HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in Crop Science Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Search for Related Content PubMed Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Agricola Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Related Collections Germplasm Enhancement Crop Genetics

PLANT GENETIC RESOURCES

Genetic Diversity in the Batini Barley Landrace from Oman

I. Spike and Seed Quantitative and Qualitative Traits

^a USDA-ARS, 803 Iowa Ave., Morris, MN 56267
^b Plant Genetic Resources Program, International Center for Biosaline Agriculture (ICBA), P.O. Box 14660, Dubai, United Arab Emirates
^c College of Agriculture, Sultan Qaboos Univ., P.O. Box 34 Al-Khod, Postal Code 123, Muscat, Oman

^* Corresponding author (Jaradat{at}morris.ars.usda.gov).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The Batini landrace of barley (Hordeum vulgare L.) is endemic to the coastal Batinah region of Oman. Although it is important to subsistence farmers, it is threatened by increasing salinity and replacement by high yielding cultivars. Seven bulk seed samples (subpopulations) of the Batini landrace were collected from farmers' fields, which provided a germplasm collection of 3191 accessions. The objectives of this study were to characterize these accessions for spike and seed qualitative and quantitative traits, quantify phenotypic diversity, and explore significant variation in seed and spike morphological traits for future selection and breeding. Variation for 26 morphological traits was assessed among the progeny of 3191 single spikes. Phenotypic diversity indices (H') differed significantly among traits and subpopulations. Weighted H' average for subpopulations was 0.501; it ranged from 0.154 for spike glaucousness to 0.853 for number of spikelets per spike. Differences in phenotypic frequencies for 20 traits were sufficient to discriminate between subpopulations. Total genetic variation (H_T) for quantitative (0.717) and qualitative (0.533) traits differed significantly. Variance component due to subpopulations was significant for seven quantitative and 12 qualitative traits, and the within-subpopulation variance component decreased in the order: qualitative (82.12%) > quantitative (78.34 ) > spike-related (68.50) > grain-related (67.25) traits. Total genetic variation and genetic differentiation estimates for qualitative traits were 25% lower than for quantitative traits. Strong, nonrandom trait associations among four seed phenotypic markers showed a hierarchical pattern, indicating an adaptive response to environmental conditions and human selection. The long history of in situ conservation of this landrace in a multitude of subsistence farming systems, undoubtedly, contributed to the high variability.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
BARLEY is one of eight founder crops of Old World agriculture (Zohary and Hopf, 1993, p. 278); the landraces of which are the evolutionary link between wild barley (Hordeum spontaneum K. Koch) and modern barley cultivars. Barley landraces are genetically heterogeneous populations comprising inbreeding lines and hybrid segregates generated by a low level of random outcrossing in each generation (Nevo, 1992). Moreover, having evolved across thousands of years in a multitude of environments and local farming systems, these landraces have developed abundant patterns of variation and would represent a largely untapped reservoir of useful genes for adaptation to biotic and abiotic stresses (Brush, 1995).

Landraces comprise the major genetic resource of cultivated barley in countries like Ethiopia (Alemayehu and Parlevliet, 1997), Iran (Brown and Munday, 1982), Jordan (Jaradat et al., 1987), Sardinia (Papa et al., 1998), Syria (Ceccarelli et al., 1987), and Turkey (Brush, 1995). Although these countries comprise a sizable part of barley's center of diversity, the land area under landraces is declining (Hammer et al., 1996). Almost 50% of all accessions maintained in ex situ collections are advanced cultivars or breeders' lines (Hammer et al., 1999), while landraces make up about 30% of these collections. One of the consequences of ex situ conservation is the loss of genetic diversity due to genetic drift after rejuvenation and seed increase cycles. The concomitant shift in population genetic structure on rejuvenation of old collections (Parzies et al., 2000a) emphasizes the need for in situ conservation of landraces and old cultivars (Brush, 1995)

Current theories (Willcox and Tengberg, 1995) suggest that barley was introduced into Oman and the southern parts of the Arabian Peninsula through trade between cultures of old Mesopotamia and other parts of the Persian Gulf. Undoubtedly, it underwent evolutionary modification resulting from artificial selection and adaptation to the harsh environment (Zohary and Hopf, 1993, p. 278). Carbonized rachis and seed of domesticated two-row and six-row barley and wheat (Triticum aestivum L.) were found in archaeological sites in southern Arabia dating back to about 5000 BP. Inscriptions were also found at a later date (about 3500 BP) indicating that wheat and barley, among other crops, had been brought down from southern Mesopotamia to this part of the Arabian Peninsula (Potts, 1993).

