HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Harris, W.G. Articles by Hollien, K.A. Search for Related Content PubMed Articles by Harris, W.G. Articles by Hollien, K.A. GeoRef GeoRef Citation Agricola Articles by Harris, W.G. Articles by Hollien, K.A.

DIVISION S-5-PEDOLOGY

Changes across Artificial E–Bh Boundaries Formed under Simulated Fluctuating Water Tables

Soil and Water Science Dep., 2169 McCarty Hall, Univ. of Florida, Gainesville, FL 32611 USA

apatite{at}ufl.edu

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Spodosols in the coastal plain of the southeastern USA are mainly restricted to zones of fluctuating water table, but mechanisms of water table linkage are not well established. A previous study showed that Aquod-like E and Bh horizons could be artificially formed only under simulated fluctuating water table conditions. The present study was conducted to characterize the redistribution of components during artificial podzolization and to compare the resultant distributions with those of an Aquod adjacent to the Psamment from which column materials were collected. A hypothesis tested is that the E–Bh formation involves colloidal as well as chemical translocation. Artificial E–Bh horizons were generated in three acrylic columns using an approximate cycle of 22 h of saturation and 2 h of free drainage. Distilled water was introduced from the bottom of columns, and oxalic acid from the top. Boundaries migrated downward at a progressively slower rate, and became increasingly abrupt and irregular with time. E horizons consisted mainly of stripped sand grains; coatings were retained in Bh horizons. Fine materials containing C and Al accumulated in the upper centimeters of the Bh, as documented by scanning electron microscopy (SEM) and energy dispersive x-ray analysis (EDX). Bulk analyses verified accumulations of C in the Bh and a movement of Al and crystalline clay from E to Bh. Artificial E–Bh formation entailed redistribution of sand grain coatings initially present on sands used in columns. Crystalline as well as noncrystalline clay accumulated in the Bh horizon, supporting the idea that E–Bh formation involved colloidal as well as chemical migration. A weatherable-mineral source of Al is not required for the artificial water table–induced formation of E–Bh sequences. Noncrystalline Al, dissolved in the grain-stripping process, is a likely source of Al in Bh horizons of the region. Distributional trends in artificial E–Bh sequences paralleled those of the natural Aquod.

Abbreviations: EDX, energy dispersive x-ray analysis • SEM, scanning electron microscopy • XRD, x-ray diffraction analysis

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
THE E AND BH HORIZONS of Spodosols (Soil Survey Staff, 1998) in the coastal plain of the southeastern USA are mainly restricted to zones of fluctuating water table (Brasfield et al., 1973; Stone et al., 1993). Such a restriction does not apply to albic-spodic sequences of northern regions, where well-drained Spodosols can be common (McKeague et al., 1983). Studies of water table dynamics document a very close correspondence between podzolic E–Bh development and relative elevation of the water table (Garman et al., 1981; Tan et al., 1999). The common trend at the transition between Aquods and better-drained soils in areas of coated sands is for the thickness and morphological prominence of E and Bh horizons to diminish with diminishing water table influence (e.g., shorter periods of saturation, greater depths to saturation). In effect, the Bh horizon fades in color and its upper boundary becomes shallower as the relative average depth to water table increases (Fig. 1b) . An equally accurate way to describe the transition is that the zone of coating-free sand grains, which constitutes the podzolic E horizon (Harris et al., 1987a), lenses out as the better-drained soil is approached.

View larger version (36K): [in this window] [in a new window]   Fig. 1 Study site information. (A) Site location and schematic of sampling layout and soil distribution. (B) Schematic of transition between Psamment (SW) and Aquod (NE). Fading upward of Bh and wedging out of E is typical of transitions between Aquods and better-drained soils in northern Florida (Garman et al., 1981; Tan et al., 1999)

Another relevant genetic consideration is the very low content of weatherable primary minerals in southern coastal Spodosols (as well as associated better-drained soils) to a depth of 2 m or more (Brasfield et al., 1973; Holzhey et al., 1975). This raises a question as to the source of Al in Bh horizons in the region, since extant theories of podzolization commonly invoke weatherable minerals as the source of metals that accumulate to form spodic horizons. A possible alternative source for Al in Aquod Bh horizons is noncrystalline forms in sand grain coatings. Harris et al. (1995) found that oxalate-extractable Al was significantly higher in the sandy soils on toeslopes adjacent to Alaquods than in the soils on summits.

