Restriction fragment length polymorphisms in landrace barleys from Ethiopia in relation to geographic, altitude and agro-ecological factors.
In barley, RFLPs have provided a class of very useful markers for assessing genetic variation of populations both in nuclear (Saghai Maroof et al., 1984; Zhang et al., 1993) and organellar genomes (Saghai Maroof et al., 1992). Very extensive RFLP based genetic linkage maps have also been constructed for barley (Graner et al., 1991; Heun et al., 1991; Kleinhofs et al., 1993).
Still, most characterization data of barley accessions in genebanks are based on morphometric characters. Knowledge of genetic diversity at the molecular level on landrace materials is important for a more profound characterization. So far data on molecular variation at the DNA level of Ethiopian landrace barley is either lacking or inadequate.
In previous papers, we have discussed the diversity of Ethiopian barley LRs based on morphological or protein markers (Demissie and Bjornstad, 1996, 1997) as well as the RFLP relationships between Ethiopian LRs and those from other origins (Bjornstad et al., 1997). In this paper, we focus on the RFLP diversity in the same LRs using a total of 31 probes (50 probe-enzyme combinations, PECs) representing all barley chromosomes. The RFLP diversity is compared among and within different accessions as well as accessions grouped by geographical origin, altitudinal range, and agroecological zones. Other major objectives are to identify sites of maximum gene diversity and possible sites for in situ conservation.
MATERIALS AND METHODS
The LRs studied are analyzed as Sets 1 and 2, which are partially overlapping. The first consists of 43 LRs, each represented by a single line. The second consists of 65 lines from 14 of these LRs, each represented by five (in some cases only four) lines, including the one in Set 1. All lines were single plant progenies. The total sample was 94 genotypes. (A comprehensive RFLP analysis of the 225 lines available from these 14 populations was considered unfeasible, although a better within-LR sampling might have been desirable. A compromise was struck from considerations of the agronomic differences between the populations.)
The accessions were divided into different groups according to three criteria: geographical origin, altitudinal ranges, and agroecological zone. The altitude range was arbitrarily classified into five altitude classes: [is less than] 2000, 2001-2500, 2501-3000, 3001-3500, and [is greater than] 3501 m. Set 1 consisted of materials collected from 10 geographic regions covering wide agro-ecological zones, while Set 2 only covered four geographical regions and more limited agro-ecological zones. Information on the accessions including region, county, altitude, and agro-ecological zone is detailed in Table 1.
[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]
RFLP Protocol and Selection of Probes
The protocol used for DNA extraction, blotting, and hybridization was as described by Kleinhofs et al. (1993). Thirty-one DNA clones used as probes in this study have been described by Bjornstad et al. (1997). They were selected from map data to provide an even distribution along the seven chromosomes and thus be representative of the overall genomic diversity.
Data Scoring and Analysis
The 50 PECs were scored for presence or absence of bands. Every clearly hybridizing fragment detected by autoradiography was treated as a unit character. Independent scoring was carried out by two persons, and bands that appeared dubious were deleted.
Genetic similarity (F) between genotypes (or "operational taxonomic units", OTUs) x and y was estimated following the measure suggested by Nei and Li (1979), i.e., F = 2[M.sub.XY]/[M.sub.X] + My (Eq. [5.53] of Nei, 1987), where [M.sub.XY] was the number of bands shared between OTUs and [M.sub.X] and My were the number of bands in each. The F was then used to estimate genetic distance d (by Eq. [5.55] of Nei, 1987) for any pair of OTUs. The "corrected" F and d values were used, subtracting the within-OTU diversity from the between-OTU comparisons.
Matrices for individual genotypes, populations, altitudinal classes, regions, and agro-ecological zones were obtained using a computer program (RESTSITE) developed by Miller (1990). The F value within each classification variable was used as a within-region diversity measure. Standard errors for F and d were obtained by the Jack-knife method. A dendrogram based on estimates of genetic distance was generated by the UP-GMA clustering procedure.
