INFORMATION TO USERS This m anuscript has been reproduced from the microfilm m aster. U M I films th e text directly from the original or copy subm itted. Thus, som e thesis and dissertation copies are in typew riter face, while others may be from any type o f com puter printer. The quality o f th is reproductio n is dependent upon th e quality o f th e copy su b m itted . B ro k en o r indistinct print, colored o r p o o r quality illustrations and photographs, print bleedthrough, substandard margins, and im proper alignm ent can adversely affect reproduction. In th e unlikely event th a t th e a u th o r did n o t send U M I a co m p lete m anuscript and th ere are missing pages, these will be noted. Also, if unauthorized copyright m aterial had to be rem oved, a note will indicate the deletion. O versize m a terials (e.g., m aps, draw ings, ch arts) a re rep ro d u c ed by sectioning th e original, beginning a t the u p p e r left-h an d c o rn e r and continuing from left to right in equal sections with small overlaps. E ach o rig in a l is a lso p h o to g ra p h e d in o n e e x p o su re a n d is in c lu d e d in reduced form at the back of the book. Photographs included in the original m anuscript have b een reproduced xerographically in this copy. H ig h er quality 6" x 9" black an d w hite photographic prints are available for any photographs o r illustrations appearing in this copy for a n additional charge. C ontact U M I directly to order. U n iv e r s ity M icrofilm s International A Bell & H o w ell Inform ation C o m p a n y 3 0 0 N o rth Z e e b R o a d . A n n Arbor. Ml 4 8 1 0 6 - 1 3 4 6 U S A 3 1 3 .7 6 1 -4 7 0 0 800 521-0600 O r d e r N u m b e r 9129459 M in e r a lo g y o f fin e -lo a m y a n d s a n d y h y d r o s e q u e n c e s in M ic h ig a n H aile-m ariam , Shawel, P h.D . Michigan State University, 1991 UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106 MINERALOGY OF FINE-LOAM Y AND SANDY HYDROSEQUENCES IN MICHIGAN By Shawel Haile-mariam A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for ilic ucgice of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 1990 ABSTRACT MINERALOGY OF FINE-LOAMY AND SANDY HYDROSEQUENCES IN MICHIGAN By Shawel Haile-mariam Two texturally different hydrosequences (fine-loamy and sandy) in replicate were selected in forested land to investigate the mineralogy of the clay, silt, and sand fractions in relation to drainage classes. The sand fraction of all soils in the hydrosequences contained quart?., K-feldspars, plagioclase, sericite, chert, pyroxene and amphibole. Weatherable minerals constitute <5% of the total minerals present. Dolomite is present in the parent material of the fine-loamy hydrosequences except in the Parkhill pedon. Quartz is the predominant mineral followed by feldspars in the silt fractions. Natural drainage does not appear to affect distribution of sand and silt size minerals. Mica, vermiculite, chlorite, kaolinite, quartz, and feldspars constitute the clay minerals present in the parent material of all the hydrosequences. Smectite is present in the fine clay of the Brookston pedon and Spodosol hydrosequences. In the fine-loamy hydrosequences, weathering of dioctahedral mica/sericite to smectite through interstratified vermiculite-smectite is the dominant trend in the Brookston pedon, where as direct transformation to vermiculite is the prevalent weathering mechanism in the other drainage classes. Except for the smectite in the Brookston pedon, the fine-loamy soils in this study do not have significant differences in the sand, silt, and clay mineralogies with respect to natural drainage. In the sandy hydrosequences, the intensity of smectite peak and the presence and absence of vermiculite in the E horizons, indicate increasing weathering intensities from poorly to well drained soils. The mineral distribution in these hydrosequences suggest a weathering sequence of dioctahedral mica/sericite -» vermiculite —» smectite with or without mixed-layer formation. Structural formulas calculated indicates 0.47 and 0.51 molc per half unit cell for the Spodosol (E) and Mollisol (Btg) horizons, respectively. The mean layer charge determined by the alkyl ammonium method is 0.41 and 0.43 mole per half unit cell for the E and Btg fine-clays, respectively. The paraffin type configuration indicates the presence of unweathered mica cores in these samples. The re-expansion property suggest the presence of hetrogeneous layer charge and a layer charge between 0.46 and 0.57 per half unit cell. The Greene-Kelly test, ie-expansion property, and 060 x-ray powder diffraction indicates the smectite is dioctahedral of beidellitic type. ACKNOWLEDGEMENTS The author wishes to express his sincere appreciation to Dr. Delbert Mokma for his guidance, constructive criticism and above all his patience. The author also extends his appreciation to his other comittee members: Dr. James Crum for the graduate assistantship he provided for two terms; Dr. Max M. Mortland for his constant support, intellectual advice and financial help he provided for laboratory supplies; and Dr. Michael A.Velbel for his helpful comments in revewing the proposal and dissertation, and also for the advice and assistance he provided during the time of making thin sections and identification of sand mineralogy. Dr. Sharon Anderson provided invaluable assistance in the often strenous computer work. Dr. Sharon Anderson and Dr. William Jaynes have provided intellectual stimulation which is necessary for an education. Finally, the author acknowledges Dr. Steve Spreacher for his creative ideas and friendship, and also for collecting field data, help with excavation and description of soil pits in Chippewa and Cheboygan Counties in Michigan. iv TABLE OF CONTENTS List of Tables vi List of Figures vii INTRODUCTION 1 LITERATURE REVIEW 3 Mineral weathering sequence Smectite formation in acidic environments Mechanism of mica weathering Clay mineral weathering in relation to drainage List of references CHAPTER 1. Sand, silt, and clay mineralogy of two fine-loamy hydrosequences in Michigan. Abstract Introduction Materials and Methods Results and Discussion Summary and Conclusions Lisi of references 3 6 7 8 11 16 16 17 18 21 51 61 CHAPTER 2. Mineralogy of two Sandy hydrosequences in Michigan. Abstract Introduction Materials and Methods Results and Discussion Conclusions List of references 65 65 66 67 69 86 95 CHAPTER 3. Characterization of smectite in a Spodosol and a Mollisol. Abstract Introduction Materials and Methods 98 99 100 101 V Results and Discussion List of references 102 112 General Conclusions 116 Apendix 1. Pedon descriptions 118 Vi LIST OF TABLES TABLE 1.1. Selected chemical properties of soils in hydrosequence 1 32 TABLE 1.2. Selected chemical properties of soils in hydrosequence 2 33 TABLE 1.3. Particle size analysis and bulk density of soils in hydrosequence 1 35 TABLE 1.4. Particle size analysis and bulk density of soils in hydrosequence 2 36 TABLE 1.5. Mineralogy of the fine sand fraction 39 TABLE 1.6. Mineral ratios for the fine sand fraction 40 TABLE 1.7. Mineralogy of coarse and fine silt fractions 45 TABLE 1.8. Clay mineralogy of total clay ffaction(<2pm) 50 TABLE 2.1. Soils of Hydrosequences 68 TABLE 2.2. Location and classification of soils in the hydrosequences 68 TABLE 2.3. Particle size analysis of soils in Chippewa hydrosequence 75 TABLE 2.4. Particle size analysis of soils in Cheboygan hydrosequence 76 TABLE 2.5. Selected chemical properties of soils in Chippewa hydrosequence 78 TABLE 2.6. Selected chemical properties of soils in Cheboygan hydrosequence 79 TABLE 2.7. Mineralogy of the sand fraction 82 TABLE 2.8. Mineralogy of coarse and fine silt fractions 83 FT' ATN* 84 A I r t D L C Z .y . Ciay mineralogy of total clay fraction(<2pm) TABLE 3.1. Elemental analysis of the Na+ saturated fine clay fraction (<0.2pm) from the E horizon of a Spodosol and Btg horizon of a Mollisol v ii 110 LIST OF FIGURES Figure 1.1. Location of the study areas in Michigan 20 Figure 1.2. Average monthly depths to water tables and average monthly 23 precipitation of hydrosequence 1, in Clinton county Figure 1.3. Average monthly depths to water tables and average monthly 24 precipitations: of hydrosequence 2, in Ionia county Figure 1.4. Percent total time saturated at given depth in poorly drained pedons 25 Figure 1.5. Average monthly depths to water tables in poorly drained pedons 26 Figure 1.6. Base saturation and pH in poorly drained pedons 27 Figure 1.7. Percent total time saturated at given depth in hydrosequence 1 28 Figure 1.8. Percent total time saturated at given depth in hydrosequence 2 29 Figure 1.9. Base saturation and pH (0. IN KC1) in pedons of hydrosequence 1 30 Figure 1.10. Base saturation and pH (0.1 N KC1) in pedons hydrosequence 2 31 Figure 1.11. Quartz to plagioclase ratios. 41 Figure 1.12. Quartz to K-feldspar ratios 42 Figure 1.13. Quartz to feldspar ratios 43 Figure 1.14. X-ray difffactograms of clay fraction (<2um) of pedon M l 52 Figure 1.15. X-ray diffractograms of clay fraction (<2um) of pedon C l 53 Figure 1.16. X-ray diffractograms of clay fraction (<2um) of pedon PI 54 Figure 1.17. X-ray diffractograms of clay fraction (<2um) of pedon M2 55 Figure 1.18. X-ray diffractograms of clay fraction (<2um) of pedon C2 56 Figure 1.19. X-ray diffractograms of clay fraction (<2um) of pedon B2 57 Figure 1.20 X-ray diffractograms of pedon B2, fine, medium, and coarse clay 58 fractions after Mg saturated and glycerol solvated. Figure 1.21 X-ray difffactogram of 060 diffraction peak of the total clay fraction 59 of the Bt horizon Figure 2.1. Average monthly depths to water tables and average monthly 70 precipitation in Chippewa hydrosequence Figure 2.2. Average monthly depths to water tables and average monthly precipitation in Cheboygan hydrosequence v iii 71 Figure 2.3. Percent total time saturated at given depth in Chippewa 72 hydrosequence Figure 2.4. Percent total time saturated at given depth in Cheboygan 73 hydrosequence. Figure 2.5. X-ray diffractograms of clay fraction (<2pm) of the Liminga 87 pedon Figure 2.6. X-ray diffractograms of clay fraction (<2pm) of the Wainola pedon. 88 Figure 2.7. X-ray diffractograms of clay fraction (<2pm) of the Kinross pedon 89 in Chippewa hydrosequence Figure 2.8. X-ray diffractograms of clay fraction (<2pm) of the Rubicon pedon 90 Figure 2.9. X-ray diffractograms of clay fraction (<2pm) of AuGres pedon 91 Figure 2.10. X-ray diffractograms of clay fraction (<2pm) of the Kinross pedon 92 in Cheboygan hydrosequence Figure 2.11. X-ray diffractograms of fine clay fraction (<0.2pm) of the E horizons 93 in Chippewa hydrosequence. Figure 2.12. X-ray diffractograms of fine clay fraction (<0.2pm) of the E horizons 94 in Cheboygan hydrosequence Figure 3.1. X-ray diffractograms of the Spodosol E and Mollisol Btg horizons 104 (<0.2pm; treatments specified at right) Figure 3.2. XRD patterns of the fine clays from E and Btg horizons of Spodosol 106 and Mollisol respectively (treatments are specified at right) Figure 3.3. XRD of the fine clays from E and Btg horizons of the Spodosol and 107 Mollisol respectively. Re-expansion properties of clays (treatments specified at right) Figure 3.4. X-ray patterns of the <0.2pm soil clay separated from the E and Btg 108 horizons of the Spodosol and Mollisol pedons and treated with various alkyammonium hydrochlorides (carbon chain length specified at right) Figure 3.5. Basal plane spacings of alkylammonium complexes with the E and Btg fine clay Spodosol and Mollisol, respectively, and the corresponding layer charge distribution ix 109 INTRODUCTION The study of weathering of primary and secondary minerals in soil hydrosequences in relation to the soil forming factors (parent material, climate, vegetation, and time) provides valuable information as to soil mineral distribution, kind and amount of secondary clay minerals formed and its variability with depth. The study of soil sequences will increase our understanding of soil distribution in a landscape and hence enable us to predict soil occurences in different areas where similar soil forming factors are present. Clay minerals, which are the most important products of weathering, are instrumental in controlling the physical and chemical characteristics of soil. Water permeability, aeration, profile development, and cation exchange capacity of soils depend on the amount and kind of clay minerals present. These properties of clay minerals affect the growth of plants and soil fertilization. The kind and amount of primary and secondary minerals present in sand, silt, and clay size affects the general nutrient reserve status of soils and also soil fertility studies. In Soil Taxonomy, sand silt and clay mineralogy data are used in soil classification at the family level and it is also used together with other features, to differentiate taxa at other categorical levels. Theoretically, internal drainage of soils is one of the major intensity factors controlling the rate cf chemical weauiciuig reactions. However, studies of the influence of drainage on weathering in soil hydrosequences and its relation to the formation of different clay minerals have been few and contradictory. Detailed study of sand, silt, and clay mineralogy of soils from the same parent material in a topographic hydrosequence occurring in a young glacial landscape and their relationship to the degree of internal drainage will increase our understanding of soil mineral distribution and genesis. The objectves of this study were to: (1) determine and compare the mineralogy of the total clay, fine silt, coarse silt, and dominant sand fraction with depth in soils of fine-loamy and sandy hydrosequences. (2) To characterize smectite present in the fine-loamy and sandy hydrosequences. 1 2 In order to accomplish these objectives, two texturally different hydrosequences (fine-loamy and sandy) in replicate were selected. These topographic hydrosequences were selected in forested land to minimize disturbances due to cultivation, artificial drainage and accelerated erosion. Sampling sites (profiles) were selected according to drainage variation in each hydrosequence. LITERATURE REVIEW Mineral weathering sequences Different weathering sequences of minerals have been postulated for different categories of minerals. Weathering sequences have been worked out for coarse-grained minerals, with heavy specific gravity, colloidal minerals and combinations of these categories. A weathering sequence of coarse-grained minerals was established by Goldich (1938). Goldich's double sequence of common minerals in the order of their decreasing vulnerabilities downwards was: O livine Ca Plagioclase 4 4 A ugite CaJ^a Plagioclase 4 4 H ornblende Na, Ca Plagioclase 4 4 Biotite Na Plagioclase 4 4 Potash feldspars X Muscovite 4 Quartz Graham (1949) also found experimentally the same kind of vulnerability sequence as observed in nature by Goldich (1938). This sequence is a generalization and may have many variations depending on weathering conditions. For instance, the weathering of plagioclase was thought to be much faster and stronger than that of alkaline feldspars in temperate climates (Millot, 1970) and tropical humid climates (Leneuf, 1959). However, in arid areas Rondeau (1958, see Millot, 1970) found strong weathering of alkaline feldspars and freshness of plagioclases. This difference was attributed to the composition of intergranular solutions which were not the same in arid areas and in humid areas, and which inverted the rates of hydrolysis. Various granular rocks reacted to different climates in different ways. Alkaline rocks were more vulnerable under humid climates and as a result acid rocks dominate the topography. However, under an arid climate the opposite holds true. In the weathering sequence of heavy minerals, the less stable minerals of high specific gravity were found to occur in decreasing amounts in rocks of increasing age. In rocks older than Pleistocene olivine was absent (Pettijohn, 1941). Augite was rare or absent in rocks of pre-Cambrian age but became more common in younger rocks. Pettijohn's weathering sequence with most stable listed first was as follows: (-3) Anatase (4) Garnet (-2) Muscovite (3) Biotite (-1) Rutile (1) Zircon (6) Apatite (7) Ilm enite (2) Tourmaline (3) Monazite (8) Magnetite (10) Kyanite (U )E p id o te (12) Hornblende (13) Andlusite (14) Topaz (16) Zoisite (17) Augite (18) Sillimanite (19) Hypersthene (20) Diopside (9) Staurolite (15) Sphene (21) Actinolite (22) Olivine The negative numbers on the first minerals were to indicate their tendency of formation rather than disappearance during long periods of burial. Weyl (see Jackson and Sherman, 1952) suggested grouping heavy minerals in four categories of stability: Extremelv Unstable Olivine Hornblende Augite Sliehtlv Stable Garnet Epidote SUM* Staurolite Kyanite Sillimanite Andalusite Magnetite Verv Stable Tourmaline Ziricon Rutile Titanite The relative stability of heavy minerals was employed to establish the age of the rock formation from which soils were derived (Buckhannan and Ham, 1942). Allen (1948) found one to five fold decreases in the amounts of heavy minerals towards the top of weathering profile. Graham (1949) utilized the presence of plagioclase feldspars as an index to determine the stage attained in soil weathering. Resistant species increased in relative abundance while less resistant ones decreased in abundance or disappeared completely (Pettijohn, 1941; Dryden and Dryden, 1946). A change in relative abundance of the minerals was used to measure their resistance to weathering. A weathering sequence of clay-size materials in soils and sediments consisted of 13 stages according to Jackson et al. (1948). 1/ Gypsum (also halite etc.) 2/ Calcite (also dolomite, aragonite, etc.) 3/ Olivine-homblende (also diopside, etc.) 4/ Biotite (also glauconite, chlorite, antigorite, nontronite, etc.) 5/ Albite (also anorthite, microcline, stilbite, etc.) 6/ Quartz (also cristobalite, etc.) 5 7/ Illite (also muscovite, sericite, etc.) 8/ Hydrous mica-intermediates 9/ Montmorillonite (also beidellite, etc.) 10/Kaolinite (also halloyite, etc.) 11/Gibbsite (also boehmite, etc.) 12/Hematite (also goethite, limonite, etc.) 13/Antase (also rutile, ilmenite, corundum, etc.) The authors also suggested one mineral may be the parent material of others in successive stages of the weathering sequence and the weathering reactions were reversible in nature. The weathering sequence proposed by Jackson et al. (1948) and Jackson and Sherman (1952) involved the transition: Mica -» Mica intermediate —» Aluminous chlorite-* Kaolinite —» Gibbsite This weathering trend was supported in another study by White et al. (1957). Mica weathering to kaolinite through vermiculite and Al-vermiculite was also observed in humid tropical regions (Tardy, 1971). Under more extreme weathering conditions mica can be converted to kaolinite without a proceeding sequence of mineralogical changes involving vermiculization (Wilson, 1974). Rice et al. (1985) proposed a weathering sequence of biotite to hydroxy-interlayered vermiculite through vermiculite and the formation of smectite from ferromagnesian minerals in soils with metagabbro and gabbro parent materials. Smectite formation from chlorite intermediates in soils developed from metagabbro parent material was also indicated. The formation of smectite from ferromagnesian minerals and from chlorite intermediates implies two kinds of smectite: transformation smectite from chlorite and neogenetic smectite formed from ions supplied by other ferromagnesian minerals (Borchardt, 1977). Transformation of micas and chlorite under beech forest and tussock grassland in New Zealand soils was observed by Churchman (1980). Alteration of chlorite follows the sequence chlorite —» interlayered hydrous mica -» chlorite-swelling chlorite —» chloritevermiculite, with increased weathering leading to destruction of the chloritic layers. The weathering of micas (muscovite) apparently occurred by either one of two distinct major pathways, both leading to beidellite: (1) mica -» mica - vermiculite -* mica-beidellite -* beidellite and (2) mica -* vermiculite -* beidellite with pedogenic chlorite as a possible intermediate which might be formed from either mica or vermiculite but which gives vermiculite on de-alumination at pH less than 4.5. Both pathways appeared to be followed concurrently in the weathering of micas in soils under beech forest, while the first pathway was the sole course followed in soils in tussock grasslands. This study suggested that the sensitivity of the transformation from pedogenic chlorite to vermiculite at this pH is consistent with the suggestion of workers (Gjems, 1970; Kapoor, 1973) that 4.5 is the critical pH for the de-alumination of pedogenic chlorite. Droste and Tharin (1958) found illite weathering through an intermediate stage of illite-smectite mixed layers to smectite in soils derived from Illinoian aged tills. Bhattacharya (1962) proposed a similar weathering sequence for dioctahedral illites but trioctahedral illites weathered to smectite through intermediate stages of illite-vermiculite and vermiculite-smectite mixed layers. Dioctahedral illites are altered to smectite by exchange of potassium from interlayer position with hydronium ions. Trioctahedral illites alter in the same way but go through intermediate stages. The minerals that result from the weathering of micas and/or chlorites depend on the environment in which the reactions occur and the time elapsed. Smectite formation in acidic enviomments Many soil mineralogists believe that smectite is unstable in acidic enviomments and restrict its formation mainly to basic environments. However, others (Brown and Jackson, 1958; Gjems, 1960; Franzmeier and Whiteside, 1963; McKeague, 1965; Ross and Mortland, 1966; Brydon et al., 1968; Malcolm et al., 1969; Coen and Arnold, 1972; Kapoor, 1972; Douglas, 1967,1987) among others, have reported the weathering of mica and chlorite through vermiculite, to smectite in eluvial or albic horizons of Spodosols. Ross and Mortland (1966) reported beidellite to be the smectite variety in their study. Gjems (1970) postulated the formation of smectite