—- MAGNESIUM RELEASE AND: 119mg- FRQM SELECIED micmem $0st - Thesis for theDegree :of Ph. D. _. MHIHiGAN STATE'VUH‘IVERSITY . DONALD iIRDBERT' VCHRISTENSON ‘1 9 6 8‘ ' 'F'fi.‘3 ’9 This is to certify that the thesis entitled Magnesium Release and Uptake From Selected Michigan Soils presented by Donald Rebert Christenson has been accepted towards fulfillment of the requirements for PhoDo degree in $01]. SCience Major professor Date April 2’43 1968 0-169 ,LIBRA Y Michigan 5- ate Universuy I WW BINDING BY- E gHUAG & suns mm mm mc. LIBR \RY BINDERS ABSTRACT MAGNESIUM RELEASE AND UPTAKE FROM SELECTED MICHIGAN SOILS By Donald Robert Christenson Selected Michigan soils were studied to evaluate: 1) the effect of soil type and rate of magnesium applica- tion on magnesium uptake by successive crops of oats and on the extractable magnesium levels of soils; 2) the effect of soil calcium level and soil pH on magnesium release from the soil and uptake by oats; and 3) the relation of type of soil minerals in different soil fractions to the availabil- ity of magnesium for plants. The seven soils studied were: Munising fine sandy loam from a cropped and an uncropped location. Karlin loamy sand from one location. Montcalm loamy sand from three different locations and Sims clay loam from one location. In the first experiment. three levels of magnesium (O. 10. and 20 ppm) were applied to these soils and to a 5% bentonite-sand mixture. Seven consecutive crops of oats were grown without any additional magnesium application. Donald Robert Christenson Yield increases due to applied magnesium were ob— tained only after the third crop. After six crops had been removed. the level of readily available magnesium was ex- tremely low on all soils. The linear correlation coefficient between the ratio of potassium:magnesium in the soil and the uptake of mag- nesium by plants was -O.418. When soil magnesium alone was used. rather than potassium:magnesium. the correlation co- efficient was 0.243. In a second experiment. different calcium levels were applied to each of three pH levels (pH 4.5. 5.5. and 6.5) and eight consecutive cr0ps of oats were grown. Soil pH had more of an effect on yield. tissue mag- nesium content (%) and magnesium uptake (mg/culture) than did calcium level. Yields at pH 6.5 were 1.3 times greater than at pH 4.5. Magnesium content of the tissue was greater at pH 4.5 than at pH 6.5. which was attributed to increased growth at the higher pH level. Release of magnesium from nonextractable forms was 1.8 times as great when calcium carbonate was applied as when calcium hydroxide was applied. while the total uptake for the eight crops was 15%iless. These differences were not explained by the data obtained. Donald Robert Christenson Sufficient quantity of different particle sized fractions from soils and clay minerals to supply 20 milli- grams of magnesium were mixed quartz sand and crOpped to two crops of oats. The relative order of magnesium availability accord- ing to particle size was as follows: < 0.08 = 0.08-0.2 > 0.2-2.0 = 2.0-20320-50u> total soil Chlorite supplied more magnesium than bentonite: bentonite approximately the same as soil clays. Release of magnesium was linearly correlated with magnesium content of the fractions for the 0.2-2.0u clay. but not for the 0.08u. 2.0-20u or 20-50u fraction. This in- dicated that magnesium was released from interlayer sites for the 0.2—2.0u clay and for the second crop for the 0.08- 0.2u clay. The release in the other fractions was predom- inantly from the crystal edges. Differential release of magnesium between the clay fractions could not be related to differences in kinds of minerals present. X-ray diffraction patterns of some soil fractions indicate that interlayer material was removed by cropping. MAGNESIUM RELEASE AND UPTAKE FROM SELECTED MICHIGAN SOILS BY Donald Robert Christenson A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science 1968 Q57¢5> (‘\ \ TO CAROL This thesis is dedicated to my wife ACKNOWLEDGEMENTS The author wishes to express his appreciation to Dr. E. C. Doll for his patient guidance in the conduct of these studies. A special note of appreciation is extended to Dr. K. V. Raman for his suggestions and assistance in the mineraloqical studies. The writer is grateful to Mr. O. G. Pierce for his assistance in the laboratory. Special acknowledgement is given to Miss. J. K. Bennett for her effort conducting the many laboratory analyses. The writer would also like to thank Mrs. Nellie Galuzzi for her assistance in the statistical analyses of the data. Financial assistance by International Minerals and Chemical Corporation is gratefully acknowledged. iii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . . iii LIST OF TABLES. . . . . . . . . . . . . . . . . . . . Vii LIST OF FIGURES . . . . . . . . . . . . . . . . . . . x INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . . . 5 Magnesium Deficiency Symptoms in Plants . . . 5 Potassium-Magnesium Relationships in Soils and Plants. . . . . . . . . . . . . . . . . 6 Calcium-Magnesium Relationships in Plants . . 8 Magnesium Fertilizer Recommendations Based on Soil Tests. . . . . . . . . . . . . . . . . 10 Fixation of Magnesium in Soils. . . . . . . . 12 Release of Nonexchangeable Magnesium. . . . . 13 Release of Lattice Magnesium by Chemical Weathering. . . . . . . . . . . . . . . . . 15 Release of Lattice Ions as a Rate Process . . 17 MATERIALS AND METHODS . . . . . . . . . . . . . . . . 21 A. Sources of Soils and Minerals . . . . . . . . 21 B. Fractionation of Soils and Minerals . . . . . 23 iv Table of Contents (cont.) C. RESULTS A. Cr0pping Procedures and Treatments. . 1. General Greenhouse Procedures . . . . 2. Intensive Cropping Study. . . . . . . . 3. pH- -Calcium Level Study. . . . . . . 4. Cropping of Soil and Clay Mineral Fractions . . . . . . . . . . . . . . . Laboratory Procedures . . . . . . . . . . 1. Soil Analysis . . . . . . . . . . . . 2. Tissue Analysis . . . . . . . . . . . 3. Total Magnesium Analysis of Fractions 4. X-Ray Characterization of Soil and Clay Mineral Fractions . . . . . . . . . . . 5. Statistical Analyses. . . . . . . . . . . AND DISCUSSION. . . . . . . . . . . . . . Intensive Cropping Study. . . . . . . . . . 1. Growth Response to Applied Magnesium and Magnesium Uptake. . . . . . . . . . . 2. Extractable Soil Magnesium and Release of Unextractable Magnesium During Cropping 3. Linear Correlations of Plant Tissue Magnesium Content and Uptake Correlated with Extractable Magnesium. Potassium. and Calcium Levels. . . . . . . . . . pH-Calcium Level Study. . . . . . . . . . . l. Yields. . . . . . . . . . . . . . . . . 2. Soil pH . . . . . . . . . . . . . 3. Soil Magnesium Levels and Magnesium Uptake. . . . . . . . . . . . . . . . . 4. Linear Correlations of Plant Tissue Magnesium Content and Uptake Correlated with Extractable Soil Magnesium. Potassium. and Calcium Levels . . . Page 24 24 25 26 27 29 29 29 30 3O 31 33 33 33 43 45 51 51 53 59 61 Table of Contents (cont.) C. CrOpping of Soil Fractions. SUMMARY LITERATURE Magnesium Content of Plant Tissue . Magnesium Uptake and Availability from Different Fractions Comparison of Magnesium Availability of the Coarse Clay Fractions from the Montcalm Soils. . Magnesium Content of the Fractions Before and After Cropping. Mineralogical Composition of Soil Frac— O O tions and Weathering Due to Cropping. REVIEWED Page 63 64 66 71 72 73 79 84 Table LIST OF TABLES Soil series. collection site. initial pH. initial Mg level. and particle size distri- bution for soils used in these studies . . . . Treatments. initial pH and initial Ca levels for the pH_Ca level StUdy. o o o o o o o o o o o Yields of oat tops as affected by soil type and rate of applied Mg under intensive cropping. Mg content of oat tops as affected by soil type and rate of applied Mg under intensive cropping . . . . . Uptake of Mg by oat tops as affected by soil type and rate of applied Mg under intensive cropping . . . . . . . . . . . . . . . . . Extractable Mg levels as affected by soil type and rate of applied Mg under intensive cropping . . . . . . . . . . . . . Probabilities for significance of differences between means using Tukey's HSD for yields. Mg content. and Mg uptake for the intensive cropping experiment. . . . . . . . Total Mg uptake and change in extractable Mg (mg/36009 soil) for crop 2 through crop 7 of the intensive cropping experiment. . . . . . Linear correlation coefficients (r) for Mg content and uptake of magnesium for each crop as a function of various independent variables. . . . . . . . . . . . vii Page 22 26 34 36 37 39 40 42 46 List of Tables (cont.) Table 10 .’ 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Linear correlation coefficients (r) for Mg content and uptake of Mg for each soil as a function of various independent variables Yields of oat tOps as affected by pH and Ca levels under intensive crOpping. Analyzed statistically as a 3x2 factorial. . . . . . Mg content of oat tops as affected by pH and Ca levels under intensive crOpping. Analyzed statistically as a 3x2 factorial. . . . . . Mg uptake by oat tops as affected by pH and Ca levels under intensive cropping. Analyzed statistically as a 3x2 factorial. . . . . . Soil pH as affected by pH and Ca level under intensive cropping. . . . . . . . . . . . . Soil Ca levels as affected by pH and Ca level under intensive cropping. . . . . . . . . . Soil Mg levels as affected by pH and Ca level under intensive cropping. . . . . . . . . . Soil K levels as affected by pH and Ca level under intensive cropping. . . . . . . . . . Total uptake of Mg and A-Extractable Mg for the pH-Ca experiment. . . . . . . . . . . . . . Linear correlation coefficients (r) for Mg content and uptake as a function of various independent variables for crops grown on the pH-Ca level experiment. . . . . . . . . . . Mg content of fractions. Mg uptake by oat tops and Mg content of oat tops as affected by soil type and particle size . . . . . . . . Means of Mg uptake by oat tops and Mg content of oat tops as affected by fraction size. . viii O Page 49 52 54 55 56 57 58 59 60 62 65 66 List of Tables (cont.) Table Page 22. Linear correlation coefficients (r) between uptake of Mg and Mg content of fractions at start of each crop. . . . . . . . . . . . . . 67 23. Minerals present in various particle size fractions of seven Michigan soils . . . . . . 69 ix LIST OF FIGURES Figure Page 1. X-ray diffraction patterns for coarse clay (0.2-2.0 u) of the Karlin loamy sand soil prior to and after cropping. . . . . . . . . . 75 2. X-ray diffraction patterns for medium clay (0.08-0.2 u) for the Montcalm loamy sand (1) prior to and after cropping. . . . . . . . . . 77 3. X-ray diffraction patterns for medium clay (0.08-0.2 u) of Montcalm loamy sand (3) prior to and after cropping. . . . . . . . . . 78 MAGNESIUM RELEASE AND UPTAKE FROM SELECTED MICHIGAN SOILS INTRODUCTION Magnesium was first shown to be essential for plant growth in the latter part of the 19th century (Loew. 1892). Since that time it has been proven to be metabolically linked in the activation of several enzymes involved in glycolysis and respiration. Magnesium is also the coordin- ating atom of the chlorophyll molecule. Geologically. magnesium ranks as the seventh most abundant element in the earth's crust. During the primor- dial stages of differentiation of the earth. i.e.. cooling of the molten magma. the distribution of the various chem- ical elements was essentially controlled by their chemical affinity for oxygen and sulfur. Crystallization involved a selection of various atomic ions according to their size and coordination number. Thus. ions of similar size and coordination number will tend to appear in the Same crystal lattice. According to Goldschmidt (1945). the possibility 1 of large scale isomorphous substitution in minerals from magmas will be limited to pairs of ions in which the radii do not differ by more than 10-15 per cent. Magnesium with an ionic radius of 0.78 A and ferrous iron with a radius of 0.83 A freely replace each other in ionic crystals. Thus. magnesium and iron are found occupying the same coordination habitat in many minerals. In soils. magnesium is a constituent of dolomitic limestone and of primary silicate minerals including bio- tite. hornblende. augite and olivine. Magnesium occupies an octahedral coordinating position in secondary silicates which include talc. serpentine. Chlorite. vermiculite. illite and montmorillonite. It is also present in trace amounts in kaolinitic minerals. In addition. magnesium occupies cation exchange sites on the surfaces of silicate minerals and organic matter. Even though magnesium deficiencies have been re- ported on acid podzols in Europe and on the acid soils of the Gulf and Atlantic coastal states. only a few instances of magnesium deficiency had been reported on Michigan soils and cr0ps until 1963. Tobin and Lawton (1962) found no yield reSponse to magnesium fertilizers on 14 different field crops on 37 different soil types. However. their re- sults did show that the uptake of magnesium was generally increased in most crops by the application of MgSO4°7H20 (Epsom salts) or K SO ~2MgSO 2 4 (sulfate of potash-magnesa). 4 They concluded that the application of magnesium containing fertilizers was not necessary at that time. The possibility that magnesium deficiencies might occur in the future was stressed because of more intensive cropping accompanied by higher yields and larger applications of high-analysis fertilizers. In 1964 in Michigan. Doll and Hossner reported a definite response to applied magnesium on Karlin and Kal- kaska soils. but no reSponse on Emmet and Montcalm soils. The pH and exchangeable magnesium potassium and calcium levels were comparable on the Kalkaska and Montcalm soils. These workers felt that factors other than soil acidity and exchangeable cation levels must have affected the availabil- ity of magnesium and thus the oat yields on these soils. This study was undertaken to evaluate the following factors on selected Michigan soils: 1. Effect of soil type and rate of magnesium applica- tion on magnesium uptake by successive crops of oats and on the extractable magnesium levels of soils in a greenhouse study. Effect of soil calcium level and soil pH on mag- nesium release from soil and uptake by oats. Relation of type of soil minerals in different soil fractions to the availability of magnesium for plants. L I TE RAT URE REV I EW Magnesium Deficiency Symptoms in Plants A description of Mg deficiency has been given by Cook and Millar (1953) as follows: Symptoms of magnesium deficiency first appear on the older leaves. Yellowing appears in patches between thexmflns and around the leaf edges. Leaf edges usually roll slightly. As the magnesium- deficient leaf becomes older. the yellow areas scattered over the leaf blade become necrotic until finally the tissue may disintegrate to leave holes in the leaf. Carolus (1933) described the condition in potatoes as char- acterized by (l) a yellowing of the leaves of the plant. the lower leaves and especially the apical leaflets yellow- ing from the margin toward the midrib. (2) stunted growth of the plant during the early stages of growth and (3) new terminal growth of the plant consists of small half-closed leaves purplish—gray in color. On oats it has been de— scribed as appearing like a "string of pearls" within the leaves (Schachtschabel. 