Levels of Probe Polymorphism
A total of 153 DNA fragment bands were identified using 51 probe-enzyme combinations (PECs). Forty-two PECs (84%) detected polymorphism; however, the level of polymorphism was 62% when the 95% criterion of polymorphism was applied. Closer study revealed considerable variation in the levels of polymorphism among the seven chromosomes. Chromosomes 3, 6 (especially the Nar1, nitrate reductase probe), and 7 appeared the most polymorphic. In contrast, chromosomes 1 and 4 were less polymorphic. Chromosome 5 displayed a variable pattern. The 5M (long arm) probes were monomorphic, while the P (short) arm probes were very polymorphic, corresponding with the previously reported diversity in the Hor-1 and Hor-2 loci (Demissie and Bjornstad, 1997) and the Ml-a locus (Negassa 1985a), all on this chromosome arm.
Genetic Diversity between and within Landraces
The F (genetic similarity) values showed that all 43 lines in Set I represented unique genotypes. The distance (d) values ranged from 0.001 [+ or -] 0.001 to 0.021 [+ or -] 0.005, the former between accessions '1837' and '1621' and the latter between the pairs '1651' vs. '1622' and '3302' vs. '3304' (Table 2). Accessions 1622 and 3302 showed unique restriction fragment patterns for many probes, indicating their relative divergence from the rest of the genotypes at these loci.
From a cluster analysis based on the data from Set 1 (not shown, because of lack of space), two major clusters of LRs were observed. The first cluster included 1622 and 3302, as well as '3291'. The second major cluster displayed a complex pattern. In general, no coherent groupings that matched the major classification variables could be perceived.
The inclusion of within-LR diversity in Set 2 altered the relationships between the LRs common to both sets. Data on genetic similarity (F) and distance (d) for a sample of the LRs included in Sets 1 or 2, are given in Table 2. The respective relationships between the 14 LRs common to both sets are presented in Fig. 1 (Set 1) and Fig. 2 (Set 2). Some conclusions may be drawn.
[Figures 1-2 ILLUSTRATION OMITTED]
First, the two dendrograms display sharp differences both in levels of genetic distance and in LR groupings, showing the importance of sampling within-LR variability. Figure 2 shows that except for the clear distinctness of the LR '3479' from Gojam, many of the distances are rather small--in the case of 3302 vs. '3304', the LRs are identical.
Second, in many cases the differences between LRs (d-estimates) in Set 1 are two to five times greater than those in Set 2 (Table 2). The maximum d in Set 1 was 0.021 (1622 vs. 1651), while in Set 2 it was only 0.005 (3302 vs. 1651). In some cases, these differences were statistically significant (e.g., the comparisons 1622 vs. 1651 or 3302 vs. 3304). However, there were also cases where estimates were identical (1651 vs. 1621). As expected, the standard errors are also greater in Set 1 estimates.
Third, in Set 2 the within-population similarity (F, Table 3) was the least for 1622 followed by 3304 and the highest for 3381 and 1651. The mean F over the 14 populations was 0.918. Thus, the level of polymorphism varies strongly between landraces, some approaching monomorphism, which may explain the sampling effects noted in Table 2. In the case of 3302 vs. 3304, the two LRs both are very variable (see Table 3) and appear very different in Set 1. In RFLP diversity, however, they appear to be duplicates of the same LR sampled twice from the same district of Kofele in Arsi (see Table 1 and Fig. 2). The same applies to 1621 and 1622, both from the district of Basona Werena in Shewa. In the case of 1622, the genotype randomly included in Set 1 turned out to be one of the most unique lines in the entire study. Obviously, the sampling effects associated with representing a LR by one individual may be very strong and may give rise to erroneous conservation strategies.
[Figure 2 ILLUSTRATION OMITTED]
Table 3. Within-population genetic similarity (F) and standard errors ([+ or -]) for the Set 2 maternal (14 populations) calculated from RFLP data of 50 PECs.
Within-pop [+ or -] Standard Accession Region F error Accession 1621 Shewa 0.899 0.028 3302 1622 Shewa 0.837 0.039 3304 1651 Shewa 0.983 0.034 3334 1706 Arsi 0.942 0.038 3371 3232 Gojam 0.951 0.032 3381 3284 Bale 0.924 0.041 3479 3295 Bale 0.943 0.034 50 Within-pop [+ or -] Standard Accession Region F error 1621 Arsi 0.848 0.035 1622 Arsi 0.847 0.042 1651 Arsi 0.901 0.023 1706 Arsi 0.951 0.039 3232 Arsi 0.987 0.033 3284 Gojam 0.918 0.038 3295 Arsi 0.928 0.034
The diversity (F) and distance (d) at the regional level was analyzed by pooling LRs from a given region. Set 1 was used to obtain a relatively balanced sampling within and between regions, and the results are given in Table 4. The within-region diversity showed a wide range--the least for Gojam (0.982 [+ or -] 0.049) and the highest for Arsi (0.871 [+ or -] 0.028). These estimates are of the same order as diversity within single LRs in Set 2 (Table 3).
