Search

Bekele, E. 1984. Relationships between morphological variance. gene diversity and flavonoid patterns in the land race populations of Ethiopian barley. —Hereditas 100: 271–294. Lund, Sweden. ISSN 0018–0661. Received February 7, 1984 An argument is forwarded that in addition to multiple genetic factors, single gene mutation with major effect could be very important in the evolution of natural populations and domesticates. The effect of null alleles and correlations between traits determined by major and minor genes are suggested to show different strategies of adaptation among different kinds of organisms. Sturtevant's hybrid dysgenesis could be the genetic reasons for Vavilov's emancipation of recessives. The three linked esterase loci are assumed to be the results of some form of gene duplication. The genetic backgrounds of the gene are reasons for unspecific polytype correlations between all states of genes and ecological variables. The rank correlation of the percentage of the contribution to the total variation by the differences between individuals in each population between enzymatic and morphological variation was not significant except for row number and Est-A locus. Diversity indices of morphological variables and enzymes are differently correlated in different regions. Enzyme variation and quantitative characters are positively correlated in some regions and negatively in others. But negative correlations are mainly concentrated into three regions. This, together with the result of kurtosis and skewness of quantitative characters are taken to be the results of multidirectional selection schemes. Correlations of quantitative characters showed an overall higher occurrence of negative correlations between plant height and number of sterile florets. Regional cluster differences based on genetic distance measures of enzymatic data and morphological data are perhaps due to differences in the fitness potential that can be related to particular allelic combinations at one level and more directly, to phenotypic properties at another. Enzyme variations and frequencies of flavonoid patterns are negatively correlated. The eleven frequent alleles that were correlated with means of six quantitative characters showed both negative and positive correlations with uneven distribution over regions. Flavonoid C pattern is negatively correlated with a number of sterile florets and positively correlated with a number of seeds.

Evolutionary approaches in gene conservation are emphasized to be faced regardless of their complexities and immediate economic return. Such emphasis is mainly based on recovery of genes eroded, endemic balance of pathogen outbreak, and on provision of information on how a population of plants becomes adapted to different regions. However, the way to continue conservation as an evolutionary process in terms of its long-term effects is not yet clarified. Measures of genetic variability not only within and across the species but also within and across phyletic lines are indicated for several ways of gene conservation and utilization of both intraspecific and interspecific variations of the various orders of gene pools. The patterns of variation of Ethiopian land race populations of Hordeum vulgare, Triticum turgidum and T. aestivum that were previously studied and reported are summarized. Hierarchical approaches to quantitatively define a center of genetic diversity into microcenters are suggested to be made among different species and among different pathogens, pests as well as parasites. The complexity of interacting genotypes in the composite populations of land races is discussed. The frequency and performance of the various genotypes of these land races varying across a given ecological gradient due to competition and/or coadaptation and being moulded by the specific effect of the environmental factors and multilocus genetic organization is presumed. The problems associated in selecting high yielding and other agronomically promising land races are pointed out. The regionally different multilocus structures and genotypic compositions, if taken with their respective possibly defined multilocus structure of parasites, pests and pathogens, will be of high interest in any selection schemes in each micro-and megacenter of the regions studied. The problems of sampling for gene conservation works are presented. Some solutions are given. Three versions of sampling problems for gene conservation studies are discussed. Solutions for two versions based on sampling one genotype at a time, and several genotypes at a time, are presented.

Bekele, E. 1984. Analysis of regional patterns of phenotypic diversity in the Ethiopian tetraploid and hexaploid wheats. —Hereditas 100:131–154. Lund, Sweden. ISSN 0018–0661. Received October 25, 1983 153 populations of tetraploid and 72 populations of hexaploid wheats collected in Ethiopia from 20 and 17 regions, respectively, were grown in a greenhouse, and 6331 and 2700 individuals of tetraploids and hexaploids were studied for basal spikelet fertility, awn colour, glume colour, glume hairiness, kernel colour, kernel weight, number of tillers, plant height, spike length, number of spikelets per spike, and number of seeds per spike. Analyses of regional patterns of character variation to determine the relative contribution of various regions to germplasm resources showed differences in the level of importance of useful germplasm material. Polymorphism of characters studied is higher in tetraploids. The contribution to the total variation varies among characters. In both tetraploids and hexaploids the contribution to the total phenotypic diversity is highest at the lowest level (within localities), followed by the differences among populations in a region, and among regions. Both tetraploids and hexaploids are assumed to equally rely on phenotypic plasticity in adapting to the multiplicity of environments, and TWINSPAN analysis showed differences in the frequencies of characters prefemng negative and positive groups differentially among the various regions. Combinations of characters showed dissimilarity among regions and between ploidy levels. This may mean particular adaptive strategies through the effect of correlated characters.