1957). In a discussion of early work on Mg (Ferrari and Sluijsmans. 1955). it was mentioned that this disorder was called "Hooghalen disease" or "acid 5 disease." In cereals it was characterized as a "curdling" of the chlorophyll giving the leaves a mottled (”tigered") appearance. As the condition progressed. the amount of chlorophyll decreased and the leaves became yellow colored. In many cases the symptoms would occur on a young growing plant. but would disappear before the plant reached maturity. Potassium-Magnesium Relationships in Soils and Plants The fact that high rates of potassium applied to soils can induce Mg deficiency in crops is well documented (Adams and Henderson. 1962; Boyton and Burrill. 1944; Car- olus. 1937; Constable. 1955; Cooper and Wallace. 1937; Drosdoff and Kenworthy. 1944; Drouineau and Nageotte. 1941; Embelton and Boyton. 1950; Hovland and Caldwell. 1960; Lucas and Scarseth. 1947; Scharrer and Mengel. 1958; Sluijsmanns §£;§1,. 1959 and many others). In many cases the deficiency of Mg was confirmed by the low content of Mg in the tissue with an accompanying high K content. In the literature re- viewed. no consistent ratio between these ions was estab- lished in the plant which was definitely related to defi— ciency or sufficiency. However. Hossner et al. (1968) reported that when soil K: Mg ratio exceeded five.-K- in- duced Mg deficiency occurred on potatoes grown on an acid sandy podzol in northern Michigan. Other reported values for this ratio have ranged from 6:1 to 3:1. The dominant factor in Mg uptake appears to be the ratio between soil.K and Mg levels or even the soil K level rather than the Mg level in the soil. Wehunt and Purvis (1954) reported a correlation coefficient (r) of -0.69 between leaf Mg and available K in the soil. Tucker and Smith (1952) stated that K exerted control over Mg rather than Mg over K. The occurrence of K-induced Mg deficiency can be explained in two ways. First. according to the law of mass action. additions of K to the soil releases exchangeable Mg to the soil solution which may be subsequently leached from the root zone. However. this does not completely explain K-induced Mg deficiency because it occurs in the greenhouse experiments where leaching is not a factor. Secondly. an antagonism between the two ions may exist within the plant (translocation mechanism) or at the root surface (uptake- mechanism). Scharrer and Mengel (1958) reported that a physiological antagonism exists which is independent of the colloidal effects of the soil and of the anion of the K salt applied. Their results suggest that this antagonism is re- stricted to the green tissues. eSpecially the leaves. ACain (1955). working with the apple trees in sand cultures. found that the effect of K in reducing the leaf concentration of Mg was much greater than was that of Mg in reducing leaf concentration of K. He suggested that the interaction or antagonism between these two nutrient elements was associ- ated entirely within the plant. Other workers reported that when large applications of potassium are applied. the Mg concentration in the leaves is decreased (Foy and Barber. 1959; Hashimoto. 1955; Hovland and Caldwell. 1960; Larsen et al.. 1959; Lucas and Scarseth. 1947; Southwick. 1943; and others). This evidence indicates that this K-Mg relationship is probably localized. for the most part. in the uptake or translocation of these ions. However. no evidence was found in the literature explaining a mechanism for this antagonism. Calcium-Magnesium Relationships in Plants Several investigators have reported that high Ca:Mg ratios suppress Mg uptake by plants (Blair et al.. 1939; Halstead. et al.. 1958; Jacoby. 1961; Salmon. 1964; and others). As in the case of potassium. this appears to be an interaction within the plant rather than completely soil related. Moore et a1. (1961). reported that Mg absorption by 6—day old excised barley roots was sensitive to low tem- perature and dinitrophenol. Excess Mg absorption was also associated with organic acid production in the root. further confirming the metabolic nature of the process. A large part of the Mg absorption was effectively blocked by Ca. This blockage was very pronounced even at low Ca concentra- tions where there was a net loss of Ca from the tissue. Jacoby (1961). using a split-root technique. demonstrated that impaired Mg uptake at the MggCa ratio of 0.05 in the medium was not due to low Mg content. but rather to an ex— cess of Ca. Komai and Noda (1959) reported that Ca decreased the rate of Mg uptake by barley roots. On the other hand. when acid soils are limed. it has been reported that Mg uptake is increased over unlimed treatments (Doll and Hossner. 1964; Korableva. 1954; Carolus. 1933). This may be attributed to two factors. First. even though calcitic sources of lime were used. these materials may contain "contaminating" Mg. Secondly. the associated increase in pH decreases the factors of H ion injury to the 10 plant root and blockage of uptake mechanisms. Rains gt_31. (1964) stated that H ions appear to block the uptake of nutrients or damaged the carrier system. The presence of Ca ions minimizes injury. Moore gt_al. (1961) reported that Mg uptake from MgBr increased with rising pH and Ca level. 2 reaching a maximum at pH 6.0-6.5. Although there is no direct evidence for such a statement. it appears that a minimum ratio of Ca:Mg is re- quired for Mg uptake. there is also a maximum ratio above which Mg uptake is suppressed. Bear and Prince (1945) pos- tulated that each cation has at least two functions in the plant. one specific and the other(s) of the type that can be performed by the other three cations (K. Ca. Mg). Once the supply of each cation is adequate to meet the Specific need for it. there can be a wide range in ratios andsquan- tities that are absorbed by the plant to meet its total needs. Magnesium Fertilizer Recommendations Based on Soil Tests In the Michigan State University Soil Testing Lab— oratory. Mg is extracted from soils with neutral M ammonium 11 acetate (Doll and Christenson. 1966). Soil pH is determined on a 1:1 soil to water ratio using a glass electrode. Mag- nesium recommendations are made on this basis. Dolomitic limestone is recommended for acid sandy soils (pH < 6.5) which have less than 75 pounds of Mg per acre. 0n sandy soils above pH 6.5. which contain less than 75 pounds Mg. soluble Mg fertilizers are recommended at a rate of 50 to 100 pounds of Mg per acre. Magnesium sulfate. sulfate of potash-magnesia or magnesium oxide are all considered satis- factory carriers of Mg. Foliar sprays are also suggested at a rate of 10 to 20 pounds of magnesium sulfate in 100 gallons applied to an acre. When soil tests indicate that the K/Mg ratio is greater than 4:1. crops should be watched for Mg deficiency. Schachtschabel (1957) reported that a good correla- tion was obtained between CaCl2 extracted Mg and availabil- ity to the plant. The critical level reported was 100 pounds per acre. Lancaster (1958) reported an excellent chance of response when less than 3%iof the exchange complex was Mg saturated. Adams and Henderson (1962) reported that soils with < 4% of the exchange capacity Mg saturated were deficient. 12 Fixation of Magnesium in Soils MacIntire and Shaw (1926) were among the first to report that magnesium was fixed in soils. Four years after the application of 3750 pounds of CaCO equivalent per acre 3 of MgO or dolomitic limestone. the amount of fixation was about 14 times the average loss due to leaching. After four years of outdoor exposure without cropping or cultivation MacIntire e£_§1. (1934) found that fixation was greater from applications of 32 tons than 8 tons per acre. The adsorbed Mg was found to be resistant to 8 successive leachings with NH4C1. An increase in soluble aluminum on the Mg treatments suggested that Mg had disrupted the A1 complex. possibly in the octahedral layer. Cheminade and Drouineau (1936) sug— gested that Mg was fixed by octahedral coordination at ex- posed edges of clay particles. Echevin (1935). in leaching studies. also suggested that Mg was adsorbed by soils. Out of 20 New Jersey soils Prince gt_§1. (1947) found that Mg fixation occurred on one-half of the soils studied. Hossner g£_gl. (1968) found that a significant portion of Mg applied in the spring was not extracted by ammonium acetate the fol- lowing fall. No data were presented to show whether or not this loss was due to fixation or leaching. 13 Release of Nonexchangeable Magnesium Release of any element to a form available for plants is essentially brought about by processes collec- tively known as weathering. Jackson et a1. (1948) described weathering as a function of capacity and intensity factors. Intensity factors are: temperature. leaching rate. acidity. the degree and fluctuation of oxidation; capacity factors: specific surface of particles and Specific nature of the mineral. Weathering of primary minerals in nature may re- lease substantial amounts of plant nutrients. Vageler (1933) states that tropical soils are worthless unless their con- tent of primary minerals is substantial. In Indonesian soils. Van Der Marel (1947) found that primary minerals .were the source of nearly all of the calcium. magnesium and potassium for plant growth. Lea and Smith (1938) found a very low availability of Mg from serpentine and olivine in pot experiments in the greenhouse. Longstaff and Graham (1951) measured the re- lease of Mg from horneblende. olivine. talc. magnesite and dolomite by cropping sand cultures with these minerals as the source of Mg. Plants Supplied with magnesite and dolo- mite were able to utilize 45 and 66.5 per cent. respectively. 14 of the total magnesium present. Olivine supplied sufficient magnesium to produce nearly the same amount of growth. Talc and hornblende supplied insufficient magnesium to produce nearly the same amount of growth. Talc and hornblend sup— plied insufficient amounts of Mg for plant growth. Colloidal inorganic soil materials could possibly break down sufficiently to result in a portion of the lat- tice Mg becoming available for plant consumption (Albrecht. 1938). Rudgers (1966) found calcined magnesite and cal- cined brucite a better source of Mg when coated on mono- ammonium phosphate or dicalcium phosphate than uncalcined brucite or serpentine applied in the same manner. Hossner §£_31. (1968) reported that from 15 to 20 pounds of Mg was released between the fall and the following spring on a Kar— lin loamy sand. Noda e£_a1 (1956) found that by increasing the K:Mg ratio. Mg released from bentonite and kaolinite was decreased. 15 Release of Lattice Magnesium by Chemical Weathering Magnesium is generally found in primary and second- ary minerals in an octahedral coordination and has been thought to be released too slowly to meet the requirements of rapidly growing plants. Personal observation. that by co-workers and reports in the literature (Ferrari and Sluijsmans (1955) support this observation. Fast growing plants will develop Mg deficiency symptoms. but later will "outgrow" these symptoms with no apparent decrease in yield. Laboratory methods of measuring the release of Mg from minerals generally consist of acid dissolution methods. Stahlberg (1961) found that augite and hornblende released more Mg than Ca when boiled in normal HCl; phlogopite and particularly biotite were less stable. Semb and Oien (1961) found that the solubility of olivine was directly propor- tional to the acidity. Octahedral cations of biotite and glauconite were completely removed by heating (74-100° C) in 2‘5 HCl for eight hours (Gastuche and Fripiat. 1962). It was demonstrated that octahedral cations were much more mobile and susceptible to acid dissolution than were those in the tetrahedral layer. Several mechanisms have been 16 proposed to describe the dissolution or removal of ions from minerals. Barshad (1960) reported that the relative proportions of Mg and A1 displaced was dependent on the total MgO and A120 contents of the crystal structure of 3 the acidified minerals. the nature of the acidifying solu- tion and the technique used to acidify the clay. He postu- lated. on the basis of geometry of the crystal. that the H ions enter the interior of the lattice as a bare proton. Droste (1960) suggests that the weathering of the brucite layer of chlorites includes a hydration envelope at the weathered edges of the lattice. This is caused by the conversion of exposed hydroxyl groups to water by hydrogen ions. Each hydroxyl would leave half of a divalent charge and one-third of a trivalent charge with a resulting posi- tive charge accumulation. Ultimately this would lead to a certain number of octahedral cations free to go into solu- tion and leave the structure. Only those cations necessary to balance the charge of the mica layers would persist in the lattice. Oxidation of iron in the lattice with the sub- sequent release of magnesium from Chlorite and illite was proposed by Murray and Leininger (1956). 17 Release of Lattice Ions as a Rate Process Release of lattice ions has been characterized as a first order kinetic reaction. Kerr et a1.