| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Biological Sciences, Marshall Univ., Huntington, WV 25755-2510
b USDA-Forest Service, Timber and Watershed Research Lab., Parsons, WV 26287
* Corresponding author (gilliam{at}marshall.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Abbreviations: FEF, Fernow Experimental Forest • WAS, Watershed Acidification Study
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Several factors potentially interact to influence spatial heterogeneity of N in forest soils. Morris and Boerner (1998) found considerable variability at the landscape scale in patterns of nitrification and microbial biomass in eastern hardwood forest soils, relating these patterns to spatial patterns of soil moisture. Some investigations have found spatial variability of N processing in forest soils to be related to leaf litter chemistry of contrasting tree species (Ferrari, 1999). Other studies suggest that spatial heterogeneity is largely determined by the direct effects of plants on soil processes (e.g., Chen and Stark, 2000), effects that are often mitigated by other spatially variable factors, such as fire (Hirobe et al., 2003b). Venterea et al. (2003) demonstrated that forest landscape-scale variation in net nitrification is best explained by a suite of vegetation, physiographic, and soil factors.
The Watershed Acidification Study (WAS) was initiated by the USDA Forest Service in 1988 at the FEF, Parsons, WV, to examine ecosystem responses to aerial applications of (NH4)2SO4 to an entire watershed (WS3). Results from this project suggest that untreated watersheds of FEF already have become N saturated from high ambient levels of atmospheric deposition of N (Gilliam et al., 1996; Peterjohn et al., 1996; Adams et al., 1997). Despite this, several studies (e.g., Peterjohn et al., 1999; Gilliam et al., 2001a, 2001b; Christ et al., 2002) found that untreated watersheds exhibited a relatively high degree of spatial heterogeneity in soil water NO3 and net N mineralization and nitrification. Gilliam et al. (2001b) further concluded that one of the more pronounced effects of experimental additions of N to the treatment watersheds was to decrease spatial heterogeneity of soil N processing.
Spatial patterns of net nitrification and NO3 in soil water of one untreated watershed in the WAS at FEF (WS4) suggest that, although WS4 shows general signs of N saturation, microenvironmental variability (e.g., soil factors) may greatly limit rates of N processing in some areas (Gilliam et al., 2001a). Some data suggest that these rates may be maintained at low levels by presence of ericaceous species, such as Vaccinium spp., in the herbaceous layer that may inhibit activity of microbial populations responsible for the conversion of organic forms of N into inorganic forms of N that are potentially available for uptake by plants (Bradley et al., 2000; Gilliam et al., 2001b; Read and Perez-Moreno, 2003). Although a valid alternative hypothesis for the lack of net nitrification in soils within this otherwise N-saturated watershed would be the absence of nitrifier populations in such soils, this was rejected by Gilliam et al. (2001a) using microbial gene amplification techniques; that is, they demonstrated evidence for the presence of nitrifying bacteria in nonnitrifying soils.
A serious consequence of N saturation is enhanced leaching of Ca from the mineral soil, usually seen as increases in stream concentrations of NO3 and Ca over time. Data gathered by the USDA Forest Service demonstrate that increases in leaching of NO3 over time have increased loss of Ca from soils at FEF (Adams et al., 1997). This relationship between NO3 and Ca was highly significant during a >30-yr period in an untreated watershed (WS4) and during the >10 yr since time of initial N treatment in an N-treated watershed (WS3), consistent with responses to N additions in soil solution chemistry at FEF (Edwards et al., 2002) and with studies of stream chemistry responses to N additions at other eastern forest sites (Fernandez et al., 2003; Jefts et al., 2004).
