| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a CIRAIG: Interuniversity Reference Center for the Life Cycle Assessment, Interpretation and Management of Products, Processes and Services, 2500 ch. Polytechnique, Montreal, Quebec, H3C 3A7, Canada
b CEAEQ: Centre d'Expertise en Analyse Environnementale du Québec, Ministère de l'Environnement du Québec, 2700, Einstein, Sainte-Foy, Quebec, G1P 3W8, Canada
c NSERC Industrial Chair in Site Remediation, Chemical Engineering Department, École Polytechnique de Montréal, 2500 ch. Polytechnique, Montreal, Quebec, H3C 3A7, Canada
* Corresponding author (louise.deschenes{at}polymtl.ca)
Received for publication March 21, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: DOC, dissolved organic carbon • DTPA, diethylene triamine pentaacetic acid • ECx, effect concentration affecting x% of the population • Kd, adsorption coefficient • LC50, lethal concentration affecting 50% of the population • OM, organic matter
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The influence of soil parameters on metal availability to organisms is frequently assessed by studying metal concentration in tissue (Peijnenburg et al., 1999; Maiz et al., 2000). For example, the influence of soil characteristics was assessed on Cu uptake by lettuce, mustard, and barley (Impellitteri et al., 2003). The work of these authors, which grouped five previous studies on 41 soils from various countries, estimated that 80% of the variability of Cu uptake could be explained by the influence of pH and organic matter. However, using the internal concentration approach to assess environmental risk is arguable (particularly for essential metals) because body concentration is often regulated (Peijnenburg et al., 1999). Furthermore, the organisms' internal concentrations are not necessarily correlated with ecotoxicological effects (Rhoads et al., 1989). Hence, estimating risk by using the internal metal concentration has some limitations.
The influence of soil physicochemical properties on Cu ecotoxicity has been studied to a lesser extent than bioaccumulation and partitioning. Observations have nevertheless been made through reported tests. It has beenshown that increase in pH significantly contributes to lowering the Cu toxic effect on Swingle citrumelo and Lumbricus rubellus in natural soils (Alva et al., 2000; Ma, 1984). Increase in organic matter and clay contents was also shown to contribute to decrease Cu toxicity as observed by Boyd and Williams (2003). They measured Caenorhabditis elegans response on Albany soil (loamy, siliceous, subactive, thermic Grossarenic Paleudults; pH 6.1, 1.4% OM [w/w], 2% clay [w/w]) and Cecil soil (fine, kaolinitic, thermic Typic Kanhapludults; pH 5.7, 5.1% OM [w/w], 10% clay [w/w]) and found respectively 24-h LC50 of 230 (SD = 51) and 548 (SD = 53) mg Cu kg–1 (where LC50 is the lethal concentration affecting 50% of the population). The most specific study on the relationship between soil physicochemical properties and Cu ecotoxicity was conducted by Lock and Janssen (2001). These authors used a central composite design of 12 artificial soils to quantify the influence of pH (4 to 7) and organic matter (0–10% [w/w]) on Cu toxicity to Enchytraeus albidus (mortality). A significant correlation was found between 14-d LC50 and pH as well as with cation exchange capacity (CEC). The cation exchange capacity was mainly attributed to organic matter (r2 = 0.99).
Aging of metal contamination, on the other hand, has been poorly studied although some observations tend to show a difference between laboratory toxicity tests and field studies covering potential risk. A previous study on the subject showed a lower toxicity associated to 70-yr-old contaminated field soils (sandy clays, pH 6.5–7.0) compared to toxicity in similar soils freshly spiked with CuCl2 in the laboratory (Scott-Fordsmand et al., 2000). No decrease in reproduction was observed for Folsomia fimetaria for field concentrations up to 2900 mg Cu kg–1, while EC10 (where ECx is the effect concentration affecting x% of the population) reproduction was observed at 340 mg Cu kg–1 in the spiked soils. Lock and Janssen (2003) also found a significantly greater Cu toxic effect in freshly spiked soils compared to field soils with similar properties, but this could be caused by a decrease in pH immediately following spiking. From a physicochemical perspective, it was demonstrated that the humidity regime has an influence on Cd, Cu, Cr, Ni, and Zn redistribution over time. Following wetting and drying cycles for over 1 yr, Han et al. (2001) found that dissolved and exchangeable Cu fractions tended to decrease over time in a sandy soil and in a loessial soil. In addition, Tom-Petersen et al. (2004) observed a decrease in the dissolved Cu fraction as well as in the toxic response of Pseudomonas fluorescens over time in a sandy loam (pH 6.1, 1.4% OM [w/w], 3.7% clay [w/w]).
