| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Plants, Soils, and Biometeorology, Utah State University, Logan, UT 84322
b Department of Civil and Environmental Engineering, Utah State University, Logan, UT 84322
* Corresponding author (bruce.bugbee{at}usu.edu)
Received for publication April 15, 2005.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: GC-MS, gas chromatography–mass spectrometry • LC-MS, liquid chromatography–mass spectrometry • TOC, total organic carbon
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
An ideal axenic plant culture system should provide: (i) continuous control of CO2, temperature, humidity, and light in the shoot zone; (ii) adequate nutrients, water, and O2 and absence of light in the root zone; (iii) appropriate mechanical impedance to root elongation; (iv) inert materials; (v) maintenance of sterility; (vi) rigorous tests to verify sterility; (vii) capability to apply treatments such as nutrient stress or inoculation; and (viii) access to collect root-zone leachate.
Although it is necessary to avoid unintentional stresses in axenic studies, not all in vitro plant culture systems promote plant health. Agar plates (Heist et al., 2002) or agar with Millipore (Billerica, MA) membranes (Meharg and Killham, 1991) have been used as simple axenic plant culture setups, but do not supply adequate airflow or allow uniform nutrient, water, or O2 delivery to the root surfaces. Agar media allow continuous monitoring of sterility but do not facilitate long-term studies. It is also difficult to extrapolate results from agar plates to the field.
Axenic liquid hydroponic cultures have been widely used (e.g., Mench and Martin, 1991; Groleau-Renaud et al., 1998), particularly for short-term (<1 wk) studies. However, root morphology and growth are significantly altered in hydroponics, compared with the field, due to reduced mechanical impedance and absence of root hairs. The viability of microbes that might be inoculated into a hydroponic system is limited by the lack of surfaces to colonize.
Soil provides a growth medium more like field conditions and has been used for axenic plant culture (Whipps and Lynch, 1983). However, soil is difficult to sterilize and structural and geochemical changes occur from autoclaving or 
irradiation.
Porous solid substrates also provide growth conditions similar to the field but are more easily sterilized. Biondini et al. (1988) enclosed plant roots in pots of sterilized fritted clay with ports for solution input and output. Sand has also been used as a growth medium for axenic plant culture (Ayers and Thornton, 1968; Lipton et al., 1987; Hodge et al., 1996). Sand has fewer reactive surfaces than soil or fritted clay. Certain types of commercially available sand are more inert than others due to increased purity of silica in their composition, which decreases reactivity. Glass beads are considerably more expensive than sand and may affect nutrient solution composition (Sandnes and Eldhuset, 2003).
Sterilization methods most commonly used include autoclaving of the system components, irradiation or autoclaving of the growth medium, and surface sterilization of seeds by soaking in dilute solutions of H2O2, NaOCl, or HgCl2.
Most previous studies have used a preliminary test for contamination by germinating surface-sterilized seeds on agar plates. Many studies also performed a secondary test for contamination using a plate count of solution at the end of the study (Mench and Martin, 1991; Basu et al., 1994; Groleau-Renaud et al., 1998). Other studies did not indicate how axenic conditions were verified after planting (Barber and Lynch 1977; Hodge et al., 1996). Since <1% of all soil microbes are capable of growing on agar plates, additional tests such as light or fluorescence microscopy are required because contamination by certain species may go undetected (Brock, 1987).