The purpose of this study was to consider the causes and consequences of spatial heterogeneity of N processing at FEF by using soils from four sites within FEF that represented a gradient in net N mineralization and nitrification measured in situ during a 3-yr period. We also examined the relationship between extractable cations and rates of net N mineralization and nitrification along this gradient. We addressed the following four questions: (i) How do rates of net N mineralization and nitrification found in the field compare to rates found under controlled conditions at varying temperatures? (ii) What is the effect of soil NO3 production on extractable Ca and how does this effect vary along the N gradient? (iii) What soil chemical variables best explain the observed N gradient? (iv) What is the effect of varying substrate quality on laboratory rates of net N mineralization and nitrification in soils of the extremes (highest vs. lowest) of this gradient?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
1900 ha of the Allegheny Mountain section of the unglaciated Allegheny Plateau in Tucker County, West Virginia (39°03' N, 79°49' W). Averaging approximately 1430 mm yr–1, precipitation at FEF varies seasonally and with elevation, generally greater during the growing season and at higher elevations (Gilliam et al., 1996).
The sample sites were located in two experimental watersheds at FEF: WS4 (100-yr-old mixed-aged stand) and WS3 (30-yr-old even-age stand); WS3 serves as the treatment watershed for the WAS, receiving three aerial applications of (NH4)2SO4 yr–1, totaling 35 kg N ha–1 yr–1 since 1989. WS4 was the untreated control watershed. Although the two watersheds vary in stand age, their soils are similar, predominantly coarse-textured Inceptisols (loamy-skeletal, mixed mesic Typic Dystrochrept) of the Berks and Calvin series, sandy loams derived from sandstone. Within each of these watersheds, sampling took place on sites designated "low N" and "high N" (see Field Sampling below).
The woody overstory is characterized by a high diversity of tree species, including several species common to all sample sites, including sugar maple (Acer saccharum Marshall), black cherry (Prunus serotina Ehrh.), and northern red oak (Quercus rubra L.). Other species, such as white ash (Fraxinus americana L.) and yellow poplar (Liriodendron tulipifera L.) are found only on N-treated sites. The herbaceous layer of these sites comprises species typical of montane eastern deciduous forests (Gilliam and Roberts, 2003). Herb layer species common to all sites include seedlings of striped maple (A. pensylvanicum L.), sugar maple, and black cherry and several species of Viola. Species with relatively high biomass are associated with single sites—southern groundcedar [Lycopodium flabeliforme (Fern.) Blanch] at the N-treated–high N site and hillside blueberry (Vaccinium vacillans Aiton) at the untreated–low N site.
Field Sampling
Sampling was done at four sites, one low-rate site and one high-rate site in each of the two watersheds, that were shown by previous investigations to represent a gradient in rates of net N mineralization and nitrification (Gilliam et al., 2001a). Determination of this gradient was based on 3-yr means of rates of net N mineralization and nitrification measured in situ and reported in Gilliam et al. (2001b). The sites were not defined by specific boundaries, thus precluding accurate estimates of area. Rather, they were essentially subcatchments within each of the two watersheds.
Mineral soil samples were taken to a 5-cm depth from three randomly located, 4- by 4-m plots within each site. Soil was taken at points randomly within each plot and combined to yield a single composite sample per plot. Surgical gloves and trowels were used for soil sampling and were washed with an antibacterial solution between plots within each site and between sites within each watershed to eliminate contamination between samples.
Laboratory Analyses
All soil samples were brought back to the laboratory at Marshall University for determination of preincubation levels of extractable ions and for incubation under controlled conditions. Soil from each plot was divided into three subsamples and placed in sterilized polyethylene bags, with each group of subsamples being incubated at three temperatures: 10, 20, and 30°C. All subsamples were extracted following incubation at 3, 7, 14, 21, and 28 d for NH4, NO3, Ca, and Al. Soil moisture was monitored for all subsamples before and during incubation, and neither varied among samples nor through time of incubation. Extraction and analysis for all ions followed methods described in Gilliam et al. (2001a). Briefly, moist soils were extracted with 1 M KCl at an extract/soil ratio of 10:1 (v/w). Extracts were analyzed colorimetrically for NH4 and NO3 with a Bran+Luebbe TrAAcs 2000 automatic analysis system. Extractable Ca and Al were analyzed with a Varian inductively coupled plasma emission spectrophotometer. Net N mineralization and nitrification rates (in µg N g–1 soil d –1) were determined via regression analysis as the slope of the line comparing extractable NH4 plus NO3 (for N mineralization) and extractable NO3 (for nitrification) vs. incubation time (in d). These relationships were linear and significant (P < 0.05) for all sites.