The influence of soil physicochemical properties on Cu ecotoxicity has been studied to a small extent considering the number of terrestrial species and soil types. Moreover, publications on aging of contaminated soils do not allow the drawing of a clear inference as to the quantification of the effect of aging on the ecotoxic responses. The present study aimed to quantify the effect of soil pH, organic matter, and clay content on the Cu ecotoxic response of two different organisms, barley and compost worm, and to analyze the effect of wetting and drying of soils on this same ecotoxic response. Artificial spiked soils were chosen for homogeneity and the absence of interfering contamination.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
A statistical design for artificial soil constitution considering pH, organic matter, and clay content was developed (Table 1). Four vertices (A, B, C, D) and a centroid soil (E) were necessary to evaluate single effects of soil components, as well as interactions between soil components and pH. Silt content remained constant at 25% (w/w) in each soil mixture.
|
Subsequently, 6 kg of each soil type were acidified to a pH of 5.5 with NaHSO4 salt (ACS grade) dissolved in deionized water (0.001 M), while the pH in the other 6 kg was increased to 7.5 with Ca(OH)2 powder (ACS grade). Both chemicals were purchased from Anachemia (Lasalle, QC, Canada). Hence, from the initial A through E soil mixtures, two series were obtained: Mixtures A1 through E1 at lower pH (5.5) and A2 through E2 at higher pH (7.5) for a total of 10 soil mixtures (Table 1).
Physicochemical Characterization of Soils
The pH of individual components (clay, silt, sand) and soil mixtures was determined according to Method D-4972-95a using a soil to distilled water ratio of 1:4 (w/v) (American Society Testing and Materials, 2000). Organic matter (Sphagnum peat moss) pH was measured according to Method D-2976-71 using a soil to water ratio of 1:16 (w/v) (American Society Testing and Materials, 2000). Tubes were sealed and agitated for 30 min on a wrist action shaker (Model 75; Burell, Pittsburgh, PA). The pH was measured in duplicate 48 h after decantation (Orion Ross 8175 BN electrode [Thermo Electron, Waltham, MA], Accumet Model 25 pH-meter [Fisher Scientific, Hampton, NH]). Organic matter content was obtained in duplicate by heating samples to 550°C (Karam, 1993). Field moisture capacity was measured by weighting the water content in the saturated soil mixtures obtained by inundation of soils with hot water (70°C) during 30 min and percolation on a Whatman (Maidstone, UK) GF/C paper filter until the interval between two drops of water reached 10 min (without any surface drying of the soil sample) (Carter, 1993). Table 1 presents soil characteristics.
Soil Contamination Procedure
From the 6 kg of each constituted soil mixture (A1 through E2), 3 kg were spiked with 6000 mg of Cu per kilogram of dry soil, while the remaining 3 kg was left uncontaminated. This concentration was chosen because it corresponds approximately to the EC50 (CI95%) values for barley growth inhibition (7-d shoot length, 7-d root length) and compost worm mortality (14-d LC50) in the centroid Soil E at pH 7.5, obtained in a preliminary test (results not shown). In addition, this level of contaminant can be compared to concentrations found on highly contaminated sites (Lock and Janssen, 2003; Kennette et al., 2002). A CuSO4 solution (4000 mg L–1) was used to spike each soil mixture using a soil to solution ratio of 1:1.5 (w/v) to get a homogeneous solution during agitation. Agitation was achieved in a customized end-over-end mixer during a 24-h mixing period followed by a 72-h resting period to allow for Cu adsorption. These time periods were chosen according to Tom-Petersen et al. (2004) who studied the influence of aging on the adsorption of Cu. For different amounts of Cu added to a sandy loam, the authors observed a mean Cu adsorption of 98% after 1 h while it reached 99% after 3 d of aging at 100% field moisture capacity (99.5% after 220 d). After contact with the solution, soil mixtures were centrifuged (Model J2-21; Beckman, Fullerton, CA) in 500-mL polypropylene bottles for 30 min (4000 x g) to remove the supernatant. The pH was readjusted following spiking. Soil mixtures were then air-dried (20 ± 2°C) for 5 d and stored in closed high density polyethylene (HDPE) jars at room temperature until the beginning of the experiment.