Also, the genetic distance (d) estimates between regions are rather low, ranging from 0.005 (between Welega and Shewa) to 0.000 (between Arsi and Bale). Based on their DNA polymorphisms, barleys from the Arsi and Bale regions may be considered as one source of germplasm. This is also apparent from the degree of overlap at the LR level (Fig. 2, see Table 1). However, Shewa appears distinct except for the LR 1651 from Basona Werena, which does not group with 1621 and 1622 from the same district (see Tables 1 and 2).
Table 4. Genetic similarity (F, below diagonal) and genetic distance (d, above diagonal), [+ or -] SE, in parentheses, between barley LR accessions from different geographical regions. Along the diagonal (italic type) the within-regional estimates are given.
Arsi Bale Gamugofa Gojam Gonder Hararge Arsi 0.871 - 0.001 0.001 0.001 0.001 0.028 - 0.001 0.001 0.001 0.001 Bale 1.000 0.871 0.001 0.001 0.001 0.002 0.006 0.039 0.001 0.001 0.001 0.001 Gamugofa 0.982 0.982 0.884 0.002 0.001 0.001 0.010 0.015 0.035 0.001 0.001 0.001 Gojam 0.986 0.979 0.965 0.982 0.001 0.003 0.011 0.012 0.019 0.049 0.001 0.001 Gonder 0.986 0.986 0.980 0.979 0.900 0.002 0.008 0.001 0.016 0.013 0.028 0.001 Hararge 0.975 0.974 0.975 0.943 0.974 0.920 0.008 0.013 0.017 0.020 0.017 0.030 Shewa 0.985 0.992 0.970 0.972 0.996 0.976 0.005 0.006 0.015 0.014 0.008 0.011 Welega 0.952 0.932 0.952 0.931 0.909 0.911 0.019 0.022 0.020 0.023 0.027 0.029 Welo 0.976 0.969 0.980 0.948 0.981 0.988 0.008 0.013 0.011 0.019 0.001 0.011 Shewa Welega Welo Arsi 0.001 0.002 0.001 0.000 0.001 0.001 Bale 0.001 0.004 0.002 0.000 0.001 0.001 Gamugofa 0.002 0.003 0.001 0.001 0.001 0.001 Gojam 0.002 0.004 0.003 0.001 0.001 0.001 Gonder 0.000 0.005 0.006 0.000 0.002 0.002 Hararge 0.001 0.005 0.001 0.001 0.002 0.001 Shewa 0.891 0.005 0.001 0.033 0.002 0.000 Welega 0.910 - 0.003 0.025 - 0.001 Welo 0.983 0.946 0.892 0.007 0.021 0.006
At the probe level, certain fragments confined to certain geographical regions could be distinguished. However, more often no regional trend was apparent, as when the same uncommon band occurred in a few LRs from widely different regions.
Set 2 was also analyzed to test the effect of sample size. The within-regional diversity values were remarkably similar. For Arsi they were nearly identical (0.878 [+ or -] 0.029 vs. 0.871 [= or -] 0.028 in Table 4, n = 10 in Set 1 vs. 34 in Set 2) and similar for other regions (Bale 0.911 [+ or -] 0.027, n = 6 vs. 9; Shewa 0.876 [+ or -] 0.027, n = 8 vs. 12). However, Gojam showed a marked difference (0.849 [+ or -] 0.036 in Set 2 vs. 0.982 [+ or -] 0.049 in Set 1), probably an effect of increasing the sample size from three to 10 lines. Thus, an underestimation of within-LR diversity due to sampling only five individuals may have occurred. However, it is significant that the general pattern of regional GS remained largely unchanged despite the numbers of LRs or genotypes assayed in the two data sets.
