Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 19 March 2008
Published in Crop Sci 48:487-494 (2008)
© 2008 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Agricola
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Related Collections
Right arrow Clover
Right arrow Root Development
Right arrow Crop Genetics

Genotypic Variation for Root Trait Morphology in a White Clover Mapping Population Grown in Sand

M. Z. Z. Jahufera,*, S. N. Nicholsb, J. R. Crushb, Li Ouyangb, A. Dunna, J. L. Forda, D. A. Careb, A. G. Griffithsa, C. S. Jonesa, C. G. Jonesc and D. R. Woodfielda

a Pastoral Genomics, AgResearch Ltd., Grasslands Research Centre, Private Bag 11008, Palmerston North, New Zealand
b Pastoral Genomics, AgResearch Ltd., Ruakura Research Centre, Private Bag 3123, Hamilton, New Zealand
c Pastoral Genomics, ViaLactia Biosciences (NZ) Ltd., Auckland, New Zealand

* Corresponding author (zulfi.jahufer{at}agresearch.co.nz).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
A study of 386 white clover (Trifolium repens L.) mapping population F1 progeny was conducted to quantify the type and magnitude of genotypic variation for a range of root morphology traits. Clones of each of the 386 white clover progeny were grown in sand. There were significant (P < 0.05) genotypic variance components and repeatability estimates for all the root traits examined. Progeny genotypes with high expression of key traits, including number of root tips and number of root forks were identified. These types may improve phosphate uptake as their highly branched roots will explore a large volume of soil per unit root weight. A strong positive phenotypic and genotypic correlation between several root traits was identified. This suggests an opportunity for indirect selection. For example, selection for high root fork number, a trait that is relatively less complicated to measure, should result in the concurrent increase in expression of the following root traits: surface area, number of tips, volume, and dry weight. Comparison of results from the sand-based trial with an earlier trial using hydroponic conditions, with clones of the same 386 progeny, showed similar correlations exist among the root traits in both systems. The progeny genotype-by-trait Best Linear Unbiased Predictor matrix generated from the sand study is currently being used for the identification of root trait quantitative trait loci.

Abbreviations: BLUP, Best Linear Unbiased Predictor • MP, mapping population • QTL, quantitative trait loci • RD, root dry weight • RF, number of root forks • RF/RL, root forks per unit length • RL, root length • RS, root surface area • RT, number of root tips per plant • RT/RL, root tips per unit root length • RV, root volume


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
MOST WHITE CLOVER (Trifolium repens L.) breeding programs in New Zealand and elsewhere have targeted aboveground plant traits (Caradus and Williams, 1989; Ayres et al., 1995). This has resulted in successful genetic gain and led to the development of new cultivars (Caradus et al., 1995). However, limited work has been performed to explore the potential for genetic improvement of the white clover root system, especially in the area of root morphology and architecture in relation to water and nutrient uptake. It is known that genetic variation in root morphology is extensive in white clover, and the narrow sense heritabilities for a range of root traits that have been described (Caradus, 1990; Caradus and Woodfield, 1998) are of sufficient magnitude to indicate a good response to selection.

The significant influence of root morphology on water uptake (Gardner, 1964; Hund, 1974), nutrient uptake (Andrews and Newman, 1970), and plant persistence (Ronningen, 1953) is well documented. Phenotypic evaluation of root traits has been performed successfully for a number of crop species, including soybean [Glycine max (L.) Merr.] (Pantalone et al., 1996), rice (Oryza sativa L.) (Champoux et al., 1995), and maize (Zea mays L.) (Costa et al., 2002). For white clover, root trait studies have primarily focused on moisture stress (Stevenson and Laidlaw, 1985; Annicchiarico and Piano, 2004) and soil nutrient uptake (Jackman and Mouat, 1970). However, the progress of root trait characterization and evaluation with regard to genetic improvement has not been as rapid as for aboveground plant traits. The complexity of root traits, their measurement, and the effect of soil environments on phenotype have limited the progress of research on root systems (O'Toole and Bland, 1987; Tuberosa et al., 2002).

