, the phosphorus level in the extract from the treated sample should fall to that of the untreated, and from then on the two should coincide. But, variation between results caused the phosphorus level in the extract from the treated sample either to sink below that of the untreated sample, or to fail to reach it. In the first case the results were unusually low, while in the latter case the end point was apparently not reached. On this account the results were discarded as being unsatis- factory. DISCUSSION. The original feature of the extraction method employed in this study is the use of carbonated water of pH 4.6-4.9. In his percolation method, Hibbard (11) uses ordinary distilled water which is in equilibrium with the carbon dioxide of the atmosphere, but in some laboratories the carbon dioxide content of the air varies widely and it was (27) felt that if a reasonably good control over the composition of more concentrated carbonated water could be effected it would be advantageous to use this latter solution. Water saturated with carbon dioxide could not be used in the per- colation apparatus since bubbles of gas appeared in the siphons and stopped the flow. Preliminary work indicated that the variations in pH values of the solutions were just as wide with solutions of lower pH as with less concentrated solutions of pH 4.6-4.9. In terms of hydrogen-ion concen- tration, the former is the greater variation and the result- ant variation in phosphorus extracted would be, according to Fig. 2, more serious. A fairly satisfactory control over the composition can be maintained when the solution is adjusted to pH 4.50; sample 8 is one of the more weakly buffered soils used in this study but the amount of phosphorus extracted by solutions of approximately pH 4.6 and 4.9 respectively is not greatly different. Accordingly, such a solution was adopted as an extracting solution. Using a continuous leaching method the effectiveness of carbonated water in bringing phosphate into solution was readily demonstrated. The first portions of the extract with either distilled water or carbonated water were not widely different in phosphorus content, which agrees with the figures obtained by Schloesing (23) and the Arizona workers (19) who used equilibrium extracts. However, in later portions of the extract the effectiveness of the carbonated water showed up. Comparative figures for calcium and phOSphorus content of the various portions of the percolate showed very clearly, the (28) reciprocal solubility relationship which exists between these two elements. This confirms the results of Hibbard (13) and McGeorge (17) and lends indirect support to Truog's theory that plants which are heavy feeders on calcium are also efficient in obtaining phosphate. McGeorge's objection (17) to leaching methods, namely the lack of accurate reproducibility of results appears to be quite valid. He attributes these variations to "channelling" and this seems logical because variations occur between duplicates which are run in a parallel series; but in spite of this limit- ation it is considered that the results are, on the whole, more informative than those of equilibrium methods. In proceeding to obtain information on a selected series of samples from the calcareous soil areas of South-Central Ontario, further experience was obtained with the method of extraction. At the outset it must be stated that duplicate results were sometimes at variance so that the extraction was repeated. The available phosphorus content of the 22 soils varied widely; the clay soils yielding the least while the remaining soils, which are mainly clay loams and loams in texture, are generally better supplied. Continuous leachings show not only the initial level of available phosphorus which is present in the soils, but also the rate at which phosphate comes into solution, and the buffering power of the soil. Equilibrium extracts correspond fairly well to the first portions of the leachate but the later portions, Which are most informative, may be widely different. (29) No. 22, a non-calcareous, neutral