| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agronomy, Throckmorton Plant Sciences Center, Kansas State Univ., Manhattan, KS 66506
b USDA-ARS, Pasture Systems and Watershed Management Research Unit, Curtin Road, University Park, PA 16802-3702
* Corresponding author (gmp{at}ksu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AEM, anion exchange membrane • epm, excursions per minute • LR, loose resin • LRBic, LR saturated with HCO3– • LRCl, LR saturated with Cl– • RS, resin strips • RSBic, RS saturated with HCO3– • RSCl, RS saturated with Cl–
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
There are several methods for estimating plant-available P in soils called ion-sink tests, which employ a P adsorbing surface that is considered to be an ion sink. This ion sink adsorbs and collects P from soil solution. One of these methods uses anion-exchange resin as the ion-sink to adsorb soil P (Buehler et al., 2002; Uusitalo et al., 2001). The second uses FeO-coated paper similarly to adsorb soil P (Chardon et al., 1996; Uusitalo and Yli-Halla, 1999). Both tests may be used as batch methods for shaking a soil solution with the appropriate ion sink.
These P testing methods have an advantage over conventional chemical extractants such as Bray-1 (Bray and Kurtz, 1945), Olsen (Watanabe and Olsen, 1965), and Mehlich-3 (Mehlich, 1984) because the ion-sink methods function similarly to a plant-root surface adsorbing available P ions from the in situ labile P pool in the soil (Menon et al., 1989). In contrast, the reactions in chemical tests for soil P may solubilize non-labile P more tightly bound to Al, Fe, or Ca complexes, which may not be plant available (Fixen and Grove, 1990; Mallarino, 1997). When this occurs, accurate interpretation of test results becomes more difficult.
Ion-sink methods have been favorably employed to estimate plant-available P for soils with large variations in physical and chemical properties (Menon et al., 1990; Sharpley et al., 1994). In contrast chemical tests for soil P are not always equally reliable over all soil types. For example, the Olsen P test was designed to extract P from calcareous soils, whereas Bray-1 and Mehlich-3 were designed for non-calcareous soil dominated by Fe- and Al-P complexes (Bray and Kurtz, 1945; Mehlich, 1984; Watanabe and Olsen, 1965). Using any of the chemical extractants beyond the range of soils for which it was developed can result in the buffering of acid or base extractants and dissolution of non-available P.
The resin method developed from the concept that a strong-base type anion-exchange resin shaken with a soil-water mixture would adsorb soil P (Amer et al., 1955). In this early study, soil was finely ground to accommodate the designed method of soil-resin separation. Problems relating to soil-resin separation and soil grinding led to the use of nylon netting bags to hold the resin to more easily retrieve, rinse, and extract P from the resin (Saggar et al., 1990; Sibbesen, 1977). To overcome these procedural difficulties, a new method using FeO-coated filter papers, shaken with a soil solution was developed (Sissingh, 1983). The FeO-coating functions as a strong P sink adsorbing P from solution onto the surface of the FeO-coating. The coating, with the adhering P is then dissolved and the P subsequently analyzed. This method rapidly increased in popularity as a soil-P test for estimating plant-available P (Chardon et al., 1996).
The FeO method is not without its problems, as the FeO-coated papers are not commercially available. This has led to various methods for their preparation and use (Kuo and Jellum, 1994; Lin et al., 1991; Myers et al., 1997). Another concern is contamination of the FeO-coated paper by fine soil particles during the shaking period (Chardon et al., 1996; Myers et al., 1995; Perrott and Wise, 1993). For instance, Uusitalo and Yli-Halla (1999) reported that soil contamination of the FeO-papers with sediment particles can overestimate desorbable P in turbid runoff water.
Anion-exchange membranes (AEM) were developed to simplify problems associated with ion-sink extraction of soil P. Saunders (1964) found that AEM could be used instead of LR beads with the advantage that soil did not have to be finely ground or placed in resin bags to facilitate separation. More recent refinements and modifications for the use of AEM as an ion-sink to extract soil P have led to increased usage (Saggar et al., 1990; Sibbesen and Rubaek, 1994).