Root morphology is a criterion associated with the efficiency of plant uptake of soil phosphates (Jackman and Mouat, 1970). Phosphates diffuse slowly in the soil (Nye, 1977). The movement of phosphate to the root surface is the rate limiting step in phosphate acquisition by plants. Plants with finely divided roots that permeate a large volume of soil are more efficient in obtaining phosphate than plants with coarse, unbranched roots. This has been shown experimentally for white clover grown with browntop (Agrostis capillaris L.) (Jackman and Mouat, 1970). Browntop has thinner roots than white clover (0.16 mm compared to 0.26 mm) with longer root hairs (0.68 mm compared to 0.23 mm), and 95% of the root length covered in hairs compared to 68% for white clover (Evans, 1977). It was shown that white clover grown with browntop required three times the superphosphate input of clover growing alone, to reach the same herbage yield. Improvement of the white clover root system in terms of its soil permeation capacity could be useful not only for phosphate uptake but also for water acquisition. Increase in the level of expression of root traits such as length, surface area, volume, number of tips, and number of forks could result in a more reticulated root system.

The availability of molecular marker technology has provided plant breeders with a new set of tools to evaluate plant diversity at a DNA level. For traits that are difficult to measure, the application of indirect selection using trait-associated molecular markers (marker-assisted selection), will improve the efficiency of applied plant breeding programs (Collard et al., 2005). The identification of quantitative trait loci (QTL) and associated molecular markers for key root traits such as length, surface area, volume, number of tips, and number of forks is likely to significantly enhance the development of new clover cultivars with improved rooting characteristics using marker-assisted selection.

We are undertaking a program of identifying QTL associated with root architectural and morphological traits in white clover. Jahufer et al. (2006) reported results on phenotyping a mapping population of 500 F1 white clover progeny, grown under hydroponic conditions, for a range of morphological root traits. This paper describes the phenotypic analysis of a sample of 386 cloned genotypes from the same mapping population for root trait expression when grown in sand.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
The mapping population (MP) used for this study was generated by pair crossing two unrelated white clover genotypes, selected from two different breeding populations, which had distinctly different root morphology and architecture (Jahufer et al., 2006). A random sample of 386 progeny clones, from a total of 500 F1 MP progeny, was used in this study.

Experimental Design
The experimental design was a row-column design with two replicates containing randomized clones of the 386 MP progeny. Each of the 386 MP progeny was represented by a single clone (stolon cutting) in each of the two replicates. Each replicate also contained repeated clonal checks (Gleeson, 1997; Kempton and Gleeson, 1997), enabling characterization of potential environmental and spatial effects across columns and down rows. Sixteen clones from each of two independent unrelated genotypes were used as the repeated checks and systematically allocated within each replicate.

Plant Establishment
Rooted clones of the 386 MP progeny genotypes were planted in 15-cm pots containing potting mix, in a glasshouse at the Ruakura Research Centre in Hamilton, New Zealand, in November–December 2005. All stock plants were maintained in a glasshouse for 2 mo. Stock plants for the two experimental checks were also identified and maintained in the same glasshouse. In February 2006, stolon tip cuttings with two to three nodes were taken from all plants and trimmed to one fully expanded leaf. This also included the stolon tips for the two experimental checks. Cuttings were planted in trays of sand and maintained with watering and half-strength nutrient solution (Hewitt, 1966; Table 1 ) in the glasshouse for 10 d, until new nodal roots were appearing.


View this table:
[in this window]
[in a new window]

  		Table 1. Nutrient solution (based on Hewitt, 1966) applied to the 386 F1 white clover mapping population progeny grown in sand. Full-strength solution was diluted 50% for application to cuttings during root formation.

