Acta agriculturae Slovenica, 121/2, 1–16, Ljubljana 2025 doi:10.14720/aas.2025.121.2.19875 Original research article / izvirni znanstveni članek Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis Gizie ABEJE BELAY 1, 2, Alemayehu ZEWODU KEBEDE 1 Received September 28, 2024, accepted April 29, 2025 Delo je prispelo 28. september 2024, sprejeto 29. april 2025 1 Ethiopian Biodiversity Institute (EBI), Department of Crop and Horticulture Biodiversity Research, P.O. Box 30726, Addis Ababa, Ethiopia 2 Corresponding Author: gizie30@gmail.com Dissecting the phenotypic diversity of Ethiopian barley (Hor- deum vulgare L.) genotypes through variance components and multivariate analysis Abstract: Exploring the genetic diversity of barley germ- plasm conserved in the gene bank is critical for climate-resilient breeding. This study explored the phenotypic diversity and as- sociations of 10 quantitative agro-morphological traits among 229 Ethiopian barley genotypes to identify desirable traits and promising accessions. The experiment was conducted at Holeta Agricultural Research Center in 2023 using an augmented de- sign. The result of the analysis of variance revealed significant variation among the studied genotypes. A significant variabil- ity and wide range of mean performance were observed for the studied traits. Key traits like heading and maturity, grain filling period, 1000-kernel mass, kernel number per spike, and pro- ductive tillers were the most discriminating traits contributing to the highest variability (46.9  %). Cluster analysis identified three major trait-specific clusters. Grain yield exhibits a moder- ate correlation with plant height and kernel number per spike. High phenotypic and genotypic coefficients of variation were obtained for grain yield, productive tillers, kernel number per spike, and spikelet per spike. In addition, high heritability and genetic advance were observed for days to heading, productive tillers, spike length, kernel number and spikelet per spike, and grain yield suggesting selective breeding for these traits is likely to be effective for crop improvement programs. Key words: barley, characterization, genetic variability, conservation, multivariate, augmented RCBD Preučevanje fenoptipske raznolikosti etiopskih genotipov je- čmena (Hordeum vulgare L.) na osnovi variance in multivari- atne analize Izvleček: Preučevanje genetske raznolikosti ječmena shranjenega v genski banki je nujno potrebno za podnebno odporno žlahtnjenje. V raziskavi je bila preučevana fenotip- ska raznolikost in z njo povezane lastnosti desetih kvantita- tivnih agro-morfoloških lastnosti med 229 etiopskimi geno- tipi ječmena z namenom določiti željene lastnosti in obetavne akcesije. Izsledki analize variance so pokazali značilne razno- likosti med preučevanimi genotipi. Med preučevanimi last- nostmi je bila ugotovljena značilna variabilnost z širokim razponom poprečij. Ključne lastnosti kot so čas do klasenja, zrelost, obdobje polnjenja zrnja, masa 1000-zrn, število zrn na klas, in število fertilnih poganjkov so se najbolj razlikovale in so največ prispevale (46.9 %). S klastersko analizo so bili prepoznani trije specifični grozdi, glede na glavne lastnosti. Pridelek zrnja je pokazal zmerno korelacijo z višino rastlin in številom zrn na klas. Doseženi so bili veliki fenotipični in genotipični koeficienti raznolikosti za pridelek zrnja, plodne poganjke, število zrn na klas in število klaskov na klas. Do- datno je bila opažena velika dednost in genetska prednost v znakih kot so število dni do klasenja, v številu plodnih po- ganjkov, v dolžini klasa, v številu zrn in klaskov na klas ter v pridelku zrnja, kar nakazuje, da bi bilo selektivno žlahtnjenje za te lastnosti verjetno učinkovito v programih izboljšanja te poljščine.  Ključne besede: ječmen, ovrednotenje, genetska spre- menljivost, ohranjevanje, multivaritna analiza, izboljšani bločni poskus Acta agriculturae Slovenica, 121/2 – 20252 G. A. BELAY and A. Z. KEBEDE ABBREVIATIONS CSA- Central Statistical Authority; DAP-Diam- monium phosphate; EBI- Ethiopian Biodiversity Insti- tute; GCV-Genotypic coefficient of variance; h2- Broad sense heritability; GAM- Genetic advance over mean; HARC- Holeta Agricultural Research Center; ICARDA- Intenational Center for Agricultural Research for the Dry Areas; IPGRI-International Plant Genetic Resource In- stitute; m. a. s. l- meter above sea level; PCV- Phenotypic coefficient of variance; PCs- Principal components 1 INTRODUCTION Barley (Hordeum vulgare L.) is one of the earliest domesticated crops that was originating from its wild progenitor H. spontaneum (K. Koch) Thell. in the Fer- tile Crescent ~10,000 years ago (Badr et al., 2000; Har- lan and Zohary, 1966). It has been cultivated in a wide range of agroecologies from high altitudes to regions with low rainfall. In Ethiopia, barley was domesticated for more than 5000 years (Harlan, 1968) and is one of the fifth most important traditional food crops next to teff, wheat, maize, and sorghum in terms of area acreage and production (CSA, 2021). Barley cultivation covers more than one million hectares in Ethiopia and con- tributing to 8 % of the nation's total cereal production (Abtew, 2019; Kaso, 2015). Due to its short maturity duration relative to other crops, it is hugely consumed by small-scale farmers in the highlands of Ethiopia for poverty reduction and hunger alleviation during the lean period (Rashid et al., 2019). More than 80 % of bar- ley grown by Ethiopian farmers is for food as compared to malt barley (Kaso, 2015). Besides, the straw is used for animal feed during the dry season and for thatching of roofs (Kaso, 2015). Cognizant of this, it is named as “Gebis ye ehil nigus” in highlanders of Ethiopia to mean “barley the king of all crops” (Mohammed et al., 2016). Ethiopia is a secondary center of genetic diversity for its native two-rowed deficient and irregular bar- ley types (Harlan, 1969; Vavilov, 1951).The Ethiopian barley landraces are diverse in morphology (two and six rows and irregular types) and color (black, white, and pink) (Asfaw, 1988) and also preferred by foreign breeders because of quality traits such as disease resist- ance, drought tolerance, high lysine, and protein con- tent (Bjørnstad et al., 1997; Engels, 1991; Milner et al., 2019). It is also reported that the diverse in altitude, soil types, climates, topography, geographical isolation for long periods together with the farming system and wide socio-cultural diversity are the main driving forces for the diversity of barley in Ethiopia (Harlan, 1969; Tekle- mariam et al., 2022). Like other cereals, barley produc- tion is heavily influenced by spatial and temporal vari- ations in the environment (Dido et al., 2021) and thus exploring and harnessing the genetic diversity available in barley germplasm conserved in the gene bank is cru- cial for the improvement of the crop. According to ICARDA (2025), around 400,000 barley accessions are conserved worldwide in gene banks, breeders, and research collections. In Ethiopia, barley has been conserved in-situ in community seed banks located in different agroecologies of the coun- try and ex-situ in gene bank. For instance, Ethiopian Biodiversity Institute (EBI) has maintained the larg- est barley farmer varieties (> 17,000 (~4.25 %)) in its cold storage facilities as compared with other crops (https://ebi.gov.et/biodiversity/conservation/genetic- material-holdings/). In Ethiopia, farmer varieties or landraces constitute 90 % of the total land devoted to barley production (Hadado et al., 2009). In fact, lan- draces or farmer varieties exhibited both within and be- tween variation as compared to the genetic uniformity of modern cultivars (Zhu et al., 2000). Thus, maintain- ing the genetic diversity in crops is required to sustain genetic improvement for polygenic traits, such as yield. Previous studies showed that landraces are the major constituent of useful genes for adaptation to biotic and abiotic stresses (Asfaw, 1988; Gegnaw & Hadado, 2014). Most specifically, recent studies by Megersa et al. (2015), Monteagudo et al. (2019), and Wosene et al. (2015) revealed that barley landraces have shown high- er yield stability and comparable yield with improved varieties. As such, characterization, and evaluation of genetic variability between and within barley landraces is an essential component for sustainable conservation and improvement. Owing to these, numerous genetic diversity studies were carried out on Ethiopian barley in relation to altitude, eco-geographical distribution of isozyme and allozyme, and association of hordein and morpho-agronomic traits (Asfaw, 1988, 1989; Assefa et al., 2016; Bedasa et al., 2015; Engels, 1994; Fantahun et al., 2023b; Gadissa et al., 2021) Multivariate analysis techniques such as principal component analysis (PCA), clustering, and correlation