Barley is traditionally grown in a mixture with alfalfa (Medicago sativa L.) as fodder in the coastal parts of Oman. The land area under barley is about 250 ha, almost 30% of which is in the coastal Batinah region, and the remaining 70% is scattered in the mountainous regions of the country. Landholdings are very small (1–2 ha, on average) with a mean grain yield of about 300 kg ha^–1 (Guarino, 1990).

The Batini landrace of barley is grown along the coast of Oman as a dual-purpose or forage crop and, occasionally, as a grain crop for animal feed and human consumption in the remote mountainous regions of Oman (Guarino, 1990). This six-row landrace is endemic to the Batinah region of Oman, which is a low-lying alluvial plain extending for about 240 km from Muscat to the border of the United Arab Emirates, and stretching for about 30 km inland from the coast (Guarino, 1989). Mean annual temperature in this region is about 29.0°C, while maximum temperature may reach 50.0°C.

Very little is known about the genetic diversity and morphological variability present in this landrace in Oman, a country experiencing loss of biodiversity, especially in the Batinah region, because of replacement of landraces with modern cultivars, increasing salinity, and decreasing fresh water resources for irrigation (ICBA, 2000, p. 38). Oman was singled out (Chapman, 1985) as one of the most important remaining sites for germplasm collection in southwest Asia during the late 1980s. To fill this gap, Guarino (1989)( 1990) collected 51 barley landrace accessions as a part of a larger collection (510 accessions and 58 species). However, the National Plant Germplasm System (NPGS, 2003) lists only 16 accessions of Omani barley introduced from Oman during the early 1990s.

Although farmers may have several socio-economic incentives to replace landraces with modern introduced varieties (Kebebew et al., 2001), this landrace is still being cultivated by subsistence farmers in Oman and the adjacent regions of the United Arab Emirates because of its high adaptability and tolerance to salinity (ICBA, 2000, p. 38). Nevertheless, the continued cultivation of this landrace and other indigenous crop genetic resources of Oman are potentially threatened and could be lost before they are adequately collected and thoroughly evaluated (Cromwell, 1996).

This paper is the first of a series of four papers on phenotypic and genetic diversity in the Batini barley landrace from Oman. Objectives of this part of the study were to: (i) characterize a barley germplasm collection for spike qualitative and quantitative traits, and (ii) quantify the phenotypic diversity and explore significant variation for future exploitation in selection and breeding.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
In 1999, a barley seed collection from the Batinah region of Oman (Fig. 1, map of Oman) was acquired by the International Center for Biosaline Agriculture (ICBA) and was comprised of seven bulk samples. Seed multiplication and plant selection, on the basis of readily observable plant and spike morphological and phenological traits during the 1999-2000 cropping season, resulted in a refined selection of 3191 single spikes representing seven distinctively different subpopulations (ICBA, 2000, p. 38).

View larger version (149K): [in this window] [in a new window]   Fig. 1. Map of Oman showing the Batinah administrative district where barley landrace was collected.

Frequency data for discrete qualitative and quantitative traits were used to carry out: (i) the Kruskal-Wallis test statistic, as described in Zar (1996), to estimate the power of the qualitative and quantitative traits for discrimination among subpopulations, (ii) discriminant analysis among the seven subpopulations, determined on the basis of discrete qualitative and quantitative traits with significant P values in the Kruskal-Wallis test, (iii) estimate the phenotypic diversity index, total diversity and level of differentiation for each subpopulation and trait, and (iv) estimate mean gene diversity, mean number of alleles (categories) per locus, and percentage of polymorphic loci by the 5% criterion for each subpopulation.

The polymorphic index, as described by Zhang and Allard (1986), was calculated from relative frequencies of the 19 qualitative, seven discrete quantitative traits and seven subpopulations. The adjustment in phenotypic diversity indices suggested by Zar (1996) was implemented to account for differences in the number of phenotypes (range from 2–4), and testing for differences between two diversity indices was performed according to Hutchison (1970).

Phenotypic Diversity Analyses
Frequency data for each quantitative trait, groups of quantitative or qualitative traits, and for each subpopulation were used to calculate the following statistics according to Nei (1973)(1987): gene diversity for each phenotypic marker, H; mean number of alleles per locus, A; effective number of phenotypic markers per locus; percent polymorphic loci, P(0.05); total gene diversity, H_T; and genetic differentiation, G_ST.

Log-Linear Models
The multitrait organization of phenotypic variation within the landrace for four seed qualitative traits showing variation controlled by single genes (Nilan, 1964) was analyzed by means of log-linear models (Agresti, 1990). The most parsimonious models with best goodness-of-fit to the frequency tables of the observed data sets, were obtained by model selection procedures based on the partitioning of likelihood-ratio statistics (G²) (Agresti, 1990).

