Religion & Spirituality

9 pages

Physiological and molecular bases of glyphosate resistance in Conyza bonariensis biotypes from Spain

of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
The physiological and molecular bases of glyphosate resistance in one susceptible (S) and four resistant (R) Conyza bonariensis biotypes (sampled in orchards from Andalusia, Spain) were investigated. Resistance index (RI) values of the four R
  Physiological and molecular bases of glyphosateresistance in  Conyza bonariensis   biotypes from Spain G DINELLI*, I MAROTTI*, A BONETTI*, P CATIZONE*, J M URBANO  & J BARNES  * Dipartimento di Scienze e Tecnologie Agroambientali, Universita ` di Bologna, Bologna, Italy,   Department of C. Agroforestales, Universidad de Sevilla, Sevilla, Spain, and    Syngenta Crop Protection, Basel, Switzerland  Received 7 May 2007Revised version accepted 2 January 2008 Summary The physiological and molecular bases of glyphosateresistance in one susceptible (S) and four resistant (R) Conyza bonariensis  biotypes (sampled in orchards fromAndalusia, Spain) were investigated. Resistance index(RI) values of the four R biotypes ranged between 2.9and 5.6. The main physiological difference between Sand R biotypes was the dissimilar mobility of glyphosatein the whole plant. In R biotypes, the herbicide wastranslocated less from leaves to culm and root, and morefrom culm to leaves compared with the S biotype. Theupward mobility of glyphosate via xylem suggests thatthe herbicide may be sequestered to the apoplast or thevacuole. The hypothesis of an insensitive 5-enolpyruv-ylshikimate-3-phosphate synthase (EPSPS) was provi-sionally discarded on the basis of shikimateaccumulation in R plant tissues after glyphosate treat-ment. At the molecular level, the relative abundance of EPSPS mRNA prior to glyphosate treatment wasapproximately double in two R biotypes compared withthe S standard and the other R biotypes. Moreover, thetwo R biotypes having both no translocation anddoubled EPSPS mRNA levels had also the highest RI.These results suggest that two factors may be related toglyphosate resistance in the R biotypes: (i) impairedtranslocation and (ii) high basal EPSPS transcript levels.The comparison between these findings and earlierresults on glyphosate resistance mechanism in  Conyzacanadensis  biotypes from the USA, suggests that similaragronomic factors (repeated application of glyphosate,no crop and herbicide rotation, no tillage) have selectedsimilar traits on different genetic pools of the resistance-prone  Conyza  genus. Keywords:  hairy fleabane, glyphosate, resistance mech-anism, uptake, metabolism, translocation, 5-enolpyruv-ylshikimate-3-phosphate synthase mRNA levels.D INELLI  G, M AROTTI  I, B ONETTI  A, C ATIZONE  P, U RBANO  JM & B ARNES  J (2008). Physiological and molecular basesof glyphosate resistance in  Conyza bonariensis  biotypes from Spain.  Weed Research  48 , 257–265. Introduction Over the last decade there has been a dramatic increasein the appearance of herbicide-resistant weed biotypesand currently herbicide resistant populations are spreadover the five continents. Resistance problems alsoinvolve glyphosate, the most widely used herbicide inthe world. For a long time, glyphosate resistance wasnot expected, as a result of the absence of evolvedglyphosate resistance in weed species (Bradshaw  et al. ,1997). However, in the last decade glyphosate resistancehas been confirmed for 12 weed species [ Amaranthus palmeri   S. Wats. , Amaranthus tuberculatus  (Moq.)Sauer , Ambrosia artemisifolia  L. , Ambrosia trifida  L .,Conyza bonariensis  (L.) Cronq ., Conyza canadensis  (L.)Cronq. , Echinochloa colona  (L.) , Eleusine indica (L.) Gaertn ., Lolium multiflorum  L. , Lolium rigidum Gaud. , Plantago lanceolata  L. , Sorghum halepense  (L.)Pers.] over a wide range of geographical locations(Argentina, Australia, Brazil, Chile, China, Colombia,France, Malaysia, South Africa, Spain, USA) (Heap,2007). Herbicide resistance is generally thought to occurwithin weed populations as a consequence of the intenseselective pressure exerted by a lack of diversity in weed Correspondence : Giovanni Dinelli, Dipartimento di Scienze e Tecnologie Agroambientali, Universita ` di Bologna, Viale Fanin 44, Bologna I-40127,Italy. Tel: (+39) 051 2096672; Fax: (+39) 051 2096241; E-mail:   2008 The AuthorsJournal Compilation    2008 European Weed Research Society  Weed Research  48,  257–265  management practices (Gressel & Segel, 1978). It is notsurprising that the majority of the 44 glyphosate-resistant biotypes have been selected in glyphosate-resistant crops or in perennial crops (Heap, 2007).Glyphosate-resistant crops and perennial crops sharesome common features: high levels of glyphosate usage,repeated applications during a growing season, adoptionof no tillage practices and lack of crop and herbiciderotation. The combination of these factors resulted inintense pressure for selecting herbicide resistant weeds.The genus  Conyza  (Asteraceae) is evidently prone inevolving resistance to glyphosate. Horseweed ( C. canad-ensis ) is a winter or summer annual North Americannative weed of importance in no-tillage crop productionsystems (Buhler & Owen, 1997).  Conyza canadensis  isconsidered one of the 10 most important herbicide-resistant weeds, evolving resistance to photosystem IIinhibitors, bipyridiliums, ureas, amides, ALS inhibitorsand glyphosate (Heap, 2007). The first reported occur-rence of glyphosate-resistant  C. canadensis  was inDelaware (USA) in 2000 (VanGessel, 2001). Six yearsafter its first occurrence, it was found on more than half-million hectares across 13 states of the USA Midwest,South and Atlantic states (Heap, 2007). Recently,resistant  C. canadensis  biotypes were recorded in Chinaand Brazil (Heap, 2007). At present  C. canadensis  is themost widespread glyphosate-resistant weed. The rela-tively abundant seed production and wind disseminationgreatly facilitate the rapid dispersal of resistant C. canadensis  plants across an agricultural landscape(Buhler & Owen, 1997; Weaver, 2001). On the basis of recent investigations on glyphosate-resistant  C. canad-ensis  biotypes from the USA, the principal resistancemechanism is the impaired translocation of the herbicideconsisting in a reduced transport towards meristematictissues and an increased accumulation in leaves (Feng et al. , 2004; Koger & Reddy, 2005; Dinelli  et al. , 2006).Additionally, the complementary role of constitutivelyhigher 5-enolpyruvylshikimate-3-phosphate synthase(EPSPS) transcript levels and enhanced ramificationwere postulated for four glyphosate-resistant  C. canad-ensis  biotypes (Dinelli  et al. , 2006).Hairy fleabane ( C. bonariensis ) is an annual or short-lived perennial weed originating in South America(Michael, 1977). As with  C. canadensis ,  C. bonariensis is predominately a summer or winter weed in no-tillagecropping systems. Reduced seed reserves and lightrequirements for germination limit seedling emergenceto 1–2 cm from the soil surface.  Conyza bonariensis evolved resistance against ALS inhibitors, bipyridiliums,photosystem II inhibitors and glycines. The firstreported cases of glyphosate-resistant  C. bonariensis were in South African orchards and vineyards in 2003,while in 2004 and 2005 resistant biotypes were recordedin Spanish and Brazilian orchards (Heap, 2007). Thefour  C. bonariensis  biotypes from southern Spain,characterised by a glyphosate resistance index (RI)ranging between 3.5 and 10.5, represent the first caseof glyphosate resistance in Europe (Urbano  et al. , 2007).Studies conducted in Spain from 1996 to 2002 showed aprogressive decrease of glyphosate efficacy against Conyza  spp. (Gil-Albarellos, pers. comm.). Outcrossinghas been documented with  Conyza  spp., increasing thepotential for spreading resistance genes among andwithin populations (VanGessel, 2001).The glyphosate resistance mechanism in  C. bonarien-sis  is still unknown to date, as no literature is availableon this topic. The present paper deals with the identi-fication of physiological and molecular bases of glypho-sate resistance in four Spanish  C. bonariensis  biotypes bydetermining: (i) the RI based on plant survival; (ii) theamount of endogenous shikimate accumulated at theshoot level after glyphosate treatment; (iii) the absorp-tion, metabolism and