were commonly employed by researchers (Abebe et al., 2010; Angassa & Tesfaye, 2019; Derbew, 2020; Enyew et al., 2019; Fantahun et al., 2023a) to investigate the ge- netic relationship between genotypes and association of traits. Besides, genetic parameters such as genotypic co- efficient of variation (GCV) and phenotypic coefficient of variation (PCV); heritability, and genetic advance were used by researchers (Addisu & Shumet, 2015; Kaur et al., 2022; Shiferaw et al., 2020; Shtaya et al., 2015) to explore genetic relationships between genotypes and Acta agriculturae Slovenica, 121/2 – 2025 3 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis the association of traits. Despite the enormous efforts made by national and international programs to con- serve barley diversity, there is, however, little informa- tion is yet available as compared with the existing huge genetic resources in Ethiopia which remained underuti- lization of germplasm accessions preserved in the gene bank. Therefore, a rigorous characterization and evalu- ation are utmost important to have sufficient charac- terized accessions to researchers and breeders. In this context, the present study was conducted to explore the genetic variability and association of yield and yield- related components among 229 barley genotypes and identify superior traits and accessions that can contrib- ute to future barley grain yield improvement program and conservation. 2 MATERIAL AND METHODS 2.1 PLANT MATERIALS AND EXPERIMENTAL SITE A total of 232 barley genotypes comprised of 229 accessions obtained from the Ethiopian Biodiversity Institute and three improved check cultivars contain- ing six-rowed food barley (HB-1307 and HB-1966) and two-rowed malt barley (Suba) (Figure 1B; Additional File 1, Sheet 1) released by Holeta agricultural research center (HARC) were characterized. The accessions were requested from the gene bank owing to the major barley growing regions, altitudinal class (1940-3546 m. a. s. l) and previous characterization history in the gene bank. The study was conducted in 2022/23 main growing sea- son at the research field of HARC located in the Oromia regional state of West Shewa zone, Walmera woreda (Fig- ure 1C) (9º00’00’’ N, 38º30’00’’ E, elevation 2400 meter above sea level. The temperature varied from 6-22  0C with an average annual rainfall of 1144 mm and light soil (Nitosols and Vertosols (http://www.eiar.gov.et/holetta/). The experiment was laid out by an Augmented block design where the check cultivars were replicated and appeared once in each block but test treatments ap- peared once in the design. The design assumes checks as fixed effects whereas entries as random effects. The ex- perimental area contained a total of seven blocks. Each block had a total of 36 entries comprised of 33 accessions and three checks. To scrutinize the bias among treat- ments, randomization were done by Agricolae package of R-software (Mendiburu, 2020). Each accession was drilled by hand in 4 rows of 2 m2 having with 2.5 m row length with 0.2 m row spaced at 85 kg ha-1 seed rate (17 g plot-1). The distance between each plot & block was 0.4 m & 1  m respectively. All the recommended agronomic practices were applied to each treatment equally before and after sowing according to the standard operational procedure of characterization (Ethiopian Biodiversity In- stitute, 2021). Artificial fertilizer was applied equally to each treatment before sowing at the rate of 36.4 g NPS and 11.2 g UREA split to 7.5 g (during sowing) and 3.7 g (during tillering after first weeding). 2.2 DATA COLLECTION All data were electronically recorded through tablet using the FieldScorer Android App (http:// www.katmandoo.org/Help/Fieldscorer4Android/in- dex.html) after developing a standardized comma de- limited (csv) trial and trait files (Tamirat Bejiga and Amare Seyoum, 2018). The traits were prepared in accordance with the barley descriptor list developed by Bioversity International (IPGRI, 1994) with minor modifications. Since three accessions were missed due to heavy water stress, data were exclusively col- lected for 229 genotypes (226 accessions and three checks). Quantitative data were collected on a plant and plot basis after randomly selecting and tagging 20 representative plants from each plot and the aver- age of these samples was used for the analysis. Three phenological and seven agro-morphological traits were recorded in this study. Days to heading (DTH) were counted as the number of days from sowing to 50 % of plants fully emerged (Z55 stage (Zadoks et al., 1974)); days to maturity (DTM) was counted as the number of days from sowing to 75 % of the plants physiologically matured); grain filling period (GFP = DTM - DTH) was measured as the number Figure 1: Map of Ethiopia showing the administrative regions (A), sample points where barley study materials were collected (B) and experimental site (C). Map was generated by ArcMap 10.2.2 (Esri, 2014). (All boundaries were prepared in accord- ance with Ethiopia Geoportal (Last updated on May 2020; htt- ps://ethiopia.africageoportal.com/) Acta agriculturae Slovenica, 121/2 – 20254 G. A. BELAY and A. Z. KEBEDE of days from the time that half of the plants have started flowering to 75 % of the plants fully matured; plant height (PH, cm) was measured at full maturity from the ground level to the top of the spike exclud- ing awns; productive tillers (PT) was counted on the number of effective tillers arising from the main plant; spike length (SL, cm) was taken from the neck to the base of the lemma awns of the uppermost grain; spikelet per spike (SPS) was counted as the number of spikelet raised from each node; kernel number per spike (KNS) was counted as the number of kernels obtained after threshing of spikes; thousand kernel mass (TKM, g) were measured after weighing and converting of 250 seeds and grain yield (GY, kg ha-1) were obtained from the field plot (g m-2) after extrap- olating the area of 2 m2 to hectare basis. TKM and GY were determined after adjusting to 12.5 % seed moisture content. 2.3 STATISTICAL DATA ANALYSIS 2.3.1 Estimates of analysis of variance and genetic variability The quantitative data obtained from augmented design were subjected to analysis of variance (ANO- VA), and genetic variability analysis using augmented RCBD bulk function of the augmented RCBD pack- age (Aravind et al., 2021) in R software (Team, 2024). The phenotypic (), genotypic (and environmental variance ( were estimated from the ANOVA tables according to the expected value of the mean square described by Federer and Searle (1976) as follows: Where MSg-mean square due to the genotype, MSe-error mean square, r-number of replications for the check (control) genotypes, -Phenotypic vari- ance, σ2g -Genotypic variance, σ2e - Environmental variance = eror mean square (MSe). The phenotypic and genotypic coefficients of variation (PCV & GCV) were estimated according to (Burton, 1951) and cat- egorized as low (< 10), medium (10-20) and high (≥ 20) according to (Sivasubramaniam S, 1973). Where σ2p = Phenotypic variance, σ2g = Genotypic vari- ance and x̄ = the grand mean of the trait under consider- ation. Heritability (h2) in the broad sense was estimat- ed according to the method of (Lush, 1940) and cat- egorized according to (Robinson, 1966) suggestion as low (< 30), medium (30-60) and high (≥ 60). Genetic advance (GA) and genetic advance as a percent of the mean (GAM) were calculated by as- suming the selection of more than 5 % of the geno- types estimated per the methods illustrated by (John- son et al., 1955). where GA - expected genetic advance, σp- phenotypic standard deviation on the mean basis, h2 - Heritability in a broad sense, K = selection differen- tial (where k = 2.06 at 5 % selection intensity). The genetic advance (as a percent of the mean) (GAM) was computed to compare the extent of the predicted genetic advance of different traits under selection and categorized as low (< 10 %), moderate (10 %-20 %), or high ( > 20%) according to (Johnson et al., 1955) Where GAM - genetic advance as a percentage of the mean, - Grand mean of the quantitative character 2.3.2 Multivariate analysis The adjusted mean values generated from the ANOVA analysis were subsequently used for the com- putation of principal component analysis, pairwise correlations, and clustering analysis. Prior to analysis, the adjusted mean values were standardized to means of zero and variances of unity to avoid variations in the scales used during data collection (Manly, 1986; Sneath and Sokal, 1973) using the scales function in R software (Wickham & Seidel, 2020). The principal component and biplot analysis was computed using FactoMineR (Lê et al., 2008), and ggbiplot (Vu, 2011). A Spearman rank correlation was generated by GGal- ly (Schloerke et al., 2024) and ggplot2 (Wickham et al., 2016) packages. A phylogenetic tree-based cluster analysis was performed after determining the cut- off point using the NbClust package (Charrad et al., 2014). The cluster analysis was hierarchically catego- Acta agriculturae Slovenica, 121/2 – 2025 5 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis rized by the Wards linkage clustering method. Besides, the genetic distance among genotypes and divergence among clusters was performed by using NbClust pack- age (Charrad et al., 2014) and clv package (Niewe- glowski & Nieweglowski, 2015) respectively. 