“sh mwniwa" ‘ .544» ‘.*l. 7". -n-— ,‘W—y fl. .-...- mu V. «mt. .p-u “I" “aru 1: ”2....3 ‘15:» “(.79 1""! char: ‘ it! ~51-.. .._ li-j?‘-tq‘ ffifl‘g‘iifd 5-1245!“- « " v“. ‘ '31 Egg Jar-«:7 <3§D§rqh ‘ . . =1 - *t'” ' ' ‘31:? L min i“ O {W 1: I 0 A44 ‘fi‘wn 1: 39m 4&1. ' ;: :46" ‘ {.351 55.9%: I“. . . dial-w” gig r ff)“; . :I u 4 . .fivlr.=.§il. ’53 2:25 * ' .57 “3% 'l‘ .mgfi.” ; A 9” - t f 15%?gé? ‘33" . -. ‘ , -.I.U‘..:L ' ‘2'“, :E b 1: ‘.. :4, 1" v - ~.o~- - -.-w mo. w —v . —v~. '9’ A ‘ . m... .. w ~ .zati‘, A .1 a .0 3., ‘.‘.‘ am? < v C :...."'—I".’ n...»- . . mnm a'yv-ov- . . -.M-—- .w‘-.—.c vm.‘ v»... 6%.. -v‘ ‘1 '23:: “ FM... 'lHESlS IIMilli!ill]Ill'lii‘i‘llulillflllI 3 1293 01710 3833 This is to certify that the thesis entitled STUDIES COMPARING THE POTASSIUM CHEMISTRY AND CLAY CONTENT OF SOME MICHIGAN, INDIANA, AND OHIO SOILS presented by James Barrett Sallee has been accepted towards fulfillment of the requirements for M.S. degree in Crop & Soil Sciences fiyb/Llw CA- Major professor Date April 27, 1998 0-7639 MSU is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State Unlverslty PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE MTE DUE DATE DUE 1!” COMM“ STUDIES COMPARING THE POTASSIUM CHEMISTRY AND CLAY CONTENT OF SOME MICHIGAN, INDIANA, AND OHIO SOILS By James Barrett Sallee A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Crop and Soil Sciences 1998 ABSTRACT STUDIES COMPARING THE POTASSIUM CHEMISTRY AND CLAY CONTENT OF SOME MICHIGAN, INDIANA, AND OHIO SOILS By James Barrett Sallee Michigan, Indiana, and Ohio share common soil series. However, few studies have been made comparing the chemistry and clay composition of these soil series with respect to potassium (K) availability. The objective of this study was to determine the degree of similarity in the K chemistry ofsoils common to this tri-state area. In 1995 and 1996, 54 soils were collected for investigation. Cation exchange capacity (CBC), total K, ammonium acetate (NI-hOAc) and sodium tetraphenylboron (NaBPh4) extractable K levels, and clay content were determined for all soils. Neubauer experiments with barley (Hordeum vulgare cv Bowers) were conducted in a growth chamber to investigate K release during intensive cropping. Four soils having a wide range in total K content were selected for clay mineralogical analysis by x—ray diffraction. Large differences in CEC, total K, and nonexchangeable K were found between major soil series. Exchangeable K levels were not significantly different between soil series. The contribution of exchangeable K to total plant uptake ranged from 17 to 75% between soil series. Soils with high clay content tended to have larger amounts of total K than soils with low clay content. The coarse clay (2-O.2um) and fine clay (<0.2um) fiactions were dominated by 2:1 clays. Illite contents ranged from 10 to 40% in the coarse clay and from 5 to 30% in the fine clay of the soils studied. Although differencesianhemistryandmineralogyexistbetweensoilseries,soilsinagivenseries generally exhibit similar K chemistry regardless of geographic location. ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Darryl Wamcke, for his guidance and friendship during the three years that I have studied under his direction at Michigan State University. I am very grateful for the financial support and graduate assistantship that made my graduate program a reality. 'I also want to thank Dr. Wamcke for suggesting the potassium studies that are embodied in this thesis. The many hours spent on this project have deepened my understanding of soils and have provided a unique opportunity to research an important soil fertility problem of great personal interest. I would like to thank the members of my guidance committee: Dr. Donald Christenson, Dr. Delbert Mokma, and Dr. Irvin Widders who provided helpful suggestions and advice as I progressed through my graduate studies. My appreciation also goes to Dr. Gary Steinhardt at Purdue University for facilitating the collection of soil samples fiom Indiana. I am indebted to the soil scientists in Ohio: Mr. Frank Gibbs, Mr. Jon Hempel, Mr. Steve Hamilton, and Mr. Terry Lutch who willingly collected soil samples for this study. Mr. Jon Dahl, Ms. Vicki Smith, Ms. Rosie Cabrera, Ms. Ten'i Janson, and Ms. Donna Ellis of the Michigan State University Soil and Plant Nutrient Laboratory provided valuable assistance in the analysis of soil and plant tissue samples. My career as a graduate student has been greatly enriched by the friendship and comradery of my fellow graduate students in the Department of Crop and Soil Sciences. iii Jose Cora, Ralph DiCosty Tom Mueller, Daniel Rasse, Djail Santos, Ayman Suleiman, and Tom Wilson all helped with various aspects of this project and their assistance is gratefully acknowledged. I am very appreciative to Mr. Cal Bricker for sharing his technical expertise and knowledge of laboratory techniques. Mr. Brian Baer and Ms. Elaine Parker were especially helpful in the preparation of slides and in helping me learn the SigmaPlot and Microsofi Ofiice 97 computer software programs. Finally, I want to express my appreciation to the undergraduate students: Chris Teboe, Charles Navratil, and Jeanette Makries, who helped me prepare and analyze the hundreds of samples necessary for this study. Without their dedicated assistance, I would have been unable to complete the pmiect. It is my sincere hope that the results from these potassium studies will set the stage for further research on the potassium chemistry of soils in Michigan, Indiana, and Ohio, and that this research will ultimately lead to the improvement of K fertilizer recommendations in the tri-state region. iv TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. vii LIST OF FIGURES ............................................................................................................. x INTRODUCTION ............................................................................................................... 1 Potassium Content, Mineralogy, and Weathering in Soils ...................................... 2 Contrasting Concepts in Soil Test Interpretation ..................................................... 6 Different Philosophies Used in Making Fertilizer Recommendations .................. ll Differences in Potassium Recommendations in the Tri-State Region ................... 12 Reasons for Tri-State Fertilizer Recommendations ............................................... 14 References .............................................................................................................. 16 CHAPTER 1 POTASSIUM RELEASE FROM EXCHANGEABLE AND NONEXCHANGEABLE FORMS IN MICHIGAN, INDIANA, AND OHIO SOILS AS MEASURED BY CHEMICAL EXTRACTION AND NEUBAUER TECHNIQUES ABSTRACT ...................................................................................................................... 20 INTRODUCTION ............................................................................................................. 21 MATERIALS AND METHODS ...................................................................................... 23 Soil Selection, Collection, and Preparation ........................................................... 23 Chemical Analyses ................................................................................................ 26 Neubauer Studies ................................................................................................... 27 RESULTS AND DISCUSSION ........................................................................................ 29 Chemical Analyses ................................................................................................ 29 Neubauer Studies ................................................................................................... 41 Factors Affecting Potassium Uptake ..................................................................... 51 Correlations between Plant Uptake and Extractable Potassium ............................ 55 CONCLUSIONS ............................................................................................................... 66 REFERENCES .................................................................................................................. 68 CHAPTER 2 THE RELATIVE IMPORTANCE OF CLAY CONTENT, CLAY MINERALOGY, AND CATION EXCHANGE CAPACITY TO THE POTASSIUM CHEMISTRY OF SOILS IN MICHIGAN, INDIANA, AND OHIO ABSTRACT ...................................................................................................................... 75 INTRODUCTION ............................................................................................................. 76 MATERIALS AND METHODS ...................................................................................... 78 Particle Size Analysis ............................................................................................ 78 Mineralogical Analysis .......................................................................................... 80 RESULTS AND DISCUSSION ........................................................................................ 82 Particle Size Analysis ............................................................................................ 82 Correlations between Clay Content, CBC, and Extractable Potassium Levels ..... 85 Relationship of Clay Mineralogy to Potassium Chemistry of Soils ...................... 95 Mineralogy of Other Soils in the Tri-state Region .............................................. 102 Relative Importance of Clay Content, Clay Mineralogy, and CEC in Establishment of Potassium Recommendations .................................. 104 CONCLUSIONS ............................................................................................................. 106 REFERENCES ................................................................................................................ 107 APPENDIX A Table A. 1. Location of soils and their taxonomic classification ......................... 111 APPENDIX B Table 8.1. Selected physical and chemical properties of all soils ....................... 114 vi LIST OF TABLES Table 1.1. Comparison of the amounts of K extracted by NaBPh4 and NI-LOAc in 54 soils from the tri-state area ....................................................................................... 29 Table 1.2. Mean values of CEC, total K, nonexchangeable K, and exchangeable K for all soil groups before Neubauer growth studies ...................................................... 31 Table 1.3. Mean values of CEC, total K, nonexchangeable K, and exchangeable K before Neubauer growth studies for the six soil series common to the tri-state area. Soils are grouped by major soil series .................................................................... 34 Table 1.4. Mean values of CEC, total K, nonexchangeable K, and exchangeable K before Neubauer growth studies for the six soil series common to the tri-state area. Soils are grouped by state ........................................................................................ 36 Table 1.5. Cation exchange capacity of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................................................................................................................................. 37 Table 1.6. Total K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................... 37 Table 1.7. Seven-day nonexchangeable K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................................................................................................................... 38 Table 1.8. Five-minute nonexchangeable K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................................................................................................................... 38 Table 1.9. Exchangeable K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................................................................................................................................. 39 Table 1.10. Mean values of CEC, total K, nonexchangeable K, and exchangeable K before Neubauer growth studies for the seven soil series common to the three glacial lobes of Michigan .................................................................................................. 40 vii Table 1.11. Amount of K taken up fiom the exchangeable and nonexchangeable soil K fractions by barley plants during Neubauer growth studies .................................... 42 Table 1.12. Average dry weight, percent K, and total K uptake of harvested barley shoot and root tissue from the 54 soils used in the Neubauer growth studies ................... 44 Table 1.13. Mean values of plant uptake K and the percent of uptake from exchangeable K for all soil groups .................................................................................... 46 Table 1.14. Mean values of plant uptake K and the percent of uptake from exchangeable K for the six soil series common to the tri-state area. Soils are grouped by major soil series ........................................................................................ 46 Table 1.15. Mean values of plant uptake K and the percent of uptake from exchangeable K for the six soil series common to the tri-state area. Soils are grouped by state ........................................................................................................... 49 Table 1.16. Plant uptake of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................... 49 Table 1.17. Exchangeable K uptake of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series .................................................................................................................................. 50 Table 1.18. Mean values of plant uptake K and the percent of uptake from exchangeable K for the seven soil series common to the three glacial lobes of Michigan ........................................................................................................................ 50 Table 1.19. Summary of coefficients of determination fi'om regression analysis of NH40Ac-K and NaBPm-K versus plant uptake K ....................................................... 63 Table 2.1. Particle size distribution of selected soils as determined by the hydrometer and pipet methods of particle size analysis ........................................................................ 83 Table 2.2. Mean clay content of the six soil series common to Michigan, Indiana, and Ohio ............................................................................................................... 84 Table 2.3. Clay content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. ................... 84 Table 2.4. Summary of coefl'rcients of determination fi'om regression analysis of percent clay and CEC versus plant uptake and extractable K levels before and after cropping .............................................................................................................. 94 viii Table 2.5. Estimated illite clay content, total soil potassium content, and total, coarse, and fine clay content of four soils fi'orn the tri-state area selected for mineralogical analysis on the basis of total K content ...................................................... 96 Table 2.6. Estimated mineral content in the coarse clay (2-0.2um) and fine clay (<0.2um) fi'actions of soils selected for mineralogical analysis from the tri-state area ..................................................................................................................................... 96 LIST OF FIGURES Figure 1.1. Tri-state map of Michigan, Indiana, and Ohio. Counties from which soil samples were collected are shown in color. ................................................................ 24 Figure 1.2. Relationship between initial NILOAc-K and plant uptake K for all soils .................................................................................................................................... 57 Figure 1.3. Relationship between delta NH40Ac-K and plant uptake K for all soils. ................................................................................................................................... 57 Figure 1.4. Relationship between initial five-minute NaBPh4-K and plant uptake K for all soils ..................................................................................................................... 58 Figure 1.5. Relationship between delta five-minute NaBPh4-K and plant uptake K for all soils ..................................................................................................................... 58 Figure 1.6. Relationship between initial NILOAc-K and plant uptake K for the Miami, Conover, Brookston soil association group .......................................................... 59 Figure 1.7. Relationship between delta NI-hOAc-K and plant uptake K for the Miami, Conover, Brookston, soil association group ......................................................... 59 Figure 1.8. Relationship between initial five-minute NaBPh4-K and plant uptake K for the Miami, Conover, Brookston soil association group ........................................... 60 Figure 1.9. Relationship between delta five-minute NaBPh4-K and plant uptake K for the Miami, Conover, Brookston soil association group ........................................... 60 Figure 1.10. Relationship between initial NI-hOAc-K and plant uptake K for the Morley, Blount, Pewamo soil association group. .............................................................. 61 Figure 1.11. Relationship between delta NI-LOAc-K and plant uptake K for the Morley, Blount, Pewamo soil association group. .............................................................. 61 Figure 1.12. Relationship between initial five-minute NaBPhr-K and plant uptake K for the Morley, Blount, Pewamo soil association group ................................................ 62 Figure 1.13. Relationship between delta five-minute NaBPh4—K and plant uptake K for the Morley, Blount, Pewamo soil association group .................................... 62 Figure 2.1. Relationship between the hydrometer and pipet methods of particle size analysis for determination of clay content in 11 soils from the tri-state area ............. 83 Figure 2.2. Relationship between the percent clay and plant uptake K for all soils. ................................................................................................................................... 86 Figure 2.3. Relationship between CBC and plant uptake K for all soils. .......................... 86 Figure 2.4. Relationship between the percent clay and initial NILOAc-K for all soils ......................................................................................................................... 87 Figure 2.5. Relationship between the percent clay and initial five-minute NaBPm-K for all soils ....................................................................................................... 87 Figure 2.6. Relationship between the percent clay and NI-LOAc-K afier cropping for all soils .......................................................................................................... 89 Figure 2.7. Relationship between the percent clay and five-minute NaBPh4-K after cropping for all soils. ................................................................................................. 89 Figure 2.8. Relationship between the percent clay and total K for all soils. ..................... 90 Figure 2.9. Relationship between the percent clay and CEC for all soils. ........................ 90 Figure 2.10. Relationship between the percent organic matter and CEC for all soils. .............................................................................................................................. 91 Figure 2.11. Relationship between CBC and initial NILOAc-K for all soils .................... 91 Figure 2.12. Relationship between CBC and initial five-minute NaBPh4-K for all soils ......................................................................................................................... 93 Figure 2.13. Relationship between CBC and NI-hOAc-K after cropping for all soils ............................................................................................................................... 93 Figure 2.14. Relationship between CBC and five-minute NaBPh4-K after cropping for all soils. ......................................................................................................... 94 Figure 2.15. X-ray difli'action patterns of coarse clay and fine clay from the Conover soil ....................................................................................................................... 98 xi Figure 2.16. X-ray diffi'action patterns of coarse clay and fine clay fi'om the Blount soil .......................................................................................................................... 99 Figure 2.17. X-ray diffi'action patterns of coarse clay and fine clay fi'om the Pewamo soil ..................................................................................................................... 100 Figure 2.18. X-ray diffi'action patterns of coarse clay and fine clay fiom the Paulding soil .................................................................................................................... 