158 land race populations of cultivated barley, represented by 3266 individuals from 19 diverse regions composed of 72 areas, were assayed electrophoretically for 5 enzyme loci (Est-A, Est-B, Est-C, Est-D and Acph-·l) to determine isozyme variability within and among populations. Differences in frequency of isozyme phenotypes were observed among populations within regions and between regions. The number of allozymes detected in the five loci studied were 5 in Est-A, 4 in Est-B, 6 in Est-C, 5 in Est-D and 6 in Acph-·l locus. The supply of alleles varied from region to region, from area to area and from locality to locality. Populations in the same region differed in their allelic composition. The highest number of alleles of the five loci studied detected in a region was 22 and the lowest, 7. A high frequency of null alleles at the Est-B locus was observed in two regions. 44.85% of the total gene diversity was found to be due to genetic variation within localities, 20.13% was due to variation between localities within areas, 18.71% was due to variation between areas within regions, while 16.31% of the total gene diversity was due to genetic variation between regions. The magnitude of the contribution of different components of the hierarchy to the total gene diversity varied from region to region. No correspondence was found between geographical distance and heterozygosity. For the genetic diversity computed Nei's and Gregorius' logics were not found to give a synonymous result. A significant heterogenous distribution of polymorphic events over regions was observed.

Amplified fragment length polymorphism (AFLP) and random amplified polymorphic DNA (RAPD) markers were used to provide estimates of the comparative genetic variation within and among populations of various Guizotia taxa with the goal of conserving and utilizing their genetic diversity. The percentage of polymorphic loci (PS) ranged from 28.5% � 90% (AFLP) and 85.6% � 99.6% (RAPD). The overall gene diversity estimate ( /H T ) has shown slight variation among taxa ranging from 0.32 � 0.37 (AFLP) and from 0.22 to 0.28 (RAPD). The within population diversity of ‘‘Chelelu’’ and ‘‘Ketcha’’ was found to be unexpectedly high. Both parameters used to estimate population differentiation (GST and FST ) revealed the highest population differentiation in G. zavattarii followed by G. arborescens. Genetic variation among populations within a taxon was highly significant for all the five taxa as revealed by AMOVA (PB0.0001). The need for immediate conservation activities for G. arborescens and G. zavattarii, and factors that contribute to the existing genetic variability and population genetic structures are discussed.

An extraction solution that is optimal for protoplast release in Eragrostis tef (Zucc.) Trotter and Sorghum bicolor (L.) Moench consists of 1% Cellulysin, 0.5 % Macerozyme and 2 % Cellulase Onozuka, 1 mM CaCl2, 5 mM 2, (n-morpholino) ethane sulfonic acid, and 0.4 M mannitol. A culture medium consisting of two parts of Gamborg et al. medium, three parts of Murashige and Skoog medium, 1 % Schenk and Hildebrandt powder vitamin, 1 mg/l matin, 10 mM glutamine, 10 mg/l of 2,4-dichlorophenoxyacetic acid, and 10 mg/l naphthaleneacetic acid, was optimal for the survival and division of a small fraction of cultured protoplasts and protoplast fusion products. Electroporation and heat shock treatments stimulated cell wall formation and cell division in both sorghum and tef. Electrical parameters and conditions for fusing tef and sorgum were determined. Field pulse conditions for eventual electrogene transfer in tef and sorghum consisted of 700 and 400 V/cm, respectively, at 20°C and 40μs field pulse duration.

In vitro somatic embryogenesis and plant regeneration from root and leaf explants and from seeds were studied using eight genotypes of Eragrostis tef at various concentrations (1–10 mg/l) of 3,6-dichloromethoxybenzoic acid (3,6-D) and 2,4-dichloromethoxybenzoicacid (2,4-D). Optimal conditions for callus growth, embryogenesis and plant regeneration were evaluated. The callus biomass and the ability to regenerate plants were found to be functions of type of explant and hormone concentration. The callus fresh weight from leaf was higher than that from root, and different genotypes differed in their callus fresh weight. Unlike leaf induced callus, the root induced callus showed no significant differences between genotypes and treatments. The fresh weight of callus from seeds of most genotypes studied was similar. The number of direct regenerants from callus maintained on callus induction medium varied among the genotypes studied. On the whole, the number of regenerants from leaf callus was higher than that from root callus at all tested hormone concentrations. 1 and 2.5 mg/l of 3, 6-D gave the highest numbers of regenerants from calli transferred to embryogenesis medium. The various tef genotypes differed in their capacity to regenerate plants. Establishment of cell suspension cultures from root and leaf explants was also accomplished.

