| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a University of Maryland at College Park, 1112 H.J. Patterson, College Park, MD 20742
b Sustainable Agricultural Systems Lab., USDA, ARS, ANRI, Bldg. 001, BARC-West, 10300 Baltimore Ave., Beltsville, MD 20705
* Corresponding author (rteller{at}wam.umd.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CEC, cation-exchange capacity • Db, bulk density • EG, eastern gamagrass • PVC, polyvinyl Cl • RCBD, randomized complete block design
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Recently, EG has received attention as a forage crop, a grass for vegetative hedges, and a crop to ameliorate marginal soils. As a forage grass, EG is relatively palatable and high yielding (Roberts, 1992). Reported yields range from 11.2 to 21.3 Mg ha-1 per year (Dewald et al., 1996; Dickerson and van der Grinten, 1990). Eastern gamagrass has a long lifespan potential, thick stems, and dense overlapping growth, making it appropriate for use in vegetative hedges, a soil conservation technique for slowing the flow of runoff water and retaining eroding soils (Ritchie et al., 2000).
Similar to the roots of wetland plant species, EG roots contain aerenchyma, providing continuous gas-filled spaces that can transport O2 for root respiration in inundated soils (Schussler and Longstreth, 1996). Drew et al. (2000) reported that aerenchyma forms constitutively in EG, without external stimulus. This is contrary to aerenchyma formation in some species which occurs when roots are subjected to hypoxia, causing cortical cells to disintegrate, leaving strips of cell wall material connecting inner and outer cortical cells (He et al., 1996; Schussler and Longstreth, 1996).
Eastern gamagrass is especially useful because of its tolerance to adverse subsoil conditions, such as extreme acidity and compaction. Foy (1997) showed that EG tolerates acid and Al-toxic conditions in both nutrient solution and soil. In the field, EG appeared to grow normally in Al-toxic compact soils and did not respond to liming at pH levels from 5.1 to 5.8 (Foy et al., 1999).
If EG roots can penetrate compacted soils, they may be able to ameliorate restrictive soils layers by providing channels for root development of subsequent crops (Clark et al., 1998; Dexter, 1991; Elkins, 1977). Clark et al. (1998) measured EG root distribution in two soil profiles and observed substantial root growth in clayey, acid, high Al subsoil layers. They also saw deeper root penetration of crop roots grown in fields subsequent to stands of EG, with crop roots following channels made by EG. Therefore, in addition to its use as a forage crop and in vegetative erosion barriers, EG may have potential for use in reclamation of marginal field soils because of its ability to penetrate layers of soil prohibitive to other crops.
The ability of EG to withstand drought and grow through soils with high soil strength and low pH needs further clarification. Kemper et al. (1998) hypothesized that EG avoids, rather than tolerates, drought because it can reach water below the claypan, a subsoil layer of clay accumulation that can impede the root growth of other species. Do EG roots penetrate compacted soils when they are water saturated and soft (exhibiting low soil strength) because of an ability to tolerate hypoxic conditions? Or do EG roots penetrate claypans and dense soil layers because of their ability to exert high penetration pressure? By studying EG root penetration through soil layers with controlled impeding conditions, the mechanisms of its root penetration may be better understood.
The objectives of this study were to determine how EG root growth is affected by extreme soil acidity (pH <4.5), compaction (Db = 1.7 g cm-3), and wetness (
m = 0 to -50 kPa). The three root inhibiting factors were evaluated both alone and in combination (Table 1). These effects also were assessed by comparison to sordan, a warm season forage grass known to be especially intolerant of the three factors studied. The expected responses were:
|
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Treatments and Experimental Design
Treatment combinations resulted from the factorial arrangement of two plant species (EG and sordan), and two levels of soil water potential (-10 and -300 kPa), soil pH (3.5 and 4.8), and soil Db (1.3 and 1.7 g cm-3). A full factorial arrangement of the three soil factors at two levels each would produce eight treatment combinations, but we used only six of them (Table 1) for each plant species. This was done to include the most important soil treatment combinations, while reducing treatments to a feasible number.
The treatments were applied to microcosms (described later) utilizing a RCBD. Four blocks (also serving as four replicates) were based on four distinct locations within the greeenhouse. This blocking was used throughout the study to block for other nontreatment factors (such as harvesting schedule and initial plant size) to minimize within-block variations. Measurements were made at different depths within individual microcosms. Therefore, each microcosm served as a main plot for the combination of plant and soil characteristics and depth was considered as a subplot factor.
