GENETIC IMPROVEMENT OF ALKALINITY TOLERANCE IN RICE IN OFFICE DU NIGER IN MALI By OUMAROU GOITA (10293989) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF DOCTOR OF PHILOSOPHY DEGREE IN PLANT BREEDING WEST AFRICA CENTRE FOR CROP IMPROVEMENT SCHOOL OF AGRICULTURE COLLEGE OF AGRICULTURE AND CONSUMER SCIENCES UNIVERSITY OF GHANA, LEGON DECEMBER 2013 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I hereby declare that except for references to works of other researchers, which have been duly cited, this work is my original research and that neither part nor whole has been presented elsewhere for the award of a degree. .................................................. Oumarou GOITA Student ....................................................... REV. Prof. Frank. K. KUMAGA Supervisor ....................................................... Prof. Kwadwo OFORI Supervisor ..................................................... Prof. Vernon GRACEN Supervisor .................................................. Dr. Mamadou M. COULIBALY Supervisor ………………………………….. Dr. Kofi BIMPONG Supervisor University of Ghana http://ugspace.ug.edu.gh ii ABSTRACT Alkalinity stress is one of the most important abiotic stresses limiting rice production in irrigated systems. Alkalinity stress during seedling stage widely affects irrigated rice production of Office du Niger in Mali. The development of rice cultivars that tolerate alkalinity stress condition has been hindered by the lack of an elaborate breeding program. The objectives of this study were: (i) to evaluate and validate production constraints and varietal preferences of rice farmers in alkalinized zones of Office du Niger; (ii) to identify alkaline tolerant accessions and their association with SSR markers and, (iii) to characterize phenotypically F2 progenies at seedling stage in alkaline hydroponic solution. Participatory rural appraisal was conducted in three (3) highly alkalinized zones of Office du Niger. Two (2) villages per zone were involved in the study. Fertilizer cost, water management, inadequate agricultural equipments and declining soil fertility were identified as the main constraints of rice production. Overall, alkalinity was the most important abiotic stress affecting rice production. Crop rotation; use of organic matters; ploughing followed by flooding and use of tolerant varieties were the strategies developed by farmers to overcome alkalinity. Taste and swelling for home consumption; grain color and size for marketing were farmers preferred traits. Kogoni 91-1, BG 90-2 and Adny 11 were identified as farmer preferred varieties. Research priorities important to the farmers, included: alkalinity tolerance at seedling and reproductive stages combined; yield and taste and yield alone. Improvement of these characters in new varieties with alkalinity tolerance would enhance productivity with likely positive impact on small scale farmers’ food security, incomes and livelihoods. Farmers had high interest in participatory varietal selection and participatory plant breeding. University of Ghana http://ugspace.ug.edu.gh iii The morphological and physiological analysis relationships between different rice accessions revealed four accessions (Sahel 210, Damodar India (IRGC 17038), CSR 10 and CSR11) in the same group of the tolerant check (Pokkali). Genetic diversity and association of Simple Sequence Repeat (SSR) markers with alkalinity tolerance were detected in a set of 26 rice genotypes. Among SSR markers used for morpho- physiological characters, RM208 was significantly (P-value = 0.01) associated with the potassium content in the flag leaves at 37% correlation coefficient. Phenotypic and genotypic association studies identified the accessions Sahel 210, Damodar India (IRGC 17038), CSR 10 and CSR11 in the same clustering as Pokkali (salt tolerant check). A protocol based on an alkaline hydroponic solution was validated and adopted. The pH 9.0 was superior to separate alkaline tolerant and sensitive rice at early seedling stage. A total of 1,800 individual F2 genotypes were screened and 1,652 were phenotyped in alkaline hydroponic solution. The total number of surviving F2 individuals at twenty five (25) days after transplantation was 34 and varied from one cross combination to another. University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGEMENT I would like to express my sincere gratitude to Reverent Prof Frank K. KUMAGA, Prof. Kwadwo OFORI, Prof Vernon GRACEN, Prof Eric Danquah, Prof Samuel K. OFFEI and Dr. Mamadou M. COULIBALY, for accepting to supervise my thesis work, for their patience, guidance and invaluable contribution to this work. My deepest gratitude goes to Dr. Kofi BIMPONG for his patience and guidance throughout my experiment and laboratory work. Special thanks and acknowledgements go to AGRA (Alliance for a Green Revolution in Africa) which provided the study grant through West Africa Center for Crop Improvement (WACCI). I am grateful to WACCI for identifying me as potential doctoral candidate and offering me the much coveted WACCI scholarship. I equally owe the gratitude of all Professors and lecturers from WACCI program, University of Ghana/Legon, who imparted interdisciplinary knowledge during the study. I am also greatly indebted to International Rice Research Institute, Philippine, for providing me some tolerant rice accessions and Africa Rice, Saint-Louis (Senegal); for also providing me tolerant rice accessions, chemicals and laboratory facilities. I am grateful to Institut d’Economie Rurale (IER) of Mali for experimental field, breeding facilities and relationship with Africa Rice for some experiment and laboratory works. I would also like to express my special thanks to Noua COULIBALY and Seydou SIDIBE for their assistance in my breeding and field work. I am happy to congratulate the three direction zones of Office du Niger where I conducted my Participatory Rural Appraisal activities. I am happy to acknowledge WACCI workers for their assistance and moral support during the entire study period. I salute the university of Accra- Legon as the umbrella institution for providing right atmosphere. I thank Dr Aboubacar Touré, Dr Issoufou Kapran, Dr Mamadou Kabirou N’Diaye, Dr Karim TRAORE and Dr Babouccar Manneh and their families, my colleagues and all workers of IER Mali for their supports. University of Ghana http://ugspace.ug.edu.gh v DEDICATION This thesis is dedicated to my wife Mrs COUMBA SAGARA and Kids SOUMAILA, SOUTOURA and MOUSSA. Your love and support made it easier for me to persevere to the end. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS DECLARATION ERREUR ! SIGNET NON DEFINI. ABSTRACT II ACKNOWLEDGEMENT IV DEDICATION V TABLE OF CONTENTS VI LIST OF FIGURES X LIST OF TABLES XI LIST OF ABREVIATIONS XII CHAPTER I 1 GENERAL INTRODUCTION 1 CHAPTER II 5 2. LITERATURE REVIEW 5 2.1 Taxonomy and origin of rice 5 2.2 Climatic conditions for rice production (Production environment) 5 2.3 Rice ecosystem 7 2.4 Role of farmer in development of new technology 7 2.5 Alkali soil characterization and effect on rice yields 8 2.6 Sensitivity of Rice to salt stress 10 2.7 Breeding for alkalinity stress tolerance 12 2.8 Molecular markers for alkalinity tolerance in rice 15 CHAPTER III 18 3. PRODUCTION CONSTRAINTS AND VARIETAL PREFERENCES OF RICE FARMERS IN THREE HIGH ALKALINITY ZONES OF OFFICE DU NIGER OF MALI 18 3.1 Introduction 18 University of Ghana http://ugspace.ug.edu.gh vii 3.2 Research Methodology 19 3.2.1 Site selection 20 3.2.2 Selection of farmers 20 3.2.3 Data collection 20 3.2.4 Data Analysis 21 3.3 Results 21 3.3.1 Rice production constraints 21 3.3.2 Importance of soil alkalinity in Office du Niger 25 3.3.3 Farmers’ varietal preference and selection criteria 26 3.3.4 Farmer priorities for research needs 30 3.4 Discussion 30 3.5 Conclusion 36 CHAPTER IV 37 4. IDENTIFICATION OF ALKALINE TOLERANT RICE ACCESSIONS AND THEIR ASSOCIATION WITH SSR MARKERS THROUGH GENETIC DIVERSITY 37 4.1 Introduction 37 4.2 Materials and methods 39 4.2.1 Plant materials 39 4.2.1.1. Identification of salt tolerant accessions under alkaline soil condition 39 4.2.1.2. Determination of genetic diversity 39 4.2.1.3. Characterization of phenotypic and genetic associations 40 4.2.2. Methodology 41 4.2.2.1 Identification of salt tolerant rice accessions under alkaline soil condition 41 4.2.2.2. Determination of genetic diversity 42 4.2.2.3. Phenotypic and genotypic association study for alkaline tolerance rice identification 43 4.2.3. Data collection 43 University of Ghana http://ugspace.ug.edu.gh viii 4.2.3.1. Identification of salt tolerant rice accessions under alkaline soil condition 43 4.2.3.2. Determination of genetic diversity 44 4.2.3.3. Phenotypic and genotypic association study for alkaline tolerance rice identification 45 4.2.4. Data Analysis 45 4.2.4.1. Identification of salt tolerant rice accessions under alkaline soil condition 45 4.2.4.2. Determination of genetic diversity 45 4.2.4.3. Phenotypic and genotypic association study for alkaline tolerance rice identification 45 4.3 Results 46 4.3.1 Identification of salt tolerant rice accessions under alkaline soil condition 46 4.3.1.1 Soil characteristics after flowering 46 4.3.1.2 Rice accessions flag leaves analyses 46 4.3.1.3 Percent germination, vigor, percent spikelet fertility, and grain weight 49 4.3.1.4 Association among traits across rice genotypes 52 4.3.1.5 Morphological and physiological analysis relationships between the rice accessions 52 4.3.2. Determination of genetic diversity 54 4.3.2.1 Number of alleles 54 4.3.2.2 Allele size range 54 4.3.2.3 Frequency of alleles 55 4.3.2.4 Polymorphism Importance Content (PIC) values 55 4.3.2.5 Euclidian mean cluster tree dendrogram 55 4.3.3. Phenotypic and genotypic association study for alkaline tolerance identification 57 4.4 Discussion 60 4.5 Conclusion 66 CHAPTER V 68 5. EVALUATION OF F2 RICE POPULATIONS FOR ALKALINITY TOLERANCE AT SEEDLING STAGE 68 University of Ghana http://ugspace.ug.edu.gh ix 5.1 Introduction 68 5.2 Materials and Methods 69 5.2.1 Parental lines 69 5.2.2. Methodology 71 5.2.2.1. Production of F1 and F2 seeds 71 5.2.2.2 Protocol optimization for alkalinity tolerance in rice 72 5.2.2.2.1 Plant material and salt treatments 72 5.2.2.2.2 Methodology, data collection and statistical analysis 75 5.2.2.3 Screening for alkalinity tolerance in F2 rice populations at seedling stage 77 5.2.2.3.1 Plant material and salt treatments 77 5.2.2.3.2 Methodology, data collection and statistical analysis 77 5.3 Results 78 5.3.1 Production of F1 and F2 seeds 78 5.3.2 Optimization for rapid alkalinity screening of Oryza sativa L. at seedling stage in Modified Yoshida’s solution 79 5.3.2.1 SES score, shoot height and root length at three levels of pH 79 5.3.2.2 Comparison of the effects of different pH levels on SES score 79 5.3.2.3 Correlation among traits across FL478 and IR29 82 5.3.3 Screening for alkalinity tolerance in F2 rice populations at seedlings stage 83 5.3.3.1 Alkalinity stress symptoms on parents and checks at seedling stage 83 5.3.3.2 Frequency of phenotyped F2 rice populations 85 5.3.3.3 Alkalinity tolerance in F2 populations means at the seedling stage using SES 86 5.3.3.4 Alkalinity stress effects on shoot and root length of F2 populations 87 5.3.3.5 Correlation among traits within F2 progenies per population 89 5.3.3.6 Survival rate twenty five (25) days after transplantation (DAT) 90 5.4 Discussion 91 University of Ghana http://ugspace.ug.edu.gh x 5.5 Conclusion 96 CHAPTER VI 97 GENERAL DISCUSSION 97 CONCLUSIONS 100 RECOMMENDATIONS 100 REFERENCES 101 APPENDICES 124 Appendix 3.1: Questionnaire Design for Individual Farmers 124 Appendix 5.1: F1 Generation to F2 in non alkaline field at Kogoni Station (Mali) in May 2012. 