from hyuroxy-aiuminium interlayered smectite, and probably also from illite in two Yugoslavian Spodosols. The strong acidity of the eluvial horizons promoted the dealumination of hydroxy-interlayered species. Gjems believed the process was further enhanced by the formation of aluminiumhumus complexes and subsequent translocation from the eluvial horizon to the B horizon. In Scandinavian Spodosols a weathering sequence of biotite —> mixed layer mineral and vermiculite (C and B horizons) -» hydrobiotite —» vermiculite -> smectite (in the eluvial horizon) was found (Kapoors, 1972). The initiating mechanism for the change from biotite to smectite under humid-temperate climates was attributed to the oxidation of Fe++ in the lattice with subsequent removal of ions both from the octahedral and interlayer positions. Coen and Arnold (1972) attributed the presence of smectite in an eluvial horizon to eolian addition. Chlorite added to the soil by dust weathered to amorphous material and 7 translocated to the spodic horizon. Mica weathered to smectite in the eluvial horizon to vermiculite in the Bh horizon, and to chloritized vermiculite in the Bs horizon. In well drained Scottish soils developed from glacial drift, a weathering sequence of biotite -» hydrobiotite (C horizon) vermiculite-chlorite (B horizon) -» vermiculite (A horizon) was observed (Wilson, 1970). In the more acid surface horizon, vermiculitechlorite interlayers tend to break down yielding more vermiculite products. The absence of smectite in soils derived from biotite-homblende rock in the above weathering sequence might suggest that the changes mica —» mica-vermiculite -» mica-smectite probably follows a time sequence as well as a precipitation sequence (Churchman, 1980). Malcolm et al. (1969) and Douglas (1987) have attributed the formation of smectite in Spodosols to high soil organic matter and low pH levels. At low pH the aluminium is complexed by organic matter and becomes unavailable for the formation of aluminium polymers. At pH below 4.5 the aluminium polymer is thermodynamically unstable and protonation of the interlayer aluminium polymer would result in the formation of exchangeable aluminium species, subject to leaching, and an expansible aluminosilicate. Ross and Mortland (1966) and Gjems (1970) have considered the presence of smectite or smectitic layers as an indication of an advanced stage of weathering in soils. In spite of many reports of the presence of smectite in Spodosols of U.S.A. and Europe, the present state of knowledge regarding its formation in acidic environment remains controversial. Most believe it is a low charge vermiculite with smectite characteristics (glycerol expansion). Mechanism of mica weathering The weathering of mica in general was thought to proceed by at least two processes (Sparks and Huang, 1985): "simple transformation" of the K bearing mica to an expandable 2:1 layer silicates by exchanging the K ions from the interlayer cation sites with hydrated exchangeable cations, and dissolution of the mica followed by the precipitation of weathering products. The relative importance of these two processes depended on the stability of the mica and the nature of the soil environment. In the simple transformation reaction a considerable portion of the primary mica structure (the 2:1 layer) was retained intact as a transformation product (Velbel, 1984; Fanning and Keramidas, 1977). This was understood mainly as the loss of interlayer K which was compensated for by replacement with hydrated exchangeable cations. In this way expandable 2:1 clay minerals such as vermiculite and smectite were formed. This simple transformation took place in two ways: (1) by edge weathering (Mortland, 1958; Scott and Smith, 1967) and (2) by layer weathering or weathering along preferential planes in the mica strucure (Fanning and Keramidas, 1977). Edge weathering resulted when the mica grains were initially opened along cleavage traces and fractures. The expanded interlayers were not continious throughout the grain but terminated in wedges toward the interior of the crystal. The core of the grain was initially unaffected. When K was released from many layers simultaneously the resultant particle had frayed, curled and expanded edges surrounding an unaltered mica core. It was by this process that intergrades of mica and expandable 2:1 phyllosilicates such as hydrobiotite and mica-smectite intergrades were thought to form. In layer weathering the interlayers were opened or expanded throughout the crystal. If all of the interlayers were not expanded then various types of mixed layer or interstratified minerals were formed (e.g. biotite-vermiculite, biotite-smectite; Bisdom et al., 1982). If all the interlayers were expanded then the biotite was completely transformed into an expandable 2:1 clay mineral. If the layer charge of the transformed mineral was greater than 1.2 per unit then the secondary mineral was a vermiculite and if it was less than 1.2 the weathering product was a smectite (Bisdom et al., 1982). Clav mineral weathering in relation to drainage Topography influences infiltration rate, run off and run on, temperature, and evapotranspiration. Each of these subfactors, in turn, influences soil development through such processes as leaching, weathering, erosion, sedimentation, and biosphere ucveiupiueiu. Tne relative water tabie position can have an effect on the magnitude of weathering and thus on the formation of different clay minerals in the soil. The rate and nature of chemical weathering were found to be influenced by the pH and the time of contact between rock and water (Drever, 1982). Weathering intensity increased as acidity and oxidation increased (Jackson et al., 1948). Well drained soils had short contact times and tended to have kaolinitic soils, whereas poorly drained soils had longer contact times and tended to have smectite as the soil clay mineral (Drever, 1982). The study of the influences of parent material and topography on soil genesis in the midwest by White et al. (1960) indicated the dominant clay mineral in the glacial tills and fine textured sedimentary rocks was mica. The influence of topography and drainage was reflected in the proportion and distribution of the vermiculite and smectite resulting from the weathering of mica. Weathering of mica to vermiculite in well drained and moderately well drained soils and alteration to smectite in poorly drained soils has been reported (Thoip et al., 1959; Smith and Wilding, 1972; McKeague et al., 1973; Ritchie et al., 1974; Schouten, 1974; Smeck et al., 1981). Differences in clay mineralogy on basic rocks were less obvious, but on the poorly drained sites the intensity of the smectite reflections were much stronger than on the well drained sites. In recent weathering products no differences in clay mineralogical composition due to differences in profile drainage status were observed (Schouten, 1974). Smeck et al.(1981) pointed out that the somewhat poorly drained soils were subjected to more severe weathering and leaching than the better drained members. The formation of smectite in the fine clays through weathering of vermiculite and or a synthesis reaction was more significant under somewhat poorly drained soils (Gleysolic) than under the well drained soils (Chemozemic). Hlitization was apparent in the fine clay fraction of the solum of the well drained soils (Huang and Lee, 1969). Bid well and Page (1951) and Martin and Russel (1952) found little difference in type of clay minerals produced between different members of a toposequence as a result of internal drainage differences. There was slight evidence of more smectite in the B horizons of the poorly drained members. Otherwise the clays had not sufficiently weathered to alter the original clay inherited from the parent materials. Johnson and Jefferies (1957) in the Allenwood catena of Pennsylvania and Sawhney (1960) in Connecticut observed a weathering sequence of mica -> mica intermediate-* vermiculite —> chlorite like material. The weathering sequence in both poorly and well drained members of the Allenwood catena was the same, however, the degree of expression of the weathering sequences was apparently controlled by drainage conditions. The extent of weathering of mica was more pronounced in ilic well uiabieu member. The poor drainage condition inhibited the transformation of vermiculite in the lower horizons of the Pennsylvania catena while it inhibited the formation of the Al interlayers in the Connecticut catena. The restricted formation of Al interlayers in these soils was probably due to less weathering associated with poor drainage. Biswas and Das (1960) and Iniguez and Scoppa (1973) found smectite contents increased and illite contents decreased in descending from well drained to the poorly drained soils. Kaolinite was only present in the well drained soils. Millot (1970), and Tardy (1971) have indicated kaolinite was the dominant weathering product in rapid leaching environments, whereas smectite was the dominant clay mineral in impervious soils even with high amounts of precipitation. 10 Drainage influenced the mica weathering product in several French soils (Millot and Gamez, 1963). Mica weathered to vermiculite through a mica-vermiculite intermediate stage in soils with open drainage, whereas in poorly drained soils it weathered to smectite. Having similar soil forming factors (time, climate, vegetation, parent material) in a hydrosequence, topography produces local differences (internal drainage) among soils and as a result the soil profile development, the type and the amount of clay minerals forming in a hydrosequence varies accordingly. LIST OF REFERENCES Alexiades, C.A. and M.L.Jackson. 1966. Quantitative clay mineralogical analysis of soils and sediments. Clays Clay Miner 14:35-52. Allen, V.T. 1948. Weathering and heavy minerals. Jour. Sed. Petrology 18:38-42. Bhattacharya, N. 1962. Weathering of glacial tills in Indiana. I. Clay minerals. Geol. Soc. Amer. Bull. 73:1007-1020. Bidwell, O.W. and J.B.Page. 1951. The effects of weatheringon the clay mineral composition of soils in the Miami catena. Soil Sci. Soc. Am. Proc. 15:314318. Bisdom, E.B., G.Stoops, J.Delvigne, P.Curmi, and H.Altemuller. 1982. Micromorphology of weathering biotite and its secondary products. Pedologie 2:225 252. Biswas, T.D and S.C.Das. 1960. Variability of clay minerals in associated soils in toposequence. VII Int. Cong. Soil Sci. p.35. Borchardt, G.A. 1977. Montmorillonite and other smectite minerals, pp. 293-330. In J.B.Dixon and S.B.Weed (ed.) Minerals in soil environments Soil Sci. Soc. Am., Madison, WI. Brown, B.E. and M.L. Jackson. 1958. Clay mineral distribution in the Hiawatha sand soils of Northern Wisconsin. Clays Clay Miner. 566:213-226. Brydon, J.E., H.Kodoma and G.J.Ross. 1968. Mineralogy and weathering of the clays in orthic podozols and other podozolic soils in Canada. 9th Intl. Congress of Soil C^.1 U W i.i i O Jl d . * > /e\ . a 1 Buckhannan, W.H. and W.E.Ham. 1942. Preliminary investigations of heavy mineral criteria as an aid in the identification of certain soils in Oklahoma. Soil Sci. Soc. Amer. Proc. 6: 63-67. Chapman, S.L., J.K.Syers, and M.L Jackson, 1968. Quantitative determination of Quartz in soils, sediments and rocks by pyrosulfate fusion and hydrofluosilicic acid treatment. Soil Sci. 107:348-355. Churchman, G.J. 1980. Clay minerals formed from micas and chlorite in some New Zealand soils. Clay Minerals. 15:59-75. Coen, G.M. and R.W.Amold. 1972. Clay mineral genesis of some New York Spodosols. Soil Sci. Soc. Amer. Proc. 36:342-350. 12 Cremeens, D.L.1983. Argillic horizon formation in the soils of a hydrosequence. MSc. Thesis, Michigan State University. Douglas, L.A. 1977. Vermiculites. p.259-292. In J.B.Dixon and S.B.Weed (ed.) Minerals in soil enviomments. Soil Sci. Soc. of Am., Madison, Wi. Drever, J.1 .1982. The geochemistry of natural waters. By Prentice-Hall, New Jersy. pp.386. Droste, J.B. and J.C. Tharin. 1958. Alteration of clay minerals in Ulionian till by weathering. Geol. Soc. Amer. Bull. 69: 61-67. Dryden, L. and C.Dryden. 1946. Comparative rates of weathering of some common heavy minerals. Jour. Sed. Petrology 16: 91-96. Eberl, Dennis D., Srodon, Jan, Lee, Mingchou, Nadeau, P.H., and Northrop, H.Roy. 1987. Sericite from the Silverton caldra, Colorado:Correlation among structure, composition, origin, and particle thickness. American Mineralogist, 72:914-934. Ellis, B.G. and M.M.Mortland. 1959. Rate of potassium release from fixed and native forms. Soil Sci. Soc. Am. Proc. 23:451-453. Fanning, D.S., and V.Z.Keramidas. 1977. Micas, p. 195-258. In J.B.Dixon and S.B.Weed (ed.) Minerals in soil environments. Soil Sci. Soc. America, Madison, WI. Franzmeier, D.P. and E.P.Whiteside. 1963. A chronosequence of podozols in Northern Michigan. Mich. St. Univ. Agr. Exp. Sta. Quar. Bull. 46:1-57. Galehouse, S.S. 1971. Point counting. P.385-407. In R.E.Carver (ed.) Procedures in sedimentary petrology. Wiley-Interscience, New York. Gilkes, R.J. and A.Suddhiprakam. 1979. Biotite alteration in deeply weathered granite. II. Oriented growth of secondary minerals. Clays Clay Miner. 27:361-367. Gjems, 0 . 1960. Some notes on clay minerals in podzol profiles in Fennoscandia. Clay Min. Bull. 4:208-211. Gjems, 0 . 1970. Mineralogical composition and pedogenic weathering of the clay fraction in podzol soil profiles in Zalesine, Yugoslavia. Soil Sci. 110:237-243. Goldich, S.S. 1938. A study in rock weathering. Jour. Geology, 46:17-58. Graham, E.R. 1949. The plagioclase feldspars as an index to soil weathering. Soil Sci. Soc. Am. Proc. 14: 300-302. 13 Huang, P.M., and S.Y.Lee. 1969. Effect of drainage onweathering transformations of mineral colloids of some Canadian prairie soils. Proc.Int. Clay Conf.(Tokyo,1969) 1:541-551. Iniguez, A.M. and C.O.Scoppa 1973. Evolution of clay minerals in an hydromorphic soil of the Pampean region of Argentina. In: Pseudogley and Gley (E.Schlichting and U.Schwertmann, eds.). Transactions of Commissions V. and VI. of the Int. Soc. Soil Sci. pp. 139-143. Ismail, F.T. 1969. Role of ferrous iron oxidation in the alteration of biotite and its effect on the type of clay minerals formed in soils of arid and humid regions. Am. Miner. 54:1460-1466. Jackson, M.L., S.A.Tyler, A.L. Willis, G.A.Bourbeau and R.P.Pennington. 1948. Weathering sequence of clay minerals in soils and sediments: Part I. Fundamental generalizations. Jour. Phys. and Colloid Chem., 52:1237-1260. Jackson, M.L., Y.Hseung, R.B.Corey, E.J.Evans and R.C.Vanderheuvel. 1952. Weathering sequences of clay size minerals in soil and sediments:Part n. Chemical weathering of layer silicates. Soil Sci. Soc. Am. Proc. 16:3-6. Jackson, M.L and G.D.Sherman. 1952. Chemical weathering of minerals in soils: Adv in Agron. Acadamic Press, New York. 5:219-318. Jackson, M.L. 1967.Soil chemical analysis-advance course (3rd. print). Published by author, Dept, of Soil Science, Univ. of Wisconsin, Madison. Johnson, L.J. and C.D Jeffries. 1957. The effect of drainage on the weathering of clay minerals in the Allenwood catena of Pennsylvania. Soil Sci. Soc. Am. Proc. 21:539-542. Kapoor, B.S. 1972. Weathering of micaceous clays in some Norwegian podzols. Clay O -'JO 'S 'i r M A*AUlWi • / Karathanasis, A.D. and B.F.Hajek. 1982. Revised methods for quantitative determination of minerals in soil clays. Soil Sci. Soc. Am. J. 46:419-425. Kiely, P.V., and M .LJackson. 1964. Selective dissolution of micas from potassium feldspars by sodium pyrosulfate fusion of soils and sediments. Am. Miner. 49:1648-1659. Leneuf, N. 1959. L1alteration des granites Calco-alcalins et des grandiorites en cote-d’ Ivori forestiere et lessols qui en sont derives. These Sci., Paris 210 page In.G. Millot, Geology of Clays, pp.89-91. Mahjoory, R. 1971. Clay mineralogy of some litho and toposequences of soils in Michigan. Ph.D.dissertation, Mich. St. Univ. 14 Mahjoory, R.1975. Clay mineralogy, physical and chemical properties of some soils in arid regions of Iran. Soil Sci. Soc. Am.Proc. 39:1157-1164. Martin, R.T.,and M.B.Russel. 1952. Clay minerals of four Southern New York soils. Soil Sci. 74:267-279. McKeague, J.A. 1965. Properties and genesis of three members of the uplands catena. Can. J. Soil Sci. 45:63-77. McKeague, J.A., J.I.MacDougall, and N.M.Miles. 1973. Micromorphological, Physical, chemical, and mineralogical properties of a catena of soils from Prince Edward Island in relation to their classification and genesis. Can. J. Soil Sci. 58:281-295. Millot, G. and T.Gamez. 1963. Genesis of vermiculite and mixed-layered vermiculite in the evolution of the soils of France. Clays Clay Miner. 10:90-95. Millot, G. 1970. Geology of clays. Springler-Verlag, New York. Mortland, M.M. 1958. Kinetics of potassium release from biotite. Soil Sci. Soc. Am. Proc. 22:503-508. Meunier, A. and B.Velde. 1982. Phengitization, Sericitization and Potassium-Beidellite in a Hydrothermally-altered Granite. Clay Minerals. 17:285-299. Nettleton, W.D., R.E.Nelson, and K.W.Flach. 1973. Formation of mica in surface horizons of dry land soils. Soil Sci. Soc. Am. Proc. 37:473-478. Newman, A.C.D. 1967. Changes in phlogopites during their artificial alteration. Clay Miner. 7:215-227. Pettijohn, F J . 1941. Resistance of heavy minerals and geologic age. J. Geology, 49:610- Rice, Jr., T.J., S.W.Buol.and S.B.Weed. 1985. Soil saprolite profiles derived from mafic rocks in the North Carolina piedmont: I. Chemical, morphological, and mineralogical characteristics and transformations. Soil Sci. Soc. Am. J. 49:171178. Ritchie, A., L.P.Wilding, G.F.Hall and C.R.Stahnke. 1974. Genetic implications of B horizons in aqualfs of northeastern Ohio. Soil Sci. Soc. Am. Proc. 38:351358. Rolfe, B.N. and C.D.Jeffries. 1953. Mica weathering in three soils in Central New York. Clay Min. Bull. 10:85-93. Ross, G J. and M.M.Mortland. 1966. A soil Beidellite. Soil Sci. Soc. Am. Proc. 30:337-343. 15 Sawhney, B.C. 1960. Weathering and aluminium interlayers in a soil catena:HollisCharlton-Sutton-Leicester. Soil Sci. Soc. Am. Proc. 24:221-226. Schouten, C.J. 1974. The application of a microdensitometer to clay mineralogy in a geomorphological invetigation in Southern France. Catena .1:257-271. Scott, A.D., and S J.Sm ith. 1967. Visible changes in macro mica particles that occur with potassium depletion. Clays Clay Miner. 15:357-373. Smeck, N.E., A.Ritchie, L.P.Wilding, and L.A.Drees. 1981. Clay accumulation in sola of poorly drained soils of Western Ohio. Soil Sci. Soc. Am. Proc. 45:95102. Smith, H. and L.P.Wilding. 1972. Genesis of argillic horizons in ochraqualfs derived from fine textured till deposits of northwestern Michigan. Soil Sci. Soc. Am. Proc. 36:808-815. Soil Conservation Service. 1984. Procedure for collecting soil samples and methods of analysis for soil survey. Soil Survey Investigations Report I. SCC-USDA, U.S. Gov. Print. Office, Washington, DC. Sparks, D.L. and P.M.Huang. 1985. Physical chemistry of soil potassium. P. 201276. In R.D. Monson (ed.) Potassium in Agriculture. Soil Sci. Soc. Am Madison, WI. Tardy, Y. 1971. Characterization of the principal weathering types by the geochemistry of waters from some European and African crystalline massifs. Chem. Geol. 7:253-271. Thorp, J., J.G.Cady and E.Gamble. 1959. Genesis of Miami silt loam. Soil Sci. Soc. Am. Proc. 23:156-161. Velbel, M.A. 1984. Weathering processes of iuck forming minerals, p. 67-i 1 i. In M.E. Fleet (ed.) Environmental Geochemistry: Mineralogic Association of Canada Short Course Notes, 10:67-111. White, J.L., G.Talvenheimo, M.G. Klages and M.M. Phillippe. 1957. A survey of the mineralogy of Indiana soils. Indiana Acad.Sci. 66:232-241. White, J.L., G.W.Bailey and J.V.Anderson. 1960. The influence of parent material and topography on soil genesis in the Midwest. Research Bull. No. 693, Purdue Univ., Lafayette, Indiana. Wilson, M J. 1970. A study of weathering in a soil derived froma biotite-homblende rock. Clay Minerals. 8:291-302. 16 CHAPTER 1 SAND, SILT, AND CLAY MINERALOGY OF TWO MICHIGAN FINE-LOAM Y HYDROSEQUENCES ABSTRACT This study was conducted to determine the effect of natural drainage on sand, silt and clay mineral distributions in two fine-loamy hydrosequences. Piezometers were installed to monitor differences in water table levels. The dominant sand fraction was used for identification of minerals by petrographic microscope. X-ray diffiractograms of the silt and clay fractions were obtained to determine the distribution of minerals. The predominant sand minerals are quartz, K-feldspars, plagioclase, sericite and dolomite. The silt fraction is dominantly quartz with small amounts of feldspars and dolomites. The clay fraction is composed of mica, vermiculite, chlorite, kaolinite, smectite, quartz and feldspars. Weathering of dioctahedral mica/sericite to smectite through interstratified vermiculite-smectite is the dominant trend in the poorly drained Brookston pedon, where as direct transformation to vermiculite is the prevalent weathering mechanism in the other drainage classes. Except for the smectite in the Brookston pedon, the fine-loamy soils in this study do not have significant differences in the sand, silt, and clay mineralogies with respect to natural drainage. 17 INTRODUCTION Few studies have been conducted to evaluate variation in clay minerals due to differences in natural drainage. These studies are contradictory and suggest further study. Bidwell and Page (1950) in Ohio, Martin and Russel (1952) in New York, and Ameman et al. (1958) in Minnesota found little or no variation in clay mineral distributions in their soil hydrosequences. Other studies however, have indicated differential clay mineral distributions with natural drainage, Johnson and Jeffries (1957) in Pennsylvania, Hutcheson et al.