A further comparison of F and d estimates was carried out with respect to altitudinal distributions of the Set 1 genotypes. In spite of an altitude span from 1650 to 3750 m, the estimated differences between altitude classes were slight: F ranged from 0.968 [+ or -] 0.020 between classes 1 and 5 to 1.00 [+ or -] 0.006 between classes 1 and 2, indicating an apparent lack of genetic differentiation linked to altitude. The least within-altitude class F (0.841 [+ or -] 0.031) was recorded in Class 3 and the highest in Class 5 (0.931 [+ or -] 0.055). However, the latter class represents the extreme margins of barley cultivation and was least well sampled (only three LRs), which may confound the results.
Agro-Ecological Zone Diversity
F values for Set 1 genotypes obtained for each paired comparison of RFLP bands among agro-ecological zones were computed (data not included). Only five zones were considered due to inadequate sample size for all zonal comparisons. The highest F (1.0 [+ or -] 0.007) was recorded between Zones M2 and SH2 and also between SH2 and SM1 while the least (0.973 [+ or -] 0.013) was between SH3 and SM1. In general, genetic differentiation was very slight, reflecting the absence of variations between the zones considered.
Correlations between Diversity Measures
Demissie and Bjornstad (1997) found generally low correlations between morphological diversity (Shannon-Weaver index, H') and Nei's isozyme diversity (H). The present data allow us to include the within-class RFLP similarity (F) values in this comparison. Based on the 14 populations in Table 3, the correlations between H', H, and F were all negative and non-significant (F vs. H r = -0.46, P = 0.10; F vs. H' r = -0.28, P = 0.33; H vs. H' r = -0.32, P = 0.26). (Negative values are expected, since F measures similarity and the other two measure distances.) Certain populations (e.g., 3302) were highly diverse in all parameters, whereas 1651 was practically monomorphic in isozymes and RFLPs, but morphologically quite diverse.
At the regional level our previous study found a positive, but non-significant correlation between H and H'. None of them were significantly correlated with Fin our data (F vs. H r = 0.10, P = 0.82; F vs. H' r = -0.40, P = 0.33). Arsi, Bale and Shewa were diverse in all kinds of traits. Welo was diverse only in RFLPs. There the within-region F was as high as that of Shewa, quite unlike the level of morphological variations.
Thus, conclusions drawn from one parameter of diversity to other levels seem very uncertain in our case. A probable cause for such lack of significant correlations in our study may be that RFLPs, isozymes, and morphological marker loci sample different parts of the genome, with very different degrees of polymorphism. In a recent study of rice diversity (Parsons et al., 1997), it was found that RAPDs identified the same relationships between the major classes of cultivated rice as isozymes, but SSRs did not. The authors related this to the chromosomal (centromeric) location of some of the SSR loci. This result parallels our finding that RFLP diversity is strongly dependent on chromosomal location of the probes used. The reason for the good correspondence between isozymes and RAPDs in this rice study, compared with our data, may be that the genetic differences between the rice lines were greater than between the Ethiopian barleys we studied.
How Should LRs Be Represented in a Core Collection?
One purpose of this study was to provide a baseline data set for conservation of the very large number ([is greater than] 12000 accessions) of LR barleys held by the Biodiversity Institute, Ethiopia. One approach to this is the development of a well-characterized `core collection', described by Brown (1989) as a subset composed of approximately 10% of the total collection, but containing [is greater than] 70% of its diversity. The FAO Global Plan of Action for the Conservation and Sustainable Utilization of Plant Genetic Resources for Food and Agriculture recommends the "establishment of genebank-based core collections for key national crop collections in national facilities" (FAO, 1996, p. 38). In connection with the development of a world-wide `Barley Core Collection', it has been suggested (van Hintum, 1994) that a LR be represented by a single fixed genotype. The rationale would be to ensure a constant reference point for future studies.
However, such an approach reflects academic concerns, not those of the germplasm curator or the plant breeder. At least, our data does not support the adoption of such an approach when developing an Ethiopian barley core collection. In a previous paper (Bjornstad et al., 1997), we used Set 1 to describe the general relationships between RFLP diversity in Ethiopian vs. European-American cultivars. For this purpose single LR-lines were adequate. Otherwise, our data support the conclusion by Brown (1995) that the single LR-line core collection would end up conserving genetic stocks, not the biological features of LRs.
Consequences for Sampling and Conservation
For rational sampling and conservation ex situ as well as in situ, a reliable classification of LRs is desirable. Where are the hot spots of diversity, and on which criteria should they be identified? Are RFLP markers appropriate in this context?