Metal oxides that cement sand grain coatings comprise only a minor proportion of the coatings by mass (Harris et al., 1987a, 1987b). Coatings are commonly dominated by kaolinite and hydroxy-interlayered minerals. These resistant secondary phyllosilicates are not prone to dissolution under weakly acidic conditions. We therefore suspect that they are subject to colloidal eluviation and illuviation when they are released from grain coatings. Others have reported evidence of particle migration in Spodosols (Calhoun and Carlisle, 1973; Guillet et al., 1975; Ugolini et al., 1977; Li et al., 1998). Also, the study by Harris et al. (1995) documented the presence of darkened lamellae below zones of stripped sands in the experimental columns, but use of small glass columns did not permit solid-state characterization of these features.

There are advantages to studying artificially induced pedologic processes. Simulation enables real-time assessment, beginning with a uniform parent material at time zero, which is generally not possible in the study of natural soils. In effect, the condition of the parent material must be inferred for a soil, along with processes that produced the soil in its current state. However, there is no uncertainty about the parent material for an artificially formed soil. Processes can be monitored, hypotheses can be experimentally tested, and comparisons can be made directly between the artificial soil and its parent material or its natural soil analogue.

The present study was conducted (i) to characterize the redistribution of components that takes place in the artificial formation of E–Bh sequences and (ii) to compare resultant distributions with those of a natural Aquod adjacent to the Psamments from which the column materials were collected (10–50 cm depth). A hypothesis tested in this study is that artificial E–Bh formation involves colloidal translocation of crystalline clay as well as the chemical translocation of C and Al.

	Materials and methods

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Sites, Soils, and Sampling
Samples were collected from a site located on the northern highlands of peninsular Florida (White, 1970) (Fig. 1a), and from soils forming in Plio-Pleistocene marine terrace deposits (Healy, 1975). The site corresponds with Site 1 as designated in the study by Harris et al. (1995). The landscape is characterized by subsurface drainage, closed depressions, and lakes. Dominant soils are Typic Quartzipsamments and Grossarenic Paleudults. Arenic Alaquods are confined mainly to some lake margins, where water tables fluctuate within the soil. The specific sampling area at the site encompassed an Aquod, a Psamment, and a transition of 10 m (Fig. 1a). The morphological nature of the transition is typical of transitions between Aquods and better-drained soils in northern Florida (Fig. 1b).

Vegetation of the sampling area was dominantly hardwood forest. Species present included live oak (Quercus virginiana Miller), laurel oak (Quercus laurafolia Michx.), sweetgum (Liquidambar styraciflua L.), and saw palmetto [Serenoa repens (W. Bartram) Small]. We did not discern a vegetation difference between Aquod and Psamment near the transition, where samples of this study were collected. However, uncleared areas approaching the summit (upslope from lake margin) were occupied by "sand-hills" species and open canopy more typical of Psamments. Aquods are commonly associated with pines (Pinus spp.) or flatwoods landscapes, but their occurrence is not restricted to pines or flatwoods.

Samples were collected by auger and stored in sealed plastic bags. Representative Psamment and Aquod profiles were described and sampled by horizon a few meters from the transition. The Psamment area was also sampled at three points parallel to the transition (West, Middle, and East in Fig. 1a), from 10 cm below the mineral soil surface (or from the base of the A if this was deeper than 10 cm) to 50-cm depth. The latter samples were collected to provide material for replicated experimental columns. No major morphological changes were observed within the upper 50-cm depth of the Psamments. Patchy, tenacious grain coatings, similar to those observed for a nearby sample (Fig. 10B in Harris et al., 1995), were visible with a hand lens in the Psamment samples.

Column Preparation and Procedure
Acrylic tubes of 85-cm length and 4-cm i.d. were packed with air-dried sample to a depth of 80 cm. The targeted column bulk density was 1.5 g cm^-3, but final densities ranged from 1.4 to 1.5 g cm^-3 due to variation in packing. Glass wool was used in the bottom of the columns as a filter. A stopper fitted with a glass tube served as outlet for the columns, as well as a means of introducing distilled water from the bottom in the simulation of a fluctuating water table. Tygon tubing was fitted over the glass tube, and valves were used with the tubing to control flow in and out of the column.