(1959) in Kentucky, and Allen and Fanning (1983). Recent studies of fine clays in two Michigan hydrosequences by Cremeens and Mokma (1987) have shown differential weathering intensity and translocation of fine clay minerals. Mineralogical studies of sand and silt fractions are also few. Yassoglou and Whiteside (1960), Cady (1960), and Sidhu and Gilkes (1977) studied individual soil series. Ameman et al. (1958) determined the mineralogical composition of all size fractions in a Mollisol catena. All of these studies found little evidence of alteration of sand and silt sized minerals. The objective of this study was to determine the effect of natural drainage on sand, silt, and clay mineral distribution of two fine-loamy hydrosequences in south-central Michigan. la MATERIAL AND METHODS Two replicated fine-loamy hydrosequences were selected in Clinton and Ionia Counties in Michigan (Figure 1.1). The soils in each hydrosequence occurred in adjacent mapping units, so as to keep the other soil forming factors relatively constant. The hydrosequences occur on forested land where disturbances due to cultivation, artificial drainage and extensive erosion are absent or minimal. The parent material of the Clinton County hydrosequence is loamy glacial till which occurs on a till plain near the Flint moraine of the Lake Border morainic system (Bretz, 1951; Ward, 1979). The Ionia County hydrosequence which also has loamy glacial till, occurs on a till plain approximately 1.5 miles north of the northern tip of the Ionia moraine of the Lake Border morainic system (Bretz, 1951). The hydrosequences are composed of moderately well drained (pedon M l) and well drained (pedon M2) Marlette soils (Glossoboric Hapludalfs; fine-loamy, mixed, mesic), somewhat poorly drained Capac soils (Pedons C l and C2) (Aerie Ochraqualfs; fine-loamy, mixed, mesic), and poorly drained Parkhill soils (Pedon P I) (Mollic Haplaquepts; fine-loamy, mixed, mesic) in the Clinton County hydrosequence and Brookston soils (Pedon B2) (Typic Argiaquolls; fine-loamy, mixed, mesic) in the Ionia County hydrosequence. Pedons were described according to Soil Survey Manual (Soil Survey Staff, 1951) and sampled by major horizons (A, B, and C). Pedon locations are given in the pedon descriptions in Appendix A. Piezometers were installed near each of the sampled pedons to monitor the apparent depth of water table. Water table levels were monitored biweekly for the period 1 January 1984 through 31 December 1987. Particle size distribution was determined by the pipette method (Soil Survey Staff, 1984), after removal of cementing agents according to Jackson (1979). The pH of each sample was determined using a 1:1 soil-solution ratio, both in water and 0.1 N KC1. 19 Sands were separated from the silts and clays of each horizon by sieving and the dominant sand fraction was considered for identification of minerals. The dominant sand fraction was impregnated with a 3M Scotchcast and #3 Resin in a test tube. Thin blocks were cut, lightly polished to make one side flat and mounted on glass slides. The mounted slides were thin sectioned to approximately 30p thickness (Carver, 1970). Quantitative determination of feldspar was made by etching the slides with hydrofluoric acid and staining with sodium cobaltinitrite, barium chloride and rhodizonate reagent as described by Bailey and Stephens (1960). Percentages of the various minerals present were determined by point counting (Galehouse, 1971) of approximately 900 grains per thin section using a standard petrographic microscope. Cation exchange capacity (CEC) and exchangeable bases of each sample were determined using ammonium acetate at pH 7 (Soil Survey Staff, 1984). Calcium carbonate equivalent (CCE) was calculated from the inorganic carbon determined using the method of Bundy and Bremner (1972). Total elemental analysis of clay samples were measured by direct analysis of clay suspensions in water using direct current plasma (DCP) emission spectrophotometry according to Spiers et al.(1983). All X-ray analyses (XRD) were performed with a Philips APD 3720 Automated X-ray Diffractometer system using C uK a radiation at 35KV and 20mA, a theta compensating slit, and diffracted beam monochromator with a graphite crystal. Diffractograms were obtained over a total scan range of 2° 20 to 40° 20 for the Mg saturated clays, coarse and fine silt at room temperature and from 2° 20 to 30° 20 for all other cation and heat treatments. All samples were scanned at a rate of 0.02°/second. Integrated peak intensities were used to semiquantitatively estimate the clay minerals present. Random powder X-ray diffraction of the coarse and fine silt was obtained by placing them in a sample holder. X-ray diffractograms of the clay fractions were obtained from basally oriented specimens by transferring clays from membrane filters to glass slides (Drever, 1973), which had been: (a) Mg-saturated and glycerol solvated, (b) Ksaturated and heated to 300°C or (c) K-saturated and heated to 550°C. Minerals were identified for XRD patterns using diffracting criteria outlined by Whitting and Allardice (1986). 20 CH IPPEW A C H EB O YG A N CLINTON IONIA Fig. 1.1. Location of the study area in Michigan. 21 RESULTS AND DISCUSSION Water Table Depth Having similar soil forming factors (time, climate, vegetation, and parent material) in the study area, topography produces local differences in depth of water table among soils. Differences in the depth of water table can bring variations in weathering, soil development, and type and amount of clay minerals. With this in mind, depths of water table in two hydrosequences were monitored for four years to see the differences within and among hydrosequences. Average monthly depths to water tables and average monthly precipitation in the two hydrosequences are presented in Figures 1.2 and 1.3 . Except for the well drained M2 pedon, all pedons showed significant changes in the average monthly depths to water table during the four years. The high water table was observed during the months of March and April in response to snow melt and spring rains. During the active growing season (May through September), when evapotranspiration and precipitation is high, the water table drops and reaches the lowest at the end of the growing season, during August and September, even though this period is one of high precipitation. On average water tables were deepest during July through September. Water tables start rising during early fall and winter due to minimum evapotranspiration, plant senescence, and moderate precipitation. Fluctuations of water tables seem to be controlled by precipitation and evapotranspiration. Well-drained (M2) and moderately well drained (M l) pedons are characterized by a deeper subsuium zone of saturation. Moderately weii drained M i pedon has a zone of saturation at the base of the solum during the middle of winter and early spring (FebruaryMay) whereas the well-drained M2 pedon was not saturated within the solum at any time of the year. When the soil is dry, dispersion upon wetting is favored (Arnold, 1965 and Daniels et al., 1967). As a consequence, the moisture fronts that result from any rainfall events will be capable of translocating suspended clay particles through the solum without the water table acting as a physical barier. On the other hand, somewhat-poorly drained pedons (Cl and C2) had persistent water tables where the B horizons are saturated throughout the year except for the period of summer and early fall (June-October). In poorly drained pedons (PI and B2) the B horizons are saturated throughout the year except for a brief period during August and September in the B2 pedon and during July through October in the PI pedon. The water table in pedon B2 was higher than that in pedon PI throughout the study period except during the month of September where the water tables 22 were at similar levels. Thus, except for this short period when the soil is dry, the throughflow of moisture fronts were restricted and as a result illuviation of suspended clay particles to the B horizon is limited. Pedons PI and B2 do differ in respect to the extent of leaching, and duration of saturation (Figure 1.4). The B horizons of the Brookston pedon was saturated longer than the Parkhill B horizon (approximately 20% longer than the B horizon of the Parkhill pedon). Both pedons are characterized by the lack of throughflow moisture fronts because of shallow depths to saturation (Figure 1.5). However, they differ by the presence of argillans and smectite in pedon B2 and their absence in pedon P I. Ciemeens and Mokma (1986) attributed the development of the Bt horizon in this pedon to relatively higher pH and presence of smectite. Smectite possibly inherited from the parent material was thought to be responsible for the argillic horizon. According to this study the high hydration condition of smectite in the natural settings, allows for its relatively easy dispersion, and its translocation with the fall rain to the zone of saturation. Base saturation and pH (Table 1.1 and 1.2; Figure 1.6) indicate these pedons are relatively unleached. The pH values of PI and B2 pedons are in the range where smectite is stable. Depths and durations of saturation were consistent with soil drainage classes as determined by profile morphology. Duration of saturation increased with depth within drainage classes and with decreasing drainage at simillar depths (Figure 1.7 and 1.8). Depths and durations of saturation were similar for soils with the same natural drainage class. Patterns of average monthly depths of water tables and durations of saturation were similar to that observed by other researchers (Thorp et al., 1972; and Mackintosh et al. 1978) on similar soils formed on glacial till materials. G - 0 MARLETTE 1 * CAPAC 1 a- a PARKHILL - 90 - -110 Average - 130 - 10 0 -. Mo n t h s Figure 1.2 Average monthly depths to water tables and average monthly precipitation of hydrosequence 1, in Clinton County. (cm) o - o MARLETTE 2 *—* CAPAC 2 a— BROOKSTOf Depth a - 3 0 - e— ©— e— © Average monthly precip (mm) -150 J F M A Months Figure 1.3 Average monthly d ep th s to water t a b le s and a v e r a g e monthly precipitation of h y d r o s e q u e n c e 2, in Ionia County. to J*. - 10 - -30- Depth (cm -50-7090- -110 -130a—a PARKHILL 1 * - * BROOKSTON 2 -150 0 20 40 60 80 100 % Time sa tu r a te d Figure 1.4. P er cen t total tim e saturated at given depth in poorly drained p e d o n s , 1 Jan. 1 9 8 4 — 31 Dec. 1 9 8 7 . - 10 - -30 Depth (cm -50-70-90—1 1 0 — -130-150 PARKHILL 1 BROOKSTON 2 M onths Figure 1.5. Average monthly d e p th s to water t a b le s in poorly drained p e d o n s. 10 - -30-50- Depth -70-90— 110 — -130-150- A—A P1 -170 0 20 40 % base Figure 1.6. 60 80 100 saturation 4 5 6 7 pH ( 0 . 1 N 8 9 KCI) B a s e saturation and pH(0.1N KCI) in poorly drain€;d p ed o n s. 10 -3 0 -50-70-9 0 - a— a 130- * PARKHILL CAPAC 1 Q—© MARLETTE 150 0 20 40 60 80 % Time s a tu r a t e d Figure 1.7. P ercen t total time satura ted at given depth in h y d r o se q u e n c e 1. © -© MARLETTE 2 * —* CAPAC 2 a— a B R 00K ST0N 2 -30 Depth (c E -7 0-90 -1 10 - -130-150 0 20 40 60 80 100 % Time s a t u r a t e d Figure 1.8. Percen t total time saturated at given depth in h y d r o se q u e n c e 2. 1 O -20- IT o -60-80- JZ Cl - 1 0 0 (V Q -120- -140G—G M1 -160- Q—© M1 * C1 A— A P1 -180- 1 0 20 1 1 40 ' 1 60 «- 80 b a s e saturation Figure 1.9. 100 (%) 3 4 5 6 7 8 pH B a s e saturation and pH (0.1N KCi) in p e d o n s of h y d r o se q u e n c e 1. - 20 - -40-60- Depth -80- 100 - 120 - - 140-160A—A B 2 -180 0 20 base 40 60 saturation 100 (%) 1 3 1 4 I 5 I 6 I 7 8 pH Figure 1.1 0. B a s e saturation and pH (0.1 N KCI) in p e d o n s of h y d r o s e q u e n c e 2. 32 Table 1.1 SELECTED CHEMICAL PROPERTIES OF SOILS IN HYDROSEQUENCE 1 DithiauecH ON H a n ts HD KEL Caibon C ta s P q . Etchanseahle Bases Base CHI me^lOOg Sat. K Ca Mg S un Fe Al Organic ■ Ca OO3 „Q M nr mcqflOOB % ta g a ric MARLETTE1 A 0-12 63 54 223 453 03 79 19 101 037 006 - m 12-26 48 36 120 15 02 _* - 038 042 006 - Btl 26-tt 48 35 194 155 03 06 23 30 081 032 - Bt2 4084 76 68 213 04 163 62 223 084 CUO 05 BC 84-108 78 72 158 03 + 44 058 006 22 C l 108-165 80 74 124 03 + 33 053 001 23 a 83 74 109 03 + 66 053 001 25 54 48 383 496 05 136 48 189 062 009 - B/E 22-32 53 39 93 140 03 - 10 13 034 003 - Bll 32-53 53 38 143 383 02 20 32 54 OS 005 - 1654- 58 13 07 07 06 05 04 CAPAC 1 A 0-22 69 04 04 BC 53-97 60 46 123 323 02 08 33 43 068 005 - BC 97-125 66 53 104 394 02 14 25 43 061 005 001 2C 79 73 no 02 + 38 066 001 19 63 53 300 590 06 124 43 173 063 010 - Bgl 23-53 59 50 228 509 04 73 39 116 OAS 007 - Bg2 5391 59 48 168 383 03 29 32 64 089 007 - 125+ Ut> 01 03 PARKHLL1 A C 023 91+ 59 * -= negligible qtanbry +=6eeCaCQ3 46 158 483 04 34 38 76 063 007 03 3.7 26 10 06 33 Table 1.2. Selected chemical properties o f soils in hydrosequence 2. B ase S a tu ra tio n S ta tu s D ithionite dH Horizon E x ch an eeab le B ases 0.1N H20 KCL B ase CEC Sat. meq/lOOg % K Ca Mg Carbon C itrate Ext. Sum Fe Al -meq/lOOg------- Inorganic Organic as C aC 0 3 as O.M % --------------------- M A RLETTE 2 A 0-12 5.1 4.6 12.6 133 0.28 B/E 12-32 5.2 4.1 6.4 3.6 0.23 * B tl 32-48 5.1 3.9 13.4 21.6 0.41 Bt2 48-86 7.2 6.6 153 94.2 BC 86-127 1.4 * 1.7 0.21 0.04 • • 0.8 03 3.8 • 0.23 0.28 0.04 1.5 1.0 2.9 0.61 0.09 • 6.6 14.6 0.69 0.09 0.6 03 0.45 7.5 + 7.9 73 7.9 0.24 23 0.38 0.04 22 0.4 8.0 12 9.9 0.24 + 2.7 0.38 0.04 23 03 0-17 6.0 5.4 20.8 44.2 0.21 7.6 1.4 9.2 0.31 0.05 • 4.0 B/E 17-33 53 4.0 123 92.0 0.26 9.0 2.2 113 0.44 0.05 * 0.6 Bt 33-58 6.4 5.1 16.1 16.8 0.21 0.8 • 03 B te 58-85 77 6A 14.0 55.7 0.22 A BC 85-160 8.0 7.4 6.1 8.0 7.4 5.8 0-25 6.1 5.4 43.7 BA 25-40 7.0 6.1 B tg l 40-58 7.1 6.0 C 127+ C A PA C 2 A C 160+ 1.7 2.7 030 0.08 2 33 n o t iU rt 4< u .'t j rt r\r w.v/v 0.28 4.8 3.4 83 0.32 0.03 3.8 03 0.19 + 2.4 0.32 0.03 22 03 72.1 0.32 26.1 5.1 313 0.16 0.05 18.8 64.4 0.37 8.1 3.6 12.1 0.21 0.07 17.2 63.9 0.36 7.3 33 11.0 0.24 0.07 • 03 0.4 0.4 BRO O K STO N 2 A * • 83 0.8 Btg2 58-95 7.1 6.0 18.7 64.7 0.37 8.1 3.6 12.1 0.29 0.08 • BC 95-170 7.4 6.6 16.9 8 8 .8 0.34 9.2 53 15.0 0.55 0.05 0.6 0.4 C 170+ 7.9 73 83 0.21 + 4.3 0.34 0.05 3.6 03 * = neg lig ib le quantity + = free C a C 0 3 p resent 34 Parent Material Homogeneity The initial step, before any soil development and mineral weathering studies, is to ascertain that the soils formed from relatively similar parent material and are not affected by the action of erosion and deposition. Glacial till is the most heterogenous sediment (Flint, 1971) and as a consequence detection of subtle lithologic discontinuities is difficult. Methods used for differentiating parent materials include soil morphology, micromorphology, particle size distribution with depth, and depth distribution of heavy minerals (ziricon, rutile, etc.) which are resistant to weathering and are considered to be stable. Cremeens and Mokma (1986), modifying Asady and Whiteside's (1982) methods, calculated the ratio of silt plus very fine sand to total sand minus very fine sand for each horizon to identify lithologic discontinuities. The ratio of an overlying horizon was divided by that of the underlying horizon and one is subtracted from the resultant ratio. A value of ± 0.60 was chosen by the investigators to identify lithologic discontinuities. Ratios between these values indicate uniformity whereas greater deviation from these values indicates nonuniformty of the parent material. Particle size data are presented in Tables 1.3 and 1.4. Lithologic discontinuities were found in pedons C l and PI but not in the other pedons. Pedon C l has a lithologic discontiniuity between the BC horizon to the 2C horizon. In pedon PI a discontinuity was identified between the A and Bg horizon. The A horizon could be deposition of erosion products from the surrounding upland soils. Hydrosequence 1, being on the edge of the Flint moraine, may have had more influence from the melting process as the glacier retreated (Cremeens, 1983). Ice contact slightly stratified drift, including flowtill phenomenon, has been described by Flint (1971). Hydrosequence 2 appears to have been developed in relatively uniform material of somewhat coarser texture than hydrosequence 1. 35 Table 1.3 Particle size analysis and bulk density of soils in hydrosequence 1. Particle Size Analysis (%)________________________ total sa n d s il t clay bulk DENS. H orizon SAND SILT CLAY v c s — — — - MS cs - - FS % VFS - CSi FSi CC FC - - - - g/cc MARLETTE 1 A 0-12 26 60 14 1 1 5 10 8 27 33 11 3 1.43 E /B 12-26 22 1 4 8 26 14 4 13 15 0 1 1 1 2 5 8 5 34 26-40 40-84 18 31 1 B tl B t2 60 56 54 25 20 6 BC Cl 84-108 108-165 17 57 1 1 3 31 38 19 22 1 2 23 1 1 2 2 5 165+ 62 62 6 5 42 41 15 14 C2 16 15 6 6 5 6 11 12 1.91 1.87 2 31 23 CA PA C 1 A 0-22 42 39 19 1 3 10 18 14 10 22-32 54 3 2 13 13 15 15 16 14 19 1 3 2 22 26 18 20 29 16 14 2 13 14 8 9 1 2 10 9 24 22 16 49 48 57 2 2 17 32-53 53-97 97-125 •* 13 18 10 B tl B t2 BC 33 33 36 10 14 25 B /E 5 11 9 48 5 29 22 28 21 31 26 2C 125+ PARKHILL 1 28 53 A 0-23 ** 25 1 2 4 6 5 23-53 18 36 57 B gl 43 21 1 3 10 15 JO k 15 n i V 1 •♦i V 7 0 39 37 1 2 12 16 8 17 ^ C . „„ „ . 91+ ,, 24 •* = Lithologic discontinuity at low er boundary o f horizon; uniform ity value exceeded ± 0.60 (Si+vfs)/(s-vfsl for overlvine horizon__________ 1 (S i+ v fs)/(s-v fs) for underlying horizon 20 21 20 19 17 9 7 11 11 8 6 1.86 1.93 2.04 2.01 9 5 1.44 9 1.91 1.86 7 7 8 1.87 1.88 2.1 14 11 1.68 12 9 1.78 n Q t 00 14 10 1.92 36 Table 1.4. Panicle size analysis of soils in hydrosequence 2. Panicle Size Analysis (%) TOTAL Horizon SAND SLT CLAY — — — 40 40 10 VCS CS MS CLAY SL.T SAND FS VFS FSI CSI CC PC BULK DENS. g/cc % M A RLETTE 2 A B /E 0-12 12-32 B tl 32-48 50 49 44 B t2 48-86 35 42 BC C 86-127 127+ 31 38 56 46 31 42 37 11 19 23 13 16 0 1 1 1 1 1 4 14 3 13 21 21 8 9 17 16 23 24 6 4 1.5 8 3 1.9 12 19 8 19 18 11 8 1.96 15 13 22 46 12 7 10 16 11 6 7 1.9 1.97 3 6 7 8 20 2 10 8 2.05 5 4 19 27 18 10 8 17 14 4 12 21 21 9 2 4 2.05 13 14 20 22 9 11 22 28 16 10 10 12 8 9 4 6 5 4 5 3 3 2 31 10 15 9 CA PA C 2 6 2 2 1 2 4 35 13 14 14 35 41 14 2 4 14 51 8 1 3 13 22 22 11 10 28 29 A 0-17 63 B /E 17-33 Bt Btg 33-58 58-85 45 48 BC 85-160 C 160+ BR OOKSTO N 2 51 54 38 4 1.49 1.91 1.86 2.03 2.09 0-25 61 30 9 5 23 8 16 14 5 4 1.46 25-40 28 16 2 17 24 16 6 40-58 58-95 31 33 19 20 2 1 4 20 13 8 11 1.82 56 31 13 2 4 20 22 8 15 14 8 9 1.92 1.87 19 18 18 10 11 4 16 15 9 8 12 B tg l Btg2 56 50 47 5 4 19 BA 17 6 JO 40 r /> A BC r* V- 95-170 t •?r>. * / VT 4r , r 4, t , r IJ A r U J 7 7 1.83 4 *.UJ | 37 Sand Mineralogy Preliminary mineral counts in triplicate samples indicated that good reproducibility could be obtained and as a result single samples were used for mineral counts. The single samples were counted three times and in each count 300 grains were counted and the average was considered. Previously established statistical computation indicates that probable error depends largely upon the number of grains counted, and is greatest for the rarer minerals and lowest for the abundant minerals (Galehouse, 1971). The mineralogy of the fine sand fraction is composed of quartz, K-feldspar, plagioclase, chert, pyroxene, amphibole, sericite, dolomite, altered minerals, and opaque minerals (Table 1.5). Quartz is the only mineral that accounts for more than 50% but less than 90% of the fine sand of these soils. Thus these soils are classified as having mixed, not siliceous, mineralogy at the family level according to Soil Taxonomy. Eberl et al. (1987) have described sericite as a petrographic term used to indicate highly bifringent, fine-grained, micaceous material that is viewed under the optical microscope. Sericite is also described by others as a fine-grained white mica present in acidic intrusive rocks which have undergone hydrothermal alteration. Sericite forms in response to changes in the H+/alkali ratio in the rock-water system (Meunier et al., 1982). Sericite can cover many mineral types (Eberl et al., 1987). It has been described as being composed of muscovite, phengite, illite, hydromica, or mixed-layer illite/smectite, with fixed interlayer cation contents which are less than the structural limit of one equivalent per O 10(OH)2 (Eberl et al., 1987). Results of the petrographic analysis of the fine sand fraction indicates the following points. A/ The quartz content for the solum (A and B horizons) is always higher than the parent material (C horizon). The differences between the solum and the parent material values are less pronounced for the poorly drained profiles particularly for Pedon PI. B/ The amount of plagioclase in the solum increases with depth while the potassium feldspar decreases or remains relatively uniform with depth. C/ The quartz to plagioclase feldspar ratios tend to decrease with depth indicating the weathering of plagioclase relative to quartz (Table 1.6 and Figure 1.11). The A horizons show the highest ratios in all pedons except the A horizon of pedon M2. D /The quartz/potassium feldspar ratios also tend to decrease with depth (Table 1.6 and Figure 1.12) but the ratios within a pedon are less than the quaitz/plagioclase ratios for the same pedon. The differences in ratios may be due to different intensities of plagioclase and K-feldspar.weathering The relatively small increase in 36 quartz/K-feldspar ratios relative to the parent material indicate the less weathering of Kfeldspar than the plagioclase feldspars. The relative similarity of the ratios in the C horizons indicates the mineral uniformity of the parent material (Table 1.6 and Figures 1.11,1.12,1.13). In pedon PI the quartz/k-feldspar ratio is higher in the B horizon than the A horizon (Table 1.6 and Figure 1.12), suggesting variability in the mineralogical composition of the material from which the A and B horizons developed relative to that of the underlying material. Differences in the parent material were also observed based on the values calculated according to Cremeens and Mokma (1986). E/ The ratios of quartz/feldspar decreased with increasing depth in all the pedons except pedon PI (Figure 1.13) suggesting differences in parent material.between A and B horizons and also indicates complementary covariance of other percentages. F/ The percentage of chert in all the profile is constant indicating the resistance of chert to chemical and physical weathering. G / Dolomite is present only in C horizons except the Parkhill profile. The absence of dolomite in the C horizon of the pedon PI could be due to the sampling depth of the C horizon 125 cm compared to 170 cm in pedon B2. Dolomite is present in C horizons of pedons M l and C l of hydrosequence 1. The three pedons are