Chemical Analyses
Total Cu content was measured following a digestion with hot acid on a heating plate: 1.2 g of soil was digested twice with 20 mL of concentrated HNO3 at 150°C for 1 h, then with concentrated HClO4 (15 mL) and HF (1 mL) for 2 h at 150°C and finally filtrated on a Whatman GF/F 0.7-µm filter and diluted to 100 mL with deionized water (Kennette et al., 2002).
Chemical extractions were also performed in triplicate using:
The pH of each solution was adjusted when necessary with either HCl or NaOH.
Extraction A was conducted following a modified procedure proposed by Scott-Fordsmand et al. (2000): 3 g of soil were added to 30 mL of deionized water in 40-mL polypropylene centrifugal tubes. The tubes were shaken on a wrist action shaker for 2 h and centrifuged 20 min (27 000 x g) before being filtered on a Whatman GF/F 0.7-µm borosilicate glass filter.
The extraction using the 0.5 M CaCl2 solution (Extraction B) was performed according to the procedure proposed by Esnaola and Millan (1998). Aside from using a different reagent, the procedure was the same as that used for the water-extracted Cu fraction.
For the purpose of the present study, the extraction method using DTPA (Extraction C), originally developed by Lindsay and Norvell (1978), was performed with 1.5 g of soil sample in 15 mL of solution (soil to solution ratio of 1:10) instead of 1:2 ratio to reduce the risk of complexing site saturation (Esnaola and Millan, 1998). The tubes were then placed on a wrist action shaker for 2 h (Lindsay and Norvell, 1978). After agitation, samples were centrifuged (27 000 x g) before being filtered on a Whatman GF/F 0.7-µm borosilicate glass filter. The DTPA extraction was originally developed to assess essential metal deficiency. Nevertheless, the chelating properties of DTPA have already been used in contaminated soils to study metal uptake in plants (Maiz et al., 2000).
Copper content in all soil mixture extracts was determined by inductively coupled plasma–mass spectrometry (ICP–MS) analysis (Philipps Analytical, Anjou, QC, Canada). Furthermore, dissolved organic carbon (DOC) content in soil water extract (Extraction A) was determined by measuring the total organic carbon content in the extract by combustion according to Standard Method 5310B (Franson, 1992).
Ecotoxicological Analyses
Two organisms from two different phyla were chosen for their direct contact with the soil matrix: compost worm and barley.
Compost Worm Mortality
Compost worm survival tests were performed according to a modified version of the standard procedure of Quebec's Ministry of the Environment (Centre d'Expertise en Analyse Environnementale du Québec, 2003). Ten adult compost worms with clitellum (300–600 mg) were placed in a 500-mL polypropylene jar containing 200 g of moist soil mixture (80% field moisture capacity). Bioassays were conducted under dark conditions in a controlled-temperature chamber set to 20 ± 2°C, for 14 d, at constant moisture content. Estimation of mortality was performed on the 14th day. Samples were checked on the seventh day for dead earthworms, which were removed if present. The experiment was performed using five replicates. Unspiked soil mixtures were used as controls. Results were expressed as percentage of effect compared to the control (unspiked soil). The LC50 determination tests were also conducted on Soil Mixtures A1, C1, D1, E1, B2, and D2. The following dilutions were used to generate a dose–response slope: 100, 75, 56, 42, and 32% (B2 and D2) and 100, 75, 50, 25, and 12.5% (A1, C1, D1, and E1). Tests on diluted soil mixtures were performed in triplicate.