Another desirable feature of the AEM method, at least from an economic viewpoint, is the durability of AEM. Reports indicate AEM have been re-used as many as 50 to 500 times without losing their extraction efficiency or showing detrimental structural effects (Saggar et al., 1990; Schoenau and Huang, 1991). If this is true it would be a distinct advantage over FeO-coated paper, which can be used only once.
Even so, a number of procedural features in the use of AEM for extraction of soil P have not been clearly defined nor standardized among laboratories. Strip size and total surface area of AEM as an ion sink have varied. Some examples of these parameters include: (i) the use of two 9 x 62 mm strips to extract 0.5 g soil, (ii) the use of one 25 x 62.5 mm strip to extract 1.0 g soil, and (iii) the use of one 20 x 60 mm strip to extract 0.5 g soil (Indiati and Singh, 2001; Sibbesen and Rubaek, 1994; Tiessen and Moir, 1993). Such disparities in exchange capacity of AEM per given amount of soil can result in different P tests for the same soil. Also, several different saturating counter ions have been used, with Cl– or HCO3– the most frequent choice (Abrams and Jarrell, 1992; Beck and Sanchez, 1994; Rubaek and Sibbesen, 1993; Uusitalo et al., 2001). The type of resin used can also lead to differences in extractable P (Rubaek and Sibbesen, 1993).
In evaluating extensive data for different P tests on plant P uptake, Sibbesen (1983) found that anion exchange resin performed slightly above sodium bicarbonate methods, and all acid methods performed worst. However, neither AEM nor FeO-coated paper were included in the evaluated tests, and in some other tests the FeO-method has outperformed all other types of P tests, including resin, in predicting P uptake from soil (Menon et al., 1989).
This study was designed to standardize an AEM extracting procedure for soil P that will give similar P tests as the traditional LR method, but will be easier and less time-consuming. We also compared the resin-P and FeO-P methods and then compared these ion-sink methods to routine chemical methods.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Soil-resin separation was accomplished by density gradients so it was unnecessary to either finely grind soil for separation or enclose the resin in nylon bags, thus avoiding the procedural hazards of both methods (Thien and Myers, 1991).
Soil Phosphorus Extraction
Dowex 1 x 8-50 basic anion-exchange resin in the Cl– form (Sigma Chemical Co., St. Louis, MO) grading 0.8 to 0.3 mm (20–50 mesh) was used. Particles <0.5 mm were separated and discarded by sieving. The resin was soaked in deionized water for 24 h before preparing it for use. To prepare the resin as an ion-sink, it was saturated with either Cl– or HCO3– by adding 250 mL of either 1.0 M KCl or 0.5 M NaHCO3 to 50 g of resin for 30 min. The procedure was repeated once. The adsorption capacity of LR was 1.58 µmolc g–1.
To extract soil P with LR, 1.0 g of soil, 1.5 g of moist resin, and 80 mL of deionized water were added to a 125-mL polyethylene wide-mouthed bottle with a screw-type closure. Bottles were laid horizontally end-to-end on a variable speed-reciprocating shaker and shaken at a speed of 130 excursions per minute (epm) for 16 h.
Except for a couple modifications, the density gradient method described by Thien and Myers (1991) was used to separate LR from the soil mixture. The modifications consisted of omitting the centrifuge steps that were unnecessary and the acetone wash, which was inappropriate in the current application.
After shaking and soil/resin separation, P was eluted from the resin beads by adding 35 mL of 0.5 M HCl to each 125-mL polyethylene bottle, which was capped and shaken end-to-end at about 125 epm on a reciprocating shaker for 30 min. The solution was decanted into a 100-mL volumetric flask and 35 mL of 0.5 M HCl was added to the resin a second time, and shaken 15 min. The solution was decanted into the same 100-mL flask, and then 25 mL of 0.5 M HCl was added to the resin in the bottle that was again shaken for 15 min. This solution was also decanted into the 100-mL flask that was made to volume with deionized water and mixed thoroughly. An aliquot of the solution was removed and the acidity neutralized with a NaOH solution. The aliquot was analyzed for inorganic P (Murphy and Riley, 1962). The adsorption capacity of the resin used in soil extractions was 1252 µmol g dry resin–1.