There are few long-term (i.e., >40 d) axenic plant studies because it is difficult to keep the systems free of microbes. Lipton et al. (1987) grew alfalfa for 24 d under sterile conditions in a completely enclosed system with a tube containing sand for the root-zone container and an inverted beaker as the shoot-zone container, but the setup did not provide airflow around the shoots. Hodge et al. (1996) developed an enclosed system that allowed inoculation of microorganisms and used radio-labeled CO2 to track the flow of assimilated carbon. Four ports were drilled into the growth container that allowed air and fluid exchange to and from the system during a 34-d growth period. Paterson and Sim (1999, 2000) modified these systems to reduce carbon contamination by the system components and to reinforce seals. Our goal was to develop a plant culture system in which axenic conditions could be maintained while facilitating optimal plant growth for several months to study root exudates. We sought to design a simple system with inert components that would be less prone to contamination with time, identify unique sterile techniques for axenic plant culture, use adequate methods to verify the absence of microbes, and include measures to promote plant health.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
and n that describe different sand grades from Schroth et al. (1996). Each layer's lower boundary corresponds with the intersection of the 90% saturation level or maximum water content of the respective retention curve. The shape of the retention curve determines the lower water content level found at the interface with the adjacent layer above. The more coarse-grained sand formed the bottom layer and increasingly finer grain (and pore) sizes formed the layers above. Intermediate water retention properties were obtained by a 50:50 mix using two adjacent sand sizes together, the water retention curves of which were assumed to be represented by the average of the neighboring layer parameters.
|
|
|
Water Availability
Transpiration tests were conducted to determine the readily available volume of water in the columns. Transpiration rates of mature crested wheatgrass grown under nonaxenic conditions were measured for up to 40 h. Water use was determined gravimetrically. The volume of plant-available water was estimated based on the water content at which transpiration rates began to decline.
Sorption to Ottawa Sand
Sorption tests were performed using unplanted sand columns with the Fe-loading treatment. Twenty milliliters of 0.02 M KCl and 0.02 M oxalic and malic acids (pH raised to 8) plus 20 mL of deionized water were added to three unplanted sand columns, followed by five rinses of 40 mL of deionized water. The recovery of the solutions was monitored using an electrical conductivity meter to determine the difference between the electrical conductivity of the solution added to the columns and leachate collected. Amounts of organic acids used for the sorption test represented cumulative amounts of C exuded by plants during the 70-d study.
Plant Growth Container: Shoot Zone
The top section of the container was made from a second 220-mm-long glass column connected to the root-zone column by a ground glass joint. The column was sealed at the top with an open-cell foam plug to prevent microbial contamination (Fig. 1).
Plant Culture
One pregerminated crested wheatgrass seed (cv. CDII, Asay et al., 1997) was planted in each root-zone column, except the unplanted column. All columns were maintained for 70 d in a laminar flow hood. The hood was modified for plant growth by installing two high-pressure sodium lamps, which supplied a PPF (photosynthetic photon flux) of 550 µmol m–2 s–1 during a 16-h photoperiod (Fig. 3). The air temperature was maintained at 25°C. Plants were watered to excess with filtered and autoclaved nutrient solution (pH = 5.5) to obtain at least 0.025 L of leachate and maintain healthy plants. The frequency of watering increased from once every 3 to 4 d at the beginning of the study to once every 2 d at the end of the study. The composition of the nutrient solution was predetermined under nonaxenic conditions for optimal growth (Table 2). Nutrient solution concentrations were reduced to half-strength after Day 35 to minimize potential salt buildup in the root zone. Filtered airflow through the shoot-zone columns was supplied during the light period. Airflow was maintained at 0.065 L min–1 after emergence and was gradually adjusted to 1.0 L min–1 at harvest to prevent condensation inside the shoot-zone columns. Airflow manifolds were mounted directly to the racks holding the columns, which improved laminar flow of sterile air through the workspace (Fig. 3).
|
|
Collection of Leachate
Amber vials containing nutrient solution leachate and root exudates were replaced with sterilized, empty vials after each watering. Leachate samples were capped and stored at 4°C until analysis within 24 h.
Procedures to Minimize Adsorption and Desorption of Carbon
Adsorption and desorption of carbon by system components would significantly interfere with analysis of root exudates. Selection of inert components and treatments of the components were necessary to assure that the plants were the only source of C in the leachate samples, and that minimal amounts of plant-derived material adsorbed to system components.