(1956) found that release from H—hectorite followed two consecutive first order reactions. First. the strong acid underwent a rapid. spontaneous reaction to weak acid. Secondly. a slower spon- taneous reacthxxof the weak acid yielded a neutral clay. For each milliequivalent of strong—acid hydrogen undergoing reaction. one millequivalent of Mg ion was released from the crystal lattice. Besides. for each millequivalent of weak acid hydrogen ion undergoing reaction. one millimole of silica was released from the lattice. They proposed that the rate determining step in the first reaction consisted of a proton attacking the monohydroxylated Mg at the crystal edge resulting in the formation of water. This would re- lease the Mg ion from the lattice and a second proton would become attached to the highly nucleophilic Si—O system. The resulting dihydroxylated silicon would be a monobasic weak acid. From this acid resulted the second-first order reac- tion. This was envisioned to be a hydrolysis or depolymeri- zation releasing a low molecular weight silicate or silicic acid. After the release of the Mg and silica the freshly 18 exposed crystal edge would be identical with the crystal before attack and the process could be repeated. Removal of lattice ions which do not grossly alter the structure have also been reported. Mortland (1958). Ellis and Mortland (1959). and Mortland and Ellis (1959) showed that removal of K from vermiculite and biotite by 0.1 E CaCl could be described by a first-order reaction; also that the rate limiting step was film diffusion. Dif- fusion of ions through a solution film is a first-order reaction and can be written as: ln Q2 - 1n B -'32 t dt VL where c is the amount of ion remaining in the mineral at time t. B is a constant. q is the cross—sectional area of the diffusion film. V is the volume of the diffusion cham- ber. L the thickness of the diffusion film. and D the dif- fusion coefficient. Meller and Bright (1958) reported that diffusion of an ion from a mineral particle could be repre- sented by a similar equation. except that a reaction con- stant. A. was substituted for the quantity. 3%. Hossner (1956) found that the logarithmic rate of release of Mg from vermiculite. mica. and prochlorite plotted against time decreased linearly after an initial l9 nonlinear decline of about 3000 minutes. Release of Mg was affected by pH and particle size. As either decreased the rate of Mg released increased. Mortland and Lawton (1961). working with biotite. noted that K release from different particle sizes depended upon stage of alteration. Potassium concentration of the solution phase was related to total K content of the biotite after equilibrating for 90 days. The rate of release of lattice potassium from 2:1 minerals was described by the following equation: r = B(Cl-C) where r is rate of release of lattice K. B is the diffusion velocity constant containing the diffusion coefficient and geometry parameters. C is the activity of K in the lattice l and C is the activity of K in the solution phase. As long as C is greater than C. release occurs. When r is 0. then 1 Cl - C allowing the determination of the activity of the lattice K by measuring the activity of the K in the solution phase at equilibrium. Doll §£_al. (1965) cropped various fractions of six soils. The K content of the silt and clay was linearly cor- related with the lOgarithm of the K uptake from each fraction. 20 except for the silt fraction of one soil. No correlation was noted between the uptake of K and the K content of the entire soil. In an equilibrium experiment. the concentra- tion of K in solution from the different clays was linearly correlated with the K content of the clay. They suggested that the plants acted as a sink to remove released K from solution in the cropping experiments. while released K re- mained in solution and would tend to depress further release of K in the equilibrium experiment. MATERIALS AND METHODS A. Sources of Soils and Minerals Soils were selected from different locations within a soil series and from different soil series with known dif- ferences in Mg level. history of response to applied Mg and different mineralogical composition. The initial Mg level and pH of the series are listed in Table 1. Samples from the Ap horizon of these soils were collected from field 10- cations during August and September. air dried and screened through a one-quarter inch screen. Two-hundred pounds of each soil was thoroughly mixed and saved for these studies. Bentonite and chlorite minerals were obtained from Ward's Natural Science Establishment. The minerals were broken into small pieces with a hammer and chisel. These pieces were then ground in a ball mill. Contents were re- moved every 8-16 hours. screened through a 12 mesh sieve and the coarse material replaced for continued grinding. Finally. the ground material was mixed and saved for further use. 21 Table l.--Soi1 series. 22 collection site. initial pH. initial Mg level and particle size distribution for soils used in these studies. Separate Soil Series Location pH Mg Sand Silt Clay ppm ------- % -------- Munising fine* Houghton county 4.7 43 72.2 23.7 4.1 sandy loam (l)** Munising fine Houghton county 5.1 30 74.0 21.0 5.0 sandy loam (2) Karlin loamy Otsego county 4.6 11 81.5 14.5 4.0 sand Montcalm Montcalm county 4.8 29 73.5 21.7 4.8 loamy sand (1) Montcalm Montcalm county 5.9 91 68.7 25.6 5.7 loamy sand (2) Montcalm Montcalm county 7.1 93 75.0 18.4 5.6 loamy sand (3) Sims clay Saginaw county 7.7 181 34.4 33.5 32.1 loam *Not recently cr0pped. surface 6 inches sampled. **Numbers in parenthesis refer to different locations within a series. 23 B. Fractionation of Soils and Minerals In order to separate the various sized fractions from the soils. sufficient soil to yield the desired quan- tity of clay was buffered with sodium acetate-acetic acid buffer. pH 4.8. and then treated with H202 to remove the organic matter. followed by leaching with more buffer and then with water to remove the salts. The treated soil was suspended in water and the pH adjusted to the phenolphthalein end-point with NaOH. Dispersion was completed by shaking the suspension for 48 hours on a reciprocating shaker. with periodic readjustments of the pH. The sand fraction (>50u) was separated by wet sieving and the silt (2—50 u) fraction separated from the clay by sedimentation. They clay was separated into fine. medium. and coarse fractions ((0.08. 0.08-0.2 and 0.2-2 u respectively) with a Sharples centri- fuge. Free iron was removed from the clay prior to this separation by treatment with sodium citrate and sodium di- thionite as outlined by Jackson (1956). Samples of the clay minerals as prepared in the ball mill. were placed in water and ground in a Waring blender. The minerals were then sodium saturated with repeated wash— ings of NaCl and then washed free of chloride; followed by 24 suspension in water and separation into the three clay sized fractions as above. The silt was separated into a fine fraction (2-20 u). and a coarse fraction (20-50 u) by sedimentation. All frac- . . ++ . tions were saturated w1th Ca and dialyzed free of excess salts against distilled water. Several drops of toluene were added to retard microbial growth and the samples were stored in that condition. C. Cropping Procedures and Treatments 1. General Greenhouse Procedure Oats (variety Garry) was used as an indicator crop in all of the studies. The seeds were planted at a depth of 1/2 inch and thinned to the required number of plants after 10-14 days of growth. The plants were harvested when the inflorescence was beginning to emerge from the sheath. except on the crOpping of soil fractions which is described in that section. In all cases. the harvested tissues were dried at 65° C. weighed. ground to pass a 20 mesh sieve and saved for chemical analysis. Cultures were watered daily as required and brought to field capacity once per week except on the cropping of soil fractions where it was required daily. 25 2. Intensive Cropping Study In order to determine the pattern of response to applied Mg and the supplying characteristics of these soils. seven successive crops were grown in the greenhouse. Ini- tially. 3600 grams of soil and the same amount of a quartz sand-vermiculite mixture (5% clay material) were weighed into gallon cans lined with plastic sacks. Rates of 0. 10. and 20 parts per million (ppm) were established using M980447H20 (Epsom salts). The granulated salt was mixed with the soil prior to placement into cans. Each treatment on each soil was replicated four times. Each pot was brought to field capacity with distilled water and seeded with 30 seeds; later thinned to 23 plants. Initially. 50 ppm of N °H O and K as KCl were added in as NH NO 4)2 2 4 3. P as Ca(H2PO solution with the initial water. Supplemental N at a rate of 25ppm was added as NH4NO3 to the first two crops; there- after supplemental N was added as Ca(NO On successive 3)2'~ crops. these amounts of N were used. but P and K were added at rates considered adequate based on soil tests. Prior to seeding the third crop. 3.6 ml of Hoaglund's (1950) micronutrient solution was added to each pot. Between seedings. the soils were removed from each pot while the soil was still moist. screened through a 26 one-quarter inch sieve. mixed and replaced with the water and nutrients added as described above. A sample of soil was removed for chemical analysis after the soil was mixed. 3. (pH—Calcium Level Study Calcium and pH levels were established on pots con- taining 3600 gm of Karlin loamy sand soil. given in Table 2. Treatments are Table 2.-—Treatments. initial pH and initial Ca levels for the pH-Ca level study. Treatment Code Treatment _ pH Ca leCal 118 ppm Al as A12(SO4)3 4.3 184 pH1Ca2 118 ppm Al as A12(SO4)3 4.2 957 1100 ppm Ca as CaSO4 pHZCal 512 ppm Ca as CaCO3 5.7 644 pHZCa2 512 ppm Ca as CaCO 5.4 975 514 ppm Ca as CaSO4 pH3Ca2a 1012 ppm Ca as CaCO3 6.7 1030 pH3Ca2b 1012 ppm Ca as Ca(OH)2 6.5 883 Check Check (No amendments) 4.6 239 27 The salts listed in Table 2 were mixed with the soil which was then placed in gallon cans lined with a plastic sack. Each treatment was replicated four times. Sufficient water was added to each pot to bring the soil to field ca— pacity. The plastic was tied around a piece of glass tubing inserted into the soil and the soil was incubated for 8 weeks. Moisture content of the soil was readjusted to field capacity periodically during the incubation period. The soil was removed from the pots and prepared for cropping as described in the previous section. Initially. 50 ppm of each—~N. P. and K-—was added as the same sources as de- scribed for the previous experiment. Eight successive crops were grown with the same gen- eral procedure as described under the intensive cropping section except that supplemental N was added as NH4NO3. Soil was kept moist between crops. except crops 1 and 2. 6 and 7. 4. Cropping of Soil and Clay Mineral Fractions In order to crOp the various sized fractions. suffi- cient amounts of each fraction and of each soil to supply 20 mg of Mg were placed in a waxed carton and mixed with 100 gm 28 of acid washed quartz sand. Five-hundred grams of sand was placed over this and sufficient water added to bring the sand to 10% moisture. Sixteen oat seeds were placed over the surface of the sand and were covered with an additional 100 gms of sand. A modified Hoagland's solution (1950) (less Mg. and 4 ppm Fe as Chel 1383) was added at a rate of 50 ml each week during the growth. Plants were harvested when the vegetative growth had reached what was considered its maximum. After the first crop was removed. the fractions were recovered by wet sieving on a 270 mesh sieve. The exchange— able Mg was removed by leaching with CaCl2 and the excess salts were removed by leaching with water. Each separate from each pot was suspended in 50 ml of water and a 5 ml aliquot was removed for chemical analysis. The remainder of the sample was replaced into the carton as described above. except that an additional 100 gm of sand plus the additional water requirement were added. The second crop was grown in the same manner and was harvested when the Mg deficiency symptoms were considered to be at the maximum. Sodium ferric ethylenediamine di-(o-hydroxyphenylacetate). Geigy ChemiCal Corporation. 29 After the second crop was removed the separates were re- covered in the same manner as above. The fractions were Ca saturated and dialyzed against distilled water. The fractions were saved for x—ray and chemical analysis. D. Laboratory Procedures 1. Soil Analysis Soil samples were air dried and sieved through a 20 mesh sieve. The cations were extracted with 1.3 ammonium acetate using a 1:8 soil to solution ratio and one hour shaking time. Clear extracts were obtained by filtering. Potassium was determined on a Coleman Model 21 flame photo- meter; Ca and Mg on a Perkin Elmer Model 290 or 303 absorp- tion Spectrophotometer using 1500 ppm La to suppress inter- fering ions. Water soluble Mg was determined in the same manner. Soil pH was determined in a 1:1 soil to water sus- pension with a glass electrode and a calomel reference cell. 2. Tissue Analysis Tissue samples were dry ashed according to the pro- cedure described by Jackson (1958). One gram of tissue was 3O ashed at 400-425° C for 15 hours. followed by cooling and the addition of 25 m1 of l‘N HNO The acid was evaporated 3. to dryness on a hot plate over a period of 2—3 hours. Reig- nition at 400° C for 10 minutes. dissolving in 25 ml of‘N HCl and filtering completed the digestion. Cations were de- termined on this solution after dilution to 100 m1. 3. Total Magnesium Analysis of Fractions Samples of soil fractions and of soils were decom- posed with HF as described by Jackson (1958). After drying at 110° C. the sample was weighed. moistened with a few dr0ps of water. then 0.5 ml of HC104 and 5 ml of Af were added. The decomposition was completed by evaporating the acids to dryness on a sand bath at a temperature of 200— 215°C. The residue was then taken up in HCl and Mg was de- termined on Model 303 Perkin Elmer Atomic Absorption Spec- trophotometer. 4. X-Ray Characterization of Soil and Clay Mineral Fractions Oriented clay specimens were prepared for x-ray dif- fraction by depositing 25-30 mg of material on ceramic plates. Specimens were then Mg saturated. glycerol solvated. 31 Diffraction patterns were made with a Phillips—Norelco X—ray unit using a c0pper source and nickle filter. Prior to heat treatments of 300° and 550° C. the Specimens were K saturated with l N KCl and washed free of chlorides. Samples were characterized prior to cropping and after the second crop. Silt sized fractions were Mg saturated. glycerol solvated and patterns were made on random powder samples before cropping only. 5. Statistical Analyses The analyses of variance for the intensive cropping study were made using a factorial analysis with 8 factors (soil) and 3 levels (Mg levels). For the pH-Ca level exper— iment factorial analyses with 3 factors (pH) and 2 levels (Ca) were used. 0n the cropping of soil fractions study. various combinations of factorial analyses were used due to the unequal replication. Tukey's honestly significant difference (HSD) as de— scribed by Steel and Torrie (1960) was used to test differ- ences between means for the first two experiments. The LSD (Steel and Torrie) was used for the fraction cropping study. 