The general daily procedure for generating artificial E–Bh sequences was to (i) introduce a water table (deionized water) slowly from the bottom of the columns upward to a specified depth; (ii) add a prescribed amount (see below) of 0.01 M oxalic acid from the top of the column, thereby bringing the water table to the soil surface or just above; (iii) incubate with the water table and acid for 22 h; (iv) drain column for 1.75 h; and (v) reintroduce the water table in the next 0.25 h; and continue the cycle. This regimen was followed in an approximate way for 81 d. Some weekends the columns were not tended, and near the end of the study more time was gradually required for drainage due to a marked decrease in hydraulic conductivity. A detailed daily log was kept of how the columns were managed, and all columns were subjected to the same treatments.

In natural soils, organic acids are released from roots (e.g., as exudate) and decomposition of organic matter incorporated to at least a few centimters depth. Thus, these acids can occur at sufficiently elevated concentrations locally with depth to continue promoting thickening of the E horizon during Aquod genesis. The way that we chose to simulate this scenario in a controlled column was to increase the volume of 0.01 M oxalic acid incrementally with time to assure that a near maximum initial concentration was attained in the zone of the downward migrating boundary within the column. Initially a volume of 40 mL of 0.01 M oxalic acid was added to the columns, but the volume was increased successively during the course of the study to 80, 120, 160, and 240 mL. These additions amounted to column concentrations ranging from 0.001 to 0.005 M (pH 3.3–3.6), taking into account the water already in the column. The increasing volumes of acid were offset by decreasing amounts of water added from the bottom of the columns.

Oxalic acid was selected because it was found to be the prevalent simple organic acid in surface horizons of selected forest soils of the southeastern USA, occurring at concentrations as high as 1 to 2 mM (Fox and Comerford, 1990). The net column concentrations of oxalic acid for this simulation (see below) were at or slightly above the high end of the forest soil range. Selection of oxalic acid concentration was a compromise to achieve a practical simulation time frame, while not grossly exceeding realistic soil conditions.

Sampling of Artificial Horizons
A window was cut from each column at the end of the study to expose the boundary region for sampling. The acrylic cut-out was removed carefully to avoid disturbance of the column material. Cylindrical metal rings of 1-cm diam. were pressed into the undisturbed column material precisely at the boundary, such that both E and Bh materials were included within the rings. The rings were then removed with the sample, thereby providing an undisturbed core for optical- and electron microscopic examination. In addition, replicate bulk samples were collected of the E and Bh within 4 cm of the boundary. The latter samples were used for determination of total C, oxalate-extractable Al and Fe, particle size, and clay mineralogy.

Chemical and Physical Analysis
Natural soil horizons and artificial horizons were subjected to essentially the same analytical procedures, but in some cases smaller aliquots were used for the latter due to limited availability of material. Total C was determined by flash combustion and CO₂ analysis using a C–N analyzer. Aluminum and Fe were extracted by acid ammonium oxalate in the dark (McKeague and Day, 1966), and extract concentrations were determined by atomic adsorption spectrometry. This extraction is selective for the noncrystalline forms of these metals in soils (McKeague et al., 1971). Ammonium oxalate extraction was also used as a pretreatment prior to particle-size determination, in order to remove noncrystalline clay components and document the distribution of crystalline clay. Clay was also treated with H₂O₂ to remove organic matter after it was separated and prior to weight determination. Particle size was determined by separating the sand (2–0.05 mm) by sieving and the silt (0.05–0.002 mm) and clay (<0.002 mm) by centrifugation, and weighing the separates. The initial weight for particle-size determination was recorded prior to removal of noncrystalline material, such that deficits in total recovery would reflect the mass of noncrystalline material removed.

Mineralogy and Microscopic Analysis
Mineralogy of the clay fraction was determined for selected soil horizons and for replicate artificial horizons. Sample preparation included pretreatment with acid ammonium oxalate in the dark to remove noncrystalline components (McKeague and Day, 1966). The silt and clay were then separated by centrifugation (Whittig and Allardice, 1986) following Na saturation to promote dispersion. The clay fraction of most soil horizons was prepared for x-ray diffraction analysis (XRD) analysis by depositing 250 mg as a suspension onto ceramic tiles under suction. Potassium and Mg saturations were performed directly on the tiles by washing with the respective Cl^- salts and rinsing. Glycerol was added to the Mg-saturated sample. Clay recoveries from the E horizon of the Aquod and for the artificial horizons were too low for tile mounting, so these clays were collected and saturated on a 0.45-µm filter and transferred to a low-background mount. Glycerol was not added to the samples on the low-background mount, because it proved to be unnecessary for identification. Samples were scanned at 2° 2 min^-1 with CuK radiation.