3 RESULTS AND DISCUSSION 3.1 ANALYSIS OF VARIANCE (ANOVA) The results of ANOVA presented in Table 1 showed a significant difference (p  ≤  0.05) to highly significant (p  ≤  0.01) difference among tests (acces- sions) and genotypes (entries) for PT, DTH, DTM, PH, SL, SPS, KNS, TGW, and GY. In contrast, a non- significant difference was observed for GFP. These notable significant differences among accessions re- vealed that substantial genetic variability exists which can be harnessed through selection. In agreement with our study, Zewodu et al. (2025) reported highly sig- nificant differences (p ≤ 0.01) for DTH, DTM, PH, SL, SPS, KNS, TKM, and GY. Similarly, (Derbew, 2020; Dido et al., 2020) also reported a non-significant dif- ference for GFP and a highly significant difference for other parameters. The two-food barley and one-malt barley improved varieties used for standard checks showed significant (p ≤  0.05) variation for GFP, PH, and PT to highly significant variation (p ≤  0.01) for the rest of the parameters except DTH which showed non-significant variation. A similar study on 102 Ethi- opian food barley landraces and five checks (Bedasa et al., 2015) reported highly significant variation for PH, SL, KNS, and TKM. In addition, the mean square due to test (accessions) v/s checks were highly significant (p ≤  0.01) for DTM, DTH, GFP, PH, and KNS traits indicating the presence of significant differences be- tween landraces and checks in these traits. Table 1: The mean square of ANOVA for 10 morpho-agronomic traits of studied barley genotypes (n = 229) resulted from aug- mented design Sources Mean squares† DTH DTM GFP PH PT SL SPS KNS TKM GY Blocks (6) 4.98ns 14.43ns 11.94ns 140.52** 0.64* 1.02* 3.25* 5.62ns 5.91* 1304592.5** Among genotypes (228) 95.73** 158.17** 24.09ns 108.73** 0.49* 1.35** 36.77** 108.9** 26.6** 690112.7** Among check (2) 11.48ns 147.57** 76.76* 136.92* 2.59** 6.53** 185** 951.9** 64.3** 951074.6** Among accessions (225) 92.43** 143.71** 20.52ns 102.22** 0.47* 1.31** 35.6** 100.3** 26.4** 688720.7** Accessions vs. Check (1) 1006.8** 3432.7** 721.4** 1517.7** 0.3ns 0.54ns 3.53ns 369.3** 4.77ns 481399.7ns Residuals (12) 4.37 17.9 12.65 21.8 0.16 0.33 0.96 3.56 1.91 136411.1 CV (%) 2.85 3.69 8.58 5.5 22.4 7.02 4.75 5.52 2.81 17.6 Standard Errors A Test Treatment and a Control Treatment 2.58 5.22 4.39 5.76 0.49 0.71 1.21 2.33 1.7 455.92 Control Treatment Means 1.12 2.26 1.9 2.5 0.21 0.31 0.52 1.01 0.74 197.42 Two Test Treat- ments (Different Blocks) 3.41 6.91 5.81 7.62 0.65 0.94 1.6 3.08 2.25 603.13 Two Test Treat- ments (Same Block) 2.95 5.98 5.03 6.6 0.56 0.82 1.39 2.67 1.95 522.32 † DTH-Days to 50 % heading, DTM-days to 75 % maturity, GFP-grain filling period, PH-plant height (cm), PT-productive tiller, SL-spike length (cm), SPS-spikelet per spike, TKM-thousand kernel mass (g) and GY-grain yield (Kg ha-1); genotypes-accession + checks. ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01; CV- coefficient of variation; Numbers in parenthesis represented degree of freedom. Acta agriculturae Slovenica, 121/2 – 20256 G. A. BELAY and A. Z. KEBEDE 3.2 AGRO-MORPHOLOGICAL TRAIT DIVERSI- TY AND MEAN PERFORMANCE OF GENO- TYPES The mean performance of the studied barley gen- otypes for 10 morpho-agronomic traits is presented in Additional File 1, Sheet 2. In general, barley geno- types exhibited a high level of variation in yield and related yield components (Figure 2; Additional File 1, Sheet 2). Meanwhile, the mean value of DTH was 72.7 days and varied from 54 days to 99 days. On the other hand, the number of days to attain 75 % physiologi- cal maturity (DTM) had a mean of 113.7 days with a range of 89 days to 146 days. In line with our work, a combined two and six-row barley study by (Gadissa & Gudeta, 2023) and (Angassa & Tesfaye, 2019) showed a comparable DTH (56-97 & 56-93 days) and DTM (99-138 & 89-140 days) respectively. In contrast, a wide range of DTH (91-116 days) and DTM (117-174 days) were also reported on Ethiopian six-row bar- ley genotypes by Alemayehu & Parlevliet (1997) and Megersa et al. (2015). In the present study, two-rowed barley accession-243314 had taken earlier heading (53 days) and maturing time (89 days) as compared to six-rowed barley accession-243209 (98 days and 146 days respectively) which is consistent with (Fantahun et al., 2023a; Kandić et al., 2018; Setotaw et al., 2010; Zewodu et al., 2025) studies. These results, however, disagree with Kaur et al. (2022) study who reported as two-rowed varieties took many days to mature as compared to six-rowed varieties. GFP on the other hand is found to be a crucial determinant of the ge- netic variation in rice (Yang et al., 2008) which ranged from 31 days to 59 days with a mean of 41 days and CV of 8.58. In a similar study, a wide range of GFP (23.88- 34.22) with a mean of 30.11 was reported by (Megersa et al., 2015). In general, the difference in the number of days (DTH, DTM, and GFP) observed in our study is mainly attributed from the genetic background of genotypes (two-rowed, six-rowed, irregular, and hul- less) and the mixture of materials used. This is from the fact that the barley genetic materials preserved in the gene bank are in the population form. The regional diversity of agronomic traits (Ta- ble 2) also revealed that accessions collected from the Amhara region showed minimum values of DTH (54 days), DTM (89 days), and GFP (31 days) as compared to the maximum values of DTH (99 days) and DTM (146 days) observed for accessions collected from Oromia and local improved varieties. In fact, earli- ness is one of the key adaptive traits as demonstrated in wheat (Hyles et al., 2020; Mondal et al., 2013) that enables crops to escape terminal moisture stress and en- suring more reliable yields under unpredictable and re- Figure 2: Diversity of some of barley genotypes based on maturity and grain filling period at 105 days after sowing (DAS) (a); row types i.e. six row (b) irregular types (c) two rowed sterile, long awn and awnless (d) two rowed deficient (e), hulless and composite barley (Acc-241790) (f &g) found in the field (top right corner) and a pot containing the same composite barley seeds planted in the Lathouse (h) and emerged two different spike forms i.e. a single spike (i) and triple-spikes (j) Acta agriculturae Slovenica, 121/2 – 2025 7 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis source-limited conditions (Kandić et al., 2018; Megersa et al., 2015; Yadav et al., 2018) and hunger alleviation during the lean period (Amri et al., 2005; Rashid et al., 2019). Therefore, the wide range of variation present in the studied genotypes across regions for phenological traits allows breeders to improve grain yield for short and extended growing seasons accordingly. Plant height (PH) is another important agronomic trait highly valued by farmers (for animal fodder, thatch- ing roofs) and breeders (lodging tolerance) (Kaso, 2015; Kaur et al., 2022). In this study, the highest (108.4 cm) PH was found for accession 243296 and the lowest (57.8 cm) being for accession 241790 with a mean of 85.6 cm and CV of 5.5. Interestingly, accession 241790 (Figure 2f) is found to be six-rowed hulless and merely has com- posite spike forms with dwarf plant height (57.8 cm) and also a minimum grain filling period (32 days). To further elucidate the mode of inheritance, we planted the seed obtained from composite barley spike (Figure 2g) in two pots in Lathouse (Figure 2h) and it was found that > 99 % of the spikes that emerged were single spikes while only a single spike showed triple-spike form (Figure 2j). This stipulated that accession 241790 could have a potential poly-row-branch-spike (prbs) gene (Poursarebani et al., 2015, 2020; Terzi et al., 2017) which needs further inves- tigation at the multi-location and genomic level. Grain yield (GY) is the consequence of multiplica- tive interaction of intricate traits such as SL, SPS, KNS, PT, TKM, and rarely on PH (Dziurdziak et al., 2020; Kaur et al., 2022). In this study, a wide range of variation was ob- served for SL (5.2-11.5 cm), SPS (12.5-33.9), KNS (14.3- 53.9), and TKM (31.9-62.4 g). Whereas, a narrow range of PT (0.23-3.7) with a mean value of 1.78 was observed for accession 243274 and 243307 respectively (Table 1). Previous studies (Angassa & Tesfaye, 2019; Dziurdziak