Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 August 2005
Published in Crop Sci 45:1844-1850 (2005)
© 2005 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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Agricola
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Related Collections
Right arrow Crop Genetics
Right arrow Clover

CROP BREEDING, GENETICS & CYTOLOGY

A QTL Analysis of White Clover Seed Production

B. A. Barretta,*, I. J. Bairdb and D. R. Woodfielda

a AgResearch Limited, Grasslands Research Centre, Private Bag 11008, Palmerston North, New Zealand
b AgResearch Limited, Canterbury Agriculture and Science Centre, Gerald Street, Lincoln, New Zealand

* Corresponding author (brent.barrett{at}agresearch.co.nz)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
White clover (Trifolium repens L.) exhibits complex inheritance of seed production potential. Genome maps allow plant breeders to dissect the genetic control of complex traits into discreet factors with Mendelian inheritance, each of which is called a quantitative trait locus (QTL). A pair-cross between two highly heterozygous white clover genotypes created an F1 mapping population (n = 182) used to investigate the genetic control of seed yield (SY) and three component traits: inflorescence density (ID), yield per inflorescence (YI), and thousand-seed weight (TSW). Analysis of 3 yr of field data from a replicated trial in the primary seed production region of New Zealand indicated no strong correlations among the component traits. The ANOVA indicated significant genotype, year, and genotype x year effects for SY, ID, and YI; and genotype and genotype x year effects for TSW. Using genotype data for 209 microsatellite loci in the allotetraploid genome, the multiple QTL model analysis resulted in the discovery of 23 QTLs related to seed production in white clover. Both parent plants harbored beneficial alleles for SY and the three component traits. A region of the genome at the distal end of linkage group D2 was implicated in a number of traits related to seed production, and was strongly implicated in the control of YI and SY. The central region of linkage group C2 was strongly implicated in the control of ID and SY. Parent trait values were similar for ID, yet the population was highly variable for this trait and segregated for multiple QTLs, indicating the utility of this experimental design for QTL discovery in crosses among heterozygous individuals with similar phenotypes. Most QTLs were dominant, suggesting suitability for marker-assisted selection (MAS) in white clover.

Abbreviations: AIL, advanced intercross line • cM, centimorgan • G x E, genotype by environment • ID, inflorescence density • IM, interval mapping • MAS, marker-assisted selection • MQM, multiple-QTL mapping • QTL, quantitative trait locus • SY, seed yield • TSW, thousand-seed weight • YI, yield per inflorescence


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
FORAGE LEGUMES are a sustainable source of high-quality feed and plant-available nitrogen. White clover (2n = 4x = 32) is a component of many intensive pastoral systems in temperate latitudes, and will continue to play an important role in sustainable production systems (Rochon et al., 2004).

Seed-propagated forage species such as white clover must deliver twice: once in the seed grower's paddock, and once again in the farmer's paddock. Assessment of breeding progress during the past 60 yr indicates increased white clover productivity has been achieved in the farmer's paddock; however, during the same period, there was no change in seed production potential (Woodfield and Caradus, 1994). Enhancement of SY performance would be attractive to seed growers, and may benefit farmers by enhanced natural reseeding of improved white clover in paddocks (Pederson and Brink, 2000).

Seed yield is a complex character showing continuously distributed values, typical of multigenic control with environmental influences on trait expression. Several component traits influence SY in white clover. Seed yield is the product of ID and YI. Yield per inflorescence is itself a product of variation in seed size and density (measured as TSW), floret number per inflorescence, and floret fertility. The use of increased YI rather than increased ID may be a preferred strategy to boost SY in white clover. Each inflorescence is at the cost of a stolon, suggesting that types with high inflorescence numbers will have compromised on-farm persistence.

The balance between SY potential and forage performance potential is yet to be clearly defined. Annicchiarico and colleagues (Annicchiarico et al., 1999) report a strong negative correlation between stolon density and SY, indicating emphasis on SY may negatively impact forage persistence and performance characteristics. In contrast, Widdup and colleagues (Widdup et al., 2004) report a successful reselection for SY components within commercial varieties with little change in plant morphology, suggesting that within some elite populations seed production potential may be significantly increased without compromising agronomic performance.

Investigators have shown that typical white clover SY levels remain below genetic potential (Williams et al., 1998), and have detected significant genetic variation for SY and yield components within elite and unimproved germplasm sources (Jahufer and Gawler, 2000). Heritability of white clover SY and components has been estimated at 0.60 to 0.70 (Annicchiarico et al., 1999; Woodfield et al., 2004), which suggests the trait is amenable to selection and genetic analysis.