about 50 meters apart, therefore it is difficult to assume the lack of dolomite in the pedon PI is due to parent material differences. H/ Pyroxenes, amphiboles, and opaque minerals are present in small amounts and are relatively uniform throughout the profile. Pyroxene and amphiboles tend to be lower in the solum than in C horizons. These minerals constitute a minor portion of ihe iotai minerals present (<5%). V Sericite is nearly uniformly distributed in the profile with a slight increasing tendency in the C horizon. Sericite viewed with a petrographic microscope was observed as shining, fine grained particles on K-feldspar and plagioclase grains and as well as a single grain. This suggests sericite may have have formed, at least partially, from feldspars. J/ The similarity in mineral contents of C horizons suggest the relative uniformity of the parent materials in each hydrosequence (Table 1.5). The decreasing quartz contents with depth indicate weatherable minerals have weathered in the solum. Plagioclase feldspar appeared to weather faster than potassium feldspar, a trend which agrees with similar results reported by other investigators 39 Table 1.5 HDRECN Mineralogy of fine sand fractions. Or K FELD. SER. N&Ca Rode FELD. FRAG — MARLETTE1 A CHE OP. ----- % PYR. AMP ALT. DoL 68 64 14 15 2 2 7 9 2 2 2 2 1 1 1 1 1 2 2 Bt 1 - C 56 13 3 11 1 3 1 2 2 1 7 CAPAC 1 A Bt 68 13 3 7 2 2 1 1 65 13 3 10 1 2 1 1 1 2 2 2 - 2C 55 13 3 13 1 2 1 2 3 1 5 2 2 1 2 - 2 2 2 - 2 1 2 2 1 - 1 1 1 3 2 1 1 6 PARKHLL A 67 14 4 9 2 Bg C MARLETTE 2 66 12 4 8 62 13 3 12 2 2 A 69 64 13 15 14 2 2 1 2 1 2 3 2 1 1 1 1 2 1 2 3 9 8 11 2 2 11 2 8 9 11 1 1 1 2 2 2 2 1 1 1 3 3 1 1 1 2 Bt C 60 - CAPAC 2 A Bt 70 68 C 58 12 13 BROOKSTON 2 A 69 12 2 9 2 1 1 1 1 2 Bt« 63 2 12 2 1 1 1 1 2 - C 58 14 14 1 12 1 1 1 2 3 1 5 3 Qsquanz; Op-opaque; OEPChea; Rockfiag^rockfiagmot k-feld^k-fcldspoi Pyr= pyroxene; Amp? amphibole: NiCa-feld?pfcigioclase; Alt? alpect Ser? seriae; Dot? dolomite; (Franzmeier et al., 1963). According to Eberl et al. (1987) and Meunieret al. (1982) sericite formation is associated with K rich and acidic rock-water system as a retrograde metamorphic product. However, its association with K poor plagioclase is something that needs further study. The very small amounts of easily weatherable minerals in C horizons - 40 of these soils indicate intense weathering either before deposition or during soil formation or low initial content in the parent rock. Table 1.6 Mineral ratios for the fine sand fraction. Q/F MARLETTE 1 A 2.4 Bt 2.7 C 23 CAPAC 1 A 3.4 Bt 2.8 2C 2.1 PARKHILL 1 A 3.2 3.3 Bg C 2.5 Q/K Q/P 4.9 4.3 4.3 9.7 7.1 5.1 52 5.0 4.2 9.7 63 4.2 5.2 53 4.8 8.4 8.3 5.2 Q/F MARLETTE 2 A Bt C CAPAC 2 A Bt C BROOKSTON A Btg C Q/K Q/P 3.1 2.8 2.4 53 43 43 7.7 8 53 3.7 3.2 2.4 6.4 5.7 43 8.8 7.6 5.3 33 2.4 2.2 5.8 43 4.1 7.7 5.3 4.8 2 SILT MINERALOGY Minerals identified in coarse (20-50 pm) and fine (2-20 pm) silt fractions are chlorite, mica, plagioclase, K-feldspar, quartz, and dolomite (Table 1.7). Generally, identification of minerals present in silt fractions was based on JCPDS (1974) powder diffraction data for standard minerals. The presence of chlorite and mica was based on 14A and 10A peaks in the x-ray diffractogram respectively. Plagioclase was identified using peaks with d-spacings of 4.03A, 4.04A, 3.68A, 3.19A, 3.18A and 3.15A, whereas Kfeldspars were identified from 4.21A, 4.22A, 3.47A, 3.46A, 3.31A, 3.25A, 3.03A, 2.99A, 2.96A, 2.9lA, 2.90A, 2.89A and 2 .1 1 k peaks. The presence of 4.26A, 3.34A, 2.46A, and 2.28A peaks were indicative of quartz in the XRD. Dolomite was identified using the 2.89A, 2.67A, and 2.54A diagonstic peaks. A A B B C C 0—0 M2 o—o M1 A— A 4 5 6 7 P1 a— a 8 Q u a r t z /P l a g io c la s e Figure 1. 11. 9 10 4 5 6 7 B2 8 9 Q u a r t z /P l a g io c la s e ratio Quartz to p lagioclase ratios. 10 A B c C ©—© M1 A—A B2 3 4 6 5 Q u a r tz /K —feldspar Figure 1. 12. 7 3 4 5 6 Q u a rtz/K —feld sp ar ratio Qua'tz to K—feldspar ratios. r- j jOA . * 1 1 w1 / 1 1 1 Horizon □u 1 1 I A; ------ r - c: Q—o M1 *—* C1 A-A P1 1------ 1------ 1-------- 1-----r 2 3 Q u a rtz/F e ld sp a r Figure 1. 13. - —o M2 * C2 A—A B2 o ' 1 1 1" ' ' 2 3 Q u a rtz/F e ld sp a r ratio Quartz to feldspar ratios. " 4 44 Quartz is the dominant mineral present in all horizons as it was in the fine sand fraction. Both silt fractions had more than 80% quartz. Chlorite and mica are present in small quantities mainly in the fine silt fraction of B and C horizons, suggesting their susceptibility to weathering in the surface. Nomberg (1980) indicated these minerals are easily weathered once they loose their protection in rock fragments and quickly reduced to finer particle sizes. Sericite was mainly observed on feldspar grains in the fine sand fraction. The low abundance of mica or sericite in the fine silt fraction indicates it is further comminuted to clay size fraction once it loses its protection on feldspar grains. Dolomite in both silt fractions has completely weathered from the solum. It is present in equal quantities in the fine and coarse silt fractions in the parent materials except pedon P I. The absence of dolomite in this pedon could be for the same reason, depth sampling, as suggested above in the sand fraction. The presence of relatively high quantities of dolomite in C horizons indicates it is less soluble than calcite which is present in the clay fraction. The amounts of plagioclase and K-feldspar are in relatively equal proportion in both silt fractions. Na-feldspar is the dominant feldspar in the silt fractions, indicating it is less susceptibile to weathering than calcium feldspar. This also suggests, once the K and Na feldspars are comminuted their weathering intensities become similar. The silt fraction is dominated by quartz comprising about 80%, followed by dolomite where present and followed by feldspars. Amphiboles and pyroxenes which were present in the fine sand fraction were not detected in the silt fraction. This could be due to their presence in very small quantities which might have limited their detection in the powder diffraction. The other explanation might be due to their high susceptibility to weathering, they might have dissolved chemically. It is evident from these data that the nutrient supply power of the silt fraction is very low. The productivity of the forest vegetation in these soils is maintained from the release of cations from the clay size fraction and as Miller et al. (1979) pointed out it is mostly due to "tight" cycling of available nutrients. 45 Table 1.7 Mineralogy of silt fractions. COA RSE-SILT FINE-SILT PEDON CHL MICA PLAG K-FELD QZ DOL PLAG K-FELD QZ DOL qL M A RLETTE 1 A Bt L L L L L+ H H • L L+ L L H H - C L L L L H H L+ M H H L L H L+ L H Bt L L L L H - L L H - 2C L L L L H H L L H H A Bg L L L L L L L H H L+ L+ H - L L H - C L L L H - L+ L H - A Bt L H H - L+ L+ L H H - C L - L L+ L L M L L+ H H L L H M L L H H L H - L L L+ - - L L M L H H L L L+ L H H L M L L H H L C A PA C 1 A PARKH ILL 1 L L M ARLETTE 2 L CA PA C 2 A Bt C BROOKSTON 2 A . L L H Btg T - L L H r T - t_ t_ u H-HIGH M -M EDIUM L-LOW T-TRACE - H u C H L =chIorite; PLA G = plagioclase: K-FELD=f-feldspar, Q Z =quartz: DOL= dolom ite II II 46 CLAY MINERALOGY X-ray Diffraction Analysis The clay minerals present in this study, based on XRD, are mica, kaolinite, vermiculite, smectite, chlorite, (interstratified vermiculite/chlorite) hydroxy-interlayered vermiculite (HIV), mica/smectite, vermiculite/smectite, hydroxy-interlayered smectite, quartz, and feldspar. These minerals were identified based on XRD patterns using Whittig and Allardice (1986) diffracting criteria Mica was identified by its (001) d-spacing of about 10A in all the treatments. Muscovite and biotite were differentiated based on the presence and intensity of the (002) peak. Presence of discrete kaolinite in all clay samples is indicated by 7.1 A (001) and 3.58A (002) basal reflections which are not present after heating to 550°c due to dehydroxylation. Vermiculite is indicated by about 14.2A peak that shifts to about 10A peak upon K-saturation and heat treatments. Discrete chlorite is identified by the presence of 14.2A peak that persists in all heat treatments. The presence of smectite is shown by about 18A d-spacing when Mg-saturated and glycerol solvated. However, when the samples are K-saturated and heated, the 18A peak collapses to about 10A peak. The presence of HTV is indicated by partial collapse of 14.2A to 10A and the asymmetrical 10A peak , when K-saturated and heated. The presence of a shoulder on the 14.2A peak indicates interstratified mica/vermiculite. Quartz was identified based on 4.26A and 3.34A peaks that persists in all heat treatments. Ciay Mineral Distribution The results of XRD analysis of the <2-pm clay from the six soils representing two hydrosequences are summarized in Table 1.8. Figures 1.14-1.21 contain the X-ray diffraction patterns obtained for these soils. Mica is the predominant mineral followed by kaolinite and vermiculite in the glacial till parent material of fine-loamy soils of Michigan. Similar mineralogical compositions of the C horizons suggest the uniformity of the parent material in which these soils have formed. All pedons in these two hydrosequences show the same kind of weathering trend but with variable intensities. The parent material of all pedons are rich in mica followed by abundant kaolinite and vermiculite, and small amounts of chlorite, and quartz (Table 1.8). Feldspars are present in very small quantities in all pedons indicating moderate weathering in these fine-loamy soils. 47 The most noticeable weathering trend observed is the decrease in amount of vemiculite with depth indicating higher weathering intensity in the A horizon than the B and C horizons. On the other hand the mica content increases with depth. This type o f trend was also observed in the fine clay mineralogy by Cremeens and Mokma (1987). The clay mineral distribution in these two hydrosequences indicates mica has weathered to vermiculite with some subsequent illuviation of vermiculite into the B horizons as shown by fine clay mineralogy of clay films (Cremeens and Mokma, 1987). The intensity of weathering seems to be influenced by drainge as evidenced by peak intensities. Similar clay mineral distributions were reported by Klages et al. (1957) and Kodama (1979). The formation of vermiculite by weathering of mica and translocation of mixed mineral suites, either downward or by lateral movement are favored by a strong leaching environment, K loss from the interlayer, oxidation of iron in iron rich mica, substitution of hydroxyl for oxygen, and loss of octahedral iron and magnesium (Douglas, 1977; Newman, 1967). The loss of charge seems to be related to the oxidation of octahedral iron in iron rich micas and to protonation of structural oxygen anions to form new hydroxyls. Pedon P I, a poorly drained soil, has slight changes in vermiculite and mica contents within the profile. Kaolinite does not show a particular trend, however in some of the pedons the solum is higher in kaolinite content than the parent material. The diffractograms do not indicate a presence of significant mixed-layer clays in all the pedons examined, although presence of minor amounts of hydroxy-interlayered vermiculite is indicated by the small shoulder in the 11A to 13A region and by the incomplete collapse of 14A peak that resulted in a broad peak towards a low angle upon K-saturation and heating to 300°c and 550°c. In the absence of interlayer contaminants, K-saturated vermiculite should collapse to 10A prior io heat treatment, but as indicated by April et ai. (1986), the necessity to heat the sample to force K into the interlayer and to dehydroxylate the interlayer is characteristic of hydroxy-interlayered vermiculite. The amount of HIV is greatest in the surface horizons, where frequent wetting and drying cycles are crucial to the formation of the mineral (Rich, 1968) and also where pH and base saturation are somewhat lower than the parent material, (Figures 1.9 and 1.10), particularly in Marlette and Capac pedons. Chlorite contents are relatively higher in the C horizons than the A horizons. In well drained and moderately well drained pedons, chlorite is present in trace amounts in the surface horizon. In general the amount of chlorite decreases by half when you go from the parent material to the B horizon. The chlorite present in C horizons of these pedons may have weathered to vermiculite (Ross et al., 1982; Ross, 1975). The amount of chlorite increases with depth whereas vermiculite decreases. However, the chlorite present in the parent material is likely trioctahedral as shown by the 060 diffraction peak of 1.54A. The 46 060 diffraction peak of the B and A horizons indicate the clay minerals present are dioctahedral. The absence of trioctahedral peak of the 060 diffraction in the solum could be due to a low quantity of chlorite present in these horizons or more likely the trioctahedral chlorite has weathered to amorphous material and/or has decomposed to soluble products. In other words trioctahedral chlorite present in the parent material eventually disappears from the solum. This leads to the conclusion that the dioctahedral vermiculite present in the solum is a weathering product of muscovite/sericite or/and neoformation. Smectite is present in the <2 pm fraction of the Btg horizon and in the fine clay fraction of all horizons in pedon B2 (Fig. 1.20). The Green-Kelly test and the 060 spacing of the fine clay of the Btg horizon indicates the smectite phase is beidellite. Smectite however, is absent from all other pedons even in the fine clay fraction. The absence of smectite in pedon P I, the comparable drainage class in hydrosequence 1, where minimum leaching occurs, is possibly attributed to differences in parent material and higher total K content in the solum than in the pedon B2 (13-18 ppm Vs 9-11 ppm). The high water table through out most of the year (Figures 1.3,1.4, and 1.5) inhibits the possibility of clay translocation in the C horizons deeper. The presence of smectite even in the deeper C horizon (250 cm) indicates it is inherited from the parent material. The distribution of smectite in the fine clay size fraction within the profile indicates this clay mineral is allogenic. The mineralogy of the clay films and the fine clay does not reflect monomineralic composition, therefore smectite is not likely the result of neoformation. The fine clay mineralogy as well as the mineralogy of the clay films (Cremeens and Mokma, 1986) of the Btg horizons indicated the existence of mixed mineral suites (mica, smectite, kaolinite, and vermiculite). In a soil where clay translocation, in situ weathering, and neoformation have occured, it is difficult to attribute the presence of smectite (Btg) to a single mechanism. Proposed Weathering Sequence Sand, silt, and clay mineral distribution suggest the following weathering sequence: Vermiculite / Beidellite T i Sericite/Muscovite -> vermiculite -^Beidellite i Hydroxy-interlayered vermiculite Plagioclase —» Dissolves Chlorite -» Dissolves 49 Chlorite -4 vermiculite The evidence for this proposed sequence is primarily based on the depth distribution of various components. The trends of weathering observed in these hydrosequences of fine-loamy soils in Michigan is in agreement with the literature. The most important changes that occur in mica during the transformation process are (1) a reduction in layer charge and (2) a replacement of interlayer K with hydrated cations (Fanning et al., 1989; Robert, 1973). Fanning et al. (1989) reviewed the literature concerning possible mechanisms for the layer charge reduction and concluded that the most probable mechanisms were (1) the oxidation of structural Fe2+ to Fe3* following the opening of the mica interlayer, (2) proton incorporation into the structure, and (3) Si replacement of Al and Fe in the tetrahedral sheets. Vermiculite derived from mica is fast forming and unstable (Kittrick, 1973). Vermiculite is often abundant in soils because the rate at which interlayer K is removed from mica to form vermiculite is often more rapid than the rate at which the 2:1 layer is broken dawn (Kittrick, 1973; Fanning et al., 1989). Robert (1973) determined that some dioctahedral micas were transformed first to dioctahedral vermiculite and then to dioctahedral smectite whereas other dioctahedral micas were transformed only to either dioctahedral smectite or vermiculite. Stoch and Sikora (1976) observed that mica weathering consisted of gradual structural transformations with removal from the structure of Mg, Fe, and K. They thought that mica weathered according to the sequence of muscovite -4 regularly interstratified mica/montmorillonite —» montmorillonite -4 kaolinite. The source of mica in this study is sericite which is a retograde metamorphic product of feidspar and iithogenic muscovite. Sand mineralogy does not indicate the presence of muscovite or biotite, however the clay fraction indicates abundant dioctahedral mica in the glacial till parent material. In the present study no attempt was made to characterize or differentiate muscovite and sericite in the clay fraction. According to Eberl et al. (1987) and Meunier et al. (1982), particle thickness of sericite influences the mineral's properties. Petrographic microscope examination reveals sericite, whereas the TEM (Transmission electron microscope) shows individual mica particles and XRD gives patterns for mixed-layer illite/smectite or muscovite. In this study the dioctahedral mica in the clay fraction is assumed to be muscovite. Elemental analysis of the fine clay fraction indicates the weathering product, smectite, is iron rich. 50 Table 1.8 Clay mineralogy of total clay. HORIZON SMBCTME CHLURIIE VERMCULTIE MCA KAOUNTIE QUARTZ O f. MARLETTE 1 A Bt C CAPAC 1 A Bt 2C PARKHLL1 A Bg C MARLETIE2 A Bt C CAPAC 2 A Bt C BROOKSTCN2 A Btg n w XXX \= APPROXIMATELY5% X=APPROXIMATELY 10% -= ABSENT T= TRACE T \ X XXX\ XX\ X XXX XXXX XXXX XXX XXv XXX \ \ X \ \ X XXXX XXv XX XXv XXXv XXXX XXv XXX XXv \ \ \ \ \ \ XXX XXv XXv XXX XXXX XXXX XXX XXv XXv \ \ \ T X X XXXX XX XX XXv XXXv XXXX XXv XXV XX X X X \ \ X xxx\ XXv XX XXv XXXX XXXX XXv xx\ XX X \ X \ \ XXv X XXX XXv V A AA AAAA XXv XX XX Xv \ X 51 SUMMARY AND CONCLUSIONS The sand, silt, and clay mineralogy of the two fine-loamy hydrosequences of southcentral Michigan suggest the glacial till parent material is relatively uniform. Depth and duration of water tables were similar for soils with the same natural drainage class. Having similar parent material, vegetation, climate, and time, topography has produced no significant differences in distribution of sand and silt size minerals within and among profiles. Weathering of muscovite/sericite to smectite (beidellite) through interstratified vermiculite-smectite seems to be the dominant trend in the poorly drained Brookston pedon whereas direct transformation to vermiculite is the prevalent weathering mechanism in the other drainage classes. Except for the smectite in the Brookston pedon, the fineloamy soils in this study do not show significant differences in sand, silt, and clay mineralogies with respect to natural drainage. 52 M IA IN T E N S IT Y , L k 550 K 300 REL. 1 Mg-Glg ^ •R W P P I i M l Bt ^K550 REL. INTENSITY ... K 300 Mg-Gl y i ..... M IC REL. INTENSITY / \ I ' ■ ‘i 1 [Lk 5 5 0 Jl A L . IlK 3 0 0 I -------- --- .--- --- -— v ,. . Mg-Glg 1------------0 11-------------1------------- 11------------10 20 -------- r1— 30 D e g r e e s 2B Fig. 1.14 X-ray diffr a ctog ra m s of clay fraction (<2jim) of pedon Ml. 53 K 550 K 300 Mg-Gly C l BT K 550 J K 300 Mg-Gly J >■ H /I (ft Z LkJ *— AjLJLuJfe Mg-Gly UJ OC 10 20 30 D eg rees 20 Fig. 1.15 X-ray d i f f r a c t o g r a m s of cla y f r a c t i o n (< 2 n m ) of pedon Cl. 54 >• H <75 ^ Mg-Glg bJ ac - A J J a K 550 >• H <75 K 300 z tu t- UJ '-MJjJC K 550 > H <75 z bJ i iy-oig 10 20 D e g re e s 28 Fig. 1.16 X-ray d if f r a ct o g ra m s of clay fra c tio n (<2um) of pedon PI. 55 REL. IN T E N S IT Y ^K550 K 300 . - A .. i ___ 1 Mg-Glg v *v A M 2Bt REL. INTENSITY , K 550 — ~J\ K 300 i v L |M g -G ly .........1 -j. - . - r— • i —^ I l M 2C REL. INTENSITY I K 550 1 Tnn 1 (vS W WW Mg-Gly 0 10 20 30 D e g re e s 26 Fig. 1.17 X-ray d iffr a cto g ra m s of clay fr a ction (<2pim) of pedon M2. 56 | C2A REL. IN T E N S IT Y K 550 V a ** K 300 M p s ty - i i i REL. INTENSITY C 2BT f e .Mg-Gly C2C ■ 1 i ' I I --------r REL. INTENSITY [ K 550 K 300 f e Mg-Gly 1 --------------------------- 1— 0 -------- 10 1--------------------------- r 20 30 D e g ree s 28 Fig. 1.18 X-ray d iffr a cto g ra m s of clay fra ctio n (<2*im) of pedon C2. 57 B2A REL. IN T E N S IT Y 550 Mg-Gly K 550 REL. INTENSITY K 300 B2Bt Mg-Gly K 550 REL. INTENSITY B2C K 300 M g-G ly 0 0 20 10 30 Degrees 2B Fig. 1.19 X-ray diffr a ctog ra m s of clay fraction (<2um) of pedon B2. 58 2A COARSE J vv | mediu; COARSE B2Bt COARSE 10 20 D e g re e s 20 Fig. 1.20 X-ray diffr a ctog ra m s of pedon B2 fine, medium, and coarse clay f r a c t i o n s a f t e r Mg satura tion and glycerol solvation. 59 4000 -r REL. IN T E N S IT Y 1 7 .7 6 A 3000 - 12.23* lio .o iX <1 7.16* 2000 - 1000 - 0 B tg (<2|im ) L i- s a t . 2 5 0 ° c + G lu. Li-sat. 250°c - ■ 0 10 20 D egrees 20 Fig. 1.21. X-ray diffractograms of the Bt horizon (<2-pm) clay fraction v h ich has been Li-saturated and glycerated. LIST OF REFERENCES 61 LIST OF REFERENCES Allen, B.L., and D.S.Fanning. 1983. Composition and soil genesis, p. 141-184. In L.P.Wilding et al. (ed) Pedogenesis and soil taxonomy 1. Concepts and Interactions. Developments in Soil Science 11 A; Elsevier, New York. April, R.H., M.H.Hluchy and R.M.Newton. 1986. The nature of vermiculite in Adirondack soils and till. Clays Clay Miner. 34:549-556. Ameman, H.F., A.D.Khan and P.R.McMiller. 1958. Physical, chemical, and mineralogical properties of related Minnesota Prairie soil. Minn. Agr. Exp. Sta. Tech. Bull. 227. Arnold, R.W. 1965. Multiple working hypothesis of soil genesis. Soil Sci. Soc. Am. Proc. 29:717-724. Asady, G.H. and E.P.Whiteside. 1982. Composition of a Conover-Brookston map unit in south eastern Michigan. Soil Sci. Soc. Am. J. 46:1043-1047. Bailey, E.E. and R.E.Stevens. 