Organic C was removed from the sand to minimize background C levels. Heating in a muffle furnace at 600°C was effective in removing C from small volumes of sand (<25 g) but was impractical for the large volumes of sand required for this study since the heat could not penetrate the entire volume of sand to effectively volatilize organic C. Sand was instead treated according to Guy (1969). Sand was washed with deionized water, dried at 90°C, washed with 30% H2O2 (v/v), and dried overnight at 90°C while stirring several times to increase the reaction of organic C with H2O2. The H2O2 treatment was repeated and the sand was given a final rinse with filtered (0.45-µm) deionized water. The sand was then poured into the glass columns and autoclaved twice, 1 d apart, for 1 h at 145 kPa and 121°C to allow any remaining endospores to germinate between autoclavings.
Silicone stoppers were used instead of rubber stoppers. All glass and silicone components of the system were rinsed with methanol, dried at 80°C to remove trace organic C, and autoclaved for 45 min at 145 kPa and 121°C.
Procedures to Minimize Microbial Contamination
Improved methods to facilitate long-term axenic plant culture were used, particularly techniques that maintained sterility while allowing optimal growth.
To achieve a high percentage of microbe-free seed, it was necessary to agitate crested wheatgrass seeds on a shaker for 60 min in a solution with 20% Clorox (Clorox Co., Oakland, CA) and 0.1% Tween 80 as a wetting agent (VWR International, Chicago, IL). Seeds were rinsed with sterilized, deionized water and placed on petri plates with 1/10 strength Difco nutrient broth (Becton, Dickinson and Co., Franklin Lakes, NJ) and Bacto agar (1.5%, Becton, Dickinson and Co.) in a 25°C incubator for 3 d to test for residual microbial activity on the seed surface. After 3 d, the radicles were 
15 mm long. On average, one seed out of 50 had residual contamination. Seed vigor compared with untreated controls was not significantly reduced. Germination rate was 72% after 6 d.
Plant-growth containers were assembled and seedlings were transplanted from the petri dishes to the root-zone columns in a laminar flow hood. Forceps and column lips were flamed before use. All manipulations were done using sterile gloves.
Air was pumped through the upper part of the glass plant-growth columns to supply CO2 and remove H2O vapor. This air was filtered through glass wool and foam plugs before and after the pump, and through a sterilized, bacterial air filter (aerosol retention = 0.3 µm, Pall Gelman Sciences Corp., Ann Arbor, MI) before entering each column.
Nutrient solution was filtered through a 0.45-µm membrane, then autoclaved at 145 kPa and 121°C for at least 45 min. After autoclaving, the nutrient solution was allowed to cool completely before using it to water the plants. Separate flasks of nutrient solution and sterile syringes were used to water each group of plants. Septa were cleaned with 70% ethanol and allowed to dry before each injection. A new sterile needle was used for each plant. A syringe was discarded immediately if it touched anything other than the nutrient solution.
The HEPA (high efficiency particulate air) filter in the laminar flow hood was replaced at the start of the first trial and was tested and certified. A calibrated pressure gauge was used to monitor the autoclave and ensure that appropriate pressure (at least 103 kPa) was maintained during the entire cycle.
Each improvement in the materials or sterile technique while developing this system increased the length of time that axenic conditions were maintained (Fig. 4). Major improvements included growing plants in a laminar flow hood in Trial 4, enclosing the entire plant in Trial 5, and wearing sterile gloves and covering seedlings with premeasured sterilized sand in Trial 6. In the most successful trial (Trial 6), 85%—or 12 out of 14—plants remained free of microbes for 70 d, after which the trial was terminated due to plant size. The materials and methods as well as results from Trial 6 are reported here.
|
In addition to plating, a direct total microbial count was used to double-check sterility. The Epifluorescent Microscopic Method (American Public Health Association et al., 1998) was used: 5mL of leachate were stained with acridine orange and filtered through a 0.2-µm nonfluorescent membrane. The membrane was then mounted on a microscope slide and observed at 100x under an ultraviolet light. Microbes could be seen in the leachate of plants that were identified as contaminated on the plates.
A rhizoplane stain using phenolic aniline blue (Rovira et al., 1974) was also performed on root samples at the end of the study. Young root segments and root tips, where microbial symbiosis or infection is most likely to occur (Curl and Truelove, 1986), were excised, stained with phenolic aniline blue solution (Schmidt and Paul, 1982) for 3 min, and mounted on slides.The root segments were then rinsed in sterile filtered water, mounted in water on a microscope slide, and observed under 100x. This confirmed the visual observations in the petri dishes.
Harvest Procedures
Plants were harvested on Day 70 in a laminar flow hood using sterile techniques. The plant growth columns were disconnected from the air flow and removed from the storage rack. After the upper glass column was removed, the plant was slowly pulled from the root-zone column. This removed much of the sand because it adhered to the roots. Sand that fell away from the roots was labeled "bulk sand" and saved in a sterile container for analysis. The roots were then gently shaken to remove additional sand, which was labeled "rhizosphere sand" and stored in a sterile container. A small segment of root was then excised for microscopic observation. The shoot was cut off, rinsed with deionized water, and dried in an 80°C oven. The roots and remaining attached sand were stored in a sterile container at –20°C. Dried shoot tissue was prepared by grinding and HNO3–H2O2 digestion (Mills and Jones, 1996) and analyzed by ICP-OES (inductively coupled plasma–optical emission spectroscopy) for P, K, Ca, Mg, S, Fe, B, Zn, Mn, and Cu at the Utah State University Analytical Laboratories, Logan.
Exudate Analysis: Leachates
Carbon in the leachate of six columns planted with one plant as well as one unplanted control column was analyzed with a total organic C analyzer (Model Phoenix-8000, Tekmar-Dohrmann, Cincinnati, OH), which uses an ultraviolet–persulfate reaction to oxidize organic C to CO2. All samples were frozen at –20°C and analyzed at the end of the study using a single standard curve.
Because of their chelating properties and ubiquitous presence in previous exudate studies, we examined organic acids for a qualitative analysis of exudates. The analytical methods used to determine exudate composition were the same as those used by Kloss et al. (1984), whereby the acids were esterified to increase their volatility, and analyzed as their methyl esters using GC-MS (gas chromatography–mass spectrometry).
To esterify the samples, 1-mL aliquots of each leachate (or extract of roots or sand, see below) were added to 3 mL of methanol and acidified with 0.6 mL of 50% H2SO4 (v/v). The acidic methanol mixture was heated for 1 h at 50°C, then cooled and diluted with an additional 3 mL of H2O. One milliliter of chloroform was added and the sample was vigorously shaken, then incubated about 30 min. The chloroform layer was removed and the components in the chloroform were quantified using GC-MS (Models 6890N/5973, Agilent Technologies, Palo Alto, CA). Chromatography conditions were: 1-µL injection; column flow rate = 0.6 mL min–1; split ratio = 3.6; column DB-624 = 30 m x 0.25 mm x 1.4 µm; temperature program = start at 50°C, increase 5°C min–1 to 200°C and hold for 5 min. Methyl esters of the dicarboxylic acids were quantified using pure compounds (Aldrich Chemical Co., Milwaukee, WI). A concentration range of 0.5 to 10 mg kg–1 showed good linearity for calibration of each organic acid (r2 = 0.997–0.999).
Exudate Analysis: Growth Column
To extract organic acids, sand attached to the roots was separated into approximately 5-g samples. Three milliliters of 0.1 M NaOH were added to each sample. The sample solutions were intermittently mixed for 1 h at room temperature and then separated from the sand matrix by filtration. The extracts were then analyzed using the esterification/GC-MS method described previously for analysis of leachates. A similar extraction was used to determine TOC in the rhizosphere sand using 50 mL 0.1 M NaOH.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
v [volumetric water content] = 0.32) based on gravimetric tests. The plant-available water in the columns was about 0.035 L, based on the volumetric water content at which transpiration rates of crested wheatgrass plants started to decline (Fig. 6). Plants were watered to maintain water loss below 0.035 L to minimize water stress.