Understanding of the genetic control of white clover SY has been advanced by a diallel among parent genotypes selected for either ID or YI. Analysis of the diallel data indicated that these two key characteristics are under independent genetic control (Woodfield et al., 2004). Further elucidation of the genetic basis of SY and its components may be best achieved using a QTL discovery approach (Liu, 1998).

A microsatellite map of the white clover genome has been developed (Barrett et al., 2004) which may be used in trait-targeted populations to discover QTLs that influence heritable phenotypes such as SY and its components. The QTL discovery will improve understanding of the regulation of SY, and may contribute to the development of MAS strategies for genetic testing of parent population performance, either before or in tandem with field trials.

The objective of this research was to use field and laboratory-based genetic analysis to discover QTLs that regulate white clover SY and three of its components: ID, YI, and TSW.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Material
A population of 182 full-sib F1 plants was created by hand pollination between two phenotypically divergent, highly heterozygous white clover genotypes: ‘P6525’, a parent of ‘Grasslands Sustain’ (Caradus et al., 1997), and ‘P3647’, an elite plant in a nematode resistance recurrent selection program at AgResearch (Mercer et al., 1999). P6525 exhibits medium leaves, cold-induced leaf fleck, leaf mark, many long thin stolons, and dense growth; whereas P3647 exhibits large leaves, no leaf fleck, no leaf mark, few short thick stolons, and open growth.

Field Trials
The two parents and 182 full-sibs were evaluated at Lincoln, New Zealand, for seed production during three growing season environments: 2001/2002, 2002/2003, and 2003/2004. Each genotype was replicated three times in a randomized complete block design. Entries were rerandomized and clonally propagated (via stolon cuttings) for each season. Each plot was a single clonal copy of a genotype, contained by a 50-cm-diam. ring set into the ground to establish a uniform plant area for observing phenotypes. During the first two seasons, weekly counts during the growing season were made of active inflorescences per plot. The sum of these counts was used as the ID (inflorescences plot–1) value. At maturity, each plant was harvested and SY (grams plot–1) and TSW (g) were recorded. Yield per inflorescence (g seed inflorescence–1) was calculated at SY/ID. In the final season, only SY values were recorded.

DNA Marker Data
Data collection and analysis for all microsatellite markers are described elsewhere (Faville et al., 2003). The medium-density white clover genome map used is that reported by Barrett et al. (2004). Two-hundred-and-nine microsatellite loci from that map were genotyped in this population, providing a mean density of 11 centimorgans per marker locus per parent for QTL discovery analyses.

Analysis of Variance
Seed yield, ID, YI, and TSW trait data were analyzed. Assuming fixed effects, a two-way ANOVA with blocking was used to test for significant genotype, environment (i.e., year), and replicate main effects, and genotype by environment (G x E) interaction effects for each of the four traits, using GenStat version 7 (Rothamsted Exp. Stn., UK) software. The (G x E) mean squares values were used as the error term to assess significance of main effects for SY, ID, and YI. The residual mean square was used as the error term to assess significance of interaction effects for all traits, and of main effects for TSW.