101 xii INTRODUCTION Potassium (K) is absorbed by plant roots in larger quantities than any other cation (Barber, 1968). The potassium content of healdry leaf tissue for most plants is in the range of 1 to 4% (Brady and Weil, 1996, p.474). The benefits of K to plant growth have been recognized since the middle of the 17th century when J. R. Glauber in the Netherlands first hypothesized that saltpeter was the “principle” of vegetation (Russell, 1988). He achieved large increases in crop growth when potassium nitrate was added to the soil. In 1840, von Leibig reported that K was an essential plant nutrient (Stewart, 1985). Since that time, the role of K in both soils and plants has been extensively studied. It is now known that K has important roles in many physiological plant processes, including: enzyme activation, protein synthesis, photosynthesis, osmoregulation, cell extension, stomatal movement, light-driven and seismonastic movements, phloem transport, and cation-anion balance (Marschner, 1995). In nature, plants obtain all of their K directly from the native K reserves of the soil or from K fertilizers that are added to the soil. Whenever nutrients of any form are added to the soil, the soil becomes both a source and a sink for these nutrients. Therefore, it is important to understand the factors that influence the availability of these nutrients to plants. This is especially true for K due to the dynamic nature of K availability in the soil. Potassium Content, Mineralogy, and Weathering in Soils Potassium makes up, on the average, 2.6% of the lithosphere. It is the seventh most abundant element and the fourth most abundant plant mineral nutrient in the earth’s crust (Schroder, 1978). The K content of mineral soils, expressed as K20, ranges fiom 0.05 to 3.5%. Most agricultural soils in the United States contain flour 1 to 2% (10,000 to 20,000 mg kg") K (Jackson, 1964). However, of the total amount of K in soils, approximately 90 to 98% is in the mineral form, 1 to 10% is fixed or nonexchangeable, and only 0.1 to 2% is exchangeable or in the soil solution (Tisdale et al., 1985). Generally, solution K is only a few percent of the exchangeable K (Schroeder, 1974). The soil organic matter usually contains less than 0.1% K while the K in living microorganisms is one order of magnitude lower (Schroeder, 1974). Thus, the contribution of K from the organic fraction of the soil is small. The total K content of a soil is determined in large measure by the mineralogy of the parent material; although, it can be affected by other factors such as the degree of weathering, K fertilization, and losses due to crop removal, erosion, and leaching. On average, the K composition of igneous (including metamorphic), shale, sandstone, and limestone rocks is 26, 27, 11, and 2.7 g K kg'1 respectively (Bertsch and Thomas, 1985). The composition of the soil mineral fraction is a mixture of both primary and secondary minerals. Primary minerals are minerals formed at high temperatures during the genesis of igneous and metamorphic rocks. Secondary minerals are minerals formed at low temperature and are derived fi'om sedimentary rocks or formed in soils by weathering (Jackson, 1964). Primary minerals comprise the main portion of the sand (particles ranging fiom 2.0 to 0.05 mm in diameter) and silt (particles ranging from 0.05 to 0.002 mm in diameter) particles of most soils. In contrast, the clay (pmicles less than 0.002mm in diameter) particles of most soils are comprised of secondary minerals except for very young soils that have undergone little weathering. One primary mineral group of major importance in soils is the K-feldspars (KAlSi303). The K-feldspars include sanidine, orthoclase, microcline, and adularia. Feldspars typically constitute 5 to 25% of the mineralogy of sand and silt particles (Jackson, 1964). The structure of the K-feldspars consists of a three dimensional fi'amework of linked SiO4 and A1203 tetrahedra which have sufficient space in the framework for K to balance electrostatic charge. Four- membered rings of tetrahedra are the basic units of the framework. These rings are then joined together to form a honeycomb type of arrangement (Sparks, 1987, Sparks and Huang 1985). As the minerals in soils weather, they release their K and form secondary minerals. Vermiculite, for example, is formed from the weathering of mica (Douglas, 1989). Huang et al. (1968) listed the following common feldspars and micas according to the rate at which they weather and release their K: biotite, muscovite, orthoclase, and microcline. Feldspars weather when protons replace structural cations. Hydrolysis of the Al-O-Si bond destabilizes the mineral structure and results in mobilization of silica (Rasmussen, 1972). This leads to formation of silicic acid which promotes the release of basic cations (Huang, 1989). Humic and fulvic acids also can weather feldspars and micas through complexation and chelation reactions (Tan, 1980). Micas are the second group of primary minerals with fundamental importance to the K chemistry of soils. Micas are more common in fine-grained sedimentary rocks such as shales and less common in coarse textured rocks such as sandstones (Sparks and Huang, 1985). In contrast to feldspars, the fraction of K in soils from mica tends to increase with decreasing particle size (Sparks, 1987). Micas are 2:1 layer silicate minerals composed of a sheet of octahedral cations sandwiched between two layers of (SiO4)“ tetrahedra. The apical hydroxyls and oxygens of the tetrahedral layer are shared by the octahedral layer. These sheets are held together electrostatically by potassium ions which neutralize the negative charge between adjacent 2:1 sheets. Both dioctahedral and trioctahedral micas exist in nature. Dioctahedral micas contain Al” as the octahedral cation; whereas trioctahedral micas have Fe or Mg as the octahedral cation (Fanning et al., 1989, Sparks and Huang, 1985). In the dioctahedral micas, such as muscovite, only two octahedral sites are filled by cations. In the trioctahedral micas, all three octahedral sites are filled by cations (Sparks and Huang, 1985). The layer charge associated with micas and other 2:1 minerals can be attributed to three factors: 1) substitution of Si“ by a lower valence cation in the tetrahedral sheet, 2) substitution of Al3+ by a lower valence cation in the tetrahedral sheet, or 3) cation vacancies in the octahedral sites (Sparks and Huang, 1985). The degree of isomorphous substitution and resultant layer charge will have a direct effect upon the cation exchange capacity (CEC) of the soil. Soils containing minerals with more isomorphous substitution will exhibit a higher CEC than soils containing minerals with less isomorphous substitution. Two mechanisms have commonly been recognized in the weathering of micas. These two mechanisms are edge weathering and layer weathering (Fanning et al., 1989). Jackson et al. (1952) reported that micas weather along preferential planes which result in expansion of the mineral. Huff (1972) observed layer weathering with several illites and a mixed layer illite-montrnorillionite. Development of fractures occurred in these samples when the interlayer K was removed by sodium tetraphenylboron (N aBPh4). Scott (1968) reported that edge weathering was the dominate mechanism of K release in > 10pm muscovite particles. In mica particles < 2pm, layer weathering was the dominate mechanism of K release. Scott and Smith (1967) obtained visual evidence of edge weathering in large mica particles weathered with NaBPhr solution. The release of K with weathering of feldspars and micas is of great practical importance to plants, animals, and humans. Although soils may contain large amounts of K, only a small portion of it is available for uptake by the plant at any given time. There are four pools of K in the soil: solution K, exchangeable K, nonexchangeable K, and mineral K (Sparks, 1987). Nonexchangeable K is distinct fi'om mineral K in that it is not bonded covalently within the crystal structure of soil minerals. Instead, it is held between adjacent tetrahedral layers of dioctahedral and trioctahedral micas, vermiculites, and intergrade clay minerals (Martin and Sparks, 1985). In all soils, there exists a dynamic equilibrium between each of these four forms of K. This equilibrium is controlled by many factors including K composition of the parent material, degree of weathering, amount of K fertilization, and intensity of cropping. The K that is immediately available for plant uptake is the K in the soil solution. As the K in the soil solution is depleted, K on the exchange sites is released into the soil solution. This release of exchangeable K buffers the concentration of K in solution. As more K is released fiorn the exchange sites, a portion of the nonexchangeable K becomes available for exchange with ions in the soil solution. The rate of this release is highly dependent upon the specific parameters of the individual soil- plant system. Over the years, much effort has gone into the development of chemical extractants that would provide a reliable measure of plant available potassium. While these efforts have met with some success, there is still a need to improve soil K tests and refine soil test interpretations for K. Contrasting Concepts in Soil Test Interpretation “The soil testing-fertilizer recommendation process involves several discrete but somewhat interrelated steps including field sampling, sample preparation, chemical analysis, interpretation of analytical results, and development of recommendations based on these results” (Eckert, 1987). Today, there are two major philosophies which guide the practice of soil testing and fertilizer recommendation in the United States. Both of these concepts originated in the 1940’s although the research was conducted in different regions of the country. In order to understand how these different concepts originated and how they affect current practice in the soil testing arena, it is beneficial to investigate each of these concepts in detail. The basic cation saturation ration (BCSR) concept originated with Bear and co- workers (Bear et al., 1945) in New Jersey from studies with alfalfa on 20 of New Jersey’s most important agricultural soils. They proposed that for an “ideal” soil the exchange complex should contain 65% Ca, 10% Mg, 5% K, and 20% H on an equivalent basis. These saturations would give cation ratios of 13:1 for Ca:K, 2:1 for Mg:K and 4:1 for H:K. These results were based on eight years of research (Bear and Toth, 1948). In reading through the series of papers by Bear et al., one is able to identify several reasons why these ratios were selected for the New Jersey soils. Hunter et a1. (1943) reported that although alfalfa could grow normally at soil Ca:K ratios from 1:1 to 100:1 the Ca:K ratios within the plant were much narrower. They observed that when the Ca:K ratio on an equivalent basis within the plant was greater than 4:1 the yields of the alfalfa were markedly depressed. A sharp decrease in alfalfa yield occurred when the Ca tissue concentration was greater than 2% or if the K tissue concentration was less than 1%. These values were then established as critical limits for alfalfa (Bear and Toth, 1948). At a soil Ca:K ratio of 32:1 the Ca:K ratio in the alfalfa was 3:1. The K concentration in the shoots was 1.15% (Bear and Prince, 1945). Bear and Toth (1948) reported that during earlier work by Bear et a1. (1944), it was found that alfalfa yields were increased in 18 out of 20 soils by adjusting the Ca:K ratio to near the “id ” 13:1 ratio. During this same period, Prince et a1. (1947) showed that when the Mg saturation of the exchange complex was less than 10%, a majority of these same soils responded to applications of Mg in the soluble form. It had been observed that crops grown on some New Jersey soils expressed symptoms of Mn and other micronutrient deficiences. Bear and Toth (1948) found that adjusting the soil pH to 6.5 could alleviate these symptoms. At a pH of 6.5 the exchange complex would contain 20% H and 65% Ca. Thus a Ca:K ratio of 13:1 would supply ample K (> 1% K in alfalfa tissue) and yet prevent luxury consumption of K. Bear and Toth (1948) also considered the fact that it was cheaper to apply Ca than K as a fertilizer. Therefore, since the plant would tolerate a wide range in soil Ca:K ratios, it made economic sense to add sufficient Ca to the soil to limit excessive uptake of K but still maintain healthy plant growth. These findings were used by Bear and his colleagues as the scientific and practical basis for the selection of a specific basic cation saturation ratio for an “ideal” soil. Later, Graham (1959) suggested that the percentage base saturation for soils in Missouri could range fi'om 65-85% for Ca, 6-12% for Mg, and 2-5% for K without affecting the soil’s ability to produce optimum yields. Since the time of Bear’s proposal of an “id ” basic cation saturation ratio, other researchers have attempted to verify the concept but have found little evidence for the existence of an ideal ratio. Hunter (1949) reported that the yield of alfalfa was not affected by varying Ca:Mg ratios in the soil, although it did have a highly significant effect upon the percentages of Ca and Mg in the alfalfa. In Indiana, Foy and Barber (195 8) found no yield reduction with corn even when Mg deficiency was induced by high rates of calcitic lime and potash on two acid soils. However, they reported that lime did produce a significant yield increase on one of the soils. McLean and Carbonell (1972) reported that yields of German millet (Seteria italica L. Beauv.) were not affected by varying Ca:Mg ratios. In growth chamber studies with German millet, Eckert and McLean (1981) found no evidence of an ideal ratio for Ca:MgzK. After an additional six years of field studies with corn, soybeans, wheat, and alfalfa, McLean et al. (1983) concluded that an “ideal” basic cation saturation ratio did not exist. They suggested that the emphasis of fertility programs should be to provide sufficient but not excessive amounts of nutrients rather than to provide an ideal ratio of basic cations. Although it would seem that the scientific evidence for the existence of an ideal basic cation saturation ratio is tenuous, McLean (1977) reported that there has been an increasing use of this concept for soil testing and fertilizer recommendations. In 1960, 1.25 million soil samples were tested in the North Central Region of the United States. Of these samples, 2/3 were analyzed by university soil testing labs. By the late 1960’s, 2 million samples were tested, of which only US were analyzed by university soil testing labs. The rest of the samples were analyzed by commercial and industrial soil testing labs. Almost all of the commercial and industrial labs used the BSCR concept when making fertility recommendations. Consequently, the total number of soils managed according to the BCSR concept is increasing. Liebhardt (1981) suggested that in Delaware 80 to 90% of the soils are managed using the BCSR concept. Although there appears to be little scientific evidence for the hypothesis that there is an “ideal” ratio of basic cations in the soil, the BCSR concept is still commonly used as the philosophical basis for making fertilizer recommendations. The second major concept used in soil fertility recommendations is the concept of a sufficiency level of available nutrients (SLAN). In a survey of 43 public universities operating soil testing labs, Eckert (1987) reported that 41 out of 43 of these public universities used the sufficiency level approach for making K fertilizer recommendations. The SLAN concept was developed by Bray (1944, 1945) in Illinois. In research with crops grown in four-year rotation, Bray found a correlation between the level of exchangeable K and crop yield when the level of exchangeable K was expressed as a percentage of K required to produce maximum yield. The maximum yield was taken to be the yield obtained on the experimental plots receiving adequate amounts of K. Bray 10 called the percentage yield obtained the “percentage sufficiency of the available potassium” (Bray, 1944). Bray observed that each additional unit of exchangeable K gave an increasingly smaller percent yield increase. Based upon his findings, Bray applied the Mitscherlich equation to describe this “law of diminishing returns” response curve and modified the equation into the following form: L08 (A'Y) = L08 A - 61b: [1] where A = yield obtained when K is not deficient, y = yield obtained at any other K soil test level, bl = the soil test K level, 01 = a proportionality constant for the crop. The proportionality constant is determined experimentally and varies by crop. Calculation of the proportionality constant, c1, requires values for A, y, and b to be known and involves the following assumptions (Bray, 1944): l) The yield obtained on the reference plot is the maximum yield obtainable, i.e. no additional yield increase would result fiom further additions of K, 2) The difference in yield between plots receiving K and plots not receiving K is due entirely to the potash, and 3) The total exchangeable K as measured by the soil test is a measure of the amount of plant available potassium. Bray (1944) also modified the Mitscherlich equation to account for x, the amount of fertilizer added to the soil. With the addition of this term, the Mitscherlich equation takes the form: L08 (A - y) = Log A — (orb: + ex) [2] where A = yield obtained when K is not deficient, y = yield obtained at any other K soil test level, b; = the soil test K level, x = amount of K added, c1 and c = proportionality constants for the crop and soil respectively. Bray (1945) noted that A is not the theoretical maximum yield when none of the essential plant nutrients are deficient, but 11 rather it is the yield obtained when the one nutrient under consideration is not deficient. Bray (1945) selected a 98% relative yield as the sufficiency level for the crops he studied. Today, most laboratories use a 95 to 99% relative yield as the sufficiency level (Eckert, 1987). Different Philosophies Used in Making Fertilizer Recommendations Three philosophies are commonly used today in the United States for making fertilizer recommendations. These are the basic cation saturation ratio concept, the nutrient sufficiency approach, and the nutrient build-up and maintenance plan. The basic cation saturation approach is used to establish designated or specific ratios among Ca, Mg, and K. Fertilizer recommendations are then made to achieve these ratios of basic cations in the soil. This philosophy is often used when lime recommendations are made and can be of benefit on soils having naturally low Mg levels. In the nutrient sufficiency approach, fertilizer recommendations are made to obtain a desired yield goal at the existing soil test level. This is a conservative strategy because the aim is to apply only the quantity of nutrient needed to achieve a desired yield goal, without attempting to increase the fertility level of the soil. In the nutrient build-up and maintenance program, a critical soil test level is established. The critical soil test level is considered to be that soil test value at which the soil can adequately supply nutrients to support optimum economic growth (V itosh et al., 1995). Below the critical level, a build-up level of fertilizer is recommended. That is, sufficient fertilizer is recommended to produce the desired yield at the existing soil test level and to raise the existing soil test value to the critical level over a period of several years. After the critical soil test level has been 12 reached, recommendations are made to compensate for crop removal and maintain the soil test value at or slightly above the critical level. This plan is generally used with three objectives in mind: 1) Increase the soil test value to the critical level and maintain it at that level. 2) “Banking” soil fertility, and 3) Build a margin of safety for protection against differential crop response in poor weather years (Eckert, 1987). Raising soil test values to the maintenance level also helps to safeguard against sampling and analytical variation (V itosh et al., 1995). Differences in Potassium Recommendations in the Tri-State Region Although 1M neutral ammonium acetate is used as a common extrath for K by all the states in the North Central Region (Brown and Wamcke, 1988) philosophical differences in soil test interpretation have resulted in markedly different K recommendations by some states. This has been the case in Michigan, Indiana, and Ohio. In the past, Michigan recommended build-up when the soil test K level was low but did not follow with a maintenance recommendation (Christenson et al., 1992). Michigan also never established critical soil test values for K. Ohio’s K guidelines included recommendations for soil buildup and crop removal. In addition, Ohio established a sufficiency level for K based upon the CEC of the soil (Anonymous, 1988). This approach was based upon the work of Fisher (1974) in Missomi. Indiana also has followed a nutrient buildup and maintenance concept and categorized their soils as deficient, adequate, optimum, or excessive according to the soil test K value. Indiana did not, however, use a CEC factor in their K recommendations except to recommend that 13 not more than 150 mm of K20 be applied to soils having a CEC of less than 6 cmoleg to prevent excessive losses of K due to leaching (Anonymous, 1992). One of the biggest differences in K recommendations between the three states has been Ohio’s use of the soil CEC to establish the sufficiency level for K. Prior to the mid 1970’s, 200 lb/A of exchangeable K was used as the sufficiency level in Ohio for all soils and all crops except alfalfa (McLean, 1976). This was based upon Bray’s data (1945) which showed that an exchangeable K value of 200 lb/A gave 95 to 98% maximum yield for corn, clover hay, soybeans, wheat, and oats. During the period from 1961 to 1971, the average level of exchangeable K of soils tested by the Ohio State University Soil Testing Laboratory increased fiorn 165 lb/A to 237 lb/A. This increase was attributed to the fact that additions of K through fertilizers and K released fi'om nonexchangeable forms was greater than K removed by crops or lost through leaching. In the four counties (Defiance, Paulding, Putnam, and Ottawa) in the Lake-Plain region of northwest Ohio, which have an average of 83% of soils that are fine textured, the level of exchangeable K did not increase (326 lb/A versus 328 lb/A) during this 10 year period. However, in the five other counties in the Lake-Plain Region, which have only 51% of soils that are fine textured, the level of exchangeable K increased from 188 lb/A to 240 lb/A. McLean (197 6) presented data, based upon the bonding strength of exchangeable K, that for Ohio soils there could be as much as a three-fold difference in the amounts of exchangeable K available to plants on soils with CEC’s ranging from 10 to 40 cmolo/kg. Earlier work by Conyers and McLean (1968) had shown that in clay separates of Ohio soils weathered with growing roots, the ability of the clay to fix K in the nonexchangeable form was greater than the clay’s ability to release K. Based upon these 14 findings and observations by university agronomists and local farmers, McLean felt that an increase in K fertilizer recommendations would be appropriate; although he admitted that it was not then known how large of an increase should be made in K recommendations or whether increased K recommendations would be suitable for all fine textured soils. In reflection of this philosophical shift, the Ohio State University Soil Testing Laboratory adopted the equation proposed by Fisher (1974) to establish the sufficiency level for K: K mm, .m. (ppm) = 110 + 2.5 x CEC [3] This formula has been in use in Ohio since the mid-1970’s and is still used as the basis for making K recommendations in that state. Reasons for Tri-State Fertilizer Recommendations There are at least three important reasons for the effort to develop common fertilizer recommendations for Michigan, Indiana, and Ohio (Vitosh, 1995). The first reason is that for many years, farmers and agribusiness companies that operate near the border areas have been interested in the development of common fertilizer recommendations that would provide for consistent management practices in neighboring areas. A second reason is that a common set of recommendations would provide more credibility for university research. The third reason is that over the past 20 years there has been a steady decline in the number of soil fertility specialists at public universities. Given the present economic and political mood in both the private and public sector, it is likely that this trend will continue. Therefore, it is very important that the scientific basis for fertililzer recommendations be well documented. A fourth reason for common 15 recommendations is the fact that Michigan, Indiana, and Ohio share common soils. Therefore, these soils should exhibit similar characteristics in K chemistry. The objective of this research project was to determine the degree of similarity in K chemistry of soils common to the three states. Examination of the K chemistry of soils collected from multiple locations throughout the tri-state area would provide support either for or against the assumptions underlying the tri-state recommendations. This objective was achieved through a combination of biological, chemical, mineralogical, and physical diagnostic approaches. The results of these studies are described in the following two chapters. 16 REFERENCES Anonymous. 1988. Ohio Agronomy Guide. 12th ed. Ohio Coop. Ext. Bull. 472. Anonymous. 1992. Fertilizer recommendations for agronomic crops. Purdue Univ. Agr. Dept. Rep. Agry 92-4. Barber, SA. 1968. Mechanism of potassium adsorption by plants. p. 293-310. In VJ. Kilmer (ed.) The role of potassium in agriculture. ASA, CSSA, and SSSA, Madison, WI. Bear, F.E., and A.L.Prince. 1945. Cation-equivalent constancy in alfalfa. J. Am. Soc. Agron. 37:217-222. Bear, FE. and SJ. Toth. 1948. Influence of calcium on availability of other soil cations. Soil Sci. 65:67-74. Bear, F.E., A.L. Prince, and J.L. Malcohn. 1944. The potassium-supplying powers of 20 New Jersey soils. Soil Sci. 58:139-149. Bear, F.E., A.L. Prince, and J .L. Malcohn. 1945. The potassium needs of New Jersey soils. New Jersey Agric. Exp. Stn. Bull. 721. Bertsch, P.M., and G.W. Thomas. 1985. Potassium status of temperate region soils. p. 131-162. In R.D. Munson (ed.) Potassium in agriculture. ASA, CSSA, and SSSA, Madison, WI. Brady, NO, and RR. Weil. 1996. The nature and properties of soils. 11th ed. Prentice- Hall, Upper Saddle River, NJ. Bray, RH. 1944. Soil-plant relations: I. The quantitative relation of exchangeable potassium to crop yields and to crop response to potash additions. Soil Sci. 58: 305-324. Bray, RH. 1945. Soil-plant relationships: 11. Balanced fertilizer use through soil tests for potassium and phosphorus. Soil Sci. 60:463-473. 17 Brown, J .R., and D. Wamcke. 1988. Recommended cation tests and measures of cation exchange capacity. p. 15-16. In W.C. Dahnke (ed.) Recommended chemical soil test procedures for the North Central Region. North Central Regional Publication No. 221 (Revised). Christenson, D.R., D.D. Wamcke, M.L. Vitosh, L.W. Jacobs, and 1.6. Dahl. 1992. Fert- ilizer recommendations for field crops in Michigan. Michigan State University Ext. Ser. Bull. E-550A. Conyers, E.S., and ED. McLean. 1968. Effect of plant weathering of soil clays on plant availability of native and added potassium and on clay mineral structure. Soil Sci. Soc. Am. Proc. 32:341-345. Douglas, LA. 1989. Vermiculites. p. 635-674. In J .B.Dixon and SB. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI. Eckert, DJ. 1987. Soil test interpretations: Basic cation saturation ratios and sufficiency levels. p. 53-64. In J.R. Brown (ed.) Soil testing: Sampling, correlation, calibration, and interpretation. SSSA Spec. Publ. 21. SSSA, Madison, WI. Eckert, D.J., and ED. McLean. 1981. Basic cation saturation ratios as a basis for fertiliz- ing and liming agronomic crops: 1. Growth chamber studies. Agron. J. 73: 795-799. Fanning, D.S., V.Z. Keramidas, and MA. El-Desoky. 1989. Micas. p. 551-634. In 1.8. Dixon and SB. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI. Fisher, TR. 1974. Some considerations for interpretation of soil tests for phosphorus and potassium. Missouri Agric. Exp. Stn. Res. Bull. 1007. F oy, CD, and SA. Barber. 1958. Magnesium deficiency and corn yield on two acid Indiana soils. Soil Sci. Soc. Am. Proc. 22:145-148. Graham, ER. 1959. An explanation of theory and methods of soil testing. Missouri Agric. Exp. Stn. Bull. 734. Haung, RM. 1989. Feldspars, Olivines, Pyroxines, and Amphiboles. p. 975-1050. In J. B. Dixon and SB. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI. Haung, P.M., L.S. Crosson, and DA. Rennie. 1968. Chemical dynamics of potassium release fi'om potassium minerals common in soils. Trans. 9th Int. Congr. Soil Sci. 2:705-712. 18 Huff, W.D. 1972. Morphological effects on illite as a result of potassium depletion. Clays Clay Miner. 20:295-301. Hunter, AS. 1949. Yield and composition of alfalfa as affected by variations in calcium- magnesium ratio in the soil. Soil Sci. 67:53-62. Hunter, A.S., SJ. Toth, and FE. Bear. 1943. Calcium-potassium ratios for alfalfa. Soil Sci. 55:61-72. Jackson, ML. 1964. Chemical composition of soils. p. 71-141. In F.E. Bear (ed) Chemistry of the soil. Reinhold Publishing Corp., New York, NY. Jackson, M.L., Y. Hseung, R.B. Corey, E.J. Evans, and RC. Vanden Heuvel. 1952. Weathering of clay-size minerals in soils and sediments: II. Chemical weathering of layer silicates. Soil Sci. Soc. Am. Proc. 16:3-6. Liebhardt, WC. 1981. The basic cation saturation ratio concept and lime and potassium recommendations on Delaware’s Coastal Plain soils. Soil Sci. Soc. Am. J. 45: 544-549. Marschner, H. 1995. Mineral nutrition of higher plants. 2nd ed. Academic Press, San Diego, CA. Martin, H.W., and D.L. Sparks. 1985. On the behavior of nonexchangeable potassium in soils. Comm. Soil Sci. Plant Anal. 16(2):133-162. McLean, ED. 1976. Exchangeable K levels for maximum crop yields on soils of diff- erent cation exchange capacities. Comm. Soil Sci. Plant Anal. 7:823-838. McLean, ED. 1977. Contrasting concepts in soil test interpretation: Sufficiency levels of available nutrients versus basic cation saturation ratios. p. 39-54. In T.R. Peck et al. (ed) Soil testing: Correlating and interpretating the analytical results. ASA Spec. Publ. 29. ASA, CSSA, and SSSA, Madison, WI. McLean, E.O., and MD. Carbonell. 1972. Calcium, magnesium and potassium ratios in two soils and their effects upon yields and nutrient contents of german millet and alfalfa. Soil Sci. Soc. Am. Proc. 36:927-930. McLean, E.O., R.C. Hartwig, D.J. Eckert, and GB. Triplett. 1983. Basic cation saturation ratios as a basis for fertilizing and liming agronomic crops: H. Field studies. Agron. J. 75:635-639. Prince, A.L., M. Zimmerman, and FE. Bear. 1947. The magnesium-supplying powers of 20 New Jersey soils. Soil Sci. 63:69-78. l9 Rasmussen, K. 1972. Potash in feldspars. Proc. Colloq. Int. Potash Inst. 9:57-60. Russell, E. W. 1988. Russell’s soil conditions and plant growth. 11th ed. John Wiley and Sons, New York, NY. Schroeder, D. 1974. Relationships between soil potassium and the potassium nutrition of the plant. Proc. Congr. Int. Potash Inst. 10:53-63. Schroeder, D. 1978. Structure and weathering of potassium containing minerals. Proc. Congr. Int. Potash Inst. 11:43-63. Scott, AD. 1968. Effect of particle size on interlayer potassium exchange in micas. Trans. 9th Int. Congr. Soil Sci. 2:649-660. Scott, A.D., and SJ. Smith. 1967. Visible changes in macro mica particles that occur with potassium depletion. Clays Clay Miner. 15:357-373. Sparks, D.L. 1987. Potassium dynamics in soils. Adv. Soil Sci. 6:1-63. Sparks, D.L., and RM. Haung. 1985. Physical chemistry of soil potassium. p. 201-276. In R.D. Munson (ed.) Potassium in agriculture. ASA, CSSA, and SSA, Madison, WI. Stewart, LA. 1985. Potassium sources, use, and potential. p. 83-98. In R.D. Munson (ed.) Potassium in agriculture. ASA, CSSA, and SSSA, Madison, WI. Tan, K.H. 1980. The release of silicon, aluminum and potassium during decomposition of soil minerals by humic acid. Soil Sci. 129:5-11. Tisdale, S.L., W.L. Nelson, and J .D. Beaton. 1985. Soil fertility and fertilizers. 4th ed. Macmillan Publishing Company, New York, NY. Vitosh, ML. 1995. Tri-state fertilizer recommendations position paper. Michigan State University Crop and Soil Sciences Dept. Vitosh, M.L., J .W. Johnson, and DB. Mengel. 1995. Tri-state fertilizer recommendations for corn, soybeans, wheat and alfalfa. Michigan State University Ext. Bull. E-2567. CHAPTER 1 POTASSIUM RELEASE FROM EXCHANGEABLE AND NONEXCHANGEABLE FORMS IN MICHIGAN, INDIANA, AND OHIO SOILS AS MEASURED BY CHEMICAL EXTRACTION AND NEUBAUER TECHNIQUES ABSTRACT There are no papers in the literature that compare the potassium (K) chemistry of soil series common to Michigan, Indiana, and Ohio. In 1995 and 1996, 54 soils from this tri-state region were collected for study. The objectives of this investigation were: 1) to determine the degree of similarity in the K chemistry of soils common to the three states; 2) to determine the nature of K release during intensive cropping; and 3) to determine whether ammonium acetate (NH40Ac) or sodium tetraphenylboron (N aBPh4) would better predict plant available K for soils of the tri-state area. Cation exchange capacity (CEC), total K, NH40Ac and NaBPh4 extractable K levels were determined for all soils. Neubauer experiments with barley (Hordeum vulgare cv. Bowers) were conducted in a growth chamber to investigate K release during intensive cropping. Large differences in CBC, total K, and nonexchangeable K release were found between soil series. The finest textured soils had higher levels of total K, and released more nonexchangeable K than soils having less clay. Exchangeable K values were much lower than nonexchangeable K values and were not significantly different between soil series regardless of texture. The contribution of exchangeable K to total plant uptake ranged from 17 to 75% between soil 20 21 series. Regression analysis did not show NaBPh4 to be a better index of plant available K than NILOAc. Although large differences in K chemistry exist between soil series, individual soils in a given series will generally exhibit similar K chemistry regardless of geographic location. These findings suggest that regional K fertilizer recommendations can be made for common soil series in the tri-state area. INTRODUCTION The improvement of analytical methodologies to accurately measure the bioavailability of soil nutrients is an important objective for soil fertility specialists and soil testing laboratories throughout the US. and the world. Accurate measurements of plant available potassium (K) in soils are complicated by fire fact that K equilibria in soils are dynamic. Potassium can be released and fixed by soil minerals (Mortland et al., 1957; Ross et al., 1989). Potassium exists in four phases in the soil: solution, exchangeable, nonexchangeable, and mineral (Sparks, 1987). Although soils may contain large amounts of K, only a small portion is immediately available to the plant. Typically, solution and exchangeable K are only 0.1 to 2% of total K (Tisdale et al., 1985). Under field conditions, the roots of annual crops occupy a volume that is only 0.4 to 2% of the soil volume in the A horizon (Barber, 1968). Therefore, if a chemical extractant is to provide a reliable index of plant available K, it must extract a quantity of K that will correlate well with the amount of K taken up by the crop. Ammonium acetate, NI-140Ac, which measures exchangeable K levels, has traditionally been used to estimate plant available K. However, plants also utilize a portion of the nonexchangeable K released from 2:1 clay minerals (Mengel and 22 Rahmatullah, 1994). In the past, various extractants have been used to estimate plant available nonexchangeable K. These have included, among others: boiling HCl and HNO;, sodium tetraphenylboron (N aBPh4), and Ca-saturated and H-saturated cation exchange resins (Sparks, 1985). Sodium tetraphenylboron was chosen in this study because it mimics the action of plant roots by lowering the concentration of K in solution. The NaBPh4 procedure was developed by Scott and co-workers at Iowa state during the late 1950’s and 1960’s. In their pioneering work, they used NaBPh4 to extract K from soil and pure mineral systems (Reed and Scott, 1966; Scott and Reed, 1962a,b; Scott and Welch; 1961, Scott et al., 1960, Smith and Scott, 1966). Sodium tetraphenylboron lowers the solution K concentration by precipitating K in solution with the BPh4' anion to form KBPh4 while the Na acts as an exchanger for K. The excess NH4 in the quenching solution terminates the reaction by saturating the exchange sites with N114, thereby inhibiting further release of K. Potassium is released into solution by dissolution of the KBPh4 precipitate by boiling, and Hg or Cu is used to destroy the RPM anion. Although common soil testing procedures for K are used throughout the North Central region of the US. (Brown and Wamcke, 1988), philosophical differences in soil test interpretations have caused some states to recommend greatly different rates of K fertilizers. Historically, this has been the case in Michigan, Indiana, and Ohio, but in the past several years, soil fertility specialists from Michigan State, Purdue, and Ohio State Universities have worked together to produce a common set of fertilizer recommendations for the tri-state region. The result of this collaborative effort has been the recent publication of an extension bulletin entitled “Tri-state fertilizer 23 recommendations for corn, soybeans, wheat and alfalfa” (V itosh et al., 1995). As the new tri-state fertilizer recommendations were developed, it was decided that additional work would be needed to determine the adequacy of the new recommendations. This work, then, is a part of the ongoing effort to provide the research data necessary to evaluate the new regional approach to K fertilizer recommendations in the tri-state area. The objectives of this study were: 1) to determine the degree of similarity in K chemistry of similar soil series from the tri-state region; 2) to determine the nature of K release of these soils during intensive cropping; and 3) to determine if NI-I40Ac or NaBPh4 provides a better index of plant available K. MATERIALS AND METHODS Soil Selection, Collection, and Preparation Soils were collected fiom sites throughout Michigan, Indiana, and Ohio in 1995 and 1996. Sites were selected by first using a generalized state soil association map to identify those counties within each state where the soil associations of interest were extensive. Individual sites were then chosen based upon the more detailed information in the respective county soil survey reports. The counties fiom which soil samples were collected are shown in Figure 1.1. The location of individual soils and their soil family classification are listed in appendix A. Samples were taken from locations as close as possible to the site of the representative pedon to minimize potential difl‘erences in soil characteristics. The soils in this study were chosen based upon the extensiveness of their distribution throughout Michigan, Indiana, and Ohio and their agricultural importance. At the center of the study are two soil association groups; the Miami, Conover, Brookston 1.wi ..,.l... I .1 ...... I it.-- . - s 9 15.01 {Fair 15.5.5.0 9 25.81 3:42)! 0 C VCPdO N on «u h. <>z 3.4+ 2.2:. «a u (Bin 32.: acres-5.: J s so a. m M W: h. 7% u. a in: 2 co» 3 An N Exam 9mm M1 .w4 h L WW .5; W 9 0.5on N E 2.0.: o w 1 z :9.sz W m 99 re .69).: 2:55 M M 3232 co en. o am. $01 E a on)»: r 1 M szx I r a 19233 on e a»: a 4.3m- a.e :2 5 0+ 009:. 03:33: m. 519 m < 953 0:232 .2302 4». arm; \o+ sea. or . N by 01):: 0 24 l .w 9 :936 as 44 .353 20+ or e eszxm 2.52 « 4+ be 2.02.7 0035 9.? m. 2:)... «o 2:25.: see o 0:21 \ e xv... :mza‘ n25: .w 235 :oznOn :>z- <§: 9. 20mm 52.32 ) 01m: 25:... N Am: 0% 03th 069 00 23:50: w 1.94.920 «:0 {new 94 w a. s a 3 Vin A. I o a o a}? p. a é um. canmzn h ~14 Q w 2622 a. s a. :6sz AW 1. m. Sim 2203 SF.» 99' J 9001 wiping r) 4 0 fl 9% vfioaa. «A? 2 4w «a 1 09 szx a. W hue wool .64 a. 0320 are, 1. .1 09 3.3 0:36 .0: Mm y 06.32 in: e m DQ— r + r o e. M $470 0%. to». a. e v 4 Emfid _. _ . H1388 Bee o». 7&2:me Fae—bu. we; OE? Oogmom HEB can: we: menu—am $08 8:88.” Ea uses: 5 one—on. 18mmOCn arm e be» ._. a. 0.830 5. >2 2. +9.00» >POOZ> 0:20 «I»: I‘M tnx‘Ol—u Ian: )0 as 9e 0 022204 as? .3. o... 022520 O>zn< M>4OZ .201)! h>oxu°2 02.1062 rn2>inn 25 soil association group and the Morley, Blount, Pewamo soil association group. These two soil groups form two toposequences, meaning that they are associated with each other in the landscape and were formed in the same parent material. Their main differences are topography and drainage class. The Miami and Morley soils are well drained, the Conover and Blount soils are somewhat poorly drained, and the Brookston and Pewamo soils are poorly drained. The Miami, Conover, and Brookston soils were formed in loam glacial till whereas the Morley, Blount, and Pewamo soils were formed in clay loam or silty clay loam glacial till (Smeck et al., 1968). The Morley, Blount, Pewamo soil association is the largest soil association in Ohio (Dr. George Hall, personal communication). The Marlette, Capac, Parkhill soil association group, which is found only in Michigan, was selected because it occupies extensive portions of central and southern Michigan where intensive agriculture is prevalent. The Hoytville, Misteguay, and Paulding soils, although not found in all three states, were selected because of their fine texture and high K contents. A representative sample from the Ap horizon of each soil was collected in 5 gallon buckets and transported to Michigan State. Bulk samples were air dried and sieved through a 1/4” wire mesh screen prior to mixing of the soil. Bulk samples were transferred to a cement mixer and mixed for 10 minutes. A representative subsample was then taken and dried at 35° C for 24 hours. After drying, samples were ground with a flail grinder to pass a 10 mesh (<2 mm) sieve. The <2mm fiaction was mixed for 10 minutes in a twin shell dry blender to ensure complete homogenization of the soil. After 26 mixing, the samples were placed in waxed cardboard cartons and stored until chemical analyses were performed. Chemical Analyses Exchangeable K (Brown and Wamcke, 1988) was determined with 2 gram samples of soil weighed in duplicate and placed into 125 mL erlenmeyer flasks. Twenty mL of 1 M neutral NH40Ac were added and samples were shaken for 5 minutes on a reciprocating shaker at 200 oscillations per minute. After extraction, samples were filtered through No. 2 Whatrnan filter paper. The K concentration in the extract solution was determined with a Technicon autoanalyzer flame photometer. Exchangeable plus nonexchangeable K was measured using a modified NaBPh4 procedure based upon Cox et a1. (1996). One-half gram samples were weighed in duplicate and placed into 50 mL folin wu tubes. Three mL of a solution containing 1.7 M NaCl, 0.167 M NaBPl'u, 0.01 M EDTA were added to the samples. The samples were swirled gently and allowed to sit for the desired extraction period (five minutes or seven days). After extraction, 25 mL of a quenching solution containing 0.5 M NH4C1 plus 0.11 M CuClz was added to terminate the release of K. Samples were then placed on a digestion block and allowed to boil gently at 150° C for 30 to 45 minutes or until the KBPh4 precipitate was dissolved. After boiling, 15 drops of 6 M HCl were added to the samples to prevent precipitation of Cu and brought to a 50 mL volume with deionized water. A 20 mL aliquot was taken after the samples had clarified sufliciently for analysis. The K concentration in solution was determined with a model 5 lCa Perkin Elmer flame photometer. 27 Finely ground (<100 mesh) 0.1 gram samples were weighed in duplicate and analyzed for total K using the HClOrzHF double acid digestion procedure as described by Knudsen et a1. (1982). Percent organic matter was determined colorimetrically with duplicate 1 gram samples by heat of dilution with concentrated H2S04 (Schulte, 1988). Duplicate 5 gram samples were used to determine soil pH using a 1:1 soil to water ratio (Eckert, 1988). Cation exchange capacity (CEC) was determined by a centrifugation method with duplicate 2 gram samples (Wamcke, 1997). Twenty mL of 1 M neutral NH40Ac were added to the soils. The samples were mixed with a vortex mixer and centrifuged for five minutes. After centrifirgation, the supernatant was discarded. This was done three times. Samples were washed free of excess NH4+ by adding 15 mL of 95% ethyl alcohol, then mixing and centrifuging the samples. The supernatant from the alcohol washings was discarded. This was done three times. The NH.+ was displaced by adding 10 mL of a 10% acidified NaCl solution and 5 mL of deionized water. The samples were mixed, centrifuged, and the supernatant transferred to a 50 mL volumetric flask. This was done three times. The samples were then brought to volume. The displaced NI-L.+ in the extract was measured with the LaChat system. Neubauer Studies A Neubauer method modified fiom that of McGeorge (1946) was utilized to study the K release characteristics of the soils during intensive cropping. One hundred grams of oven dry soil were placed into four-inch clear vinyl saucers and seeded with 110 spring barley (Hordeum vulgare cv. Bowers) seeds obtained fiom Michigan Foundation Seed, Lansing, Michigan. The barley seeds were wrapped in moist paper towel for 12 28 hours prior to seeding to imbibe water and speed germination of the seeds. Eighty-five grams of soil were placed below the seed and 15 grams of soil were placed on top of the seed to promote better contact between the soil and the seed. Twenty-eight mL of water were added to the saucers. The soil surface was then covered with 45 to 50 grams of polypropylene plastic beads to reduce evaporative losses from the soil surface and to bring all pots to a uniform weight. Each soil was replicated four times. Pots were watered twice a day with a nutrient solution containing 100 pg ml'l nitrogen as NH4N03 and 20 pg ml'1 phosphorus as Ca(H2PO4)2 to maintain optimum soil moisture content. Plants were grown in a controlled environment growth chamber for 22 days after seeding. The daylength was 16 hours with 85% relative humidity. The daytime temperature was 73° F with a nighttime temperature of 68° F. The average photon flux during the experiments ranged fiom 364 to 389 umol m‘2 5". After 22 days the plants were harvested by clipping the shoots at the soil level. The moist soil was separated fi'om the roots as efficiently as possible. The roots were then placed on a 60-mesh sieve and washed fiee of adhering soil under a stream of running tap water. The shoot and root tissue samples were placed in labeled brown paper bags and dried at 60° C for 48 hours. After drying, shoot and root tissue samples were ground using a cyclone grinder. One- half gram tissue samples were weighed into crucibles and ashed in a muffle furnace at 500° C for 5 hours. After ashing, the samples were taken up in 25 mL of 3 M HNOg, filtered, diluted, and K was determined by flame photometry. Potassium uptake was calculated according to the following formula: [(shoot dry weight "‘ shoot K concentration) + (root dry weight "' root K concentration)] - K content of the seed (15.49 mg K per seed lot). This value was then multiplied by 10 to convert the K uptake per 100 29 mg soil to K uptake per kg soil. The recovered soil from the pots was dried at 35° C, mixed, and subsamples taken to determine NI-LOAc and NaBPh4 extractable K levels after cropping. Mean separation and regression analysis of the data was done using the Statistix software program (Analytical Software, 1996). RESULTS AND DISCUSSION Chemical Analyses A summary comparison of the amounts of K extracted by NaBPh4 and NILOAc for all 54 soils is presented in Table 1.1. As expected, NaBPh4 extracted more K. The Table 1.1. Comparison of the amounts of K extracted by NaBPh. and NI-I40Ac in 54 soils from the tri-state area.1' K Extractable K Extractable K Extractable 7 Day:5 Minute 5 Minute by NaBPh, by NaBPh. by NI-LOAc NaBPh. NaBPherI-LOAC in 7 days in 5 mimrtes in 5 minutes Extractable K Extractable K % of total K Ratio low 10 1.5 0.4 7.6:] 2.1:] high 55 8.6 2.5 74.8:1 8.3:] average 29 4.6 1.1 30.8:1 4.6:1 tValues listed in table are based on extractable K levels prior to initiation of Neubauer growth studies. amount of K extractable with NaBPh4 in seven days ranged fiom 10 to 55% of the total K. These values are similar to those reported by Smith et a1. (1968). In their study, 20 to 47% of the total K was extracted by NaBPh4. Data by Cox et al. (1996) showed that a seven-day extraction period was sufficient to release nearly all of the K extractable by NaBPh4. These data are important because they demonstrate the strong K extracting 30 ability of this reagent, and they measure the long-term available K reserves of representative soils of the tri-state area. The five-minute NaBPh4 extractable K is useful fi'om a soil fertility standpoint because it more closely measures that portion of the nonexchangeable K that would be readily plant available. With a five-minute extraction, the values are much lower, ranging fi'om 1.5 to 8.6% of the total K. The average value across all soils was 4.6%. The exchangeable K was even less, ranging fiom 0.4 to 2.5% and averaging only 1% of total K. The ratio of the seven-day to five-minute NaBPh4 extractable K is interesting because it indicates the relative proportions of K released in a short time versus the amount of K released over a longer time. This ratio provides a comparative measure of the quantity of nonexchangeable K that would be more readily plant available versus the amount of nonexchangeable K that might become available over the long-term through weathering. Presumably, the lower the ratio, the greater the percentage of nonexchangeable K that would be readily plant available. It is interesting to note that although the average ratio is 30.8: 1 , there is nearly a lO-fold difference in this ratio between the highest and lowest soils. From a soil fertility standpoint, it is also important to compare the ratio of the five-minute NaBPh4zNH40Ac extractable K. On the average, NaBPh4 extracted 4.6 times as much K as NI-hOAc in five minutes. Once again, this difference demonstrates the ability of NaBPh4 to extract large quantities of nonexchangeable K from the soil. An important objective of this study was to determine the degree of similarity in the K chemistry of soils in the tri-state area. Differences in total K content are quite evident when all soil groups are compared (Table 1.2). The Paulding type soils contain the greatest amount of total K, 28,700 mg K kg", whereas the Miami soils contain the 31 as... sass: ages Ens: i=8 058.3— 00.0205 cone—Eng 230 8:00 .05 .385 33.05 05.50”» «=8 .8320 832: 382: new 3.352 as 323.. cases 32.5% .9 £5035 oaofz .33.. a: 2:52-26 0— 033530 n 03; .M 030038330 05 3:3. M 03800.30 3:502 0SEE0>E n 0.02 0552033 .M 0300w§30x0 05 2.38 M 03800.55 3:502 BOA—05m u 0.02 55.553 .508 05 .3 8:0 Rave—am H mm._. ._0>0_ b:3038a «ad 05 «a 303%? 335$?me 3: 0.3 00:0. 058 05 c3 330:8 2302... a. 88 N: 33 «2 25am «3 88.2 E on: a Fan: em 82 S £38 an 323. 2.. 88.2 3 £8. a. 3:39.85 2 8: e: 8% 3.3 084.4 «a; 38.0. 3 33.. m E0880 :. 8: a 8: a: 82... «.8 38m: 3 3n: n 0:232 8 82 a: as: S. 083 80.. 258.: 2 25.2 N :35 Q. 80. on £82 02. 2:3 82 £83. 2 3.2 m 8&0 M: «SN 3 £80 m: 2:3 N2 384.2 2 34.2 a .555 a. 88 E as: m a 88a 83 38...”. 2 3.2 a £0.32 a. 88 8 as; a... 25:6 83 88.2 2 8.: o 833$ 3 as: 2.. 8n 3 82 E85 8} 88.: 3. £3 N 053.8: m 38 we. 3%.. 83 83.2 8m 33% 3 .33 N 3523.. _. 0. v— wE _- we. 93:8 mm 5...: mm :82 mm 502 mm 50: :5 522 new no 320 Sam eaz asszaaa v.02 seesaw a .53 omo .Beaz =5 .8380 532w 0:03:02 0.6.303 3.8% =8 .3 8.. M 030033380 .05 .M 0333—3805: .M .88 .85 .3 woe—g .802 .N._ 030.—. 