For the substrate quality experiment, we utilized soils from two sites only—the extremes of low-N vs. high-N processing sites among the 21 reported in Gilliam et al. (2001b). These were found on the untreated WS4. Four replicate samples were used for each treatment per site. In addition to preincubation extractions for NH4 and NO3 as described previously, 70-g subsamples of soil from plots of each site were amended with 10 mL of sucrose, glycine, or deionized H2O. Solutions of 0.01 and 0.1 M sucrose were added to the high-N soil (representing 230 and 2300 mg C added g–1 soil, respectively), whereas 0.01 and 0.1 M glycine solutions were added to the low-N soil (representing 20 and 200 mg N and 34 and 340 mg C added g–1 soil, respectively). Soil samples were incubated at 23°C for 21 d, after which soils were extracted for NH4 and NO3. Net N mineralization rates were calculated as postincubation NH4 and NO3 minus preincubation NH4 and NO3. Nitrification rates were calculated as postincubation NO3 minus preincubation NO3.
Data Analyses
Field vs. laboratory rates (at each of the three temperatures) of net N mineralization and nitrification were compared with Pearson product moment correlation analysis. The effects of temperature on net N mineralization and nitrification were assessed by determining linear and curvilinear relationships between mineralization and nitrification as separate dependent variables vs. incubation temperature as the independent variable. The goal was to find the model that produced the best fit, defined as highest r2. Potential effects of nitrification on soil Ca extractability were similarly assessed by determining the best fit of extractable Ca vs. extractable NO3 for all extraction times (0, 7, 14, 21, and 28 d). Finally, the effects of altering substrate quality on net N mineralization and nitrification were assessed by comparing means of control (H2O added) to those of 0.01 and 0.1 M sucrose and glycine additions to the respective soils with one-way ANOVA and Bonferroni multiple range tests (Zar, 1996).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
|
|
Effects of Temperature on Net Nitrogen Mineralization and Nitrification
Net N mineralization and nitrification exhibited exponential increases with incubation temperature up to 30°C for the two high-N sites, whereas increases with temperature were linear for the low-N sites (with the exception of no net nitrification at the control–low N site) (Fig. 2a, 2b). Working with N-amended soils from a California oak woodland–annual grassland site, Stark and Firestone (1996) found curvilinear increases of nitrification in response to incubation temperature up to 30°C, finding a significant decline in nitrification at temperatures > 30°C. It was determined that temperature optima for these grassland and oak woodland soils were 35.9 and 31.8°C, respectively, a relationship that was best described by a generalized Poisson density function (Stark, 1996). Comparison of responses of nitrification to temperature between soils of the California site to those of FEF soils suggests that nitrifier populations at FEF may also have temperature optima at >30°C. This is well above the highest mean daily temperature (20°C) during a single growing season reported by Gilliam and Adams (1999) for undisturbed surface soils at another FEF site close to these study watersheds. Mean soil temperature throughout that growing season was 16°C (Gilliam and Adams, 1999).
Positive relationships between temperature and microbially mediated processes, such as N mineralization and nitrification, are certainly to be expected and indeed have been described for lab-based (e.g., Emmer and Tietema, 1990; Stark and Firestone, 1996; Niklinska et al., 1999) and field-based incubation studies (e.g., Tietema and Verstraten, 1991; Gilliam et al., 2001b), and have been demonstrated across extremely broad spatial scales (Yin, 1992). What is notable about results presented here, however, is that they confirm for N-saturated soils what was concluded by Stark (1996) for soils in woodland and grassland soils—temperature optima for nitrifier populations in the soil are far greater than temperatures typically experienced by those populations. This has important implications for the response of N-saturated forests to increases in temperature in the future that might be brought about by global warming. That is, any negative effects experienced by N-saturated forests (e.g., declines in growth rates from nutrient cation deficiencies) may be exacerbated by increases in ambient temperatures due to global warming.
Potential Effects of Nitrification on Soil Calcium
Nitrate production appeared to have a significant influence on extractable Ca, with positive curvilinear relationships across time found for all sites that exhibited net nitrification (Fig. 3) . The range and magnitude of change in extractable Ca varied consistently in the order of net nitrification along the gradient, that is, lowest for the N-treated–low N site, intermediate for the N-treated–high N site, and highest for the control–high N site (Fig. 3).
|
0.3 µg Ca g–1 soil d –1) in mineral soil beneath sugar maple. In fact, net Ca mineralization for mineral soil at our control–high N was within the range of the highest rates for forest floor (13–14 µg Ca g–1 soil d –1 for sugar maple and white ash) reported by Dijkstra (2003). Net Ca mineralization at the control–high N site is 0.31 kg Ca ha–1 yr –1, which is low relative to stream flow outputs of Ca at the study watersheds (11–12 kg Ca ha–1 yr –1) (Adams et al., 1997). Such large differences between studies are quite notable and likely arise for several reasons. One possibility is related to the conditions under which the samples were incubated (variable and lower in situ temperatures vs. constant 30°C). We suggest, however, that most of the difference is related to length of incubation period. Because we found measurable increases in extractable Ca even at the 7-d incubation period, it is probable that the relatively low values reported by Dijkstra (2003) were in part an artifact of the lengthy incubation periods (4 and 7 mo for summer and winter incubations, respectively) he used for both forest floor and mineral soil materials (i.e., although his soils may have had high short-term rates of Ca mineralization, the rates—expressed on a d–1 basis—would have been calculated by dividing by the very large number of days in the incubation period).
We found a highly significant curvilinear pattern for net Ca mineralization along the gradient of net nitrification across all four study sites at FEF (Fig. 4) , suggesting a direct connection between the two processes. Furthermore, these results are consistent with the observation that increased nitrification can increase mobility of Ca in soils at FEF, and are consistent with the findings of Edwards et al. (2002) at FEF. That study demonstrated that the (NH4)2SO4 treatments on WS3 have increased movement of NO3 in soil solution from A and B horizons to accumulate in the C horizon, and have concurrently increased Ca concentrations in soil water of the C horizon beneath the rooting zone for most plants (Edwards et al., 2002).
|
We suggest that the first potential source of Ca is not important at our N gradient sites. Because soils of the study watersheds were derived from sandstone parent materials, they are very low in Ca-bearing minerals. Indeed, an extensive mineralogical characterization of several soil types at FEF (Lusk 1998) concluded that there is little, if any, ecologically significant geochemical source of Ca in FEF soils. Lusk (1998) found a virtual absence of apatite, the Ca-bearing mineral found in some northern hardwood forests (Blum et al., 2002, Hamburg et al., 2003). Thus, most of the Ca that increases with nitrification is either being released from the cation exchange sites associated with both clay and organic colloids (Gilliam et al., 1994) or being directly mineralized from organic matter. Ultimately, the mechanism of nitrification-enhanced extractability of Ca suggested in Fig. 3 and 4 is likely one that diminishes the readily available pool for soil Ca required by plants for uptake. This conclusion is consistent with findings of Gilliam et al. (1996), who found that increased NO3 production decreased uptake of Ca by roundleaf yellow violet (Viola rotundifolia Michx.) on FEF watersheds.
Effects of Altering Substrate Quality on Net Nitrogen Mineralization and Nitrification
Because of its importance in influencing microbial processing of N, the quality of litter and soil organic matter has long been of interest in the study of N cycling of forest ecosystems, whether quality is defined by lignin/N ratio (Melillo et al., 1982; Stump and Binkley, 1993) or C:N ratio (Hirobe et al., 2003a; Springob and Kirchmann, 2003). The potential effects of variation in litter quality on N mineralization and/or nitrification are often assessed on the landscape scale by correlation among sites with contrasting litter and soil organic matter quality (Scott and Binkley, 1997; Venterea et al., 2003). Experimentally, however, this is usually assessed with manipulation of C:N ratios by adding C-rich sources, such as sugars (e.g., Davidsson and Ståhl, 2000; Magill and Aber, 2000) and sawdust (e.g., Gulledge and Schimel, 2000), or N-rich sources, such as amino acids (Jones and Shannon, 1999; Garten, 2004).
In this part of our investigation of the gradient in soil N processing, we used a C-rich source (sucrose) to determine if adding C without N (increasing soil C:N) would stimulate net immobilization in the soil with highest rates of net N mineralization and nitrification. Concurrently, we used an N-rich source (glycine) to determine if decreasing C:N would stimulate net N mineralization and initiate net nitrification in the soil exhibiting lowest rates of N mineralization and no net nitrification.
Sucrose Additions
Adding sucrose solutions to soil from the site with the highest rates of net N mineralization and nitrification significantly decreased rates of both processes, indicating that both NH4 and NO3 were immobilized by soil microbes (Fig. 5)
, although immobilization of NO2, which was not assessable in this study, is also possible (Fitzhugh et al., 2003). Net immobilization of inorganic N increased significantly as a logarithmic function of sucrose concentration.
|
Magill and Aber (2000) used two concentrations of glucose, along with leaf leachate, cinnamic acid, and humic acid, as sources of added C to soils of mixed hardwood forests and found that N immobilization responded more sensitively to glucose than to the other C sources. Overall, they found that both low and high levels of glucose additions (high glucose = 2 x low glucose) immobilized 6 and 15 mg N kg–1 mineral soil in 5 wk, respectively. Using field applications of sucrose and extracts of roots and litter to soils of a Tennessee hardwood forest, Johnson and Edwards (1979) found that adding sucrose rapidly increased net N immobilization, and did so more than less-labile sources of C, such as root and litter leachates. They found that sucrose additions decreased net N mineralization by nearly 70% in a 19-d period. Although our experiment did not include less-labile sources of C, our results are consistent with the findings of Johnson and Edwards (1979) and Magill and Aber (2000) for the effects of experimental additions of labile C on microbial immobilization of available N.
Glycine Additions
Adding glycine solutions to soil from the site with the lowest rates of net N mineralization (and no net nitrification) significantly increased net N mineralization, from 0.7 µg N g–1 soil d –1 for the control to 1.8 and 13.0 µg N g–1 soil d –1 for 0.01 M and 0.1 M glycine additions, respectively (Fig. 6)
. Thus, the 0.01 M glycine addition increased mineralization of the low-N soil to a level similar to that of the control–high N soil (Fig. 5). Net N mineralization increased significantly as a linear function of glycine concentration. By contrast, both the 0.01 M and the 0.1 M glycine concentration failed to stimulate net nitrification (Fig. 6).
|
Results for net nitrification suggest that it is limited by neither substrate quality (i.e., C:N ratio) nor availability of NH4. This is in contrast to results of Hall and Matson (2003) working in severely N-limited soils of Hawaii. Their soils were similar to soils of our untreated low-N site in exhibiting no net nitrification under field conditions. In contrast to our results, however, they were able to initiate net nitrification in their soils by adding high levels of N, concluding that their findings resulted from increases in numbers and activity of nitrifying microbes (Hall and Matson, 2003). Although the most parsimonious explanation for our results would be that FEF soils that fail to produce NO3 are lacking in nitrifier populations, this explanation was rejected by an earlier investigation (Gilliam et al., 2001a) that demonstrated for these soils the presence of the amoA gene, the enzyme that catalyzes the first step in nitrification (NH4 oxidation to NO2).
Synthesis: Causes and Consequences of Spatial Variability in Nitrogen Processing at Fernow Experimental Forest
Results of experimental investigations in this study support published results based on field studies. The four sites used in this study were chosen because they represented a gradient from extreme low to high rates of N processing, based on in situ incubations. This gradient for net mineralization or nitrification remained under controlled temperature conditions and did so across all three incubation temperatures: 10, 20, and 30°C. The significant increase in Ca extractability in response to increased NO3 production in these soils is consistent with results of previous studies at sites other than FEF (Lawrence et al., 1995; Currie et al., 1999) and with soil solution and stream chemistry data reported for FEF (Adams et al., 1997; Edwards et al., 2002). Microbial processing of N in the control–high N soil was highly sensitive to labile C, supporting findings of earlier studies (Johnson and Edwards, 1979; Magill and Aber, 2000). The additional source of immediately available energy stimulated soil microbes at this site to take up both NH4 and NO3 to meet increased microbial demand for available N. Although ammonifying microbes in soils of the control–low N site were substrate limited (especially with respect to decreasing C:N), nitrifier populations appear to be limited or inhibited by some factor (or suite of factors) other than available NH4, which increased linearly with glycine addition, but was not accompanied by increases in NO3.