32 The HSD is similar to the LSD. but is more severe and probably gives a more accurate estimate of the signifi— cant differences when there are more than a few treatments or treatment combinations in an experiment. RESULTS AND DISCUSSION A. Intensive Cropping Study Three Mg levels were applied to seven soils of dif- ferent Mg—supplying characteristics and to a bentonite-sand mixture. Seven consecutive crops were grown without any, further additions of Mg. Yields of oat tops. tissue cation content and extractable cation levels of the soil and were measured for crops 2 to 7. The tissue from crop 1 was burned in the drying oven. 1. Growth Regponse to Applied Magnesium and Magnesium Uptake Yields were different between soils for all crops (Table 3). as could be expected since soils were selected for differences in Mg—supplying characteristics. Yield differences due to Mg application were ob— tained only on crops 4 to 7. inclusive. The different soils responded differently to applied Mg from crop to crop so that no consistent trends with response and yield differences were obtained. 33 34 Table 3.--Yie1ds of oat tops as affected by soil type and rate of applied Mg under intensive cropping. .L *Tukey's honestly significant difference. w— w Soil Initial Rate of —————————— Crop Number --------------- $011 Mg Applied 2 3 4 5 6 7 Level Mg ppm ppm -------------- g/pot ----------------- -Munising 43 0 6.20 11.71 6.44 3.76 5.08 2.40 fine sandy 10 6.23 11.81 11.08 3.98 5.54 2.68 loam (1) 20 6.58 12.07 12.75 4.54 6.13 2.68 Munising 30 0 4.95 10.02 8.53 3.13 5.41 2.29 fine sandy 10 5.40 11.19 10.90 4.95 6.17 2.60 loam (2) 20 5.20 11.21 11.06 4.51 6.93 3.01 Karlin ll 0 6.61 10.46 7.38 2.00 4.57 2.33 loamy sand 10 6.93 10.53 9.33 2.22 5.41 2.70 20 6.65 11.11 9.87 2.65 5.92 2.82 Montcalm 29 0 7.75 11.65 12.68 4.23 6.75 3.46 loamy ' 10 7.58 11.39 12.14 4.00 6.84 3.14 sand (1) 20 7.26 11.02 12.18 4.18 7.06 3.23 Montcalm 91 o 6.71 11.15 5.34 3.15 5.49 2.19 loamy 10 6.00 10.84 11.32 3.91 6.39 2.24 sand (2) 20 6.39 11.24 11.63 .4.01 6.98 3.29 Montcalm 93 0 7.10 11.07 4.59 3.18 6.00 2.62 loamy 10 7.57 11.20 11.14 3.13 5.60 2.50 sand (3) 20 7.42 11.09 11.64 3.70 6.18 3.15 Sims Clay 181 0 5.60 11.23 6.16 3.16 5.84 2.54 Loam 10 6.56 11.51 12.36' 2.93 6.18 2.49 20 6.81 11.86 12.31 3.47 5.92 3.05 Bentonite 185 0 7.23 9.98 10.38 2.62 4-60 2.36 sand 10 6.95 9.92 11.18 2.19 4.19 2.54 mixture 20 7.44 9.81 10.53 2.39 3.82 2.18 HSD 05* N. 2.18 1.25 0.70 HSD 01* 2.49 1.42 0.79 35 The Mg content of the oat tissue generally decreased as cropping progressed. except for the Sims soil and the bentonite-sand mixture (Table 4). Increases in the Mg con- tent of the plant tissue due to applied Mg tended to become relatively greater as cropping progressed. which indicated that the initially available Mg was becoming depleted. The two Munising soils responded differently to applied Mg. On crop 4. yields were increased when Mg was applied to the uncropped soil (Munising 1). and yields tended to be higher. but not significantly higher. on the following crops. On the cropped Munising (Munising 2). yields were increased by the 10 ppm increment of Mg on crops 4 and 6. and were further increased by the second increment of applied Mg (20 ppm) on crop 7. Consequently. yields were increased more by applied Mg on the cropped than on the un- cropped Munising soils. This would indicate a decrease in available soil Mg plus a depletion of applied Mg under in- tensive cropping. More of the applied Mg was removed by crOps 2 to 4 from the uncropped Munising. leaving less of the applied Mg for uptake on subsequent crops. These trends were reflected in the Mg content of the tissue (Table 4) and uptake (Table 5). However. the depletion of available Mg 136 Table 4.—-Mg content of oat tops as affected by soil type and rate of applied Mg under intensive cropping. Initial Rate of ----------------- Crop Number ----------------- Soil Soil Mg Applied ' Level Mg 2 3 4 5 6 7 ppm ppm --------------------- 96-.- --------------------- Munising 43 0 0.414 0.447 0.220 0.155 0.192 0.176 fine sandy 10 0.464 0.453' 0.243 0.150 0.231 0.218 loam (l) 20 0.467 0.454 0.236 0.181 0.266 0.245 Munising 30 0 0.465 0.371 0.235 0.179 0.204 0.240 fine sandy 10 0.512 0.434 0.224 0.183 0.281 0.286 loam (2) 20 0.482 0.478 0.266 0.208 0.300 0.334 Karlin 11 0 0.161 0.206 0.135 0.100 0.122 0.157 loamy 10 0.264 0.278 0.149 0.124 0.146 0.169 sand 20 0.321 0.291 0.156 0.149 0.174 0.204 Montcalm 29 0 0.307 0.251 0.170 0.145 0.189 0.216 loamy 10 0.408 0.334 0.224 0.160 0.231 0.279 sand (1) 20 0.486 0.348 0.239 0.192 0.247 0.290 Montcalm 91 0 0.631 0.511 0.324 0.350 0.316 0.352 loamy 10 0.620 0.559 0.310 0.369 0.419 0.395 sand (2) 20 0.625 0.551 0.312 0.400 0.521 0.560 Montcalm 93 0 0.456 0.390 0.295 0.321 0.282 0.370 loamy 10 0.487 0.444 0.294 0.324 0.332 0.369 sand (3) 20 0.500 0.223 0.318 0.363 0.441 0.468 Sims clay 181 0 0.406 0.338 0.248 0.308 0.412 0.456 loam 10 0.442 0.352 0.253 0.306 0.464 0.491 20 0.416 0.376 0.270 0.296 0.508 0.575 Bentonite 185 0 0.200 0.224 0.170 0.265 0.268 0.284 sand 10 0.200 0.208 0.178 0.278 0.271 0.341 mixture 20 0.233 0.367 0.186 0.302 0.296 0.355 HSD 05* 0.080 0.104 0.056 N.S 0.075 0.074 HSD 01 0.091 0.119 0.064 N.S. 0.085 0.084 *Tukey's honestly significant difference. 37 Table 5.--Uptake of Mg by oat tops as affected by soil type and intensive crOpping. rate of applied Mg under Initial Rate of ------------- Crop Number----—------ 8011 S011 Mg Applied 2 3 4 5 6 7 Level Mg ppm ppm -------------- mg/pot -------------- Munising 43 0 25.5 52.4 14.2 5.82 9.79 4.24 fine sandy 10 29.0 53.4 27.0 5.96 12.8 5.86 loam (1) 20 30.7 55.2 30.1 8.19 16.3 6.57 Munising 30 0 22.8 37.2 20.0 5.61 11.1 5.58 fine sandy 10 27.8 48.5 24.4 9.14 17.3 7.45 loam (2) 20 25.1 53.6 29.4 9.52 20.8 10.1 Karlin 11 0 10.6 21.5 9.90 1.98 5.59 3.67 loamy 10 18.3 29.5 13.9 2.74 7.84 4.57 sand 20 21.3 32.4 15.3 3.91 10.3 5.76 Montcalm 29 0 28.7 29.3 21.5 6.10 12.8 7.51 loamy 10 31.0 38.0 27.1 6.42 15.8 8.81 sand (1) 20 35.3 38.4 29.2 8.18 17.5 9.42 Montcalm 91 0 42.4 56.9 17.2 11.0 17.4 7.74 loamy 10 37.1 60.5 35.1 14.2 26.9 8.83 sand (2) 20 40.1 62.1 36.4 15.9 36.5 18.5 Montcalm 93 0 32.4 43.2 13.5 10.2 16.9 9.71 loamy 10 36.9 49.8 33.0 10.1 18.8 9.23 sand (3) 20 37.1 24.8 37.1 13.4 27.3 14.8 Sims clay 181 0 23.0 37.8 15.1 9.76 24.1 11.7 loam 10 29.1 40.6 31.3 8.99 28.8 12.3 20 28.3 44.7 33.3 9.82 30.3 17.6 Bentonite 185 0 14.6 22.3 17.6 6.89 12.3 6.79 sand 10 14.2 ‘20.6 19.9 5.93 11.3 8.61 mixture 20 17.5, 36.0 19.6 7.15 11.4 7.73 HSD 05* 9.2 14.3 7.4 3.9 7.1 0.35 HSD 01* N.S. 16.3 8.4 N.S. 8.1 0.40 *Tukey's honestly significant difference interaction. for soils X Mg rate 38 is not reflected in extractable Mg levels of the soil (Table 6). The levels of significance (Table 7) for Mg content and uptake further reflect the depletion of available Mg on these soils. A pattern similar to the uncropped Munising soil was obtained on the Karlin soil. Yields were increased by ap- pliedeg on crop 4. probably due to the high level of K (100 ppm) applied at seeding time. Other crops received 25-50 ppm K. Magnesium deficiency symptoms were present at the 0 ppm level of applied Mg on all crOps. but were eliminated at the 10 and 20 ppm levels. Magnesium content of the tis- sue was not increased by applied Mg until crop 7. These two facts suggest that the threshold level in the plant for de- ficiency symptoms and for a decrease in yield are not the same. Furthermore. these data indicate that the difference in Mg content between producing symptoms and eliminating them is very small. These data suggest that the critical level for deficiency symptoms is approximately 0.160%.Mg. The uptake response to applied Mg on crop 7 reflects the depletion of available Mg in this soil. The three Montcalm soils also responded differently to applied Mg. Yields on the first soil (Montcalm 1) were 39 Table 6.--Extractable Mg levels as affected by soil type and rate of applied Mg under intensive cropping. Initial Rate of _______ Cro Number* ________ Soil Soil Mg applied p 2 3 4 5 6 7 Level Mg PPm PPm ------------ PPm -------------- Munising 43 0 40 44 41 26 31 30 fine sandy 10 51 50 58 30 34 35 loam (1) 20 55 53 65 32 35 34 Munising 30 0 32 42 37 24 34 36 fine sandy 10 38 41 51 24 34 33 loam (2) 20 41 59 55 32 44 35 Karlin ll 0 19 21 12 12 17 23 loamy sand 10 26 25 18 14 14 25 sand 20 32 31 23 15 21 25 Montcalm 29 0 31 19 17 21 l4 l9 loamy sand 10 42 25 22 20 18 23 (l) 20 36 37 28 25 24 27 Montcalm 91 0 92 86 72 72 74 62 loamy sand 10 95 100 78 68 78 60 (2) 20 109 104 94 77 83 58 Montalm 93 0 91 96 82 81 88 81 loamy sand 10 94 102 80 79 87 75 (3) 20 102 109 90 82 94 78 Sims clay 181 0 186 183 171 174 197 217 loam 10 194 189 188 176 200 207 20 203 194 200 184 201 211 Bentonite 185 0 70 83 81 81 92 71 sand 10 63 80 91 86 99 78 mixture 20 81 84 108 92 102 92 *Value from soil sample prior to seeding. 41) Table 7.--Probabilitlcs for significance of dlfferences between means using Tukey's HSD fcr yields. Mq content. and Mq uplakv for the intensive cropping experiment. Yield Mg Content Mg Uptake . Crop . $011 ------------------------ Means Compared (Mg applied) --------------------- Number Ovle OstO lOstO Ovle OstO 10vs20 Ovle OstO lOstO -------------------------- Probability level(2)------—-——--—-------—----- Munising 2 us us us NS us NS us NS us fine 3 us NS us NS NS NS NS us us sandy 4 0.01 0.01 us 0.05 0.01 ms NS ns ns loam(1) 5 ns NS NS NS NS NS us 0.05 NS 6 NS NS ns 0.05 0.01 us NS 0.01 as 7 NS NS NS 0.01 0.01 0.01 0.01 0.01 0.01 Munising 2 NS NS NS 0.01 0.01 NS NS us NS fine 3 as us Ns NS NS us NS ns NS candy 4 0.05 0.01 NS NS NS us NS 0.01 NS loam(2) 5 NS NS NS as us NS NS 0.05 ms 6 NS 0.01 NS 0.05 0.01 NS NS 0.01 NS 7 NS 0.05 as NS 0.01 NS 0.01 0.01 0.01 Karlin 2 NS us NS NS NS NS NS 0.05 NS loamy 3 ns us as NS NS us ns NS NS sand 4 ns 0.01 NS NS us NS NS NS us 5 as ns ns ns NS NS NS us NS 6 NS 0.05 NS ms us as us us NS 7 as NS NS NS us NS 0.01 0.01 0.01 Montcalm 2 NS NS NS NS 0.01 ns NS ns “’ns loamy 3 NS us NS NS 0.05 NS NS us as sand(l) 4 NS NS NS 0.05 0.01 NS NS 0.05 NS 5 NS us NS NS NS NS NS us NS 6 NS NS NS us NS NS NS NS NS 7 us NS NS NS NS ns 0.01 0.01 0.01 Montcalm 2 NS NS us NS NS NS 0.01 0.01 0.01 loamy 3 us MS) as us NS as 0.01 0.01 0.01 sand(2) 4 0.01 0.01 NS NS NS NS 0.01 0.01 0.01 5 NS NS . NS us NS NS NS 0.05 0.01 6 NS 0.01 NS 0.01 0.01 0.01 0.01 0.01 0.01 7 NS 0.01 0.01 NS 0.01 0.01 0.01 0.01 0.01 Montcalm 2 NS NS us NS NS NS NS NS 'II NS loamy 3 NS NS NS 0.01 NS aand(3) 4 0.01 0.01 NS NS as as 0.01 0.01 us 5 us . NS NS us us N8 NS NS NS 6 NS NS us as 0.01 0.01 NS 0.01 0.01 7 NS NS us NS 0.01 0.01 NS . 0.01 0.01 Sims 2 NS ~ us NS NS us NS NS - NS NS clay 3 us NS NS NS us NS NS NS us 4 0.01 0.01 ms NS us NS 0.01 0.01 as 5 NS NS NS NS .Ns NS NS us as 6 NS NS NS NS NS NS NS ns ns 7 NS NS NS NS 0.01 0.01 0.01 0.01 0.01 Bentonite 2 us NS NS us as NS NS us ‘I as sand 3 NS us NS NS 0.01 0.01 us as us mixture 4 NS as us NS as us as as _ NS 5 NS ‘ us as as us NS NS NS NS 6 NS NS NS us NS NS NS us NS 7 NS NS NS NS 0.05 ns 0.01 0.01 as 41 not increased by applied Mg on any crop. although Mg uptake was increased by applied Mg in crop 7. Yields on the second soil (Montcalm 2) were increased over the check by the first increment (10 ppm) of applied Mg on crOp 4 only. but were to the second increment (20 ppm) on crops 4. 6. and 7 (Table 3). A yield response was obtained between the first and second increments of applied Mg on crop 7. Magnesium con- tent of the tissue and uptake also reflect this depletion of available Mg. On the third Montcalm (Montcalm 3) soil. yield in— creases due to applied Mg were obtained only on crOp 4. However. increases in Mg content and uptake from applied Mg reflect the depletion of available Mg from this soil. Differences between Montcalm 2 and 3 were due to the more rapid depletion of available and extractable-Mg (Table 6) on the second soil than on the third soil. This is sup- ported by the average amount of Mg released from these soils (Table 8) . On the other hand. these data do not suggest any ex- planation for this lack of response to applied Mg on the first soil; even though this soil had a lower extractable Mg level and an equal or greater measure of Mg release as com- pared to Montcalm 2 and 3. 42 Table 8.--Total Mg uptake and change in extractable Mg (mg/36009 soil) for crop 2 through crop 7 of the intensive cropping experiment. Rate of A Ex- Soil applied TOt:l** tractable Tg# d Mg* Upta e, Mg*** re ease PPm mg/pot mg/pot Munising 0 106 -36 70 fine sandy 10 134 —58 76 loam (l) 20 147 -76 71 Munising 0 102 14 116 fine sandy 10 135 -18 117 loam (2) 20 149 -22 127 Karlin 0 53 11 64 loamy sand 10 77 -4 73 20 89 -25 64 Montcalm 0 106 -43 63 loamy 10 127 -68 59 sand (1) 20 138 -32 106 Montcalm 0 153 r108 45 loamy 10 183 —126 57 sand (2) 20 210 -184 26 Montcalm 0 126 -36 90 loamy 10 158 -68 90 sand (3) 20 154 -83 71 Sims clay 0 121 112 231 loam 10 151 47 198 20 164 29 193 Bentonite- 0 80 4 84 sand 10 81 54 135 mixture 20 101 40 141 *0. 10. 20 ppm equal 0. 36 and 72 mg. Mg respectively. **Total uptake crops 2-8. Table 5. ***(Extractable mg crop 2-Extractab1e Mg crop 7)x3.6 (Table 6). #Sum of total uptake and A extractable Mg. 43 A yield response due to applied Mg was obtained only on the fourth crop on the Sims soil. This was due to the large K application at seeding. These data indicate that even though a soil may have a high Mg level a deficiency can be induced by a large K application. Yields on the bentonite-sand mixture were not in- creased by applied Mg. This "soil" released more Mg than did the other soils except the Sims. This indicates that the Mg in bentonite is readily available to plants. However. the growth on this "soil" was never as vigorous as on the other soils. probably because of poor physical conditions in the sand-mineral mixture. Even though yields on the Sims soil and the bentonite- sand mixture were not increased by applied Mg. the uptake response to applied Mg on crop 7 indicates that available Mg levels were becoming depleted. 2. Extractable Soil Magnesium and Release of Unextractable Mag- nesium During Cropping The level of extractable Mg in all soils tended to decrease as cropping proceeded at all three soil Mg levels. except for the Sims soil and the bentonite-sand mixture. on which extractable Mg increased from crOp 2 to crop 7 (Table 6). 44 The level of extractable Mg increased from 2 to 24 ppm on nearly all soils between crOps 5 and 6. probably be- cause the soils were allowed to dry completely between these crops. The Sims soil released the most Mg during this per— iod. while Montcalm 1 actually fixed Mg against extraction with ammonium acetate. The decrease in extractable Mg from crop 4 to 5 was probably caused by K application (100 ppm) prior to seeding erOp 4. The algebraic sum of the change in extractable Mg between crop 2 and crop 7 and total uptake reflects the Mg released by these soils (Table 8). Averaging across Mg levels for each soil gives a measure of the relative supply- ing capability of these soils. The order was as follows: Sims clay loam >> Munising fine sandy loam 2> bentonite-sand mixture 2 Montcalm loamy sand 32 Montcalm loamy sand lg'Muni- sing fine sandy loam 12 Karlin loamy sand 2 Montcalm loamy sand 2. The data in Tables 6 and 7 suggest that Mg applied to the soils was fixed against extraction by ammonium ace- tate. However. the "fixed" forms were "released" for uptake by subsequent crops grown on these soils. This release was not reflected in the extractable Mg levels during the 45 cropping sequence. except when the soil was allowed to dry between crops 5 and 6. The mechanism of Mg fixation is not clear from the literature. However. it has been suggested that it occurs through coordination at the broken edges of the octahedral layer (Cheminade and Drouineau. 1936). A second mechanism would be the trapping of Mg ions perforations in the surface of oxygen layers due to collapse of the sheets when K is applied in large quantities. If Mg were fixed by either of these mechanisms. it could be available for plants through further "weathering” of the soil. 3. Linear Correlations of Plant Tissue Magnesium Content and Uptake Correlated with Extract- able Magnesium. Potassium. and Calcium Levels. As cropping proceeded. the correlation between ex- tractable Mg and Mg content and uptake became higher (Table 9). This indicates that the ammonium acetate extraction was more accurately predicting Mg availability of these soils as crepping progressed. However. using Mg uptake as the best indicator of Mg availability. only 32% of the variability in uptake was accounted for by the relationship between soil Mg 46 .Hm>ma occncamacman Ho.o:. .Ho>ma mocmoamacmam mo.oa nma.01 amom~.ou aaoam.01 a¢NN.OI anmm.0t omo.ou m=\mo manmuomuuxm samoo.OI asfiom.OI mamm®.0t armo¢.OI ¥%ONM.OI aaaom.OI mS\M.0HQNUUMHuNw mao.o mmo.o: sawm.o aoa~.o ¢HO.OI mom.o M manmuomuuxm «smom.o snomm.o aamo¢.o samom.o hma.o .amvm.o m2 manmuomuuxm h o m e m m . IIIIIIIIIIIIIIIIIIIIIII Honfisz QOHUJIIIIIIIIIIIIIIIIIIIIIIIIIIII oaflmaum> osmosomoUGH mmoa Dmo >9 Au0m\mav oxmumb m2 mahwm.OI ammmv.OI asehh.01 mammm.OI mammm.0l OHo.o m2\mo maflmuowuuxm ..0ss.ou .4mms.ou :.Hms.os 4.006.61 .4mmm.ou .ssmm.ou mz\m oaoncccacxm mma.o ama.01 sammm.o mamam.o nmo.ou ova.o M oaflmuomnuxm asmmhoo asmmn.o mah®0.o *«omm.o NNH.O ammm.o . m8 OHQMMUMHDNW h o m e m N IIIIIIIIIIIIIIIIIIIIIIII “09852 momUIIIIIIIIIIIIIItIIIIIIIIIIIII oHQmHHm> DamocmmmUSH mmoa umo Axg acousoo m2 .moanmanm> unoccomocsH msoflum> mo cofluossm o no mono some now Esflmocmma mo oxmums can ucoucoo m2 How AHV musofloawmooo soaumaonuoo Hmmsflqil.m magma 47 and uptake on crop 7. A lower correlation was obtained on other crOps. Even so. this correlation was better than that reported by Wehunt and Purvis (1954) who found no relation— ship between soil.Mg and Mg content of apple leaves. .The better correlation in this study was attributed reduction in variation by greenhouse control and the more accurate prediction of Mg availability by the ammonium acetate ex- traction than by the electrodialysis used in their study. The correlation between Mg uptake and the soil K:Mg ratio was higher than with the extractable soil Mg alone. Even still this correlation accounted for less than 40%.of the variation of the uptake data. However. the correlation between Mg content and soil K:Mg was higher than the corre- lation with uptake. Extractable Ca/Mg did not correlate with Mg uptake as well as did K/Mg. On crops 2 and 7 there was no correla- tion between soil Ca/Mg and uptake and r values of -0.224 to —0.510 were obtained on the remaining crops. Extractable -K/Mg correlated with uptake on all crops and accounted for more variability of the uptake of Mg. This indicates that the Ca/Mg antagonism on Mg uptake is less severe than the K-Mg antagonism. Secondly. Ca does not sterically hinder Mg release from clay minerals to the extent that K does. 48 There is no relationship between soil K and. g up- take which is contrary to the findings of Wehunt and Purvis (1954). This was attributed greenhouse control and differ— ent K fertilization rates between soils. The smaller corre- 1ation between Mg content and extractable Mg on the cropped Munising than the uncropped was attributed to Mg being fixed in forms which were not extracted with ammonium acetate. but were available to plants (Table 10). This supports the re- sponse pattern discussed earlier. Correlations between these variables in the Karlin soil were as high as for the Munising 1. Extractable-Mg and Mg content of the tissue correlated better on this soil than on Munising. "Fixed" forms were evidently more readily ex- tractable on this soil than on the other soils. On Montcalm 1 there was lower correlation between uptake of Mg and soil K/Mg and Ca/Mg than on Montcalm 2 and 3. This was borne out by the lack of reSponse on this soil. Magnesium content of the tissue was not correlated with ex— tractable Mg on Montcalm 3. Soil K/Mg level exerted the greatest influence on Mg uptake for both the bentonite-sand mixture and the Sims soil. However. only 10% of the variability in uptake was accounted 49 Table 10.—-Linear correlation coefficients (r) for Mg content and uptake of Mg for each soil as a function of various independent variables. Soil and Independent Variable Mg Content of Oat Tops Mg Uptake by Oat Tops Munising fine sandy loam (l) Munising Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K fine sandy loam (2) Extractable Extractable Extractable Extractable Karlin loamy sand Montcalm Montcalm Montcalm Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K M9 K/Mg Ca/Mg K loamy sand (1) Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K loamy sand (2) Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K loamy sand (3) Extractable Extractable Extractable Extractable Sims Clay loam Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K Mg K/Mg Ca/Mg K Bentonite-sand mixture Extractable Extractable Extractable Extractable M9 K/Mg Ca/Mg K 0.513 -0.786 —0.618 -0.801 0.317 -0.810 —0.545 -0.869 0.816 -0.760 -0.704 -O.764 0.750 -O.750 -0.641 -0.651 0.512 -O.759 -0.696 -0.700 0.142# —0.522 -0.163# —0.s35 0.677 -0.636 —0.360 -0.543 0.085 0.137 0.271 0.141 4=#=# a: 0.643 -0.649 —0.649 —0.523 0.652 —0.615 -0.713 -0.464 0.620 -0.614 -0.510 -0.647 0.567 —0.471 -0.364 -O.345 0.774. —O.783 -0.597 -0.504 0.607 -0.645 -O.385 —0.555 0.089# -0.353 0.171# —0.342 -0.020# —0.484 —0.351 -0.399 tNon-significant Significant at 0.05 level. all others highly significant. 50 for by this relationship on the Sims soil and 23% for the bentonite-sand mixture. The above data indicate that this extractant does not accurately predict the availability of Mg for plant up- take. Considering all soils. the extractable K/Mg or Ca/Mg ratio better predicts the uptake of Mg by plants than does extractable Mg alone. Extractable K correlated better with Mg uptake for individual soils than it did for all soils (Table 9). This was attributed to different K rates applied to each soil to establish a certain level of K. Thus. the relationship be- tween K and Mg was less variable within a soil than for all soils. The multiple correlation coefficient for plant Mg content as a function of plant Ca and K content was 0.493 for all soils and all crops. The associated partial corre- lation coefficients were -0.457 and 0.394 for K and Ca. re- spectively. All are highly significant. The partial corre- lation coefficients indicate that Ca and Mg content of the plant tissue vary directly with each other. while Mg content varies inversely with K content. Since for this correlation uptake and tissue content are the same. this supports the 51 content an antagonism between K and uptake of Mg. Converse- ly. the positive correlation between Mg and Ca indicates a certain dependency of Mg uptakes on the amount of Ca uptake. The linear correlation coefficients for extractable Mg. K:Mg and Ca:Mg and uptake of Mg by plants for all soils and crOps were 0.243. -0.418. and —0.313. respectively. This further indicates that cation levels extracted by ammonium acetate do not accurately predict Mg availability by plants. B. pH-Calcium Level Study Treatments were composed of two calcium levels at each of three pH levels on Karlin loamy sand. Eight suc— cessive crops of oats were grown; yields and cation contents of the oat tops from each crop were measured. Extractable cation levels and pH were measured on soil samples from each treatment before and after each of the eight crOps. 1. Yields Yields were higher for the higher pH levels (sz and pH3) than for the more acid level (le) for all cr0ps (Table 11). but there was not a consistent relationship between the yields of the two higher pH levels. These relationships are 52 Table 11.--Yields of oat t0ps as affected by pH and Ca levels under intensive crOpping. Analyzed statistically as a 3x2 factoria1.* Treatments** -------------- CrOp Number ________________ l 2 3 4 5 6 7 8 ----------------- -gms/pot----------------------- le Cal 11.09 5.38 4.68 4.15 5.99 5.93 4.42 2.39 le Ca2 12.22 5.75 4.27 4.43 6.30 6.94 3.76 2.76 pH2 Cal 13.70 5.59 5.55 6.97 7.99 6.28 5.93 2.40 pHZCa2 14.28 5.96 5.74 6.84 8.62 7.80 6.21 3.17 pH3Ca2a 13.79 7.01 5.26 6.21 8.33 6.99 6.00 2.87 pH3Ca2b 14.28 6.56 5.19 6.90 8.71 7.67 6.53 2.78 Check 13.34 4.68 5.19 4.74 6.27 4.49 4.85 2.14 Effect ------------------- Effect Means*** ------------- le 11.65 5.56 4.48 4.29 6.15 6.43 4.09 2.58 sz 13.99 5.77 5.65 6.91 8.31 7.04 6.07 2.78 pH3 14.03 6.79 5.23 6.56 8.52 7.33 6.27 2.83 Ca2 13.59 6.09 5.07 6.06 7.88 7.47 5.50 2.90 Cal 12.86 5.99 5.16 5.79 7.44 6.40 5.45 2.55 pH HSD 05# 0.61 0.29 0.20 0.25 0.27 0.19 0.29 N.S. pH HSD 01# 0.79 0.38 0.26 0.32 0.35 0.25 0.38 N.S. Ca HSD 05# N.S. N.S. N.S. N.S. N.S 0.11 N.S. 0.10 Ca HSD 01# N.S. N.S. N.S. N.S. N.S. 0.14 N.S. 0.14 *Omitting check treatment and pooling Ca2a in Ca1 level. **pH.=4.3. pH2=5.5. pH3=6.5. Cal=Low. Ca2=high. Ca2a=CaCO3. Ca2b=Ca(OH)2. see Table 2. ***pHxCa level interaction was nonsignificant. #Honestly significant difference. 53 reflected in the Mg uptake data (Table 13). but not in the ‘Mg content data (Table 12). On crOps 6 and 8. yields were higher for the high Ca level (Caz) than for the low Ca level (Cal). A decrease in soil pH for pH3 was attributed to the cause of this effect (Table 14). A decrease in differences due to dif- ferent pH levels allowed the Ca effect to become more dom- inate. Increased yields at the higher pH levels were due to more favorable soil-plant relationships at pH 6.5 than below pH 6. Nutrients are more available. fewer toxicities occur and less H-ion injury to plant tissue occurs in this pH range than at lower pH levels. When soil acidity in- creased after crOp 5. the higher Ca level probably minimized the injurious effect of the H-ion (Rains et al.. 1964). 2. Soil pH Soil pH values established by means of soil amend- ments (pH2 and pH3) were consistent for crops 1-5. inclusive (Table 14). Then the pH of these treatments decreased. tend— ing to approach the initial pH of the soil. However. the effect of the amendments could be detected in the final pH 54 Table 12. Mg content of oat tops as affected by pH and Ca levels under intensive cropping. Analyzed statistically as a 3x2 factorial.* Treat- ------------------------ Crop Number -------------------- ments** 1 2 3 4 5 6 7 8 ____________________________ %--__---___-__-______---_-_ le Cal 0.076 0.167 0.176 0.238 0.148 0.085 0.107 0.113 le Ca2 0.070 0.134 0.144 0.234 0.147 0.095 0.114 0.115 pH2 Cal 0.095 0.204 0.214 0.182 0.100 0.073 0.102 0.100 sz Ca2 0.101 0.220 0.244 0.196 0.094 0.069 0.104 0.094 pH3 Ca2a 0.072 0.131 0.156 0.146 0.090 0.079 0.123 0.113 pH3 Ca2b 0.103 0.190 0.232 0.175 0.097 0.072 0.130 0.108 Check 0.091 0.204 0.210 0.230 0.124 0.094 0.109 0.122 HSD.05# 0.008 0.012 0.026 0.012 N.S. 0.055 N.S. N.S. HSD 01# 0.010 0.015 0.033 0.015 N.S. 0.069 N.S. N.S. *Omitting check treatment and pooling Ca2a in Cal level. **le=4.3. pH2=5.5. pH3=6.5. Cal=1ow. Ca2=high. Ca2a=CaCO3. Ca2b=Ca(OH)2; see Table 2. #Honestly significant differences for pH x Ca level interaction. 55 Analyzed Table 13.--Mg uptake by oat tOpS as affected by pH and Ca levels under intensive crOpping. statistically as a 3x2 factorial.*, Treat- ------------------- CrOp Number-—--—4 ----------- ** ments 1 2 3 4 5 6 7 8, --------------------- mg/pot---——----—-——------- le Cal 8.57 8.81 8.27 9.86 8.89 5.04 4.76 2.71 le Ca2 8.59 7.71 6.13 10.4 9.29 6.58 4.30 3.18 sz Cal 13.0 11.4 11.9 12.7 8.04 4.60 6.02 2.40 pH2 Ca2 14.4 13.1 14.0 13.4 8.08 5.41 6.43 3.00 pH3 Caza 9.90 9.20 8.21 9.10 7.52 5.51 7.37 3.27 pH3 Ca2b 14.7 12.4 12.0 12.1 8.47 5.54 8.52 3.02 Check 12.1 9.46 11.0 10.9 7.79 4.21 5.24 2.59 HSD 05# 1.79 0.96 1.71 0.60 N.S. 0.38 N.S N.S. HSD 01# 2.23 1.20 2.13 0.79 N.S. 0.47 N.S N.S. * Omitting check treatment and pooling Ca a in Cal level. ** = I =.: = I =I a le pH2 5.5 pH3 6 5 Cal low Ca2 high Ca2a =CaCO 2b =Ca(0H)2; see Table 2. # Honestly significant differences for pH x Ca level inter- action. 56 Table 14.--Soi1 pH as affected by pH and Ca level under intensive cropping. m Treat- ---------------------- CrOp Number* ----------------- ments** 1 2 3 4 5 6 7 8 Final le Cal 4.3 4.3 4.5 4.6 4.5 4.4 - 4.2 4.3 le Ca2 4.1 4.2 4.2 4.2 4.3 4.2 - 4.2 4.6 pH2 Cal 5.4 5.5 5.8 5.6 5.5 5.2 - 4.8 4.8 pH2 Ca2 5.3 5.4 5.4 5.3 5.2 4.9 - 4.8 4.7 pH3 Ca2a 6.6 6.6 6.8 6.6 6.3 6.0 - 5.8 5.3 pH3 Ca2b 6.2 6.2 6.5 6.3 6.1 5.8 - 5.6 5.2 Check 4.5 4.8 4.9 4.8 4.5 4.5 - 4.5 4.3 * Value on samples taken prior to seeding. e.g.. crop 1 is initial value prior to first seeding. ** = = = = = ° , le 4.3. pH2 5.5. pH3 6.5. Cal low. Ca2 high Ca = CaCO . Ca Ca(OH)2; see Table 2. 2a 3 2b determination. Extractable Ca (Table 15) did not decrease during this period. but extractable K and Mg (Tables 16 and 17) decreased on pH levels 2 and 3. Table 15.-~Soil Ca levels as affected by pH and Ca level under intensive cropping. Treat- ----------------------- CrOp Number --------------- ments** 1 2 3 4 5 6 8 --------------------------- ppm---------—-----—-— le Cal 184 276 350 265 221 267 359 le Ca2 957 1157 1362 1141 1154 947 1067 sz Cal 644 644 569 700 589 479 644 pH2 Ca2 975 993 883 901 901 773 948 pH3 Ca23 1030 957 1214 1067 1067 837 956 pH3 Ca2b 833 1288 1040 791 791 746 821 Check 239 368 221 276 213 249 249 * Value on samples taken prior to seeding. ** le = 4.3. pH2 = 5.5. pH3 = 6.5. Ca = low. Ca = high. Ca2a = CaCO3 Ca2b = Ca(OH)2; see Table 2. 58 Table l6.--Soil Mg levels as affected by pH and Ca level under intensive cropping. Treat- - ------------------ Crop Number* ---------------- ments** 1 2 3 4 5 6 7 8 Final ---------------------- ppm--------------——----— le Cal 20 20 20 l6 l9 l7 - 24 24 le Ca2 20 20 20 16 21 18 - 23 27 sz Cal l4 16 12 9 7 9 - 12 12 sz Ca2 16 14 12 8 10 8 - 9 8 pH3 Ca2a 10 12 6 9 8 6 - 8 8 pH3 Ca2b 17 16 18 8 11 8 - 9 6 Check 18 20 20 14 ll 16 - 19 21 * Value on samples taken prior to seeding. ** = I = . I = . I = I = ' I le 4.3 pH2 5 5 pH3 6 5 Cal low Ca2 high Ca = Ca C0 . Ca = Ca(OH) 2a 3 2b see Table 2. 2? 59 Table l7.--Soil K levels as affected by pH and Ca level under intensive cropping. Treat- -------------------- Crop Number* ------------- ments** 1 2 3 4 5 6 7 8 ------------------------ ppm-----——--—--—----- le Cal 106 76 30 22 12 20 - 66 le Ca2 105 75 30 24 ll 20 - 68 sz Cal 102 70 28 26 18 20 - 27 sz Ca2 102 68 22 21 18 20 - 26 pH3 Ca2a 100 68 26 30 ll 22 - 31 pH3 Ca2b 102 77 23 28 19 22 - 30 Check 98 70 30 28 14 17 — 72 * Value on samples taken prior to seeding. ** = I =.I =.I = I le 4.3 pH2 5 5 pH3 6 5 Cal low Ca2 Ca2a = CaC03. Ca2b = Ca(0H)2; see Table 2. 3. Soil Magnesium Levels and Magnesium Uptake At the lowest pH level and on the check treatment. extractable Mg increased during the crOpping period. But. at the two higher pH levels (pH2 and pH3) extractable Mg de- creased during the same period (Table 16). Uptake of Mg was similar on both Ca levels at the two higher pH levels. except 60 for pH3Cala for all crops. However.ng release to the ex- tractable form during the cropping sequence generally in- creased with soil acidity (Table 18). Hossner (1965) re- ported that the rate of release of Mg from clay minerals in- creased with increasing acidity. In the present experiment. more Mg was released to the extractable form at pH 5.5 than Table 18.-—Total uptake of Mg and A—Extractable Mg for the pH-Ca experiment. T.......... 332:2" 3333532332121.) 1.23222... mg7pot mg/pot mg/pot leCa1 57 14 71 le Ca2 56 25 81 sz Cal 70 -7 63 sz Ca2 79 -14 65 pH3 Ca2a 60 -7 53 pH3 Ca2b 70 -4O 30 Check 63 ll 74 * le = 4.3. pH2 = 5.5. pH3 = 6.5. Cal = low. Ca2 = High. Ca2a = CaCO3. Ca2b = Ca(0H)2; see Table 2. ** Total uptake crops 1-7. Table 12. # (Final extractable Mg minus extractable Mg crop 1) x 3.6 (Table 16). ***Sum of total uptake and A-extractable Mg. 61 at higher pH levels. even though Mg uptake was essentially the same on all treatments. The greater release is reflected by higher extractable Mg levels after cropping. Release of Mg to an extractable form was greater for the pH3Ca2a (CaC03) treatment than for pH3Ca2b treatment (Table 18). But. total uptake was 15% less for crops 1—8. inclusive. The data presented here suggest no explanation for these differences and indicate the need for more defini— tive research to study these differences. 4. Linear Correlations of Plant Tissue Magnesium Content and Uptake Correlated with Extract- able Soil Magnesium. Potassium. and Calcium Levels. The poor correlations obtained between Mg uptake and content with extractable Mg. K/Mg and Ca/Mg (Table 19) are probably the result of the dominate effect of soil acidity. Extractable Mg was correlated with Mg content of the tissue on crops 4. 5. and 6. but not with uptake. The plants were unable to fully utilize available Mg at the lower pH levels (le). resulting in a negative correlation between Mg uptake by plants and extractable soil Mg. At the higher soil pH levels (which also had a lower extractable Mg 62 .mocmoHMHcmHm mo Hm>ma Ho.0¥¥ .muamoflmucmum mo Hm>ma mo.o. mmN.OI «ww¢.o aaanh.o hwo.o 5H0.OI mm~.o ona.o mo wanmuomnuxm «mm.o mam.o mv~.ou mam.o mom.ou .mmm.o mao.ou mz\mo manmuomuuxm mma.o ooo.o mom.0I omm.o mmo.o: Ho¢.o moo.o: mE\M mHQmuomuuxm mmo.on ovo.o vmm.c «mm¢.0l mma.ou «somv.01 HON.OI m2 manmuomuuxm m m m w m m H nnnnnnnnnn inluualllnlllalnumnasz monolllllanllanallnIaIInIIIIIII: mHQMAHm> ucmocmmwocH mmoa umo >9 Anom\mEv mxmumb m: Hmm.o mua.o: mmm.0| www.01 Hmm.ou ma~.on NHO.OI mo manmuomuuxm 0mm.o *Hmv.01 remoh.OI «*wwn.ou mmm.0| mma.ou mma.ou mz\mo manmuomnuxm nma.c aawam.0| *«va.OI «anom.0| Hmm.01 mvH.OI mmH.OI m£\x manmuomuuxm moa.o *ammn.o **mmm.o 44mmm.o vao.on mao.on mmo.OI m2 manmuomuuxm m o m v m m H IIIIIIIIIIIIIIIIIIIIIIIIIIII Hmflfisz QOHUIIIIIIIIIIIIIIIIIIIIIIIIIII maflmflum> unmocwmmocH mmoa umo mo ARV ucmucou m2 .ucwE Iflummxm Hm>oa moumm map so CBOHm mmouu How mmaQMHum> ucmosmmmocfl msoflum> mo cofluocsm m mm mxmums cam pamucoo m2 MOM any mucmaoflmmwoo coflumamnuoo Hmmcflalloma magma 63 level plant growth was greater. which resulted in more total uptake. but a lower Mg content (%) in the plants. Extractable K/Mg ratio was correlated with Mg con- tent on crops 4. 5. and 6. but not on any of the other crops or with uptake. Again. the lack of correlation is related to greater growth at the higher pH. Extractable Ca/Mg ratio was correlated negatively with Mg content. but was not correlated with uptake. These results do not agree with the data presented in the preced- ing section on the intensive cropping experiment. At the high pH levels. the increased plant growth resulted in a lower concentration of Mg in the plants. while at low pH levels the inverse was true. Also. a high soil Ca level (wide Ca:Mg ratio) was associated with the high pH level. C. Cropping_of Soil Fractions A sufficient quantity of five fractions of different particle size from seven soils. three fractions from benton- ite and two from chlorite to supply 20 mg of Mg was mixed with quartz sand and cropped to two crops of oats. Yield and tissue Mg content were measured on each of two successive crops. The content of Mg was measured in each 64 fraction before and after each crop. The mineralogical composition of each fraction was determined by means of X—ray diffraction before the first crop and after the second crop. The quartz sand used in the cultures contained 0.1% material less than 50 u in size which had a total Mg content of 0.4%. Approximately 2 mg of Mg was added to each culture from this source. 1. Magnesium Content of Plant Tissue Magnesium content of the oat tops was less in the second crop than in the first crop (Table 20). Oats grown on the finer fractions contained more Mg than those grown on the coarser fractions. The relatiVe Mg content of the oats between fractions varied from soil to soil for crop 1. For crop 2. it was the same from soil to soil (lack of soil by fraction interaction). following more closely to fraction size; tissue from the finer fractions having a higher Mg content. 65 Table 20.--Mg content of fractions. Mg uptake by oat tOps and Mg content of oat tops as affected by soil type and particle size. Mg Content Mg Uptake Mg Content in SOil Fractions by Oats of Oat Tops Soil Fraction Size Before After After Crop Crop Crop Crop Cropping CrOp 1 Crop 2 1 2 1 2 ------ .----—— —-——---—-----%--—--—------ —mg/cu1ture— ------%—---—-- Munising <0.08 1.15 1.08 0.769 1.65 1.45 0.129 0.0987 fine sandy 0.08-0.2 1.44 1.30 0.928 1.86 1.19 0.134 0.0877 loam(1) 0.2—2 0.699 0.637 0.498 1.77 1.20 0.138 0.0857 2—20 0.144 0.130 0.075 1.90 1.19 0.103 0.0833 20-50 0.0714 0.0645 0.0390 1.94 1.19 0.141 0.0856 Total Soil 0.0786 0.0725 - 1.59 - 0.128 — Munising <0.08 1.03 0.935 0.433 1.85 1.36 0.136 0.0977 fine sandy 0.08—0.2 1.58 1.42 0.899 1.92 1.25 0.141 0.0853 loam(2) 0.2-2.0 1.11 1.01 0.879 1.80 1.13 0.130 0.0880 2-20 0.267 0.242 0.208 1.87 1.05 0.131 0.0777 20-50 0.136 0.124 0.120 1.80 1.18 0.135 0.0853 Total Soil 0.121 0.112 - 1.54 - 0.118 - Karlin <0.08 1.10 1.01 0.145 1.61 1.23 0.124 0.0950 loamy 0.08-0.2 1.17 1.06 0.561 1.82 1.27 0.136 0.0843 sand 0.2—2.0 0.986 0.901 0.747 1.71 1.10 0.129 0.0807 2-20 0.314 0.286 0.240 1.73 1.17 0.130 0.0830 20-50 0.196 0.182 0.109 1.46 1.04 0.103 0.0773 Total Soil 0.126 0.116 - 1.52 - 0.123 — Montcalm (0.08 0.844 0.772 0.422 1.70 1.30 0.130 0.0880 loamy 0.08—0.2 1.10 0.990 0.669 1.98 1.20 0.143 0.0833 sand(1) 0.2-2.0 '0.993 0.907 0.742 1.78 1.23 0.126 0.0827 2.0—20 0.336 0.307 0.254 1.75 1.13 0.125 0.0797 20-50 0.205 0.186 0.149 1.81 1.08 0.134 0.0753 Total Soil 0.184 0.169 - 1.55 - 0.124 — Montcalm <0.08 0.828 0.749 0.538 1.88 1.34 0.143 0.0850 loamy 0.08—0.2 1.14 1.04 0.738 1.85 1.21 0.136 0.0847 sand(2) 0.2-2.0 1.08 0.993 0.696 1.62 1.09 0.126 0.0747 2.0-20 0.338 0.308 0.268 1.75 1.16 0.126 0.0817 20-50 0.194 0.178 0.132 1.71 1.03 0.128 0.0730 Total Soil 0.184 0.169 - 1.54 - 0.122 - Montcalm <0.08 0.943 0.863 0.397 1.70 1.31 0.122 0.0850 loamy 0.08-0.2 1.26 1.14 0.733 1.81 1.14 0.138 0.0797 0.2-2.0 0.970 0.882 0.786 1.83 1.04 0.128 0.0817 2.0-20 0.325 0.298 0.253 1.70 1.14 0.121 0.0763 20-50 0.200 0.182 0.139 1.76 1.07 0.119 0.0770 Total Soil 0.132 0.121 - 1.56 - 0.121 — Sims clay <0.08 1.24 1.13 0.713 1.80 1.20 0.131 0.0860 loam 0.08-0.2 1.44 1.31 0.569 1.78 1.10 0.129 0.0767 0.2—2.0 1.24 1.13 0.920 1.78 1.17 0.129 0.0833 2.0-20 0.776 0.711 0.549 1.64 1.24 0.120 0.0857 20-50 0.762 0.699 0.609 1.69 1.10 0.137 0.0803 Total Soil 0.590 0.543 - 1.59 - 0.136 - Bentonite <0.08 1.56 1.43 0.794 1.67 1.21 0.125 0.0843 0.08—0.2 1.48 1.37 0.779 1.69 1.21 0.133 0.0847 0.2-2.0 1.12 1.02 0.903 1.71 1.07 0.129 0.0753 Chlorite 0.08-0.2 3.18 2.87 1.13 1.91 1.38 0.133 0.0980 0.2-2.0 3.49 3.17 1.63 1.96 1.37 0.146 0.0927 66 2. Magnesium Uptake and Availa— bility from Different Fractions The finer fractions supplied more Mg than did the coarser fractions. Based on the total uptake for both crops (Table 21). the following relative order of supply (release from the fraction) was obtained for all soils: <0.08 = 0.08-0.2>0.2-2.0 = 2.0-20320-50 u> total soil. Table 21.—-Means of Mg uptake by oat tOps and.Mg content of oat t0ps as affected by fraction size. Fraction Mg Uptake by Mg Content of Size Oat Tgps Oat T0ps Crop 1 Crop 2 Total Crop 1 CrOp 2 —--u ----------- mg/culture---—- ----——-—% --------- <0.08 1.74 1.31 3.05 0.130 0.0908 0.08—0.2 1.86 1.19 3.05 0.137 0.0831 0.2-2.0 1.76 1.14 2.90 0.129 0.0824 2.0—20 1.76 1.15 2.91 0.126 0.0810 20-50 1.74 1.10 2.84 0.128 0.0791 LSD 0.05 0.15 0.10 — 0.008 0.008 0.01 0.20 0.13 - 0.011 0.010 67 Doll et al. reported the clay fraction supplied more K to wheat crops than did the silt fraction. Large amounts of K were removed from the coarse clay (0.2-2.0 u). The dif— ferential release was attributed to decrease in activity of interlayer K when the K content of the clay decreased. This was confirmed by a positive correlation between fraction K content and logrithm of K uptake. In the present study. Mg content of the soil frac— tions correlated with uptake on the coarse clay fraction only (Table 22). Table 22.--Linear correlation coefficients (r) between up— take of Mg and Mg content of fractions at start of each crop. Fraction Size Crop 1 Crop 2 u <0.08 —0.271 -0.428 0.08-0.2 0.111 0.512** 0.2-2.0 0.496** 0.638** 2.0-20 —0.511* 0.319 20-50 -0.l43 -0.058 Total Soil 0.130 - *0.05 level of significance. **0.01 level of significance. Each fraction had a different number of observations. 68 The lack of correlation between fraction Mg content and Mg uptake for the silt fraction and the fine and medium fractions may be explained in two ways. First. Mg release is governed by factors other than those for K release and second. Mg is present in different forms in these fractions. For example. the dominant mineral Species which would supply Mg from the coarse clays was the interstratified vermiculite- chlorite—illite systems or discrete minerals belonging to these species (Table 23). The medium clay fraction contained similar mineral species. However. the fraction Mg content correlated with uptake only on the second crop. This sug- gests that Mg on the crystal edges was easily released for plant uptake. When this was removed by the first crop Mg release was governed by the same factors as for the coarse clay. On the fine clay fraction. the lack of correlation between fraction Mg content and Mg uptake can be explained on the same basis. Magnesium released had shorter diffusion paths to move through making it more readily available. Secondly. the larger surface area allowed more of the Mg to be at the crystal edge and concurrently more surface was ex- posed for weathering and release. 69 Table 23.-—Minerals present in various particle size frac- tions of seven Michigan soils. r _ Soil Particle Predominant Minerals Size 0 Munising 0.08-0.2 Kaolinite. illite. ordered interstrati- fine fied vermiculite-chlorite-montmoril- loam(1) lonite system 0.2-2.0 Kaolinite. illite. vermiculite 2.0-20 Quartz. feldspars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite Munising 0.08-0.2 Kaolinite. interstratified chlorite- fine vermiculite—illite system sandy 0.2-2.0 Kaolinite. illite. interstratified loam(2) chlorite—vermiculite system 2.0-20 Quartz. feldspars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite Karlin 0.08-0.2 Kaolinite. interstratified illite- loamy vermiculite-chlorite system sand 0.2-2.0 Kaolinite. illite. chlorite. inter- stratified vermiculite-chlorite system 0.2—2.0 Kaolinite. illite. chlorite. inter- stratified vermiculite-chlorite system 2.0-20 Quartz. feldspars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite Montcalm 0.08-0.2 Kaolinite. interstratified vermiculite- loamy chlorite system sand(l) 0.2-2.0 Kaolinite. illite. interstratified vermiculite-chlorite system 2.0-20 Quartz. feldspars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite 70 Table 23. (Continued) Soil Particle Predominant Minerals Size 0 Montcalm 0.08-0.2 Kaolinite. illite. montmorillonite. loamy interstratified vermiculite— sand(2) chlorite system 0.2-2.0 Kaolinite. illite. interstratified vermiculite-chlorite system 2.0-20 Quartz. feldspars. goethite. boehmite 20—50 Quartz. feldspars. goethite. boehmite Montcalm 0.08—0.2 Kaolinite. illite. interstratified loamy vermiculite-chlorite system sand(3) 0.2—2.0 Kaolinite. illite. interstratified vermiculite-chlorite system 2.0-20 Quartz. feldSpars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite Sims clay 0.08-0.2 Kaolinite. illite. interstratified loam illite-vermiculite—chlorite system 0.2-2.0 Kaolinite. illite. vermiculite 2.0—20 Quartz. feldspars. goethite. boehmite 20-50 Quartz. feldspars. goethite. boehmite On the silt fraction the main source of Mg would be the feldspars and possibly goethite. These minerals are more resistant to weathering than the clay minerals. which would account for the lack of relationship between Mg con- tent and lattice activity. Mortland and Ellis (1959) reported that the rate limiting step for K release from vermiculite (K in interlayer 71 positions) was diffusion through the hydration film envelop- ing the clay particle. Doll et a1. (1965) obtained a corre- lation between clay K content and logrithm of K uptake by wheat tops. suggesting that release was related to the ac- tivity of lattice K. These results indicate that release of interlayer cations is related to activity of the lattice ion and is controlled by film diffusion. Thus. it is suggested that when the main source of Mg in the soil clay is from the interlayer position. as in the coarse clay. release is related to lattice activity sim- ilarly to K release. However. when most of the Mg is re- leased from the crystal surfaces as in the case of fine clay and the first crop on the medium clay. factors other than film diffusion and the activity of the lattice Mg control its release. These factors probably include. as the rate limiting step. diffusion through weathered residues. 