1960. Selective staining of K-feldspars and plagioclase on rock slabs and thin sections. Am. Mineral. 45:1020-1025. Bidwell, O.W. and J.B.Page. 1950. The effects of weatherin on the clay mineral composition of soils in the Miami catena. Soil Sci. Soc. Am. Proc. 15:314318. Bretz, J.J. 1951. Causes of the glacial lake stages in Saginaw basin, Michigan. Jour. Geol. 59:244-258. Bundy, L.G. and J.M.Bremner. 1972. A simple titrimetric method for determination of inorganic carbon in soils. Soil Sci. Soc. Am. Proc. 36:273-275. Cady, J.G. 1960. Mineral occurrence in relation to soil profile differentiation. Seventh Int. Congr. Soil Sci., Trans., IV: 418-424. Carver, R.E. 1970. Procedures in sedimentary petrology. Wiley-Interscience, New York. Cremeens, D.L. 1983. Argillic horizon formation in the soils of a hydrosequence. MSc. Thesis, Michigan State University. Cremeens, D.L., and D.L.Mokma. 1986. Argillic horizon expression and classification in the soils of two Michigan hydrosequences. Soil Sci. Soc. Am. J. 50:1002-1007. 62 Creemens, 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. 51:13781381. Daniels, R.B., E.E.Gamble and LJ.Bartelli. 1968. Relation between A2 horizon characteristics and drainage in some fine-loamy Ultisols. Soil Sci. 104:364369. Douglas, L.A. 1989. Vermiculites. p.635-674. In J.B.Dixon and S.B.Weed (2nd. ed.) Minerals in soil environments. Soil Sci. Soc. of Am., Madison, WI. Drever, J.1 .1973. The preparation of oriented clay mineral speciments for X-ray diffraction analysis by a filter membrane peel technique. Am. Mineral. 58:553-554. Eberl, Dennis D., J.Srodon, M.Lee, P.H.Nadeau, and H.R.Northrop. 1987. Sericite from the Silverton caldra, Colorado: Correlation among structure, composition, origin, and particle thickness. Am. Mineral., 72:914-934. Fanning, D.S., V.Z.Keramidas and M.A.El-Desoky. 1989. Micas, p. 551-624. In J.B.Dixon and S.B.Weed (2nd.ed.) Minerals in soil environments. Soil Sci. Soc. Am., Madison, WI. Flint, R.F. 1971. Glacial and quaternary geology. Chap. 7. pp. 147-197. New York: John Wiley and Sons, Inc. Franzmeier, D.P., E.P.Whiteside, and M.M.Moitland. 1963. A chronosequence of podozols in northern Michigan. Mich. State Univ. Quart. Bull., 46:37-57. Galehouse, S.S. 1971. Point counting. P.385-407. In R.E.Carver (ed.) Procedures in sedimentary petrology. Wiley-Interscience, New York. Hutcheson, J.B.Jr., RJ.Lewis ami W.A.Seay. 1959. Chemical and ciay mineraiogicai properties of certain Memphis catena soils of western Kentucky. Soil Sci. Soc. Am. Proc. Vol. 23:474:478. Jackson, M.L.,1979. Soil chemical analysis advanced course. (3rd. print). Published by author, Dept, of Soil Science, Univ.of Wisconsin, Madison. JCPDS 1974. Selected powder diffraction data for minerals: Joint committee on powder diffraction standards, Swarthmore, Pa., 262 p. Johnson, L.J. and C.DJeffries. 1957. The effect of drainage on the weathering of clay minerals in the Allenwood catena of Pennsylvania. Soil Sci. Soc. Am. Proc. 21:539-542. Kittrick, J.A. 1973. Mica-derived vermiculites as unstable intermediates. Clays Clay Miner. 21:479-488. 63 Klages, M.G. and J.L.White. 1957. A chlorite-like mineral in Indiana soils. Soil Sci. Soc. Am. Proc. 21:16-20. Kodama, H. 1979. Clay minerals in Canadian soils: Their origin, distribution, and alteration. Can. J. Soil Sci. 59:37-58. Mackintosh, E.E. and J.VanDerHulst. 1978. Soil drainage classes and soil water table relations in medium and coarse-textured soils in southern Ontario. Can. J. Soil Sci. 58:287-301. Martin, R.T. and M.B. Russel. 1952. Clay minerals of four southern New York soils. Soil Sci. 74:267-279. Miller, H.G., J.M.Cooper, J.D.Miller, and O.J.L.Pauline.1979. Nutrient cycles in pine and their adaptation to poor soils. Can. J. For. Res. 9:19-26. Meunier, A. and B.Velde. 1982. Phengitization, sericitization and potassiumbeidellite in a hydrothermally-altered granite. Clay Miner. 17:285-299. Newman, A.C.D. 1967. Changes in phlogopites during their artificial alteration. Clay Miner. 7:215-227. Norn berg, P. 1980. Mineralogy of a podozol formed in sandy materials in northern Denmark. Geoderma. 24:25-43. Rich, C .1.1968. Hydroxy interlayers in expansible layer silicates. Clays Clay Miner. 16:15-30. Robert, M, 1973. The experimental transformation of mica toward smectite; relative importance of total charge and tetrahedral substitution. Clays Clay Miner. 21:167-174. Siuhu, P.S. and R.J.Giikes. 1977. mineralogy of soils developed on alluvium in the Indo-Gangetic Plain (india). Soil Sci. Soc. Am. J. 41:1194-1201. Soil Survey Staff. 1984. Soil survey laboratory methods and procedures for collecting soil samples. Soil Surve Investigations Report I. Revised July 1984. SCS-USDA. U.S. Govt. Printing Office, Washington, D.C. Soil Survey Staff. 1951. Soil survey manual. U.S.D.A. Handbook No. 18. Agricultural Research Administration U.S.D.A. Stoch, L. and W.Sikora. 1976. Transformations of micas in the process of kaolinization of granites and gneisses. Clays Clay Miner. 24:156-162. Thorp, J. and E.E.Gamble. 1972. Annual fluctuations of water levels in soils of Miami catena. Wayne County, Indiana. Science Bull. No.5, Earlham College, Richmond, Indiana. 64 Ward, M J. 1979. The glacial history of Early Lake Saginaw. Unpublished M.Sc. Thesis. Michigan State University, E. Lansing, Michigan. Whittig, L.D. and W.R.Allardice. 1986. X-ray diffraction techniques. In C.A.Black et al., (2nd. ed.) Methods of soil analysis, Part 1-Physical and mineralogical methods. Agronomy 9:331-359. Yassoglou, N.J. and E.P.Whiteside. 1960. Morphology and genesis of some soils containing fragipans in northern Michigan. Soil Sci. Soc. Am. Proc. 24:396407. 65 CHAPTER 2 MINERALOGY OF TW O SANDY HYDROSEQUENCES IN MICHIGAN ABSTRACT 2 The aim of this study is to examine the mineralogy of sand, silt, and clay size minerals in two sandy hydrosequences of Michigan. Apparent depth of water table was monitored using piezometers. The dominant sand fraction was considered for identification of minerals using petrographic microscope. X-ray diffraction was used to examine the silt and clay mineralogy. The sand mineralogy is dominated by quartz followed by small amounts of feldspars and sericite. The silt fraction is dominantly composed of quartz and small amounts of feldspars. Differences in drainage do not indicate a particular distribution of sand and silt minerals. Chlorite, mica, kaolinite, vermiculite and quartz are the major minerals in the parent material of the clay fraction. The intensity of smectite peak and the presence and absence of vermiculite in the E horizons, indicates increasing weathering intensities from poorly drained to well drained soils in the these hydrosequences. The mineral distribution in these hydrosequences suggests the following weathering sequences: dioctahedral mica/sericite -» vermiculite -» smectite with or without mixed-layer formation. 66 INTRODUCTION In earlier work by Franzmeier et al.(1963) and Ross and Mortland (1966) in Michigan Spodosols, smectite was found to be the dominant clay mineral present in the eluvial horizon (E). Those studies were conducted to see soil development in a chronosequence of Spodosols and to characterize smectite in the E horizon of well drained Spodosols in northern Michigan respectively. Other studies have also reported the presence of 18A clay mineral in the acidic E horizon of Spodosols (Brown and Jackson, 1958; Gjems, 1960; McKeague, 1965; Coen and Arnold, 1972; and Douglas, 1982). One of the major hinderances in Michigan as in other glaciated areas in the study of weathering sequences is the lack of precise knowledge of the original nature of the parent material (Rutherford and Debenham, 1981). The formation of smectite in this acidic environment was attributed by most researchers as the weathering product of mica and vermiculite-chlorite and by others as being added as dust (Coen and Arnold, 1972) or weathering product from minerals added as dust. The aim of this study is to examine the mineralogy of sand, silt, and clay size minerals in two sandy hydrosequences of Michigan. 67 MATERIALS AND METHODS FIELD METHODOLOGY Two replicated hydrosequences in sandy outwash material were selected for this study. The sandy hydrosequences are located in Chippewa and Cheboygan Counties in Michigan (Figure 1.1). The soils in each hydrosequence occurred in adjacent mapping units (delineations), so as to keep the other soil forming factors relatively constant. These hydrosequences occur as true sequences as defined by Jenny (1941,1980). The hydrosequences were located on forested lands where disturbances due to cultivation, artificial drainage and extensive erosion are absent or minimal. Each drainage sequences consisted of three soils in the landscape with increasing heights above the local water table. According to drainage classes, these soils were classified as moderately well drained, somewhat poorly drained, and poorly drained soils and occupy, back slope, foot slope, and toe slope positions on the landscape, respectively. These hydrosequences occur as natural products of young glacial landscapes with immature drainage networks that develop on them. The sandy hydrosequences have sandy glacial drift parent materials and are both located on glacial outwash plains The soil series in the hydrosequences are shown in Table 2.1 and their classification and location in Table 2.2. Piezometers were installed in the C horizon near each of the sampled pedons to monitor the apparent depth of water table. Water table levels were monitored bi-monthly for the period i October 1985 through 31 December 1987. Piezometer data were converted to profiles of depth versus duration of saturation. Soil sampling pits were dug and soils described following standard conventions (Soil Survey Staff, 1951). Pedon descriptions are presented in Appendix 1. Approximately 1020 kg of bulk samples were collected from each mineral horizon of each pedon along the faces described, for chemical and mineralogical analysis 66 TABLE 2.1: Soils of the two sandy hydrosequences SEQUENCE CHIPPEWA HYDROSEQUENCE CHEBOYGAN HYDROSEQUENCE MODERATEL WELL DRAINED LIMINGA SOMEWHAT POORLY DRAINED WAINOLA POORLY DRAINED KINROSS 1 RUBICON AUGRES KINROSS 2 TABLE 2.2: CLASSIFICATION OF SOILS IN THE TWO HYDROSEQUENCES SOIL SERIES HYDROSEQUENCE 1 LIMINGA WAINOLA KINROSS 1 HYDROSEQUENCE 2 RUBICON AUGRES KINROSS 2 CLASSIFICATION SANDY, MIXED, FRIGID, TYPIC HAPLORTHOD SANDY, MIXED, FRIGID, TYPIC HAPLAQUOD SANDY, MIXED, FRIGID, AERIC HAPLAQUOD SANDY, MIXED, FRIGID, ENTIC HAPLORTHOD SANDY, MIXED, FRIGID, ENTIC HAPLAQUOD SANDY. MIXED. FRIGID, ENTIC HAPLAQUOD LABORATORY ANALYSIS Soil samples were air dried, crushed and sieved using a 2 mm sieve to remove coarse fragments. These samples were stored and used for the following analysis. The following procedures were performed using methods of the Soil Survey Staff (1972, 1984): particle size distribution by pipette; pH in H 20 and IN KC1; Fe, and A1 extraction with pyrophosphate, oxalate and citrate-dithonite. Total C was determined by wet oxidation using the method of Snyder and Trofymow (1984). SAND MINERALOGY The dominant sand fraction was considered for identification of minerals. The dominant sand fraction was impregnated with a 3M Scotchcast and #3 Resin in a test tube. Thin blocks were cut, lightly polished to make one side flat and mounted on glass slides. The mounted slides were thin sectioned to approximately 30p thickness (Carver, 1970). Quantitative determination of feldspar was made by etching the slides with hydrofluoric acid and staining with sodium cobaltinitrite, barium chloride and rhodizonate reagent as described by Bailey and Stevens (1960). Percentages of the various minerals 69 present were determined by point counting of approximately 900 grains using a standard petrographic microscope (Galehouse, 1971). SILT AND CLAY MINERALOGY XRD analysis were conducted using Philips APD 3720 automated X-ray diffractometer system. A theta compensating slit and diffracted beam monochromator with graphite crystal were used to maintain a constant area of irradiation and for monochromatization of the X-ray beam, respectively. All samples were scanned at a rate of 0.02°/second. Diffractograms were obtained over a total scan range of 2° 20 to 40° 20 for the Mg saturated clays, coarse and fine silt at room temperature and from 2° 20 to 30° 20 for all other cation and heat treatments. X-ray diffraction was used to examine the silt and clay mineralogy (Jackson, 1979). Random powder X-ray diffraction of the coarse and fine silt was obtained by placing them in a sample holder. Total clays and fine clays were separated by the sedimentation and centrifugation, respectively. After removal of the cementing agents (organic carbon, and iron oxide) X-ray diffractograms of the clay fractions were obtained from basally oriented specimens (Drever, 1973) which had been: (a) Mgsaturated and glycerol solvated, (b) K-saturated and heated to 300°C or (c) K-saturated and heated to 550°C. The 060 spacing of the clays was measured using a random powder diffraction. Integrated peak intensities were used to semi quantitatively estimate the clay minerals present. Minerals were identified for XRD patterns using diffracting criteria out lined by Whittig and Allardice (1986). RESULTS AND DISCUSSION VARIATION IN THE DEPTHS TO THE WATER TABLE Average monthly depths to water tables and average monthly precipitation in the two hydrosequences are presented in Figures 2.2 and 2.3. In the moderately and well drained pedons of both hydrosequences, the high water table was observed at a depth of 130cm during the month of April. During the rest of the year the water table was at or below 150cm The high water table during April is associated with snow melting rather than one of high precipitation. In the somewhat poorly drained and poorly drained pedons the water table drops and reaches the lowest between June and September. It is during these months precipitation, evaporation and plant uptake of water are the highest. 70 The Kinross and Au Gres pedons in Cheboygan County show a higher water table during fall, winter and spring than the Kinross and Wainola pedons in Chippewa County. The Cheboygan hydrosequence seems wetter than the Chippewa hydrosequence as observed by the depth to water table and the amount of precipitation during these periods. The eluvial and the spodic horizon of the Kinross pedon and the spodic horizon of AuGres pedon in Cheboygan County were under high water table for at least six months (between December and May) of the year. On average the eluvial horizon of Kinross pedon in Chippewa County was above the high water table for most of the year whereas the lowest part of the spodic horizon was under water for four months (April, May, October and November). The water table was below the spodic horizon of Wainola pedon throughout the study period. As shown by Figures 2.4 and 2.S, the somewhat poorly drained and poorly drained pedons of Cheboygan County, the eluvial and the spodic horizons had a longer duration of saturation (40% time saturated) than the corresponding pedons in Chippewa County (<20% time saturated). Much of the variation in the depths to the water table can be accounted for by the amount and frequency of precipitation.and evapotranspiration. Due to higher specific position in the landscape and more percolation of water in well drained and moderately well drained soils, thicker eluvial horizons were thought to develop in sandy materials (Lag, 1970). The thickness of the eluvial and the spodic horizons of the somewhat poorly drained pedons in both hydrosequences are deeper than the other pedons in the hydrosequences. It seems there is no direct relationship between thickness of eluvial and spodic horizons to drainage classes. Deepening of these horizons seems to be controlled by some other soil forming processes other than drainage alone. If drainage affects the rate of weathering, the Chippewa pedons should indicate a relatively higher intensity of weathering than the Cheboygan pedons (AuGres and Kinross) that has been under water for a longer period of time. Drainage is one of the most important factors in the formation of different type of soils in a landscape Jackson and Sherman (1953), Barshad (1959), Rich (1968), and Huang and Lee (1969) have discussed the relation of chemical weathering of minerals with drainage. Becouse drainage influences the moisture regime, seepage and redox potential, it is believed that the weathering of minerals and interlayer A1 compounds would also be affected (Huang and Lee, 1969). o * _c -+-> Cl Q) Q o—o Liminga * Wainola - a—a K inross ■0 - 100 CU_>N cn_c ' o -t— J c S >> o . < o CD L. CL - 806040 20 “T ” "i— M A —rM T" J J A s — I---------------- 1— 0 N M o n th s Figure 2.1. Average monthly depths to water tables and average monthly precipitation in hydrosequence 1, 1 Jan. 1985 — 31 Dec. 1987, Chippewa County. X E O X — CL CD Q o —o Limingci * Wainola a—a K inross 0= = 0) > . § crw - c o ' <1) > § .9 < EO (D 100 806040 20 ■ 0 - - -T~ M -T~ A Tu j j A t - s -I-------- 1— 0 N M o n th s Figure 2.1. Average monthly depths to water tables and average monthly precipitation in hydrosequence 1, 1 Jan. 1985 — 31 Dec. 1987, Chippewa County. ^ Liminga Wainola Kinross Depth (cm -5 0 -7 0 -9 0 —1 1 0 — -1 3 0 -1 5 0 0 20 40 60 80 100 % Tim e s a t u r a t e d Figure 2.3. Percent total time saturated at given depth in hydrosequence 1, 1 Jan. 19 85 — 31 Dec 1987, Chippewa County. - 10 Rubicon Au Gres Kinross - -3 0 - (cm -5 0 - Depth -7 0 -9 0 - 110 - -130 -150 0 20 40 60 80 100 % Time s a t u r a t e d Figure 2.4. Percent total time saturated at given depth in hydrosequence 1, 1 Jan. 19 85 — 31 Dec 1987, Cheboygan County. 75 PARTICLE SIZE DISTRIBUTION The total sand percentages of these soils range from 81-98%, and these values generally attain their maxima in the C or lower B horizons (Tables 2.3 and 2.4). The Cheboygan hydrosequence has a relatively higher sand percentage (91-98%) than the Chippewa (81-98%). In most instances the eluvial horizons tend to have lower sand contents than the horizons below, suggesting some minerals in the sand fractions might have comminuted to the silt and clay fractions The dominant sand fractions in Cheboygan and Chippewa hydrosequences are medium (39-55%) and fine (29-80%) sand fractions respectively. The total silt percentages are usually highest in the eluvial and upper B horizons and tend to decrease in lower part of the profiles. Much of the silt may have weathered from the sand fraction, since its distribution in the profile follows the pattern of increasing intensity of weathering (increasing with proximity to soil surface). The presence of altered grains in the sand fraction of the solum also suggests it is a source of silt and clay in these horizons since they could easily be comminuted to a smaller size fractions. In all the horizons the clay content is less than 5%. The total percentages of very fine sand plus silt is higher in the Chippewa than in the Cheboygan hydrosequence. The kinross 1 pedon in Chippewa has the highest percentages of very fine sand plus silt. Very fine sand plus silt have big influence on moisture holding capacity. TOTAL CARBON AND EXTRACTABLE IRON, AND ALUMINIUM All pedons show a subsurface maximum of C, Fe, and Al. The Fe, A1 and C maxima always occur in the spodic horizon (Tables 2.5 and 2.6). 76 TABLE 2.3. Particle size analysis of soils in Chippewa hydrosequence. PARTICLE SIZE ANALYSIS (%) TOTAL Horizon SAND SILT SAND SILT CLAY vcs cs MS FS VFS CSi FSi LIMINGA A 0-7 E 7-11 Bhs 11-22 Bs 22-63 C 63-192 92 92 94 94 98 6 5 3 4 2 2 3 3 2 0 0 0 0 0 0 2 1 1 0 0 12 14 13 10 12 67 66 69 75 75 10 11 11 10 12 2 1 3 4 2 4 4 0 0 0 WAINOLA Oi 7-0 E 0-15 Bhs 15-43 BC 43-77 C 77-118 88 94 95 85 12 6 5 15 1 0 0 0 3 2 0 0 15 7 0 5 14 7 1 17 46 67 80 29 10 11 15 33 9 5 4 7 3 1 0 9 KINROSS 1 Oa 10-0 E 0-8 Bhs 8-19 Bg 19-34 n ia kS««u ^ 85 86 81 OI y 15 14 19 n t 0 0 0 A 4* 0 1 1 4 4 3 h ■J 11 13 13 45 48 49 25 20 16 11 12 10 4 6 10 1 1 A A A A T T 1 1 •S 4 n A V a 1 A 77 Table 2.4. Particle size analysis of soils in Cheboygan hydrosequence. PARTICLE SIZE ANALYSIS (%) TOTAL Horizon SILT SAND SAND SILT CLAY v c s cs MS FS VFS CSi FSi 14 18 16 16 8 44 46 43 55 40 35 29 35 26 46 2 2 1 0 0 3 2 1 1 1 16 15 11 11 2 42 40 42 39 42 31 32 36 38 51 2 2 3 3 . 