 
Rooted cuttings were carefully removed from the trays and transferred into 15-cm pots of sand maintained in a glasshouse. One hundred milliliters of full-strength nutrient solution at a pH of 6.2 to 6.5 was applied three times per week to each pot. After 10 d of growth, at average day/night temperatures of 24/17°C, all the plants were harvested for their root systems. Previous experimental work had shown that this duration of root growth was satisfactory for studying root morphology and architecture in white clover (Crush et al., 2005). The harvested roots were carefully washed in water to remove the sand. Nodal roots were then excised and preserved in 70% ethanol.

Measurements
The roots harvested from each genotype were spread out on a flatbed scanner (Epson Expression 1680, Epson Inc., Long Beach, CA) and scanned using WinRhizo 2004b (Regent Instruments Inc., Quebec, QC) at 200 dpi. Top and bottom lighting systems were used to eliminate shadows and maximize contrast. The captured grayscale images were analyzed with WinRhizo to measure root length, surface area, average diameter, volume, number of tips, and number of forks. The roots were rinsed to remove ethanol, dried at 70°C for 24 h and weighed.

Traits Measured and Derived Ratios
Root diameter (mm), number of root forks, root length (cm), root surface area (cm2), number of root tips, root volume (cm3), root dry weight (g), ratio of root forks/root length (number of root forks per centimeter), ratio of root tips/root length (number of root tips per centimeter), and specific root length (cm mg–1).

Data Analysis
The data were examined using the variance component analysis procedure, Residual Maximum Likelihood (REML) option, in GenStat 7.1 (GenStat, 2003). A completely random linear model was used in the analysis using the REML algorithm. The combination of variation across columns, rows, and also the variation across the clonal checks, enabled adjustment of the genotype means for spatial variation. The analysis resulted in the generation of a progeny genotype x trait matrix consisting of adjusted means/Best Linear Unbiased Predictor (BLUP) (White and Hodge, 1989) values.

The linear model used in the analysis is:

Formula 1[1]
where Yijkl is the value of an attribute measured from genotype i in replicate j in row k and column l, and i = 1, ..., ng; j = 1, ..., nb; k = 1, ..., nr; and l = 1, ..., nc, where g, b, r, and c are, genotypes, replicates, rows, and columns, respectively; M is the overall mean; gi is the effect of progeny i, N(0,{sigma}2{varepsilon}); bj is the effect of replicate j, N(0,{sigma}2{varepsilon}); rjk is the effect of row k in replicate j, N(0,{sigma}2{varepsilon}); cjl is the effect of column l in replicate j, N(0,{sigma}2{varepsilon}); (rc)jkl is the effect of the interaction between row k and column l in replicate j, N(0,{sigma}2{varepsilon}); and {varepsilon}ijkl is the residual effect of genotype i in row k and column l in replicate j, N(0,{sigma}2{varepsilon}).

Repeatability
The genotypic variance components estimated for the root traits using REML were used to estimate repeatability (R) (Falconer, 1989) on a progeny clone mean basis.

Formula 2[2]
where the respective variance components and their divisors were defined in relation to the linear model.

Correlation Coefficients
Phenotypic and genotypic correlation coefficients for the root traits measured were estimated on a progeny clone mean basis according to Becker (1984).

Pattern Analysis
Following the variance component analysis, the progeny genotype x root trait BLUP matrix was analyzed using pattern analysis—a combination of cluster analysis and principal component analysis (Watson et al., 1995; Kroonenberg, 1994; Gabriel, 1971). The objective of conducting pattern analysis was to provide a graphical summary of the information on genotypic variation within the large data set. The biplots generated enabled assessment of the genotypic variation on a multivariate scale. Only BLUPs from root morphological traits for which there was significant (P < 0.05) genotypic variation among the MP progeny were included in the pattern analysis.