128 Appendix 5.2: Seedling float for screening at the seedling stage. Source: Gregorio (1997) 128 Appendix 5.3: Agronomic characters of survived F2 rice plants and their parents 129 LIST OF FIGURES Figure 4.1: Dendrogram of 26 rice accessions based on seedling vigor (SES), %spikelet fertility and grain weight (g) derived from UPGMA cluster analysis. 53 Figure 4.2: Cluster analysis of the 26 rice genotypes using distant coefficient based on 10 SSR markers. 56 Figure 4.3: Dendrogram of genotypic relationship in rice accessions based on RM208, RM537, RM572, RM142, -and RM493 markers, derived from UPGMA cluster analysis 59 Figure 5.1: variation of temperature (min and max) and relative humidity (%) between 01/08/12 and 01/06/2013 in the screenhouse of AfricaRice Saint Louis (Senegal). 76 Figure 5.2: Plant injury symptoms of genotypes under alkaline and non-alkaline nutrient solution. 83 Figure 5.3: Alkalinity stress injuries on selected parents used in different cross combinations and checks (tolerant and sensible) 15 Days After Alkaline stress Application (DAASA). A: SES score, B: Shoot height (cm) and C: Root length (cm) 85 Figure 5.4: Frequency of F2 populations of different cross combinations to alkalinity stress in the greenhouse of AfriaRice at Saint Louis (Senegal) in 2012 87 Figure 5.5: Tolerant and non tolerant F2 population means of different cross combination to alkalinity stress in the greenhouse of AfriaRice at Saint Louis (Senegal) in 2012. A: alkalinity stress effect on Shoot height. B: alkalinity stress effect on root length. 88 University of Ghana http://ugspace.ug.edu.gh xi LIST OF TABLES Table 2.1: Classification of salt-affected soils 9 Table 3.1: Seed source of rice farmers in three zones of Office du Niger (%) in Mali in 2011 23 Table 3.2: Rice production constraints and their relative importance (%) ranked in Office du Niger in 2011 23 Table 3.3: Causes of alkalinity affecting rice production in (%) in 2011 in Office du Niger 24 Table 3.4: Alkalinity types in rice production zones of Office du Niger in (%) in 2011 24 Table 3.5: Impact of alkalinity types on rice production in Office du Niger in 2011 25 Table 3.6: Farmers’ strategies for alkalinity control across Office du Niger villages in 2011 25 Table 3.7: Proportion of farmers (%) perception growing tolerant varieties in alkaline soils in 2011 26 Table 3.8: Percentage of farmers growing different rice varieties in Office du Niger in 2011 27 Table 3.9: Factors affecting farmer preferred rice varieties for production in Office du Niger 28 Table 3.10: Criteria for selecting rice for home consumption in Office du Niger in 2011 29 Table 3.11: Attributes of grain considered by farmers for selection for the market in Office du Niger in 2011 29 Table 3.12: Farmers’ preference for rice traits in Office du Niger in 2011 30 Table 4.1: Rice accessions for phenotypic study 39 Table 4.2: Rice genotypes for allelic diversity study 40 Table 4.3: Rice accessions for phenotypic and genotypic association study 40 Table 4.4: Selected markers used for genetic diversity for alkalinity tolerance 42 Table 4.5: Standard Evaluation System score (SES) 44 Table 4.6: The effect of rice accessions on pH, Na + and K + in alkaline soil of Office du Niger (Mali) in 2011 47 Table 4.7: Percent of Na + , K + uptake and Na + / K + ratio in the flag leaves at flowering stage in 2011 48 Table 4.8: Mean squares of 26 accessions and varieties’ for plant growth and yield characteristics 49 Table 4.9: percent germination, vigor (SES), percent spikelet fertility and grain weight 51 Table 4.10: Correlation coefficients among traits under alkaline soil of Office du Niger Mali 52 University of Ghana http://ugspace.ug.edu.gh xii Table 4.11: Allelic diversity among 28 rice genotypes for 10 SSR markers 54 Table 4.12: Marker effects on rice accessions morpho-physiological characters 58 Table 5.1: Pedigree characterization of rice genotypes tolerant and sensitive to salt used in the crosses for progeny salt tolerance in 2011 70 Table 5.2: Parental accessions and varieties used in different cross combinations 71 Table 5.3: Modified Singh et al. 2002 culture solution 75 Table 5.4: Number of F1 and weight of F2 seeds produced per cross 78 Table 5.5: Mean squares of treatments on salt injury (SES score), shoot and root length 79 Table 5.6a: pH effects on SES score 80 Table 5.6b: pH effects on shoot height (cm) 81 Table 5.6c: pH effects on root length (cm) 81 Table 5.7: Correlation among traits studied in rice entries (FL478 and IR29) 82 Table 5.8: Tolerant F2 populations from different cross-combinations 15 DAASA 86 Table 5.9: Correlation between SES score and shoot height growth and root length 90 Table 5.10: Survival of F2 progenies from different cross-combinations after 25 DAT 91 LIST OF ABREVIATIONS ANOVA: Analysis of variance bp.: base pair Chr.: Chromosome cM: centi Morgan CTAB: Cetyl-Triméthyl-Amonium Bromide CV: Coefficient of Variation DAASA: Days After Alkaline Stress Application University of Ghana http://ugspace.ug.edu.gh xiii DAT: Day After Transplantation DNA: Deoxyribo-Nucleic Acid ECe: Electrical conductivity (mmhos/cm) ESP: Exchange Sodium Percentage FAO: Food and Agriculture Organisation of the United Nations FAOSTAT: Food and Agriculture Organisation Statistical Database FGD: Focus Group Discussion GGT-2: Graphical Geno-Typer-2 IER: Institute Economy Rural IFDC: International Fertilizer Development Center IRRI: International Rice Research Institute LSD: Least Significance Difference MAS: Marker Assisted Selection NARI: National Agricultural Research Institute PAGE: Poly-Acrylamide Gel Electrophoresis PCR: Polymerase Chain Reaction pH: potential Hydrogen PIC: Polymorphism Information Content PPB: Participatory Plant Breeding PSI: Irrigated System Pole University of Ghana http://ugspace.ug.edu.gh xiv PRA: Participatory Rural Appraisal PVS: Participatory Varietal Selection QTL: Quantitative Trait Loci RYMV: Rice Yellow Mottle Virus SAR: Sodium Absorption Ratio SAS: Statistical Analysis System SES: Standard Evaluation System SSR: Simple Sequence Repeat UPGMA: Unweighted Pair Group Method with Arithmetic Mean USAID: United States Agency for International Development USDA: United States Department of Agriculture UVP: Ultra-Violet Products WARDA: West Africa Rice Development Association University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER I GENERAL INTRODUCTION Rice (Oryza sativa L.) is the staple food for half of the world’s population and the second most important crop in the world after wheat, with more than 98% currently grown in Asia (USDA, 2012). Globally it is cultivated on about 158.83 million ha and 464.87 million tons of paddy are harvested annually (USDA, 2012). Africa produces an average of 23.4 million tons of rice per year (2005-2011) on 9.7 million ha respectively equivalent to 3.5% and 6.1% of the world’s total production and rice area (FAO, 2012). Rice is the most important food crop in Mali, and is grown in all regions of the country. It is increasingly favored by consumers primarily in urban areas but also in the rural areas where it is produced. In 2011, 13.34% of the 6,221,422 ha of West African rice area were harvested in Mali (FAO, 2012). Rice production in Mali requires large amounts of water due to the high evaporative demand of the hot and dry climate. While the irrigated water is not highly mineralized, excessive concentration through evaporation can become dangerous as these waters often possess a positive calcite residual alkalinity (Bertrand et al., 1993; Vallès et al., 1992). The inventory of damaged soils in the Office of Niger reported that 62%, 54%, 32% and 27% of soils are affected by salinity and alkalinity, in Molodo, Niono, N’Debougou and Macina` respectively (N’Diaye, 1998). In Mali and particularly in Office du Niger, sodic and alkali soils are reclaimed for agricultural use particularly for rice production. Reclamation commonly starts with soil tilling followed by harrowing and treatment with gypsum and organic manures and then flooding of the fields in which the rice crop is planted. University of Ghana http://ugspace.ug.edu.gh 2 The use of inputs such as fertilizers, pesticides and weed control is seen by farmers as being risky (Cassman, 2004; Sophia, 2010). Therefore the adoption of new agronomic technologies has been limited. The most economic solution to improve crop productivity in unfavorable environments is through breeding for tolerance to the existing stress (Efisue et al., 2008; Sophia, 2010). The effects of salinity and alkalinity on the growth of rice have been found to be related to the stage of plant development, salt concentration and type, duration of rice plant exposure to salt, soil pH, water regime, temperature, humidity and solar radiation (Deepa et al., 2011). Tolerant genotypes of rice have a high germination percentage, root length, shoot height, less salt injury, amylase and dehydrogenase activity with less accumulation of anthocyanin in their roots (Janaguiraman et al., 2003). Deepa et al. (2006) have ranked salt tolerant rice lines based on germination and seedling growth under salt stress conditions. Breeding rice varieties with in- built salt tolerance is the promising, less resource consuming, economically viable and socially acceptable approach. In Mali, increased rice production can be achieved by in non alkalinized soil environments using improved cultivars and improved agronomic practices. However, such an approach may exclude many small and poor farmers in marginal environments who represent the majority of the farmers in the country and, breeders do not have a clear understanding of these farmers’ requirements and preferences (Banziger and Cooper, 2001; Cleveland and Soleri, 2002). Most of the rice varieties produced in Mali has been cultivated by farmers’ for traits such as grain size and color, taste, cooking qualities and high yielding ability. However all of these cultivated varieties are sensitive to alkalinity stress. The transfer of these preferred traits into the tolerant accessions would be the breeding program objective to develop new tolerant alkaline genotypes accepted by farmers. University of Ghana http://ugspace.ug.edu.gh 3 Screening and developing rice with increased salt tolerance and better ability to grow at seedling stage in alkaline soil will be necessary to improve rice production in Mali. The use of morphological and physiological parameters in breeding could enable rapid selection of genotypes with tolerance. A combination of selection criteria could give a good indication of the salt response of crop plants morphologically (Singh, 2002). Factors like germination, seedling stage response, vegetative vigor (tillering ability and plant height), spikelet fertility; stress rating (growth and tillering), panicle exsertion and pH of environment should be considered. To obtain accurate heritability estimates in salt stress tolerance breeding, precise phenotyping and genotyping is required. Molecular marker technology provides a powerful tool in the assessment of genetic relationships within and among species, in which differences among accessions can be revealed at the DNA level (Chakravarthi and Naravaneni, 2006). The applications of SSR markers in rice research include studies on genetic diversity of stress tolerance (Islam et al., 2012), estimation of genetic diversity in rice genotypes using SSR markers and morphological characters (Seetharam et al., 2008) and genetic diversity analysis of traditional and improved Indonesian rice germplasms (Thomson et al., 2003). The molecular characterization information as well as genetic diversity analysis could be helpful to Malian breeders for planning programs for improving several traits in rice. Thus, the seedling and reproductive stages are the most vulnerable to salt stress and plants experience greater damage from salt stress mostly in these specific stages. In this study, the seedling and reproductive stages were used to select for tolerant genotypes in alkalinized soils. The F2 individual rice plants from different cross combinations were screened at seedling stage in hydroponic alkaline solution for identification of F2 survival plants and their molecular characterization. University of Ghana http://ugspace.ug.edu.gh 4 The objectives of this study were to: - evaluate and validate production constraints and varietal preferences of rice farmers in alkalinized zones of Office du Niger; - identify alkaline tolerant accessions and establish their genetic association using SSR markers; - characterize F2 progenies phenotypically at seedling stage in alkaline hydroponic solution. University of Ghana http://ugspace.ug.edu.gh 5 CHAPTER II 2. LITERATURE REVIEW 2.1 Taxonomy and origin of rice The genus Oryza belongs to the tribe Oryzeae of the family Poaceae (http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi). There are 12 genera within the Oryzeae tribe (Vaughan, 1994). The genus Oryza contains nearly 22 species of which 20 are wild species and two, Oryza sativa and Oryza glaberrima, are cultivated (Vaughan, 1994). Oryza sativa is the most widely grown of the two cultivated species. The origin of rice can be grouped into two classes: polyphyletic origin and monophyletic origin. Polyphyletic origin refers to cultivated species that originated from several species. Monophyletic refers to cultivated species that originated from common specie. Some believe that Asian rice, Oryza sativa, and African rice, Oryza glaberrima, have evolved independently in their respective regions from several species. But others believe that both Asian rice and African rice arose from a common parent (O. perennis). The second option is the most accepted one because both Asian rice and African rice are similar except in glumes pubescence, ligules size and colour of pericarp which is red in African rice. The progenitors of Oryza sativa are considered to be the Asian AA genome diploid species (Randhawa et al., 2006). 2.2 Climatic conditions for rice production (Production environment) Most of the rice growing areas in the world are located in the tropics (Yoshida, 1981). These areas are between the tropics of Cancer (23.50N) and the tropic of Capricorn (23.50S) and extend to as far as 490N and 350S of the latitudes. Rice also grows from sea level to an altitude of 2,500m or more (Khush, 1997). Worldwide, there are about 150 million hectares of rice, which provide around 550–600 million tons of rough rice annually (Maclean et al., 2002). Rice is University of Ghana http://ugspace.ug.edu.gh 6 unique among the major food crops in its ability to grow in a wide range of hydrological situations, soil types, and climates. Although, rice is primarily a tropical and subtropical crop, the best grain yields have been obtained in the temperate region. This is attributed to lower temperature during ripening, which gives more time for grain filling, long day lengths and high levels of solar energy during the ripening period (Yoshida, 1981). These phenomena are also conducive to lower grain yield obtained during the wet season versus the dry season. However, due to the prevalence of biotic and abiotic factors during dry seasons, lower grain yield is obtained in most cases as compared to the wet season. Sensitivity to high levels of salt is a major problem for rice but degree of salt tolerance is known. Temperature and relative humidity are the most important climatic factors affecting salt tolerance (Singh et al., 2009). Temperature regime greatly influences growth duration and growth pattern of almost all crop plants under normal soil conditions. Crop plants have critical temperatures for different stages. Temperatures higher than the critical greatly affect plant growth, under salt stress. This is because high temperature enhances depletion of internal water status of the plant through high transpiration. Plant transpiration increases to maintain the plant’s internal temperature and, in the process, more salt enters into plant tissues at a rapid pace (Deepa et al., 2011). Over-accumulation of salts ultimately leads to severe injury or death of the plant (Singh et al., 2009). Relative humidity also plays an important role under salt stress. Dry weather and very low humidity increase the evapotranspiration rate of plants while the reverse is true with high humidity. Low humidity in a stress environment is detrimental to plant growth because of higher ion uptake (Singh et al., 2005). University of Ghana http://ugspace.ug.edu.gh 7 2.3 Rice ecosystem Rice ecologies are classified according to source of water supply into two broad terms as rain-fed and irrigated. In West Africa, Buddenhagen (1978) classified rice ecologies into four main types and eight sub-types based on climate, soil conditions, water regime and management practices. WARDA (1997) current AfricaRice defined six ecological areas favorable for rice production in West Africa. These ecologies are: rainfed upland, rainfed lowland, lowland irrigated, Sahelian irrigated, mangrove swamp and deep water/floating. This current study will focus on the Sahelian irrigated rice ecology. This is because it is important rice production ecology, particularly in Mali where it accounts for more than 50% (Mendez de Villar et al., 2011) and also where alkalinity stress is more pronounced and has limited rice production without soil reclamations. 2.4 Role of farmer in development of new technology Participatory rural appraisal (PRA) is one of the families of approaches and methods which enable communities to share; develop and analyze their own knowledge of life and conditions (Chambers, 1997). Essentially, participatory appraisal is employed for its adaptability and some of the non-traditional ways in which needs assessment is undertaken (Tock, 2001). It involves the researcher identifying community priorities and can include varietal development with rural people defining their ideotype. There is a wealth of knowledge and skills within a community that essentially goes unused during traditional research (Tock, 2001). The participatory approach is a rapid and cost effective method of identifying farmers-preferred rice cultivars in India and it also revealed a number of important characters that would have not been identified in breeders’ experiments (Joshi and Witcombe, 1995). To study socio-economic factors influencing the adoption of sawah rice production technology in Nigeria PRA was practical and used (Fashola et University of Ghana http://ugspace.ug.edu.gh 8 al., 2007). In Namibia farmers were engaged in pearl-millet (Pennisetum glaucum L.) cultivar selection (Monyo et al., 2000). Efisue et al. (2008) use also the PRA to identify early maturing and tall varieties among the most important traits by farmers in the rainfed production ecologies in Sikasso region of Mali. Sophia (2010) identified through PRA rice farming in Tanzania more active participation of women than men and noted that woman’s involvement and utilization of their indigenous knowledge and understanding their preferences for rice varieties may increase adoption of new varieties. Therefore PRA is considered as a tool to understand the farmers’ important information on constraint, needs and preferences and considered as vital means to collect information on farmers’ knowledge on the soil affected stress. Understand farmers’ characteristics and their involvement in the technology development process is central for success in the adoption of farm technologies (Fashola et al., 2007). The adoption or not of farmers decision will depend on their attitude to the innovation, farming experience, household size and visits by extension agents (Azuma, 2004). Neglecting farmers need may lead to low adoption rate of developed cultivars. According to Efisue et al. (2008), the low adoption rate of rice varieties by the farmers could be due to the fact that scientists are working in isolation from farmers and other stakeholders and therefore, might not identify the unique requirements for the subsistence farmers in particular niches of the diverse environments. 2.5 Alkali soil characterization and effect on rice yields Salt-affected soils are considered as degraded soils that contain an excess of water-soluble salts (saline soils), exchangeable sodium (sodic soils) or both an excess of salts and exchangeable sodium (saline-sodic soils or alkaline) (Siyal et al., 2002). University of Ghana http://ugspace.ug.edu.gh 9 Soil salinization is the accumulation of the soluble salts of sodium, calcium, and magnesium in the soil root zone (Siyal et al., 2002). The effect of these salts can degrade soils to the extent that crop production such as rice is severely decreased. High levels of salt limit plant growth by increasing osmotic pressure of the soil solution, which reduces the plant’s capacity to withdraw water from the soil (Siyal et al., 2002). Depending on salt concentration (Electrical Conductivity), Exchangeable Sodium Percentage (ESP) and pH of the soil solution, salt affected soil is divided into three categories, saline, alkali and saline-alkali soils (Table 2.1.). Table 2.1: Classification of salt-affected soils Classification Electrical conductivity mmhos/cm (ECe) Exchangeable Sodium Percentage (ESP%) pH Saline soils > 4.0 < 15 < 7.05 alkali soils < 4.0 > 15 > 8.5 Saline – alkali soils < 4.0 > 15 > 8.5 Source: (Siyal et al., 2002) In alkali soils, a sodium ion disperses the minerals colloids, which form the tight soil structure. This structure slows the infiltration and percolation of water (Siyal et al., 2002). The sodium excess effect in alkali soil does not allow soil particles to attach one another resulting in dispersion and non friability of soil. The low permeability causes by alkali soils, makes difficult water movement due to dispersed clay and humus. The alkali soils can be identified from the black color of the soil surface (Siyal et al., 2002). Salt stress disrupts the nutrient balance in the plant. In rice, a deficiency of one nutrient causes increased uptake of a nutrient of the same valence (Singh, 2001). Rice production in alkali soil is complex (Condom, 2000). Low use efficiency of nitrogen and low potential yielding of the varieties were due to the impact of alkalinity on rice production (Wopereis et al., 1999). The volatilization process of ammonia is the most noticeable effect which in reproductive stage University of Ghana http://ugspace.ug.edu.gh 10 affects spikelet fertility (Dicko, 2005). Under salt stressed conditions, K + deficiency increases and Na + uptake increases (Dobermann and Fairhurst, 2000) resulting in plant injury, sterility and yield reductions. Excess Na + may compete with K + in membrane transport and when accumulated in the cytoplasm, it inhibits many enzymes. Devitt et al. (1981) found that Na + /K + imbalance adversely affects grain yield. Due to high levels of these salts in soils, the genetic potential of the plant is not realized to its maximal extent to target grain yield (Tollenaar and Lee, 2002), hence reduction in yield. The majority of farmers in salt affected soils areas have low level of understanding of degraded soils, symptoms and their impacts in plant growth (Mnkeni, 1996). Dicko (2005) revealed further in alkalinity stress environments, the difficulty to differentiate symptoms due to salt affected soils and nutrient stresses. Thus, there is the need to understand farmers’ perception and behavior under salt affected soils community for proper planning on mitigation measures. 2.6 Sensitivity of Rice to salt stress Alkaline soils are defined as soils which contain sufficient exchangeable sodium to impair crop growth. Excess sodium results in poor soil physical properties and nutrient imbalances. The high pH causes low availability of essential nutrients such as P, Ca, Fe, and Zn and rice crop sensitivity start from or above the pH 7.5, which hinders or affects the normal plant growth. The alkalinity constraint is measured in terms of the exchangeable sodium percentage (ESP) or the sodium absorption ratio (SAR) and pH of saturated soil paste extract (Singh et al., 2011). Therefore, sodic and alkaline soils are those with ESP equal to or more than 15 % and a pH above 8.5 (Abrol, 1986; Szabolcs, 1994; Waisel, 1972). Alkalinity causes a drastic decrease in potassium content of salt-sensitive rice varieties; the salt susceptible rice cultivars have the low ratio of K + /Na + in leaves and high grain yield reduction under alkalinity (Asch et al., 2000). Salt- University of Ghana http://ugspace.ug.edu.gh 11 tolerant cultivars of rice accumulate less Na + , Cl - , Zn and proline and more K + at root and shoot levels than salt-sensitive varieties. Phosphorus transport from root to shoot is inhibited in the salt-sensitive cultivars. Accumulation of Na + , and Cl - , and decrease in K + content at the shoot level are restricted to the oldest leaves in salt-tolerant genotypes while proline is accumulated in the youngest leaves in all cultivars (Asch et al., 2000). The mechanism of salt tolerance fall into three categories: tolerance to osmotic stress, sodium exclusion from leaf blades and tolerance of tissue to accumulated sodium or chloride (Munns and Tester, 2008). The osmotic stress reduces cell expansion in root tips and young leaves and causes stomatal closure. A reduced response to the osmotic stress would result in greater leaf growth and stomatal conductance. However, the increased leaf area would benefit only plants that have sufficient soil water. Sodium exclusion by roots ensures that sodium does not accumulate to toxic concentrations within leaves. A failure in sodium exclusion manifests its toxic effect after days or weeks, depending on the species, and causes premature death of older leaves. Tolerance requires compartmentalization of sodium and chloride at cellular and intracellular level to avoid toxic concentrations within the cytoplasm, especially in mesophyll cells in the leaf. The toxicity occurs with the time, after leaf sodium increases to high concentrations in older leaves. Rice is differentially affected by alkalinity at different stages (Munns, 2002). The effects of alkalinity on the growth of rice were found to be related to the stage of plant development, salt concentration, type of salt, duration of exposure to salt, soil pH, water regime, temperature, humidity and solar radiation (Akbar, 1986). Rice is relatively tolerant to salinity and alkalinity during germination stage. The germination though affected by salinity can be appreciably higher germination percentage, root length, Shoot height, vigor index and amylase and dehydrogenase activity with less accumulation of anthocyanin in their roots (Janaguiraman et al., 2003). Salt University of Ghana http://ugspace.ug.edu.gh 12 tolerant rice lines based on germination and seedling growth has been ranked under salt stress conditions (Deepa et al., 2006). Rice is very sensitive during the early seedling stage, gains tolerance during the active tillering stage, but becomes sensitive during panicle initiation, anthesis, and fertilization, and then becomes relatively more tolerant at maturity (Rao et al., 2008). 2.7 Breeding for alkalinity stress tolerance A successful breeding program should have a wide spectrum of genetic variability for the desired trait (good donors), and a robust, repeatable, and reliable screening technique (Singh et al., 2009). A phenotype with tolerance or sensitivity is the overall manifestation of the sum of the different tolerance mechanisms operating in the genotype. Rice is an important crop and experimental evidence suggests that most rice varieties exhibit salt sensitivity, but that varietal differences tend to manifest themselves at different salt concentrations (Yeo and Flowers, 1984). A wide range of variation of sensitivity exists among different rice cultivars. There are a number of salt tolerant rice varieties available for different levels of soil stresses viz., CSR 10 for very high stress; CSR 13, CSR 27, Narendra Usar 2 and Narendra Usar 3 for moderate to high stress and a basmati CSR 30 rice for moderate stress. These are being utilized for land reclamation programs in Uttar Pradesh in India (Deepa et al., 2011). The level of salt tolerance found so far in rice is inadequate, however, the transfer of this trait to combine desirable characters has had only limited success because the knowledge of genetic basis of salt tolerance is limited (Ansari et al., 2003). The main expectation from plant physiologists is the identification of “salt markers” as defined by (Ansari et al., 2003; Epstein et al., 1979). Such markers include characters that are associated with salt stress, easily identifiable, and can be used for screening salt-tolerant plants in large breeding populations. University of Ghana http://ugspace.ug.edu.gh 13 The development of salt-tolerant varieties has been proposed as a means of expanding agriculture into the regions affected by salinity (Epstein, 1980). Breeding rice varieties with in- built salt tolerance is the most promising, less resource consuming, economically viable and socially acceptable approach. There is a need to develop rapid screening methods to evaluate tolerant plants and subsequently combine desirable characters, and identify, at an early stage of development, some markers that can play an important role in identifying superior performance at later growth stages. Some characters which manifest themselves at an early stage but are directly related to later performance of plants under salt conditions have been identified. These characters are leaf mortality, seedling vigor, leaf chlorophyll, Na + uptake, Na + /K + ratio and distribution of Na + among leaves (Ansari et al., 2003). Leaf senescence is a well-known symptom of stress and it indicates leaf tissue death resulting from the excessive accumulation of ions (Ansari et al., 2003). This can serve as an indicator of salt tolerance even at a very early stage of plant development. Salt stress through osmotic effects reduces the ability of plants to take up water and salts rise to a toxic level in older transpiring leaves causing premature senescence, and reduce the photosynthetic leaf area of a plant to a level that cannot sustain growth (Munns, 2002). Rice cultivars differ considerably in their growth rate with the most vigorous lines being the traditional varieties. Dwarfing genes were incorporated into most of the modern varieties and breeding lines to increase harvest index and reduce lodging (IRRI, 2006). Naturally occurring salt resistant varieties invariably belong to these modern varieties. Vigorous growth has a dilution effect (IRRI, 2006). Differences in vigor among rice cultivars accounted for much of the variation in survival under high salt levels. The vigorous growth is a selection criteria for University of Ghana http://ugspace.ug.edu.gh 14 alkalinity tolerance because of the good yield potential associated with short plants that do not lodge. Thus, early seedling vigor is desirable due to high sensitivity during this stage coupled with high salt levels encountered at the beginning of the season (Sexcion et al., 2009). Tolerance of one rice variety was found to be associated with its ability to restrict potentially toxic ion uptake like Na + and associated with preferential uptake of the balancing ion like K + (IRRI, 2006). It is an adaptation for the survival of plants so that the vital metabolic activities are not affected. There are larger differences in ion (Na + and K + ) uptake between the species in comparison to the genotypic differences within a crop species. These are the most studied parameters for salt tolerance in crop plants. The tolerant varieties maintained lower Na + concentrations and maintaining higher K + concentrations under high salt concentrations (Sexcion et al., 2009). One factor may be the overall control mechanism (before flowering) of sodium uptake through root properties and its subsequent distribution in different vegetative and floral parts especially in leaves where it causes leaf mortality thereby reducing transportation of total assimilates to the growing region (Munns, 2002). Tolerant genotypes like CSR1 showed regulation of distribution and accumulation of Na + taken up by the plants i.e. the delicate and vital organs like young and photosynthetically active leaves as well as the reproductive organs like panicles are kept relatively free of Na + , besides having an assured supply of K + , even under higher salt concentrations (IRRI, 2006). Thus the differences in the distribution of ions in individual plant organs and with the age and position of these organs on the plant are more important indicators of its tolerance potential than the difference in the average salt content of the plant. The uptake of both Na + and K + are entirely independent, but lower Na + /K + ratio is considered a as desirable trait as it maintains the ion balance. Na + is transported to shoots usually through University of Ghana http://ugspace.ug.edu.gh 15 apoplastic pathways (passive transport) while K + transport takes place through symplastic pathways (active transport). Younger leaves have relatively lower Na + than K + as compared to the older leaves, which in turn results in higher Na + /K + ratio in the older leaves. Thus Na + /K + ratio increases precipitously with salt concentration and leaf age (Munns, 2002). Alkalinity stress causes a drastic decrease in potassium content of salt-sensitive rice varieties; the salt susceptible rice cultivars have low ratio of K + /Na + in leaves and high grain yield reduction under alkalinity stress (Asch et al., 2000). A tolerant variety keeps its leaves relatively free of the toxic ions besides having assured K + supply. This factor, along with the higher number of leaves and higher leaf area, probably contributes to success under high salt concentration. This parameter is not a universal phenomenon hence should be used with caution as selection criteria (IRRI, 2006). 2.8 Molecular markers for alkalinity tolerance in rice A good deal of progress has been made in agriculture using conventional methods of breeding crop varieties despite non-availability of knowledge about the physiological and biochemical mechanisms. However, in some situations, genetic advance through breeding plants has been slow due to complex and ambiguous natures of the traits, like salts stress tolerance. A more comprehensive understanding of physiological and biological mechanisms would contribute to efficient breeding of crops suitable for stress conditions. Alkalinity tolerance in rice (low Na + /K + ratio in shoot) is governed by both additive and dominance gene effects (Ramayya et al., 2009). However, it is difficult to identify these genes, known as quantitative trait loci (QTLs), by using morphological and physiological markers because each of these genes singly has a relatively small effect on the phenotype (Ramayya et al., 2009). Molecular analysis of the genome at the DNA level can provide a greater advantage because DNA molecules are the same in all of the living cells of a plant, regardless of physiological or University of Ghana http://ugspace.ug.edu.gh 16 developmental state of the tissue (Mohan et al., 1997). DNA marker technology has provided a new source of information and an impetus for modifying some plant breeding methods. DNA marker technology is a new source of information, and now is being integrated into existing plant breeding programs all over the world to facilitate transferring and combining desirable genes at a rate and precision not previously thought possible (Mohan et al., 1997). Marker-assisted selection (MAS) can be used successfully to pyramid the major genes for salinity/alkalinity tolerance (Bohnert and Jensen, 1996). Simple Sequence Repeat (SSR) and Quantitative Trait Loci (QTL) analyses have been developed as advanced techniques to obtain meaningful data about the gene complement governing tolerance phenotypes. SSR or microsatellite markers are ideal for making genetic maps (Islam, 2004); assisting selection (Bhuiyan, 2005) and studying genetic diversity in germplasms. SSR markers play also an important role for identifying gene for salt tolerance or introgressing the genes to develop new cultivars. Genes/QTLs have been mapped for several important traits such as yield, quality and tolerance against biotic stresses, and abiotic stresses including drought and salinity/alkalinity. A major QTL was identified conferring salt tolerance on chromosome 1 and designated Saltol (Bonilla et al., 2002; Zhou et al., 2013). This QTL has been the target of marker assisted selection (Thomson et al., 2010). A chromosome 1 QTL SKC1 that appears to be a part of this locus was isolated by positional cloning and determined to be a protein that functions as a Na + selected transporter (Ren et al., 2005; Zhou et al., 2013). Many salt tolerant rice varieties have been developed and have also been adopted, especially in the inland saline/sodic areas of India. An association between SSRs and the alkaline tolerance has been indicated by the presence of genetic polymorphism between parents and segregation of SSRs in an F2 mapping population University of Ghana http://ugspace.ug.edu.gh 17 (Ramayya et al., 2009; Rao et al., 2008). They observed also the occurrence of transgressive segregants and demonstrated the feasibility of finding more tolerant progenies than the tolerant parent by employing proper selection from tolerant extremes of the population distribution. These techniques have the potential to make effective selection of complex traits in early generations. University of Ghana http://ugspace.ug.edu.gh 18 CHAPTER III 3. PRODUCTION CONSTRAINTS AND VARIETAL PREFERENCES OF RICE FARMERS IN THREE HIGH ALKALINITY ZONES OF OFFICE DU NIGER OF MALI 3.1 Introduction In Mali, rice is grown in every region of the country and is increasingly favored by consumers, primarily in urban zones but also in the rural areas where it is produced predominantly by resource-poor farmers. Rice production in Mali relies on large amounts of irrigation water is the main reason for soil alkalinization in Office du Niger. Water, soil and fertilizer management are the main factors used as remedies to solve the alkalinity constraint, but they were and are insufficient. Little work has been done on breeding for tolerance to alkalinity. Breeding for alkalinity tolerance is, however, regarded as an important means of bridging the yield gap between potential and actual yield. Inclusion of farmers in developing tolerant varieties is an important requirement. Farmers have preferences for varietal characteristics such as grain size and color, taste, cooking qualities and high yield. Consideration of farmer preferences in the varietal development process can help the breeders to select appropriate genetic materials (Witcombe et al., 1996; Zhou et al., 2013). Unfortunately, formal research systems in developing countries are highly centralized and do not target the problems of resource-poor farmers (Zhou et al., 2013). It is important for breeders to clearly understand farmers’ constraints and preferences for production, marketing and utilization of crop varieties they produce. Therefore, to improve rice varieties grown by farmers for alkalinity tolerance, it is imperative that farmer’s needs and preferences are understood for integration into the breeding programs. University of Ghana http://ugspace.ug.edu.gh 19 Soil degradation through salts concentration is one of the major constraints for rice production and its current status is not well known. The factors that contribute to such constraints may not be known by farmers. Rice varieties that are released in Mali are mostly farmers preferred varieties, but sensitive to alkalinity stress. These varieties are from the national institute research. Those tolerant varieties developed by Africa Rice (WARDA) and IRRI have low adoption by farmers and this could be attributed to the non integration of farmers’ needs in the breeding programs. In this study, a Participatory Rural Appraisal (PRA) was conducted to evaluate farmers’ traits preferences, perception in relation to salts stress particularly soil alkalinity and crop management practices in Office du Niger rice production zones of Mali. PRA was conducted in these zones to determine farmers’ perceptions and preferences as well as management practices under alkalinity conditions that often occur. The PRA enabled this study to determine ways of ameliorating effects of alkalinity and enhancing irrigated rice production in Office du Niger. The specific objectives of the PRA were to: - identify production constraints of rice farmers in areas with high soil alkalinity; - assess the effectiveness of strategies adopted by farmers in their production environments; - explore farmer’s preferences and adoption criteria for future varietal development. 3.2 Research Methodology PRA approaches which lead to the interaction between breeders and farmers and mutual exchange of knowledge and experience were used to facilitate the interaction with farmers in three zone of Office du Niger in 2011. University of Ghana http://ugspace.ug.edu.gh 20 3.2.1 Site selection The PRA and survey were conducted in the Office du Niger regions of Mali. Three zones were chosen because they had high rates of degraded soils (alkaline) and also offer an opportunity for comparing farmer preference for rice varieties susceptible to alkalinity in irrigated ecosystems. The selected zones were: Niono, Molodo and N’Debougou. Two villages per zones were selected and focus group discussions and structured interviews were used as tools for PRA and data collection in each village. 3.2.2 Selection of farmers Twenty farmers were selected from two villages in each zone of Office du Niger (ON). In Niono zone, the study was conducted in Gnoumanke and Kolodougou villages; in Kangaba and Fabacoura villages in Molodo zone and Banissirela and Siengo in N’Debougou zone. In addition to the twenty farmers per village, one group of fifteen to forty farmers per village was formed for focus group discussions. These groups were selected based on their experiences with rice production in alkaline soil conditions. 3.2.3 Data collection Different PRA techniques were used to obtain information about the farming problems, varietal selection criteria, research priorities and opportunities. A combination of three data collection techniques was employed. These included (1) semi-structured interviews for focus group discussion (FGD); (2) survey questionnaires for individual interviews (Appendix 3.1.); (3) and transect walks for field observation with the groups. Semi-structured interviews were conducted with 10 rice growers from two villages in each zone of Office du Niger. The discussions were followed by transect walks through the rice fields during which alkalinity problems and other biotic and abiotic constraints were identified and University of Ghana http://ugspace.ug.edu.gh 21 rated with the farmers’ groups. Group discussions were held with a selected sample of 10 farmers in each zone to confirm results from questionnaires for individual interviews. Rank matrices were drawn to rank the constraints. Individual farmers ranked biotic and abiotic constraints independently. The constraint with the highest score was considered the most important constraint. Transect walks in the fields of farmers were conducted and observations made. Different traits and plant characters that were considered by farmers in variety selection were recorded. In addition to providing information on variety preferences, special preferences of farmers were identified for home consumption and marketing quality of rice. 3.2.4 Data Analysis Data from questionnaires for individual interviews were compiled and coded and analyzed using Microsoft excel 2007. The coded information data was imported to Statistical Package for the Social Sciences (SPSS) version 17.0 for further analysis. Microsoft excel 2007 was used to calculate average scores and average ranks for data obtained from both group discussions and structure interviews. 3.3 Results 3.3.1 Rice production constraints The use of grower association’s seed for rice production was the most common practice. In general 45.5% of farmers used grower association seeds followed by seeds obtained through self or farm saved seeds (29.1%), professional private seed companies (21.8%) and research stations (3.6%) (Table3.1). Farmer association seed growers were farmers’ groups who came together to produce seeds for contract buyers without any certificate of seed inspectors. Ten (10) constraints were identified as important in rice production zone of Office du Niger. Fertilizer costs were ranked as the most important constraint (15.22%) for farmers in rice University of Ghana http://ugspace.ug.edu.gh 22 production in Office du Niger (Table 3.2). Water management (12.08 %) was ranked as the second most important constraint, followed by inadequate agricultural materials (11.80%). Varietal quality was ranked low at almost all zones. Eight causes of alkalinity were identified as being important in affecting rice production although their importance varied among locations. Nature of soil ploughing (22.5%) and sandy soils (21.4%) were identified by farmers as the first and second most important causes of alkalinity in Office du Niger (Table 3.3). Water quality (surface and ground) and non drainage of irrigated water were third and fourth alkalinity causes. At Gnoumanke, clay soils and poor drainage were the important causes besides nature of ploughing. Soil alkalinity is variable and occurs as several alkalinity types. Four types were identified in Office du Niger (Table 3.4). The most representative in Office du Niger according to farmers is the white salt (33.8%). Except for Kolodougou, black soil (23.3 – 62.5%) was high at all other locations. University of Ghana http://ugspace.ug.edu.gh 23 Table 3.1: Seed source of rice farmers in three zones of Office du Niger (%) in Mali in 2011 Source of seeds N'Debougou a Niono a Molodo a Average Siengo b Banissirila b Kolodougou b Gnoumankè b Kangaba b Fabacoura b Farm saved seed 14.3 61.9 7.7 30.4 20.0 33.3 29.1 Farmer association seed growers 71.4 33.3 38.5 56.5 40.0 16.7 45.5 Private companies 9.5 4.8 46.2 8.7 350 50.0 21.8 Research institutes 4.8 0.0 7.7 4.3 5.0 0.0 3.6 a Zones, b Villages Table 3.2: Rice production constraints and their relative importance (%) ranked in Office du Niger in 2011 Constraints N'Debougou a Niono a Molodo a Average Ranking Siengo b Banissirala b Kolodougou b Gnoumankè b Kangaba b Fabacoura b Fertilizer cost 11.29 12.63 18.29 17.54 16.09 15.50 15.22 1 Water management 12.60 11.61 9.03 14.45 16.09 8.68 12.08 2 * Inad. agricultural equip. 11.29 9.78 12.83 14.45 13.79 8.68 11.80 3 Declining soil fertility 13.65 7.74 9.03 16.59 9.20 10.33 11.09 4 Land shortage 16.27 9.78 9.03 7.11 10.34 13.84 11.06 5 Disease 8.66 14.46 9.03 11.37 9.20 11.98 10.78 6 Seed cost 12.60 12.63 9.03 5.21 9.20 11.98 10.11 7 Delay in land preparation 3.67 8.76 10.93 3.08 3.45 10.33 6.70 8 Weeds 4.99 6.72 5.46 7.11 10.34 5.17 6.63 9 Lack of improved varieties 4.99 5.91 7.36 3.08 2.30 3.51 4.52 10 a Zones, b Villages, *inadequate agricultural equipments University of Ghana http://ugspace.ug.edu.gh 24 Table 3.3: Causes of alkalinity affecting rice production in (%) in 2011 in Office du Niger Causes of alkalinity N'Debougou a Niono a Molodo a Average Siengo b Banissirela b Kolodougou b Gnoumanke b Kangaba b Fabacoura b Continuous cropping 9.7 23.7 12.1 2.7 2.6 5.6 9.4 Water Quality 19.4 7.9 9.1 16.2 15.8 11.1 13.2 * Non-drainage of Irrig. Water 12.9 5.3 9.1 24.3 15.8 5.6 12.2 Clay soils 6.5 5.3 12.1 24.3 5.3 0.0 8.9 Sandy soils 12.9 21.1 21.2 2.7 26.3 44.4 21.4 Silt soils 9.7 10.5 6.1 0.0 2.6 0.0 4.8 Fertilizer application 3.2 10.5 12.1 8.1 0.0 11.1 7.5 Nature of soil plowing 25.8 15.8 18.2 21.6 31.6 22.2 22.5 * Non drainage of irrigated water: Surface and Ground water; a Zones, b Villages Table 3.4: Alkalinity types in rice production zones of Office du Niger in (%) in 2011 Alkalinity (type) Zone N'Debougou a Zone Niono a Zone Molodo a (%) Siengo b Banissirela b Kolodougou b Gnoumanke b Kangaba b Fabacoura b White salt 33.3 32.3 33.3 30.0 57.1 12.5 33.1 Black salt 28.6 38.7 0.0 23.3 23.8 62.5 29.5 Oil salt 38.1 29.0 41.7 40.0 14.3 12.5 29.3 Red salt 0.0 0.0 25.0 6.7 4.8 12.5 8.2 a Zones, b Villages University of Ghana http://ugspace.ug.edu.gh 25 3.3.2 Importance of soil alkalinity in Office du Niger There were differences between the six villages in terms of impact of alkalinity type on rice production in Office du Niger (Table 3.5). However, according to the farmers, the white salt represented 36.1 %, followed by black (27.9%) and oil (27.7%) salts. Table 3.5: Impact of alkalinity types on rice production in Office du Niger in 2011 Salts Zone N'Debougou a Zone Niono Zone Molodo (%) Siengo b Banissirela Kolodougou Gnoumanke Kangaba Fabacoura White salt 42.1 38.7 33.3 30.8 59.1 12.5 36.1 Black salt 26.3 32.3 8.3 15.4 22.7 62.5 27.9 Oily salt 31.6 29.0 33.3 46.2 13.6 12.5 27.7 Red salt 0.0 0.0 25.0 7.7 4.5 12.5 8.3 a Zones, b Villages In Gnoumanke, the oil salt has the greatest impact of 46.2% and at Fabacoura the black salt had 62.5% (Table 3.5). Farmers’ adopted various strategies for alkalinity impacts control including crop rotations (25%), use of organic matter (24%) and puddling followed by flushing (22%). The adoptions of the above strategies were higher than the use of tolerant varieties (17%) and pre flooding (12%) (Table 3.6). Table 3.6: Farmers’ strategies for alkalinity control across Office du Niger villages in 2011 Strategies Percent Puddling following by flushing 22 Pre-flooding 12 Use of organic matters 24 Crop rotation 25 Tolerant varieties 17 Kogoni 91-1 variety was adopted by farmers as being tolerant to alkalinity in all villages except Kolodougou and Fabacoura (Table 3.7). On average, 19% of farmers selected Kogoni 91-1 as University of Ghana http://ugspace.ug.edu.gh 26 tolerant to alkalinity; and 66%, with a range of 54 to 76% did not consider any variety as being tolerant to alkalinity. Table 3.7: Proportion of farmers (%) perception growing tolerant varieties in alkaline soils in 2011 N'Debougou a Niono Molodo (%) Varieties Siengo b Banissirela Kolodougou Gnoumanke Kangaba Fabacoura Kogoni 91-1 24 33 0 22 35 0 19 Nionoka 0 0 23 0 0 0 4 Adny11 0 0 8 9 0 25 7 Nerica L2-IER 0 0 8 0 0 0 1 Gambiaka dian 0 0 8 0 0 0 1 Nerica7 0 0 0 0 0 8 1 no variety 76 67 54 69 65 67 66 a Zones, b Villages 3.3.3 Farmers’ varietal preference and selection criteria There are a number of rice varieties grown by Farmers in Office du Niger although they vary from one village to another (Table 3.8). Kogoni 91-1 (35.3%) is grown more than the other varieties. ‘‘Nionoka’’ with an average of 13.4% is also grown in all six villages. University of Ghana http://ugspace.ug.edu.gh 27 Table 3.8: Percentage of farmers growing different rice varieties in Office du Niger in 2011 varieties N'Debougou a Niono zone Molodo zone (%) Siengo b Banissirela Kolodougou Gnoumanke Kangaba Fabacoura Kogoni 91-1 36.2 25.6 27.6 40.4 46.5 40.7 35.3 BG 90-2 5.2 1.3 13.8 12.3 0.0 11.1 6.2 Sambala 19.0 10.3 0.0 10.5 0.0 14.8 9.9 Nerica L2-IER 0.0 9.0 3.4 1.8 0.0 0.0 3.1 Nerica L1-IER 0.0 1.3 0.0 0.0 0.0 0.0 0.3 Nionoka 24.1 5.1 41.4 7.0 4.7 11.1 13.4 Telimani 0.0 0.0 0.0 8.8 0.0 3.7 2.1 Wassa 5.2 21.8 0.0 5.3 18.6 18.5 12.3 Adny11 8.6 25.6 13.8 14.0 30.2 0.0 17.1 ECIA 1.7 0.0 0.0 0.0 0.0 0.0 0.3 a Zones, b Villages Farmer preferences tended to vary based on yield, market and taste (Table 3.9). The non- preference rate per variety is an important parameter to rank farmer preference for varieties. Kogoni 91-1 was the least preferred variety with 13.6%, followed by BG 90-2 (25.5%) and Adny 11 (43.5%). The other varieties had non-preference ratings of more than 50% (Table 3.9). The two Nerica varieties and ECIA were grown at rather limited locations compared to other varieties. University of Ghana http://ugspace.ug.edu.gh 28 Table 3.9: Factors affecting farmer preferred rice varieties for production in Office du Niger Factors Kogoni 91-1 BG 90-2 Sambala malo Nerica L2- IER Nerica L1- IER Nionoka Telimani Wassa Adny 11 ECIA Yield 26.4 69.1 10.9 6.4 0.0 13.6 0.0 10.9 40.0 0.9 Market 23.6 2.7 0.9 0.9 0.0 9.1 0.0 2.7 8.2 0.9 Taste 23.6 0.0 0.9 0.0 0.0 0.9 0.9 0.9 4.5 0.0 Late fertilization 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Fertilizer response 9.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Grow on sandy soil 0.9 0.0 0.0 0.0 0.0 0.9 0.0 0.0 0.0 0.0 Grain quality 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Maturity 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Swelling after cooking 0.0 2.7 0.9 0.0 0.0 0.0 0.9 0.0 2.7 0.0 Early maturity 0.0 0.0 7.3 0.0 0.0 3.6 1.8 10.9 0.0 0.0 Quality after cooking 0.0 0.0 2.7 0.0 0.0 1.8 0.0 0.0 0.0 0.0 Young seedling vigor 0.0 0.0 0.0 0.9 0.0 0.9 0.9 0.9 0.0 0.0 Low inputs 0.0 0.0 0.0 0.0 0.0 68.2 0.0 0.0 0.9 0.0 No preferences 13.6 25.5 76.4 91.8 90.9 100.0 95.5 73.6 43.6 98.2 University of Ghana http://ugspace.ug.edu.gh 29 Across the six villages, the results for home consumption criteria showed that taste and swelling after cooking (30.9%) were more appreciated by farmers than other attributes (Table 3.10). Taste (21.8%), swelling after cooking (20%) and the storage after cooking (15.5%) were identified by farmers as good criteria for rice preference for home consumption. Table 3.10: Criteria for selecting rice for home consumption in Office du Niger in 2011 Consumption criteria Percent Ability to store 15.5 Swelling after cooking 20.0 Taste 21.8 Swelling after cooking and storing 10.9 Taste and swelling after cooking 30.9 No response 0.9 For preference for marketing, grain colour and size were identified by farmers as most important (Table 3.11). For color, white (57.3%) and the white yellowish (26.4%) were preferred by rice buyers in the market. The long slender grain and less breaking were identified by 90% of farmers as rice buyer’s preference in the market. Table 3.11: Attributes of grain considered by farmers for selection for the market in Office du Niger in 2011 Grain color Percent White bluish 3.6 White yellowish 26.4 White 57.3 White greenish 2.7 No response 10.0 Grain size Long slender and less breakage 90.0 Medium and less breakage 0.9 No response 9.1 University of Ghana http://ugspace.ug.edu.gh 30 3.3.4 Farmer priorities for research needs Table 3.12 shows farmers’ suggestions of traits they prefer to be incorporated into rice varieties. A total of 58.2% of the farmers interviewed wanted vegetative tolerance to salts and spikelet fertility, 13.6% wanted yield and taste, 10% wanted yield and 13.6% did not suggest any specific trait. Table 3.12: Farmers’ preference for rice traits in Office du Niger in 2011 Suggested traits Percent Vegetative and spikelet fertility tolerance 58.2 Yield 10.0 Taste .9 Yield and taste 13.6 Avoid lodging .9 Swelling 1.8 Store after cooking .9 No proposition 13.6 3.4 Discussion Results from Participatory Rural Appraisal revealed that most farmers used seeds purchased from the farmers’ association. Farmers in the relatively homogenous irrigated zones used improved varieties because of their high value returns. The varieties grown were not improved for stress tolerance and therefore they succumbed to alkalinity damages. AfricaRice (2010) observed that most farmers selected seeds for planting from the conventional seed (seed from Farmer association seed growers) in irrigated systems. Conventional seed systems are not always well adapted to the specific abiotic stress. Hence, the production of unimproved varieties that are susceptible to abiotic stress is still continued. University of Ghana http://ugspace.ug.edu.gh 31 Some farmers obtained their seeds from seed companies because of the proximity of their village to seed company’s retail outlet. For example in Office du Niger, there are two seed companies involved in seed certification for good quality and marketing (Dembélé, 2006). One company is in Niono zone where Kolodougou village is located and another one is in Molodo zone which is close to Fabacoura village. Both of the two villages have easy access to the improved seed but the varieties are not tolerant to alkalinity. A number of factors were listed as constraints to rice production across the three zones in Office du Niger. These included fertilizer costs, water management, insufficient agricultural equipment, declining soil fertility and land shortage. These constraints are in agreement with a previous survey conducted in 1995 (Ouvry et al., 1999). In that study, the nature of plowing the soil was the most important cause of alkalinity curtailing rice production in Office du Niger, followed by sandy soil type, water quality and non drainage of irrigated water. In this study, differences were observed in causes of alkalinity ranking across the three zones of Office du Niger. In N’Debougou and Niono, nature of plowing the soil was the number one cause of alkalinity. This could probably be attributed to three factors. The first one is the impact of cropping systems (Bagayoko et al., 2010). Most farmers have continued to use the local plow for rice field plowing and this is unable to level the soil surface efficiently. It creates shallow areas in the field where salt concentration is high. This situation can explain the constraint of inadequate agricultural equipment mentioned by farmers. The second factor is the effect of soil types on evolution of water table in terms of dynamics and quality (Bagayoko et al., 2010; N’Diaye, 1998; Ouvry et al., 1999). The soil type in these two zones is largely clay soil. Sodicity and alkalinity are environmental factors that greatly affect plant growth and development in clay soils. Harmful high alkalinity is related principally to the presence of sodium carbonate and sodium bicarbonate University of Ghana http://ugspace.ug.edu.gh 32 (Bagayoko et al., 2010; Ouvry et al., 1999). Accumulation of these salts in the soil and soil water occurs when groundwater with a high content of bicarbonate ions evaporates. Initially, calcium and magnesium carbonates precipitate upon evaporation, immobilizing part of bicarbonate ions. If excess bicarbonate is present (called residual alkalinity), high pH values will develop, even at initially low concentrations of sodium carbonate. In strongly alkaline soils, sodium carbonate and bicarbonate become prevalent salts, and sodium will replace calcium and magnesium in the clay complex (sodic soils). This leads to the formation of sodium clays; those are strongly dispersed and highly impermeable to water. The third factor is the lack of tolerance to alkalinity in improved rice varieties being grown. Alkalinity has been an abiotic stress of importance in Office du Niger but there has not been any active breeding programme in Mali specifically addressing tolerance to alkalinity stress. The environmental conditions in the breeding area have not favored the selection for tolerance against alkalinity. Hence most of the materials released do not have the genetic tolerance against soil degradation by alkalinisation. In Molodo zone, the sandy soil type is the main cause of alkalinity. This situation could be attributed to soil permeability factors (Bagayoko et al., 2010; Dicko, 2005; N’Diaye, 1998). The previous studies (N'diaye et al., 2002) showed the sensitivity of sandy soil types to soil degradation and the impact of drainage as means of controlling of the ascent by capillarity. Soils of irrigated areas in Sub Sahara African countries have continued to be degraded as a result of poor irrigation management practices. In the Office du Niger zones, for example, producers and extension workers are concerned about emerging soil degradation symptoms such as salinisation/alkalinisation or sodisation. Whereas the above problems require urgent attention, the phenomenon has been treated as localised and therefore not very important. From 1995 to 1999, studies were conducted by “the Pole regional de recherche sur les Systèmes Irrigués (PSI)” University of Ghana http://ugspace.ug.edu.gh 33 which was a regional networking project, to determine the nature, importance and dynamics of soil degradation processes in irrigated zones of Mali. The results clearly show soil geochemical changes of the irrigated areas (Marlet et al., 1998). However it is important to note that, during these studies, the breeding program was not included and this could be the main reason for the lack of rice genotypes tolerant to alkalinity, although all rice varieties produced are improved for yield, grain quality and taste. Alkalinity tolerance is a polygenic trait acting both additively and with interactions between the alleles at some loci (Singh et al., 2001), which implies that there will always some symptoms even when improved materials that carry the tolerant genes are used. For farmers, most improved varieties suffer alkalinity damage and among them Kogoni 91-1 is considered as tolerant to alkalinity by 19% of farmers. All seeds for sale by certification system are considered as commercialized seeds in Mali (Dembélé, 2006). The seed that is grown for many years and not renewed is also considered as commercialized seeds. Often the commercial seeds producers don't observe the precautions that require the maintenance of the purity and the genetic identity. As a result, the commercialized seeds of most of the varieties are not genetically homogeneous. By selecting for seeds from uninfected rice plants, farmers tend to accumulate susceptible materials hence increasing alkalinity damage. Most released varieties grown by farmers were not improved for salinity and alkalinity tolerances and could have a negative impact with alkalinity problems. Soil degradation control methods being employed by the farmers are limited and consist of crop rotation, use of organic matter, puddling followed by flooding and use of tolerant varieties. Many do not practice any control method, either because they lack knowledge on control technologies, or they could not afford to use chemical control. University of Ghana http://ugspace.ug.edu.gh 34 For alkalinity, the effectiveness of these control methods depends on a number of factors including, soil and irrigated water management, crop growth stage at which the susceptibility of plant is shown and the genetic make-up of the plant variety in question. Rice is more sensitive to salts during early seedling stage than at reproductive stages (Flowers and Yeo, 1981). Practices such as pre-flooding, puddling, flushing and use of chemical and organic matters could help control yield reduction (Dicko, 2005). A more effective and practical solution for subsistence farmers is high yielding rice varieties that are tolerant to alkalinity stress (Singh and Flowers, 2010). Farmers’ preferences when choosing varieties for market and home consumption differed. Farmers identified taste and swelling after cooking as selection criteria for home consumption and grain colour and size as criteria for market. In comparison to many other nationalities in West Africa, the majority of Malian consumers prefer local rice over the imported rice, due to its preferred taste and swelling after cooking (IFDC, 2008). The most popular domestic rice variety is the Kogoni 91-1 variety, which is the dominant variety produced in the controlled irrigated systems. This variety is more appreciated by urban consumers because of its taste and aroma. However BG 90-2 variety is more preferred by rural consumers for its swelling. These two varieties are improved varieties released by Irrigated Rice Programme of National Research Institute of Mali. It is important to note that these two varieties are susceptible to alkalinity stress and to Rice Yellow Mottle Virus (RYMV) disease. Standards for grain size and shape differ among countries, although the Food and Agriculture Organization (FAO) has developed international standards. Translucency and color are affected by environment and their assessment is largely subjective. For marketing of milled rice, grain University of Ghana http://ugspace.ug.edu.gh 35 color and size were the criteria identified by farmers as attractive to rice traders in Office du Niger. The white color was preferred by majority of farmers as criteria for selling rice to the traders. White, long and slender grain rice is more preferred by Malian consumers (IFDC, 2008; USAID, 2007). There is a positive correlation between grain length and grain shape (Gupta et al., 2006) suggesting that breeders could select for both traits simultaneously, or they could choose one of the traits to indirectly select for the other. Grain size is usually evaluated by two criteria: grain weight and shape (Fan et al., 2006; Tan et al., 2000). Grain weight and grain shape (length, width, length/width ratio) are positively correlated characters (Fan et al., 2006). Besides taste and swelling after cooking for home consumption and grain colour and size for market, farmers had some other preferences that were identified by the survey. These included vegetative and spikelet fertility tolerance to alkalinity, a combination of yield and taste and yield alone. There is need to integrate such special farmer preferences into selection criteria. Involvement of farmers in varietal adoption would help bridge the information gap between farmers and breeders. It has been reported that farmers grow the same varieties for long periods of time. This could probably be because they have failed to find new varieties with special qualities they desire. Farmers’ preferences for home consumption were different from preference for market consumption. Farmer desire for market is driven by the traders and consumers needs or demands. Farmer request is based on consumers. In this regard, different breeding objectives should be formulated with input from the farmers to address these special preferences. Compared with conventional plant breeding, Participatory Plant Breeding (PPB) is more likely to produce farmer-acceptable products, particularly for marginal environments. While meeting the farmers’ needs requires continuous engagement of the farmers in the breeding process, the reality University of Ghana http://ugspace.ug.edu.gh 36 of this remains a scaring task to breeders. On the other hand, when a variety fails to meet farmers’ needs during participatory varietal selection (PVS), farmers would not adopt it and the effort and resources spent would have been wasted. 3.5 Conclusion The survey revealed four dominant factors as rice production constraints, namely fertilizer cost, water management, inadequate agricultural materials, and declining soil fertility. The combination of the four factors constitutes the main cause of soil alkalinisation in Office du Niger. Overall, alkalinity was the most important abiotic stress for rice production. The type of alkalinity varies from village to village and from zone to zone, however the white salt was more common. To overcome alkalinity constraints, farmers have developed some strategies like crop rotation, use of organic matters, plowing followed by flooding and use of tolerant varieties. The absence or weakness of a breeding program for alkalinity tolerance was apparent. Farmers have a basic preference for taste and swelling for home consumption and grain colour and size for marketing. Research priorities, as perceived by the farmers, included alkalinity tolerance of vegetative and spikelet fertility combined, combination of yield and taste and yield alone. Farmers had high interest in participatory varietal selection and participatory plant breeding. University of Ghana http://ugspace.ug.edu.gh 37 CHAPTER IV 4. IDENTIFICATION OF ALKALINE TOLERANT RICE ACCESSIONS AND THEIR ASSOCIATION WITH SSR MARKERS THROUGH GENETIC DIVERSITY 4.1 Introduction Rice production in Mali is threatened by several factors including soil degradation, alkalinity and/or use of brackish ground water. Mali experiences huge recurring losses in terms of limited productivity from the salt-affected lands, apart from serious socio-economic repercussions. With the increase in population, effective utilization of these soils has become necessary either by reclamation or by growing salt tolerant rice. Rice (Oryza sativa L.), is quite sensitive to salinity and alkalinity, particularly during early seedling and reproductive stages. Rice yield starts to decline beyond a threshold pH of 7.5, and a drastic reduction in yield occurs at pH 9.0. However rice is one of the few crops that can thrive on salt-affected soils. Being a species native to brackish ground water, it is able to grow well in standing water that can help leach salts from top soils, and therefore rice is recommended as a salts flushing crop (Marlet et al., 1998). The threat of soil alkalinity to the sustainability of rice agriculture can be overcome by developing and adopting rice-farming practices that can help prevent or arrest soil alkalinization. Soil remediation schemes can also be undertaken to rehabilitate salt-affected rice lands. However, these management and reclamation practices require considerable capital investments that are beyond the capability of the resource-poor farmers whose source of livelihood are these alkaline lands. Longer-term and affordable solutions can be achieved by developing rice varieties that are highly tolerant to salt stress. This can best be attained by profiling salt-management morpho-physiological processes in rice cultivars, which could subsequently facilitate the University of Ghana http://ugspace.ug.edu.gh 38 development of conventional and molecular breeding as well as genetic engineering approaches to combine these traits and speed up the development of varieties that are tolerant to salt stress. Salt tolerance in rice involves several physiological and adaptive mechanisms (Ismail et al., 2007; Peng and Ismail, 2004; Yeo and Flowers, 1986). The physiological bases of salt tolerance during early seedling stage involve traits such as high seedling vigor to dilute salt concentration in plant tissues, selective ion uptake by roots, compartmentalization of harmful ions in older tissues, such as older leaves, stems, leaf sheaths, and roots, responsive stomata that regulate water and salt uptake in response to increasing salt stress in the rhizosphere and re-circulating sodium back to roots to avoid accumulation of toxic concentrations in the cytoplasm. In the reproductive stage, tolerant genotypes tend to exclude salt from flag leaves and developing panicles (Khatun et al., 1995; Yeo and Flowers, 1986). Despite a wealth of knowledge on physiological mechanisms associated with salt tolerance in rice, little is known about the genetic diversity of these individual traits across rice germplasm. This is because most of the tolerant genotypes characterized to date seem to be superior in only one or a few of these traits (Ismail et al., 2007; Moradi and Ismail, 2007; Yeo and Flowers, 1986). Nonetheless, this information is essential for breeders to select appropriate parental lines by choosing genotypes that are superior in expressing a particular trait when combining or pyramiding these characters to develop highly tolerant varieties. Rice genotypes may be characterized as tolerant or sensitive based on the various responses examined. Genotypes identified could be selected for subsequent use of specific traits in breeding programs for selecting high yielding salt-tolerant varieties. Genetic characterization of the causal mechanisms related to these morpho-physiological adaptive characters would facilitate their use in developing resilient cultivars with multiple traits associated with salt tolerance. University of Ghana http://ugspace.ug.edu.gh 39 The objective of this study was to determine the variability in morpho-physiological and genotypic responses of some rice cultivars from IRRI and Africa Rice to alkalinity stress. 4.2 Materials and methods 4.2.1 Plant materials 4.2.1.1. Identification of salt tolerant accessions under alkaline soil condition Twenty six (26) rice accessions including, IR 29 (sensitive check) and Kogoni 91-1 (local checks), were used to identify salt tolerant accessions under alkaline soil conditions of Office du Niger in Mali (Table 4.1). Table 4.1: Rice accessions for phenotypic study No Accessions Sources Accessions Sources 1 D14 AfricaRice 14 IR29 (sensible check) Africa Rice 2 IR 66401-2-B-6-1-3 AfricaRice 15 (tolerant check) Africa Rice 3 IR 71829-3R-89-1-1 AfricaRice 16 Dasal IRRI 4 IR 72593-B-3-2-3-14 AfricaRice 17 CSR11 IRRI 5 ROHYB 6 AfricaRice 18 IR4630-22-2-5-1-3 IRRI 6 IR 66946-3R-178-1-1 Africa Rice 19 Damodar Haryana (CSR1) IRRI 7 IR 71991-3R-2-6-1 Africa Rice 20 Damodar India (IRGC 17038) IRRI 8 SAHEL 210 AfricaRice 21 Getu (IRGC 17041) IRRI 9 IR 76393-2B-7-1-13-1 Africa Rice 22 Getu (IRGC 22709) IRRI 10 IR 65192-4B-11-3 AfricaRice 23 CSR10 IRRI 11 IR 71895-3R-60-3-1 AfricaRice 24 CSR28 IRRI 12 IR 76346-B-B-10-1-1-1 Africa Rice 25 CSR36 IRRI 13 Kogoni 91-1 (local check) Africa Rice 26 CSR23 (SAL 332) IRRI 4.2.1.2. Determination of genetic diversity Twenty eight rice genotypes (28) of which 24 accessions and 4 varieties, were used to determine allelic diversity using molecular markers (Table 4.2). University of Ghana http://ugspace.ug.edu.gh 40 Table 4.2: Rice genotypes for allelic diversity study No Rice genotypes Sources No Rice genotypes Sources 1 Nerica L1-IER IER-Mali 15 IR29 (sensitive check) from student AfricaRice 2 BG 90-2 IER-Mali 16 Pokkali (tolerant check) from student AfricaRice 3 IR 66401-2-B-6-1-3 AfricaRice 17 Dasal IRRI 4 IR 71829-3R-89-1-1 AfricaRice 18 CSR11 IRRI 5 Nerica L2-IER IER-Mali 19 Damodar Haryana (CSR1) IRRI 6 WAT 310 IER-Mali 20 Damodar India (IRGC 17038) IRRI 7 IR 66946-3R-178-1-1 AfricaRice 21 Getu (IRGC 17041) IRRI 8 IR 71991-3R-6-3-1 AfricaRice 22 Getu (IRGC 22709) IRRI 9 SAHEL 210 AfricaRice 23 CSR10 IRRI 10 FL478 from AfricaRice AfricaRice 24 Hassawi from AfricaRice AfricaRice 11 IR 65192-4B-11-3 AfricaRice 25 CSR23 (SAL 332) IRRI 12 IR 71895-3R-60-3-1 AfricaRice 26 IR4630-22-2-5-1-3 AfricaRice 13 IR 76346-B-B-10-1-1-1 AfricaRice 27 IR29 (sensitive check) AfricaRice AfricaRice 14 Kogoni 91-1 AfricaRice 28 Pokkali (Tolerant check) AfricaRice AfricaRice 4.2.1.3. Characterization of phenotypic and genetic associations Nineteen (19) accessions were used to study the phenotypic and genotypic association (Table4.3). Table 4.3: Rice accessions for phenotypic and genotypic association study No Rice accessions Sources No Rice accessions Sources 1 Kogoni 91-1 (local check) AfricaRice 11 SAHEL 210 AfricaRice 2 CSR11 IRRI 12 IR 71829-3R-89-1-1 AfricaRice 3 CSR10 IRRI 13 IR 71895-3R-60-3-1 AfricaRice 4 Damodar Haryana (CSR1) IRRI 14 CSR23 (SAL 332) IRRI 5 IR 76346-B-B-10-1-1-1 AfricaRice 15 IR 71991-3R-2-6-1 AfricaRice 6 IR 66946-3R-178-1-1 AfricaRice 16 Dasal IRRI 7 Damodar India (IRGC 17038) IRRI 17 IR 65192-4B-11-3 AfricaRice 8 Pokkali (tolerant check) AfricaRice 18 IR29 (sensible check) AfricaRice 9 Getu (IRGC 17041) IRRI 19 Getu (IRGC 22709) IRRI 10 IR 66401-2-B-6-1-3 AfricaRice University of Ghana http://ugspace.ug.edu.gh 41 4.2.2. Methodology 4.2.2.1 Identification of salt tolerant rice accessions under alkaline soil condition Twenty six (26) accessions were screened in Randomized Complete Block Design with two replications. Direct seeding of 3 grains/pot was done after soil crushing. Plants were thinned to two seedlings per pot after germination. The genotypes were sown on June 06, 2010 in the greenhouse of Centre regional de Recherche Agronomique (CRRA) de Gao. The pot used was 20 cm in diameter and 50 cm high with 3 kg alkali soil of Office du Niger with 9.9 of pH, 8.9 mg/100g of Na + and 0.43 mg/100g of K + . The ratio Na + /K + was 20.7. Twenty days after plant emergence, three holes were made in the bottom of each pot and then placed in a hollow plate. After each, irrigation the water which was collected in the plate was returned to the pot to maintain alkalinity level in soil. Experimental pots were fertilized using: urea = 0.69g/pot (220 kg/ha) and di-ammonium phosphate (DAP) = 0.31g/pot (100 kg/ha). The DAP was used during direct seeding and urea was applied after plants had 3 to 5 leaves and the second application was during panicle initiation. Soil and plants were characterized to evaluate the nutrient uptake of each genotype. The analyses of soil were done for all genotypes in one composite soil before sowing them individually in pots. The composite soil was obtained from the mixture of soils from all pots. Soils were analyzed before sowing and after the flowering stage for the key parameters of pH, exchangeable cations (K + and Na + ) and the ratio of Na + /K + . Plant analyses on flag leaves were done on the same elements as in soil after flowering stage. These analyses were done at Soil Science Department of University of Ghana (Legon). University of Ghana http://ugspace.ug.edu.gh 42 4.2.2.2. Determination of genetic diversity Three sources of rice genotypes were used for allelic diversity for alkalinity tolerance. These were sourced from IRRI in Philippine, AfricaRice in Saint Louis (Senegal) and IER in Mali (Table 4.1). Ten flanking/nearest markers as Simple Sequence Repeat (SSR) covering the chromosomes 1, 2 4, 8 and 9 were selected for the genetic diversity analysis. Table 4.4 shows details on selected makers. The original source, repeat motifs, primer sequences and chromosomal position for these markers can be found in the RiceGenes database (http://www.gramene.org/microsat/RM_primers.html). SSR markers were obtained from AfricaRice Research Genetics and Biotechnology; Saint Louis, Senegal. Table 4.4: Selected markers used for genetic diversity for alkalinity tolerance Gene/QTLs Chr . No Traits Flanking/nearest Marker Size bp References qNa + SV1-1 1 Na + in stem at vegetative stage RM572 147 – 181 Na Ammar et al. (2009) QSst2 2 Score of salt toxicity of leaves and/ survival day of seedling RM250 RM208 154 – 174 166 – 180 Zang et al. (2008) Trait based QTL 4 K + absorption RM537 RM551 230 – 236 168 – 214 Gregorio (1997) QPh4 4 Plant High RM142 RM273 Singh and Flower (2010) qNa + /K + SV 8-1 8 Na + /K + ratio in stem at vegetative stage RM477 219 – 223 Ammar et al. (2009) Saltol (as control) RM493 RM215 211 – 259 The DNA of the 26 genotypes (accessions and varieties) was extracted from juvenile leaves of 3- week-old plants according to the simplified protocol modified CTAB method, (Bimpong et al., 2010). The Polymerase Chain Reaction (PCR) was performed in 10 μL reactions containing 2 μL of DNA template, 4.5 μL ddH2O, 1 μL 10X TB buffer (containing 200mM Tris-HCl pH 8.3, University of Ghana http://ugspace.ug.edu.gh http://www.gramene.org/microsat/RM_primers.html 43 500mM KCl, 15mM MgCl2), 1 μL of 1mM dNTP, 0.3 μL MgCl2, 0.50 μL each forward and reverse primers, and 0.5 μL of Taq DNA polymerase. The thermocycler was used to amplify the PCR product (DNA) with the program Saltol60. The PCR products were checked using the high resolution polyacrylamide gel electrophoresis (PAGE) for the primers size < 200 pb for 2hr 30mn running, and agarose gel for the primers size ≥ 200 bp for 3 hr running. The gels used are polyacrylamide 7.5% and agarose 3%. The gels were stained in 0.5 mg/mL ethidium bromide for 15 – 30 minutes and were documented using Multido lt digital Imaging System UVP. 4.2.2.3. Phenotypic and genotypic association study for alkaline tolerance rice identification Nineteen rice accessions morpho-physiological characters (% uptake of Na + and K + in the flag leaves, the ratio Na + /K + , % spikelet fertility, weight of 100 grains (g) and the grain weight/plant) were evaluated. Ten simple sequence repeats (SSR) markers (Table 4.4) were used for molecular analysis to validate the markers effects on morpho-physiological characters. 4.2.3. Data collection 4.2.3.1. Identification of salt tolerant rice accessions under alkaline soil condition The analysis of soil and plant was done in Soil Science Department laboratory of Ghana University (Legon). The data’s collected included pH, exchangeable cations (K + and Na + ) and the ratio Na + /K + . For morphological and physiological characteristics, the plants were developed in the pot and evaluated for 8 traits: 1) % Na + uptake (%Na + ) was calculated as the percent of Na + in the flag leaf. 2) % K + uptake (%K + ) was calculated as the percent of K + in the flag leaf. 3) Na + / K + ratio was calculated as the ratio Na + / K + in the flag leaf. University of Ghana http://ugspace.ug.edu.gh 44 4) Germination percentage (GM%) was calculated as the number of seeds germinated as a percentage of the total number of seed sowed. 5) Plant vigor at seedling stage (vg) was measured after 21 days after sowing (at 3 to 5) leaves to assess salt injury level in alkaline condition using the Standard Evaluation Score (SES) of IRRI, 2002 (Table 4.5). 6) The spikelet fertility in percent (sp fert%) as the percent of filled grain per plant in alkaline condition at reproductive stage. 7) The filled grain weight per plant (gw) was calculated as the weight of the fully filled grain of the plant in grams. 8) Weight of 100 grains per accession was recorded as the weight of 100 filled grains for each accession in grams. Table 4.5: Standard Evaluation System score (SES) Score Observation Vegetative Vigor 1 Fast growing; plants at 5 leaf stage have 2 or more tillers Extra vigorous 3 Fast growing plants at 4-5 leaf stage have 1-2 tillers· in majority of population Vigorous 5 Plants at 4-leaf stage Normal 7 Plants somewhat stunted; 3-4 leaves; thin population; no tiller formation Weak 9 Stunted growth; yellowing of leaves. Very weak Source: IRRI (2002) 4.2.3.2. Determination of genetic diversity The size (in nucleotide base pairs) of the most intensely amplified band for each SSR marker was scored based on its migration relative to a molecular-weight size marker (10 bp DNA ladder). University of Ghana http://ugspace.ug.edu.gh 45 4.2.3.3. Phenotypic and genotypic association study for alkaline tolerance rice identification Refer to 4.2.3.1 and 4.2.3.2 for data collection. 4.2.4. Data Analysis 4.2.4.1. Identification of salt tolerant rice accessions under alkaline soil condition Analyses of variance were conducted using the procedure General Linear Model from SAS 9.2 software with rice accessions being considered as fixed effects, and replications as random effects. The Least Significant Means of plant vigor at seedling stage, the % spikelet fertility, the grain weight (g), the weight of 100 grains (g) associated wi