clay soil tests higher in available phosphorus than any of the calcareous soils, but within this latter group there seems to be very little corre- lation between the amount of carbonates in the sample and sol- uble phosphorus; No. 9, with 1.3% is as low or lower in avail- able phosphorus as No. 20, which contains 63.0% of calcium carbonate. _ Comparing these soils, on the basis of the 1:100 water extractions, with the California soils tested by Hibbard (11), most of them are relatively better supplied with available phosphorus. They are generally less alkaline than the Arizona soils (17), none of them having a reaction of above pH 8.0 where minimum solubility of phosphates occurs, and it might also be mentioned that the local calcareous soils are not subject to puddling, like the Arizona soils, but have, on the other hand, usually, good physical condition. The laboratory tests of phosphorus fixation served to grade the 22 samples, relatively, according to their capacity in this regard. Under natural conditions in the field the results from plots on the No. 2 soil in 1933—1936 indicated that fixation was relatively mild, since there were good residual effects from applications of superphosphate. Conse- quently, as the No. 2 soil is about average in fixation capacity, among the 22 samples, it is thought that a consid- erable proportion of the calcareous soil in South-Central Ontario has relatively little power to fix phosphates in unavailable forms, while the seriousness of fixation in those (30) soils having greater capacity than No. 2 cannot be deduced from these results, and therefore awaits further study. It is recognized that more work, particularly plot work in the field needs to be done to establish the points indi- cated by these experiments on fixation, but the results already obtained will be valuable in planning future invest- igations. SUMMARY. Hibbard's percolation apparatus was modified so as to facilitate the use of carbonated water in extracting available phosphorus from calcareous soil. Available phosphorus was determined by leaching 5 grams of soil at the rate of 1 cc. per minute with carbonated water of pH 4.6-4.9 for 23 to 24 hours. An extracting solution of pH 4.6-4.9 was obtained by ad- justing the initial solution to pH 4.50. This variation in the extracting solution does not lead to serious error in the test. The 1:100 equilibrium extracts closely correspond to the first portions of the percolate, but the later portions are often quite different. By the continuous leaching method the effectiveness of carbon dioxide solutions in bringing phosphates into solution is readily demonstrated. The reciprocal solubility relationship between calcium and phosphorus is shown by the composition of successive portions of the percolate. 10. (31) Variations in results, thought to be due to channelling, is the most serious limitation of this method of deter- mining available phosphorus. The alkaline calcareous soils of South-Central Ontario vary widely in available phosphorus content. According to Hibbard's rating some of them are low but many of them are "medium" or "good" in this respect. In general, they are lower in reaction and better supplied with read- ily soluble phosphates than the Arizona soils studied by McGeorge and Breazeale. The lighter textured soils are generally less highly buffered and better supplied with available phosphates than the clays- Fixation capacity of these soils varies widely. The in- dications are that some of them possess little fixation capacity, while in others it may be more serious. The work to date will serve as a guide to future eXperiments along this line. [D 6. 7. 10. 11. 12. 13. (32) REFERENCES. Beater, B. E. Phosphate Fixation in Soils. Soil Sci. 33: 277-291, 1937. Breazeale, J. F. and McGeorge, W. T. Nutritional Disorders in Alkaline Soils as Caused by Deficiency of Carbon Dioxide. Ariz. Agr. Exp. Sta. Tech. Bul. 41, 1932. Buehrer, T. F. The Physico—Chemical Relationships of Soil Phosphates. Ariz. Agr. EXp. Sta. Tech. Bul. 42, 1932. Chapman, H. D. Studies on the Blue Colorimetric Method for the Determination of Phosphorus. Soil Sci. 33: 125-134, 1932._ Chapman, L. J. and Putnam. D. F. The Soils of South- Central Ontario. Sci. Agric. 