Barley Germination and Growth Inhibition
Barley germination and growth inhibition tests were conducted according to a modified version of the standard procedure of Quebec's Ministry of the Environment (Centre d'Expertise en Analyse Environnementale du Québec, 2000). Samples (15 g wet soil, 80% field moisture capacity) of sieved (<2 mm) soil were transferred into 25-mL sealed jars in a controlled-temperature chamber set to 20 ± 2°C. Seeds of barley (Coopérative régionale de Rivière-du-Loup, QC, Canada) were sown for germination under a photoperiod (light–dark cycle) of 16 and 8 h at a light intensity of 4300 ± 10% lux for 7 d. The experiments were conducted on five replicates using five jars per replicate and one seed per assay recipient. On Day 7, the root and shoot lengths were measured. Corresponding unspiked soil mixtures were used as controls. Results were expressed as percentage of effect compared to the control (unspiked soil). As for the earthworm bioassay, EC50 experiments were conducted with contaminated soil mixtures diluted with the corresponding unspiked soil mixture.
Aging Procedure
To simulate changing humidity conditions over time, soil mixtures at pH 7.5 were subjected to wetting and drying cycles (Han et al., 2001). Cycles began with 2 d of wet conditions (80% field moisture capacity) followed by 12 d of air-drying (20 ± 2°C). These cycles were repeated during 16 wk (8 cycles). Chemical and toxicological analysis performed at the beginning of the experiment (t = 0) were repeated at the end (t = 16 wk). During the aging period, at t = 2, 4, and 8 wk, each contaminated soil mixture was tested for pH, organic content, DOC, total Cu, and Cu fractions.
Statistical Analyses
Considering the statistical design developed for the constitution of soil mixtures, variables affecting toxic responses can be structured according to Eq. [1]. In fact, toxic responses of organisms (Y) can be explained by single effects of each soil component and total Cu content (W), as well as interaction of pH (Z) with organic matter % (X1) and clay % (X2):
 | [1] |
Mathematical transformations (log-probits) were used to provide better regressions of toxicity against soil properties. This kind of transformation is often used in ecotoxicology when studying dose–response relationships (Bliss, 1957). No toxic effect and 100% toxic effect measurements were rejected for the statistical analysis, since they could not explain the effect magnitude.
Analysis of the evolution of the parameters over time was performed using Student test (bioassays at t0 and tf) and ANOVA (physicochemical testing at t0, t1, t2, t3, and tf). Significance of parameters was determined at p < 0.05. All statistical data treatment was performed with STATISTICA 5.1 software (StatSoft, 1997).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
 | [2] |
Findings are in accordance with previous studies. Indeed, 300 Cu Kd values from published results varying by four orders of magnitude were linked to soil pH, organic matter, and total Cu (r2 = 0.61) (Sauvé et al., 2000). Another publication estimated a 70% Kd variability accountable to pH and organic matter on 41 soils with a wide range of properties (Impellitteri et al., 2003).
Equation [2] also shows a negative contribution of the interaction between pH and organic matter. In fact, while in slightly acidic soils (pH 5.5), increasing both organic matter (OM) and clay content significantly enhanced the adsorption of Cu (the greater contribution coming from OM), no significant differences were found between water-extracted Cu fractions in soils at pH 7.5 (p > 0.05). This may be explained by the enhancement of OM solubility with pH as well as by the affinity of Cu for DOC. The negative contribution of DOC to Kd (positive contribution to mobility) has already been addressed in previous studies (Janssen et al., 1997; Yin et al., 2002). Dissolved organic C in the present study is explained by pH and organic matter content (r2 = 0.86).