Albumin, globulin, and prolamin fractions were extracted from seeds of tef (Eragrostis tef (Zucc.) Trotter) and separated using several SDS-PAGE procedures and 1-D isoelectric focusing. This showed that all fractions were highly polymorphic, with differences in the prolarnin fractions allowing the 37 cultivars and lines to be classified into seven groups, each containing between one and fourteen lines. The results also indicate that there is potential to exploit the polymorphism in albumins and globulins in order to identify genotypes with enhanced nutritional quality.

Summary A total of 415 sorghum (Sorghum bicolor (L.) Moench) accessions representing different regions of Ethiopia, Eritrea and a group of introduced lines were evaluated for 15 quantitative characters to determine the extent and geographical pattern of morphological variation. The extent of variation was highly pronounced for agronomically important characters for sorghum. These characters included plant height, days for 50% flowering, peduncle exsertion, panicle length and width, number and length of primary branches per panicle and thousand seed weight. Significant regional variation was also observed for most of the characters. The results implied that environmental factors such as altitude, rainfall, temperature and growing period are important in regional variation. Mean for plant height and for days for 50% flowering showed clinal variation along the gradients of rainfall pattern and growing period in Ethiopia. Moreover, there were significant positive correlation coefficients between most of the characters. This included the correlation between agronomic characters of primary interest in sorghum breeding such as plant height and days for 50% flowering and also between various characters and the altitude of the collection sites. The implications of the results in plant breeding, germplasm collection and conservation as well as the probable sources of the wide range of variation are discussed.

The non-coding chloroplast DNA sequences of the trnL (UAA) intron and the trnL-trnF (GAA) intergeneric spacer (IGS), the coding sequences of nuclear 18S rDNA, and the transcription factor Vp1 of the cereal tef (Eragrostis tef (Zucc.) Trotter) were studied. No intraspecific variation was found among the 6 studied tef varieties. However, the study displayed that Eragrostis tef has a number of unique traits compared to other grasses. Phylogenetic analysis of the chloroplast DNA gave three grass clades, joining Eragrostis with sorghum and maize in one. In the analysis of the 18S rDNA sequences, the three grass species were joined in a monophyletic trichotomy in the cladogram, in which maize is the most divergent, rice the least and tef intermediate. The Vp1 is highly conserved. The Vp1 phylogeny showed that the tef Vp1-sequence is the hitherto most divergent Vp1-sequence reported from a grass.

Thirty-four sorghum (S. bicolor (L.) Moench) landraces consisting of 1020 individual plants (30 plants of each landrace) collected from five agroecological sites (ecosites) in North Shewa and South Welo regions of Ethiopia (Bati 8, Fontenina 5, Haik 2, Layignaw ataye 17 and Merewa Adere 2) were classified on the basis of 4 classifying variables: administrational zones; Woredas (smallest administrative unit), ecosite of origin and altitude. Morphological variation for the fourteen qualitative characters that showed two or more phenotypic classes were estimated using the Shanon–Weaver diversity index (H′). Phenotypic variation was found between and within each classifying variables. The value of H′ for all landraces varied from 0.32 to 0.98 with an overall mean of 0.77±0.04. One-way analysis of variance (ANOVA) showed significant differences between characters within all classifying variables and this contributed to the largest portion of the total variance. Cluster analysis based on ordinal variables grouped the landraces into five clusters. A higher proportion of landraces sharing similar altitude classes and similar ecosites were grouped together. Panicle compactness and shape as well as stalk juiciness were the predominant characters in grouping the landraces into their respective clusters. Panicle compactness and shape also contributed relatively more to altitudinal and ecological differentiation. This differential distribution of landraces with panicle types with respect to compactness and shape revealed the adaptive significance of panicle compactness and shape that reflected the patterns of distribution of different races in north Shewa and south Welo. χ2 test of all characters did not show significant differences between the frequencies of observed and expected characters.