QTL Analysis
The QTL discovery for each trait was done on a single-year basis using nine trait datasets named in trait–year format as SY02, SY03, SY04, ID02, ID03, YI02, YI03, TSW02, and TSW03. For each trait, the within-year mean value for each genotype was input to the software package MapQTL 4.0 (Kyazma B.V., Wageningen, the Netherlands) for interval mapping (IM) QTL analysis. Subsequently, forward selection (Jansen, 2004) was used to inform cofactor selection in a multiple-QTL mapping (MQM) approach implemented in MapQTL 4.0. Statistical significance (P < 0.05) was declared at a genome-wide LOD threshold of 4.3 based on permutation testing of the dataset (n = 1000 permutations). Quantitative trait loci were coded as having four alleles (a, b, c, and d) in the mapping population, in combination ab for parent 3647 and cd for parent 6525, which assort into genotype classes ac, ad, bc, and bd in the F1 progeny. Using MapQTL 4.0, genotype class means were calculated for all four combinations at each marker locus, and for inferred genotypes at 5-cM (centimorgan) intervals if an interlocus gap exceeded 5 cM.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Field Trials
The data exhibited variation for each trait measured. Population performance for SY decreased during the course of the 3-yr field trial, and ID and YI values also decreased in 2003 (Table 1). Thousand-seed weight was similar in 2002 and 2003. Parent 6525 exhibited higher (P < 0.05) trait values than parent 3647 for YI, and lower values for TSW. Parents were not different for ID or SY in any environment. Progeny performance values for SY02, SY03, and SY04 exhibited a wide range: 0.7 to 50.2 g plot–1, 0.5 to 34.5 g plant–1, and 0.9 to 22.0 g plant–1, respectively. A similar range of values was observed for ID02 and ID03, from 4 to 266 and 2 to 229 inflorescences plot–1, respectively. The range of values observed for YI02 and YI03 was less extreme, from 0.03 to 0.30 and 0.02 to 0.28 g seed inflorescence–1, respectively. Thousand-seed weight exhibited only modest variation, values ranged from 0.46 to 0.85 g in 2002, and 0.51 to 0.86 in 2003.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary statistics and least significant difference values for white clover seed yield and component traits in a segregating F1 QTL mapping population tested for seed yield (SY), inflorescence density (ID), yield per inflorescence (YI), and thousand-seed weight (TSW). Cultivars 3647 and 6525 were the parents of the full-sib population used in this experiment.

 
Many trait datasets were correlated; the highest correlation was between 2003 and 2004 SY values (r = 0.82) (Table 2). For SY and its components, a high correlation was observed between ID02 and SY02 (r = 0.74) and between ID03 and SY03 (r = 0.58). Moderately positive correlations were observed between YI and SY (0.45 and 0.51 in 2002 and 2003, respectively). Weak negative correlations between ID and YI were observed in both 2002 and 2003 (–0.20 and –0.33, respectively). Thousand-seed weight was not correlated with SY or its components.


View this table:
[in this window]
[in a new window]
  		Table 2. Matrix of correlation coefficient (r) values for seed yield (3-yr data) and its component traits (2-yr data) in a full-sib white clover QTL mapping population.{dagger}

 
The ANOVA indicated significant genotype, environment, replicate, and G x E effects for SY, ID, and YI (Table 3). Analysis of TSW data indicated significant genotype and G x E effects, but no significant year effect (Table 3). Year effects were of greater magnitude than the genotype effects for SY, ID, and YI.


View this table:
[in this window]
[in a new window]
  		Table 3. Analysis of variance for seed yield (SY) and component traits data (ID, inflorescence density; TSW, thousand-seed weight; YI, yield per inflorescence) in a segregating F1 population of white clover grown in experimental plots at Lincoln, New Zealand.

 
QTL Discovery
Evaluation of the dataset using IM analysis identified 12 QTLs related to SY and its components, which were distributed across six linkage groups in the white clover genome (Table 4). Subsequent forward selection of cofactors in a MQM analysis resulted in refined placement of those 12 QTLs, and indicated 11 new QTLs on nine linkage groups in total (Fig. 1, Table 4). Using MQM, LOD scores ranged from 4.4 to 15.9, and the percentage of phenotypic variation accounted for ranged from 5.8 to 29.7%. The MQM analysis accounted for 20 to 23% of the variation for SY yield in each year, and from 18% (TSW03) up to 68% (ID02) of the variation for component traits (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 4. Description of 23 QTLs for seed yield and component traits in white clover.{dagger},{ddagger}

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 1. A map of a portion of the white clover genome, indicating the positions of 23 QTLs for seed yield and its component traits. Box-bars indicate the genome region within one LOD unit of the peak LOD score, extending lines indicate genome region with LOD scores within 2 LOD units of the peak. The QTL names are to the left of each boxed bar. Only the nine linkage groups with seed production QTLs are depicted; in this experiment no QTLs were detected on the other seven linkage groups.

 
The use of MQM instead of IM resulted in the discovery of eight additional QTLs relating to ID, three of which were detected in both years. One of these QTL, ID03_D2, accounts for >15% of the trait variation. Three additional TSW QTLs were detected using MQM analysis. No additional SY or YI QTLs were detected using MQM.

Three QTLs were discovered for SY, one in each year. In each of the first 2 yr, the SY QTL location and genotype class means correspond to the predominant QTL discovered in the component trait datasets ID_02 and SP_03, respectively. In 2004, no component trait data were available for comparison; however, the SY QTL discovered in 2004 colocates with the SY QTL discovered in the 2003 dataset.