Root trait data from an earlier experiment in hydroponic culture (Jahufer et al., 2006) that included clones of the same 386 MP progeny, were combined with data from the experiment reported here, and analyzed using pattern analysis. This analysis allowed comparison of root behavior in hydroponic and sand root media.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
The differences between the minimum and maximum values for all the root traits indicated a broad range of phenotypic variation among the 386 MP progeny (Table 2 ). There was significant (P < 0.05) genotypic variation among the 386 MP progeny for all the root traits measured. The progeny mean trait values for number of root tips per plant (RT) and number of root tips per unit root length (RT/RL) measured in the sand medium experiment, based on 10-d growth, were higher than the values, 177 for RT and 0.804 for RT/RL, that were measured in the same 386 progeny clones evaluated in a hydroponic medium for over 4 to 5 wk (Jahufer et al., 2006). Similarly Crush et al. (2005), studying the root systems of five white clover lines in sand and nutrient solution, reported higher expression of RT and RT/RL in clones growing in the sand medium. These responses to the sand medium were explained as a reaction by the root tips to contact with sand particles, resulting in earlier and more frequent branching (Crush et al., 2005). Similar responses have been reported for maize roots growing in a hydroponic culture and constrained by glass beads (Groleau-Renaud et al., 1998).


View this table:
[in this window]
[in a new window]

  		Table 2. Means, ranges, variance components ({sigma}2) with associated standard errors (±SE), and progeny clone mean repeatability (R) for different root traits measured from 386 white clover F1 mapping population progeny grown in sand.{dagger}

 
There was a large component of genotypic variation ({sigma}2g) for all traits. The progeny clone mean repeatability (R) enabled estimation of an upper limit of the degree of genetic determination (Falconer, 1989), and for most traits was above 50%. Caradus (1990) reported broad sense heritability values for the white clover traits root diameter, root length, and root dry weight, which were similar to the repeatability (R) presented in Table 2. According to Caradus (1990), broad sense heritability for white clover root traits rarely exceeds 50%. Care should be taken in the interpretation of the R values in Table 2 as broad sense heritability, as the estimates were based on F1 progeny clones. However, it should also be noted that white clover is a cross-pollinating species and the 386 F1 progeny were a result of a cross between two unrelated heterozygous parents and their progeny are genetically similar to an F2 in self-pollinating species (Williams, 1987).

Pattern Analysis
Cluster analysis of the progeny genotype x root trait BLUP matrix generated five progeny groups (Table 3 ). The number of progeny in each group ranged from 49 (in Group 3) to 112 (in Group 4). Group 3 had the highest mean values for number of root forks (RF), root length (RL), root surface area (RS), number of root tips (RT), root volume (RV), and root dry weight (RD). The ratios root forks per unit length (RF/RL) and root tips per unit length (RT/RL) were also highest in Group 3. Group 1 had the highest mean root diameter.


View this table:
[in this window]
[in a new window]

  		Table 3. Within-group progeny means for each root trait based on the five clusters generated from cluster analysis of 386 white clover F1 mapping population progeny grown in sand.{dagger}

 
Cluster analysis of the combined genotype x root trait BLUP matrix consisting of data from both the hydroponic (Jahufer et al., 2006) and sand root growth media experiments, resulted in generating six progeny groups. Groups 1 and 6 contained the lowest and highest number of genotypes, respectively (Table 4 ). Group 1 consisted of 12 genotypes that showed high mean expression for the root traits RL, RS, RT, and RV, across both the hydroponic and sand root growth media. Group 1 in Table 4 showed similar characteristics to Group 3 in Table 3 in terms of high expression of the traits RL, RS, RT, and RV.


View this table:
[in this window]
[in a new window]

  		Table 4. Within-group progeny means of the six clusters, generated from cluster analysis of the 386 F1 white clover mapping population progeny, based on mean root trait expression across the hydroponic and sand growth media experiments conducted in years 2005 and 2006, respectively.{dagger}

 
Research by Jackman and Mouat (1970) and Hill et al. (2006) showed that plants with finely divided roots that provide a high root length frequency per unit volume of soil were more efficient in obtaining phosphorus than plants with coarse, unbranched roots. The characteristics of the roots of progeny in Group 3 in Table 3 and Group 1 in Table 4 indicated that these plants had robust root systems with the potential to enhance soil penetration and nutrient uptake.