32 least amount of total K, 15,900 mg K kg". The soils follow a definite trend with the finer textured soils containing more total K. This suggests that soils with a higher clay content will contain more total K than soils with a lower clay content. The cation exchange capacity does not follow such a clear trend for the soils in the middle although the soils at the extremes, the Paulding and Miami soils do show distinct differences in CEC. Cation exchange capacity, total K, seven-day and five-minute nonexchangeable K, and exchangeable K data for individual soils are presented in appendix B. The seven-day nonexchangeable K follows a pattern very similar to that exhibited by total K. There is a very consistent trend for the soils high in total K to release large amounts of nonexchangeable K to NaBPh. extraction. This would be expected since NaBPh4 is very effective in removing K from the interlayers of the 2:1 clay minerals. There is more than a three-fold difference in seven-day nonexchangeable K release between the Miami and Paulding soil groups. This indicates that the long-term K supplying power is less for the Miami soils and that these soils would require more frequent K fertilization to maintain high levels of fertility. The five-minute nonexchangeable K also exhibits a pattern similar to that for the seven-day nonexchangeable K. These trends suggest that NaBPh4-K levels could be a reliable index of the K supplying power of these soils. The exchangeable K levels of these soils show no significant differences between any of the soil groups. The soils with low nonexchangeable K values have exchangeable K values comparable to the exchangeable K values of the soils high in nonexchangeable K. These results emphasize the fact that exchangeable K values alone give little 33 indication of the ability of these soils to release nonexchangeable K nor does it provide any information on the rate of nonexchangeable K release. After consideration of the data from all the soils in the tri-state study, the second important issue is to determine what differences are present between the six soil series common to the three states. These data are presented in Table 1.3. The finer textured Morley, Blount, and Pewamo soils are higher in total K than the Miami, Conover, and Brookston soils. The CEC for these soils does not follow any clear pattern analogous to total K. However, the finer textured soils do have a slightly higher average CEC than their coarser textured counterparts. There is nearly a two-fold difference in the mean CEC, 21.0 versus 11.3 cmoL kg'l, between the Pewamo and Miami soil groups respectively. It is worthwhile to mention that the poorly drained Pewamo (4.0% average organic matter content) and Brookston soils (4.3% average organic matter content) have the highest CBC and the highest organic matter content of any of the soil groups. However, one would expect these soils to contain more smectite which would also increase the CBC. The relative importance of both the organic and inorganic contribution to total CEC is investigated in more detail in chapter two. Differences in seven-day nonexchangeable K were not statistically significant among the six soil series; although there is almost a 1.8 times difference between the highest and lowest values. This lack of significance is due to the large variability present within the soils of some groups. This is reflected in the large standard error values. This is particularly true for the Conover group where the standard error is nearly 1/4 the value of the mean. The five-minute nonexchangeable K values also do not show significant differences when only the six soil groups are considered. It is to be expected however .m=8 .3832 83.05 25:2 3 Ewen Que—ov— movaos 5.9.85 3 «:8 «2:00 28 38.5 wows—05 55.80 .2. £8 Basic 3.5.2: $82 a .0.— 038855 94552 395:: _ 05.52-0sz M oanwcusoxm u vim? .M 0383205 05 3&8 M 038035 Emmaz «SEEbzm u :02 Safizozma .v. ozwowcagoxo 05 358 M 035255 Ham—«Z hon—-533 n :02 39-553 .52: 2.. .6 85 2855 u mi .36. 3.3395 3.: 05 as Bang—u ragga»? 8: 0.8 830— 08% 05 3 332—8 2802.. 34 a. 88 N: 89. 3 an} N? .58.: E on: a $.55 a 83 S 88 an 85... was 88.2 3 £5: a. #8585 2 8: e: 8% 33 «See an? 88.2 M: on: w F388 2 88 ma 8% m; 2: 3 N2 fi§£ 2 2:..2 a .555 a. «8.6. E 8:. ma 83x 83 958.2 2 8.2 5 £3.32 a. 88 me 88 an 826 as; 83.2 mm .8: e c530; Two— M we _. wx JoEo mm :82 mm :82 mm :32 mm :8: ma 532 23 co e55 gm 302 2:52.05 v.02 balsam M 5.: 08 59:52 :8 .motom :8 SEE .3 39ch 98 £5 .83 88.0.-5 05 8 .5883 8:8 :8 iv. 05 8a mom—cam 538w 6:332 282. M oanwSono can .M oEuowaasoxoco: J— 33 .85 mo 32: :32 .m._ 035. 35 that the differences between the soil groups would be less when the groups with the highest values for total K and seven-day and five-minute nonexchangeable K (e.g. Paulding and Hoytville) are not considered. Although the seven—day and five-minute nonexchangeable K values do not show statistically significant differences, they do follow the same pattern as the total K values. The Pewamo soils that are highest in total K have greater amounts of nonexchangeable K. The exchangeable K values show no statistically significant differences when only the six soil groups are considered. Comparing the six common soil series across the three states addresses the question of whether the soils of Michigan are different from the soils of Indiana or Ohio. In Table 1.4, all soils from the six common soil series are compared by state. Table 1.4 shows that there are no statistically significant differences by state in the mean values for CBC, total K, seven-day and five-minute nonexchangeable K, or exchangeable K when the six common soil series are grouped together. To further address the question of whether the soils are similar, the common soil series were individually compared across states. These data are presented in Tables 1.5 to 1.9. There were no significant differences in CEC, total K, seven-day and five-minute nonexchangeable K, or exchangeable K values between the three states for any of the individual common soil series. These data support the philosophy that there is a sound basis for a regional approach to the development and utilization of tri-state fertilizer recommendations. Another issue of interest is the question of whether the soils in Michigan would exhibit differences in K chemistry based upon their location in the state. To address this question, seven soil series common to the Michigan, Saginaw, and Huron-Erie glacial lobes were selected. The bedrock of Michigan is shaped like a bowl. Therefore, our 36 .Q 23883 005.12 EREE 3:52-03”: 0. 0380558 n 03:. .0. 038098388 05 0:38 M 03808.28 3.582 0332-0zm n x02 3:52-350 .M 0300w830x0 05 0558 v— 0380858 37582 mam—égow u 0.02 80-5803“ .808 05 .3 09:0 @8285 u mi .35. mum—3838.3 mod 05 8 0:20.50 32805ch 8: 0.8 5:28 0:80 05 3 0802 u 0.: 8 258 a 252. 8.. . 803..“ 8... 2808 f 802 2 2.5 8 SN 8 8m 80 83. 80 83: E 2.. 2 «585 2 SN 3 08 an 9.3. m; 88.: 2 3. t 50222 7w: M we _- mo. 0380 mm :32 mm :82 mm :32 mm 80: :5 :82 88 8 3% gm 302 3:52-38 3.2 8.5.550. 0. .38. 000 30:52 .0800. 3 @0953 0.8 mzom 80.8 0088.5 05 8 .8888 00:0» :8 30. 05 com 00330. 538w 00:83:02 0.8.33 M 0380w830x0 28 .M 030032055: .v— .32 .Umu .3 00:3.» .802 .v._ 035. 37 Table 1.5. Cation exchange capacity of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. CEC Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE cmolc kg'l Miami: 7 8.2 0.6 12.8 1.9 14.6ns 2.8 Conover§ 8 15.0 5.0 12.9 0.5 14.6ns 2.0 Brookstonfi 7 15.7 2.4 19.1 0.2 26.1ns 5.0 Morley# 7 10.9 1.2 15.9 2.7 1 1.8ns 4.4 Blount 7 15.4 2.1 17.5 0.9 13.5ns 2.3 Pewamo 6 20.7 9.3 25.0 2.5 17.3ns 1.1 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean. tMiami includes Miamian soils. §Conover includes Crosby and Celina soils. 1lBrookston includes Kokomo soils. #Morley includes Glynwood soils. Table 1.6. Total K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Total K Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE mg K kg" Miami: 7 15,700 406 13,900 500 18,300ns 2,750 Conover§ 8 17,700 3,490 14,800 950 16,600ns 353 Brookston1] 7 16,000 666 15,100 700 17,700ns 150 Morley# 7 18,000 2,730 18,000 950 19,500ns 1,600 Blount 7 18,800 296 17,700 350 18,500ns 1,200 Pewamo 6 19,900 3,600 19,900 1,400 19,500ns 700 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean. IMiami includes Miamian soils. §Conover includes Crosby and Celina soils. 18mkston includes Kokomo soils. #Morley includes Glynwood soils. Table 1.7. Sevenoday nonexchangeable K content of the six soil series common to 38 Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Seven-DaLNonexchangeable K Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE mg K kg'l Miami: 7 2,890 757 2,030 210 5,640ns 2,030 Conover§ 8 6,080 2,480 1,860 390 4,410ns 640 Brookstonfl 7 4,470 580 2,660 310 5,600ns 440 Morley# 7 5,020 1,590 6,390 1,280 5,860ns 2,680 Blount 7 5,370 108 4,590 625 5,260ns 1,660 Pewamo 6 5,430 1,990 6,490 730 GfiZODS 320 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean. IMiami includes Miamian soils. §Conover includes Crosby and Celina soils. 1IBrookston includes Kokomo soils. #Morley includes Glynwood soils. Table 1.8. Five-minute nonexchangeable K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. F ive-Minute Nonexchangeable K Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE mg K ks" . Miami: 7 270 72 290 145 880ns 570 Conover§ 8 710 373 230 10 540ns l 10 Brookstonfil 7 680 106 440 75 740ns 50 Morley# 7 550 208 890 45 760ns 465 Blount 7 720 104 550 5 550ns 245 Pewamo 6 920 180 970 145 860ns 5 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean. tMiami includes Miamian soils. §Conover includes Crosby and Celina soils. 11Brookston includes Kokomo soils. #Morley includes Glynwood soils. 39 Table 1.9. Exchangeable K content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Exchangeable K Soil Number Michigan Indiana Ohio GrouL of Soils Mean SET Mean SE Mean SE mg K kg" Miamit 7 180 50 210 130 240ns 125 Conover§ 8 130 33 120 0 150ns 15 Brookstonfl 7 230 63 210 0 230ns 15 Morley# 7 220 107 200 60 170ns 90 Blount 7 240 33 200 5 180ns 40 Pewamo 6 200 20 320 140 270ns 55 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean. {Miami includes Miamian soils. §Conover includes Crosby and Celina soils. 11Brookston includes Kokomo soils. #Morley includes Glynwood soils. 40 .AM 0880880 00. 8.828.... 3.: 0... 8 80.0%... b80058... 8.. 0.8 .030. 088... 05 .3 8030...... 0.802... cm .50. em 3mm can 03 m... 89. .536. h. . and. \. 0.5-8.8: an 8ch. 3 33. .3. 83m... max .836. m. 8N.N. h Bafiwam mm 02.... on. 38 cc... .536 cm ... «and. ad .83... h 8.8.8.2 ..w.. M w8 .. mg. 0880 mm .802 Mm .802 mm .802 mm .802 "mm .802 0.8m 8 2.3 8.06 8.07. 0.8.202... 9.02 8D-..0>0m M .88... UmU .382 8.08.0 8.88.8.2 .8 00.8. 8.08.» 00...... 0... o. .8888 00.80 .80 ..0>0m 0... 8.. 00830 530.8 8.8.802 08.8.. M 0388.858 8.8 .M 038098.888: .M .88. .UmU .8 003.8. .802 .o... 088... 41 hypothesis was that the soils in the Michigan and Huron-Erie glacial lobes are similar because the Wisconsin glacier passed over the same layers of bedrock in the western and eastern parts of the state. Differences, if any, would be expected in the Saginaw lobe because younger bedrock layers are present. The data in Table 1.10 reveal that for the seven soil series selected from each lobe the only statistically significant difference was in the mean total K content of the soils. The soils from the Michigan lobe had a higher total K content than the soils from the Saginaw and Huron-Erie glacial lobes. Although the differences were not statistically significant, the soils in the Michigan lobe also had higher CBC, seven-day and five- minute nonexchangeable K, and exchangeable K values. Neubauer Studies The second objective of this study was to determine the nature of K release during intensive cropping. Many researchers have conducted intensive cropping experiments to measure the capability of soils to release K (Addiscott and Johnston, 1975; Mutscher and Tu, 1988; Pratt, 1951; Singh and Ghosh, 1984; Tabatabai and Hanway, 1969). Traditionally, these studies have often used large quantities of soil which require long cropping periods, months or years, in order to lower the soil K levels through plant uptake. The advantages of the Neubauer procedure are that it greatly reduces the length of time needed to conduct a study by eliminating the necessity of multiple harvests to achieve high rates of nutrient removal, and it does not require large quantities of soil. The Neubauer method has been used in Indiana (Thorton, 1931; Thorton, 1935) and in Ohio (Olsen and Shaw, 1943) to evaluate the fertility status of soils. However, these 42 studies were conducted over 50 years ago. Except for the work of Binnie and Barber (1964) there are few, if any, more recent reports in the literature on utilization of the Neubauer technique for studying the nutrient supplying capabilities of soils in this tri- state area. In light of the fact that many agronomic practices have changed in the intervening half-century, it is obvious that there is a need to reevaluate the fertility status of common soil series in this tri-state region. Table 1.11 presents summary data of the amounts and percentage of uptake from exchangeable and nonexchangeable K by barley plants during the Neubauer growth Table 1.11. Amount of K taken up from the exchangeable and nonexchangeable soil K fractions by barley plants during Neubauer growth studies. Exchangeable Nonexchangeable Uptake from Uptake from K uptake? K uptake: Exchangeable K§ Nonexchangeable K1 mg K kg'l % low 20 6 1 1 10 high 350 387 90 89 average 111 116 53 47 TExchangeable K uptake = Difference between NEOAc-K before and afier Neubauer growth study (Delta Nl-LOAc-K). INonexchangeable K uptake = Plant uptake K minus delta NI-hOAc-K. §Uptake from exchangeable K = (Delta NI-LOAc-K divided by plant uptake K) x 100. 1]Uptake fi'om nonexchangeable K = (N onexchangeable K uptake divided by plant uptake K) x 100. studies. Values for individual soils are listed in Appendix B. Although there was a wide range in uptake values from the exchangeable K, 11 to 90%, and nonexchangeable K, 10 to 89%, respectively, the magnitude of the differences were similar and the average values were nearly identical. On the average, the barley plants utilized the exchangeable and nonexchangeable K in nearly equal proportions, 53% versus 47% respectively. The high and low values from this study cover the range of percent nonexchangeable uptake 43 reported in the literature. Niebes et al. (1993) reported that nonexchangeable K accounted for 65 to 80% of total uptake by rape. Menon et a1. (1988) found that nonexchangeable K accounted for 22 to 79% of total uptake by the end of four harvests for grass, legume, and cereal crops grown in the greenhouse. Tarfadar and Mukhopadhyay (1989) indicated that in 19 soils from West Bengal nonexchangeable uptake averaged 63%. Tening et al. (1995) reported that in the Nigerian soils they studied stylo (Sorlosanthes hamata cv. Verano) plants received 23 to 84% of their K from the nonexchangeable forms during exhaustive cropping. In Table 1.12, the average dry weight, percent K, and total K uptake of the barley shoot and root tissue are reported. The root dry weights are slightly lower than the actual amounts due to incomplete (approximately 85%) recovery of all the roots. In these studies, the K shoot concentrations ranged from 0.72 to 2.77% and averaged 1.61%. The K root concentrations were more constant, ranging from 0.25 to 0.55% with an average value of 0.39%. Wamcke and Barber (1974) reported that the proportion of K in corn roots grown in nutrient solution was 11% of the total at 18 to 25 days. Since most of the roots were recovered and because the K concentration in the roots is very low, the K uptake values reported are probably less than 5% below the actual uptake values. In Table 1.13, mean plant uptake values and the percent of uptake from exchangeable K for all the soil groups are presented. The percent of uptake fiom exchangeable K was calculated as the difference in the exchangeable K level before and alter the Neubauer growth study, divided by the total plant uptake. The Paulding soil group had the highest plant uptake, 371 mg K kg" soil. The Conover soil group had the lowest plant uptake, 139 mg K kg'l soil. Although there were 44 Table 1.12. Average dry weight, percent K, and total K uptake of harvested barley shoot and root tissue from the 54 soils used in the Neubauer growth studies. Shoot Root Shoot Root Plant Soil County State Dry Wt‘f Dy Wt K Cone; K Conc. Uptake § —— grams — % mg K kg'l Marlette Allegan MI 1.80 2.67 1.64 0.40 247 Marlette Clinton MI 1.80 2.54 l .23 0.32 147 Marlette Sanilac M] 1.70 2.56 1.07 0.30 105 Capae Allegan M] 1.89 2.46 1.23 0.30 150 Capae Clinton MI 1.81 2.56 2.17 0.53 374 Capae Sanilac MI 1.54 2.57 0.72 0.32 38 Parkhill Clinton M] 1.68 2.66 1.94 0.47 295 Parkhill Sanilac MI 1 .80 2.50 1 .09 0.28 1 12 Miami Ottawa MI 1.66 2.31 1.03 0.32 91 Miami Genesee MI 1.64 2.60 1.40 0.32 158 Miami Washtenaw MI 1.82 2.82 1.57 0.43 250 Miami Pulaski IN 1.76 2.56 2.10 0.45 331 Miami Hendricks IN 1.50 2.28 0.77 0.29 26 Miamian Preble OH 2.08 2.47 2.63 0.50 517 Miamian Clinton OH 1.63 2.59 0.87 0.31 67 Conover Ottowa MI 1 .68 2.35 1 .88 0.36 244 Conover Genesee MI 1.73 2.59 0.91 0.28 74 Conover Washtenaw MI 1 .91 2.40 1 .29 0.29 159 Crosby Pulaski 1N 1.74 2.36 1.08 0.32 108 Crosby Hendricks IN 1.57 2.24 1.04 0.28 71 Celina Preble OH 2.09 2.34 1.18 0.34 171 Crosby Preble OH 1.83 2.61 1.16 0.37 154 Crosby Clinton OH 1.64 2.42 1.24 0.35 134 Brookston Allegan MI 2.01 2.55 2.16 0.53 412 Brookston Genesee M] 1.77 2.40 1.56 0.39 215 Brookston Washtenaw M] 1.80 2.56 1.68 0.42 254 Brookston Pulaski IN 1.71 2.52 1.83 0.41 260 Brookston Hendricks IN 1.65 2.59 1.22 0.39 148 Kokomo Preble OH 1.85 2.77 1.88 0.47 324 Kokomo Clinton OH 1 .47 2.25 1.39 0.34 125 45 Table 1.12. (cont’d) Shoot Root Shoot Root Plant Soil County State Dry Wt‘f D9! Wt K Cone: K Conc. Uptake§ -——— grams —- % mg K kg" Glynwood Allegan MI 1.84 2.54 1.58 0.34 220 Morley Genesee MI 1 .59 2.42 0.81 0.30 47 Morley Washtenaw MI 2.02 2.59 2.44 0.52 474 Morley Whitley MI 1 .86 2.75 2.48 0.52 447 Morley Randolph MI 1.77 2.59 1.77 0.47 281 Glynwood Hancock OH 1.99 2.47 0.81 0.25 69 Glynwood Seneca OH 2.00 2.31 1.98 0.41 334 Blount Allegan MI 1 .96 2.46 2.21 0.46 390 Blount Clinton OH 1.71 2.45 1.61 0.39 217 Blount Monroe MI 1.74 2.62 1.80 0.43 271 Blount Whitley IN 1.64 2.32 1.34 0.35 146 Blount Randolph IN 1.60 2.58 1.57 0.49 224 Blount Hancock OH 1 .72 2.40 1.14 0.32 l 18 Blount Seneca OH 1.75 2.32 1.83 0.37 253 Pewamo Allegan MI 1.44 2.32 2.28 0.51 290 Pewamo Monroe MI 2.06 2.38 1.75 0.34 286 Pewamo Whitley IN 1.60 2.28 1.80 0.42 228 Pewamo Randolph IN 1.77 2.31 2.77 0.55 462 Pewamo Hancock OH 1 .64 2.46 1 .61 0.38 203 Pewamo Seneca OH 1 .84 2.45 2.22 0.41 353 Hoytville Monroe MI 1.78 2.33 1.51 0.38 201 Hoytville Paulding OH 1 .57 2.30 1 .97 0.36 236 Paulding Paulding OH 1.92 2.20 2.27 0.43 372 Misteguay Saginaw MI 1.85 2.64 2.20 0.44 370 'l'The dry weight of shoot and root tissue listed in the table is the average of the four replications for each soil. IThe K concentration in shoot and root tissue listed in the table is the average of the four replications for each soil. §Plant Uptake = [{(shoot dry weight 1' shoot K cone.) + (root dry weight " root Kconc.)} - {K content of seed (15.49 mg K)}] x 10 (to convert from 100 mg soil to 1 kg soil). 46 Table 1.13. Mean values of plant uptake K and the percent of uptake from exchangeable K for all soil groups. Uptake fi'om Soil Number Plant Uptake Exchangeable K Group of Soils Mean SET Mean SE mg K kg" — —- % —— Paulding: 2 371a* 1 17b 6 Pewamo 6 304a 38 47ab 7 Morley§ 7 267a 64 Slab 10 Brookstonfi 7 248a 38 57a 8 Blount 7 231a 34 59a 6 Hoytville 2 219a 1 8 27ab 7 Miami# 7 206a 66 75a 7 Parkhill 2 204a 92 50ab 5 Capae 3 187a 99 44ab 9 Marlette 3 166a 42 44ab 13 Conovero 8 139a 20 Slab 8 *Means followed by the same letter are not significantly different at the 0.05 probability level. TSE = Standard error of the mean. tPaulding includes Paulding and Misteguay soils. §Morley includes Morley and Glynwood soils. 1! Brookston includes Kokomo soils. # Miami includes Miamian soils. '1'? Conover includes Crosby and Celina soils. Table 1.14. Mean values of plant uptake K and the percent of uptake from exchangeable K for the six soil series common to the tri-state area. Soils are grouped by major soil series. Uptake fi'om Soil Number Plant Uptake Exchangeable K Group of Soils Mean Slit Mean SE — mg K kg" — —— % —— Pewamo 6 304 38 47 7 Morley: 7 267 64 51 10 Brookston§ 7 248 38 57 8 Blount 7 231 34 59 6 Miamifil 7 206 66 75 7 Conover# 8 139ns 20 51ns 8 ns = Means in the same column are not significantly different at the 0.05 probability level. tSE = Standard error of the mean. I Morley includes Morley and Glynwood soils. § Brookston includes Kokomo soils. 1 Miami includes Miamian soils. # Conover includes Crosby and Celina soils. 47 large differences in uptake between the group means, the differences were not statistically significant because of the high degree of variability within the soils of some groups. However, significant differences do exist in the percent of uptake from exchangeable and nonexchangeable K. The Paulding soil group with 17% of uptake from exchangeable K and the Miami soil group with 75% of uptake from exchangeable K were at the low and high end of the range of exchangeable K uptake values. The difference in the percent of uptake from exchangeable K between these two soil groups is what would be expected. Recall that in Table 1.2 the nonexchangeable K values for the Paulding soils were very high while the nonexchangeable K values for the Miami soils were low. The high percentage of uptake fiom exchangeable K for the Miami soils is a reflection of the fact that there is little K supplied fiom nonexchangeable forms. Therefore, the exchangeable K values are much more important in predicting the fertility status of these soils. When exchangeable K levels are reduced by cropping, there will be a relatively smaller amount of replenishment from the nonexchangeable forms. As a result, crop yields are likely to decline more rapidly on these soils. Consequently, careful fertility management would be very important on these soils if maximum productivity is to be sustained. This condition is in contrast to the situation found with the Paulding soils. In these soils, the plants obtained 3/4 of their K from the nonexchangeable forms. This demonstrates that these soils have a large reserve of nonexchangeable K that can replenish the exchangeable K. Therefore, the fertility level of these soils is not likely to decline as quickly. Exchangeable K levels, therefore, may not give an accurate picture of the fertility status of these fine textured soils. On the majority of the soils, however, exchangeable K accounted for 40 to 60% of total uptake. 48 Table 1.14 shows the K uptake data for the six series common to the tri-state area. Although there is more than a two-fold difference in uptake between the Pewamo and Conover soils, the difi‘erence is not statistically significant because of the variability within the soil groups. Also, the percent of plant uptake fiom exchangeable K does not show statistically significant differences when only the six soil groups are considered. Potassium uptake by barley plants on the Pewamo soils averaged 304 mg K kg'l soil with 47% of uptake coming fi'om exchangeable K. On the Conover soils, K uptake averaged 139 mg K kg'l soil with 51% of uptake attributed to exchangeable K. Although the mean percent uptake from exchangeable K was similar for the two groups of soils, the absolute amount of K removal from the Pewamo soils was more than twice the K uptake on the Conover soils. Table 1.15 compares the average plant uptake and percent of uptake from exchangeable K for the six common soil series when grouped together by state. In Tables 1.16 and 1.17, the plant uptake and percent of uptake fi'om exchangeable K of the individual soil series are compared across states. No statistically significant differences in plant uptake or percent of uptake from exchangeable K were found between states when all soils were considered or if the six soil series were considered individually. These similarities would further suggest that there is a good basis for a regional approach to fertility recommendations. Data from the soils of the three glacial lobes in Michigan are presented in Table 1.18. Soils from the Saginaw lobe have a lower mean K uptake value than soils in the Michigan and Huron-Erie glacial lobes, but the difference is not statistically significant. Although not statistically significant, the soils in the Saginaw lobe obtained a higher 49 Table 1.15. Mean values of plant uptake K and the percent of uptake from exchangeable K for the six soil series common to the tri-state area. Soils are grouped by state. Uptake fi'om Soil Number Plant Uptake Exchangeable K Group of Soils Mean SET Mean SE — mg K kg" — —— % —— Michigan 17 238 28 52 6 Indiana 12 228 40 63 6 Ohio 13 217ns 37 58ns 5 ns = Means in the same column are not significantly different at the 0.05 probability level. TSE = Standard error of the mean. Table 1.16. Plant uptake of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Plant Uptake Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE mg K ks" Miami: 7 166 46 179 153 292ns 225 Conover§ 8 159 49 90 19 153ns 11 Brookston1l 7 294 60 204 56 225ns 99 Morley# 7 247 124 364 83 202ns 132 Blount 7 293 51 185 39 186ns 67 Pewamo 6 288 2 345 1 17 278ns 75 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean IMiami includes Miamian soils §Conover includes Crosby and Celina soils flBrookston includes Kokomo soils #Morley includes Glynwood soils 50 Table 1.17. Exchangeable K uptake of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Uptake from Exchangeable K Soil Number Michigan Indiana Ohio Group of Soils Mean SET Mean SE Mean SE % Miami: 7 78 10 80 3 66ns 25 Conover§ 8 39 14 72 2 49ns 1 1 BrookstonTl 7 44 12 68 14 67ns 21 Morley# 7 61 19 35 22 53ns 5 Blount 7 53 8 69 20 60ns 9 Pewamo 6 30 13 53 14 59ns 0 ns = Means in the same row are not significantly different at the 0.05 probability level. SE = Standard error of the mean tMiami includes Miamian soils §Conover includes Crosby and Celina soils TlBrookston includes Kokomo soils #Morley includes Glynwood soils Table 1.18. Mean values of plant uptake K and the percent of uptake from exchangeable K for the seven soil series common to the three glacial lobes of Michigan.T Uptake from Glacial Number Plant Uptake Exchangeable K Lobe of Soils Mean SE; Mean SE —— mg K kg'l -— —— % —— Michigan 7 251 44 48 9 Huron-Erie 7 222 53 46 10 Saginaw 7 l76ns 41 61ns 6 ns = Means in the same column are not significantly different at the 0.05 probability level. TSoil Series are Marlette, Capac, Miami, Conover, Brookston, Glynwood or Morley, and Blount. ISE = Standard error of the mean. 51 percentage of their uptake fi'om exchangeable K than the soils in the Michigan and Hmon-Erie glacial lobes. Factors Affecting Potassium Uptake The amount of K in the shoot and root tissue of the barley plants is a direct function of the quantity of K taken up by the roots. Nutrient absorption is a multiple step process. Barber (1962) considered nutrient absorption by roots to occur in three phases: 1) movement of the nutrient from the soil to the root surface; 2) movement of the ion hour the exterior of the root to the interior of the root; and 3) translocation of the nutrients from the root to the shoot. Barber (1968) indicated that root interception, mass- flow, and diffusion were the three mechanisms by which plant roots obtain nutrients from the soil. Root interception is a process of ion absorption resulting from direct physical contact and displacement of ions as the root grows through the soil. Mass flow is the movement of ions through the soil solution to the plant root as a result of water uptake by the root. Diffusion occurs when the plant root removes ions from the soil solution more quickly than the stream of water moving toward the root can replenish them. In this case, a concentration gradient is established and ions will then move by diffusion toward the root where the ion concentration is lowest. Barber (1968) indicated that root interception and mass flow may supply 15% or less of the K requirement of plants in humid region soils. Diffusion then is the major mechanism of K supply to the root. Barber (1985) stated that “the proportion of K that is supplied by each mechanism will depend on the surface area and mean radius of the root system, the K absorption characteristics of the root, the rate of water absorption by the root, the rate of K diffirsion, and the levels of 52 adsorbed K and solution K within the soil.” Soil solution K concentrations range fi'om 20 to 1,000 uM with a concentration of 100 uM being representative for humid region soils (Barber et al., 1963). Although saturation extracts were not measured on the soils used in this study, it is probable that the values reported by Barber et a1. (1963) cover the range of soil solution K concentrations actually present in these soils. One of the major reasons for the high uptake values obtainable via the Neubauer technique is the high root density achieved when a large number of plants are grown on a small quantity of soil. Heming and Rowell (1985) reported that the root density of cereal crops in soil typically ranges from 0.9 to 3.5 cm cm". They suggested that at a root density of 0.9 cm cm'3 about 50% of the exchangeable K may be unavailable to the root. They stated that if the root density was at least 3.5 cm cm‘3 diffusion alone should not limit plant growth. Root density in the Neubauer studies was greater than 3.5 cm cm'3 . The primary factor limiting the supply of K to the root was probably related to the release of K fi'om nonexchangeable or slowly available forms. Actively feeding roots cause steep K concentration gradients in the rhizosphere. Jungk et a1. (1982) reported steep gradients in a radius of approximately 3 mm around the roots of young maize and rape plants. They reported that solution K concentrations were depleted to around 2 uM. Within a few days the amount of K released in the rhizosphere was twice the amount initially exchangeable. Rao and Takkar (1996) reported lower K concentrations in the rhizosphere of corn roots in smectitic soils. Kuchenbuch (1987) reported that nearly all the nonexchangeable K uptake occurs within 2 mm of the root surface while exchangeable K and nonexchangeable K is desorbed and transported further through the soil. Root hairs greatly increase the nutrient absorbing capability of 53 roots. Drew and Nye (1969) found that root hairs increased the absorption of K by up to 77% over roots gown in solution culture, which lacked root hairs. They also stated that when root demand was high, diffusion appeared to limit K uptake. Because of the high root density and abundance of root hairs present in the soils during the Neubauer studies, it is probable that the rate of K uptake by the barley plants was limited by diffusion. Asher and Ozanne (1967) reported that for 14 plant species gown in solution culture at 1 uM K all species were K deficient. Eight species achieved maximum yield at 24 uM while the other six species achieved maximum yield at 95 uM. If we assume that the K concentration of 2 uM reported by Jungk et al. (1982) in the depleted rhizosphere zone around the young maize and rape roots is representative of the concentration of K in the rhizosphere of the barley plants in the Neubauer studies, and if we assume an initial soil solution K concentration of 100 uM to be representative of the solution K concentrations in these soils (Barber, 1968) then this would represent a 50 fold decrease in K concentration in the rhizosphere of the actively feeding roots. This estimate gives us an approximate idea of the magnitude of the concentration gadient imposed by the roots. When considering the reliability of extrapolating results hour the Neubauer studies to field conditions, it is necessary to consider differences in K uptake between species. Differences in K uptake characteristics are present within species and between species. McGeorge (1946) showed that rye removed more K than barley in Neubauer experiments. Glass et al. (1981) and Glass and Perley (1980) reported differences in K uptake between barley varieties. Cultivars that had high K influx rates at the seedling stage tended to have higher tissue K content at later stages of development and higher K levels in the seed. Schenk and Barber (1980) reported that Cm, the soil solution 54 concentration below which no net influx occurs, ranged from 2.7 to 7.3 uM for three corn genotypes. Baligar and Barber (1979) found small differences in average net influx values for P and K between corn varieties gown in Indiana and Florida. Barber (197 8) reported that net flux of K was lower for soybeans than corn early in the gowing season but geater than corn later in the gowing season. Peterson and Barber (1981) discovered that root morphology affected K influx in soybean roots. Dunlop et al. (1979) indicated that rmder conditions of low K concentration in the soil solution ryegass has a geater K influx rate than white clover. Plant age also affects K uptake. Mengel and Barber (1974) found that nutrient flux in field corn was higher in plants sampled at 20 days than in plants sampled at 70 days. These reports emphasize the fact that the results obtained with barley may not be strictly comparable to results obtained with other species. However, because all soils where treated the same, the Neubauer experiments can provide an initial reference point for making comparisons between the soils and these experiments have set the stage for further studies in the field with other crops. In addition to biological factors that may have influenced the results of the Neubauer studies, it is important to consider soil environmental factors that may afl‘ect K uptake (Barber, 1995). Soil moisture affects the rate of K uptake. Danielson and Russell (1957) reported a linear relationship between Rb"5 uptake and soil moisture. Similar results were reported by Place and Barber (1964). Soil aeration is also a factor in ion uptake by roots. If oxygen concentrations fall below 10% in soils, ion uptake will be reduced (Danielson and Russell, 1957). Oxygen levels do not generally approach 10% in soils unless they become saturated. Soil temperature can influence ion uptake by roots. 55 Ching and Barber ( 1979) found that K influx in corn roots was doubled when soil temperatures were increased from 15 and 29° C. Soil pH may also afl‘ect K availability in soils although it is not of primary importance if pH is maintained above 5 to 5.5 so that soluble and exchangeable Al become negligible (Nemeth and Grimme, 1972). For the soils used in this study, pH values ranged from 5.6 to 7.9. Therefore, it is unlikely that K availability was influenced by soluble aluminum. For experimental results to be reliable and meaningful, it is essential to minimize the potential differences in biological and environmental factors affecting nutrient uptake as much as possible. Correlations between Plant Uptake and Extractable Potassium The third objective of this study was to determine whether NH40Ac or NaBPhe would be a better index of plant available K. The release of nonexchangeable K is a difi'usion controlled process (Cox and Joem, 1997; Mortland and Ellis, 1959; Reed and Scott, 1962; Martin and Sparks, 1985). The rate of this release will depend on the concentration of K in the soil solution, the quantity of exchangeable K initially present, the rate of K removal by plant uptake and leaching, and the types of minerals present in the soil. Since NaBPh4 extracts both exchangeable K and nonexchangeable K from the interlayers of the 2:1 clay minerals, it may be a better predictor of plant available K in soils containing large amounts of 2:1 clay minerals. Wentworth and Rossi (1972) repomd that K uptake by barley and NaBPh4 extraction from several clay minerals was in the following order: vermiculite > illite > biotite > phlogopite > muscovite. Schulte and Corey (1965) found that NaBPh. gave a better estimate of K uptake by oats and ryegass during cropping studies in the geenhouse than did NI-LOAc for four Wisconsin soils and one soil from 56 Iowa. Husin et al. (1986) found that for 30 Louisiana soils there was a higher correlation between NI-LOAc and K uptake by sorghurn-sudan gass for the first two harvests in their study, but that NaBPhr gave a better correlation for the third and fourth harvests when the exchangeable K had been depleted to a low level. Jackson (1985) reported that NaBPhr generally gave higher correlations with dry matter yield, total K uptake, and percent K in ryegass than did barium acetate, nitric acid, and ammonium nitrate for 13 soils of differing K levels when the soils were cropped to exhaustion. The relationship between initial NI-LOAc-K and plant uptake for all 54 soils is shown in Figure 1.2. The coefficient of simple determination is r2 = 0.730. Figure 1.3 shows that the change in NH40Ac-K was not as well correlated with plant uptake, r2=0.500, as was the initial exchangeable K level. Considerably more scatter can be seen in Figure 1.4 which shows the relationship between initial NaBPhr-K and plant uptake, r2=0.536, for all soils. This higher variability would suggest that initial NaBPhr-K does not give as good a correlation with plant uptake as initial NILOAc-K. Interestingly, though, the relationship between the change in NaBPhr-K has a better correlation with plant uptake, r2=0.722, than does the change in exchangeable K (Figure 1.5). When these trends were discovered, it prompted further investigation to see if these same findings would hold true for the two major soil association goups in the tri- state area. The correlations for the Miami, Conover, Brookston soil association goup are shown in Figures 1.6 to 1.9. The correlations for the Morley, Blount, Pewamo soil association goup are shown in Figures 1.10 to 1.13. The correlation trends found for all soils also holds true for soils of the two major soil association goups. A summary of the coefficients of determination is presented in Table 1.19. 57 m 8 as or 8 8 Plant Uptake K (mg K kg") 8 O o§§ 1.16x - 2.47 0.730 2 O 100 I T ' 200 300 Initial NH4OAc-K (mg K kg") 500 Figure 1.2. Relationship between initial NHaoAc-K and plant uptake K for all soils. 600 - 500 . I. a To : 0 x 0 E400 - s E O 0 300 - 3 .K E 0- . 8 y= 1.14x+ 100 2 200 . r2= 0.500 g . n =54 “- 100 - I o '!. ‘ o 0 r r r 0 100 200 300 Delta NH40Ac-K (mg K kg") 400 Figure 1.3. Relationship between delta NH40Ac-K and plant uptake K for all soils. 58 600 - . l 7D X K a, 400 - E. X o 300 - x .0 3 200 - ‘5‘, y = 0.207x + 47.5 -- 2 a. 100 . r =0.536 . b. ' n = 54 0' I o 1 l I I 0 500 1000 1500 2000 Initial NaBPhg-K (mg K kg") 2500 Figure 1.4. Relationship between initial five-minute NaBPha-K and plant uptake K for all soils. 600 500 - ‘l 400 - 300 - 200 - Plant Uptake K (mg K kg 100- o 200 400 Delta NaBPhg-K (mg K kg") 800 Figure 1.5. Relationship between delta five-minute NaBPhr-K and plant uptake K for all soils. 59 Plant Uptake K (mg K kg") N w a 8 § 8 i U 200 300 400 Initial NH40Ac-K (mg K kg") Figure 1.6. Relationship between initial NH40Ac-K and plant uptake K for the Miami, Conover, Brookston soil association group. 0 § 3 Plant Uptake K (mg K kg") T I 0 50 100 150 200 250 300 Delta NH,DAc-K (mg K kg") Figure 1.7. Relationship between delta NH40Ac-K and plant uptake K for the Miami, Conover, Brookston soil association group. 600 .;- 500 - ' .9 E 400 « x 300 . 3 I a g 200 - .é y = 0.220x + 36.6 g r 2 = 0.545 100 - I I n = 22 O o I U I 0 500 1000 1500 2000 Initial NaBPhg-K (mg K kg") Figure 1.8. Relationship between initial five-minute NaBPha-K and plant uptake K for the Miami, Conover, Brookston soil association group. 600 Plant Uptake K (mg K kg") -I N U '5 U! o 8 8 8 8 8 I . y = 0.753x + 8.17 . . r2= 0.755 , n =22 ': l O 0 200 400 600 800 Delta NaBPlu-K (mg K kg") Figure 1.9. Relationship between delta five-minute NaBPhr-K and plant uptake K for the Miami, Conover, Brookston soil association group. 61 600 is x . a 400 - E. x 0 300 - '5 O. 2 20° ‘ y = 0.960x + 53.1 g , r 2 = 0.602 a. 100 . o n = 20 l o I I I I 0 100 200 300 400 500 Initial NH,DAe-K (mg K kg") Figure 1.10. Relationship between initial NH40Ac-K and plant uptake K for the Morley, Blount, Pewamo soil association group. 600 ,r‘ 500 - g '3’ E, 400 - :7 a, 300 - .K .0 5' 200 - y =1.01x +131 5 , r2 = 0.495 ! O T I I 0 100 200 300 400 Delta NH,DAo-K (mg K kg") Figure 1.11. Relationship between delta NH40Ac-K and plant uptake K for the Morley, Blount, Pewamo soil association group. 62 600 '5. 500 '- . . : Ia . . 8- s x . a 400 .5. x 0 300 - x 9. :‘11 200 - g y = 0.272x + 4.50 a r 2 = 0.553 100 - ' n = 20 2 o I I I O 500 1000 1500 2000 Initial NaBPm-K (mg K kg") Figure 1.12. Relationship between initial five-minute NaBPha-K and plant Uptake K for the Morley, Blount, Pewamo soil association group. 600 7" 500 . g 3’ 8 E, 400 - V I x 0 300 - x. :2 a. 3 20° ‘ y = 0.697x + 47.1 .3 , r 2 = 0.695 O. 100 a o n = 20 : I 0 I I’ I I I I 0 100 200 300 400 500 600 700 Delta NaBPhg-K (mg K kg") Figure 1.13. Relationship between delta five-minute NaBPm-K and plant uptake K for the Morley. Blount, Pewamo soil association group. 63 Table 1.19. Summary of coefficients of determination from regession analysis of NILOAc-K and NaBPha-K versus plant uptake K. Initial Delta Initial Delta NHaOAc-K NHaOAc-K NaBPha-K NaBPhs-K l,2 All soils 0.730*** 0,500'" '1' 0536*“! Q7220": Miami catena 0.791 *" 0614*“ 0545*“ 0755*" Morley catena 0602*" 0495"" 0553‘” 0695*” "* Significant at the 0.001 probability level. In all of these comparisons, the initial NHaOAc-K values give a higher correlation than the initial NaBPha-K values, but the delta NILOAc-K values are lower than the delta NaBPha-K values. These findings are in contrast to the results reported by other investigators (Schulte and Corey, 1985; Jackson, 1985), but a thoughtful consideration of the problem and an investigation of reports from the 1950’s may provide an explanation for the results. Conflicting reports on the suitability of exchangeable K as a predictive measure of K availability appeared in several papers from the late 1940’s to the early 1960’s. A series of field and geenhouse studies conducted in the mid to late 1950’s with alfalfa (Hanway et al., 1961), millet (Barber et al., 1961) and corn (Hanway et al., 1962) in the North Central Region generally found exchangeable K, measured with field moist samples, was a suitable index of plant available K. Breland et al. (1950) in a study of 23 Indiana soils did not find a sigrificant relationship between the initial level of exchangeable K and K uptake by ladino clover during a geenhouse cropping experiment. Pratt (1951), however, stated that exchangeable K was the single best measure of K availability for Iowa soils compared to K extractable with boiling 1M I'INO3 and Dowex 54 50 exchange resin. Schmitz and Pratt (1953) reported that in 18 soils from Ohio, 1M I-INO3 was a better index than exchangeable K in predicting crop response to fertilization. Rouse and Bertramson (1949) with the same 23 Indiana soils studied by Breland et al. (1950) reported that exchangeable K was not related to the quantity of nonexchangeable K released to I-INO3 - a term they called the “potassium supplying power” of the soil. In 1954, Pratt and Morse published the results of a study on 46 soil series from Ohio in which they reported that the correlation between exchangeable K and K release (HNO3-K - NHaOAc-K) was not high. They found the K release value to be more characteristic of soil type and soil area than exchangeable K. In this study, they gouped the soils of Ohio according to the major soil regions in the state. The soils were ranked according to their ability to release nonexchangeable K in the following order from least to geatest: l) sandy soils of the lakebed; 2) soils on Illinoian till; 3) soils formed in low-limeWisconsin till; 4) soils formed in high-lime Wisconsin till; 5) fine textured soils of the lakebed. In 1958, McLean and Simon reported the results of their study on K release and fixation in 13 representative soils fiom Ohio. They reported a highly sigrificant correlation between exchangeable K and the response of alfalfa to applied K The best explanation for these conflicting results perhaps can be found in the paper by Schmitz and Pratt (1953). They speculated that the reliability of exchangeable K as an index to K uptake by plants would vary with the release rates of K from nonexchaneable forms. They considered this release to occur at three rates. At the lowest rate, the release rate of K from nonexchangeable forms was considered to be too small to be of importance in supplying K to plants. At the highest rate, the K release rate was assumed to be high enough to maintain high levels of exchangeable K that would be 65 highly correlated with the rate of K release from nonexchangeable forms. At an intermediate release rate, the rate of release would not be fast enough to maintain high levels of exchangeable K but yet would be important in supplying K to the plants. At intermediate release rates, it was assumed that there would be a low correlation between levels of exchangeable K and rates of K release from nonexchangeable forms. Mitosis and Rowell (1987) reported that the critical K concentration in soil solution ranged fi'om 33 to 85 pM before release of nonexchangeable K occurred. Tabatabai and Hanway (1969) found that in Iowa soils plant uptake was highly correlated with the minimum level of exchangeable K. This would correspond to the low release rate suggested by Schmitz and Pratt (1953). Richards et al. (1988) reported large differences in nonexchangeable K uptake from soils that had similar exchangeable K levels. This situation would suggest intermediate rates of nonexchangeable K release. The results of Rao and Khera (1994) would support the idea of high release rates. They found higher minimal K levels in soils initially high in K. They reported a high correlation between K replenishment rate and minimum exchangeable K level. Replenishment rates varied from 0.25 to 0.67 mg K kg" soil after reaching minimum K levels. Soils with higher illite content had a higher replenishment rate. Since the late 1950’s, exchangeable K has been considered to be a satisfactory index for predicting plant availability of K. This can perhaps be explained by considering the fact that during the 1940’s and 1950’s crop yields were much lower than yields attainable today. As a result, these soils were not fertilized as heavily 40 to 50 years ago. If the exchangeable K levels were generally less than present levels, then the majority of these soils probably had intermediate rates of K replenishment and would not 66 have given consistently good correlations between exchangeable K and K uptake. As agonomic practices changed, fertility levels increased and these soils gadually shifted from intermediate to high rates of K release and are now capable of maintaining a higher rate of K release from nonexchangeable forms. The result of this interaction has been a higher correlation between exchangeable K and K uptake. The higher correlations with initial NH40Ac-K and delta NaBPhr-K over delta NILOAc-K and initial NaBPhr-K can possibly be explained by the fact that the change in exchangeable K does not provide a good index of K release from nonexchangeable forms while the change in NaBPhr-K more accurately reflects the increased utilization of nonexchangeable K. CONCLUSIONS The results of these investigations revealed large differences in CEC, total K, and nonexchangeable K release between soil series. The differences between major soil series were consistent regardless of geogaphic location. The fact that the major soil series in the three states exhibit similar K chemistry is evidence that a regional approach to fertilizer recommendations in the tri-state area is merited. The results of this study also suggest that K recommendations should be made based on major soil association goups within the three states. Even though NHaOAc proved to be a more reliable index of plant available K, more work needs to be done with NaBPhr, The ability of this reagent to measure plant available K might be improved if a lower concentration of NaBPha was used in the extracting solution or if samples were extracted for a shorter time. Long-term field experiments also need to be established on the major soil series at different sites 67 throughout Michigan, Indiana, and Ohio to more thoroughly evaluate the nature of K release fiom these soils and K uptake under different cropping systems. 68 REFERENCES Addiscott, T.M., and A.E. Johnston. 1975. Potassium in soils under different cropping systems. 3. Nonexchangeable potassium in soils from long-term experiments at Rothamsted and Woburn. J Agic. Sci. Camb. 84:513-524. Analytical Software. 1996. Statistix for Windows. Analytical Software, Tallahassee, FL. Asher, C .J ., and PG. Ozanne. 1967. Growth and potassium content of plants in solution cultures maintained at constant potassium concentrations. Soil Sci. 103:155-161. Baligar, V.C., and S.A. Barber. 1979. Genotypic differences of corn for ion uptake. Agon. J. 71:870-873. Barber, S.A. 1962. A diffusion and mass-flow concept of soil nutrient availability. Soil Sci. 93 :39-49. Barber, S.A. 1968. Mechansirn of potassium absorption by plants. p. 293-310. In V.J. Kilmer (ed.) The role of potassium in agicultrure. ASA, CSSA, and SSSA, Madison, WI. Barber, S.A. 1978. Growth and nutrient uptake of soybean roots under field conditions. Agon. J. 70:457-461. Barber, S.A. 1985. Potassium availability at the soil-root interface and factors influencing potassium uptake. p. 309-335. In R.D. Munson (ed.) Potassium in agiculture. ASA, CSSA, and SSSA, Madison, WI. Barber, S.A. 1995. Soil nutrient bioavailability: A mechanistic approach. 2nd ed. John Wiley and Sons, Inc., New York, NY. Barber, S.A., R.H. Bray, A.C. Caldwell, R.L. Fox, M. Freid, J .J . Hanway, D. Hovland, J. W. Ketcheson, W.M. Laughton, K. Lawton, R.C. Lipps, R.A. Olsen, J.T. Pesek, K. Pretty, M. Reed, F.W. Smith, and EM. Stickney. 1961. North central regional potassium studies: 11. Greenhouse experiments with millet. Indiana Ag. Exp. Sta. Res. Bull. 717. 69 Barber, S.A., J .M. Walker, and EH. Vasey. 1963. Mechanisms for the movement of plant nutrients from the soil and fertilizer to the plant root. J. Ag. Food Chem. 1 1:204-207. Binnie, R.R., and S.A. Barber. 1964. Contrasting release characteristics of potassium in alluvial and associated upland soils of Indiana. Soil Sci. Soc. Am. Proc. 28: 387-390. Breland, H.L., B.R. Bertramson, and J .W. Borland. 1950. Potassium-supplying power of several Indiana soils. Soil Sci. 69:237-247. Brown, J .R., and D. Wamcke. 1988. Recommended cation tests and measures of cation exchange capacity. p. 15-16. In W.C. Dahnke (ed.) Recommended chemical soil test procedures for the North Central Region. North Central Regional Publication No. 221 (Revised). Ching, PC, and S.A. Barber. 1979. Evaluation of temperature effects on K uptake by com. Agon. J. 71:1040-1044. Cox, A.E., and BC. Joern. 1997. Release kinetics of nonexchangeable potassium in soils using sodium tetraphenylboron. Soil Sci. 162:588-598. Cox, A.E., B.C. Joem, and CB. Roth. 1996. Nonexchangeable ammonium and potassium determination in soils with a modified sodium tetraphenylboron method. Soil Sci. Soc. Am. J. 60:114-120. Danielson, RE, and MB. Russell. 1957. Ion absorption by com roots as influenced by moisture and aeration. Soil Sci. Soc. Am. Proc. 21 :3-6. Drew, M.C., and RH. Nye. 1969. The supply of nutrient ions by diffusion to plant roots in soil. II. The effect of root hairs on the uptake of potassium by roots of rye gass (Lalium multiflorum). Plant Soil. 31 :407-424. Dunlop, J ., A.D.M. Glass, and 3D. Tomkins. 1979. The regulation of potassium uptake by ryegass and white clover roots in relation to their competition for potassium. New Phytol. 83:365-370. Eckert, DJ. 1988. Recommended pH and lime requirement tests. p. 6-8. In W.C. Dahnke (ed.) Recommended chemical soil test procedures for the North Central Region. North Central Regional Publication No. 221 (Revised). Glass, A.D.M., and J .E. Perley. 1980. Varietal differences in potassium uptake by barley. Plant Physiol. 65:160-164. 70 Glass, A.D.M., M.Y. Siddiqi, and K.I. Giles. 1981. Correlations between potassium uptake and hydrogen efflux in barley varities. A potential screening method for the isolation of nutrient efficient lines. Plant Physiol. 68 :457-459. Hanway, J.J., S.A. Barber, R.H. Bray, A.C. Caldwell, L.E. Engelbert, R.L. Fox, M. Fried, D. Hovland, J .W. Ketcheson, W.M. Laughlin, K. Lawton, R.C. Lipps, R.A. Olsen, J.T. Pesek, K. Pretty, F.W. Smith, and EM. Stickney. 1961. North central regional potassium studies. 1. Field studies with alfalfa. Iowa Ag. Home Econ. Exp. Sta. Res. Bull. 494. Hanway, J .J ., S.A. Barber, R.H. Bray, A.C. Caldwell, M. Freid, L.T. Kurtz, K. Lawton, J.T. Pesek, K. Pretty, M. Reed, and F.W. Smith. 1962. North central regional potassium studies. 111. Field studies with corn. Iowa Ag. Home Econ. Exp. Sta. Res. Bull. 503. Heming, SD, and D.L. Rowell. 1985. Soil structure and potassium supply. H. The effect of ped size and root density on the supply to plants of exchangeable and nonexchangeable potassium from a chalky boulder clay soil. 1. Soil. Sci. 36: 61-69. Husin, A.B., A.T.B. Bachik, and A.G. Caldwell. 1986. Plant response to potassium related to soil tests for potassium fractions in 30 soils from Louisiana. Plant Soil. 96:57-67. Jackson, BL. 1985. A modified sodium tetraphenylboron method for the routine determination of reserve-potassium status of soil. N.Z. J. Exp. Agic. 13:253-262. Jungk, A., N. Claassen, and R. Kuchenbuch. 1982. Potassium depletion of the soil-root interface in relation to soil parameters and root properties. In A. Scaife (ed) Plant Nutrition 82; Proc. 9th Int. Plant. Nutr. Colloq. 1:250-255. Knudsen, D., G.A. Peterson, and P.F. Pratt. 1982. Lithium, sodium, and potassium. p. 225-246. In A.L. Page (ed.) Methods of soil analysis. Part 2. 2nd ed. Agon. Monog. 9. ASA, and SSSA, Madison, WI. Kuchenbuch, RD. 1987. Potassium dynamics in the rhizosphere and potassium availability. Proc. Coll. Int. Pot. Inst. 20:215-234. Martin, H.W., and D.L. Sparks 1985. On the behavior of nonexchangeable potassium in soils. Comm. Soil Sci. Plant Anal. 16(2):133-162. McGeorge, W.T. 1946. Modified Neubauer method for soil cultures. Soil Sci. 62:61-70. 71 McLean, E.O., and RH. Simon. 1958. Potassium release and fixation in Ohio soils as measured by cropping and chemical extraction. Ohio Ag. Exp. Sta. Res. Bull. 824. Memon, Y.M., I.F. Fergus, J .D. Hughes, and D.W. Page. 1988. Utilization of nonexchangeable soil potassium in relation to soil type, plant species, and stage of gowth. Aust. J. Soil Res. 26:489-496. Mengel, DB, and S.A. Barber. 1974. Rate of nutrient uptake per unit of corn root under field conditions. Agon. J. 66:399-402. Mengel, K., and Rahmatullah. 1994. Exploitation of potassium by various crop species from primary minerals in soils rich in micas. Biol. Fert. Soils. 17:75-79. Mitsios, I.K., and D.L. Rowell. 1987. Plant uptake of exchangeable and nonexchangeable potassium. 11. Influence of soil type on uptake by onion roots. J. Soil Sci. 38: 65-70. Mortland, M.M., and B.G. Ellis. 1959. Release of fixed potssium as a diffusion controlled process. Soil Sci. Soc. Am. Proc. 23:363-364. Mortland, M.M., K. Lawton, and G. Uehara. 1957. Fixation and release of potassium by some clay minerals. Soil Sci. Soc. Am. Proc. 21:381-384. Mutscher, H., and T.V. Tu. 1988. Studies on the release of interlayer potassium of an Illitic dystric cambisol and its effect on the potassium status of the soil. 11. Rates of release of interlayer potassium and their relation to the intensity factor. Beitr. Trop. Landwirtsch. Veterinarmed. 26:107-115. Nemeth, K., and H. Grimme. 1972. Effect of soil pH on the relationship between potassium concentration in the saturation extract and potassium saturation of soils. Soil Sci. 114:349-354. Niebes, J .F ., J .E. Dufey, B. Jaillard, and P. Hinsinger. 1993. Release of nonexchangeable potassium from different size fractions of two highly potassium fertilized soils in the rhizosphere of rape (Brassica napus cv Drakkar). Plant Soil. 155/156:403-406. Olsen, S.R., and B.T. Shaw. 1943. Chemical, Mitscherlich, and Neubauer methods for determining available potassium in relation to crop response to potash fertilization. J. Am. Soc. Agon. 35:3-9. Peterson, W.R., and S.A. Barber. 1981. Soybean root morphology and potassium uptake. Agon. J. 73:316-319. 72 Place, G.A., and S.A. Barber. 1964. The effect of soil moisture and rubidium concen- tration on diffusion and uptake of rubidium-86. Soil Sci. Soc. Am. Proc. 28:239- 243. Pratt, P.F. 1951. Potassium removal fi'om Iowa soils by geenhouse and laboratory procedrn'es. Soil Sci. 72:107-117. Pratt, P.F., and H.H. Morse. 1954. Potassium release from exchangeable and nonexchangeable forms in Ohio soil. Ohio Ag. Exp. Sta. Res. Bull. 747. Rao, C.S., and M.S. Khera. 1994. Potassium replenishment capacity of illitic soils at their minimal exchangeable K in relation to clay mineralogy. Z. Pflanzenerntlhr. Bodenk. 157:467-470. Rao, C.S., and P. N. Takkar. 1996. Potassium status in maize rhizosphere of smectitic soils. 2. Plfanzenernahr. Bodenk. 160:103-106. Reed, M.G., and AD. Scott. 1962. Kinetics of potassium release from biotite and muscovite in sodium tetraphenylboron solutions. Soil Sci. Soc. Am. Proc. 26 :437-440. Reed, M.G., and AD. Scott. 1966. Chemical extraction of potassium fi-om soils and micaceous minerals with solutions containing sodium tetraphenylboron: IV. Muscovite. Soil. Sci. Soc. Am. Proc. 30:185-188. Richards, J.E., T.E. Bates, and SC. Sheppard. 1988. Studies on the potassium-supplying capacities of southern Ontario soils. 1. Field and geenhouse experiments. Can. J. Soil Sci. 68:183-197. _ Ross, G.J., R.A. Cline, and D.S. Gamble. 1989. Potassium exchange and fixation in some southern Ontario soils. Can. J. Soil. Sci. 69:649-661. Rouse, RD, and ER. Bertramson. 1949. Potassium availablility in several Indiana soils: Its nature and methods of evaluation. Soil Sci. Soc. Am. Proc. 14:113-123. Schenk, M.K., and S.A. Barber. 1980. Potassium and phosphorus uptake by com genotypes gown in the field as influenced by root characteristics. Plant Soil 54:65-76. Schmitz, G.W., and P.F. Pratt. 1953. Exchangeable and nonexchangeable potassium as indexes to yield increases and potassium absorption by com in the geenhouse. Soil Sci. 75:343-353. Schul Schu Scot Sco Ste Scc Sin Sim 51111: Still Span label: 73 Schulte, BE. 1988. Recommended soil organic matter tests. p. 29-32. In W.C. Dahnke (ed.) Recommeded chemical soil test procedures for the North Central Region. North Central Regional Publication No. 221. (Revised). Schulte, BE, and RB. Corey. 1965. Extraction of potassium from soils with sodium tetraphenylboron. Soil Sci. Soc. Am Proc. 29:33-35. Scott, A.D., R.R. Hunziker, and J .J . Hanway. 1960. Chemical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenylboron. 1. Preliminary experiments. Soil Sci. Soc. Am. Proc. 24: 191-194. Scott, A.D., and MG. Reed. 1962a. Chemical extraction of potassium from soils and micaceous mineral with solutions containg sodium tetraphenylboron: H. Biotite. Soil Sci. Soc. Am. Proc. 26:41-45. Scott, A.D., and MG. Reed. 1962b. Chemical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenyl boron: III. Illite. Soil Sci. Soc. Am. Proc. 26:45-48. Scott, A.D., and LP. Welch. 1961. Release of nonexchangeable soil potassium during short periods of cropping and sodium tetraphenylboron extraction. Soil Sci. Soc. Am. Proc. 25:128-132. Singh, D., and AB. Ghosh. 1984. Removal of nonexchangeable potassium through intensive cropping in geenhouse study with alluvial soils. J. Ind. Soc. Soil Sci. 32:303-308. Smack, N.E., L.P. Wilding, and N. Holowaychuk. 1968. Genesis of argillic horizons in Celina and Morley soils of western Ohio. Soil Soc. Am. Proc. 32:550-556. Smith, S.J., and AD. Scott. 1966. Extractable potassiumin gundite illite: 1. Method of extraction. Soil Sci. 102:115-122. ' Smith, S.J., L.J. Clark, and AD. Scott. 1968. Exchangeability of potassium in soils. Trans. Int. Cong. Soil Sci. 9(2):661-669. Sparks, D.L. 1987. Potassium dynamics in soils. Adv. Soil Sci. 6:1-63. Tabatabai, M.A., and J .J . Hanway. 1969. Potassium supplying power of Iowa soils at their “minimal” levels of exchangeable potassium. Sol Sci. Soc. Am. Proc. 33:105-109. 74 Tarafdar, P.K., and AK. Mukhopadhyay. 1989. Availability of potassium to crops at minimal exchangeable potassium level in soil by exhaustive cropping. Trop. Agric. 66:87-90. Tening, A.S., J .A.I. Omueti, G. Tarawali, and MA. Mohamed-Saleem. 1995. Potassium status of some selected soils under different land-use systems in the subhumid zone of Nigeria. Comm. Soil Sci. Plant Anal. 26:657-672. Tisdale, S.L., W.L. Nelson, and JD. Beaton. 1985. Soil fertility and fertilizers. 4th ed. MacMillan Publishing Company. New York, NY. Thorton, S.F. 1931. Experiences with the Neubauer method for determining mineral nutrient deficiencies in soils. J. Am. Soc. Agon. 23:195-208. Thorton, S.F. 1935. The available phosphorus and potassium contents of surface soils and subsoils as shown by the Neubauer method and by chemical tests. J. Am. Soc. Agon. 27:46-51. Vitosh, M.L., J .W. Johnson, and DB. Mengel. 1995. Tri-state fertilizer recommendations for corn, soybeans, wheat and alfalfa. Michigan State University Ext. Bull. E-2567. Wamcke, DD. 1997. Manual of laboratory procedures. Michigan State University Soil and Plant Nutrient Laboratory. Michigan State University. East Lansing, MI. Wamcke, DD, and S.A. Barber. 1974. Root development and nutrient uptake by com gown in solution culture. Agon. J. 66:514-516. Wentworth, S.A., and N. Rossi. 1972. Release of potassium from layer silicates by plant gowth and by NaTPB extraction. Soil Sci. 113:410-416. re] dc “P 6X Cl 0.: £01 W11 fro: CHAPTER 2 THE RELATIVE IMPORTANCE OF CLAY CONTENT, CLAY MINERALOGY, AND CATION EXCHANGE CAPACITY TO THE POTASSIUM CHEMISTRY OF SOILS IN MICHIGAN, INDIANA, AND OHIO ABSTRACT Clay content and clay composition are important factors controlling potassium (K) availability in soils. The purpose of this study was to determine the relationship between clay content, clay mineralogy, and soil cation exchange capacity (CBC) and their relative importance with regard to the K chemistry of the whole soil. Clay contents were determined by the hydrometer and pipet methods. Four soils having a wide range in total K content were selected for clay mineralogical analysis by x-ray diffraction. Clay contents were highly correlated by the hydrometer and pipet methods, r2 = 0.993. Plant uptake, ammonium acetate (NILOAc), and sodium tetraphenylboron (NaBPha) extractable K levels were better correlated with clay content than with soil CEC. Soil CEC was more highly correlated with clay content, r2 = 0.767, than organic matter, 1'2 = 0.469, but was more sensitive to changes in organic matter than to differences in clay content. Soils with high clay content tended to have larger amounts of total K than soils with low clay content. The coarse clay (2-0.2um) and fine clay (<0.2um) fractions were dominated by 2:1 clays. Illite contents ranged from 10 to 40% in the coarse clay and from 5 to 30% in the fine clay. These results suggest that K recommendations utilizing a 75 76 K sufficiency level based on soil clay content and dominate clay mineralogy may be more useful in predicting crop response to applied K than a sufficiency index based on soil CEC. INTRODUCTION The fundamental importance of soil mineralogy and its implications for soil management can hardly be over emphasized. Mineralogy profoundly affects many physical and chemical soil properties, including: cation and anion exchange reactions, soil structure, formation, and stability, water retention and movement, and suitability for construction and building purposes (Brady and Weil, p. 241, 1996). Knowledge of clay mineralogy is especially important in fertility management because the native fertility of soils is determined by the mineralogy of the soil parent material. Clays are the sites of countless physical and chemical reactions in soils. The relative abundance and types of clay minerals and their rates of weathering influence the availability of essential mineral nutrients to plants (Bertsch and Thomas, 1985). The soil is a dynamic system in which minerals undergo continuous transformations through pedogenic, biological, and chemical weathering. These weathering reactions are of geat importance for the release of nutrients to plants, particularly the reactions that result in K release from soil minerals Studies by Mortland et al. (1957) and Lui et al. (1997) have demonstrated the ability of plants to convert biotite or intergade vermicultite:illite into vermiculite. Other researchers have reported conversion of illite to smectite or interstratified illitezsmectite under intensive cropping conditions (Moberg and Nielson, 1983; Nielson and Moberg, 77 1984; Tributh et al. 1987). Moberg and Nielson (1983) found that the total K content of Danish soils was reduced four to seven percent by intensive cropping. The illite content of the soils was decreased and the smectite content was increased. They found that the magnitude of change occurring in potted soil after five years was equal in magritude to mineralogical changes occurring in the field after 60 years. Hydroxy-interlayered vermiculite may also form as a result of biological weathering of trioctahedral mica. Hinsinger et al. (1993) attributed this phenomenon to If excretion by plant roots which causes dissolution of the crystal lattice of phlogopite and release and precipitation of octahedral aluminum in the interlayer space of the mineral. The transformation of mica to vermiculite may be reversed if large amounts of K are added to the soil. Ross et al. (1985) observed conversion of vermiculite to mica after six years of heavy applications of liquid dairy manure. This transformation was attributed to fixation of K and NH: by the soil clays. Clay minerals are also susceptible to weathering with chemical reagents. Alteration of mineral structure often occurs with NaBPhr and HNO3 which have been used for extraction of nonexchangeable K. Sodium tetraphenylboron weathers mica by removing K from the interlayers of the clay. This results in an expansion of the 10A spacing of illite (Cox and Joern, 1997, Gil-Sotres and Rubio, 1992). Nitric acid is a more drastic treatment, dissolving vermiculite and clay sized chlorite (Conyers et al., 1969; Al- Kanani et al., 1984). Knowledge of the mineralogy of soil clays is important to better understanding the processes of K fixation and release in soils. The objectives of this study were: 1) to determine the correlation between the hydrometer and pipet methods of particle size 78 analysis for determination of clay content; 2) to compare the clay content of common soil series in the tri-state area; 3) to determine the relationship between clay content and CEC to plant uptake and NI-LOAc and NaBPha extractable K levels; 4) to determine the variability in clay mineralogy of soils of widely differing total K content; and 5) to determine the nature of the relationship between the clay content and mineralogy of these soils to the K chemistry of the whole soil. These data should help to further improve K recommendations in the tri-state area. MATERIALS AND METHODS Particle Size Analysis Particle size analysis was determined by both the pipet and hydrometer methods. The particle size distribution for all soils is listed in appendix B. Eleven soils, listed in Table 2.1, were selected for detailed particle size characterization. These soils were sent to the Soil Characterization Laboratory at the Ohio State University for complete characterization of the sand, silt, and clay fractions by sieving and the pipet method. Particle size analysis was also determined for all 54 soils by a modified Bouycous hydrometer method using an ASTM 152 H-type hydrometer. A procedure slightly modified from that of Gee and Bauder (1986) was used. A 40 gam sample of each soil, dried at 35° C, was placed into a 1 L beaker. Approximately 100 mL of deionized water was added to the samples. Ten mL of hydrogen peroxide (11202) were then added to initiate oxidation of organic matter. After approximately 10 minutes, when the most vigorous reaction had subsided, samples were placed on a hot plate at 70° C. More H202 was added in 10 mL increments until the strong frothing from oxidation of organic matter 79 had subsided to a moderate bubbling due to the thermal decomposition of H202. After decomposition of H202, beakers were removed from the hot plate and allowed to cool. After cooling, samples were quantitatively washed into 500 mL plastic bottles. One htmdred mL of 5% sodium hexametaphosphate (SI-IMP) solution were added to the samples. The bottles were filled 2/3 full with deionized water and shaken overnight on a reciprocating shaker at 200 oscillations per minute. Alter shaking, samples were quantitatively washed into clean 1 L sedimentation cylinders and brought to volume with deionized water. A cylinder containing 100 mL SHMP solution and 900 mL deionized water was used as a blank to correct the hydrometer readings for the backgound concentration of SI-IMP in the soil suspension. Samples were allowed to equilibrate at room temperature. Before sedimentation, samples were thoroughly mixed with a metal plunger for one minute. Timing was initiated after removal of the plunger from the soil suspension. Readings for clay were taken after 24 hours of settling. After the clay readings were taken, samples were washed through a 270 (<53um) mesh sieve to collect the sand. Sands ’were washed free of adhering silt and clay, then dried at 105° C, sieved, and subsequently weighed to determine the percent sand in the samples. A 10 gam sample of soil was also oven dried at 105° C to determine the oven dry weight of the soil. Percent sand was calculated by dividing the weight of the sand by the weight of oven dry soil. Percent clay was determined by subtracting the blank hydrometer reading from the clay suspension hydrometer reading of each soil and dividing the corrected reading by the mass of oven dry soil. The percent silt was calculated by difference as: %Silt = 100 — (%Sand + %Clay) [1] 80 Mineralogical Analysis Four soils (Table 2.5) were selected for semi-quantitative mineralogical analysis based on their range of total K contents. Thirty to 50 garns of each soil were weighed into each of four 250 mL centrifirge bottles. One hundred mL of 1 M NaOAc, pH = 5, were added to the samples to remove carbonates and soluble salts (Kunze and Dixon, 1986). Centrifuge bottles were placed in a water bath at 70° C to speed the reaction. Additional NaOAc was added, as necessary, until reaction was complete. Samples were cooled, centrifuged for 5 minutes at 2000 rpm and the supernatant discarded. Deionized water was added and samples again heated in a water bath. Hydrogen peroxide was added to the samples to oxidize organic matter. After oxidation was complete, samples were cooled and centrifuged at 2000 rpm for 5 minutes and the supernatant discarded. Deionized water was added and samples were shaken to disperse the soil suspension. Soil clays were separated into fine (<0.2um) and coarse (0.2-2pm) fiactions by swinging bucket centrifugation following methods described in Jackson (1975). After fractionation, portions of the clay separates were saturated with either MgCl2 or KCl in preparation for x-ray diffraction (XRD) analysis. Clays were washed free of excess salts by sequential washing with deionized water and 95% ethyl alcohol. The supernatant was discarded. Absence of a cloudy precipitate upon addition of silver nitrate to the supernatant was used to determine removal of chlorides. After evaporation of ethanol, the clay paste was prepared for XRD analysis. The clay paste was oriented on a glass slide with a semi-microspatula (Thiesen and Harward, 1962). When dry, the Mg-saturated clays were placed in a vented dessicator with a dish of ethylene glycol in the bottom. The dessicator was placed in an OV'C des 31 81 oven at 65° C for 4 hours to solvate the clays. After removal from the oven, the dessicator was evacuated with a vacuum pump and the vent closed until the clays were scanned. The K-saturated clays were scanned after being heated to 300° C for 5 hours in a muffle furnace. The K-saturated clays were then heated for an additional 5 hours in a muffle furnace at 550° C and scanned a second time. All x-ray analyses were performed with a Rigaku RTP 300 automated x-ray diffractometer system using CuKa radiation at 45 Kv and 100 mA, 1/2° divergence slit, 1/2° scatter slit, and a 0.03 mm receiving slit. Samples were scanned at a rate of 2° 20 min" from 2 to 32° 20 at a step interval of 005° 20. Clay minerals were identified by their first-order (001) peaks in the clay samples. Expandable minerals were identified by the presence of a > 14A peak in the Mg-EG solvated clay. Illite was identified by a 10A peak in the Mg-EG sample. The 14A peak in Mg-EG was a combination of chlorite and vermiculite. Presence of a 14A peak in K- 300 was attributed to chlorite. Vermiculite was calculated by the difference in peak area of the 14A peak in Mg-EG and K-300. In K-300, the 7A peak was attributed to the second-order (002) peak of chlorite and the fu‘st-order (001) peak of kaolinite. The disappearance of the 7A peak in K-550 was attributed to kaolinite. Kaolinite was calculated by the difference in peak area of the 7A peak in K-300 and the K-550 treatment. Quartz was identified by the presence of a 4.26A peak in all treatments. Peak areas were calculated from the x-ray difl‘ractogarns. The x-ray diffractogams were photocopied onto plain white paper. The baseline was drawn and peaks were labeled and cut out with scissors. The peaks were weighed on an analytical balance. The individual weights were summed to calculate a total peak area. Individual peak weights (areas) 82 were divided by the total weight to obtain the relative percentage of each mineral. The mineral contents were estimated to the nearest five percent. RESULTS AND DISCUSSION Particle Size Analysis Particle size analysis data, as determined by the hydrometer and pipet methods, for 11 soils in the tri-state region is presented in Table 2.1. The soils ranged from 9.1 to 61.1% clay (pipet method). This range in clay content would be representative of all but the most coarsely textured soils in the tri-state area. The ageement between the two methods is quite good. With the exception of the Paulding clay soil, the difference in clay content between the two methods is generally less than two percent. The correlation (Figure 2.1) between the two methods is very high, r2 = 0.993, and the slope is ahnost unity. The regession equation gives confidence that the clay contents of all soils, as determined by the hydrometer method, are both precise and accurate. These data show that either method may be used for routine particle size determination with highly satisfactory results. The average clay content of the six soil series common to the tri-state area is presented in Table 2.2. There are no statistically significant differences in clay content between any of the soil goups. However, the geatest differences occur between the Miami soils, 16.3% average clay content, and the Pewamo soils, 27.4% average clay content. The difference in clay content between these soils probably accounts for the differences in K chemistry that were discussed in Chapter 1. It is worthwhile to mention that the A horizon of the poorly drained Brookston and Pewamo soils have the highest 83 Table 2.1. Particle size distribution of selected soils as determined by the hydrometer and pipet methods of particle size analysis. Hydrometer Pipet Soil County State SandT Silt Clay Sand Silt Clay % % Blount Monroe MI 36.5 41.2 22.3 38.6 38.2 23.2 Conover Genesee MI 41.9 45.0 13.1 41.6 43.9 14.5 Marlette Allegan MI 37.8 43.4 18.8 36.5 43.0 20.5 Miami Ottawa MI 54.8 36.1 9.1 55.5 35.3 . 9.2 Miamian Preble OH 26.1 40.3 33 .6 25.7 40.6 33.7 Morley Randolph IN 19.5 51.6 28.9 19.8 50.2 30.0 Morley Whitley IN 27.9 50.8 21.3 26.2 52.6 21.2 Morley Genesee MI 53.0 35.4 1 1.6 55.3 33.4 1 1.3 Morley Washtenaw MI 40.6 46.8 12.6 40.8 44.4 14.8 Paulding Paulding OH 4.9 37.9 57.2 5.2 33.7 61.1 Pewamo Allegan MI 3 .0 58.5 38.5 2.5 59.9 37.6 TThe percent sand was determined by sieving in both the hydrometer and pipet methods. 70 60 .. 1:1 line 3.? 50 - 0 33. 40 - 8 "5’ 30 - e P = 0.960x + 0.093 I? 20 ‘ if? = 0.993 10 - 0 .O' I I I I I I 0 10 20 30 40 50 60 70 Pipet (% Clay) Figure 2.1. Relationship between the hydrometer and pipet methods of particle size analysis for determination of clay content in 11 soils from the tri-state area. 84 Table 2.2. Mean clay content of the six soil series common to Michigan, Indiana, and Ohio. Soil Number Clay Content Group of soils Mean SET % Miami: 7 16.3 3.1 Conover§ 8 20.1 3.0 MorleyTI 7 20.4 3.2 Blount 7 20.8 1.4 Brookston# 7 22.7 2.4 Pewamo 6 27.4ns 3.4 ns = Means in the same column are not significantly different at the 0.05 probability level. TSE = Standard error of the mean. TMiami includes Miamian soils. §Conover includes Crosby and Celina soils. TTMorley includes Glynwood soils. #Brookston includes Kokomo soils. Table 2.3. Clay content of the six soil series common to Michigan, Indiana, and Ohio. Individual series are compared across states, not between series. Clay Content Soil Number Michigan Indiana Ohio Group of soils Mean SET Mean SE Mean SE - % Miami: 7 l 1.0 0.9 15.7 2.5 24.7ns 9.0 Conover§ 8 22.5 8.0 14.7 1.5 21.3ns 2.5 BrookstonTl 7 19.9 0.9 19.1 2.8 30.5ns 4.5 Morley # 7 18.2 5.2 25.6 4.4 18.7ns 8.7 Blount 7 20.6 2.0 21.6 0.2 20.3ns 5.1 Pewamo 6 26.6 1 1.0 32.2 3.7 23.4ns 2.1 ns = Means in the same row are not significantly different at the 0.05 probability level. TSE = Standard error of the mean. TMiami includes Miamian soils. §Conover includes Crosby and Celina soils. 1| Brookston includes Kokomo soils. #Morley includes Glynwood soils. 85 clay content. This is as expected because these soils are saturated for longer periods and are less leached. Thus, less eluviation of clay from the upper portion of the soil profile has occurred. In Table 2.3, the clay contents of the six common soil series are compared across states. The clay contents of the individual soil series are not significantly different between Michigan, Indiana, or Ohio. The largest difference in clay content is seen in the Miami and Brookston soils. In both cases, the soils from Ohio have higher clay content. Correlations between Clay Content, CBC, and Extractable Potassium Levels An important objective of this study was to determine whether clay content or soil CEC is better correlated with plant uptake and extractable K levels of the whole soil. This information would help determine if K recommendations should be based in some way on clay content or CEC of the soils. The correlation between clay content and plant uptake, r2 = 0.216 (P<0.001), is shown in Figure 2.2. Although the correlation between clay content and plant uptake is low, it is better than the correlation between CBC and plant uptake, r2 = 0.129 (P<0.01), Figure 2.3. Initial exchangeable K levels were not well correlated, r2 = 0.163, with clay content (Figure 2.4) although the relationship was highly significant (P < 0.01). The correlation between initial NaBPhs-K levels and clay content, r2 = 0.764, (Figure 2.5) was much better and was very highly sigrificant (P<0.001). This indicates that the K extractable by NaBPha more closely reflects the quantity of K released from the soil clay than NILOAc. 86 600 500- ’ '29 x 400- E a 300- x s a. 2 200- 8 . E " ' , y =5.71x+99.0 100- . ' r2=0.216 . ., n=54 o I I T 1 I r 0 10 20 30 40 50 60 Percent Clay 70 Figure 2.2. Relationship between the percent clay and plant uptake K for all soils. a: 8 Plant Uptake K (mg K kg") .3 N 8 -5 U" o 8 8 o 8 8 0 Figure 2.3. Relationship between 080 and plant uptake K for all soils. 15 20 25 30 CEC (cmolc kg") 87 Initial NH4OAc-K mg os§§§ f r l I I I 30 40 50 60 Percent Clay 0 _l O N 0 Figure 2.4. Relationship between the percent clay and initial NH40Ac-K for all soils. 7O 2,500 2? 2,000 - X E 1,500 - 3.‘ .E % 1,000 - (B 2 :‘g y = 38.3): + 10.8 g 500 ' ' r2 = 0.764 . . .0 O . n = 54 o I I I I I I 0 10 20 30 40 50 60 Percent Clay Figure 2.5. Relationship between the percent clay and initial five-minute NaBPh4-K for all soils. 70 88 Exchangeable K levels afier cropping were more highly correlated with clay content, 1'2 = 0.887, (Figure 2.6) than NaBPh4-K levels after cropping, r2 = 0.848, (Figure 2.7). The dramatic improvement in the coefficient of determination for the NH40Ac-K levels after cropping suggests that the majority of these soils had been cropped to levels at or near the minimum exchangeable K level. Afier the most easily accessible K had been removed, probably the K added as fertilizer K, the replenishment of exchangeable K became directly dependent upon K release from nonexchangeable forms. Sharpley and Buol (1987) found minimum exchangeable K levels to be closely related to soil clay content. The correlation was improved when the soils were grouped by dominant mineralogy -- kaolinitic, mixed, or smectitic. Figure 2.8 shows that soils with high clay content tended to have higher total K contents than soils with low clay content. The regression equation implies that much of the total K in these soils is in the silt and sand fractions. This portion of total K represents a large reservoir of nonexchangeable K that would become slowly available for plant uptake. The clay fraction, however, contains the more labile K which is very important for supplying the K requirements of rapidly growing crops during a single growing season. The correlation between clay content and CEC, 1'2 = 0.767 is shown in Figure 2.9. There is a higher correlation between clay content and CEC than between organic matter and CBC, r2 = 0.469, (Figure 2.10). When both the organic matter and clay content are combined in a multiple regression equation, the correlation improves slightly, R2 = 0.859. The multiple regression equation takes the following form: CEC = -1.24 + 0.43clay + 2.14OM [2] 89 N 8 § 8 § NH40Ac-K (mg K kg“) after cropping ‘3 0 10 20 30 40 50 60 70 Percent Clay Figure 2.6. Relationship between the percent clay and NH40Ac-K after cropping for all soils. 2000 .‘s” 1800 - & g 1600 - 0 a 1400 J C: ,2 1200 - 3 1000 « x 800 fei’ a; 600 ‘ y = 33.5x - 160 a 400 . r2 = 0.848 g n = 54 g 200 - . . O I fl I I T I 0 10 20 30 40 50 60 70 Percent Clay Figure 2.7. Relationship between the percent clay and five-minute NaBPm—K after cropping for all soils. 35.000 30.000 « .,r~ 25,000 «- 3’ § 20.000 - I 15.000 « E 3 10.000 ' y=295x+11,466 r 2 = 0.750 5,000 - n = 54 o I— I I I fir I 0 10 20 30 40 50 60 70 Percent Clay Figure 2.8. Relationship between the percent clay and total K for all soils. 40 35 - 30 - - '3’ 25 - g 20 - 8 15 - o y = 0526:: + 4.13 10 - r2=0.767 .0 . ”'1“ 5 . o I I I T I I 0 10 20 30 40 50 60 70 Percent Clay Figure 2.9. Relationship between the percent clay and CEO for all soils. 91 35 30 - ° '- '_,. 25 - '52 g 20 - F5 15 - LL! 0 10 - y = 4.24): + 0.888 r 2 = 0.469 5 " n = 54 0 I I I I I I 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Percent Organic Matter Figure 2.10. Relationship between the percent organic matter and CEC for all soils. 500 450 s '7“ 400 T x 350 ‘ 300 -' 250 1 K 150 - 100 - Initial NH.OAc-K (mg to 8 GI 00 1— CEC (cmolc kg") Figure 2.11. Relationship between CEC and initial NH4OAc-K for all soils. 92 These simple and multiple regression equations show that although CEC is more highly correlated with clay content, the CEC is more sensitive to changes in organic matter than to changes in clay content. For soils low in clay content, organic matter is likely to contribute more to total CEC than clay content. However, as the clay content of the soil increases, the importance of the organic contribution to total CEC becomes less. The correlation between CBC and NH40Ac-K or NaBPh4-K before and after cropping is shown in Figures 2.11 to 2.14. The correlation trends found between clay content and NILOAc or NaBPh4 extractable K also hold true for CEC. The correlation between CBC and initial NH40Ac-K, although highly significant (P < 0.01) is low, r2 = 0.158. The correlation improves, however, for exchangeable K after cropping and for NaBPh4-K both before and after cropping. It is important to notice that the correlations between CBC and extractable K levels are consistently lower than the correlations between clay content and extractable K levels. These results suggest that K recommendations based on clay content might provide a more satisfactory fertility index than a K sufficiency level based on CBC. The coefficients of determination for clay content and CEC are summarized in Table 2.4. Conflicting reports on the importance of CEC in determining K availability in soils can be found in the literature. McLean (1976, 1978) considered CEC to be an important factor to consider when establishing K recommendations. Researchers in South America have found exchangeable K levels and CEC to be well correlated in soils from Argentina. Illite concentration and initial exchangeable K levels accounted for 90% of the variability in K uptake by ryegrass during cropping experiments (Zubillaga and Conti, 1996). They found that the clay contributed 73 to 80% of the exchangeable K and 93 2,500 i 1.500 1 Initial NaBPh4-K (mg K k0") § 1O 15 20 CEC (cmolc kg") Figure 2.12. Relationship between CEC and initial five-minute NaBPh4-K for all soils. 250 .s .s N O 0'! O O O O l l l NH4OAc-K (mg K kg") after cropping 01 O y = 5.21): + 3.72 r2=0.672 n=54 O O I I I 10 15 20 cec (cmolc kg") I 25 I 30 35 Figure 2.13. Relationship between CEC and NH4OAc-K after cropping for all soils. 94 2000 1800 s 1600 - 1400 n 1200 -' 1000 - 800 ' 600 - 400 ~ 200 - NaBPh4-k (mg K kg") after cropping 0 5 10 15 20 25 30 35 CEC (cmolc kg") Figure 2.14. Relationship between CEC and five-minute NaBPh4-Kafter cropping for all soils. Table 2.4. Summary of coefficients of determination from regression analysis of percent clay and CEC versus plant uptake and extractable K levels before and after cropping. Plant Initial Initial NI-hOAc-K NaBPh4-K Uptake K NI-hOAc-K NaBPh4-K alter crgpping after cropping r2 Percent Clay 0216*" 0.163” 0.764’" 0887‘" 0.848‘" CEC 0.129" 0.158" 0523*" 0672*" 0.602‘“ "3"" Significant at the 0.01 and 0.001 probability levels, respectively. 95 concluded that clay content and CEC were important considerations for K (Zubillaga and Conti, 1994). Nemeth et al. (1970) reported a positive correlation between the percent K saturation of the inorganic CBC and K concentration in the soil solution. The soil solution K concentration increased with increasing exchangeable K. McLean and Watson (1985) stated that a K sufiiciency level based on soil CEC works best when changes in CEC are a result of differences in clay content and not organic matter content. Walker and Barber (1962) considered the total amount of exchangeable K per volume of soil to be important in controlling K uptake. The percent K saturation afl'ected K availability only where small quantities of K were absorbed by roots. Barber (1981) and Shaw et a1. (1983) presented data showing that differences in K uptake of soils were not predicted by differences in CEC. Shaw et al. (1983) stated that the K concentration in solution and average rate of diffusion more nearly reflected differences in K availability. Barber (1995, p. 253-255) concluded that CBC may affect K uptake in soils having vastly different CEC but at intermediate values of CEC the importance was likely to be minimal. The relative importance of CEC as a factor controlling K availability is probably related to differences in clay mineralogy. Relationship of Clay Mineralogy to Potassium Chemistry of Soils To determine the variability in clay mineralogy of the soils in this study, four soils representing a wide range in total K content were selected for further investigation. The estimated illite content in the clay traction, total soil K content, and total, coarse, and fine clay content are presented in Table 2.5. The objective was to see what differences in the 96 Table 2.5. Estimated illite clay content, total soil potassium content, and total, coarse, and fine clay content of four soils from the tri-state area selected for mineralogical analysis on the basis of total K content? Total Coarse Fine Estimated Clay Clay Clay Soil County State Illite Clay Total K (< 2pm) (2-O.2um) (<0.2um) g kg" mg K kg'l % Conover Genesee MI 13 13,400 14.5 10.9 3.6 Blount Monroe MI 63 18,200 23.2 16.9 6.3 Pewamo Allegan MT 73 23,500 37 .6 27.1 10.5 Pauld'fl Paulding OH 240 29,500 61.1 37.8 23.3 TThe percent total, coarse, and fine clay was determined by the Ohio State University Soil Characterization Laboratory. Table 2.6. Estimated mineral content in the coarse clay (2-O.2um) and fine clay (<0.2p.m) fractions of soils selected for mineralogical analysis from the tri-state area. Conover Blount Pewamo Paulding Minerals CCj FCI CC FC CC FC CC FC % . Illite 10 5 3O 20 25 5 45 30 Vermiculite 35 4O 25 15 15 nd 15 , nd Chlorite <5 nd§ <5 nd <5 <5 <5 nd Kaolinite 30 <5 10 <5 20 <5 5 <5 Quartz <5 nd <5 nd <5 nd <5 nd Expandablefl <5 25 15 30 30 8O 15 45 Interstratified# 10 25 20 35 <5 10 15 20 TCC = Coarse Clay. tFC = Fine Clay. § nd = not detected. 1[Expandable = Minerals that have variable expansion beyond 14 A upon solvation with ethylene glycol. #Interstratified = Interstratified 10 to 14 A clay minerals. 97 K chemistry of the whole soil could be attributed to differences in the clay mineralogy of these soils. Estimated mineral contents of the coarse and fine clay are given in Table 2.6. Table 2.6 shows that these soils are dominated by 2:1 clays. The fine clay contains more expandable and interstratified minerals than the coarse clay. Illite contents are consistently higher in the coarse clay than in the fine clay of the four soils. The Conover soil has the lowest illite content, and the Paulding soil has the highest illite content. The x-ray diffiactograms from these soils are presented in Figures 2.15 to 2.18. As previously stated, an objective of this study was to determine the mineralogy of the clay fraction and determine what inferences could be made regarding the K chemistry of the whole soil. This can be achieved by assuming that one kg of pure illite contains 72g K20 (Burrass, et al. 1996). Burrass et al. (1996) used the reference illite, ITM-l (Clay Mineral Society, Columbia, M0) to estimate illite content in the fine clay fraction of soils of western Ohio because it gave diffraction properties similar to the 10A peak in the soils they studied. If we assume that the K20 content of this reference illite is an accurate estimate of the K20 content of the illite in these soils, then the following comparison can be made. An example for the Conover soil is shown although the logic would follow for the other soils. For the Conover soil, assume: 9213364212): ___§_!Finc la 1000gsoilx 10.9% CC= 109g CC 1000gsoilx3.6% FC=36g FC 109g CC x 10% illite = 10.9g illite 36g FC x 5% illite = 1.8g illite 10.9g + 1.8g = 12.7g illite Then: 1m0g illite = 2.7g illite X = 0.9g K20 in illite of Conover soil 72g K20 X g K20 98 Conover Coarse Clay 17A ' 10A 7" 5A 4.26A | | _lr I 681012l4l6l820222426283032 Degrees 29 Figure 2.15. X-ray diffraction patterns of coarse clay and fine clay from the Conover soil. 99 Blount Coarse Clay 3 3 A 14A 10A 17A I I 7A 5A 4.26A 35A Ii | IMO I K-SSO ____—__..___—1L_ y 1 I I I l I l I I I I I I 24 6 8101214161820222426283032 Degrees 29 Figure 2.16. X-ray diffraction patterns of coarse clay and fine clay from the Blount soil. 100 Pewamo Coarse Clay I 3.3 A L ‘ Mg-EG K-300 I I I I I K-SSO I I | I I H I I I I I I I i : PIfwamo Fhfile Clay; I I I I l I I | I I I I I I I I I I I I I I I I I I I I I I I I I | I I I Mg-EG I I I | I I I I K-300 I I I - | | | I K-SSO n A J I I I r I I I .I.I.I. .I,.I. ,I, .I.I., 2468101214161820222426283032 Degrees 26 Figure 2.17. X-ray diffraction patterns of coarse clay and fine clay from the Pewamo soil. 101 Paulding Coarse Clay | 3 3 A 10 A ' 17A 14A I 7 A I I 5 A 3.5 A 4.26A M -EG I I . , I I I | I I I I I I I I I ‘ ' J I I K-300 I I I I | | K-SSO I K-SSO I T I I I 2468101214161820222426283032 I I I I shin. I I I I I Degrees 26 Figure 2.18. X-ray diffraction patterns of coarse clay and fine clay fi'om the Paulding soil. 102 Using the same logic, the following values would be obtained for the soils: Conover = 0.9g K20; Blount = 4.6g K20; Pewamo = 5.3g K20; Paulding = 17.3g K20. The correlation between K20 content in the illitic clay fraction and total K is, r2 = 0.850 (P < 0.1), for these four soils. This would indicate that some prediction of total K content can be made based on the illitic K content of the clay fiaction. In reality though, the calculated K20 values are probably low because the K in the interstratified and expandable minerals was not accounted for. However, this estimate does give an approximate idea of the range in K content and mineral composition of the clay fraction that could be expected in soils with widely difi‘ering total K contents in the tri-state region. A comparison of the illitezvermiculite ratios in the coarse clay is also interesting because it reveals increasing ratios of 0.3 to 3 from the Conover to the Paulding soil. This ratio suggests that soils high in total K will have higher illitezverrniculite ratios in the coarse clay than soils low in total K. Cox et al. (unpublished data) found that the coarse clay contains larger amounts of NaBPh4 extractable K than any other size fiaction. Mineralogy of Other Soils in the Tri-State Region Burras et al. (1996) found that smectite was the dominate mineral in loess derived soils of a Birbeck-Reesville-Ragsdale toposequence in western Ohio. Loess covers much of the glaciated region of Ohio (Wilding and Drees, 1968; Hock et al. 1973). Smectite comprised a larger portion of the A horizon of the very poorly drained Ragsdale soil than of the A horizon of the Birbeck or Reesville soils. The poorly drained Pewamo soil in this study also contained large quantities of expandable minerals. Smeck et al. (1981) 103 reported that mica comprised 20%, 44%, and 76% of the total clay fiaction of the A horizon in a Brookston, Pewamo, and Hoytville soil respectively from western Ohio. They concluded that clay mineral weathering decreases in a south to north direction across Ohio. Illite content increased with depth in the soil profile. Wilding et al. (1971) found that in Celina and Morley soils, illite comprised 25% of the total clay fraction in the A horizon and increased in relative percentage with depth. It was the dominant mineral in the soil profile. Smock et al. (1968) reported similar findings in Celina and Morley soils of western Ohio. Smith and Wilding (1972) compared Napanee soils from southeastern Michigan and northwestern Ohio and found them to contain similiar quantities of mica, 30 to 40% in the A horizon. Cremeens and Mokma (1987) reported that in a Marlette, Capac, and Parkhill toposequence of south central Michigan, illite comprised 60 to 80% of the fine clay mineralogy in these soils. These values are much higher than the illite content found in the soils of this study. A second study with the Marlette, Capac, Parkhill toposequence in Clinton and Iona counties revealed that in the total clay fraction illite comprised 25 to 30% of the clay mineralogy (Hails-Mariam and Mokma, 1996). These findings are more in agreement with the results of this study. Cummings (1959) estimated illite contents in the total clay fractions of Michigan soils to range from less than 10 to 70% in the A horizon. Reports from Indiana also show a wide range in illite content. White et al. ( 1957) reported illite contents in coarse clay ranging from 17 to 50%. Fehrenbacher et al. (1965) reported illite values of 38 to 60% in the coarse clay and fiom 34 to 52% in the fine clay. Post (1967) found similar values, the coarse clay contained 25 to 60% illite and the fine clay was 22 to 45% illite. 104 Relative Importance of Clay Content, Clay Mineralogy, and CEC in Establishment of Potassium Recommendations Given the wide range in illitic clay content of soils throughout the tri-state, it is not surprising to see large differences in the K chemistry of these soils. The mineralogical composition of the clay governs, in large measure, the quantity and rate of release of K to growing plants. In a fundamental investigation, Sharpley (1990) studied the K chemistry of 102 soils representing 10 soil orders from the United States and Puerto Rico. The soils were divided into three groups (kaolinitic, mixed, or smectitic) based on their dominate clay mineralogy. Potassium was added to the soils at rates of 0 to 250 mg K kg’1 and incubated for 25 weeks. Sharpley found that indexes of water soluble, exchangeable and fixed K were closely related to the clay content for each soil group. The K buffering capacity was significantly greater for the smectitic soils than the mixed or kaolinitic soils and was closely related to the clay, CBC, and K saturation for each group of soils. Sharpley (1990) found that K added to the soil was distributed in water soluble, exchangeable, and nonexchangeable forms in proportion to the clay content and dominate mineralogy of the soil. These findings suggest that for accurate K recommendations to be made, the clay content and dominate mineralogy of the soil must be known. The results from this study and other studies in the tri-state region reveal that the soils in the tri-state area can be placed in the broad category of smectitic mineralogy. This is probably the reason why CEC is not a major factor controlling K availability in soils of the tri-state region. If one considers the dominate mineralogy of soil groups, then the relative importance of CEC is likely to be: kaolinitic > mixed > smectitic. The original concept of incorporating CEC into a sufficiency index for K recommendations 105 originated with Dr. Fred Adams in Alabama. Fisher used the concept and applied it to Missouri soils (Dr. Jay Johnson, personal communication). It is reasonable to assume that a sufficiency relationship based on CEC might be appropriate for the more highly weathered soils of the southern USA. These Ultisols would have very little illite remaining in them. Therefore, the main supply of K would be from the exchangeable K. It would logically follow that K recommendations based on a percentage saturation of the exchange complex might be a good approach under these conditions. In the tri-state, however, despite the large degree of variability in illite content, soils still contain significant reserves of nonexchangeable K in the clay and coarser fiactions which can continue to supply K to plants under intensive cropping conditions. Exchangeable K levels on research plots at the Hoytville research station in northwest Ohio have maintained an exchangeable K level of 150 ppm without fertilization for the past 35 years (Dr. Jay Johnson, personal communication). In Ohio, a response to K is not seen on silt loam soils in most years if soils have an exchangeable K level above 90 ppm. In light of this fact, it is doubtful whether the high soil test levels achieved by use of Fisher’s equation: K “ch 1",. (ppm) = 110 + 2.5 x CEC [3] are merited for soils in the tri-state. Because the soil CEC is very sensitive to changes in organic matter relative to changes in clay content, it may be more appropriate to base K recommendations on the clay content of soils in the tri-state area. The data fi'om this study and data fi-om Sharpley (1990) would support this approach. As long as soil test levels are maintained at a reasonable level, most soils will be able to supply K in amounts adequate for maximum crop yields. 106 CONCLUSIONS The results of this study suggest that clay content and mineralogy are more important factors to consider than CEC when establishing K recommendations. Soils in the tri-state region are dominated by complex mixtures of 2:1 clays and have large reserves of nonexchangeable K in the clay fiaction which can supply K to crops. The silt and sand fractions also contain large quantities of nonexchangeable K that may become slowly plant available over time. This suggests that the practice of increasing soil test values to high levels and fertilizing the soil per se rather than the crop is not generally necessary and may represent an unjustifiable economic cost to the producer. The base soil test level that is adopted for sufficiency will affect K fertilization practices. The choice of 110 ppm proposed by Fisher (1974) seems rather arbitrary and should be investigated in more detail to determine if it truly is an accurate approximation of the minimum K soil test value required for sufficiency. McLean (1985) acknowledged that the equation did overestimate the K sufficiency level on soils of low CEC but felt that this should not be a problem because leaching losses of K were more likely on soils of low CEC. Potassium recommendations calculated from a sliding scale that is related to clay content may be a closer approximation of the true sufficiency level for many soils in the tri-state area. Field experiments need to be conducted to determine which approach has the greater merit. 107 REFERENCES Al-Kanani, T., A.F. MacKenzie, and G.J. Ross. 1984. Potassium status of some Quebec soils: K released by nitric acid and sodium tetraphenylborn as related to particle size and mineralogy. Can. J. Soil Sci. 64:99-106. Barber, S.A. 1981. Soil chemistry and the availability of plant nutrients. p. 1-12. In M. Stelly (ed.) Chemistry in the soil environment. ASA Spec. Publ. 40. ASA, CSSA, and SSSA, Madison, WI. Barber, S.A. 1995. Soil nutrient bioavailability: A mechanistic approach. 2nd ed. John Wiley and Sons, Inc., New York, NY. Bertsch, P.M., and G.W. Thomas. 1985. Potassium status of temperate region soils. p. 131-162. In R.D. Munson (ed.) Potassium in agriculture. ASA, CSSA, and SSSA, Madison, WI. Brady, N.C., and RR. Weil. 1996. The nature and properties of soils. 11th ed. Prentice- Hall, Upper Saddle River, NJ. Burras, L., N.E. Smeck, and J .M. Bigham. 1996. Origin and properties of smectite in loess I -derived soils of western Ohio. Soil Sci. Soc. Am. J. 60:1961-1968. Conyers, E.S., L.P. Wilding, and E0. McLean. 1969. Influence of chemical weathering on basal spacings of clay minerals. Soil Sci. Soc. Am. Proc. 33:518-523. Cox, A.E., and BC. Joern. 1997. Release kinetics of nonexchangeable potassium in soils using sodium tetraphenylboron. Soil Sci. 162: 588-598. Cremeens, D.L., and D.L. Mokma. 1987. Fine clay mineralogy of soil matrices and clay films in two Michigan hydrosequences. Soil Sci. Soc. Am. J. 51:1378-1381. Cummings, S.L. 1959. Relationships of potassium fixation and release to the clay mineral composition of some Michigan soils. M.S. Thesis. Michigan State University, East Lansing, MI. Fehrenbacher, J .B., J .L. White, A.H. Beavers, and R.L. Jones. 1965. Loess composition in southeastern Illinois and southwestern Indiana. Soil Sci. Soc. Am. Proc. 29:572-579. 108 Fisher, TR 1974. Some considerations for interpretation of soil tests for phosphorus and potassium. Missouri Agric. Exp. Stn. Res. Bull. 1007. Gee, G.W., and J .W. Bauder. 1986. Particle size analysis. p. 383-411. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison,WI. Gil-Sotres, F., and B. Rubio. 1992. Kinetics and structural effects of the extraction of nonexchangeable potassium from the clay fraction of soils of Galacia (N .W. Spain). Comm. Soil Sci. Plant Anal. 23:143-156. Haile-Mariam, S., and D.L. Mokma. 1996. Mineralogy of two fine-loamy hydrosequences in south-central Michigan. Soil Survey Horizons. 37:65-74. Hinsinger, P., F. Elsass, B. Jaillard, and M. Robert. 1993. Root-induced irreversible transformation of a trioctahedral mica in the rhizosphere of rape. J. Soil Sci. 44:535-545. Hock, A.G., L.P. Wilding, and G.F. Hall. 1973. Loess distribution on a Wisconsin-age till plain in southwestern Ohio. Soil Sci. Soc. Amer. Proc. 37:732-738. Jackson, ML. 1975. Soil chemical analysis. Advanced course. Published by the author, University of Wisconsin, Madison, WI. Kunze, G.W., and J .B. Dixon. 1986. Pretreatment for mineralogical analysis. p. 91-100. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Lui, 6., NE. Nielsen, H.C.B. Hansen, and OK. Borggaard. 1997. Mineral changes in a Danish Alfisol caused by 30 years of potassium depletion in the field. Acta. Agric. Scand., Sect. B. Soil and Plant Sci. 47 :1-6. McLean, ED. 1976. Exchangeable K levels for maximum crop yields on soils of diff- erent cation exchange capacities. Comm. Soil Sci. Plant Anal. 7:823-838. McLean, ED. 1978. Influence of clay content and clay composition on potassium availability. p. 1-19. In G.S. Sekhon (ed.) Potassium in soils and crops. Proc. Pot. Res. Inst. India, New Delhi, India. Nov. 16-17, 1978. Pot. Res. Inst. India. New Delhi, India. McLean, E.O., and ME. Watson. 1985. Soil measurements of plant-available potassium. p. 277-308. In R.D. Munson (ed.) Potassium in agriculture. ASA. CSSA, and SSSA, Madison, WI. 109 Moberg, J .P., J .D. Nielsen. 1983. Mineralogical changes in soils used for potassium- depletion experiments for some years in pots and in the field. Acta Agric. Scand. 33:21-27. Mortland, M.M., K. Lawton, and G. Uehara. 1957. Alteration of biotite to vermiculite by plant growth. Soil Sci. 82:477-481. Nemeth, K., K. Mengel, and H. Grimme. 1970. The concentration of K, Ca, and Mg in the saturation extract in relation to exchangeable K, Ca, and Mg. Soil Sci. 109: l 79- l 85. Nielsen, J .D., and J .P. Moberg. 1984. The influence of K-depletion on mineralogical change in pedons from two field experiments and in soils from four pot experiments. Act. Agric. Scand. 34:391-399. Post, DR, and J .L. White. 1967. Clay mineralogy and mica-vermiculite layer charge density distribution in the Switzerland soils of Indiana. Soil Sci. Soc. Amer. Proc. 31:419-424. Ross, G.J., P.A. Phillips, and J .LR Culley. 1985. Transformation of vermiculite to pedogenic mica by fixation of potassium and ammonium in a six-year field manure application experiment. Can. J. Soil. Sci. 65:599-603. Sharpley, A.N. 1990. Reaction of fertilizer potassium in soils of differing mineralogy. Soil Sci. 149:44-51. Shaw, J .K., R.K. Stivers, and S.A. Barber. 1983. Evaluation of differences in potassium availability in soils of the same exchangeable potassium level. Comm. Soil Sci. Plant Anal. 14:1035-1049. Smeck, N.E., A. Ritchie, L.P. Wilding, and LR. Drees. 1981. Clay accumulation in sola of poorly drained soils of western Ohio. Soil Sci. Soc. Am. J. 45:95-102. Smeck, N.E., L.P. Wilding, and N. Holowaychuk. 1968. Genesis of agrillic horizons in Celina and Morley soils of western Ohio. Soil. Sci. Soc. Am. Proc. 32:550-556. Smith, H., and LP. Wilding. 1972. Genesis of agrillic horizons in ochraqualfs derived fi'om fine textured till deposits of northwestern Ohio and southeastern Michigan. Soil Sci. Soc. Amer. Proc. 36:808-815. Thiesen, A.A., and ME. Harward. 1962. A paste method for preparation of slides for clay mineral identification by x-ray diffraction. Soil Sci. Soc. Am. Proc. 26:90-91 . 110 Tributh, R, EV. Boguslawski, A.V. Lieres, D. Steffens, and K. Mengel. 1987. Effect of potassium removal by crops on transformation of illitic clay minerals. Soil Sci. 143 :404-409. Walker, J .M., and S.A. Barber. 1962. Absorption of potassium and rubidium from the soil by com roots. Plant Soil. 17:243-259. White, J .L., G. Talvenheimo, M.G. Klages, and M.M. Phillippe. 1957. A survey of the mineralogy of Indiana soils. Ind. Acad. Sci. 66:232-241. Wilding, LR and LR. Drees. 1968. Distribution and implications of sponge spicules in surficial deposits in Ohio. Ohio J. Sci. 68:92-99. Wilding, L.P., L.R. Drees, N.E. Smeck, and GP. Hall. 1971. Mineral and elemental Composition of Wisconsin-age till deposits in west central Ohio. p. 290-317. In R.P. Goldthwait (ed.) Till: A symposium. Ohio State Univ. Press, Columbus, OH. Zubillaga, M.M., and M. Conti. 1994. Importance of the textural fiaction and its miner- alogic characteristics in the potassium contents of different Argentine soils. Comm. Soil Sci. Plant Anal. 25:479-487. Zubillaga, M.M., and ME. Conti. 1996. Availability of exchangeable and nonexchangeable potassium in Argentine soils with different mineralogy. Z. Pflanzenernahr. Bodenk. 159:149-153. APPENDICES lll .5858 58... 585 .885 .25.. .m N 5. ..z 5 .5 .NN .8... .5 32 ..N no 58... 8.85 .3858 58< 585 .885 .25.. 3 ..N.. .55 .5. .55... z .9. .NN .55 .3 a ..o 855.0 8.85 .3558 58... 585 .885 .25.. .3 . 5. ..z 5N ..N .8 .8... .5 35 N. z. .58.... 386 55.5580 58... 585 .885 .25.. .3 . .5 ..z n. .... .5. .8. .5 32 z z. 8.558.. 8.85 .5858 5:55 585 885558.85... .m .. 5. ..m N ... .5N .8. .5 mm 5N .2 3.85.83 8.580 5.5.55.8 5.5.5 585 885555585... .5 m 5. ..z N .5 .5 .8... .5 55 MN .2 8880 8.680 .3858 5:5... 585 885558.85... .3 z 5. ..z n ... .2 .8... .5 3... N8 .2 85.5 8.6.80 .535... 5...... 585 .885 .25.. .m N 5. ..z 5 .5 .NN .8... .5 32 .8 ..o 58... 55.8 5.5.5580 58< 585 .885 558.85.. .m 2 5. ..z 5 ..5 .55 .8... .5 mm 5.8. .2 8.5.5 8.8 5.8.5580 58... 585 .885 558.85... .3 ... 5. ..z N .5 .5. .8. .5 3... ..N .2 8.5.0 8.80 58.858 58... 585 .885 5555.8... .3 m. 5. ..z .. ... .5N .8... .5 32 NN .2 .5552 856 58.8.8... 555.. 585 .885 558.85.. .3 . 5. ..z ..N ... .85 .8... .5 35 N.. 2. 8.8.5.. 8.58... 58.8.»... 5.5 585 .885 558.85.... .3 . .5 ..z 8. ... .. N .8... .5 m... ..N 2. 5555.58.. 8.8.8.... 58.5.»... 5...... 585 .885 558.85.. .m .. 5. ..m N .5 .8N .8... .5 32 NN .2 3.8.553 88.555. =85»... 5.5 5.85 .885 555585.. .m m .5 ..z N .5 .5. .8... .5 52 NN .2 888.. 88.8.5. 58.8.52 5.5 585 .885 555585.. .3 n. 5. ..z .. .... .55 .8... .5 32 NN .2 855.... 88.55.. 55.55.50 58< 585 .55... .25... 5. N. .5 ..z . ..5 .8 .8... .5 ..z N. :o .885 58:. 5555.80 58< 585 .55... .25.. .m a .5 ..m . .... .2 .8. .5 35 «N ..o .825: .55... 55.8550 58... 585 .55... .25.. .m m. .5 ..z ..N 5. .m .8... .5 55 NN 2. 55.8.5. .58... 55.5550 58... 585 .55... .25... .m. 5. 5 ..z .5 .5 ..N .8. .5 ..z 5N z. 8.553 58... 58.5580 58... 585 .55... .25.. .m 8 5. ..m 8 .5 .8. .8... .5 mm 85 .2 8552 .55... 55.85.80 58... 585 .55... .25.. .3 N 5. ..z N .... .8. .8. .5 35 ..N .2 .8556 .55... 55.55.50 58... 585 .55... .25.. .3 2 .5 ..z .. .... .5 .8. .5 32 m .2 5.8.2 .55... 8.5.... =8 55.8.... 822 2.5 35.5 8.8m 5.2 58 82.85320 £82.88. .505 use 2.8 .«o .8538: ._.< 038,—. < X—GZHA—A—< 112 5.8.5.5.... 5555 585 .55... .25... .5 8. .5 ..2 .8 .5 .5N .88 85 58 8N 2. 8.553 5582 588.88.. 5855 585.55... .25.. .5 8. .5 ..2 8N .5 .8. .88 .5 32 8N 2. 58.88.55 5582 555.88.. 5855 585.55... .255 .5 8 5 .5 . .5 .5N .8... .5 52 .. .2 358.583 5582 588...... 5855 585 .55... .25... .5 5 5 ..2 8 .5 .8N .8. .5 32 8. .2 88.8.8 5582 .8....885 58< 585 58828.88. .885 .255 .5 8 .5 ..2 .. .5 .8 .8... .5 52 N8 .2 385588 55.5352 5.55.8: 5855 585 .885 .255 .5 N 5 ..2 8 .5 .5. .88 .5 38 5N :o 582.. 55552 5.8.5.8.: 5855 585 .885 .25... 3 ..88 ..8 .8.. .583 2 ..8N .5N .8... .55 8. :0 85.5.5 55552 5.8.5.8.: 5855 585 885 558.85.. .3 . .5 ..2 8N .5 .8 .8.. .5 38 88 2. 5.8.8.. .552 5.88.8... 5855 585 885 5:55.88... .3 . 5 ..2 8. .5 ..8 .8... .5 58 ..N 2. 85558.. .552 588.88.. 5855 585 885 55.5.8.5 .5 8 5 ..8 N .5 .N8 .88 .5 38 8N .2 388.583 .552 5.88.8... 5855 585 885 558.88... .5 5 .5 ..2 8 .5 .88 .8... .5 32 8. .2 888.8 .552 8.5.8.8... 5855 585 885 558.85... .3 8. 5 ..2 8 .5 .8. .8.. .5 58 ..8 .2 58.5 .552 588.88.. 588888.88 585 .885 558.88.... .5 8. .5 ..2 8 .5 .5. .8.. .5 32 858. .2 85.58 38.82 588.88.. 8588885 585 .885 5885.855 .3 8 5 ..2 8 .5 .8 .8... .5 58 8 .2 855.5 88.82 58.5.88: 58888.0 585 .885 5:85.85... .3 .. .5 ..2 8 .5 .5N .8... .5 32 8N .2 5.55.... 28.82 588855.... 5855 585 .885 .25.. .5 N 5 ..2 8 .5 .5N .88 .5 32 ..8 ..o 582.. 858.3. 888855.... 5855 585 .885 .255 3 .8.. 88.8.. .55... 2 ..8N .88 .88 .55 . 8 ..o 88.5.5 858.3. 8.88.5.8 5.82 585 .55... .25... .5 . .5 ..2 N .5 .8N .8.. .5 52 8 :o 5.5.55... 55858.. 58885580 5882 585 .55... .255 .5 8 5 ..8 8 .5 .8. .88 .5 38 8. .2 8582 55258.. 5.8.5.85. 5888. 585 .55... .255 .5 5. 5 ..2 . .5 .8 .88 .5 52 N8 :0 8888 88385.0 8.8.5.8.... 588... 585 .55... .255 .5 8 .5 ..8 . .5 .. .8... .5 38 ..N ..o 5825.. 88385.... 555.88.. 588... 585 .55... .255 .3 8. 5 ..2 8 .5 .8. .88 .5 38 8. .2 5.55.... 8855.0 5.585 ..85 85.83 888m 8.35 5.8880 8.85 8.2 ..88 6.58. ...< 5885 113 88.85555. 5555 585 .885 .255 .5 8. .5 ..2 N .5 .5N .8.. .5 58 88 ..o 8888 8558.. 8888552 5855 585 .885 .255 .5 8 .5 ..8 . .5 .8. .88 .5 38 8N ..o 5825.. 85838.. 888858.... 5855 585 .885 .255 .5 8. 5 ..2 .8 .5 .. 8 .88 .5 52 8N 2. 58.553 85838.. 8888552 5555 585 .885 .255 .5 8. .5 ..z 8N .5 .8 .8.. .5 58 8N 2. 55.8855 8588.. 588855.... 5855 585 885.255 .5 8 .5 ..8 8 .5 .8. .8.. .5 58 8. .2 8582 8558.. 58:855.... 5855 585 .885 .255 .3 N. .5 ..2 8 .5 .. .88 .5 58 8 .2 5.55.... 8528.. 58.8.88: 5.25 585 .8828: .55... .85..-58> .5 8.5 ..2 8 .5 .8. .88 .5 58 8 ..o 558.5... 55.55... 5.88.8.8: 5.82 585 8.855 .885 5588.855 .5 8. 5 ..2 8 .5 .88 .8.. .5 58 88N. .2 25.5.8 .5585 5.3.8.5... 5882 585 5252. .885 558.855 .3 8 .5 ..2 5 .5 .8. .8.. .5 38 8N .2 5.5.... 8.5.8.. 5.5.5 ..88 85.88.. 82.5 238 5.550 8.88 8.2 =85 8.58. ...< 2885. 114 Appendix B Table B.1. Selected physical and chemical properties of all soils. Glacial Soil County State Lobe [91! OM CEC Sand Silt Clay Texturet °/o cmolc kg" % Marlette Allegan MI Michigan 5.8 4.8 17.1 36.5 43.0 20.5 L Marlette Clinton Ml Saginaw 5.8 3.7 12.4 35.7 47.6 16.7 L Marlette Sanilac Ml Huron-Erie 7.6 3.0 10.9 42.0 41.3 16.7 L Capae Allegan MI Michigan 6.9 2.5 12.5 38.6 42.2 19.2 L Capac Clinton MI Saginaw 7.0 4.1 14.3 25.8 51.4 22.8 SiL Capae Sanilac MI HuroncEI-ie 6.0 2.9 10.5 43.9 42.5 13.6 L Parkhill Clinton MI Saginaw 7.0 4.7 16.8 26.9 52.8 20.3 SiL Parkhill Sanilac MI Huron-Erie 7.3 3.0 13.1 34.4 45.9 19.7 L Miami Ottawa MI Michigan 7.1 2.8 8.1 55.5 35.3 9.2 SL Miami Genesee Ml Saginaw 7.8 1.6 7.1 56.4 31.5 12.1 SL Miami Washtenaw MI Huron-Erie 6.0 3.1 9.3 43.9 44.5 11.6 L Miami Pulaski IN 6.1 3.4 14.7 47.1 34.7 18.2 L Miami Hendricks [N 7.0 2.7 10.9 31.1 55.8 13.1 SiL Miamian Preble OH 7.7 2.4 17.4 25.7 40.6 33.7 CL Miamian Clinton OH 5.6 3.5 1 1.8 18.8 65.5 15.7 SiL Conover Ottawa MI Michigan 6.2 4.4 25.0 13.7 47.9 38.4 SiCL Conover Genesee MI Saginaw 6.4 3.1 10.2 41.6 43.9 14.5 L Conover Washtenaw MI Huron-Erie 6.6 2.3 9.7 51.0 34.4 14.6 L Crosby Pulaski IN 6.2 2.9 12.4 52.5 34.4 13.1 SL Crosby Hendricks IN 6.9 2.7 13.3 31.8 52.0 16.2 SiL Celina Preble OH 7.0 3.0 1 1.9 34.1 45.2 20.7 L Crosby Preble OH 7.0 3.4 13.4 18.8 64.0 17.2 SiL Crosby Clinton OH 7.7 2.7 18.5 16.5 57.6 25.9 SiL Brookston Allegan MI Michigan 6.8 3.8 13.0 42.1 39.7 18.2 L Brookston Genesee Ml Saginaw 5.6 3.5 13.7 40.2 39.6 20.2 L Brookston Washtenaw MI Huron-Erie 7.1 4.4 20.5 48.6 30.0 21.4 L Brookston Pulaski [N 6.5 4.5 19.3 49.6 34.2 16.2 L Brookston Hendricks IN 7.8 3.8 18.9 19.1 59.0 21.9 SiL Kokomo Preble OH 7.1 4.4 21.1 13.8 60.2 26.0 SiL Kokomo Clinton OH 5.9 5.8 31.1 9.7 55.3 35.0 SiCL 1 Texture (L = Loam, SiL = Silt Loam, CL = Clay Loam, SiCL = Silty Clay Loam, SL = Sandy Loam c = Clay, sac = Silty Clay) Table 3.1. (cont’d) 115 Total K NaBPhrK NaBPIu-K Extracted Before After I Glacial Total NaBPh4-K By NaBPh. Nek Plant Plant t Soil County State Lobe K 7 Days: In 7 Dajs 7 Dast ULtalfi Uptake LU — mg K kg'I —- — "/o — ‘ Marlette Allegan MI Michigan 18,800 5,1 10 27 4,900 830 490 Mariette Clinton Ml Saginaw 17,500 4,690 27 4,540 600 370 Mariette Sanilac Ml Huron-Erie 16,600 5,190 31 5,120 540 330 Capae Allegan MI Michigan 19,800 6,130 31 6,010 650 450 Capae Clinton M1 Saginaw 19,100 5,500 29 5,200 1,090 620 Capae Sanilac M] Huron-Erie 1 5,900 4,490 28 4,430 380 330 Parkhill Clinton M1 Saginaw 19,400 5,680 29 5,410 970 620 Parkhill Sanilac Ml Huron-Erie 16,100 5,850 36 5,740 480 310 Miami Ottawa MI Michigan 16,400 1,570 10 1,450 250 100 Miami Genesee M1 Saginaw 15,000 4,160 28 4,020 480 270 Miami Washtenaw Ml Huron-Erie 15,800 3,470 22 3,190 630 330 Miami Pulaski 1N 13,400 2,580 19 2,240 770 350 Mhml Hendricks 1N 14,400 1,900 13 1,820 220 120 Miamian Preble OH 21,000 8,030 38 7,670 1,810 1,120 1 Miamian Clinton OH 1 5,500 3,720 24 3,610 420 250 Conover Ottawa MI Michigan 24,600 1 1 ,200 46 1 1,000 1,650 1,490 I Conover Genesee Ml Saginaw 13,400 3,190 24 3,090 340 210 Conover Washtenaw Ml Huron-Erie 15,100 4,240 28 4,140 550 350 Crosby Pulaski IN 13,800 1,590 12 1,470 360 210 Crosby Hendricks IN 15,700 2,370 15 2,250 340 220 Celina Preble OH 17,300 5,200 30 5,070 740 560 Crosby Preble OH 16,500 3,270 20 3,130 460 330 Crosby Clinton OH 16,100 5,210 32 5,030 860 540 Brookston Allegan MI Michigan 16,400 4,040 25 3,680 830 340 Brookston Genesee MI Saginaw 16,900 5,770 34 5,600 980 760 Brookston Washtenaw Ml Huron-Erie 14,700 4,290 29 4,120 930 610 Brookston Pulaski IN 14,400 2,560 18 2,350 720 410 Brookston Hendricks IN 15,800 3,180 20 2,970 570 370 Kokomo Preble OH 17,800 5,400 30 5,160 930 710 Kokomo Clinton OH 17,500 6,250 36 6,040 1,000 730#_ \ INaBPh4-K 7 Days— = K extractable by NaBPh4 in 7 days §Nek 7 Days= Nonexchangeable K released in 7 days (Nonexchangeable K= NaBPh4 extractable.r 1NaBPh4-K Before Plant Uptake= K level of the soil determined by a five-minute extraction vrrithj,r #Nek Before Plant Uptake= Nonexchangeable soil K level before Neubauer growth study (five-raj TTNlhOAc-K Before Plant Uptake= ItPlant Uptake= Exchangeable K level of the soil determined by a five-mints; Amount of K taken from the soil by barley plants during Neubauer growth study?! §§Plant Uptake From Exk= The percent of K taken up by plants from the exchangeable soil K dur J: NPlant Uptake From Nek= The percent of K taken up by plants from the nonexchangeable soil K54 N ek N ek N H4OAc-K NILOAc-K Plant Plant Before After Before After Uptake Uptake belta Plant Plant Delta Plant Plant Delta Plant From From ghrK Uptake“ Uptake Nek Uptakefi Uptake NHfiAc-K flukeg Exkfi NeKfi - mg K kg'I % 340 620 410 210 210 80 130 247 53 47 230 450 310 140 150 60 90 147 61 . 39 210 470 280 190 70 50 20 105 19 81 200 530 370 160 120 80 40 150 27 73 I70 790 520 2 70 300 100 200 374 53 47 50 320 290 30 60 40 20 38 53 47 $50 700 510 190 270 110 160 295 54 46 70 370 250 120 1 10 60 50 1 12 45 55 50 130 60 70 120 40 80 91 88 12 {10 340 220 120 140 50 90 158 57 43 £00 350 270 80 280 60 220 250 88 12 20 430 280 150 340 70 270 331 82 18 00 140 60 80 80 60 20 26 77 23 90 1,450 970 480 360 150 210 517 41 59 70 310 200 110 110 50 60 67 90 10 60 1,450 1,350 100 200 140 60 244 25 75 30 240 160 80 100 50 50 74 68 32 00 450 290 160 100 60 40 159 25 75 50 240 1 70 70 1 20 40 80 108 74 26 20 220 150 70 120 70 50 71 70 30 80 610 480 130 130 80 50 171 29 71 30 320 270 50 140 60 80 154 52 48 20 680 450 230 180 90 90 134 67 33 90 470 250 220 360 90 270 412 66 34 20 810 670 140 170 90 80 215 37 63 20 760 510 250 170 100 70 254 28 72 10 510 340 170 210 70 140 260 54 46 )0 360 280 80 210 90 120 148 81 19 20 690 620 70 240 90 150 324 46 54 70 790 630 160 210 100 110 125 88 12 inus NH40Ac extractable K). Ph4 before Neubauer growth study. NaBPh4 extractable K minus five-minute N1140Ac extractable K). action with NH40Ac before Neubauer growth study. leuabuer growth study. rg Neubauer growth study. 116 Table 13.1. (cont’d) Glacial Soil County State Lobe pH OM CEC Sand Silt Clay Texturet % cmolc kg" °/o Michigan 6.8 2.6 13.2 36.3 35.3 28.4 CL Saginaw 7.1 3.6 10.2 55.3 33.4 11.3 SL Huron-Erie 5.8 2.7 9.3 40.8 44.4 14.8 L 6.5 5.6 13.2 26.2 52.6 21 .2 SiL 7.9 3.5 18.5 19.8 50.2 30.0 SiCL 7.1 1.6 7.4 58.1 31.9 10.0 SL 7.1 2.9 16.1 19.8 52.8 27.4 SiCL Michigan 5.9 3.5 11.3 48.3 35.0 16.7 L Saginaw 5.9 4.8 17.4 23.4 54.8 21.8 SiL Huron-Erie 6.5 4.4 17.5 38.6 38.2 23.2 L 6.2 3.3 18.4 27.2 51.0 21.8 SiL 7.1 4.2 16.6 10.6 68.1 21.3 SiL 6.6 2.9 1 1.2 34.4 50.4 15.2 SiL 5.8 2.7 15.7 19.7 54.9 25.4 SiL Michigan 6.8 5.9 30.0 2.5 59.9 37.6 SiCL Huron-Erie 5.7 2.8 11.4 47.1 37.3 15.6 L 6.9 3.9 22.5 30.6 40.9 28.5 CL 7.7 4.0 27.5 13.9 50.2 35.9 SiCL 5.6 4.0 18.4 23.1 52.5 24.4 SiL 5.8 3.5 16.2 19.7 58.0 22.3 SiL Huron-Erie 6.5 3.5 16.3 46.8 29.4 23.8 L Hoytville Paulding 6.2 3.6 24.2 13.6 46.5 39.9 SiCL Paulding Paulding 6.5 5.5 30.4 5.2 33.7 61.1 C Miatgxllfi’ Saginaw MI Saginaw 7.9 3.3 27.2 5.3 45.9 48.8 SiC Glynwood Allegan Morley Genesee Morley Washtenaw Morley Whitley Morley Randolph Glynwood Hancock Glynwood Seneca Blount Allegan Blount Clinton Blount Monroe Blount Whitley Blount Randolph Blount Hancock Blount Seneca Pewamo Allegan Pewamo Monroe Pewamo Whitley Pewamo Randolph Pewamo Hancock Pewamo Seneca Hoytville Monroe SEEEEEEEEEQEEEEEESEEEEE TTexture (L= Loam, SiL= Silt Loam, CL = Clay Loam, SiCL = Silty Clay Loam, SL = Sandy Loam C = Clay, SiC = Silty Clay) Table 8.1. (cont’d) 117 Total K NaBPh-K NaBPhAl Nr Extracted Before After Brf Glacial Total NaBPhrK By NaBPh. Nek Plant Plant 1 I’ll Soil County State Lobe K 7 Days; In 7 Dgs 7 Days§ UptakeL Uptake M — mg K kg"——- — % — ‘. Glynwood Allegan MI Michigan 23,100 8,250 36 8,100 1,070 880 l 9'; Morley Genesee Ml Saginaw 13,800 2,870 21 2,790 280 190 1 2r Morley Washtenaw M1 Huron-Erie 17,000 4,590 27 4,160 970 420 E 5. Morley Whitley 1N 17,000 5,250 31 5,110 1,070 550 l 9 Morley Randolph 1N 18,900 7,930 42 7,670 1,100 820 I g Glynwood Hancock OH 17,900 3,260 18 3,180 370 220 'r 2 Glynwood Seneca OH 21,100 8,790 42 8,530 1,480 850 a ll Blount Allegan MI Michigan 19,200 5,730 30 5,430 930 530 ' 6 Blount Clinton Ml Saginaw 18,900 5,380 28 5,160 830 560 ? 6 Blount Monroe MI Huron-Erie 18,200 5,710 31 5,520 1,120 920 r g Blount Whitley 1N 18,000 5,410 30 5,210 750 490 . 5 Blount Randolph 1N 17,300 4,150 24 3,960 730 510 i 5 Blount Hancock OH 17,300 3,740 22 3,600 440 250 r 1 Blount Seneca OH 19,700 7,140 36 6,920 1,010 690 . Pewamo Allegan MI Michigan 23,500 7,640 33 7,420 1,320 1,090 . 1.1 Pewamo Monroe MI Huron-Erie 16,300 3,620 22 3,440 920 520 r - Pewamo Whitley 1N 18,500 5,940 32 5,760 1,000 680 .‘ 1 Pewamo Randolph 1N 21,300 7,680 36 7,220 1,570 1,070 . l.‘ Pewamo Hancock OH 1 8,800 6,950 37 6,740 1 ,070 860 t Pewamo Seneca OH 20,200 6,420 32 6,100 1,170 830 J Hoytville Monroe MI Huron-Erie 1 7,600 5,200 30 5,070 870 730 . Hoytville Paulding OH 24,400 10,700 44 10,480 1,780 1,330 l Paulding Paulding OH 29,500 16,200 55 15,930 2,250 1.890 f. Misteguay Sgginaw Ml Sa 27,900 9,000 32 8,740 1,390 1,0'L ' INaBPh4-K 7 Days = K extractable by NaBPh4 in 7 days Q §Nek 7 Days = Nonexchangeable K released in 7 days (Nonexchangeable K = NaBPh4 extractrlblrjrjH filNaBPh4-K Before Plant Uptake = K level of the soil determined by a five-minute extraction witjrjf #Nek Before Plant Uptake = Nonexchangeable soil K level before Neubauer growth study (five-s an 0' 1'1'NH40Ac-K Before Plant Uptake= IIPIant Uptake— = Amount of K taken from the soil by barley plants during Neubauer growth stud §§Plant Uptake From Exk= The percent of K taken up by plants from the exchangeable soil K dtjr,“e “Plant Uptake From Nek= The percent of K taken up by plants from the nonexchangeable will: Tub Exchangeable K level of the soil determined by a five- -minil 5m MBMV Nek Nek - Nl-LOAc-K NILOAc-K Plant Plant | Brian 1 Before After Before After Uptake Uptake '4 N11 1111 Delta Plant Plant Delta Plant Plant Delta Plant From From MMPh-K Uptakett Uptake Nek Uptaketj Uptake NILOAc-K Uptaken Exkfi NeKfi mg K kg" % 3.100 1.011 1190 920 780 140 150 100 so 220 23 77 290 211 90 200 150 so so 40 4o 47 85 15 41700 911 550 540 340 200 430 so 350 474 74 26 5,110 1.11 :520 930 470 460 140 so 60 447 13 87 7.670 1.111 1230 840 720 120 260 100 160 281 57 43 3.10 3111150 290 180 110 so 40 40 69 58 42 1130 1.411 1630 1220 750 470 260 100 160 334 48 52 5.110 911 400 630 450 180 300 so 220 390 56 44 5150 33,0 270 610 480 130 220 so 140 217 65 35 55,0 ”3 200 930 830 100 190 90 100 271 37 63 M0 151 .260 550 420 130 200 70 130 146 89 11 3360 111 220 540 430 110 190 so 110 224 49 51 3' 111 190 300 190 110 140 60 so 118 68 32 469“ 1111 320 790 600 190 220 90 130 253 51 49 69;“ m .230 1,100 920 180 220 170 so 290 17 83 W 11 400 740 460 280 180 60 120 286 42 58 3.440 1111 320 320 590 230 180 90 90 228 39 61 5.760 1,500 1,110 920 190 460 150 310 462 67 33 .220 :3, 210 860 770 90 210 90 120 203 59 41 6.740 {m 340 350 720 130 320 110 210 353 59 41 6.100 ".0 140 740 640 100 130 90 40 201 20 so 5.070 11 450 1,560 1,190 370 220 140 so 236 34 66 10.430 ‘ 360 1,980 1,660 320 270 230 40 372 11 89 15.933 11132320 1,130 890 240 260 180 so 370 22 78 3,71 111111me NlhOAc extractable K). reable K ’ P114 before Neubauer growth study. a-five "11111le NaBPh. extractable K minus five-minute NH40Ac extractable K). leubauefg‘0w Whaction with NH40Ac before Neubauer grth study. oil determm during Neubam1,.uflfrl‘ieuabuer growth study. mthe 0116113“ Wlng Neubauer growth study. XC m 11‘ none HICHIGQN smTE UNIV. LIBRGRIES llllllllllll ill ”Illlllllllll lllllllllll Willi Willi 31293017103833