This experiment implicated nine of the 16 white clover linkage groups in the regulation of SY and component traits. Most QTLs for a trait were observed in only one of two homoeologues, with the exception of the ID02 QTLs on linkage groups E1 and E2 (Fig. 1). Another possible case are the ID02 QTLs on C1 and C2; however, they appear to be in different regions on the homoeologues. The greatest concentration of QTL effects in the genome was at the bottom of linkage group D2, which had QTLs influencing all traits measured. In this genome region, YI and TSW genotype class means are inversely correlated in both years (r = –0.97 and –0.76, respectively); however, there was no relationship between ID and YI.

All but four of the 20 QTLs for yield components were detected in both years. The SY data revealed a QTL repeated in two of the 3 yr of data, identified as SY03_D2 and SY04_D2, respectively. Quantitative trait loci that were repeated across years exhibited generally similar magnitude, and the difference in relative performance among genotype classes also was highly conserved (Table 4). In most cases, a specific allele was responsible for genotype class mean differences (e.g., genotype classes a_ vs. b_ for QTL SP02_D2, with allele c vs. allele d having little effect), suggesting dominant gene action. For a few QTLs, one allele combination was markedly better (e.g., genotype class ad for ID02_B2 vs. ac, bc, or bd).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
This report is the first QTL discovery for SY and its component traits in a forage legume, and supports the high heritability estimates previously reported for white clover seed production traits (Annicchiarico et al., 1999; Woodfield et al., 2004). Although seed production is overshadowed by the on-farm forage role, the ability to maintain a commercially acceptable seed production level is essential to a cultivar's success.

Parent plants exhibited similar trait values for SY and its component traits (Table 1). In contrast, the population exhibited a surprisingly wide range of variation, up to 6.5, 6.5, 4.2, and 5.3 times greater than the least significant difference (P < 0.05) for SY, ID, YI, and TSW, respectively (Table 1). This wide range of phenotypic variation suggests there is substantial genetic variation within and between parents for SY and that the population was well suited for QTL discovery. Although it is considered best practice to utilize parent plants with divergent genetic potential and divergent phenotypes to create QTL discovery populations, these results indicate that in allogamous species such as white clover that parent plants with similar trait values may be heterozygous at key loci and give rise to useful QTL discovery populations. These observations may be biased by the age and disease status of the parent material, which may have suppressed the observed performance substantially below their genetic potential relative to the young progeny clonal material in this experiment.

The ANOVA indicated a significant proportion of the variation was accounted for by year and G x E sources. However, most QTLs with a substantial impact on phenotype were detected in both years and exhibited similar magnitude and rankings among genotype class means within a QTL. The genotype component of G x E effects in the ANOVA is attributed to several sources, including (i) the few QTLs detected in only one environment, (ii) genetic effects that were not amenable to detection in this experiment due to limits on population size and limited replication in the field trial, (iii) the potential that the white clover reference map used does not provide complete genome coverage, and (iv) limitations of the QTL analysis in testing for epistatic effects.

The generally consistent QTL results suggest that despite white clover's reputation for phenotypic plasticity, there are genome locations which consistently affect a high heritability suite of seed production characteristics. Detection of a QTL across environments (site or years) is the first validation step for assessing robustness. The detection of most SY component QTLs in both years sets the stage for further validation of SY and component trait QTLs in independent pair crosses and in multiparent breeding populations.

Most QTLs indicated dominant rather than additive gene action. This is important for practical applications to the improvement of open-pollinated crops such as white clover. Dominant or nearly dominant gene action is well suited for MAS, as fixing desirable dominant genetic effects in open-pollinated species is exceedingly difficult by phenotypic selection alone. However, it is necessary to test these superior alleles in other genetic backgrounds to confirm their mode of gene action and to assess potential value to white clover improvement.

The observation of negative correlation between YI and TSW in linkage group D2 suggests either one QTL with pleiotropic effects, or multiple QTLs tightly linked in repulsion phase. The negative relationship between these two components also suggests that this region of the genome may influence floret number or ovule fertility. Linkage group D2 exhibited the greatest level of segregation distortion in the pair cross (Barrett et al., 2004), again suggesting that genetic factors which influence fertility may exist on this linkage group.