At pH 7.5, the CaCl2–extracted Cu fraction varied in the same way as the water-extracted fraction. However, at lower pH (5.5), increase in clay content contributed more than organic matter enhancement to the decrease in the CaCl2–extracted Cu fraction. This is probably because CaCl2 solution extracts Cu fixed on the exchangeable electrostatic sites of organic matter while water does not. Overall, the water-extracted Cu fraction explained 80% of CaCl2 (0.5 M)–extracted Cu variability, which represented the soluble and exchangeable fraction.
The DTPA-extracted Cu fraction was influenced by pH and total Cu content. Considering the similar DTPA-extracted Cu concentrations at pH 7.5, it is possible that DTPA dissolved metal precipitates formed at this higher pH (Schalscha et al., 1982). At pH 5.5, total Cu content and organic matter contributed to the difference between DTPA and CaCl2–extracted Cu fractions which is believed to express the Cu complexed with organic matter (O'Connor, 1988).
Results showed that the DTPA-extracted fraction was positively correlated to total Cu. The DTPA-extracted Cu and total Cu concentrations were linked to the Cu fixed on the solid phase because of their negative correlation to the water-extracted fraction (r = –0.873, p < 0.001). On the other hand, water-extracted and CaCl2–extracted fractions were negatively correlated to the total Cu content probably because part of the readily soluble Cu percolated during the contamination procedure.
Potentially Toxic Copper Fraction
Inhibition of barley growth and earthworm mortality in contaminated soil mixtures were compared to their respective unspiked soil mixture (Fig. 1). Soil B1, showing unexpected inhibition in both contaminated and unspiked soils, was not deemed relevant for further analysis. An unfavorable texture combined with a lower pH could be a possible explanation for the inhibition observed.
|
|
The best predictor of Cu toxicity was found to be the CaCl2–extracted Cu fraction which could explain between 30 and 44% of toxicity variability. However, CaCl2–extracted Cu could not explain toxicity in soil mixtures at pH 5.5 nor at pH 7.5 when chemical and toxicological data for pH 5.5 and 7.5 were considered separately.
Influence of Soil Properties on the Ecotoxic Response
Range finding test results assessed on Soil Mixtures A1, C1, D1, E1, B2, and D2 are presented in Table 4 as EC50 values. Results for a given species and endpoint varied up to an order of magnitude between soils, which demonstrates the important role of soil properties in toxicity testing.
|
–[5]). Results showed that the increases in pH and OM content were the most significant variables contributing to a decrease in ecotoxicity (r2 = 0.4–0.6 [barley shoot and earthworm], r2 = 0.9 [barley root]; p < 0.005). Clay content also significantly contributed to lowering the ecotoxicity of Cu expressed on barley growth and worm survival (p < 0.02):
 | [3] |
 | [4] |
 | [5] |
indicates effect on probit value; 
indicates Cu concentration, expressed in mg kg–1; 
indicates concentrations are expressed in % (w/w); and NS indicates nonsignificant (p > 0.05). Regressions for barley shoot elongation showed a r2 = 0.77 while root elongation showed a r2 = 0.92 (both p = 1 x 10–6). Soil pH, OM, and clay could explain 50% of the variability in earthworm mortality (p = 0.000059).
Spiked soils in the present study showed a lower mortality of compost worm compared to reported tests in previous publications (Neuhaser et al., 1985; Edwards and Bohlen, 1992; Spurgeon et al., 1994; Aquaterra Environmental and ESG International, 2000). In fact, LC50 values found in literature are generally two to three times lower than the predicted LC50 based on developed regression. For example, a LC50 of 836 mg Cu kg–1 estimated in Spurgeon et al. (1994) for a soil with a pH = 6.1 and constituted of 20% clay and 10% organic matter is significantly smaller than the predicted LC50 at 1854 mg Cu kg–1.
Copper bioavailability in the present study was probably lowered by the spiking procedure. Smit and Van Gestel (1998) studied the effect of zinc on the reproduction of F. candida and observed a difference by a factor of two between EC50 in a freshly spiked soil and a percolated soil while this difference rose to a factor of five between freshly spiked and field-aged soil. In a field soil, no effect was observed on L. terrestris, a more sensitive species than compost worm, up to a Cu concentration of 3000 mg kg–1 (Kennette et al., 2002).