A polymorphic distance, analogous to genetic distance based on allozyme or molecular data (Nei, 1973, 1987), was calculated for the seven subpopulations by means of frequencies of qualitative, quantitative, spike-related, and grain-related traits. The total variance in each of these four groups was partitioned into its components: among and within subpopulations.

Multivariate Graphical Analyses
The principal components analysis (PCA) was performed with the standardized mean values for each of the seven quantitative traits and subpopulation. Results of PCA were used to study the interrelationships and adjustments in the seven quantitative traits, and to detect any subpopulation groupings based on similarities in trait interrelationships and adjustments. Correspondence analysis (CA) using frequencies of the awn-, spike-, glume-, and grain-related traits was performed to reduce the dimensionality of the data matrix and to identify associations among these sets of qualitative traits. Principal Coordinate Analysis (PCoA) was performed with the intersubpopulation dissimilarity distances as data units to identify major cluster groups and to graphically display relationships among subpopulations (Gower, 1966). In addition, frequency data per subpopulation were used to calculate squared Euclidean distances between subpopulations for subsequent unweighted pair group method with arithmetic average (UPGMA) clustering of subpopulations (Sneath and Sokal, 1973).

All statistical analyses were conducted by several modules in the statistical packages STATISTICA 6.0 (StatSoft Inc., 2001), SYSTAT 10.2 (SYSTAT Software Inc., 2002), and POPGENE 1.32 (Yeh et al., 2000), unless otherwise specified.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Univariate ANOVA
Summary results of the analysis of variance, mean separation, and variance component due to subpopulations of the phenotypic diversity indices for 26 traits are presented in Table 1. A total of 54 categories were observed for the 19 qualitative traits scored on accessions of all seven subpopulations with an average of 2.84 categories per qualitative trait. However, the mean number of effective qualitative categories per trait was 2.06, or 72.8% of the observed number. On the other hand, data reduction of the seven quantitative traits resulted in a total of 21 categories, and the average effective number of phenotypes per trait was 2.195, or 73.2% of the observed number.

Phenotypic diversity index estimates ranged from 0.149 for the grain dorsal view to >0.8 for spike-related traits. The weighted average phenotypic diversity index was 0.501; however, separate phenotypic diversity indices for quantitative (0.768) and qualitative (0.403) traits differed significantly from each other (P < 0.01).

Differences among subpopulations in the phenotypic diversity index were significant for 17 of the 19 qualitative, and for all seven quantitative traits. However, the magnitude of these differences among subpopulations, as expressed by the percent significant differences for all possible 21 pairwise comparisons among seven subpopulations, varied from 0.0 for grain dorsal view to 85.7% for four spike- and rachis-related traits.

The variance component due to subpopulations ranged from 9.0 for grain dorsal view to >70% for spike-related traits. Significance tests, as seen from the level of significance associated with the percent variance due to subpopulations, suggest that a sizable portion of the variability in phenotypic diversity indices for 12 qualitative and all seven quantitative traits was accounted for by differences among subpopulations. P values for the Kruskal-Wallis test indicate that differences in phenotypic frequencies for 13 qualitative and for all seven quantitative traits were powerful enough to discriminate between subpopulations.

Total Genetic Variation and Genetic Differentiation
Mean genetic diversity (H_T) for seven quantitative traits, across all subpopulations, averaged 0.774 and ranged from 0.547 for awn length to 0.849 for number of spikelets per spike (Table 3). The extent of differentiation between the seven subpopulations for these quantitative traits (G_ST) implies that, on average, 65% of the total variation for these quantitative traits is partitioned within subpopulations, with a minimum of 50.7% for spike weight to a maximum of 79.3% for spike length.

View larger version (27K): [in this window] [in a new window]   Fig. 2. Principal component analysis (PCA) plot for seven quantitative traits measured on accessions of seven subpopulations of the Batini barley landrace from Oman.

View larger version (21K): [in this window] [in a new window]   Fig. 3. Principal component analysis (PCA) plot for seven subpopulations of the Batini barley landrace from Oman, based on seven quantitative traits.

View larger version (38K): [in this window] [in a new window]   Fig. 4. Correspondence-analysis plot of 16 Awn-, glume-, grain-, and spike-related traits in the Batini barley landrace from Oman. (See Table 1 for abbreviations).

UPGMA cluster analysis and PCoA generated two similar graphs (Fig. 5 and Fig. 6, respectively), each with two sets of subpopulations. UPGMA cluster analysis separated the seven subpopulations into two separate groups at a linkage distance of 0.44. Member subpopulations within each group were also separated at linkage distances ranging from 0.17 to 0.36, with Batini 2 and Batini 3 being the closest, and Batini 4 the farthest. The first axis of the PCoA accounted for 53.54% of total variation and separated Batini 3, Batini 2, Batini 7, and Batini 4, in this order, from the remaining three subpopulations (Fig. 6). The second axis accounted for 18.2% of total variation and separated Batini 1, Batini 3, Batini 2 and Batini 6, in this order, from the remaining three subpopulations.