et al., 2020, 2021) also reported a closer range and mean value of PH, SPS, KNS, SL and TKM A comprehensive phenotypic characterization and genetic diversity anal- ysis of 6,778 barley germplasm (Kaur et al., 2022) also revealed a wide variation in PH (45.96-171.32cm), SL (3.44-13.73 cm), KNS (10.48-82.35) and TKM (12-68.6 g). Besides, the lowest GY (292.3 kg ha-1) was recorded for two-rowed barley accession-243212 and the highest (3599.9 kg ha-1) being for six-rowed accession-243313 with an average of 2118.3 kg ha-1 (Additional File 1, Sheet 2). Derbew et al., (2013) reported a wide range of varia- tion in GY (436-3752.5 kg ha-1). It is evident from (Verma et al., 2021) study that two-rowed barley generally have a reduced number of grains per spike which is signifi- cantly affecting the final grain yield as compared to six- rowed ones. The regional diversity of morpho-agronomic parameters (Table 2) also indicated that accessions col- lected from Amhara have high SPS (50.5), TKM (62.9 g), and GY (3599.9 kg ha-1) as compared to SPS (26.2), TKM (51.6) and GY (2225.8 kg ha-1) of standard checks exploited in the study. In general, out of 229 barley geno- types, 119 (52 %) genotypes have given higher GY than the grand mean (2118.3 kg ha-1) (Additional File 1, Sheet 2). Despite the poor attitude towards landraces on their low yield potential (Azeez et al., 2018), the average GY obtained in this study, however, is significantly higher than the improved varieties HB-1307 (2102.9 kg ha-1), HB-1966 (2225.8 kg ha-1), and Suba (1534.8 kg ha-1). In agreement with our results, a multi-environment field study (Wosene et al., 2015) also indicated that Ethiopian barley landraces or farmer varieties show higher yield stability and comparable yield to improved varieties. A similar study conducted in Ethiopia (Lakew et al., 1997), Syria (Ceccarelli, 1996) and Spain (Yahiaoui et al., 2014) also supported our study and concluded that landraces outperformed improved varieties grown in favorable to modern stress conditions. 3.3 PAIRWISE CORRELATION ANALYSIS Results of the pairwise Spearman rank correlation coefficient (rs) among 10 morpho-agronomic traits are displayed in Figure 3. The estimates of trait associa- tion revealed that a strong positive association exists among pheno-agronomic traits such as DTH-DTM (rs = 0.9), DTM-GFP (rs = 0.67), and GY-PH (rs = 0.5). In contrast, a strong negative correlation was observed between KNS and TKM (r = -0.62). However, most of the trait associations showed weak correlations Figure 3: Spearman correlations of 10 morpho-agronomic traits of barley germplasm accessions. Keys: no correlation (−0.1 ≤ rs ≤ 0.1), weak correlation (−0.3 ≤ rs < −0.1 or 0.1 < rs ≤ 0.3), moderate correlation (-0.3 < rs < -0.5 or 0.3 < rs < 0.5) and strong correlation (rs < −0.5 or rs > 0.5) where rs is Spearman correlation. Trait abbreviations are given in Table 1 footnote. Acta agriculturae Slovenica, 121/2 – 20258 G. A. BELAY and A. Z. KEBEDE (44 %) followed by no correlations (29 %), moderate correlations (18 %) and strong correlations (9 %). For instance, moderate correlation was found for KNS- GY (rs = 0.4), TKM-SPS (rs = 0.4), TKM-SL (rs = 0.3), TKM-PT (rs = 0.39), DTM-KNS (rs=0.37), DTH-KNS (rs = 0.38), DTH-GFP (rs = 0.35) traits. Similarly, GY showed a weak correlation with all phenological traits. The better the correlations among traits the more likely it is that breeders can indirectly select one trait based on other traits with high heritability. From the breeders point of view, the strong correlation between GY and PH (rs = 0.5) is found to be problematic with respect to lodging and yield penalty. Therefore, reduc- ing plant height without incurring grain yield is highly recommended to escape lodging as also reported by Monteagudo et al. (2019) study that found a positive correlation between PH and GY. 3.4 PRINCIPAL COMPONENT ANALYSIS Principal component analysis (PCA) was estimat- ed to elucidate the spatial distribution and diversity of genotypes and eventually identify the key traits that con- tributed utmost to the observed variations. In this study, the principal component analysis based on the correla- tion between 10 measured traits was used to discern the pattern of variation for the studied 229 barley genotypes (Table 3). The first four principal components (PCs) having Eigenvalue ≥ 1 and explained 74.3 % of the total cumulative variance present in the studied barley germ- plasm were retained (Table 3, Figure S1 a & b). Likewise, Yadav et al., (2018)reported Eigenvalue ≥1 that explained a total variability of 79.5 %. In addition, (Bedasa et al., 2015) and (Demissie & Bjørnstad, 1996) reported four principal components explaining a total variance of 72 % and 63 % in the studied 102 and 49 barley accessions re- spectively. Region Morpho- agronomic traits† DTH DTM GFP PH PT SL SPS KNS TKM GY Amhara (n = 153) Mean 72.9 113.7 40.8 86.7 1.7 8.2 20.5 35.0 48.7 2109.5 Min 54 89 31 57.8 0.2 5.2 14.1 14.3 31.9 513.0 Max 97 141 59 108.4 3.7 11.2 50.5 52.2 62.4 3599.9 Oromia (n = 41) Mean 73.4 114.8 41.4 85.1 1.8 8.2 19.9 35.0 48.2 2190.7 Min 59 96 31 67.6 0.6 5.4 16.0 18.8 31.9 292.3 Max 99 146 55 99.7 3.0 11.5 39.9 49.0 56.3 3511.9 Tigray (n = 32) Mean 70.0 111.0 41.0 82.1 2.1 8.1 21.5 26.6 52.4 2082.7 Min 58 101 31 69.2 0.8 5.9 16.8 18.2 45.6 787.2 Max 82 126 47 100.6 3.5 10.4 32.0 53.9 60.4 3055.3 Local checks (n = 3) Mean 79.8 127 47 76.8 1.9 8.4 20.3 38.2 49.6 1954.5 Min 79 124 45 73.6 1.4 7.8 16.8 24.9 46.1 1534.8 Max 81 132 51 81.8 2.6 9.5 26.2 46.6 51.6 2225.8 Accessions (n = 226) Mean 72.6 113.5 40.9 85.7 1.8 8.2 20.6 33.8 49.1 2120.5 Min 54 89 31 57.8 0.2 5.2 14.1 14.3 31.9 292.3 Max 99 146 59 108.4 3.7 11.5 50.5 53.9 62.4 3599.9 Genotypes (n = 229) Mean 72.7 113.7 41.0 85.6 1.8 8.2 20.5 33.9 49.1 2118.3 Min 54 89 31 57.8 0.2 5.2 14.1 14.3 31.9 292.3 Max 99 146 59 108.4 3.7 11.5 50.5 53.9 62.4 3599.9 Table 2: Mean, minimum, and maximum values of agro-morphological parameters of Ethiopian barley genotype and local released varieties based on regional diversity Min-minimum, Max-maximum; †Trait abbreviations are given in Table 1 footnote Acta agriculturae Slovenica, 121/2 – 2025 9 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis Factor loadings were considered with > 10 % posi- tive contribution (Figure S1c & d, Table 3) and indicated that DTH (24.99), DTM (29.74), KNS (17.65), and GFP (13.36) were mostly contributed for PC1 in the nega- tive direction. The major traits that contributed to PC2 were TKM (30.58) in the positive direction and PT (14.4) and KNS (13.7) in the negative direction. Similarly, GY (36.88), PH (32.69), and SPS (12.2) were contributed for PC3 in the negative direction while SL (43.64) and PH (15.58) attributed for PC4 in the negative direction and SPS (23.52) in the positive direction. In line with our study (Dido et al., 2020) reported that DTH was heavily loaded in the negative direction for PC1. In contrast, the findings of (Fantahun et al., 2023a) indicated that PC1 was mostly affected by phenological traits in the positive direction while DTM was highly loaded. To visualize the associations of genotypes and mea- sured traits, a PCA scatter biplot was generated based on the first two PCs (Figure 4a). The first and the second PCA biplot accounted for 46.8 % of the total variability among the genotypes, indicating that DTH, DTM, GFP, TKM, KNS, and PT were considered as the most dis- criminating traits contributing to the highest variability (Figure 4a, Table 3). The genotypes positioned on the top right quadrant were characterized by late heading and maturity whereas, the genotypes clustered on the bottom right quadrant were associated with the highest GY and KNS; the genotypes positioned on the top left quadrant are associated with high TKM, PT and SL whereas, the genotypes assembled at the bottom left quadrant is char- Figure 4: A PCA biplot analysis based on the mean value of 10 morpho-agronomic traits depicting the scatter of 226 barley ac- cessions and three checks along with their measured traits and entire collection sites (a) and a Principal Component Analysis (PCA) biplot where the samples are colored by cluster (b). Trait abbreviations are given in Table 1. The vector indicates the traits and unknown represents improved varieties (Suba, HB-1307, and HB-1966). The more the deviation from the center towards the direction of the vector, the higher the value and vice versa. In other words, an angle of 90° or less between any two vectors (traits) indicates a positive correlation and vice versa. Traits† PC1 PC2 PC3 PC4 DTH -0.50 (24.99) 0.19 (3.80) 0.15 (2.15) -0.26 (6.81) DTM -0.55 (29.74) 0.28 (8.06) 0.05 (0.22) -0.14 (1.92) GFP -0.37 (13.36) 0.31 (9.80) -0.16 (2.68) 0.16 (2.54) PH 0.09 (0.86) -0.12 (1.35) -0.57 (32.69) -0.39 (15.58) PT 0.20 (4.12) 0.38 (14.4) -0.25 (6.36) 0.09 (0.85) SL 0.22 (5.05) 0.28 (7.89) 0.09 (0.88) -0.66 (43.64) SPS -0.10 (1.08) 0.23 (5.36) -0.35 (12.2) 0.48 (23.52) KNS -0.42 (17.65) -0.37 (13.7) -0.16 (2.52) -0.19 (3.43) TKM 0.16 (2.52) 0.55 (30.58) -0.18 (3.41) -0.10 (0.99) GY -0.08 (0.64) -0.22 (5.05) -0.61 (36.88) -0.09 (0.73) Eigenvalue 2.7 2.0 1.8 1.0 Variance (%) 27.0 19.8 17.7 9.8 Cumulative variance (%) 27.0 46.9 64.6 74.3 Table 3: Eigen values, Eigen vectors, variance, cumulative variance and contribution of variables to four dimensions obtained from the considered 10 morpho-agronomic traits Values in bold indicate the highest absolute values among all four PCs for each trait which represents relative contribution and values in parentheses indicate factor loadings with > 10 % positive contribution to the variation. †Trait abbreviations are given in Table 1 footnote. Acta agriculturae Slovenica, 121/2 – 202510 G. A. BELAY and A. Z. KEBEDE acterized by high PH, early heading and maturity. On the other hand, the genotypes concentrated around the origin had similar genetic characteristics, while the geno- types that were found far from the origin are considered as unrelated genotypes (Figure 4a). Therefore, the distri- bution of the genotype over the four quadrants indicates the presence of significant genetic diversity among the studied barley genotypes. On the other hand, a PCA clus- ter biplot (Figure 4b) indicated that 229 barley genotypes were grouped in three major clusters. 3.5 CLUSTERING ANALYSIS Clustering is grouping of individuals according to their similarities. Cluster analysis (Figure 5) based on 10 quantitative traits for 229 barley genotypes were grouped into three major clusters, each with a trait-spe- cific pattern. The mean values of the three clusters for the 10 quantitative traits of the tested barley genotypes (Table 4) showed that a high cluster mean was recorded in cluster (I) for PH, KNS, and GY and the lowest for PT and TKM. These findings suggest that the genotypes in cluster I could be preferable for improving grain yield while maintaining high plant height, making them suit- able for both animal fodder and human consumption. cluster (II) exhibited the lowest mean value for all phe- nological traits studied and the highest value for PH, PT, SL, and TKM. This indicates that the genotypes in cluster II could serve as a valuable genetic resource for developing early maturing and high-yielding barley cultivars, particularly suited for short rainy seasons. In contrast, the genotypes in cluster III were characterized by late heading and maturity, a high number of spikes per plant (SPS), and shorter plant height compared to the other clusters. These traits make cluster III geno- types a potential source to cross-hybridize and develop dwarf barley varieties suitable for extended rainy sea- sons. Among the 229 genotypes studied, 97 (42.4  %) were grouped in Cluster I, which exhibited intermediate phenological characteristics. Meanwhile, 71 genotypes (31 %) were classified as early-maturing (cluster II), and 61 genotypes (26.6 %) were late-maturing (cluster III) (Table 4). Statistical tests further confirmed that the ob- served differences among clusters were significant for DTH, DTM, GFP, PH, SL, SPS, KNS, and GY. However, no significant differences were found for TKM and PT (Table 4). Overall, the genotypes were clustered based on morphological similarity rather than their collection region. This is consistent with previous studies (Bedasa et al., 2015; Benlioglu et al., 2025; Kaur et al., 2022), which found that cluster analysis groups genotypes ac- cording to morphological traits and less on their geo- graphic location. Figure 5 further illustrates that two- Table 4: Cluster means and size among three clusters generated from 10 morpho-agronomic traits Clusters †Traits DTH DTM GFP PH PT SL SPS KNS TKM GY Cluster size I 72.4 112.0 39.5 85.6 1.51 7.9 19.9 41.3 45.3 2398.0 97 (42.4%) II 64.1 102.8 38.7 86.9 2.19 8.6 21.0 22.6 52.5 1893.5 71 (31%) III 83.0 129.1 46.1 83.6 1.74 8.2 21.7 35.1 51.4 1929.3 61 (26.6%) p-values 0.027 0.008 0.049 0.000 0.211 0.036 0.007 0.000 0.632 0.000 The means/medians of the clusters are significantly different for the traits if p < 0.01 (highly significant), p < 0.05 (significant), and p > 0.05 (non- significant); †Trait abbreviations are found in Table 1 footnote Figure 5: A circular dendrogram-based phylogenetic tree of 229 barley genotypes obtained from 10 quantitative traits. Genotypes highlighted in the blue box (cluster III) indicate six- rowed improved barley genotypes and red box (cluster II) for two-rowed ones. The prefix presented before each accession number indicates their region of collection (e.g. AM-Amhara, OR-Oromia, TG-Tigray). Acta agriculturae Slovenica, 121/2 – 2025 11 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis rowed barley accessions were predominantly grouped in cluster II, six-rowed ones in cluster III, and a mix of irregular and six-rowed barley spikes in cluster I. Simi- lar findings were reported by (Fantahun et al., 2023a). The Euclidean distance analysis revealed the high- est genetic distance (10.02) between accession numbers AM-242060 (early heading/maturing) and OR-243209 (late heading/maturing). In contrast, the smallest ge- netic distance (0.68) was observed between accessions TG-25357 and TG-35329, both collected from the Tig- ray region, indicating their similar genetic backgrounds (Additional File 1, Sheet 3). These results align with the findings of Angassa and Mohammed (2021), who re- ported a genetic distance range of 0.61 to 8.06 among 138 barley genotypes. The inter-cluster distances based on the Euclidean dissimilarity matrix (Table 5) showed the highest distance (5.06) between clusters II and III, suggesting that these clusters are genetically distinct and makes them ideal candidates for crossing to create high-yielding, early-maturing, or dwarf barley varieties with diverse traits. In contrast, the lowest (4.51) inter- cluster distance between clusters I and II revealed that the clusters are genetically similar or less divergence in traits. On the other hand, the highest intra-cluster dis- tance (4.08) observed in cluster II suggests greater ge- netic diversity or variability among the genotypes within the cluster. In contrast, the low intra-cluster distance (3.69) in cluster III indicates that the genotypes within a cluster are genetically similar or closely related. Overall, the highest and lowest values of intra-cluster and inter- cluster distances provide important insights into the ge- netic diversity and relationships among the clusters and genotypes. 3.6 ESTIMATES OF VARIANCE COMPONENTS, HERITABILITY, AND GENETIC ADVANCE The genetic variability (Table 6) was computed to estimate the level of genetic variation for environmental and genetic factors. High magnitude of phenotypic var- iance (σ2p) and genotypic variance (σ2g) were observed in GY, DTM, PH, KNS and DTH and low magnitude in SL, PT, SPS, TKM and GFP traits. Because of variations in mean and measurement units, the value of phenotyp- ic and genotypic variants cannot be directly compared across traits (Gadissa et al., 2021). Therefore, coefficient of variation at phenotype and genotype levels was used to compare variability among traits (Table 6). The coef- ficient of variation study indicated that the highest PCV and GCV were noted on GY (39.18 and 35.08), PT (38.41 and 31.19), KNS (29.56 and 29.03) and SPS (28.83 and 28.44) indicating high degree of genetic variability and Table 5: Intra (diagonal and bold) and inter (above diagonal) cluster distance (D2) among three clusters generated from 229 barley genotypes Clusters I II III I 3.83 4.51 4.57 II 4.08 5.06 III 3.69 Table 6: Genetic variability analysis of 10 yield and yield-related traits of 229 barley genotypes Trait Mean ± SE Range σ2p σ2g σ2e GCV (%) PCV (%) ECV h2b (%) GA GAM (%) DTH 72.70 ± 0.62 54-99 92.43 88.07 4.37 12.92 13.23 2.88 95.28 18.90 26.01 DTM 113.7 ± 0.79 89-146 143.71 125.80 17.90 9.87 10.55 3.72 87.54 21.65 19.04 GFP 41.0 ± 0.33 31-59 20.52 7.87 12.65 6.84 11.04 8.67 38.34 3.58 8.73 PH 85.6 ± 0.56 57.8-108.4 102.22 80.42 21.80 10.47 11.81 5.45 78.67 16.41 19.16 PT 1.78 ± 0.04 0.2-3.7 0.47 0.31 0.16 31.19 38.41 22.42 65.95 0.93 52.26 SL 8.2 ± 0.07 5.2-11.5 1.31 0.98 0.33 12.07 13.97 7.03 74.68 1.76 21.52 SPS 20.7 ± 0.4 12.5-33.9 35.60 34.63 0.96 28.44 28.83 4.75 97.29 11.98 57.87 KNS 33.9 ± 0.66 14.3-53.9 100.27 96.71 3.56 29.03 29.56 5.57 96.45 19.92 58.82 TKM 49.1 ± 0.34 31.9-62.4 26.39 24.48 1.91 10.07 10.45 2.81 92.78 9.83 20.01 GY 2118.3 ± 43.3 292.3-3599.9 688720.7 552309.5 136411.1 35.08 39.18 17.44 80.19 1372.97 64.82 Keys: σ2 p-phenotypic variance, σ2 g-genotypic variance, σ2 e-environmental variance, GCV-genotypic coefficient of variance, PCV-phenotypic coef- ficient of variance, ECV-environmental coefficient of variance, h2-heritability in the broad sense, GA-genetic advance, GAM-genetic advance over mean. PCV, GCV and GAM values < 10 % (low), 10-20 % (medium), > 20 % (high); h2 values < 30 (low), 30-60 (medium), and ≥ 60 (high). Trait abbreviations are given in Table 1 footnote. Acta agriculturae Slovenica, 121/2 – 202512 G. A. BELAY and A. Z. KEBEDE close relationship with yield attributing traits and thus could be a useful trait for screening yield traits. Whereas medium PCV and GCV values were observed on the re- maining traits except for DTM and GFP which showed low GCV. The magnitude of PCV value was generally greater than GCV for all studied traits as expected and revealed that the apparent variation was due to the envi- ronmental factors up to some extent. However, the mag- nitude of the difference was low for all studied traits sug- gesting the observed variation was due to genetic factors and less influenced by environmental factors. Moreover, the low GCV estimates for DTM and GFP suggest a di- rect selection on these traits for improvement depending on phenotype expression. In line with our findings, high coefficient of variation (PCV and GCV) were reported by (Addisu & Shumet, 2015; Angassa, 2021; Zewodu et al., 2025) in barley for SPS, KNS and GY traits. Broad sense heritability (h2) of a trait represents the extent at which the genotypes are affected by environ- ment and environmental errors (You et al., 2017). High heritability was obtained on SPS (97.29 %) followed by KNS (96.45 %), DTH (95.28 %), TKM (92.78 %), DTM (87.54 %), GY (80.19 %), PH (78.67 %), SL (74.68 %) and PT (65.95 %) while low for GFP (38.34 %) (Table 6). Similar results have been reported by Addisu and Shumet (2015) and Zewodu et al. (2025). Estimation of heritabil- ity coupled with genetic advance offers the most suitable condition for selection (Singh and Chaudhary, 1985). High heritability and genetic advance were observed for GY, KNS, SPS, PT, DTH, SL and TKM traits (Table 6) which revealed the presence of additive genetic variance in the trait (Johnson et al., 1955) and suggested that reli- able crop improvement could be effective through phe- notypic selection. Similar studies in barley (Angassa, 2021; Kaur et al., 2022; Kumar & Shekhawat, 2013; Shri- mali et al., 2017) and flax (Abeje & Zewodu, 2024; You et al., 2017) also indicated that estimates of heritability with genetic advance are more reliable than individual consideration of the parameters for selection. 4 CONCLUSION Understanding the extent and magnitude of phe- notypic diversity and trait associations is essential for the maximum utilization of gene bank materials in breeding programs and conservation. This study exam- ined the phenotypic diversity of 226 Ethiopian barley landraces and three improved varieties using variance components and multivariate analysis. Significant ge- netic variability was found in most traits, except for GFP, which had non-significance differences and medium heritability. Key traits such as grain yield (GY), days to maturity (DTM), and plant height (PH) showed nota- ble variability, aiding breeders in selection efforts. PCA biplot analysis identified early (Acc-243314, 242070 and 242060) and late heading/maturing types (Acc-243210, 243209,243273,243298 and 243074). In agreement with PCA Biplot results, cluster analysis grouped genotypes into three major clusters according to their morphologi- cal similarity and rarely on their geographical location. Notably, Accession 241790 collected from the Amhara region exhibited low GFP, low PH, and high GY (above the grand mean of 2118.3 kg ha-1) and rarely have com- posite spike forms. The study suggests further multi-lo- cation trials to explore genes linked to composite spikes and other valuable traits. In conclusion, the superior grain yield of landraces compared to improved varieties highlights the potential of Ethiopian farmer varieties for breeding programs aimed at enhancing food security. 5 ACKNOWLEDGMENTS The authors would like to thank the Ethiopian Bio- diversity Institute (EBI) for the financial support during this research and provision of barley accessions from the gene bank. The authors would also like to express their special recognition and thanks to the Holeta Agricultural Research Center Barley Research Group for the provision of experimental land and checks for our study. 6 CONFLICT OF INTEREST The authors declared that the research was conduct- ed in the absence of any commercial or financial relation- ships that could be construed as a potential conflict of interest. 7 SUPPLEMENTARY MATERIALS The following list of supplementary tables and fig- ures are attached with this manuscript. Additional File 1 (Xlsx): Geospatial data of the studied barley accessions (sheet 1; Passport), Mean per- formance of 229 barley genotypes based on 10 morpho- agronomic traits (Sheet 2; Mean performance) and Eu- clidean distance between the studied barley genotypes (Sheet 3; Genetic distance) Figure S1: A scree plot of four dimensions/PCs (a&b) and contribution of variables (c&d)retained from 10 morpho-agronomic traits Acta agriculturae Slovenica, 121/2 – 2025 13 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis 8 FUNDING This research did not receive any specific grant from funding agencies in the public, commercial, or not-for- profit sectors. 9 CREDIT AUTHORSHIP CONTRIBUTION STATEMENT Gizie Abeje: Conceptualization, Visualization, Validation, Methodology, Investigation, Data curation, Formal analysis, Resources, Project administration, Su- pervision, Funding acquisition, Writing–original draft, Writing-review, and editing. Alemayehu Zewodu: Resources, Visualization, Su- pervision, Investigation, Data curation, Validation, Writ- ing-review, and editing. 10 REFERENCES Abebe, T. D., Bauer, A. M., & Léon, J. (2010). Morphological diversity of Ethiopian barleys (Hordeum vulgare L.) in rela- tion to geographic regions and altitudes. Hereditas, 147(4), 154–164. https://doi.org/10.1111/j.1601-5223.2010.02173.x Abeje, G., & Zewodu, A. (2024). Exploring Genetic Vari- ability and Morpho-Agronomic Trait Associations Among Ethiopian Flax (Linum usitatissimum L.) Accessions for Enhanced Fiber and Seed Yield. Journal of Natural Fibers, 21(1). https://doi.org/10.1080/15440478.2024.2391005 Abtew, W. G. (2019). Characterization of genetic variation among Ethiopian barley (Hoerdeum vulgare L.) genotypes. Universität Hohenheim. https://hohpublica.uni-hohenhe- im.de/handle/123456789/6454 Addisu, F., & Shumet, T. (2015). Variability, Heritability and Genetic Advance for Some Yield and Yield Related Traits in Barley (Hordeum vulgare L.) Landraces in Ethiopia. Inter- national Journal of Plant Breeding and Genetics, 9(2), 68–76. https://doi.org/10.3923/ijpbg.2015.68.76 Alemayehu, F., & Parlevliet, J. E. (1997). Variation between and within Ethiopian barley landraces. Euphytica, 94(2), 183– 189. https://doi.org/10.1023/A:1002946004268 Angassa, D. (2021). Genetic Variability of Ethiopian Barley (Hordeum vulgare (L.)) Genotypes for yield and yield re- lated traits. Plant, 9(4), 101. https://doi.org/10.11648/j. plant.20210904.14 Angassa, D., & Mohammed, J. (2021). Genetic diversity assess- ment of Ethiopian landrace barley (Hordeum vulgare L.) genotypes through multivariate analysis. American Journal of Biological and Environmental Statistics, 7(3), 67. https:// doi.org/10.11648/j.ajbes.20210703.12 Angassa, D., & Tesfaye, K. (2019). Diversity study of some Ethiopian barley (Hordeum vulgare L.) accessions for their agronomical Important traits. International Journal of Agri- culture Innovations and Research, 8(1), 2319–1473. Aravind, J., MukeshSankar, S., & Wankhede, D. P. (2021). Aug- mented RCBD: Analysis of augmented randomised com- plete block designs. R package version 0.1. 5.9000. Asfaw, Z. (1988). Variation in the Morphology of the Spike within Ethiopian Barley, Hordeum vulgare L. (Poaceae) . Acta Agriculturae Scandinavica, 38(3), 277–288. https://doi. org/10.1080/00015128809437989 Asfaw, Z. (1989). Relationships between spike morphology, hordeins and altitude within Ethiopian barley, Hordeum vulgare L. (Poaceae). Hereditas, 110(3), 203–209. https:// doi.org/10.1111/j.1601-5223.1989.tb00782.x Assefa, A., Labuschagne, M., & Maartens, H. (2016). Genetic variability in barley (Hordeum vulgare L.) landraces from Ethiopia as measured by morphological characters and SDS-page of seed storage proteins. Ethiopian Journal of Ag- ricultural Sciences, 81, 64–81. Azeez, M. A., Adubi, A. O., & Durodola, F. A. (2018). Landraces and crop genetic improvement. In Rediscovery of Landraces as a Resource for the Future. IntechOpen. Badr, A., M, K., Sch, R., Rabey, H. El, Effgen, S., Ibrahim, H. H., Pozzi, C., Rohde, W., & Salamini, F. (2000). On the ori- gin and domestication history of barley (Hordeum vulgare). Molecular Biology and Evolution, 17(4), 499–510. https:// doi.org/10.1093/oxfordjournals.molbev.a026330 Bedasa, M., Berhane, L., & Tadesse, D. (2015). Morphological diversity and association of traits in Ethiopian food barley (Hordeum vulgare L.) landraces in relation to regions of origin and altitudes. Journal of Plant Breeding and Crop Sci- ence, 7(2), 44–54. https://doi.org/10.5897/jpbcs2014.0480 Benlioglu, B., Bilir, M., Akdogan, G., Ahmed, H. A. A., Ergun, N., Aydogan, S., & Emrebas, T. (2025). Phenotypic char- acterization of two-row barley (Hordeum vulgare L. ssp. vulgare) germplasm conserved in Osman Tosun Genebank of Türkiye by multivariate analysis model. Genetic Re- sources and Crop Evolution, 72(2), 1567-1584. https://doi. org/10.1007/s10722-024-02032-0 Bjørnstad, Å., Demissie, A., Kilian, A., & Kleinhofs, A. (1997). The distinctness and diversity of Ethiopian barleys. Theo- retical and Applied Genetics, 94(3–4), 514–521. https://doi. org/10.1007/s001220050445 Bothmer, R. von. (1992). The wild species of Hordeum: relation- ships and potential use for improvement of cultivated barley. (P. R. Shewry (ed.)). CAB International (Wallingford, UK) Burton, G. W. (1951). Quantitative inheritance in pearl millet (Pennisetum glaucum) 1. Agronomy Journal, 43(9), 409– 417. Ceccarelli, S. (1996). Euphytica, 92, 203–214. Charrad, M., Ghazzali, N., Boiteau, V., & Niknafs, A. (2014). NbClust: an R package for determining the relevant num- ber of clusters in a data set. Journal of Statistical Software, 61, 1–36. https://www.jstatsoft.org/v61/i06/ CSA. (2021). Farm Management Practices (Agricultural Sam- ple Survey) 2020/21 (2013 E.C.). In Central Statistical Agen- cy: Vol. III.https://www.statsethiopia.gov.et/wp-content/ uploads/2021/06/2020_21-2013-E.C-AgSS-Main-Season- Agricultural-Farm-Management-Report.pdf Demissie, A., & Bjørnstad, Å. (1996). Phenotypic diversity of Ethiopian barleys in relation to geographical regions, al- titudinal range, and agro-ecological zones: as an aid to Acta agriculturae Slovenica, 121/2 – 202514 G. A. BELAY and A. Z. KEBEDE germplasm collection and conservation strategy. Hereditas, 124(1), 17–29. Derbew, S. (2020). Multivariate analysis of hulled barley (Hor- deum vulgare L.) landraces of Southern Ethiopia. Cogent Food and Agriculture, 6(1). https://doi.org/10.1080/233119 32.2020.1841357 Derbew, S., Urage, E., & Mohammed, H. (2013). Genetic vari- ability in barley (Hordeum vulgare (L.)) landrace collections from Southern Ethiopia. International Journal of Science and Research, 12(2), 125–131. Dido, A. A., Degefu, D. T., Assefa, E., Krishna, M. S. R., Singh, B. J. K., & Tesfaye, K. (2021). Spatial and temporal genetic variation in Ethiopian barley (Hordeum vulgare L.) lan- draces as revealed by simple sequence repeat (SSR) mark- ers. Agriculture and Food Security, 10(1), 1–14. https://doi. org/10.1186/s40066-021-00336-3 Dido, A. A., Degefu, D. T., Singh, B. J. K., Tesfaye, K., & Krishna, M. S. R. (2020). Multivariate analysis of quantitative char- acters variability in Ethiopian barley (Hordeum vulgare L.) landrace: based on regions and altitude. Genetika, 52(2), 597–620. https://doi.org/10.2298/GENSR2002597D Dziurdziak, J., Bolc, P., Wlodarczyk, S., Puchta, M., Gryziak, G., Podyma, W., & Boczkowska, M. (2020). Multifaceted analy- sis of barley landraces collected during gene bank expedi- tions in poland at the end of the 20th century. Agronomy, 10(12). https://doi.org/10.3390/agronomy10121958 Dziurdziak, J., Gryziak, G., Groszyk, J., Podyma, W., & Boc- zkowska, M. (2021). Dartseq genotypic and phenotypic diversity of barley landraces originating from different countries. Agronomy, 11(11). https://doi.org/10.3390/ agronomy11112330 Engels, J. M. M. (1991). A diversity study in Ethiopian bar- ley. In J. M. M. Engels, J. G. Hawkes, & M. Worede (Eds.), Plant Genetic Resources of Ethiopia (pp. 131–139). Cam- bridge University Press. https://doi.org/DOI: 10.1017/ CBO9780511551543.010 Engels, J. M. M. (1994). Genetic diversity in Ethiopian barley in relation to altitude. Genetic Resources and Crop Evolution, 41(2), 67–73. https://doi.org/10.1007/BF00053050 Enyew, M., Dejene, T., Lakew, B., & Worede, F. (2019). Cluster- ing and principal component analysis of barley (Hordeum volugare L.) Landraces for major morphological traits from North Western Ethiopia. International Journal of Agricul- tural Science and Food Technology, 5(1), 058–063. https:// doi.org/10.17352/2455-815x.000043 Esri. (2014). ArcMap 10.2.2 software. https://www.esri.com/en- us/arcgis/products/arcgis-desktop/resources Ethiopian Biodiversity Institute (2021). Standard Operational Procedure (SOP) of Characterization of crop genetic resourc- es (EBI-SOP-CHA-v1.0). Fantahun, B., Woldesemayate, T., Fadda, C., Gebrehawaryat, Y., Pe, E., & Dell Acqua, M. (2023a). Multivariate analysis in the dissection of phenotypic variation of Ethiopian culti- vated barley (Hordeum vulgare ssp vulgare L.) genotypes. Cogent Food & Agriculture, 9(1), 2157104. Fantahun, B., Woldesemayate, T., & Shiferaw, E. (2023b). The association between hordein polypeptide banding and ag- ronomic traits in partitioning genetic diversity in six-rowed Ethiopian barley lines (Hordeum vulgare L.). BMC Plant Biology, 23(1), 102. Federer, W. T., & Searle, S. R. (1976). Model considerations and variance component estimation in augmented completely randomized and randomized complete blocks designs-pre- liminary version. In Technical Report BU-592-M, Cornell University, New York. Gadissa, F., Abebe, M., & Bekele, T. (2021). Agro-morpholog- ical traits-based genetic diversity assessment in Ethiopian barley (Hordeum vulgare L.) landrace collections from Bale highlands, Southeast Ethiopia. Agriculture and Food Securi- ty, 10(1), 1–14. https://doi.org/10.1186/s40066-021-00335- 4 Gadissa, F., & Gudeta, T. B. (2023). Phenotypic characterization and seed viability test in ex-situ conserved Ethiopian cul- tivated barley (Hordeum vulgare L.) landraces. BMC Plant Biology, 23(1), 1–18. https://doi.org/10.1186/s12870-023- 04628-7 Gegnaw, S. T., & Hadado, T. T. (2014). Genetic diversity of qualitative traits of barley ( Hordeum vulgare L .) landrace populations collected from Gamo Highlands of Ethiopia. 6(9), 663–673. https://doi.org/10.5897/IJBC2014.0718 Hadado, T. T., Rau, D., Bitocchi, E., & Papa, R. (2009). Genetic diversity of barley (Hordeum vulgare L.) landraces from the central highlands of Ethiopia: Comparison between the Belg and Meher growing seasons using morphological traits. Genetic Resources and Crop Evolution, 56(8), 1131– 1148. https://doi.org/10.1007/s10722-009-9437-z Harlan, J. (1968). On the origin of barley. USDA Agriculture Handbook, 338, 9–31. Harlan, J. R. (1969). Ethiopia: a center of diversity. Economic Botany, 23(4), 309–314. Harlan, J., & Zohary, D. (1966). Distribution of wild wheats and barley. Science, 153(3740), 1074–1080. Hyles, J., Bloomfield, M. T., Hunt, J. R., Trethowan, R. M., & Trevaskis, B. (2020). Phenology and related traits for wheat adaptation. Heredity, 125(6), 417–430. https://doi. org/10.1038/s41437-020-0320-1 ICARDA. (2025). Barley | Crops | Resources | CGIAR Gene- bank Initiative. https://genebanks.cgiar.org/resources/ crops/barley/ IPGRI. (1994). Descriptors for barley (Hordeum vulgare L.). International Plant Genetic Resources Institute Rome, Italy. https://alliancebioversityciat.org/publications-data/ descriptors-barley-hordeum-vulgare-l Johnson, H. W., Robinson, H. F., & Comstock, R. E. (1955). Estimates of genetic and environmental variability in soy- beans. Agronomy Journal, 47(7), 314–318. Kandić, V., Dodig, D., Nikolić, A., Kaitović, Ž., Zorić, M., Šurlan-Momirović, G., Aleksić, G., & Đurić, N. (2018). Grain filling parameters of two-and six-rowed barley genotypes in terminal drought conditions. Italian Journal of Agrometeorology-Rivista Italiana Di Agrometeorologia, 23(2), 5–14. Kaso, T. (2015). Review of Barley Value Chain Management in Ethiopia. Journal of Biology, Agriculture and Healthcare, 5(10), 84–98. Kaur, V., Aravind, J., Jacob, S. R., Kumari, J., Panwar, B. S., Pal, N., Rana, J. C., Pandey, A., & Kumar, A. (2022). Phe- Acta agriculturae Slovenica, 121/2 – 2025 15 Dissecting the phenotypic diversity of Ethiopian barley (Hordeum vulgare L.) genotypes through variance components and multivariate analysis notypic Characterization , Genetic Diversity Assessment in 6,778 Accessions of Barley (Hordeum vulgare L . ssp . vulgare ) Germplasm Conserved in National Genebank of India and Development of a Core Set. 13, 1–17. https://doi. org/10.3389/fpls.2022.771920 Kumar, M., & Shekhawat, S. S. (2013). Genetic variability in barley (Hordeum vulgare L.). Electronic Journal of Plant Breeding, 4(4), 1309–1312. Lakew, B., Semeane, Y., Alemayehu, F., Gebre, H., Grando, S., Van Leur, J. A. G., & Ceccarelli, S. (1997). Exploiting the diversity of barley landraces in Ethiopia. Genetic Re- sources and Crop Evolution, 44(2), 109–116. https://doi. org/10.1023/A:1008644901982 Lê, S., Josse, J., & Husson, F. (2008). FactoMineR: an R pack- age for multivariate analysis. Journal of Statistical Software, 25(1), 1–18. https://doi.org/10.18637/jss.v025.i01 Lev-Yadun, S., Gopher, A., & Abbo, S. (2000). The cradle of agriculture. Science, 288(5471), 1602–1603. https://doi. org/10.1126/science.288.5471.1602 Lush, J. L. (1940). Intra-sire correlations or regressions of off- spring on dam as a method of estimating heritability of characteristics. Journal of Animal Science, 1940(1), 293–301. Manly, B. F. J. (1986). Randomization and regression methods for testing for associations with geographical, environmen- tal and biological distances between populations. Research- es on Population Ecology, 28, 201–218. Megersa, G., Mekbib, F., & Lakew, B. (2015). Performance of farmers and improved varieties of barley for yield com- ponents and seed quality. Journal of Plant Breeding and Crop Science, 7(4), 107–124. https://doi.org/10.5897/ JPBCS2013.0436 Mendiburu, F. (2020). Agricolae: Statistical Procedures from Ag- ricultural Research, 1.3-3. R. Milner, S. G., Jost, M., Taketa, S., Mazón, E. R., Himmelbach, A., Oppermann, M., Weise, S., Knüpffer, H., Basterrechea, M., König, P., Schüler, D., Sharma, R., Pasam, R. K., Rut- ten, T., Guo, G., Xu, D., Zhang, J., Herren, G., Müller, T., … Stein, N. (2019). Genebank genomics highlights the di- versity of a global barley collection. Nature Genetics, 51(2), 319–326. https://doi.org/10.1038/s41588-018-0266-x Mohammed, J., Seleshi, S., Nega, F., & Lee, M. (2016). Re- visit to Ethiopian traditional barley-based food. Journal of Ethnic Foods, 3(2), 135–141. https://doi.org/10.1016/j. jef.2016.06.001 Mondal, S., Singh, R. P., Crossa, J., Huerta-Espino, J., Sharma, I., Chatrath, R., Singh, G. P., Sohu, V. S., Mavi, G. S., & Sukuru, V. S. P. (2013). Earliness in wheat: a key to adaptation under terminal and continual high temperature stress in South Asia. Field Crops Research, 151, 19–26. Monteagudo, A., Casas, A. M., Cantalapiedra, C. P., Contreras- Moreira, B., Gracia, M. P., & Igartua, E. (2019). Harnessing novel diversity from landraces to improve an elite barley variety. Frontiers in Plant Science, 10(434), 1–17. https:// doi.org/10.3389/fpls.2019.00434 Nieweglowski, L., & Nieweglowski, M. L. (2015). clv: Cluster Validation Techniques. R package version 0.3-2.4. https:// cran.r-project.org/package=clv Poursarebani, N., Seidensticker, T., Koppolu, R., Trautewig, C., Gawroński, P., Bini, F., Govind, G., Rutten, T., Sakuma, S., Tagiri, A., Wolde, G. M., Youssef, H. M., Battal, A., Cian- namea, S., Fusca, T., Nussbaumer, T., Pozzi, C., Börner, A., Lundqvist, U., … Schnurbusch, T. (2015). The genetic ba- sis of composite spike form in barley and ‘miracle-wheat.’ Genetics, 201(1), 155–165. https://doi.org/10.1534/genet- ics.115.176628 Poursarebani, N., Trautewig, C., Melzer, M., Nussbaumer, T., Lundqvist, U., Rutten, T., Schmutzer, T., Brandt, R., Him- melbach, A., Altschmied, L., Koppolu, R., Youssef, H. M., Sibout, R., Dalmais, M., Stein, N., Xin, Z., & Schnurbusch, T. (2020). COMPOSITUM 1 contributes to the architec- tural simplification of barley in florescence via meristem identity signals. Nature Communications, 11(5138). https:// doi.org/10.1038/s41467-020-18890-y Rashid, S., Abate, G. T., Lemma, S., Warner, J., Kasa, L., & Mi- not, N. (2019). The barley value chain in Ethiopia. Gates Open Res, 3(169), 169. Robinson, H. F. (1966). Quantitative genetics in relation to breeding on centennial of Mendelism. Indian Journal of Ge- netics and Plant Breeding, 171. Schloerke, B., Cook, D., Joseph Larmarange, Briatte, F., Mar- bach, M., Thoen, E., Elberg, A., & Crowley, J. (2024). GGal- ly: Extension to ’ggplot2’_. R package version 2.2.1. Setotaw, T. A., Dias, L. A. dos S., & Missio, R. F. (2010). Ge- netic divergence among barley accessions from Ethiopia. Crop Breeding and Applied Biotechnology, 10(2), 116–123. https://doi.org/10.12702/1984-7033.v10n02a03 Shiferaw, T., Abate, B., & Lakew, B. (2020). Genetic variabil- ity and association of traits in Ethiopian barley (Hordeum vulgare L.) genotypes at Holetta, Central Ethiopia. Journal of Agricultural and Crop Research, 8(1), 11–19. https://doi. org/10.33495/jacr_v8i1.19.171 Shrimali, J., Shekhawat, A. S., & Kumari, S. (2017). Genetic variation and heritability studies for yield and yield compo- nents in barley genotypes under normal and limited mois- ture conditions. Journal of Pharmacognosy and Phytochem- istry, 6(4), 233–235. Shtaya, M. J. Y., Abdallah, J., Al-Fares, H., Abu-Qaoud, H., Baker, O. A., Korf, M. V., & Haddad, M. (2015). Detect- ing genetic diversity among barley landraces grown in the West-Bank, Palestine in 2010-2011. Journal of Animal and Plant Sciences, 25(5), 1365–1370. Singh, R. K., & Chaudhary, B. D. (1985). Biometrical methods in quantitative genetic analysis. Kalyani Publishers. Sivasubramaniam S, M. P. (1973). Genotypic and phenotypic variability in rice. The Madras Agricultural Journal, 60(9– 13), 1093–1096. Sneath, P. H. A and Sokal, R. R. (1973). Numerical taxonomy. Theory and Application of Genetics, 93, 613–617. Tamirat Bejiga and Amare Seyoum. (2018). A Training Man- ual for Trainee on Knowledge Management On Electronic Data Capturing Volume 5. Team, R. C. (2024). R: A Language and Environment for Statis- tical Computing. R Foundation for Statistical Computing. https://www.r-project.org/ Teklemariam, S. S., Bayissa, K. N., Matros, A., Pillen, K., Ordon, F., & Wehner, G. (2022). The genetic diversity of Ethiopi- an barley genotypes in relation to their geographical ori- Acta agriculturae Slovenica, 121/2 – 202516 G. A. BELAY and A. Z. KEBEDE gin. PLoS ONE, 17, 1–17. https://doi.org/10.1371/journal. pone.0260422 Terzi, V., Tumino, G., Pagani, D., Rizza, F., Ghizzoni, R., Morcia, C., & Stanca, A. M. (2017). Barley developmental mutants: The high road to understand the cereal spike morphology. Diversity, 9(2), 1–16. https://doi.org/10.3390/d9020021 Vavilov, N. I. (1951). The origin, variation, immunity and breed- ing of cultivated plants, 72(6). The Chronica Botanica Co., Waltham, Mass., and Stechert-Hafner, Inc. Verma, S., Yashveer, S., Rehman, S., Gyawali, S., Kumar, Y., Chao, S., Sarker, A., & Verma, R. P. S. (2021). Genetic and Agro-morphological diversity in global barley (Hordeum vulgare L.) collection at ICARDA. Genetic Resources and Crop Evolution, 68(4), 1315–1330. https://doi.org/10.1007/ s10722-020-01063-7 Vu, V. Q. (2011). ggbiplot: A ggplot2 based biplot. R Package Version 0.55, 755. Wickham, H., Chang, W., & Wickham, M. H. (2016). Pack- age ‘ggplot2.’ Create Elegant Data Visualisations Using the Grammar of Graphics. Version, 2(1), 1–189. Wickham, H., & Seidel, D. (2020). Scales: Scale Functions for Visualization. R package version 1.1. 1. Wosene, G. A., Berhane, L., Bettina, I. G. H., & Karl, J. S. (2015). Ethiopian barley landraces show higher yield stability and comparable yield to improved varieties in multi-environ- ment field trials. Journal of Plant Breeding and Crop Science, 7(8), 275–291. https://doi.org/10.5897/jpbcs2015.0524 Yadav, R. K., Gautam, S., Palikhey, E., Joshi, B. K., Ghimire, K. H., Gurung, R., Adhikari, A. R., Pudasaini, N., & Dhakal, R. (2018). Agro-morphological diversity of Nepalese naked barley landraces. Agriculture and Food Security, 7(1), 1–12. https://doi.org/10.1186/s40066-018-0238-5 Yahiaoui, S., CuestaMarcos, A., Gracia, M. P., Medina, B., Lasa, J. M., Casas, A. M., Ciudad, F. J., Montoya, J. L., Moralejo, M., & Molina Cano, J. L. (2014). Spanish barley landrac- es outperform modern cultivars at low productivity sites. Plant Breeding, 133(2), 218–226. Yang, W., Peng, S., Dionisio-sese, M. L., Laza, R. C., & Visperas, R. M. (2008). Grain filling duration, a crucial determinant of genotypic variation of grain yield in field-grown tropical irrigated rice. Field Crops Research, 105(3), 221-227. https:// doi.org/10.1016/j.fcr.2007.10.006 You, F. M., Jia, G., Xiao, J., Duguid, S. D., Rashid, K. Y., Book- er, H. M., & Cloutier, S. (2017). Genetic variability of 27 traits in a core collection of flax (Linum usitatissimum L.). Frontiers in Plant Science, 8, 1636. https://doi.org/10.3389/ fpls.2017.01636 Zadoks, J. C., Chang, T. T., & Konzak, C. F. (1974). A deci- mal code for the growth stages of cereals. Weed Research, 14, 415–421. https://doi.org/10.1111/j.1365-3180.1974. tb01084.x Zewodu, A., Mohammed, W., & Shiferaw, E. (2025). Genetic variability and association of morpho-agronomic traits among Ethiopian barley (Hordeum vulgare L ) accessions. Scientifica, 2025(1), 3957883. https://doi.org/10.1155/ sci5/3957883 Zhu, Y., Chen, H., Fan, J., Wang, Y., Li, Y., Chen, J., Fan, J. X., Yang, S., Hu, L., Leung, H., Mew, T. W., Teng, P. S., Wang, Z., & Mundt, C. C. (2000). Genetic diversity and disease control in rice. Nature, 406(6797), 718–722. https://doi. org/10.1038/35021046