The RFLP data confirm the evidence from phenotypic (Negassa, 1985b) and isozyme studies (Bekele, 1983, Demissie and Bjornstad, 1997) that Arsi, Bale, and Shewa are key areas of LR diversity. Still, a striking result of this study is that the genetic distances detected for all the major classification variables were rather small. This supports our previous findings with morphological and protein markers that the population is the main source of variability, with higher classifications being only slightly differentiated with respect to most markers. This has the important practical implication that it may not be necessary to conserve a very large number of landraces to conserve the overall diversity. A well chosen set of well represented populations may be adequate. LRs vary greatly in diversity, but when they are diverse, our data show that they may be as diverse as a similar sized sample of LRs from a whole region. The results by Tsegaye et al. (1996), showing that diversity in Ethiopian durum wheat (Triticum durum Desf.) LRs is also primarily within populations, point to similarities between the two species. The authors ascribed this to a common ancestral population and/or adaptation to similar climatic conditions. The slight degree of overall RFLP diversity of Ethiopian barleys as compared with modem cultivars (Bjornstad et al., 1997) may support this point of view.
The lack of altitudinal differentiation in RFLPs contradicts the clear clines observed in certain morphological markers (Demissie and Bjornstad, 1996) and agronomic traits like BYDV resistance (Qualset, 1975), scald resistance and heading date (Engels, 1994). A lack of associations between isozyme markers and adaptive traits in Ethiopian LRs of durum wheat was also noted by Tsegaye et al. (1996). Such results put in doubt the utility, or at least the sufficiency, of RFLPs or other apparently neutral markers in establishing conservation strategies for Ethiopian barley LRs. For the lines included in Set 2, agronomic data from two locations in the Shewa region are available (B. Lakew, Institute of Agricultural Research, 1992, personal communication). A multivariate analysis, pooling different marker types and adaptive traits, will be undertaken to reveal which loci/traits are most valuable for characterization of LR diversity.
This study was supported by a Norwegian Committee for Development Research and Education (NUFU) scholarship grant and by the Institute for International Development (IIZ), Austria. A special thanks to Drs. Trygve Berg and Cary Fowler of NORAGRIC, Agricultural University of Norway, for facilitating the funding and administrative aspects of the project. The helpful assistance of Professor O. Nissen for data analysis is highly appreciated. The excellent technical assistance of David Kudrna, at Pullman, Washington State University, and Anne Guri Maroy, Agricultural University of Norway, is gratefully acknowledged.
Abbreviations: AFLP, amplified restriction length polymorphism; d, genetic distance; F, genetic similarity; LR, landrace; PECs, probeenzyme combinations; RAPID, random amplified polymorphic DNA; RFLP, restriction fragment length polymorphism; SSR, simple sequence repeats.
Becker, J., and M. Heun. 1994. Barley microsatellites: allele variation and mapping. Plant Mol. Biol. 27:835-845.
Becker, J., P. Vos, M. Kuiper, F. Salamini, and M. Heun. 1995. Combined mapping of AFLP and RFLP markers in barley. Mol. Gen. Genet. 249:65-75.
Beckmann, J.S., and M. Soller. 1983. Restriction fragment length polymorphisms in genetic improvements: methodologies, mapping and costs. Theor. Appl. Genet. 67:35-43.
Bekele, E. 1983. Allozyme genotypic composition and genetic distance between the Ethiopian landrace populations of barley. Hereditas 98:259-267.
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. 1989. Core collections: A practical approach to genetic resources management. Genome 31:818-824.
Brown, A.H.D. 1995. The core collection at the crossroads. In T. Hodgkin et al. (ed.) Core collections of plant genetic resources. Wiley & Sons, London.
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.
Demissie, A., and A. Bjornstad. 1997. Geographical, altitude and agroecological differentiation of isozyme and hordein genotypes of landrace barleys from Ethiopia. Genet. Resources Crop Evol. 44:43-55
Engels, J.M.M. 1994. Genetic diversity in Ethiopian barley in relation to altitude. Genet. Resources Crop Evol. 41:67-73.
FAO. 1996. GLOBAL plan of action for the conservation and sustainable utilization of plant genetic resources for food and agriculture. FAO, Rome.