A total of 1050 sorghum (Sorghum bicolor) and tef (Eragrostis tef) fields distributed in six study sites within north Shewa and south Welo (Central Highlands of Ethiopia) were systematically surveyed to examine the status of integration of edible oil crops into the cereal-based farming system. Farmers' criteria for, and perception on, integration between edible oil crops and the major cereals at field level as well as integration between the cereal grains and the oilseeds in food making at home were studied and analyzed based on formal semi-structured interview and informal discussion with local expert farmers as key informants. Farmers' traditional space optimization technique has been instrumental in rightly fitting edible oil crops (as intercrops and border crops) into the cereal-based system. Six species of edible oil crops are integrated (70.3%) in various proportions in fields of sorghum and tef. At least one oil crop was significantly intercropped and/or border cropped in sorghum fields. Noog (Guizotia abyssinica) and sesame (Sesamum indicum) were the most important edible oil crops of the study area having strong integration with sorghum both at field and home level. On average, noog was more frequently intercropped with sorghum (8.3%) than with tef (4.5%), while it was more frequently border cropped with tef (32.4%) than with sorghum (19%). Sorghum was more frequently inter/border cropped with sesame (39/2%) than with tef (10.1/0.5%). The stronger the integration of a given oilseed with sorghum-based foods, the higher the companionship between sorghum fields and the oil crop in the landscape. This cultural practice by farmers has positive contributions to on-farm conservation of oil crops along with tef and specific sorghum landraces. The central theme of this paper therefore converges on the issue of on-farm in-situ agrobiodiversity conservation that was shaped by successive ancestral generations and passed on to the present.

Amplified fragment length polymorphism (AFLP) and random amplified polymorphic DNA (RAPD) markers were used to provide estimates of the comparative genetic variation within and among populations of various Guizotia taxa with the goal of conserving and utilizing their genetic diversity. The percentage of polymorphic loci (P(S)) ranged from 28.5%-90% (AFLP) and 85.6%-99.6% (RAPD). The overall gene diversity estimate () has shown slight variation among taxa ranging from 0.32-0.37 (AFLP) and from 0.22 to 0.28 (RAPD). The within population diversity of "Chelelu" and "Ketcha" was found to be unexpectedly high. Both parameters used to estimate population differentiation (G(ST) and F(ST)) revealed the highest population differentiation G. zavattarii in followed by G. arborescens. Genetic variation among populations within a taxon was highly significant for all the five taxa as revealed by AMOVA (P0.0001). The need for immediate conservation activities for G. arborescens and G. zavattarii, and factors that contribute to the existing genetic variability and population genetic structures are discussed.

In southwest Ethiopia, the cultivation area of Ensete ventricosum (enset) overlaps with the natural distribution area of this species. Analyses of genetic diversity were undertaken using RAPD to provide information for conservation strategies as well as evidence of possible gene flow between the different gene pools, which can be of interest for future improvement of cultivated enset. The extent of RAPD variation in wild enset was investigated in 5 populations in the Bonga area (Kefficho administrative region) and 9 cultivated clones. Comparisons were also made with some Musa samples of potential relevance for crop improvement. Nine oligonucleotide primers amplified 72 polymorphic loci. Population differentiation was estimated with the Shannon index (G'(ST)=0.10), Nei's G(ST) (0.12) and AMOVA (Phi(ST)=0.12), and appears to be relatively low when compared with outbreeding, perennial species in general. Cluster analysis (UPGMA) and principal component analysis (PCA) similarly indicated low population differentiation, and also demonstrated that cultivated clones essentially clustered distinctly from wild enset samples, suggesting that the present-day cultivated enset clones have been introduced to domestication from a limited number of wild progenitors. In addition, subsequent gene flow between wild and cultivated enset may have been prohibited by differences between modes of propagation and harvesting time; cultivated enset is propagated vegetatively through sucker production and the plant is generally harvested before maturity or flower set, thereby hindering pollination by wild enset or vice versa. A significant correlation was not found between genetic and geographical distances. The relatively high total RAPD diversity suggests that wild enset populations in the Bonga area harbour genetic variability which could potentially act as a source for useful or rare genes in the improvement of cultivated enset. As expected, E. ventricosum was clearly differentiated from the analysed Musa samples, that clustered in accordance with the present morphology- and molecular marker-based taxonomy of the genus.

Genetic diversity within the forest Coffea arabica L. gene pool in Ethiopia has not been extensively examined with molecular markers. In the present study, a total of 75 polymorphic RAPD bands generated by twelve random primers were used to assess genetic diversity among 144 genotypes representing 16 C. arabica populations. The number of polymorphic bands detected with each primer ranged from 2 to 9 with a mean of 6.25 bands per primer. Banding patterns ranged in percentage polymorphism from 37% to 73% with an overall mean of 56% for the populations analyzed. The amount of genetic variation among populations estimated by Shannon-Weaver diversity index was (H = 0.30). The within population and between populations differentiation values were 0.65 and 0.35, respectively. Genetic differentiations within and between zones of sample collection sites were 0.80 and 0.20, respectively. Within population average similarities estimated by simple matching coefficients ranged from 0.72 to 0.85, with an overall average of 0.78. In the cluster analysis that used individual samples as operational taxonomic units, most of the representatives of the same population failed to cluster before they joined members of other populations. Nevertheless, most of the populations were clustered on the basis of their geographic closeness and an east west differentiation was observed at approximately 75% similarity. The results obtained provide information on how to select sites for in situ conservation of C. arabica germplasm.

The amount and distribution of genetic variation was investigated in 48 sorghum landrace accessions, representing 13 regions of origin and three adaptation zones (lowland, intermediate and highland elevation) in Ethiopia and Eritrea. Assaying 11 enzymes systems, 23 putative loci were scored for a total of 27 alleles. Nineteen loci were monomorphic and fixed for the same allele, while the remaining 4 loci, each with 2 alleles, were polymorphic across the 48 accessions. The results show significant differences in allele frequencies among the accessions, regions of origin and the adaptation zones. However, all measures of genetic variation used show that the accessions maintained much lower levels of variation than the corresponding mean values for self-pollinating crop plants, confirming previous conclusions that sorghum is depauperated in allozymic variation. The total gene diversity was 0.25, which partitioned 59% within and 41% among accessions. The latter was largely due to variation among accessions within the adaptation zones (38%), while only 3% was due to variation among accessions between the adaptation zones. Similarly, most of the total gene diversity was found within the regions of origin (80%) and within the adaptation zones (97%). Both the dendrogram constructed from NEI's unbiased genetic distance and the plot of the first two principal components distinguished three groups of regions. The level of gene flow was low among accessions, regions of origin and among accessions within adaptation zones, but high among adaptation zones. The results are discussed with emphasis on genetic resources conservation and utilization.

The various models used to test for the maintenance of allozyme variation, as they are formed in the neutralist-selectionist views, are given. The non-random allelic distributions and regional multilocus genetic organization in the Ethiopian land race populations of barley, for the five loci studied, are presented. A tight regional multilocus structure among the five loci is shown in most regions. The reasons for regional ubiquity and specificities of certain gametic combinations and multilocus linkage disequilibria are discussed. The possible maintenance of null alleles at the Est-B locus is accounted for by specific multilocus structure and, possibly, an interplay of environments, with an overriding importance of narrowly fluctuating environments. The importance of the observed multilocus structure in germplasm collection, rejuvenation, conservation, and in several aspects of plant breeding and choice of plant breeding methods is pointed out. A follow-up of the evolution of such coadapted multilocus structures in centres of genetic diversity for several hosts and parasites is recommended as a method to understand the evolutionary dynamics and management of a given gene pool for continual utility.

131 Land race populations of barley represented by 745 individuals from 16 regions in Ethiopia were investigated for a microgeographical distribution of flavonoid patterns. The differential distributions and frequencies of various flavonoid patterns observed in different regions are presumed to be the effect of natural selection, and are indicated to be used as additional characters in pinpointing areas for in situ gene conservation. A regional correlated abundance of Est-B null alleles and flavonoid patterns gave unclarified indications calling for further investigations. A numerical analysis of groupings of the various flavonoid spots recorded was found to show congruence with the already established patterns. A further split of the defined patterns and the possible utility of such subdivisions in disease resistance identification is outlined. The concept for the centre of origin of barley is discussed, the ruling out of the possibility of Ethiopia being the centre of origin for barley is questioned, and the need for archaelogical and genetical evidence on this issue is highly recommended.

158 land race populations of barley were collected from 19 regions lying between 6.10° to 12.0° N and 37.33° to 42.29° E in Ethiopia. The mixed sample of each population was rejuvenated under greenhouse conditions in Sweden. The rejuvenated 3266 individuals of the entire population were assayed electro-phoretically for 5 enzyme loci. Distance matrix and dendrograms generated from genotype and gene frequency data of the 19 regions are compared and contrasted. The diversity index worked out for regional genotypic composition, the distribution of genotypes over populations, and the distribution of populations over genotypes are presented. The genetic distance estimated using allelic or genotypic frequencies showed that D is not related to geographical distance. A total of 156 allozyme genotypes were observed, out of which 95 were unique for single populations. Two genotypes were found to have a very wide distribution over the populations. The lowest number of genotypes in a population was 1, and the highest 16. The unstandardized Shannon-Weaver index of the regions worked out on genotypic composition ranges from 3.44 to 0.80, indicating the different levels of genotypic diversity of the different regions and also the importance of different regions for in situ gene conservation.