The biplot (Fig. 1 ) generated from principal component analysis of the 386 MP progeny, based on 10 root traits, is a graphical summary of the large progeny genotype x trait BLUP matrix. There was a strong degree of separation among the progeny groups, with minor overlaps. Group 3 consisted of individuals with above average expression for the root traits: RF, RL, RS, RT, RV, RD, and the ratios RF/RL and RT/RL (Fig. 1). The progeny marked a, b, c, and d in Fig. 1 showed a much higher expression for these traits in comparison to the rest of the population of genotypes.


Figure 1
View larger version (30K):
[in this window]
[in a new window]

  		Figure 1. Biplot generated using standardized Best Linear Unbiased Predictor values of genotype root trait means from 386 F1 progeny from a white clover mapping population grown in sand. Components I and II account for 55 and 21% of total variation, respectively. The different symbols indicate progeny Groups 1 to 5 generated from cluster analysis. The vectors represent the root traits: RDI, root diameter (mm); RF, number of root forks; RL, root length (cm); RS, root surface area (cm2); RT, No. root tips; RD, root dry weight (g); RV, root volume (cm3); FL, ratio RF/RL (no. cm–1); TL, ratio RT/RL (no. cm–1); SRL, specific root length (cm mg–1). The letters a, b, c, and d, indicate individual progeny genotypes that had the highest above average expression for a range of traits. The arrow (Figure 1) indicates the label of a directional vector that is not legible.

 
The association among directional vectors is represented by the angle between them, the smaller the angle (<90°) the stronger the positive association (and vice versa). The vectors for RT, RF, RL, RS, RD, and RV indicated a strong positive association among these traits (Fig. 1). Root diameter (RDI) had a weak positive association with RV, RD, and RS. The association of RDI with RL, RF, RT, RF/RL, RT/RL, and specific root length (SRL) was negative.

The association among the root traits shown by the directional vectors is further supported by the phenotypic and genotypic correlation coefficients presented in Table 5 . There were strong positive phenotypic correlations amongst root traits RF, RL, RS, RT, RV, and RD. The genotypic correlations among these traits were also strong and positive indicating a strong positive association among these root traits at a genetic level. The limited information available on the correlation of root traits (Caradus, 1990) restricts comparison of results. However, based on results from this study, the levels of positive genotypic correlation among key root traits, suggests an opportunity for indirect selection. For example, selection for high RF, a trait that is relatively less complicated to measure, should result in the concurrent increase in expression of the traits; RL, RS, RT, RV, and RD. The repeatability R for the trait RF was also relatively high.


View this table:
[in this window]
[in a new window]

  		Table 5. Phenotypic (below diagonal) and genotypic (above diagonal) correlation coefficients among the morphological root traits measured from the 386 white clover F1 mapping population progeny grown in sand.{dagger}

 
Pattern analysis of the combined progeny genotype x root trait BLUP matrices generated from an earlier hydroponic trial (Jahufer et al., 2006) and the current sand trials, for the traits RDI, RL, RS, RT, RV, and RT/RL, resulted in the graphical summary presented in Fig. 2 . The numbers 5 and 6 included as a suffix in the root trait codes, refer to the hydroponic experiment in 2005 and the sand experiment in 2006, respectively.


Figure 2
View larger version (36K):
[in this window]
[in a new window]

  		Figure 2. Biplot generated using standardized Best Linear Unbiased Predictor values of root trait means from clones of the 386 F1 white clover mapping population progeny grown in hydroponic and sand conditions. Components I and II account for 36 and 27% of total variation, respectively. The symbols indicate progeny Groups 1 to 6 generated from cluster analysis. The vectors represent the root traits: RDI, root diameter (mm); RL, root length (cm); RS, root surface area (cm2); RT, number of root tips; RV, root volume (cm3); TL, ratio RT/RL (no. cm–1). The numbers 5 and 6 included as a suffix in the root trait codes, refer to the hydroponic experiment in year 2005 and sand experiment in year 2006, respectively. The arrows (Figure 2) indicate the labels of directional vectors that are closely aligned and those that are not legible.