Anion exchange resin membranes were used for comparison with the above LR method. This membrane product, no. 551642S, currently retains the BDH trade name, but is presently produced by VWR International Ltd., Merck House, Poole, Dorset UK. The current USA distributor of the product is Gallard-Schlesinger Industries Inc., Garden City, NY. The primary structure of the membranes consists of polystyrene cross-linked with divinylbenzene. These anion membranes are made ‘anion permeable’ by means of quaternary NH4 groups attached to the membranes. Membranes are 0.11- to 0.15-mm thick and come in the chloride form. The manufacturer states that the membranes are stable and have a shelf life of at least 2 yr.
Membrane sheets described above are 12.5 x 12.5 cm and packed moist for shipping. They were soaked in deionized water for 24 h before use. The membranes were cut into 2.08 x 4.15 cm RS so one membrane sheet yields 18 RS. One 2.08 x 4.15 cm RS has an overall surface area of 17.3 cm2, two RS has a surface area of 34.5 cm 2, and three RS a surface area of 51.8 cm2. In the current study, RS surface area was varied by using one, two, or three RS with each 1-g soil sample performed in triplicate.
Resin strips were saturated with either Cl– or HCO3– as described for LR above. The RS were stored in deionized water in a closed container until time for use.
The P-extraction procedure was the same using RS as described for the LR, except that one, two, or three RS were used in place of 1.5 g of LR. Also, the shaking period for RS was extended to 24 h to obtain P extraction more nearly identical to values obtained by LR, the standard resin method being used in this study.
At the end of the shaking period, RS were retrieved from the shaking solution and rinsed under a stream of deionized water for a few seconds. The RS from each extraction bottle (either one, two, or three strips) were placed in a 125-mL wide-mouthed bottle and 50 mL of 0.5 M HCl was added to elute P from the strips. The bottles were sealed, laid horizontally and end-to-end on a reciprocating shaker, and shaken 125 to 130 epm for 90 min (Myers et al., 1999). An aliquot of this solution was analyzed for inorganic P (Murphy and Riley, 1962). The adsorption capacity of the AEM was 1.52 µmol cm–2.
We followed the procedures described by Myers et al. (1997)(1999) to extract FeO-P from soil. We accordingly used 5.5-cm circles of FeO-coated paper as ion sinks with 1.0 g soil and 80 mL 0.01 M CaCl2. The traditional 16-h shaking period for extraction of soil P was used, but for comparison values, we also included a 24-h shaking period because of using that longer period for P extraction with RS.
At the end of the extracting period, pH of the extracting solutions was determined with a pH meter with a glass electrode.
To test the extracting efficiency of all the ion sinks including FeO-coated paper, LR, and three RS, standard P solutions (as KH2PO4 in deionized water) were made. Eighty milliliters of each solution was extracted. Eighty milliliters of each formulated solution contained the following amounts of total inorganic P: 2.45, 4.90, 7.35, 9.80, 24.50, 49.00, 98.00, 196.00, and 385 µg. Elution and analysis of P were performed for each ion sink as described above. The traditional 16-h shaking period for LR was used in these tests. A 24-h shaking period was used for all RS and FeO tests.