3 2 8 3 2 3 0 1 1 0 14 47 44 36 31 36 51 2 4 7 3 1 4 2 0 0 RUBICON A 0-10 E 10-34 Bsl 34-50 Bs2 50-113 Cl 113-153 C2 153-172 96 95 0 3 I 1 1 1 1 0 0 1 98 5 2 1 2 2 94 91 95 94 98 5 8 2 4 2 1 1 3 2 3 1 2 3 1 0 96 95 96 4 0 1 4 0 1 1 10 0 3 98 97 3 0 3 1 1 AU GRES Oi E Bsl Bs2 BC C 8-0 0-31 31-45 45-66 66-85 85-150 1 KINROSS 2 Oa E Bs C 8-0 0-18 18-50 50-125 4 76 Table 2.5. Selected chemical properties of soils in Chippewa hydrosequence. Fe Horizon LIMINGA A 0-7 E 7-11 Bhs 11-22 Bs 22-63 C 63-192 WAINOLA Oi 7-0 E 0-15 Bhs 15-43 BC 43-77 C 77-118 KINROSS 1 Oa 10-0 E 0-8 Bhs 8-19 Bg 19-34 C 34-56 PYR OX — - 0.2 3.2 1.2 0.2 0.2 3.2 2.2 0.2 A1 DITH 2.2 6.2 6.2 2.2 C pH PYR OX DITH g/kg _ _ 0.2 4.2 2.2 1.2 0.2 4.2 5.2 1.2 0.2 3.2 4.2 1.2 4.2 4.2 4.2 4.2 5.2 4.2 4.2 5.2 5.2 6.2 5.2 4.2 5.2 5.2 5.2 4.2 5.2 5.2 5.2 6.2 KC1 g/kg 0.2 5.2 1.2 1.2 0.2 6.2 1.2 1.2 4.2 14.2 3.2 4.2 0.2 7.2 1.2 1.2 0.2 9.2 2.2 1.2 0.2 8.2 1.2 1.2 5.2 3.2 4.2 5.2 5.2 0.2 1.2 0.2 0.2 0.2 0.2 0.2 1.2 1.2 1.2 1.2 3.2 0.2 2.2 3.2 2.2 0.2 2.2 3.2 3.2 0.2 2.2 32 32 4.2 4.2 4.2 4.2 52 PYR = pyrophosphate extraction OX = ammonium oxalate extraction DITH=sodium dithionite extraction pH: 1:1 1 NKChsoil, l:H 20:soil C = total carbon H20 2.2 7.2 3.2 8.2 7.2 79 Table 2.6. Selected chemical properties of soils in Cheboygan hydrosequence. A1 Fe PYR Horizon OX DITH PYR . gAg mir— RUBICON A 0-10 E 10-34 B sl 34-50 Bs2 50-113 C l 113-153 C2 153-172 AUGRES Oi 8-0 E 0-31 B sl 31-45 Bs2 45-66 BC 66-85 C 85-150 KINROSS 2 Oa 0-8 E 0-18 Bs 18-50 C 50-125 pH OX DITH C KC1 H20 53 4.4 4.3 43 4.6 5.9 5.4 53 53 5.4 0.0 1.6 0.3 0.2 0.9 0.1 1.7 0.5 03 13 1.3 2.1 13 3.1 4.7 0.0 1.8 1.1 1.3 1.2 0.0 1.7 1.6 0.6 1.6 0.0 22 0.9 13 1.1 0.0 0.1 0.2 0.2 0.1 0.0 0.1 0.3 0.2 13 0.2 0.4 1.7 1.4 1.6 0.2 2.0 2.0 1.2 0.2 0.1 1.7 2.0 2.1 1.2 0.2 23 1.8 13 03 3.9 4.0 43 4.1 43 43 4.8 4.9 4.9 4.9 5.1 53 0.1 0.9 0.1 0.1 1.0 0.0 0.4 5.0 0.7 0.2 1.2 0.2 0.1 13 0.2 03 1.0 03 3.9 4.0 43 4.4 4.6 5.0 5.4 5.4 o ii X rY K = pyrophosphate extraction DITH = sodium dithionite extraction pH: 1:1 IN K C ltso il. l:lH 2 0 :so il C = total carbon g/kg 4.2 1.5 2.8 5.2 1.8 ammonium oxalate extraction SAND AND SILT MINERALOGY Medium sand and fine sand were the dominant sand size fractions in Cheboygan and Chippewa Counties, respectively. These sand fractions were used for mineralogical studies. Results of the petrographic analysis of the major sand fraction of the profiles under study are presented in Table 2.7. Examination of the data portrays the following points. A/ Quartz accounts for more than 70% but less than 80% of the mineralogical composition 60 of these two hydrosequences. Hence, these soils are classified as having mixed, not siliceous, mineralogy at the family level (Soil Survey Staff, 1975). In all instances the solum quartz content is higher than the parent material. This could be due to a loss of weatherable minerals in the solum that causes an increase in the distribution of quartz. Another explanation is a physical break down of coarser fractions to finer size fractions. B/ The potassium feldspar(K-feldspar) distribution within a profile is somewhat uniform indicating resistance to physical and chemical weathering. The plagioclase feldspar contents, however, increase with depth indicating its susceptibility to weathering especially near the soil. This statement is supported by relatively equal amounts of these feldspars in the parent material of the Chippewa hydrosequence. Except in the well drained pedons, K-feldspar tends to be higher in the eluvial horizons than the horizons below. This may be explained in terms of weathering in the E horizons or due to physical breakdown of coarser fractions to finer fractions. C/ Chert and rock fragment contents tend to be slightly higher in the sola than in the parent materials. This could be due to high intensity of weathering (physical) at or near the surface. D/ Heavy minerals (pyroxene, hornblende and opaque) constitute 5-9% of the mineral composition in the parent materials. Their contents decrease to 3-5% in the surface due to weathering. E/ Altered (weathered) minerals are those which have aggregate extinction under strong light but which are almost opaque under normal light. These minerals were not identified as a specific mineral. These minerals are mainly identified in the solum as an alteration product of other weatherable minerals. F/' Sericite which is one of the abundant minerals next to quartz and K-feidspar shows the same distribution pattern as the weatherable minerals. As in the fine loamy hydrosequence, these minerals were observed under petrographic microscope as shiny, fine grained particles on feldspar grains and also as individual grains. G/ The distribution of the minerals identified in the parent material of each hydrosequence shows similar abundance indicating parent material uniformity within each hydrosequence. Other minerals like mica, hematite and detrital chlorite were present in very small quantities (<1%). Separation of Na and Ca feldspars in the identification process was not attempted. Another study in Michigan (Franzmeier et al, 1963) found no evidence of Kfeldspar weathering in the fine sand fraction, whereas chlorite and altered grains show the most evidence of weathering. Heavy minerals identified in their study constituted 0.5-6% of the total mineral composition. Collins (1971) found orthoclase and albite as a 61 predominant feldspars and heavy mineral contents of less than 7% in the fine sand fractions. The common results in all these studies are: 1/ K-feldspar shows little or no evidence of weathering, 2/ heavy minerals comprise a small percentage (<7%) of the total mineral assembladge and 3/ muscovite or biotite are less than 0.5% in the fine and medium sand fractions. Identification of minerals present in the coarse (20-50pm) and fine (2-20pm) silt fractions was based on JCPDS (1974) powder diffraction data for standard minerals. Minerals identified in both silt fractions are mainly quartz, plagioclase, and K-feldspar (Fig. 2.8). Chlorite was present in low amounts mainly in the fine silt fraction of Cheboygan hydrosequence. Quartz is the only mineral found in high amount. Both feldspars are present in a relatively equal proportions but low amounts. The productivity of the forest vegetation in these sandy soils is mainly maintained by "tight" cycling of available nutrients as Miller et al. (1979) pointed out in their study. Unlike the fine-loamy soils which has relatively higher silt and clay contents, these sandy soils mainly depend on the weathering of weatherable minerals and recycling of available nutrients. MINERALOGY OF CLAY FRACTION The results of XRD analyses of the <2pm clay fraction for the two hydrosequences are summarized in Table 2.9. Figures 2.6-2.11 depict the X-ray diffraction patterns obtained for these soils. Based on the X-ray diffraction patterns, the distribution of layer silicates shows the same pattern in all pedons. The non clay minerals identified were quartz and feldspars. Doublet peaks, 3.18A - 3.21 A, were the most intense part of the feldspar spectrum particularly in the parent material. The phyllosilicates of the parent material are chlorite, mica, kaolinite and small amount of vermiculite. XRDs of the C horizons show relatively high count level base lines, probably due to the presence of amorphous materials (Mehra and Jackson, 1960). Smectite is the dominant layer silicate found in the E horizons of all pedons. Other layer silicates present in this horizon are mica and kaolinite with the exception in the Kinross pedons that have vermiculite in addition to mica and kaolinite. The E horizon of Kinross pedon from the Chippewa hydrosequence has some random interstratification of smectite-vermiculite as shown by broadening towards the high angle of the 18A peak. The predominant clay mineral present in the spodic horizons is vermiculite. Other clay minerals present are mica, kaolinite, and chlorite. Random interstratification of vermiculite and mica in the spodic horizon of the Liminga pedon is shown by a broad 62 Table 2.7. Mineralogy of the sand fraction. SAND MINERALOGY HORIZON CP K FEL Ns/Cs FEL. ROC. SER FRA. CHE PY OPA. AM ALT 2 2 2 9b CHIPPEWA LIMINGA 2 2 2 2 1 2 3 3 3 2 2 1 2 2 1 2 1 1 2 2 2 1 2 1 I 6 8 8 3 I 2 I 3 2 2 2 2 2 6 8 1 1 1 1 2 2 1 I 2 2 2 2 2 9 2 2 1 3 • 2 2 6 6 1 I 2 2 7 1 1 2 2 2 5 1 1 1 I 10 2 2 1 3 ■ 73 10 10 2 2 6 6 2 2 9 5 7 I 1 1 2 1 1 2 1 2 i\i 1 2 2 10 3 3 5 6 6 2 1 1 I 1 1 2 2 7 I I 1 2 2 I • E 78 5 77 5 2 2 4 Bhi C 71 5 5 6 E Bhs 78 7 76 5 2 2 Cl C3 73 72 6 6 5 4 E Bhi 75 75 7 2 C 72 5 5 3 5 73 74 10 9 70 75 4 WAINOLA 4 1 2 • • KINROSS I CHEBOYGAN RUBICON E Bs a AUGRES E Bs C I 1 2 - KINROSS 2 E Bs 73 74 C 72 9 9 Q Z = Q uartz; k F el= K -feldspar; Ser=Sericite: R oc. F rag= R ock fragm ent; C he=chert: Opa=Opaque P y -P y ro x en e; A m =A m phibole; A lt=A ltered * = absent I 1 2 63 Table 2.8. Relative amounts of minerals in silt fractions. FINE SILT HORIZON PLAGIOCLASE K-FELDSPAR COARSE SILT QUARTZ PLAGIOCLAS E K-FELDSPAR. QUARTZ CHIPPEWA LIMINGA E L L H L M H Bhi M L+ L L H H L L L+ L H H L L H L+ L H Bhi L M L L H H L+ L L+ L+ H H Cl a L L+ L L+ H H L+ L+ L+ M H H E Bhi L+ L H L L H L+ L L H L+ C M M H L L H Cl C3 WAINOLA E KINROSS I H CHEBOYGAN RUBICON E L+ L H M M H Bs L+ H Cl L L L L L L L L H H L L H L M C3 L H H AUGRES E Bi C KINROSS 2 L L H Lt L+ • L • H • L+ L H L+ H H H E L+ L H M L+ Bi M L+ L H L* H L H L+ L L H C H-HIGH M-MEDIUM L-LOW • = ABSENT 64 Table 2.9. Relative amounts of clay minerals(<2pm). HORIZON SMECTITE CHLORITE VERMICULITE MICA KAOLINITE QUARTZ CHEBOYGAN RUBICON E XXXXX XX X/ X/ Bs XXXXX X/ XX X / Cl XXX XXX X/ XX / C3 XXX XXX X/ XX / AUGRES E XX XXXXX X/ X / Bs XXXXX XX XX X C XX/ X XXX XX/ X KINROSS 2 E XXXX XX/ X X X / Bs XX XXX/ X/ XX X C XX/ X XX/ XXX X CHIPPEWA LIMINGA E xxxxxx XX X/ / Bhs X XXX/ X/ XXX/ / XX/ Cl XX X XXX X/ XX/ X XX/ X C3 XXX WAINOLA E XX/ XXXXX X/ X XX Bhs X XXX/ XX X/ XX Cl XX X/ XXX X/ C3 XX X XXX XX/ X/ KINROSS 1 E X XXXXX X XX X Bhs XXXX XX/ XX X/ C XX/ XX/ X XX/ X/ - = not detectable / = Approximately 5% X = Approximately 10% 65 angle between 14A and 10A. This broad peak shifts to 10A with 300°C heat treatment. This indicates moderately weathered muscovite initially produces mixed-layer muscovitevermiculite and, as weathering proceeds, vermiculites and then smectite. Partial intergrade of 14A and 10A (chlorite and vermiculite or mica) with discrete chlorite appears in the Bs and C horizons of the Kinross pedon from the Cheboygan hydrosequence. Interstratification and hydroxy interlayering are relatively common especially in the spodic horizons. The presence of hydroxy-interlayering is evidenced by the fact that the vermiculite in the spodic horizons collapsed partially after K saturation at room temperature and shifted to 10A with some broading to the high angle after 300°C heat treatment. According to Bamhisel and Bertsch (1989), vermiculite containing little or no hydroxyinterlayering will collapse to 10A upon K saturation at room temperature. As shown by April et al. (1986), the necessity to heat the sample to force K into the interlayer and to dehydroxylate the interlayer is characteristic of hydroxy-interlayered vermiculite. In addition to spodic horizons this phenomenon is also observed to some extent in the E horizon of the fine clay of all pedons (Fig 2.12 and 2.13). Discrete mica occurs mostly in the C horizons of all pedons and in small amounts in the solum. The fact that the 10A peaks tend to decrease in height and symmetry towards the surface horizon is further evidence of the alteration of mica in the more acid surface horizon. The chlorite and in some cases kaolinite peaks also tend to decrease in height and symmetry with decreasing depth. In fact, very little discrete chlorite occurs in the spodic horizons and is totally absent in the E horizons. This could be due to the alteration of these minerals, probably by the loss of structural Al and Fe and desilication (Ross et al.,1982; Ross and Mortland, 1966). The absence of the 1.54A 060 diffraction peak in the spodic horizon indicates ail clay minerals present are dioctahedral and the chlorite present is pedogenic. The dioctahedral chlorite present in the C horizon might have weathered to amorphous or decomposed to soluble products by the cheluviating action of organic acids (Kodama and Schnitzer, 1973), unless chlorite vermiculitization results in a change toward dioctahedral structure as found in New Brunswick Spodosol (Ross et al.,1982) and demonstrated in the laboratory by Ross (1975). Another explanation could be due to low concentration of trioctahedral chlorite and according to Bamhisel and Bertsch (1989), if the concentration is low, peak intensities of chlorite may not be sufficient for identification. Vermiculite is the predominant clay mineral in the spodic horizon. The intensity of vermiculite reaches the maximum in the spodic horizons and disappears in the eluvial horizons of all the pedons except in the Kinross pedons. Vermiculite is present as discrete and also as randomly interstratified smectite-vermiculite in the poorly drained pedons. This indicates alteration of vermiculite in the poorly drained pedons is still in progress and also 66 suggests intensity of pedogenic weathering processes increases from poorly drained to well drained soils. The alteration of mica to vermiculite depends on a strong leaching environment to remove solution K and mixed-layer muscovite/vermiculite is common and often dominate in the weathering sequence (Sawhney, 1989). The alteration sequence mica -» vermiculite -» smectite has been observed in Spodosols (Gjems, 1960; Ross and Mortland, 1966; Franzmeier et al., 1968; Douglas and Trela, 1979; among others). These same authors and Komameni et al. (1985), among others, have attributed the transformation of smectite from mica due to: loss of K as in vermiculite formation; dealumination of the tetrahedral sheet; and silication of the tetrahedral sheet. Those workers mentioned above have considered the presence of smectite in the Spodosols as an advanced stage of weathering. The mineral distribution within a profile in these hydrosequences suggest the following weathering sequence: sericite/muscovite -> vermiculite -» smectite with or without mixed-layer formation. Trioctahedral chlorite in this study is assumed to decompose while pedogenic chlorite could have formed by neoformation which might have weathered to vermiculite in the surface horizons. CONCLUSIONS These Spodosol hydrosequences in Michigan are low in weatherable minerals. The sand mineralogy is dominated by quartz followed by small amounts of feldspar and sericite. The silt fraction is devoid of weatherable minerals including sericite. It is dominantly composed of quartz aim small amounts of feiaspar. Differences in drainage do not indicate a particular distribution of sand and silt minerals. The clay mineral distribution indicates the parent material is composed of chlorite, mica, kaolinite and small amounts of vermiculite. Smectite, dioctahedral beidellite, is the dominant layer silicate found in the E horizons of all pedons. The intensity of smectite peak and the presence and absence of vermiculite in the E horizons, suggest increasing weathering intensities from poorly drained to well drained soils in these hydrosequences. Based on the clay mineral distribution of the total clay fraction the following weathering sequence is observed: Sericite/muscovite -» vermiculite mixed-layer formation. smectite with or without 67 K 550 >H 55 Mg-Glg UJ at Bhs ► 55 UJ at K 550 i/i z K 300 UJ UJ A Mg-Glg at > H 55 10 20 D egrees 28 Fig. 2.5. X-ray d i f f r a c t o g r a m s of cla y f r a c t i o n (<2Mm) of t h e liminga pedon. 66 K 550 CO z uj K 300 UJ Mg-Gly oc K 550 > £ CO K 550 ► CO K 300 Mg-Gly ,.JuA J1 Li ... n > H K 550 CO K 300 Mg-Gly 0 10 20 30 Degrees 20 Fig. 2.6. X-ray d i f f r a c t o g r a m s of cla y f r a c t i o n (<2jjim) of t h e Wainola pedon. 69 1 0 -|----------0 . J U .u .l A . ft ft 1----------- 1----------- 1----------10 20 " I 'll 'I i Lu Um r 30 D egrees 28 Fig. 2.7. X-ray dif f r a c t o g r a m s of clay fraction (<2^m) of the Kinross pedon in Chippewa hydrosequence. K 300 V) Mg-Glg (0 z K 300 UJ Ot K 300 (ft z UJ Y•IrnW w UJ tn z UJ H UJ QC 10 20 D egrees 2B Fig. 2.8. X-ray d i f f r a c t o g r a m s of cla y f r a c t i o n (<2jim) of t h e Rubicon pedon. 91 I\ K 5 5 0 \fcii■■■■ ' flmidUftdr R EL. V , Mg-Glg IlK 550 REL. INTENSITY \V \k 3 0 0 REL. INTENSITY Mg-Glg V k.*Au,.i u i ./ K 1"“ Mg-Glg I 0 ' i 10 ■ i 20 • I 30 D egrees 20 Fig. 2.9. X-ray diffr a cto gram s of clay fraction (<2jjim) of the AuGres pedon. REL. INTENSITY REL. INTENSITY REL. IN T E N S IT Y 92 \J*aJ'v T,,*lLJ v * 0 I------------------1----------------- r ---------------- 1----------------- 1------------------ T 0 10 20 30 D egrees 20 Fig. 2.10. X-ray d iffr a cto gram s of clay fraction (<2jim) of the Kinross pedon in Cheboygan hydrosequence. 93 R E L . IN TEN SITY KINROSS K 550 K 300 IC - a D REL. INTENSITY M G -G LY K-550 ROUSSEAU REL. INTENSITY K -550 K -300 K-AD Mg-Gly D egrees 2 0 Fig. 2.11 X-ray d iffr a cto gram s of fine clay fraction ( — ; i t . . i . . ■» r \ n a o n o 1. I\C Y id C U J U i y 1 ^ 0 * + . OV^O" USDA. U.S. Govt. Printing Office, Washington, DC. Whittig, L.D. and W.R. Allardice. 1986. X-ray diffraction techniques. In C.A.Black et al., (2nd. ed.) Methods of soil analysis, Part 1-Physical and mineralogical methods. Agronomy 9:331-359. 99 CHAPTER 3 CHARACTERIZATION OF SMECTITE IN A SPODOSOL AND A MOLLISOL ABSTRACT Mineralogical studies have revealed smectites in the acidic E horizons of Spodosols and in basic Btg horizons of Mollisols. This study was conducted to characterize the smectite in fine-clays from the two environments in terms of layer charge using the Greene-Kelly test, re-expansion property, alkylammonium method (AAM), and elemental analysis. In both cases, the smectite was dioctahedral of the beidellitic type. Structural formulas calculated from the chemical data had 0.47 and 0.51 molc/half unit cell formula for the Spodosol and Mollisol clays, respectively. The mean layer charge determined by AAM was 0.41 and 0.43 molc/half unit cell formula for the E and Btg horizons, respectively. This is lower than the value determined by the structural formula. The paraffin type configuration indicates the presence of unweathered mica cores in these samples. Both samples, when K-saturated, heated at 300UC, and glycerated, re-expanded to 14A suggesting a layer charge value between 0.39 and 0.43 and part of the clay that remained at 10A indicates a charge higher than 0.43. The high octahedral iron in the Btg horizon clay is similar to a tetrahedrallycharged mica. In both samples soil mica weathers to beidellite with vermiculite being an intermediate product. 10 0 INTRODUCTION Many soil mineralogists believe that smectite is unstable in acidic environments and restrict its formation mainly to basic environments (Douglas, 1982). However, others (Brown and Jackson, 1958; Gjems, 1960; Franzmeier et al., 1963; McKeague, 1965; Ross and Mortland, 1966; Brydon et al., 1968; Malcolm et al., 1969; Coen and Arnold, 1972; Kapoor, 1972; Nash, 1979; Douglas and Treala, 1979; Douglas, 1982) among others, have reported the weathering of mica and chlorite through vermiculite, to smectite in eluvial or albic horizons of Spodosols. Ross and Mortland (1966) reported beidellite to be the smectite variety in the Spodosol E horizon of their study. Beidellite was also found to be the smectite variety in Mollisols in Iowa ( Laird et al., 1988) and Minnesota (Badraoui et al., 1987). Mineralogical studies in sandy and fine-loamy hydrosequences in Michigan have revealed smectite in the acidic E horizon of Spodosols and in the basic Btg horizon of Mollisols. Weathering of mica to smectite through vermiculite as an intermediate stage is thought to be the dominant mechanism of smectite formation in Michigan soils. The purpose of this study is to characterize these smectites in terms of layer charge using the Greene-Kelly test, re-expansion properties, alkylammonium exchange, and elemental analysis. 101 M ATERIAL AND METHODS The E and Btg horizons from pedons of the Rousseau (sandy, mixed, frigid, Entic Haplorthods) and Brookston (fine-loamy, mixed, mesic, Typic Argiaquolls) soil series, respectively, were selected for detailed clay mineralogy studies. Soil samples were taken from each horizon and the <2-mm fraction was ultrasonically dispersed and separated into sand, silt, and clay fractions after removal of cementing agents (organic matter removed by the H 2O 2 oxidation method; carbonates removed using sodium acetate adjusted to pH 4.5; and sesquioxides were extracted with a solution of citrate-bicarbonate-dithionite (CBD) using Jackson's (1979) procedure. The total clay fractions (<2pm) were further separated into fine (< 0.2pm) and coarse (2-0.2pm) fractions. The fine clay fraction was again pretreated to remove cementing agents according to Jackson (1979). Sodium hypochlorite was also used to destroy any residual organic matter in selected samples (Anderson, 1961). The fine clay fraction was saturated with 1M NaCl followed by dialysis to remove excess salts. The clays were then ultrasonically dispersed, frozen, and freeze-dried. The CEC was measured by displacing Ca from Ca-saturated clays with 0.5N MgCl2. The Ca concentrations were determined by flame emission photometry. Surface areas of the fine clays were determined using the ethylene glycol monoethyl ether (EGME) adsorption method (Carter et al., 1986). Fine clay fractions of the soil clays were saturated with Mg or K by washing 5 times with 1M M gCh or 1M KC1, respectively, followed by sequential washes with deionized H 2 O. Oriented clay films were obtained by allowing suspensions containing 25 mg of fine clay to air dry on 27-by 46-mm glass slides. Magnesium-saturated specimens were analyzed after air drying and glycerol solvation. K-saturated specimens were x-rayed after air drying, heating to 300 °C for 2h, and heating to 550 °C for 2h. Lithium charge reduction and glycerol solvation were used to determine charge location in the smectites. Reference clay samples were included for comparison. Separate Li-saturated samples were X-rayed (1) after glycerol solvation at room temperature and (2) after heating at 300 °C overnight followed by glycerol solvation. This method was used to distinguish montmorillonite from beidellite (Greene-Kelly, 1953). Potassium-saturation, heat treatment, and glycerol solvation were used to estimate total layer charge of the soil smectites. Reference clay samples were included for comparison. Clays were K-saturated and heated at 300 °C overnight followed by glycerol solvation. These samples were checked for expansion by XRD to determine the total net layer charge of the smectite. Charge estimation is based on the relationship between re­ 102 expansion properties and total layer charge (Schultz, 1969; Ross and Kodma, 1984; Malla and Douglas, 1987). Layer charge estimation was also based on XRD measurement of the (001) interplanar spacings of alkylammonium-clay complexes (Lagaly and Weiss, 1969; Ruehlicke and Kohler, 1981). Alkylammonium chloride salts with even-numbered alkyl chain lengths of 6 -18 C atoms were used. The Na- saturated and freeze-dried fine clay samples (25 mg) were treated with the various alkylamine hydrochlorides as described by Senkayi et al. (1985). All X-ray diffraction analyses were conducted with a Phillips XRD using C uK a radiation at 35 kV and 20 mA, a theta compensating slit, and a diffracted beam graphite monochromator. All samples were step-scanned using 0.05° 20 step intervals and a count time of 5 seconds per step. Total chemical analyses of the fine clay fractions were performed using fused glass disks prepared by melting a mixture of 1.000 g of an homogenized, freeze-dried clay powder at 950 °C with 9.000 g of lithium tetraborate flux and 0.16 g of ammonium nitrate in Pt-Au alloy crucibles (Hagan, 1982; Bower, 1985; and Bower and Vaentine, 1986). Analyses were performed with a Rigaku S-Max automated XRF system in the Geology department of Michigan State University. Abundances of major element oxides were calculated by the fundamental parameters matrix correction method (Criss, 1980). A U.S.G.S. reference standard (W-2) was analyzed to provide a check on the accuracy and precision of the analyses. RESULTS AND DISCUSSION The X-ray diffraction patterns of die fine ciay sampies from the E and Btg horizons of the Spodosol and the Mollisol, respectively, are presented in Fig. 3.1. The dominant clay mineral in both fine clay fractions is smectite as shown by a strong 17.8 A peak (Fig. 3.1). Based on XRD, the presence of a small amount of mica in the E horizon sample is indicated by a low-angle shoulder on the second-order smectite 9.0A peak and by the presence of peaks at 5A and 3.34A. A small amount of mica is also evident in the fine clay fraction of the Btg horizon from a broad peak at 9-10A and by peaks at 5A and 3.34A. A small amount of kaolinite is indicated in both clays by a 7.2A peak which persists after K-300 °C treatment, but disappears after K-550°C treatment. Some random interstratification of smectite and mica in the Mollisol clay (Btg) is indicated by the diffuse nature of the (001) XRD reflections and the absence of a rational series of higher-order (001) reflections (Reynolds, 1980; Laird et al., 1988). The K2 O content of the fine clay fraction of the E and Bt samples (Table 3.1) indicates that approximately 18% and 26% of 103 these samples are mica, respectively, based on the K 2O content of muscovite which is 10% (Reynolds, 1980) and the assumption that all of the K2O content in these samples is derived from mica. The (060) powder diffraction patterns of both clays show a b-dimension spacing of 1.50A, which is characteristic of dioctahedral 2:1 layer silicate clays. The fme clays were also Li+-saturated and heated to test for the Hofmann-Klemen (1950) effect. Lithiumsaturated and air-dry samples show 12.5A peaks that are indicative of smectite (water monolayer). Clay samples Li-saturated and heated to 300 °C show a diffraction peak of 10A. However, both clays (E and Btg) re-expanded to 17.8A when glycerated (Fig. 3.2). This test indicates that more of the charge is located in the tetrahedral sites than in the octahedral sites and the smectite phase has a charge distribution like beidellite (GreeneKelly, 1955). Differentiation of vermiculite and high-charge smectite from low-charge smectite in soils based on K-saturated clays and solvating them with a glycerol-ethanol solution was employed by Barshad (1960). Schultz (1969) found a proportional relationship between the re-expansion and total net layer charge of smectite that had been K-saturated, heated to 300 °C, and solvated with ethylene glycol. Recently, Malla and Douglas (1987), using Schultz's (1969) principle of re-expansion, have classified minerals into four layer-charge groups based on re-expansion. Based on the re-expansion behavior of the K-saturated fine clays of the E and Btg horizons, the expanded clay in these samples is smectite. Potassium-saturated clays that have been heated at 300 °C show a diffraction peak at 10.1 A However, after glyceration, these samples re-expanded to only 14.3A.