View larger version (22K): [in this window] [in a new window]   Fig. 5. Unweighted pair group method with arithmetic average (UPGMA) clustering based on squared Euclidean distances among seven subpopulations of the Batini barley landrace from Oman.

View larger version (21K): [in this window] [in a new window]   Fig. 6. Principal coordinates plot of the diversity, based on seven categorical and 19 qualitative traits, for seven subpopulations of the Batini barley landrace from Oman.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

Most likely, the Batini landrace evolved over many generations through the long process of natural and conscious selection (Harlan, 1968), in a multitude of local irrigated farming systems involving date palm (Phoenix dactylifera L.), alfalfa, and vegetable crops (Guarino, 1989, 1990). It is highly likely that desirable plant traits, conducive to development and survival under the harsh climatic conditions of the southern parts of the Arabian Peninsula, are available with high frequencies in this landrace.

Phenotypic Diversity Analyses
The analyses described in this paper could have been more informative had the exact geographical source of each original seed sample been known, and had individual spike or plant samples been kept separate (DeLacy et al., 2000). Although not all qualitative and quantitative traits in this study are of direct agronomic importance, most (seven quantitative traits and at least six qualitative traits) are typically used in germplasm characterization for useful attributes (Koebner et al., 2002). In addition, these morphological traits have been chosen on the basis of their usefulness for distinctness testing in barley (Lakew et al., 1997).

The high level of phenotypic diversity in this single landrace (H' = 0.501) is comparable to, or higher than, phenotypic diversity levels reported for larger germplasm collections of landraces in the barley primary (e.g., Brown and Munday, 1982; Jaradat et al., 1987; Parzies et al., 2000a) and secondary (e.g., Demissie and Bjornstad, 1996) centers of genetic diversity. Brown and Munday (1982) reported a very low average phenotypic diversity index (H' = 0.156) for the same phenotypic loci as listed in Table 7 in 12 landrace populations, presumably because of low variability for two phenotypic markers (vv and B-), as compared with 0.287 in 11 Syrian (Parzies et al., 2000a), and with 0.534 on the basis of nine phenotypic markers in Ethiopian (Kebebew et al., 2001) landraces. However, it was argued (Bjornstad et al., 1997) that phenotypic diversity does not reflect a random and chromosomally balanced sample of genetic variation. Therefore, this high phenotypic diversity may not reflect a higher average diversity at biochemical or molecular levels (Lefebvre et al., 1993). Nevertheless, the "functional genetic diversity" (i.e., based on morphological traits deliberately targeted for selection by farmers) (Koebner et al., 2002), and population differentiation (based on phenotypic markers) were found to be higher when compared with biochemical markers in cultivated barley (Jaradat et al., 1987), Plantago (Wolff, 1991), wheat (Tsegaye et al., 1996), and sorghum (Dje' et al., 1998). Moreover, Donini et al. (2000) and Koebner et al. (2002) found, in a detailed study of genetic diversity in wheat and barley varieties released in the UK during the 20th century, that the average diversity per trait for four biochemical and molecular traits was only 20 to 70% of the average diversity per morphological trait; these differences were attributed to the multigenic nature of most individual phenotypic markers, and hence variation at more than one locus is being analyzed.

The simply inherited morphological traits (Table 1) have been used to assess genetic diversity in barley and for testing the distinctiveness and uniformity of barley landraces (e.g., Brown and Munday, 1982; Alemayehu and Parlevliet, 1997; Kebebew et al., 2001) and germplasm collections (Allard, 1992). Allelic frequencies, at least for four traits (Table 7), were assumed to be equal to phenotypic frequencies based on results of parallel analysis of multilocus outcrossing rates in two barley landrace populations from Syria (Parzies et al., 2000b).

Selection by farmers and natural selection may have resulted in the high frequency of certain phenotypes adapted to the prevailing climatic and edaphic conditions (Ceccarelli et al., 1987), with farmers actively selecting for certain phenological (e.g., earliness) and agronomic (e.g., plant height, tillering capacity, grain color, grain yield and forage yield) traits (Kebebew et al., 2001).

Farmers distinguish the barley varieties by phenotypic markers, such as spike length, spike density, and spike attitude (Negassa, 1985b). The overall shape of the spike is determined, in part, by the spacing of the mature kernels on the rachis, which in turn, is greatly influenced by the rachis internode. All spike-related traits (i.e., attitude, shape, density, length, and rachis internode as expressed by number of spikelets per spike) displayed high levels of phenotypic diversity and were instrumental in separating subpopulations in multivariate analyses (Table 9, Fig. 3, 5, and 6).