33: 161-197, 1938. Collins, S. H. Jour. Soc. Chem. Ind. 33: 518—522, 1906. Das, S. The Determination of Available Phosphoric Acid of Calcareous Soils.» Mem. Dept. Agr. India. Chem. Ser. 8, Nos 6, 1926. Ford, M. C. The Nature of Phosphate Fixation in Soils. Jour. Am. Soc. Agron. 33: 134-144, 1933. Harper, H. J. Easily Soluble Phosphorus Content of Soil as Determined by Electrodialysis, Extraction with Dilute Acid Solutions and Crop Response to Fertil- ization. Soil Sci. 33: 1-16, 1933. Hendricks, S. B., Hill, W. L., Jacob, K. D., and Jefferson, M. E. Structural Characteristics of apatite-like Substances and Composition of Phos- phate Rock and Bone as Determined from Microscop- ical and X-Ray Diffraction Examination. Ind. Eng. Chem. 33, p. 1413, 1931. Hibbard, P. L. Estimation of Plant Available Phosphate in Soil. Soil Sci. 33: 17, 1933. Hibbard, P. L. An Apparatus for Percolation at a Uniform Constant Rate, etc. Ind. & Eng. Chem., Anal. Ed. 2: 404, 1930. Hibbard, P. L. Chemical Methods for Estimating the Availability of Soil Phosphate. Soil Sci. 33: 437, 1931. 14. 15. 16. 17. 18. 19. 20. 22. 23. 24. 25. (33) Hibbard, P. L. Factors Influencing Phosphate Fixation in Soils. Soil Sci. 32: 337-358, 1935. Hockensmith, R. D., Gardner, R., and Goodwin, J. B. Comparison of Methods for Estimating Available Phosphorus in Alkaline Calcareous Soils. Colo. Agr. Exp. Sta. Tech. Bul. 2, 1933. Lohse, H. W. and Ruhnke, G. N. Studies on Readily Soluble Phosphate in Soil: I. Extraction of Readily Soluble Phosphate from Soils by Means of Dilute Acid Potassium Sulphate (KHSO4) . II. The Vertical Distribution of Readily Soluble Phosphate in Some Representative Ontario Soils. Soil Sci. 33: 437-457; 459-468, 1933. McGeorge, W. T. and Breazeale, J. F. Phosphate Solu- bility Studies on Some Unproductive Calcareous Soils. Ariz. Agr. EXp. Sta. Tech. Bul. 35, 1931. McGeorge, W. T. The Determination of Phosphate Avail- ability in Calcareous Soils by Electrodialysis. Soil Sci. 33: 347-354, 1934. McGeorge, W. T. and Breazeale, J. F. The Relation of Phosphate Availability, Soil Permeability and Carbon-Dioxide to the Fertility of Calcareous Soils. Ariz. Agric. Exp. Sta. Tech. Bul. No. 36, 1931. Puri, A. N. and Asghar, A. G. Estimation of Available Phosphates in Soils by 003 Extraction. Soil Sci. 33: 39-46, 1936. Parker, F. W. Carbon Dioxide Production of Plant Roots as a Factor in The Feeding Power of Plants. Soil Sci. 31: 229-248, 1924. Russel, J. C. A Method for the Continuous Automatic Extraction of Soils. Soil Sci. 33: 447-450, 1933. Schloesing, A. Th. and Leroux, D. Sur Laction dissolvante du gaz carbonique a l'egard de l'ocide phosphoric dans les sols agricoles. Compt. Rend. Acad. Sci. 329: 989-991, 1930. Smith, F. B., Brown, P. E., Miller, H. C. and Bodily, H. L. The Effect of 009 on Soil Reaction and on The Solubility of Phosphorus in Soils. Soil Sci. 33: 93-104, 1937. Spurway, C. H. Soil Testing. Mich. Agric. Exp. Sta. Tech. Bul. 132 (revised), 1935. 28. 29. 50. 31. (54) Thomas, Walter. The Feeding Power of Plants. Plant Thornton, S. F. The Use of Rapid Chemical Tests on Soils and Plants as Aids in Determining Fertilizer Needs. Perdue Univ. Circ. No. 204, 1934. Tisdall, F. F. J. Biol. Chem. 39: 439-442, 1925. Truog, E. The Utilization of Phosphates by Agricultural CrOps, Uncluding a New Theory Regarding the Feeding Power of Plants. Wis. Res. Bul. 41, 1916. Truog, B., and Meyer, A. H. Improvements in the Deniges Colorimetric Method for Phosphorus and Arsenic. Ind. and Eng. Chem., Anal. Ed. 3: 136-139, 1929. Wilcox, J. C. Effects of Some Field Plot Treatments on Drouth Spot and Corky Core of the Apple. Sci. Agric. 33: 300-315, 1938. Wilson, P. W., Fred, E. B. and Salmon, M. R. The Re- lation Between 002 and Elemental Nitrogen Assimil- ation in Leguminous Plants. Soil Sci. 33: 145, 1933. _.m~————.— 1 115902 .f. (”‘meva 900x? a muster. _--" 'NSINC. ., . mullifiuu‘im' 3046 2018 “I! R" fill Ml.” U" 0 3 1293 umwunmmui