(Fig. 3.3) According to Schultz (1969), the reexpansion of some of the K-saturated, heated (300°C) samples by glycerol to only 14A suggests the presence of beideiiite as did the Greene-Keiiy te st Some of the ciay remains at 10 . 1A after glycerol solvation indicating a higher layer-charge component is also present. In contrast, low-charge smectite re-expands to 17A (Schultz, 1969). According to the layer charge groupings of Malla and Douglas (1987) the samples in this study have a layer charge between 0.46 and 0.57 per half u n itcell.. The presence of relatively high- and lowcharge smectite indicates the heterogeneous distribution of charge in these soil clays. 104 I7 .8 A E (< 0 .2 j im ) |l4 .4 A 3.34A 10.16A 5.05A 3.57A 7.2A 3*57A INTENSITY li K-300 C RELATIVE -2S°C Mg-Glyc. K-30CTC U K -2S°C j r,w^w. Mg-sat. 10 20 30 DEGREES 2 8 F ig .3 .1. Mg-sat. 10 20 30 40 DEGREES 2 0 X-ray diffractograms of the fine clay (<0.2|jun) fractions of the Spodosol E and Mollisol Btg horizons (treatments specified at right). 105 Figure 3.4 shows X-ray diffraction patterns of the E and Btg horizon clays that were treated with selected alkylammonium hydrochlorides. In both clays, the XRD patterns indicate the presence of kaolinite (7.2A), mica ( 10 A) and smectite (13.4A-21 A) Vermiculite should expand to 18A or more with C 6 alkylammonium treatment; no vermiculite is evident in either sample. The Spodosol clay (E) shows a low angle peak (28A-30A) with 16- and 18-carbon alkylammonium ions indicative of vermiculite. Similarly, the Mollisol (Btg) clay has peaks at 23.lA-25.4A with C12-C14 and 28-30A with C16-C18. The Btg sample C12 and C14 diffraction patterns contain two peaks representing low- and high-charge phases. According to Laird et al.( 1987,1988), Senkayi et al. (1985).and Badraoui et al. (1987), the low angle diffraction peaks were associated with K + depletion and expansion of soil mica by alkylammonium cations. The absence of vermiculite, the K content, and the presence of a weak 10 A peak in these clays suggest that they contain some un-weathered mica. Smectite in these clays may be a product of mica weathering. In Figure 3.5, the d(001) values for Spodosol E and Mollisol Btg fine clays are plotted against the chain lengths of the alkylammonium cations (nc). The smectite in both clays had basal spacings indicative of a monolayer at C 6 , a bilayer at C10-C14 and a paraffin complex configuration at C16-C18 (Lagaly and Weiss, 1969 and Laird et al., 1988). Mean layer charge was calculated using the method of Lagaly and Weiss (1976) using d-values of the alkylammonium-clay complexes in the monolayer (13.4-13.8A) to bilayer (17.5-17.7A) transition range. Because of the heterogeneous charge distribution in most soil clays, the procedure of Stul and Mortier (1974) was employed to calculate the mean layer charge. The mean layer charge was 0.41 and 0.43 mole per half unit cell for the Spodosol (E) and the Mollisol (Btg) fine clays, respectively. 106 ,7 !7* E (<0.2nm ) B tg (<0.2iim ) RELATIVE INTENSITY 17.5a 10 * 5* L i-3 0 0 C Li-300 C Li-3 0 0 C Li_2 5 C 10 20 DECREES 20 Fig. 3.2. IL L i-3 0 0 C Ll~25 C 20 DEGREES 20 XRD patterns of fine clays from the E and Btg horizons of Spodosol and Mollisol, respectively (treatments are specified at right). 3 .3 3 8 1 4 .3 4 8 3 .5 8 8 5 .0 X „ 5 .0 8 7.22, 1 7 .1 5 8 107 RELATIVE INTENSITY E (< 0 .2 jim ) 3 .3 4 8 GLY. GLY. 10 20 DEGREES 2 0 30 40 0 10 20 30 DEGREES 2 0 Fig 3.3. XRD of the fine clay from E and Btg horizons of the Spodosol and Mollisol, respectively. Re-expansion properties o f the fine clays (treatments specified a t . right.). 40 ioe E (< 0 .2 u m ) RELATIVE INTENSITY Btg (< 0 .2 jim ) 30 0 DEGREES 20 Fig. 3.4. 25 30 DEGREES 2 0 X-ray patterns of the (<0.2pm) soil clays from the E and Btg horizons of the Spodosol and Mollisol pedons, respectively, and treated with various alkylammonium hydrochlorides (carbon chain length specified at right). 109 100i 1001 E (<0.2iim ) PERCENT 50- ■ 50' ■ o 32.4% h— 1 0.3 37.2% 40 17.4% o<12 62.8% 50 50.2% 20- B tg (<0.2|im ) 60 20 00.3 1-----» - - 0.4 05 0.6 d (001) A LAYER CHARGE 35 ----- <------ 0.4 0-5 0.6 LAYER CHARGE - 0.29 0.50 13 0.50 13 • ■ -4 10 + lO 3 10 CHAIN LENGTH (n c ) Fig. 3.5. 20 3 10 19 20 CHAIN LENGTH (n c ) Basal plane spacings of alkylammonium complexes with the E and Btg fine clay Spodosol and Mollisol, respectively, and the corresponding layer charge distribution. 110 Table 3.1. Elemental analysis of the Ca++-saturated fine clay fraction (<0.2pm) from the E horizon of a Spodosol and the Btg horizon of a Mollisol. Based on 950°C weight, % Oxide E (Spodosol) Bt g (Mollisol) Si02 55.43 53.02 AI2 O 3 26.51 21.54 Fe 2 0 3 3.62 9.45 K2 0 1.85 2.63 MgO 3.57 3.52 CaO 0.06 0.10 MnO 0.03 0.03 Na 2 0 3.01 2.79 Ti02 1.62 0.50 P2O5 0.05 0.06 pH (H 20) 4.4 7.1 CEC 102 meq/lOOg 90 meq/lOOg Surface area 567.12 m2/g 497.30 m2/g The layer charge density of these fine clays was also determined from the structural formulas. Using the elemental analyses (Table 3.1),.the following structural formulas were calculated: Rousseau E Na0.477Ca0.005^A11.605Fe0.154Mg0349Mn0.002^Si3333A10.467^ 0 k / 0 H ^2 Brookston Btg N a ^ C a ^ C A J , . 150*'e0 J 90Mg0J 70M. 0 003XSi3 490A10JI0) O 10(OH )2 Assuming the mica in these samples is similar to Hower and Mowatt's (1966) Silver Hill illite, the mica content was subtracted based on the K 2O content. According to the structural formula, these soil smectites have a layer charge of 0.47 and 0.51 mole per half unit cell formula for the E and Btg samples, respectively. These layer charge values are higher than the mean layer charge values determined by the alkylammonium method which is typical according to Laird et al.(1987,1989). The chemical analyses indicate that the smectite in the Btg horizon of the Mollisol is an iron rich beidellite. From the XRD patterns of the Btg sample (Figure 3.1), the presence of randomly interstratified smectite and clay mica is evident. The presence of a small amount of mica in the fine clay fractions suggests the presence of unweathered mica, and that the smectite present might have been the Ill weathering product of mica. The layer charge values (0.41 and 0.43) determined by the alkylammonium method fall below the range (0.46 - 0.57) classified by Malla and Douglas (1987) based on re-expansion properties. However, the layer charge values determined from the structural formula fall within the range estimated by the re-expansion properties. The layer charge value 0.47 determined from the structural formula of the Spodosol fine clay is similar to the value of 0.49 obtained by Ross and Mortland (1966). The structural formulas, the re-expansion properties, the Greene-Kelly test, and the b-dimension value indicate that both smectites are beidellitic dioctahedral phyllosilicates. Smectite formation in the Spodosols is attributed to high leaching environment, low pH and high organic matter in the surface horizon. Below pH 4.5, Al is complexed by organic matter and hence the Al is not available in large amounts for the formation of polymers (Douglas, 1982). The high octahedral iron content in the fine clay of the Brookston Btg sample is similar to a tetrahedrally-charged illite. The oxidation of structural iron might have been one of the primary mechanism of layer charge reduction during weathering. A dioctahedral mica will always transform to a high-charge smectite, and the occcurence of a high-charge smectite is associated with dioctahedral mica (Egashira and Tsuda, 1983). The alteration of mica has probably occurred without a significant change in the tetrahedral sheet substitution (Badraoui et al., 1987). Vermiculite is the intermediate product in the transformation process, as shown in the total clay mineralogy and the weathering of mica to smectite might have occurred prior to deposition. Ross and Mortland (1966) concluded that soil mica in a Spodosol weathered to beidellite. However, weathering of chlorite, present in the total clay, may also have been a source for some of the smectites found in these samples. Robert (1973) pointed out that smectites derived from mica will inherit characteristics of mica such as K. fixation, and suggested the need to differentiate them from smectites found in geological deposits or formed by crystallization in the soil environment. Soils dominated by beidellite will fix K more readily than montmorillonitic soils (Ross and Mortland, 1966; Badraoui et al., 1987). As a result, identification of the right clay mineral is important to understand the dynamics of K in respect to the clay mineral present. LIST OF REFERENCES 113 LIST OF REFERENCES Anderson, J.U. 1961. An improved pretreatment for mineralogical analysis of samples containing organic matter. Clays Clay Miner. 10:380-388. Badraoui, M., P.R.Bloom, and R.H.Rust. 1987. Occurrence of high-charge beidellite in a Vertic Haplaquoll of northwestern Minnesota. Soil Sci. Soc. Am. J. 31:813-818. Barshad, I. 1960. X-ray analysis of soil colloids by modified salted paste method: in Clays and Clay Minerals, Proc. 7th Natl. Conf., Washington, D.C., 1958, Ada Swineford, ed., Pergamon Press, New York, 350-364. Bower, N.W. 1985. Optimization of precision and accuracy in X-ray fluoresence analysis of silicate rocks. Applied Spectroscopy, 39:697-703. Bower, N.W., and G. Valentine. 1986. A critical comparison of sample preparation methods for major and trace element determinations using X-ray fluorescence. Xray Spectrometry, 15:73-78. Brown, B.E. and M .LJackson 1958. Clay mineral distribution in the Hiawatha sandy soils of northern Wisconsin. Clays Clay Miner. 5:213-266. Brydon, J.E., H.Kodama, and G J.Ross. 1968. Mineralogy and weathering of the clays in the orthic Podzols and other podzolic soils in Canada. Trans. 9th Int. Congr. Soil Sci., 3:41-51. Carter, D.L., M.M.Mortland, and W.D.Kemper. 1986. Specific Surface: In A. Klute, (2nd. ed.) Methods of soil analysis, Part 1. Physical and Mineralogical MethodsAgronomy Monograph no. 9. 16:413-423. Coen, G.M. and R.W. Arnold. 1972. Clay mineral genesis of some New York Spodosols. Soil Sci. Soc. Am. Proc., 36:342-350. Criss, J. 1980. Fundamental parameters calculations on a laboratory microcomputer. Advances in X-ray Analysis, 23:93-97. Douglas, L.A. 1982. Smectite in acidic soils, p. 635-640. In H.Van Olphen and F.Veniale (ed.) Proc. Int. Clay Conf., Bologna and Pavia, Italy., 6-12 Sept. 1981. Elsevier Scientific Publishing Co., Amsterdam. Douglas, L.A. and J.Trela. 1979. Mineralogy of pine barrens soils, p. 85-109. In. R.T.T.Forman (ed.) Ecology of the pine barrens. Acadamic Press, New York. Egashira, K., and S.Tsuda. 1983. High-charge smectite found in weathered graitic rocks of Kyushu, Clay Sci. (Tokyo) 6:67-71. 1 14 Franzmeier, D.P. and E.P.Whiteside, and M.M.Mortland. 1963. A chronosequence of podzols in northern Michigan. Mich. State Univ. Quart. Bull., 46:37-57. Gjems, O. 1960. Some notes on clay minerals in Podzol profiles in Fennoscandia. Clay Miner. Bull. 4(24):208-211. Greene-Kelly, R. 1953. Irreversible dehydration in montmorillonite. Part II. Clay Miner. Bull. 2:52-56. Greene-Kelly, R. 1955. Dehydration of montmorillonite minerals. Mineral. Mag. 30:604615. Hagan, R.C. 1982. X-ray Fluorescence analysis: Major elements in silicate minerals. Los Almos Nat. Lab. Rep. LA-9400-Ms, 13pp. Hofmann, U., and R.Klemen. 1950. Verlust der Austauschfahigkeit von Iithium-Ionen an Bentonite durch Erhitzung. Z.Anorg. Allg. Chem. 262:95-99. Hower, J. and T. C.Mowatt. 1966. The mineralogy of illites and mixed-layer illite/montmorillonites. Am. Mineral., 51:825-854. Jackson, M.L.,1979. Soil chemical analysis - advanced course. (3rd. print). Published by author, Dept, of Soil Science, Univ. of Wisconsin, Madison. Kapoor, B.S. 1972. Weathering of micaceous clays in some Norwegian podzols. Clay Miner. 9:383-394. Lagaly, G. and A.Weiss. 1969. Determination of the layer charge in mica-type layer silicates, p.61-80. In L. Heller (ed.) Proc. Int. Clay Conf., Tokyo, Israel Univ. Press, Jerusalem. Lagaly, G. and A.Weiss. 1976. The layer charge of smectitic layer silicates, p. 157-172. In S.W. Baiiey (ed.) Proc. Int. Ciay Conf., Mexico City, Mexico, Applied Publishing, Wilmette, Illinois. Laird, D.A., A.D.Scott., and T.E.Fenton 1987. Interpretation of alkylammonium characterization of soil clays. Soil Sci. Soc. Am. J. 51:1659-1663. Laird, D.A., T.E.Fenton, and A.D.Scott. 1988. Layer charge of smectites in an Argialboll-Argiaquoll sequence. Soil Sci. Soc. Am. J. 52:463-467. Laird, D.A., T.E.Fenton, and A.D.Scott. 1989. Evaluation of the alkylammonium method of determining layer charge. Clays Clay Miner., 37:41-46. Malcom, R.L., W.D.Nettleton, and R.J.McCracken. 1969. Pedogenic fo rmation of montmorillonite from a 2:1-2:2 intergrade clay mineral. Clays Clay Miner., 16:405414. 115 McKeague, L.A. 1965. Properties and genesis of three members of the Upland catena. Can. J. Soil Sci. 45:63-77. Malla, P.B., and L.A.Douglas. 1987. Identification of expanding layer silicates: Layer charge vs. expansion properties, p. 277-283. In H. van Olphen and L.G. Schultz (ed.) Proc. Int. Clay Conf., Denver, CO. Clay Mineral Society, Bloomington, IN. Nash, V.E. 1979. Mineralogy of soils developed on Pliocene-Pleistocene terraces of the Tombigbee River in Mississippi. Soil Sci. Soc. Am. Proc. 43:616-623. Reynolds, R.C. 1980.1nterstratified clay minerals, p. 249-303. In G.W. Brindley and G. Brown (ed.) Crystal structures of clay minerals and their x-ray identification. Monograph 5. Mineralogical Society, London. Robert, M. 1973. The experimental transformation of mica toward smectite. Relative importance of total charge and tetrahedral substitution. Clays Clay Miner. 21:167174. Ross, G.J., and H.Kodama. 1984. Problems in differentiating soil vermiculites and soil smectites: Agron. Abstr: American Society of Agronomy, Madison, WI. p. 275. Ross, G J. and M.M.Mortland. 1966. A soil beidellite. Soil Sci. Soc. Am. Proc. 30: 337-343. Ross, G.J., C.Wang, A.L.Ozkan, and H.W.Rees. 1982. Weathering of chlorite and mica in a New Brunswick podzol developed on ill derived from chlorite-mica schist. Geoderma 27:255-267. Ruehlicke, G„ and E.E.Kohler. 1981. A simplified procedure for determining layer charge by the n-alkylammonium method. Clay Miner. 16:305-307. Schultz, G.L. 1969. Lithium and potassium adsorption, dehydroxylation temperature, and structural water coiiieui of aluminous smectites. Ciays Ciay Miner. 17:115-149. Senkayi, A.L., J.B.Dixon, L.R.Hossner, and L.A.Kippenberger. 1985. Layer charge evaluation of expandable soil clays by an alkylammonium method. Soil Sci. Soc. Am. J. 49:1054-1060. Stul, M.S., and W J.M ortier. 1974. The heterogeneity of the charge density in montmorillonites: Clays Clay Miner., 22:391-396. Weaver, C.E., and L.D.Pollard. 1973. The chemistry of clay minerals: Developments in sedimentology 15, pp. 55-77. Elsevier Scientific Publishing Co., Amsterdam, Netherlands. SUMMARY CONCLUSION The sand, silt, and clay mineralogy of the two fine-loamy hydrosequences of southcentral Michigan suggest the glacial till parent material is relatively uniform. Depth and duration of water tables were similar for soils with the same natural drainage class. Having similar parent material, vegetation, climate, and time, topography has produced some noticeable differences in distribution of minerals among and within profiles. Weathering of muscovite/sericite to smectite (beidellite) through interstratified vermiculite-smectite seems to be the dominant trend in the poorly drained Brookston pedon whereas direct transformation to vermiculite is the prevalent weathering mechanism in the other drainage classes. Except for the smectite in the Brookston pedon, the fine-loamy soils in this study do not show significant differences in sand, silt, and clay mineralogies with respect to natural drainage. The Spodosol hydrosequences in Michigan are very low in weatherable minerals. The sand mineralogy is dominated by quartz followed by small amounts of feldspar and sericite. The silt fraction is devoid of weatherable minerals including sericite. It is dominantly composed of quartz and small amounts of feldspar. Differences in drainage does not indicate a particular distribution of sand and silt minerals. The clay mineral distribution indicates the parent material is composed of chlorite, mica, kaolinite and small amounts of vermiculite. Smectite (dioctahedral beidellite) is the dominant layer silicate found in the E horizons of all pedons. The intensity of smectite peak and the presence and absence of vermiculite in the E horizons, suggest increasing weathering intensities from poorly drained to well drained soils in these hydrosequences. Based on the clay mineral distribution of the total clay fraction the following weathering sequence is observed: Sericite/muscovite -» vermiculite -» smectite.with or without mixed-layer formation. The structural formula, the re-expansion property, the Greene-Kelly test, and (060) XRD pattern indicate the smectite in the fine clay fractions of the Spodosol E and Mollisol Btg horizons are both dioctahedral phyllosilicates of beidellitic phase which is dominated by tetrahedral charge. The high octahedral iron content in the fine clay of the Btg horizon is similar to a tetrahedrally charged illite, and it seems the oxidation of structural iron might have been one of the primary mechanisms of layer charge reduction during weathering. Weathering of mica to smectite might have occurred prior to deposition. According to the 1 17 structural formula, these soil smectites have a layer charge of 0.47 and 0.51 mole per half unit cell formula for the E and Btg samples, respectively. These layer charge values are higher than the mean layer charge values determined by the alkylammonium method. APPENDIX A PEDON DESCRIPTIONS 119 Soil code: Ml Sampled as: Glossoboric Hapludalf, fine-loamy, mixed,mesic. Classification: Essex Township, Clinton county, NE 1/4 NW 1/4 NW 1/4 NW 1/4 NE 1/4 sec. 15 about 23 m south of Island Road, T. 8 N., R. 3 W. Geomorphic position: Summit position on moraine system, gently sloping. Drainage: Moderately well drained. Vegetation: Acer Saccharum. Tilia americana. and Ouercus rubura. Parent material: Calcareous glacial till. Sampled by: D. Cremeens and 0 . Doumbia. July 1,1981. Remarks: C2 sample collected with bucket auger. Thin lens (1 cm thick) of strong brown (7.5YR 4/6) loamy sand found at 85 cm depth. Less than 2% coarse fragments throughout pedon. A 0-12 cm. Dark grayish brown (10YR 4/2) silt loam (26% sand, 14% clay); moderate medium granular structure; friable; abrupt wavy boundry. E/B 12-26 cm. Yellow brown (10YR 5/6) and brownish yellow (10YR 6/ 6 ) silt loam (2 2 % sand, 18% clay); weak medium platy structure parting to moderate medium subangular blocky; friable; clear wavy boundary. B tl 26-40 cm. Dark yellow brown (10YR 4/4) silty clay loam (13% sand, 31% clay); moderate medium subangular blocky structure; firm; thin patchy dark grayish brown (10YR 4/2) clay films; gradual smooth boundary. Bt2 40-84 cm. Dark yellowish brown (10YR 4/4) silty clay loam (15% sand, 31% clay); moderate medium angular blocky structure; firm; few fine faint yellowish brown (10YR 5/4) mottles; thick continuous dark grayish brown (10YR 4/2) clay films; clear wavy boundary. BC 84-108 cm Dark yellowish brown (10YR. 4/4) silt loam (17% sand, 26% clay); strong coarse platy structure parting to moderate medium angular blocky; firm; few medium faint yellowish brown (10YR 5/4) mottles; thick continuous dark grayish brown (10YR 4/2) clay films; moderately effervescent; clear wavy boundary. Cl 108-165 cm. Grayish brown (10YR 5/2) silt loam (16% sand, 22% clay); strong coarse platy structure parting to moderate medium subangular blocky; very firm; few medium faint yellowish brown (10YR 5/4) mottles; strongly effervescent. C2 165+cm. Brown (10YR 5/3) silt loam (15% sand, 23% clay); very firm; common fine faint yellowish brown (10YR 5/4) mottles; strongly effervescent. 