Grain color is the most important selection criterion used by farmers to classify barley landraces and varieties (Ceccarelli et al., 1987; Parzies et al., 2000a; Kebebew et al., 2001). Demissie and Bjornstad (1996) suggested that human selection against pigmented barley strains might be responsible for the high frequency for the white kernel phenotype. A high number of functional phenotypes (3.6 or 90% of observed number), a high phenotypic diversity index (0.732), and the high variance component due to differences among subpopulations (65.8%) all point to the high level of variation for this trait in the Batini landrace.

Rachilla length is considered (Nilan, 1964) as one of the best taxonomic characteristics in barley; however, its adaptive significance is not yet known (Kebebew et al., 2001). Apparently (Negassa, 1985b), there is no selection pressure by farmers targeting this trait, as the short phenotype maintained a high frequency (about 70%) in the Batini and other barley landraces from Ethiopia (Demissie and Bjornstad, 1996). Both phenotypes (short and long) were represented in all subpopulations; however, subpopulation four has a higher than average frequency of the short (ss) phenotype (67.4%).

Glume hairiness is a heritable trait and was found to be associated with resistance to karnal bunt [caused by Tilletia indica Mitra = Neovossia indica (M. Mitra) Mundk.] (Warham, 1988) and powdery mildew [caused by Erysiphe graminis DC. f. sp. hordei Em. Marchal = Blumeria graminis (DC.) E. O. Speer] (Negassa, 1985a). It appeared with high frequencies (65–70%) in two subpopulations (Batini 1 and Batini 6) in this study and in most Ethiopian barley landraces (Kebebew et al., 2001). Undoubtedly, farmers have used it to identify and classify barley landrace varieties. It displayed a relatively low phenotypic diversity index in this study (0.356) because of the high frequency of the hairy phenotype in two subpopulations.

Diversity among and within Subpopulations
Lakew et al. (1997) remarked that variation for genotypic and phenotypic markers among plant populations can be from nine to 40 times higher than the variation within populations, depending on the trait and location of testing. Given the breeding system of barley (Hamrick and Godt, 1997), genetic variation is expected to be higher among than within populations. Several studies on barley landraces (e.g., Brown and Munday, 1982; Parzies et al., 2000a) indicated that 40 to 50% and 50 to 60% of the total genetic variation captured resides among and within landraces, respectively. In comparison, wheat (Donini et al., 2000) and barley (Koebner et al., 2002) varieties grown in the UK during the 20th century were monitored for their temporal flux of diversity in morphological and molecular traits. These researchers found that, in wheat, 10 and 90% of the total variation was partitioned among and within varieties, respectively; the respective values for barley varieties were 15.1 and 84.9%.

In the current study, average H_T and G_ST estimates for all traits combined are in line with values quoted for barley landraces, with 40.5 and 59.5% of total genetic variation being partitioned among and within subpopulations, respectively. In comparison, 12 Iranian (Brown and Munday, 1982), two Syrian (Parzies et al., 2000a), and 28 Syrian and Jordanian (Ceccarelli et al., 1987) barley landraces partitioned 55, and 45%, 55.3 and 44.7%, and 40.5 and 59.5% of total variation for variable numbers of phenotypic markers among and within populations, respectively. On the other hand, H_T and G_ST estimates deduced from data on six morphological traits (row type, caryopsis cover in addition to the four phenotypic markers listed in Table 7) and 51 accessions of Ethiopian barley landraces (Demissie and Bjornstad, 1996) were 0.572 and 0.428, respectively.

It was argued (Koebner et al., 2002) that morphology-based G_ST values are more likely to generate a biased picture of diversity trends than the biochemical and molecular marker-based G_ST values because phenotypic markers can be deliberately targeted for selection by farmers. Nonetheless, subsistence farmers have to keep seed of their barley crop for the next year at a seeding rate of 50 to 120 kg ha^–1 (Alemayehu and Parlevliet, 1997). They may not exercise conscious selection of specific plant type(s), and consequently seed of a large number of single, and highly variable, plants is used to grow the next crop. The exceptionally high within-landrace variation levels, found in this and other studies, is maintained, if not maximized, by a combination of genetic, environmental, and management factors.

Parzies et al. (2000a) observed a gradual and significant decline in average gene diversity, number of alleles per locus, and percentage of polymorphic loci, with a concomitant increase in genetic differentiation in barley landraces as time in ex situ storage increased. Consequently, the best protection against the loss of within landrace variation would be in situ conservation (Swanson, 1996) and the collection and maintenance of individual plants within a landrace.

Log-Linear Analysis
The pronounced nonrandom associations among, at least, four spike traits controlled by single genes (Table 7) in the whole collection may reflect either farmers' conscious selection, or an association between phenotype and utility of the grain (Kebebew et al., 2001). The persistence, during the adaptive development of landraces, of these and similar associations reported on barley landraces from Iran (Brown and Munday, 1982), Ethiopia (Asfaw, 1989; Demissie and Bjornstad, 1996), and Syria (Parzies et al., 2000a), could be due to an adaptive response to similar environmental conditions, but, nevertheless, molded by human selection (Ford-Lloyed et al., 2001).

All U values for the two-way and three-way contingency tables were significant (Table 8), implying that the nonrandom associations among alleles at these loci, generally, have a hierarchical pattern. Strong pairwise and three-way associations between all possible gene pairs may suggest that the pairwise associations are contingent on the presence of alleles at the third locus (Li and Rutger, 2000).

Multivariate Graphical Analyses
Multivariate analyses when applied to the set of 26 quantitative and qualitative traits of the Omani barley landrace were successful in discriminating among different subpopulations, and in discerning groups of quantitative and qualitative traits with similar loadings or discriminating power among these subpopulations.

PCA, when applied to single or multiple subpopulations, in two principal components, accounted for most (79.66 and 82.38%, respectively) of total variation. With 71.75% of the variation accounted for by the first two axes in the PCoA, the separation of the seven subpopulations was possible on the basis of multiple qualitative and quantitative traits. PCA operated on the correlation matrix between the seven quantitative variables, and determined the linear combinations that accounted for the greatest proportion of the variation in the data set. PCoA, on the other hand, operated on the similarity matrix or distance matrix between subpopulations, and determined the coordinate vectors that accounted for the greatest proportion of the total distance between subpopulations (Busso et al., 2000). PCoA reduced the multidimensionality of the data to a form where qualitative differences in diversity among subpopulations can be visualized. PCoA results confirmed those obtained by UPGMA cluster analysis. Several studies identified similar trait associations of spike- and seed-related morphological markers in factor (Jaradat et al., 1987) and principal component (Ruiz et al., 1997) analyses.

	CONCLUSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Landrace cultivation has been discouraged in many developing countries because of low yield potential and susceptibility to diseases. Although exotic barley cultivars out-yield local landraces under good management practices in selected testing sites (Ceccarelli et al., 1987), local landraces usually out-yield the exotic material under the low input conditions that predominate in subsistence farming systems. For such conditions (Lakew et al., 1997), native germplasm should be exploited to improve productivity.

The phenotypic and statistical evidences reported in this study indicate that farmers' selection for desirable agronomic traits is a major force shaping the dynamics of crop plant populations in subsistence agriculture in many parts of the developing world. Consequently, the in situ conservation of these landraces ensures the continuation of this dynamic process.

Received for publication April 16, 2003.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 