Plant Material
Plant material consisted of ‘Pete’ EG and Sordan 79. Seeds were germinated in trays containing a peat-vermiculite mix and placed on a thermostatically controlled propagating mat (Progrow Supply Corp., Brookfield, WI) which maintained daytime temperature at 30°C. Germtec-treated seeds of ‘Pete’ EG and Sordan 79 (Lot 9729) were obtained from the Gamagrass Seed Co. (Falls City, NE) and Norvatis Seed Inc. (New Deal, TX), respectively. On 18 Aug. 1999, 7-wk-old EG seedlings and 5-wk-old sordan seedlings were selected for uniform size and vigor and transplanted into the microcosms (hereafter referred to as columns). Because EG is generally slower than sordan to germinate and establish, the two species were nearly equal in size at this time, sordan having five to six leaves, and EG four to six leaves.
Soil Material
Seedlings were transplanted into the top sections of the columns as described below. The top and bottom column sections were filled to a Db of 1.3 g cm-3 with Galestown sandy loam (siliceous, mesic Psammentic Hapludults) collected from the Ap horizon of a field at the Beltsville Agricultural Research Center. The field was planted with corn (Zea mays L.) at the time of collection and had been used for field crops including soybean (Glycine max L. Merr.), corn and alfalfa (Medicago sativa L. ssp. sativa) rotations for the previous decade. The soil was autoclaved, then passed through a 5-mm mesh to remove crop residues. A controlled release (15-5-11) fertilizer, Osmocote Plus (Scotts-Sierra Horticultural Products Co.; Marysville, OH), was added to the Galestown soil used in the top 20-cm section at a rate of 3.1 mg cm-3 to meet plant nutrient requirements. Galestown sandy loam was used for the top and bottom sections so that its textural contrast with the Tatum soil used in the middle sections would create a barrier to capillary water flow, facilitating the maintenance of differing water regimes in the top and middle sections.
The middle 30-cm column section contained the test soil, collected from the Bt horizon of Tatum clay loam near Orange, VA, less than a month before we started the experiment. The Tatum soil was passed through a steel blade hammermill, which broke large peds into aggregates smaller than 20 mm in diameter. Rocks, roots, and debris were removed at this time. The cation-exchange capacity (CEC) of unlimed and limed Tatum soil was measured using the compulsive exchange method for acid soils described by Rhoades (1982). Particle-size distributions of the Tatum and the Galestown soils were determined with a modified sedimentation method. Particle density was determined using the pycnometer method described by Blake and Hartge (1986).
Soil pH in 1:2 soil/water suspensions (pHw) was determined for Tatum and Galestown soils using samples taken from the columns at the end of the experiment. The exchangeable Al of limed and unlimed Tatum soil samples was determined in a 1 M KCl extract using an atomic absorption spectrophotometer with a N2O flame.
For the lime treatment factor, Tatum soil for limed treatments was amended with 3.2 g kg-1 Ca(OH)2 to raise the soil pH to 4.8. This was according to a preliminary soil titration in which 25-g soil samples were mixed with 0 to 550 mg Ca(OH)2 and incubated for 18 d. Unlimed treatments were left unamended, with a pHw of 3.5 ± 0.1.
Moist Tatum soil was tamped into the middle section of each column to establish the second treatment factor, Db of either 1.3 or 1.7 g cm -3. A Db of 1.3 g cm-3 is considered to be a nonrestrictive bulk density and 1.7 g cm-3 is considered to be potentially root-restricting compaction for a clay soil (Veihmeyer and Hendrickson, 1948; Zimmerman and Kardos, 1961). The soil was incubated under greenhouse conditions for 1 wk to allow the pH of the soil to equilibrate.
The third treatment factor, soil water potential, was adjusted to create two levels of soil strength and O2 availability. Columns maintained at high water potentials close to saturation (0 to -50 kPa) were intended to contain little or no air-filled pore space, and to exhibit low soil strength. High bulk density combined with high water potential yielded potentially hypoxic conditions unsuitable for most root growth. Columns with low water potentials (-400 to -600 kPa) were intended to provide adequate water while maintaining well aerated, higher soil strength conditions.
Using a tension table and a pressure plate apparatus, soil water release curves were developed for the Tatum soil at both high and low bulk densities, and for the Galestown sandy loam at the bulk density used. The air-filled porosity was estimated from the soil water release curves and soil particle density measurements, using the equations for porosity:
 | [1] |
 | [2] |
m is gravimetric soil water content.
Columns
Twelve plant and soil treatment combinations and four blocks required 48 columns. The columns were constructed using schedule 40 PVC pipe with 7-mm thick walls and 15-cm i.d. Each column was split lengthwise to facilitate opening at harvest. Each column was divided horizontally into three sections (Fig. 1)
. The top section was 20 cm in height, the bottom was 10 cm and the middle section was 30 cm. Circular cuffs with deep flanges were fashioned from stainless steel to hold circles of screening to the base of each column. Two to three layers of 1.4-mm mesh fiberglass screen were placed within each base to allow drainage but check the loss of soil. Steel rings and plastic pull-ties were used to belt the two halves of each section of column together, and duct tape was used to connect the three sections. Portholes (13-mm diam.) were drilled into the sides of the columns for the insertion of EH and pH electrodes, and for the installation of drip emitters (Chapin Watermatics, Inc., Watertown, NY) for the irrigation system. Five holes (16-mm diam.) in each column were threaded for the installation of mini-elbow tensiometers (1.5-cm ceramic tip) (Soil Measurement Systems, Tucson, AZ). Portholes not in use were sealed with rubber stoppers. The columns were arranged 30 cm apart on two adjacent moveable greenhouse benches. Two replicates of the experiment were set up on each bench.
|
Soil water potential was monitored on randomly selected columns (representing an equal number of all treatment combinations, using three mini-elbow tensiometers in each column. Soil water potential was monitored daily using a Tensimeter transducer (Soil Measurement Systems, Tucson, AZ) to measure the vacuum under each tensiometer septum. Irrigation amount and timing were adjusted to maintain desired water potentials in the top and treatment sections. Once a week, EH and pH measurements were made on the Tatum soil by inserting either a Pt electrode (Model 96-78; Orion Research Inc., Cambridge, MS) or a flat-tip pH electrode (Gel-filled Combination Electrode, model number 34105-026; VWR Scientific Instruments, Bridgeport, NJ) into the soil through a porthole in the column and twisting the electrode to establish a soil slurry around the tip. Only columns given high water treatments were wet enough to allow measurement of EH and pH in this manner.
Plant Harvest
Harvest of plants began 4 wk after transplanting. Because of logistical limitations not all treatment combinations across the four blocks could be harvested and analyzed at once. Therefore, harvesting was done on a block per week basis. This avoided confounding of a possible time effect with treatment effects (i.e., any possible time effect was confounded with the block effect and therefore removed from experimental error).
At harvest, plant height was measured from the soil surface to the tips of the three tallest leaves. Plants were cut 1 cm above the soil surface, placed in brown paper bags, dried at 80°C in a forced draft oven for 3 d and then weighed. The tape holding the top section to the middle section was removed and a serrated knife was used to cut the soil and roots at the joint. Penetrometer readings were taken on the middle section, using a penetrometer with a 2-cm2 tip (Eijkelkamp Agrisearch Equipment, Glesbeek, The Netherlands). Readings were taken at 5-cm intervals down the length of the section. Two sets of readings were taken, in opposite halves of the column.
The middle section then was separated from the bottom section. The outer 1 cm of the middle soil section was separated from the inner core of the middle section in 5-cm horizontal increments using a steel ring and a serrated knife. The inner and outer portions of each 5-cm depth increment were sampled separately for determination of root mass and density and soil mass. Two soil samples of 
10 g each were taken from both the top and bottom sections of each column, and from each of the 5-cm layers of inner soil into which the middle section had been partitioned. The soil samples were placed in small manilla envelopes and weighed immediately. They were dried at 80°C for 3 d and then weighed again to determine soil water content. These soil samples were saved for determination of pH and exchangeable Al.
The inner and outer portions of the Tatum soil layers and the Galestown soil top and bottom sections were placed in water to soak for 48 h to loosen surrounding soil. After 48 h of soaking, the soil core with its intact segment of the root system was placed on a 20-cm diam. sieve with 2-mm openings. A stream of tap water was used to wash the soil from the roots, which were then retained on the sieve and collected with tweezers. The washed roots were soaked for an additional 24 h, rinsed, patted dry, and weighed. They were then dried at 80°C for 48 h and reweighed.
Interior root density was calculated as the dry root weight in the interior portion of the 5-cm layers of Tatum soil divided by the volume of the sample of soil. This was done because of some variation in the size of samples. This volume was calculated as (7.5 cm)2 x 3.14 x 5 cm x (grams of inner soil for layer x grams of total soil for layer-1).
Statistical Analyses
Statistical analyses were done using SYSTAT software (SPSS Inc., 1997). Treatment effects were considered significant at P 
0.10.
Since the first 5 cm of Tatum soil was mixed with varying amounts of Galestown soil from the top section, only the 5- to 30-cm depth of the test section was considered in analyses of interior root density and mean interior root density. Only in the ANOVA of total root weight in the 60-cm column were the roots of the 0- to 5-cm depth of the test section included.
Because of significant heterogeneity of variances between the two species, interior root density and total root weight in the Tatum soil test section were normalized. These normalized variables were calculated by dividing the original value for each species by the maximum species value for each block. These values then were used for direct comparison of treatment effects between the two species.
For nonnormalized variables including total root weight and interior root density, ANOVA was done separately for each species because of heterogeneity of variances of the two species. Root weight and interior density were analyzed both as the sum for a column and for individual intervals within a column. If the depth effect was significant, post hoc hypothesis testing was carried out using the Boniferroni test for significant differences between individual soil depths. Limed and unlimed treatments across the three levels of soil strength and the effect of soil strength across the liming conditions were compared using contrasts. Soil strength contrasts were based on theoretical considerations of bulk density and soil water content as well as penetrometer resistance measurements taken at harvest (Busscher et al., 1997). Using these considerations, the six treatments were assigned to either low, medium, or high soil strength (Table 1).
Soil water release curves for the Galestown soil and Tatum soil at high and low Db were determined by distance weighted least squares regression of data for soil water content and soil water potential (SPSS Inc., 1997).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
|
Soil Strength
Tatum particle density was measured to be 2.95 g cm-3. Using Eq. [1], the total porosity was calculated to be 0.42 for the high and 0.56 cm3 cm-3 for the low Db soil. Scheduled irrigation and supplemental applications of water resulted in mean matric potentials of -10.7 ± 25.7 and -299.7 ± 189.2 kPa for the high and low water treatments respectively, over the course of the experiment (Fig. 3)
. At harvest, high water treatments resulted in 18% higher average water contents than the low water treatments, with gravimetric soil water contents of 0.34 ± 0.04 cm cm-3 versus 0.28 ± 0.04 cm cm-3 (data not shown).
|
Contrary to the expected responses, changes in the soil strength of clayey subsoil layers had no significant effect on EG mean interior root density and interior root density at different depths in the Tatum soil (Tables 3 and 4, Fig. 4) . However, soil strength did affect total interior root weight and total root weight per column (Table 3). Total root weight per column was highest for EG in high soil strength treatments. Eastern gamagrass roots in high soil strength did not show stubby growth and lack of fine root hairs, morphological features associated with mechanical impedance.
|
The absence of the expected increase in interior root density in medium versus high soil strength treatments may have been because of the low air-filled porosity of medium soil strength treatments. This suggests that even with the constitutive expression of aerenchyma seen by Drew et al. (2000), EG root growth may be stunted under nearly saturated conditions at this early stage of development. Other studies presently in progress (D. T. Krizek, personal communication, 2000) confirm this likelihood.
In general, sordan responded to soil strength differently than did EG. Sordan total root weight per column was not significantly affected by soil strength because of compensatory root growth in the top and bottom sections (Tables 3 and 4). Sordan average interior root density (Table 2) was lowest in medium soil strength treatments and highest in low soil strength treatments. This may have been because of the low air-filled porosity in the medium soil strength treatments. However, for the 20- to 30-cm depths of Tatum, sordan interior root density followed the order, low > medium > high soil strength (Fig. 4). The interior root density at the 20- to 30- cm depths in Tatum soil agreed with results reported for other species, namely, that increased soil strength is inversely related to root growth (Bengough, 1991; Bengough, 1997; Ehlers et al., 1983).
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The characteristics of tolerance to acid and Al and to high soil strength conditions may make EG valuable in establishing grassed buffers, vegetative conservation barriers, and pastures. These characteristics of EG may allow EG to form root channels, which, in turn, may ameliorate subsoils for the growth of less tolerant crops, allowing land now considered unproductive to be used more profitably after the growth of EG.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication June 26, 2001.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
.