Mineralogy of the sand fractions from the E and E' horizons of the Aquod and the E1 and E4 horizons of the Psamment was determined petrographically. Grain mounts were prepared of the fine sand (codominant with medium sand in these samples) with a medium of 1.54 refractive index. Grain counts on a minimum of 300 grains were performed on a polarizing microscope.

Undisturbed core samples of the boundary region within columns were examined and photographed with a dissecting microscope and a scanning electron microscope equipped with an EDX system. Samples were prepared for SEM by coating with Au–Pd. The EDX spectra were obtained for regions at various distances from the E–Bh boundary.

	Results and discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Data for the Aquod and Psamment that were described and sampled at the site show typical morphological and distributional trends for these suborders in the region (Table 1 , Fig. 2) . The E–Bh boundary of the Aquod shows sharp contrasts in color (high to low value), crystalline clay and silt content, total C, and oxalate-extractable Al. No boundary in the Psamment exhibits such contrasts, though gradual changes in color value (increasing) and oxalate-extractable metals (decreasing) do occur with depth. The latter changes probably are the result of increasing wetness and lower redox potentials deeper in the Psamment. Total recovery of particle-size fractions was essentially complete in Psamment samples, which had low quantities of noncrystalline materials. However, the Aquod showed proportionate deficits in recovered material where noncrystalline materials were abundant (e.g., Bh horizons; Table 1). The Aquod E–Bh boundary also corresponds with sharp mineralogical change for the clay fraction (Fig. 2). The E horizon is dominated by quartz, with phyllosilicates being present in barely detectable amounts. However, subjacent horizons contain appreciable quantities of phyllosilicates. This clay mineralogical trend is consistent with other data for Florida Aquods (Harris and Carlisle, 1985). In effect, the E horizon contains a very low quantity of clay, most of which is quartz. The Psamment is essentially uniform with depth with respect to clay mineralogy. These trends suggest that a redistribution of both crystalline and noncrystalline components has taken place in the Aquod, but not in the Psamment located 20 m upslope (slope gradient 1–2 %).

View larger version (25K): [in this window] [in a new window]   Fig. 2 X-ray diffraction patterns for clay fractions of E and Bh horizons sampled from a representative Psamment and Aquod at the study site. Note that the distribution of clay-sized minerals is essentially uniform with depth in the Psamment, but that secondary phyllosilicates (kaolinite and hydroxy-interlayered minerals) abruptly increase in abundance from E to Bh in the Aquod. A similar tendency was apparent for the artificial E–Bh sequences (Fig. 6), though differences were less pronounced. Samples were Mg saturated and glycerol solvated except for the Aquod E horizon, which received no glycerol

View larger version (34K): [in this window] [in a new window]   Fig. 6 Comparison between artificial horizons with respect to crystalline clay content (after oxalate and peroxide pretreatments), total C, and ratio of kaolinite (0.358 nm) to quartz (0.334 nm) x-ray diffraction peaks. Small letters designate significant differences (Duncans Multiple Range; P < 0.05; ) between horizons, as compared according to the designated variable (SAS Institute, 1985). A different letter between the compared materials for a given variable indicates a statistically significant difference

The artificial E–Bh boundaries which formed in the initially uniform columns (Table 2) showed similar morphological, physical, and chemical contrasts to those discussed above for the Aquod located a few meters from the Psamment area where the column materials were collected. The boundaries were abrupt, with irregular topographies (Fig. 3a) . The artificial Bh horizons were darkest at the upper boundary, though some darkening occurred throughout the column below the boundary as compared with the original material. Sand grains above the boundary were mainly stripped, but grains below the boundary retained their coatings (Fig. 3b–3f). Fine material accumulated at the upper Bh boundary, which bridged sand grains and cracked upon desiccation (Fig. 4) . This fine material was shown by EDX spectra to contain C and Al, along with Si. Bulk comparison between original Psamment materials and artificial horizons shows a significant redistribution of noncrystalline (oxalate-extractable) Al from E to Bh (Fig. 5) . There was also a redistribution of crystalline clay from E to Bh (Fig. 6) and a significant accumulation of C in the Bh. The clay mineralogical contrast between E and Bh was not completely consistent, but, based on XRD peak area ratios, there was a significantly higher proportion of kaolinite to quartz in the Bh relative to the E (Fig. 6).

View larger version (110K): [in this window] [in a new window]   Fig. 3 Morphological and micromorphological features formed in Column East (similar features were observed in other columns). (A) View of the column, showing abrupt and irregular boundary between E (upper) and Bh horizons. The upper part of the Bh is darkest, but the entire column below the boundary is darker than the original Psamment material. There are also patches of darker material at various depths, but which are not readily discernible in the photograph. (B) Micrograph of the boundary, taken after cutting away a window in the acrylic column. (C, D) Scanning electron microscopy (SEM) images of quartz grains above the boundary, showing that they are essentially free of coating materials; (E, F) SEM images of grains below the boundary, showing coated surfaces

View larger version (153K): [in this window] [in a new window]   Fig. 4 Images of material accumulated in the upper part of the Bh. (A) Magnified optical image of E–Bh boundary. Area in dashed rectangle shown in B. (B) Scanning electron microscopy image showing accumulations of fine material that bridges some grains and cracks upon desiccation. (C) Closer view of area marked by dashed rectangle in B

View larger version (32K): [in this window] [in a new window]   Fig. 5 Comparison between original Psamment materials and artificial horizons formed in those materials with respect to noncrystalline Fe and Al contents (assessed by acid ammonium oxalate extraction). Small letters designate significant differences (Duncans Multiple Range; P < 0.05; ) in metal contents between Psamment, E, and Bh materials [SAS Institute, 1985]). A different letter between the compared materials for a given metal indicates a statistically significant difference

This study did not entail photographic documentation of the development of the artificial horizons, but day to day observations revealed some interesting qualitative aspects of the chronology. The first sign of change was the stripping of sand grains in the upper 5 to 10 cm of the columns. The stripped zone thickened relatively quickly in the initial stages, but slowed down with time. The decrease in rate corresponded generally with the appearance of a thin darkened zone at the boundary. Boundary topography became more and more irregular. The irregularity arose from the development of Bh pinnacles, which may have served as a hydraulic barriers such that flow was diverted around them. The migration of boundaries essentially stopped roughly midway down the columns. Hydraulic conductivity of columns eventually decreased to the point where it was not possible to maintain the same daily water table regimens, so the experiment was terminated.

Redistributions documented in this experimental study have implications for the relatively thin Bh horizons that commonly occur within a 2-m depth. However, they may be less applicable to deep, thick Bh horizons common to some areas of the coastal plain (Daniels et al., 1975). Deep sands that are perpetually saturated could possibly retain C by means of reaction with Al that is already present below the water table (Farmer et al., 1983), such that downward translocation of Al or other inorganic components may not be a significant factor.

	Conclusions

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Artificial formation of E and Bh horizons under fluctuating water table and in the presence of oxalic acid entailed a redistribution of sand grain coatings and a subsurface accumulation of C. Crystalline as well as noncrystalline clay materials were accumulated in the Bh horizon, supporting the idea that E–Bh formation involved colloidal as well as chemical migration. A weatherable-mineral source of Al is not required for the artificial water table–induced formation of E–Bh sequences. Secondary noncrystalline Al, dissolved in the grain-stripping process (E-horizon formation), is a likely source of Al that accumulates in Bh horizons of the region. Trends in distribution of components in artificial E–Bh sequences are parallel to those exhibited by natural Aquods.Guillet Rouiller Souchier 1975

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Florida Agricultural Experiment Station Journal Series no. R-06845.

Received for publication March 29, 1999.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 

• Brasfield, J.F., V.W. Carlisle, and R.W. Johnson. 1973. Spodosols—Soils with a spodic horizon. p. 57–60. In S.W. Buol (ed.) Soils of the southern states and Puerto Rico. Southern Coop. Series Bull. 174. USDA.
• Calhoun F.G., Carlisle V.W. Statistical analysis of Spodosol parameters. Soil Crop Sci. Soc. Fla. Proc. 1973;33:139-143.
• Daniels R.B., Gamble E.E., Holzhey C.S. Thick Bh horizons in the North Carolina Coastal Plain: Morphology and relation to texture and ground water. Soil Sci. Soc. Am. Proc. 1975;39:1177-1181.
• Farmer V.C., Skjemstad J.O., Thompson C.H. Genesis of humus B horizons in hydromorphic humus podzols. Nature (London) 1983;304:342-344.
• Fox T.R., Comerford N.B. Low molecular weight organic acids in selected forest soils of the southeastern USA. Soil Sci. Soc. Am. J. 1990;54:1139-1144.[Abstract/Free Full Text]
• Garman C.R., Carlisle V.W., Zelazny L.W., Beville B.C. Aquiclude-related spodic horizon development. Soil Crop Sci. Soc. Fla. Proc. 1981;40:105-110.
• Guillet B., Rouiller J., Souchier B. Podzolization and clay migration in Spodosols of eastern France. Geoderma. 1975;14:223-245.
• Harris W.G., Carlisle V.W. Clay mineralogical relationships in Florida Haplaquods. Soil Sci. Soc. Am. J. 1985;51:481-484.
• Harris W.G., Carlisle V.W., Chesser S.L. Clay mineralogy as related to morphology of Florida soils with sandy epipedons. Soil Sci. Soc. Am. J. 1987;51:1673-1677 a.[Abstract/Free Full Text]
• Harris W.G., Carlisle V.W., Van Rees K.C.J. Pedon zonation of hydroxy-interlayered minerals in Ultic Haplaquods. Soil Sci. Soc. Am. J. 1987;51:1367-1372 b.[Abstract/Free Full Text]
• Harris W.G., Crownover S.H., Comerford N.B. Experimental formation of Aquod-like features in sandy coastal plain soils. Soil Sci. Soc. Am. J. 1995;59:877-886.[Abstract/Free Full Text]
• Healy H.G. Terraces and shorelines of Florida. Map Series 71. Tallahassee, FL: Bureau of Geology, Florida Dep. Natural Resources, 1975.
• Holzhey C.S., Daniels R.B., Gamble E.E. Thick Bh horizons in the North Carolina coastal plain: II. Physical and chemical properties and rates of organic additions from surface sources. Soil Sci. Soc. Am. Proc. 1975;39:1182-1187.
• Li S.Y., Cheng Z.S., Liu J.C. Subalpine loamy Spodosols in Taiwan: Characteristics, micromorphology, and genesis. Soil Sci. Soc. Am. J. 1998;62:710-716.[Abstract/Free Full Text]
• McKeague J.A., Day J.H. Dithionite- and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Can. J. Soil Sci. 1966;46:13-22.
• McKeague J.A., Brydon J.E., Miles N.M. Differentiation of forms of extractable Fe and Al in soils. Soil Sci. Soc. Am. Proc. 1971;35:33-38.
• McKeague J.A., DeConinck F., Franzmeier D.P. Spodosols. In: Wilding L.P., et al. , ed. Pedogenesis and soil taxonomy. Vol. 2. The soil orders. New York: Elsevier, 1983:217-252.
• SAS Institute. SAS user's guide: Statistics, 5th ed Cary, NC: SAS Inst, 1985.
• Soil Survey Staff. Keys to soil taxonomy. Washington, DC: USDA-NRCS. U.S. Gov. Print. Office, 1998.
• Stone E.L., Harris W.G., Brown R.B., Kuehl R.J. Carbon storage in Florida Spodosols. Soil Sci. Soc. Am. J. 1993;57:179-182.[Abstract/Free Full Text]
• Tan Z., Harris W.G., Mansell R.S. Watertable dynamics across an Aquod-Udult transition in Florida flatwoods. Soil Sci. 1999;164:10-17.
• Ugolini F.C., Dawson H., Zachara J. Direct evidence of particle migration in the soil solution of a podzol. Science 1977;198:603-605.[Abstract/Free Full Text]
• White, W.A. 1970. The geomorphology of the Florida peninsula. Geol. Bull. 51. Bureau of Geol., Div. Interior Resources, Florida Dep. Natural Resources, Tallahassee, FL.
• Whittig L.D., Allardice W.R. X-ray diffraction techniques. In: Klute A., ed. Methods of soil analysis. Part 1. Agron. Monogr. 9. Madison, WI: ASA and SSSA, 1986:331-362.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Harris, W.G. Articles by Hollien, K.A. Search for Related Content PubMed Articles by Harris, W.G. Articles by Hollien, K.A. GeoRef GeoRef Citation Agricola Articles by Harris, W.G. Articles by Hollien, K.A.