• Agresti, A. 1990. Categorical data analysis. Wiley International, New York.
• Alemayehu, F., and J.E. Parlevliet. 1997. Variation between and within Ethiopian barley landraces. Euphytica 94:183–189.
• Allard, R.W. 1992. Predictive methods for germplasm identification. p. 119–133. In H.T. Stalker and J.P. Murphy (ed.) Plant breeding in the 1990s. CAB International, Wallingford, UK.
• Asfaw, Z. 1989. Relationship between spike morphology, hordeins and altitude within Ethiopian barley, Hordeum vulgare L. (Poaceae). Hereditas 110:203–209.
• Bjornstad, A., A. Demissie, A. Kilian, and A. Kleinhofs. 1997. The distinctness and diversity of Ethiopian barleys. Theor. Appl. Genet. 94:514–521.
• Brown, A.H.D., and J. Munday. 1982. Population genetic structure and optimal sampling of landraces of barley from Iran. Genetica (The Hague) 58:85–96.
• Brush, S.B. 1995. In situ conservation of landraces in center of crop diversity. Crop Sci. 35:346–354.[Abstract/Free Full Text]
• Busso, C.S., K.M. Devos, G. Ross, M. Mortimore, W.M. Adams, M.J. Ambrose, S. Alldrick, and M.D. Gale. 2000. Genetic diversity within and among landraces of pearl millet (Pennisetum glaucum) under farmer management in West Africa. Genet. Res. Crop Evol. 47:561–568.
• Ceccarelli, S., S. Grando, and J.A.G. van Leur. 1987. Genetic diversity in barley landraces from Syria and Jordan. Euphytica 98:269–280.
• Chapman, C.G.D. 1985. The genetic resources of wheat. A survey and strategy for collecting. International Board for Plant Genetic Resources, Rome.
• Cromwell, E. 1996. Governments, farmers and seeds in changing Africa. Overseas Development Institute with CAB International, Wallingford, Oxon, UK.
• Dje', Y., M. Ater, C. Lefebvre, and X. Vekemans. 1998. Patterns of morphological and allozyme variation in sorghum landraces of Northwestern Morocco. Genet. Res. Crop Evol. 45:541–548.
• DeLacy, I.H., B. Skovmand, and J. Huerta. 2000. Characterization of Mexican wheat landraces using agronomically useful attributes. Genet. Res. Crop Evol. 47:591–602.
• Demissie, A., and A. Bjornstad. 1996. Phenotypic diversity of Ethiopian barleys in relation to geographical regions, altitudinal range, and agro-ecological zones: As an aid to germplasm collection and conservation strategy. Hereditas 124:17–29.
• Donini, P., J.R. Law, R.M.D. Koebner, and J.C. Reeves. 2000. Temporal trends in the diversity of UK wheat. Theor. Appl. Genet. 100:912–917.
• Ford-Lloyed, B.V., H.J. Newbury, M.T. Jackson, and P.S. Virk. 2001. Genetic basis for co-adapted gene complexes in rice (Oryza sativa L.) landraces. Heredity 87:530–536.[ISI][Medline]
• Gower, J.C. 1966. Some distance properties of latent roots and vector methods used in multivariate analysis. Biometrika 53:325–338.[Abstract/Free Full Text]
• Guarino, L. 1989. Barley collecting in southern Arabia. FAO/IBPGR Plant Genet. Resour. Newsl. 73:34–36.
• Guarino, L. 1990. Crop collecting in the Sultanate of Oman in the context of the Arabian Peninsula. FAO/IBPGR Plant Genet. Resour. Newsl. 77:27–33.
• Hammer, K., A. Diederichsen, and M. Spahillar. 1999. Basic studies toward strategies for conservation of plant genetic resources. p. 29–33. In J. Serwinski and I. Faberova (ed.) Proc. technical meeting on the methodology of the FAO World Information and Early Warning System on Plant Genetic Resources. FAO, Rome.
• Hammer, K., H. Knupffer, L. Xhuveli, and P. Perrino. 1996. Estimating genetic erosion in landraces—Two case studies. Genet. Res. Crop Evol. 43:329–336.
• Hamrick, J.L., and M.J.W. Godt. 1997. Allozyme diversity in cultivated crops. Crop Sci. 37:172–176.[Abstract/Free Full Text]
• Harlan, J.R. 1968. On the origin of barley. In Barley origin: Botany, culture, winter hardness, genetics, utilization, pests. USDA-ARS Agric. Handb. 338. U.S Gov. Print. Office, Washington, DC.
• Hutchison, K. 1970. A test for comparing diversities based on the Shannon formula. J. Theor. Biol. 29:151–154.[ISI][Medline]
• International Center for Biosaline Agriculture. 2000. Annual report for 2000. ICBA, Dubai, UAE.
• Jaradat, A.A., T. Jaradat, S. Jana, and J.P. Srivastava. 1987. Diversity for quantitative characters in Jordanian landraces of barley. Barley Genet. V:109–117.
• Kebebew, F., Y. Tsehaye, and T. McNeilly. 2001. Morphological and farmers cognitive diversity of barley (Hordeum vulgare L. [Poaceae]) at Bale and North Shewa of Ethiopia. Genet. Res. Crop Evol. 48:1–10.
• Kefyalew, T., H. Tefera, A. Kebebew, and M. Ayele. 2000. Phenotypic diversity for qualitative and phenological characters in germplasm collections of tef (Eragrostis tef). Genet. Res. Crop Evol. 47:73–80.
• Koebner, R.M.D., P. Donini, J.C. Reeves, R.J. Cooke, and J.R. Law. 2002. Temporal flux in the morphological and molecular diversity of UK barley. Theor. Appl. Genet. 106:550–558.