Detection of superior alleles in inferior plants has been reported for inbred species such as rice (Oryza sativa L.) (Xiao et al., 1998), demonstrating the power of QTL experiments relative to strictly field-based analyses for deciphering genetic control of a trait. In the current study, parent 3647 was inferior to parent 6525 for YI; however, the major genetic effects for YI were identified in parent 3647 (QTLs YI02_D2 and YI03_D2). The discovery of superior alleles in inferior plants may be more common in allogamous species than in self-fertile species.

Exploration of germplasm with inferior phenotypes has proven useful for mining novel and superior alleles from accessions in forage germplasm banks, as superior alleles may go undetected using only phenotypic analyses (Tanksley and McCouch, 1997). However, interpretation of parent phenotypes in the context of full-sib designs using highly heterozygous parents is complicated, especially when compounded by the inability to accurately phenotype old parent plants relative to younger progeny plants. The discovery of major QTLs for ID (ID02_C2 and ID03_C2) given the similar parent values for ID suggests that parent phenotype of heterozygous plants is not indicative of variation potential among F1 progeny in outbred species.

Most QTL effects detected were specific to one homoeologue of white clover. This situation contrasts to other allopolyploids such as cotton, which has SY QTL in homoeologous regions of both subgenomes (Saranga et al., 2004). It is possible the white clover population parents were homozygous for some QTL in the other homoeologue, or that only one homoeologue in the white clover genome is typically involved in the control of some traits. Additional investigation is required to determine if white clover has a general tendency for only one subgenome to influence a particular trait. Definition of subgenome composition and ancestry in white clover will empower this approach.

These QTLs in the white clover genome may provide a springboard for comparative mapping and discovery of seed production QTLs in other legume species. Alfalfa (Medicago sativa L.) has high heritability for SY components (Bolanos-Aguilar et al., 2001) and a genome organization that is broadly conserved within the Papilionoideae (Kalo et al., 2004). The recent demonstration of syntenic relations among a wide array of legume species (Choi et al., 2004) also indicates potential for comparative QTL analysis for seed production traits among forages, pulse crops, and soybean [Glycine max (L.) Merr.].

Quantitative trait locus discoveries are a beginning point for implementation of MAS schemes and/or for cloning of the sequence conferring the genetic effect. The first requirement in either case is for validation of QTL effects in independent or related populations. One population structure that may prove useful is advanced intercross lines (AILs) (Darvasi and Soller, 1995), such as those reported in maize (Zea mays L.) (Dudley et al., 2004). In the case of outbred species such as white clover, AILs can be easily generated using insect pollinators. By capturing additional cycles of recombination, AILs have utility for identifying any spurious QTL effects observed in the F1, and for further refining genuine marker–QTL associations. Advanced intercross lines also provide an opportunity for testing of homozygous genotype classes (aa, bb, cc, and dd) not observed in the full-sib design currently used, which may reveal other QTLs which were fixed in one or both parents and therefore not detected in the F1.

Marker-assisted selection in outbred herbaceous plant species offers a unique array of challenges not found in major self-fertile food crops such as maize, rice, wheat (Triticum aestivum L.), and soybean. A primary challenge is the large numbers of marker and QTL alleles and their unknown allelic phase relationships in breeding populations. Considering the density of the current map, it is unlikely that direct application into MAS schemes for white clover breeding programs is economic, given the potential for highly effective phenotypic selection for seed production traits.

Focus on other economically meaningful traits and combining lab-based genome mapping resources with bioinformatic approaches to gene cloning will provide the opportunity to develop functional markers such as the quantitative trait nucleotide reported for tomato (Lycopersicon esculentum Mill.) quality (Fridman et al., 2004). Functional markers may be more suited for MAS in plant breeding programs for open-pollinated forage species such as white clover. This study provides a robust demonstration of QTL discovery in Trifolium, and indicates a useful approach for further work in forage breeding systems integrating MAS to develop forages which deliver favorable economic and environmental performance profiles for the pastoral sector.

Received for publication November 25, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2005 45: x. [Full Text]  



This article has been cited by other articles:


Home page
 Crop Sci.Home page
D. Herrmann, B. Boller, B. Studer, F. Widmer, and R. Kolliker
Improving Persistence in Red Clover: Insights from QTL Analysis and Comparative Phenotypic Evaluation
Crop Sci., January 16, 2008; 48(1): 269 - 277.
[Abstract] [Full Text] [PDF]

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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Agricola
Right arrow Articles by Barrett, B. A.
Right arrow Articles by Woodfield, D. R.
Related Collections
Right arrow Crop Genetics
Right arrow Clover

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