 
The directional vectors in Fig. 2 indicate a similar association among the root traits was maintained in both the hydroponic and sand environments. There was a strong positive association among the root traits RL, RS, RT, and RV in both the hydroponic and sand environments. Considering the expression of each of the root traits across the hydroponic and sand environments, the positive association between the traits RL 2005 and RL 2006, RS 2005 and RS 2006, RT 2005 and RT 2006, RV 2005 and RV 2006, indicated that the growth and development of the root systems in both environments followed similar trends. These data support a previous investigation of root traits in sand and hydroponics (Crush et al., 2005) that demonstrated only small differences between root systems developed in these two environments.

Of the six progeny groups, Group 6 showed a high degree of overlap with other groups (Fig. 2). The positions of Groups 1, 2, 3, 4, and 5 were distinct. Relative to the directional vectors, Groups 1, 2, 4, and 6 contained progeny that had above average expression for the root traits RL, RS, RT, and RV in both the hydroponic and sand environments. It is important to note that Groups 1 and 4 contained individuals that had high expression for the traits: RL, RS, RT, and RV in both the hydroponic and sand environments. These genotypes could be useful in further investigation of the association of white clover root permeation capacity and soil phosphate uptake. They could also be assessed as potential parental genotypes in breeding focused on the improvement of white clover phosphate uptake.

The information developed here, on genotypic variation for root traits, pattern of genotype distribution, and trait correlation, among the 386 F1 mapping population progeny, provide important criteria for breeding programs targeting objectives such as improving soil phosphorus uptake. Progeny genotypes with high expression of traits such as number of root tips and number of forks will be important for improving phosphorus uptake as they will have highly branched roots that explore a large volume of soil per unit root weight. This type of root morphology will reduce the distances that phosphate has to diffuse to the root surface and improve the plants access to soil phosphorus. Phosphate efficiency is a major issue in New Zealand because phosphate fertilizer is a leading cost in farm budgets. Optimizing nitrate interception and soil water capture are other important goals which may be achieved through manipulation of root morphology (Sullivan et al., 2000). Summer moisture stress is a significant constraint to vegetative persistence and herbage yield of white clover (Hutchinson et al., 1995; Brink and Pederson, 1998). Selection for specific root traits such as root diameter was shown to enhance white clover yield and persistence under moisture stress conditions (Caradus and Woodfield, 1998).

The strong positive phenotypic and genotypic correlation coefficients between the root traits examined in this study indicate the possibility of applying a strategy of correlated response to selection (Falconer, 1989). This will be especially useful for those traits such as number of root tips that are difficult to measure.

The significant (P < 0.05) genotypic variance components and repeatability estimates for all the root traits examined in this study indicate the potential genetic variation within the mapping population. These results provide the basis for undertaking further research and development work to identify QTL for the root traits reported in this study. The progeny genotype x trait BLUP matrix generated from this study is currently being analyzed together with the associated molecular marker information for the identification of root trait QTL, as part of the development of a marker-assisted selection breeding program for white clover (Jones et al., 2006). It is envisaged that a combination of phenotypic and molecular marker root selection techniques, will result in the development of new white clover cultivars with novel root systems with the potential for efficient nutrient and water uptake.


   ACKNOWLEDGMENTS
 
The authors wish to acknowledge Pastoral Genomics for financial support for this research project. Pastoral Genomics is a joint venture cofunded by Meat and Wool NZ, Fonterra, AgResearch Ltd., Deer Industry NZ, the New Zealand Foundation for Research, Science and Technology, and Dairy InSight Inc.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication March 21, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Agricola
Right arrow Articles by Jahufer, M. Z. Z.
Right arrow Articles by Woodfield, D. R.
Related Collections
Right arrow Clover
Right arrow Root Development
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome