Phenotypic diversity analysis of grain yield and yellow pigment content in germplasm collected from Iranian durum wheat (Triticum turgidum L.) landraces

This article was downloaded by: [Farzad Fayaz]On: 08 October 2013, At: 11:02Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Archives of Agronomy and Soil SciencePublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/gags20

Phenotypic diversity analysis of grainyield and yellow pigment content ingermplasm collected from Iraniandurum wheat (Triticum turgidum L.)landracesFarzad Fayaz a , Mohsen Mardi b , Mostafa Aghaee c , FarrokhDarvish a & Reza Talebi da Department of Plant Breeding, Science and Research Branch ,Islamic Azad University , Tehran , Iranb Department of Genomics , Agricultural Biotechnology ResearchInstitute of Iran (ABRII) , Karaj , Iranc Seed and Plant Improvement Institute (SPII) , Karaj , Irand College of Agriculture, Sanandaj Branch , Islamic AzadUniversity , Sanandaj , IranAccepted author version posted online: 10 Jul 2012.Publishedonline: 25 Sep 2012.

To cite this article: Farzad Fayaz , Mohsen Mardi , Mostafa Aghaee , Farrokh Darvish & Reza Talebi(2013) Phenotypic diversity analysis of grain yield and yellow pigment content in germplasmcollected from Iranian durum wheat (Triticum turgidum L.) landraces, Archives of Agronomy andSoil Science, 59:10, 1339-1357, DOI: 10.1080/03650340.2012.708927

To link to this article: http://dx.doi.org/10.1080/03650340.2012.708927

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or

howsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Phenotypic diversity analysis of grain yield and yellow pigment

content in germplasm collected from Iranian durum wheat

(Triticum turgidum L.) landraces

Farzad Fayaza*, Mohsen Mardib, Mostafa Aghaeec, Farrokh Darvisha andReza Talebid

aDepartment of Plant Breeding, Science and Research Branch, Islamic Azad University, Tehran,Iran; bDepartment of Genomics, Agricultural Biotechnology Research Institute of Iran (ABRII),Karaj, Iran; cSeed and Plant Improvement Institute (SPII), Karaj, Iran; dCollege of Agriculture,

Sanandaj Branch, Islamic Azad University, Sanandaj, Iran

(Received 11 May 2012; final version received 2 July 2012)

This study was undertaken to compare patterns of agronomic characters andyellow pigment content diversity between Iranian durum landraces and modernvarieties. In four field experiments, a collection of 127 accessions of landraces and23 modern cultivars tetraploid wheat representing Iranian and global geneticdiversity were tested under optimum and rain-fed conditions. Across theenvironments, the yellow pigment concentrations ranged from 1.85 mg kg71 to8.95 mg kg71 in the field-grown samples. Multivariate analysis detected fivegroups, four including landraces and one comprising modern cultivars. As agroup, modern cultivars were the most productive and showed high mean valuesfor harvest index, grains per spike and yellow pigment, but they had the lowestplant height. Landraces had the highest mean number of spikes per m2 and 1000-kernel weight, but were characterized by low plant height and yellow pigmentcontent. Clustering analysis also showed that, based on yield and yellow pigmentcontent, the accessions in different groups were of different origins, suggestingthat there was no clear relationship between accessions and geographicaldiversity. According to our results, Iranian landraces can be particularly usefulas germplasm in breeding programmes to improve spikes per m2 and 1000-kernelweights.

Keywords: durum wheat; diversity; landrace; mixed model; yellow pigment

Introduction

Durum wheat (Triticum turgidum L. var. durum Desf.) has been long grown in dryand irrigated regions of western Iran. Recent growth in the pasta industry and anincreased market demand for durum wheat in the country have led to increasinginterest in its breeding for grain yield and pasta cooking quality (Arzani 2002).

Any future improvement in durum wheat depends on the availability andidentification of genetic resources, especially in a center of origin/diversity such asthe Fertile Crescent. Iran constitutes a sizeable part of this center, where the firstcereals were domesticated by ancient civilizations *8000–10,000 years ago in the

*Corresponding author. Email: [email protected]

© 2013 Taylor & Francis

Archives of Agronomy and Soil Science, 2013http://dx.doi.org/10.1080/03650340.2012.–Vol. 59, No. , 1339 1357, 70810 927

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

late Mesolithic period and early Neolithic, and where durum wheat has beencultivated since the first millennium BC (Hawkes 1983; Harlan 1986).

Landraces or traditional cultivars are locally adapted diverse populations, whichare the result of natural selection and farmers’ cultivation methods. They havecontributed with genetic material to many breeding programmes and constituteimportant plant genetic resources. Landraces have attracted scientific interestbecause of the presence of genetic variability in well-adapted backgrounds. Despitethe great advantages of using local-adapted germplasm, the diversity of durumwheat landraces is seriously threatened by genetic erosion due to its replacement withnew, exogenous, high-yielding genetically uniform genotypes (Rashed et al. 2007).Landraces of wheat generally tolerant to biotic and abiotic stress have been grownunder low-input or sustainable farming conditions where they produce reasonableyields. A landrace, being composed a mixture of homozygous genotypes, usuallyexhibits considerable genetic variation for developmental, qualitative and quantita-tive characters (Moghaddam et al. 1997).

Local landraces may provide new alleles for the improvement of commerciallyvaluable traits. Introgression of these alleles into modern cultivars can be very useful,especially in breeding for suboptimal environments (Ceccarelli et al. 1991). Breedingactivaties carried out in durum wheat during the 20th century have resulted inincreased grain yields (De Vita et al. 2007; Royo et al. 2008), earliness, shorter plants(Royo et al. 2007) and improved biomass partitioning to the grain (Alvaro et al.2008).

Wheat is a major crop and, in addition to being an important source ofcarbohydrates in the human diet, its whole grain has been evaluated as a source ofprotective secondary plant compounds such as carotenoids (Leenhardt et al. 2006).Yellow pigment concentration (YPC) is a desirable end-use quality trait in durumwheat both visually and nutritionally, and an elevated pigment concentration hasbeen the target of durum breeding programmes worldwide (Reimer et al. 2008).Carotenoid pigments, mainly xanthophyll, present in durum wheat endosperm arethe color source. The amount of pigment varies between cultivars and fluctuatesyearly depending on environmental conditions during kernel development. Theconcentration of carotenoids is higher in durum wheat (0.5–12.13 mg kg71) than inbread wheat (0.5–2 mg kg71). Overall, modern cultivars showed significantly highervalues of YPC than old cultivars and wild accessions (Digesu et al. 2009). Braatenet al. (1962) reported the heritabilities of pigment concentration to be 0.79–0.94, withlittle evidence of a genotype–environmental interaction. Elouafi et al. (2001) found aheritability range of 0.48–0.99 for a durum by Triticum dicoccoides cross grown in 16environments.

The effective use of genetic resources in breeding programmes requires evaluationof the diversity in the gene pool and characterization of available accessions in orderto detect the presence of variants of potential interest for breeding purposes.Frequently, studies of the variation present in germplasm collections have beencarried out using the characterization of plant morphological attributes for whiteclover (Caradus et al. 1990; Jahufer et al. 1997), alfalfa (Rumbaugh et al. 1988),wheat (Pecetti et al. 1992), durum wheat (Annicchiarico et al. 2000), safflower(Jaradat and Shahid 2006) and watermelon (Szamosi et al. 2009).

Together with the important role of conserving genetic resources, gene banks alsoprovide accessions of incorporation into plant breeding programmes for thedevelopment of new cultivars of crop and pasture species (Shands and Wiesner

F. Fayaz et al.1340

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

1992). The Iran National Gene Bank holds a unique set of durum wheat landraces(700 accessions) and wild relatives collected mainly within the country. This studywas conducted with a set of 127 durum landraces from five Iranian provinces and 23modern cultivars with the following objectives: (1) to explore the genetic variabilityin YPC, yield and yield-related traits in 150 accessions of Iranian durum landracesand cultivated wheat; and (2) to characterize and classify diverse durum wheatgenotypes based on their overall similarity in terms of agronomic and YPC data.

Materials and methods

Plant materials

A collection of 150 diverse accessions of durum wheat obtained from Iran’s NationalGene Bank was used in this study. This collection was composed of landraces andcultivated varieties. Descriptions of the investigated germplasms, including theirnames, genotypic ID, donor institutions and countries of origin, are given in Table 1.

Field experiments and phenotypic data

In 2009–10 and 2010–11, the population was planted in an alpha-lattice experimentaldesign (Patterson and Williams 1976) and two replications were grown in duplicated

Table 1. List of the accessions used in this study.

Origin Number Genotype ID or name

Afghanistan 1 150a

Argentina 1 157a

Australia 1 159a

Azerbaijan, East 8 7, 444, 446, 449, 441, 442, 443, 1Bulgaria 2 155a, 160a

CIMMYT 1 502 (Yavaros)a

France 1 136a

Gene Bank 5 88, 89, 90, 91, 92Iranb 39 97, 107, 111, 130, 131, 216, 219, 221, 223, 224, 226,

228, 240, 242, 246, 249, 250, 254, 264, 268, 297,299, 312, 313, 314, 315, 341, 342, 343, 344, 361,374, 480 (Dena)a, 500 (Karghi)a, 501a (Arya), 288,

269, 415, 24 (Zardak)Italy 8 459a, 460a, 462a, 463a, 464a, 465a, 466a, 467a

Kermanshah 18 204, 205, 274, 275, 370, 371, 375, 376, 377, 378,411, 417, 504a (Syri and four), 505a (D-86-14),

506a (Sefid), 270, 423, 503 (Parisht)Khorasan 28 8, 38, 43, 44, 49, 52, 54, 55, 58, 60, 66, 316, 317,

319, 320, 321, 322, 323, 324, 325, 327, 328, 329,330, 335, 338,127, 2

Khuzestan 7 176, 177, 272, 426, 431, 438, 439Lorestan 28 68, 73, 74, 75, 76, 170, 182, 183, 186, 191, 192, 193,

194, 195, 197, 202, 206, 207, 208, 210, 215, 218,225, 349, 350, 447, 448, 456

Portugal 1 135a

Yugoslavia 1 123a

Total 150

Note: aCommercial cultivars. Genotypes with no symbol are Iranian landraces. bThe province of origin forthe genotypes in this group is unknown.

Archives of Agronomy and Soil Science 1341

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

field trials conducted at Gerize (Grz) (358160N 478010E) and Ghamloo (Ghl)(358110N 478290E), Kurdistan, Iran (Table 2). Soil samples were collected at eachexperimental site and analysed for macronutrients, CaCO3 and pH. The soils showeddifferences in macronutrients and CaCO3 concentrations at two locations (Table 2).All of the fields were prepared by ploughing to a depth of 0.25 m, mechanicallyplanted and harvested. The trials were grown in four-row plots that were seeded inan area 0.8 6 2 m. The seeding rate was *300 seeds m72 at both locations. Theentries were grown without application of external inputs. Weeds were controlledmanually. After all of the plots had reached physiological maturity, they wereharvested with a small plot combine, dried to constant moisture and stored forlaboratory analysis. Data were collected for plant height (PH; cm), grain yield(GYLD; kg ha71), biomass (Bio; g m72), harvest index (HI), grains per m2 (Grn;g m72), spikes per m2 (Spk; m2), grains per spike (GrnSpk), 1000-kernel weight(TKW; g) and culm weight (Cum; g). A subsample of 10 randomly selected plantswas used to calculate the number of plant height. Plant height was measured in cmfrom soil surface to tip of the spike excluding the awns. Yield and yield-related traitshave been calculated according to the method presented by Reynolds et al. (2001).

Yellow pigment content

The procedure for b-carotene determination used here was based on ApprovedMethod 14-50 (American Association of Cereal Chemists [AACC] 2000). Water-saturated n-butanol (WSB) was prepared by mixing n-butanol with distilled water inan 8:2 ratio. Sample lots (0.125 g of flour) were used and tested in 1.5-mLmicrocentrifuge tubes. WSB was added to each sample. The tubes containing flourand WSB were vigorously mixed by vortexing, and kept in the dark for 16–18 h toextract b-carotene. Tubes were centrifugated at 10 000 g for 10 min to recover thesupernatant. The absorbance of the supernatant was measured at 440 nm on aspectrophotometer (T80 UV/VIS Spectrophotometer) using a microcuvette. Knownquantities of pure b-carotene (Sigma C-9750) were used for calibration. Threeindividual measurements per extracted sample were recorded, and the absorbancevalues were averaged and converted into YPC (mg kg71) using the extinctioncoefficient for b-carotene. A new standard curve was prepared for each batch ofsamples observed (Santra et al. 2003).

Statistical analyses

Descriptive statistical analysis consisted of median and rank (minimum–maximum).The agronomic traits and YPC were fitted to linear mixed model. All sources ofvariance were considered random effects (Little et al. 1996). The combined analysis ofvariance (ANOVA) across environments showed that were significant genotype bylocations interaction effects (results not shown). Therefore, annual means for the singletraits were considered for combined ANOVA. The following linear model was applied:

Xijkl ¼ mþ tiðekjÞ þ bj ðekÞ þ ek þ g1 þ geðkl Þ þ eijkl

where Xijkl ¼ plot value of genotype l at year k in replication j and incompleteblock i; m ¼ overall mean; tiðekjÞ ¼ effect of incomplete block i in year k and

F. Fayaz et al.1342

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Table

2.

Experim

entaldetailsofthefourfieldtrials.

Environment

Grz-i09

Ghl-i09

Grz-r10

Ghl-r10

Environmentnumber

12

34

Country

Iran

Iran

Iran

Iran

Site

Gerize

Ghamloo

Gerize

Ghamloo

Co-ordinates

358160 N

478010 E

358110 N

478290 E

358160 N

478010 E

358110 N

478290 E

Altitude(m

)1400

1800

1400

1800

Soiltextture

Clay-loam

clayandclay-loam

withahighpercent

oflime

Clay-loam

clayandclay-loam

withahighpercent

oflime

N–P–K

(mgkg-1)

1038-0.71-3.21

580-0.64-1.3

975-0.67-3.4

486-0.51-1.5

Year

2009

2009

2010

2010

Coordinates

358160N

478010E

358110N

478290E

358160N

478010E

358110N

478290E

Seedingdate

12/11/2008

9/11/2008

13/11/2009

15/11/2009

Harvestdate

03/07/2009

13/07/2009

07/07/2011

11/07/2011

Waterinput(m

m)

378

356

211

166

Tem

perature

(8C)

Tmax

40

39

41

37

Tmean

16.46

12.72

16.14

15.16

Tmin

711

716

719

722

Meanyield

(kha7

1)

1588.5

1472.9

1387.4

1025

Note:arandiafter

thedash

indicate

rainfedandirrigatedsites,

respectively.Grz,Gerize;

Ghl,Ghamloo.Tmax,TmeanandTminare

themaxim

um,meanandminim

um

temperatures,respectively.

Archives of Agronomy and Soil Science 1343

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

replication j; bj ðekÞ ¼ effect of replication j in year k; gl ¼ effect of genoype l;ge(kl) ¼ interaction effect of genotype l with year k; eijkl ¼ error. The inspection ofagronomic and YPC broad sense heritability (H) was carried out using the followingexpression:

H ¼s2g

ðs2gþ sgy =rþ s2R =ryÞ

Where the variance component s2g equalled the genotypic variance, s2gy wasgenotype 6 year and s2R was the residual variance, r was the number of replicationand y the number of years (Milligan et al. 2003).

Principle components analysis (PCA) was performed using the PRINCOMPprocedure (SAS Institute 2000). Then, annual means for the single traits wereconsidered for PCA. PCA relies upon the eigen vector decomposition of the covarianceor correlation matrix (Granati et al. 2003). Genotype scores (eigen vectors) of the sevenprinciple components (PCs) that had eigen values greater than unity were subjected tohierarchical cluster analysis using DARwin version 5.0 and Ward’s minimum variancemethod as the clustering algorithm (Perrier et al. 2003). Ward’s minimum method is ahierarchical clustering procedure in which the similarity used to join clusters iscalculated as the sum of squares between the two clusters summed over all variables(Hair et al. 1998). It minimizes the within cluster sums of squares across all partitions.ANOVA and Duncan’s mean comparison test were performed between the clusters(groupings) by the GLM procedure (SAS Institute Inc 2000).

Results

Variation in YPC and agronomic traits

The mean grain yield performance of the durum wheat landraces was variable at thefour environments. The highest yielding experiment was Grz 2009, followed byGhl 2009, Grz 2010 and Ghl 2010. These differences may be due to somewhatdifferent soil productivity, but more importantly to seasonal weather changesbetween two regions (Table 2). For example, precipitation and average temperaturewere 3.26 mm/168C, 2.42 mm/12.48C, 3.14 mm/158C and 2.58 mm/12.68C for thefour environments of the field experiments, respectively. Therefore, precipitationdistribution and temperature were markedly different between two experimentallocations. The durum wheat crop most likely suffered more from periods of waterand temperature stress during the growing season in 2010 compared with 2009.

Impressive genetic variation was found for YPC and agronomic traits in bothyears and locations, especially in the case of agronomic traits (Table 3). In Grz 2009,the genotypic variation Grn, GrnSpk, Spk and GYLD were 10, 6.5, 6.1 and 8-foldrespectively. Therefore, these traits had the largest variation among the agronomictraits in Grz 2009. The genotypic variation of YPC, Bio and TKW were nearlythreefold. The genotypic variation HI and Cum was nearly fivefold. In the case ofPH, genotypic variation was 2.3-fold and showed the smallest genetic variation.Similar to the results found in other environments. Overall, the largest geneticvariations among traits were found in Ghl 2009. Over all of the environments, YPCranged from 1.85 to 8.95 mg kg71 in field-grown samples (Table 3). The greatestrange in YPC was 8.95 mg kg71 at Grz in 2010.

F. Fayaz et al.1344

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Table

3.

Summary

ofYPC

andagronomic

traitsmeasuredon150accessionsonfourfieldexperim

ents.

Grz

2009

Bio

(gm

72)

YPC

(mgkg7

1)

Cum

(g)

Grn

GrnSpk

HI

PH

(cm)

Spk

TKW

(g)

GYLD

(kgha7

1)

Median

3762

4.8

3.37

3839

23.19

0.26

133.9

181.4

40

1566.6

Max

7230

7.9

7.15

10879

59.53

0.58

184.6

468.5

57.67

3716

Min

2218

2.28

1.52

1068

9.13

0.11

81.5

76.79

21.46

469

Ghl2009

Bio

YPC

Cum

Grn

GrnSpk

HI

PH

Spk

TKW

GYLD

Median

2074

3.91

3.15

3187

22.55

0.35

83.3

141.7

44.72

1423

Max

3717

6.78

5.49

9055

54

0.57

122.5

395

63.9

3306

Min

852.5

1.86

1.6

916.3

9.8

0.19

42.54

62.3

30.35

347.6

Grz

2012

Bio

YPC

Cum

Grn

GrnSpk

HI

PH

Spk

TKW

GYLD

Median

3657

4.86

3.19

3807

21.11

0.24

122

172.6

35.61

1324.4

Max

6739

8.95

5.74

8499

47.87

0.5

158.4

376

54.15

3680

Min

1760

1.88

1.62

966.1

8.62

0.1

71.5

50.15

13.98

391.1

Ghl2010

Bio

YPC

Cum

Grn

GrnSpk

HI

PH

Spk

TKW

GYLD

Median

1609

3.66

2.17

3187

20

0.31

82.04

134.5

39.38

1033

Max

3497

6.23

3.83

7727

44

0.49

115.3

254.8

58

2307

Min

765

1.85

0.95

785

7.57

0.15

39.1

56

16.8

312.8

Abbreviations:YP,Yellow

pigmentcontent;GYLD,Grain

yield;TKW,Thousandkernel

weight;GrnSpk,Grainsper

spike;

HI,Harvestindex;andPH,Plantheight.

Archives of Agronomy and Soil Science 1345

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

ANOVA

ANOVA for all traits indicated highly significant variation among genotypes withineach location. The ANOVA of YPC and agronomic traits showed significant effectof genotypes for both locations. The block effect for all of the characters except YPCwas statistically significant. The genotype 6 year interaction was significant for alltraits, but the variance due to genotypes was larger than the variance due to thegenotype 6 year interaction (Table 4), resulting in high broad sense heritabilityestimates for most analyses traits, especially in Grz location. We obtained high h2 inGrz location for all traits because of the high genetic variance between genotypes.Heritability estimates were high in Ghl location for some traits such as YPC, HI andPH (Table 4). In addition, comparison of heritability of traits in the two locations(Grz and Ghl) shows that Ghl has lower values than Grz.

Classification of accessions

Based on the cluster analysis, the 150 durum wheat genotypes were separated intofive major groups that each had two or more subgroups (the genotypes werecategorized into six clusters based on an analysis of molecular data; data not shown).Figure 1 illustrates the five clusters formed by an uprooted hierarchical clustering.Table 5 summarizes the number of genotypes and their proportion in each clusterorganized by their country of origin. An ANOVA revealed significant differencesbetween the clusters for all traits. The means of the clusters and the results ofDuncan’s mean comparison tests are presented in Table 6. The first cluster (Clus1) ismainly composed of advanced genotypes from different countries of origin. Thisgroup is mainly characterized by high values of Carn, GraSpk and HI, and low PH(Table 6). The second cluster mainly comprises genotypes of the Lorestan landraces.These groups of genotypes are tall and exhibited the highest mean values for Spk.The next three clusters (Clus3, Clus4 and Clus5) form a major group of accessions ofdiverse origins. Clus3 is heterogeneous with regard to traits and origin; however, itmainly includes the Iranian landraces of unknown provincial sources and has thehighest values for GYLD, Grn and PH. Clus4 has the highest values for TKW andCum, whereas the mean values for the Carn are the lowest. Similar to Clus4, Clus5 isalso characterized by heterogeneous origins. Unlike the other clusters, the agronomictraits in this cluster have the lowest mean values. Genotypes of Clus5 arecharacterized by low GrnSpk, HI and GYLD.

Principal component analysis

Seven PCs had eigen values 41 and accounted for 88.9% of the total variance in thedata (Table 7). The proportions of the total variance attributable to the first threePCs were 25.2, 18.1 and 15.6%. The relative importance of each of the traits to thedifferent PCs is given by their corresponding eigen vectors, which are presented inTable 7. The results showed that PH had the highest loading in PC1, indicating thatit contributed significantly to this component. However, other traits were lessimportant for PC1. Cum was the main trait contributing to PC2. For PC3, traitssuch as GYLD, HI and GrnSpk were the most important, whereas Carn andGrnSpk were the main contributors to PC4. Multiple traits contributed to PC5 invarying proportions in different locations. PC6 was mainly affected by TKW. PC7was derived mainly from the variance contributed by differences in Carn and TKW.

F. Fayaz et al.1346

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Table

4.

Variance

estimatesforrandom

effects

from

thecombined

analysisofvariance

(ANOVA)ofyellow

pigmentconcentration(Y

P)(m

gkg7

1),

GYLD

(kgha7

1),thousandkernel

weight(TKW,g),grainsper

spike(G

rnSpk),culm

weight(C

um,g),harvestindex

(HI),biomass

(Bio

gm

72),grains

per

m2(G

rn),spikes

per

m2(Spk)andplantheight(PH,cm

)usingPROC

MIX

ED

across

twoyears.

df

Bio

(102)

YPC

Cum

Grn

(102)

GrnSpk

HI(107

2)

PH

Spk

TKW

GYLD

(102)

Grz

2009–10

Y1

41.58

0.006

0.02

464.68

2.34

0.1

68.4

115.4

713.83

Rep

(y)

214.33

0.06

0.04

748.53

2.64

0.1

17.6

208.5

4428.99

Blk

(y6

rep)

56

488.98***

0.05

0.08***

1967.29***

4.58***

0.3***

21.4***

568.5***

7***

374.29***

Gen

149

346.01

0.8

0.18***

2850***

18.72***

0.15***

186.1***

311.63***

10.8***

304.96**

Y6

Gen

149

462.2**

0.3

0.04

741.26

6.8

0.022

28.75**

277.5*

3.67*

261.5

Residual

242

408.51

0.1

0.04

746.34

13.6

0.2

80.6

232.2

8.42

221.26

Heritability

51%

82%

86%

83%

73%

71%

84%

61%

73%

64%

df

Bio

(102)

YPC

Cum

Grn

(102)

GrnSpk

HI(107

2)

PH

Spk

TKW

GYLD

(102)

Ghl2009–10

Y1

534

0.01

0.54

1288.64

2.63

0.6

1.75

47.24

17.95

971.88

Rep

(y)

2208.24

0.01

0.07

1002.12

8.68

0.1

10.44

19.73

2.22

128.1

Blk

(y6

rep)

56

221.96***

0.007

0.06***

732.1**

3.4*

0.4**

10.21***

186.7***

5.05***

182.46***

Gen

149

35.94***

0.48***

0.06**

387.3

2.5*

0.5***

89.15***

224.2*

4.5*

171.71**

Y6

Gen

149

8.45

0.11***

0.02

1971.3**

5.35*

0.2

26.2

359**

6.2***

193.95

Residual

242

158.04

0.2

0.2

918.5

12.4

0.6

63.7

128.8

12.3

140.28

Heritability

45%

81%

50%

24%

30%

67%

75%

51%

42%

56%

Note:*p5

0.05,**p5

0.01and***p5

0.001.Abbreviations:YP,Yellow

pigment;GYLD,Grain

yield;TKW,Thousandkernel

weight;GrnSpk,Grainsper

spike;

HI,

Harvestindex;andPH,Plantheight.

Archives of Agronomy and Soil Science 1347

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

In general, it is clear that some traits inconsistently contributed to the total variancein different locations. PCAs were performed independently for each genotype in eachexperimental location to determine the relationships between yield and yield-relatedtraits. Axes 1 and 2 of the PCA for the Grz site explained 35.5 and 24.3% of thevariability found, respectively. Given that most of the variation could be explainedby considering these two axes, we used the projections of the data points on theplanes determined by axes 1 and 2. The graphs shown in Figures 2a and 3a displaythe eigen vectors of the characteristics that most influenced each axis. The first axisof Figure 2a was positively correlated with the GYLD and its components, i.e. HI,GrnSpk and Grn, while it was negatively correlated with the PH. The second axiswas positively correlated with the spikes per m2 and the Carn, meanwhile TKW,Cum and Bio had negative effects. Figure 2a shows that there is a strong relationshipbetween yield, HI, grains per m2 and GrnSpk, whereas the length of the PH vectorwas negatively correlated with grain yield because the vectors form an angle of 1808with a negative cosine. The graphs shown in Figures 2b and 3b display the projectionof the points representing the mean genotype values across locations on planes 2

Figure 1. An un-rooted hierarchical clustering tree based on the Euclidean distances of thestandardized mean phenotypic data across environments. Adv, commercial cultivars; Afg,Afghanistan; Arg, Argentina; Aus, Australia; Azr, Azerbaijan; Bul, Bulgaria; D, D-86-14; Fra,France; Ir, Iran; It, Italy; Kar, Karghi; Kermsh, Kermanshah; Khor, Khorasan; Khuz,Khuzestan; Lor, Lorestan; Pari, Parisht; Por, Portugal; Syri, Syri and Four; Yava, Yavaros;Yug, Yugoslavia.

F. Fayaz et al.1348

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

and 3. Figure 1b shows that the genotypes form five separate clusters. Clus1 (moderncultivars) is located in the lower right quadrant and is separated by the Grn, HI,GYLD and GrnSpk vectors. The Clus2 genotypes are located in the lower leftquadrant, which corresponds to the PH vector. Clus3 is a central group that isseparated by the TKW, Cum and Bio vectors. Meanwhile, Clus4 and Clus5 wereseparated by the Carn and Spk vectors, respectively. Axes 1 and 2 of the PCA for thegenotypes at the Ghl location explained 35.8 and 19.4% of the total variation,respectively (Figure 3). The first axis was negatively correlated with the Grn, GYLD,Bio and PH. The second axis was positively correlated with the Spk and Carn andnegatively correlated with the GrnSpk, Cum, HI and TKW. Figure 3a shows thatthere is a strong association between the GYLD and the Grn, Bio and PH. The yieldand the yield components traits did not group genotypes into any clustering pattern(Figure 3b).

Discussion

The yellow pigment analysis of 150 accessions showed highly significant differences(p 5 0.001) in the grain pigment concentration between the wheat genotypes in twolocations. The results showed a wide variation in the YPC ranging from the highestvalue of commercial durum wheat genotypes (8.95 mg kg71) to the lowest value ofthe Iranian landraces (1.85 mg kg71). Overall, commercial cultivars showedsignificantly higher values compared with traditional cultivars. These results are inagreement with those of previous studies (Leenhardt et al. 2006; Digesu et al. 2009).Heritability estimates for YP were high for individual locations (Table 4), confirmingearlier reports that YP is highly heritable in durum wheat (Elouafi et al. 2001; Clarkeet al. 2006). Overall, heritability of pigment concentration is reasonably high. Thiswill permit early generation selection for the trait (Braaten et al. 1962). Reductionheritability of agronomic traits in Ghl location is probably due to stress conditions inthis region. Stress condition was due to climate variation and soil nutrient contentsbetween two locations (Table 2).

Pigment concentration demonstrated a slightly negative correlation with kernelweight. PCA analysis in two locations also confirmed this negative correlationbecause the vectors for TKW and Carn are pointing in opposite directions from one

Table 5. The number and proportion of genotypes in each cluster with respect to theirsources or provinces of origin.

Origin

Cluster1 Cluster2 Cluster3 Cluster4 Cluster5

N % N % N % N % N %

Adv 16 50 3 9 1 4 3 8Azr 3 9 5 14Ir 9 28 10 37 12 38 4 17 6 17Kermsh 2 6 5 19 3 9 3 13 3 8Khuz 1 3 1 4 1 3 2 9 2 6Khor 3 9 2 7 5 16 8 35 9 25Lor 1 3 9 33 5 16 5 22 8 22Total 32 27 32 23 36

Note: Adv, Advanced cultivars; Azr, Azerbaijan; Ir, Iran; Kermsh, Kermanshah; Khor, Khorasan; Khuz,Khuzestan; Lor, Lorestan.

Archives of Agronomy and Soil Science 1349

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Table

6.

Meansoftheagronomic

andgrain

betacarotenetraitsofthefiveclusters.

Cluster

Grn

Spk

GrnSpk

TKW

Carn

PH

HI

GYLD

Bio

Cum

13799.2ab

149.8b

25.78a

39.24b

4.85a

85.56b

0.34a

1454.57ab

2612.2a

2.93b

23419.9bc

188.72a

18.91c

39.68b

4.18bc

104a

0.28b

1352.6b

2735.4a

2.64c

33871.5a

177.13a

22.3b

39.52b

4.72a

110a

0.29b

1513.5a

3049.4a

3.06b

43206.3c

137.29b

23.66b

44.04a

3.8c

109.7a

0.3b

1393.6ab

2972.7a

3.55a

53030.9c

153.78b

20.27c

38.25b

4.58ab

108.35a

0.25c

1143.3c

2774.3a

3.1b

Note:Meansfollowed

bythesameletter

are

notsignificantlydifferentfrom

oneanother

accordingto

Duncan’smeancomparisontest

(p¼

0.01).

F. Fayaz et al.1350

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

another (Figures 2a and 3a.) This negative correlation was also found by Alvarezet al. (1999), Marais (1992), Veronico et al. (2001), Clarke et al. (2006) and Zhanget al. (2008) and could be partly attributed to dilution effects, with the increasedgrain starch found in large kernels reducing the proportion of the pigments.

On the basis of genetic diversity with regard to important agronomic traits andgrain carotene, the investigated durum wheat germplasm was grouped into five setsby a cluster analysis. This procedure revealed that the modern varieties and breedinglines differentiated more significantly from the germplasm of the Iranian landraces.Clus1 versus the other clusters is shown in Figure 1. The results of this classificationof accessions revealed that accessions within each cluster belonged to differentcountries or Iranian provinces, which suggested that there was no clear relationshipbetween accessions and geographical diversity (Table 5). Therefore, greater emphasisshould be placed on the individual accessions rather than geography as a source ofphenotypic diversity in this germplasm. Similar results have been reported indifferent crops by several studies on white clover (Jahufer et al. 1997), durum wheat(Annicchiarico et al. 2000), safflower (Jaradat and Shahid 2006; Khan et al. 2009)and watermelon (Szamosi et al. 2009).

To detect the combination of variables that best explained the variability betweengenotypes, PCA including the yield and yield-related variables was carried out; theresults are shown in Figure 2. The negative relationships among the three principal

Table 7. Eigen vectors (loadings) of the first seven principal components.

Eigen vector

Trait 6 Location PC1 PC2 PC3 PC4 PC5 PC6 PC7

BioGhl 0.33 70.01 0.30 0.14 70.01 0.14 70.16BioGrz 70.02 0.26 0.22 0.05 0.51 0.07 0.03CarnGhl 70.05 70.06 70.08 0.53 0.05 0.22 0.41CarnGrz 70.04 70.19 70.10 0.47 0.01 0.20 0.41CumGhl 0.11 0.32 0.31 0.15 70.06 70.21 0.09CumGrz 70.09 0.43 0.08 0.17 70.11 0.23 70.12GmGhl 0.25 70.28 0.30 0.14 70.11 0.04 70.16GmGrz 70.31 70.15 0.20 0.12 0.37 70.05 70.19GSpkGhl 0.00 0.01 0.39 0.20 70.23 70.50 0.13GSpkGrz 70.36 0.08 0.13 0.21 70.10 0.11 70.32HIGhl 70.13 70.20 0.31 70.21 70.21 70.11 0.34HIGrz 70.36 70.10 0.20 70.01 70.08 0.11 70.11PHGhl 0.35 0.22 0.05 0.06 0.16 70.02 0.02PHGrz 0.28 0.29 0.02 0.06 0.22 70.04 70.01SpkGhl 0.26 70.31 0.01 0.00 0.03 0.40 70.27SpkGrz 0.12 70.31 0.04 70.18 0.50 70.22 0.22TKWGhl 70.07 0.13 0.15 70.37 0.01 0.41 0.33TKWGrz 0.01 0.23 0.21 70.27 70.13 0.27 0.26GYLDGhl 0.23 70.23 0.37 70.02 70.11 0.22 70.03GYLDGrz 70.31 70.03 0.31 70.02 0.31 0.08 70.04

Eigen value 5.05 3.62 3.13 1.99 1.5 1.26 1.02% of the total

variance explained25.2 18.1 15.6 9.9 7.5 6.3 5.1

% of the cumulativevariance explained

25.2 43.3 58.9 68.8 76.3 83.8 88.9

Abbreviations: Grz: Gerize; Ghl: Ghamloo; Bio: Biomass; Carn: Grain carotene; Cum, Culm weight; Grn,Grains per m2; GSpk, Grains per spike; HI, Harvest index; PH, Plant height; Spk, Spikes per m2; TKW,Thousand kernel weight; and GYLD, Grain yield.

Archives of Agronomy and Soil Science 1351

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Figure 2. Principal component analysis (PCA) projections on axes 1 and 2, which accountfor 59.8% of the total variance for 150 genotypes at the Gerize location. (a) The eigen valuesof the correlation matrix are symbolized as vectors representing the grain yield (GYLD),harvest index (HI), grains m72 (Grn), grain carotene (Carn), spikes m72 (Spk), plant height(PH), 1000-kernel weight (TKW), culm weight (Cum), total biomass (Bio) and grains spike71

(GrnSpk). (b) Plot of the points corresponding to the genotypes from cluster 1 (þ), cluster 2(¤), cluster 3 (O), cluster 4 (*) and cluster 5 (D).

F. Fayaz et al.1352

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

yield components, i.e. Spk, GrnSpk and TKW, have been described previously(Slafer et al. 1996). Compensatory effects between Spk and GrnSpk, indicated by thedifferent directions of their eigen vectors, were observed in both locations (Figures2a, 3a and Table 6). These effects may be due to the inverse relationship betweentiller production and apical development, presumably arising from the negative

Figure 3. Principal component analysis (PCA) projections on axes 1 and 2, which accountfor 55.2% of the total variance for 150 genotypes at the Ghamloo location. (a) The eigenvalues of the correlation matrix are symbolized as vectors representing the grain yield(GYLD), harvest index (HI), grains m72 (Grn), grain carotene (Carn), spikes m72 (Spk),plant height (PH), 1000-kernel weight (TKW), culm weight (Cum), total biomass (Bio) andgrains spike71 (GrnSpk). (b) Plot of the points corresponding to genotypes from cluster 1 (þ),cluster 2 (¤), cluster 3 (O), cluster 4 (*) and cluster 5 (D).

Archives of Agronomy and Soil Science 1353

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

allometry between these traits during plant development which causes competitionfor assimilates among different structures (Moragues et al. 2006).

PCA was used for the graphical interpretation of the changes occurring in the graincarotene, yield and yield-related traits caused by the experimental locations. Figure 2bshows the genotype points plotted on the plane arranged in five clusters based on theGrn, yield and yield-related traits. The variation in Clus1 (advanced cultivars) can beexplained by Grn, HI, GYLD and GrnSpk. The mean values for these traits have beenincreased by selective breeding programmes; therefore, these traits are responsible forthe separation of this cluster from the others. The genetic gain was most clearlyassociated with a higher kernels number per m2 indicating a larger grain-sink size and ahigher number of spikes per m2. The gradual reduction in plant height associated withan increased HI has represented the main breeding goal with an effect on the sinkcapacity and on the biomass partitioning (Austin et al. 1989; Perry and D’Antuono1989; Slafer and Andrade 1989; Reynolds et al. 1999). These changes have beenassociated with the presence of the dwarfing genes Rht-B1 and Rht-D1, which conferreduced responses to gibberellins (Peng et al. 1999). Clus2 had the largest number ofSpk but fewer GrnSpk (Table 6). This component was most highly correlated with yieldamong this group of genotypes. Clus3 includes the Iranian landraces of unknownprovincial sources and has the highest values for GYLD, Grn and PH, Clus4 and Clus5were characterized based on PH, TKW, Cum and Bio (Figure 2a; Table 6). Thesecontrasting yield-formation strategies among clusters are probably a consequence ofpower adaptation or human selection for each group, which have likely conferredadaptive advantages to the genotypes of one or the other group (Moragues et al. 2006).By contrast, no clustering pattern was observed within the Ghl location (Figure 3b),probably because of the genotype by location interaction (Table 3). This fact wasrevealed by comparison of the yield and yield-related traits between the two locations(data not shown). Therefore, in this location, Iranian landraces had higher yield thanthe modern cultivars, because landraces of wheat generally tolerant to biotic andabiotic stress have been grown under low-input or sustainable farming conditionswhere they produce reasonable yield (Moghaddam et al. 1997).

The results of this study underline the substantial variability in grain b-carotene,yield and yield-related traits among Iranian durum germplasm, displaying trendsfollowed by breeding in terms of productivity and grain b-carotene traits, and allowingthe identificationofwheat landraces useful for breedingpurposes. The largest variabilityfor YPC, yield and yield-related traits was recorded in landraces from the Iranian basinin agreementwith its proximity to theFertile Crescentwhere the domestication ofwheatinitiated (Harlan 1986). As in previous studies (Motzo et al. 2004; De Vita et al. 2007;Royo et al. 2007), our work indicated that the advanced cultivars had greater yields andlower PH than landraces, but with higher TKW and Spk than Iranian landraces. Asimilar result was demonstrated by Moragues et al. (2006).

Conclusion

The selection of genotypes for a plant-breeding programme is primarily based oncertain morphological traits. Therefore, we believe that studies of molecular markersto complement the studies of the morphological traits in the field are required.Information about current genetic diversity permits the classification of our availablegermplasm into various/heterotic groups, which is particularly important for hybrid/cross breeding programmes in durum wheat.

F. Fayaz et al.1354

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Acknowledgements

The authors would like to express their appreciation to Dr Salar Shaaf for his assistance withthe statistical analyses. We gratefully acknowledge the Iran National Gene Bank and DrMostafa Aghaee for providing the germplasm for this study. We also wish to thank the staff ofthe Gerize and Ghamloo stations for their skilled technical assistance. This study was partiallyfunded by the Islamic Azad University, Sanandaj Branch.

References

Alvaro F, Isidro J, Villegas D, Garcıa del Moral LF, Royo C. 2008. Breeding effects on grainfilling, biomass partitioning, and remobilization in Mediterranean durum wheat. Agron J.100:361–370.

Alvarez JB, Martin LM, Martin A. 1999. Genetic variation for carotenoid pigment content inthe amphiploid Hordeum chilense–Triticum turgidum conv. durum. Plant Breed. 118:187–189.

American Association of Cereal Chemists (AACC). 2000. AACC official method. In:Approved methods of the American Association of Cereal Chemists. 10th ed. St. Paul(MN): AACC. p. 14–50.

Annicchiarico P, Pecetti L, Boggini G, Doust MA. 2000. Repeatability of large-scalegermplasm evaluation results in durum wheat. Crop Sci. 40:1810–1814.

Arzani A. 2002. Grain yield performance wheat germplasm under Iranian dry land andirrigated field conditions. Sabrao J Breed Gen. 34:9–18.

Austin RB, Ford MA, Morgan CL. 1989. Genetic improvement in the yield of winter wheat: afurther evaluation. J Agr Sci Camb. 112:295–301.

Braaten MO, Lebsock KL, Sibbitt LD. 1962. Intergeneration relations of physical propertiesof dough and carotenoid pigment content in durum wheat. Crop Sci. 2:277–281.

Caradus JR, Forde MB, Wewala S, Mackay AC. 1990. Description and classification of awhite clover (Trifolium repens L.) germplasm collection from southwest Europe. NZ J AgrRes. 33:367–375.

Ceccarelli S, Acevedo E, Grando S. 1991. Breeding for yield stability in unpredictableenvironments: single traits, interaction between traits, and architecture of genotypes.Euphytica. 56:169–185.

Clarke FR, Clarke JM, McCaig TN, Knox RE, DePauw RM. 2006. Inheritance of yellowpigment concentration in seven durum wheat crosses. Can J Plant Sci. 86:133–141.

De Vita P, Nicosia OL, Nigro F, Platani C, Riefolo C, Di Fonzo N, Cattivelli L. 2007.Breeding progress in morphophysiological, agronomical and qualitative traits of durumwheat cultivars released in Italy during the 20th century. Eur J Agron. 26:39–53.

Digesu AM, Platani C, Cattivelli L, Mangini G, Blanco A. 2009. Genetic variability in yellowpigment components in cultivated and wild tetraploid wheats. J Cereal Sci. 50:210–218.

Elouafi I, Nachit MM, Martin LM. 2001. Indentification of a microsatellite on chromosome7B showing a strong linkage with yellow pigment in durum wheat (Triticum turgidum L.var. durum). Hereditas. 135:255–261.

Granati E, Bisignano V, Chiaretti D, Crino P, Polignano BG. 2003. Characterization ofItalian and exotic Lathyrus germplasm for quality traits. Genet Res Crop Evol. 50:273–280.

Hair FF, Anderson RE, Tatham RL, Black WC. 1998. Multivariate data analysis. 5th ed.Upple Saddle River (NJ): Prentice Hall. p. 239–325.

Harlan JR. 1986. Plant domestication: diffuse origins and diffusion. In: Barigozzi C, editor.The origin and domestication of cultivated plants. Amsterdam (The Netherlands):Elsevier. p. 21–34.

Hawkes JG. 1983. The diversity of crop plants. Cambridge (MA): Harvard University Press.184 pp.

Jahufer MZZ, Cooper M, Harch BD. 1997. Pattern analysis of the diversity of morphologicalplant attributes and herbage yield in a world collection of white clover (Trifolium repensL.) germplasm characterised in a summer moisture stress environment of Australia. GenetResour Crop Evol. 44:289–300.

Jaradat AA, Shahid MA. 2006. Patterns of phenotypic variation in a germplasm collection of(Carthamus tinctorius L.) from the Middle East. Genet Resour Crop Evol. 53:225–244.

Archives of Agronomy and Soil Science 1355

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Khan MA, Witzke-Ebrecht S, Maass BL, Becker HC. 2009. Relationships among differentgeographical groups agro-morphology fatty acid composition and RAPD marker diversityin safflower (Carthamus tinctorius L.). Genet Resour Crop Evol. 56:19–30.

Leenhardt F, Lyan B, Rock E, Boussard A, Potus J, Chanliaud E, Remesy C. 2006.Genetic variability of carotenoid concentration, and lipoxygenase and peroxidaseactivities among cultivated wheat species and bread wheat varieties. Eur J Agron.25:170–176.

Little RC, Milliken GA, Stroup WW, Wollfinger RD. 1996. SAS system for mixed models.Cary (NC): SAS Institute.

Marais GF. 1992. Gamma irradiation induced deletions in an alien chromosome segment ofthe wheat ‘Indis’ and their use in gene mapping. Genome. 35:225–229.

Milligan SB, Balzarini M, White WH. 2003. Broad-sense heritabilities, genetic correlations,and selection indices for sugarcane borer resistance and their relation to yield loss. CropSci. 43:1729–1735.

Moghaddam M, Ehdaie B, Waines JC. 1997. Genetic variation and interrelationships ofagronomic characters in landraces of bread wheat from southeastern Iran. Euphytica.95:361–369.

Moragues M, Garcıa del Moral LF, Moralejo M, Royo C. 2006. Yield formation strategies ofdurum wheat landraces with distinct pattern of dispersal within the Mediterranean Basin.I. Yield components. Field Crops Res. 95:194–205.

Motzo R, Fois S, Giunta F. 2004. Relationship between grain yield and quality of durumwheats from different ears of breeding. Euphytica. 140:147–154.

Patterson HD, Williams ER. 1976. A new class of resolvable incomplete block designs.Biometrika. 63:83–92.

Pecetti L, Annicchiarico P, Damania AB. 1992. Biodiversity in a germplasm collection ofdurum wheat. Euphytica. 60:229–238.

Peng JR, Richards DE, Hartley NM, Murphy GP, Devos KM, Flintham JE, Beales J, FishLJ, Worland AJ, Pelica F, Sudhakar D, Christou P, Snape JW, Gale MD, Harberd NP.1999. ‘Green revolution’ genes encode mutant gibberellin response modulators. Nature.400(6741):256–261.

Perrier X, Flori A, Bonnot F. 2003. Methods of data analysis. In Hamon P, Seguin M, PerrierX, Glaszmann JC, editors. Genetic diversity of cultivated tropical plants. Montpellier(France): CIRAD. p. 33–63.

Perry MW, D’Antuono MF. 1989. Yield improvement and associated characteristics of someAustralian spring wheat cultivars introduced between 1860 and 1982. Aust J Agr Res.40:457–472.

Rashed MA, Abou-Deif MH, Sallam MA, Rizkallaa AA, Ramadan WA. 2007. Identificationand prediction of the flour quality of bread wheat by glidin electrophoresis. J Appl SciRes. 3:1393–1399.

Reimer S, Pozniak CJ, Clarke FR, Clarke JM, Somers DJ, Knox RE, Singh AK. 2008.Association mapping of yellow pigment in an elite collection of durum wheat cultivars andbreeding lines. Genome. 51:1016–1025.

Reynolds MP, Rajaram S, Sayre KD. 1999. Physiological and genetic changes of irrigatedwheat in the post-green revolution period and approaches for meeting projected globaldemand. Crop Sci. 39:1611–1621.

Reynolds MP, Ortiz-Monasterio JL, McNab A. Editors. 2001. Application of physiology inwheat breeding. Mexico, DF (Mexico): CIMMYT. 240 pp.

Royo C, Alvaro F, Martos V, Ramdani A, Isidro J, Villegas D, Garcıa del Moral LF (2007).Genetic changes in durum wheat yield components and associated traits in Italy and Spainduring the 20th century. Euphytica. 155:259–270.

Royo C, Martos V, Ramdani A, Villegas D, Rharrabti Y, Garcıa del Moral LF. 2008.Changes in yield and carbon isotope discrimination of Italian and Spanish durum wheatduring the 20th century. Agron J. 100:352–360.

Rumbaugh MD, Graves WL, Caddel JL, Mohammad RM. 1988. Variability in a collection ofalfalfa germplasm from Morocco. Crop Sci. 28:605–609.

Santra M, Rao VS, Tamhankar SA. 2003. Modification of AACC procedure for measuring b-carotene in early generation durum wheat. Cereal Chem. 80:130–131.

SAS Institute Inc. 2000. SAS online doc. R. Ver 8. Cary (NC): SAS Institute.

F. Fayaz et al.1356

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013

Shands HL, Wiesner LE. 1992. Use of plant introductions in cultivar development. Part I.CSSA special publication No. 17. Madison (WI): Crop Science Society of America.

Slafer GA, Andrade FH. 1989. Genetic improvement in bread wheat (Triticum aestivum L.)yield in Argentina. Field Crops Res. 21:289–296.

Slafer GA, Calderini DF, Miralles DJ. 1996. Yield components and compensation in wheat:opportunities for further increasing yield potential. In: Reynolds MP, Rajaram S, McNabA, editors. Increasing yield potential in wheat: breaking the barriers. Ciudad Obregon,Sonora (Mexico): CIMMYT. p. 101–133.

Szamosi C, Solmaz I, Sari N and Barsony C. 2009. Morphological characterization ofHungarian and Turkish watermelon (Citrullus lanatus (Thunb.) Matsum. et Nakai)genetic resources. Genet Resour Crop Evol. 56:1091–1105.

Veronico P, Somma S, Blanco A. 2001. Variabilita del contenuto in pigmenti carotenoidi nellemoderne varieta‘ di frumento duro. Atti IV Convegno Nazionale Biodiversita:Opportunita di sviluppo sostenibile. IAM, Valenzano; Bari, Italy; 2001 Sept 6–7. Vol. 3.p. 225–230.

Zhang W, Chao S, Manthey F, Chicaiza O, Brevis JC, Echenique W, Dubcovsky J .2008. QTLanalysis of pasta quality using a composite microsatellite and SNP map of durum wheat.Theor Appl Genet. 117:1361–1377.

Archives of Agronomy and Soil Science 1357

Dow

nloa

ded

by [

Farz

ad F

ayaz

] at

11:

02 0

8 O

ctob

er 2

013