For barley, root and shoot growth EC50 measured in this study are of the same order of magnitude than other reported tests on various soil types (Aquaterra Environmental and ESG International, 2000). For instance, reported EC50 for a 7-d exposure in a natural spiked soil containing 14.9% clay, 9% OM, and a pH of 6.1 reached 1003 and 4593 mg kg–1 for root and shoot growth, respectively.
Ali et al. (2004) measured lower EC50 values for barley root and shoot growth inhibition after 14 d in a freshly spiked artificial sand (100% sand, 0.27% OM, pH = 7.8). The root growth EC50 measured was 13.7 mg kg–1, while shoot growth EC50 exceeded the maximum concentration tested (EC50 > 305 mg kg–1).
Differences between the soils used in this study and reported toxicity tests in literature might be related to factors influencing adsorption of copper. In fact, contamination procedure is the main parameter, but also type of organic and clay material could influence retention on the solid phase. The clay material used in this study was illite which has a greater adsorption capacity than kaolinite generally used in artificial soils. The use of 25% of silt could have also enhanced adsorption. In addition, use of Ca(OH)2 to raise pH might have influenced copper partitioning. Nevertheless, spiked soils used in this study showed copper bioavailability closer to field soils than standard artificial spiked soils.
Effect of Aging on Soil Chemistry and Toxicity
The effect of aging on Cu partitioning in soil mixtures (pH 7.5) showed no significant difference in the water-extracted Cu concentrations between 0 and 16 wk (Table 2). However, a significant decrease in CaCl2–extracted fractions in organic Soil Mixtures C2 and D2 was observed. A similar decrease over time was reported for this fraction in different soil types in a previous study (Han et al., 2001). Table 2 shows a decrease in the DTPA-extracted fraction in A2 along with a slight pH decrease over time in this soil. In contrast, an increase in DTPA-extracted Cu concentration was observed in Soil D2 between Weeks 8 and 16. The unexpectedly high value at t = 16 wk might be related to an increase over time in the Cu fraction complexed to solid organic matter.
The effect of aging on the toxic response showed no significant trend. Aging at 50% of field capacity for 12 wk also showed small effect on Cu toxicity to black bindweed (Fallopia convolvulus) (Pedersen et al., 2000). However, an overall observation of the results in the present study indicated that organic soil mixtures (C2, D2) tend to become less toxic over time while non-organic soil mixtures (A2, B2) behave the opposite way; the two trends were significantly different (p = 0.002). It should be noted that the spiking procedure used in this study represented, in itself, a type of aging and that percolating the soil with water before toxicity testing reduced the difference between the aged and laboratory-treated soils (Smit and Van Gestel, 1998).
Results for compost worm mortality showed no significant difference between freshly spiked and aged mixtures, although Soil Mixtures A and B showed a slight increase in toxicity while Soil Mixtures C and E showed the opposite (Fig. 2). Barley growth showed a significant increase in root elongation inhibition for Mixtures A2 (15%) and E2 (15%) and a significant decrease in root and shoot elongation inhibition for Mixture D2 (5%; 20%).
|
Organic material properties appear to be a sensitive parameter varying with time. Measurements of DOC over time presented in Fig. 3 showed a significant increase after a first wetting and drying cycle, followed by a decrease over time (p < 0.05). The first observation is consistent with evidence that the solubility of organic matter increases when soils are remoistened following air-drying (Bartlett and James, 1980), while the decrease in DOC could be explained by the mineralization of organic matter (Sanders, 1982).
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The influence of time on ecotoxicity associated with Cu-contaminated soils at pH 7.5 showed organic soils became less toxic while other soils increased in toxicity (nonsignificant). The aged soils already contained low labile Cu concentrations making the assessment of Cu partitioning and toxicity variability difficult. Further investigation on a wider range of soil pH and longer aging period could provide more information on the significance of the trends observed for organic and non-organic soils.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS 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 | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