3. Comparison of Magnesium Availability of the Coarse Clay Fraction from the Montcalm Soils As was discussed in the section describing the in- tensive cropping experiment. the three Montcalm soils were different with reSpect to Mg supplying power and to Mg 72 response. These differences appear to be explained by dif- ferences iang release from the coarse clay fraction. More Mg was released from the coarse clay fractions of Montcalm 1 than from this same fraction on the other two soils (Table 20). The ratio between the coarse. medium. and fine clays for all of these soils was approximately 7:2.421. respec- tively. Therefore. Montcalm l with a more active coarse fraction. supplied more Mg than did the other two. These differences in Mg supplying power were not due to differ— ences in mineralogical composition or Mg content of the coarse clay. nor to mineralogical changes that could be attributed to cropping (Table 23). 4. Magnesium Content of the Fractions Before and After Cro in Magnesium content of the soil and clay fractions de- creased during crOpping. but relative loss of Mg during crop- ping did not appear to be related to either fraction size or initial Mg content. Mortland and Lawton (1961) reported that biotite lost K in the initial stages of leaching in relation to particle size; the finer fractions losing more of their K than the coarser. However. as leaching progressed to later 73 stages of weathering. the large particles lost as much as the fine. The lack of consistency in relative loss of Mg content with fraction size and Mg content can be explained as follows: Mineral species at different stages of weathering are present in various sized fractions. Release will occur. dependent upon the relative stage weathering as well as par- ticle size and lattice ion activity. Differential Mg release between mineral species will occur if the activity in one species is greater than in an- other (Mortland. 1961). Such a system was probably present in these soil clays and contributed to the inconsistent rela- tive loss between the fractions. 5. Mineralogical Composition of Soil Fractions and Weathering During Cropping The minerals occurring in the silt were quartz. feldspars. goethite. and boehmite (Table 23). The specific feldspars present were not identified by the technique em- ployed. Within these minerals. Mg would probably be present as isomorphous substitutions in octahedral positions of the feldspars and goethite. 74 Nearly all of the clay fractions contained small amounts of illite (as defined by Grim. 1953) and kaolinite. With one exception. all clay fractions contained interstrati— fied material between 10A and 14A. Depending upon the pat- tern of collapse upon K saturation and heating. this inter- stratification was placed in either a tertiary interstatifi- cation of chlorite. vermiculite and illite or a binary inter— stratification between any two of this group. The medium clay from Munising 1 contained an ordered interstratifica- tion of chlorite. vermiculite. and montmorillonite. Weathering of fractions by crOpping was expected to follow one of two patterns (or both). Removal of interlayer Mg from brucite layers of chlorites would weaken interlayer O bonding allowing these structures to expand to 17A upon glycerol solvation; or removal of the persistent 14A peak. characteristic of chlorite material. after K saturation and heating to 550° C. Since fine fractions of clay do not yield any information by X—ray diffraction analysis. only the medium and coarse clays were characterized before and after cropping. Only three gave any indication of weathering by cropping. Figure 1 shows X-ray diffraction patterns before and after cropping for the coarse clay of the Karlin soil. The 75 10.0 A 14.1 A crOpped. K saturated. 550° C cropped. 300° C K saturated. cropped. Mg saturated. 1ycerol treated uncropped. K saturated. 550°C uncrOpped. K saturated. 300° C uncropped. K saturated glycerol treated ' J I I FT I 14 12 10 8 6 4 Degrees 2 0 Figure 1.-- X-ray diffraction patterns for coarse clay (0.2—2.0 u) of the Karlin loamy sand soil prior to and after cropping. 76 O persistent 14A peak on heating to 550° was weakened by crop- ping as indicated by the broadening of the 10A peak with this treatment. Removal of the interlayer material allowed this collapse. resulting in a broadening of the 10A peak. Sufficient interlayer material was removed from the medium clay of Montcalm l to cause a sharpening of the 14A peak after crOpping (Figure 2. Mg-glycerol treatment). This indicates formation of a more discrete 14A material. Heat- ing to 550° produced a sharpening of the 10A peak along with more symmetry of this peak. which indicates formation of a more discrete material of the vermiculite nature. Changes for the medium clay of the Montcalm soil 3. followed the same pattern as described above. The degree of change was greater (Figure 3) as indicated by the formation of a more symmetrical 10A with heating to 550°. At 300° the cropped sample gave a very symmetrical 10A peak further indi- cating the removal of the interlayer material of this inter- stratified system. Enhancement of the 7A peak after cropping was attrib- uted to small amounts of kaolinitic material contaminating the sand used in the cultures. 77 10.0 A 14.1 A crOpped. K saturated. 550° C cropped. K saturated. 300° C cropped. Mg saturated glycerol treated uncrOpped. K saturated. 550° C uncropped. K saturated. 300° uncropped. k saturated. glycerol treated l I I I 14 12 10 8 6 4 Degrees 2 0 Figure 2.—-X—ray diffraction patterns for medium clay (0.08- 0.2 u) for the Montcalm loamy sand (1) prior to and after cropping. 78 10.0 A 14.1 A cropped. K saturated. 550°C \ . ‘ crOpped. K‘saturated. 300°' cropped. Mg saturated. glycerol M uncropped. K saturated. uncropped. K saturated. 300°C V I 14 12 10 8 6 4 1 Degrees 2 0 Figure 3.--X-ray diffraction patterns for medium clay (0.08- 0.2 u) of Montcalm loamy sand (3) prior to and after cropping. SUMMARY Three experiments were conducted to study some fac- tors affecting release of soil Mg and uptake by oat plants. In the first experiment. three Mg levels (0. 10. and 20 ppm) were applied to seven Michigan soils and a bentonite-sand mixture. Seven oat crops were grown in the greenhouse without any additional Mg application. Extractable cations of the soil and Ca. K. and Mg content of oat tops were de- termined for each crop. In the second experiment. two Ca levels were applied to each of three pH levels and eight consecutive crops were grown. Extractable cations. soil pH and Ca. K and Mg of the tissue were determined for each crop. In a third experiment sufficient material to supply 20 mg Mg from each of five fractions of different particle size from seven soils. three from bentonite. and two from chlorite were cropped to two crops of oats in sand cultures. Tissue ‘Mg content was measured for each crop. Fraction Mg content was measured before and after each crop. Yield re5ponse to applied Mg was not obtained on any of the soils for the first three crops. Yield responses to applied Mg varied from soil to soil in the last 4 crops. 79 80 Relative release of Mg among the soils was as follows: Sims clay loam >> Munising fine sandy loam 2 > bentonite—sand mixture 2 Montcalm loamy sand 3.2 Montcalm loamy sand 1_: Munising fine sandy loam 1'2 Karlin loamy sand 2 Montcalm loamy sand 2. Montcalm loamy sand soils from three different loca- tions responded differently to applied Mg. Yield and uptake responses to applied Mg were obtained on soils from two loca- tions with an extractable Mg level of 90 ppm. Soil from the third location. which contained 40 ppm Mg. did not give a response to applied Mg. This was attributed to the 0.2-2.0LL clay supplying more Mg for plant growth on this soil than on the other two. However. this Mg existed in a form not ex- tracted by ammonium acetate. Plant available Mg was depleted on all soils after seven crops; an uptake response to applied Mg was obtained on all soils on the seventh crop. Extractable K/Mg correlated better with Mg availabil- ity (uptake. mg/culture) than did extractable Mg. Forms of Mg are available to plants from some soils which are not extracted by the ammonium acetate extraction. It is suggested that these forms are located in interlayer 81 positions rather than at the crystal edge. The NH: ion. which fits into the perforations in the surface of the oxy- gen layers of the tetahedral layer. causes these layers to collapse trapping interlayer Mg. Where the coarse clay fractions are most active. such as the Montcalm loamy sand Which didn't respond to applied Mg. Mg is being supplied from interlayer positions and the NH ion is interfering 4 with the measurement of this plant available Mg. Other workers (Schatschabel 1957; and Gonzales. 1963) have reported that extracting with CaCl or 0.1 N 2 NaCl or 0.1 N alphanaphthylamine—HCl gave better predictions of availability to plants than does ammonium acetate. Fur- ther evaluation of various extracting solutions is required. The antagonism between K supply and Mg uptake is felt to be the predominate force in the reduction of Mg up— take rather than entrapment of interlayer Mg. However. in soils where the Mg supply is predominantly from interlayer positions. the physical entrapment becomes more important. Further research is required to evaluate the plant related K-Mg antagonism as well as the influence of K on Mg entrap- ment and subsequent Mg uptake. 82 Soil pH had more of an effect on yield. tissue Mg content. and Mg uptake than did Ca level. This was atttrib- uted to better soil-plant relationships at higher pH levels (above pH 6.0) than below. The reasons for a larger amount of Mg released from unextractable forms and an associated smaller amount of Mg uptake when CaCO was applied than when Ca(OH)2 was applied 3 were not explained by the data obtained. A more definitive study is required to clarify these differences. Magnesium release increased with increasing soil acidity. Which supported reports in the literature. Relative order for Mg release according to particle size was as follows: <0.08 = 0.08-0.2 > 0.2-2.0 = 2-0'20.Z 20-50u > total soil. Chlorite and bentonite released more Mg than did the soil clays. Magnesium uptake was linearly correlated with frac- tion Mg content for both crops on the coarse clay fraction. for the second crop on the medium clay fraction. but not for the fine clay or either silt fraction. These data indicate that interlayer Mg is released in relation to the activity of lattice Mg. but when Mg is released from the crystal 83 surface other factors govern release. It was suggested that diffusion through the weathered octahedral residue was prob— ably the rate limiting step in the release of Mg. Differential release of Mg between different clay sized fractions could not be related to different kinds of minerals present. Silt sized fractions were too resistant to weather- ing for very much Mg release from interlayer sites. which explained the lack of correlation between Mg uptake and fraction Mg content. Differential release between the silt fractions could not be related to different kinds of minerals present. X-ray diffraction patterns before and after cropping indicated that cropping had removed interlayer material from some soil fractions. LI TE RATURE REV IEWED Adams. F. and Henderson. J. B. 1962. Magnesium availabil- ity as affected by deficient and adequate levels of potassium and lime. Soil Sci. Soc. Amer. Proc° 26: 65-68. Albrech. W. A.. Graham. E. R. and Ferguson. C. E. 1939. Plant growth and the breakdown of inorganic soil colloids. Soil Sci. 47: 455-458. Barshad. I. 1960. Significance of the presence of exchange- able Mg ions in acidified clays. Science 131: 988- 990. Bear. F. E. and Prince. A. L. 1945. .Cation equivalent con- stancy for alfalfa. Jour. Amer. Soc. Agron. 37: 217—222. Beckett. P. H. T. 1965. The cation-exchange equilibria of calcium and magnesium. Soil Science 100: 118-123. Blair. A. W.. Prince. Al. and Ensminger. L. E. 1939. Ef- fect of applications of magnesium on crop yields and on the percentages of Ca and Mg oxides in the plant material. Soil Sci. 48: 59-75. Boynton. D. and Burrill. A. B. 1944. Potassium-induced Mg deficiency in the McIntosh apple tree. Soil Sci. 58: 411-454. Brown. D. A. and Wells. J. P. 1958. Ion exchange in soil- plant root environments: III The use of Mg for the measurement of magnesium absorption by plants. Soil Sci. Soc. Amer. Proc. 22: 322-326. Cain. J. C. 1955. The effect of potassium and magnesium on the absorption of nutrients by apple trees in sand culture. Amer. Soc. Hort. Sci. Proc. 65: 25-31. 84 85 Carolus. R. L. 1933. Some significant variations in the chemical composition of the plant associated with a malnutrition trouble of potatoes. Amer. Potato Jour. 10: 147—165. Carolus. R. L. 1933. Some factors affecting the absorption of Mg by the potato plant. Amer. Soc. Hort. Sci. Proc. 30: 480-484. Carolus. R. L. 1937. Relation of K. Ca and Na to Mg defi- ciency. Am. Soc. Hort. Sci. Proc. 33: ‘Cheminade. R. and G. Drouineau 1936. R echerches sur la mecanique chimique des cations 'exchangeables. Ann. Agron. 6: 677. Chittenden. E. T.. Watson. J. and Hodgson. L. 1964. Mag- nesium uptake by pastures and apple leaves from top dressings of ground serpentine and dunite. N. Z. J. Agric. Res. 7: 28-31. Coleman. N. T.. Kamprath. E. J. and Weed. S. B. 1958. Liming. Advanc. Agron. 