GENETIC IDENTITY, EPIDEMIOLOGY AND MANAGEMENT OF FABA BEAN (Vicia faba L.) GALL DISEASE IN ETHIOPIA PhD DISSERTATION BEYENE BITEW ESHETE MARCH 2022 HARAMAYA UNIVERSITY, HARAMAYA Genetic Identity, Epidemiology and Management of Faba Bean (Vicia faba L.) Gall Disease in Ethiopia A Dissertation Submitted to the Postgraduate Programs Directorate (School of Plant Sciences) Haramaya University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy (PhD) in Plant Pathology Beyene Bitew Eshete March 2022 Haramaya University, Haramaya POSTGRADUATE PROGRAMS DIRECTORATE HARAMAYA UNIVERSITY We hereby certify that we have read and evaluated this PhD dissertation, which is entitled “Genetic Identity, Epidemiology and Management of Faba Bean (Vicia faba L.) Gall Disease in Ethiopia”, and has been prepared under our guidance by Beyene Bitew Eshete. We recommend that the Dissertation be submitted as it fulfills the Dissertation reqiurments. 1. Prof. Chemeda Fininsa (PhD) Chairperson, Advisory Committee Signature Date 2. Habtamu Terefe (PhD) Member, Advisory Committee Signature Date 3. Prof. Martin Berbetti (PhD) Member, Advisory Committee Signature Date 4. Seid Ahmed Kemal (PhD) Member, Advisory Committee Signature Date As member of the Board of Exminers of the PhD Dissertation Open Defense Examination, we certify that we have read and evaluated the Dissertation prepared by Beyene Biteew Eshete and examined the candidate. We recommend that the Dissertation be accepted as fulfilling the Dissertation requirements for the Degree of Doctor of Philosophy in Plant Pathology 1. Chairperson Signature Date 2. Internal Examiner Signature Date 3. External Examiner Signature Date iii DEDICATION This dissertation work is dedicated to my late beloved parents, Mr Bitew Eshete and Mrs Denkie Kebede, my beloved wife Senait Mekonen, and my children Amanuel Beyene and Leul Beyene. iv STATEMENT OF THE AUTHOR By my signature below, I declare and confirm that this dissertation is my own work accounted for my research and that all sources of materials used for its preparation have been duly acknowledged. The dissertation has been submitted in partial fulfillment of the requirements for the Degree of Doctor of Philosophy in Plant Pathology at Haramaya University and is deposited at the University’s Library to be made available to borrowers under the rules and regulations of the Library. I solemnly declare that this dissertation has not been submitted to any other institution anywhere for the award of any academic degree, diploma or certificate. Brief quatations from this dissertation are allowable without requiring special permission provided that an accurate acknowledgement of source is made. Requests for permission for extended quotation from the reproduction of the manuscript in whole or in part may be granted by the Head of the School of Plant Sciences or the Director of the Postgraduate Programs Directorate when in his or her judgment the proposed use of the material is for a scholarly interest. In all other instances, however, permission must be obtained from the author of the dissertation. Generally, the work presented in this dissertation is entirely my own work, unless specifically acknowledged otherwise. Name: Beyene Bitew Eshete Signature: Date: March 2022 School: Plant Sciences, Haramaya University, Ethiopia v BIOGRAPHICAL SKETCH The author was born in Taem, Degadamot, West Gojjam Zone, Amhara National Regional State, in March 1968. He attended his primary and junior secondary education at Taem and Feres Bete, respectively. He pursued his senior secondary education at Debre Markos Comprehensive Secondary Schoo1 and completed in 1985. He joined the then Awassa College of Agriculture in 1985/86 and graduated with Diploma in General Agriculture in 1987. Following his graduation, the author was employed by the Ethiopian Institue of Agricultural Research and served as Technical Assistant at different positions at Mechara and Jima Agricultural Research Centers and finally transferred to Debre Birhan Agricultural Research Center. He joined the former Debub University (now Hawassa University) for his Bachlor Degree in 2003 and graduated in Plant Production and Dryland Farming in 2006. After four years of services as an Assistant Researcher at Debre Birhan Agricultural Research Center, he joined the School of Graduate Studies of Haramaya University to pursue his Master of Science (MSc) Degree in Plant Pathology and graduated in 2012. After graduation, he had been working as a Researcher and Head of Cereal and Pulse Crops Case Team Leader and as Director for Debre Birhan Agricultural Research Center for three years until he joined the Postgraduate Programs Directorate of Haramaya University to pursue for his Doctor of Philosophy (PhD) study in Plant Pathology as of October 2016. vi ACKNOWLEDGMENTS First and foremost, I would like to praise my Heavenly Father, God, for He has the supernatural power to decide on all my deeds. I am highly indebted to express my sincere gratitude and utmost respect to Prof. Chemeda Fininsa, my major supervisor and the former President of Haramaya University. I thank you wholeheartedly for your continuous guidance, mentorship and outstanding contributions during this work. You provided me with tremendous constructive and valuable comments and suggestions. I would like to express my sincere gratitude to my co-supervisor, Dr Habtamu Terefe, for his entire support during proposal development, dissertation writing and editing all my muniscripts to have international standards and repute. I also want to thank Prof. Martin Barbetti for his support in laboratory and pathogen identification works both in Ethiopia and Australia, and budget support for field experiments and during my stay at University of Western Australia. My thanks also go to Dr Seid Ahmed, my co-supervisor, for supporting me in all areas (technical and financial) during the study. I appreciate your efforts made to secure additional research supports from Africa RISING (AR) project funded by USAID through ILRI to start the Dissertation research. The research work was supported by the Australian Centre for International Agricultural Research, (ACIAR Project: CIM/2017/030 “Faba Bean in Ethiopia –Mitigating disease constraints to improve productivity and sustainability”), the Debre Birhan Agricultural Research Centre, the New South Wales Department of Primary Industries (NSW DPI) and the University of Western Australia, Australia, and the International Center for Agricultural Research in the Dry Areas (ICARDA), Morocco. I would like to thank Amhara Regional Agricultural Research Institute for sponsoring my salary during my study leave. I would like to thank Debre Birhan Agricultural Research Center for arranging vehicle for field works. My special thanks go to Ato Aleme Belete, Amare Belachew, Alemayehu Ayele, Alemnew Fantaye, Semegnew Anelye and W/t Mersha Ferdawok for assisting me in both field and laboratory works. My appreciation goes to Soil and Water Laboratory Department researchers, especially Ato Getaneh Shegaw and Lesanu Getaneh, for soil physical and chemical properties analytical services. The driving services provided by Ato Wubet Habtamu and Felek Asgelet are kindly acknowledged. Last but not least, I am grateful to all my family members, especially my soul mate Senait Mekonen and my children Amanuel and Leul Beyene, for their lovely treatments that have made me happier and more energetic on my overall study to completion. Without your vehement support and moral encouragement, the study would have not been realized. vii ACRONYMS AND ABBREVIATIONS AEA Average Environment Axis AEC Average Eenvironment Coordinates AUDPC Area Under Disease Progress Curve BLAST Basic Local Alignment Search Tools CSA Central Statistical Agency DAP Days After Planting DMRT Duncan’s Multiple Range Test DNA Deoxyribonucleic Acid FBG Faba Bean Gall EBI Ethiopian Biodiversity Institute FTA Flinders Technology Associates GLM General Linear Model HSW Hundred Seed Weight ICARDA International Center for Agricultural Research in the Dry Areas ILRI International Livestock Research Institute IPCA Interaction Principal Component Analysis/Axis ITS Internal Transcribed Spacer LRTs Likelihood Ratios LSD Least Significant Difference LSU Large Subunit MSA Multiple Sequence Alignment NCBI National Center for Biotechnology Information PCR Polymerase Chain Reaction PSI Percentage Severity Index rpm revolution per minute RYL Relative Yield Loss SAS Statistical Analysis System SSU Small Sub-Unit viii TABLE OF CONTENTS STATEMENT OF THE AUTHOR................................................................ v BIOGRAPHICAL SKETCH ........................................................................ vi ACKNOWLEDGMENTS ............................................................................ vii ACRONYMS AND ABBREVIATIONS ..................................................... viii TABLE OF CONTENTS ............................................................................... ix LIST OF ARTICLES/MANUSCRIPTS ..................................................... xiii ABSTRACT ................................................................................................. xiv 1. INTRODUCTION....................................................................................... 1 1.1. Faba Bean Productions ................................................................................. 1 1.2. Faba Bean Production Constraints ............................................................... 2 1.3. Faba Bean Gall Disease ............................................................................... 2 1.4. Distribution of FBG Disease ........................................................................ 3 1.5. Symptomatology and Host Ranges............................................................... 4 1.6. Disease Epidemiology .................................................................................. 6 1.7. Faba Bean Gall Disease Management .......................................................... 7 1.7.1. Cultural practices........................................................................................................7 1.7.2. Host plant resistance ...................................................................................................8 1.7.3. Chemical control ........................................................................................................8 1.7.4. Integrated FBG disease management ..........................................................................9 1.8. Justification .................................................................................................10 1.9. Objectives ...................................................................................................12 2. MATERIALS AND METHODS ...............................................................13 2.1. Description of the Study Areas....................................................................13 2.2. Faba Bean Gall Disease Survey ..................................................................14 2.3. Molecular Identification of FBG Disease-Causing Pathogen ......................16 2.3.1. Disease symptoms and pathogen morphology ........................................................... 16 2.3.2. Collection of FBG disease samples for DNA analyses .............................................. 16 ix TABLE OF CONTENTS (Continued) 2.3.3. Deoxyribonucleic acid (DNA) extraction and PCR conditions .................................. 17 2.3.4. Primer pairs tested .................................................................................................... 18 2.3.5. Phylogenic tree building ........................................................................................... 18 2.4. Estimating Yield Loss of Faba Bean Caused by Gall Disease in North Shoa, Ethiopia ......................................................................................................18 2.5. Evaluation of Faba Bean Genotypes for FBG Resistance Reaction .............20 2.6. Reaction of Faba Bean Genotypes to Gall Disease and Resistance Reaction Stability under Natural Infections ...............................................................21 2.7. Integrated Management of Gall Disease on Faba Bean in North Shoa, Ethiopia ......................................................................................................22 2.8. Data Analysis ..............................................................................................23 3. RESULTS ...................................................................................................26 3.1. Faba Bean Gall Disease Survey ..................................................................26 3.1.1. Symptomatology and alternate hosts ......................................................................... 26 3.1.2. Distribution, prevalence, incidence and severity of FBG ........................................... 26 3.1.3. Associations of FBG epidemics with biophysical factors .......................................... 27 3.2. Molecular Identification of FBG-Causing Pathogen....................................28 3.2.1. Disease symptoms and pathogen morphology ........................................................... 28 3.2.2 Sequence BLAST results ........................................................................................... 29 3.2.3. Phylogenetic analysis ............................................................................................... 29 3.3. Estimating Yield Loss of Faba Bean Caused by Gall Disease in North Shoa, Ethiopia ......................................................................................................30 3.3.1. Faba bean gall severity ............................................................................................. 30 3.3.2. Area under disease progress curve (AUDPC) ........................................................... 31 3.3.3. Effect of gall disease on grain yield of faba bean ...................................................... 31 3.3.4. Effect of gall disease on hundred seed weight of faba bean ....................................... 32 3.3.5. Effect of gall disease on yield components of faba bean ........................................... 32 3.3.6. Relative yield loss of faba bean due to FBG disease ................................................. 33 x TABLE OF CONTENTS (Continued) 3.3.7. Correlation of FBG disease parameters with grain yield and yield-components ........ 34 3.3.8. Regression analysis relating gall disease severity with faba bean grain yield ............. 34 3.4. Evaluation of Faba Bean Genotypes for FBG Disease Resistance Reaction 35 3.4.1. Faba bean gall disease incidence and severity ........................................................... 35 3.4.2. Area under disease progress curve (AUDPC) ........................................................... 36 3.4.3. Faba bean disease progress rate ................................................................................ 36 3.4.4. Grain yield and yield components ............................................................................. 37 3.4.5 Association of FBG disease and yield parameters ...................................................... 37 3.5. Reaction of Faba Bean Genotypes to Gall Disease and Resistance Reaction Stability under Natural Infections ...............................................................38 3.5.1. AMMI analysis of FBG disease incidence, severity and grain yield of faba bean ...... 38 3.5.2. GGE biplot analysis of FBG disease and grain yield of faba bean ............................. 39 3.6. Integrated Management of Gall Disease of Faba Fean in North Shoa, Ethiopia ......................................................................................................42 3.6.1. Faba bean gall disease epidemic ............................................................................... 42 3.6.2. Faba bean gall progress curve ................................................................................... 43 3.6.3. Area under disease progress curve (AUDPC) ........................................................... 43 3.6.4. Rate of FBG progression .......................................................................................... 44 3.6.5. Grain yield of faba bean ........................................................................................... 45 3.6.6. Association of faba bean gall disease and grain yield ................................................ 45 3.6.7. Partial budget analysis .............................................................................................. 45 4. DISCUSSION .............................................................................................47 5. CONCLUSIONS AND RECOMMENDATIONS ....................................62 5.1. Conclusions ................................................................................................62 5.2. Recommendations .......................................................................................64 6. FUTURE RESEARCH DIRECTIONS ....................................................66 7. REFERENCES...........................................................................................67 8. APPENDICES ............................................................................................77 xi TABLE OF CONTENTS (Continued) Paper I ...............................................................................................................78 Paper II ............................................................................................................ 107 Paper III ........................................................................................................... 132 Paper IV........................................................................................................... 163 Paper V ............................................................................................................ 191 Paper VI........................................................................................................... 226 xii LIST OF ARTICLES/MANUSCRIPTS The presented dissertation is based on the following papers, which are referred to in the text by their Roman numerals. I. Beyene Bitew, Chemeda Fininsa, Habtamu Terefe, Martin Berbetti and Seid Ahmed. 2021. Spatial and temporal distribution of faba bean gall (Physoderma) disease and its association with biophysical factors in Ethiopia (Published in: International Journal of Pest Management, https://doi.org/10.1080/09670874.2021.1998724). II. Ming Pei You, Beyene Bitew Eshete, Seid Ahmed Kemal, Joop van Leur and Martin J. Berbetti. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia (Published in: Plant Pathology, https://doi.org/10.1111). III. Beyene Bitew, Chemeda Fininsa and Habtamu Terefe. 2021. Estimating yield loss of faba bean (Vicia faba L.) caused by gall disease in North Shoa, Ethiopia (Published in: Crop Protection, https://doi.org/10.1016). IV. Beyene Bitew, Chemeda Fininsa and Habtamu Terefe. Evaluation of faba bean genotypes for FBG disease resistance reaction (Manuscript). V. Beyene Bitew, Chemeda Fininsa and Habtamu Terefe. AMMI and GGE biplot models reveal high genotype contribution for gall disease resistance reaction and grain yield stabilities in faba bean (Manuscript submitted to Euphytica). VI. Beyene Bitew, Chemeda Fininsa and Habtamu Terefe, Seid Ahmed. Integrating genotype, fungicide and spray schedule reduces gall (Physoderma) disease progression and enhances grain yield in faba bean (Accepted in: Journal of Crop Science and Biotechnology). xiii Genetic Identity, Epidemiology and Management of Faba Bean (Vicia faba L.) Gall Disease in Ethiopia Beyene Bitew (MSc), Haramaya University, BSc Hawassa University, Ethiopia ABSTRACT Ethiopia is the world’s second largest producer of faba bean (Vicia faba L.), and the crop has largest share of area and production of all pulses grown in Ethiopia. Faba bean is valued for its source of protein, income, and animal feedstock, and an important rotational crop. However, its current productivity is very low due to an emerging faba bean gall (FBG) disease and other biotic constraints. Initially, FBG disease-causing pathogen remained unconfirmed, but it was assumed to be caused by Olpidium sp. There also has been relatively a few data regarding the distribution of FBG, genetic identity of the pathogen, its damage on faba bean yield, reaction of genotypes and management methods in Ethiopia. Thus, the objectives of this study were to: (1) assess the distribution, disease intensity and host ranges of FBG; (2) determine the association of FBG intensity with major biophysical factors; (3) identify the genetic identity of FBG disease causal pathogen;(4) estimate yield losses caused by the disease; (5) evaluate phenotypic reactions of faba bean genotypes against the disease; (6) determine the stability of host resistance reactions to FBG disease under different agro-ecological conditions; and (7) develop an integrated FBG disease management options. In the FBG disease survey, a total of 783 faba bean fields were assessed across 14 districts (Bassona Worana, Ankober, Tarmaber, Asagert, Degem, Chole, Debay Telatgen, Sinan, Debark, Farta, Laygaint, Meket, Dessie Zuria and Enda Mehoni) and FBG disease prevalence and intensity, independent variables and alternate hosts were recorded. The associations of independent variables with disease incidence and severity were analyzed using logistic regression model. In the causative agent identification, crude DNA extraction and fixing on FTA cards and the morphological identifications were conducted at Debre Birhan Agricultural Research Center and International Livestock Research Institute (ILRI), Ethiopia. Molecular identifications were processed at Debre Birhan Agricultural Research Center (crude DNA extraction and fixing on FTA cards) and finally conducted at University of western Australia, Australia. The sequenced sample DNA were aligned using Geneious Prime version 2020.03 and then resulting consensus were BLAST in Genbank (NCBI). xiv Yield loss assessment experiments were conducted on farmers’ fields at Bassona Worana using two moderately tolerant (Degaga and Gora) faba bean varieties and one susceptible local cultivar, two systemic fungicides [Bayleton 25 WP (Triadimefon 250 g kg-1), and Ridomil Gold MZ 68 WG (Metalaxyl 40+Mancozeb 640 g kg-1)] and two application schedules (10 and 15 days interval). The treatments were arranged in factorial combinations in a randomized complete block design (RCBD) with three replications. Faba bean gall severity, grain yield and yield-components for each treatment were collected and subjected to analysis of variance (ANOVA) using the PROC GLM procedure. Similarly, 415 faba bean genotypes, including four check varieties (Degaga, Gachena, Gora and local cultivar) were evaluated under natural field infections at farmers’ fields in augmented design with ten blocks at Bassona Worana during 2018 main cropping season. Among these 415 genotypes, 104 genotypes, which showed low FBG severity during 2018, were advanced in the 2019 cropping season for further evaluation using similar design and check varieties. Faba bean gall disease incidence, severity and grain yield data were collected and computed. Multi-location experiments were conducted in six locations at Bassona Worana, Debay Telatgen and Farta districts in the 2018 and 2019 main cropping seasons under natural infections using 21 faba bean genotypes arranged in a randomized complete block design with three replications. Disease incidence, severity and grain yield data were collected during the study periods. Additive main effect and multiplicative interaction (AMMI) and genotype and genotype x environment (GGE) biplot models were used to evaluate genotype stability reactions. Field-based integrated FBG disease management experiments were also conducted on farmers’ fields in Bassona Worana district during the 2018 and 2019 main cropping seasons. The treatments were three faba bean genotypes; two moderately tolerant faba bean genotypes (Degaga and Gora), and one susceptible local cultivar, two fungicides (Bayleton and Ridomil Gold) and two application schedules (10 and 15 days), which were arranged in a factorial combination in a randomized complete block design (RCBD) with three replications. Faba bean gall severity and grain yield data were collected and subjected to analysis of variance using the PROC GLM SAS procedure. The results of the survey revealed that FBG was found as a major production constraint in all the surveyed 14 districts considered with variable levels of incidence and severity. The highest (64%) mean FBG severity was recorded in Sinan district during 2018 and in Ankober (45%) during 2019. District, altitude (≥ 2700 m), poorly drained soil, high weed and crop density, flowering growth stage, manure xv application, and early or late planting showed a highly significant (p 0.0001) association with high FBG incidence and severity. In the causal agent identification study, microscopic examination from infected faba bean leaves and stems confirmed an epibiotic phase of zoosporangia for dispersing zoospores, which are characteristic of Physoderma. The morphology did not show critical diagnostic characteristics of Olpidium viciae, such as presence of numerous short zoosporangial discharging tubes, or binucleate resting sporangia. Sequences derived from symptomatic tissue from partial ITS1-5.8S-partial ITS2, 18S-ITS1-5.8S-ITS2-part of 28S, and LSU (S28) all confirmed Physoderma and not Olpidium viciae, as the causal agent of FBG. The results of yield loss estimation study revealed that use of different faba bean varieties and applications of fungicides significantly (p0.05) reduced FBG epidemics and increased grain yield. In 2018, (62.5 and 54.1%), and in 2019 (42.2. and 45.1%), yield losses were calculated from unsprayed plots of variety Degaga and the local cultivar, respectively. Genotypes evaluation results showed that there were highly significant differences among the genotypes for FBG disease severity, AUDPC and grain yield as compared to check varieties. Low (37%) mean FBG severity and maximum (3.78 t ha-1) grain yield was recorded on faba bean accession number 1085. In the multi-location experiments, the AMMI and GGE biplot models analyses revealed highly significant (p0.001) variations among genotypes, environments and genotype x environment interactions (GEI) for FBG disease incidence. The genotypes contributed (80.32%) to the total variation observed far larger than the contributions from the environment (5.32%) and GEI (14.36%) for FBG incidence. Genotypes also showed larger (55.84%) contributions, followed by environment (37.83%) and GEI (6.33%) to the variability demonstrated for FBG severity. Similarly, the variation in grain yield was highly attributed to genotypes (50.86%), followed by environment (38.53%) and GEI (10.61%). Both in the AMMI and GGE biplot models analyses, G3, G16 and G17 showed low FBG severity, but less stable than the rest genotypes; and G7 showed low FBG severity and was stable. However, G1, G2, G5, G13, and G21 were susceptible at all test locations (Bassona Worana, Debay Telatgen and Farta districts) in main rainy seasons, in 2018 and 2019. The site Bassona Worana (E1) provided the best discriminating ability for the genotypes against FBG disease severity. On the other hand, G3, G4, G7, G16, and G17 produced high mean grain yield, but G8 gave high grain yield and was stable too. In developing IDM, integration of faba bean genotypes and fungicides significantly (p 0.05) reduced FBG disease epidemics and increased grain yield. The xvi variety Gora sprayed with Bayleton at a rate of 0.7 kg ha–1 and at 10-days interval had low mean FBG disease severity (21.67 and 10%), AUDPC value (1866.7 and 751.7%-days), low disease progress rate (0.0125 and 0.0121 units day-1) and high mean grain yield (3.70 and 5.03 t ha–1) in 2018 and 2019 main cropping seasons, respectively. Highest (3332.3%-days) AUDPC value was calculated from the unsprayed local cultivar in the 2018 main cropping season. Similarly, high marginal rates of return (7.16, 6.59 and 6.45) were recorded from the varieties Degaga, Gora and local sprayed Bayleton at 15 days interval in 2018. High marginal returns of 8.92 and 7.55 were obtained from the local cultivar and the variety Degaga in 2019 sprayed with Bayleton at 15 days interval. However, high marginal rate of return (8.85) was calculated from the variety Gora sprayed with Bayleton at 10 days interval in 2019. Generally, the findings of this study indicated that FBG is a major problem in faba bean production areas and the disease severity could be reduced by proper soil drainage; weed management, adjusting crop plant density, and following recommended planting time along with use of resistant/tolerant faba bean varieties. Symptomatology, morphological and molecular characterization confirmed that the causal agent of FBG disease is Physoderma. Integration of resistant/tolerant varieties and fungicide application reduces yield losses of faba bean due to FBG disease. Twenty-nine faba bean genotypes from the genotype evaluation experiment and four genotypes (G3, G7, G16 and G17) from multi-location experiments were identified for further evaluation and for the development of disease resistant/toleratn and high yielding varieties through crossing/breeding. Integration of moderately resistant/tolerant faba bean varieties (like Gora) with Bayleton fungicide, and spraying at the onset of the disease aligning at seedling, vegetative, flowering and podding growth stages are recommended to manage sustainably FBG disease and to increase faba bean grain yield at the study locations and other areas with similar agro-ecologies. Keywords: AUDPC, Biophysical Factors, Disease Progress Rate, Fungicide, Genotype, Incidence, Olpidium viciae, Physoderma, Severity, Yield Loss. xvii 1. INTRODUCTION 1.1. Faba Bean Productions Faba bean (Vicia faba L.) is the major cool-season food legume crop grown in many countries in the world (Jensen et al., 2010). Among legumes grown in the world, faba bean ranks fourth after field pea (Pisum sativum), chickpea (Cicer arietinum) and lentil (Lens culinaris), but third next to soyabean (Glycine max) and field pea in production per hectare (FAOSTAT, 2018). It is grown under rainfed and irrigated conditions in more than 60 countries on 2.58 million hectares with a production of nearly 5.4 million tons (FAOSTAT, 2021). The leading faba bean-producing countries are China, Ethiopia, Australia, France, United Kingdom, Egypt and Morocco (FAOSTAT, 2018). Ethiopia is considered as the second center of diversity and also one of the nine major agro-geographical production regions of faba bean in the world (Abebe et al., 2020). The crop takes the largest share of the area and production of all the pulses grown in Ethiopia with a current productivity of 2.12 t ha-1 (CSA, 2021). Over 4.3 million households grow the crop covering about half a million hectares and total production of more than one million tons of grain in 2021 main cropping season in Ethiopia. Oromia National Regional State is the leading faba bean-producer, followed by Amhara, and Southern Nations, Nationalities and Peoples’ Region (Table 1). The crop is valuable as source of protein for human food, cash and its straw is key animal feed in the crop-livestock farming system of the highlands of Ethiopia. In crop rotation practice, faba bean plays a significant role in soil fertility restoration, which contributes in reducing carbon footprints of cereal based agricultural production (Agegnehu and Fessehaie, 2006). However, there are many constraints on the productivity of faba bean in the country. Table 1. Production and yield of faba bean for private farmers’ holdings for the 2020/21 main cropping season, Ethiopia (CSA, 2021). Number of farm Area Production Productivity Administrative Region holders (ha) (tons) (t ha˗1) Oromia 1,590,965 230,114.95 536,430 2.3 Amhara 1,343,432 180,245.56 346,352 1.9 SNNPR 1,092,740 78,285.14 158,995 2.0 Tigray 214,391 11,868.22 19,709 1.7 Binishangul- Gumuz 16,685 733.64 1,327 1.8 Total 4,258,213 501,248 1,062,813 - 1 1.2. Faba Bean Production Constraints Despite its tremendous genetic diversity and multiple benefits to smallholder communities, faba bean productivity is very low in Ethiopia, mainly due to susceptibility to several biotic and abiotic stresses as well as lack of improved varieties and poor agronomic practices (Mussa et al., 2008; Tadesse et al., 2008). The major abiotic constraints for faba bean production are water-logging, soil acidity, frost and drought, and biotic stresses are caused by fungal diseases (foliar and root); parasitic weeds (Orobanche spp.) and storage insect pests (Bruchids). Plant diseases are among the most important biotic factors causing faba bean yield reduction (Yohannes, 2000). The most important diseases of faba bean in Ethiopia include chocolate spot (Botrytis fabae), ascochyta blight (Ascochyta fabae), rust (Uromyces viciae-fabae), black root rot (Fusarium solani and F. avenaceum) and viruses, like Faba bean necrotic yellows virus (FBNYV). Each of these diseases can reach a severity that causes huge yield loss and, in some cases, can cause complete crop failure (Dereje and Tesfay, 1993; Safaa and Khaled, 2007; Tadesse et al., 2008). Chocolate spot, rust and black root rot remain major and persistent yield limiting constraints in the country (Sahile et al., 2008; Terefe et al., 2015). In addition to the common faba bean disease reported, faba bean gall (FBG) is an emerging and a major production constraint in the country (Hailu et al., 2014; Beyene, 2015; Yitayih et al., 2021b). 1.3. Faba Bean Gall Disease Currently, the major faba bean-producing areas, especially central, northern and northwestern parts of Ethiopia are seriously challenged and threatened by faba bean gall (FBG) disease. Faba bean gall is an emerging and devastating disease, initially localized in North Shoa (Amhara and Oromia Regional States) and South Tigray Zones (Hailu et al., 2014; Teklay et al., 2014). However, the disease is expanding throughout all faba bean-production areas of Amhara Regional State, West Shoa and East Wellega Zones (Challa et al., 2017). In severe infections of susceptible faba bean varieties, total crop failure was observed on farmers' fields and farmers are forced to replace faba bean sown fields by other early maturing crops (Teklay et al., 2014; Beyene, 2015). Since 2010, FBG disease has been spreading and caused huge economic losses on the major faba bean-production areas of the country. The disease might have existed long in the country, but it was not recognized until 2010 (Dereje et al., 2012; Bitew and Kebede, 2012). The latter stage of the disease symptom is similar to chocolate spot, so extension workers were advising farmers to spray fungicides recommended to manage the disease (Dereje et al., 2012). Recently, FBG disease is spreading into previously non-infected new faba bean- growing areas of Amhara and Oromia Regional States. The major impact of FBG disease occurs 2 during the main rainy season, but the disease was observed at low incidence level in North Shoa during the short rainy season as well (Beyene Bitew and Nigussie personal observation). A similar disease causing gall and later named '' broad bean blister'' disease caused by Olpidium viciae Kusano was reported as a major disease on faba bean in Sichuan and neighboring provinces of China (Yan, 2012). 1.4. Distribution of FBG Disease Worldwide, the distribution and importance of FBG-forming disease remains unknown across many faba bean-producing countries, except Japan and China. According to Yan (2012) report, FBG- forming disease, later named as broad bean ‘blister disease’, is caused by Olpidium viciae and the disease remains a significant yield-limiting problem in China since 1970's (Li-juan et al., 1993; Yan, 2012). In Ethiopia, FBG-forming disease might have existed a long time in the country, but it was not recognized until 2010. FBG forming disease was observed in 2010 in Menze Mama district, Amhara National Regional State, Ethiopia (Bitew and Kebede, 2012) and in Degem district of Oromia National Regional State, Ethiopia, in 2011 after farmers brought an infected sample to Holleta Agricultural Research Center (Dereje et al., 2012). The two districts, namely Menze Mama and Degem, have similar agroecology and both are located at an altitude above 2700 meters above sea level (m.a.s.l.) and both are characterized by cool air temperature and receive high rainfall. Currently, FBG-forming disease is a major yield limiting problem in Menze Mama, Degem and other faba bean-growing districts of Ethiopia (Hailu et al., 2014; Teklay et al., 2014; Bogale et al., 2016; Anteneh et al., 2018; Yitayih et al., 2021b). Faba bean gall-forming disease has been reported from Japan in Tokyo and China in Sichuan province on spring-sown faba bean. There, the disease is prevalent at high altitudes (2500-3400 m.a.s.l.) of Songpan, Xiaojin and Maekang areas of northwestern Sichuan province of China (Xing et al., 1984). The FBG-disease causing pathogen was found infecting buckwheat, cabbage, cucumber, field pea, rapeseed and spinach (Kusano, 1912; Xing, 1984; Li-juan et al., 1993). The disease has become increasingly a serious constraint in China since the 1970s, in more than 4,000 ha of faba bean fields and a yield losses of up to 20% had been reported (Li-juan et al., 1993). However, in Ethiopia, a total crop loss has been observed in North Shoa Zone of Amhara Regional Regional State and Tigray National Regional State in 2012-2014 (Teklay et al., 2014; Beyene, 2015). Wulita (2015) also reported 28.53 and 43.79% relative yield losses on a local cultivar at Ankober and Bassona Worana districts, 3 respectively. Generally, FBG disease can cause a total crop failure under severe epidemic conditions on susceptible cultivars (Teklay et al., 2014; Beyene, 2015). Regardless of such high yield losses, the yield loss on local and improved faba bean varieties remains unknown in Ethiopia. While FBG- forming disease is reported only in China, Ethiopia and Japan, the causative agent in Ethiopia has not yet been confirmed as being the same as in the other two countries. Moreover, the distribution of FBG disease and its causal pathogen remained unknown in Ethiopia until the present time. The faba bean gall disease affects faba bean starting from seedling to maturity, and mainly attacks the aboveground parts (leaves, petioles and stems) of the faba bean crop. During severe infections, the crop becomes dwarfed and collapses before flower setting, with the total biomass and grain yield profoundly or pronouncedly reduced (Beyene, 2015). 1.5. Symptomatology and Host Ranges Faba bean gall disease symptoms varied and ranged from small green gall or sunken spots on the leaves to complete browning leaves and stem. In China, FBG-forming disease symptoms are mainly observed and reported on the leaves, stalks and petioles, but not on pods and seeds (Yan, 2012). There, the early appearance of symptoms are as light green round spots, with the surfaces of spots becoming rough with disease development, then proliferating gradually to form intumescent swollen tissues and the presence of strumae. The strumae (scrofulous tumour or goiter-like swellings) can be solitary or colonial and able to fuse into irregular shapes. Infected plants show pygmyism (dwarfish) and leaf dysmorphosis (abnormality of shape). Presence of strumae and slightly rough surfaces are considered as characteristic symptoms for FBG disease and are an important basis for its diagnosis (Li juan et al., 1993; Yan, 2012). The pathogen shows invasion and colonization within the host epidermal cells, generating surface disease verrucous (blister-like) protrusions that are hallmarks (distinguishing characteristics) of the disease. According to these characteristic symptoms, the disease is known in Japan as ‘head tilt dish’, ‘fire swollen disease’, or as ‘edema beans fire disease’; and in China it is called ‘broad bean blister disease’ (Yan, 2012). Some previous reports indicated that FBG disease is caused by O. viciae and it is a strictly obligate parasite in the host epidermal cells, and it has not yet been cultured (Yan, 2012). In Ethiopia, similar gall-forming symptoms on leaves and stems of faba bean are reported by different researchers (Beyene, 2015; Bogale et al., 2016) (Initial symptoms involve green sunken or bulged (galls) on the upper or lower sides of the leaf (Beyene, 2015). The gall symptoms on the leaf gradually 4 become brown and extend to the stem. Galls then gradually change to complete brown lesions that cover the whole leaves. Progressively, galls change from circular to slightly irregular, lesions enlarge and the plants would have a ‘crumpled appearance’ (Hailu et al., 2014; Teklay et al., 2014; Bogale et al., 2016). Severely attacked leaves customarily show large necrotic areas surrounded by white lesions. In the latter stages, all leaves become blighted and most diseased plants are shortened and collapsed. Most of the time, symptoms are observed at the middle of the stem or on the actively growing host plant leaves (Beyene, 2015). Xing (1984) confirmed by artificial inoculation that the FBG-forming pathogen can infect and buckwheat, cabbage, cucumber, faba bean, field pea, rapeseed, and spinach, but not soybean, kidney bean (common bean) or other legume crops. In Ethiopia, according to surveys conducted at North Shoa and South Gondar Zones in Amhara Region, the disease was widely observed on faba bean and field pea under natural infections (Beyene, 2015; Yitayih et al., 2021b). Additionally, the disease was reported on vetch and lentil (Bitew et al., 2021; Yitayih et al., 2021a). The host range reported in other countries are not yet confirmed in Ethiopia and the host range identification is one important knowledge gap to be addressed in the country. Widely found hosts showing gall disease symptoms are displayed or depicted here under (Figures 1 and 2). Figure 1. Gall disease symptoms on faba bean at vegetative growth stage (A), leaf and (B) and stem (C). 5 Figure 2. Gall disease symptoms on field pea (A), Trifolium sp. 1 (B) and Trifolium sp. 2 (C). 1.6. Disease Epidemiology Introduction of FBG disease into Ethiopia is not specifically well known and survival of the pathogen, and disease cycle had not been well studied in the country so far. Some reports showed that the pathogen causing FBG disease survives on infected plant debris on soil surface or buried into the soil and is believed to survive active about 1-2 years (Li juan et al., 1993). The primary inoculum source is believed to be cysts from infected debris and soil with cysts germinating and discharging zoospores during faba bean planting. These zoospores penetrate the young seedlings of faba bean and form thin- walled zoosporangia in host cells to cause disease. Yan et al. (2012) also reported that the zoospores enter the epidermal cell of broad bean or faba bean by means of germ tube. The mature zoosporangia release zoospores to penetrate deeper layers of cells in already infected plants, or to infect other healthy plants. Disease symptoms on plants usually appear 13-18 days after infection. In the late stage of crop growth, sporogenic (cystigerous) plasmodia and cysts are formed to complete the disease cycle (Li juan et al., 1993; Yan, 2012). Zoospores are constantly produced after the disease occurs in the field and liberated, and zoospores are sources of secondary infection during presence of rain or dew. The pathogen spreads short distances by wind and rain and the frequency of re-infection is very high; however, other means of pathogen transmission and spreading mechanism needs investigations. Wherever highly repeated secondary infection occurs, disease spread is rapid in the field and peak at the flowering and pod formation growth stages (Yan, 2012). Subsequently, the resting stage sporangia start to occur at podding stage and the disease progress gradually ceases. Thus, the disease is polycyclic in nature. Continuous cropping and the leftover or infected plant residue can increase spread and severity (Yan, 2012). The biology of the pathogen is not well documented in Ethiopia and preliminary tests have shown that seedlings inoculated with infected 6 fresh leaves or zoospore suspension developed symptoms within three weeks after seedling emergence (Beyene, unpublished) and Yitayih et al. (2021a) reported symptoms developed from zoospore suspension inoculation. It is recognized that environment is the key factor for FBG disease initiation and development. Integrations of high altitudes (above 2400 m.a.s.l.), high relative humidity (80- 100%), low temperature (0-18oC), and high moisture in the soil and on the plant parts favor the spread, development, and production of the pathogen’s secondary inocula within a short period (Yan, 2012; Yitayih et al., 2021a). The disease linearly increases with increase in the altitude and rainfall status (Hailu et al., 2014; Yitayih et al., 2021b) 1.7. Faba Bean Gall Disease Management Disease management practices rely on anticipating occurrence of disease and attacking vulnerable points in the disease cycle. However, FBG is an emerging and devastating disease in Ethiopia and knowledge on its epidemiology and pathogen identity was not well addressed and known. As a result, there was no proven and effective FBG management practices recommended to growers. Various researchers employed fungicides and found to reduce FBG disease incidence and severity in different parts of the country (Wulita, 2015; Bitew and Tibabie, 2016; Hailemariam et al., 2017; Teferi et al., 2018). 1.7.1. Cultural practices Cultural practices, such as crop rotation, field sanitation, row intercropping, and mixed intercropping; reducing, crop density and use of pathogen-free seeds are key components of management for many diseases. Crop rotation, deep tillage, and planting high quality seed (disease-free) interrupt the disease cycle by reducing the amount of inoculum available for infection. This often delays the onset and/or severity of a disease. Crop rotation with non-host crops for three years with cereals, or intercropping, reduces the incidence of FBG disease in China (Li juan et al., 1993). Field sanitations (residue removal, weed management, burning stubble) are also recommended to minimize FBG spread and severity (Yan, 2012). In Ethiopia, faba bean is grown in rotation with wheat, barley and tef, and faba bean straw is a key animal feed in regions where FBG is a major disease. Moreover, farmers use manure as fertilizer in their faba bean fields and this may play a role in minimizing or providing pathogen inoculum and needs to be investigated further. In general; crop rotation, sanitation and surface water drainage may lower the pathogen inoculum load and minimize consequent severity of FBG, like other diseases. The other agronomic practices, like drainage, weed management and 7 planting time that farmers utilize need to be further investigated in line with their roles on FBG disease epidemiology. A research report by Getaneh et al. (2018) showed that FBG disease is not transferred by seed. Under Ethiopian conditions, practical recommendations had not been set so far towards integrating cultural practices to manage FBG disease. 1.7.2. Host plant resistance Faba bean breeding program in Ethiopia targets grain yield and major diseases, such as chocolate spot, rust and root rots. The majority of varieties released are developed with moderate levels of resistance to these diseases. In this regard, little effort has been made to find resistant faba bean genotypes from released, local faba bean landraces or from introduced materials (DBARC, 2016; Alehegn et al., 2018). Similarly, in North Shoa and South Gondar, a few released faba bean varieties were evaluated and showed different degrees of reaction to FBG disease. Improved faba bean varieties, such as Gachena (ETH91001-13-2), Gora (EK 01024-1-2), Degaga (R878-3) and NC 58, were reported to possess moderate resistance to FBG disease (Wulita, 2015; Getenet and Yizbalem, 2017). In China, different faba bean breeding lines tested for FBG showed different degrees of resistance, but none of them were resistant or immune (Yan, 2012). All findings imply an urget and exhaustive search from local and exotic genotypes for faba bean genotypes that are tolerant/ resistant to FBG disease through the breeding programs. 1.7.3. Chemical control Fungicides were the first recommended methods forwarded by several researchers and development agents to manage FBG disease in Ethiopia (Dereje et al., 2012; Bitew and Tigabie, 2016). Fungicides tested to manage FBG disease included seed treatment, foliar spray and combination of both. Use of triadimefon (Bayleton 25%) 250 g a.i. kg-1 as foliar and seed dressing was effective to manage FBG disease (DBARC Annual Progress Report, 2015; Bitew and Tigabie, 2016; Bekele et al., 2018; Teferi et al., 2018). Use of fungicide Matco (Metalaxyl 8% + Mancozeb 64% WP) at a rate of 2.5 kg ha-1 three times at 14 days-interval was found effective against FBG disease (Wulita, 2015). In China, Carbendazim (Bavistin) and Thiram (Arasan), at a dosage of 0.6-1.0 kg 100 kg-1 of seed, 25% Bayleton or 15% Baytan (Triadimenol) at a rate of 0.3% to seed weight, were effective seed dressings for managing gall-forming disease (Li juan et al., 1993). Application of foliar fungicides, such as Bayleton, Ridomil Gold MZ 68 WG (Metalaxyl 40+Mancozeb 640 g kg-1) and Chlorothalonil (Bravo, Daconil 2787) showed promising results to manage FBG disease (Bitew and Tigabie, 2016). However, 8 the rate, time and frequency of application need further investigations to reach at a common and conclusive recommendation in the country and help growers to sustainably manage this deadly desesae in faba bean-growing agro-ecologies and to increase grain yield. 1.7.4. Integrated FBG disease management Integrated disease management is the practice of using a range of measures that are compatible to implement and manage diseases in crop production. Combination of tolerant faba bean varieties, cultural practices, such as avoiding infected plant debries, and crop rotation, may contribute to minimize FBG disease severity and needs further investigation. Integrated managements contribute to reduce other disease severities too, With regard to this; Misgana (2017) reported that integrating sowing date and mixed cropping systems of faba bean with cereals lowered chocolate spot severity. Other studies also documented that utilizing combined applications of fungicides in integration with host resistance reduced grain yield losses through lowering intensities of different diseases in faba bean (Sahile et al., 2008; Bekele et al., 2018). Integrating faba bean varieties, fungicide and intercropping with different crops suppresses FBG pathogen inoculum buildup in the soil and serve as physical barriers for spore dispersal to prevent secondary infections and disease establishment (Yitayih unpublished). Farmers in South Gondar and North Wollo Zones grow faba bean mixed with triticale, faba bean with potato and faba bean with mustard to minimize faba bean gall disease intensity (personal observation). Integrating different disease management practices also used to minimize economic loss and environmental risks. In Ethiopia, even though FBG disease is expanding at an alarming rate to new faba bean-growing areas and causing a major threat to faba bean production, efforts to identify the causative pathogen and managing the gall disease has not been coordinated and is solely dependent on few fungicides applications. However, fungicide application is not economically affordable by smallholders and may not be environmentally-safe. Since FBG disease is an emerging and most devastating disease to faba bean, assessment of FBG incidence and severity, the association of the disease intensity with major biophysical factors, identification of the causative pathogen, faba bean yield loss assessment due to FBG, developing integrated disease management method through host resistance and fungicide applications are imperative to ensure sustainable faba bean production in the country and elsewhere having similar agro-ecologies. 9 1.8. Justification In Ethiopia, there are significant knowledge gaps in relation to FBG disease-causing pathogen identity and its variability. However, the epidemiology of the disease is not well established, and it is also desirable to ensure better management and to reduce the adverse impacts of FBG disease on faba bean genetic diversity and production. Released faba bean varieties so far and the landraces tested are not resistant to FBG disease (DBARC, 2015; Wulita, 2015). However, most of released faba bean varieties and elite breeding lines are not widely tested in different agro-ecologies in the country against FBG disease. Cultural practices are not well implemented and are not effective on minimizing the gall disease. How and when the disease was introduced to the country and the status of the disease is not yet well known at the national level. Despite the existing reality, studies on identification and biology of the gall-forming pathogen are essential. In addition, conducting proper disease management experiments aimed at reducing crop losses and defining the disease aggravating factors are also very pertinent. Generally, FBG disease is an emerging disease and expanding rapidly, but the causative agent is not yet confirmed and neither the biology of the pathogen nor the epidemiology of the disease had been studied in Ethiopia. There are no detailed studies on major associated factors that influence the epidemics of FBG. Bogale et al. (2016) reported some biophysical factors associated with FBG, but did not exhaustively identify and justify how or why different factors aggravate the disease. Some factors, such as planting time, land preparation (number of plowing), slope of the farm, drainage, nutrient (fertilizer or manure) applications, weather (rainfall, relative humidity and air temperature) and other independent variables have been less studied in FBG disease-affected areas of the country. There is no identified resistant faba bean variety and most of the farmers' cultivars and released faba bean varieties are susceptible to FBG disease (Teklay et al., 2014). Ethiopia is known as the second center of diversity for faba bean and landraces are excellent sources of gene(s) for improving grain yield and disease resistance in faba bean (Gemechu et al., 2005). Evaluation of the local and introduced genotypes for FBG disease resistance is important for identifying sources of resistance(s) as a critical component of developing integrated management practices against FBG disease. There are no disease management options recommended or developed specifically to manage FBG disease except use of a very few fungicides (Wulita, 2015; Bitew and Tigabie, 2016; Teferi et al., 2018). In Ethiopia, despite a number of improved faba bean varieties have been developed and released for general production under different recommendation domains, including the mid and high altitude 10 agro-ecologies and the waterlogged vertisol areas (Tamene et al., 2015), there had been no genotype, environment and genotype x environment interaction studies conducted to know the reaction stability of improved varieties and elite breeding lines against FBG disease. And, yield loss studies using local and/or improved varieties hav been very limited or none in the country. Some faba bean varieties released in the country for different agro-ecologies have shown inconsistent reaction against FBG disease. Wulita (2015), Getenet and Yizbalem (2017) evaluated some released faba bean varieties and identified a few different FBG disease tolerant varieties for North Shoa and South Gondar, respectively. In contrast, Alehegn et al. (2018) reported that most of the released faba bean varieties tested for East Gojjam was moderately resistant to FBG disease. Contrasting reaction reports on the same varieties implied that the reaction of these varieties were not well evaluated at contrasting multi-environments over years, and highlights that further studies are needed to identify the reaction and stability of resistant/susceptible reactions of faba bean genotypes, as multi-environment testing of genotypes can show disease resistance or susceptibility reaction stability (Xu, 2010). It is also important to categorize released varieties according to their reaction groups (Sharma et al., 2015). In addition, testing released faba bean varieties at multi-environment is important to determine the genetic potential of the varieties and variation in pathogen virulence (Sharma et al., 2012). Despite the importance of FBG disease, the existing body of knowledge was limited and there was no comprehensive and empirical data on the status of FBG disease across the country. Therefore, this study was undertaken to address critical research questions, including: 1) what is the distribution and intensity of FBG disease and major and alternate hosts of FBG disease in major faba bean-growing areas of central, northern and northwestern Ethiopia? 2) What are the major biophysical factors that influence disease epidemics? 3) What is the genetic identity of the pathogen? 4) What is the yield gap between local and improved varieties? 5) What is the reaction of introduced and local genotypes to FBG disease under field conditions? 6) What is the reaction of released faba bean varieties to gall disease and what is their resistance reaction stability at different locations? 7) What are the best potential integrated FBG management options? All these abovementioned research questions were planned to be addressed in this study with the following objectives. 11 1.9. Objectives General objective: The study was launched to have deep insight into the importance, distribution, epidemiology, genetic identity and development of sustainable management strategy against faba bean gall disease and increase in faba bean grain yield in Ethiopia and elsewhere that have similar agro-ecologies. Specific objectives: The study was carried out with the specific objectives to: 1. Assess the distribution, disease incidence, severity and hosts of FBG disease in major faba bean growing areas and determine the association of FBG disease parameters with major biophysical factors in Ethiopia (Paper I). 2. Identify the genetic identity of FBG disease-causing pathogen (Paper II). 3. Determine yield losses on local and improved faba bean varieties due to FBG disease under different management options (Paper III). 4. Evaluate faba bean genotypes for resistance reaction to FBG disease (Paper IV). 5. Identify the phenotypic reactions of genotypes and determine their stability of resistance reactions to FBG disease across multiple locations (Paper V). 6. Develop integrated management options against FBG disease (Paper VI). 12 2. MATERIALS AND METHODS This PhD dissertation research had three parts: FBG disease survey, molecular identifications of FBG disease-causing pathogen and field-based experiments. Faba bean gall survey and sampling were conducted in Amhara, Oromia and Tigray National Regional States. Molecular identifications of FBG disease-causing pathogen were done at Debre Birhan Agricultural Research Center (DBARC), International Livestock Research Institute (ILRI), Ethiopia, and the University of Western Australia, Australia. Field experiments (yield loss assessment, genotype evaluation for resistance reaction, multi- location genotype evaluation and developing integrated disease management) were conducted on farmers' fields’ in North Shoa (Bassona Worana), East Gojjam (Debay Telatgen) and South Gondar (Farta), which were hot spot areas for FBG development. Local genotypes and introduced elite breeding lines as well as released faba bean varieties were used for field experiments. The details of the materials and methods are briefly described under each article and manuscript. 2.1. Description of the Study Areas Faba bean gall surveys were conducted in Amhara, Oromia, and Tigray National Regional States during June to September 2018 and 2019 main rainy seasons (Paper I). Details of the geographical locations of 14 districts covered in the survey were presented (Paper I). Similarly, surveys were also conducted in the North Shoa Zone (Ankober, Asagert, Bassona Worana, and Tarmaber districts) of Amhara Regional State in the short rainy season (January-May) in both survey years. All the surveyed districts were purposively selected based on accessibility of the areas and potential of faba bean production. The districts were characterized by light Cambisol/sandy loam and only a few of the fields were dominated by black Vertisols. In the main cropping seasons, faba bean planting was done early June to early July; however, in the short rainy seasons planting was done at the onset of rainfall starting mid-January to early February. Majorities (67%) of the farmers were not using fertilizer for faba bean cultivation, while some farmers applied NPS (25%) fertilizer, and very few (8%) applied manure at the time of planting and faba bean was cultivated with cereal rotations, mainly barley and wheat. The weather data for 2018 and 2019 were obtained from each district through Ethiopian National Meteorology Agency (Paper I). The relative humidity (%) during the main rainy season ranged from 60 to 90% in 2018 and 52 to 94% in 2019 main cropping season. In the short rainy seasons, the relative humidity ranged from 58.5 to 83.6% in 2018 and 49.6 to 82.4% in 2019. The molecular studies for identifications of the FBG disease-causal pathogen or agent were conducted at Debre Birhan Agricultural Research Center (DBARC) and International Livestock Research Institute 13 (ILRI), in Ethiopia, and University of Western Australia, Australia (Paper II). Yield loss assessment, evaluation of faba bean genotypes, reaction of faba bean genotypes to gall disease and resistance reaction stability under natural infection, and integrated FBG disease management field experiments were conducted at Bassona Worana, Debay Telatgen and Farta districts during 2018 and 2019 main cropping seasons (Papers III-VI). Bassona Worana district is located at 9o41’N latitude and 39o31’E longitudes with an altitude of 2980 m.a.s.l. The district receives bimodal rainfall. The total annual rainfall of the district reaches 1,000 mm and mean minimum and maximum temperatures are 6 and 19 °C, respectively (DBARC, 2005). However, the annual rainfall at Bassona Worana district was 1927 and 1452.9 mm during 2018 and 2019 main cropping seasons, respectively. (Meteorological data for Bassona Worana were obtained from Debre Birhan Agricultural Research Site). Debay Telatgen district is located at 10°45′ N latitude and 37°50′ E longitude with an altitude of 2400 m.a.s.l., and annual rainfall of 900 mm. Annual mean maximum and minimum temperature is 9.9 and 23.6 °C and 10.6 and 22.9 °C for the district in 2018 and 2019 cropping seasons, respectively. Farta district is located at 12°00′N latitude and 30°00′E longitude at an altitude of 2650 m.a.s.l. The relative humidity ranged from 65 to 96% in 2018 and 74 to 82% in 2019 cropping season. The temprature ranged from 9.23 to 22.58 °C. Rainfall data for 2018 and 2019 for Debay Telatgen and Farta were obtained from the Ethiopian National Meteorology Agency, Addis Ababa, Ethiopia. 2.2. Faba Bean Gall Disease Survey Faba bean gall disease surveys were conducted to assess the distribution, disease incidence, severity and alternate hosts of FBG disease in major faba bean-growing areas of Ethiopia and determine the association of FBG incidence and severity with major biophysical factors, such as altitude, drainage, weed and crop density, use of fertilizer type, growth stage, soil type and planting time (Paper I). A total of 783 fields were covered both in the main and the short rainy seasons of both 2018 and 2019 cropping years, of which 373 faba bean fields were inspected in 2018, while 410 faba bean fields were assessed in 2019 cropping season. In the main rainy season, 678 faba bean fields were surveyed, whereas 105 fields were surveyed in the short rainy seasons. Faba bean fields were selected randomly along the main roads at about 5-10 km intervals. In each field, disease prevalence, disease incidence and severity were recorded at five spots by moving in an ‘X’ pattern in the fields using 1m2 quadrat. 14 Disease prevalence (%): The FBG prevalence, i.e. the geographical distribution of the disease in the district, was determined from the total number of fields inspected per district and expressed in percentage as follows: Number of fields infected per district Disease prevalence (%) = Total number of fields assessed per district X 100 Total plants, infected and healthy plants, were counted per quadrat to estimate the FBG disease incidence. Similarly, 12 randomly selected plants per quadrat were used for severity assessment. Five records per field were averaged and mean incidence and severity were used for each field. Disease incidence (%): Disease incidence was calculated using the following formula: Number of plants showing symptoms per quadrat Disease incidence (%) = Total number of plants assessed per quadrat X 100 Disease severity (%): It was rated using a modified 0-9 disease scoring scale (Ding et al., 1993), where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. Percent severity index (PSI): Severity scores were converted into percent severity index (PSI) (Wheeler, 1969) for analysis using the following formula: Sum of numerical ratings PSI = Total number of plants scored x maximum score on scale X 100 Moreover, data on alternate host, altitude, slope, drainage, soil type, weed type, weed density, method of planting, cropping system, growth stage, variety, plant population, planting time, fertilizer type and previous crops were collected per inspected field. At the end of the season, weather data [mean maximum and minimum temperature (°C) per month, rainfall (mm) and relative humidity (%)] were collected from the nearby meteorological stations through the National Meteorology Agency for all survey districts. 15 2.3. Molecular Identification of FBG Disease-Causing Pathogen 2.3.1. Disease symptoms and pathogen morphology For FBG disease-causing pathogen identifications, faba bean leaves and stems infected by faba bean gall pathogen were collected from Bassona Worana, Debark, Debay Telatgen, Farta, and Dessie zuria districts of Amhara, and Degam and Chole districts of Oromia National Regional States. The observations were made on hand-sectioned wet mounts of infected and healthy faba bean leaves and stems using an Olympus CX33 microscope, with images captured using an Olympus EP50 digital photographic system. The molecular identifications were conducted at University of Western Australia, Australia 2.3.2. Collection of FBG disease samples for DNA analyses A total of 187 infected plants showing FBG symptoms were collected, and crude extracts of DNA from pathogen-infected materials in Ethiopia were applied onto Whatman FTA® Cards. Field sampling at each location was undertaken from two selected plants showing faba bean gall symptoms. When returned to the laboratory, for each single plant sample brought back from the field, an area of galled fresh tissue, mostly on leaves and stem, was selected and excised. Diseased tissue surfaces were first wiped with 70% ethanol for initial sterilization. The fresh galled pieces were placed into a 1.5 mL Eppendorf tube until the tube was approximately half-filled, with a separate Eppendorf tube for each sample. Gall samples were well macerated in the tubes using a plastic pestle. The upper and lower epidermal layers of infected leaf and stem tissues were separately and carefully peeled-off by hand and treated similarly. Then macerated gall extract samples were placed onto separate FTA® Cards and air- dried. (FTA® card or paper is a commercial product consisting of filter paper impregnated with a proprietary mix of chemicals, which serve to lyse cells, to prevent growth of bacteria, and to protect the DNA in the sample). The gall samples were first processed in Ethiopia to extract DNA as described in Section 2.3.3 below, and extracted DNA was applied to FTA® Cards. Impregnated FTA® Cards were exported out of Ethiopia under an Ethiopian Biodiversity Institute Material Export Permit (Reference No. EBI 71/2553/2011). Impregnated FTA® Cards were taken into Australia under an Australian Government, Department of Agriculture and Water Resources Import Permit, specifically for this purpose (Permit No. 0002826465). 16 2.3.3. Deoxyribonucleic acid (DNA) extraction and PCR conditions Deoxyribonucleic acid (DNA) was extracted from FTA® Card’s carrying pathogen DNA by cutting 2- 3 pieces each 2 mm square into a 2 mL Eppendorf tube and washing to remove inhibitor using a modified method of Ahmed et al. (2011). Briefly FTA® Card pieces in 2 mL tubes were washed twice in 200 μL of TENT (10 mM Tris-Cl, pH 8.0, 1 mM EDTA, 12 mM NaCl, 2.5% Triton X-100) and incubated for 5 minute each time with gentle agitation at 1000/min on an agitator (TTS 3 digital Yellow line). The FTA® Cards were then washed twice in 200 μL in TE 0.1 (10 mM Tris-Cl, pH 8.0, 0.1 mM EDTA) buffer and agitated for 5 min each time at 1000/min and supernatant removed each time. Ehylenediaminetetra acetic acid) (EDTA) is a white, water-soluble solid of aminopolycarboxylic acid, which is widely used to bind to iron and calcium ions and serves as a hexadentate chelating agent. The FTA® Card pieces then were left to dry at room temperature (under laminar flow for 2 hrs). Dried FTA® Card pieces were then ready for DNA extraction undertaken as follows. Two ceramic beads (2 mm) and 300 µL extraction buffer [200 mM Tris-HCl (pH 8.5), 250 mM NaCl, 25 mM EDTA], 0.5% (wt/vol) sodium dodecyl sulphate (SDS)] were added together into the tube and homogenized using a Precellys Evolution homogenizer at 10,000 rpm for 3x60 second cycles, each cycle with a 30 second pauses. Then, 150 μL of 3 M sodium acetate (pH 5.2) was added and mixed well by pipetting, left at room temperature for 10 min, then centrifuged at 13,000 rpm for 10 min. Subsequently, the supernatant was transferred to a fresh tube with an equal volume of prior-added cold isopropanol (450 µL), mixed well by pipetting and then left to stand for 15 minutes at room temperature (22 °C) or in a fridge overnight for best result. Precipitated DNA was collected by centrifugation at 13,000 rpm for 10 minutes and supernatant pipetted out. Pellet was washed with 70% (vol/vol, ethanol/water) and dried (air dry or vacuum dry), then re-suspended in 50 μL of TE (Thermo Fisher) buffer. The quantity and quality of extracted DNA were determined with a Nano Drop 1000 Spectrophotometer (Thermo Scientific). The DNA was stored in a refrigerator at 4 °C. The DNA was subjected to polymerase chain reaction (PCR) using a master mix of a total volume of 25 µL that contained 0.5 µM of each primer (primers used and PCR conditions are listed in Paper II). Polymerase chain reaction (PCR) products were subjected to agarose gel electrophoresis at 60 mV for ≥1 hr (dependent on size of the PCR products) on a 1% (w/v) agarose gel containing 0.1% GelRed™ Biotium Inc. (United States) and then visualized under UV light. PCR products were then sequenced by outsourcing to Macrogen Inc. (Korea). 17 2.3.4. Primer pairs tested The primer pairs indicated in Paper II were used to amplify small subunit (SSU) (James et al., 2006); internal transcribed spacer (ITS) region (ITS1 to ITS4) (White et al., 1990); and nrLSU (Vilgalys and Hester, 1990; Rehner and Samuels, 1994; Stielow et al., 2015). Primers EF and RPB2 were also tried, but without success. Initially assuming the pathogen was Olpidium viciae, general Olpidium specific reverse primers (Herrera-Vásquez et al., 2009) for amplifying ITS region were also used in combination with ITS1 (and with the specific primer pair OLPv4F and OLPv4R developed in the current study) for O. viciae according to the sequence of O. viciae available in GenBank (HQ677595) as had earlier been submitted by Yan (2012) (Paper II). 2.3.5. Phylogenic tree building The sequenced sample DNA were aligned using Geneious Prime version 2020.03 and then resulting consensus were BLAST in Genbank (NCBI). If percentage identity was low (≤ 97%) or in some cases higher, but with low coverage (≤ 85%), the most frequent, closest and likely-matching pathogens from Genbank were selected as references for building phylogenic trees, again using Geneious Prime version 2020.03. Sequences of morphologically similar and taxonomically-related pathogens Olpidium, Physoderma, Synchytrium and Urophlyctis were also selected from Genbank and included in building phylogenic trees for comparison. Three phylogenetic trees were built according to amplification regions. Sequences from ITS1-5.8S-ITS2 formed one tree, sequences from partial 18S- ITS1-5.8S-ITS2-partial 28S formed another tree and sequences from LSU (28S) formed a third tree. Pairwise sequence alignment [Alignment Type: global alignment with free end gaps; cost matrix: 65% similarity (5.0/-4.0)] was used with Tamura-Nei for Genetic Distance Model and Neighbor-Joining for Tree Building. 2.4. Estimating Yield Loss of Faba Bean Caused by Gall Disease in North Shoa, Ethiopia Field experiments were carried out in Bassona Worana district on farmers’ fields under natural infections in 2018 and 2019 main cropping seasons (Paper III). The experimental treatments included three faba bean varieties, two fungicides, two fungicide spray schedules and three unsprayed (control) plots with a total of 15 treatments. Two moderately resistant varieties (Gora and Degaga) and one susceptible farmers’ local cultivar and, two foliar fungicides [Bayleton 25 WP (Triadimefon 250 g kg– 1 a.i.) and Ridomil Gold MZ 68 WG (Metalaxyl 40+Mancozeb 640 g kg–1)] and two fungicide spray schedules at 10 and 15 days interval were used. The treatments were arranged in a randomized 18 complete block design with three replications in factorial combinations. A plot size of 4.8 m2 (2.4 m × 2 m) was used, and each plot was arranged to have 6 rows. Spacing between blocks and plots was 1.5 m, while spacing between rows and plants were 0.4 m and 0.1 m, respectively. Blended NPS (38 kg P2O5, 19 kg N and 7kg S) fertilizer was applied once at the time of planting at a rate of 121 kg ha–1. Bayleton and Ridomil Gold fungicides were applied according to the manufacturers' recommendation at the rates of 0.7 kg ha–1 and 2.5 kg ha–1, respectively. Five meter by three meter plastic sheet were used in the wind side direction by moving to each plot while spraying to protect draft effect of fungicides for each treatment. The first spray started at the onset of typical disease symptom both in 2018 and in 2019 cropping season and a total of 5 and 4 sprays were done for 10 and 15 days spray respectively. Seeding and fertilizer application were done manually and weeding and other cultural practices were uniformly done for each plot. Date of disease onset, disease incidence and disease severity were assessed during the epidemic periods of the disease. Faba bean gall disease incidence and severity were recorded at 10 days intervals starting from the first date of disease onset at 32 day after planting (DAP) in 2018 and 28 DAP in 2019. Disease incidence and severity were calculated using the formulae indicated on section 2.2. Area under disease progress curve (AUDPC) was calculated from FBG severity data. Percentage disease severity data recorded for each plot during the course of the experiment were summarized and used to calculate AUDPC by trapezoidal integration (Madden and Hughes, 1995) and it is expressed in % -days and calculated using the formula: 푥 + 푥 AUDPC = ( 2 )(푡 − 푡 ) Where, Xi = the disease severity at the ith measurement, ti = is the time of the ith measurement in days from the first measurement date and n = total number of disease measurements. Relative yield loss (%) due to FBG disease for all treatments were estimated based on loss compared with maximum protected plots for all tested faba bean varieties and fungicide treatments using the following formula: YSP − YNP Relative yield loss, RYL (%) = YSP X 100 Where, YSP = yield obtained from maximum protected plots and YNP = yield obtained from unsprayed and/or less protected plots. The losses in 100-seed weight were also similarly determined. Yield increase over unsprayed plots (control) was calculated from the difference between sprayed and 19 unsprayed plots in terms of grain yield and expressed in percentage (Paper III). Total costs of seed (improved and local), fertilizer, fungicides, fungicide applications, spraying equipment, weeding, harvesting, threshing and labor cost (man per days) were recorded and calculated on a per hectare basis. 2.5. Evaluation of Faba Bean Genotypes for FBG Resistance Reaction Field experiments were carried out at Bassona Worana district in the 2018 and 2019 main cropping seasons (Paper IV). Two hundred local faba bean genotypes, obtained from the Ethiopian Biodiversity Institute (EBI), and 215 breeding lines introduced from Lebanon through ICARDA were evaluated for FBG resistance under natural infections. The genotypes were planted in ten blocks using augmented experimental design with check faba bean varieties. Moderately tolerant varieties showing varying degrees of reactions; Gora (EK 01024-1-2), Gachena (ETH91001-13-2), Degaga (R878-3) and a susceptible farmers' local cultivar were used as a check. The check varieties were randomly planted in each experimental block. Each test genotype, breeding lines and checks were planted on a plot size of 1.6 m2 (0.8 m × 2 m, w*L) and each test material was planted in 2 rows with spacing of 0.4 m between rows and 2 m plot length as well as 0.1 m between plants. Farmers' local cultivar was uniformly planted as a spreader row in all directions of the experiment to increase disease pressure. Nitrogen-Phosphorus-Sulphur (NPS) (composition: 38 kg P2O5 , 19 kg N and 7 kg S ) blended fertilizer was applied at the time of planting at a rate of 121 kg ha-1. Seed sowing and fertilizer application were manually performed. Plots were hand weeded, and all other cultural practices were done uniformly across all plots. Genotypes and breeding lines showing low FBG disease severity were selected based on their reaction by comparing with each check and genotype and breeding lines with each other. Promising and selected genotypes and breeding lines were advanced and were planted in the 2019 main cropping season in the same hot spot area with the same design and further evaluated for FBG resistance and other agronomic traits. Faba bean gall disease assessments were conducted in both years and seasons. First date of disease onset, disease incidence and severity were assessed during the experimental periods. Faba bean gall disease incidence and severity were recorded at 10 days intervals starting from the first date of disease onset. For disease incidence 20 plants, and for severity measurements (l2 plants) were randomly picked from each plot and tagged prior to the appearance of typical disease symptoms. Faba bean gall incidence and severity were calculated by the formula indicated in section 20 2.2 and AUDPC values were calculated by the formula indicated in section 2.4. In addition, faba bean growth and yield parameter data were collected in both years and seasons. Growth and yield parameters, such as days to 50% seedling emergence, days to 50% flowering, days to 90% physiological maturity, plant height (cm) at maturity, number of pods per plant, number of seeds per pod, 100 seed weight (g) and grain yield (gram per plot) were recorded. Grain yield was converted into kilogram per hectare (kg ha-1) at 10% adjusted grain moisture content (Birru, 1979). 2.6. Reaction of Faba Bean Genotypes to Gall Disease and Resistance Reaction Stability under Natural Infections Field experiments were conducted on farmers’ fields at six different locations to identify the phenotypic reactions of faba bean genotypes to FBG disease and to determine the resistance reactions stability against FBG disease under natural infections (Paper V). Thirteen released faba bean varieties, which had different levels of FBG responses, seven elite breeding lines collected from the research center, and one farmers' local cultivar for each respective district, were used as treatments in the experiment. Each genotype was planted as an experimental unit with a total of 21 treatments per replication and the treatments arranged in a randomized complete block design (RCBD) with three replications. The plot size was 4.8 m2 (2.4 m × 2 m), and each plot had 6 rows. Spacing between blocks and plots were 1.5 m, while spacing between rows and plants were 0.4 m and 0.1 m, respectively. NPS (compositions: 38 kg P2O5, 19 kg N and 7 kg S) blended fertilizer was applied once at the time of planting with the rate of 121 kg ha-1. Seeding and fertilizer application was done manually at each row. Plots were hand weeded and other cultural practices were kept uniform across all plots at each location and season. Faba bean gall assessments were conducted in all locations and seasons. First date of disease onset, disease incidence and severity were recorded during experimental periods. Faba bean gall incidence and severity were recorded at 10 days intervals starting from the first date of disease onset. Percent FBG incidence and severity data were recorded and calculated by the formula indicated under section 2.2. In addition to FBG parameters, grain yield (gram per plot) was measured, and determined from the harvestable four central rows for each treatment. The grain yield per plot was converted into tons per hectare (t ha-1) at 10% adjusted grain moisture content (Birru, 1979). Moreover, composite soil samples were taken at 20 cm depth and soil type (texture) and soil pH were analyzed for each testing location (Paper V). 21 2.7. Integrated Management of Gall Disease on Faba Bean in North Shoa, Ethiopia The experiments were conducted on farmers’ fields in 2018 and 2019 at Bassona Worana district to manage FBG disease pressure (Paper VI). The treatment combinations contained three faba bean varieties, two fungicides; two spray schedules and three controls (unsprayed plots) with a total of 15 treatments in the experiment (Paper VI). Two moderately resistant varieties (Gora and Degaga) (Wulita, 2015; Getenet and Yezbalem, 2017) and one farmers’ local cultivar susceptible to FBG disease were used in the experiment. Two systemic foliar fungicides (Bayleton 25 WP and Ridomil Gold MZ 68 WG) and two fungicide spray schedules at 10 and 15 days interval were components of the treatments. The treatments were arranged in a randomized complete block design in factorial combinations with three replications. The plot size was 4.8 m2, and each plot was plant with 6 rows of faba bean. Spacing between blocks and plots was 1.5 m, while spacing between rows and plants were 0.4 m and 0.1 m, respectively. Blended NPS (composed of 38 kg P2O5, 19 kg N and 7 kg S) fertilizer was applied once at the time of planting at a rate of 121 kg ha–1. Bayleton and Ridomil Gold fungicides were applied according to the manufacturers' recommendation at the rates of 0.7 kg ha–1 and 2.5 kg ha–1, respectively. Five meter by three meter plastic sheet were used in the wind side direction by moving to each plot while spraying to protect draft effect of fungicides for each treatment. The amount of water used to spray fungicides was also based on the manufacturers’ recommendation for each fungicide. A total of 300 and 500 L ha-1 of water was used to spray Bayleton 25 WP and Ridomil Gold MZ 68 WG, respectively. The first spray was started at the onset of a typical disease symptom at 32 days after planting (DAP) in 2018 and 28 DAP in 2019. Seeding and fertilizer application were done manually, and weeding and other cultural practices were uniformly done for each plot. Date of disease onset, disease incidence and severity were repeatedly recorded during the experimental periods. Faba bean gall incidence and severity were recorded at 10 days intervals starting from the date of the disease onset. Disease incidence and severity were calculated by the formulae indicated on section 2.2. The AUDPC was calculated by the formula indicated in section 2.4. In addition, growth and yield parameters of faba bean were also collected and calculated for each plot. Partial budget analysis was performed using the market prices for inputs at planting and outputs (yield) when the crop was harvested to determine profitability of the treatments imposed based on 22 procedures of CIMMYT (1988). Marginal rate of return was calculated for a marginal increase in an investment, which was approximately the additional output resulting from a one-unit increase in the use of a variable input, while other inputs were constant. To measure the increase in net return associated with each additional unit of cost (marginal cost), the marginal rate of return (MRR) was estimated using the following formula: NB2 − NB1 MRR = TCI2 − TCI1 Where, MRR = marginal rate of return, NB1= net benefit obtained without using fungicide, NB2 = net benefit obtained after using fungicide, TCI1 = total costs of input for control treatment, and TCI2 = the total costs of input for fungicide sprayed treatment. The cost of fungicides, application costs and costs of improved varieties and local cultivar seeds were estimated based on the market price during the 2018 and 2019 cropping years. 2.8. Data Analysis In the FBG diseae surveys, the association of the mean FBG disease incidence and severity with biophysical factors was tested using the logistic regression model (Yuen, 2006) with the SAS procedure of GENMOD (SAS 2014). Mean disease incidence and severity were classified into distinct groups of binomial qualitative data (Fininsa and Yuen, 2001). Class boundaries of ≤ 50% and > 50% for incidence and severity were chosen so that groups contained approximately equal totals, thus, yielding a binary dependent variable (Paper I). The single model tested the association of each independent variable alone with the disease incidence and severity. In multiple models, the association of each independent variable with disease incidence and severity was tested when entered last into the model with all independent variables. Only the variables that showed a high association to disease incidence and severity in the single and the multiple models were tested in the reduced multiple models (Yuen, 2006). Deviance reduction and odds ratios were calculated for each independent variable as it was added to the reduced model. The odds ratios obtained by exponentiation of the parameter estimates were interpreted as the relative risk when the values are greater than one. The difference between the likelihood ratio (LRTs) statistics was used to examine the importance of the variables contributions for the case and tested against Chi-square values (McCullagh and Nelder, 1989). Mean FBG incidence 23 and severity in the short-rainy season were very low and only main-rainy season disease parameters were analyzed in the multiple regression model. In the causative agent identification sequencing and phylogenetic tree building were done. The sequenced sample DNA was aligned using Geneious Prime version 2020.03 and then resulting consensus was BLAST in Genbank (NCBI). If percentage identity was low (≤ 97%) or in some cases higher, but with low coverage (≤ 85%), the most frequent, closest and likely-matching pathogens from Genbank were selected as references for building phylogenic trees, again using Geneious Prime version 2020.03. Sequences of morphologically similar and taxonomically-related pathogens Olpidium, Physoderma, Synchytrium and Urophlyctis were also selected from Genbank and included in building phylogenic trees for comparison. For yield loss assessment, genotype evaluation and integrated FBG management, FBG severity, AUDPC, growth data, grain yield and yield-components for each treatment were subjected to analysis of variance (ANOVA) using the PROC GLM procedure (SAS, 2014) to determine the response of faba bean genotypes, effect of fungicides and spray schedules on gall disease epidemic development and grain yield accumulation. Mean separations among treatments were done using Duncan’s Multiple Range Test (DMRT) at 5% probability level (Gomez and Gomez, 1984). Mean comparison for genotype evaluation was done for checks using Least Significant Difference (LSD) at 5% probability level. The correlation of disease with growth and yield parameters was analyzed based on Pearson’s correlation coefficient to examine the relationship between FBG disease parameters, yield and yield- components of faba bean. Linear regression analysis was done by plotting grain yield data against FBG severity, and regression intercepts, slopes and coefficients of determination (R2) were computed using the Minitab 19.1 statistical package. The two seasons were considered as different environments because of heterogeneity of variances as tested using Bartlett’s test (Gomez and Gomez, 1984). Thus, separate analyses were done for each cropping season. For the reaction of faba bean genotypes to gall disease and resistance reaction stability under natural infections, disease incidence, severity and grain yield data were subjected to ANOVA using the PROC GLM procedure of SAS (SAS, 2014). Combined ANOVA was computed to determine the effect of genotypes (G), environment (E), and GxE interaction on disease reaction and grain yield. Average GxE means, and genotype ranks across environment (Falkenhagen, 1996) was computed using additive main effects and multiplicative interaction (AMMI) model. The following modified AMMI 24 model was used for the tested genotype and test environments for disease severity and grain yield (Gauch, 1992): Dij = 휇 + 푔푖 + 푒푗 + 푛 = λ푛훼푖푛푦푗푛 + 휃푖푗 Where, i = number of genotypes (1-21); j = number of locations (1-6): D푖푗 = disease severity/yield mean of ith genotype in j environment, 휇 = grand mean; 푔푖 = main effects of genotypes; 푒푗 = main effects of environments; 휆푛 = Eigen values for PCA axis n; 훼푖푛 and 훾푗푛 = the ith genotype jth environment PCA scores for the PCA axis n; 휃푖푗 = the residual effect; and n' = the number of PCA axes retained in the model. GGE biplots were generated using the first two symmetrically scaled principal components (PC) for an average tester coordinate and polygon view biplots. Genotype (G) and genotype x environment (GxE) interaction analysis were done using the GGE biplot analysis model to determine genotype performance and stability in varied environmental conditions (Yan and Kang, 2003). The model incorporates both additive and multiplicative component of the two way data structure for analysis. Genotypes reaction against FBG disease was determined based on FBG disease severity value. 25 3. RESULTS 3.1. Faba Bean Gall Disease Survey 3.1.1. Symptomatology and alternate hosts Faba bean gall symptoms were observed on different parts of faba bean. Circular or irregular green sunken or galls on the upper or lower sides of the leaf were observed on young and new leaves. The green gall gradually changed to brown, became large, and covered the majority of leaves and stems. In the latter stages of the disease development, leaves became blighted and collapsed. Severely infected plants showed short and crumpled forms before flowering and podding growth stages. Most symptoms were observed on leaves and stems, but in the current survey the symptoms were also observed on pods, flowers, and root areas (Paper I). Moreover, FBG symptoms were observed on different plants and recorded as presumably alternate host. The possible alternate hosts that showed typical FBG symptoms were field pea (Pisum sativum), lentil (Lens culinaris), smartweed (Polygonum nepalense) and Trifolium spp. The symptoms on field pea were observed on leaves and stems, but the symptom on lentil, Trifolium, and smartweed were observed on leaves only. The symptoms on Trifolium spp. and smartweed were recorded under faba bean fields at Sinan and Bassona Worana districts in East Gojjam and North Shoa Zones, respectively. Faba bean gall symptoms on lentil were observed at Bassona Worana district in North Shoa Zone. Faba bean gall symptoms on field pea were observed in all surveyed districts both in main and the short rainy seasons of the two study years. 3.1.2. Distribution, prevalence, incidence and severity of FBG Main cropping season: The survey results showed that FBG was prevalent in all faba bean surveyed fields and districts. However, FBG disease prevalence, incidence, and severity varied across districts and years (Paper I). In the 2018 main rainy season, higher (100%) FBG disease prevalence was recorded in Ankober, Asagert, Sinan, and Tarmaber districts than in other districts. The lowest (66.7%) prevalence was recorded at Chole district in Arsi Zone. Similarly, the highest (100%) FBG disease prevalence was recorded at Ankober district, while the lowest (44.5%) was at Chole district in the 2019 main rainy season. In the 2019 main cropping season, faba bean gall disease incidence was 100% in Ankober, Sinan, and Tarmaber districts. Conversely, the lowest (66.7%) mean FBG disease incidence was recorded at Asagert and Chole districts in 2018 main rainy season. Similarly, the highest (100%) and the lowest (44.2%) mean FBG disease incidence was recorded in Ankober and Chole district, respectively, in the 2019 main rainy season (Paper I). In 2018, at district level, the mean FBG incidence decreased by 33% in Asagert and Chole districts as compared to Ankober, Sinan, 26 and Tarmaber districts. Similarly, in 2019 cropping season, the mean FBG disease incidence decreased by 44.2% in Chole district as compared to Ankober district. Mean FBG disease incidence decreased by 15.5% in 2019 main cropping season as compared to the 2018 main rainy season. Generally, a similar pattern of FBG disease severity was observed in both 2018 and 2019 cropping years. In the 2018 main rainy season, high mean FBG severity was recorded in Sinan (63.9%), Bassona Worana (61.8%), Ankober (59.6%), and Tarmaber (58.7%) districts, while the lowest (20.5%) severity was recorded at Asagert district. In 2019, high mean FBG disease severity was recorded at Ankober (45.0%), Bassona Worana (43.5%), and Tarmaber (42.4%) districts, and the lowest (11.2%) FBG disease severity was recorded at Enda Mehoni district. In the 2018 main rainy season, the mean FBG severity at Asagert district decreased by 67.8% as compared to Sinan district. Similarly, in 2019, the mean FBG severity in Enda Mehoni district decreased by 75.1% as compared to Ankober district, while FBG severity in 2019 decreased by 32.4% as compared to 2018 main cropping season. Short rainy season: The highest (80%) FBG disease prevalence was recorded at Ankober district, while there was no FBG disease recorded at Bassona Worana district in the 2018 short rainy season. In 2019, the highest (52%) and the lowest (14%) mean FBG disease prevalence was recorded in Ankober and Asagert district, in that order. Faba bean fields at Ankober district obtained the highest mean FBG disease incidence (74.3% and 56.7%) and severity (16.5% and 12.1%) in 2018 and 2019, respectively. The lowest (2.7%) mean FBG severity was recorded at Bassona Worana district in 2019 short rainy season (Paper I). 3.1.3. Associations of FBG epidemics with biophysical factors Among the independent variables, district, year, altitude, drainage, soil type, weed and crop density, growth stage, fertilizer type, and planting time were the most important variables in their association with the mean FBG incidence when entered first and last into the model (Paper I). High FBG incidence (>50%) was strongly associated with Ankober, Bassona Worana, and Sinan districts, the 2018 cropping year, an altitude of > 2700 m.a.s.l., poor drainage, flowering growth stage, early planting, and manure application. Early planting showed 3 times greater probability of association with mean FBG incidence of >50% than recommended planting time. On the contrary, low mean FBG incidence (≤50%) showed high probability of association with Chole, Meket, and Enda Mehoni districts, seedling growth stage, and an altitude of < 2700 m.a.s.l. Ankober, Sinan and Bassona Worana districts showed 4.1, 2.1, and 3 times greater probability of association with high mean FBG 27 incidence than Tarmaber district, respectively (Paper I). Similarly, district, year, planting time, growth stage, altitude, drainage, weed density, and fertilizer type were the most important variables in their association with the mean FBG severity when entered first and last into the model. However, soil type and crop density lost their significance when the variables entered last into the model (Paper I). The 2018 cropping year, an altitude of > 2700 m.a.s.l., poorly drained soil, manure application, flowering growth stage, and early and late planting were significantly associated with high (>50%) mean FBG severity. Non-fertilized fields showed two times greater probability of association with high mean FBG severity than NPS fertilized faba bean fields. Low mean FBG severity (≤50%) showed high probability of association with Chole and Enda Mehoni districts, seedling growth stage, and an altitude of < 2700 m.a.s.l. On the other hand, Ankober, Bassona Worana, Laygaint, and Sinan districts showed 1.57, 1.57, 2.1 and 1.62 times greater probability of association with higher (>50%) mean FBG severity than Tarmaber district, respectively (Paper I). 3.2. Molecular Identification of FBG-Causing Pathogen 3.2.1. Disease symptoms and pathogen morphology The symptoms of FBG in Ethiopia included poor plant stand and major foliage death in severely infested fields. Sunken-well cupping indentation areas of upper leaf surface within which zoosporangia quickly develop and release zoospores into these contained water pools that maximize leaf wetness and enhance dispersal of zoospores, bulging (galling) of lower leaf surface in early symptom development, darkening or browning of such tissues as symptoms develop into more pronounced galls over time, and typical galling on stems and petioles. Symptoms were also found on field pea and clover species. In the histological studies, a range of morphological symptoms were observed that included typical leaf cross sections showing dark brown regions on underside of leaf gall, where large masses of resting spores are produced and typical masses of resting spores dried residues at the end of season. Turbinate cells with two segments and zoosporangium with opening, base of zoosporangium and zoosporangia with internal zoospore mass were also observed. In leaf sectioning, the epidictic phase of zoosporangia for dispersing zoospores was consistently observed. However, a thorough search of sections of infected leaves did not reveal either the presence of numerous short zoosporangia discharging tubes or the binucleate resting sporangia characteristic of Olpidium sp. (Paper II). 28 3.2.2 Sequence BLAST results A total of 173 sequences were obtained, 95 from ITS, 55 from LSU, 20 from SSU, 1 from TEF1α and 2 from RPB2. Nearly 40% of all sequences were close to Didymella, with percentage identity from 83 to 99% and coverage from 27 to 99%. Sequences of 11 isolates from the 2020 collection gave a result of “uncultured fungus” and there was a high similarity between them for the LSU sequences. Within sequences from the ITS region, approximately 39% were close to the genus Didymella, 20% close to Phoma, 14% close to Mycosphaerella, and 19% close to the host genus Vicia. Within sequences from the LSU gene, approximately 53% were close to the genus Didymella, 20% were close to “uncultured fungus”, 5% close to Olpidium, and 4% that were close to each of Cladosporium, and Cymadothea. For the SSU gene, 30% were close to Leucosporidium, 10% from two samples were close to Phoma, 5% each were close to Boeremia, Cryptococcus, Filobasidium, and Phaseoleae (plant) and 35% were close to the host plant genus Vicia. A large portion of sequences identified as genera Didymella, Olpidium, and “uncultured fungus” were those only showing a low percentage query coverage and identity from either or both the ITS and LSU locus/gene. Genera sequenced from SSU showed only small variations both in percentage query coverage and in identity. None of the genera identified from BLAST results with above 97% identity have previously been recorded as causing a “faba bean blister” type symptom. Despite being recorded as either producing a possible symptom with some similarity to “faba bean blister” and/or showing at least some morphological similarity, Physoderma, Synchytrium, or Urophlyctis were not found in these initial BLAST results. 3.2.3. Phylogenetic analysis The phylogenetic tree constructed from the ITS1- 5.8S- ITS2 sequences (38 test sample sequences) obtained in this study and from related isolates from the NCBI database showed that all test isolates formed a group (group 3) with Physoderma maydis (HB683909). Olpidium viciae (HQ677595) from China was close to two Didymella isolates from NCBI, forming group 1. Another four Physoderma isolates and one Synchytrium from NCBI formed group 2. When sequences from partial 18S- ITS1- 5.8S- ITS2- partial 28S (27 test sample sequences) were used to construct a phylogenetic tree, one test sequence was close to Physoderma in group 1, three test sequences in group 3 were close to Physoderma, 22 test sequences were close to Didymella in group 6, group 5 included Physoderma and Synchytrium, and group 4 included Olpidium from NCBI. In the phylogenetic tree constructed from the LSU (28S rRNA) sequences, the 47 test sequences formed four groups, with 13 sequences grouped with Physoderma (group 2); 11 sequences grouped with Physoderma in group 3, two sequences were 29 close to Cymadothea in group 4, 21 sequences grouped with Didymella in group 5, and group 1 was formed by Olpidium sequences from NCBI. Sample sequences that grouped with Physoderma have been deposited in GenBank (accession numbers from MW414613 to MW414631, from MW448404 to 448414, from MW497579 to MW497587, and from MW587325 to MW587329). 3.3. Estimating Yield Loss of Faba Bean Caused by Gall Disease in North Shoa, Ethiopia 3.3.1. Faba bean gall severity Faba bean genotypes, fungicides and spray schedules showed remarkable significant (p0.05) differences for mean FBG severity starting from 62 DAP in 2018 and 58 DAP in 2019 cropping seasons. Similarly, interaction effects of faba bean genotypes, fungicides and spray schedules showed highly significant (p0.0001) differences and influenced FBG severity starting from 72 DAP in 2018 and 68 DAP in 2019 cropping seasons (Paper III). In the 2018 cropping season, high mean FBG disease severity of 70% (local cultivar), 63% (Degaga) and 59.7% (Gora) was recorded from untreated plots of each respective faba bean genotypes. Conversely, the lowest (26.7–34.7%) mean FBG severity was obtained from test genotypes sprayed with Bayleton at 10 days interval. Application of Ridomil Gold at every 10 days interval also found to reduce disease pressure compared with 15 days interval schedule in all faba bean genotypes evaluated. It was observed that spraying fungicides at 10 days interval reduced mean FBG severity on each faba bean genotypes. Among fungicide sprayed plots, higher FBG severity of 41% was obtained from farmers’ local cultivar sprayed with Ridomil Gold at 15 day-interval than the others (Paper III). In the 2019 cropping season, higher mean FBG severity of 68, 60.3 and 55% were noted from unsprayed plots of the local cultivar, Degaga and Gora varieties, respectively. Similarly, high (32%) mean FBG severity was recorded from farmers’ local cultivar sprayed with Ridomil Gold at 15 day-interval, while low (13%) mean FBG severity was observed from Gora variety sprayed with Bayleton at 10 day-interval (Paper III). Comparatively, spraying Bayleton at every 10 days interval lowered FBG severity as compared to Ridomil Gold at similar spraying interval during the experimental periods. In this regard, spraying Bayleton at 10 days interval reduced final mean FBG severity by 48.7, 55.3 and 50.4% for unsprayed plots of the varieties Degaga and Gora and the local cultivar, respectively, in 2018. A similar trend of FBG disease decrease was also calculated in 2019 cropping season. And, genotypes recorded variable levels of mean FBG severity throughout the epidemic periods. The overall mean FBG severity in 2018 was higher than the 2019 cropping season (Paper III). 30 3.3.2. Area under disease progress curve (AUDPC) Highly significant (p0.0001) differences of AUDPC values were calculated among treatments in both the 2018 and 2019 main cropping seasons. In the 2018 cropping season, higher AUDPC values of 3332.3, 3223.5 and 3246%-days were computed from severity assessments of unsprayed plots of the local cultivar, Gora and Degaga varieties, respectively. However, the lowest values of 1915%-days (local cultivar), 1866.7%-days (Gora) and 2055.7%-days (Degaga) were recorded from plots of each faba bean genotype sprayed with Bayleton at 10 days interval for the same 2018 cropping season. Application of Ridomil Gold at 10 days interval also lowered disease pressure and AUDPC values as compared to recordings made on plots treated with both fungicides at 15 days interval. Among fungicides and spray schedules, application of Bayleton at 10 days interval reduced AUDPC by 42.5, 42.1 and 36.7% compared with unsprayed plots of the local cultivar, Gora and Degaga varieties in that order. Moreover, Bayleton fungicide showed comparative advantage over Ridomil Gold at the same spraying schedule. For example, Bayleton covered plots of Degaga (1.7 and 3.1%), Gora (3.2 and 4.8%) and the local cultivar (13.9 and 19.9%) at 10 days interval recorded noticeable AUDPC reductions as compared to Ridomil Gold at 10 and 15 days intervals, respectively. Related phenomena were observed regarding the AUDPC values and associated reductions due to fungicides during the 2019 cropping season as well (Paper III). 3.3.3. Effect of gall disease on grain yield of faba bean Analysis of variance (ANOVA) showed that treatments and their interactions established very highly significant (p0.0001) differences for mean grain yield in both cropping seasons (Paper V). Gora (3.7 t ha-1), Degaga (3.4 t ha-1) and the local (3.1 t ha-1) faba bean genotypes sprayed with Bayleton at 10 days interval attained the indicated high mean grain yields, while low mean grain yields of 1.3, 1.4 and 1.9 t ha-1 were obtained from unsprayed plots of the varieties Degaga and local cultivar, and the Gora, respectively, in 2018 main cropping seasons. Similarly, high mean grain yields of 5.0, 4.2, and 4.6 t ha-1 were obtained from the varieties Gora and Degaga, and the local cultivar sprayed with Bayleton at 10 days interval in that order in 2019 main cropping season. Of course, unsprayed plots of the varieties Gora (2.9 t ha-1) and Degaga (2.4 t ha-1), and the local cultivar (2.5 t ha-1) harvested the lowest mean grain yields compared with plots of each respective faba bean genotypes treated with either fungicide at any spraying schedule. Spraying Bayleton at 10 days interval registered mean grain yield gain of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) over unsprayed plots of the respective genotypes. Also, Bayleton application at 10 days interval recorded observable increase in mean grain 31 yield of 37.5% (Degaga), 31.7% (Gora) and 17.3% (local cultivar) as compared to plots of each faba bean genotype sprayed with Ridomil Gold at 15 days interval in the 2018 cropping season. Similar trends were noted with each test material in 2019 main cropping season. However, faba bean genotypes sprayed with Bayleton at 10 days interval revealed higher grain yield than with Ridomil Gold spray schedule at 15 days interval both in 2018 and 2019 cropping seasons. An overall higher mean grain yield was obtained in 2019 than the mean grain yield harvested in the 2018 cropping season (Paper III). 3.3.4. Effect of gall disease on hundred seed weight of faba bean The analysis of variance (ANOVA) showed significant (p0.05) differences among genotypes, fungicides and application schedules for hundred seed weight in both 2018 and 2019 main cropping seasons (Paper III). In the 2018 cropping season, high hundred seed weight was recorded from the varieties Gora (77.7 g) and Degaga (47.0 g) sprayed with Bayleton at 10 days interval, while the lowest (41.7 g) value of hundred seed weight was recorded from unsprayed plots of the local faba bean cultivar. Likewise, the variety Gora gave heavier (83.2 g) hundred seed weight, followed by the variety Degaga (57.1 g) treated with Bayleton at 10 days interval; however, unsprayed plots of the local cultivar recorded only 42.8 g hundred seed weight in the 2019 cropping season. At all spraying schedules, Bayleton application resulted in improved hundred seed weight in both cropping years. For instance, spraying Bayleton at every 10 days interval showed 11.2 and 24.9% (Degaga), 7.5 and 15.5% (Gora), and 12.3 and 18.7% (local cultivar) hundred seed weight increases over the unsprayed plots of each respective faba bean genotypes in 2018 and 2019, respectively. The results also revealed that the local cultivar and the variety Degaga were more responsive to fungicide applications than the variety Gora (Paper III). 3.3.5. Effect of gall disease on yield components of faba bean The analysis of variance (ANOVA) indicated that there was no significant (p>0.05) difference among treatments and their interactions for days to 50% seedling emergence and days to flowering; however, there was significant difference between sprayed and unsprayed plots, and in days to physiological maturity in both cropping seasons. Unsprayed plots showed relatively short maturity periods (150–153 days) as compared to fungicide sprayed plots (157–160 days). Plant height, number pods per plant and number seeds per pod showed highly significant (p0.01) differences among treatments both in the 2018 and 2019 cropping seasons, except the number of seeds per pod, which exhibited a non- 32 significant (p>0.05) variation among treatments in 2019 (Paper III). The tallest (86.1 cm) plant height was recorded on the variety Gora, followed by Degaga (77.1 cm) sprayed with Bayleton at 10 days interval in 2018 main cropping season. Unsprayed plots of the local cultivar attained the shortest (57.1 cm) plant height compared with other treatments (Paper III). In 2019, high plant heights of 114.8, 113.9 and 113 cm were recorded on the local, Degaga and Gora genotypes treated with Bayleton at 10 days interval, respectively. Moreover, high mean number of pods per plant was obtained from the variety Degaga (9.9) and the local cultivar (9.4) sprayed with Bayleton at 10 days intervals, while low (5.1) number of pods per plant was recorded from plots of unsprayed local cultivar in the 2018 cropping season. Faba bean genotypes also exhibited similar trends regarding mean number of pods per plant in the 2019 cropping season. Furthermore, mean number of seeds counted per pod was highly influenced by faba bean genotypes evaluated and fungicides applied and associated schedules used in 2019 cropping season. High number of seeds per pod was recorded from plots of the varieties Gora (3.1) and Degaga (3.0) sprayed with Bayleton at 10 days interval and low number of seeds per pod (2.1) was recorded from unsprayed local cultivar (Paper III). Though statistically insignificant, high number of seeds per pod was recorded on the variety Gora (3.3) and Degaga (3.3) sprayed with Bayleton at 10 days interval, while the lowest (2.3) number of seeds per pod was recorded from unsprayed plots of the local cultivar in 2018 cropping season. 3.3.6. Relative yield loss of faba bean due to FBG disease Among treatments compared, the genotype sprayed with Bayleton at 10 days interval was used as a reference point to calculate relative grain yield and hundred seed weight losses in both cropping years. Combined use of faba bean genotypes with fungicides at both application schedules reduced grain yield and hundred seed weight losses as compared to untreated plots of each genotype in the two testing years. The highest grain yield losses of 62.5% (Degaga), 54.2% (local cultivar) and 50% (Gora) were calculated from unsprayed plots of each respective genotype in 2018 cropping season. On the contrary, evaluated faba bean genotypes obtained yield increases of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) when treated with Bayleton at 10 days interval in 2018 main cropping season. Similarly, 45.8, 42.2 and 36% yield losses were recorded from unsprayed plots of the local cultivar, Degaga and Gora varieties in that order in 2019 main cropping season. Among fungicide application schedules, Ridomil Gold spraying at 15 days interval showed high level of 33 relative yield loss in all faba bean genotypes in both cropping seasons (Paper III). Moreover, genotypes, fungicides, spray schedules and their combinations also strongly influenced relative loss in hundred seed weight in both testing years (Paper III). In this connection, high hundred seed weight loss of 12.3% was obtained from unsprayed plots of the local cultivar, followed by variety Degaga (11.7%) and Gora (7.8%) in 2018 cropping season. In the 2019 cropping season, the untreated variety Degaga resulted in a relatively higher (24.8%) hundred seed weight loss penalty than the local cultivar (16.1%) and the variety Gora, which recorded only 15.5% (Paper III). Comparative analysis showed that Bayleton sprayed plots of each genotype attained lower grain yield and hundred seed weight losses than treated plots with Ridomil Gold at both application schedules in the two testing years. Similarly, higher grain yield losses due to FBG pressure were calculated from all genotypes tested in 2018 than in the 2019 cropping season. 3.3.7. Correlation of FBG disease parameters with grain yield and yield-components The correlation analysis showed that there was a significant and strong relationship between FBG severity, grain yield and yield-components of faba bean (Paper III). Final mean disease severity established an inverse correlation with faba bean plant height, number of pods per plant, number of seeds per pod, hundred seed weight and grain yield. In 2018, final mean disease severity showed a strong negative correlation with plant height (r = –0.84***), number pods per plant (r = –0.71**) and number of seeds per pod (r = –0.63*) and grain yield (r = –0.96***). Mean disease severity also had negative, but weak relationship with hundred seed weight (r = –0.29ns). Grain yields of the genotypes maintained a positive correlation with plant height (r = 0.86***), number of pods per plant (r = 0.65**) and number of seeds per pod (r = 0.58*) and hundred seed weight (r = 0.39ns). Moreover, yield components had variable levels of positive correlation among each other in 2018 cropping season. During the 2019 cropping season, correlations between FBG severity, grain yield and yield- components were observed to have similar trends as in the 2018 cropping season (Paper III). 3.3.8. Regression analysis relating gall disease severity with faba bean grain yield Regression analysis was done using final or terminal disease severity as a predictor variable and grain yield of the genotypes as a response variable in both 2018 and 2019 cropping years. In the analysis, fungicides and their respective application schedules were considered to describe the associations between mean values of disease severity at the final date of assessment and mean grain yield of each genotype in 2018 and 2019 main cropping seasons (Paper III). The estimated slopes of the regression 34 lines in 2018 cropping season were 0.066, 0.058 and 0.047 on the varieties Degaga and Gora, and the local cultivar, in that order. A similar trend was established regarding estimated slope of the regression line for all genotypes in 2019 cropping season (Paper III). The estimated values implied that an increase in mean values of FBG severity caused remarkable reduction in grain yield of all evaluated faba bean genotypes irrespective of the treatments in both 2018 and 2019 cropping years. Of course, the model explained that yield losses in the varieties Degaga (92.7 and 97.3%), Gora (92.0 and 87.6%) and the local cultivar (95.3 and 98.2%) were attributed to gall disease pressure in 2018 and 2019, respectively. For example, each graph shows that for each unit increase in FBG severity, there were 0.066, 0.058 and 0.047 t ha-1 grain yield loss in the tested Degaga, Gora and the local genotypes of faba bean in order of appearance in 2018 cropping season. In the 2019 cropping season, 0.044 t ha-1 (Degaga), 0.041 t ha-1 (Gora) and 0.041 t ha-1 (Local) predicted grain yield losses of faba bean were estimated at every unit increase in disease severity (Paper III). 3.4. Evaluation of Faba Bean Genotypes for FBG Disease Resistance Reaction 3.4.1. Faba bean gall disease incidence and severity Faba bean gall disease incidence reached 100% within 20 days after disease onset on 90% of the tested faba bean genotypes. Faba bean gall severity increased gradually starting from the onset of the disease to physiological maturity growth stage of the genotypes. Among the tested genotypes, majority of local genotypes (>95%) obtained from Ethiopian Biodiversity Institute (EBI) showed high mean FBG severity as compared to the introduced genotypes. In 2018 cropping season, genotypes obtained from EBI showed 38 to 87% FBG severity and only one genotype (Local collection 228186) showed 38% FBG severity. Five genotypes had less than 55% FBG severity and 195 of the genotypes showed 57 to 87% FBG severity. Similarly, introduced breeding lines showed 32 to 73% FBG severity. One hundred eleven genotypes showed 32 to 50% FBG severity and 104 genotypes had 51 to 73% FBG severity. Mean FBG severities of 58.8% (Degaga), 56% (Gachena), 54% (Gora), and 62.1% (local cultivar) were measured on check genotypes in 2018 cropping season. Among 415 tested genotypes, 116 and 113 of the genotypes showed ≤ 50% > 50 to ≤ 60% FBG severity, respectively. One hundred eighty six genotypes showed > 60 to 90% FBG severity than other genotypes evaluated in 2018 crop season (Paper IV). In the 2019 cropping season, FBG severity ranged fro 37 to 65.33%. Among the tested genotypes, five breeding lines showed > 30 to ≤ 40 FBG severity, 36 breeding lines had > 40 to ≤ 50%, 53 breeding lines attained >50 to ≤ 60% and 10 breeding lines showed > 60 to < 70% FBG severity. Mean FBG severities on check varieties were 52.30, 49.33, 46 and 58.75% on Degaga, 35 Gachena, Gora and local cultivar, respectively, in 2019. Among the selected genotypes, 13 genotypes showed < 45% FBG severity, and three breeding lines showed < 40% FBG severity. Breeding lines 1085, 1082 and 1004 had 37.0, 37.5 and 38% FBG severity, respectively. One genotype (local collection 26884) and 28 introduced breeding lines showed low FBG severity as compared to the check varieties both in the 2018 and 2019 cropping seasons (Paper IV). 3.4.2. Area under disease progress curve (AUDPC) The AUDPC values of FBG on genotype and breeding lines ranged from 900 to 2560 %-days in 2018 copping season. Among 415 genotypes and breeding lines tested, 14 of them showed ≤ 1000%-days of AUDPC value, 148 genotypes and breeding lines recorded ≤ 1500 %-days AUDPC value, 86 genotypes and breeding lines showed ≤ 2000 %-days AUDPC value, 167 genotypes and breeding lines showed > 2000 to ≤ 2655%-days AUDPC value. In 2018, the AUDPC values on check varieties were 1523, 1622, 1640 and 1909 %-days on Gora, Gachena, Degaga and local cultivar, respectively. Similarly in 2019 cropping season, the AUDPC value range on selected genotypes and breeding lines was 990 to 1720 %-days. Likewise, the AUDPC value on the check varieties were 1240, 1326.25, 1352.5 and 1363.75%-days on Gora, Gachena, Degaga and local cultivar, respectively. The AUDPC values on local genotypes were higher than the introduced breeding lines. The AUDPC value in 2018 on local genotypes ranged from 1325 to 2655%-days, whereas on introduced breeding lines, the AUDPC value range was 900 to 2130%-days. One genotype (local collection 26884) and 28 introduced breeding lines showed low AUDPC (1005 to 1295%-days) value as compared to the check varieties in 2018 and 2019 cropping seasons (Paper IV). 3.4.3. Faba bean disease progress rate High disease progress rate was calculated for genotypes and breeding lines in the 2018 cropping season. In the 2019 cropping season, among the tested 104 local genotypes and breeding lines, one genotype (local collection 26884) and 28 breeding lines showed low disease progress rate. The disease progress rate on selected genotypes and breeding lines ranged from 0.0544 to 0.0874 units day–1. The highest disease progress rate (0.0874 units day–1) was calculated for the breeding line 1117, while the lowest disease progress rate (0.0544 units day–1) was calculated for the breeding line 1107 as compared to other breeding lines tested. Low disease progress rates (0.0544, 0.0561, 0.0562, 0.0574, 0.0578, 0.0581 and 0.0588 units day–1) were recorded for the breeding lines 1107, 1004, 1047, 1085, 1082, 1002 and genotype number 28884, respectively. Disease progress rate on the check varieties 36 were 0.0641, 0.0558, 0.0411 and 0.0752 units day–1 on Degaga, Gachena, Gora and local cultivar, respectively (Paper IV). 3.4.4. Grain yield and yield components In the 2018 cropping season, significant difference was obtained on the grain yield of local genotypes and introduced breeding lines. Among the 415 genotypes and breeding lines tested, 119 genotypes obtained from EBI showed less than 0.25 t ha-1 of grain yield. Two hundred ninety-six genotypes and breeding lines showed a grain yield of 0.36 to 2.7 t ha-1. In the 2019 cropping season, grain yield ranged from 0.4 to 3.78 t ha-1. Among the tested genotypes, 28 breeding lines showed a grain yield range of 0.5 to 0.97 t ha-1. Five genotypes obtained from EBI and 44 introduced breeding lines showed 1 to 1.9 t ha-1 and 27 breeding lines showed 2 to 3.78 t ha-1 grain yields. Maximum (3.78 t ha-1) grain yield was recorded from breeding line 1085, which was among introduced genotypes. In the 2018 cropping season, the check varieties showed 1.37, 2.07, 2.25 and 1.02 t ha-1 grain yields from Degaga, Gachena, Gora and local cultivar, respectively. Similarly, 1.64, 2.18, 2.50 and 1.62 t ha-1 grain yields were obtained from Degaga, Gachena, Gora and local cultivar, respectively, during 2019 cropping season. Plant height range in 2018 cropping season was 49.9 to 92.6 cm and 38.4 to 92.1 cm in 2019 cropping season. Number of pods per plant range in 2018 was 4.6 to 13.6 and 2.8 to 10.3 in 2019 cropping season. Similarly, number of seeds per pod range in 2018 was 1.9 to 3.6 and 1.51 to 3.24 in 2019 cropping season (Paper IV). In 2018 cropping season, hundred seed weight on genotypes and breeding lines tested ranged from 34.5-104.7g. The highest (104.7 g) hundred seed weight was recorded from the introduced breeding lines. In 2019 cropping season, hundred seed weight ranged from 38.97 to 111.5g. Three breeding lines showed greater than 100 g hundred seed weight (Paper IV). 3.4.5 Association of FBG disease and yield parameters The correlation coefficient analysis showed that there was a significant and strong association between FBG parameters, grain yield, and yield-components. In 2019, FBG severity showed positive strong correlation with area under disease progress curve (r = 0.89***) and disease progress rate (r= 0.97***). Faba bean gall severity showed negative strong correlation with plant height (r = –0.74***), grain yield (r = –0.66**), and number of pods per plant (r = –0.55***). However, FBG severity showed negative 37 and weak correlation with number of seeds per pod (r= –0.17 ns) and hundred seed weight (r = –0.19 ns). Area under disease progress curve was negatively correlated with plant height (r = –0.69***), grain yield (r = –0.53***) and number of pods per plant (r = –0.69***), but showed negative and weak correlation with hundred seed weight (r= –21*) and number of seeds per pod (r = –0.16 ns). Disease progress rate attained a similar trend in its correlation with grain yield and yield-components as established by severity and AUDPC. On the contrary, grain yield attained positive significant relationship with plant height (r = 0.57***), hundred seed weight (r = 0.29**), and number of pods per plant (r = 0.37***), and a postivive, but non-significant relationship with number of seeds per pod (r = 0.17ns) (Paper IV). 3.5. Reaction of Faba Bean Genotypes to Gall Disease and Resistance Reaction Stability under Natural Infections 3.5.1. AMMI analysis of FBG disease incidence, severity and grain yield of faba bean The AMMI analysis of variance showed highly significant (p0.001) differences for genotypes, environments and genotype by environment interactions for FBG incidence. In the AMMI analysis of FBG incidence, genotype contributed about 80.32% to the variations. However, environment and genotype x environment interaction contributed only 5.32 and 14.36% of the variability, respectively. The environment (E) means for FBG incidence scores of the 21 faba bean genotypes over six environments ranged from 90.4% in Debay Telatgen–Kuy (E5) to 94.19% at Bassona Worana–Mush (E1) (Paper V). For FBG severity, 55.84% of the variations were contributed due to genotypes, and 37.83% and 6.33% contributions were due to environment and GEI, respectively (Paper V). The highest mean FBG severity (82.7%) was recorded at Bassona Worana–Mush (E1) and the lowest (35.2%) was recorded at Farta–Tiratir (E6). The combined means for FBG severity scores of the 21 faba bean genotypes over six environments (E1 to E6) ranged from 50.3% in Farta–Tiratir (E6) to 63.4% in Bassona Worana– Mush (E1). Fabab bean gall severity was high in E1 and E2 as compared to E3, E4, E5 and E6. Thus, E1 was the most favorable site for disease expression, followed by (E2). The AMMI biplot indicated that out of the 21 genotypes, G3, G7, G16 and G17 showed low FBG severity. In contrast, G1, G2, G5, G13 and G21 were susceptible to the disease in all the test environments. Genotypes with IPCA1 scores near zero had little interaction with the environment. The environment (E6) had a low environment score exhibiting little interaction with genotypes. 38 The AMMI indicated that different sources of variations differed greatly as revealed by their sums of squares (Paper V). The first four AMMI selections for mean FBG incidence showed that G5 at E1, G1 at E2 and E4, G2 at E3, E5 and E6 ranked first. The first four AMMI selections for mean FBG severity showed that G1 ranked first and G13 ranked 2nd in all the environments. The other susceptible genotypes (G2, G5 and G13) ranked 3rd and 4th (Paper V). The genotype G16 in five environments and G17 in one environment showed the last rank for FBG severity. Similarly, G16 in one environment (Farta–Gassay) and G17 in five environments showed 2nd ranks from the last. G1, G2, G5, G13 and G21 were grouped together and showed high FBG severity. Genotypes G9, G10 and G20 showed high FBG severity below, but close to mean FBG severity. The genotypes G4, G11, G12, G14, G15 and G18 showed FBG severity close, but above the mean. The genotypes G3, G6, G7, G8, G16, G17 and G19 showed low FBG severity. The AMMI analysis of variance showed highly significant (p≤0.001) differences for genotypes, environments and genotype by environment interactions for faba bean grain yield (Paper V). The relative magnitude of the different sources of variations strogly differed as revealed by their sums of squares. Genotype contributed about 50.86% to the total variations observed, while environments and genotype by environment interaction contributed 38.53 and 10.61% of the variations, respectively. The first and second interaction principal component axis contributed 78.84 and 13.12%, respectively. The average grain yield for the genotypes across environments ranged from 1.3 t ha-1 in E2 to 2.52 t ha-1 in E4. The genotype G8 ranked first at three environments (E1, E2 and E3) and G16 ranked first at three environments (E4, E5 and E6). The genotype G3 ranked second at E1 and E2 and G8 ranked second at E4. The genotype G17 ranked second at E5 and E6. The genotype G4 ranked third at E1 and E3. Similarly, the genotype G7 ranked third at E6. From the AMMI model, the environments E4, E5 and E6 were classified as high yielding, while E1, E2 and E3 were low yielding environments. The genoypes G3, G4, G6, G7, G8, G16 and G17, were grouped together and showed high grain yield from high to low, and G9, G12, G14 and G15 showed good grain yield above and close to the mean. Similarly, G10, G11, G18 and G19 had grain yield below, but close to the mean. G1, G2, G5, G13, G20 and G21 obtained low grain yield. 3.5.2. GGE biplot analysis of FBG disease and grain yield of faba bean GGE biplot analysis for FBG incidence, severity and grain yield was computed and the genotype, environment and genotype x environment interaction were further explored. The genotypes showed variable reactions to FBG incidence and severity at six locations. The first two principal components 39 of the GGE biplot accounted for a total of 64.96% (PC1 = 47.88%, PC2 = 27.08%) of the variations for FBG incidence and the main effect of genotypes and environment on FBG incidence is presented (Paper V). Similarly, in the GGE biplot analysis, majority of the faba bean genotypes evaluated at six locations showed similar FBG severity reactions. The first two principal components (PCs) of the GGE biplot accounted for a total of 92.22% (PC1 = 85.7%, PC2 = 6.52%) of the variations for FBG severity. The polygon view of the GGE biplot of 21 faba bean genotypes for FBG severity revealed the ‘which–won–where’ pattern and mega environments (Figure 5, Paper V). In the biplot view, G1 was the vertex genotypes in the sector that had E1, E3, and E5 and E6. Also, G1 was the most susceptible to FBG disease in all environments. Similarly, G9 and G21 were found in the fourth quadrat (Paper V). On the other hand, G16 and G17 were moderately tolerant genotypes to FBG as indicated by the high negative PC1 score. The other genotypes (G4, G6, G7, G10, G12 and G18 ) within the polygon view and near the origin had low positive or negative PC1 scores, indicating less tolerance to FBG disease and were less responsive than the vertex genotypes. The test environments were almost grouped in a similar category, except E2 and E4 (Figure 5, Paper V). Similarly, the GGE biplot analysis for mean performance and stability of genotypes for FBG severity was considered based on the average-environment coordinates (AEC). The genotypes G1 and G21 were susceptible and unstable, whereas G2, G5 and G13 were suceptible and stable. The genotypes G16 and G17 had low FBG severity (negative low PC1 scores) and unstable across the environments. The genotype G7 was moderately tolerant and stable. In contrast, the genotypes G8, G10 and G12 were stable, but were not tolerant to FBG (data not shown). Quadrat I had genotypes with an Interaction Principal Component Analysis/Axis (IPCA) score near the origin (zero) and average FBG severity showing stability of the genotypes across the environments tested. However, genotypes with high mean performance and large IPCA score are considered unstable across environments, but have specific adaptation to some environments. Discriminating power and representativeness of the test environments for FBG disease were also considered based on average coordinate, vector length and biplot origin. The environments E1 and E2 had the least angles to the Average Environment Axis (AEA) line indicating that there were the most representative of the test environments. In addition, the environment vector for E3 and E4 was relatively long indicating discriminating ability of the genotype. E1 and E3 had the longest vectors and high positive PC1 scores, suggesting that they were more discriminating of the genotypes than the other environments. Environment E5 had the shortest environment vector and PC2 close to zero, suggesting 40 less discriminating ability. The genotype, environment and genotype by environment interactions were further explored through GGE biplot analysis for faba bean grain yield. The AMMI2 biplot (PC1 and PC2) for grain yield of 21 genotypes at six environments are presented at Paper V. The first two principal components (PCs) of the GGE biplot accounted for a total of 97.59% for the variation of faba bean grain yield over six locations. In the ‘which –won-where’ pattern and mega environments analysis, a polygon view of the GGE biplot for grain yield of faba bean resulted in five vertex genotypes with both positive (high yielding) and negative (low yielding) PC1 scores (Paper V). These genotypes included G8 which were vertex genotype in the sector at environments E1, E2, and E3. Genotype G16 was the vertex genotype in E4, E5 and E6. The other three genotypes (G1, G13 and G15) fell in sectors with no environment markers. Three environments, namely E1, E2 and E3, fell in one sector thus encompassing one similar-environment, and E4, E5 and E6 were grouped into another similar-environment. Genotypes in the polygon with no environment indicated that such genotypes had poor performance in all environments. Similarly, the GGE biplot analysis showed mean performance and stability of genotypes for grain yield; based on an average-environment coordinates (AEC). The genotype G17 was unstable and showed high grain yield at E5. The genotypres G4, G7, G8 and G16 showed high grain yield and better adaptations at E4. The genotype G6 showed good grain yield and specific adaptation at E6. In general, G1, G2, G5, G13 and G21 were unstable genotypes identified by AMMI model, with low specific grain yields. From the AMMI biplots, the markers for genotypes were more scattered than the markers for environment indicating that the variability due to genotypes was higher than that due to environments. The biplot contains the average environment axis and the biplot origin. The environments E1 and E3 had the smallest angle with the average environment axis, indicating they were more representatives of the test environments, while E2 and E3 were the least representatives. The environment E2 had the longest vector from the biplot origin, indicating it was the most discriminating of the environments, but not representative. The environments E3 and E6 had moderate vector lengths. The environment E4 had medium vector length. With reference to correlation, none of the environments were negatively correlated as there was no obtuse angle (> 90o) observed among any of the environments (Paper V). All the environments had acute angles (< 90 o) with each other, with some of the environments, like E1 and E3; E4 and E5, having even smaller angles between them, indicating more positive correlations between the environments. The single-arrow on the AEC points out higher mean gain yield. The double-arrowed line (Paper V) was the AEC that points in either direction to greater variability (least 41 stability). The genotypes G3, G12 and G14 were stable with above average performance (Paper V). The GGE biplots showed the discriminating ability and representativeness of the test environments with the average environment and the average coordinates of all test environments. The GGE biplots enables to evaluate the environments and to identify environments that may serve to select superior genotypes in an efficient way for the mega environment. The selected test environment should have high genotype discriminativeness and representativeness. Environments with shorter vectors have less discriminativeness in relation to genotypes, that is, all genotypes tend to perform equally and almost no information about genotypic differences can be revealed by such environments. A short vector could also mean that PC1 and PC2 do not represent that environment very well in cases where G+GE have not been retained properly. The environments E1, E2 and E3 presented a long vector, which means they have high discriminativeness for the genotypes. It is also possible to identify environments with high representativeness that the smaller the angle between an environment and the mean- environment axis, the higher is its representativeness. The environment that had both characteristics more than the others was E1. Environments E2 and E3 had long vectors, but greater angles, implying that they should not be recommended for grain yield (Paper V). 3.6. Integrated Management of Gall Disease of Faba Fean in North Shoa, Ethiopia 3.6.1. Faba bean gall disease epidemic Faba bean gall symptom was appeared at 32 DAP in 2018 and 28 DAP in 2019 cropping seasons; and FBG incidence ranged from 86 to 100% in 2018 and 80.67 to 100% in 2019 (Paper VI). The highest (100%) disease incidence was recorded from unsprayed plots of all the tested faba bean genotypes in the 2018 cropping season. Faba bean genotypes, fungicides and spray schedules showed significant (p0.05) differences for mean disease severity. Significant variation was also observed between fungicides and spraying schedules. Generally, the overall mean disease severity in 2018 was higher than in the 2019 main cropping season (Paper VI). In the 2018 cropping season, the highest 75, 69 and 65.67% mean disease severities were recorded from untreated plots of the local culttivar, Degaga and Gora genotypes, respectively. Conversely, the lowest (21.67%) mean disease severity was obtained from the variety Gora sprayed with Bayleton at 10 days interval or rampart. Among fungicide sprayed plots, the highest disease severity (40%) was observed on the local cultivar sprayed with Ridomil Gold at 15 days interval at 92 DAP compared with others. Application of Ridomil Gold every 10-day interval reduced disease severity more than the 42 application made at 15 days interval on both faba bean varieties and the local cultivar. Interaction effects of faba bean genotypes, fungicides and spray schedules were significant (p0.05) on FBG severity in 2018 cropping season (Paper VI). The integration of the variety Gora and Bayleton fungicide sprayed at 10-day interval attained low disease severity compared with other treatment combinations studied. In the 2019 cropping season, the highest mean disease severities of 70% on local cultivar, 63.33% on Degaga and 60% on Gora were recorded from unsprayed plots of the test faba bean genotypes. The highest (28.9%) mean disease severity was scored from the local cultivar sprayed with Ridomil Gold at 15 days interval, while the lowest (10%) mean disease severity was registered on the variety Gora sprayed with Bayleton at 10 days interval. The variety Degaga and the local cultivar sprayed with Ridomil Gold every 15 days interval gave relatively high disease severity as compared to Gora variety. Spraying Bayleton every 10 days interval reduced FBG severity as compared to Ridomil Gold at similar spray intervals during the experimental periods. 3.6.2. Faba bean gall progress curve The depicted faba bean gall disease progress curves (Paper VI) were used to show FBG development through time on unsprayed and maximum Bayleton protected plots of the genotypes Degaga, Gora and the local faba bean cultivar. On unsprayed plots, FBG showed progression every day in all faba bean genotypes evaluated, starting from the onset of the disease to the end of the epidemic period in both cropping seasons (Paper VI). The effects of fungicide application at 10 and 15 days interval for the management of FBG disease were observed having the same trends of disease progress curves in all faba bean genotypes. Spraying Bayleton at 10 days interval showed more reduction of FBG pressure than the other spray schedule (Paper VI). Disease progress curves on genotypes sprayed with Bayleton at 15 days, and Ridomil Gold at 10 and 15 days intervals were found to lie in between unsprayed and Bayleton sprayed at 10 days interval treatment combinations. 3.6.3. Area under disease progress curve (AUDPC) There were highly significant (p0.0001) differences among treatments for AUDPC values. Decreased AUDPC values were observed on plots sprayed with Bayleton at 10 days interval. For instance, application of Bayleton at 10 days interval on the variety Gora showed the lowest AUDPC value compared with the rest of the treatment combinations evaluated in the experiment (Paper VI). In 2018 cropping season, mean AUDPC values of 3332.3, 3246.0 and 3223.3%-days were calculated from unsprayed plots of the local cultivar, Degaga and Gora variety in that order. Among fungicide sprays, 43 the highest (2390%-days) mean AUDPC value was obtained from the local cultivar sprayed with Ridomil Gold at 15 days interval and the lowest (1866.7%-day) AUDPC value was calculated from variety Gora sprayed with Bayleton at 10 days interval. Similarly, mean AUDPC values of 2204.8%- days (local cultivar), 1926.7%-days (Degaga) and 1760%-days (Gora) were computed from unsprayed plots of each respective faba bean genotype in 2019 main cropping season. Among fungicide-sprayed faba bean genotypes, the highest (1214.2%-days) mean AUDPC value was calculated from the local cultivar sprayed with Ridomil Gold at 15 days interval and the lowest (751.7%-days) AUDPC value was calculated on variety Gora sprayed with Bayleton at 10 days interval. The mean AUDPC values in 2018 were comparably higher than in the 2019 cropping season (Paper VI). 3.6.4. Rate of FBG progression Mean disease progress rates (DPRs) showed significant variations among treatments during the two cropping seasons (Paper VI). In 2018, high disease progress rates were calculated from unsprayed plots of the varieties Degaga, Gora and the local cultivar, which were 0.0571, 0.0531 and 0.0604 unit’s day–1, respectively. Among fungicide sprayed plots, the varieties Degaga, Gora and the local cultivar exhibited high disease progress rates when sprayed with Ridomil Gold at 15 days interval. Spraying Bayleton at 10 days interval lowered disease progress rate in both faba bean varieties and the local cultivar as compared to the other spray schedule (Paper VI). Similarly, in the 2019 cropping season, high disease progression rates (0.0324, 0.0322 and 0.0388 units day–1) were computed for the unsprayed plots of Degaga, Gora and local cultivar in that order. Both faba bean varieties and the local cultivar sprayed with Bayleton at 10 days interval revealed low disease progress rate. The results indicated that the disease progress rate was relatively faster on the variety Degaga and the local cultivar in both epidemic seasons. Moreover, gall disease developed at a relatively higher rate on Ridomil Gold sprayed plots than on plots sprayed with Bayleton over the two years. Spraying fungicides at interval of 10 days reduced disease progress rate, regardless of the faba bean genotypes or the fungicides evaluated in the study (Paper VI). Spraying Bayleton at 10 days interval resulted in 61.65, 76.46 and 69.87% reduction in disease progress rate on unsprayed plots of the varieties Degaga, Gora and the local cultivar, respectively, in the 2018 cropping season. Likewise, spraying Bayleton at 10 days interval obtained 55.86, 62.42 and 55.35% disease progress rate reduction on unsprayed plots of varieties Degaga, Gora and the local cultivar in that order of presentation in 2019 main cropping season. 44 3.6.5. Grain yield of faba bean Faba bean genotypes sprayed with fungicides at different spray schedules showed highly significant (p 0.0001) differences in mean grain yield in both cropping seasons. In 2018 and 2019 cropping seasons, unsprayed faba bean varieties and the local cultivar gained low grain yield as compared to fungicide sprayed plots (Paper VI). In 2018, high mean grain yields of 3.7 t ha–1 (Gora), 3.41 t ha–1 (Degaga) and 3.12 t ha–1 (local) were harvested from sprayed plots of evaluated faba bean genotypes. On the other hand, low mean grain yields of 1.28, 1.41 and 1.85 t ha–1 were harvested from unsprayed plots of the variety Degaga, Gora and the local cultivar, respectively. Closely similar trends were noticed during the 2019 cropping season (Paper VI). Faba bean genotypes sprayed with Bayleton at 10 days interval showed relatively higher mean grain yields than other spray schedule. Among the treatments imposed, integration of variety Gora with Bayleton fungicide at 10 days interval showed higher grain yield than other treatment combinations. Moreover, the mean grain yields obtained under each treatment in 2019 was higher than average grain yields recorded in the 2018 cropping season. Faba bean genotypes sprayed with fungicides at different spray schedules also showed significantly variable differences for a hundred seed weight, plant height, number of pods per plant and number of seeds per pod. 3.6.6. Association of faba bean gall disease and grain yield The correlation analysis showed that there was a significant and strong association between and among FBG parameters and grain yield of faba bean (Paper VI). The last three disease severity records (PSI1, PSI2 and PSI3) manifested positive strong correlations with AUDPC (r = 0.97***, 0.97*** and 0.99***) and disease progress rate (r = 0.97***, 0.97*** and 0.96***), respectively, in 2018. However, FBG disease severity showed negative correlation with grain yield (r = –0.94***) at the last date of disease assessment in 2018. The AUDPC and the disease progress rate also maintained positive associations (r =0.95***) between them. Conversely, AUDPC (r = –0.92***) and disease progress rate (r = –0.95***) established a strongly negative correlation with yield in 2018. In the 2019 cropping season, similar trends were observed regarding associations among and between disease parameters, and grain yield as the 2018 cropping season (Paper VI). 3.6.7. Partial budget analysis In the 2018 cropping season, the partial budge analysis showed that the highest net benefits were calculated from Gora ($1,822.64 ha–1) and Degaga ($1,612.39 ha–1) varieties sprayed with Bayleton at 45 10 days interval. In contrast, low net benefits of $274.81 and $419.82 ha–1 were obtained from unsprayed plots of the variety Degaga and the local cultivar, respectively (Paper VI). Similarly, the highest net benefits of $3,675.36 and $3,311.5 ha–1 were calculated from the variety Gora and the local cultivar sprayed with Bayleton at 10 days intervals, while the lowest net benefit of $1,616.98 ha–1 was obtained from unsprayed plots of the local cultivar, followed by the varieties Degaga ($1,773.6 ha–1) and Gora ($1773.6 ha–1) in 2019 (Paper VI). Net benefit obtained from each faba bean genotype was higher in 2019 than in the 2018 cropping season. Moreover, higher (7.16) value of the marginal rate of return was obtained from the variety Degaga, followed by variety Gora (6.59) and the local cultivar (6.45) with Bayleton sprayed at 15 days interval in 2018 (Paper VI). In the 2019 cropping season, higher (8.92) marginal rate of return was computed from the local cultivar sprayed with Bayleton at 15 days interval, followed by Gora (8.85) sprayed with Bayleton at 10 days interval and Degaga (7.55) varieties sprayed with Bayleton at 15 days interval. The marginal rate of return obtained from plots sprayed with Bayleton fungicide was higher than the rate calculated from Ridomil Gold-sprayed plots. In this regard, Bayleton sprayed plots at 15 days interval showed high marginal rate of return in Degaga and local faba bean genotypes tested both in 2018 and 2019. The marginal rate of return obtained in 2019 on variety Gora was high on Bayleton sprayed plot at 10 days interval. The mean marginal rate of return obtained in 2019 was higher than the rate recorded in 2018 main cropping season. However, frequent application of the fungicide spray revealed less marginal rate of return. 46 4. DISCUSSION The survey results revealed that FBG disease symptoms were observed in all 14 surveyed districts, all growth stages, and all plant parts both in the 2018 and 2019 cropping seasons. The presence of symptoms in all growth stages and plant parts indicated that FBG disease is a major threat at all growth stages that can limit biomass and grain yield. Yan (2012) reported symptoms of the FBG- forming disease on leaf, stem, petiole, but not on pod. However, the current study showed that pods, flowers and root areas of plants were also attacked by the FBG disease. FBG disease symptoms were observed on field pea, lentil, Trifolium spp. and smartweed (Polygonum nepalense) that commonly grow with faba bean. The presence of such alternate hosts in Ethiopia under natural conditions may increase FBG epidemics. In this regard, Xing (1984) confirmed that FBG-forming disease causing pathogen could infect buckwheat, cabbage, cucumber, faba bean, field pea, rapeseed and spinach under artificial inoculation. Field pea is grown as sole or mixed with faba bean and lentil is also grown along with faba bean fields in the same season. Likewise, Trifolium spp. and smartweed are grown under natural conditions in faba bean fields, and farmers are used to feeding these weed species to their cattle. This practice may increase inoculum spread and FBG distribution and epidemics. Remarkably, variable spatial and temporal distribution and relative importance of the FBG were observed across districts. Previously, different levels of FBG disease incidence and severity were reported across faba bean growing districts in Ethiopia (Dereje et al., 2012; Hailu et al., 2014; Teklay et al., 2014; Beyene, 2015; Bogale et al., 2016; Anteneh et al., 2018; Yitayeh et al., 2021b). The variation of the FBG disease across the surveyed districts might be related with weather conditions, occurrence of FBG for 8–9 years in the surveyed districts, diverse agronomic practices, and absence of recommended disease management practices. For instance, majority (12 districts) of the surveyed districts received more than 1000 mm rainfall, in the main rainy seasons in both cropping years. The results also demonstrated that lower FBG incidence and severity were recorded in the short rainy season than the main rainy season in both years, which could be attributed to lower relative humidity and rainfall in the short rainy season. A related conclusion was reached by Yan (2012) who noticed that high relative humidity, air temperature range of 10 to 20 °C and high amount of rainfall are conducive for the epidemics of the FBG pathogen that releases zoospores during the epidemic period. The same author confirmed that air temperature ranging from 0 to 18 °C is suitable for the germination of zoosporangia, but zoosporangia are unable to germinate or poorly germinate above 20 °C. However, at higher temperature and dry soil, most of the disease progress declines (Przetakiewicz, 47 2014). High rainfall and high relative humidity could initiate primary and secondary infections and rain splash could enhance pathogen dispersal. Similarly, Ristaino and Gumpertz (2000) noted that disease increases rapidly when rainfall droplets cause splash dispersal of primary inoculum from the soil to the aerial parts of plants in oospore-forming pathogens. Of course, the dispersal processes in pathogens are known to have major effects on both the spatial and temporal distribution of epidemics (Jeger, 1999; Rossi and Caffi, 2012). Altitudinal variation significantly influenced FBG intensity across districts both in 2018 and 2019 main cropping seasons. In the current study, the mean FBG incidence and severity increased as the altitude increased from 2043 to 3632 m.a.s.l., implying that higher FBG incidence and severity were highly associated with higher altitudinal ranges than lower altitudes of crops inspected. Of course, earlier studies in Ethiopia showed that FBG disease was positively correlated with high altitudinal range (Hailu et al., 2014; Bogle et al., 2016). Similarly, the FBG forming disease was reported at an altitude greater than 2400 m.a.s.l. in China (Xing, 1984; Yan, 2012). Contrary to this current finding, FBG disease was also detected at an altitude of 2043 m.a.s.l., which could be due to the co-occurrence of high relative humidity, rainfall, and other factors that might favor infection and subsequent development of the disease. This could be explained that altitude alone may not have effect on the development of the disease, but when combined with suitable weather conditions it may create a significant effect on the disease development and establishment. Moreover, mean FBG incidence and severity varied with drainage practices in faba bean fields. Poorly-drained fields showed higher association with high FBG components than well-drained fields, since poor drainage causes waterlogging and may create stress on faba bean plants and expose them to FBG infection and development. Proper surface and subsurface drainage of excess water from fields contributes to managing water-related (hydrophilic) diseases (Newton et al., 2010). Likewise, Bogale et al. (2016) reported that poor drainage was associated with high FBG severity. Even though the contribution of soil types in FBG intensity variation was not pronounced across districts in both years, relatively low FBG severity was recorded on black Vertisol compared with red sandy soil of faba bean fields. Light soils might ease the movement of the pathogen infecting the root areas of plants by different mechanisms as compared to heavy Vertisols. With regard to this, different pathogens, including Olpidium species infect plants at the root, stay in the soil and plant debris. However, Lay et al. (2018) indicated that Olpidium viciae transmission through soil and roots is not well studied and 48 not well known, though the causative agent of the FBG disease in Ethiopia is now known from this study to be Physoderma (You et al., 2021). Similarly, the role of different soil types for survival and establishment of Physoderma is also not well known. Throughout the survey areas, time of planting and growth stages showed different levels of FBG pressure. Early and late planting time had a strong association with high mean FBG incidence and severity in both surveying years. In early planting (early June), seeds may be exposed for infection due to long time staying in the soil, and late planting (early to mid-July) is characterized by high rainfall intensity in the study areas, which could cause excess soil moisture during seed germination and may create favorable conditions for infection. Conversely, optimum planting (mid June) could enhance seed germination soon after planting and could result in good seedling establishment though this reasoning needs further study to draw a strong conclusion. Several studies also demonstrated that planting date and crop phenology had correlation with the epidemic development of different diseases (Fininsa and Yuen, 2002; Mengesha and Yetayew, 2018; Endalew et al., 2020). In addition, high mean FBG intensity was highly associated with the flowering growth stage, where the disease started during seedling growth stage and progressively increased and reached maximum at the flowering and podding growth stages. In this regard, Yan (2012) reported that a peak epidemic of FBG-forming disease was recorded at flowering and pod formation growth stages, and then stopped at late podding growth stage. The presence of young and dense succulent leaves, the high nutrient requirement at the flowering growth stage, and the high amount of rainfall and relative humidity may together favor infection by the pathogen and subsequent disease development. In addition to planting time and growth stage, high crop and weed densities showed high mean FBG incidence and severity. Dense planting and severe weed infestation could increase crop canopy that would maintain high humidity and modify the microclimatic conditions that create a favorable environment for the development of the disease. Correspondingly, high crop and weed densities may expose plants to intensive competition for moisture, light, nutrient and space, and thus can speed up foliar (Pande and Narayana, 2002; Wakweya et al., 2016) and soilborne diseases (Yimer et al., 2018), and can cause stunting growth (Agegnehu and Fessehaie, 2006). During the current survey, it was observed that weed infested fields showed poor stand of faba bean population associated with high FBG intensity. Previous report indicates that appropriate fertilization reduces crop physiological stress and decreases disease risks (Veresoglou et 49 al., 2013). In the present FBG disease assessment, application of manure exhibited highly significant association with high mean FBG incidence and severity. Manure application increases plant nutritional status and makes the plant too succulent, thus may favor the pathogen to easily attack the faba bean plants at active growth stages, but this assertion needs further verification. Alternatively, the practice of feeding animals with faba bean straw might be associated with high disease pressure on manure- applied fields and that could be due to the survival of the pathogen (in the form of zoosporangia/zoospores) in the animal gut from feeding on the infected faba bean straw. Further to or apart from this justification, the application of manure can modify the canopy of faba bean plants and, thus, can generate more humid sites (microclimates) for the FBG development. A related study by Yan (2012) observed a similar finding that application of manure increased FBG epidemic. Conversely, NPS applied fields manifested low disease incidence and severity as NPS fertilizer may increase crop vigor, which allows the crop to escape or better tolerate the disease. Several other studies also documented that application of optimal chemical fertilizers would help to manage many crop diseases (Oborn et al., 2003; Dordas, 2008; Veresoglou et al., 2013; Mengesha and Yetayew, 2018). Furthermore, the two-year survey encountered one local cultivar and six improved faba bean varieties with different resistance reactions to the FBG disease. However, none of the genotypes were completely resistant to the disease. Despite this fact on the ground, most of the farmers used to plant susceptible landraces and seeds from previous cropping seasons or exchange seeds with each other due to lack of improved varieties. These practices could be the reason for the high infection and the wide distribution of the FBG disease in the survey areas, but it needs thorough investigation despite that Getaneh et al. (2018) noted that seeds could not be regarded as sources of FBG pathogen or disease. In general, the current study and the logistic regression analyses indicated that district, altitude, soil type, drainage, fertilizer type, planting time, growth stage, year, crop density, and weed density were strongly associated with the FBG incidence and severity, and had significant contribution to the development of the FBG epidemics. The regression model quantified the relative importance of the explanatory variables, indicating the variation of the disease epidemic either alone or in combination. This current study confirmed that the FBG is a key problem in most faba bean-growing areas of the country and now is widely spreading to new areas and becoming a major problem in previously non- infected faba bean growing areas. 50 The results of the causative agent identification, it was observed that a number of morphological characteristics, which either supported an identification of the causal pathogen as Physoderma or that were contrary to its identification as Olpidium viciae. First, in the leaf sections, although we observed the epibiotic phase of zoosporangia for dispersing zoospores, which is a diagnostic characteristic of Physoderma, we did not observe the endobiotic zoosporangia, which is a characteristic of Olpidium (Johns, 1966; Gould and Schaechter, 2009); and Physoderma, has two phases (epibiotic monocentric phase and an endobiotic polycentric phase) (Johns, 1966) that differentiates it from Olpidium. Second, the presence of numerous short zoosporangial discharge tubes or the binucleate resting sporangia, the characteristic of Olpidium were not observed (Hiruki and Alderson, 2011), or the much wider variation in size and frequency of discharge tubes as has been reported in Olpidium (Garrett and Tomlinson, 1967) compared to Physoderma. The morphological studies did not support any consensus that the causal pathogen could be an Olpidium, the symptoms and the presence of resting spores in leaves clearly indicate the pathogen to belong to a Chytridiomyceteous pathogen. Chytridiomycetes are pathogens associated with aquatic habitats, the flagellated stage (zoospores), with plants generally reacting to infection by forming gall structures around the zoosporangia and with zoospore dissemination depending on flooding or (heavy) rainfall. Chrytridiomycete genera, such as Synchytrium endobioticum (potato wart disease), while often misidentified and reclassified, contain some obligate plant pathogens of world-wide importance with resting spores persisting and remaining infective in the soil for very long periods (e.g. potato wart disease >40 years), and as assumed for the FBG pathogen surviving on infested residues in soil in China (Lang et al., 1993). The plant pathogens Olpidium, Physoderma and Synchytrium were all previously classified as Chytridiomycetes (Alexopoulos, 1962; Agrios, 2005) and all the three genera produce thick-walled resting spores and/or zoosporangia, with typical infection caused by zoospores produced from zoosporangia (within growing season) and from carryover of resting spores between seasons. However, only Physoderma and Synchytrium cause plant diseases above ground, generally as a gall symptom that occurs from cells in affected tissues being stimulated to divide repeatedly and enlarge excessively (Agrios, 2005), and as observed in the current study. In contrast, Olpidium spp. are obligate plant root pathogens found commonly throughout the world infecting the roots of wild and domesticated plants and not foliage (Maccarone et al., 2010). Further, the FBG pathogen attacks host stems and leaves, but is not known to cause disease on roots (Alexopoulos, 1962; Kusano, 1912; Lay 51 et al., 2018). In the current study, molecular phylograms were utilized for further investigation. Because multiple sequence alignment (MSA) resulted in high likelihood topology for the isolates, pairwise sequence alignment was used as it best generates optimal alignment (Xia, 2016). An interesting group from LSU matching NCBI isolates, “uncultured fungi”, showed very high similarity between themselves and grouped with the Physoderma isolate from NCBI in group 3 (Paper II). Genera resulting from analysis of SSU sequences were of high percentage identity with isolates in NCBI and were, therefore, considered true genera; however, these genera were clearly not related to the disease symptom or pathogen morphology. For example, the genera Mycosphaerella and Phoma obtained from ITS sequences and Phoma also from SSU sequences, matched with NCBI isolates with a high level of percentage identity and query coverage (>97% percentage identity and >93% query coverage) and, therefore, were assumed to be true matches (Hibbett et al., 2016). All the above- mentioned non-Olpidium and non-Physoderma genera were also isolated and identified by sequencing in the current preliminary investigations. Olpidium belong to the Zygomycota phylum, while Physoderma belong to the Blastocladiomycota phylum (James et al., 2006; Money et al., 2016). In the current phylograms, the genus Physoderma was grouped with Synchytrium in group 2, while Olpidium was further away in the ITS region and also the same applied in partial 18S-ITS1-5.8S-ITS2-partial 28S in groups 2, 4 and 5 of the tree constructed from the partial 18S- ITS1- 5.8S- ITS2- partial 28S sequences (Paper II). The findings of the current study are in contrast to those on spring-sown faba beans in high altitude regions of southwest China, where Yan (2012) used primer pair ITS1 and ITS4 to amplify the ITS region of the pathogen and concluded that the causal pathogen of faba bean blister in China was O. viciae. In the current study, the so called O. viciae (HQ677595) from China was not grouped with any other Olpidium from NCBI, but instead grouped with Didymella in group 1 (Paper II), while sequences of the current test isolates were categorized in group 3 (Paper II) with Physoderma. It is clear that this isolate of O. viciae (HQ677595) from China is neither close to Olpidium nor Physoderma, but is close to Didymella. The partial 18S- ITS1- 5.8S- ITS2- partial 28S sequences amplified from the current test isolates placed most of the isolates with Didymella group and three isolates were grouped with Physoderma; in contrast, all Olpidium isolates from NCBI were grouped in another different group. The phylogram constructed from LSU sequences showed the test isolates distributed through three main groups (2, 3, and 5), where groups 2 and 3 included Physoderma from NCBI; isolates were quite diverse within 52 group 2, but group 3 included isolates with very small differences that constituted the “uncultured fungi” group; and isolates in group 5 were grouped with Didymella from NCBI. Group 1 was Olpidium isolates from NCBI and one of the isolates (2Trl) in group 4 was from Trifolium (Paper II). From the three phylograms, FBG isolates mostly belonged to two groups, Physoderma or Didymella. Physoderma is clearly the main causal pathogen of FBG disease, while Didymella is present as an accompanying pathogen and/or from secondary infection. Didymella, Mycosphaerella, and Phoma can survive saprophytically, and the sexual/teleomorphic stage Didymella fabae was present in the faba bean fields causing blight (Ascochyta fabae, asexual/anamorphic stage). There is no available comparative sequence data for Physoderma as a legume pathogen. Some other obligate pathogens, such as those causing downy mildew and white rust diseases, produce an abundance of spores on the plant surface, making it easier to identify them morphologically and to extract DNA from spores for molecular identification. However, Physoderma, while it produces an abundance of resting spores inside the host, only produces epibiotic zoosporangia to release zoospores for a short period. This characteristic feature, not only makes identification more difficult, but it allows secondary fungal pathogens to contaminate morphological and molecular identification procedures; this has led previous attempts to determine the causal agent of FBG disease to identify either secondary pathogens or other contaminating organisms. This is illustrated by two previous unsuccessful attempts to identify molecularly the causal agent of FBG disease. The first was an investigation by CABI in 2012 from the sample sent from Debre Birhan Agricultural Research Center, where ITS rDNA analysis with FASTA showed >99% similarity to sequences assigned to Phoma and Peyronellaea, with 100% match to Peyronellaea pinodella, with a strong match with Aureobasidium/Kabatiella lini; they also used ITS rDNA analysis with BLAST to highlight similarity with Cryptococcus victoria (CABI, 2012). The second was in 2016, with FBG samples sent from Ambo University to Wageningen Plant Research International, where next generation sequencing analysis showed a low homology, but best fit to Albugo laibachii (Wageningen Plant Research International, 2016). However, neither of these organisms could possibly be the cause of FBG disease as both have significantly different morphology from that observed for FBG pathogen. Albugo species are generally not known to infect Fabaceae and produce different symptoms from FBG disease, and Olpidium and Albugo are unrelated genera despite both being favored by cool and wet environmental conditions. The genus Physoderma has been 53 reported on faba bean and other legume hosts. In Japan, Physoderma fabae Syd (1928) was reported on V. faba in 1927, but not with a gall symptom, but rather causing rusty to reddish-brown orbicular or irregular spots on leaves (Watson, 1971). Subsequently, again in Japan, Physoderma leproides (Trab) Lagerh, (1950) [Basionym: Entyloma leproideum Trab. (1894)] was reported as parasitic in leaves and stems of V. faba, again without a gall symptom, causing rusty to reddish-brown orbicular or irregular spots (Watson, 1971). Outside of Ethiopia, Physoderma is reported across a wide range of different legume hosts. For example, P. trifolii (syn. S. trifolii, O. trifolii, Urophlyctis trifolii) has been reported in China on Astragalus sinicus (Milk vetch); in Australia on Swainsona occidentalis, T. glomeratum, T. subterraneum, T. tomentosum, T. repens; and in India on T. alexandrium, T. resupinatum, T. carolinianum, T. medium, T. montanum, T. pratense, T. repens, T. resupinatum (Watson, 1951; Cunnington, 2003; Fajardo et al., 2017; Pande and Rao, 1998; Watson, 1971). Such infections by P. trifolii are relevant, because FBG pathogen not only additionally infects field pea, but also, it is found on two different Trifolium spp. A significant number of different Trifolium species are found in Ethiopia. The yield loss assessment and developing integrated disease management results revealed that use of different faba bean genotypes and applications of fungicides at different spray schedules reduced FBG disease severity, AUDPC and disease progress rate, and increased grain yield and reduced relative yield and hundred seed weight losses at variable levels. Use of host resistance is one of the key components to manage different plant diseases. Resistant varieties can be the simplest, practical, effective, time saving and economical method of disease management (Agrios, 2005). Among the evaluated varieties, Gora and Degaga were previously reported as moderately resistant to FBG disease (Wulita, 2015; Getenet and Yezbalem, 2017). Nevertheless, variety Degaga was not found to be tolerant to FBG disease in this study and no variety is registered for FBG resistance in Ethiopia so far. However, faba bean varieties tested in Ethiopia showed different degrees of reaction to FBG disease (DBARC Progress Report, 2015), and as well as faba bean breeding lines tested in China revealed variations, but were not resistant to FBG-forming disease (Yan, 2012). Regardless of the treatments imposed, the different genetic backgrounds of the faba bean genotypes could be responsible for the variation in their responses to FBG disease. For instance, higher mean disease severity and AUDPC values were recorded on the farmers’ local cultivar than the two released faba bean varieties even though unsprayed plots of Degaga and the local cultivar statistically on par to each other, but had 54 different degrees of reaction to FBG disease. On the other hand, sources of the planting materials could play undeniable role in influencing epidemic development. Accordingly, the variety Gora was obtained from hybridization and it had relatively lower FBG severity than Degaga and the local cultivar in both 2018 and 2019 testing years. The variety Degaga was also obtained from introduced faba bean lines and both had different genetic sources. Yet, further genetic background studies are mandatory to confirm the real genetic degree of tolerance to FBG epidemics in the test materials. Applications of fungicide sprays were started following the onset of typical disease symptoms at seedling growth stage, and FBG disease pressure began to decrease after three foliar applications with both Bayleton and Ridomil Gold fungicides. Both Bayleton and Ridomil Gold fungicides applied at 10 and 15 days spray intervals attained significant effect on FBG parameters. Bayleton spraying at 10 and 15 days intervals were found more effective than Ridomil Gold. This could be associated with the nature and active ingredients of the fungicides and/or differences in efficacy of the fungicides to manage the disease. Despite the variability observed among fungicides, foliar application of Bayleton at a rate of 0.7 kg ha–1 and Ridomil Gold at a rate of 2.5 kg ha–1 at 10 days interval strongly reduced FBG intensity. Similarly, Teferi et al. (2018) reported that both Bayleton and Ridomil Gold were effective to manage FBG disease in northern Ethiopia. Also, Carbendazim and Bayleton were reported to manage FBG disease in the form of foliar spray and seed dressing; however, seed dressing reduces percent seedling emergence in China (Li juan et al., 1993). Comparatively, results of the present study revealed that Bayleton sprayed plots showed relatively reduced disease severity, AUDPC and disease progress rate compared with Ridomil Gold. Other studies documented that utilizing combined applications of fungicides in integration with host resistance reduced grain yield losses through lowering intensities of different diseases in faba bean (Sahile et al., 2008; Bekele et al., 2018; Teferi et al., 2018; Wondwosen et al., 2019; Mengesha et al., 2021). Accordingly, Teferi et al. (2018) indicated that application of Bayleton (68.5 and 24.9%) and Ridomil Gold (46.1 and 32.5%) to control FBG resulted in grain yield advantages as compared to control plots of faba bean at Ofla and Enda-Mekoni, Ethiopia, respectively. Variable yield advantages over the control plots were recorded on Gora, CS20DK and faba bean gall disease were managed on local faba bean genotypes following two or three foliar sprays of Bayleton or Ridomil Gold at Degem and Mush, Ethiopia (Bekele et al., 2018). Moreover, a study conducted to assess the efficacies of six fungicides for the management of FBG disease at Lay Gorebela and Mush of Ethiopia in 2014 and 55 2015 confirmed that fungicide sprays significantly reduced disease severity and AUDPC and increased grain yield at both locations in the two cropping seasons. However, evaluated fungicides showed variable efficacies against the disease under natural infection conditions (Wondwosen et al., 2019). The findings of the current study also agrees with the observation of Li-Juan et al. (1993) who indicated that Bayleton was the most effective fungicide to manage FBG disease in China. However, findings of Wulita (2015) in central Ethiopia revealed that Matco (Metalaxyl 8%+Mancozeb 64% WP) was more effective than both Bayleton and Ridomil Gold, which could partly be attributed to rate and timing of applications of the fungicides. Integration of tolerant faba bean genotypes with fungicide applications at different spray intervals reduced disease parameters and increased grain yield and yield components, but to different or variable extents. Combination of the tolerant faba bean variety Gora with Bayleton fungicide at 10 days intervals reduced disease parameters, but resulted in low marginal rate of return in 2018. In contrast, the application of Bayleton at 15 days interval gave high marginal rate of return in all tested faba bean varieties in 2018. But the marginal rate of return on variety Gora sprayed Bayleton at 10 days interval was high in 2019. Similarly, lower marginal rate of return was recorded due to Ridomil Gold sprayed plots than treatment with Bayleton. This agrees with the report of Wulita (2015) who indicated that Ridomil Gold sprayed on a local faba bean cultivar showed only a low marginal rate of return in comparison wtih other fungicides evaluated. Low marginal rate of return from Ridomil Gold sprayed plots could be due to high rate of Ridomil Gold fungicide applied per hectare and increased associated application cost. This is because the frequency and spray interval depend on the type of fungicide applied, effectiveness of the fungicide and disease severity. During high pathogen infection and extended rainfall, FBG disease severity may increase and addditional fungicide spray frequency may be crucial. However, fungicides are not always considered environmentally safe, market price may not be affordable, and application costs are increasing from time to time. Thus, developing tolerant faba bean varieties and integrating with application of a minimum frequency of Bayleton fungicide spray is essential. The results of the genotype evaluation study for FBG resistance in the 2018 and 2019 cropping seasons indicated that FBG disease incidence and severity were very high and the test genotypes showed significant differences in FBG parameters. Most of of the local genotypes (97.5%) obtained from Oromia, Amhara, Southern Nations, Nationalities and Peoples’ (SNNP) and Tigray Regions through EBI showed high FBG disease 56 severity in the 2018 main cropping season. Very few genotypes (2.5%) and 95.2% breeding lines showed low FBG disease severity, which could be due to differences in the genetic background of the genotypes and breeding lines tested. Ethiopia has different agro-ecologies and it is expected to have various genetic diversity of faba bean genotypes growing in different parts of the country. Many cultivars are released from local landraces and introduced genotypes in the search for different disease tolerances and improved grain yields. However, there is no variety developed for resistance to FBG disease. Most of the released varieties and farmers’ local cultivars are susceptible to FBG (Teklay et al., 2014; DBARC Annual Progress Report, 2015; Wulita, 2015). On the other hand, high FBG severity recorded on the local cultivars from Ethiopia might have the same origin. Moreover, most of the genotypes obtained from different parts of the country might have originally been introduced from the same source, followed by frequent exchange of seeds among farmers from neighboring regions of the country. Gemechu et al. (2005) reported that there is a tendency, including exchanging seeds among growers, particularly among resource-poor farmers in marginal areas, of selecting for high grain yield and resistance to different common diseases, and these may contribute to disease distribution. However, FBG disease transmission through seed is not yet identified. There are different reports for some disease and grain yield that original sources of landraces might vary, due to forced genetic change by local breeding program and releasing different varieties in different areas (Tamene et al., 2015). Of course, genotypes from the same origin might have different genetic background or genotypes from different regions might have similar genetic background. Similarly, geographic diversity may not necessarily be the reason for genetic diversity, but sources or parental background could be the reason for variations (Gemechu et al., 2005). Among the tested introduced breeding lines, few of them showed low FBG severity and AUDPC values, which could be attributed to the genetic variability that the lines could have like that of the local genotypes. In this regard, research reports showed that faba bean varieties, genotypes and elite breeding lines showed different degree of reaction to FBG disease (Yan, 2012; DBARC, 2015; Alehegn et al., 2018; Negussie et al., 2018). Very few local genotypes and breeding lines selected from the 2018 cropping season and evaluated once again in 2019 showed low FBG severity and high grain yield as compared to the check varieties employed. Such findings could reveal that host resistance can be achieved through evaluation of different genotypes and breeding lines obtained from different sources. Moreover, the results might also shed light on or give a clue that further evaluation of local genotypes and breeding lines for FBG resistance and other agronomic traits should focus, not 57 only on collections across regional states, but also on introduced genotypes. It is also important to evaluate genotypes collected within a region at different agro-ecologies. Field screening is a common practice used to evaluate faba bean genotypes against different diseases of faba bean (Tivoli et al., 2006); however, screening for FBG disease resistance should consider many genotypes that possess different sources of genetic background. In multi-location experiments, the AMMI analysis showed that genotypes were the main source of variation for FBG disease incidence expression, followed by genotype x environment interaction. Similarly, the main source of variation for FBG disease severity was attributed due to genotype, followed by environment; however, the variation due to genotype x environment interaction was very low. This implies that the genetic background of the genotypes contributed more than other factors in all the test locations for the phenotypic reaction of faba bean genotypes against FBG disease. Different degrees of reactions were observed on faba bean genotypes evaluated at six locations. Tolerant faba bean genotypes remained consistently tolerant and, similarly, susceptible faba bean genotypes were consistently susceptible in all the test locations. The current results disagree with previous study reported by Alehegn et al. (2018) that most of faba bean varieties, including Degaga, Dosha and Tumsa evaluated in East Gojjam Zone, were resistant against FBG disease, but different results were obtained in the current study. Genotypes reported by Getenet and Yezbalem (2017) as moderately resistant (Degaga) were not resistant in the current study across the six test locations, but showed variable reactions. Despite the fact, different faba bean genotypes showed different degrees of reaction to FBG disease as reported by different authors (Yan, 2012; DBARC Progress Report, 2015; Wulita, 2015). Similarly, a few of the genotypes evaluated in the current study showed similar moderately tolerant reactions as reported by Negussie et al. (2018). Variability in the response of genotypes to FBG disease depends on the genotypes, environment and genotype by environment interaction. On the other hand, genotypes responded with similar reaction trends across the test environments, indicating that damage due to FBG disease was not influenced by the test environments. That is, variation in FBG severity across locations could be due to the genetic contents of the genotypes or it could be due to the fact that the current test environments have very narrow differences. In contrast, in the AMMI analysis, the contribution of environment was high as compared to genotypes for other faba bean diseases (Beyene et al., 2017). Similarly, equal contributions of the genotypes and genotype x environment interaction to disease expression were also 58 reported for faba bean genotypes evaluated in different environments (Villegas Fernandez et al., 2011). Genotypes generally remained constant from one environment to another environment; however, when the same genotype is subjected to different environments, it can produce a wide range of phenotypic reactions (Tesfaye et al., 2010; Abou-Khater et al., 2022). Differences in the genetic composition play a major role in variable response to a disease. In the current study, some of the genotypes evaluated adapted and responded to FBG disease with similar reaction in wide environments and few of them were adapted and responded differently to FBG disease in specific areas. A previous report indicated that genotypes that are closer to the origin are more stable (Yan, 2011). However, in the current study, only a few genotypes were closer to the origin and these genotypes were not tolerant to FBG disease. A genotype is regarded stable if it is not significantly influenced by the GEI and stable genotypes are those showing a consistent performance regardless of any changes in environmental conditions (Ortiz et al., 2001). In the present study, each polygon was formed by connecting the genotypes that are farthest from the biplot origin so that all other genotypes are inside the polygon and showed ‘which-won-where’ as described by Flores et al. (2013). The polygon view of the GGE biplot revealed that G3, G6, G8, G16, G17 and G19 had low FBG disease severity, but were the least stable genotypes, while G7 was tolerant and stable for FBG disease. On the other hand, G4, G6, G10, G12 and G18 were found stable, but were not tolerant to FBG disease. The test environments fell into a similar sector, except E3, which showed a few differences. In this present study, variation in genotype stability and discrimination of the environment effects for FBG disease were observed. An environment is considered ideal for genotype testing when it discriminates the genotypes and represents the environments (Yan and Kang, 2002; Yan and Kang, 2003; Hongyu et al., 2014). The environment E1 was the most discriminating and representative environment for FBG disease followed by E3 and E2. In contrast, E5, E6 and E4 had less discriminating ability for the genotypes. This indicated that E1 is the most efficient for evaluating the potentials of genotypes for FBG disease tolerance. Therefore, among the six environments, E1 represented the ideal testing environment for FBG disease and would be appropriate for selecting best faba bean genotypes against FBG disease. The findings of the current study coincides with the previous report of Beyene et al. (2017) on the identification of ideal environments for testing the reaction of potential faba bean genotypes against faba bean disease. In the AMMI analysis of variance, grain yield performances for the tested faba bean 59 genotypes were influenced highly by the genotype contribution to the total variation, followed by environment reflecting that the genotypes were highly variable in their responses to different environmental changes. In contrast, the report by Beyene et al. (2016) indicated that environment can contribute more than genotype to the total variations for faba bean grain yield when tested at different environments. There are also studies reported for high contribution of genotype by environment interactions to the variation of grain yield in faba bean (Teklay et al., 2015). Genotype, environment and genotype x environment interactions are expected for the total variations and all these different reports indicated that yield is a very complex trait which is strongly influenced by genotype, environment, or genotype x environment interaction as described by Abou-Khater et al. (2022). AMMI biplot revealed that the genotypes G3, G12 and G14 were stable with above average performance. The genotypes G3, G16 and G17 were high yielding, but adapted to specific environments. The genotypes G6 and G15 had specific adaptation at E5 and E6 with average grain yield above the mean. Similarly, G8 was adapted at E1 and E3 with high grain yield. E4 had the higher yielding genotypes, thus representing a high potential environment, while E2 was the lowest yielding environment. This could be due to the soil and weather variation between the two environments (E2 and E4). Growth and yield of faba bean are determined by climatic, edaphic, and management practices that are not independent of each other and interact to affect the chemical characteristics of the soil. Soil acidity has a dramatic impact on most chemical and biological processes of the crop. Different reports showed that faba bean grows best on heavier-textured soils, but tolerates nearly any soil type (Jenson et al., 2010). A report has also shown that faba bean grows best in soils with pH ranging from 6.5 to 9.0 (Jenson et al., 2010), but poorly at pH values of 5 or less (French and White, 2005). In the current study, the soil pH at E4 was 6.8 and 5.38 at E2 which agrees with the Jenson et al. (2010) report. The soil pH at E4 was higher than at E2 that might contribute to low grain yield at E2. The GGE biplot analysis provided a visual representation of the relationship among the genotypes and test environments and used to evaluate the environments (Yan et al., 2000). The presence of correlation between two environments indicated that similar information about the genotype performance is derived from them and, therefore, could be an option to reduce the number of test environments and, as a result, to establish a cost-effective genotype performance evaluations (Gedif and Yigzaw, 2014). The polygon view of the GGE biplot indicated the presence of a crossover 60 genotype by environment as the environments fell in two sectors of the polygon view and had different high yielding genotypes as described by Yan and Kang (2002). The genotype G8 was vertex genotype in the sector at E1, E2, and E3 environments. The genotype G16 was the vertex genotype at E4, E5 and E6 environments. The other three genotypes fell in sectors with no environment markers. Three environments (E1, E2 and E3) fell in one sector; thus, encompassing one similar environment, and E4, E5 and E6 were grouped into another similar environment category. In the present study, the majority of genotypes in the polygon were scattered with no environment that indicated such genotypes had reduced performance in all environments. This study showed which genotype performed better and stable at which environment, as described by Yan (2014). According to Yan and Tinker (2006), the test environments that are less discriminating provide little information on the genotype differences and should not be used as test environments. The environment E2 had the most discriminating environment for grain yield due to its long vector, followed by E1 and E3. In contrast, E4, E5 and E6 had the least discriminating environments. The selected test environment(s) should have high genotype discriminativeness and representativeness. The relationship between the environments was visualized by drawing a vector that connected each environment to the biplot origin and the correlation between two environments was approximated based on the angle between two vectors (Yan and Tinker, 2006; Papastylianou et al., 2021); the smaller the angle between two vectors, the higher is the correlation between the two environments. Environments with shorter vectors have less discriminatory effects in relation to genotypes, that is, all genotypes tended to perform equally and almost no information about genotypic differences was revealed by such environments. A short vector could also mean that PC1 and PC2 do not represent that environment very well in cases where G and GE have not been retained properly. The environments E1, E2 and E3 had a long vector, which means they had high discriminativeness for the genotypes. It is also possible to identify environments with high representativeness: the smaller is the angle between an environment and the mean environment axis, the higher is its representativeness (Yan and Tinker, 2006). An environment that had both characteristics more than the others was E1. The environment E2 had long vector, but greater angles, implying least representative of the test environments, indicating that it should not be discriminating and should not be recommended. On the other hand, E1 had relatively long vector and small angle and could be the most representative of the environments for grain yield. This study, along with that of Yan and Kang (2002) highlight that an 61 ideal test environment should effectively discriminate genotypes and represent the environments. Thus, in this study among all the six environments, E1 represented the ideal testing environment with high discriminating ability of the genotypes and moderate in representativeness of the test environment for faba bean grain yield. This environment can be used for selecting generally improved genotypes. An environment, such as E2, which was discriminating, but non-representative, is recommended for selecting specifically improved genotypes as described by Yan and Tinker (2005). In this current study, the angles between all the six environments were acute (< 90o), indicating positive correlations among them for both FBG disease severity and grain yield. This suggests that the same information could be obtained about the genotypes from these environments, which were closely associated. The GGE biplot analysis was used to show mega-environment analysis, genotype evaluation and test-environment evaluation, as reported by Yan et al. (2007); however, in the current study the environments were classified almost under similar category. Similarly, Yan et al. (2000) stated that the first principal component in the graphical analysis represents genotypic performance, while the second represents genotypic stability. Therefore, selecting the most representative and few test environments could be used to evaluate genotypes and minimize cost. 5. CONCLUSIONS AND RECOMMENDATIONS 5.1. Conclusions The field survey findings revealed that FBG disease was prevalent in all surveyed districts regardless of factors influencing disease initiation, development and spread. The mean FBG disease incidence and severity were very high in the main rainy season compared with the short rainy season both in 2018 and 2019 years. High altitude, poor soil drainage, flowering growth stage, high weed and crop density, manure application, and early and late planting time are key biophysical factors that aggravate FBG disease epidemics in the study areas. The study indicated that good soil drainage system, proper weed management practices, adjusting plant density, application of NPS blended fertilizer, use of recommended planting time, and developing tolerant faba bean varieties should be considered to sustainably manage FBG disease. Moreover, integrated FBG disease management strategies should be developed and implemented in the study areas and other related agro-ecologies in the production systems of faba bean. Faba bean gall disease had previously been assumed to be caused by Olpidium viciae. However, Olpidium has endobiotic phase inside the plant only and no epibiotic phase. It is a genus generally restricted to roots that does not show symptoms on above-ground plant parts and is not primarily spread by rain splash. In contrast, Physoderma was observed possessing both epibiotic 62 and endobiotic phases that were associated with this genus, with zoosporangia epibiotic in preparation to release zoospores and resting spores and endobiotic residing inside of the plant cells. Physoderma attacks and shows symptoms in all plant parts, and is a pathogen that is primarily spread by rain splash. The phylogram analyses from ITS1-5.8S-ITS2, partial 18S-ITS1-5.8S-ITS2-partial 28S and from LSU (28S) all confirmed that Physoderma is the causal agent of FBG disease. It is clearly evident from the symptom, morphological and molecular perspectives that the true causal agent of FBG disease is Physoderma. Yield loss assessment study concluded that FBG disease causes a significant yield penalty on both local and improved faba bean genotypes. Despite this fact, integration of tolerant faba bean varieties and foliar application of fungicides along with variable application schedules reduced FBG disease epidemics and increased grain and yield components. In this regard, spraying of Bayleton fungicide at 10 days interval along with the varieties Degaga, Gora and the local cultivar remarkably or considerably minimized FBG disease severities and decreased hundred seed weight and grain yield losses, and enhanced grain yield and components at different levels. On the contrary, unsprayed plots of each respective genotype resulted in the highest disease severity, AUDPC, hundred seed weight and grain yield losses in different locations and both testing years. Comparatively, variety Gora gave the highest grain yield and the lowest disease components and yield losses compared with other genotypes tested. Moreover, faba bean genotypes showed superior performance when sprayed with Bayleton to Ridomil Gold application at both spraying schedules in both cropping seasons. Application of Ridomil Gold at 10 days interval reduced disease severity and AUDPC, and increased grain yield and yield components over the 15 days interval spraying schedule studied. However, as high rate of Ridomil Gold per hectare is required for the purpose, it cannot be cost-effective. Therefore, integrating the tolerant faba bean variety Gora with Bayleton fungicide at a rate 0.7 kg ha-1 at 10 days interval application schedule reduced grain yield and hundred seed weight losses. Furthermore, evaluating more effective fungicides, efforts on developing host resistance and testing of large number of faba bean genotypes along with fungicides and other management options are indispensable for sustainabl faba bean production and productivity. Among the 415 genotypes evaluated in 2018 and 2019, one genotype (local collection 26884) and 28 breeding lines attained low FBG severity and AUDPC values. Low mean FBG parameters were recorded on breeding lines 1085, 1082 and 1004. Maximum (3.78 t ha-1) grain yield was harvested from breeding line 1085. The genotype number 26884 and 28 breeding lines that showed low FBG 63 disease parameters and good grain yield were selected for further evaluation and crossing purposes to develop host resistance. Faba bean genotypes evaluated at six locations showed variable FBG severity reactions. Faba bean gall disease was influenced largely by faba bean genotypic differences, and grain yield performances were also influenced highly by genotype. Both the AMMI and GGE biplot analyses indicated almost similar results in FBG severity as well as stability and performances of the genotypes for grain yield. The genotypes G3, G7, G16 and G17 showed low FBG severity; however, G7 was stable. In contrast, G1, G2, G5, G13 and G21 showed high FBG severity. The genotypes G3, G4, G7, G16 and G17 had high grain yield, but less stable genotypes across the test locations. The genotype G8 was stable and produced high grain yield. The GGE biplot showed that Bassona Worana- Mush (E1) was the best environment in discriminating ability and representative of the test locations to evaluate faba bean for FBG severity and grain yield. In general, genotypes such as G3, G7, G8, G16 and G17 are recommendable for further evaluation and crossing purposes. In developing integrated disease management options, considerable FBG reduction was achieved through integration of faba bean genotypes, foliar application of fungicides and different spraying schedules. The results showed that spraying of Bayleton at 10 days interval starting at the onset of the disease and aligning with seedling, vegetative, flowering and podding growth stages, was more effective to reduce FBG severity, AUDPC and rate of disease development, and resulted in increased grain yield and yield components. Low FBG disease parameters and maximum grain yield were obtained from Gora variety sprayed with Bayleton at 10 days interval. However, high marginal rate of return was obtained from spraying Bayleton at 15 days intervals in all the tested genotypes in 2018. But variety Gora sprayed Bayleton at 10 days inervals showed high marginal rate of return in 2019. This study confirmed that integration of tolerant varieties, application of effective fungicides and proper application schedules can increase grain yield and maximize net benefit. Furthermore, developing host resistance, undertaking more effective fungicides evaluation, and wider testing of cultural practice options should now be the focus to manage the FBG disease and maximise grain yields. 5.2. Recommendations 1. Good soil drainage system, proper weed management practices, appropriate adjustment of plant density, application of NPS fertilizer, use of recommended planting time, and developing and use 64 of more tolerant varieties should be considered in faba bean production to successfully minimize FBG epidemics and subsequently to sustain faba bean production in the country. 2. Physoderma is the causal agent of FBG disease and infects faba bean and other hosts, such as field pea, lentil, Trifolium and smartweed. Therefore, management of FBG should also target identified alternate hosts along with faba bean. 3. Hundred seed weight and grain yield losses can be reduced through the integration of tolerant varieties, effective fungicides and proper application schedules. Thus, faba bean growers are encouraged to take into account integrated approaches that consider tolerant varieties, effective fungicides and appropriate application schedules to reduce FBG disease pressure and enahce productivity. 4. One genotype (local collection 26884) and 28 breeding lines that showed low FBG disease parameters and high grain yield have been selected for further evaluation and crossing purposes. Therefore, continuous screening of genotypes along with the identified genotypes from different sources should be performed towards the development of more FBG disease resistant/tolerant varieties to sustain faba bean production. 5. Faba bean gall was influenced largely by faba bean genotypic differences and yield performances were influenced by genotype, environment and genotype x environment interaction. Thus, G3, G7 G8, G16 and G17, which showed low disease parameters and high grain yield performances, are recommendable for further evaluation for both disease severity and grain yield across several environments and seasons, and for crossing purposes to develop high yielding and resistant/tolerant faba bean genotypes. 6. Integration of tolerant varieties and application of effective fungicides with proper application schedules can significantly increase grain yield and maximize net benefit. Application of Bayleton at a rate of 0.7 kg ha−1 at three to four spray frequencies in conjunction with a tolerant faba bean variety can be used to significantly improve management approaches against FBG disease and increase grain yield. 65 6. FUTURE RESEARCH DIRECTIONS 1. Comprehensive investigations on the distribution and intensity of FBG disease, alternate hosts and biophysical factors that influence the disease epidemics should be conducted in the mid- and high altitude areas of faba bean-growing regions. 2. Research should be done to identify host resistance gene(s) and pathogen virulence gene(s), pathogen variability and species characterization involving many cultivars, pathogen isolates, and environmental factors. 3. There are many faba bean landraces in the country and these materials should be exhaustively evaluated against the FBG disease and crossing for FBG disease resistance should be done in consultation with other institutes which are working on FBG disease. 4. Multi-environment experiments should be strengthened across the country, and the local and introduced genotypes should be studied further to investigate the effects of weather variability on the disease development and damage levels. 5. Different integrated disease management practices should be developed and tested/verified across locations and more attention should focus on the implementation of cultural practices that are easily applicable and cost-effective by faba bean growers. 66 7. REFERENCES Abebe, G., Kindu, G., Yechale, M., Birhanu, A., Anteneh, A. and Amlaku, A. 2020. Optimization of P and K fertilizer recommendation for faba bean in Ethiopia: The case for Sekela District. International Journal, 142: 169-179. Available:www.worldnewsnaturalscience.com. Abou-Khater, L., Maalouf, F., Jighly, A., Rubiales, D. and Kumar, S. 2022. Adaptability and Stability of Faba Bean (Vicia faba L.) Accessions under Diverse Environments and Herbicide Treatments. Plants, 11: 251. https://doi.org/10.3390/ plants11030251. Agegnehu, G. and Fessahaie, R. 2006. Response of faba bean to phosphate fertilizer and weed control on Nitisols of Ethiopian highlands. Italian Journal of Agronomy, 2: 281-290. Agegnehu, G., Ghizaw, A. and Sinebo, W. 2006. Yield performance and land-use efficiency of barley and faba bean mixed cropping in Ethiopian highlands. European Journal of Agronomy, 25: 202–207. Agrios, G.A. 2005. Plant Pathology. 5th ed. Elsevier Academic Press, New York. 922 pp. Ahmed, H.A., MacLeod, E.T., Hide, G., Welburn, S.C. and Picozzi, K. 2011. The best practice for preparation of samples from FTA® Cards for diagnosis of blood borne infections using African trypanosomes as a model system. Parasites and Vectors, 4: 68–68. Alehegn, M., Melekamu, T. and Meseret, T. 2018. Screening of Faba Bean (Vicia faba) Varieties against Faba Bean Gall Disease (Olpidium viciae) in East Gojjam Zone, Ethiopia. Journal of Biology, Agriculture and Healthcare, 8: 50–55. Alexopoulos, D.J. 1962. Introductory Mycology, 2nd Edition, John Wiley and Sons, Inc, New York, 613 pp. Anteneh, A., Yohannes, E., Mesganaw, G., Solomon, G. and Getachew, T. 2018. Survey of faba bean (Vicia faba L.) diseases in major faba bean growing districts of North Gondar. African Journal of Plant Science, 12(2): 32–36. Bekele, B., Dawit, W., Kassa, B. and Salvaraj, T. 2018. Management of faba bean gall disease using cultivars and fungicides in North Showa Zone of Central Ethiopia. International Journal of Research in Agricultural Sciences, (5): 2348–3997. Beyene, B. 2015. Survey and identification of new faba bean disease (Kormid) in the highlands of North Shewa Ethiopia. Current Research in Microbiology and Biotechnology, 3: 561–563. Birru, A. 1979. Agricultural field experiment management manual, part III. Institute of Agricultural Research. Addis Ababa, Ethiopia. pp. 35-42. 67 Bitew, B., Fininsa, C., Terefe, H., Barbetti, M.J. and Ahmed, S. 2021. Spatial and temporal distribution of faba bean gall (Physoderma) disease and its association with biophysical factors in Ethiopia. International Journal of Pest Management, https://doi.org/10.1080/09670874. Bitew, B. and Kebede, W. 2012. New faba bean disease in North Shewa-Ethiopia: ‘Faba bean leaf and stem gall’. http//www.arari.gov.et/index.php option. Accessed on 01 June 2014. Bitew, B. and Tigabie, A. 2016. Management of Faba Bean Gall Disease (Kormid) in North Shewa Highlands, Ethiopia. Advances in Crop Science Technology, 4: 1-5. Bogale, N., Setu, B. and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae (Kusano) in major faba bean (Vicia faba L.) growing areas of Northeastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 110: 87-95. CABI (Centre for Agriculture and Bioscience International). 2012. Final Report for Enquiry Numbers 45/12 (W11387) and 50/12 (W11388) from Vicia faba samples showing faba bean gall symptoms as submitted by Beyene Bitew Eshete, 2 pp. Challa, D., Abraham, N., Zerihun, A. and Meseret, T. 2017. Assessment of the Occurrence and Prevalence of Faba Bean Gall (Olpidium viciae) in Western Highlands of Oromiya, Ethiopia. Journal of Natural Sciences Research, 7 (5): 63-67. CIMMYT (International Maize and Wheat Improvement Center/Centro Internacional de Mejoramiento de Maiz y Trigo). 1988. From Agronomic Data to Farmers’ Recommendations: Economic Training Mannual, 79 pp. CSA (Central Statistics Agency). 2021. Agricultural sample survey (2020/2021). Report on area and production of major crops. Statistical Bulletin 578, Central Statistical Authority, Addis Ababa, Ethiopia. Cunnington, J. 2003. Pathogenic fungi on introduced plants in Victoria. A host list and literature guide for their identification. Department of Primary Industries, Research Victoria. 57 pp. DBARC (Debre Birhan Agricultural Research Center). 2005. Debre Birhan Agricultural Research Center, Sub Center and Research sites description. Debre Birhan, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2015. Progress report (unpublished document). DBARC (Debre Birhan Agricultural Research Center). 2016. Progress report unpublished. Dereje, G. and Tesfaye, B. 19933. Faba bean disease in Ethiopia. pp.328-345. In: Asefaw, T. et al. (Eds.), Cool-Season Food Legumes of Ethiopia. Proceedings of the First National Cool-Season Food Legume Review Conference. Addis Ababa, Ethiopia, 16–20 December 1993. Addis Ababa, Ethiopia. 68 Dereje, G. Wendafrash, M. and Gemechu, K. 2012. Faba Bean Galls: A New Disease of Faba Bean in Ethiopia. Holetta Agricultural Research Center. Available at: https://docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE/edit. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean Newsletter, 32: 8–10. Dordas, C. 2008. Role of nutrients in controlling plant diseases in sustainable agriculture: a review. Agronomy and Sustainable Development, 28(1): 33–46. Endalew, E., Terefe, H., Dejene, M. and Kumar, A. 2020. Distribution and association of agro‑ecological factors influencing garlic rust (Puccinia allii) epidemics in Eastern Amhara, Ethiopia. Indian Phytopathology, https://doi.org/10.1007/s42360–020–00297. Fajardo, S.N., Valenzuela, S. Dos Santos, A.F., Gonzalez, M.P. and Sanfuentes, E.A. 2017. Phytophthora pseudosyringaeassociated with the morality of Nothofagus obliquain a pure stand in central-southern Chile. Forest Pathology, 47(6): 12361. Falkenhagen, E. 1996. Comparison of the additive main effects and multiplicative interaction methods with some classical statistical methods in provenance research. The case of South African Pinus radiata trials. Genetics, 3: 81–87. FAOSTAT (Food and Agriculture Organization of the United Nations, Statistical Database). 2018. Rome, Italy, website.faostat.fao.org. FAOSTAT (Food and Agriculture Organization of the United Nations, Statistical Database). 2021. Rome, Italy, website.faostat.fao.org. Fininsa, C. and Yuen J. 2001. Temporal progression of bean common bacterial blight (Xanthomonas campestris pv. phaseoli) in sole and intercropping systems. Europian Journal of Plant Pathology, 108: 485–495. Flores, F., Hybl, M., Knudsen, J.C., Marget, P., Muel, F., Nadal, S., Narits, L., Raffiot, B., Sass, O. and Solis, I. 2013. Adaptation of spring faba bean types across European climates. Field Crops Research, 145: 1–9. Garrett, R.G. and Tomlinson, J.A. 1967. Isolate differences in Olpidium brassicae. Transactions of the British Mycological Society, 50: 423–423. Gauch, H.G. 1992. Statistical analysis of regional yield trials: AMMI analysis of factorial designs, 1st Ed. Elsevier, Amsterdam. pp. 265–276. Gedif, M. and Yigzaw, D. 2014. Genotype x environment interaction analysis for tuber yield of potato using a GGE biplot method in Amhara Region, Ethiopia. Potato Journal, 41: 41–51. 69 Gemechu, K., Mussa, J. Tezera, W. and Getnet, D. 2005. Extent and pattern of genetic diversity for morpho-agronomic traits in Ethiopian highland pulse landraces II. Faba bean (Vicia faba L.). Genetic Resources and Crop Evolution, 52: 551–561. Getaneh, G., Hailu, E., Kassahun, S., Tesfaye, A. and Girma, M. 2018. The causal pathogen, inoculum sources and alternate hosts studies of the newly emerged gall forming faba bean (Vicia faba) disease in Ethiopia. Advances in Crop Sciences and Technologhy, 6(3): 368. DOI:10.4172/2329-8863.1000368. Getenet, Y. and Yehizbalem, A. 2017. Adaptation of faba bean varieties for yield, for yield components and against faba bean gall (Olpidium viciae Kusano) disease in South Gondar, Ethiopia. The Crop Journal, Doi.org/10.1016/j.cj.2017.05.007. Gomez, K.A. and Gomez, A.A. 1984. Statistical Procedures for Agricultural Research, 2nd Ed. John Wiley and Sons, New York. 680 pp. Gould, A.B. and Schaechter, M. 2009. "Fungi: Plant Pathogenic,". pp 457–477. In: Encyclopedia of Microbiology, 3rd Ed. Academic Press, Oxford. Hailemariam, B.N., Tilahun, M., Bazie, S., Desale, T. and Abere, E. 2017. Management of faba bean gall in faba bean producing area of Eastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 111(2): 343–350. DOI: 10.12895/jaeid.20172.641. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N. and Bitew, B., Boydom, A., Kassa, D. and Temesgen, T. 2014. Faba bean gall: A new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Sciences and Technologhy, 2(4): 1–5. Doi.org/10.4172/2329–8863, 1000144. Herrera-Vásquez, J.Á., Cebrián, M.D.C., Alfaro-Fernández, A., Córdoba-Sellés, M.d.C. and Jordá, C. 2009. Multiplex PCR assay for the simultaneous detection and differentiation of Olpidium bornovanus, O. brassicae, and O. virulentus. Mycological Research, 113: 602–610. Hibbett, D., Abarenkov, K., Kõljalg, U., Opik, M., Chai, B. and Cole, J. 2016. Sequence-Based Classification and Identification of Fungi. Mycologia , 108: 1049–1068. Hiruki, C. and Alderson, P. 2011. Morphology and distribution of resting sporangia of Olpidium brassicae in tobacco roots. Canadian Journal of Botany, 54: 2820– 2826. Hongyu, K., Penna, M.G., Araujo, L.B. and Dias, C.T.S. 2014. Statistical analysis of yield trials by AMMI analysis of genotype by environment interaction. Biometrical Letters, 51(2): 89– 102. 70 James, T.Y., Letcher, P.M., Longcore, J.E., Mozley-Standridge, S.E., Porter, D. and Powell, M.J. 2006. A molecular phylogeny of the flagellated fungi (Chytridiomycota) and description of a new phylum (Blastocladiomycota). Mycologia, 98: 860–871. Jeger, M. J. 1999. Improved understanding of dispersal in crop pest and disease management: Current status and future directions of agriculture. Meteorology, 97:331–349. Jensen, E.S., Peoples, M.B. and Hauggaard-Nielsen, H., 2010. Faba bean in cropping systems. Field Crops Research, 115: 203–216. Johns, R.M. 1966. Morphological and Ecological Study of Physoderma dulichii. American Journal of Botany , 53: 34– 45. Kusano, S. 1912. The life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of the College of Agriculture, 4: 142–199. Lay, C.Y., Hamel, C. and Arnaud, M. 2018. Taxonomy and pathogenicity of Olpidium brassicae and its allied species. Fungal Biology, 122: 837–846. Li-juan, L.. Zhao-Hai, Y., Zhao-Jie, Z., Ming-Shi, X. and Han-Qing Y. 1993. Faba bean in China: A state-of-the-art review. International Center for Agricultural Research in the Dry Areas, Aleppo, Syria, pp.127–128. Maccarone, L.D., Barbetti, M.J., Sivasithamparam, K. and Jones, R.A.C. 2010. Molecular genetic characterization of Olpidium virulentus isolates associated with big-vein diseased lettuce plants. Plant Disease, 94: 563–569. Madden, L.V. and Hughes, G. 1995. Plant disease incidence: distributions, heterogeneity, and temporal analysis. Annual Review of Phytopathology, 33: 529–564. McCullagh, P. and Nelder, J.A. 1989. Generalized Linear Models, 2nd Ed. Chapman and Hall, London , 511 pp. Mengesha, G.G. and Yetayew, H.T. 2018. Distribution and association of factors influencing bean common bacterial blight (Xanthomonas axonopodis pv. phaseoli) epidemics in Southern Ethiopia. Archives of Phytopathology and Plant Protection, 51(20): 1066–1089. Misgana, M. 2017. Integrated management of chocolate spot (Botrytis fabae) disease of faba bean (Vicia faba L.) in Ethiopia: a review. International Journal of Research, 5(9): 195–205. DOI:10.5281/zenodo.1002638. Money, N.P., Watkinson, S.C. and Boddy, L. 2016. Fungal Diversity. Academic Press, Boston. pp. 1– 36. 71 Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. 48 pp. Negussie, T., Seid, A., Gemechu, K., Wulita, W., Beyene, B., Asnakeh, T., Anteneh, A. and Yohannes, E. 2018. Faba bean gall in Ethiopia. Current situation and possible solutions. Seventh International Food Legume Research Conference, Marrakesh, Morocco. 265 pp. Newton, A.C., Duncan, J.M., Augustin, N.H., Guy, F.D.C. and Cooke, D.E.L. 2010. Survival, distribution and genetic variability of inoculum of the strawberry red core pathogen, Phytophthora fragariae var. fragariae, in soil. Plant Pathology, 59(3): 472– 479. Oborn, I. Edwards, A.C., Witter, E., Oenema, O., Ivarsson, K., Withers, P.J.A., Nilsson, S.I. and Stinzing, R.A. 2003. Element balances as a toll for sustainable nutrient management: a critical appraisal of their merits and limitations within an agronomic and environmental context. Euroupean Journal of Agronomy, 20: 211– 225. Ortiz, R., Wagoire, W., Hill, J., Chandra, S., Madsen, S. and Stolen, O. 2001. Heritability of and correlations among genotype by environment stability statistics for grain yield in bread wheat. Theoretical and Applied Genetics, 103(3): 469–474. Pande, A. and Rao, V.G. 1998. A Compendium Fungi on Legumes from India. Scientific Publishers, India. 188 pp. Pande, S. and Narayana, R.J. 2002. Effect of plant population densities on the severity of late leaf spot and rust of groundnut. The Plant Pathology Journal, 18(5): 271–278. Papastylianou, P., Vlachostergios, D.N., Dordas, C., Tigka, E., Papakaloudis, P., Kargiotidou, A., Pratsinakis, E., Koskosidis, A., Pankou, C. and Kousta, A. 2021. Genotype x Environment Interaction Analysis of Faba Bean (Vicia faba.) for Biomass and Seed Yield across Different Environments. Sustainability, 13: 2586. Przetakiewicz, J. 2014a. Synchytrium endobioticum. pp. 823–829. In: Liu D (Ed..). Manual of security sensitive microbes and toxins: CRC Press, Boca Raton, USA. Rehner, S.A. and Samuels, G.J. 1994. Taxonomy and phylogeny of Gliocladium analysed from nuclear large subunit ribosomal DNA sequences. Mycological Research, 98: 625–634. Ristaino, J. B. and Gumpertz, M. L. 2000. New frontiers in the study of dispersal and spatial analysis of epidemics caused by species in the genus Phytophthora. Journal of Phytopathology, 38: 541–576. 72 Rossi, V. and Caffi, T. 2012. The role of rain in dispersal of the primary inoculum of Plasmopara viticola. Journal of Phytopathology, 102: 158–165. Safaa, G.K. and Khaled, M.M. 2007. Virus Diseases of Faba Bean in Asia and Africa. Plant Virus, 1: 93–105. Sahile, S., Fininsa, C., Sakhuja, P.K. and Ahmed, S. 2008. Effect of mixed cropping and fungicides on chocolate spot (Botrytis fabae) of faba bean (Vicia faba) in Ethiopia. Crop Protection, 27: 275–282. SAS (Statistical Analysis System). 2014. SAS User’s Guide. SAS Institute. Cary, North Carolina, USA. Sharma, M., Kiran Babu, T., Gaur, P.M., Ghosh, R., Rameshwar, T., Chaudhary, R.G., Upadhyay, J.P., Gupta, O., Saxena, D.R., Kaur, L., Dubey, S.C., Anandani, V.P., Harer, P.N., Rathore, A. and Pande, S., 2012. Identification and multi-environment validation of resistance to Fusarium oxysprum f.sp. ciceris in chickpea. Field Crops Research, 135: 82–88. Sharma, M., Telangre, R., Ghosh, R. and Pande, S. 2015. Multi-environment field testing to identify broad, stable resistance to sterility mosaic disease of pigeon pea. Journal of Genetic Plant Pathological Society of Japan, 81: 249–259. Stielow, B., Lévesque, C.A., Seifert, K., Meyer, W., Irinyi, L., and Smits, D. 2015. One fungus, which genes? Development and assessment of universal primers for potential secondary fungal DNA barcodes. Persoonia-Molecular Phylogeny and Evolution of Fungi, 35: 242–263. Tadesse, N., Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C. and Abraham, A. 2008. Review of research on diseases of food legumes. pp. 85–132. In: Tadesse, A. (Ed.), Increasing Crop Production through Improved Plant Protection. Addis Ababa, Ethiopia. Tamene, T., Gemechu, K. and Hussein, M. 2015. Genetic progresses from over three decades of faba bean (Vicia faba L.) breeding in Ethiopia. Australian Journal of Crop Science, 9: 41–48. Teferi, T.A., Weldemichael, G.B., Wakeyo, G.K. and Mindaye, T.T. 2018. Fungicidal management of the newly emerging faba bean disease “gall” (Olpidium viciae Kusano) in Tigray, Ethiopia. Crop Protection, 107: 19–25. Teklay, A, Tsegaye, B. Ymane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Science, 2: 33–38. 73 Teklay, A., Yemane, N., Muez, M., Adhiena, M., Assefa, W. and Hadas, B. 2015. Genotype by environment interaction of some faba bean genotypes under diverse broomrape environments of Tigray, Ethiopia. Journal of Plant Breeding Crop Science, 7: 79–86. Terefe, H., Finensa, C., Sahile, S. and Tesfay, K. 2015. Effect of temperature on growth and sporulation of Botrytis fabae, and resistance reactions of faba bean against the pathogen. Journal of Plant Pathology and Microbiology, 6: 285–268. Tesfay, M., Tilahun, A, and Russell, A. 2010. Genotype x environment interaction and their translational implication. Journal of the Environment, Doi.org/10.2217/10.75. Tivoli, B., Baranger, A., Avila, C.M., Banniza, S., Barbetti, M.C.W., Davidson, J., Lindeck, K., Kharrat, M., Rubiales, D., Sadiki, M., Sillero, J.C., Sweetingham, M. and Muehlbauer F.J. 2006. Screening techniques and sources of resistance to foliar diseases caused by major necrotrophic fungi in grain legumes. Euphytica, 147: 223–253. Veresoglou, S.D., Barto, E.K., Menexes, G. and Rillig M.C. 2013. Fertilization affects severity of disease caused by fungal plant pathogens. Plant Pathology, 62(5): 961–969. Vilgalys, R. and Hester, M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. Journal of Bacteriology, 172: 4238–4246. Villegas-Fernandez, A.M., Sillero, J.C., Emeran, A.A., Flores, F. and Rubiales, D. 2011. Multiple- disease resistance in Vicia faba: Multi-environment field testing for identification of combined resistance to rust and chocolate spot. Field Crops Research, 124: 59–65. Wakweya, K., Dargie, R. and Meleta, T. 2016. Effect of sowing date and seed rate on faba bean (Vicia faba L.) growth, yield and components of yield at Sinana, highland conditions of Bale, Southeastern Ethiopia. Internationl Journal of Science, 3: 25–34. Watson, A.J. 1971. Foreign bacterial and fungus diseases of food, forage, and fiber crops. U.S.D.A. Agricultural Research Service, 111: 7505. Wheeler, B.E.J. 1969. An Introduction to Plant Diseases. Wiley and Sons, London. 374 pp. White, T.J., Bruns, T., Lee, S. and Taylor, J.W. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for polygenetics. pp. 315–322. In: Inns, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J. (eds.), PCR Protocols: A Guide to Methods and Applications. Academic Press, New York. Wondwosen, W., Dejene, M., Tadesse, N. and Ahmed, S. 2019. Fungicide Management of Faba Bean Gall (Olpidium viciae) in Ethiopia. Turkish Journal of Agriculture, 7(7): 1075–1081. 74 Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) using varieties and fungicides in North Shoa, central Ethiopia. MSc Thesis, Haramaya University, Haramaya, Ethiopia. Xia, X. 2016. Phylogenetic method with pairwise sequence alignment outperforms likelihood methods in phylogenetics involving highly diverged sequences. Molecular Phylogenetics and Evolution, 102: 331–343. Xing, Z.S. 1984. Preliminary study oiler bacteria blister prevention and treatment of disease blister beans. Plant Pathology, 114: 165–173. Xu, Y. 2010. Molecular plant breeding. Centre for Agriculture and Biosciences International, UK. 734 pp. Yan, J.M. 2012. Study on Blister Disease of Broad Bean Caused by Olpidium Viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. Yan, J.M. and Huazhi, Y. 2012. On the occurrence regularity of broad bean blaster. Journal of Sichuan Agricultural University, 30(4): 365–370. Yan, W. 2002. Singular-value partitioning in biplot analysis of multi-environmental trial data. Agronomy Journal, 94: 990–996. Yan, W. 2011. GGE Biplot and AMMI Graphs for the Genotype by Environment Data Analysis. Journal of the Indian Society of Agricultural Statistics, 65(2): 181–193. Yan, W. 2014. Crop variety trials data management and analysis. Wiley, Hoboken. Yan, W., Hunt, L.A., Sheng Q., Szlavnics Z. 2000. Cultivar evaluation and megaenvironment investigation based on the GGE biplot. Crop Sciences, 40(3): 597–605. Yan, W. and Kang, M.S. 2003. GGE biplot analysis: A Graphical Tool for Breeders, Geneticists and Agronomists. CRC press, Boca Raton. 271 pp. Yan, W., Kang, M.S., Ma, B., Woods, S. and Cornelius, P.L. 2007. GGE biplot and AMMI analysis of genotype by environment data. Crop Sciences, 47: 643–655. Yan, W. and Tinker, N.A. 2005. An integrated biplot analysis system for displaying, interpreting, and exploring genotype by environment interaction. Crop Sciences, 45: 1004–1016. Yimer, S.M., Ahmed, S., Fininsa, C., Tadesse, N., Hamwieh, A. and Cook, D.R. 2018. Distribution and factors influencing chickpea wilt and root rot epidemics in Ethiopia. Crop Protection, 106: 150–155. 75 Yitayih, G., Fininsa, C., Terefe, H. and Shibabaw, A. 2021a. Pathogenic variation of Olpidium viciae isolates on faba bean varieties and other legumes. International Journal of Pest Management, https://doi.org/10.1080/09670874.2021.1980245. Yitayih, G., Fininsa, C., Terefe, H. and Shibabaw, A. 2021b. Distribution and association of faba bean gall (Olpidium viciae) disease with agro-ecological factors in northwestern Ethiopia. Journal of Plant Diseases and Protection, https://doi.org/10.1007/s41348–021–00497–0. Yohannes, D. 2000. Faba bean (Vicia faba L.) in Ethiopia. Institute of Biodiversity, Conservation and Research, Addis Ababa, Ethiopia, 43 pp. You, M.P., Eshete, B.B., Kemal, S.A., Van Leur, J. and Barbetti, M.J. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology, https://doi.org/10.1111. 76 8. APPENDICES List of Papers from the Dissertation (I-VI) 77 Paper I Spatial and Temporal Distribution of Faba Bean Gall (Physoderma) Disease and Its Association with Biophysical Factors in Ethiopia Beyene Bitew 1*, Chemeda Fininsa 2, Habtamu Terefe 2, Martin Barbetti 3, and Seid Ahmed 4 1Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 1*Corresponding author, E-mail: beyenebitew@yahoo.com; Mobile: +251-(0)9 11 33 87 89 2School of Plant Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 3School of Plant Biology, Faculty of Science, the University of Western Australia, Australia 4Biodiversity and Crop Improvement Program, International Center for Agricultural Research in the Dry Areas (ICARDA), B.P. 6299, Rabat, Morocco Abstract Productivity of faba bean is highly constrained by an emerging and destructive faba bean gall (FBG) disease caused by the fungus Physoderma. Field surveys were conducted to assess the spatial and temporal distribution as well as intensity of FBG disease and to determine the association of biophysical factors with FBG disease epidemics in Ethiopia, during 2018 and 2019 cropping seasons. A total of 783 fields were assessed across 14 districts (Ankober, Asagert, Bassona Worana, Chole, Debark, Debay Telatgen, Degem, Dessie Zuria, Enda Mohoni, Farta, Laygayint, Meket, Sinan and Tarmaber). Faba bean gall disease incidence and severity were observed in all the surveyed districts with variable incidence and severity. The highest (64%) mean FBG disease severity was recorded in Sinan district (64%) and Ankober district (45%) during 2018 and 2019, respectively. The associations of independent variables with disease incidence and severity were analyzed using the logistic regression model. District, altitude (≥2700 m), poorly drained soil, high weed and crop density, flowering growth stage, manure application, and early or late planting showed a higjhly significant (p0.0001) association with high FBG disease incidence and severity. It is concluded that generally faba bean gall disease was found as a major production constraint of faba bean in all the surveyed districts. Therefore, proper soil drainage, application of NPS fertilizer, weed managements, adjusting crop density, recommended planting time along with tolerant varieties should be targeted. Keywords: Altitude, Assessment, Association, Disease, District, Field Survey, Incidence, Regression Analysis, Severity. 78 1. Introduction Faba bean (Vicia faba L.) is a major cool-season legume crop grown in many countries in the world. Ethiopia is the second largest faba bean producer after China (FAOSTAT, 2018) and a secondary center of diversity for faba bean (Gemechu et al., 2005). Faba bean is produced by 4.1 million smallholders covering over 466,000 ha of land with a production of over a million tons of grain and an average productivity of 2.2 t ha–1 (CSA 2019). The crop is a valuable source of plant-based protein for human food (Jensen et al., 2010; Amsalu, 2014), cash incomes and its straw is key animal feed in the crop-livestock farming system of the highlands of Ethiopia (Yohannes, 2000). Faba bean plays a significant role as a rotational crop in wheat- and barley-based cropping system and contributes to soil fertility through nitrogen fixation (Agegnehu and Fessehaie, 2006). The crop is produced as sole crop and mixed with barley and field pea under rain-fed conditions in the main and small rainy seasons. In spite of the tremendous genetic diversity and multiple benefits to smallholder communities, faba bean productivity at farm level is lower than the average productivity from improved (yielding ≥ 5 t ha–1) varieties. The low productivity is mainly due to the susceptibility of the crop to a number of biotic and abiotic constraints as well as poor access to seeds of improved cultivars (Mussa et al., 2008). Developing high-yielding and disease-resistant varieties for the user is limited in the country. On the other hand, the growing environment in Ethiopia for faba bean is highly conducive to a wide range of plant diseases, particularly in high rainfall years that are common across the highland production areas of faba bean. The most important diseases of faba bean include chocolate spot (Botrytis fabae), ascochyta blight (Ascochyta fabae), rust (Uromyces viciae-fabae), and black root rot (Fusarium solani and F. avenaceum) (Tadesse et al., 2008). Currently, the major faba bean growing areas in central, northern, and northwestern parts of the country are seriously challenged by a newly emerged faba bean gall (FBG) disease (Hailu et al., 2014; Teferi et al., 2018) caused by the fungus Physoderma (You et al.. 2021). How and when the FBG disease was introduced into Ethiopia is not well known, but was first observed in 2011 in some areas in Amhara and Oromia Regions (Dereje et al., 2012; Hailu et al. 2014; Beyene, 2015). Since then, the disease has been expanding at an alarming rate to the rest faba bean-producing zones of Amhara, Oromia, and Tigray Regions. During severe infections, a total crop failure was reported in farmers’ fields (Teklay et al., 2014; Bogale et al., 2016). Similarly, FBG-forming disease was reported in Japan in 1912 and China in the 1970s’ and the disease still remains a significant yield-limiting problem in 79 China (Kusano, 1912; Li-Juan et al., 1993; Yan 2012). Faba bean gall disease symptoms are mainly observed and reported on the leaves, stalks, and petioles (Yan, 2012). The early appearance of the symptoms is light green round spots, with the surfaces of spots becoming rough with the disease development and then proliferating gradually to form intumescent swollen tissues. The pathogen invades and colonizes host epidermal cells, generating a surface disease verrucous protrusion, which is a key symptom or characteristic feature of the disease (Yan, 2012). The pathogen survives on infected plant debris on the soil surface or buried in the soil during the off-season and is believed to survive for about 1 to 2 years (Li Juan et al., 1993). Many surveys were conducted to track the expansion of FBG disease in Ethiopia, but less research efforts have been invested to study the effects of different biophysical factors on the disease development in FBG disease affected areas of the country. Therefore, the objectives of this study were to assess the spatial and temporal distribution and to estimate the FBG disease intensity in Ethiopia as well as to determine the association of biophysical factors with FBG disease incidence and severity in the study areas. 2. Materials and Methods 2.1. Description of the Survey Areas Faba bean gall disease survey was conducted in Amhara, Oromia, and Tigray Regional States during June to September 2018 and 2019 main rainy seasons (Figure 1). The 14 survey districts during the main rainy seasons included Ankober, Asagert, Bassona Worana, Chole, Debark, Debay Telatgen, Degem, Dessie Zuria, Enda Mohoni, Farta, Laygayint, Meket, Sinan, Tarmaber. A similar assessment was also carried out in four districts, namely Ankober, Asagert, Bassona Worana, and Tarmaber districts of North Shoa Zone during the small rainy season (January–May). The surveyed districts were selected based on accessibility and area of faba bean production potential. The majority of the districts are characterized by light Cambisol/sandy loam and only a few of the fields are black Vertisols. In the main cropping seasons, faba bean planting was done early June to early July, while in the short rainy seasons, planting was done at the onset of rainfall starting from mid-January to early February. Most farmers did not apply fertilizer for faba bean cultivation; however, some and very few farmers used NPS fertilizer and manure at the time of planting, respectively. Faba bean is cultivated along with cereal in rotations, mainly, barley and wheat in the study areas. The relative humidity (%) during the main rainy seasons ranged from 60 to 90% and 52 to 94% in 2018 and 2019, respectively. In the short rainy seasons, a relative humidity range of 58.5 to 83.6% in 2018 and from 49.6 to 82.4% in 2019 was recorded. The weather data in both cropping years were obtained from each respective 80 district through the National Meteorology Agency (Table 1). A total of 783 fields were covered both in the main and the short rainy seasons of both cropping years. Of these fields, 373 faba bean fields were inspected in 2018, while 410 faba bean fields were assessed in 2019 cropping season. During the assessment periods, 678 and 105 faba bean fields were studied in the main and short rainy seasons, respectively. 2.2. Faba Bean Gall Disease Assessment Faba bean fields were identified randomly along the main roads at about 5 to 10 km intervals. The FBG disease prevalence, i.e., the geographical distribution of the disease in the district, was determined from the total number of fields inspected per districts and expressed in parcentage as follows: Number of fields infested per district Disease prevalence (%) = Total number of fields assessed per district X 100 In each field, disease incidence and severity were recorded at five spots by moving in X-shape fashion of the fields using a 1 m2 quadrat. The total plants, infected and healthy plants, were counted per quadrat to estimate the FBG disease incidence. Similarly, 12 randomly selected plants per each quadrat were used for determining FBG disease severity. Five records per field were performed and the mean incidence and severity were used for analysis per each field. Disease incidence was calculated using the following formula: Number of plants showing FBG disease symptoms per quadrat Disease incidence (%) = Total number of plants assessed per quadrat X 100 Disease severity was rated using a modified 0–9 scoring scale employed by Ding et al. (1993); where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. Severity scores were converted into percent severity index (PSI) for analysis using the following formula employed by Wheeler (1969): Sum of numerical ratings per quadrat PSI = Total number of plants scored x maximum score on scale X 100 81 Moreover, alternate host, altitude, slope, drainage status, soil type, weed type, weed density, method of planting, cropping system, growth stage, variety, plant population, planting time, fertilizer type and previous crops were collected per inspected field by interviewing the respective farm owners or farmers and by visual observation. Table 1. Altitude ranges and weather data of the districts surveyed for the FBG disease during the 2018 and 2019 cropping seasons. Annual Mean temperature (°C) a District Altitude ranges rainfall (mm) (m.a.s.l.) 2018 2019 2018 2019 Max. Min. Max. Min. Ankober 2987-3155 1507.1 1445.3 NA NA NA NA Asagert 3063-3287 1915.7 1704.3 17.1 4.5 17.13 4.30 Bassona Worana 2600-3133 1927.0 1452.9 20.9 5.5 20.21 6.94 Debark 2638-2780 1320.6 1102.0 20.8 8.3 21.07 9.26 Debay Telatgen 2507-2750 1608.5 1318.3 22.9 10.6 23.60 9.86 Dessie Zuria 2166-2750 1287.0 1155.2 24.2 9.8 22.75 7.15 Farta 2328-3000 1406.0 1345.5 23.4 9.3 21.76 9.17 Laygayint 2840-3249 1297.3 1043.6 18.1 8.6 19.03 8.79 Meket 2060-2920 828.9 764.7 23.7 13.3 26.08 14.54 Sinan 2043-3332 1499.6 1224.6 NA NA NA NA Tarmaber 2550-2866 1444.6 1311.9 NA NA NA NA Chole 2317-3093 1114.1 994.2 NA NA NA NA Degem 2510-3632 1139.4 1075.2 31.3 15.6 30.43 15.82 Enda Mohoni 2327-2916 656.1 571.2 NA NA NA NA Source of weather data: National Meteorological Agency. Altitude= recorded at each field. aNA = temperature data not available for the specific periods. 82 Figure 1. Map showing the districts of the FBG disease survey in three national regional states of Ethiopia, during the 2018 and 2019 cropping seasons. 2.3. Data Analysis The association of the mean FBG disease incidence and severity with biophysical factors was tested using the logistic regression model (Yuen, 2006) with the SAS procedure of GENMOD (SAS, 2014). Mean FBG disease incidence and severity were classified into distinct groups of binomial qualitative data (Fininsa and Yuen, 2001; Terefe and Gudero, 2019) to analyze the relationship between FBG disease parameters and biophysical factors. Class boundaries of ≤ 50% and > 50% for FBG disease incidence and severity were chosen so that groups contained approximately equal totals; thus, yielding a binary dependent variable (Table 2). The single model tested the association of each independent variable alone with the FBG disease incidence and severity. In multiple models, the association of each independent variable with disease incidence and severity was tested when entered last into the model with all independent variables. Only the variables that showed a high association to FBG disease incidence and severity in the single and the multiple models were tested in the reduced multiple models (Yuen, 2006). Deviance reduction and odds ratios were calculated for each 83 independent variable as it was added to the reduced model. The odds ratios obtained by exponentiation of the parameter estimates were interpreted as the relative risk when the values are greater than one. The difference between the likelihood ratio (LRTs) statistics was used to examine the importance of the variables contributions for the case and tested against chi-square values (McCullagh and Nelder, 1989). Mean FBG disease incidence and severity in the short rainy season were very low and only the main rainy season disease parameters were analyzed in the multiple regression models. Table 2. Independent variables and FBG disease incidence and severity classified by districts and fields (n = 678) in the 2018 and the 2019 main rainy seasons. Fields by disease parameter categories Total Variable Variable class number of FBG disease incidence FBG disease fields (%) severity (%) ≤50 >50 ≤50 >50 Bassona Worana 61 5 56 27 34 Degem 60 10 50 35 25 Chole 60 34 26 37 23 Debay Telatgegn 54 11 43 39 15 Sinan 20 1 19 10 10 Debark 67 26 41 50 17 Districts Farta 63 22 41 36 27 Laygaint 58 10 48 42 16 Meket 46 25 21 31 15 Dessie Zuria 62 28 34 45 17 Enda Mehoni 52 13 39 40 12 Ankober 25 1 24 11 14 Asagert 25 8 17 20 5 Tarmaber 25 1 24 11 14 Years 2018 328 96 232 182 146 2019 350 99 251 205 145 Altitude (m.a.s.l.) <2700 230 98 132 172 58 ≥ 2700-3000 280 75 205 191 89 >3000 168 28 140 90 78 Slope Flat 211 68 143 153 58 Gentle slope 423 118 305 303 120 Steep slope 44 13 31 36 8 Drainage 1 Poor 29 15 14 22 7 Moderate 146 30 116 95 51 Good 503 150 353 375 128 84 Table 2. Continued Total Fields by disease parameter categories Variable Variable class number of FBG disease incidence (%) FBG disease severity (%) fields ≤50 >50 ≤50 >50 Soil type Black vertisol 78 26 53 49 29 Red sandy 358 116 242 266 92 Brown Cambisol 242 62 180 177 65 Weed Low 160 46 114 116 44 density2 Optimum 239 77 162 176 63 High 279 73 206 200 79 Previous Barley 214 56 158 159 55 crops Wheat 319 111 208 240 79 Teff 27 7 20 17 10 Fallow 44 10 34 25 19 Potato 74 18 56 47 27 Cropping Sole 550 192 358 414 136 system Mixed 128 30 98 78 50 Variety Local 597 175 422 421 176 Improved 22 2 20 17 5 Unknown 59 20 39 52 7 Growth stage Seedling 19 5 14 13 6 Vegetative 40 6 34 27 13 Flowering 363 131 232 260 103 Podding 256 63 193 187 69 Fertilizer None 491 145 346 363 128 type3 Manure 62 17 45 31 31 NPS 125 45 80 98 27 Planting Flat 672 192 480 488 184 method Raised bed 6 3 3 4 2 Low 153 42 112 118 35 Crop density4 Optimum 299 80 219 208 91 High 226 76 150 163 63 Land Once 45 15 30 36 9 preparation Twice 374 117 257 265 109 Three times 259 69 190 184 75 Planting Early 267 71 196 197 70 time5 Recommended 285 113 172 211 74 Late 126 34 92 76 50 1Poor, moderate and good drainage refer to faba bean fields with standing water, fields with less standing water, and fields with no standing water, respectively; 2Low = weed free (1-2 times hand weeded), optimum = few weeds present (one time hand weeded), and high = high weed infestation (no weeding is practiced); 3Fertilizertype (NPS, none, manure), NPS = (38 kg P2O5, 19 kg N and 7 kg S); 4Crop density was determined within 1 m2 quadrat where low refers to ≤ 30 faba bean plants m–2, optimum refers to 30–40 plants m–2, and high refers to > 40 plants m–2; 5Early = early June planting, recommended = mid-June planting, and late = early to mid-July planting. 85 3. Results 3.1. Symptomatology and Alternate Hosts During the survey, FBG disease symptoms were recorded on seedlings, vegetative, flowering, podding, and maturity growth stages. Similarly, FBG disease symptoms were also observed on different plant parts of faba bean. Circular or irregular green sunken or galls on the upper or lower sides of the leaf were observed on young and new leaves. The green gall gradually changed to brown, became large, and covered the majority of leaves and stems. In the latter stages of the disease development, leaves became blighted and collapsed. Severely infected plants showed short and crumpled forms before flowering and podding growth stages. Most symptoms were recorded on leaf and stem; however, the symptoms were also observed on pods, flowers, and root areas as noted in the current survey (Figure 2 A–E). Moreover, FBG disease symptoms were observed on different plants and then recorded as presumably alternate hosts. The possible alternate hosts that showed typical FBG disease symptoms were field pea (Pisum sativum), lentil (Lens culinaris) (photo not shown), smartweed (Polygonum nepalense) and Trifolium spp. (Figure 3 A–C). The symptom on field pea was observed on leaves and stems, but the symptoms on lentil, Trifolium and smartweed were observed on leaves. The symptoms on Trifolium spp. and smartweed were recorded under faba bean fields at Sinan and Bassona Worana districts in East Gojjam and North Shoa Zones, respectively. Faba bean gall disease symptom on lentil was observed at Bassona Worana district, while it was observed on field pea in all surveyed districts both in the main and in the short rainy seasons. 86 A B C D E Figure 2. FBG disease symptoms on faba bean leaf (A), stem (B), root area (C), flower (D), and pod (E). A B C Figure 3. FBG disease symptoms observed during the survey on field pea (A), Trifolium (B) and smartweed (C) in 2018 and 2019 cropping seasons. 3.2. Distribution, Incidence and Severity of FBG Disease in the Survey Areas The field survey results showed that FBG disease was prevalent in all surveyed fields and districts. However, FBG disease prevalence, incidence, and severity varied across districts and years (Table 3). In the 2018 main rainy season, the highest 100% FBG disease prevalence was recorded in Ankober, Asagert, Sinan, and Tarmaber districts. The lowest prevalence (66.7%) was recorded in Chole district in Arsi Zone. Similarly, the highest (100%) FBG disease prevalence was recorded at Ankober district, 87 while the lowest (44.5%) at Chole district in the 2019 main rainy season. Generally, the mean FBG disease prevalence ranged from 55.6 to 100% for two-survey years. Faba bean gall disease incidence was 100% in Ankober, Sinan, and Tarmaber districts. Conversely, the lowest (66.7%) mean FBG disease incidence was recorded at Asagert and Chole districts in 2018 main rainy season. Similarly, the highest (100%) and the lowest (44.2%) mean FBG disease incidence was recorded at Ankober and Chole district, respectively, in 2019 main rainy season (Table 3). The two-year mean FBG disease incidence ranged from 55.4 to 100%. At district level, the decrease in the mean FBG disease incidence at Asagert and Chole districts was 33.3% as compared to Ankober, Sinan, and Tarmaber districts in 2018 crop season. Also, the mean FBG disease incidence reduction at Chole district was 44.2% compared with Ankober district in 2019. Moreover, there was 15.5% overall mean FBG disease incidence decrease in 2019 when compared to the 2018 main rainy season. More or less, a similar pattern of results with regard to FBG disease severity was observed in both cropping years. In the 2018 main rainy season, the highest mean FBG disease severity was recorded in Sinan (63.9%), Bassona Worana (61.8%), Ankober (59.6%), and Tarmaber (58.7%) districts, while the lowest (20.5%) was recorded at Asagert district. In 2019, the highest mean FBG disease severity was recorded at Ankober (45.0%), Bassona Worana (43.5%), and Tarmaber (42.4%) districts, and the lowest was recorded at Enda Mehoni (11.2%) district (Table 3). The two-year mean FBG disease severity ranged from 19.3% to 52.7%. A mean FBG disease severity decrease of 67.8% was computed for Asagert as compared to Sinan district in the 2018 main rainy season. Similarly, in 2019, the mean FBG disease severity decrease in Enda Mehoni district was 75.1% compared with Ankober district. Comparably, lower (about 32.4% decreases) overall mean FBG disease severity was recorded in 2019 than in the 2018 main rainy season. Additionally, FBG disease was also recorded in the short rainy season both in 2018 and in 2019. The highest (80%) FBG disease prevalence was recorded at Ankober district, while there was no FBG disease recorded in Bassona Worana district in the 2018 cropping season. In 2019, the highest (52%) and the lowest (14%) mean FBG disease prevalence was recorded in Ankober and Asagert district, in that order. Faba bean fields at Ankober district obtained the highest mean FBG disease incidence (74.3 and 56.7%) and severity (16.5 and 12.1%) in 2018 and 2019, respectively. The lowest (2.7%) mean FBG disease severity was recorded at Bassona Worana district in 2019 short rainy season (Table 3). 88 Among independent variables, altitude >2700 m.a.s.l. showed 35.4% mean FBG disease severity increase as compared to altitude ≤2700 m.a.s.l. Drainage practices influenced FBG disease development where good drainage practice reduced mean disease severity by 23.1% as compared to poor drainage. High weed infested fields revealed 14.5% mean FBG disease severity increase when compared with low weed density. Flowering growth stage and late planting increased mean FBG disease severity by 20.3 and 25.2% as compared to seedling growth stage and recommended planting date, respectively. Though a few farmers used to apply it, manure application raised mean FBG disease severity by 29.2% compared with NPS received faba bean fields. Regional states considered in the assessment recorded variable FBG disease parameters. In this regard, Amhara Regional State obtained mean FBG disease prevalence of 89.5 and 76.6%, incidence of 89.3 and 76.2%, and severity of 44 and 31.6% in 2018 and 2019 cropping years, respectively. Similarly, mean FBG disease prevalence, incidence, and severity of 78.4, 81.7, and 38% were recorded in Oromia Regional State in that order in 2018 cropping season. A closely related trend was noted in 2019. In addition, the Tigray Regional State attained mean FBG disease prevalence of 83.2%, incidence of 79.4%, and severity of 27.9% during the survey periods over the two years. 89 Table 3. FBG disease prevalence, incidence and severity on faba bean in survey districts in the 2018 and 2019 cropping seasons. Total number revalence ean ean Mean disease parameters District/season of fields P (%) M Incidence (%) M Severity (%) Incidence (%) Severity (%) a 2018 2019 2018 2019 2018 2019 2018 2019 + SE + SE a Main rainy season: Ankober 10 15 100.0 100.0 100.0 100.0 59.6 45.0 100±0 52.3±7.3 Asagert 10 15 100.0 62.0 66.7 62.9 20.5 31.3 64.8±1.9 26.0±5.5 Bassona Worana 31 30 90.3 87.1 96.7 89.1 61.8 43.1 92.9±3.8 52.7±9.2 Debark 37 30 89.2 68.3 96.7 68.4 40.4 27.7 82.5±14.1 34.0±6.3 Debay Telatgegn 24 30 87.5 78.4 83.3 72.9 45.3 28.9 78.1±5.2 37.2±8.2 Dessie Zuria 32 30 81.3 64.9 83.3 59.0 34.3 17.2 71.2±12.2 25.7±8.6 Farta 33 30 75.8 60.8 83.3 68.9 31.6 26.1 76.2±7.2 28.8±2.8 Laygaint 28 30 89.3 74.1 96.7 85.0 39.9 33.4 90.9±5.8 36.7±3.3 Meket 21 25 71.4 49.8 76.0 51.0 27.5 20.3 63.5±12.4 23.9±3.6 Sinan 10 10 100.0 96.8 100.0 87.9 63.9 31.8 94.0±6.1 47.9±16.1 Tarmaber 10 15 100.0 100.0 100.0 92.8 58.7 42.4 96.4±3.6 50.6±8.1 Chole 30 30 66.7 44.5 66.7 44.2 25.4 13.2 55.4±11.3 19.3±6.1 Degem 30 30 90.0 88.1 96.7 80.9 50.6 36.1 88.8±7.9 43.4±7.3 Enda Mehoni 22 30 86.4 80.0 83.3 76.0 44.7 11.2 79.7±3.7 27.9±16.7 Mean - - 87.7 75.3 87.8 74.2 43.2 29.1 - - 90 Table 3. Continued Total number District/season of fields Prevalence Mean disease parameters a (%) Mean Incidence (%) Mean Severity (%) Incidence (%) Severity (%) 2018 2019 2018 2019 2018 2019 2018 2019 + SE a + SE a Short rainy season: Ankober 10 15 80.0 52.0 73.3 56.7 16.5 12.1 65+8.3 14.3±2.2 Asagert 10 15 20.0 14.0 33.3 12.3 4.50 5.0 22.8+10.5 4.8±0.3 Bassona Worana 15 15 0.0 20.0 0.0 10.0 0.0 2.7 15±5 1.3±1.3 Tarmaber 10 15 50.0 26.0 33.3 17.3 8.0 5.3 25.3+8 6.7±1.3 Mean - - 23.3 20.0 22.2 13.1 4.2 4.3 - - aSE = standard error. 91 3.4. Associations of Biophysical Factors with FBG Disease Epidemics Among the independent variables, district, year, altitude, drainage, soil type, weed and crop density, growth stage, fertilizer type, and planting time, were the most important variables in their association with the mean FBG disease incidence when entered first and last into the model (Table 4). High FBG disease incidence (>50%) was strongly associated with Ankober, Bassona Worana, and Sinan districts, the 2018 cropping year, an altitude of >2700 m.a.s.l., poor drainage, flowering growth stage, early planting, and manure application (Table 5). Early planting time showed 3 times greater probability of association with mean FBG disease incidence of >50% than recommended planting time. On the contrary, low mean FBG disease incidence (≤50%) showed high probability of association with Chole, Meket, and Enda Mehoni districts, seedling growth stage, and an altitude of < 2700 m.a.s.l. Ankober, Sinan and Bassona Worana districts showed 4.1, 2.1, and 3 times greater probability of association with high mean FBG disease incidence than Tarmaber district, respectively (Table 5). Amongst independent variables, district, year, planting time, growth stage, altitude, drainage, weed density, and fertilizer type were the most important variables in their association with the mean FBG disease severity when entered first and last into the model. However, soil type and crop density lost their significance when the variables entered last into the model (Table 4). The 2018 cropping year, an altitude of > 2700 m.a.s.l., poorly drained soil, manure application, flowering growth stage, and early and late planting were significantly associated with high (>50%) mean FBG disease severity (Table 6). Non-fertilized fields showed 2 times greater probability of association with high mean FBG disease severity than NPS fertilized faba bean fields. Low mean FBG disease severity (≤50%) showed high probability of association with Chole and Enda Mehoni districts, seedling growth stage, and an altitude of < 2700 m.a.s.l. On the other hand, Ankober, Bassona Worana, Laygaint, and Sinan districts showed 1.57, 1.57, 2.1 and 1.62 times greater probability of association with high (>50%) mean FBG disease severity than Tarmaber district, respectively (Table 6). 92 Table 4. Logistic regression model for mean FBG disease incidence and severity, and likelihood ratio test on independent variables during the 2018 and the 2019 main rainy seasons. FBG disease mean incidence, LRT a FBG disease mean severity, LRT a Independent variable Df VEF VEL VEF VEL DR Pr > ꭓ2 DR Pr > ꭓ2 DR Pr > ꭓ2 DR Pr > ꭓ2 District 13 8172.2 <0.0001 8034.1 <0.0001 3214.8 <0.0001 2370.2 <0.0001 Year 1 11.3 <0.0001 46.3 <0.0001 1530.6 <0.0001 1155.5 <0.0001 Altitude 2 771.9 <0.0001 579..1 <0.0001 119.3 <0.0001 132.4 <0.0001 Drainage 2 736.9 <0.0001 667.5 <0.0001 81.4 <0.0001 86.6 <0.0001 Soil type 2 115.8 0.0004 135.1 <0.0001 11.4 0.0056 3.8 0.9082 Weed density 2 426.1 <0.0001 139.9 <0.0001 140.9 <0.0001 69.6 <.0001 Growth stage 3 1939.0 <0.0001 771.2 <0.0001 386.2 <0.0001 297.5 <0.0001 Fertilizer type 2 19.1 0.0003 85.0 <0.0001 16.9 0.0002 28.9 <0.0001 Crop density 2 87.3 <0.0001 78.7 0.0306 15.2 0.0060 12.2 0.3313 Planting time 2 3038.8 <0.0001 2562.0 <0.0001 554.3 <0.0001 487.0 <0.0001 a LRT = likelihood ratio test; VEF = variable entered first in the model; VEL= variable entered last in the model; DR = deviance reduction; Pr = probability of ꭓ2 value exceeding the deviance reduction; ꭓ2 = Chi-square; and df = degrees of freedom. 93 Table 5. Analysis of deviance, natural logarithms of odds ratio and standard error of FBG disease incidence and likelihood ratio tested on independent variables in reduced regression model, during the 2018 and the 2019 main rainy seasons. FBG disease mean Estimate Variable Residual incidence, LRT b deviance a df Variable class Loge (Odds SE Odds ratio) c ratio DR Pr > ꭓ2 Intercept 58674.59 – 0.09 0.7652 – 0.0591 0.1979 1.0608 Districts 50502.41 13 61.05 <0.0001 Bassona Worana 0.9539 0.1221 2.5958 90.00 <0.0001 Degem –1.1688 0.1232 0.3107 1590.86 <0.0001 Chole –5.1033 0.129 0.0060 290.06 <0.0001 Debay Telatgegn –2.0162 0.1184 0.1331 25.55 <0.0001 Sinan 0.743 0.147 2.1022 478.32 <0.0001 Debark –2.5621 0.1171 0.0771 1234.28 <0.0001 Farta –2.6410 0.1193 0.0656 319.90 <0.0001 Laygaint –2.2695 0.1189 0.1033 680.91 <0.0001 Meket –3.0647 0.1174 0.0466 552.83 <0.0001 Dessie Zuria –2.7239 0.1158 0.0712 493.08 <0.0001 Enda Mehoni –4.1896 0.1269 0.0151 35.89 <0.0001 Ankober 1.4081 0.2305 4.0881 389.02 <0.0001 Asagert –2.5058 0.127 0.0816 0 <0.0001 Tarmaber 0* . 1 Year 50501.11 1 46.18 <0.0001 2018 0.1353 0.0200 1.1548 . . 2019 0* . 1 Altitude 49729.22 2 188.40 <0.0001 <2700 –0.6856 0.0500 0.50378 37.57 <0.0001 2700-3000 0.2845 0.0464 1.32909 . . >3000 0* . 1 Drainage 48815.82 2 340.19 <0.0001 Poor 0.9185 0.0498 2.5055 232.34 <0.0001 Moderate 0.4251 0.0279 1.52974 . . Good 0* . 1 Soil type 48800.98 2 4.42 0.0356 Black vertisol 0.0813 0.0387 1.08469 35.01 <0.0001 Red sandy 0.1423 0.0240 1.15292 . . Brown cambisol 0* . 1 94 Table 5. Continued FBG disease mean Estimate Variable Residual incidence, LRT b deviance a DF Variable class Loge (Odds SE Odds ratio DR Pr > ꭓ2 ratio) c Weed 47992.97 2 18.17 <0.0001 Low –0.1175 0.0276 0.88914 density 139.13 <0.0001 Optimum –0.2828 0.0240 0.75367 . . High 0* . 1 Growth 43555.77 3 164.88 <0.0001 Seedling –1.1862 0.0924 0.3053 stage 0.79 0.3752 Vegetative –0.0589 0.0664 0.9428 134.73 <0.0001 Flowering 0.451 0.0389 1.5698 . <0.0001 Podding 0* . 1 Fertilizer 43501.03 2 6.88 0.0087 None –0.0881 0.03401 1.0345 type 13.32 <0.0003 Manure 0.1636 0.0448 1.1777 . . NPS 0* . 1 Crop 43171.56 2 37.77 <0.0001 Low –0.1869 0.0304 0.8295 density 5.16 0.0231 Optimum –0.0552 0.0243 0.9462 . . High 0* . 1 Planting 39901.95 2 313.12 <0.0001 Early 1.0076 0.0569 2.7390 time 368.56 <0.0001 Recommended –0.8265 0.0431 0.4375 . . Late 0* . 1 a Residual deviance = unexplained variation after fitting the model; b LRT = likelihood ratio test; DR = deviance reduction; Pr = probability of chi-square value exceeding the deviance reduction; ꭓ2 = Chi-square; c* Reference group; SE = standard error; and df = degrees of freedom. 95 Table 6. Analysis of deviance, natural logarithms of odds ratio and standard error of FBG disease severity and likelihood ratio tested on independent variables in reduced regression model, during the 2018 and the 2019 main rainy seasons. FBG disease mean Estimate Variable Residual severity, LRT b Variable class Loge SE Odds deviance a df DR Pr > ꭓ2 (Odds ratio ratio) c Intercept 18805.58 – – – 0.0714 0.0667 1.0740 District 15590.82 13 10.30 0.0014 Bassona Worana 0.4503 0.0626 1.5688 5.80 0.0162 Degem –0.1636 0.0680 0.8490 598.8 <0.0001 Chole –1.8208 0.0744 0.1618 21.20 <0.0001 Debay Telatgegn –0.2929 0.0637 0.7460 10.40 0.5284 Sinan 0. 4840 0.0768 1.6225 125.16 <0.0001 Debark –0.6556 0.0586 0.5191 322.38 <0.0001 Farta –1.1055 0.0616 0.3310 120.90 <0.0001 Laygaint 0.7436 0.0676 2.1034 284.75 <0.0001 Meket –1.0906 0.0646 0.3360 249.17 <0.0001 Dessie Zuria –0.9522 0.0603 0.3858 345.67 <0.0001 Enda Mehoni –1.1552 0.0621 0.3149 6.26 0.0124 Ankober 0.4483 0.0679 1.5658 189.43 <0.0001 Asagert –0.9878 0.0718 0.3723 . . Tarmaber 0* . 1 Year 14060.18 1 1476.68 <0.0001 2018 0.6851 0.0178 1.9839 . . 2019 0* . 1 Altitude 13940.89 2 54.80 <0.0001 <2700 –0.2284 0.0385 0.7958 0.07 <0.0001 2700-3000 0.0643 0.0341 1.0664 . . >3000 0* . 1 Drainage 13859.48 2 50.09 <0.0001 Poor 0.2011 0.0473 1.222 17.82 <0.0001 Moderate 0.1777 0.0239 1.1944 . . Good 0 0* 1 Weed 13708.24 2 16.24 <0.0001 Low –0.0976 0.0242 0.9070 density 69.10 <0.0001 Optimum –0.1724 0.0207 0.8416 High 0* . 1 96 Table 6. Continued Variable Residual df FBG disease mean Variable class Estimate SE Odds deviance a severity, LRT b Loge ratio DR Pr > ꭓ2 (Odds ratio) c Growth 13373.30 3 134.56 <0.0001 Seedling –0.9035 0.0779 0.4051 stage 26.01 <0.0001 Vegetative –0.2679 0.0525 0.7649 12.04 0.0005 Flowering 0.1061 0.0306 1.1119 . . Podding 0* . 1 Fertilizer 13356.37 2 7.90 0.0049 None 0.4761 0.0309 1.6097 type 4.78 0.0288 Manure 0.0864 0.0395 1.9090 . . NPS 0* . 1 Planting 12859.15 2 0.56 <0.0001 Early 0.0325 0.0436 1.2304 time 214.50 <0.0001 Recommended –0.5111 0.0349 0.5998 . . Late 0* . 1 aResidual deviance = unexplained variation after fitting the model; bLRT = likelihood ratio test; DR = deviance reduction; Pr = probability of chi-square value exceeding the deviance reduction; ꭓ2 = Chi-square; c* Reference group; SE = standard error; and df = degrees of freedom. 4. Discussion During the 2018 and 2019 assessment periods, FBG disease symptoms were observed in all surveyed districts, all growth stages, and all plant parts. The presence of symptoms in all growth stages and plant parts indicated that the FBG disease is a major threat in all growth stages that can limit biomass and grain yield. Yan (2012) reported symptoms of the FBG disease on leaf, stem, petiole, but not on pod. However, the current study revealed that pods, flowers and root areas of plants were also attacked by the FBG disease. In this study, FBG symptoms were observed on field pea, lentil, Trifolium spp. and smartweed commonly grown with or around faba bean. The presence of such alternate hosts in Ethiopia under natural conditions may increase FBG disease epidemics. In this regard, Xing (1984) confirmed that the FBG disease causing pathogen could infect buckwheat, cabbage, cucumber, faba bean, field pea, and rapeseed and spinach under artificial inoculation. Field pea is grown as sole or crop mixed with faba bean and lentil is also grown along with faba bean fields in the same season. Likewise, Trifolium spp. and smartweed are grown under natural conditions in faba bean fields, and farmers used to feed these 97 weed species to their cattle. This practice may increase FBG disease distribution and epidemics in the country. On the other hand, the presence of the FBG disease in all surveyed districts over the two years could confirm that the FBG disease is a major threat to faba bean production in major faba bean-growing areas. Similarly, Hailu et al. (2014) reported FBG and other foliar diseases of faba bean as a major production constraint in faba bean-growing areas of Ethiopia. However, remarkably variable spatial and temporal distribution and relative importance of the FBG disease was observed. Previously, different levels of FBG disease incidence and severity were also reported across faba bean growing districts in Ethiopia (Dereje et al., 2012; Hailu et al., 2014; Teklay et al., 2014; Beyene, 2015; Bogale et al. 2016; Anteneh et al., 2018). The variation of the FBG disease across the surveyed districts might be related with weather conditions, occurrence of FBG disease for 8–9 years in the surveyed districts, agronomic practices, and absence of disease management practices. For instance, the majority of the surveyed districts received more than 1000 mm rainfall, except Meket and Enda Mohoni districts, in the main rainy seasons of both cropping years. The results also demonstrated that lower FBG disease incidence and severity were recorded in the short rainy season than in the main rainy season in both years, which could be attributed to low relative humidity and rainfall (below 475 mm) in the short rainy season. A related conclusion was reached by Yan (2012) who noticed that high relative humidity, air temperature range of 10 to 20 oC and high amount of rainfall are conducive for the epidemics of the FBG disease that releases zoospores in the epidemic period. The same author confirmed that air temperature ranging from 0 to 18 oC is suitable for the germination of zoosporangium, but zoosporangia are unable to germinate or poorly germinate above 20 oC. However, at higher temperature and dried soil, the disease progress declines (Przetakiewicz, 2014). Altitudinal variation significantly influenced FBG disease pressure across districts both in 2018 and 2019. For example, the mean FBG disease incidence and severity increased as the altitude increased from 2043 to 3632 m.a.s.l., implying that higher FBG disease incidence and severity were highly associated with higher altitudinal ranges than lower altitudes inspected. Of course, earlier studies in Ethiopia showed that FBG disease was positively correlated with a high altitudinal range (Hailu et al., 2014; Bogle et al., 2016). Similarly, the FBG-forming disease was 98 reported at an altitude greater than 2400 m.a.s.l. in China (Xing, 1984; Yan, 2012). Contrary to this finding, FBG disease was also scored at an altitude of 2043 m.a.s.l., which could be due to the co-occurrence of high relative humidity, rainfall, and other factors that might favor infection and subsequent development of the disease. This could be explained as altitude alone may not have effect on the development of the disease, but when combined with suitable weather conditions it may create a significant effect on the disease development and establishment. As high rainfall and dew during the crop growing periods may favor and spread the pathogen to re- infect the healthy faba bean plants and increases the epidemics (Yan, 2012). Moreover, mean FBG disease incidence and severity varied with drainage practices in the faba bean fields. Poorly-drained fields showed a higher association with high FBG disease components than well-drained fields, as poor drainage causes waterlogging and may create stress on faba bean plants and expose them to FBG disease infection and development. Hence, proper surface and subsurface drainage of excess water from fields contributes to managing water related diseases (Newton et al., 2010). Likewise, Bogale et al. (2016) reported that bad drainage was associated with high FBG disease severity. Even though the contribution of soil types in FBG disease intensity variation was not pronounced across districts in both years, relatively low FBG disease severity was recorded on black Vertisol compared with red sandy soil of faba bean fields. Light soils might ease the movement of the pathogen infecting the root areas of plants by different mechanisms as compared to heavy Vertisols. With regard to it, different pathogens, including Olpidium species infect plants at the root, stay in the soil and plant debris. However, Lay et al. (2018) indicated that Olpidium viciae transmission through soil and root is not well studied and not well known though the causative agent of the FBG disease in Ethiopia is Physoderma (You et al., 2021) and the role of different soil type for survival and establishment of this pathogen is not well known. Throughout the survey areas, time of planting and growth stages showed different levels of FBG disease pressure. Early and late planning time had a strong association with high mean FBG disease incidence and severity in both years. In early planting (early June), seeds may be exposed for infection due to long time staying in the soil, and late planting (early to mid-July) is characterized by high rainfall intensity in the study areas, which could cause excess soil moisture 99 during seed germination and may create favorable conditions for infection. Conversely, optimum planting could enhance seed germination soon after planting and could result in good seedling establishment though this reasoning needs further study to draw a strong conclusion. Several studies also noted that planting date and crop phenology had correlation with the epidemic development of different diseases (Fininsa and Yuen, 2002; Mengesha and Yetayew, 2018; Endalew et al., 2020). And, high mean FBG disease intensity was highly associated with the flowering growth stage, where the disease started during seedling growth stage and progressively increased and reached maximum at the flowering growth stage. In this regard, Yan (2012) reported that a peak epidemic of FBG disease was recorded at flowering and pod formation growth stage, and then stopped at late podding growth stage. The presence of young and dense succulent leaf and high nutrient requirement at the flowering growth stage and the high amount of rainfall and relative humidity may favor the pathogen to infect. In addition to planting time and growth stage, high crop and weed densities showed high mean FBG disease incidence and severity. Dense planting and severe weed infestation could increase crop canopy that would maintain high humidity and modify the microclimatic conditions that create a favorable environment for the development of the disease. Also, high crop and weed densities may expose plants to intensive competition for moisture, light, nutrient and space, and thus, can speed up foliar (Pande and Narayana Rao, 2002; Wakweya et al., 2016)) and soilborne diseases (Yimer et al., 2018), and can cause stunting growth (Agegnehu and Fessehaie, 2006). During the survey, it was observed that weed infested fields showed poor stand of faba bean population as well as high FBG disease intensity. Appropriate fertilization reduces crop physiological stress and decreases disease risks (Veresoglou et al., 2013). In this assessment, application of manure showed highly significant association with high mean FBG disease incidence and severity. Manure application increases plant nutritional status and makes the plant too succulent, thus favoring the pathogen to easily attack the faba bean plants at active growth stages, but this assertion needs further studies. On the other way round, the practice of feeding animals with faba bean straw and high disease pressure on manure-applied fields could be due to the survival of the pathogen in the animal gut from the infected faba bean straw feeds. Further to this, the application of manure can modify the canopy 100 of faba bean plants and, thus, can generate more humid sites for the FBG disease development. A related study by Yan (2012) generated a similar finding that application of manure increased FBG disease epidemic; and Yan and Huazhi (2012) indicated that continuous cropping of faba bean and manure application could increase FBG disease severity. Conversely, NPS applied fields resulted in low disease incidence and severity as NPS fertilizer may increase crop vigor, which allows the crop to escape or tolerate the disease. Several other studies also proved that that application of optimal chemical fertilizers would help to manage many crop diseases (Oborn et al,. 2003; Dordas, 2008; Veresoglou et al., 2013; Mengesha and Yetayew, 2018). Furthermore, the two-year survey recognized one local cultivar and six improved faba bean varieties with different reactions to the FBG disease. However, none of the genotypes were completely resistant to the disease. Despite the fact, most of the farmers used to plant susceptible landraces and seeds from previous cropping seasons or exchange seeds with each other due to lack of improved and disease resistant varieties. These practices could be the reason for the high infestation and the wide distribution of the FBG disease in the assessed areas, but it needs thorough investigation though Gezahegn et al. (2018) noted that seeds could not be regarded as source of FBG disease. Regarding host resistance, there is no resistant variety developed for the FBG disease in the country; however, evaluation study by Wulita (2015) and Getenet and Yehizbalem (2017) identified a few FBG disease tolerant faba bean varieties. Similarly, Yan (2012) reported that there was variation among the tested faba bean breeding lines, but no resistant varieties. The current study and the logistic regression analyses indicated that district, year, altitude, drainage, fertilizer type, growth stage, planting time, soil type, crop density, and weed density were strongly associated with the FBG disease incidence and severity, and had significant contribution to the development of the high FBG disease epidemics. The regression model quantified the relative importance of the explanatory variables, indicating the variation of the disease epidemic either alone or in combination. This study confirmed that the FBG disease is a key problem in most faba bean growing areas of the country and now is widely spread to new areas and become a major problem in previously non-infested faba bean growing areas. The major reasons for the wide distribution and spread of the FBG disease in the country are the 101 year-round use of susceptible varieties, less research attention to the disease, lack of resistant varieties, poor field sanitation, and the absence of integrated disease management package. Thus, knowledge of the disease cycle and its epidemiology and designing proper and integrated management strategy are pertinent to sustain productivity of the crop. 5. Conclusions This study demonstrated that FBG disease was prevalent in all surveyed districts regardless of factors influencing disease spread and development. The mean FBG disease incidence and severity were very high in the main rainy season compared with the short rainy season both in 2018 and in 2019. The model singled out high altitude, poor drainage, flowering growth stage, high weed and crop density, manure application, and early and late planting time as key biophysical factors aggravating FBG disease epidemic in the study areas. Since FBG disease is highly prevalent and very severe in the central, northern, and northwestern parts of Ethiopia and faba bean is the most important pulse crop both in production volume and area coverage in the country, giving due attention in developing effective management strategy of this disease is critical. The study suggested that good soil drainage system, proper weed management practices, adjusting plant density, application of NPS fertilizer, use of recommended planting time, and developing tolerant varieties should be considered in faba bean production. Moreover, integrated FBG disease management strategy should be developed and implemented in the study areas and other related agro-ecologies. 6. Acknowledgements The study was financially supported by Amhara Regional Research Institute (ARARI), Australian Center for International Agricultural Research (ACIAR Project: CIM/2017/030 “Faba Bean in Ethiopia – Mitigating disease constraints to improve productivity and sustainability”) and the International Center for Agricultural Research in the Dry Areas (ICARDA), Africa RISING/USAID through ICARDA. Hence, the institutes are duly acknowledged for supporting the study. This work is part of a PhD Dissertation research requirement at Haramaya University and the University is duly acknowledged. We also greatly acknowledge Mr. Tameru Kibret, Mr. Alemayehu Ayele, Mr. Alemnew Fantaye, and Mr. Semegnew Anleye for their assistance during the survey periods. 102 7. References Agegnehu, G. and Fessehaie, R. 2006. Response of faba bean to phosphate fertilizer and weed control on Nitisols of Ethiopian highlands. Italian Journal of Agronomy, 2: 281–290. Amsalu, N. 2014. Role of faba bean (Vicia faba L.) for intensification of cereal-based cropping systems in the humid highlands of Ethiopia. PhD thesis, Ghent University, Ghent, Belgium, 201 pp. Anteneh, A., Yohannes, E., Mesganaw, G., Solomon, G. and Getachew, T. 2018. Survey of faba bean (Vicia faba L.) diseases in major faba bean growing districts of North Gondar. African Journal of Plant Sciences, 12: 32–36. Beyene, B. 2015. Survey and identification of new faba bean disease (Qormid) in the highlands of North Shewa, Ethiopia. Current Research in Microbiology Biotechnology, 3: 561–563. Bogale, N., Setu, B. and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae Kusano) in major faba bean (Vicia faba L.) growing areas of Northeastern Amhara, Ethiopia. Journal of Agriculture and Environment in International Development, 110: 87–95. CSA (Central Statistics Agency). 2019. Agricultural sample survey (2019). Report on area and production of major crops. Statistical Bulletin 587, Central Statistical Authority, Addis Ababa, Ethiopia. Dereje, G., Wendafrash, M. and Gemechu, K. 2012. Faba bean galls: A new disease of faba bean in Ethiopia. Holetta Agricultural Research Center, Ethiopia. Available at: https://docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE/edit. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean News, 32: 8–10. Dordas, C. 2008. Role of nutrients in controlling plant diseases in sustainable agriculture: a review. Agronomy and Sustainable Development, 28(1): 33–46. Endalew, E., Terefe, H., Dejene, M. and Kumar, A. 2020. Distribution and association of agro‑ecological factors influencing garlic rust (Puccinia allii) epidemics in Eastern Amhara, Ethiopia. Indian Phytopathology. https://doi.org/10.1007/s42360-020-00297-w. FAOSTAT (Food and Agriculture Organization of the United Nations, Statistical Database). 2018. Rome, Italy. Available at: https://faostat.fao.org. 103 Fininsa, C. and Yuen, J. 2001. Association of maize rust and leaf blight epidemics with cropping systems in Hararghe highlands, eastern Ethiopia. Crop Protection, 20: 669–678. Fininsa, C. and Yuen, J. 2002. Temporal progression of bean common bacterial blight (Xanthomonas campestris pv phaseoli) in sole and intercropping systems. European Journal of Plant Pathology, 108: 485–495. Gemechu, K, and Mussa, J. Tezera, W. Getnet, D. 2005. Extent and pattern of genetic diversity for morpho-agronomic traits in Ethiopian highland pulse landraces II. Faba bean (Vicia faba L.). General Research Crop Evolution, 52: 551–561. Getenet, Y, and Yehizbalem, A. 2017. Adaptation of faba bean varieties for yield, for yield components and against faba bean gall (Olpidium viciae Kusano) disease in South Gondar, Ethiopia. The Crop Journal, 5: 560–666. Gezahegn, G., Endale, H., Kassahu, S., Tesfaye, A. and Girma, M. 2018. The causal pathogen, inoculum sources and alternate hosts studies of the newly emerged gall-forming faba bean (Vicia faba) disease in Ethiopia. Advances in Crop Sciences and Technology. 6(3): 368, DOI:10.4172/2329-8863.1000368. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N., Bitew, B., Boydom, A., Kassa, D. and Temesgen, T. 2014. Faba bean gall; a new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Sciences and Technology, 2(4): 1–5, http://dx.doi.org/10.4172/2329– 8863, 1000144. Jensen, E.S., Peoples, M.B. and Hauggaard-Nielsen, H. 2010. Faba bean in cropping systems. Field Crops Researh, 115: 203–216. Kusano, S. 1912. On the life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of College of Agriculture, 4: 142–199. Lay, C.Y. Hamel, C. and St-Arnaud, M. 2018. Taxonomy and pathogenicity of Olpidium brassicae and its allied species. Fungal Biolology, 122: 837–846. Li-juan, L., Zhao-Hai, Y., Zhao-Jie, Z., Ming-Shi, X. and Han-Qing, Y. 1993. Faba bean in China: a state-of-the-art review. International Center for Agricultural Research in the Dry Areas, Aleppo, Syria, pp. 127–128. McCullagh, P. and Nelder, J.A. 1989. Generalized linear models. 2nd Ed. Chapman and Hall, London, 511p. 104 Mengesha, G.G. and Yetayew, H.T. 2018. Distribution and association of factors influencing bean common bacterial blight (Xanthomonas axonopodis pv. phaseoli) epidemics in southern Ethiopia. Archives of Phytopathology and Plant Protection, 51(20): 1066–1089. Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. 48 pp. Newton, A.C., Duncan, J.M., Augustin, N.H., Guy, F.D.C. and Cooke, D.E.L. 2010. Survival, distribution and genetic variability of inoculum of the strawberry red core pathogen, Phytophthora fragariae var. fragariae, in soil.Plant Pathology, 59(3): 472–479. Oborn, I., Edwards, A.C., Witter, E., Oenema, O., Ivarsson, K., Withers., P.J.A., Nilsson, S.I. and Stinzing, R.A. 2003. Element balances as a toll for sustainable nutrient management: a critical appraisal of their merits and limitations within an agronomic and environmental context. European Journal of Agronomy, 20:211–225. Pande, S. and Narayana, R.J. 2002. Effect of plant population densities on the severity of late leaf spot and rust of groundnut. The Plant Pathology Journal, 18(5): 271–278. Przetakiewicz, J. 2014. Synchytrium endobioticum. pp.823–829. In: Liu D (Ed.). Manual of security sensitive microbes and toxins: CRC Press, Boca Raton, USA. SAS (Statistical Analysis System). 2014. SAS User’s Guide. Version 9.4.2. SAS Institute Inc., Cary, NC, USA. Tadesse, N,, Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C. and AbrahamA. 2008. Review of research on diseases of food legumes. pp. 85–132. In: Abraham T (Ed.). Increasing Crop Production through Improved Plant Protection, Addis Ababa, Ethiopia: PPSE and EIAR, Addis Ababa, Ethiopia, 598 pp. Teklay, A., Tsegaye, B., Yemane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discource Journal of Agriculture and Food Science, 2: 33–38. Teferi, T.A., Weldemichael, G.B., Wakeyo, G.K. and Mindaye, T.T. 2018. Fungicidal management of the newly emerging faba bean disease “gall” (Olpidium viciae Kusano) in Tigray, Ethiopia. Crop Protection, 107: 19–25. 105 Terefe, H. and Gudero, G. 2019. Distribution and association of factors influencing emerging maize lethal necrosis disease epidemics in southern Ethiopia, Cogent Food and Agriculture. 5:1, https://doi.org/10.1080/23311932.2019.1683950. Veresoglou, S.D., Barto, E.K., Menexes, G. and Rillig, M.C. 2013. Fertilization affects severity of disease caused by fungal plant pathogens. Plant Pathology, 62(5): 961–969. Wakweya, K., Dargie, R. and Meleta, T. 2016. Effect of sowing date and seed rate on faba bean (Vicia faba L.) growth, yield and components of yield at Sinana, highland conditions of Bale, southeastern Ethiopia. International Journal of Science Research and Agricultural Sciences, 3: 25–34. Wheeler, B.E.J. 1969. An Introduction to Plant Diseases. Wiley and Sons, London. 374 pp. Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) using varieties and fungicides in North Shoa, central Ethiopia. MSc Thesis, Haramaya University, Haramaya, Ethiopia. Xing, Z.S. 1984. Preliminary study oiler bacteria blister prevention and treatment of disease blister beans. Plant Pathology, 114: 165–173. Yan, J.M. 2012. Study on blister disease of broad bean caused by Olpidium viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. Yan, J.M. and Huazhi,Y. 2012. On the occurrence regularity of broad bean blister. Journal of Sichuan Agricultural University, 30: 365–370. Yimer, S.M., Ahmed, S., Fininsa, C., Tadesse. N., Hamwieh, A. and Cook, D.R. 2018. Distribution and factors influencing chickpea wilt and root rot epidemics in Ethiopia. Crop Protection, 106: 150–155. Yohannes, D. 2000. Faba bean (Vicia faba L.) in Ethiopia. Institute of Biodiversity, Conservation and Research, Addis Ababa, Ethiopia, 43 pp. You, M.P., Eshete, B.B., Kemal, S.A., Van,Leur, J. and Barbetti, M.J. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology. https://doi.org/10.1111. Yuen, J. 2006. Deriving decision rules. The plant health instructor. Department of Forest Mycology and Pathology. Swedish University of Agricultural Sciences. 43 pp. 106 Paper II Physoderma, not Olpidium, is the True Cause of Faba Bean Gall Disease of Vicia faba in Ethiopia Ming Pei You1, Beyene Bitew Eshete2, Seid Ahmed3, Joop van Leur4, Martin J. Barbetti1* 1School of Plant Biology, Faculty of Science, the University of Western Australia, Australia 2Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 3Biodiversity and Crop Improvement Program, International Center for Agricultural Research in the Dry Areas (ICARDA), B.P. 6299, Rabat, Morocco 4New South Wales Department of Primary Industries, Tamworth Agricultural Institute, Tamworth, NSW, Australia Abstract Faba bean (Vicia faba L.) is the major food legume pulse crop grown in Ethiopia. However, the productivity is highly inhibited by faba bean gall disease. In Ethiopia, faba bean gall (FBG) is a devastating faba bean disease, which was previously assumed to be caused by Olpidim viciae, but not yet confirmed in the country. A study was conducted at Debre Biran Agricultural Research Center, International Livestock Research Institute and University of Western Australia during 2018-2020 to identity the FBG disease causal agent, through molecular techniques. Microscopic examination of specimen (infected faba bean leaves and stems) was collected from Bassona Worana, Debark, Debay Telatgen, Farta, Dessie Zuria, Degam and Chole districts. and confirmed an epibiotic phase of zoosporangia for dispersing zoospores, characteristic of Physoderma. The morphology did not show critical diagnostic characteristics, such as presence of numerous short zoosporangial discharging tubes, or binucleate resting sporangia, of Olpidium viciae. DNA extraction, DNA amplification using PCR, DNA agarose gel electrophoresis, sequencing and bioinformatics analysis were conducted. Sequences derived from symptomatic tissue from partial ITS1-5.8S-partial ITS2, from 18S-ITS1-5.8S-ITS2-part of 28S, and LSU (S28) all confirmed that Physoderma is the causal agent of FBG disease. From symptom, morphological and molecular perspectives, the causal agent of FBG disease is Physoderma, but not Olpidium viciae. Keywords: Binucleate Resting Sporangia, Diagnosis, Epibiotic Phase, Molecular Techniques, Morphology, Zoosporangia, Zoospores. 107 1. Introduction Faba bean (Vicia faba L.) is the major food and feed legume crop grown in many countries in the world (Torres et al., 2006). In Ethiopia, faba bean constitutes nearly 0.5 M ha (CSA, 2019). Faba bean has high nutritional value and, thus, it is a rich available source of food for human beings and feed for animals and it is used as an excellent component of crop rotation and green manure to improve soil fertility. However, the growing environment in Ethiopia for faba bean cropping is highly conducive to a wide range of plant diseases, particularly in high rainfall years that are common across the highland faba bean production areas. Different diseases attack faba bean in Ethiopia. The most important diseases of faba bean include chocolate spot (Botrytis fabae), Ascochyta blight (Ascochyta fabae, syn. Didymella fabae), rust (Uromyces viciae-fabae), black root rot (Fusarium solani and F. avenaceum) and viruses, like Faba Bean Necrotic Yellows Virus (FBNYV). Each of these diseases can reach severities that severely impact yield and, in some cases, can cause complete crop failure (Abraham et al., 2000; Tadesse et al., 2008). Among fungal diseases, chocolate spot was considered the most destructive faba bean disease in Ethiopia (Sahile et al., 2008). However, in 2010, a previously unknown disease, faba bean gall (FBG) was reported causing galling and distortion of faba bean foliage in North Shoa Region, Ethiopia (Dereje et al., 2012; Bitew, 2015). It then spread rapidly throughout farmers’ fields in the central highlands and became increasingly severe (Hailu et al., 2014). Surveys highlighted the rapid spread and devastating impact of FBG across the different regions of Ethiopia in subsequent years. Surveys conducted in 2012 and 2013 in the highlands of Wollo, Northern Ethiopia, and showed FBG incidence of 50-100% and severity of 10-59% (Hailemariam et al., 2016). Surveys by Abebe et al. (2014) showed severity ranging from 30 to 100% in South Tigray Zone. Hailu et al. (2014) demonstrated FBG severity in the Amhara Region 22%, but were lower at that time in Tigray and Oromiya Regions with 11 and 8% severities, respectively. Bitew (2015) highlighted the importance of FBG disease in surveys across the nine different districts of North Shoa. Further, Debela et al. (2017) reported FBG disease in the western Highlands of Oromiya Region with severity ranging from 4 to 80% and with yield losses up to 100% in some crops. Similarly, Debela and Tola (2017) also highlighted up to 100% yield losses from FBG disease and as particularly occur across the higher altitudes of 2000 to 4000 m.a.s.l. where the rainfall is greatest (Bitew and Wondosen, 2012; Abebe et al., 2014). Not only did FBG disease 108 quickly become established across all main faba bean-growing regions of Ethiopia, but also at severities surpassing all other diseases (Hailu et al., 2014). Farmers are using fungicide sprays to manage the disease, but the requirement of frequent applications prohibits economically poor farmers to save their crops from FBG disease (Teferi et al., 2018). Olpidium viciae (Kusano, 1912), belonging to the family Olpidiaceae, was accepted as the causal agent of FBG disease in Ethiopia (IFPRI, 2010; Dereje et al., 2012), the same pathogen first reported in Japan in 1912 on Vicia unijugae (Kusano, 1912) and reported subsequently as the cause of ‘blister disease’ of faba bean in China by Liang (1986), Lijuan et al., (2003) and Yan (2012). Identification in Ethiopia was based on similarity of symptoms (Earecho, 2019) and observation of resting spores within the galls of affected leaves, as reported in China through morphological studies by different researchers (Xing, 1984; Liang, 1986; Lijuan et al., 2003; Yan, 2012), whereas there have been several attempts to molecularly confirm the identity of FBG pathogen, since the morphological methods alone had not been successful, only showing similarities with non-related pathogens, such as Phoma, Peyronellaea pinodella (syn. Mycosphaerella pinodes, Didymella pinodes), Aureobasidium/Kabatiella lini; Cryptococcus victoriaeor and Albugo laibachii, none of which are known to produce the typical FBG disease symptoms on faba bean. But the molecular investigations on isolates collected from Ethiopia for the FBG pathogen were in contrast to those from China, where Yan (2012) used primer pair ITS1 and ITS4 to amplify the ITS region of the pathogen and confirmed the identity of faba bean blister causal pathogen in China as O. viciae according to pathogen sequence. This would be the first ever Olpidium species to operate via foliar disease cycles rather than root infection. Hence, the current study was carried out to identity the FBG pathogen causal agent through molecular studies or techniques. 2. Materials and Methods 2.1. Disease Symptoms and Pathogen Morphology Faba bean leaves and stems infected by faba bean gall disease were collected from different districts of Amhara and Oromia Regional States. The morphological and molecular identifications were conducted in Ethiopia at Debre Birhan Agricultural Research Center and International Livestock Research Institute, Addis Ababa, Ethiopia, and University of Western 109 Australia, Australia. In relation to FBG field symptoms and pathogen morphology, investigations were undertaken in Ethiopia, in September 2019 in the Debre Birhan Agricultural Research Centre Laboratory and later in March 2020 in the International Livestock Research Institute Laboratories in Addis Ababa, Ethiopia. The observations were made on hand-sectioned wet mounts of infected and healthy faba bean leaves and stems using an Olympus CX33 microscope, with images captured using an Olympus EP50 digital photographic system. 2.2. Collection of FBG DNA Samples A total of 187 infected plants showing FBG disease symptoms were collected from Amhara and Oromia Regional States (Ankober, Asagert, Bassona Worana, Chole, Debark, Debay Telatgen, Degem, Dessie Zuria, Farta, Laygayint, Meket, Sinan and Tarmaber districts), and crude extracts of DNA from pathogen-infected materials in Ethiopia were applied onto Whatman® FTA® Cards. Field sampling at each location was undertaken from two selected plants showing gall symptoms. When returned to the laboratory, for each single plant sample brought back from the field, an area of galled fresh tissue, mostly on leaves and stem, were excised. Diseased tissue surfaces were first wiped with 70% ethanol for initial sterilization. The fresh galled pieces were placed into a 1.5 mL Eppendorf tube until the tube was approximately half-filled, with a separate Eppendorf tube for each sample. Faba bean gall samples were well macerated in the tubes using a plastic pestle. The upper and lower epidermal layers of infected leaf and stem tissues were separately and carefully peeled off by hand and treated similarly. Then macerated gall extract samples were placed onto separate FTA® Cards and air-dried. The gall samples were first processed in Ethiopia to extract DNA as described in Section 2.3 below, and extracted DNA was applied to FTA® Cards. Impregnated FTA® Cards were exported out of Ethiopia under an Ethiopian Biodiversity Institute Material Export Permit (Reference No.EBI 71/2553/2011). Impregnated FTA® Cards were then taken into Australia under an Australian Government, Department of Agriculture and Water Resources Import Permit specifically for this purpose (Permit No. 0002826465). 110 2.3. Deoxyribonucleic acid (DNA) extraction and PCR conditions In Australia, DNA was extracted from FTA® Card’s carrying pathogen DNA by cutting 2-3 pieces each 2 mm square into a 2 mL Eppendorf tube and washing to remove inhibitor using a modified method of Ahmed et al. (2011). Briefly FTA® Card pieces in 2 mL tubes were washed twice in 200 μL of TENT (10 mM Tris-Cl, pH 8.0, 1 mM EDTA, 12 mM NaCl, 2.5% Triton X- 100) and incubated for 5 minutes each time with gentle agitation at 1000 per minute on an agitator (TTS 3 digital Yellow line). The FTA® Cards were then washed twice in 200 μL in TE 0.1 (10 mM Tris-Cl, pH 8.0, 0.1 mM EDTA) buffer and agitated for 5 minutes each time at 1000 per minute and supernatant removed each time. The FTA® Card pieces then were left to dry at room temperature (laminar flow for 2 hrs). Dried FTA® Card pieces were then ready for DNA extraction undertaken as follows. Two ceramic beads (2 mm) and 300 µL extraction buffer [200 mM Tris-HCl (pH 8.5), 250 mM NaCl, 25 mM EDTA], 0.5% (wt/vol) sodium dodecyl sulphate (SDS)] were added together into the tube and homogenized using a Precellys Evolution homogenizer at 10,000 rpm for 3x60 second cycles, each cycle with a 30 second pause. Then, 150 μL of 3 M sodium acetate (pH 5.2) was added and mixed well by pipetting, left at room temperature for 10 min, then centrifuged at 13,000 rpm for 10 min. Subsequently, supernatant was transferred into a fresh tube with an equal volume of prior-added cold isopropanol (450 µL), mixed well by pipetting and then left to stand for 15 minutes at room temperature (22 °C) or in a fridge overnight for best result. Precipitated DNA was collected by centrifugation at 13,000 rpm for 10 minutes and supernatant pipetted out. Pellet was washed with 70% (vol/vol, ethanol/water) and dried (air dry or vacuum dry), then re-suspended in 50 μL of TE buffer. Quantity and the quality of extracted DNA were determined with a Nano Drop 1000 Spectrophotometer (Thermo Scientific). The DNA was stored at 4 °C. Then the DNA was subjected to PCR using a master mix of a total volume of 25 µL that contained 0.5 µM of each primer (primers used and PCR conditions are listed in Table 1). PCR products were subjected to agarose gel electrophoresis at 60 mV for ≥1 hr (dependent on size of the PCR products) on a 1% (w/v) agarose gel containing 0.1% GelRed™ Biotium Inc. (United States) and then visualized under UV light. Polymerase chain reaction (PCR) products were then sequenced by outsourcing via Macrogen Inc. (Korea). 111 2.4. Primer Pairs Primer pairs (Table 1) were used to amplify small subunit (SSU) (James et al., 2006); internal transcribed spacer (ITS) region (ITS1 to ITS4) (White et al., 1990); and nrLSU (Vilgalys and Hester, 1990; Rehner and Samuels, 1994; Stielow et al., 2015). Primers EF and RPB2 were also tried, but without success. Initially assuming the pathogen was O. viciae, general Olpidium specific reverse primers (Herrera-Vásquez et al., 2009) for amplifying ITS region were also used in combination with ITS1 (and with the specific primer pair OLPv4F and OLPv4R developed in the current study) for O. viciae according to the sequence of O. viciae available in GenBank (HQ677595) as had earlier been submitted by Yan et al. (2012) (Table 1). 2.5. Phylogenic Tree Building The sequenced samples DNA were aligned using “Geneious Prime version 2020.03” and then resulting consensus were BLAST in Genbank (NCBI –NIH). If percentage identity was low (≤97%) and in some cases higher, but low coverage (≤85%), the most frequent, closest and likely-matching pathogens from Genbank were selected as references for building phylogenic trees, again using “Geneious Prime version 2020.03”. Sequences of morphologically similar and taxonomically-related pathogens Olpidium, Physoderma, Synchytrium and Urophlyctis were also selected from Genbank and included in building phylogenic trees for comparison. Three phylogenetic trees were built according to amplification regions. Sequences from ITS1- 5.8S-ITS2 formed one tree (Figure 3), sequences from partial 18S-ITS1-5.8S-ITS2-partial 28S formed another tree (Figure 4) and sequences from LSU (28S) formed a third tree (Figure 5). Pairwise sequence alignment [Alignment Type: global alignment with free end gaps; cost matrix: 65% similarity (5.0/-4.0)] was used with Tamura-Nei for Genetic Distance Model and Neighbor- Joining for Tree Building. 112 Table 1. Primers and polymerase chain reaction (PCR) conditions used in this study Gene/ primer Sequence(5’-3’) Initial Duratio Denatura Durati Annealin Duratio Elong. Duratio cycle Extensio Duration locus T(0c) n tion T on (s) g T(oc) n (s) T(oc) n (s) n T(oc) (min) amplified (min) (oc) ITS ITS1 TCCGTAGGTGAACCTGCGG 94 1 94 30 60 30 72 60 35 72 10 ITS4 TCCTCCGCTTATTGATATGC ITS1 TCCGTAGGTGAACCTGCGC 94 5 94 45 55 60 72 60 35 72 10 OLPR TCCTCCGCTTATTGATATGCTTA OLPvic4F ATCGATGAAGAACGCAGCGA 95 5 95 45 55 40 72 45 35 72 10 OLPvic4R TTCAGCGGGTATCCCTACCT TWB1 GTTTCCGTAGGTGAACCTGC 94 2 94 60 55 60 72 90 35 72 10 OLPvic4R TTCAGCGGGTATCCCTACCT SSU rRNA SR1-R TACCTGGTTGATTCTGC 95 2 94 60 55 60 72 210 38 72 10 NSR CTTCCGTCAATTCCTTTAAG SR1-R TACCTGGTTGATTCTGC 95 2 94 60 55 60 72 210 38 72 10 SR6 TGTTACGACTTTTACTT SR1-R TACCTGGTTGATTCTGC 94 1 94 30 55 30 72 300 35 72 10 LR12 GACTTAGAGGCGTTCAG 18S-Cs-1F GAGGCCTACCATGGTGAT 95 2 94 60 55 60 72 210 38 72 10 NS6 GCATCACAGACCTGTTATTGCCTC LSU LROR ACCCGCTGAACTTAAGC 95 2 94 60 55 60 72 60 38 72 10 LRS TCCTGAGGGAAACTTCG RBP1 RBP1af GAATGTCCAGGACATTTCGG 94 1 94 30 52 30 72 90 38 72 10 RBP1cr CCTGCAATTTCATTATCCATGTA RBP2a RBP2-5f GATGATAGAGATCACTTTGG 94 5 95 30 55 60 72 60 35 72 7 RBP2-7cr CCCATAGCTTGTTTACCCAT RBP2-7cF ATGGGTAAGCAAGCCATGGG TEF1αb EF1- GACTTCATCAAGAACATGAT 94 5 94 30 48 50 72 50 40 72 7 1018F(AI33F) EF1- GACGTTGAATCCAACATTGTC 1620R(AI33R) EF1- TTCATCAAGAACATGAT 94 5 94 20 48 50 72 50 40 72 7 1002F(AI34F) EF1- CTATCATCACAATGGACGTTCTTG 1688R(AI34R) GAG ITS, internal transcribed spacer region; SSU rRNA, small subunit rRNA gene; LSU, nuclear large ribosomal subunit gene; RPB1and RPB2, DNA- directed RNA polymerase II subunit genes; TEF1α, translation elongation factor 1α. aRPB2- 5f can be used with RPB2- 7cr. bIf not successful then repeated using (Al34F/R) 113 3. Results 3.1. Disease Symptoms and Pathogen Morphology and Histology Field and plant host symptoms of FBG disease is presented hereunder (Figure 1 a– k). Figure 1. Field and plant host symptoms for FBG disease (a–g) showing poor plant growth and major foliage death. A severely infected farm field (a); impact of disease demonstrated by fungicide control on left and no fungicides on right (b); "sunken- well" indentations on upper leaf surface (arrows), within which zoosporangia quickly develop and release zoospores enhancing their dispersal (c); areas of bulging (galling) of lower leaf surface (arrows) in early symptom development (d), and darkening of such tissues as symptoms develop into more pronounced galls over time (arrows) (e, f); typical galling on stem (g). Symptoms (arrowed) were also found, to a much lesser degree, on field pea sown together with faba bean (h), and on two different unidentified clover species growing in close proximity to infected faba bean crops (i, j), sunken and gall (k). When histological studies were conducted, a range of morphological symptoms were observed that included dark brown regions on the underside of the leaf gall seen in typical leaf cross sections (Figure 2a); here, large masses of resting spores were produced (Figure 2b) and typically, masses of resting spores in the end-of-season dried residues (Figure 2c). We also observed turbinate cells (Sparrow et al., 1961) with two segments (Figure 2d) and zoosporangia with openings, inside which zoospore masses developed (Figure 2e– f). Importantly, in our leaf sections, we consistently observed the epibiotic phase of zoosporangia for dispersing zoospores (Figure 2e, f). However, a thorough search of sections of affected leaves did not reveal either the presence of numerous short zoosporangia discharging tubes or the binucleate resting sporangia characteristic of Olpidium spp. 114 Figure 2. Typical cross section of a faba bean leaf with symptoms of faba bean gall disease: Lesions showing dark brown regions (arrows) on underside of leaf gall (a); in which large masses of resting spores are produced (arrow) (b); and typical masses of resting spores in end- of- season dried residues (c). Turbinate cells with two segments were also observed (d). Zoosporangium with opening (e, arrow i) and base of zoosporangium (e, arrow ii) and Zoosporangia develop internal zoospores masses (arrow) (f), zoospore and zoosporangium (g). 3.2. Sequence BLAST Results A total of 173 sequences obtained, 95 from ITS, 55 from LSU, 20 from SSU, 1 from TEF1α and 2 from RPB2. Nearly 40% of all sequences were close to Didymella, with percentage identity from 83 to 99% and coverage from 27 to 99% (Table 2). Sequences of 11 isolates from the 2020 collection gave a result of “uncultured fungus” and there was a high similarity between them for the LSU sequences. Within sequences from the ITS region, approximately 39% were close to the genus Didymella, 20% close to Phoma, 14% close to Mycosphaerella, and 19% close to the host Vicia. Within sequences from the LSU gene, approximately 53% were close to the genus Didymella, 20% were close to “uncultured fungus”, 5% close to Olpidium, and 4% that were close to each of Cladosporium, Cymadothea, “no significant result obtained”. For the SSU gene, 30% were close to Leucosporidium, 10% from two samples were close to Phoma, 5% (one sample) each were close to Boeremia, Cryptococcus, Filobasidium, and Phaseoleae (plant) and 35% close to the host plant Vicia (Table 2). A large portion of sequences identified as genera Didymella, Olpidium, and “uncultured fungus” included those showing a low percentage query coverage and identity from either or both the ITS and LSU locus/gene. Genera sequenced from SSU showed only small variations both in percentage query coverage and in identity. None of the genera identified from BLAST results with above 97% identity have previously been recorded as causing a “faba bean blister” type symptom. Despite being recorded as either producing a possible symptom with some similarity to “faba bean blister” and/or showing at least some morphological similarity, we did not find Physoderma, Synchytrium, or Urophlyctis in these BLAST results. 115 3.3. Phylogenetic Analysis The phylogenetic tree constructed from the ITS1- 5.8S- ITS2 sequences (38 test sample sequences) obtained in this study and from related isolates from the NCBI database showed that all test isolates formed a group (group 3) with Physoderma maydis (HB683909). Olpidium viciae (HQ677595) from China was close to two Didymella isolates from NCBI, forming group 1. Another four Physoderma isolates and one Synchytrium from NCBI formed group 2 (Figure 3). When sequences from partial 18S- ITS1- 5.8S- ITS2- partial 28S (27 test sample sequences) were used to construct a phylogenetic tree, one test sequence was close to Physoderma in group 1, three test sequences in group 3 were close to Physoderma, 22 test sequences were close to Didymella in group 6, group 5 included Physoderma and Synchytrium, and group 4 included Olpidium from NCBI (Figure 4). In the phylogenetic tree constructed from the LSU (28S rRNA) sequences, the 47 test sequences formed four groups, with 13 sequences grouped with Physoderma (group 2); 11 sequences grouped with Physoderma in group 3, two sequences were close to Cymadothea in group 4, 21 sequences grouped with Didymella in group 5, and group 1 was formed by Olpidium sequences from NCBI (Figure 5). Sample sequences that grouped with Physoderma have been deposited in GenBank (accession numbers from MW414613 to MW414631, from MW448404 to 448414, from MW497579 to MW497587, and from MW587325 to MW587329). 116 Table 2. Genera obtained from BLAST search results using successfully sequenced genes/loci from isolates obtained from plants with faba bean gall disease. Gene/locus Percentage of successful isolates per genus Genus ITS LSU SSU TEF1a RPB2 Total Within Within Within (%) ITS (%) LSU (%) SSU (%) Abrothrix 0 1 0 0 0 0.6 0 1.8 0 Ascochyta 1 0 0 0 0 0.6 1.0 0 0 Aspergillus 1 0 0 0 0 0.6 1.0 0 0 Boeremia 0 0 1 0 0 0.6 0 0 5 Cladosporium 0 2 0 0 0 1.2 0 3.6 0 Coniothyrium 1 0 0 0 0 0.6 1.0 0 0 Cryptococcus 1 1 1 0 0 1.7 1.0 1.8 5 Cyamadothea 0 2 0 0 0 1.2 0 3.6 0 Didymella 37 29 0 0 0 38.2 39.0 52.7 0 Didymosphaeria 1 0 0 0 0 0.6 1.0 0 0 Epicoccum 1 0 0 0 0 0.6 1.0 0 0 Filobasidium 0 0 1 0 0 0.6 0 0 5 Leucosporidium 0 0 6 0 0 3.5 0 0 30 Mycosphaerella 13 0 0 0 0 7.5 13.7 0 0 No significant 1 2 0 0 0 2.3 1.0 3.6 0 Olpidium 0 3 0 0 1 1.7 0 5.4 0 Penicillium 0 1 0 1 0 1.7 0 1.8 0 Phaseoleae 0 0 1 0 1 0.6 0 0 5 Phoma 19 1 2 0 0 12.7 20.0 1.8 10 Uncultured 1 11 1 0 0 7.5 1.0 20.0 5 Vicia 18 2 7 0 0 15.6 19.0 3.6 35 Total 95 55 20 1 2 100 100 100 100 aSequences from all gene regions considered together 117 Figure 3. Phylogram based on sequences of the ITS1- 5.8S- ITS2 region, showing the relationship between isolates obtained from plants with faba bean gall disease in the current study and other related isolates from the NCBI database. Reference isolates from NCBI are shown in bold type. 118 Figure 4. Phylogram based on sequences from partial 18S- ITS1- 5.8S- ITS2- partial 28S, showing the relationship between isolates obtained from plants with faba bean gall disease in the current study and other related isolates from the NCBI database. Reference isolates from NCBI are shown in bold type. 119 Figure 5. Phylogram based on sequences from the large subunit rRNA gene (LSU[28S rRNA]) showing the relationship between isolates obtained from plants with faba bean gall disease in the current study and other related isolates from the NCBI database. Reference isolates from NCBI are shown in bold type. 120 4. Discussion The FBG disease symptom development both in Ethiopia (Hailu et al., 2014; Bitew, 2015; Bogale et al., 2016) and China (Yan, 2012) are similar that showed typical galling of leaves and stems and browning of the infected tissues over time. In the current study, the upper leaf surfaces of infected leaves showed distinct sunken-well cupping (puckering) indentations before any leaf browning occurs. At the margins of these indentations, prolific zoosporangia formation and releasing of zoospores was observed. The indentations naturally retain moisture, fostering multiple cycles of zoosporangia production and zoospore dispersal by rain-splash to rapidly progress the disease epidemic under conducive environments. This was quickly followed by production of masses of sporangia observed, particularly on the undersides of leaves. In the current studies, there were a number of morphological characteristics observed that either supported an identification of the causal pathogen as Physoderma or were contrary to an identification of it as Olpidium. First, in the leaf structure sections, the epibiotic phase of zoosporangia for dispersing zoospores, a diagnostic characteristic of Physoderma was observed, but the the endobiotic zoosporangia, characteristic of Olpidium (Johns, 1966; Gould and Schaechter, 2009) was not observed. Physoderma has two phases (epibiotic monocentric phase and an endobiotic polycentric phase) (Johns, 1966). Second, we did not observe the presence of numerous short zoosporangia discharge tubes or the binucleate resting sporangia so characteristic of Olpidium (Hiruki and Alderson, 2011), or the much wider variation in size and frequency of discharge tubes as it has been reported in Olpidium (Garrett and Tomlinson, 1967) compared to Physoderma. Thirdly, while we observed masses of large resting spores, it appears that the true identity as Physoderma was missed from mistakenly assuming that the resting spores observed in FBG disease belonged to the holocarpic genus Olpidium, particularly when observed at the advanced stages of infections where the rhizomycelial elements would not have been apparent (Johns, 1966). The morphological studies did not support any consensus that the causal pathogen could be an Olpidium, the symptoms and presence of resting spores in leaves clearly indicates it to be a Chytridiomycete pathogen. Chytridiomycetes are pathogens associated with aquatic habitats, the flagellated stage (zoospores), with plants generally reacting to infection by forming gall structures around the zoosporangia and with zoospore dissemination depending on flooding or heavy rainfall. Chrytridiomycete genera, while often misidentified and reclassified, contain some obligate plant 121 pathogens of world-wide importance (Synchytrium endobioticum, potato wart disease), with resting spores persisting and remaining infective in soil for very long periods (Potato wart disease >40 years), and as assumed for the FBG pathogen surviving on infected residues in soil in China (Lang et al.,1993). Plant pathogens Olpidium, Physoderma and Synchytrium were all previously classified as Chytridiomycetes (Alexopoulos, 1962; Agrios, 2005) and all three genera produce thick-walled resting spores and/or zoosporangia, with typical infection caused by zoospores produced from zoosporangia (within growing season) and from carryover of resting spores between seasons. However, only Physoderma and Synchytrium cause plant diseases above ground, generally as a gall symptom that occurs from cells in affected tissues being stimulated to divide repeatedly and enlarge excessively (Agrios, 2005), and as observed in the current study. In contrast, Olpidium spp. are obligate plant root pathogens found commonly throughout the world infecting the roots of wild and domesticated plants and not foliage (Maccarone et al., 2010). Further, the FBG pathogen in Ethiopia and elsewhere attacks host stems and leaves, but is not known to cause disease on roots (Alexopoulos, 1962; Kusano, 1912; Lay et al., 2018). There are only three Olpidium spp. recognized plant pathogenic because of their ability to vector viruses, the first two being O. brassicae, O. bornovanus (=O. radicale) (Campbell, 1985, 1996; Rochon et al., 2004; Herrera-Vasquez et al., 2010). However, Sahtiyanci (1962) had earlier separated O. brassicae (formerly Pleotrachelus brassicae) into two species (O. brassicae being crucifer- infecting and O. virulentus non-crucifer-infecting). All three of these Olpidium spp. show morphological features clearly distinct from those of Physoderma. In the current molecular identity studies, phylograms were utilized for further investigation. Because multiple sequence alignment (MSA) resulted in high likelihood topology for the isolates, we used pairwise sequence alignment as it best generates optimal alignment (Xia, 2016). An interesting group from LSU matching NCBI isolates, “uncultured fungi”, showed very high similarity between them and grouped with the Physoderma isolate from NCBI in group 3. Genera resulting from analysis of SSU sequences were of high percentage identity with isolates in NCBI and were, therefore, considered true genera; however, these genera were clearly not related to the disease symptom or pathogen morphology. The genera Mycosphaerella and Phoma obtained from ITS sequences and Phoma also from SSU sequences, matched with NCBI isolates with a high level of percentage identity and query coverage (>97% percentage identity and >93% query coverage) and, therefore, were assumed to be 122 true matches (Hibbett et al., 2016). All the above mentioned non-Olpidium and non-Physoderma genera were also isolated and identified by sequencing in our preliminary investigations, including our initial samplings (data not shown). Olpidium belongs to the phylum Zygomycota, while Physoderma belong to the phylum Blastocladiomycota (James et al., 2006; Money et al., 2016). In our phylograms, Physoderma were grouped with Synchytrium in group 2, while Olpidium were further away in the ITS region (Figure 3) and also the same applied in partial 18S-ITS1-5.8S-ITS2-partial 28S in groups 2, 4 and 5 of the tree constructed from the partial 18S- ITS1- 5.8S- ITS2- partial 28S sequences. The findings of the current study are in contrast to those on spring-sown faba beans in high altitude regions of south- west China, where Yan (2012) used primer pair ITS1 and ITS4 to amplify the ITS region of the pathogen and concluded that the causal pathogen of faba bean blister in China was O. viciae. In the present study, O. viciae (HQ677595) from China did not group with any other Olpidium from NCBI, but instead grouped with Didymella in group 1, while sequences of our test isolates were grouped in group 3 with Physoderma. It is clear that this isolate of O. viciae (HQ677595) from China is neither close to Olpidium nor Physoderma, but is close to Didymella. The partial 18S- ITS1- 5.8S- ITS2- partial 28S sequences amplified from our test isolates placed most of the isolates with Didymella in group 5. However, critically, the isolate sequence in group 1 and three isolates in group 3 grouped with Physoderma; in contrast, all Olpidium isolates from NCBI were grouped in groups 2 and 4. The phylogram constructed from LSU sequences showed the test isolates distributed through three main groups (2, 3, and 5), where groups 2 and 3 included Physoderma from NCBI; isolates were quite diverse within group 2, but group 3 included isolates with very small differences that constituted the “uncultured fungi” group; and isolates in group 5 grouped with Didymella from NCBI. Group 1 was Olpidium isolates from NCBI and one of the isolates (2Trl) in group 4 was from Trifolium. From our three phylograms, FBG isolates mostly belonged to two groups, Physoderma or Didymella. Physoderma is clearly the main causal pathogen of faba bean gall, while Didymella is present as an accompanying pathogen and/or from secondary infection. Didymella, Mycosphaerella, and Phoma can survive saprophytically, and Didymella was present in the faba bean fields causing blight (Ascochyta fabae, anamorphic; syn. Didymella fabae, teleomorphic state). There is no available comparative sequence data for Physoderma as a legume pathogen. Some other obligate pathogens, such as those causing downy mildew and white rust diseases, produce an abundance of spores on the plant surface, making it easier to identify them morphologically and to extract DNA from spores for molecular identification. 123 However, Physoderma, while it produces an abundance of resting spores inside the host, only produces epibiotic zoosporangia to release zoospores for a short period. This characteristic feature, not only makes identification more difficult, but also allows secondary fungal pathogens to contaminate morphological and molecular identification procedures; this has led previous attempts to determine the causal agent of FBG to identify either secondary pathogens or other contaminating organisms. This is illustrated by two previous unsuccessful attempts to identify molecularly the causal agent of FBG disease. The first was an investigation by CABI in 2012, where ITS rDNA analysis with FASTA showed >99% similarity to sequences assigned to Phoma and Peyronellaea, with 100% match to Peyronellaea pinodella (syn. Mycosphaerella pinodes, Didymella pinodes), with a strong match with Aureobasidium/Kabatiella lini; they also used ITS rDNA analysis with BLAST to highlight similarity with Cryptococcus victoria (CABI, 2012). The second was in 2016, with FBG samples sent from Ambo University to Wageningen Plant Research International, where next generation sequencing analysis showed a low homology, but best fit to Albugo laibachii (Wageningen Plant Research International, 2016). However, neither of these organisms could possibly be the cause of FBG disease as both have significantly different morphology from that observed for FBG. For example, Albugo species are generally not known to infect Fabaceae and produce different symptoms to FBG, and Olpidium and Albugo are unrelated genera despite both being favored by cool and wet environmental conditions. Thus, PCR has limitations as a means of obtaining sequences for identification, but it is an important method for confirmation of morphological observations. The genus Physoderma has been reported on faba bean and other legume hosts. In Japan, Physoderma fabae Syd (1928) was reported on V. faba in 1927, but not with a gall symptom, but rather causing rusty to reddish-brown orbicular or irregular spots on leaves (Watson, 1971). Subsequently, again in Japan, Physoderma leproides (Trab) Lagerh, (1950) [Basionym: Entyloma leproideum Trab. (1894)] was reported as parasitic in leaves and stems of V. faba, again without a gall symptom, causing rusty to reddish-brown orbicular or irregular spots (Watson, 1971). Outside of Ethiopia, Physoderma is reported across a wide range of different legume hosts. For example, P. trifolii (syn. S. trifolii, O. trifolii, Urophlyctis trifolii) has been reported in China on Astragalus sinicus (Milk vetch); in Australia on Swainsona occidentalis, Trifolium glomeratum, T. subterraneum, T. tomentosum, T. repens; and in India on T. alexandrium, T. resupinatum, T. carolinianum, T. medium, T. montanum, T. pratense, T. repens, T. resupinatum (Watson, 1951; Allison et al., 1952; Anonymous, 1960; Butler and Hall, 1966; French, 1989; Shivas, 1989; Pande and Rao, 1998; Cunnington, 2003; Fajardo et al., 124 2017). Such infections by P. trifolii are relevant for Ethiopia, as it is clear that the FBG pathogen not only additionally infects field pea, but also found on two different Trifolium spp. and smartweed. A significant number of Trifolium species occur in Ethiopia, including T. baccarinii, T. fragiferum, T. polystachyum, T. pratense, T. repens, T. semipilosum, T. simense, T. subterraneum, T. tembense, and T. usambarense (ILRI Genetic Resources, unpublished data). That the FBG pathogen can cross over between different host genera and species increases the biosecurity risk of accidental introduction of FBG disease for countries growing faba bean and field pea crops that are currently free of FBG disease. In addition, as Trifolium spp. can be a host, this also has significant biosecurity implications. Trifolium subterraneum is an important annual forage legume in Mediterranean-type climatic regions of southern Australia and parts of Africa, Asia, Europe, and North and South America (Nichols et al., 2007, 2014), with an estimated 29 million hectares sown in Australia alone (Hill and Donald, 1998; Nichols et al., 2013). Distinguishing Physoderma from Olpidium has a number of challenges. It is noteworthy that on analysis of sequences from the ITS region, our Physoderma isolates were grouped with Synchytrium in group 2, while Olpidium isolates were further away; the same applied to the partial 18S-ITS1-5.8S- ITS2-partial sequences in groups 2, 4, and 5. In Japan, on T. repens, swellings, like blisters on leaves, and galls occurring in the stems and petioles, have been mistakenly reported as caused by O. trifolii, but in fact are caused by P. trifolii (Anonymous). This is not surprising, as Olpidium spp. have frequently been misreported to cause various gall-type symptoms on leaves. Other species of Physodermataceae, like S. trifolii and Urophlyctis trifolii, have also been misidentified as O. trifolii instead of the correct name P. trifolii. Thus, it is not unexpected that the original description of the FBG pathogen in Japan was given as O. viciae. As already noted above, Physoderma isolates were closely grouped with the genus Synchytrium. The species S. aureum has long been known to attack and cause galls on various Trifolium spp. (Saccardo, 1898; Oudemans, 1921; Chilton et al., 1943). It has been widely reported on T. subterraneum and other clovers in eastern Australia (White et al., 1956). Walker (1957) reported both S. aureum and P. trifolii occurring together on the same T. subterraneum plants. Other Synchytrium spp. occurring in Australia include S. decipiens as false rust on Amphicarpaea, S. desmodii as wart disease on tropical legumes, and S. psophocarpi as false rust on winged bean (Psophocarpus tetragonolobus) (Price, 1987; Price and Lenne, 1987, 1988). 125 5. Conclusions The causal agent of faba bean gall had previously been identified as O. viciae; however, Olpidium only has endobiotic phase inside the plant and no epibiotic phase and it is a genus generally restricted to roots that does not show any symptoms on the above-ground plant parts and is not primarily spread by rain splash. In contrast, for Physoderma have both epibiotic and endobiotic phases as associated with this genus, with zoosporangia epibiotic in preparation to release of zoospores and resting spores endobiotic residing inside of the plant cells. Physoderma is a genus that generally attacks and shows symptoms on many plant parts, and is a pathogen that is primarily spread by rain splash. Recognising the epibiotic phase is foundational, not only for comprehending the disease epidemiology, but also for achieving future disease forecasting for prediction of zoospore release and consequent best timings for application of chemical sprays to reduce reinfection following initial infection cycle of faba bean. The phylogram analyses from ITS1-5.8S-ITS2, from partial18S-ITS1-5.8S-ITS2-partial 28S and from LSU (28S) all confirmed Physoderma is the causal agent of faba bean gall disease. Sample sequences were either close to Physoderma or the ascochyta pathogen Didymella. It is clearly evident from the symptom, morphological and molecular perspective that the true causal agent of FBG disease is Physoderma. This Physoderma causal agent of FBG disease can crossover between different legume host genera highlights both challenges for its management and increased biosecurity risk. 6. Acknowledgements These studies were supported by the Australian Centre for International Agricultural Research, Australia, (ACIAR Project: CIM/2017/030 “Faba Bean in Ethiopia–Mitigating disease constraints to improve productivity and sustainability”), the Debre Birhan Agricultural Research Centre (DBARC), the Ethiopian Institute for Agricultural Research (EIAR), the New South Wales Department of Primary Industries (NSW DPI) and the University of Western Australia (UWA), Australia, and the International Center for Agricultural Research in the Dry Areas (ICARDA), Morocco. We are grateful to Dr Chris Jones at the International Livestock Institute (ILRI) in Nairobi, Kenya for granting us permission to use the ILRI Plant Pathology and Molecular Laboratories at ILRI, Addis Ababa. We are especially grateful to Dr Alemayehu Teressa, ILRI Laboratories, Addis Ababa, for providing us access to all plant pathology and molecular facilities in these laboratories at all times and with whatever assistance and requirements we needed. 126 7. References Abebe, T., Birhane, T., Nega, Y. and Workineh, A. 2014. The Prevalence and Importance of Faba Bean Diseases with Special Consideration to the Newly Emerging “Faba Bean Gall” in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Sciences, 2: 33-38. Abraham, A., Makkouk, K.M., Gorfu, D., Lencho, A.G., Ali, K., Tadess,e N., Yusuf, A. and Lencho, A. 2000. Survey of faba bean (Vicia faba L.) virus diseases in Ethiopia. Phytopathologia Mediterranea, 39: 277-282. Agrios, G.N. 2005. Plant Pathology (5th ed). Academic Press, 922 pp. Ahmed, H.A., MacLeod, E.T., Hide, G., Welburn, S.C. and Picozzi, K. 2011. The best practice for preparation of samples from FTA® Cards for diagnosis of blood borne infections using African trypanosomes as a model system. Parasites and Vectors, 4: 68-68. Alexopoulos, D.J. 1962. Introductory Mycology, 2nd Edition, John Wiley and Sons, Inc, New York and London. 613 pp. Allison, J.L., Wells, J.C. and Wells, H.D. 1952. Pasture and forage legume and grass diseases in North Carolina in 1950 and 1951. Plant Disease Reporter, 36 (60): 2578. Bitew, B. 2015. Survey and identification of new faba bean disease (Qormid) in the highlands of North Shewa, Ethiopia. Current Research Microbiology and Biotechnology, 3: 561–563. Bitew, B. and Wondosen, W. 2012. New faba bean disease in North Shewa-Ethiopia: ‘Faba bean leaf and stem gall’. http//www.arari.gov.et/index.php?option Accessed on 01 June 2014. Bogale, N.H., Tagele, S.B.T. and Meiaku, M.T. 2016. Assessment of faba bean gall [Olpidium viciae (Kusano)] in major faba bean (Vicia faba L.) growing areas of northeastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 110: 87-95. Butler, E.E. and Hall, D.H. 1966. Occurrence in California of wart disease of subterranean clover caused by Physoderma trifolii. Plant Disease Reporter, 50: 796–797. CABI (Centre for Agriculture and Bioscience International ). 2012. Final Report for Enquiry Numbers 45/12 (W11387) and 50/12 (W11388) from Vicia faba samples showing faba bean gall symptoms as submitted by Beyene Bitew Eshete. 2 pp. Campbell, R.N. 1985. Longevity of Olpidium brassicae in air-dried soil and the persistence of the lettuce big-vein agent. Canadian Journal of Botany , 63: 2288–2289. Campbell, R.N. 1996. Fungal transmission of plant viruses. Annual Review of Phytopathology, 34: 87- 108. 127 CSA (Central Statistical Agency – Federal Democratic Republic of Ethiopia). 2019. Agricultural sample survey 2018 / 2019 (2011 E.C.): Report on area and production of major crops (private peasant holdings, Meher season). Chilton, S.J.P., Henson, l. and Johnson, H.W. 1943. U.S. Department of Agriculture Agriculture Miscelaneous Publications 499, 152 pp. Cook, R.P. and Dubé, A.J. 1989. Host-pathogen index of plant diseases in South Australia. South Australian Department of Agriculture, pp.1-142. Cunnington, J. 2003. Pathogenic fungi on introduced plants in Victoria. A host list and literature guide for their identification.Department of Primary Industries, Research Victoria. 57 pp. Debela, C. and Tola, M. 2017. A review on faba bean gall (Olpidium vicia) ‘Qormid’ disease and its management in Ethiopia. Journal of Natural Sciences Research, 7: 24-27. Debela, C., Zerihun, A.N., and Tola, A.M. 2017. Assessment of the occurrence and prevalence of faba bean gall (Olpidium viciae) in western highlands of Oromiya, Ethiopia. Journal of Natural Sciences Research, 7: 5. Dereje, G., Wendafrash and Kenei, G. 2012. Faba bean gall: a new disease of faba bean in Ethiopia. docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE. Earecho, M.K. 2019. Extent and Management Strategies of Faba Bean Gall Disease (Olpidium viciae Kusano) in Ethiopia "A Review". Agriculture, 5:13-24. Fajardo, S.N., Valenzuela, S., Dos Santos, A.F., Gonzalez, M.P. and Sanfuentes, E.A. 2017. Phytophthora pseudosyringae associated with the mortality of Nothofagus obliquain a pure stand in central-southern Chile. Forest Pathology, 47 (6): 12361. French, A.M. 1989. California Plant Disease Host Index. California Department of Food and Agriculture, Sacramento, 394 pp. Garrett, R.G. and Tomlinson, J.A. 1967. Isolate differences in Olpidium brassicae. Transactions of the British Mycological Society, 50: 423–423. Gould, A.B. and Schaechter, M. 2009. "Fungi: Plant Pathogenic,". pp.457-477. In: Encyclopedia of Microbiology (Third Edition). Oxford: Academic Press. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N. and Bitew, B. 2014. Faba bean gall; a new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Science and Technology, 2: 5. Herrera-Vasquez, J.A., Cordoba-Selles, M.D., Cebrian, M.C., Rosello, J.A. and Jorda, C. 2010. Genetic diversity of Melon necrotic spot virus and Olpidium isolates from different origins. Plant Pathology, 59: 240–251. 128 Herrera-Vásquez, J.Á., Cebrián, M.d.C., Alfaro-Fernández, A., Córdoba-Sellés, M.D.C. and Jordá, C. 2009. Multiplex PCR assay for the simultaneous detection and differentiation of Olpidium bornovanus, O. brassicae, and O. virulentus. Mycological Research, 113: 602–610. Hibbett, D., Abarenkov, K., Kõljalg, U., Öpik, M., Chai, B. and Cole, J.2016. Sequence-based classification and identification of fungi. Mycologia, 108: 1049–1068. Hill, M.J. and Donald, G.E. 1998. Australian Temperate Forages Database. National Forage Improvement Coordinating Committee/CSIRO Division of Animal Production. [CD‐ROM]. Hiruki, C. and Alderson, P. 2011. Morphology and distribution of resting sporangia of Olpidium brassicae in tobacco roots. Canadian Journal of Botany, 54: 2820–2826. IFPRI (International Food Policy Research Institute). 2010. Pulses Value Chain Potential in Ethiopia. Constraints and Opportunities for Enhancing Exports. International Food Policy Research Institute. 58 pp. James, T.Y., Letcher, P.M., Longcore, J.E., Mozley-Standridge, S.E., Porter, D., Powell, M.J. .2006. A molecular phylogeny of the flagellated fungi (Chytridiomycota) and description of a new phylum (Blastocladiomycota). Mycologia, 98: 860-871. Johns, R.M. 1966. Morphological and Ecological Study of Physoderma dulichii. American Journal of Botany, 53:34–45. Kusano, S. 1912. On the life history and cytology of a new species of Olpidium with special reference to the copulation of motile isogametes. Journal of the College of Agriculture, Imperial University, 4: 141-199. Lay, C.Y., Hamel, C. and St-Arnaud, M. 2018. Taxonomy and pathogenicity of Olpidium brassicae and its allied species. Fungal Biology, 122: 837–846. Lang L-J, Yu Z-H, Zheng Z-J, Xu M-S., Ying H-G. 1993. Faba bean in Chine: state-of-the-arts review. ICARDA, Aleppo, Syria. 154 pp. Liang, X.Y. 1986. Faba bean (Vicia faba) diseases in China. FABIS Newsletter, Faba Bean Information Service, ICARDA No.15, pp.49–51. Maccarone, L.D., Barbetti, M.J., Sivasithamparam, K. and Jones, R.A.C. 2010. Molecular genetic characterization of Olpidium virulentus isolates associated with big-vein diseased lettuce plants. Plant Disease, 94: 563–569. Money, N.P., Watkinson, S.C., Boddy, L. and Money, N.P. 2016. "Chapter 1 - Fungal Diversity," pp.1-36. In: The Fungi (Third Edition). (Boston: Academic Press). 129 Nichols, P.G.H., Foster, K.J., Piano, E., Pecetti, L., Kaur, P. and Ghamkhar, K. . 2013. Genetic improvement of subterranean clover (Trifolium subterraneum L.). 1. Germplasm, traits and future prospects. Crop and Pasture Science, 64: 312–346. Nichols, P.G.H., Jones, R.A.C., Ridsdill-Smith, T.J. and Barbetti, M.J. 2014. Genetic improvement of subterranean clover (Trifolium subterraneum L.). Breeding for disease and pest resistance. Crop and Pasture Science, 65: 1207–1229. Oudemans, C.A.J.A. 1921. Ennumeratio Systematica Fungorum III, pp. 872–76. Groningen, Martinus Nijhoff. Pande, A. and Rao, V.G. 1998 A Compendium of Fungi on legumes from India. Scientific Publishers India, Jodhpur, 188 pp. Price, T.V. 1987. Liberation and germination of Synchytrium desmodiisporangia. Transactions of the British Mycological Society, 89: 333–40. Price, T.V. and Lenne, J.M. 1988. Infection of Desmodium ovalifolium by Synchytrium desmodii. Transactions of the British Mycological Society, 90: 502–504. Rehner, S.A. and Samuels, G.J. 1994. Taxonomy and phylogeny of Gliocladium analysed from nuclear large subunit ribosomal DNA sequences. Mycological Research, 98: 625–634. Rochon, D’A., Kakani, K., Robbins, M. and Reade, R. 2004. Molecular aspects of plant virus transmission by Olpidium and Plasmodiophorid vectors. Annual Review of Phytopathology, 42: 211–241. Saccardo, P.A. 1898. Sylloge Fungorum XIII: pp. 1251-1252. Berlin, Borntraeger Brothers. Sahile, S., Ahmed, S., Fininsa, C., Abang, M.M. and Sakhuja, P.K. 2008. Survey of chocolate spot (Botrytis fabae) disease of faba bean (Vicia faba L.) and assessment of factors influencing disease epidemics in northern Ethiopia. Crop Protection, 27: 1457–1463. Sahtiyanci, S. 1962. Studien über einige wurzelparasitäre Olpidiaceen. Archives in Microbiology, 41: 187–228. Shivas, R.G. 1989. Fungal and bacterial diseases of plants in Western Australia. Journal of the Royal Society of Western Australia, 72: 1–62. Sparrow, F.K. Griffin, J.E. and Johns, R. M. 1961. Observations on Chytridiaceous parasites of Phanerogams. XI. A Physoderma on Agropyron repens. American Journal of Botany, 48: 850– 858. 130 Stielow, B., Lévesque, C.A., Seifert, K., Meyer, W., Irinyi, L., Smits, D., .2015. One fungus, which genes? Development and assessment of universal primers for potential secondary fungal DNA barcodes. Persoonia - Molecular Phylogeny and Evolution of Fungi, 35: 242–263. Tadesse, N., Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C., and Abraham, A., 2008. Review of research on diseases of food legumes, in Increasing Crop Production through Improved Plant Protection ed. T. Abraham. (Addis Ababa, Ethiopia: PPSE and EIAR, Addis Ababa, Ethiopia), 598 pp Teferi, T.A., Gurja, B.W., Gemechu, K.W. and Taye, T.M. 2018. Fungicidal management of the newly emerging faba bean disease “gall” (Olpidium viciae Kusano) in Tigray, Ethiopia. Crop Protection, 107: 19–25. Vilgalys, R. and Hester, M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. Journal of Bacteriology, 172: 4238–4246. Wageningen Plant Research International. 2016. Report “Faba Bean Gall – NGS analysis to identify a putative causal agent”, from Vicia faba samples showing faba bean gall symptoms as submitted by Woubit Dawit, Ambo University, and as assessed by Ilse Houwers and Theo van der Lee, Wageningen UR. 17 pp. Walker, J. 1957. Two chytrid diseases of subterranean clover in New South Wales. Australian Journal of Science, 19: 207. Walker, J. 1983. Pacific mycogeography: deficiencies and irregularities in the distribution of plant parasitic fungi. Australian Journal of Botany , 10: 89–136. White, T.J., Bruns, T., Lee, S., Taylor, J., Innis, M.A., Gelfand, D.H., .1990. "Amplification and Direct Sequencing of Fungal Ribosomal RNA Genes for Phylogenetics”, pp.315-322 in PCR Protocols, a Guide to Methods and Applications,eds. M.A. Innis, D.H. Gelfand, J.J. Sninsky and T.J. White. Academic Press, 482 pp. Xia, X. 2016. PhyPA: Phylogenetic method with pairwise sequence alignment outperforms likelihood methods in phylogenetics involving highly diverged sequences. Molecular Phylogenetics and Evolution, 102: 331–343. Xing, Z. 1984. Faba bean gall disease caused by Olpidium and its control. Acta Phytopathologica Sinica, 14: 165–173. Yan, J. 2012. Study on blister disease of broad bean caused by Olpidium viciae Kusano. Ph.D Thesis, Sichuan Agricultural University. 131 Paper III Estimating Yield Loss of Faba Bean (Vicia faba L.) Caused by Gall Disease in North Shoa, Ethiopia Beyene Bitew 1*, Chemeda Fininsa 2 and Habtamu Terefe 2 1Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 2School of Plant Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 1* Corresponding author, E-mail: beyenebitew@yahoo.com; Mobile: +251-(0)9 11 33 87 89 Abstract Faba bean is the most important food legume crop; however, its productivity is threatened by faba bean gall (FBG) disease. Faba bean gall is an emerging and destructive disease in the highlands of Ethiopia. Field experiments were conducted on farmers’ fields in Bassona Worana district, Ethiopia, under natural condition during the 2018 and 2019 cropping seasons to estimate yield losses on local cultivar and improved faba bean varieties and to determine fungicides spray schedules. The experiments consisted of two moderately tolerant (Degaga and Gora) faba bean varieties and one susceptible local cultivar, two fungicides [Bayleton 25 WP (Triadimefon 250 g kg–1) and Ridomil Gold MZ 68 WG (Metalaxyl 40+Mancozeb 640 g kg–1)] and two application schedules (10 and 15 days interval). The treatments were arranged in randomized complete block design with three replications in factorial combinations. Faba bean gall disease severity, grain yield and yield- components data were collected and subjected to analysis of variance using the PROC GLM SAS procedure. The result showed that use of different faba bean genotypes and applications of fungicides highly and significantly (p0.001) reduced FBG disease epidemics and minimized grain yield loss. The variety Gora sprayed with Bayleton at 10 days interval resulted low mean FBG disease severity of 26.7 and 13.0% and AUDPC of 1866.7 and 751.7%-days in 2018 and 2019, respectively. High mean grain yields of 3.7 t ha–1 (2018) and 5.0 t ha–1 (2019) were also obtained from the variety Gora sprayed with Bayleton at 10 days interval. High grain yield losses of 62.5% (Degaga), 54.1% (local cultivar) and 50.0% (Gora) were obtained from unsprayed plots of each genotype. Conversely, faba bean genotypes obtained yield increases of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) when treated with Bayleton at 10 days interval in 2018. Similar trends were observed regarding 100 seed weight loss and in 2019. Moreover, unsprayed plots of each genotype gave the highest disease severity and AUDPC and the lowest grain yield in both cropping years. On the other hand, faba bean grain yield and yield components demonstrated an inverse relationship with disease severity upon 132 association studies. Hence, use of tolerant faba bean varieties, like Gora, and application Bayleton at rate of 0.7 kg ha–1 at 10 days interval is essential to manage FBG disease and reduce yield losses of faba bean. Keywords: Faba bean, Fungicide, Gall, Severity, Vicia faba, Yield loss 1. Introduction Faba bean (Vicia faba L.) is the most important cool-season food and feed legume crop grown in many countries across the world. The crop is adapted to a wide range of diverse soil types and environmental conditions (Jensen et al., 2010; Stoddard et al., 2010; Duc et al., 2015). Ethiopia is considered as the secondary center of diversity and also one of the nine major agro-geographical production regions of faba bean in the world (Abebe et al., 2020). Faba bean takes the largest share of the area and production of all the pulses grown with a national productivity of 2.2 t ha–1. More than 4.1 million households are reported to grow the crop covering nearly 0.5 million hectares of land and total production of over one million tons of grain yield (CSA, 2019). The crop is the key and cheap source of good quality dietary protein for human food (Jensen et al., 2010; Amsalu, 2014), cash and its straw is used for animal feed in the crop-livestock mixed farming system of the highlands of Ethiopia. Also, faba bean plays a significant role in soil fertility restorations under rotation farming practices (Agegnehu and Fessehaie, 2006; Jensen et al., 2010; Sillero et al., 2010). In spite of its multiple benefits to smallholder farming communities, faba bean productivity in the country is very low. Reduced faba bean productivity in Ethiopia compared with worldwide average is associated several biotic and abiotic constraints, lack of improved varieties and poor agronomic practices (Mussa et al., 2008; Tadesse et al., 2008). The major abiotic constraints include water logging, frost, moisture stress, poor soil fertility, soil acidity and poor cultural practices. Chocolate spot (Botrytis fabae), ascochyta blight (Aschochyta fabae, syn. Didymella fabae), rust (Uromyces-viciae fabae) and black root rot (Fusarium solani and F. avenaceum) constituted the most economically important fungal diseases that have been causing significant grain yield losses in Ethiopia (Sahile et al., 2008; Terefe et al., 2015; Gemechu et al., 2016). More recently, faba bean production is seriously challenged and threatened by faba bean gall (FBG) disease, which was assumed to be caused by Olpidium viciae. However, the causal agent identified as Physoderma causes FBG disease in Ethiopia (You et al., 2021). Faba bean gall disease is regarded as an emerging and devastating disease in major faba bean-growing areas of the country (Hailu et al., 2014; Beyene, 2015; Bogale et al., 2016; Bitew et al., 2021; Yitayih et al., 133 2021b). Some reports showed that FBG-forming disease was reported in Japan in 1912 and became increasingly a serious problem in China since the 1970s, where more than 4,000 ha of faba bean fields were affected, and yield losses of up to 20% had been reported (Kusano, 1912; Li-juan et al., 1993; Yan, 2012). Similarly, FBG disease in Ethiopia, in hot spot areas and farms cultivated with susceptible faba bean varieties, the total crop failure was observed on farmers' fields (Teklay et al., 2014). Nearly 40% yield loss was also reported on local susceptible cultivar in the country (Wulita, 2015). Moreover, faba bean gall disease is continuing to expand to new faba bean-growing areas of Amhara and Oromia Regional States of Ethiopia, and poses a major threat to faba bean production (Dereje et al., 2012; Hailu et al., 2014; Anteneh et al., 2018; Bitew et al., 2021). However, there has been a body of knowledge that confirms a huge knowledge gap with regard to the epidemiology and management of the disease. Hence, yield losses have not been yet quantified on local cultivars and improved faba bean varieties, and cost-effective and comprehensive management options that target the host, pathogen and their interactions have not been developed and recommended for the growers in the study areas. Therefore, the objectives of the study were to: (1) estimate the amount of relative yield loss caused by FBG disease on local and improved faba bean varieties; and (2) determine the efficacies of two fungicides spray schedules for the management of FBG disease and reduction of yield loss in faba bean. 2. Materials and Methods 2.1. Description of the Study Area The experiments were carried out on farmers’ fields under natural epiphytotic conditions.in Bassona Worana district, North Shoa, Ethiopia, during the 2018 and 2019 main cropping seasons. Bassona Worana district is located at 9o41′ N latitude and 39o31′ E longitude, and at an altitude of 2980 meters above sea level. The area is characterized by light Cambisol and faba bean is the major pulse crop cultivated and yearly rotated with cereals, mainly barley and wheat crops. The district receives bimodal rainfall (the main and short rainy seasons). The main rainy season covers June to September and the short rainy season starts from January and ends in May. Faba bean grows both in the main and short rainy seasons. The rainfall in the main rainy cropping season reaches 1000 to 1200 mm, and the mean minimum and maximum temperatures are 6 and 19 oC, respectively (DBARC, 2005). The monthly mean maximum and minimum temperatures (oC) and total rainfall (mm) during 2018 and 2019 cropping seasons were obtained from Debre Birhan Agricultural Research Center weather station 134 and depicted hereunder (Figure 1). The relative humidity (%) ranged from 49.5 to 90.5% and 58.5 to 90.5% in the 2018 and 2019 main cropping seasons, respectively. 500 A 35 450 30 400 350 25 300 20 250 200 15 150 10 100 50 5 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax 500 B 35 450 30 400 350 25 300 20 250 200 15 150 10 100 50 5 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax Figure 1. Monthly mean minimum (Tmin) and maximum temperatures (Tmax) (°C) and rainfall (RF) (mm) data at Bassona Worana district, North Shoa Zone, Ethiopia, in 2018 (A) and 2019 (B). 2.2. Treatments and Experimental Design Treatment combinations contained three faba bean genotypes, two fungicides, two spray schedules and three controls (unsprayed plots) with a total of 15 treatments (Table 1). The faba bean genotypes consisted of two moderately resistant (Gora and Degaga) varieties (Wulita, 2015; Getenet and Yezbalem, 2017) and one susceptible farmers’ local cultivar to FBG disease were used in the 135 Rainfall (mm) Rainfall (mm) Temperature (oC) Temperature (oC) experiment. Genotypes’ reactions were determined only based on field evaluations at different times. Two systemic foliar fungicides [Bayleton 25 WP (Triadimefon 250 g kg–1 a.i.) and Ridomil Gold MZ 68 WG (Metalaxyl 40 + Mancozeb 640 g kg–1)] and two fungicide spray schedules, which were made at 10 and 15 days interval were used in the experimental setup. The treatments and their combinations were factorial arranged in a randomized complete block design (RCBD) with three replications. A plot size of 4.8 m2 (2.4 m × 2 m) was used in the study, and each plot was arranged to have 6 rows. Spacing between blocks and plots was 1.5 m, while rows and plants were separated by spacing of 0.4 m and 0.1 m, respectively. Planting was done on 14 June 2018 and 18 June 2019 main cropping seasons. Blended NPS fertilizer was applied once at the time of planting at a rate of 121 kg ha–1. Bayleton and Ridomil Gold fungicides were applied according to the manufacturers' recommendations at a rate of 0.7 kg ha–1 and 2.5 kg ha–1, respectively. The volume of water used to spray the fungicides was also based on the manufacturers’ recommendations per fungicide. A total of 300 and 500 L ha–1 water was used to spray Bayleton and Ridomil Gold in that order. Separate hand operated knapsack sprayer with 16 L and hollow cone nozzles was used for each fungicide. The first fungicide spray was started at the onset of typical disease symptoms at 32 days after planting (DAP) in 2018 and 28 DAP in 2019 cropping season. Five meter by three meter plastic sheet were used in the wind side direction by moving to each plot while spraying to protect draft effect of fungicides for each treatment. Seeding and fertilizer application were done manually, and weeding and other cultural practices were uniformly done for each experimental plot. 136 Table 1. Treatment combinations of faba bean genotype x fungicide x spray schedule to minimize faba bean gall disease epidemics and grain yield in Bassona Worana, north Shoa, Ethiopia, during the 2018 and 2019 main cropping seasons. S/N Treatments and their combinations 1 Degaga + Bayleton 25 WP + 10 days interval 2 Degaga + Bayleton 25 WP + 15 days interval 3 Degaga + Ridomil Gold MZ 68 WG + 10 days interval 4 Degaga + Ridomil Gold MZ 68 WG + 15 days interval 5 Degaga control (no spray) 6 Gora + Bayleton 25 WP + 10 days interval 7 Gora + Bayleton 25 WP + 15 days interval 8 Gora + Ridomil Gold MZ 68 WG + 10 days interval 9 Gora + Ridomil Gold MZ 68 WG + 15 days interval 10 Gora control (no spray) 11 Local + Bayleton 25 WP + 10 days interval 12 Local + Bayleton 25 WP + 15 days interval 13 Local+ Ridomil Gold MZ 68 WG + 10 days interval 14 Local + Ridomil Gold MZ 68 WG + 15 days interval 15 Local control (no spray) 2.3. Disease Data Assessment Date of disease onset and disease severity (leaf and stem area infected among sample plants) data were assessed during the epidemic periods of both cropping seasons. Faba bean gall disease severity was recorded at 10 days interval starting from the date of disease onset at 32 and 28 days after planting in 2018 and 2019, respectively. Twelve plants from the central rows of each plot were randomly tagged and rated using a modified 0–9 scoring scale of Ding et al. (1993); where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. A total of seven severity assessments were made until the disease starts to decline and changes 137 diminished between recording dates. Severity scores were converted into percentage severity index (PSI) for analysis (Wheeler, 1969) as follows: Sum of numerical ratings PSI = Total number of plants scored x maximum score on the scale X 100 Area under disease progress curve (AUDPC) was calculated from FBG disease severity data (Campbell and Madden, 1990) using the following formula: 푥 + 푥 AUDPC = ( 2 )(푡 − 푡 ) Where, xi = disease severity at the ith assessment, ti = the time of the ith assessment in days from the first measurement date and n = total number of disease assessments made during the epidemic period. Area under disease progress curve was expressed in %-days, as disease severity was recorded in percentage and time of assessment (t) in days. 2.4. Growth and Yield Parameters Data Growth and yield parameters, such as plant height (cm), number of pods per plant (NPP) and number of seeds per pod (NSP), 100 seed weight (g) and grain yield (gram per plot) were recorded during the study. Plant height (cm) was measured using a meter at physiological maturity stage from the harvestable rows of 10 randomly taken plants per plot. Numbers of pods per plant were recorded by counting number of pods from 10 randomly taken plants per plot. Numbers of seeds per pod were measured by counting number of seeds from pods of 10 plants per plot. Hundred seed weight (HSW) was determined from composite samples taken from each plot of the harvested total grain yield after adjusting to 10% moisture content. Also, grain yield was determined from harvestable four central rows for each treatment per plot, and converted into t ha–1 at 10% adjusted grain moisture content (Birru, 1979). 2.5. Yield Loss Assessment Yield loss assessment was determined to evaluate the effect of FBG disease pressure on faba bean genotypes and efficacy of the treatments imposed. Total costs of seed (improved varieties and local 138 cultivar), fertilizer, fungicides, fungicide applications, spraying equipment, and weeding, harvesting, threshing and other labor costs (man per days) were recorded and calculated on a hectare basis to estimate production cost and to set grain yield loss. Thus, the relative yield loss (%) due to FBG disease for each treatment was estimated as percentage yield reduction (%) of the less protected plots as compared to the maximum protected plots for all tested faba bean genotypes and fungicides over the two years using the following formula: Ymp− Ylp Relative yield loss (%) = Ymp X 100 Where, Ymp = mean yield obtained from the maximum protected plots (mean yield obtained from the best treatment) and Ylp = mean yield obtained from unsprayed and/or less protected plots (mean yield obtained from none to medium protected plots). The losses in 100 seed weight were also similarly determined. Yield increase over unsprayed plots (control) was calculated from the difference in grain yield between sprayed and unsprayed plots and expressed in percentage. 2.6. Data Analysis Faba bean gall disease severity, AUDPC, growth data, grain yield and yield components for each treatment were subjected to analysis of variance (ANOVA) using the PROC GLM procedure (SAS, 2014) to determine the response of faba bean genotypes, effect of fungicides and spray schedules on gall disease epidemic development and grain yield accumulation. Mean separations among treatments were done using Duncan's Multiple Range Test (DMRT) at 5% probability level (Gomez and Gomez, 1984). The correlation of disease with growth and yield parameters was analyzed based on Pearson's correlation coefficient to examine the relationship between faba bean gall disease parameters, growth and yield and yield components of faba bean. Linear regression analysis was done by plotting grain yield data against FBG disease severity, and regression intercepts, slopes and coefficients of determination (R2) were computed using the Minitab 19.1 statistical package. The two seasons were considered as different environment because of heterogeneity of variances as tested using Bartlett’s test (Gomez and Gomez, 1984). Thus, separate analyses were done for each cropping season. 3. Results 3.1. Analysis of Variance Test of homogeneity of error variance showed that the mean squares were heterogeneous for severity, AUDPC, disease progress rate, grain yield, hundred seed weight, plant height, number pods per plant 139 and number seeds per pod in the experimental years. Hence, separate data analysis was done for such data over years (Table 2). There was a highly significant (p0.001) difference for all traits among the test genotypes, except number of seeds per pod, which was only significant at p0.05. The treatment x year interaction for faba bean gall and agronomic traits also showed very highly significant (p0.001) variations among treatments for severity, disease progress rate and grain yield. Moreover, interaction effects attained a highly significant (p0.01) and significant (p0.05) differences among treatments for plant height and number of pods per plant, and hundred seed weight, respectively. Table 2. Combined analysis of mean square values in the integration of faba bean genotypes, fungicides and spray schedules evaluated for the management of faba bean gall disease under field conditions at Bassona Worana, North Shoa Zone, Ethiopia, during the 2018 and 2019 main cropping seasons. Parameter a Y(1) R(2) T(14) Y*R(2) Y*T(14) E(56) CV (%) Severity 2529.16*** 13.57* 2048.38*** 4.68 26.22*** 3.33 5.55 AUDPC 29104928.67*** 22304.10 1379756.98*** 122132.54* 22449.95 25904.75 9.30 DPR 8.55*** 0.03567* 4.36*** 0.007693 0.158474*** 0.0092 11.77 GY 29.42*** 0.2076*** 3.03*** 0.0638 0.2195*** 0.0410 6.02 HSW 643.20*** 3.0650 1354.03*** 8.14 36.92* 16.39 7.10 PH 21896.16*** 818.48*** 296.84*** 66.49 161.55** 65.91 9.23 NPPP 598.56*** 13.18* 27.94*** 3.68 8.83** 3.54 18.75 NSPP 0.6316* 0.7237** 0.3436* 0.4277 0.1337 0.1434 14.09 aY = Year effect; R = Replication; T = Treatment; Y*R = Year x replication interaction; Y*T = Year x treatment interaction; E = Experimental error; CV = Coefficient of variation; AUDPC = Area under disease progress curve; DPR = Disease progress rate; GY = Grain yield; HSW = Hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. Numbers in parenthesis refer to degree(s) of freedom to each respective parameter. *, ** and *** refer to level of significance at p0.05, 0.01 and 0.001 according to Duncan’s Multiple Range Test (DMRT). 3.2. Faba Bean Gall Disease Severity Faba bean gall symptoms were observed at 32 DAP in 2018 and 28 DAP in 2019 main cropping seasons. Faba bean varieties, fungicides and spray schedules showed remarkable significant (p0.05) differences for mean FBG disease severity starting from 62 DAP in 2018 and 58 DAP in 2019. Similarly, interaction effects of faba bean genotypes, fungicides and spray schedules showed very highly significant (p0.0001) differences and influenced FBG disease severity starting from 72 DAP in 2018 and 68 DAP in 2019 cropping seasons (Table 3). In the 2018 cropping season, high mean FBG disease severity of 70% (local), 63% (Degaga) and 59.7% (Gora) was recorded from untreated plots of 140 each respective faba bean genotypes. Conversely, the lowest (26.7–34.7%) mean FBG disease severity was obtained from test genotypes sprayed with Bayleton at 10 days interval. Application of Ridomil Gold at every 10 days interval was also found to reduce disease pressure compared with 15 days interval schedule in all faba bean genotypes evaluated. It was observed that spraying fungicides at 10 days interval reduced mean FBG disease severity on each faba bean genotypes. Among fungicide sprayed plots, the highest (41%) FBG disease severity was obtained from farmers’ local cultivar sprayed with Ridomil Gold at 15 days interval (Table 3). In the 2019 cropping season, the highest mean FBG severity of 68, 60.3 and 55% was noted from unsprayed plots of local cultivar, Degaga and Gora varieties, respectively. Similarly, the highest (32%) mean FBG severity was recorded from farmers’ local cultivar sprayed with Ridomil Gold at 15 days interval, while the lowest (13%) mean FBG severity was observed from Gora variety sprayed with Bayleton at 10 days interval (Table 3). Comparatively, spraying Bayleton every 10 days interval lowered FBG severity as compared to Ridomil Gold at similar spraying interval during the experimental periods. In this regard, spraying Bayleton at 10 days interval reduced final mean FBG severity by 48.7, 55.3 and 50.4% per respective unsprayed plots of variety Degaga, Gora and the local cultivar, respectively, in 2018. A similar trend of FBG disease reduction was also calculated in 2019 cropping season. And, genotypes had variable levels of mean FBG disease severity throughout the epidemic periods. The overall mean FBG severity in 2018 was higher than in 2019 cropping season (Table 3). 3.3. Area under Disease Progress Curve (AUDPC) The AUDPC (%-days) values of the tested faba bean genotypes in both cropping seasons are presented hereundr (Table 3). Analysis of variance (ANOVA) revealed very highly significant (p0.0001) variations among treatments and their interactions for AUDPC values both in 2018 and 2019 cropping seasons. In the 2018 cropping season, the highest AUDPC values of 3332.3, 3223.5 and 3246%-days were computed from severity assessments of unsprayed plots of the local cultivar, Gora and Degaga varieties, respectively. However, the lowest values of 1915 (local cultivar), 1866.7 (Gora) and 2055.7%-days (Degaga) were computed from plots of each faba bean genotype sprayed with Bayleton at 10 days interval. Application of Ridomil Gold at 10 days interval also lowered disease pressure and AUDPC values as compared to values calculated from data on plots treated with both fungicides at 15 days interval. Among fungicides and spray schedules, application of Bayleton at 10 days interval 141 reduced AUDPC by 42.5, 42.1 and 36.7% compared with unsprayed plots of the local cultivar, Gora and Degaga faba bean varieties in that order. Moreover, Bayleton fungicide showed comparative advantage over Ridomil Gold at the same spraying schedule. For example, Bayleton sprayed plots of Degaga (1.7 and 3.1 %), Gora (3.2 and 4.8%) and the local cultivar (13.9 and 19.9%) at 10 days interval gave noticeable AUDPC reductions as compared to Ridomil Gold at 10 and 15 days intervals, respectively (Table 3). Table 3. Effects of faba bean genotypes, fungicides and spray schedules on mean faba bean gall disease (FBG) severity and AUDPC at Bassona Worana, North Shoa, Ethiopia, during the 2018 and 2019 main cropping seasons. Treatment combinations Severity (%) AUDPC (%-days) Genotype Fungicide Spray schedule (Trade name) (days) 2018 2019 2018 2019 Degaga Bayleton 10 32.3cd 16.3bcd 2055.7b–d 936.7d–g 15 36.0bcd 23.0bcd 2056b–d 1002.5c–g Ridomil Gold 10 37.0bcd 26.3bc 2090b–d 1103c–e 15 38.1bcd 26.3bc 2120b–d 1134.2cd Control Unsprayed 63.0a 60.3a 3246a 1926.7b Gora Bayleton 10 26.7d 13.0d 1866.7d 751.7g 15 31.7cd 16.3cd 2008.3cd 850e–g Ridomil Gold 10 34.3cd 13.1d 1928.3cd 778.3g 15 36.7cd 16.7cd 1961.7cd 802.5fg Control Unsprayed 59.7a 55.0a 3223.3a 1760b Local Bayleton 10 34.7cd 17.3cd 1915cd 775g 15 36.7cd 26.3bcd 2085bcd 1054.2c–f Ridomil Gold 10 38.0bc 29.7bc 2223.3bc 1138.3cd 15 41.0b 32.0bc 2390b 1214.2c Control Unsprayed 70a 68.0a 3332.3a 2204.8a Mean 38.2 27.6 2299.56 1162.21 CV (%) 18.6 24.5 7.67 11.94 Genotype P < 0.0001 P 0.0001 P 0.0001 P 0.0001 Genotype x Fungicide P 0.0001 P 0.0018 P 0.0301 P 0.0110 Genotype x Fungicide x Spray schedule P 0.0037 P 0.0010 P 0.0725 P 0.0601 Mean values within a column followed by different letter(s) are significantly different at 5% level of probability using Duncan’s Multiple Range Test (DMRT). 3.4. Effect of Gall Disease on Grain Yield of Faba Bean The analysis of variance showed that treatments and their interactions established very highly significant (p0.0001) differences for mean grain yield in both cropping seasons (Table 4). Gora (3.7 t ha–1), Degaga (3.4 t ha–1) and the local (3.1 t ha–1) faba bean genotypes sprayed with Bayleton at 10 142 days interval attained high mean grain yield, while low mean grain yields of 1.3, 1.4 and 1.9 t ha–1 were obtained from unsprayed plots of the varieties Degaga, local and the Gora, respectively, in 2018 cropping season. Similarly, in the 2019 cropping season, high mean grain yields of 5.0, 4.2, and 4.6 t ha–1 were obtained from the varieties Gora, Degaga and the local cultivar sprayed with Bayleton at 10 days interval in that order. Of course, unsprayed plots of the varieties Gora (2.9 t ha–1), Degaga (2.4 t ha–1) and the local cultivar (2.5 t ha–1) produced the lowest mean grain yields compared with plots of each respective faba bean genotype treated with either fungicide at any spraying schedule in 2019 (Table 4). Comparatively, application of Bayleton fungicide resulted in mean grain yield advantages over both the control and Ridomil Gold sprayed plots of each faba bean genotype in both cropping years. In this regard, spraying Bayleton at 10 days interval registered mean grain yield gain of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) over unsprayed plots of respective genotypes. Also, Bayleton application at 10 days interval recorded observable increase in mean grain yield of 37.5% (Degaga), 31.7% (Gora) and 17.3% (local cultivar) as compared to plots of each faba bean genotype sprayed with Ridomil Gold at 15 days interval in the 2018 cropping season. Similar trends were noted in 2019. However, faba bean genotypes sprayed with Bayleton at 10 days interval relatively revealed higher grain yield than a Ridomil Gold schedule at 15 days interval both in 2018 and 2019 cropping seasons. An overall higher mean grain yield was obtained in 2019 than the mean grain yield harvested in the 2018 cropping season (Table 4). 3.5. Effect of Gall Disease on Hundred Seed Weight of Faba Bean The analysis of variance result showed significant (p≤0.05) differences among genotypes, fungicides and application schedules for hundred seeds weight in both 2018 and 2019 cropping seasons (Table 4). In the 2018 cropping season, high hundred seed weight was obtained from the varieties Gora (77.7 g) and Degaga (47.0 g) sprayed with Bayleton at 10 days interval, while the lowest (41.7 g) value of hundred seed weight was recorded from unsprayed plots of the local faba bean cultivar. Likewise, the variety Gora (83.2 g) showed heavier hundred seed weight, followed by the variety Degaga (57.1 g) treated with Bayleton at 10 days interval; however, unsprayed plots of the local cultivar produced only 42.8 g hundred seed weight in the 2019 cropping season. At all spraying schedules, Bayleton application achieved improved hundred seed weight in both testing years. For instance, spraying Bayleton at every 10 days interval showed 11.2 and 24.9% (Degaga), 7.5 and 15.5% (Gora) and 12.3 143 and 18.7% (local cultivar) hundred seed weight increase over unsprayed plots of each respective faba bean genotypes in 2018 and 2019, respectively. The results also revealed that the local cultivar and variety Degaga were more responsive to fungicides applications than the variety Gora (Table 4). Table 4. Effect of faba bean genotypes, fungicides and spray schedules on grain yield and hundred seed weight in Bassona Worana District, North Shoa, Ethiopia, during the 2018 and 2019 cropping seasons. Treatment combinations Grain yield (t ha–1) 100-seed weight (g) Spray Genotype Fungicide (Trade name) schedule 2018 2019 2018 2019 (days) Degage Bayleton 10 3.41ab 4.22b–d 47.03b 57.1bc 15 3.14cd 4.20b–d 41.77b 54.4bc Ridomil Gold 10 3.06bc 4.00cd 47.02b 50.3c–e 15 2.48f 3.96cd 46.09b 48.0c–e Control Unsprayed 1.28h 2.4e 41.75b 42.9e Gora Bayleton 10 3.70a 5.03a 77.68a 83.2a 15 3.58a 4.48b 76.72a 81.2a Ridomil Gold 10 3.53a 4.30bc 73.80a 81.2a 15 2.81de 4.24b–d 73.10a 80.4a Control Unsprayed 1.85g 2.87e 71.84a 70.3b Local Bayleton 10 days 3.12bc 4.58b 47.54b 52.2cd 15 days 3.11c 4.22b–d 43.14b 50.3c–e Ridomil Gold 10 days 2.66ef 3.84d 46.02b 49.8c–e 15 days 2.66ef 3.82d 43.64b 44.0de Control Unsprayed 1.41h 2.48e 41.71b 42.8e Mean 2.79 3.93 54.32 59.5 CV (%) 5.86 5.96 6.76 7.4 Genotype P0.0001 P0.0001 P 0.0001 P 0.0001 Genotype x Fungicide P0.0165 P0.0253 P 0.0110 P 0.0340 Genotype x Fungicide x Spray schedule P0.0621 P0.0442 P 0.0338 P 0.0721 3.6. Effect Gall Disease on Yield Components of Faba Bean Plant height, number pods per plant and number seeds per pod showed highly significant (p0.01) differences between treatments both in 2018 and 2019 cropping seasons, except number seeds per pod, which showed a non-significant variation among treatments in 2019 (Table 5). In 2018 cropping season, the tallest (86.1 cm) plant height was recorded on the variety Gora, followed by Degaga (77.1 cm) sprayed with Bayleton at 10 days interval. Unsprayed plots of the local cultivar attained the 144 shortest (57.1 cm) plant height compared with other treatments (Table 5). In 2019, the tallest plant heights of 114.8, 113.9 and 113 cm were measured the local, Degaga and Gora genotypes treated with Bayleton at 10 days interval, respectively. Moreover, high mean number of pods per plant was obtained from variety Degaga (9.9) and the local cultivar (9.4) sprayed with Bayleton at 10 days interval, while lowest (5.1) number of pods per plant was recorded from plots of unsprayed local cultivar in the 2018 cropping season. Faba bean genotypes also exhibited similar trends regarding mean number of pods per plant in 2019 crop season. Furthermore, the mean number of seeds counted per pod was highly influenced by faba bean genotypes evaluated and fungicides applied and associated schedules used in 2019 cropping season. High number of seeds per pod was recorded from plots of the varieties Gora (3.1) and Degaga (3.0) sprayed with Bayleton at 10 days interval and low number of seeds per pod (2.1) was recorded from unsprayed local cultivar (Table 5). Though statistically insignificant, high number of seeds per pod was recorded on variety Gora (3.3) and Degaga (3.3) sprayed with Bayleton at 10 days interval, while the lowest (2.3) number of seeds per pod was recorded from unsprayed plots of the local cultivar in 2018 crop season. Note that combined analysis of variance indicated that there was no significant (p>0.05) difference among treatments and their interactions for days to 50% seedling emergence and days to flowering; however, there was significant difference between sprayed and unsprayed plots in days to physiological maturity in both cropping seasons. Unsprayed plots showed relatively short maturity periods (150–153 days) as compared to fungicide sprayed plots (157–160 days) (data not shown). 145 Table 5. Effect of faba bean genotypes, fungicides and spray schedules on mean plant height, number of pods per plant and number of seeds per pod in, Bassona Worana, North Shoa, Ethiopia, in 2018 and 2019 main cropping seasons. Treatment combinations Plant height (cm) No. of pods plant–1 No. of seeds pod–1 Genotype Fungicide Spray schedule (Trade name) (days) 2018 2019 2018 2019 2018 2019 Degaga Bayleton 10 77.2ab 113.9a 9.9a 16.5ab 3.3a 3.0ab 15 70.1b–d 107.5a–d 8.3a-d 12.8c–f 2.6ab 2.4b–d Ridomil Gold 10 72.1a–d 105.8a–d 8.9a–c 15.1cd 2.6ab 2.6a–d 15 69.1b–d 101.6a–d 7.9a–e 11.7c–f 2.7ab 2.5b–d Control Unsprayed 67.1cd 94.9c–e 5.6de 10.3d–f 2.4b 2.2cd Gora Bayleton 10 86.1a 113.0a 8.0a–d 11.2c–f 3.3a 3.1a 15 81.7ab 112.4ab 6.7b–e 10.1d–f 2.9ab 2.8a–c Ridomil Gold 10 82.2a 96.6b–e 6.9b–e 10.0d–f 2.9ab 2.9ab 15 74.2a–c 94.2c–e 6.1c–e 8.9ef 2.6ab 2.6a–d Control Unsprayed 61.6cd 92.0de 5.6de 8.1f 2.6ab 2.3cd Local Bayleton 10 75.3a–c 114.8a 9.4ab 19.5a 3.0ab 2.7a–d 15 69.3b–d 109.1a–c 8.6a–c 13.4b–d 2.7ab 2.6a–d Ridomil Gold 10 69.9b–d 105.4a–d 8.6a–c 16.8ab 2.9ab 2.7a–d 15 68.7b–d 98.8a–e 7.0b–e 12.9b–d 2.7ab 2.6a–d Control Unsprayed 57.1d 83.6e 5.1e 12.9b–d 2.3b 2.1d Mean 71.7 102.9 7.5 12.6 2.8 2.6 CV (%) 11.00 8.11 19.60 17.63 15.56 12.20 Genotype P 0.0046 P 0.0001 P 0.0078 P <0.0001 P 0.280 P 0.0019 Genotype x Fungicide P <0.0001 P 0.0001 P 0.001 P 0.0201 P <0.072 P <0.055 Genotype x Fungicide x Spray schedule P 0.001 P 0.0101 P 0.0420 P 0.001 P 0.6001 P 0.832 Mean values within a column followed by different letters are statistically significant at 5% probability level by Duncan’s Multiple Range Test (DMRT). 146 3.7. Relative Grain Yield and Hundred Seed Weight Losses of Faba Bean due to Gall Disease The mean relative grain yield and hundred seed weight losses computed for each treatment combination against respective unsprayed plots over years are presented hereunder (Tables 6 and 7). Among treatments compared, genotypes sprayed with Bayleton at 10 days interval were used as reference points to calculate relative grain yield and hundred seed weight losses in both cropping years. Combined use of faba bean genotypes with fungicides at both application schedules reduced grain yield and hundred seed weight losses as compared to untreated plots of each genotype in the two years. The highest grain yield losses of 62.5% (for Degaga), 54.2% (for local cultivar) and 50% (for Gora) were calculated from unsprayed plots of each respective genotype in 2018 main crop season. On the contrary, evaluated faba bean genotypes obtained yield increases of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) when treated with Bayleton at 10 days interval in 2018 crop season. Similarly, 45.8, 42.2 and 36% yield losses were recorded from unsprayed plots of the local cultivar, Degaga and Gora varieties in that order in 2019 crop season. Among fungicide application schedules, Ridomil Gold spraying at 15 days interval showed high level of relative yield losses in all faba bean genotypes in both cropping seasons (Table 6). Moreover, genotypes, fungicides, spray schedules and their combinations also strongly influenced relative losses in hundred seed weights both in 2018 and 2019 cropping seasons (Table 7). In this connection, the highest (12.3%) hundred seed weight loss was computed for unsprayed plots of the local cultivar, followed by the varieties Degaga (11.7%) and Gora (7.8%) in 2018 crop season. In the 2019 cropping season, the untreated variety Degaga received relatively the highest (24.8%) hundred seed weight loss penalty compared with the local cultivar (16.1%) and variety Gora 15.5% (Table 7). The comparative analysis exhibited that Bayleton sprayed plots of each genotype attained lower grain yield and hundred seed weight losses than plots treated with Ridomil Gold at both application schedules in both testing years. Similarly, higher grain yield losses due to faba bean gall pressure were calculated from all genotypes tested in 2018 than in the 2019 cropping season (Tables 6). 147 Table 6. Effect of genotypes, fungicides and fungicide spray schedules on relative grain yield losses of faba bean due to faba bean gall disease in Bassona Worana District, North Shoa, Ethiopia, in 2018 and 2019 main cropping seasons. Treatment combinations Relative grain yield loss, 2018 a Relative grain yield loss, 2019 a Genotype Fungicide Spray schedule Grain yield Relative Yield Yield Grain yield Relative Yield loss Yield (Trade name) (days) (t ha–1) yield (%) loss (%) increase (%) (t ha–1) yield (%) (%) increase (%) Degaga Bayleton 10 3.41 100 0.0 166.4 4.22 100 0.0 56.3 15 3.14 92.1 –7.9 145.3 4.20 99.57 –0.43 55.7 Ridomil Gold 10 3.06 89.7 –10.3 139.1 4.00 94.7 –5.3 48.0 15 2.48 72.7 –27.3 93.8 3.96 93.8 –6.2 46.7 Control Unsprayed 1.28 37.5 –62.5 0.0 2.40 57.8 –42.2 0.0 Gora Bayleton 10 3.70 100 0.0 100 5.03 100 0.0 75.2 15 3.58 96.8 –3.2 93.5 4.48 89.1 –10.9 56.1 Ridomil Gold 10 3.53 95.4 –4.6 90.8 4.30 85.5 –14.5 49.8 15 2.81 75.9 –24.1 51.9 4.24 84.3 –15.7 47.7 Control Unsprayed 1.85 50 –50.0 0.0 2.87 64 –36.0 0.0 Local Bayleton 10 3.12 100 0.0 121.3 4.58 100 0.0 84.4 15 3.11 99.68 –0.32 120.6 4.22 92.1 –7.9 69.9 Ridomil Gold 10 2.66 85.3 –14.7 88.7 3.84 83.9 –16.1 54.7 15 2.66 85.3 –14.7 88.7 3.82 83.3 –16.7 53.6 Control Unsprayed 1.41 45.9 –54.1 0.0 2.48 54.2 –45.8 0.0 a Yield increase over control plots was determined as the difference between sprayed and unsprayed plots in each year. 148 Table 7. Effect of genotypes, fungicides and fungicide spray schedules on relative 100-seed weight losses of faba bean due to faba bean gall disease in Bassona Worana, oNrth Shoa, Ethiopia, in 2018 and 2019 main cropping seasons. Treatment combinations Relative 100-seed weight loss, 2018 Relative 100 seed weight loss, 2019 Genotype Fungicide Spray schedule 100-seed Relative 100seed 100 seed weight 100 seed Relative 100 100 seed (Trade name) (days) weight (g) weight (%) loss (%) weight (g) seed weight (%) weight loss (%) Degaga Bayleton 10 47.03 100.00 0.0 57.1 100 0.0 15 41.77 88.77 –11.23 54.4 95.27 –4.73 Ridomil Gold 10 47.02 99.98 –0.02 50.3 88.09 –11.91 15 46.09 97.99 –2.01 48 84.07 –15.93 Control Unsprayed 41.75 88.24 –11.76 42.9 75.13 –24.87 Gora Bayleton 10 77.68 100.00 0.0 83.2 100 0.0 15 76.72 98.76 –1.24 81.2 97.6 –2.4 Ridomil Gold 10 73.8 95.01 –4.99 81.2 97.6 –2.4 15 73.1 94.10 –5.9 80.37 96.6 –3.4 Control Unsprayed 71.84 92.18 –7.82 70.3 84.5 –15.5 Local Bayleton 10 47.54 100.00 0.0 52.2 100 0.0 15 43.14 90.74 –9.26 50.3 96.36 –3.64 Ridomil Gold 10 46.02 96.80 –3.2 49.8 95.4 –4.6 15 43.64 91.80 –8.2 44 84.29 –15.71 Control Unsprayed 41.71 87.74 –12.26 43.8 83.91 –16.09 149 3.8. Correlation of FBG Disease Parameters with Grain Yield and Yield Components The correlation analysis showed that there was a significant and strong relationship between FBG disease severity, grain yield and yield components of faba bean (Table 8). Final mean disease severity established an inverse correlation with faba bean plant height, number of pods per plant, number of seeds per pod, grain yield and hundred seed weight. In the 2018 cropping season, final mean disease severity showed a strong negative correlation with grain yield (r = –0.96***), plant height (r = – 0.84***), number pods per plant (r = –0.71**) and number of seeds per pod (r = –0.63*). Final mean disease severity also had negative, but weak relationship with hundred seed weight (r = –0.29ns). Grain yields of the genotypes maintained a positive correlation with plant height (r = 0.86***), number of pods per plant (r = 0.65**) and number of seeds per pod (r = 0.58*) and hundred seed weight (r = 0.39ns). Moreover, yield components had variable levels of positive correlation among each other in 2018. During the 2019 cropping season, correlations among FBG disease severity, grain yield and yield components were observed to have similar trends as in the 2018 cropping season (Table 8). Table 8. Coefficients of correlation (r) and level of significance between and among gall disease parameters, grain yield and yield components of faba bean in 2018 (upper diagonal) and 2019 (lower diagonal), Bassona Worana, Northern Shoa, Ethiopia. Parameter a PSI GY HSW PH NPPP NSPP PSI – –0.95*** –0.29ns –0.84*** –0.71** –0.63* GY 0.96*** – 0.39ns 0.86*** 0.65** 0.58* HSW –0.51ns 0.49* – 0.58* 0.32ns 0.37ns PH –0.73** 0.80*** 0.14ns – 0.42* 0.57* NPPP –0.22ns 0.23ns 0.52* 0.53* – 0.48* NSPP –0.84*** 0.83*** 0.54* 0.67** 0.37* – a PSI = Final percent severity index, GY = Grain yield, HSW = Hundred seed weight, PH = Plant height, NPPP = Number of pods per plant and NSPP = Number seeds per pod. ***, **, * and ns refer to level of significance at p0.001, p0.01, p< 0.05 and p>0.05, respectively. 3.9. Regression Analysis Relating Faba Bean Grain Yield with Gall Disease Severity The regression presented (Figure 2) indicated that high grain yield and low severity was obtained from genotype sprayed with Bayleton at 10 days interval, and low grain yield and high severity was obtained from unsprayed control plots of all genotypes in both years. The linear regression analysis was run to describe the associations between faba bean grain yield and disease severity. The mean 150 values of disease severity at the final date of assessment were used to predict yield loss per treatment combination at each location over the two years (Figure 2). The estimated slope of the regression line in 2018 cropping season was –0.066, –0.058 and –0.047 on varieties Degaga, Gora and the local cultivar in that order. A similar trend regarding estimated slope of the regression line was established for all genotypes in 2019 (Figure 2). The estimated values implied that an increase in mean values of FBG disease severity caused remarkable reduction in grain yield of all evaluated faba bean genotypes irrespective of the treatments in both years. Of course, the model explained that yield losses in variety Degaga (92.7 and 97.3%), Gora (92.0 and 87.6%) and the local cultivar (95.3 and 98.2%) were attributed to gall disease pressure in 2018 and 2019, respectively. For example, each graph tended to show that for each unit increase in FBG disease severity, there were 0.066, 0.058 and 0.047 t ha–1 grain yield loss in the tested Degaga, Gora and the local genotypes of faba bean in order of appearance in 2018. In the 2019 cropping season, 0.044 t ha–1 (Degaga), 0.041 t ha–1 (Gora) and 0.041 t ha–1 (Local) predicted grain yield losses of faba bean were noticed at every unit increase in disease severity (Figure 2). 151 Degaga, 2018 y = 5.41 – 0.0663x Degaga, 2019 y = 5.13 – 0.0443x R2 = 92.7% R2 = 97.3% 4 P = 0.008 4.5 P = 0.002 3.5 4 3 2.5 3.5 2 3 1.5 2.5 1 2 30 35 40 45 50 55 60 65 15 20 25 30 35 40 45 50 55 60 65 Final FBG severity (%) Final FBG severity (%) Gora, 2018 y = 5.30 – 0.0584x Gora, 2019 y = 5.13 – 0.0413x R2 = 92.0% R2 = 87.6% 4 P = 0.01 5.5 5 P = 0.019 3.5 3 4.5 2.5 4 3.5 2 3 1.5 2.5 25 30 35 40 45 50 55 60 10 15 20 25 30 35 40 45 50 55 60 Final FBG severity (%) Final FBG severity (%) 3.5 3 Local, 2018 y = 4.64 – 0.0465x Local, 2019 y = 5.19 – 0.0405x R2 2.5 R2 = 95.3% = 98.2% 2 P = 0.004 5 4.5 P = 0.001 1.5 4 1 3.5 30 35 40 45 50 55 60 65 70 75 3 2.5 2 15 20 25 30 35 40 45 50 55 60 65 70 Final FBG severity (%) Final FBG severity (%) Figure 2. Linear regression relating final FBG disease severity with grain yield of faba bean (Degaga, Gora and local) genotypes sprayed with Bayleton and Ridomil Gold fungicides at 10 and 15 days intervals in Bassona Worana, North Shoa, Ethiopia, in the 2018 and 2019 cropping seasons. 152 Yield (t ha–1) Yield (t ha–1) Yield (t ha–1) Yield (t ha–1) Yield (t ha–1) Yield (t ha–1) 4. Discussion In the present study, typical FBG disease symptoms started at 28 and 32 days after planting during seedling growth stage, in 2018 and 2019 cropping seasons, respectively, and application of fungicide spray began from the onset of the disease. Disease severity was observed to decrease after three consecutive foliar applications of the fungicides. In this study, two moderately tolerant (Degaga and Gora) varieties and one susceptible local cultivar faba bean, two fungicides (Bayleton and Ridomil Gold) and two spray schedules (10 and 15 days interval) were compared, and the results revealed that use of different tolerant faba bean varieties and applications of fungicides reduced FBG disease severity and AUDPC, and increased grain yield and reduced relative yield and hundred seed weight losses. A relatively higher FBG disease severity and AUDPC and lower yield losses were computed for 2018 crop season than for the 2019 cropping season. High FBG disease pressure in the former cropping season could be due to environmental conditions, such as high rainfall and relative humidity, and low temperature (Figugre 1). Similarly, Teferi et al. (2018) reported that variability in weather conditions showed marked differences in FBG disease components between seasons; and also high relative humidity, air temperature within the range of 10–20 ◦C and high amount of rainfall are reported as conducive for zoospore germinations and subsequent high epidemics of FBG-forming diseases (Yan, 2012). Moreover, high rainfall and relative humidity could enhance spore germination and initiate infections of some pathogen types and oospore-producing pathogens could easily be dispersed with the help of rain splash. That is, dispersal processes in pathogens have major effects on both the spatial and temporal distribution of epidemics (Jeger, 1999; Bitew et al., 2021). Note that splash dispersal of inocula from leaves to leaves or from soil to leaves increases the infections by pathogens, such as Plasmopara viticola of grapevines and other oospore-forming pathogens (Rossi and Caffi, 2012) as observed in this investigation. A study on Phytophthora spp., an oospore forming-pathogen belonging to Oomyceteous group, noticed that disease intensity increases rapidly when rainfall causes splash dispersal of primary inocula from the soil to the aerial parts of the plants (Ristaino and Gumpertz, 2000). Of course, the findings of the current study confirmed that FBG disease intensity was very high at the seedling growth stage where the amount and frequency of rainfall was maximum that could create splash and dispersal of the 153 pathogen for further infections, as also documented by reports of related studies conducted both in China and Ethiopia (Yan, 2012; Teferi et al., 2018). The results of the current study showed that use of different faba bean genotypes and applications of fungicides at different spray schedules reduced FBG disease parameters. Amongst the tested genotypes, Gora (Wulita, 2015) and Degaga (Getenet and Yezbalem, 2017) were reported as moderately tolerant to FBG disease. Nevertheless, the variety Degaga was not found to be tolerant to FBG disease in this study. Whatsoever the case, the use of host resistance is the easiest and the most economical strategy of crop disease management. Unfortunately, no variety is registered for FBG disease resistance in the country. On the other hand, Ethiopia is considered as secondary center of diversity and a number of improved faba bean varieties have been developed and released to be used against many major diseases, like chocolate spot and for general production purpose under different recommendation domains (Gemechu et al., 2005; Tamene et al., 2015). A few released faba bean varieties and germplasms evaluated against FBG disease in Ethiopia showed differences in degree of resistance reaction to FBG disease (DBARC, 2015; Wulita, 2015; Getenet and Yezbalem, 2017; Alehegn et al., 2018), and this variation might be due to the genetic background of the genotypes evaluated as noted in the study. In this current study, integration of Bayleton and Ridomil Gold fungicides at different spray schedules with the faba bean varieties Degaga, Gora and the local cultivar showed significant differences in FBG disease parameters at different levels, and this could indicate that FBG disease management using tolerant varieties and fungicides could reduce the FBG disease epidemics and yield losses of faba bean. Other reports also showed that application of Bayleton at different rates and modes of applications reduced FBG severity (Li juan et al., 1993; Bitew and Tigabie, 2016; Hailemariam et al., 2017). For example, Hailemariam et al. (2017) used Bayleton as seed dressing fungicide at a rate of 0.3 kg 100 kg-1 of faba bean seed; Teferi et al. (2018) applied Bayleton at a rate of 0.3 kg ha-1 as foliar spray; and Bitew and Tigabie (2016) applied Bayleton as foliar application at a rate of 0.7 kg ha-1. In all application practices, disease parameters were significantly lowered and grain yield and yield components were improved. The effectiveness of Bayleton at both application intervals as compared to Ridomil Gold in minimizing the disease and relative yield losses could be attributed to differences on the active 154 ingredients and modes of actions among the fungicides that would affect the efficacy of the fungicides tested. Ridomil Gold is a semi-systemic fungicide and it is a mixture of two fungicides with low amount of Metalaxyl and high amount of Mancozeb. Mancozeb is a preventive contact fungicide that needs to be applied before the onset of the disease and requires high frequency of application at shorter intervals. It was also observed that application of Ridomil Gold at 10 days interval lowered disease intensity and increased grain yield as compared to 15 days spraying interval and unsprayed plots though a relatively higher rate of application is needed. Similarly, Wondwosen et al. (2019) reported that fungicides showed different levels of efficiency across locations and over seasons, and Teferi et al. (2018) also noted that both Bayleton and Ridomil Gold are effective to manage FBG disease in Tigray areas. The highest grain yield losses of 62.5% (Degaga), 54.2% (local cultivar) and 50% (Gora) were obtained from unsprayed plots of each respective genotype in 2018 crop season. Conversely, evaluated faba bean genotypes gained yield increases of 166.4% (Degaga), 100% (Gora) and 121.3% (local cultivar) when treated with Bayleton at 10 days interval in 2018 cropping season. All these results indicated that fungicides and tolerant genotypes were key component of FBG management. Among the fungicide application schedules, Ridomil Gold spraying at 15 days interval caused high level of relative yield losses in all faba bean genotypes in 2018 crop season. Similar trends were noticed regarding hundred seed weight loss in 2019. Previous studies also documented that combined applications of host resistance and fungicides found to reduce grain yield losses through lowering intensities of different diseases in faba bean (Sahile et al., 2010; El-Sayed et al., 2011; Bekele et al., 2018; Teferi et al., 2018; Wondwosen et al., 2019; Mengesha et al., 2021). Accordingly, Teferi et al. (2018) indicated that application of Bayleton (68.5 and 24.9%) and Ridomil Gold (46.1 and 32.5%) to control FBG resulted in grain yield advantages as compared to control plots of faba bean at Ofla and Enda-Mekoni, Ethiopia, respectively. Variable yield advantages over the control plots were recorded on Gora, CS20DK and local faba bean genotypes due to 2–3 foliar spray frequencies of both Bayleton and Ridomil Gold to manage FBG at Degem and Mush, Ethiopia (Bekele et al., 2018). Moreover, a study conducted to assess the efficacies of six fungicides for the management of FBG at Lay Gorebela and Mush of Ethiopia in 2014 and 2015 confirmed that fungicide sprays significantly increased grain yield 155 and reduced disease severity and AUDPC at both locations in the two cropping seasons (Wulita, 2015). However, evaluated fungicides showed variable efficacies against the disease under natural infection conditions (Wondwosen et al., 2019). The correlation analysis revealed that terminal disease severity had negative correlations with plant height (r =0.84***), number pods per plant (r = 0.71**) and number of seeds per pod (r = 0.63*) and grain yield (r = 0.96***). On the other hand, grain yield established a positive and significant correlation with plant height (r =0.86***), number of pods per plant (r =0.65**) and number of seeds per pod (r =0.58*) in 2018 cropping season. The same trends were observed in 2019 crop season too. In addition, the regression analysis also demonstrated an inverse association between terminal severity and grain yield in both cropping years. Several other studies also reported similar findings in the associations of FBG disease and yield parameters in faba bean (Wulita, 2015; Bitew and Tigabie, 2016; Bekele et al., 2018; Wondwosen et al., 2019). The negative relationships between disease severity and yield parameters could imply that disease severity is an important disease component to estimate loss levels in grain yield of faba bean genotypes. Similarly, Su et al. (2006) noted that terminal disease severity and AUDPC had considerable negative adverse effects on grain yield of the crops and, hence, important in determining the magnitude of losses in yield and yield components. Of course, FBG disease has been reported to infect leaves, stems, flowers, and pods of the crop (Bitew et al., 2021), and subsequently supposed to diminish grain yield accumulation. The findings of the current study generally confirmed that spraying Bayleton significantly minimized FBG disease parameters and grain and hundred seed weight losses though both fungicides played notable roles in the management of the disease in both years. Yet, Bayleton is registered to manage wheat rust diseases and Ridomil Gold is registered to manage late blight of potato and downy mildew on different vegetable and fruit crops in Ethiopia (Belay et al., 2015). That is, methods and rates of applications for both fungicides to manage FBG disease are not uniform across different areas in the country. For instance, Bayleton was applied as seed dressing and foliar spray at a rate of 0.3 to 2 kg ha-1 and Ridomil Gold was applied at a rate of 2 to 3 kg ha-1as foliar application to manage FBG disease (Wulita, 2015; Bitew and Tigabie, 2016; Hailemariam et al., 2017; Bekele et al., 2018; Teferi et al., 2018). Such disparities could indicate 156 that there is no agreed rate and schedule of application in the tested fungicides to manage FBG disease in hot spot areas. Even though the rates of both fungicides were based on manufacturers’ recommendation for cereal and horticultural crops in this study, the fungicides were able to reduce disease epidemics and reduced yield losses, and maximized grain yield. However, both scenarios stressed the need for further study to determine proper rate, method and frequency of application in response to the type of fungicides, the genetic nature of faba bean genotypes, growth stages and weather conditions of the production areas and cropping seasons. 5. Conclusions The findings of the present study confirmed that FBG disease causes a significant yield penalty on both local and improved faba bean genotypes evaluated. Despite the fact, integration of tolerant faba bean varieties and foliar application of fungicides reduced FBG disease epidemics and increased grain and yield components. In this regard, spraying of Bayleton fungicide at 10 days interval along with the varieties Degaga, Gora and the local cultivar strongly minimized disease parameters and yield losses, and enhanced grain yield and yield components at different levels. On the contrary, unsprayed plots of each respective genotype resulted in the highest disease severity, AUDPC, grain yield and hundred seed weight losses in both testing years. Comparatively, the faba bean variety Gora produced the highest grain yield and the lowest disease components and yield losses compared with other genotypes tested. Moreover, when sprayed with Bayleton, the faba bean genotypes showed superior performance to Ridomil Gold application at both spraying schedules (10 and 15 days interval) in both cropping seasons. Application of Ridomil Gold at 10 days interval reduced disease severity and AUDPC, and increased grain yield and yield components over the 15 days interval spraying schedule studied; however, very high rate of Ridomil Gold per hectare is required for the purpose that could not be cost effective. Therefore, integrating the tolerant faba bean variety Gora with Bayleton fungicide at a rate 0.7 kg ha-1 at 10 days interval application schedule could be recommended as the best option for FBG disease management, thereby reducing grain yield and hundred seed weight losses. Furthermore, evaluating more effective fungicides, efforts on developing host resistance and testing of large number of faba bean genotypes along with fungicides and other management options are indispensable to sustainably produce the crop in the study areas and elsewhere in other locations with similar agro-ecologies. 157 6. Acknowledgements The study was financially supported by Amhara Regional Agricultural Research Institute (ARARI), Australian Center for International Agricultural Research (ACIAR Project: CIM/2017/030 “Faba Bean in Ethiopia–Mitigating disease constraints to improve productivity and sustainability”) and the International Center for Agricultural Research in the Dry Areas (ICARDA), and Africa RISING/USAID through ICARDA. The work is part of a PhD diseartation research at Haramaya University and the University is especially acknowledged for facilitating the overall research activities. 7. References Abebe, G., Kindu, G. Yechale, M. Birhanu, A. Anteneh, A. and Amlaku, A. 2020. Optimization of P and K fertilizer recommendation for faba bean in Ethiopia: the case for Sekela District. International Journal, 142: 169–179. www.worldnewsnaturalscience.com. Agegnehu, G. and Fessehaie, R. 2006. Response of faba bean to phosphate fertilizer and weed control on Nitisols of Ethiopian highlands. Journal of Agronomy, 2: 281–290. Alehegn, M., Tiru, M. and Taddess, M. 2018. Screening of faba bean (Vicia faba) varieties against faba bean gall disease (Olpidium viciae) in East Gojjam Zone, Ethiopia. Journal of Biology and Agriculture and Healthcare, 8(1): 50–55. Amsalu, N. 2014. Role of faba bean (Vicia faba L.) for intensification of cereal-based cropping systems in the humid highlands of Ethiopia. PhD thesis, Ghent University, Gent, Belgium, 201 pp. Anteneh, A., Yohannes, E., Mesganaw, G., Solomon, G. and Getachew, T. 2018. Survey of faba bean (Vicia faba L.) diseases in major faba bean-growing districts of North Gondar, African Journal of Plant Sciences, 12(2): 32–36. Bekele, B., Dawit, W., Kassa, B. and Salvaraj, T. 2018. Management of faba bean gall disease using cultivars and fungicides in North Showa Zone of Central Ethiopia. International Journal of Research in Agricultural Sciences, (5): 2348–3997. Belay, T. M., Arthur, P.J. Mol, and Peter, O. 2015. Pesticide use practices among smallholder vegetable farmers in Ethiopian Central Rift Valley. Journal of Environment, Development and Sustainability, 19: 301–324. 158 Beyene, B. 2015. Survey and identification of new faba bean disease (Qormid) in the highlands of North Shewa, Ethiopia. Current Research in Microbiology and Biotechnology, 3: 561– 563. Birru, A. 1979. Agricultural field experiment management manual part III. Institute of Agricultural Research. Addis Ababa, Ethiopia. pp. 35-42. Bitew, B., Fininsa, C., Terefe, H., Barbetti, M.J. and Ahmed, S. 2021. Spatial and temporal distribution of faba bean gall (Physoderma) disease and its association with biophysical factors in Ethiopia. International Journal of Pest Management. https://doi.org/10.1080/09670874.2021.1998724. Bitew, B. and Tigabe, A. 2016. Management of Faba Bean Gall Disease (Kormid) in North Shewa Highlands, Ethiopia. Advances in Crop Sciences and Technology, 4: 1–5. Bogale, N., Setu, B. and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae (Kusano) in major faba bean (Vicia faba L.) growing areas of Northeastern Amhara, Ethiopia. Journal of Agricture and Environment, 110: 87–95. Campbell, C.L. and Madden, V.L. 1990. Introduction to Plant Disease Epidemiology, New York: John Wiley and Sons, Inc. New York, USA. 532 pp. CSA (Central Statistical Agency). 2019. Agricultural sample survey (2019). Report on area and production of major crops. Statistical Bulletin 587, Central Statistical Agency, Addis Ababa, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2005. Debre Birhan Agricultural Research Center, Sub-Center and Research sites description. Debre Birhan, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2015. Annual Progress Report, Debre Birhan, Ethiopia. Dereje, G., Wendafrash, M. and Gemechu, K. 2012. Faba Bean Galls: A new disease of faba bean in Ethiopia. Holetta Agricultural Research Center. Availableat: https://docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE/edit. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean Newsletter, 32: 8-10. Duc, G., Aleksic, J.M., Marget, P., Mikić, A., Paull, J. and Redden, R.J. 2015. Faba bean, pp. 141–178. In: Grain Legumes, A.M. De Ron (ed.). https:// doi: 10.1007/978-1-4939-2797- 5-5. 159 El-Sayed, S.A., El-Shennawy, R.Z. and Ismail, A.I. 2011. Fungicidal management of chocolate spot of faba bean and assessment of yield losses due to the disease. Annual Agricultural Sciences. 56 (1): 27–35. https://doi.org/10.1016/j.aoas.2011.05.004. Gemechu, K., Asnake, F and Million, E. 2016. Reflections on highland pulses improvement research in Ethiopia. Past achievements and future directions: Ethiopian Journal of Agricultural Sciences. EIAR 50th Year Jubilee Anniversary Special Issue: 17–50. Gemechu, K., Mussa, J., Tezera, W. and Getnet, D. 2005. Extent and pattern of genetic diversity for morpho-agronomic traits in Ethiopian highland pulse landraces II. Faba bean (Vicia faba L.). General Research and Crop Evolution, 52: 551–561. Getenet, Y. and Yehizbalem, A. 2017. Adaptation of faba bean varieties for yield, for yield components and against faba bean gall (Olpidium viciae Kusano) disease in South Gondar, Ethiopia. Crop Journal. https:// doi:10.1016/j.cj.2017.05.00. Gomez, K.A. and Gomez, A.A. 1984. Statistical procedures for agricultural research, 2nd Edition. John Wiley and Sons Inc., New York, USA. Hailemariam, B.N., Tilahun, M., Bazie, S., Desale, T. and Abere, E. 2017. Management of faba bean gall in faba bean-producing areas of Eastern Amhara, Ethiopia. Journal of Agricture and Environment, for International Development. 111(2): 343–350. DOI: 10.12895/jaeid.20172.641. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N., Bitew, B., Boydom, A., Kassa, D., and Temesgen, T. 2014. Faba bean gall; a new threat for faba bean (Vicia faba) production in Ethiopia. Advances Crop Science and Technology, 2(4): 1-5, http://dx.doi.org/10.4172/2329–8863, 1000144. Jeger, M.J. 1999. Improved understanding of dispersal in crop pest and disease management: Current status and future directions of agriculture. Meteorology, 97: 331-349. Jensen, E.S., Peoples, M.B. and Hauggaard-Nielsen, H., 2010. Faba bean in cropping systems. Field Crops Research, 115: 203 –216. Kusano, S. 1912. The life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of the College of Agriculture, 4: 142–199. Li-juan, L., Zhao-hai, Y., Zhao-Jie, Z., Ming-shi, X. and Han-qing, Y. 1993. Faba Bean in China: a State-of-the-art Review. International Center for Agricultural Research in the Dry Areas, Aleppo, Syria, pp. 127–128. 160 Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. 48 pp. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. Ristaino, J.B. and Gumpertz, M.L. 2000. New frontiers in the study of dispersal and spatial analysis of epidemics caused by species in the genus Phytophthora. Annual Review. Journal of Phytopathology, 38: 541–576. Rossi, V. and Caffi, T. 2012. The role of rain in dispersal of the primary inoculum of Plasmopara viticola. Journal of Phytopathology, 102: 158–165. Sahile, S., Fininsa, F., Sakhuja, P.K. and Ahmed, S. 2010. Yield loss of faba bean (Vicia faba) due to chocolate spot (Botrytis fabae) in sole and mixed cropping systems in Ethiopia. Archieve Phytopathology and Plant Protection. 43 (12): 1144–1159. https://doi.org/10.1080/ 0323540080234379. Sahile, S., Fininsa, C. Sakhuja, P.K. and Seid, A. 2008. Effect of mixed cropping and fungicides on chocolate spot (Botrytis fabae) of faba bean (Vicia faba) in Ethiopia. Crop Protection, 27: 275–282. SAS (Statistical Analysis System). 2014. SAS User’s Guide. Version 9.4.2. SAS Institute, Inc., Cary, NC, USA. Sillero, J.C., Villegas-Fernández, A.M., Thomas, J., Rojas-Molina, M.M., Emeran, A.A., Fernández-Aparicio, M., Rubiales, D. 2010. Faba bean breeding for disease resistance. Field Crop Research,115(3): 297–307. Stoddard, F.L., Nicholas, A.H., Rubiales, D., Thomas, J. and Villegas-Fernández, A.M. 2010. Integrated pest management in faba bean. Field Crops Research, 115: 308-318. Su, H., Hwang, S.F., Chang´e, K.F., Conner, R.L., Xue, A.G., Warkentin, T.D., Blade, S.F. and Turnbull, G.D. 2006. Assessment of yield loss caused by mycosphaerella blight in field pea crops in western Canada. Journal of Plant Disease and Protection, 113 (6): 267– 274. Tadesse, N., Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C. and Abraham, A. 2008. Review of research on diseases of food legumes. pp. 85– 132. In: Tadesse, A. (Eds.) Increasing Crop Production through Improved Plant Protection. Addis Ababa, Ethiopia: PPSE and EIAR. 161 Tamene, T., Gemechu, K. and Hussein, M. 2015. Genetic progresses from over three decades of faba bean (Vicia faba L.) breeding in Ethiopia. Australian Journal of Crop Sciences, 9: 41-48. Teferi, T.A., Weldemichael, G.B., Wakeyo, G.K. and Mindaye, T.T. 2018. Fungicidal management of the newly emerging faba bean disease “gall” (Olpidium viciae Kusano) in Tigray, Ethiopia. Crop Protection, 107: 19–25. Teklay, A., Tsegaye, B., Yemane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Science, 2: 33–38. Terefe, H., Fininsa, C., Sahile, S., Dejene, M. and Tesfaye, K. 2015. Effect of integrated cultural practices on the epidemics of chocolate spot (Botryitis fabae) of faba bean (Vicae fabae L.) in Hararghe Highlands, Ethiopia. Journal of Pests, Diseases and Crop Protection, 3(4): 113–123. Wheeler, B.J. 1969. An introduction to plant diseases. John Wiley and Sons Ltd., London, 374pp. Wondwosen, W., Dejene, M., Tadesse, N. and Ahmed, S., 2019. Fungicide management of faba bean gall (Olpidium viciae) in Ethiopia. Turkish Journal of Agriculture, 7 (7): 1075– 1081. https:// doi.org/10.24925/turjaf.v7i7.1075-1081.2189. Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) Using varieties and fungicides in North Shoa, central Ethiopia. MSc Thesis, Haramaya University, Haramaya, Ethiopia. pp 1-65. Yan, J.M. 2012. Study on blister disease of broad bean caused by Olpidium Viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. Yitayih, G., Fininsa, C., Terefe, H. and Shibabaw, A. 2021. Distribution and association of faba bean gall (Olpidium viciae) disease with agro-ecological factors in northwestern Ethiopia. Journal of Plant Diseases and Protection, 128: 1603–1615. https://doi.org/10.1007/s41348-021-00497-0. You, M.P, Eshete, B.B., Kemal, S.A., van Leur, J. and Barbetti, M.J. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology. https://doi.org/10.1111/ppa.13359. 162 Paper IV Evaluation of Faba Bean (Vicia faba L.) Genotypes for FBG Disease Resistance Reaction at Basona Worana, North Show, Ethiopia Beyene Bitew 1*, Chemeda Fininsa 2, Habtamu Terefe 2, Martin Barbetti 3, and Seid Ahmed 4 1Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 1*Corresponding author, E-mail: beyenebitew@yahoo.com; Mobile: +251-(0)9 11 33 87 89 2School of Plant Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 3School of Plant Biology, Faculty of Science, the University of Western Australia, Australia 4Biodiversity and Crop Improvement Program, International Center for Agricultural Research in the Dry Areas (ICARDA), B.P. 6299, Rabat, Morocco Abstract Faba bean (Vicia faba L.) is the most important grain legume crop grown in Ethiopia. However, the productivity is highly inhibited by faba bean gall (FBG) disease. Field experiment was conducted at Bassona Worana District, North Shoa, Ethiopia, in 2018 and 2019 to evaluate faba bean genotypes for their reaction to FBG disease. A total of 415 faba bean genotypes and breeding lines obtained from Ethiopian Biodiversity Institute (EBI) and International Center for Agricultural Research in the Dry Areas (ICARDA), including four check varieties, were evaluated under natural field conditions in augmented design with ten blocks in 2018 cropping season. One hundred four genotypes, which showed low FBG severity in 2018 cropping season, were advanced in 2019 cropping season in four blocks with four check varieties in augmented design. Faba bean gall disease parameters and grain yield data were collected. Check varieties were analyzed using statistical software and compared with test genotypes. The result showed that there were highly significant differences among the test genotypes and check varieties in FBG severity, AUDPC and grain yield. Low mean FBG severity of 37.0, 37.5 and 38.0% were computed for accession numbers 1085, 1082 and 1004 and AUDPC values of 1005, 1010 and 1025%-day were calculated for the same accession numbers in the same order. Maximum (3.78 t ha-1) grain yield was recorded from accession number 1085. One genotype (local collection 26884) and 28 breeding lines, which showed low FBG severities and good grain yields, were 163 selected and and these test genotypes should be further evaluated and advanced to develop disease resistant and high yielder faba bean varieties through crossing. Keywords: Accession number, AUDPC, Augmented design, Breeding lines, Grain yield, Severity, Varieties. 1. Introduction Faba bean (Vicia faba L.) is the major cool-season food and feed legume grown in many countries in the world. Ethiopia is the world’s second producer of faba bean next to China (FAOSTAT, 2018) and faba bean is the leading pulse crop produced next to cereals (Asfaw et al., 1994). The crop takes the largest share of the area and production of all the pulses grown in Ethiopia with a current productivity of 2.12 t ha-1 (CSA, 2021). Over 4.1 million households grow the crop covering 0.5 M hectares and total production of more than one million tons of grain in 2021 main cropping season in the country (CSA, 2021). The crop is valuable as the cheap source of protein for human food, cash and its straw is key animal feed in the crop- livestock farming system of the highlands of Ethiopia. In crop rotation practice, faba bean plays a significant role in soil fertility restoration, which contributes in reducing carbon footprint of cereal-based agricultural production (Agegnehu et al., 2006; Jensen et al., 2010). However, the productivity of faba bean in Ethiopia is far below its potential, mainly due to susceptibility to a number of biotic and abiotic stresses as well as lack of improved varieties and poor agronomic practices (Mussa et al., 2008). The major abiotic constraints for faba bean productions are water logging, soil acidity, frost and water stress. Fungal diseases (both foliar and root); broad and grass weed, parasitic weeds (especially Orobanche spp.) and storage pests, such as Bruchids (Callosobruchus spp.) are the most prevalent biotic stresses (Dereje, 1999). Among fungal diseases, chocolate spot (Botrytis fabae), Ascochyta blight (Ascochyta fabae, syn. Didymella fabae, teleomorphic stage), rust (Uromyces viciae-fabae), black root rot (Fusarium solani, F. avenaceum) and virus diseases (Safa and Khaled, 2007) are major diseases limiting faba bean productivity (Sahile et al., 2008; Terefe et al., 2015). In addition to the common diseases occurring in the country, the major faba bean-producing areas are seriously challenged and threatened by an emerging faba bean gall (FBG) disease, which was assumed to be caused by Olpidium viciae (Dereje et al., 2012; Hailu et al., 2014; 164 Beyene, 2015; Bogale et al., 2016); however, the causative agent of FBG in Ethiopia is Physoderma (You et al., 2021). Due to FBG severe infections and susceptible varieties, total crop failures were observed on farmers' fields and farmers were forced to replace faba bean sown fields by other early maturing crops (Teklay et al., 2014). Similarly, FBG-forming disease caused by Olpidium viciae was reported from Japan and China with considerable yield losses (Kusano, 1912; Xing, 1984). In Ethiopia, FBG increases very rapidly; however, there are no well-known FBG management recommendations and no FBG resistant varieties have been developed to date. Ethiopia is known as a secondary center of diversity for faba bean and landraces are excellent sources of gene(s) for improving grain yield and disease resistance (Gemechu et al., 2005). Despite a number of improved faba bean varieties have been developed and released for general production under different recommendation domains, including the mid and high altitude agro- ecologies and the waterlogged vertisol areas (Tamene et al., 2015), there are no resistant varieties recommended to manage FBG disease. Most of farmers are using a few types of fungicides to manage FBG; however, use of fungicides is not environmentally safe and economically affordable. Host resistance may be one of the best options to manage the FBG for long-term program. Developing host resistance has been considered as one of the most economically and environmentally sounds strategy in managing many diseases. Evaluating different genotypes and identifying sources of resistant genes could be an important and critical component to develop FBG management practices. Some reports showed that some faba bean varieties released in the country at different agro-ecologies have shown different reactions against FBG (Wulita (2015; Getenet and Yizblem, 2017), but most of the released varieties tested are not, in the real sense, resistant to FBG, but showed different degree of reactions or tolerance (DBARC, 2015). However, different genotypes are not tested widely in the country against FBG. Hence, this study was undertaken to evaluate faba bean genotypes for their reaction to FBG disease. 165 2. Materials and methods 2.1. Description of the Study Area Field experiments were carried out on farmers’ fields in Bassona Worana district during the 2018 and 2019 main cropping seasons. Bassona Worana is located at 9o41′N latitude and 39o31′E longitude, and at an altitude of 2980 meters above sea level, North Shoa, Ethiopia. The area is characterized by light Cambisol and faba bean is the major pulse crop cultivated and yearly rotated with cereals, mainly barley and wheat crops. The district receives bimodal rainfall where the main rainy season covers June to September and the short rainy season starts from January and ends in May. Faba bean grows in both seasons, and under supplementary irrigation during short rainy season, when rainfall is scarce in the area. The rainfall in the main cropping season reaches 1,000 to 1,200 mm and the mean minimum and maximum temperatures are reported as 6 and 19 oC, respectively (DBARC, 2015). The monthly mean maximum and minimum temperatures (oC) and total rainfall (mm) during 2018 and 2019 were obtained from the weather station of Debre Birhan Agricultural Research Site and illustrated hereunder (Figure 1). The relative humidity ranged from 49.5 to 90.5% in 2018 and 58.5 to 90.5% in the 2019 cropping seasons. 500 A 35 400 30 25 300 20 200 15 10 100 5 0 0 Months RF Tmin Tmax 166 Rainfall (mm) Temperature (oC) 500 B 35 400 30 25 300 20 200 15 10 100 5 0 0 Months RF Tmin Tmax Figure 1. Monthly mean minimum (Tmin) and maximum (Tmax) temperatures (°C) and rainfall (RF) (mm) data at Bassona Worana district of North Shoa, Ethiopia, in 2018 (A) and 2019 (B). 2.2. Experimental Materials and Planting Methods Two hundred local faba bean genotypes collected from different parts of the country [Amhara, Oromia, Tigray and Southern Nations and Nationalities People’s Region (SNNPR) National Regional States], obtained through Ethiopian Biodiversity Institute (EBI), and 215 breeding lines introduced from ICARDA, Jordan, Lebanon was evaluated for FBG disease resistance under natural conditions at Bassona Worana (hot spot area). Moderately resistant varieties showing varying degrees of reactions; Gora, Gachena, Degaga and the susceptible farmers' local cultivar were randomized and used as check materials and planted in each experimental block. Each test genotype and checks were planted continuously in 2 rows having 2 m plot length with spacing of 0.4 m and 0.1 m between rows and between plants, respectively, with a total of 1.6 m2 plot size or area. The local cultivar was planted as spreader row in all directions of the experimental units to uniformly increase disease pressure. Recommended seed rate (200 kg ha-1) of faba bean and NPS (compositions: 38 kg P2O5, 19 kg N and 7 kg S) blended fertilizer was applied once at the time of planting with the rate of 121 kg ha-1 was applied at the time of planting. The seed and fertilizer were applied manually. Both genotypes and breeding lines were arranged as one set of experiment and planted in ten blocks using augmented experimental design. Plots were hand weeded and all cultural practices were kept uniform across all plots. Genotypes and breeding lines, which showed low FBG severity, were selected based on their reaction and by comparing with each check and the tested materials with each other. Genotypes and breeding lines, which 167 Rainfall (mm) Temperature (oC) showed low FBG severity and high grain yield, were selected and planted in the 2019 main cropping season in the same hot spot area with the same design, like that in 2018, and further evaluated for FBG resistance and grain yield. 2.3. Disease Assessment Date of disease onset, disease incidence (number of plants infected per total number of plants inspected) and disease severity (leaf and stem area damaged among sample plants) were assessed during the epidemic periods of the disease. Faba bean gall disease severity was recorded at 10 days interval starting from date of disease onset, which were at 30 days after planting (DAP) in 2018 and 28 DAP in 2019 cropping seasons. Disease incidence was calculated using the following formula: Number of plants showing disease symptoms per plot Disease incidence (%) = Total number of plants assessed per plot X 100 For disease severity measurements, 12 plants from each plot were randomly tagged prior to the appearance of typical disease symptoms and used for the assessment. Disease severity was rated using a modified 0–9 scoring scale of Ding et al. (1993); where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. A total of seven severity assessments were made until the disease starts to decline and changes diminished between records. Severity scores were converted into percentage severity index (PSI) for analysis (Wheeler, 1969) as follows: Sum of numerical ratings PSI = Number of plants scored x maximum score on the scale X 100 Area under disease progress curve (AUDPC) was calculated from disease severity data. Percent disease severity calculated for each plot during the course of the experiment was summarized and used to calculate AUDPC (Campbell and Madden, 1990) based on number of ratings at the 168 intervals of 10 day disease assessment and expressed in %-days and computed using the following formula: (xi + xi + 1) AUDPC = 2 (ti + 1 − ti) Where, Xi =disease severity at the ith assessment, ti = the time of the ith assessment in days from the first measurement date and n = total number of disease assessments made during the epidemic period. 2.4. Growth and Yield Parameters Assessment Growth and yield parameters, including days to 50% seedling emergence, days to 50% flowering, days to 90% physiological maturity, plant height (cm), number of pods per plant, number of seeds per pod, 100 seed weight (g) and grain yield (g plot–1), were recorded from each experimental plot. Plant height per plot was taken at harvest from 10 pre-tagged plants and number of pods per plant and number of seeds per pod was recorded from randomly identified five plants, and then averaged for final data records. Grain yield was determined from each plot, and was converted into tons per hectare (t ha–1) at 10% adjusted grain moisture content (Birru, 1979). Hundred seed weight was also determined from randomly taken 100 seeds from each plot grain yield at 10% moisture content and expressed in grams (g). 2.5. Data Analysis The data of check varieties were subjected to analysis of variance (ANOVA) using the PROC GLM procedure of Statistical Analysis System (SAS, 2014). Faba bean gall disease percent severity index (PSI) was used for FBG parameters analysis. Disease severity, AUDPC, rate of disease development, growth data, grain yield and yield components of checks were analyzed and compared with test genotypes and breeding lines. To determine disease progress rate from the linear regression, fitness of different models was tested based on the magnitude of coefficient of determination (R2) and residuals standard error (SE) obtained per model (Campbell and Madden, 1990) and finally logistic model, ln[(y/1-y)] (Van der Plank, 1963) showed high R2 and used to determine the disease progression from the linear regression for each test material. The correlation and regression analyses were used to examine the association between and among 169 FBG disease parameters and yield, and yield components of the test genotypes. Mean comparison was done for checks using Least Significant Difference (LSD) at 5% probability level. Genotypes and breeding lines showing low FBG severity and high grain yield were compared with checks and selected for further studies based on the data analysis results. 3. Results 3.1. Faba bean Gall Disease Incidence and Severity Faba bean gall disease incidence reached 100% within 20 days after disease onset on 90% of the tested faba bean genotypes. Faba bean gall severity increased gradually starting from the onset of the disease to physiological maturity growth stage of the genotypes. Among the tested genotypes, majority of local genotypes (>95%) obtained from Ethiopian Biodiversity Institute (EBI) showed high mean FBG severity as compared to the introduced genotypes. In 2018 cropping season, genotypes obtained from EBI showed 38 to 87% FBG severity and only one genotype (local collection number 228186) showed 38% FBG severity. Five genotypes had less than 55% FBG severity and 195 genotypes showed 57 to 87% FBG severity. Similarly, introduced breeding lines showed 32 to 73% FBG severity. One hundred eleven genotypes showed 32 to 50% FBG severity and 104 genotypes had 51 to 73% FBG severity. Mean FBG severity on check varieties on Degaga (58.8%), Gachena (56%), Gora (54%) and local (62.1%) were measured in 2018 cropping season. Among 415 tested genotypes, 116 showed ≤50% FBG severity score, and 113 genotypes showed >50 to ≤60%. One hundred eighty-six genotypes showed greater than 60 to 90% FBG severity than other genotypes evaluated in 2018 main cropping season (Table 3). In 2019 cropping season, FBG severity ranged from 37 to 65.33%. Among the tested genotypes, five breeding lines showed greater than 30 ≤40 FBG severity, thirty-six breeding lines had >40 to ≤50%, 53 breeding lines showed greater than 50 to ≤60% and 10 breeding lines showed greater than 60 to <70% FBG severity. Mean FBG severities on check varieties in 2019 were 52.30, 49.33, 46 and 58.75% on Degaga, Gachena, Gora and local cultivar, respectively. Among the selected genotypes, 13 genotypes showed less than 45% FBG severity and three breeding lines showed less than 40% FBG severity. Breeding lines 1085, 1082 and 1004 showed 37.0, 37.5 and 38% FBG severity, respectively. One genotype (local collection-26884) and 28 introduced 170 breeding lines showed low FBG severity as compared to check varieties both in 2018 and 2019 cropping seasons. Summary of FBG disease parameters and grain yield and yield components at each block are tabulated hereunder (Table 1 and 2). Table 1. Summary of mean FBG disease parameters, grain yield and yield components on 415 test genotypes and breeding lines at Bassona Worana, Ethiopia, in 2018 main cropping season. FBG disease parameters1 Yield and yield components2 Block/ No of check genotypes Incidence Severity AUDPC (%) (%) (%-days) GY HSW PH (t ha-1) (g) (cm) NPPP NSPP 1 42 97.06 72.29 2256.07 0.65 48.45 62.58 9.56 2.63 2 42 100.00 73.26 2401.43 1.07 66.79 68.79 9.94 2.71 3 42 96.67 67.50 2310.24 1.65 49.15 69.92 9.40 2.63 4 42 90.89 63.93 2089.40 0.85 34.82 62.05 7.65 2.74 5 42 93.59 58.19 1761.31 2.63 48.12 67.59 8.95 2.70 6 42 97.67 46.31 1175.00 2.81 73.28 73.88 8.15 2.78 7 42 90.88 51.98 1447.56 2.53 73.17 69.67 7.07 2.75 8 42 88.79 49.48 1332.02 2.93 72.23 73.33 6.90 2.85 9 42 91.68 51.88 1439.29 2.49 71.37 73.74 7.26 2.81 10 37 95.33 50.54 1449.05 2.88 70.17 69.17 6.38 2.86 Mean 94.26 58.54 1766.14 2.05 60.76 69.07 8.13 2.75 SE 1.15 3.16 144.95 0.17 4.47 1.32 0.40 0.03 Checks Degaga 1 94.33 58.80 1640.00 1.37 50.9 66.90 7.50 2.76 Gachena 1 90.67 56.00 1622.00 2.07 61.1 68.7 7.20 2.67 Gora 1 88.50 54.00 1523.00 2.25 76.40 71.1 7.90 2.90 Local 1 97.67 62.10 1902.00 1.02 50.50 64.7 5.30 2.40 1AUDPC = Area under disease progress curv. 2GY = grain yield; HSW = hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. SE = Standard error. 171 Table 2. Summary of mean FBG disease parameters, grain yield and yield components on 104 test genotypes and breeding lines at Bassona Worana, Ethiopia, in 2019 cropping season. FBG disease parameters Yield and yield components Block/ No. of check genotypes Incidence Severity AUDPC GY HSW PH (%) (%) (%-day) (t ha-1) (g) (cm) NPPL NSPP 1 26 91.67 49.57 1268.65 1.43 76.51 69.43 6.48 2.31 2 26 95.88 49.15 1291.54 1.51 83.68 67.03 6.36 2.38 3 26 90.67 53.56 1375.77 1.37 80.60 61.75 5.60 2.38 4 26 96.33 54.56 1402.50 1.30 77.91 59.57 5.99 2.33 Mean 93.64 51.71 1334.62 1.40 79.68 64.45 6.11 2.35 SE 0.91 1.37 32.29 0.04 1.58 2.28 0.20 0.02 Checks Degaga 1 88.33 52.00 1352.50 1.64 55.30 65.35 6.50 2.30 Gachena 1 85.67 49.33 1326.25 2.18 73.43 70.33 6.30 2.10 Gora 1 83.50 46.00 1240.00 2.50 78.27 71.90 5.90 2.30 Local 1 91.67 58.75 1363.75 1.62 48.05 71.38 7.60 2.32 1AUDPC = Area under disease progress curve. 2GY = grain yield; HSW = hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. SE = Standard error. 172 Table 3. Summary of mean FBG severity of 415 and 104 genotypes and breeding lines tested at Bassona Worana district, Ethiopia, in 2018 and 2019, respectively. Faba bean gall disease severity (%) 2018 2019 No. of genotype/ FBG breeding severity Mean FBG No. of FBG type/breeding severity Mean FBG lines range Severity geno lines range severity >30 ≤40 38.14 5 >30≤40 38.5 94 >40 ≤50 46.39 36 >40≤50 46.56 113 >50 ≤60 55.04 53 >50≤60 54.58 186 >60 <90 74.72 10 >60≤80 62.00 Mean 53.57 50.41 SE 7.85 5.01 Checks: Check Degaga - 58.80 Degaga - 52.30 Gachena - 56.00 Gachena - 49.33 Gora - 54.00 Gora - 46.00 Local - 62.10 Local - 58.75 Mean 57.73 51.60 CV (%) 5.08 3.51 LSD (0.05) *** * 3.2. Area under Disease Progress Curve (AUDPC) The AUDPC values of FBG on genotypes and breeding lines in 2018 ranged from 900 to 2560%-days. Among 415 genotypes and breeding lines tested, 14 of them showed ≤ 1,000 AUDPC value, 148 genotypes and breeding lines showed ≤ 1500 AUDPC value, 86 genotypes and breeding lines showed ≤ 2000 AUDPC value, 167 genotypes and breeding lines showed greater than 2000 to ≤ 2655%-days AUDPC value. In 2018, the AUDPC values on check varieties were 1523, 1622, 1640 and 1909 %-days on Gora, Gachena, Degaga and local cultivar, respectively. Similarly, in 2019 cropping season, the AUDPC value range on selected genotypes and breeding lines was 990-1720%-days. Likewise, the AUDPC values on check varieties were 1240, 1326.25, 1352.5 and 1363.75%-days on Gora, Gachena, Degaga and local cultivar, respectively. The AUDPC values on local genotypes were higher than that of introduced breeding lines. The AUDPC value in 2018 on local genotypes ranged from 1325-2655%-days, 173 whereas on introduced breeding lines, the AUDPC value range was 900-2130%-days. One local genotype and 28 introduced breeding lines showed low AUDPC value as compared to check varieties in 2018 and 2019 cropping seasons (Table 4). Table 4. Summary of genotypes /breeding lines tested and AUDPC values in 2018 and 2019 cropping season at Bassona Worana district, Ethiopia Area under disease progress curve (%-day) No. of genotype/breeding lines 2018 2019 AUDPC (%- day) range No. of genotype / AUDPC (%- day) breeding lines range 14 >900 ≤1000 11 >900≤1100 148 >1000 ≤1500 17 >1101≤1200 86 >1500 ≤2000 14 >1205≤1300 167 >2000 ≤2655 62 >1310≤1720 Mean 1770 1334.62 SE 37.75 17.90 Checks: Check Degaga 1640.0 Degaga 1352.50 Gachena 1622.0 Gachena 1326.25 Gora 1523.0 Gora 1240.00 Local 1909.0 Local 1363.75 Mean 1673.50 1320.63 174 Table 5. Mean faba bean gall disease parameters, grain yield and yield components of faba bean genotypes and breeding lines tested at Bassona Worana, north Shoa, Ethiopia in 2019 cropping seasons. S/N Genotype / FBG AUDPC GY HSW PH Breeding line Severity (%) (%-day) (t ha-1) (g) (cm) NPPP NSPP 1 26884* 40.5 1110 1.73 45.0 72.5 6.5 2.91 2 26879 49.33 1295 1.72 42.02 70.7 9.1 2.01 3 28776 52.11 1465 0.73 38.97 65.5 5.6 2.44 4 30017 49.33 1310 1.32 39 68.5 6.9 2.61 5 30016 55.33 1525 0.95 47.79 72.5 6.4 2.16 6 1002* 42.00 1030 1.87 90.4 85.1 6.1 2.24 7 1004* 38.00 1025 2.32 80.3 81.9 7.7 2.41 8 1005* 47.00 1205 2.47 80.0 83.1 6.7 2.95 9 1006* 48.00 1210 2.26 97.5 84.2 7.3 2.52 10 1009* 41.00 1030 2.74 91.4 85.8 10.3 2.12 11 1010 57.21 1590 1.89 85.19 57.3 2.8 2.57 12 1011 56.34 1575 0.69 93.55 63.1 4.4 2.15 13 1015 62.22 1670 0.46 70.48 60.6 4.1 2.45 14 1017 53.33 1320 0.69 77.68 63.1 5.4 2.29 15 1018 55.21 1370 1.16 71.55 53.9 6.3 2.21 16 1019 56.67 1490 1.64 87.37 59.9 5.2 2.38 17 1020 55.33 1380 0.60 56.97 46.6 4.9 2.93 18 1024 51.33 1345 1.79 94.15 64.9 5.3 2.30 19 1025* 45.67 1090 1.96 93.9 68.6 8.9 2.18 20 1028 50.67 1325 1.07 77.53 64.7 5.0 2.90 21 1029 48.67 1345 1.20 80.25 61.1 4.8 2.54 22 1030* 47.00 1045 1.36 111.5 83.6 6.3 2.40 23 1031* 48.00 1155 1.38 92.6 63.0 7.7 2.44 24 1032* 47.00 1035 1.26 66.6 65.7 8.4 2.23 25 1034* 42.00 1055 1.35 98.1 79.7 10.2 2.27 26 1036 49.67 990 0.50 79.5 79.5 6.3 1.51 27 1038 52.33 1360 0.49 72.92 52.2 5.5 2.92 28 1039 60.55 1580 0.70 84.74 59.3 5.5 2.28 29 1041 53.33 1345 1.00 92.88 68.1 6.2 2.10 30 1042 57.33 1465 0.54 74.53 50.8 5.0 2.15 32 1045* 52.00 1060 0.87 77.2 58.0 9.9 2.21 33 1047* 47.00 1195 0.97 80.9 78.7 7.9 2.33 34 1057 56.25 1530 0.97 85.87 53.6 3.7 2.53 35 1054 55.33 1495 1.37 72.38 61.6 5.2 2.44 36 1069 56.33 1490 1.10 90.12 55.9 4.6 2.58 37 1070 48.56 1260 1.13 81.48 72.9 5.8 2.63 38 1071 55.67 1460 0.71 74.42 58.6 4.7 2.10 39 1072 47.67 1230 1.62 83.7 56.7 5.3 2.33 175 Table 5 continued 40 1077 54.55 1375 1.61 91.81 63.8 5.8 2.21 41 1079 48.67 1280 1.71 90.87 63.9 5.5 2.74 42 1080 51.33 1370 1.87 92.33 59.1 5.5 2.56 43 1081 46.67 1395 1.93 76.84 67.3 4.9 2.28 44 1082* 37.50 1010 2.09 89.04 68.8 8.5 2.48 45 1084* 41.00 1105 1.96 90.9 83.3 6.1 2.44 46 1085* 37.00 1005 3.78 101.4 92.1 6.9 2.88 47 1090* 47.00 1295 1.45 86.6 82.9 6.5 2.26 48 1091 48.67 1355 2.00 81.72 77.2 6.7 2.14 49 1094* 45.00 1155 2.22 81.7 72.4 5.4 2.48 50 1096* 42.00 1160 2.06 76.4 70.4 8.4 2.91 51 1098* 47.00 1125 1.22 80.9 78.4 10.3 2.64 52 1099* 40.00 1150 2.68 79.2 83.2 9.7 2.10 53 1105* 42.00 1180 2.17 65.2 65.2 7.0 2.10 54 1106 49.67 1380 1.57 81.16 79.1 5.1 2.39 55 1107* 47.00 1120 2.06 78.6 78.6 7.7 2.37 56 1111 47.87 1345 1.06 84.92 69.0 3.5 2.43 57 1114* 43.00 1160 2.03 70.6 70.6 7.7 2.11 58 1115* 52.00 1160 2.16 72.3 72.3 6.6 2.19 59 1117* 52.00 1110 1.53 82.3 82.3 7.4 2.20 60 1121 49.67 1300 1.16 87.34 69.0 6.4 3.03 61 1122 54.67 1420 1.19 90.17 72.4 5.9 2.03 62 1127 53.33 1375 1.37 109.61 60.0 6.2 2.69 63 1131* 48.00 1195 1.56 82.44 55.1 5.8 2.55 64 1132 58.33 1550 2.01 67.86 54.2 4.8 2.29 65 1136 58,45 1560 1.18 85.87 70.5 2.8 2.71 66 1137 57.67 1600 2.02 95.52 50.3 4.8 2.78 67 1147 59.33 1560 0.60 81.80 40.7 4.6 2.10 68 1151 51,45 1335 0.80 74.52 60.3 4.7 2.59 69 1156 55.67 1430 1.06 81.7 43.4 5.8 2.04 70 1157 61.33 1560 0.61 72.24 51.4 4.2 2.18 71 1158 61.67 1560 0.83 69.29 51.1 5.9 2.21 72 1159 58.33 1525 1.09 90.45 59.4 5.8 1.71 73 1160 49.87 1200 2.02 96.28 70.4 5.1 2.04 74 1161 55.63 1420 1.19 50.73 56.4 6.4 2.53 75 1164 50.33 1170 1.94 90.82 66.2 4.6 2.96 76 1165 59.25 1490 0.82 65.48 49.1 5.5 2.32 77 1166 62.33 1600 0.85 71.76 38.4 4.7 3.24 78 1167 53.67 1465 0.79 96.68 70.0 6.5 2.20 79 1168 50.67 1320 0.57 68.05 55.2 4.1 2.81 80 1172 51.33 1310 1.10 79.62 53.2 6.9 2.41 176 Table 5. Continued 81 1173 57.33 1455 0.71 60.64 53.1 6.2 2.29 82 1174 53.33 1345 1.32 89.63 61.1 6.6 2.19 83 1175 58.88 1470 0.65 69.80 51.6 5.2 2.27 84 1176 60.33 1550 0.84 71.08 58.9 5.5 2.16 85 1178 54.67 1380 1.45 80.63 50.8 6.5 2.04 86 1179 57.25 1440 1.59 99.2 58.2 4.1 2.19 87 1181 55.33 1430 1.65 92.63 61.1 6.0 2.17 88 1182 55.67 1450 2.03 79.69 57.0 5.6 2.10 89 1183 50.67 1300 1.88 66.61 61.9 6.3 2.64 90 1184 60.33 1530 1.40 71.5 56.7 7.0 2.18 91 1185 64.25 1720 1.34 72.04 46.2 5.1 2.29 92 1186 61.33 1610 1.01 85.12 58.7 6.3 2.45 93 1188 53.67 1440 1.24 82.45 56.5 5.8 2.57 94 1189 51.87 1180 1.13 90.44 63.2 6.2 2.23 95 1190* 46.00 1195 2.56 79.3 69.8 8.0 3.00 96 1191 59.33 1660 0.80 69.24 55.9 6.2 2.63 97 1192 65.33 1720 0.49 56.82 48.0 5.3 2.00 98 1193 54.33 1420 1.38 90.97 57.4 6.1 2.00 99 1194 53.67 1440 0.97 67.32 56.4 4.2 1.95 100 1195 48.67 1175 1.38 73.56 63.9 6.8 2.11 101 1207 49.87 1195 1.39 82.35 61.8 4.2 2.50 102 1213 51.67 1405 1.17 95.56 60.5 6.6 2.19 103 1215 52.88 1255 1.28 60.09 80.4 7.6 2.47 104 1216* 40.00 1070 2.49 91.3 91.4 7.3 2.68 Mean 51.74 1335 1.4 79.68 64.5 6.1 2.38 Std Dev 6.23 182.58 0.61 13.87 11.5 1.5 0.30 Check variety 1 Degaga 52.30 1352.5 1.64 55.3 65.4 6.5 2.3 2 Gachena 49.33 1326.3 2.18 73.4 70.3 6.3 2.1 3 Gora 46.00 1240 2.50 78.3 71.9 5.9 2.3 4 Local 58.75 1363.8 1.62 48.1 71.4 7.6 2.3 Mean 51.60 1320.6 1.99 63.8 69.8 6.6 2.3 Cv (%) 3.51 3.39 11.2 2.28 8.26 25.9 10.6 LSD (0.05) ns * ** *** ns ns ns AUDPC = Area under disease progress curve. GY = grain yield; HSW = hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. StDev = Standard devation and R2 =coefficient of determination. * Represents genotypes selected for low FBG disease severity and AUDPC. 177 Table 6. Faba bean genotypes and breeding lines selected for low FBG disease severity and AUDPC at Bassona Worana district, Ethiopia in 2018 and 2019 main cropping seasons. 2018 2019 Genotype/ FBG severity (%) AUDPC FBG severity (%) AUDPC breeding line Initial Final (%-day) Initial Final (%-day) 26884 15.00 46.00 1315 6.50 40.50 1110 1002 3.00 44.00 1010 8.50 42.00 1030 1004 5.00 38.00 1020 10.00 38.00 1025 1005 5.50 50.00 1140 10.00 47.00 1205 1006 10.00 46.00 1230 8.50 48.00 1210 1009 4.00 40.00 950 7.00 41.00 1030 1025 5.00 44.00 980 6.00 52.00 1090 1030 10.00 38.00 1110 5.50 47.00 1045 1031 5.00 40.00 1050 5.50 48.00 1155 1032 11.50 44.00 1350 5.50 47.00 1035 1034 13.00 45.00 1285 5.50 42.00 1055 1045 5.50 40.00 850 7.00 52.00 1060 1047 7.00 45.00 1275 12.00 47.00 1195 1082 18.00 45.00 975 10.00 37.50 1010 1084 12.00 37.00 1155 7.00 41.00 1105 1085 10.00 39.00 935 7.50 37.00 1005 1090 13.00 50.00 950 6.50 47.00 1295 1094 5.00 43.00 1200 10.00 45.00 1155 1096 5.00 44.00 1160 10.00 42.00 1160 1098 7.50 47.00 1175 7.00 47.00 1125 1099 17.00 37.00 1365 7.00 40.00 1150 1105 8.00 38.00 1070 10.00 42.00 1180 1107 16.00 46.00 1385 10.00 47.00 1120 1114 10.00 32.00 940 6.50 43.00 1160 1115 7.00 45.00 1025 6.50 52.00 1160 178 Table 6 continued 1117 8.50 46.00 1040 5.50 52.00 1110 1131 13.50 47.00 1225 7.00 48.00 1195 1190 12.00 48.00 1260 6.50 46.00 1195 1216 18.00 42.00 1390 5.50 40.00 1070 Mean 9.50 43.20 1141.0 7.67 44.63 1133.83 Check: Degaga 17.70 58.80 1640 8.00 52.30 1352.5 Gachena 16.60 56.00 1622 9.88 49.33 1326.25 Gora 11.13 54.00 1523 8.25 46.00 1240 Local 18.88 62.10 1909 9.25 58.75 1363.75 Mean 16.08 57.35 1681 8.85 51.60 1320.63 Cv (%) 11.32 5.08 7.06 12.05 3.51 3.39 LSD (0.05) * *** *** ns ns * 3.3. Disease Progress Rate High disease progress rate was calculated for the genotypes and breeding lines in 2018 cropping season (data not shown). In 2019 cropping season, among the tested 104 local genotypes and breeding lines, one genotype (local collection-26884) and 28 breeding lines showed low disease progress rate. The disease progress rate on selected genotypes and breeding lines ranged from 0.0544 to 0.0874 unit days-1. High disease progress rate (0.0874 unit days-1) was calculated for breeding line 1117 and low disease progress rate (0.0544 unit days-1) was calculated for breeding line 1107. Low disease progress rates (0.0544, 0.0561, 0.0562, 0.0574, and 0.0578, 0.0581 and 0.0588 unit days-1) were computed for breeding lines 1107, 1004, 1047, 1085, 1082, 1002 and genotypes number 28884, respectively. Disease progress rate on the check varieties were 0.0641, 0.0558, 0.0411 and 0.0752 unit days-1 for Degaga, Gachena, Gora and local cultivar, respectively (Table 7). 179 Table 7. Disease progress rates calculated for selected faba bean genotypes at Bassona Worana, North Shoa, Ethiopia, in 2019 main cropping season. Genotype / Severity (%) DPR SE of Intercept SE of R2 breeding line Initial Final (units day-1) DPR intercept 26884 6.50 40.50 0.0588 0.0072 -3.725 0.361 73.11 1002 8.50 42.00 0.0581 0.0080 -3.934 0.417 92.76 1004 10.00 38.00 0.0561 0.0097 -3.814 0.503 89.06 1005 10.00 47.00 0.0621 0.0061 -4.272 0.319 96.21 1006 8.50 48.00 0.0683 0.0032 -4.335 0.421 67.32 1009 7.00 41.00 0.0658 0.0037 -4.484 0.192 98.76 1025 6.00 45.67 0.0831 0.0059 -5.220 0.309 97.98 1030 5.50 47.00 0.0773 0.0022 -5.144 0.115 99.67 1031 5.50 48.00 0.0810 0.0087 -5.192 0.452 95.54 1032 5.50 47.00 0.0786 0.0036 -5.182 0.189 99.15 1034 5.50 42.00 0.0747 0.0025 -5.151 0.132 99.54 1045 7.00 52.00 0.0796 0.0106 -4.845 0.551 93.26 1047 12.00 47.00 0.0562 0.0071 -3.497 0.371 93.85 1082 10.00 37.50 0.0578 0.0106 -4.081 0.551 87.80 1084 7.00 41.00 0.0650 0.0085 -4.372 0.441 93.51 1085 7.50 37.00 0.0574 0.0090 -4.033 0.468 90.83 1090 6.50 47.00 0.0779 0.0121 -4.863 0.628 91.02 1094 10.00 45.00 0.0663 0.0033 -5.211 0.446 74.11 1096 10.00 42.00 0.0605 0.0072 -4.074 0.374 94.58 1098 7.00 47.00 0.0715 0.0041 -4.662 0.214 98.68 1099 7.00 40.00 0.0626 0.0115 -4.249 0.596 87.80 1105 10.00 42.00 0.0613 0.0096 -3.993 0.498 90.88 1107 10.00 47.00 0.0544 0.0034 -4.453 0.346 59.01 1114 6.50 43.00 0.0711 0.0099 -4.654 0.515 92.66 1115 6.50 52.00 0.0768 0.0062 -4.896 0.320 97.47 1117 5.50 52.00 0.0874 0.0109 -5.471 0.564 94.10 180 Table 7 continued 1131 7.00 48.00 0.0707 0.0094 -4.554 0.490 93.25 1190 6.50 46.00 0.0747 0.0086 -4.784 0.447 94.90 1216 5.50 40.00 0.0669 0.0155 -4.609 0.805 81.50 Degaga 8.00 52.30 0.0641 0.0078 -5.372 0.413 96.83 Gachena 9.88 49.33 0.0558 0.0099 -5.227 0.512 94.92 Gora 8.25 46.00 0.0411 0.0084 -4.395 0.464 93.98 Local 9.25 58.75 0.0752 0.0092 -4.633 0.477 94.29 3.4. Grain Yield and Yield Components In 2018 cropping season, a significant difference was obtained on the grain yield of local genotypes and introduced breeding lines as compared to check varieties (Table 8). Among 415 genotypes and breeding lines tested, 119 genotypes obtained from EBI showed less than 0.25 t ha-1 grain yields. Two hundred ninety-six genotypes and breeding lines showed a grain yield of 0.36 t to 2.7 t ha-1 (data not shown). In 2019 cropping season, grain yield ranged from 0.4 to 3.78 t ha-1. Among the tested genotypes, 28 breeding lines showed a grain yield of 0.5 to 0.97 t ha-1. Five genotypes obtained from EBI and 44 introduced breeding lines showed 1 to 1.9 t ha-1 and 27 breeding lines showed 2 to 3.78 t ha-1 grain yields. Maximum (3.78 t ha-1) grain yield was recorded from breeding line 1085, which was from introduced genotypes. In 2018 cropping season, check varieties showed 1.37, 2.07, 2.25 and 1.02 t ha-1 grain yields from Degaga, Gachena, Gora and local cultivar, respectively. Similarly, in 2019 cropping season, 1.64, 2.18, 2.50 and 1.62 t ha-1 grain yields were obtained from Degaga, Gachena, Gora and local cultivar, respectively. In 2018 cropping season, hundred seed weight from genotypes and breeding lines tested ranged from 34.5-104.7 g. The highest (104.7 g) hundred seed weight was recorded from introduced breeding lines. In 2019, hundred seed weight showed a range of 38.97 to 111.5 g. Three breeding lines showed greater than 100 g hundred seed weight. Plant height range in 2018 was 49.9 to 92.6 cm and 38.4 to 92.1 cm in 2019 cropping season. The number of pods per plant range in 2018 was 4.6 to 13.6 and 2.8 to 10.3 in 2019 cropping season. Similarly, number of seeds per pod range in 2018 was 1.9 to 3.6 and 1.51 to 3.25 in 2019 cropping season (Table 8). 181 Table 8. Mean grain yield and yield components of selected faba bean genotypes and breeding lines in Bassona Worana, North Shoa, Ethiopia in 2018 and 2019 cropping seasons. Genotype/ GY (t ha-1) HSW (g) PH (cm) No. of Pod No. of Seed pod-1 breeding lines plant-1 2018 2019 2018 2019 2018 2019 2018 2019 2018 2019 26884 1.39 1.73 46.7 45.0 69.6 72.5 8.4 6.5 3.2 2.91 1002 2.70 1.87 98.2 90.4 74.8 85.1 8.6 6.1 2.8 2.24 1004 2.28 2.47 62.9 80.3 77.2 81.9 7.8 7.7 2.6 2.41 1005 2.36 2.32 75.2 80.0 77.2 83.1 5.0 6.7 2.9 2.95 1006 2.09 2.26 79.5 97.5 84.6 84.2 8.4 7.3 3.0 2.52 1009 2.70 2.74 83.2 91.4 85.0 85.8 13.6 10.3 2.9 2.12 1025 2.12 1.96 90.9 93.9 64.4 68.6 9.8 8.9 2.7 2.18 1030 2.46 1.36 104.7 111.5 86.8 83.6 5.4 6.3 3.0 2.40 1031 1.37 1.38 83.4 92.6 73.0 63.0 6.8 7.7 2.9 2.44 1032 1.77 1.26 68.6 66.6 70.2 65.7 7.0 8.4 2.8 2.23 1034 1.97 1.35 95.2 98.1 92.6 79.7 11.4 10.2 2.8 2.27 1045 1.51 0.87 66.5 77.2 85.0 58.0 11.2 9.9 2.5 2.21 1047 1.41 0.97 84.6 80.9 80.2 78.7 8.4 7.9 3.6 2.33 1082 1.91 2.68 80.1 89.04 71.6 68.8 9.2 8.5 2.3 2.48 1084 1.56 1.96 93.3 90.9 75.4 83.3 8.4 6.1 2.7 2.44 1085 2.45 3.78 94.2 101.4 76.2 92.1 8.0 6.9 3.1 2.88 182 Table 8. Continued 1090 1.37 1.45 72.4 86.6 85.4 82.9 6.6 6.5 3.0 2.26 1094 2.10 2.22 89.1 81.7 89.2 72.4 4.8 5.4 3.0 2.48 1096 2.49 2.06 91.7 76.4 73.6 70.4 7.4 8.4 3.0 2.91 1098 1.36 1.22 70.6 80.9 84.0 78.4 12.2 10.3 3.3 2.64 1099 1.98 2.09 70.1 79.2 84.8 83.2 10.2 9.7 2.8 2.10 1105 2.12 2.17 64.4 73.6 64.8 65.2 9.4 8.6 3.1 2.10 1107 2.12 2.06 80.6 97.6 80.2 78.6 8.0 7.0 3.4 2.37 1114 1.75 2.03 81.7 90.3 80.8 70.6 6.8 7.7 2.6 2.11 1115 1.23 2.16 46.9 85.2 68.2 72.3 5.2 6.6 3.0 2.19 1117 2.99 1.53 79.3 72.1 71.4 82.3 8.4 7.4 3.1 2.20 1131 1.23 1.56 61.0 82.4 70.0 55.1 4.6 5.8 3.0 2.55 1190 2.24 2.56 72.2 79.3 74.0 69.8 7.0 8.0 3.1 3.00 1216 1.62 2.49 87.4 91.3 74.8 91.4 8.8 7.3 3.5 2.68 Mean 1.95 1.95 78.43 84.94 77.41 76.09 8.17 7.73 2.96 2.5 Checks: Degaga 1.37 1.64 50.9 55.3 66.9 65.4 7.5 6.5 2.7 2.3 Gachena 2.07 2.18 61.1 73.4 68.7 70.3 7.2 6.3 2.6 2.1 Gora 2.25 2.50 76.4 78.3 71.1 71.9 7.9 5.9 2.9 2.3 Local 1.02 1.62 50.5 48.1 64.7 71.4 5.3 7.6 2.4 2.3 Mean 1.68 1.99 59.7 63.8 67.9 69.8 6.9 6.6 2.7 2.3 Cv (%) 19.5 11.2 11.3 2.28 11.4 8.26 26.5 25.9 12.4 10.6 LSD (0.05) *** ** *** *** * ns ns ns ns ns 183 3.5. Association of FBG Disease and Yield Parameters The correlation coefficient analysis showed that there was a significant and strong association between FBG parameters, grain yield, and yield components. In 2019 cropping season, FBG severity showed positive strong (r = 0.89***) correlation with AUDPC and DPR (r = 0.97***). Faba bean gall severity showed negative strong correlation with plant height (r = -0.74***), number of pods per plant (r = -0.55***) and grain yield (r = -0.66**). However, hundred seed weight showed negative weak correlation (r = -0.19 ns) and number of seeds per pod (r = -0.17 ns). Area under disease progress curve showed negatively correlated with plant height (r = -0.69***), number of pods per plant (r = -0.69***) and grain yield (r = -0.53***), but showed negative weak correlation with hundred seed weight (r = -21*) and number of seeds per pod (r = -0.16 ns). Disease progress rate showed similar trends of correlation with FBG severity, AUDPC, grain yield and yield-components. On the contrary, grain yield attained positive relationship with plant height (r = 0.57***), number of pods per plant (r = 0.37***) and number of seeds per pod (r = 0.17ns) and hundred seeds weight (r = 0.29**). The relationship between plant height, number of pods per plant and number of seeds per pod and hundred seed weight were presented hereunder (Table 9). Table 9. Coefficients of correlation (r) and level of significance between FBG disease parameters, grain yield and yield components of faba bean genotypes and breeding lines at Bassona Worana, North Shoa Zone, Ethiopia, in 2019 cropping season. Parametera PSI AUDPC DPR GY HSW PH5 NPPP NSPP PSI 1 0.89*** 0.97*** -0.66*** -0.19 ns -0.74*** -0.55*** -0.17 ns AUDPC 0.89*** 1 0.92*** -0.53*** -0.21* -0.69*** -0.62*** -0.16 ns DPR 0.76** 0.75** 1 –0.87*** –0.35ns –0.66*** –0.71** –0.55* GY -0.66*** -0.53*** –0.80*** 1 0.29** 0.57*** 0.37*** 0.17 ns HSW -0.19 ns -0.219* –0.45ns 0.29** 1 0.24* -0.18 ns -0.13 ns PH -0.74*** -0.69*** –0.66*** 0.57*** 0.24 ns 1 0.49*** 0.13 ns NPPP -0.55*** -0.63*** –0.37ns 0.37*** -0.19 ns 0.49*** 1 -0.12 ns NSPP -0.17 ns -0.067 ns –0.75** 0.17 ns -0.13 ns 0.13 ns -0.12 ns 1 aPSI = Percent severity index, AUDPC = Area under disease progress curve; DPR = Disease progress rate; GY = Grain yield; HSW = Hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. *** = Very highly significant; ** = highly significant; * = Significant; and ns = Non-significant at p0.001, p0.01, p<0.05 and p>0.05, respectively. 184 4. Discussion The results of genotype and breeding lines evaluations for FBG resistance in 2018 and 2018 cropping season indicated that FBG incidence and severity were very high and the test genotypes showed significant differences in FBG parameters when compared with check varieties. Most of local genotypes obtained from Oromia, Amhara, Southern Nations, Nationalities and People’s Region (SNNPR) and Tigray Regions through EBI showed high FBG severity in 2018 cropping season. Very few genotypes and breeding lines showed low FBG severity. Low FBG severity on few genotypes and breeding lines could be due to differences in genetic variations. Ethiopia has different agro-ecologies and it is expected various genetic diversity of faba bean genotypes growing in different parts of the country. Similarly, many cultivars are released for different disease tolerant and grain yield from local landraces and introduced genotypes. However, there is no variety developed for resistance to FBG. Most of released varieties and farmers’ local cultivars are susceptible to FBG (Teklay et al., 2014; DBARC, 2015; Wulita, 2015). In the current study, local genotypes showed high FBG severity. The reason for high FBG severity could be due to several factors. Among these factors, most of genotypes from Ethiopia might have the same origin. Moreover, most of the genotypes obtained from different parts of the country might have originally been introduced from the same source, followed by frequent exchange of seeds among farmers from neighboring regions of the country. Gemechu et al. (2005) reported that there is a tendency, particularly among resource-poor farmers in marginal areas, of selecting for high grain yield and resistance to different common diseases, including exchanging seeds among growers and these may contribute to disease distribution. There are different reports for some disease and grain yield that original sources of landraces might vary, due to forced genetic change by local breeding programs and releasing different varieties for different areas (Tamene et al., 2015). Genotypes from the same origin might have different genetic background or genotypes from different regions might have similar genetic background. Similarly, geographic diversity may not necessarily be the reason for genetic diversity, but sources or parental background could be the reason for variation (Gemechu et al., 2005). The reason for high FBG severity among genotypes could be due to similar genetic background as reported by Gemechu et al., (2005). Introduced breeding lines showed significant differences in FBG parameters in both cropping seasons. Among the tested introduced breeding lines, a few of 185 them showed low FBG severity and AUDPC values. Differences in FBG severities and AUDPC values among introduced breeding lines could be due to genetic variability like that of local genotypes. Different reports showed that faba bean varieties, genotypes and elite breeding lines showed different degrees of reaction to FBG (Yan, 2012; Alehegn et al., 2018; Negussie et al., 2018). The findings in current study are in conformity with the observations of some authors and organizations (Yan, 2012; DBARC Annual Progress Report, 2015; Negussie et al., 2018) who reported that different genotypes and breeding lines showed different degrees of FBG reactions. Very few local genotypes and breeding lines selected from 2018 cropping season and evaluated in 2019 showed low FBG severity and high grain yield as compared to the check faba bean varieties. Low FBG severity and high grain yield of and breeding lines could be due to their genetic background. The current study revealed that host resistance can be achieved through evaluation of different genotypes and breeding lines obtained from different sources. The results indicated that further genotypes and breeding lines evaluation for FBG resistance and other agronomic traits should focus not only across regional states, but should be also on introduced genotypes. It is also important to evaluate genotypes collected within a region at different agro-ecologies. Field screening is a common practice used to evaluate faba bean genotypes against different diseases of faba bean (Tivoli et al., 2006); however, screening for FBG resistant should focus on many genotypes having different sources of genetic background. 5. Conclusions Field experiments conducted in 2018 and 2019 main cropping seasons on 415 faba bean genotypes and breeding lines showed that FBG disease infected most of the genotypes, and few breeding lines showed low FBG severity and high grain yield. Among genotypes and breeding lines tested in 2019, one genotype (local collection-26884) and 28 breeding lines showed low FBG severity and low AUDPC values. Low mean FBG disease parameters were recorded on breeding line number 1085, 1082 and 1004. Maximum (3.78 t ha-1) grain yield was recorded on breeding line number 1085. Genotype number 26884 and 28 breeding lines, which showed low FBG severity, AUDPC values and genotypes with good grain yield was selected for further evaluation and crossing purpose. 186 6. Acknowledgements The study was financially supported by Amhara Regional Research Institute (ARARI), Australian Center for International Agricultural Research (ACIAR) and the International Center for Agricultural Research in the Dry Areas (ICARDA), Africa RISING/USAID through ICARDA. Hence the institutes are duly acknowledged for supporting the study. This piece of work is part of a PhD Dissertation research requirement at Haramaya University and the University is duly acknowledged. We also greatly acknowledge, Mr Aleme Belete and Amare Belachew for their assistances during field data collection. 7. References Agegnehu, G., Ghizaw, A. and Sinebo W. 2006. Yield performance and land-use efficiency of barley and faba bean mixed cropping in Ethiopian highlands. European Journal of Agronomy, 25: 202–207. Alehegn, M., Melekamu, T. and Meseret, T. 2018. Screening of faba bean (Vicia faba) varieties against Faba bean Gall Disease (Olpidium viciae) in East Gojjam Zone, Ethiopia. Journal of Biology, Agriculture and Healthcare, 8: 50–55. Asfaw, T., Beyene, D. and Tesfaye, G. 1994. Genetic and breeding of field pea. First National Cool-season Food Legumes Conference. Addis Ababa, Ethiopia. ICARDA/ IAR. ICARDA: Aleppo, Syria. pp. 122–137. Beyene, B. 2015. Survey and identification of new faba bean disease (Qormid) in the highlands of North Shewa, Ethiopia. Current Research Microbiology and Biotechnology, 3: 561– 563. Birru, A. 1979. Agricultural field experiment management manual part III. Institute of Agricultural Research. Addis Ababa, Ethiopia. pp. 35-42. Bogale, N., Setu, B., and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae Kusano) in major faba bean growing areas of northeastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 110: 87 – 95. Campbell, C. L. and Madden, V. L. 1990. Introduction to Plant Disease Epidemiology, New York: John Wiley and Sons, Inc. New York, USA. 532 pp. 187 CSA (Central Statistical Agency). 2021. Agricultural sample survey (2020/2021). Report on area and production of major crops. Statistical Bulletin 578, Central Statistical Authority, Addis Ababa, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2005. Debre Birhan Agricultural Research Center, Sub Center and Research sites description. Debre Birhan, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2015. Progress report unpublished. Dereje, G. Wendafrash, M. and Gemechu, K. 2012. Faba Bean Galls: A New Disease of Faba Bean in Ethiopia. Holetta Agricultural Research Center. Available at: https://docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE/edit. Dereje, G. 1999. Survival of Botrytis fabae Sard. between seasons on crop debris in field soils at Holetta, Ethiopia. Phytopathologia Mediterranean, 38: 68–75. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean Newsletter, 32: 8–10. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N., Bitew, B., Boydom, A, Kassa D, and Temesgen T. 2014. Faba bean gall; A new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Science and Technology. 2(4): doi.org/10.4172/2329-8863.1000144. FAOSTAT (Food and Agriculture Organization of the United Nations, Statistical Database). 2018. Rome, Italy, website.faostat.fao.org. Gemechu, K., Mussa, J. Tezera, W. and Getnet, D. 2005. Extent and pattern of genetic diversity for morpho-agronomic traits in Ethiopian highland pulse landraces II. Faba bean (Vicia faba L.). Genetic Resources and Crop Evolution, 52: 551–561. Getaneh, G., Hailu, E., Kassahun, S., Tesfaye, A. and Girma, M. 2018. The causal pathogen, inoculum sources and alternate hosts studies of the newly emerged gall forming faba bean (Vicia faba) disease in Ethiopia. Advances in Crop Sciences and Technologhy, 6(3): 368. DOI:10.4172/2329-8863.1000368. Jensen, E.S., Peoples, M.B., and Hauggaard-Nielsen, H., 2010. Faba bean in cropping systems. Field Crops Research, 115: 203-216. Kusano, S. 1912. The life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of the College of Agriculture, 4: 142–199. 188 Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. 48 pp. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. Negussie, T., Seid, A., Gemechu, K., Wulita, W., Beyene, B., Asnakeh, T., Anteneh, A. and Yohannes, E. 2018. Faba bean gall in Ethiopia. Current situation and possible solutions. Seventh International Food Legume Research Conference, Marrakesh, Morocco. Safa, G. k. and Khaled M. M. 2007. Virus Diseases of Faba Bean in Asia and Africa. Plant Virus, 1: 93-105. Sahile, S., Chemeda, F., Sakhuja, P.K. and Seid, A. 2008. Effect of mixed cropping and fungicides on chocolate spot (Botrytis fabae) of faba bean (Vicia faba) in Ethiopia. Crop Protection, 27: 275–282. SAS (Statistical Analysis System). 2014. SAS User’s Guide. Version 9.2.4. SAS Institute. Cary, North Carolina, USA. Tamene, T., Gemechu, K. and Hussein, M. 2015. Genetic progresses from over three decades of faba bean (Vicia faba L.) breeding in Ethiopia. Australian Journal of Crop Science, 9: 41– 48. Teklay, A., Tsegaye, B.. Ymane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Science, 2: 33–38. Terefe, H., Fininsa. C., Sahile, S. and Tesfay, K. 2015. Effect of temperature on growth and sporulation of Botrytis fabae, and resistance reactions of faba bean against the pathogen. Journal of Plant Pathology and Microbiology, 6: 285–268. Tivoli, B., Baranger, A., Avila, C.M., Banniza, S., Barbetti, M.C.W., Davidson, J., Lindeck, K., Kharrat, M., Rubiales, D., Sadiki, M., Sillero, J.C., Sweetingham, M. and Muehlbauer, F.J. .2006. Screening techniques and sources of resistance to foliar diseases caused by major necrotrophic fungi in grain legumes. Euphytica, 147: 223–253. Van der Plank, J.E. 1963. Plant diseases: Epidemics and control. Academic Press, New York. 334 pp. Wheeler, B.J. 1969. An introduction to plant diseases. John Wiley and Sons Ltd., London, 374pp. 189 Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) Using varieties and fungicides in north Shoa, central Ethiopia. Msc Thesis, Haramaya University, Haramaya, Ethiopia. Xing, Z.S. 1984. Preliminary study oiler bacteria blister prevention and treatment of disease blister beans. Plant Pathology, 114: 165–173. Yan, J.M. 2012. Study on Blister Disease of Broad Bean Caused by Olpidium Viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. You, M.P., Eshete, B.B., Kemal, S.A., Van Leur, J. and Barbetti, M.J. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology, 70(5): 1180-1194. 190 Paper V AMMI and GGE biplot Models Reveal High Genotype Contribution for Gall Disease Resistance Reaction and Grain Yield Stabilities in Faba Bean Beyene Bitew 1*, Chemeda Fininsa 2, Habtamu Terefe 2, Martin Barbetti 3, and Seid Ahmed 4 1Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 1*Corresponding author, E-mail: beyenebitew@yahoo.com; Mobile: +251-(0)9 11 33 87 89 2School of Plant Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 3School of Plant Biology, Faculty of Science, the University of Western Australia, Australia 4Biodiversity and Crop Improvement Program, International Center for Agricultural Research in the Dry Areas (ICARDA), B.P. 6299, Rabat, Morocco Abstract Faba bean gall (FBG), is an emerging and devastating disease of faba bean in central, northern and northwestern areas of Ethiopia. Faba bean genotypes have been reported to show different responses to FBG infections. However, the resistance reactions of faba bean genotypes to FBG disease are not yet evaluated and documented across multiple locations in Ethiopia. Field experiments were conducted at Bassona Worana, Debay Telatgen and Farta districts in the 2018 and 2019 main cropping seasons to identify the phenotypic reactions of genotypes and determine the resistance reactions stability against FBG disease under natural infections across locations. A total of 21 faba bean genotypes were evaluated in randomized complete block design with three replications. Disease parameters and grain yield data were collected during the study periods. Additive main effect and multiplicative interaction (AMMI) and genotype and genotype x environment (GGE) biplot models were used to evaluate genotype reactions. The AMMI and GGE biplot analyses revealed highly significant ((P<0.001) differences among genotypes, environments and genotype x environment interactions (GEI) for FBG disease incidence. The genotypes contributed (80.32%) to the total variation observed far larger than the contributions from the environment (5.32%) and GEI (14.36%) for FBG disease incidence. Genotypes also showed the largest (55.84%) contributions, followed by environment (37.83%) and GEI (6.33%) to the variability demonstrated for FBG disease severity. Similarly, the variation in grain yield was highly attributed to genotypes (50.86%), followed by environment (38.53%) and GEI, which contributed about 10.61% of the variation for grain yield. Both in the AMMI and GGE biplot 191 analysis, G16, G17 and G3 showed low FBG disease severity and less stable, but G7 showed low FBG severity and stable. On the other hand, G16, G17, G3, G7, G4 and G8 showed high mean grain yield, and G8 was relatively stable. However, G1, G21, G13, G2 and G5 were susceptible at all test locations. The site Bassona Worana (E1) provided the best discriminating ability for the genotypes against FBG disease severity. G16, G17, G3, G7 and G8 which showed low FBG disease severity reaction and high grain yield are recommended. Keywords: AMMI model, Disease severity, Environments, GGE biplot, Grain yield, Genotypes, Stability. 1. Introduction Faba bean (Vicia faba L.) is a major cool season legume crop grown in many countries (Torres et al., 2006; Duc et al., 2010; Jensen et al). Ethiopia is the largest producer of faba bean next to China (FAOSTAT, 2018). Faba bean mainly grows in the mid and high altitude areas of the country with an elevation ranging from 1800–3000 meters above sea level (Gemechu et al., 2006), and best grows in nitosols, cambisol and vertisol types in Ethiopia (Gemechu and Mussa, 2002). Faba bean is a leading pulse crop grown next to cereals and takes the largest share of the area and production of all the pulses grown in Ethiopia with a current productivity of 2.12 t ha–1; and over 4.1 million households grow the crop covering closely 0.5 M ha and a total production of more than one million tons of grain in the main cropping season in the country (CSA, 2021). Faba bean is valuable as a cheap source of protein for human food, good sources of cash to the farmers and generate foreign currency to the country. Straw of the crop is a key animal feed in the crop-livestock farming systems in the midland and highlands of Ethiopia. The crop also plays an important role in soil fertility restoration as component of crop rotation practices, which contributes in reducing carbon footprints of cereal-based agricultural production (Agegnehu and Fessehaie, 2006). Despite the fact that the crop has production potentials, several benefits to the farming communities and tremendous genetic diversity for pest and agronomic traits, faba bean productivity is very low in Ethiopia (Sahile et al., 2008; Terefe et al., 2015). Such low production and productivity of the crop could be attributed to many biotic and abiotic stresses, and as well as lack of improved varieties and poor agronomic practices (Mussa et al., 2008; 192 Tadesse et al., 2008). Among biotic stresses, fungal diseases namely, chocolate spot (Botrytis fabae), ascochyta blight (Ascochyta fabae), rust (Uromyces viciae-fabae), zonate leaf spot (Cercospora zonatae), and black root rot (Fusarium solani F. avenaceum) have been the major limiting factors of faba bean productivity in the country (Dereje, 1999; Sahile et al., 2008; Terefe et al., 2015). In addition to the already documented diseases of the crop, major faba bean- producing areas, especially, central, northern and northwestern parts of the country are seriously threatened by an emerging and more devastating faba bean gall (FBG) disease. Similar gall causing disease, later named as ''broad bean blister disease'', was reported as a major disease on faba bean in Japan, and Sichuan and neighboring provinces of China; and the disease noticed to be caused by Olpidium viciae (Kusano, 1912; Xing, 1984; Yan, 2012) though Physoderma is identified as the true causative agent of FBG disease in Ethiopia (You et al., 2021). Previously, FBG was localized in North Shoa of Amhara and Oromia Regional States and South Tigray (Dereje et al., 2012; Hailu et al., 2014; Beyene, 2015; Bogale et al., 2016). However, the disease continued to expand to other faba bean production areas of Amhara Regional State, and West Shoa and East Wollega zones of Oromia Regional State through time (Challa et al., 2017), implying to cause huge production challenge all over the country. For instance, during severe infections and repeated cultivation of susceptible varieties, total crop failure was observed on farmers' field and farmers had been forced to replace faba bean sown fields with other early maturing crops (Teklay et al., 2014). In Ethiopia, even though FBG disease is overstretching to several faba bean growing areas and causing major devastation to faba bean production and productivity, the efforts in managing the gall disease have not been well coordinated and knowledge-based; hence, smallholder farmers, research centers and other stakeholders are solely dependent on few available fungicides as solution to contain the disease and sustain productivity. On the other hand, the majority of released faba bean varieties, and cultivars and landraces tested are not resistant to FBG disease though showed different levels of reaction (Wulita, 2015; DBARC, 2015). However, there is lack of comprehensive evaluation of all released faba bean varieties or elite breeding lines to FBG disease across the country. 193 Reports confirmed that some faba bean varieties released in the country for different agro- ecological production areas showed inconsistent reaction against FBG disease. In this regard, Wulita (2015) and Getenet and Yizblem (2017) evaluated few similar faba bean varieties at North Shoa and South Gondar, and recommended three different tolerant varieties for each zone for the disease. In contrast, Alehegn et al. (2018) screened seventeen faba bean genotypes and reported that majority of the released varieties were resistant to FBG disease in East Gojjam. Contrasting reports on the same faba bean varieties could demonstrate that the reactions of the varieties were not well evaluated across contrasting and multi-environments over years that highlight the need for further studies to identify the reactions and stability of resistance/susceptible reactions. Multi-location testing of genotypes can help to identify disease resistance or susceptibility reaction stability (Xu, 2010). It is also important to categorize released varieties according to their reaction classes (Sharma et al., 2015). In addition, testing released faba bean varieties for the reaction stability at multi-environment is pertinent to determine not only the genetic potential of the varieties, but also variation in pathogen virulence (Sharma et al., 2012). In Ethiopia, a number of improved faba bean varieties have been developed and released for general production purposes and for major diseases under different recommendation domains (Tamene et al., 2015); however, there are less or no genotype, environment and genotype by environment interaction studies conducted to know the reaction stability of faba bean varieties against FBG disease. Therefore, the objectives of the study were to identify the phenotypic reactions of faba bean genotypes and determine the resistance reactions stability of the genotypes against FBG disease under different environments in Ethiopia. 2. Materials and Methods 2.1. Description of the Study Areas Field experiments were conducted on farmers’ field at Basona Worana (North Shoa), Debay Telatgen (East Gojjam) and Farta (South Gondar) districts of Ethiopia for two consecutive main cropping seasons. That is, the study was performed at six locations during the 2018 and 2019 main cropping seasons under natural infections. The testing locations are the main faba bean growing areas and different in temperature, rainfall and soil pH features. Also, the test locations are known hot spot areas for FBG disease epidemics. Soil texture, pH and specific location 194 altitude are presented in table 1. Basona Worana district is located at 9°41' N latitude and 39o31' E longitude and an altitude of 2980 meter above sea level (m.a.s.l.). The district receives bimodal rainfall, where the main rainy season covers June to September and the short rainy season starts in January and ends in May. The rainfall in the main rainy cropping season reaches 1000 to 1200 mm and the mean minimum and maximum temperatures are 6 oC and 19 oC, respectively (DBARC, 2005). The daily maximum and minimum temperatures (oC), relative humidity (%) and total rainfall (mm) in 2018 and 2019 for Bassona Worana district were obtained from Debre Birhan Agricultural Research weather station (Figure 1). The relative humidity range was 49.5 to 90.5% in 2018 and 58.5 to 90.5% in 2019. Debay Telatgen district is located at 10°45' N latitude and 37°50' E longitude with an altitude of 2600 m.a.s.l. and the annual rainfall was 900 mm. The annual mean maximum and minimum temperatures were 9.9 °C and 23.6 °C in 2018 and 10.6 °C and 22.9°C in 2019. Moreover, Farta district is located at 12°00' N latitude and 30°00' E longitude with an altitude of 2800 m.a.s.l. The relative humidity ranged 65 to 96% and 74 to 82% in 2018 and 2019, respectively. Rainfall data for both cropping seasons at Debay Telatgen and Farta were obtained from national meteorology agency, Addis Ababa, Ethiopia (Figure 2). Table1. Composite soil samples taken at 20 cm depth for soil pH and texture analysis for faba bean genotype evaluation against FBG disease at six locations in 2018 and 2019 main cropping seasons District Specific Year Altitude pH Texture Texture location (m.a.s.l.) % sand % clay % silt class Bassona Mush 2018 2966 6.44 24 46 30 Clay Worana Gudoberet 2019 3039 6.80 4 76 20 Clay Debay Asendabo 2018 2638 5.38 26 42 32 Clay Telatgen Kuy 2019 2650 6.26 22 46 32 Clay Farta Gassay 2018 2809 5.45 26 36 38 Clay loam Tiratir 2019 2844 6.07 14 54 32 Clay 195 500 A 35 450 30 400 350 25 300 20 250 200 15 150 10 100 5 50 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax 500 B 35 450 30 400 350 25 300 20 250 200 15 150 10 100 5 50 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax Figure 1. Monthly mean minimum (Tmin) and maximum (Tmax) temperatures (°C) and rainfall (RF) (mm) data at Bassona Worana district of Northern Shoa, Ethiopia in 2018 (A) and 2019 (B). 196 Rainfall (mm) Rainfall (mm) Temperature (oC) Temperature (oC) 350 A 300 250 200 150 Yr-2018 100 Yr- 2019 50 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months 500 B 400 300 200 Yr-2018 100 Yr- 2019 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months Figure 2. Rainfall (mm) data at Debay Telatgen (A) and Farta (B) districts, Ethiopia in 2018 and 2019 cropping season. 2.2. Treatments, Experimental Design and Procedures Thirteen released faba bean varieties, which had different FBG disease reactions, seven elite breeding lines collected from research centers, and one farmers' local cultivar for each respective district, were used as treatment in the experiment and grouped as genotypes (Table 2). Each genotype was planted as an experimental unit with a total of 21 treatments per replication and the experiment was laid out in a randomized complete block design with three replications. A plot size of 4.8 m2 (2.4 m × 2 m) with 6 rows was used in the study. Spacing between blocks and plots were 1.5 m, while spacing between rows and plants were 0.4 m and 0.1 m, respectively. Planting was done on 15–22 June of 2018 and 2019. NPS (compositions: 38 kg P2O5, 19 kg N and 7 kg S) blended fertilizer was applied once at the time of planting with the rate of 121 kg ha- 1. Seeding and fertilizer application was done manually throughout each row. Plots were hand weeded and other cultural practices were kept uniform at plot level across all locations. 197 Rainfall (mm) Rainfall (mm) Table 2. Description of faba bean genotypes evaluated for faba bean gall disease reaction at Bassona Worana, Debay Telatgen and Farta districts, Ethiopia in 2018 and 2019 cropping season S/N Genotype Pedigree Source Year of Reaction to Reference Release FBG 1 Wayu 089-5 Collection 2002 Unknown NA 2 Mosisa EH99047-1 Hybridization 2012 Unknown NA 3 Numan EH 06007-2 Hybridization 2016 Unknown NA 4 Ashebeka EH 01075-4 Hybridization 2015 Unknown NA 5 Shallo EH011-22-1 Hybridization 1999 Unknown NA 6 Didea EH01048-1 Hybridization 2014 Unknown NA 7 Gora EK 01024-1-2 Hybridization 2013 Moderately R Wondwosen (2015) 8 Gachena ETH91001-13-2 Hybridization 2008 Moderately R Wondwosen (2015) 9 Degaga R878-3 Introduction 2002 Moderately R Yetayeh and Yezbalem (2017) 10 Dosha Coll. 155/00-3 Collection 2009 Moderately R Alehegn (2018) 11 Tumsa Tesfa x ILB 4726 Hybridization 2010 Moderately R Alehegn l (2018) 12 Hachalu EH960091-1 Hybridization 2010 Moderately R Alehegn l (2018) 13 Wolki Bulga-70 x ILB4615 Hybridization 2008 Susceptible Observation 14 EH01005-1 Elite breeding line Hybridization NR Unknown NA 15 EH09028-3 Elite breeding line Hybridization NR Unknown NA 16 EH010008-5 Elite breeding line Hybridization NR Resistant Nigussie (2018) 17 EH010058-1 Elite breeding line Hybridization NR Resistant Nigussie (2018) 18 EH09017-5 Elite breeding line Hybridization NR Unknown NA 19 EH09021-1 Elite breeding line Hybridization NR Unknown NA 20 EH09046-3 Elite breeding line Hybridization NR Unknown NA 21 Local NA Local cultivar LC susceptible NA Source: MoARD, Variety Registration Book, Issue numbers 4-19; NA= not available; LC = local cultivar; NR= not released. Genotypes are coded G1-G21 in the text as indicated from 1-21in the table. 2.3. Disease Incidence and Severity Assessment First date of disease onset, disease incidence (number of plants infected per total number of plants inspected), and disease severity (leaf and stem area infected) were assessed during the experimental periods. Faba bean gall disease incidence and severity were recorded at 10 days interval starting from the first date of disease onset at 32 days after planting (DAP) in 2018 and at 28 DAP in 2019 at Bassona Worana. Similarly, disease assessment started at 25 and 29 DAP at Debay Telatgen and 32 and 35 DAP at Farta districts in the 2018 and 2019 cropping seasons, respectively. For disease incidence measurements, 20 plants from the central four rows of each 198 plot were randomly tagged prior to the appearance of disease symptoms, and percent disease incidence was calculated using the following formula: Number of plants showing disease symptoms Disease incidence (%) = Total number of plants assessed X 100 For disease severity measurements, l2 plants from the central rows of each plot were randomly tagged prior to the appearance of typical disease symptoms. Disease severity was rated using a modified 0-9 scoring scale of Ding et al. (1993); where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. A total of seven severity assessments were made and disease severity scores were converted into percentage severity index (PSI) for analysis as follows: Sum of numerical ratings PSI = Number of plants scored x maximum score on the scale X 100 2.4. Grain Yield Data Grain yield (g plot–1) of faba bean was recorded from each experimental plot. Grain yield was determined from four harvestable central rows for each treatment, and converted to t ha–1 at 10% adjusted grain moisture content (Birru, 1979). 2.5. Data Analysis Disease incidence, severity and grain yield data were computed for each tested genotypes. The combined ANOVA of genotypes, environment and genotype by environment interaction was combined over locations and years on the basis of plot means and pooled over locations and seasons using generalized linear model procedures of the Genstat 14th edition (2011). The general ANOVA (SAS, 2014) was used to disaggregate the environment components in to location, years and their interaction with the genotype based on mean square. Both additive main effects and multiplicative interaction (AMMI) and the genotype and (genotype by environment) (GGE) biplot methods were used to investigate the genotype (G), 199 environment (E) and genotype by environment interaction (GEI) effects for Faba bean gall disease incidence, severity and grain yield of the faba bean genotypes. The AMMI model which combines the standard analysis of variance with principal component analysis was used to compute the average genotype by environment interaction and genotype rank across environments (Zobel et al., 1988; Falkenhagen, 1996). The measured FBG disease incidence, severity and grain yield of each genotype in each test environment was a measure of genotype main effect, environment main effect, and GE interaction. The modified AMMI model was used for the tested 21 genotypes and six test environments for disease parameters and grain yield (Gauch, 1992): Dij = 휇 + 푔푖 + 푒푗 + 푛 = λ푛훼푖푛푦푗푛 + 푅푖푗 Where, i = number of genotypes (1-21); j= number of locations (1-6) Where: D푖푗 = disease data/yield mean of ith genotype in j environment/, 휇 = grand mean; 푔푖 = main effect of genotypes; 푒푗 = main effects of environments; 휆푛 = Eigen values for PCA axis n; 훼푖푛 and 훾푗푛 = the ith genotype jth environment PCA scores for the PCA axis n; R푖푗 = the residual effect; n' = the number of PCA axes retained in the model. Genotype and genotype by environment interaction analysis were also computed. The GGE biplots were generated using the first two symmetrically scaled principal components (PC) for an average tester coordinate and polygon view biplots. Genotype and genotype by environment GGE biplot analysis model was used due to its statistical power to determine genotype resistant reaction stability, genotype performance in diverse environmental conditions, and the discriminatory power of environments and characterization of the environments (Yan and Kang, 2003). The GGE model was used to construct GGE biplot. The stability of the genotype or environment was determined by the length of the vector from the genotype marker to the average environment coordinate (AEC) abscissa. The vector which was closer to the AEC abscissa was considered to have less interaction effect and hence regarded as stable. A genotype rank, responsiveness and stability were determined based on all environments and closeness of the origin. The discriminatory power was detected by the length of the vector from the origin of the GGE biplot to coordinate of the location, and the longer the vector, the more discriminatory power. 200 3. Results 3.1. AMMI Analysis of FBG Disease Incidence, Severity and Grain Yield of Faba Bean The AMMI analysis of variance showed highly significant (P < 0.001) differences for genotypes, environments and genotype by environment interactions for FBG incidence. In the AMMI analysis of FBG incidence, genotype contributed about 80.32% of the variations. However, environment and genotype x environment interaction contributed only 5.32% and 14.36% of the variability, respectively. The environment (E) means for FBG incidence scores of the 21 faba bean genotypes over six environments ranged from 90.4% in Debay Telatgen– Kuy (E5) to 94.19% at Bassona Worana– Mush (E1) (Table 4). For FBG severity, 55.84% of the variations were contributed due to genotypes, and 37.83% and 6.33% contributions were due to environment and GEI, respectively (Table 3). The highest mean FBG severity (82.7%) was recorded at Bassona Worana– Mush (E1) and the lowest (35.2%) was recorded at Farta–Tiratir (E6). The combined means for FBG severity scores of the 21 faba bean genotypes over six environments ranged from 50.3% in Farta–Tiratir (E6) to 63.4% in Bassona Worana– Mush (E1) (Table 4). FBG severity was high in E1 and E2 as compared to E3, E4, E5 and E6. Thus, E1 was the most favorable site for disease expression, followed by (E2). The AMMI biplot indicated that of the 21 genotypes, G16, G17, G7 and G3 showed low FBG severity. In contrast, G1, G21, G13, G2 and G5 were susceptible to the disease in all the test environments. Genotypes with IPCA1 scores near zero had little interaction with the environment. E6 had a low environment score exhibiting little interaction with genotypes. The AMMI indicated that different sources of variations differed greatly as revealed by their sums of squares. The first four AMMI selections for mean FBG incidence showed that G5 at E1, G1 at E2 and E4, G2 at E3, E5 and E6 ranked first. The first four AMMI selections for mean FBG severity showed that G1 ranked first and G13 ranked 2nd in all the environments. The other susceptible genotypes (G13, G2 and G5) ranked 3rd and 4th (Table 4). G16 in five environments and G17 in one environment showed the last rank for FBG severity. Similarly, G16 in one environment (Farta– Gassay) and G17 in five environments showed 2nd ranks from the last. G1, G21, G13, G2 and G5 were grouped together and showed high FBG severity. G20, G10 and G9 showed high FBG severity below, but close to mean FBG severity. G11, G4, G12, G18, G15 and 201 G14 showed FBG severity close, but above the mean. G16, G17, G7, G3, G19, G8 and G6 showed low FBG severity. The genotype rank across six locations for FBG incidence and seceriry was presented in Table 7 and 8, respectively. The AMMI analysis of variance showed highly significant (P ≤ 0.001) differences for genotypes, environments and genotype by environment interactions for faba bean grain yield (Table 5). The relative magnitude of the different sources of variations strogly differed as revealed by their sums of squares. Genotype contributed about 50.86% to the total variations observed, while environments and genotype by environment interaction contributed 38.53% and 10.61% of the variations, respectively. The first and second interaction principal component axis contributed 78.84% and 13.12%, respectively. The average grain yield for the genotypes across environments ranged from 1.3 t ha-1 in E2 to 2.52 t ha-1 in E4. G8 ranked first at three environments (E1, E2 and E3) and G16 ranked first at three environments (E4, E5 and E6). G3 ranked second at E1 and E2 and G8 ranked second at E4. G17 ranked second at E5 and E6. G4 ranked third at E1 and E3. Similarly G7 ranked third at E6 (Table 6). From the AMMI model, E4, E5 and E6 were classified as high yielding, while E1, E2 and E3 were low yielding environments. G16, G8, G17, G3, G7, G4 and G6 were grouped together and showed high grain yield from high to low, and G12, G9, G14 and G15 showed good grain yield above and close to the mean. Similarly, G19, G10, G18 and G11 had grain yield below, but close to the mean (1.95 t ha–1). G1, G5, G21, G2, G13 and G20 obtained low grain yield. The genotype rank for grain yield across six location were presented in Table 9. 202 Table 3. Analysis of variance of additive main effect and multiplicative interaction on the genotype, environment and genotype x environment interaction for faba bean gall disease incidence and severity at six locations, Ethiopia in 2018 and 2019 cropping season Sources FBG disease incidence (%) FBG disease severity (%) df SS MS VE (%) P value df SS MS VE (%) P value Total 377 15250 40.45 – 377 27029 162 – <0.001 Treatment 125 12628 101.03 – <0.001 125 21759 2268 – <0.001 Genotype (G) 20 10144 507.18 80.32 <0.001 20 12150 608 55.84 <0.001 Environment (E) 5 672 134.31 5.32 <0.001 5 8231 1646 37.83 <0.001 G x E 100 1813 18.13 14.36 <0.001 100 1378 14 6.33 0.0475 IPCA1 24 868 36.18 47.90 <0.001 24 115 107 85.70 <0.0281 IPCA2 22 491 22.34 27.10 0.0020 22 103 143 6.52 <0.0301 Residual 54 453 8.39 – 0.7940 54 215 103 – – Error 240 2432 10.13 – - – 3322 14 – – Total = Total treatment; G x E = genotype x environment interactions; and IPCA1 = Interaction principal component analysis axis. df = degrees of freedom; SS = Sum of squares; MS = Means of sum of squares; VE (%) = Variance explained; and P = Probability. Table 4. The first four AMMI selections for mean faba bean gall disease incidence and severity per environment Environment/Location Mean FBG Rank of genotypes FBG disease Rank of genotypes disease 1st 2nd 3rd 4th severity (%) 1st 2nd 3rd 4th incidence (%) E1. Bassona Worana -Mush 94.19 G5 G21 G1 G2 63.4 G1 G21 G13 G5 E2. Debay Telatgen -Asendabo 92.68 G1 G2 G11 G21 61.8 G1 G21 G13 G2 E3. Farta – Gassay 92.22 G2 G21 G1 G5 58.9 G1 G21 G2 G13 E4. Bassona Worana -Gudoberet 91.16 G1 G2 G21 G18 54.9 G1 G21 G13 G2 E5. Debay Telatgen -Kuy 90.02 G2 G13 G1 G21 53.1 G1 G21 G5 G2 E6. Farta –Tiratir 91.13 G2 G21 G1 G13 50.3 G1 G21 G5 G13 Ranks of genotype for FBG disease incidence and severity standing 1st to 4th refer to high to low performances of the genotypes. 203 Table 5. Analysis of variance of additive main effect and multiplicative interaction on the genotype, environment and genotype x environment interaction for faba bean grain yield at six locations, Ethiopia in 2018 and 2019 cropping season Sources Faba bean grain yield df SS MS VE % P value Total 377 279556960 741530 – <0.001 Treatment 125 270321444 2162572 – <0.001 Genotype (G) 20 137473865 6873693 50.86 <0.001 Environment (E) 5 104159900 20831980 38.53 <0.001 G x E 100 28687679 286877 10.61 <0.001 IPCA1 24 22617360 286877 78.84 <0.001 IPCA2 22 3760955 171040 13.12 <0.001 Residual 54 2308795 42755 1.18 0.1983 Error 240 – – – – df = degree of freedom; SS = sum of square; MS = means sum of square; VE % = variance explained; P = probability; G x E = genotype by environment interactions; IPCA1 = interaction principal component analysis axis. Table 6. The first four AMMI selections for mean faba bean grain yield per environment Environment Grain yield Rank of genotypes (high to low) ( t ha-1) 1 2 3 4 E1. Bassona Worana–Mush 1.71 G8 G3 G4 G16 E2. Debay Telatgen–Asendabo 1.30 G8 G3 G16 G17 E3. Farta–Gassay 1.51 G8 G16 G4 G7 E4. Bassona Worana–Gudoberet 2.52 G16 G8 G17 G3 E5. Debay Telatgen–Kuy 2.46 G16 G17 G3 G8 E6. Farta–Tiratir 2.23 G16 G17 G7 G3 Ranks of genotype for FBG disease incidence and severity standing 1st to 4th refer to high to low performances of the genotypes. 204 Table 7. Mean faba bean gall disease incidence (%) and rank of faba bean genotypes evaluated at six locations in Ethiopia, in 2018 and 2019 main cropping seasons Genotype Test environment (location) by year Bassona Worana Debay Telatgen Farta Mush Gudoberet Asendabo Kuy Gassay Tiratir (2018) (2019) (2018) (2019) (2018) (2019) Wayu 99.4 (3) 97.4 (1) 98.7 (1) 96.7 (3) 98.7 (3) 96.0 (3) Mosisa 98.3 (4) 96.8 (2) 98.3 (2) 100 (1) 99.4 (1) 100 (1) Numan 81.3 (20) 81.7 (18) 79.0 (21) 76.7 (20) 85.3 (18) 84.0 (20) Ashebeka 96.7 (10) 91.0 (14) 96.4 (5) 91.3 (9) 93.7 (8) 91.7 (8) Shallo 100 (1) 94.0 (8) 96.3 (6) 95.0 (5) 97.7 (4) 94.3 (4) Didea 97.7 (7) 89.3 (16) 94.3 (12) 92.0 (8) 92.0 (12) 91.7 (8) Gora 82.3 (19) 81.7 (18) 83.3 (18) 81.0 (19) 85.3 (18) 89.7 (16) Gachena 92.3 (17) 89.3 (16) 95.0 (11) 90.7 (12) 89.3 (16) 84.7 (19) Degaga 95.3 (15) 92.3 (11) 92.7 (15) 90.7 (11) 90.7 (15) 87.7 (18) Dosha 97.3 (8) 96.5 (5) 96.3 (6) 91.0 (10) 95.0 (6) 92.7 (6) Tumsa 96.3 (12) 93.0 (9) 97.0 (3) 92.3 (7) 94.7 (7) 91.7 (8) Hachalu 95.7 (13) 91.3 (13) 92.3 (16) 89.3 (16) 89.3 (16) 88.7 (17) Wolki 97.3 (8) 95.3 (7) 96.0 (9) 98.3 (2) 96.7 (5) 94.3 (4) EH01005-1 96.7 (10) 95.7 (6) 95.3 (10) 92.7 (6) 93.7 (8) 91.0 (13) EH09028-3 95.7 (13) 93.0 (9) 94.3 (13) 89.7 (13) 91.3 (13) 91.3 (12) EH010008-5 81.3 (20) 75.0 (21) 80.0 (20) 75.0 (21) 79.0 (21) 82.7 (21) EH010058-1 83.0 (18) 79.3 (20) 82.0 (19) 83.0 (18) 83.7 (20) 90.3 (15) EH09017-5 98,0 (5) 96.6 (4) 93.0 (14) 89.7 (13) 91.0 (14) 91.0 (13) EH09021-1 95.3 (15) 92.0 (12) 92.0 (17) 89.7 (13) 93.3 (11) 91.7 (8) EH09046-3 98.0 (5) 91.0 (14) 96.3 (6) 89.0 (17) 93.7 (8) 92.3 (7) Local 99.9 (2) 96.7 (3) 96.7 (4) 96.7 (4) 99.2 (2) 96.3 (2) Mean 94.19 91.16 92.68 90.02 92.22 91.13 SE 1.38 1.39 1.32 1.41 1.17 0.90 SE = refers standard error and the numbers in brackets indicate rank of the genotypes at each testing location. 205 Table 8. Mean faba bean gall disease severity (%) and rank of faba bean genotypes evaluated at six locations in Ethiopia, in 2018 and 2019 main cropping seasons Genotype Test environment (location) by year Bassona Worana Debay Telatgen Farta Mush Gudoberet Asendabo Kuy Gassay Tiratir (2018) (2019) (2018) (2019) (2018) (2019) Wayu 82.7 (1) 71.3 (1) 75.8 (1) 69.4 (1) 72.9 (1) 66.8 (1) Mosisa 67.4 (5) 58.9 (6) 65.3 (4) 57.6 (4) 64.0 (3) 53.9 (7) Numan 57.2 (18) 48.6 (18) 55.9 (19) 46.8 (18) 55.1 (18) 44.1 (18) Ashebeka 63.2 (10) 54.4 (12) 60.0 (16) 53.6 (11) 58.0 (9) 50.6 (11) Shallo 67.7 (4) 59.4 (4) 64.6 (5) 58.2 (3) 57.8 (10) 55.7 (3) Didea 60.0 (14) 53.3 (15) 63.1 (7) 51.1 (15) 59.9 (6) 48.3 (15) Gora 57.0 (19) 48.3 (19) 57.9 (18) 46.1 (19) 57.3 (11) 41.2 (19) Gachena 61.6 (13) 51.3 (17) 61.1 (12) 49.6 (17) 57.1 (14) 46.8 (17) Degaga 67.2 (7) 58.8 (7) 59.4 (17) 56.7 (6) 57.1 (14) 54.0 (6) Dosha 67.4 (5) 56.4 (8) 63.2 (6) 54.4 (8) 62.1 (5) 51.8 (8) Tumsa 62.1 (12) 54.8 (11) 62.4 (8) 53.9 (9) 56.4 (16) 51.0 (10) Hachalu 64.2 (9) 54.4 (12) 61.3 (11) 51.7 (14) 57.2 (12) 50.2 (13) Wolki 68.9 (3) 60.3 (3) 66.2 (3) 56.7 (6) 63.1 (4) 55.0 (4) EH01005-1 60.0 (15) 55.9 (9) 60.2 (15) 53.9 (9) 57.2 (12) 51.1 (9) EH09028-3 62.9 (11) 54.4 (12) 62.1 (9) 52.2 (13) 55.8 (17) 49.4 (14) EH010008-5 51.4 (21) 39.4 (21) 54.0 (21) 37.9 (21) 52.9 (20) 35.2 (21) EH010058-1 51.4 (20) 45.6 (20) 54.7 (20) 44.3 (20) 52.1 (21) 40.8 (20) EH09017-5 59.8 (16) 55.0 (10) 60.9 (13) 53.1 (12) 58.7 (8) 50.4 (12) EH09021-1 58.9 (17) 52.2 (16) 62.0 (10) 50.0 (16) 53.4 (19) 47.3 (16) EH09046-3 66.8 (8) 59.4 (5) 60.9 (14) 57.4 (5) 59.7 (7) 54.8 (5) Local 73.2 (2) 61.7 (2) 66.4 (2) 60.3 (2) 68.1 (2) 57.5 (2) Mean 63.4 54.9 61.8 53.1 58.9 50.3 SE 1.55 1.42 1.02 1.41 1.09 1.46 SE = refers standard error and the numbers in brackets indicate rank of the genotypes at each testing location. 206 Table 9. Mean grain yield and rank of faba bean genotypes evaluated for faba bean gall disease at six locations in Ethiopia, in 2018 and 2019 main cropping seasons Genotype Test environment (location) by year Bassona Worana Debay Telatgen Farta Mush Gudoberet Asendabo Kuy Gassay Tiratir (2018) (2019) (2018) (2019) (2018) (2019) Wayu 0.87 (21) 0.84 (21) 0.77 (21) 0.78 (21) 0.76 (21) 0.79 (21) Mosisa 1.14 (17) 1.64 (19) 0.91 (20) 1.68 (17) 1.20 (15) 1.59 (17) Numan 2.44 (2) 3.24 (4) 1.81 (2) 3.25 (3) 1.98 (5) 2.84 (4) Ashebeka 2.39 (3) 3.14 (6) 1.44 (7) 3.08 (7) 2.13 (3) 2.69 (7) Shallo 1.13 (18) 1.23 (20) 0.97 (18) 1.29 (20) 1.00 (18) 1.14 (20) Didea 1.64 (12) 2.95 (7) 1.33 (10) 3.12 (6) 1.43 (10) 2.75 (6) Gora 2.10 (6) 3.18 (5) 1.40 (8) 3.18 (5) 2.03 (4) 3.02 (3) Gachena 2.77 (1) 3.38 (2) 1.94 (1) 3.18 (4) 2.62 (1) 2.82 (5) Degaga 1.92 (8) 2.76 (10) 1.04 (16) 2.76 (9) 1.58 (8) 2.47 (9) Dosha 1.58 (13) 2.63 (11) 1.05 (15) 2.48 (11) 1.19 (16) 2.21 (13) Tumsa 1.67 (10) 1.94 (16) 1.51 (5) 2.20 (15) 1.39 (12) 1.89 (16) Hachalu 1.95 (7) 2.86 (9) 1.21 (13) 2.67 (10) 1.42 (11) 2.43 (10) Wolki 1.04 (19) 2.05 (15) 0.94 (19) 2.01 (16) 0.98 (19) 1.90 (15) EH01005-1 1.92 (8) 2.60 (12) 1.39 (9) 2.45 (12) 1.68 (7) 2.30 (11) EH09028-3 1.25 (16) 2.93 (8) 1.27 (12) 2.87 (8) 1.17 (17) 2.62 (8) EH010008-5 2.38 (4) 3.65 (1) 1.59 (3) 3.53 (1) 2.21 (2) 3.26 (1) EH010058-1 2.10 (5) 3.33 (3) 1.55 (4) 3.45 (2) 1.96 (6) 3.15 (2) EH09017-5 1.37 (15) 2.24 (14) 1.31 (11) 2.40 (13) 1.34 (13) 2.21 (12) EH09021-1 1.67 (10) 2.52 (13) 1.51 (5) 2.30 (14) 1.47 (9) 2.04 (14) EH09046-3 1.48 (14) 1.90 (17) 1.02 (17) 1.54 (18) 1.30 (14) 1.49 (18) Local 1.04 (20) 1.84 (18) 1.19 (14) 1.36 (19) 0.93 (20) 1.24 (19) Mean 1.71 2.52 1.30 2.46 1.51 2.23 SE 0.12 0.16 0.07 0.17 0.11 0.15 SE = refers standard error and the numbers in brackets indicate rank of the genotypes at each testing location. 3.2. GGE biplot Analysis of FBG Disease and Grain Yield of Faba Bean GGE biplot analysis for FBG incidence, severity and grain yield were computed and the genotype, environment and genotype x environment interaction were further explored. The genotypes showed variable reactions to FBG incidence and severity at six locations. The first two principal components of the GGE biplot accounted for a total of 64.96% (PC1 = 47.88%, PC2 = 27.08%) of the variations for FBG incidence and the main effect of genotypes and environment on FBG incidence is presented in Figures 3 and 4. Similarly, in the GGE biplot analysis, majority of the faba bean genotypes evaluated at six locations showed similar FBG severity reactions. The first 207 two principal components (PCs) of the GGE biplot accounted for a total of 92.22% (PC1 = 85.7%, PC2 = 6.52%) of the variations for FBG severity. Figure 5 displayed the polygon view of the GGE biplot of 21 faba bean genotypes for FBG severity to determine the ‘which–won–where’ pattern and mega environments. In the biplot view, G1 was the vertex genotypes in the sector that had E1, E3, and E5 and E6. G1 was the most susceptible to FBG in all environments. G21 and G9 were found in the fourth quadrat. G16 and G17, on the other hand, were moderately tolerant genotypes to FBG as indicated by the high negative PC1 score. The other genotypes (G10, G12, G4, G18 and G6) within the polygon view and near the origin had low positive or negative PC1 scores indicating less tolerant to FBG and were less responsive compared with the vertex genotypes. The test environments were almost grouped in similar category, except E2 and E4 as indicated on Figure 5. Similarly, the GGE biplot analysis, for mean performance and stability of genotypes for FBG severity was considered based on an average-environment coordinates (AEC). G1and G21 were susceptible and unstable where as G2, G5 and G13 were suceptible and stable. G16 and G17 had low FBG severity (negative low PC1 scores) and unstable across the environments. G7 was moderately tolerant and stable. In contrast G8, G10 and G12 were stable, but were not tolerant to FBG (data not shown). Quadrat I had genotypes with an IPCA score near the origin (zero) and average FBG severity showing stability of the genotypes across the environments tested. However, genotypes with high mean performance and large IPCA score are considered unstable across environments, but have specific adaptation to some environments. Discriminating power and representativeness of the test environments for FBG were also considered based on average coordinate, vector length and biplot origin. E1 and E2 had the least angles to the Average Environment Axis (AEA) line indicating that there were the most representative of the test environments. In addition, the environment vector for E3 and E4 was relatively long vector indicating discriminating ability of the genotype. E1and E3 had the longest vectors and high positive PC1 scores, suggesting that they were more discriminating of the genotypes than the other environments. Environment E5 had the shortest environment vector and PC2 close to zero, suggesting less discriminating ability. 208 The genotype, environment and genotype by environment interactions were further explored through GGE biplot analysis for faba bean grain yield. The AMMI2 biplot (PC1 and PC2) for grain yield of 21 genotypes at six environments are presented Figure 7. The first two principal components (PCs) of the GGE biplot accounted for a total of 97.59% for the variation of faba bean grain yield over six locations. In the ‘which –won-where’ pattern and mega environments analysis, a polygon view of the GGE biplot for grain yield of faba bean resulted in five vertex genotypes with both positive (high yielding) and negative (low yielding) PC1 scores (Figure 8). These genotypes included G8 which were vertex genotype in the sector at environments E1, E2, and E3. Genotype G16 was the vertex genotype in E4, E5 and E6. The other three genotypes (G1, G13 and G15) fell in sectors with no environment markers. Three environments E1, E2 and E3 fell in one sector thus encompassing one similar-environment, and E4, E5 and E6 were grouped into another similar-environment. Genotypes in the polygon with no environment indicated that such genotypes had poor performance in all environments. Similarly, the GGE biplot analysis, showed mean performance and stability of genotypes for grain yield; based on an average-environment coordinates (AEC). G17 was unstable and showed high grain yield at E5. G16, G7, G4 and G8 showed high grain yield and better adaptations at E4 and other environments. G6 showed good grain yield and specific adaptation at E6. In general, G1, G5, G2, G21 and G13 were unstable genotype identified by AMMI model, with low grain yield. From the AMMI biplots, the markers for genotype were more scattered than the markers for environment indicating that the variability due to genotypes were higher than that due to environments (Figure 6). The biplot contains the average environment axis and the biplot origin. E1 and E3 had the smallest angle with the average environment axis indicating it was more representative of the test environments while E3 and E2 was the least representative. E2 had the longest vector from the biplot origin indicating it was the most discriminating of the environments, but not representative. E3 and E6 had moderate vector lengths. E4 had medium vector lengths. In terms of correlation, none of the environments were negatively correlated as there was no obtuse angle (> 90o) observed between any of the environments (Figure 7). All the environments had acute angles (< 90 o) with each other, with some of the environments like, E1 and E3; E4 and E5 having even smaller angles between them indicating more positive correlations between the environments. The single-arrow on the AEC points higher mean gain yield (Figure 9 A). The double-arrowed line 209 (Figure 9 B) is the AEC that points in either direction to greater variability (least stability). G3, G14 and G12 were stable with above average performance. The GGE biplots showed the discriminating ability and representativeness of the test environments with the average environment and the average coordinates of all test environments. The GGE biplots enables to evaluate the environments and to identify environments that may serve to select superior genotypes in an efficient way for the mega environment. The selected test environment should have high genotype discriminativeness and representativeness. Environments with shorter vectors have less discriminativeness in relation to genotypes, that is, all genotypes tend to perform equally and almost no information about genotypic differences can be revealed by such environments. A short vector could also mean that PC1 and PC2 do not represent that environment very well in cases where G + GE have not been retained properly. E2, E3, and E1 presented a long vector, which means they have high discriminativeness for the genotypes. It is also possible to identify environments with high representativeness that the smaller the angle between an environment and the mean-environment axis, the higher is its representativeness. An environment that has both characteristics more than the others is E1. Environments E2 and E3 have long vectors, but greater angles, implying that they should not be recommended (Figure 10). Main effect: - Genotype + - Environment Figure 3. Mean FBG disease incidence; AMMI biplot of the first interaction principal component axis (IPCA1) versus mean FBG disease incidence (%) of 21 faba bean genotypes at E1= Bassona Worana (Mush), E2 = Debay Telatgen (Asendabo), E3 = Farta (Gassay), E4 = Bassona Worana (Gudoberet), E5 = Debay Telatgen (Kuy), E6 = Farta (Tiratir), G = Genotype 210 P C1- 47.90 Figure 4. FBG disease incidence AMMI2 biplot displaying genotype and environment in the first and second principal component (PC) axis; E1= Bassona Worana (Mush), E2 = Debay Telatgen (Asendabo), E3 = Farta (Gassay), E4 = Bassona Worana (Gudoberet), E5 = Debay Telatgen (Kuy), E6 = Farta (Tiratir), G = Genotype Scatter plot (total 92.22%) PC1- 85.70% Figure 5. Which-won-where or which is best at what: A polygon view of GGE biplot, showing which genotype showed better reaction for FBG severity on genotypes, environment and genotype x environment on 21 faba bean genotypes evaluated in six environments. E1= Bassona Worana (Mush), E2 = Debay Telatgen (Asendabo), E3 = Farta (Gassay), E4 = Bassona Worana (Gudoberet), E5 = Debay Telatgen (Kuy), E6 = Farta (Tiratir), G = Genotype 211 PC2- 27.10% PC2- 6.52% Plot of Genotype and environment IPCA2 score versus mean grain yield Main effect: - Genotype + - Environment Figure 6. AMMI biplot of the first interaction principal component axis (IPCA1) versus mean grain yields (t ha-1) of faba bean genotype; E1= Bassona Worana (Mush), E2 = Debay Telatgen (Asendabo), E3 = Farta (Gassay), E4 = Bassona Worana (Gudoberet), E5 = Debay Telatgen (Kuy), E6 = Farta (Tiratir), G = Genotype PC1- 78.84% Figure 7. The AMMI2 biplot displaying 21 faba bean genotypes grain yield at six environments in the first and second principal component (PC) axis; E1= Bassona Worana (Mush), E2 = Debay Telatgen (Asendabo), E3 = Farta (Gassay), E4 = Bassona Worana (Gudoberet), E5 = Debay Telatgen (Kuy), E6 = Farta (Tiratir), G = Genotype 212 PC2-12.13% Scatter plot (total 97.59%) PC1- 91.7 3% Figure 8. ‘Which-won-where or which is best at what’: for grain yield, GGE biplot, showing which genotype showed better grain yield on genotypes, environment and genotype x environment on 21 faba bean genotypes tested in six environments. E1 = Bassona Worana 2018, E2 = Debay Telatgen 2018, E3 = Farta 2018, E4 = Bassona Worana 2019, E5 = Debay Telatgen 2019, E6 = Farta 2019, G = Genotype 213 PC2- 5.87% Ranking biplot (Total 97.59) Ranking biplot (Total 97.59) PC1 - 91.73% PC1 - 91.73% Figure 9. (A and B) Average environment coordination (AEC) views of the GGE biplot on environment-focused scaling for faba bean grain yield means and stability. 214 PC2- 5.87% PC2- -5.87% Comparison biplot (total 97.59%) PC1 – 91.73 Figure 10. Discriminitiveness and represintativeness of the GGE biplot shows interrelationships among the test environment and comparison of environment with ‘the ideal environment’ based on a genotype x environment for faba bean grain yield. 4. Discussion The AMMI analysis revealed highly significant effects for genotypes (G), environments (E), and genotype x environment interactions (GEI) both for FBG disease and grain yield of faba bean. The AMMI analysis showed that genotypes were the main source of variation for FBG incidence expression, followed by genotype x environment interaction. Similarly, the main source of variation for FBG severity was attributed due to genotype followed by environment; however, the variation due to genotype x environment interaction was very low. This implies that the genetic background of the genotypes contributed more thatn other factors in all the test locations for the phenotypic reaction of faba bean genotypes against FBG disease. Different degrees of reactions were observed on faba bean genotypes evaluated at six locations. 215 PC2 – 5.87% Tolerant faba bean genotypes were consistently tolerant and susceptible faba bean genotypes were consistently susceptible in all the test locations. The results disagree with previous study reported by Alehegn et al. (2018) that most of faba bean varieties including Degaga, Dosha and Tumsa evaluated in east Gojjam were resistant against FBG, but different results were obtained in the current study. Genotypes reported by Getenet and Yezbalem (2017) as moderately resistant (Degaga) were not resistant in the current study across the six test locations, but showed variable reactions. Despite the fact, different faba bean genotypes showed different degrees of FBG reaction as reported by Wulita (2015), DBARC progress report (2015) and Yan (2012). Similarly, some of the genotypes evaluated in the current study showed similar moderately tolerant reactions as reported by Negussie et al. (2018). Variability in the response of genotypes to FBG disease depends more on the genotypes than environment and genotype x environment interaction. On the other hand, genotypes responded similar reaction trends across the test environments indicating that damage of FBG disease was not influenced by the current test environments. That is, variation in FBG severity across locations could be due to the genetic contents of the genotypes or it could be due to the fact that the current test environments have very narrow differences. In contrast, in the AMMI analysis of previous researches, the contribution of environment was high as compared to genotypes for other faba bean diseases conducted by Beyene et al. (2017). Similarly, equal contributions of the genotypes and genotype x environment interaction to disease expression were also reported for faba bean genotypes evaluated in different environments (Villegas Fernandez et al., 2011). Generally, genotypes remained constant from one environment to another environment; however, when the same genotype is subjected to different environments, it can produce a wide range of phenotypic reactions (Tesfaye et al., 2010; Abou-Khater et al., 2022). Differences in the genetic composition play a major role in variable response to a disease. In the current study, some of the genotypes evaluated adapted and responded to FBG disease similar reaction in wide environments and few of them were adapted and responded differently to FBG disease in specific areas. Report showed that genotypes that are closer to the origin are more stable (Yan, 2011). However, in the current study, only few genotypes were closer to the origin and these genotypes were not tolerant to FBG. 216 A genotype is regarded stable, if it has low contribution to the G × E interaction and stable genotypes are those showing a consistent performance regardless of any changes in environmental conditions (Ortiz et al., 2001). In the present study, each polygon was formed by connecting the genotypes that are farthest from the biplot origin so that all other genotypes are inside the polygon and showed which won where as described by Flores et al., (2013). The polygon view of the GGE biplot revealed that G16, G17, G3, G19 and G8 had low FBG severity, but were the least stable genotypes, while G7 was tolerant and stable for FBG disease. The test environments fell into a similar sector except E3 which showed few differences. G16, G17 and G3 were desirable genotypes for their low disease severity. Conversely G1 was the most susceptible among the tested genotypes followed by genotypes G21, G13, G2 and G5. In this study, variations of genotype stability and discrimination of the environmenteffects for FBG disease were observed. An environment is considered ideal for genotype testing when it discriminates the genotypes and represents the environments (Yan and Kang, 2003; Yan and Kang, 2002; Hongyu et al., 2014). E1 was the most discriminating and representative environment for FBG disease followed by E3 and E2. In contrast, E5, E6 and E4 had less discriminating ability for the genotypes. This indicated that E1 is the most efficient for evaluating the potential of genotypes for FBG disease reactions. Therefore, among the six environments, E1 represented the ideal testing environment for FBG disease and would be appropriate for selecting best faba bean genotypes against FBG disease. The findings of the current study could coincide with the reports of Beyene et al. (2017) on other faba bean disease on the identification of ideal environments. In the AMMI analysis of variance, grain yield performances for the tested faba bean genotypes were influenced highly by the genotype contribution to the total variation, followed by environment reflecting that the genotypes were highly variable in their responses to different environmental changes. In contrast, the report by Beyene et al. (2016) indicated that environment can contribute more than genotype to the total variations for faba bean grain yield when tested at different environments. There are also studies reported for high contribution of genotype by environment interactions for the variation of grain yield in faba bean (Teklay et al., 2015). Genotype, environment and genotype x environment interaction are expected for the total variations and all these different reports indicated that yield is a very complex trait which is strongly influenced by genotype, environment or genotype x environment interaction as described by Abou-Khater et al. (2022). 217 AMMI biplot revealed that genotypes (G3, G14 and G12) were stable with above average performance. G16, G17 and G3 were high yielding, but adapted in specific environments. G6 and G15 had specific adaptation at E5 and E6 with average grain yield above the mean. Similarly, G8 was adapted at E1 and E3 with high grain yield. E4 had the higher yielding genotypes, thus representing a high potential environment while E2 was the lowest yielding environment. This could be due to the soil and weather variation between the two environments (E2 and E4). The soil pH at E4 was higher than E2 that might contribute for low grain yield at E2. Different reports showed that faba bean grows best on heavier-textured soils, but tolerates nearly any soil type (Jenson et al., 2010). Growth and yield of faba bean are determined by climatic, edaphic, and management practices that are not independent of each other and interact to affect the chemical characteristics of the soil. Soil acidity has a dramatic impact on most chemical and biological processes of the crop. A report has also shown that faba bean grows best in soils with pH ranging from 6.5 to 9.0 (Jenson et al., 2010), but poorly at pH values of 5 or less (French and White, 2005). In the current study, the soil pH at E4 was 6.8 and 5.38 at E2 which agrees with the Jenson et al. (2010) report. The GGE biplot analysis provided a visual representation of the relationship among the genotypes and test environments and used to evaluate the environments (Yan et al., 2000). The presence of correlation between two environments indicates that similar information about the genotype performance is derived from them and therefore could be an option to reduce the number of test environments and, as a result, to establish a cost-effective genotype performance evaluations (Gedif and Yigzaw, 2014). The polygon view of the GGE biplot indicated the presence of a crossover genotype by environment as the environments fell in two sectors of the polygon view and had different high yielding genotypes as described by Yan and Kang (2002). G8 were vertex genotype in the sector at E1, E2, and E3. Genotype G16 was the vertex genotype at E4, E5 and E6. The other three genotypes (G1 G13 and G15) fell in sectors with no environment markers. Three environments (E1, E2 and E3) fell in one sector thus encompassing one similar environment, and E4, E5 and E6 were grouped into another similar environment. In the present study, majority of genotypes in the polygon were scattered with no environment that indicated such genotypes had reduced performance in all environments. This study tried to show which genotype performed better and stable at which environment, as described by Yan (2014). According to Yan and Tinker (2006), the test environments that are less 218 discriminating provide little information on the genotype differences and should not be used as test environments. E2 had the most discriminating environment for grain yield due to its long vector followed by E3 and E1. In contrast, E4, E5 and E6 had the least discriminating environments. The selected test environment should have high genotype discriminativeness and representativeness. The relationship between the environments was visualized by drawing a vector that connected each environment to the biplot origin and the correlation between two environments was approximated based on the angle between two vectors (Yan and Tinker, 2006; Papastylianou et al., 2021); the smaller the angle between two vectors the higher is the correlation between the two environments. Environments with shorter vectors have less discrimination in relation to genotypes, that is, all genotypes tend to perform equally and almost no information about genotypic differences revealed by such environments. A short vector could also mean that PC1 and PC2 do not represent that environment very well in cases where G and GE have not been retained properly. E2, E3, and E1 had a long vector, which means they have high discriminativeness for the genotypes. It is also possible to identify environments with high representativeness: the smaller is the angle between an environment and the mean environment axis, the higher is its representativeness (Yan and Tinker, 2006). An environment that had both characteristics more than the others is E1. E2 had long vector, but greater angles, implying least representative of the test environments indicating that it should not be discriminating and recommended. On the other hand, E1 had relatively long vector and small angle and could be the most representative of the environments for grain yield. This study, along with that of Yan and Kang (2002) highlight that an ideal test environment should effectively discriminate genotypes and represent the environments. Thus, in this study among all the six environments, E1 represented the ideal testing environment with high discriminating ability of the genotypes and moderate in representativeness of the test environment for faba bean grain yield. This environment can be used for selecting generally improved genotypes. Environments such as E2, which was discriminating, but non-representative, are recommended for selecting specifically improved genotypes as described by Yan and Tinker (2005). In this study, the angles between all the six environments were acute (< 90o) indicating positive correlations among them for both grain yield and FBG disease severity. This suggests that the same information could be obtained about the genotypes from these environments which are closely associated. The GGE biplot analysis used to show mega-environment analysis, genotype evaluation 219 and test-environment evaluation as reported by Yan et al. (2007), however in the current study the environments were classified almost under similar category. Similarly, Yan et al. (2000) stated that the first principal component in the graphical analysis represents genotypic performance, while the second represents genotypic stability. Therefore selecting the most representative and few test environments could be used to evaluate genotypes and minimize cost. 5. Conclusions Faba bean genotypes evaluated at six locations showed different FBG disease severity reactions. In the present study, FBG disease was influenced largely by faba bean genotypic differences. Similarly, yield performances were influenced by genotype and environment. Both the AMMI and GGE biplot analysis indicated almost similar results in terms of FBG disease severity as well as stability and performances of the genotypes grain yield. Among the genotypes evaluated, G16, G17 smf G3 showed low FBG severity. G7 showed low FBG severity and stable. In contrast, G1, G21 G13, G2 and G5 showed high FBG disease severity. G16, G17, G3, G7 and G4 had high grain yield, but less stable genotypes across the test locations. G8 was stable and showed high grain yield. Genotype by environment interaction on FBG disease severity was very low. The GGE biplot showed that Bassona Worana-Mush (E1) was the best in terms of discriminating ability and representative of the test location to evaluate faba bean for FBG disease severity and grain yield respectively. In general genotype G16, G17, G3, G7 and G8 are recommendable for further evaluation and crossing purpose. 6. Acknowledgements The study was financially supported by Amhara Regional Research Institute (ARARI), Australian Center for International Agricultural Research (ACIAR) and the International Center for Agricultural Research in the Dry Areas (ICARDA), Africa RISING/USAID through ICARDA. Hence the institutes are duly acknowledged for supporting the study. This work is part of a PhD Dissertation research requirement at Haramaya University and the University is duly acknowledged. We also greatly acknowledge, Mr. Aleme Belete and Amare Belachew for their assistances in field and laboratory data collection. 220 7. References Abou-Khater, L., Maalouf, F., Jighly, A., Rubiales, D. and Kumar, S. 2022. Adaptability and Stability of Faba Bean (Vicia faba L.) Accessions under Diverse Environments and Herbicide Treatments. Plants, 11, 251. https://doi.org/10.3390/ plants11030251. Agegnehu, G. and Fessehaie, R. 2006. Response of faba bean to phosphate fertilizer and weed control on nitisols of Ethiopian highlands. Italian Journal of Agronomy, 2: 281-290. Alehegn, M., Melekamu, T. and Meseret, T. 2018. Screening of faba bean (Vicia faba) varieties against Faba bean Gall Disease (Olpidium viciae) in East Gojjam Zone, Ethiopia. Journal of Biology, Agriculture and Healthcare, 8: 50-55. Asfaw, T., Beyene, D. and Tesfaye, G. 1994. Genetic and breeding of field pea. First National Cool- season Food Legumes Conference. Addis Ababa, Ethiopia. ICARDA/ IAR. ICARDA: Aleppo, Syria. pp. 122-137. Beyene, B. 2015. Survey and identification of new faba bean disease (Qormid) in the highlands of North Shewa, Ethiopia. Current Research and Microbio. Biotechnology, 3: 561–563. Beyene, A., John, D., Sibiya, J. and Fikre, A. 2016. Gene action determining grain yield and chocolate spot (Botrytis fabae) resistance in a faba bean. Euphytica, 207: 293–304. Beyene, A., John, D., Sibiya, J. and Fikre, A. 2017. Analysis of genotype × environment interaction and stability for grain yield and chocolate spot (Botrytis fabae) disease resistance in faba bean (Vicia faba). Australian Journal of crop science 11(10):1228–1235. Birru, A. 1979. Agricultural field experiment management manual part III. Institute of Agricultural Research. Addis Ababa, Ethiopia. Bogale, N., Setu, B. and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae (Kusano) in major faba bean (Vicia faba L.) growing areas of Northeastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 110: 87–95. Challa, D., Abraham, N., Zerihun, A. and Meseret, T. 2017. Assessment of the Occurrence and Prevalence of Faba Bean Gall (Olpidium viciae) in Western Highlands of Oromiya, Ethiopia. Journal of Natural Sciences Research, 7 (5): 63–67. CSA (Central Statistics Agency). 2021. Agricultural sample survey (2020/2021). Report on area and production of major crops. Statistical Bulletin 578, Central Statistical Authority, Addis Ababa, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2015. Annual progress report. 221 Dereje, G. 1999. Survival of Botrytis fabae Sard. between seasons on crop debris in field soils at Holetta, Ethiopia. Phytopathologia Mediterranean, 38: 68–75. Dereje, G., Wendafrash, M. and Gemechu, K. 2012. Faba Bean Galls: A new disease of faba bean in Ethiopia. Holetta Agricultural Research Center. Available at: https://docs.google.com/file/d/0B6qFhb82WH8MWXJxdFJZZXlRcEE/edit. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean Newsletter, 32: 8–10. Duc, G., Aleksic, J. M., Marget, P., Mikić, A., Paull, J., Redden, R. J. 2015. Faba bean, in Grain Legumes, ed. A.M. De Ron, 141–178. https:// doi: 10.1007/978-1-4939-2797-5-5. Falkenhagen, E. 1996. Comparison of the additive main effects and multiplicative interaction methods with some classical statistical methods in provenance research. The case of South African Pinus Radiata trials. Genetics, 3: 81–87. FAOSTAT (Food and Agricultural Organization of the United Nations, Statistical Data Base). 2018. Rome, Italy, Available at: https://faostat.fao.org. Flores, F., Hybl, M., Knudsen, J.C., Marget, P., Muel, F., Nadal, S., Narits, L., Raffiot, B., Sass, O. and Solis, I. 2013. Adaptation of spring faba bean types across European climates. Field Crops Research, 145, 1–9. French, B. and White, P. 2005. Soil and environmental factors affecting pulse adaptation in Western Australia. Australia Journal of Agricultural Research, 50:375–387. Gauch, H.G. 1992. Statistical analysis of regional yield trials. AMMI analysis of factorial designs. First editions, P.265–276, Elsevier, Amsterdam. Gedif, M. and Yigzaw, D. 2014. Genotype environment interaction analysis for tuber yield of potato using a GGE biplot method in Amhara region, Ethiopia. Potato Journal, 41: 41–51. Gemechu, K. and Mussa, J. 2002. Comparison of three secondary traits as determinants of grain yield in faba bean on waterlogged vertisols. Jounal of Genet Breed, 56: 317–326. Gemechu, K., Mussa, J. and Tezera W. 2006. Faba Bean (Vicia faba L.) Genetics and Breeding Research in Ethiopia: A Review. pp. 42–52 In: Kemal Ali, Gemechu Keneni, Seid Ahmed, Rajendra Malhotra, Surendra Beniwal, Khaled Makkouk and M.H. Halila (eds.). Food and forage legumes of Ethiopia: Progress and prospects. Proceedings of a Workshop on Food and Forage Legumes. 22-26 Sept 2003, Addis Ababa, Ethiopia. ICARDA, Aleppo, Syria. ISBN 92- 9127–185–4. pp351. 222 Getenet, Y. and Yehizbalem, A. 2017. Adaptation of faba bean varieties for yield, for yield components and against faba bean gall (Olpidium viciae Kusano) disease in South Gondar, Ethiopia. The Crop Journal. https://Doi.org/10.1016/j.cj.2017.05.007. Hailu, E., Getaneh, G., Sefera, T., Tadesse, N., Bitew, B., Boydom, A., Kassa D. and Temesgen T. 2014. Faba bean gall; A new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Sciences and Technology. doi.org/10.4172/2329-8863.1000144. Hongyu, K., Penna, M.G., Araujo, LB. and Dias, C.T.S. 2014. Statistical analysis of yield trials by AMMI analysis of genotype by environment interaction. Biometrical Letters 51(2): 89–102. Jensen, E.S., Peoples, M.B., Hauggaard-Nielsen, H., 2010. Faba bean in cropping systems. Field Crops Research, 115: 203-216. Kusano, S. 1912. The life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of the College of Agriculture, 4: 142–199. Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. 48p. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. Negussie, T., Seid, A.. Gemechu, K.. Wulita, W., Beyene, B., Asnakeh, T., Anteneh, A. and Yohannes, E. 2018. Faba bean gall in Ethiopia. Current situation and possible solutions. Seventh International Food Legume Research Conference, Marrakesh, Morocco. Ortiz, R., Wagoire, W., Hill, J., Chandra, S., Madsen, S. and Stolen, O. 2001. Heritability of and correlations among genotype by environment stability statistics for grain yield in bread wheat. TAG Theoretical and Applied Genetics, 103 (3: 469–474. Papastylianou, P., Vlachostergios, D.N., Dordas, C., Tigka, E., Papakaloudis, P., Kargiotidou, A., Pratsinakis, E., Koskosidis, A., Pankou, C. and Kousta, A. 2021. Genotype X Environment Interaction Analysis of Faba Bean (Vicia faba.) for Biomass and Seed Yield across Different Environments. Sustainability, 13: 2586 Sahile, S., Fininsa, C., Sakhuja, P.K. and Seid, A. 2008. Effect of mixed cropping and fungicides on chocolate spot (Botrytis fabae) of faba bean (Vicia faba) in Ethiopia. Crop Protection, 27: 275–282. SAS (Statistical Analysis System) software,. 2014. SAS User’s Guide. SAS Institute. Cary, North Carolina, USA. Sharma, M., Kiran Babu, T., Gaur, P.M., Ghosh, R., Rameshwar, T., Chaudhary, R.G., Upadhyay, J.P., Gupta, O., Saxena, D.R., Kaur, L., Dubey, S.C., Anandani, V.P., Harer, P.N., Rathore, A. 223 and Pande, S. 2012. Identification and multi-environment validation of resistance to Fusarium oxysprum f.sp. ciceris in chickpea. Field Crops Research, 135: 82–88. Sharma, M., Telangre, R., Ghosh, R. and Pande, S. 2015. Multi-environment field testing to identify broad, stable resistance to sterility mosaic disease of pigeon pea. Journal of Genetic Plant Pathological Society of Japan, 81:249–259. Tadesse, N., Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C. and Abraham, A. 2008. Review of research on diseases of food legumes. In Abraham T (Ed.). Increasing Crop Production through Improved Plant Protection. Addis Ababa, Ethiopia: PPSE and EIAR, Addis Ababa, Ethiopia, 598 pp. Tamene, T., Gemechu, K. and Hussein, M. 2015. Genetic progresses from over three decades of faba bean (Vicia faba L.) breeding in Ethiopia. Australian Journal of Crop Science, 9:41–48. Teklay, A., Tsegaye, B., Ymane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Science, 2: 33–38. Teklay, A., Yemane, N., Muez, M., Adhiena, M., Assef, W. and Hadas, B. 2015. Genotype by environment interaction of some faba bean genotypes under diverse broomrape environments of Tigray, Ethiopia. Journal of Plant Breeding Crop Science, 7: 79–86. Terefe, H., Fininsa, C., Sahile, S. and Tesfaye, K. 2015. Effect of temperature on growth and sporulation of Botrytis fabae, and resistance reactions of faba bean against the pathogen. Journal of Plant Pathology and Microbiology, 6: 285–268. Tesfay, M., Tilahun, A. and Russell, A. 2010. Genotype x envieronment interaction and their translational implication. Journal of the environment. https://doi.org/10.2217/pme.10.75. Torres, A.M., Roman, B., Avila, C.M., Satovic, Z., Rubiales, D., Sillero, J.C., Cubero, J.I., Moreno, M. T., 2006. Faba bean breeding for resistance against biotic stresses: Towards application of marker technology. Euphytica, 147: 67–80. Villegas-Fernandez, A.M., Sillero, J.C., Emeran, A.A., Flores, F. and Rubiales, D. 2011. Multiple-disease resistance in Vicia faba: Multi-environment field testing for identification of combined resistance to rust and chocolate spot. Field Crops Research, 124:59–65. Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) Using varieties and fungicides in north Shoa, central Ethiopia. Msc Thesis, Haramaya University, Haramaya, Ethiopia. 224 Xing, Z.S. 1984. Preliminary study oiler bacteria blister prevention and treatment of disease blister beans. Plant Pathology, 114: 165–173. Xu, Y. 2010. Molecular plant breeding. CABI (Centre for Agriculture and Biosciences International), P. 734. Yan, W. 2002. Singular-value partitioning in biplot analysis of multi-environmental trial data. Agronomy Journal, 94: 990–996. Yan, J.M. 2012. Study on Blister Disease of Broad Bean Caused by Olpidium Viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. Yan, W. 2011. GGE Biplot and AMMI Graphs for the Genotype by Environment Data Analysis. Journal of the Indian Society of Agricultural Statistics, 65(2): 181–193. Yan, W. 2014. Crop variety trials data management and analysis. Wiley, Hoboken. Yan, W. Hunt L.A., Sheng Q., Szlavnics Z. 2000. Cultivar evaluation and megaenvironment investigation based on the GGE biplot. Crop Sciences, 40(3): 597–605. Yan, W. and Kang, M.S. 2003. GGE biplot analysis: A Graphical Tool for Breeders, Geneticists and Agronomist. CRC press, Boca Raton. 271 pp. Yan, W. Kang, M.S. Ma, B. Woods, S. Cornelius, P.L. 2007. GGE biplot and AMMI analysis of genotype by environment data. Crop Sciences, 47: 643–655. Yan, W. Tinker NA. 2005. An integrated biplot analysis system for displaying, interpreting, and exploring genotype by environment interaction. Crop Sciences, 45:1004–1016. You, M.P., Eshete, B.B., Kemal, S.A., van Leur, J. and Barbetti, M.J. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology. https://doi.org/10.1111. Zobel, R.W., Wright, M.S., and Gauch, H.G. 1988. Statistical analysis of a yield trial. Agronomy Journal, 80:388–393. 225 Paper VI Integrating Genotype, Fungicide and Spray Schedule Reduces Gall (Physoderma) Disease Progression and Enhances Grain Yield in Faba Bean Beyene Bitew 1, Chemeda Fininsa 2, Habtamu Terefe 2*, and Seid Ahmed3 1Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia 2School of Plant Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 3Biodiversity and Crop Improvement Program, International Center for Agricultural Research in the Dry Areas (ICARDA), B.P. 6299, Rabat, Morocco *Corresponding author: habmam21@gmail.com; Tel.: +251911743338 Abstract Faba bean is the main cool season food legume crop in Ethiopia. However, productivity of the crop is constrained by faba bean gall (FBG), which is an emerging and the most destructive disease in the highland major faba bean growing areas of Ethiopia. Field experiments were conducted on farmers’ field during the 2018 and 2019 main cropping seasons in Bassona Worana district, Ethiopia; to develop integrated management options that combine host resistance and fungicides to reduce FBG disease and minimize yield losses, and to determine fungicide spraying schedules under natural conditions. The treatments included three faba bean genotypes (Degaga and Gora as moderately tolerant varieties and local as susceptible cultivar), two fungicides and two application schedules, which were factorial arranged in randomized complete block design with three replications. Faba bean gall disease severity, grain yield and yield-components data were collected and subjected to analysis of variance using the PROC GLM SAS procedure. The result revealed that integration of faba bean genotypes and fungicides significantly (P < 0.05) reduced FBG disease epidemics and increased grain yield. Faba bean varieties showed different responses in affecting disease parameters. Variety Gora sprayed Bayleton with the rate of 0.7 kg ha–1 at 10 days interval showed low mean FBG disease severity (21.67 and 10%), AUDPC value (1866.7 and 751.7%-days), low disease progress rate (0.0125 and 0.0121units day-1) and high mean grain yield (3.7 and 5.03 t ha–1) in 2018 and 2019, respectively. High AUDPC value of 3332.3%-days was calculated from the local unsprayed cultivar in the 2018 cropping season. Similarly, high net benefit and marginal rates of return (7.16 and 8.92) were obtained from variety Degaga and local cultivar sprayed Bayleton in 2018 and 2019 respectively. High marginal rates of return of (6.59 and 8.85) were obtained from varaity Gora in 2018 and 2019 in that order. The study showed that integration of moderately tolerant faba bean 226 varieties and Bayleton fungicide, and spraying at the onset of the disease at 10 days intervals/aligning with seedling, vegetative, flowering and podding growth stages are recommendable to manage FBG disease progression and increase grain yield of faba bean. Keywords: AUDPC, Disease progress rate, Faba bean, Fungicide, Gall disease, Severity. 1. Introduction Faba bean (Vicia faba L.) is one of the earliest domesticated and the major cool season food and feed legume crop grown in many countries worldwide (Duc et al., 2015). It is grown and adapted under a wide range of environmental conditions (Jensen et al., 2010; Stoddard et al., 2010). In Ethiopia, the crop takes the largest share of the area and production of pulse crops grown with a current productivity of 2.12 t ha–1 (CSA, 2021). Over 4.1 million households grow the crop and covering more than half a million hectare, and a total production of more than one million tons grain yield (CSA, 2021). The crop is the key and cheap source of good quality dietary protein for human food, and the straw is used as animal feed in the crop-livestock farming systems of the highlands of Ethiopia. In crop rotation practices, faba bean plays a significant role in soil fertility restoration for the subsequent cropping seasons (Agegnehu and Fessehaie, 2006; Jensen et al., 2010; Sillero et al., 2010), and breaking of disease and other pest cycles are among the main benefits of faba bean– cereal rotations. In spite of its multiple benefits to smallholder farming communities, faba bean productivity in Ethiopia is very low as compared to international average mean grain yield of 3.7 t ha– 1 (FAOSTAT, 2018). Low productivity of faba bean in Ethiopia is due to the susceptibility of the crop to a number of biotic and abiotic constraints and lack of improved varieties, and poor agronomic practices (Mussa et al., 2008). The major abiotic constraints for faba bean productions included water logging, frost, moisture stress, poor soil fertility, soil acidity and poor cultural practices. On the other hand, faba bean is highly affected by biotic factors particularly, fungal diseases (foliar and root), weeds (broad and grass), parasitic weeds (Orobanche spp.), and field and storage pests (Bruchides). Among the diseases, chocolate spot (Botrytis fabae), ascochyta blight (Aschochyta fabae), rust (Uromyces fabae) and black root rot (Fusarium solani and F. avenaceum) are the most economically important diseases that cause significant yield losses in Ethiopia (Sahile et al., 2008; Terefe et al., 2015; Gemechu et al., 2016). At present, faba bean production is seriously challenged and threatened by faba bean gall (FBG) disease which was reported to be caused by Olpidum viciae; however, it is identified that Physoderma causes FBG disease in Ethiopia (You et al., 2021). Faba bean gall is an emerging and 227 devastating disease in major faba bean growing areas of the country, mainly in Amhara, Oromia and some part of South Tigray (Hailu et al., 2014). In hot spot areas and susceptible faba bean varieties, total crop failure was observed on farmers' fields (Teklay et al., 2014). Similarly, reports showed that FBG forming disease has become increasingly a serious problem in China since the 1970s and more than 4000 ha of faba bean fields were affected, and yield losses of up to 20% has been reported (Li- juan et al., 1993). In Ethiopia, complete crop failure were reported in North Shoa (Amhara and Oromia), South Gondar, South Tigray and East Gojjam (Hailu et al, 2014, Teklay et al., 2014). At the moment, the disease is expanding to new faba bean growing areas of West Shoa and East Wollega (Challa et al., 2017), and North Gondar (Anteneh et al., 2018). Even though the disease causes a major threat to faba bean production, managing the gall disease has not been seriously attempted. There are notable knowledge gaps in relation to the disease epidemiology and its management. Released faba bean varieties and the landraces are not resistant to FBG disease. Despite the fact, management options have not been developed and recommended for severely affected areas. Therefore, the objectives of this study were to (1) develop integrated management options that combine host resistance and fungicides to reduce FBG disease progression under field conditions and (2) determine fungicide spray schedules for the management of FBG disease. 2. Materials and Methods 2.1. Description of the Study Area Field experiments were carried out in Bassona Worana district on farmers’ field during the 2018 and 2019 main cropping seasons. Bassona Worana is located at 9o 41′ N latitude and 39o 31′ E longitude, and at an altitude of 2980 meters above sea level (m.a.s.l.), North Shoa, Ethiopia. The area is characterized by light cambisol and faba bean is the major pulse crop cultivated and yearly rotated with cereals, mainly barley and wheat crops. The district receives bimodal rainfall where the main rainy season covers June to September and the short rainy season starts from January and ends in May. Faba bean grows in both seasons, and under supplementary irrigation during short rainy season, when there is deficient rainfall distribution. The rainfall in the main cropping season reaches 1000 to 1200 mm and the mean minimum and maximum temperatures are reported as 6 oC and 19 oC, respectively (DBARC, 2005). The monthly mean maximum and minimum temperatures (oC) and total rainfall (mm) during 2018 and 2019 were obtained from the weather station of Debre Birhan Agricultural Research Site and depicted hereunder Figure 1. The relative humidity (%) ranged from 49.5 to 90.5% in 2018 and 58.5 to 90.5% in the 2019 cropping seasons. 228 500 2018 35 450 30 400 350 25 300 20 250 200 15 150 10 100 50 5 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax 500 2019 35 450 30 400 350 25 300 20 250 200 15 150 10 100 50 5 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months RF Tmin Tmax Figure 1 Monthly mean minimum and maximum temperatures (°C) and rainfall (mm) data during the 2018 and 2019 cropping seasons at Bassona Worana district of North Shoa, Ethiopia. 2.2. Experimental Materials, Treatments and Design Treatment combinations consisted of three faba bean genotypes, two fungicides, two spray schedules, and three controls (unsprayed plots) with a total of 15 treatments (Table 1). Two moderately tolerant (Gora and Degaga) faba bean varieties-based on field evaluation (Wulita, 2015; Getenet and Yezbalem, 2017) and one susceptible farmers’ local cultivar to FBG disease were used in the experiment. Two systemic foliar fungicides Bayleton 25 WP and Ridomil Gold MZ 68 WG and two 229 Rainfall (mm) Rainfall (mm) Temperature (oC) Temperature (oC) fungicide spray schedules, which were set at 10 and 15 days interval, were also used in the trial. The treatments were factorial arranged in a randomized complete block design (RCBD) with three replications. The plot size employed was 4.8 m2 (2.4 m × 2 m), and each plot was arranged to have 6 rows of faba bean. Spacing between blocks and plots was 1.5 m, while spacing between rows and plants were 0.4 m and 0.1 m, respectively. Planting was done on 14 June 2018 and 18 June 2019. Blended NPS (38 kg P2O5, 19 kg N and 7kg S) fertilizer was applied once at the time of planting at a rate of 121 kg ha–1. Bayleton and Ridomil Gold fungicides were applied according to the manufacturers' recommendations at the rates of 0.7 kg ha–1 and 2.5 kg ha–1, respectively. The amount of water used to spray fungicides was also based on the manufacturers’ recommendations for each fungicide. The first spray started at the onset of typical disease symptom at 32 days after planting (DAP) in 2018 and 28 DAP 2019 cropping seasons. Five meter by three meter plastic sheet were used in the wind side direction by moving to each plot while spraying to protect draft effect of fungicides for each treatment. Seeding and fertilizer applications were done manually and weeding and other cultural practices were uniformly done for each plot. Table 1. Treatment combinations of faba bean genotype x fungicide x spray schedule for the management of faba bean gall disease in Bassona Worana district of North Shoa, Ethiopia, during the 2018 and 2019 main cropping seasons S/N Treatments and their combinations 1 Degaga + Bayleton 25 WP + 10 days interval 2 Degaga + Bayleton 25 WP + 15 days interval 3 Degaga + Ridomil Gold MZ 68 WG + 10 days interval 4 Degaga + Ridomil Gold MZ 68 WG + 15 days interval 5 Degaga + no spray (control) 6 Gora + Bayleton 25 WP + 10 days interval 7 Gora + Bayleton 25 WP + 15 days interval 8 Gora + Ridomil Gold MZ 68 WG + 10 days interval 9 Gora + Ridomil Gold MZ 68 WG + 15 days interval 10 Gora + no spray (control) 11 Local + Bayleton 25 WP + 10 days interval 12 Local + Bayleton 25 WP + 15 days interval 13 Local + Ridomil Gold MZ 68 WG + 10 days interval 14 Local + Ridomil Gold MZ 68 WG + 15 days interval 15 Local + no spray (control) 230 2.3. Disease Assessment Date of disease onset, disease incidence and severity were assessed during the epidemic periods of the disease. Faba bean gall disease severity was recorded at 10 days interval starting from date of disease onset, which were at 32 and 28 DAP in 2018 and 2019, respectively. Disease incidence was calculated using the following formula: Number of plants showing disease symptoms per plot Disease incidence (%) = Total number of plants assessed per plot X 100 For disease severity measurements, 12 plants from the central rows of each plot were randomly tagged prior to the appearance of typical disease symptoms and used for the assessment. Disease severity was rated using a modified 0–9 scoring scale of Ding et al. (1993); where 0 = no symptom, 1 = very small and few green gall and sunken lesions on the leaves, 2 = very small and green gall and sunken lesions, 3 = many green gall and sunken small lesions, 4 = many small gall and sunken lesions, and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants. A total of seven severity assessments were made until the disease starts to decline and changes diminished between records. Severity scores were converted into percentage severity index (PSI) for analysis (Wheeler, 1969) as follows: Sum of numerical ratings PSI = Number of plants scored x maximum score on the scale X 100 Area under disease progress curve (AUDPC) and disease progress rate were calculated from disease severity data. Percent disease severity calculated for each plot during the course of the experiment was summarized and used to calculate AUDPC (Campbell and Madden, 1990) based on number of ratings at the intervals of 10 day disease assessment and expressed in %-days and computed using the following formula: (xi + xi + 1) AUDPC = 2 (ti + 1 − ti) Where, Xi = disease severity at the ith assessment, ti = the time of the ith assessment in days from the first measurement date and n = total number of disease assessments made during the epidemic period. 231 2.4. Growth and Yield Parameters Assessment Growth and yield parameters such as days to 50% seedling emergence, days to 50% flowering, days to 90% physiological maturity, plant height (cm), number of pods per plant, number of seeds per pod, 100 seed weight (g) and grain yield (g plot–1) were recorded from each experimental plot. Plant height per plot was taken at harvest from 10 pre-tagged plants and number of pod per plant and number of seed per pod was recorded from randomly selected five plants, and then averaged for final data records. Grain yield was determined from four harvestable central rows for each treatment, and was converted to t ha–1 at 10% adjusted grain moisture content (Birru, 1979). Hundred seed weight was also determined from randomly taken 100 seeds from each plot grain yield at 10% moisture content and expressed in grams. 2.5. Data Analysis The two years data was analyzed separately due to wide differences between the two years of disease parameters, yield, and yield component data. Faba bean gall disease incidence, severity, AUDPC, grain yield, and yield components for each treatment were subjected to analysis of variance (ANOVA) using the PROC GLM procedure (SAS, 2014) to determine the response of faba bean varieties, effect of fungicides and spray schedules. Mean separations among treatments were done using Duncan's multiple range test (DMRT) at 5% probability level (Gomez and Gomez, 1984). To determine disease progress rate from the linear regression, fitness of two epidemiological models were tested based on the magnitude of coefficient of determination (R2) and residuals standard error (SE) obtained per model (Campbell and Madden, 1990), and finally, the Logistic model, ln [(y/1-y)] (Van der Plank, 1963) showed higher R2 and lower SE than the Gompertz model, and used to determine the disease progression from the linear regression for each separate treatment. The correlation analysis was employed based on Pearson's correlation coefficient to examine the association between faba bean gall disease parameters and yield and yield components of faba bean. 2.6. Partial Budget Analysis Cost-benefit analysis was done to determine profitability of the study based on the procedure employed by CIMMYT (1988). Marginal return is the rate of return for a marginal increase in an investment. Approximately, this is the additional output resulting from a one-unit increase in the use of a variable input, while other inputs are constant. To measure the increase in net return associated with each additional unit of cost (marginal cost), the marginal rate of return (MRR) was estimated using the following formula: 232 NB2 − NB1 MRR = TCI2 − TCI1 Where, MRR = marginal rate of return; NB1 = net benefit obtained without using fungicide; NB2 = net benefit obtained after using fungicide; and TCI1 = total costs of input for control treatment, and TCI2 = the total costs of input for fungicide sprayed treatments. The costs of fungicides, improved varieties, and local cultivar used in the study were estimated based on the local market price value during the harvest periods each cropping season. 3. Results 3.1. Rate of Faba Bean Gall Disease Development Mean disease progress rates showed significant variations among treatments during the two consecutive cropping seasons (Table 2). In 2018, the highest disease progress rates were calculated from unsprayed plots of variety Degaga, Gora and the local cultivar, which were 0.0571, 0.0531 and 0.0604 unit’s day–1, respectively. Among fungicide sprayed plots, variety Degaga, Gora and the local cultivar showed the highest disease progress rates when sprayed with Ridomil Gold at 15 days interval. Spraying Bayleton at 10 days interval showed lower disease progress rate in both faba bean varieties and the local cultivar than the other spray schedule (Table 2). Similarly, in the 2019 cropping season, the highest disease progress rates (0.0324, 0.0322 and 0.0388 units day–1) were noted on unsprayed plots of Degaga, Gora and local cultivar in that order. Both faba bean varieties and the local cultivar sprayed with Bayleton at 10 days interval showed low disease progress rate. The results showed that the disease progress rate was relatively faster on the variety Degaga and the local cultivar in both epidemic seasons. Moreover, gall disease progressed relatively at a higher rate on Ridomil Gold sprayed plots than on plots sprayed with Bayleton over the two cropping seasons. Spraying fungicides at every 10 days interval reduced disease progress rate, regardless of the faba bean genotypes and the fungicides evaluated in the study (Table 2). Spraying Bayleton at 10 days interval recorded 61.65%, 76.46% and 69.87% reduction in disease progress rate on unsprayed plots of variety Degaga, Gora and the local cultivar, respectively, in the 2018 cropping season. Likewise, spraying Bayleton at 10 days interval obtained 55.86%, 62.42% and 55.35% disease progress rate reduction on unsprayed plots of variety Degaga, Gora and the local cultivar, respectively, in 2019. 233 Table 2. Effects of faba bean genotypes, fungicides and spray schedules on disease progress rate (units day–1) at Bassona Worana, North Shoa, Ethiopia in the 2018 and 2019 main cropping seasons. Treatment combination 2018 cropping season a 2019 cropping season a Genotype Fungicide Spray Disease SE Intercept SE of R2 Disease SE Intercept SE of R2 schedule progress of rate intercept (%) progress of rate intercept (%) rate (r) rate (r) Degaga Bayleton 10 0.0219 0.0053 –3.215 0.335 71.0 0.0143 0.0059 –0.670 0.377 62.6 15 0.0243 0.0072 –3.044 0.454 76.7 0.0241 0.0072 –0.761 0.453 68.0 Ridomil 10 0.0295 0.0042 –3.422 0.267 71.7 0.0224 0.0071 –0.541 0.446 75.3 Gold 15 0.0301 0.0033 –3.412 0.208 62.2 0.0281 0.0057 –0.524 0.360 72.6 Control Unsprayed 0.0571 0.0041 –4.397 0.257 82.2 0.0324 0.0044 –1.834 0.276 80.6 Gora Bayleton 10 0.0125 0.0044 –2.227 0.275 73.4 0.0121 0.0058 –1.44 0.368 91.2 15 0.0233 0.0051 –2.700 0.324 93.9 0.0183 0.0079 –0.480 0.505 83.3 Ridomil 10 0.0221 0.0047 –2.237 0.298 83.3 0.0248 0.0058 –0.832 0.369 82.3 Gold 15 0.0341 0.0044 –2.503 0.276 83.4 0.0274 0.0084 –1.198 0.529 82.8 Control Unsprayed 0.0531 0.0032 –4.264 0.200 85.6 0.0322 0.0051 –1.874 0.322 93.6 Local Bayleton 10 0.0182 0.0049 –3.124 0.031 86.3 0.0181 0.0078 –1.151 0.495 82.3 15 0.0237 0.0057 –3.107 0.363 80.8 0.0306 0.0075 –1.162 0.471 77.8 Ridomil 10 0.0292 0.0036 –3.362 0.226 77.6 0.0292 0.0062 –0.850 0.392 85.8 Gold 15 0.0294 0.0030 –3.274 0.191 83.3 0.0343 0.0060 –0.779 0.382 89.8 Control Unsprayed 0.0604 0.0033 –4.357 0.206 87.5 0.0388 0.0049 –2.179 0.312 94.7 a Disease progress rate obtained from regression line of severity (%) against time of disease assessment (days); SE = standard error of rate and parameter estimate (intercept); and R2 = coefficient of determination for the Logistic epidemiological model. 234 3.2. Faba Bean Gall Disease Epidemic Faba bean gall disease symptom was observed at 32 DAP in 2018 and 28 DAP in 2019 cropping seasons; and FBG disease incidence recorded at 52 DAP and 48 DAP ranged from 86 to 100% in 2018 and 80.67 to 100% in 2019 cropping seasons (Tables 3 and 4). The highest disease incidence (100%) was recorded from unsprayed plots in all the tested faba bean genotypes in the 2018 cropping season. Faba bean genotypes, fungicides and spray schedules showed remarkably significant (P<0.05) differences for mean disease severity. Significant variation was also observed between fungicides and spraying schedules (Table 3). The overall mean disease severity in 2018 was higher than in the 2019 cropping season (Tables 3 and 4). Mean disease severity assessments showed significant differences among treatments starting from 62 DAP in 2018 and 58 DAP in 2019. In the 2018 cropping season, the highest (75%, 69% and 65.67%) mean disease severity was recorded at 92 DAP from untreated plots of the local, Degaga and Gora genotypes, respectively. Conversely, the lowest (21.67%) mean disease severity was obtained from variety Gora sprayed with Bayleton at 10 days interval at 92 DAP. Among fungicide sprayed plots, the highest disease severity (40%) was observed on the local cultivar sprayed with Ridomil Gold at 15 days interval at 92 DAP compared with others. Application of Ridomil Gold at every 10 days interval reduced disease severity more than application made at 15 days interval on both faba bean varieties and the local cultivar. Interaction effect of faba bean genotypes, fungicides and spray schedules were highly significant (P<0.05) on FBG disease severity in 2018 cropping season (Table 3). Integration of variety Gora and Bayleton fungicide sprayed at 10 days interval showed low disease severity as compared to other treatment combinations studied. In the 2019 cropping season, the highest mean disease severities of 70% (local cultivar), 63.33% (Degaga) and 60% (Gora) were recorded from unsprayed plots of test faba bean genotypes at 88 DAP. The local cultivar noted the highest mean disease severity (28.9%) when sprayed with Ridomil Gold at 15 days interval, while the lowest mean disease severity of 10% was registered on variety Gora variety sprayed with Bayleton at 10 days interval at 88 DAP. Variety Degaga and the local cultivar sprayed with Ridomil Gold at every 15 day interval showed relatively high disease severity as compared to Gora variety. Spraying Bayleton at every 10 days interval reduced faba bean gall disease severity as compared to Ridomil Gold at similar spray intervals during the experimental periods (Table 4). The interaction effects of test genotypes, fungicides and fungicide application schedules on disease severity were highly significant in the 2019 cropping season. 235 Table 3. Interaction effects of faba bean genotypes, fungicides and spray schedules on mean FBG disease incidence and severity at Bassona Worana, North Shoa, Ethiopia in the 2018 cropping season Treatment combination FBG disease FBG disease severity (%) 1 incidence (%) 1 Genotype Fungicide Spray schedule 52 DAP 62 DAP 72 DAP 82 DAP 92 DAP Degaga Bayleton 10 88.3cd 40.7bc 32.3cd 32.3b 26.3cd 15 90.3cd 39.3bc 36.0bcd 32.7b 30.0bcd Ridomil Gold 10 92.3bcd 41.0bc 37.0bcd 36.7b 30.9bcd 15 95.0abc 40.7bc 38.1bcd 36.0b 32.1bcd Control Unsprayed 100.0a 59.3a 63.0a 64.0a 69.0a Gora Bayleton 10 86.0d 34.3c 26.7d 26.7c 21.7e 15 87.0cd 41.3bc 31.7cd 30.7bc 26.7d Ridomil Gold 10 90.7d 37.76bc 34.3cd 34.3b 28.3cd 15 92.7bcd 39.7c 36.7cd 36.3b 31.7cd Control Unsprayed 100.0a 57.0a 59.7a 60.0a 65.7a Local Bayleton 10 90.0bcd 39.7bc 34.7cd 32.3b 28.7cd 15 91.7cd 41.0bc 36.7cd 35.7b 31.7cd Ridomil Gold 10 93.3bc 44.3b 38.0bc 36.3b 35.0bc 15 97.3ab 45.0b 41.0b 40.7b 40.0b Control Unsprayed 100.0a 61.0a 70a 72.7a 75.0a Mean 93.75 44.1 41.1 40.5 38.2 CV (%) 3.70 8.34 17.5 15.85 18.62 Genotype <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Genotype x fungicide ns ns ns ns 0.0001 Genotype x fungicide x spray schedule ns ns ns ns 0.0037 1FBG = faba bean gall disease and DAP = days after planting on which disease was assessed. Values within a column followed by different letter(s) are significantly varied at 5% level of probability using DMRT. 236 Table 4. Interaction effect of faba bean genotype, fungicides and spray schedules on mean FBG disease incidence and severity at Bassona Worana, North Shoa, Ethiopia during the 2019 cropping seasons Treatment combination FBG disease FBG disease severity (%)1 incidence (%)1 Genotype Fungicide Spray schedule 48 DAP 58 DAP 68 DAP 78 DAP 88 DAP Degaga Bayleton 10 86.3c 13.3cd 16.3bcd 14.3ef 13.3bcd 15 88.0bc 15.3cd 23.0bcd 20.0cdef 20.0bcd Ridomil Gold 10 91.3bc 15.7cd 26.3bc 24.0c 23.3bc 15 91.7bc 17.3cd 26.3bc 23.7c 23.3bc Control Unsprayed 99.3a 26.7b 60.3a 61.0ab 63.3a Gora Bayleton 10 80.7d 12.7d 13.0d 10.0f 10.0e 15 86.7c 13.3cd 16.3cd 16.0def 13.3cd Ridomil Gold 10 87.0c 13.0d 13.1d 10.0f 10.1e 15 91.7bc 14.3cd 16.7cd 13.7ef 13.7cd Control Unsprayed 97.2ab 26.3b 55.0b 57.0b 60.0a Local Bayleton 10 89.0c 13.0d 17.3cd 15.3ef 14.3cd 15 90.7bc 13.3cd 26.3bcd 26.0cd 23.3bcd Ridomil Gold 10 93.0bc 15.7cd 29.7bc 27.0c 26.7bc 15 93.0bc 19.3c 32.0bc 30.0c 28.9bc Control Unsprayed 100.0a 35.3a 68.0a 69.7a 70.0a Mean 91.50 17.6 29.3 27.8 27.6 CV (%) 3.74 17.6 25.0 23.9 24.9 Genotype <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Genotype x Fungicide ns ns ns ns 0.0018 Genotype x Fungicide x Spray schedule ns ns ns ns 0.0010 1FBG = faba bean gall disease and DAP = days after planting on which disease was assessed. Values within a column followed by different letter(s) are significantly varied at 5% level of probability using DMRT. 237 3.3. Area under Disease Progress Curve (AUDPC) The AUDPC values for the tested faba bean varieties and the local cultivar in both cropping seasons are displayed in Figure 2. There were highly significant (P<0.0001) differences among treatments for AUDPC values. Reduced AUDPC values were observed on plots sprayed with Bayleton at 10 days interval. For instance, application of Bayleton at 10 days interval on variety Gora showed the lowest AUDPC value compared with the rest of the treatment combinations evaluated in the experiment (Figure 2). In 2018 cropping season, the highest mean AUDPC values of 3332.3, 3246.0 and 3223.3%-days were calculated from unsprayed plots of the local cultivar, Degaga and Gora varieties in that order. Among fungicide sprays, the highest (2390%-days) mean AUDPC value was obtained from the local cultivar sprayed with Ridomil Gold at 15 days interval and the lowest AUDPC value was recorded from variety Gora sprayed with Bayleton at 10 days intervals, which were 1866.7%-days. Similarly, the highest mean AUDPC values of 2204.8%-days (local cultivar), 1926.7%-days (Degaga) and 1760%- days (Gora) were computed from unsprayed plots of each respective faba bean genotype in the 2019 cropping season. Among fungicide sprayed faba bean genotypes, the highest mean AUDPC value of 1214.2%-days was recorded from the local cultivar sprayed with Ridomil Gold at 15 days interval and the lowest AUDPC value of 751.7%-days was calculated on variety Gora sprayed with Bayleton at 10 days interval. The mean AUDPC values in 2018 were comparably higher than in the 2019 cropping season. 238 2018 4000 3500 3000 2500 2000 Degaga 1500 1000 Gora 500 Local 0 10 days 15 days 10 days 15 days Bayleton Ridomil Gold Controls Fungicide and application schedules 2019 3000 2500 2000 1500 Degaga 1000 Gora 500 Local 0 10d ays 15 days 10 days 15 days Bayleton Ridomil Gold Controls Fungicides and application schedules Figure 2. Mean AUDPC (%-days) values of faba bean gall disease as influenced by faba bean genotypes, fungicides and spray schedules at Bassona Worana, North Shoa, Ethiopia in 2018 and 2019. 3.4. Disease Progress Curve The disease progress curve presented at figure 3 and 4 were used to show FBG disease severity on unsprayed and maximum Bayleton 25 WP sprayed plots on variety Degaga, Gora and local cultivar. On unsprayed plots, FBG disease severity showed progression at every day in all faba bean genotypes tested starting from the onset of the disease in both cropping seasons (Figure 3). The effects of 239 AUDPC (%-days) AUDPC (%-days) fungicide application at 10 and 15 days interval for the management of FBG were observed on trends of disease progression curves in all faba bean genotypes. Spraying Bayleton 25 WP at 10 days interval showed more reduction of FBG disease pressure (Figure 4). Disease progress curve on genotypes sprayed with Bayleton 25 WP at 15 days, Ridomil Gold MZ 68 WG at 10 and 15 days intervals were laid between unsprayed and Bayleton 25 WP sprayed at 10 days intervals. 2018 80 60 40 Degaga 20 Gora Local 0 32 42 52 62 72 82 92 Days after planting 2019 80 60 40 Degaga 20 Gora Local 0 28 38 48 58 68 78 88 Days after planting Figure 3. Disease progress curve of faba bean gall (FBG) on unsprayed faba genotypes during the 2018 and 2019 main cropping seasons, Bassona Worana, North Shoa, Ethiopia. 240 Disease severity (%) Disease severity (%) 2018 50 40 30 Degaga 20 10 Gora 0 Local 32 42 52 62 72 82 92 Days after planting 2019 20 15 10 Degaga 5 Gora 0 Local 28 38 48 58 68 78 88 Days after planting Figure 4. Disease progress curve of faba bean gall (FBG) during 2018 and 2019 main cropping seasons on Bayleton fungicide sprayed plots at 10 days interval in Bassona Worana, North Shoa, Ethiopia 3.5. Faba Bean Grain Yield Faba bean genotypes sprayed with fungicides at different spray schedules showed very highly significant (P<0.0001) differences in both cropping seasons. In the two years, unsprayed faba bean varieties and the local cultivar gained low grain yield as compared to fungicide sprayed plots. Bayleton and Ridomil Gold spraying at 10 and 15 days intervals showed very highly significant (P<0.0001) differences among genotypes for mean grain yield in both cropping years (Table 5). In 2018, high mean grain yields of 3.7 t ha–1 (Gora), 3.41 t ha–1 (Degaga) and 3.12 t ha–1 (local) were harvested from sprayed plots of evaluated faba bean genotypes. On the other hand, low mean grain yields of 1.28, 1.41 and 1.85 t ha–1 were harvested from unsprayed plots of the variety Degaga, local cultivar and Gora respectively. Closely similar trends were noticed during the 2019 cropping season (Table 5). Faba bean genotypes sprayed with Bayleton at 10 days interval showed relatively higher mean grain yields than other spray schedule. The mean grain yields obtained in 2019 was higher than 241 Disease severity (%) Disease severity (%) average grain yields recorded in the 2018 cropping season. Integration of variety Gora with Bayleton fungicide at 10 days interval showed higher grain yield than other treatment combinations. 3.6. Faba Bean Hundred Seed Weight In the 2018 cropping season, hundred seed weight showed very highly significant (P<0.0001) variability among the treatments imposed during the epidemic period. Similarly, spraying Bayleton fungicide on faba bean genotypes at 10 and 15 days intervals showed significant difference in hundred seed weight during the 2019 cropping season. High hundred seed weights of 77.68 g (2018) and 81.2 g (2019) were measured from Gora variety sprayed with Bayleton at 10 days interval, while low hundred seed weights (41.71 and 41.75g) were recorded on variety Degaga and the local cultivar in that order in 2018. Low hundred seed weights were gained from variety Degaga (42.9 g) and the local cultivar, which measured 43.8 g in 2019 (Table 5). 3.7. Faba Bean Yield Components The analysis of variance showed that there were no significant differences between days to 50% seedling emergence and days to flowering; however, there was strong difference between sprayed and unsprayed plots in days to physiological maturity both in 2018 and 2019 cropping seasons. Unsprayed plots showed relatively short maturity periods as compared to fungicide sprayed plots (data not shown). Significantly variable plant heights were noted between genotypes, fungicides and unsprayed plots in 2018. Among the tested faba bean genotypes, the highest plant height (86.1 cm) was recorded on variety Gora, followed by Degaga (77.1 cm) sprayed with Bayleton at 10 days interval. On the other hand, the lowest plant height (57.1 cm) was recorded on unsprayed local cultivar (Table 6). Also, the highest plant heights of 114, 113.9 and 113 cm were recorded on the local, Degaga and Gora genotypes, respectively, and the lowest (83.6 cm) was recorded on unsprayed local cultivar in 2019. In all evaluated genotypes, unsprayed plots showed low plant height in both cropping seasons. Moreover, high mean number of pods per plant was obtained from Degaga (9.9) and local (9.4) genotypes sprayed Bayleton at 10 days interval in 2018. In 2019 cropping season, high number of pod per plant was recorded on local (19.5) cultivar. Similarly, among the tested faba bean genotypes, high mean number of seeds per pod was recorded on variety Gora (3.3) and Degaga (3.3) sprayed fungicides at 10 days interval in 2018. Likewise, high mean number of seeds per pod (3.2) was recorded on variety Gora in 2019. Low mean number of seeds per pod was recorded on unsprayed local cultivar, Degaga and Gora varieties in both cropping seasons (Table 6). 242 Table 5. Interaction effects of faba bean genotypes, fungicides and spray schedules on grain yield and hundred seed weight of faba bean during the 2018 and 2019 cropping seasons, Bassona Worana, north Shoa, Ethiopia Treatment combination Grain yield (t ha–1) 100 seed weight (g) Genotype Fungicide Spray schedule 2018 2019 2018 2019 Degaga Bayleton 10 3.41ab 4.22bcd 47.03b 57.1bc 15 3.14cd 4.20bcd 46.09b 54.4bc Ridomil Gold 10 3.06bc 4.00cd 47.02b 50.3cde 15 2.48f 3.96cd 41.77b 48.0cde Control Unsprayed 1.28h 2.7e 41.75b 42.9e Gora Bayleton 10 3.70a 5.03a 77.68a 83.2a 15 3.58a 4.48b 76.72a 81.2a Ridomil Gold 10 3.53a 4.30bc 73.8a 81.2a 15 2.81de 4.24bcd 73.1a 80.37a Control Unsprayed 1.85g 2.87e 71.84a 70.3b Local Bayleton 10 3.12bc 4.58b 47.54b 52.2cd 15 3.11c 4.22bcd 43.14b 50.3cde Ridomil Gold 10 2.66ef 3.84d 46.02b 49.8cde 15 2.66ef 3.82d 43.64b 44.0de Control Unsprayed 1.41h 2.48e 41.71b 43.8e Mean 2.79 3.93 54.32 59.5 CV (%) 5.86 5.96 6.76 7.4 P (0.05) <0.0001 <0.0001 <0.0001 0.0001 Mean values within a column followed by different letter(s) vary significantly at 5% probability level using DMRT. 243 Table 6. Interaction effects of faba bean genotypes, fungicides and spray schedules on mean plant height, number of pods per plant and number of seeds per pod of faba bean during the 2018 and 2019 main cropping seasons, Bassona Worana, north Shoa, Ethiopia Treatment combination Plant height (cm) Number of pods plant–1 Number of seeds pod–1 Genotype Fungicide Spray schedule 2018 2019 2018 2019 2018 2019 Degaga Bayleton 10 77.2ab 113.9a 9.9a 16.5ab 2.6ab 3.0ab 15 70.1bcd 107.5abcd 8.3abcd 12.8cdef 2.6ab 2.4bcd Ridomil Gold 10 72.1abcd 105.8abcd 8.9abc 15.1cd 3.3a 2.6abcd 15 69.1bcd 101.6abcd 7.9abcde 11.7cdef 2.7ab 2.5bcd Control Unsprayed 67.1cd 94.9cde 5.6de 10.3def 2.4b 2.2cd Gora Bayleton 10 86.1a 113.0a 8.0abcd 11.2cdef 3.3a 3.2a 15 81.7ab 112.4ab 6.7bcde 10.1def 2.6ab 2.8abc Ridomil Gold 10 82.2a 96.6bcde 6.9bcde 10.0def 2.9ab 2.9ab 15 74.2abc 94.2cde 6.1cde 8.9ef 2.9ab 2.6abcd Control Unsprayed 61.6cd 92.0de 5.6de 8.1f 2.7ab 2.3cd Local Bayleton 10 75.3abc 114.8a 9.4ab 19.5a 3.0ab 2.6abcd 15 69.3bcd 109.1abc 8.6abc 13.4bcd 2.7ab 2.6abcd Ridomil Gold 10 69.9bcd 105.4abcd 8.6abc 16.8ab 2.9ab 2.7abcd 15 68.7bcd 98.8abcde 7.0bcde 12.9bcd 2.7ab 2.7abcd Control Unsprayed 57.1d 83.6e 5.1e 12.9bcd 2.3b 2.1d Mean 71.7 102.9 7.5 12.6 2.8 2.6 CV (%) 11.00 8.11 19.60 17.63 15.56 12.20 P (0.05) 0.0046 0.0001 0.0078 <0.0001 0.2800 0.0019 Mean values within a column followed by different letter(s) vary significantly at 5% probability level using DMRT. 244 3.8. Association of FBG Disease and Yield Parameters The correlation analysis showed that there was a significant and strong association between and among FBG disease parameters and grain yield and yield components and correlation coefficients are presented hereunder Table 7. The last disease severity showed positive strong correlation with AUDPC (r = 0.99***) and disease progress rate (r = 0.96***) in 2018. FBG disease severity showed negative correlation with grain yield (r = –0.94***), plant height (r = –0.81***), number of pod per plant (r = –0.73**) and number of seed per pod (r = –0.62*) at the last date of disease assessment in 2018. FBG disease severity and hundred seed weight attained negative (r = –0.25ns) weak correlation. The AUDPC and disease progress rate also maintained positive association among each other and negatively correlated with yield and yield components in 2018. The other disease severity records also showed closely similar trends as the correlations displayed by the last severity record. On the contrary, grain yield attained positive relationship with hundred seed weight (r = 0.39ns), plant height (r = 0.86***), number of pod per plant (r = 0.65**) and number of seed per pod (r = 0.58*). A positive correlation was also established among yield components in the 2018 cropping season. In the 2019 cropping season, similar trends were observed regarding associations among and between disease parameters, grain yield and yield-related components as the 2018 cropping season (Table 7). 245 Table 7. Coefficients of correlation (r) between faba bean gall disease parameters, grain yield and yield components of faba bean genotypes under different fungicides and spray schedules at Bassona Worana, North Shoa, Ethiopia, during the 2018 (upper diagonal) and 2019 (lower diagonal) cropping seasons Parameter a PSI1 PSI2 PSI3 AUDPC DPR GY HSW PH NPPP NSPP PSI1 1 0.99*** 0.99*** 0.97*** 0.97*** –0.95*** –0.29ns –0.84*** –0.71** –0.63* PSI2 0.99*** 1 0.99*** 0.97*** 0.97*** –0.93*** –0.26ns –0.81*** –0.73** –0.61* PSI3 0.99*** 0.99*** 1 0.99*** 0.96*** –0.94*** –0.25ns –0.81*** –0.73** –0.62* AUDPC 0.99*** 0.99*** 0.99*** 1 0.95*** –0.92*** –0.25ns –0.79*** –0.70** –0.61* DPR 0.77** 0.76** 0.74** 0.75*** 1 –0.95*** –0.22ns –0.82*** –0.76** –0.62* GY 0.96*** –0.96*** –0.97*** –0.97*** –0.81*** 1 0.39ns 0.86*** 0.65** 0.58* HSW –0.51ns –0.49ns –0.46ns –0.51ns –0.51ns 0.49* 1 0.58* 0.32ns 0.37ns PH –0.73** –0.72** –0.74** –0.74** –0.83*** 0.80*** 0.14ns 1 0.42* 0.57* NPPP –0.22ns –0.23ns –0.25ns –0.22ns –0.27ns 0.23ns 0.52* 0.53* 1 0.48* NSPP –0.84*** –0.83*** –0.83*** –0.82*** –0.76** 0.83*** 0.54* 0.67** 0.37* 1 a PSI1–PSI3 refer to percent severity index at the last three dates of disease assessments (72, 82 and 92 DAP in 2018, and 68, 78 and 88 DAP in 2019); AUDPC = Area under disease progress curve; DPR = Disease progress rate; GY = Grain yield; HSW = Hundred seed weight; PH = Plant height; NPPP = Number of pods per plant; and NSPP = Number of seeds per pod. *** = Very highly significant; ** = Highly significant; * = Significant; and ns = Non-significant at P<0.001, <0.01, <0.05 and P >0.05, respectively. 246 3.9. Partial Budget Analysis In the 2018 cropping season, the partial budge analysis showed that the highest net benefits were calculated from Gora ($1822.64 ha–1 ), Degaga ($1612.37 ha–1) varieties sprayed with Bayleton at 10 days interval. The net benefit of $1486.86 were obtained from local cultivar sprayed Bayleton at 15 days intervals. In contrast, low net benefits of $274.81 and $411.44 ha–1 were obtained from unsprayed plots of variety Degaga and the local cultivar, respectively (Table 8). Similarly, the highest net benefits ($3675.36 and $3311.5 ha–1) were calculated from variety Gora and the local cultivar sprayed with Bayleton at 10 days intervals, while the lowest net benefit of $1616.98 ha–1 was incurred from unsprayed plots of local cultivar, followed by variety Degaga ($1773.6 ha–1) and Gora ($1926.77 ha–1) in 2019 cropping season (Table 9). Net benefit obtained from each faba bean genotype was higher in 2019 than in 2018 cropping season. Moreover, the highest value of the marginal rate of return (7.16) was obtained from variety Degaga, followed by variety Gora (6.59) and the local cultivar (6.45) with Bayleton sprayed at 15 days interval in 2018 (Table 8). In the 2019 cropping season, the highest (8.92) marginal rate of return was computed from the local cultivar, followed by Degaga (7.55) varieties sprayed with Bayleton at 15 days interval (Table 9). However, high marginal rate of return (8.85) was calculated from Gora variety sprayed with Bayleton at 10 days intervals in 2019. The marginal rate of return obtained from plots sprayed with Bayleton fungicide was higher than the rate calculated from Ridomil Gold sprayed plots. In this regard, Bayleton sprayed plots at 15 days interval showed high marginal rate of return in all faba bean genotypes tested, and the mean marginal rate of return obtained in 2019 was higher than the rate recorded in 2018. Frequent application of fungicide spray showed less marginal rate of return. 247 Table 8. Partial budget analysis for the management of faba bean gall disease using genotypes, fungicides and spray schedules in Bassona Worana, north Shoa, Ethiopia, during the 2018 cropping season Treatment combination Grain Sales Production Marginal Total Net Marginal Marginal Genotype Fungicide Spray yield revenue cost cost cost benefit benefit rate of schedule (t ha–1) ($ t–1) ($ ha–1) ($ ha–1) ($ ha–1) ($ ha–1) ($ ha–1) return Degaga Bayleton 10 3.41 2472.25 653.19 206.67 859.86 1612.39 1337.58 6.47 15 3.14 2276.50 653.19 165.34 818.53 1457.97 1183.16 7.16 Ridomil G 10 3.06 2218.50 653.19 450.73 1103.92 1114.58 839.77 1.86 15 2.48 1798.00 653.19 360.59 1013.78 784.22 509.41 1.41 Control Unsprayed 1.28 928.00 653.19 0.0 653.19 274.81 0.0 0.0 Gora Bayleton 10 3.70 2682.50 653.19 206.67 859.86 1822.64 1134.58 5.49 15 3.58 2595.50 653.19 165.34 818.53 1776.97 1088.91 6.59 Ridomil G 10 3.53 2559.25 653.19 450.73 1103.92 1455.33 767.27 1.70 15 2.81 2037.25 653.19 360.59 1013.78 1023.47 335.41 0.93 Control Unsprayed 1.85 1341.25 653.19 0.0 653.19 688.06 0.0 0.00 Local Bayleton 10 3.12 2262.0 602.43 206.67 809.10 1452.9 1033.08 5.00 15 3.11 2254.75 602.43 165.34 767.77 1486.98 1067.16 6.45 Ridomil G 10 2.66 1928.50 602.43 450.73 1053.16 875.34 455.52 1.01 15 2.66 1928.50 602.43 360.59 963.02 965.48 545.66 1.51 Control Unsprayed 1.41 1022.25 602.43 0.0 602.43 419.82 0.0 0.00 Mean unit of price of grain yield per ton during 2018 was 725$ (at the exchange rate of 1$ = ETB 27.58) at the time of grain selling, and price of Bayleton and Ridomil Gold fungicides were $42.51 and $30.82 per each kg respectively. 248 Table 9. Partial budget analysis for the management of faba bean gall disease using genotypes, fungicides and spray schedules in Bassona Worana, north Shoa, Ethiopia, during the 2019 cropping season. Treatment combination Grain Sales Production Marginal Total Net Marginal Marginal Genotype Fungicide Spray yield revenue Cost cost cost benefit benefit rate of schedule ($ t–1) ($ ha–1 (t ha–1) ) ($ ha–1) ($ ha–1) ($ ha–1) ($ ha–1) return Degaga Bayleton 10 4.22 3802.22 659.10 197.57 856.67 2945.55 1171.95 5.93 15 4.20 3784.20 659.10 158.06 817.16 2967.04 1193.44 7.55 Ridomil G 10 4.00 3604.00 659.10 571.92 1231.02 2372.98 599.38 1.05 15 3.96 3567.96 659.10 457.54 1116.64 2451.32 677.72 1.48 Control Unsprayed 2.70 2432.70 659.10 0.0 659.10 1773.6 0.00 0.00 Gora Bayleton 10 5.03 4532.03 659.10 197.57 856.67 3675.36 1748.59 8.85 15 4.48 4036.48 659.10 158.06 817.16 3219.32 1292.55 8.18 Ridomil G 10 4.30 3874.30 659.10 571.92 1231.02 2643.28 716.51 1.25 15 4.24 3820.24 659.10 457.54 1116.64 2703.6 776.83 1.70 Control Unsprayed 2.87 2585.87 659.10 0.0 659.10 1926.77 0.00 0.00 Local Bayleton 10 4.58 4126.58 617.50 197.57 815.08 3311.5 1694.52 8.58 15 4.22 3802.22 617.50 158.06 775.56 3026.66 1409.68 8.92 Ridomil G 10 3.84 3459.84 617.50 571.92 1189.43 2270.41 653.43 1.14 15 3.82 3441.82 617.50 457.54 1075.04 2366.78 749.80 1.64 Control Unsprayed 2.48 2234.48 617.50 0.0 617.50 1616.98 0.00 0.00 Mean unit of price of grain yield per ton during 2019 was 901$ (at the exchange rate of 1$ = ETB 28.85) at the time of grain selling, and price of both Bayleton and Ridomil Gold per kg fungicides were $41.59. 249 4. Discussion The results of the current study revealed that a relatively high FBG disease epidemic was detected in 2018 than in the 2019 cropping season. Such high variation in gall pressure among cropping seasons might be due to variations on the amount of rainfall, relative humidity and air temperatures between the two seasons. The amount and frequency of rainfall, relative humidity and air temperatures in 2018 was higher than in 2019 as indicated in figure 1. In this regard, Teklay et al. (2018) reported that seasonal variation of FBG disease could be due to the prevailing weather conditions; and Yan (2012) also indicated that high relative humidity, air temperatures of 10 to 20 ℃, and high amount of rainfall favor zoospore releasing pathogen and increase epidemics of FBG disease. On the other hand, high rainfall and relative humidity could initiate primary and secondary infections in the pathogen and rain splash could enhance pathogen dispersal. Similarly, Ristaino and Gumpertz (2000) noted that disease increases rapidly when rainfall droplets cause splash dispersal of primary inoculum from the soil to the aerial parts of plants in oospore-forming pathogens. Of course, the dispersal processes in pathogens are known to have major effects on both the spatial and temporal distribution of epidemics (Jeger, 1999; Rossi and Caffi, 2012). Use of host resistant is one of the key methods to manage different plant diseases. Resistant varieties can be the simplest, practical, effective, time saving and economical method of disease management. The results of the study showed that use of different faba bean varieties and application of fungicides at different spray schedules reduced FBG disease severity, AUDPC and disease progress rate at variable levels. Among the evaluated varieties, Gora and Degaga were previously reported as moderately resistant to FBG disease (Wulita, 2015; Getenet and Yezbalem, 2017) though no variety is registered for FBG disease resistance in the country. However, faba bean varieties tested in Ethiopia showed different degrees of FBG disease reaction (DBARC progress report, 2015), and as well as faba bean breeding lines tested in China revealed variations, but were not resistant to FBG (Yan, 2012). Regardless of the treatments imposed, the genetic backgrounds of the faba bean genotypes could be responsible for the variation in their responses to FBG disease. For instance, higher mean disease severity and AUDPC values were recorded on the farmers’ local cultivar than the two released faba bean varieties even though unsprayed plots of Degaga and the local cultivar statistically on par to each 250 other, but different degrees of FBG disease reaction. On the other hand, sources of the planting materials could play undeniable role in influencing development of epidemics. Accordingly, variety Gora is obtained from hybridization and it showed relatively lower FBG disease severity than Degaga and the local cultivar in both years. Degaga is also obtained from introduced faba bean lines and both have different genetic sources. However, further genetic background studies are recommended to confirm the real genetic degree of tolerance to FBG disease epidemics in the test materials. In the present study, applications of fungicides sprays were started following the onset of typical disease symptom at seedling growth stage, and the FBG disease pressure began to reduce after three foliar applications of both fungicides. Both Bayleton and Ridomil Gold fungicides applied at 10 and 15 days spray intervals attained significant effect on FBG disease parameters. Bayleton spraying at 10 and 15 days intervals were found more effective than Ridomil Gold application among genotypes in the two years. This could be associated with the nature and active ingredients of the fungicides and/or differences in efficacy of the fungicides to manage the disease. That is, Ridomil Gold is a mixture of Metalaxyl and Mancozeb, where Mancozeb is a preventive fungicide that needs to be applied before the onset of a disease and requires very frequent spraying than Bayleton fungicide. Despite the variability observed among fungicides, foliar application of Bayleton at a rate of 0.7 kg ha–1 and Ridomil Gold at a rate of 2.5 kg ha–1 at 10 days interval strongly reduced FBG disease components. Similarly, Teferi et al. (2018) reported that both Bayleton and Ridomil Gold are effective to manage FBG disease in northern Ethiopia. Also, Carbendazim and Bayleton were reported to manage FBG disease in the form of foliar spray and seed dressing; however, seed dressing reduces percent seedling emergence in China (Li juan et al., 1993). Comparatively, results of the study revealed that Bayleton sprayed plots showed relatively reduced disease severity, AUDPC and disease progress rate compared with Ridomil Gold. The findings of the current study agreed with Li-Juan et al. (1993) who indicated that Bayleton was more effective fungicide to manage FBG disease than others in China. However, findings of Wulita (2015) noted that Matco (Metalaxyl 8% + Mancozeb 64% WP) was more effective than Bayleton and Ridomil Gold in central Ethiopia, which could partly be attributed to rate and time of applications of the fungicides. Despite the fact, Bayleton is registered for wheat rust 251 management and Ridomil Gold is registered to manage late blight of potato and downy mildew on different vegetable and fruit crops in Ethiopia (Belay et al., 2015). In addition, optimum application rate and mode of application of both fungicides to manage faba bean gall disease is yet not determined and lacks uniformity across different faba bean growing areas over the years in the country (Wulita, 2015; Bogale et al., 2016; Bekele et al., 2018; Teferi et al., 2018), implying that further studies are required to determine and recommend feasible, economical and appropriate rates of Bayleton and Ridomil Gold fungicides to manage FBG disease and reduce yield loss in the production conditions of Ethiopia. In the present study, integration of relatively tolerant faba bean genotype with fungicide at different spray intervals reduced disease parameters and increased grain yield and yield components at different level. Combination of the tolerant faba bean variety Gora and Bayleton fungicide at 10 days intervals showed reduced disease parameters, but showed low marginal rate of return in 2018, however application of Bayleton at 15 days interval gained high marginal rate of return in all tested faba bean genotypes. In contrast, Gora variety sprayed Bayleton at 10 days intervals showed high marginal rate of return in 2019. Similarly, low marginal rate of return was recorded on Ridomil Gold sprayed plots than Bayleton. This study agree with Wulita (2015) report that Ridomil Gold sprayed on local faba bean cultivar showed low marginal rate of return than other fungicides evaluated. Low marginal rate of return on Ridomil Gold sprayed plots could due to high rate of Ridomil Gold fungicide per hectare that increases amount of fungicide cost. The frequency and spray interval depends on the type of fungicide applied, effectiveness of the fungicide and disease infestation. During high pathogen infestation and extended rainfall, FBG disease severity may increases and more fungicide spray frequency may be crucial. However fungicides are not environmentally safe and market price may not be affordable and application costs are increasing every year, hence developing tolerant faba bean varieties and integrating with Bayleton fungicide with minimum frequency of spray is essential. 5. Conclusions Recently, FBG disease is a serious threat to faba bean production in the highland faba bean growing areas of Ethiopia and causes significant yield reduction on local and improved faba bean varieties. In this study, considerable level of FBG disease reduction was achieved through integration of faba bean genotype, foliar application of fungicides and spraying schedules. The 252 result showed that spraying of Bayleton at 10 days interval starting at the onset of the disease aligning with seedling, vegetative, flowering and podding growth stages was more effective to reduce FBG disease severity, AUDPC and rate of disease development, and increased grain yield and yield components. Low FBG disease parameters and maximum grain yield was obtained from Gora variety sprayed with Bayleton at 10 days interval both in 2018 and 2019 main cropping seasons. Similarly, variety Degaga and local cultivar sprayed Bayleton at 10 days intervals showed FBG disease reductions. High FBG disease parameters and lower grain yield was obtained from local unsprayed plots in both seasons. However, high marginal rate of return was obtained from spraying Bayleton at 15 days intervals in all tested genotypes in 2018, and low marginal rate of return was obtained from Ridomil Gold sprayed plots in both seasons. This study confirmed that integration of tolerant varieties, application of effective fungicides and proper application schedules can increase grain yield and maximize net benefit. For immediate solution application of Bayleton with the rate of 0.7 kg ha−1 three to four sprays frequency with tolerant faba bean variety can be used to manage FBG disease and increase grain yield. Furthermore developing host resistance and more effective fungicides evaluation, and different cultural practices should be tested further to manage the disease and increase grain yield. 6. Acknowledgements The study was financially supported by Amhara Regional Research Institute (ARARI), Australian Center for International Agricultural Research (ACIAR) and the International Center for Agricultural Research in the Dryland Areas (ICARDA), and the institutes are duly acknowledged for supporting the study. This work is part of a PhD Dissertation research requirement at Haramaya University and the University is duly acknowledged. We also greatly acknowledge Mr. Aleme Belete and Amare Belachew for their assistances during field and laboratory data assessment and organization. 253 7. References Agegnehu, G. and Fessehaie, R. 2006. Response of faba bean to phosphate fertilizer and weed control on nitisols of Ethiopian highlands. Italian Journal of Agronomy, 2: 281–290. Anteneh, A., Yohannes, E., Mesganaw, G., Solomon, G. and Getachew, T. 2018. Survey of faba bean (Vicia faba L.) diseases in major faba bean growing districts of North Gondar, African Journal of Plant Science, 12(2): 32–36. Bekele, B., Wubit, D., Bekele, K. and Thangavel, S. 2018. Management of Faba Bean Gall Disease using Cultivars and Fungicides in North Showa Zone of Central Ethiopia. International Journal of Research in Agricultural Sciences, (5): 2348–3997. Belay, T.M., Mol, A.P.J. and Oosterveer, P. 2015. Pesticide use practices among smallholder vegetable farmers in Ethiopian Central Rift Valley. Journal of Environment, Development and Sustainability, 19:301–324 Birru, A. 1979. Agricultural field experiment management manual part III. Institute of Agricultural Research. Addis Ababa, Ethiopia. Bogale, N., Setu, B. and Mulugeta, T. 2016. Assessment of faba bean gall (Olpidium viciae (Kusano) in major faba bean (Vicia faba L.) growing areas of Northeastern Amhara, Ethiopia. Journal of Agriculture and Environment for International Development, 110: 87–95. Campbell, C.L. and Madden, V.L. 1990. Introduction to Plant Disease Epidemiology, New York: John Wiley and Sons, Inc. New York, USA. 532 pp. Challa, D. Abraham, N. Zerihun, A. and Meseret, T. 2017. Assessment of the Occurrence and Prevalence of Faba Bean Gall (Olpidium viciae) in Western Highlands of Oromiya, Ethiopia. Journal of Natural Sciences Research, 7(5): 63–67. CIMMYT (International Maize and Wheat Improvement Center). 1988. From Agronomic Data to Farmers’ Recommendations: Economic Training Manual. 79 pp. CSA (Central Statistics Agency). 2021. Agricultural sample survey (2021). Report on area and production of major crops. Statistical bulletin 578, Central Statistical Authority, Addis Ababa, Ethiopia. DBARC (Debre Birhan Agricultural Research Center). 2005. Debre Birhan Agricultural Research Center, Sub Center and Research sites description. Debre Birhan, Ethiopia. 254 DBARC (Debre Birhan Agricultural Research Center). 2015. Annual progress report, Debre Birhan, Ethiopia. Ding, G., Xung, L., Oifang, G., Pingxi, L., Dazaho, Y. and Ronghai, H. 1993. Evaluation and screening of faba bean germplasm in China. Faba Bean Newsletter, 32: 8–10. Duc, G., Aleksic, J.M., Marget, P., Mikic, A., Paull, J., Redden, R.J., Sass, O., Stoddard, F.L., Vandenberg, A., Vishnyakova, M. and Torres, A.M. 2015. Faba bean. Pp. 141–178. In: de Ron, AM. (eds.), Grain Legumes. Handbook of Plant Breeding, No. 10, Springer, New York, https://doi.org/10.1007/978-1-4939-2797-5_5. FAOSTAT (Food and Agriculture Organization of the United Nations, Statistical Database). 2018. Rome, Italy, website.faostat.fao.org. Gemechu, K., Asnake, F. and Million, E. 2016. Reflections on highland pulses improvement research in Ethiopia. Past achievements and future directions. Ethiopian Journal of Agricultural Sciences (EIAR 50th Year Jubilee Anniversary Special Issue), 17–50. Getenet, Y. and Yehizbalem, A. 2017. Adaptation of faba bean varieties for yield, for yield components and against faba bean gall (Olpidium viciae Kusano) disease in South Gondar, Ethiopia. The Crop Journal, 5(6): 560–566. Gomez, K.A. and Gomez, AA. 1984. Statistical Procedures for Agricultural Research. John Wiley and Sons, Singapore. Hailu E., Getaneh, G., Sefera, T., Tadesse, N., Bitew, B., Boydom, A., Kassa, D. and Temesgen, T. 2014. Faba bean gall; A new threat for faba bean (Vicia faba) production in Ethiopia. Advances in Crop Science Technology, 2(4): doi.org/10.4172/2329-8863.1000144. Jeger, M.J. 1999. Improved understanding of dispersal in crop pest and disease management: Current status and future directions of Agriculture. Meteorology, 97:331–349. Jensen, E.S., Peoples, M.B., Hauggaard-Nielsen, H. 2010. Faba bean in cropping systems. Field Crops Research, 115: 203–216. Kusano, S. 1912. The life history of cytology and a new Olpidium with special reference to the copulation of motile isogametes. Journal of The College of Agriculture, 4: 142–199. Li-Juan, L., Zhao hai, Y., Zhao-Jie, Z., Ming-shi, X. and Han-qing, Y. 1993. Faba Bean in China: A State of the Art Review. International Center for Agricultural Research in the Dry Areas, Aleppo, Syria, pp. 127–128. 255 Mussa, J., Dereje, G. and Gemechu, K. 2008. Procedures of Faba Bean Improvement through Hybridization. 48p. Technical Manual No. 21, Ethiopian Institute of Agricultural Research. Addis Ababa, Ethiopia. Ristaino, J.B. and Gumpertz, ML. 2000. New frontiers in the study of dispersal and spatial analysis of epidemics caused by species in the genus Phytophthora. Annual. Review. Phytopathology, 38:541–576. Rossi, V. and Caffi, T. 2012. The role of rain in dispersal of the primary inoculum of Plasmopara viticola. Phytopathology, 102:158–165. Sahile, S., Fininsa, C., Sakhuja, P.K. and Seid, A. 2008. Effect of mixed cropping and fungicides on chocolate spot (Botrytis fabae) of faba bean (Vicia faba) in Ethiopia. Crop Protection, 27: 275–282. SAS (Statistical Analysis System). 2014. SAS User’s Guide. SAS Institute. Cary, North Carolina, USA. Sillero, J.C., Villegas-Fernandez, A.M. Thomas, J., Rojas-Molina, M.M., Emeran, A.A., Fernandez-Aparicio, M. and Rubiales, D. 2010. Faba bean breeding for disease resistance. Field Crops Research, 115(3): 297–307. Stoddard, F.L., Nicholas, A.H., Rubiales, D., Thomas, J. and Villegas-Fernández, A.M. 2010. Integrated pest management in faba bean. Field Crops Research, 115: 308–318. Tadesse, N., Ahmed, S., Gorfu, D., Beshir, T., Fininsa, C., Abraham, A. 2008. Review of research on diseases of food legumes. In: Tadesse, A. (Ed.) Increasing Crop Production through Improved Plant Protection. Addis Ababa, Ethiopia: PPSE and EIAR, pp. 85– 132. Teklay, A., Tsegaye, B., Yemane, N. and Assefa, W. 2014. The prevalence and importance of faba bean diseases with special consideration to faba bean gall in Tigray, Ethiopia. Discourse Journal of Agriculture and Food Science, 2: 33–38. Teferi, T.A., Weldemichael, G.B., Wakeyo, G.K. and Mindaye TT. 2018. Fungicidal management of the newly emerging faba bean disease “gall” (Olpidium viciae Kusano) in Tigray, Ethiopia. Crop Protection, 107: 19–25. Terefe, H., Fininsa, C., Sahile, S., Dejene, M. and Tesfaye, K. 2015. Effect of integrated cultural practices on the epidemics of chocolate spot (Botryitis fabae) of faba bean (vicae fabae 256 L.) in Hararghe Highlands, Ethiopia. Global Journal of Pests, Diseases and Crop Protection 3(4):113–123. Van der Plank, J.E. 1963. Plant diseases: Epidemics and control. Academic Press, New York. 334 pp. Wheeler, B.J. 1969. An introduction to plant diseases. John Wiley and Sons Ltd., London. Wulita, W. 2015. Management of the newly emerged disease “qormid” on faba bean (Vicia faba L.) Using varieties and fungicides in north Shoa, central Ethiopia. MSc Thesis, Haramaya University, Ethiopia. Yan, J.M. 2012. Study on blister disease of broad bean caused by Olpidium Viciae Kusano. Doctoral Dissertation, Sichuan Agricultural University. Sichuan, China. You, M.P., Eshete, B.B., Kemal, S.A., Van Leur, J. and Barbetti, MJ. 2021. Physoderma, not Olpidium, is the true cause of faba bean gall disease of Vicia faba in Ethiopia. Plant Pathology, 70(5): 1180-1194. 257