Graner, A., A. Jahoor, J. Schondelmaier, S. Siedler, K. Pillen, G. Wenzel, R.G. Hermann. 1990. Construction of an RFLP map for barley. Theor. Appl. Genet. 83:250-56.
Heun, M., A.E. Kennedy, J.A. Anderson, N.L.V. Lapitan, M.E. Sorrels, and S.D. Tanksley. 1991. Construction of a restriction fragment length polymorphism map for barley (Hordeum vulgare). Genome 34:437--447.
van Hintum, T. 1994. Drowning in the gene pool. Ph.D. Dissertation, Swedish University of Agricultural Sciences, Umea.
Kleinhofs, A., A., Kilian, M.A. Saghai Maroof, R.M. Biyashev, P. Hayes, F.Q. Chen, N. Lapitan, Fenwick, T.K. Blake, V. Kanazin, E. Ananiev, L. Dahleen, D. Kudrna, J. Bollinger, S.J. Knapp, B. Liu, M. Sorrels, M. Heun, J.D. Franckowiak, D. Hoffman, R. Skadsen, and B.J. Steffensen. 1993. A molecular, isozyme and morphological map of barley (Hordeum vulgare) genome. Theor. Appl. Genet. 86:705-712.
Miller, C.J. 1990. Programs for analyzing restriction site or fragment data. Dep. of Agronomy, Univ. of Wisconsin, Madison.
Negassa, M. 1985a. Genetics of resistance to powdery mildew in some Ethiopian barleys. Hereditas 102:123-138.
Negassa, M. 1985b. Pattern of phenotypic diversity in an Ethiopian barley collection, and the Arsi-Bale Highland as a center of origin of barley. Hereditas 102:139-150.
Nei, M. 1987. Molecular evolutionary genetics. Columbia University Press, New York.
Nei, M., and W. Li. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. (USA) 76:5269-5273.
Parsons, B.J., H.J. Newbury, M.T. Jackson, and B.V. Ford-Lloyd. 1997. Contrasting genetic diversity relationships are revealed in rice (Oryza sativa L.) using different marker types. Mol. Breed. 3:115-125.
Qualset, C.O. 1975. Sampling germplasm in a center of diversity: An example of disease resistance in Ethiopian barley. p. 81-96. In O.H. Frankel et al. (ed.) Crop genetic resources for today and tomorrow. Cambridge University Press, Cambridge, England.
Saghai Maroof, M.A., K.M. Soliman, R.A Jorgensen, and R.W. Allard. 1984. Ribosomal spacer length polymorphism in barley: Mendelian inheritance, chromosomal location and population dynamics. Proc. Natl. Acad. Sci. (USA) 81:8014-8018.
Saghai Maroof, M.A., Q. Zhang, D.B. Neale, and R.W. Allard. 1992. Association between nuclear loci and chloroplast DNA genotypes in wild barley. Genetics 131:225-231.
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.
Williams, J.G., A.R. Kubelik, K.J. Livak, J.A. Rafalski, and S.V. Tingey. 1990. DNA polymorphism amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531-6535.
Zhang, Q., M.A. Saghai Maroof, and A. Kleinhofs. 1993. Comparative diversity analysis of RFLPs and isozymes within and among populations of Hordeum vulgare ssp. spontaneum. Genetics 134:909-916.
Abebe Demissie,(*) Asmund Bjornstad, and Andris Kleinhofs
A. Demissie, Biodiversity Institute of Ethiopia, P.O. Box 30726, Addis Ababa, Ethiopia, and Agricultural Univ. of Norway, Dep. of Horticulture and Crop Sciences, N-1432, P.O. Box 5022, Aas, Norway; A. Bjornstad, Agricultural Univ. of Norway, Dep. of Horticulture and Crop Sciences, N-1432, P.O. Box 5022, Aas, Norway; A. Kleinhofs, Dep. of Crop and Soil Sciences, Washington State Univ., Pullman, WA 99163-6420. Received 15 April 1997. (*)Corresponding author.
Published in Crop Sci. 38:237-243 (1998).
|Printer friendly Cite/link Email Feedback|
|Author:||Demissie, Abebe; Bjornstad, Asmund; Kleinhofs, Andris|
|Date:||Jan 1, 1998|
|Previous Article:||Chromosomal location of genes for resistance to Karnal bunt in wheat.|
|Next Article:||Predicting progeny variance from parental divergence in hard red spring wheat.|