120 Soil code: Cl Sampled as: Capac, Aerie Ochraqualf, fine-loamy, mixed, mesic. Classification: Aerie Haplaquept, coarse-loamy, mixed, mesic. Location: Essex Township, Clinton County, MI. NW1/4NW1/4NE1/4NE1/4NW1/4 sec.15 about 30 m south of Island Road, T. 8 N„ R. 3 W. Geomorphic position: Foot slope position on moraine system, gently sloping. Drainage: Somewhat poorly drained. Vegetaion: Acer rubrum. Acer saccharum. Ouercus rubura. and Carva ovata. Parent material: Calcareous glacial till. Sampled by: D. Cremeens. July 2,1981. Remarks: A horizon could have been divided in to A 1 and A2 based on size of structural units. Augered to approximately 2.5 m no C2 horizon detected.water table at 1.8 m. About 10% coarse fragments throughout pedon except BC which contained slightly more and 2C which contained approximately 5%. A 0-22 cm. Very dark gray (10YR 3 /1 ,10YR 5/2) loam (42 % sand, 19% clay); moderate medium granular structure; friable; clear wavy boundry. B/E 22-32 cm. Brown (10YR 5/3) and light yellowish brown (10YR 6/4) sandy loam (54 % sand, 13% clay); weak fine subangular blocky structure; friable; few fine faint mottles; clear smooth boundry. B tl 32-53 cm. Brown (10YR 5/3) loam (49 % sand, 18 % clay); moderate medium angular blocky structure; firm; moderate medium distinct strong brown (7.5YR 4/4) and few medium faint grayish brown (10YR 5/2) mottles; medium continuos dark brown (7.5YR 4/2) clay films; clear wavy boundry. Bt2 53-97 cm. Dark grayish brown (2.5Y 4/2) loam (48 % sand, 16 % clay); moderate medium subangular blocky structure; friable; common medium distinct strong brown (7.5YR 4/4) and few medium distinct brown (10YR 5/3) mottles; medium discontinuous grayish brown (10YR 5/2) clay films; gradual smooth i/V U U U l j # BC 97-125 cm. Dark grayish brown (2.5Y 4/2) sandy loam (57 % sand, 14 % clay); weak medium subangular blocky structure; friable; many medium prominent strong brown (7.5YR 5/6) and common fine distinct gray (N 5/0) mottles; thin discontinuous grayish brown (10YR 5/2) lining root channels; gradual smooth boundary. 2C 125 + cm. Olive brown (2.5Y 4/4) silt loam (28 % sand, 19 % clay); moderate coarse platy structure breaking to weak coarse angular blocky; firm; common coarse distinct gray (N 5/0) and many fine distinct dark yellowish brown (10YR 4/6) mottles; moderately effervescent 121 Soil code: PI Sampled as: Parkhill, Mollic Haplaquept, fine-loamy, mixed, mesic. Classification: Humic Haplaquept, fine-loamy, mixed, mesic. Location: Essex Township, Clinton County, NE1/4 NE1/4 NW1/4 NE1/4 NW1/4 sec. 15 about 38 m south of Island Road, T. 8 N., R. 3 W. Geomorphic position: Toe slope position on moraine system, nearly level Drainage: Poorly drained. Vegetation: Fraxinus americana. Ouercus alba. Ouercus bicolor, and Cvoerus esculentus. Parent material: Non calcareous glacial till. Sampled by: D. Cremeens and G. Lemme. July 3,1981. Remarks: No carbonates detected down to 2.4 m. C horizon delineated due to structureless condition. Approximately 5-10% coarse fragments throughout pedon. A 0-23 cm. Very dark grayish brown (10YR 3 /2 ,10YR 5/1 dry) silt loam (19 % sand, 25 % clay); strong very fine subangular blocky structure; firm; clear smooth boundary. B gl 23-53 cm. Dark gray (5Y 4/1) loam (36 % sand, 21% clay); moderate fine subangular blocky structure; firm; few fine distinct brown (10YR 4/3) and few fine prominent yellowish red (5YR 4/6) mottles; clear wavy boundary. Bg2 53-91 cm. Gray (5Y 5/1) loam (38 % sand, 21% clay); moderate medium subangular blocky structure parting to moderate fine subangular blocky; common fine prominent strong brown (7.5YR 5/6) mottles; gradual wavy boundary. C 91 + cm. Dark grayish brown (2.5Y 4/2) loam (39 % sand, 24 % clay); structureless, massive; firm;common medium distinct brown (10YR 5/3) and strong brown (7.5YR 5/6) mettles with gray (N 6/0) coatings along root channels. 122 Soil Code: M2 Sampled as: Marlette Classification: Fine-loamy, mixed, mesic Glossoboric Hapludalf Location: Ronald Township, Ionia County, MI. S. NE 1/4 of NW 1/4 of NW 1/4 of Sec. 3, T. 8 N., R. 6 W. Physiographic position: Glacial till plain Topography: 3 percent slope, NE aspect; linear across slope and convex down slope Elevation: 253 m (4 m above soil I-FL-L) Distance from low member of catena: 250 m W MAST: 9.8 °C (mean air temperature 0.4 °C above normal) Drainage: Well drained Vegetation: Forest. White oak (Ouercus albaL.l. pin oak (Ouercus palustris Muenchh.), and big shellbark hickory (Carva lacinosa (Michx. f.) Loud.) Parent material: Glacial till Sampled by: D. Cremeens, 0 . Doumbia, and L. Gloden; 8 July 1981. A 0-12 cm. Dark grayish brown (10YR 4/2) loam; moderate medium granular structure; friable; strongly acid; clear wavy boundary. B/E 12-32 cm. Dark brown (7.5YR 4/4) and grayish brown (10YR 5/2) loam; weak medium platy structure parting to moderate fine subangular blocky; friable; strongly acid; clear wavy boundary. B tl 32-48 cm. Strong brown (7.5YR 4/6) loam; thin discontinuous dark brown (10YR 4/3) clay skins; strong fine subangular blocky structure; friable; strongly acid; clear wavy boundary. Bt2 48-86 cm. Brown (7.5YR 5/4) loam; thick continuous dark brown (7.5YR 4/3) clay skins; moderate medium subangular blocky; firm; neutral; clear wavy boundary. BC 86-127 cm. Biuwn (10YR 5/3) silt ioam; ihin discontinuous dark brown (7.5 YR 3/4) clay skins; strong medium platy structure parting to moderate medium angular blocky; friable; few fine roots; moderately alkaline; clear wavy boundary. C 127+ cm. Dark brown (10YR 4/3) loam; few fine distinct yellowish red (5YR 5/6) mottles; strong coarse platy structure parting to moderate coarse angular blocky; friable; moderately alkaline. 123 Soil Code: C2 Sampled as: Capac Classification: Fine-loamy, mixed, mesic Udollic Ochraqualf Location: Ronald Township, Ionia County, MI. NE 1/4 of NE 1/4 of NW 1/4 of Sec. 3, T. 8 N., R. 6 W. Physiographic position: Glacial till plain Topography: <1 percent slope, W. aspect; linear across and down slope Elevation: 250 m (1.5 m above soil I-FL-L) Distance from low member of catena: 150 m E MAST: 9.8 °C (mean air temperature 0.4 °C above normal) Drainage: Somewhat poorly drained Vegetation: Forest. Bittemut (Carva cordiformis /Wang.') K. Koch.), shagbark hickory (Carva ovata (Mill.) K. Koch.), white oak (Ouercus alba L.), quaking aspen (Pooulus tremuloides Michx.). Parent material: Glacial till. Sampled by: D.Cremeens, and O.Doumbia, 8 July 1981; mottle organization sampled and described by S.W.Sprecher and D.L.Mokma, 12 August 1986. A 0-17 cm. Very dark gray (10YR 3/1) sandy loam; moderate fine granular structure; friable; medium acid; abrupt wavy boundary. B/E 17-33 cm. Dark yellowish brown (10YR 4/4) loam and brown (10YR 5/3) loam; many coarse prominent yellowish brown (10YR 5/6) and common coarse faint (10YR 5/3) mottles; thin discontinuous dark grayish brown (10YR 4/2) clay skins in root channels in bottom of horizon; moderate fine subangular blocky structure; friable; acid; clear wavy boundary. Bt 33-58 cm. Dark yellowish brown (10YR 4/4) loam; thin discontinuous dark grayish brown (10YR 4/2) clay skins; common fine distinct grayish brown (2.5Y 5/2) and common medium distinct strong brown (7.5YR 4/6) mottles; moderate fine subangular blocky structure; firm; slightly acid; clear wavy boundary. Btg 58-85 cm. Grayish brown (2.5Y 5/2) loam; thick discontinuous dark grayish brown (10YR 4/2) clay skins; common medium distinct dark yellowish brwon (10YR 4/4) and common fine distinct grayish brown (10YR 5/2) mottles; moderate medium subangular blocky structure; firm; neutral; clear wavy boundary. BC 85-160 cm. Yellowish brown (10YR 5/6) sandy loam; thin patchy reddish brown (5YR 5/4) clay skins; many fine distinct gray (10YR 6/1) mottles (some as weakly developed neoalbans) and common medium distinct reddish brown (5YR 5/4) mottles (mainly ped interiors); weak coarse angular blocky structure; friable; moderately alkaline; clear smooth boundary. C 160+ cm. Pale brown (10YR 6/3) loam; moderate medium faint gray (10YR 6/1) and few coarse distinct dark brown (7.5YR 4/4) mottles; friable; moderately alkaline. 124 Soil Code: B2 Sampled as: Brookston Classification: Fine-loamy, mixed, mesic Typic Argiaquoll Location: Ronald Township, Ionia County, MI.NE 1/4 of NE 1/4 of NW 1/4 of Sec. 3, T. 8 N., R. 6 W. Physiographic position: Glacial till plain Topography: <1 percent slope; concave across and down slope Elevation: 248 m MAST: 9.9 °C (mean air temperature 0.4 °C above normal) Drainage: Poorly drained Vegetation: Forest Sugar maple (Acer saccharum Marsh.), slippery elm (Ulmus rubra Muhl.), basswood (Tilia americana LA Parent material: Glacial till Sampled by: D. L.Cremeens, 3 Sept. 1981; mottle organization sampled and described by S.W.Sprecher and D.L.Mokma, 12 August 1986. A 0-25 cm. Black (10YR 2/1) and dark gray (10YR 4/1) sandy loam; moderate medium granular structure; friable; slightly acid; abrupt wavy boundary. BA 25-40 cm. Dark gray (10YR 4/1) sandy loam; few fine distinct dark grayish brown (10YR 4/2) mottles and common fine distinct strong brown (7.5YR 4/6) mottles organized as channel and ped ferrans; moderate fine subangular blocky structure; friable; neutral; clear smooth boundary. Btg 40-58 cm. Olive gray (5Y 5/2) loam; common medium distinct yellowish brown (10YR 5/4) and common fine distinct strong brown (7.5YR 5/6) mottles (weakly developed ferrans); medium discontinuous gray (N 5/0) clay skins; weak medium subangular blocky structure parting to moderate fine subangular blocky; friable; neutral; clear smooth boundary. Btg 58-95 cm. Gray (5Y 5/1) loam; few fine distinct dark yellowish brown (10YR 4/4) mottles; medium discontinuous gray (N 5/0) clay skins; weak coarse prismatic structure parting to moderate medium subangular blocky; friable; neutral; gradual wavy boundary. BC 95-170 cm. Gray (5Y 5/1) sandy loam; thin discontinuous olive gray (5Y 5/2) clay skins; many fine prominent strong brown (7.5YR 5/6) mottles; weak fine subangular blocky structure; friable; mildly alkaline; clear wavy boundary. C 170+ cm. Dark yellowish brown (10YR 4/4) loam; common medium distinct gray (N 5/0) mottles; massive; firm; moderately alkaline. 125 Soil Code: B-S-H Sampled as: Rubicon (mesic taxadjunct) Classification: Sandy, mixed, mesic Entic Haplorthod Location: Inverness Township, Cheboygan County, MI. NW 1/4 of NW 1/4 of SW 1/4 of Sec. 14, T. 37 N .,R .2 W . Physiographic position: Glacial outwash plain Topography: 5 percent slope, NE aspect; concave across and down slope Elevation: 209 m (2.2 m above soil B-S-L) Distance from low member of catena: 49 m S MAST: 8.2 °C (mean air temperature 0.1 °C below normal) Drainage: Well drained Vegetation: Forest. Red maple (Acer rubrum L.l. Eastern hemlock (Tsuga canadensis (L.) Carr), yellow birch (Betula lutea Michx. f.), bracken fern (Pteridium aouilinum (L.) Kuhn), tree clubmoss (Lvcopodium obscurum L.). Clintonia (Clintonia borealis (Ait.) Raf.), starflower (Trientalis borealis Raf.) Parent material: Sandy glacial drift Sampled by: S.W.Sprecher, D.L.Mokma, and S.Haile-Mariam; 20 Aug. 1985. A 0-10 cm. Black (1OYR 2/1) sand; moderate medium granular structure; very friable; many very fine and fine and few medium and coarse roots; medium acid; abrupt smooth boundary. E 10-34 cm. Grayish brown (10YR 5/2) sand; single grain structure; loose; few roots; strongly acid; abrupt wavy boundary. B sl 34-50 cm, tongues penetrating to 84 cm. Yellowish red (5YR 5/8) sand; patches of dark reddish brown (5YR 2/2) weakly cemented ortstein; weak coarse subangular blocky structure; very friable; cracked coatings on sand grains; few fine medium and coarse roots; strongly acid; clear irregular boundary. Bs2 50-113 cm. Strong brown (7.5YR 4/6) sand; weak medium subangular blocky structure; very rriabie; cracked coatings on sand grains; few fine roots; strongly acid; gradual wavy boundary. Cl 113-153 cm. Yellowish brown (10YR 5/6) sand; common fine distinct dark yellowish brown (10YR 3/4) mottles; moderate coarse subangular blocky structure; very friable; strongly acid; clear wavy boundary. C2 153-175 cm. Yellowish brown (10YR 5/8) sand; single grain structure; loose. 12 6 Soil Code: B-S-M Sampled as: Au Gres (mesic taxadjunct) Classification: Sandy, mixed, mesic Entic Haplaquod Location: Inverness Township, Cheboygan County, MI. NW 1/4 of NW 1/4 of SW 1/4 of Sec. 14, T. 37 N., R . 2 W . Physiographic position: Glacial outwash plain Topography: 3 percent slope, SE aspect; linear across slope and convex down slope Elevation: 208 m (0.6 m above soil B-S-L) Distance from low member of catena: 21 m S MAST: 8.3 °C (mean air temperature 0.1 °C below normal) Drainage: Somewhat poorly drained Vegetation: Forest. Red pine (Pinus resinosa Ait.), red maple (Acer rubrum L.), Eastern hemlock (Tsuea canadensis (L.) Carr), yellow birch (Betula lutea Michx. f.), bracken fern (Pteridium aouilinum (L.) Kuhn) Parent material: Sandy glacial drift Sampled by: S. W. Sprecher and S. Haile-Mariam; 4 July 1985. Oi 8-0 cm. Very dark brown (10YR 2/2) partially decomposed leaf litter; moderate coarse platy structure parting to moderate coarse granular; friable;Je.w medium and coarse roots; very stronglyacid; abrupt wavy boundary. E 0-31 cm. Light gray (10YR 7/1) sand; many medium distinct light brownish gray (10YR 6/2) mottles; weak coarse subangular blocky structure; very friable; few very fine and common fine roots; very strongly acid; abrupt wavy boundary. B sl 31-45 cm. Strong brown (7.5YR 5/6) sand; few medium distinct dark brown (7.5YR 3/4) and few medium fine light yellowish brown (2.5Y 6/4) mottles; weak coarse subangular blocky structure in discontinuous strong brown (7.5YR 4/6) ortstein; cracked coatings on sand grains; very firm; few very fine and fine roots; verystrongly acid; clear broken boundary. Bs2 45-66 cm. Dark reddish brown (5YR 3/4) and strong brown (7.5YR 4/6) sand; weak coarse subangular blocky structure; firm; cracked coatings on sand grains; few fine roots; very strongly acid; clear wavy boundary. BC 66-85 cm. Dark brown (7.5YR 4/4) sand; common coarse distinct yellowish brown (10YR 5/4) mottles; weak medium subangular blocky structure; very friable; strongly acid; clear wavy boundary. C 85-150 cm. Yellowish brown (10YR 5/4) sand; single grain structure; loose; strongly acid. 12 7 B-S-L Sampled as : Kinross (mesic taxadjunct) Classification : Sandy, mixed, mesic Entic Haplaquod. Location : Inverness Township, Cheboygan County, MI. NW 1/4 NW 1/4 SW 1/4 Sec. 14, T. 37 N., R. 2 W. Physiographic position: Glacial outwash plain Topography : 0 % slope; concave across and down slope Elevation: 207 m MAST : 8.8 °C (mean air temperature 0.1 °C below normal) D rainage: Poorly drained Vegetation : Forest. Red maple ( Acer rubrum L.L Eastern Hemlock (Tsuea canadensis (L.) Carr.); paperbirch (Betula papvrifera Marsha.), bracken fem(Pteridium aauilinum (L.) Kuhn), bunchbeny (Comus canadensis L.), staghom clubmoss (Lvcopodium clavatum L.), star flower (Trientalis borealis Raf.) Parent m aterial: Sandy glacial drift Sampled by : S. W. Sprecher, D. L. Mokma, and S. Haile-Mariam; 20 August 1985. Oa 8-0 cm. Very dark brown (10YR 2/2) organic matter; moderate medium platy structure; very strongly acid; abrupt smooth boundary. E 0-18 cm. Light gray (10YR 7/1) sand; single grain structure; loose; common fine and few medium roots; very strongly acid; abrupt wavy boundary. Bs 18-50 cm. Strong brown (7.5YR 5/6) and yellowish red (SYR 4/6) sand; common medium distinct dark brown (7.5YR 4/6) mottles; weak coarse subangular blocky structure; friable; cracked coatings on sand grains; few medium roots; strongly acid; gradual wavy boundary. C 50-125 cm. Brown (10YR 5/3) sand; single grain structure; loose; strongly acid. 128 Soil Code: P-S-H Sampled as: Liminga Classification: Sandy, mixed, frigid Typic Haplorthod Location: Munuscong State Forest, Chippewa County, Michigan. NE 1/4 of NW 1/4 of NE 1/4 of Sec. 20, T. 45 N., R. 1 W. Physiographic position: Glacial outwash plain Topography: 15 percent slope, SE aspect; convex across and down slope Elevation: 224 m (1.6 m above soil P-S-L) Distance from low member of catena: 30 m E MAST: 8.3 °C (mean air temperature 1.1 °C above normal) Drainage: Well drained Vegetation: Forest. Red maple (Acer rubrum L,), Eastern hemlock (Tsuea canadensis (L.) Carr.), quaking aspen (Ponulus tremuloides Michx.). bracken fern (Pteridiuir. aouilinum L.), bunchberry (Comus canadensis L.), wintergreen (Pvrola secunda L.). Parent material: Sandy glacial drift Sampled by: S.W.Sprecher, D.L.Mokma, and S.Haile-Mariam; 20 Aug. 1985 A 0-7 cm. Black (N 2/0) fine sand; weak medium granular structure; very friable; many fine roots; extremely acid; clear wavy boundary. E 7-11 cm. Brown (7.5YR 5/2) fine sand; weak medium platy structure; very friable; extremely acid; clear irregular boundary. Bhs 11-22 cm, tonguing to 97 cm. Dark reddish brown (5YR 3/2) fine sand; many coarse prominent strong brown (7.5YR 4/6) mottles; very firm; discontinuous ortstein; cracked coatings on sand grains; extremely acid; clear irregular boundary. Bs 22-63 cm. Strong brown (7.5YR 4/6) fine sand; firm; cracked coatings on sand grains; extremely acid; diffuse irregular boundary. C 63-192 cm. Yellowish brown (10YR 5/4) fine sand; single grain structure with broken patches of weak medium subangular blocky material in upper 10 cm; loose; strongly acid. 129 Soil Code: P-S-M Sampled as: Wainola Classification: Sandy, mixed, frigid Entic Haplaquod Location: Munuscong State Forest, Chippewa County, Michigan. NE 1/4 of NW 1/4 of NE 1/4 of Sec. 20, T. 45 N., R. 1 W. Physiographic position: Glacial outwash plain Topography: 1 percent slope, N aspect; linear across slope, concave down slope Elevation: 223 m (0.5 m above soil P-S-L) Distance from low member of catena: 24 m W MAST: 7.8 °C (mean air temperature 1.1 °C above normal) Drainage: Somewhat poorly drained Vegetation: Forest. Red maple (Acer rubrum L.), Eastern hemlock (Tsuga canadensis (L.) Carr.), black spruce (Picea mariana (Mill.) BSP.), bracken fern (Pteridium aquilinum L.), bunchberry (Comus canadensis L.), tree clubmoss (Lycopodium obscurum L.), starflower (Trientalis borealis Raf), wintergreen (Pyrola secunda L.). Parent material: Sandy glacial drift Sampled by: S. W. Sprecher, D. L. Mokma, and S. Haile-Mariam; 20 Aug. 1985 Oi 7-0 cm. Black (N 2/0) fibric organic materials mixed with fine sand; weak medium granular structure; friable; many very fine and fine roots and common medium roots; very strongly acid; abrupt wavy boundary. E 0-15 cm. Light gray (7.5YR 6/1) fine sand; weak medium platy structure extremely acid; abrupt wavy boundary. Bhs 15-43 cm. Dark brown (7.5YR 3/2) and brown (7.5YR 4/4) fine sand;common medium distinct dark brown (10YR 4/3) mottles; weak medium subangular blocky structure; pockets of weakly cemented ortstein; cracked coatings on sand grains; firm; common fine and medium roots; very strongly acid; abrupt irregular boundary. BC 43-77 cm. Yellowish brown (10YR 5/8) fine sand; common fine distinct very dark gray (10YR 3/1) streaks; weak coarse platy structure parting to weak medium subangular blocky; friable; few fine and medium roots; very strongly acid; gradual wavy boundary. C 77-118 cm. Dark yellowish brown (10YR 4/6) fine sand; few fine distinct light brownish gray (10YR 6/2) and common medium distinct strong brown (7.5YR 5/6) mottles; weak coarse platy structure; very friable; strongly acid. 130 Soil Code: P-S-L Sampled as: Kinross Classification: Sandy, mixed, frigid Aerie Haplaquod Location: Munuscong State Forest, Chippewa County, Michigan.NE 1/4 of NW 1/4 of NE 1/4 of Sec. 20, T. 45 N., R. 1 W. Physiographic position: Glacial outwash plain Topography: 1/2 percent slope, N. aspect; linear across slope, concave down slope Elevation: 222 m MAST: 8.4 °C (mean air temperature 1.1 °C above normal) Drainage: Poorly drained Vegetation: Forest. Red maple (Acer rubrum L.). Eastern hemlock (Tsuga canadensis (L.) Carr.), cinnamon fem (Osmunda cinnamomea L.), bunchberry (Comus canadensis L.), clubmosses (Lvcopodium L. spp.), starflower (Trientalis borealis Raf), mosses. Parent material: Sandy glacial drift Sampled by: S.W.Sprecher, D.L.Mokma, and S.Haile-Mariam; 20 Aug.1985 Oa 0-10 cm. Black (N 2/0) and very dark brown (10YR 2/2) decomposed organic matter; moderate very coarse platy structure parting to moderate fine platy; very friable; many roots; extremely acid; abrupt wavy boundary. E 0-8 cm. Gray (5Y 6/1) loamy fine sand; common coarse distinct light brownish gray (10YR 6/2) mottles; weak very coarse platy structure; very friable; few roots; very strongly acid; abrupt wavy boundary. Bhs 8-19 cm. Dark brown (7.5YR 3/2) loamy fine sand; common medium faint grayishbrown (10YR 5/2) mottles; weak medium subangular blocky structure; few ine roots; very friable; cracked coatings on sand grains; very strongly acid; clear broken boundary. Bg 19-34 cm. Light brownish gray (2.5Y 6/2) loamy fine sand; common medium faint grayish brown (10YR 5/2) mottles; weak medium subangular blocky structure; few fine roots; very friable; very strongly acid; clear broken boundary. C 34-56 cm. Dark brown (7.5YR 4/4) fine sand; common fine distinct dark gray (10YR 4/1) and strong brown (7.5YR 5/6) mottles; single grain structure; loose; strongly acid.