10: 475-552. Cook. R. L. and Millar. C. E. 1955. Plant Nutrient Defi- ciencies. Michigan State Agri. Exp. Sta. Spec. Bul. 353. Cornfield. A. H. and Pollard. A. G. 1952. The relative ratios of release of K. Ca and Mg from soils during electrodialysis. J. Sci Food Agr. 3: 613-615. Doll. E. C. and Christenson. D. R. 1966. Routine soil test determination of Mg using an atomic absorption Spec- trophotometer. .Mich. State Univ. Agri. Exp. Sta. Quart. Bul. 49: 216-220. Doll. E. C. and Hossner. L. R. 1964. Magnesium deficiency as related to liming and potassium in acid sandy podzol. 8th Intern. Congress of Soil Science. Bucharest. Romania. 4: 907-912. 86 Doll. E. C.. Mortland. M. M.. Lawton K. and Ellis. B. G. 1965. Release potassium from soil fractions during cropping. Soil Sci. Soc. Amer. Proc. 29: 699-702. Doll. E. C. and Thurlow. D. L. 1965. Soil management and fertilization for potatoes grown on mineral soils. Mich. State Univ. Agri. Exp. Sta. Res. Report 31. Drosdoff. M. and Kenworthy. A. L. 1944. Magnesium defi— ciency of Tung trees. Proc. Amer. Soc. Hort. Sci. 44: 1-7. Droste. J. B. 1960. Clay mineral alteration in some Indiana soils: in Clay and Clay Minerals. 9th Conf. Pergamon .Pres. pp. 329-342. Drouineau. G. and Nageotte.~M. 1941. Antagonism of certain cations in plant nutrition. Trav. membres. soc. Chim. biol 23: 1073-1077. Echevin. R. 1935. The absorbing power of soils with respect to MgClZ. Compt. Red. 200: 1243—1245. Elgabaly. M. M. 1955. Specific effects of adsorbed ions on plant growth: I. Effect of different combinations of Ca. Mg and Na on barley seedings. Soil Sci. 80: 235-248. Ellis. B. G. and Mortland. M. M. 1959. Rate of potassium release from fixed and native forms. Soil Sci. Soc. Amer. Proc. 23: 451-453. Embelton. T. W.. and Boynton. D. 1950. Effects of Spray residues on the pH and the basic cations of the soil in northeastern McIntosh Apple orchards. Soil Sci. Soc. Amer. Proc. 14: 105-109. Epstien. E. and Leggett. J. E. 1954. The absorption of alkaline earth cations by barley roots: Kinetics and mechanism. Amer. Jour. Bot. 41: 785-791. 87 Ferrari. T. J.. and Sluijsmans. C. M. J. 1955. Mottling and magnesium deficiency in oats and their depend- ence on various factors. Plant and Soil. 6: 262- 289. Foy. C. D. and Barber. 8. A. 1958. Magnesium deficiency and corn yields on two acid Indiana soils. Soil Sci. Soc. Amer. Proc. 22: 145-148. Fried. M. and Dean. L. A. 1952. A concept concerning the measurement of available nutrients. Soil Sci. 73: 263-271. Gastuche. M. C. and Fripiat. J. J. 1962. Acid dissolution techniques applied to clay structure determinations and controlled by physical methods. In Science of Ceramics. G. Stewart. Ed. 1: 121-138. Gautier. P. 1966. Effect of magnesium on various crops in different French soils. Soils and Fert. 29: 465. Goldschmidt. V. M. 1945. The geochemical background of minor elements distribution. Soil Sci. 60: 1-7. Gonzales. M. A. 1963. Relation of soil reaction to availa- bility of Mg from different sources. M. S. Thesis. Mich. State Univ. Graham. E. R. 1941. Soil development and plant nutrition: I. Nutrient delivery to plants by the sand and silt separates. Soil Sci. Soc. Amer. Proc. 6: 259-262. Grim. R. E. 1933. Clay-Minerology. McGraw Hill. New YOrk. Halstead. R. L.. Maclean. A. J. and Nielsen. K. F. 1958. Ca:Mg ratios in soil and yield and composition of alfalfa. Can Jour. Soil Sci. 38: 85-93. Harding. R. 1945. Exchangeable cations in soil of Cali- fornia orange orchards in relation to yield and size of fruit and leaf composition. Soil Sci. 77: 119— 127. 88 Hashimoto. T. 1957. Studies on the Mg metabolism of crops. I: The balance among Mg. Ca. and K in free and bound forms at the flowering stage of soybean plants. Soil and Plant Food. 2: 123-130. Hashimoto. T. 1955. Studies on the magnesium nutrition of crops. IV: Interrelationships of magnesium. cal- cium and potassium in crOps. J. Sci. Soil Tokyo. 26: 139—142. Hoagland. D. R. and Arnon. E. I. 1950. The water culture method for growing plants without soil. Calic. Agri. Expt. Sta. Cir. 347. Hossner. L. R. 1965. Release of Mg by leaching from vermic— ulite. mica and prochlorite. Ph.D. Thesis. Mich. State Univ . Hossner. L. R.. Doll. E. C. and Thurlow. D. L. 1968. Ef- fect of magnesium fertilization on yield and mag- nesium uptake by potatoes. (Manuscript in prepara- tion.) Hovland. D. and Caldwell. A. C. 1960. Potassium and mag— nesium relationships in soils and plants. Soil Sci. 89: 92—96. Jackson. M. L. 1956. Soil Chemical Analysis. Advanced Course. (Second Printing.) Published by the author. Department of Soil Science. University of Wisconsin. Madison. Jackson. M. L. 1958. Soil Chemical Analysis. Prentice- Hall. Inc.. Englewood Cliffs. New Jersey. Jackson. M. L. Soil clay mineralogical analysis. p. 245-294. Ig_Rich. C. I. and Kunze. G. S. Soil clay mineralogy. University of North Carolina Press. Chapel Hill. Jackson. M. L. Tyler. S. A.. Willis. A. L. Bourbeau. G. A. and Pennington. R. P. 1948. Weathering sequence of clay-size minerals in soils and sediments. Jour. Phy. and Coll. Chem. 52: 1237-1260. 89 Jacoby. B. 1961. Ca—Mg ratios in the root medium as re- lated to Mg uptake by citrus seedlings. Plant and Soil. 15: 74-80. Kardos. L. T. and Joffe. J. S. 1938. The preparation. composition and chemical behavior of the complex silicates of Mg. Ca. Sr and Ba. Soil Sci. 45: 293-307. Kerr. G. T.. Zimmerman. R. H.. Fox. H. A. and Wells. F. H. 1956. Degradation of hectorite by hydrogen ion: In Clays and Clay minerals. Natl. Acad. Sci.-Natl. Res. Council Pub. 456: 322-329. Key. J. L. and Kurtz. L. T. 1960. ResPonse of corn and soybeans to magnesium fertilizers. Agron. Jour. 52: 300- Kolthoff. I. M. and Sandell. E. B. 1949. Textbook of Quan- titative Inorganic Analysis. The Macmillan Co. New York. Komai. Y. and Noda. Y. 1957. Studies on the complementary ion effect between magnesium and other cations in the ionic environment of plant roots. III: The ad- sorption of cations by barley roots. J. Sci. Soil Tokyo. 28: 279-282. Korableva. L. I. 1954. The effect of lime and magnesium fertilizers on the fertility of sod-podzolized sandy loam acid soils. Pochvovedenie No. 3: 17-32. Korableva. L. I. 1966. Vermiculite as a source of Mg for plants. Soils and Fert. 29: 467. Lancaster. J. O. 1958. Magnesium status of Blackland soils of Northeast Mississippi for cotton production. Miss State Univ. Agri. Exp. Sta. Bul. 560. Larsen. R. P. et a1. 1959. Effects of N. P. K. and Mg fer- tilizers on yield and petiole nutrient content of a Concord grape vineyard. Mich. State Univ. Agr. Exp. Sta. Quart. Bull. 41: 812-819. 90 Lea. G. G. and Smith. J. B. 1938. Availability studies on Magnesium in magnesic limestones. serpentine. olivine and magnesium ammonium phosphate. Am. Potato Jour. 15: 4-9. Loew. O. 1892. Uber die physiologischen Functionen der Calcium-und Magnesiumsalze in Pflazenorganisms. Flora 75: 368-394. Loew. O. 1931. The effect of Ca upon the physiological function of Mg. Ernahr Pflaze 27: 97-101. 121—122. Abstracted Chem Abstracts 25: 5498. Longstaff. W. H. and Graham. E. R. 1951. Release of mineral magnesium and its effect on growth and composition of soybeans. Soil Sci. 71: 167-174. Lucas. R. E. and Scarseth. G. D. 1947. Potassium. calcium and magnesium balance and reciprocal relationship in plants. Jour. Amer. Soc. Agron. 39: 887-896. MacIntire. W. H. and Shaw. W. M. 1926. Fixation of Ca-Mg from burnt limes. limestone and dolomite incorpora— tion in two soil zones. Soil Sci. 22: 109-121. MacIntire. W..H.. Shaw. W. H. and Robinson. B. 1934. Dis- tinction between Mg absorbed and that exchangeable. four years after lysimeter incorporations of oxides and carbonates. Soil Sci. 37: 289-303. McCart. G. D. and Kamprath. E. J. 1965. Supplying calcium and magnesium for cotton on sandy. low cation ex- change capacity soils. Agron. Jour. 57: 404-406. McLean. E. O. 1950. Reciprocal effects of Mg and K as shown by their cationic activities in four clays. Soil Sci. Soc. Amer. Proc. 14: 89—93. Magnitskie. K. P. 1950. Mg deficiency in winter rye on a podzolized sandy loam. Agrobiologiya. 1: 96-99. Mazaeva. M. M. 1950. Sensitivity to Mg deficiency and photoperiodic reaction in plants. Doklady Akad. Nauk S.S.S.R. 73: 191-194. 91 Meller. A. and Bright. J. E. 1958. The rate of reactions as a function of time. Jour. Phys. Chem. 62: 495—497. Miller. G. W. 1960. Carbon dioxide-bicarbonate absorption. accumulation. effects on various plant metabolic reactions and possible relations to lime induced chlorosis. Soil Sci. 89-241-245. Michael. G. and Schilling. G. 1957. Magnesium levels in the arable soils central Germany. A. PflErnahr. Dung. 79: 31-50. Milam. F. M. and Mehlich. A. 1954. Effect of soil-root ionic environment on growth and mineral content of Crotalara straita. Soil Sci. 77: 227-236. Moore. D. P.. Overstreet. R. and Jacobson. L. 1961. Uptake of Mg and its interaction with Ca in excised barley roots. Plant Physiol. 36: 290-295. Mortland. M. M. 1958. Kinetics of potassium release from biotite. Soil Sci. Soc. Amer. Proc. 22: 503-508. Mortland. M. M. 1961. The dynamic character of potassium release and fixation. Soil Sci. 91: 11—13. Mortland. M. M. and Ellis. Boyd. 1959. Release of fixed potassium as a diffusion controlled process. Soil Sci. Soc. Amer. Proc. 23: 363-364. Mortland. M. M. and Lawton. K. 1961. Relationships between particle size and potassium release from biotite and its analogues. Soil Sci. Soc. Amer. Proc. 25: 473-476. Murray. H. H. and Leininger. R. K. 1956. Effect of weather- ing on clay minerals: In clays and clay minerals. Natl. Acad. Sci. Nat. Res. Council. pub. 456. 340- 347. Nearpass. D. C. and Drisdoff. M. 1952. Potassium. calcium and magnesium in tung leaves as related to these inons in the soil. Soil Sci. 74: 295. 92 NOdaI M.I Komai. Y.I and Iba. T. 19560 Reciprocal effeCtS of magnesium with other cations in the vicinity of roots. 1. Release of exchangeable Mg. K and Ca from clays. J. Sci. Soil Tokyo. 27: 245-248. Pratt. P. F. et a1. 1957. Magnesium and potassium content of orange leaves in relation to exchangeable Mg and K in the soil at various depths. Proc. Amer. Soc. Hort. Sci. 70: 245-251. Prince. A. L.. Zimmerman. M. and Bear. F. E. 1947. The magnesium-supplying powers twenty New Jersey soils. Soil Sci. 63: 69-78. Rains. D. W.. Schmid. W. E. and Epstein. E. 1964. Absorp- tion of cations by roots. Effects of hydrOgen ions and essential role of calcium. Plant Physiol. 39: 274-278. Rudgers. L. A. 1967. The evaluation of various magnesium carriers when applied with mixed fertilizers. M.S. Thesis. Mich. State Univ. Salmon. R. C. 1964. Cation-activity in equilibrium soil solutions and the availability of magnesium. Soil Science. 98: 213-221. Sanik. J.. Perkins. A. T. and Schrenk. W. G. 1952. The effect of the Ca-Mg ratio on the solubility and availability of plant nutrients. Soil Sci. Soc. Amer. Proc. 16: 263-267. Schachtschabel. P. 1957. Magnesium in Boden und Pflanze. Vortrag gehalten auf der Sitzung der Sektion 10 (Landw. Versuch—und Untersuch—ungswesen) am 24.1. Berlin 507—523. Scharrer. K. and Mengel. K. 1958. The K-Mg antagonism in maze and sunflower. Z. PflErahr. Dung 83: 149-162. Scharrer. K. and Mengel. K. 1959. Transitory occurrence of visible signs of magnesium deficiency in oats. Agro- chimica 4: 3-24. 93 Schilling. G. 1958. The form of Mg. Ca and K occurring in soils of central Germany. Chem d. Erde. 19: 170- 197. Schnorr. H. 1963. (The effect of increasing magnesium ap- plications in relation to soil and fertilizing. on the yield and mineral uptake in maize and annual Bernburg Ryegrass) Albrecht-Thaer-Arch. 7: 811-831. Selke. W. 1960. The importance of magnesium fertilizing in the Potsdam Cottbus and Frankfurt (Oder) districts. Dtsch. Landw. 11: 450-454. Selke. W. and Ebert. E. 1962. (Relationships between mag- nesium fertilizing liming and form of nitrogen.) Z landw. Vers. Untersuch Wer. 8: 473-497. Selke. W. et a1. 1960. The relationship between Mg content and some yield determining properties of soils in Brandenburg. Z. Landw. Vers.-Untersuchwes. 6: 374-395. Semb. G. and Oien. A. 1961. Preliminary investigations on the liberation of Mg from olivine. Soils and Fert. 24: 3282. Sluijsmans. C. M. J.. et al. 1959. Potassium fertilization and Mg deficiency of potatoes. Kali No. 39. 331-335. Soofi. G. S. and Fuehring. H. D. 1964. Nutrition of corn on a calcareous soil: I. Inter-relationships of N. P. K. Mg and S on the growth and composition. Soil Sci. Soc. Amer. Proc. 28: 76-79. Southwick. Lawrence. 1943. Magnesium deficiency in Massa— chusetts Apple Orchards. Proc. Am. Soc. Hort. Sci. 42: 85-94. Spencer. W. F. 1954. Influence of cation-exchange reaction on retention and availability of cations in sandy soils. Soil Sci. 77: 129-136. 94 Stahlberg. S. 1961. Studies on the release of bases from minerals and soils. IV. The release of Ca and Mg by boiling normal hydrochloric acid. Soils and Fert. 24: 37. ’ Stalkov. Ts. 1942. The occurrence of difficulty soluble magnesium salts that affect the exchangeable magnes- ium. Bodenkunde u. Pflazenerahr. 28: 105-118. Steel. R. G. D. and Torrie. J. H. 1960. Principles and procedures of statistics. McGraw-Hill. New York. Steinberg. R. A. 1951. Influence of acidity. Ca and Mg on growth and culture of xanthi tobacco in water cul- ture. Plant Physiol. 26: 37-44. Temme. J. 1952. Some results of two fertilizer experiments. Landbouwk. Tijdschr. 62: 362-368. Tobin. L. W. 1960. The Mg status of Michigan soils and effect of several magnesium sources on the yield and chemical composition of crops. Ph.D. Thesis. Mich. State Univ. Tobin. L. W. and Lawton. K. 1962. The magnesium status of Michigan soils and the effect of magnesium sources on the yield and chemical composition of crops. Mich. State Univ. Agri. Exp. Sta. Quart. Bul. 44: 442-454. Tucker. T. C. and Smith. F. W. 1952. The influence of B. Mg. and K on the growth and chemical composition of red clover grown under greenhouse conditions. Soil Sci. Soc. Amer. Proc. 16: 252-255. Vageler. P. 1933. An Introduction to Tropical Soils (translated by H. Greene). The Macmillan Co. 240pp. Vander Marel. H. W. 1947. TrOpical soils in relation to plant nutrition. Soil Sci. 64: 445-451. Viets. F. G. 1944. Calcium and other polyvalent cations as accelerators of ion accumulation by excised barley roots. Plant Physiol. 19: 466-480. 95 Walker. R. B.. Walker. H. M. and Ashworth. P. R. 1955. Calcium-magnesium nutrition with Special reference to serpentine soils. Plant Physiol. 30: 214-221. Walsh. T. and Calrke. E. J. 1945. Chlorosis of tomatoes with particular reference to K-Mg relations. Proc. Roy. Irish Acad. SOB: 245-263. Wehunt. R. L. and Purvis. E. R. 1954. Mineral composition of apple leaves in relation to available nutrient content of the soil. Soil Sci. 77: 215-218. 27 LEM)gimmigiiflmmm IIHIWHH . 3