We hypothesize that factors limiting net N mineralization and/or inhibiting net nitrification at the control–low N site may, in large part, explain the great spatial heterogeneity in N processing within untreated WS4. These include soil chemistry, which is often spatially quite variable (Johnson et al., 2000). This is especially seen in the extremes of high Al and low Ca concentrations, which themselves are often a function of weathering rates of mineral soil. Thus, spatial heterogeneity in N processing at FEF may have arisen, at least in part, from spatially variable rates of weathering in watershed soils.
Net nitrification was negatively related to preincubation levels of Al across the spatial gradient, with the relationship being fit most closely with a logarithmic curve [y = 36.37 – 5.66ln(x), r2 = 0.959, P < 0.01] (Fig. 7) . This apparent inhibition of nitrification via Al toxicity is consistent with findings of Illmer et al. (2003), who demonstrated that increases in KCl-extractable Al had a highly significant negative effect on several microbial parameters, including microbial biomass and respiration, in 95 soils from throughout Tyrol, Austria.
|
Roots of ericaceous plants support a special type of mycorrhizae—ericoid mycorrhizae—that secrete organic acids which limit N-mineralizing microbes in general and inhibit nitrifying bacteria in particular (Straker, 1996). Read and Perez-Moreno (2003) showed that, by attacking structural polymers, ericoid mycorrhizae have the potential to render nutrients unavailable to microbes, and that natural selection favors ericoid mycorrhizae with well-developed saprotrophic capabilities to retain N as organic complexes in the soil. Ericaceous plant detritus contains high concentrations of secondary compounds (e.g., polyphenols) that have been shown to shift the pathway of N cycling from inorganic to organic forms (Northup et al., 1998; Schimel et al., 1998). Thus, we hypothesize that the predominance of hillside blueberry on WS4 is part of a positive feedback mechanism between soils with low mineral N and plants that simultaneously (i) are selected for and highly competitive in low-N soils, and (ii) maintain low-N conditions.
An additional positive feedback mechanism at FEF may be related to soil chemistry. Ericoid mycorrhizae have been shown to survive high levels of toxic metals, including Cd, Zn, and Al, in soils (Perotto et al., 2002). Although the specific mechanisms for survival are not completely understood, ericoid fungi have the capacity for rapid adaptation to high metal concentrations. For example, when strains of the ericoid fungus Oidiodendron maius were isolated from Cd-, Zn-, and Al-polluted soils, they were found to grow better in vitro in media with those metals than did strains isolated from nonpolluted soils (Lacourt et al., 2000). Furthermore, these mycorrhizae can both (i) increase mobility of toxic metals, which increases toxicity to soil microbes (e.g., Illmer et al., 2003), and (ii) reduce susceptibility of host plants to metal toxicity, which allows ericaceous species to survive in soils with levels of metals that are toxic to other, nonericoid species (e.g., Sharples et al., 2000). Consequently, via its ericoid mycorrhizae, hillside blueberry that dominates the herbaceous layer of the control–low N site at FEF can tolerate the high levels of extractable Al in those soils that exhibited no net nitrification.
Understanding spatial variability in N processing has important practical applications. These include predicting the response of soil N to disturbance and the challenge of sustainable management of N-saturated forest stands (Fenn et al., 1998; Adams 1999; Gilliam et al., 2004). Vitousek et al. (1982) concluded that sites with low rates of N processing would be highly resistant to disturbance, whereas sites with high rates would be much less resistant. Thus, we would expect a disturbance (e.g., wind damage, forest harvesting) involving only the low-N portion of WS4 to bring about few changes in soil N. By contrast, we predict that similar disturbances to other areas of WS4 with high rates of N turnover would further enhance N mobility, which may exacerbate loss of Ca.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 8, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