Statistical Analysis
Analysis of variance and regression analysis were performed using the Statistical Analysis System (SAS Institute, 1982). When ratios of areas under given curves were determined, the no-intercept option of regression analysis for those curves was used.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Shaking Times
Although shaking times for resin extraction have traditionally been around 16 h, in this study we extended the shaking time to 24 h for RS to increase extractable P to levels obtained by the traditional method using LR. The customary shaking time for FeO-coated paper in the extraction of soil P is 16 h, but since we increased the shaking time to 24 h for the RS, we also included a 24 h shaking time for FeO-paper to compare values on an equal shaking time basis. The rate of increase in P extraction with increase in shaking time was nearly the same for both 3 RSBic and FeO with slopes of 1.14 and 1.16, respectively (Fig. 1a and b).
|
|
The data also show for all soils with LR P tests of 10 mg kg–1 or more that 3 RSBic P tests average over 100% of the LRBic P tests for those soils. This favors use of the 3 RSBic method instead of the more time-consuming LR method for soils with P tests of 10 mg kg–1 or more. However, compared with the LR method, the 3 RSBic performed poorly when LR extractable P was 7.0 mg kg–1 (Table 2). This may be due to a more effective recovery of P from soil by LR than RS as a result of the greater adsorption capacity of LR than RS and more intimate mixing of soil with LR than RS.
Extracted soil P was less for most soils for one RSCl than for LRCl (Table 3). The mean P value for LRCl was 63.2 mg kg–1 compared with 48.7 mg kg–1 for one RSCl, and the corresponding percentage mean over all soils showed about a 29% difference. Extraction efficiency improved with two RSCl for most soils, however, extracted P was usually below that for LRCl (Table 3). The mean P extracted over the 24 soils of 54.8 mg kg–1 for two RSCl was less than the 63.2 mg kg–1 for LRCl, and a corresponding percentage mean difference of about 20% between amount of P extracted by LRCl and two RSCl. Some soil P tests for three RSCl compared well with those for LRCl, but considerable variation often occurred over the 24 soils tested (Table 3). The mean P extracted by three RSCl of 57.3 mg kg–1 compared with 63.2 mg kg–1 by LRCl. The corresponding percentage mean over all soils was 81.5%, which was only about 1% increase over the 80.1% with two RS noted above.
|
|
The RS procedure using Cl– as the saturating counter ion is less satisfactory than when HCO3– is the saturating counter ion, such that replicate variability was greater for RSCl–P than LRCl–P. The regression of one RSCl–P on LRCl–P is: Y = 0.76X with R2 = 0.98. With two RSCl, the corresponding regression equation is: Y = 0.87X with R2 = 0.99; and with three RSCl, the corresponding regression equation is: Y = 0.91X with R2 = 0.99. Although these comparisons indicate that P extraction with HCO3––saturated RS extracts more soil P than extraction with Cl––saturated RS, they don't prove that the increased amounts of extracted P are available to plants.
In previous studies, Sibbesen (1978) noted that resins in the HCO3– form stabilize the system, so that the amount of P extracted and the suspension pH are almost independent of the type of resin and soil/water ratio used. However, he also found in 34 soils with known available P levels that soil P extraction with resin in the HCO3– form was only slightly more closely correlated with known P levels than when resin was in the Cl– form. This agrees with our current data showing that RSBic extracted P more efficiently from most soils than did RSCl.
Another positive feature of HCO3––saturated resin is its similarity to natural, chemical conditions that prevail in the rhizosphere of actively growing plant roots in soil. Sibbesen (1978) notes that investigations have shown plant roots accumulate bicarbonate in the rhizosphere resulting in an increase in rhizosphere pH in acid to neutral soils and a decrease in rhizosphere-pH in calcareous soils, and felt this concept indirectly supported findings that the amount of P extracted by HCO3––resin was better correlated with plant P uptake, than was the amount of P extracted with Cl– resin. One reason HCO3––saturated resins perform better than Cl–-saturated resins is that the HCO3– released from the resin equilibrates with carbon dioxide–carbonic acid–bicarbonate–carbonate equilibrium in the soil system (Sibbesen, 1978). This results in low concentrations of HCO3– in the soil solution, which favors release of mobile P ions from soil surfaces. In contrast, Cl– accumulates in solution, thereby inhibiting the ion exchange processes. In the HCO3––form, efficiency of extraction of soil P with resins has shown little variation between the different anion-exchange resins, and in addition, extracted P has been greater than with resins in chloride form (Rubaek and Sibbesen, 1993). Our results agreed with their observations.