• Lakew, B., Y. Semeane, F. Alemayehu, H. Gebre, S. Grando, J.A.G. van Leur, and S. Ceccarelli. 1997. Exploiting the diversity of barley landraces in Ethiopia. Genet. Res. Crop Evol. 44:109–116.
• Lefebvre, V., A. Palloix, and M. Reives. 1993. Nuclear RFLP between pepper cultivars (Capsicum annuum L.). Euphytica 71:189–199.
• Li, Z., and J.N. Rutger. 2000. Geographic distribution and multilocus organization of isozyme variation in rice (Oryza sativa L.). Theor. Appl. Genet. 101:379–387.
• National Plant Germplasm System. 2003. Germplasm Resources Information Network (GRIN). Database management Unit (DBMU), NPGS, USDA, Beltsville, MD.
• Negassa, M. 1985a. Geographical distribution and genotypic diversity of resistance to powdery mildew of barley in Ethiopia. Hereditas 102:113–121.
• Negassa, M. 1985b. Pattern of phenotypic diversity in an Ethiopian barley collection, and the Arusi-Bale Highlands a center of diversity of barley. Hereditas 102:139–150.
• Nei, E. 1973. Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci. (USA) 70:3321–3323.[Abstract/Free Full Text]
• Nei, E. 1987. Molecular evolutionary genetics. Columbia University Press, New York.
• Nevo, E. 1992. Origin, evolution, population genetics and resources of wild barley, Hordeum spontaneum, in the Fertile Crescent. p. 19–43. In P.R. Shewry (ed.) Barley: Genetics, biochemistry, molecular biology and biotechnology. CAB International, Wallingford, UK.
• Nilan, R.A. 1964. The cytology and genetics of barley. Res. Studies 32(1), Washington State University, Pullman, WA.
• Papa, R., G. Attene, G. Barcaccia, A. Ohgata, and T. Konishi. 1998. Genetic diversity in landrace populations of Hordeum vulgare L. from Sardinia, Italy, as revealed by RAPDs, isozymes and morphophenological traits. Plant Breed. 117:523–530.
• Parzies, H.K., W. Spoor, and R.A. Ennos. 2000a. Genetic diversity of barley landrace accessions (Hordeum vulgare ssp. vulgare) conserved for different lengths of time in ex situ gene banks. Heredity 84:476–486.
• Parzies, H.K., W. Spoor, and R.A. Ennos. 2000b. Outcrossing rates of barley landraces from Syria. Plant Breed. 119:520–522.
• Potts, D.T. 1993. Rethinking some aspects of trade in the Arabian Gulf. World Archaeol. 24:423–440.
• Ruiz, M., F. Varela, and J.M. Carrilo. 1997. Analysis of the discriminating power of agro/morphological and biochemical descriptors in a sample of Spanish collection of barley (Hordeum vulgare L.). Genet. Res. Crop Evol. 44:247–255.
• Sneath, P.H.A., and R.R. Sokal. 1973. Numerical taxonomy. W.H. Freemen, San Francisco.
• StatSoft, Inc. 2001. STATISTICA (data analysis software systems), version 6. StatSoft, Inc. (www.statsoft.com), Tulsa, OK.
• Swanson, T. 1996. Global values of biological diversity: The public interest in the conservation of plant genetic resources for agriculture. FAO/IPGRI Plant Genet. Resour. Newsl. 105:1–7.
• SYSTAT Software Inc. 2002. SYSTAT 10.2 Statistics I, p. 665. Systat Software Inc., Richmond, CA.
• Tsegaye, S., T. Tesemma, and G. Belay. 1996. Relationships among tetraploid wheat (Triticum turgidum L.) landrace populations revealed by isozyme markers and agronomic traits. Theor. Appl. Genet. 93:600–605.
• Warham, E.J. 1988. Screening for karnal bunt (Tilletia indica) resistance in wheat, triticale, rye, and barley. Can. J. Plant Pathol. 10:57–70.
• Wolff, K. 1991. Analysis of allozyme variability in three Plantago species and a comparison to morphological variability. Theor. Appl. Genet. 81:119–126.
• Willcox, G., and M. Tengberg. 1995. Preliminary report on the archaeobotanical investigations at Tell Abraq with special attention to the chaff impressions in mud brick. Arab. Archaeol. Epigraphy 6:129–138.
• Yeh, F.C., R.-C. Yang, and T.B.J. Boyle. Z.–H., Ye, and J.X. Mao. 2000. POPGENE, the user-friendly shareware for population genetic analysis. Molecular Biology and Biotechnology Center, Univ. of Alberta, Edmonton AB.
• Zar, J.H. 1996. Biostatistical analysis. 3rd ed. Prentice Hall, Engelwood Cliffs, NJ.
• Zhang, Q., and R.W. Allard. 1986. Sampling variance of the genetic diversity index. J. Hered. 77:54–55.[Abstract/Free Full Text]
• Zohary, D., and M. Hopf. 1993. Domestication of plants in the Old World: The origin and spread of cultivated plants in West Asia, Europe, and the Nile Valley. Clarendon Press, Oxford, England.

Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2004 44: 1-4. [Full Text]  

This article has been cited by other articles:

				A. A. Jaradat, M. Shahid, and A. Al-Maskri Genetic Diversity in the Batini Barley Landrace from Oman: II. Response to Salinity Stress Crop Sci., May 1, 2004; 44(3): 997 - 1007. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in Crop Science Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Search for Related Content PubMed Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Agricola Articles by Jaradat, A. A. Articles by Al Maskri, A. Y. Related Collections Germplasm Enhancement Crop Genetics