|
|
Recovery of Compounds from the System
As an alternative to collecting exudates directly in the nutrient solution, transfer of roots from nutrient solution to deionized water for 1 to 24 h to collect root exudates has been used in some short-term solution culture studies (Fan et al., 1997; Ratnayake et al., 1978; Ström et al., 1994). This procedure would reduce salt interference with analyses but may bias results since the amount and composition of root exudates can reflect the nutritional status of the plant. Nutrient deficiencies may change the composition or induce increased amounts of exudates (Kraffczyk et al., 1984; Römheld, 1991; Ratnayake et al., 1978). Wang et al. (2002) have shown that activation of genes associated with P, K, and Fe deficiencies can be induced within 1 h after withholding these nutrients from the plant. Since this response was confined to the root, a rapid response in root exudate composition to nutrient deficiency is also probable. Aulakh et al. (2001) found 0.01 M CaSO4 to be a good leachate collection medium that did not interfere with TOC or HPLC analyses and did not increase TOC or the proportion of sugars exuded, as was observed with deionized water. We recommend an appropriate qualitative analysis that is not affected by salt content of the exudate sample.
Achieving and Maintaining Axenic Conditions
Since the stage when contamination is most likely and plants are most susceptible to damage occurs when seedlings are being planted, rapid transfer of the germinated seed from the agar plate to the moistened sand is necessary. Initial sterility was improved by minimizing the exposure time of the growth columns and germinated seeds to the surrounding air during planting by pre-assembling components before sterilizing. It is not necessary that the entire seedling be covered by sand, just the seed coat and root initial. Additional time spent burying the seedling during planting (i.e., with a spatula) increases exposure time and increases potential for microbial contamination. Extra sand was autoclaved separately to pour on top of seedlings during planting. This eliminated the need to bury the seedling with a spatula and reduced the number of manipulations and length of time the column was exposed to the surrounding air. It was useful to practice the planting techniques under nonsterile conditions before doing the sterile manipulations.
In both of the contaminated plants, roots grew through the glass wool, emerged out of the drain tube, and were briefly exposed when the surrounding vial was changed. Because a rhizoplane stain of roots from within the system was free of microbial growth, it is likely that only the emerging root segment was contaminated.
Although plants were kept in the laminar flow hood and sterile gloves were always worn when watering and replacing vials, direct exposure of the plant to the surrounding air resulted in contamination. This implies that long-term axenic plant culture would be unsuccessful in a nonsterile growth chamber.
Procedures for cleaning the system components probably reduced microbial contamination. The 20% H2O2 treatment used to reduce residual TOC on the sand was powerful enough to kill most microbial cells, and the deionized water washes physically removed microbial cells and spores from all components. All water used for cleaning was filtered, since any debris would interfere with microscopic observations.
Verification of Sterility
The use of three separate methods to confirm sterility provides a high level of confidence in the results, but all methods have limitations. Since acridine orange stains DNA, all living material is stained, including root cells. This might result in false positive identification of contamination. However, contaminating microbes would most likely be attached to surfaces, and the sand and glass wool could have filtered these microbial cells before they reached the leachate. Additionally, this method does not allow time for microbial multiplication so the microscopic observations might miss single cells, compared with agar plates in which growth makes the microbes more visible. The acridine orange method was effective in identifying contaminating microbes in the leachate from the two plants that had roots emerging from the bottom of the column.
The phenolic aniline blue stain of root segments was the most direct means of assessing contamination but only one slide was made for each plant and this slide represented only a small part of the root system. This method could miss contamination if it was confined to one part of the root zone.
Exudate Analysis
Total organic C was monitored in the leachates throughout the trial. Exudate concentrations were converted to micrograms of C per plant per day based on values in milligrams per kilogram provided by the TOC analyzer and the time between waterings (Fig. 7). Leachate TOC was highly correlated with plant size (r2 = 0.93). Carbon leached from the unplanted control ranged from 15 to 60 µg C d–1, or about 10% of the planted systems. The TOC remaining on the sand at the end of the study was, on average, 26% of the leachate TOC (Table 5). The 17% TOC found in the rhizosphere (sand immediately adjacent to the root) is probably an overestimate because small root pieces and root hairs contaminated the sand. The presence of root hairs, small pieces of roots, border cells, and compounds that reacted with the Fe may all contribute to insoluble organic compounds that were detected as residual TOC on the sand.
|
The lower amounts of exudate in this study compared with the literature may be due to plant age or growth conditions. Plants in other studies were all younger than in this study, and most plants in previous studies were grown hydroponically. Average exudation rates in this study would have been higher if the study had been terminated earlier since exudation rates decreased at the end of the study. This study was continued as long as possible to evaluate exudation on a long-term basis.
Similar to the rates of TOC release, the rates of organic acid release in the exudates also peaked before the end of the study. Fumaric, succinic, oxalic, malonic, and malic acids were quantified (Table 6). Organic acids detected in the rhizosphere sand were minimal.
|
0.4 mg L–1 for the leachates using 5-mL samples, and 0.6 mg kg–1 for the sand using 1-g samples. The esterification had poor efficiencies for hydroxy-diacids (carboxylic acids with a hydroxyl group) since it was not designed for these compounds. It is possible that hydroxyacids other than those detected were present in the exudates. Organic acids were corrected for recovery efficiencies, which were: oxalic acid 22 ± 3%, n = 10; malonic acid 50 ± 6%, n = 10; fumaric acid 51 ± 6%, n = 10; succinic acid 56 ± 4%, n = 10; and malic acid 15 ± 7%, n = 8. There were several challenges with the organic acid analysis. Acids were present in low concentrations in the leachate, and sample volumes were too low to be concentrated, for example by lyophilization. The biggest difficulty was in derivatizing the acids to methyl esters for analysis, especially acids like malic acid that contain a hydroxy group in their structure. The presence of the hydroxy group on a neighboring C atom changes the reactivity of the carboxyl group, making the esterfication reaction less likely to go to completion and thus leaving some acid in the sample unreacted. Low concentrations coupled with low derivatization rates may have reduced the number of acids. For example, citric acid is a common acid found in root exudates. Citric acid mobilizes phosphate and chelates Fe and Al in the rhizosphere, but it was not detected in this study. Citric acid is also a hydroxy acid, thus its decreased ability to be esterified and its presence in low concentrations may have contributed to its absence of detection in this study. The absence of citric acid may also be due to adequate Fe supplied to the plants, thus reducing the need to exude Fe-sequestering compounds.
The composition of the other exudates is unknown. Other compounds such as sugars and amino acids may also have properties useful for phytoremediation.
The variability associated with the organic acid analysis (Table 6) is probably due to the poor esterification efficiency. High performance liquid chromatography and ion chromatography have been used for exudate analysis, but it is more difficult to separate the different acids with these methods and they are less sensitive than gas chromatography (Szmigielska et al., 1995). Ion chromatography is typically used for general anion analysis and does not require derivatization of the acids but requires higher concentrations than those in the leachates from this study. A more comprehensive method was used by Fan et al. (1997) with nuclear magnetic resonance and GC-MS for de novo identification of exudate components, which is useful for analyzing unknown compounds. Liquid chromatography–mass spectrometry allows analysis without derivatization and provides structural information for positive identification of unknowns, but LC-MS instruments are very expensive. Depending on exudate concentrations and instrument availability, these methods may provide analyses with less variability.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Since this system allows the application of treatments and collection of leachates to monitor root exudates, it is suitable for many types of studies, including the effects of inoculation with pure microbial cultures and the effects of biotic and abiotic stresses such as heavy metals, all of which can provide insight into the dynamics of root exudation.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Child, C. D. Miller, Y. Liang, R. C. Sims, and A. J. Anderson Pyrene Mineralization by Mycobacterium sp. Strain KMS in a Barley Rhizosphere J. Environ. Qual., July 17, 2007; 36(5): 1260 - 1265. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Henry, W. Doucette, J. Norton, and B. Bugbee Changes in Crested Wheatgrass Root Exudation Caused by Flood, Drought, and Nutrient Stress J. Environ. Qual., May 7, 2007; 36(3): 904 - 912. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