pH of Shaking Solutions
At the end of the shaking period, pH of the extracting matrix was tested to determine whether the ion-sink extracting agents had affected solution pH. Data showed a significant buffering effect on the system pH when HCO3– was the saturating counter ion. The effect was similar for LRBic and three RSBic (Fig. 3a). The regression of solution pH on soil pH is described by the equation: Y = 0.53X + 3.48 with R2 = 0.91 when three RSBic were used and Y = 0.50X + 3.33 with R2 = 0.88 when LRBic was used. This buffering effect of the HCO3– concurs with observations made by Sibbesen (1978) that resins in the HCO3––form buffer the system.
|
Resin membranes in the Cl–-form have sometimes been used instead of the HCO3– form, to minimize pH changes near the soil-resin membrane interface (Abrams and Jarrell, 1992). Our results indicated that the HCO3––saturated resin minimized pH change in the extracting solutions (Fig. 3a), however, at the soil–solution interface the superior buffering effect of the HCO3––saturated resin would result in a greater disparity between the ambient soil pH and the ambient solution pH than with Cl– saturated resin with its accompanying negligible buffering effect as Abrams and Jarrell (1992) have noted.
When soil P was extracted with FeO-coated paper, the buffering effect was intermediate of Cl––resin and HCO3––resin (Fig. 3c). With FeO-coated paper, the equation for the regression of the shaking solution pH on soil pH was Y = 0.71X + 1.73 with R2 = 0.93.
Ion Sinks: Resin vs. FeO-Coated Paper
Anion-exchange resin and FeO-coated paper have traditionally been considered similar methods for extracting and estimating plant available P in soils. However, in this present study, LR and RS both extracted greater amounts of soil P than did FeO (Fig. 2a, 2b, and 4). The regression equation for soil P extracted by three RSBic on soil P extracted by LRBic was: Y = 0.99X with R2 = 0.99; whereas, the regression equation for FeO-P on soil P extracted by LRBic was: Y = 0.47X with R2 = 0.84 for the 16-h shaking period, the customary length of shaking time in tests for FeO-P (Fig. 4). Some improvement in the relative extraction efficiency of FeO occurred when the shaking period was increased to 24 h, then: Y = 0.54X with R2 = 0.84. However, the difference between the three RSBic–P slope (0.99; Fig. 2a) and the FeO-P slope (0.54; Fig. 4) still remained relatively large. The RS-P slope in Fig. 2a was different than either of the FeO slopes in Fig. 4. The reason for the disparity between these tests is unclear. The total area under the FeO-P (24 h) curve is 54% of the total area under the RS-P curve, and for the FeO-P (16 h) 47%, (Fig. 2a and 4).
|
However, the results did lead us to test the extracting efficiency of all ion sinks used in the present study with standard P solutions of inorganic P, and in that way to rate their extracting efficiencies. The tests were made in aqueous solutions.
The efficiency and accuracy for all ion sinks tested including LR, three RS, and FeO-coated paper with standard inorganic P solutions showed similar recovery for all ion-sink methods tested except for RSCl with a slope of only 0.84 compared with 0.95 to 0.99 for all the other methods including FeO-coated paper (Fig. 5). This showed that the soil P extraction efficiency of FeO-coated paper is not inferior to the resin ion sinks over the range of soil P values tested.
|
Although the present data do not explain the quantitative differences between FeO-P and resin-P in this study, it is well known that CaCl2 solutions have a stabilizing effect on soil flocculation, and that in fact is the reason it is used in the FeO method to minimize soil dispersion, which increases soil particle contamination of the FeO-coated filter paper. However, as soil disperses to a greater extent in the resin extractions with its shaking matrix of water, then the magnitude of soil/resin contact may be enhanced resulting in increased extraction of soil P due to the increased soil–resin contact compared with the more stable FeO–CaCl2 systems. This is one possible reason for higher P tests for resin-P than for FeO-P in the current study, however, we made no tests to verify that theory.
Previous reports have shown decreased extraction of soil P when the extracting matrix is 0.01 M CaCl2 compared with an aqueous matrix (Koopmans et al., 2001). Their curve slope for soil P extraction in the CaCl2 solution compared with that extracted in the aqueous solution was 0.39. Our curves in the present study for FeO-P (0.01 M CaCl2 matrix) extraction compared with resin P (aqueous matrix) extraction were 0.54 and 0.47 (Fig. 4). In reviewing the literature, Koopmans et al. (2001) noted that desorption of P is known to decrease with an increase in CaCl2 concentration because of an exchange of H+ from the soil with Ca2+ in solution, which causes a decrease of the solution pH and an increase in net positive charge of the soil surface and that increases P retention at the soil surface. If CaCl2 always has an acidifying effect on solution pH, then in the current study the pH of extracting solutions for low pH soils extracted by the FeO (in CaCl2 solution) method should have been lower than that for the low pH soils extracted with RSCl (in aqueous solution), which had no buffering effect (Fig. 3b and c). It appears the converse is true in our soils. While the reasons for this are unclear, other mechanisms were probably also influencing soil P release.
An earlier study found AER in the HCO3– form are highly efficient in releasing P in acid soils rich in organic matter and in calcareous soils (Delgado and Torrent, 2001). These factors may have been enhancing the magnitude of P tests from soils with high pH and also those with highest levels of organic C. In any case the higher P tests for resin than FeO methods do not indicate resin methods are superior to FeO methods, but do indicate that when such tests are used to make fertilizer calibrations and recommendations, that it is critically important to establish plant-available P indices separately for the particular and specific P-testing method being employed because of the large differences which occur in some soils between the different P tests.
Comparison of Ion-Sink Methods with Chemical Methods
Correlation of Olsen P with 3 RSBic-P was high with R2 = 0.96 (Fig. 6a) and about the same for Olsen P with 3 RSCl–P with R2 = 0.95 (Fig. 6b). The correlation for Olsen P with FeO-P was also good with R2 = 0.90 and nearly a 1:1 slope of 1.07 (Fig. 6c). With the slope of 1.07, it appears that plant-available P indices might be used interchangeably for Olsen P and FeO-P, but pot studies with controlled conditions would need to be used to verify that. In contrast, the slopes for Olsen P on 3 RSBic and 3 RSCl are only 0.61 and 0.66, respectively, indicating major differences in P-test results (Fig. 6a and b).
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The current study showed that P tests from the Olsen, FeO, RSBic, and RSCl methods were all well correlated. The P tests for Bray-1 and Mehlich-3 methods were less well correlated with the ion sink methods.
Amounts of P extracted by FeO-coated paper only averaged about 50% of that extracted by resin methods. This difference is surprising and cannot be explained by our experimental design. However, the finding does not indicate that resin is superior to FeO methods in calibrating P fertilizer recommendations. The disparity between the resin- and FeO-P indicates that separate plant-available-P indices need to be determined for each method.
Pot or field studies need to compare all ion-sink and chemical extraction methods to with different soils and crops to compile fertilizer P recommendations for P. Chardon et al. (1996) standardized guidelines for the methodology of the FeO-method. Standardized procedures for the chemical P methods are established and followed. A standardized procedure for resin-P procedure similar to that outlined above needs to be adopted and followed.
By using standardized methods for these different P tests for pot studies under controlled conditions, the results should help determine whether one of these methods is more effective to calibrate fertilizer P recommendations. Conversely, if correlations for the different tests for plant-available P uptake were all the same, then that would indicate the methods might be used interchangeably. In that case, the preference and choice to use the easiest, quickest, most economical, and most convenient method would probably be fully justified. However, regardless of correlation considerations, when a large disparity between P tests from different methods occurs, the plant-available P indices need to be determined separately.
Resin methods need to be standardized between different labs all over the world and it is hoped that this current study will provide a modicum of progress in that direction.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication March 30, 2004.
|
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 | |||