translocation of glyphosateand (iv) the constitutive expression levels of EPSPStranscripts. Materials and methods Plant material and growth conditions  In 2004, susceptible (04-033, S) and resistant (04-019,R1; 04-050, R2; 04-015, R3; 04-038, R4)  C. bonariensis biotypes were collected from orchards located in theprovinces of Seville, Huelva and Cordoba (Spain). S andR biotypes were originally characterised by Urbano et al. ( 2007). The S biotype was sampled in a peach[ Prunus persica  (L.) Bat.] orchard (no-tillage, dripirrigation) and was never treated with glyphosate inthe 5 years before the sampling. In the peach orchard,weed control was based on the application of glufosinateand oxyfluorfen. The R biotypes, collected in olive ( Oleaeuropea  L.) groves (no-tillage, drip irrigation), wererepeatedly treated with glyphosate (at least one appli-cation per year for a period of 4–6 years). In the 5 yearsbefore the sampling in olive groves, other herbicideswere applied for weed control: oxyfluorfen (R1, R3, R4),amitrol (R2), diuron and MCPA (R3). In the four olivegroves, farmers complained of the lack of glyphosateactivity against  C. bonariensis , which was the mostabundant and troublesome weed species. A variety of glyphosate resistant (GR) soyabean ( Glycine max  L.)was also included in the experiments.For all five  C. bonariensis  biotypes and the GRsoyabean, seeds were germinated in Petri dishes withmoist filter paper under controlled environment condi-tions (25  C, 12-h photoperiod, 250  l mol photonsm ) 2 s ) 1 ). Seedlings in the cotyledon growth stage were 258  G Dinelli  et al.   2008 The AuthorsJournal Compilation    2008 European Weed Research Society  Weed Research  48,  257–265  singly transplanted into pots (3 cm radius; 350 mLvolume) containing a 1:1 ( V   ⁄   V) peat:sand sterile pottingmix. Plants were placed in a growth chamber set at 28  Cand 70% relative humidity (RH) day and 22  C and 50%RH night conditions; light was supplemented to a 12-hphotoperiod with artificial illumination at 550  l molphotons m ) 2 s ) 1 . Plants were sub-irrigated as needed. Dose–response assay  Dose–response curves of   C. bonariensis  biotypes weredetermined at the rosette stage (diameter of   c.  8 cm;18–22 leaves). Plants were selected within each biotypeto match in size and growth stage and sprayed withglyphosate at the doses of 0, 180, 360, 740, 1480and 3000 g ae ha ) 1 (Roundup Bioflow, 360 g ae L ) 1 ,Monsanto). The sprayer was equipped with a flat-fannozzle at a height of 50 cm with an output volumeequivalent to 185 L ha ) 1 . The experimental design fordose–response tests was a randomised complete blockwith five biotypes, six herbicide rates and three repli-cates of 25 plants. Plants were assessed 28 days aftertreatment (DAT) and were scored as dead or alive.Non-linear regression analysis and  ANOVA  were used todetermine the effect of glyphosate dose on plantsurvival of each  C. bonariensis  biotype. A sigmoidallog–logistic model (Seber & Wild, 1989) was used torelate number of live plants as a percentage of theuntreated check ( Y  ) to glyphosate dose ( x ) according tothe following formula: Y    ¼ a = 1 þ e ð  X    LD 50 Þ = b ð 1 Þ In this equation,  a  is the difference of the upper andlower response limits (asymptotes), LD 50  is the herbicidedose required to kill 50% of individuals and  b  is theslope of the curve around LD 50.  The RI was determinedby dividing the LD 50  value of each R biotype by theLD 50  of the S biotype. The LD 50  and RI values wereseparated by Fisher  s least significant difference (LSD)( P  = 0.05). Shikimic acid accumulation  The accumulation of shikimic acid in green tissues wasinvestigated according to the procedure described byDinelli  et al.  (2006). The experiment was designed as a2  ·  3  ·  6 factorial, replicated five times, with glyphosatedose (0, 360 g ae ha ) 1 ), time of plant harvesting (3, 7and 10 DAT) and plant accession (S, R1, R2, R3, R4and GR soyabean) as factors.  Conyza bonariensis  plantsat the rosette stage (diameter of   c.  8 cm; 18–22 leaves)and soyabean plants at 4–6 leaf stage were sprayed aspreviously described (section  Dose–response assay ).Shikimate extraction and analysis were carried outaccording to the method proposed by Mueller  et al. (2003) and modified by Dinelli  et al.  (2006). Data wereanalysedforcompletelyrandomiseddesignandthewholeexperiment was repeated twice. As the effect of repetitionover time was not significant, data of the two runs werecombined. Values were reported as mean concentrations( l g shikimate g ) 1 shoot fresh weight) ± SE. [  14  C]-glyphosate dose and application  Glyphosate-(phosphonomethyl- 14 C) (Syngenta, Basel,Switzerland) solution was prepared and applied, asreported in Dinelli  et al.  (2006). The final concentrationof the radiolabel solution (0.5 kBq  l L ) 1 ) was1.5  l g ae  l L ) 1 . Two 1  l L of droplets were applied witha micro-applicator (Dispenser PB 6000) to five leaves of  C. bonariensis  plants at the rosette stage (diameter of  c.  8 cm; 18–22 leaves) and GR soyabean plants at 6–8leaf stage. For absorption and metabolism studies, two1  l L of droplets were applied in the mid of adaxial leaf surface on the opposite sites of the major vein. Fortranslocation study in the downward direction two 1  l Lof droplets were applied on the adaxial leaf surface(opposite sites of the major vein) at  c.  0.5 cm from thetop of the leaf. For translocation study in the upwarddirection two 1  l L of droplets were applied on the baseof the leaf petiole (i.e. insertion of the leaf on the culm).Each plant treated with [ 14 C]-glyphosate received 15  l gae, which is approximately equivalent to a rate of 350 g ae ha ) 1 at 187 L ha ) 1 spray volume, according tothe estimation reported by Feng  et al.  (2004). [  14  C]-glyphosate absorption and metabolism Glyphosate absorption and metabolism in  C. bonariensis biotypes was determined according to the procedurereported by Dinelli  et al.  (2006). Thirty plants perbiotype were treated. At 3, 7 and 10 DAT, 10 plants perbiotype were harvested. After washing leaf surfaces withmethanol–water (1:9,  V   ⁄   V), unabsorbed radioactivitywas quantified by liquid scintillation spectroscopy (LSS)(1409 Liquid Scintillation Analyzer; Wallac). Plantswere then dissected into leaves, culm (crown meristemsand immature leaves) and roots. The different plantparts were weighed, frozen in liquid nitrogen, powderedand extracted with ultra pure water (1:4 g freshweight mL ) 1 ). After centrifugation (15 000  g , 10 min),supernatant radioactivity was determined by LSS. Plantdebris contained in the centrifugation pellet were driedand combusted in a biological oxidiser (Packard 387).The unextracted radioactivity was then quantified byLSS. Glyphosate and its main metabolite, namelyaminomethylphosphonic acid (AMPA), were separated Glyphosate resistance mechanisms of hairy fleabane  259   2008 The AuthorsJournal Compilation    2008 European Weed Research Society  Weed Research  48,  257–265  by thin layer chromatography (TLC) plates (SG60 withfluorescent marker; Merck). Electronic auto-radiogra-phy and image analysis of TLC plates were performedusing a Molecular Imager (Bio-RadZ). Glyphosate andAMPA were identified by comparing their  R f   valueswith those of authentic commercial standards. Percent-age data were arcsine square-root transformed to respectthe assumption of variance homogeneity and analysedas a completely randomised design. Mean values wereexpressed as percentages of absorbed radioactiv-ity ± SE. Considering all biotypes and sampling times,the mean radioactivity recovery was 90 ± 5% of theapplied dose, while the mean unextracted radioactivitywas 4 ± 2% of the applied dose. No significant differ-ence in recovered and unextracted radioactivity wasobserved between S and R biotypes. [  14  C]-glyphosate translocation in whole plants  Herbicide translocation was investigated in two differ-ent experiments, hereinafter called   translocation in thedownward direction   and   translocation in the upwarddirection  , according to the procedures reported byDinelli  et al.  (2006). Except the different application of the radiolabel (see section   [ 14 C]-glyphosate dose and application  ; adaxial leaf surface and leaf petiole appli-cation for downward and upward translocation, respec-tively), the two experiments were identical. For eachexperiment 30 plants per accession were treated. At7 DAT, three replicates of 10 plants per biotype wereharvested. Plants were then dissected into treatedleaves, untreated leaves, culm (crown meristems andimmature leaves) and roots. Unabsorbed, total extract-able and unextracted  14 C in the different plant partswere determined by LSS, as previously described(section   [ 14 C]-glyphosate absorption and metabolism  ).The different plant parts were placed on a glass supportand covered with a fine plastic film. Electronic autora-diography and image analysis were performed aspreviously described (section   [ 14 C]-glyphosate absorp-tion and metabolism  ). Percentage data were arcsinesquare-root transformed to respect the assumption of variance homogeneity and analysed as a completelyrandomised design. Mean values were expressed aspercentages of absorbed radioactivity ± SE. Eachexperiment was repeated twice. As the effect of repe-tition over time was not significant, data of the tworuns were combined. Considering all biotypes andsampling times, the mean radioactivity recovery was92 ± 6% of the applied dose, while the mean unex-tracted radioactivity was 3 ± 1% of the applieddose. No significant difference in recovered and unex-tracted radioactivity was observed between S and R C. bonariensis  biotypes. Semiquantitative RT-PCR analysis of EPSPS transcripts  Leaf samples of each  C. bonariensis  biotype werecollected from four replicates of 10 untreated plants atthe rosette stage (diameter of   c.  8 cm; 18–22 leaves).Total RNA was isolated from leaf samples and com-plementary DNA (cDNA) synthesised according to themethod reported in Dinelli  et al.  (2006). EPSPS geneexpression determinations were conducted by the multi-plex titration RT-PCR procedure described in Neb-enfuhr and Lomax (1998) with minor modifications(Dinelli  et al. , 2006). For PCR amplification of eachbulk cDNA sample, a series of ten 0.5-fold serialdilutions was constructed. The first sample in eachseries contained a constant amount (20 ng) of cDNA.Primers used were synthesised by Genenco (M-Medical)and the sequence was determined using the software PRIMER  3 (developed by Rozen & Skaletsky, 2000)available online at The design of EPSP primerswas based on the published 1766 bp mRNA sequencefor EPSPS2 of   C. canadensis  (GenBank accessionnumber AY545667) while for the internal control (ahousekeeping gene, translation elongation factor 1- a ,EF1- a ) primers construction was based on the EF1- a gene sequence (GenBank accession number AK221176)of   Arabidopsis thaliana.  The following primers wereused: EPSPS2-F 5 ¢ -GTTGCGGGACAAGCA-3 ¢  (Tm50.6  C) and EPSPS2-R 5 ¢ -AGGGCAACCACAGCAA-3 ¢  (Tm 51.8  C); EF1 a -F 5 ¢ -TTAAGGCCGAGCGTG-3 ¢ (Tm 50.7) and EF1 a -R 5 ¢ -CGAAGGGGCTTGTCT-GA-3 ¢  (Tm 54.8). The Tm s  indicated in brackets werecalculated according to the formula reported by Sam-brook  et al.  (1989). Two microlitre of cDNA wereamplified with 200  l M  deoxyribonucleotide triphos-phate, 5.6 m M  MgCl 2 , 0.4 and 0.3  l M  of each EPSPand EF primers, respectively, 1 unit of   Taq  polymerase(M-Medical). Amplification was as follows: 95  C for10 min; 35 cycles at 95  C for 40 s, 59.2  C for 40 s and72  C for 40 s; and a final extension at 72  C for 5 min.Reactions were carried out in a T-Gradient ThermalCycler (Biometra). RT-PCR products were separated on1.5% agarose gels, scanned and saved in tagged infor-mation file format. EPSPS2 and EF1 a  primers yieldedamplification products of 180 and 500 bp respectively.Both PCR fragments were sequenced (Genenco;M-Medical) and the comparison of their nucleotidecomposition, by using the NCBI GenBank BLASTprograms, confirmed that the 180 and the 500 bpfragment contained part of the selected EPSPS andhousekeeping genes respectively. Furthermore, the180 bp fragment showed a high matching degree withpart of the mRNA sequences of EPSPS1 (GenBank 260  G Dinelli  et al.   2008 The AuthorsJournal Compilation    2008 European Weed Research Society  Weed Research  48,  257–265  accession number AY545666) and EPSPS3 (GenBankaccession number AY545668) of   Conyza  spp .  For eachR accession the relative transcript level of EPSPS gene(EPSPS mRNA abundance of R biotype divided byEPSPS mRNA abundance of S biotype) was calculated.Percentage data were arcsine square-root transformed torespect the assumption of variance homogeneity andanalysed for a completely randomised design. Meanvalues were separated by Fisher  s LSD ( P  = 0.05) andreported as percentage of relative transcript levels of EPSPS gene ± SE. Results Dose–response assay  The response of each  C. bonariensis  biotype toincreasing glyphosate dose was best fit to a sigmoidallog–logistic model, with regression coefficients rangingbetween 0.92 and 0.97. LD 50  values for the R biotypeswere higher than that for the susceptible standard,indicating that the four biotypes were resistant toglyphosate (Fig. 1). The R biotypes were characterisedby different resistance levels. The RI for R1 and R2biotypes was  c.  3, whereas the RI for R3 and R4biotypes was 4.4 and 5.6 respectively. No individuals of susceptible biotype survived at doses >740 g ae ha ) 1 .On the basis of sigmoidal log–logistic equations, at therecommended field dosage rate (1080 g glypho-sate ha ) 1 ) for the control of   C. bonariensis ,  c.  19%,20%, 29% and 37% of R1, R2, R3 and R4 plants werenot killed. Shikimic acid accumulation  Shikimate accumulated in concentrations significantlygreater than background levels after glyphosate treat-ment in all  C. bonariensis  biotypes (Fig. 2). Three DATno significant differences ( P  = 0.05) in shikimate levelsbetween S and R biotypes were observed and, across allbiotypes, the increase of shikimic acid in green tissuesranged between 6 and 10 times with respect to untreatedcontrols. The 7 and 10 DAT the concentration of shikimate in the susceptible standard was significantly( P  < 0.05) higher than that observed in R biotypes.There were no significant differences ( P  = 0.05) inshikimate levels among R biotypes 7 and 10 DAT. Conyza bonariensis  biotypes differed in the trend overthe time in shikimate concentration; it remained con-stant from 7 to 10 DAT in R plants but increased from 7to 10 DAT in S plants. The 7 and 10 DAT in R biotypes,the level of shikimate was respectively 13–16 and 9–14times higher than untreated controls, while for the Sbiotype it was respectively 37 and 47 times higher thanuntreated controls. After glyphosate treatment (3, 7 and10 DAT), only a slight increase of shikimate concentra-tion ( c.  two times with respect to untreated controls) wasobserved in the green tissues of GR soyabean, notablycharacterised by a glyphosate resistant EPSPS. [  14  C]-glyphosate absorption and metabolism No significant differences ( P  = 0.05) were observed inthe leaf uptake of [ 14 C]-glyphosate between S and Rbiotypes. Additionally, no significant change in the Glyphosate dose (g ae ha –1 ) 10 100 1000    S  u  r  v   i  v  a   l   (   %   o   f  u  n   t  r  e  a   t  e   d  c  o  n   t  r  o   l   ) 0 20 40 60 80 100 S 154 1.0 (c) R1 461 2.9 (b) R2 482 3.1 (b) R3 686 4.4 (ab) R4 868 5.6 (a) LD 50 RI Fig. 1  Dose–response assay of glyphosate-resistant (R1, R2, R3,R4) and susceptible (S)  Conyza bonariensis  biotypes treated at therosette stage with different glyphosate doses. Lines describe thepredicted survival responses according to the equation reported inthe Materials and methods section. Data are mean ± SE. In theinset, for each biotype the LD 50  value (g ae ha ) 1 ), resistance index(RI) and statistical significance (within brackets different lettersrepresent values significantly different at  P  = 0.05 based onFisher  s LSD) are reported. Days after treatment 0246810   µ  g  s   h   i   k   i  m  a   t  e  g   –   1     f  r  e  s   h  w  e   i  g   h   t 050010001500200025003000S R1R2R3R4Soybean (GR) Fig. 2  Shikimate content ( l g shikimate g ) 1 plant fresh weight) inshoot of   Conyza bonariensis  biotypes (rosette stage) and glyphosateresistant (GR) soyabean (4–6 true leaves) after the treatment with360 g glyphosate ha ) 1 . Data are mean ± SE.Glyphosate resistance mechanisms of hairy fleabane  261   2008 The AuthorsJournal Compilation    2008 European Weed Research Society  Weed Research  48,  257–265
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks